From 80492e7d49bee0a280a84a39075a7857b92836b2 Mon Sep 17 00:00:00 2001
From: Roel Kluin <roel.kluin@gmail.com>
Date: Tue, 21 Apr 2009 16:08:39 +0200
Subject: rpcgss: remove redundant test on unsigned

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/export.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c
index 5839b229cd0e..6eb918153fd4 100644
--- a/fs/nfsd/export.c
+++ b/fs/nfsd/export.c
@@ -464,16 +464,11 @@ static int secinfo_parse(char **mesg, char *buf, struct svc_export *exp)
 		if (err)
 			return err;
 		/*
-		 * Just a quick sanity check; we could also try to check
-		 * whether this pseudoflavor is supported, but at worst
-		 * an unsupported pseudoflavor on the export would just
-		 * be a pseudoflavor that won't match the flavor of any
-		 * authenticated request.  The administrator will
-		 * probably discover the problem when someone fails to
-		 * authenticate.
+		 * XXX: It would be nice to also check whether this
+		 * pseudoflavor is supported, so we can discover the
+		 * problem at export time instead of when a client fails
+		 * to authenticate.
 		 */
-		if (f->pseudoflavor < 0)
-			return -EINVAL;
 		err = get_int(mesg, &f->flags);
 		if (err)
 			return err;
-- 
cgit v1.2.3-59-g8ed1b


From a9e61e25f9d2e7e43bf17625f5cb56c9e0a89b17 Mon Sep 17 00:00:00 2001
From: Felix Blyakher <felixb@sgi.com>
Date: Tue, 31 Mar 2009 15:12:56 -0500
Subject: lockd: call locks_release_private to cleanup per-filesystem state

For every lock request lockd creates a new file_lock object
in nlmsvc_setgrantargs() by copying the passed in file_lock with
locks_copy_lock(). A filesystem can attach it's own lock_operations
vector to the file_lock. It has to be cleaned up at the end of the
file_lock's life. However, lockd doesn't do it today, yet it
asserts in nlmclnt_release_lockargs() that the per-filesystem
state is clean.
This patch fixes it by exporting locks_release_private() and adding
it to nlmsvc_freegrantargs(), to be symmetrical to creating a
file_lock in nlmsvc_setgrantargs().

Signed-off-by: Felix Blyakher <felixb@sgi.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/lockd/svclock.c | 2 ++
 fs/locks.c         | 3 ++-
 include/linux/fs.h | 1 +
 3 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c
index 83ee34203bd7..e577a78d7bac 100644
--- a/fs/lockd/svclock.c
+++ b/fs/lockd/svclock.c
@@ -326,6 +326,8 @@ static void nlmsvc_freegrantargs(struct nlm_rqst *call)
 {
 	if (call->a_args.lock.oh.data != call->a_owner)
 		kfree(call->a_args.lock.oh.data);
+
+	locks_release_private(&call->a_args.lock.fl);
 }
 
 /*
diff --git a/fs/locks.c b/fs/locks.c
index ec3deea29e37..b6440f52178f 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -151,7 +151,7 @@ static struct file_lock *locks_alloc_lock(void)
 	return kmem_cache_alloc(filelock_cache, GFP_KERNEL);
 }
 
-static void locks_release_private(struct file_lock *fl)
+void locks_release_private(struct file_lock *fl)
 {
 	if (fl->fl_ops) {
 		if (fl->fl_ops->fl_release_private)
@@ -165,6 +165,7 @@ static void locks_release_private(struct file_lock *fl)
 	}
 
 }
+EXPORT_SYMBOL_GPL(locks_release_private);
 
 /* Free a lock which is not in use. */
 static void locks_free_lock(struct file_lock *fl)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 5bed436f4353..5ba615e8f533 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1108,6 +1108,7 @@ extern void locks_copy_lock(struct file_lock *, struct file_lock *);
 extern void __locks_copy_lock(struct file_lock *, const struct file_lock *);
 extern void locks_remove_posix(struct file *, fl_owner_t);
 extern void locks_remove_flock(struct file *);
+extern void locks_release_private(struct file_lock *);
 extern void posix_test_lock(struct file *, struct file_lock *);
 extern int posix_lock_file(struct file *, struct file_lock *, struct file_lock *);
 extern int posix_lock_file_wait(struct file *, struct file_lock *);
-- 
cgit v1.2.3-59-g8ed1b


From 78155ed75f470710f2aecb3e75e3d97107ba8374 Mon Sep 17 00:00:00 2001
From: Bian Naimeng <biannm@cn.fujitsu.com>
Date: Wed, 22 Apr 2009 18:25:37 +0800
Subject: nfsd4: distinguish expired from stale stateids

If we encode the time of client creation into the stateid instead of the
time of server boot, then we can determine whether that stateid is from
a previous instance of the a server, or from a client that has expired,
and return an appropriate error to the client.

Signed-off-by: Bian Naimeng <biannm@cn.fujitsu.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4state.c | 62 ++++++++++++++++++++++++++++++++++++++++++-----------
 1 file changed, 49 insertions(+), 13 deletions(-)

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index c65a27b76a9d..74e822ec34cb 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -206,7 +206,7 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f
 	dp->dl_recall.cbr_dp = NULL;
 	dp->dl_recall.cbr_ident = cb->cb_ident;
 	dp->dl_recall.cbr_trunc = 0;
-	dp->dl_stateid.si_boot = boot_time;
+	dp->dl_stateid.si_boot = get_seconds();
 	dp->dl_stateid.si_stateownerid = current_delegid++;
 	dp->dl_stateid.si_fileid = 0;
 	dp->dl_stateid.si_generation = 0;
@@ -1883,7 +1883,7 @@ init_stateid(struct nfs4_stateid *stp, struct nfs4_file *fp, struct nfsd4_open *
 	stp->st_stateowner = sop;
 	get_nfs4_file(fp);
 	stp->st_file = fp;
-	stp->st_stateid.si_boot = boot_time;
+	stp->st_stateid.si_boot = get_seconds();
 	stp->st_stateid.si_stateownerid = sop->so_id;
 	stp->st_stateid.si_fileid = fp->fi_id;
 	stp->st_stateid.si_generation = 0;
@@ -2739,12 +2739,42 @@ nfs4_check_fh(struct svc_fh *fhp, struct nfs4_stateid *stp)
 static int
 STALE_STATEID(stateid_t *stateid)
 {
-	if (stateid->si_boot == boot_time)
-		return 0;
-	dprintk("NFSD: stale stateid (%08x/%08x/%08x/%08x)!\n",
-		stateid->si_boot, stateid->si_stateownerid, stateid->si_fileid,
-		stateid->si_generation);
-	return 1;
+	if (time_after((unsigned long)boot_time,
+			(unsigned long)stateid->si_boot)) {
+		dprintk("NFSD: stale stateid (%08x/%08x/%08x/%08x)!\n",
+			stateid->si_boot, stateid->si_stateownerid,
+			stateid->si_fileid, stateid->si_generation);
+		return 1;
+	}
+	return 0;
+}
+
+static int
+EXPIRED_STATEID(stateid_t *stateid)
+{
+	if (time_before((unsigned long)boot_time,
+			((unsigned long)stateid->si_boot)) &&
+	    time_before((stateid->si_boot + lease_time), get_seconds())) {
+		dprintk("NFSD: expired stateid (%08x/%08x/%08x/%08x)!\n",
+			stateid->si_boot, stateid->si_stateownerid,
+			stateid->si_fileid, stateid->si_generation);
+		return 1;
+	}
+	return 0;
+}
+
+static __be32
+stateid_error_map(stateid_t *stateid)
+{
+	if (STALE_STATEID(stateid))
+		return nfserr_stale_stateid;
+	if (EXPIRED_STATEID(stateid))
+		return nfserr_expired;
+
+	dprintk("NFSD: bad stateid (%08x/%08x/%08x/%08x)!\n",
+		stateid->si_boot, stateid->si_stateownerid,
+		stateid->si_fileid, stateid->si_generation);
+	return nfserr_bad_stateid;
 }
 
 static inline int
@@ -2868,8 +2898,10 @@ nfs4_preprocess_stateid_op(struct nfsd4_compound_state *cstate,
 	status = nfserr_bad_stateid;
 	if (is_delegation_stateid(stateid)) {
 		dp = find_delegation_stateid(ino, stateid);
-		if (!dp)
+		if (!dp) {
+			status = stateid_error_map(stateid);
 			goto out;
+		}
 		status = check_stateid_generation(stateid, &dp->dl_stateid,
 						  flags);
 		if (status)
@@ -2882,8 +2914,10 @@ nfs4_preprocess_stateid_op(struct nfsd4_compound_state *cstate,
 			*filpp = dp->dl_vfs_file;
 	} else { /* open or lock stateid */
 		stp = find_stateid(stateid, flags);
-		if (!stp)
+		if (!stp) {
+			status = stateid_error_map(stateid);
 			goto out;
+		}
 		if (nfs4_check_fh(current_fh, stp))
 			goto out;
 		if (!stp->st_stateowner->so_confirmed)
@@ -2957,7 +2991,7 @@ nfs4_preprocess_seqid_op(struct nfsd4_compound_state *cstate, u32 seqid,
 		 */
 		sop = search_close_lru(stateid->si_stateownerid, flags);
 		if (sop == NULL)
-			return nfserr_bad_stateid;
+			return stateid_error_map(stateid);
 		*sopp = sop;
 		goto check_replay;
 	}
@@ -3228,8 +3262,10 @@ nfsd4_delegreturn(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (!is_delegation_stateid(stateid))
 		goto out;
 	dp = find_delegation_stateid(inode, stateid);
-	if (!dp)
+	if (!dp) {
+		status = stateid_error_map(stateid);
 		goto out;
+	}
 	status = check_stateid_generation(stateid, &dp->dl_stateid, flags);
 	if (status)
 		goto out;
@@ -3456,7 +3492,7 @@ alloc_init_lock_stateid(struct nfs4_stateowner *sop, struct nfs4_file *fp, struc
 	stp->st_stateowner = sop;
 	get_nfs4_file(fp);
 	stp->st_file = fp;
-	stp->st_stateid.si_boot = boot_time;
+	stp->st_stateid.si_boot = get_seconds();
 	stp->st_stateid.si_stateownerid = sop->so_id;
 	stp->st_stateid.si_fileid = fp->fi_id;
 	stp->st_stateid.si_generation = 0;
-- 
cgit v1.2.3-59-g8ed1b


From 4ba170c2bb77713e999ac6fb9ffe6ddd99d2f25a Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Tue, 7 Apr 2009 21:45:37 +0300
Subject: update Documentation/filesystems/00-INDEX with new nfsd related docs.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Cc: James Lentini <jlentini@netapp.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 Documentation/filesystems/00-INDEX | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/filesystems/00-INDEX b/Documentation/filesystems/00-INDEX
index 8dd6db76171d..f15621ee5599 100644
--- a/Documentation/filesystems/00-INDEX
+++ b/Documentation/filesystems/00-INDEX
@@ -66,6 +66,10 @@ mandatory-locking.txt
 	- info on the Linux implementation of Sys V mandatory file locking.
 ncpfs.txt
 	- info on Novell Netware(tm) filesystem using NCP protocol.
+nfs41-server.txt
+	- info on the Linux server implementation of NFSv4 minor version 1.
+nfs-rdma.txt
+	- how to install and setup the Linux NFS/RDMA client and server software.
 nfsroot.txt
 	- short guide on setting up a diskless box with NFS root filesystem.
 nilfs2.txt
-- 
cgit v1.2.3-59-g8ed1b


From dcf1a3573eae69937fb14462369c4d3e6f4a37f1 Mon Sep 17 00:00:00 2001
From: H Hartley Sweeten <hartleys@visionengravers.com>
Date: Wed, 22 Apr 2009 20:18:19 -0400
Subject: net/sunrpc/svc_xprt.c: fix sparse warnings

Fix the following sparse warnings in net/sunrpc/svc_xprt.c.

  warning: symbol 'svc_recv' was not declared. Should it be static?
  warning: symbol 'svc_drop' was not declared. Should it be static?
  warning: symbol 'svc_send' was not declared. Should it be static?
  warning: symbol 'svc_close_all' was not declared. Should it be static?

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 net/sunrpc/svc_xprt.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c
index c200d92e57e4..f200393ac877 100644
--- a/net/sunrpc/svc_xprt.c
+++ b/net/sunrpc/svc_xprt.c
@@ -11,6 +11,7 @@
 #include <net/sock.h>
 #include <linux/sunrpc/stats.h>
 #include <linux/sunrpc/svc_xprt.h>
+#include <linux/sunrpc/svcsock.h>
 
 #define RPCDBG_FACILITY	RPCDBG_SVCXPRT
 
-- 
cgit v1.2.3-59-g8ed1b


From abc5c44d6284fab8fb21bcfc52c0f16f980637df Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:31:25 -0400
Subject: SUNRPC: Fix error return value of svc_addr_len()

The svc_addr_len() helper function returns -EAFNOSUPPORT if it doesn't
recognize the address family of the passed-in socket address.  However,
the return type of this function is size_t, which means -EAFNOSUPPORT
is turned into a very large positive value in this case.

The check in svc_udp_recvfrom() to see if the return value is less
than zero therefore won't work at all.

Additionally, handle_connect_req() passes this value directly to
memset().  This could cause memset() to clobber a large chunk of memory
if svc_addr_len() has returned an error.  Currently the address family
of these addresses, however, is known to be supported long before
handle_connect_req() is called, so this isn't a real risk.

Change the error return value of svc_addr_len() to zero, which fits in
the range of size_t, and is safer to pass to memset() directly.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 include/linux/sunrpc/svc_xprt.h | 5 +++--
 net/sunrpc/svcsock.c            | 7 ++++---
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h
index 0d9cb6ef28b0..d790c52525cc 100644
--- a/include/linux/sunrpc/svc_xprt.h
+++ b/include/linux/sunrpc/svc_xprt.h
@@ -118,7 +118,7 @@ static inline unsigned short svc_addr_port(const struct sockaddr *sa)
 	return 0;
 }
 
-static inline size_t svc_addr_len(struct sockaddr *sa)
+static inline size_t svc_addr_len(const struct sockaddr *sa)
 {
 	switch (sa->sa_family) {
 	case AF_INET:
@@ -126,7 +126,8 @@ static inline size_t svc_addr_len(struct sockaddr *sa)
 	case AF_INET6:
 		return sizeof(struct sockaddr_in6);
 	}
-	return -EAFNOSUPPORT;
+
+	return 0;
 }
 
 static inline unsigned short svc_xprt_local_port(const struct svc_xprt *xprt)
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index af3198814c15..8b0832834135 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -426,13 +426,14 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 		long		all[SVC_PKTINFO_SPACE / sizeof(long)];
 	} buffer;
 	struct cmsghdr *cmh = &buffer.hdr;
-	int		err, len;
 	struct msghdr msg = {
 		.msg_name = svc_addr(rqstp),
 		.msg_control = cmh,
 		.msg_controllen = sizeof(buffer),
 		.msg_flags = MSG_DONTWAIT,
 	};
+	size_t len;
+	int err;
 
 	if (test_and_clear_bit(XPT_CHNGBUF, &svsk->sk_xprt.xpt_flags))
 	    /* udp sockets need large rcvbuf as all pending
@@ -464,8 +465,8 @@ static int svc_udp_recvfrom(struct svc_rqst *rqstp)
 		return -EAGAIN;
 	}
 	len = svc_addr_len(svc_addr(rqstp));
-	if (len < 0)
-		return len;
+	if (len == 0)
+		return -EAFNOSUPPORT;
 	rqstp->rq_addrlen = len;
 	if (skb->tstamp.tv64 == 0) {
 		skb->tstamp = ktime_get_real();
-- 
cgit v1.2.3-59-g8ed1b


From 4cd5dc751a5889b5b37aa88752e33a58d8f6b624 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:31:32 -0400
Subject: NFSD: Refactor transport removal out of __write_ports()

Clean up: Refactor transport removal out of __write_ports() to make it
easier to understand and maintain.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c | 53 +++++++++++++++++++++++++++++------------------------
 1 file changed, 29 insertions(+), 24 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index af16849d243a..2c1dce8e9a06 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -910,6 +910,31 @@ static ssize_t write_versions(struct file *file, char *buf, size_t size)
 	return rv;
 }
 
+/*
+ * A transport listener is removed by writing a "-", it's transport
+ * name, and it's port number.
+ */
+static ssize_t __write_ports_delxprt(char *buf)
+{
+	struct svc_xprt *xprt;
+	char transport[16];
+	int port;
+
+	if (sscanf(&buf[1], "%15s %4u", transport, &port) != 2)
+		return -EINVAL;
+
+	if (port < 1 || port > USHORT_MAX || nfsd_serv == NULL)
+		return -EINVAL;
+
+	xprt = svc_find_xprt(nfsd_serv, transport, AF_UNSPEC, port);
+	if (xprt == NULL)
+		return -ENOTCONN;
+
+	svc_close_xprt(xprt);
+	svc_xprt_put(xprt);
+	return 0;
+}
+
 static ssize_t __write_ports(struct file *file, char *buf, size_t size)
 {
 	if (size == 0) {
@@ -984,30 +1009,10 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size)
 			return err < 0 ? err : 0;
 		}
 	}
-	/*
-	 * Remove a transport by writing it's transport name and port number
-	 */
-	if (buf[0] == '-' && isalpha(buf[1])) {
-		struct svc_xprt *xprt;
-		int err = -EINVAL;
-		char transport[16];
-		int port;
-		if (sscanf(&buf[1], "%15s %4d", transport, &port) == 2) {
-			if (port < 1 || port > 65535)
-				return -EINVAL;
-			if (nfsd_serv) {
-				xprt = svc_find_xprt(nfsd_serv, transport,
-						     AF_UNSPEC, port);
-				if (xprt) {
-					svc_close_xprt(xprt);
-					svc_xprt_put(xprt);
-					err = 0;
-				} else
-					err = -ENOTCONN;
-			}
-			return err < 0 ? err : 0;
-		}
-	}
+
+	if (buf[0] == '-' && isalpha(buf[1]))
+		return __write_ports_delxprt(buf);
+
 	return -EINVAL;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 4eb68c266cb1754ffa0040e882882680ece8cf34 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:31:40 -0400
Subject: NFSD: Refactor transport addition out of __write_ports()

Clean up: Refactor transport addition out of __write_ports() to make
it easier to understand and maintain.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c | 56 +++++++++++++++++++++++++++++++++-----------------------
 1 file changed, 33 insertions(+), 23 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 2c1dce8e9a06..748532b93fd9 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -910,6 +910,36 @@ static ssize_t write_versions(struct file *file, char *buf, size_t size)
 	return rv;
 }
 
+/*
+ * A transport listener is added by writing it's transport name and
+ * a port number.
+ */
+static ssize_t __write_ports_addxprt(char *buf)
+{
+	char transport[16];
+	int port, err;
+
+	if (sscanf(buf, "%15s %4u", transport, &port) != 2)
+		return -EINVAL;
+
+	if (port < 1 || port > USHORT_MAX)
+		return -EINVAL;
+
+	err = nfsd_create_serv();
+	if (err != 0)
+		return err;
+
+	err = svc_create_xprt(nfsd_serv, transport,
+				PF_INET, port, SVC_SOCK_ANONYMOUS);
+	if (err < 0) {
+		/* Give a reasonable perror msg for bad transport string */
+		if (err == -ENOENT)
+			err = -EPROTONOSUPPORT;
+		return err;
+	}
+	return 0;
+}
+
 /*
  * A transport listener is removed by writing a "-", it's transport
  * name, and it's port number.
@@ -986,29 +1016,9 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size)
 		kfree(toclose);
 		return len;
 	}
-	/*
-	 * Add a transport listener by writing it's transport name
-	 */
-	if (isalpha(buf[0])) {
-		int err;
-		char transport[16];
-		int port;
-		if (sscanf(buf, "%15s %4d", transport, &port) == 2) {
-			if (port < 1 || port > 65535)
-				return -EINVAL;
-			err = nfsd_create_serv();
-			if (!err) {
-				err = svc_create_xprt(nfsd_serv,
-						      transport, PF_INET, port,
-						      SVC_SOCK_ANONYMOUS);
-				if (err == -ENOENT)
-					/* Give a reasonable perror msg for
-					 * bad transport string */
-					err = -EPROTONOSUPPORT;
-			}
-			return err < 0 ? err : 0;
-		}
-	}
+
+	if (isalpha(buf[0]))
+		return __write_ports_addxprt(buf);
 
 	if (buf[0] == '-' && isalpha(buf[1]))
 		return __write_ports_delxprt(buf);
-- 
cgit v1.2.3-59-g8ed1b


From 82d565919aa4138bc481a7473491b71385af4018 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:31:48 -0400
Subject: NFSD: Refactor portlist socket closing into a helper

Clean up: Refactor the socket closing logic out of __write_ports() to
make it easier to understand and maintain.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c | 36 ++++++++++++++++++++++++------------
 1 file changed, 24 insertions(+), 12 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 748532b93fd9..fa268d1b775f 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -910,6 +910,27 @@ static ssize_t write_versions(struct file *file, char *buf, size_t size)
 	return rv;
 }
 
+/*
+ * A '-' followed by the 'name' of a socket means we close the socket.
+ */
+static ssize_t __write_ports_delfd(char *buf)
+{
+	char *toclose;
+	int len = 0;
+
+	toclose = kstrdup(buf + 1, GFP_KERNEL);
+	if (toclose == NULL)
+		return -ENOMEM;
+
+	if (nfsd_serv != NULL)
+		len = svc_sock_names(buf, nfsd_serv, toclose);
+	if (len >= 0)
+		lockd_down();
+
+	kfree(toclose);
+	return len;
+}
+
 /*
  * A transport listener is added by writing it's transport name and
  * a port number.
@@ -1004,18 +1025,9 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size)
 		}
 		return err < 0 ? err : 0;
 	}
-	if (buf[0] == '-' && isdigit(buf[1])) {
-		char *toclose = kstrdup(buf+1, GFP_KERNEL);
-		int len = 0;
-		if (!toclose)
-			return -ENOMEM;
-		if (nfsd_serv)
-			len = svc_sock_names(buf, nfsd_serv, toclose);
-		if (len >= 0)
-			lockd_down();
-		kfree(toclose);
-		return len;
-	}
+
+	if (buf[0] == '-' && isdigit(buf[1]))
+		return __write_ports_delfd(buf);
 
 	if (isalpha(buf[0]))
 		return __write_ports_addxprt(buf);
-- 
cgit v1.2.3-59-g8ed1b


From 0b7c2f6fc7e06867885c7b0f256ff5bd494e0653 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:31:55 -0400
Subject: NFSD: Refactor socket creation out of __write_ports()

Clean up: Refactor the socket creation logic out of __write_ports() to
make it easier to understand and maintain.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c | 64 ++++++++++++++++++++++++++++++--------------------------
 1 file changed, 34 insertions(+), 30 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index fa268d1b775f..b6a847faa0bb 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -910,6 +910,37 @@ static ssize_t write_versions(struct file *file, char *buf, size_t size)
 	return rv;
 }
 
+/*
+ * A single 'fd' number was written, in which case it must be for
+ * a socket of a supported family/protocol, and we use it as an
+ * nfsd listener.
+ */
+static ssize_t __write_ports_addfd(char *buf)
+{
+	char *mesg = buf;
+	int fd, err;
+
+	err = get_int(&mesg, &fd);
+	if (err != 0 || fd < 0)
+		return -EINVAL;
+
+	err = nfsd_create_serv();
+	if (err != 0)
+		return err;
+
+	err = svc_addsock(nfsd_serv, fd, buf);
+	if (err >= 0) {
+		err = lockd_up();
+		if (err < 0)
+			svc_sock_names(buf + strlen(buf) + 1, nfsd_serv, buf);
+
+		/* Decrease the count, but don't shut down the service */
+		nfsd_serv->sv_nrthreads--;
+	}
+
+	return err < 0 ? err : 0;
+}
+
 /*
  * A '-' followed by the 'name' of a socket means we close the socket.
  */
@@ -995,36 +1026,9 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size)
 			len = svc_xprt_names(nfsd_serv, buf, 0);
 		return len;
 	}
-	/* Either a single 'fd' number is written, in which
-	 * case it must be for a socket of a supported family/protocol,
-	 * and we use it as an nfsd socket, or
-	 * A '-' followed by the 'name' of a socket in which case
-	 * we close the socket.
-	 */
-	if (isdigit(buf[0])) {
-		char *mesg = buf;
-		int fd;
-		int err;
-		err = get_int(&mesg, &fd);
-		if (err)
-			return -EINVAL;
-		if (fd < 0)
-			return -EINVAL;
-		err = nfsd_create_serv();
-		if (!err) {
-			err = svc_addsock(nfsd_serv, fd, buf);
-			if (err >= 0) {
-				err = lockd_up();
-				if (err < 0)
-					svc_sock_names(buf+strlen(buf)+1, nfsd_serv, buf);
-			}
-			/* Decrease the count, but don't shutdown the
-			 * the service
-			 */
-			nfsd_serv->sv_nrthreads--;
-		}
-		return err < 0 ? err : 0;
-	}
+
+	if (isdigit(buf[0]))
+		return __write_ports_addfd(buf);
 
 	if (buf[0] == '-' && isdigit(buf[1]))
 		return __write_ports_delfd(buf);
-- 
cgit v1.2.3-59-g8ed1b


From c71206a7b4e829601e44c49d8a0f4df0fdfe9302 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:32:03 -0400
Subject: NFSD: Note an additional requirement when passing TCP sockets to
 portlist

User space must call listen(3) on SOCK_STREAM sockets passed into
/proc/fs/nfsd/portlist, otherwise that listener is ignored.  Document
this.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index b6a847faa0bb..d491fa9df9bd 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -1061,7 +1061,9 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size)
  *			buf:		C string containing an unsigned
  *					integer value representing a bound
  *					but unconnected socket that is to be
- *					used as an NFSD listener
+ *					used as an NFSD listener; listen(3)
+ *					must be called for a SOCK_STREAM
+ *					socket, otherwise it is ignored
  *			size:		non-zero length of C string in @buf
  * Output:
  *	On success:	NFS service is started;
-- 
cgit v1.2.3-59-g8ed1b


From 0a5372d8a13c3f670bf7c9c4a68ab04c3664fa28 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:32:10 -0400
Subject: NFSD: Finish refactoring __write_ports()

Clean up: Refactor transport name listing out of __write_ports() to
make it easier to understand and maintain.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index d491fa9df9bd..caf4fdc674c0 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -910,6 +910,17 @@ static ssize_t write_versions(struct file *file, char *buf, size_t size)
 	return rv;
 }
 
+/*
+ * Zero-length write.  Return a list of NFSD's current listener
+ * transports.
+ */
+static ssize_t __write_ports_names(char *buf)
+{
+	if (nfsd_serv == NULL)
+		return 0;
+	return svc_xprt_names(nfsd_serv, buf, 0);
+}
+
 /*
  * A single 'fd' number was written, in which case it must be for
  * a socket of a supported family/protocol, and we use it as an
@@ -1019,13 +1030,8 @@ static ssize_t __write_ports_delxprt(char *buf)
 
 static ssize_t __write_ports(struct file *file, char *buf, size_t size)
 {
-	if (size == 0) {
-		int len = 0;
-
-		if (nfsd_serv)
-			len = svc_xprt_names(nfsd_serv, buf, 0);
-		return len;
-	}
+	if (size == 0)
+		return __write_ports_names(buf);
 
 	if (isdigit(buf[0]))
 		return __write_ports_addfd(buf);
-- 
cgit v1.2.3-59-g8ed1b


From ea068bad27cefc71ab03230dbf01a8f8d98da5ba Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:32:18 -0400
Subject: NFSD: move lockd_up() before svc_addsock()

Clean up.

A couple of years ago, a series of commits, finishing with commit
5680c446, swapped the order of the lockd_up() and svc_addsock() calls
in __write_ports().  At that time lockd_up() needed to know the
transport protocol of the passed-in socket to start a listener on the
same transport protocol.

These days, lockd_up() doesn't take a protocol argument; it always
starts both a UDP and TCP listener.  It's now more straightforward to
try the lockd_up() first, then do a lockd_down() if the svc_addsock()
fails.

Careful review of this code shows that the svc_sock_names() call is
used only to close the just-opened socket in case lockd_up() fails.
So it is no longer needed if lockd_up() is done first.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c | 19 ++++++++++---------
 1 file changed, 10 insertions(+), 9 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index caf4fdc674c0..e051847b93fb 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -939,17 +939,18 @@ static ssize_t __write_ports_addfd(char *buf)
 	if (err != 0)
 		return err;
 
-	err = svc_addsock(nfsd_serv, fd, buf);
-	if (err >= 0) {
-		err = lockd_up();
-		if (err < 0)
-			svc_sock_names(buf + strlen(buf) + 1, nfsd_serv, buf);
+	err = lockd_up();
+	if (err != 0)
+		goto out;
 
-		/* Decrease the count, but don't shut down the service */
-		nfsd_serv->sv_nrthreads--;
-	}
+	err = svc_addsock(nfsd_serv, fd, buf);
+	if (err < 0)
+		lockd_down();
 
-	return err < 0 ? err : 0;
+out:
+	/* Decrease the count, but don't shut down the service */
+	nfsd_serv->sv_nrthreads--;
+	return err;
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 335c54bdc4d3bacdbd619ec95cd0b352435bd37f Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:32:25 -0400
Subject: NFSD: Prevent a buffer overflow in svc_xprt_names()

The svc_xprt_names() function can overflow its buffer if it's so near
the end of the passed in buffer that the "name too long" string still
doesn't fit.  Of course, it could never tell if it was near the end
of the passed in buffer, since its only caller passes in zero as the
buffer length.

Let's make this API a little safer.

Change svc_xprt_names() so it *always* checks for a buffer overflow,
and change its only caller to pass in the correct buffer length.

If svc_xprt_names() does overflow its buffer, it now fails with an
ENAMETOOLONG errno, instead of trying to write a message at the end
of the buffer.  I don't like this much, but I can't figure out a clean
way that's always safe to return some of the names, *and* an
indication that the buffer was not long enough.

The displayed error when doing a 'cat /proc/fs/nfsd/portlist' is
"File name too long".

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c                |  2 +-
 include/linux/sunrpc/svc_xprt.h |  2 +-
 net/sunrpc/svc_xprt.c           | 56 ++++++++++++++++++++++++++++-------------
 3 files changed, 41 insertions(+), 19 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index e051847b93fb..6a1cd908e6bc 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -918,7 +918,7 @@ static ssize_t __write_ports_names(char *buf)
 {
 	if (nfsd_serv == NULL)
 		return 0;
-	return svc_xprt_names(nfsd_serv, buf, 0);
+	return svc_xprt_names(nfsd_serv, buf, SIMPLE_TRANSACTION_LIMIT);
 }
 
 /*
diff --git a/include/linux/sunrpc/svc_xprt.h b/include/linux/sunrpc/svc_xprt.h
index d790c52525cc..2223ae0b5ed5 100644
--- a/include/linux/sunrpc/svc_xprt.h
+++ b/include/linux/sunrpc/svc_xprt.h
@@ -83,7 +83,7 @@ int	svc_port_is_privileged(struct sockaddr *sin);
 int	svc_print_xprts(char *buf, int maxlen);
 struct	svc_xprt *svc_find_xprt(struct svc_serv *serv, const char *xcl_name,
 			const sa_family_t af, const unsigned short port);
-int	svc_xprt_names(struct svc_serv *serv, char *buf, int buflen);
+int	svc_xprt_names(struct svc_serv *serv, char *buf, const int buflen);
 
 static inline void svc_xprt_get(struct svc_xprt *xprt)
 {
diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c
index f200393ac877..6f33d33cc064 100644
--- a/net/sunrpc/svc_xprt.c
+++ b/net/sunrpc/svc_xprt.c
@@ -1098,36 +1098,58 @@ struct svc_xprt *svc_find_xprt(struct svc_serv *serv, const char *xcl_name,
 }
 EXPORT_SYMBOL_GPL(svc_find_xprt);
 
-/*
- * Format a buffer with a list of the active transports. A zero for
- * the buflen parameter disables target buffer overflow checking.
+static int svc_one_xprt_name(const struct svc_xprt *xprt,
+			     char *pos, int remaining)
+{
+	int len;
+
+	len = snprintf(pos, remaining, "%s %u\n",
+			xprt->xpt_class->xcl_name,
+			svc_xprt_local_port(xprt));
+	if (len >= remaining)
+		return -ENAMETOOLONG;
+	return len;
+}
+
+/**
+ * svc_xprt_names - format a buffer with a list of transport names
+ * @serv: pointer to an RPC service
+ * @buf: pointer to a buffer to be filled in
+ * @buflen: length of buffer to be filled in
+ *
+ * Fills in @buf with a string containing a list of transport names,
+ * each name terminated with '\n'.
+ *
+ * Returns positive length of the filled-in string on success; otherwise
+ * a negative errno value is returned if an error occurs.
  */
-int svc_xprt_names(struct svc_serv *serv, char *buf, int buflen)
+int svc_xprt_names(struct svc_serv *serv, char *buf, const int buflen)
 {
 	struct svc_xprt *xprt;
-	char xprt_str[64];
-	int totlen = 0;
-	int len;
+	int len, totlen;
+	char *pos;
 
 	/* Sanity check args */
 	if (!serv)
 		return 0;
 
 	spin_lock_bh(&serv->sv_lock);
+
+	pos = buf;
+	totlen = 0;
 	list_for_each_entry(xprt, &serv->sv_permsocks, xpt_list) {
-		len = snprintf(xprt_str, sizeof(xprt_str),
-			       "%s %d\n", xprt->xpt_class->xcl_name,
-			       svc_xprt_local_port(xprt));
-		/* If the string was truncated, replace with error string */
-		if (len >= sizeof(xprt_str))
-			strcpy(xprt_str, "name-too-long\n");
-		/* Don't overflow buffer */
-		len = strlen(xprt_str);
-		if (buflen && (len + totlen >= buflen))
+		len = svc_one_xprt_name(xprt, pos, buflen - totlen);
+		if (len < 0) {
+			*buf = '\0';
+			totlen = len;
+		}
+		if (len <= 0)
 			break;
-		strcpy(buf+totlen, xprt_str);
+
+		pos += len;
 		totlen += len;
 	}
+
 	spin_unlock_bh(&serv->sv_lock);
 	return totlen;
 }
-- 
cgit v1.2.3-59-g8ed1b


From bfba9ab4c64f0e5c33930711e6c073c285e01fcf Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:32:33 -0400
Subject: SUNRPC: pass buffer size to svc_addsock()

Adjust the synopsis of svc_addsock() to pass in the size of the output
buffer.  Add a documenting comment.

This is a cosmetic change for now.  A subsequent patch will make sure
the buffer length is passed to one_sock_name(), where the length will
actually be useful.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c               |  2 +-
 include/linux/sunrpc/svcsock.h |  3 ++-
 net/sunrpc/svcsock.c           | 16 +++++++++++++---
 3 files changed, 16 insertions(+), 5 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 6a1cd908e6bc..1f1c2159b802 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -943,7 +943,7 @@ static ssize_t __write_ports_addfd(char *buf)
 	if (err != 0)
 		goto out;
 
-	err = svc_addsock(nfsd_serv, fd, buf);
+	err = svc_addsock(nfsd_serv, fd, buf, SIMPLE_TRANSACTION_LIMIT);
 	if (err < 0)
 		lockd_down();
 
diff --git a/include/linux/sunrpc/svcsock.h b/include/linux/sunrpc/svcsock.h
index 483e10380aae..e23241c53f42 100644
--- a/include/linux/sunrpc/svcsock.h
+++ b/include/linux/sunrpc/svcsock.h
@@ -39,7 +39,8 @@ int		svc_send(struct svc_rqst *);
 void		svc_drop(struct svc_rqst *);
 void		svc_sock_update_bufs(struct svc_serv *serv);
 int		svc_sock_names(char *buf, struct svc_serv *serv, char *toclose);
-int		svc_addsock(struct svc_serv *serv, int fd, char *name_return);
+int		svc_addsock(struct svc_serv *serv, const int fd,
+					char *name_return, const size_t len);
 void		svc_init_xprt_sock(void);
 void		svc_cleanup_xprt_sock(void);
 
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 8b0832834135..6bec1e25b542 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -1128,9 +1128,19 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv,
 	return svsk;
 }
 
-int svc_addsock(struct svc_serv *serv,
-		int fd,
-		char *name_return)
+/**
+ * svc_addsock - add a listener socket to an RPC service
+ * @serv: pointer to RPC service to which to add a new listener
+ * @fd: file descriptor of the new listener
+ * @name_return: pointer to buffer to fill in with name of listener
+ * @len: size of the buffer
+ *
+ * Fills in socket name and returns positive length of name if successful.
+ * Name is terminated with '\n'.  On error, returns a negative errno
+ * value.
+ */
+int svc_addsock(struct svc_serv *serv, const int fd, char *name_return,
+		const size_t len)
 {
 	int err = 0;
 	struct socket *so = sockfd_lookup(fd, &err);
-- 
cgit v1.2.3-59-g8ed1b


From 8435d34dbbe75678c3cdad3d53b1e7996a79b3bf Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:32:40 -0400
Subject: SUNRPC: pass buffer size to svc_sock_names()

Adjust the synopsis of svc_sock_names() to pass in the size of the
output buffer.  Add a documenting comment.

This is a cosmetic change for now.  A subsequent patch will make sure
the buffer length is passed to one_sock_name(), where the length will
actually be useful.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c               |  3 ++-
 include/linux/sunrpc/svcsock.h |  4 +++-
 net/sunrpc/svcsock.c           | 19 +++++++++++++++++--
 3 files changed, 22 insertions(+), 4 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 1f1c2159b802..b64a7fbfccf5 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -966,7 +966,8 @@ static ssize_t __write_ports_delfd(char *buf)
 		return -ENOMEM;
 
 	if (nfsd_serv != NULL)
-		len = svc_sock_names(buf, nfsd_serv, toclose);
+		len = svc_sock_names(nfsd_serv, buf,
+					SIMPLE_TRANSACTION_LIMIT, toclose);
 	if (len >= 0)
 		lockd_down();
 
diff --git a/include/linux/sunrpc/svcsock.h b/include/linux/sunrpc/svcsock.h
index e23241c53f42..827163138949 100644
--- a/include/linux/sunrpc/svcsock.h
+++ b/include/linux/sunrpc/svcsock.h
@@ -38,7 +38,9 @@ int		svc_recv(struct svc_rqst *, long);
 int		svc_send(struct svc_rqst *);
 void		svc_drop(struct svc_rqst *);
 void		svc_sock_update_bufs(struct svc_serv *serv);
-int		svc_sock_names(char *buf, struct svc_serv *serv, char *toclose);
+int		svc_sock_names(struct svc_serv *serv, char *buf,
+					const size_t buflen,
+					const char *toclose);
 int		svc_addsock(struct svc_serv *serv, const int fd,
 					char *name_return, const size_t len);
 void		svc_init_xprt_sock(void);
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 6bec1e25b542..032b52ea9541 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -259,8 +259,23 @@ static int one_sock_name(char *buf, struct svc_sock *svsk)
 	return len;
 }
 
-int
-svc_sock_names(char *buf, struct svc_serv *serv, char *toclose)
+/**
+ * svc_sock_names - construct a list of listener names in a string
+ * @serv: pointer to RPC service
+ * @buf: pointer to a buffer to fill in with socket names
+ * @buflen: size of the buffer to be filled
+ * @toclose: pointer to '\0'-terminated C string containing the name
+ *		of a listener to be closed
+ *
+ * Fills in @buf with a '\n'-separated list of names of listener
+ * sockets.  If @toclose is not NULL, the socket named by @toclose
+ * is closed, and is not included in the output list.
+ *
+ * Returns positive length of the socket name string, or a negative
+ * errno value on error.
+ */
+int svc_sock_names(struct svc_serv *serv, char *buf, const size_t buflen,
+		   const char *toclose)
 {
 	struct svc_sock *svsk, *closesk = NULL;
 	int len = 0;
-- 
cgit v1.2.3-59-g8ed1b


From e7942b9f2586fb15e1b898acc7c198ffd60aee4e Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:32:48 -0400
Subject: SUNRPC: Switch one_sock_name() to use snprintf()

Use snprintf() in one_sock_name() to prevent overflowing the output
buffer.  If the name doesn't fit in the buffer, the buffer is filled
in with an empty string, and -ENAMETOOLONG is returned.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 net/sunrpc/svcsock.c | 25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)

diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 032b52ea9541..61d4a3281f94 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -240,22 +240,27 @@ out:
 /*
  * Report socket names for nfsdfs
  */
-static int one_sock_name(char *buf, struct svc_sock *svsk)
+static int svc_one_sock_name(struct svc_sock *svsk, char *buf, int remaining)
 {
 	int len;
 
 	switch(svsk->sk_sk->sk_family) {
-	case AF_INET:
-		len = sprintf(buf, "ipv4 %s %pI4 %d\n",
+	case PF_INET:
+		len = snprintf(buf, remaining, "ipv4 %s %pI4 %d\n",
 			      svsk->sk_sk->sk_protocol == IPPROTO_UDP ?
 			      "udp" : "tcp",
 			      &inet_sk(svsk->sk_sk)->rcv_saddr,
 			      inet_sk(svsk->sk_sk)->num);
 		break;
 	default:
-		len = sprintf(buf, "*unknown-%d*\n",
+		len = snprintf(buf, remaining, "*unknown-%d*\n",
 			       svsk->sk_sk->sk_family);
 	}
+
+	if (len >= remaining) {
+		*buf = '\0';
+		return -ENAMETOOLONG;
+	}
 	return len;
 }
 
@@ -282,15 +287,21 @@ int svc_sock_names(struct svc_serv *serv, char *buf, const size_t buflen,
 
 	if (!serv)
 		return 0;
+
 	spin_lock_bh(&serv->sv_lock);
 	list_for_each_entry(svsk, &serv->sv_permsocks, sk_xprt.xpt_list) {
-		int onelen = one_sock_name(buf+len, svsk);
-		if (toclose && strcmp(toclose, buf+len) == 0)
+		int onelen = svc_one_sock_name(svsk, buf + len, buflen - len);
+		if (onelen < 0) {
+			len = onelen;
+			break;
+		}
+		if (toclose && strcmp(toclose, buf + len) == 0)
 			closesk = svsk;
 		else
 			len += onelen;
 	}
 	spin_unlock_bh(&serv->sv_lock);
+
 	if (closesk)
 		/* Should unregister with portmap, but you cannot
 		 * unregister just one protocol...
@@ -1195,7 +1206,7 @@ int svc_addsock(struct svc_serv *serv, const int fd, char *name_return,
 		sockfd_put(so);
 		return err;
 	}
-	return one_sock_name(name_return, svsk);
+	return svc_one_sock_name(svsk, name_return, len);
 }
 EXPORT_SYMBOL_GPL(svc_addsock);
 
-- 
cgit v1.2.3-59-g8ed1b


From 58de2f86585dd8fc785aca6625adee32af84b8d7 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:32:55 -0400
Subject: SUNRPC: Support PF_INET6 in one_sock_name()

Add an arm to the switch statement in svc_one_sock_name() so it can
construct the name of PF_INET6 sockets properly.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Cc: Aime Le Rouzic <aime.le-rouzic@bull.net>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 net/sunrpc/svcsock.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 61d4a3281f94..983bfa9f3102 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -252,6 +252,13 @@ static int svc_one_sock_name(struct svc_sock *svsk, char *buf, int remaining)
 			      &inet_sk(svsk->sk_sk)->rcv_saddr,
 			      inet_sk(svsk->sk_sk)->num);
 		break;
+	case PF_INET6:
+		len = snprintf(buf, remaining, "ipv6 %s %pI6 %d\n",
+				svsk->sk_sk->sk_protocol == IPPROTO_UDP ?
+				"udp" : "tcp",
+				&inet6_sk(svsk->sk_sk)->rcv_saddr,
+				inet_sk(svsk->sk_sk)->num);
+		break;
 	default:
 		len = snprintf(buf, remaining, "*unknown-%d*\n",
 			       svsk->sk_sk->sk_family);
-- 
cgit v1.2.3-59-g8ed1b


From 017cb47f46722f29d101a709a2094d151111ed6d Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:33:03 -0400
Subject: SUNRPC: Clean up one_sock_name()

Clean up svc_one_sock_name() by setting up automatic variables for
frequently used expressions.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 net/sunrpc/svcsock.c | 21 +++++++++++----------
 1 file changed, 11 insertions(+), 10 deletions(-)

diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 983bfa9f3102..4e6d406264a0 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -242,26 +242,27 @@ out:
  */
 static int svc_one_sock_name(struct svc_sock *svsk, char *buf, int remaining)
 {
+	const struct sock *sk = svsk->sk_sk;
+	const char *proto_name = sk->sk_protocol == IPPROTO_UDP ?
+							"udp" : "tcp";
 	int len;
 
-	switch(svsk->sk_sk->sk_family) {
+	switch (sk->sk_family) {
 	case PF_INET:
 		len = snprintf(buf, remaining, "ipv4 %s %pI4 %d\n",
-			      svsk->sk_sk->sk_protocol == IPPROTO_UDP ?
-			      "udp" : "tcp",
-			      &inet_sk(svsk->sk_sk)->rcv_saddr,
-			      inet_sk(svsk->sk_sk)->num);
+				proto_name,
+				&inet_sk(sk)->rcv_saddr,
+				inet_sk(sk)->num);
 		break;
 	case PF_INET6:
 		len = snprintf(buf, remaining, "ipv6 %s %pI6 %d\n",
-				svsk->sk_sk->sk_protocol == IPPROTO_UDP ?
-				"udp" : "tcp",
-				&inet6_sk(svsk->sk_sk)->rcv_saddr,
-				inet_sk(svsk->sk_sk)->num);
+				proto_name,
+				&inet6_sk(sk)->rcv_saddr,
+				inet_sk(sk)->num);
 		break;
 	default:
 		len = snprintf(buf, remaining, "*unknown-%d*\n",
-			       svsk->sk_sk->sk_family);
+				sk->sk_family);
 	}
 
 	if (len >= remaining) {
-- 
cgit v1.2.3-59-g8ed1b


From 3d72ab8fdd44c872633b210dd1a4afd2910d0bbb Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:33:10 -0400
Subject: NFSD: Stricter buffer size checking in write_recoverydir()

While it's not likely a pathname will be longer than
SIMPLE_TRANSACTION_SIZE, we should be more careful about just
plopping it into the output buffer without bounds checking.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index b64a7fbfccf5..c4843467cfd4 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -1260,8 +1260,9 @@ static ssize_t __write_recoverydir(struct file *file, char *buf, size_t size)
 
 		status = nfs4_reset_recoverydir(recdir);
 	}
-	sprintf(buf, "%s\n", nfs4_recoverydir());
-	return strlen(buf);
+
+	return scnprintf(buf, SIMPLE_TRANSACTION_LIMIT, "%s\n",
+							nfs4_recoverydir());
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 261758b5c3dfeac73ca364c47ed538f5ce4250ee Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:33:18 -0400
Subject: NFSD: Stricter buffer size checking in write_versions()

While it's not likely today that there are enough NFS versions to
overflow the output buffer in write_versions(), we should be more
careful about detecting the end of the buffer.

The number of NFS versions will only increase as NFSv4 minor versions
are added.

Note that this API doesn't behave the same as portlist.  Here we
attempt to display as many versions as will fit in the buffer, and do
not provide any indication that an overflow would have occurred.  I
don't have any good rationale for that.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c | 32 +++++++++++++++++++++++++-------
 1 file changed, 25 insertions(+), 7 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index c4843467cfd4..a152694e016e 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -793,7 +793,7 @@ static ssize_t __write_versions(struct file *file, char *buf, size_t size)
 {
 	char *mesg = buf;
 	char *vers, *minorp, sign;
-	int len, num;
+	int len, num, remaining;
 	unsigned minor;
 	ssize_t tlen = 0;
 	char *sep;
@@ -840,32 +840,50 @@ static ssize_t __write_versions(struct file *file, char *buf, size_t size)
 			}
 		next:
 			vers += len + 1;
-			tlen += len;
 		} while ((len = qword_get(&mesg, vers, size)) > 0);
 		/* If all get turned off, turn them back on, as
 		 * having no versions is BAD
 		 */
 		nfsd_reset_versions();
 	}
+
 	/* Now write current state into reply buffer */
 	len = 0;
 	sep = "";
+	remaining = SIMPLE_TRANSACTION_LIMIT;
 	for (num=2 ; num <= 4 ; num++)
 		if (nfsd_vers(num, NFSD_AVAIL)) {
-			len += sprintf(buf+len, "%s%c%d", sep,
+			len = snprintf(buf, remaining, "%s%c%d", sep,
 				       nfsd_vers(num, NFSD_TEST)?'+':'-',
 				       num);
 			sep = " ";
+
+			if (len > remaining)
+				break;
+			remaining -= len;
+			buf += len;
+			tlen += len;
 		}
 	if (nfsd_vers(4, NFSD_AVAIL))
-		for (minor = 1; minor <= NFSD_SUPPORTED_MINOR_VERSION; minor++)
-			len += sprintf(buf+len, " %c4.%u",
+		for (minor = 1; minor <= NFSD_SUPPORTED_MINOR_VERSION;
+		     minor++) {
+			len = snprintf(buf, remaining, " %c4.%u",
 					(nfsd_vers(4, NFSD_TEST) &&
 					 nfsd_minorversion(minor, NFSD_TEST)) ?
 						'+' : '-',
 					minor);
-	len += sprintf(buf+len, "\n");
-	return len;
+
+			if (len > remaining)
+				break;
+			remaining -= len;
+			buf += len;
+			tlen += len;
+		}
+
+	len = snprintf(buf, remaining, "\n");
+	if (len > remaining)
+		return -EINVAL;
+	return tlen + len;
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From e06b64050ebea1220bca3c291d0431f5206b5c95 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Thu, 23 Apr 2009 19:33:25 -0400
Subject: NFSD: Stricter buffer size checking in fs/nfsd/nfsctl.c

Clean up: For consistency, handle output buffer size checking in a
other nfsctl functions the same way it's done for write_versions().

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsctl.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index a152694e016e..877e713a0fd6 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -695,8 +695,9 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size)
 		if (rv)
 			return rv;
 	}
-	sprintf(buf, "%d\n", nfsd_nrthreads());
-	return strlen(buf);
+
+	return scnprintf(buf, SIMPLE_TRANSACTION_LIMIT, "%d\n",
+							nfsd_nrthreads());
 }
 
 /**
@@ -1197,7 +1198,9 @@ static ssize_t write_maxblksize(struct file *file, char *buf, size_t size)
 		nfsd_max_blksize = bsize;
 		mutex_unlock(&nfsd_mutex);
 	}
-	return sprintf(buf, "%d\n", nfsd_max_blksize);
+
+	return scnprintf(buf, SIMPLE_TRANSACTION_LIMIT, "%d\n",
+							nfsd_max_blksize);
 }
 
 #ifdef CONFIG_NFSD_V4
@@ -1221,8 +1224,9 @@ static ssize_t __write_leasetime(struct file *file, char *buf, size_t size)
 			return -EINVAL;
 		nfs4_reset_lease(lease);
 	}
-	sprintf(buf, "%ld\n", nfs4_lease_time());
-	return strlen(buf);
+
+	return scnprintf(buf, SIMPLE_TRANSACTION_LIMIT, "%ld\n",
+							nfs4_lease_time());
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 4ed0d3e6c64cfd9ba4ceb2099b10d1cf8ece4320 Mon Sep 17 00:00:00 2001
From: Fenghua Yu <fenghua.yu@intel.com>
Date: Fri, 24 Apr 2009 17:30:20 -0700
Subject: Intel IOMMU Pass Through Support

The patch adds kernel parameter intel_iommu=pt to set up pass through
mode in context mapping entry. This disables DMAR in linux kernel; but
KVM still runs on VT-d and interrupt remapping still works.

In this mode, kernel uses swiotlb for DMA API functions but other VT-d
functionalities are enabled for KVM. KVM always uses multi level
translation page table in VT-d. By default, pass though mode is disabled
in kernel.

This is useful when people don't want to enable VT-d DMAR in kernel but
still want to use KVM and interrupt remapping for reasons like DMAR
performance concern or debug purpose.

Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Acked-by: Weidong Han <weidong@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 Documentation/kernel-parameters.txt |   1 +
 arch/ia64/include/asm/iommu.h       |   1 +
 arch/ia64/kernel/pci-swiotlb.c      |   2 +-
 arch/x86/include/asm/iommu.h        |   1 +
 arch/x86/kernel/pci-dma.c           |   6 ++
 arch/x86/kernel/pci-swiotlb.c       |   3 +-
 drivers/pci/dmar.c                  |  11 ++-
 drivers/pci/intel-iommu.c           | 180 ++++++++++++++++++++++++++----------
 include/linux/dma_remapping.h       |   8 ++
 include/linux/intel-iommu.h         |   2 +
 10 files changed, 165 insertions(+), 50 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 600cdd72900c..fa4faeb7597f 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -965,6 +965,7 @@ and is between 256 and 4096 characters. It is defined in the file
 		nomerge
 		forcesac
 		soft
+		pt	[x86, IA64]
 
 	io7=		[HW] IO7 for Marvel based alpha systems
 			See comment before marvel_specify_io7 in
diff --git a/arch/ia64/include/asm/iommu.h b/arch/ia64/include/asm/iommu.h
index 0490794fe4aa..37d41ca5645a 100644
--- a/arch/ia64/include/asm/iommu.h
+++ b/arch/ia64/include/asm/iommu.h
@@ -9,6 +9,7 @@ extern void pci_iommu_shutdown(void);
 extern void no_iommu_init(void);
 extern int force_iommu, no_iommu;
 extern int iommu_detected;
+extern int iommu_pass_through;
 extern void iommu_dma_init(void);
 extern void machvec_init(const char *name);
 
diff --git a/arch/ia64/kernel/pci-swiotlb.c b/arch/ia64/kernel/pci-swiotlb.c
index 285aae8431c6..223abb134105 100644
--- a/arch/ia64/kernel/pci-swiotlb.c
+++ b/arch/ia64/kernel/pci-swiotlb.c
@@ -46,7 +46,7 @@ void __init swiotlb_dma_init(void)
 
 void __init pci_swiotlb_init(void)
 {
-	if (!iommu_detected) {
+	if (!iommu_detected || iommu_pass_through) {
 #ifdef CONFIG_IA64_GENERIC
 		swiotlb = 1;
 		printk(KERN_INFO "PCI-DMA: Re-initialize machine vector.\n");
diff --git a/arch/x86/include/asm/iommu.h b/arch/x86/include/asm/iommu.h
index af326a2975b5..fd6d21bbee6c 100644
--- a/arch/x86/include/asm/iommu.h
+++ b/arch/x86/include/asm/iommu.h
@@ -6,6 +6,7 @@ extern void no_iommu_init(void);
 extern struct dma_map_ops nommu_dma_ops;
 extern int force_iommu, no_iommu;
 extern int iommu_detected;
+extern int iommu_pass_through;
 
 /* 10 seconds */
 #define DMAR_OPERATION_TIMEOUT ((cycles_t) tsc_khz*10*1000)
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 745579bc8256..8cad0d854242 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -160,6 +160,8 @@ again:
 	return page_address(page);
 }
 
+extern int iommu_pass_through;
+
 /*
  * See <Documentation/x86_64/boot-options.txt> for the iommu kernel parameter
  * documentation.
@@ -209,6 +211,10 @@ static __init int iommu_setup(char *p)
 #ifdef CONFIG_SWIOTLB
 		if (!strncmp(p, "soft", 4))
 			swiotlb = 1;
+		if (!strncmp(p, "pt", 2)) {
+			iommu_pass_through = 1;
+			return 1;
+		}
 #endif
 
 		gart_parse_options(p);
diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
index 221a3853e268..3a0c51e0ba6d 100644
--- a/arch/x86/kernel/pci-swiotlb.c
+++ b/arch/x86/kernel/pci-swiotlb.c
@@ -71,7 +71,8 @@ void __init pci_swiotlb_init(void)
 {
 	/* don't initialize swiotlb if iommu=off (no_iommu=1) */
 #ifdef CONFIG_X86_64
-	if (!iommu_detected && !no_iommu && max_pfn > MAX_DMA32_PFN)
+	if ((!iommu_detected && !no_iommu && max_pfn > MAX_DMA32_PFN) ||
+		iommu_pass_through)
 	       swiotlb = 1;
 #endif
 	if (swiotlb_force)
diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index fa3a11365ec3..d3d86b749eee 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -515,6 +515,7 @@ int alloc_iommu(struct dmar_drhd_unit *drhd)
 	u32 ver;
 	static int iommu_allocated = 0;
 	int agaw = 0;
+	int msagaw = 0;
 
 	iommu = kzalloc(sizeof(*iommu), GFP_KERNEL);
 	if (!iommu)
@@ -535,12 +536,20 @@ int alloc_iommu(struct dmar_drhd_unit *drhd)
 	agaw = iommu_calculate_agaw(iommu);
 	if (agaw < 0) {
 		printk(KERN_ERR
-			"Cannot get a valid agaw for iommu (seq_id = %d)\n",
+		       "Cannot get a valid agaw for iommu (seq_id = %d)\n",
+		       iommu->seq_id);
+		goto error;
+	}
+	msagaw = iommu_calculate_max_sagaw(iommu);
+	if (msagaw < 0) {
+		printk(KERN_ERR
+			"Cannot get a valid max agaw for iommu (seq_id = %d)\n",
 			iommu->seq_id);
 		goto error;
 	}
 #endif
 	iommu->agaw = agaw;
+	iommu->msagaw = msagaw;
 
 	/* the registers might be more than one page */
 	map_size = max_t(int, ecap_max_iotlb_offset(iommu->ecap),
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 001b328adf80..13121821db7f 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -53,6 +53,8 @@
 
 #define DEFAULT_DOMAIN_ADDRESS_WIDTH 48
 
+#define MAX_AGAW_WIDTH 64
+
 #define DOMAIN_MAX_ADDR(gaw) ((((u64)1) << gaw) - 1)
 
 #define IOVA_PFN(addr)		((addr) >> PAGE_SHIFT)
@@ -127,8 +129,6 @@ static inline void context_set_fault_enable(struct context_entry *context)
 	context->lo &= (((u64)-1) << 2) | 1;
 }
 
-#define CONTEXT_TT_MULTI_LEVEL 0
-
 static inline void context_set_translation_type(struct context_entry *context,
 						unsigned long value)
 {
@@ -288,6 +288,7 @@ int dmar_disabled = 1;
 static int __initdata dmar_map_gfx = 1;
 static int dmar_forcedac;
 static int intel_iommu_strict;
+int iommu_pass_through;
 
 #define DUMMY_DEVICE_DOMAIN_INFO ((struct device_domain_info *)(-1))
 static DEFINE_SPINLOCK(device_domain_lock);
@@ -397,17 +398,13 @@ void free_iova_mem(struct iova *iova)
 
 static inline int width_to_agaw(int width);
 
-/* calculate agaw for each iommu.
- * "SAGAW" may be different across iommus, use a default agaw, and
- * get a supported less agaw for iommus that don't support the default agaw.
- */
-int iommu_calculate_agaw(struct intel_iommu *iommu)
+static int __iommu_calculate_agaw(struct intel_iommu *iommu, int max_gaw)
 {
 	unsigned long sagaw;
 	int agaw = -1;
 
 	sagaw = cap_sagaw(iommu->cap);
-	for (agaw = width_to_agaw(DEFAULT_DOMAIN_ADDRESS_WIDTH);
+	for (agaw = width_to_agaw(max_gaw);
 	     agaw >= 0; agaw--) {
 		if (test_bit(agaw, &sagaw))
 			break;
@@ -416,6 +413,24 @@ int iommu_calculate_agaw(struct intel_iommu *iommu)
 	return agaw;
 }
 
+/*
+ * Calculate max SAGAW for each iommu.
+ */
+int iommu_calculate_max_sagaw(struct intel_iommu *iommu)
+{
+	return __iommu_calculate_agaw(iommu, MAX_AGAW_WIDTH);
+}
+
+/*
+ * calculate agaw for each iommu.
+ * "SAGAW" may be different across iommus, use a default agaw, and
+ * get a supported less agaw for iommus that don't support the default agaw.
+ */
+int iommu_calculate_agaw(struct intel_iommu *iommu)
+{
+	return __iommu_calculate_agaw(iommu, DEFAULT_DOMAIN_ADDRESS_WIDTH);
+}
+
 /* in native case, each domain is related to only one iommu */
 static struct intel_iommu *domain_get_iommu(struct dmar_domain *domain)
 {
@@ -1321,8 +1336,8 @@ static void domain_exit(struct dmar_domain *domain)
 	free_domain_mem(domain);
 }
 
-static int domain_context_mapping_one(struct dmar_domain *domain,
-				      int segment, u8 bus, u8 devfn)
+static int domain_context_mapping_one(struct dmar_domain *domain, int segment,
+				 u8 bus, u8 devfn, int translation)
 {
 	struct context_entry *context;
 	unsigned long flags;
@@ -1335,7 +1350,10 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
 
 	pr_debug("Set context mapping for %02x:%02x.%d\n",
 		bus, PCI_SLOT(devfn), PCI_FUNC(devfn));
+
 	BUG_ON(!domain->pgd);
+	BUG_ON(translation != CONTEXT_TT_PASS_THROUGH &&
+	       translation != CONTEXT_TT_MULTI_LEVEL);
 
 	iommu = device_to_iommu(segment, bus, devfn);
 	if (!iommu)
@@ -1395,9 +1413,18 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
 	}
 
 	context_set_domain_id(context, id);
-	context_set_address_width(context, iommu->agaw);
-	context_set_address_root(context, virt_to_phys(pgd));
-	context_set_translation_type(context, CONTEXT_TT_MULTI_LEVEL);
+
+	/*
+	 * In pass through mode, AW must be programmed to indicate the largest
+	 * AGAW value supported by hardware. And ASR is ignored by hardware.
+	 */
+	if (likely(translation == CONTEXT_TT_MULTI_LEVEL)) {
+		context_set_address_width(context, iommu->agaw);
+		context_set_address_root(context, virt_to_phys(pgd));
+	} else
+		context_set_address_width(context, iommu->msagaw);
+
+	context_set_translation_type(context, translation);
 	context_set_fault_enable(context);
 	context_set_present(context);
 	domain_flush_cache(domain, context, sizeof(*context));
@@ -1422,13 +1449,15 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
 }
 
 static int
-domain_context_mapping(struct dmar_domain *domain, struct pci_dev *pdev)
+domain_context_mapping(struct dmar_domain *domain, struct pci_dev *pdev,
+			int translation)
 {
 	int ret;
 	struct pci_dev *tmp, *parent;
 
 	ret = domain_context_mapping_one(domain, pci_domain_nr(pdev->bus),
-					 pdev->bus->number, pdev->devfn);
+					 pdev->bus->number, pdev->devfn,
+					 translation);
 	if (ret)
 		return ret;
 
@@ -1442,7 +1471,7 @@ domain_context_mapping(struct dmar_domain *domain, struct pci_dev *pdev)
 		ret = domain_context_mapping_one(domain,
 						 pci_domain_nr(parent->bus),
 						 parent->bus->number,
-						 parent->devfn);
+						 parent->devfn, translation);
 		if (ret)
 			return ret;
 		parent = parent->bus->self;
@@ -1450,12 +1479,14 @@ domain_context_mapping(struct dmar_domain *domain, struct pci_dev *pdev)
 	if (tmp->is_pcie) /* this is a PCIE-to-PCI bridge */
 		return domain_context_mapping_one(domain,
 					pci_domain_nr(tmp->subordinate),
-					tmp->subordinate->number, 0);
+					tmp->subordinate->number, 0,
+					translation);
 	else /* this is a legacy PCI bridge */
 		return domain_context_mapping_one(domain,
 						  pci_domain_nr(tmp->bus),
 						  tmp->bus->number,
-						  tmp->devfn);
+						  tmp->devfn,
+						  translation);
 }
 
 static int domain_context_mapped(struct pci_dev *pdev)
@@ -1752,7 +1783,7 @@ static int iommu_prepare_identity_map(struct pci_dev *pdev,
 		goto error;
 
 	/* context entry init */
-	ret = domain_context_mapping(domain, pdev);
+	ret = domain_context_mapping(domain, pdev, CONTEXT_TT_MULTI_LEVEL);
 	if (!ret)
 		return 0;
 error:
@@ -1853,6 +1884,23 @@ static inline void iommu_prepare_isa(void)
 }
 #endif /* !CONFIG_DMAR_FLPY_WA */
 
+/* Initialize each context entry as pass through.*/
+static int __init init_context_pass_through(void)
+{
+	struct pci_dev *pdev = NULL;
+	struct dmar_domain *domain;
+	int ret;
+
+	for_each_pci_dev(pdev) {
+		domain = get_domain_for_dev(pdev, DEFAULT_DOMAIN_ADDRESS_WIDTH);
+		ret = domain_context_mapping(domain, pdev,
+					     CONTEXT_TT_PASS_THROUGH);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
 static int __init init_dmars(void)
 {
 	struct dmar_drhd_unit *drhd;
@@ -1860,6 +1908,7 @@ static int __init init_dmars(void)
 	struct pci_dev *pdev;
 	struct intel_iommu *iommu;
 	int i, ret;
+	int pass_through = 1;
 
 	/*
 	 * for each drhd
@@ -1913,7 +1962,15 @@ static int __init init_dmars(void)
 			printk(KERN_ERR "IOMMU: allocate root entry failed\n");
 			goto error;
 		}
+		if (!ecap_pass_through(iommu->ecap))
+			pass_through = 0;
 	}
+	if (iommu_pass_through)
+		if (!pass_through) {
+			printk(KERN_INFO
+			       "Pass Through is not supported by hardware.\n");
+			iommu_pass_through = 0;
+		}
 
 	/*
 	 * Start from the sane iommu hardware state.
@@ -1976,37 +2033,57 @@ static int __init init_dmars(void)
 			       "IOMMU: enable interrupt remapping failed\n");
 	}
 #endif
+	/*
+	 * If pass through is set and enabled, context entries of all pci
+	 * devices are intialized by pass through translation type.
+	 */
+	if (iommu_pass_through) {
+		ret = init_context_pass_through();
+		if (ret) {
+			printk(KERN_ERR "IOMMU: Pass through init failed.\n");
+			iommu_pass_through = 0;
+		}
+	}
 
 	/*
-	 * For each rmrr
-	 *   for each dev attached to rmrr
-	 *   do
-	 *     locate drhd for dev, alloc domain for dev
-	 *     allocate free domain
-	 *     allocate page table entries for rmrr
-	 *     if context not allocated for bus
-	 *           allocate and init context
-	 *           set present in root table for this bus
-	 *     init context with domain, translation etc
-	 *    endfor
-	 * endfor
+	 * If pass through is not set or not enabled, setup context entries for
+	 * identity mappings for rmrr, gfx, and isa.
 	 */
-	for_each_rmrr_units(rmrr) {
-		for (i = 0; i < rmrr->devices_cnt; i++) {
-			pdev = rmrr->devices[i];
-			/* some BIOS lists non-exist devices in DMAR table */
-			if (!pdev)
-				continue;
-			ret = iommu_prepare_rmrr_dev(rmrr, pdev);
-			if (ret)
-				printk(KERN_ERR
+	if (!iommu_pass_through) {
+		/*
+		 * For each rmrr
+		 *   for each dev attached to rmrr
+		 *   do
+		 *     locate drhd for dev, alloc domain for dev
+		 *     allocate free domain
+		 *     allocate page table entries for rmrr
+		 *     if context not allocated for bus
+		 *           allocate and init context
+		 *           set present in root table for this bus
+		 *     init context with domain, translation etc
+		 *    endfor
+		 * endfor
+		 */
+		for_each_rmrr_units(rmrr) {
+			for (i = 0; i < rmrr->devices_cnt; i++) {
+				pdev = rmrr->devices[i];
+				/*
+				 * some BIOS lists non-exist devices in DMAR
+				 * table.
+				 */
+				if (!pdev)
+					continue;
+				ret = iommu_prepare_rmrr_dev(rmrr, pdev);
+				if (ret)
+					printk(KERN_ERR
 				 "IOMMU: mapping reserved region failed\n");
+			}
 		}
-	}
 
-	iommu_prepare_gfx_mapping();
+		iommu_prepare_gfx_mapping();
 
-	iommu_prepare_isa();
+		iommu_prepare_isa();
+	}
 
 	/*
 	 * for each drhd
@@ -2117,7 +2194,8 @@ get_valid_domain_for_dev(struct pci_dev *pdev)
 
 	/* make sure context mapping is ok */
 	if (unlikely(!domain_context_mapped(pdev))) {
-		ret = domain_context_mapping(domain, pdev);
+		ret = domain_context_mapping(domain, pdev,
+					     CONTEXT_TT_MULTI_LEVEL);
 		if (ret) {
 			printk(KERN_ERR
 				"Domain context map for %s failed",
@@ -2786,7 +2864,7 @@ int __init intel_iommu_init(void)
 	 * Check the need for DMA-remapping initialization now.
 	 * Above initialization will also be used by Interrupt-remapping.
 	 */
-	if (no_iommu || swiotlb || dmar_disabled)
+	if (no_iommu || (swiotlb && !iommu_pass_through) || dmar_disabled)
 		return -ENODEV;
 
 	iommu_init_mempool();
@@ -2806,7 +2884,15 @@ int __init intel_iommu_init(void)
 
 	init_timer(&unmap_timer);
 	force_iommu = 1;
-	dma_ops = &intel_dma_ops;
+
+	if (!iommu_pass_through) {
+		printk(KERN_INFO
+		       "Multi-level page-table translation for DMAR.\n");
+		dma_ops = &intel_dma_ops;
+	} else
+		printk(KERN_INFO
+		       "DMAR: Pass through translation for DMAR.\n");
+
 	init_iommu_sysfs();
 
 	register_iommu(&intel_iommu_ops);
@@ -3146,7 +3232,7 @@ static int intel_iommu_attach_device(struct iommu_domain *domain,
 		return -EFAULT;
 	}
 
-	ret = domain_context_mapping(dmar_domain, pdev);
+	ret = domain_context_mapping(dmar_domain, pdev, CONTEXT_TT_MULTI_LEVEL);
 	if (ret)
 		return ret;
 
diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index 1a455f1f86d7..e0a03aff63d9 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -13,6 +13,9 @@
 #define DMA_PTE_WRITE (2)
 #define DMA_PTE_SNP (1 << 11)
 
+#define CONTEXT_TT_MULTI_LEVEL	0
+#define CONTEXT_TT_PASS_THROUGH 2
+
 struct intel_iommu;
 struct dmar_domain;
 struct root_entry;
@@ -21,11 +24,16 @@ extern void free_dmar_iommu(struct intel_iommu *iommu);
 
 #ifdef CONFIG_DMAR
 extern int iommu_calculate_agaw(struct intel_iommu *iommu);
+extern int iommu_calculate_max_sagaw(struct intel_iommu *iommu);
 #else
 static inline int iommu_calculate_agaw(struct intel_iommu *iommu)
 {
 	return 0;
 }
+static inline int iommu_calculate_max_sagaw(struct intel_iommu *iommu)
+{
+	return 0;
+}
 #endif
 
 extern int dmar_disabled;
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index aa8c53171233..7246971a7feb 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -120,6 +120,7 @@ static inline void dmar_writeq(void __iomem *addr, u64 val)
 	(ecap_iotlb_offset(e) + ecap_niotlb_iunits(e) * 16)
 #define ecap_coherent(e)	((e) & 0x1)
 #define ecap_qis(e)		((e) & 0x2)
+#define ecap_pass_through(e)	((e >> 6) & 0x1)
 #define ecap_eim_support(e)	((e >> 4) & 0x1)
 #define ecap_ir_support(e)	((e >> 3) & 0x1)
 #define ecap_max_handle_mask(e) ((e >> 20) & 0xf)
@@ -302,6 +303,7 @@ struct intel_iommu {
 	spinlock_t	register_lock; /* protect register handling */
 	int		seq_id;	/* sequence id of the iommu */
 	int		agaw; /* agaw of this iommu */
+	int		msagaw; /* max sagaw of this iommu */
 	unsigned int 	irq;
 	unsigned char 	name[13];    /* Device Name */
 
-- 
cgit v1.2.3-59-g8ed1b


From bc749ca4c405d507e6ec6e3f3e5475e9a09faf0a Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Tue, 7 Apr 2009 16:55:27 -0700
Subject: nfsd: eliminate ENCODE_HEAD macro

This macro doesn't serve any useful purpose.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4xdr.c | 46 ++++++++++++++++++++++------------------------
 1 file changed, 22 insertions(+), 24 deletions(-)

diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index b820c311931c..fe46edeab2f6 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -1503,8 +1503,6 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
  * task to translate them into Linux-specific versions which are more
  * consistent with the style used in NFSv2/v3...
  */
-#define ENCODE_HEAD              __be32 *p
-
 #define WRITE32(n)               *p++ = htonl(n)
 #define WRITE64(n)               do {				\
 	*p++ = htonl((u32)((n) >> 32));				\
@@ -2334,7 +2332,7 @@ fail:
 static void
 nfsd4_encode_stateid(struct nfsd4_compoundres *resp, stateid_t *sid)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	RESERVE_SPACE(sizeof(stateid_t));
 	WRITE32(sid->si_generation);
@@ -2345,7 +2343,7 @@ nfsd4_encode_stateid(struct nfsd4_compoundres *resp, stateid_t *sid)
 static __be32
 nfsd4_encode_access(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_access *access)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (!nfserr) {
 		RESERVE_SPACE(8);
@@ -2372,7 +2370,7 @@ nfsd4_encode_close(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_c
 static __be32
 nfsd4_encode_commit(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_commit *commit)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (!nfserr) {
 		RESERVE_SPACE(8);
@@ -2385,7 +2383,7 @@ nfsd4_encode_commit(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_
 static __be32
 nfsd4_encode_create(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_create *create)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (!nfserr) {
 		RESERVE_SPACE(32);
@@ -2421,7 +2419,7 @@ nfsd4_encode_getfh(struct nfsd4_compoundres *resp, __be32 nfserr, struct svc_fh
 {
 	struct svc_fh *fhp = *fhpp;
 	unsigned int len;
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (!nfserr) {
 		len = fhp->fh_handle.fh_size;
@@ -2440,7 +2438,7 @@ nfsd4_encode_getfh(struct nfsd4_compoundres *resp, __be32 nfserr, struct svc_fh
 static void
 nfsd4_encode_lock_denied(struct nfsd4_compoundres *resp, struct nfsd4_lock_denied *ld)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	RESERVE_SPACE(32 + XDR_LEN(ld->ld_sop ? ld->ld_sop->so_owner.len : 0));
 	WRITE64(ld->ld_start);
@@ -2496,7 +2494,7 @@ nfsd4_encode_locku(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_l
 static __be32
 nfsd4_encode_link(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_link *link)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (!nfserr) {
 		RESERVE_SPACE(20);
@@ -2510,7 +2508,7 @@ nfsd4_encode_link(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_li
 static __be32
 nfsd4_encode_open(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open *open)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 	ENCODE_SEQID_OP_HEAD;
 
 	if (nfserr)
@@ -2605,7 +2603,7 @@ nfsd4_encode_read(struct nfsd4_compoundres *resp, __be32 nfserr,
 	int v, pn;
 	unsigned long maxcount; 
 	long len;
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (nfserr)
 		return nfserr;
@@ -2667,7 +2665,7 @@ nfsd4_encode_readlink(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd
 {
 	int maxcount;
 	char *page;
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (nfserr)
 		return nfserr;
@@ -2716,7 +2714,7 @@ nfsd4_encode_readdir(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4
 	int maxcount;
 	loff_t offset;
 	__be32 *page, *savep, *tailbase;
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (nfserr)
 		return nfserr;
@@ -2792,7 +2790,7 @@ err_no_verf:
 static __be32
 nfsd4_encode_remove(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_remove *remove)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (!nfserr) {
 		RESERVE_SPACE(20);
@@ -2805,7 +2803,7 @@ nfsd4_encode_remove(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_
 static __be32
 nfsd4_encode_rename(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_rename *rename)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (!nfserr) {
 		RESERVE_SPACE(40);
@@ -2825,7 +2823,7 @@ nfsd4_encode_secinfo(struct nfsd4_compoundres *resp, __be32 nfserr,
 	u32 nflavs;
 	struct exp_flavor_info *flavs;
 	struct exp_flavor_info def_flavs[2];
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (nfserr)
 		goto out;
@@ -2890,7 +2888,7 @@ out:
 static __be32
 nfsd4_encode_setattr(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_setattr *setattr)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	RESERVE_SPACE(12);
 	if (nfserr) {
@@ -2910,7 +2908,7 @@ nfsd4_encode_setattr(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4
 static __be32
 nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_setclientid *scd)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (!nfserr) {
 		RESERVE_SPACE(8 + sizeof(nfs4_verifier));
@@ -2930,7 +2928,7 @@ nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, __be32 nfserr, struct n
 static __be32
 nfsd4_encode_write(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_write *write)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (!nfserr) {
 		RESERVE_SPACE(16);
@@ -2946,7 +2944,7 @@ static __be32
 nfsd4_encode_exchange_id(struct nfsd4_compoundres *resp, int nfserr,
 			 struct nfsd4_exchange_id *exid)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 	char *major_id;
 	char *server_scope;
 	int major_id_sz;
@@ -3001,7 +2999,7 @@ static __be32
 nfsd4_encode_create_session(struct nfsd4_compoundres *resp, int nfserr,
 			    struct nfsd4_create_session *sess)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (nfserr)
 		return nfserr;
@@ -3057,7 +3055,7 @@ __be32
 nfsd4_encode_sequence(struct nfsd4_compoundres *resp, int nfserr,
 		      struct nfsd4_sequence *seq)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 
 	if (nfserr)
 		return nfserr;
@@ -3205,7 +3203,7 @@ void
 nfsd4_encode_operation(struct nfsd4_compoundres *resp, struct nfsd4_op *op)
 {
 	__be32 *statp;
-	ENCODE_HEAD;
+	__be32 *p;
 
 	RESERVE_SPACE(8);
 	WRITE32(op->opnum);
@@ -3239,7 +3237,7 @@ status:
 void
 nfsd4_encode_replay(struct nfsd4_compoundres *resp, struct nfsd4_op *op)
 {
-	ENCODE_HEAD;
+	__be32 *p;
 	struct nfs4_replay *rp = op->replay;
 
 	BUG_ON(!rp);
-- 
cgit v1.2.3-59-g8ed1b


From 3352d2c2d0540955a7bbb3421a28330af7f9d79c Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Tue, 7 Apr 2009 17:03:19 -0700
Subject: nfsd4: delete obsolete xdr comments

We don't need comments to tell us these macros are ugly.  And we're long
past trying to share any of this code with the BSD's.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4xdr.c | 27 ---------------------------
 1 file changed, 27 deletions(-)

diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index fe46edeab2f6..4a71fcd3f036 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -83,16 +83,6 @@ check_filename(char *str, int len, __be32 err)
 	return 0;
 }
 
-/*
- * START OF "GENERIC" DECODE ROUTINES.
- *   These may look a little ugly since they are imported from a "generic"
- * set of XDR encode/decode routines which are intended to be shared by
- * all of our NFSv4 implementations (OpenBSD, MacOS X...).
- *
- * If the pain of reading these is too great, it should be a straightforward
- * task to translate them into Linux-specific versions which are more
- * consistent with the style used in NFSv2/v3...
- */
 #define DECODE_HEAD				\
 	__be32 *p;				\
 	__be32 status
@@ -1489,20 +1479,7 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
 
 	DECODE_TAIL;
 }
-/*
- * END OF "GENERIC" DECODE ROUTINES.
- */
 
-/*
- * START OF "GENERIC" ENCODE ROUTINES.
- *   These may look a little ugly since they are imported from a "generic"
- * set of XDR encode/decode routines which are intended to be shared by
- * all of our NFSv4 implementations (OpenBSD, MacOS X...).
- *
- * If the pain of reading these is too great, it should be a straightforward
- * task to translate them into Linux-specific versions which are more
- * consistent with the style used in NFSv2/v3...
- */
 #define WRITE32(n)               *p++ = htonl(n)
 #define WRITE64(n)               do {				\
 	*p++ = htonl((u32)((n) >> 32));				\
@@ -3252,10 +3229,6 @@ nfsd4_encode_replay(struct nfsd4_compoundres *resp, struct nfsd4_op *op)
 	ADJUST_ARGS();
 }
 
-/*
- * END OF "GENERIC" ENCODE ROUTINES.
- */
-
 int
 nfs4svc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy)
 {
-- 
cgit v1.2.3-59-g8ed1b


From c654b8a9cba6002aad1c01919e4928a79a4a6dcf Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Thu, 16 Apr 2009 17:33:25 -0400
Subject: nfsd: support ext4 i_version

ext4 supports a real NFSv4 change attribute, which is bumped whenever
the ctime would be updated, including times when two updates arrive
within a jiffy of each other.  (Note that although ext4 has space for
nanosecond-precision ctime, the real resolution is lower: it actually
uses jiffies as the time-source.)  This ensures clients will invalidate
their caches when they need to.

There is some fear that keeping the i_version up-to-date could have
performance drawbacks, so for now it's turned on only by a mount option.
We hope to do something better eventually.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Cc: Theodore Tso <tytso@mit.edu>
---
 fs/nfsd/nfs3xdr.c          |  1 +
 fs/nfsd/nfs4xdr.c          | 63 ++++++++++++++++++++++++++++++----------------
 include/linux/nfsd/nfsfh.h |  7 ++++++
 include/linux/nfsd/xdr4.h  | 17 ++++++++++---
 4 files changed, 63 insertions(+), 25 deletions(-)

diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c
index 17d0dd997204..01d4ec1c88e0 100644
--- a/fs/nfsd/nfs3xdr.c
+++ b/fs/nfsd/nfs3xdr.c
@@ -272,6 +272,7 @@ void fill_post_wcc(struct svc_fh *fhp)
 
 	err = vfs_getattr(fhp->fh_export->ex_path.mnt, fhp->fh_dentry,
 			&fhp->fh_post_attr);
+	fhp->fh_post_change = fhp->fh_dentry->d_inode->i_version;
 	if (err)
 		fhp->fh_post_saved = 0;
 	else
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 4a71fcd3f036..12d36a7361cd 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -1490,13 +1490,41 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
 	memcpy(p, ptr, nbytes);					\
 	p += XDR_QUADLEN(nbytes);				\
 }} while (0)
-#define WRITECINFO(c)		do {				\
-	*p++ = htonl(c.atomic);					\
-	*p++ = htonl(c.before_ctime_sec);				\
-	*p++ = htonl(c.before_ctime_nsec);				\
-	*p++ = htonl(c.after_ctime_sec);				\
-	*p++ = htonl(c.after_ctime_nsec);				\
-} while (0)
+
+static void write32(__be32 **p, u32 n)
+{
+	*(*p)++ = n;
+}
+
+static void write64(__be32 **p, u64 n)
+{
+	write32(p, (u32)(n >> 32));
+	write32(p, (u32)n);
+}
+
+static void write_change(__be32 **p, struct kstat *stat, struct inode *inode)
+{
+	if (IS_I_VERSION(inode)) {
+		write64(p, inode->i_version);
+	} else {
+		write32(p, stat->ctime.tv_sec);
+		write32(p, stat->ctime.tv_nsec);
+	}
+}
+
+static void write_cinfo(__be32 **p, struct nfsd4_change_info *c)
+{
+	write32(p, c->atomic);
+	if (c->change_supported) {
+		write64(p, c->before_change);
+		write64(p, c->after_change);
+	} else {
+		write32(p, c->before_ctime_sec);
+		write32(p, c->before_ctime_nsec);
+		write32(p, c->after_ctime_sec);
+		write32(p, c->after_ctime_nsec);
+	}
+}
 
 #define RESERVE_SPACE(nbytes)	do {				\
 	p = resp->p;						\
@@ -1849,16 +1877,9 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp,
 			WRITE32(NFS4_FH_PERSISTENT|NFS4_FH_VOL_RENAME);
 	}
 	if (bmval0 & FATTR4_WORD0_CHANGE) {
-		/*
-		 * Note: This _must_ be consistent with the scheme for writing
-		 * change_info, so any changes made here must be reflected there
-		 * as well.  (See xdr4.h:set_change_info() and the WRITECINFO()
-		 * macro above.)
-		 */
 		if ((buflen -= 8) < 0)
 			goto out_resource;
-		WRITE32(stat.ctime.tv_sec);
-		WRITE32(stat.ctime.tv_nsec);
+		write_change(&p, &stat, dentry->d_inode);
 	}
 	if (bmval0 & FATTR4_WORD0_SIZE) {
 		if ((buflen -= 8) < 0)
@@ -2364,7 +2385,7 @@ nfsd4_encode_create(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_
 
 	if (!nfserr) {
 		RESERVE_SPACE(32);
-		WRITECINFO(create->cr_cinfo);
+		write_cinfo(&p, &create->cr_cinfo);
 		WRITE32(2);
 		WRITE32(create->cr_bmval[0]);
 		WRITE32(create->cr_bmval[1]);
@@ -2475,7 +2496,7 @@ nfsd4_encode_link(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_li
 
 	if (!nfserr) {
 		RESERVE_SPACE(20);
-		WRITECINFO(link->li_cinfo);
+		write_cinfo(&p, &link->li_cinfo);
 		ADJUST_ARGS();
 	}
 	return nfserr;
@@ -2493,7 +2514,7 @@ nfsd4_encode_open(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_op
 
 	nfsd4_encode_stateid(resp, &open->op_stateid);
 	RESERVE_SPACE(40);
-	WRITECINFO(open->op_cinfo);
+	write_cinfo(&p, &open->op_cinfo);
 	WRITE32(open->op_rflags);
 	WRITE32(2);
 	WRITE32(open->op_bmval[0]);
@@ -2771,7 +2792,7 @@ nfsd4_encode_remove(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_
 
 	if (!nfserr) {
 		RESERVE_SPACE(20);
-		WRITECINFO(remove->rm_cinfo);
+		write_cinfo(&p, &remove->rm_cinfo);
 		ADJUST_ARGS();
 	}
 	return nfserr;
@@ -2784,8 +2805,8 @@ nfsd4_encode_rename(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_
 
 	if (!nfserr) {
 		RESERVE_SPACE(40);
-		WRITECINFO(rename->rn_sinfo);
-		WRITECINFO(rename->rn_tinfo);
+		write_cinfo(&p, &rename->rn_sinfo);
+		write_cinfo(&p, &rename->rn_tinfo);
 		ADJUST_ARGS();
 	}
 	return nfserr;
diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h
index afa19016c4a8..8f641c908450 100644
--- a/include/linux/nfsd/nfsfh.h
+++ b/include/linux/nfsd/nfsfh.h
@@ -151,9 +151,15 @@ typedef struct svc_fh {
 	__u64			fh_pre_size;	/* size before operation */
 	struct timespec		fh_pre_mtime;	/* mtime before oper */
 	struct timespec		fh_pre_ctime;	/* ctime before oper */
+	/*
+	 * pre-op nfsv4 change attr: note must check IS_I_VERSION(inode)
+	 *  to find out if it is valid.
+	 */
+	u64			fh_pre_change;
 
 	/* Post-op attributes saved in fh_unlock */
 	struct kstat		fh_post_attr;	/* full attrs after operation */
+	u64			fh_post_change; /* nfsv4 change; see above */
 #endif /* CONFIG_NFSD_V3 */
 
 } svc_fh;
@@ -298,6 +304,7 @@ fill_pre_wcc(struct svc_fh *fhp)
 		fhp->fh_pre_mtime = inode->i_mtime;
 		fhp->fh_pre_ctime = inode->i_ctime;
 		fhp->fh_pre_size  = inode->i_size;
+		fhp->fh_pre_change = inode->i_version;
 		fhp->fh_pre_saved = 1;
 	}
 }
diff --git a/include/linux/nfsd/xdr4.h b/include/linux/nfsd/xdr4.h
index f80d6013fdc3..d0f050f01eca 100644
--- a/include/linux/nfsd/xdr4.h
+++ b/include/linux/nfsd/xdr4.h
@@ -64,10 +64,13 @@ static inline bool nfsd4_has_session(struct nfsd4_compound_state *cs)
 
 struct nfsd4_change_info {
 	u32		atomic;
+	bool		change_supported;
 	u32		before_ctime_sec;
 	u32		before_ctime_nsec;
+	u64		before_change;
 	u32		after_ctime_sec;
 	u32		after_ctime_nsec;
+	u64		after_change;
 };
 
 struct nfsd4_access {
@@ -503,10 +506,16 @@ set_change_info(struct nfsd4_change_info *cinfo, struct svc_fh *fhp)
 {
 	BUG_ON(!fhp->fh_pre_saved || !fhp->fh_post_saved);
 	cinfo->atomic = 1;
-	cinfo->before_ctime_sec = fhp->fh_pre_ctime.tv_sec;
-	cinfo->before_ctime_nsec = fhp->fh_pre_ctime.tv_nsec;
-	cinfo->after_ctime_sec = fhp->fh_post_attr.ctime.tv_sec;
-	cinfo->after_ctime_nsec = fhp->fh_post_attr.ctime.tv_nsec;
+	cinfo->change_supported = IS_I_VERSION(fhp->fh_dentry->d_inode);
+	if (cinfo->change_supported) {
+		cinfo->before_change = fhp->fh_pre_change;
+		cinfo->after_change = fhp->fh_post_change;
+	} else {
+		cinfo->before_ctime_sec = fhp->fh_pre_ctime.tv_sec;
+		cinfo->before_ctime_nsec = fhp->fh_pre_ctime.tv_nsec;
+		cinfo->after_ctime_sec = fhp->fh_post_attr.ctime.tv_sec;
+		cinfo->after_ctime_nsec = fhp->fh_post_attr.ctime.tv_nsec;
+	}
 }
 
 int nfs4svc_encode_voidres(struct svc_rqst *, __be32 *, void *);
-- 
cgit v1.2.3-59-g8ed1b


From b8fd47aefa5f13df1edacbc7e68d9874635109e5 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Wed, 29 Apr 2009 11:36:17 -0400
Subject: nfsd: quiet compile warning

Stephen Rothwell said:
"Today's linux-next build (powerpc ppc64_defconfig) produced this new
warning:

fs/nfsd/nfs4state.c: In function 'EXPIRED_STATEID':
fs/nfsd/nfs4state.c:2757: warning: comparison of distinct pointer types lacks a cast

Caused by commit 78155ed75f470710f2aecb3e75e3d97107ba8374 ("nfsd4:
distinguish expired from stale stateids")."

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 fs/nfsd/nfs4state.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 74e822ec34cb..d24dd12ddb4d 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -2754,7 +2754,7 @@ EXPIRED_STATEID(stateid_t *stateid)
 {
 	if (time_before((unsigned long)boot_time,
 			((unsigned long)stateid->si_boot)) &&
-	    time_before((stateid->si_boot + lease_time), get_seconds())) {
+	    time_before((unsigned long)(stateid->si_boot + lease_time), get_seconds())) {
 		dprintk("NFSD: expired stateid (%08x/%08x/%08x/%08x)!\n",
 			stateid->si_boot, stateid->si_stateownerid,
 			stateid->si_fileid, stateid->si_generation);
-- 
cgit v1.2.3-59-g8ed1b


From f64f79ea5f5e02ba8585f35a10b4a3bcab0cea52 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Wed, 29 Apr 2009 13:45:36 -0400
Subject: nfsd4: setclientid_confirm callback-change fixes

This setclientid_confirm case should allow the client to change
callbacks, but it currently has a dummy implementation that just turns
off callbacks completely.  That dummy implementation isn't completely
correct either, though:

	- There's no need to remove any client recovery directory in
	  this case.
	- New clientid confirm verifiers should be generated (and
	  returned) in setclientid; there's no need to generate a new
	  one here.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4state.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index d24dd12ddb4d..7e1fcc3aade4 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -1688,8 +1688,6 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp,
 			/* XXX: We just turn off callbacks until we can handle
 			  * change request correctly. */
 			atomic_set(&conf->cl_callback.cb_set, 0);
-			gen_confirm(conf);
-			nfsd4_remove_clid_dir(unconf);
 			expire_client(unconf);
 			status = nfs_ok;
 
-- 
cgit v1.2.3-59-g8ed1b


From 595947acaaef373445131471a78650003f5d8e7d Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Thu, 5 Mar 2009 17:18:10 -0500
Subject: nfsd4: set shorter timeout

We tried to do something overly complicated with the callback rpc
timeouts here.  And they're wrong--the result is that by the time a
single callback times out, it's already too late to tell the client
(using the cb_path_down return to RENEW) that the callback is down.

Use a much shorter, simpler timeout.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4callback.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 290289bd44f7..049f052a6eb3 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -358,6 +358,11 @@ static struct rpc_program cb_program = {
 		.pipe_dir_name  = "/nfsd4_cb",
 };
 
+static int max_cb_time(void)
+{
+	return max(NFSD_LEASE_TIME/10, (time_t)1) * HZ;
+}
+
 /* Reference counting, callback cleanup, etc., all look racy as heck.
  * And why is cb_set an atomic? */
 
@@ -366,10 +371,8 @@ static struct rpc_clnt *setup_callback_client(struct nfs4_client *clp)
 	struct sockaddr_in	addr;
 	struct nfs4_callback    *cb = &clp->cl_callback;
 	struct rpc_timeout	timeparms = {
-		.to_initval	= (NFSD_LEASE_TIME/4) * HZ,
-		.to_retries	= 5,
-		.to_maxval	= (NFSD_LEASE_TIME/2) * HZ,
-		.to_exponential	= 1,
+		.to_initval	= max_cb_time(),
+		.to_retries	= 0,
 	};
 	struct rpc_create_args args = {
 		.protocol	= IPPROTO_TCP,
-- 
cgit v1.2.3-59-g8ed1b


From e1cab5a5896e142190cd66a8287099b52e5855a7 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Mon, 23 Feb 2009 10:45:27 -0800
Subject: nfsd4: set cb_client inside setup_callback_client

This is just a minor code simplification.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4callback.c | 26 ++++++++++++--------------
 1 file changed, 12 insertions(+), 14 deletions(-)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 049f052a6eb3..4788d09d9bec 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -366,7 +366,7 @@ static int max_cb_time(void)
 /* Reference counting, callback cleanup, etc., all look racy as heck.
  * And why is cb_set an atomic? */
 
-static struct rpc_clnt *setup_callback_client(struct nfs4_client *clp)
+int setup_callback_client(struct nfs4_client *clp)
 {
 	struct sockaddr_in	addr;
 	struct nfs4_callback    *cb = &clp->cl_callback;
@@ -389,7 +389,7 @@ static struct rpc_clnt *setup_callback_client(struct nfs4_client *clp)
 	struct rpc_clnt *client;
 
 	if (!clp->cl_principal && (clp->cl_flavor >= RPC_AUTH_GSS_KRB5))
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
 
 	/* Initialize address */
 	memset(&addr, 0, sizeof(addr));
@@ -399,10 +399,13 @@ static struct rpc_clnt *setup_callback_client(struct nfs4_client *clp)
 
 	/* Create RPC client */
 	client = rpc_create(&args);
-	if (IS_ERR(client))
+	if (IS_ERR(client)) {
 		dprintk("NFSD: couldn't create callback client: %ld\n",
 			PTR_ERR(client));
-	return client;
+		return PTR_ERR(client);
+	}
+	cb->cb_client = client;
+	return 0;
 
 }
 
@@ -414,28 +417,23 @@ static int do_probe_callback(void *data)
 		.rpc_proc       = &nfs4_cb_procedures[NFSPROC4_CLNT_CB_NULL],
 		.rpc_argp       = clp,
 	};
-	struct rpc_clnt *client;
 	int status;
 
-	client = setup_callback_client(clp);
-	if (IS_ERR(client)) {
-		status = PTR_ERR(client);
-		dprintk("NFSD: couldn't create callback client: %d\n",
-								status);
+	status = setup_callback_client(clp);
+	if (status)
 		goto out_err;
-	}
 
-	status = rpc_call_sync(client, &msg, RPC_TASK_SOFT);
+	status = rpc_call_sync(cb->cb_client, &msg, RPC_TASK_SOFT);
 
 	if (status)
 		goto out_release_client;
 
-	cb->cb_client = client;
 	atomic_set(&cb->cb_set, 1);
 	put_nfs4_client(clp);
 	return 0;
 out_release_client:
-	rpc_shutdown_client(client);
+	rpc_shutdown_client(cb->cb_client);
+	cb->cb_client = NULL;
 out_err:
 	dprintk("NFSD: warning: no callback path to client %.*s: error %d\n",
 		(int)clp->cl_name.len, clp->cl_name.data, status);
-- 
cgit v1.2.3-59-g8ed1b


From ecdd03b7914c91ef849e49c4d466c87f4981b5cd Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Mon, 23 Feb 2009 19:35:22 -0800
Subject: nfsd4: create rpc callback client from server thread

The code is a little simpler, and it should be easier to avoid races, if
we just do all rpc client creation/destruction from nfsd or laundromat
threads and do only the rpc calls themselves asynchronously.  The rpc
creation doesn't involve any significant waiting (it doesn't call the
client, for example), so there's no reason not to do this.

Also don't bother destroying the client on failure of the rpc null
probe.  We may want to retry the probe later anyway.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4callback.c | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 4788d09d9bec..711c6282151f 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -409,6 +409,12 @@ int setup_callback_client(struct nfs4_client *clp)
 
 }
 
+static void warn_no_callback_path(struct nfs4_client *clp, int reason)
+{
+	dprintk("NFSD: warning: no callback path to client %.*s: error %d\n",
+		(int)clp->cl_name.len, clp->cl_name.data, reason);
+}
+
 static int do_probe_callback(void *data)
 {
 	struct nfs4_client *clp = data;
@@ -419,24 +425,12 @@ static int do_probe_callback(void *data)
 	};
 	int status;
 
-	status = setup_callback_client(clp);
-	if (status)
-		goto out_err;
-
 	status = rpc_call_sync(cb->cb_client, &msg, RPC_TASK_SOFT);
-
 	if (status)
-		goto out_release_client;
+		warn_no_callback_path(clp, status);
+	else
+		atomic_set(&cb->cb_set, 1);
 
-	atomic_set(&cb->cb_set, 1);
-	put_nfs4_client(clp);
-	return 0;
-out_release_client:
-	rpc_shutdown_client(cb->cb_client);
-	cb->cb_client = NULL;
-out_err:
-	dprintk("NFSD: warning: no callback path to client %.*s: error %d\n",
-		(int)clp->cl_name.len, clp->cl_name.data, status);
 	put_nfs4_client(clp);
 	return 0;
 }
@@ -448,9 +442,16 @@ void
 nfsd4_probe_callback(struct nfs4_client *clp)
 {
 	struct task_struct *t;
+	int status;
 
 	BUG_ON(atomic_read(&clp->cl_callback.cb_set));
 
+	status = setup_callback_client(clp);
+	if (status) {
+		warn_no_callback_path(clp, status);
+		return;
+	}
+
 	/* the task holds a reference to the nfs4_client struct */
 	atomic_inc(&clp->cl_count);
 
-- 
cgit v1.2.3-59-g8ed1b


From 3cef9ab266a932899e756f7e1ea7a988a97bf3b2 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Mon, 23 Feb 2009 21:42:10 -0800
Subject: nfsd4: lookup up callback cred only once

Lookup the callback cred once and then use it for all subsequent
callbacks.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4callback.c     | 26 ++++++++++++++++++++++++++
 fs/nfsd/nfs4state.c        |  4 ++++
 include/linux/nfsd/state.h |  1 +
 3 files changed, 31 insertions(+)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 711c6282151f..cc10ed35ac81 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -415,6 +415,22 @@ static void warn_no_callback_path(struct nfs4_client *clp, int reason)
 		(int)clp->cl_name.len, clp->cl_name.data, reason);
 }
 
+static struct rpc_cred *lookup_cb_cred(struct nfs4_callback *cb)
+{
+	struct auth_cred acred = {
+		.machine_cred = 1
+	};
+
+	/*
+	 * Note in the gss case this doesn't actually have to wait for a
+	 * gss upcall (or any calls to the client); this just creates a
+	 * non-uptodate cred which the rpc state machine will fill in with
+	 * a refresh_upcall later.
+	 */
+	return rpcauth_lookup_credcache(cb->cb_client->cl_auth, &acred,
+							RPCAUTH_LOOKUP_NEW);
+}
+
 static int do_probe_callback(void *data)
 {
 	struct nfs4_client *clp = data;
@@ -423,9 +439,18 @@ static int do_probe_callback(void *data)
 		.rpc_proc       = &nfs4_cb_procedures[NFSPROC4_CLNT_CB_NULL],
 		.rpc_argp       = clp,
 	};
+	struct rpc_cred *cred;
 	int status;
 
+	cred = lookup_cb_cred(cb);
+	if (IS_ERR(cred)) {
+		status = PTR_ERR(cred);
+		goto out;
+	}
+	cb->cb_cred = cred;
+	msg.rpc_cred = cb->cb_cred;
 	status = rpc_call_sync(cb->cb_client, &msg, RPC_TASK_SOFT);
+out:
 	if (status)
 		warn_no_callback_path(clp, status);
 	else
@@ -475,6 +500,7 @@ nfsd4_cb_recall(struct nfs4_delegation *dp)
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_cb_procedures[NFSPROC4_CLNT_CB_RECALL],
 		.rpc_argp = cbr,
+		.rpc_cred = clp->cl_callback.cb_cred
 	};
 	int retries = 1;
 	int status = 0;
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 7e1fcc3aade4..b205c7d7bc6a 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -643,6 +643,10 @@ shutdown_callback_client(struct nfs4_client *clp)
 		clp->cl_callback.cb_client = NULL;
 		rpc_shutdown_client(clnt);
 	}
+	if (clp->cl_callback.cb_cred) {
+		put_rpccred(clp->cl_callback.cb_cred);
+		clp->cl_callback.cb_cred = NULL;
+	}
 }
 
 static inline void
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 4d61c873feed..8d882a3eb4b9 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -97,6 +97,7 @@ struct nfs4_callback {
 	/* RPC client info */
 	atomic_t		cb_set;     /* successful CB_NULL call */
 	struct rpc_clnt *       cb_client;
+	struct rpc_cred	*	cb_cred;
 };
 
 /* Maximum number of slots per session. 128 is useful for long haul TCP */
-- 
cgit v1.2.3-59-g8ed1b


From e300a63ce4ccec073d254d883a3638d5dca1d771 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Thu, 5 Mar 2009 15:01:11 -0500
Subject: nfsd4: replace callback thread by asynchronous rpc

We don't really need a synchronous rpc, and moving to an asynchronous
rpc allows us to do without this extra kthread.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4callback.c | 38 ++++++++++++++++++++++----------------
 1 file changed, 22 insertions(+), 16 deletions(-)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index cc10ed35ac81..0aaf68beedbd 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -415,6 +415,21 @@ static void warn_no_callback_path(struct nfs4_client *clp, int reason)
 		(int)clp->cl_name.len, clp->cl_name.data, reason);
 }
 
+static void nfsd4_cb_probe_done(struct rpc_task *task, void *calldata)
+{
+	struct nfs4_client *clp = calldata;
+
+	if (task->tk_status)
+		warn_no_callback_path(clp, task->tk_status);
+	else
+		atomic_set(&clp->cl_callback.cb_set, 1);
+	put_nfs4_client(clp);
+}
+
+static const struct rpc_call_ops nfsd4_cb_probe_ops = {
+	.rpc_call_done = nfsd4_cb_probe_done,
+};
+
 static struct rpc_cred *lookup_cb_cred(struct nfs4_callback *cb)
 {
 	struct auth_cred acred = {
@@ -431,9 +446,8 @@ static struct rpc_cred *lookup_cb_cred(struct nfs4_callback *cb)
 							RPCAUTH_LOOKUP_NEW);
 }
 
-static int do_probe_callback(void *data)
+void do_probe_callback(struct nfs4_client *clp)
 {
-	struct nfs4_client *clp = data;
 	struct nfs4_callback    *cb = &clp->cl_callback;
 	struct rpc_message msg = {
 		.rpc_proc       = &nfs4_cb_procedures[NFSPROC4_CLNT_CB_NULL],
@@ -449,15 +463,13 @@ static int do_probe_callback(void *data)
 	}
 	cb->cb_cred = cred;
 	msg.rpc_cred = cb->cb_cred;
-	status = rpc_call_sync(cb->cb_client, &msg, RPC_TASK_SOFT);
+	status = rpc_call_async(cb->cb_client, &msg, RPC_TASK_SOFT,
+				&nfsd4_cb_probe_ops, (void *)clp);
 out:
-	if (status)
+	if (status) {
 		warn_no_callback_path(clp, status);
-	else
-		atomic_set(&cb->cb_set, 1);
-
-	put_nfs4_client(clp);
-	return 0;
+		put_nfs4_client(clp);
+	}
 }
 
 /*
@@ -466,7 +478,6 @@ out:
 void
 nfsd4_probe_callback(struct nfs4_client *clp)
 {
-	struct task_struct *t;
 	int status;
 
 	BUG_ON(atomic_read(&clp->cl_callback.cb_set));
@@ -480,12 +491,7 @@ nfsd4_probe_callback(struct nfs4_client *clp)
 	/* the task holds a reference to the nfs4_client struct */
 	atomic_inc(&clp->cl_count);
 
-	t = kthread_run(do_probe_callback, clp, "nfs4_cb_probe");
-
-	if (IS_ERR(t))
-		atomic_dec(&clp->cl_count);
-
-	return;
+	do_probe_callback(clp);
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From aed5d5f4c5ea5da01a774e42cff08c4b4fa59072 Mon Sep 17 00:00:00 2001
From: Fenghua Yu <fenghua.yu@intel.com>
Date: Thu, 30 Apr 2009 17:57:11 -0700
Subject: Fix !CONFIG_DMAR build failure introduced by Intel IOMMU Pass Through
 Support

This updated patch should fix the compiling errors and remove the extern
iommu_pass_through from drivers/pci/intel-iommu.c file.

Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 arch/ia64/kernel/pci-dma.c | 2 ++
 arch/x86/kernel/pci-dma.c  | 4 ++--
 drivers/pci/intel-iommu.c  | 1 -
 3 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/arch/ia64/kernel/pci-dma.c b/arch/ia64/kernel/pci-dma.c
index eb987386f691..ecdde25d0d18 100644
--- a/arch/ia64/kernel/pci-dma.c
+++ b/arch/ia64/kernel/pci-dma.c
@@ -32,6 +32,8 @@ int force_iommu __read_mostly = 1;
 int force_iommu __read_mostly;
 #endif
 
+int iommu_pass_through;
+
 /* Dummy device used for NULL arguments (normally ISA). Better would
    be probably a smaller DMA mask, but this is bug-to-bug compatible
    to i386. */
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 8cad0d854242..049005e82178 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -32,6 +32,8 @@ int no_iommu __read_mostly;
 /* Set this to 1 if there is a HW IOMMU in the system */
 int iommu_detected __read_mostly = 0;
 
+int iommu_pass_through;
+
 dma_addr_t bad_dma_address __read_mostly = 0;
 EXPORT_SYMBOL(bad_dma_address);
 
@@ -160,8 +162,6 @@ again:
 	return page_address(page);
 }
 
-extern int iommu_pass_through;
-
 /*
  * See <Documentation/x86_64/boot-options.txt> for the iommu kernel parameter
  * documentation.
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 13121821db7f..d3edd6aa82ce 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -288,7 +288,6 @@ int dmar_disabled = 1;
 static int __initdata dmar_map_gfx = 1;
 static int dmar_forcedac;
 static int intel_iommu_strict;
-int iommu_pass_through;
 
 #define DUMMY_DEVICE_DOMAIN_INFO ((struct device_domain_info *)(-1))
 static DEFINE_SPINLOCK(device_domain_lock);
-- 
cgit v1.2.3-59-g8ed1b


From c237dc0303bcf6f4cc2e0efe4fe4e341c6f34dac Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Wed, 29 Apr 2009 19:09:19 -0400
Subject: nfsd4: rename callback struct to cb_conn

I want to use the name for a struct that actually does represent a
single callback.

(Actually, I've never been sure it helps to a separate struct for the
callback information.  Some day maybe those fields could just be dumped
into struct nfs4_client.  I don't know.)

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4callback.c     | 16 ++++++++--------
 fs/nfsd/nfs4state.c        | 22 +++++++++++-----------
 include/linux/nfsd/state.h |  4 ++--
 3 files changed, 21 insertions(+), 21 deletions(-)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 0aaf68beedbd..ed860d7ddd19 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -369,7 +369,7 @@ static int max_cb_time(void)
 int setup_callback_client(struct nfs4_client *clp)
 {
 	struct sockaddr_in	addr;
-	struct nfs4_callback    *cb = &clp->cl_callback;
+	struct nfs4_cb_conn *cb = &clp->cl_cb_conn;
 	struct rpc_timeout	timeparms = {
 		.to_initval	= max_cb_time(),
 		.to_retries	= 0,
@@ -422,7 +422,7 @@ static void nfsd4_cb_probe_done(struct rpc_task *task, void *calldata)
 	if (task->tk_status)
 		warn_no_callback_path(clp, task->tk_status);
 	else
-		atomic_set(&clp->cl_callback.cb_set, 1);
+		atomic_set(&clp->cl_cb_conn.cb_set, 1);
 	put_nfs4_client(clp);
 }
 
@@ -430,7 +430,7 @@ static const struct rpc_call_ops nfsd4_cb_probe_ops = {
 	.rpc_call_done = nfsd4_cb_probe_done,
 };
 
-static struct rpc_cred *lookup_cb_cred(struct nfs4_callback *cb)
+static struct rpc_cred *lookup_cb_cred(struct nfs4_cb_conn *cb)
 {
 	struct auth_cred acred = {
 		.machine_cred = 1
@@ -448,7 +448,7 @@ static struct rpc_cred *lookup_cb_cred(struct nfs4_callback *cb)
 
 void do_probe_callback(struct nfs4_client *clp)
 {
-	struct nfs4_callback    *cb = &clp->cl_callback;
+	struct nfs4_cb_conn *cb = &clp->cl_cb_conn;
 	struct rpc_message msg = {
 		.rpc_proc       = &nfs4_cb_procedures[NFSPROC4_CLNT_CB_NULL],
 		.rpc_argp       = clp,
@@ -480,7 +480,7 @@ nfsd4_probe_callback(struct nfs4_client *clp)
 {
 	int status;
 
-	BUG_ON(atomic_read(&clp->cl_callback.cb_set));
+	BUG_ON(atomic_read(&clp->cl_cb_conn.cb_set));
 
 	status = setup_callback_client(clp);
 	if (status) {
@@ -501,12 +501,12 @@ void
 nfsd4_cb_recall(struct nfs4_delegation *dp)
 {
 	struct nfs4_client *clp = dp->dl_client;
-	struct rpc_clnt *clnt = clp->cl_callback.cb_client;
+	struct rpc_clnt *clnt = clp->cl_cb_conn.cb_client;
 	struct nfs4_cb_recall *cbr = &dp->dl_recall;
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_cb_procedures[NFSPROC4_CLNT_CB_RECALL],
 		.rpc_argp = cbr,
-		.rpc_cred = clp->cl_callback.cb_cred
+		.rpc_cred = clp->cl_cb_conn.cb_cred
 	};
 	int retries = 1;
 	int status = 0;
@@ -519,7 +519,7 @@ nfsd4_cb_recall(struct nfs4_delegation *dp)
 		switch (status) {
 			case -EIO:
 				/* Network partition? */
-				atomic_set(&clp->cl_callback.cb_set, 0);
+				atomic_set(&clp->cl_cb_conn.cb_set, 0);
 			case -EBADHANDLE:
 			case -NFS4ERR_BAD_STATEID:
 				/* Race: client probably got cb_recall
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index b205c7d7bc6a..d7b5e6b89207 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -182,7 +182,7 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f
 {
 	struct nfs4_delegation *dp;
 	struct nfs4_file *fp = stp->st_file;
-	struct nfs4_callback *cb = &stp->st_stateowner->so_client->cl_callback;
+	struct nfs4_cb_conn *cb = &stp->st_stateowner->so_client->cl_cb_conn;
 
 	dprintk("NFSD alloc_init_deleg\n");
 	if (fp->fi_had_conflict)
@@ -633,19 +633,19 @@ static struct nfs4_client *alloc_client(struct xdr_netobj name)
 static void
 shutdown_callback_client(struct nfs4_client *clp)
 {
-	struct rpc_clnt *clnt = clp->cl_callback.cb_client;
+	struct rpc_clnt *clnt = clp->cl_cb_conn.cb_client;
 
 	if (clnt) {
 		/*
 		 * Callback threads take a reference on the client, so there
 		 * should be no outstanding callbacks at this point.
 		 */
-		clp->cl_callback.cb_client = NULL;
+		clp->cl_cb_conn.cb_client = NULL;
 		rpc_shutdown_client(clnt);
 	}
-	if (clp->cl_callback.cb_cred) {
-		put_rpccred(clp->cl_callback.cb_cred);
-		clp->cl_callback.cb_cred = NULL;
+	if (clp->cl_cb_conn.cb_cred) {
+		put_rpccred(clp->cl_cb_conn.cb_cred);
+		clp->cl_cb_conn.cb_cred = NULL;
 	}
 }
 
@@ -719,7 +719,7 @@ static struct nfs4_client *create_client(struct xdr_netobj name, char *recdir)
 		return NULL;
 	memcpy(clp->cl_recdir, recdir, HEXDIR_LEN);
 	atomic_set(&clp->cl_count, 1);
-	atomic_set(&clp->cl_callback.cb_set, 0);
+	atomic_set(&clp->cl_cb_conn.cb_set, 0);
 	INIT_LIST_HEAD(&clp->cl_idhash);
 	INIT_LIST_HEAD(&clp->cl_strhash);
 	INIT_LIST_HEAD(&clp->cl_openowners);
@@ -971,7 +971,7 @@ parse_ipv4(unsigned int addr_len, char *addr_val, unsigned int *cbaddrp, unsigne
 static void
 gen_callback(struct nfs4_client *clp, struct nfsd4_setclientid *se)
 {
-	struct nfs4_callback *cb = &clp->cl_callback;
+	struct nfs4_cb_conn *cb = &clp->cl_cb_conn;
 
 	/* Currently, we only support tcp for the callback channel */
 	if ((se->se_callback_netid_len != 3) || memcmp((char *)se->se_callback_netid_val, "tcp", 3))
@@ -1691,7 +1691,7 @@ nfsd4_setclientid_confirm(struct svc_rqst *rqstp,
 		else {
 			/* XXX: We just turn off callbacks until we can handle
 			  * change request correctly. */
-			atomic_set(&conf->cl_callback.cb_set, 0);
+			atomic_set(&conf->cl_cb_conn.cb_set, 0);
 			expire_client(unconf);
 			status = nfs_ok;
 
@@ -2425,7 +2425,7 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_sta
 {
 	struct nfs4_delegation *dp;
 	struct nfs4_stateowner *sop = stp->st_stateowner;
-	struct nfs4_callback *cb = &sop->so_client->cl_callback;
+	struct nfs4_cb_conn *cb = &sop->so_client->cl_cb_conn;
 	struct file_lock fl, *flp = &fl;
 	int status, flag = 0;
 
@@ -2617,7 +2617,7 @@ nfsd4_renew(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	renew_client(clp);
 	status = nfserr_cb_path_down;
 	if (!list_empty(&clp->cl_delegations)
-			&& !atomic_read(&clp->cl_callback.cb_set))
+			&& !atomic_read(&clp->cl_cb_conn.cb_set))
 		goto out;
 	status = nfs_ok;
 out:
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 8d882a3eb4b9..563c367a3013 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -88,7 +88,7 @@ struct nfs4_delegation {
 #define dl_fh           dl_recall.cbr_fh
 
 /* client delegation callback info */
-struct nfs4_callback {
+struct nfs4_cb_conn {
 	/* SETCLIENTID info */
 	u32                     cb_addr;
 	unsigned short          cb_port;
@@ -186,7 +186,7 @@ struct nfs4_client {
 	struct svc_cred		cl_cred; 	/* setclientid principal */
 	clientid_t		cl_clientid;	/* generated by server */
 	nfs4_verifier		cl_confirm;	/* generated by server */
-	struct nfs4_callback	cl_callback;    /* callback info */
+	struct nfs4_cb_conn	cl_cb_conn;     /* callback info */
 	atomic_t		cl_count;	/* ref count */
 	u32			cl_firststate;	/* recovery dir creation */
 
-- 
cgit v1.2.3-59-g8ed1b


From b53d40c5070bffde1b2bcaf848412a50d8894794 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Fri, 1 May 2009 19:50:00 -0400
Subject: nfsd4: eliminate struct nfs4_cb_recall

The nfs4_cb_recall struct is used only in nfs4_delegation, so its
pointer to the containing delegation is unnecessary--we could just use
container_of().

But there's no real reason to have this a separate struct at all--just
move these fields to nfs4_delegation.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4callback.c     | 24 +++++++++++-------------
 fs/nfsd/nfs4state.c        |  5 ++---
 include/linux/nfsd/state.h | 18 +++++-------------
 3 files changed, 18 insertions(+), 29 deletions(-)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index ed860d7ddd19..2509305f6f53 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -215,18 +215,18 @@ encode_cb_compound_hdr(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr)
 }
 
 static int
-encode_cb_recall(struct xdr_stream *xdr, struct nfs4_cb_recall *cb_rec)
+encode_cb_recall(struct xdr_stream *xdr, struct nfs4_delegation *dp)
 {
 	__be32 *p;
-	int len = cb_rec->cbr_fh.fh_size;
+	int len = dp->dl_fh.fh_size;
 
-	RESERVE_SPACE(12+sizeof(cb_rec->cbr_stateid) + len);
+	RESERVE_SPACE(12+sizeof(dp->dl_stateid) + len);
 	WRITE32(OP_CB_RECALL);
-	WRITE32(cb_rec->cbr_stateid.si_generation);
-	WRITEMEM(&cb_rec->cbr_stateid.si_opaque, sizeof(stateid_opaque_t));
-	WRITE32(cb_rec->cbr_trunc);
+	WRITE32(dp->dl_stateid.si_generation);
+	WRITEMEM(&dp->dl_stateid.si_opaque, sizeof(stateid_opaque_t));
+	WRITE32(dp->dl_trunc);
 	WRITE32(len);
-	WRITEMEM(&cb_rec->cbr_fh.fh_base, len);
+	WRITEMEM(&dp->dl_fh.fh_base, len);
 	return 0;
 }
 
@@ -241,11 +241,11 @@ nfs4_xdr_enc_cb_null(struct rpc_rqst *req, __be32 *p)
 }
 
 static int
-nfs4_xdr_enc_cb_recall(struct rpc_rqst *req, __be32 *p, struct nfs4_cb_recall *args)
+nfs4_xdr_enc_cb_recall(struct rpc_rqst *req, __be32 *p, struct nfs4_delegation *args)
 {
 	struct xdr_stream xdr;
 	struct nfs4_cb_compound_hdr hdr = {
-		.ident = args->cbr_ident,
+		.ident = args->dl_ident,
 		.nops   = 1,
 	};
 
@@ -502,17 +502,15 @@ nfsd4_cb_recall(struct nfs4_delegation *dp)
 {
 	struct nfs4_client *clp = dp->dl_client;
 	struct rpc_clnt *clnt = clp->cl_cb_conn.cb_client;
-	struct nfs4_cb_recall *cbr = &dp->dl_recall;
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_cb_procedures[NFSPROC4_CLNT_CB_RECALL],
-		.rpc_argp = cbr,
+		.rpc_argp = dp,
 		.rpc_cred = clp->cl_cb_conn.cb_cred
 	};
 	int retries = 1;
 	int status = 0;
 
-	cbr->cbr_trunc = 0; /* XXX need to implement truncate optimization */
-	cbr->cbr_dp = dp;
+	dp->dl_trunc = 0; /* XXX need to implement truncate optimization */
 
 	status = rpc_call_sync(clnt, &msg, RPC_TASK_SOFT);
 	while (retries--) {
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index d7b5e6b89207..3e5345e01b13 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -203,9 +203,8 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f
 	get_file(stp->st_vfs_file);
 	dp->dl_vfs_file = stp->st_vfs_file;
 	dp->dl_type = type;
-	dp->dl_recall.cbr_dp = NULL;
-	dp->dl_recall.cbr_ident = cb->cb_ident;
-	dp->dl_recall.cbr_trunc = 0;
+	dp->dl_ident = cb->cb_ident;
+	dp->dl_trunc = 0;
 	dp->dl_stateid.si_boot = get_seconds();
 	dp->dl_stateid.si_stateownerid = current_delegid++;
 	dp->dl_stateid.si_fileid = 0;
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 563c367a3013..233b60d39b84 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -61,15 +61,6 @@ typedef struct {
 #define si_stateownerid   si_opaque.so_stateownerid
 #define si_fileid         si_opaque.so_fileid
 
-
-struct nfs4_cb_recall {
-	u32			cbr_ident;
-	int			cbr_trunc;
-	stateid_t		cbr_stateid;
-	struct knfsd_fh		cbr_fh;
-	struct nfs4_delegation	*cbr_dp;
-};
-
 struct nfs4_delegation {
 	struct list_head	dl_perfile;
 	struct list_head	dl_perclnt;
@@ -81,12 +72,13 @@ struct nfs4_delegation {
 	struct file		*dl_vfs_file;
 	u32			dl_type;
 	time_t			dl_time;
-	struct nfs4_cb_recall	dl_recall;
+/* For recall: */
+	u32			dl_ident;
+	int			dl_trunc;
+	stateid_t		dl_stateid;
+	struct knfsd_fh		dl_fh;
 };
 
-#define dl_stateid      dl_recall.cbr_stateid
-#define dl_fh           dl_recall.cbr_fh
-
 /* client delegation callback info */
 struct nfs4_cb_conn {
 	/* SETCLIENTID info */
-- 
cgit v1.2.3-59-g8ed1b


From 6707bd3d420f53ae8f090dac871843f6f43c9980 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Fri, 1 May 2009 19:57:46 -0400
Subject: nfsd4: remove unused dl_trunc

There's no point in keeping this field around--it's always zero.

(Background: the protocol allows you to tell the client that the file is
about to be truncated, as an optimization to save the client from
writing back dirty pages that will just be discarded.  We don't
implement this hint.  If we do some day, adding this field back in will
be the least of the work involved.)

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4callback.c     | 4 +---
 fs/nfsd/nfs4state.c        | 1 -
 include/linux/nfsd/state.h | 1 -
 3 files changed, 1 insertion(+), 5 deletions(-)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 2509305f6f53..0420b5e6e20d 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -224,7 +224,7 @@ encode_cb_recall(struct xdr_stream *xdr, struct nfs4_delegation *dp)
 	WRITE32(OP_CB_RECALL);
 	WRITE32(dp->dl_stateid.si_generation);
 	WRITEMEM(&dp->dl_stateid.si_opaque, sizeof(stateid_opaque_t));
-	WRITE32(dp->dl_trunc);
+	WRITE32(0); /* truncate optimization not implemented */
 	WRITE32(len);
 	WRITEMEM(&dp->dl_fh.fh_base, len);
 	return 0;
@@ -510,8 +510,6 @@ nfsd4_cb_recall(struct nfs4_delegation *dp)
 	int retries = 1;
 	int status = 0;
 
-	dp->dl_trunc = 0; /* XXX need to implement truncate optimization */
-
 	status = rpc_call_sync(clnt, &msg, RPC_TASK_SOFT);
 	while (retries--) {
 		switch (status) {
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 3e5345e01b13..cbb16e191d5b 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -204,7 +204,6 @@ alloc_init_deleg(struct nfs4_client *clp, struct nfs4_stateid *stp, struct svc_f
 	dp->dl_vfs_file = stp->st_vfs_file;
 	dp->dl_type = type;
 	dp->dl_ident = cb->cb_ident;
-	dp->dl_trunc = 0;
 	dp->dl_stateid.si_boot = get_seconds();
 	dp->dl_stateid.si_stateownerid = current_delegid++;
 	dp->dl_stateid.si_fileid = 0;
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 233b60d39b84..346b603072ce 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -74,7 +74,6 @@ struct nfs4_delegation {
 	time_t			dl_time;
 /* For recall: */
 	u32			dl_ident;
-	int			dl_trunc;
 	stateid_t		dl_stateid;
 	struct knfsd_fh		dl_fh;
 };
-- 
cgit v1.2.3-59-g8ed1b


From 3aea09dc9106407d8bc18e593fbffda9ad632844 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Fri, 1 May 2009 20:11:12 -0400
Subject: nfsd4: track recall retries in nfs4_delegation

Move this out of a local variable into the nfs4_delegation object in
preparation for making this an async rpc call (at which point we'll need
any state like this in a common object that's preserved across function
calls).

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4callback.c     | 4 ++--
 include/linux/nfsd/state.h | 1 +
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 0420b5e6e20d..b88b207d75d9 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -507,11 +507,11 @@ nfsd4_cb_recall(struct nfs4_delegation *dp)
 		.rpc_argp = dp,
 		.rpc_cred = clp->cl_cb_conn.cb_cred
 	};
-	int retries = 1;
 	int status = 0;
 
+	dp->dl_retries = 1;
 	status = rpc_call_sync(clnt, &msg, RPC_TASK_SOFT);
-	while (retries--) {
+	while (dp->dl_retries--) {
 		switch (status) {
 			case -EIO:
 				/* Network partition? */
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 346b603072ce..c0c49215ddc5 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -76,6 +76,7 @@ struct nfs4_delegation {
 	u32			dl_ident;
 	stateid_t		dl_stateid;
 	struct knfsd_fh		dl_fh;
+	int			dl_retries;
 };
 
 /* client delegation callback info */
-- 
cgit v1.2.3-59-g8ed1b


From 63e4863fabc6e165a6ca813051305be58966da45 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Fri, 1 May 2009 22:36:55 -0400
Subject: nfsd4: make recall callback an asynchronous rpc

As with the probe, this removes the need for another kthread.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4callback.c | 74 +++++++++++++++++++++++++++++++++-----------------
 fs/nfsd/nfs4state.c    | 28 ++-----------------
 2 files changed, 52 insertions(+), 50 deletions(-)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index b88b207d75d9..f4fab69a8c30 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -494,6 +494,49 @@ nfsd4_probe_callback(struct nfs4_client *clp)
 	do_probe_callback(clp);
 }
 
+static void nfsd4_cb_recall_done(struct rpc_task *task, void *calldata)
+{
+	struct nfs4_delegation *dp = calldata;
+	struct nfs4_client *clp = dp->dl_client;
+
+	switch (task->tk_status) {
+	case -EIO:
+		/* Network partition? */
+		atomic_set(&clp->cl_cb_conn.cb_set, 0);
+		warn_no_callback_path(clp, task->tk_status);
+	case -EBADHANDLE:
+	case -NFS4ERR_BAD_STATEID:
+		/* Race: client probably got cb_recall
+		 * before open reply granting delegation */
+		break;
+	default:
+		/* success, or error we can't handle */
+		return;
+	}
+	if (dp->dl_retries--) {
+		rpc_delay(task, 2*HZ);
+		task->tk_status = 0;
+		rpc_restart_call(task);
+	} else {
+		atomic_set(&clp->cl_cb_conn.cb_set, 0);
+		warn_no_callback_path(clp, task->tk_status);
+	}
+}
+
+static void nfsd4_cb_recall_release(void *calldata)
+{
+	struct nfs4_delegation *dp = calldata;
+	struct nfs4_client *clp = dp->dl_client;
+
+	nfs4_put_delegation(dp);
+	put_nfs4_client(clp);
+}
+
+static const struct rpc_call_ops nfsd4_cb_recall_ops = {
+	.rpc_call_done = nfsd4_cb_recall_done,
+	.rpc_release = nfsd4_cb_recall_release,
+};
+
 /*
  * called with dp->dl_count inc'ed.
  */
@@ -507,32 +550,13 @@ nfsd4_cb_recall(struct nfs4_delegation *dp)
 		.rpc_argp = dp,
 		.rpc_cred = clp->cl_cb_conn.cb_cred
 	};
-	int status = 0;
+	int status;
 
 	dp->dl_retries = 1;
-	status = rpc_call_sync(clnt, &msg, RPC_TASK_SOFT);
-	while (dp->dl_retries--) {
-		switch (status) {
-			case -EIO:
-				/* Network partition? */
-				atomic_set(&clp->cl_cb_conn.cb_set, 0);
-			case -EBADHANDLE:
-			case -NFS4ERR_BAD_STATEID:
-				/* Race: client probably got cb_recall
-				 * before open reply granting delegation */
-				break;
-			default:
-				goto out_put_cred;
-		}
-		ssleep(2);
-		status = rpc_call_sync(clnt, &msg, RPC_TASK_SOFT);
+	status = rpc_call_async(clnt, &msg, RPC_TASK_SOFT,
+				&nfsd4_cb_recall_ops, dp);
+	if (status) {
+		put_nfs4_client(clp);
+		nfs4_put_delegation(dp);
 	}
-out_put_cred:
-	/*
-	 * Success or failure, now we're either waiting for lease expiration
-	 * or deleg_return.
-	 */
-	put_nfs4_client(clp);
-	nfs4_put_delegation(dp);
-	return;
 }
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index cbb16e191d5b..a4bdf2589b41 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -2059,19 +2059,6 @@ nfs4_file_downgrade(struct file *filp, unsigned int share_access)
 	}
 }
 
-/*
- * Recall a delegation
- */
-static int
-do_recall(void *__dp)
-{
-	struct nfs4_delegation *dp = __dp;
-
-	dp->dl_file->fi_had_conflict = true;
-	nfsd4_cb_recall(dp);
-	return 0;
-}
-
 /*
  * Spawn a thread to perform a recall on the delegation represented
  * by the lease (file_lock)
@@ -2083,8 +2070,7 @@ do_recall(void *__dp)
 static
 void nfsd_break_deleg_cb(struct file_lock *fl)
 {
-	struct nfs4_delegation *dp=  (struct nfs4_delegation *)fl->fl_owner;
-	struct task_struct *t;
+	struct nfs4_delegation *dp = (struct nfs4_delegation *)fl->fl_owner;
 
 	dprintk("NFSD nfsd_break_deleg_cb: dp %p fl %p\n",dp,fl);
 	if (!dp)
@@ -2112,16 +2098,8 @@ void nfsd_break_deleg_cb(struct file_lock *fl)
 	 */
 	fl->fl_break_time = 0;
 
-	t = kthread_run(do_recall, dp, "%s", "nfs4_cb_recall");
-	if (IS_ERR(t)) {
-		struct nfs4_client *clp = dp->dl_client;
-
-		printk(KERN_INFO "NFSD: Callback thread failed for "
-			"for client (clientid %08x/%08x)\n",
-			clp->cl_clientid.cl_boot, clp->cl_clientid.cl_id);
-		put_nfs4_client(dp->dl_client);
-		nfs4_put_delegation(dp);
-	}
+	dp->dl_file->fi_had_conflict = true;
+	nfsd4_cb_recall(dp);
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 9064caae8f47bb7ed5d91712e81f01c1aba2fa3c Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Tue, 28 Apr 2009 16:48:25 -0700
Subject: nfsd: use C99 struct initializers

Eliminate 56 sparse warnings like this one:

fs/nfsd/nfs4xdr.c:1331:15: warning: obsolete array initializer, use C99 syntax

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4xdr.c | 112 +++++++++++++++++++++++++++---------------------------
 1 file changed, 56 insertions(+), 56 deletions(-)

diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 12d36a7361cd..ab005fc637e1 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -1318,64 +1318,64 @@ static nfsd4_dec nfsd4_dec_ops[] = {
 };
 
 static nfsd4_dec nfsd41_dec_ops[] = {
-	[OP_ACCESS]		(nfsd4_dec)nfsd4_decode_access,
-	[OP_CLOSE]		(nfsd4_dec)nfsd4_decode_close,
-	[OP_COMMIT]		(nfsd4_dec)nfsd4_decode_commit,
-	[OP_CREATE]		(nfsd4_dec)nfsd4_decode_create,
-	[OP_DELEGPURGE]		(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_DELEGRETURN]	(nfsd4_dec)nfsd4_decode_delegreturn,
-	[OP_GETATTR]		(nfsd4_dec)nfsd4_decode_getattr,
-	[OP_GETFH]		(nfsd4_dec)nfsd4_decode_noop,
-	[OP_LINK]		(nfsd4_dec)nfsd4_decode_link,
-	[OP_LOCK]		(nfsd4_dec)nfsd4_decode_lock,
-	[OP_LOCKT]		(nfsd4_dec)nfsd4_decode_lockt,
-	[OP_LOCKU]		(nfsd4_dec)nfsd4_decode_locku,
-	[OP_LOOKUP]		(nfsd4_dec)nfsd4_decode_lookup,
-	[OP_LOOKUPP]		(nfsd4_dec)nfsd4_decode_noop,
-	[OP_NVERIFY]		(nfsd4_dec)nfsd4_decode_verify,
-	[OP_OPEN]		(nfsd4_dec)nfsd4_decode_open,
-	[OP_OPENATTR]		(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_OPEN_CONFIRM]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_OPEN_DOWNGRADE]	(nfsd4_dec)nfsd4_decode_open_downgrade,
-	[OP_PUTFH]		(nfsd4_dec)nfsd4_decode_putfh,
-	[OP_PUTPUBFH]		(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_PUTROOTFH]		(nfsd4_dec)nfsd4_decode_noop,
-	[OP_READ]		(nfsd4_dec)nfsd4_decode_read,
-	[OP_READDIR]		(nfsd4_dec)nfsd4_decode_readdir,
-	[OP_READLINK]		(nfsd4_dec)nfsd4_decode_noop,
-	[OP_REMOVE]		(nfsd4_dec)nfsd4_decode_remove,
-	[OP_RENAME]		(nfsd4_dec)nfsd4_decode_rename,
-	[OP_RENEW]		(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_RESTOREFH]		(nfsd4_dec)nfsd4_decode_noop,
-	[OP_SAVEFH]		(nfsd4_dec)nfsd4_decode_noop,
-	[OP_SECINFO]		(nfsd4_dec)nfsd4_decode_secinfo,
-	[OP_SETATTR]		(nfsd4_dec)nfsd4_decode_setattr,
-	[OP_SETCLIENTID]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_SETCLIENTID_CONFIRM](nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_VERIFY]		(nfsd4_dec)nfsd4_decode_verify,
-	[OP_WRITE]		(nfsd4_dec)nfsd4_decode_write,
-	[OP_RELEASE_LOCKOWNER]	(nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_ACCESS]		= (nfsd4_dec)nfsd4_decode_access,
+	[OP_CLOSE]		= (nfsd4_dec)nfsd4_decode_close,
+	[OP_COMMIT]		= (nfsd4_dec)nfsd4_decode_commit,
+	[OP_CREATE]		= (nfsd4_dec)nfsd4_decode_create,
+	[OP_DELEGPURGE]		= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_DELEGRETURN]	= (nfsd4_dec)nfsd4_decode_delegreturn,
+	[OP_GETATTR]		= (nfsd4_dec)nfsd4_decode_getattr,
+	[OP_GETFH]		= (nfsd4_dec)nfsd4_decode_noop,
+	[OP_LINK]		= (nfsd4_dec)nfsd4_decode_link,
+	[OP_LOCK]		= (nfsd4_dec)nfsd4_decode_lock,
+	[OP_LOCKT]		= (nfsd4_dec)nfsd4_decode_lockt,
+	[OP_LOCKU]		= (nfsd4_dec)nfsd4_decode_locku,
+	[OP_LOOKUP]		= (nfsd4_dec)nfsd4_decode_lookup,
+	[OP_LOOKUPP]		= (nfsd4_dec)nfsd4_decode_noop,
+	[OP_NVERIFY]		= (nfsd4_dec)nfsd4_decode_verify,
+	[OP_OPEN]		= (nfsd4_dec)nfsd4_decode_open,
+	[OP_OPENATTR]		= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_OPEN_CONFIRM]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_OPEN_DOWNGRADE]	= (nfsd4_dec)nfsd4_decode_open_downgrade,
+	[OP_PUTFH]		= (nfsd4_dec)nfsd4_decode_putfh,
+	[OP_PUTPUBFH]		= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_PUTROOTFH]		= (nfsd4_dec)nfsd4_decode_noop,
+	[OP_READ]		= (nfsd4_dec)nfsd4_decode_read,
+	[OP_READDIR]		= (nfsd4_dec)nfsd4_decode_readdir,
+	[OP_READLINK]		= (nfsd4_dec)nfsd4_decode_noop,
+	[OP_REMOVE]		= (nfsd4_dec)nfsd4_decode_remove,
+	[OP_RENAME]		= (nfsd4_dec)nfsd4_decode_rename,
+	[OP_RENEW]		= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_RESTOREFH]		= (nfsd4_dec)nfsd4_decode_noop,
+	[OP_SAVEFH]		= (nfsd4_dec)nfsd4_decode_noop,
+	[OP_SECINFO]		= (nfsd4_dec)nfsd4_decode_secinfo,
+	[OP_SETATTR]		= (nfsd4_dec)nfsd4_decode_setattr,
+	[OP_SETCLIENTID]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_SETCLIENTID_CONFIRM]= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_VERIFY]		= (nfsd4_dec)nfsd4_decode_verify,
+	[OP_WRITE]		= (nfsd4_dec)nfsd4_decode_write,
+	[OP_RELEASE_LOCKOWNER]	= (nfsd4_dec)nfsd4_decode_notsupp,
 
 	/* new operations for NFSv4.1 */
-	[OP_BACKCHANNEL_CTL]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_BIND_CONN_TO_SESSION](nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_EXCHANGE_ID]	(nfsd4_dec)nfsd4_decode_exchange_id,
-	[OP_CREATE_SESSION]	(nfsd4_dec)nfsd4_decode_create_session,
-	[OP_DESTROY_SESSION]	(nfsd4_dec)nfsd4_decode_destroy_session,
-	[OP_FREE_STATEID]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_GET_DIR_DELEGATION]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_GETDEVICEINFO]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_GETDEVICELIST]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_LAYOUTCOMMIT]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_LAYOUTGET]		(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_LAYOUTRETURN]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_SECINFO_NO_NAME]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_SEQUENCE]		(nfsd4_dec)nfsd4_decode_sequence,
-	[OP_SET_SSV]		(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_TEST_STATEID]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_WANT_DELEGATION]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_DESTROY_CLIENTID]	(nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_RECLAIM_COMPLETE]	(nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_BACKCHANNEL_CTL]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_BIND_CONN_TO_SESSION]= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_EXCHANGE_ID]	= (nfsd4_dec)nfsd4_decode_exchange_id,
+	[OP_CREATE_SESSION]	= (nfsd4_dec)nfsd4_decode_create_session,
+	[OP_DESTROY_SESSION]	= (nfsd4_dec)nfsd4_decode_destroy_session,
+	[OP_FREE_STATEID]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_GET_DIR_DELEGATION]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_GETDEVICEINFO]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_GETDEVICELIST]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_LAYOUTCOMMIT]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_LAYOUTGET]		= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_LAYOUTRETURN]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_SECINFO_NO_NAME]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_SEQUENCE]		= (nfsd4_dec)nfsd4_decode_sequence,
+	[OP_SET_SSV]		= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_TEST_STATEID]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_WANT_DELEGATION]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_DESTROY_CLIENTID]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_RECLAIM_COMPLETE]	= (nfsd4_dec)nfsd4_decode_notsupp,
 };
 
 struct nfsd4_minorversion_ops {
-- 
cgit v1.2.3-59-g8ed1b


From 02cb2858dbfe216bd4a91c4afb4e259ef3b9e151 Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Fri, 24 Apr 2009 15:41:57 +0800
Subject: nfsd: nfs4_stat_init cleanup

Save some loop time.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4state.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index a4bdf2589b41..b276624be66e 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -4002,6 +4002,7 @@ nfs4_state_init(void)
 		INIT_LIST_HEAD(&conf_str_hashtbl[i]);
 		INIT_LIST_HEAD(&unconf_str_hashtbl[i]);
 		INIT_LIST_HEAD(&unconf_id_hashtbl[i]);
+		INIT_LIST_HEAD(&reclaim_str_hashtbl[i]);
 	}
 	for (i = 0; i < SESSION_HASH_SIZE; i++)
 		INIT_LIST_HEAD(&sessionid_hashtbl[i]);
@@ -4024,8 +4025,6 @@ nfs4_state_init(void)
 	INIT_LIST_HEAD(&close_lru);
 	INIT_LIST_HEAD(&client_lru);
 	INIT_LIST_HEAD(&del_recall_lru);
-	for (i = 0; i < CLIENT_HASH_SIZE; i++)
-		INIT_LIST_HEAD(&reclaim_str_hashtbl[i]);
 	reclaim_str_hashtbl_size = 0;
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From fa3b6dcd5298db2e7b63c17795c9e5570d3df8d9 Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Fri, 8 May 2009 10:33:38 +0800
Subject: VT-d: fix invalid domain id for KVM context flush

The domain->id is a sequence number associated with the KVM guest
and should not be used for the context flush. This patch replaces
the domain->id with a proper id value for both bare metal and KVM.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Acked-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index d3edd6aa82ce..d6f4ee50924c 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1429,7 +1429,7 @@ static int domain_context_mapping_one(struct dmar_domain *domain, int segment,
 	domain_flush_cache(domain, context, sizeof(*context));
 
 	/* it's a non-present to present mapping */
-	if (iommu->flush.flush_context(iommu, domain->id,
+	if (iommu->flush.flush_context(iommu, id,
 		(((u16)bus) << 8) | devfn, DMA_CCMD_MASK_NOBIT,
 		DMA_CCMD_DEVICE_INVL, 1))
 		iommu_flush_write_buffer(iommu);
-- 
cgit v1.2.3-59-g8ed1b


From 4c25a2c1b90bf785fc2e2f0f0c74a80b3e070d39 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 10 May 2009 17:16:06 +0100
Subject: intel-iommu: Clean up handling of "caching mode" vs. context
 flushing.

It really doesn't make a lot of sense to have some of the logic to
handle caching vs. non-caching mode duplicated in qi_flush_context() and
__iommu_flush_context(), while the return value indicates whether the
caller should take other action which depends on the same thing.

Especially since qi_flush_context() thought it was returning something
entirely different anyway.

This patch makes qi_flush_context() and __iommu_flush_context() both
return void, removes the 'non_present_entry_flush' argument and makes
the only call site which _set_ that argument to 1 do the right thing.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/dmar.c          | 13 +++---------
 drivers/pci/intel-iommu.c   | 52 ++++++++++++++++++---------------------------
 include/linux/intel-iommu.h |  8 +++----
 3 files changed, 28 insertions(+), 45 deletions(-)

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index d3d86b749eee..10a071ba3232 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -723,23 +723,16 @@ void qi_global_iec(struct intel_iommu *iommu)
 	qi_submit_sync(&desc, iommu);
 }
 
-int qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid, u8 fm,
-		     u64 type, int non_present_entry_flush)
+void qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid, u8 fm,
+		      u64 type)
 {
 	struct qi_desc desc;
 
-	if (non_present_entry_flush) {
-		if (!cap_caching_mode(iommu->cap))
-			return 1;
-		else
-			did = 0;
-	}
-
 	desc.low = QI_CC_FM(fm) | QI_CC_SID(sid) | QI_CC_DID(did)
 			| QI_CC_GRAN(type) | QI_CC_TYPE;
 	desc.high = 0;
 
-	return qi_submit_sync(&desc, iommu);
+	qi_submit_sync(&desc, iommu);
 }
 
 int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index d6f4ee50924c..9f5d9151edc9 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -857,26 +857,13 @@ static void iommu_flush_write_buffer(struct intel_iommu *iommu)
 }
 
 /* return value determine if we need a write buffer flush */
-static int __iommu_flush_context(struct intel_iommu *iommu,
-	u16 did, u16 source_id, u8 function_mask, u64 type,
-	int non_present_entry_flush)
+static void __iommu_flush_context(struct intel_iommu *iommu,
+				  u16 did, u16 source_id, u8 function_mask,
+				  u64 type)
 {
 	u64 val = 0;
 	unsigned long flag;
 
-	/*
-	 * In the non-present entry flush case, if hardware doesn't cache
-	 * non-present entry we do nothing and if hardware cache non-present
-	 * entry, we flush entries of domain 0 (the domain id is used to cache
-	 * any non-present entries)
-	 */
-	if (non_present_entry_flush) {
-		if (!cap_caching_mode(iommu->cap))
-			return 1;
-		else
-			did = 0;
-	}
-
 	switch (type) {
 	case DMA_CCMD_GLOBAL_INVL:
 		val = DMA_CCMD_GLOBAL_INVL;
@@ -901,9 +888,6 @@ static int __iommu_flush_context(struct intel_iommu *iommu,
 		dmar_readq, (!(val & DMA_CCMD_ICC)), val);
 
 	spin_unlock_irqrestore(&iommu->register_lock, flag);
-
-	/* flush context entry will implicitly flush write buffer */
-	return 0;
 }
 
 /* return value determine if we need a write buffer flush */
@@ -1428,14 +1412,21 @@ static int domain_context_mapping_one(struct dmar_domain *domain, int segment,
 	context_set_present(context);
 	domain_flush_cache(domain, context, sizeof(*context));
 
-	/* it's a non-present to present mapping */
-	if (iommu->flush.flush_context(iommu, id,
-		(((u16)bus) << 8) | devfn, DMA_CCMD_MASK_NOBIT,
-		DMA_CCMD_DEVICE_INVL, 1))
-		iommu_flush_write_buffer(iommu);
-	else
+	/*
+	 * It's a non-present to present mapping. If hardware doesn't cache
+	 * non-present entry we only need to flush the write-buffer. If the
+	 * _does_ cache non-present entries, then it does so in the special
+	 * domain #0, which we have to flush:
+	 */
+	if (cap_caching_mode(iommu->cap)) {
+		iommu->flush.flush_context(iommu, 0,
+					   (((u16)bus) << 8) | devfn,
+					   DMA_CCMD_MASK_NOBIT,
+					   DMA_CCMD_DEVICE_INVL);
 		iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_DSI_FLUSH, 0);
-
+	} else {
+		iommu_flush_write_buffer(iommu);
+	}
 	spin_unlock_irqrestore(&iommu->lock, flags);
 
 	spin_lock_irqsave(&domain->iommu_lock, flags);
@@ -1566,7 +1557,7 @@ static void iommu_detach_dev(struct intel_iommu *iommu, u8 bus, u8 devfn)
 
 	clear_context_table(iommu, bus, devfn);
 	iommu->flush.flush_context(iommu, 0, 0, 0,
-					   DMA_CCMD_GLOBAL_INVL, 0);
+					   DMA_CCMD_GLOBAL_INVL);
 	iommu->flush.flush_iotlb(iommu, 0, 0, 0,
 					 DMA_TLB_GLOBAL_FLUSH, 0);
 }
@@ -2104,8 +2095,7 @@ static int __init init_dmars(void)
 
 		iommu_set_root_entry(iommu);
 
-		iommu->flush.flush_context(iommu, 0, 0, 0, DMA_CCMD_GLOBAL_INVL,
-					   0);
+		iommu->flush.flush_context(iommu, 0, 0, 0, DMA_CCMD_GLOBAL_INVL);
 		iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_GLOBAL_FLUSH,
 					 0);
 		iommu_disable_protect_mem_regions(iommu);
@@ -2721,7 +2711,7 @@ static int init_iommu_hw(void)
 		iommu_set_root_entry(iommu);
 
 		iommu->flush.flush_context(iommu, 0, 0, 0,
-						DMA_CCMD_GLOBAL_INVL, 0);
+						DMA_CCMD_GLOBAL_INVL);
 		iommu->flush.flush_iotlb(iommu, 0, 0, 0,
 						DMA_TLB_GLOBAL_FLUSH, 0);
 		iommu_disable_protect_mem_regions(iommu);
@@ -2738,7 +2728,7 @@ static void iommu_flush_all(void)
 
 	for_each_active_iommu(iommu, drhd) {
 		iommu->flush.flush_context(iommu, 0, 0, 0,
-						DMA_CCMD_GLOBAL_INVL, 0);
+						DMA_CCMD_GLOBAL_INVL);
 		iommu->flush.flush_iotlb(iommu, 0, 0, 0,
 						DMA_TLB_GLOBAL_FLUSH, 0);
 	}
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 7246971a7feb..f2b94dafbf38 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -281,8 +281,8 @@ struct ir_table {
 #endif
 
 struct iommu_flush {
-	int (*flush_context)(struct intel_iommu *iommu, u16 did, u16 sid, u8 fm,
-		u64 type, int non_present_entry_flush);
+	void (*flush_context)(struct intel_iommu *iommu, u16 did, u16 sid,
+			      u8 fm, u64 type);
 	int (*flush_iotlb)(struct intel_iommu *iommu, u16 did, u64 addr,
 		unsigned int size_order, u64 type, int non_present_entry_flush);
 };
@@ -339,8 +339,8 @@ extern void dmar_disable_qi(struct intel_iommu *iommu);
 extern int dmar_reenable_qi(struct intel_iommu *iommu);
 extern void qi_global_iec(struct intel_iommu *iommu);
 
-extern int qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid,
-			        u8 fm, u64 type, int non_present_entry_flush);
+extern void qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid,
+			     u8 fm, u64 type);
 extern int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
 			  unsigned int size_order, u64 type,
 			  int non_present_entry_flush);
-- 
cgit v1.2.3-59-g8ed1b


From 1f0ef2aa18802a8ce7eb5a5164aaaf4d59073801 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 10 May 2009 19:58:49 +0100
Subject: intel-iommu: Clean up handling of "caching mode" vs. IOTLB flushing.

As we just did for context cache flushing, clean up the logic around
whether we need to flush the iotlb or just the write-buffer, depending
on caching mode.

Fix the same bug in qi_flush_iotlb() that qi_flush_context() had -- it
isn't supposed to be returning an error; it's supposed to be returning a
flag which triggers a write-buffer flush.

Remove some superfluous conditional write-buffer flushes which could
never have happened because they weren't for non-present-to-present
mapping changes anyway.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/dmar.c          | 14 ++------
 drivers/pci/intel-iommu.c   | 78 +++++++++++++++++----------------------------
 include/linux/intel-iommu.h |  9 +++---
 3 files changed, 37 insertions(+), 64 deletions(-)

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index 10a071ba3232..df6af0d4ec03 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -735,22 +735,14 @@ void qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid, u8 fm,
 	qi_submit_sync(&desc, iommu);
 }
 
-int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
-		   unsigned int size_order, u64 type,
-		   int non_present_entry_flush)
+void qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
+		    unsigned int size_order, u64 type)
 {
 	u8 dw = 0, dr = 0;
 
 	struct qi_desc desc;
 	int ih = 0;
 
-	if (non_present_entry_flush) {
-		if (!cap_caching_mode(iommu->cap))
-			return 1;
-		else
-			did = 0;
-	}
-
 	if (cap_write_drain(iommu->cap))
 		dw = 1;
 
@@ -762,7 +754,7 @@ int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
 	desc.high = QI_IOTLB_ADDR(addr) | QI_IOTLB_IH(ih)
 		| QI_IOTLB_AM(size_order);
 
-	return qi_submit_sync(&desc, iommu);
+	qi_submit_sync(&desc, iommu);
 }
 
 /*
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 9f5d9151edc9..f47d04aced87 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -891,27 +891,13 @@ static void __iommu_flush_context(struct intel_iommu *iommu,
 }
 
 /* return value determine if we need a write buffer flush */
-static int __iommu_flush_iotlb(struct intel_iommu *iommu, u16 did,
-	u64 addr, unsigned int size_order, u64 type,
-	int non_present_entry_flush)
+static void __iommu_flush_iotlb(struct intel_iommu *iommu, u16 did,
+				u64 addr, unsigned int size_order, u64 type)
 {
 	int tlb_offset = ecap_iotlb_offset(iommu->ecap);
 	u64 val = 0, val_iva = 0;
 	unsigned long flag;
 
-	/*
-	 * In the non-present entry flush case, if hardware doesn't cache
-	 * non-present entry we do nothing and if hardware cache non-present
-	 * entry, we flush entries of domain 0 (the domain id is used to cache
-	 * any non-present entries)
-	 */
-	if (non_present_entry_flush) {
-		if (!cap_caching_mode(iommu->cap))
-			return 1;
-		else
-			did = 0;
-	}
-
 	switch (type) {
 	case DMA_TLB_GLOBAL_FLUSH:
 		/* global flush doesn't need set IVA_REG */
@@ -959,12 +945,10 @@ static int __iommu_flush_iotlb(struct intel_iommu *iommu, u16 did,
 		pr_debug("IOMMU: tlb flush request %Lx, actual %Lx\n",
 			(unsigned long long)DMA_TLB_IIRG(type),
 			(unsigned long long)DMA_TLB_IAIG(val));
-	/* flush iotlb entry will implicitly flush write buffer */
-	return 0;
 }
 
-static int iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did,
-	u64 addr, unsigned int pages, int non_present_entry_flush)
+static void iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did,
+				  u64 addr, unsigned int pages)
 {
 	unsigned int mask;
 
@@ -974,8 +958,7 @@ static int iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did,
 	/* Fallback to domain selective flush if no PSI support */
 	if (!cap_pgsel_inv(iommu->cap))
 		return iommu->flush.flush_iotlb(iommu, did, 0, 0,
-						DMA_TLB_DSI_FLUSH,
-						non_present_entry_flush);
+						DMA_TLB_DSI_FLUSH);
 
 	/*
 	 * PSI requires page size to be 2 ^ x, and the base address is naturally
@@ -985,11 +968,10 @@ static int iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did,
 	/* Fallback to domain selective flush if size is too big */
 	if (mask > cap_max_amask_val(iommu->cap))
 		return iommu->flush.flush_iotlb(iommu, did, 0, 0,
-			DMA_TLB_DSI_FLUSH, non_present_entry_flush);
+						DMA_TLB_DSI_FLUSH);
 
 	return iommu->flush.flush_iotlb(iommu, did, addr, mask,
-					DMA_TLB_PSI_FLUSH,
-					non_present_entry_flush);
+					DMA_TLB_PSI_FLUSH);
 }
 
 static void iommu_disable_protect_mem_regions(struct intel_iommu *iommu)
@@ -1423,7 +1405,7 @@ static int domain_context_mapping_one(struct dmar_domain *domain, int segment,
 					   (((u16)bus) << 8) | devfn,
 					   DMA_CCMD_MASK_NOBIT,
 					   DMA_CCMD_DEVICE_INVL);
-		iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_DSI_FLUSH, 0);
+		iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_DSI_FLUSH);
 	} else {
 		iommu_flush_write_buffer(iommu);
 	}
@@ -1558,8 +1540,7 @@ static void iommu_detach_dev(struct intel_iommu *iommu, u8 bus, u8 devfn)
 	clear_context_table(iommu, bus, devfn);
 	iommu->flush.flush_context(iommu, 0, 0, 0,
 					   DMA_CCMD_GLOBAL_INVL);
-	iommu->flush.flush_iotlb(iommu, 0, 0, 0,
-					 DMA_TLB_GLOBAL_FLUSH, 0);
+	iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_GLOBAL_FLUSH);
 }
 
 static void domain_remove_dev_info(struct dmar_domain *domain)
@@ -2096,8 +2077,7 @@ static int __init init_dmars(void)
 		iommu_set_root_entry(iommu);
 
 		iommu->flush.flush_context(iommu, 0, 0, 0, DMA_CCMD_GLOBAL_INVL);
-		iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_GLOBAL_FLUSH,
-					 0);
+		iommu->flush.flush_iotlb(iommu, 0, 0, 0, DMA_TLB_GLOBAL_FLUSH);
 		iommu_disable_protect_mem_regions(iommu);
 
 		ret = iommu_enable_translation(iommu);
@@ -2244,10 +2224,11 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr,
 	if (ret)
 		goto error;
 
-	/* it's a non-present to present mapping */
-	ret = iommu_flush_iotlb_psi(iommu, domain->id,
-			start_paddr, size >> VTD_PAGE_SHIFT, 1);
-	if (ret)
+	/* it's a non-present to present mapping. Only flush if caching mode */
+	if (cap_caching_mode(iommu->cap))
+		iommu_flush_iotlb_psi(iommu, 0, start_paddr,
+				      size >> VTD_PAGE_SHIFT);
+	else
 		iommu_flush_write_buffer(iommu);
 
 	return start_paddr + ((u64)paddr & (~PAGE_MASK));
@@ -2283,7 +2264,7 @@ static void flush_unmaps(void)
 
 		if (deferred_flush[i].next) {
 			iommu->flush.flush_iotlb(iommu, 0, 0, 0,
-						 DMA_TLB_GLOBAL_FLUSH, 0);
+						 DMA_TLB_GLOBAL_FLUSH);
 			for (j = 0; j < deferred_flush[i].next; j++) {
 				__free_iova(&deferred_flush[i].domain[j]->iovad,
 						deferred_flush[i].iova[j]);
@@ -2362,9 +2343,8 @@ static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr,
 	/* free page tables */
 	dma_pte_free_pagetable(domain, start_addr, start_addr + size);
 	if (intel_iommu_strict) {
-		if (iommu_flush_iotlb_psi(iommu,
-			domain->id, start_addr, size >> VTD_PAGE_SHIFT, 0))
-			iommu_flush_write_buffer(iommu);
+		iommu_flush_iotlb_psi(iommu, domain->id, start_addr,
+				      size >> VTD_PAGE_SHIFT);
 		/* free iova */
 		__free_iova(&domain->iovad, iova);
 	} else {
@@ -2455,9 +2435,8 @@ static void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist,
 	/* free page tables */
 	dma_pte_free_pagetable(domain, start_addr, start_addr + size);
 
-	if (iommu_flush_iotlb_psi(iommu, domain->id, start_addr,
-			size >> VTD_PAGE_SHIFT, 0))
-		iommu_flush_write_buffer(iommu);
+	iommu_flush_iotlb_psi(iommu, domain->id, start_addr,
+			      size >> VTD_PAGE_SHIFT);
 
 	/* free iova */
 	__free_iova(&domain->iovad, iova);
@@ -2549,10 +2528,13 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne
 		offset += size;
 	}
 
-	/* it's a non-present to present mapping */
-	if (iommu_flush_iotlb_psi(iommu, domain->id,
-			start_addr, offset >> VTD_PAGE_SHIFT, 1))
+	/* it's a non-present to present mapping. Only flush if caching mode */
+	if (cap_caching_mode(iommu->cap))
+		iommu_flush_iotlb_psi(iommu, 0, start_addr,
+				      offset >> VTD_PAGE_SHIFT);
+	else
 		iommu_flush_write_buffer(iommu);
+
 	return nelems;
 }
 
@@ -2711,9 +2693,9 @@ static int init_iommu_hw(void)
 		iommu_set_root_entry(iommu);
 
 		iommu->flush.flush_context(iommu, 0, 0, 0,
-						DMA_CCMD_GLOBAL_INVL);
+					   DMA_CCMD_GLOBAL_INVL);
 		iommu->flush.flush_iotlb(iommu, 0, 0, 0,
-						DMA_TLB_GLOBAL_FLUSH, 0);
+					 DMA_TLB_GLOBAL_FLUSH);
 		iommu_disable_protect_mem_regions(iommu);
 		iommu_enable_translation(iommu);
 	}
@@ -2728,9 +2710,9 @@ static void iommu_flush_all(void)
 
 	for_each_active_iommu(iommu, drhd) {
 		iommu->flush.flush_context(iommu, 0, 0, 0,
-						DMA_CCMD_GLOBAL_INVL);
+					   DMA_CCMD_GLOBAL_INVL);
 		iommu->flush.flush_iotlb(iommu, 0, 0, 0,
-						DMA_TLB_GLOBAL_FLUSH, 0);
+					 DMA_TLB_GLOBAL_FLUSH);
 	}
 }
 
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index f2b94dafbf38..29e05a034c09 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -283,8 +283,8 @@ struct ir_table {
 struct iommu_flush {
 	void (*flush_context)(struct intel_iommu *iommu, u16 did, u16 sid,
 			      u8 fm, u64 type);
-	int (*flush_iotlb)(struct intel_iommu *iommu, u16 did, u64 addr,
-		unsigned int size_order, u64 type, int non_present_entry_flush);
+	void (*flush_iotlb)(struct intel_iommu *iommu, u16 did, u64 addr,
+			    unsigned int size_order, u64 type);
 };
 
 enum {
@@ -341,9 +341,8 @@ extern void qi_global_iec(struct intel_iommu *iommu);
 
 extern void qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid,
 			     u8 fm, u64 type);
-extern int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
-			  unsigned int size_order, u64 type,
-			  int non_present_entry_flush);
+extern void qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
+			  unsigned int size_order, u64 type);
 
 extern int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu);
 
-- 
cgit v1.2.3-59-g8ed1b


From 462b60f6ccc685f7e8aa04ff430e6b4ffedf629f Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 10 May 2009 20:18:18 +0100
Subject: intel-iommu: Fix tiny theoretical race in write-buffer flush.

In iommu_flush_write_buffer() we read iommu->gcmd before taking the
register_lock, and then we mask in the WBF bit and write it to the
register.

There is a tiny chance that something else could have _changed_
iommu->gcmd before we take the lock, but after we read it. So we could
be undoing that change.

Never actually going to have happened in practice, since nothing else
changes that register at runtime -- aside from the write-buffer flush
it's only ever touched at startup for enabling translation, etc.

But worth fixing anyway.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index f47d04aced87..2e2c7406131d 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -844,10 +844,8 @@ static void iommu_flush_write_buffer(struct intel_iommu *iommu)
 
 	if (!rwbf_quirk && !cap_rwbf(iommu->cap))
 		return;
-	val = iommu->gcmd | DMA_GCMD_WBF;
-
 	spin_lock_irqsave(&iommu->register_lock, flag);
-	writel(val, iommu->reg + DMAR_GCMD_REG);
+	writel(iommu->gcmd | DMA_GCMD_WBF, iommu->reg + DMAR_GCMD_REG);
 
 	/* Make sure hardware complete it */
 	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG,
-- 
cgit v1.2.3-59-g8ed1b


From c416daa98a584596df21ee2c26fac6579ee58f57 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 10 May 2009 20:30:58 +0100
Subject: intel-iommu: Tidy up iommu->gcmd handling

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/dmar.c           |  6 ++----
 drivers/pci/intel-iommu.c    | 18 +++++++++---------
 drivers/pci/intr_remapping.c | 11 ++++-------
 3 files changed, 15 insertions(+), 20 deletions(-)

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index df6af0d4ec03..faf77a00cafe 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -784,7 +784,6 @@ void dmar_disable_qi(struct intel_iommu *iommu)
 		cpu_relax();
 
 	iommu->gcmd &= ~DMA_GCMD_QIE;
-
 	writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG);
 
 	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, readl,
@@ -798,7 +797,7 @@ end:
  */
 static void __dmar_enable_qi(struct intel_iommu *iommu)
 {
-	u32 cmd, sts;
+	u32 sts;
 	unsigned long flags;
 	struct q_inval *qi = iommu->qi;
 
@@ -812,9 +811,8 @@ static void __dmar_enable_qi(struct intel_iommu *iommu)
 
 	dmar_writeq(iommu->reg + DMAR_IQA_REG, virt_to_phys(qi->desc));
 
-	cmd = iommu->gcmd | DMA_GCMD_QIE;
 	iommu->gcmd |= DMA_GCMD_QIE;
-	writel(cmd, iommu->reg + DMAR_GCMD_REG);
+	writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG);
 
 	/* Make sure hardware complete it */
 	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, readl, (sts & DMA_GSTS_QIES), sts);
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 2e2c7406131d..bc99b1e47fbc 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -819,7 +819,7 @@ static int iommu_alloc_root_entry(struct intel_iommu *iommu)
 static void iommu_set_root_entry(struct intel_iommu *iommu)
 {
 	void *addr;
-	u32 cmd, sts;
+	u32 sts;
 	unsigned long flag;
 
 	addr = iommu->root_entry;
@@ -827,12 +827,11 @@ static void iommu_set_root_entry(struct intel_iommu *iommu)
 	spin_lock_irqsave(&iommu->register_lock, flag);
 	dmar_writeq(iommu->reg + DMAR_RTADDR_REG, virt_to_phys(addr));
 
-	cmd = iommu->gcmd | DMA_GCMD_SRTP;
-	writel(cmd, iommu->reg + DMAR_GCMD_REG);
+	writel(iommu->gcmd | DMA_GCMD_SRTP, iommu->reg + DMAR_GCMD_REG);
 
 	/* Make sure hardware complete it */
 	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG,
-		readl, (sts & DMA_GSTS_RTPS), sts);
+		      readl, (sts & DMA_GSTS_RTPS), sts);
 
 	spin_unlock_irqrestore(&iommu->register_lock, flag);
 }
@@ -844,12 +843,13 @@ static void iommu_flush_write_buffer(struct intel_iommu *iommu)
 
 	if (!rwbf_quirk && !cap_rwbf(iommu->cap))
 		return;
+
 	spin_lock_irqsave(&iommu->register_lock, flag);
 	writel(iommu->gcmd | DMA_GCMD_WBF, iommu->reg + DMAR_GCMD_REG);
 
 	/* Make sure hardware complete it */
 	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG,
-			readl, (!(val & DMA_GSTS_WBFS)), val);
+		      readl, (!(val & DMA_GSTS_WBFS)), val);
 
 	spin_unlock_irqrestore(&iommu->register_lock, flag);
 }
@@ -995,13 +995,13 @@ static int iommu_enable_translation(struct intel_iommu *iommu)
 	unsigned long flags;
 
 	spin_lock_irqsave(&iommu->register_lock, flags);
-	writel(iommu->gcmd|DMA_GCMD_TE, iommu->reg + DMAR_GCMD_REG);
+	iommu->gcmd |= DMA_GCMD_TE;
+	writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG);
 
 	/* Make sure hardware complete it */
 	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG,
-		readl, (sts & DMA_GSTS_TES), sts);
+		      readl, (sts & DMA_GSTS_TES), sts);
 
-	iommu->gcmd |= DMA_GCMD_TE;
 	spin_unlock_irqrestore(&iommu->register_lock, flags);
 	return 0;
 }
@@ -1017,7 +1017,7 @@ static int iommu_disable_translation(struct intel_iommu *iommu)
 
 	/* Make sure hardware complete it */
 	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG,
-		readl, (!(sts & DMA_GSTS_TES)), sts);
+		      readl, (!(sts & DMA_GSTS_TES)), sts);
 
 	spin_unlock_irqrestore(&iommu->register_lock, flag);
 	return 0;
diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index f5e0ea724a6f..166959614087 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -404,7 +404,7 @@ int free_irte(int irq)
 static void iommu_set_intr_remapping(struct intel_iommu *iommu, int mode)
 {
 	u64 addr;
-	u32 cmd, sts;
+	u32 sts;
 	unsigned long flags;
 
 	addr = virt_to_phys((void *)iommu->ir_table->base);
@@ -415,9 +415,8 @@ static void iommu_set_intr_remapping(struct intel_iommu *iommu, int mode)
 		    (addr) | IR_X2APIC_MODE(mode) | INTR_REMAP_TABLE_REG_SIZE);
 
 	/* Set interrupt-remapping table pointer */
-	cmd = iommu->gcmd | DMA_GCMD_SIRTP;
 	iommu->gcmd |= DMA_GCMD_SIRTP;
-	writel(cmd, iommu->reg + DMAR_GCMD_REG);
+	writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG);
 
 	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG,
 		      readl, (sts & DMA_GSTS_IRTPS), sts);
@@ -427,9 +426,8 @@ static void iommu_set_intr_remapping(struct intel_iommu *iommu, int mode)
 		spin_lock_irqsave(&iommu->register_lock, flags);
 
 		/* enable comaptiblity format interrupt pass through */
-		cmd = iommu->gcmd | DMA_GCMD_CFI;
 		iommu->gcmd |= DMA_GCMD_CFI;
-		writel(cmd, iommu->reg + DMAR_GCMD_REG);
+		writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG);
 
 		IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG,
 			      readl, (sts & DMA_GSTS_CFIS), sts);
@@ -446,9 +444,8 @@ static void iommu_set_intr_remapping(struct intel_iommu *iommu, int mode)
 	spin_lock_irqsave(&iommu->register_lock, flags);
 
 	/* Enable interrupt-remapping */
-	cmd = iommu->gcmd | DMA_GCMD_IRE;
 	iommu->gcmd |= DMA_GCMD_IRE;
-	writel(cmd, iommu->reg + DMAR_GCMD_REG);
+	writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG);
 
 	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG,
 		      readl, (sts & DMA_GSTS_IRES), sts);
-- 
cgit v1.2.3-59-g8ed1b


From dd4dc82d4c129babca9b48b219cff0add5ff4132 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Tue, 12 May 2009 13:28:09 -0700
Subject: lockd: fix FILE_LOCKING=n build error

lockd/svclock.c is missing a header file <linux/fs.h>.

<linux/fs.h> is missing a definition of locks_release_private()
for the config case of FILE_LOCKING=n, causing a build error:

fs/lockd/svclock.c:330: error: implicit declaration of function 'locks_release_private'

lockd without FILE_LOCKING doesn't make sense, so make LOCKD and LOCKD_V4
depend on FILE_LOCKING, and make NFS depend on FILE_LOCKING.

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/Kconfig     | 2 ++
 fs/nfs/Kconfig | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/Kconfig b/fs/Kconfig
index 9f7270f36b2a..44ab328ceb2a 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -226,10 +226,12 @@ source "fs/nfsd/Kconfig"
 
 config LOCKD
 	tristate
+	depends on FILE_LOCKING
 
 config LOCKD_V4
 	bool
 	depends on NFSD_V3 || NFS_V3
+	depends on FILE_LOCKING
 	default y
 
 config EXPORTFS
diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig
index e67f3ec07736..7dbb8f27b9d6 100644
--- a/fs/nfs/Kconfig
+++ b/fs/nfs/Kconfig
@@ -1,6 +1,6 @@
 config NFS_FS
 	tristate "NFS client support"
-	depends on INET
+	depends on INET && FILE_LOCKING
 	select LOCKD
 	select SUNRPC
 	select NFS_ACL_SUPPORT if NFS_V3_ACL
-- 
cgit v1.2.3-59-g8ed1b


From dd7264355a203c3456dbba04db471947d3b55e7e Mon Sep 17 00:00:00 2001
From: Chris Wright <chrisw@sous-sol.org>
Date: Wed, 13 May 2009 15:55:52 -0700
Subject: intel-iommu: dmar_set_interrupt return error value

dmar_set_interrupt feigns success when arch_setup_dmar_msi
fails, return error value.

Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/dmar.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index faf77a00cafe..f23460a5d106 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -1088,7 +1088,7 @@ int dmar_set_interrupt(struct intel_iommu *iommu)
 		set_irq_data(irq, NULL);
 		iommu->irq = 0;
 		destroy_irq(irq);
-		return 0;
+		return ret;
 	}
 
 	ret = request_irq(irq, dmar_fault, 0, iommu->name, iommu);
-- 
cgit v1.2.3-59-g8ed1b


From 302b4215daa0a704c843da40fd2529e5757a72da Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Mon, 18 May 2009 13:51:32 +0800
Subject: PCI: support the ATS capability

The PCIe ATS capability makes the Endpoint be able to request the
DMA address translation from the IOMMU and cache the translation
in the device side, thus alleviate IOMMU pressure and improve the
hardware performance in the I/O virtualization environment.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/iov.c        | 105 +++++++++++++++++++++++++++++++++++++++++++++++
 drivers/pci/pci.h        |  37 +++++++++++++++++
 include/linux/pci.h      |   2 +
 include/linux/pci_regs.h |  10 +++++
 4 files changed, 154 insertions(+)

diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c
index b497daab3d4a..0a7a1b40286f 100644
--- a/drivers/pci/iov.c
+++ b/drivers/pci/iov.c
@@ -5,6 +5,7 @@
  *
  * PCI Express I/O Virtualization (IOV) support.
  *   Single Root IOV 1.0
+ *   Address Translation Service 1.0
  */
 
 #include <linux/pci.h>
@@ -679,3 +680,107 @@ irqreturn_t pci_sriov_migration(struct pci_dev *dev)
 	return sriov_migration(dev) ? IRQ_HANDLED : IRQ_NONE;
 }
 EXPORT_SYMBOL_GPL(pci_sriov_migration);
+
+static int ats_alloc_one(struct pci_dev *dev, int ps)
+{
+	int pos;
+	u16 cap;
+	struct pci_ats *ats;
+
+	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ATS);
+	if (!pos)
+		return -ENODEV;
+
+	ats = kzalloc(sizeof(*ats), GFP_KERNEL);
+	if (!ats)
+		return -ENOMEM;
+
+	ats->pos = pos;
+	ats->stu = ps;
+	pci_read_config_word(dev, pos + PCI_ATS_CAP, &cap);
+	ats->qdep = PCI_ATS_CAP_QDEP(cap) ? PCI_ATS_CAP_QDEP(cap) :
+					    PCI_ATS_MAX_QDEP;
+	dev->ats = ats;
+
+	return 0;
+}
+
+static void ats_free_one(struct pci_dev *dev)
+{
+	kfree(dev->ats);
+	dev->ats = NULL;
+}
+
+/**
+ * pci_enable_ats - enable the ATS capability
+ * @dev: the PCI device
+ * @ps: the IOMMU page shift
+ *
+ * Returns 0 on success, or negative on failure.
+ */
+int pci_enable_ats(struct pci_dev *dev, int ps)
+{
+	int rc;
+	u16 ctrl;
+
+	BUG_ON(dev->ats);
+
+	if (ps < PCI_ATS_MIN_STU)
+		return -EINVAL;
+
+	rc = ats_alloc_one(dev, ps);
+	if (rc)
+		return rc;
+
+	ctrl = PCI_ATS_CTRL_ENABLE;
+	ctrl |= PCI_ATS_CTRL_STU(ps - PCI_ATS_MIN_STU);
+	pci_write_config_word(dev, dev->ats->pos + PCI_ATS_CTRL, ctrl);
+
+	return 0;
+}
+
+/**
+ * pci_disable_ats - disable the ATS capability
+ * @dev: the PCI device
+ */
+void pci_disable_ats(struct pci_dev *dev)
+{
+	u16 ctrl;
+
+	BUG_ON(!dev->ats);
+
+	pci_read_config_word(dev, dev->ats->pos + PCI_ATS_CTRL, &ctrl);
+	ctrl &= ~PCI_ATS_CTRL_ENABLE;
+	pci_write_config_word(dev, dev->ats->pos + PCI_ATS_CTRL, ctrl);
+
+	ats_free_one(dev);
+}
+
+/**
+ * pci_ats_queue_depth - query the ATS Invalidate Queue Depth
+ * @dev: the PCI device
+ *
+ * Returns the queue depth on success, or negative on failure.
+ *
+ * The ATS spec uses 0 in the Invalidate Queue Depth field to
+ * indicate that the function can accept 32 Invalidate Request.
+ * But here we use the `real' values (i.e. 1~32) for the Queue
+ * Depth.
+ */
+int pci_ats_queue_depth(struct pci_dev *dev)
+{
+	int pos;
+	u16 cap;
+
+	if (dev->ats)
+		return dev->ats->qdep;
+
+	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ATS);
+	if (!pos)
+		return -ENODEV;
+
+	pci_read_config_word(dev, pos + PCI_ATS_CAP, &cap);
+
+	return PCI_ATS_CAP_QDEP(cap) ? PCI_ATS_CAP_QDEP(cap) :
+				       PCI_ATS_MAX_QDEP;
+}
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index d03f6b99f292..3c2ec64f78e9 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -229,6 +229,13 @@ struct pci_sriov {
 	u8 __iomem *mstate;	/* VF Migration State Array */
 };
 
+/* Address Translation Service */
+struct pci_ats {
+	int pos;	/* capability position */
+	int stu;	/* Smallest Translation Unit */
+	int qdep;	/* Invalidate Queue Depth */
+};
+
 #ifdef CONFIG_PCI_IOV
 extern int pci_iov_init(struct pci_dev *dev);
 extern void pci_iov_release(struct pci_dev *dev);
@@ -236,6 +243,20 @@ extern int pci_iov_resource_bar(struct pci_dev *dev, int resno,
 				enum pci_bar_type *type);
 extern void pci_restore_iov_state(struct pci_dev *dev);
 extern int pci_iov_bus_range(struct pci_bus *bus);
+
+extern int pci_enable_ats(struct pci_dev *dev, int ps);
+extern void pci_disable_ats(struct pci_dev *dev);
+extern int pci_ats_queue_depth(struct pci_dev *dev);
+/**
+ * pci_ats_enabled - query the ATS status
+ * @dev: the PCI device
+ *
+ * Returns 1 if ATS capability is enabled, or 0 if not.
+ */
+static inline int pci_ats_enabled(struct pci_dev *dev)
+{
+	return !!dev->ats;
+}
 #else
 static inline int pci_iov_init(struct pci_dev *dev)
 {
@@ -257,6 +278,22 @@ static inline int pci_iov_bus_range(struct pci_bus *bus)
 {
 	return 0;
 }
+
+static inline int pci_enable_ats(struct pci_dev *dev, int ps)
+{
+	return -ENODEV;
+}
+static inline void pci_disable_ats(struct pci_dev *dev)
+{
+}
+static inline int pci_ats_queue_depth(struct pci_dev *dev)
+{
+	return -ENODEV;
+}
+static inline int pci_ats_enabled(struct pci_dev *dev)
+{
+	return 0;
+}
 #endif /* CONFIG_PCI_IOV */
 
 #endif /* DRIVERS_PCI_H */
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 72698d89e767..bd3e4a798c43 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -188,6 +188,7 @@ struct pci_cap_saved_state {
 struct pcie_link_state;
 struct pci_vpd;
 struct pci_sriov;
+struct pci_ats;
 
 /*
  * The pci_dev structure is used to describe PCI devices.
@@ -285,6 +286,7 @@ struct pci_dev {
 		struct pci_sriov *sriov;	/* SR-IOV capability related */
 		struct pci_dev *physfn;	/* the PF this VF is associated with */
 	};
+	struct pci_ats	*ats;	/* Address Translation Service */
 #endif
 };
 
diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h
index e4d08c1b2e0b..c03189c56c7a 100644
--- a/include/linux/pci_regs.h
+++ b/include/linux/pci_regs.h
@@ -501,6 +501,7 @@
 #define PCI_EXT_CAP_ID_DSN	3
 #define PCI_EXT_CAP_ID_PWR	4
 #define PCI_EXT_CAP_ID_ARI	14
+#define PCI_EXT_CAP_ID_ATS	15
 #define PCI_EXT_CAP_ID_SRIOV	16
 
 /* Advanced Error Reporting */
@@ -619,6 +620,15 @@
 #define  PCI_ARI_CTRL_ACS	0x0002	/* ACS Function Groups Enable */
 #define  PCI_ARI_CTRL_FG(x)	(((x) >> 4) & 7) /* Function Group */
 
+/* Address Translation Service */
+#define PCI_ATS_CAP		0x04	/* ATS Capability Register */
+#define  PCI_ATS_CAP_QDEP(x)	((x) & 0x1f)	/* Invalidate Queue Depth */
+#define  PCI_ATS_MAX_QDEP	32	/* Max Invalidate Queue Depth */
+#define PCI_ATS_CTRL		0x06	/* ATS Control Register */
+#define  PCI_ATS_CTRL_ENABLE	0x8000	/* ATS Enable */
+#define  PCI_ATS_CTRL_STU(x)	((x) & 0x1f)	/* Smallest Translation Unit */
+#define  PCI_ATS_MIN_STU	12	/* shift of minimum STU block */
+
 /* Single Root I/O Virtualization */
 #define PCI_SRIOV_CAP		0x04	/* SR-IOV Capabilities */
 #define  PCI_SRIOV_CAP_VFM	0x01	/* VF Migration Capable */
-- 
cgit v1.2.3-59-g8ed1b


From e277d2fc79d6abb86fafadb58dca0b9c498a9aa7 Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Mon, 18 May 2009 13:51:33 +0800
Subject: PCI: handle Virtual Function ATS enabling

The SR-IOV spec requires that the Smallest Translation Unit and
the Invalidate Queue Depth fields in the Virtual Function ATS
capability are hardwired to 0. If a function is a Virtual Function,
then and set its Physical Function's STU before enabling the ATS.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/iov.c | 66 +++++++++++++++++++++++++++++++++++++++++++------------
 drivers/pci/pci.h |  4 +++-
 2 files changed, 55 insertions(+), 15 deletions(-)

diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c
index 0a7a1b40286f..415140499ffb 100644
--- a/drivers/pci/iov.c
+++ b/drivers/pci/iov.c
@@ -491,10 +491,10 @@ found:
 
 	if (pdev)
 		iov->dev = pci_dev_get(pdev);
-	else {
+	else
 		iov->dev = dev;
-		mutex_init(&iov->lock);
-	}
+
+	mutex_init(&iov->lock);
 
 	dev->sriov = iov;
 	dev->is_physfn = 1;
@@ -514,11 +514,11 @@ static void sriov_release(struct pci_dev *dev)
 {
 	BUG_ON(dev->sriov->nr_virtfn);
 
-	if (dev == dev->sriov->dev)
-		mutex_destroy(&dev->sriov->lock);
-	else
+	if (dev != dev->sriov->dev)
 		pci_dev_put(dev->sriov->dev);
 
+	mutex_destroy(&dev->sriov->lock);
+
 	kfree(dev->sriov);
 	dev->sriov = NULL;
 }
@@ -723,19 +723,40 @@ int pci_enable_ats(struct pci_dev *dev, int ps)
 	int rc;
 	u16 ctrl;
 
-	BUG_ON(dev->ats);
+	BUG_ON(dev->ats && dev->ats->is_enabled);
 
 	if (ps < PCI_ATS_MIN_STU)
 		return -EINVAL;
 
-	rc = ats_alloc_one(dev, ps);
-	if (rc)
-		return rc;
+	if (dev->is_physfn || dev->is_virtfn) {
+		struct pci_dev *pdev = dev->is_physfn ? dev : dev->physfn;
+
+		mutex_lock(&pdev->sriov->lock);
+		if (pdev->ats)
+			rc = pdev->ats->stu == ps ? 0 : -EINVAL;
+		else
+			rc = ats_alloc_one(pdev, ps);
+
+		if (!rc)
+			pdev->ats->ref_cnt++;
+		mutex_unlock(&pdev->sriov->lock);
+		if (rc)
+			return rc;
+	}
+
+	if (!dev->is_physfn) {
+		rc = ats_alloc_one(dev, ps);
+		if (rc)
+			return rc;
+	}
 
 	ctrl = PCI_ATS_CTRL_ENABLE;
-	ctrl |= PCI_ATS_CTRL_STU(ps - PCI_ATS_MIN_STU);
+	if (!dev->is_virtfn)
+		ctrl |= PCI_ATS_CTRL_STU(ps - PCI_ATS_MIN_STU);
 	pci_write_config_word(dev, dev->ats->pos + PCI_ATS_CTRL, ctrl);
 
+	dev->ats->is_enabled = 1;
+
 	return 0;
 }
 
@@ -747,13 +768,26 @@ void pci_disable_ats(struct pci_dev *dev)
 {
 	u16 ctrl;
 
-	BUG_ON(!dev->ats);
+	BUG_ON(!dev->ats || !dev->ats->is_enabled);
 
 	pci_read_config_word(dev, dev->ats->pos + PCI_ATS_CTRL, &ctrl);
 	ctrl &= ~PCI_ATS_CTRL_ENABLE;
 	pci_write_config_word(dev, dev->ats->pos + PCI_ATS_CTRL, ctrl);
 
-	ats_free_one(dev);
+	dev->ats->is_enabled = 0;
+
+	if (dev->is_physfn || dev->is_virtfn) {
+		struct pci_dev *pdev = dev->is_physfn ? dev : dev->physfn;
+
+		mutex_lock(&pdev->sriov->lock);
+		pdev->ats->ref_cnt--;
+		if (!pdev->ats->ref_cnt)
+			ats_free_one(pdev);
+		mutex_unlock(&pdev->sriov->lock);
+	}
+
+	if (!dev->is_physfn)
+		ats_free_one(dev);
 }
 
 /**
@@ -765,13 +799,17 @@ void pci_disable_ats(struct pci_dev *dev)
  * The ATS spec uses 0 in the Invalidate Queue Depth field to
  * indicate that the function can accept 32 Invalidate Request.
  * But here we use the `real' values (i.e. 1~32) for the Queue
- * Depth.
+ * Depth; and 0 indicates the function shares the Queue with
+ * other functions (doesn't exclusively own a Queue).
  */
 int pci_ats_queue_depth(struct pci_dev *dev)
 {
 	int pos;
 	u16 cap;
 
+	if (dev->is_virtfn)
+		return 0;
+
 	if (dev->ats)
 		return dev->ats->qdep;
 
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index 3c2ec64f78e9..f73bcbedf37c 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -234,6 +234,8 @@ struct pci_ats {
 	int pos;	/* capability position */
 	int stu;	/* Smallest Translation Unit */
 	int qdep;	/* Invalidate Queue Depth */
+	int ref_cnt;	/* Physical Function reference count */
+	int is_enabled:1;	/* Enable bit is set */
 };
 
 #ifdef CONFIG_PCI_IOV
@@ -255,7 +257,7 @@ extern int pci_ats_queue_depth(struct pci_dev *dev);
  */
 static inline int pci_ats_enabled(struct pci_dev *dev)
 {
-	return !!dev->ats;
+	return dev->ats && dev->ats->is_enabled;
 }
 #else
 static inline int pci_iov_init(struct pci_dev *dev)
-- 
cgit v1.2.3-59-g8ed1b


From aa5d2b515b6fca5f8a56eac84f7fa0a68c1ce9b7 Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Mon, 18 May 2009 13:51:34 +0800
Subject: VT-d: parse ATSR in DMA Remapping Reporting Structure

Parse the Root Port ATS Capability Reporting Structure in the DMA
Remapping Reporting Structure ACPI table.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/dmar.c          | 112 +++++++++++++++++++++++++++++++++++++++++---
 include/linux/dmar.h        |   9 ++++
 include/linux/intel-iommu.h |   1 +
 3 files changed, 116 insertions(+), 6 deletions(-)

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index f23460a5d106..6d7f9619b8a9 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -267,6 +267,84 @@ rmrr_parse_dev(struct dmar_rmrr_unit *rmrru)
 	}
 	return ret;
 }
+
+static LIST_HEAD(dmar_atsr_units);
+
+static int __init dmar_parse_one_atsr(struct acpi_dmar_header *hdr)
+{
+	struct acpi_dmar_atsr *atsr;
+	struct dmar_atsr_unit *atsru;
+
+	atsr = container_of(hdr, struct acpi_dmar_atsr, header);
+	atsru = kzalloc(sizeof(*atsru), GFP_KERNEL);
+	if (!atsru)
+		return -ENOMEM;
+
+	atsru->hdr = hdr;
+	atsru->include_all = atsr->flags & 0x1;
+
+	list_add(&atsru->list, &dmar_atsr_units);
+
+	return 0;
+}
+
+static int __init atsr_parse_dev(struct dmar_atsr_unit *atsru)
+{
+	int rc;
+	struct acpi_dmar_atsr *atsr;
+
+	if (atsru->include_all)
+		return 0;
+
+	atsr = container_of(atsru->hdr, struct acpi_dmar_atsr, header);
+	rc = dmar_parse_dev_scope((void *)(atsr + 1),
+				(void *)atsr + atsr->header.length,
+				&atsru->devices_cnt, &atsru->devices,
+				atsr->segment);
+	if (rc || !atsru->devices_cnt) {
+		list_del(&atsru->list);
+		kfree(atsru);
+	}
+
+	return rc;
+}
+
+int dmar_find_matched_atsr_unit(struct pci_dev *dev)
+{
+	int i;
+	struct pci_bus *bus;
+	struct acpi_dmar_atsr *atsr;
+	struct dmar_atsr_unit *atsru;
+
+	list_for_each_entry(atsru, &dmar_atsr_units, list) {
+		atsr = container_of(atsru->hdr, struct acpi_dmar_atsr, header);
+		if (atsr->segment == pci_domain_nr(dev->bus))
+			goto found;
+	}
+
+	return 0;
+
+found:
+	for (bus = dev->bus; bus; bus = bus->parent) {
+		struct pci_dev *bridge = bus->self;
+
+		if (!bridge || !bridge->is_pcie ||
+		    bridge->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE)
+			return 0;
+
+		if (bridge->pcie_type == PCI_EXP_TYPE_ROOT_PORT) {
+			for (i = 0; i < atsru->devices_cnt; i++)
+				if (atsru->devices[i] == bridge)
+					return 1;
+			break;
+		}
+	}
+
+	if (atsru->include_all)
+		return 1;
+
+	return 0;
+}
 #endif
 
 static void __init
@@ -274,22 +352,28 @@ dmar_table_print_dmar_entry(struct acpi_dmar_header *header)
 {
 	struct acpi_dmar_hardware_unit *drhd;
 	struct acpi_dmar_reserved_memory *rmrr;
+	struct acpi_dmar_atsr *atsr;
 
 	switch (header->type) {
 	case ACPI_DMAR_TYPE_HARDWARE_UNIT:
-		drhd = (struct acpi_dmar_hardware_unit *)header;
+		drhd = container_of(header, struct acpi_dmar_hardware_unit,
+				    header);
 		printk (KERN_INFO PREFIX
-			"DRHD (flags: 0x%08x)base: 0x%016Lx\n",
-			drhd->flags, (unsigned long long)drhd->address);
+			"DRHD base: %#016Lx flags: %#x\n",
+			(unsigned long long)drhd->address, drhd->flags);
 		break;
 	case ACPI_DMAR_TYPE_RESERVED_MEMORY:
-		rmrr = (struct acpi_dmar_reserved_memory *)header;
-
+		rmrr = container_of(header, struct acpi_dmar_reserved_memory,
+				    header);
 		printk (KERN_INFO PREFIX
-			"RMRR base: 0x%016Lx end: 0x%016Lx\n",
+			"RMRR base: %#016Lx end: %#016Lx\n",
 			(unsigned long long)rmrr->base_address,
 			(unsigned long long)rmrr->end_address);
 		break;
+	case ACPI_DMAR_TYPE_ATSR:
+		atsr = container_of(header, struct acpi_dmar_atsr, header);
+		printk(KERN_INFO PREFIX "ATSR flags: %#x\n", atsr->flags);
+		break;
 	}
 }
 
@@ -361,6 +445,11 @@ parse_dmar_table(void)
 		case ACPI_DMAR_TYPE_RESERVED_MEMORY:
 #ifdef CONFIG_DMAR
 			ret = dmar_parse_one_rmrr(entry_header);
+#endif
+			break;
+		case ACPI_DMAR_TYPE_ATSR:
+#ifdef CONFIG_DMAR
+			ret = dmar_parse_one_atsr(entry_header);
 #endif
 			break;
 		default:
@@ -431,11 +520,19 @@ int __init dmar_dev_scope_init(void)
 #ifdef CONFIG_DMAR
 	{
 		struct dmar_rmrr_unit *rmrr, *rmrr_n;
+		struct dmar_atsr_unit *atsr, *atsr_n;
+
 		list_for_each_entry_safe(rmrr, rmrr_n, &dmar_rmrr_units, list) {
 			ret = rmrr_parse_dev(rmrr);
 			if (ret)
 				return ret;
 		}
+
+		list_for_each_entry_safe(atsr, atsr_n, &dmar_atsr_units, list) {
+			ret = atsr_parse_dev(atsr);
+			if (ret)
+				return ret;
+		}
 	}
 #endif
 
@@ -468,6 +565,9 @@ int __init dmar_table_init(void)
 #ifdef CONFIG_DMAR
 	if (list_empty(&dmar_rmrr_units))
 		printk(KERN_INFO PREFIX "No RMRR found\n");
+
+	if (list_empty(&dmar_atsr_units))
+		printk(KERN_INFO PREFIX "No ATSR found\n");
 #endif
 
 #ifdef CONFIG_INTR_REMAP
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index e397dc342cda..7c9a207e5da6 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -185,6 +185,15 @@ struct dmar_rmrr_unit {
 
 #define for_each_rmrr_units(rmrr) \
 	list_for_each_entry(rmrr, &dmar_rmrr_units, list)
+
+struct dmar_atsr_unit {
+	struct list_head list;		/* list of ATSR units */
+	struct acpi_dmar_header *hdr;	/* ACPI header */
+	struct pci_dev **devices;	/* target devices */
+	int devices_cnt;		/* target device count */
+	u8 include_all:1;		/* include all ports */
+};
+
 /* Intel DMAR  initialization functions */
 extern int intel_iommu_init(void);
 #else
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 29e05a034c09..0a1939f200fc 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -331,6 +331,7 @@ static inline void __iommu_flush_cache(
 }
 
 extern struct dmar_drhd_unit * dmar_find_matched_drhd_unit(struct pci_dev *dev);
+extern int dmar_find_matched_atsr_unit(struct pci_dev *dev);
 
 extern int alloc_iommu(struct dmar_drhd_unit *drhd);
 extern void free_iommu(struct intel_iommu *iommu);
-- 
cgit v1.2.3-59-g8ed1b


From 6ba6c3a4cacfd68bf970e3e04e2ff0d66fa0f695 Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Mon, 18 May 2009 13:51:35 +0800
Subject: VT-d: add device IOTLB invalidation support

Support device IOTLB invalidation to flush the translation cached
in the Endpoint.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/dmar.c          | 77 ++++++++++++++++++++++++++++++++++++++++-----
 include/linux/intel-iommu.h | 14 ++++++++-
 2 files changed, 82 insertions(+), 9 deletions(-)

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index 6d7f9619b8a9..7b287cb38b7a 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -699,7 +699,8 @@ void free_iommu(struct intel_iommu *iommu)
  */
 static inline void reclaim_free_desc(struct q_inval *qi)
 {
-	while (qi->desc_status[qi->free_tail] == QI_DONE) {
+	while (qi->desc_status[qi->free_tail] == QI_DONE ||
+	       qi->desc_status[qi->free_tail] == QI_ABORT) {
 		qi->desc_status[qi->free_tail] = QI_FREE;
 		qi->free_tail = (qi->free_tail + 1) % QI_LENGTH;
 		qi->free_cnt++;
@@ -709,10 +710,13 @@ static inline void reclaim_free_desc(struct q_inval *qi)
 static int qi_check_fault(struct intel_iommu *iommu, int index)
 {
 	u32 fault;
-	int head;
+	int head, tail;
 	struct q_inval *qi = iommu->qi;
 	int wait_index = (index + 1) % QI_LENGTH;
 
+	if (qi->desc_status[wait_index] == QI_ABORT)
+		return -EAGAIN;
+
 	fault = readl(iommu->reg + DMAR_FSTS_REG);
 
 	/*
@@ -722,7 +726,11 @@ static int qi_check_fault(struct intel_iommu *iommu, int index)
 	 */
 	if (fault & DMA_FSTS_IQE) {
 		head = readl(iommu->reg + DMAR_IQH_REG);
-		if ((head >> 4) == index) {
+		if ((head >> DMAR_IQ_SHIFT) == index) {
+			printk(KERN_ERR "VT-d detected invalid descriptor: "
+				"low=%llx, high=%llx\n",
+				(unsigned long long)qi->desc[index].low,
+				(unsigned long long)qi->desc[index].high);
 			memcpy(&qi->desc[index], &qi->desc[wait_index],
 					sizeof(struct qi_desc));
 			__iommu_flush_cache(iommu, &qi->desc[index],
@@ -732,6 +740,32 @@ static int qi_check_fault(struct intel_iommu *iommu, int index)
 		}
 	}
 
+	/*
+	 * If ITE happens, all pending wait_desc commands are aborted.
+	 * No new descriptors are fetched until the ITE is cleared.
+	 */
+	if (fault & DMA_FSTS_ITE) {
+		head = readl(iommu->reg + DMAR_IQH_REG);
+		head = ((head >> DMAR_IQ_SHIFT) - 1 + QI_LENGTH) % QI_LENGTH;
+		head |= 1;
+		tail = readl(iommu->reg + DMAR_IQT_REG);
+		tail = ((tail >> DMAR_IQ_SHIFT) - 1 + QI_LENGTH) % QI_LENGTH;
+
+		writel(DMA_FSTS_ITE, iommu->reg + DMAR_FSTS_REG);
+
+		do {
+			if (qi->desc_status[head] == QI_IN_USE)
+				qi->desc_status[head] = QI_ABORT;
+			head = (head - 2 + QI_LENGTH) % QI_LENGTH;
+		} while (head != tail);
+
+		if (qi->desc_status[wait_index] == QI_ABORT)
+			return -EAGAIN;
+	}
+
+	if (fault & DMA_FSTS_ICE)
+		writel(DMA_FSTS_ICE, iommu->reg + DMAR_FSTS_REG);
+
 	return 0;
 }
 
@@ -741,7 +775,7 @@ static int qi_check_fault(struct intel_iommu *iommu, int index)
  */
 int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu)
 {
-	int rc = 0;
+	int rc;
 	struct q_inval *qi = iommu->qi;
 	struct qi_desc *hw, wait_desc;
 	int wait_index, index;
@@ -752,6 +786,9 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu)
 
 	hw = qi->desc;
 
+restart:
+	rc = 0;
+
 	spin_lock_irqsave(&qi->q_lock, flags);
 	while (qi->free_cnt < 3) {
 		spin_unlock_irqrestore(&qi->q_lock, flags);
@@ -782,7 +819,7 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu)
 	 * update the HW tail register indicating the presence of
 	 * new descriptors.
 	 */
-	writel(qi->free_head << 4, iommu->reg + DMAR_IQT_REG);
+	writel(qi->free_head << DMAR_IQ_SHIFT, iommu->reg + DMAR_IQT_REG);
 
 	while (qi->desc_status[wait_index] != QI_DONE) {
 		/*
@@ -794,18 +831,21 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu)
 		 */
 		rc = qi_check_fault(iommu, index);
 		if (rc)
-			goto out;
+			break;
 
 		spin_unlock(&qi->q_lock);
 		cpu_relax();
 		spin_lock(&qi->q_lock);
 	}
-out:
-	qi->desc_status[index] = qi->desc_status[wait_index] = QI_DONE;
+
+	qi->desc_status[index] = QI_DONE;
 
 	reclaim_free_desc(qi);
 	spin_unlock_irqrestore(&qi->q_lock, flags);
 
+	if (rc == -EAGAIN)
+		goto restart;
+
 	return rc;
 }
 
@@ -857,6 +897,27 @@ void qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
 	qi_submit_sync(&desc, iommu);
 }
 
+void qi_flush_dev_iotlb(struct intel_iommu *iommu, u16 sid, u16 qdep,
+			u64 addr, unsigned mask)
+{
+	struct qi_desc desc;
+
+	if (mask) {
+		BUG_ON(addr & ((1 << (VTD_PAGE_SHIFT + mask)) - 1));
+		addr |= (1 << (VTD_PAGE_SHIFT + mask - 1)) - 1;
+		desc.high = QI_DEV_IOTLB_ADDR(addr) | QI_DEV_IOTLB_SIZE;
+	} else
+		desc.high = QI_DEV_IOTLB_ADDR(addr);
+
+	if (qdep >= QI_DEV_IOTLB_MAX_INVS)
+		qdep = 0;
+
+	desc.low = QI_DEV_IOTLB_SID(sid) | QI_DEV_IOTLB_QDEP(qdep) |
+		   QI_DIOTLB_TYPE;
+
+	qi_submit_sync(&desc, iommu);
+}
+
 /*
  * Disable Queued Invalidation interface.
  */
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 0a1939f200fc..40561b224a17 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -53,6 +53,7 @@
 #define	DMAR_PHMLIMIT_REG 0x78	/* pmrr high limit */
 #define DMAR_IQH_REG	0x80	/* Invalidation queue head register */
 #define DMAR_IQT_REG	0x88	/* Invalidation queue tail register */
+#define DMAR_IQ_SHIFT	4	/* Invalidation queue head/tail shift */
 #define DMAR_IQA_REG	0x90	/* Invalidation queue addr register */
 #define DMAR_ICS_REG	0x98	/* Invalidation complete status register */
 #define DMAR_IRTA_REG	0xb8    /* Interrupt remapping table addr register */
@@ -198,6 +199,8 @@ static inline void dmar_writeq(void __iomem *addr, u64 val)
 #define DMA_FSTS_PPF ((u32)2)
 #define DMA_FSTS_PFO ((u32)1)
 #define DMA_FSTS_IQE (1 << 4)
+#define DMA_FSTS_ICE (1 << 5)
+#define DMA_FSTS_ITE (1 << 6)
 #define dma_fsts_fault_record_index(s) (((s) >> 8) & 0xff)
 
 /* FRCD_REG, 32 bits access */
@@ -226,7 +229,8 @@ do {									\
 enum {
 	QI_FREE,
 	QI_IN_USE,
-	QI_DONE
+	QI_DONE,
+	QI_ABORT
 };
 
 #define QI_CC_TYPE		0x1
@@ -255,6 +259,12 @@ enum {
 #define QI_CC_DID(did)		(((u64)did) << 16)
 #define QI_CC_GRAN(gran)	(((u64)gran) >> (DMA_CCMD_INVL_GRANU_OFFSET-4))
 
+#define QI_DEV_IOTLB_SID(sid)	((u64)((sid) & 0xffff) << 32)
+#define QI_DEV_IOTLB_QDEP(qdep)	(((qdep) & 0x1f) << 16)
+#define QI_DEV_IOTLB_ADDR(addr)	((u64)(addr) & VTD_PAGE_MASK)
+#define QI_DEV_IOTLB_SIZE	1
+#define QI_DEV_IOTLB_MAX_INVS	32
+
 struct qi_desc {
 	u64 low, high;
 };
@@ -344,6 +354,8 @@ extern void qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid,
 			     u8 fm, u64 type);
 extern void qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
 			  unsigned int size_order, u64 type);
+extern void qi_flush_dev_iotlb(struct intel_iommu *iommu, u16 sid, u16 qdep,
+			       u64 addr, unsigned mask);
 
 extern int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu);
 
-- 
cgit v1.2.3-59-g8ed1b


From 9dd2fe89062c90a964d122b8be5615d6f2203bbe Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Mon, 18 May 2009 13:51:36 +0800
Subject: VT-d: cleanup iommu_flush_iotlb_psi and flush_unmaps

Make iommu_flush_iotlb_psi() and flush_unmaps() more readable.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 38 +++++++++++++++++---------------------
 1 file changed, 17 insertions(+), 21 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index bc99b1e47fbc..6d7cb84c63ea 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -948,28 +948,23 @@ static void __iommu_flush_iotlb(struct intel_iommu *iommu, u16 did,
 static void iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did,
 				  u64 addr, unsigned int pages)
 {
-	unsigned int mask;
+	unsigned int mask = ilog2(__roundup_pow_of_two(pages));
 
 	BUG_ON(addr & (~VTD_PAGE_MASK));
 	BUG_ON(pages == 0);
 
-	/* Fallback to domain selective flush if no PSI support */
-	if (!cap_pgsel_inv(iommu->cap))
-		return iommu->flush.flush_iotlb(iommu, did, 0, 0,
-						DMA_TLB_DSI_FLUSH);
-
 	/*
+	 * Fallback to domain selective flush if no PSI support or the size is
+	 * too big.
 	 * PSI requires page size to be 2 ^ x, and the base address is naturally
 	 * aligned to the size
 	 */
-	mask = ilog2(__roundup_pow_of_two(pages));
-	/* Fallback to domain selective flush if size is too big */
-	if (mask > cap_max_amask_val(iommu->cap))
-		return iommu->flush.flush_iotlb(iommu, did, 0, 0,
+	if (!cap_pgsel_inv(iommu->cap) || mask > cap_max_amask_val(iommu->cap))
+		iommu->flush.flush_iotlb(iommu, did, 0, 0,
 						DMA_TLB_DSI_FLUSH);
-
-	return iommu->flush.flush_iotlb(iommu, did, addr, mask,
-					DMA_TLB_PSI_FLUSH);
+	else
+		iommu->flush.flush_iotlb(iommu, did, addr, mask,
+						DMA_TLB_PSI_FLUSH);
 }
 
 static void iommu_disable_protect_mem_regions(struct intel_iommu *iommu)
@@ -2260,15 +2255,16 @@ static void flush_unmaps(void)
 		if (!iommu)
 			continue;
 
-		if (deferred_flush[i].next) {
-			iommu->flush.flush_iotlb(iommu, 0, 0, 0,
-						 DMA_TLB_GLOBAL_FLUSH);
-			for (j = 0; j < deferred_flush[i].next; j++) {
-				__free_iova(&deferred_flush[i].domain[j]->iovad,
-						deferred_flush[i].iova[j]);
-			}
-			deferred_flush[i].next = 0;
+		if (!deferred_flush[i].next)
+			continue;
+
+		iommu->flush.flush_iotlb(iommu, 0, 0, 0,
+					 DMA_TLB_GLOBAL_FLUSH, 0);
+		for (j = 0; j < deferred_flush[i].next; j++) {
+			__free_iova(&deferred_flush[i].domain[j]->iovad,
+					deferred_flush[i].iova[j]);
 		}
+		deferred_flush[i].next = 0;
 	}
 
 	list_size = 0;
-- 
cgit v1.2.3-59-g8ed1b


From 93a23a7271dfb811b3adb72779054c3a24433112 Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Mon, 18 May 2009 13:51:37 +0800
Subject: VT-d: support the device IOTLB

Enable the device IOTLB (i.e. ATS) for both the bare metal and KVM
environments.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c     | 109 ++++++++++++++++++++++++++++++++++++++----
 include/linux/dma_remapping.h |   1 +
 include/linux/intel-iommu.h   |   1 +
 3 files changed, 102 insertions(+), 9 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 6d7cb84c63ea..c3cdfc90c13a 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -252,6 +252,7 @@ struct device_domain_info {
 	u8 bus;			/* PCI bus number */
 	u8 devfn;		/* PCI devfn number */
 	struct pci_dev *dev; /* it's NULL for PCIE-to-PCI bridge */
+	struct intel_iommu *iommu; /* IOMMU used by this device */
 	struct dmar_domain *domain; /* pointer to domain */
 };
 
@@ -945,6 +946,77 @@ static void __iommu_flush_iotlb(struct intel_iommu *iommu, u16 did,
 			(unsigned long long)DMA_TLB_IAIG(val));
 }
 
+static struct device_domain_info *iommu_support_dev_iotlb(
+	struct dmar_domain *domain, int segment, u8 bus, u8 devfn)
+{
+	int found = 0;
+	unsigned long flags;
+	struct device_domain_info *info;
+	struct intel_iommu *iommu = device_to_iommu(segment, bus, devfn);
+
+	if (!ecap_dev_iotlb_support(iommu->ecap))
+		return NULL;
+
+	if (!iommu->qi)
+		return NULL;
+
+	spin_lock_irqsave(&device_domain_lock, flags);
+	list_for_each_entry(info, &domain->devices, link)
+		if (info->bus == bus && info->devfn == devfn) {
+			found = 1;
+			break;
+		}
+	spin_unlock_irqrestore(&device_domain_lock, flags);
+
+	if (!found || !info->dev)
+		return NULL;
+
+	if (!pci_find_ext_capability(info->dev, PCI_EXT_CAP_ID_ATS))
+		return NULL;
+
+	if (!dmar_find_matched_atsr_unit(info->dev))
+		return NULL;
+
+	info->iommu = iommu;
+
+	return info;
+}
+
+static void iommu_enable_dev_iotlb(struct device_domain_info *info)
+{
+	if (!info)
+		return;
+
+	pci_enable_ats(info->dev, VTD_PAGE_SHIFT);
+}
+
+static void iommu_disable_dev_iotlb(struct device_domain_info *info)
+{
+	if (!info->dev || !pci_ats_enabled(info->dev))
+		return;
+
+	pci_disable_ats(info->dev);
+}
+
+static void iommu_flush_dev_iotlb(struct dmar_domain *domain,
+				  u64 addr, unsigned mask)
+{
+	u16 sid, qdep;
+	unsigned long flags;
+	struct device_domain_info *info;
+
+	spin_lock_irqsave(&device_domain_lock, flags);
+	list_for_each_entry(info, &domain->devices, link) {
+		if (!info->dev || !pci_ats_enabled(info->dev))
+			continue;
+
+		sid = info->bus << 8 | info->devfn;
+		qdep = pci_ats_queue_depth(info->dev);
+		qi_flush_dev_iotlb(info->iommu, sid, qdep, addr, mask);
+	}
+	spin_unlock_irqrestore(&device_domain_lock, flags);
+}
+
 static void iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did,
 				  u64 addr, unsigned int pages)
 {
@@ -965,6 +1037,8 @@ static void iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did,
 	else
 		iommu->flush.flush_iotlb(iommu, did, addr, mask,
 						DMA_TLB_PSI_FLUSH);
+	if (did)
+		iommu_flush_dev_iotlb(iommu->domains[did], addr, mask);
 }
 
 static void iommu_disable_protect_mem_regions(struct intel_iommu *iommu)
@@ -1305,6 +1379,7 @@ static int domain_context_mapping_one(struct dmar_domain *domain, int segment,
 	unsigned long ndomains;
 	int id;
 	int agaw;
+	struct device_domain_info *info = NULL;
 
 	pr_debug("Set context mapping for %02x:%02x.%d\n",
 		bus, PCI_SLOT(devfn), PCI_FUNC(devfn));
@@ -1372,15 +1447,21 @@ static int domain_context_mapping_one(struct dmar_domain *domain, int segment,
 
 	context_set_domain_id(context, id);
 
+	if (translation != CONTEXT_TT_PASS_THROUGH) {
+		info = iommu_support_dev_iotlb(domain, segment, bus, devfn);
+		translation = info ? CONTEXT_TT_DEV_IOTLB :
+				     CONTEXT_TT_MULTI_LEVEL;
+	}
 	/*
 	 * In pass through mode, AW must be programmed to indicate the largest
 	 * AGAW value supported by hardware. And ASR is ignored by hardware.
 	 */
-	if (likely(translation == CONTEXT_TT_MULTI_LEVEL)) {
-		context_set_address_width(context, iommu->agaw);
-		context_set_address_root(context, virt_to_phys(pgd));
-	} else
+	if (unlikely(translation == CONTEXT_TT_PASS_THROUGH))
 		context_set_address_width(context, iommu->msagaw);
+	else {
+		context_set_address_root(context, virt_to_phys(pgd));
+		context_set_address_width(context, iommu->agaw);
+	}
 
 	context_set_translation_type(context, translation);
 	context_set_fault_enable(context);
@@ -1402,6 +1483,7 @@ static int domain_context_mapping_one(struct dmar_domain *domain, int segment,
 	} else {
 		iommu_flush_write_buffer(iommu);
 	}
+	iommu_enable_dev_iotlb(info);
 	spin_unlock_irqrestore(&iommu->lock, flags);
 
 	spin_lock_irqsave(&domain->iommu_lock, flags);
@@ -1552,6 +1634,7 @@ static void domain_remove_dev_info(struct dmar_domain *domain)
 			info->dev->dev.archdata.iommu = NULL;
 		spin_unlock_irqrestore(&device_domain_lock, flags);
 
+		iommu_disable_dev_iotlb(info);
 		iommu = device_to_iommu(info->segment, info->bus, info->devfn);
 		iommu_detach_dev(iommu, info->bus, info->devfn);
 		free_devinfo_mem(info);
@@ -2259,10 +2342,16 @@ static void flush_unmaps(void)
 			continue;
 
 		iommu->flush.flush_iotlb(iommu, 0, 0, 0,
-					 DMA_TLB_GLOBAL_FLUSH, 0);
+					 DMA_TLB_GLOBAL_FLUSH);
 		for (j = 0; j < deferred_flush[i].next; j++) {
-			__free_iova(&deferred_flush[i].domain[j]->iovad,
-					deferred_flush[i].iova[j]);
+			unsigned long mask;
+			struct iova *iova = deferred_flush[i].iova[j];
+
+			mask = (iova->pfn_hi - iova->pfn_lo + 1) << PAGE_SHIFT;
+			mask = ilog2(mask >> VTD_PAGE_SHIFT);
+			iommu_flush_dev_iotlb(deferred_flush[i].domain[j],
+					iova->pfn_lo << PAGE_SHIFT, mask);
+			__free_iova(&deferred_flush[i].domain[j]->iovad, iova);
 		}
 		deferred_flush[i].next = 0;
 	}
@@ -2943,6 +3032,7 @@ static void vm_domain_remove_one_dev_info(struct dmar_domain *domain,
 				info->dev->dev.archdata.iommu = NULL;
 			spin_unlock_irqrestore(&device_domain_lock, flags);
 
+			iommu_disable_dev_iotlb(info);
 			iommu_detach_dev(iommu, info->bus, info->devfn);
 			iommu_detach_dependent_devices(iommu, pdev);
 			free_devinfo_mem(info);
@@ -2993,6 +3083,7 @@ static void vm_domain_remove_all_dev_info(struct dmar_domain *domain)
 
 		spin_unlock_irqrestore(&device_domain_lock, flags1);
 
+		iommu_disable_dev_iotlb(info);
 		iommu = device_to_iommu(info->segment, info->bus, info->devfn);
 		iommu_detach_dev(iommu, info->bus, info->devfn);
 		iommu_detach_dependent_devices(iommu, info->dev);
@@ -3197,11 +3288,11 @@ static int intel_iommu_attach_device(struct iommu_domain *domain,
 		return -EFAULT;
 	}
 
-	ret = domain_context_mapping(dmar_domain, pdev, CONTEXT_TT_MULTI_LEVEL);
+	ret = vm_domain_add_dev_info(dmar_domain, pdev);
 	if (ret)
 		return ret;
 
-	ret = vm_domain_add_dev_info(dmar_domain, pdev);
+	ret = domain_context_mapping(dmar_domain, pdev, CONTEXT_TT_MULTI_LEVEL);
 	return ret;
 }
 
diff --git a/include/linux/dma_remapping.h b/include/linux/dma_remapping.h
index e0a03aff63d9..5619f8522738 100644
--- a/include/linux/dma_remapping.h
+++ b/include/linux/dma_remapping.h
@@ -14,6 +14,7 @@
 #define DMA_PTE_SNP (1 << 11)
 
 #define CONTEXT_TT_MULTI_LEVEL	0
+#define CONTEXT_TT_DEV_IOTLB	1
 #define CONTEXT_TT_PASS_THROUGH 2
 
 struct intel_iommu;
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 40561b224a17..482dc91fd53a 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -124,6 +124,7 @@ static inline void dmar_writeq(void __iomem *addr, u64 val)
 #define ecap_pass_through(e)	((e >> 6) & 0x1)
 #define ecap_eim_support(e)	((e >> 4) & 0x1)
 #define ecap_ir_support(e)	((e >> 3) & 0x1)
+#define ecap_dev_iotlb_support(e)	(((e) >> 2) & 0x1)
 #define ecap_max_handle_mask(e) ((e >> 20) & 0xf)
 #define ecap_sc_support(e)	((e >> 7) & 0x1) /* Snooping Control */
 
-- 
cgit v1.2.3-59-g8ed1b


From d694846b6b1c92bcc946b6ffb0a5ea25d5df1014 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 5 Apr 2009 07:38:33 -0700
Subject: [MTD] set blkdev parent to the mtd device, not its parent

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtd_blkdevs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/mtd_blkdevs.c b/drivers/mtd/mtd_blkdevs.c
index a49a9c8f2cb1..8773481a30c2 100644
--- a/drivers/mtd/mtd_blkdevs.c
+++ b/drivers/mtd/mtd_blkdevs.c
@@ -286,7 +286,7 @@ int add_mtd_blktrans_dev(struct mtd_blktrans_dev *new)
 	gd->private_data = new;
 	new->blkcore_priv = gd;
 	gd->queue = tr->blkcore_priv->rq;
-	gd->driverfs_dev = new->mtd->dev.parent;
+	gd->driverfs_dev = &new->mtd->dev;
 
 	if (new->readonly)
 		set_disk_ro(gd, 1);
-- 
cgit v1.2.3-59-g8ed1b


From 15bce40cb3133bcc07d548013df97e4653d363c1 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 5 Apr 2009 07:40:58 -0700
Subject: [MTD] Restore suspend/resume support for mtd devices

This is intended to suspend/resume the _chip_, while we leave board
drivers to handle their own suspend/resume for the controller.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdcore.c | 47 +++++++++++++++++++++++++++++++++++++----------
 1 file changed, 37 insertions(+), 10 deletions(-)

diff --git a/drivers/mtd/mtdcore.c b/drivers/mtd/mtdcore.c
index bccb4b1ffc46..fac54a3fa3f1 100644
--- a/drivers/mtd/mtdcore.c
+++ b/drivers/mtd/mtdcore.c
@@ -23,8 +23,15 @@
 
 #include "mtdcore.h"
 
-
-static struct class *mtd_class;
+static int mtd_cls_suspend(struct device *dev, pm_message_t state);
+static int mtd_cls_resume(struct device *dev);
+
+static struct class mtd_class = {
+	.name = "mtd",
+	.owner = THIS_MODULE,
+	.suspend = mtd_cls_suspend,
+	.resume = mtd_cls_resume,
+};
 
 /* These are exported solely for the purpose of mtd_blkdevs.c. You
    should not use them for _anything_ else */
@@ -52,7 +59,26 @@ static void mtd_release(struct device *dev)
 
 	/* remove /dev/mtdXro node if needed */
 	if (index)
-		device_destroy(mtd_class, index + 1);
+		device_destroy(&mtd_class, index + 1);
+}
+
+static int mtd_cls_suspend(struct device *dev, pm_message_t state)
+{
+	struct mtd_info *mtd = dev_to_mtd(dev);
+	
+	if (mtd->suspend)
+		return mtd->suspend(mtd);
+	else
+		return 0;
+}
+
+static int mtd_cls_resume(struct device *dev)
+{
+	struct mtd_info *mtd = dev_to_mtd(dev);
+	
+	if (mtd->resume)
+		mtd->resume(mtd);
+	return 0;
 }
 
 static ssize_t mtd_type_show(struct device *dev,
@@ -269,7 +295,7 @@ int add_mtd_device(struct mtd_info *mtd)
 			 * physical device.
 			 */
 			mtd->dev.type = &mtd_devtype;
-			mtd->dev.class = mtd_class;
+			mtd->dev.class = &mtd_class;
 			mtd->dev.devt = MTD_DEVT(i);
 			dev_set_name(&mtd->dev, "mtd%d", i);
 			if (device_register(&mtd->dev) != 0) {
@@ -278,7 +304,7 @@ int add_mtd_device(struct mtd_info *mtd)
 			}
 
 			if (MTD_DEVT(i))
-				device_create(mtd_class, mtd->dev.parent,
+				device_create(&mtd_class, mtd->dev.parent,
 						MTD_DEVT(i) + 1,
 						NULL, "mtd%dro", i);
 
@@ -604,11 +630,12 @@ done:
 
 static int __init init_mtd(void)
 {
-	mtd_class = class_create(THIS_MODULE, "mtd");
+	int ret;
+	ret = class_register(&mtd_class);
 
-	if (IS_ERR(mtd_class)) {
-		pr_err("Error creating mtd class.\n");
-		return PTR_ERR(mtd_class);
+	if (ret) {
+		pr_err("Error registering mtd class: %d\n", ret);
+		return ret;
 	}
 #ifdef CONFIG_PROC_FS
 	if ((proc_mtd = create_proc_entry( "mtd", 0, NULL )))
@@ -623,7 +650,7 @@ static void __exit cleanup_mtd(void)
         if (proc_mtd)
 		remove_proc_entry( "mtd", NULL);
 #endif /* CONFIG_PROC_FS */
-	class_destroy(mtd_class);
+	class_unregister(&mtd_class);
 }
 
 module_init(init_mtd);
-- 
cgit v1.2.3-59-g8ed1b


From ccd93854d44710adaa02cecf0ef5f24ab383dd20 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 5 Apr 2009 07:49:48 -0700
Subject: [MTD] Remove mtd->{suspend,resume} calls from board drivers

Now the MTD core will do this for us, we don't need to hook it up from
the board drivers.

Shame we can't do shutdown from the class too...

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/physmap.c        | 40 ---------------------------------------
 drivers/mtd/maps/pxa2xx-flash.c   | 22 ---------------------
 drivers/mtd/maps/rbtx4939-flash.c | 23 ----------------------
 drivers/mtd/maps/sa1100-flash.c   | 23 ----------------------
 drivers/mtd/nand/mxc_nand.c       |  5 -----
 5 files changed, 113 deletions(-)

diff --git a/drivers/mtd/maps/physmap.c b/drivers/mtd/maps/physmap.c
index 29a901157352..380648e9051a 100644
--- a/drivers/mtd/maps/physmap.c
+++ b/drivers/mtd/maps/physmap.c
@@ -195,42 +195,6 @@ err_out:
 }
 
 #ifdef CONFIG_PM
-static int physmap_flash_suspend(struct platform_device *dev, pm_message_t state)
-{
-	struct physmap_flash_info *info = platform_get_drvdata(dev);
-	int ret = 0;
-	int i;
-
-	for (i = 0; i < MAX_RESOURCES && info->mtd[i]; i++)
-		if (info->mtd[i]->suspend) {
-			ret = info->mtd[i]->suspend(info->mtd[i]);
-			if (ret)
-				goto fail;
-		}
-
-	return 0;
-fail:
-	for (--i; i >= 0; --i)
-		if (info->mtd[i]->suspend) {
-			BUG_ON(!info->mtd[i]->resume);
-			info->mtd[i]->resume(info->mtd[i]);
-		}
-
-	return ret;
-}
-
-static int physmap_flash_resume(struct platform_device *dev)
-{
-	struct physmap_flash_info *info = platform_get_drvdata(dev);
-	int i;
-
-	for (i = 0; i < MAX_RESOURCES && info->mtd[i]; i++)
-		if (info->mtd[i]->resume)
-			info->mtd[i]->resume(info->mtd[i]);
-
-	return 0;
-}
-
 static void physmap_flash_shutdown(struct platform_device *dev)
 {
 	struct physmap_flash_info *info = platform_get_drvdata(dev);
@@ -242,16 +206,12 @@ static void physmap_flash_shutdown(struct platform_device *dev)
 				info->mtd[i]->resume(info->mtd[i]);
 }
 #else
-#define physmap_flash_suspend NULL
-#define physmap_flash_resume NULL
 #define physmap_flash_shutdown NULL
 #endif
 
 static struct platform_driver physmap_flash_driver = {
 	.probe		= physmap_flash_probe,
 	.remove		= physmap_flash_remove,
-	.suspend	= physmap_flash_suspend,
-	.resume		= physmap_flash_resume,
 	.shutdown	= physmap_flash_shutdown,
 	.driver		= {
 		.name	= "physmap-flash",
diff --git a/drivers/mtd/maps/pxa2xx-flash.c b/drivers/mtd/maps/pxa2xx-flash.c
index 572d32fdf38a..643aa06b599e 100644
--- a/drivers/mtd/maps/pxa2xx-flash.c
+++ b/drivers/mtd/maps/pxa2xx-flash.c
@@ -140,24 +140,6 @@ static int __devexit pxa2xx_flash_remove(struct platform_device *dev)
 }
 
 #ifdef CONFIG_PM
-static int pxa2xx_flash_suspend(struct platform_device *dev, pm_message_t state)
-{
-	struct pxa2xx_flash_info *info = platform_get_drvdata(dev);
-	int ret = 0;
-
-	if (info->mtd && info->mtd->suspend)
-		ret = info->mtd->suspend(info->mtd);
-	return ret;
-}
-
-static int pxa2xx_flash_resume(struct platform_device *dev)
-{
-	struct pxa2xx_flash_info *info = platform_get_drvdata(dev);
-
-	if (info->mtd && info->mtd->resume)
-		info->mtd->resume(info->mtd);
-	return 0;
-}
 static void pxa2xx_flash_shutdown(struct platform_device *dev)
 {
 	struct pxa2xx_flash_info *info = platform_get_drvdata(dev);
@@ -166,8 +148,6 @@ static void pxa2xx_flash_shutdown(struct platform_device *dev)
 		info->mtd->resume(info->mtd);
 }
 #else
-#define pxa2xx_flash_suspend NULL
-#define pxa2xx_flash_resume NULL
 #define pxa2xx_flash_shutdown NULL
 #endif
 
@@ -178,8 +158,6 @@ static struct platform_driver pxa2xx_flash_driver = {
 	},
 	.probe		= pxa2xx_flash_probe,
 	.remove		= __devexit_p(pxa2xx_flash_remove),
-	.suspend	= pxa2xx_flash_suspend,
-	.resume		= pxa2xx_flash_resume,
 	.shutdown	= pxa2xx_flash_shutdown,
 };
 
diff --git a/drivers/mtd/maps/rbtx4939-flash.c b/drivers/mtd/maps/rbtx4939-flash.c
index d39f0adac846..83ed64512c5e 100644
--- a/drivers/mtd/maps/rbtx4939-flash.c
+++ b/drivers/mtd/maps/rbtx4939-flash.c
@@ -145,25 +145,6 @@ err_out:
 }
 
 #ifdef CONFIG_PM
-static int rbtx4939_flash_suspend(struct platform_device *dev,
-				  pm_message_t state)
-{
-	struct rbtx4939_flash_info *info = platform_get_drvdata(dev);
-
-	if (info->mtd->suspend)
-		return info->mtd->suspend(info->mtd);
-	return 0;
-}
-
-static int rbtx4939_flash_resume(struct platform_device *dev)
-{
-	struct rbtx4939_flash_info *info = platform_get_drvdata(dev);
-
-	if (info->mtd->resume)
-		info->mtd->resume(info->mtd);
-	return 0;
-}
-
 static void rbtx4939_flash_shutdown(struct platform_device *dev)
 {
 	struct rbtx4939_flash_info *info = platform_get_drvdata(dev);
@@ -173,16 +154,12 @@ static void rbtx4939_flash_shutdown(struct platform_device *dev)
 			info->mtd->resume(info->mtd);
 }
 #else
-#define rbtx4939_flash_suspend NULL
-#define rbtx4939_flash_resume NULL
 #define rbtx4939_flash_shutdown NULL
 #endif
 
 static struct platform_driver rbtx4939_flash_driver = {
 	.probe		= rbtx4939_flash_probe,
 	.remove		= rbtx4939_flash_remove,
-	.suspend	= rbtx4939_flash_suspend,
-	.resume		= rbtx4939_flash_resume,
 	.shutdown	= rbtx4939_flash_shutdown,
 	.driver		= {
 		.name	= "rbtx4939-flash",
diff --git a/drivers/mtd/maps/sa1100-flash.c b/drivers/mtd/maps/sa1100-flash.c
index 05e9362dc7f0..c6210f5118d1 100644
--- a/drivers/mtd/maps/sa1100-flash.c
+++ b/drivers/mtd/maps/sa1100-flash.c
@@ -415,25 +415,6 @@ static int __exit sa1100_mtd_remove(struct platform_device *pdev)
 }
 
 #ifdef CONFIG_PM
-static int sa1100_mtd_suspend(struct platform_device *dev, pm_message_t state)
-{
-	struct sa_info *info = platform_get_drvdata(dev);
-	int ret = 0;
-
-	if (info)
-		ret = info->mtd->suspend(info->mtd);
-
-	return ret;
-}
-
-static int sa1100_mtd_resume(struct platform_device *dev)
-{
-	struct sa_info *info = platform_get_drvdata(dev);
-	if (info)
-		info->mtd->resume(info->mtd);
-	return 0;
-}
-
 static void sa1100_mtd_shutdown(struct platform_device *dev)
 {
 	struct sa_info *info = platform_get_drvdata(dev);
@@ -441,16 +422,12 @@ static void sa1100_mtd_shutdown(struct platform_device *dev)
 		info->mtd->resume(info->mtd);
 }
 #else
-#define sa1100_mtd_suspend NULL
-#define sa1100_mtd_resume  NULL
 #define sa1100_mtd_shutdown NULL
 #endif
 
 static struct platform_driver sa1100_mtd_driver = {
 	.probe		= sa1100_mtd_probe,
 	.remove		= __exit_p(sa1100_mtd_remove),
-	.suspend	= sa1100_mtd_suspend,
-	.resume		= sa1100_mtd_resume,
 	.shutdown	= sa1100_mtd_shutdown,
 	.driver		= {
 		.name	= "sa1100-mtd",
diff --git a/drivers/mtd/nand/mxc_nand.c b/drivers/mtd/nand/mxc_nand.c
index f3548d048014..65040de54bb7 100644
--- a/drivers/mtd/nand/mxc_nand.c
+++ b/drivers/mtd/nand/mxc_nand.c
@@ -1015,8 +1015,6 @@ static int mxcnd_suspend(struct platform_device *pdev, pm_message_t state)
 	int ret = 0;
 
 	DEBUG(MTD_DEBUG_LEVEL0, "MXC_ND : NAND suspend\n");
-	if (info)
-		ret = info->suspend(info);
 
 	/* Disable the NFC clock */
 	clk_disable(nfc_clk);	/* FIXME */
@@ -1033,9 +1031,6 @@ static int mxcnd_resume(struct platform_device *pdev)
 	/* Enable the NFC clock */
 	clk_enable(nfc_clk);	/* FIXME */
 
-	if (info)
-		info->resume(info);
-
 	return ret;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 4704a78472cd5c58f6b4c4f8c04d32de2da3f20a Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 5 Apr 2009 07:56:23 -0700
Subject: [MTD] Only set partition suspend/resume method if parent not
 registered

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdpart.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c
index 29675edb44b4..63d1cd2c17be 100644
--- a/drivers/mtd/mtdpart.c
+++ b/drivers/mtd/mtdpart.c
@@ -395,7 +395,7 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 		slave->mtd.get_fact_prot_info = part_get_fact_prot_info;
 	if (master->sync)
 		slave->mtd.sync = part_sync;
-	if (!partno && master->suspend && master->resume) {
+	if (!partno && !master->dev.class && master->suspend && master->resume) {
 			slave->mtd.suspend = part_suspend;
 			slave->mtd.resume = part_resume;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From b90cf6681f4f6263920616e7ca2fd09130e4143a Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 5 Apr 2009 08:23:44 -0700
Subject: [MTD] Remove option for add_mtd_partitions() to not register
 partitions.

This breaks the dilnetpc map driver, but it could be fixed not to use
that option. We want to simplify the partition handling, and this is a
step towards that.

Remove superfluous 'index' field from private struct mtd_part too, while
we're at it.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdpart.c          | 18 ++++--------------
 include/linux/mtd/partitions.h |  1 -
 2 files changed, 4 insertions(+), 15 deletions(-)

diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c
index 63d1cd2c17be..349fcbe5cc0f 100644
--- a/drivers/mtd/mtdpart.c
+++ b/drivers/mtd/mtdpart.c
@@ -27,9 +27,7 @@ struct mtd_part {
 	struct mtd_info mtd;
 	struct mtd_info *master;
 	uint64_t offset;
-	int index;
 	struct list_head list;
-	int registered;
 };
 
 /*
@@ -321,8 +319,7 @@ int del_mtd_partitions(struct mtd_info *master)
 	list_for_each_entry_safe(slave, next, &mtd_partitions, list)
 		if (slave->master == master) {
 			list_del(&slave->list);
-			if (slave->registered)
-				del_mtd_device(&slave->mtd);
+			del_mtd_device(&slave->mtd);
 			kfree(slave);
 		}
 
@@ -412,7 +409,6 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 	slave->mtd.erase = part_erase;
 	slave->master = master;
 	slave->offset = part->offset;
-	slave->index = partno;
 
 	if (slave->offset == MTDPART_OFS_APPEND)
 		slave->offset = cur_offset;
@@ -500,15 +496,9 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 	}
 
 out_register:
-	if (part->mtdp) {
-		/* store the object pointer (caller may or may not register it*/
-		*part->mtdp = &slave->mtd;
-		slave->registered = 0;
-	} else {
-		/* register our partition */
-		add_mtd_device(&slave->mtd);
-		slave->registered = 1;
-	}
+	/* register our partition */
+	add_mtd_device(&slave->mtd);
+
 	return slave;
 }
 
diff --git a/include/linux/mtd/partitions.h b/include/linux/mtd/partitions.h
index 7535a74083b9..af6dcb992bc3 100644
--- a/include/linux/mtd/partitions.h
+++ b/include/linux/mtd/partitions.h
@@ -40,7 +40,6 @@ struct mtd_partition {
 	uint64_t offset;		/* offset within the master MTD space */
 	uint32_t mask_flags;		/* master MTD flags to mask out for this partition */
 	struct nand_ecclayout *ecclayout;	/* out of band layout for this partition (NAND only)*/
-	struct mtd_info **mtdp;		/* pointer to store the MTD object */
 };
 
 #define MTDPART_OFS_NXTBLK	(-2)
-- 
cgit v1.2.3-59-g8ed1b


From fca4217c5bab31019b5247e977673c9fcc385f6b Mon Sep 17 00:00:00 2001
From: Greg Banks <gnb@sgi.com>
Date: Wed, 1 Apr 2009 07:28:13 +1100
Subject: knfsd: reply cache cleanups

Make REQHASH() an inline function.  Rename hash_list to cache_hash.
Fix an obsolete comment.

Signed-off-by: Greg Banks <gnb@sgi.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfscache.c         | 29 +++++++++++++++++++----------
 include/linux/nfsd/cache.h |  3 +--
 2 files changed, 20 insertions(+), 12 deletions(-)

diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c
index 5bfc2ac60d54..6f0aa4989c61 100644
--- a/fs/nfsd/nfscache.c
+++ b/fs/nfsd/nfscache.c
@@ -29,15 +29,24 @@
  */
 #define CACHESIZE		1024
 #define HASHSIZE		64
-#define REQHASH(xid)		(((((__force __u32)xid) >> 24) ^ ((__force __u32)xid)) & (HASHSIZE-1))
 
-static struct hlist_head *	hash_list;
+static struct hlist_head *	cache_hash;
 static struct list_head 	lru_head;
 static int			cache_disabled = 1;
 
+/*
+ * Calculate the hash index from an XID.
+ */
+static inline u32 request_hash(u32 xid)
+{
+	u32 h = xid;
+	h ^= (xid >> 24);
+	return h & (HASHSIZE-1);
+}
+
 static int	nfsd_cache_append(struct svc_rqst *rqstp, struct kvec *vec);
 
-/* 
+/*
  * locking for the reply cache:
  * A cache entry is "single use" if c_state == RC_INPROG
  * Otherwise, it when accessing _prev or _next, the lock must be held.
@@ -62,8 +71,8 @@ int nfsd_reply_cache_init(void)
 		i--;
 	}
 
-	hash_list = kcalloc (HASHSIZE, sizeof(struct hlist_head), GFP_KERNEL);
-	if (!hash_list)
+	cache_hash = kcalloc (HASHSIZE, sizeof(struct hlist_head), GFP_KERNEL);
+	if (!cache_hash)
 		goto out_nomem;
 
 	cache_disabled = 0;
@@ -88,8 +97,8 @@ void nfsd_reply_cache_shutdown(void)
 
 	cache_disabled = 1;
 
-	kfree (hash_list);
-	hash_list = NULL;
+	kfree (cache_hash);
+	cache_hash = NULL;
 }
 
 /*
@@ -108,7 +117,7 @@ static void
 hash_refile(struct svc_cacherep *rp)
 {
 	hlist_del_init(&rp->c_hash);
-	hlist_add_head(&rp->c_hash, hash_list + REQHASH(rp->c_xid));
+	hlist_add_head(&rp->c_hash, cache_hash + request_hash(rp->c_xid));
 }
 
 /*
@@ -138,7 +147,7 @@ nfsd_cache_lookup(struct svc_rqst *rqstp, int type)
 	spin_lock(&cache_lock);
 	rtn = RC_DOIT;
 
-	rh = &hash_list[REQHASH(xid)];
+	rh = &cache_hash[request_hash(xid)];
 	hlist_for_each_entry(rp, hn, rh, c_hash) {
 		if (rp->c_state != RC_UNUSED &&
 		    xid == rp->c_xid && proc == rp->c_proc &&
@@ -264,7 +273,7 @@ nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, __be32 *statp)
 
 	len = resv->iov_len - ((char*)statp - (char*)resv->iov_base);
 	len >>= 2;
-	
+
 	/* Don't cache excessive amounts of data and XDR failures */
 	if (!statp || len > (256 >> 2)) {
 		rp->c_state = RC_UNUSED;
diff --git a/include/linux/nfsd/cache.h b/include/linux/nfsd/cache.h
index 5bccaab81056..3a3f58934f5e 100644
--- a/include/linux/nfsd/cache.h
+++ b/include/linux/nfsd/cache.h
@@ -14,8 +14,7 @@
 #include <linux/uio.h>
 
 /*
- * Representation of a reply cache entry. The first two members *must*
- * be hash_next and hash_prev.
+ * Representation of a reply cache entry.
  */
 struct svc_cacherep {
 	struct hlist_node	c_hash;
-- 
cgit v1.2.3-59-g8ed1b


From cf0a586cf41a1779edeee7562afb5d0ab46c7cf4 Mon Sep 17 00:00:00 2001
From: Greg Banks <gnb@sgi.com>
Date: Wed, 1 Apr 2009 07:28:15 +1100
Subject: knfsd: fix reply cache memory corruption

Fix a regression in the reply cache introduced when the code was
converted to use proper Linux lists.  When a new entry needs to be
inserted, the case where all the entries are currently being used
by threads is not correctly detected.  This can result in memory
corruption and a crash.  In the current code this is an extremely
unlikely corner case; it would require the machine to have 1024
nfsd threads and all of them to be busy at the same time.  However,
upcoming reply cache changes make this more likely; a crash due to
this problem was actually observed in field.

Signed-off-by: Greg Banks <gnb@sgi.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfscache.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c
index 6f0aa4989c61..4638635c5d87 100644
--- a/fs/nfsd/nfscache.c
+++ b/fs/nfsd/nfscache.c
@@ -174,8 +174,8 @@ nfsd_cache_lookup(struct svc_rqst *rqstp, int type)
 	}
 	}
 
-	/* This should not happen */
-	if (rp == NULL) {
+	/* All entries on the LRU are in-progress. This should not happen */
+	if (&rp->c_lru == &lru_head) {
 		static int	complaints;
 
 		printk(KERN_WARNING "nfsd: all repcache entries locked!\n");
-- 
cgit v1.2.3-59-g8ed1b


From 1dbd0d53f394cd9a86fc801dd68fdbcbcdb45718 Mon Sep 17 00:00:00 2001
From: Greg Banks <gnb@sgi.com>
Date: Wed, 1 Apr 2009 07:28:21 +1100
Subject: knfsd: remove unreported filehandle stats counters

The file nfsfh.c contains two static variables nfsd_nr_verified and
nfsd_nr_put.  These are counters which are incremented as a side
effect of the fh_verify() fh_compose() and fh_put() operations,
i.e. at least twice per NFS call for any non-trivial workload.
Needless to say this makes the cacheline that contains them (and any
other innocent victims) a very hot contention point indeed under high
call-rate workloads on multiprocessor NFS server.  It also turns out
that these counters are not used anywhere.  They're not reported to
userspace, they're not used in logic, they're not even exported from
the object file (let alone the module).  All they do is waste CPU time.

So this patch removes them.

Tests on a 16 CPU Altix A4700 with 2 10gige Myricom cards, configured
separately (no bonding).  Workload is 640 client threads doing directory
traverals with random small reads, from server RAM.

Before
======

Kernel profile:

  %   cumulative   self              self     total
 time   samples   samples    calls   1/call   1/call  name
  6.05   2716.00  2716.00    30406     0.09     1.02  svc_process
  4.44   4706.00  1990.00     1975     1.01     1.01  spin_unlock_irqrestore
  3.72   6376.00  1670.00     1666     1.00     1.00  svc_export_put
  3.41   7907.00  1531.00     1786     0.86     1.02  nfsd_ofcache_lookup
  3.25   9363.00  1456.00    10965     0.13     1.01  nfsd_dispatch
  3.10  10752.00  1389.00     1376     1.01     1.01  nfsd_cache_lookup
  2.57  11907.00  1155.00     4517     0.26     1.03  svc_tcp_recvfrom
  ...
  2.21  15352.00  1003.00     1081     0.93     1.00  nfsd_choose_ofc  <----
  ^^^^

Here the function nfsd_choose_ofc() reads a global variable
which by accident happened to be located in the same cacheline as
nfsd_nr_verified.

Call rate:

nullarbor:~ # pmdumptext nfs3.server.calls
...
Thu Dec 13 00:15:27     184780.663
Thu Dec 13 00:15:28     184885.881
Thu Dec 13 00:15:29     184449.215
Thu Dec 13 00:15:30     184971.058
Thu Dec 13 00:15:31     185036.052
Thu Dec 13 00:15:32     185250.475
Thu Dec 13 00:15:33     184481.319
Thu Dec 13 00:15:34     185225.737
Thu Dec 13 00:15:35     185408.018
Thu Dec 13 00:15:36     185335.764

After
=====

kernel profile:

  %   cumulative   self              self     total
 time   samples   samples    calls   1/call   1/call  name
  6.33   2813.00  2813.00    29979     0.09     1.01  svc_process
  4.66   4883.00  2070.00     2065     1.00     1.00  spin_unlock_irqrestore
  4.06   6687.00  1804.00     2182     0.83     1.00  nfsd_ofcache_lookup
  3.20   8110.00  1423.00    10932     0.13     1.00  nfsd_dispatch
  3.03   9456.00  1346.00     1343     1.00     1.00  nfsd_cache_lookup
  2.62  10622.00  1166.00     4645     0.25     1.01  svc_tcp_recvfrom
[...]
  0.10  42586.00    44.00       74     0.59     1.00  nfsd_choose_ofc  <--- HA!!
  ^^^^

Call rate:

nullarbor:~ # pmdumptext nfs3.server.calls
...
Thu Dec 13 01:45:28     194677.118
Thu Dec 13 01:45:29     193932.692
Thu Dec 13 01:45:30     194294.364
Thu Dec 13 01:45:31     194971.276
Thu Dec 13 01:45:32     194111.207
Thu Dec 13 01:45:33     194999.635
Thu Dec 13 01:45:34     195312.594
Thu Dec 13 01:45:35     195707.293
Thu Dec 13 01:45:36     194610.353
Thu Dec 13 01:45:37     195913.662
Thu Dec 13 01:45:38     194808.675

i.e. about a 5.3% improvement in call rate.

Signed-off-by: Greg Banks <gnb@melbourne.sgi.com>
Reviewed-by: David Chinner <dgc@sgi.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfsfh.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c
index 9f1ca17293d3..8847f3fbfc1e 100644
--- a/fs/nfsd/nfsfh.c
+++ b/fs/nfsd/nfsfh.c
@@ -27,9 +27,6 @@
 #define NFSDDBG_FACILITY		NFSDDBG_FH
 
 
-static int nfsd_nr_verified;
-static int nfsd_nr_put;
-
 /*
  * our acceptability function.
  * if NOSUBTREECHECK, accept anything
@@ -251,7 +248,6 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp)
 
 	fhp->fh_dentry = dentry;
 	fhp->fh_export = exp;
-	nfsd_nr_verified++;
 	return 0;
 out:
 	exp_put(exp);
@@ -552,7 +548,6 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry,
 			return nfserr_opnotsupp;
 	}
 
-	nfsd_nr_verified++;
 	return 0;
 }
 
@@ -609,7 +604,6 @@ fh_put(struct svc_fh *fhp)
 		fhp->fh_pre_saved = 0;
 		fhp->fh_post_saved = 0;
 #endif
-		nfsd_nr_put++;
 	}
 	if (exp) {
 		cache_put(&exp->h, &svc_export_cache);
-- 
cgit v1.2.3-59-g8ed1b


From b9417f84e17b93a6976a8a88b38bf9567975cb38 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Mon, 27 Apr 2009 16:33:36 -0600
Subject: ACPI: use LNXCPU, not ACPI_CPU, for Linux-specific processor _HID

ACPI_PROCESSOR_OBJECT_HID is a synthetic _HID that Linux generates
for "Processor" definitions.  Unlike "Device" definitions, "Processor"
definitions do not have a _HID in the namespace, so we generate a
fake _HID.  By convention, all these fake _HIDs begin with "LNX".

This does change the user-visible _HID for "Processor" objects --
previously, we used "ACPI_CPU" and this changes that to "LNXCPU",
which starts with "LNX" as do all the other made-up _HIDs.  This
change is visible in processor filenames and "hid" files under
/sys/devices/LNXSYSTM:00/.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 include/acpi/acpi_drivers.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/acpi/acpi_drivers.h b/include/acpi/acpi_drivers.h
index 0352c8f0b05b..c9922b362005 100644
--- a/include/acpi/acpi_drivers.h
+++ b/include/acpi/acpi_drivers.h
@@ -57,7 +57,7 @@
  */
 
 #define ACPI_POWER_HID			"LNXPOWER"
-#define ACPI_PROCESSOR_OBJECT_HID	"ACPI_CPU"
+#define ACPI_PROCESSOR_OBJECT_HID	"LNXCPU"
 #define ACPI_PROCESSOR_HID		"ACPI0007"
 #define ACPI_SYSTEM_HID			"LNXSYSTM"
 #define ACPI_THERMAL_HID		"LNXTHERM"
-- 
cgit v1.2.3-59-g8ed1b


From 6cc73b4806c07b4207780f6d85c456b4c5b29d71 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Mon, 27 Apr 2009 16:33:41 -0600
Subject: ACPI: processor: check for synthetic _HID, default to "Device"
 declaration

This patch inverts the logic that distinguishes "Processor" statements
from "Device" statements, so we now check explicitly for "Processor" and
default to "Device".  This removes the only real use of ACPI_PROCESSOR_HID,
so we can then remove the #define.  It also has the theoretical advantage
that if a new processor _HID were ever added, we wouldn't have to change
the code here.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/processor_core.c | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c
index 45ad3288c5ff..cf627d64cde9 100644
--- a/drivers/acpi/processor_core.c
+++ b/drivers/acpi/processor_core.c
@@ -596,7 +596,21 @@ static int acpi_processor_get_info(struct acpi_device *device)
 		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
 				  "No bus mastering arbitration control\n"));
 
-	if (!strcmp(acpi_device_hid(device), ACPI_PROCESSOR_HID)) {
+	if (!strcmp(acpi_device_hid(device), ACPI_PROCESSOR_OBJECT_HID)) {
+		/* Declared with "Processor" statement; match ProcessorID */
+		status = acpi_evaluate_object(pr->handle, NULL, NULL, &buffer);
+		if (ACPI_FAILURE(status)) {
+			printk(KERN_ERR PREFIX "Evaluating processor object\n");
+			return -ENODEV;
+		}
+
+		/*
+		 * TBD: Synch processor ID (via LAPIC/LSAPIC structures) on SMP.
+		 *      >>> 'acpi_get_processor_id(acpi_id, &id)' in
+		 *      arch/xxx/acpi.c
+		 */
+		pr->acpi_id = object.processor.proc_id;
+	} else {
 		/*
 		 * Declared with "Device" statement; match _UID.
 		 * Note that we don't handle string _UIDs yet.
@@ -611,20 +625,6 @@ static int acpi_processor_get_info(struct acpi_device *device)
 		}
 		device_declaration = 1;
 		pr->acpi_id = value;
-	} else {
-		/* Declared with "Processor" statement; match ProcessorID */
-		status = acpi_evaluate_object(pr->handle, NULL, NULL, &buffer);
-		if (ACPI_FAILURE(status)) {
-			printk(KERN_ERR PREFIX "Evaluating processor object\n");
-			return -ENODEV;
-		}
-
-		/*
-		 * TBD: Synch processor ID (via LAPIC/LSAPIC structures) on SMP.
-		 *      >>> 'acpi_get_processor_id(acpi_id, &id)' in
-		 *      arch/xxx/acpi.c
-		 */
-		pr->acpi_id = object.processor.proc_id;
 	}
 	cpu_index = get_cpu_id(pr->handle, device_declaration, pr->acpi_id);
 
-- 
cgit v1.2.3-59-g8ed1b


From 9eccbc2f67efd0d19c47f40182abf2965c287add Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Mon, 27 Apr 2009 16:33:46 -0600
Subject: ACPI: processor: move device _HID into driver

The ACPI0007 _HID used for processor "Device" objects in the namespace
is not needed outside the processor driver, so move it there.  Also, the
#define is only used once, so just remove it and hard-code "ACPI0007".

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/processor_core.c | 2 +-
 include/acpi/acpi_drivers.h   | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c
index cf627d64cde9..cabff4cb21f0 100644
--- a/drivers/acpi/processor_core.c
+++ b/drivers/acpi/processor_core.c
@@ -89,7 +89,7 @@ static int acpi_processor_handle_eject(struct acpi_processor *pr);
 
 static const struct acpi_device_id processor_device_ids[] = {
 	{ACPI_PROCESSOR_OBJECT_HID, 0},
-	{ACPI_PROCESSOR_HID, 0},
+	{"ACPI0007", 0},
 	{"", 0},
 };
 MODULE_DEVICE_TABLE(acpi, processor_device_ids);
diff --git a/include/acpi/acpi_drivers.h b/include/acpi/acpi_drivers.h
index c9922b362005..5e8ed3a78cd0 100644
--- a/include/acpi/acpi_drivers.h
+++ b/include/acpi/acpi_drivers.h
@@ -58,7 +58,6 @@
 
 #define ACPI_POWER_HID			"LNXPOWER"
 #define ACPI_PROCESSOR_OBJECT_HID	"LNXCPU"
-#define ACPI_PROCESSOR_HID		"ACPI0007"
 #define ACPI_SYSTEM_HID			"LNXSYSTM"
 #define ACPI_THERMAL_HID		"LNXTHERM"
 #define ACPI_BUTTON_HID_POWERF		"LNXPWRBN"
-- 
cgit v1.2.3-59-g8ed1b


From 8cb24c8fd70ea8431744de1ca0ca34ab45fbbdaa Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 21 May 2009 15:49:59 -0600
Subject: PNPACPI: parse Extended Address Space Descriptors

Extended Address Space Descriptors are new in ACPI 3.0 and allow the
BIOS to communicate device resource cacheability attributes (write-back,
write-through, uncacheable, etc) to the OS.

Previously, PNPACPI ignored these descriptors, so if a BIOS used them,
a device could be responding at addresses the OS doesn't know about.
This patch adds support for these descriptors in _CRS and _PRS.  We
don't attempt to encode them for _SRS (just like we don't attempt to
encode the existing 16-, 32-, and 64-bit Address Space Descriptors).

Unfortunately, I don't have a way to test this.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/pnp/pnpacpi/rsparser.c | 46 ++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 44 insertions(+), 2 deletions(-)

diff --git a/drivers/pnp/pnpacpi/rsparser.c b/drivers/pnp/pnpacpi/rsparser.c
index adf17856bacc..e2a87fcfa6cf 100644
--- a/drivers/pnp/pnpacpi/rsparser.c
+++ b/drivers/pnp/pnpacpi/rsparser.c
@@ -287,6 +287,25 @@ static void pnpacpi_parse_allocated_address_space(struct pnp_dev *dev,
 				ACPI_DECODE_16);
 }
 
+static void pnpacpi_parse_allocated_ext_address_space(struct pnp_dev *dev,
+						      struct acpi_resource *res)
+{
+	struct acpi_resource_extended_address64 *p = &res->data.ext_address64;
+
+	if (p->producer_consumer == ACPI_PRODUCER)
+		return;
+
+	if (p->resource_type == ACPI_MEMORY_RANGE)
+		pnpacpi_parse_allocated_memresource(dev,
+			p->minimum, p->address_length,
+			p->info.mem.write_protect);
+	else if (p->resource_type == ACPI_IO_RANGE)
+		pnpacpi_parse_allocated_ioresource(dev,
+			p->minimum, p->address_length,
+			p->granularity == 0xfff ? ACPI_DECODE_10 :
+				ACPI_DECODE_16);
+}
+
 static acpi_status pnpacpi_allocated_resource(struct acpi_resource *res,
 					      void *data)
 {
@@ -400,8 +419,7 @@ static acpi_status pnpacpi_allocated_resource(struct acpi_resource *res,
 		break;
 
 	case ACPI_RESOURCE_TYPE_EXTENDED_ADDRESS64:
-		if (res->data.ext_address64.producer_consumer == ACPI_PRODUCER)
-			return AE_OK;
+		pnpacpi_parse_allocated_ext_address_space(dev, res);
 		break;
 
 	case ACPI_RESOURCE_TYPE_EXTENDED_IRQ:
@@ -630,6 +648,28 @@ static __init void pnpacpi_parse_address_option(struct pnp_dev *dev,
 					   IORESOURCE_IO_FIXED);
 }
 
+static __init void pnpacpi_parse_ext_address_option(struct pnp_dev *dev,
+						    unsigned int option_flags,
+						    struct acpi_resource *r)
+{
+	struct acpi_resource_extended_address64 *p = &r->data.ext_address64;
+	unsigned char flags = 0;
+
+	if (p->address_length == 0)
+		return;
+
+	if (p->resource_type == ACPI_MEMORY_RANGE) {
+		if (p->info.mem.write_protect == ACPI_READ_WRITE_MEMORY)
+			flags = IORESOURCE_MEM_WRITEABLE;
+		pnp_register_mem_resource(dev, option_flags, p->minimum,
+					  p->minimum, 0, p->address_length,
+					  flags);
+	} else if (p->resource_type == ACPI_IO_RANGE)
+		pnp_register_port_resource(dev, option_flags, p->minimum,
+					   p->minimum, 0, p->address_length,
+					   IORESOURCE_IO_FIXED);
+}
+
 struct acpipnp_parse_option_s {
 	struct pnp_dev *dev;
 	unsigned int option_flags;
@@ -711,6 +751,7 @@ static __init acpi_status pnpacpi_option_resource(struct acpi_resource *res,
 		break;
 
 	case ACPI_RESOURCE_TYPE_EXTENDED_ADDRESS64:
+		pnpacpi_parse_ext_address_option(dev, option_flags, res);
 		break;
 
 	case ACPI_RESOURCE_TYPE_EXTENDED_IRQ:
@@ -765,6 +806,7 @@ static int pnpacpi_supported_resource(struct acpi_resource *res)
 	case ACPI_RESOURCE_TYPE_ADDRESS16:
 	case ACPI_RESOURCE_TYPE_ADDRESS32:
 	case ACPI_RESOURCE_TYPE_ADDRESS64:
+	case ACPI_RESOURCE_TYPE_EXTENDED_ADDRESS64:
 	case ACPI_RESOURCE_TYPE_EXTENDED_IRQ:
 		return 1;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From bdf43bbf2e19952d82995a50e00cb4b66afa4f0c Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 21 May 2009 17:28:53 -0600
Subject: ACPI: don't check power state after _ON/_OFF

We used to evaluate _STA to check the power state of a device after
running _ON or _OFF.  But as far as I can tell, there's no benefit
to evaluating _STA, and sometimes we trip over bugs when BIOSes don't
implement _STA correctly.

Yakui says Windows XP doesn't evaluate _STA during power transition.
So let's skip it in Linux, too.  It's conceivable that we'll need to
check _STA in the future for some reason, but until we do, I don't
see a reason to clutter this code path.

References:
    http://bugzilla.kernel.org/show_bug.cgi?id=13243
    http://marc.info/?l=linux-acpi&m=124166053803753&w=2
    http://marc.info/?l=linux-acpi&m=124175761408256&w=2
    http://marc.info/?l=linux-acpi&m=124210593114061&w=2

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Acked-by: Matthew Garrett <mjg59@srcf.ucam.org>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/power.c | 28 ++--------------------------
 1 file changed, 2 insertions(+), 26 deletions(-)

diff --git a/drivers/acpi/power.c b/drivers/acpi/power.c
index 56665a63bf19..d74365d4a6e7 100644
--- a/drivers/acpi/power.c
+++ b/drivers/acpi/power.c
@@ -194,7 +194,7 @@ static int acpi_power_get_list_state(struct acpi_handle_list *list, int *state)
 
 static int acpi_power_on(acpi_handle handle, struct acpi_device *dev)
 {
-	int result = 0, state;
+	int result = 0;
 	int found = 0;
 	acpi_status status = AE_OK;
 	struct acpi_power_resource *resource = NULL;
@@ -236,18 +236,6 @@ static int acpi_power_on(acpi_handle handle, struct acpi_device *dev)
 	if (ACPI_FAILURE(status))
 		return -ENODEV;
 
-	if (!acpi_power_nocheck) {
-		/*
-		 * If acpi_power_nocheck is set, it is unnecessary to check
-		 * the power state after power transition.
-		 */
-		result = acpi_power_get_state(resource->device->handle,
-				&state);
-		if (result)
-			return result;
-		if (state != ACPI_POWER_RESOURCE_STATE_ON)
-			return -ENOEXEC;
-	}
 	/* Update the power resource's _device_ power state */
 	resource->device->power.state = ACPI_STATE_D0;
 
@@ -258,7 +246,7 @@ static int acpi_power_on(acpi_handle handle, struct acpi_device *dev)
 
 static int acpi_power_off_device(acpi_handle handle, struct acpi_device *dev)
 {
-	int result = 0, state;
+	int result = 0;
 	acpi_status status = AE_OK;
 	struct acpi_power_resource *resource = NULL;
 	struct list_head *node, *next;
@@ -293,18 +281,6 @@ static int acpi_power_off_device(acpi_handle handle, struct acpi_device *dev)
 	if (ACPI_FAILURE(status))
 		return -ENODEV;
 
-	if (!acpi_power_nocheck) {
-		/*
-		 * If acpi_power_nocheck is set, it is unnecessary to check
-		 * the power state after power transition.
-		 */
-		result = acpi_power_get_state(handle, &state);
-		if (result)
-			return result;
-		if (state != ACPI_POWER_RESOURCE_STATE_OFF)
-			return -ENOEXEC;
-	}
-
 	/* Update the power resource's _device_ power state */
 	resource->device->power.state = ACPI_STATE_D3;
 
-- 
cgit v1.2.3-59-g8ed1b


From 113b3a2b901573961509e81a28e9546cf9defef0 Mon Sep 17 00:00:00 2001
From: Len Brown <len.brown@intel.com>
Date: Wed, 27 May 2009 21:46:11 -0400
Subject: ACPI: delete acpi.power_nocheck from kernel-parameters.txt

Signed-off-by: Len Brown <len.brown@intel.com>
---
 Documentation/kernel-parameters.txt | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index fd5cac013037..ed3d8b966c02 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -229,14 +229,6 @@ and is between 256 and 4096 characters. It is defined in the file
 			to assume that this machine's pmtimer latches its value
 			and always returns good values.
 
- 	acpi.power_nocheck=	[HW,ACPI]
- 			Format: 1/0 enable/disable the check of power state.
- 			On some bogus BIOS the _PSC object/_STA object of
- 			power resource can't return the correct device power
- 			state. In such case it is unneccessary to check its
- 			power state again in power transition.
- 			1 : disable the power state check
-
 	acpi_sci=	[HW,ACPI] ACPI System Control Interrupt trigger mode
 			Format: { level | edge | high | low }
 
-- 
cgit v1.2.3-59-g8ed1b


From ee1ca48fae7e575d5e399d4fdcfe0afc1212a64c Mon Sep 17 00:00:00 2001
From: "Pallipadi, Venkatesh" <venkatesh.pallipadi@intel.com>
Date: Thu, 21 May 2009 17:09:10 -0700
Subject: ACPI: Disable ARB_DISABLE on platforms where it is not needed

ARB_DISABLE is a NOP on all of the recent Intel platforms.

For such platforms, reduce contention on c3_lock
by skipping the fake ARB_DISABLE.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 arch/x86/kernel/acpi/cstate.c | 16 +++++++++++++---
 drivers/acpi/processor_idle.c |  7 +++++--
 2 files changed, 18 insertions(+), 5 deletions(-)

diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c
index bbbe4bbb6f34..8c44c232efcb 100644
--- a/arch/x86/kernel/acpi/cstate.c
+++ b/arch/x86/kernel/acpi/cstate.c
@@ -34,12 +34,22 @@ void acpi_processor_power_init_bm_check(struct acpi_processor_flags *flags,
 		flags->bm_check = 1;
 	else if (c->x86_vendor == X86_VENDOR_INTEL) {
 		/*
-		 * Today all CPUs that support C3 share cache.
-		 * TBD: This needs to look at cache shared map, once
-		 * multi-core detection patch makes to the base.
+		 * Today all MP CPUs that support C3 share cache.
+		 * And caches should not be flushed by software while
+		 * entering C3 type state.
 		 */
 		flags->bm_check = 1;
 	}
+
+	/*
+	 * On all recent Intel platforms, ARB_DISABLE is a nop.
+	 * So, set bm_control to zero to indicate that ARB_DISABLE
+	 * is not required while entering C3 type state on
+	 * P4, Core and beyond CPUs
+	 */
+	if (c->x86_vendor == X86_VENDOR_INTEL &&
+	    (c->x86 > 0x6 || (c->x86 == 6 && c->x86_model >= 14)))
+			flags->bm_control = 0;
 }
 EXPORT_SYMBOL(acpi_processor_power_init_bm_check);
 
diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index 72069ba5f1ed..4840c79fd8e0 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -512,7 +512,8 @@ static void acpi_processor_power_verify_c2(struct acpi_processor_cx *cx)
 static void acpi_processor_power_verify_c3(struct acpi_processor *pr,
 					   struct acpi_processor_cx *cx)
 {
-	static int bm_check_flag;
+	static int bm_check_flag = -1;
+	static int bm_control_flag = -1;
 
 
 	if (!cx->address)
@@ -542,12 +543,14 @@ static void acpi_processor_power_verify_c3(struct acpi_processor *pr,
 	}
 
 	/* All the logic here assumes flags.bm_check is same across all CPUs */
-	if (!bm_check_flag) {
+	if (bm_check_flag == -1) {
 		/* Determine whether bm_check is needed based on CPU  */
 		acpi_processor_power_init_bm_check(&(pr->flags), pr->id);
 		bm_check_flag = pr->flags.bm_check;
+		bm_control_flag = pr->flags.bm_control;
 	} else {
 		pr->flags.bm_check = bm_check_flag;
+		pr->flags.bm_control = bm_control_flag;
 	}
 
 	if (pr->flags.bm_check) {
-- 
cgit v1.2.3-59-g8ed1b


From 8d42b524f4323cdf49872fe5e3a0e824f32feb51 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Fri, 29 May 2009 13:57:13 +0100
Subject: mtd: DIL/NetPC broken for now

We'll fix it up again, but for now I don't think anyone really cares.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/maps/Kconfig b/drivers/mtd/maps/Kconfig
index 82923bd2d9c5..cdd54cb4d224 100644
--- a/drivers/mtd/maps/Kconfig
+++ b/drivers/mtd/maps/Kconfig
@@ -270,7 +270,7 @@ config MTD_ALCHEMY
 
 config MTD_DILNETPC
 	tristate "CFI Flash device mapped on DIL/Net PC"
-	depends on X86 && MTD_CONCAT && MTD_PARTITIONS && MTD_CFI_INTELEXT
+	depends on X86 && MTD_CONCAT && MTD_PARTITIONS && MTD_CFI_INTELEXT && BROKEN
 	help
 	  MTD map driver for SSV DIL/Net PC Boards "DNP" and "ADNP".
 	  For details, see <http://www.ssv-embedded.de/ssv/pc104/p169.htm>
-- 
cgit v1.2.3-59-g8ed1b


From 9fd1e8f92ad17b3bc94245fee9b4e4bfea0dba0e Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Thu, 23 Apr 2009 17:41:10 +0100
Subject: mtd: Add armflash support for multiple blocks of flash

This patch adds MTD concatenation support to integrator-flash.c for
platforms with more than one block of flash memory (e.g. RealView
PB11MPCore). The implementation is based on the sa1100-flash.c one.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/integrator-flash.c | 226 ++++++++++++++++++++++++++----------
 1 file changed, 164 insertions(+), 62 deletions(-)

diff --git a/drivers/mtd/maps/integrator-flash.c b/drivers/mtd/maps/integrator-flash.c
index c9681a339a59..b08a798ee254 100644
--- a/drivers/mtd/maps/integrator-flash.c
+++ b/drivers/mtd/maps/integrator-flash.c
@@ -36,27 +36,33 @@
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/map.h>
 #include <linux/mtd/partitions.h>
+#include <linux/mtd/concat.h>
 
 #include <asm/mach/flash.h>
 #include <mach/hardware.h>
 #include <asm/system.h>
 
-#ifdef CONFIG_ARCH_P720T
-#define FLASH_BASE		(0x04000000)
-#define FLASH_SIZE		(64*1024*1024)
-#endif
+#define SUBDEV_NAME_SIZE	(BUS_ID_SIZE + 2)
 
-struct armflash_info {
+struct armflash_subdev_info {
+	char			name[SUBDEV_NAME_SIZE];
+	struct mtd_info		*mtd;
+	struct map_info		map;
 	struct flash_platform_data *plat;
+};
+
+struct armflash_info {
 	struct resource		*res;
 	struct mtd_partition	*parts;
 	struct mtd_info		*mtd;
-	struct map_info		map;
+	int			nr_subdev;
+	struct armflash_subdev_info subdev[0];
 };
 
 static void armflash_set_vpp(struct map_info *map, int on)
 {
-	struct armflash_info *info = container_of(map, struct armflash_info, map);
+	struct armflash_subdev_info *info =
+		container_of(map, struct armflash_subdev_info, map);
 
 	if (info->plat && info->plat->set_vpp)
 		info->plat->set_vpp(on);
@@ -64,32 +70,17 @@ static void armflash_set_vpp(struct map_info *map, int on)
 
 static const char *probes[] = { "cmdlinepart", "RedBoot", "afs", NULL };
 
-static int armflash_probe(struct platform_device *dev)
+static int armflash_subdev_probe(struct armflash_subdev_info *subdev,
+				 struct resource *res)
 {
-	struct flash_platform_data *plat = dev->dev.platform_data;
-	struct resource *res = dev->resource;
-	unsigned int size = res->end - res->start + 1;
-	struct armflash_info *info;
-	int err;
+	struct flash_platform_data *plat = subdev->plat;
+	resource_size_t size = res->end - res->start + 1;
 	void __iomem *base;
+	int err = 0;
 
-	info = kzalloc(sizeof(struct armflash_info), GFP_KERNEL);
-	if (!info) {
-		err = -ENOMEM;
-		goto out;
-	}
-
-	info->plat = plat;
-	if (plat && plat->init) {
-		err = plat->init();
-		if (err)
-			goto no_resource;
-	}
-
-	info->res = request_mem_region(res->start, size, "armflash");
-	if (!info->res) {
+	if (!request_mem_region(res->start, size, subdev->name)) {
 		err = -EBUSY;
-		goto no_resource;
+		goto out;
 	}
 
 	base = ioremap(res->start, size);
@@ -101,27 +92,132 @@ static int armflash_probe(struct platform_device *dev)
 	/*
 	 * look for CFI based flash parts fitted to this board
 	 */
-	info->map.size		= size;
-	info->map.bankwidth	= plat->width;
-	info->map.phys		= res->start;
-	info->map.virt		= base;
-	info->map.name		= dev_name(&dev->dev);
-	info->map.set_vpp	= armflash_set_vpp;
+	subdev->map.size	= size;
+	subdev->map.bankwidth	= plat->width;
+	subdev->map.phys	= res->start;
+	subdev->map.virt	= base;
+	subdev->map.name	= subdev->name;
+	subdev->map.set_vpp	= armflash_set_vpp;
 
-	simple_map_init(&info->map);
+	simple_map_init(&subdev->map);
 
 	/*
 	 * Also, the CFI layer automatically works out what size
 	 * of chips we have, and does the necessary identification
 	 * for us automatically.
 	 */
-	info->mtd = do_map_probe(plat->map_name, &info->map);
-	if (!info->mtd) {
+	subdev->mtd = do_map_probe(plat->map_name, &subdev->map);
+	if (!subdev->mtd) {
 		err = -ENXIO;
 		goto no_device;
 	}
 
-	info->mtd->owner = THIS_MODULE;
+	subdev->mtd->owner = THIS_MODULE;
+
+	/* Successful? */
+	if (err == 0)
+		return err;
+
+	if (subdev->mtd)
+		map_destroy(subdev->mtd);
+ no_device:
+	iounmap(base);
+ no_mem:
+	release_mem_region(res->start, size);
+ out:
+	return err;
+}
+
+static void armflash_subdev_remove(struct armflash_subdev_info *subdev)
+{
+	if (subdev->mtd)
+		map_destroy(subdev->mtd);
+	if (subdev->map.virt)
+		iounmap(subdev->map.virt);
+	release_mem_region(subdev->map.phys, subdev->map.size);
+}
+
+static int armflash_probe(struct platform_device *dev)
+{
+	struct flash_platform_data *plat = dev->dev.platform_data;
+	unsigned int size;
+	struct armflash_info *info;
+	int i, nr, err;
+
+	/* Count the number of devices */
+	for (nr = 0; ; nr++)
+		if (!platform_get_resource(dev, IORESOURCE_MEM, nr))
+			break;
+	if (nr == 0) {
+		err = -ENODEV;
+		goto out;
+	}
+
+	size = sizeof(struct armflash_info) +
+		sizeof(struct armflash_subdev_info) * nr;
+	info = kzalloc(size, GFP_KERNEL);
+	if (!info) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	if (plat && plat->init) {
+		err = plat->init();
+		if (err)
+			goto no_resource;
+	}
+
+	for (i = 0; i < nr; i++) {
+		struct armflash_subdev_info *subdev = &info->subdev[i];
+		struct resource *res;
+
+		res = platform_get_resource(dev, IORESOURCE_MEM, i);
+		if (!res)
+			break;
+
+		if (nr == 1)
+			/* No MTD concatenation, just use the default name */
+			snprintf(subdev->name, SUBDEV_NAME_SIZE, "%s",
+				 dev_name(&dev->dev));
+		else
+			snprintf(subdev->name, SUBDEV_NAME_SIZE, "%s-%d",
+				 dev_name(&dev->dev), i);
+		subdev->plat = plat;
+
+		err = armflash_subdev_probe(subdev, res);
+		if (err)
+			break;
+	}
+	info->nr_subdev = i;
+
+	if (err)
+		goto subdev_err;
+
+	if (info->nr_subdev == 1)
+		info->mtd = info->subdev[0].mtd;
+	else if (info->nr_subdev > 1) {
+#ifdef CONFIG_MTD_CONCAT
+		struct mtd_info *cdev[info->nr_subdev];
+
+		/*
+		 * We detected multiple devices.  Concatenate them together.
+		 */
+		for (i = 0; i < info->nr_subdev; i++)
+			cdev[i] = info->subdev[i].mtd;
+
+		info->mtd = mtd_concat_create(cdev, info->nr_subdev,
+					      dev_name(&dev->dev));
+		if (info->mtd == NULL)
+			err = -ENXIO;
+#else
+		printk(KERN_ERR "armflash: multiple devices found but "
+		       "MTD concat support disabled.\n");
+		err = -ENXIO;
+#endif
+	}
+
+	if (err < 0)
+		goto cleanup;
 
 	err = parse_mtd_partitions(info->mtd, probes, &info->parts, 0);
 	if (err > 0) {
@@ -131,28 +227,30 @@ static int armflash_probe(struct platform_device *dev)
 			       "mtd partition registration failed: %d\n", err);
 	}
 
-	if (err == 0)
+	if (err == 0) {
 		platform_set_drvdata(dev, info);
+		return err;
+	}
 
 	/*
-	 * If we got an error, free all resources.
+	 * We got an error, free all resources.
 	 */
-	if (err < 0) {
-		if (info->mtd) {
-			del_mtd_partitions(info->mtd);
-			map_destroy(info->mtd);
-		}
-		kfree(info->parts);
-
- no_device:
-		iounmap(base);
- no_mem:
-		release_mem_region(res->start, size);
- no_resource:
-		if (plat && plat->exit)
-			plat->exit();
-		kfree(info);
+ cleanup:
+	if (info->mtd) {
+		del_mtd_partitions(info->mtd);
+#ifdef CONFIG_MTD_CONCAT
+		if (info->mtd != info->subdev[0].mtd)
+			mtd_concat_destroy(info->mtd);
+#endif
 	}
+	kfree(info->parts);
+ subdev_err:
+	for (i = info->nr_subdev - 1; i >= 0; i--)
+		armflash_subdev_remove(&info->subdev[i]);
+ no_resource:
+	if (plat && plat->exit)
+		plat->exit();
+	kfree(info);
  out:
 	return err;
 }
@@ -160,22 +258,26 @@ static int armflash_probe(struct platform_device *dev)
 static int armflash_remove(struct platform_device *dev)
 {
 	struct armflash_info *info = platform_get_drvdata(dev);
+	struct flash_platform_data *plat = dev->dev.platform_data;
+	int i;
 
 	platform_set_drvdata(dev, NULL);
 
 	if (info) {
 		if (info->mtd) {
 			del_mtd_partitions(info->mtd);
-			map_destroy(info->mtd);
+#ifdef CONFIG_MTD_CONCAT
+			if (info->mtd != info->subdev[0].mtd)
+				mtd_concat_destroy(info->mtd);
+#endif
 		}
 		kfree(info->parts);
 
-		iounmap(info->map.virt);
-		release_resource(info->res);
-		kfree(info->res);
+		for (i = info->nr_subdev - 1; i >= 0; i--)
+			armflash_subdev_remove(&info->subdev[i]);
 
-		if (info->plat && info->plat->exit)
-			info->plat->exit();
+		if (plat && plat->exit)
+			plat->exit();
 
 		kfree(info);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 81933046ef2a615031c46171013bde2c5225ee69 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Fri, 29 May 2009 14:26:23 +0100
Subject: mtd: Fix handling of mtdname in txx9ndfmc.c

As pointed out by Kay Sievers, the name size limit is gone
from the driver-core, and BUS_ID_SIZE is obsolescent.

Rather than just papering over the problem by replacing the mtdname
array size with an arbitrary '20 + 2', fix the problem properly and
handle arbitrary name sizes.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/txx9ndfmc.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/nand/txx9ndfmc.c b/drivers/mtd/nand/txx9ndfmc.c
index 812479264896..5f919e63b29b 100644
--- a/drivers/mtd/nand/txx9ndfmc.c
+++ b/drivers/mtd/nand/txx9ndfmc.c
@@ -64,7 +64,7 @@ struct txx9ndfmc_priv {
 	struct nand_chip chip;
 	struct mtd_info mtd;
 	int cs;
-	char mtdname[BUS_ID_SIZE + 2];
+	const char *mtdname;
 };
 
 #define MAX_TXX9NDFMC_DEV	4
@@ -334,11 +334,17 @@ static int __init txx9ndfmc_probe(struct platform_device *dev)
 
 		if (plat->ch_mask != 1) {
 			txx9_priv->cs = i;
-			sprintf(txx9_priv->mtdname, "%s.%u",
-				dev_name(&dev->dev), i);
+			txx9_priv->mtdname = kasprintf(GFP_KERNEL, "%s.%u",
+						       dev_name(&dev->dev), i);
+			if (!txx9_priv->mtdname) {
+				kfree(txx9_priv);
+				dev_err(&dev->dev,
+					"Unable to allocate TXx9 NDFMC MTD device name.\n");
+				continue;
+			}
 		} else {
 			txx9_priv->cs = -1;
-			strcpy(txx9_priv->mtdname, dev_name(&dev->dev));
+			txx9_priv->mtdname = dev_name(&dev->dev);
 		}
 		if (plat->wide_mask & (1 << i))
 			chip->options |= NAND_BUSWIDTH_16;
@@ -385,6 +391,8 @@ static int __exit txx9ndfmc_remove(struct platform_device *dev)
 		kfree(drvdata->parts[i]);
 #endif
 		del_mtd_device(mtd);
+		if (txx9_priv->mtdname != dev_name(&dev->dev))
+			kfree(txx9_priv->mtdname);
 		kfree(txx9_priv);
 	}
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 19fe7f1a00023d2aa97617655b7ea56eb72f4db8 Mon Sep 17 00:00:00 2001
From: Kevin Cernekee <kpc.mtd@gmail.com>
Date: Wed, 8 Apr 2009 22:51:43 -0700
Subject: Documentation: add MTD sysfs docs

Signed-off-by: Kevin Cernekee <kpc.mtd@gmail.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 Documentation/ABI/testing/sysfs-class-mtd | 125 ++++++++++++++++++++++++++++++
 1 file changed, 125 insertions(+)
 create mode 100644 Documentation/ABI/testing/sysfs-class-mtd

diff --git a/Documentation/ABI/testing/sysfs-class-mtd b/Documentation/ABI/testing/sysfs-class-mtd
new file mode 100644
index 000000000000..4d55a1888981
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-class-mtd
@@ -0,0 +1,125 @@
+What:		/sys/class/mtd/
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		The mtd/ class subdirectory belongs to the MTD subsystem
+		(MTD core).
+
+What:		/sys/class/mtd/mtdX/
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		The /sys/class/mtd/mtd{0,1,2,3,...} directories correspond
+		to each /dev/mtdX character device.  These may represent
+		physical/simulated flash devices, partitions on a flash
+		device, or concatenated flash devices.  They exist regardless
+		of whether CONFIG_MTD_CHAR is actually enabled.
+
+What:		/sys/class/mtd/mtdXro/
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		These directories provide the corresponding read-only device
+		nodes for /sys/class/mtd/mtdX/ .  They are only created
+		(for the benefit of udev) if CONFIG_MTD_CHAR is enabled.
+
+What:		/sys/class/mtd/mtdX/dev
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		Major and minor numbers of the character device corresponding
+		to this MTD device (in <major>:<minor> format).  This is the
+		read-write device so <minor> will be even.
+
+What:		/sys/class/mtd/mtdXro/dev
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		Major and minor numbers of the character device corresponding
+		to the read-only variant of thie MTD device (in
+		<major>:<minor> format).  In this case <minor> will be odd.
+
+What:		/sys/class/mtd/mtdX/erasesize
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		"Major" erase size for the device.  If numeraseregions is
+		zero, this is the eraseblock size for the entire device.
+		Otherwise, the MEMGETREGIONCOUNT/MEMGETREGIONINFO ioctls
+		can be used to determine the actual eraseblock layout.
+
+What:		/sys/class/mtd/mtdX/flags
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		A hexadecimal value representing the device flags, ORed
+		together:
+
+		0x0400: MTD_WRITEABLE - device is writable
+		0x0800: MTD_BIT_WRITEABLE - single bits can be flipped
+		0x1000: MTD_NO_ERASE - no erase necessary
+		0x2000: MTD_POWERUP_LOCK - always locked after reset
+
+What:		/sys/class/mtd/mtdX/name
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		A human-readable ASCII name for the device or partition.
+		This will match the name in /proc/mtd .
+
+What:		/sys/class/mtd/mtdX/numeraseregions
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		For devices that have variable eraseblock sizes, this
+		provides the total number of erase regions.  Otherwise,
+		it will read back as zero.
+
+What:		/sys/class/mtd/mtdX/oobsize
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		Number of OOB bytes per page.
+
+What:		/sys/class/mtd/mtdX/size
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		Total size of the device/partition, in bytes.
+
+What:		/sys/class/mtd/mtdX/type
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		One of the following ASCII strings, representing the device
+		type:
+
+		absent, ram, rom, nor, nand, dataflash, ubi, unknown
+
+What:		/sys/class/mtd/mtdX/writesize
+Date:		April 2009
+KernelVersion:	2.6.29
+Contact:	linux-mtd@lists.infradead.org
+Description:
+		Minimal writable flash unit size.  This will always be
+		a positive integer.
+
+		In the case of NOR flash it is 1 (even though individual
+		bits can be cleared).
+
+		In the case of NAND flash it is one NAND page (or a
+		half page, or a quarter page).
+
+		In the case of ECC NOR, it is the ECC block size.
-- 
cgit v1.2.3-59-g8ed1b


From 0dc54e9f33e2fbcea28356bc2c8c931cb307d3b3 Mon Sep 17 00:00:00 2001
From: Kevin Cernekee <kpc.mtd@gmail.com>
Date: Wed, 8 Apr 2009 22:52:28 -0700
Subject: mtd: add MEMERASE64 ioctl for >4GiB devices

New MEMERASE/MEMREADOOB/MEMWRITEOOB ioctls are needed in order to support
64-bit offsets into large NAND flash devices.

Signed-off-by: Kevin Cernekee <kpc.mtd@gmail.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdchar.c | 29 +++++++++++++++++++++--------
 fs/compat_ioctl.c     |  1 +
 include/mtd/mtd-abi.h |  6 ++++++
 3 files changed, 28 insertions(+), 8 deletions(-)

diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
index 763d3f0a1f42..ad4b8618977d 100644
--- a/drivers/mtd/mtdchar.c
+++ b/drivers/mtd/mtdchar.c
@@ -417,6 +417,7 @@ static int mtd_ioctl(struct inode *inode, struct file *file,
 		break;
 
 	case MEMERASE:
+	case MEMERASE64:
 	{
 		struct erase_info *erase;
 
@@ -427,20 +428,32 @@ static int mtd_ioctl(struct inode *inode, struct file *file,
 		if (!erase)
 			ret = -ENOMEM;
 		else {
-			struct erase_info_user einfo;
-
 			wait_queue_head_t waitq;
 			DECLARE_WAITQUEUE(wait, current);
 
 			init_waitqueue_head(&waitq);
 
-			if (copy_from_user(&einfo, argp,
-				    sizeof(struct erase_info_user))) {
-				kfree(erase);
-				return -EFAULT;
+			if (cmd == MEMERASE64) {
+				struct erase_info_user64 einfo64;
+
+				if (copy_from_user(&einfo64, argp,
+					    sizeof(struct erase_info_user64))) {
+					kfree(erase);
+					return -EFAULT;
+				}
+				erase->addr = einfo64.start;
+				erase->len = einfo64.length;
+			} else {
+				struct erase_info_user einfo32;
+
+				if (copy_from_user(&einfo32, argp,
+					    sizeof(struct erase_info_user))) {
+					kfree(erase);
+					return -EFAULT;
+				}
+				erase->addr = einfo32.start;
+				erase->len = einfo32.length;
 			}
-			erase->addr = einfo.start;
-			erase->len = einfo.length;
 			erase->mtd = mtd;
 			erase->callback = mtdchar_erase_callback;
 			erase->priv = (unsigned long)&waitq;
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index b83f6bcfa51a..c603ca2c223a 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -2441,6 +2441,7 @@ COMPATIBLE_IOCTL(MEMGETREGIONCOUNT)
 COMPATIBLE_IOCTL(MEMGETREGIONINFO)
 COMPATIBLE_IOCTL(MEMGETBADBLOCK)
 COMPATIBLE_IOCTL(MEMSETBADBLOCK)
+COMPATIBLE_IOCTL(MEMERASE64)
 /* NBD */
 ULONG_IOCTL(NBD_SET_SOCK)
 ULONG_IOCTL(NBD_SET_BLKSIZE)
diff --git a/include/mtd/mtd-abi.h b/include/mtd/mtd-abi.h
index b6595b3c68b6..2e32be1e3a1e 100644
--- a/include/mtd/mtd-abi.h
+++ b/include/mtd/mtd-abi.h
@@ -12,6 +12,11 @@ struct erase_info_user {
 	__u32 length;
 };
 
+struct erase_info_user64 {
+	__u64 start;
+	__u64 length;
+};
+
 struct mtd_oob_buf {
 	__u32 start;
 	__u32 length;
@@ -95,6 +100,7 @@ struct otp_info {
 #define ECCGETLAYOUT		_IOR('M', 17, struct nand_ecclayout)
 #define ECCGETSTATS		_IOR('M', 18, struct mtd_ecc_stats)
 #define MTDFILEMODE		_IO('M', 19)
+#define MEMERASE64		_IOW('M', 20, struct erase_info_user64)
 
 /*
  * Obsolete legacy interface. Keep it in order not to break userspace
-- 
cgit v1.2.3-59-g8ed1b


From 977185404046afb31d2e18fac0a076de1a20bf0e Mon Sep 17 00:00:00 2001
From: Kevin Cernekee <kpc.mtd@gmail.com>
Date: Wed, 8 Apr 2009 22:53:13 -0700
Subject: mtd: compat_ioctl cleanup

1) Move the MEMREADOOB/MEMWRITEOOB compat_ioctl wrappers from
fs/compat_ioctl.c into mtdchar.c .  Original request was here:

http://lkml.org/lkml/2009/4/1/295

2) Add missing COMPATIBLE_IOCTL lines, so that mtd-utils does not error
out when running in 64/32 compatibility mode.

LKML-Reference: <200904011650.22928.arnd@arndb.de>
Signed-off-by: Kevin Cernekee <kpc.mtd@gmail.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdchar.c | 255 +++++++++++++++++++++++++++++++++-----------------
 fs/compat_ioctl.c     |  51 ++--------
 2 files changed, 180 insertions(+), 126 deletions(-)

diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
index ad4b8618977d..51bb0b092003 100644
--- a/drivers/mtd/mtdchar.c
+++ b/drivers/mtd/mtdchar.c
@@ -14,6 +14,7 @@
 #include <linux/sched.h>
 #include <linux/smp_lock.h>
 #include <linux/backing-dev.h>
+#include <linux/compat.h>
 
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/compatmac.h>
@@ -355,6 +356,100 @@ static int otp_select_filemode(struct mtd_file_info *mfi, int mode)
 # define otp_select_filemode(f,m)	-EOPNOTSUPP
 #endif
 
+static int mtd_do_writeoob(struct file *file, struct mtd_info *mtd,
+	uint64_t start, uint32_t length, void __user *ptr,
+	uint32_t __user *retp)
+{
+	struct mtd_oob_ops ops;
+	uint32_t retlen;
+	int ret = 0;
+
+	if (!(file->f_mode & FMODE_WRITE))
+		return -EPERM;
+
+	if (length > 4096)
+		return -EINVAL;
+
+	if (!mtd->write_oob)
+		ret = -EOPNOTSUPP;
+	else
+		ret = access_ok(VERIFY_READ, ptr, length) ? 0 : EFAULT;
+
+	if (ret)
+		return ret;
+
+	ops.ooblen = length;
+	ops.ooboffs = start & (mtd->oobsize - 1);
+	ops.datbuf = NULL;
+	ops.mode = MTD_OOB_PLACE;
+
+	if (ops.ooboffs && ops.ooblen > (mtd->oobsize - ops.ooboffs))
+		return -EINVAL;
+
+	ops.oobbuf = kmalloc(length, GFP_KERNEL);
+	if (!ops.oobbuf)
+		return -ENOMEM;
+
+	if (copy_from_user(ops.oobbuf, ptr, length)) {
+		kfree(ops.oobbuf);
+		return -EFAULT;
+	}
+
+	start &= ~((uint64_t)mtd->oobsize - 1);
+	ret = mtd->write_oob(mtd, start, &ops);
+
+	if (ops.oobretlen > 0xFFFFFFFFU)
+		ret = -EOVERFLOW;
+	retlen = ops.oobretlen;
+	if (copy_to_user(retp, &retlen, sizeof(length)))
+		ret = -EFAULT;
+
+	kfree(ops.oobbuf);
+	return ret;
+}
+
+static int mtd_do_readoob(struct mtd_info *mtd, uint64_t start,
+	uint32_t length, void __user *ptr, uint32_t __user *retp)
+{
+	struct mtd_oob_ops ops;
+	int ret = 0;
+
+	if (length > 4096)
+		return -EINVAL;
+
+	if (!mtd->read_oob)
+		ret = -EOPNOTSUPP;
+	else
+		ret = access_ok(VERIFY_WRITE, ptr,
+				length) ? 0 : -EFAULT;
+	if (ret)
+		return ret;
+
+	ops.ooblen = length;
+	ops.ooboffs = start & (mtd->oobsize - 1);
+	ops.datbuf = NULL;
+	ops.mode = MTD_OOB_PLACE;
+
+	if (ops.ooboffs && ops.ooblen > (mtd->oobsize - ops.ooboffs))
+		return -EINVAL;
+
+	ops.oobbuf = kmalloc(length, GFP_KERNEL);
+	if (!ops.oobbuf)
+		return -ENOMEM;
+
+	start &= ~((uint64_t)mtd->oobsize - 1);
+	ret = mtd->read_oob(mtd, start, &ops);
+
+	if (put_user(ops.oobretlen, retp))
+		ret = -EFAULT;
+	else if (ops.oobretlen && copy_to_user(ptr, ops.oobbuf,
+					    ops.oobretlen))
+		ret = -EFAULT;
+
+	kfree(ops.oobbuf);
+	return ret;
+}
+
 static int mtd_ioctl(struct inode *inode, struct file *file,
 		     u_int cmd, u_long arg)
 {
@@ -487,100 +582,28 @@ static int mtd_ioctl(struct inode *inode, struct file *file,
 	case MEMWRITEOOB:
 	{
 		struct mtd_oob_buf buf;
-		struct mtd_oob_ops ops;
-		struct mtd_oob_buf __user *user_buf = argp;
-	        uint32_t retlen;
-
-		if(!(file->f_mode & FMODE_WRITE))
-			return -EPERM;
-
-		if (copy_from_user(&buf, argp, sizeof(struct mtd_oob_buf)))
-			return -EFAULT;
-
-		if (buf.length > 4096)
-			return -EINVAL;
-
-		if (!mtd->write_oob)
-			ret = -EOPNOTSUPP;
-		else
-			ret = access_ok(VERIFY_READ, buf.ptr,
-					buf.length) ? 0 : EFAULT;
-
-		if (ret)
-			return ret;
-
-		ops.ooblen = buf.length;
-		ops.ooboffs = buf.start & (mtd->oobsize - 1);
-		ops.datbuf = NULL;
-		ops.mode = MTD_OOB_PLACE;
-
-		if (ops.ooboffs && ops.ooblen > (mtd->oobsize - ops.ooboffs))
-			return -EINVAL;
-
-		ops.oobbuf = kmalloc(buf.length, GFP_KERNEL);
-		if (!ops.oobbuf)
-			return -ENOMEM;
-
-		if (copy_from_user(ops.oobbuf, buf.ptr, buf.length)) {
-			kfree(ops.oobbuf);
-			return -EFAULT;
-		}
-
-		buf.start &= ~(mtd->oobsize - 1);
-		ret = mtd->write_oob(mtd, buf.start, &ops);
+		struct mtd_oob_buf __user *buf_user = argp;
 
-		if (ops.oobretlen > 0xFFFFFFFFU)
-			ret = -EOVERFLOW;
-		retlen = ops.oobretlen;
-		if (copy_to_user(&user_buf->length, &retlen, sizeof(buf.length)))
+		/* NOTE: writes return length to buf_user->length */
+		if (copy_from_user(&buf, argp, sizeof(buf)))
 			ret = -EFAULT;
-
-		kfree(ops.oobbuf);
+		else
+			ret = mtd_do_writeoob(file, mtd, buf.start, buf.length,
+				buf.ptr, &buf_user->length);
 		break;
-
 	}
 
 	case MEMREADOOB:
 	{
 		struct mtd_oob_buf buf;
-		struct mtd_oob_ops ops;
-
-		if (copy_from_user(&buf, argp, sizeof(struct mtd_oob_buf)))
-			return -EFAULT;
-
-		if (buf.length > 4096)
-			return -EINVAL;
-
-		if (!mtd->read_oob)
-			ret = -EOPNOTSUPP;
-		else
-			ret = access_ok(VERIFY_WRITE, buf.ptr,
-					buf.length) ? 0 : -EFAULT;
-		if (ret)
-			return ret;
-
-		ops.ooblen = buf.length;
-		ops.ooboffs = buf.start & (mtd->oobsize - 1);
-		ops.datbuf = NULL;
-		ops.mode = MTD_OOB_PLACE;
+		struct mtd_oob_buf __user *buf_user = argp;
 
-		if (ops.ooboffs && ops.ooblen > (mtd->oobsize - ops.ooboffs))
-			return -EINVAL;
-
-		ops.oobbuf = kmalloc(buf.length, GFP_KERNEL);
-		if (!ops.oobbuf)
-			return -ENOMEM;
-
-		buf.start &= ~(mtd->oobsize - 1);
-		ret = mtd->read_oob(mtd, buf.start, &ops);
-
-		if (put_user(ops.oobretlen, (uint32_t __user *)argp))
-			ret = -EFAULT;
-		else if (ops.oobretlen && copy_to_user(buf.ptr, ops.oobbuf,
-						    ops.oobretlen))
+		/* NOTE: writes return length to buf_user->start */
+		if (copy_from_user(&buf, argp, sizeof(buf)))
 			ret = -EFAULT;
-
-		kfree(ops.oobbuf);
+		else
+			ret = mtd_do_readoob(mtd, buf.start, buf.length,
+				buf.ptr, &buf_user->start);
 		break;
 	}
 
@@ -771,6 +794,67 @@ static int mtd_ioctl(struct inode *inode, struct file *file,
 	return ret;
 } /* memory_ioctl */
 
+#ifdef CONFIG_COMPAT
+
+struct mtd_oob_buf32 {
+	u_int32_t start;
+	u_int32_t length;
+	compat_caddr_t ptr;	/* unsigned char* */
+};
+
+#define MEMWRITEOOB32		_IOWR('M', 3, struct mtd_oob_buf32)
+#define MEMREADOOB32		_IOWR('M', 4, struct mtd_oob_buf32)
+
+static long mtd_compat_ioctl(struct file *file, unsigned int cmd,
+	unsigned long arg)
+{
+	struct mtd_file_info *mfi = file->private_data;
+	struct mtd_info *mtd = mfi->mtd;
+	void __user *argp = (void __user *)arg;
+	int ret = 0;
+
+	lock_kernel();
+
+	switch (cmd) {
+	case MEMWRITEOOB32:
+	{
+		struct mtd_oob_buf32 buf;
+		struct mtd_oob_buf32 __user *buf_user = argp;
+
+		if (copy_from_user(&buf, argp, sizeof(buf)))
+			ret = -EFAULT;
+		else
+			ret = mtd_do_writeoob(file, mtd, buf.start,
+				buf.length, compat_ptr(buf.ptr),
+				&buf_user->length);
+		break;
+	}
+
+	case MEMREADOOB32:
+	{
+		struct mtd_oob_buf32 buf;
+		struct mtd_oob_buf32 __user *buf_user = argp;
+
+		/* NOTE: writes return length to buf->start */
+		if (copy_from_user(&buf, argp, sizeof(buf)))
+			ret = -EFAULT;
+		else
+			ret = mtd_do_readoob(mtd, buf.start,
+				buf.length, compat_ptr(buf.ptr),
+				&buf_user->start);
+		break;
+	}
+	default:
+		ret = -ENOIOCTLCMD;
+	}
+
+	unlock_kernel();
+
+	return ret;
+}
+
+#endif /* CONFIG_COMPAT */
+
 /*
  * try to determine where a shared mapping can be made
  * - only supported for NOMMU at the moment (MMU can't doesn't copy private
@@ -830,6 +914,9 @@ static const struct file_operations mtd_fops = {
 	.read		= mtd_read,
 	.write		= mtd_write,
 	.ioctl		= mtd_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl	= mtd_compat_ioctl,
+#endif
 	.open		= mtd_open,
 	.release	= mtd_close,
 	.mmap		= mtd_mmap,
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index c603ca2c223a..196397bff086 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -1411,46 +1411,6 @@ static int ioc_settimeout(unsigned int fd, unsigned int cmd, unsigned long arg)
 #define HIDPGETCONNLIST	_IOR('H', 210, int)
 #define HIDPGETCONNINFO	_IOR('H', 211, int)
 
-struct mtd_oob_buf32 {
-	u_int32_t start;
-	u_int32_t length;
-	compat_caddr_t ptr;	/* unsigned char* */
-};
-
-#define MEMWRITEOOB32 	_IOWR('M',3,struct mtd_oob_buf32)
-#define MEMREADOOB32 	_IOWR('M',4,struct mtd_oob_buf32)
-
-static int mtd_rw_oob(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	struct mtd_oob_buf __user *buf = compat_alloc_user_space(sizeof(*buf));
-	struct mtd_oob_buf32 __user *buf32 = compat_ptr(arg);
-	u32 data;
-	char __user *datap;
-	unsigned int real_cmd;
-	int err;
-
-	real_cmd = (cmd == MEMREADOOB32) ?
-		MEMREADOOB : MEMWRITEOOB;
-
-	if (copy_in_user(&buf->start, &buf32->start,
-			 2 * sizeof(u32)) ||
-	    get_user(data, &buf32->ptr))
-		return -EFAULT;
-	datap = compat_ptr(data);
-	if (put_user(datap, &buf->ptr))
-		return -EFAULT;
-
-	err = sys_ioctl(fd, real_cmd, (unsigned long) buf);
-
-	if (!err) {
-		if (copy_in_user(&buf32->start, &buf->start,
-				 2 * sizeof(u32)))
-			err = -EFAULT;
-	}
-
-	return err;
-}	
-
 #ifdef CONFIG_BLOCK
 struct raw32_config_request
 {
@@ -2439,8 +2399,17 @@ COMPATIBLE_IOCTL(MEMLOCK)
 COMPATIBLE_IOCTL(MEMUNLOCK)
 COMPATIBLE_IOCTL(MEMGETREGIONCOUNT)
 COMPATIBLE_IOCTL(MEMGETREGIONINFO)
+COMPATIBLE_IOCTL(MEMSETOOBSEL)
+COMPATIBLE_IOCTL(MEMGETOOBSEL)
 COMPATIBLE_IOCTL(MEMGETBADBLOCK)
 COMPATIBLE_IOCTL(MEMSETBADBLOCK)
+COMPATIBLE_IOCTL(OTPSELECT)
+COMPATIBLE_IOCTL(OTPGETREGIONCOUNT)
+COMPATIBLE_IOCTL(OTPGETREGIONINFO)
+COMPATIBLE_IOCTL(OTPLOCK)
+COMPATIBLE_IOCTL(ECCGETLAYOUT)
+COMPATIBLE_IOCTL(ECCGETSTATS)
+COMPATIBLE_IOCTL(MTDFILEMODE)
 COMPATIBLE_IOCTL(MEMERASE64)
 /* NBD */
 ULONG_IOCTL(NBD_SET_SOCK)
@@ -2551,8 +2520,6 @@ COMPATIBLE_IOCTL(JSIOCGBUTTONS)
 COMPATIBLE_IOCTL(JSIOCGNAME(0))
 
 /* now things that need handlers */
-HANDLE_IOCTL(MEMREADOOB32, mtd_rw_oob)
-HANDLE_IOCTL(MEMWRITEOOB32, mtd_rw_oob)
 #ifdef CONFIG_NET
 HANDLE_IOCTL(SIOCGIFNAME, dev_ifname32)
 HANDLE_IOCTL(SIOCGIFCONF, dev_ifconf)
-- 
cgit v1.2.3-59-g8ed1b


From aea7cea9fa9e39e71f95ad70b3daf98ba9972587 Mon Sep 17 00:00:00 2001
From: Kevin Cernekee <kpc.mtd@gmail.com>
Date: Wed, 8 Apr 2009 22:53:49 -0700
Subject: mtd: add OOB ioctls for >4GiB devices

Signed-off-by: Kevin Cernekee <kpc.mtd@gmail.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdchar.c | 28 ++++++++++++++++++++++++++++
 fs/compat_ioctl.c     |  2 ++
 include/mtd/mtd-abi.h |  9 +++++++++
 3 files changed, 39 insertions(+)

diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
index 51bb0b092003..99d1fbc95011 100644
--- a/drivers/mtd/mtdchar.c
+++ b/drivers/mtd/mtdchar.c
@@ -607,6 +607,34 @@ static int mtd_ioctl(struct inode *inode, struct file *file,
 		break;
 	}
 
+	case MEMWRITEOOB64:
+	{
+		struct mtd_oob_buf64 buf;
+		struct mtd_oob_buf64 __user *buf_user = argp;
+
+		if (copy_from_user(&buf, argp, sizeof(buf)))
+			ret = -EFAULT;
+		else
+			ret = mtd_do_writeoob(file, mtd, buf.start, buf.length,
+				(void __user *)(uintptr_t)buf.usr_ptr,
+				&buf_user->length);
+		break;
+	}
+
+	case MEMREADOOB64:
+	{
+		struct mtd_oob_buf64 buf;
+		struct mtd_oob_buf64 __user *buf_user = argp;
+
+		if (copy_from_user(&buf, argp, sizeof(buf)))
+			ret = -EFAULT;
+		else
+			ret = mtd_do_readoob(mtd, buf.start, buf.length,
+				(void __user *)(uintptr_t)buf.usr_ptr,
+				&buf_user->length);
+		break;
+	}
+
 	case MEMLOCK:
 	{
 		struct erase_info_user einfo;
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 196397bff086..8da222eacbaf 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -2411,6 +2411,8 @@ COMPATIBLE_IOCTL(ECCGETLAYOUT)
 COMPATIBLE_IOCTL(ECCGETSTATS)
 COMPATIBLE_IOCTL(MTDFILEMODE)
 COMPATIBLE_IOCTL(MEMERASE64)
+COMPATIBLE_IOCTL(MEMREADOOB64)
+COMPATIBLE_IOCTL(MEMWRITEOOB64)
 /* NBD */
 ULONG_IOCTL(NBD_SET_SOCK)
 ULONG_IOCTL(NBD_SET_BLKSIZE)
diff --git a/include/mtd/mtd-abi.h b/include/mtd/mtd-abi.h
index 2e32be1e3a1e..be51ae2bd0ff 100644
--- a/include/mtd/mtd-abi.h
+++ b/include/mtd/mtd-abi.h
@@ -23,6 +23,13 @@ struct mtd_oob_buf {
 	unsigned char __user *ptr;
 };
 
+struct mtd_oob_buf64 {
+	__u64 start;
+	__u32 pad;
+	__u32 length;
+	__u64 usr_ptr;
+};
+
 #define MTD_ABSENT		0
 #define MTD_RAM			1
 #define MTD_ROM			2
@@ -101,6 +108,8 @@ struct otp_info {
 #define ECCGETSTATS		_IOR('M', 18, struct mtd_ecc_stats)
 #define MTDFILEMODE		_IO('M', 19)
 #define MEMERASE64		_IOW('M', 20, struct erase_info_user64)
+#define MEMWRITEOOB64		_IOWR('M', 21, struct mtd_oob_buf64)
+#define MEMREADOOB64		_IOWR('M', 22, struct mtd_oob_buf64)
 
 /*
  * Obsolete legacy interface. Keep it in order not to break userspace
-- 
cgit v1.2.3-59-g8ed1b


From 668ff9ab45d595222d3f90d7974ccba3518e3bb3 Mon Sep 17 00:00:00 2001
From: Kevin Cernekee <kpc.mtd@gmail.com>
Date: Tue, 14 Apr 2009 21:59:22 -0700
Subject: mtd: Handle compat ioctls directly; remove all trace from
 compat_ioctl.c

Remove all references to MTD ioctls from fs/compat_ioctl.c and let
them all be handled by mtd_compat_ioctl().

Signed-off-by: Kevin Cernekee <kpc.mtd@gmail.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdchar.c |  3 ++-
 fs/compat_ioctl.c     | 22 ----------------------
 2 files changed, 2 insertions(+), 23 deletions(-)

diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
index 99d1fbc95011..5fff04f3303d 100644
--- a/drivers/mtd/mtdchar.c
+++ b/drivers/mtd/mtdchar.c
@@ -836,6 +836,7 @@ struct mtd_oob_buf32 {
 static long mtd_compat_ioctl(struct file *file, unsigned int cmd,
 	unsigned long arg)
 {
+	struct inode *inode = file->f_path.dentry->d_inode;
 	struct mtd_file_info *mfi = file->private_data;
 	struct mtd_info *mtd = mfi->mtd;
 	void __user *argp = (void __user *)arg;
@@ -873,7 +874,7 @@ static long mtd_compat_ioctl(struct file *file, unsigned int cmd,
 		break;
 	}
 	default:
-		ret = -ENOIOCTLCMD;
+		ret = mtd_ioctl(inode, file, cmd, arg);
 	}
 
 	unlock_kernel();
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 8da222eacbaf..aa6ba39ff370 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -94,7 +94,6 @@
 #include <linux/atm_tcp.h>
 #include <linux/sonet.h>
 #include <linux/atm_suni.h>
-#include <linux/mtd/mtd.h>
 
 #include <linux/usb.h>
 #include <linux/usbdevice_fs.h>
@@ -2392,27 +2391,6 @@ COMPATIBLE_IOCTL(USBDEVFS_SUBMITURB32)
 COMPATIBLE_IOCTL(USBDEVFS_REAPURB32)
 COMPATIBLE_IOCTL(USBDEVFS_REAPURBNDELAY32)
 COMPATIBLE_IOCTL(USBDEVFS_CLEAR_HALT)
-/* MTD */
-COMPATIBLE_IOCTL(MEMGETINFO)
-COMPATIBLE_IOCTL(MEMERASE)
-COMPATIBLE_IOCTL(MEMLOCK)
-COMPATIBLE_IOCTL(MEMUNLOCK)
-COMPATIBLE_IOCTL(MEMGETREGIONCOUNT)
-COMPATIBLE_IOCTL(MEMGETREGIONINFO)
-COMPATIBLE_IOCTL(MEMSETOOBSEL)
-COMPATIBLE_IOCTL(MEMGETOOBSEL)
-COMPATIBLE_IOCTL(MEMGETBADBLOCK)
-COMPATIBLE_IOCTL(MEMSETBADBLOCK)
-COMPATIBLE_IOCTL(OTPSELECT)
-COMPATIBLE_IOCTL(OTPGETREGIONCOUNT)
-COMPATIBLE_IOCTL(OTPGETREGIONINFO)
-COMPATIBLE_IOCTL(OTPLOCK)
-COMPATIBLE_IOCTL(ECCGETLAYOUT)
-COMPATIBLE_IOCTL(ECCGETSTATS)
-COMPATIBLE_IOCTL(MTDFILEMODE)
-COMPATIBLE_IOCTL(MEMERASE64)
-COMPATIBLE_IOCTL(MEMREADOOB64)
-COMPATIBLE_IOCTL(MEMWRITEOOB64)
 /* NBD */
 ULONG_IOCTL(NBD_SET_SOCK)
 ULONG_IOCTL(NBD_SET_BLKSIZE)
-- 
cgit v1.2.3-59-g8ed1b


From 0b6585ce05f169f10ce74329e87bd8c5070b4bb9 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Fri, 29 May 2009 16:09:08 +0100
Subject: mtd: Fix pointer handling in compat ioctls to use compat_ptr()

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdchar.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
index 5fff04f3303d..5b081cb84351 100644
--- a/drivers/mtd/mtdchar.c
+++ b/drivers/mtd/mtdchar.c
@@ -839,7 +839,7 @@ static long mtd_compat_ioctl(struct file *file, unsigned int cmd,
 	struct inode *inode = file->f_path.dentry->d_inode;
 	struct mtd_file_info *mfi = file->private_data;
 	struct mtd_info *mtd = mfi->mtd;
-	void __user *argp = (void __user *)arg;
+	void __user *argp = compat_ptr(arg);
 	int ret = 0;
 
 	lock_kernel();
@@ -874,7 +874,7 @@ static long mtd_compat_ioctl(struct file *file, unsigned int cmd,
 		break;
 	}
 	default:
-		ret = mtd_ioctl(inode, file, cmd, arg);
+		ret = mtd_ioctl(inode, file, cmd, (unsigned long)argp);
 	}
 
 	unlock_kernel();
-- 
cgit v1.2.3-59-g8ed1b


From ec0482e6cfbd460bc69a9073ffbef4c2f3422fdf Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sat, 30 May 2009 16:55:29 +0100
Subject: [MTD] [NAND] S3C2410: Move to using platform device table

Commit 57fee4a58fe802272742caae248872c392a60670 added an
method to specify the platform device compatibility by using
an id-table instead of registering multiple drivers.

Move the S3C24XX NAND driver to using this ID table.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
CC: Eric Miao <eric.miao@marvell.com>
---
 drivers/mtd/nand/s3c2410.c | 78 ++++++++++++++++------------------------------
 1 file changed, 26 insertions(+), 52 deletions(-)

diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index 8e375d5fe231..b7f0740d842f 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -593,7 +593,7 @@ static inline void s3c2410_nand_cpufreq_deregister(struct s3c2410_nand_info *inf
 
 /* device management functions */
 
-static int s3c2410_nand_remove(struct platform_device *pdev)
+static int s3c24xx_nand_remove(struct platform_device *pdev)
 {
 	struct s3c2410_nand_info *info = to_nand_info(pdev);
 
@@ -788,18 +788,17 @@ static void s3c2410_nand_update_chip(struct s3c2410_nand_info *info,
 	}
 }
 
-/* s3c2410_nand_probe
+/* s3c24xx_nand_probe
  *
  * called by device layer when it finds a device matching
  * one our driver can handled. This code checks to see if
  * it can allocate all necessary resources then calls the
  * nand layer to look for devices
 */
-
-static int s3c24xx_nand_probe(struct platform_device *pdev,
-			      enum s3c_cpu_type cpu_type)
+static int s3c24xx_nand_probe(struct platform_device *pdev)
 {
 	struct s3c2410_platform_nand *plat = to_nand_plat(pdev);
+	enum s3c_cpu_type cpu_type; 
 	struct s3c2410_nand_info *info;
 	struct s3c2410_nand_mtd *nmtd;
 	struct s3c2410_nand_set *sets;
@@ -809,6 +808,8 @@ static int s3c24xx_nand_probe(struct platform_device *pdev,
 	int nr_sets;
 	int setno;
 
+	cpu_type = platform_get_device_id(pdev)->driver_data;
+
 	pr_debug("s3c2410_nand_probe(%p)\n", pdev);
 
 	info = kmalloc(sizeof(*info), GFP_KERNEL);
@@ -922,7 +923,7 @@ static int s3c24xx_nand_probe(struct platform_device *pdev,
 	return 0;
 
  exit_error:
-	s3c2410_nand_remove(pdev);
+	s3c24xx_nand_remove(pdev);
 
 	if (err == 0)
 		err = -EINVAL;
@@ -983,50 +984,30 @@ static int s3c24xx_nand_resume(struct platform_device *dev)
 
 /* driver device registration */
 
-static int s3c2410_nand_probe(struct platform_device *dev)
-{
-	return s3c24xx_nand_probe(dev, TYPE_S3C2410);
-}
-
-static int s3c2440_nand_probe(struct platform_device *dev)
-{
-	return s3c24xx_nand_probe(dev, TYPE_S3C2440);
-}
-
-static int s3c2412_nand_probe(struct platform_device *dev)
-{
-	return s3c24xx_nand_probe(dev, TYPE_S3C2412);
-}
-
-static struct platform_driver s3c2410_nand_driver = {
-	.probe		= s3c2410_nand_probe,
-	.remove		= s3c2410_nand_remove,
-	.suspend	= s3c24xx_nand_suspend,
-	.resume		= s3c24xx_nand_resume,
-	.driver		= {
-		.name	= "s3c2410-nand",
-		.owner	= THIS_MODULE,
+static struct platform_device_id s3c24xx_driver_ids[] = {
+	{
+		.name		= "s3c2410-nand",
+		.driver_data	= TYPE_S3C2410,
+	}, {
+		.name		= "s3c2440-nand",
+		.driver_data	= TYPE_S3C2440,
+	}, {
+		.name		= "s3c2412-nand",
+		.driver_data	= TYPE_S3C2412,
 	},
+	{ }
 };
 
-static struct platform_driver s3c2440_nand_driver = {
-	.probe		= s3c2440_nand_probe,
-	.remove		= s3c2410_nand_remove,
-	.suspend	= s3c24xx_nand_suspend,
-	.resume		= s3c24xx_nand_resume,
-	.driver		= {
-		.name	= "s3c2440-nand",
-		.owner	= THIS_MODULE,
-	},
-};
+MODULE_DEVICE_TABLE(platform, s3c24xx_driver_ids);
 
-static struct platform_driver s3c2412_nand_driver = {
-	.probe		= s3c2412_nand_probe,
-	.remove		= s3c2410_nand_remove,
+static struct platform_driver s3c24xx_nand_driver = {
+	.probe		= s3c24xx_nand_probe,
+	.remove		= s3c24xx_nand_remove,
 	.suspend	= s3c24xx_nand_suspend,
 	.resume		= s3c24xx_nand_resume,
+	.id_table	= s3c24xx_driver_ids,
 	.driver		= {
-		.name	= "s3c2412-nand",
+		.name	= "s3c24xx-nand",
 		.owner	= THIS_MODULE,
 	},
 };
@@ -1035,16 +1016,12 @@ static int __init s3c2410_nand_init(void)
 {
 	printk("S3C24XX NAND Driver, (c) 2004 Simtec Electronics\n");
 
-	platform_driver_register(&s3c2412_nand_driver);
-	platform_driver_register(&s3c2440_nand_driver);
-	return platform_driver_register(&s3c2410_nand_driver);
+	return platform_driver_register(&s3c24xx_nand_driver);
 }
 
 static void __exit s3c2410_nand_exit(void)
 {
-	platform_driver_unregister(&s3c2412_nand_driver);
-	platform_driver_unregister(&s3c2440_nand_driver);
-	platform_driver_unregister(&s3c2410_nand_driver);
+	platform_driver_unregister(&s3c24xx_nand_driver);
 }
 
 module_init(s3c2410_nand_init);
@@ -1053,6 +1030,3 @@ module_exit(s3c2410_nand_exit);
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Ben Dooks <ben@simtec.co.uk>");
 MODULE_DESCRIPTION("S3C24XX MTD NAND driver");
-MODULE_ALIAS("platform:s3c2410-nand");
-MODULE_ALIAS("platform:s3c2412-nand");
-MODULE_ALIAS("platform:s3c2440-nand");
-- 
cgit v1.2.3-59-g8ed1b


From 3db72151aa4c246f8bdb8b3501972e1f1b32fe0d Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sat, 30 May 2009 17:18:15 +0100
Subject: [MTD] [NAND] S3C2410: Basic kerneldoc comment updates

Move to using kerneldoc style commenting in the driver

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 drivers/mtd/nand/s3c2410.c | 89 ++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 78 insertions(+), 11 deletions(-)

diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index b7f0740d842f..a2d1c70c5227 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -74,6 +74,14 @@ static struct nand_ecclayout nand_hw_eccoob = {
 
 struct s3c2410_nand_info;
 
+/**
+ * struct s3c2410_nand_mtd - driver MTD structure
+ * @mtd: The MTD instance to pass to the MTD layer.
+ * @chip: The NAND chip information.
+ * @set: The platform information supplied for this set of NAND chips.
+ * @info: Link back to the hardware information.
+ * @scan_res: The result from calling nand_scan_ident().
+*/
 struct s3c2410_nand_mtd {
 	struct mtd_info			mtd;
 	struct nand_chip		chip;
@@ -90,6 +98,21 @@ enum s3c_cpu_type {
 
 /* overview of the s3c2410 nand state */
 
+/**
+ * struct s3c2410_nand_info - NAND controller state.
+ * @mtds: An array of MTD instances on this controoler.
+ * @platform: The platform data for this board.
+ * @device: The platform device we bound to.
+ * @area: The IO area resource that came from request_mem_region().
+ * @clk: The clock resource for this controller.
+ * @regs: The area mapped for the hardware registers described by @area.
+ * @sel_reg: Pointer to the register controlling the NAND selection.
+ * @sel_bit: The bit in @sel_reg to select the NAND chip.
+ * @mtd_count: The number of MTDs created from this controller.
+ * @save_sel: The contents of @sel_reg to be saved over suspend.
+ * @clk_rate: The clock rate from @clk.
+ * @cpu_type: The exact type of this controller.
+ */
 struct s3c2410_nand_info {
 	/* mtd info */
 	struct nand_hw_control		controller;
@@ -145,6 +168,14 @@ static inline int allow_clk_stop(struct s3c2410_nand_info *info)
 
 #define NS_IN_KHZ 1000000
 
+/**
+ * s3c_nand_calc_rate - calculate timing data.
+ * @wanted: The cycle time in nanoseconds.
+ * @clk: The clock rate in kHz.
+ * @max: The maximum divider value.
+ *
+ * Calculate the timing value from the given parameters.
+ */
 static int s3c_nand_calc_rate(int wanted, unsigned long clk, int max)
 {
 	int result;
@@ -169,6 +200,14 @@ static int s3c_nand_calc_rate(int wanted, unsigned long clk, int max)
 
 /* controller setup */
 
+/**
+ * s3c2410_nand_setrate - setup controller timing information.
+ * @info: The controller instance.
+ *
+ * Given the information supplied by the platform, calculate and set
+ * the necessary timing registers in the hardware to generate the
+ * necessary timing cycles to the hardware.
+ */
 static int s3c2410_nand_setrate(struct s3c2410_nand_info *info)
 {
 	struct s3c2410_platform_nand *plat = info->platform;
@@ -245,6 +284,13 @@ static int s3c2410_nand_setrate(struct s3c2410_nand_info *info)
 	return 0;
 }
 
+/**
+ * s3c2410_nand_inithw - basic hardware initialisation
+ * @info: The hardware state.
+ *
+ * Do the basic initialisation of the hardware, using s3c2410_nand_setrate()
+ * to setup the hardware access speeds and set the controller to be enabled.
+*/
 static int s3c2410_nand_inithw(struct s3c2410_nand_info *info)
 {
 	int ret;
@@ -268,8 +314,19 @@ static int s3c2410_nand_inithw(struct s3c2410_nand_info *info)
 	return 0;
 }
 
-/* select chip */
-
+/**
+ * s3c2410_nand_select_chip - select the given nand chip
+ * @mtd: The MTD instance for this chip.
+ * @chip: The chip number.
+ *
+ * This is called by the MTD layer to either select a given chip for the
+ * @mtd instance, or to indicate that the access has finished and the
+ * chip can be de-selected.
+ *
+ * The routine ensures that the nFCE line is correctly setup, and any
+ * platform specific selection code is called to route nFCE to the specific
+ * chip.
+ */
 static void s3c2410_nand_select_chip(struct mtd_info *mtd, int chip)
 {
 	struct s3c2410_nand_info *info;
@@ -667,11 +724,16 @@ static int s3c2410_nand_add_partition(struct s3c2410_nand_info *info,
 }
 #endif
 
-/* s3c2410_nand_init_chip
+/**
+ * s3c2410_nand_init_chip - initialise a single instance of an chip
+ * @info: The base NAND controller the chip is on.
+ * @nmtd: The new controller MTD instance to fill in.
+ * @set: The information passed from the board specific platform data.
  *
- * init a single instance of an chip
-*/
-
+ * Initialise the given @nmtd from the information in @info and @set. This
+ * readies the structure for use with the MTD layer functions by ensuring
+ * all pointers are setup and the necessary control routines selected.
+ */
 static void s3c2410_nand_init_chip(struct s3c2410_nand_info *info,
 				   struct s3c2410_nand_mtd *nmtd,
 				   struct s3c2410_nand_set *set)
@@ -759,12 +821,17 @@ static void s3c2410_nand_init_chip(struct s3c2410_nand_info *info,
 		chip->ecc.mode	= NAND_ECC_NONE;
 }
 
-/* s3c2410_nand_update_chip
+/**
+ * s3c2410_nand_update_chip - post probe update
+ * @info: The controller instance.
+ * @nmtd: The driver version of the MTD instance.
  *
- * post-probe chip update, to change any items, such as the
- * layout for large page nand
- */
-
+ * This routine is called after the chip probe has succesfully completed
+ * and the relevant per-chip information updated. This call ensure that
+ * we update the internal state accordingly.
+ *
+ * The internal state is currently limited to the ECC state information.
+*/
 static void s3c2410_nand_update_chip(struct s3c2410_nand_info *info,
 				     struct s3c2410_nand_mtd *nmtd)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 8c3e843d56f74889f3ff32202e82e3bc16d0d552 Mon Sep 17 00:00:00 2001
From: Andy Green <andy@warmcat.com>
Date: Sun, 10 May 2009 15:41:25 -0500
Subject: [MTD] [NAND] S3C2410: NAND ECC by chip rather than global

This makes us take note about the chosen ECC mode per-chip and
not the one set globally.

Signed-off-by: Andy Green <andy@warmcat.com>
Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>
[ben-linux@fluff.org: andy@openmoko.com => andy@warmcat.com, rewrite subject]
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 drivers/mtd/nand/s3c2410.c | 34 +++++++++++++++++++++++++---------
 1 file changed, 25 insertions(+), 9 deletions(-)

diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index a2d1c70c5227..daa4af918205 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -819,6 +819,21 @@ static void s3c2410_nand_init_chip(struct s3c2410_nand_info *info,
 
 	if (set->disable_ecc)
 		chip->ecc.mode	= NAND_ECC_NONE;
+
+	switch (chip->ecc.mode) {
+	case NAND_ECC_NONE:
+		dev_info(info->device, "NAND ECC disabled\n");
+		break;
+	case NAND_ECC_SOFT:
+		dev_info(info->device, "NAND soft ECC\n");
+		break;
+	case NAND_ECC_HW:
+		dev_info(info->device, "NAND hardware ECC\n");
+		break;
+	default:
+		dev_info(info->device, "NAND ECC UNKNOWN\n");
+		break;
+	}
 }
 
 /**
@@ -840,18 +855,19 @@ static void s3c2410_nand_update_chip(struct s3c2410_nand_info *info,
 	dev_dbg(info->device, "chip %p => page shift %d\n",
 		chip, chip->page_shift);
 
-	if (hardware_ecc) {
+	if (chip->ecc.mode != NAND_ECC_HW)
+		return;
+
 		/* change the behaviour depending on wether we are using
 		 * the large or small page nand device */
 
-		if (chip->page_shift > 10) {
-			chip->ecc.size	    = 256;
-			chip->ecc.bytes	    = 3;
-		} else {
-			chip->ecc.size	    = 512;
-			chip->ecc.bytes	    = 3;
-			chip->ecc.layout    = &nand_hw_eccoob;
-		}
+	if (chip->page_shift > 10) {
+		chip->ecc.size	    = 256;
+		chip->ecc.bytes	    = 3;
+	} else {
+		chip->ecc.size	    = 512;
+		chip->ecc.bytes	    = 3;
+		chip->ecc.layout    = &nand_hw_eccoob;
 	}
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From fb6ea3258742f41be8dd527ebd58a1a1e2cef23a Mon Sep 17 00:00:00 2001
From: Nelson Castillo <arhuaco@freaks-unidos.net>
Date: Sun, 10 May 2009 15:41:32 -0500
Subject: [MTD] [NAND] S3C2410: Small colon cleanup

checkpatch would complain with "disable_ecc : 1".

Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>
[ben-linux@fluff.org: subject cleanup]
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c/include/plat/nand.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm/plat-s3c/include/plat/nand.h b/arch/arm/plat-s3c/include/plat/nand.h
index f4dcd14af059..985546750c6d 100644
--- a/arch/arm/plat-s3c/include/plat/nand.h
+++ b/arch/arm/plat-s3c/include/plat/nand.h
@@ -22,7 +22,7 @@
 */
 
 struct s3c2410_nand_set {
-	unsigned int		disable_ecc : 1;
+	unsigned int		disable_ecc:1;
 
 	int			nr_chips;
 	int			nr_partitions;
@@ -39,7 +39,7 @@ struct s3c2410_platform_nand {
 	int	twrph0;	/* active time for nWE/nOE */
 	int	twrph1;	/* time for release CLE/ALE from nWE/nOE inactive */
 
-	unsigned int	ignore_unset_ecc : 1;
+	unsigned int	ignore_unset_ecc:1;
 
 	int			nr_sets;
 	struct s3c2410_nand_set *sets;
-- 
cgit v1.2.3-59-g8ed1b


From 2612e523dc3695df319662ff279806a3d74de375 Mon Sep 17 00:00:00 2001
From: Nelson Castillo <arhuaco@freaks-unidos.net>
Date: Sun, 10 May 2009 15:41:54 -0500
Subject: [MTD] [NAND] S3C2410: Uninitialised variable cleanup

~ Avoid warning without generating code.
  (I don't even get the warning without the macro uninitialized_var).

Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>
[ben-linux@fluff.org: subject cleanup]
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 drivers/mtd/nand/s3c2410.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index daa4af918205..7be3663df1e5 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -214,7 +214,7 @@ static int s3c2410_nand_setrate(struct s3c2410_nand_info *info)
 	int tacls_max = (info->cpu_type == TYPE_S3C2412) ? 8 : 4;
 	int tacls, twrph0, twrph1;
 	unsigned long clkrate = clk_get_rate(info->clk);
-	unsigned long set, cfg, mask;
+	unsigned long uninitialized_var(set), cfg, uninitialized_var(mask);
 	unsigned long flags;
 
 	/* calculate the timing information for the controller */
@@ -264,9 +264,6 @@ static int s3c2410_nand_setrate(struct s3c2410_nand_info *info)
 		break;
 
 	default:
-		/* keep compiler happy */
-		mask = 0;
-		set = 0;
 		BUG();
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From ae7304e554642d57993b32265b817e6ae80787de Mon Sep 17 00:00:00 2001
From: Andy Green <andy@warmcat.com>
Date: Sun, 10 May 2009 15:42:02 -0500
Subject: [MTD] [NAND] S3C2410: Fix CFG debug order

Fix NAND CFG debug order.

Signed-off-by: Andy Green <andy@warmcat.com>
Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>
[ben-linux@fluff.org: Change andy@openmoko.com to andy@warmcat.com, subject cleanup]
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 drivers/mtd/nand/s3c2410.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index 7be3663df1e5..87c40deb27bb 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -267,8 +267,6 @@ static int s3c2410_nand_setrate(struct s3c2410_nand_info *info)
 		BUG();
 	}
 
-	dev_dbg(info->device, "NF_CONF is 0x%lx\n", cfg);
-
 	local_irq_save(flags);
 
 	cfg = readl(info->regs + S3C2410_NFCONF);
@@ -278,6 +276,8 @@ static int s3c2410_nand_setrate(struct s3c2410_nand_info *info)
 
 	local_irq_restore(flags);
 
+	dev_dbg(info->device, "NF_CONF is 0x%lx\n", cfg);
+
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From ed27f0287062236d50190d7447f6377ff4acdfad Mon Sep 17 00:00:00 2001
From: Andy Green <andy@warmcat.com>
Date: Sun, 10 May 2009 15:42:09 -0500
Subject: [MTD] [NAND] S3C2410: Allow commandline partition processing

This patch allows commandline partition processing to
work with the s3c2410 NAND platform driver.

Signed-off-by: Andy Green <andy@warcat.com>
Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>
[ben-linux@fluff.org: Change andy@openmoko.com to andy@warmcat.com]
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 drivers/mtd/nand/s3c2410.c | 18 ++++++++++++++++--
 1 file changed, 16 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index 87c40deb27bb..ef5665258968 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -699,17 +699,31 @@ static int s3c24xx_nand_remove(struct platform_device *pdev)
 }
 
 #ifdef CONFIG_MTD_PARTITIONS
+const char *part_probes[] = { "cmdlinepart", NULL };
 static int s3c2410_nand_add_partition(struct s3c2410_nand_info *info,
 				      struct s3c2410_nand_mtd *mtd,
 				      struct s3c2410_nand_set *set)
 {
+	struct mtd_partition *part_info;
+	int nr_part = 0;
+
 	if (set == NULL)
 		return add_mtd_device(&mtd->mtd);
 
-	if (set->nr_partitions > 0 && set->partitions != NULL) {
-		return add_mtd_partitions(&mtd->mtd, set->partitions, set->nr_partitions);
+	if (set->nr_partitions == 0) {
+		mtd->mtd.name = set->name;
+		nr_part = parse_mtd_partitions(&mtd->mtd, part_probes,
+						&part_info, 0);
+	} else {
+		if (set->nr_partitions > 0 && set->partitions != NULL) {
+			nr_part = set->nr_partitions;
+			part_info = set->partitions;
+		}
 	}
 
+	if (nr_part > 0 && part_info)
+		return add_mtd_partitions(&mtd->mtd, part_info, nr_part);
+
 	return add_mtd_device(&mtd->mtd);
 }
 #else
-- 
cgit v1.2.3-59-g8ed1b


From 94d72176f69954d7a20e95e97dc101a4b521ce57 Mon Sep 17 00:00:00 2001
From: Roel Kluin <roel.kluin@gmail.com>
Date: Tue, 12 May 2009 13:47:21 -0700
Subject: uwb: event_size should be signed

event_size should be ssize_t to notice when hwarc_get_event_size() returns
-ENOSPC.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Cc: David Vrabel <david.vrabel@csr.com>
Cc: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/uwb/hwa-rc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/uwb/hwa-rc.c b/drivers/uwb/hwa-rc.c
index 559f8784acf3..9052bcb4f528 100644
--- a/drivers/uwb/hwa-rc.c
+++ b/drivers/uwb/hwa-rc.c
@@ -501,7 +501,7 @@ int hwarc_filter_event_WUSB_0100(struct uwb_rc *rc, struct uwb_rceb **header,
 	int result = -ENOANO;
 	struct uwb_rceb *rceb = *header;
 	int event = le16_to_cpu(rceb->wEvent);
-	size_t event_size;
+	ssize_t event_size;
 	size_t core_size, offset;
 
 	if (rceb->bEventType != UWB_RC_CET_GENERAL)
-- 
cgit v1.2.3-59-g8ed1b


From b81c087f6deb049023e41ce00717202a953f3939 Mon Sep 17 00:00:00 2001
From: Frank Leipold <frank.leipold@eads.net>
Date: Mon, 1 Jun 2009 12:03:15 +0100
Subject: uwb: allow WLP to be used with IPv6.

Ethernet multicast addresses are supported by mapping them to
broadcast WLP frames. These are frequently used in IPv6 traffic.

Signed-off-by: Frank Leipold <frank.leipold@eads.net>
Signed-off-by: David Vrabel <david.vrabel@csr.com>
---
 drivers/uwb/wlp/txrx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/uwb/wlp/txrx.c b/drivers/uwb/wlp/txrx.c
index cd2035768b47..86a853b84119 100644
--- a/drivers/uwb/wlp/txrx.c
+++ b/drivers/uwb/wlp/txrx.c
@@ -326,7 +326,7 @@ int wlp_prepare_tx_frame(struct device *dev, struct wlp *wlp,
 	int result = -EINVAL;
 	struct ethhdr *eth_hdr = (void *) skb->data;
 
-	if (is_broadcast_ether_addr(eth_hdr->h_dest)) {
+	if (is_multicast_ether_addr(eth_hdr->h_dest)) {
 		result = wlp_eda_for_each(&wlp->eda, wlp_wss_send_copy, skb);
 		if (result < 0) {
 			if (printk_ratelimit())
-- 
cgit v1.2.3-59-g8ed1b


From 3c8e03166ae234d16e7871f8009638e0946d303c Mon Sep 17 00:00:00 2001
From: Yu Zhiguo <yuzg@cn.fujitsu.com>
Date: Sat, 16 May 2009 16:22:31 +0800
Subject: NFSv4: do exact check about attribute specified

Server should return NFS4ERR_ATTRNOTSUPP if an attribute specified is
not supported in current environment.
Operations CREATE, NVERIFY, OPEN, SETATTR and VERIFY should do this check.

This bug is found when do newpynfs tests. The names of the tests that failed
are following:
  CR12 NVF7a NVF7b NVF7c NVF7d NVF7f NVF7r NVF7s
  OPEN15 VF7a VF7b VF7c VF7d VF7f VF7r VF7s

Add function do_check_fattr() to do exact check:
1, Check attribute specified is supported by the NFSv4 server or not.
2, Check FATTR4_WORD0_ACL & FATTR4_WORD0_FS_LOCATIONS are supported
   in current environment or not.
3, Check attribute specified is writable or not.

step 1 and 3 are done in function nfsd4_decode_fattr() but removed
to this function now.

Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4proc.c | 95 +++++++++++++++++++++++++++++++++++++++++++++++++++---
 fs/nfsd/nfs4xdr.c  | 46 +++++++-------------------
 2 files changed, 103 insertions(+), 38 deletions(-)

diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index b2883e9c6381..9272e1f806fc 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -51,6 +51,78 @@
 
 #define NFSDDBG_FACILITY		NFSDDBG_PROC
 
+static u32 nfsd_attrmask[] = {
+	NFSD_WRITEABLE_ATTRS_WORD0,
+	NFSD_WRITEABLE_ATTRS_WORD1,
+	NFSD_WRITEABLE_ATTRS_WORD2
+};
+
+static u32 nfsd41_ex_attrmask[] = {
+	NFSD_SUPPATTR_EXCLCREAT_WORD0,
+	NFSD_SUPPATTR_EXCLCREAT_WORD1,
+	NFSD_SUPPATTR_EXCLCREAT_WORD2
+};
+
+static __be32
+check_attr_support(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
+		   u32 *bmval, u32 *writable)
+{
+	struct dentry *dentry = cstate->current_fh.fh_dentry;
+	struct svc_export *exp = cstate->current_fh.fh_export;
+
+	/*
+	 * Check about attributes are supported by the NFSv4 server or not.
+	 * According to spec, unsupported attributes return ERR_ATTRNOTSUPP.
+	 */
+	if ((bmval[0] & ~nfsd_suppattrs0(cstate->minorversion)) ||
+	    (bmval[1] & ~nfsd_suppattrs1(cstate->minorversion)) ||
+	    (bmval[2] & ~nfsd_suppattrs2(cstate->minorversion)))
+		return nfserr_attrnotsupp;
+
+	/*
+	 * Check FATTR4_WORD0_ACL & FATTR4_WORD0_FS_LOCATIONS can be supported
+	 * in current environment or not.
+	 */
+	if (bmval[0] & FATTR4_WORD0_ACL) {
+		if (!IS_POSIXACL(dentry->d_inode))
+			return nfserr_attrnotsupp;
+	}
+	if (bmval[0] & FATTR4_WORD0_FS_LOCATIONS) {
+		if (exp->ex_fslocs.locations == NULL)
+			return nfserr_attrnotsupp;
+	}
+
+	/*
+	 * According to spec, read-only attributes return ERR_INVAL.
+	 */
+	if (writable) {
+		if ((bmval[0] & ~writable[0]) || (bmval[1] & ~writable[1]) ||
+		    (bmval[2] & ~writable[2]))
+			return nfserr_inval;
+	}
+
+	return nfs_ok;
+}
+
+static __be32
+nfsd4_check_open_attributes(struct svc_rqst *rqstp,
+	struct nfsd4_compound_state *cstate, struct nfsd4_open *open)
+{
+	__be32 status = nfs_ok;
+
+	if (open->op_create == NFS4_OPEN_CREATE) {
+		if (open->op_createmode == NFS4_CREATE_UNCHECKED
+		    || open->op_createmode == NFS4_CREATE_GUARDED)
+			status = check_attr_support(rqstp, cstate,
+					open->op_bmval, nfsd_attrmask);
+		else if (open->op_createmode == NFS4_CREATE_EXCLUSIVE4_1)
+			status = check_attr_support(rqstp, cstate,
+					open->op_bmval, nfsd41_ex_attrmask);
+	}
+
+	return status;
+}
+
 static inline void
 fh_dup2(struct svc_fh *dst, struct svc_fh *src)
 {
@@ -225,6 +297,10 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (status)
 		goto out;
 
+	status = nfsd4_check_open_attributes(rqstp, cstate, open);
+	if (status)
+		goto out;
+
 	/* Openowner is now set, so sequence id will get bumped.  Now we need
 	 * these checks before we do any creates: */
 	status = nfserr_grace;
@@ -395,6 +471,11 @@ nfsd4_create(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (status)
 		return status;
 
+	status = check_attr_support(rqstp, cstate, create->cr_bmval,
+				    nfsd_attrmask);
+	if (status)
+		return status;
+
 	switch (create->cr_type) {
 	case NF4LNK:
 		/* ugh! we have to null-terminate the linktext, or
@@ -689,6 +770,12 @@ nfsd4_setattr(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (status)
 		return status;
 	status = nfs_ok;
+
+	status = check_attr_support(rqstp, cstate, setattr->sa_bmval,
+				    nfsd_attrmask);
+	if (status)
+		goto out;
+
 	if (setattr->sa_acl != NULL)
 		status = nfsd4_set_nfs4_acl(rqstp, &cstate->current_fh,
 					    setattr->sa_acl);
@@ -763,10 +850,10 @@ _nfsd4_verify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	if (status)
 		return status;
 
-	if ((verify->ve_bmval[0] & ~nfsd_suppattrs0(cstate->minorversion))
-	    || (verify->ve_bmval[1] & ~nfsd_suppattrs1(cstate->minorversion))
-	    || (verify->ve_bmval[2] & ~nfsd_suppattrs2(cstate->minorversion)))
-		return nfserr_attrnotsupp;
+	status = check_attr_support(rqstp, cstate, verify->ve_bmval, NULL);
+	if (status)
+		return status;
+
 	if ((verify->ve_bmval[0] & FATTR4_WORD0_RDATTR_ERROR)
 	    || (verify->ve_bmval[1] & NFSD_WRITEONLY_ATTRS_WORD1))
 		return nfserr_inval;
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index ab005fc637e1..254e5b21b915 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -244,20 +244,8 @@ nfsd4_decode_bitmap(struct nfsd4_compoundargs *argp, u32 *bmval)
 	DECODE_TAIL;
 }
 
-static u32 nfsd_attrmask[] = {
-	NFSD_WRITEABLE_ATTRS_WORD0,
-	NFSD_WRITEABLE_ATTRS_WORD1,
-	NFSD_WRITEABLE_ATTRS_WORD2
-};
-
-static u32 nfsd41_ex_attrmask[] = {
-	NFSD_SUPPATTR_EXCLCREAT_WORD0,
-	NFSD_SUPPATTR_EXCLCREAT_WORD1,
-	NFSD_SUPPATTR_EXCLCREAT_WORD2
-};
-
 static __be32
-nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, u32 *writable,
+nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval,
 		   struct iattr *iattr, struct nfs4_acl **acl)
 {
 	int expected_len, len = 0;
@@ -270,18 +258,6 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, u32 *writable,
 	if ((status = nfsd4_decode_bitmap(argp, bmval)))
 		return status;
 
-	/*
-	 * According to spec, unsupported attributes return ERR_ATTRNOTSUPP;
-	 * read-only attributes return ERR_INVAL.
-	 */
-	if ((bmval[0] & ~nfsd_suppattrs0(argp->minorversion)) ||
-	    (bmval[1] & ~nfsd_suppattrs1(argp->minorversion)) ||
-	    (bmval[2] & ~nfsd_suppattrs2(argp->minorversion)))
-		return nfserr_attrnotsupp;
-	if ((bmval[0] & ~writable[0]) || (bmval[1] & ~writable[1]) ||
-	    (bmval[2] & ~writable[2]))
-		return nfserr_inval;
-
 	READ_BUF(4);
 	READ32(expected_len);
 
@@ -414,8 +390,11 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, u32 *writable,
 			goto xdr_error;
 		}
 	}
-	BUG_ON(bmval[2]);	/* no such writeable attr supported yet */
-	if (len != expected_len)
+	if (bmval[0] & ~NFSD_WRITEABLE_ATTRS_WORD0
+	    || bmval[1] & ~NFSD_WRITEABLE_ATTRS_WORD1
+	    || bmval[2] & ~NFSD_WRITEABLE_ATTRS_WORD2)
+		READ_BUF(expected_len - len);
+	else if (len != expected_len)
 		goto xdr_error;
 
 	DECODE_TAIL;
@@ -508,8 +487,8 @@ nfsd4_decode_create(struct nfsd4_compoundargs *argp, struct nfsd4_create *create
 	if ((status = check_filename(create->cr_name, create->cr_namelen, nfserr_inval)))
 		return status;
 
-	status = nfsd4_decode_fattr(argp, create->cr_bmval, nfsd_attrmask,
-				    &create->cr_iattr, &create->cr_acl);
+	status = nfsd4_decode_fattr(argp, create->cr_bmval, &create->cr_iattr,
+				    &create->cr_acl);
 	if (status)
 		goto out;
 
@@ -672,7 +651,7 @@ nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open)
 		case NFS4_CREATE_UNCHECKED:
 		case NFS4_CREATE_GUARDED:
 			status = nfsd4_decode_fattr(argp, open->op_bmval,
-				nfsd_attrmask, &open->op_iattr, &open->op_acl);
+				&open->op_iattr, &open->op_acl);
 			if (status)
 				goto out;
 			break;
@@ -686,8 +665,7 @@ nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open)
 			READ_BUF(8);
 			COPYMEM(open->op_verf.data, 8);
 			status = nfsd4_decode_fattr(argp, open->op_bmval,
-				nfsd41_ex_attrmask, &open->op_iattr,
-				&open->op_acl);
+				&open->op_iattr, &open->op_acl);
 			if (status)
 				goto out;
 			break;
@@ -883,8 +861,8 @@ nfsd4_decode_setattr(struct nfsd4_compoundargs *argp, struct nfsd4_setattr *seta
 	status = nfsd4_decode_stateid(argp, &setattr->sa_stateid);
 	if (status)
 		return status;
-	return nfsd4_decode_fattr(argp, setattr->sa_bmval, nfsd_attrmask,
-				  &setattr->sa_iattr, &setattr->sa_acl);
+	return nfsd4_decode_fattr(argp, setattr->sa_bmval, &setattr->sa_iattr,
+				  &setattr->sa_acl);
 }
 
 static __be32
-- 
cgit v1.2.3-59-g8ed1b


From 0a93a47f042c459f0f46942c3a920e3c81878031 Mon Sep 17 00:00:00 2001
From: Yu Zhiguo <yuzg@cn.fujitsu.com>
Date: Tue, 19 May 2009 14:09:54 +0800
Subject: NFSv4: kill off complicated macro 'PROC'

J. Bruce Fields wrote:
...
> (This is extremely confusing code to track down: note that
> proc->pc_decode is set to nfs4svc_decode_compoundargs() by the PROC()
> macro at the end of fs/nfsd/nfs4proc.c.  Which means, for example, that
> grepping for nfs4svc_decode_compoundargs() gets you nowhere.  Patches to
> kill off that macro would be welcomed....)

the macro 'PROC' is complicated and obscure, it had better
be killed off in order to make the code more clear.

Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4proc.c | 34 +++++++++++++++++-----------------
 1 file changed, 17 insertions(+), 17 deletions(-)

diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index 9272e1f806fc..7c8801769a3c 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -1313,24 +1313,9 @@ static const char *nfsd4_op_name(unsigned opnum)
 	return "unknown_operation";
 }
 
-#define nfs4svc_decode_voidargs		NULL
-#define nfs4svc_release_void		NULL
 #define nfsd4_voidres			nfsd4_voidargs
-#define nfs4svc_release_compound	NULL
 struct nfsd4_voidargs { int dummy; };
 
-#define PROC(name, argt, rest, relt, cache, respsize)	\
- { (svc_procfunc) nfsd4_proc_##name,		\
-   (kxdrproc_t) nfs4svc_decode_##argt##args,	\
-   (kxdrproc_t) nfs4svc_encode_##rest##res,	\
-   (kxdrproc_t) nfs4svc_release_##relt,		\
-   sizeof(struct nfsd4_##argt##args),		\
-   sizeof(struct nfsd4_##rest##res),		\
-   0,						\
-   cache,					\
-   respsize,					\
- }
-
 /*
  * TODO: At the present time, the NFSv4 server does not do XID caching
  * of requests.  Implementing XID caching would not be a serious problem,
@@ -1342,8 +1327,23 @@ struct nfsd4_voidargs { int dummy; };
  * better XID's.
  */
 static struct svc_procedure		nfsd_procedures4[2] = {
-  PROC(null,	 void,		void,		void,	  RC_NOCACHE, 1),
-  PROC(compound, compound,	compound,	compound, RC_NOCACHE, NFSD_BUFSIZE/4)
+	[NFSPROC4_NULL] = {
+		.pc_func = (svc_procfunc) nfsd4_proc_null,
+		.pc_encode = (kxdrproc_t) nfs4svc_encode_voidres,
+		.pc_argsize = sizeof(struct nfsd4_voidargs),
+		.pc_ressize = sizeof(struct nfsd4_voidres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = 1,
+	},
+	[NFSPROC4_COMPOUND] = {
+		.pc_func = (svc_procfunc) nfsd4_proc_compound,
+		.pc_decode = (kxdrproc_t) nfs4svc_decode_compoundargs,
+		.pc_encode = (kxdrproc_t) nfs4svc_encode_compoundres,
+		.pc_argsize = sizeof(struct nfsd4_compoundargs),
+		.pc_ressize = sizeof(struct nfsd4_compoundres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = NFSD_BUFSIZE/4,
+	},
 };
 
 struct svc_version	nfsd_version4 = {
-- 
cgit v1.2.3-59-g8ed1b


From 24b5ce20cc75603ce7c03a42116c30a17bce509a Mon Sep 17 00:00:00 2001
From: Thomas Chou <thomas@wytron.com.tw>
Date: Tue, 21 Apr 2009 12:27:34 +0800
Subject: mtd: plat_nand: fix section error

With CONFIG_HOTPLUG=n, the following eror occurred during link:
local symbol 0: discarded in section `.devexit.text' from
drivers/built-in.o

It was caused by improper section reference. The __devexit_p()
should be added to the .remove function.

Signed-off-by: Thomas Chou <thomas@wytron.com.tw>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/plat_nand.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/nand/plat_nand.c b/drivers/mtd/nand/plat_nand.c
index 86e1d08eee00..28ffd4e8bb2f 100644
--- a/drivers/mtd/nand/plat_nand.c
+++ b/drivers/mtd/nand/plat_nand.c
@@ -128,7 +128,7 @@ static int __devexit plat_nand_remove(struct platform_device *pdev)
 
 static struct platform_driver plat_nand_driver = {
 	.probe		= plat_nand_probe,
-	.remove		= plat_nand_remove,
+	.remove		= __devexit_p(plat_nand_remove),
 	.driver		= {
 		.name	= "gen_nand",
 		.owner	= THIS_MODULE,
-- 
cgit v1.2.3-59-g8ed1b


From 24430abc88c67e3df2f06c96a6ccd73bda4c92f0 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Tue, 21 Apr 2009 11:02:01 +0300
Subject: Kill jffs2-user.h

This file does not define any kernel-userspace API, all
it does it defines few helpers for userspace. Instead,
userspace should have a private copy of this file.

The main (if not the only) user is the mtd-utils package, but
it already has a private copy of this file.

This patch also removes references to 'jffs2-user.h' from
'Kbuild' and MAINTAINERS' files.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 MAINTAINERS              |  1 -
 include/mtd/Kbuild       |  1 -
 include/mtd/jffs2-user.h | 34 ----------------------------------
 3 files changed, 36 deletions(-)
 delete mode 100644 include/mtd/jffs2-user.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 2b349ba4add4..3276e1c256b8 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3171,7 +3171,6 @@ W:	http://www.linux-mtd.infradead.org/doc/jffs2.html
 S:	Maintained
 F:	fs/jffs2/
 F:	include/linux/jffs2.h
-F:	include/mtd/jffs2-user.h
 
 JOURNALLING LAYER FOR BLOCK DEVICES (JBD)
 P:	Stephen Tweedie
diff --git a/include/mtd/Kbuild b/include/mtd/Kbuild
index 8eb018f96002..192f8fb7d546 100644
--- a/include/mtd/Kbuild
+++ b/include/mtd/Kbuild
@@ -1,5 +1,4 @@
 header-y += inftl-user.h
-header-y += jffs2-user.h
 header-y += mtd-abi.h
 header-y += mtd-user.h
 header-y += nftl-user.h
diff --git a/include/mtd/jffs2-user.h b/include/mtd/jffs2-user.h
deleted file mode 100644
index fa94b0eb67c1..000000000000
--- a/include/mtd/jffs2-user.h
+++ /dev/null
@@ -1,34 +0,0 @@
-/*
- * JFFS2 definitions for use in user space only
- */
-
-#ifndef __JFFS2_USER_H__
-#define __JFFS2_USER_H__
-
-/* This file is blessed for inclusion by userspace */
-#include <linux/jffs2.h>
-#include <linux/types.h>
-#include <endian.h>
-#include <byteswap.h>
-
-#undef cpu_to_je16
-#undef cpu_to_je32
-#undef cpu_to_jemode
-#undef je16_to_cpu
-#undef je32_to_cpu
-#undef jemode_to_cpu
-
-extern int target_endian;
-
-#define t16(x) ({ __u16 __b = (x); (target_endian==__BYTE_ORDER)?__b:bswap_16(__b); })
-#define t32(x) ({ __u32 __b = (x); (target_endian==__BYTE_ORDER)?__b:bswap_32(__b); })
-
-#define cpu_to_je16(x) ((jint16_t){t16(x)})
-#define cpu_to_je32(x) ((jint32_t){t32(x)})
-#define cpu_to_jemode(x) ((jmode_t){t32(x)})
-
-#define je16_to_cpu(x) (t16((x).v16))
-#define je32_to_cpu(x) (t32((x).v32))
-#define jemode_to_cpu(x) (t32((x).m))
-
-#endif /* __JFFS2_USER_H__ */
-- 
cgit v1.2.3-59-g8ed1b


From 81d19b04a865f9fcc0ca01b20be806169ff9efb3 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Tue, 21 Apr 2009 19:51:20 -0700
Subject: mtd: nand: don't walk past end of oobfree[]

Resolve issue noted by Sneha:  when computing oobavail from
the list of free areas in the OOB, don't assume there will
always be an unused slot at the end.  With ECC_HW_SYNDROME
and 4KiB page chips, it's fairly likely there *won't* be one.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: "Narnakaje, Snehaprabha" <nsnehaprabha@ti.com>"
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/nand_base.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c
index 3d7ed432fa41..8c21b89d2d0c 100644
--- a/drivers/mtd/nand/nand_base.c
+++ b/drivers/mtd/nand/nand_base.c
@@ -2756,7 +2756,8 @@ int nand_scan_tail(struct mtd_info *mtd)
 	 * the out of band area
 	 */
 	chip->ecc.layout->oobavail = 0;
-	for (i = 0; chip->ecc.layout->oobfree[i].length; i++)
+	for (i = 0; chip->ecc.layout->oobfree[i].length
+			&& i < ARRAY_SIZE(chip->ecc.layout->oobfree); i++)
 		chip->ecc.layout->oobavail +=
 			chip->ecc.layout->oobfree[i].length;
 	mtd->oobavail = chip->ecc.layout->oobavail;
-- 
cgit v1.2.3-59-g8ed1b


From 533a0149148ccaa0199a1ee6492cd860e3c8b456 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Tue, 21 Apr 2009 19:51:31 -0700
Subject: mtd: nand: minor davinci_nand cleanup

Make the DaVinci NAND driver require platform_data with
board-specific configuration.  We can't actually do any
kind of sane job of configuring it otherwise.

Also fix the comment about picking the "best" ECC mode.

We can't do those any more; that relied on knowing what kind
of CPU we're using (they don't all support 4-bit ECC), and
current policy is that drivers not have cpu_is_*() checks.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/davinci_nand.c | 27 +++++++++++----------------
 1 file changed, 11 insertions(+), 16 deletions(-)

diff --git a/drivers/mtd/nand/davinci_nand.c b/drivers/mtd/nand/davinci_nand.c
index 0119220de7d0..68b747584bc8 100644
--- a/drivers/mtd/nand/davinci_nand.c
+++ b/drivers/mtd/nand/davinci_nand.c
@@ -306,6 +306,10 @@ static int __init nand_davinci_probe(struct platform_device *pdev)
 	uint32_t			val;
 	nand_ecc_modes_t		ecc_mode;
 
+	/* insist on board-specific configuration */
+	if (!pdata)
+		return -ENODEV;
+
 	/* which external chipselect will we be managing? */
 	if (pdev->id < 0 || pdev->id > 3)
 		return -ENODEV;
@@ -351,7 +355,7 @@ static int __init nand_davinci_probe(struct platform_device *pdev)
 	info->chip.select_chip	= nand_davinci_select_chip;
 
 	/* options such as NAND_USE_FLASH_BBT or 16-bit widths */
-	info->chip.options	= pdata ? pdata->options : 0;
+	info->chip.options	= pdata->options;
 
 	info->ioaddr		= (uint32_t __force) vaddr;
 
@@ -360,14 +364,8 @@ static int __init nand_davinci_probe(struct platform_device *pdev)
 	info->mask_chipsel	= pdata->mask_chipsel;
 
 	/* use nandboot-capable ALE/CLE masks by default */
-	if (pdata && pdata->mask_ale)
-		info->mask_ale	= pdata->mask_cle;
-	else
-		info->mask_ale	= MASK_ALE;
-	if (pdata && pdata->mask_cle)
-		info->mask_cle	= pdata->mask_cle;
-	else
-		info->mask_cle	= MASK_CLE;
+	info->mask_ale		= pdata->mask_cle ? : MASK_ALE;
+	info->mask_cle		= pdata->mask_cle ? : MASK_CLE;
 
 	/* Set address of hardware control function */
 	info->chip.cmd_ctrl	= nand_davinci_hwcontrol;
@@ -377,11 +375,8 @@ static int __init nand_davinci_probe(struct platform_device *pdev)
 	info->chip.read_buf     = nand_davinci_read_buf;
 	info->chip.write_buf    = nand_davinci_write_buf;
 
-	/* use board-specific ECC config; else, the best available */
-	if (pdata)
-		ecc_mode = pdata->ecc_mode;
-	else
-		ecc_mode = NAND_ECC_HW;
+	/* Use board-specific ECC config */
+	ecc_mode		= pdata->ecc_mode;
 
 	switch (ecc_mode) {
 	case NAND_ECC_NONE:
@@ -469,7 +464,7 @@ static int __init nand_davinci_probe(struct platform_device *pdev)
 			info->mtd.name = master_name;
 		}
 
-		if (mtd_parts_nb <= 0 && pdata) {
+		if (mtd_parts_nb <= 0) {
 			mtd_parts = pdata->parts;
 			mtd_parts_nb = pdata->nr_parts;
 		}
@@ -482,7 +477,7 @@ static int __init nand_davinci_probe(struct platform_device *pdev)
 				info->partitioned = true;
 		}
 
-	} else if (pdata && pdata->nr_parts) {
+	} else if (pdata->nr_parts) {
 		dev_warn(&pdev->dev, "ignoring %d default partitions on %s\n",
 				pdata->nr_parts, info->mtd.name);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 6a4123e581b3112ff4ea7439ab9ae5cb271a9dbd Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Tue, 21 Apr 2009 19:58:13 -0700
Subject: mtd: nand: davinci_nand, 4-bit ECC for smallpage

Minimal support for the 4-bit ECC engine found on DM355, DM365,
DA830/OMAP-L137, and similar recent DaVinci-family chips.

This is limited to small-page flash for now; there are some page
layout issues for large page chips.  Note that most boards using
this engine (like the DM355 EVM) include 2GiB large page chips.

Sanity tested on DM355 EVM after swapping the socketed NAND for
a small-page one.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 arch/arm/mach-davinci/include/mach/nand.h |   8 +-
 drivers/mtd/nand/davinci_nand.c           | 304 ++++++++++++++++++++++++++++--
 2 files changed, 297 insertions(+), 15 deletions(-)

diff --git a/arch/arm/mach-davinci/include/mach/nand.h b/arch/arm/mach-davinci/include/mach/nand.h
index aa482841270b..b520c4b5678a 100644
--- a/arch/arm/mach-davinci/include/mach/nand.h
+++ b/arch/arm/mach-davinci/include/mach/nand.h
@@ -68,10 +68,14 @@ struct davinci_nand_pdata {		/* platform_data */
 
 	/* none  == NAND_ECC_NONE (strongly *not* advised!!)
 	 * soft  == NAND_ECC_SOFT
-	 * 1-bit == NAND_ECC_HW
-	 * 4-bit == NAND_ECC_HW_SYNDROME (not on all chips)
+	 * else  == NAND_ECC_HW, according to ecc_bits
+	 *
+	 * All DaVinci-family chips support 1-bit hardware ECC.
+	 * Newer ones also support 4-bit ECC, but are awkward
+	 * using it with large page chips.
 	 */
 	nand_ecc_modes_t	ecc_mode;
+	u8			ecc_bits;
 
 	/* e.g. NAND_BUSWIDTH_16 or NAND_USE_FLASH_BBT */
 	unsigned		options;
diff --git a/drivers/mtd/nand/davinci_nand.c b/drivers/mtd/nand/davinci_nand.c
index 68b747584bc8..cb5a05e01531 100644
--- a/drivers/mtd/nand/davinci_nand.c
+++ b/drivers/mtd/nand/davinci_nand.c
@@ -44,7 +44,7 @@
  * and some flavors of secondary chipselect (e.g. based on A12) as used
  * with multichip packages.
  *
- * The 1-bit ECC hardware is supported, but not yet the newer 4-bit ECC
+ * The 1-bit ECC hardware is supported, as well as the newer 4-bit ECC
  * available on chips like the DM355 and OMAP-L137 and needed with the
  * more error-prone MLC NAND chips.
  *
@@ -54,11 +54,14 @@
 struct davinci_nand_info {
 	struct mtd_info		mtd;
 	struct nand_chip	chip;
+	struct nand_ecclayout	ecclayout;
 
 	struct device		*dev;
 	struct clk		*clk;
 	bool			partitioned;
 
+	bool			is_readmode;
+
 	void __iomem		*base;
 	void __iomem		*vaddr;
 
@@ -73,6 +76,7 @@ struct davinci_nand_info {
 };
 
 static DEFINE_SPINLOCK(davinci_nand_lock);
+static bool ecc4_busy;
 
 #define to_davinci_nand(m) container_of(m, struct davinci_nand_info, mtd)
 
@@ -217,6 +221,192 @@ static int nand_davinci_correct_1bit(struct mtd_info *mtd, u_char *dat,
 
 /*----------------------------------------------------------------------*/
 
+/*
+ * 4-bit hardware ECC ... context maintained over entire AEMIF
+ *
+ * This is a syndrome engine, but we avoid NAND_ECC_HW_SYNDROME
+ * since that forces use of a problematic "infix OOB" layout.
+ * Among other things, it trashes manufacturer bad block markers.
+ * Also, and specific to this hardware, it ECC-protects the "prepad"
+ * in the OOB ... while having ECC protection for parts of OOB would
+ * seem useful, the current MTD stack sometimes wants to update the
+ * OOB without recomputing ECC.
+ */
+
+static void nand_davinci_hwctl_4bit(struct mtd_info *mtd, int mode)
+{
+	struct davinci_nand_info *info = to_davinci_nand(mtd);
+	unsigned long flags;
+	u32 val;
+
+	spin_lock_irqsave(&davinci_nand_lock, flags);
+
+	/* Start 4-bit ECC calculation for read/write */
+	val = davinci_nand_readl(info, NANDFCR_OFFSET);
+	val &= ~(0x03 << 4);
+	val |= (info->core_chipsel << 4) | BIT(12);
+	davinci_nand_writel(info, NANDFCR_OFFSET, val);
+
+	info->is_readmode = (mode == NAND_ECC_READ);
+
+	spin_unlock_irqrestore(&davinci_nand_lock, flags);
+}
+
+/* Read raw ECC code after writing to NAND. */
+static void
+nand_davinci_readecc_4bit(struct davinci_nand_info *info, u32 code[4])
+{
+	const u32 mask = 0x03ff03ff;
+
+	code[0] = davinci_nand_readl(info, NAND_4BIT_ECC1_OFFSET) & mask;
+	code[1] = davinci_nand_readl(info, NAND_4BIT_ECC2_OFFSET) & mask;
+	code[2] = davinci_nand_readl(info, NAND_4BIT_ECC3_OFFSET) & mask;
+	code[3] = davinci_nand_readl(info, NAND_4BIT_ECC4_OFFSET) & mask;
+}
+
+/* Terminate read ECC; or return ECC (as bytes) of data written to NAND. */
+static int nand_davinci_calculate_4bit(struct mtd_info *mtd,
+		const u_char *dat, u_char *ecc_code)
+{
+	struct davinci_nand_info *info = to_davinci_nand(mtd);
+	u32 raw_ecc[4], *p;
+	unsigned i;
+
+	/* After a read, terminate ECC calculation by a dummy read
+	 * of some 4-bit ECC register.  ECC covers everything that
+	 * was read; correct() just uses the hardware state, so
+	 * ecc_code is not needed.
+	 */
+	if (info->is_readmode) {
+		davinci_nand_readl(info, NAND_4BIT_ECC1_OFFSET);
+		return 0;
+	}
+
+	/* Pack eight raw 10-bit ecc values into ten bytes, making
+	 * two passes which each convert four values (in upper and
+	 * lower halves of two 32-bit words) into five bytes.  The
+	 * ROM boot loader uses this same packing scheme.
+	 */
+	nand_davinci_readecc_4bit(info, raw_ecc);
+	for (i = 0, p = raw_ecc; i < 2; i++, p += 2) {
+		*ecc_code++ =   p[0]        & 0xff;
+		*ecc_code++ = ((p[0] >>  8) & 0x03) | ((p[0] >> 14) & 0xfc);
+		*ecc_code++ = ((p[0] >> 22) & 0x0f) | ((p[1] <<  4) & 0xf0);
+		*ecc_code++ = ((p[1] >>  4) & 0x3f) | ((p[1] >> 10) & 0xc0);
+		*ecc_code++ =  (p[1] >> 18) & 0xff;
+	}
+
+	return 0;
+}
+
+/* Correct up to 4 bits in data we just read, using state left in the
+ * hardware plus the ecc_code computed when it was first written.
+ */
+static int nand_davinci_correct_4bit(struct mtd_info *mtd,
+		u_char *data, u_char *ecc_code, u_char *null)
+{
+	int i;
+	struct davinci_nand_info *info = to_davinci_nand(mtd);
+	unsigned short ecc10[8];
+	unsigned short *ecc16;
+	u32 syndrome[4];
+	unsigned num_errors, corrected;
+
+	/* All bytes 0xff?  It's an erased page; ignore its ECC. */
+	for (i = 0; i < 10; i++) {
+		if (ecc_code[i] != 0xff)
+			goto compare;
+	}
+	return 0;
+
+compare:
+	/* Unpack ten bytes into eight 10 bit values.  We know we're
+	 * little-endian, and use type punning for less shifting/masking.
+	 */
+	if (WARN_ON(0x01 & (unsigned) ecc_code))
+		return -EINVAL;
+	ecc16 = (unsigned short *)ecc_code;
+
+	ecc10[0] =  (ecc16[0] >>  0) & 0x3ff;
+	ecc10[1] = ((ecc16[0] >> 10) & 0x3f) | ((ecc16[1] << 6) & 0x3c0);
+	ecc10[2] =  (ecc16[1] >>  4) & 0x3ff;
+	ecc10[3] = ((ecc16[1] >> 14) & 0x3)  | ((ecc16[2] << 2) & 0x3fc);
+	ecc10[4] =  (ecc16[2] >>  8)         | ((ecc16[3] << 8) & 0x300);
+	ecc10[5] =  (ecc16[3] >>  2) & 0x3ff;
+	ecc10[6] = ((ecc16[3] >> 12) & 0xf)  | ((ecc16[4] << 4) & 0x3f0);
+	ecc10[7] =  (ecc16[4] >>  6) & 0x3ff;
+
+	/* Tell ECC controller about the expected ECC codes. */
+	for (i = 7; i >= 0; i--)
+		davinci_nand_writel(info, NAND_4BIT_ECC_LOAD_OFFSET, ecc10[i]);
+
+	/* Allow time for syndrome calculation ... then read it.
+	 * A syndrome of all zeroes 0 means no detected errors.
+	 */
+	davinci_nand_readl(info, NANDFSR_OFFSET);
+	nand_davinci_readecc_4bit(info, syndrome);
+	if (!(syndrome[0] | syndrome[1] | syndrome[2] | syndrome[3]))
+		return 0;
+
+	/* Start address calculation, and wait for it to complete.
+	 * We _could_ start reading more data while this is working,
+	 * to speed up the overall page read.
+	 */
+	davinci_nand_writel(info, NANDFCR_OFFSET,
+			davinci_nand_readl(info, NANDFCR_OFFSET) | BIT(13));
+	for (;;) {
+		u32	fsr = davinci_nand_readl(info, NANDFSR_OFFSET);
+
+		switch ((fsr >> 8) & 0x0f) {
+		case 0:		/* no error, should not happen */
+			return 0;
+		case 1:		/* five or more errors detected */
+			return -EIO;
+		case 2:		/* error addresses computed */
+		case 3:
+			num_errors = 1 + ((fsr >> 16) & 0x03);
+			goto correct;
+		default:	/* still working on it */
+			cpu_relax();
+			continue;
+		}
+	}
+
+correct:
+	/* correct each error */
+	for (i = 0, corrected = 0; i < num_errors; i++) {
+		int error_address, error_value;
+
+		if (i > 1) {
+			error_address = davinci_nand_readl(info,
+						NAND_ERR_ADD2_OFFSET);
+			error_value = davinci_nand_readl(info,
+						NAND_ERR_ERRVAL2_OFFSET);
+		} else {
+			error_address = davinci_nand_readl(info,
+						NAND_ERR_ADD1_OFFSET);
+			error_value = davinci_nand_readl(info,
+						NAND_ERR_ERRVAL1_OFFSET);
+		}
+
+		if (i & 1) {
+			error_address >>= 16;
+			error_value >>= 16;
+		}
+		error_address &= 0x3ff;
+		error_address = (512 + 7) - error_address;
+
+		if (error_address < 512) {
+			data[error_address] ^= error_value;
+			corrected++;
+		}
+	}
+
+	return corrected;
+}
+
+/*----------------------------------------------------------------------*/
+
 /*
  * NOTE:  NAND boot requires ALE == EM_A[1], CLE == EM_A[2], so that's
  * how these chips are normally wired.  This translates to both 8 and 16
@@ -294,6 +484,23 @@ static void __init nand_dm6446evm_flash_init(struct davinci_nand_info *info)
 
 /*----------------------------------------------------------------------*/
 
+/* An ECC layout for using 4-bit ECC with small-page flash, storing
+ * ten ECC bytes plus the manufacturer's bad block marker byte, and
+ * and not overlapping the default BBT markers.
+ */
+static struct nand_ecclayout hwecc4_small __initconst = {
+	.eccbytes = 10,
+	.eccpos = { 0, 1, 2, 3, 4,
+		/* offset 5 holds the badblock marker */
+		6, 7,
+		13, 14, 15, },
+	.oobfree = {
+		{.offset = 8, .length = 5, },
+		{.offset = 16, },
+	},
+};
+
+
 static int __init nand_davinci_probe(struct platform_device *pdev)
 {
 	struct davinci_nand_pdata	*pdata = pdev->dev.platform_data;
@@ -378,24 +585,41 @@ static int __init nand_davinci_probe(struct platform_device *pdev)
 	/* Use board-specific ECC config */
 	ecc_mode		= pdata->ecc_mode;
 
+	ret = -EINVAL;
 	switch (ecc_mode) {
 	case NAND_ECC_NONE:
 	case NAND_ECC_SOFT:
+		pdata->ecc_bits = 0;
 		break;
 	case NAND_ECC_HW:
-		info->chip.ecc.calculate = nand_davinci_calculate_1bit;
-		info->chip.ecc.correct = nand_davinci_correct_1bit;
-		info->chip.ecc.hwctl = nand_davinci_hwctl_1bit;
+		if (pdata->ecc_bits == 4) {
+			/* No sanity checks:  CPUs must support this,
+			 * and the chips may not use NAND_BUSWIDTH_16.
+			 */
+
+			/* No sharing 4-bit hardware between chipselects yet */
+			spin_lock_irq(&davinci_nand_lock);
+			if (ecc4_busy)
+				ret = -EBUSY;
+			else
+				ecc4_busy = true;
+			spin_unlock_irq(&davinci_nand_lock);
+
+			if (ret == -EBUSY)
+				goto err_ecc;
+
+			info->chip.ecc.calculate = nand_davinci_calculate_4bit;
+			info->chip.ecc.correct = nand_davinci_correct_4bit;
+			info->chip.ecc.hwctl = nand_davinci_hwctl_4bit;
+			info->chip.ecc.bytes = 10;
+		} else {
+			info->chip.ecc.calculate = nand_davinci_calculate_1bit;
+			info->chip.ecc.correct = nand_davinci_correct_1bit;
+			info->chip.ecc.hwctl = nand_davinci_hwctl_1bit;
+			info->chip.ecc.bytes = 3;
+		}
 		info->chip.ecc.size = 512;
-		info->chip.ecc.bytes = 3;
 		break;
-	case NAND_ECC_HW_SYNDROME:
-		/* FIXME implement */
-		info->chip.ecc.size = 512;
-		info->chip.ecc.bytes = 10;
-
-		dev_warn(&pdev->dev, "4-bit ECC nyet supported\n");
-		/* FALL THROUGH */
 	default:
 		ret = -EINVAL;
 		goto err_ecc;
@@ -435,12 +659,56 @@ static int __init nand_davinci_probe(struct platform_device *pdev)
 	spin_unlock_irq(&davinci_nand_lock);
 
 	/* Scan to find existence of the device(s) */
-	ret = nand_scan(&info->mtd, pdata->mask_chipsel ? 2 : 1);
+	ret = nand_scan_ident(&info->mtd, pdata->mask_chipsel ? 2 : 1);
 	if (ret < 0) {
 		dev_dbg(&pdev->dev, "no NAND chip(s) found\n");
 		goto err_scan;
 	}
 
+	/* Update ECC layout if needed ... for 1-bit HW ECC, the default
+	 * is OK, but it allocates 6 bytes when only 3 are needed (for
+	 * each 512 bytes).  For the 4-bit HW ECC, that default is not
+	 * usable:  10 bytes are needed, not 6.
+	 */
+	if (pdata->ecc_bits == 4) {
+		int	chunks = info->mtd.writesize / 512;
+
+		if (!chunks || info->mtd.oobsize < 16) {
+			dev_dbg(&pdev->dev, "too small\n");
+			ret = -EINVAL;
+			goto err_scan;
+		}
+
+		/* For small page chips, preserve the manufacturer's
+		 * badblock marking data ... and make sure a flash BBT
+		 * table marker fits in the free bytes.
+		 */
+		if (chunks == 1) {
+			info->ecclayout = hwecc4_small;
+			info->ecclayout.oobfree[1].length =
+				info->mtd.oobsize - 16;
+			goto syndrome_done;
+		}
+
+		/* For large page chips we'll be wanting to use a
+		 * not-yet-implemented mode that reads OOB data
+		 * before reading the body of the page, to avoid
+		 * the "infix OOB" model of NAND_ECC_HW_SYNDROME
+		 * (and preserve manufacturer badblock markings).
+		 */
+		dev_warn(&pdev->dev, "no 4-bit ECC support yet "
+				"for large page NAND\n");
+		ret = -EIO;
+		goto err_scan;
+
+syndrome_done:
+		info->chip.ecc.layout = &info->ecclayout;
+	}
+
+	ret = nand_scan_tail(&info->mtd);
+	if (ret < 0)
+		goto err_scan;
+
 	if (mtd_has_partitions()) {
 		struct mtd_partition	*mtd_parts = NULL;
 		int			mtd_parts_nb = 0;
@@ -503,6 +771,11 @@ err_scan:
 err_clk_enable:
 	clk_put(info->clk);
 
+	spin_lock_irq(&davinci_nand_lock);
+	if (ecc_mode == NAND_ECC_HW_SYNDROME)
+		ecc4_busy = false;
+	spin_unlock_irq(&davinci_nand_lock);
+
 err_ecc:
 err_clk:
 err_ioremap:
@@ -526,6 +799,11 @@ static int __exit nand_davinci_remove(struct platform_device *pdev)
 	else
 		status = del_mtd_device(&info->mtd);
 
+	spin_lock_irq(&davinci_nand_lock);
+	if (info->chip.ecc.mode == NAND_ECC_HW_SYNDROME)
+		ecc4_busy = false;
+	spin_unlock_irq(&davinci_nand_lock);
+
 	iounmap(info->base);
 	iounmap(info->vaddr);
 
-- 
cgit v1.2.3-59-g8ed1b


From 260dc003e9fd41a6cac64308e63dea37ea5de13f Mon Sep 17 00:00:00 2001
From: Vimal Singh <vimalsingh@ti.com>
Date: Mon, 23 Feb 2009 13:46:08 +0530
Subject: mtd: nand: fix 512 byte software ecc support

Type of 'byte_addr' needes to be 'unsigned int' for 512 byte
ECC support.

Signed-off-by: Vimal Singh <vimalsingh@ti.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/nand_ecc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/nand_ecc.c b/drivers/mtd/nand/nand_ecc.c
index 868147acce2c..c0cb87d6d16e 100644
--- a/drivers/mtd/nand/nand_ecc.c
+++ b/drivers/mtd/nand/nand_ecc.c
@@ -428,8 +428,8 @@ EXPORT_SYMBOL(nand_calculate_ecc);
 int nand_correct_data(struct mtd_info *mtd, unsigned char *buf,
 		      unsigned char *read_ecc, unsigned char *calc_ecc)
 {
-	unsigned char b0, b1, b2;
-	unsigned char byte_addr, bit_addr;
+	unsigned char b0, b1, b2, bit_addr;
+	unsigned int byte_addr;
 	/* 256 or 512 bytes/ecc  */
 	const uint32_t eccsize_mult =
 			(((struct nand_chip *)mtd->priv)->ecc.size) >> 8;
-- 
cgit v1.2.3-59-g8ed1b


From b258fd8d0470c65fef5231887d7e97cb246da0d0 Mon Sep 17 00:00:00 2001
From: Magnus Lilja <lilja.magnus@gmail.com>
Date: Fri, 8 May 2009 21:57:47 +0200
Subject: mtd: mxc_nand: add correct dev_id parameter to free_irq() calls

Make sure to pass the same dev_id data to free_irq() that was
used when calling request_irq(), otherwise we get a warning about
freeing an already free IRQ.

Signed-off-by: Magnus Lilja <lilja.magnus@gmail.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/mxc_nand.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/mxc_nand.c b/drivers/mtd/nand/mxc_nand.c
index 65040de54bb7..d5aea6951d0b 100644
--- a/drivers/mtd/nand/mxc_nand.c
+++ b/drivers/mtd/nand/mxc_nand.c
@@ -981,7 +981,7 @@ static int __init mxcnd_probe(struct platform_device *pdev)
 	return 0;
 
 escan:
-	free_irq(host->irq, NULL);
+	free_irq(host->irq, host);
 eirq:
 	iounmap(host->regs);
 eres:
@@ -1001,7 +1001,7 @@ static int __devexit mxcnd_remove(struct platform_device *pdev)
 	platform_set_drvdata(pdev, NULL);
 
 	nand_release(&host->mtd);
-	free_irq(host->irq, NULL);
+	free_irq(host->irq, host);
 	iounmap(host->regs);
 	kfree(host);
 
-- 
cgit v1.2.3-59-g8ed1b


From 1e42d142e65ebdef38fb399b421d04e092ad1c6e Mon Sep 17 00:00:00 2001
From: Matthieu CASTET <matthieu.castet@parrot.com>
Date: Tue, 28 Apr 2009 18:15:31 +0200
Subject: mtd: m25p80 nand: add m45pe10 ids

this patch add m45pe10 [1] chip support to the m25p80 driver.

[1] http://www.numonyx.com/Documents/Datasheets/M45PE10.pdf

Signed-off-by: Matthieu CASTET <matthieu.castet@parrot.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/devices/m25p80.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/mtd/devices/m25p80.c b/drivers/mtd/devices/m25p80.c
index cc6369ea67dd..982572c1056c 100644
--- a/drivers/mtd/devices/m25p80.c
+++ b/drivers/mtd/devices/m25p80.c
@@ -528,6 +528,7 @@ static struct flash_info __devinitdata m25p_data [] = {
 	{ "m25p64",  0x202017,  0, 64 * 1024, 128, },
 	{ "m25p128", 0x202018, 0, 256 * 1024, 64, },
 
+	{ "m45pe10", 0x204011,  0, 64 * 1024, 2, },
 	{ "m45pe80", 0x204014,  0, 64 * 1024, 16, },
 	{ "m45pe16", 0x204015,  0, 64 * 1024, 32, },
 
-- 
cgit v1.2.3-59-g8ed1b


From ee8f37688966ab1438d0cf42e0cb7c6595d9592c Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Tue, 5 May 2009 11:04:19 +0300
Subject: mtd: OneNAND: add support for OneNAND manufactured by Numonyx

In addition to adding the Numonyx manufacturer code, this patch
also ensures 'sync. write' is disabled when reading identification
data - something that the Numonyx chip objects to, but the
Samsung chip seems to ignore.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/onenand/onenand_base.c | 3 ++-
 include/linux/mtd/onenand.h        | 1 +
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
index 30d6999e5f9f..2346857a275d 100644
--- a/drivers/mtd/onenand/onenand_base.c
+++ b/drivers/mtd/onenand/onenand_base.c
@@ -2576,6 +2576,7 @@ static void onenand_print_device_info(int device, int version)
 
 static const struct onenand_manufacturers onenand_manuf_ids[] = {
         {ONENAND_MFR_SAMSUNG, "Samsung"},
+	{ONENAND_MFR_NUMONYX, "Numonyx"},
 };
 
 /**
@@ -2621,7 +2622,7 @@ static int onenand_probe(struct mtd_info *mtd)
 	/* Save system configuration 1 */
 	syscfg = this->read_word(this->base + ONENAND_REG_SYS_CFG1);
 	/* Clear Sync. Burst Read mode to read BootRAM */
-	this->write_word((syscfg & ~ONENAND_SYS_CFG1_SYNC_READ), this->base + ONENAND_REG_SYS_CFG1);
+	this->write_word((syscfg & ~ONENAND_SYS_CFG1_SYNC_READ & ~ONENAND_SYS_CFG1_SYNC_WRITE), this->base + ONENAND_REG_SYS_CFG1);
 
 	/* Send the command for reading device ID from BootRAM */
 	this->write_word(ONENAND_CMD_READID, this->base + ONENAND_BOOTRAM);
diff --git a/include/linux/mtd/onenand.h b/include/linux/mtd/onenand.h
index 9aa2a9149b58..0fa3ac4ad576 100644
--- a/include/linux/mtd/onenand.h
+++ b/include/linux/mtd/onenand.h
@@ -176,6 +176,7 @@ struct onenand_chip {
  * OneNAND Flash Manufacturer ID Codes
  */
 #define ONENAND_MFR_SAMSUNG	0xec
+#define ONENAND_MFR_NUMONYX	0x20
 
 /**
  * struct onenand_manufacturers - NAND Flash Manufacturer ID Structure
-- 
cgit v1.2.3-59-g8ed1b


From f19e8999a5631e3af0e1ca5127af80a25aba1fd7 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Tue, 12 May 2009 16:23:32 -0700
Subject: mtd: davinci_nand: cmdlinepart uses MTD IDs

Remove some legacy code from the davinci_nand driver, which made
cmdlinepart ignore the the MTD ID passed to it.  Boards can have
multiple NAND chips, and some do (like the DM357 EVM), so this
dated hack is undesirable.

Correct labels are like "davinci_nand.0" (for chipselect 0).

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/davinci_nand.c | 11 -----------
 1 file changed, 11 deletions(-)

diff --git a/drivers/mtd/nand/davinci_nand.c b/drivers/mtd/nand/davinci_nand.c
index cb5a05e01531..ba6940d1d3d1 100644
--- a/drivers/mtd/nand/davinci_nand.c
+++ b/drivers/mtd/nand/davinci_nand.c
@@ -717,19 +717,8 @@ syndrome_done:
 			static const char *probes[] __initconst =
 				{ "cmdlinepart", NULL };
 
-			const char		*master_name;
-
-			/* Set info->mtd.name = 0 temporarily */
-			master_name		= info->mtd.name;
-			info->mtd.name		= (char *)0;
-
-			/* info->mtd.name == 0, means: don't bother checking
-			   <mtd-id> */
 			mtd_parts_nb = parse_mtd_partitions(&info->mtd, probes,
 							    &mtd_parts, 0);
-
-			/* Restore info->mtd.name */
-			info->mtd.name = master_name;
 		}
 
 		if (mtd_parts_nb <= 0) {
-- 
cgit v1.2.3-59-g8ed1b


From a0645ce9ba2e40fb2e2d74e47c90063015ee4527 Mon Sep 17 00:00:00 2001
From: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Date: Wed, 13 May 2009 00:37:18 +0200
Subject: mtd: add SST39SF040 chip to jedec_probe
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add SST39SF040 chip (like SST39SF020A but bigger - 4Mbit).

Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/chips/jedec_probe.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/mtd/chips/jedec_probe.c b/drivers/mtd/chips/jedec_probe.c
index e824b9b9b056..ccc4cfc7e4b5 100644
--- a/drivers/mtd/chips/jedec_probe.c
+++ b/drivers/mtd/chips/jedec_probe.c
@@ -166,6 +166,7 @@
 #define SST39LF040	0x00D7
 #define SST39SF010A	0x00B5
 #define SST39SF020A	0x00B6
+#define SST39SF040	0x00B7
 #define SST49LF004B	0x0060
 #define SST49LF040B	0x0050
 #define SST49LF008A	0x005a
@@ -1391,6 +1392,18 @@ static const struct amd_flash_info jedec_table[] = {
 		.regions	= {
 			ERASEINFO(0x01000,64),
 		}
+	}, {
+		.mfr_id		= MANUFACTURER_SST,
+		.dev_id		= SST39SF040,
+		.name		= "SST 39SF040",
+		.devtypes	= CFI_DEVICETYPE_X8,
+		.uaddr		= MTD_UADDR_0x5555_0x2AAA,
+		.dev_size	= SIZE_512KiB,
+		.cmd_set	= P_ID_AMD_STD,
+		.nr_regions	= 1,
+		.regions	= {
+			ERASEINFO(0x01000,128),
+		}
 	}, {
 		.mfr_id		= MANUFACTURER_SST,
 		.dev_id		= SST49LF040B,
-- 
cgit v1.2.3-59-g8ed1b


From d6fed9e9fc5eefae5be0ecf222bac7e7496e8e74 Mon Sep 17 00:00:00 2001
From: Alexander Clouter <alex@digriz.org.uk>
Date: Mon, 11 May 2009 19:28:01 +0100
Subject: mtd: extend plat_nand for (read|write)_buf

This patch adds (write|read)_buf callbacks to plat_nand.

The NAND on the TS-7800 provisioned by the FPGA allows readw() and
readl() to be used which gives a 2.5x speed up.  To be able to use this
from the plat_nand driver a hook for read_buf (and also write_buf whilst
we are in there) need to be made available.  This patch adds the hook.

Signed-off-by: Alexander Clouter <alex@digriz.org.uk>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/plat_nand.c | 2 ++
 include/linux/mtd/nand.h     | 6 ++++++
 2 files changed, 8 insertions(+)

diff --git a/drivers/mtd/nand/plat_nand.c b/drivers/mtd/nand/plat_nand.c
index 28ffd4e8bb2f..47a2105b9671 100644
--- a/drivers/mtd/nand/plat_nand.c
+++ b/drivers/mtd/nand/plat_nand.c
@@ -61,6 +61,8 @@ static int __devinit plat_nand_probe(struct platform_device *pdev)
 	data->chip.cmd_ctrl = pdata->ctrl.cmd_ctrl;
 	data->chip.dev_ready = pdata->ctrl.dev_ready;
 	data->chip.select_chip = pdata->ctrl.select_chip;
+	data->chip.write_buf = pdata->ctrl.write_buf;
+	data->chip.read_buf = pdata->ctrl.read_buf;
 	data->chip.chip_delay = pdata->chip.chip_delay;
 	data->chip.options |= pdata->chip.options;
 
diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h
index 7efb9be34662..0e35375ea795 100644
--- a/include/linux/mtd/nand.h
+++ b/include/linux/mtd/nand.h
@@ -584,6 +584,8 @@ struct platform_nand_chip {
  * @select_chip:	platform specific chip select function
  * @cmd_ctrl:		platform specific function for controlling
  *			ALE/CLE/nCE. Also used to write command and address
+ * @write_buf:		platform specific function for write buffer
+ * @read_buf:		platform specific function for read buffer
  * @priv:		private data to transport driver specific settings
  *
  * All fields are optional and depend on the hardware driver requirements
@@ -594,6 +596,10 @@ struct platform_nand_ctrl {
 	void		(*select_chip)(struct mtd_info *mtd, int chip);
 	void		(*cmd_ctrl)(struct mtd_info *mtd, int dat,
 				    unsigned int ctrl);
+	void		(*write_buf)(struct mtd_info *mtd,
+				    const uint8_t *buf, int len);
+	void		(*read_buf)(struct mtd_info *mtd,
+				    uint8_t *buf, int len);
 	void		*priv;
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From ec2d0d842577854eee18f0dc06bd48fe17189b54 Mon Sep 17 00:00:00 2001
From: Daniel Ribeiro <drwyrm@gmail.com>
Date: Sun, 17 May 2009 08:02:17 -0300
Subject: mtd: CFI: quirk for PF38F4476.

This chip reports CFI 1.3, but the CFI PRI is like CFI 1.1. Add a quirk
to pass probe on this chip.

This patch depends on "MTD: CFI 1.0 and CFI 1.1"

Signed-off-by: Daniel Ribeiro <drwyrm@gmail.com>
Acked-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/chips/cfi_cmdset_0001.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/mtd/chips/cfi_cmdset_0001.c b/drivers/mtd/chips/cfi_cmdset_0001.c
index c240454fd113..e772803b4959 100644
--- a/drivers/mtd/chips/cfi_cmdset_0001.c
+++ b/drivers/mtd/chips/cfi_cmdset_0001.c
@@ -46,6 +46,7 @@
 #define MANUFACTURER_INTEL	0x0089
 #define I82802AB	0x00ad
 #define I82802AC	0x00ac
+#define PF38F4476	0x881c
 #define MANUFACTURER_ST         0x0020
 #define M50LPW080       0x002F
 #define M50FLW080A	0x0080
@@ -315,9 +316,18 @@ static struct cfi_fixup fixup_table[] = {
 	{ 0, 0, NULL, NULL }
 };
 
+static void cfi_fixup_major_minor(struct cfi_private *cfi,
+						struct cfi_pri_intelext *extp)
+{
+	if (cfi->mfr == MANUFACTURER_INTEL &&
+			cfi->id == PF38F4476 && extp->MinorVersion == '3')
+		extp->MinorVersion = '1';
+}
+
 static inline struct cfi_pri_intelext *
 read_pri_intelext(struct map_info *map, __u16 adr)
 {
+	struct cfi_private *cfi = map->fldrv_priv;
 	struct cfi_pri_intelext *extp;
 	unsigned int extp_size = sizeof(*extp);
 
@@ -326,6 +336,8 @@ read_pri_intelext(struct map_info *map, __u16 adr)
 	if (!extp)
 		return NULL;
 
+	cfi_fixup_major_minor(cfi, extp);
+
 	if (extp->MajorVersion != '1' ||
 	    (extp->MinorVersion < '0' || extp->MinorVersion > '5')) {
 		printk(KERN_ERR "  Unknown Intel/Sharp Extended Query "
-- 
cgit v1.2.3-59-g8ed1b


From e1b158abc532f5a9d355c187583038c4f75ab11d Mon Sep 17 00:00:00 2001
From: Daniel Ribeiro <drwyrm@gmail.com>
Date: Sun, 17 May 2009 08:02:08 -0300
Subject: mtd: CFI 1.0 and CFI 1.1

This patch allows otpinfo for CFI >= 1.0 and burst read for CFI >= 1.1.

references:
1.0:	http://www.datasheetcatalog.org/datasheets2/81/816884_1.pdf

1.1:	http://milkymist.org/doc/MT28F640J3.pdf
	http://www.delorie.com/agenda/specs/29066709.pdf

Signed-off-by: Daniel Ribeiro <drwyrm@gmail.com>
Acked-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/chips/cfi_cmdset_0001.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/chips/cfi_cmdset_0001.c b/drivers/mtd/chips/cfi_cmdset_0001.c
index e772803b4959..8664feebc93b 100644
--- a/drivers/mtd/chips/cfi_cmdset_0001.c
+++ b/drivers/mtd/chips/cfi_cmdset_0001.c
@@ -329,6 +329,7 @@ read_pri_intelext(struct map_info *map, __u16 adr)
 {
 	struct cfi_private *cfi = map->fldrv_priv;
 	struct cfi_pri_intelext *extp;
+	unsigned int extra_size = 0;
 	unsigned int extp_size = sizeof(*extp);
 
  again:
@@ -352,19 +353,24 @@ read_pri_intelext(struct map_info *map, __u16 adr)
 	extp->BlkStatusRegMask = le16_to_cpu(extp->BlkStatusRegMask);
 	extp->ProtRegAddr = le16_to_cpu(extp->ProtRegAddr);
 
-	if (extp->MajorVersion == '1' && extp->MinorVersion >= '3') {
-		unsigned int extra_size = 0;
-		int nb_parts, i;
+	if (extp->MinorVersion >= '0') {
+		extra_size = 0;
 
 		/* Protection Register info */
 		extra_size += (extp->NumProtectionFields - 1) *
 			      sizeof(struct cfi_intelext_otpinfo);
+	}
 
+	if (extp->MinorVersion >= '1') {
 		/* Burst Read info */
 		extra_size += 2;
 		if (extp_size < sizeof(*extp) + extra_size)
 			goto need_more;
-		extra_size += extp->extra[extra_size-1];
+		extra_size += extp->extra[extra_size - 1];
+	}
+
+	if (extp->MinorVersion >= '3') {
+		int nb_parts, i;
 
 		/* Number of hardware-partitions */
 		extra_size += 1;
-- 
cgit v1.2.3-59-g8ed1b


From ab1ff210a86ae4ab5990b7bd2dc69fafbfa2355a Mon Sep 17 00:00:00 2001
From: Lennert Buytenhek <buytenh@wantstofly.org>
Date: Wed, 20 May 2009 13:07:11 +0200
Subject: mtd: m25p80: add support for Macronix MX25L12805D

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/devices/m25p80.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/mtd/devices/m25p80.c b/drivers/mtd/devices/m25p80.c
index 982572c1056c..59c46126a5ce 100644
--- a/drivers/mtd/devices/m25p80.c
+++ b/drivers/mtd/devices/m25p80.c
@@ -500,6 +500,9 @@ static struct flash_info __devinitdata m25p_data [] = {
 	{ "at26df161a", 0x1f4601, 0, 64 * 1024, 32, SECT_4K, },
 	{ "at26df321",  0x1f4701, 0, 64 * 1024, 64, SECT_4K, },
 
+	/* Macronix */
+	{ "mx25l12805d", 0xc22018, 0, 64 * 1024, 256, },
+
 	/* Spansion -- single (large) sector size only, at least
 	 * for the chips listed here (without boot sectors).
 	 */
-- 
cgit v1.2.3-59-g8ed1b


From d3412dbd721c0136379d86242297d19399f0c05f Mon Sep 17 00:00:00 2001
From: Mika Korhonen <mika.j.korhonen@gmail.com>
Date: Thu, 21 May 2009 23:09:42 +0300
Subject: mtd: OneNAND: add missing __devexit_p

Add missing __devexit_p wrapper and no more mark shutdown with __devexit.
Fixes build in configurations where devexit functions get discarded.

Signed-off-by: Mika Korhonen <mika.j.korhonen@gmail.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/onenand/omap2.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/onenand/omap2.c b/drivers/mtd/onenand/omap2.c
index f2e9de1414df..df26db863f2f 100644
--- a/drivers/mtd/onenand/omap2.c
+++ b/drivers/mtd/onenand/omap2.c
@@ -566,7 +566,7 @@ int omap2_onenand_rephase(void)
 				      NULL, __adjust_timing);
 }
 
-static void __devexit omap2_onenand_shutdown(struct platform_device *pdev)
+static void omap2_onenand_shutdown(struct platform_device *pdev)
 {
 	struct omap2_onenand *c = dev_get_drvdata(&pdev->dev);
 
@@ -778,7 +778,7 @@ static int __devexit omap2_onenand_remove(struct platform_device *pdev)
 
 static struct platform_driver omap2_onenand_driver = {
 	.probe		= omap2_onenand_probe,
-	.remove		= omap2_onenand_remove,
+	.remove		= __devexit_p(omap2_onenand_remove),
 	.shutdown	= omap2_onenand_shutdown,
 	.driver		= {
 		.name	= DRIVER_NAME,
-- 
cgit v1.2.3-59-g8ed1b


From 4d964824ec826ed97bdde10bc8d8c4ce10540a98 Mon Sep 17 00:00:00 2001
From: Shane McDonald <mcdonald.shane@gmail.com>
Date: Sat, 2 May 2009 09:40:06 -0600
Subject: mtd: remove pmcmsp-ramroot.c

The RAMROOT function was a successful but non-portable attempt to append
the root filesystem to the end of the kernel image.  The preferred and
portable solution is to use an initramfs instead.

The only user of this function was the msp71xx configuration
in the MIPS architecture; as the use of the RAMROOT has been removed
from that configuration, there are no more users, so this code
can be removed.

Signed-off-by: Shane McDonald <mcdonald.shane@gmail.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/Kconfig          |   9 ----
 drivers/mtd/maps/Makefile         |   1 -
 drivers/mtd/maps/pmcmsp-ramroot.c | 104 --------------------------------------
 3 files changed, 114 deletions(-)
 delete mode 100644 drivers/mtd/maps/pmcmsp-ramroot.c

diff --git a/drivers/mtd/maps/Kconfig b/drivers/mtd/maps/Kconfig
index cdd54cb4d224..91544c2c99a6 100644
--- a/drivers/mtd/maps/Kconfig
+++ b/drivers/mtd/maps/Kconfig
@@ -105,15 +105,6 @@ config MSP_FLASH_MAP_LIMIT
 	default "0x02000000"
 	depends on MSP_FLASH_MAP_LIMIT_32M
 
-config MTD_PMC_MSP_RAMROOT
-	tristate "Embedded RAM block device for root on PMC-Sierra MSP"
-	depends on PMC_MSP_EMBEDDED_ROOTFS && \
-			(MTD_BLOCK || MTD_BLOCK_RO) && \
-			MTD_RAM
-	help
-	  This provides support for the embedded root file system
-          on PMC MSP devices.  This memory is mapped as a MTD block device.
-
 config MTD_SUN_UFLASH
 	tristate "Sun Microsystems userflash support"
 	depends on SPARC && MTD_CFI && PCI
diff --git a/drivers/mtd/maps/Makefile b/drivers/mtd/maps/Makefile
index 2dbc1bec8488..8bae7f9850c0 100644
--- a/drivers/mtd/maps/Makefile
+++ b/drivers/mtd/maps/Makefile
@@ -25,7 +25,6 @@ obj-$(CONFIG_MTD_OCTAGON)	+= octagon-5066.o
 obj-$(CONFIG_MTD_PHYSMAP)	+= physmap.o
 obj-$(CONFIG_MTD_PHYSMAP_OF)	+= physmap_of.o
 obj-$(CONFIG_MTD_PMC_MSP_EVM)   += pmcmsp-flash.o
-obj-$(CONFIG_MTD_PMC_MSP_RAMROOT)+= pmcmsp-ramroot.o
 obj-$(CONFIG_MTD_PCMCIA)	+= pcmciamtd.o
 obj-$(CONFIG_MTD_RPXLITE)	+= rpxlite.o
 obj-$(CONFIG_MTD_TQM8XXL)	+= tqm8xxl.o
diff --git a/drivers/mtd/maps/pmcmsp-ramroot.c b/drivers/mtd/maps/pmcmsp-ramroot.c
deleted file mode 100644
index 30de5c0c09a9..000000000000
--- a/drivers/mtd/maps/pmcmsp-ramroot.c
+++ /dev/null
@@ -1,104 +0,0 @@
-/*
- * Mapping of the rootfs in a physical region of memory
- *
- * Copyright (C) 2005-2007 PMC-Sierra Inc.
- * Author: Andrew Hughes, Andrew_Hughes@pmc-sierra.com
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-#include <linux/module.h>
-#include <linux/types.h>
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/slab.h>
-#include <linux/fs.h>
-#include <linux/root_dev.h>
-#include <linux/mtd/mtd.h>
-#include <linux/mtd/map.h>
-
-#include <asm/io.h>
-
-#include <msp_prom.h>
-
-static struct mtd_info *rr_mtd;
-
-struct map_info rr_map = {
-	.name = "ramroot",
-	.bankwidth = 4,
-};
-
-static int __init init_rrmap(void)
-{
-	void *ramroot_start;
-	unsigned long ramroot_size;
-
-	/* Check for supported rootfs types */
-	if (get_ramroot(&ramroot_start, &ramroot_size)) {
-		rr_map.phys = CPHYSADDR(ramroot_start);
-		rr_map.size = ramroot_size;
-
-		printk(KERN_NOTICE
-			"PMC embedded root device: 0x%08lx @ 0x%08lx\n",
-			rr_map.size, (unsigned long)rr_map.phys);
-	} else {
-		printk(KERN_ERR
-			"init_rrmap: no supported embedded rootfs detected!\n");
-		return -ENXIO;
-	}
-
-	/* Map rootfs to I/O space for block device driver */
-	rr_map.virt = ioremap(rr_map.phys, rr_map.size);
-	if (!rr_map.virt) {
-		printk(KERN_ERR "Failed to ioremap\n");
-		return -EIO;
-	}
-
-	simple_map_init(&rr_map);
-
-	rr_mtd = do_map_probe("map_ram", &rr_map);
-	if (rr_mtd) {
-		rr_mtd->owner = THIS_MODULE;
-
-		add_mtd_device(rr_mtd);
-
-		return 0;
-	}
-
-	iounmap(rr_map.virt);
-	return -ENXIO;
-}
-
-static void __exit cleanup_rrmap(void)
-{
-	del_mtd_device(rr_mtd);
-	map_destroy(rr_mtd);
-
-	iounmap(rr_map.virt);
-	rr_map.virt = NULL;
-}
-
-MODULE_AUTHOR("PMC-Sierra, Inc");
-MODULE_DESCRIPTION("MTD map driver for embedded PMC-Sierra MSP filesystem");
-MODULE_LICENSE("GPL");
-
-module_init(init_rrmap);
-module_exit(cleanup_rrmap);
-- 
cgit v1.2.3-59-g8ed1b


From bf95efd41b1a760128eb25402791b0a4941eb655 Mon Sep 17 00:00:00 2001
From: H Hartley Sweeten <hsweeten@visionengravers.com>
Date: Tue, 12 May 2009 13:46:58 -0700
Subject: mtd: plat_nand: add platform probe/remove callbacks

Add optional probe and remove callbacks to the plat_nand driver.

Some platforms may require additional setup, such as configuring the
memory controller, before the nand device can be accessed.  This patch
provides an optional callback to handle this setup as well as a callback
to teardown the setup.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Tested-by: Alexander Clouter <alex@digriz.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/plat_nand.c | 13 +++++++++++--
 include/linux/mtd/nand.h     |  7 +++++++
 2 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/plat_nand.c b/drivers/mtd/nand/plat_nand.c
index 47a2105b9671..22e0ce788419 100644
--- a/drivers/mtd/nand/plat_nand.c
+++ b/drivers/mtd/nand/plat_nand.c
@@ -72,6 +72,13 @@ static int __devinit plat_nand_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, data);
 
+	/* Handle any platform specific setup */
+	if (pdata->ctrl.probe) {
+		res = pdata->ctrl.probe(pdev);
+		if (res)
+			goto out;
+	}
+
 	/* Scan to find existance of the device */
 	if (nand_scan(&data->mtd, 1)) {
 		res = -ENXIO;
@@ -101,6 +108,8 @@ static int __devinit plat_nand_probe(struct platform_device *pdev)
 
 	nand_release(&data->mtd);
 out:
+	if (pdata->ctrl.remove)
+		pdata->ctrl.remove(pdev);
 	platform_set_drvdata(pdev, NULL);
 	iounmap(data->io_base);
 	kfree(data);
@@ -113,15 +122,15 @@ out:
 static int __devexit plat_nand_remove(struct platform_device *pdev)
 {
 	struct plat_nand_data *data = platform_get_drvdata(pdev);
-#ifdef CONFIG_MTD_PARTITIONS
 	struct platform_nand_data *pdata = pdev->dev.platform_data;
-#endif
 
 	nand_release(&data->mtd);
 #ifdef CONFIG_MTD_PARTITIONS
 	if (data->parts && data->parts != pdata->chip.partitions)
 		kfree(data->parts);
 #endif
+	if (pdata->ctrl.remove)
+		pdata->ctrl.remove(pdev);
 	iounmap(data->io_base);
 	kfree(data);
 
diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h
index 0e35375ea795..7f2d69356554 100644
--- a/include/linux/mtd/nand.h
+++ b/include/linux/mtd/nand.h
@@ -577,8 +577,13 @@ struct platform_nand_chip {
 	void			*priv;
 };
 
+/* Keep gcc happy */
+struct platform_device;
+
 /**
  * struct platform_nand_ctrl - controller level device structure
+ * @probe:		platform specific function to probe/setup hardware
+ * @remove:		platform specific function to remove/teardown hardware
  * @hwcontrol:		platform specific hardware control structure
  * @dev_ready:		platform specific function to read ready/busy pin
  * @select_chip:	platform specific chip select function
@@ -591,6 +596,8 @@ struct platform_nand_chip {
  * All fields are optional and depend on the hardware driver requirements
  */
 struct platform_nand_ctrl {
+	int		(*probe)(struct platform_device *pdev);
+	void		(*remove)(struct platform_device *pdev);
 	void		(*hwcontrol)(struct mtd_info *mtd, int cmd);
 	int		(*dev_ready)(struct mtd_info *mtd);
 	void		(*select_chip)(struct mtd_info *mtd, int chip);
-- 
cgit v1.2.3-59-g8ed1b


From f36e20c01ad0104688f2eaebdf2213e749929c97 Mon Sep 17 00:00:00 2001
From: H Hartley Sweeten <hsweeten@visionengravers.com>
Date: Tue, 12 May 2009 13:46:59 -0700
Subject: mtd: plat_nand: allow platform to set partitions

Add optional callback to allow platform to initialize partitions.

Static partitions on a nand device could vary depending on the size of the
device.  This patch allows an optional platform callback to be used to
setup this partition information at runtime.

Scan order is:
	1) chip.part_probe_types
	2) chip.set_parts
	3) chip.partitions
	4) full mtd device (fallback for no partitions)

Some of the existing nand drivers could possibly be replaced by the
plat_nand driver by using this patch.  These include autcpu12.c and
ts7250.c drivers.

Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/plat_nand.c | 2 ++
 include/linux/mtd/nand.h     | 3 +++
 2 files changed, 5 insertions(+)

diff --git a/drivers/mtd/nand/plat_nand.c b/drivers/mtd/nand/plat_nand.c
index 22e0ce788419..4e16c6f5bdd5 100644
--- a/drivers/mtd/nand/plat_nand.c
+++ b/drivers/mtd/nand/plat_nand.c
@@ -95,6 +95,8 @@ static int __devinit plat_nand_probe(struct platform_device *pdev)
 			return 0;
 		}
 	}
+	if (pdata->chip.set_parts)
+		pdata->chip.set_parts(data->mtd.size, &pdata->chip);
 	if (pdata->chip.partitions) {
 		data->parts = pdata->chip.partitions;
 		res = add_mtd_partitions(&data->mtd, data->parts,
diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h
index 7f2d69356554..4030ebada49e 100644
--- a/include/linux/mtd/nand.h
+++ b/include/linux/mtd/nand.h
@@ -563,6 +563,7 @@ extern int nand_do_read(struct mtd_info *mtd, loff_t from, size_t len,
  * @options:		Option flags, e.g. 16bit buswidth
  * @ecclayout:		ecc layout info structure
  * @part_probe_types:	NULL-terminated array of probe types
+ * @set_parts:		platform specific function to set partitions
  * @priv:		hardware controller specific settings
  */
 struct platform_nand_chip {
@@ -574,6 +575,8 @@ struct platform_nand_chip {
 	int			chip_delay;
 	unsigned int		options;
 	const char		**part_probe_types;
+	void			(*set_parts)(uint64_t size,
+					struct platform_nand_chip *chip);
 	void			*priv;
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From bd3fd62ecc99c709739cb969be76f44903a4043b Mon Sep 17 00:00:00 2001
From: Vladimir Barinov <vova.barinov@gmail.com>
Date: Mon, 25 May 2009 13:06:17 +0400
Subject: mtd: MXC NAND support for 2KiB page size flashes

- Add support for 2KiB page size flashes
- Fix page address access for large pages
- Detect oob layout at runtime
- handle pagesize_2k variable
- Fix oob16 layout: reserve location 5 of oob area since it's used for bbt

Signed-off-by: Vladimir Barinov <vova.barinov@gmail.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/mxc_nand.c | 60 ++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 51 insertions(+), 9 deletions(-)

diff --git a/drivers/mtd/nand/mxc_nand.c b/drivers/mtd/nand/mxc_nand.c
index d5aea6951d0b..d03bd4eff722 100644
--- a/drivers/mtd/nand/mxc_nand.c
+++ b/drivers/mtd/nand/mxc_nand.c
@@ -138,7 +138,14 @@ static struct nand_ecclayout nand_hw_eccoob_8 = {
 static struct nand_ecclayout nand_hw_eccoob_16 = {
 	.eccbytes = 5,
 	.eccpos = {6, 7, 8, 9, 10},
-	.oobfree = {{0, 6}, {12, 4}, }
+	.oobfree = {{0, 5}, {11, 5}, }
+};
+
+static struct nand_ecclayout nand_hw_eccoob_64 = {
+	.eccbytes = 20,
+	.eccpos = {6, 7, 8, 9, 10, 22, 23, 24, 25, 26,
+		   38, 39, 40, 41, 42, 54, 55, 56, 57, 58},
+	.oobfree = {{2, 4}, {11, 10}, {27, 10}, {43, 10}, {59, 5}, }
 };
 
 #ifdef CONFIG_MTD_PARTITIONS
@@ -795,9 +802,13 @@ static void mxc_nand_command(struct mtd_info *mtd, unsigned command,
 		send_addr(host, (page_addr & 0xff), false);
 
 		if (host->pagesize_2k) {
-			send_addr(host, (page_addr >> 8) & 0xFF, false);
-			if (mtd->size >= 0x40000000)
+			if (mtd->size >= 0x10000000) {
+				/* paddr_8 - paddr_15 */
+				send_addr(host, (page_addr >> 8) & 0xff, false);
 				send_addr(host, (page_addr >> 16) & 0xff, true);
+			} else
+				/* paddr_8 - paddr_15 */
+				send_addr(host, (page_addr >> 8) & 0xff, true);
 		} else {
 			/* One more address cycle for higher density devices */
 			if (mtd->size >= 0x4000000) {
@@ -919,7 +930,6 @@ static int __init mxcnd_probe(struct platform_device *pdev)
 		this->ecc.mode = NAND_ECC_HW;
 		this->ecc.size = 512;
 		this->ecc.bytes = 3;
-		this->ecc.layout = &nand_hw_eccoob_8;
 		tmp = readw(host->regs + NFC_CONFIG1);
 		tmp |= NFC_ECC_EN;
 		writew(tmp, host->regs + NFC_CONFIG1);
@@ -953,12 +963,44 @@ static int __init mxcnd_probe(struct platform_device *pdev)
 		this->ecc.layout = &nand_hw_eccoob_16;
 	}
 
-	host->pagesize_2k = 0;
+	/* first scan to find the device and get the page size */
+	if (nand_scan_ident(mtd, 1)) {
+		err = -ENXIO;
+		goto escan;
+	}
 
-	/* Scan to find existence of the device */
-	if (nand_scan(mtd, 1)) {
-		DEBUG(MTD_DEBUG_LEVEL0,
-		      "MXC_ND: Unable to find any NAND device.\n");
+	host->pagesize_2k = (mtd->writesize == 2048) ? 1 : 0;
+
+	if (this->ecc.mode == NAND_ECC_HW) {
+		switch (mtd->oobsize) {
+		case 8:
+			this->ecc.layout = &nand_hw_eccoob_8;
+			break;
+		case 16:
+			this->ecc.layout = &nand_hw_eccoob_16;
+			break;
+		case 64:
+			this->ecc.layout = &nand_hw_eccoob_64;
+			break;
+		default:
+			/* page size not handled by HW ECC */
+			/* switching back to soft ECC */
+			this->ecc.size = 512;
+			this->ecc.bytes = 3;
+			this->ecc.layout = &nand_hw_eccoob_8;
+			this->ecc.mode = NAND_ECC_SOFT;
+			this->ecc.calculate = NULL;
+			this->ecc.correct = NULL;
+			this->ecc.hwctl = NULL;
+			tmp = readw(host->regs + NFC_CONFIG1);
+			tmp &= ~NFC_ECC_EN;
+			writew(tmp, host->regs + NFC_CONFIG1);
+			break;
+		}
+	}
+
+	/* second phase scan */
+	if (nand_scan_tail(mtd)) {
 		err = -ENXIO;
 		goto escan;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From f4fa697c26bcd9e942de26bad970f4de1da5a49b Mon Sep 17 00:00:00 2001
From: Simon Polette <spolette@adetelgroup.com>
Date: Wed, 27 May 2009 18:19:39 +0300
Subject: mtd: add on-flash BBT support for Atmel NAND driver

Just add a new on-flash-bbt module parameter.

Signed-off-by: Simon Polette <spolette@adetelgroup.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/atmel_nand.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/nand/atmel_nand.c b/drivers/mtd/nand/atmel_nand.c
index 47a33cec3793..2802992b39da 100644
--- a/drivers/mtd/nand/atmel_nand.c
+++ b/drivers/mtd/nand/atmel_nand.c
@@ -24,6 +24,7 @@
 
 #include <linux/slab.h>
 #include <linux/module.h>
+#include <linux/moduleparam.h>
 #include <linux/platform_device.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/nand.h>
@@ -47,6 +48,9 @@
 #define no_ecc		0
 #endif
 
+static int on_flash_bbt = 0;
+module_param(on_flash_bbt, int, 0);
+
 /* Register access macros */
 #define ecc_readl(add, reg)				\
 	__raw_readl(add + ATMEL_ECC_##reg)
@@ -459,12 +463,17 @@ static int __init atmel_nand_probe(struct platform_device *pdev)
 
 	if (host->board->det_pin) {
 		if (gpio_get_value(host->board->det_pin)) {
-			printk("No SmartMedia card inserted.\n");
+			printk(KERN_INFO "No SmartMedia card inserted.\n");
 			res = ENXIO;
 			goto err_no_card;
 		}
 	}
 
+	if (on_flash_bbt) {
+		printk(KERN_INFO "atmel_nand: Use On Flash BBT\n");
+		nand_chip->options |= NAND_USE_FLASH_BBT;
+	}
+
 	/* first scan to find the device and get the page size */
 	if (nand_scan_ident(mtd, 1)) {
 		res = -ENXIO;
-- 
cgit v1.2.3-59-g8ed1b


From 530c3b60658687e2ad7bf98ef83631a8280ae8a6 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Tue, 26 May 2009 06:24:13 -0400
Subject: mtd: blackfin NFC: remove pointless return value in bf5xx_nand_dma_rw

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/bf5xx_nand.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/mtd/nand/bf5xx_nand.c b/drivers/mtd/nand/bf5xx_nand.c
index 4c2a67ca801e..2ab42d2d4f91 100644
--- a/drivers/mtd/nand/bf5xx_nand.c
+++ b/drivers/mtd/nand/bf5xx_nand.c
@@ -458,7 +458,7 @@ static irqreturn_t bf5xx_nand_dma_irq(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static int bf5xx_nand_dma_rw(struct mtd_info *mtd,
+static void bf5xx_nand_dma_rw(struct mtd_info *mtd,
 				uint8_t *buf, int is_read)
 {
 	struct bf5xx_nand_info *info = mtd_to_nand_info(mtd);
@@ -512,8 +512,6 @@ static int bf5xx_nand_dma_rw(struct mtd_info *mtd,
 	else
 		bfin_write_NFC_PGCTL(0x2);
 	wait_for_completion(&info->dma_completion);
-
-	return 0;
 }
 
 static void bf5xx_nand_dma_read_buf(struct mtd_info *mtd,
-- 
cgit v1.2.3-59-g8ed1b


From c3a9f35673290f49ec115d36ad283961c82c135a Mon Sep 17 00:00:00 2001
From: Cliff Cai <cliff.cai@analog.com>
Date: Tue, 26 May 2009 06:24:14 -0400
Subject: mtd: blackfin NFC: fix hang when using NAND on BF527-EZKITs

The DMAs have different bit sizes on BF52x and BF54x.  From the PHRM:

"The 16-bit DMA Access Bus (DAB) connects the DMA controller to the
on-chip peripherals, PPI, SPI, Ethernet MAC, the SPORTs, NFC,
HOSTDP and the UARTs."

32-bit DMA won't work for BF52x.

Signed-off-by: Cliff Cai <cliff.cai@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/bf5xx_nand.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/bf5xx_nand.c b/drivers/mtd/nand/bf5xx_nand.c
index 2ab42d2d4f91..8506e7e606fd 100644
--- a/drivers/mtd/nand/bf5xx_nand.c
+++ b/drivers/mtd/nand/bf5xx_nand.c
@@ -496,11 +496,20 @@ static void bf5xx_nand_dma_rw(struct mtd_info *mtd,
 	/* setup DMA register with Blackfin DMA API */
 	set_dma_config(CH_NFC, 0x0);
 	set_dma_start_addr(CH_NFC, (unsigned long) buf);
+
+/* The DMAs have different size on BF52x and BF54x */
+#ifdef CONFIG_BF52x
+	set_dma_x_count(CH_NFC, (page_size >> 1));
+	set_dma_x_modify(CH_NFC, 2);
+	val = DI_EN | WDSIZE_16;
+#endif
+
+#ifdef CONFIG_BF54x
 	set_dma_x_count(CH_NFC, (page_size >> 2));
 	set_dma_x_modify(CH_NFC, 4);
-
-	/* setup write or read operation */
 	val = DI_EN | WDSIZE_32;
+#endif
+	/* setup write or read operation */
 	if (is_read)
 		val |= WNR;
 	set_dma_config(CH_NFC, val);
-- 
cgit v1.2.3-59-g8ed1b


From fa254ecbcca713a4aec99478e79f858942b3d4e0 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Tue, 26 May 2009 19:33:16 -0400
Subject: mtd: uclinux: allow systems to override map addr/size

Due to a processor anomaly (05000263 to be exact), most Blackfin parts
cannot keep the embedded filesystem image directly after the kernel in
RAM.  Instead, the filesystem needs to be relocated to the end of memory.
As such, we need to tweak the map addr/size during boot for Blackfin
systems.  This can be done in any early arch/board init code.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
CC: Paul Mundt <lethal@linux-sh.org>
CC: Greg Ungerer <gerg@uclinux.org>
CC: uclinux-dev@uclinux.org
CC: linux-mtd@lists.infradead.org
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/uclinux.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/maps/uclinux.c b/drivers/mtd/maps/uclinux.c
index 81756e397711..61d4087dd8f7 100644
--- a/drivers/mtd/maps/uclinux.c
+++ b/drivers/mtd/maps/uclinux.c
@@ -22,8 +22,12 @@
 
 /****************************************************************************/
 
+extern char _ebss;
+
 struct map_info uclinux_ram_map = {
 	.name = "RAM",
+	.phys = (unsigned long)&_ebss,
+	.size = 0,
 };
 
 struct mtd_info *uclinux_ram_mtdinfo;
@@ -55,12 +59,10 @@ static int __init uclinux_mtd_init(void)
 {
 	struct mtd_info *mtd;
 	struct map_info *mapp;
-	extern char _ebss;
-	unsigned long addr = (unsigned long) &_ebss;
 
 	mapp = &uclinux_ram_map;
-	mapp->phys = addr;
-	mapp->size = PAGE_ALIGN(ntohl(*((unsigned long *)(addr + 8))));
+	if (!mapp->size)
+		mapp->size = PAGE_ALIGN(ntohl(*((unsigned long *)(mapp->phys + 8))));
 	mapp->bankwidth = 4;
 
 	printk("uclinux[mtd]: RAM probe address=0x%x size=0x%x\n",
-- 
cgit v1.2.3-59-g8ed1b


From 6ae392ccadbc622d58a9d01a7ee59e340f82fe85 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Tue, 26 May 2009 19:33:18 -0400
Subject: mtd: uclinux: do not allow to be built as a module

There isn't any benefit to building the uClinux MTD map as a module as the
rootfs it requires in order to actually be usable is appended to the
kernel image, not the module.  No known system builds it this way either,
so change the option to "bool".

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
CC: Paul Mundt <lethal@linux-sh.org>
CC: Greg Ungerer <gerg@uclinux.org>
CC: uclinux-dev@uclinux.org
CC: linux-mtd@lists.infradead.org
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/maps/Kconfig b/drivers/mtd/maps/Kconfig
index 91544c2c99a6..0b98654d8eed 100644
--- a/drivers/mtd/maps/Kconfig
+++ b/drivers/mtd/maps/Kconfig
@@ -492,7 +492,7 @@ config MTD_BFIN_ASYNC
 	  If compiled as a module, it will be called bfin-async-flash.
 
 config MTD_UCLINUX
-	tristate "Generic uClinux RAM/ROM filesystem support"
+	bool "Generic uClinux RAM/ROM filesystem support"
 	depends on MTD_PARTITIONS && MTD_RAM && !MMU
 	help
 	  Map driver to support image based filesystems for uClinux.
-- 
cgit v1.2.3-59-g8ed1b


From 9f31f4b9dccf6fa4a606ae04602ec232b94727fb Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Tue, 26 May 2009 19:33:17 -0400
Subject: mtd: uclinux: mark local stuff static

The uclinux_ram_mtdinfo, uclinux_romfs, and uclinux_point symbols do not
need to be visible outside of this module, so mark them static.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
CC: Greg Ungerer <gerg@uclinux.org>
CC: uclinux-dev@uclinux.org
CC: linux-mtd@lists.infradead.org
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/uclinux.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/mtd/maps/uclinux.c b/drivers/mtd/maps/uclinux.c
index 61d4087dd8f7..d4314fb88212 100644
--- a/drivers/mtd/maps/uclinux.c
+++ b/drivers/mtd/maps/uclinux.c
@@ -30,11 +30,11 @@ struct map_info uclinux_ram_map = {
 	.size = 0,
 };
 
-struct mtd_info *uclinux_ram_mtdinfo;
+static struct mtd_info *uclinux_ram_mtdinfo;
 
 /****************************************************************************/
 
-struct mtd_partition uclinux_romfs[] = {
+static struct mtd_partition uclinux_romfs[] = {
 	{ .name = "ROMfs" }
 };
 
@@ -42,7 +42,7 @@ struct mtd_partition uclinux_romfs[] = {
 
 /****************************************************************************/
 
-int uclinux_point(struct mtd_info *mtd, loff_t from, size_t len,
+static int uclinux_point(struct mtd_info *mtd, loff_t from, size_t len,
 	size_t *retlen, void **virt, resource_size_t *phys)
 {
 	struct map_info *map = mtd->priv;
-- 
cgit v1.2.3-59-g8ed1b


From 4938c88c922fad23f0a9f404eeda0207a819e4df Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Tue, 2 Jun 2009 00:06:23 -0400
Subject: mtd: maps: Blackfin async: fix memory leaks in probe/remove funcs

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/bfin-async-flash.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/mtd/maps/bfin-async-flash.c b/drivers/mtd/maps/bfin-async-flash.c
index 576611f605db..365c77b1b871 100644
--- a/drivers/mtd/maps/bfin-async-flash.c
+++ b/drivers/mtd/maps/bfin-async-flash.c
@@ -40,6 +40,9 @@ struct async_state {
 	uint32_t flash_ambctl0, flash_ambctl1;
 	uint32_t save_ambctl0, save_ambctl1;
 	unsigned long irq_flags;
+#ifdef CONFIG_MTD_PARTITIONS
+	struct mtd_partition *parts;
+#endif
 };
 
 static void switch_to_flash(struct async_state *state)
@@ -170,6 +173,7 @@ static int __devinit bfin_flash_probe(struct platform_device *pdev)
 	if (ret > 0) {
 		pr_devinit(KERN_NOTICE DRIVER_NAME ": Using commandline partition definition\n");
 		add_mtd_partitions(state->mtd, pdata->parts, ret);
+		state->parts = pdata->parts;
 
 	} else if (pdata->nr_parts) {
 		pr_devinit(KERN_NOTICE DRIVER_NAME ": Using board partition definition\n");
@@ -193,6 +197,7 @@ static int __devexit bfin_flash_remove(struct platform_device *pdev)
 	gpio_free(state->enet_flash_pin);
 #ifdef CONFIG_MTD_PARTITIONS
 	del_mtd_partitions(state->mtd);
+	kfree(state->parts);
 #endif
 	map_destroy(state->mtd);
 	kfree(state);
-- 
cgit v1.2.3-59-g8ed1b


From 67ce04bf2746f8a1f8c2a104b313d20c63f68378 Mon Sep 17 00:00:00 2001
From: Vimal Singh <vimalsingh@ti.com>
Date: Tue, 12 May 2009 13:47:03 -0700
Subject: mtd: nand: add OMAP2/OMAP3 NAND driver

This driver is present in the OMAP tree, now pushing it to MTD.

Original author(s):
       Jian Zhang <jzhang@ti.com>

Signed-off-by: Vimal Singh <vimalsingh@ti.com>
Cc: Jian Zhang <jzhang@ti.com>
Cc: Artem Bityutskiy <dedekind@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/Kconfig  |   6 +
 drivers/mtd/nand/Makefile |   1 +
 drivers/mtd/nand/omap2.c  | 776 ++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 783 insertions(+)
 create mode 100644 drivers/mtd/nand/omap2.c

diff --git a/drivers/mtd/nand/Kconfig b/drivers/mtd/nand/Kconfig
index 890936d0275e..3b3a21d1a6ba 100644
--- a/drivers/mtd/nand/Kconfig
+++ b/drivers/mtd/nand/Kconfig
@@ -74,6 +74,12 @@ config MTD_NAND_AMS_DELTA
 	help
 	  Support for NAND flash on Amstrad E3 (Delta).
 
+config MTD_NAND_OMAP2
+	tristate "NAND Flash device on OMAP2 and OMAP3"
+	depends on ARM && MTD_NAND && (ARCH_OMAP2 || ARCH_OMAP3)
+	help
+          Support for NAND flash on Texas Instruments OMAP2 and OMAP3 platforms.
+
 config MTD_NAND_TS7250
 	tristate "NAND Flash device on TS-7250 board"
 	depends on MACH_TS72XX
diff --git a/drivers/mtd/nand/Makefile b/drivers/mtd/nand/Makefile
index d33860ac42c3..f3a786b3cff3 100644
--- a/drivers/mtd/nand/Makefile
+++ b/drivers/mtd/nand/Makefile
@@ -25,6 +25,7 @@ obj-$(CONFIG_MTD_NAND_CS553X)		+= cs553x_nand.o
 obj-$(CONFIG_MTD_NAND_NDFC)		+= ndfc.o
 obj-$(CONFIG_MTD_NAND_ATMEL)		+= atmel_nand.o
 obj-$(CONFIG_MTD_NAND_GPIO)		+= gpio.o
+obj-$(CONFIG_MTD_NAND_OMAP2) 		+= omap2.o
 obj-$(CONFIG_MTD_NAND_CM_X270)		+= cmx270_nand.o
 obj-$(CONFIG_MTD_NAND_BASLER_EXCITE)	+= excite_nandflash.o
 obj-$(CONFIG_MTD_NAND_PXA3xx)		+= pxa3xx_nand.o
diff --git a/drivers/mtd/nand/omap2.c b/drivers/mtd/nand/omap2.c
new file mode 100644
index 000000000000..0cd76f89f4b0
--- /dev/null
+++ b/drivers/mtd/nand/omap2.c
@@ -0,0 +1,776 @@
+/*
+ * Copyright © 2004 Texas Instruments, Jian Zhang <jzhang@ti.com>
+ * Copyright © 2004 Micron Technology Inc.
+ * Copyright © 2004 David Brownell
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/platform_device.h>
+#include <linux/dma-mapping.h>
+#include <linux/delay.h>
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/nand.h>
+#include <linux/mtd/partitions.h>
+#include <linux/io.h>
+
+#include <asm/dma.h>
+
+#include <mach/gpmc.h>
+#include <mach/nand.h>
+
+#define GPMC_IRQ_STATUS		0x18
+#define GPMC_ECC_CONFIG		0x1F4
+#define GPMC_ECC_CONTROL	0x1F8
+#define GPMC_ECC_SIZE_CONFIG	0x1FC
+#define GPMC_ECC1_RESULT	0x200
+
+#define	DRIVER_NAME	"omap2-nand"
+
+/* size (4 KiB) for IO mapping */
+#define	NAND_IO_SIZE	SZ_4K
+
+#define	NAND_WP_OFF	0
+#define NAND_WP_BIT	0x00000010
+#define WR_RD_PIN_MONITORING	0x00600000
+
+#define	GPMC_BUF_FULL	0x00000001
+#define	GPMC_BUF_EMPTY	0x00000000
+
+#define NAND_Ecc_P1e		(1 << 0)
+#define NAND_Ecc_P2e		(1 << 1)
+#define NAND_Ecc_P4e		(1 << 2)
+#define NAND_Ecc_P8e		(1 << 3)
+#define NAND_Ecc_P16e		(1 << 4)
+#define NAND_Ecc_P32e		(1 << 5)
+#define NAND_Ecc_P64e		(1 << 6)
+#define NAND_Ecc_P128e		(1 << 7)
+#define NAND_Ecc_P256e		(1 << 8)
+#define NAND_Ecc_P512e		(1 << 9)
+#define NAND_Ecc_P1024e		(1 << 10)
+#define NAND_Ecc_P2048e		(1 << 11)
+
+#define NAND_Ecc_P1o		(1 << 16)
+#define NAND_Ecc_P2o		(1 << 17)
+#define NAND_Ecc_P4o		(1 << 18)
+#define NAND_Ecc_P8o		(1 << 19)
+#define NAND_Ecc_P16o		(1 << 20)
+#define NAND_Ecc_P32o		(1 << 21)
+#define NAND_Ecc_P64o		(1 << 22)
+#define NAND_Ecc_P128o		(1 << 23)
+#define NAND_Ecc_P256o		(1 << 24)
+#define NAND_Ecc_P512o		(1 << 25)
+#define NAND_Ecc_P1024o		(1 << 26)
+#define NAND_Ecc_P2048o		(1 << 27)
+
+#define TF(value)	(value ? 1 : 0)
+
+#define P2048e(a)	(TF(a & NAND_Ecc_P2048e)	<< 0)
+#define P2048o(a)	(TF(a & NAND_Ecc_P2048o)	<< 1)
+#define P1e(a)		(TF(a & NAND_Ecc_P1e)		<< 2)
+#define P1o(a)		(TF(a & NAND_Ecc_P1o)		<< 3)
+#define P2e(a)		(TF(a & NAND_Ecc_P2e)		<< 4)
+#define P2o(a)		(TF(a & NAND_Ecc_P2o)		<< 5)
+#define P4e(a)		(TF(a & NAND_Ecc_P4e)		<< 6)
+#define P4o(a)		(TF(a & NAND_Ecc_P4o)		<< 7)
+
+#define P8e(a)		(TF(a & NAND_Ecc_P8e)		<< 0)
+#define P8o(a)		(TF(a & NAND_Ecc_P8o)		<< 1)
+#define P16e(a)		(TF(a & NAND_Ecc_P16e)		<< 2)
+#define P16o(a)		(TF(a & NAND_Ecc_P16o)		<< 3)
+#define P32e(a)		(TF(a & NAND_Ecc_P32e)		<< 4)
+#define P32o(a)		(TF(a & NAND_Ecc_P32o)		<< 5)
+#define P64e(a)		(TF(a & NAND_Ecc_P64e)		<< 6)
+#define P64o(a)		(TF(a & NAND_Ecc_P64o)		<< 7)
+
+#define P128e(a)	(TF(a & NAND_Ecc_P128e)		<< 0)
+#define P128o(a)	(TF(a & NAND_Ecc_P128o)		<< 1)
+#define P256e(a)	(TF(a & NAND_Ecc_P256e)		<< 2)
+#define P256o(a)	(TF(a & NAND_Ecc_P256o)		<< 3)
+#define P512e(a)	(TF(a & NAND_Ecc_P512e)		<< 4)
+#define P512o(a)	(TF(a & NAND_Ecc_P512o)		<< 5)
+#define P1024e(a)	(TF(a & NAND_Ecc_P1024e)	<< 6)
+#define P1024o(a)	(TF(a & NAND_Ecc_P1024o)	<< 7)
+
+#define P8e_s(a)	(TF(a & NAND_Ecc_P8e)		<< 0)
+#define P8o_s(a)	(TF(a & NAND_Ecc_P8o)		<< 1)
+#define P16e_s(a)	(TF(a & NAND_Ecc_P16e)		<< 2)
+#define P16o_s(a)	(TF(a & NAND_Ecc_P16o)		<< 3)
+#define P1e_s(a)	(TF(a & NAND_Ecc_P1e)		<< 4)
+#define P1o_s(a)	(TF(a & NAND_Ecc_P1o)		<< 5)
+#define P2e_s(a)	(TF(a & NAND_Ecc_P2e)		<< 6)
+#define P2o_s(a)	(TF(a & NAND_Ecc_P2o)		<< 7)
+
+#define P4e_s(a)	(TF(a & NAND_Ecc_P4e)		<< 0)
+#define P4o_s(a)	(TF(a & NAND_Ecc_P4o)		<< 1)
+
+#ifdef CONFIG_MTD_PARTITIONS
+static const char *part_probes[] = { "cmdlinepart", NULL };
+#endif
+
+struct omap_nand_info {
+	struct nand_hw_control		controller;
+	struct omap_nand_platform_data	*pdata;
+	struct mtd_info			mtd;
+	struct mtd_partition		*parts;
+	struct nand_chip		nand;
+	struct platform_device		*pdev;
+
+	int				gpmc_cs;
+	unsigned long			phys_base;
+	void __iomem			*gpmc_cs_baseaddr;
+	void __iomem			*gpmc_baseaddr;
+};
+
+/**
+ * omap_nand_wp - This function enable or disable the Write Protect feature
+ * @mtd: MTD device structure
+ * @mode: WP ON/OFF
+ */
+static void omap_nand_wp(struct mtd_info *mtd, int mode)
+{
+	struct omap_nand_info *info = container_of(mtd,
+						struct omap_nand_info, mtd);
+
+	unsigned long config = __raw_readl(info->gpmc_baseaddr + GPMC_CONFIG);
+
+	if (mode)
+		config &= ~(NAND_WP_BIT);	/* WP is ON */
+	else
+		config |= (NAND_WP_BIT);	/* WP is OFF */
+
+	__raw_writel(config, (info->gpmc_baseaddr + GPMC_CONFIG));
+}
+
+/**
+ * omap_hwcontrol - hardware specific access to control-lines
+ * @mtd: MTD device structure
+ * @cmd: command to device
+ * @ctrl:
+ * NAND_NCE: bit 0 -> don't care
+ * NAND_CLE: bit 1 -> Command Latch
+ * NAND_ALE: bit 2 -> Address Latch
+ *
+ * NOTE: boards may use different bits for these!!
+ */
+static void omap_hwcontrol(struct mtd_info *mtd, int cmd, unsigned int ctrl)
+{
+	struct omap_nand_info *info = container_of(mtd,
+					struct omap_nand_info, mtd);
+	switch (ctrl) {
+	case NAND_CTRL_CHANGE | NAND_CTRL_CLE:
+		info->nand.IO_ADDR_W = info->gpmc_cs_baseaddr +
+						GPMC_CS_NAND_COMMAND;
+		info->nand.IO_ADDR_R = info->gpmc_cs_baseaddr +
+						GPMC_CS_NAND_DATA;
+		break;
+
+	case NAND_CTRL_CHANGE | NAND_CTRL_ALE:
+		info->nand.IO_ADDR_W = info->gpmc_cs_baseaddr +
+						GPMC_CS_NAND_ADDRESS;
+		info->nand.IO_ADDR_R = info->gpmc_cs_baseaddr +
+						GPMC_CS_NAND_DATA;
+		break;
+
+	case NAND_CTRL_CHANGE | NAND_NCE:
+		info->nand.IO_ADDR_W = info->gpmc_cs_baseaddr +
+						GPMC_CS_NAND_DATA;
+		info->nand.IO_ADDR_R = info->gpmc_cs_baseaddr +
+						GPMC_CS_NAND_DATA;
+		break;
+	}
+
+	if (cmd != NAND_CMD_NONE)
+		__raw_writeb(cmd, info->nand.IO_ADDR_W);
+}
+
+/**
+ * omap_read_buf16 - read data from NAND controller into buffer
+ * @mtd: MTD device structure
+ * @buf: buffer to store date
+ * @len: number of bytes to read
+ */
+static void omap_read_buf16(struct mtd_info *mtd, u_char *buf, int len)
+{
+	struct nand_chip *nand = mtd->priv;
+
+	__raw_readsw(nand->IO_ADDR_R, buf, len / 2);
+}
+
+/**
+ * omap_write_buf16 - write buffer to NAND controller
+ * @mtd: MTD device structure
+ * @buf: data buffer
+ * @len: number of bytes to write
+ */
+static void omap_write_buf16(struct mtd_info *mtd, const u_char * buf, int len)
+{
+	struct omap_nand_info *info = container_of(mtd,
+						struct omap_nand_info, mtd);
+	u16 *p = (u16 *) buf;
+
+	/* FIXME try bursts of writesw() or DMA ... */
+	len >>= 1;
+
+	while (len--) {
+		writew(*p++, info->nand.IO_ADDR_W);
+
+		while (GPMC_BUF_EMPTY == (readl(info->gpmc_baseaddr +
+						GPMC_STATUS) & GPMC_BUF_FULL))
+			;
+	}
+}
+/**
+ * omap_verify_buf - Verify chip data against buffer
+ * @mtd: MTD device structure
+ * @buf: buffer containing the data to compare
+ * @len: number of bytes to compare
+ */
+static int omap_verify_buf(struct mtd_info *mtd, const u_char * buf, int len)
+{
+	struct omap_nand_info *info = container_of(mtd, struct omap_nand_info,
+							mtd);
+	u16 *p = (u16 *) buf;
+
+	len >>= 1;
+	while (len--) {
+		if (*p++ != cpu_to_le16(readw(info->nand.IO_ADDR_R)))
+			return -EFAULT;
+	}
+
+	return 0;
+}
+
+#ifdef CONFIG_MTD_NAND_OMAP_HWECC
+/**
+ * omap_hwecc_init - Initialize the HW ECC for NAND flash in GPMC controller
+ * @mtd: MTD device structure
+ */
+static void omap_hwecc_init(struct mtd_info *mtd)
+{
+	struct omap_nand_info *info = container_of(mtd, struct omap_nand_info,
+							mtd);
+	struct nand_chip *chip = mtd->priv;
+	unsigned long val = 0x0;
+
+	/* Read from ECC Control Register */
+	val = __raw_readl(info->gpmc_baseaddr + GPMC_ECC_CONTROL);
+	/* Clear all ECC | Enable Reg1 */
+	val = ((0x00000001<<8) | 0x00000001);
+	__raw_writel(val, info->gpmc_baseaddr + GPMC_ECC_CONTROL);
+
+	/* Read from ECC Size Config Register */
+	val = __raw_readl(info->gpmc_baseaddr + GPMC_ECC_SIZE_CONFIG);
+	/* ECCSIZE1=512 | Select eccResultsize[0-3] */
+	val = ((((chip->ecc.size >> 1) - 1) << 22) | (0x0000000F));
+	__raw_writel(val, info->gpmc_baseaddr + GPMC_ECC_SIZE_CONFIG);
+}
+
+/**
+ * gen_true_ecc - This function will generate true ECC value
+ * @ecc_buf: buffer to store ecc code
+ *
+ * This generated true ECC value can be used when correcting
+ * data read from NAND flash memory core
+ */
+static void gen_true_ecc(u8 *ecc_buf)
+{
+	u32 tmp = ecc_buf[0] | (ecc_buf[1] << 16) |
+		((ecc_buf[2] & 0xF0) << 20) | ((ecc_buf[2] & 0x0F) << 8);
+
+	ecc_buf[0] = ~(P64o(tmp) | P64e(tmp) | P32o(tmp) | P32e(tmp) |
+			P16o(tmp) | P16e(tmp) | P8o(tmp) | P8e(tmp));
+	ecc_buf[1] = ~(P1024o(tmp) | P1024e(tmp) | P512o(tmp) | P512e(tmp) |
+			P256o(tmp) | P256e(tmp) | P128o(tmp) | P128e(tmp));
+	ecc_buf[2] = ~(P4o(tmp) | P4e(tmp) | P2o(tmp) | P2e(tmp) | P1o(tmp) |
+			P1e(tmp) | P2048o(tmp) | P2048e(tmp));
+}
+
+/**
+ * omap_compare_ecc - Detect (2 bits) and correct (1 bit) error in data
+ * @ecc_data1:  ecc code from nand spare area
+ * @ecc_data2:  ecc code from hardware register obtained from hardware ecc
+ * @page_data:  page data
+ *
+ * This function compares two ECC's and indicates if there is an error.
+ * If the error can be corrected it will be corrected to the buffer.
+ */
+static int omap_compare_ecc(u8 *ecc_data1,	/* read from NAND memory */
+			    u8 *ecc_data2,	/* read from register */
+			    u8 *page_data)
+{
+	uint	i;
+	u8	tmp0_bit[8], tmp1_bit[8], tmp2_bit[8];
+	u8	comp0_bit[8], comp1_bit[8], comp2_bit[8];
+	u8	ecc_bit[24];
+	u8	ecc_sum = 0;
+	u8	find_bit = 0;
+	uint	find_byte = 0;
+	int	isEccFF;
+
+	isEccFF = ((*(u32 *)ecc_data1 & 0xFFFFFF) == 0xFFFFFF);
+
+	gen_true_ecc(ecc_data1);
+	gen_true_ecc(ecc_data2);
+
+	for (i = 0; i <= 2; i++) {
+		*(ecc_data1 + i) = ~(*(ecc_data1 + i));
+		*(ecc_data2 + i) = ~(*(ecc_data2 + i));
+	}
+
+	for (i = 0; i < 8; i++) {
+		tmp0_bit[i]     = *ecc_data1 % 2;
+		*ecc_data1	= *ecc_data1 / 2;
+	}
+
+	for (i = 0; i < 8; i++) {
+		tmp1_bit[i]	 = *(ecc_data1 + 1) % 2;
+		*(ecc_data1 + 1) = *(ecc_data1 + 1) / 2;
+	}
+
+	for (i = 0; i < 8; i++) {
+		tmp2_bit[i]	 = *(ecc_data1 + 2) % 2;
+		*(ecc_data1 + 2) = *(ecc_data1 + 2) / 2;
+	}
+
+	for (i = 0; i < 8; i++) {
+		comp0_bit[i]     = *ecc_data2 % 2;
+		*ecc_data2       = *ecc_data2 / 2;
+	}
+
+	for (i = 0; i < 8; i++) {
+		comp1_bit[i]     = *(ecc_data2 + 1) % 2;
+		*(ecc_data2 + 1) = *(ecc_data2 + 1) / 2;
+	}
+
+	for (i = 0; i < 8; i++) {
+		comp2_bit[i]     = *(ecc_data2 + 2) % 2;
+		*(ecc_data2 + 2) = *(ecc_data2 + 2) / 2;
+	}
+
+	for (i = 0; i < 6; i++)
+		ecc_bit[i] = tmp2_bit[i + 2] ^ comp2_bit[i + 2];
+
+	for (i = 0; i < 8; i++)
+		ecc_bit[i + 6] = tmp0_bit[i] ^ comp0_bit[i];
+
+	for (i = 0; i < 8; i++)
+		ecc_bit[i + 14] = tmp1_bit[i] ^ comp1_bit[i];
+
+	ecc_bit[22] = tmp2_bit[0] ^ comp2_bit[0];
+	ecc_bit[23] = tmp2_bit[1] ^ comp2_bit[1];
+
+	for (i = 0; i < 24; i++)
+		ecc_sum += ecc_bit[i];
+
+	switch (ecc_sum) {
+	case 0:
+		/* Not reached because this function is not called if
+		 *  ECC values are equal
+		 */
+		return 0;
+
+	case 1:
+		/* Uncorrectable error */
+		DEBUG(MTD_DEBUG_LEVEL0, "ECC UNCORRECTED_ERROR 1\n");
+		return -1;
+
+	case 11:
+		/* UN-Correctable error */
+		DEBUG(MTD_DEBUG_LEVEL0, "ECC UNCORRECTED_ERROR B\n");
+		return -1;
+
+	case 12:
+		/* Correctable error */
+		find_byte = (ecc_bit[23] << 8) +
+			    (ecc_bit[21] << 7) +
+			    (ecc_bit[19] << 6) +
+			    (ecc_bit[17] << 5) +
+			    (ecc_bit[15] << 4) +
+			    (ecc_bit[13] << 3) +
+			    (ecc_bit[11] << 2) +
+			    (ecc_bit[9]  << 1) +
+			    ecc_bit[7];
+
+		find_bit = (ecc_bit[5] << 2) + (ecc_bit[3] << 1) + ecc_bit[1];
+
+		DEBUG(MTD_DEBUG_LEVEL0, "Correcting single bit ECC error at "
+				"offset: %d, bit: %d\n", find_byte, find_bit);
+
+		page_data[find_byte] ^= (1 << find_bit);
+
+		return 0;
+	default:
+		if (isEccFF) {
+			if (ecc_data2[0] == 0 &&
+			    ecc_data2[1] == 0 &&
+			    ecc_data2[2] == 0)
+				return 0;
+		}
+		DEBUG(MTD_DEBUG_LEVEL0, "UNCORRECTED_ERROR default\n");
+		return -1;
+	}
+}
+
+/**
+ * omap_correct_data - Compares the ECC read with HW generated ECC
+ * @mtd: MTD device structure
+ * @dat: page data
+ * @read_ecc: ecc read from nand flash
+ * @calc_ecc: ecc read from HW ECC registers
+ *
+ * Compares the ecc read from nand spare area with ECC registers values
+ * and if ECC's mismached, it will call 'omap_compare_ecc' for error detection
+ * and correction.
+ */
+static int omap_correct_data(struct mtd_info *mtd, u_char *dat,
+				u_char *read_ecc, u_char *calc_ecc)
+{
+	struct omap_nand_info *info = container_of(mtd, struct omap_nand_info,
+							mtd);
+	int blockCnt = 0, i = 0, ret = 0;
+
+	/* Ex NAND_ECC_HW12_2048 */
+	if ((info->nand.ecc.mode == NAND_ECC_HW) &&
+			(info->nand.ecc.size  == 2048))
+		blockCnt = 4;
+	else
+		blockCnt = 1;
+
+	for (i = 0; i < blockCnt; i++) {
+		if (memcmp(read_ecc, calc_ecc, 3) != 0) {
+			ret = omap_compare_ecc(read_ecc, calc_ecc, dat);
+			if (ret < 0)
+				return ret;
+		}
+		read_ecc += 3;
+		calc_ecc += 3;
+		dat      += 512;
+	}
+	return 0;
+}
+
+/**
+ * omap_calcuate_ecc - Generate non-inverted ECC bytes.
+ * @mtd: MTD device structure
+ * @dat: The pointer to data on which ecc is computed
+ * @ecc_code: The ecc_code buffer
+ *
+ * Using noninverted ECC can be considered ugly since writing a blank
+ * page ie. padding will clear the ECC bytes. This is no problem as long
+ * nobody is trying to write data on the seemingly unused page. Reading
+ * an erased page will produce an ECC mismatch between generated and read
+ * ECC bytes that has to be dealt with separately.
+ */
+static int omap_calculate_ecc(struct mtd_info *mtd, const u_char *dat,
+				u_char *ecc_code)
+{
+	struct omap_nand_info *info = container_of(mtd, struct omap_nand_info,
+							mtd);
+	unsigned long val = 0x0;
+	unsigned long reg;
+
+	/* Start Reading from HW ECC1_Result = 0x200 */
+	reg = (unsigned long)(info->gpmc_baseaddr + GPMC_ECC1_RESULT);
+	val = __raw_readl(reg);
+	*ecc_code++ = val;          /* P128e, ..., P1e */
+	*ecc_code++ = val >> 16;    /* P128o, ..., P1o */
+	/* P2048o, P1024o, P512o, P256o, P2048e, P1024e, P512e, P256e */
+	*ecc_code++ = ((val >> 8) & 0x0f) | ((val >> 20) & 0xf0);
+	reg += 4;
+
+	return 0;
+}
+
+/**
+ * omap_enable_hwecc - This function enables the hardware ecc functionality
+ * @mtd: MTD device structure
+ * @mode: Read/Write mode
+ */
+static void omap_enable_hwecc(struct mtd_info *mtd, int mode)
+{
+	struct omap_nand_info *info = container_of(mtd, struct omap_nand_info,
+							mtd);
+	struct nand_chip *chip = mtd->priv;
+	unsigned int dev_width = (chip->options & NAND_BUSWIDTH_16) ? 1 : 0;
+	unsigned long val = __raw_readl(info->gpmc_baseaddr + GPMC_ECC_CONFIG);
+
+	switch (mode) {
+	case NAND_ECC_READ:
+		__raw_writel(0x101, info->gpmc_baseaddr + GPMC_ECC_CONTROL);
+		/* (ECC 16 or 8 bit col) | ( CS  )  | ECC Enable */
+		val = (dev_width << 7) | (info->gpmc_cs << 1) | (0x1);
+		break;
+	case NAND_ECC_READSYN:
+		 __raw_writel(0x100, info->gpmc_baseaddr + GPMC_ECC_CONTROL);
+		/* (ECC 16 or 8 bit col) | ( CS  )  | ECC Enable */
+		val = (dev_width << 7) | (info->gpmc_cs << 1) | (0x1);
+		break;
+	case NAND_ECC_WRITE:
+		__raw_writel(0x101, info->gpmc_baseaddr + GPMC_ECC_CONTROL);
+		/* (ECC 16 or 8 bit col) | ( CS  )  | ECC Enable */
+		val = (dev_width << 7) | (info->gpmc_cs << 1) | (0x1);
+		break;
+	default:
+		DEBUG(MTD_DEBUG_LEVEL0, "Error: Unrecognized Mode[%d]!\n",
+					mode);
+		break;
+	}
+
+	__raw_writel(val, info->gpmc_baseaddr + GPMC_ECC_CONFIG);
+}
+#endif
+
+/**
+ * omap_wait - wait until the command is done
+ * @mtd: MTD device structure
+ * @chip: NAND Chip structure
+ *
+ * Wait function is called during Program and erase operations and
+ * the way it is called from MTD layer, we should wait till the NAND
+ * chip is ready after the programming/erase operation has completed.
+ *
+ * Erase can take up to 400ms and program up to 20ms according to
+ * general NAND and SmartMedia specs
+ */
+static int omap_wait(struct mtd_info *mtd, struct nand_chip *chip)
+{
+	struct nand_chip *this = mtd->priv;
+	struct omap_nand_info *info = container_of(mtd, struct omap_nand_info,
+							mtd);
+	unsigned long timeo = jiffies;
+	int status, state = this->state;
+
+	if (state == FL_ERASING)
+		timeo += (HZ * 400) / 1000;
+	else
+		timeo += (HZ * 20) / 1000;
+
+	this->IO_ADDR_W = (void *) info->gpmc_cs_baseaddr +
+						GPMC_CS_NAND_COMMAND;
+	this->IO_ADDR_R = (void *) info->gpmc_cs_baseaddr + GPMC_CS_NAND_DATA;
+
+	__raw_writeb(NAND_CMD_STATUS & 0xFF, this->IO_ADDR_W);
+
+	while (time_before(jiffies, timeo)) {
+		status = __raw_readb(this->IO_ADDR_R);
+		if (!(status & 0x40))
+			break;
+	}
+	return status;
+}
+
+/**
+ * omap_dev_ready - calls the platform specific dev_ready function
+ * @mtd: MTD device structure
+ */
+static int omap_dev_ready(struct mtd_info *mtd)
+{
+	struct omap_nand_info *info = container_of(mtd, struct omap_nand_info,
+							mtd);
+	unsigned int val = __raw_readl(info->gpmc_baseaddr + GPMC_IRQ_STATUS);
+
+	if ((val & 0x100) == 0x100) {
+		/* Clear IRQ Interrupt */
+		val |= 0x100;
+		val &= ~(0x0);
+		__raw_writel(val, info->gpmc_baseaddr + GPMC_IRQ_STATUS);
+	} else {
+		unsigned int cnt = 0;
+		while (cnt++ < 0x1FF) {
+			if  ((val & 0x100) == 0x100)
+				return 0;
+			val = __raw_readl(info->gpmc_baseaddr +
+							GPMC_IRQ_STATUS);
+		}
+	}
+
+	return 1;
+}
+
+static int __devinit omap_nand_probe(struct platform_device *pdev)
+{
+	struct omap_nand_info		*info;
+	struct omap_nand_platform_data	*pdata;
+	int				err;
+	unsigned long 			val;
+
+
+	pdata = pdev->dev.platform_data;
+	if (pdata == NULL) {
+		dev_err(&pdev->dev, "platform data missing\n");
+		return -ENODEV;
+	}
+
+	info = kzalloc(sizeof(struct omap_nand_info), GFP_KERNEL);
+	if (!info)
+		return -ENOMEM;
+
+	platform_set_drvdata(pdev, info);
+
+	spin_lock_init(&info->controller.lock);
+	init_waitqueue_head(&info->controller.wq);
+
+	info->pdev = pdev;
+
+	info->gpmc_cs		= pdata->cs;
+	info->gpmc_baseaddr	= pdata->gpmc_baseaddr;
+	info->gpmc_cs_baseaddr	= pdata->gpmc_cs_baseaddr;
+
+	info->mtd.priv		= &info->nand;
+	info->mtd.name		= dev_name(&pdev->dev);
+	info->mtd.owner		= THIS_MODULE;
+
+	err = gpmc_cs_request(info->gpmc_cs, NAND_IO_SIZE, &info->phys_base);
+	if (err < 0) {
+		dev_err(&pdev->dev, "Cannot request GPMC CS\n");
+		goto out_free_info;
+	}
+
+	/* Enable RD PIN Monitoring Reg */
+	if (pdata->dev_ready) {
+		val  = gpmc_cs_read_reg(info->gpmc_cs, GPMC_CS_CONFIG1);
+		val |= WR_RD_PIN_MONITORING;
+		gpmc_cs_write_reg(info->gpmc_cs, GPMC_CS_CONFIG1, val);
+	}
+
+	val  = gpmc_cs_read_reg(info->gpmc_cs, GPMC_CS_CONFIG7);
+	val &= ~(0xf << 8);
+	val |=  (0xc & 0xf) << 8;
+	gpmc_cs_write_reg(info->gpmc_cs, GPMC_CS_CONFIG7, val);
+
+	/* NAND write protect off */
+	omap_nand_wp(&info->mtd, NAND_WP_OFF);
+
+	if (!request_mem_region(info->phys_base, NAND_IO_SIZE,
+				pdev->dev.driver->name)) {
+		err = -EBUSY;
+		goto out_free_cs;
+	}
+
+	info->nand.IO_ADDR_R = ioremap(info->phys_base, NAND_IO_SIZE);
+	if (!info->nand.IO_ADDR_R) {
+		err = -ENOMEM;
+		goto out_release_mem_region;
+	}
+	info->nand.controller = &info->controller;
+
+	info->nand.IO_ADDR_W = info->nand.IO_ADDR_R;
+	info->nand.cmd_ctrl  = omap_hwcontrol;
+
+	/* REVISIT:  only supports 16-bit NAND flash */
+
+	info->nand.read_buf   = omap_read_buf16;
+	info->nand.write_buf  = omap_write_buf16;
+	info->nand.verify_buf = omap_verify_buf;
+
+	/*
+	 * If RDY/BSY line is connected to OMAP then use the omap ready
+	 * funcrtion and the generic nand_wait function which reads the status
+	 * register after monitoring the RDY/BSY line.Otherwise use a standard
+	 * chip delay which is slightly more than tR (AC Timing) of the NAND
+	 * device and read status register until you get a failure or success
+	 */
+	if (pdata->dev_ready) {
+		info->nand.dev_ready = omap_dev_ready;
+		info->nand.chip_delay = 0;
+	} else {
+		info->nand.waitfunc = omap_wait;
+		info->nand.chip_delay = 50;
+	}
+
+	info->nand.options  |= NAND_SKIP_BBTSCAN;
+	if ((gpmc_cs_read_reg(info->gpmc_cs, GPMC_CS_CONFIG1) & 0x3000)
+								== 0x1000)
+		info->nand.options  |= NAND_BUSWIDTH_16;
+
+#ifdef CONFIG_MTD_NAND_OMAP_HWECC
+	info->nand.ecc.bytes		= 3;
+	info->nand.ecc.size		= 512;
+	info->nand.ecc.calculate	= omap_calculate_ecc;
+	info->nand.ecc.hwctl		= omap_enable_hwecc;
+	info->nand.ecc.correct		= omap_correct_data;
+	info->nand.ecc.mode		= NAND_ECC_HW;
+
+	/* init HW ECC */
+	omap_hwecc_init(&info->mtd);
+#else
+	info->nand.ecc.mode = NAND_ECC_SOFT;
+#endif
+
+	/* DIP switches on some boards change between 8 and 16 bit
+	 * bus widths for flash.  Try the other width if the first try fails.
+	 */
+	if (nand_scan(&info->mtd, 1)) {
+		info->nand.options ^= NAND_BUSWIDTH_16;
+		if (nand_scan(&info->mtd, 1)) {
+			err = -ENXIO;
+			goto out_release_mem_region;
+		}
+	}
+
+#ifdef CONFIG_MTD_PARTITIONS
+	err = parse_mtd_partitions(&info->mtd, part_probes, &info->parts, 0);
+	if (err > 0)
+		add_mtd_partitions(&info->mtd, info->parts, err);
+	else if (pdata->parts)
+		add_mtd_partitions(&info->mtd, pdata->parts, pdata->nr_parts);
+	else
+#endif
+		add_mtd_device(&info->mtd);
+
+	platform_set_drvdata(pdev, &info->mtd);
+
+	return 0;
+
+out_release_mem_region:
+	release_mem_region(info->phys_base, NAND_IO_SIZE);
+out_free_cs:
+	gpmc_cs_free(info->gpmc_cs);
+out_free_info:
+	kfree(info);
+
+	return err;
+}
+
+static int omap_nand_remove(struct platform_device *pdev)
+{
+	struct mtd_info *mtd = platform_get_drvdata(pdev);
+	struct omap_nand_info *info = mtd->priv;
+
+	platform_set_drvdata(pdev, NULL);
+	/* Release NAND device, its internal structures and partitions */
+	nand_release(&info->mtd);
+	iounmap(info->nand.IO_ADDR_R);
+	kfree(&info->mtd);
+	return 0;
+}
+
+static struct platform_driver omap_nand_driver = {
+	.probe		= omap_nand_probe,
+	.remove		= omap_nand_remove,
+	.driver		= {
+		.name	= DRIVER_NAME,
+		.owner	= THIS_MODULE,
+	},
+};
+
+static int __init omap_nand_init(void)
+{
+	printk(KERN_INFO "%s driver initializing\n", DRIVER_NAME);
+	return platform_driver_register(&omap_nand_driver);
+}
+
+static void __exit omap_nand_exit(void)
+{
+	platform_driver_unregister(&omap_nand_driver);
+}
+
+module_init(omap_nand_init);
+module_exit(omap_nand_exit);
+
+MODULE_ALIAS(DRIVER_NAME);
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Glue layer for NAND flash on TI OMAP boards");
-- 
cgit v1.2.3-59-g8ed1b


From 5988af2319781bc8e0ce418affec4e09cfa77907 Mon Sep 17 00:00:00 2001
From: Rohit Hagargundgi <h.rohit@samsung.com>
Date: Tue, 12 May 2009 13:46:57 -0700
Subject: mtd: Flex-OneNAND support

Add support for Samsung Flex-OneNAND devices.

Flex-OneNAND combines SLC and MLC technologies into a single device.
SLC area provides increased reliability and speed, suitable for storing
code such as bootloader, kernel and root file system.  MLC area
provides high density and is suitable for storing user data.

SLC and MLC regions can be configured through kernel parameter.

[akpm@linux-foundation.org: export flexoand_region and onenand_addr]
Signed-off-by: Rohit Hagargundgi <h.rohit@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Vishak G <vishak.g@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 Documentation/kernel-parameters.txt |  10 +
 drivers/mtd/onenand/onenand_base.c  | 857 ++++++++++++++++++++++++++++++++----
 drivers/mtd/onenand/onenand_bbt.c   |  14 +-
 drivers/mtd/onenand/onenand_sim.c   |  81 +++-
 include/linux/mtd/onenand.h         |  18 +
 include/linux/mtd/onenand_regs.h    |  20 +-
 6 files changed, 913 insertions(+), 87 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index e87bdbfbcc75..12df135f8af9 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1380,6 +1380,16 @@ and is between 256 and 4096 characters. It is defined in the file
 	mtdparts=	[MTD]
 			See drivers/mtd/cmdlinepart.c.
 
+	onenand.bdry=	[HW,MTD] Flex-OneNAND Boundary Configuration
+
+			Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock]
+
+			boundary - index of last SLC block on Flex-OneNAND.
+				   The remaining blocks are configured as MLC blocks.
+			lock	 - Configure if Flex-OneNAND boundary should be locked.
+				   Once locked, the boundary cannot be changed.
+				   1 indicates lock status, 0 indicates unlock status.
+
 	mtdset=		[ARM]
 			ARM/S3C2412 JIVE boot control
 
diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
index 2346857a275d..8d4c9c253732 100644
--- a/drivers/mtd/onenand/onenand_base.c
+++ b/drivers/mtd/onenand/onenand_base.c
@@ -9,6 +9,10 @@
  *	auto-placement support, read-while load support, various fixes
  *	Copyright (C) Nokia Corporation, 2007
  *
+ *	Vishak G <vishak.g at samsung.com>, Rohit Hagargundgi <h.rohit at samsung.com>
+ *	Flex-OneNAND support
+ *	Copyright (C) Samsung Electronics, 2008
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
@@ -27,6 +31,30 @@
 
 #include <asm/io.h>
 
+/* Default Flex-OneNAND boundary and lock respectively */
+static int flex_bdry[MAX_DIES * 2] = { -1, 0, -1, 0 };
+
+/**
+ *  onenand_oob_128 - oob info for Flex-Onenand with 4KB page
+ *  For now, we expose only 64 out of 80 ecc bytes
+ */
+static struct nand_ecclayout onenand_oob_128 = {
+	.eccbytes	= 64,
+	.eccpos		= {
+		6, 7, 8, 9, 10, 11, 12, 13, 14, 15,
+		22, 23, 24, 25, 26, 27, 28, 29, 30, 31,
+		38, 39, 40, 41, 42, 43, 44, 45, 46, 47,
+		54, 55, 56, 57, 58, 59, 60, 61, 62, 63,
+		70, 71, 72, 73, 74, 75, 76, 77, 78, 79,
+		86, 87, 88, 89, 90, 91, 92, 93, 94, 95,
+		102, 103, 104, 105
+		},
+	.oobfree	= {
+		{2, 4}, {18, 4}, {34, 4}, {50, 4},
+		{66, 4}, {82, 4}, {98, 4}, {114, 4}
+	}
+};
+
 /**
  * onenand_oob_64 - oob info for large (2KB) page
  */
@@ -65,6 +93,14 @@ static const unsigned char ffchars[] = {
 	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,	/* 48 */
 	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
 	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,	/* 64 */
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,	/* 80 */
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,	/* 96 */
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,	/* 112 */
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,	/* 128 */
 };
 
 /**
@@ -170,6 +206,70 @@ static int onenand_buffer_address(int dataram1, int sectors, int count)
 	return ((bsa << ONENAND_BSA_SHIFT) | bsc);
 }
 
+/**
+ * flexonenand_block- For given address return block number
+ * @param this         - OneNAND device structure
+ * @param addr		- Address for which block number is needed
+ */
+static unsigned flexonenand_block(struct onenand_chip *this, loff_t addr)
+{
+	unsigned boundary, blk, die = 0;
+
+	if (ONENAND_IS_DDP(this) && addr >= this->diesize[0]) {
+		die = 1;
+		addr -= this->diesize[0];
+	}
+
+	boundary = this->boundary[die];
+
+	blk = addr >> (this->erase_shift - 1);
+	if (blk > boundary)
+		blk = (blk + boundary + 1) >> 1;
+
+	blk += die ? this->density_mask : 0;
+	return blk;
+}
+
+inline unsigned onenand_block(struct onenand_chip *this, loff_t addr)
+{
+	if (!FLEXONENAND(this))
+		return addr >> this->erase_shift;
+	return flexonenand_block(this, addr);
+}
+
+/**
+ * flexonenand_addr - Return address of the block
+ * @this:		OneNAND device structure
+ * @block:		Block number on Flex-OneNAND
+ *
+ * Return address of the block
+ */
+static loff_t flexonenand_addr(struct onenand_chip *this, int block)
+{
+	loff_t ofs = 0;
+	int die = 0, boundary;
+
+	if (ONENAND_IS_DDP(this) && block >= this->density_mask) {
+		block -= this->density_mask;
+		die = 1;
+		ofs = this->diesize[0];
+	}
+
+	boundary = this->boundary[die];
+	ofs += (loff_t)block << (this->erase_shift - 1);
+	if (block > (boundary + 1))
+		ofs += (loff_t)(block - boundary - 1) << (this->erase_shift - 1);
+	return ofs;
+}
+
+loff_t onenand_addr(struct onenand_chip *this, int block)
+{
+	if (!FLEXONENAND(this))
+		return (loff_t)block << this->erase_shift;
+	return flexonenand_addr(this, block);
+}
+EXPORT_SYMBOL(onenand_addr);
+
 /**
  * onenand_get_density - [DEFAULT] Get OneNAND density
  * @param dev_id	OneNAND device ID
@@ -182,6 +282,22 @@ static inline int onenand_get_density(int dev_id)
 	return (density & ONENAND_DEVICE_DENSITY_MASK);
 }
 
+/**
+ * flexonenand_region - [Flex-OneNAND] Return erase region of addr
+ * @param mtd		MTD device structure
+ * @param addr		address whose erase region needs to be identified
+ */
+int flexonenand_region(struct mtd_info *mtd, loff_t addr)
+{
+	int i;
+
+	for (i = 0; i < mtd->numeraseregions; i++)
+		if (addr < mtd->eraseregions[i].offset)
+			break;
+	return i - 1;
+}
+EXPORT_SYMBOL(flexonenand_region);
+
 /**
  * onenand_command - [DEFAULT] Send command to OneNAND device
  * @param mtd		MTD device structure
@@ -207,16 +323,28 @@ static int onenand_command(struct mtd_info *mtd, int cmd, loff_t addr, size_t le
 		page = -1;
 		break;
 
+	case FLEXONENAND_CMD_PI_ACCESS:
+		/* addr contains die index */
+		block = addr * this->density_mask;
+		page = -1;
+		break;
+
 	case ONENAND_CMD_ERASE:
 	case ONENAND_CMD_BUFFERRAM:
 	case ONENAND_CMD_OTP_ACCESS:
-		block = (int) (addr >> this->erase_shift);
+		block = onenand_block(this, addr);
 		page = -1;
 		break;
 
+	case FLEXONENAND_CMD_READ_PI:
+		cmd = ONENAND_CMD_READ;
+		block = addr * this->density_mask;
+		page = 0;
+		break;
+
 	default:
-		block = (int) (addr >> this->erase_shift);
-		page = (int) (addr >> this->page_shift);
+		block = onenand_block(this, addr);
+		page = (int) (addr - onenand_addr(this, block)) >> this->page_shift;
 
 		if (ONENAND_IS_2PLANE(this)) {
 			/* Make the even block number */
@@ -236,7 +364,7 @@ static int onenand_command(struct mtd_info *mtd, int cmd, loff_t addr, size_t le
 		value = onenand_bufferram_address(this, block);
 		this->write_word(value, this->base + ONENAND_REG_START_ADDRESS2);
 
-		if (ONENAND_IS_2PLANE(this))
+		if (ONENAND_IS_MLC(this) || ONENAND_IS_2PLANE(this))
 			/* It is always BufferRAM0 */
 			ONENAND_SET_BUFFERRAM0(this);
 		else
@@ -258,13 +386,18 @@ static int onenand_command(struct mtd_info *mtd, int cmd, loff_t addr, size_t le
 
 	if (page != -1) {
 		/* Now we use page size operation */
-		int sectors = 4, count = 4;
+		int sectors = 0, count = 0;
 		int dataram;
 
 		switch (cmd) {
+		case FLEXONENAND_CMD_RECOVER_LSB:
 		case ONENAND_CMD_READ:
 		case ONENAND_CMD_READOOB:
-			dataram = ONENAND_SET_NEXT_BUFFERRAM(this);
+			if (ONENAND_IS_MLC(this))
+				/* It is always BufferRAM0 */
+				dataram = ONENAND_SET_BUFFERRAM0(this);
+			else
+				dataram = ONENAND_SET_NEXT_BUFFERRAM(this);
 			break;
 
 		default:
@@ -292,6 +425,30 @@ static int onenand_command(struct mtd_info *mtd, int cmd, loff_t addr, size_t le
 	return 0;
 }
 
+/**
+ * onenand_read_ecc - return ecc status
+ * @param this		onenand chip structure
+ */
+static inline int onenand_read_ecc(struct onenand_chip *this)
+{
+	int ecc, i, result = 0;
+
+	if (!FLEXONENAND(this))
+		return this->read_word(this->base + ONENAND_REG_ECC_STATUS);
+
+	for (i = 0; i < 4; i++) {
+		ecc = this->read_word(this->base + ONENAND_REG_ECC_STATUS + i);
+		if (likely(!ecc))
+			continue;
+		if (ecc & FLEXONENAND_UNCORRECTABLE_ERROR)
+			return ONENAND_ECC_2BIT_ALL;
+		else
+			result = ONENAND_ECC_1BIT_ALL;
+	}
+
+	return result;
+}
+
 /**
  * onenand_wait - [DEFAULT] wait until the command is done
  * @param mtd		MTD device structure
@@ -331,14 +488,14 @@ static int onenand_wait(struct mtd_info *mtd, int state)
 	 * power off recovery (POR) test, it should read ECC status first
 	 */
 	if (interrupt & ONENAND_INT_READ) {
-		int ecc = this->read_word(this->base + ONENAND_REG_ECC_STATUS);
+		int ecc = onenand_read_ecc(this);
 		if (ecc) {
 			if (ecc & ONENAND_ECC_2BIT_ALL) {
 				printk(KERN_ERR "onenand_wait: ECC error = 0x%04x\n", ecc);
 				mtd->ecc_stats.failed++;
 				return -EBADMSG;
 			} else if (ecc & ONENAND_ECC_1BIT_ALL) {
-				printk(KERN_INFO "onenand_wait: correctable ECC error = 0x%04x\n", ecc);
+				printk(KERN_DEBUG "onenand_wait: correctable ECC error = 0x%04x\n", ecc);
 				mtd->ecc_stats.corrected++;
 			}
 		}
@@ -656,7 +813,7 @@ static int onenand_check_bufferram(struct mtd_info *mtd, loff_t addr)
 
 	if (found && ONENAND_IS_DDP(this)) {
 		/* Select DataRAM for DDP */
-		int block = (int) (addr >> this->erase_shift);
+		int block = onenand_block(this, addr);
 		int value = onenand_bufferram_address(this, block);
 		this->write_word(value, this->base + ONENAND_REG_START_ADDRESS2);
 	}
@@ -815,6 +972,149 @@ static int onenand_transfer_auto_oob(struct mtd_info *mtd, uint8_t *buf, int col
 	return 0;
 }
 
+/**
+ * onenand_recover_lsb - [Flex-OneNAND] Recover LSB page data
+ * @param mtd		MTD device structure
+ * @param addr		address to recover
+ * @param status	return value from onenand_wait / onenand_bbt_wait
+ *
+ * MLC NAND Flash cell has paired pages - LSB page and MSB page. LSB page has
+ * lower page address and MSB page has higher page address in paired pages.
+ * If power off occurs during MSB page program, the paired LSB page data can
+ * become corrupt. LSB page recovery read is a way to read LSB page though page
+ * data are corrupted. When uncorrectable error occurs as a result of LSB page
+ * read after power up, issue LSB page recovery read.
+ */
+static int onenand_recover_lsb(struct mtd_info *mtd, loff_t addr, int status)
+{
+	struct onenand_chip *this = mtd->priv;
+	int i;
+
+	/* Recovery is only for Flex-OneNAND */
+	if (!FLEXONENAND(this))
+		return status;
+
+	/* check if we failed due to uncorrectable error */
+	if (status != -EBADMSG && status != ONENAND_BBT_READ_ECC_ERROR)
+		return status;
+
+	/* check if address lies in MLC region */
+	i = flexonenand_region(mtd, addr);
+	if (mtd->eraseregions[i].erasesize < (1 << this->erase_shift))
+		return status;
+
+	/* We are attempting to reread, so decrement stats.failed
+	 * which was incremented by onenand_wait due to read failure
+	 */
+	printk(KERN_INFO "onenand_recover_lsb: Attempting to recover from uncorrectable read\n");
+	mtd->ecc_stats.failed--;
+
+	/* Issue the LSB page recovery command */
+	this->command(mtd, FLEXONENAND_CMD_RECOVER_LSB, addr, this->writesize);
+	return this->wait(mtd, FL_READING);
+}
+
+/**
+ * onenand_mlc_read_ops_nolock - MLC OneNAND read main and/or out-of-band
+ * @param mtd		MTD device structure
+ * @param from		offset to read from
+ * @param ops:		oob operation description structure
+ *
+ * MLC OneNAND / Flex-OneNAND has 4KB page size and 4KB dataram.
+ * So, read-while-load is not present.
+ */
+static int onenand_mlc_read_ops_nolock(struct mtd_info *mtd, loff_t from,
+				struct mtd_oob_ops *ops)
+{
+	struct onenand_chip *this = mtd->priv;
+	struct mtd_ecc_stats stats;
+	size_t len = ops->len;
+	size_t ooblen = ops->ooblen;
+	u_char *buf = ops->datbuf;
+	u_char *oobbuf = ops->oobbuf;
+	int read = 0, column, thislen;
+	int oobread = 0, oobcolumn, thisooblen, oobsize;
+	int ret = 0;
+	int writesize = this->writesize;
+
+	DEBUG(MTD_DEBUG_LEVEL3, "onenand_mlc_read_ops_nolock: from = 0x%08x, len = %i\n", (unsigned int) from, (int) len);
+
+	if (ops->mode == MTD_OOB_AUTO)
+		oobsize = this->ecclayout->oobavail;
+	else
+		oobsize = mtd->oobsize;
+
+	oobcolumn = from & (mtd->oobsize - 1);
+
+	/* Do not allow reads past end of device */
+	if (from + len > mtd->size) {
+		printk(KERN_ERR "onenand_mlc_read_ops_nolock: Attempt read beyond end of device\n");
+		ops->retlen = 0;
+		ops->oobretlen = 0;
+		return -EINVAL;
+	}
+
+	stats = mtd->ecc_stats;
+
+	while (read < len) {
+		cond_resched();
+
+		thislen = min_t(int, writesize, len - read);
+
+		column = from & (writesize - 1);
+		if (column + thislen > writesize)
+			thislen = writesize - column;
+
+		if (!onenand_check_bufferram(mtd, from)) {
+			this->command(mtd, ONENAND_CMD_READ, from, writesize);
+
+			ret = this->wait(mtd, FL_READING);
+			if (unlikely(ret))
+				ret = onenand_recover_lsb(mtd, from, ret);
+			onenand_update_bufferram(mtd, from, !ret);
+			if (ret == -EBADMSG)
+				ret = 0;
+		}
+
+		this->read_bufferram(mtd, ONENAND_DATARAM, buf, column, thislen);
+		if (oobbuf) {
+			thisooblen = oobsize - oobcolumn;
+			thisooblen = min_t(int, thisooblen, ooblen - oobread);
+
+			if (ops->mode == MTD_OOB_AUTO)
+				onenand_transfer_auto_oob(mtd, oobbuf, oobcolumn, thisooblen);
+			else
+				this->read_bufferram(mtd, ONENAND_SPARERAM, oobbuf, oobcolumn, thisooblen);
+			oobread += thisooblen;
+			oobbuf += thisooblen;
+			oobcolumn = 0;
+		}
+
+		read += thislen;
+		if (read == len)
+			break;
+
+		from += thislen;
+		buf += thislen;
+	}
+
+	/*
+	 * Return success, if no ECC failures, else -EBADMSG
+	 * fs driver will take care of that, because
+	 * retlen == desired len and result == -EBADMSG
+	 */
+	ops->retlen = read;
+	ops->oobretlen = oobread;
+
+	if (ret)
+		return ret;
+
+	if (mtd->ecc_stats.failed - stats.failed)
+		return -EBADMSG;
+
+	return mtd->ecc_stats.corrected - stats.corrected ? -EUCLEAN : 0;
+}
+
 /**
  * onenand_read_ops_nolock - [OneNAND Interface] OneNAND read main and/or out-of-band
  * @param mtd		MTD device structure
@@ -962,7 +1262,7 @@ static int onenand_read_oob_nolock(struct mtd_info *mtd, loff_t from,
 	size_t len = ops->ooblen;
 	mtd_oob_mode_t mode = ops->mode;
 	u_char *buf = ops->oobbuf;
-	int ret = 0;
+	int ret = 0, readcmd;
 
 	from += ops->ooboffs;
 
@@ -993,17 +1293,22 @@ static int onenand_read_oob_nolock(struct mtd_info *mtd, loff_t from,
 
 	stats = mtd->ecc_stats;
 
+	readcmd = ONENAND_IS_MLC(this) ? ONENAND_CMD_READ : ONENAND_CMD_READOOB;
+
 	while (read < len) {
 		cond_resched();
 
 		thislen = oobsize - column;
 		thislen = min_t(int, thislen, len);
 
-		this->command(mtd, ONENAND_CMD_READOOB, from, mtd->oobsize);
+		this->command(mtd, readcmd, from, mtd->oobsize);
 
 		onenand_update_bufferram(mtd, from, 0);
 
 		ret = this->wait(mtd, FL_READING);
+		if (unlikely(ret))
+			ret = onenand_recover_lsb(mtd, from, ret);
+
 		if (ret && ret != -EBADMSG) {
 			printk(KERN_ERR "onenand_read_oob_nolock: read failed = 0x%x\n", ret);
 			break;
@@ -1053,6 +1358,7 @@ static int onenand_read_oob_nolock(struct mtd_info *mtd, loff_t from,
 static int onenand_read(struct mtd_info *mtd, loff_t from, size_t len,
 	size_t *retlen, u_char *buf)
 {
+	struct onenand_chip *this = mtd->priv;
 	struct mtd_oob_ops ops = {
 		.len	= len,
 		.ooblen	= 0,
@@ -1062,7 +1368,9 @@ static int onenand_read(struct mtd_info *mtd, loff_t from, size_t len,
 	int ret;
 
 	onenand_get_device(mtd, FL_READING);
-	ret = onenand_read_ops_nolock(mtd, from, &ops);
+	ret = ONENAND_IS_MLC(this) ?
+		onenand_mlc_read_ops_nolock(mtd, from, &ops) :
+		onenand_read_ops_nolock(mtd, from, &ops);
 	onenand_release_device(mtd);
 
 	*retlen = ops.retlen;
@@ -1080,6 +1388,7 @@ static int onenand_read(struct mtd_info *mtd, loff_t from, size_t len,
 static int onenand_read_oob(struct mtd_info *mtd, loff_t from,
 			    struct mtd_oob_ops *ops)
 {
+	struct onenand_chip *this = mtd->priv;
 	int ret;
 
 	switch (ops->mode) {
@@ -1094,7 +1403,9 @@ static int onenand_read_oob(struct mtd_info *mtd, loff_t from,
 
 	onenand_get_device(mtd, FL_READING);
 	if (ops->datbuf)
-		ret = onenand_read_ops_nolock(mtd, from, ops);
+		ret = ONENAND_IS_MLC(this) ?
+			onenand_mlc_read_ops_nolock(mtd, from, ops) :
+			onenand_read_ops_nolock(mtd, from, ops);
 	else
 		ret = onenand_read_oob_nolock(mtd, from, ops);
 	onenand_release_device(mtd);
@@ -1128,11 +1439,11 @@ static int onenand_bbt_wait(struct mtd_info *mtd, int state)
 	ctrl = this->read_word(this->base + ONENAND_REG_CTRL_STATUS);
 
 	if (interrupt & ONENAND_INT_READ) {
-		int ecc = this->read_word(this->base + ONENAND_REG_ECC_STATUS);
+		int ecc = onenand_read_ecc(this);
 		if (ecc & ONENAND_ECC_2BIT_ALL) {
 			printk(KERN_INFO "onenand_bbt_wait: ecc error = 0x%04x"
 				", controller error 0x%04x\n", ecc, ctrl);
-			return ONENAND_BBT_READ_ERROR;
+			return ONENAND_BBT_READ_ECC_ERROR;
 		}
 	} else {
 		printk(KERN_ERR "onenand_bbt_wait: read timeout!"
@@ -1163,7 +1474,7 @@ int onenand_bbt_read_oob(struct mtd_info *mtd, loff_t from,
 {
 	struct onenand_chip *this = mtd->priv;
 	int read = 0, thislen, column;
-	int ret = 0;
+	int ret = 0, readcmd;
 	size_t len = ops->ooblen;
 	u_char *buf = ops->oobbuf;
 
@@ -1183,17 +1494,22 @@ int onenand_bbt_read_oob(struct mtd_info *mtd, loff_t from,
 
 	column = from & (mtd->oobsize - 1);
 
+	readcmd = ONENAND_IS_MLC(this) ? ONENAND_CMD_READ : ONENAND_CMD_READOOB;
+
 	while (read < len) {
 		cond_resched();
 
 		thislen = mtd->oobsize - column;
 		thislen = min_t(int, thislen, len);
 
-		this->command(mtd, ONENAND_CMD_READOOB, from, mtd->oobsize);
+		this->command(mtd, readcmd, from, mtd->oobsize);
 
 		onenand_update_bufferram(mtd, from, 0);
 
 		ret = onenand_bbt_wait(mtd, FL_READING);
+		if (unlikely(ret))
+			ret = onenand_recover_lsb(mtd, from, ret);
+
 		if (ret)
 			break;
 
@@ -1230,9 +1546,11 @@ static int onenand_verify_oob(struct mtd_info *mtd, const u_char *buf, loff_t to
 {
 	struct onenand_chip *this = mtd->priv;
 	u_char *oob_buf = this->oob_buf;
-	int status, i;
+	int status, i, readcmd;
+
+	readcmd = ONENAND_IS_MLC(this) ? ONENAND_CMD_READ : ONENAND_CMD_READOOB;
 
-	this->command(mtd, ONENAND_CMD_READOOB, to, mtd->oobsize);
+	this->command(mtd, readcmd, to, mtd->oobsize);
 	onenand_update_bufferram(mtd, to, 0);
 	status = this->wait(mtd, FL_READING);
 	if (status)
@@ -1633,7 +1951,7 @@ static int onenand_write_oob_nolock(struct mtd_info *mtd, loff_t to,
 {
 	struct onenand_chip *this = mtd->priv;
 	int column, ret = 0, oobsize;
-	int written = 0;
+	int written = 0, oobcmd;
 	u_char *oobbuf;
 	size_t len = ops->ooblen;
 	const u_char *buf = ops->oobbuf;
@@ -1675,6 +1993,8 @@ static int onenand_write_oob_nolock(struct mtd_info *mtd, loff_t to,
 
 	oobbuf = this->oob_buf;
 
+	oobcmd = ONENAND_IS_MLC(this) ? ONENAND_CMD_PROG : ONENAND_CMD_PROGOOB;
+
 	/* Loop until all data write */
 	while (written < len) {
 		int thislen = min_t(int, oobsize, len - written);
@@ -1692,7 +2012,14 @@ static int onenand_write_oob_nolock(struct mtd_info *mtd, loff_t to,
 			memcpy(oobbuf + column, buf, thislen);
 		this->write_bufferram(mtd, ONENAND_SPARERAM, oobbuf, 0, mtd->oobsize);
 
-		this->command(mtd, ONENAND_CMD_PROGOOB, to, mtd->oobsize);
+		if (ONENAND_IS_MLC(this)) {
+			/* Set main area of DataRAM to 0xff*/
+			memset(this->page_buf, 0xff, mtd->writesize);
+			this->write_bufferram(mtd, ONENAND_DATARAM,
+					 this->page_buf, 0, mtd->writesize);
+		}
+
+		this->command(mtd, oobcmd, to, mtd->oobsize);
 
 		onenand_update_bufferram(mtd, to, 0);
 		if (ONENAND_IS_2PLANE(this)) {
@@ -1815,29 +2142,48 @@ static int onenand_erase(struct mtd_info *mtd, struct erase_info *instr)
 {
 	struct onenand_chip *this = mtd->priv;
 	unsigned int block_size;
-	loff_t addr;
-	int len;
-	int ret = 0;
+	loff_t addr = instr->addr;
+	loff_t len = instr->len;
+	int ret = 0, i;
+	struct mtd_erase_region_info *region = NULL;
+	loff_t region_end = 0;
 
 	DEBUG(MTD_DEBUG_LEVEL3, "onenand_erase: start = 0x%012llx, len = %llu\n", (unsigned long long) instr->addr, (unsigned long long) instr->len);
 
-	block_size = (1 << this->erase_shift);
-
-	/* Start address must align on block boundary */
-	if (unlikely(instr->addr & (block_size - 1))) {
-		printk(KERN_ERR "onenand_erase: Unaligned address\n");
+	/* Do not allow erase past end of device */
+	if (unlikely((len + addr) > mtd->size)) {
+		printk(KERN_ERR "onenand_erase: Erase past end of device\n");
 		return -EINVAL;
 	}
 
-	/* Length must align on block boundary */
-	if (unlikely(instr->len & (block_size - 1))) {
-		printk(KERN_ERR "onenand_erase: Length not block aligned\n");
-		return -EINVAL;
+	if (FLEXONENAND(this)) {
+		/* Find the eraseregion of this address */
+		i = flexonenand_region(mtd, addr);
+		region = &mtd->eraseregions[i];
+
+		block_size = region->erasesize;
+		region_end = region->offset + region->erasesize * region->numblocks;
+
+		/* Start address within region must align on block boundary.
+		 * Erase region's start offset is always block start address.
+		 */
+		if (unlikely((addr - region->offset) & (block_size - 1))) {
+			printk(KERN_ERR "onenand_erase: Unaligned address\n");
+			return -EINVAL;
+		}
+	} else {
+		block_size = 1 << this->erase_shift;
+
+		/* Start address must align on block boundary */
+		if (unlikely(addr & (block_size - 1))) {
+			printk(KERN_ERR "onenand_erase: Unaligned address\n");
+			return -EINVAL;
+		}
 	}
 
-	/* Do not allow erase past end of device */
-	if (unlikely((instr->len + instr->addr) > mtd->size)) {
-		printk(KERN_ERR "onenand_erase: Erase past end of device\n");
+	/* Length must align on block boundary */
+	if (unlikely(len & (block_size - 1))) {
+		printk(KERN_ERR "onenand_erase: Length not block aligned\n");
 		return -EINVAL;
 	}
 
@@ -1847,9 +2193,6 @@ static int onenand_erase(struct mtd_info *mtd, struct erase_info *instr)
 	onenand_get_device(mtd, FL_ERASING);
 
 	/* Loop throught the pages */
-	len = instr->len;
-	addr = instr->addr;
-
 	instr->state = MTD_ERASING;
 
 	while (len) {
@@ -1869,7 +2212,8 @@ static int onenand_erase(struct mtd_info *mtd, struct erase_info *instr)
 		ret = this->wait(mtd, FL_ERASING);
 		/* Check, if it is write protected */
 		if (ret) {
-			printk(KERN_ERR "onenand_erase: Failed erase, block %d\n", (unsigned) (addr >> this->erase_shift));
+			printk(KERN_ERR "onenand_erase: Failed erase, block %d\n",
+						 onenand_block(this, addr));
 			instr->state = MTD_ERASE_FAILED;
 			instr->fail_addr = addr;
 			goto erase_exit;
@@ -1877,6 +2221,22 @@ static int onenand_erase(struct mtd_info *mtd, struct erase_info *instr)
 
 		len -= block_size;
 		addr += block_size;
+
+		if (addr == region_end) {
+			if (!len)
+				break;
+			region++;
+
+			block_size = region->erasesize;
+			region_end = region->offset + region->erasesize * region->numblocks;
+
+			if (len & (block_size - 1)) {
+				/* FIXME: This should be handled at MTD partitioning level. */
+				printk(KERN_ERR "onenand_erase: Unaligned address\n");
+				goto erase_exit;
+			}
+		}
+
 	}
 
 	instr->state = MTD_ERASE_DONE;
@@ -1955,13 +2315,17 @@ static int onenand_default_block_markbad(struct mtd_info *mtd, loff_t ofs)
 	int block;
 
 	/* Get block number */
-	block = ((int) ofs) >> bbm->bbt_erase_shift;
+	block = onenand_block(this, ofs);
         if (bbm->bbt)
                 bbm->bbt[block >> 2] |= 0x01 << ((block & 0x03) << 1);
 
         /* We write two bytes, so we dont have to mess with 16 bit access */
         ofs += mtd->oobsize + (bbm->badblockpos & ~0x01);
-        return onenand_write_oob_nolock(mtd, ofs, &ops);
+	/* FIXME : What to do when marking SLC block in partition
+	 * 	   with MLC erasesize? For now, it is not advisable to
+	 *	   create partitions containing both SLC and MLC regions.
+	 */
+	return onenand_write_oob_nolock(mtd, ofs, &ops);
 }
 
 /**
@@ -2005,8 +2369,8 @@ static int onenand_do_lock_cmd(struct mtd_info *mtd, loff_t ofs, size_t len, int
 	int start, end, block, value, status;
 	int wp_status_mask;
 
-	start = ofs >> this->erase_shift;
-	end = len >> this->erase_shift;
+	start = onenand_block(this, ofs);
+	end = onenand_block(this, ofs + len) - 1;
 
 	if (cmd == ONENAND_CMD_LOCK)
 		wp_status_mask = ONENAND_WP_LS;
@@ -2018,7 +2382,7 @@ static int onenand_do_lock_cmd(struct mtd_info *mtd, loff_t ofs, size_t len, int
 		/* Set start block address */
 		this->write_word(start, this->base + ONENAND_REG_START_BLOCK_ADDRESS);
 		/* Set end block address */
-		this->write_word(start + end - 1, this->base + ONENAND_REG_END_BLOCK_ADDRESS);
+		this->write_word(end, this->base +  ONENAND_REG_END_BLOCK_ADDRESS);
 		/* Write lock command */
 		this->command(mtd, cmd, 0, 0);
 
@@ -2039,7 +2403,7 @@ static int onenand_do_lock_cmd(struct mtd_info *mtd, loff_t ofs, size_t len, int
 	}
 
 	/* Block lock scheme */
-	for (block = start; block < start + end; block++) {
+	for (block = start; block < end + 1; block++) {
 		/* Set block address */
 		value = onenand_block_address(this, block);
 		this->write_word(value, this->base + ONENAND_REG_START_ADDRESS1);
@@ -2147,7 +2511,7 @@ static void onenand_unlock_all(struct mtd_info *mtd)
 {
 	struct onenand_chip *this = mtd->priv;
 	loff_t ofs = 0;
-	size_t len = this->chipsize;
+	loff_t len = mtd->size;
 
 	if (this->options & ONENAND_HAS_UNLOCK_ALL) {
 		/* Set start block address */
@@ -2168,7 +2532,7 @@ static void onenand_unlock_all(struct mtd_info *mtd)
 			return;
 
 		/* Workaround for all block unlock in DDP */
-		if (ONENAND_IS_DDP(this)) {
+		if (ONENAND_IS_DDP(this) && !FLEXONENAND(this)) {
 			/* All blocks on another chip */
 			ofs = this->chipsize >> 1;
 			len = this->chipsize >> 1;
@@ -2210,7 +2574,9 @@ static int do_otp_read(struct mtd_info *mtd, loff_t from, size_t len,
 	this->command(mtd, ONENAND_CMD_OTP_ACCESS, 0, 0);
 	this->wait(mtd, FL_OTPING);
 
-	ret = onenand_read_ops_nolock(mtd, from, &ops);
+	ret = ONENAND_IS_MLC(this) ?
+		onenand_mlc_read_ops_nolock(mtd, from, &ops) :
+		onenand_read_ops_nolock(mtd, from, &ops);
 
 	/* Exit OTP access mode */
 	this->command(mtd, ONENAND_CMD_RESET, 0, 0);
@@ -2277,21 +2643,32 @@ static int do_otp_lock(struct mtd_info *mtd, loff_t from, size_t len,
 		size_t *retlen, u_char *buf)
 {
 	struct onenand_chip *this = mtd->priv;
-	struct mtd_oob_ops ops = {
-		.mode = MTD_OOB_PLACE,
-		.ooblen = len,
-		.oobbuf = buf,
-		.ooboffs = 0,
-	};
+	struct mtd_oob_ops ops;
 	int ret;
 
 	/* Enter OTP access mode */
 	this->command(mtd, ONENAND_CMD_OTP_ACCESS, 0, 0);
 	this->wait(mtd, FL_OTPING);
 
-	ret = onenand_write_oob_nolock(mtd, from, &ops);
-
-	*retlen = ops.oobretlen;
+	if (FLEXONENAND(this)) {
+		/*
+		 * For Flex-OneNAND, we write lock mark to 1st word of sector 4 of
+		 * main area of page 49.
+		 */
+		ops.len = mtd->writesize;
+		ops.ooblen = 0;
+		ops.datbuf = buf;
+		ops.oobbuf = NULL;
+		ret = onenand_write_ops_nolock(mtd, mtd->writesize * 49, &ops);
+		*retlen = ops.retlen;
+	} else {
+		ops.mode = MTD_OOB_PLACE;
+		ops.ooblen = len;
+		ops.oobbuf = buf;
+		ops.ooboffs = 0;
+		ret = onenand_write_oob_nolock(mtd, from, &ops);
+		*retlen = ops.oobretlen;
+	}
 
 	/* Exit OTP access mode */
 	this->command(mtd, ONENAND_CMD_RESET, 0, 0);
@@ -2475,27 +2852,34 @@ static int onenand_lock_user_prot_reg(struct mtd_info *mtd, loff_t from,
 			size_t len)
 {
 	struct onenand_chip *this = mtd->priv;
-	u_char *oob_buf = this->oob_buf;
+	u_char *buf = FLEXONENAND(this) ? this->page_buf : this->oob_buf;
 	size_t retlen;
 	int ret;
 
-	memset(oob_buf, 0xff, mtd->oobsize);
+	memset(buf, 0xff, FLEXONENAND(this) ? this->writesize
+						 : mtd->oobsize);
 	/*
 	 * Note: OTP lock operation
 	 *       OTP block : 0xXXFC
 	 *       1st block : 0xXXF3 (If chip support)
 	 *       Both      : 0xXXF0 (If chip support)
 	 */
-	oob_buf[ONENAND_OTP_LOCK_OFFSET] = 0xFC;
+	if (FLEXONENAND(this))
+		buf[FLEXONENAND_OTP_LOCK_OFFSET] = 0xFC;
+	else
+		buf[ONENAND_OTP_LOCK_OFFSET] = 0xFC;
 
 	/*
 	 * Write lock mark to 8th word of sector0 of page0 of the spare0.
 	 * We write 16 bytes spare area instead of 2 bytes.
+	 * For Flex-OneNAND, we write lock mark to 1st word of sector 4 of
+	 * main area of page 49.
 	 */
+
 	from = 0;
-	len = 16;
+	len = FLEXONENAND(this) ? mtd->writesize : 16;
 
-	ret = onenand_otp_walk(mtd, from, len, &retlen, oob_buf, do_otp_lock, MTD_OTP_USER);
+	ret = onenand_otp_walk(mtd, from, len, &retlen, buf, do_otp_lock, MTD_OTP_USER);
 
 	return ret ? : retlen;
 }
@@ -2542,6 +2926,14 @@ static void onenand_check_features(struct mtd_info *mtd)
 		break;
 	}
 
+	if (ONENAND_IS_MLC(this))
+		this->options &= ~ONENAND_HAS_2PLANE;
+
+	if (FLEXONENAND(this)) {
+		this->options &= ~ONENAND_HAS_CONT_LOCK;
+		this->options |= ONENAND_HAS_UNLOCK_ALL;
+	}
+
 	if (this->options & ONENAND_HAS_CONT_LOCK)
 		printk(KERN_DEBUG "Lock scheme is Continuous Lock\n");
 	if (this->options & ONENAND_HAS_UNLOCK_ALL)
@@ -2559,14 +2951,16 @@ static void onenand_check_features(struct mtd_info *mtd)
  */
 static void onenand_print_device_info(int device, int version)
 {
-        int vcc, demuxed, ddp, density;
+	int vcc, demuxed, ddp, density, flexonenand;
 
         vcc = device & ONENAND_DEVICE_VCC_MASK;
         demuxed = device & ONENAND_DEVICE_IS_DEMUX;
         ddp = device & ONENAND_DEVICE_IS_DDP;
         density = onenand_get_density(device);
-        printk(KERN_INFO "%sOneNAND%s %dMB %sV 16-bit (0x%02x)\n",
-                demuxed ? "" : "Muxed ",
+	flexonenand = device & DEVICE_IS_FLEXONENAND;
+	printk(KERN_INFO "%s%sOneNAND%s %dMB %sV 16-bit (0x%02x)\n",
+		demuxed ? "" : "Muxed ",
+		flexonenand ? "Flex-" : "",
                 ddp ? "(DDP)" : "",
                 (16 << density),
                 vcc ? "2.65/3.3" : "1.8",
@@ -2605,6 +2999,280 @@ static int onenand_check_maf(int manuf)
 	return (i == size);
 }
 
+/**
+* flexonenand_get_boundary	- Reads the SLC boundary
+* @param onenand_info		- onenand info structure
+**/
+static int flexonenand_get_boundary(struct mtd_info *mtd)
+{
+	struct onenand_chip *this = mtd->priv;
+	unsigned die, bdry;
+	int ret, syscfg, locked;
+
+	/* Disable ECC */
+	syscfg = this->read_word(this->base + ONENAND_REG_SYS_CFG1);
+	this->write_word((syscfg | 0x0100), this->base + ONENAND_REG_SYS_CFG1);
+
+	for (die = 0; die < this->dies; die++) {
+		this->command(mtd, FLEXONENAND_CMD_PI_ACCESS, die, 0);
+		this->wait(mtd, FL_SYNCING);
+
+		this->command(mtd, FLEXONENAND_CMD_READ_PI, die, 0);
+		ret = this->wait(mtd, FL_READING);
+
+		bdry = this->read_word(this->base + ONENAND_DATARAM);
+		if ((bdry >> FLEXONENAND_PI_UNLOCK_SHIFT) == 3)
+			locked = 0;
+		else
+			locked = 1;
+		this->boundary[die] = bdry & FLEXONENAND_PI_MASK;
+
+		this->command(mtd, ONENAND_CMD_RESET, 0, 0);
+		ret = this->wait(mtd, FL_RESETING);
+
+		printk(KERN_INFO "Die %d boundary: %d%s\n", die,
+		       this->boundary[die], locked ? "(Locked)" : "(Unlocked)");
+	}
+
+	/* Enable ECC */
+	this->write_word(syscfg, this->base + ONENAND_REG_SYS_CFG1);
+	return 0;
+}
+
+/**
+ * flexonenand_get_size - Fill up fields in onenand_chip and mtd_info
+ * 			  boundary[], diesize[], mtd->size, mtd->erasesize
+ * @param mtd		- MTD device structure
+ */
+static void flexonenand_get_size(struct mtd_info *mtd)
+{
+	struct onenand_chip *this = mtd->priv;
+	int die, i, eraseshift, density;
+	int blksperdie, maxbdry;
+	loff_t ofs;
+
+	density = onenand_get_density(this->device_id);
+	blksperdie = ((loff_t)(16 << density) << 20) >> (this->erase_shift);
+	blksperdie >>= ONENAND_IS_DDP(this) ? 1 : 0;
+	maxbdry = blksperdie - 1;
+	eraseshift = this->erase_shift - 1;
+
+	mtd->numeraseregions = this->dies << 1;
+
+	/* This fills up the device boundary */
+	flexonenand_get_boundary(mtd);
+	die = ofs = 0;
+	i = -1;
+	for (; die < this->dies; die++) {
+		if (!die || this->boundary[die-1] != maxbdry) {
+			i++;
+			mtd->eraseregions[i].offset = ofs;
+			mtd->eraseregions[i].erasesize = 1 << eraseshift;
+			mtd->eraseregions[i].numblocks =
+							this->boundary[die] + 1;
+			ofs += mtd->eraseregions[i].numblocks << eraseshift;
+			eraseshift++;
+		} else {
+			mtd->numeraseregions -= 1;
+			mtd->eraseregions[i].numblocks +=
+							this->boundary[die] + 1;
+			ofs += (this->boundary[die] + 1) << (eraseshift - 1);
+		}
+		if (this->boundary[die] != maxbdry) {
+			i++;
+			mtd->eraseregions[i].offset = ofs;
+			mtd->eraseregions[i].erasesize = 1 << eraseshift;
+			mtd->eraseregions[i].numblocks = maxbdry ^
+							 this->boundary[die];
+			ofs += mtd->eraseregions[i].numblocks << eraseshift;
+			eraseshift--;
+		} else
+			mtd->numeraseregions -= 1;
+	}
+
+	/* Expose MLC erase size except when all blocks are SLC */
+	mtd->erasesize = 1 << this->erase_shift;
+	if (mtd->numeraseregions == 1)
+		mtd->erasesize >>= 1;
+
+	printk(KERN_INFO "Device has %d eraseregions\n", mtd->numeraseregions);
+	for (i = 0; i < mtd->numeraseregions; i++)
+		printk(KERN_INFO "[offset: 0x%08x, erasesize: 0x%05x,"
+			" numblocks: %04u]\n",
+			(unsigned int) mtd->eraseregions[i].offset,
+			mtd->eraseregions[i].erasesize,
+			mtd->eraseregions[i].numblocks);
+
+	for (die = 0, mtd->size = 0; die < this->dies; die++) {
+		this->diesize[die] = (loff_t)blksperdie << this->erase_shift;
+		this->diesize[die] -= (loff_t)(this->boundary[die] + 1)
+						 << (this->erase_shift - 1);
+		mtd->size += this->diesize[die];
+	}
+}
+
+/**
+ * flexonenand_check_blocks_erased - Check if blocks are erased
+ * @param mtd_info	- mtd info structure
+ * @param start		- first erase block to check
+ * @param end		- last erase block to check
+ *
+ * Converting an unerased block from MLC to SLC
+ * causes byte values to change. Since both data and its ECC
+ * have changed, reads on the block give uncorrectable error.
+ * This might lead to the block being detected as bad.
+ *
+ * Avoid this by ensuring that the block to be converted is
+ * erased.
+ */
+static int flexonenand_check_blocks_erased(struct mtd_info *mtd, int start, int end)
+{
+	struct onenand_chip *this = mtd->priv;
+	int i, ret;
+	int block;
+	struct mtd_oob_ops ops = {
+		.mode = MTD_OOB_PLACE,
+		.ooboffs = 0,
+		.ooblen	= mtd->oobsize,
+		.datbuf	= NULL,
+		.oobbuf	= this->oob_buf,
+	};
+	loff_t addr;
+
+	printk(KERN_DEBUG "Check blocks from %d to %d\n", start, end);
+
+	for (block = start; block <= end; block++) {
+		addr = flexonenand_addr(this, block);
+		if (onenand_block_isbad_nolock(mtd, addr, 0))
+			continue;
+
+		/*
+		 * Since main area write results in ECC write to spare,
+		 * it is sufficient to check only ECC bytes for change.
+		 */
+		ret = onenand_read_oob_nolock(mtd, addr, &ops);
+		if (ret)
+			return ret;
+
+		for (i = 0; i < mtd->oobsize; i++)
+			if (this->oob_buf[i] != 0xff)
+				break;
+
+		if (i != mtd->oobsize) {
+			printk(KERN_WARNING "Block %d not erased.\n", block);
+			return 1;
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * flexonenand_set_boundary	- Writes the SLC boundary
+ * @param mtd			- mtd info structure
+ */
+int flexonenand_set_boundary(struct mtd_info *mtd, int die,
+				    int boundary, int lock)
+{
+	struct onenand_chip *this = mtd->priv;
+	int ret, density, blksperdie, old, new, thisboundary;
+	loff_t addr;
+
+	/* Change only once for SDP Flex-OneNAND */
+	if (die && (!ONENAND_IS_DDP(this)))
+		return 0;
+
+	/* boundary value of -1 indicates no required change */
+	if (boundary < 0 || boundary == this->boundary[die])
+		return 0;
+
+	density = onenand_get_density(this->device_id);
+	blksperdie = ((16 << density) << 20) >> this->erase_shift;
+	blksperdie >>= ONENAND_IS_DDP(this) ? 1 : 0;
+
+	if (boundary >= blksperdie) {
+		printk(KERN_ERR "flexonenand_set_boundary: Invalid boundary value. "
+				"Boundary not changed.\n");
+		return -EINVAL;
+	}
+
+	/* Check if converting blocks are erased */
+	old = this->boundary[die] + (die * this->density_mask);
+	new = boundary + (die * this->density_mask);
+	ret = flexonenand_check_blocks_erased(mtd, min(old, new) + 1, max(old, new));
+	if (ret) {
+		printk(KERN_ERR "flexonenand_set_boundary: Please erase blocks before boundary change\n");
+		return ret;
+	}
+
+	this->command(mtd, FLEXONENAND_CMD_PI_ACCESS, die, 0);
+	this->wait(mtd, FL_SYNCING);
+
+	/* Check is boundary is locked */
+	this->command(mtd, FLEXONENAND_CMD_READ_PI, die, 0);
+	ret = this->wait(mtd, FL_READING);
+
+	thisboundary = this->read_word(this->base + ONENAND_DATARAM);
+	if ((thisboundary >> FLEXONENAND_PI_UNLOCK_SHIFT) != 3) {
+		printk(KERN_ERR "flexonenand_set_boundary: boundary locked\n");
+		ret = 1;
+		goto out;
+	}
+
+	printk(KERN_INFO "flexonenand_set_boundary: Changing die %d boundary: %d%s\n",
+			die, boundary, lock ? "(Locked)" : "(Unlocked)");
+
+	addr = die ? this->diesize[0] : 0;
+
+	boundary &= FLEXONENAND_PI_MASK;
+	boundary |= lock ? 0 : (3 << FLEXONENAND_PI_UNLOCK_SHIFT);
+
+	this->command(mtd, ONENAND_CMD_ERASE, addr, 0);
+	ret = this->wait(mtd, FL_ERASING);
+	if (ret) {
+		printk(KERN_ERR "flexonenand_set_boundary: Failed PI erase for Die %d\n", die);
+		goto out;
+	}
+
+	this->write_word(boundary, this->base + ONENAND_DATARAM);
+	this->command(mtd, ONENAND_CMD_PROG, addr, 0);
+	ret = this->wait(mtd, FL_WRITING);
+	if (ret) {
+		printk(KERN_ERR "flexonenand_set_boundary: Failed PI write for Die %d\n", die);
+		goto out;
+	}
+
+	this->command(mtd, FLEXONENAND_CMD_PI_UPDATE, die, 0);
+	ret = this->wait(mtd, FL_WRITING);
+out:
+	this->write_word(ONENAND_CMD_RESET, this->base + ONENAND_REG_COMMAND);
+	this->wait(mtd, FL_RESETING);
+	if (!ret)
+		/* Recalculate device size on boundary change*/
+		flexonenand_get_size(mtd);
+
+	return ret;
+}
+
+/**
+ * flexonenand_setup - 	capture Flex-OneNAND boundary and lock
+ * 			values  passed as kernel parameters
+ * @param s	kernel parameter string
+ */
+static int flexonenand_setup(char *s)
+{
+	int ints[5], i;
+
+	s = get_options(s, 5, ints);
+
+	for (i = 0; i < ints[0]; i++)
+		flex_bdry[i] = ints[i + 1];
+
+	return 1;
+}
+
+__setup("onenand.bdry=", flexonenand_setup);
+
 /**
  * onenand_probe - [OneNAND Interface] Probe the OneNAND device
  * @param mtd		MTD device structure
@@ -2647,6 +3315,7 @@ static int onenand_probe(struct mtd_info *mtd)
 	maf_id = this->read_word(this->base + ONENAND_REG_MANUFACTURER_ID);
 	dev_id = this->read_word(this->base + ONENAND_REG_DEVICE_ID);
 	ver_id = this->read_word(this->base + ONENAND_REG_VERSION_ID);
+	this->technology = this->read_word(this->base + ONENAND_REG_TECHNOLOGY);
 
 	/* Check OneNAND device */
 	if (maf_id != bram_maf_id || dev_id != bram_dev_id)
@@ -2658,29 +3327,55 @@ static int onenand_probe(struct mtd_info *mtd)
 	this->version_id = ver_id;
 
 	density = onenand_get_density(dev_id);
+	if (FLEXONENAND(this)) {
+		this->dies = ONENAND_IS_DDP(this) ? 2 : 1;
+		/* Maximum possible erase regions */
+		mtd->numeraseregions = this->dies << 1;
+		mtd->eraseregions = kzalloc(sizeof(struct mtd_erase_region_info)
+					* (this->dies << 1), GFP_KERNEL);
+		if (!mtd->eraseregions)
+			return -ENOMEM;
+	}
+
+	/*
+	 * For Flex-OneNAND, chipsize represents maximum possible device size.
+	 * mtd->size represents the actual device size.
+	 */
 	this->chipsize = (16 << density) << 20;
-	/* Set density mask. it is used for DDP */
-	if (ONENAND_IS_DDP(this))
-		this->density_mask = (1 << (density + 6));
-	else
-		this->density_mask = 0;
 
 	/* OneNAND page size & block size */
 	/* The data buffer size is equal to page size */
 	mtd->writesize = this->read_word(this->base + ONENAND_REG_DATA_BUFFER_SIZE);
+	/* We use the full BufferRAM */
+	if (ONENAND_IS_MLC(this))
+		mtd->writesize <<= 1;
+
 	mtd->oobsize = mtd->writesize >> 5;
 	/* Pages per a block are always 64 in OneNAND */
 	mtd->erasesize = mtd->writesize << 6;
+	/*
+	 * Flex-OneNAND SLC area has 64 pages per block.
+	 * Flex-OneNAND MLC area has 128 pages per block.
+	 * Expose MLC erase size to find erase_shift and page_mask.
+	 */
+	if (FLEXONENAND(this))
+		mtd->erasesize <<= 1;
 
 	this->erase_shift = ffs(mtd->erasesize) - 1;
 	this->page_shift = ffs(mtd->writesize) - 1;
 	this->page_mask = (1 << (this->erase_shift - this->page_shift)) - 1;
+	/* Set density mask. it is used for DDP */
+	if (ONENAND_IS_DDP(this))
+		this->density_mask = this->chipsize >> (this->erase_shift + 1);
 	/* It's real page size */
 	this->writesize = mtd->writesize;
 
 	/* REVIST: Multichip handling */
 
-	mtd->size = this->chipsize;
+	if (FLEXONENAND(this))
+		flexonenand_get_size(mtd);
+	else
+		mtd->size = this->chipsize;
 
 	/* Check OneNAND features */
 	onenand_check_features(mtd);
@@ -2735,7 +3430,7 @@ static void onenand_resume(struct mtd_info *mtd)
  */
 int onenand_scan(struct mtd_info *mtd, int maxchips)
 {
-	int i;
+	int i, ret;
 	struct onenand_chip *this = mtd->priv;
 
 	if (!this->read_word)
@@ -2797,6 +3492,10 @@ int onenand_scan(struct mtd_info *mtd, int maxchips)
 	 * Allow subpage writes up to oobsize.
 	 */
 	switch (mtd->oobsize) {
+	case 128:
+		this->ecclayout = &onenand_oob_128;
+		mtd->subpage_sft = 0;
+		break;
 	case 64:
 		this->ecclayout = &onenand_oob_64;
 		mtd->subpage_sft = 2;
@@ -2862,7 +3561,16 @@ int onenand_scan(struct mtd_info *mtd, int maxchips)
 	/* Unlock whole block */
 	onenand_unlock_all(mtd);
 
-	return this->scan_bbt(mtd);
+	ret = this->scan_bbt(mtd);
+	if ((!FLEXONENAND(this)) || ret)
+		return ret;
+
+	/* Change Flex-OneNAND boundaries if required */
+	for (i = 0; i < MAX_DIES; i++)
+		flexonenand_set_boundary(mtd, i, flex_bdry[2 * i],
+						 flex_bdry[(2 * i) + 1]);
+
+	return 0;
 }
 
 /**
@@ -2891,6 +3599,7 @@ void onenand_release(struct mtd_info *mtd)
 		kfree(this->page_buf);
 	if (this->options & ONENAND_OOBBUF_ALLOC)
 		kfree(this->oob_buf);
+	kfree(mtd->eraseregions);
 }
 
 EXPORT_SYMBOL_GPL(onenand_scan);
diff --git a/drivers/mtd/onenand/onenand_bbt.c b/drivers/mtd/onenand/onenand_bbt.c
index 2f53b51c6805..a91fcac1af01 100644
--- a/drivers/mtd/onenand/onenand_bbt.c
+++ b/drivers/mtd/onenand/onenand_bbt.c
@@ -63,6 +63,7 @@ static int create_bbt(struct mtd_info *mtd, uint8_t *buf, struct nand_bbt_descr
 	loff_t from;
 	size_t readlen, ooblen;
 	struct mtd_oob_ops ops;
+	int rgn;
 
 	printk(KERN_INFO "Scanning device for bad blocks\n");
 
@@ -76,7 +77,7 @@ static int create_bbt(struct mtd_info *mtd, uint8_t *buf, struct nand_bbt_descr
 	/* Note that numblocks is 2 * (real numblocks) here;
 	 * see i += 2 below as it makses shifting and masking less painful
 	 */
-	numblocks = mtd->size >> (bbm->bbt_erase_shift - 1);
+	numblocks = this->chipsize >> (bbm->bbt_erase_shift - 1);
 	startblock = 0;
 	from = 0;
 
@@ -106,7 +107,12 @@ static int create_bbt(struct mtd_info *mtd, uint8_t *buf, struct nand_bbt_descr
 			}
 		}
 		i += 2;
-		from += (1 << bbm->bbt_erase_shift);
+
+		if (FLEXONENAND(this)) {
+			rgn = flexonenand_region(mtd, from);
+			from += mtd->eraseregions[rgn].erasesize;
+		} else
+			from += (1 << bbm->bbt_erase_shift);
 	}
 
 	return 0;
@@ -143,7 +149,7 @@ static int onenand_isbad_bbt(struct mtd_info *mtd, loff_t offs, int allowbbt)
 	uint8_t res;
 
 	/* Get block number * 2 */
-	block = (int) (offs >> (bbm->bbt_erase_shift - 1));
+	block = (int) (onenand_block(this, offs) << 1);
 	res = (bbm->bbt[block >> 3] >> (block & 0x06)) & 0x03;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "onenand_isbad_bbt: bbt info for offs 0x%08x: (block %d) 0x%02x\n",
@@ -178,7 +184,7 @@ int onenand_scan_bbt(struct mtd_info *mtd, struct nand_bbt_descr *bd)
 	struct bbm_info *bbm = this->bbm;
 	int len, ret = 0;
 
-	len = mtd->size >> (this->erase_shift + 2);
+	len = this->chipsize >> (this->erase_shift + 2);
 	/* Allocate memory (2bit per block) and clear the memory bad block table */
 	bbm->bbt = kzalloc(len, GFP_KERNEL);
 	if (!bbm->bbt) {
diff --git a/drivers/mtd/onenand/onenand_sim.c b/drivers/mtd/onenand/onenand_sim.c
index d64200b7c94b..f6e3c8aebd3a 100644
--- a/drivers/mtd/onenand/onenand_sim.c
+++ b/drivers/mtd/onenand/onenand_sim.c
@@ -6,6 +6,10 @@
  *  Copyright © 2005-2007 Samsung Electronics
  *  Kyungmin Park <kyungmin.park@samsung.com>
  *
+ *  Vishak G <vishak.g at samsung.com>, Rohit Hagargundgi <h.rohit at samsung.com>
+ *  Flex-OneNAND simulator support
+ *  Copyright (C) Samsung Electronics, 2008
+ *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
@@ -24,16 +28,38 @@
 #ifndef CONFIG_ONENAND_SIM_MANUFACTURER
 #define CONFIG_ONENAND_SIM_MANUFACTURER         0xec
 #endif
+
 #ifndef CONFIG_ONENAND_SIM_DEVICE_ID
 #define CONFIG_ONENAND_SIM_DEVICE_ID            0x04
 #endif
+
+#define CONFIG_FLEXONENAND ((CONFIG_ONENAND_SIM_DEVICE_ID >> 9) & 1)
+
 #ifndef CONFIG_ONENAND_SIM_VERSION_ID
 #define CONFIG_ONENAND_SIM_VERSION_ID           0x1e
 #endif
 
+#ifndef CONFIG_ONENAND_SIM_TECHNOLOGY_ID
+#define CONFIG_ONENAND_SIM_TECHNOLOGY_ID CONFIG_FLEXONENAND
+#endif
+
+/* Initial boundary values for Flex-OneNAND Simulator */
+#ifndef CONFIG_FLEXONENAND_SIM_DIE0_BOUNDARY
+#define CONFIG_FLEXONENAND_SIM_DIE0_BOUNDARY	0x01
+#endif
+
+#ifndef CONFIG_FLEXONENAND_SIM_DIE1_BOUNDARY
+#define CONFIG_FLEXONENAND_SIM_DIE1_BOUNDARY	0x01
+#endif
+
 static int manuf_id	= CONFIG_ONENAND_SIM_MANUFACTURER;
 static int device_id	= CONFIG_ONENAND_SIM_DEVICE_ID;
 static int version_id	= CONFIG_ONENAND_SIM_VERSION_ID;
+static int technology_id = CONFIG_ONENAND_SIM_TECHNOLOGY_ID;
+static int boundary[] = {
+	CONFIG_FLEXONENAND_SIM_DIE0_BOUNDARY,
+	CONFIG_FLEXONENAND_SIM_DIE1_BOUNDARY,
+};
 
 struct onenand_flash {
 	void __iomem *base;
@@ -57,12 +83,18 @@ struct onenand_flash {
 	(writew(v, this->base + ONENAND_REG_WP_STATUS))
 
 /* It has all 0xff chars */
-#define MAX_ONENAND_PAGESIZE		(2048 + 64)
+#define MAX_ONENAND_PAGESIZE		(4096 + 128)
 static unsigned char *ffchars;
 
+#if CONFIG_FLEXONENAND
+#define PARTITION_NAME "Flex-OneNAND simulator partition"
+#else
+#define PARTITION_NAME "OneNAND simulator partition"
+#endif
+
 static struct mtd_partition os_partitions[] = {
 	{
-		.name		= "OneNAND simulator partition",
+		.name		= PARTITION_NAME,
 		.offset		= 0,
 		.size		= MTDPART_SIZ_FULL,
 	},
@@ -104,6 +136,7 @@ static void onenand_lock_handle(struct onenand_chip *this, int cmd)
 
 	switch (cmd) {
 	case ONENAND_CMD_UNLOCK:
+	case ONENAND_CMD_UNLOCK_ALL:
 		if (block_lock_scheme)
 			ONENAND_SET_WP_STATUS(ONENAND_WP_US, this);
 		else
@@ -228,10 +261,12 @@ static void onenand_data_handle(struct onenand_chip *this, int cmd,
 {
 	struct mtd_info *mtd = &info->mtd;
 	struct onenand_flash *flash = this->priv;
-	int main_offset, spare_offset;
+	int main_offset, spare_offset, die = 0;
 	void __iomem *src;
 	void __iomem *dest;
 	unsigned int i;
+	static int pi_operation;
+	int erasesize, rgn;
 
 	if (dataram) {
 		main_offset = mtd->writesize;
@@ -241,10 +276,27 @@ static void onenand_data_handle(struct onenand_chip *this, int cmd,
 		spare_offset = 0;
 	}
 
+	if (pi_operation) {
+		die = readw(this->base + ONENAND_REG_START_ADDRESS2);
+		die >>= ONENAND_DDP_SHIFT;
+	}
+
 	switch (cmd) {
+	case FLEXONENAND_CMD_PI_ACCESS:
+		pi_operation = 1;
+		break;
+
+	case ONENAND_CMD_RESET:
+		pi_operation = 0;
+		break;
+
 	case ONENAND_CMD_READ:
 		src = ONENAND_CORE(flash) + offset;
 		dest = ONENAND_MAIN_AREA(this, main_offset);
+		if (pi_operation) {
+			writew(boundary[die], this->base + ONENAND_DATARAM);
+			break;
+		}
 		memcpy(dest, src, mtd->writesize);
 		/* Fall through */
 
@@ -257,6 +309,10 @@ static void onenand_data_handle(struct onenand_chip *this, int cmd,
 	case ONENAND_CMD_PROG:
 		src = ONENAND_MAIN_AREA(this, main_offset);
 		dest = ONENAND_CORE(flash) + offset;
+		if (pi_operation) {
+			boundary[die] = readw(this->base + ONENAND_DATARAM);
+			break;
+		}
 		/* To handle partial write */
 		for (i = 0; i < (1 << mtd->subpage_sft); i++) {
 			int off = i * this->subpagesize;
@@ -284,9 +340,18 @@ static void onenand_data_handle(struct onenand_chip *this, int cmd,
 		break;
 
 	case ONENAND_CMD_ERASE:
-		memset(ONENAND_CORE(flash) + offset, 0xff, mtd->erasesize);
+		if (pi_operation)
+			break;
+
+		if (FLEXONENAND(this)) {
+			rgn = flexonenand_region(mtd, offset);
+			erasesize = mtd->eraseregions[rgn].erasesize;
+		} else
+			erasesize = mtd->erasesize;
+
+		memset(ONENAND_CORE(flash) + offset, 0xff, erasesize);
 		memset(ONENAND_CORE_SPARE(flash, this, offset), 0xff,
-		       (mtd->erasesize >> 5));
+		       (erasesize >> 5));
 		break;
 
 	default:
@@ -339,7 +404,7 @@ static void onenand_command_handle(struct onenand_chip *this, int cmd)
 	}
 
 	if (block != -1)
-		offset += block << this->erase_shift;
+		offset = onenand_addr(this, block);
 
 	if (page != -1)
 		offset += page << this->page_shift;
@@ -390,6 +455,7 @@ static int __init flash_init(struct onenand_flash *flash)
 	}
 
 	density = device_id >> ONENAND_DEVICE_DENSITY_SHIFT;
+	density &= ONENAND_DEVICE_DENSITY_MASK;
 	size = ((16 << 20) << density);
 
 	ONENAND_CORE(flash) = vmalloc(size + (size >> 5));
@@ -405,8 +471,9 @@ static int __init flash_init(struct onenand_flash *flash)
 	writew(manuf_id, flash->base + ONENAND_REG_MANUFACTURER_ID);
 	writew(device_id, flash->base + ONENAND_REG_DEVICE_ID);
 	writew(version_id, flash->base + ONENAND_REG_VERSION_ID);
+	writew(technology_id, flash->base + ONENAND_REG_TECHNOLOGY);
 
-	if (density < 2)
+	if (density < 2 && (!CONFIG_FLEXONENAND))
 		buffer_size = 0x0400;	/* 1KiB page */
 	else
 		buffer_size = 0x0800;	/* 2KiB page */
diff --git a/include/linux/mtd/onenand.h b/include/linux/mtd/onenand.h
index 0fa3ac4ad576..9aab82c1c743 100644
--- a/include/linux/mtd/onenand.h
+++ b/include/linux/mtd/onenand.h
@@ -17,6 +17,7 @@
 #include <linux/mtd/onenand_regs.h>
 #include <linux/mtd/bbm.h>
 
+#define MAX_DIES		2
 #define MAX_BUFFERRAM		2
 
 /* Scan and identify a OneNAND device */
@@ -51,7 +52,12 @@ struct onenand_bufferram {
 /**
  * struct onenand_chip - OneNAND Private Flash Chip Data
  * @base:		[BOARDSPECIFIC] address to access OneNAND
+ * @dies:		[INTERN][FLEX-ONENAND] number of dies on chip
+ * @boundary:		[INTERN][FLEX-ONENAND] Boundary of the dies
+ * @diesize:		[INTERN][FLEX-ONENAND] Size of the dies
  * @chipsize:		[INTERN] the size of one chip for multichip arrays
+ *			FIXME For Flex-OneNAND, chipsize holds maximum possible
+ *			device size ie when all blocks are considered MLC
  * @device_id:		[INTERN] device ID
  * @density_mask:	chip density, used for DDP devices
  * @verstion_id:	[INTERN] version ID
@@ -92,9 +98,13 @@ struct onenand_bufferram {
  */
 struct onenand_chip {
 	void __iomem		*base;
+	unsigned		dies;
+	unsigned		boundary[MAX_DIES];
+	loff_t			diesize[MAX_DIES];
 	unsigned int		chipsize;
 	unsigned int		device_id;
 	unsigned int		version_id;
+	unsigned int		technology;
 	unsigned int		density_mask;
 	unsigned int		options;
 
@@ -145,6 +155,8 @@ struct onenand_chip {
 #define ONENAND_SET_BUFFERRAM0(this)		(this->bufferram_index = 0)
 #define ONENAND_SET_BUFFERRAM1(this)		(this->bufferram_index = 1)
 
+#define FLEXONENAND(this)						\
+	(this->device_id & DEVICE_IS_FLEXONENAND)
 #define ONENAND_GET_SYS_CFG1(this)					\
 	(this->read_word(this->base + ONENAND_REG_SYS_CFG1))
 #define ONENAND_SET_SYS_CFG1(v, this)					\
@@ -153,6 +165,9 @@ struct onenand_chip {
 #define ONENAND_IS_DDP(this)						\
 	(this->device_id & ONENAND_DEVICE_IS_DDP)
 
+#define ONENAND_IS_MLC(this)						\
+	(this->technology & ONENAND_TECHNOLOGY_IS_MLC)
+
 #ifdef CONFIG_MTD_ONENAND_2X_PROGRAM
 #define ONENAND_IS_2PLANE(this)						\
 	(this->options & ONENAND_HAS_2PLANE)
@@ -190,5 +205,8 @@ struct onenand_manufacturers {
 
 int onenand_bbt_read_oob(struct mtd_info *mtd, loff_t from,
 			 struct mtd_oob_ops *ops);
+unsigned onenand_block(struct onenand_chip *this, loff_t addr);
+loff_t onenand_addr(struct onenand_chip *this, int block);
+int flexonenand_region(struct mtd_info *mtd, loff_t addr);
 
 #endif	/* __LINUX_MTD_ONENAND_H */
diff --git a/include/linux/mtd/onenand_regs.h b/include/linux/mtd/onenand_regs.h
index 0c6bbe28f38c..86a6bbef6465 100644
--- a/include/linux/mtd/onenand_regs.h
+++ b/include/linux/mtd/onenand_regs.h
@@ -67,6 +67,9 @@
 /*
  * Device ID Register F001h (R)
  */
+#define DEVICE_IS_FLEXONENAND		(1 << 9)
+#define FLEXONENAND_PI_MASK		(0x3ff)
+#define FLEXONENAND_PI_UNLOCK_SHIFT	(14)
 #define ONENAND_DEVICE_DENSITY_MASK	(0xf)
 #define ONENAND_DEVICE_DENSITY_SHIFT	(4)
 #define ONENAND_DEVICE_IS_DDP		(1 << 3)
@@ -83,6 +86,11 @@
  */
 #define ONENAND_VERSION_PROCESS_SHIFT	(8)
 
+/*
+ * Technology Register F006h (R)
+ */
+#define ONENAND_TECHNOLOGY_IS_MLC	(1 << 0)
+
 /*
  * Start Address 1 F100h (R/W) & Start Address 2 F101h (R/W)
  */
@@ -93,7 +101,8 @@
 /*
  * Start Address 8 F107h (R/W)
  */
-#define ONENAND_FPA_MASK		(0x3f)
+/* Note: It's actually 0x3f in case of SLC */
+#define ONENAND_FPA_MASK		(0x7f)
 #define ONENAND_FPA_SHIFT		(2)
 #define ONENAND_FSA_MASK		(0x03)
 
@@ -105,7 +114,8 @@
 #define ONENAND_BSA_BOOTRAM		(0 << 2)
 #define ONENAND_BSA_DATARAM0		(2 << 2)
 #define ONENAND_BSA_DATARAM1		(3 << 2)
-#define ONENAND_BSC_MASK		(0x03)
+/* Note: It's actually 0x03 in case of SLC */
+#define ONENAND_BSC_MASK		(0x07)
 
 /*
  * Command Register F220h (R/W)
@@ -124,9 +134,13 @@
 #define ONENAND_CMD_RESET		(0xF0)
 #define ONENAND_CMD_OTP_ACCESS		(0x65)
 #define ONENAND_CMD_READID		(0x90)
+#define FLEXONENAND_CMD_PI_UPDATE	(0x05)
+#define FLEXONENAND_CMD_PI_ACCESS	(0x66)
+#define FLEXONENAND_CMD_RECOVER_LSB	(0x05)
 
 /* NOTE: Those are not *REAL* commands */
 #define ONENAND_CMD_BUFFERRAM		(0x1978)
+#define FLEXONENAND_CMD_READ_PI		(0x1985)
 
 /*
  * System Configuration 1 Register F221h (R, R/W)
@@ -192,10 +206,12 @@
 #define ONENAND_ECC_1BIT_ALL		(0x5555)
 #define ONENAND_ECC_2BIT		(1 << 1)
 #define ONENAND_ECC_2BIT_ALL		(0xAAAA)
+#define FLEXONENAND_UNCORRECTABLE_ERROR	(0x1010)
 
 /*
  * One-Time Programmable (OTP)
  */
+#define FLEXONENAND_OTP_LOCK_OFFSET		(2048)
 #define ONENAND_OTP_LOCK_OFFSET		(14)
 
 #endif	/* __ONENAND_REG_H */
-- 
cgit v1.2.3-59-g8ed1b


From 31bb999ee73748068ddc271dd99b22dcc418efe3 Mon Sep 17 00:00:00 2001
From: Kyungmin Park <kmpark@infradead.org>
Date: Tue, 12 May 2009 13:46:57 -0700
Subject: mtd: onenand: add bbt_wait & unlock_all as replaceable for some
 platform

Add bbt_wait & unlock_all as replaceable for some platform such as
  s3c64xx s3c64xx has its own OneNAND controller and another interface

Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/onenand/onenand_base.c | 12 ++++++++++--
 include/linux/mtd/onenand.h        |  5 +++++
 2 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
index 8d4c9c253732..864327ed7fb3 100644
--- a/drivers/mtd/onenand/onenand_base.c
+++ b/drivers/mtd/onenand/onenand_base.c
@@ -1506,7 +1506,7 @@ int onenand_bbt_read_oob(struct mtd_info *mtd, loff_t from,
 
 		onenand_update_bufferram(mtd, from, 0);
 
-		ret = onenand_bbt_wait(mtd, FL_READING);
+		ret = this->bbt_wait(mtd, FL_READING);
 		if (unlikely(ret))
 			ret = onenand_recover_lsb(mtd, from, ret);
 
@@ -2527,6 +2527,10 @@ static void onenand_unlock_all(struct mtd_info *mtd)
 		    & ONENAND_CTRL_ONGO)
 			continue;
 
+		/* Don't check lock status */
+		if (this->options & ONENAND_SKIP_UNLOCK_CHECK)
+			return;
+
 		/* Check lock status */
 		if (onenand_check_lock_status(this))
 			return;
@@ -3442,6 +3446,10 @@ int onenand_scan(struct mtd_info *mtd, int maxchips)
 		this->command = onenand_command;
 	if (!this->wait)
 		onenand_setup_wait(mtd);
+	if (!this->bbt_wait)
+		this->bbt_wait = onenand_bbt_wait;
+	if (!this->unlock_all)
+		this->unlock_all = onenand_unlock_all;
 
 	if (!this->read_bufferram)
 		this->read_bufferram = onenand_read_bufferram;
@@ -3559,7 +3567,7 @@ int onenand_scan(struct mtd_info *mtd, int maxchips)
 	mtd->owner = THIS_MODULE;
 
 	/* Unlock whole block */
-	onenand_unlock_all(mtd);
+	this->unlock_all(mtd);
 
 	ret = this->scan_bbt(mtd);
 	if ((!FLEXONENAND(this)) || ret)
diff --git a/include/linux/mtd/onenand.h b/include/linux/mtd/onenand.h
index 9aab82c1c743..8ed873374381 100644
--- a/include/linux/mtd/onenand.h
+++ b/include/linux/mtd/onenand.h
@@ -74,6 +74,8 @@ struct onenand_bufferram {
  * @command:		[REPLACEABLE] hardware specific function for writing
  *			commands to the chip
  * @wait:		[REPLACEABLE] hardware specific function for wait on ready
+ * @bbt_wait:		[REPLACEABLE] hardware specific function for bbt wait on ready
+ * @unlock_all:		[REPLACEABLE] hardware specific function for unlock all
  * @read_bufferram:	[REPLACEABLE] hardware specific function for BufferRAM Area
  * @write_bufferram:	[REPLACEABLE] hardware specific function for BufferRAM Area
  * @read_word:		[REPLACEABLE] hardware specific function for read
@@ -118,6 +120,8 @@ struct onenand_chip {
 
 	int (*command)(struct mtd_info *mtd, int cmd, loff_t address, size_t len);
 	int (*wait)(struct mtd_info *mtd, int state);
+	int (*bbt_wait)(struct mtd_info *mtd, int state);
+	void (*unlock_all)(struct mtd_info *mtd);
 	int (*read_bufferram)(struct mtd_info *mtd, int area,
 			unsigned char *buffer, int offset, size_t count);
 	int (*write_bufferram)(struct mtd_info *mtd, int area,
@@ -184,6 +188,7 @@ struct onenand_chip {
 #define ONENAND_HAS_CONT_LOCK		(0x0001)
 #define ONENAND_HAS_UNLOCK_ALL		(0x0002)
 #define ONENAND_HAS_2PLANE		(0x0004)
+#define ONENAND_SKIP_UNLOCK_CHECK	(0x0100)
 #define ONENAND_PAGEBUF_ALLOC		(0x1000)
 #define ONENAND_OOBBUF_ALLOC		(0x2000)
 
-- 
cgit v1.2.3-59-g8ed1b


From 008fe148cb0fb51d266baabe2c09997b21cf90c6 Mon Sep 17 00:00:00 2001
From: "Luck, Tony" <tony.luck@intel.com>
Date: Tue, 26 May 2009 15:45:12 -0700
Subject: intel-iommu: Fix one last ia64 build problem in Pass Through Support

On ia64 with CONFIG_DMAR=n and CONFIG_SWIOTLB=y (as used in
arch/ia64/configs/tiger_defconfig) there is still a link
error with iommu_pass_through listed as an undefined symbol:

arch/ia64/kernel/built-in.o: In function `pci_swiotlb_init':
(.init.text+0x7f70): undefined reference to `iommu_pass_through'

Fix it by #defining iommu_pass_through away in asm/iommu.h

Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 arch/ia64/include/asm/iommu.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/ia64/include/asm/iommu.h b/arch/ia64/include/asm/iommu.h
index 37d41ca5645a..745e095fe82e 100644
--- a/arch/ia64/include/asm/iommu.h
+++ b/arch/ia64/include/asm/iommu.h
@@ -9,7 +9,11 @@ extern void pci_iommu_shutdown(void);
 extern void no_iommu_init(void);
 extern int force_iommu, no_iommu;
 extern int iommu_detected;
+#ifdef	CONFIG_DMAR
 extern int iommu_pass_through;
+#else
+#define iommu_pass_through	(0)
+#endif
 extern void iommu_dma_init(void);
 extern void machvec_init(const char *name);
 
-- 
cgit v1.2.3-59-g8ed1b


From a755a3858f96ea7e8762ecaac451adfad45321bd Mon Sep 17 00:00:00 2001
From: Peter Korsgaard <jacmet@sunsite.dk>
Date: Wed, 3 Jun 2009 13:46:54 +0200
Subject: mtd: nand: s3c2410_nand_setrate(): use correct macros for 2412/2440

Use the correct S3C2440_NFCONF_* macros for the mask for the 2412/2440
variants instead of the 2410 ones which use wrong bit positions.

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/s3c2410.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index 8e375d5fe231..776756e4ebe5 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -215,9 +215,9 @@ static int s3c2410_nand_setrate(struct s3c2410_nand_info *info)
 
 	case TYPE_S3C2440:
 	case TYPE_S3C2412:
-		mask = (S3C2410_NFCONF_TACLS(tacls_max - 1) |
-			S3C2410_NFCONF_TWRPH0(7) |
-			S3C2410_NFCONF_TWRPH1(7));
+		mask = (S3C2440_NFCONF_TACLS(tacls_max - 1) |
+			S3C2440_NFCONF_TWRPH0(7) |
+			S3C2440_NFCONF_TWRPH1(7));
 
 		set = S3C2440_NFCONF_TACLS(tacls - 1);
 		set |= S3C2440_NFCONF_TWRPH0(twrph0 - 1);
-- 
cgit v1.2.3-59-g8ed1b


From 43950a605dc76677f0c74dcd818a57d4df040e12 Mon Sep 17 00:00:00 2001
From: Roel Kluin <roel.kluin@gmail.com>
Date: Thu, 4 Jun 2009 16:24:59 +0200
Subject: mtd: nand: max_retries off by one in mxc_nand

with `while (max_retries-- > 0)' max_retries reaches -1 after the loop.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/mxc_nand.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/nand/mxc_nand.c b/drivers/mtd/nand/mxc_nand.c
index d03bd4eff722..da41ea82941a 100644
--- a/drivers/mtd/nand/mxc_nand.c
+++ b/drivers/mtd/nand/mxc_nand.c
@@ -199,7 +199,7 @@ static void wait_op_done(struct mxc_nand_host *host, int max_retries,
 			}
 			udelay(1);
 		}
-		if (max_retries <= 0)
+		if (max_retries < 0)
 			DEBUG(MTD_DEBUG_LEVEL0, "%s(%d): INT not set\n",
 			      __func__, param);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 143070e74630b9557e1bb64d899ff2cc5a1dcb48 Mon Sep 17 00:00:00 2001
From: Stefan Roese <sr@denx.de>
Date: Thu, 16 Apr 2009 14:10:45 +0200
Subject: mtd: physmap_of: Add multiple regions and concatenation support

This patch adds support to handle multiple non-identical chips in one
flash device tree node. It also adds concat support to physmap_of. This
makes it possible to support e.g. the Intel P30 48F4400 chips which
internally consists of 2 non-identical NOR chips on one die. Additionally
partitions now can span over multiple chips.

To describe such a chip's, multiple "reg" tuples are now supported in one
flash device tree node. Here an dts example:

        flash@f0000000,0 {
                #address-cells = <1>;
                #size-cells = <1>;
                compatible = "cfi-flash";
                reg = <0 0x00000000 0x02000000
                       0 0x02000000 0x02000000>;
                bank-width = <2>;
                partition@0 {
                        label = "test-part1";
                        reg = <0 0x04000000>;
                };
        };

Signed-off-by: Stefan Roese <sr@denx.de>
Reviewed-by: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/physmap_of.c | 199 ++++++++++++++++++++++++++++++------------
 1 file changed, 143 insertions(+), 56 deletions(-)

diff --git a/drivers/mtd/maps/physmap_of.c b/drivers/mtd/maps/physmap_of.c
index c83a60fada53..39d357b2eb47 100644
--- a/drivers/mtd/maps/physmap_of.c
+++ b/drivers/mtd/maps/physmap_of.c
@@ -20,16 +20,23 @@
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/map.h>
 #include <linux/mtd/partitions.h>
+#include <linux/mtd/concat.h>
 #include <linux/of.h>
 #include <linux/of_platform.h>
 
+struct of_flash_list {
+	struct mtd_info *mtd;
+	struct map_info map;
+	struct resource *res;
+};
+
 struct of_flash {
-	struct mtd_info		*mtd;
-	struct map_info		map;
-	struct resource		*res;
+	struct mtd_info		*cmtd;
 #ifdef CONFIG_MTD_PARTITIONS
 	struct mtd_partition	*parts;
 #endif
+	int list_size; /* number of elements in of_flash_list */
+	struct of_flash_list	list[0];
 };
 
 #ifdef CONFIG_MTD_PARTITIONS
@@ -88,30 +95,44 @@ static int parse_obsolete_partitions(struct of_device *dev,
 static int of_flash_remove(struct of_device *dev)
 {
 	struct of_flash *info;
+	int i;
 
 	info = dev_get_drvdata(&dev->dev);
 	if (!info)
 		return 0;
 	dev_set_drvdata(&dev->dev, NULL);
 
-	if (info->mtd) {
+#ifdef CONFIG_MTD_CONCAT
+	if (info->cmtd != info->list[0].mtd) {
+		del_mtd_device(info->cmtd);
+		mtd_concat_destroy(info->cmtd);
+	}
+#endif
+
+	if (info->cmtd) {
 		if (OF_FLASH_PARTS(info)) {
-			del_mtd_partitions(info->mtd);
+			del_mtd_partitions(info->cmtd);
 			kfree(OF_FLASH_PARTS(info));
 		} else {
-			del_mtd_device(info->mtd);
+			del_mtd_device(info->cmtd);
 		}
-		map_destroy(info->mtd);
 	}
 
-	if (info->map.virt)
-		iounmap(info->map.virt);
+	for (i = 0; i < info->list_size; i++) {
+		if (info->list[i].mtd)
+			map_destroy(info->list[i].mtd);
 
-	if (info->res) {
-		release_resource(info->res);
-		kfree(info->res);
+		if (info->list[i].map.virt)
+			iounmap(info->list[i].map.virt);
+
+		if (info->list[i].res) {
+			release_resource(info->list[i].res);
+			kfree(info->list[i].res);
+		}
 	}
 
+	kfree(info);
+
 	return 0;
 }
 
@@ -164,68 +185,130 @@ static int __devinit of_flash_probe(struct of_device *dev,
 	const char *probe_type = match->data;
 	const u32 *width;
 	int err;
-
-	err = -ENXIO;
-	if (of_address_to_resource(dp, 0, &res)) {
-		dev_err(&dev->dev, "Can't get IO address from device tree\n");
+	int i;
+	int count;
+	const u32 *p;
+	int reg_tuple_size;
+	struct mtd_info **mtd_list = NULL;
+
+	reg_tuple_size = (of_n_addr_cells(dp) + of_n_size_cells(dp)) * sizeof(u32);
+
+	/*
+	 * Get number of "reg" tuples. Scan for MTD devices on area's
+	 * described by each "reg" region. This makes it possible (including
+	 * the concat support) to support the Intel P30 48F4400 chips which
+	 * consists internally of 2 non-identical NOR chips on one die.
+	 */
+	p = of_get_property(dp, "reg", &count);
+	if (count % reg_tuple_size != 0) {
+		dev_err(&dev->dev, "Malformed reg property on %s\n",
+				dev->node->full_name);
+		err = -EINVAL;
 		goto err_out;
 	}
-
-       	dev_dbg(&dev->dev, "of_flash device: %.8llx-%.8llx\n",
-		(unsigned long long)res.start, (unsigned long long)res.end);
+	count /= reg_tuple_size;
 
 	err = -ENOMEM;
-	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	info = kzalloc(sizeof(struct of_flash) +
+		       sizeof(struct of_flash_list) * count, GFP_KERNEL);
+	if (!info)
+		goto err_out;
+
+	mtd_list = kzalloc(sizeof(struct mtd_info) * count, GFP_KERNEL);
 	if (!info)
 		goto err_out;
 
 	dev_set_drvdata(&dev->dev, info);
 
-	err = -EBUSY;
-	info->res = request_mem_region(res.start, res.end - res.start + 1,
-				       dev_name(&dev->dev));
-	if (!info->res)
-		goto err_out;
+	for (i = 0; i < count; i++) {
+		err = -ENXIO;
+		if (of_address_to_resource(dp, i, &res)) {
+			dev_err(&dev->dev, "Can't get IO address from device"
+				" tree\n");
+			goto err_out;
+		}
 
-	err = -ENXIO;
-	width = of_get_property(dp, "bank-width", NULL);
-	if (!width) {
-		dev_err(&dev->dev, "Can't get bank width from device tree\n");
-		goto err_out;
-	}
+		dev_dbg(&dev->dev, "of_flash device: %.8llx-%.8llx\n",
+			(unsigned long long)res.start,
+			(unsigned long long)res.end);
+
+		err = -EBUSY;
+		info->list[i].res = request_mem_region(res.start, res.end -
+						       res.start + 1,
+						       dev_name(&dev->dev));
+		if (!info->list[i].res)
+			goto err_out;
+
+		err = -ENXIO;
+		width = of_get_property(dp, "bank-width", NULL);
+		if (!width) {
+			dev_err(&dev->dev, "Can't get bank width from device"
+				" tree\n");
+			goto err_out;
+		}
 
-	info->map.name = dev_name(&dev->dev);
-	info->map.phys = res.start;
-	info->map.size = res.end - res.start + 1;
-	info->map.bankwidth = *width;
+		info->list[i].map.name = dev_name(&dev->dev);
+		info->list[i].map.phys = res.start;
+		info->list[i].map.size = res.end - res.start + 1;
+		info->list[i].map.bankwidth = *width;
+
+		err = -ENOMEM;
+		info->list[i].map.virt = ioremap(info->list[i].map.phys,
+						 info->list[i].map.size);
+		if (!info->list[i].map.virt) {
+			dev_err(&dev->dev, "Failed to ioremap() flash"
+				" region\n");
+			goto err_out;
+		}
 
-	err = -ENOMEM;
-	info->map.virt = ioremap(info->map.phys, info->map.size);
-	if (!info->map.virt) {
-		dev_err(&dev->dev, "Failed to ioremap() flash region\n");
-		goto err_out;
-	}
+		simple_map_init(&info->list[i].map);
 
-	simple_map_init(&info->map);
+		if (probe_type) {
+			info->list[i].mtd = do_map_probe(probe_type,
+							 &info->list[i].map);
+		} else {
+			info->list[i].mtd = obsolete_probe(dev,
+							   &info->list[i].map);
+		}
+		mtd_list[i] = info->list[i].mtd;
 
-	if (probe_type)
-		info->mtd = do_map_probe(probe_type, &info->map);
-	else
-		info->mtd = obsolete_probe(dev, &info->map);
+		err = -ENXIO;
+		if (!info->list[i].mtd) {
+			dev_err(&dev->dev, "do_map_probe() failed\n");
+			goto err_out;
+		} else {
+			info->list_size++;
+		}
+		info->list[i].mtd->owner = THIS_MODULE;
+		info->list[i].mtd->dev.parent = &dev->dev;
+	}
 
-	err = -ENXIO;
-	if (!info->mtd) {
-		dev_err(&dev->dev, "do_map_probe() failed\n");
-		goto err_out;
+	err = 0;
+	if (info->list_size == 1) {
+		info->cmtd = info->list[0].mtd;
+	} else if (info->list_size > 1) {
+		/*
+		 * We detected multiple devices. Concatenate them together.
+		 */
+#ifdef CONFIG_MTD_CONCAT
+		info->cmtd = mtd_concat_create(mtd_list, info->list_size,
+					       dev_name(&dev->dev));
+		if (info->cmtd == NULL)
+			err = -ENXIO;
+#else
+		printk(KERN_ERR "physmap_of: multiple devices "
+		       "found but MTD concat support disabled.\n");
+		err = -ENXIO;
+#endif
 	}
-	info->mtd->owner = THIS_MODULE;
-	info->mtd->dev.parent = &dev->dev;
+	if (err)
+		goto err_out;
 
 #ifdef CONFIG_MTD_PARTITIONS
 	/* First look for RedBoot table or partitions on the command
 	 * line, these take precedence over device tree information */
-	err = parse_mtd_partitions(info->mtd, part_probe_types,
-	                           &info->parts, 0);
+	err = parse_mtd_partitions(info->cmtd, part_probe_types,
+				   &info->parts, 0);
 	if (err < 0)
 		return err;
 
@@ -244,15 +327,19 @@ static int __devinit of_flash_probe(struct of_device *dev,
 	}
 
 	if (err > 0)
-		add_mtd_partitions(info->mtd, info->parts, err);
+		add_mtd_partitions(info->cmtd, info->parts, err);
 	else
 #endif
-		add_mtd_device(info->mtd);
+		add_mtd_device(info->cmtd);
+
+	kfree(mtd_list);
 
 	return 0;
 
 err_out:
+	kfree(mtd_list);
 	of_flash_remove(dev);
+
 	return err;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From b0903ee8e4853eea35cb24cfb4dda1b60c7948dc Mon Sep 17 00:00:00 2001
From: Roel Kluin <roel.kluin@gmail.com>
Date: Mon, 8 Jun 2009 13:04:43 +0200
Subject: CRIS: assignment/is equal confusion

Somehow, the assignments of rw->lock got written as comparisons.
The effect probably was that the read-write spinlocks never locked.
However, since the locks are only used in CONFIG_SMP and SMP CRIS is
experimental, it has never been identified as a problem.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Acked-by: Jesper Nilsson <jesper.nilsson@axis.com>
---
 arch/cris/include/arch-v32/arch/spinlock.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/cris/include/arch-v32/arch/spinlock.h b/arch/cris/include/arch-v32/arch/spinlock.h
index 129756b96661..367a53ea10c5 100644
--- a/arch/cris/include/arch-v32/arch/spinlock.h
+++ b/arch/cris/include/arch-v32/arch/spinlock.h
@@ -78,7 +78,7 @@ static  inline void __raw_write_lock(raw_rwlock_t *rw)
 {
 	__raw_spin_lock(&rw->slock);
 	while (rw->lock != RW_LOCK_BIAS);
-	rw->lock == 0;
+	rw->lock = 0;
 	__raw_spin_unlock(&rw->slock);
 }
 
@@ -93,7 +93,7 @@ static  inline void __raw_write_unlock(raw_rwlock_t *rw)
 {
 	__raw_spin_lock(&rw->slock);
 	while (rw->lock != RW_LOCK_BIAS);
-	rw->lock == RW_LOCK_BIAS;
+	rw->lock = RW_LOCK_BIAS;
 	__raw_spin_unlock(&rw->slock);
 }
 
@@ -114,7 +114,7 @@ static  inline int __raw_write_trylock(raw_rwlock_t *rw)
 	int ret = 0;
 	__raw_spin_lock(&rw->slock);
 	if (rw->lock == RW_LOCK_BIAS) {
-		rw->lock == 0;
+		rw->lock = 0;
 		ret = 1;
 	}
 	__raw_spin_unlock(&rw->slock);
-- 
cgit v1.2.3-59-g8ed1b


From a4536b19df92adda215f6fc225b52dc4cee4cf83 Mon Sep 17 00:00:00 2001
From: Michel Pollet <buserror@gmail.com>
Date: Wed, 13 May 2009 16:54:13 +0100
Subject: [MTD] [NAND] S3C2410: Added a kerneldoc for s3c2410_nand_set

Converted the old comnent to kerneldoc.

Signed-off-by: Michel Pollet <buserror@gmail.com>
[ben-linux@fluff.org: updated subject, spello fix]
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c/include/plat/nand.h | 22 ++++++++++++----------
 1 file changed, 12 insertions(+), 10 deletions(-)

diff --git a/arch/arm/plat-s3c/include/plat/nand.h b/arch/arm/plat-s3c/include/plat/nand.h
index 985546750c6d..935651664969 100644
--- a/arch/arm/plat-s3c/include/plat/nand.h
+++ b/arch/arm/plat-s3c/include/plat/nand.h
@@ -10,17 +10,19 @@
  * published by the Free Software Foundation.
 */
 
-/* struct s3c2410_nand_set
+/**
+ * struct s3c2410_nand_set - define a set of one or more nand chips
+ * @disable_ecc:	Entirely disable ECC - Dangerous
+ * @nr_chips:		Number of chips in this set
+ * @nr_partitions:	Number of partitions pointed to by @partitions
+ * @name:		Name of set (optional)
+ * @nr_map:		Map for low-layer logical to physical chip numbers (option)
+ * @partitions:		The mtd partition list
  *
- * define an set of one or more nand chips registered with an unique mtd
- *
- * nr_chips	 = number of chips in this set
- * nr_partitions = number of partitions pointed to be partitoons (or zero)
- * name		 = name of set (optional)
- * nr_map	 = map for low-layer logical to physical chip numbers (option)
- * partitions	 = mtd partition list
-*/
-
+ * define a set of one or more nand chips registered with an unique mtd. Also
+ * allows to pass flag to the underlying NAND layer. 'disable_ecc' will trigger
+ * a warning at boot time.
+ */
 struct s3c2410_nand_set {
 	unsigned int		disable_ecc:1;
 
-- 
cgit v1.2.3-59-g8ed1b


From 9db41f9edcb87ae050fcb171c44be7f212728d54 Mon Sep 17 00:00:00 2001
From: Michel Pollet <buserror@gmail.com>
Date: Wed, 13 May 2009 16:54:14 +0100
Subject: [MTD] [NAND] S3C2410: Allow the machine code to get the BBT table
 from NAND

Added a flag to allow the machine code to tell the NAND
subsystem that it should try to pickup a BBT from the flash,
and also skip the NAND full scan at startup.

Signed-off-by: Michel Pollet <buserror@gmail.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c/include/plat/nand.h | 5 +++++
 drivers/mtd/nand/s3c2410.c            | 6 ++++++
 2 files changed, 11 insertions(+)

diff --git a/arch/arm/plat-s3c/include/plat/nand.h b/arch/arm/plat-s3c/include/plat/nand.h
index 935651664969..18f958801e64 100644
--- a/arch/arm/plat-s3c/include/plat/nand.h
+++ b/arch/arm/plat-s3c/include/plat/nand.h
@@ -13,6 +13,10 @@
 /**
  * struct s3c2410_nand_set - define a set of one or more nand chips
  * @disable_ecc:	Entirely disable ECC - Dangerous
+ * @flash_bbt: 		Openmoko u-boot can create a Bad Block Table
+ *			Setting this flag will allow the kernel to
+ *			look for it at boot time and also skip the NAND
+ *			scan.
  * @nr_chips:		Number of chips in this set
  * @nr_partitions:	Number of partitions pointed to by @partitions
  * @name:		Name of set (optional)
@@ -25,6 +29,7 @@
  */
 struct s3c2410_nand_set {
 	unsigned int		disable_ecc:1;
+	unsigned int		flash_bbt:1;
 
 	int			nr_chips;
 	int			nr_partitions;
diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index ef5665258968..d315b513db5c 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -845,6 +845,12 @@ static void s3c2410_nand_init_chip(struct s3c2410_nand_info *info,
 		dev_info(info->device, "NAND ECC UNKNOWN\n");
 		break;
 	}
+
+	/* If you use u-boot BBT creation code, specifying this flag will
+	 * let the kernel fish out the BBT from the NAND, and also skip the
+	 * full NAND scan that can take 1/2s or so. Little things... */
+	if (set->flash_bbt)
+		chip->options |= NAND_USE_FLASH_BBT | NAND_SKIP_BBTSCAN;
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From dea2aa6fd7d46c43c840ad77905f3c161d5bc59d Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sat, 30 May 2009 18:30:18 +0100
Subject: [MTD] [NAND] S3C2410: Deal with unaligned lengths in S3C2440 buffer
 read/write

Add code to deal with fractional lengths, as reported by
Werner Almesberger. Re-work of his original patch.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 drivers/mtd/nand/s3c2410.c | 22 ++++++++++++++++++++--
 1 file changed, 20 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index d315b513db5c..8a7f960a0df5 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -584,7 +584,16 @@ static void s3c2410_nand_read_buf(struct mtd_info *mtd, u_char *buf, int len)
 static void s3c2440_nand_read_buf(struct mtd_info *mtd, u_char *buf, int len)
 {
 	struct s3c2410_nand_info *info = s3c2410_nand_mtd_toinfo(mtd);
-	readsl(info->regs + S3C2440_NFDATA, buf, len / 4);
+
+	readsl(info->regs + S3C2440_NFDATA, buf, len >> 2);
+
+	/* cleanup if we've got less than a word to do */
+	if (len & 3) {
+		buf += len & ~3;
+
+		for (; len & 3; len--)
+			*buf++ = readb(info->regs + S3C2440_NFDATA);
+	}
 }
 
 static void s3c2410_nand_write_buf(struct mtd_info *mtd, const u_char *buf, int len)
@@ -596,7 +605,16 @@ static void s3c2410_nand_write_buf(struct mtd_info *mtd, const u_char *buf, int
 static void s3c2440_nand_write_buf(struct mtd_info *mtd, const u_char *buf, int len)
 {
 	struct s3c2410_nand_info *info = s3c2410_nand_mtd_toinfo(mtd);
-	writesl(info->regs + S3C2440_NFDATA, buf, len / 4);
+
+	writesl(info->regs + S3C2440_NFDATA, buf, len >> 2);
+
+	/* cleanup any fractional write */
+	if (len & 3) {
+		buf += len & ~3;
+
+		for (; len & 3; len--, buf++)
+			writeb(*buf, info->regs + S3C2440_NFDATA);
+	}
 }
 
 /* cpufreq driver support */
-- 
cgit v1.2.3-59-g8ed1b


From 947391cfbaa3b08558844c0b187bcd0223c3f660 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sat, 30 May 2009 18:34:16 +0100
Subject: [MTD] [NAND] S3C2410: Use DIV_ROUND_UP

Change to using DIV_ROUND_UP() in the timing calculation
instead of blindly doing result++

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 drivers/mtd/nand/s3c2410.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index 8a7f960a0df5..89b79051cc68 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -180,8 +180,7 @@ static int s3c_nand_calc_rate(int wanted, unsigned long clk, int max)
 {
 	int result;
 
-	result = (wanted * clk) / NS_IN_KHZ;
-	result++;
+	result = DIV_ROUND_UP((wanted * clk), NS_IN_KHZ);
 
 	pr_debug("result %d from %ld, %d\n", result, clk, wanted);
 
-- 
cgit v1.2.3-59-g8ed1b


From 9dbc090274668abe3fc9f3a5de490d7d412cd74a Mon Sep 17 00:00:00 2001
From: Peter Korsgaard <jacmet@sunsite.dk>
Date: Sun, 7 Jun 2009 06:04:23 -0700
Subject: mtd/nand: s3c6400 support for s3c2410 driver

Add s3c6400 support to the s3c2410 driver. The nand controller in
the s3c64xx devices is compatible with the one in the s3c2412, so
simply reuse that code.

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Acked-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/Kconfig   | 18 +++++++++---------
 drivers/mtd/nand/s3c2410.c |  3 +++
 2 files changed, 12 insertions(+), 9 deletions(-)

diff --git a/drivers/mtd/nand/Kconfig b/drivers/mtd/nand/Kconfig
index 3b3a21d1a6ba..1823212c6b46 100644
--- a/drivers/mtd/nand/Kconfig
+++ b/drivers/mtd/nand/Kconfig
@@ -145,27 +145,27 @@ config MTD_NAND_PPCHAMELEONEVB
 	  This enables the NAND flash driver on the PPChameleon EVB Board.
 
 config MTD_NAND_S3C2410
-	tristate "NAND Flash support for S3C2410/S3C2440 SoC"
-	depends on ARCH_S3C2410
+	tristate "NAND Flash support for Samsung S3C SoCs"
+	depends on ARCH_S3C2410 || ARCH_S3C64XX
 	help
-	  This enables the NAND flash controller on the S3C2410 and S3C2440
+	  This enables the NAND flash controller on the S3C24xx and S3C64xx
 	  SoCs
 
 	  No board specific support is done by this driver, each board
 	  must advertise a platform_device for the driver to attach.
 
 config MTD_NAND_S3C2410_DEBUG
-	bool "S3C2410 NAND driver debug"
+	bool "Samsung S3C NAND driver debug"
 	depends on MTD_NAND_S3C2410
 	help
-	  Enable debugging of the S3C2410 NAND driver
+	  Enable debugging of the S3C NAND driver
 
 config MTD_NAND_S3C2410_HWECC
-	bool "S3C2410 NAND Hardware ECC"
+	bool "Samsung S3C NAND Hardware ECC"
 	depends on MTD_NAND_S3C2410
 	help
-	  Enable the use of the S3C2410's internal ECC generator when
-	  using NAND. Early versions of the chip have had problems with
+	  Enable the use of the controller's internal ECC generator when
+	  using NAND. Early versions of the chips have had problems with
 	  incorrect ECC generation, and if using these, the default of
 	  software ECC is preferable.
 
@@ -177,7 +177,7 @@ config MTD_NAND_NDFC
 	 NDFC Nand Flash Controllers are integrated in IBM/AMCC's 4xx SoCs
 
 config MTD_NAND_S3C2410_CLKSTOP
-	bool "S3C2410 NAND IDLE clock stop"
+	bool "Samsung S3C NAND IDLE clock stop"
 	depends on MTD_NAND_S3C2410
 	default n
 	help
diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index 01a105eda3f4..11dc7e69c4fb 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -1111,6 +1111,9 @@ static struct platform_device_id s3c24xx_driver_ids[] = {
 	}, {
 		.name		= "s3c2412-nand",
 		.driver_data	= TYPE_S3C2412,
+	}, {
+		.name		= "s3c6400-nand",
+		.driver_data	= TYPE_S3C2412, /* compatible with 2412 */
 	},
 	{ }
 };
-- 
cgit v1.2.3-59-g8ed1b


From bfee1a4311702c9fdecd8264ffd1126fd0ce92fb Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@cam.org>
Date: Sun, 31 May 2009 22:25:40 -0400
Subject: mtd: orion_nand: use burst reads with double word accesses

This is not 8 times faster than byte access, but still around 60% faster.

Signed-off-by: Nicolas Pitre <nico@marvell.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/orion_nand.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/drivers/mtd/nand/orion_nand.c b/drivers/mtd/nand/orion_nand.c
index c2dfd3ea353d..7ad972229db4 100644
--- a/drivers/mtd/nand/orion_nand.c
+++ b/drivers/mtd/nand/orion_nand.c
@@ -47,6 +47,28 @@ static void orion_nand_cmd_ctrl(struct mtd_info *mtd, int cmd, unsigned int ctrl
 	writeb(cmd, nc->IO_ADDR_W + offs);
 }
 
+static void orion_nand_read_buf(struct mtd_info *mtd, uint8_t *buf, int len)
+{
+	struct nand_chip *chip = mtd->priv;
+	void __iomem *io_base = chip->IO_ADDR_R;
+	uint64_t *buf64;
+	int i = 0;
+
+	while (len && (unsigned long)buf & 7) {
+		*buf++ = readb(io_base);
+		len--;
+	}
+	buf64 = (uint64_t *)buf;
+	while (i < len/8) {
+		uint64_t x;
+		asm ("ldrd\t%0, [%1]" : "=r" (x) : "r" (io_base));
+		buf64[i++] = x;
+	}
+	i *= 8;
+	while (i < len)
+		buf[i++] = readb(io_base);
+}
+
 static int __init orion_nand_probe(struct platform_device *pdev)
 {
 	struct mtd_info *mtd;
@@ -83,6 +105,7 @@ static int __init orion_nand_probe(struct platform_device *pdev)
 	nc->priv = board;
 	nc->IO_ADDR_R = nc->IO_ADDR_W = io_base;
 	nc->cmd_ctrl = orion_nand_cmd_ctrl;
+	nc->read_buf = orion_nand_read_buf;
 	nc->ecc.mode = NAND_ECC_SOFT;
 
 	if (board->chip_delay)
-- 
cgit v1.2.3-59-g8ed1b


From ae9fb6e814ecede683bcd404910085cea3ab1260 Mon Sep 17 00:00:00 2001
From: Daniel Mack <daniel@caiaq.de>
Date: Tue, 28 Apr 2009 10:55:00 +0200
Subject: ds2760_battery: cleanups in ds2760_battery_probe()

Removed struct ds2760_platform_data which wasn't defined anywhere.
Indentation cleanups.

Signed-off-by: Daniel Mack <daniel@caiaq.de>
Cc: Szabolcs Gyurko <szabolcs.gyurko@tlt.hu>
Acked-by: Matt Reimer <mreimer@vpop.net>
Signed-off-by: Anton Vorontsov <cbouatmailru@gmail.com>
---
 drivers/power/ds2760_battery.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/drivers/power/ds2760_battery.c b/drivers/power/ds2760_battery.c
index a52d4a11652d..9331009090f1 100644
--- a/drivers/power/ds2760_battery.c
+++ b/drivers/power/ds2760_battery.c
@@ -344,7 +344,6 @@ static int ds2760_battery_probe(struct platform_device *pdev)
 {
 	int retval = 0;
 	struct ds2760_device_info *di;
-	struct ds2760_platform_data *pdata;
 
 	di = kzalloc(sizeof(*di), GFP_KERNEL);
 	if (!di) {
@@ -354,14 +353,13 @@ static int ds2760_battery_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, di);
 
-	pdata = pdev->dev.platform_data;
-	di->dev		= &pdev->dev;
-	di->w1_dev	     = pdev->dev.parent;
-	di->bat.name	   = dev_name(&pdev->dev);
-	di->bat.type	   = POWER_SUPPLY_TYPE_BATTERY;
-	di->bat.properties     = ds2760_battery_props;
-	di->bat.num_properties = ARRAY_SIZE(ds2760_battery_props);
-	di->bat.get_property   = ds2760_battery_get_property;
+	di->dev			= &pdev->dev;
+	di->w1_dev		= pdev->dev.parent;
+	di->bat.name		= dev_name(&pdev->dev);
+	di->bat.type		= POWER_SUPPLY_TYPE_BATTERY;
+	di->bat.properties	= ds2760_battery_props;
+	di->bat.num_properties	= ARRAY_SIZE(ds2760_battery_props);
+	di->bat.get_property	= ds2760_battery_get_property;
 	di->bat.external_power_changed =
 				  ds2760_battery_external_power_changed;
 
-- 
cgit v1.2.3-59-g8ed1b


From 0b47b5703b1cc6c3aa89663ac70e28dadedf6ccc Mon Sep 17 00:00:00 2001
From: Daniel Mack <daniel@caiaq.de>
Date: Tue, 28 Apr 2009 10:55:01 +0200
Subject: w1: ds2760: add support for EEPROM read and write

In order to modify the DS2762's status registers and to add support for
sleep mode, there is need for functions to write the internal EEPROM.

Signed-off-by: Daniel Mack <daniel@caiaq.de>
Acked-by: Matt Reimer <mreimer@vpop.net>
Acked-by: Szabolcs Gyurko <szabolcs.gyurko@tlt.hu>
Acked-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Anton Vorontsov <cbouatmailru@gmail.com>
---
 drivers/w1/slaves/w1_ds2760.c | 30 ++++++++++++++++++++++++++++++
 drivers/w1/slaves/w1_ds2760.h |  2 ++
 2 files changed, 32 insertions(+)

diff --git a/drivers/w1/slaves/w1_ds2760.c b/drivers/w1/slaves/w1_ds2760.c
index 1f09d4e4144c..59f708efe25f 100644
--- a/drivers/w1/slaves/w1_ds2760.c
+++ b/drivers/w1/slaves/w1_ds2760.c
@@ -68,6 +68,34 @@ int w1_ds2760_write(struct device *dev, char *buf, int addr, size_t count)
 	return w1_ds2760_io(dev, buf, addr, count, 1);
 }
 
+static int w1_ds2760_eeprom_cmd(struct device *dev, int addr, int cmd)
+{
+	struct w1_slave *sl = container_of(dev, struct w1_slave, dev);
+
+	if (!dev)
+		return -EINVAL;
+
+	mutex_lock(&sl->master->mutex);
+
+	if (w1_reset_select_slave(sl) == 0) {
+		w1_write_8(sl->master, cmd);
+		w1_write_8(sl->master, addr);
+	}
+
+	mutex_unlock(&sl->master->mutex);
+	return 0;
+}
+
+int w1_ds2760_store_eeprom(struct device *dev, int addr)
+{
+	return w1_ds2760_eeprom_cmd(dev, addr, W1_DS2760_COPY_DATA);
+}
+
+int w1_ds2760_recall_eeprom(struct device *dev, int addr)
+{
+	return w1_ds2760_eeprom_cmd(dev, addr, W1_DS2760_RECALL_DATA);
+}
+
 static ssize_t w1_ds2760_read_bin(struct kobject *kobj,
 				  struct bin_attribute *bin_attr,
 				  char *buf, loff_t off, size_t count)
@@ -200,6 +228,8 @@ static void __exit w1_ds2760_exit(void)
 
 EXPORT_SYMBOL(w1_ds2760_read);
 EXPORT_SYMBOL(w1_ds2760_write);
+EXPORT_SYMBOL(w1_ds2760_store_eeprom);
+EXPORT_SYMBOL(w1_ds2760_recall_eeprom);
 
 module_init(w1_ds2760_init);
 module_exit(w1_ds2760_exit);
diff --git a/drivers/w1/slaves/w1_ds2760.h b/drivers/w1/slaves/w1_ds2760.h
index f1302429cb02..ea39419172a6 100644
--- a/drivers/w1/slaves/w1_ds2760.h
+++ b/drivers/w1/slaves/w1_ds2760.h
@@ -46,5 +46,7 @@ extern int w1_ds2760_read(struct device *dev, char *buf, int addr,
 			  size_t count);
 extern int w1_ds2760_write(struct device *dev, char *buf, int addr,
 			   size_t count);
+extern int w1_ds2760_store_eeprom(struct device *dev, int addr);
+extern int w1_ds2760_recall_eeprom(struct device *dev, int addr);
 
 #endif /* !__w1_ds2760_h__ */
-- 
cgit v1.2.3-59-g8ed1b


From cef437e3a9b6d229d4ed3730cde047007267df6d Mon Sep 17 00:00:00 2001
From: Daniel Mack <daniel@caiaq.de>
Date: Tue, 28 Apr 2009 10:55:02 +0200
Subject: w1: ds2760_battery: add support for sleep mode feature

This adds support for ds2760's sleep mode feature. With this feature
enabled, the chip enters a deep sleep mode and disconnects from the
battery when the w1 line is held down for more than 2 seconds.

This new behaviour can be switched on and off using a new module
parameter.

Signed-off-by: Daniel Mack <daniel@caiaq.de>
Cc: Szabolcs Gyurko <szabolcs.gyurko@tlt.hu>
Acked-by: Matt Reimer <mreimer@vpop.net>
Acked-by: Evgeniy Polyakov <zbr@ioremap.net>
Signed-off-by: Anton Vorontsov <cbouatmailru@gmail.com>
---
 drivers/power/ds2760_battery.c | 26 ++++++++++++++++++++++++++
 drivers/w1/slaves/w1_ds2760.h  |  5 +++++
 2 files changed, 31 insertions(+)

diff --git a/drivers/power/ds2760_battery.c b/drivers/power/ds2760_battery.c
index 9331009090f1..520b5c49ff30 100644
--- a/drivers/power/ds2760_battery.c
+++ b/drivers/power/ds2760_battery.c
@@ -62,6 +62,10 @@ static unsigned int cache_time = 1000;
 module_param(cache_time, uint, 0644);
 MODULE_PARM_DESC(cache_time, "cache time in milliseconds");
 
+static unsigned int pmod_enabled;
+module_param(pmod_enabled, bool, 0644);
+MODULE_PARM_DESC(pmod_enabled, "PMOD enable bit");
+
 /* Some batteries have their rated capacity stored a N * 10 mAh, while
  * others use an index into this table. */
 static int rated_capacities[] = {
@@ -259,6 +263,17 @@ static void ds2760_battery_update_status(struct ds2760_device_info *di)
 		power_supply_changed(&di->bat);
 }
 
+static void ds2760_battery_write_status(struct ds2760_device_info *di,
+					char status)
+{
+	if (status == di->raw[DS2760_STATUS_REG])
+		return;
+
+	w1_ds2760_write(di->w1_dev, &status, DS2760_STATUS_WRITE_REG, 1);
+	w1_ds2760_store_eeprom(di->w1_dev, DS2760_EEPROM_BLOCK1);
+	w1_ds2760_recall_eeprom(di->w1_dev, DS2760_EEPROM_BLOCK1);
+}
+
 static void ds2760_battery_work(struct work_struct *work)
 {
 	struct ds2760_device_info *di = container_of(work,
@@ -342,6 +357,7 @@ static enum power_supply_property ds2760_battery_props[] = {
 
 static int ds2760_battery_probe(struct platform_device *pdev)
 {
+	char status;
 	int retval = 0;
 	struct ds2760_device_info *di;
 
@@ -371,6 +387,16 @@ static int ds2760_battery_probe(struct platform_device *pdev)
 		goto batt_failed;
 	}
 
+	/* enable sleep mode feature */
+	ds2760_battery_read_status(di);
+	status = di->raw[DS2760_STATUS_REG];
+	if (pmod_enabled)
+		status |= DS2760_STATUS_PMOD;
+	else
+		status &= ~DS2760_STATUS_PMOD;
+
+	ds2760_battery_write_status(di, status);
+
 	INIT_DELAYED_WORK(&di->monitor_work, ds2760_battery_work);
 	di->monitor_wqueue = create_singlethread_workqueue(dev_name(&pdev->dev));
 	if (!di->monitor_wqueue) {
diff --git a/drivers/w1/slaves/w1_ds2760.h b/drivers/w1/slaves/w1_ds2760.h
index ea39419172a6..58e774141568 100644
--- a/drivers/w1/slaves/w1_ds2760.h
+++ b/drivers/w1/slaves/w1_ds2760.h
@@ -25,6 +25,10 @@
 
 #define DS2760_PROTECTION_REG		0x00
 #define DS2760_STATUS_REG		0x01
+	#define DS2760_STATUS_IE	(1 << 2)
+	#define DS2760_STATUS_SWEN	(1 << 3)
+	#define DS2760_STATUS_RNAOP	(1 << 4)
+	#define DS2760_STATUS_PMOD	(1 << 5)
 #define DS2760_EEPROM_REG		0x07
 #define DS2760_SPECIAL_FEATURE_REG	0x08
 #define DS2760_VOLTAGE_MSB		0x0c
@@ -38,6 +42,7 @@
 #define DS2760_EEPROM_BLOCK0		0x20
 #define DS2760_ACTIVE_FULL		0x20
 #define DS2760_EEPROM_BLOCK1		0x30
+#define DS2760_STATUS_WRITE_REG		0x31
 #define DS2760_RATED_CAPACITY		0x32
 #define DS2760_CURRENT_OFFSET_BIAS	0x33
 #define DS2760_ACTIVE_EMPTY		0x3b
-- 
cgit v1.2.3-59-g8ed1b


From c6f4a42de60b981dd210de01cd3e575835e3158e Mon Sep 17 00:00:00 2001
From: Minkyu Kang <mk7.kang@samsung.com>
Date: Fri, 5 Jun 2009 15:33:04 +0900
Subject: Add MAX17040 Fuel Gauge driver

The MAX17040 is a I2C interfaced Fuel Gauge systems for lithium-ion
batteries This patch adds support the MAX17040 Fuel Gauge

Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Signed-off-by: Anton Vorontsov <cbouatmailru@gmail.com>
---
 drivers/power/Kconfig            |   8 +
 drivers/power/Makefile           |   3 +-
 drivers/power/max17040_battery.c | 309 +++++++++++++++++++++++++++++++++++++++
 include/linux/max17040_battery.h |  19 +++
 4 files changed, 338 insertions(+), 1 deletion(-)
 create mode 100644 drivers/power/max17040_battery.c
 create mode 100644 include/linux/max17040_battery.h

diff --git a/drivers/power/Kconfig b/drivers/power/Kconfig
index 33da1127992a..7eda34838bfe 100644
--- a/drivers/power/Kconfig
+++ b/drivers/power/Kconfig
@@ -82,6 +82,14 @@ config BATTERY_DA9030
 	  Say Y here to enable support for batteries charger integrated into
 	  DA9030 PMIC.
 
+config BATTERY_MAX17040
+	tristate "Maxim MAX17040 Fuel Gauge"
+	depends on I2C
+	help
+	  MAX17040 is fuel-gauge systems for lithium-ion (Li+) batteries
+	  in handheld and portable equipment. The MAX17040 is configured
+	  to operate with a single lithium cell
+
 config CHARGER_PCF50633
 	tristate "NXP PCF50633 MBC"
 	depends on MFD_PCF50633
diff --git a/drivers/power/Makefile b/drivers/power/Makefile
index 2fcf41d13e5c..daf3179689aa 100644
--- a/drivers/power/Makefile
+++ b/drivers/power/Makefile
@@ -25,4 +25,5 @@ obj-$(CONFIG_BATTERY_TOSA)	+= tosa_battery.o
 obj-$(CONFIG_BATTERY_WM97XX)	+= wm97xx_battery.o
 obj-$(CONFIG_BATTERY_BQ27x00)	+= bq27x00_battery.o
 obj-$(CONFIG_BATTERY_DA9030)	+= da9030_battery.o
-obj-$(CONFIG_CHARGER_PCF50633)	+= pcf50633-charger.o
\ No newline at end of file
+obj-$(CONFIG_BATTERY_MAX17040)	+= max17040_battery.o
+obj-$(CONFIG_CHARGER_PCF50633)	+= pcf50633-charger.o
diff --git a/drivers/power/max17040_battery.c b/drivers/power/max17040_battery.c
new file mode 100644
index 000000000000..87b98bf27ae1
--- /dev/null
+++ b/drivers/power/max17040_battery.c
@@ -0,0 +1,309 @@
+/*
+ *  max17040_battery.c
+ *  fuel-gauge systems for lithium-ion (Li+) batteries
+ *
+ *  Copyright (C) 2009 Samsung Electronics
+ *  Minkyu Kang <mk7.kang@samsung.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/mutex.h>
+#include <linux/err.h>
+#include <linux/i2c.h>
+#include <linux/delay.h>
+#include <linux/power_supply.h>
+#include <linux/max17040_battery.h>
+
+#define MAX17040_VCELL_MSB	0x02
+#define MAX17040_VCELL_LSB	0x03
+#define MAX17040_SOC_MSB	0x04
+#define MAX17040_SOC_LSB	0x05
+#define MAX17040_MODE_MSB	0x06
+#define MAX17040_MODE_LSB	0x07
+#define MAX17040_VER_MSB	0x08
+#define MAX17040_VER_LSB	0x09
+#define MAX17040_RCOMP_MSB	0x0C
+#define MAX17040_RCOMP_LSB	0x0D
+#define MAX17040_CMD_MSB	0xFE
+#define MAX17040_CMD_LSB	0xFF
+
+#define MAX17040_DELAY		1000
+#define MAX17040_BATTERY_FULL	95
+
+struct max17040_chip {
+	struct i2c_client		*client;
+	struct delayed_work		work;
+	struct power_supply		battery;
+	struct max17040_platform_data	*pdata;
+
+	/* State Of Connect */
+	int online;
+	/* battery voltage */
+	int vcell;
+	/* battery capacity */
+	int soc;
+	/* State Of Charge */
+	int status;
+};
+
+static int max17040_get_property(struct power_supply *psy,
+			    enum power_supply_property psp,
+			    union power_supply_propval *val)
+{
+	struct max17040_chip *chip = container_of(psy,
+				struct max17040_chip, battery);
+
+	switch (psp) {
+	case POWER_SUPPLY_PROP_STATUS:
+		val->intval = chip->status;
+		break;
+	case POWER_SUPPLY_PROP_ONLINE:
+		val->intval = chip->online;
+		break;
+	case POWER_SUPPLY_PROP_VOLTAGE_NOW:
+		val->intval = chip->vcell;
+		break;
+	case POWER_SUPPLY_PROP_CAPACITY:
+		val->intval = chip->soc;
+		break;
+	default:
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int max17040_write_reg(struct i2c_client *client, int reg, u8 value)
+{
+	int ret;
+
+	ret = i2c_smbus_write_byte_data(client, reg, value);
+
+	if (ret < 0)
+		dev_err(&client->dev, "%s: err %d\n", __func__, ret);
+
+	return ret;
+}
+
+static int max17040_read_reg(struct i2c_client *client, int reg)
+{
+	int ret;
+
+	ret = i2c_smbus_read_byte_data(client, reg);
+
+	if (ret < 0)
+		dev_err(&client->dev, "%s: err %d\n", __func__, ret);
+
+	return ret;
+}
+
+static void max17040_reset(struct i2c_client *client)
+{
+	max17040_write_reg(client, MAX17040_CMD_MSB, 0x54);
+	max17040_write_reg(client, MAX17040_CMD_LSB, 0x00);
+}
+
+static void max17040_get_vcell(struct i2c_client *client)
+{
+	struct max17040_chip *chip = i2c_get_clientdata(client);
+	u8 msb;
+	u8 lsb;
+
+	msb = max17040_read_reg(client, MAX17040_VCELL_MSB);
+	lsb = max17040_read_reg(client, MAX17040_VCELL_LSB);
+
+	chip->vcell = (msb << 4) + (lsb >> 4);
+}
+
+static void max17040_get_soc(struct i2c_client *client)
+{
+	struct max17040_chip *chip = i2c_get_clientdata(client);
+	u8 msb;
+	u8 lsb;
+
+	msb = max17040_read_reg(client, MAX17040_SOC_MSB);
+	lsb = max17040_read_reg(client, MAX17040_SOC_LSB);
+
+	chip->soc = msb;
+}
+
+static void max17040_get_version(struct i2c_client *client)
+{
+	u8 msb;
+	u8 lsb;
+
+	msb = max17040_read_reg(client, MAX17040_VER_MSB);
+	lsb = max17040_read_reg(client, MAX17040_VER_LSB);
+
+	dev_info(&client->dev, "MAX17040 Fuel-Gauge Ver %d%d\n", msb, lsb);
+}
+
+static void max17040_get_online(struct i2c_client *client)
+{
+	struct max17040_chip *chip = i2c_get_clientdata(client);
+
+	if (chip->pdata->battery_online)
+		chip->online = chip->pdata->battery_online();
+	else
+		chip->online = 1;
+}
+
+static void max17040_get_status(struct i2c_client *client)
+{
+	struct max17040_chip *chip = i2c_get_clientdata(client);
+
+	if (!chip->pdata->charger_online || !chip->pdata->charger_enable) {
+		chip->status = POWER_SUPPLY_STATUS_UNKNOWN;
+		return;
+	}
+
+	if (chip->pdata->charger_online()) {
+		if (chip->pdata->charger_enable())
+			chip->status = POWER_SUPPLY_STATUS_CHARGING;
+		else
+			chip->status = POWER_SUPPLY_STATUS_NOT_CHARGING;
+	} else {
+		chip->status = POWER_SUPPLY_STATUS_DISCHARGING;
+	}
+
+	if (chip->soc > MAX17040_BATTERY_FULL)
+		chip->status = POWER_SUPPLY_STATUS_FULL;
+}
+
+static void max17040_work(struct work_struct *work)
+{
+	struct max17040_chip *chip;
+
+	chip = container_of(work, struct max17040_chip, work.work);
+
+	max17040_get_vcell(chip->client);
+	max17040_get_soc(chip->client);
+	max17040_get_online(chip->client);
+	max17040_get_status(chip->client);
+
+	schedule_delayed_work(&chip->work, MAX17040_DELAY);
+}
+
+static enum power_supply_property max17040_battery_props[] = {
+	POWER_SUPPLY_PROP_STATUS,
+	POWER_SUPPLY_PROP_ONLINE,
+	POWER_SUPPLY_PROP_VOLTAGE_NOW,
+	POWER_SUPPLY_PROP_CAPACITY,
+};
+
+static int __devinit max17040_probe(struct i2c_client *client,
+			const struct i2c_device_id *id)
+{
+	struct i2c_adapter *adapter = to_i2c_adapter(client->dev.parent);
+	struct max17040_chip *chip;
+	int ret;
+
+	if (!i2c_check_functionality(adapter, I2C_FUNC_SMBUS_BYTE))
+		return -EIO;
+
+	chip = kzalloc(sizeof(*chip), GFP_KERNEL);
+	if (!chip)
+		return -ENOMEM;
+
+	chip->client = client;
+	chip->pdata = client->dev.platform_data;
+
+	i2c_set_clientdata(client, chip);
+
+	chip->battery.name		= "battery";
+	chip->battery.type		= POWER_SUPPLY_TYPE_BATTERY;
+	chip->battery.get_property	= max17040_get_property;
+	chip->battery.properties	= max17040_battery_props;
+	chip->battery.num_properties	= ARRAY_SIZE(max17040_battery_props);
+
+	ret = power_supply_register(&client->dev, &chip->battery);
+	if (ret) {
+		dev_err(&client->dev, "failed: power supply register\n");
+		i2c_set_clientdata(client, NULL);
+		kfree(chip);
+		return ret;
+	}
+
+	max17040_reset(client);
+	max17040_get_version(client);
+
+	INIT_DELAYED_WORK_DEFERRABLE(&chip->work, max17040_work);
+	schedule_delayed_work(&chip->work, MAX17040_DELAY);
+
+	return 0;
+}
+
+static int __devexit max17040_remove(struct i2c_client *client)
+{
+	struct max17040_chip *chip = i2c_get_clientdata(client);
+
+	power_supply_unregister(&chip->battery);
+	cancel_delayed_work(&chip->work);
+	i2c_set_clientdata(client, NULL);
+	kfree(chip);
+	return 0;
+}
+
+#ifdef CONFIG_PM
+
+static int max17040_suspend(struct i2c_client *client,
+		pm_message_t state)
+{
+	struct max17040_chip *chip = i2c_get_clientdata(client);
+
+	cancel_delayed_work(&chip->work);
+	return 0;
+}
+
+static int max17040_resume(struct i2c_client *client)
+{
+	struct max17040_chip *chip = i2c_get_clientdata(client);
+
+	schedule_delayed_work(&chip->work, MAX17040_DELAY);
+	return 0;
+}
+
+#else
+
+#define max17040_suspend NULL
+#define max17040_resume NULL
+
+#endif /* CONFIG_PM */
+
+static const struct i2c_device_id max17040_id[] = {
+	{ "max17040", 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(i2c, max17040_id);
+
+static struct i2c_driver max17040_i2c_driver = {
+	.driver	= {
+		.name	= "max17040",
+	},
+	.probe		= max17040_probe,
+	.remove		= __devexit_p(max17040_remove),
+	.suspend	= max17040_suspend,
+	.resume		= max17040_resume,
+	.id_table	= max17040_id,
+};
+
+static int __init max17040_init(void)
+{
+	return i2c_add_driver(&max17040_i2c_driver);
+}
+module_init(max17040_init);
+
+static void __exit max17040_exit(void)
+{
+	i2c_del_driver(&max17040_i2c_driver);
+}
+module_exit(max17040_exit);
+
+MODULE_AUTHOR("Minkyu Kang <mk7.kang@samsung.com>");
+MODULE_DESCRIPTION("MAX17040 Fuel Gauge");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/max17040_battery.h b/include/linux/max17040_battery.h
new file mode 100644
index 000000000000..ad97b06cf930
--- /dev/null
+++ b/include/linux/max17040_battery.h
@@ -0,0 +1,19 @@
+/*
+ *  Copyright (C) 2009 Samsung Electronics
+ *  Minkyu Kang <mk7.kang@samsung.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __MAX17040_BATTERY_H_
+#define __MAX17040_BATTERY_H_
+
+struct max17040_platform_data {
+	int (*battery_online)(void);
+	int (*charger_online)(void);
+	int (*charger_enable)(void);
+};
+
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From a35d01a5d2ac533edab94a8e3b6749ab213c91c5 Mon Sep 17 00:00:00 2001
From: Mike Rapoport <mike@compulab.co.il>
Date: Tue, 9 Jun 2009 01:09:45 +0400
Subject: da9030_battery: Fix race between event handler and monitor

There are cases when charging monitor and the event handler try to
change the charger state simultaneously. For instance, a charger is
connected to the system, there's the detection event and the event
handler tries to enable charging. It is possible that the periodic
charging monitor runs at the same time and it still thinks there's
no external charger. So it tries to disable the charging. As the
result, even if the conditions necessary to charge the battery hold,
there will be no actual charging.

The patch changes the event handler so that instead of enabling/
disabling the charger immediately it would rather make the monitor
run. The monitor code then decides what should be the charger state.

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Anton Vorontsov <cbouatmailru@gmail.com>
---
 drivers/power/da9030_battery.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)

diff --git a/drivers/power/da9030_battery.c b/drivers/power/da9030_battery.c
index 1662bb0f23a5..3364198134a1 100644
--- a/drivers/power/da9030_battery.c
+++ b/drivers/power/da9030_battery.c
@@ -22,8 +22,6 @@
 #include <linux/debugfs.h>
 #include <linux/seq_file.h>
 
-#define DA9030_STATUS_CHDET	(1 << 3)
-
 #define DA9030_FAULT_LOG		0x0a
 #define DA9030_FAULT_LOG_OVER_TEMP	(1 << 7)
 #define DA9030_FAULT_LOG_VBAT_OVER	(1 << 4)
@@ -244,6 +242,8 @@ static void da9030_set_charge(struct da9030_charger *charger, int on)
 	}
 
 	da903x_write(charger->master, DA9030_CHARGE_CONTROL, val);
+
+	power_supply_changed(&charger->psy);
 }
 
 static void da9030_charger_check_state(struct da9030_charger *charger)
@@ -258,6 +258,12 @@ static void da9030_charger_check_state(struct da9030_charger *charger)
 			da9030_set_charge(charger, 1);
 		}
 	} else {
+		/* Charger has been pulled out */
+		if (!charger->chdet) {
+			da9030_set_charge(charger, 0);
+			return;
+		}
+
 		if (charger->adc.vbat_res >=
 		    charger->thresholds.vbat_charge_stop) {
 			da9030_set_charge(charger, 0);
@@ -395,13 +401,11 @@ static int da9030_battery_event(struct notifier_block *nb, unsigned long event,
 {
 	struct da9030_charger *charger =
 		container_of(nb, struct da9030_charger, nb);
-	int status;
 
 	switch (event) {
 	case DA9030_EVENT_CHDET:
-		status = da903x_query_status(charger->master,
-					     DA9030_STATUS_CHDET);
-		da9030_set_charge(charger, status);
+		cancel_delayed_work_sync(&charger->work);
+		schedule_work(&charger->work.work);
 		break;
 	case DA9030_EVENT_VBATMON:
 		da9030_battery_vbat_event(charger);
@@ -565,7 +569,8 @@ static int da9030_battery_remove(struct platform_device *dev)
 	da903x_unregister_notifier(charger->master, &charger->nb,
 				   DA9030_EVENT_CHDET | DA9030_EVENT_VBATMON |
 				   DA9030_EVENT_CHIOVER | DA9030_EVENT_TBAT);
-	cancel_delayed_work(&charger->work);
+	cancel_delayed_work_sync(&charger->work);
+	da9030_set_charge(charger, 0);
 	power_supply_unregister(&charger->psy);
 
 	kfree(charger);
-- 
cgit v1.2.3-59-g8ed1b


From 272023df26da2668ecc3937f8eeb48c8683b64fa Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Tue, 9 Jun 2009 14:31:15 +0100
Subject: mtd: nand: Fix memory leak on txx9ndfmc probe failure.

Commit 81933046ef2a615031c46171013bde2c5225ee69 ('mtd: Fix handling of
mtdname in txx9ndfmc.c') introduced a potential memory leak. The
'mtdname' member of the private data structure is now allocated
separately, but was not freed on certain error paths.

Fix that, and make things simpler by _always_ allocating it separately
so that we don't need 'if (mtdname != dev_name()) kfree(mtdname);'...
which gets ugly now that we're doing it more than once, and more likely
that we'll get it wrong some time.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/txx9ndfmc.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/mtd/nand/txx9ndfmc.c b/drivers/mtd/nand/txx9ndfmc.c
index 5f919e63b29b..488088eff2ca 100644
--- a/drivers/mtd/nand/txx9ndfmc.c
+++ b/drivers/mtd/nand/txx9ndfmc.c
@@ -336,20 +336,21 @@ static int __init txx9ndfmc_probe(struct platform_device *dev)
 			txx9_priv->cs = i;
 			txx9_priv->mtdname = kasprintf(GFP_KERNEL, "%s.%u",
 						       dev_name(&dev->dev), i);
-			if (!txx9_priv->mtdname) {
-				kfree(txx9_priv);
-				dev_err(&dev->dev,
-					"Unable to allocate TXx9 NDFMC MTD device name.\n");
-				continue;
-			}
 		} else {
 			txx9_priv->cs = -1;
-			txx9_priv->mtdname = dev_name(&dev->dev);
+			txx9_priv->mtdname = kstrdup(dev_name(&dev->dev),
+						     GFP_KERNEL);
+		}
+		if (!txx9_priv->mtdname) {
+			kfree(txx9_priv);
+			dev_err(&dev->dev, "Unable to allocate MTD name.\n");
+			continue;
 		}
 		if (plat->wide_mask & (1 << i))
 			chip->options |= NAND_BUSWIDTH_16;
 
 		if (nand_scan(mtd, 1)) {
+			kfree(txx9_priv->mtdname);
 			kfree(txx9_priv);
 			continue;
 		}
@@ -391,8 +392,7 @@ static int __exit txx9ndfmc_remove(struct platform_device *dev)
 		kfree(drvdata->parts[i]);
 #endif
 		del_mtd_device(mtd);
-		if (txx9_priv->mtdname != dev_name(&dev->dev))
-			kfree(txx9_priv->mtdname);
+		kfree(txx9_priv->mtdname);
 		kfree(txx9_priv);
 	}
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 7f2ff23db1de53ea8695bb4a7c1cfab88886e3fd Mon Sep 17 00:00:00 2001
From: Jesper Nilsson <jesper.nilsson@axis.com>
Date: Wed, 10 Jun 2009 15:06:55 +0200
Subject: CRISv32: Add arch optimized strcmp.

Add an optimized strcmp for CRISv32. This improves strcmp performance
with about 25% when comparing a 55 character string with itself.

Signed-off-by: Edgar Iglesias <edgar@axis.com>
Acked-by: Jesper Nilsson <jesper.nilsson@axis.com>
---
 arch/cris/arch-v32/lib/Makefile |  2 +-
 arch/cris/arch-v32/lib/strcmp.S | 21 +++++++++++++++++++++
 arch/cris/include/asm/string.h  |  6 ++++++
 3 files changed, 28 insertions(+), 1 deletion(-)
 create mode 100644 arch/cris/arch-v32/lib/strcmp.S

diff --git a/arch/cris/arch-v32/lib/Makefile b/arch/cris/arch-v32/lib/Makefile
index eb4aad1f1158..dd296b9db034 100644
--- a/arch/cris/arch-v32/lib/Makefile
+++ b/arch/cris/arch-v32/lib/Makefile
@@ -3,5 +3,5 @@
 #
 
 lib-y  = checksum.o checksumcopy.o string.o usercopy.o memset.o \
-	csumcpfruser.o spinlock.o delay.o
+	csumcpfruser.o spinlock.o delay.o strcmp.o
 
diff --git a/arch/cris/arch-v32/lib/strcmp.S b/arch/cris/arch-v32/lib/strcmp.S
new file mode 100644
index 000000000000..8f7a1ee62591
--- /dev/null
+++ b/arch/cris/arch-v32/lib/strcmp.S
@@ -0,0 +1,21 @@
+; strcmp.S -- CRISv32 version.
+; Copyright (C) 2008 AXIS Communications AB
+; Written by Edgar E. Iglesias
+;
+; This source code is licensed under the GNU General Public License,
+; Version 2.  See the file COPYING for more details.
+
+	.global	strcmp
+	.type	strcmp,@function
+strcmp:
+1:
+	move.b	[$r10+], $r12
+	seq	$r13
+	sub.b	[$r11+], $r12
+	or.b	$r12, $r13
+	beq	1b
+	nop
+
+	ret
+	movs.b	$r12, $r10
+	.size	strcmp, . - strcmp
diff --git a/arch/cris/include/asm/string.h b/arch/cris/include/asm/string.h
index 691190e99a27..d5db39f9eea1 100644
--- a/arch/cris/include/asm/string.h
+++ b/arch/cris/include/asm/string.h
@@ -11,4 +11,10 @@ extern void *memcpy(void *, const void *, size_t);
 #define __HAVE_ARCH_MEMSET
 extern void *memset(void *, int, size_t);
 
+#ifdef CONFIG_ETRAX_ARCH_V32
+/* For v32 we provide strcmp.  */
+#define __HAVE_ARCH_STRCMP
+extern int strcmp(const char *s1, const char *s2);
+#endif
+
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 91a120d03fd901fc8b95e85af7903358c5862d65 Mon Sep 17 00:00:00 2001
From: Jesper Nilsson <jesper.nilsson@axis.com>
Date: Thu, 11 Jun 2009 19:09:00 +0200
Subject: CRISv32: Fix potential null reference in cryptocop driver.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The code didn't test the pointer to the newly allocated
memory, but a parameter sent in as value.
Since the input parameter was most often set, the code
would have used a null pointer if the kmalloc failed.
If the input parameter was not set, the code would
leak the allocated buffer.

http://bugzilla.kernel.org/show_bug.cgi?id=11363

Reported-by: Daniel Marjamäki <danielm77@spray.se>
Signed-off-by: Jesper Nilsson <jesper.nilsson@axis.com>
---
 arch/cris/arch-v32/drivers/cryptocop.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/cris/arch-v32/drivers/cryptocop.c b/arch/cris/arch-v32/drivers/cryptocop.c
index 67c61ea86813..fd529a0ec758 100644
--- a/arch/cris/arch-v32/drivers/cryptocop.c
+++ b/arch/cris/arch-v32/drivers/cryptocop.c
@@ -1395,7 +1395,7 @@ static int create_md5_pad(int alloc_flag, unsigned long long hashed_length, char
 	if (padlen < MD5_MIN_PAD_LENGTH) padlen += MD5_BLOCK_LENGTH;
 
 	p = kmalloc(padlen, alloc_flag);
-	if (!pad) return -ENOMEM;
+	if (!p) return -ENOMEM;
 
 	*p = 0x80;
 	memset(p+1, 0, padlen - 1);
@@ -1427,7 +1427,7 @@ static int create_sha1_pad(int alloc_flag, unsigned long long hashed_length, cha
 	if (padlen < SHA1_MIN_PAD_LENGTH) padlen += SHA1_BLOCK_LENGTH;
 
 	p = kmalloc(padlen, alloc_flag);
-	if (!pad) return -ENOMEM;
+	if (!p) return -ENOMEM;
 
 	*p = 0x80;
 	memset(p+1, 0, padlen - 1);
-- 
cgit v1.2.3-59-g8ed1b


From 768c31495abe49227a96261b5aa19f34b30882b1 Mon Sep 17 00:00:00 2001
From: Jesper Nilsson <jesper.nilsson@axis.com>
Date: Wed, 10 Jun 2009 11:45:47 +0200
Subject: CRISv32: irq.c - Move end brace outside #endif

The end brace for a larger for statement was placed inside the #else
part of #ifdef TIMER_VECT1. However, for all current chips, the
define TIMER_VECT1 is always unset, and the error was never triggered.

Move the brace down below the #endif.

Fixes:
http://bugzilla.kernel.org/show_bug.cgi?id=13476

Reported-by: Martin Ettl <ettl.martin@gmx.de>
Signed-off-by: Jesper Nilsson <jesper.nilsson@axis.com>
Acked-by: Mikael Starvik <mikael.starvik@axis.com>
---
 arch/cris/arch-v32/kernel/irq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/cris/arch-v32/kernel/irq.c b/arch/cris/arch-v32/kernel/irq.c
index df3925cb1c7f..9c9eb5b708fc 100644
--- a/arch/cris/arch-v32/kernel/irq.c
+++ b/arch/cris/arch-v32/kernel/irq.c
@@ -428,8 +428,8 @@ crisv32_do_multiple(struct pt_regs* regs)
 			 masked[i] &= ~TIMER_MASK;
 			 do_IRQ(TIMER0_INTR_VECT, regs);
 		}
-	}
 #endif
+	}
 
 #ifdef IGNORE_MASK
 	/* Remove IRQs that can't be handled as multiple. */
-- 
cgit v1.2.3-59-g8ed1b


From 04846b5b8112e53b588038349b3e92b8485c1807 Mon Sep 17 00:00:00 2001
From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Date: Mon, 20 Apr 2009 10:54:52 +0900
Subject: PCI MSI: Remove unused/obsolete macros and definitions

Impact: cleanup, spec compliance

This patch does:

 - Remove unused msi/msix_enable/disable macros.
   User should use msi/msix_set_enable() functions instead.

 - Remove unused msix_mask/unmask/pending macros.
   These macros are useless because they are not based on any of
   the PCI Local Bus Specifications properly.
   It seems that they were written based on a draft of PCI spec,
   and that the draft was the MSI-X ECN that underwent membership
   review in September 2002.
   (* In the draft, the size of a entry in MSI-X table was 64bit,
      containing 32bit message data and DWORD aligned lower address
      plus a pending bit and a mask bit.(30+1+1bit)  The higher
      address was placed in MSI-X capability structure and shared
      by all entries.)

 - Remove PCI_MSIX_FLAGS_BITMASK.
   This definition also come from the draft ECN.

Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Reviewed-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/msi.h        | 8 +-------
 include/linux/pci_regs.h | 1 -
 2 files changed, 1 insertion(+), 8 deletions(-)

diff --git a/drivers/pci/msi.h b/drivers/pci/msi.h
index 71f4df2ef654..4fed59261952 100644
--- a/drivers/pci/msi.h
+++ b/drivers/pci/msi.h
@@ -19,18 +19,12 @@
 	( (is64bit == 1) ? base+PCI_MSI_DATA_64 : base+PCI_MSI_DATA_32 )
 #define msi_mask_bits_reg(base, is64bit) \
 	( (is64bit == 1) ? base+PCI_MSI_MASK_BIT : base+PCI_MSI_MASK_BIT-4)
-#define msi_disable(control)		control &= ~PCI_MSI_FLAGS_ENABLE
 #define is_64bit_address(control)	(!!(control & PCI_MSI_FLAGS_64BIT))
 #define is_mask_bit_support(control)	(!!(control & PCI_MSI_FLAGS_MASKBIT))
 
 #define msix_table_offset_reg(base)	(base + 0x04)
 #define msix_pba_offset_reg(base)	(base + 0x08)
-#define msix_enable(control)	 	control |= PCI_MSIX_FLAGS_ENABLE
-#define msix_disable(control)	 	control &= ~PCI_MSIX_FLAGS_ENABLE
 #define msix_table_size(control) 	((control & PCI_MSIX_FLAGS_QSIZE)+1)
-#define multi_msix_capable		msix_table_size
-#define msix_unmask(address)	 	(address & ~PCI_MSIX_FLAGS_BITMASK)
-#define msix_mask(address)		(address | PCI_MSIX_FLAGS_BITMASK)
-#define msix_is_pending(address) 	(address & PCI_MSIX_FLAGS_PENDMASK)
+#define multi_msix_capable(control)	msix_table_size((control))
 
 #endif /* MSI_H */
diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h
index 616bf8b3c8b5..dcba7668e0cd 100644
--- a/include/linux/pci_regs.h
+++ b/include/linux/pci_regs.h
@@ -304,7 +304,6 @@
 #define  PCI_MSIX_FLAGS_ENABLE	(1 << 15)
 #define  PCI_MSIX_FLAGS_MASKALL	(1 << 14)
 #define PCI_MSIX_FLAGS_BIRMASK	(7 << 0)
-#define PCI_MSIX_FLAGS_BITMASK	(1 << 0)
 
 /* CompactPCI Hotswap Register */
 
-- 
cgit v1.2.3-59-g8ed1b


From ad4efa359dcded8f76c94768f9deb1f79f950090 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 30 Apr 2009 15:20:11 -0700
Subject: PCIE: remove driver_data direct access of struct device

In the near future, the driver core is going to not allow direct access
to the driver_data pointer in struct device.  Instead, the functions
dev_get_drvdata() and dev_set_drvdata() should be used.  These functions
have been around since the beginning, so are backwards compatible with
all older kernel versions.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/portdrv_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c
index e39982503863..13ffdc35ea0e 100644
--- a/drivers/pci/pcie/portdrv_core.c
+++ b/drivers/pci/pcie/portdrv_core.c
@@ -275,7 +275,7 @@ static void pcie_device_init(struct pci_dev *parent, struct pcie_device *dev,
 	memset(device, 0, sizeof(struct device));
 	device->bus = &pcie_port_bus_type;
 	device->driver = NULL;
-	device->driver_data = NULL;
+	dev_set_drvdata(device, NULL);
 	device->release = release_pcie_device;	/* callback to free pcie dev */
 	dev_set_name(device, "%s:pcie%02x",
 		 pci_name(parent), get_descriptor_id(port_type, service_type));
-- 
cgit v1.2.3-59-g8ed1b


From 64f039d3d7f574943165b1afb72ee25caa1a9a91 Mon Sep 17 00:00:00 2001
From: "akpm@linux-foundation.org" <akpm@linux-foundation.org>
Date: Wed, 15 Apr 2009 14:24:08 -0700
Subject: PCI: ibmphp_core.c: fix warning due to missing module_exit()

drivers/pci/hotplug/ibmphp_core.c:1414: warning: `ibmphp_exit' defined but not used

Signed-off-by: Zhenwen Xu <helight.xu@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/ibmphp_core.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/pci/hotplug/ibmphp_core.c b/drivers/pci/hotplug/ibmphp_core.c
index dd18f857dfb0..29ccb8a6da8a 100644
--- a/drivers/pci/hotplug/ibmphp_core.c
+++ b/drivers/pci/hotplug/ibmphp_core.c
@@ -1419,3 +1419,4 @@ static void __exit ibmphp_exit(void)
 }
 
 module_init(ibmphp_init);
+module_exit(ibmphp_exit);
-- 
cgit v1.2.3-59-g8ed1b


From 67b5db6502ddd27d65dea43bf036abbd82d0dfc9 Mon Sep 17 00:00:00 2001
From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Date: Mon, 20 Apr 2009 10:54:59 +0900
Subject: PCI MSI: Define PCI_MSI_MASK_32/64

Impact: cleanup, improve readability

Define PCI_MSI_MASK_32/64 for 32/64bit devices, instead of using
implicit offset (-4), "PCI_MSI_MASK_BIT - 4" and "PCI_MSI_MASK_BIT".

Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Reviewed-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/msi.c        | 2 +-
 drivers/pci/msi.h        | 6 +++---
 include/linux/pci_regs.h | 3 ++-
 3 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 362773247fbf..7ffac27d5d4a 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -381,7 +381,7 @@ static int msi_capability_init(struct pci_dev *dev, int nvec)
 	entry->msi_attrib.default_irq = dev->irq;	/* Save IOAPIC IRQ */
 	entry->msi_attrib.pos = pos;
 
-	entry->mask_pos = msi_mask_bits_reg(pos, entry->msi_attrib.is_64);
+	entry->mask_pos = msi_mask_reg(pos, entry->msi_attrib.is_64);
 	/* All MSIs are unmasked by default, Mask them all */
 	if (entry->msi_attrib.maskbit)
 		pci_read_config_dword(dev, entry->mask_pos, &entry->masked);
diff --git a/drivers/pci/msi.h b/drivers/pci/msi.h
index 4fed59261952..a0662842550b 100644
--- a/drivers/pci/msi.h
+++ b/drivers/pci/msi.h
@@ -16,9 +16,9 @@
 #define msi_lower_address_reg(base)	(base + PCI_MSI_ADDRESS_LO)
 #define msi_upper_address_reg(base)	(base + PCI_MSI_ADDRESS_HI)
 #define msi_data_reg(base, is64bit)	\
-	( (is64bit == 1) ? base+PCI_MSI_DATA_64 : base+PCI_MSI_DATA_32 )
-#define msi_mask_bits_reg(base, is64bit) \
-	( (is64bit == 1) ? base+PCI_MSI_MASK_BIT : base+PCI_MSI_MASK_BIT-4)
+	(base + ((is64bit == 1) ? PCI_MSI_DATA_64 : PCI_MSI_DATA_32))
+#define msi_mask_reg(base, is64bit)	\
+	(base + ((is64bit == 1) ? PCI_MSI_MASK_64 : PCI_MSI_MASK_32))
 #define is_64bit_address(control)	(!!(control & PCI_MSI_FLAGS_64BIT))
 #define is_mask_bit_support(control)	(!!(control & PCI_MSI_FLAGS_MASKBIT))
 
diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h
index dcba7668e0cd..83b02f5a25b2 100644
--- a/include/linux/pci_regs.h
+++ b/include/linux/pci_regs.h
@@ -295,8 +295,9 @@
 #define PCI_MSI_ADDRESS_LO	4	/* Lower 32 bits */
 #define PCI_MSI_ADDRESS_HI	8	/* Upper 32 bits (if PCI_MSI_FLAGS_64BIT set) */
 #define PCI_MSI_DATA_32		8	/* 16 bits of data for 32-bit devices */
+#define PCI_MSI_MASK_32		12	/* Mask bits register for 32-bit devices */
 #define PCI_MSI_DATA_64		12	/* 16 bits of data for 64-bit devices */
-#define PCI_MSI_MASK_BIT	16	/* Mask bits register */
+#define PCI_MSI_MASK_64		16	/* Mask bits register for 64-bit devices */
 
 /* MSI-X registers (these are at offset PCI_MSIX_FLAGS) */
 #define PCI_MSIX_FLAGS		2
-- 
cgit v1.2.3-59-g8ed1b


From 1f82de10d6b1d845155363c895c552e61b36b51a Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Thu, 23 Apr 2009 20:48:32 -0700
Subject: PCI/x86: don't assume prefetchable ranges are 64bit

We should not assign 64bit ranges to PCI devices that only take 32bit
prefetchable addresses.

Try to set IORESOURCE_MEM_64 in 64bit resource of pci_device/pci_bridge
and make the bus resource only have that bit set when all devices under
it support 64bit prefetchable memory.  Use that flag to allocate
resources from that range.

Reported-by: Yannick <yannick.roehlly@free.fr>
Reviewed-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/x86/include/asm/pci.h |  1 +
 drivers/pci/bus.c          |  7 ++++++-
 drivers/pci/probe.c        |  9 ++++++--
 drivers/pci/setup-bus.c    | 52 +++++++++++++++++++++++++++++++++++++---------
 include/linux/ioport.h     |  2 ++
 include/linux/pci.h        |  4 ++++
 6 files changed, 62 insertions(+), 13 deletions(-)

diff --git a/arch/x86/include/asm/pci.h b/arch/x86/include/asm/pci.h
index b51a1e8b0baf..927958d13c19 100644
--- a/arch/x86/include/asm/pci.h
+++ b/arch/x86/include/asm/pci.h
@@ -130,6 +130,7 @@ extern void pci_iommu_alloc(void);
 
 /* generic pci stuff */
 #include <asm-generic/pci.h>
+#define PCIBIOS_MAX_MEM_32 0xffffffff
 
 #ifdef CONFIG_NUMA
 /* Returns the node based on pci bus */
diff --git a/drivers/pci/bus.c b/drivers/pci/bus.c
index 97a8194063b5..40af27f31043 100644
--- a/drivers/pci/bus.c
+++ b/drivers/pci/bus.c
@@ -41,9 +41,14 @@ pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res,
 		void *alignf_data)
 {
 	int i, ret = -ENOMEM;
+	resource_size_t max = -1;
 
 	type_mask |= IORESOURCE_IO | IORESOURCE_MEM;
 
+	/* don't allocate too high if the pref mem doesn't support 64bit*/
+	if (!(res->flags & IORESOURCE_MEM_64))
+		max = PCIBIOS_MAX_MEM_32;
+
 	for (i = 0; i < PCI_BUS_NUM_RESOURCES; i++) {
 		struct resource *r = bus->resource[i];
 		if (!r)
@@ -62,7 +67,7 @@ pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res,
 		/* Ok, try it out.. */
 		ret = allocate_resource(r, res, size,
 					r->start ? : min,
-					-1, align,
+					max, align,
 					alignf, alignf_data);
 		if (ret == 0)
 			break;
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index f1ae2475ffff..b962326e3d95 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -193,7 +193,7 @@ int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type,
 		res->flags |= pci_calc_resource_flags(l) | IORESOURCE_SIZEALIGN;
 		if (type == pci_bar_io) {
 			l &= PCI_BASE_ADDRESS_IO_MASK;
-			mask = PCI_BASE_ADDRESS_IO_MASK & 0xffff;
+			mask = PCI_BASE_ADDRESS_IO_MASK & IO_SPACE_LIMIT;
 		} else {
 			l &= PCI_BASE_ADDRESS_MEM_MASK;
 			mask = (u32)PCI_BASE_ADDRESS_MEM_MASK;
@@ -237,6 +237,8 @@ int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type,
 			dev_printk(KERN_DEBUG, &dev->dev,
 				"reg %x 64bit mmio: %pR\n", pos, res);
 		}
+
+		res->flags |= IORESOURCE_MEM_64;
 	} else {
 		sz = pci_size(l, sz, mask);
 
@@ -362,7 +364,10 @@ void __devinit pci_read_bridge_bases(struct pci_bus *child)
 		}
 	}
 	if (base <= limit) {
-		res->flags = (mem_base_lo & PCI_MEMORY_RANGE_TYPE_MASK) | IORESOURCE_MEM | IORESOURCE_PREFETCH;
+		res->flags = (mem_base_lo & PCI_PREF_RANGE_TYPE_MASK) |
+					 IORESOURCE_MEM | IORESOURCE_PREFETCH;
+		if (res->flags & PCI_PREF_RANGE_TYPE_64)
+			res->flags |= IORESOURCE_MEM_64;
 		res->start = base;
 		res->end = limit + 0xfffff;
 		dev_printk(KERN_DEBUG, &dev->dev, "bridge %sbit mmio pref: %pR\n",
diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c
index a00f85471b6e..e1c360a5b0db 100644
--- a/drivers/pci/setup-bus.c
+++ b/drivers/pci/setup-bus.c
@@ -143,6 +143,7 @@ static void pci_setup_bridge(struct pci_bus *bus)
 	struct pci_dev *bridge = bus->self;
 	struct pci_bus_region region;
 	u32 l, bu, lu, io_upper16;
+	int pref_mem64;
 
 	if (pci_is_enabled(bridge))
 		return;
@@ -198,16 +199,22 @@ static void pci_setup_bridge(struct pci_bus *bus)
 	pci_write_config_dword(bridge, PCI_PREF_LIMIT_UPPER32, 0);
 
 	/* Set up PREF base/limit. */
+	pref_mem64 = 0;
 	bu = lu = 0;
 	pcibios_resource_to_bus(bridge, &region, bus->resource[2]);
 	if (bus->resource[2]->flags & IORESOURCE_PREFETCH) {
+		int width = 8;
 		l = (region.start >> 16) & 0xfff0;
 		l |= region.end & 0xfff00000;
-		bu = upper_32_bits(region.start);
-		lu = upper_32_bits(region.end);
-		dev_info(&bridge->dev, "  PREFETCH window: %#016llx-%#016llx\n",
-		    (unsigned long long)region.start,
-		    (unsigned long long)region.end);
+		if (bus->resource[2]->flags & IORESOURCE_MEM_64) {
+			pref_mem64 = 1;
+			bu = upper_32_bits(region.start);
+			lu = upper_32_bits(region.end);
+			width = 16;
+		}
+		dev_info(&bridge->dev, "  PREFETCH window: %#0*llx-%#0*llx\n",
+				width, (unsigned long long)region.start,
+				width, (unsigned long long)region.end);
 	}
 	else {
 		l = 0x0000fff0;
@@ -215,9 +222,11 @@ static void pci_setup_bridge(struct pci_bus *bus)
 	}
 	pci_write_config_dword(bridge, PCI_PREF_MEMORY_BASE, l);
 
-	/* Set the upper 32 bits of PREF base & limit. */
-	pci_write_config_dword(bridge, PCI_PREF_BASE_UPPER32, bu);
-	pci_write_config_dword(bridge, PCI_PREF_LIMIT_UPPER32, lu);
+	if (pref_mem64) {
+		/* Set the upper 32 bits of PREF base & limit. */
+		pci_write_config_dword(bridge, PCI_PREF_BASE_UPPER32, bu);
+		pci_write_config_dword(bridge, PCI_PREF_LIMIT_UPPER32, lu);
+	}
 
 	pci_write_config_word(bridge, PCI_BRIDGE_CONTROL, bus->bridge_ctl);
 }
@@ -255,8 +264,25 @@ static void pci_bridge_check_ranges(struct pci_bus *bus)
 		pci_read_config_dword(bridge, PCI_PREF_MEMORY_BASE, &pmem);
 		pci_write_config_dword(bridge, PCI_PREF_MEMORY_BASE, 0x0);
 	}
-	if (pmem)
+	if (pmem) {
 		b_res[2].flags |= IORESOURCE_MEM | IORESOURCE_PREFETCH;
+		if ((pmem & PCI_PREF_RANGE_TYPE_MASK) == PCI_PREF_RANGE_TYPE_64)
+			b_res[2].flags |= IORESOURCE_MEM_64;
+	}
+
+	/* double check if bridge does support 64 bit pref */
+	if (b_res[2].flags & IORESOURCE_MEM_64) {
+		u32 mem_base_hi, tmp;
+		pci_read_config_dword(bridge, PCI_PREF_BASE_UPPER32,
+					 &mem_base_hi);
+		pci_write_config_dword(bridge, PCI_PREF_BASE_UPPER32,
+					       0xffffffff);
+		pci_read_config_dword(bridge, PCI_PREF_BASE_UPPER32, &tmp);
+		if (!tmp)
+			b_res[2].flags &= ~IORESOURCE_MEM_64;
+		pci_write_config_dword(bridge, PCI_PREF_BASE_UPPER32,
+				       mem_base_hi);
+	}
 }
 
 /* Helper function for sizing routines: find first available
@@ -336,6 +362,7 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask, unsigned long
 	resource_size_t aligns[12];	/* Alignments from 1Mb to 2Gb */
 	int order, max_order;
 	struct resource *b_res = find_free_bus_resource(bus, type);
+	unsigned int mem64_mask = 0;
 
 	if (!b_res)
 		return 0;
@@ -344,9 +371,12 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask, unsigned long
 	max_order = 0;
 	size = 0;
 
+	mem64_mask = b_res->flags & IORESOURCE_MEM_64;
+	b_res->flags &= ~IORESOURCE_MEM_64;
+
 	list_for_each_entry(dev, &bus->devices, bus_list) {
 		int i;
-		
+
 		for (i = 0; i < PCI_NUM_RESOURCES; i++) {
 			struct resource *r = &dev->resource[i];
 			resource_size_t r_size;
@@ -372,6 +402,7 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask, unsigned long
 				aligns[order] += align;
 			if (order > max_order)
 				max_order = order;
+			mem64_mask &= r->flags & IORESOURCE_MEM_64;
 		}
 	}
 
@@ -396,6 +427,7 @@ static int pbus_size_mem(struct pci_bus *bus, unsigned long mask, unsigned long
 	b_res->start = min_align;
 	b_res->end = size + min_align - 1;
 	b_res->flags |= IORESOURCE_STARTALIGN;
+	b_res->flags |= mem64_mask;
 	return 1;
 }
 
diff --git a/include/linux/ioport.h b/include/linux/ioport.h
index 32e4b2f72294..786e7b8cece9 100644
--- a/include/linux/ioport.h
+++ b/include/linux/ioport.h
@@ -49,6 +49,8 @@ struct resource_list {
 #define IORESOURCE_SIZEALIGN	0x00020000	/* size indicates alignment */
 #define IORESOURCE_STARTALIGN	0x00040000	/* start field is alignment */
 
+#define IORESOURCE_MEM_64	0x00100000
+
 #define IORESOURCE_EXCLUSIVE	0x08000000	/* Userland may not map this resource */
 #define IORESOURCE_DISABLED	0x10000000
 #define IORESOURCE_UNSET	0x20000000
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 72698d89e767..6dfa47d25ba4 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -1097,6 +1097,10 @@ static inline struct pci_dev *pci_get_bus_and_slot(unsigned int bus,
 
 #include <asm/pci.h>
 
+#ifndef PCIBIOS_MAX_MEM_32
+#define PCIBIOS_MAX_MEM_32 (-1)
+#endif
+
 /* these helpers provide future and backwards compatibility
  * for accessing popular PCI BAR info */
 #define pci_resource_start(dev, bar)	((dev)->resource[(bar)].start)
-- 
cgit v1.2.3-59-g8ed1b


From d09ee9687e027fc7d2c6b95daf05a8ef3ff06340 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Thu, 23 Apr 2009 20:49:25 -0700
Subject: PCI: improve resource allocation under transparent bridges

We could run out of space under under 4g, but devices under transparent
bridges can use 64bit resources, so keep trying on the parent bus until
we hit a non-transparent bridge.

Impact: better support for assigning unassigned resources

Reviewed-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/setup-bus.c |  1 -
 drivers/pci/setup-res.c | 49 ++++++++++++++++++++++++++++++++++++-------------
 2 files changed, 36 insertions(+), 14 deletions(-)

diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c
index e1c360a5b0db..b636e245445d 100644
--- a/drivers/pci/setup-bus.c
+++ b/drivers/pci/setup-bus.c
@@ -58,7 +58,6 @@ static void pbus_assign_resources_sorted(const struct pci_bus *bus)
 		res = list->res;
 		idx = res - &list->dev->resource[0];
 		if (pci_assign_resource(list->dev, idx)) {
-			/* FIXME: get rid of this */
 			res->start = 0;
 			res->end = 0;
 			res->flags = 0;
diff --git a/drivers/pci/setup-res.c b/drivers/pci/setup-res.c
index 3039fcb86afc..0b6908b48935 100644
--- a/drivers/pci/setup-res.c
+++ b/drivers/pci/setup-res.c
@@ -135,23 +135,16 @@ void pci_disable_bridge_window(struct pci_dev *dev)
 }
 #endif	/* CONFIG_PCI_QUIRKS */
 
-int pci_assign_resource(struct pci_dev *dev, int resno)
+static int __pci_assign_resource(struct pci_bus *bus, struct pci_dev *dev,
+				 int resno)
 {
-	struct pci_bus *bus = dev->bus;
 	struct resource *res = dev->resource + resno;
 	resource_size_t size, min, align;
 	int ret;
 
 	size = resource_size(res);
 	min = (res->flags & IORESOURCE_IO) ? PCIBIOS_MIN_IO : PCIBIOS_MIN_MEM;
-
 	align = resource_alignment(res);
-	if (!align) {
-		dev_info(&dev->dev, "BAR %d: can't allocate resource (bogus "
-			"alignment) %pR flags %#lx\n",
-			resno, res, res->flags);
-		return -EINVAL;
-	}
 
 	/* First, try exact prefetching match.. */
 	ret = pci_bus_alloc_resource(bus, res, size, align, min,
@@ -169,10 +162,7 @@ int pci_assign_resource(struct pci_dev *dev, int resno)
 					     pcibios_align_resource, dev);
 	}
 
-	if (ret) {
-		dev_info(&dev->dev, "BAR %d: can't allocate %s resource %pR\n",
-			resno, res->flags & IORESOURCE_IO ? "I/O" : "mem", res);
-	} else {
+	if (!ret) {
 		res->flags &= ~IORESOURCE_STARTALIGN;
 		if (resno < PCI_BRIDGE_RESOURCES)
 			pci_update_resource(dev, resno);
@@ -181,6 +171,39 @@ int pci_assign_resource(struct pci_dev *dev, int resno)
 	return ret;
 }
 
+int pci_assign_resource(struct pci_dev *dev, int resno)
+{
+	struct resource *res = dev->resource + resno;
+	resource_size_t align;
+	struct pci_bus *bus;
+	int ret;
+
+	align = resource_alignment(res);
+	if (!align) {
+		dev_info(&dev->dev, "BAR %d: can't allocate resource (bogus "
+			"alignment) %pR flags %#lx\n",
+			resno, res, res->flags);
+		return -EINVAL;
+	}
+
+	bus = dev->bus;
+	while ((ret = __pci_assign_resource(bus, dev, resno))) {
+		if (bus->parent && bus->self->transparent)
+			bus = bus->parent;
+		else
+			bus = NULL;
+		if (bus)
+			continue;
+		break;
+	}
+
+	if (ret)
+		dev_info(&dev->dev, "BAR %d: can't allocate %s resource %pR\n",
+			resno, res->flags & IORESOURCE_IO ? "I/O" : "mem", res);
+
+	return ret;
+}
+
 #if 0
 int pci_assign_resource_fixed(struct pci_dev *dev, int resno)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 861fefbf550d41e7a4f44584f3ec35977fa08bf1 Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:23:11 -0600
Subject: PCI Hotplug: cpqphp: stray whitespace cleanups

Clean up all stray whitespace issues, such as trailing whitespace,
spaces before tabs, etc. and whatever else vim's c_space_errors
highlights in red.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp.h       | 58 ++++++++++++-------------
 drivers/pci/hotplug/cpqphp_core.c  | 56 ++++++++++---------------
 drivers/pci/hotplug/cpqphp_ctrl.c  | 86 +++++++++++++++++++-------------------
 drivers/pci/hotplug/cpqphp_nvram.c | 18 ++++----
 drivers/pci/hotplug/cpqphp_pci.c   | 14 +++----
 5 files changed, 110 insertions(+), 122 deletions(-)

diff --git a/drivers/pci/hotplug/cpqphp.h b/drivers/pci/hotplug/cpqphp.h
index afaf8f69f73e..b627dfd74500 100644
--- a/drivers/pci/hotplug/cpqphp.h
+++ b/drivers/pci/hotplug/cpqphp.h
@@ -150,25 +150,25 @@ struct ctrl_reg {			/* offset */
 
 /* offsets to the controller registers based on the above structure layout */
 enum ctrl_offsets {
-	SLOT_RST = 		offsetof(struct ctrl_reg, slot_RST),
+	SLOT_RST =		offsetof(struct ctrl_reg, slot_RST),
 	SLOT_ENABLE =		offsetof(struct ctrl_reg, slot_enable),
 	MISC =			offsetof(struct ctrl_reg, misc),
 	LED_CONTROL =		offsetof(struct ctrl_reg, led_control),
 	INT_INPUT_CLEAR =	offsetof(struct ctrl_reg, int_input_clear),
-	INT_MASK = 		offsetof(struct ctrl_reg, int_mask),
-	CTRL_RESERVED0 = 	offsetof(struct ctrl_reg, reserved0),
+	INT_MASK =		offsetof(struct ctrl_reg, int_mask),
+	CTRL_RESERVED0 =	offsetof(struct ctrl_reg, reserved0),
 	CTRL_RESERVED1 =	offsetof(struct ctrl_reg, reserved1),
 	CTRL_RESERVED2 =	offsetof(struct ctrl_reg, reserved1),
-	GEN_OUTPUT_AB = 	offsetof(struct ctrl_reg, gen_output_AB),
-	NON_INT_INPUT = 	offsetof(struct ctrl_reg, non_int_input),
+	GEN_OUTPUT_AB =		offsetof(struct ctrl_reg, gen_output_AB),
+	NON_INT_INPUT =		offsetof(struct ctrl_reg, non_int_input),
 	CTRL_RESERVED3 =	offsetof(struct ctrl_reg, reserved3),
 	CTRL_RESERVED4 =	offsetof(struct ctrl_reg, reserved4),
 	CTRL_RESERVED5 =	offsetof(struct ctrl_reg, reserved5),
 	CTRL_RESERVED6 =	offsetof(struct ctrl_reg, reserved6),
 	CTRL_RESERVED7 =	offsetof(struct ctrl_reg, reserved7),
 	CTRL_RESERVED8 =	offsetof(struct ctrl_reg, reserved8),
-	SLOT_MASK = 		offsetof(struct ctrl_reg, slot_mask),
-	CTRL_RESERVED9 = 	offsetof(struct ctrl_reg, reserved9),
+	SLOT_MASK =		offsetof(struct ctrl_reg, slot_mask),
+	CTRL_RESERVED9 =	offsetof(struct ctrl_reg, reserved9),
 	CTRL_RESERVED10 =	offsetof(struct ctrl_reg, reserved10),
 	CTRL_RESERVED11 =	offsetof(struct ctrl_reg, reserved11),
 	SLOT_SERR =		offsetof(struct ctrl_reg, slot_SERR),
@@ -220,15 +220,15 @@ struct slot_rt {
 /* offsets to the hotplug slot resource table registers based on the above structure layout */
 enum slot_rt_offsets {
 	DEV_FUNC =		offsetof(struct slot_rt, dev_func),
-	PRIMARY_BUS = 		offsetof(struct slot_rt, primary_bus),
-	SECONDARY_BUS = 	offsetof(struct slot_rt, secondary_bus),
-	MAX_BUS = 		offsetof(struct slot_rt, max_bus),
-	IO_BASE = 		offsetof(struct slot_rt, io_base),
-	IO_LENGTH = 		offsetof(struct slot_rt, io_length),
-	MEM_BASE = 		offsetof(struct slot_rt, mem_base),
-	MEM_LENGTH = 		offsetof(struct slot_rt, mem_length),
-	PRE_MEM_BASE = 		offsetof(struct slot_rt, pre_mem_base),
-	PRE_MEM_LENGTH = 	offsetof(struct slot_rt, pre_mem_length),
+	PRIMARY_BUS =		offsetof(struct slot_rt, primary_bus),
+	SECONDARY_BUS =		offsetof(struct slot_rt, secondary_bus),
+	MAX_BUS =		offsetof(struct slot_rt, max_bus),
+	IO_BASE =		offsetof(struct slot_rt, io_base),
+	IO_LENGTH =		offsetof(struct slot_rt, io_length),
+	MEM_BASE =		offsetof(struct slot_rt, mem_base),
+	MEM_LENGTH =		offsetof(struct slot_rt, mem_length),
+	PRE_MEM_BASE =		offsetof(struct slot_rt, pre_mem_base),
+	PRE_MEM_LENGTH =	offsetof(struct slot_rt, pre_mem_length),
 };
 
 struct pci_func {
@@ -471,7 +471,7 @@ static inline void return_resource(struct pci_resource **head, struct pci_resour
 static inline void set_SOGO(struct controller *ctrl)
 {
 	u16 misc;
-	
+
 	misc = readw(ctrl->hpc_reg + MISC);
 	misc = (misc | 0x0001) & 0xFFFB;
 	writew(misc, ctrl->hpc_reg + MISC);
@@ -481,7 +481,7 @@ static inline void set_SOGO(struct controller *ctrl)
 static inline void amber_LED_on(struct controller *ctrl, u8 slot)
 {
 	u32 led_control;
-	
+
 	led_control = readl(ctrl->hpc_reg + LED_CONTROL);
 	led_control |= (0x01010000L << slot);
 	writel(led_control, ctrl->hpc_reg + LED_CONTROL);
@@ -491,7 +491,7 @@ static inline void amber_LED_on(struct controller *ctrl, u8 slot)
 static inline void amber_LED_off(struct controller *ctrl, u8 slot)
 {
 	u32 led_control;
-	
+
 	led_control = readl(ctrl->hpc_reg + LED_CONTROL);
 	led_control &= ~(0x01010000L << slot);
 	writel(led_control, ctrl->hpc_reg + LED_CONTROL);
@@ -504,7 +504,7 @@ static inline int read_amber_LED(struct controller *ctrl, u8 slot)
 
 	led_control = readl(ctrl->hpc_reg + LED_CONTROL);
 	led_control &= (0x01010000L << slot);
-	
+
 	return led_control ? 1 : 0;
 }
 
@@ -512,7 +512,7 @@ static inline int read_amber_LED(struct controller *ctrl, u8 slot)
 static inline void green_LED_on(struct controller *ctrl, u8 slot)
 {
 	u32 led_control;
-	
+
 	led_control = readl(ctrl->hpc_reg + LED_CONTROL);
 	led_control |= 0x0101L << slot;
 	writel(led_control, ctrl->hpc_reg + LED_CONTROL);
@@ -521,7 +521,7 @@ static inline void green_LED_on(struct controller *ctrl, u8 slot)
 static inline void green_LED_off(struct controller *ctrl, u8 slot)
 {
 	u32 led_control;
-	
+
 	led_control = readl(ctrl->hpc_reg + LED_CONTROL);
 	led_control &= ~(0x0101L << slot);
 	writel(led_control, ctrl->hpc_reg + LED_CONTROL);
@@ -531,7 +531,7 @@ static inline void green_LED_off(struct controller *ctrl, u8 slot)
 static inline void green_LED_blink(struct controller *ctrl, u8 slot)
 {
 	u32 led_control;
-	
+
 	led_control = readl(ctrl->hpc_reg + LED_CONTROL);
 	led_control &= ~(0x0101L << slot);
 	led_control |= (0x0001L << slot);
@@ -586,11 +586,11 @@ static inline u8 read_slot_enable(struct controller *ctrl)
 static inline u8 get_controller_speed(struct controller *ctrl)
 {
 	u8 curr_freq;
- 	u16 misc;
- 	
+	u16 misc;
+
 	if (ctrl->pcix_support) {
 		curr_freq = readb(ctrl->hpc_reg + NEXT_CURR_FREQ);
-		if ((curr_freq & 0xB0) == 0xB0) 
+		if ((curr_freq & 0xB0) == 0xB0)
 			return PCI_SPEED_133MHz_PCIX;
 		if ((curr_freq & 0xA0) == 0xA0)
 			return PCI_SPEED_100MHz_PCIX;
@@ -602,10 +602,10 @@ static inline u8 get_controller_speed(struct controller *ctrl)
 		return PCI_SPEED_33MHz;
 	}
 
- 	misc = readw(ctrl->hpc_reg + MISC);
- 	return (misc & 0x0800) ? PCI_SPEED_66MHz : PCI_SPEED_33MHz;
+	misc = readw(ctrl->hpc_reg + MISC);
+	return (misc & 0x0800) ? PCI_SPEED_66MHz : PCI_SPEED_33MHz;
 }
- 
+
 
 /*
  * get_adapter_speed - find the max supported frequency/mode of adapter.
diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c
index c2e1bcbb28a7..55eae4c233c9 100644
--- a/drivers/pci/hotplug/cpqphp_core.c
+++ b/drivers/pci/hotplug/cpqphp_core.c
@@ -25,7 +25,7 @@
  * Send feedback to <greg@kroah.com>
  *
  * Jan 12, 2003 -	Added 66/100/133MHz PCI-X support,
- * 			Torben Mathiasen <torben.mathiasen@hp.com>
+ *			Torben Mathiasen <torben.mathiasen@hp.com>
  *
  */
 
@@ -100,8 +100,8 @@ static struct hotplug_slot_ops cpqphp_hotplug_slot_ops = {
 	.get_attention_status =	get_attention_status,
 	.get_latch_status =	get_latch_status,
 	.get_adapter_status =	get_adapter_status,
-  	.get_max_bus_speed =	get_max_bus_speed,
-  	.get_cur_bus_speed =	get_cur_bus_speed,
+	.get_max_bus_speed =	get_max_bus_speed,
+	.get_cur_bus_speed =	get_cur_bus_speed,
 };
 
 
@@ -144,7 +144,7 @@ static void __iomem * detect_SMBIOS_pointer(void __iomem *begin, void __iomem *e
 			break;
 		}
 	}
-	
+
 	if (!status)
 		fp = NULL;
 
@@ -292,7 +292,7 @@ static void __iomem *get_SMBIOS_entry(void __iomem *smbios_start,
 	if (!smbios_table)
 		return NULL;
 
-	if (!previous) {		  
+	if (!previous) {
 		previous = smbios_start;
 	} else {
 		previous = get_subsequent_smbios_entry(smbios_start,
@@ -300,7 +300,7 @@ static void __iomem *get_SMBIOS_entry(void __iomem *smbios_start,
 	}
 
 	while (previous) {
-	       	if (readb(previous + SMBIOS_GENERIC_TYPE) != type) {
+		if (readb(previous + SMBIOS_GENERIC_TYPE) != type) {
 			previous = get_subsequent_smbios_entry(smbios_start,
 						smbios_table, previous);
 		} else {
@@ -426,7 +426,7 @@ static int ctrl_slot_setup(struct controller *ctrl,
 			cpq_get_latch_status(ctrl, slot);
 		hotplug_slot_info->adapter_status =
 			get_presence_status(ctrl, slot);
-		
+
 		dbg("registering bus %d, dev %d, number %d, "
 				"ctrl->slot_device_offset %d, slot %d\n",
 				slot->bus, slot->device,
@@ -440,7 +440,7 @@ static int ctrl_slot_setup(struct controller *ctrl,
 			err("pci_hp_register failed with error %d\n", result);
 			goto error_info;
 		}
-		
+
 		slot->next = ctrl->slot;
 		ctrl->slot = slot;
 
@@ -563,9 +563,9 @@ get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot)
 
 	}
 
-	// If we got here, we didn't find an entry in the IRQ mapping table 
-	// for the target PCI device.  If we did determine that the target 
-	// device is on the other side of a PCI-to-PCI bridge, return the 
+	// If we got here, we didn't find an entry in the IRQ mapping table
+	// for the target PCI device.  If we did determine that the target
+	// device is on the other side of a PCI-to-PCI bridge, return the
 	// slot number for the bridge.
 	if (bridgeSlot != 0xFF) {
 		*slot = bridgeSlot;
@@ -719,7 +719,7 @@ static int hardware_test(struct hotplug_slot *hotplug_slot, u32 value)
 
 	dbg("%s - physical_slot = %s\n", __func__, slot_name(slot));
 
-	return cpqhp_hardware_test(ctrl, value);	
+	return cpqhp_hardware_test(ctrl, value);
 }
 
 
@@ -738,7 +738,7 @@ static int get_attention_status(struct hotplug_slot *hotplug_slot, u8 *value)
 {
 	struct slot *slot = hotplug_slot->private;
 	struct controller *ctrl = slot->ctrl;
-	
+
 	dbg("%s - physical_slot = %s\n", __func__, slot_name(slot));
 
 	*value = cpq_get_attention_status(ctrl, slot);
@@ -832,7 +832,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	}
 
 	/* Check for the proper subsytem ID's
-	 * Intel uses a different SSID programming model than Compaq.  
+	 * Intel uses a different SSID programming model than Compaq.
 	 * For Intel, each SSID bit identifies a PHP capability.
 	 * Also Intel HPC's may have RID=0.
 	 */
@@ -1087,7 +1087,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (rc) {
 		goto err_free_bus;
 	}
-	
+
 	dbg("pdev = %p\n", pdev);
 	dbg("pci resource start %llx\n", (unsigned long long)pci_resource_start(pdev, 0));
 	dbg("pci resource len %llx\n", (unsigned long long)pci_resource_len(pdev, 0));
@@ -1182,7 +1182,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 			__func__, rc);
 		goto err_iounmap;
 	}
-	
+
 	/* Mask all general input interrupts */
 	writel(0xFFFFFFFFL, ctrl->hpc_reg + INT_MASK);
 
@@ -1291,7 +1291,6 @@ err_disable_device:
 	return rc;
 }
 
-
 static int one_time_init(void)
 {
 	int loop;
@@ -1328,10 +1327,10 @@ static int one_time_init(void)
 		retval = -EIO;
 		goto error;
 	}
-	
+
 	/* Now, map the int15 entry point if we are on compaq specific hardware */
 	compaq_nvram_init(cpqhp_rom_start);
-	
+
 	/* Map smbios table entry point structure */
 	smbios_table = detect_SMBIOS_pointer(cpqhp_rom_start,
 					cpqhp_rom_start + ROM_PHY_LEN);
@@ -1361,7 +1360,6 @@ error:
 	return retval;
 }
 
-
 static void __exit unload_cpqphpd(void)
 {
 	struct pci_func *next;
@@ -1381,10 +1379,10 @@ static void __exit unload_cpqphpd(void)
 		if (ctrl->hpc_reg) {
 			u16 misc;
 			rc = read_slot_enable (ctrl);
-			
+
 			writeb(0, ctrl->hpc_reg + SLOT_SERR);
 			writel(0xFFFFFFC0L | ~rc, ctrl->hpc_reg + INT_MASK);
-			
+
 			misc = readw(ctrl->hpc_reg + MISC);
 			misc &= 0xFFFD;
 			writew(misc, ctrl->hpc_reg + MISC);
@@ -1475,27 +1473,23 @@ static void __exit unload_cpqphpd(void)
 		iounmap(smbios_start);
 }
 
-
-
 static struct pci_device_id hpcd_pci_tbl[] = {
 	{
 	/* handle any PCI Hotplug controller */
 	.class =        ((PCI_CLASS_SYSTEM_PCI_HOTPLUG << 8) | 0x00),
 	.class_mask =   ~0,
-	
+
 	/* no matter who makes it */
 	.vendor =       PCI_ANY_ID,
 	.device =       PCI_ANY_ID,
 	.subvendor =    PCI_ANY_ID,
 	.subdevice =    PCI_ANY_ID,
-	
+
 	}, { /* end: all zeroes */ }
 };
 
 MODULE_DEVICE_TABLE(pci, hpcd_pci_tbl);
 
-
-
 static struct pci_driver cpqhpc_driver = {
 	.name =		"compaq_pci_hotplug",
 	.id_table =	hpcd_pci_tbl,
@@ -1503,8 +1497,6 @@ static struct pci_driver cpqhpc_driver = {
 	/* remove:	cpqhpc_remove_one, */
 };
 
-
-
 static int __init cpqhpc_init(void)
 {
 	int result;
@@ -1518,7 +1510,6 @@ static int __init cpqhpc_init(void)
 	return result;
 }
 
-
 static void __exit cpqhpc_cleanup(void)
 {
 	dbg("unload_cpqphpd()\n");
@@ -1529,8 +1520,5 @@ static void __exit cpqhpc_cleanup(void)
 	cpqhp_shutdown_debugfs();
 }
 
-
 module_init(cpqhpc_init);
 module_exit(cpqhpc_cleanup);
-
-
diff --git a/drivers/pci/hotplug/cpqphp_ctrl.c b/drivers/pci/hotplug/cpqphp_ctrl.c
index cc227a8c4b11..ec3a76519af2 100644
--- a/drivers/pci/hotplug/cpqphp_ctrl.c
+++ b/drivers/pci/hotplug/cpqphp_ctrl.c
@@ -642,7 +642,7 @@ static struct pci_resource *get_max_resource(struct pci_resource **head, u32 siz
 		return NULL;
 
 	for (max = *head; max; max = max->next) {
-		/* If not big enough we could probably just bail, 
+		/* If not big enough we could probably just bail,
 		 * instead we'll continue to the next. */
 		if (max->length < size)
 			continue;
@@ -886,7 +886,7 @@ irqreturn_t cpqhp_ctrl_intr(int IRQ, void *data)
 	u32 Diff;
 	u32 temp_dword;
 
-	
+
 	misc = readw(ctrl->hpc_reg + MISC);
 	/***************************************
 	 * Check to see if it was our interrupt
@@ -1130,33 +1130,33 @@ static u8 set_controller_speed(struct controller *ctrl, u8 adapter_speed, u8 hp_
 	u8 slot_power = readb(ctrl->hpc_reg + SLOT_POWER);
 	u16 reg16;
 	u32 leds = readl(ctrl->hpc_reg + LED_CONTROL);
-	
+
 	if (ctrl->speed == adapter_speed)
 		return 0;
-	
+
 	/* We don't allow freq/mode changes if we find another adapter running
 	 * in another slot on this controller */
 	for(slot = ctrl->slot; slot; slot = slot->next) {
-		if (slot->device == (hp_slot + ctrl->slot_device_offset)) 
+		if (slot->device == (hp_slot + ctrl->slot_device_offset))
 			continue;
 		if (!slot->hotplug_slot || !slot->hotplug_slot->info)
 			continue;
-		if (slot->hotplug_slot->info->adapter_status == 0) 
+		if (slot->hotplug_slot->info->adapter_status == 0)
 			continue;
 		/* If another adapter is running on the same segment but at a
 		 * lower speed/mode, we allow the new adapter to function at
 		 * this rate if supported */
-		if (ctrl->speed < adapter_speed) 
+		if (ctrl->speed < adapter_speed)
 			return 0;
 
 		return 1;
 	}
-	
+
 	/* If the controller doesn't support freq/mode changes and the
 	 * controller is running at a higher mode, we bail */
 	if ((ctrl->speed > adapter_speed) && (!ctrl->pcix_speed_capability))
 		return 1;
-	
+
 	/* But we allow the adapter to run at a lower rate if possible */
 	if ((ctrl->speed < adapter_speed) && (!ctrl->pcix_speed_capability))
 		return 0;
@@ -1171,22 +1171,22 @@ static u8 set_controller_speed(struct controller *ctrl, u8 adapter_speed, u8 hp_
 
 	writel(0x0L, ctrl->hpc_reg + LED_CONTROL);
 	writeb(0x00, ctrl->hpc_reg + SLOT_ENABLE);
-	
-	set_SOGO(ctrl); 
+
+	set_SOGO(ctrl);
 	wait_for_ctrl_irq(ctrl);
-	
+
 	if (adapter_speed != PCI_SPEED_133MHz_PCIX)
 		reg = 0xF5;
 	else
-		reg = 0xF4;	
+		reg = 0xF4;
 	pci_write_config_byte(ctrl->pci_dev, 0x41, reg);
-	
+
 	reg16 = readw(ctrl->hpc_reg + NEXT_CURR_FREQ);
 	reg16 &= ~0x000F;
 	switch(adapter_speed) {
-		case(PCI_SPEED_133MHz_PCIX): 
+		case(PCI_SPEED_133MHz_PCIX):
 			reg = 0x75;
-			reg16 |= 0xB; 
+			reg16 |= 0xB;
 			break;
 		case(PCI_SPEED_100MHz_PCIX):
 			reg = 0x74;
@@ -1203,47 +1203,47 @@ static u8 set_controller_speed(struct controller *ctrl, u8 adapter_speed, u8 hp_
 		default: /* 33MHz PCI 2.2 */
 			reg = 0x71;
 			break;
-			
+
 	}
 	reg16 |= 0xB << 12;
 	writew(reg16, ctrl->hpc_reg + NEXT_CURR_FREQ);
-	
-	mdelay(5); 
-	
+
+	mdelay(5);
+
 	/* Reenable interrupts */
 	writel(0, ctrl->hpc_reg + INT_MASK);
 
-	pci_write_config_byte(ctrl->pci_dev, 0x41, reg); 
-	
+	pci_write_config_byte(ctrl->pci_dev, 0x41, reg);
+
 	/* Restart state machine */
 	reg = ~0xF;
 	pci_read_config_byte(ctrl->pci_dev, 0x43, &reg);
 	pci_write_config_byte(ctrl->pci_dev, 0x43, reg);
-	
+
 	/* Only if mode change...*/
 	if (((ctrl->speed == PCI_SPEED_66MHz) && (adapter_speed == PCI_SPEED_66MHz_PCIX)) ||
 		((ctrl->speed == PCI_SPEED_66MHz_PCIX) && (adapter_speed == PCI_SPEED_66MHz))) 
 			set_SOGO(ctrl);
-	
+
 	wait_for_ctrl_irq(ctrl);
 	mdelay(1100);
-	
+
 	/* Restore LED/Slot state */
 	writel(leds, ctrl->hpc_reg + LED_CONTROL);
 	writeb(slot_power, ctrl->hpc_reg + SLOT_ENABLE);
-	
+
 	set_SOGO(ctrl);
 	wait_for_ctrl_irq(ctrl);
 
 	ctrl->speed = adapter_speed;
 	slot = cpqhp_find_slot(ctrl, hp_slot + ctrl->slot_device_offset);
 
-	info("Successfully changed frequency/mode for adapter in slot %d\n", 
+	info("Successfully changed frequency/mode for adapter in slot %d\n",
 			slot->number);
 	return 0;
 }
 
-/* the following routines constitute the bulk of the 
+/* the following routines constitute the bulk of the
    hotplug controller logic
  */
 
@@ -1299,7 +1299,7 @@ static u32 board_replaced(struct pci_func *func, struct controller *ctrl)
 
 		/* Wait for SOBS to be unset */
 		wait_for_ctrl_irq (ctrl);
-		
+
 		adapter_speed = get_adapter_speed(ctrl, hp_slot);
 		if (ctrl->speed != adapter_speed)
 			if (set_controller_speed(ctrl, adapter_speed, hp_slot))
@@ -1443,12 +1443,12 @@ static u32 board_added(struct pci_func *func, struct controller *ctrl)
 
 	/* Wait for SOBS to be unset */
 	wait_for_ctrl_irq (ctrl);
-	
+
 	adapter_speed = get_adapter_speed(ctrl, hp_slot);
 	if (ctrl->speed != adapter_speed)
 		if (set_controller_speed(ctrl, adapter_speed, hp_slot))
 			rc = WRONG_BUS_FREQUENCY;
-	
+
 	/* turn off board without attaching to the bus */
 	disable_slot_power (ctrl, hp_slot);
 
@@ -1461,7 +1461,7 @@ static u32 board_added(struct pci_func *func, struct controller *ctrl)
 
 	if (rc)
 		return rc;
-	
+
 	p_slot = cpqhp_find_slot(ctrl, hp_slot + ctrl->slot_device_offset);
 
 	/* turn on board and blink green LED */
@@ -1859,12 +1859,12 @@ static void interrupt_event_handler(struct controller *ctrl)
 						info(msg_button_on, p_slot->number);
 					}
 					mutex_lock(&ctrl->crit_sect);
-					
+
 					dbg("blink green LED and turn off amber\n");
-					
+
 					amber_LED_off (ctrl, hp_slot);
 					green_LED_blink (ctrl, hp_slot);
-					
+
 					set_SOGO(ctrl);
 
 					/* Wait for SOBS to be unset */
@@ -1958,7 +1958,7 @@ void cpqhp_pushbutton_thread(unsigned long slot)
 			if (cpqhp_process_SI(ctrl, func) != 0) {
 				amber_LED_on(ctrl, hp_slot);
 				green_LED_off(ctrl, hp_slot);
-				
+
 				set_SOGO(ctrl);
 
 				/* Wait for SOBS to be unset */
@@ -2079,7 +2079,7 @@ int cpqhp_process_SS(struct controller *ctrl, struct pci_func *func)
 	struct pci_bus *pci_bus = ctrl->pci_bus;
 	int physical_slot=0;
 
-	device = func->device; 
+	device = func->device;
 	func = cpqhp_slot_find(ctrl->bus, device, index++);
 	p_slot = cpqhp_find_slot(ctrl, device);
 	if (p_slot) {
@@ -2216,7 +2216,7 @@ int cpqhp_hardware_test(struct controller *ctrl, int test_num)
 				long_delay((3*HZ)/10);
 				work_LED = work_LED >> 16;
 				writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
-				
+
 				set_SOGO(ctrl);
 
 				/* Wait for SOGO interrupt */
@@ -2339,7 +2339,7 @@ static u32 configure_new_device(struct controller * ctrl, struct pci_func * func
 
 
 /*
-  Configuration logic that involves the hotplug data structures and 
+  Configuration logic that involves the hotplug data structures and
   their bookkeeping
  */
 
@@ -2917,17 +2917,17 @@ static int configure_new_function(struct controller *ctrl, struct pci_func *func
 		}		/* End of base register loop */
 		if (cpqhp_legacy_mode) {
 			/* Figure out which interrupt pin this function uses */
-			rc = pci_bus_read_config_byte (pci_bus, devfn, 
+			rc = pci_bus_read_config_byte (pci_bus, devfn,
 				PCI_INTERRUPT_PIN, &temp_byte);
 
 			/* If this function needs an interrupt and we are behind
 			 * a bridge and the pin is tied to something that's
 			 * alread mapped, set this one the same */
-			if (temp_byte && resources->irqs && 
-			    (resources->irqs->valid_INT & 
+			if (temp_byte && resources->irqs &&
+			    (resources->irqs->valid_INT &
 			     (0x01 << ((temp_byte + resources->irqs->barber_pole - 1) & 0x03)))) {
 				/* We have to share with something already set up */
-				IRQ = resources->irqs->interrupt[(temp_byte + 
+				IRQ = resources->irqs->interrupt[(temp_byte +
 					resources->irqs->barber_pole - 1) & 0x03];
 			} else {
 				/* Program IRQ based on card type */
diff --git a/drivers/pci/hotplug/cpqphp_nvram.c b/drivers/pci/hotplug/cpqphp_nvram.c
index cb174888002b..76e110f0e3a6 100644
--- a/drivers/pci/hotplug/cpqphp_nvram.c
+++ b/drivers/pci/hotplug/cpqphp_nvram.c
@@ -113,7 +113,7 @@ static u32 add_byte( u32 **p_buffer, u8 value, u32 *used, u32 *avail)
 
 	if ((*used + 1) > *avail)
 		return(1);
-	
+
 	*((u8*)*p_buffer) = value;
 	tByte = (u8**)p_buffer;
 	(*tByte)++;
@@ -170,10 +170,10 @@ static u32 access_EV (u16 operation, u8 *ev_name, u8 *buffer, u32 *buf_size)
 	unsigned long flags;
 	int op = operation;
 	int ret_val;
-	
+
 	if (!compaq_int15_entry_point)
 		return -ENODEV;
-	
+
 	spin_lock_irqsave(&int15_lock, flags);
 	__asm__ (
 		"xorl   %%ebx,%%ebx\n" \
@@ -187,7 +187,7 @@ static u32 access_EV (u16 operation, u8 *ev_name, u8 *buffer, u32 *buf_size)
 		"D" (buffer), "m" (compaq_int15_entry_point)
 		: "%ebx", "%edx");
 	spin_unlock_irqrestore(&int15_lock, flags);
-	
+
 	return((ret_val & 0xFF00) >> 8);
 }
 
@@ -263,7 +263,7 @@ static u32 store_HRT (void __iomem *rom_start)
 	p_EV_header = (struct ev_hrt_header *) pFill;
 
 	ctrl = cpqhp_ctrl_list;
-	
+
 	// The revision of this structure
 	rc = add_byte( &pFill, 1 + ctrl->push_flag, &usedbytes, &available);
 	if (rc)
@@ -401,7 +401,7 @@ static u32 store_HRT (void __iomem *rom_start)
 
 		ctrl = ctrl->next;
 	}
-	
+
 	p_EV_header->num_of_ctrl = numCtrl;
 
 	// Now store the EV
@@ -479,7 +479,7 @@ int compaq_nvram_load (void __iomem *rom_start, struct controller *ctrl)
 		function = p_ev_ctrl->function;
 
 		while ((bus != ctrl->bus) ||
-		       (device != PCI_SLOT(ctrl->pci_dev->devfn)) || 
+		       (device != PCI_SLOT(ctrl->pci_dev->devfn)) ||
 		       (function != PCI_FUNC(ctrl->pci_dev->devfn))) {
 			nummem = p_ev_ctrl->mem_avail;
 			numpmem = p_ev_ctrl->p_mem_avail;
@@ -640,14 +640,14 @@ int compaq_nvram_load (void __iomem *rom_start, struct controller *ctrl)
 		if (rc)
 			return(rc);
 	} else {
-		if ((evbuffer[0] != 0) && (!ctrl->push_flag)) 
+		if ((evbuffer[0] != 0) && (!ctrl->push_flag))
 			return 1;
 	}
 
 	return 0;
 }
 
-	
+
 int compaq_nvram_store (void __iomem *rom_start)
 {
 	int rc = 1;
diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c
index 6c0ed0fcb8ee..573a2702fb6a 100644
--- a/drivers/pci/hotplug/cpqphp_pci.c
+++ b/drivers/pci/hotplug/cpqphp_pci.c
@@ -82,7 +82,7 @@ static void __iomem *detect_HRT_floating_pointer(void __iomem *begin, void __iom
 }
 
 
-int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func)  
+int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func)
 {
 	unsigned char bus;
 	struct pci_bus *child;
@@ -116,10 +116,10 @@ int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func)
 }
 
 
-int cpqhp_unconfigure_device(struct pci_func* func) 
+int cpqhp_unconfigure_device(struct pci_func* func)
 {
 	int j;
-	
+
 	dbg("%s: bus/dev/func = %x/%x/%x\n", __func__, func->bus, func->device, func->function);
 
 	for (j=0; j<8 ; j++) {
@@ -195,8 +195,8 @@ int cpqhp_set_irq (u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num)
 
 
 /*
- * WTF??? This function isn't in the code, yet a function calls it, but the 
- * compiler optimizes it away?  strange.  Here as a placeholder to keep the 
+ * WTF??? This function isn't in the code, yet a function calls it, but the
+ * compiler optimizes it away?  strange.  Here as a placeholder to keep the
  * compiler happy.
  */
 static int PCI_ScanBusNonBridge (u8 bus, u8 device)
@@ -398,7 +398,7 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 
 				index = 0;
 				new_slot = cpqhp_slot_find(busnumber, device, index++);
-				while (new_slot && 
+				while (new_slot &&
 				       (new_slot->function != (u8) function))
 					new_slot = cpqhp_slot_find(busnumber, device, index++);
 
@@ -1168,7 +1168,7 @@ int cpqhp_valid_replace(struct controller *ctrl, struct pci_func * func)
  * this function is for hot plug ADD!
  *
  * returns 0 if success
- */  
+ */
 int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_start)
 {
 	u8 temp;
-- 
cgit v1.2.3-59-g8ed1b


From 427438c61b0083a60bb953cb36cfdc5841f0bb03 Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:23:16 -0600
Subject: PCI Hotplug: cpqphp: fix comment style

Fix up comments from C++ to C-style, wrapping if necessary, etc.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp.h       |  22 +--
 drivers/pci/hotplug/cpqphp_core.c  | 157 ++++++++++---------
 drivers/pci/hotplug/cpqphp_ctrl.c  | 131 +++++++++-------
 drivers/pci/hotplug/cpqphp_nvram.c |  79 +++++-----
 drivers/pci/hotplug/cpqphp_pci.c   | 313 ++++++++++++++++++++-----------------
 5 files changed, 387 insertions(+), 315 deletions(-)

diff --git a/drivers/pci/hotplug/cpqphp.h b/drivers/pci/hotplug/cpqphp.h
index b627dfd74500..e15d657368f0 100644
--- a/drivers/pci/hotplug/cpqphp.h
+++ b/drivers/pci/hotplug/cpqphp.h
@@ -190,7 +190,9 @@ struct hrt {
 	u32 reserved2;
 } __attribute__ ((packed));
 
-/* offsets to the hotplug resource table registers based on the above structure layout */
+/* offsets to the hotplug resource table registers based on the above
+ * structure layout
+ */
 enum hrt_offsets {
 	SIG0 =			offsetof(struct hrt, sig0),
 	SIG1 =			offsetof(struct hrt, sig1),
@@ -217,7 +219,9 @@ struct slot_rt {
 	u16 pre_mem_length;
 } __attribute__ ((packed));
 
-/* offsets to the hotplug slot resource table registers based on the above structure layout */
+/* offsets to the hotplug slot resource table registers based on the above
+ * structure layout
+ */
 enum slot_rt_offsets {
 	DEV_FUNC =		offsetof(struct slot_rt, dev_func),
 	PRIMARY_BUS =		offsetof(struct slot_rt, primary_bus),
@@ -286,8 +290,8 @@ struct event_info {
 struct controller {
 	struct controller *next;
 	u32 ctrl_int_comp;
-	struct mutex crit_sect;		/* critical section mutex */
-	void __iomem *hpc_reg;		/* cookie for our pci controller location */
+	struct mutex crit_sect;	/* critical section mutex */
+	void __iomem *hpc_reg;	/* cookie for our pci controller location */
 	struct pci_resource *mem_head;
 	struct pci_resource *p_mem_head;
 	struct pci_resource *io_head;
@@ -299,7 +303,7 @@ struct controller {
 	u8 next_event;
 	u8 interrupt;
 	u8 cfgspc_irq;
-	u8 bus;				/* bus number for the pci hotplug controller */
+	u8 bus;			/* bus number for the pci hotplug controller */
 	u8 rev;
 	u8 slot_device_offset;
 	u8 first_slot;
@@ -458,7 +462,6 @@ static inline char *slot_name(struct slot *slot)
  * return_resource
  *
  * Puts node back in the resource list pointed to by head
- *
  */
 static inline void return_resource(struct pci_resource **head, struct pci_resource *node)
 {
@@ -575,13 +578,12 @@ static inline u8 read_slot_enable(struct controller *ctrl)
 }
 
 
-/*
+/**
  * get_controller_speed - find the current frequency/mode of controller.
  *
  * @ctrl: controller to get frequency/mode for.
  *
  * Returns controller speed.
- *
  */
 static inline u8 get_controller_speed(struct controller *ctrl)
 {
@@ -607,14 +609,13 @@ static inline u8 get_controller_speed(struct controller *ctrl)
 }
 
 
-/*
+/**
  * get_adapter_speed - find the max supported frequency/mode of adapter.
  *
  * @ctrl: hotplug controller.
  * @hp_slot: hotplug slot where adapter is installed.
  *
  * Returns adapter speed.
- *
  */
 static inline u8 get_adapter_speed(struct controller *ctrl, u8 hp_slot)
 {
@@ -719,4 +720,3 @@ static inline int wait_for_ctrl_irq(struct controller *ctrl)
 }
 
 #endif
-
diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c
index 55eae4c233c9..91dc95850cce 100644
--- a/drivers/pci/hotplug/cpqphp_core.c
+++ b/drivers/pci/hotplug/cpqphp_core.c
@@ -26,7 +26,6 @@
  *
  * Jan 12, 2003 -	Added 66/100/133MHz PCI-X support,
  *			Torben Mathiasen <torben.mathiasen@hp.com>
- *
  */
 
 #include <linux/module.h>
@@ -171,7 +170,7 @@ static int init_SERR(struct controller * ctrl)
 	tempdword = ctrl->first_slot;
 
 	number_of_slots = readb(ctrl->hpc_reg + SLOT_MASK) & 0x0F;
-	// Loop through slots
+	/* Loop through slots */
 	while (number_of_slots) {
 		physical_slot = tempdword;
 		writeb(0, ctrl->hpc_reg + SLOT_SERR);
@@ -200,7 +199,7 @@ static int pci_print_IRQ_route (void)
 
 	len = (routing_table->size - sizeof(struct irq_routing_table)) /
 			sizeof(struct irq_info);
-	// Make sure I got at least one entry
+	/* Make sure I got at least one entry */
 	if (len == 0) {
 		kfree(routing_table);
 		return -1;
@@ -244,7 +243,7 @@ static void __iomem *get_subsequent_smbios_entry(void __iomem *smbios_start,
 	if (!smbios_table || !curr)
 		return(NULL);
 
-	// set p_max to the end of the table
+	/* set p_max to the end of the table */
 	p_max = smbios_start + readw(smbios_table + ST_LENGTH);
 
 	p_temp = curr;
@@ -253,7 +252,8 @@ static void __iomem *get_subsequent_smbios_entry(void __iomem *smbios_start,
 	while ((p_temp < p_max) && !bail) {
 		/* Look for the double NULL terminator
 		 * The first condition is the previous byte
-		 * and the second is the curr */
+		 * and the second is the curr
+		 */
 		if (!previous_byte && !(readb(p_temp))) {
 			bail = 1;
 		}
@@ -387,8 +387,9 @@ static int ctrl_slot_setup(struct controller *ctrl,
 		slot->task_event.expires = jiffies + 5 * HZ;
 		slot->task_event.function = cpqhp_pushbutton_thread;
 
-		//FIXME: these capabilities aren't used but if they are
-		//       they need to be correctly implemented
+		/*FIXME: these capabilities aren't used but if they are
+		 *	 they need to be correctly implemented
+		 */
 		slot->capabilities |= PCISLOT_REPLACE_SUPPORTED;
 		slot->capabilities |= PCISLOT_INTERLOCK_SUPPORTED;
 
@@ -402,14 +403,14 @@ static int ctrl_slot_setup(struct controller *ctrl,
 		ctrl_slot =
 			slot_device - (readb(ctrl->hpc_reg + SLOT_MASK) >> 4);
 
-		// Check presence
+		/* Check presence */
 		slot->capabilities |=
 			((((~tempdword) >> 23) |
 			 ((~tempdword) >> 15)) >> ctrl_slot) & 0x02;
-		// Check the switch state
+		/* Check the switch state */
 		slot->capabilities |=
 			((~tempdword & 0xFF) >> ctrl_slot) & 0x01;
-		// Check the slot enable
+		/* Check the slot enable */
 		slot->capabilities |=
 			((read_slot_enable(ctrl) << 2) >> ctrl_slot) & 0x04;
 
@@ -476,11 +477,11 @@ static int ctrl_slot_cleanup (struct controller * ctrl)
 
 	cpqhp_remove_debugfs_files(ctrl);
 
-	//Free IRQ associated with hot plug device
+	/* Free IRQ associated with hot plug device */
 	free_irq(ctrl->interrupt, ctrl);
-	//Unmap the memory
+	/* Unmap the memory */
 	iounmap(ctrl->hpc_reg);
-	//Finally reclaim PCI mem
+	/* Finally reclaim PCI mem */
 	release_mem_region(pci_resource_start(ctrl->pci_dev, 0),
 			   pci_resource_len(ctrl->pci_dev, 0));
 
@@ -488,20 +489,17 @@ static int ctrl_slot_cleanup (struct controller * ctrl)
 }
 
 
-//============================================================================
-// function:	get_slot_mapping
-//
-// Description: Attempts to determine a logical slot mapping for a PCI
-//		device.  Won't work for more than one PCI-PCI bridge
-//		in a slot.
-//
-// Input:	u8 bus_num - bus number of PCI device
-//		u8 dev_num - device number of PCI device
-//		u8 *slot - Pointer to u8 where slot number will
-//			be returned
-//
-// Output:	SUCCESS or FAILURE
-//=============================================================================
+/**
+ * get_slot_mapping - determine logical slot mapping for PCI device
+ *
+ * Won't work for more than one PCI-PCI bridge in a slot.
+ *
+ * @bus_num - bus number of PCI device
+ * @dev_num - device number of PCI device
+ * @slot - Pointer to u8 where slot number will	be returned
+ *
+ * Output:	SUCCESS or FAILURE
+ */
 static int
 get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot)
 {
@@ -522,7 +520,7 @@ get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot)
 
 	len = (PCIIRQRoutingInfoLength->size -
 	       sizeof(struct irq_routing_table)) / sizeof(struct irq_info);
-	// Make sure I got at least one entry
+	/* Make sure I got at least one entry */
 	if (len == 0) {
 		kfree(PCIIRQRoutingInfoLength);
 		return -1;
@@ -539,13 +537,14 @@ get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot)
 			return 0;
 		} else {
 			/* Did not get a match on the target PCI device. Check
-			 * if the current IRQ table entry is a PCI-to-PCI bridge
-			 * device.  If so, and it's secondary bus matches the
-			 * bus number for the target device, I need to save the
-			 * bridge's slot number.  If I can not find an entry for
-			 * the target device, I will have to assume it's on the
-			 * other side of the bridge, and assign it the bridge's
-			 * slot. */
+			 * if the current IRQ table entry is a PCI-to-PCI
+			 * bridge device.  If so, and it's secondary bus
+			 * matches the bus number for the target device, I need
+			 * to save the bridge's slot number.  If I can not find
+			 * an entry for the target device, I will have to
+			 * assume it's on the other side of the bridge, and
+			 * assign it the bridge's slot.
+			 */
 			bus->number = tbus;
 			pci_bus_read_config_dword(bus, PCI_DEVFN(tdevice, 0),
 						PCI_CLASS_REVISION, &work);
@@ -563,17 +562,18 @@ get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot)
 
 	}
 
-	// If we got here, we didn't find an entry in the IRQ mapping table
-	// for the target PCI device.  If we did determine that the target
-	// device is on the other side of a PCI-to-PCI bridge, return the
-	// slot number for the bridge.
+	/* If we got here, we didn't find an entry in the IRQ mapping table for
+	 * the target PCI device.  If we did determine that the target device
+	 * is on the other side of a PCI-to-PCI bridge, return the slot number
+	 * for the bridge.
+	 */
 	if (bridgeSlot != 0xFF) {
 		*slot = bridgeSlot;
 		kfree(PCIIRQRoutingInfoLength);
 		return 0;
 	}
 	kfree(PCIIRQRoutingInfoLength);
-	// Couldn't find an entry in the routing table for this PCI device
+	/* Couldn't find an entry in the routing table for this PCI device */
 	return -1;
 }
 
@@ -595,7 +595,7 @@ cpqhp_set_attention_status(struct controller *ctrl, struct pci_func *func,
 
 	hp_slot = func->device - ctrl->slot_device_offset;
 
-	// Wait for exclusive access to hardware
+	/* Wait for exclusive access to hardware */
 	mutex_lock(&ctrl->crit_sect);
 
 	if (status == 1) {
@@ -603,17 +603,17 @@ cpqhp_set_attention_status(struct controller *ctrl, struct pci_func *func,
 	} else if (status == 0) {
 		amber_LED_off (ctrl, hp_slot);
 	} else {
-		// Done with exclusive hardware access
+		/* Done with exclusive hardware access */
 		mutex_unlock(&ctrl->crit_sect);
 		return(1);
 	}
 
 	set_SOGO(ctrl);
 
-	// Wait for SOBS to be unset
+	/* Wait for SOBS to be unset */
 	wait_for_ctrl_irq (ctrl);
 
-	// Done with exclusive hardware access
+	/* Done with exclusive hardware access */
 	mutex_unlock(&ctrl->crit_sect);
 
 	return(0);
@@ -815,7 +815,9 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		return err;
 	}
 
-	// Need to read VID early b/c it's used to differentiate CPQ and INTC discovery
+	/* Need to read VID early b/c it's used to differentiate CPQ and INTC
+	 * discovery
+	 */
 	rc = pci_read_config_word(pdev, PCI_VENDOR_ID, &vendor_id);
 	if (rc || ((vendor_id != PCI_VENDOR_ID_COMPAQ) && (vendor_id != PCI_VENDOR_ID_INTEL))) {
 		err(msg_HPC_non_compaq_or_intel);
@@ -837,7 +839,9 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * Also Intel HPC's may have RID=0.
 	 */
 	if ((pdev->revision > 2) || (vendor_id == PCI_VENDOR_ID_INTEL)) {
-		// TODO: This code can be made to support non-Compaq or Intel subsystem IDs
+		/* TODO: This code can be made to support non-Compaq or Intel
+		 * subsystem IDs
+		 */
 		rc = pci_read_config_word(pdev, PCI_SUBSYSTEM_VENDOR_ID, &subsystem_vid);
 		if (rc) {
 			err("%s : pci_read_config_word failed\n", __func__);
@@ -865,7 +869,9 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 		info("Hot Plug Subsystem Device ID: %x\n", subsystem_deviceid);
 
-		/* Set Vendor ID, so it can be accessed later from other functions */
+		/* Set Vendor ID, so it can be accessed later from other
+		 * functions
+		 */
 		ctrl->vendor_id = vendor_id;
 
 		switch (subsystem_vid) {
@@ -992,23 +998,23 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 				/* PHP Status (0=De-feature PHP, 1=Normal operation) */
 				if (subsystem_deviceid & 0x0008) {
-					ctrl->defeature_PHP = 1;	// PHP supported
+					ctrl->defeature_PHP = 1;	/* PHP supported */
 				} else {
-					ctrl->defeature_PHP = 0;	// PHP not supported
+					ctrl->defeature_PHP = 0;	/* PHP not supported */
 				}
 
 				/* Alternate Base Address Register Interface (0=not supported, 1=supported) */
 				if (subsystem_deviceid & 0x0010) {
-					ctrl->alternate_base_address = 1;	// supported
+					ctrl->alternate_base_address = 1;	/* supported */
 				} else {
-					ctrl->alternate_base_address = 0;	// not supported
+					ctrl->alternate_base_address = 0;	/* not supported */
 				}
 
 				/* PCI Config Space Index (0=not supported, 1=supported) */
 				if (subsystem_deviceid & 0x0020) {
-					ctrl->pci_config_space = 1;		// supported
+					ctrl->pci_config_space = 1;		/* supported */
 				} else {
-					ctrl->pci_config_space = 0;		// not supported
+					ctrl->pci_config_space = 0;		/* not supported */
 				}
 
 				/* PCI-X support */
@@ -1042,7 +1048,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		return -ENODEV;
 	}
 
-	// Tell the user that we found one.
+	/* Tell the user that we found one. */
 	info("Initializing the PCI hot plug controller residing on PCI bus %d\n",
 					pdev->bus->number);
 
@@ -1120,7 +1126,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 *
 	 ********************************************************/
 
-	// find the physical slot number of the first hot plug slot
+	/* find the physical slot number of the first hot plug slot */
 
 	/* Get slot won't work for devices behind bridges, but
 	 * in this case it will always be called for the "base"
@@ -1137,7 +1143,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_iounmap;
 	}
 
-	// Store PCI Config Space for all devices on this bus
+	/* Store PCI Config Space for all devices on this bus */
 	rc = cpqhp_save_config(ctrl, ctrl->bus, readb(ctrl->hpc_reg + SLOT_MASK));
 	if (rc) {
 		err("%s: unable to save PCI configuration data, error %d\n",
@@ -1148,7 +1154,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/*
 	 * Get IO, memory, and IRQ resources for new devices
 	 */
-	// The next line is required for cpqhp_find_available_resources
+	/* The next line is required for cpqhp_find_available_resources */
 	ctrl->interrupt = pdev->irq;
 	if (ctrl->interrupt < 0x10) {
 		cpqhp_legacy_mode = 1;
@@ -1196,12 +1202,14 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_iounmap;
 	}
 
-	/* Enable Shift Out interrupt and clear it, also enable SERR on power fault */
+	/* Enable Shift Out interrupt and clear it, also enable SERR on power
+	 * fault
+	 */
 	temp_word = readw(ctrl->hpc_reg + MISC);
 	temp_word |= 0x4006;
 	writew(temp_word, ctrl->hpc_reg + MISC);
 
-	// Changed 05/05/97 to clear all interrupts at start
+	/* Changed 05/05/97 to clear all interrupts at start */
 	writel(0xFFFFFFFFL, ctrl->hpc_reg + INT_INPUT_CLEAR);
 
 	ctrl->ctrl_int_comp = readl(ctrl->hpc_reg + INT_INPUT_CLEAR);
@@ -1216,13 +1224,14 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		cpqhp_ctrl_list = ctrl;
 	}
 
-	// turn off empty slots here unless command line option "ON" set
-	// Wait for exclusive access to hardware
+	/* turn off empty slots here unless command line option "ON" set
+	 * Wait for exclusive access to hardware
+	 */
 	mutex_lock(&ctrl->crit_sect);
 
 	num_of_slots = readb(ctrl->hpc_reg + SLOT_MASK) & 0x0F;
 
-	// find first device number for the ctrl
+	/* find first device number for the ctrl */
 	device = readb(ctrl->hpc_reg + SLOT_MASK) >> 4;
 
 	while (num_of_slots) {
@@ -1234,7 +1243,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		hp_slot = func->device - ctrl->slot_device_offset;
 		dbg("hp_slot: %d\n", hp_slot);
 
-		// We have to save the presence info for these slots
+		/* We have to save the presence info for these slots */
 		temp_word = ctrl->ctrl_int_comp >> 16;
 		func->presence_save = (temp_word >> hp_slot) & 0x01;
 		func->presence_save |= (temp_word >> (hp_slot + 7)) & 0x02;
@@ -1258,7 +1267,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	if (!power_mode) {
 		set_SOGO(ctrl);
-		// Wait for SOBS to be unset
+		/* Wait for SOBS to be unset */
 		wait_for_ctrl_irq(ctrl);
 	}
 
@@ -1269,7 +1278,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_free_irq;
 	}
 
-	// Done with exclusive hardware access
+	/* Done with exclusive hardware access */
 	mutex_unlock(&ctrl->crit_sect);
 
 	cpqhp_create_debugfs_files(ctrl);
@@ -1316,11 +1325,11 @@ static int one_time_init(void)
 		cpqhp_slot_list[loop] = NULL;
 	}
 
-	// FIXME: We also need to hook the NMI handler eventually.
-	// this also needs to be worked with Christoph
-	// register_NMI_handler();
-
-	// Map rom address
+	/* FIXME: We also need to hook the NMI handler eventually.
+	 * this also needs to be worked with Christoph
+	 * register_NMI_handler();
+	 */
+	/* Map rom address */
 	cpqhp_rom_start = ioremap(ROM_PHY_ADDR, ROM_PHY_LEN);
 	if (!cpqhp_rom_start) {
 		err ("Could not ioremap memory region for ROM\n");
@@ -1328,7 +1337,9 @@ static int one_time_init(void)
 		goto error;
 	}
 
-	/* Now, map the int15 entry point if we are on compaq specific hardware */
+	/* Now, map the int15 entry point if we are on compaq specific
+	 * hardware
+	 */
 	compaq_nvram_init(cpqhp_rom_start);
 
 	/* Map smbios table entry point structure */
@@ -1462,11 +1473,11 @@ static void __exit unload_cpqphpd(void)
 		}
 	}
 
-	// Stop the notification mechanism
+	/* Stop the notification mechanism */
 	if (initialized)
 		cpqhp_event_stop_thread();
 
-	//unmap the rom address
+	/* unmap the rom address */
 	if (cpqhp_rom_start)
 		iounmap(cpqhp_rom_start);
 	if (smbios_start)
diff --git a/drivers/pci/hotplug/cpqphp_ctrl.c b/drivers/pci/hotplug/cpqphp_ctrl.c
index ec3a76519af2..b02b8dddcf9f 100644
--- a/drivers/pci/hotplug/cpqphp_ctrl.c
+++ b/drivers/pci/hotplug/cpqphp_ctrl.c
@@ -81,14 +81,15 @@ static u8 handle_switch_change(u8 change, struct controller * ctrl)
 
 	for (hp_slot = 0; hp_slot < 6; hp_slot++) {
 		if (change & (0x1L << hp_slot)) {
-			/**********************************
+			/*
 			 * this one changed.
-			 **********************************/
+			 */
 			func = cpqhp_slot_find(ctrl->bus,
 				(hp_slot + ctrl->slot_device_offset), 0);
 
 			/* this is the structure that tells the worker thread
-			 *what to do */
+			 * what to do
+			 */
 			taskInfo = &(ctrl->event_queue[ctrl->next_event]);
 			ctrl->next_event = (ctrl->next_event + 1) % 10;
 			taskInfo->hp_slot = hp_slot;
@@ -100,17 +101,17 @@ static u8 handle_switch_change(u8 change, struct controller * ctrl)
 			func->presence_save |= (temp_word >> (hp_slot + 7)) & 0x02;
 
 			if (ctrl->ctrl_int_comp & (0x1L << hp_slot)) {
-				/**********************************
+				/*
 				 * Switch opened
-				 **********************************/
+				 */
 
 				func->switch_save = 0;
 
 				taskInfo->event_type = INT_SWITCH_OPEN;
 			} else {
-				/**********************************
+				/*
 				 * Switch closed
-				 **********************************/
+				 */
 
 				func->switch_save = 0x10;
 
@@ -152,17 +153,17 @@ static u8 handle_presence_change(u16 change, struct controller * ctrl)
 	if (!change)
 		return 0;
 
-	/**********************************
+	/*
 	 * Presence Change
-	 **********************************/
+	 */
 	dbg("cpqsbd:  Presence/Notify input change.\n");
 	dbg("         Changed bits are 0x%4.4x\n", change );
 
 	for (hp_slot = 0; hp_slot < 6; hp_slot++) {
 		if (change & (0x0101 << hp_slot)) {
-			/**********************************
+			/*
 			 * this one changed.
-			 **********************************/
+			 */
 			func = cpqhp_slot_find(ctrl->bus,
 				(hp_slot + ctrl->slot_device_offset), 0);
 
@@ -177,22 +178,23 @@ static u8 handle_presence_change(u16 change, struct controller * ctrl)
 				return 0;
 
 			/* If the switch closed, must be a button
-			 * If not in button mode, nevermind */
+			 * If not in button mode, nevermind
+			 */
 			if (func->switch_save && (ctrl->push_button == 1)) {
 				temp_word = ctrl->ctrl_int_comp >> 16;
 				temp_byte = (temp_word >> hp_slot) & 0x01;
 				temp_byte |= (temp_word >> (hp_slot + 7)) & 0x02;
 
 				if (temp_byte != func->presence_save) {
-					/**************************************
+					/*
 					 * button Pressed (doesn't do anything)
-					 **************************************/
+					 */
 					dbg("hp_slot %d button pressed\n", hp_slot);
 					taskInfo->event_type = INT_BUTTON_PRESS;
 				} else {
-					/**********************************
+					/*
 					 * button Released - TAKE ACTION!!!!
-					 **********************************/
+					 */
 					dbg("hp_slot %d button released\n", hp_slot);
 					taskInfo->event_type = INT_BUTTON_RELEASE;
 
@@ -210,7 +212,8 @@ static u8 handle_presence_change(u16 change, struct controller * ctrl)
 				}
 			} else {
 				/* Switch is open, assume a presence change
-				 * Save the presence state */
+				 * Save the presence state
+				 */
 				temp_word = ctrl->ctrl_int_comp >> 16;
 				func->presence_save = (temp_word >> hp_slot) & 0x01;
 				func->presence_save |= (temp_word >> (hp_slot + 7)) & 0x02;
@@ -241,17 +244,17 @@ static u8 handle_power_fault(u8 change, struct controller * ctrl)
 	if (!change)
 		return 0;
 
-	/**********************************
+	/*
 	 * power fault
-	 **********************************/
+	 */
 
 	info("power fault interrupt\n");
 
 	for (hp_slot = 0; hp_slot < 6; hp_slot++) {
 		if (change & (0x01 << hp_slot)) {
-			/**********************************
+			/*
 			 * this one changed.
-			 **********************************/
+			 */
 			func = cpqhp_slot_find(ctrl->bus,
 				(hp_slot + ctrl->slot_device_offset), 0);
 
@@ -262,16 +265,16 @@ static u8 handle_power_fault(u8 change, struct controller * ctrl)
 			rc++;
 
 			if (ctrl->ctrl_int_comp & (0x00000100 << hp_slot)) {
-				/**********************************
+				/*
 				 * power fault Cleared
-				 **********************************/
+				 */
 				func->status = 0x00;
 
 				taskInfo->event_type = INT_POWER_FAULT_CLEAR;
 			} else {
-				/**********************************
+				/*
 				 * power fault
-				 **********************************/
+				 */
 				taskInfo->event_type = INT_POWER_FAULT;
 
 				if (ctrl->rev < 4) {
@@ -432,13 +435,15 @@ static struct pci_resource *do_pre_bridge_resource_split(struct pci_resource **h
 
 
 	/* If we got here, there the bridge requires some of the resource, but
-	 * we may be able to split some off of the front */
+	 * we may be able to split some off of the front
+	 */
 
 	node = *head;
 
 	if (node->length & (alignment -1)) {
 		/* this one isn't an aligned length, so we'll make a new entry
-		 * and split it up. */
+		 * and split it up.
+		 */
 		split_node = kmalloc(sizeof(*split_node), GFP_KERNEL);
 
 		if (!split_node)
@@ -556,7 +561,8 @@ static struct pci_resource *get_io_resource(struct pci_resource **head, u32 size
 
 		if (node->base & (size - 1)) {
 			/* this one isn't base aligned properly
-			 * so we'll make a new entry and split it up */
+			 * so we'll make a new entry and split it up
+			 */
 			temp_dword = (node->base | (size-1)) + 1;
 
 			/* Short circuit if adjusted size is too small */
@@ -581,7 +587,8 @@ static struct pci_resource *get_io_resource(struct pci_resource **head, u32 size
 		/* Don't need to check if too small since we already did */
 		if (node->length > size) {
 			/* this one is longer than we need
-			 * so we'll make a new entry and split it up */
+			 * so we'll make a new entry and split it up
+			 */
 			split_node = kmalloc(sizeof(*split_node), GFP_KERNEL);
 
 			if (!split_node)
@@ -601,7 +608,8 @@ static struct pci_resource *get_io_resource(struct pci_resource **head, u32 size
 			continue;
 
 		/* If we got here, then it is the right size
-		 * Now take it out of the list and break */
+		 * Now take it out of the list and break
+		 */
 		if (*head == node) {
 			*head = node->next;
 		} else {
@@ -643,13 +651,15 @@ static struct pci_resource *get_max_resource(struct pci_resource **head, u32 siz
 
 	for (max = *head; max; max = max->next) {
 		/* If not big enough we could probably just bail,
-		 * instead we'll continue to the next. */
+		 * instead we'll continue to the next.
+		 */
 		if (max->length < size)
 			continue;
 
 		if (max->base & (size - 1)) {
 			/* this one isn't base aligned properly
-			 * so we'll make a new entry and split it up */
+			 * so we'll make a new entry and split it up
+			 */
 			temp_dword = (max->base | (size-1)) + 1;
 
 			/* Short circuit if adjusted size is too small */
@@ -672,7 +682,8 @@ static struct pci_resource *get_max_resource(struct pci_resource **head, u32 siz
 
 		if ((max->base + max->length) & (size - 1)) {
 			/* this one isn't end aligned properly at the top
-			 * so we'll make a new entry and split it up */
+			 * so we'll make a new entry and split it up
+			 */
 			split_node = kmalloc(sizeof(*split_node), GFP_KERNEL);
 
 			if (!split_node)
@@ -744,7 +755,8 @@ static struct pci_resource *get_resource(struct pci_resource **head, u32 size)
 		if (node->base & (size - 1)) {
 			dbg("%s: not aligned\n", __func__);
 			/* this one isn't base aligned properly
-			 * so we'll make a new entry and split it up */
+			 * so we'll make a new entry and split it up
+			 */
 			temp_dword = (node->base | (size-1)) + 1;
 
 			/* Short circuit if adjusted size is too small */
@@ -769,7 +781,8 @@ static struct pci_resource *get_resource(struct pci_resource **head, u32 size)
 		if (node->length > size) {
 			dbg("%s: too big\n", __func__);
 			/* this one is longer than we need
-			 * so we'll make a new entry and split it up */
+			 * so we'll make a new entry and split it up
+			 */
 			split_node = kmalloc(sizeof(*split_node), GFP_KERNEL);
 
 			if (!split_node)
@@ -888,17 +901,17 @@ irqreturn_t cpqhp_ctrl_intr(int IRQ, void *data)
 
 
 	misc = readw(ctrl->hpc_reg + MISC);
-	/***************************************
+	/*
 	 * Check to see if it was our interrupt
-	 ***************************************/
+	 */
 	if (!(misc & 0x000C)) {
 		return IRQ_NONE;
 	}
 
 	if (misc & 0x0004) {
-		/**********************************
+		/*
 		 * Serial Output interrupt Pending
-		 **********************************/
+		 */
 
 		/* Clear the interrupt */
 		misc |= 0x0004;
@@ -963,7 +976,8 @@ struct pci_func *cpqhp_slot_create(u8 busnumber)
 	new_slot = kzalloc(sizeof(*new_slot), GFP_KERNEL);
 	if (new_slot == NULL) {
 		/* I'm not dead yet!
-		 * You will be. */
+		 * You will be.
+		 */
 		return new_slot;
 	}
 
@@ -1135,7 +1149,8 @@ static u8 set_controller_speed(struct controller *ctrl, u8 adapter_speed, u8 hp_
 		return 0;
 
 	/* We don't allow freq/mode changes if we find another adapter running
-	 * in another slot on this controller */
+	 * in another slot on this controller
+	 */
 	for(slot = ctrl->slot; slot; slot = slot->next) {
 		if (slot->device == (hp_slot + ctrl->slot_device_offset))
 			continue;
@@ -1145,7 +1160,8 @@ static u8 set_controller_speed(struct controller *ctrl, u8 adapter_speed, u8 hp_
 			continue;
 		/* If another adapter is running on the same segment but at a
 		 * lower speed/mode, we allow the new adapter to function at
-		 * this rate if supported */
+		 * this rate if supported
+		 */
 		if (ctrl->speed < adapter_speed)
 			return 0;
 
@@ -1153,7 +1169,8 @@ static u8 set_controller_speed(struct controller *ctrl, u8 adapter_speed, u8 hp_
 	}
 
 	/* If the controller doesn't support freq/mode changes and the
-	 * controller is running at a higher mode, we bail */
+	 * controller is running at a higher mode, we bail
+	 */
 	if ((ctrl->speed > adapter_speed) && (!ctrl->pcix_speed_capability))
 		return 1;
 
@@ -1162,7 +1179,8 @@ static u8 set_controller_speed(struct controller *ctrl, u8 adapter_speed, u8 hp_
 		return 0;
 
 	/* We try to set the max speed supported by both the adapter and
-	 * controller */
+	 * controller
+	 */
 	if (ctrl->speed_capability < adapter_speed) {
 		if (ctrl->speed == ctrl->speed_capability)
 			return 0;
@@ -1244,7 +1262,7 @@ static u8 set_controller_speed(struct controller *ctrl, u8 adapter_speed, u8 hp_
 }
 
 /* the following routines constitute the bulk of the
-   hotplug controller logic
+ * hotplug controller logic
  */
 
 
@@ -1269,14 +1287,14 @@ static u32 board_replaced(struct pci_func *func, struct controller *ctrl)
 	hp_slot = func->device - ctrl->slot_device_offset;
 
 	if (readl(ctrl->hpc_reg + INT_INPUT_CLEAR) & (0x01L << hp_slot)) {
-		/**********************************
+		/*
 		 * The switch is open.
-		 **********************************/
+		 */
 		rc = INTERLOCK_OPEN;
 	} else if (is_slot_enabled (ctrl, hp_slot)) {
-		/**********************************
+		/*
 		 * The board is already on
-		 **********************************/
+		 */
 		rc = CARD_FUNCTIONING;
 	} else {
 		mutex_lock(&ctrl->crit_sect);
@@ -1352,7 +1370,8 @@ static u32 board_replaced(struct pci_func *func, struct controller *ctrl)
 			 * Get slot won't work for devices behind
 			 * bridges, but in this case it will always be
 			 * called for the "base" bus/dev/func of an
-			 * adapter. */
+			 * adapter.
+			 */
 
 			mutex_lock(&ctrl->crit_sect);
 
@@ -1377,7 +1396,8 @@ static u32 board_replaced(struct pci_func *func, struct controller *ctrl)
 
 			 * Get slot won't work for devices behind bridges, but
 			 * in this case it will always be called for the "base"
-			 * bus/dev/func of an adapter. */
+			 * bus/dev/func of an adapter.
+			 */
 
 			mutex_lock(&ctrl->crit_sect);
 
@@ -1434,7 +1454,8 @@ static u32 board_added(struct pci_func *func, struct controller *ctrl)
 	wait_for_ctrl_irq (ctrl);
 
 	/* Change bits in slot power register to force another shift out
-	 * NOTE: this is to work around the timer bug */
+	 * NOTE: this is to work around the timer bug
+	 */
 	temp_byte = readb(ctrl->hpc_reg + SLOT_POWER);
 	writeb(0x00, ctrl->hpc_reg + SLOT_POWER);
 	writeb(temp_byte, ctrl->hpc_reg + SLOT_POWER);
@@ -2484,7 +2505,8 @@ static int configure_new_function(struct controller *ctrl, struct pci_func *func
 		temp_resources.irqs = &irqs;
 
 		/* Make copies of the nodes we are going to pass down so that
-		 * if there is a problem,we can just use these to free resources */
+		 * if there is a problem,we can just use these to free resources
+		 */
 		hold_bus_node = kmalloc(sizeof(*hold_bus_node), GFP_KERNEL);
 		hold_IO_node = kmalloc(sizeof(*hold_IO_node), GFP_KERNEL);
 		hold_mem_node = kmalloc(sizeof(*hold_mem_node), GFP_KERNEL);
@@ -2556,7 +2578,8 @@ static int configure_new_function(struct controller *ctrl, struct pci_func *func
 		temp_word = (p_mem_node->base + p_mem_node->length - 1) >> 16;
 		rc = pci_bus_write_config_word (pci_bus, devfn, PCI_PREF_MEMORY_LIMIT, temp_word);
 
-		/* Adjust this to compensate for extra adjustment in first loop */
+		/* Adjust this to compensate for extra adjustment in first loop
+		 */
 		irqs.barber_pole--;
 
 		rc = 0;
diff --git a/drivers/pci/hotplug/cpqphp_nvram.c b/drivers/pci/hotplug/cpqphp_nvram.c
index 76e110f0e3a6..76ba8a1c774d 100644
--- a/drivers/pci/hotplug/cpqphp_nvram.c
+++ b/drivers/pci/hotplug/cpqphp_nvram.c
@@ -94,12 +94,13 @@ static u8 evbuffer[1024];
 
 static void __iomem *compaq_int15_entry_point;
 
-static spinlock_t int15_lock;		/* lock for ordering int15_bios_call() */
+/* lock for ordering int15_bios_call() */
+static spinlock_t int15_lock;
 
 
 /* This is a series of function that deals with
-   setting & getting the hotplug resource table in some environment variable.
-*/
+ * setting & getting the hotplug resource table in some environment variable.
+ */
 
 /*
  * We really shouldn't be doing this unless there is a _very_ good reason to!!!
@@ -210,14 +211,16 @@ static int load_HRT (void __iomem *rom_start)
 
 	available = 1024;
 
-	// Now load the EV
+	/* Now load the EV */
 	temp_dword = available;
 
 	rc = access_EV(READ_EV, "CQTHPS", evbuffer, &temp_dword);
 
 	evbuffer_length = temp_dword;
 
-	// We're maintaining the resource lists so write FF to invalidate old info
+	/* We're maintaining the resource lists so write FF to invalidate old
+	 * info
+	 */
 	temp_dword = 1;
 
 	rc = access_EV(WRITE_EV, "CQTHPS", &temp_byte, &temp_dword);
@@ -264,12 +267,12 @@ static u32 store_HRT (void __iomem *rom_start)
 
 	ctrl = cpqhp_ctrl_list;
 
-	// The revision of this structure
+	/* The revision of this structure */
 	rc = add_byte( &pFill, 1 + ctrl->push_flag, &usedbytes, &available);
 	if (rc)
 		return(rc);
 
-	// The number of controllers
+	/* The number of controllers */
 	rc = add_byte( &pFill, 1, &usedbytes, &available);
 	if (rc)
 		return(rc);
@@ -279,27 +282,27 @@ static u32 store_HRT (void __iomem *rom_start)
 
 		numCtrl++;
 
-		// The bus number
+		/* The bus number */
 		rc = add_byte( &pFill, ctrl->bus, &usedbytes, &available);
 		if (rc)
 			return(rc);
 
-		// The device Number
+		/* The device Number */
 		rc = add_byte( &pFill, PCI_SLOT(ctrl->pci_dev->devfn), &usedbytes, &available);
 		if (rc)
 			return(rc);
 
-		// The function Number
+		/* The function Number */
 		rc = add_byte( &pFill, PCI_FUNC(ctrl->pci_dev->devfn), &usedbytes, &available);
 		if (rc)
 			return(rc);
 
-		// Skip the number of available entries
+		/* Skip the number of available entries */
 		rc = add_dword( &pFill, 0, &usedbytes, &available);
 		if (rc)
 			return(rc);
 
-		// Figure out memory Available
+		/* Figure out memory Available */
 
 		resNode = ctrl->mem_head;
 
@@ -308,12 +311,12 @@ static u32 store_HRT (void __iomem *rom_start)
 		while (resNode) {
 			loop ++;
 
-			// base
+			/* base */
 			rc = add_dword( &pFill, resNode->base, &usedbytes, &available);
 			if (rc)
 				return(rc);
 
-			// length
+			/* length */
 			rc = add_dword( &pFill, resNode->length, &usedbytes, &available);
 			if (rc)
 				return(rc);
@@ -321,10 +324,10 @@ static u32 store_HRT (void __iomem *rom_start)
 			resNode = resNode->next;
 		}
 
-		// Fill in the number of entries
+		/* Fill in the number of entries */
 		p_ev_ctrl->mem_avail = loop;
 
-		// Figure out prefetchable memory Available
+		/* Figure out prefetchable memory Available */
 
 		resNode = ctrl->p_mem_head;
 
@@ -333,12 +336,12 @@ static u32 store_HRT (void __iomem *rom_start)
 		while (resNode) {
 			loop ++;
 
-			// base
+			/* base */
 			rc = add_dword( &pFill, resNode->base, &usedbytes, &available);
 			if (rc)
 				return(rc);
 
-			// length
+			/* length */
 			rc = add_dword( &pFill, resNode->length, &usedbytes, &available);
 			if (rc)
 				return(rc);
@@ -346,10 +349,10 @@ static u32 store_HRT (void __iomem *rom_start)
 			resNode = resNode->next;
 		}
 
-		// Fill in the number of entries
+		/* Fill in the number of entries */
 		p_ev_ctrl->p_mem_avail = loop;
 
-		// Figure out IO Available
+		/* Figure out IO Available */
 
 		resNode = ctrl->io_head;
 
@@ -358,12 +361,12 @@ static u32 store_HRT (void __iomem *rom_start)
 		while (resNode) {
 			loop ++;
 
-			// base
+			/* base */
 			rc = add_dword( &pFill, resNode->base, &usedbytes, &available);
 			if (rc)
 				return(rc);
 
-			// length
+			/* length */
 			rc = add_dword( &pFill, resNode->length, &usedbytes, &available);
 			if (rc)
 				return(rc);
@@ -371,10 +374,10 @@ static u32 store_HRT (void __iomem *rom_start)
 			resNode = resNode->next;
 		}
 
-		// Fill in the number of entries
+		/* Fill in the number of entries */
 		p_ev_ctrl->io_avail = loop;
 
-		// Figure out bus Available
+		/* Figure out bus Available */
 
 		resNode = ctrl->bus_head;
 
@@ -383,12 +386,12 @@ static u32 store_HRT (void __iomem *rom_start)
 		while (resNode) {
 			loop ++;
 
-			// base
+			/* base */
 			rc = add_dword( &pFill, resNode->base, &usedbytes, &available);
 			if (rc)
 				return(rc);
 
-			// length
+			/* length */
 			rc = add_dword( &pFill, resNode->length, &usedbytes, &available);
 			if (rc)
 				return(rc);
@@ -396,7 +399,7 @@ static u32 store_HRT (void __iomem *rom_start)
 			resNode = resNode->next;
 		}
 
-		// Fill in the number of entries
+		/* Fill in the number of entries */
 		p_ev_ctrl->bus_avail = loop;
 
 		ctrl = ctrl->next;
@@ -404,7 +407,7 @@ static u32 store_HRT (void __iomem *rom_start)
 
 	p_EV_header->num_of_ctrl = numCtrl;
 
-	// Now store the EV
+	/* Now store the EV */
 
 	temp_dword = usedbytes;
 
@@ -449,20 +452,21 @@ int compaq_nvram_load (void __iomem *rom_start, struct controller *ctrl)
 	struct ev_hrt_header *p_EV_header;
 
 	if (!evbuffer_init) {
-		// Read the resource list information in from NVRAM
+		/* Read the resource list information in from NVRAM */
 		if (load_HRT(rom_start))
 			memset (evbuffer, 0, 1024);
 
 		evbuffer_init = 1;
 	}
 
-	// If we saved information in NVRAM, use it now
+	/* If we saved information in NVRAM, use it now */
 	p_EV_header = (struct ev_hrt_header *) evbuffer;
 
-	// The following code is for systems where version 1.0 of this
-	// driver has been loaded, but doesn't support the hardware.
-	// In that case, the driver would incorrectly store something
-	// in NVRAM.
+	/* The following code is for systems where version 1.0 of this
+	 * driver has been loaded, but doesn't support the hardware.
+	 * In that case, the driver would incorrectly store something
+	 * in NVRAM.
+	 */
 	if ((p_EV_header->Version == 2) ||
 	    ((p_EV_header->Version == 1) && !ctrl->push_flag)) {
 		p_byte = &(p_EV_header->next);
@@ -491,7 +495,7 @@ int compaq_nvram_load (void __iomem *rom_start, struct controller *ctrl)
 			if (p_byte > ((u8*)p_EV_header + evbuffer_length))
 				return 2;
 
-			// Skip forward to the next entry
+			/* Skip forward to the next entry */
 			p_byte += (nummem + numpmem + numio + numbus) * 8;
 
 			if (p_byte > ((u8*)p_EV_header + evbuffer_length))
@@ -629,8 +633,9 @@ int compaq_nvram_load (void __iomem *rom_start, struct controller *ctrl)
 			ctrl->bus_head = bus_node;
 		}
 
-		// If all of the following fail, we don't have any resources for
-		// hot plug add
+		/* If all of the following fail, we don't have any resources for
+		 * hot plug add
+		 */
 		rc = 1;
 		rc &= cpqhp_resource_sort_and_combine(&(ctrl->mem_head));
 		rc &= cpqhp_resource_sort_and_combine(&(ctrl->p_mem_head));
diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c
index 573a2702fb6a..2e96bae3c82a 100644
--- a/drivers/pci/hotplug/cpqphp_pci.c
+++ b/drivers/pci/hotplug/cpqphp_pci.c
@@ -178,17 +178,17 @@ int cpqhp_set_irq (u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num)
 		if (!rc)
 			return !rc;
 
-		// set the Edge Level Control Register (ELCR)
+		/* set the Edge Level Control Register (ELCR) */
 		temp_word = inb(0x4d0);
 		temp_word |= inb(0x4d1) << 8;
 
 		temp_word |= 0x01 << irq_num;
 
-		// This should only be for x86 as it sets the Edge Level Control Register
-		outb((u8) (temp_word & 0xFF), 0x4d0);
-		outb((u8) ((temp_word & 0xFF00) >> 8), 0x4d1);
-		rc = 0;
-	}
+		/* This should only be for x86 as it sets the Edge Level
+		 * Control Register
+		 */
+		outb((u8) (temp_word & 0xFF), 0x4d0); outb((u8) ((temp_word &
+		0xFF00) >> 8), 0x4d1); rc = 0; }
 
 	return rc;
 }
@@ -213,11 +213,11 @@ static int PCI_ScanBusForNonBridge(struct controller *ctrl, u8 bus_num, u8 * dev
 	ctrl->pci_bus->number = bus_num;
 
 	for (tdevice = 0; tdevice < 0xFF; tdevice++) {
-		//Scan for access first
+		/* Scan for access first */
 		if (PCI_RefinedAccessConfig(ctrl->pci_bus, tdevice, 0x08, &work) == -1)
 			continue;
 		dbg("Looking for nonbridge bus_num %d dev_num %d\n", bus_num, tdevice);
-		//Yep we got one. Not a bridge ?
+		/* Yep we got one. Not a bridge ? */
 		if ((work >> 8) != PCI_TO_PCI_BRIDGE_CLASS) {
 			*dev_num = tdevice;
 			dbg("found it !\n");
@@ -225,11 +225,11 @@ static int PCI_ScanBusForNonBridge(struct controller *ctrl, u8 bus_num, u8 * dev
 		}
 	}
 	for (tdevice = 0; tdevice < 0xFF; tdevice++) {
-		//Scan for access first
+		/* Scan for access first */
 		if (PCI_RefinedAccessConfig(ctrl->pci_bus, tdevice, 0x08, &work) == -1)
 			continue;
 		dbg("Looking for bridge bus_num %d dev_num %d\n", bus_num, tdevice);
-		//Yep we got one. bridge ?
+		/* Yep we got one. bridge ? */
 		if ((work >> 8) == PCI_TO_PCI_BRIDGE_CLASS) {
 			pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(tdevice, 0), PCI_SECONDARY_BUS, &tbus);
 			dbg("Recurse on bus_num %d tdevice %d\n", tbus, tdevice);
@@ -257,7 +257,7 @@ static int PCI_GetBusDevHelper(struct controller *ctrl, u8 *bus_num, u8 *dev_num
 
 	len = (PCIIRQRoutingInfoLength->size -
 	       sizeof(struct irq_routing_table)) / sizeof(struct irq_info);
-	// Make sure I got at least one entry
+	/* Make sure I got at least one entry */
 	if (len == 0) {
 		kfree(PCIIRQRoutingInfoLength );
 		return -1;
@@ -304,11 +304,14 @@ static int PCI_GetBusDevHelper(struct controller *ctrl, u8 *bus_num, u8 *dev_num
 
 int cpqhp_get_bus_dev (struct controller *ctrl, u8 * bus_num, u8 * dev_num, u8 slot)
 {
-	return PCI_GetBusDevHelper(ctrl, bus_num, dev_num, slot, 0);	//plain (bridges allowed)
+	/* plain (bridges allowed) */
+	return PCI_GetBusDevHelper(ctrl, bus_num, dev_num, slot, 0);
 }
 
 
-/* More PCI configuration routines; this time centered around hotplug controller */
+/* More PCI configuration routines; this time centered around hotplug
+ * controller
+ */
 
 
 /*
@@ -339,12 +342,12 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 	int stop_it;
 	int index;
 
-	//              Decide which slots are supported
+	/* Decide which slots are supported */
 
 	if (is_hot_plug) {
-		//*********************************
-		// is_hot_plug is the slot mask
-		//*********************************
+		/*
+		 * is_hot_plug is the slot mask
+		 */
 		FirstSupported = is_hot_plug >> 4;
 		LastSupported = FirstSupported + (is_hot_plug & 0x0F) - 1;
 	} else {
@@ -352,13 +355,13 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 		LastSupported = 0x1F;
 	}
 
-	//     Save PCI configuration space for all devices in supported slots
+	/* Save PCI configuration space for all devices in supported slots */
 	ctrl->pci_bus->number = busnumber;
 	for (device = FirstSupported; device <= LastSupported; device++) {
 		ID = 0xFFFFFFFF;
 		rc = pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(device, 0), PCI_VENDOR_ID, &ID);
 
-		if (ID != 0xFFFFFFFF) {	  //  device in slot
+		if (ID != 0xFFFFFFFF) {	  /* device in slot */
 			rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, 0), 0x0B, &class_code);
 			if (rc)
 				return rc;
@@ -367,7 +370,7 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 			if (rc)
 				return rc;
 
-			// If multi-function device, set max_functions to 8
+			/* If multi-function device, set max_functions to 8 */
 			if (header_type & 0x80)
 				max_functions = 8;
 			else
@@ -377,18 +380,19 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 
 			do {
 				DevError = 0;
-
-				if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {   // P-P Bridge
-					//  Recurse the subordinate bus
-					//  get the subordinate bus number
+				if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
+					/* Recurse the subordinate bus
+					 * get the subordinate bus number
+					 */
 					rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, function), PCI_SECONDARY_BUS, &secondary_bus);
 					if (rc) {
 						return rc;
 					} else {
 						sub_bus = (int) secondary_bus;
 
-						// Save secondary bus cfg spc
-						// with this recursive call.
+						/* Save secondary bus cfg spc
+						 * with this recursive call.
+						 */
 						rc = cpqhp_save_config(ctrl, sub_bus, 0);
 						if (rc)
 							return rc;
@@ -403,7 +407,7 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 					new_slot = cpqhp_slot_find(busnumber, device, index++);
 
 				if (!new_slot) {
-					// Setup slot structure.
+					/* Setup slot structure. */
 					new_slot = cpqhp_slot_create(busnumber);
 
 					if (new_slot == NULL)
@@ -415,7 +419,7 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 				new_slot->function = (u8) function;
 				new_slot->is_a_board = 1;
 				new_slot->switch_save = 0x10;
-				// In case of unsupported board
+				/* In case of unsupported board */
 				new_slot->status = DevError;
 				new_slot->pci_dev = pci_find_slot(new_slot->bus, (new_slot->device << 3) | new_slot->function);
 
@@ -429,14 +433,15 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 
 				stop_it = 0;
 
-				//  this loop skips to the next present function
-				//  reading in Class Code and Header type.
+				/* this loop skips to the next present function
+				 * reading in Class Code and Header type.
+				 */
 
 				while ((function < max_functions)&&(!stop_it)) {
 					rc = pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(device, function), PCI_VENDOR_ID, &ID);
-					if (ID == 0xFFFFFFFF) {	 // nothing there.
+					if (ID == 0xFFFFFFFF) {	 /* nothing there. */
 						function++;
-					} else {  // Something there
+					} else {  /* Something there */
 						rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, function), 0x0B, &class_code);
 						if (rc)
 							return rc;
@@ -450,9 +455,9 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 				}
 
 			} while (function < max_functions);
-		}		// End of IF (device in slot?)
+		}		/* End of IF (device in slot?) */
 		else if (is_hot_plug) {
-			// Setup slot structure with entry for empty slot
+			/* Setup slot structure with entry for empty slot */
 			new_slot = cpqhp_slot_create(busnumber);
 
 			if (new_slot == NULL) {
@@ -466,7 +471,7 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 			new_slot->presence_save = 0;
 			new_slot->switch_save = 0;
 		}
-	}			// End of FOR loop
+	}			/* End of FOR loop */
 
 	return(0);
 }
@@ -498,11 +503,11 @@ int cpqhp_save_slot_config (struct controller *ctrl, struct pci_func * new_slot)
 	ctrl->pci_bus->number = new_slot->bus;
 	pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), PCI_VENDOR_ID, &ID);
 
-	if (ID != 0xFFFFFFFF) {	  //  device in slot
+	if (ID != 0xFFFFFFFF) {	  /*  device in slot */
 		pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), 0x0B, &class_code);
 		pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), PCI_HEADER_TYPE, &header_type);
 
-		if (header_type & 0x80)	// Multi-function device
+		if (header_type & 0x80)	/* Multi-function device */
 			max_functions = 8;
 		else
 			max_functions = 1;
@@ -510,19 +515,21 @@ int cpqhp_save_slot_config (struct controller *ctrl, struct pci_func * new_slot)
 		function = 0;
 
 		do {
-			if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {	  // PCI-PCI Bridge
-				//  Recurse the subordinate bus
+			if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
+				/*  Recurse the subordinate bus */
 				pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_SECONDARY_BUS, &secondary_bus);
 
 				sub_bus = (int) secondary_bus;
 
-				// Save the config headers for the secondary bus.
+				/* Save the config headers for the secondary
+				 * bus.
+				 */
 				rc = cpqhp_save_config(ctrl, sub_bus, 0);
 				if (rc)
 					return(rc);
 				ctrl->pci_bus->number = new_slot->bus;
 
-			}	// End of IF
+			}	/* End of IF */
 
 			new_slot->status = 0;
 
@@ -534,15 +541,15 @@ int cpqhp_save_slot_config (struct controller *ctrl, struct pci_func * new_slot)
 
 			stop_it = 0;
 
-			//  this loop skips to the next present function
-			//  reading in the Class Code and the Header type.
-
+			/* this loop skips to the next present function
+			 * reading in the Class Code and the Header type.
+			 */
 			while ((function < max_functions) && (!stop_it)) {
 				pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_VENDOR_ID, &ID);
 
-				if (ID == 0xFFFFFFFF) {	 // nothing there.
+				if (ID == 0xFFFFFFFF) {	 /* nothing there. */
 					function++;
-				} else {  // Something there
+				} else {  /* Something there */
 					pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), 0x0B, &class_code);
 
 					pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_HEADER_TYPE, &header_type);
@@ -552,7 +559,7 @@ int cpqhp_save_slot_config (struct controller *ctrl, struct pci_func * new_slot)
 			}
 
 		} while (function < max_functions);
-	}			// End of IF (device in slot?)
+	}			/* End of IF (device in slot?) */
 	else {
 		return 2;
 	}
@@ -590,11 +597,10 @@ int cpqhp_save_base_addr_length(struct controller *ctrl, struct pci_func * func)
 		pci_bus->number = func->bus;
 		devfn = PCI_DEVFN(func->device, func->function);
 
-		// Check for Bridge
+		/* Check for Bridge */
 		pci_bus_read_config_byte (pci_bus, devfn, PCI_HEADER_TYPE, &header_type);
 
 		if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
-			// PCI-PCI Bridge
 			pci_bus_read_config_byte (pci_bus, devfn, PCI_SECONDARY_BUS, &secondary_bus);
 
 			sub_bus = (int) secondary_bus;
@@ -610,23 +616,27 @@ int cpqhp_save_base_addr_length(struct controller *ctrl, struct pci_func * func)
 			}
 			pci_bus->number = func->bus;
 
-			//FIXME: this loop is duplicated in the non-bridge case.  The two could be rolled together
-			// Figure out IO and memory base lengths
+			/* FIXME: this loop is duplicated in the non-bridge
+			 * case.  The two could be rolled together Figure out
+			 * IO and memory base lengths
+			 */
 			for (cloop = 0x10; cloop <= 0x14; cloop += 4) {
 				temp_register = 0xFFFFFFFF;
 				pci_bus_write_config_dword (pci_bus, devfn, cloop, temp_register);
 				pci_bus_read_config_dword (pci_bus, devfn, cloop, &base);
-
-				if (base) {  // If this register is implemented
+				/* If this register is implemented */
+				if (base) {
 					if (base & 0x01L) {
-						// IO base
-						// set base = amount of IO space requested
+						/* IO base
+						 * set base = amount of IO space
+						 * requested
+						 */
 						base = base & 0xFFFFFFFE;
 						base = (~base) + 1;
 
 						type = 1;
 					} else {
-						// memory base
+						/* memory base */
 						base = base & 0xFFFFFFF0;
 						base = (~base) + 1;
 
@@ -637,32 +647,36 @@ int cpqhp_save_base_addr_length(struct controller *ctrl, struct pci_func * func)
 					type = 0;
 				}
 
-				// Save information in slot structure
+				/* Save information in slot structure */
 				func->base_length[(cloop - 0x10) >> 2] =
 				base;
 				func->base_type[(cloop - 0x10) >> 2] = type;
 
-			}	// End of base register loop
-
+			}	/* End of base register loop */
 
-		} else if ((header_type & 0x7F) == 0x00) {	  // PCI-PCI Bridge
-			// Figure out IO and memory base lengths
+		} else if ((header_type & 0x7F) == 0x00) {
+			/* Figure out IO and memory base lengths */
 			for (cloop = 0x10; cloop <= 0x24; cloop += 4) {
 				temp_register = 0xFFFFFFFF;
 				pci_bus_write_config_dword (pci_bus, devfn, cloop, temp_register);
 				pci_bus_read_config_dword (pci_bus, devfn, cloop, &base);
 
-				if (base) {  // If this register is implemented
+				/* If this register is implemented */
+				if (base) {
 					if (base & 0x01L) {
-						// IO base
-						// base = amount of IO space requested
+						/* IO base
+						 * base = amount of IO space
+						 * requested
+						 */
 						base = base & 0xFFFFFFFE;
 						base = (~base) + 1;
 
 						type = 1;
 					} else {
-						// memory base
-						// base = amount of memory space requested
+						/* memory base
+						 * base = amount of memory
+						 * space requested
+						 */
 						base = base & 0xFFFFFFF0;
 						base = (~base) + 1;
 
@@ -673,16 +687,16 @@ int cpqhp_save_base_addr_length(struct controller *ctrl, struct pci_func * func)
 					type = 0;
 				}
 
-				// Save information in slot structure
+				/* Save information in slot structure */
 				func->base_length[(cloop - 0x10) >> 2] = base;
 				func->base_type[(cloop - 0x10) >> 2] = type;
 
-			}	// End of base register loop
+			}	/* End of base register loop */
 
-		} else {	  // Some other unknown header type
+		} else {	  /* Some other unknown header type */
 		}
 
-		// find the next device in this slot
+		/* find the next device in this slot */
 		func = cpqhp_slot_find(func->bus, func->device, index++);
 	}
 
@@ -728,18 +742,18 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 		pci_bus->number = func->bus;
 		devfn = PCI_DEVFN(func->device, func->function);
 
-		// Save the command register
+		/* Save the command register */
 		pci_bus_read_config_word(pci_bus, devfn, PCI_COMMAND, &save_command);
 
-		// disable card
+		/* disable card */
 		command = 0x00;
 		pci_bus_write_config_word(pci_bus, devfn, PCI_COMMAND, command);
 
-		// Check for Bridge
+		/* Check for Bridge */
 		pci_bus_read_config_byte(pci_bus, devfn, PCI_HEADER_TYPE, &header_type);
 
-		if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {	  // PCI-PCI Bridge
-			// Clear Bridge Control Register
+		if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
+			/* Clear Bridge Control Register */
 			command = 0x00;
 			pci_bus_write_config_word(pci_bus, devfn, PCI_BRIDGE_CONTROL, command);
 			pci_bus_read_config_byte(pci_bus, devfn, PCI_SECONDARY_BUS, &secondary_bus);
@@ -755,7 +769,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 			bus_node->next = func->bus_head;
 			func->bus_head = bus_node;
 
-			// Save IO base and Limit registers
+			/* Save IO base and Limit registers */
 			pci_bus_read_config_byte(pci_bus, devfn, PCI_IO_BASE, &b_base);
 			pci_bus_read_config_byte(pci_bus, devfn, PCI_IO_LIMIT, &b_length);
 
@@ -771,7 +785,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 				func->io_head = io_node;
 			}
 
-			// Save memory base and Limit registers
+			/* Save memory base and Limit registers */
 			pci_bus_read_config_word(pci_bus, devfn, PCI_MEMORY_BASE, &w_base);
 			pci_bus_read_config_word(pci_bus, devfn, PCI_MEMORY_LIMIT, &w_length);
 
@@ -787,7 +801,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 				func->mem_head = mem_node;
 			}
 
-			// Save prefetchable memory base and Limit registers
+			/* Save prefetchable memory base and Limit registers */
 			pci_bus_read_config_word(pci_bus, devfn, PCI_PREF_MEMORY_BASE, &w_base);
 			pci_bus_read_config_word(pci_bus, devfn, PCI_PREF_MEMORY_LIMIT, &w_length);
 
@@ -802,7 +816,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 				p_mem_node->next = func->p_mem_head;
 				func->p_mem_head = p_mem_node;
 			}
-			// Figure out IO and memory base lengths
+			/* Figure out IO and memory base lengths */
 			for (cloop = 0x10; cloop <= 0x14; cloop += 4) {
 				pci_bus_read_config_dword (pci_bus, devfn, cloop, &save_base);
 
@@ -812,11 +826,14 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 
 				temp_register = base;
 
-				if (base) {  // If this register is implemented
+				/* If this register is implemented */
+				if (base) {
 					if (((base & 0x03L) == 0x01)
 					    && (save_command & 0x01)) {
-						// IO base
-						// set temp_register = amount of IO space requested
+						/* IO base
+						 * set temp_register = amount
+						 * of IO space requested
+						 */
 						temp_register = base & 0xFFFFFFFE;
 						temp_register = (~temp_register) + 1;
 
@@ -834,7 +851,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 					} else
 						if (((base & 0x0BL) == 0x08)
 						    && (save_command & 0x02)) {
-						// prefetchable memory base
+						/* prefetchable memory base */
 						temp_register = base & 0xFFFFFFF0;
 						temp_register = (~temp_register) + 1;
 
@@ -851,7 +868,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 					} else
 						if (((base & 0x0BL) == 0x00)
 						    && (save_command & 0x02)) {
-						// prefetchable memory base
+						/* prefetchable memory base */
 						temp_register = base & 0xFFFFFFF0;
 						temp_register = (~temp_register) + 1;
 
@@ -868,9 +885,10 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 					} else
 						return(1);
 				}
-			}	// End of base register loop
-		} else if ((header_type & 0x7F) == 0x00) {	  // Standard header
-			// Figure out IO and memory base lengths
+			}	/* End of base register loop */
+		/* Standard header */
+		} else if ((header_type & 0x7F) == 0x00) {
+			/* Figure out IO and memory base lengths */
 			for (cloop = 0x10; cloop <= 0x24; cloop += 4) {
 				pci_bus_read_config_dword(pci_bus, devfn, cloop, &save_base);
 
@@ -880,11 +898,14 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 
 				temp_register = base;
 
-				if (base) {	  // If this register is implemented
+				/* If this register is implemented */
+				if (base) {
 					if (((base & 0x03L) == 0x01)
 					    && (save_command & 0x01)) {
-						// IO base
-						// set temp_register = amount of IO space requested
+						/* IO base
+						 * set temp_register = amount
+						 * of IO space requested
+						 */
 						temp_register = base & 0xFFFFFFFE;
 						temp_register = (~temp_register) + 1;
 
@@ -901,7 +922,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 					} else
 						if (((base & 0x0BL) == 0x08)
 						    && (save_command & 0x02)) {
-						// prefetchable memory base
+						/* prefetchable memory base */
 						temp_register = base & 0xFFFFFFF0;
 						temp_register = (~temp_register) + 1;
 
@@ -918,7 +939,7 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 					} else
 						if (((base & 0x0BL) == 0x00)
 						    && (save_command & 0x02)) {
-						// prefetchable memory base
+						/* prefetchable memory base */
 						temp_register = base & 0xFFFFFFF0;
 						temp_register = (~temp_register) + 1;
 
@@ -935,11 +956,12 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 					} else
 						return(1);
 				}
-			}	// End of base register loop
-		} else {	  // Some other unknown header type
+			}	/* End of base register loop */
+		/* Some other unknown header type */
+		} else {
 		}
 
-		// find the next device in this slot
+		/* find the next device in this slot */
 		func = cpqhp_slot_find(func->bus, func->device, index++);
 	}
 
@@ -975,16 +997,17 @@ int cpqhp_configure_board(struct controller *ctrl, struct pci_func * func)
 		pci_bus->number = func->bus;
 		devfn = PCI_DEVFN(func->device, func->function);
 
-		// Start at the top of config space so that the control
-		// registers are programmed last
+		/* Start at the top of config space so that the control
+		 * registers are programmed last
+		 */
 		for (cloop = 0x3C; cloop > 0; cloop -= 4) {
 			pci_bus_write_config_dword (pci_bus, devfn, cloop, func->config_space[cloop >> 2]);
 		}
 
 		pci_bus_read_config_byte (pci_bus, devfn, PCI_HEADER_TYPE, &header_type);
 
-		// If this is a bridge device, restore subordinate devices
-		if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {	  // PCI-PCI Bridge
+		/* If this is a bridge device, restore subordinate devices */
+		if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
 			pci_bus_read_config_byte (pci_bus, devfn, PCI_SECONDARY_BUS, &secondary_bus);
 
 			sub_bus = (int) secondary_bus;
@@ -1000,8 +1023,9 @@ int cpqhp_configure_board(struct controller *ctrl, struct pci_func * func)
 			}
 		} else {
 
-			// Check all the base Address Registers to make sure
-			// they are the same.  If not, the board is different.
+			/* Check all the base Address Registers to make sure
+			 * they are the same.  If not, the board is different.
+			 */
 
 			for (cloop = 16; cloop < 40; cloop += 4) {
 				pci_bus_read_config_dword (pci_bus, devfn, cloop, &temp);
@@ -1058,27 +1082,28 @@ int cpqhp_valid_replace(struct controller *ctrl, struct pci_func * func)
 
 		pci_bus_read_config_dword (pci_bus, devfn, PCI_VENDOR_ID, &temp_register);
 
-		// No adapter present
+		/* No adapter present */
 		if (temp_register == 0xFFFFFFFF)
 			return(NO_ADAPTER_PRESENT);
 
 		if (temp_register != func->config_space[0])
 			return(ADAPTER_NOT_SAME);
 
-		// Check for same revision number and class code
+		/* Check for same revision number and class code */
 		pci_bus_read_config_dword (pci_bus, devfn, PCI_CLASS_REVISION, &temp_register);
 
-		// Adapter not the same
+		/* Adapter not the same */
 		if (temp_register != func->config_space[0x08 >> 2])
 			return(ADAPTER_NOT_SAME);
 
-		// Check for Bridge
+		/* Check for Bridge */
 		pci_bus_read_config_byte (pci_bus, devfn, PCI_HEADER_TYPE, &header_type);
 
-		if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {	  // PCI-PCI Bridge
-			// In order to continue checking, we must program the
-			// bus registers in the bridge to respond to accesses
-			// for it's subordinate bus(es)
+		if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
+			/* In order to continue checking, we must program the
+			 * bus registers in the bridge to respond to accesses
+			 * for its subordinate bus(es)
+			 */
 
 			temp_register = func->config_space[0x18 >> 2];
 			pci_bus_write_config_dword (pci_bus, devfn, PCI_PRIMARY_BUS, temp_register);
@@ -1096,35 +1121,39 @@ int cpqhp_valid_replace(struct controller *ctrl, struct pci_func * func)
 			}
 
 		}
-		// Check to see if it is a standard config header
+		/* Check to see if it is a standard config header */
 		else if ((header_type & 0x7F) == PCI_HEADER_TYPE_NORMAL) {
-			// Check subsystem vendor and ID
+			/* Check subsystem vendor and ID */
 			pci_bus_read_config_dword (pci_bus, devfn, PCI_SUBSYSTEM_VENDOR_ID, &temp_register);
 
 			if (temp_register != func->config_space[0x2C >> 2]) {
-				// If it's a SMART-2 and the register isn't filled
-				// in, ignore the difference because
-				// they just have an old rev of the firmware
-
+				/* If it's a SMART-2 and the register isn't
+				 * filled in, ignore the difference because
+				 * they just have an old rev of the firmware
+				 */
 				if (!((func->config_space[0] == 0xAE100E11)
 				      && (temp_register == 0x00L)))
 					return(ADAPTER_NOT_SAME);
 			}
-			// Figure out IO and memory base lengths
+			/* Figure out IO and memory base lengths */
 			for (cloop = 0x10; cloop <= 0x24; cloop += 4) {
 				temp_register = 0xFFFFFFFF;
 				pci_bus_write_config_dword (pci_bus, devfn, cloop, temp_register);
 				pci_bus_read_config_dword (pci_bus, devfn, cloop, &base);
-				if (base) {	  // If this register is implemented
+
+				/* If this register is implemented */
+				if (base) {
 					if (base & 0x01L) {
-						// IO base
-						// set base = amount of IO space requested
+						/* IO base
+						 * set base = amount of IO
+						 * space requested
+						 */
 						base = base & 0xFFFFFFFE;
 						base = (~base) + 1;
 
 						type = 1;
 					} else {
-						// memory base
+						/* memory base */
 						base = base & 0xFFFFFFF0;
 						base = (~base) + 1;
 
@@ -1135,23 +1164,24 @@ int cpqhp_valid_replace(struct controller *ctrl, struct pci_func * func)
 					type = 0;
 				}
 
-				// Check information in slot structure
+				/* Check information in slot structure */
 				if (func->base_length[(cloop - 0x10) >> 2] != base)
 					return(ADAPTER_NOT_SAME);
 
 				if (func->base_type[(cloop - 0x10) >> 2] != type)
 					return(ADAPTER_NOT_SAME);
 
-			}	// End of base register loop
+			}	/* End of base register loop */
 
-		}		// End of (type 0 config space) else
+		}		/* End of (type 0 config space) else */
 		else {
-			// this is not a type 0 or 1 config space header so
-			// we don't know how to do it
+			/* this is not a type 0 or 1 config space header so
+			 * we don't know how to do it
+			 */
 			return(DEVICE_TYPE_NOT_SUPPORTED);
 		}
 
-		// Get the next function
+		/* Get the next function */
 		func = cpqhp_slot_find(func->bus, func->device, index++);
 	}
 
@@ -1190,7 +1220,7 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st
 	if (rom_resource_table == NULL) {
 		return -ENODEV;
 	}
-	// Sum all resources and setup resource maps
+	/* Sum all resources and setup resource maps */
 	unused_IRQ = readl(rom_resource_table + UNUSED_IRQ);
 	dbg("unused_IRQ = %x\n", unused_IRQ);
 
@@ -1262,13 +1292,13 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st
 		    dev_func, io_base, io_length, mem_base, mem_length, pre_mem_base, pre_mem_length,
 		    primary_bus, secondary_bus, max_bus);
 
-		// If this entry isn't for our controller's bus, ignore it
+		/* If this entry isn't for our controller's bus, ignore it */
 		if (primary_bus != ctrl->bus) {
 			i--;
 			one_slot += sizeof (struct slot_rt);
 			continue;
 		}
-		// find out if this entry is for an occupied slot
+		/* find out if this entry is for an occupied slot */
 		ctrl->pci_bus->number = primary_bus;
 		pci_bus_read_config_dword (ctrl->pci_bus, dev_func, PCI_VENDOR_ID, &temp_dword);
 		dbg("temp_D_word = %x\n", temp_dword);
@@ -1282,13 +1312,13 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st
 				func = cpqhp_slot_find(primary_bus, dev_func >> 3, index++);
 			}
 
-			// If we can't find a match, skip this table entry
+			/* If we can't find a match, skip this table entry */
 			if (!func) {
 				i--;
 				one_slot += sizeof (struct slot_rt);
 				continue;
 			}
-			// this may not work and shouldn't be used
+			/* this may not work and shouldn't be used */
 			if (secondary_bus != primary_bus)
 				bridged_slot = 1;
 			else
@@ -1301,7 +1331,7 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st
 		}
 
 
-		// If we've got a valid IO base, use it
+		/* If we've got a valid IO base, use it */
 
 		temp_dword = io_base + io_length;
 
@@ -1325,7 +1355,7 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st
 			}
 		}
 
-		// If we've got a valid memory base, use it
+		/* If we've got a valid memory base, use it */
 		temp_dword = mem_base + mem_length;
 		if ((mem_base) && (temp_dword < 0x10000)) {
 			mem_node = kmalloc(sizeof(*mem_node), GFP_KERNEL);
@@ -1348,8 +1378,9 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st
 			}
 		}
 
-		// If we've got a valid prefetchable memory base, and
-		// the base + length isn't greater than 0xFFFF
+		/* If we've got a valid prefetchable memory base, and
+		 * the base + length isn't greater than 0xFFFF
+		 */
 		temp_dword = pre_mem_base + pre_mem_length;
 		if ((pre_mem_base) && (temp_dword < 0x10000)) {
 			p_mem_node = kmalloc(sizeof(*p_mem_node), GFP_KERNEL);
@@ -1372,9 +1403,10 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st
 			}
 		}
 
-		// If we've got a valid bus number, use it
-		// The second condition is to ignore bus numbers on
-		// populated slots that don't have PCI-PCI bridges
+		/* If we've got a valid bus number, use it
+		 * The second condition is to ignore bus numbers on
+		 * populated slots that don't have PCI-PCI bridges
+		 */
 		if (secondary_bus && (secondary_bus != primary_bus)) {
 			bus_node = kmalloc(sizeof(*bus_node), GFP_KERNEL);
 			if (!bus_node)
@@ -1398,8 +1430,9 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st
 		one_slot += sizeof (struct slot_rt);
 	}
 
-	// If all of the following fail, we don't have any resources for
-	// hot plug add
+	/* If all of the following fail, we don't have any resources for
+	 * hot plug add
+	 */
 	rc = 1;
 	rc &= cpqhp_resource_sort_and_combine(&(ctrl->mem_head));
 	rc &= cpqhp_resource_sort_and_combine(&(ctrl->p_mem_head));
-- 
cgit v1.2.3-59-g8ed1b


From 86a58023e4078a843f8ca8a9b6fa23542d881f99 Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:23:21 -0600
Subject: PCI Hotplug: cpqphp: obey 80 column convention in cpqphp.h

Clean up cpqphp.h to follow 80 column convention.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp.h | 76 ++++++++++++++++++++++++++------------------
 1 file changed, 45 insertions(+), 31 deletions(-)

diff --git a/drivers/pci/hotplug/cpqphp.h b/drivers/pci/hotplug/cpqphp.h
index e15d657368f0..308f82b1fc9a 100644
--- a/drivers/pci/hotplug/cpqphp.h
+++ b/drivers/pci/hotplug/cpqphp.h
@@ -405,40 +405,50 @@ struct resource_lists {
 
 
 /* debugfs functions for the hotplug controller info */
-extern void cpqhp_initialize_debugfs		(void);
-extern void cpqhp_shutdown_debugfs		(void);
-extern void cpqhp_create_debugfs_files		(struct controller *ctrl);
-extern void cpqhp_remove_debugfs_files		(struct controller *ctrl);
+extern void cpqhp_initialize_debugfs(void);
+extern void cpqhp_shutdown_debugfs(void);
+extern void cpqhp_create_debugfs_files(struct controller *ctrl);
+extern void cpqhp_remove_debugfs_files(struct controller *ctrl);
 
 /* controller functions */
-extern void	cpqhp_pushbutton_thread		(unsigned long event_pointer);
-extern irqreturn_t cpqhp_ctrl_intr		(int IRQ, void *data);
-extern int	cpqhp_find_available_resources	(struct controller *ctrl, void __iomem *rom_start);
-extern int	cpqhp_event_start_thread	(void);
-extern void	cpqhp_event_stop_thread		(void);
-extern struct pci_func *cpqhp_slot_create	(unsigned char busnumber);
-extern struct pci_func *cpqhp_slot_find		(unsigned char bus, unsigned char device, unsigned char index);
-extern int	cpqhp_process_SI		(struct controller *ctrl, struct pci_func *func);
-extern int	cpqhp_process_SS		(struct controller *ctrl, struct pci_func *func);
-extern int	cpqhp_hardware_test		(struct controller *ctrl, int test_num);
+extern void cpqhp_pushbutton_thread(unsigned long event_pointer);
+extern irqreturn_t cpqhp_ctrl_intr(int IRQ, void *data);
+extern int cpqhp_find_available_resources(struct controller *ctrl,
+					  void __iomem *rom_start);
+extern int cpqhp_event_start_thread(void);
+extern void cpqhp_event_stop_thread(void);
+extern struct pci_func *cpqhp_slot_create(unsigned char busnumber);
+extern struct pci_func *cpqhp_slot_find(unsigned char bus, unsigned char device,
+					unsigned char index);
+extern int cpqhp_process_SI(struct controller *ctrl, struct pci_func *func);
+extern int cpqhp_process_SS(struct controller *ctrl, struct pci_func *func);
+extern int cpqhp_hardware_test(struct controller *ctrl, int test_num);
 
 /* resource functions */
 extern int	cpqhp_resource_sort_and_combine	(struct pci_resource **head);
 
 /* pci functions */
-extern int	cpqhp_set_irq			(u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num);
-extern int	cpqhp_get_bus_dev		(struct controller *ctrl, u8 *bus_num, u8 *dev_num, u8 slot);
-extern int	cpqhp_save_config		(struct controller *ctrl, int busnumber, int is_hot_plug);
-extern int	cpqhp_save_base_addr_length	(struct controller *ctrl, struct pci_func * func);
-extern int	cpqhp_save_used_resources	(struct controller *ctrl, struct pci_func * func);
-extern int	cpqhp_configure_board		(struct controller *ctrl, struct pci_func * func);
-extern int	cpqhp_save_slot_config		(struct controller *ctrl, struct pci_func * new_slot);
-extern int	cpqhp_valid_replace		(struct controller *ctrl, struct pci_func * func);
-extern void	cpqhp_destroy_board_resources	(struct pci_func * func);
-extern int	cpqhp_return_board_resources	(struct pci_func * func, struct resource_lists * resources);
-extern void	cpqhp_destroy_resource_list	(struct resource_lists * resources);
-extern int	cpqhp_configure_device		(struct controller* ctrl, struct pci_func* func);
-extern int	cpqhp_unconfigure_device	(struct pci_func* func);
+extern int cpqhp_set_irq(u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num);
+extern int cpqhp_get_bus_dev(struct controller *ctrl, u8 *bus_num, u8 *dev_num,
+			     u8 slot);
+extern int cpqhp_save_config(struct controller *ctrl, int busnumber,
+			     int is_hot_plug);
+extern int cpqhp_save_base_addr_length(struct controller *ctrl,
+				       struct pci_func *func);
+extern int cpqhp_save_used_resources(struct controller *ctrl,
+				     struct pci_func *func);
+extern int cpqhp_configure_board(struct controller *ctrl,
+				 struct pci_func *func);
+extern int cpqhp_save_slot_config(struct controller *ctrl,
+				  struct pci_func *new_slot);
+extern int cpqhp_valid_replace(struct controller *ctrl, struct pci_func *func);
+extern void cpqhp_destroy_board_resources(struct pci_func *func);
+extern int cpqhp_return_board_resources	(struct pci_func *func,
+					 struct resource_lists *resources);
+extern void cpqhp_destroy_resource_list(struct resource_lists *resources);
+extern int cpqhp_configure_device(struct controller *ctrl,
+				  struct pci_func *func);
+extern int cpqhp_unconfigure_device(struct pci_func *func);
 
 /* Global variables */
 extern int cpqhp_debug;
@@ -463,7 +473,8 @@ static inline char *slot_name(struct slot *slot)
  *
  * Puts node back in the resource list pointed to by head
  */
-static inline void return_resource(struct pci_resource **head, struct pci_resource *node)
+static inline void return_resource(struct pci_resource **head,
+				   struct pci_resource *node)
 {
 	if (!node || !head)
 		return;
@@ -673,7 +684,8 @@ static inline int get_slot_enabled(struct controller *ctrl, struct slot *slot)
 }
 
 
-static inline int cpq_get_latch_status(struct controller *ctrl, struct slot *slot)
+static inline int cpq_get_latch_status(struct controller *ctrl,
+				       struct slot *slot)
 {
 	u32 status;
 	u8 hp_slot;
@@ -688,7 +700,8 @@ static inline int cpq_get_latch_status(struct controller *ctrl, struct slot *slo
 }
 
 
-static inline int get_presence_status(struct controller *ctrl, struct slot *slot)
+static inline int get_presence_status(struct controller *ctrl,
+				      struct slot *slot)
 {
 	int presence_save = 0;
 	u8 hp_slot;
@@ -697,7 +710,8 @@ static inline int get_presence_status(struct controller *ctrl, struct slot *slot
 	hp_slot = slot->device - ctrl->slot_device_offset;
 
 	tempdword = readl(ctrl->hpc_reg + INT_INPUT_CLEAR);
-	presence_save = (int) ((((~tempdword) >> 23) | ((~tempdword) >> 15)) >> hp_slot) & 0x02;
+	presence_save = (int) ((((~tempdword) >> 23) | ((~tempdword) >> 15))
+				>> hp_slot) & 0x02;
 
 	return presence_save;
 }
-- 
cgit v1.2.3-59-g8ed1b


From b4d897a48d451db0ab6a4ebf8c28eb314eba0280 Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:23:26 -0600
Subject: PCI Hotplug: cpqphp: remove useless prototypes in cpqphp_core.c

Impact: refactor

Refactor code to follow convention more closely and eliminate the need
for some useless prototypes.

No functional change.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp_core.c | 461 ++++++++++++++++++--------------------
 1 file changed, 224 insertions(+), 237 deletions(-)

diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c
index 91dc95850cce..f05ea7a5606b 100644
--- a/drivers/pci/hotplug/cpqphp_core.c
+++ b/drivers/pci/hotplug/cpqphp_core.c
@@ -77,33 +77,6 @@ MODULE_PARM_DESC(debug, "Debugging mode enabled or not");
 
 #define CPQHPC_MODULE_MINOR 208
 
-static int one_time_init	(void);
-static int set_attention_status	(struct hotplug_slot *slot, u8 value);
-static int process_SI		(struct hotplug_slot *slot);
-static int process_SS		(struct hotplug_slot *slot);
-static int hardware_test	(struct hotplug_slot *slot, u32 value);
-static int get_power_status	(struct hotplug_slot *slot, u8 *value);
-static int get_attention_status	(struct hotplug_slot *slot, u8 *value);
-static int get_latch_status	(struct hotplug_slot *slot, u8 *value);
-static int get_adapter_status	(struct hotplug_slot *slot, u8 *value);
-static int get_max_bus_speed	(struct hotplug_slot *slot, enum pci_bus_speed *value);
-static int get_cur_bus_speed	(struct hotplug_slot *slot, enum pci_bus_speed *value);
-
-static struct hotplug_slot_ops cpqphp_hotplug_slot_ops = {
-	.owner =		THIS_MODULE,
-	.set_attention_status =	set_attention_status,
-	.enable_slot =		process_SI,
-	.disable_slot =		process_SS,
-	.hardware_test =	hardware_test,
-	.get_power_status =	get_power_status,
-	.get_attention_status =	get_attention_status,
-	.get_latch_status =	get_latch_status,
-	.get_adapter_status =	get_adapter_status,
-	.get_max_bus_speed =	get_max_bus_speed,
-	.get_cur_bus_speed =	get_cur_bus_speed,
-};
-
-
 static inline int is_slot64bit(struct slot *slot)
 {
 	return (readb(slot->p_sm_slot + SMBIOS_SLOT_WIDTH) == 0x06) ? 1 : 0;
@@ -322,145 +295,6 @@ static void release_slot(struct hotplug_slot *hotplug_slot)
 	kfree(slot);
 }
 
-#define SLOT_NAME_SIZE 10
-
-static int ctrl_slot_setup(struct controller *ctrl,
-			void __iomem *smbios_start,
-			void __iomem *smbios_table)
-{
-	struct slot *slot;
-	struct hotplug_slot *hotplug_slot;
-	struct hotplug_slot_info *hotplug_slot_info;
-	u8 number_of_slots;
-	u8 slot_device;
-	u8 slot_number;
-	u8 ctrl_slot;
-	u32 tempdword;
-	char name[SLOT_NAME_SIZE];
-	void __iomem *slot_entry= NULL;
-	int result = -ENOMEM;
-
-	dbg("%s\n", __func__);
-
-	tempdword = readl(ctrl->hpc_reg + INT_INPUT_CLEAR);
-
-	number_of_slots = readb(ctrl->hpc_reg + SLOT_MASK) & 0x0F;
-	slot_device = readb(ctrl->hpc_reg + SLOT_MASK) >> 4;
-	slot_number = ctrl->first_slot;
-
-	while (number_of_slots) {
-		slot = kzalloc(sizeof(*slot), GFP_KERNEL);
-		if (!slot)
-			goto error;
-
-		slot->hotplug_slot = kzalloc(sizeof(*(slot->hotplug_slot)),
-						GFP_KERNEL);
-		if (!slot->hotplug_slot)
-			goto error_slot;
-		hotplug_slot = slot->hotplug_slot;
-
-		hotplug_slot->info =
-				kzalloc(sizeof(*(hotplug_slot->info)),
-							GFP_KERNEL);
-		if (!hotplug_slot->info)
-			goto error_hpslot;
-		hotplug_slot_info = hotplug_slot->info;
-
-		slot->ctrl = ctrl;
-		slot->bus = ctrl->bus;
-		slot->device = slot_device;
-		slot->number = slot_number;
-		dbg("slot->number = %u\n", slot->number);
-
-		slot_entry = get_SMBIOS_entry(smbios_start, smbios_table, 9,
-					slot_entry);
-
-		while (slot_entry && (readw(slot_entry + SMBIOS_SLOT_NUMBER) !=
-				slot->number)) {
-			slot_entry = get_SMBIOS_entry(smbios_start,
-						smbios_table, 9, slot_entry);
-		}
-
-		slot->p_sm_slot = slot_entry;
-
-		init_timer(&slot->task_event);
-		slot->task_event.expires = jiffies + 5 * HZ;
-		slot->task_event.function = cpqhp_pushbutton_thread;
-
-		/*FIXME: these capabilities aren't used but if they are
-		 *	 they need to be correctly implemented
-		 */
-		slot->capabilities |= PCISLOT_REPLACE_SUPPORTED;
-		slot->capabilities |= PCISLOT_INTERLOCK_SUPPORTED;
-
-		if (is_slot64bit(slot))
-			slot->capabilities |= PCISLOT_64_BIT_SUPPORTED;
-		if (is_slot66mhz(slot))
-			slot->capabilities |= PCISLOT_66_MHZ_SUPPORTED;
-		if (ctrl->speed == PCI_SPEED_66MHz)
-			slot->capabilities |= PCISLOT_66_MHZ_OPERATION;
-
-		ctrl_slot =
-			slot_device - (readb(ctrl->hpc_reg + SLOT_MASK) >> 4);
-
-		/* Check presence */
-		slot->capabilities |=
-			((((~tempdword) >> 23) |
-			 ((~tempdword) >> 15)) >> ctrl_slot) & 0x02;
-		/* Check the switch state */
-		slot->capabilities |=
-			((~tempdword & 0xFF) >> ctrl_slot) & 0x01;
-		/* Check the slot enable */
-		slot->capabilities |=
-			((read_slot_enable(ctrl) << 2) >> ctrl_slot) & 0x04;
-
-		/* register this slot with the hotplug pci core */
-		hotplug_slot->release = &release_slot;
-		hotplug_slot->private = slot;
-		snprintf(name, SLOT_NAME_SIZE, "%u", slot->number);
-		hotplug_slot->ops = &cpqphp_hotplug_slot_ops;
-
-		hotplug_slot_info->power_status = get_slot_enabled(ctrl, slot);
-		hotplug_slot_info->attention_status =
-			cpq_get_attention_status(ctrl, slot);
-		hotplug_slot_info->latch_status =
-			cpq_get_latch_status(ctrl, slot);
-		hotplug_slot_info->adapter_status =
-			get_presence_status(ctrl, slot);
-
-		dbg("registering bus %d, dev %d, number %d, "
-				"ctrl->slot_device_offset %d, slot %d\n",
-				slot->bus, slot->device,
-				slot->number, ctrl->slot_device_offset,
-				slot_number);
-		result = pci_hp_register(hotplug_slot,
-					 ctrl->pci_dev->bus,
-					 slot->device,
-					 name);
-		if (result) {
-			err("pci_hp_register failed with error %d\n", result);
-			goto error_info;
-		}
-
-		slot->next = ctrl->slot;
-		ctrl->slot = slot;
-
-		number_of_slots--;
-		slot_device++;
-		slot_number++;
-	}
-
-	return 0;
-error_info:
-	kfree(hotplug_slot_info);
-error_hpslot:
-	kfree(hotplug_slot);
-error_slot:
-	kfree(slot);
-error:
-	return result;
-}
-
 static int ctrl_slot_cleanup (struct controller * ctrl)
 {
 	struct slot *old_slot, *next_slot;
@@ -793,6 +627,230 @@ static int get_cur_bus_speed (struct hotplug_slot *hotplug_slot, enum pci_bus_sp
 	return 0;
 }
 
+static struct hotplug_slot_ops cpqphp_hotplug_slot_ops = {
+	.owner =		THIS_MODULE,
+	.set_attention_status =	set_attention_status,
+	.enable_slot =		process_SI,
+	.disable_slot =		process_SS,
+	.hardware_test =	hardware_test,
+	.get_power_status =	get_power_status,
+	.get_attention_status =	get_attention_status,
+	.get_latch_status =	get_latch_status,
+	.get_adapter_status =	get_adapter_status,
+	.get_max_bus_speed =	get_max_bus_speed,
+	.get_cur_bus_speed =	get_cur_bus_speed,
+};
+
+#define SLOT_NAME_SIZE 10
+
+static int ctrl_slot_setup(struct controller *ctrl,
+			void __iomem *smbios_start,
+			void __iomem *smbios_table)
+{
+	struct slot *slot;
+	struct hotplug_slot *hotplug_slot;
+	struct hotplug_slot_info *hotplug_slot_info;
+	u8 number_of_slots;
+	u8 slot_device;
+	u8 slot_number;
+	u8 ctrl_slot;
+	u32 tempdword;
+	char name[SLOT_NAME_SIZE];
+	void __iomem *slot_entry= NULL;
+	int result = -ENOMEM;
+
+	dbg("%s\n", __func__);
+
+	tempdword = readl(ctrl->hpc_reg + INT_INPUT_CLEAR);
+
+	number_of_slots = readb(ctrl->hpc_reg + SLOT_MASK) & 0x0F;
+	slot_device = readb(ctrl->hpc_reg + SLOT_MASK) >> 4;
+	slot_number = ctrl->first_slot;
+
+	while (number_of_slots) {
+		slot = kzalloc(sizeof(*slot), GFP_KERNEL);
+		if (!slot)
+			goto error;
+
+		slot->hotplug_slot = kzalloc(sizeof(*(slot->hotplug_slot)),
+						GFP_KERNEL);
+		if (!slot->hotplug_slot)
+			goto error_slot;
+		hotplug_slot = slot->hotplug_slot;
+
+		hotplug_slot->info =
+				kzalloc(sizeof(*(hotplug_slot->info)),
+							GFP_KERNEL);
+		if (!hotplug_slot->info)
+			goto error_hpslot;
+		hotplug_slot_info = hotplug_slot->info;
+
+		slot->ctrl = ctrl;
+		slot->bus = ctrl->bus;
+		slot->device = slot_device;
+		slot->number = slot_number;
+		dbg("slot->number = %u\n", slot->number);
+
+		slot_entry = get_SMBIOS_entry(smbios_start, smbios_table, 9,
+					slot_entry);
+
+		while (slot_entry && (readw(slot_entry + SMBIOS_SLOT_NUMBER) !=
+				slot->number)) {
+			slot_entry = get_SMBIOS_entry(smbios_start,
+						smbios_table, 9, slot_entry);
+		}
+
+		slot->p_sm_slot = slot_entry;
+
+		init_timer(&slot->task_event);
+		slot->task_event.expires = jiffies + 5 * HZ;
+		slot->task_event.function = cpqhp_pushbutton_thread;
+
+		/*FIXME: these capabilities aren't used but if they are
+		 *	 they need to be correctly implemented
+		 */
+		slot->capabilities |= PCISLOT_REPLACE_SUPPORTED;
+		slot->capabilities |= PCISLOT_INTERLOCK_SUPPORTED;
+
+		if (is_slot64bit(slot))
+			slot->capabilities |= PCISLOT_64_BIT_SUPPORTED;
+		if (is_slot66mhz(slot))
+			slot->capabilities |= PCISLOT_66_MHZ_SUPPORTED;
+		if (ctrl->speed == PCI_SPEED_66MHz)
+			slot->capabilities |= PCISLOT_66_MHZ_OPERATION;
+
+		ctrl_slot =
+			slot_device - (readb(ctrl->hpc_reg + SLOT_MASK) >> 4);
+
+		/* Check presence */
+		slot->capabilities |=
+			((((~tempdword) >> 23) |
+			 ((~tempdword) >> 15)) >> ctrl_slot) & 0x02;
+		/* Check the switch state */
+		slot->capabilities |=
+			((~tempdword & 0xFF) >> ctrl_slot) & 0x01;
+		/* Check the slot enable */
+		slot->capabilities |=
+			((read_slot_enable(ctrl) << 2) >> ctrl_slot) & 0x04;
+
+		/* register this slot with the hotplug pci core */
+		hotplug_slot->release = &release_slot;
+		hotplug_slot->private = slot;
+		snprintf(name, SLOT_NAME_SIZE, "%u", slot->number);
+		hotplug_slot->ops = &cpqphp_hotplug_slot_ops;
+
+		hotplug_slot_info->power_status = get_slot_enabled(ctrl, slot);
+		hotplug_slot_info->attention_status =
+			cpq_get_attention_status(ctrl, slot);
+		hotplug_slot_info->latch_status =
+			cpq_get_latch_status(ctrl, slot);
+		hotplug_slot_info->adapter_status =
+			get_presence_status(ctrl, slot);
+
+		dbg("registering bus %d, dev %d, number %d, "
+				"ctrl->slot_device_offset %d, slot %d\n",
+				slot->bus, slot->device,
+				slot->number, ctrl->slot_device_offset,
+				slot_number);
+		result = pci_hp_register(hotplug_slot,
+					 ctrl->pci_dev->bus,
+					 slot->device,
+					 name);
+		if (result) {
+			err("pci_hp_register failed with error %d\n", result);
+			goto error_info;
+		}
+
+		slot->next = ctrl->slot;
+		ctrl->slot = slot;
+
+		number_of_slots--;
+		slot_device++;
+		slot_number++;
+	}
+
+	return 0;
+error_info:
+	kfree(hotplug_slot_info);
+error_hpslot:
+	kfree(hotplug_slot);
+error_slot:
+	kfree(slot);
+error:
+	return result;
+}
+
+static int one_time_init(void)
+{
+	int loop;
+	int retval = 0;
+
+	if (initialized)
+		return 0;
+
+	power_mode = 0;
+
+	retval = pci_print_IRQ_route();
+	if (retval)
+		goto error;
+
+	dbg("Initialize + Start the notification mechanism \n");
+
+	retval = cpqhp_event_start_thread();
+	if (retval)
+		goto error;
+
+	dbg("Initialize slot lists\n");
+	for (loop = 0; loop < 256; loop++) {
+		cpqhp_slot_list[loop] = NULL;
+	}
+
+	/* FIXME: We also need to hook the NMI handler eventually.
+	 * this also needs to be worked with Christoph
+	 * register_NMI_handler();
+	 */
+	/* Map rom address */
+	cpqhp_rom_start = ioremap(ROM_PHY_ADDR, ROM_PHY_LEN);
+	if (!cpqhp_rom_start) {
+		err ("Could not ioremap memory region for ROM\n");
+		retval = -EIO;
+		goto error;
+	}
+
+	/* Now, map the int15 entry point if we are on compaq specific
+	 * hardware
+	 */
+	compaq_nvram_init(cpqhp_rom_start);
+
+	/* Map smbios table entry point structure */
+	smbios_table = detect_SMBIOS_pointer(cpqhp_rom_start,
+					cpqhp_rom_start + ROM_PHY_LEN);
+	if (!smbios_table) {
+		err ("Could not find the SMBIOS pointer in memory\n");
+		retval = -EIO;
+		goto error_rom_start;
+	}
+
+	smbios_start = ioremap(readl(smbios_table + ST_ADDRESS),
+					readw(smbios_table + ST_LENGTH));
+	if (!smbios_start) {
+		err ("Could not ioremap memory region taken from SMBIOS values\n");
+		retval = -EIO;
+		goto error_smbios_start;
+	}
+
+	initialized = 1;
+
+	return retval;
+
+error_smbios_start:
+	iounmap(smbios_start);
+error_rom_start:
+	iounmap(cpqhp_rom_start);
+error:
+	return retval;
+}
+
 static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	u8 num_of_slots = 0;
@@ -1300,77 +1358,6 @@ err_disable_device:
 	return rc;
 }
 
-static int one_time_init(void)
-{
-	int loop;
-	int retval = 0;
-
-	if (initialized)
-		return 0;
-
-	power_mode = 0;
-
-	retval = pci_print_IRQ_route();
-	if (retval)
-		goto error;
-
-	dbg("Initialize + Start the notification mechanism \n");
-
-	retval = cpqhp_event_start_thread();
-	if (retval)
-		goto error;
-
-	dbg("Initialize slot lists\n");
-	for (loop = 0; loop < 256; loop++) {
-		cpqhp_slot_list[loop] = NULL;
-	}
-
-	/* FIXME: We also need to hook the NMI handler eventually.
-	 * this also needs to be worked with Christoph
-	 * register_NMI_handler();
-	 */
-	/* Map rom address */
-	cpqhp_rom_start = ioremap(ROM_PHY_ADDR, ROM_PHY_LEN);
-	if (!cpqhp_rom_start) {
-		err ("Could not ioremap memory region for ROM\n");
-		retval = -EIO;
-		goto error;
-	}
-
-	/* Now, map the int15 entry point if we are on compaq specific
-	 * hardware
-	 */
-	compaq_nvram_init(cpqhp_rom_start);
-
-	/* Map smbios table entry point structure */
-	smbios_table = detect_SMBIOS_pointer(cpqhp_rom_start,
-					cpqhp_rom_start + ROM_PHY_LEN);
-	if (!smbios_table) {
-		err ("Could not find the SMBIOS pointer in memory\n");
-		retval = -EIO;
-		goto error_rom_start;
-	}
-
-	smbios_start = ioremap(readl(smbios_table + ST_ADDRESS),
-					readw(smbios_table + ST_LENGTH));
-	if (!smbios_start) {
-		err ("Could not ioremap memory region taken from SMBIOS values\n");
-		retval = -EIO;
-		goto error_smbios_start;
-	}
-
-	initialized = 1;
-
-	return retval;
-
-error_smbios_start:
-	iounmap(smbios_start);
-error_rom_start:
-	iounmap(cpqhp_rom_start);
-error:
-	return retval;
-}
-
 static void __exit unload_cpqphpd(void)
 {
 	struct pci_func *next;
-- 
cgit v1.2.3-59-g8ed1b


From 04225fe7e6877493765b9cfa3092524e21e020d7 Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:23:31 -0600
Subject: PCI Hotplug: cpqphp: eliminate stray braces

Clean up style and eliminate superfluous braces and parens.

No functional change.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp_core.c | 55 ++++++++++++++++-----------------------
 1 file changed, 22 insertions(+), 33 deletions(-)

diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c
index f05ea7a5606b..195c8c9b33e0 100644
--- a/drivers/pci/hotplug/cpqphp_core.c
+++ b/drivers/pci/hotplug/cpqphp_core.c
@@ -154,7 +154,6 @@ static int init_SERR(struct controller * ctrl)
 	return 0;
 }
 
-
 /* nice debugging output */
 static int pci_print_IRQ_route (void)
 {
@@ -214,7 +213,7 @@ static void __iomem *get_subsequent_smbios_entry(void __iomem *smbios_start,
 	void __iomem *p_max;
 
 	if (!smbios_table || !curr)
-		return(NULL);
+		return NULL;
 
 	/* set p_max to the end of the table */
 	p_max = smbios_start + readw(smbios_table + ST_LENGTH);
@@ -227,19 +226,17 @@ static void __iomem *get_subsequent_smbios_entry(void __iomem *smbios_start,
 		 * The first condition is the previous byte
 		 * and the second is the curr
 		 */
-		if (!previous_byte && !(readb(p_temp))) {
+		if (!previous_byte && !(readb(p_temp)))
 			bail = 1;
-		}
 
 		previous_byte = readb(p_temp);
 		p_temp++;
 	}
 
-	if (p_temp < p_max) {
+	if (p_temp < p_max)
 		return p_temp;
-	} else {
+	else
 		return NULL;
-	}
 }
 
 
@@ -265,21 +262,18 @@ static void __iomem *get_SMBIOS_entry(void __iomem *smbios_start,
 	if (!smbios_table)
 		return NULL;
 
-	if (!previous) {
+	if (!previous)
 		previous = smbios_start;
-	} else {
+	else
 		previous = get_subsequent_smbios_entry(smbios_start,
 					smbios_table, previous);
-	}
 
-	while (previous) {
-		if (readb(previous + SMBIOS_GENERIC_TYPE) != type) {
+	while (previous)
+		if (readb(previous + SMBIOS_GENERIC_TYPE) != type)
 			previous = get_subsequent_smbios_entry(smbios_start,
 						smbios_table, previous);
-		} else {
+		else
 			break;
-		}
-	}
 
 	return previous;
 }
@@ -319,7 +313,7 @@ static int ctrl_slot_cleanup (struct controller * ctrl)
 	release_mem_region(pci_resource_start(ctrl->pci_dev, 0),
 			   pci_resource_len(ctrl->pci_dev, 0));
 
-	return(0);
+	return 0;
 }
 
 
@@ -388,9 +382,8 @@ get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot)
 							PCI_DEVFN(tdevice, 0),
 							PCI_PRIMARY_BUS, &work);
 				// See if bridge's secondary bus matches target bus.
-				if (((work >> 8) & 0x000000FF) == (long) bus_num) {
+				if (((work >> 8) & 0x000000FF) == (long) bus_num)
 					bridgeSlot = tslot;
-				}
 			}
 		}
 
@@ -425,21 +418,21 @@ cpqhp_set_attention_status(struct controller *ctrl, struct pci_func *func,
 	u8 hp_slot;
 
 	if (func == NULL)
-		return(1);
+		return 1;
 
 	hp_slot = func->device - ctrl->slot_device_offset;
 
 	/* Wait for exclusive access to hardware */
 	mutex_lock(&ctrl->crit_sect);
 
-	if (status == 1) {
+	if (status == 1)
 		amber_LED_on (ctrl, hp_slot);
-	} else if (status == 0) {
+	else if (status == 0)
 		amber_LED_off (ctrl, hp_slot);
-	} else {
+	else {
 		/* Done with exclusive hardware access */
 		mutex_unlock(&ctrl->crit_sect);
-		return(1);
+		return 1;
 	}
 
 	set_SOGO(ctrl);
@@ -450,7 +443,7 @@ cpqhp_set_attention_status(struct controller *ctrl, struct pci_func *func,
 	/* Done with exclusive hardware access */
 	mutex_unlock(&ctrl->crit_sect);
 
-	return(0);
+	return 0;
 }
 
 
@@ -678,8 +671,7 @@ static int ctrl_slot_setup(struct controller *ctrl,
 			goto error_slot;
 		hotplug_slot = slot->hotplug_slot;
 
-		hotplug_slot->info =
-				kzalloc(sizeof(*(hotplug_slot->info)),
+		hotplug_slot->info = kzalloc(sizeof(*(hotplug_slot->info)),
 							GFP_KERNEL);
 		if (!hotplug_slot->info)
 			goto error_hpslot;
@@ -801,9 +793,8 @@ static int one_time_init(void)
 		goto error;
 
 	dbg("Initialize slot lists\n");
-	for (loop = 0; loop < 256; loop++) {
+	for (loop = 0; loop < 256; loop++)
 		cpqhp_slot_list[loop] = NULL;
-	}
 
 	/* FIXME: We also need to hook the NMI handler eventually.
 	 * this also needs to be worked with Christoph
@@ -1306,18 +1297,16 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		func->presence_save = (temp_word >> hp_slot) & 0x01;
 		func->presence_save |= (temp_word >> (hp_slot + 7)) & 0x02;
 
-		if (ctrl->ctrl_int_comp & (0x1L << hp_slot)) {
+		if (ctrl->ctrl_int_comp & (0x1L << hp_slot))
 			func->switch_save = 0;
-		} else {
+		else
 			func->switch_save = 0x10;
-		}
 
-		if (!power_mode) {
+		if (!power_mode)
 			if (!func->is_a_board) {
 				green_LED_off(ctrl, hp_slot);
 				slot_disable(ctrl, hp_slot);
 			}
-		}
 
 		device++;
 		num_of_slots--;
-- 
cgit v1.2.3-59-g8ed1b


From 867556fe740d0d29a05fce99d2d960625077ed45 Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:23:36 -0600
Subject: PCI Hotplug: cpqphp: refactor cpqhp_probe

Apply DeMorgan's theorem:

	if ((pdev->revision > 2) || (vendor_id == PCI_VENDOR_ID_INTEL))

turns into

	if ((pdev->revision <= 2) && (vendor_id != PCI_VENDOR_ID_INTEL))

Now we can bail out early from the function if the controller is not
supported.

This allows us to un-indent the remainder of the function quite a bit and
make it much more readable.

Fix up some extra braces, and un-indent the 'case' labels in the switch
statement as per CodingStyle.

No functional change.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp_core.c | 382 ++++++++++++++++++--------------------
 1 file changed, 185 insertions(+), 197 deletions(-)

diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c
index 195c8c9b33e0..857e466df71d 100644
--- a/drivers/pci/hotplug/cpqphp_core.c
+++ b/drivers/pci/hotplug/cpqphp_core.c
@@ -887,214 +887,202 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * For Intel, each SSID bit identifies a PHP capability.
 	 * Also Intel HPC's may have RID=0.
 	 */
-	if ((pdev->revision > 2) || (vendor_id == PCI_VENDOR_ID_INTEL)) {
-		/* TODO: This code can be made to support non-Compaq or Intel
-		 * subsystem IDs
-		 */
-		rc = pci_read_config_word(pdev, PCI_SUBSYSTEM_VENDOR_ID, &subsystem_vid);
-		if (rc) {
-			err("%s : pci_read_config_word failed\n", __func__);
-			goto err_disable_device;
-		}
-		dbg("Subsystem Vendor ID: %x\n", subsystem_vid);
-		if ((subsystem_vid != PCI_VENDOR_ID_COMPAQ) && (subsystem_vid != PCI_VENDOR_ID_INTEL)) {
-			err(msg_HPC_non_compaq_or_intel);
-			rc = -ENODEV;
-			goto err_disable_device;
-		}
+	if ((pdev->revision <= 2) && (vendor_id != PCI_VENDOR_ID_INTEL)) {
+		err(msg_HPC_not_supported);
+		return -ENODEV;
+	}
+
+	/* TODO: This code can be made to support non-Compaq or Intel
+	 * subsystem IDs
+	 */
+	rc = pci_read_config_word(pdev, PCI_SUBSYSTEM_VENDOR_ID, &subsystem_vid);
+	if (rc) {
+		err("%s : pci_read_config_word failed\n", __func__);
+		goto err_disable_device;
+	}
+	dbg("Subsystem Vendor ID: %x\n", subsystem_vid);
+	if ((subsystem_vid != PCI_VENDOR_ID_COMPAQ) && (subsystem_vid != PCI_VENDOR_ID_INTEL)) {
+		err(msg_HPC_non_compaq_or_intel);
+		rc = -ENODEV;
+		goto err_disable_device;
+	}
+
+	ctrl = kzalloc(sizeof(struct controller), GFP_KERNEL);
+	if (!ctrl) {
+		err("%s : out of memory\n", __func__);
+		rc = -ENOMEM;
+		goto err_disable_device;
+	}
 
-		ctrl = kzalloc(sizeof(struct controller), GFP_KERNEL);
-		if (!ctrl) {
-			err("%s : out of memory\n", __func__);
-			rc = -ENOMEM;
-			goto err_disable_device;
+	rc = pci_read_config_word(pdev, PCI_SUBSYSTEM_ID, &subsystem_deviceid);
+	if (rc) {
+		err("%s : pci_read_config_word failed\n", __func__);
+		goto err_free_ctrl;
+	}
+
+	info("Hot Plug Subsystem Device ID: %x\n", subsystem_deviceid);
+
+	/* Set Vendor ID, so it can be accessed later from other
+	 * functions
+	 */
+	ctrl->vendor_id = vendor_id;
+
+	switch (subsystem_vid) {
+	case PCI_VENDOR_ID_COMPAQ:
+		if (pdev->revision >= 0x13) { /* CIOBX */
+			ctrl->push_flag = 1;
+			ctrl->slot_switch_type = 1;
+			ctrl->push_button = 1;
+			ctrl->pci_config_space = 1;
+			ctrl->defeature_PHP = 1;
+			ctrl->pcix_support = 1;
+			ctrl->pcix_speed_capability = 1;
+			pci_read_config_byte(pdev, 0x41, &bus_cap);
+			if (bus_cap & 0x80) {
+				dbg("bus max supports 133MHz PCI-X\n");
+				ctrl->speed_capability = PCI_SPEED_133MHz_PCIX;
+				break;
+			}
+			if (bus_cap & 0x40) {
+				dbg("bus max supports 100MHz PCI-X\n");
+				ctrl->speed_capability = PCI_SPEED_100MHz_PCIX;
+				break;
+			}
+			if (bus_cap & 20) {
+				dbg("bus max supports 66MHz PCI-X\n");
+				ctrl->speed_capability = PCI_SPEED_66MHz_PCIX;
+				break;
+			}
+			if (bus_cap & 10) {
+				dbg("bus max supports 66MHz PCI\n");
+				ctrl->speed_capability = PCI_SPEED_66MHz;
+				break;
+			}
+
+			break;
 		}
 
-		rc = pci_read_config_word(pdev, PCI_SUBSYSTEM_ID, &subsystem_deviceid);
-		if (rc) {
-			err("%s : pci_read_config_word failed\n", __func__);
+		switch (subsystem_deviceid) {
+		case PCI_SUB_HPC_ID:
+			/* Original 6500/7000 implementation */
+			ctrl->slot_switch_type = 1;
+			ctrl->speed_capability = PCI_SPEED_33MHz;
+			ctrl->push_button = 0;
+			ctrl->pci_config_space = 1;
+			ctrl->defeature_PHP = 1;
+			ctrl->pcix_support = 0;
+			ctrl->pcix_speed_capability = 0;
+			break;
+		case PCI_SUB_HPC_ID2:
+			/* First Pushbutton implementation */
+			ctrl->push_flag = 1;
+			ctrl->slot_switch_type = 1;
+			ctrl->speed_capability = PCI_SPEED_33MHz;
+			ctrl->push_button = 1;
+			ctrl->pci_config_space = 1;
+			ctrl->defeature_PHP = 1;
+			ctrl->pcix_support = 0;
+			ctrl->pcix_speed_capability = 0;
+			break;
+		case PCI_SUB_HPC_ID_INTC:
+			/* Third party (6500/7000) */
+			ctrl->slot_switch_type = 1;
+			ctrl->speed_capability = PCI_SPEED_33MHz;
+			ctrl->push_button = 0;
+			ctrl->pci_config_space = 1;
+			ctrl->defeature_PHP = 1;
+			ctrl->pcix_support = 0;
+			ctrl->pcix_speed_capability = 0;
+			break;
+		case PCI_SUB_HPC_ID3:
+			/* First 66 Mhz implementation */
+			ctrl->push_flag = 1;
+			ctrl->slot_switch_type = 1;
+			ctrl->speed_capability = PCI_SPEED_66MHz;
+			ctrl->push_button = 1;
+			ctrl->pci_config_space = 1;
+			ctrl->defeature_PHP = 1;
+			ctrl->pcix_support = 0;
+			ctrl->pcix_speed_capability = 0;
+			break;
+		case PCI_SUB_HPC_ID4:
+			/* First PCI-X implementation, 100MHz */
+			ctrl->push_flag = 1;
+			ctrl->slot_switch_type = 1;
+			ctrl->speed_capability = PCI_SPEED_100MHz_PCIX;
+			ctrl->push_button = 1;
+			ctrl->pci_config_space = 1;
+			ctrl->defeature_PHP = 1;
+			ctrl->pcix_support = 1;
+			ctrl->pcix_speed_capability = 0;
+			break;
+		default:
+			err(msg_HPC_not_supported);
+			rc = -ENODEV;
 			goto err_free_ctrl;
 		}
+		break;
 
-		info("Hot Plug Subsystem Device ID: %x\n", subsystem_deviceid);
+	case PCI_VENDOR_ID_INTEL:
+		/* Check for speed capability (0=33, 1=66) */
+		if (subsystem_deviceid & 0x0001)
+			ctrl->speed_capability = PCI_SPEED_66MHz;
+		else
+			ctrl->speed_capability = PCI_SPEED_33MHz;
 
-		/* Set Vendor ID, so it can be accessed later from other
-		 * functions
-		 */
-		ctrl->vendor_id = vendor_id;
-
-		switch (subsystem_vid) {
-			case PCI_VENDOR_ID_COMPAQ:
-				if (pdev->revision >= 0x13) { /* CIOBX */
-					ctrl->push_flag = 1;
-					ctrl->slot_switch_type = 1;
-					ctrl->push_button = 1;
-					ctrl->pci_config_space = 1;
-					ctrl->defeature_PHP = 1;
-					ctrl->pcix_support = 1;
-					ctrl->pcix_speed_capability = 1;
-					pci_read_config_byte(pdev, 0x41, &bus_cap);
-					if (bus_cap & 0x80) {
-						dbg("bus max supports 133MHz PCI-X\n");
-						ctrl->speed_capability = PCI_SPEED_133MHz_PCIX;
-						break;
-					}
-					if (bus_cap & 0x40) {
-						dbg("bus max supports 100MHz PCI-X\n");
-						ctrl->speed_capability = PCI_SPEED_100MHz_PCIX;
-						break;
-					}
-					if (bus_cap & 20) {
-						dbg("bus max supports 66MHz PCI-X\n");
-						ctrl->speed_capability = PCI_SPEED_66MHz_PCIX;
-						break;
-					}
-					if (bus_cap & 10) {
-						dbg("bus max supports 66MHz PCI\n");
-						ctrl->speed_capability = PCI_SPEED_66MHz;
-						break;
-					}
-
-					break;
-				}
-
-				switch (subsystem_deviceid) {
-					case PCI_SUB_HPC_ID:
-						/* Original 6500/7000 implementation */
-						ctrl->slot_switch_type = 1;
-						ctrl->speed_capability = PCI_SPEED_33MHz;
-						ctrl->push_button = 0;
-						ctrl->pci_config_space = 1;
-						ctrl->defeature_PHP = 1;
-						ctrl->pcix_support = 0;
-						ctrl->pcix_speed_capability = 0;
-						break;
-					case PCI_SUB_HPC_ID2:
-						/* First Pushbutton implementation */
-						ctrl->push_flag = 1;
-						ctrl->slot_switch_type = 1;
-						ctrl->speed_capability = PCI_SPEED_33MHz;
-						ctrl->push_button = 1;
-						ctrl->pci_config_space = 1;
-						ctrl->defeature_PHP = 1;
-						ctrl->pcix_support = 0;
-						ctrl->pcix_speed_capability = 0;
-						break;
-					case PCI_SUB_HPC_ID_INTC:
-						/* Third party (6500/7000) */
-						ctrl->slot_switch_type = 1;
-						ctrl->speed_capability = PCI_SPEED_33MHz;
-						ctrl->push_button = 0;
-						ctrl->pci_config_space = 1;
-						ctrl->defeature_PHP = 1;
-						ctrl->pcix_support = 0;
-						ctrl->pcix_speed_capability = 0;
-						break;
-					case PCI_SUB_HPC_ID3:
-						/* First 66 Mhz implementation */
-						ctrl->push_flag = 1;
-						ctrl->slot_switch_type = 1;
-						ctrl->speed_capability = PCI_SPEED_66MHz;
-						ctrl->push_button = 1;
-						ctrl->pci_config_space = 1;
-						ctrl->defeature_PHP = 1;
-						ctrl->pcix_support = 0;
-						ctrl->pcix_speed_capability = 0;
-						break;
-					case PCI_SUB_HPC_ID4:
-						/* First PCI-X implementation, 100MHz */
-						ctrl->push_flag = 1;
-						ctrl->slot_switch_type = 1;
-						ctrl->speed_capability = PCI_SPEED_100MHz_PCIX;
-						ctrl->push_button = 1;
-						ctrl->pci_config_space = 1;
-						ctrl->defeature_PHP = 1;
-						ctrl->pcix_support = 1;
-						ctrl->pcix_speed_capability = 0;	
-						break;
-					default:
-						err(msg_HPC_not_supported);
-						rc = -ENODEV;
-						goto err_free_ctrl;
-				}
-				break;
+		/* Check for push button */
+		if (subsystem_deviceid & 0x0002)
+			ctrl->push_button = 0;
+		else
+			ctrl->push_button = 1;
 
-			case PCI_VENDOR_ID_INTEL:
-				/* Check for speed capability (0=33, 1=66) */
-				if (subsystem_deviceid & 0x0001) {
-					ctrl->speed_capability = PCI_SPEED_66MHz;
-				} else {
-					ctrl->speed_capability = PCI_SPEED_33MHz;
-				}
-
-				/* Check for push button */
-				if (subsystem_deviceid & 0x0002) {
-					/* no push button */
-					ctrl->push_button = 0;
-				} else {
-					/* push button supported */
-					ctrl->push_button = 1;
-				}
-
-				/* Check for slot switch type (0=mechanical, 1=not mechanical) */
-				if (subsystem_deviceid & 0x0004) {
-					/* no switch */
-					ctrl->slot_switch_type = 0;
-				} else {
-					/* switch */
-					ctrl->slot_switch_type = 1;
-				}
-
-				/* PHP Status (0=De-feature PHP, 1=Normal operation) */
-				if (subsystem_deviceid & 0x0008) {
-					ctrl->defeature_PHP = 1;	/* PHP supported */
-				} else {
-					ctrl->defeature_PHP = 0;	/* PHP not supported */
-				}
-
-				/* Alternate Base Address Register Interface (0=not supported, 1=supported) */
-				if (subsystem_deviceid & 0x0010) {
-					ctrl->alternate_base_address = 1;	/* supported */
-				} else {
-					ctrl->alternate_base_address = 0;	/* not supported */
-				}
-
-				/* PCI Config Space Index (0=not supported, 1=supported) */
-				if (subsystem_deviceid & 0x0020) {
-					ctrl->pci_config_space = 1;		/* supported */
-				} else {
-					ctrl->pci_config_space = 0;		/* not supported */
-				}
-
-				/* PCI-X support */
-				if (subsystem_deviceid & 0x0080) {
-					/* PCI-X capable */
-					ctrl->pcix_support = 1;
-					/* Frequency of operation in PCI-X mode */
-					if (subsystem_deviceid & 0x0040) {
-						/* 133MHz PCI-X if bit 7 is 1 */
-						ctrl->pcix_speed_capability = 1;
-					} else {
-						/* 100MHz PCI-X if bit 7 is 1 and bit 0 is 0, */
-						/* 66MHz PCI-X if bit 7 is 1 and bit 0 is 1 */
-						ctrl->pcix_speed_capability = 0;
-					}
-				} else {
-					/* Conventional PCI */
-					ctrl->pcix_support = 0;
-					ctrl->pcix_speed_capability = 0;
-				}
-				break;
+		/* Check for slot switch type (0=mechanical, 1=not mechanical) */
+		if (subsystem_deviceid & 0x0004)
+			ctrl->slot_switch_type = 0;
+		else
+			ctrl->slot_switch_type = 1;
+
+		/* PHP Status (0=De-feature PHP, 1=Normal operation) */
+		if (subsystem_deviceid & 0x0008)
+			ctrl->defeature_PHP = 1;	/* PHP supported */
+		else
+			ctrl->defeature_PHP = 0;	/* PHP not supported */
+
+		/* Alternate Base Address Register Interface
+		 * (0=not supported, 1=supported)
+		 */
+		if (subsystem_deviceid & 0x0010)
+			ctrl->alternate_base_address = 1;
+		else
+			ctrl->alternate_base_address = 0;
 
-			default:
-				err(msg_HPC_not_supported);
-				rc = -ENODEV;
-				goto err_free_ctrl;
+		/* PCI Config Space Index (0=not supported, 1=supported) */
+		if (subsystem_deviceid & 0x0020)
+			ctrl->pci_config_space = 1;
+		else
+			ctrl->pci_config_space = 0;
+
+		/* PCI-X support */
+		if (subsystem_deviceid & 0x0080) {
+			ctrl->pcix_support = 1;
+			if (subsystem_deviceid & 0x0040)
+				/* 133MHz PCI-X if bit 7 is 1 */
+				ctrl->pcix_speed_capability = 1;
+			else
+				/* 100MHz PCI-X if bit 7 is 1 and bit 0 is 0, */
+				/* 66MHz PCI-X if bit 7 is 1 and bit 0 is 1 */
+				ctrl->pcix_speed_capability = 0;
+		} else {
+			/* Conventional PCI */
+			ctrl->pcix_support = 0;
+			ctrl->pcix_speed_capability = 0;
 		}
+		break;
 
-	} else {
+	default:
 		err(msg_HPC_not_supported);
-		return -ENODEV;
+		rc = -ENODEV;
+		goto err_free_ctrl;
 	}
 
 	/* Tell the user that we found one. */
@@ -1164,7 +1152,7 @@ static int cpqhpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_free_mem_region;
 	}
 
-	// Check for 66Mhz operation
+	/* Check for 66Mhz operation */
 	ctrl->speed = get_controller_speed(ctrl);
 
 
-- 
cgit v1.2.3-59-g8ed1b


From 1d3ecf1376bf22be57c6138e7cdf425c6027b115 Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:23:41 -0600
Subject: PCI Hotplug: cpqphp: clean up cpqphp_ctrl.c

Style and whitespace cleanups, no functional change.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp_ctrl.c | 170 ++++++++++++++++++--------------------
 1 file changed, 79 insertions(+), 91 deletions(-)

diff --git a/drivers/pci/hotplug/cpqphp_ctrl.c b/drivers/pci/hotplug/cpqphp_ctrl.c
index b02b8dddcf9f..2fa47af992a8 100644
--- a/drivers/pci/hotplug/cpqphp_ctrl.c
+++ b/drivers/pci/hotplug/cpqphp_ctrl.c
@@ -132,9 +132,8 @@ static struct slot *cpqhp_find_slot(struct controller *ctrl, u8 device)
 {
 	struct slot *slot = ctrl->slot;
 
-	while (slot && (slot->device != device)) {
+	while (slot && (slot->device != device))
 		slot = slot->next;
-	}
 
 	return slot;
 }
@@ -549,10 +548,10 @@ static struct pci_resource *get_io_resource(struct pci_resource **head, u32 size
 	if (!(*head))
 		return NULL;
 
-	if ( cpqhp_resource_sort_and_combine(head) )
+	if (cpqhp_resource_sort_and_combine(head))
 		return NULL;
 
-	if ( sort_by_size(head) )
+	if (sort_by_size(head))
 		return NULL;
 
 	for (node = *head; node; node = node->next) {
@@ -974,12 +973,8 @@ struct pci_func *cpqhp_slot_create(u8 busnumber)
 	struct pci_func *next;
 
 	new_slot = kzalloc(sizeof(*new_slot), GFP_KERNEL);
-	if (new_slot == NULL) {
-		/* I'm not dead yet!
-		 * You will be.
-		 */
+	if (new_slot == NULL)
 		return new_slot;
-	}
 
 	new_slot->next = NULL;
 	new_slot->configured = 1;
@@ -1010,10 +1005,8 @@ static int slot_remove(struct pci_func * old_slot)
 		return 1;
 
 	next = cpqhp_slot_list[old_slot->bus];
-
-	if (next == NULL) {
+	if (next == NULL)
 		return 1;
-	}
 
 	if (next == old_slot) {
 		cpqhp_slot_list[old_slot->bus] = old_slot->next;
@@ -1022,9 +1015,8 @@ static int slot_remove(struct pci_func * old_slot)
 		return 0;
 	}
 
-	while ((next->next != old_slot) && (next->next != NULL)) {
+	while ((next->next != old_slot) && (next->next != NULL))
 		next = next->next;
-	}
 
 	if (next->next == old_slot) {
 		next->next = old_slot->next;
@@ -1054,9 +1046,8 @@ static int bridge_slot_remove(struct pci_func *bridge)
 	for (tempBus = secondaryBus; tempBus <= subordinateBus; tempBus++) {
 		next = cpqhp_slot_list[tempBus];
 
-		while (!slot_remove(next)) {
+		while (!slot_remove(next))
 			next = cpqhp_slot_list[tempBus];
-		}
 	}
 
 	next = cpqhp_slot_list[bridge->bus];
@@ -1286,17 +1277,17 @@ static u32 board_replaced(struct pci_func *func, struct controller *ctrl)
 
 	hp_slot = func->device - ctrl->slot_device_offset;
 
-	if (readl(ctrl->hpc_reg + INT_INPUT_CLEAR) & (0x01L << hp_slot)) {
-		/*
-		 * The switch is open.
-		 */
+	/*
+	 * The switch is open.
+	 */
+	if (readl(ctrl->hpc_reg + INT_INPUT_CLEAR) & (0x01L << hp_slot))
 		rc = INTERLOCK_OPEN;
-	} else if (is_slot_enabled (ctrl, hp_slot)) {
-		/*
-		 * The board is already on
-		 */
+	/*
+	 * The board is already on
+	 */
+	else if (is_slot_enabled (ctrl, hp_slot))
 		rc = CARD_FUNCTIONING;
-	} else {
+	else {
 		mutex_lock(&ctrl->crit_sect);
 
 		/* turn on board without attaching to the bus */
@@ -1542,7 +1533,7 @@ static u32 board_added(struct pci_func *func, struct controller *ctrl)
 	}
 
 	/* All F's is an empty slot or an invalid board */
-	if (temp_register != 0xFFFFFFFF) {	  /* Check for a board in the slot */
+	if (temp_register != 0xFFFFFFFF) {
 		res_lists.io_head = ctrl->io_head;
 		res_lists.mem_head = ctrl->mem_head;
 		res_lists.p_mem_head = ctrl->p_mem_head;
@@ -1591,9 +1582,8 @@ static u32 board_added(struct pci_func *func, struct controller *ctrl)
 		index = 0;
 		do {
 			new_slot = cpqhp_slot_find(ctrl->bus, func->device, index++);
-			if (new_slot && !new_slot->pci_dev) {
+			if (new_slot && !new_slot->pci_dev)
 				cpqhp_configure_device(ctrl, new_slot);
-			}
 		} while (new_slot);
 
 		mutex_lock(&ctrl->crit_sect);
@@ -2134,9 +2124,8 @@ int cpqhp_process_SS(struct controller *ctrl, struct pci_func *func)
 
 				/* If the VGA Enable bit is set, remove isn't
 				 * supported */
-				if (BCR & PCI_BRIDGE_CTL_VGA) {
+				if (BCR & PCI_BRIDGE_CTL_VGA)
 					rc = REMOVE_NOT_SUPPORTED;
-				}
 			}
 		}
 
@@ -2204,67 +2193,67 @@ int cpqhp_hardware_test(struct controller *ctrl, int test_num)
 	num_of_slots = readb(ctrl->hpc_reg + SLOT_MASK) & 0x0f;
 
 	switch (test_num) {
-		case 1:
-			/* Do stuff here! */
-
-			/* Do that funky LED thing */
-			/* so we can restore them later */
-			save_LED = readl(ctrl->hpc_reg + LED_CONTROL);
-			work_LED = 0x01010101;
-			switch_leds(ctrl, num_of_slots, &work_LED, 0);
-			switch_leds(ctrl, num_of_slots, &work_LED, 1);
-			switch_leds(ctrl, num_of_slots, &work_LED, 0);
-			switch_leds(ctrl, num_of_slots, &work_LED, 1);
-
-			work_LED = 0x01010000;
-			writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
-			switch_leds(ctrl, num_of_slots, &work_LED, 0);
-			switch_leds(ctrl, num_of_slots, &work_LED, 1);
-			work_LED = 0x00000101;
-			writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
-			switch_leds(ctrl, num_of_slots, &work_LED, 0);
-			switch_leds(ctrl, num_of_slots, &work_LED, 1);
-
-			work_LED = 0x01010000;
-			writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
-			for (loop = 0; loop < num_of_slots; loop++) {
-				set_SOGO(ctrl);
+	case 1:
+		/* Do stuff here! */
+
+		/* Do that funky LED thing */
+		/* so we can restore them later */
+		save_LED = readl(ctrl->hpc_reg + LED_CONTROL);
+		work_LED = 0x01010101;
+		switch_leds(ctrl, num_of_slots, &work_LED, 0);
+		switch_leds(ctrl, num_of_slots, &work_LED, 1);
+		switch_leds(ctrl, num_of_slots, &work_LED, 0);
+		switch_leds(ctrl, num_of_slots, &work_LED, 1);
+
+		work_LED = 0x01010000;
+		writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
+		switch_leds(ctrl, num_of_slots, &work_LED, 0);
+		switch_leds(ctrl, num_of_slots, &work_LED, 1);
+		work_LED = 0x00000101;
+		writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
+		switch_leds(ctrl, num_of_slots, &work_LED, 0);
+		switch_leds(ctrl, num_of_slots, &work_LED, 1);
+
+		work_LED = 0x01010000;
+		writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
+		for (loop = 0; loop < num_of_slots; loop++) {
+			set_SOGO(ctrl);
 
-				/* Wait for SOGO interrupt */
-				wait_for_ctrl_irq (ctrl);
+			/* Wait for SOGO interrupt */
+			wait_for_ctrl_irq (ctrl);
 
-				/* Get ready for next iteration */
-				long_delay((3*HZ)/10);
-				work_LED = work_LED >> 16;
-				writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
+			/* Get ready for next iteration */
+			long_delay((3*HZ)/10);
+			work_LED = work_LED >> 16;
+			writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
 
-				set_SOGO(ctrl);
+			set_SOGO(ctrl);
 
-				/* Wait for SOGO interrupt */
-				wait_for_ctrl_irq (ctrl);
+			/* Wait for SOGO interrupt */
+			wait_for_ctrl_irq (ctrl);
 
-				/* Get ready for next iteration */
-				long_delay((3*HZ)/10);
-				work_LED = work_LED << 16;
-				writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
-				work_LED = work_LED << 1;
-				writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
-			}
+			/* Get ready for next iteration */
+			long_delay((3*HZ)/10);
+			work_LED = work_LED << 16;
+			writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
+			work_LED = work_LED << 1;
+			writel(work_LED, ctrl->hpc_reg + LED_CONTROL);
+		}
 
-			/* put it back the way it was */
-			writel(save_LED, ctrl->hpc_reg + LED_CONTROL);
+		/* put it back the way it was */
+		writel(save_LED, ctrl->hpc_reg + LED_CONTROL);
 
-			set_SOGO(ctrl);
+		set_SOGO(ctrl);
 
-			/* Wait for SOBS to be unset */
-			wait_for_ctrl_irq (ctrl);
-			break;
-		case 2:
-			/* Do other stuff here! */
-			break;
-		case 3:
-			/* and more... */
-			break;
+		/* Wait for SOBS to be unset */
+		wait_for_ctrl_irq (ctrl);
+		break;
+	case 2:
+		/* Do other stuff here! */
+		break;
+	case 3:
+		/* and more... */
+		break;
 	}
 	return 0;
 }
@@ -2333,9 +2322,9 @@ static u32 configure_new_device(struct controller * ctrl, struct pci_func * func
 		while ((function < max_functions) && (!stop_it)) {
 			pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(func->device, function), 0x00, &ID);
 
-			if (ID == 0xFFFFFFFF) {	  /* There's nothing there. */
+			if (ID == 0xFFFFFFFF) {
 				function++;
-			} else {  /* There's something there */
+			} else {
 				/* Setup slot structure. */
 				new_slot = cpqhp_slot_create(func->bus);
 
@@ -2360,8 +2349,8 @@ static u32 configure_new_device(struct controller * ctrl, struct pci_func * func
 
 
 /*
-  Configuration logic that involves the hotplug data structures and
-  their bookkeeping
+ * Configuration logic that involves the hotplug data structures and
+ * their bookkeeping
  */
 
 
@@ -2414,7 +2403,7 @@ static int configure_new_function(struct controller *ctrl, struct pci_func *func
 	if (rc)
 		return rc;
 
-	if ((temp_byte & 0x7F) == PCI_HEADER_TYPE_BRIDGE) { /* PCI-PCI Bridge */
+	if ((temp_byte & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
 		/* set Primary bus */
 		dbg("set Primary bus = %d\n", func->bus);
 		rc = pci_bus_write_config_byte(pci_bus, devfn, PCI_PRIMARY_BUS, func->bus);
@@ -2956,11 +2945,10 @@ static int configure_new_function(struct controller *ctrl, struct pci_func *func
 				/* Program IRQ based on card type */
 				rc = pci_bus_read_config_byte (pci_bus, devfn, 0x0B, &class_code);
 
-				if (class_code == PCI_BASE_CLASS_STORAGE) {
+				if (class_code == PCI_BASE_CLASS_STORAGE)
 					IRQ = cpqhp_disk_irq;
-				} else {
+				else
 					IRQ = cpqhp_nic_irq;
-				}
 			}
 
 			/* IRQ Line */
-- 
cgit v1.2.3-59-g8ed1b


From de86ae16d55a23315cdf1dae68df9de55312cf25 Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:23:46 -0600
Subject: PCI Hotplug: cpqphp: refactor cpqphp_save_slot_config

Check for an empty slot, and return early if so.

This allows us to un-indent the rest of the function by one level.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp_pci.c | 85 ++++++++++++++++++----------------------
 1 file changed, 39 insertions(+), 46 deletions(-)

diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c
index 2e96bae3c82a..1f1c90dd791d 100644
--- a/drivers/pci/hotplug/cpqphp_pci.c
+++ b/drivers/pci/hotplug/cpqphp_pci.c
@@ -494,7 +494,7 @@ int cpqhp_save_slot_config (struct controller *ctrl, struct pci_func * new_slot)
 	u8 secondary_bus;
 	int sub_bus;
 	int max_functions;
-	int function;
+	int function = 0;
 	int cloop = 0;
 	int stop_it;
 
@@ -503,65 +503,58 @@ int cpqhp_save_slot_config (struct controller *ctrl, struct pci_func * new_slot)
 	ctrl->pci_bus->number = new_slot->bus;
 	pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), PCI_VENDOR_ID, &ID);
 
-	if (ID != 0xFFFFFFFF) {	  /*  device in slot */
-		pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), 0x0B, &class_code);
-		pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), PCI_HEADER_TYPE, &header_type);
-
-		if (header_type & 0x80)	/* Multi-function device */
-			max_functions = 8;
-		else
-			max_functions = 1;
-
-		function = 0;
+	if (ID == 0xFFFFFFFF)
+		return 2;
 
-		do {
-			if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
-				/*  Recurse the subordinate bus */
-				pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_SECONDARY_BUS, &secondary_bus);
+	pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), 0x0B, &class_code);
+	pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(new_slot->device, 0), PCI_HEADER_TYPE, &header_type);
 
-				sub_bus = (int) secondary_bus;
+	if (header_type & 0x80)	/* Multi-function device */
+		max_functions = 8;
+	else
+		max_functions = 1;
 
-				/* Save the config headers for the secondary
-				 * bus.
-				 */
-				rc = cpqhp_save_config(ctrl, sub_bus, 0);
-				if (rc)
-					return(rc);
-				ctrl->pci_bus->number = new_slot->bus;
+	while (function < max_functions) {
+		if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
+			/*  Recurse the subordinate bus */
+			pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_SECONDARY_BUS, &secondary_bus);
 
-			}	/* End of IF */
+			sub_bus = (int) secondary_bus;
 
-			new_slot->status = 0;
+			/* Save the config headers for the secondary
+			 * bus.
+			 */
+			rc = cpqhp_save_config(ctrl, sub_bus, 0);
+			if (rc)
+				return(rc);
+			ctrl->pci_bus->number = new_slot->bus;
 
-			for (cloop = 0; cloop < 0x20; cloop++) {
-				pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), cloop << 2, (u32 *) & (new_slot-> config_space [cloop]));
-			}
+		}
 
-			function++;
+		new_slot->status = 0;
 
-			stop_it = 0;
+		for (cloop = 0; cloop < 0x20; cloop++)
+			pci_bus_read_config_dword(ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), cloop << 2, (u32 *) & (new_slot-> config_space [cloop]));
 
-			/* this loop skips to the next present function
-			 * reading in the Class Code and the Header type.
-			 */
-			while ((function < max_functions) && (!stop_it)) {
-				pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_VENDOR_ID, &ID);
+		function++;
 
-				if (ID == 0xFFFFFFFF) {	 /* nothing there. */
-					function++;
-				} else {  /* Something there */
-					pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), 0x0B, &class_code);
+		stop_it = 0;
 
-					pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_HEADER_TYPE, &header_type);
+		/* this loop skips to the next present function
+		 * reading in the Class Code and the Header type.
+		 */
+		while ((function < max_functions) && (!stop_it)) {
+			pci_bus_read_config_dword(ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_VENDOR_ID, &ID);
 
-					stop_it++;
-				}
+			if (ID == 0xFFFFFFFF)
+				function++;
+			else {
+				pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), 0x0B, &class_code);
+				pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(new_slot->device, function), PCI_HEADER_TYPE, &header_type);
+				stop_it++;
 			}
+		}
 
-		} while (function < max_functions);
-	}			/* End of IF (device in slot?) */
-	else {
-		return 2;
 	}
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 4aabb58e1f544e97dbb97d0ce29bdfc9108f2f2c Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:23:52 -0600
Subject: PCI Hotplug: cpqphp: style cleanups

Clean up style, whitespace in cpqphp_pci.c

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp_pci.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c
index 1f1c90dd791d..bd384e94f212 100644
--- a/drivers/pci/hotplug/cpqphp_pci.c
+++ b/drivers/pci/hotplug/cpqphp_pci.c
@@ -950,15 +950,13 @@ int cpqhp_save_used_resources (struct controller *ctrl, struct pci_func * func)
 						return(1);
 				}
 			}	/* End of base register loop */
-		/* Some other unknown header type */
-		} else {
 		}
 
 		/* find the next device in this slot */
 		func = cpqhp_slot_find(func->bus, func->device, index++);
 	}
 
-	return(0);
+	return 0;
 }
 
 
@@ -993,9 +991,8 @@ int cpqhp_configure_board(struct controller *ctrl, struct pci_func * func)
 		/* Start at the top of config space so that the control
 		 * registers are programmed last
 		 */
-		for (cloop = 0x3C; cloop > 0; cloop -= 4) {
+		for (cloop = 0x3C; cloop > 0; cloop -= 4)
 			pci_bus_write_config_dword (pci_bus, devfn, cloop, func->config_space[cloop >> 2]);
-		}
 
 		pci_bus_read_config_byte (pci_bus, devfn, PCI_HEADER_TYPE, &header_type);
 
@@ -1210,9 +1207,9 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st
 	rom_resource_table = detect_HRT_floating_pointer(rom_start, rom_start+0xffff);
 	dbg("rom_resource_table = %p\n", rom_resource_table);
 
-	if (rom_resource_table == NULL) {
+	if (rom_resource_table == NULL)
 		return -ENODEV;
-	}
+
 	/* Sum all resources and setup resource maps */
 	unused_IRQ = readl(rom_resource_table + UNUSED_IRQ);
 	dbg("unused_IRQ = %x\n", unused_IRQ);
@@ -1245,13 +1242,11 @@ int cpqhp_find_available_resources(struct controller *ctrl, void __iomem *rom_st
 
 	temp = 0;
 
-	if (!cpqhp_nic_irq) {
+	if (!cpqhp_nic_irq)
 		cpqhp_nic_irq = ctrl->cfgspc_irq;
-	}
 
-	if (!cpqhp_disk_irq) {
+	if (!cpqhp_disk_irq)
 		cpqhp_disk_irq = ctrl->cfgspc_irq;
-	}
 
 	dbg("cpqhp_disk_irq, cpqhp_nic_irq= %d, %d\n", cpqhp_disk_irq, cpqhp_nic_irq);
 
-- 
cgit v1.2.3-59-g8ed1b


From 1d2e8b1c58ef96b8834b139caf4357effedcb5ab Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:23:57 -0600
Subject: PCI Hotplug: cpqphp: refactor cpqhp_save_config

Handle an empty slot at the top of the loop, and continue early.

This allows us to un-indent the rest of the function by one level.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp_pci.c | 182 +++++++++++++++++++--------------------
 1 file changed, 91 insertions(+), 91 deletions(-)

diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c
index bd384e94f212..2909e3f6caa7 100644
--- a/drivers/pci/hotplug/cpqphp_pci.c
+++ b/drivers/pci/hotplug/cpqphp_pci.c
@@ -359,121 +359,121 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 	ctrl->pci_bus->number = busnumber;
 	for (device = FirstSupported; device <= LastSupported; device++) {
 		ID = 0xFFFFFFFF;
-		rc = pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(device, 0), PCI_VENDOR_ID, &ID);
+		rc = pci_bus_read_config_dword(ctrl->pci_bus, PCI_DEVFN(device, 0), PCI_VENDOR_ID, &ID);
 
-		if (ID != 0xFFFFFFFF) {	  /* device in slot */
-			rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, 0), 0x0B, &class_code);
-			if (rc)
-				return rc;
-
-			rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, 0), PCI_HEADER_TYPE, &header_type);
-			if (rc)
-				return rc;
-
-			/* If multi-function device, set max_functions to 8 */
-			if (header_type & 0x80)
-				max_functions = 8;
-			else
-				max_functions = 1;
+		if (ID == 0xFFFFFFFF) {
+			if (is_hot_plug) {
+				/* Setup slot structure with entry for empty
+				 * slot
+				 */
+				new_slot = cpqhp_slot_create(busnumber);
+				if (new_slot == NULL)
+					return 1;
 
-			function = 0;
+				new_slot->bus = (u8) busnumber;
+				new_slot->device = (u8) device;
+				new_slot->function = 0;
+				new_slot->is_a_board = 0;
+				new_slot->presence_save = 0;
+				new_slot->switch_save = 0;
+			}
+			continue;
+		}
 
-			do {
-				DevError = 0;
-				if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
-					/* Recurse the subordinate bus
-					 * get the subordinate bus number
-					 */
-					rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, function), PCI_SECONDARY_BUS, &secondary_bus);
-					if (rc) {
-						return rc;
-					} else {
-						sub_bus = (int) secondary_bus;
+		rc = pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(device, 0), 0x0B, &class_code);
+		if (rc)
+			return rc;
 
-						/* Save secondary bus cfg spc
-						 * with this recursive call.
-						 */
-						rc = cpqhp_save_config(ctrl, sub_bus, 0);
-						if (rc)
-							return rc;
-						ctrl->pci_bus->number = busnumber;
-					}
-				}
+		rc = pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(device, 0), PCI_HEADER_TYPE, &header_type);
+		if (rc)
+			return rc;
 
-				index = 0;
-				new_slot = cpqhp_slot_find(busnumber, device, index++);
-				while (new_slot &&
-				       (new_slot->function != (u8) function))
-					new_slot = cpqhp_slot_find(busnumber, device, index++);
+		/* If multi-function device, set max_functions to 8 */
+		if (header_type & 0x80)
+			max_functions = 8;
+		else
+			max_functions = 1;
 
-				if (!new_slot) {
-					/* Setup slot structure. */
-					new_slot = cpqhp_slot_create(busnumber);
+		function = 0;
 
-					if (new_slot == NULL)
-						return(1);
-				}
+		do {
+			DevError = 0;
+			if ((header_type & 0x7F) == PCI_HEADER_TYPE_BRIDGE) {
+				/* Recurse the subordinate bus
+				 * get the subordinate bus number
+				 */
+				rc = pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(device, function), PCI_SECONDARY_BUS, &secondary_bus);
+				if (rc) {
+					return rc;
+				} else {
+					sub_bus = (int) secondary_bus;
 
-				new_slot->bus = (u8) busnumber;
-				new_slot->device = (u8) device;
-				new_slot->function = (u8) function;
-				new_slot->is_a_board = 1;
-				new_slot->switch_save = 0x10;
-				/* In case of unsupported board */
-				new_slot->status = DevError;
-				new_slot->pci_dev = pci_find_slot(new_slot->bus, (new_slot->device << 3) | new_slot->function);
-
-				for (cloop = 0; cloop < 0x20; cloop++) {
-					rc = pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(device, function), cloop << 2, (u32 *) & (new_slot-> config_space [cloop]));
+					/* Save secondary bus cfg spc
+					 * with this recursive call.
+					 */
+					rc = cpqhp_save_config(ctrl, sub_bus, 0);
 					if (rc)
 						return rc;
+					ctrl->pci_bus->number = busnumber;
 				}
+			}
 
-				function++;
+			index = 0;
+			new_slot = cpqhp_slot_find(busnumber, device, index++);
+			while (new_slot &&
+			       (new_slot->function != (u8) function))
+				new_slot = cpqhp_slot_find(busnumber, device, index++);
 
-				stop_it = 0;
+			if (!new_slot) {
+				/* Setup slot structure. */
+				new_slot = cpqhp_slot_create(busnumber);
+				if (new_slot == NULL)
+					return 1;
+			}
 
-				/* this loop skips to the next present function
-				 * reading in Class Code and Header type.
-				 */
+			new_slot->bus = (u8) busnumber;
+			new_slot->device = (u8) device;
+			new_slot->function = (u8) function;
+			new_slot->is_a_board = 1;
+			new_slot->switch_save = 0x10;
+			/* In case of unsupported board */
+			new_slot->status = DevError;
+			new_slot->pci_dev = pci_find_slot(new_slot->bus, (new_slot->device << 3) | new_slot->function);
+
+			for (cloop = 0; cloop < 0x20; cloop++) {
+				rc = pci_bus_read_config_dword(ctrl->pci_bus, PCI_DEVFN(device, function), cloop << 2, (u32 *) & (new_slot-> config_space [cloop]));
+				if (rc)
+					return rc;
+			}
 
-				while ((function < max_functions)&&(!stop_it)) {
-					rc = pci_bus_read_config_dword (ctrl->pci_bus, PCI_DEVFN(device, function), PCI_VENDOR_ID, &ID);
-					if (ID == 0xFFFFFFFF) {	 /* nothing there. */
-						function++;
-					} else {  /* Something there */
-						rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, function), 0x0B, &class_code);
-						if (rc)
-							return rc;
+			function++;
 
-						rc = pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(device, function), PCI_HEADER_TYPE, &header_type);
-						if (rc)
-							return rc;
+			stop_it = 0;
 
-						stop_it++;
-					}
+			/* this loop skips to the next present function
+			 * reading in Class Code and Header type.
+			 */
+			while ((function < max_functions) && (!stop_it)) {
+				rc = pci_bus_read_config_dword(ctrl->pci_bus, PCI_DEVFN(device, function), PCI_VENDOR_ID, &ID);
+				if (ID == 0xFFFFFFFF) {
+					function++;
+					continue;
 				}
+				rc = pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(device, function), 0x0B, &class_code);
+				if (rc)
+					return rc;
 
-			} while (function < max_functions);
-		}		/* End of IF (device in slot?) */
-		else if (is_hot_plug) {
-			/* Setup slot structure with entry for empty slot */
-			new_slot = cpqhp_slot_create(busnumber);
+				rc = pci_bus_read_config_byte(ctrl->pci_bus, PCI_DEVFN(device, function), PCI_HEADER_TYPE, &header_type);
+				if (rc)
+					return rc;
 
-			if (new_slot == NULL) {
-				return(1);
+				stop_it++;
 			}
 
-			new_slot->bus = (u8) busnumber;
-			new_slot->device = (u8) device;
-			new_slot->function = 0;
-			new_slot->is_a_board = 0;
-			new_slot->presence_save = 0;
-			new_slot->switch_save = 0;
-		}
+		} while (function < max_functions);
 	}			/* End of FOR loop */
 
-	return(0);
+	return 0;
 }
 
 
-- 
cgit v1.2.3-59-g8ed1b


From b019ee679afde950f2d01b1af0530453aa60af3f Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:24:02 -0600
Subject: PCI Hotplug: cpqphp: clean up accesses to
 pcibios_get_irq_routing_table()

Instead of making multiple calls to pcibios_get_irq_routing_table, let's
just do it once and save the answer.

The reason we were making multiple calls is because we liked to calculate
its length and perform some loop over it. Instead of open-coding the length
calculation every time, provide it in an inline helper function.

Finally, since pci_print_IRQ_route() is used only for debug, let's only
do it when cpqhp_debug is set.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp.h      |  9 +++++
 drivers/pci/hotplug/cpqphp_core.c | 72 +++++++++++++++++----------------------
 drivers/pci/hotplug/cpqphp_pci.c  | 36 ++++----------------
 3 files changed, 47 insertions(+), 70 deletions(-)

diff --git a/drivers/pci/hotplug/cpqphp.h b/drivers/pci/hotplug/cpqphp.h
index 308f82b1fc9a..1d5561c6bad4 100644
--- a/drivers/pci/hotplug/cpqphp.h
+++ b/drivers/pci/hotplug/cpqphp.h
@@ -455,6 +455,7 @@ extern int cpqhp_debug;
 extern int cpqhp_legacy_mode;
 extern struct controller *cpqhp_ctrl_list;
 extern struct pci_func *cpqhp_slot_list[256];
+extern struct irq_routing_table *cpqhp_routing_table;
 
 /* these can be gotten rid of, but for debugging they are purty */
 extern u8 cpqhp_nic_irq;
@@ -733,4 +734,12 @@ static inline int wait_for_ctrl_irq(struct controller *ctrl)
 	return retval;
 }
 
+#include <asm/pci_x86.h>
+static inline int cpqhp_routing_table_length(void)
+{
+	BUG_ON(cpqhp_routing_table == NULL);
+	return ((cpqhp_routing_table->size - sizeof(struct irq_routing_table)) /
+		sizeof(struct irq_info));
+}
+
 #endif
diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c
index 857e466df71d..7888b37c6c8e 100644
--- a/drivers/pci/hotplug/cpqphp_core.c
+++ b/drivers/pci/hotplug/cpqphp_core.c
@@ -44,7 +44,6 @@
 
 #include "cpqphp.h"
 #include "cpqphp_nvram.h"
-#include <asm/pci_x86.h>
 
 
 /* Global variables */
@@ -52,6 +51,7 @@ int cpqhp_debug;
 int cpqhp_legacy_mode;
 struct controller *cpqhp_ctrl_list;	/* = NULL */
 struct pci_func *cpqhp_slot_list[256];
+struct irq_routing_table *cpqhp_routing_table;
 
 /* local variables */
 static void __iomem *smbios_table;
@@ -154,40 +154,42 @@ static int init_SERR(struct controller * ctrl)
 	return 0;
 }
 
-/* nice debugging output */
-static int pci_print_IRQ_route (void)
+static int init_cpqhp_routing_table(void)
 {
-	struct irq_routing_table *routing_table;
 	int len;
-	int loop;
-
-	u8 tbus, tdevice, tslot;
 
-	routing_table = pcibios_get_irq_routing_table();
-	if (routing_table == NULL) {
-		err("No BIOS Routing Table??? Not good\n");
+	cpqhp_routing_table = pcibios_get_irq_routing_table();
+	if (cpqhp_routing_table == NULL)
 		return -ENOMEM;
-	}
 
-	len = (routing_table->size - sizeof(struct irq_routing_table)) /
-			sizeof(struct irq_info);
-	/* Make sure I got at least one entry */
+	len = cpqhp_routing_table_length();
 	if (len == 0) {
-		kfree(routing_table);
+		kfree(cpqhp_routing_table);
+		cpqhp_routing_table = NULL;
 		return -1;
 	}
 
-	dbg("bus dev func slot\n");
+	return 0;
+}
+
+/* nice debugging output */
+static void pci_print_IRQ_route(void)
+{
+	int len;
+	int loop;
+	u8 tbus, tdevice, tslot;
+
+	len = cpqhp_routing_table_length();
 
+	dbg("bus dev func slot\n");
 	for (loop = 0; loop < len; ++loop) {
-		tbus = routing_table->slots[loop].bus;
-		tdevice = routing_table->slots[loop].devfn;
-		tslot = routing_table->slots[loop].slot;
+		tbus = cpqhp_routing_table->slots[loop].bus;
+		tdevice = cpqhp_routing_table->slots[loop].devfn;
+		tslot = cpqhp_routing_table->slots[loop].slot;
 		dbg("%d %d %d %d\n", tbus, tdevice >> 3, tdevice & 0x7, tslot);
 
 	}
-	kfree(routing_table);
-	return 0;
+	return;
 }
 
 
@@ -331,7 +333,6 @@ static int ctrl_slot_cleanup (struct controller * ctrl)
 static int
 get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot)
 {
-	struct irq_routing_table *PCIIRQRoutingInfoLength;
 	u32 work;
 	long len;
 	long loop;
@@ -342,26 +343,14 @@ get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot)
 
 	bridgeSlot = 0xFF;
 
-	PCIIRQRoutingInfoLength = pcibios_get_irq_routing_table();
-	if (!PCIIRQRoutingInfoLength)
-		return -1;
-
-	len = (PCIIRQRoutingInfoLength->size -
-	       sizeof(struct irq_routing_table)) / sizeof(struct irq_info);
-	/* Make sure I got at least one entry */
-	if (len == 0) {
-		kfree(PCIIRQRoutingInfoLength);
-		return -1;
-	}
-
+	len = cpqhp_routing_table_length();
 	for (loop = 0; loop < len; ++loop) {
-		tbus = PCIIRQRoutingInfoLength->slots[loop].bus;
-		tdevice = PCIIRQRoutingInfoLength->slots[loop].devfn >> 3;
-		tslot = PCIIRQRoutingInfoLength->slots[loop].slot;
+		tbus = cpqhp_routing_table->slots[loop].bus;
+		tdevice = cpqhp_routing_table->slots[loop].devfn >> 3;
+		tslot = cpqhp_routing_table->slots[loop].slot;
 
 		if ((tbus == bus_num) && (tdevice == dev_num)) {
 			*slot = tslot;
-			kfree(PCIIRQRoutingInfoLength);
 			return 0;
 		} else {
 			/* Did not get a match on the target PCI device. Check
@@ -396,10 +385,8 @@ get_slot_mapping(struct pci_bus *bus, u8 bus_num, u8 dev_num, u8 *slot)
 	 */
 	if (bridgeSlot != 0xFF) {
 		*slot = bridgeSlot;
-		kfree(PCIIRQRoutingInfoLength);
 		return 0;
 	}
-	kfree(PCIIRQRoutingInfoLength);
 	/* Couldn't find an entry in the routing table for this PCI device */
 	return -1;
 }
@@ -782,10 +769,13 @@ static int one_time_init(void)
 
 	power_mode = 0;
 
-	retval = pci_print_IRQ_route();
+	retval = init_cpqhp_routing_table();
 	if (retval)
 		goto error;
 
+	if (cpqhp_debug)
+		pci_print_IRQ_route();
+
 	dbg("Initialize + Start the notification mechanism \n");
 
 	retval = cpqhp_event_start_thread();
diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c
index 2909e3f6caa7..6201281b9dab 100644
--- a/drivers/pci/hotplug/cpqphp_pci.c
+++ b/drivers/pci/hotplug/cpqphp_pci.c
@@ -37,7 +37,6 @@
 #include "../pci.h"
 #include "cpqphp.h"
 #include "cpqphp_nvram.h"
-#include <asm/pci_x86.h>
 
 
 u8 cpqhp_nic_irq;
@@ -244,39 +243,23 @@ static int PCI_ScanBusForNonBridge(struct controller *ctrl, u8 bus_num, u8 * dev
 
 static int PCI_GetBusDevHelper(struct controller *ctrl, u8 *bus_num, u8 *dev_num, u8 slot, u8 nobridge)
 {
-	struct irq_routing_table *PCIIRQRoutingInfoLength;
-	long len;
-	long loop;
+	int loop, len;
 	u32 work;
-
 	u8 tbus, tdevice, tslot;
 
-	PCIIRQRoutingInfoLength = pcibios_get_irq_routing_table();
-	if (!PCIIRQRoutingInfoLength)
-		return -1;
-
-	len = (PCIIRQRoutingInfoLength->size -
-	       sizeof(struct irq_routing_table)) / sizeof(struct irq_info);
-	/* Make sure I got at least one entry */
-	if (len == 0) {
-		kfree(PCIIRQRoutingInfoLength );
-		return -1;
-	}
-
+	len = cpqhp_routing_table_length();
 	for (loop = 0; loop < len; ++loop) {
-		tbus = PCIIRQRoutingInfoLength->slots[loop].bus;
-		tdevice = PCIIRQRoutingInfoLength->slots[loop].devfn;
-		tslot = PCIIRQRoutingInfoLength->slots[loop].slot;
+		tbus = cpqhp_routing_table->slots[loop].bus;
+		tdevice = cpqhp_routing_table->slots[loop].devfn;
+		tslot = cpqhp_routing_table->slots[loop].slot;
 
 		if (tslot == slot) {
 			*bus_num = tbus;
 			*dev_num = tdevice;
 			ctrl->pci_bus->number = tbus;
 			pci_bus_read_config_dword (ctrl->pci_bus, *dev_num, PCI_VENDOR_ID, &work);
-			if (!nobridge || (work == 0xffffffff)) {
-				kfree(PCIIRQRoutingInfoLength );
+			if (!nobridge || (work == 0xffffffff))
 				return 0;
-			}
 
 			dbg("bus_num %d devfn %d\n", *bus_num, *dev_num);
 			pci_bus_read_config_dword (ctrl->pci_bus, *dev_num, PCI_CLASS_REVISION, &work);
@@ -287,17 +270,12 @@ static int PCI_GetBusDevHelper(struct controller *ctrl, u8 *bus_num, u8 *dev_num
 				dbg("Scan bus for Non Bridge: bus %d\n", tbus);
 				if (PCI_ScanBusForNonBridge(ctrl, tbus, dev_num) == 0) {
 					*bus_num = tbus;
-					kfree(PCIIRQRoutingInfoLength );
 					return 0;
 				}
-			} else {
-				kfree(PCIIRQRoutingInfoLength );
+			} else
 				return 0;
-			}
-
 		}
 	}
-	kfree(PCIIRQRoutingInfoLength );
 	return -1;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From e3265ea364c7ed5accc9955f8b805c380149870f Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:24:07 -0600
Subject: PCI Hotplug: cpqphp: eliminate dead code - PCI_ScanBusNonBridge

I have no clue what the original intent here was, but the code as
written is useless.

The old dbg() statement above the old callsite might lead one to think
that at one point, there was supposed to be some recursion, but any
sense of sanity here has been lost to the ravages of time.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp_pci.c | 14 ++------------
 1 file changed, 2 insertions(+), 12 deletions(-)

diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c
index 6201281b9dab..3e3acc7e7461 100644
--- a/drivers/pci/hotplug/cpqphp_pci.c
+++ b/drivers/pci/hotplug/cpqphp_pci.c
@@ -193,16 +193,6 @@ int cpqhp_set_irq (u8 bus_num, u8 dev_num, u8 int_pin, u8 irq_num)
 }
 
 
-/*
- * WTF??? This function isn't in the code, yet a function calls it, but the
- * compiler optimizes it away?  strange.  Here as a placeholder to keep the
- * compiler happy.
- */
-static int PCI_ScanBusNonBridge (u8 bus, u8 device)
-{
-	return 0;
-}
-
 static int PCI_ScanBusForNonBridge(struct controller *ctrl, u8 bus_num, u8 * dev_num)
 {
 	u16 tdevice;
@@ -231,9 +221,9 @@ static int PCI_ScanBusForNonBridge(struct controller *ctrl, u8 bus_num, u8 * dev
 		/* Yep we got one. bridge ? */
 		if ((work >> 8) == PCI_TO_PCI_BRIDGE_CLASS) {
 			pci_bus_read_config_byte (ctrl->pci_bus, PCI_DEVFN(tdevice, 0), PCI_SECONDARY_BUS, &tbus);
+			/* XXX: no recursion, wtf? */
 			dbg("Recurse on bus_num %d tdevice %d\n", tbus, tdevice);
-			if (PCI_ScanBusNonBridge(tbus, tdevice) == 0)
-				return 0;
+			return 0;
 		}
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 6d1e87daeeba864a18535b7f3aed0e65f5f52275 Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:24:12 -0600
Subject: PCI Hotplug: cpqphp: constify slot_name()

Eliminate this warning:
	warning: return discards qualifiers from pointer target type

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/cpqphp.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/hotplug/cpqphp.h b/drivers/pci/hotplug/cpqphp.h
index 1d5561c6bad4..53836001d511 100644
--- a/drivers/pci/hotplug/cpqphp.h
+++ b/drivers/pci/hotplug/cpqphp.h
@@ -464,7 +464,7 @@ extern u8 cpqhp_disk_irq;
 
 /* inline functions */
 
-static inline char *slot_name(struct slot *slot)
+static inline const char *slot_name(struct slot *slot)
 {
 	return hotplug_slot_name(slot->hotplug_slot);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 12a9da0fcb147b46de33bb919b1de2bb92c9e2a9 Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:24:17 -0600
Subject: PCI Hotplug: cpqphp: don't use pci_find_slot()

Convert uses of pci_find_slot to modern API.

In the conversion sites, we end up calling pci_dev_put() right away.

This may seem like it misses the entire point of doing something like
pci_get_bus_and_slot(), since we drop the reference so soon, but it turns
out we don't actually do much with the returned pci_dev.

I plan on untangling cpqphp further, but clearly cpqphp never worried too
much about a properly refcounted pci_dev anyway. For now, this conversion
seems reasonable, as it gets rid of the last in-tree caller of pci_find_slot.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/Kconfig      |  2 +-
 drivers/pci/hotplug/cpqphp_pci.c | 16 +++++++++++-----
 2 files changed, 12 insertions(+), 6 deletions(-)

diff --git a/drivers/pci/hotplug/Kconfig b/drivers/pci/hotplug/Kconfig
index 9aa4fe100a0d..ac888ccfa161 100644
--- a/drivers/pci/hotplug/Kconfig
+++ b/drivers/pci/hotplug/Kconfig
@@ -41,7 +41,7 @@ config HOTPLUG_PCI_FAKE
 
 config HOTPLUG_PCI_COMPAQ
 	tristate "Compaq PCI Hotplug driver"
-	depends on X86 && PCI_BIOS && PCI_LEGACY
+	depends on X86 && PCI_BIOS
 	help
 	  Say Y here if you have a motherboard with a Compaq PCI Hotplug
 	  controller.
diff --git a/drivers/pci/hotplug/cpqphp_pci.c b/drivers/pci/hotplug/cpqphp_pci.c
index 3e3acc7e7461..6173b9a4544e 100644
--- a/drivers/pci/hotplug/cpqphp_pci.c
+++ b/drivers/pci/hotplug/cpqphp_pci.c
@@ -88,7 +88,7 @@ int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func)
 	int num;
 
 	if (func->pci_dev == NULL)
-		func->pci_dev = pci_find_slot(func->bus, PCI_DEVFN(func->device, func->function));
+		func->pci_dev = pci_get_bus_and_slot(func->bus,PCI_DEVFN(func->device, func->function));
 
 	/* No pci device, we need to create it then */
 	if (func->pci_dev == NULL) {
@@ -98,7 +98,7 @@ int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func)
 		if (num)
 			pci_bus_add_devices(ctrl->pci_dev->bus);
 
-		func->pci_dev = pci_find_slot(func->bus, PCI_DEVFN(func->device, func->function));
+		func->pci_dev = pci_get_bus_and_slot(func->bus, PCI_DEVFN(func->device, func->function));
 		if (func->pci_dev == NULL) {
 			dbg("ERROR: pci_dev still null\n");
 			return 0;
@@ -111,6 +111,8 @@ int cpqhp_configure_device (struct controller* ctrl, struct pci_func* func)
 		pci_do_scan_bus(child);
 	}
 
+	pci_dev_put(func->pci_dev);
+
 	return 0;
 }
 
@@ -122,9 +124,11 @@ int cpqhp_unconfigure_device(struct pci_func* func)
 	dbg("%s: bus/dev/func = %x/%x/%x\n", __func__, func->bus, func->device, func->function);
 
 	for (j=0; j<8 ; j++) {
-		struct pci_dev* temp = pci_find_slot(func->bus, PCI_DEVFN(func->device, j));
-		if (temp)
+		struct pci_dev* temp = pci_get_bus_and_slot(func->bus, PCI_DEVFN(func->device, j));
+		if (temp) {
+			pci_dev_put(temp);
 			pci_remove_bus_device(temp);
+		}
 	}
 	return 0;
 }
@@ -406,7 +410,7 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 			new_slot->switch_save = 0x10;
 			/* In case of unsupported board */
 			new_slot->status = DevError;
-			new_slot->pci_dev = pci_find_slot(new_slot->bus, (new_slot->device << 3) | new_slot->function);
+			new_slot->pci_dev = pci_get_bus_and_slot(new_slot->bus, (new_slot->device << 3) | new_slot->function);
 
 			for (cloop = 0; cloop < 0x20; cloop++) {
 				rc = pci_bus_read_config_dword(ctrl->pci_bus, PCI_DEVFN(device, function), cloop << 2, (u32 *) & (new_slot-> config_space [cloop]));
@@ -414,6 +418,8 @@ int cpqhp_save_config(struct controller *ctrl, int busnumber, int is_hot_plug)
 					return rc;
 			}
 
+			pci_dev_put(new_slot->pci_dev);
+
 			function++;
 
 			stop_it = 0;
-- 
cgit v1.2.3-59-g8ed1b


From 3b073eda9557975a87a27b08a46a545fe8da66fb Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Tue, 31 Mar 2009 09:24:22 -0600
Subject: PCI: remove deprecated pci_find_slot() interface

The last in-tree caller of pci_find_slot has been converted, so
let's get rid of this deprecated interface.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/search.c | 30 ------------------------------
 include/linux/pci.h  |  8 --------
 2 files changed, 38 deletions(-)

diff --git a/drivers/pci/search.c b/drivers/pci/search.c
index 710d4ea69568..650bc0a538dc 100644
--- a/drivers/pci/search.c
+++ b/drivers/pci/search.c
@@ -114,36 +114,6 @@ pci_find_next_bus(const struct pci_bus *from)
 }
 
 #ifdef CONFIG_PCI_LEGACY
-/**
- * pci_find_slot - locate PCI device from a given PCI slot
- * @bus: number of PCI bus on which desired PCI device resides
- * @devfn: encodes number of PCI slot in which the desired PCI
- * device resides and the logical device number within that slot
- * in case of multi-function devices.
- *
- * Given a PCI bus and slot/function number, the desired PCI device
- * is located in system global list of PCI devices.  If the device
- * is found, a pointer to its data structure is returned.  If no
- * device is found, %NULL is returned.
- *
- * NOTE: Do not use this function any more; use pci_get_slot() instead, as
- * the PCI device returned by this function can disappear at any moment in
- * time.
- */
-struct pci_dev *pci_find_slot(unsigned int bus, unsigned int devfn)
-{
-	struct pci_dev *dev = NULL;
-
-	while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
-		if (dev->bus->number == bus && dev->devfn == devfn) {
-			pci_dev_put(dev);
-			return dev;
-		}
-	}
-	return NULL;
-}
-EXPORT_SYMBOL(pci_find_slot);
-
 /**
  * pci_find_device - begin or continue searching for a PCI device by vendor/device id
  * @vendor: PCI vendor id to match, or %PCI_ANY_ID to match all vendor ids
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 6dfa47d25ba4..19ee92c53ef7 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -599,8 +599,6 @@ extern void pci_sort_breadthfirst(void);
 struct pci_dev __deprecated *pci_find_device(unsigned int vendor,
 					     unsigned int device,
 					     struct pci_dev *from);
-struct pci_dev __deprecated *pci_find_slot(unsigned int bus,
-					   unsigned int devfn);
 #endif /* CONFIG_PCI_LEGACY */
 
 enum pci_lost_interrupt_reason {
@@ -936,12 +934,6 @@ static inline struct pci_dev *pci_find_device(unsigned int vendor,
 	return NULL;
 }
 
-static inline struct pci_dev *pci_find_slot(unsigned int bus,
-					    unsigned int devfn)
-{
-	return NULL;
-}
-
 static inline struct pci_dev *pci_get_device(unsigned int vendor,
 					     unsigned int device,
 					     struct pci_dev *from)
-- 
cgit v1.2.3-59-g8ed1b


From 9e9f46c44e487af0a82eb61b624553e2f7118f5b Mon Sep 17 00:00:00 2001
From: Jesse Barnes <jbarnes@virtuousgeek.org>
Date: Thu, 11 Jun 2009 10:58:28 -0700
Subject: PCI: use ACPI _CRS data by default

At this point, it seems to solve more problems than it causes, so let's try using it by default.  It's an easy revert if it ends up causing trouble.

Reviewed-by: Yinghai Lu <yhlu.kernel@gmail.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 Documentation/kernel-parameters.txt | 2 +-
 arch/x86/include/asm/pci_x86.h      | 2 +-
 arch/x86/pci/acpi.c                 | 2 +-
 arch/x86/pci/amd_bus.c              | 2 +-
 arch/x86/pci/common.c               | 4 ++--
 5 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index fd5cac013037..7bdaf5080408 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1787,7 +1787,7 @@ and is between 256 and 4096 characters. It is defined in the file
 				IRQ routing is enabled.
 		noacpi		[X86] Do not use ACPI for IRQ routing
 				or for PCI scanning.
-		use_crs		[X86] Use _CRS for PCI resource
+		nocrs		[X86] Don't use _CRS for PCI resource
 				allocation.
 		routeirq	Do IRQ routing for all PCI devices.
 				This is normally done in pci_enable_device(),
diff --git a/arch/x86/include/asm/pci_x86.h b/arch/x86/include/asm/pci_x86.h
index e60fd3e14bdf..cb739cc0a080 100644
--- a/arch/x86/include/asm/pci_x86.h
+++ b/arch/x86/include/asm/pci_x86.h
@@ -25,7 +25,7 @@
 #define PCI_BIOS_IRQ_SCAN	0x2000
 #define PCI_ASSIGN_ALL_BUSSES	0x4000
 #define PCI_CAN_SKIP_ISA_ALIGN	0x8000
-#define PCI_USE__CRS		0x10000
+#define PCI_NO_ROOT_CRS		0x10000
 #define PCI_CHECK_ENABLE_AMD_MMCONF	0x20000
 #define PCI_HAS_IO_ECS		0x40000
 #define PCI_NOASSIGN_ROMS	0x80000
diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c
index c0ecf250fe51..8d898e0d3609 100644
--- a/arch/x86/pci/acpi.c
+++ b/arch/x86/pci/acpi.c
@@ -217,7 +217,7 @@ struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_device *device, int do
 #endif
 	}
 
-	if (bus && (pci_probe & PCI_USE__CRS))
+	if (bus && !(pci_probe & PCI_NO_ROOT_CRS))
 		get_current_resources(device, busnum, domain, bus);
 	return bus;
 }
diff --git a/arch/x86/pci/amd_bus.c b/arch/x86/pci/amd_bus.c
index f893d6a6e803..2255f880678b 100644
--- a/arch/x86/pci/amd_bus.c
+++ b/arch/x86/pci/amd_bus.c
@@ -101,7 +101,7 @@ void x86_pci_root_bus_res_quirks(struct pci_bus *b)
 	struct pci_root_info *info;
 
 	/* don't go for it if _CRS is used */
-	if (pci_probe & PCI_USE__CRS)
+	if (!(pci_probe & PCI_NO_ROOT_CRS))
 		return;
 
 	/* if only one root bus, don't need to anything */
diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c
index 2202b6257b82..4740119e4bb7 100644
--- a/arch/x86/pci/common.c
+++ b/arch/x86/pci/common.c
@@ -515,8 +515,8 @@ char * __devinit  pcibios_setup(char *str)
 	} else if (!strcmp(str, "assign-busses")) {
 		pci_probe |= PCI_ASSIGN_ALL_BUSSES;
 		return NULL;
-	} else if (!strcmp(str, "use_crs")) {
-		pci_probe |= PCI_USE__CRS;
+	} else if (!strcmp(str, "nocrs")) {
+		pci_probe |= PCI_NO_ROOT_CRS;
 		return NULL;
 	} else if (!strcmp(str, "earlydump")) {
 		pci_early_dump_regs = 1;
-- 
cgit v1.2.3-59-g8ed1b


From 8e822df700694ca6850d1e0c122fd7004b2778d8 Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Mon, 8 Jun 2009 09:27:25 +0800
Subject: PCI: disable ASPM on VIA root-port-under-bridge configurations

VIA has a strange chipset, it has root port under a bridge. Disable ASPM
for such strange chipset.

Cc: stable@kernel.org
Tested-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index b0367f168af4..777b2c76caf5 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -638,6 +638,10 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
 	if (pdev->pcie_type != PCI_EXP_TYPE_ROOT_PORT &&
 		pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM)
 		return;
+	/* VIA has a strange chipset, root port is under a bridge */
+	if (pdev->pcie_type == PCI_EXP_TYPE_ROOT_PORT &&
+		pdev->bus->self)
+		return;
 	down_read(&pci_bus_sem);
 	if (list_empty(&pdev->subordinate->devices))
 		goto out;
-- 
cgit v1.2.3-59-g8ed1b


From 57fbf52c86addd8e25d1975fac0d59d982d1f6ec Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@redhat.com>
Date: Thu, 7 May 2009 11:28:41 +0300
Subject: PCI MSI: let drivers retry when not enough vectors

pci_enable_msix currently returns -EINVAL if you ask
for more vectors than supported by the device, which would
typically cause fallback to regular interrupts.

It's better to return the table size, making the driver retry
MSI-X with less vectors.

Reviewed-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/msi.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 7ffac27d5d4a..f2725710593a 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -691,8 +691,8 @@ int pci_msix_table_size(struct pci_dev *dev)
  * indicates the successful configuration of MSI-X capability structure
  * with new allocated MSI-X irqs. A return of < 0 indicates a failure.
  * Or a return of > 0 indicates that driver request is exceeding the number
- * of irqs available. Driver should use the returned value to re-send
- * its request.
+ * of irqs or MSI-X vectors available. Driver should use the returned value to
+ * re-send its request.
  **/
 int pci_enable_msix(struct pci_dev* dev, struct msix_entry *entries, int nvec)
 {
@@ -708,7 +708,7 @@ int pci_enable_msix(struct pci_dev* dev, struct msix_entry *entries, int nvec)
 
 	nr_entries = pci_msix_table_size(dev);
 	if (nvec > nr_entries)
-		return -EINVAL;
+		return nr_entries;
 
 	/* Check for any invalid entries */
 	for (i = 0; i < nvec; i++) {
-- 
cgit v1.2.3-59-g8ed1b


From af4c5f985afd8d4cfdf402aaa03677f2cb96e37c Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Mon, 18 May 2009 19:02:38 -0600
Subject: PCI: eliminate redundant pci_stop_dev() call from pci_destroy_dev()

We always call pci_stop_bus_device before calling pci_destroy_dev.

Since pci_stop_bus_device calls pci_stop_dev, there is no need
for pci_destroy_dev to repeat the call.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/remove.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/pci/remove.c b/drivers/pci/remove.c
index 86503c14ce7e..176615e7231f 100644
--- a/drivers/pci/remove.c
+++ b/drivers/pci/remove.c
@@ -32,8 +32,6 @@ static void pci_stop_dev(struct pci_dev *dev)
 
 static void pci_destroy_dev(struct pci_dev *dev)
 {
-	pci_stop_dev(dev);
-
 	/* Remove the device from the device lists, and prevent any further
 	 * list accesses from this device */
 	down_write(&pci_bus_sem);
-- 
cgit v1.2.3-59-g8ed1b


From 4d135dbee7b0a89e946f7ba284f2b957505a2c3a Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Wed, 20 May 2009 17:11:57 +0800
Subject: PCI: fix SR-IOV function dependency link problem

PCIe root complex integrated endpoint does not implement ARI, so this
kind of endpoint uses 3-bit function number. The function dependency
link of the integrated endpoint should be calculated using the device
number plus the value from function dependency link register.

Normal endpoint always implements ARI and the function dependency link
register contains 8-bit function number (i.e. `devfn' from software's
perspective).

Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/iov.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c
index b497daab3d4a..e87fe95da814 100644
--- a/drivers/pci/iov.c
+++ b/drivers/pci/iov.c
@@ -487,6 +487,8 @@ found:
 	iov->self = dev;
 	pci_read_config_dword(dev, pos + PCI_SRIOV_CAP, &iov->cap);
 	pci_read_config_byte(dev, pos + PCI_SRIOV_FUNC_LINK, &iov->link);
+	if (dev->pcie_type == PCI_EXP_TYPE_RC_END)
+		iov->link = PCI_DEVFN(PCI_SLOT(dev->devfn), iov->link);
 
 	if (pdev)
 		iov->dev = pci_dev_get(pdev);
-- 
cgit v1.2.3-59-g8ed1b


From f62795f1e892ca9269849fa83de97621da7e02c0 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 18 May 2009 22:51:12 +0200
Subject: PCI PM: Follow PCI_PM_CTRL_NO_SOFT_RESET during transitions from D3

According to the PCI PM specification (PCI Bus Power Management
Interface Specification, Rev. 1.2, Section 5.4.1) we are supposed to
reinitialize devices that have PCI_PM_CTRL_NO_SOFT_RESET clear during
all transitions from PCI_D3hot to PCI_D0, but we only do it if the
device's current_state field is equal to PCI_UNKNOWN.

This may lead to problems if a device with PCI_PM_CTRL_NO_SOFT_RESET
unset is put into PCI_D3hot at run time by its driver and
pci_set_power_state() is used to put it back into PCI_D0, because in
that case the device will remain uninitialized after
pci_set_power_state() has returned.  Prevent that from happening by
modifying pci_raw_set_power_state() to reinitialize devices with
PCI_PM_CTRL_NO_SOFT_RESET unset during all transitions from D3 to D0.

Cc: stable@kernel.org
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 1a91bf9687af..761557688b18 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -480,6 +480,8 @@ static int pci_raw_set_power_state(struct pci_dev *dev, pci_power_t state)
 		pmcsr &= ~PCI_PM_CTRL_STATE_MASK;
 		pmcsr |= state;
 		break;
+	case PCI_D3hot:
+	case PCI_D3cold:
 	case PCI_UNKNOWN: /* Boot-up */
 		if ((pmcsr & PCI_PM_CTRL_STATE_MASK) == PCI_D3hot
 		 && !(pmcsr & PCI_PM_CTRL_NO_SOFT_RESET))
-- 
cgit v1.2.3-59-g8ed1b


From 43c16408842b0eeb367c23a6fa540ce69f99e347 Mon Sep 17 00:00:00 2001
From: Andrew Patterson <andrew.patterson@hp.com>
Date: Wed, 22 Apr 2009 16:52:09 -0600
Subject: PCI: Add support for turning PCIe ECRC on or off

Adds support for PCI Express transaction layer end-to-end CRC checking
(ECRC).  This patch will enable/disable ECRC checking by setting/clearing
the ECRC Check Enable and/or ECRC Generation Enable bits for devices that
support ECRC.

The ECRC setting is controlled by the "pci=ecrc=<policy>" command-line
option. If this option is not set or is set to 'bios", the enable and
generation bits are left in whatever state that firmware/BIOS set them to.
The "off" setting turns them off, and the "on" option turns them on (if the
device supports it).

Turning ECRC on or off can be a data integrity versus performance
tradeoff.  In theory, turning it on will catch more data errors, turning
it off means possibly better performance since CRC does not need to be
calculated by the PCIe hardware and packet sizes are reduced.

Signed-off-by: Andrew Patterson <andrew.patterson@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 Documentation/kernel-parameters.txt |   6 ++
 drivers/pci/pci.c                   |   2 +
 drivers/pci/pcie/aer/Kconfig        |  13 ++++
 drivers/pci/pcie/aer/Makefile       |   2 +
 drivers/pci/pcie/aer/aerdrv_core.c  |  16 +++--
 drivers/pci/pcie/aer/ecrc.c         | 131 ++++++++++++++++++++++++++++++++++++
 include/linux/pci.h                 |  11 +++
 7 files changed, 174 insertions(+), 7 deletions(-)
 create mode 100644 drivers/pci/pcie/aer/ecrc.c

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 7bdaf5080408..395d1a013ebb 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1824,6 +1824,12 @@ and is between 256 and 4096 characters. It is defined in the file
 				PAGE_SIZE is used as alignment.
 				PCI-PCI bridge can be specified, if resource
 				windows need to be expanded.
+		ecrc=		Enable/disable PCIe ECRC (transaction layer
+				end-to-end CRC checking).
+				bios: Use BIOS/firmware settings. This is the
+				the default.
+				off: Turn ECRC off
+				on: Turn ECRC on.
 
 	pcie_aspm=	[PCIE] Forcibly enable or disable PCIe Active State Power
 			Management.
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 761557688b18..56fb18d2cb52 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -2588,6 +2588,8 @@ static int __init pci_setup(char *str)
 			} else if (!strncmp(str, "resource_alignment=", 19)) {
 				pci_set_resource_alignment_param(str + 19,
 							strlen(str + 19));
+			} else if (!strncmp(str, "ecrc=", 5)) {
+				pcie_ecrc_get_policy(str + 5);
 			} else {
 				printk(KERN_ERR "PCI: Unknown option `%s'\n",
 						str);
diff --git a/drivers/pci/pcie/aer/Kconfig b/drivers/pci/pcie/aer/Kconfig
index c3bde588aa13..db4cb950933a 100644
--- a/drivers/pci/pcie/aer/Kconfig
+++ b/drivers/pci/pcie/aer/Kconfig
@@ -10,3 +10,16 @@ config PCIEAER
 	  This enables PCI Express Root Port Advanced Error Reporting
 	  (AER) driver support. Error reporting messages sent to Root
 	  Port will be handled by PCI Express AER driver.
+
+
+#
+# PCI Express ECRC
+#
+config PCIE_ECRC
+	bool "PCI Express ECRC settings control"
+	depends on PCIEAER
+	help
+	  Used to override firmware/bios settings for PCI Express ECRC
+	  (transaction layer end-to-end CRC checking).
+
+	  When in doubt, say N.
diff --git a/drivers/pci/pcie/aer/Makefile b/drivers/pci/pcie/aer/Makefile
index 8da3bd8455a8..7f93411c56e5 100644
--- a/drivers/pci/pcie/aer/Makefile
+++ b/drivers/pci/pcie/aer/Makefile
@@ -4,6 +4,8 @@
 
 obj-$(CONFIG_PCIEAER) += aerdriver.o
 
+obj-$(CONFIG_PCIE_ECRC)	+= ecrc.o
+
 aerdriver-objs := aerdrv_errprint.o aerdrv_core.o aerdrv.o
 aerdriver-$(CONFIG_ACPI) += aerdrv_acpi.o
 
diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c
index 307452f30035..dd3829e68e3f 100644
--- a/drivers/pci/pcie/aer/aerdrv_core.c
+++ b/drivers/pci/pcie/aer/aerdrv_core.c
@@ -113,15 +113,17 @@ static void set_device_error_reporting(struct pci_dev *dev, void *data)
 {
 	bool enable = *((bool *)data);
 
-	if (dev->pcie_type != PCIE_RC_PORT &&
-	    dev->pcie_type != PCIE_SW_UPSTREAM_PORT &&
-	    dev->pcie_type != PCIE_SW_DOWNSTREAM_PORT)
-		return;
+	if (dev->pcie_type == PCIE_RC_PORT ||
+	    dev->pcie_type == PCIE_SW_UPSTREAM_PORT ||
+	    dev->pcie_type == PCIE_SW_DOWNSTREAM_PORT) {
+		if (enable)
+			pci_enable_pcie_error_reporting(dev);
+		else
+			pci_disable_pcie_error_reporting(dev);
+	}
 
 	if (enable)
-		pci_enable_pcie_error_reporting(dev);
-	else
-		pci_disable_pcie_error_reporting(dev);
+		pcie_set_ecrc_checking(dev);
 }
 
 /**
diff --git a/drivers/pci/pcie/aer/ecrc.c b/drivers/pci/pcie/aer/ecrc.c
new file mode 100644
index 000000000000..ece97df4df6d
--- /dev/null
+++ b/drivers/pci/pcie/aer/ecrc.c
@@ -0,0 +1,131 @@
+/*
+ *    Enables/disables PCIe ECRC checking.
+ *
+ *    (C) Copyright 2009 Hewlett-Packard Development Company, L.P.
+ *    Andrew Patterson <andrew.patterson@hp.com>
+ *
+ *    This program is free software; you can redistribute it and/or modify
+ *    it under the terms of the GNU General Public License as published by
+ *    the Free Software Foundation; version 2 of the License.
+ *
+ *    This program is distributed in the hope that it will be useful,
+ *    but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *    MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ *    General Public License for more details.
+ *
+ *    You should have received a copy of the GNU General Public License
+ *    along with this program; if not, write to the Free Software
+ *    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+ *    02111-1307, USA.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/pci.h>
+#include <linux/pci_regs.h>
+#include <linux/errno.h>
+#include "../../pci.h"
+
+#define ECRC_POLICY_DEFAULT 0		/* ECRC set by BIOS */
+#define ECRC_POLICY_OFF     1		/* ECRC off for performance */
+#define ECRC_POLICY_ON      2		/* ECRC on for data integrity */
+
+static int ecrc_policy = ECRC_POLICY_DEFAULT;
+
+static const char *ecrc_policy_str[] = {
+	[ECRC_POLICY_DEFAULT] = "bios",
+	[ECRC_POLICY_OFF] = "off",
+	[ECRC_POLICY_ON] = "on"
+};
+
+/**
+ * enable_ercr_checking - enable PCIe ECRC checking for a device
+ * @dev: the PCI device
+ *
+ * Returns 0 on success, or negative on failure.
+ */
+static int enable_ecrc_checking(struct pci_dev *dev)
+{
+	int pos;
+	u32 reg32;
+
+	if (!dev->is_pcie)
+		return -ENODEV;
+
+	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR);
+	if (!pos)
+		return -ENODEV;
+
+	pci_read_config_dword(dev, pos + PCI_ERR_CAP, &reg32);
+	if (reg32 & PCI_ERR_CAP_ECRC_GENC)
+		reg32 |= PCI_ERR_CAP_ECRC_GENE;
+	if (reg32 & PCI_ERR_CAP_ECRC_CHKC)
+		reg32 |= PCI_ERR_CAP_ECRC_CHKE;
+	pci_write_config_dword(dev, pos + PCI_ERR_CAP, reg32);
+
+	return 0;
+}
+
+/**
+ * disable_ercr_checking - disables PCIe ECRC checking for a device
+ * @dev: the PCI device
+ *
+ * Returns 0 on success, or negative on failure.
+ */
+static int disable_ecrc_checking(struct pci_dev *dev)
+{
+	int pos;
+	u32 reg32;
+
+	if (!dev->is_pcie)
+		return -ENODEV;
+
+	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR);
+	if (!pos)
+		return -ENODEV;
+
+	pci_read_config_dword(dev, pos + PCI_ERR_CAP, &reg32);
+	reg32 &= ~(PCI_ERR_CAP_ECRC_GENE | PCI_ERR_CAP_ECRC_CHKE);
+	pci_write_config_dword(dev, pos + PCI_ERR_CAP, reg32);
+
+	return 0;
+}
+
+/**
+ * pcie_set_ecrc_checking - set/unset PCIe ECRC checking for a device based on global policy
+ * @dev: the PCI device
+ */
+void pcie_set_ecrc_checking(struct pci_dev *dev)
+{
+	switch (ecrc_policy) {
+	case ECRC_POLICY_DEFAULT:
+		return;
+	case ECRC_POLICY_OFF:
+		disable_ecrc_checking(dev);
+		break;
+	case ECRC_POLICY_ON:
+		enable_ecrc_checking(dev);;
+		break;
+	default:
+		return;
+	}
+}
+
+/**
+ * pcie_ecrc_get_policy - parse kernel command-line ecrc option
+ */
+void pcie_ecrc_get_policy(char *str)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(ecrc_policy_str); i++)
+		if (!strncmp(str, ecrc_policy_str[i],
+			     strlen(ecrc_policy_str[i])))
+			break;
+	if (i >= ARRAY_SIZE(ecrc_policy_str))
+		return;
+
+	ecrc_policy = i;
+}
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 19ee92c53ef7..ec03b90d3510 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -878,6 +878,17 @@ static inline int pcie_aspm_enabled(void)
 extern int pcie_aspm_enabled(void);
 #endif
 
+#ifndef CONFIG_PCIE_ECRC
+static inline void pcie_set_ecrc_checking(struct pci_dev *dev)
+{
+	return;
+}
+static inline void pcie_ecrc_get_policy(char *str) {};
+#else
+extern void pcie_set_ecrc_checking(struct pci_dev *dev);
+extern void pcie_ecrc_get_policy(char *str);
+#endif
+
 #define pci_enable_msi(pdev)	pci_enable_msi_block(pdev, 1)
 
 #ifdef CONFIG_HT_IRQ
-- 
cgit v1.2.3-59-g8ed1b


From 4096ed0fc7c6ed83da44ab27ab7d1c7cd6e8e175 Mon Sep 17 00:00:00 2001
From: Mats Erik Andersson <mats.andersson@gisladisker.se>
Date: Tue, 12 May 2009 12:05:23 +0200
Subject: PCI: expose SMBus on Asus notebook A6L

Addition of one unknown subsystem identifier to the quirks handler for
chipset i82855GM_HB on notebook Asus A6L. This exposes the otherwise
hidden SMBus controller within the south bridge ICH4-M.

Signed-off-by: Mats Erik Andersson <mats.andersson@gisladisker.se>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/quirks.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 3067673d54f6..a778c84d04a9 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -1133,6 +1133,7 @@ static void __init asus_hides_smbus_hostbridge(struct pci_dev *dev)
 			switch (dev->subsystem_device) {
 			case 0x1751: /* M2N notebook */
 			case 0x1821: /* M5N notebook */
+			case 0x1897: /* A6L notebook */
 				asus_hides_smbus = 1;
 			}
 		else if (dev->device == PCI_DEVICE_ID_INTEL_82855PM_HB)
-- 
cgit v1.2.3-59-g8ed1b


From c636f753b5b943f08fb3490e7f1a9b47aa5cc7d3 Mon Sep 17 00:00:00 2001
From: Len Brown <len.brown@intel.com>
Date: Tue, 19 May 2009 23:47:38 -0400
Subject: ACPI: delete dead acpi_disabled setting code

Testing CONFIG_ACPI inside boot.c is a waste of text,
since boot.c is built only when CONFIG_ACPI=y

Signed-off-by: Len Brown <len.brown@intel.com>
---
 arch/x86/kernel/acpi/boot.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 723989d7f802..d4acedf07866 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -43,11 +43,7 @@
 
 static int __initdata acpi_force = 0;
 u32 acpi_rsdt_forced;
-#ifdef	CONFIG_ACPI
-int acpi_disabled = 0;
-#else
-int acpi_disabled = 1;
-#endif
+int acpi_disabled;
 EXPORT_SYMBOL(acpi_disabled);
 
 #ifdef	CONFIG_X86_64
-- 
cgit v1.2.3-59-g8ed1b


From d023e49118b9c93bbab9aaf798b25f78f1a5803c Mon Sep 17 00:00:00 2001
From: Olivier Berger <oberger@ouvaton.org>
Date: Thu, 21 May 2009 16:07:38 +0200
Subject: ACPI: Remove Asus P4B266 from blacklist

See http://marc.info/?l=linux-acpi&m=124068823904429&w=2 for discussion

Signed-off-by: Olivier Berger <oberger@ouvaton.org>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 arch/x86/kernel/acpi/boot.c | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index d4acedf07866..817d6a5e115d 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -1563,14 +1563,6 @@ static struct dmi_system_id __initdata acpi_dmi_table[] = {
 		     DMI_MATCH(DMI_PRODUCT_NAME, "Workstation W8000"),
 		     },
 	 },
-	{
-	 .callback = force_acpi_ht,
-	 .ident = "ASUS P4B266",
-	 .matches = {
-		     DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
-		     DMI_MATCH(DMI_BOARD_NAME, "P4B266"),
-		     },
-	 },
 	{
 	 .callback = force_acpi_ht,
 	 .ident = "ASUS P2B-DS",
-- 
cgit v1.2.3-59-g8ed1b


From 24c5c4c2f506bf87ef2343669fb892c944c3fdde Mon Sep 17 00:00:00 2001
From: Zhao Yakui <yakui.zhao@intel.com>
Date: Thu, 21 May 2009 16:25:35 +0800
Subject: ACPI: increase size of acpi_bus_id[]

Previously [5], now [8].

sprintf(acpi_device_bid(device), "CPU%X", cpu_id)
now looks better on systems with more than 0xFF processors.

Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 include/acpi/acpi_bus.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index c34b11022908..0be24101a48f 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -168,7 +168,7 @@ struct acpi_device_dir {
 
 /* Plug and Play */
 
-typedef char acpi_bus_id[5];
+typedef char acpi_bus_id[8];
 typedef unsigned long acpi_bus_address;
 typedef char acpi_hardware_id[15];
 typedef char acpi_unique_id[9];
-- 
cgit v1.2.3-59-g8ed1b


From c4bf2f372db09ef8d16a25a60d523bfa1c50f7b5 Mon Sep 17 00:00:00 2001
From: Len Brown <len.brown@intel.com>
Date: Thu, 11 Jun 2009 23:53:55 -0400
Subject: ACPI, PCI, x86: move MCFG parsing routine from ACPI to PCI file

Move
arch/x86/kernel/acpi/boot.c: acpi_parse_mcfg()
to
arch/x86/pci/mmconfig-shared.c: pci_parse_mcfg()
where it is used, and make it static.

Move associated globals and helper routine with it.

No functional change.

This code move is in preparation for SFI support,
which will allow the PCI code to find the MCFG table
on systems which do not support ACPI.

Signed-off-by: Len Brown <len.brown@intel.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/x86/include/asm/pci_x86.h |  3 ++
 arch/x86/kernel/acpi/boot.c    | 66 ------------------------------------------
 arch/x86/pci/mmconfig-shared.c | 65 ++++++++++++++++++++++++++++++++++++++++-
 include/linux/acpi.h           |  3 --
 4 files changed, 67 insertions(+), 70 deletions(-)

diff --git a/arch/x86/include/asm/pci_x86.h b/arch/x86/include/asm/pci_x86.h
index e60fd3e14bdf..b399988eee3a 100644
--- a/arch/x86/include/asm/pci_x86.h
+++ b/arch/x86/include/asm/pci_x86.h
@@ -121,6 +121,9 @@ extern int __init pcibios_init(void);
 extern int __init pci_mmcfg_arch_init(void);
 extern void __init pci_mmcfg_arch_free(void);
 
+extern struct acpi_mcfg_allocation *pci_mmcfg_config;
+extern int pci_mmcfg_config_num;
+
 /*
  * AMD Fam10h CPUs are buggy, and cannot access MMIO config space
  * on their northbrige except through the * %eax register. As such, you MUST
diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index 817d6a5e115d..f54e0e557cd2 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -117,72 +117,6 @@ void __init __acpi_unmap_table(char *map, unsigned long size)
 	early_iounmap(map, size);
 }
 
-#ifdef CONFIG_PCI_MMCONFIG
-
-static int acpi_mcfg_64bit_base_addr __initdata = FALSE;
-
-/* The physical address of the MMCONFIG aperture.  Set from ACPI tables. */
-struct acpi_mcfg_allocation *pci_mmcfg_config;
-int pci_mmcfg_config_num;
-
-static int __init acpi_mcfg_oem_check(struct acpi_table_mcfg *mcfg)
-{
-	if (!strcmp(mcfg->header.oem_id, "SGI"))
-		acpi_mcfg_64bit_base_addr = TRUE;
-
-	return 0;
-}
-
-int __init acpi_parse_mcfg(struct acpi_table_header *header)
-{
-	struct acpi_table_mcfg *mcfg;
-	unsigned long i;
-	int config_size;
-
-	if (!header)
-		return -EINVAL;
-
-	mcfg = (struct acpi_table_mcfg *)header;
-
-	/* how many config structures do we have */
-	pci_mmcfg_config_num = 0;
-	i = header->length - sizeof(struct acpi_table_mcfg);
-	while (i >= sizeof(struct acpi_mcfg_allocation)) {
-		++pci_mmcfg_config_num;
-		i -= sizeof(struct acpi_mcfg_allocation);
-	};
-	if (pci_mmcfg_config_num == 0) {
-		printk(KERN_ERR PREFIX "MMCONFIG has no entries\n");
-		return -ENODEV;
-	}
-
-	config_size = pci_mmcfg_config_num * sizeof(*pci_mmcfg_config);
-	pci_mmcfg_config = kmalloc(config_size, GFP_KERNEL);
-	if (!pci_mmcfg_config) {
-		printk(KERN_WARNING PREFIX
-		       "No memory for MCFG config tables\n");
-		return -ENOMEM;
-	}
-
-	memcpy(pci_mmcfg_config, &mcfg[1], config_size);
-
-	acpi_mcfg_oem_check(mcfg);
-
-	for (i = 0; i < pci_mmcfg_config_num; ++i) {
-		if ((pci_mmcfg_config[i].address > 0xFFFFFFFF) &&
-		    !acpi_mcfg_64bit_base_addr) {
-			printk(KERN_ERR PREFIX
-			       "MMCONFIG not in low 4GB of memory\n");
-			kfree(pci_mmcfg_config);
-			pci_mmcfg_config_num = 0;
-			return -ENODEV;
-		}
-	}
-
-	return 0;
-}
-#endif				/* CONFIG_PCI_MMCONFIG */
-
 #ifdef CONFIG_X86_LOCAL_APIC
 static int __init acpi_parse_madt(struct acpi_table_header *table)
 {
diff --git a/arch/x86/pci/mmconfig-shared.c b/arch/x86/pci/mmconfig-shared.c
index 8766b0e216c5..712443ec6d43 100644
--- a/arch/x86/pci/mmconfig-shared.c
+++ b/arch/x86/pci/mmconfig-shared.c
@@ -523,6 +523,69 @@ reject:
 
 static int __initdata known_bridge;
 
+static int acpi_mcfg_64bit_base_addr __initdata = FALSE;
+
+/* The physical address of the MMCONFIG aperture.  Set from ACPI tables. */
+struct acpi_mcfg_allocation *pci_mmcfg_config;
+int pci_mmcfg_config_num;
+
+static int __init acpi_mcfg_oem_check(struct acpi_table_mcfg *mcfg)
+{
+	if (!strcmp(mcfg->header.oem_id, "SGI"))
+		acpi_mcfg_64bit_base_addr = TRUE;
+
+	return 0;
+}
+
+static int __init pci_parse_mcfg(struct acpi_table_header *header)
+{
+	struct acpi_table_mcfg *mcfg;
+	unsigned long i;
+	int config_size;
+
+	if (!header)
+		return -EINVAL;
+
+	mcfg = (struct acpi_table_mcfg *)header;
+
+	/* how many config structures do we have */
+	pci_mmcfg_config_num = 0;
+	i = header->length - sizeof(struct acpi_table_mcfg);
+	while (i >= sizeof(struct acpi_mcfg_allocation)) {
+		++pci_mmcfg_config_num;
+		i -= sizeof(struct acpi_mcfg_allocation);
+	};
+	if (pci_mmcfg_config_num == 0) {
+		printk(KERN_ERR PREFIX "MMCONFIG has no entries\n");
+		return -ENODEV;
+	}
+
+	config_size = pci_mmcfg_config_num * sizeof(*pci_mmcfg_config);
+	pci_mmcfg_config = kmalloc(config_size, GFP_KERNEL);
+	if (!pci_mmcfg_config) {
+		printk(KERN_WARNING PREFIX
+		       "No memory for MCFG config tables\n");
+		return -ENOMEM;
+	}
+
+	memcpy(pci_mmcfg_config, &mcfg[1], config_size);
+
+	acpi_mcfg_oem_check(mcfg);
+
+	for (i = 0; i < pci_mmcfg_config_num; ++i) {
+		if ((pci_mmcfg_config[i].address > 0xFFFFFFFF) &&
+		    !acpi_mcfg_64bit_base_addr) {
+			printk(KERN_ERR PREFIX
+			       "MMCONFIG not in low 4GB of memory\n");
+			kfree(pci_mmcfg_config);
+			pci_mmcfg_config_num = 0;
+			return -ENODEV;
+		}
+	}
+
+	return 0;
+}
+
 static void __init __pci_mmcfg_init(int early)
 {
 	/* MMCONFIG disabled */
@@ -543,7 +606,7 @@ static void __init __pci_mmcfg_init(int early)
 	}
 
 	if (!known_bridge)
-		acpi_table_parse(ACPI_SIG_MCFG, acpi_parse_mcfg);
+		acpi_table_parse(ACPI_SIG_MCFG, pci_parse_mcfg);
 
 	pci_mmcfg_reject_broken(early);
 
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index 88be890ee3c7..73cb141150df 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -113,9 +113,6 @@ void acpi_irq_stats_init(void);
 extern u32 acpi_irq_handled;
 extern u32 acpi_irq_not_handled;
 
-extern struct acpi_mcfg_allocation *pci_mmcfg_config;
-extern int pci_mmcfg_config_num;
-
 extern int sbf_port;
 extern unsigned long acpi_realmode_flags;
 
-- 
cgit v1.2.3-59-g8ed1b


From 4a7a16dc061e4c57bf288150f51bd4c2ace33723 Mon Sep 17 00:00:00 2001
From: Len Brown <len.brown@intel.com>
Date: Fri, 12 Jun 2009 20:42:08 -0400
Subject: ACPI: move declaration acpi_early_init() to acpi.h

Signed-off-by: Len Brown <len.brown@intel.com>
---
 include/linux/acpi.h | 3 +++
 init/main.c          | 6 +-----
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index 73cb141150df..bf17681cb06f 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -290,7 +290,10 @@ void __init acpi_s4_no_nvs(void);
 				OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL)
 
 extern acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 flags);
+extern void acpi_early_init(void);
+
 #else	/* CONFIG_ACPI */
+static inline void acpi_early_init(void) { }
 
 static inline int early_acpi_boot_init(void)
 {
diff --git a/init/main.c b/init/main.c
index d721dad05dd7..f1b9f0fdb1b4 100644
--- a/init/main.c
+++ b/init/main.c
@@ -24,6 +24,7 @@
 #include <linux/smp_lock.h>
 #include <linux/initrd.h>
 #include <linux/bootmem.h>
+#include <linux/acpi.h>
 #include <linux/tty.h>
 #include <linux/gfp.h>
 #include <linux/percpu.h>
@@ -86,11 +87,6 @@ extern void sbus_init(void);
 extern void prio_tree_init(void);
 extern void radix_tree_init(void);
 extern void free_initmem(void);
-#ifdef	CONFIG_ACPI
-extern void acpi_early_init(void);
-#else
-static inline void acpi_early_init(void) { }
-#endif
 #ifndef CONFIG_DEBUG_RODATA
 static inline void mark_rodata_ro(void) { }
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From ab46feae865c5b96dbf5e261be8638165932bfb1 Mon Sep 17 00:00:00 2001
From: Len Brown <len.brown@intel.com>
Date: Fri, 12 Jun 2009 20:47:50 -0400
Subject: ACPI: #define acpi_disabled 1 for CONFIG_ACPI=n

SFI will need to test acpi_disabled no matter
the value of CONFIG_ACPI.

Signed-off-by: Len Brown <len.brown@intel.com>
---
 arch/x86/include/asm/acpi.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h
index 4518dc500903..20d1465a2ab0 100644
--- a/arch/x86/include/asm/acpi.h
+++ b/arch/x86/include/asm/acpi.h
@@ -144,6 +144,7 @@ static inline unsigned int acpi_processor_cstate_check(unsigned int max_cstate)
 
 #else /* !CONFIG_ACPI */
 
+#define acpi_disabled 1
 #define acpi_lapic 0
 #define acpi_ioapic 0
 static inline void acpi_noirq_set(void) { }
-- 
cgit v1.2.3-59-g8ed1b


From 8616e0fc1e27295316f9821a883f0e9fa6f8200f Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Thu, 11 Jun 2009 10:27:28 -0400
Subject: cifs: remove unneeded NULL checks from cifs_show_options

show_options is always called with the namespace_sem held. Therefore we
don't need to worry about the vfsmount being NULL, or it vanishing while
the function is running. By the same token, there's no need to worry
about the superblock, tcon, smb or tcp sessions being NULL on entry.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/cifsfs.c | 130 +++++++++++++++++++++++++------------------------------
 1 file changed, 59 insertions(+), 71 deletions(-)

diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 0d92114195ab..3f121fbd6c56 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -346,80 +346,68 @@ cifs_show_options(struct seq_file *s, struct vfsmount *m)
 	struct TCP_Server_Info *server;
 
 	cifs_sb = CIFS_SB(m->mnt_sb);
+	tcon = cifs_sb->tcon;
 
-	if (cifs_sb) {
-		tcon = cifs_sb->tcon;
-		if (tcon) {
-			seq_printf(s, ",unc=%s", cifs_sb->tcon->treeName);
-			if (tcon->ses) {
-				if (tcon->ses->userName)
-					seq_printf(s, ",username=%s",
-					   tcon->ses->userName);
-				if (tcon->ses->domainName)
-					seq_printf(s, ",domain=%s",
-					   tcon->ses->domainName);
-				server = tcon->ses->server;
-				if (server) {
-					seq_printf(s, ",addr=");
-					switch (server->addr.sockAddr6.
-						sin6_family) {
-					case AF_INET6:
-						seq_printf(s, "%pI6",
-							   &server->addr.sockAddr6.sin6_addr);
-						break;
-					case AF_INET:
-						seq_printf(s, "%pI4",
-							   &server->addr.sockAddr.sin_addr.s_addr);
-						break;
-					}
-				}
-			}
-			if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID) ||
-			   !(tcon->unix_ext))
-				seq_printf(s, ",uid=%d", cifs_sb->mnt_uid);
-			if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID) ||
-			   !(tcon->unix_ext))
-				seq_printf(s, ",gid=%d", cifs_sb->mnt_gid);
-			if (!tcon->unix_ext) {
-				seq_printf(s, ",file_mode=0%o,dir_mode=0%o",
+	seq_printf(s, ",unc=%s", cifs_sb->tcon->treeName);
+	if (tcon->ses->userName)
+		seq_printf(s, ",username=%s", tcon->ses->userName);
+	if (tcon->ses->domainName)
+		seq_printf(s, ",domain=%s", tcon->ses->domainName);
+
+	cifs_show_address(s, tcon->ses->server);
+
+	seq_printf(s, ",uid=%d", cifs_sb->mnt_uid);
+	seq_printf(s, ",gid=%d", cifs_sb->mnt_gid);
+
+	server = tcon->ses->server;
+	seq_printf(s, ",addr=");
+	switch (server->addr.sockAddr6.sin6_family) {
+	case AF_INET6:
+		seq_printf(s, "%pI6", &server->addr.sockAddr6.sin6_addr);
+		break;
+	case AF_INET:
+		seq_printf(s, "%pI4", &server->addr.sockAddr.sin_addr.s_addr);
+		break;
+	}
+
+	if (!tcon->unix_ext)
+		seq_printf(s, ",file_mode=0%o,dir_mode=0%o",
 					   cifs_sb->mnt_file_mode,
 					   cifs_sb->mnt_dir_mode);
-			}
-			if (tcon->seal)
-				seq_printf(s, ",seal");
-			if (tcon->nocase)
-				seq_printf(s, ",nocase");
-			if (tcon->retry)
-				seq_printf(s, ",hard");
-		}
-		if (cifs_sb->prepath)
-			seq_printf(s, ",prepath=%s", cifs_sb->prepath);
-		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS)
-			seq_printf(s, ",posixpaths");
-		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID)
-			seq_printf(s, ",setuids");
-		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)
-			seq_printf(s, ",serverino");
-		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO)
-			seq_printf(s, ",directio");
-		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR)
-			seq_printf(s, ",nouser_xattr");
-		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR)
-			seq_printf(s, ",mapchars");
-		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL)
-			seq_printf(s, ",sfu");
-		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL)
-			seq_printf(s, ",nobrl");
-		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL)
-			seq_printf(s, ",cifsacl");
-		if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM)
-			seq_printf(s, ",dynperm");
-		if (m->mnt_sb->s_flags & MS_POSIXACL)
-			seq_printf(s, ",acl");
-
-		seq_printf(s, ",rsize=%d", cifs_sb->rsize);
-		seq_printf(s, ",wsize=%d", cifs_sb->wsize);
-	}
+	if (tcon->seal)
+		seq_printf(s, ",seal");
+	if (tcon->nocase)
+		seq_printf(s, ",nocase");
+	if (tcon->retry)
+		seq_printf(s, ",hard");
+	if (cifs_sb->prepath)
+		seq_printf(s, ",prepath=%s", cifs_sb->prepath);
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS)
+		seq_printf(s, ",posixpaths");
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID)
+		seq_printf(s, ",setuids");
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)
+		seq_printf(s, ",serverino");
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO)
+		seq_printf(s, ",directio");
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_XATTR)
+		seq_printf(s, ",nouser_xattr");
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR)
+		seq_printf(s, ",mapchars");
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL)
+		seq_printf(s, ",sfu");
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL)
+		seq_printf(s, ",nobrl");
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL)
+		seq_printf(s, ",cifsacl");
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DYNPERM)
+		seq_printf(s, ",dynperm");
+	if (m->mnt_sb->s_flags & MS_POSIXACL)
+		seq_printf(s, ",acl");
+
+	seq_printf(s, ",rsize=%d", cifs_sb->rsize);
+	seq_printf(s, ",wsize=%d", cifs_sb->wsize);
+
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 340481a36498bf3fe404bcecb2e2d6188e950bff Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Thu, 11 Jun 2009 10:27:29 -0400
Subject: cifs: have cifs_show_options show forceuid/forcegid options

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/cifsfs.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 3f121fbd6c56..8b315708cb3f 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -357,7 +357,12 @@ cifs_show_options(struct seq_file *s, struct vfsmount *m)
 	cifs_show_address(s, tcon->ses->server);
 
 	seq_printf(s, ",uid=%d", cifs_sb->mnt_uid);
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID)
+		seq_printf(s, ",forceuid");
+
 	seq_printf(s, ",gid=%d", cifs_sb->mnt_gid);
+	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID)
+		seq_printf(s, ",forcegid");
 
 	server = tcon->ses->server;
 	seq_printf(s, ",addr=");
-- 
cgit v1.2.3-59-g8ed1b


From 1e68b2b2756fc3488ecbade5ad5f13302b3aaafc Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Thu, 11 Jun 2009 10:27:30 -0400
Subject: cifs: add new routine for converting AF_INET and AF_INET6 addrs

...to consolidate some logic used in more than one place.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/cifsproto.h   |  2 +-
 fs/cifs/connect.c     | 21 ++++-----------------
 fs/cifs/dns_resolve.c | 21 +++------------------
 fs/cifs/netmisc.c     | 34 ++++++++++++++++++++++++++++++----
 4 files changed, 38 insertions(+), 40 deletions(-)

diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h
index f9452329bcce..c419416a42ee 100644
--- a/fs/cifs/cifsproto.h
+++ b/fs/cifs/cifsproto.h
@@ -74,7 +74,7 @@ extern unsigned int smbCalcSize(struct smb_hdr *ptr);
 extern unsigned int smbCalcSize_LE(struct smb_hdr *ptr);
 extern int decode_negTokenInit(unsigned char *security_blob, int length,
 			enum securityEnum *secType);
-extern int cifs_inet_pton(const int, const char *source, void *dst);
+extern int cifs_convert_address(char *src, void *dst);
 extern int map_smb_to_linux_error(struct smb_hdr *smb, int logErr);
 extern void header_assemble(struct smb_hdr *, char /* command */ ,
 			    const struct cifsTconInfo *, int /* length of
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 97f4311b9a8e..c368ad658236 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -1433,28 +1433,15 @@ cifs_get_tcp_session(struct smb_vol *volume_info)
 
 	memset(&addr, 0, sizeof(struct sockaddr_storage));
 
-	if (volume_info->UNCip && volume_info->UNC) {
-		rc = cifs_inet_pton(AF_INET, volume_info->UNCip,
-				    &sin_server->sin_addr.s_addr);
-
-		if (rc <= 0) {
-			/* not ipv4 address, try ipv6 */
-			rc = cifs_inet_pton(AF_INET6, volume_info->UNCip,
-					    &sin_server6->sin6_addr.in6_u);
-			if (rc > 0)
-				addr.ss_family = AF_INET6;
-		} else {
-			addr.ss_family = AF_INET;
-		}
+	cFYI(1, ("UNC: %s ip: %s", volume_info->UNC, volume_info->UNCip));
 
-		if (rc <= 0) {
+	if (volume_info->UNCip && volume_info->UNC) {
+		rc = cifs_convert_address(volume_info->UNCip, &addr);
+		if (!rc) {
 			/* we failed translating address */
 			rc = -EINVAL;
 			goto out_err;
 		}
-
-		cFYI(1, ("UNC: %s ip: %s", volume_info->UNC,
-			 volume_info->UNCip));
 	} else if (volume_info->UNCip) {
 		/* BB using ip addr as tcp_ses name to connect to the
 		   DFS root below */
diff --git a/fs/cifs/dns_resolve.c b/fs/cifs/dns_resolve.c
index df4a306f697e..91b5500755bf 100644
--- a/fs/cifs/dns_resolve.c
+++ b/fs/cifs/dns_resolve.c
@@ -37,24 +37,9 @@
 static int
 is_ip(const char *name)
 {
-	int rc;
-	struct sockaddr_in sin_server;
-	struct sockaddr_in6 sin_server6;
-
-	rc = cifs_inet_pton(AF_INET, name,
-			&sin_server.sin_addr.s_addr);
-
-	if (rc <= 0) {
-		/* not ipv4 address, try ipv6 */
-		rc = cifs_inet_pton(AF_INET6, name,
-				&sin_server6.sin6_addr.in6_u);
-		if (rc > 0)
-			return 1;
-	} else {
-		return 1;
-	}
-	/* we failed translating address */
-	return 0;
+	struct sockaddr_storage ss;
+
+	return cifs_convert_address(name, &ss);
 }
 
 static int
diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c
index 32d6baa0a54f..00e6e357ae88 100644
--- a/fs/cifs/netmisc.c
+++ b/fs/cifs/netmisc.c
@@ -133,10 +133,12 @@ static const struct smb_to_posix_error mapping_table_ERRHRD[] = {
 	{0, 0}
 };
 
-/* Convert string containing dotted ip address to binary form */
-/* returns 0 if invalid address */
-
-int
+/*
+ * Convert a string containing text IPv4 or IPv6 address to binary form.
+ *
+ * Returns 0 on failure.
+ */
+static int
 cifs_inet_pton(const int address_family, const char *cp, void *dst)
 {
 	int ret = 0;
@@ -153,6 +155,30 @@ cifs_inet_pton(const int address_family, const char *cp, void *dst)
 	return ret;
 }
 
+/*
+ * Try to convert a string to an IPv4 address and then attempt to convert
+ * it to an IPv6 address if that fails. Set the family field if either
+ * succeeds.
+ *
+ * Returns 0 on failure.
+ */
+int
+cifs_convert_address(char *src, void *dst)
+{
+	struct sockaddr_in *s4 = (struct sockaddr_in *) dst;
+	struct sockaddr_in6 *s6 = (Struct sockaddr_in6 *) dst;
+
+	if (cifs_inet_pton(AF_INET, src, &s4->sin_addr.s_addr)) {
+		s4->sin_family = AF_INET;
+		return 1;
+	} else if (cifs_inet_pton(AF_INET6, src, &s6->sin6_addr.s6_addr)) {
+		s6->sin6_family = AF_INET6;
+		return 1;
+	}
+
+	return 0;
+}
+
 /*****************************************************************************
 convert a NT status code to a dos class/code
  *****************************************************************************/
-- 
cgit v1.2.3-59-g8ed1b


From 3a45c9e4b192c9e96f65a8473c7a28e09cf2ac2a Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sat, 13 Jun 2009 10:43:19 +0100
Subject: MAINTAINERS: Update file list for ARM/S3C2410 and ARM/S3C2440

Add F: entries for ARM/S3C2410 and ARM/S3C2440 to update the
entries.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 MAINTAINERS | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 90f81283b722..fa9377df160d 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -830,6 +830,7 @@ M:	ben-linux@fluff.org
 L:	linux-arm-kernel@lists.arm.linux.org.uk	(subscribers-only)
 W:	http://www.fluff.org/ben/linux/
 S:	Maintained
+F:	arch/arm/mach-s3c2410/
 
 ARM/S3C2440 ARM ARCHITECTURE
 P:	Ben Dooks
@@ -837,6 +838,7 @@ M:	ben-linux@fluff.org
 L:	linux-arm-kernel@lists.arm.linux.org.uk	(subscribers-only)
 W:	http://www.fluff.org/ben/linux/
 S:	Maintained
+F:	arch/arm/mach-s3c2440/
 
 ARM/TECHNOLOGIC SYSTEMS TS7250 MACHINE SUPPORT
 P:	Lennert Buytenhek
-- 
cgit v1.2.3-59-g8ed1b


From b21477f9d257cd8d45f7df32a9ebed45dbdfa4b5 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Sat, 13 Jun 2009 10:46:34 +0100
Subject: MAINTAINERS: Add ARM S3C2442, S3C2443, S3C6400, S3C6410 and
 ARM/SAMSUNG

Add entries for the ARM architectures and platform support that are
currently being maintained by myself.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 MAINTAINERS | 41 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 41 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index fa9377df160d..293cba0dc121 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -824,6 +824,15 @@ M:	alex@shark-linux.de
 W:	http://www.shark-linux.de/shark.html
 S:	Maintained
 
+ARM/SAMSUNG ARM ARCHITECTURES
+P:	Ben Dooks
+M:	ben-linux@fluff.org
+L:	linux-arm-kernel@lists.arm.linux.org.uk	(subscribers-only)
+W:	http://www.fluff.org/ben/linux/
+S:	Maintained
+F:	arch/arm/plat-s3c/
+F:	arch/arm/plat-s3c24xx/
+
 ARM/S3C2410 ARM ARCHITECTURE
 P:	Ben Dooks
 M:	ben-linux@fluff.org
@@ -840,6 +849,38 @@ W:	http://www.fluff.org/ben/linux/
 S:	Maintained
 F:	arch/arm/mach-s3c2440/
 
+ARM/S3C2442 ARM ARCHITECTURE
+P:	Ben Dooks
+M:	ben-linux@fluff.org
+L:	linux-arm-kernel@lists.arm.linux.org.uk	(subscribers-only)
+W:	http://www.fluff.org/ben/linux/
+S:	Maintained
+F:	arch/arm/mach-s3c2442/
+
+ARM/S3C2443 ARM ARCHITECTURE
+P:	Ben Dooks
+M:	ben-linux@fluff.org
+L:	linux-arm-kernel@lists.arm.linux.org.uk	(subscribers-only)
+W:	http://www.fluff.org/ben/linux/
+S:	Maintained
+F:	arch/arm/mach-s3c2443/
+
+ARM/S3C6400 ARM ARCHITECTURE
+P:	Ben Dooks
+M:	ben-linux@fluff.org
+L:	linux-arm-kernel@lists.arm.linux.org.uk	(subscribers-only)
+W:	http://www.fluff.org/ben/linux/
+S:	Maintained
+F:	arch/arm/mach-s3c6400/
+
+ARM/S3C6410 ARM ARCHITECTURE
+P:	Ben Dooks
+M:	ben-linux@fluff.org
+L:	linux-arm-kernel@lists.arm.linux.org.uk	(subscribers-only)
+W:	http://www.fluff.org/ben/linux/
+S:	Maintained
+F:	arch/arm/mach-s3c6410/
+
 ARM/TECHNOLOGIC SYSTEMS TS7250 MACHINE SUPPORT
 P:	Lennert Buytenhek
 M:	kernel@wantstofly.org
-- 
cgit v1.2.3-59-g8ed1b


From c76acec6d55107b652a37c90b36c00bc8b04dabb Mon Sep 17 00:00:00 2001
From: Jay Fenlason <fenlason@redhat.com>
Date: Mon, 18 May 2009 13:08:06 -0400
Subject: firewire: add IPv4 support

Implement IPv4 over IEEE 1394 as per RFC 2734 for the newer firewire
stack.  This feature has only been present in the older ieee1394 stack
via the eth1394 driver.

Still to do:
  - fix ipv4_priv and ipv4_node lifetime logic
  - fix determination of speeds and max payloads
  - fix bus reset handling
  - fix unaligned memory accesses
  - fix coding style
  - further testing/ improvement of fragment reassembly
  - perhaps multicast support

Signed-off-by: Jay Fenlason <fenlason@redhat.com>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de> (rebased, copyright note, changelog)
---
 drivers/firewire/Makefile    |    2 +
 drivers/firewire/core-card.c |    4 +
 drivers/firewire/core-iso.c  |    7 +
 drivers/firewire/core.h      |   87 --
 drivers/firewire/fw-ipv4.c   | 1819 ++++++++++++++++++++++++++++++++++++++++++
 include/linux/firewire.h     |   94 +++
 6 files changed, 1926 insertions(+), 87 deletions(-)
 create mode 100644 drivers/firewire/fw-ipv4.c

diff --git a/drivers/firewire/Makefile b/drivers/firewire/Makefile
index bc3b9bf822bf..31edf30c558d 100644
--- a/drivers/firewire/Makefile
+++ b/drivers/firewire/Makefile
@@ -6,7 +6,9 @@ firewire-core-y += core-card.o core-cdev.o core-device.o \
                    core-iso.o core-topology.o core-transaction.o
 firewire-ohci-y += ohci.o
 firewire-sbp2-y += sbp2.o
+firewire-ipv4-y += fw-ipv4.o
 
 obj-$(CONFIG_FIREWIRE) += firewire-core.o
 obj-$(CONFIG_FIREWIRE_OHCI) += firewire-ohci.o
 obj-$(CONFIG_FIREWIRE_SBP2) += firewire-sbp2.o
+obj-$(CONFIG_FIREWIRE_IPV4) += firewire-ipv4.o
diff --git a/drivers/firewire/core-card.c b/drivers/firewire/core-card.c
index 4c1be64fdddd..cdab32b20675 100644
--- a/drivers/firewire/core-card.c
+++ b/drivers/firewire/core-card.c
@@ -176,6 +176,7 @@ int fw_core_add_descriptor(struct fw_descriptor *desc)
 
 	return 0;
 }
+EXPORT_SYMBOL(fw_core_add_descriptor);
 
 void fw_core_remove_descriptor(struct fw_descriptor *desc)
 {
@@ -189,6 +190,7 @@ void fw_core_remove_descriptor(struct fw_descriptor *desc)
 
 	mutex_unlock(&card_mutex);
 }
+EXPORT_SYMBOL(fw_core_remove_descriptor);
 
 static void allocate_broadcast_channel(struct fw_card *card, int generation)
 {
@@ -427,6 +429,8 @@ void fw_card_initialize(struct fw_card *card,
 	card->local_node = NULL;
 
 	INIT_DELAYED_WORK(&card->work, fw_card_bm_work);
+	card->netdev = NULL;
+	INIT_LIST_HEAD(&card->ipv4_nodes);
 }
 EXPORT_SYMBOL(fw_card_initialize);
 
diff --git a/drivers/firewire/core-iso.c b/drivers/firewire/core-iso.c
index 28076c892d7e..448ddd7d887b 100644
--- a/drivers/firewire/core-iso.c
+++ b/drivers/firewire/core-iso.c
@@ -80,6 +80,7 @@ int fw_iso_buffer_init(struct fw_iso_buffer *buffer, struct fw_card *card,
 
 	return -ENOMEM;
 }
+EXPORT_SYMBOL(fw_iso_buffer_init);
 
 int fw_iso_buffer_map(struct fw_iso_buffer *buffer, struct vm_area_struct *vma)
 {
@@ -114,6 +115,7 @@ void fw_iso_buffer_destroy(struct fw_iso_buffer *buffer,
 	kfree(buffer->pages);
 	buffer->pages = NULL;
 }
+EXPORT_SYMBOL(fw_iso_buffer_destroy);
 
 struct fw_iso_context *fw_iso_context_create(struct fw_card *card,
 		int type, int channel, int speed, size_t header_size,
@@ -136,6 +138,7 @@ struct fw_iso_context *fw_iso_context_create(struct fw_card *card,
 
 	return ctx;
 }
+EXPORT_SYMBOL(fw_iso_context_create);
 
 void fw_iso_context_destroy(struct fw_iso_context *ctx)
 {
@@ -143,12 +146,14 @@ void fw_iso_context_destroy(struct fw_iso_context *ctx)
 
 	card->driver->free_iso_context(ctx);
 }
+EXPORT_SYMBOL(fw_iso_context_destroy);
 
 int fw_iso_context_start(struct fw_iso_context *ctx,
 			 int cycle, int sync, int tags)
 {
 	return ctx->card->driver->start_iso(ctx, cycle, sync, tags);
 }
+EXPORT_SYMBOL(fw_iso_context_start);
 
 int fw_iso_context_queue(struct fw_iso_context *ctx,
 			 struct fw_iso_packet *packet,
@@ -159,11 +164,13 @@ int fw_iso_context_queue(struct fw_iso_context *ctx,
 
 	return card->driver->queue_iso(ctx, packet, buffer, payload);
 }
+EXPORT_SYMBOL(fw_iso_context_queue);
 
 int fw_iso_context_stop(struct fw_iso_context *ctx)
 {
 	return ctx->card->driver->stop_iso(ctx);
 }
+EXPORT_SYMBOL(fw_iso_context_stop);
 
 /*
  * Isochronous bus resource management (channels, bandwidth), client side
diff --git a/drivers/firewire/core.h b/drivers/firewire/core.h
index 0a25a7b38a80..c3cfc647e5e3 100644
--- a/drivers/firewire/core.h
+++ b/drivers/firewire/core.h
@@ -1,7 +1,6 @@
 #ifndef _FIREWIRE_CORE_H
 #define _FIREWIRE_CORE_H
 
-#include <linux/dma-mapping.h>
 #include <linux/fs.h>
 #include <linux/list.h>
 #include <linux/idr.h>
@@ -97,17 +96,6 @@ int fw_core_initiate_bus_reset(struct fw_card *card, int short_reset);
 int fw_compute_block_crc(u32 *block);
 void fw_schedule_bm_work(struct fw_card *card, unsigned long delay);
 
-struct fw_descriptor {
-	struct list_head link;
-	size_t length;
-	u32 immediate;
-	u32 key;
-	const u32 *data;
-};
-
-int fw_core_add_descriptor(struct fw_descriptor *desc);
-void fw_core_remove_descriptor(struct fw_descriptor *desc);
-
 
 /* -cdev */
 
@@ -130,77 +118,7 @@ void fw_node_event(struct fw_card *card, struct fw_node *node, int event);
 
 /* -iso */
 
-/*
- * The iso packet format allows for an immediate header/payload part
- * stored in 'header' immediately after the packet info plus an
- * indirect payload part that is pointer to by the 'payload' field.
- * Applications can use one or the other or both to implement simple
- * low-bandwidth streaming (e.g. audio) or more advanced
- * scatter-gather streaming (e.g. assembling video frame automatically).
- */
-struct fw_iso_packet {
-	u16 payload_length;	/* Length of indirect payload. */
-	u32 interrupt:1;	/* Generate interrupt on this packet */
-	u32 skip:1;		/* Set to not send packet at all. */
-	u32 tag:2;
-	u32 sy:4;
-	u32 header_length:8;	/* Length of immediate header. */
-	u32 header[0];
-};
-
-#define FW_ISO_CONTEXT_TRANSMIT	0
-#define FW_ISO_CONTEXT_RECEIVE	1
-
-#define FW_ISO_CONTEXT_MATCH_TAG0	 1
-#define FW_ISO_CONTEXT_MATCH_TAG1	 2
-#define FW_ISO_CONTEXT_MATCH_TAG2	 4
-#define FW_ISO_CONTEXT_MATCH_TAG3	 8
-#define FW_ISO_CONTEXT_MATCH_ALL_TAGS	15
-
-/*
- * An iso buffer is just a set of pages mapped for DMA in the
- * specified direction.  Since the pages are to be used for DMA, they
- * are not mapped into the kernel virtual address space.  We store the
- * DMA address in the page private. The helper function
- * fw_iso_buffer_map() will map the pages into a given vma.
- */
-struct fw_iso_buffer {
-	enum dma_data_direction direction;
-	struct page **pages;
-	int page_count;
-};
-
-typedef void (*fw_iso_callback_t)(struct fw_iso_context *context,
-				  u32 cycle, size_t header_length,
-				  void *header, void *data);
-
-struct fw_iso_context {
-	struct fw_card *card;
-	int type;
-	int channel;
-	int speed;
-	size_t header_size;
-	fw_iso_callback_t callback;
-	void *callback_data;
-};
-
-int fw_iso_buffer_init(struct fw_iso_buffer *buffer, struct fw_card *card,
-		       int page_count, enum dma_data_direction direction);
 int fw_iso_buffer_map(struct fw_iso_buffer *buffer, struct vm_area_struct *vma);
-void fw_iso_buffer_destroy(struct fw_iso_buffer *buffer, struct fw_card *card);
-
-struct fw_iso_context *fw_iso_context_create(struct fw_card *card,
-		int type, int channel, int speed, size_t header_size,
-		fw_iso_callback_t callback, void *callback_data);
-int fw_iso_context_queue(struct fw_iso_context *ctx,
-			 struct fw_iso_packet *packet,
-			 struct fw_iso_buffer *buffer,
-			 unsigned long payload);
-int fw_iso_context_start(struct fw_iso_context *ctx,
-			 int cycle, int sync, int tags);
-int fw_iso_context_stop(struct fw_iso_context *ctx);
-void fw_iso_context_destroy(struct fw_iso_context *ctx);
-
 void fw_iso_resource_manage(struct fw_card *card, int generation,
 		u64 channels_mask, int *channel, int *bandwidth, bool allocate);
 
@@ -285,9 +203,4 @@ void fw_flush_transactions(struct fw_card *card);
 void fw_send_phy_config(struct fw_card *card,
 			int node_id, int generation, int gap_count);
 
-static inline int fw_stream_packet_destination_id(int tag, int channel, int sy)
-{
-	return tag << 14 | channel << 8 | sy;
-}
-
 #endif /* _FIREWIRE_CORE_H */
diff --git a/drivers/firewire/fw-ipv4.c b/drivers/firewire/fw-ipv4.c
new file mode 100644
index 000000000000..4de6dbb95f0c
--- /dev/null
+++ b/drivers/firewire/fw-ipv4.c
@@ -0,0 +1,1819 @@
+/*
+ * IPv4 over IEEE 1394, per RFC 2734
+ *
+ * Copyright (C) 2009 Jay Fenlason <fenlason@redhat.com>
+ *
+ * based on eth1394 by Ben Collins et al
+ */
+
+#include <linux/device.h>
+#include <linux/ethtool.h>
+#include <linux/firewire.h>
+#include <linux/firewire-constants.h>
+#include <linux/highmem.h>
+#include <linux/in.h>
+#include <linux/ip.h>
+#include <linux/mod_devicetable.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/netdevice.h>
+#include <linux/skbuff.h>
+
+#include <asm/unaligned.h>
+#include <net/arp.h>
+
+/* Things to potentially make runtime cofigurable */
+/* must be at least as large as our maximum receive size */
+#define FIFO_SIZE 4096
+/* Network timeout in glibbles */
+#define IPV4_TIMEOUT       100000
+
+/* Runitme configurable paramaters */
+static int ipv4_mpd = 25;
+static int ipv4_max_xmt = 0;
+/* 16k for receiving arp and broadcast packets.  Enough? */
+static int ipv4_iso_page_count = 4;
+
+MODULE_AUTHOR("Jay Fenlason (fenlason@redhat.com)");
+MODULE_DESCRIPTION("Firewire IPv4 Driver (IPv4-over-IEEE1394 as per RFC 2734)");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(ieee1394, ipv4_id_table);
+module_param_named(max_partial_datagrams, ipv4_mpd, int, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(max_partial_datagrams, "Maximum number of received"
+ " incomplete fragmented datagrams (default = 25).");
+
+/* Max xmt is useful for forcing fragmentation, which makes testing easier. */
+module_param_named(max_transmit, ipv4_max_xmt, int, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(max_transmit, "Maximum datagram size to transmit"
+ " (larger datagrams will be fragmented) (default = 0 (use hardware defaults).");
+
+/* iso page count controls how many pages will be used for receiving broadcast packets. */
+module_param_named(iso_pages, ipv4_iso_page_count, int, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(iso_pages, "Number of pages to use for receiving broadcast packets"
+ " (default = 4).");
+
+/* uncomment this line to do debugging */
+#define fw_debug(s, args...) printk(KERN_DEBUG KBUILD_MODNAME ": " s, ## args)
+
+/* comment out these lines to do debugging. */
+/* #undef fw_debug */
+/* #define fw_debug(s...) */
+/* #define print_hex_dump(l...) */
+
+/* Define a fake hardware header format for the networking core.  Note that
+ * header size cannot exceed 16 bytes as that is the size of the header cache.
+ * Also, we do not need the source address in the header so we omit it and
+ * keep the header to under 16 bytes */
+#define IPV4_ALEN (8)
+/* This must equal sizeof(struct ipv4_ether_hdr) */
+#define IPV4_HLEN (10)
+
+/* FIXME: what's a good size for this? */
+#define INVALID_FIFO_ADDR (u64)~0ULL
+
+/* Things specified by standards */
+#define BROADCAST_CHANNEL 31
+
+#define S100_BUFFER_SIZE 512
+#define MAX_BUFFER_SIZE 4096
+
+#define IPV4_GASP_SPECIFIER_ID	0x00005EU
+#define IPV4_GASP_VERSION	0x00000001U
+
+#define IPV4_GASP_OVERHEAD (2 * sizeof(u32)) /* for GASP header */
+
+#define IPV4_UNFRAG_HDR_SIZE	sizeof(u32)
+#define IPV4_FRAG_HDR_SIZE	(2 * sizeof(u32))
+#define IPV4_FRAG_OVERHEAD	sizeof(u32)
+
+#define ALL_NODES (0xffc0 | 0x003f)
+
+#define IPV4_HDR_UNFRAG		0	/* unfragmented		*/
+#define IPV4_HDR_FIRSTFRAG	1	/* first fragment	*/
+#define IPV4_HDR_LASTFRAG	2	/* last fragment	*/
+#define IPV4_HDR_INTFRAG	3	/* interior fragment	*/
+
+/* Our arp packet (ARPHRD_IEEE1394) */
+/* FIXME: note that this is probably bogus on weird-endian machines */
+struct ipv4_arp {
+	u16 hw_type;		/* 0x0018	*/
+	u16 proto_type;		/* 0x0806       */
+	u8 hw_addr_len;		/* 16		*/
+	u8 ip_addr_len;         /* 4		*/
+	u16 opcode;	        /* ARP Opcode	*/
+	/* Above is exactly the same format as struct arphdr */
+
+	u64 s_uniq_id;		/* Sender's 64bit EUI			*/
+	u8 max_rec;             /* Sender's max packet size		*/
+	u8 sspd;		/* Sender's max speed			*/
+	u16 fifo_hi;            /* hi 16bits of sender's FIFO addr	*/
+	u32 fifo_lo;            /* lo 32bits of sender's FIFO addr	*/
+	u32 sip;		/* Sender's IP Address			*/
+	u32 tip;		/* IP Address of requested hw addr	*/
+} __attribute__((packed));
+
+struct ipv4_ether_hdr {
+	unsigned char	h_dest[IPV4_ALEN];	/* destination address */
+	unsigned short  h_proto;                /* packet type ID field */
+}  __attribute__((packed));
+
+static inline struct ipv4_ether_hdr *ipv4_ether_hdr(const struct sk_buff *skb)
+{
+	return (struct ipv4_ether_hdr *)skb_mac_header(skb);
+}
+
+enum ipv4_tx_type {
+	IPV4_UNKNOWN = 0,
+	IPV4_GASP = 1,
+	IPV4_WRREQ = 2,
+};
+
+enum ipv4_broadcast_state {
+	IPV4_BROADCAST_ERROR,
+	IPV4_BROADCAST_RUNNING,
+	IPV4_BROADCAST_STOPPED,
+};
+
+#define ipv4_get_hdr_lf(h)		(((h)->w0&0xC0000000)>>30)
+#define ipv4_get_hdr_ether_type(h)	(((h)->w0&0x0000FFFF)    )
+#define ipv4_get_hdr_dg_size(h)		(((h)->w0&0x0FFF0000)>>16)
+#define ipv4_get_hdr_fg_off(h)		(((h)->w0&0x00000FFF)    )
+#define ipv4_get_hdr_dgl(h)		(((h)->w1&0xFFFF0000)>>16)
+
+#define ipv4_set_hdr_lf(lf)		(( lf)<<30)
+#define ipv4_set_hdr_ether_type(et)	(( et)    )
+#define ipv4_set_hdr_dg_size(dgs)	((dgs)<<16)
+#define ipv4_set_hdr_fg_off(fgo)	((fgo)    )
+
+#define ipv4_set_hdr_dgl(dgl)		((dgl)<<16)
+
+struct ipv4_hdr {
+	u32 w0;
+	u32 w1;
+};
+
+static inline void ipv4_make_uf_hdr( struct ipv4_hdr *hdr, unsigned ether_type) {
+	hdr->w0 = ipv4_set_hdr_lf(IPV4_HDR_UNFRAG)
+		   |ipv4_set_hdr_ether_type(ether_type);
+	fw_debug ( "Setting unfragmented header %p to %x\n", hdr, hdr->w0 );
+}
+
+static inline void ipv4_make_ff_hdr ( struct ipv4_hdr *hdr, unsigned ether_type, unsigned dg_size, unsigned dgl ) {
+	hdr->w0 = ipv4_set_hdr_lf(IPV4_HDR_FIRSTFRAG)
+		   |ipv4_set_hdr_dg_size(dg_size)
+		   |ipv4_set_hdr_ether_type(ether_type);
+	hdr->w1 = ipv4_set_hdr_dgl(dgl);
+	fw_debug ( "Setting fragmented header %p to first_frag %x,%x (et %x, dgs %x, dgl %x)\n", hdr, hdr->w0, hdr->w1,
+ ether_type, dg_size, dgl );
+}
+
+static inline void ipv4_make_sf_hdr ( struct ipv4_hdr *hdr, unsigned lf, unsigned dg_size, unsigned fg_off, unsigned dgl) {
+	hdr->w0 = ipv4_set_hdr_lf(lf)
+		 |ipv4_set_hdr_dg_size(dg_size)
+		 |ipv4_set_hdr_fg_off(fg_off);
+	hdr->w1 = ipv4_set_hdr_dgl(dgl);
+	fw_debug ( "Setting fragmented header %p to %x,%x (lf %x, dgs %x, fo %x dgl %x)\n",
+ hdr, hdr->w0, hdr->w1,
+ lf, dg_size, fg_off, dgl );
+}
+
+/* End of IP1394 headers */
+
+/* Fragment types */
+#define ETH1394_HDR_LF_UF	0	/* unfragmented		*/
+#define ETH1394_HDR_LF_FF	1	/* first fragment	*/
+#define ETH1394_HDR_LF_LF	2	/* last fragment	*/
+#define ETH1394_HDR_LF_IF	3	/* interior fragment	*/
+
+#define IP1394_HW_ADDR_LEN	16	/* As per RFC		*/
+
+/* This list keeps track of what parts of the datagram have been filled in */
+struct ipv4_fragment_info {
+        struct list_head fragment_info;
+	u16 offset;
+	u16 len;
+};
+
+struct ipv4_partial_datagram {
+	struct list_head pdg_list;
+	struct list_head fragment_info;
+	struct sk_buff *skb;
+	/* FIXME Why not use skb->data? */
+	char *pbuf;
+	u16 datagram_label;
+	u16 ether_type;
+	u16 datagram_size;
+};
+
+/*
+ * We keep one of these for each IPv4 capable device attached to a fw_card.
+ * The list of them is stored in the fw_card structure rather than in the
+ * ipv4_priv because the remote IPv4 nodes may be probed before the card is,
+ * so we need a place to store them before the ipv4_priv structure is
+ * allocated.
+ */
+struct ipv4_node {
+	struct list_head ipv4_nodes;
+	/* guid of the remote node */
+	u64 guid;
+	/* FIFO address to transmit datagrams to, or INVALID_FIFO_ADDR */
+	u64 fifo;
+
+	spinlock_t pdg_lock;	/* partial datagram lock		*/
+	/* List of partial datagrams received from this node */
+	struct list_head pdg_list;
+	/* Number of entries in pdg_list at the moment */
+	unsigned pdg_size;
+
+	/* max payload to transmit to this remote node */
+	/* This already includes the IPV4_FRAG_HDR_SIZE overhead */
+	u16 max_payload;
+	/* outgoing datagram label */
+	u16 datagram_label;
+	/* Current node_id of the remote node */
+	u16 nodeid;
+	/* current generation of the remote node */
+	u8 generation;
+	/* max speed that this node can receive at */
+	u8 xmt_speed;
+};
+
+struct ipv4_priv {
+	spinlock_t lock;
+
+	enum ipv4_broadcast_state broadcast_state;
+	struct fw_iso_context *broadcast_rcv_context;
+	struct fw_iso_buffer broadcast_rcv_buffer;
+	void **broadcast_rcv_buffer_ptrs;
+	unsigned broadcast_rcv_next_ptr;
+	unsigned num_broadcast_rcv_ptrs;
+	unsigned rcv_buffer_size;
+	/*
+	 * This value is the maximum unfragmented datagram size that can be
+	 * sent by the hardware.  It already has the GASP overhead and the
+	 * unfragmented datagram header overhead calculated into it.
+	 */
+	unsigned broadcast_xmt_max_payload;
+	u16 broadcast_xmt_datagramlabel;
+
+	/*
+	 * The csr address that remote nodes must send datagrams to for us to
+	 * receive them.
+	 */
+	struct fw_address_handler handler;
+	u64 local_fifo;
+
+	/* Wake up to xmt	 */
+        /* struct work_struct wake;*/
+	/* List of packets to be sent */
+	struct list_head packet_list;
+	/*
+	 * List of packets that were broadcasted.  When we get an ISO interrupt
+	 * one of them has been sent
+	 */
+	struct list_head broadcasted_list;
+	/* List of packets that have been sent but not yet acked */
+	struct list_head sent_list;
+
+	struct fw_card *card;
+};
+
+/* This is our task struct. It's used for the packet complete callback.  */
+struct ipv4_packet_task {
+	/*
+	 * ptask can actually be on priv->packet_list, priv->broadcasted_list,
+	 * or priv->sent_list depending on its current state.
+	 */
+	struct list_head packet_list;
+	struct fw_transaction transaction;
+	struct ipv4_hdr hdr;
+	struct sk_buff *skb;
+	struct ipv4_priv *priv;
+	enum ipv4_tx_type tx_type;
+	int outstanding_pkts;
+	unsigned max_payload;
+	u64 fifo_addr;
+	u16 dest_node;
+	u8 generation;
+	u8 speed;
+};
+
+static struct kmem_cache *ipv4_packet_task_cache;
+
+static const char ipv4_driver_name[] = "firewire-ipv4";
+
+static const struct ieee1394_device_id ipv4_id_table[] = {
+	{
+		.match_flags  = IEEE1394_MATCH_SPECIFIER_ID |
+				IEEE1394_MATCH_VERSION,
+		.specifier_id = IPV4_GASP_SPECIFIER_ID,
+		.version      = IPV4_GASP_VERSION,
+	},
+	{ }
+};
+
+static u32 ipv4_unit_directory_data[] = {
+	0x00040000,					/* unit directory */
+	0x12000000 | IPV4_GASP_SPECIFIER_ID,	/* specifier ID */
+	0x81000003,					/* text descriptor */
+	0x13000000 | IPV4_GASP_VERSION,		/* version */
+	0x81000005,					/* text descriptor */
+
+	0x00030000,					/* Three quadlets */
+	0x00000000,					/* Text */
+	0x00000000,					/* Language 0 */
+	0x49414e41,					/* I A N A */
+	0x00030000,					/* Three quadlets */
+	0x00000000,					/* Text */
+	0x00000000,					/* Language 0 */
+	0x49507634,					/* I P v 4 */
+};
+
+static struct fw_descriptor ipv4_unit_directory = {
+	.length = ARRAY_SIZE(ipv4_unit_directory_data),
+	.key = 0xd1000000,
+	.data = ipv4_unit_directory_data
+};
+
+static int ipv4_send_packet(struct ipv4_packet_task *ptask );
+
+/* ------------------------------------------------------------------ */
+/******************************************
+ * HW Header net device functions
+ ******************************************/
+  /* These functions have been adapted from net/ethernet/eth.c */
+
+/* Create a fake MAC header for an arbitrary protocol layer.
+ * saddr=NULL means use device source address
+ * daddr=NULL means leave destination address (eg unresolved arp). */
+
+static int ipv4_header ( struct sk_buff *skb, struct net_device *dev,
+		       unsigned short type, const void *daddr,
+		       const void *saddr, unsigned len) {
+	struct ipv4_ether_hdr *eth;
+
+	eth = (struct ipv4_ether_hdr *)skb_push(skb, sizeof(*eth));
+	eth->h_proto = htons(type);
+
+	if (dev->flags & (IFF_LOOPBACK | IFF_NOARP)) {
+		memset(eth->h_dest, 0, dev->addr_len);
+		return dev->hard_header_len;
+	}
+
+	if (daddr) {
+		memcpy(eth->h_dest, daddr, dev->addr_len);
+		return dev->hard_header_len;
+	}
+
+	return -dev->hard_header_len;
+}
+
+/* Rebuild the faked MAC header. This is called after an ARP
+ * (or in future other address resolution) has completed on this
+ * sk_buff. We now let ARP fill in the other fields.
+ *
+ * This routine CANNOT use cached dst->neigh!
+ * Really, it is used only when dst->neigh is wrong.
+ */
+
+static int ipv4_rebuild_header(struct sk_buff *skb)
+{
+	struct ipv4_ether_hdr *eth;
+
+	eth = (struct ipv4_ether_hdr *)skb->data;
+	if (eth->h_proto == htons(ETH_P_IP))
+		return arp_find((unsigned char *)&eth->h_dest, skb);
+
+	fw_notify ( "%s: unable to resolve type %04x addresses\n",
+		   skb->dev->name,ntohs(eth->h_proto) );
+	return 0;
+}
+
+static int ipv4_header_cache(const struct neighbour *neigh, struct hh_cache *hh) {
+	unsigned short type = hh->hh_type;
+	struct net_device *dev;
+	struct ipv4_ether_hdr *eth;
+
+	if (type == htons(ETH_P_802_3))
+		return -1;
+	dev = neigh->dev;
+	eth = (struct ipv4_ether_hdr *)((u8 *)hh->hh_data + 16 - sizeof(*eth));
+	eth->h_proto = type;
+	memcpy(eth->h_dest, neigh->ha, dev->addr_len);
+
+	hh->hh_len = IPV4_HLEN;
+	return 0;
+}
+
+/* Called by Address Resolution module to notify changes in address. */
+static void ipv4_header_cache_update(struct hh_cache *hh, const struct net_device *dev, const unsigned char * haddr ) {
+	memcpy((u8 *)hh->hh_data + 16 - IPV4_HLEN, haddr, dev->addr_len);
+}
+
+static int ipv4_header_parse(const struct sk_buff *skb, unsigned char *haddr) {
+	memcpy(haddr, skb->dev->dev_addr, IPV4_ALEN);
+	return IPV4_ALEN;
+}
+
+static const struct header_ops ipv4_header_ops = {
+	.create         = ipv4_header,
+	.rebuild        = ipv4_rebuild_header,
+	.cache		= ipv4_header_cache,
+	.cache_update	= ipv4_header_cache_update,
+	.parse          = ipv4_header_parse,
+};
+
+/* ------------------------------------------------------------------ */
+
+/* FIXME: is this correct for all cases? */
+static bool ipv4_frag_overlap(struct ipv4_partial_datagram *pd, unsigned offset, unsigned len)
+{
+        struct ipv4_fragment_info *fi;
+	unsigned end = offset + len;
+
+	list_for_each_entry(fi, &pd->fragment_info, fragment_info) {
+		if (offset < fi->offset + fi->len && end > fi->offset) {
+			fw_debug ( "frag_overlap pd %p fi %p (%x@%x) with %x@%x\n", pd, fi, fi->len, fi->offset, len, offset );
+			return true;
+		}
+	}
+	fw_debug ( "frag_overlap %p does not overlap with %x@%x\n", pd, len, offset );
+	return false;
+}
+
+/* Assumes that new fragment does not overlap any existing fragments */
+static struct ipv4_fragment_info *ipv4_frag_new ( struct ipv4_partial_datagram *pd, unsigned offset, unsigned len ) {
+	struct ipv4_fragment_info *fi, *fi2, *new;
+	struct list_head *list;
+
+	fw_debug ( "frag_new pd %p %x@%x\n", pd, len, offset );
+	list = &pd->fragment_info;
+	list_for_each_entry(fi, &pd->fragment_info, fragment_info) {
+		if (fi->offset + fi->len == offset) {
+			/* The new fragment can be tacked on to the end */
+			/* Did the new fragment plug a hole? */
+			fi2 = list_entry(fi->fragment_info.next, struct ipv4_fragment_info, fragment_info);
+			if (fi->offset + fi->len == fi2->offset) {
+				fw_debug ( "pd %p: hole filling %p (%x@%x) and %p(%x@%x): now %x@%x\n", pd, fi, fi->len, fi->offset,
+				fi2, fi2->len, fi2->offset, fi->len + len + fi2->len, fi->offset );
+				/* glue fragments together */
+				fi->len += len + fi2->len;
+				list_del(&fi2->fragment_info);
+				kfree(fi2);
+			} else {
+				fw_debug ( "pd %p: extending %p from %x@%x to %x@%x\n", pd, fi, fi->len, fi->offset, fi->len+len, fi->offset );
+				fi->len += len;
+			}
+			return fi;
+		}
+		if (offset + len == fi->offset) {
+			/* The new fragment can be tacked on to the beginning */
+			/* Did the new fragment plug a hole? */
+			fi2 = list_entry(fi->fragment_info.prev, struct ipv4_fragment_info, fragment_info);
+			if (fi2->offset + fi2->len == fi->offset) {
+				/* glue fragments together */
+				fw_debug ( "pd %p: extending %p and merging with %p from %x@%x to %x@%x\n",
+ pd, fi2, fi, fi2->len, fi2->offset, fi2->len + fi->len + len, fi2->offset );
+				fi2->len += fi->len + len;
+				list_del(&fi->fragment_info);
+				kfree(fi);
+				return fi2;
+			}
+			fw_debug ( "pd %p: extending %p from %x@%x to %x@%x\n", pd, fi, fi->len, fi->offset, offset, fi->len + len );
+			fi->offset = offset;
+			fi->len += len;
+			return fi;
+		}
+		if (offset > fi->offset + fi->len) {
+			list = &fi->fragment_info;
+			break;
+		}
+		if (offset + len < fi->offset) {
+			list = fi->fragment_info.prev;
+			break;
+		}
+	}
+
+	new = kmalloc(sizeof(*new), GFP_ATOMIC);
+	if (!new) {
+		fw_error ( "out of memory in fragment handling!\n" );
+		return NULL;
+	}
+
+	new->offset = offset;
+	new->len = len;
+	list_add(&new->fragment_info, list);
+	fw_debug ( "pd %p: new frag %p %x@%x\n", pd, new, new->len, new->offset );
+	list_for_each_entry( fi, &pd->fragment_info, fragment_info )
+		fw_debug ( "fi %p %x@%x\n", fi, fi->len, fi->offset );
+	return new;
+}
+
+/* ------------------------------------------------------------------ */
+
+static struct ipv4_partial_datagram *ipv4_pd_new(struct net_device *netdev,
+ struct ipv4_node *node, u16 datagram_label, unsigned dg_size, u32 *frag_buf,
+ unsigned frag_off, unsigned frag_len) {
+	struct ipv4_partial_datagram *new;
+	struct ipv4_fragment_info *fi;
+
+	new = kmalloc(sizeof(*new), GFP_ATOMIC);
+	if (!new)
+		goto fail;
+	INIT_LIST_HEAD(&new->fragment_info);
+	fi = ipv4_frag_new ( new, frag_off, frag_len);
+	if ( fi == NULL )
+		goto fail_w_new;
+	new->datagram_label = datagram_label;
+	new->datagram_size = dg_size;
+	new->skb = dev_alloc_skb(dg_size + netdev->hard_header_len + 15);
+	if ( new->skb == NULL )
+		goto fail_w_fi;
+	skb_reserve(new->skb, (netdev->hard_header_len + 15) & ~15);
+	new->pbuf = skb_put(new->skb, dg_size);
+	memcpy(new->pbuf + frag_off, frag_buf, frag_len);
+	list_add_tail(&new->pdg_list, &node->pdg_list);
+	fw_debug ( "pd_new: new pd %p { dgl %u, dg_size %u, skb %p, pbuf %p } on node %p\n",
+ new, new->datagram_label, new->datagram_size, new->skb, new->pbuf, node );
+	return new;
+
+fail_w_fi:
+	kfree(fi);
+fail_w_new:
+	kfree(new);
+fail:
+	fw_error("ipv4_pd_new: no memory\n");
+	return NULL;
+}
+
+static struct ipv4_partial_datagram *ipv4_pd_find(struct ipv4_node *node, u16 datagram_label) {
+	struct ipv4_partial_datagram *pd;
+
+	list_for_each_entry(pd, &node->pdg_list, pdg_list) {
+	        if ( pd->datagram_label == datagram_label ) {
+			fw_debug ( "pd_find(node %p, label %u): pd %p\n", node, datagram_label, pd );
+			return pd;
+		}
+	}
+	fw_debug ( "pd_find(node %p, label %u) no entry\n", node, datagram_label );
+	return NULL;
+}
+
+
+static void ipv4_pd_delete ( struct ipv4_partial_datagram *old ) {
+	struct ipv4_fragment_info *fi, *n;
+
+	fw_debug ( "pd_delete %p\n", old );
+	list_for_each_entry_safe(fi, n, &old->fragment_info, fragment_info) {
+		fw_debug ( "Freeing fi %p\n", fi );
+		kfree(fi);
+	}
+	list_del(&old->pdg_list);
+	dev_kfree_skb_any(old->skb);
+	kfree(old);
+}
+
+static bool ipv4_pd_update ( struct ipv4_node *node, struct ipv4_partial_datagram *pd,
+ u32 *frag_buf, unsigned frag_off, unsigned frag_len) {
+	fw_debug ( "pd_update node %p, pd %p, frag_buf %p, %x@%x\n", node, pd, frag_buf, frag_len, frag_off );
+	if ( ipv4_frag_new ( pd, frag_off, frag_len ) == NULL)
+		return false;
+	memcpy(pd->pbuf + frag_off, frag_buf, frag_len);
+
+	/*
+	 * Move list entry to beginnig of list so that oldest partial
+	 * datagrams percolate to the end of the list
+	 */
+	list_move_tail(&pd->pdg_list, &node->pdg_list);
+	fw_debug ( "New pd list:\n" );
+	list_for_each_entry ( pd, &node->pdg_list, pdg_list ) {
+		fw_debug ( "pd %p\n", pd );
+	}
+	return true;
+}
+
+static bool ipv4_pd_is_complete ( struct ipv4_partial_datagram *pd ) {
+	struct ipv4_fragment_info *fi;
+	bool ret;
+
+	fi = list_entry(pd->fragment_info.next, struct ipv4_fragment_info, fragment_info);
+
+	ret = (fi->len == pd->datagram_size);
+	fw_debug ( "pd_is_complete (pd %p, dgs %x): fi %p (%x@%x) %s\n", pd, pd->datagram_size, fi, fi->len, fi->offset, ret ? "yes" : "no" );
+	return ret;
+}
+
+/* ------------------------------------------------------------------ */
+
+static int ipv4_node_new ( struct fw_card *card, struct fw_device *device ) {
+	struct ipv4_node *node;
+
+	node = kmalloc ( sizeof(*node), GFP_KERNEL );
+	if ( ! node ) {
+		fw_error ( "allocate new node failed\n" );
+		return -ENOMEM;
+	}
+	node->guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
+	node->fifo = INVALID_FIFO_ADDR;
+	INIT_LIST_HEAD(&node->pdg_list);
+	spin_lock_init(&node->pdg_lock);
+	node->pdg_size = 0;
+	node->generation = device->generation;
+	rmb();
+	node->nodeid = device->node_id;
+	 /* FIXME what should it really be? */
+	node->max_payload = S100_BUFFER_SIZE - IPV4_UNFRAG_HDR_SIZE;
+	node->datagram_label = 0U;
+	node->xmt_speed = device->max_speed;
+	list_add_tail ( &node->ipv4_nodes, &card->ipv4_nodes );
+	fw_debug ( "node_new: %p { guid %016llx, generation %u, nodeid %x, max_payload %x, xmt_speed %x } added\n",
+ node, (unsigned long long)node->guid, node->generation, node->nodeid, node->max_payload, node->xmt_speed );
+	return 0;
+}
+
+static struct ipv4_node *ipv4_node_find_by_guid(struct ipv4_priv *priv, u64 guid) {
+	struct ipv4_node *node;
+	unsigned long flags;
+
+	spin_lock_irqsave(&priv->lock, flags);
+	list_for_each_entry(node, &priv->card->ipv4_nodes, ipv4_nodes)
+		if (node->guid == guid) {
+			/* FIXME: lock the node first? */
+			spin_unlock_irqrestore ( &priv->lock, flags );
+			fw_debug ( "node_find_by_guid (%016llx) found %p\n", (unsigned long long)guid, node );
+			return node;
+		}
+
+	spin_unlock_irqrestore ( &priv->lock, flags );
+	fw_debug ( "node_find_by_guid (%016llx) not found\n", (unsigned long long)guid );
+	return NULL;
+}
+
+static struct ipv4_node *ipv4_node_find_by_nodeid(struct ipv4_priv *priv, u16 nodeid) {
+	struct ipv4_node *node;
+	unsigned long flags;
+
+	spin_lock_irqsave(&priv->lock, flags);
+	list_for_each_entry(node, &priv->card->ipv4_nodes, ipv4_nodes)
+		if (node->nodeid == nodeid) {
+			/* FIXME: lock the node first? */
+			spin_unlock_irqrestore ( &priv->lock, flags );
+			fw_debug ( "node_find_by_nodeid (%x) found %p\n", nodeid, node );
+			return node;
+		}
+	fw_debug ( "node_find_by_nodeid (%x) not found\n", nodeid );
+	spin_unlock_irqrestore ( &priv->lock, flags );
+	return NULL;
+}
+
+/* This is only complicated because we can't assume priv exists */
+static void ipv4_node_delete ( struct fw_card *card, struct fw_device *device ) {
+	struct net_device *netdev;
+	struct ipv4_priv *priv;
+	struct ipv4_node *node;
+	u64 guid;
+	unsigned long flags;
+	struct ipv4_partial_datagram *pd, *pd_next;
+
+	guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
+	netdev = card->netdev;
+	if ( netdev )
+		priv = netdev_priv ( netdev );
+	else
+		priv = NULL;
+	if ( priv )
+		spin_lock_irqsave ( &priv->lock, flags );
+	list_for_each_entry( node, &card->ipv4_nodes, ipv4_nodes ) {
+		if ( node->guid == guid ) {
+			list_del ( &node->ipv4_nodes );
+			list_for_each_entry_safe( pd, pd_next, &node->pdg_list, pdg_list )
+				ipv4_pd_delete ( pd );
+			break;
+		}
+	}
+	if ( priv )
+		spin_unlock_irqrestore ( &priv->lock, flags );
+}
+
+/* ------------------------------------------------------------------ */
+
+
+static int ipv4_finish_incoming_packet ( struct net_device *netdev,
+ struct sk_buff *skb, u16 source_node_id, bool is_broadcast, u16 ether_type ) {
+	struct ipv4_priv *priv;
+	static u64 broadcast_hw = ~0ULL;
+	int status;
+	u64 guid;
+
+	fw_debug ( "ipv4_finish_incoming_packet(%p, %p, %x, %s, %x\n",
+ netdev, skb, source_node_id, is_broadcast ? "true" : "false", ether_type );
+	priv = netdev_priv(netdev);
+	/* Write metadata, and then pass to the receive level */
+	skb->dev = netdev;
+	skb->ip_summed = CHECKSUM_UNNECESSARY;  /* don't check it */
+
+	/*
+	 * Parse the encapsulation header. This actually does the job of
+	 * converting to an ethernet frame header, as well as arp
+	 * conversion if needed. ARP conversion is easier in this
+	 * direction, since we are using ethernet as our backend.
+	 */
+	/*
+	 * If this is an ARP packet, convert it. First, we want to make
+	 * use of some of the fields, since they tell us a little bit
+	 * about the sending machine.
+	 */
+	if (ether_type == ETH_P_ARP) {
+		struct ipv4_arp *arp1394;
+		struct arphdr *arp;
+		unsigned char *arp_ptr;
+		u64 fifo_addr;
+		u8 max_rec;
+		u8 sspd;
+		u16 max_payload;
+		struct ipv4_node *node;
+		static const u16 ipv4_speed_to_max_payload[] = {
+			/* S100, S200, S400, S800, S1600, S3200 */
+			    512, 1024, 2048, 4096,  4096,  4096
+		};
+
+		/* fw_debug ( "ARP packet\n" ); */
+		arp1394 = (struct ipv4_arp *)skb->data;
+		arp = (struct arphdr *)skb->data;
+		arp_ptr = (unsigned char *)(arp + 1);
+		fifo_addr = (u64)ntohs(arp1394->fifo_hi) << 32 |
+ ntohl(arp1394->fifo_lo);
+		max_rec = priv->card->max_receive;
+		if ( arp1394->max_rec < max_rec )
+			max_rec = arp1394->max_rec;
+		sspd = arp1394->sspd;
+		/*
+		 * Sanity check. MacOSX seems to be sending us 131 in this
+		 * field (atleast on my Panther G5). Not sure why.
+		 */
+		if (sspd > 5 ) {
+			fw_notify ( "sspd %x out of range\n", sspd );
+			sspd = 0;
+		}
+
+		max_payload = min(ipv4_speed_to_max_payload[sspd],
+ (u16)(1 << (max_rec + 1))) - IPV4_UNFRAG_HDR_SIZE;
+
+		guid = be64_to_cpu(get_unaligned(&arp1394->s_uniq_id));
+		node = ipv4_node_find_by_guid(priv, guid);
+		if (!node) {
+			fw_notify ( "No node for ARP packet from %llx\n", guid );
+			goto failed_proto;
+		}
+		if ( node->nodeid != source_node_id || node->generation != priv->card->generation ) {
+			fw_notify ( "Internal error: node->nodeid (%x) != soucre_node_id (%x) or node->generation (%x) != priv->card->generation(%x)\n",
+ node->nodeid, source_node_id, node->generation, priv->card->generation );
+			node->nodeid = source_node_id;
+			node->generation = priv->card->generation;
+		}
+
+		/* FIXME: for debugging */
+		if ( sspd > SCODE_400 )
+			sspd = SCODE_400;
+		/* Update our speed/payload/fifo_offset table */
+		/*
+		 * FIXME: this does not handle cases where two high-speed endpoints must use a slower speed because of
+		 * a lower speed hub between them.  We need to look at the actual topology map here.
+		 */
+		fw_debug ( "Setting node %p fifo %llx (was %llx), max_payload %x (was %x), speed %x (was %x)\n",
+ node, fifo_addr, node->fifo, max_payload, node->max_payload, sspd, node->xmt_speed );
+		node->fifo =	fifo_addr;
+		node->max_payload = max_payload;
+		/*
+		 * Only allow speeds to go down from their initial value.
+		 * Otherwise a local node that can only do S400 or slower may
+		 * be told to transmit at S800 to a faster remote node.
+		 */
+		if ( node->xmt_speed > sspd )
+			node->xmt_speed = sspd;
+
+		/*
+		 * Now that we're done with the 1394 specific stuff, we'll
+		 * need to alter some of the data.  Believe it or not, all
+		 * that needs to be done is sender_IP_address needs to be
+		 * moved, the destination hardware address get stuffed
+		 * in and the hardware address length set to 8.
+		 *
+		 * IMPORTANT: The code below overwrites 1394 specific data
+		 * needed above so keep the munging of the data for the
+		 * higher level IP stack last.
+		 */
+
+		arp->ar_hln = 8;
+		arp_ptr += arp->ar_hln;		/* skip over sender unique id */
+		*(u32 *)arp_ptr = arp1394->sip; /* move sender IP addr */
+		arp_ptr += arp->ar_pln;		/* skip over sender IP addr */
+
+		if (arp->ar_op == htons(ARPOP_REQUEST))
+			memset(arp_ptr, 0, sizeof(u64));
+		else
+			memcpy(arp_ptr, netdev->dev_addr, sizeof(u64));
+	}
+
+	/* Now add the ethernet header. */
+	guid = cpu_to_be64(priv->card->guid);
+	if (dev_hard_header(skb, netdev, ether_type, is_broadcast ? &broadcast_hw : &guid, NULL,
+ skb->len) >= 0) {
+		struct ipv4_ether_hdr *eth;
+		u16 *rawp;
+		__be16 protocol;
+
+		skb_reset_mac_header(skb);
+		skb_pull(skb, sizeof(*eth));
+		eth = ipv4_ether_hdr(skb);
+		if (*eth->h_dest & 1) {
+			if (memcmp(eth->h_dest, netdev->broadcast, netdev->addr_len) == 0) {
+				fw_debug ( "Broadcast\n" );
+				skb->pkt_type = PACKET_BROADCAST;
+			}
+#if 0
+			else
+				skb->pkt_type = PACKET_MULTICAST;
+#endif
+		} else {
+			if (memcmp(eth->h_dest, netdev->dev_addr, netdev->addr_len)) {
+				u64 a1, a2;
+
+				memcpy ( &a1, eth->h_dest, sizeof(u64));
+				memcpy ( &a2, netdev->dev_addr, sizeof(u64));
+				fw_debug ( "Otherhost %llx %llx %x\n", a1, a2, netdev->addr_len );
+				skb->pkt_type = PACKET_OTHERHOST;
+			}
+		}
+		if (ntohs(eth->h_proto) >= 1536) {
+			fw_debug ( " proto %x %x\n", eth->h_proto, ntohs(eth->h_proto) );
+			protocol = eth->h_proto;
+		} else {
+			rawp = (u16 *)skb->data;
+			if (*rawp == 0xFFFF) {
+				fw_debug ( "proto 802_3\n" );
+				protocol = htons(ETH_P_802_3);
+			} else {
+				fw_debug ( "proto 802_2\n" );
+				protocol = htons(ETH_P_802_2);
+			}
+		}
+		skb->protocol = protocol;
+	}
+	status = netif_rx(skb);
+	if ( status == NET_RX_DROP) {
+		netdev->stats.rx_errors++;
+		netdev->stats.rx_dropped++;
+	} else {
+		netdev->stats.rx_packets++;
+		netdev->stats.rx_bytes += skb->len;
+	}
+	if (netif_queue_stopped(netdev))
+		netif_wake_queue(netdev);
+	return 0;
+
+ failed_proto:
+	netdev->stats.rx_errors++;
+	netdev->stats.rx_dropped++;
+	dev_kfree_skb_any(skb);
+	if (netif_queue_stopped(netdev))
+		netif_wake_queue(netdev);
+	netdev->last_rx = jiffies;
+	return 0;
+}
+
+/* ------------------------------------------------------------------ */
+
+static int ipv4_incoming_packet ( struct ipv4_priv *priv, u32 *buf, int len, u16 source_node_id, bool is_broadcast ) {
+	struct sk_buff *skb;
+	struct net_device *netdev;
+	struct ipv4_hdr hdr;
+	unsigned lf;
+	unsigned long flags;
+	struct ipv4_node *node;
+	struct ipv4_partial_datagram *pd;
+	int fg_off;
+	int dg_size;
+	u16 datagram_label;
+	int retval;
+	u16 ether_type;
+
+	fw_debug ( "ipv4_incoming_packet(%p, %p, %d, %x, %s)\n", priv, buf, len, source_node_id, is_broadcast ? "true" : "false" );
+	netdev = priv->card->netdev;
+
+	hdr.w0 = ntohl(buf[0]);
+	lf = ipv4_get_hdr_lf(&hdr);
+	if ( lf == IPV4_HDR_UNFRAG ) {
+		/*
+		 * An unfragmented datagram has been received by the ieee1394
+		 * bus. Build an skbuff around it so we can pass it to the
+		 * high level network layer.
+		 */
+		ether_type = ipv4_get_hdr_ether_type(&hdr);
+		fw_debug ( "header w0 = %x, lf = %x, ether_type = %x\n", hdr.w0, lf, ether_type );
+		buf++;
+		len -= IPV4_UNFRAG_HDR_SIZE;
+
+		skb = dev_alloc_skb(len + netdev->hard_header_len + 15);
+		if (unlikely(!skb)) {
+			fw_error ( "Out of memory for incoming packet\n");
+			netdev->stats.rx_dropped++;
+			return -1;
+		}
+		skb_reserve(skb, (netdev->hard_header_len + 15) & ~15);
+		memcpy(skb_put(skb, len), buf, len );
+		return ipv4_finish_incoming_packet(netdev, skb, source_node_id, is_broadcast, ether_type );
+	}
+	/* A datagram fragment has been received, now the fun begins. */
+	hdr.w1 = ntohl(buf[1]);
+	buf +=2;
+	len -= IPV4_FRAG_HDR_SIZE;
+	if ( lf ==IPV4_HDR_FIRSTFRAG ) {
+		ether_type = ipv4_get_hdr_ether_type(&hdr);
+		fg_off = 0;
+	} else {
+		fg_off = ipv4_get_hdr_fg_off(&hdr);
+		ether_type = 0; /* Shut up compiler! */
+	}
+	datagram_label = ipv4_get_hdr_dgl(&hdr);
+	dg_size = ipv4_get_hdr_dg_size(&hdr); /* ??? + 1 */
+	fw_debug ( "fragmented: %x.%x = lf %x, ether_type %x, fg_off %x, dgl %x, dg_size %x\n", hdr.w0, hdr.w1, lf, ether_type, fg_off, datagram_label, dg_size );
+	node = ipv4_node_find_by_nodeid ( priv, source_node_id);
+	spin_lock_irqsave(&node->pdg_lock, flags);
+	pd = ipv4_pd_find( node, datagram_label );
+	if (pd == NULL) {
+		while ( node->pdg_size >= ipv4_mpd ) {
+			/* remove the oldest */
+			ipv4_pd_delete ( list_first_entry(&node->pdg_list, struct ipv4_partial_datagram, pdg_list) );
+			node->pdg_size--;
+		}
+		pd = ipv4_pd_new ( netdev, node, datagram_label, dg_size,
+ buf, fg_off, len);
+		if ( pd == NULL) {
+			retval = -ENOMEM;
+			goto bad_proto;
+		}
+		node->pdg_size++;
+	} else {
+		if (ipv4_frag_overlap(pd, fg_off, len) || pd->datagram_size != dg_size) {
+			/*
+			 * Differing datagram sizes or overlapping fragments,
+			 * Either way the remote machine is playing silly buggers
+			 * with us: obliterate the old datagram and start a new one.
+			 */
+			ipv4_pd_delete ( pd );
+			pd = ipv4_pd_new ( netdev, node, datagram_label,
+ dg_size, buf, fg_off, len);
+			if ( pd == NULL ) {
+				retval = -ENOMEM;
+				node->pdg_size--;
+				goto bad_proto;
+			}
+		} else {
+			bool worked;
+
+			worked = ipv4_pd_update ( node, pd,
+ buf, fg_off, len );
+			if ( ! worked ) {
+				/*
+				 * Couldn't save off fragment anyway
+				 * so might as well obliterate the
+				 * datagram now.
+				 */
+				ipv4_pd_delete ( pd );
+				node->pdg_size--;
+				goto bad_proto;
+			}
+		}
+	} /* new datagram or add to existing one */
+
+	if ( lf == IPV4_HDR_FIRSTFRAG )
+		pd->ether_type = ether_type;
+	if ( ipv4_pd_is_complete ( pd ) ) {
+		ether_type = pd->ether_type;
+		node->pdg_size--;
+		skb = skb_get(pd->skb);
+		ipv4_pd_delete ( pd );
+		spin_unlock_irqrestore(&node->pdg_lock, flags);
+		return ipv4_finish_incoming_packet ( netdev, skb, source_node_id, false, ether_type );
+	}
+	/*
+	 * Datagram is not complete, we're done for the
+	 * moment.
+	 */
+	spin_unlock_irqrestore(&node->pdg_lock, flags);
+	return 0;
+
+ bad_proto:
+	spin_unlock_irqrestore(&node->pdg_lock, flags);
+	if (netif_queue_stopped(netdev))
+		netif_wake_queue(netdev);
+	return 0;
+}
+
+static void ipv4_receive_packet ( struct fw_card *card, struct fw_request *r,
+ int tcode, int destination, int source, int generation, int speed,
+ unsigned long long offset, void *payload, size_t length, void *callback_data ) {
+	struct ipv4_priv *priv;
+	int status;
+
+	fw_debug ( "ipv4_receive_packet(%p,%p,%x,%x,%x,%x,%x,%llx,%p,%lx,%p)\n",
+ card, r, tcode, destination, source, generation, speed, offset, payload,
+ (unsigned long)length, callback_data);
+	print_hex_dump ( KERN_DEBUG, "header: ", DUMP_PREFIX_OFFSET, 32, 1, payload, length, false );
+	priv = callback_data;
+	if (   tcode != TCODE_WRITE_BLOCK_REQUEST
+	    || destination != card->node_id
+	    || generation != card->generation
+	    || offset != priv->handler.offset ) {
+		fw_send_response(card, r, RCODE_CONFLICT_ERROR);
+		fw_debug("Conflict error card node_id=%x, card generation=%x, local offset %llx\n",
+ card->node_id, card->generation, (unsigned long long)priv->handler.offset );
+		return;
+	}
+	status = ipv4_incoming_packet ( priv, payload, length, source, false );
+	if ( status != 0 ) {
+		fw_error ( "Incoming packet failure\n" );
+		fw_send_response ( card, r, RCODE_CONFLICT_ERROR );
+		return;
+	}
+	fw_send_response ( card, r, RCODE_COMPLETE );
+}
+
+static void ipv4_receive_broadcast(struct fw_iso_context *context, u32 cycle,
+ size_t header_length, void *header, void *data) {
+	struct ipv4_priv *priv;
+	struct fw_iso_packet packet;
+	struct fw_card *card;
+	u16 *hdr_ptr;
+	u32 *buf_ptr;
+	int retval;
+	u32 length;
+	u16 source_node_id;
+	u32 specifier_id;
+	u32 ver;
+	unsigned long offset;
+	unsigned long flags;
+
+	fw_debug ( "ipv4_receive_broadcast ( context=%p, cycle=%x, header_length=%lx, header=%p, data=%p )\n", context, cycle, (unsigned long)header_length, header, data );
+	print_hex_dump ( KERN_DEBUG, "header: ", DUMP_PREFIX_OFFSET, 32, 1, header, header_length, false );
+	priv = data;
+	card = priv->card;
+	hdr_ptr = header;
+	length = ntohs(hdr_ptr[0]);
+	spin_lock_irqsave(&priv->lock,flags);
+	offset = priv->rcv_buffer_size * priv->broadcast_rcv_next_ptr;
+	buf_ptr = priv->broadcast_rcv_buffer_ptrs[priv->broadcast_rcv_next_ptr++];
+	if ( priv->broadcast_rcv_next_ptr == priv->num_broadcast_rcv_ptrs )
+		priv->broadcast_rcv_next_ptr = 0;
+	spin_unlock_irqrestore(&priv->lock,flags);
+	fw_debug ( "length %u at %p\n", length, buf_ptr );
+	print_hex_dump ( KERN_DEBUG, "buffer: ", DUMP_PREFIX_OFFSET, 32, 1, buf_ptr, length, false );
+
+	specifier_id =    (be32_to_cpu(buf_ptr[0]) & 0xffff) << 8
+			| (be32_to_cpu(buf_ptr[1]) & 0xff000000) >> 24;
+	ver = be32_to_cpu(buf_ptr[1]) & 0xFFFFFF;
+	source_node_id = be32_to_cpu(buf_ptr[0]) >> 16;
+	/* fw_debug ( "source %x SpecID %x ver %x\n", source_node_id, specifier_id, ver ); */
+	if ( specifier_id == IPV4_GASP_SPECIFIER_ID && ver == IPV4_GASP_VERSION ) {
+		buf_ptr += 2;
+		length -= IPV4_GASP_OVERHEAD;
+		ipv4_incoming_packet(priv, buf_ptr, length, source_node_id, true);
+	} else
+		fw_debug ( "Ignoring packet: not GASP\n" );
+	packet.payload_length = priv->rcv_buffer_size;
+	packet.interrupt = 1;
+	packet.skip = 0;
+	packet.tag = 3;
+	packet.sy = 0;
+	packet.header_length = IPV4_GASP_OVERHEAD;
+	spin_lock_irqsave(&priv->lock,flags);
+	retval = fw_iso_context_queue ( priv->broadcast_rcv_context, &packet,
+ &priv->broadcast_rcv_buffer, offset );
+	spin_unlock_irqrestore(&priv->lock,flags);
+	if ( retval < 0 )
+		fw_error ( "requeue failed\n" );
+}
+
+static void debug_ptask ( struct ipv4_packet_task *ptask ) {
+	static const char *tx_types[] = { "Unknown", "GASP", "Write" };
+
+	fw_debug ( "packet %p { hdr { w0 %x w1 %x }, skb %p, priv %p,"
+ " tx_type %s, outstanding_pkts %d, max_payload %x, fifo %llx,"
+ " speed %x, dest_node %x, generation %x }\n",
+ ptask, ptask->hdr.w0, ptask->hdr.w1, ptask->skb, ptask->priv,
+ ptask->tx_type > IPV4_WRREQ ? "Invalid" : tx_types[ptask->tx_type],
+ ptask->outstanding_pkts,  ptask->max_payload,
+ ptask->fifo_addr, ptask->speed, ptask->dest_node, ptask->generation );
+	print_hex_dump ( KERN_DEBUG, "packet :", DUMP_PREFIX_OFFSET, 32, 1,
+ ptask->skb->data, ptask->skb->len, false );
+}
+
+static void ipv4_transmit_packet_done ( struct ipv4_packet_task *ptask ) {
+	struct ipv4_priv *priv;
+	unsigned long flags;
+
+	priv = ptask->priv;
+	spin_lock_irqsave ( &priv->lock, flags );
+	list_del ( &ptask->packet_list );
+	spin_unlock_irqrestore ( &priv->lock, flags );
+	ptask->outstanding_pkts--;
+	if ( ptask->outstanding_pkts > 0 ) {
+		u16 dg_size;
+		u16 fg_off;
+		u16 datagram_label;
+		u16 lf;
+		struct sk_buff *skb;
+
+		/* Update the ptask to point to the next fragment and send it */
+		lf = ipv4_get_hdr_lf(&ptask->hdr);
+		switch (lf) {
+		case IPV4_HDR_LASTFRAG:
+		case IPV4_HDR_UNFRAG:
+		default:
+			fw_error ( "Outstanding packet %x lf %x, header %x,%x\n", ptask->outstanding_pkts, lf, ptask->hdr.w0, ptask->hdr.w1 );
+			BUG();
+
+		case IPV4_HDR_FIRSTFRAG:
+			/* Set frag type here for future interior fragments */
+			dg_size = ipv4_get_hdr_dg_size(&ptask->hdr);
+			fg_off = ptask->max_payload - IPV4_FRAG_HDR_SIZE;
+			datagram_label = ipv4_get_hdr_dgl(&ptask->hdr);
+			break;
+
+		case IPV4_HDR_INTFRAG:
+			dg_size = ipv4_get_hdr_dg_size(&ptask->hdr);
+			fg_off = ipv4_get_hdr_fg_off(&ptask->hdr) + ptask->max_payload - IPV4_FRAG_HDR_SIZE;
+			datagram_label = ipv4_get_hdr_dgl(&ptask->hdr);
+			break;
+		}
+		skb = ptask->skb;
+		skb_pull ( skb, ptask->max_payload );
+		if ( ptask->outstanding_pkts > 1 ) {
+			ipv4_make_sf_hdr ( &ptask->hdr,
+  IPV4_HDR_INTFRAG, dg_size, fg_off, datagram_label );
+		} else {
+			ipv4_make_sf_hdr ( &ptask->hdr,
+  IPV4_HDR_LASTFRAG, dg_size, fg_off, datagram_label );
+			ptask->max_payload = skb->len + IPV4_FRAG_HDR_SIZE;
+
+		}
+		ipv4_send_packet ( ptask );
+	} else {
+		dev_kfree_skb_any ( ptask->skb );
+		kmem_cache_free( ipv4_packet_task_cache, ptask );
+	}
+}
+
+static void ipv4_write_complete ( struct fw_card *card, int rcode,
+ void *payload, size_t length, void *data ) {
+	struct ipv4_packet_task *ptask;
+
+	ptask = data;
+	fw_debug ( "ipv4_write_complete ( %p, %x, %p, %lx, %p )\n",
+ card, rcode, payload, (unsigned long)length, data );
+	debug_ptask ( ptask );
+
+	if ( rcode == RCODE_COMPLETE ) {
+		ipv4_transmit_packet_done ( ptask );
+	} else {
+		fw_error ( "ipv4_write_complete: failed: %x\n", rcode );
+		/* ??? error recovery */
+	}
+}
+
+static int ipv4_send_packet ( struct ipv4_packet_task *ptask ) {
+	struct ipv4_priv *priv;
+	unsigned tx_len;
+	struct ipv4_hdr *bufhdr;
+	unsigned long flags;
+	struct net_device *netdev;
+#if 0 /* stefanr */
+	int retval;
+#endif
+
+	fw_debug ( "ipv4_send_packet\n" );
+	debug_ptask ( ptask );
+	priv = ptask->priv;
+	tx_len = ptask->max_payload;
+	switch (ipv4_get_hdr_lf(&ptask->hdr)) {
+	case IPV4_HDR_UNFRAG:
+		bufhdr = (struct ipv4_hdr *)skb_push(ptask->skb, IPV4_UNFRAG_HDR_SIZE);
+		bufhdr->w0 = htonl(ptask->hdr.w0);
+		break;
+
+	case IPV4_HDR_FIRSTFRAG:
+	case IPV4_HDR_INTFRAG:
+	case IPV4_HDR_LASTFRAG:
+		bufhdr = (struct ipv4_hdr *)skb_push(ptask->skb, IPV4_FRAG_HDR_SIZE);
+		bufhdr->w0 = htonl(ptask->hdr.w0);
+		bufhdr->w1 = htonl(ptask->hdr.w1);
+		break;
+
+	default:
+		BUG();
+	}
+	if ( ptask->tx_type == IPV4_GASP ) {
+		u32 *packets;
+		int generation;
+		int nodeid;
+
+		/* ptask->generation may not have been set yet */
+		generation = priv->card->generation;
+		smp_rmb();
+		nodeid = priv->card->node_id;
+		packets = (u32 *)skb_push(ptask->skb, sizeof(u32)*2);
+		packets[0] = htonl(nodeid << 16 | (IPV4_GASP_SPECIFIER_ID>>8));
+		packets[1] = htonl((IPV4_GASP_SPECIFIER_ID & 0xFF) << 24 | IPV4_GASP_VERSION);
+		fw_send_request ( priv->card, &ptask->transaction, TCODE_STREAM_DATA,
+ fw_stream_packet_destination_id(3, BROADCAST_CHANNEL, 0),
+ generation, SCODE_100, 0ULL, ptask->skb->data, tx_len + 8, ipv4_write_complete, ptask );
+		spin_lock_irqsave(&priv->lock,flags);
+		list_add_tail ( &ptask->packet_list, &priv->broadcasted_list );
+		spin_unlock_irqrestore(&priv->lock,flags);
+#if 0 /* stefanr */
+		return retval;
+#else
+		return 0;
+#endif
+	}
+	fw_debug("send_request (%p, %p, WRITE_BLOCK, %x, %x, %x, %llx, %p, %d, %p, %p\n",
+ priv->card, &ptask->transaction, ptask->dest_node, ptask->generation,
+ ptask->speed, (unsigned long long)ptask->fifo_addr, ptask->skb->data, tx_len,
+ ipv4_write_complete, ptask );
+	fw_send_request ( priv->card, &ptask->transaction,
+ TCODE_WRITE_BLOCK_REQUEST, ptask->dest_node, ptask->generation, ptask->speed,
+ ptask->fifo_addr, ptask->skb->data, tx_len, ipv4_write_complete, ptask );
+	spin_lock_irqsave(&priv->lock,flags);
+	list_add_tail ( &ptask->packet_list, &priv->sent_list );
+	spin_unlock_irqrestore(&priv->lock,flags);
+	netdev = priv->card->netdev;
+	netdev->trans_start = jiffies;
+	return 0;
+}
+
+static int ipv4_broadcast_start ( struct ipv4_priv *priv ) {
+	struct fw_iso_context *context;
+	int retval;
+	unsigned num_packets;
+	unsigned max_receive;
+	struct fw_iso_packet packet;
+	unsigned long offset;
+	unsigned u;
+	/* unsigned transmit_speed; */
+
+#if 0 /* stefanr */
+	if ( priv->card->broadcast_channel != (BROADCAST_CHANNEL_VALID|BROADCAST_CHANNEL_INITIAL)) {
+		fw_notify ( "Invalid broadcast channel %x\n", priv->card->broadcast_channel );
+		/* FIXME: try again later? */
+		/* return -EINVAL; */
+	}
+#endif
+	if ( priv->local_fifo == INVALID_FIFO_ADDR ) {
+		struct fw_address_region region;
+
+		priv->handler.length = FIFO_SIZE;
+		priv->handler.address_callback = ipv4_receive_packet;
+		priv->handler.callback_data = priv;
+		/* FIXME: this is OHCI, but what about others? */
+		region.start = 0xffff00000000ULL;
+		region.end =   0xfffffffffffcULL;
+
+		retval = fw_core_add_address_handler ( &priv->handler, &region );
+		if ( retval < 0 )
+			goto failed_initial;
+		priv->local_fifo = priv->handler.offset;
+	}
+
+	/*
+	 * FIXME: rawiso limits us to PAGE_SIZE.  This only matters if we ever have
+	 * a machine with PAGE_SIZE < 4096
+	 */
+	max_receive = 1U << (priv->card->max_receive + 1);
+	num_packets = ( ipv4_iso_page_count * PAGE_SIZE ) / max_receive;
+	if ( ! priv->broadcast_rcv_context ) {
+		void **ptrptr;
+
+		context = fw_iso_context_create ( priv->card,
+ FW_ISO_CONTEXT_RECEIVE, BROADCAST_CHANNEL,
+ priv->card->link_speed, 8, ipv4_receive_broadcast, priv );
+		if (IS_ERR(context)) {
+			retval = PTR_ERR(context);
+			goto failed_context_create;
+		}
+		retval = fw_iso_buffer_init ( &priv->broadcast_rcv_buffer,
+ priv->card, ipv4_iso_page_count, DMA_FROM_DEVICE );
+		if ( retval < 0 )
+			goto failed_buffer_init;
+		ptrptr = kmalloc ( sizeof(void*)*num_packets, GFP_KERNEL );
+		if ( ! ptrptr ) {
+			retval = -ENOMEM;
+			goto failed_ptrs_alloc;
+		}
+		priv->broadcast_rcv_buffer_ptrs = ptrptr;
+		for ( u = 0; u < ipv4_iso_page_count; u++ ) {
+			void *ptr;
+			unsigned v;
+
+			ptr = kmap ( priv->broadcast_rcv_buffer.pages[u] );
+			for ( v = 0; v < num_packets / ipv4_iso_page_count; v++ )
+				*ptrptr++ = (void *)((char *)ptr + v * max_receive);
+		}
+		priv->broadcast_rcv_context = context;
+	} else
+		context = priv->broadcast_rcv_context;
+
+	packet.payload_length = max_receive;
+	packet.interrupt = 1;
+	packet.skip = 0;
+	packet.tag = 3;
+	packet.sy = 0;
+	packet.header_length = IPV4_GASP_OVERHEAD;
+	offset = 0;
+	for ( u = 0; u < num_packets; u++ ) {
+		retval = fw_iso_context_queue ( context, &packet,
+ &priv->broadcast_rcv_buffer, offset );
+		if ( retval < 0 )
+			goto failed_rcv_queue;
+		offset += max_receive;
+	}
+	priv->num_broadcast_rcv_ptrs = num_packets;
+	priv->rcv_buffer_size = max_receive;
+	priv->broadcast_rcv_next_ptr = 0U;
+	retval = fw_iso_context_start ( context, -1, 0, FW_ISO_CONTEXT_MATCH_ALL_TAGS ); /* ??? sync */
+	if ( retval < 0 )
+		goto failed_rcv_queue;
+	/* FIXME: adjust this when we know the max receive speeds of all other IP nodes on the bus. */
+	/* since we only xmt at S100 ??? */
+	priv->broadcast_xmt_max_payload = S100_BUFFER_SIZE - IPV4_GASP_OVERHEAD - IPV4_UNFRAG_HDR_SIZE;
+	priv->broadcast_state = IPV4_BROADCAST_RUNNING;
+	return 0;
+
+ failed_rcv_queue:
+	kfree ( priv->broadcast_rcv_buffer_ptrs );
+	priv->broadcast_rcv_buffer_ptrs = NULL;
+ failed_ptrs_alloc:
+	fw_iso_buffer_destroy ( &priv->broadcast_rcv_buffer, priv->card );
+ failed_buffer_init:
+	fw_iso_context_destroy ( context );
+	priv->broadcast_rcv_context = NULL;
+ failed_context_create:
+	fw_core_remove_address_handler ( &priv->handler );
+ failed_initial:
+	priv->local_fifo = INVALID_FIFO_ADDR;
+	return retval;
+}
+
+/* This is called after an "ifup" */
+static int ipv4_open(struct net_device *dev) {
+	struct ipv4_priv *priv;
+	int ret;
+
+	priv = netdev_priv(dev);
+	if (priv->broadcast_state == IPV4_BROADCAST_ERROR) {
+		ret = ipv4_broadcast_start ( priv );
+		if (ret)
+			return ret;
+	}
+	netif_start_queue(dev);
+	return 0;
+}
+
+/* This is called after an "ifdown" */
+static int ipv4_stop(struct net_device *netdev)
+{
+	/* flush priv->wake */
+	/* flush_scheduled_work(); */
+
+	netif_stop_queue(netdev);
+	return 0;
+}
+
+/* Transmit a packet (called by kernel) */
+static int ipv4_tx(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct ipv4_ether_hdr hdr_buf;
+	struct ipv4_priv *priv = netdev_priv(netdev);
+	__be16 proto;
+	u16 dest_node;
+	enum ipv4_tx_type tx_type;
+	unsigned max_payload;
+	u16 dg_size;
+	u16 *datagram_label_ptr;
+	struct ipv4_packet_task *ptask;
+	struct ipv4_node *node = NULL;
+
+	ptask = kmem_cache_alloc(ipv4_packet_task_cache, GFP_ATOMIC);
+	if (ptask == NULL)
+		goto fail;
+
+	skb = skb_share_check(skb, GFP_ATOMIC);
+	if (!skb)
+		goto fail;
+
+	/*
+	 * Get rid of the fake ipv4 header, but first make a copy.
+	 * We might need to rebuild the header on tx failure.
+	 */
+	memcpy(&hdr_buf, skb->data, sizeof(hdr_buf));
+	skb_pull(skb, sizeof(hdr_buf));
+
+	proto = hdr_buf.h_proto;
+	dg_size = skb->len;
+
+	/*
+	 * Set the transmission type for the packet.  ARP packets and IP
+	 * broadcast packets are sent via GASP.
+	 */
+	if (   memcmp(hdr_buf.h_dest, netdev->broadcast, IPV4_ALEN) == 0
+	    || proto == htons(ETH_P_ARP)
+	    || (   proto == htons(ETH_P_IP)
+		&& IN_MULTICAST(ntohl(ip_hdr(skb)->daddr)) ) ) {
+		/* fw_debug ( "transmitting arp or multicast packet\n" );*/
+		tx_type = IPV4_GASP;
+		dest_node = ALL_NODES;
+		max_payload = priv->broadcast_xmt_max_payload;
+		/* BUG_ON(max_payload < S100_BUFFER_SIZE - IPV4_GASP_OVERHEAD); */
+		datagram_label_ptr = &priv->broadcast_xmt_datagramlabel;
+		ptask->fifo_addr = INVALID_FIFO_ADDR;
+		ptask->generation = 0U;
+		ptask->dest_node = 0U;
+		ptask->speed = 0;
+	} else {
+		__be64 guid = get_unaligned((u64 *)hdr_buf.h_dest);
+		u8 generation;
+
+		node = ipv4_node_find_by_guid(priv, be64_to_cpu(guid));
+		if (!node) {
+			fw_debug ( "Normal packet but no node\n" );
+			goto fail;
+		}
+
+		if (node->fifo == INVALID_FIFO_ADDR) {
+			fw_debug ( "Normal packet but no fifo addr\n" );
+			goto fail;
+		}
+
+		/* fw_debug ( "Transmitting normal packet to %x at %llxx\n", node->nodeid, node->fifo ); */
+		generation = node->generation;
+		dest_node = node->nodeid;
+		max_payload = node->max_payload;
+		/* BUG_ON(max_payload < S100_BUFFER_SIZE - IPV4_FRAG_HDR_SIZE); */
+
+		datagram_label_ptr = &node->datagram_label;
+		tx_type = IPV4_WRREQ;
+		ptask->fifo_addr = node->fifo;
+		ptask->generation = generation;
+		ptask->dest_node = dest_node;
+		ptask->speed = node->xmt_speed;
+	}
+
+	/* If this is an ARP packet, convert it */
+	if (proto == htons(ETH_P_ARP)) {
+		/* Convert a standard ARP packet to 1394 ARP. The first 8 bytes (the entire
+		 * arphdr) is the same format as the ip1394 header, so they overlap.  The rest
+		 * needs to be munged a bit.  The remainder of the arphdr is formatted based
+		 * on hwaddr len and ipaddr len.  We know what they'll be, so it's easy to
+		 * judge.
+		 *
+		 * Now that the EUI is used for the hardware address all we need to do to make
+		 * this work for 1394 is to insert 2 quadlets that contain max_rec size,
+		 * speed, and unicast FIFO address information between the sender_unique_id
+		 * and the IP addresses.
+		 */
+		struct arphdr *arp = (struct arphdr *)skb->data;
+		unsigned char *arp_ptr = (unsigned char *)(arp + 1);
+		struct ipv4_arp *arp1394 = (struct ipv4_arp *)skb->data;
+		u32 ipaddr;
+
+		ipaddr = *(u32*)(arp_ptr + IPV4_ALEN);
+		arp1394->hw_addr_len    = 16;
+		arp1394->max_rec        = priv->card->max_receive;
+		arp1394->sspd		= priv->card->link_speed;
+		arp1394->fifo_hi	= htons(priv->local_fifo >> 32);
+		arp1394->fifo_lo        = htonl(priv->local_fifo & 0xFFFFFFFF);
+		arp1394->sip		= ipaddr;
+	}
+	if ( ipv4_max_xmt && max_payload > ipv4_max_xmt )
+		max_payload = ipv4_max_xmt;
+
+	ptask->hdr.w0 = 0;
+	ptask->hdr.w1 = 0;
+	ptask->skb = skb;
+	ptask->priv = priv;
+        ptask->tx_type = tx_type;
+	/* Does it all fit in one packet? */
+	if ( dg_size <= max_payload ) {
+		ipv4_make_uf_hdr(&ptask->hdr, be16_to_cpu(proto));
+		ptask->outstanding_pkts = 1;
+		max_payload = dg_size + IPV4_UNFRAG_HDR_SIZE;
+	} else {
+		u16 datagram_label;
+
+		max_payload -= IPV4_FRAG_OVERHEAD;
+		datagram_label = (*datagram_label_ptr)++;
+		ipv4_make_ff_hdr(&ptask->hdr, be16_to_cpu(proto), dg_size, datagram_label );
+		ptask->outstanding_pkts = DIV_ROUND_UP(dg_size, max_payload);
+		max_payload += IPV4_FRAG_HDR_SIZE;
+	}
+	ptask->max_payload = max_payload;
+	ipv4_send_packet ( ptask );
+	return NETDEV_TX_OK;
+
+ fail:
+	if (ptask)
+		kmem_cache_free(ipv4_packet_task_cache, ptask);
+
+	if (skb != NULL)
+		dev_kfree_skb(skb);
+
+	netdev->stats.tx_dropped++;
+	netdev->stats.tx_errors++;
+
+	/*
+	 * FIXME: According to a patch from 2003-02-26, "returning non-zero
+	 * causes serious problems" here, allegedly.  Before that patch,
+	 * -ERRNO was returned which is not appropriate under Linux 2.6.
+	 * Perhaps more needs to be done?  Stop the queue in serious
+	 * conditions and restart it elsewhere?
+	 */
+	return NETDEV_TX_OK;
+}
+
+/*
+ * FIXME: What to do if we timeout? I think a host reset is probably in order,
+ * so that's what we do. Should we increment the stat counters too?
+ */
+static void ipv4_tx_timeout(struct net_device *dev) {
+	struct ipv4_priv *priv;
+
+	priv = netdev_priv(dev);
+	fw_error ( "%s: Timeout, resetting host\n", dev->name );
+#if 0 /* stefanr */
+	fw_core_initiate_bus_reset ( priv->card, 1 );
+#endif
+}
+
+static int ipv4_change_mtu ( struct net_device *dev, int new_mtu ) {
+#if 0
+	int max_mtu;
+	struct ipv4_priv *priv;
+#endif
+
+	if (new_mtu < 68)
+		return -EINVAL;
+
+#if 0
+	priv = netdev_priv(dev);
+	/* This is not actually true because we can fragment packets at the firewire layer */
+	max_mtu = (1 << (priv->card->max_receive + 1))
+		                - sizeof(struct ipv4_hdr) - IPV4_GASP_OVERHEAD;
+	if (new_mtu > max_mtu) {
+		fw_notify ( "%s: Local node constrains MTU to %d\n", dev->name, max_mtu);
+		return -ERANGE;
+	}
+#endif
+	dev->mtu = new_mtu;
+	return 0;
+}
+
+static void ipv4_get_drvinfo(struct net_device *dev,
+struct ethtool_drvinfo *info) {
+	strcpy(info->driver, ipv4_driver_name);
+	strcpy(info->bus_info, "ieee1394"); /* FIXME provide more detail? */
+}
+
+static struct ethtool_ops ipv4_ethtool_ops = {
+	.get_drvinfo = ipv4_get_drvinfo,
+};
+
+static const struct net_device_ops ipv4_netdev_ops = {
+	.ndo_open       = ipv4_open,
+	.ndo_stop	= ipv4_stop,
+	.ndo_start_xmit = ipv4_tx,
+	.ndo_tx_timeout = ipv4_tx_timeout,
+	.ndo_change_mtu = ipv4_change_mtu,
+};
+
+static void ipv4_init_dev ( struct net_device *dev ) {
+	dev->header_ops		= &ipv4_header_ops;
+	dev->netdev_ops         = &ipv4_netdev_ops;
+	SET_ETHTOOL_OPS(dev, &ipv4_ethtool_ops);
+
+	dev->watchdog_timeo	= IPV4_TIMEOUT;
+	dev->flags		= IFF_BROADCAST | IFF_MULTICAST;
+	dev->features		= NETIF_F_HIGHDMA;
+	dev->addr_len		= IPV4_ALEN;
+	dev->hard_header_len	= IPV4_HLEN;
+	dev->type		= ARPHRD_IEEE1394;
+
+	/* FIXME: This value was copied from ether_setup(). Is it too much? */
+	dev->tx_queue_len	= 1000;
+}
+
+static int ipv4_probe ( struct device *dev ) {
+	struct fw_unit * unit;
+	struct fw_device *device;
+	struct fw_card *card;
+	struct net_device *netdev;
+	struct ipv4_priv *priv;
+	unsigned max_mtu;
+	__be64 guid;
+
+	fw_debug("ipv4 Probing\n" );
+	unit = fw_unit ( dev );
+	device = fw_device ( unit->device.parent );
+	card = device->card;
+
+	if ( ! device->is_local ) {
+		int added;
+
+		fw_debug ( "Non-local, adding remote node entry\n" );
+		added = ipv4_node_new ( card, device );
+		return added;
+	}
+	fw_debug("ipv4 Local: adding netdev\n" );
+	netdev = alloc_netdev ( sizeof(*priv), "fw-ipv4-%d", ipv4_init_dev );
+	if ( netdev == NULL) {
+		fw_error( "Out of memory\n");
+		goto out;
+	}
+
+	SET_NETDEV_DEV(netdev, card->device);
+	priv = netdev_priv(netdev);
+
+	spin_lock_init(&priv->lock);
+	priv->broadcast_state = IPV4_BROADCAST_ERROR;
+	priv->broadcast_rcv_context = NULL;
+	priv->broadcast_xmt_max_payload = 0;
+	priv->broadcast_xmt_datagramlabel = 0;
+
+	priv->local_fifo = INVALID_FIFO_ADDR;
+
+	/* INIT_WORK(&priv->wake, ipv4_handle_queue);*/
+	INIT_LIST_HEAD(&priv->packet_list);
+	INIT_LIST_HEAD(&priv->broadcasted_list);
+	INIT_LIST_HEAD(&priv->sent_list );
+
+	priv->card = card;
+
+	/*
+	 * Use the RFC 2734 default 1500 octets or the maximum payload
+	 * as initial MTU
+	 */
+	max_mtu = (1 << (card->max_receive + 1))
+		  - sizeof(struct ipv4_hdr) - IPV4_GASP_OVERHEAD;
+	netdev->mtu = min(1500U, max_mtu);
+
+	/* Set our hardware address while we're at it */
+	guid = cpu_to_be64(card->guid);
+	memcpy(netdev->dev_addr, &guid, sizeof(u64));
+	memset(netdev->broadcast, 0xff, sizeof(u64));
+	if ( register_netdev ( netdev ) ) {
+		fw_error ( "Cannot register the driver\n");
+		goto out;
+	}
+
+	fw_notify ( "%s: IPv4 over Firewire on device %016llx\n",
+ netdev->name, card->guid );
+	card->netdev = netdev;
+
+	return 0 /* ipv4_new_node ( ud ) */;
+ out:
+	if ( netdev )
+		free_netdev ( netdev );
+	return -ENOENT;
+}
+
+
+static int ipv4_remove ( struct device *dev ) {
+	struct fw_unit * unit;
+	struct fw_device *device;
+	struct fw_card *card;
+	struct net_device *netdev;
+	struct ipv4_priv *priv;
+	struct ipv4_node *node;
+	struct ipv4_partial_datagram *pd, *pd_next;
+	struct ipv4_packet_task *ptask, *pt_next;
+
+	fw_debug("ipv4 Removing\n" );
+	unit = fw_unit ( dev );
+	device = fw_device ( unit->device.parent );
+	card = device->card;
+
+	if ( ! device->is_local ) {
+		fw_debug ( "Node %x is non-local, removing remote node entry\n", device->node_id );
+		ipv4_node_delete ( card, device );
+		return 0;
+	}
+	netdev = card->netdev;
+	if ( netdev ) {
+		fw_debug ( "Node %x is local: deleting netdev\n", device->node_id );
+		priv = netdev_priv ( netdev );
+		unregister_netdev ( netdev );
+		fw_debug ( "unregistered\n" );
+		if ( priv->local_fifo != INVALID_FIFO_ADDR )
+			fw_core_remove_address_handler ( &priv->handler );
+		fw_debug ( "address handler gone\n" );
+		if ( priv->broadcast_rcv_context ) {
+			fw_iso_context_stop ( priv->broadcast_rcv_context );
+			fw_iso_buffer_destroy ( &priv->broadcast_rcv_buffer, priv->card );
+			fw_iso_context_destroy ( priv->broadcast_rcv_context );
+			fw_debug ( "rcv stopped\n" );
+		}
+		list_for_each_entry_safe( ptask, pt_next, &priv->packet_list, packet_list ) {
+			dev_kfree_skb_any ( ptask->skb );
+			kmem_cache_free( ipv4_packet_task_cache, ptask );
+		}
+		list_for_each_entry_safe( ptask, pt_next, &priv->broadcasted_list, packet_list ) {
+			dev_kfree_skb_any ( ptask->skb );
+			kmem_cache_free( ipv4_packet_task_cache, ptask );
+		}
+		list_for_each_entry_safe( ptask, pt_next, &priv->sent_list, packet_list ) {
+			dev_kfree_skb_any ( ptask->skb );
+			kmem_cache_free( ipv4_packet_task_cache, ptask );
+		}
+		fw_debug ( "lists emptied\n" );
+		list_for_each_entry( node, &card->ipv4_nodes, ipv4_nodes ) {
+			if ( node->pdg_size ) {
+				list_for_each_entry_safe( pd, pd_next, &node->pdg_list, pdg_list )
+					ipv4_pd_delete ( pd );
+				node->pdg_size = 0;
+			}
+			node->fifo = INVALID_FIFO_ADDR;
+		}
+		fw_debug ( "nodes cleaned up\n" );
+		free_netdev ( netdev );
+		card->netdev = NULL;
+		fw_debug ( "done\n" );
+	}
+	return 0;
+}
+
+static void ipv4_update ( struct fw_unit *unit ) {
+	struct fw_device *device;
+	struct fw_card *card;
+
+	fw_debug ( "ipv4_update unit %p\n", unit );
+	device = fw_device ( unit->device.parent );
+	card = device->card;
+	if ( ! device->is_local ) {
+		struct ipv4_node *node;
+		u64 guid;
+		struct net_device *netdev;
+		struct ipv4_priv *priv;
+
+		netdev = card->netdev;
+		if ( netdev ) {
+			priv = netdev_priv ( netdev );
+			guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
+			node = ipv4_node_find_by_guid ( priv, guid );
+			if ( ! node ) {
+				fw_error ( "ipv4_update: no node for device %llx\n", guid );
+				return;
+			}
+			fw_debug ( "Non-local, updating remote node entry for guid %llx old generation %x, old nodeid %x\n", guid, node->generation, node->nodeid );
+			node->generation = device->generation;
+			rmb();
+			node->nodeid = device->node_id;
+			fw_debug ( "New generation %x, new nodeid %x\n", node->generation, node->nodeid );
+		} else
+			fw_error ( "nonlocal, but no netdev?  How can that be?\n" );
+	} else {
+		/* FIXME: What do we need to do on bus reset? */
+		fw_debug ( "Local, doing nothing\n" );
+	}
+}
+
+static struct fw_driver ipv4_driver = {
+	.driver = {
+		.owner = THIS_MODULE,
+		.name = ipv4_driver_name,
+		.bus = &fw_bus_type,
+		.probe = ipv4_probe,
+		.remove = ipv4_remove,
+	},
+	.update = ipv4_update,
+	.id_table = ipv4_id_table,
+};
+
+static int __init ipv4_init ( void ) {
+	int added;
+
+	added = fw_core_add_descriptor ( &ipv4_unit_directory );
+	if ( added < 0 )
+		fw_error ( "Failed to add descriptor" );
+	ipv4_packet_task_cache = kmem_cache_create("packet_task",
+ sizeof(struct ipv4_packet_task), 0, 0, NULL);
+	fw_debug("Adding ipv4 module\n" );
+	return driver_register ( &ipv4_driver.driver );
+}
+
+static void __exit ipv4_cleanup ( void ) {
+	fw_core_remove_descriptor ( &ipv4_unit_directory );
+	fw_debug("Removing ipv4 module\n" );
+	driver_unregister ( &ipv4_driver.driver );
+}
+
+module_init(ipv4_init);
+module_exit(ipv4_cleanup);
diff --git a/include/linux/firewire.h b/include/linux/firewire.h
index e584b7215e8b..d44f47d3b2d9 100644
--- a/include/linux/firewire.h
+++ b/include/linux/firewire.h
@@ -3,6 +3,7 @@
 
 #include <linux/completion.h>
 #include <linux/device.h>
+#include <linux/dma-mapping.h>
 #include <linux/kernel.h>
 #include <linux/kref.h>
 #include <linux/list.h>
@@ -130,6 +131,13 @@ struct fw_card {
 	bool broadcast_channel_allocated;
 	u32 broadcast_channel;
 	u32 topology_map[(CSR_TOPOLOGY_MAP_END - CSR_TOPOLOGY_MAP) / 4];
+	/* Only non-NULL if firewire-ipv4 is active on this card. */
+	void *netdev;
+	/*
+	 * The nodes get probed before the card, so we need a place to store
+	 * them independent of card->netdev
+	 */
+	struct list_head ipv4_nodes;
 };
 
 static inline struct fw_card *fw_card_get(struct fw_card *card)
@@ -355,4 +363,90 @@ int fw_run_transaction(struct fw_card *card, int tcode, int destination_id,
 		       int generation, int speed, unsigned long long offset,
 		       void *payload, size_t length);
 
+static inline int fw_stream_packet_destination_id(int tag, int channel, int sy)
+{
+	return tag << 14 | channel << 8 | sy;
+}
+
+struct fw_descriptor {
+	struct list_head link;
+	size_t length;
+	u32 immediate;
+	u32 key;
+	const u32 *data;
+};
+
+int fw_core_add_descriptor(struct fw_descriptor *desc);
+void fw_core_remove_descriptor(struct fw_descriptor *desc);
+
+/*
+ * The iso packet format allows for an immediate header/payload part
+ * stored in 'header' immediately after the packet info plus an
+ * indirect payload part that is pointer to by the 'payload' field.
+ * Applications can use one or the other or both to implement simple
+ * low-bandwidth streaming (e.g. audio) or more advanced
+ * scatter-gather streaming (e.g. assembling video frame automatically).
+ */
+struct fw_iso_packet {
+	u16 payload_length;	/* Length of indirect payload. */
+	u32 interrupt:1;	/* Generate interrupt on this packet */
+	u32 skip:1;		/* Set to not send packet at all. */
+	u32 tag:2;
+	u32 sy:4;
+	u32 header_length:8;	/* Length of immediate header. */
+	u32 header[0];
+};
+
+#define FW_ISO_CONTEXT_TRANSMIT	0
+#define FW_ISO_CONTEXT_RECEIVE	1
+
+#define FW_ISO_CONTEXT_MATCH_TAG0	 1
+#define FW_ISO_CONTEXT_MATCH_TAG1	 2
+#define FW_ISO_CONTEXT_MATCH_TAG2	 4
+#define FW_ISO_CONTEXT_MATCH_TAG3	 8
+#define FW_ISO_CONTEXT_MATCH_ALL_TAGS	15
+
+/*
+ * An iso buffer is just a set of pages mapped for DMA in the
+ * specified direction.  Since the pages are to be used for DMA, they
+ * are not mapped into the kernel virtual address space.  We store the
+ * DMA address in the page private. The helper function
+ * fw_iso_buffer_map() will map the pages into a given vma.
+ */
+struct fw_iso_buffer {
+	enum dma_data_direction direction;
+	struct page **pages;
+	int page_count;
+};
+
+int fw_iso_buffer_init(struct fw_iso_buffer *buffer, struct fw_card *card,
+		       int page_count, enum dma_data_direction direction);
+void fw_iso_buffer_destroy(struct fw_iso_buffer *buffer, struct fw_card *card);
+
+struct fw_iso_context;
+typedef void (*fw_iso_callback_t)(struct fw_iso_context *context,
+				  u32 cycle, size_t header_length,
+				  void *header, void *data);
+struct fw_iso_context {
+	struct fw_card *card;
+	int type;
+	int channel;
+	int speed;
+	size_t header_size;
+	fw_iso_callback_t callback;
+	void *callback_data;
+};
+
+struct fw_iso_context *fw_iso_context_create(struct fw_card *card,
+		int type, int channel, int speed, size_t header_size,
+		fw_iso_callback_t callback, void *callback_data);
+int fw_iso_context_queue(struct fw_iso_context *ctx,
+			 struct fw_iso_packet *packet,
+			 struct fw_iso_buffer *buffer,
+			 unsigned long payload);
+int fw_iso_context_start(struct fw_iso_context *ctx,
+			 int cycle, int sync, int tags);
+int fw_iso_context_stop(struct fw_iso_context *ctx);
+void fw_iso_context_destroy(struct fw_iso_context *ctx);
+
 #endif /* _LINUX_FIREWIRE_H */
-- 
cgit v1.2.3-59-g8ed1b


From b9530fd6c3f057bda258c8e2631ad1a25959f4a2 Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Sun, 7 Jun 2009 22:57:53 +0200
Subject: firewire: net: add Kconfig item, rename driver

The driver is now called firewire-net.  It might implement the transport
of other networking protocols in the future, notably IPv6 per RFC 3146.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/Kconfig   |   12 +
 drivers/firewire/Makefile  |    6 +-
 drivers/firewire/fw-ipv4.c | 1819 --------------------------------------------
 drivers/firewire/net.c     | 1819 ++++++++++++++++++++++++++++++++++++++++++++
 drivers/ieee1394/Kconfig   |    2 +-
 5 files changed, 1835 insertions(+), 1823 deletions(-)
 delete mode 100644 drivers/firewire/fw-ipv4.c
 create mode 100644 drivers/firewire/net.c

diff --git a/drivers/firewire/Kconfig b/drivers/firewire/Kconfig
index 450902438208..d6b1721e52ab 100644
--- a/drivers/firewire/Kconfig
+++ b/drivers/firewire/Kconfig
@@ -77,3 +77,15 @@ config FIREWIRE_SBP2
 
 	  You should also enable support for disks, CD-ROMs, etc. in the SCSI
 	  configuration section.
+
+config FIREWIRE_NET
+	tristate "IP networking over 1394"
+	depends on FIREWIRE && INET
+	help
+	  This enables IPv4 over IEEE 1394, providing IP connectivity with
+	  other implementations of RFC 2734 as found on several operating
+	  systems.  Multicast support is currently limited.
+
+	  To compile this driver as a module, say M here:  The module will be
+	  called firewire-net.  It replaces eth1394 of the classic IEEE 1394
+	  stack.
diff --git a/drivers/firewire/Makefile b/drivers/firewire/Makefile
index 31edf30c558d..a8f9bb6d9fdf 100644
--- a/drivers/firewire/Makefile
+++ b/drivers/firewire/Makefile
@@ -6,9 +6,9 @@ firewire-core-y += core-card.o core-cdev.o core-device.o \
                    core-iso.o core-topology.o core-transaction.o
 firewire-ohci-y += ohci.o
 firewire-sbp2-y += sbp2.o
-firewire-ipv4-y += fw-ipv4.o
+firewire-net-y  += net.o
 
-obj-$(CONFIG_FIREWIRE) += firewire-core.o
+obj-$(CONFIG_FIREWIRE)      += firewire-core.o
 obj-$(CONFIG_FIREWIRE_OHCI) += firewire-ohci.o
 obj-$(CONFIG_FIREWIRE_SBP2) += firewire-sbp2.o
-obj-$(CONFIG_FIREWIRE_IPV4) += firewire-ipv4.o
+obj-$(CONFIG_FIREWIRE_NET)  += firewire-net.o
diff --git a/drivers/firewire/fw-ipv4.c b/drivers/firewire/fw-ipv4.c
deleted file mode 100644
index 4de6dbb95f0c..000000000000
--- a/drivers/firewire/fw-ipv4.c
+++ /dev/null
@@ -1,1819 +0,0 @@
-/*
- * IPv4 over IEEE 1394, per RFC 2734
- *
- * Copyright (C) 2009 Jay Fenlason <fenlason@redhat.com>
- *
- * based on eth1394 by Ben Collins et al
- */
-
-#include <linux/device.h>
-#include <linux/ethtool.h>
-#include <linux/firewire.h>
-#include <linux/firewire-constants.h>
-#include <linux/highmem.h>
-#include <linux/in.h>
-#include <linux/ip.h>
-#include <linux/mod_devicetable.h>
-#include <linux/module.h>
-#include <linux/moduleparam.h>
-#include <linux/netdevice.h>
-#include <linux/skbuff.h>
-
-#include <asm/unaligned.h>
-#include <net/arp.h>
-
-/* Things to potentially make runtime cofigurable */
-/* must be at least as large as our maximum receive size */
-#define FIFO_SIZE 4096
-/* Network timeout in glibbles */
-#define IPV4_TIMEOUT       100000
-
-/* Runitme configurable paramaters */
-static int ipv4_mpd = 25;
-static int ipv4_max_xmt = 0;
-/* 16k for receiving arp and broadcast packets.  Enough? */
-static int ipv4_iso_page_count = 4;
-
-MODULE_AUTHOR("Jay Fenlason (fenlason@redhat.com)");
-MODULE_DESCRIPTION("Firewire IPv4 Driver (IPv4-over-IEEE1394 as per RFC 2734)");
-MODULE_LICENSE("GPL");
-MODULE_DEVICE_TABLE(ieee1394, ipv4_id_table);
-module_param_named(max_partial_datagrams, ipv4_mpd, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(max_partial_datagrams, "Maximum number of received"
- " incomplete fragmented datagrams (default = 25).");
-
-/* Max xmt is useful for forcing fragmentation, which makes testing easier. */
-module_param_named(max_transmit, ipv4_max_xmt, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(max_transmit, "Maximum datagram size to transmit"
- " (larger datagrams will be fragmented) (default = 0 (use hardware defaults).");
-
-/* iso page count controls how many pages will be used for receiving broadcast packets. */
-module_param_named(iso_pages, ipv4_iso_page_count, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(iso_pages, "Number of pages to use for receiving broadcast packets"
- " (default = 4).");
-
-/* uncomment this line to do debugging */
-#define fw_debug(s, args...) printk(KERN_DEBUG KBUILD_MODNAME ": " s, ## args)
-
-/* comment out these lines to do debugging. */
-/* #undef fw_debug */
-/* #define fw_debug(s...) */
-/* #define print_hex_dump(l...) */
-
-/* Define a fake hardware header format for the networking core.  Note that
- * header size cannot exceed 16 bytes as that is the size of the header cache.
- * Also, we do not need the source address in the header so we omit it and
- * keep the header to under 16 bytes */
-#define IPV4_ALEN (8)
-/* This must equal sizeof(struct ipv4_ether_hdr) */
-#define IPV4_HLEN (10)
-
-/* FIXME: what's a good size for this? */
-#define INVALID_FIFO_ADDR (u64)~0ULL
-
-/* Things specified by standards */
-#define BROADCAST_CHANNEL 31
-
-#define S100_BUFFER_SIZE 512
-#define MAX_BUFFER_SIZE 4096
-
-#define IPV4_GASP_SPECIFIER_ID	0x00005EU
-#define IPV4_GASP_VERSION	0x00000001U
-
-#define IPV4_GASP_OVERHEAD (2 * sizeof(u32)) /* for GASP header */
-
-#define IPV4_UNFRAG_HDR_SIZE	sizeof(u32)
-#define IPV4_FRAG_HDR_SIZE	(2 * sizeof(u32))
-#define IPV4_FRAG_OVERHEAD	sizeof(u32)
-
-#define ALL_NODES (0xffc0 | 0x003f)
-
-#define IPV4_HDR_UNFRAG		0	/* unfragmented		*/
-#define IPV4_HDR_FIRSTFRAG	1	/* first fragment	*/
-#define IPV4_HDR_LASTFRAG	2	/* last fragment	*/
-#define IPV4_HDR_INTFRAG	3	/* interior fragment	*/
-
-/* Our arp packet (ARPHRD_IEEE1394) */
-/* FIXME: note that this is probably bogus on weird-endian machines */
-struct ipv4_arp {
-	u16 hw_type;		/* 0x0018	*/
-	u16 proto_type;		/* 0x0806       */
-	u8 hw_addr_len;		/* 16		*/
-	u8 ip_addr_len;         /* 4		*/
-	u16 opcode;	        /* ARP Opcode	*/
-	/* Above is exactly the same format as struct arphdr */
-
-	u64 s_uniq_id;		/* Sender's 64bit EUI			*/
-	u8 max_rec;             /* Sender's max packet size		*/
-	u8 sspd;		/* Sender's max speed			*/
-	u16 fifo_hi;            /* hi 16bits of sender's FIFO addr	*/
-	u32 fifo_lo;            /* lo 32bits of sender's FIFO addr	*/
-	u32 sip;		/* Sender's IP Address			*/
-	u32 tip;		/* IP Address of requested hw addr	*/
-} __attribute__((packed));
-
-struct ipv4_ether_hdr {
-	unsigned char	h_dest[IPV4_ALEN];	/* destination address */
-	unsigned short  h_proto;                /* packet type ID field */
-}  __attribute__((packed));
-
-static inline struct ipv4_ether_hdr *ipv4_ether_hdr(const struct sk_buff *skb)
-{
-	return (struct ipv4_ether_hdr *)skb_mac_header(skb);
-}
-
-enum ipv4_tx_type {
-	IPV4_UNKNOWN = 0,
-	IPV4_GASP = 1,
-	IPV4_WRREQ = 2,
-};
-
-enum ipv4_broadcast_state {
-	IPV4_BROADCAST_ERROR,
-	IPV4_BROADCAST_RUNNING,
-	IPV4_BROADCAST_STOPPED,
-};
-
-#define ipv4_get_hdr_lf(h)		(((h)->w0&0xC0000000)>>30)
-#define ipv4_get_hdr_ether_type(h)	(((h)->w0&0x0000FFFF)    )
-#define ipv4_get_hdr_dg_size(h)		(((h)->w0&0x0FFF0000)>>16)
-#define ipv4_get_hdr_fg_off(h)		(((h)->w0&0x00000FFF)    )
-#define ipv4_get_hdr_dgl(h)		(((h)->w1&0xFFFF0000)>>16)
-
-#define ipv4_set_hdr_lf(lf)		(( lf)<<30)
-#define ipv4_set_hdr_ether_type(et)	(( et)    )
-#define ipv4_set_hdr_dg_size(dgs)	((dgs)<<16)
-#define ipv4_set_hdr_fg_off(fgo)	((fgo)    )
-
-#define ipv4_set_hdr_dgl(dgl)		((dgl)<<16)
-
-struct ipv4_hdr {
-	u32 w0;
-	u32 w1;
-};
-
-static inline void ipv4_make_uf_hdr( struct ipv4_hdr *hdr, unsigned ether_type) {
-	hdr->w0 = ipv4_set_hdr_lf(IPV4_HDR_UNFRAG)
-		   |ipv4_set_hdr_ether_type(ether_type);
-	fw_debug ( "Setting unfragmented header %p to %x\n", hdr, hdr->w0 );
-}
-
-static inline void ipv4_make_ff_hdr ( struct ipv4_hdr *hdr, unsigned ether_type, unsigned dg_size, unsigned dgl ) {
-	hdr->w0 = ipv4_set_hdr_lf(IPV4_HDR_FIRSTFRAG)
-		   |ipv4_set_hdr_dg_size(dg_size)
-		   |ipv4_set_hdr_ether_type(ether_type);
-	hdr->w1 = ipv4_set_hdr_dgl(dgl);
-	fw_debug ( "Setting fragmented header %p to first_frag %x,%x (et %x, dgs %x, dgl %x)\n", hdr, hdr->w0, hdr->w1,
- ether_type, dg_size, dgl );
-}
-
-static inline void ipv4_make_sf_hdr ( struct ipv4_hdr *hdr, unsigned lf, unsigned dg_size, unsigned fg_off, unsigned dgl) {
-	hdr->w0 = ipv4_set_hdr_lf(lf)
-		 |ipv4_set_hdr_dg_size(dg_size)
-		 |ipv4_set_hdr_fg_off(fg_off);
-	hdr->w1 = ipv4_set_hdr_dgl(dgl);
-	fw_debug ( "Setting fragmented header %p to %x,%x (lf %x, dgs %x, fo %x dgl %x)\n",
- hdr, hdr->w0, hdr->w1,
- lf, dg_size, fg_off, dgl );
-}
-
-/* End of IP1394 headers */
-
-/* Fragment types */
-#define ETH1394_HDR_LF_UF	0	/* unfragmented		*/
-#define ETH1394_HDR_LF_FF	1	/* first fragment	*/
-#define ETH1394_HDR_LF_LF	2	/* last fragment	*/
-#define ETH1394_HDR_LF_IF	3	/* interior fragment	*/
-
-#define IP1394_HW_ADDR_LEN	16	/* As per RFC		*/
-
-/* This list keeps track of what parts of the datagram have been filled in */
-struct ipv4_fragment_info {
-        struct list_head fragment_info;
-	u16 offset;
-	u16 len;
-};
-
-struct ipv4_partial_datagram {
-	struct list_head pdg_list;
-	struct list_head fragment_info;
-	struct sk_buff *skb;
-	/* FIXME Why not use skb->data? */
-	char *pbuf;
-	u16 datagram_label;
-	u16 ether_type;
-	u16 datagram_size;
-};
-
-/*
- * We keep one of these for each IPv4 capable device attached to a fw_card.
- * The list of them is stored in the fw_card structure rather than in the
- * ipv4_priv because the remote IPv4 nodes may be probed before the card is,
- * so we need a place to store them before the ipv4_priv structure is
- * allocated.
- */
-struct ipv4_node {
-	struct list_head ipv4_nodes;
-	/* guid of the remote node */
-	u64 guid;
-	/* FIFO address to transmit datagrams to, or INVALID_FIFO_ADDR */
-	u64 fifo;
-
-	spinlock_t pdg_lock;	/* partial datagram lock		*/
-	/* List of partial datagrams received from this node */
-	struct list_head pdg_list;
-	/* Number of entries in pdg_list at the moment */
-	unsigned pdg_size;
-
-	/* max payload to transmit to this remote node */
-	/* This already includes the IPV4_FRAG_HDR_SIZE overhead */
-	u16 max_payload;
-	/* outgoing datagram label */
-	u16 datagram_label;
-	/* Current node_id of the remote node */
-	u16 nodeid;
-	/* current generation of the remote node */
-	u8 generation;
-	/* max speed that this node can receive at */
-	u8 xmt_speed;
-};
-
-struct ipv4_priv {
-	spinlock_t lock;
-
-	enum ipv4_broadcast_state broadcast_state;
-	struct fw_iso_context *broadcast_rcv_context;
-	struct fw_iso_buffer broadcast_rcv_buffer;
-	void **broadcast_rcv_buffer_ptrs;
-	unsigned broadcast_rcv_next_ptr;
-	unsigned num_broadcast_rcv_ptrs;
-	unsigned rcv_buffer_size;
-	/*
-	 * This value is the maximum unfragmented datagram size that can be
-	 * sent by the hardware.  It already has the GASP overhead and the
-	 * unfragmented datagram header overhead calculated into it.
-	 */
-	unsigned broadcast_xmt_max_payload;
-	u16 broadcast_xmt_datagramlabel;
-
-	/*
-	 * The csr address that remote nodes must send datagrams to for us to
-	 * receive them.
-	 */
-	struct fw_address_handler handler;
-	u64 local_fifo;
-
-	/* Wake up to xmt	 */
-        /* struct work_struct wake;*/
-	/* List of packets to be sent */
-	struct list_head packet_list;
-	/*
-	 * List of packets that were broadcasted.  When we get an ISO interrupt
-	 * one of them has been sent
-	 */
-	struct list_head broadcasted_list;
-	/* List of packets that have been sent but not yet acked */
-	struct list_head sent_list;
-
-	struct fw_card *card;
-};
-
-/* This is our task struct. It's used for the packet complete callback.  */
-struct ipv4_packet_task {
-	/*
-	 * ptask can actually be on priv->packet_list, priv->broadcasted_list,
-	 * or priv->sent_list depending on its current state.
-	 */
-	struct list_head packet_list;
-	struct fw_transaction transaction;
-	struct ipv4_hdr hdr;
-	struct sk_buff *skb;
-	struct ipv4_priv *priv;
-	enum ipv4_tx_type tx_type;
-	int outstanding_pkts;
-	unsigned max_payload;
-	u64 fifo_addr;
-	u16 dest_node;
-	u8 generation;
-	u8 speed;
-};
-
-static struct kmem_cache *ipv4_packet_task_cache;
-
-static const char ipv4_driver_name[] = "firewire-ipv4";
-
-static const struct ieee1394_device_id ipv4_id_table[] = {
-	{
-		.match_flags  = IEEE1394_MATCH_SPECIFIER_ID |
-				IEEE1394_MATCH_VERSION,
-		.specifier_id = IPV4_GASP_SPECIFIER_ID,
-		.version      = IPV4_GASP_VERSION,
-	},
-	{ }
-};
-
-static u32 ipv4_unit_directory_data[] = {
-	0x00040000,					/* unit directory */
-	0x12000000 | IPV4_GASP_SPECIFIER_ID,	/* specifier ID */
-	0x81000003,					/* text descriptor */
-	0x13000000 | IPV4_GASP_VERSION,		/* version */
-	0x81000005,					/* text descriptor */
-
-	0x00030000,					/* Three quadlets */
-	0x00000000,					/* Text */
-	0x00000000,					/* Language 0 */
-	0x49414e41,					/* I A N A */
-	0x00030000,					/* Three quadlets */
-	0x00000000,					/* Text */
-	0x00000000,					/* Language 0 */
-	0x49507634,					/* I P v 4 */
-};
-
-static struct fw_descriptor ipv4_unit_directory = {
-	.length = ARRAY_SIZE(ipv4_unit_directory_data),
-	.key = 0xd1000000,
-	.data = ipv4_unit_directory_data
-};
-
-static int ipv4_send_packet(struct ipv4_packet_task *ptask );
-
-/* ------------------------------------------------------------------ */
-/******************************************
- * HW Header net device functions
- ******************************************/
-  /* These functions have been adapted from net/ethernet/eth.c */
-
-/* Create a fake MAC header for an arbitrary protocol layer.
- * saddr=NULL means use device source address
- * daddr=NULL means leave destination address (eg unresolved arp). */
-
-static int ipv4_header ( struct sk_buff *skb, struct net_device *dev,
-		       unsigned short type, const void *daddr,
-		       const void *saddr, unsigned len) {
-	struct ipv4_ether_hdr *eth;
-
-	eth = (struct ipv4_ether_hdr *)skb_push(skb, sizeof(*eth));
-	eth->h_proto = htons(type);
-
-	if (dev->flags & (IFF_LOOPBACK | IFF_NOARP)) {
-		memset(eth->h_dest, 0, dev->addr_len);
-		return dev->hard_header_len;
-	}
-
-	if (daddr) {
-		memcpy(eth->h_dest, daddr, dev->addr_len);
-		return dev->hard_header_len;
-	}
-
-	return -dev->hard_header_len;
-}
-
-/* Rebuild the faked MAC header. This is called after an ARP
- * (or in future other address resolution) has completed on this
- * sk_buff. We now let ARP fill in the other fields.
- *
- * This routine CANNOT use cached dst->neigh!
- * Really, it is used only when dst->neigh is wrong.
- */
-
-static int ipv4_rebuild_header(struct sk_buff *skb)
-{
-	struct ipv4_ether_hdr *eth;
-
-	eth = (struct ipv4_ether_hdr *)skb->data;
-	if (eth->h_proto == htons(ETH_P_IP))
-		return arp_find((unsigned char *)&eth->h_dest, skb);
-
-	fw_notify ( "%s: unable to resolve type %04x addresses\n",
-		   skb->dev->name,ntohs(eth->h_proto) );
-	return 0;
-}
-
-static int ipv4_header_cache(const struct neighbour *neigh, struct hh_cache *hh) {
-	unsigned short type = hh->hh_type;
-	struct net_device *dev;
-	struct ipv4_ether_hdr *eth;
-
-	if (type == htons(ETH_P_802_3))
-		return -1;
-	dev = neigh->dev;
-	eth = (struct ipv4_ether_hdr *)((u8 *)hh->hh_data + 16 - sizeof(*eth));
-	eth->h_proto = type;
-	memcpy(eth->h_dest, neigh->ha, dev->addr_len);
-
-	hh->hh_len = IPV4_HLEN;
-	return 0;
-}
-
-/* Called by Address Resolution module to notify changes in address. */
-static void ipv4_header_cache_update(struct hh_cache *hh, const struct net_device *dev, const unsigned char * haddr ) {
-	memcpy((u8 *)hh->hh_data + 16 - IPV4_HLEN, haddr, dev->addr_len);
-}
-
-static int ipv4_header_parse(const struct sk_buff *skb, unsigned char *haddr) {
-	memcpy(haddr, skb->dev->dev_addr, IPV4_ALEN);
-	return IPV4_ALEN;
-}
-
-static const struct header_ops ipv4_header_ops = {
-	.create         = ipv4_header,
-	.rebuild        = ipv4_rebuild_header,
-	.cache		= ipv4_header_cache,
-	.cache_update	= ipv4_header_cache_update,
-	.parse          = ipv4_header_parse,
-};
-
-/* ------------------------------------------------------------------ */
-
-/* FIXME: is this correct for all cases? */
-static bool ipv4_frag_overlap(struct ipv4_partial_datagram *pd, unsigned offset, unsigned len)
-{
-        struct ipv4_fragment_info *fi;
-	unsigned end = offset + len;
-
-	list_for_each_entry(fi, &pd->fragment_info, fragment_info) {
-		if (offset < fi->offset + fi->len && end > fi->offset) {
-			fw_debug ( "frag_overlap pd %p fi %p (%x@%x) with %x@%x\n", pd, fi, fi->len, fi->offset, len, offset );
-			return true;
-		}
-	}
-	fw_debug ( "frag_overlap %p does not overlap with %x@%x\n", pd, len, offset );
-	return false;
-}
-
-/* Assumes that new fragment does not overlap any existing fragments */
-static struct ipv4_fragment_info *ipv4_frag_new ( struct ipv4_partial_datagram *pd, unsigned offset, unsigned len ) {
-	struct ipv4_fragment_info *fi, *fi2, *new;
-	struct list_head *list;
-
-	fw_debug ( "frag_new pd %p %x@%x\n", pd, len, offset );
-	list = &pd->fragment_info;
-	list_for_each_entry(fi, &pd->fragment_info, fragment_info) {
-		if (fi->offset + fi->len == offset) {
-			/* The new fragment can be tacked on to the end */
-			/* Did the new fragment plug a hole? */
-			fi2 = list_entry(fi->fragment_info.next, struct ipv4_fragment_info, fragment_info);
-			if (fi->offset + fi->len == fi2->offset) {
-				fw_debug ( "pd %p: hole filling %p (%x@%x) and %p(%x@%x): now %x@%x\n", pd, fi, fi->len, fi->offset,
-				fi2, fi2->len, fi2->offset, fi->len + len + fi2->len, fi->offset );
-				/* glue fragments together */
-				fi->len += len + fi2->len;
-				list_del(&fi2->fragment_info);
-				kfree(fi2);
-			} else {
-				fw_debug ( "pd %p: extending %p from %x@%x to %x@%x\n", pd, fi, fi->len, fi->offset, fi->len+len, fi->offset );
-				fi->len += len;
-			}
-			return fi;
-		}
-		if (offset + len == fi->offset) {
-			/* The new fragment can be tacked on to the beginning */
-			/* Did the new fragment plug a hole? */
-			fi2 = list_entry(fi->fragment_info.prev, struct ipv4_fragment_info, fragment_info);
-			if (fi2->offset + fi2->len == fi->offset) {
-				/* glue fragments together */
-				fw_debug ( "pd %p: extending %p and merging with %p from %x@%x to %x@%x\n",
- pd, fi2, fi, fi2->len, fi2->offset, fi2->len + fi->len + len, fi2->offset );
-				fi2->len += fi->len + len;
-				list_del(&fi->fragment_info);
-				kfree(fi);
-				return fi2;
-			}
-			fw_debug ( "pd %p: extending %p from %x@%x to %x@%x\n", pd, fi, fi->len, fi->offset, offset, fi->len + len );
-			fi->offset = offset;
-			fi->len += len;
-			return fi;
-		}
-		if (offset > fi->offset + fi->len) {
-			list = &fi->fragment_info;
-			break;
-		}
-		if (offset + len < fi->offset) {
-			list = fi->fragment_info.prev;
-			break;
-		}
-	}
-
-	new = kmalloc(sizeof(*new), GFP_ATOMIC);
-	if (!new) {
-		fw_error ( "out of memory in fragment handling!\n" );
-		return NULL;
-	}
-
-	new->offset = offset;
-	new->len = len;
-	list_add(&new->fragment_info, list);
-	fw_debug ( "pd %p: new frag %p %x@%x\n", pd, new, new->len, new->offset );
-	list_for_each_entry( fi, &pd->fragment_info, fragment_info )
-		fw_debug ( "fi %p %x@%x\n", fi, fi->len, fi->offset );
-	return new;
-}
-
-/* ------------------------------------------------------------------ */
-
-static struct ipv4_partial_datagram *ipv4_pd_new(struct net_device *netdev,
- struct ipv4_node *node, u16 datagram_label, unsigned dg_size, u32 *frag_buf,
- unsigned frag_off, unsigned frag_len) {
-	struct ipv4_partial_datagram *new;
-	struct ipv4_fragment_info *fi;
-
-	new = kmalloc(sizeof(*new), GFP_ATOMIC);
-	if (!new)
-		goto fail;
-	INIT_LIST_HEAD(&new->fragment_info);
-	fi = ipv4_frag_new ( new, frag_off, frag_len);
-	if ( fi == NULL )
-		goto fail_w_new;
-	new->datagram_label = datagram_label;
-	new->datagram_size = dg_size;
-	new->skb = dev_alloc_skb(dg_size + netdev->hard_header_len + 15);
-	if ( new->skb == NULL )
-		goto fail_w_fi;
-	skb_reserve(new->skb, (netdev->hard_header_len + 15) & ~15);
-	new->pbuf = skb_put(new->skb, dg_size);
-	memcpy(new->pbuf + frag_off, frag_buf, frag_len);
-	list_add_tail(&new->pdg_list, &node->pdg_list);
-	fw_debug ( "pd_new: new pd %p { dgl %u, dg_size %u, skb %p, pbuf %p } on node %p\n",
- new, new->datagram_label, new->datagram_size, new->skb, new->pbuf, node );
-	return new;
-
-fail_w_fi:
-	kfree(fi);
-fail_w_new:
-	kfree(new);
-fail:
-	fw_error("ipv4_pd_new: no memory\n");
-	return NULL;
-}
-
-static struct ipv4_partial_datagram *ipv4_pd_find(struct ipv4_node *node, u16 datagram_label) {
-	struct ipv4_partial_datagram *pd;
-
-	list_for_each_entry(pd, &node->pdg_list, pdg_list) {
-	        if ( pd->datagram_label == datagram_label ) {
-			fw_debug ( "pd_find(node %p, label %u): pd %p\n", node, datagram_label, pd );
-			return pd;
-		}
-	}
-	fw_debug ( "pd_find(node %p, label %u) no entry\n", node, datagram_label );
-	return NULL;
-}
-
-
-static void ipv4_pd_delete ( struct ipv4_partial_datagram *old ) {
-	struct ipv4_fragment_info *fi, *n;
-
-	fw_debug ( "pd_delete %p\n", old );
-	list_for_each_entry_safe(fi, n, &old->fragment_info, fragment_info) {
-		fw_debug ( "Freeing fi %p\n", fi );
-		kfree(fi);
-	}
-	list_del(&old->pdg_list);
-	dev_kfree_skb_any(old->skb);
-	kfree(old);
-}
-
-static bool ipv4_pd_update ( struct ipv4_node *node, struct ipv4_partial_datagram *pd,
- u32 *frag_buf, unsigned frag_off, unsigned frag_len) {
-	fw_debug ( "pd_update node %p, pd %p, frag_buf %p, %x@%x\n", node, pd, frag_buf, frag_len, frag_off );
-	if ( ipv4_frag_new ( pd, frag_off, frag_len ) == NULL)
-		return false;
-	memcpy(pd->pbuf + frag_off, frag_buf, frag_len);
-
-	/*
-	 * Move list entry to beginnig of list so that oldest partial
-	 * datagrams percolate to the end of the list
-	 */
-	list_move_tail(&pd->pdg_list, &node->pdg_list);
-	fw_debug ( "New pd list:\n" );
-	list_for_each_entry ( pd, &node->pdg_list, pdg_list ) {
-		fw_debug ( "pd %p\n", pd );
-	}
-	return true;
-}
-
-static bool ipv4_pd_is_complete ( struct ipv4_partial_datagram *pd ) {
-	struct ipv4_fragment_info *fi;
-	bool ret;
-
-	fi = list_entry(pd->fragment_info.next, struct ipv4_fragment_info, fragment_info);
-
-	ret = (fi->len == pd->datagram_size);
-	fw_debug ( "pd_is_complete (pd %p, dgs %x): fi %p (%x@%x) %s\n", pd, pd->datagram_size, fi, fi->len, fi->offset, ret ? "yes" : "no" );
-	return ret;
-}
-
-/* ------------------------------------------------------------------ */
-
-static int ipv4_node_new ( struct fw_card *card, struct fw_device *device ) {
-	struct ipv4_node *node;
-
-	node = kmalloc ( sizeof(*node), GFP_KERNEL );
-	if ( ! node ) {
-		fw_error ( "allocate new node failed\n" );
-		return -ENOMEM;
-	}
-	node->guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
-	node->fifo = INVALID_FIFO_ADDR;
-	INIT_LIST_HEAD(&node->pdg_list);
-	spin_lock_init(&node->pdg_lock);
-	node->pdg_size = 0;
-	node->generation = device->generation;
-	rmb();
-	node->nodeid = device->node_id;
-	 /* FIXME what should it really be? */
-	node->max_payload = S100_BUFFER_SIZE - IPV4_UNFRAG_HDR_SIZE;
-	node->datagram_label = 0U;
-	node->xmt_speed = device->max_speed;
-	list_add_tail ( &node->ipv4_nodes, &card->ipv4_nodes );
-	fw_debug ( "node_new: %p { guid %016llx, generation %u, nodeid %x, max_payload %x, xmt_speed %x } added\n",
- node, (unsigned long long)node->guid, node->generation, node->nodeid, node->max_payload, node->xmt_speed );
-	return 0;
-}
-
-static struct ipv4_node *ipv4_node_find_by_guid(struct ipv4_priv *priv, u64 guid) {
-	struct ipv4_node *node;
-	unsigned long flags;
-
-	spin_lock_irqsave(&priv->lock, flags);
-	list_for_each_entry(node, &priv->card->ipv4_nodes, ipv4_nodes)
-		if (node->guid == guid) {
-			/* FIXME: lock the node first? */
-			spin_unlock_irqrestore ( &priv->lock, flags );
-			fw_debug ( "node_find_by_guid (%016llx) found %p\n", (unsigned long long)guid, node );
-			return node;
-		}
-
-	spin_unlock_irqrestore ( &priv->lock, flags );
-	fw_debug ( "node_find_by_guid (%016llx) not found\n", (unsigned long long)guid );
-	return NULL;
-}
-
-static struct ipv4_node *ipv4_node_find_by_nodeid(struct ipv4_priv *priv, u16 nodeid) {
-	struct ipv4_node *node;
-	unsigned long flags;
-
-	spin_lock_irqsave(&priv->lock, flags);
-	list_for_each_entry(node, &priv->card->ipv4_nodes, ipv4_nodes)
-		if (node->nodeid == nodeid) {
-			/* FIXME: lock the node first? */
-			spin_unlock_irqrestore ( &priv->lock, flags );
-			fw_debug ( "node_find_by_nodeid (%x) found %p\n", nodeid, node );
-			return node;
-		}
-	fw_debug ( "node_find_by_nodeid (%x) not found\n", nodeid );
-	spin_unlock_irqrestore ( &priv->lock, flags );
-	return NULL;
-}
-
-/* This is only complicated because we can't assume priv exists */
-static void ipv4_node_delete ( struct fw_card *card, struct fw_device *device ) {
-	struct net_device *netdev;
-	struct ipv4_priv *priv;
-	struct ipv4_node *node;
-	u64 guid;
-	unsigned long flags;
-	struct ipv4_partial_datagram *pd, *pd_next;
-
-	guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
-	netdev = card->netdev;
-	if ( netdev )
-		priv = netdev_priv ( netdev );
-	else
-		priv = NULL;
-	if ( priv )
-		spin_lock_irqsave ( &priv->lock, flags );
-	list_for_each_entry( node, &card->ipv4_nodes, ipv4_nodes ) {
-		if ( node->guid == guid ) {
-			list_del ( &node->ipv4_nodes );
-			list_for_each_entry_safe( pd, pd_next, &node->pdg_list, pdg_list )
-				ipv4_pd_delete ( pd );
-			break;
-		}
-	}
-	if ( priv )
-		spin_unlock_irqrestore ( &priv->lock, flags );
-}
-
-/* ------------------------------------------------------------------ */
-
-
-static int ipv4_finish_incoming_packet ( struct net_device *netdev,
- struct sk_buff *skb, u16 source_node_id, bool is_broadcast, u16 ether_type ) {
-	struct ipv4_priv *priv;
-	static u64 broadcast_hw = ~0ULL;
-	int status;
-	u64 guid;
-
-	fw_debug ( "ipv4_finish_incoming_packet(%p, %p, %x, %s, %x\n",
- netdev, skb, source_node_id, is_broadcast ? "true" : "false", ether_type );
-	priv = netdev_priv(netdev);
-	/* Write metadata, and then pass to the receive level */
-	skb->dev = netdev;
-	skb->ip_summed = CHECKSUM_UNNECESSARY;  /* don't check it */
-
-	/*
-	 * Parse the encapsulation header. This actually does the job of
-	 * converting to an ethernet frame header, as well as arp
-	 * conversion if needed. ARP conversion is easier in this
-	 * direction, since we are using ethernet as our backend.
-	 */
-	/*
-	 * If this is an ARP packet, convert it. First, we want to make
-	 * use of some of the fields, since they tell us a little bit
-	 * about the sending machine.
-	 */
-	if (ether_type == ETH_P_ARP) {
-		struct ipv4_arp *arp1394;
-		struct arphdr *arp;
-		unsigned char *arp_ptr;
-		u64 fifo_addr;
-		u8 max_rec;
-		u8 sspd;
-		u16 max_payload;
-		struct ipv4_node *node;
-		static const u16 ipv4_speed_to_max_payload[] = {
-			/* S100, S200, S400, S800, S1600, S3200 */
-			    512, 1024, 2048, 4096,  4096,  4096
-		};
-
-		/* fw_debug ( "ARP packet\n" ); */
-		arp1394 = (struct ipv4_arp *)skb->data;
-		arp = (struct arphdr *)skb->data;
-		arp_ptr = (unsigned char *)(arp + 1);
-		fifo_addr = (u64)ntohs(arp1394->fifo_hi) << 32 |
- ntohl(arp1394->fifo_lo);
-		max_rec = priv->card->max_receive;
-		if ( arp1394->max_rec < max_rec )
-			max_rec = arp1394->max_rec;
-		sspd = arp1394->sspd;
-		/*
-		 * Sanity check. MacOSX seems to be sending us 131 in this
-		 * field (atleast on my Panther G5). Not sure why.
-		 */
-		if (sspd > 5 ) {
-			fw_notify ( "sspd %x out of range\n", sspd );
-			sspd = 0;
-		}
-
-		max_payload = min(ipv4_speed_to_max_payload[sspd],
- (u16)(1 << (max_rec + 1))) - IPV4_UNFRAG_HDR_SIZE;
-
-		guid = be64_to_cpu(get_unaligned(&arp1394->s_uniq_id));
-		node = ipv4_node_find_by_guid(priv, guid);
-		if (!node) {
-			fw_notify ( "No node for ARP packet from %llx\n", guid );
-			goto failed_proto;
-		}
-		if ( node->nodeid != source_node_id || node->generation != priv->card->generation ) {
-			fw_notify ( "Internal error: node->nodeid (%x) != soucre_node_id (%x) or node->generation (%x) != priv->card->generation(%x)\n",
- node->nodeid, source_node_id, node->generation, priv->card->generation );
-			node->nodeid = source_node_id;
-			node->generation = priv->card->generation;
-		}
-
-		/* FIXME: for debugging */
-		if ( sspd > SCODE_400 )
-			sspd = SCODE_400;
-		/* Update our speed/payload/fifo_offset table */
-		/*
-		 * FIXME: this does not handle cases where two high-speed endpoints must use a slower speed because of
-		 * a lower speed hub between them.  We need to look at the actual topology map here.
-		 */
-		fw_debug ( "Setting node %p fifo %llx (was %llx), max_payload %x (was %x), speed %x (was %x)\n",
- node, fifo_addr, node->fifo, max_payload, node->max_payload, sspd, node->xmt_speed );
-		node->fifo =	fifo_addr;
-		node->max_payload = max_payload;
-		/*
-		 * Only allow speeds to go down from their initial value.
-		 * Otherwise a local node that can only do S400 or slower may
-		 * be told to transmit at S800 to a faster remote node.
-		 */
-		if ( node->xmt_speed > sspd )
-			node->xmt_speed = sspd;
-
-		/*
-		 * Now that we're done with the 1394 specific stuff, we'll
-		 * need to alter some of the data.  Believe it or not, all
-		 * that needs to be done is sender_IP_address needs to be
-		 * moved, the destination hardware address get stuffed
-		 * in and the hardware address length set to 8.
-		 *
-		 * IMPORTANT: The code below overwrites 1394 specific data
-		 * needed above so keep the munging of the data for the
-		 * higher level IP stack last.
-		 */
-
-		arp->ar_hln = 8;
-		arp_ptr += arp->ar_hln;		/* skip over sender unique id */
-		*(u32 *)arp_ptr = arp1394->sip; /* move sender IP addr */
-		arp_ptr += arp->ar_pln;		/* skip over sender IP addr */
-
-		if (arp->ar_op == htons(ARPOP_REQUEST))
-			memset(arp_ptr, 0, sizeof(u64));
-		else
-			memcpy(arp_ptr, netdev->dev_addr, sizeof(u64));
-	}
-
-	/* Now add the ethernet header. */
-	guid = cpu_to_be64(priv->card->guid);
-	if (dev_hard_header(skb, netdev, ether_type, is_broadcast ? &broadcast_hw : &guid, NULL,
- skb->len) >= 0) {
-		struct ipv4_ether_hdr *eth;
-		u16 *rawp;
-		__be16 protocol;
-
-		skb_reset_mac_header(skb);
-		skb_pull(skb, sizeof(*eth));
-		eth = ipv4_ether_hdr(skb);
-		if (*eth->h_dest & 1) {
-			if (memcmp(eth->h_dest, netdev->broadcast, netdev->addr_len) == 0) {
-				fw_debug ( "Broadcast\n" );
-				skb->pkt_type = PACKET_BROADCAST;
-			}
-#if 0
-			else
-				skb->pkt_type = PACKET_MULTICAST;
-#endif
-		} else {
-			if (memcmp(eth->h_dest, netdev->dev_addr, netdev->addr_len)) {
-				u64 a1, a2;
-
-				memcpy ( &a1, eth->h_dest, sizeof(u64));
-				memcpy ( &a2, netdev->dev_addr, sizeof(u64));
-				fw_debug ( "Otherhost %llx %llx %x\n", a1, a2, netdev->addr_len );
-				skb->pkt_type = PACKET_OTHERHOST;
-			}
-		}
-		if (ntohs(eth->h_proto) >= 1536) {
-			fw_debug ( " proto %x %x\n", eth->h_proto, ntohs(eth->h_proto) );
-			protocol = eth->h_proto;
-		} else {
-			rawp = (u16 *)skb->data;
-			if (*rawp == 0xFFFF) {
-				fw_debug ( "proto 802_3\n" );
-				protocol = htons(ETH_P_802_3);
-			} else {
-				fw_debug ( "proto 802_2\n" );
-				protocol = htons(ETH_P_802_2);
-			}
-		}
-		skb->protocol = protocol;
-	}
-	status = netif_rx(skb);
-	if ( status == NET_RX_DROP) {
-		netdev->stats.rx_errors++;
-		netdev->stats.rx_dropped++;
-	} else {
-		netdev->stats.rx_packets++;
-		netdev->stats.rx_bytes += skb->len;
-	}
-	if (netif_queue_stopped(netdev))
-		netif_wake_queue(netdev);
-	return 0;
-
- failed_proto:
-	netdev->stats.rx_errors++;
-	netdev->stats.rx_dropped++;
-	dev_kfree_skb_any(skb);
-	if (netif_queue_stopped(netdev))
-		netif_wake_queue(netdev);
-	netdev->last_rx = jiffies;
-	return 0;
-}
-
-/* ------------------------------------------------------------------ */
-
-static int ipv4_incoming_packet ( struct ipv4_priv *priv, u32 *buf, int len, u16 source_node_id, bool is_broadcast ) {
-	struct sk_buff *skb;
-	struct net_device *netdev;
-	struct ipv4_hdr hdr;
-	unsigned lf;
-	unsigned long flags;
-	struct ipv4_node *node;
-	struct ipv4_partial_datagram *pd;
-	int fg_off;
-	int dg_size;
-	u16 datagram_label;
-	int retval;
-	u16 ether_type;
-
-	fw_debug ( "ipv4_incoming_packet(%p, %p, %d, %x, %s)\n", priv, buf, len, source_node_id, is_broadcast ? "true" : "false" );
-	netdev = priv->card->netdev;
-
-	hdr.w0 = ntohl(buf[0]);
-	lf = ipv4_get_hdr_lf(&hdr);
-	if ( lf == IPV4_HDR_UNFRAG ) {
-		/*
-		 * An unfragmented datagram has been received by the ieee1394
-		 * bus. Build an skbuff around it so we can pass it to the
-		 * high level network layer.
-		 */
-		ether_type = ipv4_get_hdr_ether_type(&hdr);
-		fw_debug ( "header w0 = %x, lf = %x, ether_type = %x\n", hdr.w0, lf, ether_type );
-		buf++;
-		len -= IPV4_UNFRAG_HDR_SIZE;
-
-		skb = dev_alloc_skb(len + netdev->hard_header_len + 15);
-		if (unlikely(!skb)) {
-			fw_error ( "Out of memory for incoming packet\n");
-			netdev->stats.rx_dropped++;
-			return -1;
-		}
-		skb_reserve(skb, (netdev->hard_header_len + 15) & ~15);
-		memcpy(skb_put(skb, len), buf, len );
-		return ipv4_finish_incoming_packet(netdev, skb, source_node_id, is_broadcast, ether_type );
-	}
-	/* A datagram fragment has been received, now the fun begins. */
-	hdr.w1 = ntohl(buf[1]);
-	buf +=2;
-	len -= IPV4_FRAG_HDR_SIZE;
-	if ( lf ==IPV4_HDR_FIRSTFRAG ) {
-		ether_type = ipv4_get_hdr_ether_type(&hdr);
-		fg_off = 0;
-	} else {
-		fg_off = ipv4_get_hdr_fg_off(&hdr);
-		ether_type = 0; /* Shut up compiler! */
-	}
-	datagram_label = ipv4_get_hdr_dgl(&hdr);
-	dg_size = ipv4_get_hdr_dg_size(&hdr); /* ??? + 1 */
-	fw_debug ( "fragmented: %x.%x = lf %x, ether_type %x, fg_off %x, dgl %x, dg_size %x\n", hdr.w0, hdr.w1, lf, ether_type, fg_off, datagram_label, dg_size );
-	node = ipv4_node_find_by_nodeid ( priv, source_node_id);
-	spin_lock_irqsave(&node->pdg_lock, flags);
-	pd = ipv4_pd_find( node, datagram_label );
-	if (pd == NULL) {
-		while ( node->pdg_size >= ipv4_mpd ) {
-			/* remove the oldest */
-			ipv4_pd_delete ( list_first_entry(&node->pdg_list, struct ipv4_partial_datagram, pdg_list) );
-			node->pdg_size--;
-		}
-		pd = ipv4_pd_new ( netdev, node, datagram_label, dg_size,
- buf, fg_off, len);
-		if ( pd == NULL) {
-			retval = -ENOMEM;
-			goto bad_proto;
-		}
-		node->pdg_size++;
-	} else {
-		if (ipv4_frag_overlap(pd, fg_off, len) || pd->datagram_size != dg_size) {
-			/*
-			 * Differing datagram sizes or overlapping fragments,
-			 * Either way the remote machine is playing silly buggers
-			 * with us: obliterate the old datagram and start a new one.
-			 */
-			ipv4_pd_delete ( pd );
-			pd = ipv4_pd_new ( netdev, node, datagram_label,
- dg_size, buf, fg_off, len);
-			if ( pd == NULL ) {
-				retval = -ENOMEM;
-				node->pdg_size--;
-				goto bad_proto;
-			}
-		} else {
-			bool worked;
-
-			worked = ipv4_pd_update ( node, pd,
- buf, fg_off, len );
-			if ( ! worked ) {
-				/*
-				 * Couldn't save off fragment anyway
-				 * so might as well obliterate the
-				 * datagram now.
-				 */
-				ipv4_pd_delete ( pd );
-				node->pdg_size--;
-				goto bad_proto;
-			}
-		}
-	} /* new datagram or add to existing one */
-
-	if ( lf == IPV4_HDR_FIRSTFRAG )
-		pd->ether_type = ether_type;
-	if ( ipv4_pd_is_complete ( pd ) ) {
-		ether_type = pd->ether_type;
-		node->pdg_size--;
-		skb = skb_get(pd->skb);
-		ipv4_pd_delete ( pd );
-		spin_unlock_irqrestore(&node->pdg_lock, flags);
-		return ipv4_finish_incoming_packet ( netdev, skb, source_node_id, false, ether_type );
-	}
-	/*
-	 * Datagram is not complete, we're done for the
-	 * moment.
-	 */
-	spin_unlock_irqrestore(&node->pdg_lock, flags);
-	return 0;
-
- bad_proto:
-	spin_unlock_irqrestore(&node->pdg_lock, flags);
-	if (netif_queue_stopped(netdev))
-		netif_wake_queue(netdev);
-	return 0;
-}
-
-static void ipv4_receive_packet ( struct fw_card *card, struct fw_request *r,
- int tcode, int destination, int source, int generation, int speed,
- unsigned long long offset, void *payload, size_t length, void *callback_data ) {
-	struct ipv4_priv *priv;
-	int status;
-
-	fw_debug ( "ipv4_receive_packet(%p,%p,%x,%x,%x,%x,%x,%llx,%p,%lx,%p)\n",
- card, r, tcode, destination, source, generation, speed, offset, payload,
- (unsigned long)length, callback_data);
-	print_hex_dump ( KERN_DEBUG, "header: ", DUMP_PREFIX_OFFSET, 32, 1, payload, length, false );
-	priv = callback_data;
-	if (   tcode != TCODE_WRITE_BLOCK_REQUEST
-	    || destination != card->node_id
-	    || generation != card->generation
-	    || offset != priv->handler.offset ) {
-		fw_send_response(card, r, RCODE_CONFLICT_ERROR);
-		fw_debug("Conflict error card node_id=%x, card generation=%x, local offset %llx\n",
- card->node_id, card->generation, (unsigned long long)priv->handler.offset );
-		return;
-	}
-	status = ipv4_incoming_packet ( priv, payload, length, source, false );
-	if ( status != 0 ) {
-		fw_error ( "Incoming packet failure\n" );
-		fw_send_response ( card, r, RCODE_CONFLICT_ERROR );
-		return;
-	}
-	fw_send_response ( card, r, RCODE_COMPLETE );
-}
-
-static void ipv4_receive_broadcast(struct fw_iso_context *context, u32 cycle,
- size_t header_length, void *header, void *data) {
-	struct ipv4_priv *priv;
-	struct fw_iso_packet packet;
-	struct fw_card *card;
-	u16 *hdr_ptr;
-	u32 *buf_ptr;
-	int retval;
-	u32 length;
-	u16 source_node_id;
-	u32 specifier_id;
-	u32 ver;
-	unsigned long offset;
-	unsigned long flags;
-
-	fw_debug ( "ipv4_receive_broadcast ( context=%p, cycle=%x, header_length=%lx, header=%p, data=%p )\n", context, cycle, (unsigned long)header_length, header, data );
-	print_hex_dump ( KERN_DEBUG, "header: ", DUMP_PREFIX_OFFSET, 32, 1, header, header_length, false );
-	priv = data;
-	card = priv->card;
-	hdr_ptr = header;
-	length = ntohs(hdr_ptr[0]);
-	spin_lock_irqsave(&priv->lock,flags);
-	offset = priv->rcv_buffer_size * priv->broadcast_rcv_next_ptr;
-	buf_ptr = priv->broadcast_rcv_buffer_ptrs[priv->broadcast_rcv_next_ptr++];
-	if ( priv->broadcast_rcv_next_ptr == priv->num_broadcast_rcv_ptrs )
-		priv->broadcast_rcv_next_ptr = 0;
-	spin_unlock_irqrestore(&priv->lock,flags);
-	fw_debug ( "length %u at %p\n", length, buf_ptr );
-	print_hex_dump ( KERN_DEBUG, "buffer: ", DUMP_PREFIX_OFFSET, 32, 1, buf_ptr, length, false );
-
-	specifier_id =    (be32_to_cpu(buf_ptr[0]) & 0xffff) << 8
-			| (be32_to_cpu(buf_ptr[1]) & 0xff000000) >> 24;
-	ver = be32_to_cpu(buf_ptr[1]) & 0xFFFFFF;
-	source_node_id = be32_to_cpu(buf_ptr[0]) >> 16;
-	/* fw_debug ( "source %x SpecID %x ver %x\n", source_node_id, specifier_id, ver ); */
-	if ( specifier_id == IPV4_GASP_SPECIFIER_ID && ver == IPV4_GASP_VERSION ) {
-		buf_ptr += 2;
-		length -= IPV4_GASP_OVERHEAD;
-		ipv4_incoming_packet(priv, buf_ptr, length, source_node_id, true);
-	} else
-		fw_debug ( "Ignoring packet: not GASP\n" );
-	packet.payload_length = priv->rcv_buffer_size;
-	packet.interrupt = 1;
-	packet.skip = 0;
-	packet.tag = 3;
-	packet.sy = 0;
-	packet.header_length = IPV4_GASP_OVERHEAD;
-	spin_lock_irqsave(&priv->lock,flags);
-	retval = fw_iso_context_queue ( priv->broadcast_rcv_context, &packet,
- &priv->broadcast_rcv_buffer, offset );
-	spin_unlock_irqrestore(&priv->lock,flags);
-	if ( retval < 0 )
-		fw_error ( "requeue failed\n" );
-}
-
-static void debug_ptask ( struct ipv4_packet_task *ptask ) {
-	static const char *tx_types[] = { "Unknown", "GASP", "Write" };
-
-	fw_debug ( "packet %p { hdr { w0 %x w1 %x }, skb %p, priv %p,"
- " tx_type %s, outstanding_pkts %d, max_payload %x, fifo %llx,"
- " speed %x, dest_node %x, generation %x }\n",
- ptask, ptask->hdr.w0, ptask->hdr.w1, ptask->skb, ptask->priv,
- ptask->tx_type > IPV4_WRREQ ? "Invalid" : tx_types[ptask->tx_type],
- ptask->outstanding_pkts,  ptask->max_payload,
- ptask->fifo_addr, ptask->speed, ptask->dest_node, ptask->generation );
-	print_hex_dump ( KERN_DEBUG, "packet :", DUMP_PREFIX_OFFSET, 32, 1,
- ptask->skb->data, ptask->skb->len, false );
-}
-
-static void ipv4_transmit_packet_done ( struct ipv4_packet_task *ptask ) {
-	struct ipv4_priv *priv;
-	unsigned long flags;
-
-	priv = ptask->priv;
-	spin_lock_irqsave ( &priv->lock, flags );
-	list_del ( &ptask->packet_list );
-	spin_unlock_irqrestore ( &priv->lock, flags );
-	ptask->outstanding_pkts--;
-	if ( ptask->outstanding_pkts > 0 ) {
-		u16 dg_size;
-		u16 fg_off;
-		u16 datagram_label;
-		u16 lf;
-		struct sk_buff *skb;
-
-		/* Update the ptask to point to the next fragment and send it */
-		lf = ipv4_get_hdr_lf(&ptask->hdr);
-		switch (lf) {
-		case IPV4_HDR_LASTFRAG:
-		case IPV4_HDR_UNFRAG:
-		default:
-			fw_error ( "Outstanding packet %x lf %x, header %x,%x\n", ptask->outstanding_pkts, lf, ptask->hdr.w0, ptask->hdr.w1 );
-			BUG();
-
-		case IPV4_HDR_FIRSTFRAG:
-			/* Set frag type here for future interior fragments */
-			dg_size = ipv4_get_hdr_dg_size(&ptask->hdr);
-			fg_off = ptask->max_payload - IPV4_FRAG_HDR_SIZE;
-			datagram_label = ipv4_get_hdr_dgl(&ptask->hdr);
-			break;
-
-		case IPV4_HDR_INTFRAG:
-			dg_size = ipv4_get_hdr_dg_size(&ptask->hdr);
-			fg_off = ipv4_get_hdr_fg_off(&ptask->hdr) + ptask->max_payload - IPV4_FRAG_HDR_SIZE;
-			datagram_label = ipv4_get_hdr_dgl(&ptask->hdr);
-			break;
-		}
-		skb = ptask->skb;
-		skb_pull ( skb, ptask->max_payload );
-		if ( ptask->outstanding_pkts > 1 ) {
-			ipv4_make_sf_hdr ( &ptask->hdr,
-  IPV4_HDR_INTFRAG, dg_size, fg_off, datagram_label );
-		} else {
-			ipv4_make_sf_hdr ( &ptask->hdr,
-  IPV4_HDR_LASTFRAG, dg_size, fg_off, datagram_label );
-			ptask->max_payload = skb->len + IPV4_FRAG_HDR_SIZE;
-
-		}
-		ipv4_send_packet ( ptask );
-	} else {
-		dev_kfree_skb_any ( ptask->skb );
-		kmem_cache_free( ipv4_packet_task_cache, ptask );
-	}
-}
-
-static void ipv4_write_complete ( struct fw_card *card, int rcode,
- void *payload, size_t length, void *data ) {
-	struct ipv4_packet_task *ptask;
-
-	ptask = data;
-	fw_debug ( "ipv4_write_complete ( %p, %x, %p, %lx, %p )\n",
- card, rcode, payload, (unsigned long)length, data );
-	debug_ptask ( ptask );
-
-	if ( rcode == RCODE_COMPLETE ) {
-		ipv4_transmit_packet_done ( ptask );
-	} else {
-		fw_error ( "ipv4_write_complete: failed: %x\n", rcode );
-		/* ??? error recovery */
-	}
-}
-
-static int ipv4_send_packet ( struct ipv4_packet_task *ptask ) {
-	struct ipv4_priv *priv;
-	unsigned tx_len;
-	struct ipv4_hdr *bufhdr;
-	unsigned long flags;
-	struct net_device *netdev;
-#if 0 /* stefanr */
-	int retval;
-#endif
-
-	fw_debug ( "ipv4_send_packet\n" );
-	debug_ptask ( ptask );
-	priv = ptask->priv;
-	tx_len = ptask->max_payload;
-	switch (ipv4_get_hdr_lf(&ptask->hdr)) {
-	case IPV4_HDR_UNFRAG:
-		bufhdr = (struct ipv4_hdr *)skb_push(ptask->skb, IPV4_UNFRAG_HDR_SIZE);
-		bufhdr->w0 = htonl(ptask->hdr.w0);
-		break;
-
-	case IPV4_HDR_FIRSTFRAG:
-	case IPV4_HDR_INTFRAG:
-	case IPV4_HDR_LASTFRAG:
-		bufhdr = (struct ipv4_hdr *)skb_push(ptask->skb, IPV4_FRAG_HDR_SIZE);
-		bufhdr->w0 = htonl(ptask->hdr.w0);
-		bufhdr->w1 = htonl(ptask->hdr.w1);
-		break;
-
-	default:
-		BUG();
-	}
-	if ( ptask->tx_type == IPV4_GASP ) {
-		u32 *packets;
-		int generation;
-		int nodeid;
-
-		/* ptask->generation may not have been set yet */
-		generation = priv->card->generation;
-		smp_rmb();
-		nodeid = priv->card->node_id;
-		packets = (u32 *)skb_push(ptask->skb, sizeof(u32)*2);
-		packets[0] = htonl(nodeid << 16 | (IPV4_GASP_SPECIFIER_ID>>8));
-		packets[1] = htonl((IPV4_GASP_SPECIFIER_ID & 0xFF) << 24 | IPV4_GASP_VERSION);
-		fw_send_request ( priv->card, &ptask->transaction, TCODE_STREAM_DATA,
- fw_stream_packet_destination_id(3, BROADCAST_CHANNEL, 0),
- generation, SCODE_100, 0ULL, ptask->skb->data, tx_len + 8, ipv4_write_complete, ptask );
-		spin_lock_irqsave(&priv->lock,flags);
-		list_add_tail ( &ptask->packet_list, &priv->broadcasted_list );
-		spin_unlock_irqrestore(&priv->lock,flags);
-#if 0 /* stefanr */
-		return retval;
-#else
-		return 0;
-#endif
-	}
-	fw_debug("send_request (%p, %p, WRITE_BLOCK, %x, %x, %x, %llx, %p, %d, %p, %p\n",
- priv->card, &ptask->transaction, ptask->dest_node, ptask->generation,
- ptask->speed, (unsigned long long)ptask->fifo_addr, ptask->skb->data, tx_len,
- ipv4_write_complete, ptask );
-	fw_send_request ( priv->card, &ptask->transaction,
- TCODE_WRITE_BLOCK_REQUEST, ptask->dest_node, ptask->generation, ptask->speed,
- ptask->fifo_addr, ptask->skb->data, tx_len, ipv4_write_complete, ptask );
-	spin_lock_irqsave(&priv->lock,flags);
-	list_add_tail ( &ptask->packet_list, &priv->sent_list );
-	spin_unlock_irqrestore(&priv->lock,flags);
-	netdev = priv->card->netdev;
-	netdev->trans_start = jiffies;
-	return 0;
-}
-
-static int ipv4_broadcast_start ( struct ipv4_priv *priv ) {
-	struct fw_iso_context *context;
-	int retval;
-	unsigned num_packets;
-	unsigned max_receive;
-	struct fw_iso_packet packet;
-	unsigned long offset;
-	unsigned u;
-	/* unsigned transmit_speed; */
-
-#if 0 /* stefanr */
-	if ( priv->card->broadcast_channel != (BROADCAST_CHANNEL_VALID|BROADCAST_CHANNEL_INITIAL)) {
-		fw_notify ( "Invalid broadcast channel %x\n", priv->card->broadcast_channel );
-		/* FIXME: try again later? */
-		/* return -EINVAL; */
-	}
-#endif
-	if ( priv->local_fifo == INVALID_FIFO_ADDR ) {
-		struct fw_address_region region;
-
-		priv->handler.length = FIFO_SIZE;
-		priv->handler.address_callback = ipv4_receive_packet;
-		priv->handler.callback_data = priv;
-		/* FIXME: this is OHCI, but what about others? */
-		region.start = 0xffff00000000ULL;
-		region.end =   0xfffffffffffcULL;
-
-		retval = fw_core_add_address_handler ( &priv->handler, &region );
-		if ( retval < 0 )
-			goto failed_initial;
-		priv->local_fifo = priv->handler.offset;
-	}
-
-	/*
-	 * FIXME: rawiso limits us to PAGE_SIZE.  This only matters if we ever have
-	 * a machine with PAGE_SIZE < 4096
-	 */
-	max_receive = 1U << (priv->card->max_receive + 1);
-	num_packets = ( ipv4_iso_page_count * PAGE_SIZE ) / max_receive;
-	if ( ! priv->broadcast_rcv_context ) {
-		void **ptrptr;
-
-		context = fw_iso_context_create ( priv->card,
- FW_ISO_CONTEXT_RECEIVE, BROADCAST_CHANNEL,
- priv->card->link_speed, 8, ipv4_receive_broadcast, priv );
-		if (IS_ERR(context)) {
-			retval = PTR_ERR(context);
-			goto failed_context_create;
-		}
-		retval = fw_iso_buffer_init ( &priv->broadcast_rcv_buffer,
- priv->card, ipv4_iso_page_count, DMA_FROM_DEVICE );
-		if ( retval < 0 )
-			goto failed_buffer_init;
-		ptrptr = kmalloc ( sizeof(void*)*num_packets, GFP_KERNEL );
-		if ( ! ptrptr ) {
-			retval = -ENOMEM;
-			goto failed_ptrs_alloc;
-		}
-		priv->broadcast_rcv_buffer_ptrs = ptrptr;
-		for ( u = 0; u < ipv4_iso_page_count; u++ ) {
-			void *ptr;
-			unsigned v;
-
-			ptr = kmap ( priv->broadcast_rcv_buffer.pages[u] );
-			for ( v = 0; v < num_packets / ipv4_iso_page_count; v++ )
-				*ptrptr++ = (void *)((char *)ptr + v * max_receive);
-		}
-		priv->broadcast_rcv_context = context;
-	} else
-		context = priv->broadcast_rcv_context;
-
-	packet.payload_length = max_receive;
-	packet.interrupt = 1;
-	packet.skip = 0;
-	packet.tag = 3;
-	packet.sy = 0;
-	packet.header_length = IPV4_GASP_OVERHEAD;
-	offset = 0;
-	for ( u = 0; u < num_packets; u++ ) {
-		retval = fw_iso_context_queue ( context, &packet,
- &priv->broadcast_rcv_buffer, offset );
-		if ( retval < 0 )
-			goto failed_rcv_queue;
-		offset += max_receive;
-	}
-	priv->num_broadcast_rcv_ptrs = num_packets;
-	priv->rcv_buffer_size = max_receive;
-	priv->broadcast_rcv_next_ptr = 0U;
-	retval = fw_iso_context_start ( context, -1, 0, FW_ISO_CONTEXT_MATCH_ALL_TAGS ); /* ??? sync */
-	if ( retval < 0 )
-		goto failed_rcv_queue;
-	/* FIXME: adjust this when we know the max receive speeds of all other IP nodes on the bus. */
-	/* since we only xmt at S100 ??? */
-	priv->broadcast_xmt_max_payload = S100_BUFFER_SIZE - IPV4_GASP_OVERHEAD - IPV4_UNFRAG_HDR_SIZE;
-	priv->broadcast_state = IPV4_BROADCAST_RUNNING;
-	return 0;
-
- failed_rcv_queue:
-	kfree ( priv->broadcast_rcv_buffer_ptrs );
-	priv->broadcast_rcv_buffer_ptrs = NULL;
- failed_ptrs_alloc:
-	fw_iso_buffer_destroy ( &priv->broadcast_rcv_buffer, priv->card );
- failed_buffer_init:
-	fw_iso_context_destroy ( context );
-	priv->broadcast_rcv_context = NULL;
- failed_context_create:
-	fw_core_remove_address_handler ( &priv->handler );
- failed_initial:
-	priv->local_fifo = INVALID_FIFO_ADDR;
-	return retval;
-}
-
-/* This is called after an "ifup" */
-static int ipv4_open(struct net_device *dev) {
-	struct ipv4_priv *priv;
-	int ret;
-
-	priv = netdev_priv(dev);
-	if (priv->broadcast_state == IPV4_BROADCAST_ERROR) {
-		ret = ipv4_broadcast_start ( priv );
-		if (ret)
-			return ret;
-	}
-	netif_start_queue(dev);
-	return 0;
-}
-
-/* This is called after an "ifdown" */
-static int ipv4_stop(struct net_device *netdev)
-{
-	/* flush priv->wake */
-	/* flush_scheduled_work(); */
-
-	netif_stop_queue(netdev);
-	return 0;
-}
-
-/* Transmit a packet (called by kernel) */
-static int ipv4_tx(struct sk_buff *skb, struct net_device *netdev)
-{
-	struct ipv4_ether_hdr hdr_buf;
-	struct ipv4_priv *priv = netdev_priv(netdev);
-	__be16 proto;
-	u16 dest_node;
-	enum ipv4_tx_type tx_type;
-	unsigned max_payload;
-	u16 dg_size;
-	u16 *datagram_label_ptr;
-	struct ipv4_packet_task *ptask;
-	struct ipv4_node *node = NULL;
-
-	ptask = kmem_cache_alloc(ipv4_packet_task_cache, GFP_ATOMIC);
-	if (ptask == NULL)
-		goto fail;
-
-	skb = skb_share_check(skb, GFP_ATOMIC);
-	if (!skb)
-		goto fail;
-
-	/*
-	 * Get rid of the fake ipv4 header, but first make a copy.
-	 * We might need to rebuild the header on tx failure.
-	 */
-	memcpy(&hdr_buf, skb->data, sizeof(hdr_buf));
-	skb_pull(skb, sizeof(hdr_buf));
-
-	proto = hdr_buf.h_proto;
-	dg_size = skb->len;
-
-	/*
-	 * Set the transmission type for the packet.  ARP packets and IP
-	 * broadcast packets are sent via GASP.
-	 */
-	if (   memcmp(hdr_buf.h_dest, netdev->broadcast, IPV4_ALEN) == 0
-	    || proto == htons(ETH_P_ARP)
-	    || (   proto == htons(ETH_P_IP)
-		&& IN_MULTICAST(ntohl(ip_hdr(skb)->daddr)) ) ) {
-		/* fw_debug ( "transmitting arp or multicast packet\n" );*/
-		tx_type = IPV4_GASP;
-		dest_node = ALL_NODES;
-		max_payload = priv->broadcast_xmt_max_payload;
-		/* BUG_ON(max_payload < S100_BUFFER_SIZE - IPV4_GASP_OVERHEAD); */
-		datagram_label_ptr = &priv->broadcast_xmt_datagramlabel;
-		ptask->fifo_addr = INVALID_FIFO_ADDR;
-		ptask->generation = 0U;
-		ptask->dest_node = 0U;
-		ptask->speed = 0;
-	} else {
-		__be64 guid = get_unaligned((u64 *)hdr_buf.h_dest);
-		u8 generation;
-
-		node = ipv4_node_find_by_guid(priv, be64_to_cpu(guid));
-		if (!node) {
-			fw_debug ( "Normal packet but no node\n" );
-			goto fail;
-		}
-
-		if (node->fifo == INVALID_FIFO_ADDR) {
-			fw_debug ( "Normal packet but no fifo addr\n" );
-			goto fail;
-		}
-
-		/* fw_debug ( "Transmitting normal packet to %x at %llxx\n", node->nodeid, node->fifo ); */
-		generation = node->generation;
-		dest_node = node->nodeid;
-		max_payload = node->max_payload;
-		/* BUG_ON(max_payload < S100_BUFFER_SIZE - IPV4_FRAG_HDR_SIZE); */
-
-		datagram_label_ptr = &node->datagram_label;
-		tx_type = IPV4_WRREQ;
-		ptask->fifo_addr = node->fifo;
-		ptask->generation = generation;
-		ptask->dest_node = dest_node;
-		ptask->speed = node->xmt_speed;
-	}
-
-	/* If this is an ARP packet, convert it */
-	if (proto == htons(ETH_P_ARP)) {
-		/* Convert a standard ARP packet to 1394 ARP. The first 8 bytes (the entire
-		 * arphdr) is the same format as the ip1394 header, so they overlap.  The rest
-		 * needs to be munged a bit.  The remainder of the arphdr is formatted based
-		 * on hwaddr len and ipaddr len.  We know what they'll be, so it's easy to
-		 * judge.
-		 *
-		 * Now that the EUI is used for the hardware address all we need to do to make
-		 * this work for 1394 is to insert 2 quadlets that contain max_rec size,
-		 * speed, and unicast FIFO address information between the sender_unique_id
-		 * and the IP addresses.
-		 */
-		struct arphdr *arp = (struct arphdr *)skb->data;
-		unsigned char *arp_ptr = (unsigned char *)(arp + 1);
-		struct ipv4_arp *arp1394 = (struct ipv4_arp *)skb->data;
-		u32 ipaddr;
-
-		ipaddr = *(u32*)(arp_ptr + IPV4_ALEN);
-		arp1394->hw_addr_len    = 16;
-		arp1394->max_rec        = priv->card->max_receive;
-		arp1394->sspd		= priv->card->link_speed;
-		arp1394->fifo_hi	= htons(priv->local_fifo >> 32);
-		arp1394->fifo_lo        = htonl(priv->local_fifo & 0xFFFFFFFF);
-		arp1394->sip		= ipaddr;
-	}
-	if ( ipv4_max_xmt && max_payload > ipv4_max_xmt )
-		max_payload = ipv4_max_xmt;
-
-	ptask->hdr.w0 = 0;
-	ptask->hdr.w1 = 0;
-	ptask->skb = skb;
-	ptask->priv = priv;
-        ptask->tx_type = tx_type;
-	/* Does it all fit in one packet? */
-	if ( dg_size <= max_payload ) {
-		ipv4_make_uf_hdr(&ptask->hdr, be16_to_cpu(proto));
-		ptask->outstanding_pkts = 1;
-		max_payload = dg_size + IPV4_UNFRAG_HDR_SIZE;
-	} else {
-		u16 datagram_label;
-
-		max_payload -= IPV4_FRAG_OVERHEAD;
-		datagram_label = (*datagram_label_ptr)++;
-		ipv4_make_ff_hdr(&ptask->hdr, be16_to_cpu(proto), dg_size, datagram_label );
-		ptask->outstanding_pkts = DIV_ROUND_UP(dg_size, max_payload);
-		max_payload += IPV4_FRAG_HDR_SIZE;
-	}
-	ptask->max_payload = max_payload;
-	ipv4_send_packet ( ptask );
-	return NETDEV_TX_OK;
-
- fail:
-	if (ptask)
-		kmem_cache_free(ipv4_packet_task_cache, ptask);
-
-	if (skb != NULL)
-		dev_kfree_skb(skb);
-
-	netdev->stats.tx_dropped++;
-	netdev->stats.tx_errors++;
-
-	/*
-	 * FIXME: According to a patch from 2003-02-26, "returning non-zero
-	 * causes serious problems" here, allegedly.  Before that patch,
-	 * -ERRNO was returned which is not appropriate under Linux 2.6.
-	 * Perhaps more needs to be done?  Stop the queue in serious
-	 * conditions and restart it elsewhere?
-	 */
-	return NETDEV_TX_OK;
-}
-
-/*
- * FIXME: What to do if we timeout? I think a host reset is probably in order,
- * so that's what we do. Should we increment the stat counters too?
- */
-static void ipv4_tx_timeout(struct net_device *dev) {
-	struct ipv4_priv *priv;
-
-	priv = netdev_priv(dev);
-	fw_error ( "%s: Timeout, resetting host\n", dev->name );
-#if 0 /* stefanr */
-	fw_core_initiate_bus_reset ( priv->card, 1 );
-#endif
-}
-
-static int ipv4_change_mtu ( struct net_device *dev, int new_mtu ) {
-#if 0
-	int max_mtu;
-	struct ipv4_priv *priv;
-#endif
-
-	if (new_mtu < 68)
-		return -EINVAL;
-
-#if 0
-	priv = netdev_priv(dev);
-	/* This is not actually true because we can fragment packets at the firewire layer */
-	max_mtu = (1 << (priv->card->max_receive + 1))
-		                - sizeof(struct ipv4_hdr) - IPV4_GASP_OVERHEAD;
-	if (new_mtu > max_mtu) {
-		fw_notify ( "%s: Local node constrains MTU to %d\n", dev->name, max_mtu);
-		return -ERANGE;
-	}
-#endif
-	dev->mtu = new_mtu;
-	return 0;
-}
-
-static void ipv4_get_drvinfo(struct net_device *dev,
-struct ethtool_drvinfo *info) {
-	strcpy(info->driver, ipv4_driver_name);
-	strcpy(info->bus_info, "ieee1394"); /* FIXME provide more detail? */
-}
-
-static struct ethtool_ops ipv4_ethtool_ops = {
-	.get_drvinfo = ipv4_get_drvinfo,
-};
-
-static const struct net_device_ops ipv4_netdev_ops = {
-	.ndo_open       = ipv4_open,
-	.ndo_stop	= ipv4_stop,
-	.ndo_start_xmit = ipv4_tx,
-	.ndo_tx_timeout = ipv4_tx_timeout,
-	.ndo_change_mtu = ipv4_change_mtu,
-};
-
-static void ipv4_init_dev ( struct net_device *dev ) {
-	dev->header_ops		= &ipv4_header_ops;
-	dev->netdev_ops         = &ipv4_netdev_ops;
-	SET_ETHTOOL_OPS(dev, &ipv4_ethtool_ops);
-
-	dev->watchdog_timeo	= IPV4_TIMEOUT;
-	dev->flags		= IFF_BROADCAST | IFF_MULTICAST;
-	dev->features		= NETIF_F_HIGHDMA;
-	dev->addr_len		= IPV4_ALEN;
-	dev->hard_header_len	= IPV4_HLEN;
-	dev->type		= ARPHRD_IEEE1394;
-
-	/* FIXME: This value was copied from ether_setup(). Is it too much? */
-	dev->tx_queue_len	= 1000;
-}
-
-static int ipv4_probe ( struct device *dev ) {
-	struct fw_unit * unit;
-	struct fw_device *device;
-	struct fw_card *card;
-	struct net_device *netdev;
-	struct ipv4_priv *priv;
-	unsigned max_mtu;
-	__be64 guid;
-
-	fw_debug("ipv4 Probing\n" );
-	unit = fw_unit ( dev );
-	device = fw_device ( unit->device.parent );
-	card = device->card;
-
-	if ( ! device->is_local ) {
-		int added;
-
-		fw_debug ( "Non-local, adding remote node entry\n" );
-		added = ipv4_node_new ( card, device );
-		return added;
-	}
-	fw_debug("ipv4 Local: adding netdev\n" );
-	netdev = alloc_netdev ( sizeof(*priv), "fw-ipv4-%d", ipv4_init_dev );
-	if ( netdev == NULL) {
-		fw_error( "Out of memory\n");
-		goto out;
-	}
-
-	SET_NETDEV_DEV(netdev, card->device);
-	priv = netdev_priv(netdev);
-
-	spin_lock_init(&priv->lock);
-	priv->broadcast_state = IPV4_BROADCAST_ERROR;
-	priv->broadcast_rcv_context = NULL;
-	priv->broadcast_xmt_max_payload = 0;
-	priv->broadcast_xmt_datagramlabel = 0;
-
-	priv->local_fifo = INVALID_FIFO_ADDR;
-
-	/* INIT_WORK(&priv->wake, ipv4_handle_queue);*/
-	INIT_LIST_HEAD(&priv->packet_list);
-	INIT_LIST_HEAD(&priv->broadcasted_list);
-	INIT_LIST_HEAD(&priv->sent_list );
-
-	priv->card = card;
-
-	/*
-	 * Use the RFC 2734 default 1500 octets or the maximum payload
-	 * as initial MTU
-	 */
-	max_mtu = (1 << (card->max_receive + 1))
-		  - sizeof(struct ipv4_hdr) - IPV4_GASP_OVERHEAD;
-	netdev->mtu = min(1500U, max_mtu);
-
-	/* Set our hardware address while we're at it */
-	guid = cpu_to_be64(card->guid);
-	memcpy(netdev->dev_addr, &guid, sizeof(u64));
-	memset(netdev->broadcast, 0xff, sizeof(u64));
-	if ( register_netdev ( netdev ) ) {
-		fw_error ( "Cannot register the driver\n");
-		goto out;
-	}
-
-	fw_notify ( "%s: IPv4 over Firewire on device %016llx\n",
- netdev->name, card->guid );
-	card->netdev = netdev;
-
-	return 0 /* ipv4_new_node ( ud ) */;
- out:
-	if ( netdev )
-		free_netdev ( netdev );
-	return -ENOENT;
-}
-
-
-static int ipv4_remove ( struct device *dev ) {
-	struct fw_unit * unit;
-	struct fw_device *device;
-	struct fw_card *card;
-	struct net_device *netdev;
-	struct ipv4_priv *priv;
-	struct ipv4_node *node;
-	struct ipv4_partial_datagram *pd, *pd_next;
-	struct ipv4_packet_task *ptask, *pt_next;
-
-	fw_debug("ipv4 Removing\n" );
-	unit = fw_unit ( dev );
-	device = fw_device ( unit->device.parent );
-	card = device->card;
-
-	if ( ! device->is_local ) {
-		fw_debug ( "Node %x is non-local, removing remote node entry\n", device->node_id );
-		ipv4_node_delete ( card, device );
-		return 0;
-	}
-	netdev = card->netdev;
-	if ( netdev ) {
-		fw_debug ( "Node %x is local: deleting netdev\n", device->node_id );
-		priv = netdev_priv ( netdev );
-		unregister_netdev ( netdev );
-		fw_debug ( "unregistered\n" );
-		if ( priv->local_fifo != INVALID_FIFO_ADDR )
-			fw_core_remove_address_handler ( &priv->handler );
-		fw_debug ( "address handler gone\n" );
-		if ( priv->broadcast_rcv_context ) {
-			fw_iso_context_stop ( priv->broadcast_rcv_context );
-			fw_iso_buffer_destroy ( &priv->broadcast_rcv_buffer, priv->card );
-			fw_iso_context_destroy ( priv->broadcast_rcv_context );
-			fw_debug ( "rcv stopped\n" );
-		}
-		list_for_each_entry_safe( ptask, pt_next, &priv->packet_list, packet_list ) {
-			dev_kfree_skb_any ( ptask->skb );
-			kmem_cache_free( ipv4_packet_task_cache, ptask );
-		}
-		list_for_each_entry_safe( ptask, pt_next, &priv->broadcasted_list, packet_list ) {
-			dev_kfree_skb_any ( ptask->skb );
-			kmem_cache_free( ipv4_packet_task_cache, ptask );
-		}
-		list_for_each_entry_safe( ptask, pt_next, &priv->sent_list, packet_list ) {
-			dev_kfree_skb_any ( ptask->skb );
-			kmem_cache_free( ipv4_packet_task_cache, ptask );
-		}
-		fw_debug ( "lists emptied\n" );
-		list_for_each_entry( node, &card->ipv4_nodes, ipv4_nodes ) {
-			if ( node->pdg_size ) {
-				list_for_each_entry_safe( pd, pd_next, &node->pdg_list, pdg_list )
-					ipv4_pd_delete ( pd );
-				node->pdg_size = 0;
-			}
-			node->fifo = INVALID_FIFO_ADDR;
-		}
-		fw_debug ( "nodes cleaned up\n" );
-		free_netdev ( netdev );
-		card->netdev = NULL;
-		fw_debug ( "done\n" );
-	}
-	return 0;
-}
-
-static void ipv4_update ( struct fw_unit *unit ) {
-	struct fw_device *device;
-	struct fw_card *card;
-
-	fw_debug ( "ipv4_update unit %p\n", unit );
-	device = fw_device ( unit->device.parent );
-	card = device->card;
-	if ( ! device->is_local ) {
-		struct ipv4_node *node;
-		u64 guid;
-		struct net_device *netdev;
-		struct ipv4_priv *priv;
-
-		netdev = card->netdev;
-		if ( netdev ) {
-			priv = netdev_priv ( netdev );
-			guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
-			node = ipv4_node_find_by_guid ( priv, guid );
-			if ( ! node ) {
-				fw_error ( "ipv4_update: no node for device %llx\n", guid );
-				return;
-			}
-			fw_debug ( "Non-local, updating remote node entry for guid %llx old generation %x, old nodeid %x\n", guid, node->generation, node->nodeid );
-			node->generation = device->generation;
-			rmb();
-			node->nodeid = device->node_id;
-			fw_debug ( "New generation %x, new nodeid %x\n", node->generation, node->nodeid );
-		} else
-			fw_error ( "nonlocal, but no netdev?  How can that be?\n" );
-	} else {
-		/* FIXME: What do we need to do on bus reset? */
-		fw_debug ( "Local, doing nothing\n" );
-	}
-}
-
-static struct fw_driver ipv4_driver = {
-	.driver = {
-		.owner = THIS_MODULE,
-		.name = ipv4_driver_name,
-		.bus = &fw_bus_type,
-		.probe = ipv4_probe,
-		.remove = ipv4_remove,
-	},
-	.update = ipv4_update,
-	.id_table = ipv4_id_table,
-};
-
-static int __init ipv4_init ( void ) {
-	int added;
-
-	added = fw_core_add_descriptor ( &ipv4_unit_directory );
-	if ( added < 0 )
-		fw_error ( "Failed to add descriptor" );
-	ipv4_packet_task_cache = kmem_cache_create("packet_task",
- sizeof(struct ipv4_packet_task), 0, 0, NULL);
-	fw_debug("Adding ipv4 module\n" );
-	return driver_register ( &ipv4_driver.driver );
-}
-
-static void __exit ipv4_cleanup ( void ) {
-	fw_core_remove_descriptor ( &ipv4_unit_directory );
-	fw_debug("Removing ipv4 module\n" );
-	driver_unregister ( &ipv4_driver.driver );
-}
-
-module_init(ipv4_init);
-module_exit(ipv4_cleanup);
diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c
new file mode 100644
index 000000000000..15353886bd80
--- /dev/null
+++ b/drivers/firewire/net.c
@@ -0,0 +1,1819 @@
+/*
+ * IPv4 over IEEE 1394, per RFC 2734
+ *
+ * Copyright (C) 2009 Jay Fenlason <fenlason@redhat.com>
+ *
+ * based on eth1394 by Ben Collins et al
+ */
+
+#include <linux/device.h>
+#include <linux/ethtool.h>
+#include <linux/firewire.h>
+#include <linux/firewire-constants.h>
+#include <linux/highmem.h>
+#include <linux/in.h>
+#include <linux/ip.h>
+#include <linux/mod_devicetable.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/netdevice.h>
+#include <linux/skbuff.h>
+
+#include <asm/unaligned.h>
+#include <net/arp.h>
+
+/* Things to potentially make runtime cofigurable */
+/* must be at least as large as our maximum receive size */
+#define FIFO_SIZE 4096
+/* Network timeout in glibbles */
+#define IPV4_TIMEOUT       100000
+
+/* Runitme configurable paramaters */
+static int ipv4_mpd = 25;
+static int ipv4_max_xmt = 0;
+/* 16k for receiving arp and broadcast packets.  Enough? */
+static int ipv4_iso_page_count = 4;
+
+MODULE_AUTHOR("Jay Fenlason (fenlason@redhat.com)");
+MODULE_DESCRIPTION("Firewire IPv4 Driver (IPv4-over-IEEE1394 as per RFC 2734)");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(ieee1394, ipv4_id_table);
+module_param_named(max_partial_datagrams, ipv4_mpd, int, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(max_partial_datagrams, "Maximum number of received"
+ " incomplete fragmented datagrams (default = 25).");
+
+/* Max xmt is useful for forcing fragmentation, which makes testing easier. */
+module_param_named(max_transmit, ipv4_max_xmt, int, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(max_transmit, "Maximum datagram size to transmit"
+ " (larger datagrams will be fragmented) (default = 0 (use hardware defaults).");
+
+/* iso page count controls how many pages will be used for receiving broadcast packets. */
+module_param_named(iso_pages, ipv4_iso_page_count, int, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(iso_pages, "Number of pages to use for receiving broadcast packets"
+ " (default = 4).");
+
+/* uncomment this line to do debugging */
+#define fw_debug(s, args...) printk(KERN_DEBUG KBUILD_MODNAME ": " s, ## args)
+
+/* comment out these lines to do debugging. */
+/* #undef fw_debug */
+/* #define fw_debug(s...) */
+/* #define print_hex_dump(l...) */
+
+/* Define a fake hardware header format for the networking core.  Note that
+ * header size cannot exceed 16 bytes as that is the size of the header cache.
+ * Also, we do not need the source address in the header so we omit it and
+ * keep the header to under 16 bytes */
+#define IPV4_ALEN (8)
+/* This must equal sizeof(struct ipv4_ether_hdr) */
+#define IPV4_HLEN (10)
+
+/* FIXME: what's a good size for this? */
+#define INVALID_FIFO_ADDR (u64)~0ULL
+
+/* Things specified by standards */
+#define BROADCAST_CHANNEL 31
+
+#define S100_BUFFER_SIZE 512
+#define MAX_BUFFER_SIZE 4096
+
+#define IPV4_GASP_SPECIFIER_ID	0x00005EU
+#define IPV4_GASP_VERSION	0x00000001U
+
+#define IPV4_GASP_OVERHEAD (2 * sizeof(u32)) /* for GASP header */
+
+#define IPV4_UNFRAG_HDR_SIZE	sizeof(u32)
+#define IPV4_FRAG_HDR_SIZE	(2 * sizeof(u32))
+#define IPV4_FRAG_OVERHEAD	sizeof(u32)
+
+#define ALL_NODES (0xffc0 | 0x003f)
+
+#define IPV4_HDR_UNFRAG		0	/* unfragmented		*/
+#define IPV4_HDR_FIRSTFRAG	1	/* first fragment	*/
+#define IPV4_HDR_LASTFRAG	2	/* last fragment	*/
+#define IPV4_HDR_INTFRAG	3	/* interior fragment	*/
+
+/* Our arp packet (ARPHRD_IEEE1394) */
+/* FIXME: note that this is probably bogus on weird-endian machines */
+struct ipv4_arp {
+	u16 hw_type;		/* 0x0018	*/
+	u16 proto_type;		/* 0x0806       */
+	u8 hw_addr_len;		/* 16		*/
+	u8 ip_addr_len;         /* 4		*/
+	u16 opcode;	        /* ARP Opcode	*/
+	/* Above is exactly the same format as struct arphdr */
+
+	u64 s_uniq_id;		/* Sender's 64bit EUI			*/
+	u8 max_rec;             /* Sender's max packet size		*/
+	u8 sspd;		/* Sender's max speed			*/
+	u16 fifo_hi;            /* hi 16bits of sender's FIFO addr	*/
+	u32 fifo_lo;            /* lo 32bits of sender's FIFO addr	*/
+	u32 sip;		/* Sender's IP Address			*/
+	u32 tip;		/* IP Address of requested hw addr	*/
+} __attribute__((packed));
+
+struct ipv4_ether_hdr {
+	unsigned char	h_dest[IPV4_ALEN];	/* destination address */
+	unsigned short  h_proto;                /* packet type ID field */
+}  __attribute__((packed));
+
+static inline struct ipv4_ether_hdr *ipv4_ether_hdr(const struct sk_buff *skb)
+{
+	return (struct ipv4_ether_hdr *)skb_mac_header(skb);
+}
+
+enum ipv4_tx_type {
+	IPV4_UNKNOWN = 0,
+	IPV4_GASP = 1,
+	IPV4_WRREQ = 2,
+};
+
+enum ipv4_broadcast_state {
+	IPV4_BROADCAST_ERROR,
+	IPV4_BROADCAST_RUNNING,
+	IPV4_BROADCAST_STOPPED,
+};
+
+#define ipv4_get_hdr_lf(h)		(((h)->w0&0xC0000000)>>30)
+#define ipv4_get_hdr_ether_type(h)	(((h)->w0&0x0000FFFF)    )
+#define ipv4_get_hdr_dg_size(h)		(((h)->w0&0x0FFF0000)>>16)
+#define ipv4_get_hdr_fg_off(h)		(((h)->w0&0x00000FFF)    )
+#define ipv4_get_hdr_dgl(h)		(((h)->w1&0xFFFF0000)>>16)
+
+#define ipv4_set_hdr_lf(lf)		(( lf)<<30)
+#define ipv4_set_hdr_ether_type(et)	(( et)    )
+#define ipv4_set_hdr_dg_size(dgs)	((dgs)<<16)
+#define ipv4_set_hdr_fg_off(fgo)	((fgo)    )
+
+#define ipv4_set_hdr_dgl(dgl)		((dgl)<<16)
+
+struct ipv4_hdr {
+	u32 w0;
+	u32 w1;
+};
+
+static inline void ipv4_make_uf_hdr( struct ipv4_hdr *hdr, unsigned ether_type) {
+	hdr->w0 = ipv4_set_hdr_lf(IPV4_HDR_UNFRAG)
+		   |ipv4_set_hdr_ether_type(ether_type);
+	fw_debug ( "Setting unfragmented header %p to %x\n", hdr, hdr->w0 );
+}
+
+static inline void ipv4_make_ff_hdr ( struct ipv4_hdr *hdr, unsigned ether_type, unsigned dg_size, unsigned dgl ) {
+	hdr->w0 = ipv4_set_hdr_lf(IPV4_HDR_FIRSTFRAG)
+		   |ipv4_set_hdr_dg_size(dg_size)
+		   |ipv4_set_hdr_ether_type(ether_type);
+	hdr->w1 = ipv4_set_hdr_dgl(dgl);
+	fw_debug ( "Setting fragmented header %p to first_frag %x,%x (et %x, dgs %x, dgl %x)\n", hdr, hdr->w0, hdr->w1,
+ ether_type, dg_size, dgl );
+}
+
+static inline void ipv4_make_sf_hdr ( struct ipv4_hdr *hdr, unsigned lf, unsigned dg_size, unsigned fg_off, unsigned dgl) {
+	hdr->w0 = ipv4_set_hdr_lf(lf)
+		 |ipv4_set_hdr_dg_size(dg_size)
+		 |ipv4_set_hdr_fg_off(fg_off);
+	hdr->w1 = ipv4_set_hdr_dgl(dgl);
+	fw_debug ( "Setting fragmented header %p to %x,%x (lf %x, dgs %x, fo %x dgl %x)\n",
+ hdr, hdr->w0, hdr->w1,
+ lf, dg_size, fg_off, dgl );
+}
+
+/* End of IP1394 headers */
+
+/* Fragment types */
+#define ETH1394_HDR_LF_UF	0	/* unfragmented		*/
+#define ETH1394_HDR_LF_FF	1	/* first fragment	*/
+#define ETH1394_HDR_LF_LF	2	/* last fragment	*/
+#define ETH1394_HDR_LF_IF	3	/* interior fragment	*/
+
+#define IP1394_HW_ADDR_LEN	16	/* As per RFC		*/
+
+/* This list keeps track of what parts of the datagram have been filled in */
+struct ipv4_fragment_info {
+        struct list_head fragment_info;
+	u16 offset;
+	u16 len;
+};
+
+struct ipv4_partial_datagram {
+	struct list_head pdg_list;
+	struct list_head fragment_info;
+	struct sk_buff *skb;
+	/* FIXME Why not use skb->data? */
+	char *pbuf;
+	u16 datagram_label;
+	u16 ether_type;
+	u16 datagram_size;
+};
+
+/*
+ * We keep one of these for each IPv4 capable device attached to a fw_card.
+ * The list of them is stored in the fw_card structure rather than in the
+ * ipv4_priv because the remote IPv4 nodes may be probed before the card is,
+ * so we need a place to store them before the ipv4_priv structure is
+ * allocated.
+ */
+struct ipv4_node {
+	struct list_head ipv4_nodes;
+	/* guid of the remote node */
+	u64 guid;
+	/* FIFO address to transmit datagrams to, or INVALID_FIFO_ADDR */
+	u64 fifo;
+
+	spinlock_t pdg_lock;	/* partial datagram lock		*/
+	/* List of partial datagrams received from this node */
+	struct list_head pdg_list;
+	/* Number of entries in pdg_list at the moment */
+	unsigned pdg_size;
+
+	/* max payload to transmit to this remote node */
+	/* This already includes the IPV4_FRAG_HDR_SIZE overhead */
+	u16 max_payload;
+	/* outgoing datagram label */
+	u16 datagram_label;
+	/* Current node_id of the remote node */
+	u16 nodeid;
+	/* current generation of the remote node */
+	u8 generation;
+	/* max speed that this node can receive at */
+	u8 xmt_speed;
+};
+
+struct ipv4_priv {
+	spinlock_t lock;
+
+	enum ipv4_broadcast_state broadcast_state;
+	struct fw_iso_context *broadcast_rcv_context;
+	struct fw_iso_buffer broadcast_rcv_buffer;
+	void **broadcast_rcv_buffer_ptrs;
+	unsigned broadcast_rcv_next_ptr;
+	unsigned num_broadcast_rcv_ptrs;
+	unsigned rcv_buffer_size;
+	/*
+	 * This value is the maximum unfragmented datagram size that can be
+	 * sent by the hardware.  It already has the GASP overhead and the
+	 * unfragmented datagram header overhead calculated into it.
+	 */
+	unsigned broadcast_xmt_max_payload;
+	u16 broadcast_xmt_datagramlabel;
+
+	/*
+	 * The csr address that remote nodes must send datagrams to for us to
+	 * receive them.
+	 */
+	struct fw_address_handler handler;
+	u64 local_fifo;
+
+	/* Wake up to xmt	 */
+        /* struct work_struct wake;*/
+	/* List of packets to be sent */
+	struct list_head packet_list;
+	/*
+	 * List of packets that were broadcasted.  When we get an ISO interrupt
+	 * one of them has been sent
+	 */
+	struct list_head broadcasted_list;
+	/* List of packets that have been sent but not yet acked */
+	struct list_head sent_list;
+
+	struct fw_card *card;
+};
+
+/* This is our task struct. It's used for the packet complete callback.  */
+struct ipv4_packet_task {
+	/*
+	 * ptask can actually be on priv->packet_list, priv->broadcasted_list,
+	 * or priv->sent_list depending on its current state.
+	 */
+	struct list_head packet_list;
+	struct fw_transaction transaction;
+	struct ipv4_hdr hdr;
+	struct sk_buff *skb;
+	struct ipv4_priv *priv;
+	enum ipv4_tx_type tx_type;
+	int outstanding_pkts;
+	unsigned max_payload;
+	u64 fifo_addr;
+	u16 dest_node;
+	u8 generation;
+	u8 speed;
+};
+
+static struct kmem_cache *ipv4_packet_task_cache;
+
+static const char ipv4_driver_name[] = "firewire-ipv4";
+
+static const struct ieee1394_device_id ipv4_id_table[] = {
+	{
+		.match_flags  = IEEE1394_MATCH_SPECIFIER_ID |
+				IEEE1394_MATCH_VERSION,
+		.specifier_id = IPV4_GASP_SPECIFIER_ID,
+		.version      = IPV4_GASP_VERSION,
+	},
+	{ }
+};
+
+static u32 ipv4_unit_directory_data[] = {
+	0x00040000,					/* unit directory */
+	0x12000000 | IPV4_GASP_SPECIFIER_ID,	/* specifier ID */
+	0x81000003,					/* text descriptor */
+	0x13000000 | IPV4_GASP_VERSION,		/* version */
+	0x81000005,					/* text descriptor */
+
+	0x00030000,					/* Three quadlets */
+	0x00000000,					/* Text */
+	0x00000000,					/* Language 0 */
+	0x49414e41,					/* I A N A */
+	0x00030000,					/* Three quadlets */
+	0x00000000,					/* Text */
+	0x00000000,					/* Language 0 */
+	0x49507634,					/* I P v 4 */
+};
+
+static struct fw_descriptor ipv4_unit_directory = {
+	.length = ARRAY_SIZE(ipv4_unit_directory_data),
+	.key = 0xd1000000,
+	.data = ipv4_unit_directory_data
+};
+
+static int ipv4_send_packet(struct ipv4_packet_task *ptask );
+
+/* ------------------------------------------------------------------ */
+/******************************************
+ * HW Header net device functions
+ ******************************************/
+  /* These functions have been adapted from net/ethernet/eth.c */
+
+/* Create a fake MAC header for an arbitrary protocol layer.
+ * saddr=NULL means use device source address
+ * daddr=NULL means leave destination address (eg unresolved arp). */
+
+static int ipv4_header ( struct sk_buff *skb, struct net_device *dev,
+		       unsigned short type, const void *daddr,
+		       const void *saddr, unsigned len) {
+	struct ipv4_ether_hdr *eth;
+
+	eth = (struct ipv4_ether_hdr *)skb_push(skb, sizeof(*eth));
+	eth->h_proto = htons(type);
+
+	if (dev->flags & (IFF_LOOPBACK | IFF_NOARP)) {
+		memset(eth->h_dest, 0, dev->addr_len);
+		return dev->hard_header_len;
+	}
+
+	if (daddr) {
+		memcpy(eth->h_dest, daddr, dev->addr_len);
+		return dev->hard_header_len;
+	}
+
+	return -dev->hard_header_len;
+}
+
+/* Rebuild the faked MAC header. This is called after an ARP
+ * (or in future other address resolution) has completed on this
+ * sk_buff. We now let ARP fill in the other fields.
+ *
+ * This routine CANNOT use cached dst->neigh!
+ * Really, it is used only when dst->neigh is wrong.
+ */
+
+static int ipv4_rebuild_header(struct sk_buff *skb)
+{
+	struct ipv4_ether_hdr *eth;
+
+	eth = (struct ipv4_ether_hdr *)skb->data;
+	if (eth->h_proto == htons(ETH_P_IP))
+		return arp_find((unsigned char *)&eth->h_dest, skb);
+
+	fw_notify ( "%s: unable to resolve type %04x addresses\n",
+		   skb->dev->name,ntohs(eth->h_proto) );
+	return 0;
+}
+
+static int ipv4_header_cache(const struct neighbour *neigh, struct hh_cache *hh) {
+	unsigned short type = hh->hh_type;
+	struct net_device *dev;
+	struct ipv4_ether_hdr *eth;
+
+	if (type == htons(ETH_P_802_3))
+		return -1;
+	dev = neigh->dev;
+	eth = (struct ipv4_ether_hdr *)((u8 *)hh->hh_data + 16 - sizeof(*eth));
+	eth->h_proto = type;
+	memcpy(eth->h_dest, neigh->ha, dev->addr_len);
+
+	hh->hh_len = IPV4_HLEN;
+	return 0;
+}
+
+/* Called by Address Resolution module to notify changes in address. */
+static void ipv4_header_cache_update(struct hh_cache *hh, const struct net_device *dev, const unsigned char * haddr ) {
+	memcpy((u8 *)hh->hh_data + 16 - IPV4_HLEN, haddr, dev->addr_len);
+}
+
+static int ipv4_header_parse(const struct sk_buff *skb, unsigned char *haddr) {
+	memcpy(haddr, skb->dev->dev_addr, IPV4_ALEN);
+	return IPV4_ALEN;
+}
+
+static const struct header_ops ipv4_header_ops = {
+	.create         = ipv4_header,
+	.rebuild        = ipv4_rebuild_header,
+	.cache		= ipv4_header_cache,
+	.cache_update	= ipv4_header_cache_update,
+	.parse          = ipv4_header_parse,
+};
+
+/* ------------------------------------------------------------------ */
+
+/* FIXME: is this correct for all cases? */
+static bool ipv4_frag_overlap(struct ipv4_partial_datagram *pd, unsigned offset, unsigned len)
+{
+        struct ipv4_fragment_info *fi;
+	unsigned end = offset + len;
+
+	list_for_each_entry(fi, &pd->fragment_info, fragment_info) {
+		if (offset < fi->offset + fi->len && end > fi->offset) {
+			fw_debug ( "frag_overlap pd %p fi %p (%x@%x) with %x@%x\n", pd, fi, fi->len, fi->offset, len, offset );
+			return true;
+		}
+	}
+	fw_debug ( "frag_overlap %p does not overlap with %x@%x\n", pd, len, offset );
+	return false;
+}
+
+/* Assumes that new fragment does not overlap any existing fragments */
+static struct ipv4_fragment_info *ipv4_frag_new ( struct ipv4_partial_datagram *pd, unsigned offset, unsigned len ) {
+	struct ipv4_fragment_info *fi, *fi2, *new;
+	struct list_head *list;
+
+	fw_debug ( "frag_new pd %p %x@%x\n", pd, len, offset );
+	list = &pd->fragment_info;
+	list_for_each_entry(fi, &pd->fragment_info, fragment_info) {
+		if (fi->offset + fi->len == offset) {
+			/* The new fragment can be tacked on to the end */
+			/* Did the new fragment plug a hole? */
+			fi2 = list_entry(fi->fragment_info.next, struct ipv4_fragment_info, fragment_info);
+			if (fi->offset + fi->len == fi2->offset) {
+				fw_debug ( "pd %p: hole filling %p (%x@%x) and %p(%x@%x): now %x@%x\n", pd, fi, fi->len, fi->offset,
+				fi2, fi2->len, fi2->offset, fi->len + len + fi2->len, fi->offset );
+				/* glue fragments together */
+				fi->len += len + fi2->len;
+				list_del(&fi2->fragment_info);
+				kfree(fi2);
+			} else {
+				fw_debug ( "pd %p: extending %p from %x@%x to %x@%x\n", pd, fi, fi->len, fi->offset, fi->len+len, fi->offset );
+				fi->len += len;
+			}
+			return fi;
+		}
+		if (offset + len == fi->offset) {
+			/* The new fragment can be tacked on to the beginning */
+			/* Did the new fragment plug a hole? */
+			fi2 = list_entry(fi->fragment_info.prev, struct ipv4_fragment_info, fragment_info);
+			if (fi2->offset + fi2->len == fi->offset) {
+				/* glue fragments together */
+				fw_debug ( "pd %p: extending %p and merging with %p from %x@%x to %x@%x\n",
+ pd, fi2, fi, fi2->len, fi2->offset, fi2->len + fi->len + len, fi2->offset );
+				fi2->len += fi->len + len;
+				list_del(&fi->fragment_info);
+				kfree(fi);
+				return fi2;
+			}
+			fw_debug ( "pd %p: extending %p from %x@%x to %x@%x\n", pd, fi, fi->len, fi->offset, offset, fi->len + len );
+			fi->offset = offset;
+			fi->len += len;
+			return fi;
+		}
+		if (offset > fi->offset + fi->len) {
+			list = &fi->fragment_info;
+			break;
+		}
+		if (offset + len < fi->offset) {
+			list = fi->fragment_info.prev;
+			break;
+		}
+	}
+
+	new = kmalloc(sizeof(*new), GFP_ATOMIC);
+	if (!new) {
+		fw_error ( "out of memory in fragment handling!\n" );
+		return NULL;
+	}
+
+	new->offset = offset;
+	new->len = len;
+	list_add(&new->fragment_info, list);
+	fw_debug ( "pd %p: new frag %p %x@%x\n", pd, new, new->len, new->offset );
+	list_for_each_entry( fi, &pd->fragment_info, fragment_info )
+		fw_debug ( "fi %p %x@%x\n", fi, fi->len, fi->offset );
+	return new;
+}
+
+/* ------------------------------------------------------------------ */
+
+static struct ipv4_partial_datagram *ipv4_pd_new(struct net_device *netdev,
+ struct ipv4_node *node, u16 datagram_label, unsigned dg_size, u32 *frag_buf,
+ unsigned frag_off, unsigned frag_len) {
+	struct ipv4_partial_datagram *new;
+	struct ipv4_fragment_info *fi;
+
+	new = kmalloc(sizeof(*new), GFP_ATOMIC);
+	if (!new)
+		goto fail;
+	INIT_LIST_HEAD(&new->fragment_info);
+	fi = ipv4_frag_new ( new, frag_off, frag_len);
+	if ( fi == NULL )
+		goto fail_w_new;
+	new->datagram_label = datagram_label;
+	new->datagram_size = dg_size;
+	new->skb = dev_alloc_skb(dg_size + netdev->hard_header_len + 15);
+	if ( new->skb == NULL )
+		goto fail_w_fi;
+	skb_reserve(new->skb, (netdev->hard_header_len + 15) & ~15);
+	new->pbuf = skb_put(new->skb, dg_size);
+	memcpy(new->pbuf + frag_off, frag_buf, frag_len);
+	list_add_tail(&new->pdg_list, &node->pdg_list);
+	fw_debug ( "pd_new: new pd %p { dgl %u, dg_size %u, skb %p, pbuf %p } on node %p\n",
+ new, new->datagram_label, new->datagram_size, new->skb, new->pbuf, node );
+	return new;
+
+fail_w_fi:
+	kfree(fi);
+fail_w_new:
+	kfree(new);
+fail:
+	fw_error("ipv4_pd_new: no memory\n");
+	return NULL;
+}
+
+static struct ipv4_partial_datagram *ipv4_pd_find(struct ipv4_node *node, u16 datagram_label) {
+	struct ipv4_partial_datagram *pd;
+
+	list_for_each_entry(pd, &node->pdg_list, pdg_list) {
+	        if ( pd->datagram_label == datagram_label ) {
+			fw_debug ( "pd_find(node %p, label %u): pd %p\n", node, datagram_label, pd );
+			return pd;
+		}
+	}
+	fw_debug ( "pd_find(node %p, label %u) no entry\n", node, datagram_label );
+	return NULL;
+}
+
+
+static void ipv4_pd_delete ( struct ipv4_partial_datagram *old ) {
+	struct ipv4_fragment_info *fi, *n;
+
+	fw_debug ( "pd_delete %p\n", old );
+	list_for_each_entry_safe(fi, n, &old->fragment_info, fragment_info) {
+		fw_debug ( "Freeing fi %p\n", fi );
+		kfree(fi);
+	}
+	list_del(&old->pdg_list);
+	dev_kfree_skb_any(old->skb);
+	kfree(old);
+}
+
+static bool ipv4_pd_update ( struct ipv4_node *node, struct ipv4_partial_datagram *pd,
+ u32 *frag_buf, unsigned frag_off, unsigned frag_len) {
+	fw_debug ( "pd_update node %p, pd %p, frag_buf %p, %x@%x\n", node, pd, frag_buf, frag_len, frag_off );
+	if ( ipv4_frag_new ( pd, frag_off, frag_len ) == NULL)
+		return false;
+	memcpy(pd->pbuf + frag_off, frag_buf, frag_len);
+
+	/*
+	 * Move list entry to beginnig of list so that oldest partial
+	 * datagrams percolate to the end of the list
+	 */
+	list_move_tail(&pd->pdg_list, &node->pdg_list);
+	fw_debug ( "New pd list:\n" );
+	list_for_each_entry ( pd, &node->pdg_list, pdg_list ) {
+		fw_debug ( "pd %p\n", pd );
+	}
+	return true;
+}
+
+static bool ipv4_pd_is_complete ( struct ipv4_partial_datagram *pd ) {
+	struct ipv4_fragment_info *fi;
+	bool ret;
+
+	fi = list_entry(pd->fragment_info.next, struct ipv4_fragment_info, fragment_info);
+
+	ret = (fi->len == pd->datagram_size);
+	fw_debug ( "pd_is_complete (pd %p, dgs %x): fi %p (%x@%x) %s\n", pd, pd->datagram_size, fi, fi->len, fi->offset, ret ? "yes" : "no" );
+	return ret;
+}
+
+/* ------------------------------------------------------------------ */
+
+static int ipv4_node_new ( struct fw_card *card, struct fw_device *device ) {
+	struct ipv4_node *node;
+
+	node = kmalloc ( sizeof(*node), GFP_KERNEL );
+	if ( ! node ) {
+		fw_error ( "allocate new node failed\n" );
+		return -ENOMEM;
+	}
+	node->guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
+	node->fifo = INVALID_FIFO_ADDR;
+	INIT_LIST_HEAD(&node->pdg_list);
+	spin_lock_init(&node->pdg_lock);
+	node->pdg_size = 0;
+	node->generation = device->generation;
+	rmb();
+	node->nodeid = device->node_id;
+	 /* FIXME what should it really be? */
+	node->max_payload = S100_BUFFER_SIZE - IPV4_UNFRAG_HDR_SIZE;
+	node->datagram_label = 0U;
+	node->xmt_speed = device->max_speed;
+	list_add_tail ( &node->ipv4_nodes, &card->ipv4_nodes );
+	fw_debug ( "node_new: %p { guid %016llx, generation %u, nodeid %x, max_payload %x, xmt_speed %x } added\n",
+ node, (unsigned long long)node->guid, node->generation, node->nodeid, node->max_payload, node->xmt_speed );
+	return 0;
+}
+
+static struct ipv4_node *ipv4_node_find_by_guid(struct ipv4_priv *priv, u64 guid) {
+	struct ipv4_node *node;
+	unsigned long flags;
+
+	spin_lock_irqsave(&priv->lock, flags);
+	list_for_each_entry(node, &priv->card->ipv4_nodes, ipv4_nodes)
+		if (node->guid == guid) {
+			/* FIXME: lock the node first? */
+			spin_unlock_irqrestore ( &priv->lock, flags );
+			fw_debug ( "node_find_by_guid (%016llx) found %p\n", (unsigned long long)guid, node );
+			return node;
+		}
+
+	spin_unlock_irqrestore ( &priv->lock, flags );
+	fw_debug ( "node_find_by_guid (%016llx) not found\n", (unsigned long long)guid );
+	return NULL;
+}
+
+static struct ipv4_node *ipv4_node_find_by_nodeid(struct ipv4_priv *priv, u16 nodeid) {
+	struct ipv4_node *node;
+	unsigned long flags;
+
+	spin_lock_irqsave(&priv->lock, flags);
+	list_for_each_entry(node, &priv->card->ipv4_nodes, ipv4_nodes)
+		if (node->nodeid == nodeid) {
+			/* FIXME: lock the node first? */
+			spin_unlock_irqrestore ( &priv->lock, flags );
+			fw_debug ( "node_find_by_nodeid (%x) found %p\n", nodeid, node );
+			return node;
+		}
+	fw_debug ( "node_find_by_nodeid (%x) not found\n", nodeid );
+	spin_unlock_irqrestore ( &priv->lock, flags );
+	return NULL;
+}
+
+/* This is only complicated because we can't assume priv exists */
+static void ipv4_node_delete ( struct fw_card *card, struct fw_device *device ) {
+	struct net_device *netdev;
+	struct ipv4_priv *priv;
+	struct ipv4_node *node;
+	u64 guid;
+	unsigned long flags;
+	struct ipv4_partial_datagram *pd, *pd_next;
+
+	guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
+	netdev = card->netdev;
+	if ( netdev )
+		priv = netdev_priv ( netdev );
+	else
+		priv = NULL;
+	if ( priv )
+		spin_lock_irqsave ( &priv->lock, flags );
+	list_for_each_entry( node, &card->ipv4_nodes, ipv4_nodes ) {
+		if ( node->guid == guid ) {
+			list_del ( &node->ipv4_nodes );
+			list_for_each_entry_safe( pd, pd_next, &node->pdg_list, pdg_list )
+				ipv4_pd_delete ( pd );
+			break;
+		}
+	}
+	if ( priv )
+		spin_unlock_irqrestore ( &priv->lock, flags );
+}
+
+/* ------------------------------------------------------------------ */
+
+
+static int ipv4_finish_incoming_packet ( struct net_device *netdev,
+ struct sk_buff *skb, u16 source_node_id, bool is_broadcast, u16 ether_type ) {
+	struct ipv4_priv *priv;
+	static u64 broadcast_hw = ~0ULL;
+	int status;
+	u64 guid;
+
+	fw_debug ( "ipv4_finish_incoming_packet(%p, %p, %x, %s, %x\n",
+ netdev, skb, source_node_id, is_broadcast ? "true" : "false", ether_type );
+	priv = netdev_priv(netdev);
+	/* Write metadata, and then pass to the receive level */
+	skb->dev = netdev;
+	skb->ip_summed = CHECKSUM_UNNECESSARY;  /* don't check it */
+
+	/*
+	 * Parse the encapsulation header. This actually does the job of
+	 * converting to an ethernet frame header, as well as arp
+	 * conversion if needed. ARP conversion is easier in this
+	 * direction, since we are using ethernet as our backend.
+	 */
+	/*
+	 * If this is an ARP packet, convert it. First, we want to make
+	 * use of some of the fields, since they tell us a little bit
+	 * about the sending machine.
+	 */
+	if (ether_type == ETH_P_ARP) {
+		struct ipv4_arp *arp1394;
+		struct arphdr *arp;
+		unsigned char *arp_ptr;
+		u64 fifo_addr;
+		u8 max_rec;
+		u8 sspd;
+		u16 max_payload;
+		struct ipv4_node *node;
+		static const u16 ipv4_speed_to_max_payload[] = {
+			/* S100, S200, S400, S800, S1600, S3200 */
+			    512, 1024, 2048, 4096,  4096,  4096
+		};
+
+		/* fw_debug ( "ARP packet\n" ); */
+		arp1394 = (struct ipv4_arp *)skb->data;
+		arp = (struct arphdr *)skb->data;
+		arp_ptr = (unsigned char *)(arp + 1);
+		fifo_addr = (u64)ntohs(arp1394->fifo_hi) << 32 |
+ ntohl(arp1394->fifo_lo);
+		max_rec = priv->card->max_receive;
+		if ( arp1394->max_rec < max_rec )
+			max_rec = arp1394->max_rec;
+		sspd = arp1394->sspd;
+		/*
+		 * Sanity check. MacOSX seems to be sending us 131 in this
+		 * field (atleast on my Panther G5). Not sure why.
+		 */
+		if (sspd > 5 ) {
+			fw_notify ( "sspd %x out of range\n", sspd );
+			sspd = 0;
+		}
+
+		max_payload = min(ipv4_speed_to_max_payload[sspd],
+ (u16)(1 << (max_rec + 1))) - IPV4_UNFRAG_HDR_SIZE;
+
+		guid = be64_to_cpu(get_unaligned(&arp1394->s_uniq_id));
+		node = ipv4_node_find_by_guid(priv, guid);
+		if (!node) {
+			fw_notify ( "No node for ARP packet from %llx\n", guid );
+			goto failed_proto;
+		}
+		if ( node->nodeid != source_node_id || node->generation != priv->card->generation ) {
+			fw_notify ( "Internal error: node->nodeid (%x) != soucre_node_id (%x) or node->generation (%x) != priv->card->generation(%x)\n",
+ node->nodeid, source_node_id, node->generation, priv->card->generation );
+			node->nodeid = source_node_id;
+			node->generation = priv->card->generation;
+		}
+
+		/* FIXME: for debugging */
+		if ( sspd > SCODE_400 )
+			sspd = SCODE_400;
+		/* Update our speed/payload/fifo_offset table */
+		/*
+		 * FIXME: this does not handle cases where two high-speed endpoints must use a slower speed because of
+		 * a lower speed hub between them.  We need to look at the actual topology map here.
+		 */
+		fw_debug ( "Setting node %p fifo %llx (was %llx), max_payload %x (was %x), speed %x (was %x)\n",
+ node, fifo_addr, node->fifo, max_payload, node->max_payload, sspd, node->xmt_speed );
+		node->fifo =	fifo_addr;
+		node->max_payload = max_payload;
+		/*
+		 * Only allow speeds to go down from their initial value.
+		 * Otherwise a local node that can only do S400 or slower may
+		 * be told to transmit at S800 to a faster remote node.
+		 */
+		if ( node->xmt_speed > sspd )
+			node->xmt_speed = sspd;
+
+		/*
+		 * Now that we're done with the 1394 specific stuff, we'll
+		 * need to alter some of the data.  Believe it or not, all
+		 * that needs to be done is sender_IP_address needs to be
+		 * moved, the destination hardware address get stuffed
+		 * in and the hardware address length set to 8.
+		 *
+		 * IMPORTANT: The code below overwrites 1394 specific data
+		 * needed above so keep the munging of the data for the
+		 * higher level IP stack last.
+		 */
+
+		arp->ar_hln = 8;
+		arp_ptr += arp->ar_hln;		/* skip over sender unique id */
+		*(u32 *)arp_ptr = arp1394->sip; /* move sender IP addr */
+		arp_ptr += arp->ar_pln;		/* skip over sender IP addr */
+
+		if (arp->ar_op == htons(ARPOP_REQUEST))
+			memset(arp_ptr, 0, sizeof(u64));
+		else
+			memcpy(arp_ptr, netdev->dev_addr, sizeof(u64));
+	}
+
+	/* Now add the ethernet header. */
+	guid = cpu_to_be64(priv->card->guid);
+	if (dev_hard_header(skb, netdev, ether_type, is_broadcast ? &broadcast_hw : &guid, NULL,
+ skb->len) >= 0) {
+		struct ipv4_ether_hdr *eth;
+		u16 *rawp;
+		__be16 protocol;
+
+		skb_reset_mac_header(skb);
+		skb_pull(skb, sizeof(*eth));
+		eth = ipv4_ether_hdr(skb);
+		if (*eth->h_dest & 1) {
+			if (memcmp(eth->h_dest, netdev->broadcast, netdev->addr_len) == 0) {
+				fw_debug ( "Broadcast\n" );
+				skb->pkt_type = PACKET_BROADCAST;
+			}
+#if 0
+			else
+				skb->pkt_type = PACKET_MULTICAST;
+#endif
+		} else {
+			if (memcmp(eth->h_dest, netdev->dev_addr, netdev->addr_len)) {
+				u64 a1, a2;
+
+				memcpy ( &a1, eth->h_dest, sizeof(u64));
+				memcpy ( &a2, netdev->dev_addr, sizeof(u64));
+				fw_debug ( "Otherhost %llx %llx %x\n", a1, a2, netdev->addr_len );
+				skb->pkt_type = PACKET_OTHERHOST;
+			}
+		}
+		if (ntohs(eth->h_proto) >= 1536) {
+			fw_debug ( " proto %x %x\n", eth->h_proto, ntohs(eth->h_proto) );
+			protocol = eth->h_proto;
+		} else {
+			rawp = (u16 *)skb->data;
+			if (*rawp == 0xFFFF) {
+				fw_debug ( "proto 802_3\n" );
+				protocol = htons(ETH_P_802_3);
+			} else {
+				fw_debug ( "proto 802_2\n" );
+				protocol = htons(ETH_P_802_2);
+			}
+		}
+		skb->protocol = protocol;
+	}
+	status = netif_rx(skb);
+	if ( status == NET_RX_DROP) {
+		netdev->stats.rx_errors++;
+		netdev->stats.rx_dropped++;
+	} else {
+		netdev->stats.rx_packets++;
+		netdev->stats.rx_bytes += skb->len;
+	}
+	if (netif_queue_stopped(netdev))
+		netif_wake_queue(netdev);
+	return 0;
+
+ failed_proto:
+	netdev->stats.rx_errors++;
+	netdev->stats.rx_dropped++;
+	dev_kfree_skb_any(skb);
+	if (netif_queue_stopped(netdev))
+		netif_wake_queue(netdev);
+	netdev->last_rx = jiffies;
+	return 0;
+}
+
+/* ------------------------------------------------------------------ */
+
+static int ipv4_incoming_packet ( struct ipv4_priv *priv, u32 *buf, int len, u16 source_node_id, bool is_broadcast ) {
+	struct sk_buff *skb;
+	struct net_device *netdev;
+	struct ipv4_hdr hdr;
+	unsigned lf;
+	unsigned long flags;
+	struct ipv4_node *node;
+	struct ipv4_partial_datagram *pd;
+	int fg_off;
+	int dg_size;
+	u16 datagram_label;
+	int retval;
+	u16 ether_type;
+
+	fw_debug ( "ipv4_incoming_packet(%p, %p, %d, %x, %s)\n", priv, buf, len, source_node_id, is_broadcast ? "true" : "false" );
+	netdev = priv->card->netdev;
+
+	hdr.w0 = ntohl(buf[0]);
+	lf = ipv4_get_hdr_lf(&hdr);
+	if ( lf == IPV4_HDR_UNFRAG ) {
+		/*
+		 * An unfragmented datagram has been received by the ieee1394
+		 * bus. Build an skbuff around it so we can pass it to the
+		 * high level network layer.
+		 */
+		ether_type = ipv4_get_hdr_ether_type(&hdr);
+		fw_debug ( "header w0 = %x, lf = %x, ether_type = %x\n", hdr.w0, lf, ether_type );
+		buf++;
+		len -= IPV4_UNFRAG_HDR_SIZE;
+
+		skb = dev_alloc_skb(len + netdev->hard_header_len + 15);
+		if (unlikely(!skb)) {
+			fw_error ( "Out of memory for incoming packet\n");
+			netdev->stats.rx_dropped++;
+			return -1;
+		}
+		skb_reserve(skb, (netdev->hard_header_len + 15) & ~15);
+		memcpy(skb_put(skb, len), buf, len );
+		return ipv4_finish_incoming_packet(netdev, skb, source_node_id, is_broadcast, ether_type );
+	}
+	/* A datagram fragment has been received, now the fun begins. */
+	hdr.w1 = ntohl(buf[1]);
+	buf +=2;
+	len -= IPV4_FRAG_HDR_SIZE;
+	if ( lf ==IPV4_HDR_FIRSTFRAG ) {
+		ether_type = ipv4_get_hdr_ether_type(&hdr);
+		fg_off = 0;
+	} else {
+		fg_off = ipv4_get_hdr_fg_off(&hdr);
+		ether_type = 0; /* Shut up compiler! */
+	}
+	datagram_label = ipv4_get_hdr_dgl(&hdr);
+	dg_size = ipv4_get_hdr_dg_size(&hdr); /* ??? + 1 */
+	fw_debug ( "fragmented: %x.%x = lf %x, ether_type %x, fg_off %x, dgl %x, dg_size %x\n", hdr.w0, hdr.w1, lf, ether_type, fg_off, datagram_label, dg_size );
+	node = ipv4_node_find_by_nodeid ( priv, source_node_id);
+	spin_lock_irqsave(&node->pdg_lock, flags);
+	pd = ipv4_pd_find( node, datagram_label );
+	if (pd == NULL) {
+		while ( node->pdg_size >= ipv4_mpd ) {
+			/* remove the oldest */
+			ipv4_pd_delete ( list_first_entry(&node->pdg_list, struct ipv4_partial_datagram, pdg_list) );
+			node->pdg_size--;
+		}
+		pd = ipv4_pd_new ( netdev, node, datagram_label, dg_size,
+ buf, fg_off, len);
+		if ( pd == NULL) {
+			retval = -ENOMEM;
+			goto bad_proto;
+		}
+		node->pdg_size++;
+	} else {
+		if (ipv4_frag_overlap(pd, fg_off, len) || pd->datagram_size != dg_size) {
+			/*
+			 * Differing datagram sizes or overlapping fragments,
+			 * Either way the remote machine is playing silly buggers
+			 * with us: obliterate the old datagram and start a new one.
+			 */
+			ipv4_pd_delete ( pd );
+			pd = ipv4_pd_new ( netdev, node, datagram_label,
+ dg_size, buf, fg_off, len);
+			if ( pd == NULL ) {
+				retval = -ENOMEM;
+				node->pdg_size--;
+				goto bad_proto;
+			}
+		} else {
+			bool worked;
+
+			worked = ipv4_pd_update ( node, pd,
+ buf, fg_off, len );
+			if ( ! worked ) {
+				/*
+				 * Couldn't save off fragment anyway
+				 * so might as well obliterate the
+				 * datagram now.
+				 */
+				ipv4_pd_delete ( pd );
+				node->pdg_size--;
+				goto bad_proto;
+			}
+		}
+	} /* new datagram or add to existing one */
+
+	if ( lf == IPV4_HDR_FIRSTFRAG )
+		pd->ether_type = ether_type;
+	if ( ipv4_pd_is_complete ( pd ) ) {
+		ether_type = pd->ether_type;
+		node->pdg_size--;
+		skb = skb_get(pd->skb);
+		ipv4_pd_delete ( pd );
+		spin_unlock_irqrestore(&node->pdg_lock, flags);
+		return ipv4_finish_incoming_packet ( netdev, skb, source_node_id, false, ether_type );
+	}
+	/*
+	 * Datagram is not complete, we're done for the
+	 * moment.
+	 */
+	spin_unlock_irqrestore(&node->pdg_lock, flags);
+	return 0;
+
+ bad_proto:
+	spin_unlock_irqrestore(&node->pdg_lock, flags);
+	if (netif_queue_stopped(netdev))
+		netif_wake_queue(netdev);
+	return 0;
+}
+
+static void ipv4_receive_packet ( struct fw_card *card, struct fw_request *r,
+ int tcode, int destination, int source, int generation, int speed,
+ unsigned long long offset, void *payload, size_t length, void *callback_data ) {
+	struct ipv4_priv *priv;
+	int status;
+
+	fw_debug ( "ipv4_receive_packet(%p,%p,%x,%x,%x,%x,%x,%llx,%p,%lx,%p)\n",
+ card, r, tcode, destination, source, generation, speed, offset, payload,
+ (unsigned long)length, callback_data);
+	print_hex_dump ( KERN_DEBUG, "header: ", DUMP_PREFIX_OFFSET, 32, 1, payload, length, false );
+	priv = callback_data;
+	if (   tcode != TCODE_WRITE_BLOCK_REQUEST
+	    || destination != card->node_id
+	    || generation != card->generation
+	    || offset != priv->handler.offset ) {
+		fw_send_response(card, r, RCODE_CONFLICT_ERROR);
+		fw_debug("Conflict error card node_id=%x, card generation=%x, local offset %llx\n",
+ card->node_id, card->generation, (unsigned long long)priv->handler.offset );
+		return;
+	}
+	status = ipv4_incoming_packet ( priv, payload, length, source, false );
+	if ( status != 0 ) {
+		fw_error ( "Incoming packet failure\n" );
+		fw_send_response ( card, r, RCODE_CONFLICT_ERROR );
+		return;
+	}
+	fw_send_response ( card, r, RCODE_COMPLETE );
+}
+
+static void ipv4_receive_broadcast(struct fw_iso_context *context, u32 cycle,
+ size_t header_length, void *header, void *data) {
+	struct ipv4_priv *priv;
+	struct fw_iso_packet packet;
+	struct fw_card *card;
+	u16 *hdr_ptr;
+	u32 *buf_ptr;
+	int retval;
+	u32 length;
+	u16 source_node_id;
+	u32 specifier_id;
+	u32 ver;
+	unsigned long offset;
+	unsigned long flags;
+
+	fw_debug ( "ipv4_receive_broadcast ( context=%p, cycle=%x, header_length=%lx, header=%p, data=%p )\n", context, cycle, (unsigned long)header_length, header, data );
+	print_hex_dump ( KERN_DEBUG, "header: ", DUMP_PREFIX_OFFSET, 32, 1, header, header_length, false );
+	priv = data;
+	card = priv->card;
+	hdr_ptr = header;
+	length = ntohs(hdr_ptr[0]);
+	spin_lock_irqsave(&priv->lock,flags);
+	offset = priv->rcv_buffer_size * priv->broadcast_rcv_next_ptr;
+	buf_ptr = priv->broadcast_rcv_buffer_ptrs[priv->broadcast_rcv_next_ptr++];
+	if ( priv->broadcast_rcv_next_ptr == priv->num_broadcast_rcv_ptrs )
+		priv->broadcast_rcv_next_ptr = 0;
+	spin_unlock_irqrestore(&priv->lock,flags);
+	fw_debug ( "length %u at %p\n", length, buf_ptr );
+	print_hex_dump ( KERN_DEBUG, "buffer: ", DUMP_PREFIX_OFFSET, 32, 1, buf_ptr, length, false );
+
+	specifier_id =    (be32_to_cpu(buf_ptr[0]) & 0xffff) << 8
+			| (be32_to_cpu(buf_ptr[1]) & 0xff000000) >> 24;
+	ver = be32_to_cpu(buf_ptr[1]) & 0xFFFFFF;
+	source_node_id = be32_to_cpu(buf_ptr[0]) >> 16;
+	/* fw_debug ( "source %x SpecID %x ver %x\n", source_node_id, specifier_id, ver ); */
+	if ( specifier_id == IPV4_GASP_SPECIFIER_ID && ver == IPV4_GASP_VERSION ) {
+		buf_ptr += 2;
+		length -= IPV4_GASP_OVERHEAD;
+		ipv4_incoming_packet(priv, buf_ptr, length, source_node_id, true);
+	} else
+		fw_debug ( "Ignoring packet: not GASP\n" );
+	packet.payload_length = priv->rcv_buffer_size;
+	packet.interrupt = 1;
+	packet.skip = 0;
+	packet.tag = 3;
+	packet.sy = 0;
+	packet.header_length = IPV4_GASP_OVERHEAD;
+	spin_lock_irqsave(&priv->lock,flags);
+	retval = fw_iso_context_queue ( priv->broadcast_rcv_context, &packet,
+ &priv->broadcast_rcv_buffer, offset );
+	spin_unlock_irqrestore(&priv->lock,flags);
+	if ( retval < 0 )
+		fw_error ( "requeue failed\n" );
+}
+
+static void debug_ptask ( struct ipv4_packet_task *ptask ) {
+	static const char *tx_types[] = { "Unknown", "GASP", "Write" };
+
+	fw_debug ( "packet %p { hdr { w0 %x w1 %x }, skb %p, priv %p,"
+ " tx_type %s, outstanding_pkts %d, max_payload %x, fifo %llx,"
+ " speed %x, dest_node %x, generation %x }\n",
+ ptask, ptask->hdr.w0, ptask->hdr.w1, ptask->skb, ptask->priv,
+ ptask->tx_type > IPV4_WRREQ ? "Invalid" : tx_types[ptask->tx_type],
+ ptask->outstanding_pkts,  ptask->max_payload,
+ ptask->fifo_addr, ptask->speed, ptask->dest_node, ptask->generation );
+	print_hex_dump ( KERN_DEBUG, "packet :", DUMP_PREFIX_OFFSET, 32, 1,
+ ptask->skb->data, ptask->skb->len, false );
+}
+
+static void ipv4_transmit_packet_done ( struct ipv4_packet_task *ptask ) {
+	struct ipv4_priv *priv;
+	unsigned long flags;
+
+	priv = ptask->priv;
+	spin_lock_irqsave ( &priv->lock, flags );
+	list_del ( &ptask->packet_list );
+	spin_unlock_irqrestore ( &priv->lock, flags );
+	ptask->outstanding_pkts--;
+	if ( ptask->outstanding_pkts > 0 ) {
+		u16 dg_size;
+		u16 fg_off;
+		u16 datagram_label;
+		u16 lf;
+		struct sk_buff *skb;
+
+		/* Update the ptask to point to the next fragment and send it */
+		lf = ipv4_get_hdr_lf(&ptask->hdr);
+		switch (lf) {
+		case IPV4_HDR_LASTFRAG:
+		case IPV4_HDR_UNFRAG:
+		default:
+			fw_error ( "Outstanding packet %x lf %x, header %x,%x\n", ptask->outstanding_pkts, lf, ptask->hdr.w0, ptask->hdr.w1 );
+			BUG();
+
+		case IPV4_HDR_FIRSTFRAG:
+			/* Set frag type here for future interior fragments */
+			dg_size = ipv4_get_hdr_dg_size(&ptask->hdr);
+			fg_off = ptask->max_payload - IPV4_FRAG_HDR_SIZE;
+			datagram_label = ipv4_get_hdr_dgl(&ptask->hdr);
+			break;
+
+		case IPV4_HDR_INTFRAG:
+			dg_size = ipv4_get_hdr_dg_size(&ptask->hdr);
+			fg_off = ipv4_get_hdr_fg_off(&ptask->hdr) + ptask->max_payload - IPV4_FRAG_HDR_SIZE;
+			datagram_label = ipv4_get_hdr_dgl(&ptask->hdr);
+			break;
+		}
+		skb = ptask->skb;
+		skb_pull ( skb, ptask->max_payload );
+		if ( ptask->outstanding_pkts > 1 ) {
+			ipv4_make_sf_hdr ( &ptask->hdr,
+  IPV4_HDR_INTFRAG, dg_size, fg_off, datagram_label );
+		} else {
+			ipv4_make_sf_hdr ( &ptask->hdr,
+  IPV4_HDR_LASTFRAG, dg_size, fg_off, datagram_label );
+			ptask->max_payload = skb->len + IPV4_FRAG_HDR_SIZE;
+
+		}
+		ipv4_send_packet ( ptask );
+	} else {
+		dev_kfree_skb_any ( ptask->skb );
+		kmem_cache_free( ipv4_packet_task_cache, ptask );
+	}
+}
+
+static void ipv4_write_complete ( struct fw_card *card, int rcode,
+ void *payload, size_t length, void *data ) {
+	struct ipv4_packet_task *ptask;
+
+	ptask = data;
+	fw_debug ( "ipv4_write_complete ( %p, %x, %p, %lx, %p )\n",
+ card, rcode, payload, (unsigned long)length, data );
+	debug_ptask ( ptask );
+
+	if ( rcode == RCODE_COMPLETE ) {
+		ipv4_transmit_packet_done ( ptask );
+	} else {
+		fw_error ( "ipv4_write_complete: failed: %x\n", rcode );
+		/* ??? error recovery */
+	}
+}
+
+static int ipv4_send_packet ( struct ipv4_packet_task *ptask ) {
+	struct ipv4_priv *priv;
+	unsigned tx_len;
+	struct ipv4_hdr *bufhdr;
+	unsigned long flags;
+	struct net_device *netdev;
+#if 0 /* stefanr */
+	int retval;
+#endif
+
+	fw_debug ( "ipv4_send_packet\n" );
+	debug_ptask ( ptask );
+	priv = ptask->priv;
+	tx_len = ptask->max_payload;
+	switch (ipv4_get_hdr_lf(&ptask->hdr)) {
+	case IPV4_HDR_UNFRAG:
+		bufhdr = (struct ipv4_hdr *)skb_push(ptask->skb, IPV4_UNFRAG_HDR_SIZE);
+		bufhdr->w0 = htonl(ptask->hdr.w0);
+		break;
+
+	case IPV4_HDR_FIRSTFRAG:
+	case IPV4_HDR_INTFRAG:
+	case IPV4_HDR_LASTFRAG:
+		bufhdr = (struct ipv4_hdr *)skb_push(ptask->skb, IPV4_FRAG_HDR_SIZE);
+		bufhdr->w0 = htonl(ptask->hdr.w0);
+		bufhdr->w1 = htonl(ptask->hdr.w1);
+		break;
+
+	default:
+		BUG();
+	}
+	if ( ptask->tx_type == IPV4_GASP ) {
+		u32 *packets;
+		int generation;
+		int nodeid;
+
+		/* ptask->generation may not have been set yet */
+		generation = priv->card->generation;
+		smp_rmb();
+		nodeid = priv->card->node_id;
+		packets = (u32 *)skb_push(ptask->skb, sizeof(u32)*2);
+		packets[0] = htonl(nodeid << 16 | (IPV4_GASP_SPECIFIER_ID>>8));
+		packets[1] = htonl((IPV4_GASP_SPECIFIER_ID & 0xFF) << 24 | IPV4_GASP_VERSION);
+		fw_send_request ( priv->card, &ptask->transaction, TCODE_STREAM_DATA,
+ fw_stream_packet_destination_id(3, BROADCAST_CHANNEL, 0),
+ generation, SCODE_100, 0ULL, ptask->skb->data, tx_len + 8, ipv4_write_complete, ptask );
+		spin_lock_irqsave(&priv->lock,flags);
+		list_add_tail ( &ptask->packet_list, &priv->broadcasted_list );
+		spin_unlock_irqrestore(&priv->lock,flags);
+#if 0 /* stefanr */
+		return retval;
+#else
+		return 0;
+#endif
+	}
+	fw_debug("send_request (%p, %p, WRITE_BLOCK, %x, %x, %x, %llx, %p, %d, %p, %p\n",
+ priv->card, &ptask->transaction, ptask->dest_node, ptask->generation,
+ ptask->speed, (unsigned long long)ptask->fifo_addr, ptask->skb->data, tx_len,
+ ipv4_write_complete, ptask );
+	fw_send_request ( priv->card, &ptask->transaction,
+ TCODE_WRITE_BLOCK_REQUEST, ptask->dest_node, ptask->generation, ptask->speed,
+ ptask->fifo_addr, ptask->skb->data, tx_len, ipv4_write_complete, ptask );
+	spin_lock_irqsave(&priv->lock,flags);
+	list_add_tail ( &ptask->packet_list, &priv->sent_list );
+	spin_unlock_irqrestore(&priv->lock,flags);
+	netdev = priv->card->netdev;
+	netdev->trans_start = jiffies;
+	return 0;
+}
+
+static int ipv4_broadcast_start ( struct ipv4_priv *priv ) {
+	struct fw_iso_context *context;
+	int retval;
+	unsigned num_packets;
+	unsigned max_receive;
+	struct fw_iso_packet packet;
+	unsigned long offset;
+	unsigned u;
+	/* unsigned transmit_speed; */
+
+#if 0 /* stefanr */
+	if ( priv->card->broadcast_channel != (BROADCAST_CHANNEL_VALID|BROADCAST_CHANNEL_INITIAL)) {
+		fw_notify ( "Invalid broadcast channel %x\n", priv->card->broadcast_channel );
+		/* FIXME: try again later? */
+		/* return -EINVAL; */
+	}
+#endif
+	if ( priv->local_fifo == INVALID_FIFO_ADDR ) {
+		struct fw_address_region region;
+
+		priv->handler.length = FIFO_SIZE;
+		priv->handler.address_callback = ipv4_receive_packet;
+		priv->handler.callback_data = priv;
+		/* FIXME: this is OHCI, but what about others? */
+		region.start = 0xffff00000000ULL;
+		region.end =   0xfffffffffffcULL;
+
+		retval = fw_core_add_address_handler ( &priv->handler, &region );
+		if ( retval < 0 )
+			goto failed_initial;
+		priv->local_fifo = priv->handler.offset;
+	}
+
+	/*
+	 * FIXME: rawiso limits us to PAGE_SIZE.  This only matters if we ever have
+	 * a machine with PAGE_SIZE < 4096
+	 */
+	max_receive = 1U << (priv->card->max_receive + 1);
+	num_packets = ( ipv4_iso_page_count * PAGE_SIZE ) / max_receive;
+	if ( ! priv->broadcast_rcv_context ) {
+		void **ptrptr;
+
+		context = fw_iso_context_create ( priv->card,
+ FW_ISO_CONTEXT_RECEIVE, BROADCAST_CHANNEL,
+ priv->card->link_speed, 8, ipv4_receive_broadcast, priv );
+		if (IS_ERR(context)) {
+			retval = PTR_ERR(context);
+			goto failed_context_create;
+		}
+		retval = fw_iso_buffer_init ( &priv->broadcast_rcv_buffer,
+ priv->card, ipv4_iso_page_count, DMA_FROM_DEVICE );
+		if ( retval < 0 )
+			goto failed_buffer_init;
+		ptrptr = kmalloc ( sizeof(void*)*num_packets, GFP_KERNEL );
+		if ( ! ptrptr ) {
+			retval = -ENOMEM;
+			goto failed_ptrs_alloc;
+		}
+		priv->broadcast_rcv_buffer_ptrs = ptrptr;
+		for ( u = 0; u < ipv4_iso_page_count; u++ ) {
+			void *ptr;
+			unsigned v;
+
+			ptr = kmap ( priv->broadcast_rcv_buffer.pages[u] );
+			for ( v = 0; v < num_packets / ipv4_iso_page_count; v++ )
+				*ptrptr++ = (void *)((char *)ptr + v * max_receive);
+		}
+		priv->broadcast_rcv_context = context;
+	} else
+		context = priv->broadcast_rcv_context;
+
+	packet.payload_length = max_receive;
+	packet.interrupt = 1;
+	packet.skip = 0;
+	packet.tag = 3;
+	packet.sy = 0;
+	packet.header_length = IPV4_GASP_OVERHEAD;
+	offset = 0;
+	for ( u = 0; u < num_packets; u++ ) {
+		retval = fw_iso_context_queue ( context, &packet,
+ &priv->broadcast_rcv_buffer, offset );
+		if ( retval < 0 )
+			goto failed_rcv_queue;
+		offset += max_receive;
+	}
+	priv->num_broadcast_rcv_ptrs = num_packets;
+	priv->rcv_buffer_size = max_receive;
+	priv->broadcast_rcv_next_ptr = 0U;
+	retval = fw_iso_context_start ( context, -1, 0, FW_ISO_CONTEXT_MATCH_ALL_TAGS ); /* ??? sync */
+	if ( retval < 0 )
+		goto failed_rcv_queue;
+	/* FIXME: adjust this when we know the max receive speeds of all other IP nodes on the bus. */
+	/* since we only xmt at S100 ??? */
+	priv->broadcast_xmt_max_payload = S100_BUFFER_SIZE - IPV4_GASP_OVERHEAD - IPV4_UNFRAG_HDR_SIZE;
+	priv->broadcast_state = IPV4_BROADCAST_RUNNING;
+	return 0;
+
+ failed_rcv_queue:
+	kfree ( priv->broadcast_rcv_buffer_ptrs );
+	priv->broadcast_rcv_buffer_ptrs = NULL;
+ failed_ptrs_alloc:
+	fw_iso_buffer_destroy ( &priv->broadcast_rcv_buffer, priv->card );
+ failed_buffer_init:
+	fw_iso_context_destroy ( context );
+	priv->broadcast_rcv_context = NULL;
+ failed_context_create:
+	fw_core_remove_address_handler ( &priv->handler );
+ failed_initial:
+	priv->local_fifo = INVALID_FIFO_ADDR;
+	return retval;
+}
+
+/* This is called after an "ifup" */
+static int ipv4_open(struct net_device *dev) {
+	struct ipv4_priv *priv;
+	int ret;
+
+	priv = netdev_priv(dev);
+	if (priv->broadcast_state == IPV4_BROADCAST_ERROR) {
+		ret = ipv4_broadcast_start ( priv );
+		if (ret)
+			return ret;
+	}
+	netif_start_queue(dev);
+	return 0;
+}
+
+/* This is called after an "ifdown" */
+static int ipv4_stop(struct net_device *netdev)
+{
+	/* flush priv->wake */
+	/* flush_scheduled_work(); */
+
+	netif_stop_queue(netdev);
+	return 0;
+}
+
+/* Transmit a packet (called by kernel) */
+static int ipv4_tx(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct ipv4_ether_hdr hdr_buf;
+	struct ipv4_priv *priv = netdev_priv(netdev);
+	__be16 proto;
+	u16 dest_node;
+	enum ipv4_tx_type tx_type;
+	unsigned max_payload;
+	u16 dg_size;
+	u16 *datagram_label_ptr;
+	struct ipv4_packet_task *ptask;
+	struct ipv4_node *node = NULL;
+
+	ptask = kmem_cache_alloc(ipv4_packet_task_cache, GFP_ATOMIC);
+	if (ptask == NULL)
+		goto fail;
+
+	skb = skb_share_check(skb, GFP_ATOMIC);
+	if (!skb)
+		goto fail;
+
+	/*
+	 * Get rid of the fake ipv4 header, but first make a copy.
+	 * We might need to rebuild the header on tx failure.
+	 */
+	memcpy(&hdr_buf, skb->data, sizeof(hdr_buf));
+	skb_pull(skb, sizeof(hdr_buf));
+
+	proto = hdr_buf.h_proto;
+	dg_size = skb->len;
+
+	/*
+	 * Set the transmission type for the packet.  ARP packets and IP
+	 * broadcast packets are sent via GASP.
+	 */
+	if (   memcmp(hdr_buf.h_dest, netdev->broadcast, IPV4_ALEN) == 0
+	    || proto == htons(ETH_P_ARP)
+	    || (   proto == htons(ETH_P_IP)
+		&& IN_MULTICAST(ntohl(ip_hdr(skb)->daddr)) ) ) {
+		/* fw_debug ( "transmitting arp or multicast packet\n" );*/
+		tx_type = IPV4_GASP;
+		dest_node = ALL_NODES;
+		max_payload = priv->broadcast_xmt_max_payload;
+		/* BUG_ON(max_payload < S100_BUFFER_SIZE - IPV4_GASP_OVERHEAD); */
+		datagram_label_ptr = &priv->broadcast_xmt_datagramlabel;
+		ptask->fifo_addr = INVALID_FIFO_ADDR;
+		ptask->generation = 0U;
+		ptask->dest_node = 0U;
+		ptask->speed = 0;
+	} else {
+		__be64 guid = get_unaligned((u64 *)hdr_buf.h_dest);
+		u8 generation;
+
+		node = ipv4_node_find_by_guid(priv, be64_to_cpu(guid));
+		if (!node) {
+			fw_debug ( "Normal packet but no node\n" );
+			goto fail;
+		}
+
+		if (node->fifo == INVALID_FIFO_ADDR) {
+			fw_debug ( "Normal packet but no fifo addr\n" );
+			goto fail;
+		}
+
+		/* fw_debug ( "Transmitting normal packet to %x at %llxx\n", node->nodeid, node->fifo ); */
+		generation = node->generation;
+		dest_node = node->nodeid;
+		max_payload = node->max_payload;
+		/* BUG_ON(max_payload < S100_BUFFER_SIZE - IPV4_FRAG_HDR_SIZE); */
+
+		datagram_label_ptr = &node->datagram_label;
+		tx_type = IPV4_WRREQ;
+		ptask->fifo_addr = node->fifo;
+		ptask->generation = generation;
+		ptask->dest_node = dest_node;
+		ptask->speed = node->xmt_speed;
+	}
+
+	/* If this is an ARP packet, convert it */
+	if (proto == htons(ETH_P_ARP)) {
+		/* Convert a standard ARP packet to 1394 ARP. The first 8 bytes (the entire
+		 * arphdr) is the same format as the ip1394 header, so they overlap.  The rest
+		 * needs to be munged a bit.  The remainder of the arphdr is formatted based
+		 * on hwaddr len and ipaddr len.  We know what they'll be, so it's easy to
+		 * judge.
+		 *
+		 * Now that the EUI is used for the hardware address all we need to do to make
+		 * this work for 1394 is to insert 2 quadlets that contain max_rec size,
+		 * speed, and unicast FIFO address information between the sender_unique_id
+		 * and the IP addresses.
+		 */
+		struct arphdr *arp = (struct arphdr *)skb->data;
+		unsigned char *arp_ptr = (unsigned char *)(arp + 1);
+		struct ipv4_arp *arp1394 = (struct ipv4_arp *)skb->data;
+		u32 ipaddr;
+
+		ipaddr = *(u32*)(arp_ptr + IPV4_ALEN);
+		arp1394->hw_addr_len    = 16;
+		arp1394->max_rec        = priv->card->max_receive;
+		arp1394->sspd		= priv->card->link_speed;
+		arp1394->fifo_hi	= htons(priv->local_fifo >> 32);
+		arp1394->fifo_lo        = htonl(priv->local_fifo & 0xFFFFFFFF);
+		arp1394->sip		= ipaddr;
+	}
+	if ( ipv4_max_xmt && max_payload > ipv4_max_xmt )
+		max_payload = ipv4_max_xmt;
+
+	ptask->hdr.w0 = 0;
+	ptask->hdr.w1 = 0;
+	ptask->skb = skb;
+	ptask->priv = priv;
+        ptask->tx_type = tx_type;
+	/* Does it all fit in one packet? */
+	if ( dg_size <= max_payload ) {
+		ipv4_make_uf_hdr(&ptask->hdr, be16_to_cpu(proto));
+		ptask->outstanding_pkts = 1;
+		max_payload = dg_size + IPV4_UNFRAG_HDR_SIZE;
+	} else {
+		u16 datagram_label;
+
+		max_payload -= IPV4_FRAG_OVERHEAD;
+		datagram_label = (*datagram_label_ptr)++;
+		ipv4_make_ff_hdr(&ptask->hdr, be16_to_cpu(proto), dg_size, datagram_label );
+		ptask->outstanding_pkts = DIV_ROUND_UP(dg_size, max_payload);
+		max_payload += IPV4_FRAG_HDR_SIZE;
+	}
+	ptask->max_payload = max_payload;
+	ipv4_send_packet ( ptask );
+	return NETDEV_TX_OK;
+
+ fail:
+	if (ptask)
+		kmem_cache_free(ipv4_packet_task_cache, ptask);
+
+	if (skb != NULL)
+		dev_kfree_skb(skb);
+
+	netdev->stats.tx_dropped++;
+	netdev->stats.tx_errors++;
+
+	/*
+	 * FIXME: According to a patch from 2003-02-26, "returning non-zero
+	 * causes serious problems" here, allegedly.  Before that patch,
+	 * -ERRNO was returned which is not appropriate under Linux 2.6.
+	 * Perhaps more needs to be done?  Stop the queue in serious
+	 * conditions and restart it elsewhere?
+	 */
+	return NETDEV_TX_OK;
+}
+
+/*
+ * FIXME: What to do if we timeout? I think a host reset is probably in order,
+ * so that's what we do. Should we increment the stat counters too?
+ */
+static void ipv4_tx_timeout(struct net_device *dev) {
+	struct ipv4_priv *priv;
+
+	priv = netdev_priv(dev);
+	fw_error ( "%s: Timeout, resetting host\n", dev->name );
+#if 0 /* stefanr */
+	fw_core_initiate_bus_reset ( priv->card, 1 );
+#endif
+}
+
+static int ipv4_change_mtu ( struct net_device *dev, int new_mtu ) {
+#if 0
+	int max_mtu;
+	struct ipv4_priv *priv;
+#endif
+
+	if (new_mtu < 68)
+		return -EINVAL;
+
+#if 0
+	priv = netdev_priv(dev);
+	/* This is not actually true because we can fragment packets at the firewire layer */
+	max_mtu = (1 << (priv->card->max_receive + 1))
+		                - sizeof(struct ipv4_hdr) - IPV4_GASP_OVERHEAD;
+	if (new_mtu > max_mtu) {
+		fw_notify ( "%s: Local node constrains MTU to %d\n", dev->name, max_mtu);
+		return -ERANGE;
+	}
+#endif
+	dev->mtu = new_mtu;
+	return 0;
+}
+
+static void ipv4_get_drvinfo(struct net_device *dev,
+struct ethtool_drvinfo *info) {
+	strcpy(info->driver, ipv4_driver_name);
+	strcpy(info->bus_info, "ieee1394"); /* FIXME provide more detail? */
+}
+
+static struct ethtool_ops ipv4_ethtool_ops = {
+	.get_drvinfo = ipv4_get_drvinfo,
+};
+
+static const struct net_device_ops ipv4_netdev_ops = {
+	.ndo_open       = ipv4_open,
+	.ndo_stop	= ipv4_stop,
+	.ndo_start_xmit = ipv4_tx,
+	.ndo_tx_timeout = ipv4_tx_timeout,
+	.ndo_change_mtu = ipv4_change_mtu,
+};
+
+static void ipv4_init_dev ( struct net_device *dev ) {
+	dev->header_ops		= &ipv4_header_ops;
+	dev->netdev_ops         = &ipv4_netdev_ops;
+	SET_ETHTOOL_OPS(dev, &ipv4_ethtool_ops);
+
+	dev->watchdog_timeo	= IPV4_TIMEOUT;
+	dev->flags		= IFF_BROADCAST | IFF_MULTICAST;
+	dev->features		= NETIF_F_HIGHDMA;
+	dev->addr_len		= IPV4_ALEN;
+	dev->hard_header_len	= IPV4_HLEN;
+	dev->type		= ARPHRD_IEEE1394;
+
+	/* FIXME: This value was copied from ether_setup(). Is it too much? */
+	dev->tx_queue_len	= 1000;
+}
+
+static int ipv4_probe ( struct device *dev ) {
+	struct fw_unit * unit;
+	struct fw_device *device;
+	struct fw_card *card;
+	struct net_device *netdev;
+	struct ipv4_priv *priv;
+	unsigned max_mtu;
+	__be64 guid;
+
+	fw_debug("ipv4 Probing\n" );
+	unit = fw_unit ( dev );
+	device = fw_device ( unit->device.parent );
+	card = device->card;
+
+	if ( ! device->is_local ) {
+		int added;
+
+		fw_debug ( "Non-local, adding remote node entry\n" );
+		added = ipv4_node_new ( card, device );
+		return added;
+	}
+	fw_debug("ipv4 Local: adding netdev\n" );
+	netdev = alloc_netdev ( sizeof(*priv), "firewire%d", ipv4_init_dev );
+	if ( netdev == NULL) {
+		fw_error( "Out of memory\n");
+		goto out;
+	}
+
+	SET_NETDEV_DEV(netdev, card->device);
+	priv = netdev_priv(netdev);
+
+	spin_lock_init(&priv->lock);
+	priv->broadcast_state = IPV4_BROADCAST_ERROR;
+	priv->broadcast_rcv_context = NULL;
+	priv->broadcast_xmt_max_payload = 0;
+	priv->broadcast_xmt_datagramlabel = 0;
+
+	priv->local_fifo = INVALID_FIFO_ADDR;
+
+	/* INIT_WORK(&priv->wake, ipv4_handle_queue);*/
+	INIT_LIST_HEAD(&priv->packet_list);
+	INIT_LIST_HEAD(&priv->broadcasted_list);
+	INIT_LIST_HEAD(&priv->sent_list );
+
+	priv->card = card;
+
+	/*
+	 * Use the RFC 2734 default 1500 octets or the maximum payload
+	 * as initial MTU
+	 */
+	max_mtu = (1 << (card->max_receive + 1))
+		  - sizeof(struct ipv4_hdr) - IPV4_GASP_OVERHEAD;
+	netdev->mtu = min(1500U, max_mtu);
+
+	/* Set our hardware address while we're at it */
+	guid = cpu_to_be64(card->guid);
+	memcpy(netdev->dev_addr, &guid, sizeof(u64));
+	memset(netdev->broadcast, 0xff, sizeof(u64));
+	if ( register_netdev ( netdev ) ) {
+		fw_error ( "Cannot register the driver\n");
+		goto out;
+	}
+
+	fw_notify ( "%s: IPv4 over Firewire on device %016llx\n",
+ netdev->name, card->guid );
+	card->netdev = netdev;
+
+	return 0 /* ipv4_new_node ( ud ) */;
+ out:
+	if ( netdev )
+		free_netdev ( netdev );
+	return -ENOENT;
+}
+
+
+static int ipv4_remove ( struct device *dev ) {
+	struct fw_unit * unit;
+	struct fw_device *device;
+	struct fw_card *card;
+	struct net_device *netdev;
+	struct ipv4_priv *priv;
+	struct ipv4_node *node;
+	struct ipv4_partial_datagram *pd, *pd_next;
+	struct ipv4_packet_task *ptask, *pt_next;
+
+	fw_debug("ipv4 Removing\n" );
+	unit = fw_unit ( dev );
+	device = fw_device ( unit->device.parent );
+	card = device->card;
+
+	if ( ! device->is_local ) {
+		fw_debug ( "Node %x is non-local, removing remote node entry\n", device->node_id );
+		ipv4_node_delete ( card, device );
+		return 0;
+	}
+	netdev = card->netdev;
+	if ( netdev ) {
+		fw_debug ( "Node %x is local: deleting netdev\n", device->node_id );
+		priv = netdev_priv ( netdev );
+		unregister_netdev ( netdev );
+		fw_debug ( "unregistered\n" );
+		if ( priv->local_fifo != INVALID_FIFO_ADDR )
+			fw_core_remove_address_handler ( &priv->handler );
+		fw_debug ( "address handler gone\n" );
+		if ( priv->broadcast_rcv_context ) {
+			fw_iso_context_stop ( priv->broadcast_rcv_context );
+			fw_iso_buffer_destroy ( &priv->broadcast_rcv_buffer, priv->card );
+			fw_iso_context_destroy ( priv->broadcast_rcv_context );
+			fw_debug ( "rcv stopped\n" );
+		}
+		list_for_each_entry_safe( ptask, pt_next, &priv->packet_list, packet_list ) {
+			dev_kfree_skb_any ( ptask->skb );
+			kmem_cache_free( ipv4_packet_task_cache, ptask );
+		}
+		list_for_each_entry_safe( ptask, pt_next, &priv->broadcasted_list, packet_list ) {
+			dev_kfree_skb_any ( ptask->skb );
+			kmem_cache_free( ipv4_packet_task_cache, ptask );
+		}
+		list_for_each_entry_safe( ptask, pt_next, &priv->sent_list, packet_list ) {
+			dev_kfree_skb_any ( ptask->skb );
+			kmem_cache_free( ipv4_packet_task_cache, ptask );
+		}
+		fw_debug ( "lists emptied\n" );
+		list_for_each_entry( node, &card->ipv4_nodes, ipv4_nodes ) {
+			if ( node->pdg_size ) {
+				list_for_each_entry_safe( pd, pd_next, &node->pdg_list, pdg_list )
+					ipv4_pd_delete ( pd );
+				node->pdg_size = 0;
+			}
+			node->fifo = INVALID_FIFO_ADDR;
+		}
+		fw_debug ( "nodes cleaned up\n" );
+		free_netdev ( netdev );
+		card->netdev = NULL;
+		fw_debug ( "done\n" );
+	}
+	return 0;
+}
+
+static void ipv4_update ( struct fw_unit *unit ) {
+	struct fw_device *device;
+	struct fw_card *card;
+
+	fw_debug ( "ipv4_update unit %p\n", unit );
+	device = fw_device ( unit->device.parent );
+	card = device->card;
+	if ( ! device->is_local ) {
+		struct ipv4_node *node;
+		u64 guid;
+		struct net_device *netdev;
+		struct ipv4_priv *priv;
+
+		netdev = card->netdev;
+		if ( netdev ) {
+			priv = netdev_priv ( netdev );
+			guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
+			node = ipv4_node_find_by_guid ( priv, guid );
+			if ( ! node ) {
+				fw_error ( "ipv4_update: no node for device %llx\n", guid );
+				return;
+			}
+			fw_debug ( "Non-local, updating remote node entry for guid %llx old generation %x, old nodeid %x\n", guid, node->generation, node->nodeid );
+			node->generation = device->generation;
+			rmb();
+			node->nodeid = device->node_id;
+			fw_debug ( "New generation %x, new nodeid %x\n", node->generation, node->nodeid );
+		} else
+			fw_error ( "nonlocal, but no netdev?  How can that be?\n" );
+	} else {
+		/* FIXME: What do we need to do on bus reset? */
+		fw_debug ( "Local, doing nothing\n" );
+	}
+}
+
+static struct fw_driver ipv4_driver = {
+	.driver = {
+		.owner = THIS_MODULE,
+		.name = ipv4_driver_name,
+		.bus = &fw_bus_type,
+		.probe = ipv4_probe,
+		.remove = ipv4_remove,
+	},
+	.update = ipv4_update,
+	.id_table = ipv4_id_table,
+};
+
+static int __init ipv4_init ( void ) {
+	int added;
+
+	added = fw_core_add_descriptor ( &ipv4_unit_directory );
+	if ( added < 0 )
+		fw_error ( "Failed to add descriptor" );
+	ipv4_packet_task_cache = kmem_cache_create("packet_task",
+ sizeof(struct ipv4_packet_task), 0, 0, NULL);
+	fw_debug("Adding ipv4 module\n" );
+	return driver_register ( &ipv4_driver.driver );
+}
+
+static void __exit ipv4_cleanup ( void ) {
+	fw_core_remove_descriptor ( &ipv4_unit_directory );
+	fw_debug("Removing ipv4 module\n" );
+	driver_unregister ( &ipv4_driver.driver );
+}
+
+module_init(ipv4_init);
+module_exit(ipv4_cleanup);
diff --git a/drivers/ieee1394/Kconfig b/drivers/ieee1394/Kconfig
index 95f45f9b8e5e..584245881f4a 100644
--- a/drivers/ieee1394/Kconfig
+++ b/drivers/ieee1394/Kconfig
@@ -105,7 +105,7 @@ config IEEE1394_ETH1394_ROM_ENTRY
 	default n
 
 config IEEE1394_ETH1394
-	tristate "IP over 1394"
+	tristate "IP networking over 1394"
 	depends on IEEE1394 && EXPERIMENTAL && INET
 	select IEEE1394_ETH1394_ROM_ENTRY
 	help
-- 
cgit v1.2.3-59-g8ed1b


From f91e3bd842ec6f5cea245993926ee8ff26250467 Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Sun, 7 Jun 2009 22:57:53 +0200
Subject: firewire: net: style changes

Change names of types, variables, functions.
Omit debug code.
Use get_unaligned*, put_unaligned*.
Annotate big endian data.
Handle errors in __init.
Change whitespace.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/core-card.c |    2 +-
 drivers/firewire/net.c       | 2041 ++++++++++++++++++++----------------------
 include/linux/firewire.h     |    9 +-
 3 files changed, 969 insertions(+), 1083 deletions(-)

diff --git a/drivers/firewire/core-card.c b/drivers/firewire/core-card.c
index cdab32b20675..8c45e43da7c5 100644
--- a/drivers/firewire/core-card.c
+++ b/drivers/firewire/core-card.c
@@ -430,7 +430,7 @@ void fw_card_initialize(struct fw_card *card,
 
 	INIT_DELAYED_WORK(&card->work, fw_card_bm_work);
 	card->netdev = NULL;
-	INIT_LIST_HEAD(&card->ipv4_nodes);
+	INIT_LIST_HEAD(&card->peer_list);
 }
 EXPORT_SYMBOL(fw_card_initialize);
 
diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c
index 15353886bd80..ba6f924b1b13 100644
--- a/drivers/firewire/net.c
+++ b/drivers/firewire/net.c
@@ -6,6 +6,7 @@
  * based on eth1394 by Ben Collins et al
  */
 
+#include <linux/bug.h>
 #include <linux/device.h>
 #include <linux/ethtool.h>
 #include <linux/firewire.h>
@@ -13,6 +14,7 @@
 #include <linux/highmem.h>
 #include <linux/in.h>
 #include <linux/ip.h>
+#include <linux/jiffies.h>
 #include <linux/mod_devicetable.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
@@ -22,181 +24,109 @@
 #include <asm/unaligned.h>
 #include <net/arp.h>
 
-/* Things to potentially make runtime cofigurable */
-/* must be at least as large as our maximum receive size */
-#define FIFO_SIZE 4096
-/* Network timeout in glibbles */
-#define IPV4_TIMEOUT       100000
+#define FWNET_MAX_FRAGMENTS	25	/* arbitrary limit */
+#define FWNET_ISO_PAGE_COUNT	(PAGE_SIZE < 16 * 1024 ? 4 : 2)
 
-/* Runitme configurable paramaters */
-static int ipv4_mpd = 25;
-static int ipv4_max_xmt = 0;
-/* 16k for receiving arp and broadcast packets.  Enough? */
-static int ipv4_iso_page_count = 4;
+#define IEEE1394_BROADCAST_CHANNEL	31
+#define IEEE1394_ALL_NODES		(0xffc0 | 0x003f)
+#define IEEE1394_MAX_PAYLOAD_S100	512
+#define FWNET_NO_FIFO_ADDR		(~0ULL)
 
-MODULE_AUTHOR("Jay Fenlason (fenlason@redhat.com)");
-MODULE_DESCRIPTION("Firewire IPv4 Driver (IPv4-over-IEEE1394 as per RFC 2734)");
-MODULE_LICENSE("GPL");
-MODULE_DEVICE_TABLE(ieee1394, ipv4_id_table);
-module_param_named(max_partial_datagrams, ipv4_mpd, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(max_partial_datagrams, "Maximum number of received"
- " incomplete fragmented datagrams (default = 25).");
-
-/* Max xmt is useful for forcing fragmentation, which makes testing easier. */
-module_param_named(max_transmit, ipv4_max_xmt, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(max_transmit, "Maximum datagram size to transmit"
- " (larger datagrams will be fragmented) (default = 0 (use hardware defaults).");
-
-/* iso page count controls how many pages will be used for receiving broadcast packets. */
-module_param_named(iso_pages, ipv4_iso_page_count, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(iso_pages, "Number of pages to use for receiving broadcast packets"
- " (default = 4).");
-
-/* uncomment this line to do debugging */
-#define fw_debug(s, args...) printk(KERN_DEBUG KBUILD_MODNAME ": " s, ## args)
-
-/* comment out these lines to do debugging. */
-/* #undef fw_debug */
-/* #define fw_debug(s...) */
-/* #define print_hex_dump(l...) */
-
-/* Define a fake hardware header format for the networking core.  Note that
- * header size cannot exceed 16 bytes as that is the size of the header cache.
- * Also, we do not need the source address in the header so we omit it and
- * keep the header to under 16 bytes */
-#define IPV4_ALEN (8)
-/* This must equal sizeof(struct ipv4_ether_hdr) */
-#define IPV4_HLEN (10)
-
-/* FIXME: what's a good size for this? */
-#define INVALID_FIFO_ADDR (u64)~0ULL
-
-/* Things specified by standards */
-#define BROADCAST_CHANNEL 31
-
-#define S100_BUFFER_SIZE 512
-#define MAX_BUFFER_SIZE 4096
-
-#define IPV4_GASP_SPECIFIER_ID	0x00005EU
-#define IPV4_GASP_VERSION	0x00000001U
-
-#define IPV4_GASP_OVERHEAD (2 * sizeof(u32)) /* for GASP header */
-
-#define IPV4_UNFRAG_HDR_SIZE	sizeof(u32)
-#define IPV4_FRAG_HDR_SIZE	(2 * sizeof(u32))
-#define IPV4_FRAG_OVERHEAD	sizeof(u32)
-
-#define ALL_NODES (0xffc0 | 0x003f)
-
-#define IPV4_HDR_UNFRAG		0	/* unfragmented		*/
-#define IPV4_HDR_FIRSTFRAG	1	/* first fragment	*/
-#define IPV4_HDR_LASTFRAG	2	/* last fragment	*/
-#define IPV4_HDR_INTFRAG	3	/* interior fragment	*/
-
-/* Our arp packet (ARPHRD_IEEE1394) */
-/* FIXME: note that this is probably bogus on weird-endian machines */
-struct ipv4_arp {
-	u16 hw_type;		/* 0x0018	*/
-	u16 proto_type;		/* 0x0806       */
-	u8 hw_addr_len;		/* 16		*/
-	u8 ip_addr_len;         /* 4		*/
-	u16 opcode;	        /* ARP Opcode	*/
-	/* Above is exactly the same format as struct arphdr */
-
-	u64 s_uniq_id;		/* Sender's 64bit EUI			*/
-	u8 max_rec;             /* Sender's max packet size		*/
-	u8 sspd;		/* Sender's max speed			*/
-	u16 fifo_hi;            /* hi 16bits of sender's FIFO addr	*/
-	u32 fifo_lo;            /* lo 32bits of sender's FIFO addr	*/
-	u32 sip;		/* Sender's IP Address			*/
-	u32 tip;		/* IP Address of requested hw addr	*/
-} __attribute__((packed));
+#define IANA_SPECIFIER_ID		0x00005eU
+#define RFC2734_SW_VERSION		0x000001U
 
-struct ipv4_ether_hdr {
-	unsigned char	h_dest[IPV4_ALEN];	/* destination address */
-	unsigned short  h_proto;                /* packet type ID field */
-}  __attribute__((packed));
+#define IEEE1394_GASP_HDR_SIZE	8
 
-static inline struct ipv4_ether_hdr *ipv4_ether_hdr(const struct sk_buff *skb)
-{
-	return (struct ipv4_ether_hdr *)skb_mac_header(skb);
-}
+#define RFC2374_UNFRAG_HDR_SIZE	4
+#define RFC2374_FRAG_HDR_SIZE	8
+#define RFC2374_FRAG_OVERHEAD	4
 
-enum ipv4_tx_type {
-	IPV4_UNKNOWN = 0,
-	IPV4_GASP = 1,
-	IPV4_WRREQ = 2,
-};
+#define RFC2374_HDR_UNFRAG	0	/* unfragmented		*/
+#define RFC2374_HDR_FIRSTFRAG	1	/* first fragment	*/
+#define RFC2374_HDR_LASTFRAG	2	/* last fragment	*/
+#define RFC2374_HDR_INTFRAG	3	/* interior fragment	*/
 
-enum ipv4_broadcast_state {
-	IPV4_BROADCAST_ERROR,
-	IPV4_BROADCAST_RUNNING,
-	IPV4_BROADCAST_STOPPED,
-};
+#define RFC2734_HW_ADDR_LEN	16
 
-#define ipv4_get_hdr_lf(h)		(((h)->w0&0xC0000000)>>30)
-#define ipv4_get_hdr_ether_type(h)	(((h)->w0&0x0000FFFF)    )
-#define ipv4_get_hdr_dg_size(h)		(((h)->w0&0x0FFF0000)>>16)
-#define ipv4_get_hdr_fg_off(h)		(((h)->w0&0x00000FFF)    )
-#define ipv4_get_hdr_dgl(h)		(((h)->w1&0xFFFF0000)>>16)
+struct rfc2734_arp {
+	__be16 hw_type;		/* 0x0018	*/
+	__be16 proto_type;	/* 0x0806       */
+	u8 hw_addr_len;		/* 16		*/
+	u8 ip_addr_len;		/* 4		*/
+	__be16 opcode;		/* ARP Opcode	*/
+	/* Above is exactly the same format as struct arphdr */
 
-#define ipv4_set_hdr_lf(lf)		(( lf)<<30)
-#define ipv4_set_hdr_ether_type(et)	(( et)    )
-#define ipv4_set_hdr_dg_size(dgs)	((dgs)<<16)
-#define ipv4_set_hdr_fg_off(fgo)	((fgo)    )
+	__be64 s_uniq_id;	/* Sender's 64bit EUI			*/
+	u8 max_rec;		/* Sender's max packet size		*/
+	u8 sspd;		/* Sender's max speed			*/
+	__be16 fifo_hi;		/* hi 16bits of sender's FIFO addr	*/
+	__be32 fifo_lo;		/* lo 32bits of sender's FIFO addr	*/
+	__be32 sip;		/* Sender's IP Address			*/
+	__be32 tip;		/* IP Address of requested hw addr	*/
+} __attribute__((packed));
 
-#define ipv4_set_hdr_dgl(dgl)		((dgl)<<16)
+/* This header format is specific to this driver implementation. */
+#define FWNET_ALEN	8
+#define FWNET_HLEN	10
+struct fwnet_header {
+	u8 h_dest[FWNET_ALEN];	/* destination address */
+	__be16 h_proto;		/* packet type ID field */
+} __attribute__((packed));
 
-struct ipv4_hdr {
+/* IPv4 and IPv6 encapsulation header */
+struct rfc2734_header {
 	u32 w0;
 	u32 w1;
 };
 
-static inline void ipv4_make_uf_hdr( struct ipv4_hdr *hdr, unsigned ether_type) {
-	hdr->w0 = ipv4_set_hdr_lf(IPV4_HDR_UNFRAG)
-		   |ipv4_set_hdr_ether_type(ether_type);
-	fw_debug ( "Setting unfragmented header %p to %x\n", hdr, hdr->w0 );
-}
+#define fwnet_get_hdr_lf(h)		(((h)->w0 & 0xc0000000) >> 30)
+#define fwnet_get_hdr_ether_type(h)	(((h)->w0 & 0x0000ffff))
+#define fwnet_get_hdr_dg_size(h)	(((h)->w0 & 0x0fff0000) >> 16)
+#define fwnet_get_hdr_fg_off(h)		(((h)->w0 & 0x00000fff))
+#define fwnet_get_hdr_dgl(h)		(((h)->w1 & 0xffff0000) >> 16)
 
-static inline void ipv4_make_ff_hdr ( struct ipv4_hdr *hdr, unsigned ether_type, unsigned dg_size, unsigned dgl ) {
-	hdr->w0 = ipv4_set_hdr_lf(IPV4_HDR_FIRSTFRAG)
-		   |ipv4_set_hdr_dg_size(dg_size)
-		   |ipv4_set_hdr_ether_type(ether_type);
-	hdr->w1 = ipv4_set_hdr_dgl(dgl);
-	fw_debug ( "Setting fragmented header %p to first_frag %x,%x (et %x, dgs %x, dgl %x)\n", hdr, hdr->w0, hdr->w1,
- ether_type, dg_size, dgl );
-}
+#define fwnet_set_hdr_lf(lf)		((lf)  << 30)
+#define fwnet_set_hdr_ether_type(et)	(et)
+#define fwnet_set_hdr_dg_size(dgs)	((dgs) << 16)
+#define fwnet_set_hdr_fg_off(fgo)	(fgo)
 
-static inline void ipv4_make_sf_hdr ( struct ipv4_hdr *hdr, unsigned lf, unsigned dg_size, unsigned fg_off, unsigned dgl) {
-	hdr->w0 = ipv4_set_hdr_lf(lf)
-		 |ipv4_set_hdr_dg_size(dg_size)
-		 |ipv4_set_hdr_fg_off(fg_off);
-	hdr->w1 = ipv4_set_hdr_dgl(dgl);
-	fw_debug ( "Setting fragmented header %p to %x,%x (lf %x, dgs %x, fo %x dgl %x)\n",
- hdr, hdr->w0, hdr->w1,
- lf, dg_size, fg_off, dgl );
-}
+#define fwnet_set_hdr_dgl(dgl)		((dgl) << 16)
 
-/* End of IP1394 headers */
+static inline void fwnet_make_uf_hdr(struct rfc2734_header *hdr,
+		unsigned ether_type)
+{
+	hdr->w0 = fwnet_set_hdr_lf(RFC2374_HDR_UNFRAG)
+		  | fwnet_set_hdr_ether_type(ether_type);
+}
 
-/* Fragment types */
-#define ETH1394_HDR_LF_UF	0	/* unfragmented		*/
-#define ETH1394_HDR_LF_FF	1	/* first fragment	*/
-#define ETH1394_HDR_LF_LF	2	/* last fragment	*/
-#define ETH1394_HDR_LF_IF	3	/* interior fragment	*/
+static inline void fwnet_make_ff_hdr(struct rfc2734_header *hdr,
+		unsigned ether_type, unsigned dg_size, unsigned dgl)
+{
+	hdr->w0 = fwnet_set_hdr_lf(RFC2374_HDR_FIRSTFRAG)
+		  | fwnet_set_hdr_dg_size(dg_size)
+		  | fwnet_set_hdr_ether_type(ether_type);
+	hdr->w1 = fwnet_set_hdr_dgl(dgl);
+}
 
-#define IP1394_HW_ADDR_LEN	16	/* As per RFC		*/
+static inline void fwnet_make_sf_hdr(struct rfc2734_header *hdr,
+		unsigned lf, unsigned dg_size, unsigned fg_off, unsigned dgl)
+{
+	hdr->w0 = fwnet_set_hdr_lf(lf)
+		  | fwnet_set_hdr_dg_size(dg_size)
+		  | fwnet_set_hdr_fg_off(fg_off);
+	hdr->w1 = fwnet_set_hdr_dgl(dgl);
+}
 
 /* This list keeps track of what parts of the datagram have been filled in */
-struct ipv4_fragment_info {
-        struct list_head fragment_info;
+struct fwnet_fragment_info {
+	struct list_head fi_link;
 	u16 offset;
 	u16 len;
 };
 
-struct ipv4_partial_datagram {
-	struct list_head pdg_list;
-	struct list_head fragment_info;
+struct fwnet_partial_datagram {
+	struct list_head pd_link;
+	struct list_head fi_list;
 	struct sk_buff *skb;
 	/* FIXME Why not use skb->data? */
 	char *pbuf;
@@ -208,40 +138,43 @@ struct ipv4_partial_datagram {
 /*
  * We keep one of these for each IPv4 capable device attached to a fw_card.
  * The list of them is stored in the fw_card structure rather than in the
- * ipv4_priv because the remote IPv4 nodes may be probed before the card is,
- * so we need a place to store them before the ipv4_priv structure is
+ * fwnet_device because the remote IPv4 nodes may be probed before the card is,
+ * so we need a place to store them before the fwnet_device structure is
  * allocated.
  */
-struct ipv4_node {
-	struct list_head ipv4_nodes;
-	/* guid of the remote node */
+struct fwnet_peer {
+	struct list_head peer_link;
+	/* guid of the remote peer */
 	u64 guid;
-	/* FIFO address to transmit datagrams to, or INVALID_FIFO_ADDR */
+	/* FIFO address to transmit datagrams to, or FWNET_NO_FIFO_ADDR */
 	u64 fifo;
 
 	spinlock_t pdg_lock;	/* partial datagram lock		*/
-	/* List of partial datagrams received from this node */
-	struct list_head pdg_list;
-	/* Number of entries in pdg_list at the moment */
+	/* List of partial datagrams received from this peer */
+	struct list_head pd_list;
+	/* Number of entries in pd_list at the moment */
 	unsigned pdg_size;
 
-	/* max payload to transmit to this remote node */
-	/* This already includes the IPV4_FRAG_HDR_SIZE overhead */
+	/* max payload to transmit to this remote peer */
+	/* This already includes the RFC2374_FRAG_HDR_SIZE overhead */
 	u16 max_payload;
 	/* outgoing datagram label */
 	u16 datagram_label;
-	/* Current node_id of the remote node */
-	u16 nodeid;
-	/* current generation of the remote node */
+	/* Current node_id of the remote peer */
+	u16 node_id;
+	/* current generation of the remote peer */
 	u8 generation;
-	/* max speed that this node can receive at */
+	/* max speed that this peer can receive at */
 	u8 xmt_speed;
 };
 
-struct ipv4_priv {
+struct fwnet_device {
 	spinlock_t lock;
-
-	enum ipv4_broadcast_state broadcast_state;
+	enum {
+		FWNET_BROADCAST_ERROR,
+		FWNET_BROADCAST_RUNNING,
+		FWNET_BROADCAST_STOPPED,
+	} broadcast_state;
 	struct fw_iso_context *broadcast_rcv_context;
 	struct fw_iso_buffer broadcast_rcv_buffer;
 	void **broadcast_rcv_buffer_ptrs;
@@ -257,14 +190,12 @@ struct ipv4_priv {
 	u16 broadcast_xmt_datagramlabel;
 
 	/*
-	 * The csr address that remote nodes must send datagrams to for us to
+	 * The CSR address that remote nodes must send datagrams to for us to
 	 * receive them.
 	 */
 	struct fw_address_handler handler;
 	u64 local_fifo;
 
-	/* Wake up to xmt	 */
-        /* struct work_struct wake;*/
 	/* List of packets to be sent */
 	struct list_head packet_list;
 	/*
@@ -279,17 +210,17 @@ struct ipv4_priv {
 };
 
 /* This is our task struct. It's used for the packet complete callback.  */
-struct ipv4_packet_task {
+struct fwnet_packet_task {
 	/*
-	 * ptask can actually be on priv->packet_list, priv->broadcasted_list,
-	 * or priv->sent_list depending on its current state.
+	 * ptask can actually be on dev->packet_list, dev->broadcasted_list,
+	 * or dev->sent_list depending on its current state.
 	 */
-	struct list_head packet_list;
+	struct list_head pt_link;
 	struct fw_transaction transaction;
-	struct ipv4_hdr hdr;
+	struct rfc2734_header hdr;
 	struct sk_buff *skb;
-	struct ipv4_priv *priv;
-	enum ipv4_tx_type tx_type;
+	struct fwnet_device *dev;
+
 	int outstanding_pkts;
 	unsigned max_payload;
 	u64 fifo_addr;
@@ -298,243 +229,192 @@ struct ipv4_packet_task {
 	u8 speed;
 };
 
-static struct kmem_cache *ipv4_packet_task_cache;
-
-static const char ipv4_driver_name[] = "firewire-ipv4";
-
-static const struct ieee1394_device_id ipv4_id_table[] = {
-	{
-		.match_flags  = IEEE1394_MATCH_SPECIFIER_ID |
-				IEEE1394_MATCH_VERSION,
-		.specifier_id = IPV4_GASP_SPECIFIER_ID,
-		.version      = IPV4_GASP_VERSION,
-	},
-	{ }
-};
-
-static u32 ipv4_unit_directory_data[] = {
-	0x00040000,					/* unit directory */
-	0x12000000 | IPV4_GASP_SPECIFIER_ID,	/* specifier ID */
-	0x81000003,					/* text descriptor */
-	0x13000000 | IPV4_GASP_VERSION,		/* version */
-	0x81000005,					/* text descriptor */
-
-	0x00030000,					/* Three quadlets */
-	0x00000000,					/* Text */
-	0x00000000,					/* Language 0 */
-	0x49414e41,					/* I A N A */
-	0x00030000,					/* Three quadlets */
-	0x00000000,					/* Text */
-	0x00000000,					/* Language 0 */
-	0x49507634,					/* I P v 4 */
-};
-
-static struct fw_descriptor ipv4_unit_directory = {
-	.length = ARRAY_SIZE(ipv4_unit_directory_data),
-	.key = 0xd1000000,
-	.data = ipv4_unit_directory_data
-};
-
-static int ipv4_send_packet(struct ipv4_packet_task *ptask );
-
-/* ------------------------------------------------------------------ */
-/******************************************
- * HW Header net device functions
- ******************************************/
-  /* These functions have been adapted from net/ethernet/eth.c */
-
-/* Create a fake MAC header for an arbitrary protocol layer.
- * saddr=NULL means use device source address
- * daddr=NULL means leave destination address (eg unresolved arp). */
+/*
+ * saddr == NULL means use device source address.
+ * daddr == NULL means leave destination address (eg unresolved arp).
+ */
+static int fwnet_header_create(struct sk_buff *skb, struct net_device *net,
+			unsigned short type, const void *daddr,
+			const void *saddr, unsigned len)
+{
+	struct fwnet_header *h;
 
-static int ipv4_header ( struct sk_buff *skb, struct net_device *dev,
-		       unsigned short type, const void *daddr,
-		       const void *saddr, unsigned len) {
-	struct ipv4_ether_hdr *eth;
+	h = (struct fwnet_header *)skb_push(skb, sizeof(*h));
+	put_unaligned_be16(type, &h->h_proto);
 
-	eth = (struct ipv4_ether_hdr *)skb_push(skb, sizeof(*eth));
-	eth->h_proto = htons(type);
+	if (net->flags & (IFF_LOOPBACK | IFF_NOARP)) {
+		memset(h->h_dest, 0, net->addr_len);
 
-	if (dev->flags & (IFF_LOOPBACK | IFF_NOARP)) {
-		memset(eth->h_dest, 0, dev->addr_len);
-		return dev->hard_header_len;
+		return net->hard_header_len;
 	}
 
 	if (daddr) {
-		memcpy(eth->h_dest, daddr, dev->addr_len);
-		return dev->hard_header_len;
+		memcpy(h->h_dest, daddr, net->addr_len);
+
+		return net->hard_header_len;
 	}
 
-	return -dev->hard_header_len;
+	return -net->hard_header_len;
 }
 
-/* Rebuild the faked MAC header. This is called after an ARP
- * (or in future other address resolution) has completed on this
- * sk_buff. We now let ARP fill in the other fields.
- *
- * This routine CANNOT use cached dst->neigh!
- * Really, it is used only when dst->neigh is wrong.
- */
-
-static int ipv4_rebuild_header(struct sk_buff *skb)
+static int fwnet_header_rebuild(struct sk_buff *skb)
 {
-	struct ipv4_ether_hdr *eth;
+	struct fwnet_header *h = (struct fwnet_header *)skb->data;
 
-	eth = (struct ipv4_ether_hdr *)skb->data;
-	if (eth->h_proto == htons(ETH_P_IP))
-		return arp_find((unsigned char *)&eth->h_dest, skb);
+	if (get_unaligned_be16(&h->h_proto) == ETH_P_IP)
+		return arp_find((unsigned char *)&h->h_dest, skb);
 
-	fw_notify ( "%s: unable to resolve type %04x addresses\n",
-		   skb->dev->name,ntohs(eth->h_proto) );
+	fw_notify("%s: unable to resolve type %04x addresses\n",
+		  skb->dev->name, be16_to_cpu(h->h_proto));
 	return 0;
 }
 
-static int ipv4_header_cache(const struct neighbour *neigh, struct hh_cache *hh) {
-	unsigned short type = hh->hh_type;
-	struct net_device *dev;
-	struct ipv4_ether_hdr *eth;
+static int fwnet_header_cache(const struct neighbour *neigh,
+			      struct hh_cache *hh)
+{
+	struct net_device *net;
+	struct fwnet_header *h;
 
-	if (type == htons(ETH_P_802_3))
+	if (hh->hh_type == cpu_to_be16(ETH_P_802_3))
 		return -1;
-	dev = neigh->dev;
-	eth = (struct ipv4_ether_hdr *)((u8 *)hh->hh_data + 16 - sizeof(*eth));
-	eth->h_proto = type;
-	memcpy(eth->h_dest, neigh->ha, dev->addr_len);
+	net = neigh->dev;
+	h = (struct fwnet_header *)((u8 *)hh->hh_data + 16 - sizeof(*h));
+	h->h_proto = hh->hh_type;
+	memcpy(h->h_dest, neigh->ha, net->addr_len);
+	hh->hh_len = FWNET_HLEN;
 
-	hh->hh_len = IPV4_HLEN;
 	return 0;
 }
 
 /* Called by Address Resolution module to notify changes in address. */
-static void ipv4_header_cache_update(struct hh_cache *hh, const struct net_device *dev, const unsigned char * haddr ) {
-	memcpy((u8 *)hh->hh_data + 16 - IPV4_HLEN, haddr, dev->addr_len);
+static void fwnet_header_cache_update(struct hh_cache *hh,
+		const struct net_device *net, const unsigned char *haddr)
+{
+	memcpy((u8 *)hh->hh_data + 16 - FWNET_HLEN, haddr, net->addr_len);
 }
 
-static int ipv4_header_parse(const struct sk_buff *skb, unsigned char *haddr) {
-	memcpy(haddr, skb->dev->dev_addr, IPV4_ALEN);
-	return IPV4_ALEN;
+static int fwnet_header_parse(const struct sk_buff *skb, unsigned char *haddr)
+{
+	memcpy(haddr, skb->dev->dev_addr, FWNET_ALEN);
+
+	return FWNET_ALEN;
 }
 
-static const struct header_ops ipv4_header_ops = {
-	.create         = ipv4_header,
-	.rebuild        = ipv4_rebuild_header,
-	.cache		= ipv4_header_cache,
-	.cache_update	= ipv4_header_cache_update,
-	.parse          = ipv4_header_parse,
+static const struct header_ops fwnet_header_ops = {
+	.create         = fwnet_header_create,
+	.rebuild        = fwnet_header_rebuild,
+	.cache		= fwnet_header_cache,
+	.cache_update	= fwnet_header_cache_update,
+	.parse          = fwnet_header_parse,
 };
 
-/* ------------------------------------------------------------------ */
-
 /* FIXME: is this correct for all cases? */
-static bool ipv4_frag_overlap(struct ipv4_partial_datagram *pd, unsigned offset, unsigned len)
+static bool fwnet_frag_overlap(struct fwnet_partial_datagram *pd,
+			       unsigned offset, unsigned len)
 {
-        struct ipv4_fragment_info *fi;
+	struct fwnet_fragment_info *fi;
 	unsigned end = offset + len;
 
-	list_for_each_entry(fi, &pd->fragment_info, fragment_info) {
-		if (offset < fi->offset + fi->len && end > fi->offset) {
-			fw_debug ( "frag_overlap pd %p fi %p (%x@%x) with %x@%x\n", pd, fi, fi->len, fi->offset, len, offset );
+	list_for_each_entry(fi, &pd->fi_list, fi_link)
+		if (offset < fi->offset + fi->len && end > fi->offset)
 			return true;
-		}
-	}
-	fw_debug ( "frag_overlap %p does not overlap with %x@%x\n", pd, len, offset );
+
 	return false;
 }
 
 /* Assumes that new fragment does not overlap any existing fragments */
-static struct ipv4_fragment_info *ipv4_frag_new ( struct ipv4_partial_datagram *pd, unsigned offset, unsigned len ) {
-	struct ipv4_fragment_info *fi, *fi2, *new;
+static struct fwnet_fragment_info *fwnet_frag_new(
+	struct fwnet_partial_datagram *pd, unsigned offset, unsigned len)
+{
+	struct fwnet_fragment_info *fi, *fi2, *new;
 	struct list_head *list;
 
-	fw_debug ( "frag_new pd %p %x@%x\n", pd, len, offset );
-	list = &pd->fragment_info;
-	list_for_each_entry(fi, &pd->fragment_info, fragment_info) {
+	list = &pd->fi_list;
+	list_for_each_entry(fi, &pd->fi_list, fi_link) {
 		if (fi->offset + fi->len == offset) {
 			/* The new fragment can be tacked on to the end */
 			/* Did the new fragment plug a hole? */
-			fi2 = list_entry(fi->fragment_info.next, struct ipv4_fragment_info, fragment_info);
+			fi2 = list_entry(fi->fi_link.next,
+					 struct fwnet_fragment_info, fi_link);
 			if (fi->offset + fi->len == fi2->offset) {
-				fw_debug ( "pd %p: hole filling %p (%x@%x) and %p(%x@%x): now %x@%x\n", pd, fi, fi->len, fi->offset,
-				fi2, fi2->len, fi2->offset, fi->len + len + fi2->len, fi->offset );
 				/* glue fragments together */
 				fi->len += len + fi2->len;
-				list_del(&fi2->fragment_info);
+				list_del(&fi2->fi_link);
 				kfree(fi2);
 			} else {
-				fw_debug ( "pd %p: extending %p from %x@%x to %x@%x\n", pd, fi, fi->len, fi->offset, fi->len+len, fi->offset );
 				fi->len += len;
 			}
+
 			return fi;
 		}
 		if (offset + len == fi->offset) {
 			/* The new fragment can be tacked on to the beginning */
 			/* Did the new fragment plug a hole? */
-			fi2 = list_entry(fi->fragment_info.prev, struct ipv4_fragment_info, fragment_info);
+			fi2 = list_entry(fi->fi_link.prev,
+					 struct fwnet_fragment_info, fi_link);
 			if (fi2->offset + fi2->len == fi->offset) {
 				/* glue fragments together */
-				fw_debug ( "pd %p: extending %p and merging with %p from %x@%x to %x@%x\n",
- pd, fi2, fi, fi2->len, fi2->offset, fi2->len + fi->len + len, fi2->offset );
 				fi2->len += fi->len + len;
-				list_del(&fi->fragment_info);
+				list_del(&fi->fi_link);
 				kfree(fi);
+
 				return fi2;
 			}
-			fw_debug ( "pd %p: extending %p from %x@%x to %x@%x\n", pd, fi, fi->len, fi->offset, offset, fi->len + len );
 			fi->offset = offset;
 			fi->len += len;
+
 			return fi;
 		}
 		if (offset > fi->offset + fi->len) {
-			list = &fi->fragment_info;
+			list = &fi->fi_link;
 			break;
 		}
 		if (offset + len < fi->offset) {
-			list = fi->fragment_info.prev;
+			list = fi->fi_link.prev;
 			break;
 		}
 	}
 
 	new = kmalloc(sizeof(*new), GFP_ATOMIC);
 	if (!new) {
-		fw_error ( "out of memory in fragment handling!\n" );
+		fw_error("out of memory\n");
 		return NULL;
 	}
 
 	new->offset = offset;
 	new->len = len;
-	list_add(&new->fragment_info, list);
-	fw_debug ( "pd %p: new frag %p %x@%x\n", pd, new, new->len, new->offset );
-	list_for_each_entry( fi, &pd->fragment_info, fragment_info )
-		fw_debug ( "fi %p %x@%x\n", fi, fi->len, fi->offset );
+	list_add(&new->fi_link, list);
+
 	return new;
 }
 
-/* ------------------------------------------------------------------ */
-
-static struct ipv4_partial_datagram *ipv4_pd_new(struct net_device *netdev,
- struct ipv4_node *node, u16 datagram_label, unsigned dg_size, u32 *frag_buf,
- unsigned frag_off, unsigned frag_len) {
-	struct ipv4_partial_datagram *new;
-	struct ipv4_fragment_info *fi;
+static struct fwnet_partial_datagram *fwnet_pd_new(struct net_device *net,
+		struct fwnet_peer *peer, u16 datagram_label, unsigned dg_size,
+		void *frag_buf, unsigned frag_off, unsigned frag_len)
+{
+	struct fwnet_partial_datagram *new;
+	struct fwnet_fragment_info *fi;
 
 	new = kmalloc(sizeof(*new), GFP_ATOMIC);
 	if (!new)
 		goto fail;
-	INIT_LIST_HEAD(&new->fragment_info);
-	fi = ipv4_frag_new ( new, frag_off, frag_len);
-	if ( fi == NULL )
+
+	INIT_LIST_HEAD(&new->fi_list);
+	fi = fwnet_frag_new(new, frag_off, frag_len);
+	if (fi == NULL)
 		goto fail_w_new;
+
 	new->datagram_label = datagram_label;
 	new->datagram_size = dg_size;
-	new->skb = dev_alloc_skb(dg_size + netdev->hard_header_len + 15);
-	if ( new->skb == NULL )
+	new->skb = dev_alloc_skb(dg_size + net->hard_header_len + 15);
+	if (new->skb == NULL)
 		goto fail_w_fi;
-	skb_reserve(new->skb, (netdev->hard_header_len + 15) & ~15);
+
+	skb_reserve(new->skb, (net->hard_header_len + 15) & ~15);
 	new->pbuf = skb_put(new->skb, dg_size);
 	memcpy(new->pbuf + frag_off, frag_buf, frag_len);
-	list_add_tail(&new->pdg_list, &node->pdg_list);
-	fw_debug ( "pd_new: new pd %p { dgl %u, dg_size %u, skb %p, pbuf %p } on node %p\n",
- new, new->datagram_label, new->datagram_size, new->skb, new->pbuf, node );
+	list_add_tail(&new->pd_link, &peer->pd_list);
+
 	return new;
 
 fail_w_fi:
@@ -542,174 +422,171 @@ fail_w_fi:
 fail_w_new:
 	kfree(new);
 fail:
-	fw_error("ipv4_pd_new: no memory\n");
+	fw_error("out of memory\n");
+
 	return NULL;
 }
 
-static struct ipv4_partial_datagram *ipv4_pd_find(struct ipv4_node *node, u16 datagram_label) {
-	struct ipv4_partial_datagram *pd;
+static struct fwnet_partial_datagram *fwnet_pd_find(struct fwnet_peer *peer,
+						    u16 datagram_label)
+{
+	struct fwnet_partial_datagram *pd;
 
-	list_for_each_entry(pd, &node->pdg_list, pdg_list) {
-	        if ( pd->datagram_label == datagram_label ) {
-			fw_debug ( "pd_find(node %p, label %u): pd %p\n", node, datagram_label, pd );
+	list_for_each_entry(pd, &peer->pd_list, pd_link)
+		if (pd->datagram_label == datagram_label)
 			return pd;
-		}
-	}
-	fw_debug ( "pd_find(node %p, label %u) no entry\n", node, datagram_label );
+
 	return NULL;
 }
 
 
-static void ipv4_pd_delete ( struct ipv4_partial_datagram *old ) {
-	struct ipv4_fragment_info *fi, *n;
+static void fwnet_pd_delete(struct fwnet_partial_datagram *old)
+{
+	struct fwnet_fragment_info *fi, *n;
 
-	fw_debug ( "pd_delete %p\n", old );
-	list_for_each_entry_safe(fi, n, &old->fragment_info, fragment_info) {
-		fw_debug ( "Freeing fi %p\n", fi );
+	list_for_each_entry_safe(fi, n, &old->fi_list, fi_link)
 		kfree(fi);
-	}
-	list_del(&old->pdg_list);
+
+	list_del(&old->pd_link);
 	dev_kfree_skb_any(old->skb);
 	kfree(old);
 }
 
-static bool ipv4_pd_update ( struct ipv4_node *node, struct ipv4_partial_datagram *pd,
- u32 *frag_buf, unsigned frag_off, unsigned frag_len) {
-	fw_debug ( "pd_update node %p, pd %p, frag_buf %p, %x@%x\n", node, pd, frag_buf, frag_len, frag_off );
-	if ( ipv4_frag_new ( pd, frag_off, frag_len ) == NULL)
+static bool fwnet_pd_update(struct fwnet_peer *peer,
+		struct fwnet_partial_datagram *pd, void *frag_buf,
+		unsigned frag_off, unsigned frag_len)
+{
+	if (fwnet_frag_new(pd, frag_off, frag_len) == NULL)
 		return false;
+
 	memcpy(pd->pbuf + frag_off, frag_buf, frag_len);
 
 	/*
 	 * Move list entry to beginnig of list so that oldest partial
 	 * datagrams percolate to the end of the list
 	 */
-	list_move_tail(&pd->pdg_list, &node->pdg_list);
-	fw_debug ( "New pd list:\n" );
-	list_for_each_entry ( pd, &node->pdg_list, pdg_list ) {
-		fw_debug ( "pd %p\n", pd );
-	}
+	list_move_tail(&pd->pd_link, &peer->pd_list);
+
 	return true;
 }
 
-static bool ipv4_pd_is_complete ( struct ipv4_partial_datagram *pd ) {
-	struct ipv4_fragment_info *fi;
-	bool ret;
+static bool fwnet_pd_is_complete(struct fwnet_partial_datagram *pd)
+{
+	struct fwnet_fragment_info *fi;
 
-	fi = list_entry(pd->fragment_info.next, struct ipv4_fragment_info, fragment_info);
+	fi = list_entry(pd->fi_list.next, struct fwnet_fragment_info, fi_link);
 
-	ret = (fi->len == pd->datagram_size);
-	fw_debug ( "pd_is_complete (pd %p, dgs %x): fi %p (%x@%x) %s\n", pd, pd->datagram_size, fi, fi->len, fi->offset, ret ? "yes" : "no" );
-	return ret;
+	return fi->len == pd->datagram_size;
 }
 
-/* ------------------------------------------------------------------ */
+static int fwnet_peer_new(struct fw_card *card, struct fw_device *device)
+{
+	struct fwnet_peer *peer;
 
-static int ipv4_node_new ( struct fw_card *card, struct fw_device *device ) {
-	struct ipv4_node *node;
+	peer = kmalloc(sizeof(*peer), GFP_KERNEL);
+	if (!peer) {
+		fw_error("out of memory\n");
 
-	node = kmalloc ( sizeof(*node), GFP_KERNEL );
-	if ( ! node ) {
-		fw_error ( "allocate new node failed\n" );
 		return -ENOMEM;
 	}
-	node->guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
-	node->fifo = INVALID_FIFO_ADDR;
-	INIT_LIST_HEAD(&node->pdg_list);
-	spin_lock_init(&node->pdg_lock);
-	node->pdg_size = 0;
-	node->generation = device->generation;
+	peer->guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
+	peer->fifo = FWNET_NO_FIFO_ADDR;
+	INIT_LIST_HEAD(&peer->pd_list);
+	spin_lock_init(&peer->pdg_lock);
+	peer->pdg_size = 0;
+	peer->generation = device->generation;
 	rmb();
-	node->nodeid = device->node_id;
+	peer->node_id = device->node_id;
 	 /* FIXME what should it really be? */
-	node->max_payload = S100_BUFFER_SIZE - IPV4_UNFRAG_HDR_SIZE;
-	node->datagram_label = 0U;
-	node->xmt_speed = device->max_speed;
-	list_add_tail ( &node->ipv4_nodes, &card->ipv4_nodes );
-	fw_debug ( "node_new: %p { guid %016llx, generation %u, nodeid %x, max_payload %x, xmt_speed %x } added\n",
- node, (unsigned long long)node->guid, node->generation, node->nodeid, node->max_payload, node->xmt_speed );
+	peer->max_payload = IEEE1394_MAX_PAYLOAD_S100 - RFC2374_UNFRAG_HDR_SIZE;
+	peer->datagram_label = 0U;
+	peer->xmt_speed = device->max_speed;
+	list_add_tail(&peer->peer_link, &card->peer_list);
+
 	return 0;
 }
 
-static struct ipv4_node *ipv4_node_find_by_guid(struct ipv4_priv *priv, u64 guid) {
-	struct ipv4_node *node;
+/* FIXME caller must take the lock, or peer needs to be reference-counted */
+static struct fwnet_peer *fwnet_peer_find_by_guid(struct fwnet_device *dev,
+						  u64 guid)
+{
+	struct fwnet_peer *p, *peer = NULL;
 	unsigned long flags;
 
-	spin_lock_irqsave(&priv->lock, flags);
-	list_for_each_entry(node, &priv->card->ipv4_nodes, ipv4_nodes)
-		if (node->guid == guid) {
-			/* FIXME: lock the node first? */
-			spin_unlock_irqrestore ( &priv->lock, flags );
-			fw_debug ( "node_find_by_guid (%016llx) found %p\n", (unsigned long long)guid, node );
-			return node;
+	spin_lock_irqsave(&dev->lock, flags);
+	list_for_each_entry(p, &dev->card->peer_list, peer_link)
+		if (p->guid == guid) {
+			peer = p;
+			break;
 		}
+	spin_unlock_irqrestore(&dev->lock, flags);
 
-	spin_unlock_irqrestore ( &priv->lock, flags );
-	fw_debug ( "node_find_by_guid (%016llx) not found\n", (unsigned long long)guid );
-	return NULL;
+	return peer;
 }
 
-static struct ipv4_node *ipv4_node_find_by_nodeid(struct ipv4_priv *priv, u16 nodeid) {
-	struct ipv4_node *node;
+/* FIXME caller must take the lock, or peer needs to be reference-counted */
+/* FIXME node_id doesn't mean anything without generation */
+static struct fwnet_peer *fwnet_peer_find_by_node_id(struct fwnet_device *dev,
+						     u16 node_id)
+{
+	struct fwnet_peer *p, *peer = NULL;
 	unsigned long flags;
 
-	spin_lock_irqsave(&priv->lock, flags);
-	list_for_each_entry(node, &priv->card->ipv4_nodes, ipv4_nodes)
-		if (node->nodeid == nodeid) {
-			/* FIXME: lock the node first? */
-			spin_unlock_irqrestore ( &priv->lock, flags );
-			fw_debug ( "node_find_by_nodeid (%x) found %p\n", nodeid, node );
-			return node;
+	spin_lock_irqsave(&dev->lock, flags);
+	list_for_each_entry(p, &dev->card->peer_list, peer_link)
+		if (p->node_id == node_id) {
+			peer = p;
+			break;
 		}
-	fw_debug ( "node_find_by_nodeid (%x) not found\n", nodeid );
-	spin_unlock_irqrestore ( &priv->lock, flags );
-	return NULL;
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	return peer;
 }
 
-/* This is only complicated because we can't assume priv exists */
-static void ipv4_node_delete ( struct fw_card *card, struct fw_device *device ) {
-	struct net_device *netdev;
-	struct ipv4_priv *priv;
-	struct ipv4_node *node;
+/* FIXME */
+static void fwnet_peer_delete(struct fw_card *card, struct fw_device *device)
+{
+	struct net_device *net;
+	struct fwnet_device *dev;
+	struct fwnet_peer *peer;
 	u64 guid;
 	unsigned long flags;
-	struct ipv4_partial_datagram *pd, *pd_next;
+	struct fwnet_partial_datagram *pd, *pd_next;
 
 	guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
-	netdev = card->netdev;
-	if ( netdev )
-		priv = netdev_priv ( netdev );
+	net = card->netdev;
+	if (net)
+		dev = netdev_priv(net);
 	else
-		priv = NULL;
-	if ( priv )
-		spin_lock_irqsave ( &priv->lock, flags );
-	list_for_each_entry( node, &card->ipv4_nodes, ipv4_nodes ) {
-		if ( node->guid == guid ) {
-			list_del ( &node->ipv4_nodes );
-			list_for_each_entry_safe( pd, pd_next, &node->pdg_list, pdg_list )
-				ipv4_pd_delete ( pd );
+		dev = NULL;
+	if (dev)
+		spin_lock_irqsave(&dev->lock, flags);
+
+	list_for_each_entry(peer, &card->peer_list, peer_link) {
+		if (peer->guid == guid) {
+			list_del(&peer->peer_link);
+			list_for_each_entry_safe(pd, pd_next, &peer->pd_list,
+						 pd_link)
+				fwnet_pd_delete(pd);
 			break;
 		}
 	}
-	if ( priv )
-		spin_unlock_irqrestore ( &priv->lock, flags );
+	if (dev)
+		spin_unlock_irqrestore(&dev->lock, flags);
 }
 
-/* ------------------------------------------------------------------ */
-
-
-static int ipv4_finish_incoming_packet ( struct net_device *netdev,
- struct sk_buff *skb, u16 source_node_id, bool is_broadcast, u16 ether_type ) {
-	struct ipv4_priv *priv;
-	static u64 broadcast_hw = ~0ULL;
+static int fwnet_finish_incoming_packet(struct net_device *net,
+					struct sk_buff *skb, u16 source_node_id,
+					bool is_broadcast, u16 ether_type)
+{
+	struct fwnet_device *dev;
+	static const __be64 broadcast_hw = cpu_to_be64(~0ULL);
 	int status;
-	u64 guid;
+	__be64 guid;
 
-	fw_debug ( "ipv4_finish_incoming_packet(%p, %p, %x, %s, %x\n",
- netdev, skb, source_node_id, is_broadcast ? "true" : "false", ether_type );
-	priv = netdev_priv(netdev);
+	dev = netdev_priv(net);
 	/* Write metadata, and then pass to the receive level */
-	skb->dev = netdev;
+	skb->dev = net;
 	skb->ip_summed = CHECKSUM_UNNECESSARY;  /* don't check it */
 
 	/*
@@ -724,73 +601,75 @@ static int ipv4_finish_incoming_packet ( struct net_device *netdev,
 	 * about the sending machine.
 	 */
 	if (ether_type == ETH_P_ARP) {
-		struct ipv4_arp *arp1394;
+		struct rfc2734_arp *arp1394;
 		struct arphdr *arp;
 		unsigned char *arp_ptr;
 		u64 fifo_addr;
+		u64 peer_guid;
 		u8 max_rec;
 		u8 sspd;
 		u16 max_payload;
-		struct ipv4_node *node;
-		static const u16 ipv4_speed_to_max_payload[] = {
+		struct fwnet_peer *peer;
+		static const u16 fwnet_speed_to_max_payload[] = {
 			/* S100, S200, S400, S800, S1600, S3200 */
 			    512, 1024, 2048, 4096,  4096,  4096
 		};
 
-		/* fw_debug ( "ARP packet\n" ); */
-		arp1394 = (struct ipv4_arp *)skb->data;
+		arp1394 = (struct rfc2734_arp *)skb->data;
 		arp = (struct arphdr *)skb->data;
 		arp_ptr = (unsigned char *)(arp + 1);
-		fifo_addr = (u64)ntohs(arp1394->fifo_hi) << 32 |
- ntohl(arp1394->fifo_lo);
-		max_rec = priv->card->max_receive;
-		if ( arp1394->max_rec < max_rec )
+		fifo_addr = (u64)ntohs(arp1394->fifo_hi) << 32
+				| ntohl(arp1394->fifo_lo);
+		max_rec = dev->card->max_receive;
+		if (arp1394->max_rec < max_rec)
 			max_rec = arp1394->max_rec;
 		sspd = arp1394->sspd;
-		/*
-		 * Sanity check. MacOSX seems to be sending us 131 in this
-		 * field (atleast on my Panther G5). Not sure why.
-		 */
-		if (sspd > 5 ) {
-			fw_notify ( "sspd %x out of range\n", sspd );
+		/* Sanity check.  OS X 10.3 PPC reportedly sends 131. */
+		if (sspd > SCODE_3200) {
+			fw_notify("sspd %x out of range\n", sspd);
 			sspd = 0;
 		}
 
-		max_payload = min(ipv4_speed_to_max_payload[sspd],
- (u16)(1 << (max_rec + 1))) - IPV4_UNFRAG_HDR_SIZE;
+		max_payload = min(fwnet_speed_to_max_payload[sspd],
+			(u16)(1 << (max_rec + 1))) - RFC2374_UNFRAG_HDR_SIZE;
 
-		guid = be64_to_cpu(get_unaligned(&arp1394->s_uniq_id));
-		node = ipv4_node_find_by_guid(priv, guid);
-		if (!node) {
-			fw_notify ( "No node for ARP packet from %llx\n", guid );
+		peer_guid = get_unaligned_be64(&arp1394->s_uniq_id);
+		peer = fwnet_peer_find_by_guid(dev, peer_guid);
+		if (!peer) {
+			fw_notify("No peer for ARP packet from %016llx\n",
+				  (unsigned long long)peer_guid);
 			goto failed_proto;
 		}
-		if ( node->nodeid != source_node_id || node->generation != priv->card->generation ) {
-			fw_notify ( "Internal error: node->nodeid (%x) != soucre_node_id (%x) or node->generation (%x) != priv->card->generation(%x)\n",
- node->nodeid, source_node_id, node->generation, priv->card->generation );
-			node->nodeid = source_node_id;
-			node->generation = priv->card->generation;
+
+		/* FIXME don't use card->generation */
+		if (peer->node_id != source_node_id ||
+		    peer->generation != dev->card->generation) {
+			fw_notify("Internal error: peer->node_id (%x) != "
+				  "source_node_id (%x) or peer->generation (%x)"
+				  " != dev->card->generation(%x)\n",
+				  peer->node_id, source_node_id,
+				  peer->generation, dev->card->generation);
+			peer->node_id = source_node_id;
+			peer->generation = dev->card->generation;
 		}
 
 		/* FIXME: for debugging */
-		if ( sspd > SCODE_400 )
+		if (sspd > SCODE_400)
 			sspd = SCODE_400;
 		/* Update our speed/payload/fifo_offset table */
 		/*
 		 * FIXME: this does not handle cases where two high-speed endpoints must use a slower speed because of
 		 * a lower speed hub between them.  We need to look at the actual topology map here.
 		 */
-		fw_debug ( "Setting node %p fifo %llx (was %llx), max_payload %x (was %x), speed %x (was %x)\n",
- node, fifo_addr, node->fifo, max_payload, node->max_payload, sspd, node->xmt_speed );
-		node->fifo =	fifo_addr;
-		node->max_payload = max_payload;
+		peer->fifo = fifo_addr;
+		peer->max_payload = max_payload;
 		/*
 		 * Only allow speeds to go down from their initial value.
-		 * Otherwise a local node that can only do S400 or slower may
-		 * be told to transmit at S800 to a faster remote node.
+		 * Otherwise a local peer that can only do S400 or slower may
+		 * be told to transmit at S800 to a faster remote peer.
 		 */
-		if ( node->xmt_speed > sspd )
-			node->xmt_speed = sspd;
+		if (peer->xmt_speed > sspd)
+			peer->xmt_speed = sspd;
 
 		/*
 		 * Now that we're done with the 1394 specific stuff, we'll
@@ -805,248 +684,257 @@ static int ipv4_finish_incoming_packet ( struct net_device *netdev,
 		 */
 
 		arp->ar_hln = 8;
-		arp_ptr += arp->ar_hln;		/* skip over sender unique id */
-		*(u32 *)arp_ptr = arp1394->sip; /* move sender IP addr */
-		arp_ptr += arp->ar_pln;		/* skip over sender IP addr */
+		/* skip over sender unique id */
+		arp_ptr += arp->ar_hln;
+		/* move sender IP addr */
+		put_unaligned(arp1394->sip, (u32 *)arp_ptr);
+		/* skip over sender IP addr */
+		arp_ptr += arp->ar_pln;
 
 		if (arp->ar_op == htons(ARPOP_REQUEST))
 			memset(arp_ptr, 0, sizeof(u64));
 		else
-			memcpy(arp_ptr, netdev->dev_addr, sizeof(u64));
+			memcpy(arp_ptr, net->dev_addr, sizeof(u64));
 	}
 
 	/* Now add the ethernet header. */
-	guid = cpu_to_be64(priv->card->guid);
-	if (dev_hard_header(skb, netdev, ether_type, is_broadcast ? &broadcast_hw : &guid, NULL,
- skb->len) >= 0) {
-		struct ipv4_ether_hdr *eth;
+	guid = cpu_to_be64(dev->card->guid);
+	if (dev_hard_header(skb, net, ether_type,
+			   is_broadcast ? &broadcast_hw : &guid,
+			   NULL, skb->len) >= 0) {
+		struct fwnet_header *eth;
 		u16 *rawp;
 		__be16 protocol;
 
 		skb_reset_mac_header(skb);
 		skb_pull(skb, sizeof(*eth));
-		eth = ipv4_ether_hdr(skb);
+		eth = (struct fwnet_header *)skb_mac_header(skb);
 		if (*eth->h_dest & 1) {
-			if (memcmp(eth->h_dest, netdev->broadcast, netdev->addr_len) == 0) {
-				fw_debug ( "Broadcast\n" );
+			if (memcmp(eth->h_dest, net->broadcast,
+				   net->addr_len) == 0)
 				skb->pkt_type = PACKET_BROADCAST;
-			}
 #if 0
 			else
 				skb->pkt_type = PACKET_MULTICAST;
 #endif
 		} else {
-			if (memcmp(eth->h_dest, netdev->dev_addr, netdev->addr_len)) {
+			if (memcmp(eth->h_dest, net->dev_addr, net->addr_len)) {
 				u64 a1, a2;
 
-				memcpy ( &a1, eth->h_dest, sizeof(u64));
-				memcpy ( &a2, netdev->dev_addr, sizeof(u64));
-				fw_debug ( "Otherhost %llx %llx %x\n", a1, a2, netdev->addr_len );
+				memcpy(&a1, eth->h_dest, sizeof(u64));
+				memcpy(&a2, net->dev_addr, sizeof(u64));
 				skb->pkt_type = PACKET_OTHERHOST;
 			}
 		}
 		if (ntohs(eth->h_proto) >= 1536) {
-			fw_debug ( " proto %x %x\n", eth->h_proto, ntohs(eth->h_proto) );
 			protocol = eth->h_proto;
 		} else {
 			rawp = (u16 *)skb->data;
-			if (*rawp == 0xFFFF) {
-				fw_debug ( "proto 802_3\n" );
+			if (*rawp == 0xffff)
 				protocol = htons(ETH_P_802_3);
-			} else {
-				fw_debug ( "proto 802_2\n" );
+			else
 				protocol = htons(ETH_P_802_2);
-			}
 		}
 		skb->protocol = protocol;
 	}
 	status = netif_rx(skb);
-	if ( status == NET_RX_DROP) {
-		netdev->stats.rx_errors++;
-		netdev->stats.rx_dropped++;
+	if (status == NET_RX_DROP) {
+		net->stats.rx_errors++;
+		net->stats.rx_dropped++;
 	} else {
-		netdev->stats.rx_packets++;
-		netdev->stats.rx_bytes += skb->len;
+		net->stats.rx_packets++;
+		net->stats.rx_bytes += skb->len;
 	}
-	if (netif_queue_stopped(netdev))
-		netif_wake_queue(netdev);
+	if (netif_queue_stopped(net))
+		netif_wake_queue(net);
+
 	return 0;
 
  failed_proto:
-	netdev->stats.rx_errors++;
-	netdev->stats.rx_dropped++;
+	net->stats.rx_errors++;
+	net->stats.rx_dropped++;
+
 	dev_kfree_skb_any(skb);
-	if (netif_queue_stopped(netdev))
-		netif_wake_queue(netdev);
-	netdev->last_rx = jiffies;
+	if (netif_queue_stopped(net))
+		netif_wake_queue(net);
+
+	net->last_rx = jiffies;
+
 	return 0;
 }
 
-/* ------------------------------------------------------------------ */
-
-static int ipv4_incoming_packet ( struct ipv4_priv *priv, u32 *buf, int len, u16 source_node_id, bool is_broadcast ) {
+static int fwnet_incoming_packet(struct fwnet_device *dev, __be32 *buf, int len,
+				 u16 source_node_id, bool is_broadcast)
+{
 	struct sk_buff *skb;
-	struct net_device *netdev;
-	struct ipv4_hdr hdr;
+	struct net_device *net;
+	struct rfc2734_header hdr;
 	unsigned lf;
 	unsigned long flags;
-	struct ipv4_node *node;
-	struct ipv4_partial_datagram *pd;
+	struct fwnet_peer *peer;
+	struct fwnet_partial_datagram *pd;
 	int fg_off;
 	int dg_size;
 	u16 datagram_label;
 	int retval;
 	u16 ether_type;
 
-	fw_debug ( "ipv4_incoming_packet(%p, %p, %d, %x, %s)\n", priv, buf, len, source_node_id, is_broadcast ? "true" : "false" );
-	netdev = priv->card->netdev;
+	net = dev->card->netdev;
 
-	hdr.w0 = ntohl(buf[0]);
-	lf = ipv4_get_hdr_lf(&hdr);
-	if ( lf == IPV4_HDR_UNFRAG ) {
+	hdr.w0 = be32_to_cpu(buf[0]);
+	lf = fwnet_get_hdr_lf(&hdr);
+	if (lf == RFC2374_HDR_UNFRAG) {
 		/*
 		 * An unfragmented datagram has been received by the ieee1394
 		 * bus. Build an skbuff around it so we can pass it to the
 		 * high level network layer.
 		 */
-		ether_type = ipv4_get_hdr_ether_type(&hdr);
-		fw_debug ( "header w0 = %x, lf = %x, ether_type = %x\n", hdr.w0, lf, ether_type );
+		ether_type = fwnet_get_hdr_ether_type(&hdr);
 		buf++;
-		len -= IPV4_UNFRAG_HDR_SIZE;
+		len -= RFC2374_UNFRAG_HDR_SIZE;
 
-		skb = dev_alloc_skb(len + netdev->hard_header_len + 15);
+		skb = dev_alloc_skb(len + net->hard_header_len + 15);
 		if (unlikely(!skb)) {
-			fw_error ( "Out of memory for incoming packet\n");
-			netdev->stats.rx_dropped++;
+			fw_error("out of memory\n");
+			net->stats.rx_dropped++;
+
 			return -1;
 		}
-		skb_reserve(skb, (netdev->hard_header_len + 15) & ~15);
-		memcpy(skb_put(skb, len), buf, len );
-		return ipv4_finish_incoming_packet(netdev, skb, source_node_id, is_broadcast, ether_type );
+		skb_reserve(skb, (net->hard_header_len + 15) & ~15);
+		memcpy(skb_put(skb, len), buf, len);
+
+		return fwnet_finish_incoming_packet(net, skb, source_node_id,
+						    is_broadcast, ether_type);
 	}
 	/* A datagram fragment has been received, now the fun begins. */
 	hdr.w1 = ntohl(buf[1]);
-	buf +=2;
-	len -= IPV4_FRAG_HDR_SIZE;
-	if ( lf ==IPV4_HDR_FIRSTFRAG ) {
-		ether_type = ipv4_get_hdr_ether_type(&hdr);
+	buf += 2;
+	len -= RFC2374_FRAG_HDR_SIZE;
+	if (lf == RFC2374_HDR_FIRSTFRAG) {
+		ether_type = fwnet_get_hdr_ether_type(&hdr);
 		fg_off = 0;
 	} else {
-		fg_off = ipv4_get_hdr_fg_off(&hdr);
-		ether_type = 0; /* Shut up compiler! */
+		ether_type = 0;
+		fg_off = fwnet_get_hdr_fg_off(&hdr);
 	}
-	datagram_label = ipv4_get_hdr_dgl(&hdr);
-	dg_size = ipv4_get_hdr_dg_size(&hdr); /* ??? + 1 */
-	fw_debug ( "fragmented: %x.%x = lf %x, ether_type %x, fg_off %x, dgl %x, dg_size %x\n", hdr.w0, hdr.w1, lf, ether_type, fg_off, datagram_label, dg_size );
-	node = ipv4_node_find_by_nodeid ( priv, source_node_id);
-	spin_lock_irqsave(&node->pdg_lock, flags);
-	pd = ipv4_pd_find( node, datagram_label );
+	datagram_label = fwnet_get_hdr_dgl(&hdr);
+	dg_size = fwnet_get_hdr_dg_size(&hdr); /* ??? + 1 */
+	peer = fwnet_peer_find_by_node_id(dev, source_node_id);
+
+	spin_lock_irqsave(&peer->pdg_lock, flags);
+
+	pd = fwnet_pd_find(peer, datagram_label);
 	if (pd == NULL) {
-		while ( node->pdg_size >= ipv4_mpd ) {
+		while (peer->pdg_size >= FWNET_MAX_FRAGMENTS) {
 			/* remove the oldest */
-			ipv4_pd_delete ( list_first_entry(&node->pdg_list, struct ipv4_partial_datagram, pdg_list) );
-			node->pdg_size--;
+			fwnet_pd_delete(list_first_entry(&peer->pd_list,
+				struct fwnet_partial_datagram, pd_link));
+			peer->pdg_size--;
 		}
-		pd = ipv4_pd_new ( netdev, node, datagram_label, dg_size,
- buf, fg_off, len);
-		if ( pd == NULL) {
+		pd = fwnet_pd_new(net, peer, datagram_label,
+				  dg_size, buf, fg_off, len);
+		if (pd == NULL) {
 			retval = -ENOMEM;
 			goto bad_proto;
 		}
-		node->pdg_size++;
+		peer->pdg_size++;
 	} else {
-		if (ipv4_frag_overlap(pd, fg_off, len) || pd->datagram_size != dg_size) {
+		if (fwnet_frag_overlap(pd, fg_off, len) ||
+		    pd->datagram_size != dg_size) {
 			/*
 			 * Differing datagram sizes or overlapping fragments,
-			 * Either way the remote machine is playing silly buggers
-			 * with us: obliterate the old datagram and start a new one.
+			 * discard old datagram and start a new one.
 			 */
-			ipv4_pd_delete ( pd );
-			pd = ipv4_pd_new ( netdev, node, datagram_label,
- dg_size, buf, fg_off, len);
-			if ( pd == NULL ) {
+			fwnet_pd_delete(pd);
+			pd = fwnet_pd_new(net, peer, datagram_label,
+					  dg_size, buf, fg_off, len);
+			if (pd == NULL) {
 				retval = -ENOMEM;
-				node->pdg_size--;
+				peer->pdg_size--;
 				goto bad_proto;
 			}
 		} else {
-			bool worked;
-
-			worked = ipv4_pd_update ( node, pd,
- buf, fg_off, len );
-			if ( ! worked ) {
+			if (!fwnet_pd_update(peer, pd, buf, fg_off, len)) {
 				/*
 				 * Couldn't save off fragment anyway
 				 * so might as well obliterate the
 				 * datagram now.
 				 */
-				ipv4_pd_delete ( pd );
-				node->pdg_size--;
+				fwnet_pd_delete(pd);
+				peer->pdg_size--;
 				goto bad_proto;
 			}
 		}
 	} /* new datagram or add to existing one */
 
-	if ( lf == IPV4_HDR_FIRSTFRAG )
+	if (lf == RFC2374_HDR_FIRSTFRAG)
 		pd->ether_type = ether_type;
-	if ( ipv4_pd_is_complete ( pd ) ) {
+
+	if (fwnet_pd_is_complete(pd)) {
 		ether_type = pd->ether_type;
-		node->pdg_size--;
+		peer->pdg_size--;
 		skb = skb_get(pd->skb);
-		ipv4_pd_delete ( pd );
-		spin_unlock_irqrestore(&node->pdg_lock, flags);
-		return ipv4_finish_incoming_packet ( netdev, skb, source_node_id, false, ether_type );
+		fwnet_pd_delete(pd);
+
+		spin_unlock_irqrestore(&peer->pdg_lock, flags);
+
+		return fwnet_finish_incoming_packet(net, skb, source_node_id,
+						    false, ether_type);
 	}
 	/*
 	 * Datagram is not complete, we're done for the
 	 * moment.
 	 */
-	spin_unlock_irqrestore(&node->pdg_lock, flags);
+	spin_unlock_irqrestore(&peer->pdg_lock, flags);
+
 	return 0;
 
  bad_proto:
-	spin_unlock_irqrestore(&node->pdg_lock, flags);
-	if (netif_queue_stopped(netdev))
-		netif_wake_queue(netdev);
+	spin_unlock_irqrestore(&peer->pdg_lock, flags);
+
+	if (netif_queue_stopped(net))
+		netif_wake_queue(net);
+
 	return 0;
 }
 
-static void ipv4_receive_packet ( struct fw_card *card, struct fw_request *r,
- int tcode, int destination, int source, int generation, int speed,
- unsigned long long offset, void *payload, size_t length, void *callback_data ) {
-	struct ipv4_priv *priv;
+static void fwnet_receive_packet(struct fw_card *card, struct fw_request *r,
+		int tcode, int destination, int source, int generation,
+		int speed, unsigned long long offset, void *payload,
+		size_t length, void *callback_data)
+{
+	struct fwnet_device *dev;
 	int status;
 
-	fw_debug ( "ipv4_receive_packet(%p,%p,%x,%x,%x,%x,%x,%llx,%p,%lx,%p)\n",
- card, r, tcode, destination, source, generation, speed, offset, payload,
- (unsigned long)length, callback_data);
-	print_hex_dump ( KERN_DEBUG, "header: ", DUMP_PREFIX_OFFSET, 32, 1, payload, length, false );
-	priv = callback_data;
-	if (   tcode != TCODE_WRITE_BLOCK_REQUEST
-	    || destination != card->node_id
-	    || generation != card->generation
-	    || offset != priv->handler.offset ) {
+	dev = callback_data;
+	if (tcode != TCODE_WRITE_BLOCK_REQUEST
+	    || destination != card->node_id	/* <- FIXME */
+	    || generation != card->generation	/* <- FIXME */
+	    || offset != dev->handler.offset) {
 		fw_send_response(card, r, RCODE_CONFLICT_ERROR);
-		fw_debug("Conflict error card node_id=%x, card generation=%x, local offset %llx\n",
- card->node_id, card->generation, (unsigned long long)priv->handler.offset );
+
 		return;
 	}
-	status = ipv4_incoming_packet ( priv, payload, length, source, false );
-	if ( status != 0 ) {
-		fw_error ( "Incoming packet failure\n" );
-		fw_send_response ( card, r, RCODE_CONFLICT_ERROR );
+
+	status = fwnet_incoming_packet(dev, payload, length, source, false);
+	if (status != 0) {
+		fw_error("Incoming packet failure\n");
+		fw_send_response(card, r, RCODE_CONFLICT_ERROR);
+
 		return;
 	}
-	fw_send_response ( card, r, RCODE_COMPLETE );
+
+	fw_send_response(card, r, RCODE_COMPLETE);
 }
 
-static void ipv4_receive_broadcast(struct fw_iso_context *context, u32 cycle,
- size_t header_length, void *header, void *data) {
-	struct ipv4_priv *priv;
+static void fwnet_receive_broadcast(struct fw_iso_context *context,
+		u32 cycle, size_t header_length, void *header, void *data)
+{
+	struct fwnet_device *dev;
 	struct fw_iso_packet packet;
 	struct fw_card *card;
-	u16 *hdr_ptr;
-	u32 *buf_ptr;
+	__be16 *hdr_ptr;
+	__be32 *buf_ptr;
 	int retval;
 	u32 length;
 	u16 source_node_id;
@@ -1055,70 +943,68 @@ static void ipv4_receive_broadcast(struct fw_iso_context *context, u32 cycle,
 	unsigned long offset;
 	unsigned long flags;
 
-	fw_debug ( "ipv4_receive_broadcast ( context=%p, cycle=%x, header_length=%lx, header=%p, data=%p )\n", context, cycle, (unsigned long)header_length, header, data );
-	print_hex_dump ( KERN_DEBUG, "header: ", DUMP_PREFIX_OFFSET, 32, 1, header, header_length, false );
-	priv = data;
-	card = priv->card;
+	dev = data;
+	card = dev->card;
 	hdr_ptr = header;
-	length = ntohs(hdr_ptr[0]);
-	spin_lock_irqsave(&priv->lock,flags);
-	offset = priv->rcv_buffer_size * priv->broadcast_rcv_next_ptr;
-	buf_ptr = priv->broadcast_rcv_buffer_ptrs[priv->broadcast_rcv_next_ptr++];
-	if ( priv->broadcast_rcv_next_ptr == priv->num_broadcast_rcv_ptrs )
-		priv->broadcast_rcv_next_ptr = 0;
-	spin_unlock_irqrestore(&priv->lock,flags);
-	fw_debug ( "length %u at %p\n", length, buf_ptr );
-	print_hex_dump ( KERN_DEBUG, "buffer: ", DUMP_PREFIX_OFFSET, 32, 1, buf_ptr, length, false );
+	length = be16_to_cpup(hdr_ptr);
+
+	spin_lock_irqsave(&dev->lock, flags);
+
+	offset = dev->rcv_buffer_size * dev->broadcast_rcv_next_ptr;
+	buf_ptr = dev->broadcast_rcv_buffer_ptrs[dev->broadcast_rcv_next_ptr++];
+	if (dev->broadcast_rcv_next_ptr == dev->num_broadcast_rcv_ptrs)
+		dev->broadcast_rcv_next_ptr = 0;
+
+	spin_unlock_irqrestore(&dev->lock, flags);
 
 	specifier_id =    (be32_to_cpu(buf_ptr[0]) & 0xffff) << 8
 			| (be32_to_cpu(buf_ptr[1]) & 0xff000000) >> 24;
-	ver = be32_to_cpu(buf_ptr[1]) & 0xFFFFFF;
+	ver = be32_to_cpu(buf_ptr[1]) & 0xffffff;
 	source_node_id = be32_to_cpu(buf_ptr[0]) >> 16;
-	/* fw_debug ( "source %x SpecID %x ver %x\n", source_node_id, specifier_id, ver ); */
-	if ( specifier_id == IPV4_GASP_SPECIFIER_ID && ver == IPV4_GASP_VERSION ) {
+
+	if (specifier_id == IANA_SPECIFIER_ID && ver == RFC2734_SW_VERSION) {
 		buf_ptr += 2;
-		length -= IPV4_GASP_OVERHEAD;
-		ipv4_incoming_packet(priv, buf_ptr, length, source_node_id, true);
-	} else
-		fw_debug ( "Ignoring packet: not GASP\n" );
-	packet.payload_length = priv->rcv_buffer_size;
+		length -= IEEE1394_GASP_HDR_SIZE;
+		fwnet_incoming_packet(dev, buf_ptr, length,
+				      source_node_id, true);
+	}
+
+	packet.payload_length = dev->rcv_buffer_size;
 	packet.interrupt = 1;
 	packet.skip = 0;
 	packet.tag = 3;
 	packet.sy = 0;
-	packet.header_length = IPV4_GASP_OVERHEAD;
-	spin_lock_irqsave(&priv->lock,flags);
-	retval = fw_iso_context_queue ( priv->broadcast_rcv_context, &packet,
- &priv->broadcast_rcv_buffer, offset );
-	spin_unlock_irqrestore(&priv->lock,flags);
-	if ( retval < 0 )
-		fw_error ( "requeue failed\n" );
-}
+	packet.header_length = IEEE1394_GASP_HDR_SIZE;
+
+	spin_lock_irqsave(&dev->lock, flags);
 
-static void debug_ptask ( struct ipv4_packet_task *ptask ) {
-	static const char *tx_types[] = { "Unknown", "GASP", "Write" };
-
-	fw_debug ( "packet %p { hdr { w0 %x w1 %x }, skb %p, priv %p,"
- " tx_type %s, outstanding_pkts %d, max_payload %x, fifo %llx,"
- " speed %x, dest_node %x, generation %x }\n",
- ptask, ptask->hdr.w0, ptask->hdr.w1, ptask->skb, ptask->priv,
- ptask->tx_type > IPV4_WRREQ ? "Invalid" : tx_types[ptask->tx_type],
- ptask->outstanding_pkts,  ptask->max_payload,
- ptask->fifo_addr, ptask->speed, ptask->dest_node, ptask->generation );
-	print_hex_dump ( KERN_DEBUG, "packet :", DUMP_PREFIX_OFFSET, 32, 1,
- ptask->skb->data, ptask->skb->len, false );
+	retval = fw_iso_context_queue(dev->broadcast_rcv_context, &packet,
+				      &dev->broadcast_rcv_buffer, offset);
+
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	if (retval < 0)
+		fw_error("requeue failed\n");
 }
 
-static void ipv4_transmit_packet_done ( struct ipv4_packet_task *ptask ) {
-	struct ipv4_priv *priv;
+static struct kmem_cache *fwnet_packet_task_cache;
+
+static int fwnet_send_packet(struct fwnet_packet_task *ptask);
+
+static void fwnet_transmit_packet_done(struct fwnet_packet_task *ptask)
+{
+	struct fwnet_device *dev;
 	unsigned long flags;
 
-	priv = ptask->priv;
-	spin_lock_irqsave ( &priv->lock, flags );
-	list_del ( &ptask->packet_list );
-	spin_unlock_irqrestore ( &priv->lock, flags );
-	ptask->outstanding_pkts--;
-	if ( ptask->outstanding_pkts > 0 ) {
+	dev = ptask->dev;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	list_del(&ptask->pt_link);
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	ptask->outstanding_pkts--; /* FIXME access inside lock */
+
+	if (ptask->outstanding_pkts > 0) {
 		u16 dg_size;
 		u16 fg_off;
 		u16 datagram_label;
@@ -1126,133 +1012,139 @@ static void ipv4_transmit_packet_done ( struct ipv4_packet_task *ptask ) {
 		struct sk_buff *skb;
 
 		/* Update the ptask to point to the next fragment and send it */
-		lf = ipv4_get_hdr_lf(&ptask->hdr);
+		lf = fwnet_get_hdr_lf(&ptask->hdr);
 		switch (lf) {
-		case IPV4_HDR_LASTFRAG:
-		case IPV4_HDR_UNFRAG:
+		case RFC2374_HDR_LASTFRAG:
+		case RFC2374_HDR_UNFRAG:
 		default:
-			fw_error ( "Outstanding packet %x lf %x, header %x,%x\n", ptask->outstanding_pkts, lf, ptask->hdr.w0, ptask->hdr.w1 );
+			fw_error("Outstanding packet %x lf %x, header %x,%x\n",
+				 ptask->outstanding_pkts, lf, ptask->hdr.w0,
+				 ptask->hdr.w1);
 			BUG();
 
-		case IPV4_HDR_FIRSTFRAG:
+		case RFC2374_HDR_FIRSTFRAG:
 			/* Set frag type here for future interior fragments */
-			dg_size = ipv4_get_hdr_dg_size(&ptask->hdr);
-			fg_off = ptask->max_payload - IPV4_FRAG_HDR_SIZE;
-			datagram_label = ipv4_get_hdr_dgl(&ptask->hdr);
+			dg_size = fwnet_get_hdr_dg_size(&ptask->hdr);
+			fg_off = ptask->max_payload - RFC2374_FRAG_HDR_SIZE;
+			datagram_label = fwnet_get_hdr_dgl(&ptask->hdr);
 			break;
 
-		case IPV4_HDR_INTFRAG:
-			dg_size = ipv4_get_hdr_dg_size(&ptask->hdr);
-			fg_off = ipv4_get_hdr_fg_off(&ptask->hdr) + ptask->max_payload - IPV4_FRAG_HDR_SIZE;
-			datagram_label = ipv4_get_hdr_dgl(&ptask->hdr);
+		case RFC2374_HDR_INTFRAG:
+			dg_size = fwnet_get_hdr_dg_size(&ptask->hdr);
+			fg_off = fwnet_get_hdr_fg_off(&ptask->hdr)
+				  + ptask->max_payload - RFC2374_FRAG_HDR_SIZE;
+			datagram_label = fwnet_get_hdr_dgl(&ptask->hdr);
 			break;
 		}
 		skb = ptask->skb;
-		skb_pull ( skb, ptask->max_payload );
-		if ( ptask->outstanding_pkts > 1 ) {
-			ipv4_make_sf_hdr ( &ptask->hdr,
-  IPV4_HDR_INTFRAG, dg_size, fg_off, datagram_label );
+		skb_pull(skb, ptask->max_payload);
+		if (ptask->outstanding_pkts > 1) {
+			fwnet_make_sf_hdr(&ptask->hdr, RFC2374_HDR_INTFRAG,
+					  dg_size, fg_off, datagram_label);
 		} else {
-			ipv4_make_sf_hdr ( &ptask->hdr,
-  IPV4_HDR_LASTFRAG, dg_size, fg_off, datagram_label );
-			ptask->max_payload = skb->len + IPV4_FRAG_HDR_SIZE;
-
+			fwnet_make_sf_hdr(&ptask->hdr, RFC2374_HDR_LASTFRAG,
+					  dg_size, fg_off, datagram_label);
+			ptask->max_payload = skb->len + RFC2374_FRAG_HDR_SIZE;
 		}
-		ipv4_send_packet ( ptask );
+		fwnet_send_packet(ptask);
 	} else {
-		dev_kfree_skb_any ( ptask->skb );
-		kmem_cache_free( ipv4_packet_task_cache, ptask );
+		dev_kfree_skb_any(ptask->skb);
+		kmem_cache_free(fwnet_packet_task_cache, ptask);
 	}
 }
 
-static void ipv4_write_complete ( struct fw_card *card, int rcode,
- void *payload, size_t length, void *data ) {
-	struct ipv4_packet_task *ptask;
+static void fwnet_write_complete(struct fw_card *card, int rcode,
+				 void *payload, size_t length, void *data)
+{
+	struct fwnet_packet_task *ptask;
 
 	ptask = data;
-	fw_debug ( "ipv4_write_complete ( %p, %x, %p, %lx, %p )\n",
- card, rcode, payload, (unsigned long)length, data );
-	debug_ptask ( ptask );
 
-	if ( rcode == RCODE_COMPLETE ) {
-		ipv4_transmit_packet_done ( ptask );
-	} else {
-		fw_error ( "ipv4_write_complete: failed: %x\n", rcode );
+	if (rcode == RCODE_COMPLETE)
+		fwnet_transmit_packet_done(ptask);
+	else
+		fw_error("fwnet_write_complete: failed: %x\n", rcode);
 		/* ??? error recovery */
-	}
 }
 
-static int ipv4_send_packet ( struct ipv4_packet_task *ptask ) {
-	struct ipv4_priv *priv;
+static int fwnet_send_packet(struct fwnet_packet_task *ptask)
+{
+	struct fwnet_device *dev;
 	unsigned tx_len;
-	struct ipv4_hdr *bufhdr;
+	struct rfc2734_header *bufhdr;
 	unsigned long flags;
-	struct net_device *netdev;
-#if 0 /* stefanr */
-	int retval;
-#endif
+	struct net_device *net;
 
-	fw_debug ( "ipv4_send_packet\n" );
-	debug_ptask ( ptask );
-	priv = ptask->priv;
+	dev = ptask->dev;
 	tx_len = ptask->max_payload;
-	switch (ipv4_get_hdr_lf(&ptask->hdr)) {
-	case IPV4_HDR_UNFRAG:
-		bufhdr = (struct ipv4_hdr *)skb_push(ptask->skb, IPV4_UNFRAG_HDR_SIZE);
-		bufhdr->w0 = htonl(ptask->hdr.w0);
+	switch (fwnet_get_hdr_lf(&ptask->hdr)) {
+	case RFC2374_HDR_UNFRAG:
+		bufhdr = (struct rfc2734_header *)
+				skb_push(ptask->skb, RFC2374_UNFRAG_HDR_SIZE);
+		put_unaligned_be32(ptask->hdr.w0, &bufhdr->w0);
 		break;
 
-	case IPV4_HDR_FIRSTFRAG:
-	case IPV4_HDR_INTFRAG:
-	case IPV4_HDR_LASTFRAG:
-		bufhdr = (struct ipv4_hdr *)skb_push(ptask->skb, IPV4_FRAG_HDR_SIZE);
-		bufhdr->w0 = htonl(ptask->hdr.w0);
-		bufhdr->w1 = htonl(ptask->hdr.w1);
+	case RFC2374_HDR_FIRSTFRAG:
+	case RFC2374_HDR_INTFRAG:
+	case RFC2374_HDR_LASTFRAG:
+		bufhdr = (struct rfc2734_header *)
+				skb_push(ptask->skb, RFC2374_FRAG_HDR_SIZE);
+		put_unaligned_be32(ptask->hdr.w0, &bufhdr->w0);
+		put_unaligned_be32(ptask->hdr.w1, &bufhdr->w1);
 		break;
 
 	default:
 		BUG();
 	}
-	if ( ptask->tx_type == IPV4_GASP ) {
-		u32 *packets;
+	if (ptask->dest_node == IEEE1394_ALL_NODES) {
+		u8 *p;
 		int generation;
-		int nodeid;
+		int node_id;
 
 		/* ptask->generation may not have been set yet */
-		generation = priv->card->generation;
+		generation = dev->card->generation;
 		smp_rmb();
-		nodeid = priv->card->node_id;
-		packets = (u32 *)skb_push(ptask->skb, sizeof(u32)*2);
-		packets[0] = htonl(nodeid << 16 | (IPV4_GASP_SPECIFIER_ID>>8));
-		packets[1] = htonl((IPV4_GASP_SPECIFIER_ID & 0xFF) << 24 | IPV4_GASP_VERSION);
-		fw_send_request ( priv->card, &ptask->transaction, TCODE_STREAM_DATA,
- fw_stream_packet_destination_id(3, BROADCAST_CHANNEL, 0),
- generation, SCODE_100, 0ULL, ptask->skb->data, tx_len + 8, ipv4_write_complete, ptask );
-		spin_lock_irqsave(&priv->lock,flags);
-		list_add_tail ( &ptask->packet_list, &priv->broadcasted_list );
-		spin_unlock_irqrestore(&priv->lock,flags);
-#if 0 /* stefanr */
-		return retval;
-#else
+		node_id = dev->card->node_id;
+
+		p = skb_push(ptask->skb, 8);
+		put_unaligned_be32(node_id << 16 | IANA_SPECIFIER_ID >> 8, p);
+		put_unaligned_be32((IANA_SPECIFIER_ID & 0xff) << 24
+						| RFC2734_SW_VERSION, &p[4]);
+
+		/* We should not transmit if broadcast_channel.valid == 0. */
+		fw_send_request(dev->card, &ptask->transaction,
+				TCODE_STREAM_DATA,
+				fw_stream_packet_destination_id(3,
+						IEEE1394_BROADCAST_CHANNEL, 0),
+				generation, SCODE_100, 0ULL, ptask->skb->data,
+				tx_len + 8, fwnet_write_complete, ptask);
+
+		/* FIXME race? */
+		spin_lock_irqsave(&dev->lock, flags);
+		list_add_tail(&ptask->pt_link, &dev->broadcasted_list);
+		spin_unlock_irqrestore(&dev->lock, flags);
+
 		return 0;
-#endif
 	}
-	fw_debug("send_request (%p, %p, WRITE_BLOCK, %x, %x, %x, %llx, %p, %d, %p, %p\n",
- priv->card, &ptask->transaction, ptask->dest_node, ptask->generation,
- ptask->speed, (unsigned long long)ptask->fifo_addr, ptask->skb->data, tx_len,
- ipv4_write_complete, ptask );
-	fw_send_request ( priv->card, &ptask->transaction,
- TCODE_WRITE_BLOCK_REQUEST, ptask->dest_node, ptask->generation, ptask->speed,
- ptask->fifo_addr, ptask->skb->data, tx_len, ipv4_write_complete, ptask );
-	spin_lock_irqsave(&priv->lock,flags);
-	list_add_tail ( &ptask->packet_list, &priv->sent_list );
-	spin_unlock_irqrestore(&priv->lock,flags);
-	netdev = priv->card->netdev;
-	netdev->trans_start = jiffies;
+
+	fw_send_request(dev->card, &ptask->transaction,
+			TCODE_WRITE_BLOCK_REQUEST, ptask->dest_node,
+			ptask->generation, ptask->speed, ptask->fifo_addr,
+			ptask->skb->data, tx_len, fwnet_write_complete, ptask);
+
+	/* FIXME race? */
+	spin_lock_irqsave(&dev->lock, flags);
+	list_add_tail(&ptask->pt_link, &dev->sent_list);
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	net = dev->card->netdev;
+	net->trans_start = jiffies;
+
 	return 0;
 }
 
-static int ipv4_broadcast_start ( struct ipv4_priv *priv ) {
+static int fwnet_broadcast_start(struct fwnet_device *dev)
+{
 	struct fw_iso_context *context;
 	int retval;
 	unsigned num_packets;
@@ -1260,150 +1152,151 @@ static int ipv4_broadcast_start ( struct ipv4_priv *priv ) {
 	struct fw_iso_packet packet;
 	unsigned long offset;
 	unsigned u;
-	/* unsigned transmit_speed; */
 
-#if 0 /* stefanr */
-	if ( priv->card->broadcast_channel != (BROADCAST_CHANNEL_VALID|BROADCAST_CHANNEL_INITIAL)) {
-		fw_notify ( "Invalid broadcast channel %x\n", priv->card->broadcast_channel );
-		/* FIXME: try again later? */
-		/* return -EINVAL; */
-	}
-#endif
-	if ( priv->local_fifo == INVALID_FIFO_ADDR ) {
-		struct fw_address_region region;
-
-		priv->handler.length = FIFO_SIZE;
-		priv->handler.address_callback = ipv4_receive_packet;
-		priv->handler.callback_data = priv;
-		/* FIXME: this is OHCI, but what about others? */
-		region.start = 0xffff00000000ULL;
-		region.end =   0xfffffffffffcULL;
-
-		retval = fw_core_add_address_handler ( &priv->handler, &region );
-		if ( retval < 0 )
+	if (dev->local_fifo == FWNET_NO_FIFO_ADDR) {
+		/* outside OHCI posted write area? */
+		static const struct fw_address_region region = {
+			.start = 0xffff00000000ULL,
+			.end   = CSR_REGISTER_BASE,
+		};
+
+		dev->handler.length = 4096;
+		dev->handler.address_callback = fwnet_receive_packet;
+		dev->handler.callback_data = dev;
+
+		retval = fw_core_add_address_handler(&dev->handler, &region);
+		if (retval < 0)
 			goto failed_initial;
-		priv->local_fifo = priv->handler.offset;
+
+		dev->local_fifo = dev->handler.offset;
 	}
 
-	/*
-	 * FIXME: rawiso limits us to PAGE_SIZE.  This only matters if we ever have
-	 * a machine with PAGE_SIZE < 4096
-	 */
-	max_receive = 1U << (priv->card->max_receive + 1);
-	num_packets = ( ipv4_iso_page_count * PAGE_SIZE ) / max_receive;
-	if ( ! priv->broadcast_rcv_context ) {
+	max_receive = 1U << (dev->card->max_receive + 1);
+	num_packets = (FWNET_ISO_PAGE_COUNT * PAGE_SIZE) / max_receive;
+
+	if (!dev->broadcast_rcv_context) {
 		void **ptrptr;
 
-		context = fw_iso_context_create ( priv->card,
- FW_ISO_CONTEXT_RECEIVE, BROADCAST_CHANNEL,
- priv->card->link_speed, 8, ipv4_receive_broadcast, priv );
+		context = fw_iso_context_create(dev->card,
+		    FW_ISO_CONTEXT_RECEIVE, IEEE1394_BROADCAST_CHANNEL,
+		    dev->card->link_speed, 8, fwnet_receive_broadcast, dev);
 		if (IS_ERR(context)) {
 			retval = PTR_ERR(context);
 			goto failed_context_create;
 		}
-		retval = fw_iso_buffer_init ( &priv->broadcast_rcv_buffer,
- priv->card, ipv4_iso_page_count, DMA_FROM_DEVICE );
-		if ( retval < 0 )
+
+		retval = fw_iso_buffer_init(&dev->broadcast_rcv_buffer,
+		    dev->card, FWNET_ISO_PAGE_COUNT, DMA_FROM_DEVICE);
+		if (retval < 0)
 			goto failed_buffer_init;
-		ptrptr = kmalloc ( sizeof(void*)*num_packets, GFP_KERNEL );
-		if ( ! ptrptr ) {
+
+		ptrptr = kmalloc(sizeof(void *) * num_packets, GFP_KERNEL);
+		if (!ptrptr) {
 			retval = -ENOMEM;
 			goto failed_ptrs_alloc;
 		}
-		priv->broadcast_rcv_buffer_ptrs = ptrptr;
-		for ( u = 0; u < ipv4_iso_page_count; u++ ) {
+
+		dev->broadcast_rcv_buffer_ptrs = ptrptr;
+		for (u = 0; u < FWNET_ISO_PAGE_COUNT; u++) {
 			void *ptr;
 			unsigned v;
 
-			ptr = kmap ( priv->broadcast_rcv_buffer.pages[u] );
-			for ( v = 0; v < num_packets / ipv4_iso_page_count; v++ )
-				*ptrptr++ = (void *)((char *)ptr + v * max_receive);
+			ptr = kmap(dev->broadcast_rcv_buffer.pages[u]);
+			for (v = 0; v < num_packets / FWNET_ISO_PAGE_COUNT; v++)
+				*ptrptr++ = (void *)
+						((char *)ptr + v * max_receive);
 		}
-		priv->broadcast_rcv_context = context;
-	} else
-		context = priv->broadcast_rcv_context;
+		dev->broadcast_rcv_context = context;
+	} else {
+		context = dev->broadcast_rcv_context;
+	}
 
 	packet.payload_length = max_receive;
 	packet.interrupt = 1;
 	packet.skip = 0;
 	packet.tag = 3;
 	packet.sy = 0;
-	packet.header_length = IPV4_GASP_OVERHEAD;
+	packet.header_length = IEEE1394_GASP_HDR_SIZE;
 	offset = 0;
-	for ( u = 0; u < num_packets; u++ ) {
-		retval = fw_iso_context_queue ( context, &packet,
- &priv->broadcast_rcv_buffer, offset );
-		if ( retval < 0 )
+
+	for (u = 0; u < num_packets; u++) {
+		retval = fw_iso_context_queue(context, &packet,
+				&dev->broadcast_rcv_buffer, offset);
+		if (retval < 0)
 			goto failed_rcv_queue;
+
 		offset += max_receive;
 	}
-	priv->num_broadcast_rcv_ptrs = num_packets;
-	priv->rcv_buffer_size = max_receive;
-	priv->broadcast_rcv_next_ptr = 0U;
-	retval = fw_iso_context_start ( context, -1, 0, FW_ISO_CONTEXT_MATCH_ALL_TAGS ); /* ??? sync */
-	if ( retval < 0 )
+	dev->num_broadcast_rcv_ptrs = num_packets;
+	dev->rcv_buffer_size = max_receive;
+	dev->broadcast_rcv_next_ptr = 0U;
+	retval = fw_iso_context_start(context, -1, 0,
+			FW_ISO_CONTEXT_MATCH_ALL_TAGS); /* ??? sync */
+	if (retval < 0)
 		goto failed_rcv_queue;
-	/* FIXME: adjust this when we know the max receive speeds of all other IP nodes on the bus. */
-	/* since we only xmt at S100 ??? */
-	priv->broadcast_xmt_max_payload = S100_BUFFER_SIZE - IPV4_GASP_OVERHEAD - IPV4_UNFRAG_HDR_SIZE;
-	priv->broadcast_state = IPV4_BROADCAST_RUNNING;
+
+	/* FIXME: adjust it according to the min. speed of all known peers? */
+	dev->broadcast_xmt_max_payload = IEEE1394_MAX_PAYLOAD_S100
+			- IEEE1394_GASP_HDR_SIZE - RFC2374_UNFRAG_HDR_SIZE;
+	dev->broadcast_state = FWNET_BROADCAST_RUNNING;
+
 	return 0;
 
  failed_rcv_queue:
-	kfree ( priv->broadcast_rcv_buffer_ptrs );
-	priv->broadcast_rcv_buffer_ptrs = NULL;
+	kfree(dev->broadcast_rcv_buffer_ptrs);
+	dev->broadcast_rcv_buffer_ptrs = NULL;
  failed_ptrs_alloc:
-	fw_iso_buffer_destroy ( &priv->broadcast_rcv_buffer, priv->card );
+	fw_iso_buffer_destroy(&dev->broadcast_rcv_buffer, dev->card);
  failed_buffer_init:
-	fw_iso_context_destroy ( context );
-	priv->broadcast_rcv_context = NULL;
+	fw_iso_context_destroy(context);
+	dev->broadcast_rcv_context = NULL;
  failed_context_create:
-	fw_core_remove_address_handler ( &priv->handler );
+	fw_core_remove_address_handler(&dev->handler);
  failed_initial:
-	priv->local_fifo = INVALID_FIFO_ADDR;
+	dev->local_fifo = FWNET_NO_FIFO_ADDR;
+
 	return retval;
 }
 
-/* This is called after an "ifup" */
-static int ipv4_open(struct net_device *dev) {
-	struct ipv4_priv *priv;
+/* ifup */
+static int fwnet_open(struct net_device *net)
+{
+	struct fwnet_device *dev = netdev_priv(net);
 	int ret;
 
-	priv = netdev_priv(dev);
-	if (priv->broadcast_state == IPV4_BROADCAST_ERROR) {
-		ret = ipv4_broadcast_start ( priv );
+	if (dev->broadcast_state == FWNET_BROADCAST_ERROR) {
+		ret = fwnet_broadcast_start(dev);
 		if (ret)
 			return ret;
 	}
-	netif_start_queue(dev);
+	netif_start_queue(net);
+
 	return 0;
 }
 
-/* This is called after an "ifdown" */
-static int ipv4_stop(struct net_device *netdev)
+/* ifdown */
+static int fwnet_stop(struct net_device *net)
 {
-	/* flush priv->wake */
-	/* flush_scheduled_work(); */
+	netif_stop_queue(net);
+
+	/* Deallocate iso context for use by other applications? */
 
-	netif_stop_queue(netdev);
 	return 0;
 }
 
-/* Transmit a packet (called by kernel) */
-static int ipv4_tx(struct sk_buff *skb, struct net_device *netdev)
+static int fwnet_tx(struct sk_buff *skb, struct net_device *net)
 {
-	struct ipv4_ether_hdr hdr_buf;
-	struct ipv4_priv *priv = netdev_priv(netdev);
+	struct fwnet_header hdr_buf;
+	struct fwnet_device *dev = netdev_priv(net);
 	__be16 proto;
 	u16 dest_node;
-	enum ipv4_tx_type tx_type;
 	unsigned max_payload;
 	u16 dg_size;
 	u16 *datagram_label_ptr;
-	struct ipv4_packet_task *ptask;
-	struct ipv4_node *node = NULL;
+	struct fwnet_packet_task *ptask;
+	struct fwnet_peer *peer = NULL;
 
-	ptask = kmem_cache_alloc(ipv4_packet_task_cache, GFP_ATOMIC);
+	ptask = kmem_cache_alloc(fwnet_packet_task_cache, GFP_ATOMIC);
 	if (ptask == NULL)
 		goto fail;
 
@@ -1412,7 +1305,7 @@ static int ipv4_tx(struct sk_buff *skb, struct net_device *netdev)
 		goto fail;
 
 	/*
-	 * Get rid of the fake ipv4 header, but first make a copy.
+	 * Make a copy of the driver-specific header.
 	 * We might need to rebuild the header on tx failure.
 	 */
 	memcpy(&hdr_buf, skb->data, sizeof(hdr_buf));
@@ -1425,110 +1318,95 @@ static int ipv4_tx(struct sk_buff *skb, struct net_device *netdev)
 	 * Set the transmission type for the packet.  ARP packets and IP
 	 * broadcast packets are sent via GASP.
 	 */
-	if (   memcmp(hdr_buf.h_dest, netdev->broadcast, IPV4_ALEN) == 0
+	if (memcmp(hdr_buf.h_dest, net->broadcast, FWNET_ALEN) == 0
 	    || proto == htons(ETH_P_ARP)
-	    || (   proto == htons(ETH_P_IP)
-		&& IN_MULTICAST(ntohl(ip_hdr(skb)->daddr)) ) ) {
-		/* fw_debug ( "transmitting arp or multicast packet\n" );*/
-		tx_type = IPV4_GASP;
-		dest_node = ALL_NODES;
-		max_payload = priv->broadcast_xmt_max_payload;
-		/* BUG_ON(max_payload < S100_BUFFER_SIZE - IPV4_GASP_OVERHEAD); */
-		datagram_label_ptr = &priv->broadcast_xmt_datagramlabel;
-		ptask->fifo_addr = INVALID_FIFO_ADDR;
-		ptask->generation = 0U;
-		ptask->dest_node = 0U;
-		ptask->speed = 0;
+	    || (proto == htons(ETH_P_IP)
+		&& IN_MULTICAST(ntohl(ip_hdr(skb)->daddr)))) {
+		max_payload = dev->broadcast_xmt_max_payload;
+		datagram_label_ptr = &dev->broadcast_xmt_datagramlabel;
+
+		ptask->fifo_addr = FWNET_NO_FIFO_ADDR;
+		ptask->generation = 0;
+		ptask->dest_node = IEEE1394_ALL_NODES;
+		ptask->speed = SCODE_100;
 	} else {
-		__be64 guid = get_unaligned((u64 *)hdr_buf.h_dest);
+		__be64 guid = get_unaligned((__be64 *)hdr_buf.h_dest);
 		u8 generation;
 
-		node = ipv4_node_find_by_guid(priv, be64_to_cpu(guid));
-		if (!node) {
-			fw_debug ( "Normal packet but no node\n" );
+		peer = fwnet_peer_find_by_guid(dev, be64_to_cpu(guid));
+		if (!peer)
 			goto fail;
-		}
 
-		if (node->fifo == INVALID_FIFO_ADDR) {
-			fw_debug ( "Normal packet but no fifo addr\n" );
+		if (peer->fifo == FWNET_NO_FIFO_ADDR)
 			goto fail;
-		}
 
-		/* fw_debug ( "Transmitting normal packet to %x at %llxx\n", node->nodeid, node->fifo ); */
-		generation = node->generation;
-		dest_node = node->nodeid;
-		max_payload = node->max_payload;
-		/* BUG_ON(max_payload < S100_BUFFER_SIZE - IPV4_FRAG_HDR_SIZE); */
+		generation = peer->generation;
+		smp_rmb();
+		dest_node = peer->node_id;
+
+		max_payload = peer->max_payload;
+		datagram_label_ptr = &peer->datagram_label;
 
-		datagram_label_ptr = &node->datagram_label;
-		tx_type = IPV4_WRREQ;
-		ptask->fifo_addr = node->fifo;
+		ptask->fifo_addr = peer->fifo;
 		ptask->generation = generation;
 		ptask->dest_node = dest_node;
-		ptask->speed = node->xmt_speed;
+		ptask->speed = peer->xmt_speed;
 	}
 
 	/* If this is an ARP packet, convert it */
 	if (proto == htons(ETH_P_ARP)) {
-		/* Convert a standard ARP packet to 1394 ARP. The first 8 bytes (the entire
-		 * arphdr) is the same format as the ip1394 header, so they overlap.  The rest
-		 * needs to be munged a bit.  The remainder of the arphdr is formatted based
-		 * on hwaddr len and ipaddr len.  We know what they'll be, so it's easy to
-		 * judge.
-		 *
-		 * Now that the EUI is used for the hardware address all we need to do to make
-		 * this work for 1394 is to insert 2 quadlets that contain max_rec size,
-		 * speed, and unicast FIFO address information between the sender_unique_id
-		 * and the IP addresses.
-		 */
 		struct arphdr *arp = (struct arphdr *)skb->data;
 		unsigned char *arp_ptr = (unsigned char *)(arp + 1);
-		struct ipv4_arp *arp1394 = (struct ipv4_arp *)skb->data;
-		u32 ipaddr;
-
-		ipaddr = *(u32*)(arp_ptr + IPV4_ALEN);
-		arp1394->hw_addr_len    = 16;
-		arp1394->max_rec        = priv->card->max_receive;
-		arp1394->sspd		= priv->card->link_speed;
-		arp1394->fifo_hi	= htons(priv->local_fifo >> 32);
-		arp1394->fifo_lo        = htonl(priv->local_fifo & 0xFFFFFFFF);
-		arp1394->sip		= ipaddr;
+		struct rfc2734_arp *arp1394 = (struct rfc2734_arp *)skb->data;
+		__be32 ipaddr;
+
+		ipaddr = get_unaligned((__be32 *)(arp_ptr + FWNET_ALEN));
+
+		arp1394->hw_addr_len    = RFC2734_HW_ADDR_LEN;
+		arp1394->max_rec        = dev->card->max_receive;
+		arp1394->sspd		= dev->card->link_speed;
+
+		put_unaligned_be16(dev->local_fifo >> 32,
+				   &arp1394->fifo_hi);
+		put_unaligned_be32(dev->local_fifo & 0xffffffff,
+				   &arp1394->fifo_lo);
+		put_unaligned(ipaddr, &arp1394->sip);
 	}
-	if ( ipv4_max_xmt && max_payload > ipv4_max_xmt )
-		max_payload = ipv4_max_xmt;
 
 	ptask->hdr.w0 = 0;
 	ptask->hdr.w1 = 0;
 	ptask->skb = skb;
-	ptask->priv = priv;
-        ptask->tx_type = tx_type;
+	ptask->dev = dev;
+
 	/* Does it all fit in one packet? */
-	if ( dg_size <= max_payload ) {
-		ipv4_make_uf_hdr(&ptask->hdr, be16_to_cpu(proto));
+	if (dg_size <= max_payload) {
+		fwnet_make_uf_hdr(&ptask->hdr, ntohs(proto));
 		ptask->outstanding_pkts = 1;
-		max_payload = dg_size + IPV4_UNFRAG_HDR_SIZE;
+		max_payload = dg_size + RFC2374_UNFRAG_HDR_SIZE;
 	} else {
 		u16 datagram_label;
 
-		max_payload -= IPV4_FRAG_OVERHEAD;
+		max_payload -= RFC2374_FRAG_OVERHEAD;
 		datagram_label = (*datagram_label_ptr)++;
-		ipv4_make_ff_hdr(&ptask->hdr, be16_to_cpu(proto), dg_size, datagram_label );
+		fwnet_make_ff_hdr(&ptask->hdr, ntohs(proto), dg_size,
+				  datagram_label);
 		ptask->outstanding_pkts = DIV_ROUND_UP(dg_size, max_payload);
-		max_payload += IPV4_FRAG_HDR_SIZE;
+		max_payload += RFC2374_FRAG_HDR_SIZE;
 	}
 	ptask->max_payload = max_payload;
-	ipv4_send_packet ( ptask );
+	fwnet_send_packet(ptask);
+
 	return NETDEV_TX_OK;
 
  fail:
 	if (ptask)
-		kmem_cache_free(ipv4_packet_task_cache, ptask);
+		kmem_cache_free(fwnet_packet_task_cache, ptask);
 
 	if (skb != NULL)
 		dev_kfree_skb(skb);
 
-	netdev->stats.tx_dropped++;
-	netdev->stats.tx_errors++;
+	net->stats.tx_dropped++;
+	net->stats.tx_errors++;
 
 	/*
 	 * FIXME: According to a patch from 2003-02-26, "returning non-zero
@@ -1540,280 +1418,291 @@ static int ipv4_tx(struct sk_buff *skb, struct net_device *netdev)
 	return NETDEV_TX_OK;
 }
 
-/*
- * FIXME: What to do if we timeout? I think a host reset is probably in order,
- * so that's what we do. Should we increment the stat counters too?
- */
-static void ipv4_tx_timeout(struct net_device *dev) {
-	struct ipv4_priv *priv;
+static void fwnet_tx_timeout(struct net_device *net)
+{
+	fw_error("%s: timeout\n", net->name);
 
-	priv = netdev_priv(dev);
-	fw_error ( "%s: Timeout, resetting host\n", dev->name );
-#if 0 /* stefanr */
-	fw_core_initiate_bus_reset ( priv->card, 1 );
-#endif
+	/* FIXME: What to do if we timeout? */
 }
 
-static int ipv4_change_mtu ( struct net_device *dev, int new_mtu ) {
-#if 0
-	int max_mtu;
-	struct ipv4_priv *priv;
-#endif
-
+static int fwnet_change_mtu(struct net_device *net, int new_mtu)
+{
 	if (new_mtu < 68)
 		return -EINVAL;
 
-#if 0
-	priv = netdev_priv(dev);
-	/* This is not actually true because we can fragment packets at the firewire layer */
-	max_mtu = (1 << (priv->card->max_receive + 1))
-		                - sizeof(struct ipv4_hdr) - IPV4_GASP_OVERHEAD;
-	if (new_mtu > max_mtu) {
-		fw_notify ( "%s: Local node constrains MTU to %d\n", dev->name, max_mtu);
-		return -ERANGE;
-	}
-#endif
-	dev->mtu = new_mtu;
+	net->mtu = new_mtu;
 	return 0;
 }
 
-static void ipv4_get_drvinfo(struct net_device *dev,
-struct ethtool_drvinfo *info) {
-	strcpy(info->driver, ipv4_driver_name);
-	strcpy(info->bus_info, "ieee1394"); /* FIXME provide more detail? */
+static void fwnet_get_drvinfo(struct net_device *net,
+			      struct ethtool_drvinfo *info)
+{
+	strcpy(info->driver, KBUILD_MODNAME);
+	strcpy(info->bus_info, "ieee1394");
 }
 
-static struct ethtool_ops ipv4_ethtool_ops = {
-	.get_drvinfo = ipv4_get_drvinfo,
+static struct ethtool_ops fwnet_ethtool_ops = {
+	.get_drvinfo = fwnet_get_drvinfo,
 };
 
-static const struct net_device_ops ipv4_netdev_ops = {
-	.ndo_open       = ipv4_open,
-	.ndo_stop	= ipv4_stop,
-	.ndo_start_xmit = ipv4_tx,
-	.ndo_tx_timeout = ipv4_tx_timeout,
-	.ndo_change_mtu = ipv4_change_mtu,
+static const struct net_device_ops fwnet_netdev_ops = {
+	.ndo_open       = fwnet_open,
+	.ndo_stop	= fwnet_stop,
+	.ndo_start_xmit = fwnet_tx,
+	.ndo_tx_timeout = fwnet_tx_timeout,
+	.ndo_change_mtu = fwnet_change_mtu,
 };
 
-static void ipv4_init_dev ( struct net_device *dev ) {
-	dev->header_ops		= &ipv4_header_ops;
-	dev->netdev_ops         = &ipv4_netdev_ops;
-	SET_ETHTOOL_OPS(dev, &ipv4_ethtool_ops);
-
-	dev->watchdog_timeo	= IPV4_TIMEOUT;
-	dev->flags		= IFF_BROADCAST | IFF_MULTICAST;
-	dev->features		= NETIF_F_HIGHDMA;
-	dev->addr_len		= IPV4_ALEN;
-	dev->hard_header_len	= IPV4_HLEN;
-	dev->type		= ARPHRD_IEEE1394;
-
-	/* FIXME: This value was copied from ether_setup(). Is it too much? */
-	dev->tx_queue_len	= 1000;
+static void fwnet_init_dev(struct net_device *net)
+{
+	net->header_ops		= &fwnet_header_ops;
+	net->netdev_ops		= &fwnet_netdev_ops;
+	net->watchdog_timeo	= 100000; /* ? FIXME */
+	net->flags		= IFF_BROADCAST | IFF_MULTICAST;
+	net->features		= NETIF_F_HIGHDMA;
+	net->addr_len		= FWNET_ALEN;
+	net->hard_header_len	= FWNET_HLEN;
+	net->type		= ARPHRD_IEEE1394;
+	net->tx_queue_len	= 1000; /* ? FIXME */
+	SET_ETHTOOL_OPS(net, &fwnet_ethtool_ops);
 }
 
-static int ipv4_probe ( struct device *dev ) {
-	struct fw_unit * unit;
-	struct fw_device *device;
-	struct fw_card *card;
-	struct net_device *netdev;
-	struct ipv4_priv *priv;
+/* FIXME create netdev upon first fw_unit of a card, not upon local fw_unit */
+static int fwnet_probe(struct device *_dev)
+{
+	struct fw_unit *unit = fw_unit(_dev);
+	struct fw_device *device = fw_parent_device(unit);
+	struct fw_card *card = device->card;
+	struct net_device *net;
+	struct fwnet_device *dev;
 	unsigned max_mtu;
-	__be64 guid;
-
-	fw_debug("ipv4 Probing\n" );
-	unit = fw_unit ( dev );
-	device = fw_device ( unit->device.parent );
-	card = device->card;
 
-	if ( ! device->is_local ) {
+	if (!device->is_local) {
 		int added;
 
-		fw_debug ( "Non-local, adding remote node entry\n" );
-		added = ipv4_node_new ( card, device );
+		added = fwnet_peer_new(card, device);
 		return added;
 	}
-	fw_debug("ipv4 Local: adding netdev\n" );
-	netdev = alloc_netdev ( sizeof(*priv), "firewire%d", ipv4_init_dev );
-	if ( netdev == NULL) {
-		fw_error( "Out of memory\n");
+	net = alloc_netdev(sizeof(*dev), "firewire%d", fwnet_init_dev);
+	if (net == NULL) {
+		fw_error("out of memory\n");
 		goto out;
 	}
 
-	SET_NETDEV_DEV(netdev, card->device);
-	priv = netdev_priv(netdev);
+	SET_NETDEV_DEV(net, card->device);
+	dev = netdev_priv(net);
 
-	spin_lock_init(&priv->lock);
-	priv->broadcast_state = IPV4_BROADCAST_ERROR;
-	priv->broadcast_rcv_context = NULL;
-	priv->broadcast_xmt_max_payload = 0;
-	priv->broadcast_xmt_datagramlabel = 0;
+	spin_lock_init(&dev->lock);
+	dev->broadcast_state = FWNET_BROADCAST_ERROR;
+	dev->broadcast_rcv_context = NULL;
+	dev->broadcast_xmt_max_payload = 0;
+	dev->broadcast_xmt_datagramlabel = 0;
 
-	priv->local_fifo = INVALID_FIFO_ADDR;
+	dev->local_fifo = FWNET_NO_FIFO_ADDR;
 
-	/* INIT_WORK(&priv->wake, ipv4_handle_queue);*/
-	INIT_LIST_HEAD(&priv->packet_list);
-	INIT_LIST_HEAD(&priv->broadcasted_list);
-	INIT_LIST_HEAD(&priv->sent_list );
+	/* INIT_WORK(&dev->wake, fwnet_handle_queue);*/
+	INIT_LIST_HEAD(&dev->packet_list);
+	INIT_LIST_HEAD(&dev->broadcasted_list);
+	INIT_LIST_HEAD(&dev->sent_list);
 
-	priv->card = card;
+	dev->card = card;
 
 	/*
 	 * Use the RFC 2734 default 1500 octets or the maximum payload
 	 * as initial MTU
 	 */
 	max_mtu = (1 << (card->max_receive + 1))
-		  - sizeof(struct ipv4_hdr) - IPV4_GASP_OVERHEAD;
-	netdev->mtu = min(1500U, max_mtu);
+		  - sizeof(struct rfc2734_header) - IEEE1394_GASP_HDR_SIZE;
+	net->mtu = min(1500U, max_mtu);
 
 	/* Set our hardware address while we're at it */
-	guid = cpu_to_be64(card->guid);
-	memcpy(netdev->dev_addr, &guid, sizeof(u64));
-	memset(netdev->broadcast, 0xff, sizeof(u64));
-	if ( register_netdev ( netdev ) ) {
-		fw_error ( "Cannot register the driver\n");
+	put_unaligned_be64(card->guid, net->dev_addr);
+	put_unaligned_be64(~0ULL, net->broadcast);
+	if (register_netdev(net)) {
+		fw_error("Cannot register the driver\n");
 		goto out;
 	}
 
-	fw_notify ( "%s: IPv4 over Firewire on device %016llx\n",
- netdev->name, card->guid );
-	card->netdev = netdev;
+	fw_notify("%s: IPv4 over FireWire on device %016llx\n",
+		  net->name, (unsigned long long)card->guid);
+	card->netdev = net;
 
-	return 0 /* ipv4_new_node ( ud ) */;
+	return 0;
  out:
-	if ( netdev )
-		free_netdev ( netdev );
+	if (net)
+		free_netdev(net);
+
 	return -ENOENT;
 }
 
+static int fwnet_remove(struct device *_dev)
+{
+	struct fw_unit *unit = fw_unit(_dev);
+	struct fw_device *device = fw_parent_device(unit);
+	struct fw_card *card = device->card;
+	struct net_device *net;
+	struct fwnet_device *dev;
+	struct fwnet_peer *peer;
+	struct fwnet_partial_datagram *pd, *pd_next;
+	struct fwnet_packet_task *ptask, *pt_next;
+
+	if (!device->is_local) {
+		fwnet_peer_delete(card, device);
 
-static int ipv4_remove ( struct device *dev ) {
-	struct fw_unit * unit;
-	struct fw_device *device;
-	struct fw_card *card;
-	struct net_device *netdev;
-	struct ipv4_priv *priv;
-	struct ipv4_node *node;
-	struct ipv4_partial_datagram *pd, *pd_next;
-	struct ipv4_packet_task *ptask, *pt_next;
-
-	fw_debug("ipv4 Removing\n" );
-	unit = fw_unit ( dev );
-	device = fw_device ( unit->device.parent );
-	card = device->card;
-
-	if ( ! device->is_local ) {
-		fw_debug ( "Node %x is non-local, removing remote node entry\n", device->node_id );
-		ipv4_node_delete ( card, device );
 		return 0;
 	}
-	netdev = card->netdev;
-	if ( netdev ) {
-		fw_debug ( "Node %x is local: deleting netdev\n", device->node_id );
-		priv = netdev_priv ( netdev );
-		unregister_netdev ( netdev );
-		fw_debug ( "unregistered\n" );
-		if ( priv->local_fifo != INVALID_FIFO_ADDR )
-			fw_core_remove_address_handler ( &priv->handler );
-		fw_debug ( "address handler gone\n" );
-		if ( priv->broadcast_rcv_context ) {
-			fw_iso_context_stop ( priv->broadcast_rcv_context );
-			fw_iso_buffer_destroy ( &priv->broadcast_rcv_buffer, priv->card );
-			fw_iso_context_destroy ( priv->broadcast_rcv_context );
-			fw_debug ( "rcv stopped\n" );
+
+	net = card->netdev;
+	if (net) {
+		dev = netdev_priv(net);
+		unregister_netdev(net);
+
+		if (dev->local_fifo != FWNET_NO_FIFO_ADDR)
+			fw_core_remove_address_handler(&dev->handler);
+		if (dev->broadcast_rcv_context) {
+			fw_iso_context_stop(dev->broadcast_rcv_context);
+			fw_iso_buffer_destroy(&dev->broadcast_rcv_buffer,
+					      dev->card);
+			fw_iso_context_destroy(dev->broadcast_rcv_context);
 		}
-		list_for_each_entry_safe( ptask, pt_next, &priv->packet_list, packet_list ) {
-			dev_kfree_skb_any ( ptask->skb );
-			kmem_cache_free( ipv4_packet_task_cache, ptask );
+		list_for_each_entry_safe(ptask, pt_next,
+					 &dev->packet_list, pt_link) {
+			dev_kfree_skb_any(ptask->skb);
+			kmem_cache_free(fwnet_packet_task_cache, ptask);
 		}
-		list_for_each_entry_safe( ptask, pt_next, &priv->broadcasted_list, packet_list ) {
-			dev_kfree_skb_any ( ptask->skb );
-			kmem_cache_free( ipv4_packet_task_cache, ptask );
+		list_for_each_entry_safe(ptask, pt_next,
+					 &dev->broadcasted_list, pt_link) {
+			dev_kfree_skb_any(ptask->skb);
+			kmem_cache_free(fwnet_packet_task_cache, ptask);
 		}
-		list_for_each_entry_safe( ptask, pt_next, &priv->sent_list, packet_list ) {
-			dev_kfree_skb_any ( ptask->skb );
-			kmem_cache_free( ipv4_packet_task_cache, ptask );
+		list_for_each_entry_safe(ptask, pt_next,
+					 &dev->sent_list, pt_link) {
+			dev_kfree_skb_any(ptask->skb);
+			kmem_cache_free(fwnet_packet_task_cache, ptask);
 		}
-		fw_debug ( "lists emptied\n" );
-		list_for_each_entry( node, &card->ipv4_nodes, ipv4_nodes ) {
-			if ( node->pdg_size ) {
-				list_for_each_entry_safe( pd, pd_next, &node->pdg_list, pdg_list )
-					ipv4_pd_delete ( pd );
-				node->pdg_size = 0;
+		list_for_each_entry(peer, &card->peer_list, peer_link) {
+			if (peer->pdg_size) {
+				list_for_each_entry_safe(pd, pd_next,
+						&peer->pd_list, pd_link)
+					fwnet_pd_delete(pd);
+				peer->pdg_size = 0;
 			}
-			node->fifo = INVALID_FIFO_ADDR;
+			peer->fifo = FWNET_NO_FIFO_ADDR;
 		}
-		fw_debug ( "nodes cleaned up\n" );
-		free_netdev ( netdev );
+		free_netdev(net);
 		card->netdev = NULL;
-		fw_debug ( "done\n" );
 	}
+
 	return 0;
 }
 
-static void ipv4_update ( struct fw_unit *unit ) {
-	struct fw_device *device;
-	struct fw_card *card;
+/*
+ * FIXME abort partially sent fragmented datagrams,
+ * discard partially received fragmented datagrams
+ */
+static void fwnet_update(struct fw_unit *unit)
+{
+	struct fw_device *device = fw_parent_device(unit);
+	struct net_device *net = device->card->netdev;
+	struct fwnet_device *dev;
+	struct fwnet_peer *peer;
+	u64 guid;
 
-	fw_debug ( "ipv4_update unit %p\n", unit );
-	device = fw_device ( unit->device.parent );
-	card = device->card;
-	if ( ! device->is_local ) {
-		struct ipv4_node *node;
-		u64 guid;
-		struct net_device *netdev;
-		struct ipv4_priv *priv;
-
-		netdev = card->netdev;
-		if ( netdev ) {
-			priv = netdev_priv ( netdev );
-			guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
-			node = ipv4_node_find_by_guid ( priv, guid );
-			if ( ! node ) {
-				fw_error ( "ipv4_update: no node for device %llx\n", guid );
-				return;
-			}
-			fw_debug ( "Non-local, updating remote node entry for guid %llx old generation %x, old nodeid %x\n", guid, node->generation, node->nodeid );
-			node->generation = device->generation;
-			rmb();
-			node->nodeid = device->node_id;
-			fw_debug ( "New generation %x, new nodeid %x\n", node->generation, node->nodeid );
-		} else
-			fw_error ( "nonlocal, but no netdev?  How can that be?\n" );
-	} else {
-		/* FIXME: What do we need to do on bus reset? */
-		fw_debug ( "Local, doing nothing\n" );
+	if (net && !device->is_local) {
+		dev = netdev_priv(net);
+		guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
+		peer = fwnet_peer_find_by_guid(dev, guid);
+		if (!peer) {
+			fw_error("fwnet_update: no peer for device %016llx\n",
+				 (unsigned long long)guid);
+			return;
+		}
+		peer->generation = device->generation;
+		rmb();
+		peer->node_id = device->node_id;
 	}
 }
 
-static struct fw_driver ipv4_driver = {
+static const struct ieee1394_device_id fwnet_id_table[] = {
+	{
+		.match_flags  = IEEE1394_MATCH_SPECIFIER_ID |
+				IEEE1394_MATCH_VERSION,
+		.specifier_id = IANA_SPECIFIER_ID,
+		.version      = RFC2734_SW_VERSION,
+	},
+	{ }
+};
+
+static struct fw_driver fwnet_driver = {
 	.driver = {
-		.owner = THIS_MODULE,
-		.name = ipv4_driver_name,
-		.bus = &fw_bus_type,
-		.probe = ipv4_probe,
-		.remove = ipv4_remove,
+		.owner  = THIS_MODULE,
+		.name   = "net",
+		.bus    = &fw_bus_type,
+		.probe  = fwnet_probe,
+		.remove = fwnet_remove,
 	},
-	.update = ipv4_update,
-	.id_table = ipv4_id_table,
+	.update   = fwnet_update,
+	.id_table = fwnet_id_table,
+};
+
+static const u32 rfc2374_unit_directory_data[] = {
+	0x00040000,	/* directory_length		*/
+	0x1200005e,	/* unit_specifier_id: IANA	*/
+	0x81000003,	/* textual descriptor offset	*/
+	0x13000001,	/* unit_sw_version: RFC 2734	*/
+	0x81000005,	/* textual descriptor offset	*/
+	0x00030000,	/* descriptor_length		*/
+	0x00000000,	/* text				*/
+	0x00000000,	/* minimal ASCII, en		*/
+	0x49414e41,	/* I A N A			*/
+	0x00030000,	/* descriptor_length		*/
+	0x00000000,	/* text				*/
+	0x00000000,	/* minimal ASCII, en		*/
+	0x49507634,	/* I P v 4			*/
+};
+
+static struct fw_descriptor rfc2374_unit_directory = {
+	.length = ARRAY_SIZE(rfc2374_unit_directory_data),
+	.key    = (CSR_DIRECTORY | CSR_UNIT) << 24,
+	.data   = rfc2374_unit_directory_data
 };
 
-static int __init ipv4_init ( void ) {
-	int added;
+static int __init fwnet_init(void)
+{
+	int err;
+
+	err = fw_core_add_descriptor(&rfc2374_unit_directory);
+	if (err)
+		return err;
 
-	added = fw_core_add_descriptor ( &ipv4_unit_directory );
-	if ( added < 0 )
-		fw_error ( "Failed to add descriptor" );
-	ipv4_packet_task_cache = kmem_cache_create("packet_task",
- sizeof(struct ipv4_packet_task), 0, 0, NULL);
-	fw_debug("Adding ipv4 module\n" );
-	return driver_register ( &ipv4_driver.driver );
+	fwnet_packet_task_cache = kmem_cache_create("packet_task",
+			sizeof(struct fwnet_packet_task), 0, 0, NULL);
+	if (!fwnet_packet_task_cache) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	err = driver_register(&fwnet_driver.driver);
+	if (!err)
+		return 0;
+
+	kmem_cache_destroy(fwnet_packet_task_cache);
+out:
+	fw_core_remove_descriptor(&rfc2374_unit_directory);
+
+	return err;
 }
+module_init(fwnet_init);
 
-static void __exit ipv4_cleanup ( void ) {
-	fw_core_remove_descriptor ( &ipv4_unit_directory );
-	fw_debug("Removing ipv4 module\n" );
-	driver_unregister ( &ipv4_driver.driver );
+static void __exit fwnet_cleanup(void)
+{
+	driver_unregister(&fwnet_driver.driver);
+	kmem_cache_destroy(fwnet_packet_task_cache);
+	fw_core_remove_descriptor(&rfc2374_unit_directory);
 }
+module_exit(fwnet_cleanup);
 
-module_init(ipv4_init);
-module_exit(ipv4_cleanup);
+MODULE_AUTHOR("Jay Fenlason <fenlason@redhat.com>");
+MODULE_DESCRIPTION("IPv4 over IEEE1394 as per RFC 2734");
+MODULE_LICENSE("GPL");
+MODULE_DEVICE_TABLE(ieee1394, fwnet_id_table);
diff --git a/include/linux/firewire.h b/include/linux/firewire.h
index d44f47d3b2d9..5cb0c1549ff1 100644
--- a/include/linux/firewire.h
+++ b/include/linux/firewire.h
@@ -131,13 +131,10 @@ struct fw_card {
 	bool broadcast_channel_allocated;
 	u32 broadcast_channel;
 	u32 topology_map[(CSR_TOPOLOGY_MAP_END - CSR_TOPOLOGY_MAP) / 4];
-	/* Only non-NULL if firewire-ipv4 is active on this card. */
+
+	/* firewire-net driver data */
 	void *netdev;
-	/*
-	 * The nodes get probed before the card, so we need a place to store
-	 * them independent of card->netdev
-	 */
-	struct list_head ipv4_nodes;
+	struct list_head peer_list;
 };
 
 static inline struct fw_card *fw_card_get(struct fw_card *card)
-- 
cgit v1.2.3-59-g8ed1b


From 5a124d382ea5c97be43c779e4f481455e0287654 Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Sun, 14 Jun 2009 11:45:27 +0200
Subject: firewire: net: allow for unordered unit discovery

Decouple the creation and destruction of the net_device from the order
of discovery and removal of nodes with RFC 2734 unit directories since
there is no reliable order.  The net_device is now created when the
first RFC 2734 unit on a card is discovered, and destroyed when the last
RFC 2734 unit on a card went away.  This includes all remote units as
well as the local unit, which is therefore tracked as a peer now too.

Also, locking around the list of peers is slightly extended to guard
against peer removal.  As a side effect, fwnet_peer.pdg_lock has become
superfluous and is deleted.

Peer data (max_rec, speed, node ID, generation) are updated more
carefully.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/core-card.c |   2 -
 drivers/firewire/net.c       | 454 ++++++++++++++++++++-----------------------
 include/linux/firewire.h     |   4 -
 3 files changed, 207 insertions(+), 253 deletions(-)

diff --git a/drivers/firewire/core-card.c b/drivers/firewire/core-card.c
index 8c45e43da7c5..667603ac14b1 100644
--- a/drivers/firewire/core-card.c
+++ b/drivers/firewire/core-card.c
@@ -429,8 +429,6 @@ void fw_card_initialize(struct fw_card *card,
 	card->local_node = NULL;
 
 	INIT_DELAYED_WORK(&card->work, fw_card_bm_work);
-	card->netdev = NULL;
-	INIT_LIST_HEAD(&card->peer_list);
 }
 EXPORT_SYMBOL(fw_card_initialize);
 
diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c
index ba6f924b1b13..d83c54587a63 100644
--- a/drivers/firewire/net.c
+++ b/drivers/firewire/net.c
@@ -18,8 +18,10 @@
 #include <linux/mod_devicetable.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
+#include <linux/mutex.h>
 #include <linux/netdevice.h>
 #include <linux/skbuff.h>
+#include <linux/spinlock.h>
 
 #include <asm/unaligned.h>
 #include <net/arp.h>
@@ -135,40 +137,11 @@ struct fwnet_partial_datagram {
 	u16 datagram_size;
 };
 
-/*
- * We keep one of these for each IPv4 capable device attached to a fw_card.
- * The list of them is stored in the fw_card structure rather than in the
- * fwnet_device because the remote IPv4 nodes may be probed before the card is,
- * so we need a place to store them before the fwnet_device structure is
- * allocated.
- */
-struct fwnet_peer {
-	struct list_head peer_link;
-	/* guid of the remote peer */
-	u64 guid;
-	/* FIFO address to transmit datagrams to, or FWNET_NO_FIFO_ADDR */
-	u64 fifo;
-
-	spinlock_t pdg_lock;	/* partial datagram lock		*/
-	/* List of partial datagrams received from this peer */
-	struct list_head pd_list;
-	/* Number of entries in pd_list at the moment */
-	unsigned pdg_size;
-
-	/* max payload to transmit to this remote peer */
-	/* This already includes the RFC2374_FRAG_HDR_SIZE overhead */
-	u16 max_payload;
-	/* outgoing datagram label */
-	u16 datagram_label;
-	/* Current node_id of the remote peer */
-	u16 node_id;
-	/* current generation of the remote peer */
-	u8 generation;
-	/* max speed that this peer can receive at */
-	u8 xmt_speed;
-};
+static DEFINE_MUTEX(fwnet_device_mutex);
+static LIST_HEAD(fwnet_device_list);
 
 struct fwnet_device {
+	struct list_head dev_link;
 	spinlock_t lock;
 	enum {
 		FWNET_BROADCAST_ERROR,
@@ -206,7 +179,26 @@ struct fwnet_device {
 	/* List of packets that have been sent but not yet acked */
 	struct list_head sent_list;
 
+	struct list_head peer_list;
 	struct fw_card *card;
+	struct net_device *netdev;
+};
+
+struct fwnet_peer {
+	struct list_head peer_link;
+	struct fwnet_device *dev;
+	u64 guid;
+	u64 fifo;
+
+	/* guarded by dev->lock */
+	struct list_head pd_list; /* received partial datagrams */
+	unsigned pdg_size;        /* pd_list size */
+
+	u16 datagram_label;       /* outgoing datagram label */
+	unsigned max_payload;     /* includes RFC2374_FRAG_HDR_SIZE overhead */
+	int node_id;
+	int generation;
+	unsigned speed;
 };
 
 /* This is our task struct. It's used for the packet complete callback.  */
@@ -479,102 +471,47 @@ static bool fwnet_pd_is_complete(struct fwnet_partial_datagram *pd)
 	return fi->len == pd->datagram_size;
 }
 
-static int fwnet_peer_new(struct fw_card *card, struct fw_device *device)
-{
-	struct fwnet_peer *peer;
-
-	peer = kmalloc(sizeof(*peer), GFP_KERNEL);
-	if (!peer) {
-		fw_error("out of memory\n");
-
-		return -ENOMEM;
-	}
-	peer->guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
-	peer->fifo = FWNET_NO_FIFO_ADDR;
-	INIT_LIST_HEAD(&peer->pd_list);
-	spin_lock_init(&peer->pdg_lock);
-	peer->pdg_size = 0;
-	peer->generation = device->generation;
-	rmb();
-	peer->node_id = device->node_id;
-	 /* FIXME what should it really be? */
-	peer->max_payload = IEEE1394_MAX_PAYLOAD_S100 - RFC2374_UNFRAG_HDR_SIZE;
-	peer->datagram_label = 0U;
-	peer->xmt_speed = device->max_speed;
-	list_add_tail(&peer->peer_link, &card->peer_list);
-
-	return 0;
-}
-
-/* FIXME caller must take the lock, or peer needs to be reference-counted */
+/* caller must hold dev->lock */
 static struct fwnet_peer *fwnet_peer_find_by_guid(struct fwnet_device *dev,
 						  u64 guid)
 {
-	struct fwnet_peer *p, *peer = NULL;
-	unsigned long flags;
+	struct fwnet_peer *peer;
 
-	spin_lock_irqsave(&dev->lock, flags);
-	list_for_each_entry(p, &dev->card->peer_list, peer_link)
-		if (p->guid == guid) {
-			peer = p;
-			break;
-		}
-	spin_unlock_irqrestore(&dev->lock, flags);
+	list_for_each_entry(peer, &dev->peer_list, peer_link)
+		if (peer->guid == guid)
+			return peer;
 
-	return peer;
+	return NULL;
 }
 
-/* FIXME caller must take the lock, or peer needs to be reference-counted */
-/* FIXME node_id doesn't mean anything without generation */
+/* caller must hold dev->lock */
 static struct fwnet_peer *fwnet_peer_find_by_node_id(struct fwnet_device *dev,
-						     u16 node_id)
+						int node_id, int generation)
 {
-	struct fwnet_peer *p, *peer = NULL;
-	unsigned long flags;
+	struct fwnet_peer *peer;
 
-	spin_lock_irqsave(&dev->lock, flags);
-	list_for_each_entry(p, &dev->card->peer_list, peer_link)
-		if (p->node_id == node_id) {
-			peer = p;
-			break;
-		}
-	spin_unlock_irqrestore(&dev->lock, flags);
+	list_for_each_entry(peer, &dev->peer_list, peer_link)
+		if (peer->node_id    == node_id &&
+		    peer->generation == generation)
+			return peer;
 
-	return peer;
+	return NULL;
 }
 
-/* FIXME */
-static void fwnet_peer_delete(struct fw_card *card, struct fw_device *device)
+/* See IEEE 1394-2008 table 6-4, table 8-8, table 16-18. */
+static unsigned fwnet_max_payload(unsigned max_rec, unsigned speed)
 {
-	struct net_device *net;
-	struct fwnet_device *dev;
-	struct fwnet_peer *peer;
-	u64 guid;
-	unsigned long flags;
-	struct fwnet_partial_datagram *pd, *pd_next;
-
-	guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
-	net = card->netdev;
-	if (net)
-		dev = netdev_priv(net);
-	else
-		dev = NULL;
-	if (dev)
-		spin_lock_irqsave(&dev->lock, flags);
-
-	list_for_each_entry(peer, &card->peer_list, peer_link) {
-		if (peer->guid == guid) {
-			list_del(&peer->peer_link);
-			list_for_each_entry_safe(pd, pd_next, &peer->pd_list,
-						 pd_link)
-				fwnet_pd_delete(pd);
-			break;
-		}
+	max_rec = min(max_rec, speed + 8);
+	max_rec = min(max_rec, 0xbU); /* <= 4096 */
+	if (max_rec < 8) {
+		fw_notify("max_rec %x out of range\n", max_rec);
+		max_rec = 8;
 	}
-	if (dev)
-		spin_unlock_irqrestore(&dev->lock, flags);
+
+	return (1 << (max_rec + 1)) - RFC2374_FRAG_HDR_SIZE;
 }
 
+
 static int fwnet_finish_incoming_packet(struct net_device *net,
 					struct sk_buff *skb, u16 source_node_id,
 					bool is_broadcast, u16 ether_type)
@@ -606,71 +543,44 @@ static int fwnet_finish_incoming_packet(struct net_device *net,
 		unsigned char *arp_ptr;
 		u64 fifo_addr;
 		u64 peer_guid;
-		u8 max_rec;
-		u8 sspd;
+		unsigned sspd;
 		u16 max_payload;
 		struct fwnet_peer *peer;
-		static const u16 fwnet_speed_to_max_payload[] = {
-			/* S100, S200, S400, S800, S1600, S3200 */
-			    512, 1024, 2048, 4096,  4096,  4096
-		};
+		unsigned long flags;
+
+		arp1394   = (struct rfc2734_arp *)skb->data;
+		arp       = (struct arphdr *)skb->data;
+		arp_ptr   = (unsigned char *)(arp + 1);
+		peer_guid = get_unaligned_be64(&arp1394->s_uniq_id);
+		fifo_addr = (u64)get_unaligned_be16(&arp1394->fifo_hi) << 32
+				| get_unaligned_be32(&arp1394->fifo_lo);
 
-		arp1394 = (struct rfc2734_arp *)skb->data;
-		arp = (struct arphdr *)skb->data;
-		arp_ptr = (unsigned char *)(arp + 1);
-		fifo_addr = (u64)ntohs(arp1394->fifo_hi) << 32
-				| ntohl(arp1394->fifo_lo);
-		max_rec = dev->card->max_receive;
-		if (arp1394->max_rec < max_rec)
-			max_rec = arp1394->max_rec;
 		sspd = arp1394->sspd;
 		/* Sanity check.  OS X 10.3 PPC reportedly sends 131. */
 		if (sspd > SCODE_3200) {
 			fw_notify("sspd %x out of range\n", sspd);
-			sspd = 0;
+			sspd = SCODE_3200;
 		}
+		max_payload = fwnet_max_payload(arp1394->max_rec, sspd);
 
-		max_payload = min(fwnet_speed_to_max_payload[sspd],
-			(u16)(1 << (max_rec + 1))) - RFC2374_UNFRAG_HDR_SIZE;
-
-		peer_guid = get_unaligned_be64(&arp1394->s_uniq_id);
+		spin_lock_irqsave(&dev->lock, flags);
 		peer = fwnet_peer_find_by_guid(dev, peer_guid);
+		if (peer) {
+			peer->fifo = fifo_addr;
+
+			if (peer->speed > sspd)
+				peer->speed = sspd;
+			if (peer->max_payload > max_payload)
+				peer->max_payload = max_payload;
+		}
+		spin_unlock_irqrestore(&dev->lock, flags);
+
 		if (!peer) {
 			fw_notify("No peer for ARP packet from %016llx\n",
 				  (unsigned long long)peer_guid);
 			goto failed_proto;
 		}
 
-		/* FIXME don't use card->generation */
-		if (peer->node_id != source_node_id ||
-		    peer->generation != dev->card->generation) {
-			fw_notify("Internal error: peer->node_id (%x) != "
-				  "source_node_id (%x) or peer->generation (%x)"
-				  " != dev->card->generation(%x)\n",
-				  peer->node_id, source_node_id,
-				  peer->generation, dev->card->generation);
-			peer->node_id = source_node_id;
-			peer->generation = dev->card->generation;
-		}
-
-		/* FIXME: for debugging */
-		if (sspd > SCODE_400)
-			sspd = SCODE_400;
-		/* Update our speed/payload/fifo_offset table */
-		/*
-		 * FIXME: this does not handle cases where two high-speed endpoints must use a slower speed because of
-		 * a lower speed hub between them.  We need to look at the actual topology map here.
-		 */
-		peer->fifo = fifo_addr;
-		peer->max_payload = max_payload;
-		/*
-		 * Only allow speeds to go down from their initial value.
-		 * Otherwise a local peer that can only do S400 or slower may
-		 * be told to transmit at S800 to a faster remote peer.
-		 */
-		if (peer->xmt_speed > sspd)
-			peer->xmt_speed = sspd;
-
 		/*
 		 * Now that we're done with the 1394 specific stuff, we'll
 		 * need to alter some of the data.  Believe it or not, all
@@ -764,10 +674,11 @@ static int fwnet_finish_incoming_packet(struct net_device *net,
 }
 
 static int fwnet_incoming_packet(struct fwnet_device *dev, __be32 *buf, int len,
-				 u16 source_node_id, bool is_broadcast)
+				 int source_node_id, int generation,
+				 bool is_broadcast)
 {
 	struct sk_buff *skb;
-	struct net_device *net;
+	struct net_device *net = dev->netdev;
 	struct rfc2734_header hdr;
 	unsigned lf;
 	unsigned long flags;
@@ -779,8 +690,6 @@ static int fwnet_incoming_packet(struct fwnet_device *dev, __be32 *buf, int len,
 	int retval;
 	u16 ether_type;
 
-	net = dev->card->netdev;
-
 	hdr.w0 = be32_to_cpu(buf[0]);
 	lf = fwnet_get_hdr_lf(&hdr);
 	if (lf == RFC2374_HDR_UNFRAG) {
@@ -819,9 +728,12 @@ static int fwnet_incoming_packet(struct fwnet_device *dev, __be32 *buf, int len,
 	}
 	datagram_label = fwnet_get_hdr_dgl(&hdr);
 	dg_size = fwnet_get_hdr_dg_size(&hdr); /* ??? + 1 */
-	peer = fwnet_peer_find_by_node_id(dev, source_node_id);
 
-	spin_lock_irqsave(&peer->pdg_lock, flags);
+	spin_lock_irqsave(&dev->lock, flags);
+
+	peer = fwnet_peer_find_by_node_id(dev, source_node_id, generation);
+	if (!peer)
+		goto bad_proto;
 
 	pd = fwnet_pd_find(peer, datagram_label);
 	if (pd == NULL) {
@@ -876,7 +788,7 @@ static int fwnet_incoming_packet(struct fwnet_device *dev, __be32 *buf, int len,
 		skb = skb_get(pd->skb);
 		fwnet_pd_delete(pd);
 
-		spin_unlock_irqrestore(&peer->pdg_lock, flags);
+		spin_unlock_irqrestore(&dev->lock, flags);
 
 		return fwnet_finish_incoming_packet(net, skb, source_node_id,
 						    false, ether_type);
@@ -885,12 +797,12 @@ static int fwnet_incoming_packet(struct fwnet_device *dev, __be32 *buf, int len,
 	 * Datagram is not complete, we're done for the
 	 * moment.
 	 */
-	spin_unlock_irqrestore(&peer->pdg_lock, flags);
+	spin_unlock_irqrestore(&dev->lock, flags);
 
 	return 0;
 
  bad_proto:
-	spin_unlock_irqrestore(&peer->pdg_lock, flags);
+	spin_unlock_irqrestore(&dev->lock, flags);
 
 	if (netif_queue_stopped(net))
 		netif_wake_queue(net);
@@ -916,7 +828,8 @@ static void fwnet_receive_packet(struct fw_card *card, struct fw_request *r,
 		return;
 	}
 
-	status = fwnet_incoming_packet(dev, payload, length, source, false);
+	status = fwnet_incoming_packet(dev, payload, length,
+				       source, generation, false);
 	if (status != 0) {
 		fw_error("Incoming packet failure\n");
 		fw_send_response(card, r, RCODE_CONFLICT_ERROR);
@@ -966,7 +879,7 @@ static void fwnet_receive_broadcast(struct fw_iso_context *context,
 		buf_ptr += 2;
 		length -= IEEE1394_GASP_HDR_SIZE;
 		fwnet_incoming_packet(dev, buf_ptr, length,
-				      source_node_id, true);
+				      source_node_id, -1, true);
 	}
 
 	packet.payload_length = dev->rcv_buffer_size;
@@ -1073,7 +986,6 @@ static int fwnet_send_packet(struct fwnet_packet_task *ptask)
 	unsigned tx_len;
 	struct rfc2734_header *bufhdr;
 	unsigned long flags;
-	struct net_device *net;
 
 	dev = ptask->dev;
 	tx_len = ptask->max_payload;
@@ -1137,8 +1049,7 @@ static int fwnet_send_packet(struct fwnet_packet_task *ptask)
 	list_add_tail(&ptask->pt_link, &dev->sent_list);
 	spin_unlock_irqrestore(&dev->lock, flags);
 
-	net = dev->card->netdev;
-	net->trans_start = jiffies;
+	dev->netdev->trans_start = jiffies;
 
 	return 0;
 }
@@ -1294,7 +1205,8 @@ static int fwnet_tx(struct sk_buff *skb, struct net_device *net)
 	u16 dg_size;
 	u16 *datagram_label_ptr;
 	struct fwnet_packet_task *ptask;
-	struct fwnet_peer *peer = NULL;
+	struct fwnet_peer *peer;
+	unsigned long flags;
 
 	ptask = kmem_cache_alloc(fwnet_packet_task_cache, GFP_ATOMIC);
 	if (ptask == NULL)
@@ -1314,6 +1226,9 @@ static int fwnet_tx(struct sk_buff *skb, struct net_device *net)
 	proto = hdr_buf.h_proto;
 	dg_size = skb->len;
 
+	/* serialize access to peer, including peer->datagram_label */
+	spin_lock_irqsave(&dev->lock, flags);
+
 	/*
 	 * Set the transmission type for the packet.  ARP packets and IP
 	 * broadcast packets are sent via GASP.
@@ -1322,35 +1237,30 @@ static int fwnet_tx(struct sk_buff *skb, struct net_device *net)
 	    || proto == htons(ETH_P_ARP)
 	    || (proto == htons(ETH_P_IP)
 		&& IN_MULTICAST(ntohl(ip_hdr(skb)->daddr)))) {
-		max_payload = dev->broadcast_xmt_max_payload;
+		max_payload        = dev->broadcast_xmt_max_payload;
 		datagram_label_ptr = &dev->broadcast_xmt_datagramlabel;
 
-		ptask->fifo_addr = FWNET_NO_FIFO_ADDR;
-		ptask->generation = 0;
-		ptask->dest_node = IEEE1394_ALL_NODES;
-		ptask->speed = SCODE_100;
+		ptask->fifo_addr   = FWNET_NO_FIFO_ADDR;
+		ptask->generation  = 0;
+		ptask->dest_node   = IEEE1394_ALL_NODES;
+		ptask->speed       = SCODE_100;
 	} else {
 		__be64 guid = get_unaligned((__be64 *)hdr_buf.h_dest);
 		u8 generation;
 
 		peer = fwnet_peer_find_by_guid(dev, be64_to_cpu(guid));
-		if (!peer)
-			goto fail;
-
-		if (peer->fifo == FWNET_NO_FIFO_ADDR)
-			goto fail;
+		if (!peer || peer->fifo == FWNET_NO_FIFO_ADDR)
+			goto fail_unlock;
 
-		generation = peer->generation;
-		smp_rmb();
-		dest_node = peer->node_id;
-
-		max_payload = peer->max_payload;
+		generation         = peer->generation;
+		dest_node          = peer->node_id;
+		max_payload        = peer->max_payload;
 		datagram_label_ptr = &peer->datagram_label;
 
-		ptask->fifo_addr = peer->fifo;
-		ptask->generation = generation;
-		ptask->dest_node = dest_node;
-		ptask->speed = peer->xmt_speed;
+		ptask->fifo_addr   = peer->fifo;
+		ptask->generation  = generation;
+		ptask->dest_node   = dest_node;
+		ptask->speed       = peer->speed;
 	}
 
 	/* If this is an ARP packet, convert it */
@@ -1393,11 +1303,16 @@ static int fwnet_tx(struct sk_buff *skb, struct net_device *net)
 		ptask->outstanding_pkts = DIV_ROUND_UP(dg_size, max_payload);
 		max_payload += RFC2374_FRAG_HDR_SIZE;
 	}
+
+	spin_unlock_irqrestore(&dev->lock, flags);
+
 	ptask->max_payload = max_payload;
 	fwnet_send_packet(ptask);
 
 	return NETDEV_TX_OK;
 
+ fail_unlock:
+	spin_unlock_irqrestore(&dev->lock, flags);
  fail:
 	if (ptask)
 		kmem_cache_free(fwnet_packet_task_cache, ptask);
@@ -1467,7 +1382,48 @@ static void fwnet_init_dev(struct net_device *net)
 	SET_ETHTOOL_OPS(net, &fwnet_ethtool_ops);
 }
 
-/* FIXME create netdev upon first fw_unit of a card, not upon local fw_unit */
+/* caller must hold fwnet_device_mutex */
+static struct fwnet_device *fwnet_dev_find(struct fw_card *card)
+{
+	struct fwnet_device *dev;
+
+	list_for_each_entry(dev, &fwnet_device_list, dev_link)
+		if (dev->card == card)
+			return dev;
+
+	return NULL;
+}
+
+static int fwnet_add_peer(struct fwnet_device *dev,
+			  struct fw_unit *unit, struct fw_device *device)
+{
+	struct fwnet_peer *peer;
+
+	peer = kmalloc(sizeof(*peer), GFP_KERNEL);
+	if (!peer)
+		return -ENOMEM;
+
+	unit->device.driver_data = peer;
+	peer->dev = dev;
+	peer->guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
+	peer->fifo = FWNET_NO_FIFO_ADDR;
+	INIT_LIST_HEAD(&peer->pd_list);
+	peer->pdg_size = 0;
+	peer->datagram_label = 0;
+	peer->speed = device->max_speed;
+	peer->max_payload = fwnet_max_payload(device->max_rec, peer->speed);
+
+	peer->generation = device->generation;
+	smp_rmb();
+	peer->node_id = device->node_id;
+
+	spin_lock_irq(&dev->lock);
+	list_add_tail(&peer->peer_link, &dev->peer_list);
+	spin_unlock_irq(&dev->lock);
+
+	return 0;
+}
+
 static int fwnet_probe(struct device *_dev)
 {
 	struct fw_unit *unit = fw_unit(_dev);
@@ -1476,16 +1432,22 @@ static int fwnet_probe(struct device *_dev)
 	struct net_device *net;
 	struct fwnet_device *dev;
 	unsigned max_mtu;
+	bool new_netdev;
+	int ret;
 
-	if (!device->is_local) {
-		int added;
+	mutex_lock(&fwnet_device_mutex);
 
-		added = fwnet_peer_new(card, device);
-		return added;
+	dev = fwnet_dev_find(card);
+	if (dev) {
+		new_netdev = false;
+		net = dev->netdev;
+		goto have_dev;
 	}
+
+	new_netdev = true;
 	net = alloc_netdev(sizeof(*dev), "firewire%d", fwnet_init_dev);
 	if (net == NULL) {
-		fw_error("out of memory\n");
+		ret = -ENOMEM;
 		goto out;
 	}
 
@@ -1500,12 +1462,13 @@ static int fwnet_probe(struct device *_dev)
 
 	dev->local_fifo = FWNET_NO_FIFO_ADDR;
 
-	/* INIT_WORK(&dev->wake, fwnet_handle_queue);*/
 	INIT_LIST_HEAD(&dev->packet_list);
 	INIT_LIST_HEAD(&dev->broadcasted_list);
 	INIT_LIST_HEAD(&dev->sent_list);
+	INIT_LIST_HEAD(&dev->peer_list);
 
 	dev->card = card;
+	dev->netdev = net;
 
 	/*
 	 * Use the RFC 2734 default 1500 octets or the maximum payload
@@ -1518,43 +1481,57 @@ static int fwnet_probe(struct device *_dev)
 	/* Set our hardware address while we're at it */
 	put_unaligned_be64(card->guid, net->dev_addr);
 	put_unaligned_be64(~0ULL, net->broadcast);
-	if (register_netdev(net)) {
+	ret = register_netdev(net);
+	if (ret) {
 		fw_error("Cannot register the driver\n");
 		goto out;
 	}
 
+	list_add_tail(&dev->dev_link, &fwnet_device_list);
 	fw_notify("%s: IPv4 over FireWire on device %016llx\n",
 		  net->name, (unsigned long long)card->guid);
-	card->netdev = net;
-
-	return 0;
+ have_dev:
+	ret = fwnet_add_peer(dev, unit, device);
+	if (ret && new_netdev) {
+		unregister_netdev(net);
+		list_del(&dev->dev_link);
+	}
  out:
-	if (net)
+	if (ret && new_netdev)
 		free_netdev(net);
 
-	return -ENOENT;
+	mutex_unlock(&fwnet_device_mutex);
+
+	return ret;
+}
+
+static void fwnet_remove_peer(struct fwnet_peer *peer)
+{
+	struct fwnet_partial_datagram *pd, *pd_next;
+
+	spin_lock_irq(&peer->dev->lock);
+	list_del(&peer->peer_link);
+	spin_unlock_irq(&peer->dev->lock);
+
+	list_for_each_entry_safe(pd, pd_next, &peer->pd_list, pd_link)
+		fwnet_pd_delete(pd);
+
+	kfree(peer);
 }
 
 static int fwnet_remove(struct device *_dev)
 {
-	struct fw_unit *unit = fw_unit(_dev);
-	struct fw_device *device = fw_parent_device(unit);
-	struct fw_card *card = device->card;
+	struct fwnet_peer *peer = _dev->driver_data;
+	struct fwnet_device *dev = peer->dev;
 	struct net_device *net;
-	struct fwnet_device *dev;
-	struct fwnet_peer *peer;
-	struct fwnet_partial_datagram *pd, *pd_next;
 	struct fwnet_packet_task *ptask, *pt_next;
 
-	if (!device->is_local) {
-		fwnet_peer_delete(card, device);
+	mutex_lock(&fwnet_device_mutex);
 
-		return 0;
-	}
+	fwnet_remove_peer(peer);
 
-	net = card->netdev;
-	if (net) {
-		dev = netdev_priv(net);
+	if (list_empty(&dev->peer_list)) {
+		net = dev->netdev;
 		unregister_netdev(net);
 
 		if (dev->local_fifo != FWNET_NO_FIFO_ADDR)
@@ -1580,19 +1557,11 @@ static int fwnet_remove(struct device *_dev)
 			dev_kfree_skb_any(ptask->skb);
 			kmem_cache_free(fwnet_packet_task_cache, ptask);
 		}
-		list_for_each_entry(peer, &card->peer_list, peer_link) {
-			if (peer->pdg_size) {
-				list_for_each_entry_safe(pd, pd_next,
-						&peer->pd_list, pd_link)
-					fwnet_pd_delete(pd);
-				peer->pdg_size = 0;
-			}
-			peer->fifo = FWNET_NO_FIFO_ADDR;
-		}
 		free_netdev(net);
-		card->netdev = NULL;
 	}
 
+	mutex_unlock(&fwnet_device_mutex);
+
 	return 0;
 }
 
@@ -1603,24 +1572,15 @@ static int fwnet_remove(struct device *_dev)
 static void fwnet_update(struct fw_unit *unit)
 {
 	struct fw_device *device = fw_parent_device(unit);
-	struct net_device *net = device->card->netdev;
-	struct fwnet_device *dev;
-	struct fwnet_peer *peer;
-	u64 guid;
+	struct fwnet_peer *peer = unit->device.driver_data;
+	int generation;
 
-	if (net && !device->is_local) {
-		dev = netdev_priv(net);
-		guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
-		peer = fwnet_peer_find_by_guid(dev, guid);
-		if (!peer) {
-			fw_error("fwnet_update: no peer for device %016llx\n",
-				 (unsigned long long)guid);
-			return;
-		}
-		peer->generation = device->generation;
-		rmb();
-		peer->node_id = device->node_id;
-	}
+	generation = device->generation;
+
+	spin_lock_irq(&peer->dev->lock);
+	peer->node_id    = device->node_id;
+	peer->generation = generation;
+	spin_unlock_irq(&peer->dev->lock);
 }
 
 static const struct ieee1394_device_id fwnet_id_table[] = {
diff --git a/include/linux/firewire.h b/include/linux/firewire.h
index 5cb0c1549ff1..9823946adbc5 100644
--- a/include/linux/firewire.h
+++ b/include/linux/firewire.h
@@ -131,10 +131,6 @@ struct fw_card {
 	bool broadcast_channel_allocated;
 	u32 broadcast_channel;
 	u32 topology_map[(CSR_TOPOLOGY_MAP_END - CSR_TOPOLOGY_MAP) / 4];
-
-	/* firewire-net driver data */
-	void *netdev;
-	struct list_head peer_list;
 };
 
 static inline struct fw_card *fw_card_get(struct fw_card *card)
-- 
cgit v1.2.3-59-g8ed1b


From 156ce867a6725ea8a24b452469a6dc9f3fa6a161 Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Sun, 14 Jun 2009 11:46:57 +0200
Subject: firewire: net: remove unused code

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/net.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c
index d83c54587a63..d67e8d957d21 100644
--- a/drivers/firewire/net.c
+++ b/drivers/firewire/net.c
@@ -628,13 +628,8 @@ static int fwnet_finish_incoming_packet(struct net_device *net,
 				skb->pkt_type = PACKET_MULTICAST;
 #endif
 		} else {
-			if (memcmp(eth->h_dest, net->dev_addr, net->addr_len)) {
-				u64 a1, a2;
-
-				memcpy(&a1, eth->h_dest, sizeof(u64));
-				memcpy(&a2, net->dev_addr, sizeof(u64));
+			if (memcmp(eth->h_dest, net->dev_addr, net->addr_len))
 				skb->pkt_type = PACKET_OTHERHOST;
-			}
 		}
 		if (ntohs(eth->h_proto) >= 1536) {
 			protocol = eth->h_proto;
-- 
cgit v1.2.3-59-g8ed1b


From 1337f8535ac1f41915d9e8aa03d5a3edf2f7c0a5 Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Sun, 14 Jun 2009 11:47:44 +0200
Subject: firewire: net: adjust net_device ops

The .ndo_tx_timeout callback is currently without function; delete it.
Give .watchdog_timeo a proper time value; lower it to 2 seconds.

Decrease the .tx_queue_len from 1000 (as in Ethernet card drivers) to 10
because we have only 64 transaction labels available, and responders
might have further limits of their AR req contexts.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/net.c | 12 ++----------
 1 file changed, 2 insertions(+), 10 deletions(-)

diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c
index d67e8d957d21..95e35a36b01f 100644
--- a/drivers/firewire/net.c
+++ b/drivers/firewire/net.c
@@ -1328,13 +1328,6 @@ static int fwnet_tx(struct sk_buff *skb, struct net_device *net)
 	return NETDEV_TX_OK;
 }
 
-static void fwnet_tx_timeout(struct net_device *net)
-{
-	fw_error("%s: timeout\n", net->name);
-
-	/* FIXME: What to do if we timeout? */
-}
-
 static int fwnet_change_mtu(struct net_device *net, int new_mtu)
 {
 	if (new_mtu < 68)
@@ -1359,7 +1352,6 @@ static const struct net_device_ops fwnet_netdev_ops = {
 	.ndo_open       = fwnet_open,
 	.ndo_stop	= fwnet_stop,
 	.ndo_start_xmit = fwnet_tx,
-	.ndo_tx_timeout = fwnet_tx_timeout,
 	.ndo_change_mtu = fwnet_change_mtu,
 };
 
@@ -1367,13 +1359,13 @@ static void fwnet_init_dev(struct net_device *net)
 {
 	net->header_ops		= &fwnet_header_ops;
 	net->netdev_ops		= &fwnet_netdev_ops;
-	net->watchdog_timeo	= 100000; /* ? FIXME */
+	net->watchdog_timeo	= 2 * HZ;
 	net->flags		= IFF_BROADCAST | IFF_MULTICAST;
 	net->features		= NETIF_F_HIGHDMA;
 	net->addr_len		= FWNET_ALEN;
 	net->hard_header_len	= FWNET_HLEN;
 	net->type		= ARPHRD_IEEE1394;
-	net->tx_queue_len	= 1000; /* ? FIXME */
+	net->tx_queue_len	= 10;
 	SET_ETHTOOL_OPS(net, &fwnet_ethtool_ops);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 97f8a27a5cfb08c9ce3abc90aaafb791759aed94 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Date: Thu, 11 Jun 2009 22:51:12 -0400
Subject: [ARM] orion5x: increment window counter after adding sram mapping

Without incrementing the counter the next window setup will overwrite
the SRAM mapping.

Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
Signed-off-by: Nicolas Pitre <nico@marvell.com>
---
 arch/arm/mach-orion5x/addr-map.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/mach-orion5x/addr-map.c b/arch/arm/mach-orion5x/addr-map.c
index 6f3f77d031d0..d78731edebb6 100644
--- a/arch/arm/mach-orion5x/addr-map.c
+++ b/arch/arm/mach-orion5x/addr-map.c
@@ -200,6 +200,6 @@ void __init orion5x_setup_pcie_wa_win(u32 base, u32 size)
 
 int __init orion5x_setup_sram_win(void)
 {
-	return setup_cpu_win(win_alloc_count, ORION5X_SRAM_PHYS_BASE,
+	return setup_cpu_win(win_alloc_count++, ORION5X_SRAM_PHYS_BASE,
 			ORION5X_SRAM_SIZE, TARGET_SRAM, ATTR_SRAM, -1);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 3fade49b734cca2d8c4f1bcd7c3023302b557f3b Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@cam.org>
Date: Thu, 11 Jun 2009 22:27:20 +0200
Subject: [ARM] orion5x: register the crypto device on SOCs that support it

Not all Orion variants do implement the crypto unit.

Signed-off-by: Nicolas Pitre <nico@marvell.com>
---
 arch/arm/mach-orion5x/common.c | 10 +++++++++-
 arch/arm/mach-orion5x/common.h |  1 -
 2 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-orion5x/common.c b/arch/arm/mach-orion5x/common.c
index eafcc49009ea..f87fa1253803 100644
--- a/arch/arm/mach-orion5x/common.c
+++ b/arch/arm/mach-orion5x/common.c
@@ -562,7 +562,7 @@ static struct platform_device orion5x_crypto_device = {
 	.resource       = orion5x_crypto_res,
 };
 
-int __init orion5x_crypto_init(void)
+static int __init orion5x_crypto_init(void)
 {
 	int ret;
 
@@ -696,6 +696,14 @@ void __init orion5x_init(void)
 		disable_hlt();
 	}
 
+	/*
+	 * The 5082/5181l/5182/6082/6082l/6183 have crypto
+	 * while 5180n/5181/5281 don't have crypto.
+	 */
+	if ((dev == MV88F5181_DEV_ID && rev >= MV88F5181L_REV_A0) ||
+	    dev == MV88F5182_DEV_ID || dev == MV88F6183_DEV_ID)
+		orion5x_crypto_init();
+
 	/*
 	 * Register watchdog driver
 	 */
diff --git a/arch/arm/mach-orion5x/common.h b/arch/arm/mach-orion5x/common.h
index de483e83edd7..8f004503c96d 100644
--- a/arch/arm/mach-orion5x/common.h
+++ b/arch/arm/mach-orion5x/common.h
@@ -38,7 +38,6 @@ void orion5x_spi_init(void);
 void orion5x_uart0_init(void);
 void orion5x_uart1_init(void);
 void orion5x_xor_init(void);
-int orion5x_crypto_init(void);
 
 /*
  * PCIe/PCI functions.
-- 
cgit v1.2.3-59-g8ed1b


From a2ab0ce09edf20b5228208405dd14bc8790fbdbd Mon Sep 17 00:00:00 2001
From: Christian Engelmayer <christian.engelmayer@frequentis.com>
Date: Sat, 13 Jun 2009 23:06:29 +0200
Subject: jffs2: leaking jffs2_summary in function jffs2_scan_medium

In case of an error returned by file_dirty() 's' is not freed as the cleanup
path is skipped.

Reported by Coverity.

Signed-off-by: Christian Engelmayer <christian.engelmayer@frequentis.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 fs/jffs2/scan.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/jffs2/scan.c b/fs/jffs2/scan.c
index 1d437de1e9a8..7515e73e2bfb 100644
--- a/fs/jffs2/scan.c
+++ b/fs/jffs2/scan.c
@@ -196,7 +196,7 @@ int jffs2_scan_medium(struct jffs2_sb_info *c)
 				if (c->nextblock) {
 					ret = file_dirty(c, c->nextblock);
 					if (ret)
-						return ret;
+						goto out;
 					/* deleting summary information of the old nextblock */
 					jffs2_sum_reset_collected(c->summary);
 				}
@@ -207,7 +207,7 @@ int jffs2_scan_medium(struct jffs2_sb_info *c)
 			} else {
 				ret = file_dirty(c, jeb);
 				if (ret)
-					return ret;
+					goto out;
 			}
 			break;
 
-- 
cgit v1.2.3-59-g8ed1b


From 61f98ffd74254a95871168bd5a6646b4f3002e31 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Thu, 11 Jun 2009 10:27:32 -0400
Subject: cifs: display scopeid in /proc/mounts

Move address display into a new function and display the scopeid as part
of the address in /proc/mounts.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/cifsfs.c | 33 ++++++++++++++++++++++-----------
 1 file changed, 22 insertions(+), 11 deletions(-)

diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 8b315708cb3f..ddef913ff3e6 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -333,6 +333,27 @@ cifs_destroy_inode(struct inode *inode)
 	kmem_cache_free(cifs_inode_cachep, CIFS_I(inode));
 }
 
+static void
+cifs_show_address(struct seq_file *s, struct TCP_Server_Info *server)
+{
+	seq_printf(s, ",addr=");
+
+	switch (server->addr.sockAddr.sin_family) {
+	case AF_INET:
+		seq_printf(s, "%pI4", &server->addr.sockAddr.sin_addr.s_addr);
+		break;
+	case AF_INET6:
+		seq_printf(s, "%pI6",
+			   &server->addr.sockAddr6.sin6_addr.s6_addr);
+		if (server->addr.sockAddr6.sin6_scope_id)
+			seq_printf(s, "%%%u",
+				   server->addr.sockAddr6.sin6_scope_id);
+		break;
+	default:
+		seq_printf(s, "(unknown)");
+	}
+}
+
 /*
  * cifs_show_options() is for displaying mount options in /proc/mounts.
  * Not all settable options are displayed but most of the important
@@ -343,7 +364,6 @@ cifs_show_options(struct seq_file *s, struct vfsmount *m)
 {
 	struct cifs_sb_info *cifs_sb;
 	struct cifsTconInfo *tcon;
-	struct TCP_Server_Info *server;
 
 	cifs_sb = CIFS_SB(m->mnt_sb);
 	tcon = cifs_sb->tcon;
@@ -364,16 +384,7 @@ cifs_show_options(struct seq_file *s, struct vfsmount *m)
 	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID)
 		seq_printf(s, ",forcegid");
 
-	server = tcon->ses->server;
-	seq_printf(s, ",addr=");
-	switch (server->addr.sockAddr6.sin6_family) {
-	case AF_INET6:
-		seq_printf(s, "%pI6", &server->addr.sockAddr6.sin6_addr);
-		break;
-	case AF_INET:
-		seq_printf(s, "%pI4", &server->addr.sockAddr.sin_addr.s_addr);
-		break;
-	}
+	cifs_show_address(s, tcon->ses->server);
 
 	if (!tcon->unix_ext)
 		seq_printf(s, ",file_mode=0%o,dir_mode=0%o",
-- 
cgit v1.2.3-59-g8ed1b


From 361ea1ae5451040cd254eee0b6df64581080b2cc Mon Sep 17 00:00:00 2001
From: Steve French <sfrench@us.ibm.com>
Date: Mon, 15 Jun 2009 13:46:12 +0000
Subject: [CIFS] Fix build break

Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/netmisc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c
index 00e6e357ae88..f9a54da97d34 100644
--- a/fs/cifs/netmisc.c
+++ b/fs/cifs/netmisc.c
@@ -166,7 +166,7 @@ int
 cifs_convert_address(char *src, void *dst)
 {
 	struct sockaddr_in *s4 = (struct sockaddr_in *) dst;
-	struct sockaddr_in6 *s6 = (Struct sockaddr_in6 *) dst;
+	struct sockaddr_in6 *s6 = (struct sockaddr_in6 *) dst;
 
 	if (cifs_inet_pton(AF_INET, src, &s4->sin_addr.s_addr)) {
 		s4->sin_family = AF_INET;
-- 
cgit v1.2.3-59-g8ed1b


From 5102aa4acdbde305646ae60c5eec7042b5016a0e Mon Sep 17 00:00:00 2001
From: Michel Pollet <buserror@gmail.com>
Date: Wed, 25 Mar 2009 17:19:14 +0000
Subject: [ARM] MINI2440: Add machine support

The MINI2440 is a chinese made s3c2440 development board with
a large set of peripherals.
This patch provides machine support for almost all the features
of the board.

Since it can come with various "options" fitted, a kernel parameter
is used to specify the lcd size, backlight control and touchscreen.

Signed-off-by: Michel Pollet <buserror@gmail.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/mach-s3c2440/Kconfig         |  10 +
 arch/arm/mach-s3c2440/Makefile        |   1 +
 arch/arm/mach-s3c2440/mach-mini2440.c | 706 ++++++++++++++++++++++++++++++++++
 3 files changed, 717 insertions(+)
 create mode 100644 arch/arm/mach-s3c2440/mach-mini2440.c

diff --git a/arch/arm/mach-s3c2440/Kconfig b/arch/arm/mach-s3c2440/Kconfig
index 5df73cbf2b40..8cfeaec37306 100644
--- a/arch/arm/mach-s3c2440/Kconfig
+++ b/arch/arm/mach-s3c2440/Kconfig
@@ -84,5 +84,15 @@ config MACH_AT2440EVB
 	help
 	  Say Y here if you are using the AT2440EVB development board
 
+config MACH_MINI2440
+	bool "MINI2440 development board"
+	select CPU_S3C2440
+	select EEPROM_AT24
+	select LEDS_TRIGGER_BACKLIGHT
+	select SND_S3C24XX_SOC_S3C24XX_UDA134X
+	help
+	  Say Y here to select support for the MINI2440. Is a 10cm x 10cm board
+	  available via various sources. It can come with a 3.5" or 7" touch LCD.
+
 endmenu
 
diff --git a/arch/arm/mach-s3c2440/Makefile b/arch/arm/mach-s3c2440/Makefile
index 0b4440e79b90..bfadcf684a2a 100644
--- a/arch/arm/mach-s3c2440/Makefile
+++ b/arch/arm/mach-s3c2440/Makefile
@@ -22,3 +22,4 @@ obj-$(CONFIG_MACH_RX3715)	+= mach-rx3715.o
 obj-$(CONFIG_ARCH_S3C2440)	+= mach-smdk2440.o
 obj-$(CONFIG_MACH_NEXCODER_2440) += mach-nexcoder.o
 obj-$(CONFIG_MACH_AT2440EVB) += mach-at2440evb.o
+obj-$(CONFIG_MACH_MINI2440) += mach-mini2440.o
diff --git a/arch/arm/mach-s3c2440/mach-mini2440.c b/arch/arm/mach-s3c2440/mach-mini2440.c
new file mode 100644
index 000000000000..eb4b24501a9f
--- /dev/null
+++ b/arch/arm/mach-s3c2440/mach-mini2440.c
@@ -0,0 +1,706 @@
+/* linux/arch/arm/mach-s3c2440/mach-mini2440.c
+ *
+ * Copyright (c) 2008 Ramax Lo <ramaxlo@gmail.com>
+ *      Based on mach-anubis.c by Ben Dooks <ben@simtec.co.uk>
+ *      and modifications by SBZ <sbz@spgui.org> and
+ *      Weibing <http://weibing.blogbus.com> and
+ *      Michel Pollet <buserror@gmail.com>
+ *
+ * For product information, visit http://code.google.com/p/mini2440/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/interrupt.h>
+#include <linux/list.h>
+#include <linux/timer.h>
+#include <linux/init.h>
+#include <linux/gpio.h>
+#include <linux/input.h>
+#include <linux/io.h>
+#include <linux/serial_core.h>
+#include <linux/dm9000.h>
+#include <linux/i2c/at24.h>
+#include <linux/platform_device.h>
+#include <linux/gpio_keys.h>
+#include <linux/i2c.h>
+#include <linux/mmc/host.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+
+#include <mach/hardware.h>
+#include <mach/fb.h>
+#include <asm/mach-types.h>
+
+#include <plat/regs-serial.h>
+#include <mach/regs-gpio.h>
+#include <mach/leds-gpio.h>
+#include <mach/regs-mem.h>
+#include <mach/regs-lcd.h>
+#include <mach/irqs.h>
+#include <plat/nand.h>
+#include <plat/iic.h>
+#include <plat/mci.h>
+#include <plat/udc.h>
+
+#include <plat/regs-serial.h>
+#include <mach/regs-gpio.h>
+#include <mach/regs-mem.h>
+#include <mach/regs-lcd.h>
+
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/nand.h>
+#include <linux/mtd/nand_ecc.h>
+#include <linux/mtd/partitions.h>
+
+#include <plat/clock.h>
+#include <plat/devs.h>
+#include <plat/cpu.h>
+
+#include <sound/s3c24xx_uda134x.h>
+
+#define MACH_MINI2440_DM9K_BASE (S3C2410_CS4 + 0x300)
+
+static struct map_desc mini2440_iodesc[] __initdata = {
+	/* nothing to declare, move along */
+};
+
+#define UCON S3C2410_UCON_DEFAULT
+#define ULCON S3C2410_LCON_CS8 | S3C2410_LCON_PNONE | S3C2410_LCON_STOPB
+#define UFCON S3C2410_UFCON_RXTRIG8 | S3C2410_UFCON_FIFOMODE
+
+
+static struct s3c2410_uartcfg mini2440_uartcfgs[] __initdata = {
+	[0] = {
+		.hwport	     = 0,
+		.flags	     = 0,
+		.ucon	     = UCON,
+		.ulcon	     = ULCON,
+		.ufcon	     = UFCON,
+	},
+	[1] = {
+		.hwport	     = 1,
+		.flags	     = 0,
+		.ucon	     = UCON,
+		.ulcon	     = ULCON,
+		.ufcon	     = UFCON,
+	},
+	[2] = {
+		.hwport	     = 2,
+		.flags	     = 0,
+		.ucon	     = UCON,
+		.ulcon	     = ULCON,
+		.ufcon	     = UFCON,
+	},
+};
+
+/* USB device UDC support */
+
+static void mini2440_udc_pullup(enum s3c2410_udc_cmd_e cmd)
+{
+	pr_debug("udc: pullup(%d)\n", cmd);
+
+	switch (cmd) {
+		case S3C2410_UDC_P_ENABLE :
+			s3c2410_gpio_setpin(S3C2410_GPC(5), 1);
+			break;
+		case S3C2410_UDC_P_DISABLE :
+			s3c2410_gpio_setpin(S3C2410_GPC(5), 0);
+			break;
+		case S3C2410_UDC_P_RESET :
+			break;
+		default:
+			break;
+	}
+}
+
+static struct s3c2410_udc_mach_info mini2440_udc_cfg __initdata = {
+	.udc_command		= mini2440_udc_pullup,
+};
+
+
+/* LCD timing and setup */
+
+/*
+ * This macro simplifies the table bellow
+ */
+#define _LCD_DECLARE(_clock,_xres,margin_left,margin_right,hsync, \
+			_yres,margin_top,margin_bottom,vsync, refresh) \
+	.width = _xres, \
+	.xres = _xres, \
+	.height = _yres, \
+	.yres = _yres, \
+	.left_margin	= margin_left,	\
+	.right_margin	= margin_right,	\
+	.upper_margin	= margin_top,	\
+	.lower_margin	= margin_bottom,	\
+	.hsync_len	= hsync,	\
+	.vsync_len	= vsync,	\
+	.pixclock	= ((_clock*100000000000LL) /	\
+			   ((refresh) * \
+			   (hsync + margin_left + _xres + margin_right) * \
+			   (vsync + margin_top + _yres + margin_bottom))), \
+	.bpp		= 16,\
+	.type		= (S3C2410_LCDCON1_TFT16BPP |\
+			   S3C2410_LCDCON1_TFT)
+
+struct s3c2410fb_display mini2440_lcd_cfg[] __initdata = {
+	[0] = {	/* mini2440 + 3.5" TFT + touchscreen */
+		_LCD_DECLARE(
+			7,			/* The 3.5 is quite fast */
+			240, 21, 38, 6, 	/* x timing */
+			320, 4, 4, 2,		/* y timing */
+			60),			/* refresh rate */
+		.lcdcon5	= (S3C2410_LCDCON5_FRM565 |
+				   S3C2410_LCDCON5_INVVLINE |
+				   S3C2410_LCDCON5_INVVFRAME |
+				   S3C2410_LCDCON5_INVVDEN |
+				   S3C2410_LCDCON5_PWREN),
+	},
+	[1] = { /* mini2440 + 7" TFT + touchscreen */
+		_LCD_DECLARE(
+			10,			/* the 7" runs slower */
+			800, 40, 40, 48, 	/* x timing */
+			480, 29, 3, 3,		/* y timing */
+			50),			/* refresh rate */
+		.lcdcon5	= (S3C2410_LCDCON5_FRM565 |
+				   S3C2410_LCDCON5_INVVLINE |
+				   S3C2410_LCDCON5_INVVFRAME |
+				   S3C2410_LCDCON5_PWREN),
+	},
+	/* The VGA shield can outout at several resolutions. All share 
+	 * the same timings, however, anything smaller than 1024x768
+	 * will only be displayed in the top left corner of a 1024x768
+	 * XGA output unless you add optional dip switches to the shield.
+	 * Therefore timings for other resolutions have been ommited here.
+	 */
+	[2] = {
+		_LCD_DECLARE(
+			10,
+			1024, 1, 2, 2,		/* y timing */
+			768, 200, 16, 16, 	/* x timing */
+			24),	/* refresh rate, maximum stable,
+				 tested with the FPGA shield */
+		.lcdcon5	= (S3C2410_LCDCON5_FRM565 |
+				   S3C2410_LCDCON5_HWSWP),
+	},
+};
+
+/* todo - put into gpio header */
+
+#define S3C2410_GPCCON_MASK(x)	(3 << ((x) * 2))
+#define S3C2410_GPDCON_MASK(x)	(3 << ((x) * 2))
+
+struct s3c2410fb_mach_info mini2440_fb_info __initdata = {
+	.displays	 = &mini2440_lcd_cfg[0], /* not constant! see init */
+	.num_displays	 = 1,
+	.default_display = 0,
+
+	/* Enable VD[2..7], VD[10..15], VD[18..23] and VCLK, syncs, VDEN
+	 * and disable the pull down resistors on pins we are using for LCD
+	 * data. */
+
+	.gpcup		= (0xf << 1) | (0x3f << 10),
+
+	.gpccon		= (S3C2410_GPC1_VCLK   | S3C2410_GPC2_VLINE |
+			   S3C2410_GPC3_VFRAME | S3C2410_GPC4_VM |
+			   S3C2410_GPC10_VD2   | S3C2410_GPC11_VD3 |
+			   S3C2410_GPC12_VD4   | S3C2410_GPC13_VD5 |
+			   S3C2410_GPC14_VD6   | S3C2410_GPC15_VD7),
+
+	.gpccon_mask	= (S3C2410_GPCCON_MASK(1)  | S3C2410_GPCCON_MASK(2)  |
+			   S3C2410_GPCCON_MASK(3)  | S3C2410_GPCCON_MASK(4)  |
+			   S3C2410_GPCCON_MASK(10) | S3C2410_GPCCON_MASK(11) |
+			   S3C2410_GPCCON_MASK(12) | S3C2410_GPCCON_MASK(13) |
+			   S3C2410_GPCCON_MASK(14) | S3C2410_GPCCON_MASK(15)),
+
+	.gpdup		= (0x3f << 2) | (0x3f << 10),
+
+	.gpdcon		= (S3C2410_GPD2_VD10  | S3C2410_GPD3_VD11 |
+			   S3C2410_GPD4_VD12  | S3C2410_GPD5_VD13 |
+			   S3C2410_GPD6_VD14  | S3C2410_GPD7_VD15 |
+			   S3C2410_GPD10_VD18 | S3C2410_GPD11_VD19 |
+			   S3C2410_GPD12_VD20 | S3C2410_GPD13_VD21 |
+			   S3C2410_GPD14_VD22 | S3C2410_GPD15_VD23),
+
+	.gpdcon_mask	= (S3C2410_GPDCON_MASK(2)  | S3C2410_GPDCON_MASK(3) |
+			   S3C2410_GPDCON_MASK(4)  | S3C2410_GPDCON_MASK(5) |
+			   S3C2410_GPDCON_MASK(6)  | S3C2410_GPDCON_MASK(7) |
+			   S3C2410_GPDCON_MASK(10) | S3C2410_GPDCON_MASK(11)|
+			   S3C2410_GPDCON_MASK(12) | S3C2410_GPDCON_MASK(13)|
+			   S3C2410_GPDCON_MASK(14) | S3C2410_GPDCON_MASK(15)),
+};
+
+/* MMC/SD  */
+
+static struct s3c24xx_mci_pdata mini2440_mmc_cfg __initdata = {
+   .gpio_detect   = S3C2410_GPG(8),
+   .gpio_wprotect = S3C2410_GPH(8),
+   .set_power     = NULL,
+   .ocr_avail     = MMC_VDD_32_33|MMC_VDD_33_34,
+};
+
+/* NAND Flash on MINI2440 board */
+
+static struct mtd_partition mini2440_default_nand_part[] __initdata = {
+	[0] = {
+		.name	= "u-boot",
+		.size	= SZ_256K,
+		.offset	= 0,
+	},
+	[1] = {
+		.name	= "u-boot-env",
+		.size	= SZ_128K,
+		.offset	= SZ_256K,
+	},
+	[2] = {
+		.name	= "kernel",
+		/* 5 megabytes, for a kernel with no modules
+		 * or a uImage with a ramdisk attached */
+		.size	= 0x00500000,
+		.offset	= SZ_256K + SZ_128K,
+	},
+	[3] = {
+		.name	= "root",
+		.offset	= SZ_256K + SZ_128K + 0x00500000,
+		.size	= MTDPART_SIZ_FULL,
+	},
+};
+
+static struct s3c2410_nand_set mini2440_nand_sets[] __initdata = {
+	[0] = {
+		.name		= "nand",
+		.nr_chips	= 1,
+		.nr_partitions	= ARRAY_SIZE(mini2440_default_nand_part),
+		.partitions	= mini2440_default_nand_part,
+	},
+};
+
+static struct s3c2410_platform_nand mini2440_nand_info __initdata = {
+	.tacls		= 0,
+	.twrph0		= 25,
+	.twrph1		= 15,
+	.nr_sets	= ARRAY_SIZE(mini2440_nand_sets),
+	.sets		= mini2440_nand_sets,
+	.ignore_unset_ecc = 1,
+};
+
+/* DM9000AEP 10/100 ethernet controller */
+
+static struct resource mini2440_dm9k_resource[] __initdata = {
+	[0] = {
+		.start = MACH_MINI2440_DM9K_BASE,
+		.end   = MACH_MINI2440_DM9K_BASE + 3,
+		.flags = IORESOURCE_MEM
+	},
+	[1] = {
+		.start = MACH_MINI2440_DM9K_BASE + 4,
+		.end   = MACH_MINI2440_DM9K_BASE + 7,
+		.flags = IORESOURCE_MEM
+	},
+	[2] = {
+		.start = IRQ_EINT7,
+		.end   = IRQ_EINT7,
+		.flags = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHEDGE,
+	}
+};
+
+/*
+ * The DM9000 has no eeprom, and it's MAC address is set by
+ * the bootloader before starting the kernel.
+ */
+static struct dm9000_plat_data mini2440_dm9k_pdata __initdata = {
+	.flags		= (DM9000_PLATF_16BITONLY | DM9000_PLATF_NO_EEPROM),
+};
+
+static struct platform_device mini2440_device_eth __initdata = {
+	.name		= "dm9000",
+	.id		= -1,
+	.num_resources	= ARRAY_SIZE(mini2440_dm9k_resource),
+	.resource	= mini2440_dm9k_resource,
+	.dev		= {
+		.platform_data	= &mini2440_dm9k_pdata,
+	},
+};
+
+/*  CON5
+ *	+--+	 /-----\
+ *	|  |    |	|
+ *	|  |	|  BAT	|
+ *	|  |	 \_____/
+ *	|  |
+ *	|  |  +----+  +----+
+ *	|  |  | K5 |  | K1 |
+ *	|  |  +----+  +----+
+ *	|  |  +----+  +----+
+ *	|  |  | K4 |  | K2 |
+ *	|  |  +----+  +----+
+ *	|  |  +----+  +----+
+ *	|  |  | K6 |  | K3 |
+ *	|  |  +----+  +----+
+ *	  .....
+ */
+static struct gpio_keys_button mini2440_buttons[] __initdata = {
+	{
+		.gpio		= S3C2410_GPG(0),		/* K1 */
+		.code		= KEY_F1,
+		.desc		= "Button 1",
+		.active_low	= 1,
+	},
+	{
+		.gpio		= S3C2410_GPG(3),		/* K2 */
+		.code		= KEY_F2,
+		.desc		= "Button 2",
+		.active_low	= 1,
+	},
+	{
+		.gpio		= S3C2410_GPG(5),		/* K3 */
+		.code		= KEY_F3,
+		.desc		= "Button 3",
+		.active_low	= 1,
+	},
+	{
+		.gpio		= S3C2410_GPG(6),		/* K4 */
+		.code		= KEY_POWER,
+		.desc		= "Power",
+		.active_low	= 1,
+	},
+	{
+		.gpio		= S3C2410_GPG(7),		/* K5 */
+		.code		= KEY_F5,
+		.desc		= "Button 5",
+		.active_low	= 1,
+	},
+#if 0
+	/* this pin is also known as TCLK1 and seems to already
+	 * marked as "in use" somehow in the kernel -- possibly wrongly */
+	{
+		.gpio		= S3C2410_GPG(11),	/* K6 */
+		.code		= KEY_F6,
+		.desc		= "Button 6",
+		.active_low	= 1,
+	},
+#endif
+};
+
+static struct gpio_keys_platform_data mini2440_button_data __initdata = {
+	.buttons	= mini2440_buttons,
+	.nbuttons	= ARRAY_SIZE(mini2440_buttons),
+};
+
+static struct platform_device mini2440_button_device __initdata = {
+	.name		= "gpio-keys",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= &mini2440_button_data,
+	}
+};
+
+/* LEDS */
+
+static struct s3c24xx_led_platdata mini2440_led1_pdata __initdata = {
+	.name		= "led1",
+	.gpio		= S3C2410_GPB(5),
+	.flags		= S3C24XX_LEDF_ACTLOW | S3C24XX_LEDF_TRISTATE,
+	.def_trigger	= "heartbeat",
+};
+
+static struct s3c24xx_led_platdata mini2440_led2_pdata __initdata = {
+	.name		= "led2",
+	.gpio		= S3C2410_GPB(6),
+	.flags		= S3C24XX_LEDF_ACTLOW | S3C24XX_LEDF_TRISTATE,
+	.def_trigger	= "nand-disk",
+};
+
+static struct s3c24xx_led_platdata mini2440_led3_pdata __initdata = {
+	.name		= "led3",
+	.gpio		= S3C2410_GPB(7),
+	.flags		= S3C24XX_LEDF_ACTLOW | S3C24XX_LEDF_TRISTATE,
+	.def_trigger	= "mmc0",
+};
+
+static struct s3c24xx_led_platdata mini2440_led4_pdata __initdata = {
+	.name		= "led4",
+	.gpio		= S3C2410_GPB(8),
+	.flags		= S3C24XX_LEDF_ACTLOW | S3C24XX_LEDF_TRISTATE,
+	.def_trigger	= "",
+};
+
+static struct s3c24xx_led_platdata mini2440_led_backlight_pdata __initdata = {
+	.name		= "backlight",
+	.gpio		= S3C2410_GPG(4),
+	.def_trigger	= "backlight",
+};
+
+static struct platform_device mini2440_led1 __initdata = {
+	.name		= "s3c24xx_led",
+	.id		= 1,
+	.dev		= {
+		.platform_data	= &mini2440_led1_pdata,
+	},
+};
+
+static struct platform_device mini2440_led2 __initdata = {
+	.name		= "s3c24xx_led",
+	.id		= 2,
+	.dev		= {
+		.platform_data	= &mini2440_led2_pdata,
+	},
+};
+
+static struct platform_device mini2440_led3 __initdata = {
+	.name		= "s3c24xx_led",
+	.id		= 3,
+	.dev		= {
+		.platform_data	= &mini2440_led3_pdata,
+	},
+};
+
+static struct platform_device mini2440_led4 __initdata = {
+	.name		= "s3c24xx_led",
+	.id		= 4,
+	.dev		= {
+		.platform_data	= &mini2440_led4_pdata,
+	},
+};
+
+static struct platform_device mini2440_led_backlight __initdata = {
+	.name		= "s3c24xx_led",
+	.id		= 5,
+	.dev		= {
+		.platform_data	= &mini2440_led_backlight_pdata,
+	},
+};
+
+/* AUDIO */
+
+static struct s3c24xx_uda134x_platform_data mini2440_audio_pins __initdata = {
+	.l3_clk = S3C2410_GPB(4),
+	.l3_mode = S3C2410_GPB(2),
+	.l3_data = S3C2410_GPB(3),
+	.model = UDA134X_UDA1341
+};
+
+static struct platform_device mini2440_audio __initdata = {
+	.name		= "s3c24xx_uda134x",
+	.id		= 0,
+	.dev		= {
+		.platform_data	= &mini2440_audio_pins,
+	},
+};
+
+/*
+ * I2C devices
+ */
+static struct at24_platform_data at24c08 = {
+	.byte_len	= SZ_8K / 8,
+	.page_size	= 16,
+};
+
+static struct i2c_board_info mini2440_i2c_devs[] __initdata = {
+	{
+		I2C_BOARD_INFO("24c08", 0x50),
+		.platform_data = &at24c08,
+	},
+};
+
+static struct platform_device *mini2440_devices[] __initdata = {
+	&s3c_device_usb,
+	&s3c_device_wdt,
+/*	&s3c_device_adc,*/ /* ADC doesn't like living with touchscreen ! */
+	&s3c_device_i2c0,
+	&s3c_device_rtc,
+	&s3c_device_usbgadget,
+	&mini2440_device_eth,
+	&mini2440_led1,
+	&mini2440_led2,
+	&mini2440_led3,
+	&mini2440_led4,
+	&mini2440_button_device,
+	&s3c_device_nand,
+	&s3c_device_sdi,
+	&s3c_device_iis,
+	&mini2440_audio,
+/*	&s3c_device_timer[0],*/	/* buzzer pwm, no API for it */
+	/* remaining devices are optional */
+};
+
+static void __init mini2440_map_io(void)
+{
+	s3c24xx_init_io(mini2440_iodesc, ARRAY_SIZE(mini2440_iodesc));
+	s3c24xx_init_clocks(12000000);
+	s3c24xx_init_uarts(mini2440_uartcfgs, ARRAY_SIZE(mini2440_uartcfgs));
+
+	s3c_device_nand.dev.platform_data = &mini2440_nand_info;
+	s3c_device_sdi.dev.platform_data = &mini2440_mmc_cfg;
+}
+
+/*
+ * mini2440_features string
+ *
+ * t = Touchscreen present
+ * b = backlight control
+ * c = camera [TODO]
+ * 0-9 LCD configuration
+ *
+ */
+static char mini2440_features_str[12] __initdata = "0tb";
+
+static int __init mini2440_features_setup(char *str)
+{
+	if (str)
+		strlcpy(mini2440_features_str, str, sizeof(mini2440_features_str));
+	return 1;
+}
+
+__setup("mini2440=", mini2440_features_setup);
+
+#define FEATURE_SCREEN (1 << 0)
+#define FEATURE_BACKLIGHT (1 << 1)
+#define FEATURE_TOUCH (1 << 2)
+#define FEATURE_CAMERA (1 << 3)
+
+struct mini2440_features_t {
+	int count;
+	int done;
+	int lcd_index;
+	struct platform_device *optional[8];
+};
+
+static void mini2440_parse_features(
+		struct mini2440_features_t * features,
+		const char * features_str )
+{
+	const char * fp = features_str;
+
+	features->count = 0;
+	features->done = 0;
+	features->lcd_index = -1;
+
+	while (*fp) {
+		char f = *fp++;
+
+		switch (f) {
+		case '0'...'9':	/* tft screen */
+			if (features->done & FEATURE_SCREEN) {
+				printk(KERN_INFO "MINI2440: '%c' ignored, "
+					"screen type already set\n", f);
+			} else {
+				int li = f - '0';
+				if (li >= ARRAY_SIZE(mini2440_lcd_cfg))
+					printk(KERN_INFO "MINI2440: "
+						"'%c' out of range LCD mode\n", f);
+				else {
+					features->optional[features->count++] =
+							&s3c_device_lcd;
+					features->lcd_index = li;
+				}
+			}
+			features->done |= FEATURE_SCREEN;
+			break;
+		case 'b':
+			if (features->done & FEATURE_BACKLIGHT)
+				printk(KERN_INFO "MINI2440: '%c' ignored, "
+					"backlight already set\n", f);
+			else {
+				features->optional[features->count++] =
+						&mini2440_led_backlight;
+			}
+			features->done |= FEATURE_BACKLIGHT;
+			break;
+		case 't':
+			printk(KERN_INFO "MINI2440: '%c' ignored, "
+				"touchscreen not compiled in\n", f);
+			break;
+		case 'c':
+			if (features->done & FEATURE_CAMERA)
+				printk(KERN_INFO "MINI2440: '%c' ignored, "
+					"camera already registered\n", f);
+			else
+				features->optional[features->count++] =
+					&s3c_device_camif;
+			features->done |= FEATURE_CAMERA;
+			break;
+		}
+	}
+}
+
+static void __init mini2440_init(void)
+{
+	struct mini2440_features_t features = { 0 };
+	int i;
+
+	printk(KERN_INFO "MINI2440: Option string mini2440=%s\n",
+			mini2440_features_str);
+
+	/* Parse the feature string */
+	mini2440_parse_features(&features, mini2440_features_str);
+
+	/* turn LCD on */
+	s3c2410_gpio_cfgpin(S3C2410_GPC(0), S3C2410_GPC0_LEND);
+
+	/* Turn the backlight early on */
+	s3c2410_gpio_setpin(S3C2410_GPG(4), 1);
+	s3c2410_gpio_cfgpin(S3C2410_GPG(4), S3C2410_GPIO_OUTPUT);
+
+	/* remove pullup on optional PWM backlight -- unused on 3.5 and 7"s */
+	s3c2410_gpio_pullup(S3C2410_GPB(1), 0);
+	s3c2410_gpio_setpin(S3C2410_GPB(1), 0);
+	s3c2410_gpio_cfgpin(S3C2410_GPB(1), S3C2410_GPIO_INPUT);
+
+	/* Make sure the D+ pullup pin is output */
+	s3c2410_gpio_cfgpin(S3C2410_GPC(5), S3C2410_GPIO_OUTPUT);
+
+	/* mark the key as input, without pullups (there is one on the board) */
+	for (i = 0; i < ARRAY_SIZE(mini2440_buttons); i++) {
+		s3c2410_gpio_pullup(mini2440_buttons[i].gpio, 0);
+		s3c2410_gpio_cfgpin(mini2440_buttons[i].gpio,
+					S3C2410_GPIO_INPUT);
+	}
+	if (features.lcd_index != -1) {
+		int li;
+
+		mini2440_fb_info.displays =
+			&mini2440_lcd_cfg[features.lcd_index];
+
+		printk(KERN_INFO "MINI2440: LCD");
+		for (li = 0; li < ARRAY_SIZE(mini2440_lcd_cfg); li++)
+			if (li == features.lcd_index)
+				printk(" [%d:%dx%d]", li,
+					mini2440_lcd_cfg[li].width,
+					mini2440_lcd_cfg[li].height);
+			else
+				printk(" %d:%dx%d", li,
+					mini2440_lcd_cfg[li].width,
+					mini2440_lcd_cfg[li].height);
+		printk("\n");
+		s3c24xx_fb_set_platdata(&mini2440_fb_info);
+	}
+	s3c24xx_udc_set_platdata(&mini2440_udc_cfg);
+	s3c_i2c0_set_platdata(NULL);
+	i2c_register_board_info(0, mini2440_i2c_devs,
+				ARRAY_SIZE(mini2440_i2c_devs));
+
+	platform_add_devices(mini2440_devices, ARRAY_SIZE(mini2440_devices));
+
+	if (features.count)	/* the optional features */
+		platform_add_devices(features.optional, features.count);
+
+}
+
+
+MACHINE_START(MINI2440, "MINI2440")
+	/* Maintainer: Michel Pollet <buserror@gmail.com> */
+	.phys_io	= S3C2410_PA_UART,
+	.io_pg_offst	= (((u32)S3C24XX_VA_UART) >> 18) & 0xfffc,
+	.boot_params	= S3C2410_SDRAM_PA + 0x100,
+	.map_io		= mini2440_map_io,
+	.init_machine	= mini2440_init,
+	.init_irq	= s3c24xx_init_irq,
+	.timer		= &s3c24xx_timer,
+MACHINE_END
-- 
cgit v1.2.3-59-g8ed1b


From 39f45d7bc0df52db8a7dd099ac02fef89514b7fc Mon Sep 17 00:00:00 2001
From: Michel Pollet <buserror@gmail.com>
Date: Wed, 20 May 2009 11:10:31 +0100
Subject: [ARM] MINI2440: Document the mini2440= kernel parameter

Also explains why the touchscreen support is not present,
and were to find a working one...

Signed-off-by: Michel Pollet <buserror@gmail.com>
[ben-linux@fluff.org: Fix header description]
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 Documentation/kernel-parameters.txt | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 5f66ba295c5d..4b765b7ded9f 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1351,6 +1351,27 @@ and is between 256 and 4096 characters. It is defined in the file
 	min_addr=nn[KMG]	[KNL,BOOT,ia64] All physical memory below this
 			physical address is ignored.
 
+	mini2440=	[ARM,HW,KNL]
+			Format:[0..2][b][c][t]
+			Default: "0tb"
+			MINI2440 configuration specification:
+			0 - The attached screen is the 3.5" TFT
+			1 - The attached screen is the 7" TFT
+			2 - The VGA Shield is attached (1024x768)
+			Leaving out the screen size parameter will not load
+			the TFT driver, and the framebuffer will be left
+			unconfigured.
+			b - Enable backlight. The TFT backlight pin will be
+			linked to the kernel VESA blanking code and a GPIO
+			LED. This parameter is not necessary when using the
+			VGA shield.
+			c - Enable the s3c camera interface.
+			t - Reserved for enabling touchscreen support. The
+			touchscreen support is not enabled in the mainstream
+			kernel as of 2.6.30, a preliminary port can be found
+			in the "bleeding edge" mini2440 support kernel at
+			http://repo.or.cz/w/linux-2.6/mini2440.git
+
 	mminit_loglevel=
 			[KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
 			parameter allows control of the logging verbosity for
-- 
cgit v1.2.3-59-g8ed1b


From bf4a59f1524202f887c4595aedb1aa489bcd9b64 Mon Sep 17 00:00:00 2001
From: Michel Pollet <buserror@gmail.com>
Date: Wed, 25 Mar 2009 17:24:08 +0000
Subject: [ARM] MINI2440: Create a mini2440_defconfig file

Just a working, starting point config file for the board.

Signed-off-by: Michel Pollet <buserror@gmail.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/configs/mini2440_defconfig | 2097 +++++++++++++++++++++++++++++++++++
 1 file changed, 2097 insertions(+)
 create mode 100644 arch/arm/configs/mini2440_defconfig

diff --git a/arch/arm/configs/mini2440_defconfig b/arch/arm/configs/mini2440_defconfig
new file mode 100644
index 000000000000..e49ed40f3be7
--- /dev/null
+++ b/arch/arm/configs/mini2440_defconfig
@@ -0,0 +1,2097 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.30-rc6
+# Wed May 20 12:29:51 2009
+#
+CONFIG_ARM=y
+CONFIG_HAVE_PWM=y
+CONFIG_SYS_SUPPORTS_APM_EMULATION=y
+CONFIG_GENERIC_GPIO=y
+# CONFIG_GENERIC_TIME is not set
+# CONFIG_GENERIC_CLOCKEVENTS is not set
+CONFIG_MMU=y
+CONFIG_NO_IOPORT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_STACKTRACE_SUPPORT=y
+CONFIG_HAVE_LATENCYTOP_SUPPORT=y
+CONFIG_LOCKDEP_SUPPORT=y
+CONFIG_TRACE_IRQFLAGS_SUPPORT=y
+CONFIG_HARDIRQS_SW_RESEND=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+# CONFIG_ARCH_HAS_ILOG2_U32 is not set
+# CONFIG_ARCH_HAS_ILOG2_U64 is not set
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ=y
+CONFIG_VECTORS_BASE=0xffff0000
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+
+#
+# General setup
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+CONFIG_LOCALVERSION=""
+# CONFIG_LOCALVERSION_AUTO is not set
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+CONFIG_SYSVIPC_SYSCTL=y
+CONFIG_POSIX_MQUEUE=y
+CONFIG_POSIX_MQUEUE_SYSCTL=y
+# CONFIG_BSD_PROCESS_ACCT is not set
+# CONFIG_TASKSTATS is not set
+# CONFIG_AUDIT is not set
+
+#
+# RCU Subsystem
+#
+CONFIG_CLASSIC_RCU=y
+# CONFIG_TREE_RCU is not set
+# CONFIG_PREEMPT_RCU is not set
+# CONFIG_TREE_RCU_TRACE is not set
+# CONFIG_PREEMPT_RCU_TRACE is not set
+# CONFIG_IKCONFIG is not set
+CONFIG_LOG_BUF_SHIFT=17
+# CONFIG_GROUP_SCHED is not set
+# CONFIG_CGROUPS is not set
+# CONFIG_SYSFS_DEPRECATED_V2 is not set
+CONFIG_RELAY=y
+CONFIG_NAMESPACES=y
+CONFIG_UTS_NS=y
+CONFIG_IPC_NS=y
+# CONFIG_USER_NS is not set
+# CONFIG_PID_NS is not set
+# CONFIG_NET_NS is not set
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_INITRAMFS_SOURCE=""
+CONFIG_RD_GZIP=y
+CONFIG_RD_BZIP2=y
+CONFIG_RD_LZMA=y
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_SYSCTL=y
+CONFIG_ANON_INODES=y
+# CONFIG_EMBEDDED is not set
+CONFIG_UID16=y
+CONFIG_SYSCTL_SYSCALL=y
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_ALL is not set
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_STRIP_ASM_SYMS=y
+CONFIG_HOTPLUG=y
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+CONFIG_BASE_FULL=y
+CONFIG_FUTEX=y
+CONFIG_EPOLL=y
+CONFIG_SIGNALFD=y
+CONFIG_TIMERFD=y
+CONFIG_EVENTFD=y
+CONFIG_SHMEM=y
+CONFIG_AIO=y
+CONFIG_VM_EVENT_COUNTERS=y
+CONFIG_SLUB_DEBUG=y
+# CONFIG_COMPAT_BRK is not set
+# CONFIG_SLAB is not set
+CONFIG_SLUB=y
+# CONFIG_SLOB is not set
+# CONFIG_PROFILING is not set
+# CONFIG_MARKERS is not set
+CONFIG_HAVE_OPROFILE=y
+# CONFIG_KPROBES is not set
+CONFIG_HAVE_KPROBES=y
+CONFIG_HAVE_KRETPROBES=y
+CONFIG_HAVE_CLK=y
+# CONFIG_SLOW_WORK is not set
+CONFIG_HAVE_GENERIC_DMA_COHERENT=y
+CONFIG_SLABINFO=y
+CONFIG_RT_MUTEXES=y
+CONFIG_BASE_SMALL=0
+CONFIG_MODULES=y
+CONFIG_MODULE_FORCE_LOAD=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODULE_FORCE_UNLOAD=y
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+CONFIG_BLOCK=y
+CONFIG_LBD=y
+# CONFIG_BLK_DEV_BSG is not set
+CONFIG_BLK_DEV_INTEGRITY=y
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+CONFIG_IOSCHED_AS=y
+CONFIG_IOSCHED_DEADLINE=y
+CONFIG_IOSCHED_CFQ=y
+CONFIG_DEFAULT_AS=y
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+# CONFIG_DEFAULT_NOOP is not set
+CONFIG_DEFAULT_IOSCHED="anticipatory"
+CONFIG_FREEZER=y
+
+#
+# System Type
+#
+# CONFIG_ARCH_AAEC2000 is not set
+# CONFIG_ARCH_INTEGRATOR is not set
+# CONFIG_ARCH_REALVIEW is not set
+# CONFIG_ARCH_VERSATILE is not set
+# CONFIG_ARCH_AT91 is not set
+# CONFIG_ARCH_CLPS711X is not set
+# CONFIG_ARCH_EBSA110 is not set
+# CONFIG_ARCH_EP93XX is not set
+# CONFIG_ARCH_GEMINI is not set
+# CONFIG_ARCH_FOOTBRIDGE is not set
+# CONFIG_ARCH_NETX is not set
+# CONFIG_ARCH_H720X is not set
+# CONFIG_ARCH_IMX is not set
+# CONFIG_ARCH_IOP13XX is not set
+# CONFIG_ARCH_IOP32X is not set
+# CONFIG_ARCH_IOP33X is not set
+# CONFIG_ARCH_IXP23XX is not set
+# CONFIG_ARCH_IXP2000 is not set
+# CONFIG_ARCH_IXP4XX is not set
+# CONFIG_ARCH_L7200 is not set
+# CONFIG_ARCH_KIRKWOOD is not set
+# CONFIG_ARCH_KS8695 is not set
+# CONFIG_ARCH_NS9XXX is not set
+# CONFIG_ARCH_LOKI is not set
+# CONFIG_ARCH_MV78XX0 is not set
+# CONFIG_ARCH_MXC is not set
+# CONFIG_ARCH_ORION5X is not set
+# CONFIG_ARCH_PNX4008 is not set
+# CONFIG_ARCH_PXA is not set
+# CONFIG_ARCH_MMP is not set
+# CONFIG_ARCH_RPC is not set
+# CONFIG_ARCH_SA1100 is not set
+CONFIG_ARCH_S3C2410=y
+# CONFIG_ARCH_S3C64XX is not set
+# CONFIG_ARCH_SHARK is not set
+# CONFIG_ARCH_LH7A40X is not set
+# CONFIG_ARCH_DAVINCI is not set
+# CONFIG_ARCH_OMAP is not set
+# CONFIG_ARCH_MSM is not set
+# CONFIG_ARCH_W90X900 is not set
+CONFIG_PLAT_S3C24XX=y
+CONFIG_S3C2410_CLOCK=y
+CONFIG_CPU_S3C244X=y
+CONFIG_S3C24XX_PWM=y
+CONFIG_S3C24XX_GPIO_EXTRA=0
+CONFIG_S3C2410_DMA=y
+# CONFIG_S3C2410_DMA_DEBUG is not set
+CONFIG_S3C24XX_ADC=y
+CONFIG_PLAT_S3C=y
+CONFIG_CPU_LLSERIAL_S3C2440_ONLY=y
+CONFIG_CPU_LLSERIAL_S3C2440=y
+
+#
+# Boot options
+#
+# CONFIG_S3C_BOOT_WATCHDOG is not set
+# CONFIG_S3C_BOOT_ERROR_RESET is not set
+CONFIG_S3C_BOOT_UART_FORCE_FIFO=y
+
+#
+# Power management
+#
+# CONFIG_S3C2410_PM_DEBUG is not set
+# CONFIG_S3C2410_PM_CHECK is not set
+CONFIG_S3C_LOWLEVEL_UART_PORT=0
+CONFIG_S3C_GPIO_SPACE=0
+
+#
+# S3C2400 Machines
+#
+CONFIG_S3C2410_PM=y
+CONFIG_S3C2410_GPIO=y
+
+#
+# S3C2410 Machines
+#
+# CONFIG_ARCH_SMDK2410 is not set
+# CONFIG_ARCH_H1940 is not set
+# CONFIG_MACH_N30 is not set
+# CONFIG_ARCH_BAST is not set
+# CONFIG_MACH_OTOM is not set
+# CONFIG_MACH_AML_M5900 is not set
+# CONFIG_MACH_TCT_HAMMER is not set
+# CONFIG_MACH_VR1000 is not set
+# CONFIG_MACH_QT2410 is not set
+
+#
+# S3C2412 Machines
+#
+# CONFIG_MACH_JIVE is not set
+# CONFIG_MACH_SMDK2413 is not set
+# CONFIG_MACH_SMDK2412 is not set
+# CONFIG_MACH_VSTMS is not set
+CONFIG_CPU_S3C2440=y
+CONFIG_S3C2440_DMA=y
+
+#
+# S3C2440 Machines
+#
+# CONFIG_MACH_ANUBIS is not set
+# CONFIG_MACH_OSIRIS is not set
+# CONFIG_MACH_RX3715 is not set
+# CONFIG_ARCH_S3C2440 is not set
+# CONFIG_MACH_NEXCODER_2440 is not set
+# CONFIG_MACH_AT2440EVB is not set
+CONFIG_MACH_MINI2440=y
+
+#
+# S3C2442 Machines
+#
+
+#
+# S3C2443 Machines
+#
+# CONFIG_MACH_SMDK2443 is not set
+
+#
+# Processor Type
+#
+CONFIG_CPU_32=y
+CONFIG_CPU_ARM920T=y
+CONFIG_CPU_32v4T=y
+CONFIG_CPU_ABRT_EV4T=y
+CONFIG_CPU_PABRT_NOIFAR=y
+CONFIG_CPU_CACHE_V4WT=y
+CONFIG_CPU_CACHE_VIVT=y
+CONFIG_CPU_COPY_V4WB=y
+CONFIG_CPU_TLB_V4WBI=y
+CONFIG_CPU_CP15=y
+CONFIG_CPU_CP15_MMU=y
+
+#
+# Processor Features
+#
+CONFIG_ARM_THUMB=y
+# CONFIG_CPU_ICACHE_DISABLE is not set
+# CONFIG_CPU_DCACHE_DISABLE is not set
+# CONFIG_CPU_DCACHE_WRITETHROUGH is not set
+# CONFIG_OUTER_CACHE is not set
+
+#
+# Bus support
+#
+# CONFIG_PCI_SYSCALL is not set
+# CONFIG_ARCH_SUPPORTS_MSI is not set
+# CONFIG_PCCARD is not set
+
+#
+# Kernel Features
+#
+CONFIG_VMSPLIT_3G=y
+# CONFIG_VMSPLIT_2G is not set
+# CONFIG_VMSPLIT_1G is not set
+CONFIG_PAGE_OFFSET=0xC0000000
+# CONFIG_PREEMPT is not set
+CONFIG_HZ=200
+CONFIG_AEABI=y
+# CONFIG_OABI_COMPAT is not set
+CONFIG_ARCH_FLATMEM_HAS_HOLES=y
+# CONFIG_ARCH_SPARSEMEM_DEFAULT is not set
+# CONFIG_ARCH_SELECT_MEMORY_MODEL is not set
+# CONFIG_HIGHMEM is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+CONFIG_PAGEFLAGS_EXTENDED=y
+CONFIG_SPLIT_PTLOCK_CPUS=4096
+# CONFIG_PHYS_ADDR_T_64BIT is not set
+CONFIG_ZONE_DMA_FLAG=0
+CONFIG_VIRT_TO_BUS=y
+CONFIG_UNEVICTABLE_LRU=y
+CONFIG_HAVE_MLOCK=y
+CONFIG_HAVE_MLOCKED_PAGE_BIT=y
+CONFIG_ALIGNMENT_TRAP=y
+
+#
+# Boot options
+#
+CONFIG_ZBOOT_ROM_TEXT=0
+CONFIG_ZBOOT_ROM_BSS=0
+CONFIG_CMDLINE=""
+# CONFIG_XIP_KERNEL is not set
+CONFIG_KEXEC=y
+CONFIG_ATAGS_PROC=y
+
+#
+# CPU Power Management
+#
+CONFIG_CPU_IDLE=y
+CONFIG_CPU_IDLE_GOV_LADDER=y
+
+#
+# Floating point emulation
+#
+
+#
+# At least one emulation must be selected
+#
+
+#
+# Userspace binary formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set
+CONFIG_HAVE_AOUT=y
+CONFIG_BINFMT_AOUT=m
+CONFIG_BINFMT_MISC=m
+
+#
+# Power management options
+#
+CONFIG_PM=y
+# CONFIG_PM_DEBUG is not set
+CONFIG_PM_SLEEP=y
+CONFIG_SUSPEND=y
+CONFIG_SUSPEND_FREEZER=y
+CONFIG_APM_EMULATION=y
+CONFIG_ARCH_SUSPEND_POSSIBLE=y
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+CONFIG_PACKET_MMAP=y
+CONFIG_UNIX=y
+CONFIG_XFRM=y
+CONFIG_XFRM_USER=m
+# CONFIG_XFRM_SUB_POLICY is not set
+# CONFIG_XFRM_MIGRATE is not set
+# CONFIG_XFRM_STATISTICS is not set
+CONFIG_NET_KEY=m
+# CONFIG_NET_KEY_MIGRATE is not set
+CONFIG_INET=y
+CONFIG_IP_MULTICAST=y
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_ASK_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
+CONFIG_IP_FIB_HASH=y
+CONFIG_IP_MULTIPLE_TABLES=y
+CONFIG_IP_ROUTE_MULTIPATH=y
+CONFIG_IP_ROUTE_VERBOSE=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+CONFIG_IP_PNP_RARP=y
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+CONFIG_IP_MROUTE=y
+CONFIG_IP_PIMSM_V1=y
+CONFIG_IP_PIMSM_V2=y
+# CONFIG_ARPD is not set
+CONFIG_SYN_COOKIES=y
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
+# CONFIG_INET_LRO is not set
+CONFIG_INET_DIAG=m
+CONFIG_INET_TCP_DIAG=m
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_CUBIC=y
+CONFIG_DEFAULT_TCP_CONG="cubic"
+# CONFIG_TCP_MD5SIG is not set
+# CONFIG_IPV6 is not set
+# CONFIG_NETWORK_SECMARK is not set
+CONFIG_NETFILTER=y
+# CONFIG_NETFILTER_DEBUG is not set
+CONFIG_NETFILTER_ADVANCED=y
+CONFIG_BRIDGE_NETFILTER=y
+
+#
+# Core Netfilter Configuration
+#
+# CONFIG_NETFILTER_NETLINK_QUEUE is not set
+# CONFIG_NETFILTER_NETLINK_LOG is not set
+# CONFIG_NF_CONNTRACK is not set
+# CONFIG_NETFILTER_XTABLES is not set
+# CONFIG_IP_VS is not set
+
+#
+# IP: Netfilter Configuration
+#
+# CONFIG_NF_DEFRAG_IPV4 is not set
+# CONFIG_IP_NF_QUEUE is not set
+# CONFIG_IP_NF_IPTABLES is not set
+# CONFIG_IP_NF_ARPTABLES is not set
+# CONFIG_BRIDGE_NF_EBTABLES is not set
+# CONFIG_IP_DCCP is not set
+# CONFIG_IP_SCTP is not set
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+CONFIG_STP=m
+CONFIG_GARP=m
+CONFIG_BRIDGE=m
+# CONFIG_NET_DSA is not set
+CONFIG_VLAN_8021Q=m
+CONFIG_VLAN_8021Q_GVRP=y
+# CONFIG_DECNET is not set
+CONFIG_LLC=m
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+# CONFIG_PHONET is not set
+# CONFIG_NET_SCHED is not set
+# CONFIG_DCB is not set
+
+#
+# Network testing
+#
+CONFIG_NET_PKTGEN=m
+# CONFIG_HAMRADIO is not set
+# CONFIG_CAN is not set
+# CONFIG_IRDA is not set
+CONFIG_BT=m
+CONFIG_BT_L2CAP=m
+CONFIG_BT_SCO=m
+CONFIG_BT_RFCOMM=m
+CONFIG_BT_RFCOMM_TTY=y
+CONFIG_BT_BNEP=m
+CONFIG_BT_BNEP_MC_FILTER=y
+CONFIG_BT_BNEP_PROTO_FILTER=y
+CONFIG_BT_HIDP=m
+
+#
+# Bluetooth device drivers
+#
+CONFIG_BT_HCIBTUSB=m
+CONFIG_BT_HCIBTSDIO=m
+CONFIG_BT_HCIUART=m
+CONFIG_BT_HCIUART_H4=y
+CONFIG_BT_HCIUART_BCSP=y
+CONFIG_BT_HCIUART_LL=y
+CONFIG_BT_HCIBCM203X=m
+CONFIG_BT_HCIBPA10X=m
+CONFIG_BT_HCIBFUSB=m
+CONFIG_BT_HCIVHCI=m
+# CONFIG_AF_RXRPC is not set
+CONFIG_FIB_RULES=y
+CONFIG_WIRELESS=y
+CONFIG_CFG80211=m
+CONFIG_CFG80211_REG_DEBUG=y
+CONFIG_WIRELESS_OLD_REGULATORY=y
+CONFIG_WIRELESS_EXT=y
+CONFIG_WIRELESS_EXT_SYSFS=y
+CONFIG_LIB80211=m
+CONFIG_LIB80211_CRYPT_WEP=m
+CONFIG_LIB80211_CRYPT_CCMP=m
+CONFIG_LIB80211_CRYPT_TKIP=m
+# CONFIG_LIB80211_DEBUG is not set
+CONFIG_MAC80211=m
+
+#
+# Rate control algorithm selection
+#
+CONFIG_MAC80211_RC_MINSTREL=y
+# CONFIG_MAC80211_RC_DEFAULT_PID is not set
+CONFIG_MAC80211_RC_DEFAULT_MINSTREL=y
+CONFIG_MAC80211_RC_DEFAULT="minstrel"
+CONFIG_MAC80211_MESH=y
+CONFIG_MAC80211_LEDS=y
+# CONFIG_MAC80211_DEBUGFS is not set
+# CONFIG_MAC80211_DEBUG_MENU is not set
+# CONFIG_WIMAX is not set
+# CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=y
+# CONFIG_FIRMWARE_IN_KERNEL is not set
+CONFIG_EXTRA_FIRMWARE=""
+# CONFIG_DEBUG_DRIVER is not set
+# CONFIG_DEBUG_DEVRES is not set
+# CONFIG_SYS_HYPERVISOR is not set
+CONFIG_CONNECTOR=m
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+CONFIG_MTD_CONCAT=y
+CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_TESTS is not set
+# CONFIG_MTD_REDBOOT_PARTS is not set
+CONFIG_MTD_CMDLINE_PARTS=y
+# CONFIG_MTD_AFS_PARTS is not set
+# CONFIG_MTD_AR7_PARTS is not set
+
+#
+# User Modules And Translation Layers
+#
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLKDEVS=y
+CONFIG_MTD_BLOCK=y
+CONFIG_FTL=y
+CONFIG_NFTL=y
+CONFIG_NFTL_RW=y
+CONFIG_INFTL=y
+CONFIG_RFD_FTL=y
+# CONFIG_SSFDC is not set
+# CONFIG_MTD_OOPS is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+CONFIG_MTD_CFI=y
+CONFIG_MTD_JEDECPROBE=y
+CONFIG_MTD_GEN_PROBE=y
+# CONFIG_MTD_CFI_ADV_OPTIONS is not set
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+# CONFIG_MTD_CFI_INTELEXT is not set
+CONFIG_MTD_CFI_AMDSTD=y
+CONFIG_MTD_CFI_STAA=y
+CONFIG_MTD_CFI_UTIL=y
+CONFIG_MTD_RAM=y
+CONFIG_MTD_ROM=y
+# CONFIG_MTD_ABSENT is not set
+
+#
+# Mapping drivers for chip access
+#
+# CONFIG_MTD_COMPLEX_MAPPINGS is not set
+# CONFIG_MTD_PHYSMAP is not set
+# CONFIG_MTD_ARM_INTEGRATOR is not set
+# CONFIG_MTD_IMPA7 is not set
+# CONFIG_MTD_PLATRAM is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_DATAFLASH is not set
+# CONFIG_MTD_M25P80 is not set
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+# CONFIG_MTD_BLOCK2MTD is not set
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+CONFIG_MTD_NAND=y
+CONFIG_MTD_NAND_VERIFY_WRITE=y
+# CONFIG_MTD_NAND_ECC_SMC is not set
+# CONFIG_MTD_NAND_MUSEUM_IDS is not set
+# CONFIG_MTD_NAND_GPIO is not set
+CONFIG_MTD_NAND_IDS=y
+CONFIG_MTD_NAND_S3C2410=y
+# CONFIG_MTD_NAND_S3C2410_DEBUG is not set
+# CONFIG_MTD_NAND_S3C2410_HWECC is not set
+# CONFIG_MTD_NAND_S3C2410_CLKSTOP is not set
+# CONFIG_MTD_NAND_DISKONCHIP is not set
+# CONFIG_MTD_NAND_NANDSIM is not set
+CONFIG_MTD_NAND_PLATFORM=y
+# CONFIG_MTD_ALAUDA is not set
+# CONFIG_MTD_ONENAND is not set
+
+#
+# LPDDR flash memory drivers
+#
+CONFIG_MTD_LPDDR=y
+CONFIG_MTD_QINFO_PROBE=y
+
+#
+# UBI - Unsorted block images
+#
+# CONFIG_MTD_UBI is not set
+# CONFIG_PARPORT is not set
+CONFIG_BLK_DEV=y
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=m
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+CONFIG_BLK_DEV_NBD=m
+# CONFIG_BLK_DEV_UB is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=65536
+# CONFIG_BLK_DEV_XIP is not set
+CONFIG_CDROM_PKTCDVD=m
+CONFIG_CDROM_PKTCDVD_BUFFERS=8
+# CONFIG_CDROM_PKTCDVD_WCACHE is not set
+# CONFIG_ATA_OVER_ETH is not set
+CONFIG_MISC_DEVICES=y
+# CONFIG_ICS932S401 is not set
+# CONFIG_ENCLOSURE_SERVICES is not set
+# CONFIG_ISL29003 is not set
+# CONFIG_C2PORT is not set
+
+#
+# EEPROM support
+#
+CONFIG_EEPROM_AT24=y
+# CONFIG_EEPROM_AT25 is not set
+# CONFIG_EEPROM_LEGACY is not set
+# CONFIG_EEPROM_93CX6 is not set
+CONFIG_HAVE_IDE=y
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+CONFIG_SCSI=m
+CONFIG_SCSI_DMA=y
+# CONFIG_SCSI_TGT is not set
+# CONFIG_SCSI_NETLINK is not set
+# CONFIG_SCSI_PROC_FS is not set
+
+#
+# SCSI support type (disk, tape, CD-ROM)
+#
+CONFIG_BLK_DEV_SD=m
+# CONFIG_CHR_DEV_ST is not set
+# CONFIG_CHR_DEV_OSST is not set
+# CONFIG_BLK_DEV_SR is not set
+CONFIG_CHR_DEV_SG=m
+# CONFIG_CHR_DEV_SCH is not set
+
+#
+# Some SCSI devices (e.g. CD jukebox) support multiple LUNs
+#
+# CONFIG_SCSI_MULTI_LUN is not set
+# CONFIG_SCSI_CONSTANTS is not set
+# CONFIG_SCSI_LOGGING is not set
+# CONFIG_SCSI_SCAN_ASYNC is not set
+CONFIG_SCSI_WAIT_SCAN=m
+
+#
+# SCSI Transports
+#
+# CONFIG_SCSI_SPI_ATTRS is not set
+# CONFIG_SCSI_FC_ATTRS is not set
+# CONFIG_SCSI_ISCSI_ATTRS is not set
+# CONFIG_SCSI_SAS_LIBSAS is not set
+# CONFIG_SCSI_SRP_ATTRS is not set
+# CONFIG_SCSI_LOWLEVEL is not set
+# CONFIG_SCSI_DH is not set
+# CONFIG_SCSI_OSD_INITIATOR is not set
+# CONFIG_ATA is not set
+# CONFIG_MD is not set
+CONFIG_NETDEVICES=y
+CONFIG_COMPAT_NET_DEV_OPS=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
+# CONFIG_EQUALIZER is not set
+CONFIG_TUN=m
+# CONFIG_VETH is not set
+# CONFIG_PHYLIB is not set
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+# CONFIG_AX88796 is not set
+# CONFIG_SMC91X is not set
+CONFIG_DM9000=y
+CONFIG_DM9000_DEBUGLEVEL=4
+# CONFIG_DM9000_FORCE_SIMPLE_PHY_POLL is not set
+# CONFIG_ENC28J60 is not set
+# CONFIG_ETHOC is not set
+# CONFIG_SMC911X is not set
+# CONFIG_SMSC911X is not set
+# CONFIG_DNET is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_IBM_NEW_EMAC_NO_FLOW_CTRL is not set
+# CONFIG_IBM_NEW_EMAC_MAL_CLR_ICINTSTAT is not set
+# CONFIG_IBM_NEW_EMAC_MAL_COMMON_ERR is not set
+# CONFIG_B44 is not set
+# CONFIG_NETDEV_1000 is not set
+# CONFIG_NETDEV_10000 is not set
+
+#
+# Wireless LAN
+#
+# CONFIG_WLAN_PRE80211 is not set
+CONFIG_WLAN_80211=y
+CONFIG_LIBERTAS=m
+# CONFIG_LIBERTAS_USB is not set
+CONFIG_LIBERTAS_SDIO=m
+# CONFIG_LIBERTAS_SPI is not set
+# CONFIG_LIBERTAS_DEBUG is not set
+# CONFIG_LIBERTAS_THINFIRM is not set
+# CONFIG_AT76C50X_USB is not set
+# CONFIG_USB_ZD1201 is not set
+# CONFIG_USB_NET_RNDIS_WLAN is not set
+# CONFIG_RTL8187 is not set
+# CONFIG_MAC80211_HWSIM is not set
+# CONFIG_P54_COMMON is not set
+# CONFIG_AR9170_USB is not set
+CONFIG_HOSTAP=m
+CONFIG_HOSTAP_FIRMWARE=y
+CONFIG_HOSTAP_FIRMWARE_NVRAM=y
+# CONFIG_B43 is not set
+# CONFIG_B43LEGACY is not set
+CONFIG_ZD1211RW=m
+CONFIG_ZD1211RW_DEBUG=y
+# CONFIG_RT2X00 is not set
+
+#
+# Enable WiMAX (Networking options) to see the WiMAX drivers
+#
+
+#
+# USB Network Adapters
+#
+# CONFIG_USB_CATC is not set
+# CONFIG_USB_KAWETH is not set
+# CONFIG_USB_PEGASUS is not set
+# CONFIG_USB_RTL8150 is not set
+# CONFIG_USB_USBNET is not set
+# CONFIG_WAN is not set
+CONFIG_PPP=m
+CONFIG_PPP_MULTILINK=y
+CONFIG_PPP_FILTER=y
+CONFIG_PPP_ASYNC=m
+CONFIG_PPP_SYNC_TTY=m
+CONFIG_PPP_DEFLATE=m
+CONFIG_PPP_BSDCOMP=m
+CONFIG_PPP_MPPE=m
+# CONFIG_PPPOE is not set
+# CONFIG_PPPOL2TP is not set
+# CONFIG_SLIP is not set
+CONFIG_SLHC=m
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_ISDN is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+CONFIG_INPUT_FF_MEMLESS=y
+# CONFIG_INPUT_POLLDEV is not set
+
+#
+# Userland interfaces
+#
+CONFIG_INPUT_MOUSEDEV=y
+CONFIG_INPUT_MOUSEDEV_PSAUX=y
+CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
+CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
+# CONFIG_INPUT_JOYDEV is not set
+CONFIG_INPUT_EVDEV=y
+CONFIG_INPUT_EVBUG=m
+
+#
+# Input Device Drivers
+#
+CONFIG_INPUT_KEYBOARD=y
+# CONFIG_KEYBOARD_ATKBD is not set
+# CONFIG_KEYBOARD_SUNKBD is not set
+# CONFIG_KEYBOARD_LKKBD is not set
+# CONFIG_KEYBOARD_XTKBD is not set
+# CONFIG_KEYBOARD_NEWTON is not set
+# CONFIG_KEYBOARD_STOWAWAY is not set
+CONFIG_KEYBOARD_GPIO=y
+CONFIG_INPUT_MOUSE=y
+CONFIG_MOUSE_PS2=y
+CONFIG_MOUSE_PS2_ALPS=y
+CONFIG_MOUSE_PS2_LOGIPS2PP=y
+CONFIG_MOUSE_PS2_SYNAPTICS=y
+CONFIG_MOUSE_PS2_TRACKPOINT=y
+# CONFIG_MOUSE_PS2_ELANTECH is not set
+# CONFIG_MOUSE_PS2_TOUCHKIT is not set
+# CONFIG_MOUSE_SERIAL is not set
+# CONFIG_MOUSE_APPLETOUCH is not set
+# CONFIG_MOUSE_BCM5974 is not set
+# CONFIG_MOUSE_VSXXXAA is not set
+# CONFIG_MOUSE_GPIO is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TABLET is not set
+CONFIG_INPUT_TOUCHSCREEN=y
+# CONFIG_TOUCHSCREEN_ADS7846 is not set
+# CONFIG_TOUCHSCREEN_AD7877 is not set
+# CONFIG_TOUCHSCREEN_AD7879_I2C is not set
+# CONFIG_TOUCHSCREEN_AD7879_SPI is not set
+# CONFIG_TOUCHSCREEN_AD7879 is not set
+# CONFIG_TOUCHSCREEN_FUJITSU is not set
+# CONFIG_TOUCHSCREEN_GUNZE is not set
+# CONFIG_TOUCHSCREEN_ELO is not set
+# CONFIG_TOUCHSCREEN_WACOM_W8001 is not set
+# CONFIG_TOUCHSCREEN_MTOUCH is not set
+# CONFIG_TOUCHSCREEN_INEXIO is not set
+# CONFIG_TOUCHSCREEN_MK712 is not set
+# CONFIG_TOUCHSCREEN_PENMOUNT is not set
+# CONFIG_TOUCHSCREEN_TOUCHRIGHT is not set
+# CONFIG_TOUCHSCREEN_TOUCHWIN is not set
+# CONFIG_TOUCHSCREEN_USB_COMPOSITE is not set
+# CONFIG_TOUCHSCREEN_TOUCHIT213 is not set
+# CONFIG_TOUCHSCREEN_TSC2007 is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+CONFIG_SERIO=y
+CONFIG_SERIO_SERPORT=y
+CONFIG_SERIO_LIBPS2=y
+CONFIG_SERIO_RAW=y
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+CONFIG_VT=y
+CONFIG_CONSOLE_TRANSLATIONS=y
+CONFIG_VT_CONSOLE=y
+CONFIG_HW_CONSOLE=y
+CONFIG_VT_HW_CONSOLE_BINDING=y
+CONFIG_DEVKMEM=y
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SAMSUNG=y
+CONFIG_SERIAL_SAMSUNG_UARTS=3
+CONFIG_SERIAL_SAMSUNG_CONSOLE=y
+CONFIG_SERIAL_S3C2440=y
+# CONFIG_SERIAL_MAX3100 is not set
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+CONFIG_UNIX98_PTYS=y
+# CONFIG_DEVPTS_MULTIPLE_INSTANCES is not set
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=128
+CONFIG_IPMI_HANDLER=m
+# CONFIG_IPMI_PANIC_EVENT is not set
+CONFIG_IPMI_DEVICE_INTERFACE=m
+CONFIG_IPMI_SI=m
+CONFIG_IPMI_WATCHDOG=m
+CONFIG_IPMI_POWEROFF=m
+CONFIG_HW_RANDOM=y
+# CONFIG_HW_RANDOM_TIMERIOMEM is not set
+# CONFIG_R3964 is not set
+# CONFIG_RAW_DRIVER is not set
+# CONFIG_TCG_TPM is not set
+CONFIG_I2C=y
+CONFIG_I2C_BOARDINFO=y
+CONFIG_I2C_CHARDEV=y
+CONFIG_I2C_HELPER_AUTO=y
+CONFIG_I2C_ALGOBIT=y
+
+#
+# I2C Hardware Bus support
+#
+
+#
+# I2C system bus drivers (mostly embedded / system-on-chip)
+#
+# CONFIG_I2C_GPIO is not set
+# CONFIG_I2C_OCORES is not set
+CONFIG_I2C_S3C2410=y
+CONFIG_I2C_SIMTEC=y
+
+#
+# External I2C/SMBus adapter drivers
+#
+# CONFIG_I2C_PARPORT_LIGHT is not set
+# CONFIG_I2C_TAOS_EVM is not set
+# CONFIG_I2C_TINY_USB is not set
+
+#
+# Other I2C/SMBus bus drivers
+#
+# CONFIG_I2C_PCA_PLATFORM is not set
+# CONFIG_I2C_STUB is not set
+
+#
+# Miscellaneous I2C Chip support
+#
+# CONFIG_DS1682 is not set
+# CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_PCF8575 is not set
+# CONFIG_SENSORS_PCA9539 is not set
+# CONFIG_SENSORS_MAX6875 is not set
+CONFIG_SENSORS_TSL2550=m
+# CONFIG_I2C_DEBUG_CORE is not set
+# CONFIG_I2C_DEBUG_ALGO is not set
+# CONFIG_I2C_DEBUG_BUS is not set
+# CONFIG_I2C_DEBUG_CHIP is not set
+CONFIG_SPI=y
+# CONFIG_SPI_DEBUG is not set
+CONFIG_SPI_MASTER=y
+
+#
+# SPI Master Controller Drivers
+#
+CONFIG_SPI_BITBANG=y
+# CONFIG_SPI_GPIO is not set
+CONFIG_SPI_S3C24XX=y
+# CONFIG_SPI_S3C24XX_GPIO is not set
+
+#
+# SPI Protocol Masters
+#
+CONFIG_SPI_SPIDEV=y
+# CONFIG_SPI_TLE62X0 is not set
+CONFIG_ARCH_REQUIRE_GPIOLIB=y
+CONFIG_GPIOLIB=y
+# CONFIG_DEBUG_GPIO is not set
+CONFIG_GPIO_SYSFS=y
+
+#
+# Memory mapped GPIO expanders:
+#
+
+#
+# I2C GPIO expanders:
+#
+# CONFIG_GPIO_MAX732X is not set
+# CONFIG_GPIO_PCA953X is not set
+# CONFIG_GPIO_PCF857X is not set
+
+#
+# PCI GPIO expanders:
+#
+
+#
+# SPI GPIO expanders:
+#
+# CONFIG_GPIO_MAX7301 is not set
+# CONFIG_GPIO_MCP23S08 is not set
+# CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
+CONFIG_HWMON=y
+# CONFIG_HWMON_VID is not set
+# CONFIG_SENSORS_AD7414 is not set
+# CONFIG_SENSORS_AD7418 is not set
+# CONFIG_SENSORS_ADCXX is not set
+# CONFIG_SENSORS_ADM1021 is not set
+# CONFIG_SENSORS_ADM1025 is not set
+# CONFIG_SENSORS_ADM1026 is not set
+# CONFIG_SENSORS_ADM1029 is not set
+# CONFIG_SENSORS_ADM1031 is not set
+# CONFIG_SENSORS_ADM9240 is not set
+# CONFIG_SENSORS_ADT7462 is not set
+# CONFIG_SENSORS_ADT7470 is not set
+# CONFIG_SENSORS_ADT7473 is not set
+# CONFIG_SENSORS_ADT7475 is not set
+# CONFIG_SENSORS_ATXP1 is not set
+# CONFIG_SENSORS_DS1621 is not set
+# CONFIG_SENSORS_F71805F is not set
+# CONFIG_SENSORS_F71882FG is not set
+# CONFIG_SENSORS_F75375S is not set
+# CONFIG_SENSORS_G760A is not set
+# CONFIG_SENSORS_GL518SM is not set
+# CONFIG_SENSORS_GL520SM is not set
+# CONFIG_SENSORS_IBMAEM is not set
+# CONFIG_SENSORS_IBMPEX is not set
+# CONFIG_SENSORS_IT87 is not set
+# CONFIG_SENSORS_LM63 is not set
+# CONFIG_SENSORS_LM70 is not set
+CONFIG_SENSORS_LM75=y
+# CONFIG_SENSORS_LM77 is not set
+# CONFIG_SENSORS_LM78 is not set
+# CONFIG_SENSORS_LM80 is not set
+# CONFIG_SENSORS_LM83 is not set
+# CONFIG_SENSORS_LM85 is not set
+# CONFIG_SENSORS_LM87 is not set
+# CONFIG_SENSORS_LM90 is not set
+# CONFIG_SENSORS_LM92 is not set
+# CONFIG_SENSORS_LM93 is not set
+# CONFIG_SENSORS_LTC4215 is not set
+# CONFIG_SENSORS_LTC4245 is not set
+# CONFIG_SENSORS_LM95241 is not set
+# CONFIG_SENSORS_MAX1111 is not set
+# CONFIG_SENSORS_MAX1619 is not set
+# CONFIG_SENSORS_MAX6650 is not set
+# CONFIG_SENSORS_PC87360 is not set
+# CONFIG_SENSORS_PC87427 is not set
+# CONFIG_SENSORS_PCF8591 is not set
+# CONFIG_SENSORS_SHT15 is not set
+# CONFIG_SENSORS_DME1737 is not set
+# CONFIG_SENSORS_SMSC47M1 is not set
+# CONFIG_SENSORS_SMSC47M192 is not set
+# CONFIG_SENSORS_SMSC47B397 is not set
+# CONFIG_SENSORS_ADS7828 is not set
+# CONFIG_SENSORS_THMC50 is not set
+# CONFIG_SENSORS_VT1211 is not set
+# CONFIG_SENSORS_W83781D is not set
+# CONFIG_SENSORS_W83791D is not set
+# CONFIG_SENSORS_W83792D is not set
+# CONFIG_SENSORS_W83793 is not set
+# CONFIG_SENSORS_W83L785TS is not set
+# CONFIG_SENSORS_W83L786NG is not set
+# CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
+# CONFIG_SENSORS_LIS3_SPI is not set
+# CONFIG_HWMON_DEBUG_CHIP is not set
+CONFIG_THERMAL=m
+# CONFIG_THERMAL_HWMON is not set
+CONFIG_WATCHDOG=y
+# CONFIG_WATCHDOG_NOWAYOUT is not set
+
+#
+# Watchdog Device Drivers
+#
+# CONFIG_SOFT_WATCHDOG is not set
+CONFIG_S3C2410_WATCHDOG=y
+
+#
+# USB-based Watchdog Cards
+#
+# CONFIG_USBPCWATCHDOG is not set
+CONFIG_SSB_POSSIBLE=y
+
+#
+# Sonics Silicon Backplane
+#
+# CONFIG_SSB is not set
+
+#
+# Multifunction device drivers
+#
+# CONFIG_MFD_CORE is not set
+# CONFIG_MFD_SM501 is not set
+# CONFIG_MFD_ASIC3 is not set
+# CONFIG_HTC_EGPIO is not set
+# CONFIG_HTC_PASIC3 is not set
+# CONFIG_TPS65010 is not set
+# CONFIG_TWL4030_CORE is not set
+# CONFIG_MFD_TMIO is not set
+# CONFIG_MFD_T7L66XB is not set
+# CONFIG_MFD_TC6387XB is not set
+# CONFIG_MFD_TC6393XB is not set
+# CONFIG_PMIC_DA903X is not set
+# CONFIG_MFD_WM8400 is not set
+# CONFIG_MFD_WM8350_I2C is not set
+# CONFIG_MFD_PCF50633 is not set
+
+#
+# Multimedia devices
+#
+
+#
+# Multimedia core support
+#
+CONFIG_VIDEO_DEV=m
+CONFIG_VIDEO_V4L2_COMMON=m
+CONFIG_VIDEO_ALLOW_V4L1=y
+CONFIG_VIDEO_V4L1_COMPAT=y
+CONFIG_DVB_CORE=m
+CONFIG_VIDEO_MEDIA=m
+
+#
+# Multimedia drivers
+#
+# CONFIG_MEDIA_ATTACH is not set
+CONFIG_MEDIA_TUNER=m
+# CONFIG_MEDIA_TUNER_CUSTOMISE is not set
+CONFIG_MEDIA_TUNER_SIMPLE=m
+CONFIG_MEDIA_TUNER_TDA8290=m
+CONFIG_MEDIA_TUNER_TDA9887=m
+CONFIG_MEDIA_TUNER_TEA5761=m
+CONFIG_MEDIA_TUNER_TEA5767=m
+CONFIG_MEDIA_TUNER_MT20XX=m
+CONFIG_MEDIA_TUNER_XC2028=m
+CONFIG_MEDIA_TUNER_XC5000=m
+CONFIG_MEDIA_TUNER_MC44S803=m
+CONFIG_VIDEO_V4L2=m
+CONFIG_VIDEO_V4L1=m
+CONFIG_VIDEOBUF_GEN=m
+CONFIG_VIDEO_CAPTURE_DRIVERS=y
+# CONFIG_VIDEO_ADV_DEBUG is not set
+# CONFIG_VIDEO_FIXED_MINOR_RANGES is not set
+CONFIG_VIDEO_HELPER_CHIPS_AUTO=y
+# CONFIG_VIDEO_VIVI is not set
+# CONFIG_VIDEO_CPIA is not set
+# CONFIG_VIDEO_CPIA2 is not set
+# CONFIG_VIDEO_SAA5246A is not set
+# CONFIG_VIDEO_SAA5249 is not set
+# CONFIG_VIDEO_AU0828 is not set
+CONFIG_SOC_CAMERA=m
+# CONFIG_SOC_CAMERA_MT9M001 is not set
+# CONFIG_SOC_CAMERA_MT9M111 is not set
+# CONFIG_SOC_CAMERA_MT9T031 is not set
+# CONFIG_SOC_CAMERA_MT9V022 is not set
+# CONFIG_SOC_CAMERA_TW9910 is not set
+CONFIG_SOC_CAMERA_PLATFORM=m
+# CONFIG_SOC_CAMERA_OV772X is not set
+# CONFIG_VIDEO_SH_MOBILE_CEU is not set
+CONFIG_V4L_USB_DRIVERS=y
+# CONFIG_USB_VIDEO_CLASS is not set
+CONFIG_USB_VIDEO_CLASS_INPUT_EVDEV=y
+CONFIG_USB_GSPCA=m
+# CONFIG_USB_M5602 is not set
+# CONFIG_USB_STV06XX is not set
+# CONFIG_USB_GSPCA_CONEX is not set
+# CONFIG_USB_GSPCA_ETOMS is not set
+# CONFIG_USB_GSPCA_FINEPIX is not set
+# CONFIG_USB_GSPCA_MARS is not set
+# CONFIG_USB_GSPCA_MR97310A is not set
+# CONFIG_USB_GSPCA_OV519 is not set
+# CONFIG_USB_GSPCA_OV534 is not set
+# CONFIG_USB_GSPCA_PAC207 is not set
+# CONFIG_USB_GSPCA_PAC7311 is not set
+# CONFIG_USB_GSPCA_SONIXB is not set
+# CONFIG_USB_GSPCA_SONIXJ is not set
+# CONFIG_USB_GSPCA_SPCA500 is not set
+# CONFIG_USB_GSPCA_SPCA501 is not set
+# CONFIG_USB_GSPCA_SPCA505 is not set
+# CONFIG_USB_GSPCA_SPCA506 is not set
+# CONFIG_USB_GSPCA_SPCA508 is not set
+# CONFIG_USB_GSPCA_SPCA561 is not set
+# CONFIG_USB_GSPCA_SQ905 is not set
+# CONFIG_USB_GSPCA_SQ905C is not set
+# CONFIG_USB_GSPCA_STK014 is not set
+# CONFIG_USB_GSPCA_SUNPLUS is not set
+# CONFIG_USB_GSPCA_T613 is not set
+# CONFIG_USB_GSPCA_TV8532 is not set
+# CONFIG_USB_GSPCA_VC032X is not set
+CONFIG_USB_GSPCA_ZC3XX=m
+# CONFIG_VIDEO_PVRUSB2 is not set
+# CONFIG_VIDEO_HDPVR is not set
+# CONFIG_VIDEO_EM28XX is not set
+# CONFIG_VIDEO_CX231XX is not set
+# CONFIG_VIDEO_USBVISION is not set
+# CONFIG_USB_VICAM is not set
+# CONFIG_USB_IBMCAM is not set
+# CONFIG_USB_KONICAWC is not set
+# CONFIG_USB_QUICKCAM_MESSENGER is not set
+# CONFIG_USB_ET61X251 is not set
+# CONFIG_VIDEO_OVCAMCHIP is not set
+# CONFIG_USB_OV511 is not set
+# CONFIG_USB_SE401 is not set
+# CONFIG_USB_SN9C102 is not set
+# CONFIG_USB_STV680 is not set
+# CONFIG_USB_ZC0301 is not set
+# CONFIG_USB_PWC is not set
+# CONFIG_USB_PWC_INPUT_EVDEV is not set
+# CONFIG_USB_ZR364XX is not set
+# CONFIG_USB_STKWEBCAM is not set
+# CONFIG_USB_S2255 is not set
+CONFIG_RADIO_ADAPTERS=y
+# CONFIG_USB_DSBR is not set
+# CONFIG_USB_SI470X is not set
+# CONFIG_USB_MR800 is not set
+# CONFIG_RADIO_TEA5764 is not set
+# CONFIG_DVB_DYNAMIC_MINORS is not set
+CONFIG_DVB_CAPTURE_DRIVERS=y
+# CONFIG_TTPCI_EEPROM is not set
+
+#
+# Supported USB Adapters
+#
+# CONFIG_DVB_USB is not set
+# CONFIG_DVB_SIANO_SMS1XXX is not set
+
+#
+# Supported FlexCopII (B2C2) Adapters
+#
+# CONFIG_DVB_B2C2_FLEXCOP is not set
+
+#
+# Supported DVB Frontends
+#
+# CONFIG_DVB_FE_CUSTOMISE is not set
+# CONFIG_DAB is not set
+
+#
+# Graphics support
+#
+# CONFIG_VGASTATE is not set
+CONFIG_VIDEO_OUTPUT_CONTROL=y
+CONFIG_FB=y
+CONFIG_FIRMWARE_EDID=y
+# CONFIG_FB_DDC is not set
+# CONFIG_FB_BOOT_VESA_SUPPORT is not set
+CONFIG_FB_CFB_FILLRECT=y
+CONFIG_FB_CFB_COPYAREA=y
+CONFIG_FB_CFB_IMAGEBLIT=y
+# CONFIG_FB_CFB_REV_PIXELS_IN_BYTE is not set
+# CONFIG_FB_SYS_FILLRECT is not set
+# CONFIG_FB_SYS_COPYAREA is not set
+# CONFIG_FB_SYS_IMAGEBLIT is not set
+# CONFIG_FB_FOREIGN_ENDIAN is not set
+# CONFIG_FB_SYS_FOPS is not set
+# CONFIG_FB_SVGALIB is not set
+# CONFIG_FB_MACMODES is not set
+# CONFIG_FB_BACKLIGHT is not set
+CONFIG_FB_MODE_HELPERS=y
+CONFIG_FB_TILEBLITTING=y
+
+#
+# Frame buffer hardware drivers
+#
+# CONFIG_FB_UVESA is not set
+# CONFIG_FB_S1D13XXX is not set
+CONFIG_FB_S3C2410=y
+# CONFIG_FB_S3C2410_DEBUG is not set
+# CONFIG_FB_VIRTUAL is not set
+# CONFIG_FB_METRONOME is not set
+# CONFIG_FB_MB862XX is not set
+# CONFIG_FB_BROADSHEET is not set
+CONFIG_BACKLIGHT_LCD_SUPPORT=y
+CONFIG_LCD_CLASS_DEVICE=y
+# CONFIG_LCD_LTV350QV is not set
+# CONFIG_LCD_ILI9320 is not set
+# CONFIG_LCD_TDO24M is not set
+# CONFIG_LCD_VGG2432A4 is not set
+CONFIG_LCD_PLATFORM=y
+CONFIG_BACKLIGHT_CLASS_DEVICE=y
+# CONFIG_BACKLIGHT_GENERIC is not set
+CONFIG_BACKLIGHT_PWM=y
+
+#
+# Display device support
+#
+CONFIG_DISPLAY_SUPPORT=y
+
+#
+# Display hardware drivers
+#
+
+#
+# Console display driver support
+#
+# CONFIG_VGA_CONSOLE is not set
+CONFIG_DUMMY_CONSOLE=y
+CONFIG_FRAMEBUFFER_CONSOLE=y
+CONFIG_FRAMEBUFFER_CONSOLE_DETECT_PRIMARY=y
+CONFIG_FRAMEBUFFER_CONSOLE_ROTATION=y
+CONFIG_FONTS=y
+CONFIG_FONT_8x8=y
+# CONFIG_FONT_8x16 is not set
+# CONFIG_FONT_6x11 is not set
+# CONFIG_FONT_7x14 is not set
+# CONFIG_FONT_PEARL_8x8 is not set
+# CONFIG_FONT_ACORN_8x8 is not set
+CONFIG_FONT_MINI_4x6=y
+# CONFIG_FONT_SUN8x16 is not set
+# CONFIG_FONT_SUN12x22 is not set
+# CONFIG_FONT_10x18 is not set
+CONFIG_LOGO=y
+# CONFIG_LOGO_LINUX_MONO is not set
+# CONFIG_LOGO_LINUX_VGA16 is not set
+CONFIG_LOGO_LINUX_CLUT224=y
+CONFIG_SOUND=y
+CONFIG_SOUND_OSS_CORE=y
+CONFIG_SND=y
+CONFIG_SND_TIMER=y
+CONFIG_SND_PCM=y
+CONFIG_SND_HWDEP=m
+CONFIG_SND_RAWMIDI=m
+CONFIG_SND_JACK=y
+CONFIG_SND_SEQUENCER=m
+CONFIG_SND_SEQ_DUMMY=m
+CONFIG_SND_OSSEMUL=y
+CONFIG_SND_MIXER_OSS=m
+CONFIG_SND_PCM_OSS=m
+CONFIG_SND_PCM_OSS_PLUGINS=y
+CONFIG_SND_SEQUENCER_OSS=y
+CONFIG_SND_DYNAMIC_MINORS=y
+CONFIG_SND_SUPPORT_OLD_API=y
+CONFIG_SND_VERBOSE_PROCFS=y
+# CONFIG_SND_VERBOSE_PRINTK is not set
+# CONFIG_SND_DEBUG is not set
+# CONFIG_SND_DRIVERS is not set
+# CONFIG_SND_ARM is not set
+# CONFIG_SND_SPI is not set
+CONFIG_SND_USB=y
+CONFIG_SND_USB_AUDIO=m
+CONFIG_SND_USB_CAIAQ=m
+CONFIG_SND_USB_CAIAQ_INPUT=y
+CONFIG_SND_SOC=y
+CONFIG_SND_S3C24XX_SOC=y
+CONFIG_SND_S3C24XX_SOC_I2S=y
+# CONFIG_SND_S3C24XX_SOC_LN2440SBC_ALC650 is not set
+CONFIG_SND_S3C24XX_SOC_S3C24XX_UDA134X=y
+CONFIG_SND_SOC_I2C_AND_SPI=y
+# CONFIG_SND_SOC_ALL_CODECS is not set
+CONFIG_SND_SOC_L3=y
+CONFIG_SND_SOC_UDA134X=y
+# CONFIG_SOUND_PRIME is not set
+CONFIG_HID_SUPPORT=y
+CONFIG_HID=y
+# CONFIG_HID_DEBUG is not set
+CONFIG_HIDRAW=y
+
+#
+# USB Input Devices
+#
+CONFIG_USB_HID=y
+CONFIG_HID_PID=y
+CONFIG_USB_HIDDEV=y
+
+#
+# Special HID drivers
+#
+CONFIG_HID_A4TECH=y
+CONFIG_HID_APPLE=y
+CONFIG_HID_BELKIN=y
+CONFIG_HID_CHERRY=y
+CONFIG_HID_CHICONY=y
+CONFIG_HID_CYPRESS=y
+# CONFIG_DRAGONRISE_FF is not set
+CONFIG_HID_EZKEY=y
+CONFIG_HID_KYE=y
+CONFIG_HID_GYRATION=y
+CONFIG_HID_KENSINGTON=y
+CONFIG_HID_LOGITECH=y
+# CONFIG_LOGITECH_FF is not set
+# CONFIG_LOGIRUMBLEPAD2_FF is not set
+CONFIG_HID_MICROSOFT=y
+CONFIG_HID_MONTEREY=y
+CONFIG_HID_NTRIG=y
+CONFIG_HID_PANTHERLORD=y
+# CONFIG_PANTHERLORD_FF is not set
+CONFIG_HID_PETALYNX=y
+CONFIG_HID_SAMSUNG=y
+CONFIG_HID_SONY=y
+CONFIG_HID_SUNPLUS=y
+# CONFIG_GREENASIA_FF is not set
+CONFIG_HID_TOPSEED=y
+# CONFIG_THRUSTMASTER_FF is not set
+# CONFIG_ZEROPLUS_FF is not set
+CONFIG_USB_SUPPORT=y
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
+# CONFIG_USB_ARCH_HAS_EHCI is not set
+CONFIG_USB=y
+# CONFIG_USB_DEBUG is not set
+# CONFIG_USB_ANNOUNCE_NEW_DEVICES is not set
+
+#
+# Miscellaneous USB options
+#
+CONFIG_USB_DEVICEFS=y
+# CONFIG_USB_DEVICE_CLASS is not set
+# CONFIG_USB_DYNAMIC_MINORS is not set
+# CONFIG_USB_SUSPEND is not set
+# CONFIG_USB_OTG is not set
+# CONFIG_USB_MON is not set
+# CONFIG_USB_WUSB is not set
+# CONFIG_USB_WUSB_CBAF is not set
+
+#
+# USB Host Controller Drivers
+#
+# CONFIG_USB_C67X00_HCD is not set
+# CONFIG_USB_OXU210HP_HCD is not set
+# CONFIG_USB_ISP116X_HCD is not set
+# CONFIG_USB_ISP1760_HCD is not set
+CONFIG_USB_OHCI_HCD=y
+# CONFIG_USB_OHCI_BIG_ENDIAN_DESC is not set
+# CONFIG_USB_OHCI_BIG_ENDIAN_MMIO is not set
+CONFIG_USB_OHCI_LITTLE_ENDIAN=y
+# CONFIG_USB_SL811_HCD is not set
+# CONFIG_USB_R8A66597_HCD is not set
+# CONFIG_USB_HWA_HCD is not set
+# CONFIG_USB_MUSB_HDRC is not set
+# CONFIG_USB_GADGET_MUSB_HDRC is not set
+
+#
+# USB Device Class drivers
+#
+CONFIG_USB_ACM=m
+# CONFIG_USB_PRINTER is not set
+CONFIG_USB_WDM=m
+# CONFIG_USB_TMC is not set
+
+#
+# NOTE: USB_STORAGE depends on SCSI but BLK_DEV_SD may
+#
+
+#
+# also be needed; see USB_STORAGE Help for more info
+#
+CONFIG_USB_STORAGE=m
+# CONFIG_USB_STORAGE_DEBUG is not set
+CONFIG_USB_STORAGE_DATAFAB=m
+# CONFIG_USB_STORAGE_FREECOM is not set
+CONFIG_USB_STORAGE_ISD200=m
+CONFIG_USB_STORAGE_USBAT=m
+CONFIG_USB_STORAGE_SDDR09=m
+CONFIG_USB_STORAGE_SDDR55=m
+CONFIG_USB_STORAGE_JUMPSHOT=m
+CONFIG_USB_STORAGE_ALAUDA=m
+# CONFIG_USB_STORAGE_ONETOUCH is not set
+# CONFIG_USB_STORAGE_KARMA is not set
+# CONFIG_USB_STORAGE_CYPRESS_ATACB is not set
+CONFIG_USB_LIBUSUAL=y
+
+#
+# USB Imaging devices
+#
+# CONFIG_USB_MDC800 is not set
+# CONFIG_USB_MICROTEK is not set
+
+#
+# USB port drivers
+#
+CONFIG_USB_SERIAL=m
+# CONFIG_USB_EZUSB is not set
+# CONFIG_USB_SERIAL_GENERIC is not set
+# CONFIG_USB_SERIAL_AIRCABLE is not set
+# CONFIG_USB_SERIAL_ARK3116 is not set
+# CONFIG_USB_SERIAL_BELKIN is not set
+# CONFIG_USB_SERIAL_CH341 is not set
+# CONFIG_USB_SERIAL_WHITEHEAT is not set
+# CONFIG_USB_SERIAL_DIGI_ACCELEPORT is not set
+CONFIG_USB_SERIAL_CP210X=m
+# CONFIG_USB_SERIAL_CYPRESS_M8 is not set
+# CONFIG_USB_SERIAL_EMPEG is not set
+CONFIG_USB_SERIAL_FTDI_SIO=m
+# CONFIG_USB_SERIAL_FUNSOFT is not set
+# CONFIG_USB_SERIAL_VISOR is not set
+# CONFIG_USB_SERIAL_IPAQ is not set
+# CONFIG_USB_SERIAL_IR is not set
+# CONFIG_USB_SERIAL_EDGEPORT is not set
+# CONFIG_USB_SERIAL_EDGEPORT_TI is not set
+# CONFIG_USB_SERIAL_GARMIN is not set
+# CONFIG_USB_SERIAL_IPW is not set
+# CONFIG_USB_SERIAL_IUU is not set
+# CONFIG_USB_SERIAL_KEYSPAN_PDA is not set
+# CONFIG_USB_SERIAL_KEYSPAN is not set
+# CONFIG_USB_SERIAL_KLSI is not set
+# CONFIG_USB_SERIAL_KOBIL_SCT is not set
+# CONFIG_USB_SERIAL_MCT_U232 is not set
+# CONFIG_USB_SERIAL_MOS7720 is not set
+# CONFIG_USB_SERIAL_MOS7840 is not set
+# CONFIG_USB_SERIAL_MOTOROLA is not set
+# CONFIG_USB_SERIAL_NAVMAN is not set
+# CONFIG_USB_SERIAL_PL2303 is not set
+# CONFIG_USB_SERIAL_OTI6858 is not set
+# CONFIG_USB_SERIAL_QUALCOMM is not set
+CONFIG_USB_SERIAL_SPCP8X5=m
+# CONFIG_USB_SERIAL_HP4X is not set
+# CONFIG_USB_SERIAL_SAFE is not set
+# CONFIG_USB_SERIAL_SIEMENS_MPI is not set
+# CONFIG_USB_SERIAL_SIERRAWIRELESS is not set
+# CONFIG_USB_SERIAL_SYMBOL is not set
+# CONFIG_USB_SERIAL_TI is not set
+# CONFIG_USB_SERIAL_CYBERJACK is not set
+# CONFIG_USB_SERIAL_XIRCOM is not set
+# CONFIG_USB_SERIAL_OPTION is not set
+# CONFIG_USB_SERIAL_OMNINET is not set
+# CONFIG_USB_SERIAL_OPTICON is not set
+# CONFIG_USB_SERIAL_DEBUG is not set
+
+#
+# USB Miscellaneous drivers
+#
+# CONFIG_USB_EMI62 is not set
+# CONFIG_USB_EMI26 is not set
+# CONFIG_USB_ADUTUX is not set
+# CONFIG_USB_SEVSEG is not set
+# CONFIG_USB_RIO500 is not set
+# CONFIG_USB_LEGOTOWER is not set
+# CONFIG_USB_LCD is not set
+# CONFIG_USB_BERRY_CHARGE is not set
+# CONFIG_USB_LED is not set
+# CONFIG_USB_CYPRESS_CY7C63 is not set
+# CONFIG_USB_CYTHERM is not set
+# CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_FTDI_ELAN is not set
+# CONFIG_USB_APPLEDISPLAY is not set
+# CONFIG_USB_LD is not set
+# CONFIG_USB_TRANCEVIBRATOR is not set
+# CONFIG_USB_IOWARRIOR is not set
+# CONFIG_USB_TEST is not set
+# CONFIG_USB_ISIGHTFW is not set
+# CONFIG_USB_VST is not set
+CONFIG_USB_GADGET=y
+# CONFIG_USB_GADGET_DEBUG is not set
+# CONFIG_USB_GADGET_DEBUG_FILES is not set
+# CONFIG_USB_GADGET_DEBUG_FS is not set
+CONFIG_USB_GADGET_VBUS_DRAW=2
+CONFIG_USB_GADGET_SELECTED=y
+# CONFIG_USB_GADGET_AT91 is not set
+# CONFIG_USB_GADGET_ATMEL_USBA is not set
+# CONFIG_USB_GADGET_FSL_USB2 is not set
+# CONFIG_USB_GADGET_LH7A40X is not set
+# CONFIG_USB_GADGET_OMAP is not set
+# CONFIG_USB_GADGET_PXA25X is not set
+# CONFIG_USB_GADGET_PXA27X is not set
+CONFIG_USB_GADGET_S3C2410=y
+CONFIG_USB_S3C2410=y
+# CONFIG_USB_S3C2410_DEBUG is not set
+# CONFIG_USB_GADGET_IMX is not set
+# CONFIG_USB_GADGET_M66592 is not set
+# CONFIG_USB_GADGET_AMD5536UDC is not set
+# CONFIG_USB_GADGET_FSL_QE is not set
+# CONFIG_USB_GADGET_CI13XXX is not set
+# CONFIG_USB_GADGET_NET2280 is not set
+# CONFIG_USB_GADGET_GOKU is not set
+# CONFIG_USB_GADGET_DUMMY_HCD is not set
+# CONFIG_USB_GADGET_DUALSPEED is not set
+CONFIG_USB_ZERO=m
+CONFIG_USB_ETH=m
+CONFIG_USB_ETH_RNDIS=y
+CONFIG_USB_GADGETFS=m
+CONFIG_USB_FILE_STORAGE=m
+# CONFIG_USB_FILE_STORAGE_TEST is not set
+CONFIG_USB_G_SERIAL=m
+# CONFIG_USB_MIDI_GADGET is not set
+# CONFIG_USB_G_PRINTER is not set
+CONFIG_USB_CDC_COMPOSITE=m
+
+#
+# OTG and related infrastructure
+#
+# CONFIG_USB_GPIO_VBUS is not set
+# CONFIG_NOP_USB_XCEIV is not set
+CONFIG_MMC=y
+# CONFIG_MMC_DEBUG is not set
+# CONFIG_MMC_UNSAFE_RESUME is not set
+
+#
+# MMC/SD/SDIO Card Drivers
+#
+CONFIG_MMC_BLOCK=y
+CONFIG_MMC_BLOCK_BOUNCE=y
+CONFIG_SDIO_UART=y
+# CONFIG_MMC_TEST is not set
+
+#
+# MMC/SD/SDIO Host Controller Drivers
+#
+CONFIG_MMC_SDHCI=y
+CONFIG_MMC_SPI=y
+CONFIG_MMC_S3C=y
+# CONFIG_MEMSTICK is not set
+# CONFIG_ACCESSIBILITY is not set
+CONFIG_NEW_LEDS=y
+CONFIG_LEDS_CLASS=y
+
+#
+# LED drivers
+#
+CONFIG_LEDS_S3C24XX=y
+# CONFIG_LEDS_PCA9532 is not set
+CONFIG_LEDS_GPIO=y
+CONFIG_LEDS_GPIO_PLATFORM=y
+# CONFIG_LEDS_LP5521 is not set
+# CONFIG_LEDS_PCA955X is not set
+# CONFIG_LEDS_DAC124S085 is not set
+# CONFIG_LEDS_PWM is not set
+# CONFIG_LEDS_BD2802 is not set
+
+#
+# LED Triggers
+#
+CONFIG_LEDS_TRIGGERS=y
+CONFIG_LEDS_TRIGGER_TIMER=y
+CONFIG_LEDS_TRIGGER_HEARTBEAT=y
+CONFIG_LEDS_TRIGGER_BACKLIGHT=y
+CONFIG_LEDS_TRIGGER_GPIO=y
+CONFIG_LEDS_TRIGGER_DEFAULT_ON=y
+
+#
+# iptables trigger is under Netfilter config (LED target)
+#
+CONFIG_RTC_LIB=y
+CONFIG_RTC_CLASS=y
+CONFIG_RTC_HCTOSYS=y
+CONFIG_RTC_HCTOSYS_DEVICE="rtc0"
+# CONFIG_RTC_DEBUG is not set
+
+#
+# RTC interfaces
+#
+CONFIG_RTC_INTF_SYSFS=y
+CONFIG_RTC_INTF_PROC=y
+CONFIG_RTC_INTF_DEV=y
+CONFIG_RTC_INTF_DEV_UIE_EMUL=y
+# CONFIG_RTC_DRV_TEST is not set
+
+#
+# I2C RTC drivers
+#
+# CONFIG_RTC_DRV_DS1307 is not set
+# CONFIG_RTC_DRV_DS1374 is not set
+# CONFIG_RTC_DRV_DS1672 is not set
+# CONFIG_RTC_DRV_MAX6900 is not set
+# CONFIG_RTC_DRV_RS5C372 is not set
+# CONFIG_RTC_DRV_ISL1208 is not set
+# CONFIG_RTC_DRV_X1205 is not set
+# CONFIG_RTC_DRV_PCF8563 is not set
+# CONFIG_RTC_DRV_PCF8583 is not set
+# CONFIG_RTC_DRV_M41T80 is not set
+# CONFIG_RTC_DRV_S35390A is not set
+# CONFIG_RTC_DRV_FM3130 is not set
+# CONFIG_RTC_DRV_RX8581 is not set
+
+#
+# SPI RTC drivers
+#
+# CONFIG_RTC_DRV_M41T94 is not set
+# CONFIG_RTC_DRV_DS1305 is not set
+# CONFIG_RTC_DRV_DS1390 is not set
+# CONFIG_RTC_DRV_MAX6902 is not set
+# CONFIG_RTC_DRV_R9701 is not set
+# CONFIG_RTC_DRV_RS5C348 is not set
+# CONFIG_RTC_DRV_DS3234 is not set
+
+#
+# Platform RTC drivers
+#
+# CONFIG_RTC_DRV_CMOS is not set
+# CONFIG_RTC_DRV_DS1286 is not set
+# CONFIG_RTC_DRV_DS1511 is not set
+# CONFIG_RTC_DRV_DS1553 is not set
+# CONFIG_RTC_DRV_DS1742 is not set
+# CONFIG_RTC_DRV_STK17TA8 is not set
+# CONFIG_RTC_DRV_M48T86 is not set
+# CONFIG_RTC_DRV_M48T35 is not set
+# CONFIG_RTC_DRV_M48T59 is not set
+# CONFIG_RTC_DRV_BQ4802 is not set
+# CONFIG_RTC_DRV_V3020 is not set
+
+#
+# on-CPU RTC drivers
+#
+CONFIG_RTC_DRV_S3C=y
+CONFIG_DMADEVICES=y
+
+#
+# DMA Devices
+#
+# CONFIG_AUXDISPLAY is not set
+# CONFIG_REGULATOR is not set
+# CONFIG_UIO is not set
+# CONFIG_STAGING is not set
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=m
+CONFIG_EXT2_FS_XATTR=y
+CONFIG_EXT2_FS_POSIX_ACL=y
+CONFIG_EXT2_FS_SECURITY=y
+# CONFIG_EXT2_FS_XIP is not set
+CONFIG_EXT3_FS=y
+# CONFIG_EXT3_DEFAULTS_TO_ORDERED is not set
+CONFIG_EXT3_FS_XATTR=y
+CONFIG_EXT3_FS_POSIX_ACL=y
+CONFIG_EXT3_FS_SECURITY=y
+# CONFIG_EXT4_FS is not set
+CONFIG_JBD=y
+# CONFIG_JBD_DEBUG is not set
+CONFIG_FS_MBCACHE=y
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+CONFIG_FS_POSIX_ACL=y
+CONFIG_FILE_LOCKING=y
+# CONFIG_XFS_FS is not set
+# CONFIG_GFS2_FS is not set
+# CONFIG_OCFS2_FS is not set
+# CONFIG_BTRFS_FS is not set
+CONFIG_DNOTIFY=y
+CONFIG_INOTIFY=y
+CONFIG_INOTIFY_USER=y
+# CONFIG_QUOTA is not set
+CONFIG_AUTOFS_FS=y
+CONFIG_AUTOFS4_FS=y
+# CONFIG_FUSE_FS is not set
+CONFIG_GENERIC_ACL=y
+
+#
+# Caches
+#
+# CONFIG_FSCACHE is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=y
+CONFIG_MSDOS_FS=y
+CONFIG_VFAT_FS=y
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_SYSCTL=y
+CONFIG_PROC_PAGE_MONITOR=y
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+CONFIG_TMPFS_POSIX_ACL=y
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_CONFIGFS_FS=m
+CONFIG_MISC_FILESYSTEMS=y
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_ECRYPT_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+CONFIG_JFFS2_FS=y
+CONFIG_JFFS2_FS_DEBUG=0
+CONFIG_JFFS2_FS_WRITEBUFFER=y
+# CONFIG_JFFS2_FS_WBUF_VERIFY is not set
+# CONFIG_JFFS2_SUMMARY is not set
+# CONFIG_JFFS2_FS_XATTR is not set
+# CONFIG_JFFS2_COMPRESSION_OPTIONS is not set
+CONFIG_JFFS2_ZLIB=y
+# CONFIG_JFFS2_LZO is not set
+CONFIG_JFFS2_RTIME=y
+# CONFIG_JFFS2_RUBIN is not set
+CONFIG_CRAMFS=y
+# CONFIG_SQUASHFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_OMFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+CONFIG_ROMFS_FS=y
+# CONFIG_ROMFS_BACKED_BY_BLOCK is not set
+# CONFIG_ROMFS_BACKED_BY_MTD is not set
+CONFIG_ROMFS_BACKED_BY_BOTH=y
+CONFIG_ROMFS_ON_BLOCK=y
+CONFIG_ROMFS_ON_MTD=y
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+# CONFIG_NILFS2_FS is not set
+CONFIG_NETWORK_FILESYSTEMS=y
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+CONFIG_NFS_V3_ACL=y
+CONFIG_NFS_V4=y
+CONFIG_ROOT_NFS=y
+# CONFIG_NFSD is not set
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_NFS_ACL_SUPPORT=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+CONFIG_SUNRPC_GSS=y
+CONFIG_RPCSEC_GSS_KRB5=y
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+
+#
+# Partition Types
+#
+CONFIG_PARTITION_ADVANCED=y
+# CONFIG_ACORN_PARTITION is not set
+# CONFIG_OSF_PARTITION is not set
+# CONFIG_AMIGA_PARTITION is not set
+# CONFIG_ATARI_PARTITION is not set
+# CONFIG_MAC_PARTITION is not set
+CONFIG_MSDOS_PARTITION=y
+CONFIG_BSD_DISKLABEL=y
+CONFIG_MINIX_SUBPARTITION=y
+CONFIG_SOLARIS_X86_PARTITION=y
+CONFIG_UNIXWARE_DISKLABEL=y
+CONFIG_LDM_PARTITION=y
+# CONFIG_LDM_DEBUG is not set
+# CONFIG_SGI_PARTITION is not set
+# CONFIG_ULTRIX_PARTITION is not set
+# CONFIG_SUN_PARTITION is not set
+# CONFIG_KARMA_PARTITION is not set
+CONFIG_EFI_PARTITION=y
+# CONFIG_SYSV68_PARTITION is not set
+CONFIG_NLS=y
+CONFIG_NLS_DEFAULT="cp437"
+CONFIG_NLS_CODEPAGE_437=m
+CONFIG_NLS_CODEPAGE_737=m
+CONFIG_NLS_CODEPAGE_775=m
+CONFIG_NLS_CODEPAGE_850=m
+CONFIG_NLS_CODEPAGE_852=m
+CONFIG_NLS_CODEPAGE_855=m
+CONFIG_NLS_CODEPAGE_857=m
+CONFIG_NLS_CODEPAGE_860=m
+CONFIG_NLS_CODEPAGE_861=m
+CONFIG_NLS_CODEPAGE_862=m
+CONFIG_NLS_CODEPAGE_863=m
+CONFIG_NLS_CODEPAGE_864=m
+CONFIG_NLS_CODEPAGE_865=m
+CONFIG_NLS_CODEPAGE_866=m
+CONFIG_NLS_CODEPAGE_869=m
+CONFIG_NLS_CODEPAGE_936=m
+CONFIG_NLS_CODEPAGE_950=m
+CONFIG_NLS_CODEPAGE_932=m
+CONFIG_NLS_CODEPAGE_949=m
+CONFIG_NLS_CODEPAGE_874=m
+CONFIG_NLS_ISO8859_8=m
+CONFIG_NLS_CODEPAGE_1250=m
+CONFIG_NLS_CODEPAGE_1251=m
+CONFIG_NLS_ASCII=m
+CONFIG_NLS_ISO8859_1=m
+CONFIG_NLS_ISO8859_2=m
+CONFIG_NLS_ISO8859_3=m
+CONFIG_NLS_ISO8859_4=m
+CONFIG_NLS_ISO8859_5=m
+CONFIG_NLS_ISO8859_6=m
+CONFIG_NLS_ISO8859_7=m
+CONFIG_NLS_ISO8859_9=m
+CONFIG_NLS_ISO8859_13=m
+CONFIG_NLS_ISO8859_14=m
+CONFIG_NLS_ISO8859_15=m
+CONFIG_NLS_KOI8_R=m
+CONFIG_NLS_KOI8_U=m
+CONFIG_NLS_UTF8=m
+# CONFIG_DLM is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+# CONFIG_ENABLE_WARN_DEPRECATED is not set
+# CONFIG_ENABLE_MUST_CHECK is not set
+CONFIG_FRAME_WARN=1024
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_UNUSED_SYMBOLS is not set
+CONFIG_DEBUG_FS=y
+# CONFIG_HEADERS_CHECK is not set
+CONFIG_DEBUG_KERNEL=y
+# CONFIG_DEBUG_SHIRQ is not set
+CONFIG_DETECT_SOFTLOCKUP=y
+# CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC is not set
+CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC_VALUE=0
+CONFIG_DETECT_HUNG_TASK=y
+# CONFIG_BOOTPARAM_HUNG_TASK_PANIC is not set
+CONFIG_BOOTPARAM_HUNG_TASK_PANIC_VALUE=0
+# CONFIG_SCHED_DEBUG is not set
+# CONFIG_SCHEDSTATS is not set
+# CONFIG_TIMER_STATS is not set
+# CONFIG_DEBUG_OBJECTS is not set
+# CONFIG_SLUB_DEBUG_ON is not set
+# CONFIG_SLUB_STATS is not set
+# CONFIG_DEBUG_RT_MUTEXES is not set
+# CONFIG_RT_MUTEX_TESTER is not set
+# CONFIG_DEBUG_SPINLOCK is not set
+# CONFIG_DEBUG_MUTEXES is not set
+# CONFIG_DEBUG_LOCK_ALLOC is not set
+# CONFIG_PROVE_LOCKING is not set
+# CONFIG_LOCK_STAT is not set
+# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
+# CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
+# CONFIG_DEBUG_KOBJECT is not set
+CONFIG_DEBUG_BUGVERBOSE=y
+CONFIG_DEBUG_INFO=y
+# CONFIG_DEBUG_VM is not set
+# CONFIG_DEBUG_WRITECOUNT is not set
+CONFIG_DEBUG_MEMORY_INIT=y
+# CONFIG_DEBUG_LIST is not set
+# CONFIG_DEBUG_SG is not set
+# CONFIG_DEBUG_NOTIFIERS is not set
+# CONFIG_BOOT_PRINTK_DELAY is not set
+# CONFIG_RCU_TORTURE_TEST is not set
+# CONFIG_RCU_CPU_STALL_DETECTOR is not set
+# CONFIG_BACKTRACE_SELF_TEST is not set
+# CONFIG_DEBUG_BLOCK_EXT_DEVT is not set
+# CONFIG_FAULT_INJECTION is not set
+# CONFIG_LATENCYTOP is not set
+CONFIG_SYSCTL_SYSCALL_CHECK=y
+# CONFIG_PAGE_POISONING is not set
+CONFIG_HAVE_FUNCTION_TRACER=y
+CONFIG_TRACING_SUPPORT=y
+
+#
+# Tracers
+#
+# CONFIG_FUNCTION_TRACER is not set
+# CONFIG_SCHED_TRACER is not set
+# CONFIG_CONTEXT_SWITCH_TRACER is not set
+# CONFIG_EVENT_TRACER is not set
+# CONFIG_BOOT_TRACER is not set
+# CONFIG_TRACE_BRANCH_PROFILING is not set
+# CONFIG_STACK_TRACER is not set
+# CONFIG_KMEMTRACE is not set
+# CONFIG_WORKQUEUE_TRACER is not set
+# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_DYNAMIC_DEBUG is not set
+# CONFIG_SAMPLES is not set
+CONFIG_HAVE_ARCH_KGDB=y
+# CONFIG_KGDB is not set
+CONFIG_ARM_UNWIND=y
+CONFIG_DEBUG_USER=y
+# CONFIG_DEBUG_ERRORS is not set
+# CONFIG_DEBUG_STACK_USAGE is not set
+# CONFIG_DEBUG_LL is not set
+CONFIG_DEBUG_S3C_UART=0
+
+#
+# Security options
+#
+CONFIG_KEYS=y
+# CONFIG_KEYS_DEBUG_PROC_KEYS is not set
+# CONFIG_SECURITY is not set
+# CONFIG_SECURITYFS is not set
+CONFIG_SECURITY_FILE_CAPABILITIES=y
+CONFIG_CRYPTO=y
+
+#
+# Crypto core or helper
+#
+CONFIG_CRYPTO_FIPS=y
+CONFIG_CRYPTO_ALGAPI=y
+CONFIG_CRYPTO_ALGAPI2=y
+CONFIG_CRYPTO_AEAD=m
+CONFIG_CRYPTO_AEAD2=y
+CONFIG_CRYPTO_BLKCIPHER=y
+CONFIG_CRYPTO_BLKCIPHER2=y
+CONFIG_CRYPTO_HASH=y
+CONFIG_CRYPTO_HASH2=y
+CONFIG_CRYPTO_RNG=m
+CONFIG_CRYPTO_RNG2=y
+CONFIG_CRYPTO_PCOMP=y
+CONFIG_CRYPTO_MANAGER=y
+CONFIG_CRYPTO_MANAGER2=y
+CONFIG_CRYPTO_GF128MUL=m
+CONFIG_CRYPTO_NULL=m
+CONFIG_CRYPTO_WORKQUEUE=y
+CONFIG_CRYPTO_CRYPTD=m
+CONFIG_CRYPTO_AUTHENC=m
+CONFIG_CRYPTO_TEST=m
+
+#
+# Authenticated Encryption with Associated Data
+#
+CONFIG_CRYPTO_CCM=m
+CONFIG_CRYPTO_GCM=m
+CONFIG_CRYPTO_SEQIV=m
+
+#
+# Block modes
+#
+CONFIG_CRYPTO_CBC=y
+CONFIG_CRYPTO_CTR=m
+CONFIG_CRYPTO_CTS=m
+CONFIG_CRYPTO_ECB=y
+CONFIG_CRYPTO_LRW=m
+CONFIG_CRYPTO_PCBC=m
+CONFIG_CRYPTO_XTS=m
+
+#
+# Hash modes
+#
+CONFIG_CRYPTO_HMAC=y
+CONFIG_CRYPTO_XCBC=m
+
+#
+# Digest
+#
+CONFIG_CRYPTO_CRC32C=m
+CONFIG_CRYPTO_MD4=m
+CONFIG_CRYPTO_MD5=y
+CONFIG_CRYPTO_MICHAEL_MIC=y
+CONFIG_CRYPTO_RMD128=m
+CONFIG_CRYPTO_RMD160=m
+CONFIG_CRYPTO_RMD256=m
+CONFIG_CRYPTO_RMD320=m
+CONFIG_CRYPTO_SHA1=m
+CONFIG_CRYPTO_SHA256=m
+CONFIG_CRYPTO_SHA512=m
+CONFIG_CRYPTO_TGR192=m
+CONFIG_CRYPTO_WP512=m
+
+#
+# Ciphers
+#
+CONFIG_CRYPTO_AES=y
+CONFIG_CRYPTO_ANUBIS=m
+CONFIG_CRYPTO_ARC4=y
+CONFIG_CRYPTO_BLOWFISH=m
+CONFIG_CRYPTO_CAMELLIA=m
+CONFIG_CRYPTO_CAST5=m
+CONFIG_CRYPTO_CAST6=m
+CONFIG_CRYPTO_DES=y
+CONFIG_CRYPTO_FCRYPT=m
+CONFIG_CRYPTO_KHAZAD=m
+CONFIG_CRYPTO_SALSA20=m
+CONFIG_CRYPTO_SEED=m
+CONFIG_CRYPTO_SERPENT=m
+CONFIG_CRYPTO_TEA=m
+CONFIG_CRYPTO_TWOFISH=m
+CONFIG_CRYPTO_TWOFISH_COMMON=m
+
+#
+# Compression
+#
+CONFIG_CRYPTO_DEFLATE=m
+CONFIG_CRYPTO_ZLIB=m
+CONFIG_CRYPTO_LZO=m
+
+#
+# Random Number Generation
+#
+CONFIG_CRYPTO_ANSI_CPRNG=m
+CONFIG_CRYPTO_HW=y
+# CONFIG_BINARY_PRINTF is not set
+
+#
+# Library routines
+#
+CONFIG_BITREVERSE=y
+CONFIG_GENERIC_FIND_LAST_BIT=y
+CONFIG_CRC_CCITT=m
+CONFIG_CRC16=m
+CONFIG_CRC_T10DIF=y
+CONFIG_CRC_ITU_T=y
+CONFIG_CRC32=y
+CONFIG_CRC7=y
+CONFIG_LIBCRC32C=m
+CONFIG_ZLIB_INFLATE=y
+CONFIG_ZLIB_DEFLATE=y
+CONFIG_LZO_COMPRESS=m
+CONFIG_LZO_DECOMPRESS=m
+CONFIG_DECOMPRESS_GZIP=y
+CONFIG_DECOMPRESS_BZIP2=y
+CONFIG_DECOMPRESS_LZMA=y
+CONFIG_HAS_IOMEM=y
+CONFIG_HAS_DMA=y
+CONFIG_NLATTR=y
-- 
cgit v1.2.3-59-g8ed1b


From 799ca065213ab5df92d787c36c99cb4222f4e717 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 15 Jun 2009 22:07:48 +0100
Subject: [ARM] MINI2440: Cleanup duplicated header includes

Remove duplicate mach headers for gpio.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/mach-s3c2440/mach-mini2440.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/arch/arm/mach-s3c2440/mach-mini2440.c b/arch/arm/mach-s3c2440/mach-mini2440.c
index eb4b24501a9f..6a5bc3021bdb 100644
--- a/arch/arm/mach-s3c2440/mach-mini2440.c
+++ b/arch/arm/mach-s3c2440/mach-mini2440.c
@@ -49,9 +49,6 @@
 #include <plat/udc.h>
 
 #include <plat/regs-serial.h>
-#include <mach/regs-gpio.h>
-#include <mach/regs-mem.h>
-#include <mach/regs-lcd.h>
 
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/nand.h>
-- 
cgit v1.2.3-59-g8ed1b


From f25f0b9ca48848632f19e6616bd01550e3c0fc0e Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Thu, 4 Jun 2009 18:00:25 +0100
Subject: s3cmci: fix dma configuration call

This was missed in the DMA changes during the s3c24xx
updates in commit 8970ef47d56fd3db28ee798b9d400caf08abd924.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 drivers/mmc/host/s3cmci.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mmc/host/s3cmci.c b/drivers/mmc/host/s3cmci.c
index 4eb4f37544ab..8c08cd7efa7f 100644
--- a/drivers/mmc/host/s3cmci.c
+++ b/drivers/mmc/host/s3cmci.c
@@ -794,7 +794,7 @@ static void s3cmci_dma_setup(struct s3cmci_host *host,
 			      host->mem->start + host->sdidata);
 
 	if (!setup_ok) {
-		s3c2410_dma_config(host->dma, 4, 0);
+		s3c2410_dma_config(host->dma, 4);
 		s3c2410_dma_set_buffdone_fn(host->dma,
 					    s3cmci_dma_done_callback);
 		s3c2410_dma_setflags(host->dma, S3C2410_DMAF_AUTOSTART);
-- 
cgit v1.2.3-59-g8ed1b


From 48e03bc515cff75718de5460458680a230ae997e Mon Sep 17 00:00:00 2001
From: Trond Myklebust <trond.myklebust@fys.uio.no>
Date: Fri, 5 Jun 2009 12:35:15 -0400
Subject: nfsd: Use write gathering only with NFSv2

NFSv3 and above can use unstable writes whenever they are sending more
than one write, rather than relying on the flaky write gathering
heuristics. More often than not, write gathering is currently getting it
wrong when the NFSv3 clients are sending a single write with FILE_SYNC
for efficiency reasons.

This patch turns off write gathering for NFSv3/v4, and ensures that
it only applies to the one case that can actually benefit: namely NFSv2.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/vfs.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index b660435978d2..f30cc4eadb0a 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -975,6 +975,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 	__be32			err = 0;
 	int			host_err;
 	int			stable = *stablep;
+	int			use_wgather;
 
 #ifdef MSNFS
 	err = nfserr_perm;
@@ -993,9 +994,10 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 	 *  -	the sync export option has been set, or
 	 *  -	the client requested O_SYNC behavior (NFSv3 feature).
 	 *  -   The file system doesn't support fsync().
-	 * When gathered writes have been configured for this volume,
+	 * When NFSv2 gathered writes have been configured for this volume,
 	 * flushing the data to disk is handled separately below.
 	 */
+	use_wgather = (rqstp->rq_vers == 2) && EX_WGATHER(exp);
 
 	if (!file->f_op->fsync) {/* COMMIT3 cannot work */
 	       stable = 2;
@@ -1004,7 +1006,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 
 	if (!EX_ISSYNC(exp))
 		stable = 0;
-	if (stable && !EX_WGATHER(exp)) {
+	if (stable && !use_wgather) {
 		spin_lock(&file->f_lock);
 		file->f_flags |= O_SYNC;
 		spin_unlock(&file->f_lock);
@@ -1040,7 +1042,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 		 * nice and simple solution (IMHO), and it seems to
 		 * work:-)
 		 */
-		if (EX_WGATHER(exp)) {
+		if (use_wgather) {
 			if (atomic_read(&inode->i_writecount) > 1
 			    || (last_ino == inode->i_ino && last_dev == inode->i_sb->s_dev)) {
 				dprintk("nfsd: write defer %d\n", task_pid_nr(current));
-- 
cgit v1.2.3-59-g8ed1b


From 6aad89c8376e432231b78d1bdbcc10ef7708b011 Mon Sep 17 00:00:00 2001
From: Anton Blanchard <anton@samba.org>
Date: Wed, 10 Jun 2009 12:52:21 -0700
Subject: sunrpc: align cache_clean work's timer

Align cache_clean work.

Signed-off-by: Anton Blanchard <anton@samba.org>
Cc: Neil Brown <neilb@suse.de>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 net/sunrpc/cache.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c
index 20029a79a5de..ff0c23053d2f 100644
--- a/net/sunrpc/cache.c
+++ b/net/sunrpc/cache.c
@@ -488,7 +488,7 @@ static void do_cache_clean(struct work_struct *work)
 {
 	int delay = 5;
 	if (cache_clean() == -1)
-		delay = 30*HZ;
+		delay = round_jiffies_relative(30*HZ);
 
 	if (list_empty(&cache_list))
 		delay = 0;
-- 
cgit v1.2.3-59-g8ed1b


From 9d2a3f31d6d7832cd441eeda08bc2266cdd5d972 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Mon, 15 Jun 2009 18:47:56 -0700
Subject: nfsd: track last inode only in use_wgather case

Updating last_ino and last_dev probably isn't useful in the !use_wgather
case.

Also remove some pointless ifdef'd-out code.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/vfs.c | 25 ++++++++++---------------
 1 file changed, 10 insertions(+), 15 deletions(-)

diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index f30cc4eadb0a..ebf56c6040ca 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1026,7 +1026,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 	if (host_err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID)))
 		kill_suid(dentry);
 
-	if (host_err >= 0 && stable) {
+	if (host_err >= 0 && stable && use_wgather) {
 		static ino_t	last_ino;
 		static dev_t	last_dev;
 
@@ -1042,21 +1042,16 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 		 * nice and simple solution (IMHO), and it seems to
 		 * work:-)
 		 */
-		if (use_wgather) {
-			if (atomic_read(&inode->i_writecount) > 1
-			    || (last_ino == inode->i_ino && last_dev == inode->i_sb->s_dev)) {
-				dprintk("nfsd: write defer %d\n", task_pid_nr(current));
-				msleep(10);
-				dprintk("nfsd: write resume %d\n", task_pid_nr(current));
-			}
+		if (atomic_read(&inode->i_writecount) > 1
+		    || (last_ino == inode->i_ino && last_dev == inode->i_sb->s_dev)) {
+			dprintk("nfsd: write defer %d\n", task_pid_nr(current));
+			msleep(10);
+			dprintk("nfsd: write resume %d\n", task_pid_nr(current));
+		}
 
-			if (inode->i_state & I_DIRTY) {
-				dprintk("nfsd: write sync %d\n", task_pid_nr(current));
-				host_err=nfsd_sync(file);
-			}
-#if 0
-			wake_up(&inode->i_wait);
-#endif
+		if (inode->i_state & I_DIRTY) {
+			dprintk("nfsd: write sync %d\n", task_pid_nr(current));
+			host_err=nfsd_sync(file);
 		}
 		last_ino = inode->i_ino;
 		last_dev = inode->i_sb->s_dev;
-- 
cgit v1.2.3-59-g8ed1b


From d911df7b8d44de41661363a4e29ee710180ba025 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Mon, 15 Jun 2009 16:03:53 -0700
Subject: nfsd: Pull write-gathering code out of nfsd_vfs_write

This is a relatively self-contained piece of code that handles a special
case--move it to its own function.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/vfs.c | 69 +++++++++++++++++++++++++++++++++--------------------------
 1 file changed, 39 insertions(+), 30 deletions(-)

diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index ebf56c6040ca..6ad76a4cfc01 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -963,6 +963,43 @@ static void kill_suid(struct dentry *dentry)
 	mutex_unlock(&dentry->d_inode->i_mutex);
 }
 
+/*
+ * Gathered writes: If another process is currently writing to the file,
+ * there's a high chance this is another nfsd (triggered by a bulk write
+ * from a client's biod). Rather than syncing the file with each write
+ * request, we sleep for 10 msec.
+ *
+ * I don't know if this roughly approximates C. Juszak's idea of
+ * gathered writes, but it's a nice and simple solution (IMHO), and it
+ * seems to work:-)
+ *
+ * Note: we do this only in the NFSv2 case, since v3 and higher have a
+ * better tool (separate unstable writes and commits) for solving this
+ * problem.
+ */
+static int wait_for_concurrent_writes(struct file *file)
+{
+	struct inode *inode = file->f_path.dentry->d_inode;
+	static ino_t last_ino;
+	static dev_t last_dev;
+	int err = 0;
+
+	if (atomic_read(&inode->i_writecount) > 1
+	    || (last_ino == inode->i_ino && last_dev == inode->i_sb->s_dev)) {
+		dprintk("nfsd: write defer %d\n", task_pid_nr(current));
+		msleep(10);
+		dprintk("nfsd: write resume %d\n", task_pid_nr(current));
+	}
+
+	if (inode->i_state & I_DIRTY) {
+		dprintk("nfsd: write sync %d\n", task_pid_nr(current));
+		err = nfsd_sync(file);
+	}
+	last_ino = inode->i_ino;
+	last_dev = inode->i_sb->s_dev;
+	return err;
+}
+
 static __be32
 nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 				loff_t offset, struct kvec *vec, int vlen,
@@ -1026,36 +1063,8 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 	if (host_err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID)))
 		kill_suid(dentry);
 
-	if (host_err >= 0 && stable && use_wgather) {
-		static ino_t	last_ino;
-		static dev_t	last_dev;
-
-		/*
-		 * Gathered writes: If another process is currently
-		 * writing to the file, there's a high chance
-		 * this is another nfsd (triggered by a bulk write
-		 * from a client's biod). Rather than syncing the
-		 * file with each write request, we sleep for 10 msec.
-		 *
-		 * I don't know if this roughly approximates
-		 * C. Juszak's idea of gathered writes, but it's a
-		 * nice and simple solution (IMHO), and it seems to
-		 * work:-)
-		 */
-		if (atomic_read(&inode->i_writecount) > 1
-		    || (last_ino == inode->i_ino && last_dev == inode->i_sb->s_dev)) {
-			dprintk("nfsd: write defer %d\n", task_pid_nr(current));
-			msleep(10);
-			dprintk("nfsd: write resume %d\n", task_pid_nr(current));
-		}
-
-		if (inode->i_state & I_DIRTY) {
-			dprintk("nfsd: write sync %d\n", task_pid_nr(current));
-			host_err=nfsd_sync(file);
-		}
-		last_ino = inode->i_ino;
-		last_dev = inode->i_sb->s_dev;
-	}
+	if (host_err >= 0 && stable && use_wgather)
+		host_err = wait_for_concurrent_writes(file);
 
 	dprintk("nfsd: write complete host_err=%d\n", host_err);
 	if (host_err >= 0)
-- 
cgit v1.2.3-59-g8ed1b


From e4636d535e32768c8c500641ddb144f56e3dc5c0 Mon Sep 17 00:00:00 2001
From: "J. Bruce Fields" <bfields@citi.umich.edu>
Date: Mon, 15 Jun 2009 19:07:13 -0700
Subject: nfsd: minor nfsd_vfs_write cleanup

There's no need to check host_err >= 0 every time here when we could
check host_err < 0 once, following the usual kernel style.

Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/vfs.c | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 6ad76a4cfc01..1cf70616a11e 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1053,19 +1053,20 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file,
 	oldfs = get_fs(); set_fs(KERNEL_DS);
 	host_err = vfs_writev(file, (struct iovec __user *)vec, vlen, &offset);
 	set_fs(oldfs);
-	if (host_err >= 0) {
-		*cnt = host_err;
-		nfsdstats.io_write += host_err;
-		fsnotify_modify(file->f_path.dentry);
-	}
+	if (host_err < 0)
+		goto out_nfserr;
+	*cnt = host_err;
+	nfsdstats.io_write += host_err;
+	fsnotify_modify(file->f_path.dentry);
 
 	/* clear setuid/setgid flag after write */
-	if (host_err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID)))
+	if (inode->i_mode & (S_ISUID | S_ISGID))
 		kill_suid(dentry);
 
-	if (host_err >= 0 && stable && use_wgather)
+	if (stable && use_wgather)
 		host_err = wait_for_concurrent_writes(file);
 
+out_nfserr:
 	dprintk("nfsd: write complete host_err=%d\n", host_err);
 	if (host_err >= 0)
 		err = 0;
-- 
cgit v1.2.3-59-g8ed1b


From 59fb30660be3f3ead54b3850c401d462449caaaa Mon Sep 17 00:00:00 2001
From: Christian Engelmayer <christian.engelmayer@frequentis.com>
Date: Sun, 14 Jun 2009 00:05:26 +0200
Subject: sunrpc: potential memory leak in function rdma_read_xdr

In case the check on ch_count fails the cleanup path is skipped and the
previously allocated memory 'rpl_map', 'chl_map' is not freed.

Reported by Coverity.

Signed-off-by: Christian Engelmayer <christian.engelmayer@frequentis.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 net/sunrpc/xprtrdma/svc_rdma_recvfrom.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
index 42a6f9f20285..9e884383134f 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
@@ -397,14 +397,14 @@ static int rdma_read_xdr(struct svcxprt_rdma *xprt,
 	if (!ch)
 		return 0;
 
-	/* Allocate temporary reply and chunk maps */
-	rpl_map = svc_rdma_get_req_map();
-	chl_map = svc_rdma_get_req_map();
-
 	svc_rdma_rcl_chunk_counts(ch, &ch_count, &byte_count);
 	if (ch_count > RPCSVC_MAXPAGES)
 		return -EINVAL;
 
+	/* Allocate temporary reply and chunk maps */
+	rpl_map = svc_rdma_get_req_map();
+	chl_map = svc_rdma_get_req_map();
+
 	if (!xprt->sc_frmr_pg_list_len)
 		sge_count = map_read_chunks(xprt, rqstp, hdr_ctxt, rmsgp,
 					    rpl_map, chl_map, ch_count,
-- 
cgit v1.2.3-59-g8ed1b


From b9081d90f5b989cd927052084b16b4f950c8c8d7 Mon Sep 17 00:00:00 2001
From: Yu Zhiguo <yuzg@cn.fujitsu.com>
Date: Tue, 9 Jun 2009 17:33:34 +0800
Subject: NFS: kill off complicated macro 'PROC'

kill off obscure macro 'PROC' of NFSv2&3 in order to make the code more clear.

Among other things, this makes it simpler to grep for callers of these
functions--something which has frequently caused confusion among nfs
developers.

Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs3proc.c | 237 +++++++++++++++++++++++++++++++++++++++++++++++------
 fs/nfsd/nfsproc.c  | 198 ++++++++++++++++++++++++++++++++++++--------
 2 files changed, 379 insertions(+), 56 deletions(-)

diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c
index 7c9fe838f038..a713c418a922 100644
--- a/fs/nfsd/nfs3proc.c
+++ b/fs/nfsd/nfs3proc.c
@@ -652,8 +652,6 @@ nfsd3_proc_commit(struct svc_rqst * rqstp, struct nfsd3_commitargs *argp,
  * NFSv3 Server procedures.
  * Only the results of non-idempotent operations are cached.
  */
-#define nfs3svc_decode_voidargs		NULL
-#define nfs3svc_release_void		NULL
 #define nfs3svc_decode_fhandleargs	nfs3svc_decode_fhandle
 #define nfs3svc_encode_attrstatres	nfs3svc_encode_attrstat
 #define nfs3svc_encode_wccstatres	nfs3svc_encode_wccstat
@@ -686,28 +684,219 @@ struct nfsd3_voidargs { int dummy; };
 #define WC (7+pAT)	/* WCC attributes */
 
 static struct svc_procedure		nfsd_procedures3[22] = {
-  PROC(null,	 void,		void,		void,	  RC_NOCACHE, ST),
-  PROC(getattr,	 fhandle,	attrstat,	fhandle,  RC_NOCACHE, ST+AT),
-  PROC(setattr,  sattr,		wccstat,	fhandle,  RC_REPLBUFF, ST+WC),
-  PROC(lookup,	 dirop,		dirop,		fhandle2, RC_NOCACHE, ST+FH+pAT+pAT),
-  PROC(access,	 access,	access,		fhandle,  RC_NOCACHE, ST+pAT+1),
-  PROC(readlink, readlink,	readlink,	fhandle,  RC_NOCACHE, ST+pAT+1+NFS3_MAXPATHLEN/4),
-  PROC(read,	 read,		read,		fhandle,  RC_NOCACHE, ST+pAT+4+NFSSVC_MAXBLKSIZE/4),
-  PROC(write,	 write,		write,		fhandle,  RC_REPLBUFF, ST+WC+4),
-  PROC(create,	 create,	create,		fhandle2, RC_REPLBUFF, ST+(1+FH+pAT)+WC),
-  PROC(mkdir,	 mkdir,		create,		fhandle2, RC_REPLBUFF, ST+(1+FH+pAT)+WC),
-  PROC(symlink,	 symlink,	create,		fhandle2, RC_REPLBUFF, ST+(1+FH+pAT)+WC),
-  PROC(mknod,	 mknod,		create,		fhandle2, RC_REPLBUFF, ST+(1+FH+pAT)+WC),
-  PROC(remove,	 dirop,		wccstat,	fhandle,  RC_REPLBUFF, ST+WC),
-  PROC(rmdir,	 dirop,		wccstat,	fhandle,  RC_REPLBUFF, ST+WC),
-  PROC(rename,	 rename,	rename,		fhandle2, RC_REPLBUFF, ST+WC+WC),
-  PROC(link,	 link,		link,		fhandle2, RC_REPLBUFF, ST+pAT+WC),
-  PROC(readdir,	 readdir,	readdir,	fhandle,  RC_NOCACHE, 0),
-  PROC(readdirplus,readdirplus,	readdir,	fhandle,  RC_NOCACHE, 0),
-  PROC(fsstat,	 fhandle,	fsstat,		void,     RC_NOCACHE, ST+pAT+2*6+1),
-  PROC(fsinfo,   fhandle,	fsinfo,		void,     RC_NOCACHE, ST+pAT+12),
-  PROC(pathconf, fhandle,	pathconf,	void,     RC_NOCACHE, ST+pAT+6),
-  PROC(commit,	 commit,	commit,		fhandle,  RC_NOCACHE, ST+WC+2),
+	[NFS3PROC_NULL] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_null,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_voidres,
+		.pc_argsize = sizeof(struct nfsd3_voidargs),
+		.pc_ressize = sizeof(struct nfsd3_voidres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST,
+	},
+	[NFS3PROC_GETATTR] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_getattr,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_fhandleargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_attrstatres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd3_fhandleargs),
+		.pc_ressize = sizeof(struct nfsd3_attrstatres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+AT,
+	},
+	[NFS3PROC_SETATTR] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_setattr,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_sattrargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_wccstatres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd3_sattrargs),
+		.pc_ressize = sizeof(struct nfsd3_wccstatres),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+WC,
+	},
+	[NFS3PROC_LOOKUP] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_lookup,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_diropargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_diropres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle2,
+		.pc_argsize = sizeof(struct nfsd3_diropargs),
+		.pc_ressize = sizeof(struct nfsd3_diropres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+FH+pAT+pAT,
+	},
+	[NFS3PROC_ACCESS] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_access,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_accessargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_accessres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd3_accessargs),
+		.pc_ressize = sizeof(struct nfsd3_accessres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+pAT+1,
+	},
+	[NFS3PROC_READLINK] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_readlink,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_readlinkargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_readlinkres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd3_readlinkargs),
+		.pc_ressize = sizeof(struct nfsd3_readlinkres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+pAT+1+NFS3_MAXPATHLEN/4,
+	},
+	[NFS3PROC_READ] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_read,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_readargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_readres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd3_readargs),
+		.pc_ressize = sizeof(struct nfsd3_readres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+pAT+4+NFSSVC_MAXBLKSIZE/4,
+	},
+	[NFS3PROC_WRITE] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_write,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_writeargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_writeres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd3_writeargs),
+		.pc_ressize = sizeof(struct nfsd3_writeres),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+WC+4,
+	},
+	[NFS3PROC_CREATE] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_create,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_createargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_createres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle2,
+		.pc_argsize = sizeof(struct nfsd3_createargs),
+		.pc_ressize = sizeof(struct nfsd3_createres),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+(1+FH+pAT)+WC,
+	},
+	[NFS3PROC_MKDIR] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_mkdir,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_mkdirargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_createres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle2,
+		.pc_argsize = sizeof(struct nfsd3_mkdirargs),
+		.pc_ressize = sizeof(struct nfsd3_createres),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+(1+FH+pAT)+WC,
+	},
+	[NFS3PROC_SYMLINK] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_symlink,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_symlinkargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_createres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle2,
+		.pc_argsize = sizeof(struct nfsd3_symlinkargs),
+		.pc_ressize = sizeof(struct nfsd3_createres),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+(1+FH+pAT)+WC,
+	},
+	[NFS3PROC_MKNOD] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_mknod,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_mknodargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_createres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle2,
+		.pc_argsize = sizeof(struct nfsd3_mknodargs),
+		.pc_ressize = sizeof(struct nfsd3_createres),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+(1+FH+pAT)+WC,
+	},
+	[NFS3PROC_REMOVE] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_remove,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_diropargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_wccstatres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd3_diropargs),
+		.pc_ressize = sizeof(struct nfsd3_wccstatres),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+WC,
+	},
+	[NFS3PROC_RMDIR] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_rmdir,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_diropargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_wccstatres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd3_diropargs),
+		.pc_ressize = sizeof(struct nfsd3_wccstatres),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+WC,
+	},
+	[NFS3PROC_RENAME] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_rename,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_renameargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_renameres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle2,
+		.pc_argsize = sizeof(struct nfsd3_renameargs),
+		.pc_ressize = sizeof(struct nfsd3_renameres),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+WC+WC,
+	},
+	[NFS3PROC_LINK] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_link,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_linkargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_linkres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle2,
+		.pc_argsize = sizeof(struct nfsd3_linkargs),
+		.pc_ressize = sizeof(struct nfsd3_linkres),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+pAT+WC,
+	},
+	[NFS3PROC_READDIR] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_readdir,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_readdirargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_readdirres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd3_readdirargs),
+		.pc_ressize = sizeof(struct nfsd3_readdirres),
+		.pc_cachetype = RC_NOCACHE,
+	},
+	[NFS3PROC_READDIRPLUS] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_readdirplus,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_readdirplusargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_readdirres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd3_readdirplusargs),
+		.pc_ressize = sizeof(struct nfsd3_readdirres),
+		.pc_cachetype = RC_NOCACHE,
+	},
+	[NFS3PROC_FSSTAT] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_fsstat,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_fhandleargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_fsstatres,
+		.pc_argsize = sizeof(struct nfsd3_fhandleargs),
+		.pc_ressize = sizeof(struct nfsd3_fsstatres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+pAT+2*6+1,
+	},
+	[NFS3PROC_FSINFO] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_fsinfo,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_fhandleargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_fsinfores,
+		.pc_argsize = sizeof(struct nfsd3_fhandleargs),
+		.pc_ressize = sizeof(struct nfsd3_fsinfores),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+pAT+12,
+	},
+	[NFS3PROC_PATHCONF] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_pathconf,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_fhandleargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_pathconfres,
+		.pc_argsize = sizeof(struct nfsd3_fhandleargs),
+		.pc_ressize = sizeof(struct nfsd3_pathconfres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+pAT+6,
+	},
+	[NFS3PROC_COMMIT] = {
+		.pc_func = (svc_procfunc) nfsd3_proc_commit,
+		.pc_decode = (kxdrproc_t) nfs3svc_decode_commitargs,
+		.pc_encode = (kxdrproc_t) nfs3svc_encode_commitres,
+		.pc_release = (kxdrproc_t) nfs3svc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd3_commitargs),
+		.pc_ressize = sizeof(struct nfsd3_commitres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+WC+2,
+	},
 };
 
 struct svc_version	nfsd_version3 = {
diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c
index e298e260b5f1..0eb9c820b7a6 100644
--- a/fs/nfsd/nfsproc.c
+++ b/fs/nfsd/nfsproc.c
@@ -533,45 +533,179 @@ nfsd_proc_statfs(struct svc_rqst * rqstp, struct nfsd_fhandle   *argp,
  * NFSv2 Server procedures.
  * Only the results of non-idempotent operations are cached.
  */
-#define nfsd_proc_none		NULL
-#define nfssvc_release_none	NULL
 struct nfsd_void { int dummy; };
 
-#define PROC(name, argt, rest, relt, cache, respsize)	\
- { (svc_procfunc) nfsd_proc_##name,		\
-   (kxdrproc_t) nfssvc_decode_##argt,		\
-   (kxdrproc_t) nfssvc_encode_##rest,		\
-   (kxdrproc_t) nfssvc_release_##relt,		\
-   sizeof(struct nfsd_##argt),			\
-   sizeof(struct nfsd_##rest),			\
-   0,						\
-   cache,					\
-   respsize,				       	\
- }
-
 #define ST 1		/* status */
 #define FH 8		/* filehandle */
 #define	AT 18		/* attributes */
 
 static struct svc_procedure		nfsd_procedures2[18] = {
-  PROC(null,	 void,		void,		none,		RC_NOCACHE, ST),
-  PROC(getattr,	 fhandle,	attrstat,	fhandle,	RC_NOCACHE, ST+AT),
-  PROC(setattr,  sattrargs,	attrstat,	fhandle,	RC_REPLBUFF, ST+AT),
-  PROC(none,	 void,		void,		none,		RC_NOCACHE, ST),
-  PROC(lookup,	 diropargs,	diropres,	fhandle,	RC_NOCACHE, ST+FH+AT),
-  PROC(readlink, readlinkargs,	readlinkres,	none,		RC_NOCACHE, ST+1+NFS_MAXPATHLEN/4),
-  PROC(read,	 readargs,	readres,	fhandle,	RC_NOCACHE, ST+AT+1+NFSSVC_MAXBLKSIZE_V2/4),
-  PROC(none,	 void,		void,		none,		RC_NOCACHE, ST),
-  PROC(write,	 writeargs,	attrstat,	fhandle,	RC_REPLBUFF, ST+AT),
-  PROC(create,	 createargs,	diropres,	fhandle,	RC_REPLBUFF, ST+FH+AT),
-  PROC(remove,	 diropargs,	void,		none,		RC_REPLSTAT, ST),
-  PROC(rename,	 renameargs,	void,		none,		RC_REPLSTAT, ST),
-  PROC(link,	 linkargs,	void,		none,		RC_REPLSTAT, ST),
-  PROC(symlink,	 symlinkargs,	void,		none,		RC_REPLSTAT, ST),
-  PROC(mkdir,	 createargs,	diropres,	fhandle,	RC_REPLBUFF, ST+FH+AT),
-  PROC(rmdir,	 diropargs,	void,		none,		RC_REPLSTAT, ST),
-  PROC(readdir,	 readdirargs,	readdirres,	none,		RC_NOCACHE, 0),
-  PROC(statfs,	 fhandle,	statfsres,	none,		RC_NOCACHE, ST+5),
+	[NFSPROC_NULL] = {
+		.pc_func = (svc_procfunc) nfsd_proc_null,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_void,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_void,
+		.pc_argsize = sizeof(struct nfsd_void),
+		.pc_ressize = sizeof(struct nfsd_void),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST,
+	},
+	[NFSPROC_GETATTR] = {
+		.pc_func = (svc_procfunc) nfsd_proc_getattr,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_fhandle,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_attrstat,
+		.pc_release = (kxdrproc_t) nfssvc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd_fhandle),
+		.pc_ressize = sizeof(struct nfsd_attrstat),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+AT,
+	},
+	[NFSPROC_SETATTR] = {
+		.pc_func = (svc_procfunc) nfsd_proc_setattr,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_sattrargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_attrstat,
+		.pc_release = (kxdrproc_t) nfssvc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd_sattrargs),
+		.pc_ressize = sizeof(struct nfsd_attrstat),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+AT,
+	},
+	[NFSPROC_ROOT] = {
+		.pc_decode = (kxdrproc_t) nfssvc_decode_void,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_void,
+		.pc_argsize = sizeof(struct nfsd_void),
+		.pc_ressize = sizeof(struct nfsd_void),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST,
+	},
+	[NFSPROC_LOOKUP] = {
+		.pc_func = (svc_procfunc) nfsd_proc_lookup,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_diropargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_diropres,
+		.pc_release = (kxdrproc_t) nfssvc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd_diropargs),
+		.pc_ressize = sizeof(struct nfsd_diropres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+FH+AT,
+	},
+	[NFSPROC_READLINK] = {
+		.pc_func = (svc_procfunc) nfsd_proc_readlink,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_readlinkargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_readlinkres,
+		.pc_argsize = sizeof(struct nfsd_readlinkargs),
+		.pc_ressize = sizeof(struct nfsd_readlinkres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+1+NFS_MAXPATHLEN/4,
+	},
+	[NFSPROC_READ] = {
+		.pc_func = (svc_procfunc) nfsd_proc_read,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_readargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_readres,
+		.pc_release = (kxdrproc_t) nfssvc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd_readargs),
+		.pc_ressize = sizeof(struct nfsd_readres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+AT+1+NFSSVC_MAXBLKSIZE_V2/4,
+	},
+	[NFSPROC_WRITECACHE] = {
+		.pc_decode = (kxdrproc_t) nfssvc_decode_void,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_void,
+		.pc_argsize = sizeof(struct nfsd_void),
+		.pc_ressize = sizeof(struct nfsd_void),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST,
+	},
+	[NFSPROC_WRITE] = {
+		.pc_func = (svc_procfunc) nfsd_proc_write,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_writeargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_attrstat,
+		.pc_release = (kxdrproc_t) nfssvc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd_writeargs),
+		.pc_ressize = sizeof(struct nfsd_attrstat),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+AT,
+	},
+	[NFSPROC_CREATE] = {
+		.pc_func = (svc_procfunc) nfsd_proc_create,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_createargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_diropres,
+		.pc_release = (kxdrproc_t) nfssvc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd_createargs),
+		.pc_ressize = sizeof(struct nfsd_diropres),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+FH+AT,
+	},
+	[NFSPROC_REMOVE] = {
+		.pc_func = (svc_procfunc) nfsd_proc_remove,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_diropargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_void,
+		.pc_argsize = sizeof(struct nfsd_diropargs),
+		.pc_ressize = sizeof(struct nfsd_void),
+		.pc_cachetype = RC_REPLSTAT,
+		.pc_xdrressize = ST,
+	},
+	[NFSPROC_RENAME] = {
+		.pc_func = (svc_procfunc) nfsd_proc_rename,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_renameargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_void,
+		.pc_argsize = sizeof(struct nfsd_renameargs),
+		.pc_ressize = sizeof(struct nfsd_void),
+		.pc_cachetype = RC_REPLSTAT,
+		.pc_xdrressize = ST,
+	},
+	[NFSPROC_LINK] = {
+		.pc_func = (svc_procfunc) nfsd_proc_link,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_linkargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_void,
+		.pc_argsize = sizeof(struct nfsd_linkargs),
+		.pc_ressize = sizeof(struct nfsd_void),
+		.pc_cachetype = RC_REPLSTAT,
+		.pc_xdrressize = ST,
+	},
+	[NFSPROC_SYMLINK] = {
+		.pc_func = (svc_procfunc) nfsd_proc_symlink,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_symlinkargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_void,
+		.pc_argsize = sizeof(struct nfsd_symlinkargs),
+		.pc_ressize = sizeof(struct nfsd_void),
+		.pc_cachetype = RC_REPLSTAT,
+		.pc_xdrressize = ST,
+	},
+	[NFSPROC_MKDIR] = {
+		.pc_func = (svc_procfunc) nfsd_proc_mkdir,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_createargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_diropres,
+		.pc_release = (kxdrproc_t) nfssvc_release_fhandle,
+		.pc_argsize = sizeof(struct nfsd_createargs),
+		.pc_ressize = sizeof(struct nfsd_diropres),
+		.pc_cachetype = RC_REPLBUFF,
+		.pc_xdrressize = ST+FH+AT,
+	},
+	[NFSPROC_RMDIR] = {
+		.pc_func = (svc_procfunc) nfsd_proc_rmdir,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_diropargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_void,
+		.pc_argsize = sizeof(struct nfsd_diropargs),
+		.pc_ressize = sizeof(struct nfsd_void),
+		.pc_cachetype = RC_REPLSTAT,
+		.pc_xdrressize = ST,
+	},
+	[NFSPROC_READDIR] = {
+		.pc_func = (svc_procfunc) nfsd_proc_readdir,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_readdirargs,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_readdirres,
+		.pc_argsize = sizeof(struct nfsd_readdirargs),
+		.pc_ressize = sizeof(struct nfsd_readdirres),
+		.pc_cachetype = RC_NOCACHE,
+	},
+	[NFSPROC_STATFS] = {
+		.pc_func = (svc_procfunc) nfsd_proc_statfs,
+		.pc_decode = (kxdrproc_t) nfssvc_decode_fhandle,
+		.pc_encode = (kxdrproc_t) nfssvc_encode_statfsres,
+		.pc_argsize = sizeof(struct nfsd_fhandle),
+		.pc_ressize = sizeof(struct nfsd_statfsres),
+		.pc_cachetype = RC_NOCACHE,
+		.pc_xdrressize = ST+5,
+	},
 };
 
 
-- 
cgit v1.2.3-59-g8ed1b


From c90173f0907486fe4010c2a8cef534e2473db43f Mon Sep 17 00:00:00 2001
From: Amul Saha <amul.saha@samsung.com>
Date: Tue, 16 Jun 2009 11:24:01 +0530
Subject: mtd: OneNAND: Allow setting of boundary information when built as
 module

This patch unifies the flex_bdry setting for module vs. built-in
configuration of OneNAND.

Signed-off-by: Amul Kumar Saha <amul.saha@samsung.com>
Signed-off-by: Vishak G <vishak.g@samsung.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/onenand/onenand_base.c | 28 +++++++++-------------------
 1 file changed, 9 insertions(+), 19 deletions(-)

diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
index 864327ed7fb3..6e829095ea9d 100644
--- a/drivers/mtd/onenand/onenand_base.c
+++ b/drivers/mtd/onenand/onenand_base.c
@@ -20,6 +20,7 @@
 
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/moduleparam.h>
 #include <linux/init.h>
 #include <linux/sched.h>
 #include <linux/delay.h>
@@ -34,6 +35,14 @@
 /* Default Flex-OneNAND boundary and lock respectively */
 static int flex_bdry[MAX_DIES * 2] = { -1, 0, -1, 0 };
 
+module_param_array(flex_bdry, int, NULL, 0400);
+MODULE_PARM_DESC(flex_bdry,	"SLC Boundary information for Flex-OneNAND"
+				"Syntax:flex_bdry=DIE_BDRY,LOCK,..."
+				"DIE_BDRY: SLC boundary of the die"
+				"LOCK: Locking information for SLC boundary"
+				"    : 0->Set boundary in unlocked status"
+				"    : 1->Set boundary in locked status");
+
 /**
  *  onenand_oob_128 - oob info for Flex-Onenand with 4KB page
  *  For now, we expose only 64 out of 80 ecc bytes
@@ -3258,25 +3267,6 @@ out:
 	return ret;
 }
 
-/**
- * flexonenand_setup - 	capture Flex-OneNAND boundary and lock
- * 			values  passed as kernel parameters
- * @param s	kernel parameter string
- */
-static int flexonenand_setup(char *s)
-{
-	int ints[5], i;
-
-	s = get_options(s, 5, ints);
-
-	for (i = 0; i < ints[0]; i++)
-		flex_bdry[i] = ints[i + 1];
-
-	return 1;
-}
-
-__setup("onenand.bdry=", flexonenand_setup);
-
 /**
  * onenand_probe - [OneNAND Interface] Probe the OneNAND device
  * @param mtd		MTD device structure
-- 
cgit v1.2.3-59-g8ed1b


From eb73e6221c5013b32647bcf44bcc570232ba2804 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marek.vasut@gmail.com>
Date: Mon, 1 Jun 2009 22:34:14 +0200
Subject: [ARM] pxa/palm: various fixes for PalmZ72 (mostly audio asoc and usb)

Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
---
 arch/arm/mach-pxa/include/mach/palmz72.h |  5 +--
 arch/arm/mach-pxa/palmz72.c              | 65 +++++++++++++++++++++++++++++++-
 2 files changed, 65 insertions(+), 5 deletions(-)

diff --git a/arch/arm/mach-pxa/include/mach/palmz72.h b/arch/arm/mach-pxa/include/mach/palmz72.h
index 5032307ebf7d..2806ef69ba5a 100644
--- a/arch/arm/mach-pxa/include/mach/palmz72.h
+++ b/arch/arm/mach-pxa/include/mach/palmz72.h
@@ -21,7 +21,7 @@
 /* SD/MMC */
 #define GPIO_NR_PALMZ72_SD_DETECT_N		14
 #define GPIO_NR_PALMZ72_SD_POWER_N		98
-#define GPIO_NR_PALMZ72_SD_RO 115
+#define GPIO_NR_PALMZ72_SD_RO			115
 
 /* Touchscreen */
 #define GPIO_NR_PALMZ72_WM9712_IRQ		27
@@ -31,8 +31,7 @@
 
 /* USB */
 #define GPIO_NR_PALMZ72_USB_DETECT_N		15
-#define GPIO_NR_PALMZ72_USB_POWER		95
-#define GPIO_NR_PALMZ72_USB_PULLUP		12
+#define GPIO_NR_PALMZ72_USB_PULLUP		95
 
 /* LCD/Backlight */
 #define GPIO_NR_PALMZ72_BL_POWER		20
diff --git a/arch/arm/mach-pxa/palmz72.c b/arch/arm/mach-pxa/palmz72.c
index b88eb4dd2c84..c3645aa3fa3d 100644
--- a/arch/arm/mach-pxa/palmz72.c
+++ b/arch/arm/mach-pxa/palmz72.c
@@ -27,7 +27,9 @@
 #include <linux/pda_power.h>
 #include <linux/pwm_backlight.h>
 #include <linux/gpio.h>
+#include <linux/wm97xx_batt.h>
 #include <linux/power_supply.h>
+#include <linux/usb/gpio_vbus.h>
 
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
@@ -41,6 +43,8 @@
 #include <mach/irda.h>
 #include <mach/pxa27x_keypad.h>
 #include <mach/udc.h>
+#include <mach/palmasoc.h>
+
 #include <mach/pm.h>
 
 #include "generic.h"
@@ -66,6 +70,8 @@ static unsigned long palmz72_pin_config[] __initdata = {
 	GPIO29_AC97_SDATA_IN_0,
 	GPIO30_AC97_SDATA_OUT,
 	GPIO31_AC97_SYNC,
+	GPIO89_AC97_SYSCLK,
+	GPIO113_AC97_nRESET,
 
 	/* IrDA */
 	GPIO49_GPIO,	/* ir disable */
@@ -77,8 +83,7 @@ static unsigned long palmz72_pin_config[] __initdata = {
 
 	/* USB */
 	GPIO15_GPIO,	/* usb detect */
-	GPIO12_GPIO,	/* usb pullup */
-	GPIO95_GPIO,	/* usb power */
+	GPIO95_GPIO,	/* usb pullup */
 
 	/* Matrix keypad */
 	GPIO100_KP_MKIN_0	| WAKEUP_ON_LEVEL_HIGH,
@@ -354,6 +359,22 @@ static struct platform_device palmz72_leds = {
 	}
 };
 
+/******************************************************************************
+ * UDC
+ ******************************************************************************/
+static struct gpio_vbus_mach_info palmz72_udc_info = {
+	.gpio_vbus		= GPIO_NR_PALMZ72_USB_DETECT_N,
+	.gpio_pullup		= GPIO_NR_PALMZ72_USB_PULLUP,
+};
+
+static struct platform_device palmz72_gpio_vbus = {
+	.name	= "gpio-vbus",
+	.id	= -1,
+	.dev	= {
+		.platform_data	= &palmz72_udc_info,
+	},
+};
+
 /******************************************************************************
  * Power supply
  ******************************************************************************/
@@ -421,6 +442,31 @@ static struct platform_device power_supply = {
 	},
 };
 
+/******************************************************************************
+ * WM97xx battery
+ ******************************************************************************/
+static struct wm97xx_batt_info wm97xx_batt_pdata = {
+	.batt_aux	= WM97XX_AUX_ID3,
+	.temp_aux	= WM97XX_AUX_ID2,
+	.charge_gpio	= -1,
+	.max_voltage	= PALMZ72_BAT_MAX_VOLTAGE,
+	.min_voltage	= PALMZ72_BAT_MIN_VOLTAGE,
+	.batt_mult	= 1000,
+	.batt_div	= 414,
+	.temp_mult	= 1,
+	.temp_div	= 1,
+	.batt_tech	= POWER_SUPPLY_TECHNOLOGY_LIPO,
+	.batt_name	= "main-batt",
+};
+
+/******************************************************************************
+ * aSoC audio
+ ******************************************************************************/
+static struct platform_device palmz72_asoc = {
+	.name = "palm27x-asoc",
+	.id   = -1,
+};
+
 /******************************************************************************
  * Framebuffer
  ******************************************************************************/
@@ -527,17 +573,32 @@ device_initcall(palmz72_pm_init);
 static struct platform_device *devices[] __initdata = {
 	&palmz72_backlight,
 	&palmz72_leds,
+	&palmz72_asoc,
 	&power_supply,
+	&palmz72_gpio_vbus,
 };
 
+/* setup udc GPIOs initial state */
+static void __init palmz72_udc_init(void)
+{
+	if (!gpio_request(GPIO_NR_PALMZ72_USB_PULLUP, "USB Pullup")) {
+		gpio_direction_output(GPIO_NR_PALMZ72_USB_PULLUP, 0);
+		gpio_free(GPIO_NR_PALMZ72_USB_PULLUP);
+	}
+}
+
 static void __init palmz72_init(void)
 {
 	pxa2xx_mfp_config(ARRAY_AND_SIZE(palmz72_pin_config));
+
 	set_pxa_fb_info(&palmz72_lcd_screen);
 	pxa_set_mci_info(&palmz72_mci_platform_data);
+	palmz72_udc_init();
 	pxa_set_ac97_info(NULL);
 	pxa_set_ficp_info(&palmz72_ficp_platform_data);
 	pxa_set_keypad_info(&palmz72_keypad_platform_data);
+	wm97xx_bat_set_pdata(&wm97xx_batt_pdata);
+
 	platform_add_devices(devices, ARRAY_SIZE(devices));
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 3cb05c97ad38aab0858fc176fd1577ac4a947bad Mon Sep 17 00:00:00 2001
From: Mike Rapoport <mike@compulab.co.il>
Date: Tue, 16 Jun 2009 15:19:24 +0300
Subject: [ARM] pxa/em-x270: remove .gpio_cs from em_x270_libertas_pdata

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
---
 arch/arm/mach-pxa/em-x270.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/mach-pxa/em-x270.c b/arch/arm/mach-pxa/em-x270.c
index 243e0802b5f4..b2d5a5394e47 100644
--- a/arch/arm/mach-pxa/em-x270.c
+++ b/arch/arm/mach-pxa/em-x270.c
@@ -735,6 +735,7 @@ static struct pxa2xx_spi_chip em_x270_libertas_chip = {
 	.rx_threshold	= 1,
 	.tx_threshold	= 1,
 	.timeout	= 1000,
+	.gpio_cs	= 14,
 };
 
 static unsigned long em_x270_libertas_pin_config[] = {
@@ -803,7 +804,6 @@ static int em_x270_libertas_teardown(struct spi_device *spi)
 
 struct libertas_spi_platform_data em_x270_libertas_pdata = {
 	.use_dummy_writes	= 1,
-	.gpio_cs		= 14,
 	.setup			= em_x270_libertas_setup,
 	.teardown		= em_x270_libertas_teardown,
 };
-- 
cgit v1.2.3-59-g8ed1b


From a37ed43115b70e832c7e8cc7dabd026d127ad9db Mon Sep 17 00:00:00 2001
From: Mike Rapoport <mike@compulab.co.il>
Date: Mon, 18 May 2009 08:53:56 +0300
Subject: [ARM] pxa/em-x270: add ability to control GPS and GPRS power

Register userspace-consumer devices to allow userspace control of
voltage regulators supplying GPS and GPRS modules

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Signed-off-by: Eric Miao <eric.miao@marvell.com>
---
 arch/arm/mach-pxa/em-x270.c | 53 +++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 51 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-pxa/em-x270.c b/arch/arm/mach-pxa/em-x270.c
index b2d5a5394e47..5184a570ac5f 100644
--- a/arch/arm/mach-pxa/em-x270.c
+++ b/arch/arm/mach-pxa/em-x270.c
@@ -30,6 +30,7 @@
 #include <linux/apm-emulation.h>
 #include <linux/i2c.h>
 #include <linux/i2c/pca953x.h>
+#include <linux/regulator/userspace-consumer.h>
 
 #include <media/soc_camera.h>
 
@@ -1038,6 +1039,52 @@ static void  __init em_x270_init_camera(void)
 static inline void em_x270_init_camera(void) {}
 #endif
 
+static struct regulator_bulk_data em_x270_gps_consumer_supply = {
+	.supply		= "vcc gps",
+};
+
+static struct regulator_userspace_consumer_data em_x270_gps_consumer_data = {
+	.name		= "vcc gps",
+	.num_supplies	= 1,
+	.supplies	= &em_x270_gps_consumer_supply,
+};
+
+static struct platform_device em_x270_gps_userspace_consumer = {
+	.name		= "reg-userspace-consumer",
+	.id		= 0,
+	.dev		= {
+		.platform_data = &em_x270_gps_consumer_data,
+	},
+};
+
+static struct regulator_bulk_data em_x270_gprs_consumer_supply = {
+	.supply		= "vcc gprs",
+};
+
+static struct regulator_userspace_consumer_data em_x270_gprs_consumer_data = {
+	.name		= "vcc gprs",
+	.num_supplies	= 1,
+	.supplies	= &em_x270_gprs_consumer_supply
+};
+
+static struct platform_device em_x270_gprs_userspace_consumer = {
+	.name		= "reg-userspace-consumer",
+	.id		= 1,
+	.dev		= {
+		.platform_data = &em_x270_gprs_consumer_data,
+	}
+};
+
+static struct platform_device *em_x270_userspace_consumers[] = {
+	&em_x270_gps_userspace_consumer,
+	&em_x270_gprs_userspace_consumer,
+};
+
+static void __init em_x270_userspace_consumers_init(void)
+{
+	platform_add_devices(ARRAY_AND_SIZE(em_x270_userspace_consumers));
+}
+
 /* DA9030 related initializations */
 #define REGULATOR_CONSUMER(_name, _dev, _supply)			       \
 	static struct regulator_consumer_supply _name##_consumers[] = {	\
@@ -1047,11 +1094,11 @@ static inline void em_x270_init_camera(void) {}
 		},							\
 	}
 
-REGULATOR_CONSUMER(ldo3, NULL, "vcc gps");
+REGULATOR_CONSUMER(ldo3, &em_x270_gps_userspace_consumer.dev, "vcc gps");
 REGULATOR_CONSUMER(ldo5, NULL, "vcc cam");
 REGULATOR_CONSUMER(ldo10, &pxa_device_mci.dev, "vcc sdio");
 REGULATOR_CONSUMER(ldo12, NULL, "vcc usb");
-REGULATOR_CONSUMER(ldo19, NULL, "vcc gprs");
+REGULATOR_CONSUMER(ldo19, &em_x270_gprs_userspace_consumer.dev, "vcc gprs");
 
 #define REGULATOR_INIT(_ldo, _min_uV, _max_uV, _ops_mask)		\
 	static struct regulator_init_data _ldo##_data = {		\
@@ -1062,6 +1109,7 @@ REGULATOR_CONSUMER(ldo19, NULL, "vcc gprs");
 				.enabled = 0,				\
 			},						\
 			.valid_ops_mask = _ops_mask,			\
+			.apply_uV = 1,					\
 		},							\
 		.num_consumer_supplies = ARRAY_SIZE(_ldo##_consumers),	\
 		.consumer_supplies = _ldo##_consumers,			\
@@ -1240,6 +1288,7 @@ static void __init em_x270_init(void)
 	em_x270_init_spi();
 	em_x270_init_i2c();
 	em_x270_init_camera();
+	em_x270_userspace_consumers_init();
 }
 
 MACHINE_START(EM_X270, "Compulab EM-X270")
-- 
cgit v1.2.3-59-g8ed1b


From c63a5751d971b07d8cd80f4d4aae75c676665a3b Mon Sep 17 00:00:00 2001
From: Mike Rapoport <mike@compulab.co.il>
Date: Mon, 11 May 2009 15:06:16 +0300
Subject: [ARM] pxa/em-x270: always register AC97 controller device

After commit 6b849bcff0004aa5dd216b4d3eb56f51c9df8a72 (ASoC: Convert PXA
AC97 driver to probe with the platform device) AC97 controller device
should be registered explicitly for both ASoC and generic AC97 drivers

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
---
 arch/arm/mach-pxa/em-x270.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-pxa/em-x270.c b/arch/arm/mach-pxa/em-x270.c
index 5184a570ac5f..5343d1f339d6 100644
--- a/arch/arm/mach-pxa/em-x270.c
+++ b/arch/arm/mach-pxa/em-x270.c
@@ -839,10 +839,14 @@ static void __init em_x270_init_spi(void)
 static inline void em_x270_init_spi(void) {}
 #endif
 
-#if defined(CONFIG_SND_PXA2XX_AC97) || defined(CONFIG_SND_PXA2XX_AC97_MODULE)
+#if defined(CONFIG_SND_PXA2XX_AC97_LIB)
+static pxa2xx_audio_ops_t em_x270_ac97_info = {
+	.reset_gpio = 113,
+};
+
 static void __init em_x270_init_ac97(void)
 {
-	pxa_set_ac97_info(NULL);
+	pxa_set_ac97_info(&em_x270_ac97_info);
 }
 #else
 static inline void em_x270_init_ac97(void) {}
-- 
cgit v1.2.3-59-g8ed1b


From 303f2d7dc8eddc65de27d97eab0fc1c04ec4d2e1 Mon Sep 17 00:00:00 2001
From: Mike Rapoport <mike@compulab.co.il>
Date: Sun, 14 Jun 2009 12:04:19 +0300
Subject: [ARM] pxa/em-x270: fix type in SND_PXA2XX_LIB_AC97

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
---
 arch/arm/mach-pxa/em-x270.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/mach-pxa/em-x270.c b/arch/arm/mach-pxa/em-x270.c
index 5343d1f339d6..63b10d9bb1d3 100644
--- a/arch/arm/mach-pxa/em-x270.c
+++ b/arch/arm/mach-pxa/em-x270.c
@@ -839,7 +839,7 @@ static void __init em_x270_init_spi(void)
 static inline void em_x270_init_spi(void) {}
 #endif
 
-#if defined(CONFIG_SND_PXA2XX_AC97_LIB)
+#if defined(CONFIG_SND_PXA2XX_LIB_AC97)
 static pxa2xx_audio_ops_t em_x270_ac97_info = {
 	.reset_gpio = 113,
 };
-- 
cgit v1.2.3-59-g8ed1b


From e6c3f4b89bf3698a6994d30de7a16ae395a81dab Mon Sep 17 00:00:00 2001
From: Tomas 'Sleep_Walker' Cech <sleep_walker@suse.cz>
Date: Mon, 18 May 2009 15:24:14 +0200
Subject: [ARM] pxa/treo680: initial support
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Tomáš Čech <sleep_walker@suse.cz>
Acked-by: Marek Vasut <marek.vasut@gmail.com>
Signed-off-by: Eric Miao <eric.miao@marvell.com>
---
 arch/arm/mach-pxa/Kconfig                |  10 +
 arch/arm/mach-pxa/Makefile               |   1 +
 arch/arm/mach-pxa/include/mach/treo680.h |  49 +++
 arch/arm/mach-pxa/treo680.c              | 612 +++++++++++++++++++++++++++++++
 arch/arm/mm/mmu.c                        |   7 +
 5 files changed, 679 insertions(+)
 create mode 100644 arch/arm/mach-pxa/include/mach/treo680.h
 create mode 100644 arch/arm/mach-pxa/treo680.c

diff --git a/arch/arm/mach-pxa/Kconfig b/arch/arm/mach-pxa/Kconfig
index f4533f8ff4e8..89c992b8f75b 100644
--- a/arch/arm/mach-pxa/Kconfig
+++ b/arch/arm/mach-pxa/Kconfig
@@ -401,6 +401,16 @@ config MACH_PALMZ72
 	  Say Y here if you intend to run this kernel on Palm Zire 72
 	  handheld computer.
 
+config MACH_TREO680
+	bool "Palm Treo 680"
+	default y
+	depends on ARCH_PXA_PALM
+	select PXA27x
+	select IWMMXT
+	help
+	  Say Y here if you intend to run this kernel on Palm Treo 680
+	  smartphone.
+
 config MACH_PALMLD
 	bool "Palm LifeDrive"
 	default y
diff --git a/arch/arm/mach-pxa/Makefile b/arch/arm/mach-pxa/Makefile
index d18ffef44b8c..d4c6122a342f 100644
--- a/arch/arm/mach-pxa/Makefile
+++ b/arch/arm/mach-pxa/Makefile
@@ -62,6 +62,7 @@ obj-$(CONFIG_MACH_PALMT5)	+= palmt5.o
 obj-$(CONFIG_MACH_PALMTX)	+= palmtx.o
 obj-$(CONFIG_MACH_PALMLD)	+= palmld.o
 obj-$(CONFIG_MACH_PALMZ72)	+= palmz72.o
+obj-$(CONFIG_MACH_TREO680)	+= treo680.o
 obj-$(CONFIG_ARCH_VIPER)	+= viper.o
 
 ifeq ($(CONFIG_MACH_ZYLONITE),y)
diff --git a/arch/arm/mach-pxa/include/mach/treo680.h b/arch/arm/mach-pxa/include/mach/treo680.h
new file mode 100644
index 000000000000..af443b24d99a
--- /dev/null
+++ b/arch/arm/mach-pxa/include/mach/treo680.h
@@ -0,0 +1,49 @@
+/*
+ * GPIOs and interrupts for Palm Treo 680 smartphone
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#ifndef _INCLUDE_TREO680_H_
+#define _INCLUDE_TREO680_H_
+
+/* GPIOs */
+#define GPIO_NR_TREO680_POWER_DETECT	0
+#define GPIO_NR_TREO680_AMP_EN		27
+#define GPIO_NR_TREO680_KEYB_BL		24
+#define GPIO_NR_TREO680_VIBRATE_EN	44
+#define GPIO_NR_TREO680_GREEN_LED	20
+#define GPIO_NR_TREO680_RED_LED		79
+#define GPIO_NR_TREO680_SD_DETECT_N	113
+#define GPIO_NR_TREO680_SD_READONLY	33
+#define GPIO_NR_TREO680_EP_DETECT_N	116
+#define GPIO_NR_TREO680_SD_POWER	42
+#define GPIO_NR_TREO680_USB_DETECT	1
+#define GPIO_NR_TREO680_USB_PULLUP	114
+#define GPIO_NR_TREO680_GSM_POWER	40
+#define GPIO_NR_TREO680_GSM_RESET	87
+#define GPIO_NR_TREO680_GSM_WAKE	57
+#define GPIO_NR_TREO680_GSM_HOST_WAKE	14
+#define GPIO_NR_TREO680_GSM_TRIGGER	10
+#define GPIO_NR_TREO680_BT_EN		43
+#define GPIO_NR_TREO680_IR_EN		115
+#define GPIO_NR_TREO680_IR_TXD		47
+#define GPIO_NR_TREO680_BL_POWER	38
+#define GPIO_NR_TREO680_LCD_POWER	25
+
+/* Various addresses  */
+#define TREO680_PHYS_RAM_START	0xa0000000
+#define TREO680_PHYS_IO_START	0x40000000
+#define TREO680_STR_BASE	0xa2000000
+
+/* BACKLIGHT */
+#define TREO680_MAX_INTENSITY		254
+#define TREO680_DEFAULT_INTENSITY	160
+#define TREO680_LIMIT_MASK		0x7F
+#define TREO680_PRESCALER		63
+#define TREO680_PERIOD_NS		3500
+
+#endif
diff --git a/arch/arm/mach-pxa/treo680.c b/arch/arm/mach-pxa/treo680.c
new file mode 100644
index 000000000000..a06f19edebb3
--- /dev/null
+++ b/arch/arm/mach-pxa/treo680.c
@@ -0,0 +1,612 @@
+/*
+ * Hardware definitions for Palm Treo 680
+ *
+ * Author:     Tomas Cech <sleep_walker@suse.cz>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * (find more info at www.hackndev.com)
+ *
+ */
+
+#include <linux/platform_device.h>
+#include <linux/delay.h>
+#include <linux/irq.h>
+#include <linux/gpio_keys.h>
+#include <linux/input.h>
+#include <linux/pda_power.h>
+#include <linux/pwm_backlight.h>
+#include <linux/gpio.h>
+#include <linux/wm97xx_batt.h>
+#include <linux/power_supply.h>
+#include <linux/sysdev.h>
+#include <linux/w1-gpio.h>
+
+#include <asm/mach-types.h>
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+
+#include <mach/pxa27x.h>
+#include <mach/pxa27x-udc.h>
+#include <mach/audio.h>
+#include <mach/treo680.h>
+#include <mach/mmc.h>
+#include <mach/pxafb.h>
+#include <mach/irda.h>
+#include <mach/pxa27x_keypad.h>
+#include <mach/udc.h>
+#include <mach/ohci.h>
+#include <mach/pxa2xx-regs.h>
+#include <mach/palmasoc.h>
+#include <mach/camera.h>
+
+#include <sound/pxa2xx-lib.h>
+
+#include "generic.h"
+#include "devices.h"
+
+/******************************************************************************
+ * Pin configuration
+ ******************************************************************************/
+static unsigned long treo680_pin_config[] __initdata = {
+	/* MMC */
+	GPIO32_MMC_CLK,
+	GPIO92_MMC_DAT_0,
+	GPIO109_MMC_DAT_1,
+	GPIO110_MMC_DAT_2,
+	GPIO111_MMC_DAT_3,
+	GPIO112_MMC_CMD,
+	GPIO33_GPIO,				/* SD read only */
+	GPIO113_GPIO,				/* SD detect */
+
+	/* AC97 */
+	GPIO28_AC97_BITCLK,
+	GPIO29_AC97_SDATA_IN_0,
+	GPIO30_AC97_SDATA_OUT,
+	GPIO31_AC97_SYNC,
+	GPIO89_AC97_SYSCLK,
+	GPIO95_AC97_nRESET,
+
+	/* IrDA */
+	GPIO46_FICP_RXD,
+	GPIO47_FICP_TXD,
+
+	/* PWM */
+	GPIO16_PWM0_OUT,
+
+	/* USB */
+	GPIO1_GPIO | WAKEUP_ON_EDGE_BOTH,	/* usb detect */
+
+	/* MATRIX KEYPAD */
+	GPIO100_KP_MKIN_0 | WAKEUP_ON_LEVEL_HIGH,
+	GPIO101_KP_MKIN_1,
+	GPIO102_KP_MKIN_2,
+	GPIO97_KP_MKIN_3,
+	GPIO98_KP_MKIN_4,
+	GPIO99_KP_MKIN_5,
+	GPIO91_KP_MKIN_6,
+	GPIO13_KP_MKIN_7,
+	GPIO103_KP_MKOUT_0 | MFP_LPM_DRIVE_HIGH,
+	GPIO104_KP_MKOUT_1,
+	GPIO105_KP_MKOUT_2,
+	GPIO106_KP_MKOUT_3,
+	GPIO107_KP_MKOUT_4,
+	GPIO108_KP_MKOUT_5,
+	GPIO96_KP_MKOUT_6,
+	GPIO93_KP_DKIN_0 | WAKEUP_ON_LEVEL_HIGH,	/* Hotsync button */
+
+	/* LCD */
+	GPIO58_LCD_LDD_0,
+	GPIO59_LCD_LDD_1,
+	GPIO60_LCD_LDD_2,
+	GPIO61_LCD_LDD_3,
+	GPIO62_LCD_LDD_4,
+	GPIO63_LCD_LDD_5,
+	GPIO64_LCD_LDD_6,
+	GPIO65_LCD_LDD_7,
+	GPIO66_LCD_LDD_8,
+	GPIO67_LCD_LDD_9,
+	GPIO68_LCD_LDD_10,
+	GPIO69_LCD_LDD_11,
+	GPIO70_LCD_LDD_12,
+	GPIO71_LCD_LDD_13,
+	GPIO72_LCD_LDD_14,
+	GPIO73_LCD_LDD_15,
+	GPIO74_LCD_FCLK,
+	GPIO75_LCD_LCLK,
+	GPIO76_LCD_PCLK,
+
+	/* Quick Capture Interface */
+	GPIO84_CIF_FV,
+	GPIO85_CIF_LV,
+	GPIO53_CIF_MCLK,
+	GPIO54_CIF_PCLK,
+	GPIO81_CIF_DD_0,
+	GPIO55_CIF_DD_1,
+	GPIO51_CIF_DD_2,
+	GPIO50_CIF_DD_3,
+	GPIO52_CIF_DD_4,
+	GPIO48_CIF_DD_5,
+	GPIO17_CIF_DD_6,
+	GPIO12_CIF_DD_7,
+
+	/* I2C */
+	GPIO117_I2C_SCL,
+	GPIO118_I2C_SDA,
+
+	/* GSM */
+	GPIO14_GPIO | WAKEUP_ON_EDGE_BOTH,	/* GSM host wake up */
+	GPIO34_FFUART_RXD,
+	GPIO35_FFUART_CTS,
+	GPIO39_FFUART_TXD,
+	GPIO41_FFUART_RTS,
+
+	/* MISC. */
+	GPIO0_GPIO | WAKEUP_ON_EDGE_BOTH,	/* external power detect */
+	GPIO15_GPIO | WAKEUP_ON_EDGE_BOTH,	/* silent switch */
+	GPIO116_GPIO,				/* headphone detect */
+	GPIO11_GPIO | WAKEUP_ON_EDGE_BOTH,	/* bluetooth host wake up */
+};
+
+/******************************************************************************
+ * SD/MMC card controller
+ ******************************************************************************/
+static int treo680_mci_init(struct device *dev,
+		irq_handler_t treo680_detect_int, void *data)
+{
+	int err = 0;
+
+	/* Setup an interrupt for detecting card insert/remove events */
+	err = gpio_request(GPIO_NR_TREO680_SD_DETECT_N, "SD IRQ");
+
+	if (err)
+		goto err;
+
+	err = gpio_direction_input(GPIO_NR_TREO680_SD_DETECT_N);
+	if (err)
+		goto err2;
+
+	err = request_irq(gpio_to_irq(GPIO_NR_TREO680_SD_DETECT_N),
+			treo680_detect_int, IRQF_DISABLED | IRQF_SAMPLE_RANDOM |
+			IRQF_TRIGGER_FALLING | IRQF_TRIGGER_RISING,
+			"SD/MMC card detect", data);
+
+	if (err) {
+		dev_err(dev, "%s: cannot request SD/MMC card detect IRQ\n",
+			     __func__);
+		goto err2;
+	}
+
+	err = gpio_request(GPIO_NR_TREO680_SD_POWER, "SD_POWER");
+	if (err)
+		goto err3;
+
+	err = gpio_direction_output(GPIO_NR_TREO680_SD_POWER, 1);
+	if (err)
+		goto err4;
+
+	err = gpio_request(GPIO_NR_TREO680_SD_READONLY, "SD_READONLY");
+	if (err)
+		goto err4;
+
+	err = gpio_direction_input(GPIO_NR_TREO680_SD_READONLY);
+	if (err)
+		goto err5;
+
+	return 0;
+
+err5:
+	gpio_free(GPIO_NR_TREO680_SD_READONLY);
+err4:
+	gpio_free(GPIO_NR_TREO680_SD_POWER);
+err3:
+	free_irq(gpio_to_irq(GPIO_NR_TREO680_SD_DETECT_N), data);
+err2:
+	gpio_free(GPIO_NR_TREO680_SD_DETECT_N);
+err:
+	return err;
+}
+
+static void treo680_mci_exit(struct device *dev, void *data)
+{
+	gpio_free(GPIO_NR_TREO680_SD_READONLY);
+	gpio_free(GPIO_NR_TREO680_SD_POWER);
+	free_irq(gpio_to_irq(GPIO_NR_TREO680_SD_DETECT_N), data);
+	gpio_free(GPIO_NR_TREO680_SD_DETECT_N);
+}
+
+static void treo680_mci_power(struct device *dev, unsigned int vdd)
+{
+	struct pxamci_platform_data *p_d = dev->platform_data;
+	gpio_set_value(GPIO_NR_TREO680_SD_POWER, p_d->ocr_mask & (1 << vdd));
+}
+
+static int treo680_mci_get_ro(struct device *dev)
+{
+	return gpio_get_value(GPIO_NR_TREO680_SD_READONLY);
+}
+
+static struct pxamci_platform_data treo680_mci_platform_data = {
+	.ocr_mask	= MMC_VDD_32_33 | MMC_VDD_33_34,
+	.setpower	= treo680_mci_power,
+	.get_ro		= treo680_mci_get_ro,
+	.init 		= treo680_mci_init,
+	.exit		= treo680_mci_exit,
+};
+
+/******************************************************************************
+ * GPIO keyboard
+ ******************************************************************************/
+static unsigned int treo680_matrix_keys[] = {
+	KEY(0, 0, KEY_F8),		/* Red/Off/Power */
+	KEY(0, 1, KEY_LEFT),
+	KEY(0, 2, KEY_LEFTCTRL),	/* Alternate */
+	KEY(0, 3, KEY_L),
+	KEY(0, 4, KEY_A),
+	KEY(0, 5, KEY_Q),
+	KEY(0, 6, KEY_P),
+
+	KEY(1, 0, KEY_RIGHTCTRL),	/* Menu */
+	KEY(1, 1, KEY_RIGHT),
+	KEY(1, 2, KEY_LEFTSHIFT),	/* Left shift */
+	KEY(1, 3, KEY_Z),
+	KEY(1, 4, KEY_S),
+	KEY(1, 5, KEY_W),
+
+	KEY(2, 0, KEY_F1),		/* Phone */
+	KEY(2, 1, KEY_UP),
+	KEY(2, 2, KEY_0),
+	KEY(2, 3, KEY_X),
+	KEY(2, 4, KEY_D),
+	KEY(2, 5, KEY_E),
+
+	KEY(3, 0, KEY_F10),		/* Calendar */
+	KEY(3, 1, KEY_DOWN),
+	KEY(3, 2, KEY_SPACE),
+	KEY(3, 3, KEY_C),
+	KEY(3, 4, KEY_F),
+	KEY(3, 5, KEY_R),
+
+	KEY(4, 0, KEY_F12),		/* Mail */
+	KEY(4, 1, KEY_KPENTER),
+	KEY(4, 2, KEY_RIGHTALT),	/* Alt */
+	KEY(4, 3, KEY_V),
+	KEY(4, 4, KEY_G),
+	KEY(4, 5, KEY_T),
+
+	KEY(5, 0, KEY_F9),		/* Home */
+	KEY(5, 1, KEY_PAGEUP),		/* Side up */
+	KEY(5, 2, KEY_DOT),
+	KEY(5, 3, KEY_B),
+	KEY(5, 4, KEY_H),
+	KEY(5, 5, KEY_Y),
+
+	KEY(6, 0, KEY_TAB),		/* Side Activate */
+	KEY(6, 1, KEY_PAGEDOWN),	/* Side down */
+	KEY(6, 2, KEY_ENTER),
+	KEY(6, 3, KEY_N),
+	KEY(6, 4, KEY_J),
+	KEY(6, 5, KEY_U),
+
+	KEY(7, 0, KEY_F6),		/* Green/Call */
+	KEY(7, 1, KEY_O),
+	KEY(7, 2, KEY_BACKSPACE),
+	KEY(7, 3, KEY_M),
+	KEY(7, 4, KEY_K),
+	KEY(7, 5, KEY_I),
+};
+
+static struct pxa27x_keypad_platform_data treo680_keypad_platform_data = {
+	.matrix_key_rows	= 8,
+	.matrix_key_cols	= 7,
+	.matrix_key_map		= treo680_matrix_keys,
+	.matrix_key_map_size	= ARRAY_SIZE(treo680_matrix_keys),
+	.direct_key_map		= { KEY_CONNECT },
+	.direct_key_num		= 1,
+
+	.debounce_interval	= 30,
+};
+
+/******************************************************************************
+ * aSoC audio
+ ******************************************************************************/
+
+static pxa2xx_audio_ops_t treo680_ac97_pdata = {
+	.reset_gpio	= 95,
+};
+
+/******************************************************************************
+ * Backlight
+ ******************************************************************************/
+static int treo680_backlight_init(struct device *dev)
+{
+	int ret;
+
+	ret = gpio_request(GPIO_NR_TREO680_BL_POWER, "BL POWER");
+	if (ret)
+		goto err;
+	ret = gpio_direction_output(GPIO_NR_TREO680_BL_POWER, 0);
+	if (ret)
+		goto err2;
+	ret = gpio_request(GPIO_NR_TREO680_LCD_POWER, "LCD POWER");
+	if (ret)
+		goto err2;
+	ret = gpio_direction_output(GPIO_NR_TREO680_LCD_POWER, 0);
+	if (ret)
+		goto err3;
+
+	return 0;
+err3:
+	gpio_free(GPIO_NR_TREO680_LCD_POWER);
+err2:
+	gpio_free(GPIO_NR_TREO680_BL_POWER);
+err:
+	return ret;
+}
+
+static int treo680_backlight_notify(int brightness)
+{
+	gpio_set_value(GPIO_NR_TREO680_BL_POWER, brightness);
+	return TREO680_MAX_INTENSITY - brightness;
+};
+
+static void treo680_backlight_exit(struct device *dev)
+{
+	gpio_free(GPIO_NR_TREO680_BL_POWER);
+	gpio_free(GPIO_NR_TREO680_LCD_POWER);
+}
+
+static struct platform_pwm_backlight_data treo680_backlight_data = {
+	.pwm_id		= 0,
+	.max_brightness	= TREO680_MAX_INTENSITY,
+	.dft_brightness	= TREO680_DEFAULT_INTENSITY,
+	.pwm_period_ns	= TREO680_PERIOD_NS,
+	.init		= treo680_backlight_init,
+	.notify		= treo680_backlight_notify,
+	.exit		= treo680_backlight_exit,
+};
+
+static struct platform_device treo680_backlight = {
+	.name	= "pwm-backlight",
+	.dev	= {
+		.parent		= &pxa27x_device_pwm0.dev,
+		.platform_data	= &treo680_backlight_data,
+	},
+};
+
+/******************************************************************************
+ * IrDA
+ ******************************************************************************/
+static void treo680_transceiver_mode(struct device *dev, int mode)
+{
+	gpio_set_value(GPIO_NR_TREO680_IR_EN, mode & IR_OFF);
+	pxa2xx_transceiver_mode(dev, mode);
+}
+
+static int treo680_irda_startup(struct device *dev)
+{
+	int err;
+
+	err = gpio_request(GPIO_NR_TREO680_IR_EN, "Ir port disable");
+	if (err)
+		goto err1;
+
+	err = gpio_direction_output(GPIO_NR_TREO680_IR_EN, 1);
+	if (err)
+		goto err2;
+
+	return 0;
+
+err2:
+	dev_err(dev, "treo680_irda: cannot change IR gpio direction\n");
+	gpio_free(GPIO_NR_TREO680_IR_EN);
+err1:
+	dev_err(dev, "treo680_irda: cannot allocate IR gpio\n");
+	return err;
+}
+
+static void treo680_irda_shutdown(struct device *dev)
+{
+	gpio_free(GPIO_NR_TREO680_AMP_EN);
+}
+
+static struct pxaficp_platform_data treo680_ficp_info = {
+	.transceiver_cap  = IR_FIRMODE | IR_SIRMODE | IR_OFF,
+	.startup          = treo680_irda_startup,
+	.shutdown         = treo680_irda_shutdown,
+	.transceiver_mode = treo680_transceiver_mode,
+};
+
+/******************************************************************************
+ * UDC
+ ******************************************************************************/
+static struct pxa2xx_udc_mach_info treo680_udc_info __initdata = {
+	.gpio_vbus		= GPIO_NR_TREO680_USB_DETECT,
+	.gpio_vbus_inverted	= 1,
+	.gpio_pullup		= GPIO_NR_TREO680_USB_PULLUP,
+};
+
+
+/******************************************************************************
+ * USB host
+ ******************************************************************************/
+static struct pxaohci_platform_data treo680_ohci_info = {
+	.port_mode    = PMM_PERPORT_MODE,
+	.flags        = ENABLE_PORT1 | ENABLE_PORT3,
+	.power_budget = 0,
+};
+
+/******************************************************************************
+ * Power supply
+ ******************************************************************************/
+static int power_supply_init(struct device *dev)
+{
+	int ret;
+
+	ret = gpio_request(GPIO_NR_TREO680_POWER_DETECT, "CABLE_STATE_AC");
+	if (ret)
+		goto err1;
+	ret = gpio_direction_input(GPIO_NR_TREO680_POWER_DETECT);
+	if (ret)
+		goto err2;
+
+	return 0;
+
+err2:
+	gpio_free(GPIO_NR_TREO680_POWER_DETECT);
+err1:
+	return ret;
+}
+
+static int treo680_is_ac_online(void)
+{
+	return gpio_get_value(GPIO_NR_TREO680_POWER_DETECT);
+}
+
+static void power_supply_exit(struct device *dev)
+{
+	gpio_free(GPIO_NR_TREO680_POWER_DETECT);
+}
+
+static char *treo680_supplicants[] = {
+	"main-battery",
+};
+
+static struct pda_power_pdata power_supply_info = {
+	.init		 = power_supply_init,
+	.is_ac_online    = treo680_is_ac_online,
+	.exit		 = power_supply_exit,
+	.supplied_to     = treo680_supplicants,
+	.num_supplicants = ARRAY_SIZE(treo680_supplicants),
+};
+
+static struct platform_device power_supply = {
+	.name = "pda-power",
+	.id   = -1,
+	.dev  = {
+		.platform_data = &power_supply_info,
+	},
+};
+
+/******************************************************************************
+ * Vibra and LEDs
+ ******************************************************************************/
+static struct gpio_led gpio_leds[] = {
+	{
+		.name			= "treo680:vibra:vibra",
+		.default_trigger	= "none",
+		.gpio			= GPIO_NR_TREO680_VIBRATE_EN,
+	},
+	{
+		.name			= "treo680:green:led",
+		.default_trigger	= "mmc0",
+		.gpio			= GPIO_NR_TREO680_GREEN_LED,
+	},
+	{
+		.name			= "treo680:keybbl:keybbl",
+		.default_trigger	= "none",
+		.gpio			= GPIO_NR_TREO680_KEYB_BL,
+	},
+};
+
+static struct gpio_led_platform_data gpio_led_info = {
+	.leds		= gpio_leds,
+	.num_leds	= ARRAY_SIZE(gpio_leds),
+};
+
+static struct platform_device treo680_leds = {
+	.name   = "leds-gpio",
+	.id     = -1,
+	.dev    = {
+		.platform_data  = &gpio_led_info,
+	}
+};
+
+
+/******************************************************************************
+ * Framebuffer
+ ******************************************************************************/
+/* TODO: add support for 324x324 */
+static struct pxafb_mode_info treo680_lcd_modes[] = {
+{
+	.pixclock		= 86538,
+	.xres			= 320,
+	.yres			= 320,
+	.bpp			= 16,
+
+	.left_margin		= 20,
+	.right_margin		= 8,
+	.upper_margin		= 8,
+	.lower_margin		= 5,
+
+	.hsync_len		= 4,
+	.vsync_len		= 1,
+},
+};
+
+static struct pxafb_mach_info treo680_lcd_screen = {
+	.modes		= treo680_lcd_modes,
+	.num_modes	= ARRAY_SIZE(treo680_lcd_modes),
+	.lcd_conn	= LCD_COLOR_TFT_16BPP | LCD_PCLK_EDGE_FALL,
+};
+
+/******************************************************************************
+ * Power management - standby
+ ******************************************************************************/
+static void __init treo680_pm_init(void)
+{
+	static u32 resume[] = {
+		0xe3a00101,	/* mov	r0,	#0x40000000 */
+		0xe380060f,	/* orr	r0, r0, #0x00f00000 */
+		0xe590f008,	/* ldr	pc, [r0, #0x08] */
+	};
+
+	/* this is where the bootloader jumps */
+	memcpy(phys_to_virt(TREO680_STR_BASE), resume, sizeof(resume));
+}
+
+/******************************************************************************
+ * Machine init
+ ******************************************************************************/
+static struct platform_device *devices[] __initdata = {
+	&treo680_backlight,
+	&treo680_leds,
+	&power_supply,
+};
+
+/* setup udc GPIOs initial state */
+static void __init treo680_udc_init(void)
+{
+	if (!gpio_request(GPIO_NR_TREO680_USB_PULLUP, "UDC Vbus")) {
+		gpio_direction_output(GPIO_NR_TREO680_USB_PULLUP, 1);
+		gpio_free(GPIO_NR_TREO680_USB_PULLUP);
+	}
+}
+
+static void __init treo680_init(void)
+{
+	treo680_pm_init();
+	pxa2xx_mfp_config(ARRAY_AND_SIZE(treo680_pin_config));
+	pxa_set_keypad_info(&treo680_keypad_platform_data);
+	set_pxa_fb_info(&treo680_lcd_screen);
+	pxa_set_mci_info(&treo680_mci_platform_data);
+	treo680_udc_init();
+	pxa_set_udc_info(&treo680_udc_info);
+	pxa_set_ac97_info(&treo680_ac97_pdata);
+	pxa_set_ficp_info(&treo680_ficp_info);
+	pxa_set_ohci_info(&treo680_ohci_info);
+
+	platform_add_devices(devices, ARRAY_SIZE(devices));
+}
+
+MACHINE_START(TREO680, "Palm Treo 680")
+	.phys_io	= TREO680_PHYS_IO_START,
+	.io_pg_offst	= io_p2v(0x40000000),
+	.boot_params	= 0xa0000100,
+	.map_io		= pxa_map_io,
+	.init_irq	= pxa27x_init_irq,
+	.timer		= &pxa_timer,
+	.init_machine	= treo680_init,
+MACHINE_END
diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c
index e6344ece00ce..0b91bb20cf95 100644
--- a/arch/arm/mm/mmu.c
+++ b/arch/arm/mm/mmu.c
@@ -835,6 +835,13 @@ void __init reserve_node_zero(pg_data_t *pgdat)
 				BOOTMEM_EXCLUSIVE);
 	}
 
+	if (machine_is_treo680()) {
+		reserve_bootmem_node(pgdat, 0xa0000000, 0x1000,
+				BOOTMEM_EXCLUSIVE);
+		reserve_bootmem_node(pgdat, 0xa2000000, 0x1000,
+				BOOTMEM_EXCLUSIVE);
+	}
+
 	if (machine_is_palmt5())
 		reserve_bootmem_node(pgdat, 0xa0200000, 0x1000,
 				BOOTMEM_EXCLUSIVE);
-- 
cgit v1.2.3-59-g8ed1b


From 6ea0414fc748ab5b1d83a414c7ee3a60190363aa Mon Sep 17 00:00:00 2001
From: Philipp Zabel <philipp.zabel@gmail.com>
Date: Thu, 28 May 2009 07:15:18 +0200
Subject: [ARM] pxa/hx4700: add Maxim 1587A voltage regulator

On this board, the PXA270 CPU voltage VCC_CORE is provided
by a Maxim 1587A voltage regulator configured to provide
1.55 V maximum voltage for 624 MHz operation.

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Signed-off-by: Eric Miao <eric.miao@marvell.com>
---
 arch/arm/mach-pxa/hx4700.c | 41 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 41 insertions(+)

diff --git a/arch/arm/mach-pxa/hx4700.c b/arch/arm/mach-pxa/hx4700.c
index 7fff467e84fc..81359d574f88 100644
--- a/arch/arm/mach-pxa/hx4700.c
+++ b/arch/arm/mach-pxa/hx4700.c
@@ -30,6 +30,7 @@
 #include <linux/pwm_backlight.h>
 #include <linux/regulator/bq24022.h>
 #include <linux/regulator/machine.h>
+#include <linux/regulator/max1586.h>
 #include <linux/spi/ads7846.h>
 #include <linux/spi/spi.h>
 #include <linux/usb/gpio_vbus.h>
@@ -774,6 +775,45 @@ static struct platform_device strataflash = {
 	},
 };
 
+/*
+ * Maxim MAX1587A on PI2C
+ */
+
+static struct regulator_consumer_supply max1587a_consumer = {
+	.supply = "vcc_core",
+};
+
+static struct regulator_init_data max1587a_v3_info = {
+	.constraints = {
+		.name = "vcc_core range",
+		.min_uV =  900000,
+		.max_uV = 1705000,
+		.always_on = 1,
+		.valid_ops_mask = REGULATOR_CHANGE_VOLTAGE,
+	},
+	.num_consumer_supplies = 1,
+	.consumer_supplies     = &max1587a_consumer,
+};
+
+static struct max1586_subdev_data max1587a_subdev = {
+	.name = "vcc_core",
+	.id   = MAX1586_V3,
+	.platform_data = &max1587a_v3_info,
+};
+
+static struct max1586_platform_data max1587a_info = {
+	.num_subdevs = 1,
+	.subdevs     = &max1587a_subdev,
+	.v3_gain     = MAX1586_GAIN_R24_3k32, /* 730..1550 mV */
+};
+
+static struct i2c_board_info __initdata pi2c_board_info[] = {
+	{
+		I2C_BOARD_INFO("max1586", 0x14),
+		.platform_data = &max1587a_info,
+	},
+};
+
 /*
  * PCMCIA
  */
@@ -828,6 +868,7 @@ static void __init hx4700_init(void)
 	pxa_set_ficp_info(&ficp_info);
 	pxa27x_set_i2c_power_info(NULL);
 	pxa_set_i2c_info(NULL);
+	i2c_register_board_info(1, ARRAY_AND_SIZE(pi2c_board_info));
 	pxa2xx_set_spi_info(2, &pxa_ssp2_master_info);
 	spi_register_board_info(ARRAY_AND_SIZE(tsc2046_board_info));
 
-- 
cgit v1.2.3-59-g8ed1b


From 9752e14898102063a67b44e6db5b758092e769a4 Mon Sep 17 00:00:00 2001
From: Robert Jarzmik <robert.jarzmik@free.fr>
Date: Thu, 30 Apr 2009 11:53:30 +0800
Subject: [ARM] pxa/mioa701: add Maxim 1586 voltage regulator

On this board, the PXA272 CPU voltage VCC_CORE is provided
by a Maxim 1586 voltage regulator. Use the regulator
framework to provide VCC_CORE control. When cpufreq will be
updated to ask for vcc_core, this will optimize power
drained by the board.

Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Eric Miao <eric.miao@marvell.com>
---
 arch/arm/mach-pxa/mioa701.c | 41 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 41 insertions(+)

diff --git a/arch/arm/mach-pxa/mioa701.c b/arch/arm/mach-pxa/mioa701.c
index 4dc8c2ec40a9..99e5a880a4ad 100644
--- a/arch/arm/mach-pxa/mioa701.c
+++ b/arch/arm/mach-pxa/mioa701.c
@@ -37,6 +37,7 @@
 #include <linux/wm97xx_batt.h>
 #include <linux/mtd/physmap.h>
 #include <linux/usb/gpio_vbus.h>
+#include <linux/regulator/max1586.h>
 
 #include <asm/mach-types.h>
 #include <asm/mach/arch.h>
@@ -716,6 +717,37 @@ static struct wm97xx_batt_info mioa701_battery_data = {
 	.batt_name	= "mioa701_battery",
 };
 
+/*
+ * Voltage regulation
+ */
+static struct regulator_consumer_supply max1586_consumers[] = {
+	{
+		.supply = "vcc_core",
+	}
+};
+
+static struct regulator_init_data max1586_v3_info = {
+	.constraints = {
+		.name = "vcc_core range",
+		.min_uV = 1000000,
+		.max_uV = 1705000,
+		.always_on = 1,
+		.valid_ops_mask = REGULATOR_CHANGE_VOLTAGE,
+	},
+	.num_consumer_supplies = ARRAY_SIZE(max1586_consumers),
+	.consumer_supplies = max1586_consumers,
+};
+
+static struct max1586_subdev_data max1586_subdevs[] = {
+	{ .name = "vcc_core", .id = MAX1586_V3,
+	  .platform_data = &max1586_v3_info },
+};
+
+static struct max1586_platform_data max1586_info = {
+	.subdevs = max1586_subdevs,
+	.num_subdevs = ARRAY_SIZE(max1586_subdevs),
+};
+
 /*
  * Camera interface
  */
@@ -725,6 +757,13 @@ struct pxacamera_platform_data mioa701_pxacamera_platform_data = {
 	.mclk_10khz = 5000,
 };
 
+static struct i2c_board_info __initdata mioa701_pi2c_devices[] = {
+	{
+		I2C_BOARD_INFO("max1586", 0x14),
+		.platform_data = &max1586_info,
+	},
+};
+
 static struct soc_camera_link iclink = {
 	.bus_id	= 0, /* Must match id in pxa27x_device_camera in device.c */
 };
@@ -825,7 +864,9 @@ static void __init mioa701_machine_init(void)
 	platform_add_devices(devices, ARRAY_SIZE(devices));
 	gsm_init();
 
+	i2c_register_board_info(1, ARRAY_AND_SIZE(mioa701_pi2c_devices));
 	pxa_set_i2c_info(&i2c_pdata);
+	pxa27x_set_i2c_power_info(NULL);
 	pxa_set_camera_info(&mioa701_pxacamera_platform_data);
 	i2c_register_board_info(0, ARRAY_AND_SIZE(mioa701_i2c_devices));
 }
-- 
cgit v1.2.3-59-g8ed1b


From e567b74caff03c0ed66b3d3e11c0acd17fbcfb9a Mon Sep 17 00:00:00 2001
From: Philipp Zabel <philipp.zabel@gmail.com>
Date: Thu, 28 May 2009 07:15:17 +0200
Subject: [ARM] pxa/mioa701: add V3 gain configuration for Maxim 1586 voltage
 regulator

Maxim 1586 V3 gain can be configured by external resistors
(R24, R25) connected to the FB3 pin. To avoid accidental
overvoltage, it now requires this gain to be configured
explicitly.

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Acked-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Eric Miao <eric.miao@marvell.com>
---
 arch/arm/mach-pxa/mioa701.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm/mach-pxa/mioa701.c b/arch/arm/mach-pxa/mioa701.c
index 99e5a880a4ad..2d28132c725b 100644
--- a/arch/arm/mach-pxa/mioa701.c
+++ b/arch/arm/mach-pxa/mioa701.c
@@ -746,6 +746,7 @@ static struct max1586_subdev_data max1586_subdevs[] = {
 static struct max1586_platform_data max1586_info = {
 	.subdevs = max1586_subdevs,
 	.num_subdevs = ARRAY_SIZE(max1586_subdevs),
+	.v3_gain = MAX1586_GAIN_NO_R24, /* 700..1475 mV */
 };
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 8776b268c6b0b671c2c744b06fee021b6da3e45f Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Sun, 26 Apr 2009 12:53:58 +0100
Subject: [ARM] pxa: register wm8731 explicitly for corgi and poodle

The wm8731 driver has been converted to register using the standard I2C
device registration mechanism so we can now do the registration in the
arch/arm code as normal.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
---
 arch/arm/mach-pxa/corgi.c  |  6 ++++++
 arch/arm/mach-pxa/poodle.c |  6 ++++++
 sound/soc/pxa/corgi.c      | 36 ------------------------------------
 sound/soc/pxa/poodle.c     | 36 ------------------------------------
 4 files changed, 12 insertions(+), 72 deletions(-)

diff --git a/arch/arm/mach-pxa/corgi.c b/arch/arm/mach-pxa/corgi.c
index 962dda2e154a..5363e1aea3fb 100644
--- a/arch/arm/mach-pxa/corgi.c
+++ b/arch/arm/mach-pxa/corgi.c
@@ -23,6 +23,7 @@
 #include <linux/pm.h>
 #include <linux/gpio.h>
 #include <linux/backlight.h>
+#include <linux/i2c.h>
 #include <linux/io.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/ads7846.h>
@@ -600,6 +601,10 @@ static struct platform_device *devices[] __initdata = {
 	&sharpsl_rom_device,
 };
 
+static struct i2c_board_info __initdata corgi_i2c_devices[] = {
+	{ I2C_BOARD_INFO("wm8731", 0x1b) },
+};
+
 static void corgi_poweroff(void)
 {
 	if (!machine_is_corgi())
@@ -634,6 +639,7 @@ static void __init corgi_init(void)
 	pxa_set_mci_info(&corgi_mci_platform_data);
 	pxa_set_ficp_info(&corgi_ficp_platform_data);
 	pxa_set_i2c_info(NULL);
+	i2c_register_board_info(0, ARRAY_AND_SIZE(corgi_i2c_devices));
 
 	platform_scoop_config = &corgi_pcmcia_config;
 
diff --git a/arch/arm/mach-pxa/poodle.c b/arch/arm/mach-pxa/poodle.c
index ac431ed10399..9352d4a34837 100644
--- a/arch/arm/mach-pxa/poodle.c
+++ b/arch/arm/mach-pxa/poodle.c
@@ -22,6 +22,7 @@
 #include <linux/delay.h>
 #include <linux/mtd/physmap.h>
 #include <linux/gpio.h>
+#include <linux/i2c.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/ads7846.h>
 #include <linux/mtd/sharpsl.h>
@@ -486,6 +487,10 @@ static struct platform_device *devices[] __initdata = {
 	&sharpsl_rom_device,
 };
 
+static struct i2c_board_info __initdata poodle_i2c_devices[] = {
+	{ I2C_BOARD_INFO("wm8731", 0x1b) },
+};
+
 static void poodle_poweroff(void)
 {
 	arm_machine_restart('h', NULL);
@@ -519,6 +524,7 @@ static void __init poodle_init(void)
 	pxa_set_mci_info(&poodle_mci_platform_data);
 	pxa_set_ficp_info(&poodle_ficp_platform_data);
 	pxa_set_i2c_info(NULL);
+	i2c_register_board_info(0, ARRAY_AND_SIZE(poodle_i2c_devices));
 	poodle_init_spi();
 }
 
diff --git a/sound/soc/pxa/corgi.c b/sound/soc/pxa/corgi.c
index d5be2b30cda5..fefe1a57f31a 100644
--- a/sound/soc/pxa/corgi.c
+++ b/sound/soc/pxa/corgi.c
@@ -320,38 +320,6 @@ static struct snd_soc_device corgi_snd_devdata = {
 	.codec_dev = &soc_codec_dev_wm8731,
 };
 
-/*
- * FIXME: This is a temporary bodge to avoid cross-tree merge issues.
- * New drivers should register the wm8731 I2C device in the machine
- * setup code (under arch/arm for ARM systems).
- */
-static int wm8731_i2c_register(void)
-{
-	struct i2c_board_info info;
-	struct i2c_adapter *adapter;
-	struct i2c_client *client;
-
-	memset(&info, 0, sizeof(struct i2c_board_info));
-	info.addr = 0x1b;
-	strlcpy(info.type, "wm8731", I2C_NAME_SIZE);
-
-	adapter = i2c_get_adapter(0);
-	if (!adapter) {
-		printk(KERN_ERR "can't get i2c adapter 0\n");
-		return -ENODEV;
-	}
-
-	client = i2c_new_device(adapter, &info);
-	i2c_put_adapter(adapter);
-	if (!client) {
-		printk(KERN_ERR "can't add i2c device at 0x%x\n",
-			(unsigned int)info.addr);
-		return -ENODEV;
-	}
-
-	return 0;
-}
-
 static struct platform_device *corgi_snd_device;
 
 static int __init corgi_init(void)
@@ -362,10 +330,6 @@ static int __init corgi_init(void)
 	      machine_is_husky()))
 		return -ENODEV;
 
-	ret = wm8731_i2c_register();
-	if (ret != 0)
-		return ret;
-
 	corgi_snd_device = platform_device_alloc("soc-audio", -1);
 	if (!corgi_snd_device)
 		return -ENOMEM;
diff --git a/sound/soc/pxa/poodle.c b/sound/soc/pxa/poodle.c
index a51058f66747..c5f36e0eab58 100644
--- a/sound/soc/pxa/poodle.c
+++ b/sound/soc/pxa/poodle.c
@@ -280,38 +280,6 @@ static struct snd_soc_card snd_soc_poodle = {
 	.num_links = 1,
 };
 
-/*
- * FIXME: This is a temporary bodge to avoid cross-tree merge issues.
- * New drivers should register the wm8731 I2C device in the machine
- * setup code (under arch/arm for ARM systems).
- */
-static int wm8731_i2c_register(void)
-{
-	struct i2c_board_info info;
-	struct i2c_adapter *adapter;
-	struct i2c_client *client;
-
-	memset(&info, 0, sizeof(struct i2c_board_info));
-	info.addr = 0x1b;
-	strlcpy(info.type, "wm8731", I2C_NAME_SIZE);
-
-	adapter = i2c_get_adapter(0);
-	if (!adapter) {
-		printk(KERN_ERR "can't get i2c adapter 0\n");
-		return -ENODEV;
-	}
-
-	client = i2c_new_device(adapter, &info);
-	i2c_put_adapter(adapter);
-	if (!client) {
-		printk(KERN_ERR "can't add i2c device at 0x%x\n",
-			(unsigned int)info.addr);
-		return -ENODEV;
-	}
-
-	return 0;
-}
-
 /* poodle audio subsystem */
 static struct snd_soc_device poodle_snd_devdata = {
 	.card = &snd_soc_poodle,
@@ -327,10 +295,6 @@ static int __init poodle_init(void)
 	if (!machine_is_poodle())
 		return -ENODEV;
 
-	ret = wm8731_i2c_register();
-	if (ret != 0)
-		return ret;
-
 	locomo_gpio_set_dir(&poodle_locomo_device.dev,
 		POODLE_LOCOMO_GPIO_AMP_ON, 0);
 	/* should we mute HP at startup - burning power ?*/
-- 
cgit v1.2.3-59-g8ed1b


From 90af5811eb1fd69968f7c99e0cae6a5825dcf229 Mon Sep 17 00:00:00 2001
From: Tomas 'Sleep_Walker' Cech <sleep_walker@suse.cz>
Date: Fri, 22 May 2009 16:05:58 +0200
Subject: MAINTAINERS: add entry for Palm Treo680
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Tomáš Čech <sleep_walker@suse.cz>
Signed-off-by: Eric Miao <eric.miao@marvell.com>
---
 MAINTAINERS | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 9146a1c2e67d..7f132ae527f0 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -784,6 +784,12 @@ M:	marek.vasut@gmail.com
 W:	http://hackndev.com
 S:	Maintained
 
+ARM/PALM TREO 680 SUPPORT
+P:	Tomas Cech
+M:	sleep_walker@suse.cz
+W:	http://hackndev.com
+S:	Maintained
+
 ARM/PALMZ72 SUPPORT
 P:	Sergey Lapin
 M:	slapin@ossfans.org
-- 
cgit v1.2.3-59-g8ed1b


From cafc22658e85f65df72bee31c31b1fb337e7e606 Mon Sep 17 00:00:00 2001
From: Stefan Schmidt <stefan@datenfreihafen.org>
Date: Sun, 14 Jun 2009 01:08:36 +0200
Subject: MAINTAINERS: Update entry with file and SCM for EZX

Add file entry for easier mainatiner detection and make SCM more visible.

Signed-off-by: Stefan Schmidt <stefan@datenfreihafen.org>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
---
 MAINTAINERS | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 7f132ae527f0..852859ecbbce 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -645,6 +645,8 @@ M:	laforge@openezx.org
 L:	openezx-devel@lists.openezx.org (subscribers-only)
 W:	http://www.openezx.org/
 S:	Maintained
+T:	topgit git://git.openezx.org/openezx.git
+F:	arch/arm/mach-pxa/ezx.c
 
 ARM/FARADAY FA526 PORT
 P:	Paulius Zaleckas
-- 
cgit v1.2.3-59-g8ed1b


From d78ff0a50aac6a1bfe445969dd963e6486e49f56 Mon Sep 17 00:00:00 2001
From: Robert Jarzmik <robert.jarzmik@free.fr>
Date: Tue, 16 Jun 2009 07:51:05 +0200
Subject: MAINTAINERS: add entry for Mitac Mio A701 board

Add maintainer entry for Mitac Mio A701 board.

Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Eric Miao <eric.y.miao@gmail.com>
---
 MAINTAINERS | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 852859ecbbce..ad716d00bdfb 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -768,6 +768,13 @@ P:	Philipp Zabel
 M:	philipp.zabel@gmail.com
 S:	Maintained
 
+ARM/MIOA701 MACHINE SUPPORT
+P:	Robert Jarzmik
+M:	robert.jarzmik@free.fr
+L:	linux-arm-kernel@lists.arm.linux.org.uk (subscribers-only)
+F:	arch/arm/mach-pxa/mioa701.c
+S:	Maintained
+
 ARM/NEC MOBILEPRO 900/c MACHINE SUPPORT
 P:	Michael Petchkovsky
 M:	mkpetch@internode.on.net
-- 
cgit v1.2.3-59-g8ed1b


From 6c18ba9f5e506b8115b89b1aa7bdc25178f40b0a Mon Sep 17 00:00:00 2001
From: Alexandros Batsakis <Alexandros.Batsakis@netapp.com>
Date: Tue, 16 Jun 2009 04:19:13 +0300
Subject: nfsd41: move channel attributes from nfsd4_session to a
 nfsd4_channel_attr struct

the change is valid for both the forechannel and the backchannel (currently dummy)

Signed-off-by: Alexandros Batsakis <Alexandros.Batsakis@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4state.c        | 28 +++++++++++++++-------------
 fs/nfsd/nfs4xdr.c          |  2 +-
 include/linux/nfsd/state.h | 18 +++++++++++++-----
 include/linux/nfsd/xdr4.h  | 11 -----------
 4 files changed, 29 insertions(+), 30 deletions(-)

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 89d9ac55c034..d5caf2a709d2 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -444,8 +444,8 @@ static int set_forechannel_maxreqs(struct nfsd4_channel_attrs *fchan)
  * fchan holds the client values on input, and the server values on output
  */
 static int init_forechannel_attrs(struct svc_rqst *rqstp,
-				    struct nfsd4_session *session,
-				    struct nfsd4_channel_attrs *fchan)
+				  struct nfsd4_channel_attrs *session_fchan,
+				  struct nfsd4_channel_attrs *fchan)
 {
 	int status = 0;
 	__u32   maxcount = svc_max_payload(rqstp);
@@ -455,21 +455,21 @@ static int init_forechannel_attrs(struct svc_rqst *rqstp,
 	/* Use the client's max request and max response size if possible */
 	if (fchan->maxreq_sz > maxcount)
 		fchan->maxreq_sz = maxcount;
-	session->se_fmaxreq_sz = fchan->maxreq_sz;
+	session_fchan->maxreq_sz = fchan->maxreq_sz;
 
 	if (fchan->maxresp_sz > maxcount)
 		fchan->maxresp_sz = maxcount;
-	session->se_fmaxresp_sz = fchan->maxresp_sz;
+	session_fchan->maxresp_sz = fchan->maxresp_sz;
 
 	/* Set the max response cached size our default which is
 	 * a multiple of PAGE_SIZE and small */
-	session->se_fmaxresp_cached = NFSD_PAGES_PER_SLOT * PAGE_SIZE;
-	fchan->maxresp_cached = session->se_fmaxresp_cached;
+	session_fchan->maxresp_cached = NFSD_PAGES_PER_SLOT * PAGE_SIZE;
+	fchan->maxresp_cached = session_fchan->maxresp_cached;
 
 	/* Use the client's maxops if possible */
 	if (fchan->maxops > NFSD_MAX_OPS_PER_COMPOUND)
 		fchan->maxops = NFSD_MAX_OPS_PER_COMPOUND;
-	session->se_fmaxops = fchan->maxops;
+	session_fchan->maxops = fchan->maxops;
 
 	/* try to use the client requested number of slots */
 	if (fchan->maxreqs > NFSD_MAX_SLOTS_PER_SESSION)
@@ -481,7 +481,7 @@ static int init_forechannel_attrs(struct svc_rqst *rqstp,
 	 */
 	status = set_forechannel_maxreqs(fchan);
 
-	session->se_fnumslots = fchan->maxreqs;
+	session_fchan->maxreqs = fchan->maxreqs;
 	return status;
 }
 
@@ -495,12 +495,14 @@ alloc_init_session(struct svc_rqst *rqstp, struct nfs4_client *clp,
 	memset(&tmp, 0, sizeof(tmp));
 
 	/* FIXME: For now, we just accept the client back channel attributes. */
-	status = init_forechannel_attrs(rqstp, &tmp, &cses->fore_channel);
+	tmp.se_bchannel = cses->back_channel;
+	status = init_forechannel_attrs(rqstp, &tmp.se_fchannel,
+					&cses->fore_channel);
 	if (status)
 		goto out;
 
 	/* allocate struct nfsd4_session and slot table in one piece */
-	slotsize = tmp.se_fnumslots * sizeof(struct nfsd4_slot);
+	slotsize = tmp.se_fchannel.maxreqs * sizeof(struct nfsd4_slot);
 	new = kzalloc(sizeof(*new) + slotsize, GFP_KERNEL);
 	if (!new)
 		goto out;
@@ -574,7 +576,7 @@ free_session(struct kref *kref)
 	int i;
 
 	ses = container_of(kref, struct nfsd4_session, se_ref);
-	for (i = 0; i < ses->se_fnumslots; i++) {
+	for (i = 0; i < ses->se_fchannel.maxreqs; i++) {
 		struct nfsd4_cache_entry *e = &ses->se_slots[i].sl_cache_entry;
 		nfsd4_release_respages(e->ce_respages, e->ce_resused);
 	}
@@ -1130,7 +1132,7 @@ nfsd4_replay_cache_entry(struct nfsd4_compoundres *resp,
 	 * is sent (lease renewal).
 	 */
 	if (seq && nfsd4_not_cached(resp)) {
-		seq->maxslots = resp->cstate.session->se_fnumslots;
+		seq->maxslots = resp->cstate.session->se_fchannel.maxreqs;
 		return nfs_ok;
 	}
 
@@ -1473,7 +1475,7 @@ nfsd4_sequence(struct svc_rqst *rqstp,
 		goto out;
 
 	status = nfserr_badslot;
-	if (seq->slotid >= session->se_fnumslots)
+	if (seq->slotid >= session->se_fchannel.maxreqs)
 		goto out;
 
 	slot = &session->se_slots[seq->slotid];
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index d07f704a2ac9..2dcc7feaa6ff 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -3183,7 +3183,7 @@ static int nfsd4_check_drc_limit(struct nfsd4_compoundres *resp)
 	dprintk("%s length %u, xb->page_len %u tlen %u pad %u\n", __func__,
 		length, xb->page_len, tlen, pad);
 
-	if (length <= session->se_fmaxresp_cached)
+	if (length <= session->se_fchannel.maxresp_cached)
 		return status;
 	else
 		return nfserr_rep_too_big_to_cache;
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index c0c49215ddc5..105cc100de05 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -115,6 +115,17 @@ struct nfsd4_slot {
 	struct nfsd4_cache_entry	sl_cache_entry;
 };
 
+struct nfsd4_channel_attrs {
+	u32		headerpadsz;
+	u32		maxreq_sz;
+	u32		maxresp_sz;
+	u32		maxresp_cached;
+	u32		maxops;
+	u32		maxreqs;
+	u32		nr_rdma_attrs;
+	u32		rdma_attrs;
+};
+
 struct nfsd4_session {
 	struct kref		se_ref;
 	struct list_head	se_hash;	/* hash by sessionid */
@@ -122,11 +133,8 @@ struct nfsd4_session {
 	u32			se_flags;
 	struct nfs4_client	*se_client;	/* for expire_client */
 	struct nfs4_sessionid	se_sessionid;
-	u32			se_fmaxreq_sz;
-	u32			se_fmaxresp_sz;
-	u32			se_fmaxresp_cached;
-	u32			se_fmaxops;
-	u32			se_fnumslots;
+	struct nfsd4_channel_attrs se_fchannel;
+	struct nfsd4_channel_attrs se_bchannel;
 	struct nfsd4_slot	se_slots[];	/* forward channel slots */
 };
 
diff --git a/include/linux/nfsd/xdr4.h b/include/linux/nfsd/xdr4.h
index d0f050f01eca..2bacf7535069 100644
--- a/include/linux/nfsd/xdr4.h
+++ b/include/linux/nfsd/xdr4.h
@@ -366,17 +366,6 @@ struct nfsd4_exchange_id {
 	int		spa_how;
 };
 
-struct nfsd4_channel_attrs {
-	u32		headerpadsz;
-	u32		maxreq_sz;
-	u32		maxresp_sz;
-	u32		maxresp_cached;
-	u32		maxops;
-	u32		maxreqs;
-	u32		nr_rdma_attrs;
-	u32		rdma_attrs;
-};
-
 struct nfsd4_create_session {
 	clientid_t		clientid;
 	struct nfs4_sessionid	sessionid;
-- 
cgit v1.2.3-59-g8ed1b


From 29ad14cddd6246d17ff22f496363dfd6b3de8964 Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Mon, 15 Jun 2009 00:38:50 +0200
Subject: firewire: core: fix DMA unmapping in iso buffer removal

dmap_unmap_page() shall use the same direction as dma_map_page().

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/core-iso.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/firewire/core-iso.c b/drivers/firewire/core-iso.c
index 448ddd7d887b..166f19c6d38d 100644
--- a/drivers/firewire/core-iso.c
+++ b/drivers/firewire/core-iso.c
@@ -71,7 +71,7 @@ int fw_iso_buffer_init(struct fw_iso_buffer *buffer, struct fw_card *card,
 	for (j = 0; j < i; j++) {
 		address = page_private(buffer->pages[j]);
 		dma_unmap_page(card->device, address,
-			       PAGE_SIZE, DMA_TO_DEVICE);
+			       PAGE_SIZE, direction);
 		__free_page(buffer->pages[j]);
 	}
 	kfree(buffer->pages);
@@ -108,7 +108,7 @@ void fw_iso_buffer_destroy(struct fw_iso_buffer *buffer,
 	for (i = 0; i < buffer->page_count; i++) {
 		address = page_private(buffer->pages[i]);
 		dma_unmap_page(card->device, address,
-			       PAGE_SIZE, DMA_TO_DEVICE);
+			       PAGE_SIZE, buffer->direction);
 		__free_page(buffer->pages[i]);
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From d645f4dad056a98089df904294f66b96d04e91b6 Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Tue, 16 Jun 2009 19:15:25 +0200
Subject: firewire: core: fix iso context shutdown on card removal

If isochronous contexts existed when firewire-ohci was unloaded, the
core iso shutdown functions crashed with NULL dereferences, and buffers
etc. weren't released.

How the fix works:  We first copy the card driver's iso shutdown hooks
into the dummy driver, then fw_destroy_nodes notifies upper layers of
devices going away, these should shut down (including their iso
contexts), wait_for_completion(&card->done) will be triggered after
upper layers gave up all fw_device references, after which the card
driver's shutdown proceeds.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/core-card.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/drivers/firewire/core-card.c b/drivers/firewire/core-card.c
index 667603ac14b1..543fccac81bb 100644
--- a/drivers/firewire/core-card.c
+++ b/drivers/firewire/core-card.c
@@ -461,11 +461,11 @@ EXPORT_SYMBOL(fw_card_add);
 
 
 /*
- * The next few functions implements a dummy driver that use once a
- * card driver shuts down an fw_card.  This allows the driver to
- * cleanly unload, as all IO to the card will be handled by the dummy
- * driver instead of calling into the (possibly) unloaded module.  The
- * dummy driver just fails all IO.
+ * The next few functions implement a dummy driver that is used once a card
+ * driver shuts down an fw_card.  This allows the driver to cleanly unload,
+ * as all IO to the card will be handled (and failed) by the dummy driver
+ * instead of calling into the module.  Only functions for iso context
+ * shutdown still need to be provided by the card driver.
  */
 
 static int dummy_enable(struct fw_card *card, u32 *config_rom, size_t length)
@@ -512,7 +512,7 @@ static int dummy_enable_phys_dma(struct fw_card *card,
 	return -ENODEV;
 }
 
-static struct fw_card_driver dummy_driver = {
+static const struct fw_card_driver dummy_driver_template = {
 	.enable          = dummy_enable,
 	.update_phy_reg  = dummy_update_phy_reg,
 	.set_config_rom  = dummy_set_config_rom,
@@ -531,6 +531,8 @@ void fw_card_release(struct kref *kref)
 
 void fw_core_remove_card(struct fw_card *card)
 {
+	struct fw_card_driver dummy_driver = dummy_driver_template;
+
 	card->driver->update_phy_reg(card, 4,
 				     PHY_LINK_ACTIVE | PHY_CONTENDER, 0);
 	fw_core_initiate_bus_reset(card, 1);
@@ -539,7 +541,9 @@ void fw_core_remove_card(struct fw_card *card)
 	list_del_init(&card->link);
 	mutex_unlock(&card_mutex);
 
-	/* Set up the dummy driver. */
+	/* Switch off most of the card driver interface. */
+	dummy_driver.free_iso_context	= card->driver->free_iso_context;
+	dummy_driver.stop_iso		= card->driver->stop_iso;
 	card->driver = &dummy_driver;
 
 	fw_destroy_nodes(card);
-- 
cgit v1.2.3-59-g8ed1b


From b01b4babbf204443b5a846a7494546501614cefc Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Tue, 16 Jun 2009 20:43:55 +0200
Subject: firewire: net: fix card driver reloading

Fix some problems from "firewire: net: allow for unordered unit
discovery":
  - fwnet_remove was missing a list_del, causing fwnet_probe to crash if
    called after fwnet_remove, e.g. if firewire-ohci was unloaded and
    reloaded.
  - fwnet_probe should set its new_netdev flag only if it actually
    allocated a net_device.
  - Use dev_set_drvdata and dev_get_drvdata instead of deprecated direct
    access to device.driver_data.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/net.c | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c
index 95e35a36b01f..47dcb45d720e 100644
--- a/drivers/firewire/net.c
+++ b/drivers/firewire/net.c
@@ -1390,7 +1390,8 @@ static int fwnet_add_peer(struct fwnet_device *dev,
 	if (!peer)
 		return -ENOMEM;
 
-	unit->device.driver_data = peer;
+	dev_set_drvdata(&unit->device, peer);
+
 	peer->dev = dev;
 	peer->guid = (u64)device->config_rom[3] << 32 | device->config_rom[4];
 	peer->fifo = FWNET_NO_FIFO_ADDR;
@@ -1417,27 +1418,26 @@ static int fwnet_probe(struct device *_dev)
 	struct fw_device *device = fw_parent_device(unit);
 	struct fw_card *card = device->card;
 	struct net_device *net;
+	bool allocated_netdev = false;
 	struct fwnet_device *dev;
 	unsigned max_mtu;
-	bool new_netdev;
 	int ret;
 
 	mutex_lock(&fwnet_device_mutex);
 
 	dev = fwnet_dev_find(card);
 	if (dev) {
-		new_netdev = false;
 		net = dev->netdev;
 		goto have_dev;
 	}
 
-	new_netdev = true;
 	net = alloc_netdev(sizeof(*dev), "firewire%d", fwnet_init_dev);
 	if (net == NULL) {
 		ret = -ENOMEM;
 		goto out;
 	}
 
+	allocated_netdev = true;
 	SET_NETDEV_DEV(net, card->device);
 	dev = netdev_priv(net);
 
@@ -1479,12 +1479,12 @@ static int fwnet_probe(struct device *_dev)
 		  net->name, (unsigned long long)card->guid);
  have_dev:
 	ret = fwnet_add_peer(dev, unit, device);
-	if (ret && new_netdev) {
+	if (ret && allocated_netdev) {
 		unregister_netdev(net);
 		list_del(&dev->dev_link);
 	}
  out:
-	if (ret && new_netdev)
+	if (ret && allocated_netdev)
 		free_netdev(net);
 
 	mutex_unlock(&fwnet_device_mutex);
@@ -1508,7 +1508,7 @@ static void fwnet_remove_peer(struct fwnet_peer *peer)
 
 static int fwnet_remove(struct device *_dev)
 {
-	struct fwnet_peer *peer = _dev->driver_data;
+	struct fwnet_peer *peer = dev_get_drvdata(_dev);
 	struct fwnet_device *dev = peer->dev;
 	struct net_device *net;
 	struct fwnet_packet_task *ptask, *pt_next;
@@ -1544,6 +1544,8 @@ static int fwnet_remove(struct device *_dev)
 			dev_kfree_skb_any(ptask->skb);
 			kmem_cache_free(fwnet_packet_task_cache, ptask);
 		}
+		list_del(&dev->dev_link);
+
 		free_netdev(net);
 	}
 
@@ -1559,7 +1561,7 @@ static int fwnet_remove(struct device *_dev)
 static void fwnet_update(struct fw_unit *unit)
 {
 	struct fw_device *device = fw_parent_device(unit);
-	struct fwnet_peer *peer = unit->device.driver_data;
+	struct fwnet_peer *peer = dev_get_drvdata(&unit->device);
 	int generation;
 
 	generation = device->generation;
-- 
cgit v1.2.3-59-g8ed1b


From 00635b8ee2b5650fd01f5602ecfa289db336b570 Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Tue, 16 Jun 2009 22:35:32 +0200
Subject: firewire: net: better FIFO address range check and rcodes

The AR req handler should not check the generation; higher level code
is the better place to handle bus generation changes.  The target node
ID just needs to be checked for not being the "all nodes" address; in
this case don't handle the request and don't respond.

Use Address_Error and Type_Error rcodes as appropriate.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/net.c | 30 ++++++++++++++----------------
 1 file changed, 14 insertions(+), 16 deletions(-)

diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c
index 47dcb45d720e..a42209a73aed 100644
--- a/drivers/firewire/net.c
+++ b/drivers/firewire/net.c
@@ -810,29 +810,27 @@ static void fwnet_receive_packet(struct fw_card *card, struct fw_request *r,
 		int speed, unsigned long long offset, void *payload,
 		size_t length, void *callback_data)
 {
-	struct fwnet_device *dev;
-	int status;
+	struct fwnet_device *dev = callback_data;
+	int rcode;
 
-	dev = callback_data;
-	if (tcode != TCODE_WRITE_BLOCK_REQUEST
-	    || destination != card->node_id	/* <- FIXME */
-	    || generation != card->generation	/* <- FIXME */
-	    || offset != dev->handler.offset) {
-		fw_send_response(card, r, RCODE_CONFLICT_ERROR);
+	if (destination == IEEE1394_ALL_NODES) {
+		kfree(r);
 
 		return;
 	}
 
-	status = fwnet_incoming_packet(dev, payload, length,
-				       source, generation, false);
-	if (status != 0) {
+	if (offset != dev->handler.offset)
+		rcode = RCODE_ADDRESS_ERROR;
+	else if (tcode != TCODE_WRITE_BLOCK_REQUEST)
+		rcode = RCODE_TYPE_ERROR;
+	else if (fwnet_incoming_packet(dev, payload, length,
+				       source, generation, false) != 0) {
 		fw_error("Incoming packet failure\n");
-		fw_send_response(card, r, RCODE_CONFLICT_ERROR);
-
-		return;
-	}
+		rcode = RCODE_CONFLICT_ERROR;
+	} else
+		rcode = RCODE_COMPLETE;
 
-	fw_send_response(card, r, RCODE_COMPLETE);
+	fw_send_response(card, r, rcode);
 }
 
 static void fwnet_receive_broadcast(struct fw_iso_context *context,
-- 
cgit v1.2.3-59-g8ed1b


From 26c56dc0c4fbdf3af12ebc48278f09ef65ddb4dd Mon Sep 17 00:00:00 2001
From: Michal Miroslaw <mirq-linux@rere.qmqm.pl>
Date: Tue, 12 May 2009 13:49:26 -0700
Subject: PCI quirk: unhide 'Overflow' device on i828{6,7}5P/PE chipsets

Some BIOSes hide 'overflow' device (dev #6) for i82875P/PE chipsets.
The same happens for i82865P/PE. Add a quirk to enable this device.
This allows i82875 EDAC driver to bind to chipset's dev #6 and not
dev #0 as the latter is used by AGP driver.

On my laptop (i82865P based) ACPI code is disabling this device
again in \_SB.PCI0._CRS method (called at least at PNP init time).
This can be easily worked around by patching DSDT.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Michal Miroslaw <mirq-linux@rere.qmqm.pl>
Acked-by: Doug Thompson <norsk5@yahoo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/quirks.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index a778c84d04a9..15a8ab7ea911 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -2017,6 +2017,28 @@ DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_BROADCOM,
 			PCI_DEVICE_ID_NX2_5709S,
 			quirk_brcm_570x_limit_vpd);
 
+/* Originally in EDAC sources for i82875P:
+ * Intel tells BIOS developers to hide device 6 which
+ * configures the overflow device access containing
+ * the DRBs - this is where we expose device 6.
+ * http://www.x86-secret.com/articles/tweak/pat/patsecrets-2.htm
+ */
+static void __devinit quirk_unhide_mch_dev6(struct pci_dev *dev)
+{
+	u8 reg;
+
+	if (pci_read_config_byte(dev, 0xF4, &reg) == 0 && !(reg & 0x02)) {
+		dev_info(&dev->dev, "Enabling MCH 'Overflow' Device\n");
+		pci_write_config_byte(dev, 0xF4, reg | 0x02);
+	}
+}
+
+DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82865_HB,
+			quirk_unhide_mch_dev6);
+DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82875_HB,
+			quirk_unhide_mch_dev6);
+
+
 #ifdef CONFIG_PCI_MSI
 /* Some chipsets do not support MSI. We cannot easily rely on setting
  * PCI_BUS_FLAGS_NO_MSI in its bus flags because there are actually
-- 
cgit v1.2.3-59-g8ed1b


From 74c57428971d28863ec1804b1dd453930bbe1306 Mon Sep 17 00:00:00 2001
From: Michal Miroslaw <mirq-linux@rere.qmqm.pl>
Date: Tue, 12 May 2009 13:49:25 -0700
Subject: PCI quirk: HP hides SMBus controller in Compaq nx9500 laptops

I found no references to SMBus in ACPI DSDT disassembly on my laptop
so this should be safe.

Signed-off-by: Michal Miroslaw <mirq-linux@rere.qmqm.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/quirks.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 15a8ab7ea911..47d62084d5b7 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -1164,6 +1164,7 @@ static void __init asus_hides_smbus_hostbridge(struct pci_dev *dev)
 			switch (dev->subsystem_device) {
 			case 0x12bc: /* HP D330L */
 			case 0x12bd: /* HP D530 */
+			case 0x006a: /* HP Compaq nx9500 */
 				asus_hides_smbus = 1;
 			}
 		else if (dev->device == PCI_DEVICE_ID_INTEL_82875_HB)
-- 
cgit v1.2.3-59-g8ed1b


From 84845c070ce3ac4d3bd2c148fa20ba8ce5409167 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Tue, 26 May 2009 16:05:06 +0900
Subject: PCI: use pci_is_root_bus() in acpi_pci_get_bridge_handle()

Use pci_is_root_bus() in acpi_pci_get_bridge_handle() to check if the
pci bus is root, for code consistency.

Reviewed-by: Grant Grundler <grundler@parisc-linux.org>
Reviewed-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 include/linux/pci-acpi.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/pci-acpi.h b/include/linux/pci-acpi.h
index 092e82e0048c..df67c78dfe24 100644
--- a/include/linux/pci-acpi.h
+++ b/include/linux/pci-acpi.h
@@ -23,7 +23,7 @@ static inline acpi_handle acpi_find_root_bridge_handle(struct pci_dev *pdev)
 
 static inline acpi_handle acpi_pci_get_bridge_handle(struct pci_bus *pbus)
 {
-	if (pbus->parent)
+	if (!pci_is_root_bus(pbus))
 		return DEVICE_ACPI_HANDLE(&(pbus->self->dev));
 	return acpi_get_pci_rootbridge_handle(pci_domain_nr(pbus),
 					      pbus->number);
-- 
cgit v1.2.3-59-g8ed1b


From a222b8f83b995e9c6fe2aff2a8125facb49f658e Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Tue, 26 May 2009 16:05:33 +0900
Subject: PCI: use pci_is_root_bus() in acpi_find_root_bridge_handle()

Use pci_is_root_bus() in acpi_find_root_bridge_handle() to check if
the pci bus is root, for code consistency.

Reviewed-by: Alex Chiang <achiang@hp.com>
Reviewed-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 include/linux/pci-acpi.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/pci-acpi.h b/include/linux/pci-acpi.h
index df67c78dfe24..93a7c08f869d 100644
--- a/include/linux/pci-acpi.h
+++ b/include/linux/pci-acpi.h
@@ -15,7 +15,7 @@ static inline acpi_handle acpi_find_root_bridge_handle(struct pci_dev *pdev)
 {
 	struct pci_bus *pbus = pdev->bus;
 	/* Find a PCI root bus */
-	while (pbus->parent)
+	while (!pci_is_root_bus(pbus))
 		pbus = pbus->parent;
 	return acpi_get_pci_rootbridge_handle(pci_domain_nr(pbus),
 					      pbus->number);
-- 
cgit v1.2.3-59-g8ed1b


From 6e3f36df0ffa433e273c89f1447c94382a9db49e Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Tue, 26 May 2009 16:06:10 +0900
Subject: PCI: use pci_is_root_bus() in pci_find_upstream_pcie_bridge()

Use pci_is_root_bus() in pci_find_upstream_pcie_bridge() to check if
the pci bus is root, for code consistency.

Reviewed-by: Alex Chiang <achiang@hp.com>
Reviewed-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/search.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/search.c b/drivers/pci/search.c
index 650bc0a538dc..e8cb5051c311 100644
--- a/drivers/pci/search.c
+++ b/drivers/pci/search.c
@@ -29,7 +29,7 @@ pci_find_upstream_pcie_bridge(struct pci_dev *pdev)
 	if (pdev->is_pcie)
 		return NULL;
 	while (1) {
-		if (!pdev->bus->parent)
+		if (pci_is_root_bus(pdev->bus))
 			break;
 		pdev = pdev->bus->self;
 		/* a p2p bridge */
-- 
cgit v1.2.3-59-g8ed1b


From 9fc39256508c18d2861de11622183dfb6e79de87 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Tue, 26 May 2009 16:06:48 +0900
Subject: PCI: use pci_is_root_bus() in pci_read_bridge_bases()

Use pci_is_root_bus() in pci_read_bridge_bases() to check if the pci
bus is root, for code consistency.

Reviewed-by: Alex Chiang <achiang@hp.com>
Reviewed-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/probe.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index b962326e3d95..40e75f6a5056 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -289,7 +289,7 @@ void __devinit pci_read_bridge_bases(struct pci_bus *child)
 	struct resource *res;
 	int i;
 
-	if (!child->parent)	/* It's a host bus, nothing to read */
+	if (pci_is_root_bus(child))	/* It's a host bus, nothing to read */
 		return;
 
 	if (dev->transparent) {
-- 
cgit v1.2.3-59-g8ed1b


From 8784fd4d497171882319e4b513f5a5949fc8ab43 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Tue, 26 May 2009 16:07:33 +0900
Subject: PCI: use pci_is_root_bus() in pci_get_interrupt_pin()

Use pci_is_root_bus() in pci_get_interrupt_pin() for checking if the
pci bus is root, for code consistency.

Reviewed-by: Alex Chiang <achiang@hp.com>
Reviewed-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 56fb18d2cb52..70b3b44a8b6c 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -1529,7 +1529,7 @@ pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge)
 	if (!pin)
 		return -1;
 
-	while (dev->bus->parent) {
+	while (!pci_is_root_bus(dev->bus)) {
 		pin = pci_swizzle_interrupt_pin(dev, pin);
 		dev = dev->bus->self;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 1eb3948716f68bdb71509d0175765295f1aca23d Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Tue, 26 May 2009 16:08:36 +0900
Subject: PCI: use pci_is_root_bus() in pci_common_swizzle()

Use pci_is_root_bus() in pci_common_swizzle() for checking if the pci
bus is root, for code consistency.

Reviewed-by: Alex Chiang <achiang@hp.com>
Reviewed-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 70b3b44a8b6c..8ea911e55722 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -1549,7 +1549,7 @@ u8 pci_common_swizzle(struct pci_dev *dev, u8 *pinp)
 {
 	u8 pin = *pinp;
 
-	while (dev->bus->parent) {
+	while (!pci_is_root_bus(dev->bus)) {
 		pin = pci_swizzle_interrupt_pin(dev, pin);
 		dev = dev->bus->self;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From a72b46c3849cdb05993015991bde548ab8b6d7ac Mon Sep 17 00:00:00 2001
From: Huang Ying <ying.huang@intel.com>
Date: Fri, 24 Apr 2009 10:45:17 +0800
Subject: PCI: Add pci_bus_set_ops

pci_bus_set_ops changes pci_ops associated with a pci_bus. This can be
used by debug tools such as PCIE AER error injection to fake some PCI
configuration registers.

Acked-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/access.c | 19 +++++++++++++++++++
 include/linux/pci.h  |  1 +
 2 files changed, 20 insertions(+)

diff --git a/drivers/pci/access.c b/drivers/pci/access.c
index 0f3706512686..db23200c4874 100644
--- a/drivers/pci/access.c
+++ b/drivers/pci/access.c
@@ -66,6 +66,25 @@ EXPORT_SYMBOL(pci_bus_write_config_byte);
 EXPORT_SYMBOL(pci_bus_write_config_word);
 EXPORT_SYMBOL(pci_bus_write_config_dword);
 
+/**
+ * pci_bus_set_ops - Set raw operations of pci bus
+ * @bus:	pci bus struct
+ * @ops:	new raw operations
+ *
+ * Return previous raw operations
+ */
+struct pci_ops *pci_bus_set_ops(struct pci_bus *bus, struct pci_ops *ops)
+{
+	struct pci_ops *old_ops;
+	unsigned long flags;
+
+	spin_lock_irqsave(&pci_lock, flags);
+	old_ops = bus->ops;
+	bus->ops = ops;
+	spin_unlock_irqrestore(&pci_lock, flags);
+	return old_ops;
+}
+EXPORT_SYMBOL(pci_bus_set_ops);
 
 /**
  * pci_read_vpd - Read one entry from Vital Product Data
diff --git a/include/linux/pci.h b/include/linux/pci.h
index ec03b90d3510..ea2a153a9126 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -637,6 +637,7 @@ int pci_bus_write_config_word(struct pci_bus *bus, unsigned int devfn,
 			      int where, u16 val);
 int pci_bus_write_config_dword(struct pci_bus *bus, unsigned int devfn,
 			       int where, u32 val);
+struct pci_ops *pci_bus_set_ops(struct pci_bus *bus, struct pci_ops *ops);
 
 static inline int pci_read_config_byte(struct pci_dev *dev, int where, u8 *val)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 634deb028c9188b4144863ea87dde5457fb93e92 Mon Sep 17 00:00:00 2001
From: Huang Ying <ying.huang@intel.com>
Date: Fri, 24 Apr 2009 10:45:23 +0800
Subject: PCI: PCIE AER: export aer_irq

This is used by PCIE AER error injection to fake an PCI AER interrupt.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aer/aerdrv.c | 3 ++-
 drivers/pci/pcie/aer/aerdrv.h | 2 ++
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/pcie/aer/aerdrv.c b/drivers/pci/pcie/aer/aerdrv.c
index 32ade5af927e..4770f13b3ca1 100644
--- a/drivers/pci/pcie/aer/aerdrv.c
+++ b/drivers/pci/pcie/aer/aerdrv.c
@@ -77,7 +77,7 @@ void pci_no_aer(void)
  *
  * Invoked when Root Port detects AER messages.
  **/
-static irqreturn_t aer_irq(int irq, void *context)
+irqreturn_t aer_irq(int irq, void *context)
 {
 	unsigned int status, id;
 	struct pcie_device *pdev = (struct pcie_device *)context;
@@ -126,6 +126,7 @@ static irqreturn_t aer_irq(int irq, void *context)
 
 	return IRQ_HANDLED;
 }
+EXPORT_SYMBOL_GPL(aer_irq);
 
 /**
  * aer_alloc_rpc - allocate Root Port data structure
diff --git a/drivers/pci/pcie/aer/aerdrv.h b/drivers/pci/pcie/aer/aerdrv.h
index aa14482a4779..3a69ddefe361 100644
--- a/drivers/pci/pcie/aer/aerdrv.h
+++ b/drivers/pci/pcie/aer/aerdrv.h
@@ -11,6 +11,7 @@
 #include <linux/workqueue.h>
 #include <linux/pcieport_if.h>
 #include <linux/aer.h>
+#include <linux/interrupt.h>
 
 #define AER_NONFATAL			0
 #define AER_FATAL			1
@@ -120,6 +121,7 @@ extern void aer_delete_rootport(struct aer_rpc *rpc);
 extern int aer_init(struct pcie_device *dev);
 extern void aer_isr(struct work_struct *work);
 extern void aer_print_error(struct pci_dev *dev, struct aer_err_info *info);
+extern irqreturn_t aer_irq(int irq, void *context);
 
 #ifdef CONFIG_ACPI
 extern int aer_osc_setup(struct pcie_device *pciedev);
-- 
cgit v1.2.3-59-g8ed1b


From bfe5a7401785f864a4f30b4f066bfa045a28b9f6 Mon Sep 17 00:00:00 2001
From: Huang Ying <ying.huang@intel.com>
Date: Fri, 24 Apr 2009 10:45:31 +0800
Subject: PCI: PCIE AER: Document for PCIE AER software error injection

This patch adds a minimal HOWTO for PCIE AER software error injection
in Documentation/PCI/pcieaer-howto.txt.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 Documentation/PCI/pcieaer-howto.txt | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/Documentation/PCI/pcieaer-howto.txt b/Documentation/PCI/pcieaer-howto.txt
index ddeb14beacc8..f6b1ba7464dc 100644
--- a/Documentation/PCI/pcieaer-howto.txt
+++ b/Documentation/PCI/pcieaer-howto.txt
@@ -246,3 +246,24 @@ with the PCI Express AER Root driver?
 A: It could call the helper functions to enable AER in devices and
 cleanup uncorrectable status register. Pls. refer to section 3.3.
 
+
+4. Software error injection
+
+Debugging PCIE AER error recovery code is quite difficult because it
+is hard to trigger real hardware errors. Software based error
+injection can be used to fake various kinds of PCIE errors.
+
+First you should enable PCIE AER software error injection in kernel
+configuration, that is, following item should be in your .config.
+
+CONFIG_PCIEAER_INJECT=y or CONFIG_PCIEAER_INJECT=m
+
+After reboot with new kernel or insert the module, a device file named
+/dev/aer_inject should be created.
+
+Then, you need a user space tool named aer-inject, which can be gotten
+from:
+    http://www.kernel.org/pub/linux/kernel/people/yhuang/
+
+More information about aer-inject can be found in the document comes
+with its source code.
-- 
cgit v1.2.3-59-g8ed1b


From 1298307736a5882352418b4dc7c65442ab3b8bd4 Mon Sep 17 00:00:00 2001
From: Andreas Herrmann <andreas.herrmann3@amd.com>
Date: Sun, 7 Jun 2009 16:15:16 +0200
Subject: x86/PCI: add description for check_enable_amd_mmconf boot parameter

Describe check_enable_amd_mmconf.

Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 Documentation/kernel-parameters.txt | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 395d1a013ebb..8a82489d9a81 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1735,6 +1735,9 @@ and is between 256 and 4096 characters. It is defined in the file
 				root domains (aka PCI segments, in ACPI-speak).
 		nommconf	[X86] Disable use of MMCONFIG for PCI
 				Configuration
+		check_enable_amd_mmconf [X86] check for and enable
+				properly configured MMIO access to PCI
+				config space on AMD family 10h CPU
 		nomsi		[MSI] If the PCI_MSI kernel config parameter is
 				enabled, this kernel boot option can be used to
 				disable the use of MSI interrupts system-wide.
-- 
cgit v1.2.3-59-g8ed1b


From bd3d99c17039fd05a29587db3f4a180c48da115a Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Tue, 2 Jun 2009 13:52:26 +0900
Subject: PCI: Remove untested Electromechanical Interlock (EMI) support in
 pciehp.

The EMI support in pciehp is obviously broken. It is implemented using
struct hotplug_slot_attribute, but sysfs_ops for pci_slot_ktype is NOT
for struct hotplug_slot_attribute, but for struct pci_slot_attribute.
This bug had been there for a long time, maybe it was introduced when
PCI slot framework was introduced. The reason why this bug didn't
cause any problem is maybe the EMI support is not tested at all
because of lack of test environment.

As described above, the EMI support in pciehp seems not to be tested
at all. So this patch removes EMI support from pciehp, instead of
fixing the bug.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/pciehp.h      |   3 --
 drivers/pci/hotplug/pciehp_core.c | 111 +-------------------------------------
 drivers/pci/hotplug/pciehp_hpc.c  |  31 -----------
 include/linux/pci_hotplug.h       |   8 ---
 4 files changed, 1 insertion(+), 152 deletions(-)

diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h
index 0a368547e633..e6cf096498be 100644
--- a/drivers/pci/hotplug/pciehp.h
+++ b/drivers/pci/hotplug/pciehp.h
@@ -81,7 +81,6 @@ struct slot {
 	struct hpc_ops *hpc_ops;
 	struct hotplug_slot *hotplug_slot;
 	struct list_head	slot_list;
-	unsigned long last_emi_toggle;
 	struct delayed_work work;	/* work for button event */
 	struct mutex lock;
 };
@@ -203,8 +202,6 @@ struct hpc_ops {
 	int (*set_attention_status)(struct slot *slot, u8 status);
 	int (*get_latch_status)(struct slot *slot, u8 *status);
 	int (*get_adapter_status)(struct slot *slot, u8 *status);
-	int (*get_emi_status)(struct slot *slot, u8 *status);
-	int (*toggle_emi)(struct slot *slot);
 	int (*get_max_bus_speed)(struct slot *slot, enum pci_bus_speed *speed);
 	int (*get_cur_bus_speed)(struct slot *slot, enum pci_bus_speed *speed);
 	int (*get_max_lnk_width)(struct slot *slot, enum pcie_link_width *val);
diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c
index fb254b2454de..eb183d1d0912 100644
--- a/drivers/pci/hotplug/pciehp_core.c
+++ b/drivers/pci/hotplug/pciehp_core.c
@@ -85,99 +85,6 @@ static struct hotplug_slot_ops pciehp_hotplug_slot_ops = {
   	.get_cur_bus_speed =	get_cur_bus_speed,
 };
 
-/*
- * Check the status of the Electro Mechanical Interlock (EMI)
- */
-static int get_lock_status(struct hotplug_slot *hotplug_slot, u8 *value)
-{
-	struct slot *slot = hotplug_slot->private;
-	return (slot->hpc_ops->get_emi_status(slot, value));
-}
-
-/*
- * sysfs interface for the Electro Mechanical Interlock (EMI)
- * 1 == locked, 0 == unlocked
- */
-static ssize_t lock_read_file(struct hotplug_slot *slot, char *buf)
-{
-	int retval;
-	u8 value;
-
-	retval = get_lock_status(slot, &value);
-	if (retval)
-		goto lock_read_exit;
-	retval = sprintf (buf, "%d\n", value);
-
-lock_read_exit:
-	return retval;
-}
-
-/*
- * Change the status of the Electro Mechanical Interlock (EMI)
- * This is a toggle - in addition there must be at least 1 second
- * in between toggles.
- */
-static int set_lock_status(struct hotplug_slot *hotplug_slot, u8 status)
-{
-	struct slot *slot = hotplug_slot->private;
-	int retval;
-	u8 value;
-
-	mutex_lock(&slot->ctrl->crit_sect);
-
-	/* has it been >1 sec since our last toggle? */
-	if ((get_seconds() - slot->last_emi_toggle) < 1) {
-		mutex_unlock(&slot->ctrl->crit_sect);
-		return -EINVAL;
-	}
-
-	/* see what our current state is */
-	retval = get_lock_status(hotplug_slot, &value);
-	if (retval || (value == status))
-		goto set_lock_exit;
-
-	slot->hpc_ops->toggle_emi(slot);
-set_lock_exit:
-	mutex_unlock(&slot->ctrl->crit_sect);
-	return 0;
-}
-
-/*
- * sysfs interface which allows the user to toggle the Electro Mechanical
- * Interlock.  Valid values are either 0 or 1.  0 == unlock, 1 == lock
- */
-static ssize_t lock_write_file(struct hotplug_slot *hotplug_slot,
-		const char *buf, size_t count)
-{
-	struct slot *slot = hotplug_slot->private;
-	unsigned long llock;
-	u8 lock;
-	int retval = 0;
-
-	llock = simple_strtoul(buf, NULL, 10);
-	lock = (u8)(llock & 0xff);
-
-	switch (lock) {
-		case 0:
-		case 1:
-			retval = set_lock_status(hotplug_slot, lock);
-			break;
-		default:
-			ctrl_err(slot->ctrl, "%d is an invalid lock value\n",
-				 lock);
-			retval = -EINVAL;
-	}
-	if (retval)
-		return retval;
-	return count;
-}
-
-static struct hotplug_slot_attribute hotplug_slot_attr_lock = {
-	.attr = {.name = "lock", .mode = S_IFREG | S_IRUGO | S_IWUSR},
-	.show = lock_read_file,
-	.store = lock_write_file
-};
-
 /**
  * release_slot - free up the memory used by a slot
  * @hotplug_slot: slot to free
@@ -236,17 +143,6 @@ static int init_slots(struct controller *ctrl)
 		get_attention_status(hotplug_slot, &info->attention_status);
 		get_latch_status(hotplug_slot, &info->latch_status);
 		get_adapter_status(hotplug_slot, &info->adapter_status);
-		/* create additional sysfs entries */
-		if (EMI(ctrl)) {
-			retval = sysfs_create_file(&hotplug_slot->pci_slot->kobj,
-				&hotplug_slot_attr_lock.attr);
-			if (retval) {
-				pci_hp_deregister(hotplug_slot);
-				ctrl_err(ctrl, "Cannot create additional sysfs "
-					 "entries\n");
-				goto error_info;
-			}
-		}
 	}
 
 	return 0;
@@ -261,13 +157,8 @@ error:
 static void cleanup_slots(struct controller *ctrl)
 {
 	struct slot *slot;
-
-	list_for_each_entry(slot, &ctrl->slot_list, slot_list) {
-		if (EMI(ctrl))
-			sysfs_remove_file(&slot->hotplug_slot->pci_slot->kobj,
-				&hotplug_slot_attr_lock.attr);
+	list_for_each_entry(slot, &ctrl->slot_list, slot_list)
 		pci_hp_deregister(slot->hotplug_slot);
-	}
 }
 
 /*
diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c
index 07bd32151146..52813257e5bf 100644
--- a/drivers/pci/hotplug/pciehp_hpc.c
+++ b/drivers/pci/hotplug/pciehp_hpc.c
@@ -422,35 +422,6 @@ static int hpc_query_power_fault(struct slot *slot)
 	return !!(slot_status & PCI_EXP_SLTSTA_PFD);
 }
 
-static int hpc_get_emi_status(struct slot *slot, u8 *status)
-{
-	struct controller *ctrl = slot->ctrl;
-	u16 slot_status;
-	int retval;
-
-	retval = pciehp_readw(ctrl, PCI_EXP_SLTSTA, &slot_status);
-	if (retval) {
-		ctrl_err(ctrl, "Cannot check EMI status\n");
-		return retval;
-	}
-	*status = !!(slot_status & PCI_EXP_SLTSTA_EIS);
-	return retval;
-}
-
-static int hpc_toggle_emi(struct slot *slot)
-{
-	u16 slot_cmd;
-	u16 cmd_mask;
-	int rc;
-
-	slot_cmd = PCI_EXP_SLTCTL_EIC;
-	cmd_mask = PCI_EXP_SLTCTL_EIC;
-	rc = pcie_write_cmd(slot->ctrl, slot_cmd, cmd_mask);
-	slot->last_emi_toggle = get_seconds();
-
-	return rc;
-}
-
 static int hpc_set_attention_status(struct slot *slot, u8 value)
 {
 	struct controller *ctrl = slot->ctrl;
@@ -874,8 +845,6 @@ static struct hpc_ops pciehp_hpc_ops = {
 	.get_attention_status		= hpc_get_attention_status,
 	.get_latch_status		= hpc_get_latch_status,
 	.get_adapter_status		= hpc_get_adapter_status,
-	.get_emi_status			= hpc_get_emi_status,
-	.toggle_emi			= hpc_toggle_emi,
 
 	.get_max_bus_speed		= hpc_get_max_lnk_speed,
 	.get_cur_bus_speed		= hpc_get_cur_lnk_speed,
diff --git a/include/linux/pci_hotplug.h b/include/linux/pci_hotplug.h
index 20998746518e..11936fd0b56d 100644
--- a/include/linux/pci_hotplug.h
+++ b/include/linux/pci_hotplug.h
@@ -66,14 +66,6 @@ enum pcie_link_speed {
 	PCIE_LNK_SPEED_UNKNOWN	= 0xFF,
 };
 
-struct hotplug_slot;
-struct hotplug_slot_attribute {
-	struct attribute attr;
-	ssize_t (*show)(struct hotplug_slot *, char *);
-	ssize_t (*store)(struct hotplug_slot *, const char *, size_t);
-};
-#define to_hotplug_attr(n) container_of(n, struct hotplug_slot_attribute, attr);
-
 /**
  * struct hotplug_slot_ops -the callbacks that the hotplug pci core can use
  * @owner: The module owner of this structure
-- 
cgit v1.2.3-59-g8ed1b


From 498a8faf2c7eb974f70b7c5a60a31f0d48c35d44 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Tue, 16 Jun 2009 11:00:47 +0900
Subject: PCI hotplug: fix return value of has_foo() functions

Current has_foo() functions in pci_hotplug_core.c returns 0 if the
"foo" property is true. It would cause misunderstanding. In addition,
the error code of those functions is never checked, so this patch
changes those functions' error code to 'bool' and return true if the
property "foo" is true.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Reviewed-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/pci_hotplug_core.c | 132 +++++++++++++++++----------------
 1 file changed, 68 insertions(+), 64 deletions(-)

diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c
index 535fce0f07f9..ff32c6b4ae13 100644
--- a/drivers/pci/hotplug/pci_hotplug_core.c
+++ b/drivers/pci/hotplug/pci_hotplug_core.c
@@ -347,125 +347,126 @@ static struct pci_slot_attribute hotplug_slot_attr_test = {
 	.store = test_write_file
 };
 
-static int has_power_file(struct pci_slot *pci_slot)
+static bool has_power_file(struct pci_slot *pci_slot)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
-		return -ENODEV;
+		return false;
 	if ((slot->ops->enable_slot) ||
 	    (slot->ops->disable_slot) ||
 	    (slot->ops->get_power_status))
-		return 0;
-	return -ENOENT;
+		return true;
+	return false;
 }
 
-static int has_attention_file(struct pci_slot *pci_slot)
+static bool has_attention_file(struct pci_slot *pci_slot)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
-		return -ENODEV;
+		return false;
 	if ((slot->ops->set_attention_status) ||
 	    (slot->ops->get_attention_status))
-		return 0;
-	return -ENOENT;
+		return true;
+	return false;
 }
 
-static int has_latch_file(struct pci_slot *pci_slot)
+static bool has_latch_file(struct pci_slot *pci_slot)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
-		return -ENODEV;
+		return false;
 	if (slot->ops->get_latch_status)
-		return 0;
-	return -ENOENT;
+		return true;
+	return false;
 }
 
-static int has_adapter_file(struct pci_slot *pci_slot)
+static bool has_adapter_file(struct pci_slot *pci_slot)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
-		return -ENODEV;
+		return false;
 	if (slot->ops->get_adapter_status)
-		return 0;
-	return -ENOENT;
+		return true;
+	return false;
 }
 
-static int has_max_bus_speed_file(struct pci_slot *pci_slot)
+static bool has_max_bus_speed_file(struct pci_slot *pci_slot)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
-		return -ENODEV;
+		return false;
 	if (slot->ops->get_max_bus_speed)
-		return 0;
-	return -ENOENT;
+		return true;
+	return false;
 }
 
-static int has_cur_bus_speed_file(struct pci_slot *pci_slot)
+static bool has_cur_bus_speed_file(struct pci_slot *pci_slot)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
-		return -ENODEV;
+		return false;
 	if (slot->ops->get_cur_bus_speed)
-		return 0;
-	return -ENOENT;
+		return true;
+	return false;
 }
 
-static int has_test_file(struct pci_slot *pci_slot)
+static bool has_test_file(struct pci_slot *pci_slot)
 {
 	struct hotplug_slot *slot = pci_slot->hotplug;
 	if ((!slot) || (!slot->ops))
-		return -ENODEV;
+		return false;
 	if (slot->ops->hardware_test)
-		return 0;
-	return -ENOENT;
+		return true;
+	return false;
 }
 
 static int fs_add_slot(struct pci_slot *slot)
 {
 	int retval = 0;
 
-	if (has_power_file(slot) == 0) {
-		retval = sysfs_create_file(&slot->kobj, &hotplug_slot_attr_power.attr);
+	if (has_power_file(slot)) {
+		retval = sysfs_create_file(&slot->kobj,
+					   &hotplug_slot_attr_power.attr);
 		if (retval)
 			goto exit_power;
 	}
 
-	if (has_attention_file(slot) == 0) {
+	if (has_attention_file(slot)) {
 		retval = sysfs_create_file(&slot->kobj,
 					   &hotplug_slot_attr_attention.attr);
 		if (retval)
 			goto exit_attention;
 	}
 
-	if (has_latch_file(slot) == 0) {
+	if (has_latch_file(slot)) {
 		retval = sysfs_create_file(&slot->kobj,
 					   &hotplug_slot_attr_latch.attr);
 		if (retval)
 			goto exit_latch;
 	}
 
-	if (has_adapter_file(slot) == 0) {
+	if (has_adapter_file(slot)) {
 		retval = sysfs_create_file(&slot->kobj,
 					   &hotplug_slot_attr_presence.attr);
 		if (retval)
 			goto exit_adapter;
 	}
 
-	if (has_max_bus_speed_file(slot) == 0) {
+	if (has_max_bus_speed_file(slot)) {
 		retval = sysfs_create_file(&slot->kobj,
-					   &hotplug_slot_attr_max_bus_speed.attr);
+					&hotplug_slot_attr_max_bus_speed.attr);
 		if (retval)
 			goto exit_max_speed;
 	}
 
-	if (has_cur_bus_speed_file(slot) == 0) {
+	if (has_cur_bus_speed_file(slot)) {
 		retval = sysfs_create_file(&slot->kobj,
-					   &hotplug_slot_attr_cur_bus_speed.attr);
+					&hotplug_slot_attr_cur_bus_speed.attr);
 		if (retval)
 			goto exit_cur_speed;
 	}
 
-	if (has_test_file(slot) == 0) {
+	if (has_test_file(slot)) {
 		retval = sysfs_create_file(&slot->kobj,
 					   &hotplug_slot_attr_test.attr);
 		if (retval)
@@ -475,27 +476,26 @@ static int fs_add_slot(struct pci_slot *slot)
 	goto exit;
 
 exit_test:
-	if (has_cur_bus_speed_file(slot) == 0)
-		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_cur_bus_speed.attr);
-
+	if (has_cur_bus_speed_file(slot))
+		sysfs_remove_file(&slot->kobj,
+				  &hotplug_slot_attr_cur_bus_speed.attr);
 exit_cur_speed:
-	if (has_max_bus_speed_file(slot) == 0)
-		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_max_bus_speed.attr);
-
+	if (has_max_bus_speed_file(slot))
+		sysfs_remove_file(&slot->kobj,
+				  &hotplug_slot_attr_max_bus_speed.attr);
 exit_max_speed:
-	if (has_adapter_file(slot) == 0)
-		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_presence.attr);
-
+	if (has_adapter_file(slot))
+		sysfs_remove_file(&slot->kobj,
+				  &hotplug_slot_attr_presence.attr);
 exit_adapter:
-	if (has_latch_file(slot) == 0)
+	if (has_latch_file(slot))
 		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_latch.attr);
-
 exit_latch:
-	if (has_attention_file(slot) == 0)
-		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_attention.attr);
-
+	if (has_attention_file(slot))
+		sysfs_remove_file(&slot->kobj,
+				  &hotplug_slot_attr_attention.attr);
 exit_attention:
-	if (has_power_file(slot) == 0)
+	if (has_power_file(slot))
 		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_power.attr);
 exit_power:
 exit:
@@ -504,25 +504,29 @@ exit:
 
 static void fs_remove_slot(struct pci_slot *slot)
 {
-	if (has_power_file(slot) == 0)
+	if (has_power_file(slot))
 		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_power.attr);
 
-	if (has_attention_file(slot) == 0)
-		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_attention.attr);
+	if (has_attention_file(slot))
+		sysfs_remove_file(&slot->kobj,
+				  &hotplug_slot_attr_attention.attr);
 
-	if (has_latch_file(slot) == 0)
+	if (has_latch_file(slot))
 		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_latch.attr);
 
-	if (has_adapter_file(slot) == 0)
-		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_presence.attr);
+	if (has_adapter_file(slot))
+		sysfs_remove_file(&slot->kobj,
+				  &hotplug_slot_attr_presence.attr);
 
-	if (has_max_bus_speed_file(slot) == 0)
-		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_max_bus_speed.attr);
+	if (has_max_bus_speed_file(slot))
+		sysfs_remove_file(&slot->kobj,
+				  &hotplug_slot_attr_max_bus_speed.attr);
 
-	if (has_cur_bus_speed_file(slot) == 0)
-		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_cur_bus_speed.attr);
+	if (has_cur_bus_speed_file(slot))
+		sysfs_remove_file(&slot->kobj,
+				  &hotplug_slot_attr_cur_bus_speed.attr);
 
-	if (has_test_file(slot) == 0)
+	if (has_test_file(slot))
 		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_test.attr);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From c825bc94c8c1908750ab20413eb639c6be029e2d Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Tue, 16 Jun 2009 11:01:25 +0900
Subject: PCI hotplug: create symlink to hotplug driver module

Create symbolic link to hotplug driver module in the PCI slot
directory (/sys/bus/pci/slots/<SLOT#>). In the past, we need to load
hotplug drivers one by one to identify the hotplug driver that handles
the slot, and it was very inconvenient especially for trouble shooting.
With this change, we can easily identify the hotplug driver.

Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Reviewed-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 Documentation/ABI/testing/sysfs-bus-pci |  7 ++++++
 drivers/pci/hotplug/pci_hotplug_core.c  | 23 +++++++++++++------
 drivers/pci/slot.c                      | 39 +++++++++++++++++++++++++++++++++
 include/linux/pci.h                     |  5 +++++
 include/linux/pci_hotplug.h             | 15 +++++++++++--
 5 files changed, 80 insertions(+), 9 deletions(-)

diff --git a/Documentation/ABI/testing/sysfs-bus-pci b/Documentation/ABI/testing/sysfs-bus-pci
index 97ad190e13af..6bf68053e4b8 100644
--- a/Documentation/ABI/testing/sysfs-bus-pci
+++ b/Documentation/ABI/testing/sysfs-bus-pci
@@ -122,3 +122,10 @@ Description:
 		This symbolic link appears when a device is a Virtual Function.
 		The symbolic link points to the PCI device sysfs entry of the
 		Physical Function this device associates with.
+
+What:		/sys/bus/pci/slots/.../module
+Date:		June 2009
+Contact:	linux-pci@vger.kernel.org
+Description:
+		This symbolic link points to the PCI hotplug controller driver
+		module that manages the hotplug slot.
diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c
index ff32c6b4ae13..844580489d4d 100644
--- a/drivers/pci/hotplug/pci_hotplug_core.c
+++ b/drivers/pci/hotplug/pci_hotplug_core.c
@@ -424,6 +424,9 @@ static int fs_add_slot(struct pci_slot *slot)
 {
 	int retval = 0;
 
+	/* Create symbolic link to the hotplug driver module */
+	pci_hp_create_module_link(slot);
+
 	if (has_power_file(slot)) {
 		retval = sysfs_create_file(&slot->kobj,
 					   &hotplug_slot_attr_power.attr);
@@ -498,6 +501,7 @@ exit_attention:
 	if (has_power_file(slot))
 		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_power.attr);
 exit_power:
+	pci_hp_remove_module_link(slot);
 exit:
 	return retval;
 }
@@ -528,6 +532,8 @@ static void fs_remove_slot(struct pci_slot *slot)
 
 	if (has_test_file(slot))
 		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_test.attr);
+
+	pci_hp_remove_module_link(slot);
 }
 
 static struct hotplug_slot *get_slot_from_name (const char *name)
@@ -544,10 +550,10 @@ static struct hotplug_slot *get_slot_from_name (const char *name)
 }
 
 /**
- * pci_hp_register - register a hotplug_slot with the PCI hotplug subsystem
+ * __pci_hp_register - register a hotplug_slot with the PCI hotplug subsystem
  * @bus: bus this slot is on
  * @slot: pointer to the &struct hotplug_slot to register
- * @slot_nr: slot number
+ * @devnr: device number
  * @name: name registered with kobject core
  *
  * Registers a hotplug slot with the pci hotplug subsystem, which will allow
@@ -555,8 +561,9 @@ static struct hotplug_slot *get_slot_from_name (const char *name)
  *
  * Returns 0 if successful, anything else for an error.
  */
-int pci_hp_register(struct hotplug_slot *slot, struct pci_bus *bus, int slot_nr,
-			const char *name)
+int __pci_hp_register(struct hotplug_slot *slot, struct pci_bus *bus,
+		      int devnr, const char *name,
+		      struct module *owner, const char *mod_name)
 {
 	int result;
 	struct pci_slot *pci_slot;
@@ -571,14 +578,16 @@ int pci_hp_register(struct hotplug_slot *slot, struct pci_bus *bus, int slot_nr,
 		return -EINVAL;
 	}
 
-	mutex_lock(&pci_hp_mutex);
+	slot->ops->owner = owner;
+	slot->ops->mod_name = mod_name;
 
+	mutex_lock(&pci_hp_mutex);
 	/*
 	 * No problems if we call this interface from both ACPI_PCI_SLOT
 	 * driver and call it here again. If we've already created the
 	 * pci_slot, the interface will simply bump the refcount.
 	 */
-	pci_slot = pci_create_slot(bus, slot_nr, name, slot);
+	pci_slot = pci_create_slot(bus, devnr, name, slot);
 	if (IS_ERR(pci_slot)) {
 		result = PTR_ERR(pci_slot);
 		goto out;
@@ -688,6 +697,6 @@ MODULE_LICENSE("GPL");
 module_param(debug, bool, 0644);
 MODULE_PARM_DESC(debug, "Debugging mode enabled or not");
 
-EXPORT_SYMBOL_GPL(pci_hp_register);
+EXPORT_SYMBOL_GPL(__pci_hp_register);
 EXPORT_SYMBOL_GPL(pci_hp_deregister);
 EXPORT_SYMBOL_GPL(pci_hp_change_slot_info);
diff --git a/drivers/pci/slot.c b/drivers/pci/slot.c
index fe95ce20bcbd..eddb0748b0ea 100644
--- a/drivers/pci/slot.c
+++ b/drivers/pci/slot.c
@@ -307,6 +307,45 @@ void pci_destroy_slot(struct pci_slot *slot)
 }
 EXPORT_SYMBOL_GPL(pci_destroy_slot);
 
+#if defined(CONFIG_HOTPLUG_PCI) || defined(CONFIG_HOTPLUG_PCI_MODULE)
+#include <linux/pci_hotplug.h>
+/**
+ * pci_hp_create_link - create symbolic link to the hotplug driver module.
+ * @slot: struct pci_slot
+ *
+ * Helper function for pci_hotplug_core.c to create symbolic link to
+ * the hotplug driver module.
+ */
+void pci_hp_create_module_link(struct pci_slot *pci_slot)
+{
+	struct hotplug_slot *slot = pci_slot->hotplug;
+	struct kobject *kobj = NULL;
+	int no_warn;
+
+	if (!slot || !slot->ops)
+		return;
+	kobj = kset_find_obj(module_kset, slot->ops->mod_name);
+	if (!kobj)
+		return;
+	no_warn = sysfs_create_link(&pci_slot->kobj, kobj, "module");
+	kobject_put(kobj);
+}
+EXPORT_SYMBOL_GPL(pci_hp_create_module_link);
+
+/**
+ * pci_hp_remove_link - remove symbolic link to the hotplug driver module.
+ * @slot: struct pci_slot
+ *
+ * Helper function for pci_hotplug_core.c to remove symbolic link to
+ * the hotplug driver module.
+ */
+void pci_hp_remove_module_link(struct pci_slot *pci_slot)
+{
+	sysfs_remove_link(&pci_slot->kobj, "module");
+}
+EXPORT_SYMBOL_GPL(pci_hp_remove_module_link);
+#endif
+
 static int pci_slot_init(void)
 {
 	struct kset *pci_bus_kset;
diff --git a/include/linux/pci.h b/include/linux/pci.h
index ea2a153a9126..6a1800ecd95d 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -1261,5 +1261,10 @@ static inline irqreturn_t pci_sriov_migration(struct pci_dev *dev)
 }
 #endif
 
+#if defined(CONFIG_HOTPLUG_PCI) || defined(CONFIG_HOTPLUG_PCI_MODULE)
+extern void pci_hp_create_module_link(struct pci_slot *pci_slot);
+extern void pci_hp_remove_module_link(struct pci_slot *pci_slot);
+#endif
+
 #endif /* __KERNEL__ */
 #endif /* LINUX_PCI_H */
diff --git a/include/linux/pci_hotplug.h b/include/linux/pci_hotplug.h
index 11936fd0b56d..b3646cd7fd5a 100644
--- a/include/linux/pci_hotplug.h
+++ b/include/linux/pci_hotplug.h
@@ -69,6 +69,7 @@ enum pcie_link_speed {
 /**
  * struct hotplug_slot_ops -the callbacks that the hotplug pci core can use
  * @owner: The module owner of this structure
+ * @mod_name: The module name (KBUILD_MODNAME) of this structure
  * @enable_slot: Called when the user wants to enable a specific pci slot
  * @disable_slot: Called when the user wants to disable a specific pci slot
  * @set_attention_status: Called to set the specific slot's attention LED to
@@ -101,6 +102,7 @@ enum pcie_link_speed {
  */
 struct hotplug_slot_ops {
 	struct module *owner;
+	const char *mod_name;
 	int (*enable_slot)		(struct hotplug_slot *slot);
 	int (*disable_slot)		(struct hotplug_slot *slot);
 	int (*set_attention_status)	(struct hotplug_slot *slot, u8 value);
@@ -159,12 +161,21 @@ static inline const char *hotplug_slot_name(const struct hotplug_slot *slot)
 	return pci_slot_name(slot->pci_slot);
 }
 
-extern int pci_hp_register(struct hotplug_slot *, struct pci_bus *, int nr,
-			   const char *name);
+extern int __pci_hp_register(struct hotplug_slot *slot, struct pci_bus *pbus,
+			     int nr, const char *name,
+			     struct module *owner, const char *mod_name);
 extern int pci_hp_deregister(struct hotplug_slot *slot);
 extern int __must_check pci_hp_change_slot_info	(struct hotplug_slot *slot,
 						 struct hotplug_slot_info *info);
 
+static inline int pci_hp_register(struct hotplug_slot *slot,
+				  struct pci_bus *pbus,
+				  int devnr, const char *name)
+{
+	return __pci_hp_register(slot, pbus, devnr, name,
+				 THIS_MODULE, KBUILD_MODNAME);
+}
+
 /* PCI Setting Record (Type 0) */
 struct hpp_type0 {
 	u32 revision;
-- 
cgit v1.2.3-59-g8ed1b


From a6c0d5c6ebb3d988b1f18a1612b5188f3f555637 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Tue, 16 Jun 2009 11:02:02 +0900
Subject: PCI hotplug: remove redundant .owner initializations

The "owner" field in struct hotplug_slot_ops is initialized by PCI
hotplug core. So each hotplug controller driver doesn't need to
initialize it.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Reviewed-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/hotplug/acpiphp_core.c      | 1 -
 drivers/pci/hotplug/cpci_hotplug_core.c | 1 -
 drivers/pci/hotplug/cpqphp_core.c       | 1 -
 drivers/pci/hotplug/ibmphp_core.c       | 1 -
 drivers/pci/hotplug/pciehp_core.c       | 1 -
 drivers/pci/hotplug/pcihp_skeleton.c    | 1 -
 drivers/pci/hotplug/rpaphp_core.c       | 1 -
 drivers/pci/hotplug/sgi_hotplug.c       | 1 -
 drivers/pci/hotplug/shpchp_core.c       | 1 -
 9 files changed, 9 deletions(-)

diff --git a/drivers/pci/hotplug/acpiphp_core.c b/drivers/pci/hotplug/acpiphp_core.c
index 43c10bd261b4..4dd7114964ac 100644
--- a/drivers/pci/hotplug/acpiphp_core.c
+++ b/drivers/pci/hotplug/acpiphp_core.c
@@ -77,7 +77,6 @@ static int get_latch_status	(struct hotplug_slot *slot, u8 *value);
 static int get_adapter_status	(struct hotplug_slot *slot, u8 *value);
 
 static struct hotplug_slot_ops acpi_hotplug_slot_ops = {
-	.owner			= THIS_MODULE,
 	.enable_slot		= enable_slot,
 	.disable_slot		= disable_slot,
 	.set_attention_status	= set_attention_status,
diff --git a/drivers/pci/hotplug/cpci_hotplug_core.c b/drivers/pci/hotplug/cpci_hotplug_core.c
index de94f4feef8c..a5b9f6ae507b 100644
--- a/drivers/pci/hotplug/cpci_hotplug_core.c
+++ b/drivers/pci/hotplug/cpci_hotplug_core.c
@@ -72,7 +72,6 @@ static int get_adapter_status(struct hotplug_slot *slot, u8 * value);
 static int get_latch_status(struct hotplug_slot *slot, u8 * value);
 
 static struct hotplug_slot_ops cpci_hotplug_slot_ops = {
-	.owner = THIS_MODULE,
 	.enable_slot = enable_slot,
 	.disable_slot = disable_slot,
 	.set_attention_status = set_attention_status,
diff --git a/drivers/pci/hotplug/cpqphp_core.c b/drivers/pci/hotplug/cpqphp_core.c
index 7888b37c6c8e..075b4f4b6e0d 100644
--- a/drivers/pci/hotplug/cpqphp_core.c
+++ b/drivers/pci/hotplug/cpqphp_core.c
@@ -608,7 +608,6 @@ static int get_cur_bus_speed (struct hotplug_slot *hotplug_slot, enum pci_bus_sp
 }
 
 static struct hotplug_slot_ops cpqphp_hotplug_slot_ops = {
-	.owner =		THIS_MODULE,
 	.set_attention_status =	set_attention_status,
 	.enable_slot =		process_SI,
 	.disable_slot =		process_SS,
diff --git a/drivers/pci/hotplug/ibmphp_core.c b/drivers/pci/hotplug/ibmphp_core.c
index 29ccb8a6da8a..b3d5d47e6980 100644
--- a/drivers/pci/hotplug/ibmphp_core.c
+++ b/drivers/pci/hotplug/ibmphp_core.c
@@ -1316,7 +1316,6 @@ error:
 }
 
 struct hotplug_slot_ops ibmphp_hotplug_slot_ops = {
-	.owner =			THIS_MODULE,
 	.set_attention_status =		set_attention_status,
 	.enable_slot =			enable_slot,
 	.disable_slot =			ibmphp_disable_slot,
diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c
index eb183d1d0912..2317557fdee6 100644
--- a/drivers/pci/hotplug/pciehp_core.c
+++ b/drivers/pci/hotplug/pciehp_core.c
@@ -73,7 +73,6 @@ static int get_max_bus_speed	(struct hotplug_slot *slot, enum pci_bus_speed *val
 static int get_cur_bus_speed	(struct hotplug_slot *slot, enum pci_bus_speed *value);
 
 static struct hotplug_slot_ops pciehp_hotplug_slot_ops = {
-	.owner =		THIS_MODULE,
 	.set_attention_status =	set_attention_status,
 	.enable_slot =		enable_slot,
 	.disable_slot =		disable_slot,
diff --git a/drivers/pci/hotplug/pcihp_skeleton.c b/drivers/pci/hotplug/pcihp_skeleton.c
index e3dd6cf9e89f..5175d9b26f0b 100644
--- a/drivers/pci/hotplug/pcihp_skeleton.c
+++ b/drivers/pci/hotplug/pcihp_skeleton.c
@@ -82,7 +82,6 @@ static int get_latch_status	(struct hotplug_slot *slot, u8 *value);
 static int get_adapter_status	(struct hotplug_slot *slot, u8 *value);
 
 static struct hotplug_slot_ops skel_hotplug_slot_ops = {
-	.owner =		THIS_MODULE,
 	.enable_slot =		enable_slot,
 	.disable_slot =		disable_slot,
 	.set_attention_status =	set_attention_status,
diff --git a/drivers/pci/hotplug/rpaphp_core.c b/drivers/pci/hotplug/rpaphp_core.c
index 95d02a08fdc7..c159223389ec 100644
--- a/drivers/pci/hotplug/rpaphp_core.c
+++ b/drivers/pci/hotplug/rpaphp_core.c
@@ -423,7 +423,6 @@ static int disable_slot(struct hotplug_slot *hotplug_slot)
 }
 
 struct hotplug_slot_ops rpaphp_hotplug_slot_ops = {
-	.owner = THIS_MODULE,
 	.enable_slot = enable_slot,
 	.disable_slot = disable_slot,
 	.set_attention_status = set_attention_status,
diff --git a/drivers/pci/hotplug/sgi_hotplug.c b/drivers/pci/hotplug/sgi_hotplug.c
index 3eee70928d45..25a48b9f625e 100644
--- a/drivers/pci/hotplug/sgi_hotplug.c
+++ b/drivers/pci/hotplug/sgi_hotplug.c
@@ -83,7 +83,6 @@ static int disable_slot(struct hotplug_slot *slot);
 static inline int get_power_status(struct hotplug_slot *slot, u8 *value);
 
 static struct hotplug_slot_ops sn_hotplug_slot_ops = {
-	.owner                  = THIS_MODULE,
 	.enable_slot            = enable_slot,
 	.disable_slot           = disable_slot,
 	.get_power_status       = get_power_status,
diff --git a/drivers/pci/hotplug/shpchp_core.c b/drivers/pci/hotplug/shpchp_core.c
index fe8d149c2293..8a520a3d0f59 100644
--- a/drivers/pci/hotplug/shpchp_core.c
+++ b/drivers/pci/hotplug/shpchp_core.c
@@ -69,7 +69,6 @@ static int get_max_bus_speed	(struct hotplug_slot *slot, enum pci_bus_speed *val
 static int get_cur_bus_speed	(struct hotplug_slot *slot, enum pci_bus_speed *value);
 
 static struct hotplug_slot_ops shpchp_hotplug_slot_ops = {
-	.owner =		THIS_MODULE,
 	.set_attention_status =	set_attention_status,
 	.enable_slot =		enable_slot,
 	.disable_slot =		disable_slot,
-- 
cgit v1.2.3-59-g8ed1b


From 70298c6e6c1ba68346336b4ea54bd5c0abbf73c8 Mon Sep 17 00:00:00 2001
From: "Zhang, Yanmin" <yanmin_zhang@linux.intel.com>
Date: Tue, 16 Jun 2009 13:34:38 +0800
Subject: PCI AER: support Multiple Error Received and no error source id

Based on PCI Express AER specs, a root port might receive multiple
TLP errors while it could only save a correctable error source id
and an uncorrectable error source id at the same time. In addition,
some root port hardware might be unable to provide a correct source
id, i.e., the source id, or the bus id part of the source id provided
by root port might be equal to 0.

The patchset implements the support in kernel by searching the device
tree under the root port.

Patch 1 changes parameter cb of function pci_walk_bus to return a value.
When cb return non-zero, pci_walk_bus stops more searching on the
device tree.

Reviewed-by: Andrew Patterson <andrew.patterson@hp.com>
Signed-off-by: Zhang Yanmin <yanmin_zhang@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/powerpc/platforms/pseries/eeh_driver.c | 38 ++++++++++++++++++-----------
 drivers/pci/bus.c                           | 11 +++++++--
 drivers/pci/pcie/aer/aerdrv_core.c          | 30 ++++++++++++-----------
 include/linux/pci.h                         |  2 +-
 4 files changed, 50 insertions(+), 31 deletions(-)

diff --git a/arch/powerpc/platforms/pseries/eeh_driver.c b/arch/powerpc/platforms/pseries/eeh_driver.c
index 9a2a6e32f00f..0e8db6771252 100644
--- a/arch/powerpc/platforms/pseries/eeh_driver.c
+++ b/arch/powerpc/platforms/pseries/eeh_driver.c
@@ -122,7 +122,7 @@ static void eeh_enable_irq(struct pci_dev *dev)
  * passed back in "userdata".
  */
 
-static void eeh_report_error(struct pci_dev *dev, void *userdata)
+static int eeh_report_error(struct pci_dev *dev, void *userdata)
 {
 	enum pci_ers_result rc, *res = userdata;
 	struct pci_driver *driver = dev->driver;
@@ -130,19 +130,21 @@ static void eeh_report_error(struct pci_dev *dev, void *userdata)
 	dev->error_state = pci_channel_io_frozen;
 
 	if (!driver)
-		return;
+		return 0;
 
 	eeh_disable_irq(dev);
 
 	if (!driver->err_handler ||
 	    !driver->err_handler->error_detected)
-		return;
+		return 0;
 
 	rc = driver->err_handler->error_detected (dev, pci_channel_io_frozen);
 
 	/* A driver that needs a reset trumps all others */
 	if (rc == PCI_ERS_RESULT_NEED_RESET) *res = rc;
 	if (*res == PCI_ERS_RESULT_NONE) *res = rc;
+
+	return 0;
 }
 
 /**
@@ -153,7 +155,7 @@ static void eeh_report_error(struct pci_dev *dev, void *userdata)
  * Cumulative response passed back in "userdata".
  */
 
-static void eeh_report_mmio_enabled(struct pci_dev *dev, void *userdata)
+static int eeh_report_mmio_enabled(struct pci_dev *dev, void *userdata)
 {
 	enum pci_ers_result rc, *res = userdata;
 	struct pci_driver *driver = dev->driver;
@@ -161,26 +163,28 @@ static void eeh_report_mmio_enabled(struct pci_dev *dev, void *userdata)
 	if (!driver ||
 	    !driver->err_handler ||
 	    !driver->err_handler->mmio_enabled)
-		return;
+		return 0;
 
 	rc = driver->err_handler->mmio_enabled (dev);
 
 	/* A driver that needs a reset trumps all others */
 	if (rc == PCI_ERS_RESULT_NEED_RESET) *res = rc;
 	if (*res == PCI_ERS_RESULT_NONE) *res = rc;
+
+	return 0;
 }
 
 /**
  * eeh_report_reset - tell device that slot has been reset
  */
 
-static void eeh_report_reset(struct pci_dev *dev, void *userdata)
+static int eeh_report_reset(struct pci_dev *dev, void *userdata)
 {
 	enum pci_ers_result rc, *res = userdata;
 	struct pci_driver *driver = dev->driver;
 
 	if (!driver)
-		return;
+		return 0;
 
 	dev->error_state = pci_channel_io_normal;
 
@@ -188,35 +192,39 @@ static void eeh_report_reset(struct pci_dev *dev, void *userdata)
 
 	if (!driver->err_handler ||
 	    !driver->err_handler->slot_reset)
-		return;
+		return 0;
 
 	rc = driver->err_handler->slot_reset(dev);
 	if ((*res == PCI_ERS_RESULT_NONE) ||
 	    (*res == PCI_ERS_RESULT_RECOVERED)) *res = rc;
 	if (*res == PCI_ERS_RESULT_DISCONNECT &&
 	     rc == PCI_ERS_RESULT_NEED_RESET) *res = rc;
+
+	return 0;
 }
 
 /**
  * eeh_report_resume - tell device to resume normal operations
  */
 
-static void eeh_report_resume(struct pci_dev *dev, void *userdata)
+static int eeh_report_resume(struct pci_dev *dev, void *userdata)
 {
 	struct pci_driver *driver = dev->driver;
 
 	dev->error_state = pci_channel_io_normal;
 
 	if (!driver)
-		return;
+		return 0;
 
 	eeh_enable_irq(dev);
 
 	if (!driver->err_handler ||
 	    !driver->err_handler->resume)
-		return;
+		return 0;
 
 	driver->err_handler->resume(dev);
+
+	return 0;
 }
 
 /**
@@ -226,22 +234,24 @@ static void eeh_report_resume(struct pci_dev *dev, void *userdata)
  * dead, and that no further recovery attempts will be made on it.
  */
 
-static void eeh_report_failure(struct pci_dev *dev, void *userdata)
+static int eeh_report_failure(struct pci_dev *dev, void *userdata)
 {
 	struct pci_driver *driver = dev->driver;
 
 	dev->error_state = pci_channel_io_perm_failure;
 
 	if (!driver)
-		return;
+		return 0;
 
 	eeh_disable_irq(dev);
 
 	if (!driver->err_handler ||
 	    !driver->err_handler->error_detected)
-		return;
+		return 0;
 
 	driver->err_handler->error_detected(dev, pci_channel_io_perm_failure);
+
+	return 0;
 }
 
 /* ------------------------------------------------------- */
diff --git a/drivers/pci/bus.c b/drivers/pci/bus.c
index 40af27f31043..cef28a79103f 100644
--- a/drivers/pci/bus.c
+++ b/drivers/pci/bus.c
@@ -206,13 +206,18 @@ void pci_enable_bridges(struct pci_bus *bus)
  *  Walk the given bus, including any bridged devices
  *  on buses under this bus.  Call the provided callback
  *  on each device found.
+ *
+ *  We check the return of @cb each time. If it returns anything
+ *  other than 0, we break out.
+ *
  */
-void pci_walk_bus(struct pci_bus *top, void (*cb)(struct pci_dev *, void *),
+void pci_walk_bus(struct pci_bus *top, int (*cb)(struct pci_dev *, void *),
 		  void *userdata)
 {
 	struct pci_dev *dev;
 	struct pci_bus *bus;
 	struct list_head *next;
+	int retval;
 
 	bus = top;
 	down_read(&pci_bus_sem);
@@ -236,8 +241,10 @@ void pci_walk_bus(struct pci_bus *top, void (*cb)(struct pci_dev *, void *),
 
 		/* Run device routines with the device locked */
 		down(&dev->dev.sem);
-		cb(dev, userdata);
+		retval = cb(dev, userdata);
 		up(&dev->dev.sem);
+		if (retval)
+			break;
 	}
 	up_read(&pci_bus_sem);
 }
diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c
index dd3829e68e3f..a7a3919904bb 100644
--- a/drivers/pci/pcie/aer/aerdrv_core.c
+++ b/drivers/pci/pcie/aer/aerdrv_core.c
@@ -109,7 +109,7 @@ int pci_cleanup_aer_correct_error_status(struct pci_dev *dev)
 #endif  /*  0  */
 
 
-static void set_device_error_reporting(struct pci_dev *dev, void *data)
+static int set_device_error_reporting(struct pci_dev *dev, void *data)
 {
 	bool enable = *((bool *)data);
 
@@ -124,6 +124,8 @@ static void set_device_error_reporting(struct pci_dev *dev, void *data)
 
 	if (enable)
 		pcie_set_ecrc_checking(dev);
+
+	return 0;
 }
 
 /**
@@ -207,7 +209,7 @@ static struct device* find_source_device(struct pci_dev *parent, u16 id)
 	return NULL;
 }
 
-static void report_error_detected(struct pci_dev *dev, void *data)
+static int report_error_detected(struct pci_dev *dev, void *data)
 {
 	pci_ers_result_t vote;
 	struct pci_error_handlers *err_handler;
@@ -232,16 +234,16 @@ static void report_error_detected(struct pci_dev *dev, void *data)
 				   dev->driver ?
 				   "no AER-aware driver" : "no driver");
 		}
-		return;
+		return 0;
 	}
 
 	err_handler = dev->driver->err_handler;
 	vote = err_handler->error_detected(dev, result_data->state);
 	result_data->result = merge_result(result_data->result, vote);
-	return;
+	return 0;
 }
 
-static void report_mmio_enabled(struct pci_dev *dev, void *data)
+static int report_mmio_enabled(struct pci_dev *dev, void *data)
 {
 	pci_ers_result_t vote;
 	struct pci_error_handlers *err_handler;
@@ -251,15 +253,15 @@ static void report_mmio_enabled(struct pci_dev *dev, void *data)
 	if (!dev->driver ||
 		!dev->driver->err_handler ||
 		!dev->driver->err_handler->mmio_enabled)
-		return;
+		return 0;
 
 	err_handler = dev->driver->err_handler;
 	vote = err_handler->mmio_enabled(dev);
 	result_data->result = merge_result(result_data->result, vote);
-	return;
+	return 0;
 }
 
-static void report_slot_reset(struct pci_dev *dev, void *data)
+static int report_slot_reset(struct pci_dev *dev, void *data)
 {
 	pci_ers_result_t vote;
 	struct pci_error_handlers *err_handler;
@@ -269,15 +271,15 @@ static void report_slot_reset(struct pci_dev *dev, void *data)
 	if (!dev->driver ||
 		!dev->driver->err_handler ||
 		!dev->driver->err_handler->slot_reset)
-		return;
+		return 0;
 
 	err_handler = dev->driver->err_handler;
 	vote = err_handler->slot_reset(dev);
 	result_data->result = merge_result(result_data->result, vote);
-	return;
+	return 0;
 }
 
-static void report_resume(struct pci_dev *dev, void *data)
+static int report_resume(struct pci_dev *dev, void *data)
 {
 	struct pci_error_handlers *err_handler;
 
@@ -286,11 +288,11 @@ static void report_resume(struct pci_dev *dev, void *data)
 	if (!dev->driver ||
 		!dev->driver->err_handler ||
 		!dev->driver->err_handler->resume)
-		return;
+		return 0;
 
 	err_handler = dev->driver->err_handler;
 	err_handler->resume(dev);
-	return;
+	return 0;
 }
 
 /**
@@ -307,7 +309,7 @@ static void report_resume(struct pci_dev *dev, void *data)
 static pci_ers_result_t broadcast_error_message(struct pci_dev *dev,
 	enum pci_channel_state state,
 	char *error_mesg,
-	void (*cb)(struct pci_dev *, void *))
+	int (*cb)(struct pci_dev *, void *))
 {
 	struct aer_broadcast_data result_data;
 
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 6a1800ecd95d..61d9b790d21c 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -789,7 +789,7 @@ const struct pci_device_id *pci_match_id(const struct pci_device_id *ids,
 int pci_scan_bridge(struct pci_bus *bus, struct pci_dev *dev, int max,
 		    int pass);
 
-void pci_walk_bus(struct pci_bus *top, void (*cb)(struct pci_dev *, void *),
+void pci_walk_bus(struct pci_bus *top, int (*cb)(struct pci_dev *, void *),
 		  void *userdata);
 int pci_cfg_space_size_ext(struct pci_dev *dev);
 int pci_cfg_space_size(struct pci_dev *dev);
-- 
cgit v1.2.3-59-g8ed1b


From 28eb27cf0839a30948335f9b2edda739f48b7a2e Mon Sep 17 00:00:00 2001
From: "Zhang, Yanmin" <yanmin_zhang@linux.intel.com>
Date: Tue, 16 Jun 2009 13:35:11 +0800
Subject: PCI AER: support invalid error source IDs

When the bus id part of error source id is equal to 0 or nosourceid=1,
make the kernel probe the AER status registers of all devices under the
root port to find the initial error reporter.

Reviewed-by: Andrew Patterson <andrew.patterson@hp.com>
Signed-off-by: Zhang Yanmin <yanmin_zhang@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 Documentation/PCI/pcieaer-howto.txt |   4 +
 drivers/pci/pcie/aer/aerdrv.h       |   2 +
 drivers/pci/pcie/aer/aerdrv_core.c  | 176 ++++++++++++++++++++++++------------
 3 files changed, 122 insertions(+), 60 deletions(-)

diff --git a/Documentation/PCI/pcieaer-howto.txt b/Documentation/PCI/pcieaer-howto.txt
index f6b1ba7464dc..5408b9b39d89 100644
--- a/Documentation/PCI/pcieaer-howto.txt
+++ b/Documentation/PCI/pcieaer-howto.txt
@@ -61,6 +61,10 @@ be initiated although firmwares have no _OSC support. To enable the
 walkaround, pls. add aerdriver.forceload=y to kernel boot parameter line
 when booting kernel. Note that forceload=n by default.
 
+nosourceid, another parameter of type bool, can be used when broken
+hardware (mostly chipsets) has root ports that cannot obtain the reporting
+source ID. nosourceid=n by default.
+
 2.3 AER error output
 When a PCI-E AER error is captured, an error message will be outputed to
 console. If it's a correctable error, it is outputed as a warning.
diff --git a/drivers/pci/pcie/aer/aerdrv.h b/drivers/pci/pcie/aer/aerdrv.h
index 3a69ddefe361..dadf492e9ce9 100644
--- a/drivers/pci/pcie/aer/aerdrv.h
+++ b/drivers/pci/pcie/aer/aerdrv.h
@@ -58,6 +58,8 @@ struct header_log_regs {
 };
 
 struct aer_err_info {
+	struct pci_dev *dev;
+	u16 id;
 	int severity;			/* 0:NONFATAL | 1:FATAL | 2:COR */
 	int flags;
 	unsigned int status;		/* COR/UNCOR Error Status */
diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c
index a7a3919904bb..2750e7b266b4 100644
--- a/drivers/pci/pcie/aer/aerdrv_core.c
+++ b/drivers/pci/pcie/aer/aerdrv_core.c
@@ -26,7 +26,9 @@
 #include "aerdrv.h"
 
 static int forceload;
+static int nosourceid;
 module_param(forceload, bool, 0);
+module_param(nosourceid, bool, 0);
 
 int pci_enable_pcie_error_reporting(struct pci_dev *dev)
 {
@@ -143,34 +145,87 @@ static void set_downstream_devices_error_reporting(struct pci_dev *dev,
 	pci_walk_bus(dev->subordinate, set_device_error_reporting, &enable);
 }
 
-static int find_device_iter(struct device *device, void *data)
+static inline int compare_device_id(struct pci_dev *dev,
+			struct aer_err_info *e_info)
 {
-	struct pci_dev *dev;
-	u16 id = *(unsigned long *)data;
-	u8 secondary, subordinate, d_bus = id >> 8;
+	if (e_info->id == ((dev->bus->number << 8) | dev->devfn)) {
+		/*
+		 * Device ID match
+		 */
+		return 1;
+	}
 
-	if (device->bus == &pci_bus_type) {
-		dev = to_pci_dev(device);
-		if (id == ((dev->bus->number << 8) | dev->devfn)) {
-			/*
-			 * Device ID match
-			 */
-			*(unsigned long*)data = (unsigned long)device;
+	return 0;
+}
+
+#define	PCI_BUS(x)	(((x) >> 8) & 0xff)
+
+static int find_device_iter(struct pci_dev *dev, void *data)
+{
+	int pos;
+	u32 status;
+	u32 mask;
+	u16 reg16;
+	int result;
+	struct aer_err_info *e_info = (struct aer_err_info *)data;
+
+	/*
+	 * When bus id is equal to 0, it might be a bad id
+	 * reported by root port.
+	 */
+	if (!nosourceid && (PCI_BUS(e_info->id) != 0)) {
+		result = compare_device_id(dev, e_info);
+		if (result)
+			e_info->dev = dev;
+		return result;
+	}
+
+	/*
+	 * Next is to check when bus id is equal to 0 or
+	 * nosourceid==y. Some ports might lose the bus
+	 * id of error source id. We check AER status
+	 * registers to find the initial reporter.
+	 */
+	if (atomic_read(&dev->enable_cnt) == 0)
+		return 0;
+	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	if (!pos)
+		return 0;
+	/* Check if AER is enabled */
+	pci_read_config_word(dev, pos+PCI_EXP_DEVCTL, &reg16);
+	if (!(reg16 & (
+		PCI_EXP_DEVCTL_CERE |
+		PCI_EXP_DEVCTL_NFERE |
+		PCI_EXP_DEVCTL_FERE |
+		PCI_EXP_DEVCTL_URRE)))
+		return 0;
+	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR);
+	if (!pos)
+		return 0;
+
+	status = 0;
+	mask = 0;
+	if (e_info->severity == AER_CORRECTABLE) {
+		pci_read_config_dword(dev,
+				pos + PCI_ERR_COR_STATUS,
+				&status);
+		pci_read_config_dword(dev,
+				pos + PCI_ERR_COR_MASK,
+				&mask);
+		if (status & ERR_CORRECTABLE_ERROR_MASK & ~mask) {
+			e_info->dev = dev;
 			return 1;
 		}
-
-		/*
-		 * If device is P2P, check if it is an upstream?
-		 */
-		if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE) {
-			pci_read_config_byte(dev, PCI_SECONDARY_BUS,
-				&secondary);
-			pci_read_config_byte(dev, PCI_SUBORDINATE_BUS,
-				&subordinate);
-			if (d_bus >= secondary && d_bus <= subordinate) {
-				*(unsigned long*)data = (unsigned long)device;
-				return 1;
-			}
+	} else {
+		pci_read_config_dword(dev,
+				pos + PCI_ERR_UNCOR_STATUS,
+				&status);
+		pci_read_config_dword(dev,
+				pos + PCI_ERR_UNCOR_MASK,
+				&mask);
+		if (status & ERR_UNCORRECTABLE_ERROR_MASK & ~mask) {
+			e_info->dev = dev;
+			return 1;
 		}
 	}
 
@@ -180,33 +235,22 @@ static int find_device_iter(struct device *device, void *data)
 /**
  * find_source_device - search through device hierarchy for source device
  * @parent: pointer to Root Port pci_dev data structure
- * @id: device ID of agent who sends an error message to this Root Port
+ * @err_info: including detailed error information such like id
  *
  * Invoked when error is detected at the Root Port.
  */
-static struct device* find_source_device(struct pci_dev *parent, u16 id)
+static void find_source_device(struct pci_dev *parent,
+		struct aer_err_info *e_info)
 {
 	struct pci_dev *dev = parent;
-	struct device *device;
-	unsigned long device_addr;
-	int status;
+	int result;
 
 	/* Is Root Port an agent that sends error message? */
-	if (id == ((dev->bus->number << 8) | dev->devfn))
-		return &dev->dev;
-
-	do {
-		device_addr = id;
- 		if ((status = device_for_each_child(&dev->dev,
-			&device_addr, find_device_iter))) {
-			device = (struct device*)device_addr;
-			dev = to_pci_dev(device);
-			if (id == ((dev->bus->number << 8) | dev->devfn))
-				return device;
-		}
- 	}while (status);
+	result = find_device_iter(dev, e_info);
+	if (result)
+		return;
 
-	return NULL;
+	pci_walk_bus(parent->subordinate, find_device_iter, e_info);
 }
 
 static int report_error_detected(struct pci_dev *dev, void *data)
@@ -501,12 +545,12 @@ static pci_ers_result_t do_recovery(struct pcie_device *aerdev,
  */
 static void handle_error_source(struct pcie_device * aerdev,
 	struct pci_dev *dev,
-	struct aer_err_info info)
+	struct aer_err_info *info)
 {
 	pci_ers_result_t status = 0;
 	int pos;
 
-	if (info.severity == AER_CORRECTABLE) {
+	if (info->severity == AER_CORRECTABLE) {
 		/*
 		 * Correctable error does not need software intevention.
 		 * No need to go through error recovery process.
@@ -514,9 +558,9 @@ static void handle_error_source(struct pcie_device * aerdev,
 		pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR);
 		if (pos)
 			pci_write_config_dword(dev, pos + PCI_ERR_COR_STATUS,
-					info.status);
+					info->status);
 	} else {
-		status = do_recovery(aerdev, dev, info.severity);
+		status = do_recovery(aerdev, dev, info->severity);
 		if (status == PCI_ERS_RESULT_RECOVERED) {
 			dev_printk(KERN_DEBUG, &dev->dev, "AER driver "
 				   "successfully recovered\n");
@@ -673,10 +717,16 @@ static int get_device_error_info(struct pci_dev *dev, struct aer_err_info *info)
 static void aer_isr_one_error(struct pcie_device *p_device,
 		struct aer_err_source *e_src)
 {
-	struct device *s_device;
-	struct aer_err_info e_info = {0, 0, 0,};
+	struct aer_err_info *e_info;
 	int i;
-	u16 id;
+
+	/* struct aer_err_info might be big, so we allocate it with slab */
+	e_info = kmalloc(sizeof(struct aer_err_info), GFP_KERNEL);
+	if (e_info == NULL) {
+		dev_printk(KERN_DEBUG, &p_device->port->dev,
+			"Can't allocate mem when processing AER errors\n");
+		return;
+	}
 
 	/*
 	 * There is a possibility that both correctable error and
@@ -688,31 +738,37 @@ static void aer_isr_one_error(struct pcie_device *p_device,
 		if (!(e_src->status & i))
 			continue;
 
+		memset(e_info, 0, sizeof(struct aer_err_info));
+
 		/* Init comprehensive error information */
 		if (i & PCI_ERR_ROOT_COR_RCV) {
-			id = ERR_COR_ID(e_src->id);
-			e_info.severity = AER_CORRECTABLE;
+			e_info->id = ERR_COR_ID(e_src->id);
+			e_info->severity = AER_CORRECTABLE;
 		} else {
-			id = ERR_UNCOR_ID(e_src->id);
-			e_info.severity = ((e_src->status >> 6) & 1);
+			e_info->id = ERR_UNCOR_ID(e_src->id);
+			e_info->severity = ((e_src->status >> 6) & 1);
 		}
 		if (e_src->status &
 			(PCI_ERR_ROOT_MULTI_COR_RCV |
 			 PCI_ERR_ROOT_MULTI_UNCOR_RCV))
-			e_info.flags |= AER_MULTI_ERROR_VALID_FLAG;
-		if (!(s_device = find_source_device(p_device->port, id))) {
+			e_info->flags |= AER_MULTI_ERROR_VALID_FLAG;
+
+		find_source_device(p_device->port, e_info);
+		if (e_info->dev == NULL) {
 			printk(KERN_DEBUG "%s->can't find device of ID%04x\n",
-				__func__, id);
+				__func__, e_info->id);
 			continue;
 		}
-		if (get_device_error_info(to_pci_dev(s_device), &e_info) ==
+		if (get_device_error_info(e_info->dev, e_info) ==
 				AER_SUCCESS) {
-			aer_print_error(to_pci_dev(s_device), &e_info);
+			aer_print_error(e_info->dev, e_info);
 			handle_error_source(p_device,
-				to_pci_dev(s_device),
+				e_info->dev,
 				e_info);
 		}
 	}
+
+	kfree(e_info);
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 3d5505c56db5c8d1eeca45c325b19e95115afdea Mon Sep 17 00:00:00 2001
From: "Zhang, Yanmin" <yanmin_zhang@linux.intel.com>
Date: Tue, 16 Jun 2009 13:35:16 +0800
Subject: PCI AER: multiple error support

When a root port receives the same errors more than once before the
kernel process them, the Multiple Error Messages Received flags are set
by hardware. Because the root port could only save one kind of
correctable error source id and another uncorrectable error source id at
the same time, the second message sender id is lost if the 2 messages
are sent from 2 different devices. This patch makes the kernel search
all devices under the root port when multiple messages are received.

Reviewed-by: Andrew Patterson <andrew.patterson@hp.com>
Signed-off-by: Zhang Yanmin <yanmin_zhang@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aer/aerdrv.h      |  4 +-
 drivers/pci/pcie/aer/aerdrv_core.c | 88 ++++++++++++++++++++++++++++----------
 2 files changed, 69 insertions(+), 23 deletions(-)

diff --git a/drivers/pci/pcie/aer/aerdrv.h b/drivers/pci/pcie/aer/aerdrv.h
index dadf492e9ce9..bbd7428ca2d0 100644
--- a/drivers/pci/pcie/aer/aerdrv.h
+++ b/drivers/pci/pcie/aer/aerdrv.h
@@ -57,8 +57,10 @@ struct header_log_regs {
 	unsigned int dw3;
 };
 
+#define AER_MAX_MULTI_ERR_DEVICES	5	/* Not likely to have more */
 struct aer_err_info {
-	struct pci_dev *dev;
+	struct pci_dev *dev[AER_MAX_MULTI_ERR_DEVICES];
+	int error_dev_num;
 	u16 id;
 	int severity;			/* 0:NONFATAL | 1:FATAL | 2:COR */
 	int flags;
diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c
index 2750e7b266b4..3d8872704a58 100644
--- a/drivers/pci/pcie/aer/aerdrv_core.c
+++ b/drivers/pci/pcie/aer/aerdrv_core.c
@@ -158,6 +158,17 @@ static inline int compare_device_id(struct pci_dev *dev,
 	return 0;
 }
 
+static int add_error_device(struct aer_err_info *e_info, struct pci_dev *dev)
+{
+	if (e_info->error_dev_num < AER_MAX_MULTI_ERR_DEVICES) {
+		e_info->dev[e_info->error_dev_num] = dev;
+		e_info->error_dev_num++;
+		return 1;
+	} else
+		return 0;
+}
+
+
 #define	PCI_BUS(x)	(((x) >> 8) & 0xff)
 
 static int find_device_iter(struct pci_dev *dev, void *data)
@@ -176,15 +187,31 @@ static int find_device_iter(struct pci_dev *dev, void *data)
 	if (!nosourceid && (PCI_BUS(e_info->id) != 0)) {
 		result = compare_device_id(dev, e_info);
 		if (result)
-			e_info->dev = dev;
-		return result;
+			add_error_device(e_info, dev);
+
+		/*
+		 * If there is no multiple error, we stop
+		 * or continue based on the id comparing.
+		 */
+		if (!(e_info->flags & AER_MULTI_ERROR_VALID_FLAG))
+			return result;
+
+		/*
+		 * If there are multiple errors and id does match,
+		 * We need continue to search other devices under
+		 * the root port. Return 0 means that.
+		 */
+		if (result)
+			return 0;
 	}
 
 	/*
-	 * Next is to check when bus id is equal to 0 or
-	 * nosourceid==y. Some ports might lose the bus
-	 * id of error source id. We check AER status
-	 * registers to find the initial reporter.
+	 * When either
+	 *      1) nosourceid==y;
+	 *      2) bus id is equal to 0. Some ports might lose the bus
+	 *              id of error source id;
+	 *      3) There are multiple errors and prior id comparing fails;
+	 * We check AER status registers to find the initial reporter.
 	 */
 	if (atomic_read(&dev->enable_cnt) == 0)
 		return 0;
@@ -213,8 +240,8 @@ static int find_device_iter(struct pci_dev *dev, void *data)
 				pos + PCI_ERR_COR_MASK,
 				&mask);
 		if (status & ERR_CORRECTABLE_ERROR_MASK & ~mask) {
-			e_info->dev = dev;
-			return 1;
+			add_error_device(e_info, dev);
+			goto added;
 		}
 	} else {
 		pci_read_config_dword(dev,
@@ -224,12 +251,18 @@ static int find_device_iter(struct pci_dev *dev, void *data)
 				pos + PCI_ERR_UNCOR_MASK,
 				&mask);
 		if (status & ERR_UNCORRECTABLE_ERROR_MASK & ~mask) {
-			e_info->dev = dev;
-			return 1;
+			add_error_device(e_info, dev);
+			goto added;
 		}
 	}
 
 	return 0;
+
+added:
+	if (e_info->flags & AER_MULTI_ERROR_VALID_FLAG)
+		return 0;
+	else
+		return 1;
 }
 
 /**
@@ -709,6 +742,28 @@ static int get_device_error_info(struct pci_dev *dev, struct aer_err_info *info)
 	return AER_SUCCESS;
 }
 
+static inline void aer_process_err_devices(struct pcie_device *p_device,
+			struct aer_err_info *e_info)
+{
+	int i;
+
+	if (!e_info->dev[0]) {
+		dev_printk(KERN_DEBUG, &p_device->port->dev,
+				"can't find device of ID%04x\n",
+				e_info->id);
+	}
+
+	for (i = 0; i < e_info->error_dev_num && e_info->dev[i]; i++) {
+		if (get_device_error_info(e_info->dev[i], e_info) ==
+				AER_SUCCESS) {
+			aer_print_error(e_info->dev[i], e_info);
+			handle_error_source(p_device,
+					e_info->dev[i],
+					e_info);
+		}
+	}
+}
+
 /**
  * aer_isr_one_error - consume an error detected by root port
  * @p_device: pointer to error root port service device
@@ -754,18 +809,7 @@ static void aer_isr_one_error(struct pcie_device *p_device,
 			e_info->flags |= AER_MULTI_ERROR_VALID_FLAG;
 
 		find_source_device(p_device->port, e_info);
-		if (e_info->dev == NULL) {
-			printk(KERN_DEBUG "%s->can't find device of ID%04x\n",
-				__func__, e_info->id);
-			continue;
-		}
-		if (get_device_error_info(e_info->dev, e_info) ==
-				AER_SUCCESS) {
-			aer_print_error(e_info->dev, e_info);
-			handle_error_source(p_device,
-				e_info->dev,
-				e_info);
-		}
+		aer_process_err_devices(p_device, e_info);
 	}
 
 	kfree(e_info);
-- 
cgit v1.2.3-59-g8ed1b


From c465def6bfe834b62623caa9b98f2d4f4739875a Mon Sep 17 00:00:00 2001
From: Huang Ying <ying.huang@intel.com>
Date: Mon, 15 Jun 2009 10:42:57 +0800
Subject: PCI AER: software error injection

Debugging PCIE AER code can be very difficult because it is hard
to trigger various real hardware errors. This patch provide a
software based error injection tool, which can fake various PCIE
errors with a user space helper tool named "aer-inject". Which
can be gotten from:

  http://www.kernel.org/pub/linux/kernel/people/yhuang/

The patch fakes AER error by faking some PCIE AER related
registers and an AER interrupt for specified the PCIE device.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 Documentation/PCI/pcieaer-howto.txt |   2 +-
 drivers/pci/pcie/aer/Kconfig        |   2 +
 drivers/pci/pcie/aer/Kconfig.debug  |  18 ++
 drivers/pci/pcie/aer/Makefile       |   1 +
 drivers/pci/pcie/aer/aer_inject.c   | 473 ++++++++++++++++++++++++++++++++++++
 5 files changed, 495 insertions(+), 1 deletion(-)
 create mode 100644 drivers/pci/pcie/aer/Kconfig.debug
 create mode 100644 drivers/pci/pcie/aer/aer_inject.c

diff --git a/Documentation/PCI/pcieaer-howto.txt b/Documentation/PCI/pcieaer-howto.txt
index 5408b9b39d89..be21001ab144 100644
--- a/Documentation/PCI/pcieaer-howto.txt
+++ b/Documentation/PCI/pcieaer-howto.txt
@@ -267,7 +267,7 @@ After reboot with new kernel or insert the module, a device file named
 
 Then, you need a user space tool named aer-inject, which can be gotten
 from:
-    http://www.kernel.org/pub/linux/kernel/people/yhuang/
+    http://www.kernel.org/pub/linux/utils/pci/aer-inject/
 
 More information about aer-inject can be found in the document comes
 with its source code.
diff --git a/drivers/pci/pcie/aer/Kconfig b/drivers/pci/pcie/aer/Kconfig
index db4cb950933a..50e94e02378a 100644
--- a/drivers/pci/pcie/aer/Kconfig
+++ b/drivers/pci/pcie/aer/Kconfig
@@ -23,3 +23,5 @@ config PCIE_ECRC
 	  (transaction layer end-to-end CRC checking).
 
 	  When in doubt, say N.
+
+source "drivers/pci/pcie/aer/Kconfig.debug"
diff --git a/drivers/pci/pcie/aer/Kconfig.debug b/drivers/pci/pcie/aer/Kconfig.debug
new file mode 100644
index 000000000000..b8c925c1f6aa
--- /dev/null
+++ b/drivers/pci/pcie/aer/Kconfig.debug
@@ -0,0 +1,18 @@
+#
+# PCI Express Root Port Device AER Debug Configuration
+#
+
+config PCIEAER_INJECT
+	tristate "PCIE AER error injector support"
+	depends on PCIEAER
+	default n
+	help
+	  This enables PCI Express Root Port Advanced Error Reporting
+	  (AER) software error injector.
+
+	  Debuging PCIE AER code is quite difficult because it is hard
+	  to trigger various real hardware errors. Software based
+	  error injection can fake almost all kinds of errors with the
+	  help of a user space helper tool aer-inject, which can be
+	  gotten from:
+	     http://www.kernel.org/pub/linux/utils/pci/aer-inject/
diff --git a/drivers/pci/pcie/aer/Makefile b/drivers/pci/pcie/aer/Makefile
index 7f93411c56e5..2cba67510dc8 100644
--- a/drivers/pci/pcie/aer/Makefile
+++ b/drivers/pci/pcie/aer/Makefile
@@ -9,3 +9,4 @@ obj-$(CONFIG_PCIE_ECRC)	+= ecrc.o
 aerdriver-objs := aerdrv_errprint.o aerdrv_core.o aerdrv.o
 aerdriver-$(CONFIG_ACPI) += aerdrv_acpi.o
 
+obj-$(CONFIG_PCIEAER_INJECT) += aer_inject.o
diff --git a/drivers/pci/pcie/aer/aer_inject.c b/drivers/pci/pcie/aer/aer_inject.c
new file mode 100644
index 000000000000..d92ae21a59d8
--- /dev/null
+++ b/drivers/pci/pcie/aer/aer_inject.c
@@ -0,0 +1,473 @@
+/*
+ * PCIE AER software error injection support.
+ *
+ * Debuging PCIE AER code is quite difficult because it is hard to
+ * trigger various real hardware errors. Software based error
+ * injection can fake almost all kinds of errors with the help of a
+ * user space helper tool aer-inject, which can be gotten from:
+ *   http://www.kernel.org/pub/linux/utils/pci/aer-inject/
+ *
+ * Copyright 2009 Intel Corporation.
+ *     Huang Ying <ying.huang@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/miscdevice.h>
+#include <linux/pci.h>
+#include <linux/fs.h>
+#include <asm/uaccess.h>
+#include "aerdrv.h"
+
+struct aer_error_inj
+{
+	u8 bus;
+	u8 dev;
+	u8 fn;
+	u32 uncor_status;
+	u32 cor_status;
+	u32 header_log0;
+	u32 header_log1;
+	u32 header_log2;
+	u32 header_log3;
+};
+
+struct aer_error
+{
+	struct list_head list;
+	unsigned int bus;
+	unsigned int devfn;
+	int pos_cap_err;
+
+	u32 uncor_status;
+	u32 cor_status;
+	u32 header_log0;
+	u32 header_log1;
+	u32 header_log2;
+	u32 header_log3;
+	u32 root_status;
+	u32 source_id;
+};
+
+struct pci_bus_ops
+{
+	struct list_head list;
+	struct pci_bus *bus;
+	struct pci_ops *ops;
+};
+
+static LIST_HEAD(einjected);
+
+static LIST_HEAD(pci_bus_ops_list);
+
+/* Protect einjected and pci_bus_ops_list */
+static DEFINE_SPINLOCK(inject_lock);
+
+static void aer_error_init(struct aer_error *err, unsigned int bus,
+			   unsigned int devfn, int pos_cap_err)
+{
+	INIT_LIST_HEAD(&err->list);
+	err->bus = bus;
+	err->devfn = devfn;
+	err->pos_cap_err = pos_cap_err;
+}
+
+/* inject_lock must be held before calling */
+static struct aer_error *__find_aer_error(unsigned int bus, unsigned int devfn)
+{
+	struct aer_error *err;
+
+	list_for_each_entry(err, &einjected, list) {
+		if (bus == err->bus && devfn == err->devfn)
+			return err;
+	}
+	return NULL;
+}
+
+/* inject_lock must be held before calling */
+static struct aer_error *__find_aer_error_by_dev(struct pci_dev *dev)
+{
+	return __find_aer_error(dev->bus->number, dev->devfn);
+}
+
+/* inject_lock must be held before calling */
+static struct pci_ops *__find_pci_bus_ops(struct pci_bus *bus)
+{
+	struct pci_bus_ops *bus_ops;
+
+	list_for_each_entry(bus_ops, &pci_bus_ops_list, list) {
+		if (bus_ops->bus == bus)
+			return bus_ops->ops;
+	}
+	return NULL;
+}
+
+static struct pci_bus_ops *pci_bus_ops_pop(void)
+{
+	unsigned long flags;
+	struct pci_bus_ops *bus_ops = NULL;
+
+	spin_lock_irqsave(&inject_lock, flags);
+	if (list_empty(&pci_bus_ops_list))
+		bus_ops = NULL;
+	else {
+		struct list_head *lh = pci_bus_ops_list.next;
+		list_del(lh);
+		bus_ops = list_entry(lh, struct pci_bus_ops, list);
+	}
+	spin_unlock_irqrestore(&inject_lock, flags);
+	return bus_ops;
+}
+
+static u32 *find_pci_config_dword(struct aer_error *err, int where,
+				  int *prw1cs)
+{
+	int rw1cs = 0;
+	u32 *target = NULL;
+
+	if (err->pos_cap_err == -1)
+		return NULL;
+
+	switch (where - err->pos_cap_err) {
+	case PCI_ERR_UNCOR_STATUS:
+		target = &err->uncor_status;
+		rw1cs = 1;
+		break;
+	case PCI_ERR_COR_STATUS:
+		target = &err->cor_status;
+		rw1cs = 1;
+		break;
+	case PCI_ERR_HEADER_LOG:
+		target = &err->header_log0;
+		break;
+	case PCI_ERR_HEADER_LOG+4:
+		target = &err->header_log1;
+		break;
+	case PCI_ERR_HEADER_LOG+8:
+	        target = &err->header_log2;
+		break;
+	case PCI_ERR_HEADER_LOG+12:
+		target = &err->header_log3;
+		break;
+	case PCI_ERR_ROOT_STATUS:
+		target = &err->root_status;
+		rw1cs = 1;
+		break;
+	case PCI_ERR_ROOT_COR_SRC:
+		target = &err->source_id;
+		break;
+	}
+	if (prw1cs)
+		*prw1cs = rw1cs;
+	return target;
+}
+
+static int pci_read_aer(struct pci_bus *bus, unsigned int devfn, int where,
+			int size, u32 *val)
+{
+	u32 *sim;
+	struct aer_error *err;
+	unsigned long flags;
+	struct pci_ops *ops;
+
+	spin_lock_irqsave(&inject_lock, flags);
+	if (size != sizeof(u32))
+		goto out;
+	err = __find_aer_error(bus->number, devfn);
+	if (!err)
+		goto out;
+
+	sim = find_pci_config_dword(err, where, NULL);
+	if (sim) {
+		*val = *sim;
+		spin_unlock_irqrestore(&inject_lock, flags);
+		return 0;
+	}
+out:
+	ops = __find_pci_bus_ops(bus);
+	spin_unlock_irqrestore(&inject_lock, flags);
+	return ops->read(bus, devfn, where, size, val);
+}
+
+int pci_write_aer(struct pci_bus *bus, unsigned int devfn, int where, int size,
+		  u32 val)
+{
+	u32 *sim;
+	struct aer_error *err;
+	unsigned long flags;
+	int rw1cs;
+	struct pci_ops *ops;
+
+	spin_lock_irqsave(&inject_lock, flags);
+	if (size != sizeof(u32))
+		goto out;
+	err = __find_aer_error(bus->number, devfn);
+	if (!err)
+		goto out;
+
+	sim = find_pci_config_dword(err, where, &rw1cs);
+	if (sim) {
+		if (rw1cs)
+			*sim ^= val;
+		else
+			*sim = val;
+		spin_unlock_irqrestore(&inject_lock, flags);
+		return 0;
+	}
+out:
+	ops = __find_pci_bus_ops(bus);
+	spin_unlock_irqrestore(&inject_lock, flags);
+	return ops->write(bus, devfn, where, size, val);
+}
+
+static struct pci_ops pci_ops_aer = {
+	.read = pci_read_aer,
+	.write = pci_write_aer,
+};
+
+static void pci_bus_ops_init(struct pci_bus_ops *bus_ops,
+			     struct pci_bus *bus,
+			     struct pci_ops *ops)
+{
+	INIT_LIST_HEAD(&bus_ops->list);
+	bus_ops->bus = bus;
+	bus_ops->ops = ops;
+}
+
+static int pci_bus_set_aer_ops(struct pci_bus *bus)
+{
+	struct pci_ops *ops;
+	struct pci_bus_ops *bus_ops;
+	unsigned long flags;
+
+	bus_ops = kmalloc(sizeof(*bus_ops), GFP_KERNEL);
+	if (!bus_ops)
+		return -ENOMEM;
+	ops = pci_bus_set_ops(bus, &pci_ops_aer);
+	spin_lock_irqsave(&inject_lock, flags);
+	if (ops == &pci_ops_aer)
+		goto out;
+	pci_bus_ops_init(bus_ops, bus, ops);
+	list_add(&bus_ops->list, &pci_bus_ops_list);
+	bus_ops = NULL;
+out:
+	spin_unlock_irqrestore(&inject_lock, flags);
+	if (bus_ops)
+		kfree(bus_ops);
+	return 0;
+}
+
+static struct pci_dev *pcie_find_root_port(struct pci_dev *dev)
+{
+	while (1) {
+		if (!dev->is_pcie)
+			break;
+		if (dev->pcie_type == PCI_EXP_TYPE_ROOT_PORT)
+			return dev;
+		if (!dev->bus->self)
+			break;
+		dev = dev->bus->self;
+	}
+	return NULL;
+}
+
+static int find_aer_device_iter(struct device *device, void *data)
+{
+	struct pcie_device **result = data;
+	struct pcie_device *pcie_dev;
+
+	if (device->bus == &pcie_port_bus_type) {
+		pcie_dev = to_pcie_device(device);
+		if (pcie_dev->service & PCIE_PORT_SERVICE_AER) {
+			*result = pcie_dev;
+			return 1;
+		}
+	}
+	return 0;
+}
+
+static int find_aer_device(struct pci_dev *dev, struct pcie_device **result)
+{
+	return device_for_each_child(&dev->dev, result, find_aer_device_iter);
+}
+
+static int aer_inject(struct aer_error_inj *einj)
+{
+	struct aer_error *err, *rperr;
+	struct aer_error *err_alloc = NULL, *rperr_alloc = NULL;
+	struct pci_dev *dev, *rpdev;
+	struct pcie_device *edev;
+	unsigned long flags;
+	unsigned int devfn = PCI_DEVFN(einj->dev, einj->fn);
+	int pos_cap_err, rp_pos_cap_err;
+	u32 sever;
+	int ret = 0;
+
+	dev = pci_get_bus_and_slot(einj->bus, devfn);
+	if (!dev)
+		return -EINVAL;
+	rpdev = pcie_find_root_port(dev);
+	if (!rpdev) {
+		ret = -EINVAL;
+		goto out_put;
+	}
+
+	pos_cap_err = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR);
+	if (!pos_cap_err) {
+		ret = -EIO;
+		goto out_put;
+	}
+	pci_read_config_dword(dev, pos_cap_err + PCI_ERR_UNCOR_SEVER, &sever);
+
+	rp_pos_cap_err = pci_find_ext_capability(rpdev, PCI_EXT_CAP_ID_ERR);
+	if (!rp_pos_cap_err) {
+		ret = -EIO;
+		goto out_put;
+	}
+
+	err_alloc =  kzalloc(sizeof(struct aer_error), GFP_KERNEL);
+	if (!err_alloc) {
+		ret = -ENOMEM;
+		goto out_put;
+	}
+	rperr_alloc =  kzalloc(sizeof(struct aer_error), GFP_KERNEL);
+	if (!rperr_alloc) {
+		ret = -ENOMEM;
+		goto out_put;
+	}
+
+	spin_lock_irqsave(&inject_lock, flags);
+
+	err = __find_aer_error_by_dev(dev);
+	if (!err) {
+		err = err_alloc;
+		err_alloc = NULL;
+		aer_error_init(err, einj->bus, devfn, pos_cap_err);
+		list_add(&err->list, &einjected);
+	}
+	err->uncor_status |= einj->uncor_status;
+	err->cor_status |= einj->cor_status;
+	err->header_log0 = einj->header_log0;
+	err->header_log1 = einj->header_log1;
+	err->header_log2 = einj->header_log2;
+	err->header_log3 = einj->header_log3;
+
+	rperr = __find_aer_error_by_dev(rpdev);
+	if (!rperr) {
+		rperr = rperr_alloc;
+		rperr_alloc = NULL;
+		aer_error_init(rperr, rpdev->bus->number, rpdev->devfn,
+			       rp_pos_cap_err);
+		list_add(&rperr->list, &einjected);
+	}
+	if (einj->cor_status) {
+		if (rperr->root_status & PCI_ERR_ROOT_COR_RCV)
+			rperr->root_status |= PCI_ERR_ROOT_MULTI_COR_RCV;
+		else
+			rperr->root_status |= PCI_ERR_ROOT_COR_RCV;
+		rperr->source_id &= 0xffff0000;
+		rperr->source_id |= (einj->bus << 8) | devfn;
+	}
+	if (einj->uncor_status) {
+		if (rperr->root_status & PCI_ERR_ROOT_UNCOR_RCV)
+			rperr->root_status |= PCI_ERR_ROOT_MULTI_UNCOR_RCV;
+		if (sever & einj->uncor_status) {
+			rperr->root_status |= PCI_ERR_ROOT_FATAL_RCV;
+			if (!(rperr->root_status & PCI_ERR_ROOT_UNCOR_RCV))
+				rperr->root_status |= PCI_ERR_ROOT_FIRST_FATAL;
+		} else
+			rperr->root_status |= PCI_ERR_ROOT_NONFATAL_RCV;
+		rperr->root_status |= PCI_ERR_ROOT_UNCOR_RCV;
+		rperr->source_id &= 0x0000ffff;
+		rperr->source_id |= ((einj->bus << 8) | devfn) << 16;
+	}
+	spin_unlock_irqrestore(&inject_lock, flags);
+
+	ret = pci_bus_set_aer_ops(dev->bus);
+	if (ret)
+		goto out_put;
+	ret = pci_bus_set_aer_ops(rpdev->bus);
+	if (ret)
+		goto out_put;
+
+	if (find_aer_device(rpdev, &edev))
+		aer_irq(-1, edev);
+	else
+		ret = -EINVAL;
+out_put:
+	if (err_alloc)
+		kfree(err_alloc);
+	if (rperr_alloc)
+		kfree(rperr_alloc);
+	pci_dev_put(dev);
+	return ret;
+}
+
+static ssize_t aer_inject_write(struct file *filp, const char __user *ubuf,
+				size_t usize, loff_t *off)
+{
+	struct aer_error_inj einj;
+	int ret;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	if (usize != sizeof(struct aer_error_inj))
+		return -EINVAL;
+
+	if (copy_from_user(&einj, ubuf, usize))
+		return -EFAULT;
+
+	ret = aer_inject(&einj);
+	return ret ? ret : usize;
+}
+
+static const struct file_operations aer_inject_fops = {
+	.write = aer_inject_write,
+	.owner = THIS_MODULE,
+};
+
+static struct miscdevice aer_inject_device = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "aer_inject",
+	.fops = &aer_inject_fops,
+};
+
+static int __init aer_inject_init(void)
+{
+	return misc_register(&aer_inject_device);
+}
+
+static void __exit aer_inject_exit(void)
+{
+	struct aer_error *err, *err_next;
+	unsigned long flags;
+	struct pci_bus_ops *bus_ops;
+
+	misc_deregister(&aer_inject_device);
+
+	while ((bus_ops = pci_bus_ops_pop())) {
+		pci_bus_set_ops(bus_ops->bus, bus_ops->ops);
+		kfree(bus_ops);
+	}
+
+	spin_lock_irqsave(&inject_lock, flags);
+	list_for_each_entry_safe(err, err_next,
+				 &pci_bus_ops_list, list) {
+		list_del(&err->list);
+		kfree(err);
+	}
+	spin_unlock_irqrestore(&inject_lock, flags);
+}
+
+module_init(aer_inject_init);
+module_exit(aer_inject_exit);
+
+MODULE_DESCRIPTION("PCIE AER software error injector");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-59-g8ed1b


From 8c1c699fec9e9021bf6ff0285dee086bb27aec90 Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Sat, 13 Jun 2009 15:52:13 +0800
Subject: PCI: cleanup Function Level Reset

This patch enhances the FLR functions:
  1) remove disable_irq() so the shared IRQ won't be disabled.
  2) replace the 1s wait with 100, 200 and 400ms wait intervals
     for the Pending Transaction.
  3) replace mdelay() with msleep().
  4) add might_sleep().
  5) lock the device to prevent PM suspend from accessing the CSRs
     during the reset.
  6) coding style fixes.

Reviewed-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/iov.c   |   4 +-
 drivers/pci/pci.c   | 166 ++++++++++++++++++++++++++--------------------------
 include/linux/pci.h |   2 +-
 3 files changed, 87 insertions(+), 85 deletions(-)

diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c
index e87fe95da814..03c7706c0a09 100644
--- a/drivers/pci/iov.c
+++ b/drivers/pci/iov.c
@@ -110,7 +110,7 @@ static int virtfn_add(struct pci_dev *dev, int id, int reset)
 	}
 
 	if (reset)
-		pci_execute_reset_function(virtfn);
+		__pci_reset_function(virtfn);
 
 	pci_device_add(virtfn, virtfn->bus);
 	mutex_unlock(&iov->dev->sriov->lock);
@@ -164,7 +164,7 @@ static void virtfn_remove(struct pci_dev *dev, int id, int reset)
 
 	if (reset) {
 		device_release_driver(&virtfn->dev);
-		pci_execute_reset_function(virtfn);
+		__pci_reset_function(virtfn);
 	}
 
 	sprintf(buf, "virtfn%u", id);
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 8ea911e55722..6a052ada3fe8 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -2055,111 +2055,112 @@ int pci_set_dma_seg_boundary(struct pci_dev *dev, unsigned long mask)
 EXPORT_SYMBOL(pci_set_dma_seg_boundary);
 #endif
 
-static int __pcie_flr(struct pci_dev *dev, int probe)
+static int pcie_flr(struct pci_dev *dev, int probe)
 {
-	u16 status;
+	int i;
+	int pos;
 	u32 cap;
-	int exppos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	u16 status;
 
-	if (!exppos)
+	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	if (!pos)
 		return -ENOTTY;
-	pci_read_config_dword(dev, exppos + PCI_EXP_DEVCAP, &cap);
+
+	pci_read_config_dword(dev, pos + PCI_EXP_DEVCAP, &cap);
 	if (!(cap & PCI_EXP_DEVCAP_FLR))
 		return -ENOTTY;
 
 	if (probe)
 		return 0;
 
-	pci_block_user_cfg_access(dev);
-
 	/* Wait for Transaction Pending bit clean */
-	pci_read_config_word(dev, exppos + PCI_EXP_DEVSTA, &status);
-	if (!(status & PCI_EXP_DEVSTA_TRPND))
-		goto transaction_done;
+	for (i = 0; i < 4; i++) {
+		if (i)
+			msleep((1 << (i - 1)) * 100);
 
-	msleep(100);
-	pci_read_config_word(dev, exppos + PCI_EXP_DEVSTA, &status);
-	if (!(status & PCI_EXP_DEVSTA_TRPND))
-		goto transaction_done;
-
-	dev_info(&dev->dev, "Busy after 100ms while trying to reset; "
-			"sleeping for 1 second\n");
-	ssleep(1);
-	pci_read_config_word(dev, exppos + PCI_EXP_DEVSTA, &status);
-	if (status & PCI_EXP_DEVSTA_TRPND)
-		dev_info(&dev->dev, "Still busy after 1s; "
-				"proceeding with reset anyway\n");
-
-transaction_done:
-	pci_write_config_word(dev, exppos + PCI_EXP_DEVCTL,
+		pci_read_config_word(dev, pos + PCI_EXP_DEVSTA, &status);
+		if (!(status & PCI_EXP_DEVSTA_TRPND))
+			goto clear;
+	}
+
+	dev_err(&dev->dev, "transaction is not cleared; "
+			"proceeding with reset anyway\n");
+
+clear:
+	pci_write_config_word(dev, pos + PCI_EXP_DEVCTL,
 				PCI_EXP_DEVCTL_BCR_FLR);
-	mdelay(100);
+	msleep(100);
 
-	pci_unblock_user_cfg_access(dev);
 	return 0;
 }
 
-static int __pci_af_flr(struct pci_dev *dev, int probe)
+static int pci_af_flr(struct pci_dev *dev, int probe)
 {
-	int cappos = pci_find_capability(dev, PCI_CAP_ID_AF);
-	u8 status;
+	int i;
+	int pos;
 	u8 cap;
+	u8 status;
 
-	if (!cappos)
+	pos = pci_find_capability(dev, PCI_CAP_ID_AF);
+	if (!pos)
 		return -ENOTTY;
-	pci_read_config_byte(dev, cappos + PCI_AF_CAP, &cap);
+
+	pci_read_config_byte(dev, pos + PCI_AF_CAP, &cap);
 	if (!(cap & PCI_AF_CAP_TP) || !(cap & PCI_AF_CAP_FLR))
 		return -ENOTTY;
 
 	if (probe)
 		return 0;
 
-	pci_block_user_cfg_access(dev);
-
 	/* Wait for Transaction Pending bit clean */
-	pci_read_config_byte(dev, cappos + PCI_AF_STATUS, &status);
-	if (!(status & PCI_AF_STATUS_TP))
-		goto transaction_done;
+	for (i = 0; i < 4; i++) {
+		if (i)
+			msleep((1 << (i - 1)) * 100);
+
+		pci_read_config_byte(dev, pos + PCI_AF_STATUS, &status);
+		if (!(status & PCI_AF_STATUS_TP))
+			goto clear;
+	}
+
+	dev_err(&dev->dev, "transaction is not cleared; "
+			"proceeding with reset anyway\n");
 
+clear:
+	pci_write_config_byte(dev, pos + PCI_AF_CTRL, PCI_AF_CTRL_FLR);
 	msleep(100);
-	pci_read_config_byte(dev, cappos + PCI_AF_STATUS, &status);
-	if (!(status & PCI_AF_STATUS_TP))
-		goto transaction_done;
-
-	dev_info(&dev->dev, "Busy after 100ms while trying to"
-			" reset; sleeping for 1 second\n");
-	ssleep(1);
-	pci_read_config_byte(dev, cappos + PCI_AF_STATUS, &status);
-	if (status & PCI_AF_STATUS_TP)
-		dev_info(&dev->dev, "Still busy after 1s; "
-				"proceeding with reset anyway\n");
-
-transaction_done:
-	pci_write_config_byte(dev, cappos + PCI_AF_CTRL, PCI_AF_CTRL_FLR);
-	mdelay(100);
-
-	pci_unblock_user_cfg_access(dev);
+
 	return 0;
 }
 
-static int __pci_reset_function(struct pci_dev *pdev, int probe)
+static int pci_dev_reset(struct pci_dev *dev, int probe)
 {
-	int res;
+	int rc;
+
+	might_sleep();
+
+	if (!probe) {
+		pci_block_user_cfg_access(dev);
+		/* block PM suspend, driver probe, etc. */
+		down(&dev->dev.sem);
+	}
 
-	res = __pcie_flr(pdev, probe);
-	if (res != -ENOTTY)
-		return res;
+	rc = pcie_flr(dev, probe);
+	if (rc != -ENOTTY)
+		goto done;
 
-	res = __pci_af_flr(pdev, probe);
-	if (res != -ENOTTY)
-		return res;
+	rc = pci_af_flr(dev, probe);
+done:
+	if (!probe) {
+		up(&dev->dev.sem);
+		pci_unblock_user_cfg_access(dev);
+	}
 
-	return res;
+	return rc;
 }
 
 /**
- * pci_execute_reset_function() - Reset a PCI device function
- * @dev: Device function to reset
+ * __pci_reset_function - reset a PCI device function
+ * @dev: PCI device to reset
  *
  * Some devices allow an individual function to be reset without affecting
  * other functions in the same device.  The PCI device must be responsive
@@ -2171,18 +2172,18 @@ static int __pci_reset_function(struct pci_dev *pdev, int probe)
  * device including MSI, bus mastering, BARs, decoding IO and memory spaces,
  * etc.
  *
- * Returns 0 if the device function was successfully reset or -ENOTTY if the
+ * Returns 0 if the device function was successfully reset or negative if the
  * device doesn't support resetting a single function.
  */
-int pci_execute_reset_function(struct pci_dev *dev)
+int __pci_reset_function(struct pci_dev *dev)
 {
-	return __pci_reset_function(dev, 0);
+	return pci_dev_reset(dev, 0);
 }
-EXPORT_SYMBOL_GPL(pci_execute_reset_function);
+EXPORT_SYMBOL_GPL(__pci_reset_function);
 
 /**
- * pci_reset_function() - quiesce and reset a PCI device function
- * @dev: Device function to reset
+ * pci_reset_function - quiesce and reset a PCI device function
+ * @dev: PCI device to reset
  *
  * Some devices allow an individual function to be reset without affecting
  * other functions in the same device.  The PCI device must be responsive
@@ -2190,32 +2191,33 @@ EXPORT_SYMBOL_GPL(pci_execute_reset_function);
  *
  * This function does not just reset the PCI portion of a device, but
  * clears all the state associated with the device.  This function differs
- * from pci_execute_reset_function in that it saves and restores device state
+ * from __pci_reset_function in that it saves and restores device state
  * over the reset.
  *
- * Returns 0 if the device function was successfully reset or -ENOTTY if the
+ * Returns 0 if the device function was successfully reset or negative if the
  * device doesn't support resetting a single function.
  */
 int pci_reset_function(struct pci_dev *dev)
 {
-	int r = __pci_reset_function(dev, 1);
+	int rc;
 
-	if (r < 0)
-		return r;
+	rc = pci_dev_reset(dev, 1);
+	if (rc)
+		return rc;
 
-	if (!dev->msi_enabled && !dev->msix_enabled && dev->irq != 0)
-		disable_irq(dev->irq);
 	pci_save_state(dev);
 
+	/*
+	 * both INTx and MSI are disabled after the Interrupt Disable bit
+	 * is set and the Bus Master bit is cleared.
+	 */
 	pci_write_config_word(dev, PCI_COMMAND, PCI_COMMAND_INTX_DISABLE);
 
-	r = pci_execute_reset_function(dev);
+	rc = pci_dev_reset(dev, 0);
 
 	pci_restore_state(dev);
-	if (!dev->msi_enabled && !dev->msix_enabled && dev->irq != 0)
-		enable_irq(dev->irq);
 
-	return r;
+	return rc;
 }
 EXPORT_SYMBOL_GPL(pci_reset_function);
 
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 61d9b790d21c..91b06be2f01e 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -702,8 +702,8 @@ int pcix_get_mmrbc(struct pci_dev *dev);
 int pcix_set_mmrbc(struct pci_dev *dev, int mmrbc);
 int pcie_get_readrq(struct pci_dev *dev);
 int pcie_set_readrq(struct pci_dev *dev, int rq);
+int __pci_reset_function(struct pci_dev *dev);
 int pci_reset_function(struct pci_dev *dev);
-int pci_execute_reset_function(struct pci_dev *dev);
 void pci_update_resource(struct pci_dev *dev, int resno);
 int __must_check pci_assign_resource(struct pci_dev *dev, int i);
 int pci_select_bars(struct pci_dev *dev, unsigned long flags);
-- 
cgit v1.2.3-59-g8ed1b


From f85876ba82281f15bc4da11e41b94243a8b2b5b4 Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Sat, 13 Jun 2009 15:52:14 +0800
Subject: PCI: support PM D0hot->D3 transition reset

PCI PM 1.2 specifies that the device will perform an internal reset upon
transitioning from D3hot to D0 when the NO_SOFT_RESET bit is clear. This
method can be used to reset a function if neither PCIe FLR nor PCI AF FLR
are supported.

Reviewed-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c | 34 ++++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 6a052ada3fe8..2e58acc66a8c 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -2132,6 +2132,36 @@ clear:
 	return 0;
 }
 
+static int pci_pm_reset(struct pci_dev *dev, int probe)
+{
+	u16 csr;
+
+	if (!dev->pm_cap)
+		return -ENOTTY;
+
+	pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &csr);
+	if (csr & PCI_PM_CTRL_NO_SOFT_RESET)
+		return -ENOTTY;
+
+	if (probe)
+		return 0;
+
+	if (dev->current_state != PCI_D0)
+		return -EINVAL;
+
+	csr &= ~PCI_PM_CTRL_STATE_MASK;
+	csr |= PCI_D3hot;
+	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, csr);
+	msleep(pci_pm_d3_delay);
+
+	csr &= ~PCI_PM_CTRL_STATE_MASK;
+	csr |= PCI_D0;
+	pci_write_config_word(dev, dev->pm_cap + PCI_PM_CTRL, csr);
+	msleep(pci_pm_d3_delay);
+
+	return 0;
+}
+
 static int pci_dev_reset(struct pci_dev *dev, int probe)
 {
 	int rc;
@@ -2149,6 +2179,10 @@ static int pci_dev_reset(struct pci_dev *dev, int probe)
 		goto done;
 
 	rc = pci_af_flr(dev, probe);
+	if (rc != -ENOTTY)
+		goto done;
+
+	rc = pci_pm_reset(dev, probe);
 done:
 	if (!probe) {
 		up(&dev->dev.sem);
-- 
cgit v1.2.3-59-g8ed1b


From c12ff1df5f114484e3d8abd028769a624cc3399f Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Sat, 13 Jun 2009 15:52:15 +0800
Subject: PCI: support Secondary Bus Reset

PCI-to-PCI Bridge 1.2 specifies that the Secondary Bus Reset bit can
force the assertion of RST# on the secondary interface, which can be
used to reset all devices including subordinates under this bus. This
can be used to reset a function if this function is the only device
under this bus.

Reviewed-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 2e58acc66a8c..c56a4a0355a8 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -2162,6 +2162,33 @@ static int pci_pm_reset(struct pci_dev *dev, int probe)
 	return 0;
 }
 
+static int pci_parent_bus_reset(struct pci_dev *dev, int probe)
+{
+	u16 ctrl;
+	struct pci_dev *pdev;
+
+	if (dev->subordinate)
+		return -ENOTTY;
+
+	list_for_each_entry(pdev, &dev->bus->devices, bus_list)
+		if (pdev != dev)
+			return -ENOTTY;
+
+	if (probe)
+		return 0;
+
+	pci_read_config_word(dev->bus->self, PCI_BRIDGE_CONTROL, &ctrl);
+	ctrl |= PCI_BRIDGE_CTL_BUS_RESET;
+	pci_write_config_word(dev->bus->self, PCI_BRIDGE_CONTROL, ctrl);
+	msleep(100);
+
+	ctrl &= ~PCI_BRIDGE_CTL_BUS_RESET;
+	pci_write_config_word(dev->bus->self, PCI_BRIDGE_CONTROL, ctrl);
+	msleep(100);
+
+	return 0;
+}
+
 static int pci_dev_reset(struct pci_dev *dev, int probe)
 {
 	int rc;
@@ -2183,6 +2210,10 @@ static int pci_dev_reset(struct pci_dev *dev, int probe)
 		goto done;
 
 	rc = pci_pm_reset(dev, probe);
+	if (rc != -ENOTTY)
+		goto done;
+
+	rc = pci_parent_bus_reset(dev, probe);
 done:
 	if (!probe) {
 		up(&dev->dev.sem);
-- 
cgit v1.2.3-59-g8ed1b


From d2abdf62882d982c58e7a6b09ecdcfcc28075e2e Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Sun, 14 Jun 2009 21:25:02 +0200
Subject: PCI PM: Fix handling of devices without PM support by
 pci_target_state()

If a PCI device is not power-manageable either by the platform, or
with the help of the native PCI PM interface, pci_target_state() will
return either PCI_D3hot, or PCI_POWER_ERROR for it, depending on
whether or not the device is configured to wake up the system.  Alas,
none of these return values is correct, because each of them causes
pci_prepare_to_sleep() to return error code, although it should
complete successfully in such a case.

Fix this problem by making pci_target_state() always return PCI_D0
for devices that cannot be power managed.

Cc: stable@kernel.org
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index c56a4a0355a8..7b59fd7c9575 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -1284,15 +1284,14 @@ pci_power_t pci_target_state(struct pci_dev *dev)
 		default:
 			target_state = state;
 		}
+	} else if (!dev->pm_cap) {
+		target_state = PCI_D0;
 	} else if (device_may_wakeup(&dev->dev)) {
 		/*
 		 * Find the deepest state from which the device can generate
 		 * wake-up events, make it the target state and enable device
 		 * to generate PME#.
 		 */
-		if (!dev->pm_cap)
-			return PCI_POWER_ERROR;
-
 		if (dev->pme_support) {
 			while (target_state
 			      && !(dev->pme_support & (1 << target_state)))
-- 
cgit v1.2.3-59-g8ed1b


From ab7de999a2c771482698efa6fe7c7b7fcb1d482a Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Tue, 16 Jun 2009 16:25:40 +0900
Subject: PCI: remove invalid comment of msi_mask_irq()

Remove invalid comment of msi_mask_irq().

Reviewed-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/msi.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index f2725710593a..f48f7550b4a7 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -131,9 +131,6 @@ static inline __attribute_const__ u32 msi_enabled_mask(u16 control)
  * mask all MSI interrupts by clearing the MSI enable bit does not work
  * reliably as devices without an INTx disable bit will then generate a
  * level IRQ which will never be cleared.
- *
- * Returns 1 if it succeeded in masking the interrupt and 0 if the device
- * doesn't support MSI masking.
  */
 static void msi_mask_irq(struct msi_desc *desc, u32 mask, u32 flag)
 {
-- 
cgit v1.2.3-59-g8ed1b


From f9cde5ffed17bf74f6bef042d99edb0622f58576 Mon Sep 17 00:00:00 2001
From: Gary Hade <garyhade@us.ibm.com>
Date: Wed, 27 May 2009 12:41:44 -0700
Subject: x86/ACPI: Correct maximum allowed _CRS returned resources and warn if
 exceeded

Issue a warning if _CRS returns too many resource descriptors to be
accommodated by the fixed size resource array instances.  If there is no
transparent bridge on the root bus "too many" is the
PCI_BUS_NUM_RESOURCES size of the resource array.  Otherwise, the last 3
slots of the resource array must be excluded making the maximum
(PCI_BUS_NUM_RESOURCES - 3).

The current code:
 - is silent when _CRS returns too many resource descriptors and
 - incorrectly allows use of the last 3 slots of the resource array
   for a root bus with a transparent bridge

Signed-off-by: Gary Hade <garyhade@us.ibm.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 arch/x86/pci/acpi.c | 33 +++++++++++++++++++++++++++------
 1 file changed, 27 insertions(+), 6 deletions(-)

diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c
index 8d898e0d3609..16c3fda85bba 100644
--- a/arch/x86/pci/acpi.c
+++ b/arch/x86/pci/acpi.c
@@ -38,15 +38,26 @@ count_resource(struct acpi_resource *acpi_res, void *data)
 	struct acpi_resource_address64 addr;
 	acpi_status status;
 
-	if (info->res_num >= PCI_BUS_NUM_RESOURCES)
-		return AE_OK;
-
 	status = resource_to_addr(acpi_res, &addr);
 	if (ACPI_SUCCESS(status))
 		info->res_num++;
 	return AE_OK;
 }
 
+static int
+bus_has_transparent_bridge(struct pci_bus *bus)
+{
+	struct pci_dev *dev;
+
+	list_for_each_entry(dev, &bus->devices, bus_list) {
+		u16 class = dev->class >> 8;
+
+		if (class == PCI_CLASS_BRIDGE_PCI && dev->transparent)
+			return true;
+	}
+	return false;
+}
+
 static acpi_status
 setup_resource(struct acpi_resource *acpi_res, void *data)
 {
@@ -56,9 +67,7 @@ setup_resource(struct acpi_resource *acpi_res, void *data)
 	acpi_status status;
 	unsigned long flags;
 	struct resource *root;
-
-	if (info->res_num >= PCI_BUS_NUM_RESOURCES)
-		return AE_OK;
+	int max_root_bus_resources = PCI_BUS_NUM_RESOURCES;
 
 	status = resource_to_addr(acpi_res, &addr);
 	if (!ACPI_SUCCESS(status))
@@ -82,6 +91,18 @@ setup_resource(struct acpi_resource *acpi_res, void *data)
 	res->end = res->start + addr.address_length - 1;
 	res->child = NULL;
 
+	if (bus_has_transparent_bridge(info->bus))
+		max_root_bus_resources -= 3;
+	if (info->res_num >= max_root_bus_resources) {
+		printk(KERN_WARNING "PCI: Failed to allocate 0x%lx-0x%lx "
+			"from %s for %s due to _CRS returning more than "
+			"%d resource descriptors\n", (unsigned long) res->start,
+			(unsigned long) res->end, root->name, info->name,
+			max_root_bus_resources);
+		info->res_num++;
+		return AE_OK;
+	}
+
 	if (insert_resource(root, res)) {
 		printk(KERN_ERR "PCI: Failed to allocate 0x%lx-0x%lx "
 			"from %s for %s\n", (unsigned long) res->start,
-- 
cgit v1.2.3-59-g8ed1b


From 7d9a73f6dcf4390d256bf19330c81e91523a26d5 Mon Sep 17 00:00:00 2001
From: Frans Pop <elendil@planet.nl>
Date: Wed, 17 Jun 2009 00:16:15 +0200
Subject: PCI PM: consistently use type bool for wake enable variable

Other functions use type bool, so use that for pci_enable_wake as well.

Signed-off-by: Frans Pop <elendil@planet.nl>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c   | 2 +-
 include/linux/pci.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 7b59fd7c9575..ccc0a0ccbef9 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -1205,7 +1205,7 @@ void pci_pme_active(struct pci_dev *dev, bool enable)
  * Error code depending on the platform is returned if both the platform and
  * the native mechanism fail to enable the generation of wake-up events
  */
-int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable)
+int pci_enable_wake(struct pci_dev *dev, pci_power_t state, bool enable)
 {
 	int error = 0;
 	bool pme_done = false;
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 91b06be2f01e..62e8452c2ec6 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -723,7 +723,7 @@ int pci_set_power_state(struct pci_dev *dev, pci_power_t state);
 pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state);
 bool pci_pme_capable(struct pci_dev *dev, pci_power_t state);
 void pci_pme_active(struct pci_dev *dev, bool enable);
-int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable);
+int pci_enable_wake(struct pci_dev *dev, pci_power_t state, bool enable);
 int pci_wake_from_d3(struct pci_dev *dev, bool enable);
 pci_power_t pci_target_state(struct pci_dev *dev);
 int pci_prepare_to_sleep(struct pci_dev *dev);
-- 
cgit v1.2.3-59-g8ed1b


From 3f1a567d8a4ed7a5d105bd049343606f5273b603 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Tue, 2 Jun 2009 09:31:03 +0100
Subject: [ARM] VIC: Fix resume sources usage

The resume_mask wasn't being checked in vic_set_wake()
to see if the IRQ was a valid wakeup source.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/common/vic.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/arch/arm/common/vic.c b/arch/arm/common/vic.c
index 887c6eb3a18a..6ed89836e908 100644
--- a/arch/arm/common/vic.c
+++ b/arch/arm/common/vic.c
@@ -229,14 +229,18 @@ static int vic_set_wake(unsigned int irq, unsigned int on)
 {
 	struct vic_device *v = vic_from_irq(irq);
 	unsigned int off = irq & 31;
+	u32 bit = 1 << off;
 
 	if (!v)
 		return -EINVAL;
 
+	if (!(bit & v->resume_sources))
+		return -EINVAL;
+
 	if (on)
-		v->resume_irqs |= 1 << off;
+		v->resume_irqs |= bit;
 	else
-		v->resume_irqs &= ~(1 << off);
+		v->resume_irqs &= ~bit;
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 9492b272820ac8f32a6ad4fcabd571d8305a6d47 Mon Sep 17 00:00:00 2001
From: Huang Weiyi <weiyi.huang@gmail.com>
Date: Tue, 16 Jun 2009 16:56:34 +0800
Subject: [ARM] S3C24XX: remove duplicated #include

Remove duplicated #include in arch/arm/mach-s3c2410/usb-simtec.c.

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/mach-s3c2410/usb-simtec.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/arm/mach-s3c2410/usb-simtec.c b/arch/arm/mach-s3c2410/usb-simtec.c
index 6cd9377ddb82..50e25fc5f8ab 100644
--- a/arch/arm/mach-s3c2410/usb-simtec.c
+++ b/arch/arm/mach-s3c2410/usb-simtec.c
@@ -22,7 +22,6 @@
 #include <linux/timer.h>
 #include <linux/init.h>
 #include <linux/device.h>
-#include <linux/gpio.h>
 #include <linux/io.h>
 
 #include <asm/mach/arch.h>
-- 
cgit v1.2.3-59-g8ed1b


From 52da219e9664e537a745877b0efa7cf2b1ff2996 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Wed, 3 Jun 2009 20:08:38 +0100
Subject: [ARM] S3C64XX: Provide device for IIS ports

This allows the S3C64XX IIS drivers to be converted to the standard
driver model and allows fixes there for problems with attempting to
acquire the clocks for the IIS blocks via the clock API.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c/Makefile            |  1 +
 arch/arm/plat-s3c/dev-audio.c         | 52 +++++++++++++++++++++++++++++++++++
 arch/arm/plat-s3c/include/plat/devs.h |  4 ++-
 3 files changed, 56 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/plat-s3c/dev-audio.c

diff --git a/arch/arm/plat-s3c/Makefile b/arch/arm/plat-s3c/Makefile
index 610651455a78..74bb7cb5da49 100644
--- a/arch/arm/plat-s3c/Makefile
+++ b/arch/arm/plat-s3c/Makefile
@@ -34,6 +34,7 @@ obj-$(CONFIG_S3C_DEV_HSMMC)	+= dev-hsmmc.o
 obj-$(CONFIG_S3C_DEV_HSMMC1)	+= dev-hsmmc1.o
 obj-y				+= dev-i2c0.o
 obj-$(CONFIG_S3C_DEV_I2C1)	+= dev-i2c1.o
+obj-$(CONFIG_SND_S3C24XX_SOC)	+= dev-audio.o
 obj-$(CONFIG_S3C_DEV_FB)	+= dev-fb.o
 obj-$(CONFIG_S3C_DEV_USB_HOST)	+= dev-usb.o
 obj-$(CONFIG_S3C_DEV_USB_HSOTG)	+= dev-usb-hsotg.o
diff --git a/arch/arm/plat-s3c/dev-audio.c b/arch/arm/plat-s3c/dev-audio.c
new file mode 100644
index 000000000000..ccdd81584310
--- /dev/null
+++ b/arch/arm/plat-s3c/dev-audio.c
@@ -0,0 +1,52 @@
+/* linux/arch/arm/plat-s3c/dev-audio.c
+ *
+ * Copyright 2009 Wolfson Microelectronics
+ *      Mark Brown <broonie@opensource.wolfsonmicro.com>
+ *
+
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/platform_device.h>
+
+#include <mach/irqs.h>
+#include <mach/map.h>
+
+#include <plat/devs.h>
+
+
+static struct resource s3c64xx_iis0_resource[] = {
+	[0] = {
+		.start = S3C64XX_PA_IIS0,
+		.end   = S3C64XX_PA_IIS0 + 0x100 - 1,
+		.flags = IORESOURCE_MEM,
+	},
+};
+
+struct platform_device s3c64xx_device_iis0 = {
+	.name		  = "s3c64xx-iis",
+	.id		  = 0,
+	.num_resources	  = ARRAY_SIZE(s3c64xx_iis0_resource),
+	.resource	  = s3c64xx_iis0_resource,
+};
+EXPORT_SYMBOL(s3c64xx_device_iis0);
+
+static struct resource s3c64xx_iis1_resource[] = {
+	[0] = {
+		.start = S3C64XX_PA_IIS1,
+		.end   = S3C64XX_PA_IIS1 + 0x100 - 1,
+		.flags = IORESOURCE_MEM,
+	},
+};
+
+struct platform_device s3c64xx_device_iis1 = {
+	.name		  = "s3c64xx-iis",
+	.id		  = 1,
+	.num_resources	  = ARRAY_SIZE(s3c64xx_iis1_resource),
+	.resource	  = s3c64xx_iis1_resource,
+};
+EXPORT_SYMBOL(s3c64xx_device_iis1);
diff --git a/arch/arm/plat-s3c/include/plat/devs.h b/arch/arm/plat-s3c/include/plat/devs.h
index a0b6768fddcf..91de11be274a 100644
--- a/arch/arm/plat-s3c/include/plat/devs.h
+++ b/arch/arm/plat-s3c/include/plat/devs.h
@@ -24,13 +24,15 @@ extern struct platform_device *s3c24xx_uart_src[];
 
 extern struct platform_device s3c_device_timer[];
 
+extern struct platform_device s3c64xx_device_iis0;
+extern struct platform_device s3c64xx_device_iis1;
+
 extern struct platform_device s3c_device_fb;
 extern struct platform_device s3c_device_usb;
 extern struct platform_device s3c_device_lcd;
 extern struct platform_device s3c_device_wdt;
 extern struct platform_device s3c_device_i2c0;
 extern struct platform_device s3c_device_i2c1;
-extern struct platform_device s3c_device_iis;
 extern struct platform_device s3c_device_rtc;
 extern struct platform_device s3c_device_adc;
 extern struct platform_device s3c_device_sdi;
-- 
cgit v1.2.3-59-g8ed1b


From d06a49eec97718949acfdc26110701d28b1872c0 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Wed, 3 Jun 2009 20:08:39 +0100
Subject: [ARM] S3C64XX: Add device for IISv4 port

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c/dev-audio.c         | 16 ++++++++++++++++
 arch/arm/plat-s3c/include/plat/devs.h |  1 +
 2 files changed, 17 insertions(+)

diff --git a/arch/arm/plat-s3c/dev-audio.c b/arch/arm/plat-s3c/dev-audio.c
index ccdd81584310..1322beb40dd7 100644
--- a/arch/arm/plat-s3c/dev-audio.c
+++ b/arch/arm/plat-s3c/dev-audio.c
@@ -50,3 +50,19 @@ struct platform_device s3c64xx_device_iis1 = {
 	.resource	  = s3c64xx_iis1_resource,
 };
 EXPORT_SYMBOL(s3c64xx_device_iis1);
+
+static struct resource s3c64xx_iisv4_resource[] = {
+	[0] = {
+		.start = S3C64XX_PA_IISV4,
+		.end   = S3C64XX_PA_IISV4 + 0x100 - 1,
+		.flags = IORESOURCE_MEM,
+	},
+};
+
+struct platform_device s3c64xx_device_iisv4 = {
+	.name		  = "s3c64xx-iis-v4",
+	.id		  = -1,
+	.num_resources	  = ARRAY_SIZE(s3c64xx_iisv4_resource),
+	.resource	  = s3c64xx_iisv4_resource,
+};
+EXPORT_SYMBOL(s3c64xx_device_iisv4);
diff --git a/arch/arm/plat-s3c/include/plat/devs.h b/arch/arm/plat-s3c/include/plat/devs.h
index 91de11be274a..b5b9c4d46e9a 100644
--- a/arch/arm/plat-s3c/include/plat/devs.h
+++ b/arch/arm/plat-s3c/include/plat/devs.h
@@ -26,6 +26,7 @@ extern struct platform_device s3c_device_timer[];
 
 extern struct platform_device s3c64xx_device_iis0;
 extern struct platform_device s3c64xx_device_iis1;
+extern struct platform_device s3c64xx_device_iisv4;
 
 extern struct platform_device s3c_device_fb;
 extern struct platform_device s3c_device_usb;
-- 
cgit v1.2.3-59-g8ed1b


From b3748ddd80569ec753f62e709629b8c639143222 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Mon, 15 Jun 2009 11:23:20 +0100
Subject: [ARM] S3C64XX: Initial support for DVFS

This patch provides initial support for CPU frequency scaling on the
Samsung S3C ARM processors. Currently only S3C6410 processors are
supported, though addition of another data table with supported clock
rates should be sufficient to enable support for further CPUs.

Use the regulator framework to provide optional support for DVFS in
the S3C cpufreq driver. When a software controllable regulator is
configured the driver will use it to lower the supply voltage when
running at a lower frequency, giving improved power savings.

When regulator support is disabled or no regulator can be obtained
for VDDARM the driver will fall back to scaling only the frequency.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/Kconfig                |   6 +-
 arch/arm/plat-s3c64xx/Makefile  |   1 +
 arch/arm/plat-s3c64xx/cpufreq.c | 262 ++++++++++++++++++++++++++++++++++++++++
 3 files changed, 268 insertions(+), 1 deletion(-)
 create mode 100644 arch/arm/plat-s3c64xx/cpufreq.c

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 29475101a7b3..aef63c8e3d2d 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -1241,7 +1241,7 @@ endmenu
 
 menu "CPU Power Management"
 
-if (ARCH_SA1100 || ARCH_INTEGRATOR || ARCH_OMAP || ARCH_PXA)
+if (ARCH_SA1100 || ARCH_INTEGRATOR || ARCH_OMAP || ARCH_PXA || ARCH_S3C64XX)
 
 source "drivers/cpufreq/Kconfig"
 
@@ -1272,6 +1272,10 @@ config CPU_FREQ_PXA
 	default y
 	select CPU_FREQ_DEFAULT_GOV_USERSPACE
 
+config CPU_FREQ_S3C64XX
+	bool "CPUfreq support for Samsung S3C64XX CPUs"
+	depends on CPU_FREQ && CPU_S3C6410
+
 endif
 
 source "drivers/cpuidle/Kconfig"
diff --git a/arch/arm/plat-s3c64xx/Makefile b/arch/arm/plat-s3c64xx/Makefile
index 2ed5df34f9ea..3c8882cd6268 100644
--- a/arch/arm/plat-s3c64xx/Makefile
+++ b/arch/arm/plat-s3c64xx/Makefile
@@ -23,6 +23,7 @@ obj-y				+= gpiolib.o
 
 obj-$(CONFIG_CPU_S3C6400_INIT)	+= s3c6400-init.o
 obj-$(CONFIG_CPU_S3C6400_CLOCK)	+= s3c6400-clock.o
+obj-$(CONFIG_CPU_FREQ_S3C64XX)  += cpufreq.o
 
 # PM support
 
diff --git a/arch/arm/plat-s3c64xx/cpufreq.c b/arch/arm/plat-s3c64xx/cpufreq.c
new file mode 100644
index 000000000000..e6e0843215df
--- /dev/null
+++ b/arch/arm/plat-s3c64xx/cpufreq.c
@@ -0,0 +1,262 @@
+/* linux/arch/arm/plat-s3c64xx/cpufreq.c
+ *
+ * Copyright 2009 Wolfson Microelectronics plc
+ *
+ * S3C64xx CPUfreq Support
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/cpufreq.h>
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/regulator/consumer.h>
+
+static struct clk *armclk;
+static struct regulator *vddarm;
+
+#ifdef CONFIG_CPU_S3C6410
+struct s3c64xx_dvfs {
+	unsigned int vddarm_min;
+	unsigned int vddarm_max;
+};
+
+static struct s3c64xx_dvfs s3c64xx_dvfs_table[] = {
+	[0] = { 1000000, 1000000 },
+	[1] = { 1000000, 1050000 },
+	[2] = { 1050000, 1100000 },
+	[3] = { 1050000, 1150000 },
+	[4] = { 1250000, 1350000 },
+};
+
+static struct cpufreq_frequency_table s3c64xx_freq_table[] = {
+	{ 0,  66000 },
+	{ 0, 133000 },
+	{ 1, 222000 },
+	{ 1, 266000 },
+	{ 2, 333000 },
+	{ 2, 400000 },
+	{ 3, 532000 },
+	{ 3, 533000 },
+	{ 4, 667000 },
+	{ 0, CPUFREQ_TABLE_END },
+};
+#endif
+
+static int s3c64xx_cpufreq_verify_speed(struct cpufreq_policy *policy)
+{
+	if (policy->cpu != 0)
+		return -EINVAL;
+
+	return cpufreq_frequency_table_verify(policy, s3c64xx_freq_table);
+}
+
+static unsigned int s3c64xx_cpufreq_get_speed(unsigned int cpu)
+{
+	if (cpu != 0)
+		return 0;
+
+	return clk_get_rate(armclk) / 1000;
+}
+
+static int s3c64xx_cpufreq_set_target(struct cpufreq_policy *policy,
+				      unsigned int target_freq,
+				      unsigned int relation)
+{
+	int ret;
+	unsigned int i;
+	struct cpufreq_freqs freqs;
+	struct s3c64xx_dvfs *dvfs;
+
+	ret = cpufreq_frequency_table_target(policy, s3c64xx_freq_table,
+					     target_freq, relation, &i);
+	if (ret != 0)
+		return ret;
+
+	freqs.cpu = 0;
+	freqs.old = clk_get_rate(armclk) / 1000;
+	freqs.new = s3c64xx_freq_table[i].frequency;
+	freqs.flags = 0;
+	dvfs = &s3c64xx_dvfs_table[s3c64xx_freq_table[i].index];
+
+	if (freqs.old == freqs.new)
+		return 0;
+
+	pr_debug("cpufreq: Transition %d-%dkHz\n", freqs.old, freqs.new);
+
+	cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
+
+#ifdef CONFIG_REGULATOR
+	if (vddarm && freqs.new > freqs.old) {
+		ret = regulator_set_voltage(vddarm,
+					    dvfs->vddarm_min,
+					    dvfs->vddarm_max);
+		if (ret != 0) {
+			pr_err("cpufreq: Failed to set VDDARM for %dkHz: %d\n",
+			       freqs.new, ret);
+			goto err;
+		}
+	}
+#endif
+
+	ret = clk_set_rate(armclk, freqs.new * 1000);
+	if (ret < 0) {
+		pr_err("cpufreq: Failed to set rate %dkHz: %d\n",
+		       freqs.new, ret);
+		goto err;
+	}
+
+#ifdef CONFIG_REGULATOR
+	if (vddarm && freqs.new < freqs.old) {
+		ret = regulator_set_voltage(vddarm,
+					    dvfs->vddarm_min,
+					    dvfs->vddarm_max);
+		if (ret != 0) {
+			pr_err("cpufreq: Failed to set VDDARM for %dkHz: %d\n",
+			       freqs.new, ret);
+			goto err_clk;
+		}
+	}
+#endif
+
+	cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
+
+	pr_debug("cpufreq: Set actual frequency %lukHz\n",
+		 clk_get_rate(armclk) / 1000);
+
+	return 0;
+
+err_clk:
+	if (clk_set_rate(armclk, freqs.old * 1000) < 0)
+		pr_err("Failed to restore original clock rate\n");
+err:
+	cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
+
+	return ret;
+}
+
+#ifdef CONFIG_REGULATOR
+static void __init s3c64xx_cpufreq_constrain_voltages(void)
+{
+	int count, v, i, found;
+	struct cpufreq_frequency_table *freq;
+	struct s3c64xx_dvfs *dvfs;
+
+	count = regulator_count_voltages(vddarm);
+	if (count < 0) {
+		pr_err("cpufreq: Unable to check supported voltages\n");
+		return;
+	}
+
+	freq = s3c64xx_freq_table;
+	while (freq->frequency != CPUFREQ_TABLE_END) {
+		if (freq->frequency == CPUFREQ_ENTRY_INVALID)
+			continue;
+
+		dvfs = &s3c64xx_dvfs_table[freq->index];
+		found = 0;
+
+		for (i = 0; i < count; i++) {
+			v = regulator_list_voltage(vddarm, i);
+			if (v >= dvfs->vddarm_min && v <= dvfs->vddarm_max)
+				found = 1;
+		}
+
+		if (!found) {
+			pr_debug("cpufreq: %dkHz unsupported by regulator\n",
+				 freq->frequency);
+			freq->frequency = CPUFREQ_ENTRY_INVALID;
+		}
+
+		freq++;
+	}
+}
+#endif
+
+static int __init s3c64xx_cpufreq_driver_init(struct cpufreq_policy *policy)
+{
+	int ret;
+	struct cpufreq_frequency_table *freq;
+
+	if (policy->cpu != 0)
+		return -EINVAL;
+
+	if (s3c64xx_freq_table == NULL) {
+		pr_err("cpufreq: No frequency information for this CPU\n");
+		return -ENODEV;
+	}
+
+	armclk = clk_get(NULL, "armclk");
+	if (IS_ERR(armclk)) {
+		pr_err("cpufreq: Unable to obtain ARMCLK: %ld\n",
+		       PTR_ERR(armclk));
+		return PTR_ERR(armclk);
+	}
+
+#ifdef CONFIG_REGULATOR
+	vddarm = regulator_get(NULL, "vddarm");
+	if (IS_ERR(vddarm)) {
+		ret = PTR_ERR(vddarm);
+		pr_err("cpufreq: Failed to obtain VDDARM: %d\n", ret);
+		pr_err("cpufreq: Only frequency scaling available\n");
+		vddarm = NULL;
+	} else {
+		s3c64xx_cpufreq_constrain_voltages();
+	}
+#endif
+
+	freq = s3c64xx_freq_table;
+	while (freq->frequency != CPUFREQ_TABLE_END) {
+		unsigned long r;
+
+		/* Check for frequencies we can generate */
+		r = clk_round_rate(armclk, freq->frequency * 1000);
+		r /= 1000;
+		if (r != freq->frequency)
+			freq->frequency = CPUFREQ_ENTRY_INVALID;
+
+		/* If we have no regulator then assume startup
+		 * frequency is the maximum we can support. */
+		if (!vddarm && freq->frequency > s3c64xx_cpufreq_get_speed(0))
+			freq->frequency = CPUFREQ_ENTRY_INVALID;
+
+		freq++;
+	}
+
+	policy->cur = clk_get_rate(armclk) / 1000;
+
+	/* Pick a conservative guess in ns: we'll need ~1 I2C/SPI
+	 * write plus clock reprogramming. */
+	policy->cpuinfo.transition_latency = 2 * 1000 * 1000;
+
+	ret = cpufreq_frequency_table_cpuinfo(policy, s3c64xx_freq_table);
+	if (ret != 0) {
+		pr_err("cpufreq: Failed to configure frequency table: %d\n",
+		       ret);
+		regulator_put(vddarm);
+		clk_put(armclk);
+	}
+
+	return ret;
+}
+
+static struct cpufreq_driver s3c64xx_cpufreq_driver = {
+	.owner		= THIS_MODULE,
+	.flags          = 0,
+	.verify		= s3c64xx_cpufreq_verify_speed,
+	.target		= s3c64xx_cpufreq_set_target,
+	.get		= s3c64xx_cpufreq_get_speed,
+	.init		= s3c64xx_cpufreq_driver_init,
+	.name		= "s3c",
+};
+
+static int __init s3c64xx_cpufreq_init(void)
+{
+	return cpufreq_register_driver(&s3c64xx_cpufreq_driver);
+}
+module_init(s3c64xx_cpufreq_init);
-- 
cgit v1.2.3-59-g8ed1b


From 9d76295ac6082449a64fd6fa981d2615a34e243d Mon Sep 17 00:00:00 2001
From: Andy Green <andy@warmcat.com>
Date: Tue, 19 May 2009 06:41:42 -0300
Subject: [ARM] GTA02/FreeRunner: Add machine definition

This patch introduces the Openmoko GTA02 machine definition.

Much of the code is based on Harald Welte's work, although it
has been largely rewritten several times.

This is intended to be the minimum machine definition to boot and
be able to run a rootfs from NAND on GTA02 / FreeRunner.  It does
not bring up the framebuffer / Glamo and lacks many other peripheral
drivers from outside the SoC.  But once we have this basis in
mainline kernel, we will be able to introduce the other drivers
and add them here.

Thanks to Sven Rebhan <odinshorse@googlemail.com> for his fixes to this
patch (Kconfig and defconfig files).

Signed-off-by: Andy Green <andy@warmcat.com>
Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>
[ben-linux@fluff.org: Fix the GPIO definitions]
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 MAINTAINERS                                |   7 +
 arch/arm/mach-s3c2442/Kconfig              |  12 +
 arch/arm/mach-s3c2442/Makefile             |   2 +
 arch/arm/mach-s3c2442/include/mach/gta02.h |  84 ++++
 arch/arm/mach-s3c2442/mach-gta02.c         | 646 +++++++++++++++++++++++++++++
 5 files changed, 751 insertions(+)
 create mode 100644 arch/arm/mach-s3c2442/include/mach/gta02.h
 create mode 100644 arch/arm/mach-s3c2442/mach-gta02.c

diff --git a/MAINTAINERS b/MAINTAINERS
index 2cb7566904b1..efa7bfa39f84 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -772,6 +772,13 @@ P:	Michael Petchkovsky
 M:	mkpetch@internode.on.net
 S:	Maintained
 
+ARM/OPENMOKO NEO FREERUNNER (GTA02) MACHINE SUPPORT
+P:	Nelson Castillo
+M:	arhuaco@freaks-unidos.net
+L:	openmoko-kernel@lists.openmoko.org (subscribers-only)
+W:	http://wiki.openmoko.org/wiki/Neo_FreeRunner
+S:	Supported
+
 ARM/TOSA MACHINE SUPPORT
 P:	Dmitry Eremin-Solenikov
 M:	dbaryshkov@gmail.com
diff --git a/arch/arm/mach-s3c2442/Kconfig b/arch/arm/mach-s3c2442/Kconfig
index b289d198020e..103e913f2258 100644
--- a/arch/arm/mach-s3c2442/Kconfig
+++ b/arch/arm/mach-s3c2442/Kconfig
@@ -24,6 +24,18 @@ config SMDK2440_CPU2442
 	depends on ARCH_S3C2440
 	select CPU_S3C2442
 
+config MACH_NEO1973_GTA02
+	bool "Openmoko GTA02 / Freerunner phone"
+	select CPU_S3C2442
+	select MFD_PCF50633
+	select PCF50633_GPIO
+	select I2C
+	select POWER_SUPPLY
+	select MACH_NEO1973
+	select S3C2410_PWM
+	help
+	   Say Y here if you are using the Openmoko GTA02 / Freerunner GSM Phone
+
 
 endmenu
 
diff --git a/arch/arm/mach-s3c2442/Makefile b/arch/arm/mach-s3c2442/Makefile
index 2a909c6c5798..2a19113a5769 100644
--- a/arch/arm/mach-s3c2442/Makefile
+++ b/arch/arm/mach-s3c2442/Makefile
@@ -12,5 +12,7 @@ obj-				:=
 obj-$(CONFIG_CPU_S3C2442)	+= s3c2442.o
 obj-$(CONFIG_CPU_S3C2442)	+= clock.o
 
+obj-$(CONFIG_MACH_NEO1973_GTA02) += mach-gta02.o
+
 # Machine support
 
diff --git a/arch/arm/mach-s3c2442/include/mach/gta02.h b/arch/arm/mach-s3c2442/include/mach/gta02.h
new file mode 100644
index 000000000000..953331d8d56a
--- /dev/null
+++ b/arch/arm/mach-s3c2442/include/mach/gta02.h
@@ -0,0 +1,84 @@
+#ifndef _GTA02_H
+#define _GTA02_H
+
+#include <mach/regs-gpio.h>
+
+/* Different hardware revisions, passed in ATAG_REVISION by u-boot */
+#define GTA02v1_SYSTEM_REV	0x00000310
+#define GTA02v2_SYSTEM_REV	0x00000320
+#define GTA02v3_SYSTEM_REV	0x00000330
+#define GTA02v4_SYSTEM_REV	0x00000340
+#define GTA02v5_SYSTEM_REV	0x00000350
+/* since A7 is basically same as A6, we use A6 PCB ID */
+#define GTA02v6_SYSTEM_REV	0x00000360
+
+#define GTA02_GPIO_n3DL_GSM	S3C2410_GPA(13)	/* v1 + v2 + v3 only */
+
+#define GTA02_GPIO_PWR_LED1	S3C2410_GPB(0)
+#define GTA02_GPIO_PWR_LED2	S3C2410_GPB(1)
+#define GTA02_GPIO_AUX_LED	S3C2410_GPB(2)
+#define GTA02_GPIO_VIBRATOR_ON	S3C2410_GPB(3)
+#define GTA02_GPIO_MODEM_RST	S3C2410_GPB(5)
+#define GTA02_GPIO_BT_EN	S3C2410_GPB(6)
+#define GTA02_GPIO_MODEM_ON	S3C2410_GPB(7)
+#define GTA02_GPIO_EXTINT8	S3C2410_GPB(8)
+#define GTA02_GPIO_USB_PULLUP	S3C2410_GPB(9)
+
+#define GTA02_GPIO_PIO5		S3C2410_GPC(5)	/* v3 + v4 only */
+
+#define GTA02v3_GPIO_nG1_CS	S3C2410_GPD(12)	/* v3 + v4 only */
+#define GTA02v3_GPIO_nG2_CS	S3C2410_GPD(13)	/* v3 + v4 only */
+#define GTA02v5_GPIO_HDQ	S3C2410_GPD(14)   /* v5 + */
+
+#define GTA02_GPIO_nG1_INT	S3C2410_GPF(0)
+#define GTA02_GPIO_IO1		S3C2410_GPF(1)
+#define GTA02_GPIO_PIO_2	S3C2410_GPF(2)	/* v2 + v3 + v4 only */
+#define GTA02_GPIO_JACK_INSERT	S3C2410_GPF(4)
+#define GTA02_GPIO_WLAN_GPIO1	S3C2410_GPF(5)	/* v2 + v3 + v4 only */
+#define GTA02_GPIO_AUX_KEY	S3C2410_GPF(6)
+#define GTA02_GPIO_HOLD_KEY	S3C2410_GPF(7)
+
+#define GTA02_GPIO_3D_IRQ	S3C2410_GPG(4)
+#define GTA02v2_GPIO_nG2_INT	S3C2410_GPG(8)	/* v2 + v3 + v4 only */
+#define GTA02v3_GPIO_nUSB_OC	S3C2410_GPG(9)	/* v3 + v4 only */
+#define GTA02v3_GPIO_nUSB_FLT	S3C2410_GPG(10)	/* v3 + v4 only */
+#define GTA02v3_GPIO_nGSM_OC	S3C2410_GPG(11)	/* v3 + v4 only */
+
+#define GTA02_GPIO_AMP_SHUT	S3C2440_GPJ1	/* v2 + v3 + v4 only */
+#define GTA02v1_GPIO_WLAN_GPIO10	S3C2440_GPJ2
+#define GTA02_GPIO_HP_IN	S3C2440_GPJ2	/* v2 + v3 + v4 only */
+#define GTA02_GPIO_INT0		S3C2440_GPJ3	/* v2 + v3 + v4 only */
+#define GTA02_GPIO_nGSM_EN	S3C2440_GPJ4
+#define GTA02_GPIO_3D_RESET	S3C2440_GPJ5
+#define GTA02_GPIO_nDL_GSM	S3C2440_GPJ6	/* v4 + v5 only */
+#define GTA02_GPIO_WLAN_GPIO0	S3C2440_GPJ7
+#define GTA02v1_GPIO_BAT_ID	S3C2440_GPJ8
+#define GTA02_GPIO_KEEPACT	S3C2440_GPJ8
+#define GTA02v1_GPIO_HP_IN	S3C2440_GPJ10
+#define GTA02_CHIP_PWD		S3C2440_GPJ11	/* v2 + v3 + v4 only */
+#define GTA02_GPIO_nWLAN_RESET	S3C2440_GPJ12	/* v2 + v3 + v4 only */
+
+#define GTA02_IRQ_GSENSOR_1	IRQ_EINT0
+#define GTA02_IRQ_MODEM		IRQ_EINT1
+#define GTA02_IRQ_PIO_2		IRQ_EINT2	/* v2 + v3 + v4 only */
+#define GTA02_IRQ_nJACK_INSERT	IRQ_EINT4
+#define GTA02_IRQ_WLAN_GPIO1	IRQ_EINT5
+#define GTA02_IRQ_AUX		IRQ_EINT6
+#define GTA02_IRQ_nHOLD		IRQ_EINT7
+#define GTA02_IRQ_PCF50633	IRQ_EINT9
+#define GTA02_IRQ_3D		IRQ_EINT12
+#define GTA02_IRQ_GSENSOR_2	IRQ_EINT16	/* v2 + v3 + v4 only */
+#define GTA02v3_IRQ_nUSB_OC	IRQ_EINT17	/* v3 + v4 only */
+#define GTA02v3_IRQ_nUSB_FLT	IRQ_EINT18	/* v3 + v4 only */
+#define GTA02v3_IRQ_nGSM_OC	IRQ_EINT19	/* v3 + v4 only */
+
+/* returns 00 000 on GTA02 A5 and earlier, A6 returns 01 001 */
+#define GTA02_PCB_ID1_0		S3C2410_GPC(13)
+#define GTA02_PCB_ID1_1		S3C2410_GPC(15)
+#define GTA02_PCB_ID1_2		S3C2410_GPD(0)
+#define GTA02_PCB_ID2_0		S3C2410_GPD(3)
+#define GTA02_PCB_ID2_1		S3C2410_GPD(4)
+
+int gta02_get_pcb_revision(void);
+
+#endif /* _GTA02_H */
diff --git a/arch/arm/mach-s3c2442/mach-gta02.c b/arch/arm/mach-s3c2442/mach-gta02.c
new file mode 100644
index 000000000000..e23b581aa0e1
--- /dev/null
+++ b/arch/arm/mach-s3c2442/mach-gta02.c
@@ -0,0 +1,646 @@
+/*
+ * linux/arch/arm/mach-s3c2442/mach-gta02.c
+ *
+ * S3C2442 Machine Support for Openmoko GTA02 / FreeRunner.
+ *
+ * Copyright (C) 2006-2009 by Openmoko, Inc.
+ * Authors: Harald Welte <laforge@openmoko.org>
+ *          Andy Green <andy@openmoko.org>
+ *          Werner Almesberger <werner@openmoko.org>
+ * All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of
+ * the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston,
+ * MA 02111-1307 USA
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/interrupt.h>
+#include <linux/list.h>
+#include <linux/delay.h>
+#include <linux/timer.h>
+#include <linux/init.h>
+#include <linux/gpio.h>
+#include <linux/workqueue.h>
+#include <linux/platform_device.h>
+#include <linux/serial_core.h>
+#include <linux/spi/spi.h>
+
+#include <linux/mmc/host.h>
+
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/nand.h>
+#include <linux/mtd/nand_ecc.h>
+#include <linux/mtd/partitions.h>
+#include <linux/mtd/physmap.h>
+#include <linux/io.h>
+
+#include <linux/i2c.h>
+#include <linux/backlight.h>
+#include <linux/regulator/machine.h>
+
+#include <linux/mfd/pcf50633/core.h>
+#include <linux/mfd/pcf50633/mbc.h>
+#include <linux/mfd/pcf50633/adc.h>
+#include <linux/mfd/pcf50633/gpio.h>
+#include <linux/mfd/pcf50633/pmic.h>
+
+#include <asm/mach/arch.h>
+#include <asm/mach/map.h>
+#include <asm/mach/irq.h>
+
+#include <asm/irq.h>
+#include <asm/mach-types.h>
+
+#include <mach/regs-irq.h>
+#include <mach/regs-gpio.h>
+#include <mach/regs-gpioj.h>
+#include <mach/fb.h>
+
+#include <mach/spi.h>
+#include <mach/spi-gpio.h>
+#include <plat/usb-control.h>
+#include <mach/regs-mem.h>
+#include <mach/hardware.h>
+
+#include <mach/gta02.h>
+
+#include <plat/regs-serial.h>
+#include <plat/nand.h>
+#include <plat/devs.h>
+#include <plat/cpu.h>
+#include <plat/pm.h>
+#include <plat/udc.h>
+#include <plat/gpio-cfg.h>
+#include <plat/iic.h>
+
+static struct pcf50633 *gta02_pcf;
+
+/*
+ * This gets called every 1ms when we paniced.
+ */
+
+static long gta02_panic_blink(long count)
+{
+	long delay = 0;
+	static long last_blink;
+	static char led;
+
+	/* Fast blink: 200ms period. */
+	if (count - last_blink < 100)
+		return 0;
+
+	led ^= 1;
+	gpio_direction_output(GTA02_GPIO_AUX_LED, led);
+
+	last_blink = count;
+
+	return delay;
+}
+
+
+static struct map_desc gta02_iodesc[] __initdata = {
+	{
+		.virtual	= 0xe0000000,
+		.pfn		= __phys_to_pfn(S3C2410_CS3 + 0x01000000),
+		.length		= SZ_1M,
+		.type		= MT_DEVICE
+	},
+};
+
+#define UCON (S3C2410_UCON_DEFAULT | S3C2443_UCON_RXERR_IRQEN)
+#define ULCON (S3C2410_LCON_CS8 | S3C2410_LCON_PNONE | S3C2410_LCON_STOPB)
+#define UFCON (S3C2410_UFCON_RXTRIG8 | S3C2410_UFCON_FIFOMODE)
+
+static struct s3c2410_uartcfg gta02_uartcfgs[] = {
+	[0] = {
+		.hwport		= 0,
+		.flags		= 0,
+		.ucon		= UCON,
+		.ulcon		= ULCON,
+		.ufcon		= UFCON,
+	},
+	[1] = {
+		.hwport		= 1,
+		.flags		= 0,
+		.ucon		= UCON,
+		.ulcon		= ULCON,
+		.ufcon		= UFCON,
+	},
+	[2] = {
+		.hwport		= 2,
+		.flags		= 0,
+		.ucon		= UCON,
+		.ulcon		= ULCON,
+		.ufcon		= UFCON,
+	},
+};
+
+#ifdef CONFIG_CHARGER_PCF50633
+/*
+ * On GTA02 the 1A charger features a 48K resistor to 0V on the ID pin.
+ * We use this to recognize that we can pull 1A from the USB socket.
+ *
+ * These constants are the measured pcf50633 ADC levels with the 1A
+ * charger / 48K resistor, and with no pulldown resistor.
+ */
+
+#define ADC_NOM_CHG_DETECT_1A 6
+#define ADC_NOM_CHG_DETECT_USB 43
+
+static void
+gta02_configure_pmu_for_charger(struct pcf50633 *pcf, void *unused, int res)
+{
+	int  ma;
+
+	/* Interpret charger type */
+	if (res < ((ADC_NOM_CHG_DETECT_USB + ADC_NOM_CHG_DETECT_1A) / 2)) {
+
+		/*
+		 * Sanity - stop GPO driving out now that we have a 1A charger
+		 * GPO controls USB Host power generation on GTA02
+		 */
+		pcf50633_gpio_set(pcf, PCF50633_GPO, 0);
+
+		ma = 1000;
+	} else
+		ma = 100;
+
+	pcf50633_mbc_usb_curlim_set(pcf, ma);
+}
+
+static struct delayed_work gta02_charger_work;
+static int gta02_usb_vbus_draw;
+
+static void gta02_charger_worker(struct work_struct *work)
+{
+	if (gta02_usb_vbus_draw) {
+		pcf50633_mbc_usb_curlim_set(gta02_pcf, gta02_usb_vbus_draw);
+		return;
+	}
+
+#ifdef CONFIG_PCF50633_ADC
+	pcf50633_adc_async_read(gta02_pcf,
+				PCF50633_ADCC1_MUX_ADCIN1,
+				PCF50633_ADCC1_AVERAGE_16,
+				gta02_configure_pmu_for_charger,
+				NULL);
+#else
+	/*
+	 * If the PCF50633 ADC is disabled we fallback to a
+	 * 100mA limit for safety.
+	 */
+	pcf50633_mbc_usb_curlim_set(pcf, 100);
+#endif
+}
+
+#define GTA02_CHARGER_CONFIGURE_TIMEOUT ((3000 * HZ) / 1000)
+
+static void gta02_pmu_event_callback(struct pcf50633 *pcf, int irq)
+{
+	if (irq == PCF50633_IRQ_USBINS) {
+		schedule_delayed_work(&gta02_charger_work,
+				      GTA02_CHARGER_CONFIGURE_TIMEOUT);
+
+		return;
+	}
+
+	if (irq == PCF50633_IRQ_USBREM) {
+		cancel_delayed_work_sync(&gta02_charger_work);
+		gta02_usb_vbus_draw = 0;
+	}
+}
+
+static void gta02_udc_vbus_draw(unsigned int ma)
+{
+	if (!gta02_pcf)
+		return;
+
+	gta02_usb_vbus_draw = ma;
+
+	schedule_delayed_work(&gta02_charger_work,
+			      GTA02_CHARGER_CONFIGURE_TIMEOUT);
+}
+#else /* !CONFIG_CHARGER_PCF50633 */
+#define gta02_pmu_event_callback	NULL
+#define gta02_udc_vbus_draw		NULL
+#endif
+
+/*
+ * This is called when pc50633 is probed, unfortunately quite late in the
+ * day since it is an I2C bus device. Here we can belatedly define some
+ * platform devices with the advantage that we can mark the pcf50633 as the
+ * parent. This makes them get suspended and resumed with their parent
+ * the pcf50633 still around.
+ */
+
+static void gta02_pmu_attach_child_devices(struct pcf50633 *pcf);
+
+
+static char *gta02_batteries[] = {
+	"battery",
+};
+
+struct pcf50633_platform_data gta02_pcf_pdata = {
+	.resumers = {
+		[0] =	PCF50633_INT1_USBINS |
+			PCF50633_INT1_USBREM |
+			PCF50633_INT1_ALARM,
+		[1] =	PCF50633_INT2_ONKEYF,
+		[2] =	PCF50633_INT3_ONKEY1S,
+		[3] =	PCF50633_INT4_LOWSYS |
+			PCF50633_INT4_LOWBAT |
+			PCF50633_INT4_HIGHTMP,
+	},
+
+	.batteries = gta02_batteries,
+	.num_batteries = ARRAY_SIZE(gta02_batteries),
+	.reg_init_data = {
+		[PCF50633_REGULATOR_AUTO] = {
+			.constraints = {
+				.min_uV = 3300000,
+				.max_uV = 3300000,
+				.valid_modes_mask = REGULATOR_MODE_NORMAL,
+				.always_on = 1,
+				.apply_uV = 1,
+				.state_mem = {
+					.enabled = 1,
+				},
+			},
+		},
+		[PCF50633_REGULATOR_DOWN1] = {
+			.constraints = {
+				.min_uV = 1300000,
+				.max_uV = 1600000,
+				.valid_modes_mask = REGULATOR_MODE_NORMAL,
+				.always_on = 1,
+				.apply_uV = 1,
+			},
+		},
+		[PCF50633_REGULATOR_DOWN2] = {
+			.constraints = {
+				.min_uV = 1800000,
+				.max_uV = 1800000,
+				.valid_modes_mask = REGULATOR_MODE_NORMAL,
+				.apply_uV = 1,
+				.always_on = 1,
+				.state_mem = {
+					.enabled = 1,
+				},
+			},
+		},
+		[PCF50633_REGULATOR_HCLDO] = {
+			.constraints = {
+				.min_uV = 2000000,
+				.max_uV = 3300000,
+				.valid_modes_mask = REGULATOR_MODE_NORMAL,
+				.valid_ops_mask = REGULATOR_CHANGE_VOLTAGE,
+				.always_on = 1,
+			},
+		},
+		[PCF50633_REGULATOR_LDO1] = {
+			.constraints = {
+				.min_uV = 3300000,
+				.max_uV = 3300000,
+				.valid_modes_mask = REGULATOR_MODE_NORMAL,
+				.apply_uV = 1,
+				.state_mem = {
+					.enabled = 0,
+				},
+			},
+		},
+		[PCF50633_REGULATOR_LDO2] = {
+			.constraints = {
+				.min_uV = 3300000,
+				.max_uV = 3300000,
+				.valid_modes_mask = REGULATOR_MODE_NORMAL,
+				.apply_uV = 1,
+			},
+		},
+		[PCF50633_REGULATOR_LDO3] = {
+			.constraints = {
+				.min_uV = 3000000,
+				.max_uV = 3000000,
+				.valid_modes_mask = REGULATOR_MODE_NORMAL,
+				.apply_uV = 1,
+			},
+		},
+		[PCF50633_REGULATOR_LDO4] = {
+			.constraints = {
+				.min_uV = 3200000,
+				.max_uV = 3200000,
+				.valid_modes_mask = REGULATOR_MODE_NORMAL,
+				.apply_uV = 1,
+			},
+		},
+		[PCF50633_REGULATOR_LDO5] = {
+			.constraints = {
+				.min_uV = 3000000,
+				.max_uV = 3000000,
+				.valid_modes_mask = REGULATOR_MODE_NORMAL,
+				.apply_uV = 1,
+				.state_mem = {
+					.enabled = 1,
+				},
+			},
+		},
+		[PCF50633_REGULATOR_LDO6] = {
+			.constraints = {
+				.min_uV = 3000000,
+				.max_uV = 3000000,
+				.valid_modes_mask = REGULATOR_MODE_NORMAL,
+			},
+		},
+		[PCF50633_REGULATOR_MEMLDO] = {
+			.constraints = {
+				.min_uV = 1800000,
+				.max_uV = 1800000,
+				.valid_modes_mask = REGULATOR_MODE_NORMAL,
+				.state_mem = {
+					.enabled = 1,
+				},
+			},
+		},
+
+	},
+	.probe_done = gta02_pmu_attach_child_devices,
+	.mbc_event_callback = gta02_pmu_event_callback,
+};
+
+
+/* NOR Flash. */
+
+#define GTA02_FLASH_BASE	0x18000000 /* GCS3 */
+#define GTA02_FLASH_SIZE	0x200000 /* 2MBytes */
+
+static struct physmap_flash_data gta02_nor_flash_data = {
+	.width		= 2,
+};
+
+static struct resource gta02_nor_flash_resource = {
+	.start		= GTA02_FLASH_BASE,
+	.end		= GTA02_FLASH_BASE + GTA02_FLASH_SIZE - 1,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct platform_device gta02_nor_flash = {
+	.name		= "physmap-flash",
+	.id		= 0,
+	.dev		= {
+		.platform_data	= &gta02_nor_flash_data,
+	},
+	.resource	= &gta02_nor_flash_resource,
+	.num_resources	= 1,
+};
+
+
+struct platform_device s3c24xx_pwm_device = {
+	.name		= "s3c24xx_pwm",
+	.num_resources	= 0,
+};
+
+static struct i2c_board_info gta02_i2c_devs[] __initdata = {
+	{
+		I2C_BOARD_INFO("pcf50633", 0x73),
+		.irq = GTA02_IRQ_PCF50633,
+		.platform_data = &gta02_pcf_pdata,
+	},
+	{
+		I2C_BOARD_INFO("wm8753", 0x1a),
+	},
+};
+
+static struct s3c2410_nand_set gta02_nand_sets[] = {
+	[0] = {
+		/*
+		 * This name is also hard-coded in the boot loaders, so
+		 * changing it would would require all users to upgrade
+		 * their boot loaders, some of which are stored in a NOR
+		 * that is considered to be immutable.
+		 */
+		.name		= "neo1973-nand",
+		.nr_chips	= 1,
+		.use_bbt	= 1,
+		.force_soft_ecc	= 1,
+	},
+};
+
+/*
+ * Choose a set of timings derived from S3C@2442B MCP54
+ * data sheet (K5D2G13ACM-D075 MCP Memory).
+ */
+
+static struct s3c2410_platform_nand gta02_nand_info = {
+	.tacls		= 0,
+	.twrph0		= 25,
+	.twrph1		= 15,
+	.nr_sets	= ARRAY_SIZE(gta02_nand_sets),
+	.sets		= gta02_nand_sets,
+};
+
+
+static void gta02_udc_command(enum s3c2410_udc_cmd_e cmd)
+{
+	switch (cmd) {
+	case S3C2410_UDC_P_ENABLE:
+		pr_debug("%s S3C2410_UDC_P_ENABLE\n", __func__);
+		gpio_direction_output(GTA02_GPIO_USB_PULLUP, 1);
+		break;
+	case S3C2410_UDC_P_DISABLE:
+		pr_debug("%s S3C2410_UDC_P_DISABLE\n", __func__);
+		gpio_direction_output(GTA02_GPIO_USB_PULLUP, 0);
+		break;
+	case S3C2410_UDC_P_RESET:
+		pr_debug("%s S3C2410_UDC_P_RESET\n", __func__);
+		/* FIXME: Do something here. */
+	}
+}
+
+/* Get PMU to set USB current limit accordingly. */
+static struct s3c2410_udc_mach_info gta02_udc_cfg = {
+	.vbus_draw	= gta02_udc_vbus_draw,
+	.udc_command	= gta02_udc_command,
+
+};
+
+
+
+static void gta02_bl_set_intensity(int intensity)
+{
+	struct pcf50633 *pcf = gta02_pcf;
+	int old_intensity = pcf50633_reg_read(pcf, PCF50633_REG_LEDOUT);
+
+	/* We map 8-bit intensity to 6-bit intensity in hardware. */
+	intensity >>= 2;
+
+	/*
+	 * This can happen during, eg, print of panic on blanked console,
+	 * but we can't service i2c without interrupts active, so abort.
+	 */
+	if (in_atomic()) {
+		printk(KERN_ERR "gta02_bl_set_intensity called while atomic\n");
+		return;
+	}
+
+	old_intensity = pcf50633_reg_read(pcf, PCF50633_REG_LEDOUT);
+	if (intensity == old_intensity)
+		return;
+
+	/* We can't do this anywhere else. */
+	pcf50633_reg_write(pcf, PCF50633_REG_LEDDIM, 5);
+
+	if (!(pcf50633_reg_read(pcf, PCF50633_REG_LEDENA) & 3))
+		old_intensity = 0;
+
+	/*
+	 * The PCF50633 cannot handle LEDOUT = 0 (datasheet p60)
+	 * if seen, you have to re-enable the LED unit.
+	 */
+	if (!intensity || !old_intensity)
+		pcf50633_reg_write(pcf, PCF50633_REG_LEDENA, 0);
+
+	/* Illegal to set LEDOUT to 0. */
+	if (!intensity)
+		pcf50633_reg_set_bit_mask(pcf, PCF50633_REG_LEDOUT, 0x3f, 2);
+	else
+		pcf50633_reg_set_bit_mask(pcf, PCF50633_REG_LEDOUT, 0x3f,
+					  intensity);
+
+	if (intensity)
+		pcf50633_reg_write(pcf, PCF50633_REG_LEDENA, 2);
+
+}
+
+static struct generic_bl_info gta02_bl_info = {
+	.name			= "gta02-bl",
+	.max_intensity		= 0xff,
+	.default_intensity	= 0xff,
+	.set_bl_intensity	= gta02_bl_set_intensity,
+};
+
+static struct platform_device gta02_bl_dev = {
+	.name			= "generic-bl",
+	.id			= 1,
+	.dev = {
+		.platform_data = &gta02_bl_info,
+	},
+};
+
+
+
+/* USB */
+static struct s3c2410_hcd_info gta02_usb_info = {
+	.port[0]	= {
+		.flags	= S3C_HCDFLG_USED,
+	},
+	.port[1]	= {
+		.flags	= 0,
+	},
+};
+
+
+static void __init gta02_map_io(void)
+{
+	s3c24xx_init_io(gta02_iodesc, ARRAY_SIZE(gta02_iodesc));
+	s3c24xx_init_clocks(12000000);
+	s3c24xx_init_uarts(gta02_uartcfgs, ARRAY_SIZE(gta02_uartcfgs));
+}
+
+
+/* These are the guys that don't need to be children of PMU. */
+
+static struct platform_device *gta02_devices[] __initdata = {
+	&s3c_device_usb,
+	&s3c_device_wdt,
+	&s3c_device_sdi,
+	&s3c_device_usbgadget,
+	&s3c_device_nand,
+	&gta02_nor_flash,
+	&s3c24xx_pwm_device,
+	&s3c_device_iis,
+	&s3c_device_i2c0,
+};
+
+/* These guys DO need to be children of PMU. */
+
+static struct platform_device *gta02_devices_pmu_children[] = {
+	&gta02_bl_dev,
+};
+
+
+/*
+ * This is called when pc50633 is probed, quite late in the day since it is an
+ * I2C bus device.  Here we can define platform devices with the advantage that
+ * we can mark the pcf50633 as the parent.  This makes them get suspended and
+ * resumed with their parent the pcf50633 still around.  All devices whose
+ * operation depends on something from pcf50633 must have this relationship
+ * made explicit like this, or suspend and resume will become an unreliable
+ * hellworld.
+ */
+
+static void gta02_pmu_attach_child_devices(struct pcf50633 *pcf)
+{
+	int n;
+
+	/* Grab a copy of the now probed PMU pointer. */
+	gta02_pcf = pcf;
+
+	for (n = 0; n < ARRAY_SIZE(gta02_devices_pmu_children); n++)
+		gta02_devices_pmu_children[n]->dev.parent = pcf->dev;
+
+	platform_add_devices(gta02_devices_pmu_children,
+			     ARRAY_SIZE(gta02_devices_pmu_children));
+}
+
+static void gta02_poweroff(void)
+{
+	pcf50633_reg_set_bit_mask(gta02_pcf, PCF50633_REG_OOCSHDWN, 1, 1);
+}
+
+static void __init gta02_machine_init(void)
+{
+	/* Set the panic callback to make AUX LED blink at ~5Hz. */
+	panic_blink = gta02_panic_blink;
+
+	s3c_pm_init();
+
+#ifdef CONFIG_CHARGER_PCF50633
+	INIT_DELAYED_WORK(&gta02_charger_work, gta02_charger_worker);
+#endif
+
+	s3c_device_usb.dev.platform_data = &gta02_usb_info;
+	s3c_device_nand.dev.platform_data = &gta02_nand_info;
+
+	s3c24xx_udc_set_platdata(&gta02_udc_cfg);
+	s3c_i2c0_set_platdata(NULL);
+
+	i2c_register_board_info(0, gta02_i2c_devs, ARRAY_SIZE(gta02_i2c_devs));
+
+	platform_add_devices(gta02_devices, ARRAY_SIZE(gta02_devices));
+	pm_power_off = gta02_poweroff;
+}
+
+
+MACHINE_START(NEO1973_GTA02, "GTA02")
+	/* Maintainer: Nelson Castillo <arhuaco@freaks-unidos.net> */
+	.phys_io	= S3C2410_PA_UART,
+	.io_pg_offst	= (((u32)S3C24XX_VA_UART) >> 18) & 0xfffc,
+	.boot_params	= S3C2410_SDRAM_PA + 0x100,
+	.map_io		= gta02_map_io,
+	.init_irq	= s3c24xx_init_irq,
+	.init_machine	= gta02_machine_init,
+	.timer		= &s3c24xx_timer,
+MACHINE_END
-- 
cgit v1.2.3-59-g8ed1b


From 5d77ddfbcb062f2617ea79d7a371b4bc78f28417 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Tue, 16 Jun 2009 04:19:38 +0300
Subject: nfsd41: sanity check client drc maxreqs

Ensure the client requested maximum requests are between 1 and
NFSD_MAX_SLOTS_PER_SESSION

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4state.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index d5caf2a709d2..99570c49add5 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -425,6 +425,11 @@ static int set_forechannel_maxreqs(struct nfsd4_channel_attrs *fchan)
 {
 	int status = 0, np = fchan->maxreqs * NFSD_PAGES_PER_SLOT;
 
+	if (fchan->maxreqs < 1)
+		return nfserr_inval;
+	else if (fchan->maxreqs > NFSD_MAX_SLOTS_PER_SESSION)
+		fchan->maxreqs = NFSD_MAX_SLOTS_PER_SESSION;
+
 	spin_lock(&nfsd_serv->sv_lock);
 	if (np + nfsd_serv->sv_drc_pages_used > nfsd_serv->sv_drc_max_pages)
 		np = nfsd_serv->sv_drc_max_pages - nfsd_serv->sv_drc_pages_used;
-- 
cgit v1.2.3-59-g8ed1b


From 1efae38140546db403845d628db9f2d608caa87e Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:21:46 -0400
Subject: nfs41: Add Kconfig symbols for NFSv4.1

Added CONFIG_NFS_V4_1 and made it depend upon CONFIG_NFS_V4 and EXPERIMENTAL.
Indicate that CONFIG_NFS_V4_1 is for NFS developers at the moment

At the moment we're expecting folks trying out nfs41 to
actively participate in the development process by helping us
debug issues and ideally send patches to fix problems.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/Kconfig | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig
index e67f3ec07736..5d6d6f415935 100644
--- a/fs/nfs/Kconfig
+++ b/fs/nfs/Kconfig
@@ -74,6 +74,15 @@ config NFS_V4
 
 	  If unsure, say N.
 
+config NFS_V4_1
+	bool "NFS client support for NFSv4.1 (DEVELOPER ONLY)"
+	depends on NFS_V4 && EXPERIMENTAL
+	help
+	  This option enables support for minor version 1 of the NFSv4 protocol
+	  (draft-ietf-nfsv4-minorversion1) in the kernel's NFS client.
+
+	  Unless you're an NFS developer, say N.
+
 config ROOT_NFS
 	bool "Root file system on NFS"
 	depends on NFS_FS=y && IP_PNP
-- 
cgit v1.2.3-59-g8ed1b


From 44549dff82753b6a5ffabcefeead34be63e95d96 Mon Sep 17 00:00:00 2001
From: Mike Sager <sager@netapp.com>
Date: Wed, 1 Apr 2009 09:21:47 -0400
Subject: nfs41: define NFS4_MAX_MINOR_VERSION based on CONFIG_NFS_V4_1

If 4.1 isn't supported, NFS4_MAX_MINOR_VERSION will be 0.

Signed-off-by: Mike Sager <sager@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/nfs4.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index e3f0cbcbd0db..7c36fcf2dfb7 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -462,6 +462,13 @@ enum lock_type4 {
 #define NFSPROC4_NULL 0
 #define NFSPROC4_COMPOUND 1
 #define NFS4_MINOR_VERSION 0
+
+#if defined(CONFIG_NFS_V4_1)
+#define NFS4_MAX_MINOR_VERSION 1
+#else
+#define NFS4_MAX_MINOR_VERSION 0
+#endif /* CONFIG_NFS_V4_1 */
+
 #define NFS4_DEBUG 1
 
 /* Index of predefined Linux client operations */
-- 
cgit v1.2.3-59-g8ed1b


From 3fd5be9e19921a89d9ed78d6a708a379a6c3c76a Mon Sep 17 00:00:00 2001
From: Mike Sager <sager@netapp.com>
Date: Wed, 1 Apr 2009 09:21:48 -0400
Subject: nfs41: add mount command option minorversion

mount -t nfs4 -o minorversion=[0|1] specifies whether to use 4.0 or 4.1.
By default, the minorversion is set to 0.

Signed-off-by: Mike Sager <sager@netapp.com>
[set default minorversion to 0 as per Trond and SteveD's request]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/internal.h |  1 +
 fs/nfs/super.c    | 10 ++++++++++
 2 files changed, 11 insertions(+)

diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index e4d6a8348adf..ffa6bd54d439 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -44,6 +44,7 @@ struct nfs_parsed_mount_data {
 	unsigned int		auth_flavor_len;
 	rpc_authflavor_t	auth_flavors[1];
 	char			*client_address;
+	unsigned int		minorversion;
 	char			*fscache_uniq;
 
 	struct {
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index d2d67781c579..5a8fdc791cc1 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -90,6 +90,7 @@ enum {
 	Opt_mountport,
 	Opt_mountvers,
 	Opt_nfsvers,
+	Opt_minorversion,
 
 	/* Mount options that take string arguments */
 	Opt_sec, Opt_proto, Opt_mountproto, Opt_mounthost,
@@ -155,6 +156,7 @@ static const match_table_t nfs_mount_option_tokens = {
 	{ Opt_mountvers, "mountvers=%u" },
 	{ Opt_nfsvers, "nfsvers=%u" },
 	{ Opt_nfsvers, "vers=%u" },
+	{ Opt_minorversion, "minorversion=%u" },
 
 	{ Opt_sec, "sec=%s" },
 	{ Opt_proto, "proto=%s" },
@@ -1211,6 +1213,13 @@ static int nfs_parse_mount_options(char *raw,
 				nfs_parse_invalid_value("nfsvers");
 			}
 			break;
+		case Opt_minorversion:
+			if (match_int(args, &option))
+				return 0;
+			if (option < 0 || option > NFS4_MAX_MINOR_VERSION)
+				return 0;
+			mnt->minorversion = option;
+			break;
 
 		/*
 		 * options that take text values
@@ -2261,6 +2270,7 @@ static int nfs4_validate_mount_data(void *options,
 	args->nfs_server.port	= NFS_PORT; /* 2049 unless user set port= */
 	args->auth_flavors[0]	= RPC_AUTH_UNIX;
 	args->auth_flavor_len	= 0;
+	args->minorversion	= 0;
 
 	switch (data->version) {
 	case 1:
-- 
cgit v1.2.3-59-g8ed1b


From 94a417f3d7a02478a2d7842e693a61339fb54ea4 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:21:49 -0400
Subject: nfs41: nfs_client.cl_minorversion

This field is set to the nfsv4 minor version for this mount.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>

Note: This patch sets the referral to the same minorversion as the
current mount. Revisit in future patch.

Signed-off-by: Andy Adamson <andros@netapp.com>
[removed cl_minorversion assignment in nfs_set_client]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[always define nfs_client.cl_minorversion]
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c           | 9 ++++++---
 include/linux/nfs_fs_sb.h | 3 ++-
 2 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 75c9cd2aa119..0efcb55c2caa 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -1101,7 +1101,8 @@ static int nfs4_set_client(struct nfs_server *server,
 		const size_t addrlen,
 		const char *ip_addr,
 		rpc_authflavor_t authflavour,
-		int proto, const struct rpc_timeout *timeparms)
+		int proto, const struct rpc_timeout *timeparms,
+		u32 minorversion)
 {
 	struct nfs_client_initdata cl_init = {
 		.hostname = hostname,
@@ -1164,7 +1165,8 @@ static int nfs4_init_server(struct nfs_server *server,
 			data->client_address,
 			data->auth_flavors[0],
 			data->nfs_server.protocol,
-			&timeparms);
+			&timeparms,
+			data->minorversion);
 	if (error < 0)
 		goto error;
 
@@ -1282,7 +1284,8 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data,
 				parent_client->cl_ipaddr,
 				data->authflavor,
 				parent_server->client->cl_xprt->prot,
-				parent_server->client->cl_timeout);
+				parent_server->client->cl_timeout,
+				parent_client->cl_minorversion);
 	if (error < 0)
 		goto error;
 
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 6ad75948cbf7..e9a51fe46aa3 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -32,6 +32,7 @@ struct nfs_client {
 	const struct nfs_rpc_ops *rpc_ops;	/* NFS protocol vector */
 	int			cl_proto;	/* Network transport protocol */
 
+	u32			cl_minorversion;/* NFSv4 minorversion */
 	struct rpc_cred		*cl_machine_cred;
 
 #ifdef CONFIG_NFS_V4
@@ -63,7 +64,7 @@ struct nfs_client {
 	 */
 	char			cl_ipaddr[48];
 	unsigned char		cl_id_uniquifier;
-#endif
+#endif /* CONFIG_NFS_V4 */
 
 #ifdef CONFIG_NFS_FSCACHE
 	struct fscache_cookie	*fscache;	/* client index cache cookie */
-- 
cgit v1.2.3-59-g8ed1b


From 5aae4a9ae0dd55409a42ca61b82ef1f5a840091e Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:21:50 -0400
Subject: nfs41: Use mount minorversion option

Use the mount minorversion option to initialize the nfs_client cl_minorversion
and match it in nfs_match_client() when looking up a nfs_client.

[nfs41: remove ifdefs around nfs_client_initdata.minorversion]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 0efcb55c2caa..a736160046c3 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -102,6 +102,7 @@ struct nfs_client_initdata {
 	size_t addrlen;
 	const struct nfs_rpc_ops *rpc_ops;
 	int proto;
+	u32 minorversion;
 };
 
 /*
@@ -150,6 +151,7 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 	rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS client");
 	clp->cl_boot_time = CURRENT_TIME;
 	clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED;
+	clp->cl_minorversion = cl_init->minorversion;
 #endif
 	cred = rpc_lookup_machine_cred();
 	if (!IS_ERR(cred))
@@ -420,7 +422,9 @@ static struct nfs_client *nfs_match_client(const struct nfs_client_initdata *dat
 
 		if (clp->cl_proto != data->proto)
 			continue;
-
+		/* Match nfsv4 minorversion */
+		if (clp->cl_minorversion != data->minorversion)
+			continue;
 		/* Match the full socket address */
 		if (!nfs_sockaddr_cmp(sap, clap))
 			continue;
@@ -1110,6 +1114,7 @@ static int nfs4_set_client(struct nfs_server *server,
 		.addrlen = addrlen,
 		.rpc_ops = &nfs_v4_clientops,
 		.proto = proto,
+		.minorversion = minorversion,
 	};
 	struct nfs_client *clp;
 	int error;
-- 
cgit v1.2.3-59-g8ed1b


From c2e713dd83dcabb2ae951ea572e7de68c96e2d48 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:21:51 -0400
Subject: nfs41: translate NFS4ERR_MINOR_VERS_MISMATCH to EPROTONOSUPPORT

To be returned to the mount command when trying to mount a v4 server
using minorversion 1.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4state.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 0298e909559f..bc683ed477e1 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1107,6 +1107,8 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 			nfs4_clear_machine_cred(clp);
 			status = -EAGAIN;
 		}
+		if (status == -NFS4ERR_MINOR_VERS_MISMATCH)
+			status = -EPROTONOSUPPORT;
 	}
 	return status;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 9ff71c3a9827b99699510076dffa0bbe7c36bfd4 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:21:52 -0400
Subject: nfs41: client xdr definitions

Define stubs for sequence args and res data structures and embed
them in all other nfs4 and nfs41 xdr types.  They are needed for
sending any op in a nfs41 compound rpc.

Signed-off-by: Andy Adamson<andros@netapp.com>
[moved new args/res definitions away, to where they're first used]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/nfs_xdr.h | 52 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 52 insertions(+)

diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index b89c34e40bc2..f2c5700c7b6e 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -145,6 +145,15 @@ struct nfs4_change_info {
 };
 
 struct nfs_seqid;
+
+struct nfs4_sequence_args {
+	/* stub */
+};
+
+struct nfs4_sequence_res {
+	/* stub */
+};
+
 /*
  * Arguments to the open call.
  */
@@ -165,6 +174,7 @@ struct nfs_openargs {
 	const struct nfs_server *server;	 /* Needed for ID mapping */
 	const u32 *		bitmask;
 	__u32			claim;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_openres {
@@ -181,6 +191,7 @@ struct nfs_openres {
 	__u32			do_recall;
 	__u64			maxsize;
 	__u32			attrset[NFS4_BITMAP_SIZE];
+	struct nfs4_sequence_res	seq_res;
 };
 
 /*
@@ -206,6 +217,7 @@ struct nfs_closeargs {
 	struct nfs_seqid *	seqid;
 	fmode_t			fmode;
 	const u32 *		bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_closeres {
@@ -213,6 +225,7 @@ struct nfs_closeres {
 	struct nfs_fattr *	fattr;
 	struct nfs_seqid *	seqid;
 	const struct nfs_server *server;
+	struct nfs4_sequence_res	seq_res;
 };
 /*
  *  * Arguments to the lock,lockt, and locku call.
@@ -233,12 +246,14 @@ struct nfs_lock_args {
 	unsigned char		block : 1;
 	unsigned char		reclaim : 1;
 	unsigned char		new_lock_owner : 1;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_lock_res {
 	nfs4_stateid		stateid;
 	struct nfs_seqid *	lock_seqid;
 	struct nfs_seqid *	open_seqid;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs_locku_args {
@@ -246,32 +261,38 @@ struct nfs_locku_args {
 	struct file_lock *	fl;
 	struct nfs_seqid *	seqid;
 	nfs4_stateid *		stateid;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_locku_res {
 	nfs4_stateid		stateid;
 	struct nfs_seqid *	seqid;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs_lockt_args {
 	struct nfs_fh *		fh;
 	struct file_lock *	fl;
 	struct nfs_lowner	lock_owner;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_lockt_res {
 	struct file_lock *	denied; /* LOCK, LOCKT failed */
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_delegreturnargs {
 	const struct nfs_fh *fhandle;
 	const nfs4_stateid *stateid;
 	const u32 * bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_delegreturnres {
 	struct nfs_fattr * fattr;
 	const struct nfs_server *server;
+	struct nfs4_sequence_res	seq_res;
 };
 
 /*
@@ -284,12 +305,14 @@ struct nfs_readargs {
 	__u32			count;
 	unsigned int		pgbase;
 	struct page **		pages;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_readres {
 	struct nfs_fattr *	fattr;
 	__u32			count;
 	int                     eof;
+	struct nfs4_sequence_res	seq_res;
 };
 
 /*
@@ -304,6 +327,7 @@ struct nfs_writeargs {
 	unsigned int		pgbase;
 	struct page **		pages;
 	const u32 *		bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_writeverf {
@@ -316,6 +340,7 @@ struct nfs_writeres {
 	struct nfs_writeverf *	verf;
 	__u32			count;
 	const struct nfs_server *server;
+	struct nfs4_sequence_res	seq_res;
 };
 
 /*
@@ -325,12 +350,14 @@ struct nfs_removeargs {
 	const struct nfs_fh	*fh;
 	struct qstr		name;
 	const u32 *		bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_removeres {
 	const struct nfs_server *server;
 	struct nfs4_change_info	cinfo;
 	struct nfs_fattr	dir_attr;
+	struct nfs4_sequence_res 	seq_res;
 };
 
 /*
@@ -383,6 +410,7 @@ struct nfs_setattrargs {
 	struct iattr *                  iap;
 	const struct nfs_server *	server; /* Needed for name mapping */
 	const u32 *			bitmask;
+	struct nfs4_sequence_args 	seq_args;
 };
 
 struct nfs_setaclargs {
@@ -390,6 +418,7 @@ struct nfs_setaclargs {
 	size_t				acl_len;
 	unsigned int			acl_pgbase;
 	struct page **			acl_pages;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs_getaclargs {
@@ -397,11 +426,13 @@ struct nfs_getaclargs {
 	size_t				acl_len;
 	unsigned int			acl_pgbase;
 	struct page **			acl_pages;
+	struct nfs4_sequence_args 	seq_args;
 };
 
 struct nfs_setattrres {
 	struct nfs_fattr *              fattr;
 	const struct nfs_server *	server;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs_linkargs {
@@ -583,6 +614,7 @@ struct nfs4_accessargs {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;
 	u32				access;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_accessres {
@@ -590,6 +622,7 @@ struct nfs4_accessres {
 	struct nfs_fattr *		fattr;
 	u32				supported;
 	u32				access;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_create_arg {
@@ -609,6 +642,7 @@ struct nfs4_create_arg {
 	const struct iattr *		attrs;
 	const struct nfs_fh *		dir_fh;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args 	seq_args;
 };
 
 struct nfs4_create_res {
@@ -617,21 +651,25 @@ struct nfs4_create_res {
 	struct nfs_fattr *		fattr;
 	struct nfs4_change_info		dir_cinfo;
 	struct nfs_fattr *		dir_fattr;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_fsinfo_arg {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_getattr_arg {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_getattr_res {
 	const struct nfs_server *	server;
 	struct nfs_fattr *		fattr;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_link_arg {
@@ -639,6 +677,7 @@ struct nfs4_link_arg {
 	const struct nfs_fh *		dir_fh;
 	const struct qstr *		name;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args 	seq_args;
 };
 
 struct nfs4_link_res {
@@ -646,6 +685,7 @@ struct nfs4_link_res {
 	struct nfs_fattr *		fattr;
 	struct nfs4_change_info		cinfo;
 	struct nfs_fattr *		dir_attr;
+	struct nfs4_sequence_res	seq_res;
 };
 
 
@@ -653,21 +693,25 @@ struct nfs4_lookup_arg {
 	const struct nfs_fh *		dir_fh;
 	const struct qstr *		name;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_lookup_res {
 	const struct nfs_server *	server;
 	struct nfs_fattr *		fattr;
 	struct nfs_fh *			fh;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_lookup_root_arg {
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_pathconf_arg {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_readdir_arg {
@@ -678,11 +722,13 @@ struct nfs4_readdir_arg {
 	struct page **			pages;	/* zero-copy data */
 	unsigned int			pgbase;	/* zero-copy data */
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_readdir_res {
 	nfs4_verifier			verifier;
 	unsigned int			pgbase;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_readlink {
@@ -690,6 +736,7 @@ struct nfs4_readlink {
 	unsigned int			pgbase;
 	unsigned int			pglen;   /* zero-copy data */
 	struct page **			pages;   /* zero-copy data */
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_rename_arg {
@@ -698,6 +745,7 @@ struct nfs4_rename_arg {
 	const struct qstr *		old_name;
 	const struct qstr *		new_name;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_rename_res {
@@ -706,6 +754,7 @@ struct nfs4_rename_res {
 	struct nfs_fattr *		old_fattr;
 	struct nfs4_change_info		new_cinfo;
 	struct nfs_fattr *		new_fattr;
+	struct nfs4_sequence_res	seq_res;
 };
 
 #define NFS4_SETCLIENTID_NAMELEN	(127)
@@ -724,6 +773,7 @@ struct nfs4_setclientid {
 struct nfs4_statfs_arg {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 struct nfs4_server_caps_res {
@@ -731,6 +781,7 @@ struct nfs4_server_caps_res {
 	u32				acl_bitmask;
 	u32				has_links;
 	u32				has_symlinks;
+	struct nfs4_sequence_res	seq_res;
 };
 
 struct nfs4_string {
@@ -765,6 +816,7 @@ struct nfs4_fs_locations_arg {
 	const struct qstr *name;
 	struct page *page;
 	const u32 *bitmask;
+	struct nfs4_sequence_args	seq_args;
 };
 
 #endif /* CONFIG_NFS_V4 */
-- 
cgit v1.2.3-59-g8ed1b


From 557134a39c8d2ab79d8b8d53438e03e29feb5ec4 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:21:53 -0400
Subject: nfs41: sessions client infrastructure

NFSv4.1 Sessions basic data types, initialization, and destruction.

The session is always associated with a struct nfs_client that holds
the exchange_id results.

Signed-off-by: Rahul Iyer <iyer@netapp.com>
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[remove extraneous rpc_clnt pointer, use the struct nfs_client cl_rpcclient.
remove the rpc_clnt parameter from nfs4 nfs4_init_session]
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Use the presence of a session to determine behaviour instead of the
minorversion number.]
Signed-off-by: Andy Adamson <andros@netapp.com>
[constified nfs4_has_session's struct nfs_client parameter]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Rename nfs4_put_session() to nfs4_destroy_session() and call it from nfs4_free_client() not nfs4_free_server().
Also get rid of nfs4_get_session() and the ref_count in nfs4_session struct as keeping track of nfs_client should be sufficient]
Signed-off-by: Alexandros Batsakis <Alexandros.Batsakis@netapp.com>
[nfs41: pass rsize and wsize into nfs4_init_session]
Signed-off-by: Andy Adamson <andros@netapp.com>
[separated out removal of rpc_clnt parameter from nfs4_init_session ot a
 patch of its own]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Pass the nfs_client pointer into nfs4_alloc_session]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: don't assign to session->clp->cl_session in nfs4_destroy_session]
[nfs41: fixup nfs4_clear_client_minor_version]
[introduce nfs4_clear_client_minor_version() in this patch]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Refactor nfs4_init_session]
    Moved session allocation into nfs4_init_client_minor_version, called from
    nfs4_init_client.
    Leave rwise and wsize initialization in nfs4_init_session, called from
    nfs4_init_server.
    Reverted moving of nfs_fsid definition to nfs_fs_sb.h
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Move NFS4_MAX_SLOT_TABLE define from under CONFIG_NFS_V4_1]
[Fix comile error when CONFIG_NFS_V4_1 is not set.]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[moved nfs4_init_slot_table definition to "create_session operation"]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: alloc session with GFP_KERNEL]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c           | 60 +++++++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/internal.h         | 12 ++++++++++
 fs/nfs/nfs4_fs.h          |  4 ++++
 fs/nfs/nfs4proc.c         | 34 +++++++++++++++++++++++++++
 include/linux/nfs_fs_sb.h | 49 ++++++++++++++++++++++++++++++++++++++
 include/linux/nfs_xdr.h   | 15 ++++++++++++
 6 files changed, 174 insertions(+)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index a736160046c3..f1506f148521 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -183,6 +183,20 @@ static void nfs4_shutdown_client(struct nfs_client *clp)
 #endif
 }
 
+/*
+ * Clears/puts all minor version specific parts from an nfs_client struct
+ * reverting it to minorversion 0.
+ */
+static void nfs4_clear_client_minor_version(struct nfs_client *clp)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (nfs4_has_session(clp)) {
+		nfs4_destroy_session(clp->cl_session);
+		clp->cl_session = NULL;
+	}
+#endif /* CONFIG_NFS_V4_1 */
+}
+
 /*
  * Destroy a shared client record
  */
@@ -190,6 +204,7 @@ static void nfs_free_client(struct nfs_client *clp)
 {
 	dprintk("--> nfs_free_client(%u)\n", clp->rpc_ops->version);
 
+	nfs4_clear_client_minor_version(clp);
 	nfs4_shutdown_client(clp);
 
 	nfs_fscache_release_client_cookie(clp);
@@ -1053,6 +1068,30 @@ error:
 }
 
 #ifdef CONFIG_NFS_V4
+/*
+ * Initialize the minor version specific parts of an NFS4 client record
+ */
+static int nfs4_init_client_minor_version(struct nfs_client *clp)
+{
+#if defined(CONFIG_NFS_V4_1)
+	if (clp->cl_minorversion) {
+		struct nfs4_session *session = NULL;
+		/*
+		 * Create the session and mark it expired.
+		 * When a SEQUENCE operation encounters the expired session
+		 * it will do session recovery to initialize it.
+		 */
+		session = nfs4_alloc_session(clp);
+		if (!session)
+			return -ENOMEM;
+
+		clp->cl_session = session;
+	}
+#endif /* CONFIG_NFS_V4_1 */
+
+	return 0;
+}
+
 /*
  * Initialise an NFS4 client record
  */
@@ -1087,6 +1126,10 @@ static int nfs4_init_client(struct nfs_client *clp,
 	}
 	__set_bit(NFS_CS_IDMAP, &clp->cl_res_state);
 
+	error = nfs4_init_client_minor_version(clp);
+	if (error < 0)
+		goto error;
+
 	nfs_mark_client_ready(clp, NFS_CS_READY);
 	return 0;
 
@@ -1143,6 +1186,21 @@ error:
 	return error;
 }
 
+/*
+ * Initialize a session.
+ * Note: save the mount rsize and wsize for create_server negotiation.
+ */
+static void nfs4_init_session(struct nfs_client *clp,
+			      unsigned int wsize, unsigned int rsize)
+{
+#if defined(CONFIG_NFS_V4_1)
+	if (nfs4_has_session(clp)) {
+		clp->cl_session->fc_attrs.max_rqst_sz = wsize;
+		clp->cl_session->fc_attrs.max_resp_sz = rsize;
+	}
+#endif /* CONFIG_NFS_V4_1 */
+}
+
 /*
  * Create a version 4 volume record
  */
@@ -1221,6 +1279,8 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
 	BUG_ON(!server->nfs_client->rpc_ops);
 	BUG_ON(!server->nfs_client->rpc_ops->file_inode_ops);
 
+	nfs4_init_session(server->nfs_client, server->wsize, server->rsize);
+
 	/* Probe the root fh to retrieve its FSID */
 	error = nfs4_path_walk(server, mntfh, data->nfs_server.export_path);
 	if (error < 0)
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index ffa6bd54d439..7cef45db9257 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -206,6 +206,18 @@ extern int nfs4_path_walk(struct nfs_server *server,
 			  const char *path);
 #endif
 
+/*
+ * Determine if sessions are in use.
+ */
+static inline int nfs4_has_session(const struct nfs_client *clp)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (clp->cl_session)
+		return 1;
+#endif /* CONFIG_NFS_V4_1 */
+	return 0;
+}
+
 /*
  * Determine the device name as a string
  */
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 84345deab26f..acac6f8c3d39 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -202,6 +202,10 @@ extern int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 
 extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops;
 extern struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops;
+#if defined(CONFIG_NFS_V4_1)
+extern void nfs4_destroy_session(struct nfs4_session *session);
+extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp);
+#endif /* CONFIG_NFS_V4_1 */
 
 extern const u32 nfs4_fattr_bitmap[2];
 extern const u32 nfs4_statfs_bitmap[2];
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 4674f8092da8..cdd8e74c47d0 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3723,6 +3723,40 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 	return status;
 }
 
+#ifdef CONFIG_NFS_V4_1
+/* Destroy the slot table */
+static void nfs4_destroy_slot_table(struct nfs4_session *session)
+{
+	if (session->fc_slot_table.slots == NULL)
+		return;
+	kfree(session->fc_slot_table.slots);
+	session->fc_slot_table.slots = NULL;
+	return;
+}
+
+struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
+{
+	struct nfs4_session *session;
+	struct nfs4_slot_table *tbl;
+
+	session = kzalloc(sizeof(struct nfs4_session), GFP_KERNEL);
+	if (!session)
+		return NULL;
+	tbl = &session->fc_slot_table;
+	spin_lock_init(&tbl->slot_tbl_lock);
+	rpc_init_wait_queue(&tbl->slot_tbl_waitq, "Slot table");
+	session->clp = clp;
+	return session;
+}
+
+void nfs4_destroy_session(struct nfs4_session *session)
+{
+	nfs4_destroy_slot_table(session);
+	kfree(session);
+}
+
+#endif /* CONFIG_NFS_V4_1 */
+
 struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = {
 	.owner_flag_bit = NFS_OWNER_RECLAIM_REBOOT,
 	.state_flag_bit	= NFS_STATE_RECLAIM_REBOOT,
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index e9a51fe46aa3..b47c0fc55d42 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -4,9 +4,12 @@
 #include <linux/list.h>
 #include <linux/backing-dev.h>
 #include <linux/wait.h>
+#include <linux/nfs_xdr.h>
+#include <linux/sunrpc/xprt.h>
 
 #include <asm/atomic.h>
 
+struct nfs4_session;
 struct nfs_iostats;
 struct nlm_host;
 
@@ -66,6 +69,10 @@ struct nfs_client {
 	unsigned char		cl_id_uniquifier;
 #endif /* CONFIG_NFS_V4 */
 
+#ifdef CONFIG_NFS_V4_1
+	struct nfs4_session	*cl_session; 	/* sharred session */
+#endif /* CONFIG_NFS_V4_1 */
+
 #ifdef CONFIG_NFS_FSCACHE
 	struct fscache_cookie	*fscache;	/* client index cache cookie */
 #endif
@@ -146,4 +153,46 @@ struct nfs_server {
 #define NFS_CAP_ACLS		(1U << 3)
 #define NFS_CAP_ATOMIC_OPEN	(1U << 4)
 
+
+/* maximum number of slots to use */
+#define NFS4_MAX_SLOT_TABLE RPC_MAX_SLOT_TABLE
+
+#if defined(CONFIG_NFS_V4_1)
+
+/* Sessions */
+#define SLOT_TABLE_SZ (NFS4_MAX_SLOT_TABLE/(8*sizeof(long)))
+struct nfs4_slot_table {
+	struct nfs4_slot *slots;		/* seqid per slot */
+	unsigned long   used_slots[SLOT_TABLE_SZ]; /* used/unused bitmap */
+	spinlock_t	slot_tbl_lock;
+	struct rpc_wait_queue	slot_tbl_waitq;	/* allocators may wait here */
+	int		max_slots;		/* # slots in table */
+	int		highest_used_slotid;	/* sent to server on each SEQ.
+						 * op for dynamic resizing */
+};
+
+static inline int slot_idx(struct nfs4_slot_table *tbl, struct nfs4_slot *sp)
+{
+	return sp - tbl->slots;
+}
+
+/*
+ * Session related parameters
+ */
+struct nfs4_session {
+	struct nfs4_sessionid		sess_id;
+	u32				flags;
+	unsigned long			session_state;
+	u32				hash_alg;
+	u32				ssv_len;
+
+	/* The fore and back channel */
+	struct nfs4_channel_attrs	fc_attrs;
+	struct nfs4_slot_table		fc_slot_table;
+	struct nfs4_channel_attrs	bc_attrs;
+					/* back channel has one slot */
+	struct nfs_client		*clp;
+};
+
+#endif /* CONFIG_NFS_V4_1 */
 #endif
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index f2c5700c7b6e..8f8c026c558e 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -146,6 +146,21 @@ struct nfs4_change_info {
 
 struct nfs_seqid;
 
+/* nfs41 sessions channel attributes */
+struct nfs4_channel_attrs {
+	u32			headerpadsz;
+	u32			max_rqst_sz;
+	u32			max_resp_sz;
+	u32			max_resp_sz_cached;
+	u32			max_ops;
+	u32			max_reqs;
+};
+
+/* nfs41 sessions slot seqid */
+struct nfs4_slot {
+	u32		 	seq_nr;
+};
+
 struct nfs4_sequence_args {
 	/* stub */
 };
-- 
cgit v1.2.3-59-g8ed1b


From 43652ad55342d9146d8035932101a5814b22315a Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:21:54 -0400
Subject: nfs41: use nfs4_server_caps_arg

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[define nfs4_server_caps_arg]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 5 ++++-
 fs/nfs/nfs4xdr.c        | 5 +++--
 include/linux/nfs_xdr.h | 5 +++++
 3 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index cdd8e74c47d0..5ef1022b7e6f 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1584,10 +1584,13 @@ void nfs4_close_context(struct nfs_open_context *ctx, int is_sync)
 
 static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle)
 {
+	struct nfs4_server_caps_arg args = {
+		.fhandle = fhandle,
+	};
 	struct nfs4_server_caps_res res = {};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SERVER_CAPS],
-		.rpc_argp = fhandle,
+		.rpc_argp = &args,
 		.rpc_resp = &res,
 	};
 	int status;
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 1690f0e44b91..91305861037a 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -1900,7 +1900,8 @@ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, __be32 *p, const struct nfs
 /*
  * GETATTR_BITMAP request
  */
-static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p, const struct nfs_fh *fhandle)
+static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p,
+				    struct nfs4_server_caps_arg *args)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
@@ -1909,7 +1910,7 @@ static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p, const struc
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, &hdr);
-	encode_putfh(&xdr, fhandle, &hdr);
+	encode_putfh(&xdr, args->fhandle, &hdr);
 	encode_getattr_one(&xdr, FATTR4_WORD0_SUPPORTED_ATTRS|
 			   FATTR4_WORD0_LINK_SUPPORT|
 			   FATTR4_WORD0_SYMLINK_SUPPORT|
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 8f8c026c558e..a7b7f2a059cc 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -791,6 +791,11 @@ struct nfs4_statfs_arg {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs4_server_caps_arg {
+	struct nfs_fh		       *fhandle;
+	struct nfs4_sequence_args	seq_args;
+};
+
 struct nfs4_server_caps_res {
 	u32				attr_bitmask[2];
 	u32				acl_bitmask;
-- 
cgit v1.2.3-59-g8ed1b


From f50c7000817e7cb4e676ac5d911a82c0f3fd226f Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:21:55 -0400
Subject: nfs41: use nfs4_readlink_res

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[define nfs4_readlink_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 3 ++-
 fs/nfs/nfs4xdr.c        | 3 ++-
 include/linux/nfs_xdr.h | 4 ++++
 3 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 5ef1022b7e6f..b0ec8ff96eb7 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1960,10 +1960,11 @@ static int _nfs4_proc_readlink(struct inode *inode, struct page *page,
 		.pglen    = pglen,
 		.pages    = &page,
 	};
+	struct nfs4_readlink_res res;
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READLINK],
 		.rpc_argp = &args,
-		.rpc_resp = NULL,
+		.rpc_resp = &res,
 	};
 
 	return rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 91305861037a..1e41420916ad 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4235,7 +4235,8 @@ out:
 /*
  * Decode READLINK response
  */
-static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, __be32 *p, void *res)
+static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, __be32 *p,
+				 struct nfs4_readlink_res *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index a7b7f2a059cc..f71260aeb803 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -754,6 +754,10 @@ struct nfs4_readlink {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs4_readlink_res {
+	struct nfs4_sequence_res	seq_res;
+};
+
 struct nfs4_rename_arg {
 	const struct nfs_fh *		old_dir;
 	const struct nfs_fh *		new_dir;
-- 
cgit v1.2.3-59-g8ed1b


From 24ad148a0ff74b1e703a8bc5b3e0793dc7d4e3a9 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:21:56 -0400
Subject: nfs41: use nfs4_statfs_res

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[define nfs4_statfs_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 5 ++++-
 fs/nfs/nfs4xdr.c        | 5 +++--
 include/linux/nfs_xdr.h | 5 +++++
 3 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index b0ec8ff96eb7..b715f6057611 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2426,10 +2426,13 @@ static int _nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle,
 		.fh = fhandle,
 		.bitmask = server->attr_bitmask,
 	};
+	struct nfs4_statfs_res res = {
+		.fsstat = fsstat,
+	};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_STATFS],
 		.rpc_argp = &args,
-		.rpc_resp = fsstat,
+		.rpc_resp = &res,
 	};
 
 	nfs_fattr_init(fsstat->fattr);
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 1e41420916ad..b7871ad82aac 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4387,7 +4387,8 @@ static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p, struct nfs_pat
 /*
  * STATFS request
  */
-static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p, struct nfs_fsstat *fsstat)
+static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p,
+			       struct nfs4_statfs_res *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
@@ -4398,7 +4399,7 @@ static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p, struct nfs_fssta
 	if (!status)
 		status = decode_putfh(&xdr);
 	if (!status)
-		status = decode_statfs(&xdr, fsstat);
+		status = decode_statfs(&xdr, res->fsstat);
 	return status;
 }
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index f71260aeb803..4dac59ef6f4f 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -795,6 +795,11 @@ struct nfs4_statfs_arg {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs4_statfs_res {
+	struct nfs_fsstat	       *fsstat;
+	struct nfs4_sequence_res	seq_res;
+};
+
 struct nfs4_server_caps_arg {
 	struct nfs_fh		       *fhandle;
 	struct nfs4_sequence_args	seq_args;
-- 
cgit v1.2.3-59-g8ed1b


From 3dda5e434721f942870ee30bc6103761618d410f Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:21:57 -0400
Subject: nfs41: use nfs4_fsinfo_res

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[define nfs4_fsinfo_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 5 ++++-
 fs/nfs/nfs4xdr.c        | 5 +++--
 include/linux/nfs_xdr.h | 5 +++++
 3 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index b715f6057611..b8915ef533da 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2458,10 +2458,13 @@ static int _nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
 		.fh = fhandle,
 		.bitmask = server->attr_bitmask,
 	};
+	struct nfs4_fsinfo_res res = {
+		.fsinfo = fsinfo,
+	};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FSINFO],
 		.rpc_argp = &args,
-		.rpc_resp = fsinfo,
+		.rpc_resp = &res,
 	};
 
 	return rpc_call_sync(server->client, &msg, 0);
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index b7871ad82aac..d9ab8209c286 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4351,7 +4351,8 @@ out:
 /*
  * FSINFO request
  */
-static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs_fsinfo *fsinfo)
+static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p,
+			       struct nfs4_fsinfo_res *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
@@ -4362,7 +4363,7 @@ static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs_fsinf
 	if (!status)
 		status = decode_putfh(&xdr);
 	if (!status)
-		status = decode_fsinfo(&xdr, fsinfo);
+		status = decode_fsinfo(&xdr, res->fsinfo);
 	return status;
 }
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 4dac59ef6f4f..7d64913cbb1b 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -675,6 +675,11 @@ struct nfs4_fsinfo_arg {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs4_fsinfo_res {
+	struct nfs_fsinfo	       *fsinfo;
+	struct nfs4_sequence_res	seq_res;
+};
+
 struct nfs4_getattr_arg {
 	const struct nfs_fh *		fh;
 	const u32 *			bitmask;
-- 
cgit v1.2.3-59-g8ed1b


From d45b2989a7956ae9e71d584ceac942278c0371c7 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:21:58 -0400
Subject: nfs41: use nfs4_pathconf_res

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[define nfs4_pathconf_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 5 ++++-
 fs/nfs/nfs4xdr.c        | 5 +++--
 include/linux/nfs_xdr.h | 5 +++++
 3 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index b8915ef533da..aea2e83d3939 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2496,10 +2496,13 @@ static int _nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle
 		.fh = fhandle,
 		.bitmask = server->attr_bitmask,
 	};
+	struct nfs4_pathconf_res res = {
+		.pathconf = pathconf,
+	};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_PATHCONF],
 		.rpc_argp = &args,
-		.rpc_resp = pathconf,
+		.rpc_resp = &res,
 	};
 
 	/* None of the pathconf attributes are mandatory to implement */
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index d9ab8209c286..a77ee3dd0b3d 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4370,7 +4370,8 @@ static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p,
 /*
  * PATHCONF request
  */
-static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p, struct nfs_pathconf *pathconf)
+static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p,
+				 struct nfs4_pathconf_res *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
@@ -4381,7 +4382,7 @@ static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p, struct nfs_pat
 	if (!status)
 		status = decode_putfh(&xdr);
 	if (!status)
-		status = decode_pathconf(&xdr, pathconf);
+		status = decode_pathconf(&xdr, res->pathconf);
 	return status;
 }
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 7d64913cbb1b..56523319e14c 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -734,6 +734,11 @@ struct nfs4_pathconf_arg {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs4_pathconf_res {
+	struct nfs_pathconf	       *pathconf;
+	struct nfs4_sequence_res	seq_res;
+};
+
 struct nfs4_readdir_arg {
 	const struct nfs_fh *		fh;
 	u64				cookie;
-- 
cgit v1.2.3-59-g8ed1b


From 663c79b3cd8f5fe21fe7d7565fec0072e3234ddc Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:21:59 -0400
Subject: nfs41: use nfs4_getaclres

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: embed resp_len in nfs_getaclres]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 20 +++++++++++---------
 fs/nfs/nfs4xdr.c        |  5 +++--
 include/linux/nfs_xdr.h |  5 +++++
 3 files changed, 19 insertions(+), 11 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index aea2e83d3939..20c9acf689fd 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2755,12 +2755,14 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
 		.acl_pages = pages,
 		.acl_len = buflen,
 	};
-	size_t resp_len = buflen;
+	struct nfs_getaclres res = {
+		.acl_len = buflen,
+	};
 	void *resp_buf;
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GETACL],
 		.rpc_argp = &args,
-		.rpc_resp = &resp_len,
+		.rpc_resp = &res,
 	};
 	struct page *localpage = NULL;
 	int ret;
@@ -2774,7 +2776,7 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
 			return -ENOMEM;
 		args.acl_pages[0] = localpage;
 		args.acl_pgbase = 0;
-		resp_len = args.acl_len = PAGE_SIZE;
+		args.acl_len = PAGE_SIZE;
 	} else {
 		resp_buf = buf;
 		buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase);
@@ -2782,18 +2784,18 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
 	ret = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
 	if (ret)
 		goto out_free;
-	if (resp_len > args.acl_len)
-		nfs4_write_cached_acl(inode, NULL, resp_len);
+	if (res.acl_len > args.acl_len)
+		nfs4_write_cached_acl(inode, NULL, res.acl_len);
 	else
-		nfs4_write_cached_acl(inode, resp_buf, resp_len);
+		nfs4_write_cached_acl(inode, resp_buf, res.acl_len);
 	if (buf) {
 		ret = -ERANGE;
-		if (resp_len > buflen)
+		if (res.acl_len > buflen)
 			goto out_free;
 		if (localpage)
-			memcpy(buf, resp_buf, resp_len);
+			memcpy(buf, resp_buf, res.acl_len);
 	}
-	ret = resp_len;
+	ret = res.acl_len;
 out_free:
 	if (localpage)
 		__free_page(localpage);
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index a77ee3dd0b3d..3e777893e2b0 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4018,7 +4018,8 @@ out:
  * Decode GETACL response
  */
 static int
-nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p, size_t *acl_len)
+nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p,
+		    struct nfs_getaclres *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
@@ -4031,7 +4032,7 @@ nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p, size_t *acl_len)
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
-	status = decode_getacl(&xdr, rqstp, acl_len);
+	status = decode_getacl(&xdr, rqstp, &res->acl_len);
 
 out:
 	return status;
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 56523319e14c..6e9ee2848606 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -444,6 +444,11 @@ struct nfs_getaclargs {
 	struct nfs4_sequence_args 	seq_args;
 };
 
+struct nfs_getaclres {
+	size_t				acl_len;
+	struct nfs4_sequence_res	seq_res;
+};
+
 struct nfs_setattrres {
 	struct nfs_fattr *              fattr;
 	const struct nfs_server *	server;
-- 
cgit v1.2.3-59-g8ed1b


From 9e9ecc03d6b37a5d85e16b357751e3b341de0fef Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:00 -0400
Subject: NFS: get rid of unused xdr decode_setattr(, res) argument

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4xdr.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 3e777893e2b0..27dd25d9ad42 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -3658,7 +3658,7 @@ decode_savefh(struct xdr_stream *xdr)
 	return decode_op_hdr(xdr, OP_SAVEFH);
 }
 
-static int decode_setattr(struct xdr_stream *xdr, struct nfs_setattrres *res)
+static int decode_setattr(struct xdr_stream *xdr)
 {
 	__be32 *p;
 	uint32_t bmlen;
@@ -4009,7 +4009,7 @@ nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p, void *res)
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
-	status = decode_setattr(&xdr, res);
+	status = decode_setattr(&xdr);
 out:
 	return status;
 }
@@ -4162,7 +4162,7 @@ static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs_se
 	status = decode_putfh(&xdr);
 	if (status)
 		goto out;
-	status = decode_setattr(&xdr, res);
+	status = decode_setattr(&xdr);
 	if (status)
 		goto out;
 	decode_getfattr(&xdr, res->fattr, res->server);
-- 
cgit v1.2.3-59-g8ed1b


From 73c403a9a93743b068103c13c05ed136dc687d05 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:01 -0400
Subject: nfs41: use nfs4_setaclres

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[define nfs_setaclres]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 3 ++-
 fs/nfs/nfs4xdr.c        | 3 ++-
 include/linux/nfs_xdr.h | 4 ++++
 3 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 20c9acf689fd..62bbe25d9423 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2842,10 +2842,11 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl
 		.acl_pages	= pages,
 		.acl_len	= buflen,
 	};
+	struct nfs_setaclres res;
 	struct rpc_message msg = {
 		.rpc_proc	= &nfs4_procedures[NFSPROC4_CLNT_SETACL],
 		.rpc_argp	= &arg,
-		.rpc_resp	= NULL,
+		.rpc_resp	= &res,
 	};
 	int ret;
 
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 27dd25d9ad42..aa350d5bf207 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -3996,7 +3996,8 @@ nfs4_xdr_enc_setacl(struct rpc_rqst *req, __be32 *p, struct nfs_setaclargs *args
  * Decode SETACL response
  */
 static int
-nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p, void *res)
+nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p,
+		    struct nfs_setaclres *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 6e9ee2848606..0f2dc8f4cc36 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -436,6 +436,10 @@ struct nfs_setaclargs {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs_setaclres {
+	struct nfs4_sequence_res	seq_res;
+};
+
 struct nfs_getaclargs {
 	struct nfs_fh *			fh;
 	size_t				acl_len;
-- 
cgit v1.2.3-59-g8ed1b


From 22958463d5dca8548e19430779f379e66fd6e4a4 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:02 -0400
Subject: nfs41: use nfs4_fs_locations_res

In preparation for nfs41 sequence processing.

Signed-off-by: Andy Admason <andros@netapp.com>
[find nfs4_fs_locations_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 5 ++++-
 fs/nfs/nfs4xdr.c        | 6 ++++--
 include/linux/nfs_xdr.h | 5 +++++
 3 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 62bbe25d9423..e08edc99faac 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3722,10 +3722,13 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 		.page = page,
 		.bitmask = bitmask,
 	};
+	struct nfs4_fs_locations_res res = {
+		.fs_locations = fs_locations,
+	};
 	struct rpc_message msg = {
 		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_FS_LOCATIONS],
 		.rpc_argp = &args,
-		.rpc_resp = fs_locations,
+		.rpc_resp = &res,
 	};
 	int status;
 
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index aa350d5bf207..e448e33b4d05 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -4504,7 +4504,8 @@ out:
 /*
  * FS_LOCATIONS request
  */
-static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs4_fs_locations *res)
+static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p,
+				     struct nfs4_fs_locations_res *res)
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr;
@@ -4519,7 +4520,8 @@ static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 	if ((status = decode_lookup(&xdr)) != 0)
 		goto out;
 	xdr_enter_page(&xdr, PAGE_SIZE);
-	status = decode_getfattr(&xdr, &res->fattr, res->server);
+	status = decode_getfattr(&xdr, &res->fs_locations->fattr,
+				 res->fs_locations->server);
 out:
 	return status;
 }
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 0f2dc8f4cc36..d837f10c49ef 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -867,6 +867,11 @@ struct nfs4_fs_locations_arg {
 	struct nfs4_sequence_args	seq_args;
 };
 
+struct nfs4_fs_locations_res {
+	struct nfs4_fs_locations       *fs_locations;
+	struct nfs4_sequence_res	seq_res;
+};
+
 #endif /* CONFIG_NFS_V4 */
 
 struct nfs_page;
-- 
cgit v1.2.3-59-g8ed1b


From cccef3b96a4759ae0790452280c00ea505412157 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:03 -0400
Subject: nfs41: introduce nfs4_call_sync

Use nfs4_call_sync rather than rpc_call_sync to provide
for a nfs41 sessions-enabled interface for sessions manipulation.

The nfs41 rpc logic uses the rpc_call_prepare method to
recover and create the session, as well as selecting a free slot id
and the rpc_call_done to free the slot and update slot table
related metadata.

In the coming patches we'll add rpc prepare and done routines
for setting up the sequence op and processing the sequence result.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_call_sync]
As per 11-14-08 review.
Squash into "nfs41: introduce nfs4_call_sync" and "nfs41: nfs4_setup_sequence"
Define two functions one for v4 and one for v41
add a pointer to struct nfs4_client to the correct one.
Signed-off-by: Andy Adamson <andros@netapp.com>
[added BUG() in _nfs4_call_sync_session if !CONFIG_NFS_V4_1]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: check for session not minorversion]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[group minorversion specific stuff together]
Signed-off-by: Alexandros Batsakis <Alexandros.Batsakis@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: fixup nfs4_clear_client_minor_version]
[introduce nfs4_init_client_minor_version() in this patch]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[cleaned-up patch: got rid of nfs_call_sync_t, dprintks, cosmetics, extra server defs]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c           |  5 ++++
 fs/nfs/internal.h         | 12 +++++++++
 fs/nfs/nfs4proc.c         | 67 +++++++++++++++++++++++++++++++++--------------
 include/linux/nfs_fs_sb.h |  8 ++++++
 4 files changed, 73 insertions(+), 19 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index f1506f148521..a9828baaa445 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -194,6 +194,8 @@ static void nfs4_clear_client_minor_version(struct nfs_client *clp)
 		nfs4_destroy_session(clp->cl_session);
 		clp->cl_session = NULL;
 	}
+
+	clp->cl_call_sync = _nfs4_call_sync;
 #endif /* CONFIG_NFS_V4_1 */
 }
 
@@ -1073,6 +1075,8 @@ error:
  */
 static int nfs4_init_client_minor_version(struct nfs_client *clp)
 {
+	clp->cl_call_sync = _nfs4_call_sync;
+
 #if defined(CONFIG_NFS_V4_1)
 	if (clp->cl_minorversion) {
 		struct nfs4_session *session = NULL;
@@ -1086,6 +1090,7 @@ static int nfs4_init_client_minor_version(struct nfs_client *clp)
 			return -ENOMEM;
 
 		clp->cl_session = session;
+		clp->cl_call_sync = _nfs4_call_sync_session;
 	}
 #endif /* CONFIG_NFS_V4_1 */
 
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 7cef45db9257..8d67c2865dc3 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -206,6 +206,18 @@ extern int nfs4_path_walk(struct nfs_server *server,
 			  const char *path);
 #endif
 
+/* nfs4proc.c */
+extern int _nfs4_call_sync(struct nfs_server *server,
+			   struct rpc_message *msg,
+			   struct nfs4_sequence_args *args,
+			   struct nfs4_sequence_res *res,
+			   int cache_reply);
+extern int _nfs4_call_sync_session(struct nfs_server *server,
+				   struct rpc_message *msg,
+				   struct nfs4_sequence_args *args,
+				   struct nfs4_sequence_res *res,
+				   int cache_reply);
+
 /*
  * Determine if sessions are in use.
  */
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index e08edc99faac..4fc5b385f61e 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -271,6 +271,33 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp
 	spin_unlock(&clp->cl_lock);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+
+int _nfs4_call_sync_session(struct nfs_server *server,
+			    struct rpc_message *msg,
+			    struct nfs4_sequence_args *args,
+			    struct nfs4_sequence_res *res,
+			    int cache_reply)
+{
+	/* in preparation for setting up the sequence op */
+	return rpc_call_sync(server->client, msg, 0);
+}
+
+#endif /* CONFIG_NFS_V4_1 */
+
+int _nfs4_call_sync(struct nfs_server *server,
+		    struct rpc_message *msg,
+		    struct nfs4_sequence_args *args,
+		    struct nfs4_sequence_res *res,
+		    int cache_reply)
+{
+	return rpc_call_sync(server->client, msg, 0);
+}
+
+#define nfs4_call_sync(server, msg, args, res, cache_reply) \
+	(server)->nfs_client->cl_call_sync((server), (msg), &(args)->seq_args, \
+			&(res)->seq_res, (cache_reply))
+
 static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo)
 {
 	struct nfs_inode *nfsi = NFS_I(dir);
@@ -1269,7 +1296,7 @@ static int _nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred,
 	} else
 		memcpy(&arg.stateid, &zero_stateid, sizeof(arg.stateid));
 
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &arg, &res, 1);
 	if (status == 0 && state != NULL)
 		renew_lease(server, timestamp);
 	return status;
@@ -1595,7 +1622,7 @@ static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *f
 	};
 	int status;
 
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &args, &res, 0);
 	if (status == 0) {
 		memcpy(server->attr_bitmask, res.attr_bitmask, sizeof(server->attr_bitmask));
 		if (res.attr_bitmask[0] & FATTR4_WORD0_ACL)
@@ -1609,6 +1636,7 @@ static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *f
 		server->cache_consistency_bitmask[1] &= FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY;
 		server->acl_bitmask = res.acl_bitmask;
 	}
+
 	return status;
 }
 
@@ -1641,7 +1669,7 @@ static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
 		.rpc_resp = &res,
 	};
 	nfs_fattr_init(info->fattr);
-	return rpc_call_sync(server->client, &msg, 0);
+	return nfs4_call_sync(server, &msg, &args, &res, 0);
 }
 
 static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
@@ -1731,7 +1759,7 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
 	};
 	
 	nfs_fattr_init(fattr);
-	return rpc_call_sync(server->client, &msg, 0);
+	return nfs4_call_sync(server, &msg, &args, &res, 0);
 }
 
 static int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr)
@@ -1815,7 +1843,7 @@ static int _nfs4_proc_lookupfh(struct nfs_server *server, const struct nfs_fh *d
 	nfs_fattr_init(fattr);
 
 	dprintk("NFS call  lookupfh %s\n", name->name);
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &args, &res, 0);
 	dprintk("NFS reply lookupfh: %d\n", status);
 	return status;
 }
@@ -1901,7 +1929,7 @@ static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry
 			args.access |= NFS4_ACCESS_EXECUTE;
 	}
 	nfs_fattr_init(&fattr);
-	status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	status = nfs4_call_sync(server, &msg, &args, &res, 0);
 	if (!status) {
 		entry->mask = 0;
 		if (res.access & NFS4_ACCESS_READ)
@@ -1967,7 +1995,7 @@ static int _nfs4_proc_readlink(struct inode *inode, struct page *page,
 		.rpc_resp = &res,
 	};
 
-	return rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	return nfs4_call_sync(NFS_SERVER(inode), &msg, &args, &res, 0);
 }
 
 static int nfs4_proc_readlink(struct inode *inode, struct page *page,
@@ -2061,7 +2089,7 @@ static int _nfs4_proc_remove(struct inode *dir, struct qstr *name)
 	int			status;
 
 	nfs_fattr_init(&res.dir_attr);
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &args, &res, 1);
 	if (status == 0) {
 		update_changeattr(dir, &res.cinfo);
 		nfs_post_op_update_inode(dir, &res.dir_attr);
@@ -2129,7 +2157,7 @@ static int _nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name,
 	
 	nfs_fattr_init(res.old_fattr);
 	nfs_fattr_init(res.new_fattr);
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &arg, &res, 1);
 
 	if (!status) {
 		update_changeattr(old_dir, &res.old_cinfo);
@@ -2178,7 +2206,7 @@ static int _nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *
 
 	nfs_fattr_init(res.fattr);
 	nfs_fattr_init(res.dir_attr);
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &arg, &res, 1);
 	if (!status) {
 		update_changeattr(dir, &res.cinfo);
 		nfs_post_op_update_inode(dir, res.dir_attr);
@@ -2239,7 +2267,8 @@ static struct nfs4_createdata *nfs4_alloc_createdata(struct inode *dir,
 
 static int nfs4_do_create(struct inode *dir, struct dentry *dentry, struct nfs4_createdata *data)
 {
-	int status = rpc_call_sync(NFS_CLIENT(dir), &data->msg, 0);
+	int status = nfs4_call_sync(NFS_SERVER(dir), &data->msg,
+				    &data->arg, &data->res, 1);
 	if (status == 0) {
 		update_changeattr(dir, &data->res.dir_cinfo);
 		nfs_post_op_update_inode(dir, data->res.dir_fattr);
@@ -2348,7 +2377,7 @@ static int _nfs4_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
 			(unsigned long long)cookie);
 	nfs4_setup_readdir(cookie, NFS_COOKIEVERF(dir), dentry, &args);
 	res.pgbase = args.pgbase;
-	status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0);
+	status = nfs4_call_sync(NFS_SERVER(dir), &msg, &args, &res, 0);
 	if (status == 0)
 		memcpy(NFS_COOKIEVERF(dir), res.verifier.data, NFS4_VERIFIER_SIZE);
 
@@ -2436,7 +2465,7 @@ static int _nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle,
 	};
 
 	nfs_fattr_init(fsstat->fattr);
-	return rpc_call_sync(server->client, &msg, 0);
+	return  nfs4_call_sync(server, &msg, &args, &res, 0);
 }
 
 static int nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsstat *fsstat)
@@ -2467,7 +2496,7 @@ static int _nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle,
 		.rpc_resp = &res,
 	};
 
-	return rpc_call_sync(server->client, &msg, 0);
+	return nfs4_call_sync(server, &msg, &args, &res, 0);
 }
 
 static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *fsinfo)
@@ -2512,7 +2541,7 @@ static int _nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle
 	}
 
 	nfs_fattr_init(pathconf->fattr);
-	return rpc_call_sync(server->client, &msg, 0);
+	return nfs4_call_sync(server, &msg, &args, &res, 0);
 }
 
 static int nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle,
@@ -2781,7 +2810,7 @@ static ssize_t __nfs4_get_acl_uncached(struct inode *inode, void *buf, size_t bu
 		resp_buf = buf;
 		buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase);
 	}
-	ret = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	ret = nfs4_call_sync(NFS_SERVER(inode), &msg, &args, &res, 0);
 	if (ret)
 		goto out_free;
 	if (res.acl_len > args.acl_len)
@@ -2854,7 +2883,7 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl
 		return -EOPNOTSUPP;
 	nfs_inode_return_delegation(inode);
 	buf_to_pages(buf, buflen, arg.acl_pages, &arg.acl_pgbase);
-	ret = rpc_call_sync(NFS_CLIENT(inode), &msg, 0);
+	ret = nfs4_call_sync(server, &msg, &arg, &res, 1);
 	nfs_access_zap_cache(inode);
 	nfs_zap_acl_cache(inode);
 	return ret;
@@ -3143,7 +3172,7 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock
 		goto out;
 	lsp = request->fl_u.nfs4_fl.owner;
 	arg.lock_owner.id = lsp->ls_id.id;
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &arg, &res, 1);
 	switch (status) {
 		case 0:
 			request->fl_type = F_UNLCK;
@@ -3736,7 +3765,7 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 	nfs_fattr_init(&fs_locations->fattr);
 	fs_locations->server = server;
 	fs_locations->nlocations = 0;
-	status = rpc_call_sync(server->client, &msg, 0);
+	status = nfs4_call_sync(server, &msg, &args, &res, 0);
 	nfs_fixup_referral_attributes(&fs_locations->fattr);
 	dprintk("%s: returned status = %d\n", __func__, status);
 	return status;
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index b47c0fc55d42..206485e5082f 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -12,6 +12,9 @@
 struct nfs4_session;
 struct nfs_iostats;
 struct nlm_host;
+struct nfs4_sequence_args;
+struct nfs4_sequence_res;
+struct nfs_server;
 
 /*
  * The nfs_client identifies our client state to the server.
@@ -67,6 +70,11 @@ struct nfs_client {
 	 */
 	char			cl_ipaddr[48];
 	unsigned char		cl_id_uniquifier;
+	int		     (* cl_call_sync)(struct nfs_server *server,
+					      struct rpc_message *msg,
+					      struct nfs4_sequence_args *args,
+					      struct nfs4_sequence_res *res,
+					      int cache_reply);
 #endif /* CONFIG_NFS_V4 */
 
 #ifdef CONFIG_NFS_V4_1
-- 
cgit v1.2.3-59-g8ed1b


From f3752975caa716709c5ea0b0820b86111d921df4 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:04 -0400
Subject: nfs41: nfs41: pass *session in seq_args and seq_res

To be used for getting the rpc's minorversion and for nfs41 xdr
{en,de}coding of the sequence operation.
Reset the seq session ptrs for minorversion=0 rpc calls.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 1 +
 include/linux/nfs_xdr.h | 4 ++--
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 4fc5b385f61e..49cf969417c0 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -291,6 +291,7 @@ int _nfs4_call_sync(struct nfs_server *server,
 		    struct nfs4_sequence_res *res,
 		    int cache_reply)
 {
+	args->sa_session = res->sr_session = NULL;
 	return rpc_call_sync(server->client, msg, 0);
 }
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index d837f10c49ef..f5675063f951 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -162,11 +162,11 @@ struct nfs4_slot {
 };
 
 struct nfs4_sequence_args {
-	/* stub */
+	struct nfs4_session	*sa_session;
 };
 
 struct nfs4_sequence_res {
-	/* stub */
+	struct nfs4_session	*sr_session;
 };
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 5f7dbd5c752d88310d8fe1feedefd5c6496eff48 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:05 -0400
Subject: nfs41: set up seq_res.sr_slotid

Initialize nfs4_sequence_res sr_slotid to NFS4_MAX_SLOT_TABLE.

[was nfs41: sequence res use slotid]
Signed-off-by: Andy Adamson <andros@netapp.com>
[pulled definition of struct nfs4_sequence_res.sr_slotid to here]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 5 +++++
 fs/nfs/read.c           | 1 +
 fs/nfs/unlink.c         | 1 +
 fs/nfs/write.c          | 2 ++
 include/linux/nfs_xdr.h | 1 +
 5 files changed, 10 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 49cf969417c0..5878930b4c3b 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -371,6 +371,7 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct path *path,
 	p->o_arg.server = server;
 	p->o_arg.bitmask = server->attr_bitmask;
 	p->o_arg.claim = NFS4_OPEN_CLAIM_NULL;
+	p->o_res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	if (flags & O_EXCL) {
 		u32 *s = (u32 *) p->o_arg.u.verifier.data;
 		s[0] = jiffies;
@@ -1463,6 +1464,7 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 	calldata->res.fattr = &calldata->fattr;
 	calldata->res.seqid = calldata->arg.seqid;
 	calldata->res.server = server;
+	calldata->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	calldata->path.mnt = mntget(path->mnt);
 	calldata->path.dentry = dget(path->dentry);
 
@@ -3088,6 +3090,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co
 	memcpy(&data->stateid, stateid, sizeof(data->stateid));
 	data->res.fattr = &data->fattr;
 	data->res.server = server;
+	data->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	nfs_fattr_init(data->res.fattr);
 	data->timestamp = jiffies;
 	data->rpc_status = 0;
@@ -3240,6 +3243,7 @@ static struct nfs4_unlockdata *nfs4_alloc_unlockdata(struct file_lock *fl,
 	p->arg.fl = &p->fl;
 	p->arg.seqid = seqid;
 	p->res.seqid = seqid;
+	p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	p->arg.stateid = &lsp->ls_stateid;
 	p->lsp = lsp;
 	atomic_inc(&lsp->ls_count);
@@ -3412,6 +3416,7 @@ static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl,
 	p->arg.lock_owner.clientid = server->nfs_client->cl_clientid;
 	p->arg.lock_owner.id = lsp->ls_id.id;
 	p->res.lock_seqid = p->arg.lock_seqid;
+	p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	p->lsp = lsp;
 	atomic_inc(&lsp->ls_count);
 	p->ctx = get_nfs_open_context(ctx);
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 4ace3c50a8eb..70ba2b4cb9a4 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -46,6 +46,7 @@ struct nfs_read_data *nfs_readdata_alloc(unsigned int pagecount)
 		memset(p, 0, sizeof(*p));
 		INIT_LIST_HEAD(&p->pages);
 		p->npages = pagecount;
+		p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 		if (pagecount <= ARRAY_SIZE(p->page_array))
 			p->pagevec = p->page_array;
 		else {
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index ecc295347775..83d0ce2600ab 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -241,6 +241,7 @@ nfs_async_unlink(struct inode *dir, struct dentry *dentry)
 		status = PTR_ERR(data->cred);
 		goto out_free;
 	}
+	data->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 
 	status = -EBUSY;
 	spin_lock(&dentry->d_lock);
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index e560a78995a3..035e6fb9f57e 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -52,6 +52,7 @@ struct nfs_write_data *nfs_commitdata_alloc(void)
 	if (p) {
 		memset(p, 0, sizeof(*p));
 		INIT_LIST_HEAD(&p->pages);
+		p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	}
 	return p;
 }
@@ -71,6 +72,7 @@ struct nfs_write_data *nfs_writedata_alloc(unsigned int pagecount)
 		memset(p, 0, sizeof(*p));
 		INIT_LIST_HEAD(&p->pages);
 		p->npages = pagecount;
+		p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 		if (pagecount <= ARRAY_SIZE(p->page_array))
 			p->pagevec = p->page_array;
 		else {
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index f5675063f951..db0d1236aae7 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -167,6 +167,7 @@ struct nfs4_sequence_args {
 
 struct nfs4_sequence_res {
 	struct nfs4_session	*sr_session;
+	u8			sr_slotid;	/* slot used to send request */
 };
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 6ce183919b4a09289cb0fe4fce960a9faa1e7c6b Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:06 -0400
Subject: NFS: use decode_change_info_maxsz for xdr maxsz calculations

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4xdr.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index e448e33b4d05..487a197231fd 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -192,12 +192,16 @@ static int nfs4_stat_to_errno(int);
 				 decode_verifier_maxsz)
 #define encode_remove_maxsz	(op_encode_hdr_maxsz + \
 				nfs4_name_maxsz)
+#define decode_remove_maxsz	(op_decode_hdr_maxsz + \
+				 decode_change_info_maxsz)
 #define encode_rename_maxsz	(op_encode_hdr_maxsz + \
 				2 * nfs4_name_maxsz)
-#define decode_rename_maxsz	(op_decode_hdr_maxsz + 5 + 5)
+#define decode_rename_maxsz	(op_decode_hdr_maxsz + \
+				 decode_change_info_maxsz + \
+				 decode_change_info_maxsz)
 #define encode_link_maxsz	(op_encode_hdr_maxsz + \
 				nfs4_name_maxsz)
-#define decode_link_maxsz	(op_decode_hdr_maxsz + 5)
+#define decode_link_maxsz	(op_decode_hdr_maxsz + decode_change_info_maxsz)
 #define encode_lock_maxsz	(op_encode_hdr_maxsz + \
 				 7 + \
 				 1 + encode_stateid_maxsz + 8)
@@ -414,7 +418,7 @@ static int nfs4_stat_to_errno(int);
 				encode_getattr_maxsz)
 #define NFS4_dec_remove_sz	(compound_decode_hdr_maxsz + \
 				decode_putfh_maxsz + \
-				op_decode_hdr_maxsz + 5 + \
+				decode_remove_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_rename_sz	(compound_encode_hdr_maxsz + \
 				encode_putfh_maxsz + \
-- 
cgit v1.2.3-59-g8ed1b


From 0c4e8c187758258ec58842384fe6a99cf1ce16c7 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:07 -0400
Subject: NFS: define and initialize compound_hdr.replen

replen holds the running count of expected reply bytes.
repl will then be used by encoding routines for xdr_inline_pages offset
after which data bytes are to be received directly into the xdr
buffer pages.

NOTE: According to the nfsv4 and v4.1 RFCs, the replied tag SHOULD be the same
is the one sent, but this is not required as a MUST for the server to do so.
The server may screw us if it replies a tag of a different length in the
compound result.

[NFS: cb_compoundhdr.replen is in words not bytes]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4xdr.c | 77 +++++++++++++++++++++++++++++++-------------------------
 1 file changed, 43 insertions(+), 34 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 487a197231fd..c85dbee34b8a 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -545,6 +545,7 @@ struct compound_hdr {
 	__be32 *	nops_p;
 	uint32_t	taglen;
 	char *		tag;
+	uint32_t	replen;		/* expected reply words */
 };
 
 /*
@@ -580,9 +581,17 @@ static void encode_string(struct xdr_stream *xdr, unsigned int len, const char *
 	xdr_encode_opaque(p, str, len);
 }
 
-static void encode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr)
+static void encode_compound_hdr(struct xdr_stream *xdr,
+				struct rpc_rqst *req,
+				struct compound_hdr *hdr)
 {
 	__be32 *p;
+	struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth;
+
+	/* initialize running count of expected bytes in reply.
+	 * NOTE: the replied tag SHOULD be the same is the one sent,
+	 * but this is not required as a MUST for the server to do so. */
+	hdr->replen = RPC_REPHDRSIZE + auth->au_rslack + 3 + hdr->taglen;
 
 	dprintk("encode_compound: tag=%.*s\n", (int)hdr->taglen, hdr->tag);
 	BUG_ON(hdr->taglen > NFS4_MAXTAGLEN);
@@ -1359,7 +1368,7 @@ static int nfs4_xdr_enc_access(struct rpc_rqst *req, __be32 *p, const struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_access(&xdr, args->access, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1378,7 +1387,7 @@ static int nfs4_xdr_enc_lookup(struct rpc_rqst *req, __be32 *p, const struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_lookup(&xdr, args->name, &hdr);
 	encode_getfh(&xdr, &hdr);
@@ -1398,7 +1407,7 @@ static int nfs4_xdr_enc_lookup_root(struct rpc_rqst *req, __be32 *p, const struc
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putrootfh(&xdr, &hdr);
 	encode_getfh(&xdr, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1417,7 +1426,7 @@ static int nfs4_xdr_enc_remove(struct rpc_rqst *req, __be32 *p, const struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_remove(&xdr, &args->name, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1436,7 +1445,7 @@ static int nfs4_xdr_enc_rename(struct rpc_rqst *req, __be32 *p, const struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->old_dir, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_putfh(&xdr, args->new_dir, &hdr);
@@ -1459,7 +1468,7 @@ static int nfs4_xdr_enc_link(struct rpc_rqst *req, __be32 *p, const struct nfs4_
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
@@ -1482,7 +1491,7 @@ static int nfs4_xdr_enc_create(struct rpc_rqst *req, __be32 *p, const struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_create(&xdr, args, &hdr);
@@ -1513,7 +1522,7 @@ static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, __be32 *p, const struct nf
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
 	encode_nops(&hdr);
@@ -1531,7 +1540,7 @@ static int nfs4_xdr_enc_close(struct rpc_rqst *req, __be32 *p, struct nfs_closea
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_close(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1550,7 +1559,7 @@ static int nfs4_xdr_enc_open(struct rpc_rqst *req, __be32 *p, struct nfs_openarg
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_open(&xdr, args, &hdr);
@@ -1573,7 +1582,7 @@ static int nfs4_xdr_enc_open_confirm(struct rpc_rqst *req, __be32 *p, struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_open_confirm(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1591,7 +1600,7 @@ static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, __be32 *p, struct nfs_
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_open(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1610,7 +1619,7 @@ static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, __be32 *p, struct n
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_open_downgrade(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1629,7 +1638,7 @@ static int nfs4_xdr_enc_lock(struct rpc_rqst *req, __be32 *p, struct nfs_lock_ar
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_lock(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1647,7 +1656,7 @@ static int nfs4_xdr_enc_lockt(struct rpc_rqst *req, __be32 *p, struct nfs_lockt_
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_lockt(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1665,7 +1674,7 @@ static int nfs4_xdr_enc_locku(struct rpc_rqst *req, __be32 *p, struct nfs_locku_
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_locku(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1685,7 +1694,7 @@ static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct n
 	unsigned int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_readlink(&xdr, args, req, &hdr);
 
@@ -1713,7 +1722,7 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf
 	int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_readdir(&xdr, args, req, &hdr);
 
@@ -1744,7 +1753,7 @@ static int nfs4_xdr_enc_read(struct rpc_rqst *req, __be32 *p, struct nfs_readarg
 	int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_read(&xdr, args, &hdr);
 
@@ -1771,7 +1780,7 @@ static int nfs4_xdr_enc_setattr(struct rpc_rqst *req, __be32 *p, struct nfs_seta
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_setattr(&xdr, args, args->server, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1794,7 +1803,7 @@ nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p,
 	int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getattr_two(&xdr, FATTR4_WORD0_ACL, 0, &hdr);
 
@@ -1817,7 +1826,7 @@ static int nfs4_xdr_enc_write(struct rpc_rqst *req, __be32 *p, struct nfs_writea
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_write(&xdr, args, &hdr);
 	req->rq_snd_buf.flags |= XDRBUF_WRITE;
@@ -1837,7 +1846,7 @@ static int nfs4_xdr_enc_commit(struct rpc_rqst *req, __be32 *p, struct nfs_write
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_commit(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1856,7 +1865,7 @@ static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs4_fsin
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_fsinfo(&xdr, args->bitmask, &hdr);
 	encode_nops(&hdr);
@@ -1874,7 +1883,7 @@ static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, __be32 *p, const struct n
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getattr_one(&xdr, args->bitmask[0] & nfs4_pathconf_bitmap[0],
 			   &hdr);
@@ -1893,7 +1902,7 @@ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, __be32 *p, const struct nfs
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getattr_two(&xdr, args->bitmask[0] & nfs4_statfs_bitmap[0],
 			   args->bitmask[1] & nfs4_statfs_bitmap[1], &hdr);
@@ -1913,7 +1922,7 @@ static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p,
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fhandle, &hdr);
 	encode_getattr_one(&xdr, FATTR4_WORD0_SUPPORTED_ATTRS|
 			   FATTR4_WORD0_LINK_SUPPORT|
@@ -1934,7 +1943,7 @@ static int nfs4_xdr_enc_renew(struct rpc_rqst *req, __be32 *p, struct nfs_client
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_renew(&xdr, clp, &hdr);
 	encode_nops(&hdr);
 	return 0;
@@ -1951,7 +1960,7 @@ static int nfs4_xdr_enc_setclientid(struct rpc_rqst *req, __be32 *p, struct nfs4
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_setclientid(&xdr, sc, &hdr);
 	encode_nops(&hdr);
 	return 0;
@@ -1969,7 +1978,7 @@ static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, __be32 *p, str
 	const u32 lease_bitmap[2] = { FATTR4_WORD0_LEASE_TIME, 0 };
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_setclientid_confirm(&xdr, clp, &hdr);
 	encode_putrootfh(&xdr, &hdr);
 	encode_fsinfo(&xdr, lease_bitmap, &hdr);
@@ -1988,7 +1997,7 @@ static int nfs4_xdr_enc_delegreturn(struct rpc_rqst *req, __be32 *p, const struc
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fhandle, &hdr);
 	encode_delegreturn(&xdr, args->stateid, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -2009,7 +2018,7 @@ static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 	int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_lookup(&xdr, args->name, &hdr);
 	encode_fs_locations(&xdr, args->bitmask, &hdr);
@@ -3989,7 +3998,7 @@ nfs4_xdr_enc_setacl(struct rpc_rqst *req, __be32 *p, struct nfs_setaclargs *args
 	int status;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
-	encode_compound_hdr(&xdr, &hdr);
+	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	status = encode_setacl(&xdr, args, &hdr);
 	encode_nops(&hdr);
-- 
cgit v1.2.3-59-g8ed1b


From dadf0c2767ce7772fc4ff82044f3ba5823e5b79f Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:08 -0400
Subject: NFS: update hdr->replen for every encode op

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4xdr.c | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index c85dbee34b8a..746c6a52794c 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -749,6 +749,7 @@ static void encode_access(struct xdr_stream *xdr, u32 access, struct compound_hd
 	WRITE32(OP_ACCESS);
 	WRITE32(access);
 	hdr->nops++;
+	hdr->replen += decode_access_maxsz;
 }
 
 static void encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg, struct compound_hdr *hdr)
@@ -760,6 +761,7 @@ static void encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg
 	WRITE32(arg->seqid->sequence->counter);
 	WRITEMEM(arg->stateid->data, NFS4_STATEID_SIZE);
 	hdr->nops++;
+	hdr->replen += decode_close_maxsz;
 }
 
 static void encode_commit(struct xdr_stream *xdr, const struct nfs_writeargs *args, struct compound_hdr *hdr)
@@ -771,6 +773,7 @@ static void encode_commit(struct xdr_stream *xdr, const struct nfs_writeargs *ar
 	WRITE64(args->offset);
 	WRITE32(args->count);
 	hdr->nops++;
+	hdr->replen += decode_commit_maxsz;
 }
 
 static void encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg *create, struct compound_hdr *hdr)
@@ -802,6 +805,7 @@ static void encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg *
 	WRITE32(create->name->len);
 	WRITEMEM(create->name->name, create->name->len);
 	hdr->nops++;
+	hdr->replen += decode_create_maxsz;
 
 	encode_attrs(xdr, create->attrs, create->server);
 }
@@ -815,6 +819,7 @@ static void encode_getattr_one(struct xdr_stream *xdr, uint32_t bitmap, struct c
 	WRITE32(1);
 	WRITE32(bitmap);
 	hdr->nops++;
+	hdr->replen += decode_getattr_maxsz;
 }
 
 static void encode_getattr_two(struct xdr_stream *xdr, uint32_t bm0, uint32_t bm1, struct compound_hdr *hdr)
@@ -827,6 +832,7 @@ static void encode_getattr_two(struct xdr_stream *xdr, uint32_t bm0, uint32_t bm
 	WRITE32(bm0);
 	WRITE32(bm1);
 	hdr->nops++;
+	hdr->replen += decode_getattr_maxsz;
 }
 
 static void encode_getfattr(struct xdr_stream *xdr, const u32* bitmask, struct compound_hdr *hdr)
@@ -854,6 +860,7 @@ static void encode_getfh(struct xdr_stream *xdr, struct compound_hdr *hdr)
 	RESERVE_SPACE(4);
 	WRITE32(OP_GETFH);
 	hdr->nops++;
+	hdr->replen += decode_getfh_maxsz;
 }
 
 static void encode_link(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr)
@@ -865,6 +872,7 @@ static void encode_link(struct xdr_stream *xdr, const struct qstr *name, struct
 	WRITE32(name->len);
 	WRITEMEM(name->name, name->len);
 	hdr->nops++;
+	hdr->replen += decode_link_maxsz;
 }
 
 static inline int nfs4_lock_type(struct file_lock *fl, int block)
@@ -912,6 +920,7 @@ static void encode_lock(struct xdr_stream *xdr, const struct nfs_lock_args *args
 		WRITE32(args->lock_seqid->sequence->counter);
 	}
 	hdr->nops++;
+	hdr->replen += decode_lock_maxsz;
 }
 
 static void encode_lockt(struct xdr_stream *xdr, const struct nfs_lockt_args *args, struct compound_hdr *hdr)
@@ -928,6 +937,7 @@ static void encode_lockt(struct xdr_stream *xdr, const struct nfs_lockt_args *ar
 	WRITEMEM("lock id:", 8);
 	WRITE64(args->lock_owner.id);
 	hdr->nops++;
+	hdr->replen += decode_lockt_maxsz;
 }
 
 static void encode_locku(struct xdr_stream *xdr, const struct nfs_locku_args *args, struct compound_hdr *hdr)
@@ -942,6 +952,7 @@ static void encode_locku(struct xdr_stream *xdr, const struct nfs_locku_args *ar
 	WRITE64(args->fl->fl_start);
 	WRITE64(nfs4_lock_length(args->fl));
 	hdr->nops++;
+	hdr->replen += decode_locku_maxsz;
 }
 
 static void encode_lookup(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr)
@@ -954,6 +965,7 @@ static void encode_lookup(struct xdr_stream *xdr, const struct qstr *name, struc
 	WRITE32(len);
 	WRITEMEM(name->name, len);
 	hdr->nops++;
+	hdr->replen += decode_lookup_maxsz;
 }
 
 static void encode_share_access(struct xdr_stream *xdr, fmode_t fmode)
@@ -1093,6 +1105,7 @@ static void encode_open(struct xdr_stream *xdr, const struct nfs_openargs *arg,
 		BUG();
 	}
 	hdr->nops++;
+	hdr->replen += decode_open_maxsz;
 }
 
 static void encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_confirmargs *arg, struct compound_hdr *hdr)
@@ -1104,6 +1117,7 @@ static void encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_co
 	WRITEMEM(arg->stateid->data, NFS4_STATEID_SIZE);
 	WRITE32(arg->seqid->sequence->counter);
 	hdr->nops++;
+	hdr->replen += decode_open_confirm_maxsz;
 }
 
 static void encode_open_downgrade(struct xdr_stream *xdr, const struct nfs_closeargs *arg, struct compound_hdr *hdr)
@@ -1116,6 +1130,7 @@ static void encode_open_downgrade(struct xdr_stream *xdr, const struct nfs_close
 	WRITE32(arg->seqid->sequence->counter);
 	encode_share_access(xdr, arg->fmode);
 	hdr->nops++;
+	hdr->replen += decode_open_downgrade_maxsz;
 }
 
 static void
@@ -1129,6 +1144,7 @@ encode_putfh(struct xdr_stream *xdr, const struct nfs_fh *fh, struct compound_hd
 	WRITE32(len);
 	WRITEMEM(fh->data, len);
 	hdr->nops++;
+	hdr->replen += decode_putfh_maxsz;
 }
 
 static void encode_putrootfh(struct xdr_stream *xdr, struct compound_hdr *hdr)
@@ -1138,6 +1154,7 @@ static void encode_putrootfh(struct xdr_stream *xdr, struct compound_hdr *hdr)
 	RESERVE_SPACE(4);
 	WRITE32(OP_PUTROOTFH);
 	hdr->nops++;
+	hdr->replen += decode_putrootfh_maxsz;
 }
 
 static void encode_stateid(struct xdr_stream *xdr, const struct nfs_open_context *ctx)
@@ -1166,6 +1183,7 @@ static void encode_read(struct xdr_stream *xdr, const struct nfs_readargs *args,
 	WRITE64(args->offset);
 	WRITE32(args->count);
 	hdr->nops++;
+	hdr->replen += decode_read_maxsz;
 }
 
 static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg *readdir, struct rpc_rqst *req, struct compound_hdr *hdr)
@@ -1191,6 +1209,7 @@ static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg
 	WRITE32(attrs[0] & readdir->bitmask[0]);
 	WRITE32(attrs[1] & readdir->bitmask[1]);
 	hdr->nops++;
+	hdr->replen += decode_readdir_maxsz;
 	dprintk("%s: cookie = %Lu, verifier = %08x:%08x, bitmap = %08x:%08x\n",
 			__func__,
 			(unsigned long long)readdir->cookie,
@@ -1207,6 +1226,7 @@ static void encode_readlink(struct xdr_stream *xdr, const struct nfs4_readlink *
 	RESERVE_SPACE(4);
 	WRITE32(OP_READLINK);
 	hdr->nops++;
+	hdr->replen += decode_readlink_maxsz;
 }
 
 static void encode_remove(struct xdr_stream *xdr, const struct qstr *name, struct compound_hdr *hdr)
@@ -1218,6 +1238,7 @@ static void encode_remove(struct xdr_stream *xdr, const struct qstr *name, struc
 	WRITE32(name->len);
 	WRITEMEM(name->name, name->len);
 	hdr->nops++;
+	hdr->replen += decode_remove_maxsz;
 }
 
 static void encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, const struct qstr *newname, struct compound_hdr *hdr)
@@ -1233,6 +1254,7 @@ static void encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, co
 	WRITE32(newname->len);
 	WRITEMEM(newname->name, newname->len);
 	hdr->nops++;
+	hdr->replen += decode_rename_maxsz;
 }
 
 static void encode_renew(struct xdr_stream *xdr, const struct nfs_client *client_stateid, struct compound_hdr *hdr)
@@ -1243,6 +1265,7 @@ static void encode_renew(struct xdr_stream *xdr, const struct nfs_client *client
 	WRITE32(OP_RENEW);
 	WRITE64(client_stateid->cl_clientid);
 	hdr->nops++;
+	hdr->replen += decode_renew_maxsz;
 }
 
 static void
@@ -1253,6 +1276,7 @@ encode_restorefh(struct xdr_stream *xdr, struct compound_hdr *hdr)
 	RESERVE_SPACE(4);
 	WRITE32(OP_RESTOREFH);
 	hdr->nops++;
+	hdr->replen += decode_restorefh_maxsz;
 }
 
 static int
@@ -1272,6 +1296,7 @@ encode_setacl(struct xdr_stream *xdr, struct nfs_setaclargs *arg, struct compoun
 	WRITE32(arg->acl_len);
 	xdr_write_pages(xdr, arg->acl_pages, arg->acl_pgbase, arg->acl_len);
 	hdr->nops++;
+	hdr->replen += decode_setacl_maxsz;
 	return 0;
 }
 
@@ -1283,6 +1308,7 @@ encode_savefh(struct xdr_stream *xdr, struct compound_hdr *hdr)
 	RESERVE_SPACE(4);
 	WRITE32(OP_SAVEFH);
 	hdr->nops++;
+	hdr->replen += decode_savefh_maxsz;
 }
 
 static void encode_setattr(struct xdr_stream *xdr, const struct nfs_setattrargs *arg, const struct nfs_server *server, struct compound_hdr *hdr)
@@ -1293,6 +1319,7 @@ static void encode_setattr(struct xdr_stream *xdr, const struct nfs_setattrargs
 	WRITE32(OP_SETATTR);
 	WRITEMEM(arg->stateid.data, NFS4_STATEID_SIZE);
 	hdr->nops++;
+	hdr->replen += decode_setattr_maxsz;
 	encode_attrs(xdr, arg->iap, server);
 }
 
@@ -1312,6 +1339,7 @@ static void encode_setclientid(struct xdr_stream *xdr, const struct nfs4_setclie
 	RESERVE_SPACE(4);
 	WRITE32(setclientid->sc_cb_ident);
 	hdr->nops++;
+	hdr->replen += decode_setclientid_maxsz;
 }
 
 static void encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_client *client_state, struct compound_hdr *hdr)
@@ -1323,6 +1351,7 @@ static void encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_
 	WRITE64(client_state->cl_clientid);
 	WRITEMEM(client_state->cl_confirm.data, NFS4_VERIFIER_SIZE);
 	hdr->nops++;
+	hdr->replen += decode_setclientid_confirm_maxsz;
 }
 
 static void encode_write(struct xdr_stream *xdr, const struct nfs_writeargs *args, struct compound_hdr *hdr)
@@ -1341,6 +1370,7 @@ static void encode_write(struct xdr_stream *xdr, const struct nfs_writeargs *arg
 
 	xdr_write_pages(xdr, args->pages, args->pgbase, args->count);
 	hdr->nops++;
+	hdr->replen += decode_write_maxsz;
 }
 
 static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *stateid, struct compound_hdr *hdr)
@@ -1352,6 +1382,7 @@ static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *state
 	WRITE32(OP_DELEGRETURN);
 	WRITEMEM(stateid->data, NFS4_STATEID_SIZE);
 	hdr->nops++;
+	hdr->replen += decode_delegreturn_maxsz;
 }
 /*
  * END OF "GENERIC" ENCODE ROUTINES.
-- 
cgit v1.2.3-59-g8ed1b


From 28f566942c6b1d929f5e240e69e7081b77b238d3 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:09 -0400
Subject: NFS: use dynamically computed compound_hdr.replen for
 xdr_inline_pages offset

As Trond suggested, rather than passing a constant to xdr_inline_pages,
keep a running count of the expected reply bytes.  In preparation for
nfs41, where additional op sequence are expteced when talking to nfs41
servers.

[NFS: cb_compoundhdr.replen is in words not bytes]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: get fs_locations replen before encoding the GETATTR]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: get getacl replen before encoding the GETATTR]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4xdr.c | 48 ++++++++++--------------------------------------
 1 file changed, 10 insertions(+), 38 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 746c6a52794c..68889469d114 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -1721,20 +1721,13 @@ static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct n
 	struct compound_hdr hdr = {
 		.nops = 0,
 	};
-	struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth;
-	unsigned int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_readlink(&xdr, args, req, &hdr);
 
-	/* set up reply kvec
-	 *    toplevel_status + taglen + rescount + OP_PUTFH + status
-	 *      + OP_READLINK + status + string length = 8
-	 */
-	replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_readlink_sz) << 2;
-	xdr_inline_pages(&req->rq_rcv_buf, replen, args->pages,
+	xdr_inline_pages(&req->rq_rcv_buf, hdr.replen << 2, args->pages,
 			args->pgbase, args->pglen);
 	encode_nops(&hdr);
 	return 0;
@@ -1749,23 +1742,16 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf
 	struct compound_hdr hdr = {
 		.nops = 0,
 	};
-	struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth;
-	int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_readdir(&xdr, args, req, &hdr);
 
-	/* set up reply kvec
-	 *    toplevel_status + taglen + rescount + OP_PUTFH + status
-	 *      + OP_READDIR + status + verifer(2)  = 9
-	 */
-	replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_readdir_sz) << 2;
-	xdr_inline_pages(&req->rq_rcv_buf, replen, args->pages,
+	xdr_inline_pages(&req->rq_rcv_buf, hdr.replen << 2, args->pages,
 			 args->pgbase, args->count);
 	dprintk("%s: inlined page args = (%u, %p, %u, %u)\n",
-			__func__, replen, args->pages,
+			__func__, hdr.replen << 2, args->pages,
 			args->pgbase, args->count);
 	encode_nops(&hdr);
 	return 0;
@@ -1776,24 +1762,17 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf
  */
 static int nfs4_xdr_enc_read(struct rpc_rqst *req, __be32 *p, struct nfs_readargs *args)
 {
-	struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth;
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
 		.nops = 0,
 	};
-	int replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_read(&xdr, args, &hdr);
 
-	/* set up reply kvec
-	 *    toplevel status + taglen=0 + rescount + OP_PUTFH + status
-	 *       + OP_READ + status + eof + datalen = 9
-	 */
-	replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_read_sz) << 2;
-	xdr_inline_pages(&req->rq_rcv_buf, replen,
+	xdr_inline_pages(&req->rq_rcv_buf, hdr.replen << 2,
 			 args->pages, args->pgbase, args->count);
 	req->rq_rcv_buf.flags |= XDRBUF_READ;
 	encode_nops(&hdr);
@@ -1827,20 +1806,18 @@ nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p,
 		struct nfs_getaclargs *args)
 {
 	struct xdr_stream xdr;
-	struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth;
 	struct compound_hdr hdr = {
 		.nops   = 0,
 	};
-	int replen;
+	uint32_t replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
+	replen = hdr.replen + nfs4_fattr_bitmap_maxsz + 1;
 	encode_getattr_two(&xdr, FATTR4_WORD0_ACL, 0, &hdr);
 
-	/* set up reply buffer: */
-	replen = (RPC_REPHDRSIZE + auth->au_rslack + NFS4_dec_getacl_sz) << 2;
-	xdr_inline_pages(&req->rq_rcv_buf, replen,
+	xdr_inline_pages(&req->rq_rcv_buf, replen << 2,
 		args->acl_pages, args->acl_pgbase, args->acl_len);
 	encode_nops(&hdr);
 	return 0;
@@ -2045,21 +2022,16 @@ static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 	struct compound_hdr hdr = {
 		.nops = 0,
 	};
-	struct rpc_auth *auth = req->rq_task->tk_msg.rpc_cred->cr_auth;
-	int replen;
+	uint32_t replen;
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_lookup(&xdr, args->name, &hdr);
+	replen = hdr.replen;	/* get the attribute into args->page */
 	encode_fs_locations(&xdr, args->bitmask, &hdr);
 
-	/* set up reply
-	 *   toplevel_status + OP_PUTFH + status
-	 *   + OP_LOOKUP + status + OP_GETATTR + status = 7
-	 */
-	replen = (RPC_REPHDRSIZE + auth->au_rslack + 7) << 2;
-	xdr_inline_pages(&req->rq_rcv_buf, replen, &args->page,
+	xdr_inline_pages(&req->rq_rcv_buf, replen << 2, &args->page,
 			0, PAGE_SIZE);
 	encode_nops(&hdr);
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 66cc042970f7077c66be65a780eb3a60a9bcbf0b Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:10 -0400
Subject: nfs41: encode minorversion in compound header

Signed-off-by: Andy Adamdon <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4xdr.c | 70 ++++++++++++++++++++++++++++++++------------------------
 1 file changed, 40 insertions(+), 30 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 68889469d114..19ca0b519a1d 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -546,6 +546,7 @@ struct compound_hdr {
 	uint32_t	taglen;
 	char *		tag;
 	uint32_t	replen;		/* expected reply words */
+	u32		minorversion;
 };
 
 /*
@@ -598,7 +599,7 @@ static void encode_compound_hdr(struct xdr_stream *xdr,
 	RESERVE_SPACE(12+(XDR_QUADLEN(hdr->taglen)<<2));
 	WRITE32(hdr->taglen);
 	WRITEMEM(hdr->tag, hdr->taglen);
-	WRITE32(NFS4_MINOR_VERSION);
+	WRITE32(hdr->minorversion);
 	hdr->nops_p = p;
 	WRITE32(hdr->nops);
 }
@@ -1388,6 +1389,15 @@ static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *state
  * END OF "GENERIC" ENCODE ROUTINES.
  */
 
+static u32 nfs4_xdr_minorversion(const struct nfs4_sequence_args *args)
+{
+#if defined(CONFIG_NFS_V4_1)
+	if (args->sa_session)
+		return args->sa_session->clp->cl_minorversion;
+#endif /* CONFIG_NFS_V4_1 */
+	return 0;
+}
+
 /*
  * Encode an ACCESS request
  */
@@ -1395,7 +1405,7 @@ static int nfs4_xdr_enc_access(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1414,7 +1424,7 @@ static int nfs4_xdr_enc_lookup(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1434,7 +1444,7 @@ static int nfs4_xdr_enc_lookup_root(struct rpc_rqst *req, __be32 *p, const struc
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1453,7 +1463,7 @@ static int nfs4_xdr_enc_remove(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1472,7 +1482,7 @@ static int nfs4_xdr_enc_rename(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1495,7 +1505,7 @@ static int nfs4_xdr_enc_link(struct rpc_rqst *req, __be32 *p, const struct nfs4_
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1518,7 +1528,7 @@ static int nfs4_xdr_enc_create(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1549,7 +1559,7 @@ static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, __be32 *p, const struct nf
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1567,7 +1577,7 @@ static int nfs4_xdr_enc_close(struct rpc_rqst *req, __be32 *p, struct nfs_closea
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1586,7 +1596,7 @@ static int nfs4_xdr_enc_open(struct rpc_rqst *req, __be32 *p, struct nfs_openarg
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1627,7 +1637,7 @@ static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, __be32 *p, struct nfs_
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1646,7 +1656,7 @@ static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, __be32 *p, struct n
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops	= 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1665,7 +1675,7 @@ static int nfs4_xdr_enc_lock(struct rpc_rqst *req, __be32 *p, struct nfs_lock_ar
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1683,7 +1693,7 @@ static int nfs4_xdr_enc_lockt(struct rpc_rqst *req, __be32 *p, struct nfs_lockt_
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1701,7 +1711,7 @@ static int nfs4_xdr_enc_locku(struct rpc_rqst *req, __be32 *p, struct nfs_locku_
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1719,7 +1729,7 @@ static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct n
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1740,7 +1750,7 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1764,7 +1774,7 @@ static int nfs4_xdr_enc_read(struct rpc_rqst *req, __be32 *p, struct nfs_readarg
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1786,7 +1796,7 @@ static int nfs4_xdr_enc_setattr(struct rpc_rqst *req, __be32 *p, struct nfs_seta
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1807,7 +1817,7 @@ nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p,
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 	uint32_t replen;
 
@@ -1830,7 +1840,7 @@ static int nfs4_xdr_enc_write(struct rpc_rqst *req, __be32 *p, struct nfs_writea
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1850,7 +1860,7 @@ static int nfs4_xdr_enc_commit(struct rpc_rqst *req, __be32 *p, struct nfs_write
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1869,7 +1879,7 @@ static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs4_fsin
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops	= 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1887,7 +1897,7 @@ static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, __be32 *p, const struct n
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1906,7 +1916,7 @@ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, __be32 *p, const struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -1926,7 +1936,7 @@ static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p,
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -2001,7 +2011,7 @@ static int nfs4_xdr_enc_delegreturn(struct rpc_rqst *req, __be32 *p, const struc
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
@@ -2020,7 +2030,7 @@ static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 	uint32_t replen;
 
@@ -3996,7 +4006,7 @@ nfs4_xdr_enc_setacl(struct rpc_rqst *req, __be32 *p, struct nfs_setaclargs *args
 {
 	struct xdr_stream xdr;
 	struct compound_hdr hdr = {
-		.nops   = 0,
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args),
 	};
 	int status;
 
-- 
cgit v1.2.3-59-g8ed1b


From 9b7b9fcc9c124b8a2a079f748239ce9b7a8d8304 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:11 -0400
Subject: nfs41: xdr {encode,decode}_sequence

Implement stubs for encode and decode sequence, defined as no-ops when
CONFIG_NFS_V4_1 is not defined.
Add the nfsv41 encode and decode sizes. Add encode_sequence to all
nfs4_enc_* routines and decode_sequence to all nfs4_dec_* routines as required
by v41.

[was nfs41: minorversion support for xdr]
[added nfs_client argument to encode_sequence so not to use sequence_args to pass sa_session]
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4xdr.c | 242 ++++++++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 232 insertions(+), 10 deletions(-)

diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 19ca0b519a1d..5b944cd57218 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -244,43 +244,63 @@ static int nfs4_stat_to_errno(int);
 				(encode_getattr_maxsz)
 #define decode_fs_locations_maxsz \
 				(0)
+
+#if defined(CONFIG_NFS_V4_1)
+#define encode_sequence_maxsz	0 /* stub */
+#define decode_sequence_maxsz	0 /* stub */
+#else /* CONFIG_NFS_V4_1 */
+#define encode_sequence_maxsz	0
+#define decode_sequence_maxsz	0
+#endif /* CONFIG_NFS_V4_1 */
+
 #define NFS4_enc_compound_sz	(1024)  /* XXX: large enough? */
 #define NFS4_dec_compound_sz	(1024)  /* XXX: large enough? */
 #define NFS4_enc_read_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_read_maxsz)
 #define NFS4_dec_read_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_read_maxsz)
 #define NFS4_enc_readlink_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_readlink_maxsz)
 #define NFS4_dec_readlink_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_readlink_maxsz)
 #define NFS4_enc_readdir_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_readdir_maxsz)
 #define NFS4_dec_readdir_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_readdir_maxsz)
 #define NFS4_enc_write_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_write_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_write_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_write_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_commit_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_commit_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_commit_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_commit_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_open_sz        (compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_savefh_maxsz + \
 				encode_open_maxsz + \
@@ -289,6 +309,7 @@ static int nfs4_stat_to_errno(int);
 				encode_restorefh_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_open_sz        (compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_savefh_maxsz + \
 				decode_open_maxsz + \
@@ -305,43 +326,53 @@ static int nfs4_stat_to_errno(int);
 				 decode_putfh_maxsz + \
 				 decode_open_confirm_maxsz)
 #define NFS4_enc_open_noattr_sz	(compound_encode_hdr_maxsz + \
+					encode_sequence_maxsz + \
 					encode_putfh_maxsz + \
 					encode_open_maxsz + \
 					encode_getattr_maxsz)
 #define NFS4_dec_open_noattr_sz	(compound_decode_hdr_maxsz + \
+					decode_sequence_maxsz + \
 					decode_putfh_maxsz + \
 					decode_open_maxsz + \
 					decode_getattr_maxsz)
 #define NFS4_enc_open_downgrade_sz \
 				(compound_encode_hdr_maxsz + \
+				 encode_sequence_maxsz + \
 				 encode_putfh_maxsz + \
 				 encode_open_downgrade_maxsz + \
 				 encode_getattr_maxsz)
 #define NFS4_dec_open_downgrade_sz \
 				(compound_decode_hdr_maxsz + \
+				 decode_sequence_maxsz + \
 				 decode_putfh_maxsz + \
 				 decode_open_downgrade_maxsz + \
 				 decode_getattr_maxsz)
 #define NFS4_enc_close_sz	(compound_encode_hdr_maxsz + \
+				 encode_sequence_maxsz + \
 				 encode_putfh_maxsz + \
 				 encode_close_maxsz + \
 				 encode_getattr_maxsz)
 #define NFS4_dec_close_sz	(compound_decode_hdr_maxsz + \
+				 decode_sequence_maxsz + \
 				 decode_putfh_maxsz + \
 				 decode_close_maxsz + \
 				 decode_getattr_maxsz)
 #define NFS4_enc_setattr_sz	(compound_encode_hdr_maxsz + \
+				 encode_sequence_maxsz + \
 				 encode_putfh_maxsz + \
 				 encode_setattr_maxsz + \
 				 encode_getattr_maxsz)
 #define NFS4_dec_setattr_sz	(compound_decode_hdr_maxsz + \
+				 decode_sequence_maxsz + \
 				 decode_putfh_maxsz + \
 				 decode_setattr_maxsz + \
 				 decode_getattr_maxsz)
 #define NFS4_enc_fsinfo_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_fsinfo_maxsz)
 #define NFS4_dec_fsinfo_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_fsinfo_maxsz)
 #define NFS4_enc_renew_sz	(compound_encode_hdr_maxsz + \
@@ -363,64 +394,81 @@ static int nfs4_stat_to_errno(int);
 				decode_putrootfh_maxsz + \
 				decode_fsinfo_maxsz)
 #define NFS4_enc_lock_sz        (compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_lock_maxsz)
 #define NFS4_dec_lock_sz        (compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_lock_maxsz)
 #define NFS4_enc_lockt_sz       (compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_lockt_maxsz)
 #define NFS4_dec_lockt_sz       (compound_decode_hdr_maxsz + \
+				 decode_sequence_maxsz + \
 				 decode_putfh_maxsz + \
 				 decode_lockt_maxsz)
 #define NFS4_enc_locku_sz       (compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_locku_maxsz)
 #define NFS4_dec_locku_sz       (compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_locku_maxsz)
 #define NFS4_enc_access_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_access_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_access_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_access_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_getattr_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_getattr_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_lookup_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_lookup_maxsz + \
 				encode_getattr_maxsz + \
 				encode_getfh_maxsz)
 #define NFS4_dec_lookup_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_lookup_maxsz + \
 				decode_getattr_maxsz + \
 				decode_getfh_maxsz)
 #define NFS4_enc_lookup_root_sz (compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putrootfh_maxsz + \
 				encode_getattr_maxsz + \
 				encode_getfh_maxsz)
 #define NFS4_dec_lookup_root_sz (compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putrootfh_maxsz + \
 				decode_getattr_maxsz + \
 				decode_getfh_maxsz)
 #define NFS4_enc_remove_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_remove_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_remove_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_remove_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_rename_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_savefh_maxsz + \
 				encode_putfh_maxsz + \
@@ -429,6 +477,7 @@ static int nfs4_stat_to_errno(int);
 				encode_restorefh_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_rename_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_savefh_maxsz + \
 				decode_putfh_maxsz + \
@@ -437,6 +486,7 @@ static int nfs4_stat_to_errno(int);
 				decode_restorefh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_link_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_savefh_maxsz + \
 				encode_putfh_maxsz + \
@@ -445,6 +495,7 @@ static int nfs4_stat_to_errno(int);
 				encode_restorefh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_dec_link_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_savefh_maxsz + \
 				decode_putfh_maxsz + \
@@ -453,16 +504,19 @@ static int nfs4_stat_to_errno(int);
 				decode_restorefh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_symlink_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_symlink_maxsz + \
 				encode_getattr_maxsz + \
 				encode_getfh_maxsz)
 #define NFS4_dec_symlink_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_symlink_maxsz + \
 				decode_getattr_maxsz + \
 				decode_getfh_maxsz)
 #define NFS4_enc_create_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_savefh_maxsz + \
 				encode_create_maxsz + \
@@ -471,6 +525,7 @@ static int nfs4_stat_to_errno(int);
 				encode_restorefh_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_create_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_savefh_maxsz + \
 				decode_create_maxsz + \
@@ -479,49 +534,63 @@ static int nfs4_stat_to_errno(int);
 				decode_restorefh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_pathconf_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_pathconf_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_statfs_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_statfs_maxsz)
 #define NFS4_dec_statfs_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_statfs_maxsz)
 #define NFS4_enc_server_caps_sz (compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_server_caps_sz (compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_delegreturn_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_delegreturn_maxsz + \
 				encode_getattr_maxsz)
 #define NFS4_dec_delegreturn_sz (compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_delegreturn_maxsz + \
 				decode_getattr_maxsz)
 #define NFS4_enc_getacl_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_getacl_maxsz)
 #define NFS4_dec_getacl_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_getacl_maxsz)
 #define NFS4_enc_setacl_sz	(compound_encode_hdr_maxsz + \
+				encode_sequence_maxsz + \
 				encode_putfh_maxsz + \
 				encode_setacl_maxsz)
 #define NFS4_dec_setacl_sz	(compound_decode_hdr_maxsz + \
+				decode_sequence_maxsz + \
 				decode_putfh_maxsz + \
 				decode_setacl_maxsz)
 #define NFS4_enc_fs_locations_sz \
 				(compound_encode_hdr_maxsz + \
+				 encode_sequence_maxsz + \
 				 encode_putfh_maxsz + \
 				 encode_lookup_maxsz + \
 				 encode_fs_locations_maxsz)
 #define NFS4_dec_fs_locations_sz \
 				(compound_decode_hdr_maxsz + \
+				 decode_sequence_maxsz + \
 				 decode_putfh_maxsz + \
 				 decode_lookup_maxsz + \
 				 decode_fs_locations_maxsz)
@@ -1385,6 +1454,24 @@ static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *state
 	hdr->nops++;
 	hdr->replen += decode_delegreturn_maxsz;
 }
+
+/* NFSv4.1 operations */
+static void encode_sequence(struct xdr_stream *xdr,
+			    const struct nfs4_sequence_args *args,
+			    struct compound_hdr *hdr)
+{
+#if defined(CONFIG_NFS_V4_1)
+	struct nfs4_session *session = args->sa_session;
+
+	if (!session)
+		return;
+
+	/* stub */
+	hdr->nops++;
+	hdr->replen += decode_sequence_maxsz;
+#endif /* CONFIG_NFS_V4_1 */
+}
+
 /*
  * END OF "GENERIC" ENCODE ROUTINES.
  */
@@ -1410,6 +1497,7 @@ static int nfs4_xdr_enc_access(struct rpc_rqst *req, __be32 *p, const struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_access(&xdr, args->access, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1429,6 +1517,7 @@ static int nfs4_xdr_enc_lookup(struct rpc_rqst *req, __be32 *p, const struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_lookup(&xdr, args->name, &hdr);
 	encode_getfh(&xdr, &hdr);
@@ -1449,6 +1538,7 @@ static int nfs4_xdr_enc_lookup_root(struct rpc_rqst *req, __be32 *p, const struc
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putrootfh(&xdr, &hdr);
 	encode_getfh(&xdr, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1468,6 +1558,7 @@ static int nfs4_xdr_enc_remove(struct rpc_rqst *req, __be32 *p, const struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_remove(&xdr, &args->name, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1487,6 +1578,7 @@ static int nfs4_xdr_enc_rename(struct rpc_rqst *req, __be32 *p, const struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->old_dir, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_putfh(&xdr, args->new_dir, &hdr);
@@ -1510,6 +1602,7 @@ static int nfs4_xdr_enc_link(struct rpc_rqst *req, __be32 *p, const struct nfs4_
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
@@ -1533,6 +1626,7 @@ static int nfs4_xdr_enc_create(struct rpc_rqst *req, __be32 *p, const struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_create(&xdr, args, &hdr);
@@ -1564,6 +1658,7 @@ static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, __be32 *p, const struct nf
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
 	encode_nops(&hdr);
@@ -1582,6 +1677,7 @@ static int nfs4_xdr_enc_close(struct rpc_rqst *req, __be32 *p, struct nfs_closea
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_close(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1601,6 +1697,7 @@ static int nfs4_xdr_enc_open(struct rpc_rqst *req, __be32 *p, struct nfs_openarg
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_savefh(&xdr, &hdr);
 	encode_open(&xdr, args, &hdr);
@@ -1642,6 +1739,7 @@ static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, __be32 *p, struct nfs_
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_open(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1661,6 +1759,7 @@ static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, __be32 *p, struct n
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_open_downgrade(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1680,6 +1779,7 @@ static int nfs4_xdr_enc_lock(struct rpc_rqst *req, __be32 *p, struct nfs_lock_ar
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_lock(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1698,6 +1798,7 @@ static int nfs4_xdr_enc_lockt(struct rpc_rqst *req, __be32 *p, struct nfs_lockt_
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_lockt(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1716,6 +1817,7 @@ static int nfs4_xdr_enc_locku(struct rpc_rqst *req, __be32 *p, struct nfs_locku_
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_locku(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -1734,6 +1836,7 @@ static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct n
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_readlink(&xdr, args, req, &hdr);
 
@@ -1755,6 +1858,7 @@ static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nf
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_readdir(&xdr, args, req, &hdr);
 
@@ -1779,6 +1883,7 @@ static int nfs4_xdr_enc_read(struct rpc_rqst *req, __be32 *p, struct nfs_readarg
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_read(&xdr, args, &hdr);
 
@@ -1801,6 +1906,7 @@ static int nfs4_xdr_enc_setattr(struct rpc_rqst *req, __be32 *p, struct nfs_seta
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_setattr(&xdr, args, args->server, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1823,6 +1929,7 @@ nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p,
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	replen = hdr.replen + nfs4_fattr_bitmap_maxsz + 1;
 	encode_getattr_two(&xdr, FATTR4_WORD0_ACL, 0, &hdr);
@@ -1845,6 +1952,7 @@ static int nfs4_xdr_enc_write(struct rpc_rqst *req, __be32 *p, struct nfs_writea
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_write(&xdr, args, &hdr);
 	req->rq_snd_buf.flags |= XDRBUF_WRITE;
@@ -1865,6 +1973,7 @@ static int nfs4_xdr_enc_commit(struct rpc_rqst *req, __be32 *p, struct nfs_write
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_commit(&xdr, args, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -1884,6 +1993,7 @@ static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs4_fsin
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_fsinfo(&xdr, args->bitmask, &hdr);
 	encode_nops(&hdr);
@@ -1902,6 +2012,7 @@ static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, __be32 *p, const struct n
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getattr_one(&xdr, args->bitmask[0] & nfs4_pathconf_bitmap[0],
 			   &hdr);
@@ -1921,6 +2032,7 @@ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, __be32 *p, const struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	encode_getattr_two(&xdr, args->bitmask[0] & nfs4_statfs_bitmap[0],
 			   args->bitmask[1] & nfs4_statfs_bitmap[1], &hdr);
@@ -1941,6 +2053,7 @@ static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p,
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fhandle, &hdr);
 	encode_getattr_one(&xdr, FATTR4_WORD0_SUPPORTED_ATTRS|
 			   FATTR4_WORD0_LINK_SUPPORT|
@@ -2016,6 +2129,7 @@ static int nfs4_xdr_enc_delegreturn(struct rpc_rqst *req, __be32 *p, const struc
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fhandle, &hdr);
 	encode_delegreturn(&xdr, args->stateid, &hdr);
 	encode_getfattr(&xdr, args->bitmask, &hdr);
@@ -2036,6 +2150,7 @@ static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->dir_fh, &hdr);
 	encode_lookup(&xdr, args->name, &hdr);
 	replen = hdr.replen;	/* get the attribute into args->page */
@@ -3762,6 +3877,20 @@ static int decode_delegreturn(struct xdr_stream *xdr)
 	return decode_op_hdr(xdr, OP_DELEGRETURN);
 }
 
+static int decode_sequence(struct xdr_stream *xdr,
+			   struct nfs4_sequence_res *res,
+			   struct rpc_rqst *rqstp)
+{
+#if defined(CONFIG_NFS_V4_1)
+	if (!res->sr_session)
+		return 0;
+
+	/* stub */
+#endif /* CONFIG_NFS_V4_1 */
+
+	return 0;
+}
+
 /*
  * END OF "GENERIC" DECODE ROUTINES.
  */
@@ -3777,6 +3906,9 @@ static int nfs4_xdr_dec_open_downgrade(struct rpc_rqst *rqstp, __be32 *p, struct
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -3800,7 +3932,11 @@ static int nfs4_xdr_dec_access(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_ac
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
 	if (status != 0)
@@ -3823,7 +3959,11 @@ static int nfs4_xdr_dec_lookup(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_lo
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
@@ -3846,7 +3986,11 @@ static int nfs4_xdr_dec_lookup_root(struct rpc_rqst *rqstp, __be32 *p, struct nf
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	if ((status = decode_putrootfh(&xdr)) != 0)
 		goto out;
@@ -3866,7 +4010,11 @@ static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, __be32 *p, struct nfs_rem
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
@@ -3887,7 +4035,11 @@ static int nfs4_xdr_dec_rename(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_re
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
@@ -3917,7 +4069,11 @@ static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_link
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
@@ -3950,7 +4106,11 @@ static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_cr
 	int status;
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
@@ -3988,6 +4148,9 @@ static int nfs4_xdr_dec_getattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_g
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4012,6 +4175,7 @@ nfs4_xdr_enc_setacl(struct rpc_rqst *req, __be32 *p, struct nfs_setaclargs *args
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
 	status = encode_setacl(&xdr, args, &hdr);
 	encode_nops(&hdr);
@@ -4031,6 +4195,9 @@ nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p,
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4054,6 +4221,9 @@ nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p,
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4076,6 +4246,9 @@ static int nfs4_xdr_dec_close(struct rpc_rqst *rqstp, __be32 *p, struct nfs_clos
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4106,6 +4279,9 @@ static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, __be32 *p, struct nfs_openr
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4160,6 +4336,9 @@ static int nfs4_xdr_dec_open_noattr(struct rpc_rqst *rqstp, __be32 *p, struct nf
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4184,6 +4363,9 @@ static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs_se
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4208,6 +4390,9 @@ static int nfs4_xdr_dec_lock(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lock_
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4229,6 +4414,9 @@ static int nfs4_xdr_dec_lockt(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lock
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4250,6 +4438,9 @@ static int nfs4_xdr_dec_locku(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lock
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4272,6 +4463,9 @@ static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, __be32 *p,
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4293,6 +4487,9 @@ static int nfs4_xdr_dec_readdir(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_r
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4314,6 +4511,9 @@ static int nfs4_xdr_dec_read(struct rpc_rqst *rqstp, __be32 *p, struct nfs_readr
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4337,6 +4537,9 @@ static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, __be32 *p, struct nfs_writ
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4363,6 +4566,9 @@ static int nfs4_xdr_dec_commit(struct rpc_rqst *rqstp, __be32 *p, struct nfs_wri
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
 	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
@@ -4388,6 +4594,8 @@ static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p,
 
 	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_sequence(&xdr, &res->seq_res, req);
 	if (!status)
 		status = decode_putfh(&xdr);
 	if (!status)
@@ -4407,6 +4615,8 @@ static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p,
 
 	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_sequence(&xdr, &res->seq_res, req);
 	if (!status)
 		status = decode_putfh(&xdr);
 	if (!status)
@@ -4426,6 +4636,8 @@ static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p,
 
 	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_sequence(&xdr, &res->seq_res, req);
 	if (!status)
 		status = decode_putfh(&xdr);
 	if (!status)
@@ -4443,7 +4655,11 @@ static int nfs4_xdr_dec_server_caps(struct rpc_rqst *req, __be32 *p, struct nfs4
 	int status;
 
 	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
-	if ((status = decode_compound_hdr(&xdr, &hdr)) != 0)
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, req);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
@@ -4516,7 +4732,10 @@ static int nfs4_xdr_dec_delegreturn(struct rpc_rqst *rqstp, __be32 *p, struct nf
 
 	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
-	if (status != 0)
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (status)
 		goto out;
 	status = decode_putfh(&xdr);
 	if (status != 0)
@@ -4539,7 +4758,10 @@ static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p,
 
 	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
 	status = decode_compound_hdr(&xdr, &hdr);
-	if (status != 0)
+	if (status)
+		goto out;
+	status = decode_sequence(&xdr, &res->seq_res, req);
+	if (status)
 		goto out;
 	if ((status = decode_putfh(&xdr)) != 0)
 		goto out;
-- 
cgit v1.2.3-59-g8ed1b


From ce5039c1be1fd04c2d9b1af80d41a6d003f2e20c Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:13 -0400
Subject: nfs41: nfs4_setup_sequence

Perform the nfs4_setup_sequence in the rpc_call_prepare state.

If a session slot is not available, we will rpc_sleep_on the
slot wait queue leaving the tk_action as rpc_call_prepare.

Once we have a session slot, hang on to it even through rpc_restart_calls.
Ensure the nfs41_sequence_res sr_slot pointer is NULL before rpc_run_task is
called as nfs41_setup_sequence will only find a new slot if it is NULL.

A future patch will call free slot after any rpc_restart_calls, and handle the
rpc restart that result from a sequence operation error.

Signed-off-by: Rahul Iyer <iyer@netapp.com>
[nfs41: sequence res use slotid]
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: simplify nfs4_call_sync]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_call_sync]
[nfs41: check for session not minorversion]
[nfs41: remove rpc_message from nfs41_call_sync_args]
[moved NFS4_MAX_SLOT_TABLE logic into nfs41_setup_sequence]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs41_call_sync_data use nfs_client not nfs_server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: expose nfs4_call_sync_session for lease renewal]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove unnecessary return check]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 100 ++++++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 98 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 5878930b4c3b..90aed76e593b 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -273,14 +273,110 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp
 
 #if defined(CONFIG_NFS_V4_1)
 
+static int nfs41_setup_sequence(struct nfs4_session *session,
+				struct nfs4_sequence_args *args,
+				struct nfs4_sequence_res *res,
+				int cache_reply,
+				struct rpc_task *task)
+{
+	/* slot already allocated? */
+	if (res->sr_slotid != NFS4_MAX_SLOT_TABLE)
+		return 0;
+
+	memset(res, 0, sizeof(*res));
+	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
+	return 0;
+}
+
+int nfs4_setup_sequence(struct nfs_client *clp,
+			struct nfs4_sequence_args *args,
+			struct nfs4_sequence_res *res,
+			int cache_reply,
+			struct rpc_task *task)
+{
+	int ret = 0;
+
+	dprintk("--> %s clp %p session %p sr_slotid %d\n",
+		__func__, clp, clp->cl_session, res->sr_slotid);
+
+	if (!nfs4_has_session(clp))
+		goto out;
+	ret = nfs41_setup_sequence(clp->cl_session, args, res, cache_reply,
+				   task);
+	if (ret != -EAGAIN) {
+		/* terminate rpc task */
+		task->tk_status = ret;
+		task->tk_action = NULL;
+	}
+out:
+	dprintk("<-- %s status=%d\n", __func__, ret);
+	return ret;
+}
+
+struct nfs41_call_sync_data {
+	struct nfs_client *clp;
+	struct nfs4_sequence_args *seq_args;
+	struct nfs4_sequence_res *seq_res;
+	int cache_reply;
+};
+
+static void nfs41_call_sync_prepare(struct rpc_task *task, void *calldata)
+{
+	struct nfs41_call_sync_data *data = calldata;
+
+	dprintk("--> %s data->clp->cl_session %p\n", __func__,
+		data->clp->cl_session);
+	if (nfs4_setup_sequence(data->clp, data->seq_args,
+				data->seq_res, data->cache_reply, task))
+		return;
+	rpc_call_start(task);
+}
+
+struct rpc_call_ops nfs41_call_sync_ops = {
+	.rpc_call_prepare = nfs41_call_sync_prepare,
+};
+
+static int nfs4_call_sync_sequence(struct nfs_client *clp,
+				   struct rpc_clnt *clnt,
+				   struct rpc_message *msg,
+				   struct nfs4_sequence_args *args,
+				   struct nfs4_sequence_res *res,
+				   int cache_reply)
+{
+	int ret;
+	struct rpc_task *task;
+	struct nfs41_call_sync_data data = {
+		.clp = clp,
+		.seq_args = args,
+		.seq_res = res,
+		.cache_reply = cache_reply,
+	};
+	struct rpc_task_setup task_setup = {
+		.rpc_client = clnt,
+		.rpc_message = msg,
+		.callback_ops = &nfs41_call_sync_ops,
+		.callback_data = &data
+	};
+
+	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
+	task = rpc_run_task(&task_setup);
+	if (IS_ERR(task))
+		ret = PTR_ERR(task);
+	else {
+		ret = task->tk_status;
+		rpc_put_task(task);
+	}
+	return ret;
+}
+
 int _nfs4_call_sync_session(struct nfs_server *server,
 			    struct rpc_message *msg,
 			    struct nfs4_sequence_args *args,
 			    struct nfs4_sequence_res *res,
 			    int cache_reply)
 {
-	/* in preparation for setting up the sequence op */
-	return rpc_call_sync(server->client, msg, 0);
+	return nfs4_call_sync_sequence(server->nfs_client, server->client,
+				       msg, args, res, cache_reply);
 }
 
 #endif /* CONFIG_NFS_V4_1 */
-- 
cgit v1.2.3-59-g8ed1b


From 510b81756f18922a4c5b555e8145f4fed5beb569 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:14 -0400
Subject: nfs41: find slot

Find a free slot using bitmap-based allocation.
Use the optimized ffz function to find a zero bit
in the bitmap that indicates a free slot, starting
the search from the 'lowest_free_slotid' position.

If found, mark the slot as used in the bitmap, get
the slot's slotid and seqid, and update max_slotid
to be used by the SEQUENCE operation.

Also, update lowest_free_slotid for next search.

If no free slot was found the caller has to wait
for a free slot (outside the scope of this function)

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: find slot return slotid]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[use find_first_zero_bit for nfs4_find_slot as per review comment 21/85.]
[use NFS4_MAX_SLOT_TABLE rather than NFS4_NO_SLOT]
[nfs41: rpc_sleep_on slot_tbl_waitq must be called under slot_tbl_lock]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 33 +++++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 90aed76e593b..a0946a0d116e 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -273,6 +273,39 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp
 
 #if defined(CONFIG_NFS_V4_1)
 
+/*
+ * nfs4_find_slot - efficiently look for a free slot
+ *
+ * nfs4_find_slot looks for an unset bit in the used_slots bitmap.
+ * If found, we mark the slot as used, update the highest_used_slotid,
+ * and respectively set up the sequence operation args.
+ * The slot number is returned if found, or NFS4_MAX_SLOT_TABLE otherwise.
+ */
+static u8
+nfs4_find_slot(struct nfs4_slot_table *tbl, struct rpc_task *task)
+{
+	int slotid;
+	u8 ret_id = NFS4_MAX_SLOT_TABLE;
+	BUILD_BUG_ON((u8)NFS4_MAX_SLOT_TABLE != (int)NFS4_MAX_SLOT_TABLE);
+
+	spin_lock(&tbl->slot_tbl_lock);
+	dprintk("--> %s used_slots=%04lx highest_used=%d max_slots=%d\n",
+		__func__, tbl->used_slots[0], tbl->highest_used_slotid,
+		tbl->max_slots);
+	slotid = find_first_zero_bit(tbl->used_slots, tbl->max_slots);
+	if (slotid >= tbl->max_slots)
+		goto out;
+	__set_bit(slotid, tbl->used_slots);
+	if (slotid > tbl->highest_used_slotid)
+		tbl->highest_used_slotid = slotid;
+	ret_id = slotid;
+out:
+	dprintk("<-- %s used_slots=%04lx highest_used=%d slotid=%d \n",
+		__func__, tbl->used_slots[0], tbl->highest_used_slotid, ret_id);
+	spin_unlock(&tbl->slot_tbl_lock);
+	return ret_id;
+}
+
 static int nfs41_setup_sequence(struct nfs4_session *session,
 				struct nfs4_sequence_args *args,
 				struct nfs4_sequence_res *res,
-- 
cgit v1.2.3-59-g8ed1b


From fbcd4abcb3841f85578985c09c6df85aa41b0ae8 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:15 -0400
Subject: nfs41: setup_sequence method

Allocate a slot in the session slot table and set the sequence op arguments.

Called at the rpc prepare stage.

Add a status to nfs41_sequence_res, initialize it to one so that we catch
rpc level failures which do not go through decode_sequence which sets
the new status field.

Note that upon an rpc level failure, we don't know if the server processed the
sequence operation or not. Proceed as if the server did process the sequence
operation.

Signed-off-by: Rahul Iyer <iyer@netapp.com>
[nfs41: sequence args use slotid]
[nfs41: find slot return slotid]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove SEQ4_STATUS_USE_TK_STATUS]
As per 11-14-08 review
[move extern declaration from nfs41: sequence setup/done support]
[removed sa_session definition, changed sa_cache_this into a u8 to reduce footprint]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: rpc_sleep_on slot_tbl_waitq must be called under slot_tbl_lock]
    Otherwise there's a race (we've hit) with nfs4_free_slot where
    nfs41_setup_sequence sees a full slot table, unlocks slot_tbl_lock,
    nfs4_free_slots happen concurrently and call rpc_wake_up_next
    where there's nobody to wake up yet, context goes back to
    nfs41_setup_sequence which goes to sleep when the slot table
    is actually empty now and there's no-one to wake it up anymore.
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4_fs.h        | 10 ++++++++++
 fs/nfs/nfs4proc.c       | 34 ++++++++++++++++++++++++++++++++--
 include/linux/nfs_xdr.h |  4 ++++
 3 files changed, 46 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index acac6f8c3d39..eccf4e93e7d7 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -203,8 +203,18 @@ extern int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops;
 extern struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops;
 #if defined(CONFIG_NFS_V4_1)
+extern int nfs4_setup_sequence(struct nfs_client *clp,
+		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
+		int cache_reply, struct rpc_task *task);
 extern void nfs4_destroy_session(struct nfs4_session *session);
 extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp);
+#else /* CONFIG_NFS_v4_1 */
+static inline int nfs4_setup_sequence(struct nfs_client *clp,
+		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
+		int cache_reply, struct rpc_task *task)
+{
+	return 0;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 extern const u32 nfs4_fattr_bitmap[2];
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index a0946a0d116e..c9618080317e 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -280,6 +280,8 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp
  * If found, we mark the slot as used, update the highest_used_slotid,
  * and respectively set up the sequence operation args.
  * The slot number is returned if found, or NFS4_MAX_SLOT_TABLE otherwise.
+ *
+ * Note: must be called with under the slot_tbl_lock.
  */
 static u8
 nfs4_find_slot(struct nfs4_slot_table *tbl, struct rpc_task *task)
@@ -288,7 +290,6 @@ nfs4_find_slot(struct nfs4_slot_table *tbl, struct rpc_task *task)
 	u8 ret_id = NFS4_MAX_SLOT_TABLE;
 	BUILD_BUG_ON((u8)NFS4_MAX_SLOT_TABLE != (int)NFS4_MAX_SLOT_TABLE);
 
-	spin_lock(&tbl->slot_tbl_lock);
 	dprintk("--> %s used_slots=%04lx highest_used=%d max_slots=%d\n",
 		__func__, tbl->used_slots[0], tbl->highest_used_slotid,
 		tbl->max_slots);
@@ -302,7 +303,6 @@ nfs4_find_slot(struct nfs4_slot_table *tbl, struct rpc_task *task)
 out:
 	dprintk("<-- %s used_slots=%04lx highest_used=%d slotid=%d \n",
 		__func__, tbl->used_slots[0], tbl->highest_used_slotid, ret_id);
-	spin_unlock(&tbl->slot_tbl_lock);
 	return ret_id;
 }
 
@@ -312,12 +312,42 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 				int cache_reply,
 				struct rpc_task *task)
 {
+	struct nfs4_slot *slot;
+	struct nfs4_slot_table *tbl;
+	u8 slotid;
+
+	dprintk("--> %s\n", __func__);
 	/* slot already allocated? */
 	if (res->sr_slotid != NFS4_MAX_SLOT_TABLE)
 		return 0;
 
 	memset(res, 0, sizeof(*res));
 	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
+	tbl = &session->fc_slot_table;
+
+	spin_lock(&tbl->slot_tbl_lock);
+	slotid = nfs4_find_slot(tbl, task);
+	if (slotid == NFS4_MAX_SLOT_TABLE) {
+		rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
+		spin_unlock(&tbl->slot_tbl_lock);
+		dprintk("<-- %s: no free slots\n", __func__);
+		return -EAGAIN;
+	}
+	spin_unlock(&tbl->slot_tbl_lock);
+
+	slot = tbl->slots + slotid;
+	args->sa_slotid = slotid;
+	args->sa_cache_this = cache_reply;
+
+	dprintk("<-- %s slotid=%d seqid=%d\n", __func__, slotid, slot->seq_nr);
+
+	res->sr_slotid = slotid;
+	res->sr_renewal_time = jiffies;
+	/*
+	 * sr_status is only set in decode_sequence, and so will remain
+	 * set to 1 if an rpc level failure occurs.
+	 */
+	res->sr_status = 1;
 	return 0;
 }
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index db0d1236aae7..4ac14b40efc9 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -163,11 +163,15 @@ struct nfs4_slot {
 
 struct nfs4_sequence_args {
 	struct nfs4_session	*sa_session;
+	u8			sa_slotid;
+	u8			sa_cache_this;
 };
 
 struct nfs4_sequence_res {
 	struct nfs4_session	*sr_session;
 	u8			sr_slotid;	/* slot used to send request */
+	unsigned long		sr_renewal_time;
+	int			sr_status;	/* sequence operation status */
 };
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From e2c4ab3ce2ecc527672bd7e29a594c50d3ec0477 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:16 -0400
Subject: nfs41: free slot

Free a slot in the slot table.

Mark the slot as free in the bitmap-based allocation table
by clearing a bit corresponding to the slotid.

Update lowest_free_slotid if freed slotid is lower than that.
Update highest_used_slotid.  In the case the freed slotid
equals the highest_used_slotid, scan downwards for the next
highest used slotid using the optimized fls* functions.

Finally, wake up thread waiting on slot_tbl_waitq for a free slot
to become available.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: free slot use slotid]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use find_first_zero_bit for nfs4_find_slot]
    While at it, obliterate lowest_free_slotid and fix-up related comments.
    As per review comment 21/85.
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use __clear_bit for nfs4_free_slot]
    While at it, fix-up function comment.
    Part of review comment 22/85.
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use find_last_bit in nfs4_free_slot to determine highest used slot.]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: rpc_sleep_on slot_tbl_waitq must be called under slot_tbl_lock]
    Otherwise there's a race (we've hit) with nfs4_free_slot where
    nfs41_setup_sequence sees a full slot table, unlocks slot_tbl_lock,
    nfs4_free_slots happen concurrently and call rpc_wake_up_next
    where there's nobody to wake up yet, context goes back to
    nfs41_setup_sequence which goes to sleep when the slot table
    is actually empty now and there's no-one to wake it up anymore.
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index c9618080317e..b6308f6740ec 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -273,6 +273,42 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp
 
 #if defined(CONFIG_NFS_V4_1)
 
+/*
+ * nfs4_free_slot - free a slot and efficiently update slot table.
+ *
+ * freeing a slot is trivially done by clearing its respective bit
+ * in the bitmap.
+ * If the freed slotid equals highest_used_slotid we want to update it
+ * so that the server would be able to size down the slot table if needed,
+ * otherwise we know that the highest_used_slotid is still in use.
+ * When updating highest_used_slotid there may be "holes" in the bitmap
+ * so we need to scan down from highest_used_slotid to 0 looking for the now
+ * highest slotid in use.
+ * If none found, highest_used_slotid is set to -1.
+ */
+static void
+nfs4_free_slot(struct nfs4_slot_table *tbl, u8 free_slotid)
+{
+	int slotid = free_slotid;
+
+	spin_lock(&tbl->slot_tbl_lock);
+	/* clear used bit in bitmap */
+	__clear_bit(slotid, tbl->used_slots);
+
+	/* update highest_used_slotid when it is freed */
+	if (slotid == tbl->highest_used_slotid) {
+		slotid = find_last_bit(tbl->used_slots, tbl->max_slots);
+		if (slotid >= 0 && slotid < tbl->max_slots)
+			tbl->highest_used_slotid = slotid;
+		else
+			tbl->highest_used_slotid = -1;
+	}
+	rpc_wake_up_next(&tbl->slot_tbl_waitq);
+	spin_unlock(&tbl->slot_tbl_lock);
+	dprintk("%s: free_slotid %u highest_used_slotid %d\n", __func__,
+		free_slotid, tbl->highest_used_slotid);
+}
+
 /*
  * nfs4_find_slot - efficiently look for a free slot
  *
-- 
cgit v1.2.3-59-g8ed1b


From 13615871cdf8e8263626a817c350c6978a6483fe Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:17 -0400
Subject: nfs41: nfs41_sequence_free_slot

[from nfs41: separate free slot from sequence done]

Don't free the slot until after all rpc_restart_calls have completed.
Session reset will require more work.

As noted by Trond, since we're using rpc_wake_up_next rather than
rpc_wake_up() we must always wake up the next task in the queue
either by going through nfs4_free_slot, or just calling
rpc_wake_up_next if no slot is to be freed.

[nfs41: sequence res use slotid]
[nfs41: remove SEQ4_STATUS_USE_TK_STATUS]
[got rid of nfs4_sequence_res.sr_session, use nfs_client.cl_session instead]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: rpc_wake_up_next if sessions slot was not consumed.]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/internal.h | 14 ++++++++++++++
 fs/nfs/nfs4proc.c | 21 +++++++++++++++++++++
 2 files changed, 35 insertions(+)

diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 8d67c2865dc3..12f4c5e6fd39 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -230,6 +230,20 @@ static inline int nfs4_has_session(const struct nfs_client *clp)
 	return 0;
 }
 
+#ifdef CONFIG_NFS_V4_1
+extern void nfs41_sequence_free_slot(const struct nfs_client *,
+				     struct nfs4_sequence_res *res);
+#endif /* CONFIG_NFS_V4_1 */
+
+static inline void nfs4_sequence_free_slot(const struct nfs_client *clp,
+					   struct nfs4_sequence_res *res)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (nfs4_has_session(clp))
+		nfs41_sequence_free_slot(clp, res);
+#endif /* CONFIG_NFS_V4_1 */
+}
+
 /*
  * Determine the device name as a string
  */
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index b6308f6740ec..861b103c9e3c 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -309,6 +309,27 @@ nfs4_free_slot(struct nfs4_slot_table *tbl, u8 free_slotid)
 		free_slotid, tbl->highest_used_slotid);
 }
 
+void nfs41_sequence_free_slot(const struct nfs_client *clp,
+			      struct nfs4_sequence_res *res)
+{
+	struct nfs4_slot_table *tbl;
+
+	if (!nfs4_has_session(clp)) {
+		dprintk("%s: No session\n", __func__);
+		return;
+	}
+	tbl = &clp->cl_session->fc_slot_table;
+	if (res->sr_slotid == NFS4_MAX_SLOT_TABLE) {
+		dprintk("%s: No slot\n", __func__);
+		/* just wake up the next guy waiting since
+		 * we may have not consumed a slot after all */
+		rpc_wake_up_next(&tbl->slot_tbl_waitq);
+		return;
+	}
+	nfs4_free_slot(tbl, res->sr_slotid);
+	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
+}
+
 /*
  * nfs4_find_slot - efficiently look for a free slot
  *
-- 
cgit v1.2.3-59-g8ed1b


From b0df806c0f9d61e5fcfdb0f4e606e1a59303aaa6 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:18 -0400
Subject: nfs41: nfs41_sequence_done

Handle session level errors, update slot sequence id and
sessions bookeeping, free slot.

[nfs41: sequence res use slotid]
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove SEQ4_STATUS_USE_TK_STATUS]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: check for session not minorversion]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: bail out early out of nfs41_sequence_done if !res->sr_session]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[move nfs4_sequence_done from nfs41: nfs41_call_sync_done]
Signed-off-by: Andy Adamson <andros@netapp.com>
[move nfs4_sequence_free_slot from nfs41: separate free slot from sequence done]
    Don't free the slot until after all rpc_restart_calls have completed.
    Session reset will require more work.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[moved reset sr_slotid to nfs41_sequence_free_slot]
[free slot also on unexpectecd error]
[remove seq_res.sr_session member, use nfs_client's instead]
[ditch seq_res.sr_flags until used]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[look at sr_slotid for bailing out early from nfs41_sequence_done]
[nfs41: rpc_wake_up_next if sessions slot was not consumed.]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove unused error checking in nfs41_sequence_done]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: remove nfs4_has_session check in nfs41_sequence_done]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: remove nfs_client pointer check]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 58 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 861b103c9e3c..188f2ce593ec 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -330,6 +330,46 @@ void nfs41_sequence_free_slot(const struct nfs_client *clp,
 	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
 }
 
+static void nfs41_sequence_done(struct nfs_client *clp,
+				struct nfs4_sequence_res *res,
+				int rpc_status)
+{
+	unsigned long timestamp;
+	struct nfs4_slot_table *tbl;
+	struct nfs4_slot *slot;
+
+	/*
+	 * sr_status remains 1 if an RPC level error occurred. The server
+	 * may or may not have processed the sequence operation..
+	 * Proceed as if the server received and processed the sequence
+	 * operation.
+	 */
+	if (res->sr_status == 1)
+		res->sr_status = NFS_OK;
+
+	/* -ERESTARTSYS can result in skipping nfs41_sequence_setup */
+	if (res->sr_slotid == NFS4_MAX_SLOT_TABLE)
+		goto out;
+
+	tbl = &clp->cl_session->fc_slot_table;
+	slot = tbl->slots + res->sr_slotid;
+
+	if (res->sr_status == 0) {
+		/* Update the slot's sequence and clientid lease timer */
+		++slot->seq_nr;
+		timestamp = res->sr_renewal_time;
+		spin_lock(&clp->cl_lock);
+		if (time_before(clp->cl_last_renewal, timestamp))
+			clp->cl_last_renewal = timestamp;
+		spin_unlock(&clp->cl_lock);
+		return;
+	}
+out:
+	/* The session may be reset by one of the error handlers. */
+	dprintk("%s: Error %d free the slot \n", __func__, res->sr_status);
+	nfs41_sequence_free_slot(clp, res);
+}
+
 /*
  * nfs4_find_slot - efficiently look for a free slot
  *
@@ -515,6 +555,24 @@ int _nfs4_call_sync(struct nfs_server *server,
 	(server)->nfs_client->cl_call_sync((server), (msg), &(args)->seq_args, \
 			&(res)->seq_res, (cache_reply))
 
+static void nfs4_sequence_done(const struct nfs_server *server,
+			       struct nfs4_sequence_res *res, int rpc_status)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (nfs4_has_session(server->nfs_client))
+		nfs41_sequence_done(server->nfs_client, res, rpc_status);
+#endif /* CONFIG_NFS_V4_1 */
+}
+
+/* no restart, therefore free slot here */
+static void nfs4_sequence_done_free_slot(const struct nfs_server *server,
+					 struct nfs4_sequence_res *res,
+					 int rpc_status)
+{
+	nfs4_sequence_done(server, res, rpc_status);
+	nfs4_sequence_free_slot(server->nfs_client, res);
+}
+
 static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo)
 {
 	struct nfs_inode *nfsi = NFS_I(dir);
-- 
cgit v1.2.3-59-g8ed1b


From 69ab40c4c3b9622232e31e0ee145875a2853c113 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:19 -0400
Subject: nfs41: nfs41_call_sync_done

Implement nfs4.1 synchronous rpc_call_done method
that essentially just calls nfs4_sequence_done, that turns
around and calls nfs41_sequence_done for minorversion1 rpcs.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: check for session not minorversion]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[move adding nfs4_sequence_free_slot from nfs41-separate-free-slot-from-sequence-done]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs41_call_sync_data use nfs_client not nfs_server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 188f2ce593ec..a8e0b3573594 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -492,8 +492,17 @@ static void nfs41_call_sync_prepare(struct rpc_task *task, void *calldata)
 	rpc_call_start(task);
 }
 
+static void nfs41_call_sync_done(struct rpc_task *task, void *calldata)
+{
+	struct nfs41_call_sync_data *data = calldata;
+
+	nfs41_sequence_done(data->clp, data->seq_res, task->tk_status);
+	nfs41_sequence_free_slot(data->clp, data->seq_res);
+}
+
 struct rpc_call_ops nfs41_call_sync_ops = {
 	.rpc_call_prepare = nfs41_call_sync_prepare,
+	.rpc_call_done = nfs41_call_sync_done,
 };
 
 static int nfs4_call_sync_sequence(struct nfs_client *clp,
-- 
cgit v1.2.3-59-g8ed1b


From 19ddab06ed923d27339b8009d37b0ab3de93cadf Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:20 -0400
Subject: nfs41: close sequence setup/done support

Separate nfs4_close calls from nfs41: sequence setup/done support
Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: separate free slot from sequence done]
[nfs41: sequence res use slotid]
[nfs41: remove SEQ4_STATUS_USE_TK_STATUS]
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index a8e0b3573594..a34fada51446 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1630,6 +1630,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 	struct nfs4_state *state = calldata->state;
 	struct nfs_server *server = NFS_SERVER(calldata->inode);
 
+	nfs4_sequence_done(server, &calldata->res.seq_res, task->tk_status);
 	if (RPC_ASSASSINATED(task))
 		return;
         /* hmm. we are done with the inode, and in the process of freeing
@@ -1652,6 +1653,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 				return;
 			}
 	}
+	nfs4_sequence_free_slot(server->nfs_client, &calldata->res.seq_res);
 	nfs_refresh_inode(calldata->inode, calldata->res.fattr);
 }
 
@@ -1692,6 +1694,10 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data)
 		calldata->arg.fmode = FMODE_WRITE;
 	}
 	calldata->timestamp = jiffies;
+	if (nfs4_setup_sequence((NFS_SERVER(calldata->inode))->nfs_client,
+				&calldata->arg.seq_args, &calldata->res.seq_res,
+				1, task))
+		return;
 	rpc_call_start(task);
 }
 
@@ -1731,7 +1737,7 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 	};
 	int status = -ENOMEM;
 
-	calldata = kmalloc(sizeof(*calldata), GFP_KERNEL);
+	calldata = kzalloc(sizeof(*calldata), GFP_KERNEL);
 	if (calldata == NULL)
 		goto out;
 	calldata->inode = state->inode;
-- 
cgit v1.2.3-59-g8ed1b


From d898528cdb338b145c71049c3cb7b1dfa542a48a Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:21 -0400
Subject: nfs41: open sequence setup/done support

Separate nfs4_open calls from nfs41: sequence setup/done support
Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
[use nfs4_sequence_done_free_slot]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index a34fada51446..46e2509452a5 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1241,6 +1241,10 @@ static void nfs4_open_prepare(struct rpc_task *task, void *calldata)
 		nfs_copy_fh(&data->o_res.fh, data->o_arg.fh);
 	}
 	data->timestamp = jiffies;
+	if (nfs4_setup_sequence(data->o_arg.server->nfs_client,
+				&data->o_arg.seq_args,
+				&data->o_res.seq_res, 1, task))
+		return;
 	rpc_call_start(task);
 	return;
 out_no_action:
@@ -1253,6 +1257,10 @@ static void nfs4_open_done(struct rpc_task *task, void *calldata)
 	struct nfs4_opendata *data = calldata;
 
 	data->rpc_status = task->tk_status;
+
+	nfs4_sequence_done_free_slot(data->o_arg.server, &data->o_res.seq_res,
+				     task->tk_status);
+
 	if (RPC_ASSASSINATED(task))
 		return;
 	if (task->tk_status == 0) {
-- 
cgit v1.2.3-59-g8ed1b


From 66179efee31f431ab538016b4ac05edaf87525dc Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:22 -0400
Subject: nfs41: lock sequence setup/done support

Separate nfs4_lock calls from nfs41: sequence setup/done support
Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
[use nfs4_sequence_done_free_slot]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 46e2509452a5..dc070819096a 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3688,6 +3688,7 @@ struct nfs4_lockdata {
 	unsigned long timestamp;
 	int rpc_status;
 	int cancelled;
+	struct nfs_server *server;
 };
 
 static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl,
@@ -3715,6 +3716,7 @@ static struct nfs4_lockdata *nfs4_alloc_lockdata(struct file_lock *fl,
 	p->res.lock_seqid = p->arg.lock_seqid;
 	p->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	p->lsp = lsp;
+	p->server = server;
 	atomic_inc(&lsp->ls_count);
 	p->ctx = get_nfs_open_context(ctx);
 	memcpy(&p->fl, fl, sizeof(p->fl));
@@ -3744,6 +3746,9 @@ static void nfs4_lock_prepare(struct rpc_task *task, void *calldata)
 	} else
 		data->arg.new_lock_owner = 0;
 	data->timestamp = jiffies;
+	if (nfs4_setup_sequence(data->server->nfs_client, &data->arg.seq_args,
+				&data->res.seq_res, 1, task))
+		return;
 	rpc_call_start(task);
 	dprintk("%s: done!, ret = %d\n", __func__, data->rpc_status);
 }
@@ -3754,6 +3759,9 @@ static void nfs4_lock_done(struct rpc_task *task, void *calldata)
 
 	dprintk("%s: begin!\n", __func__);
 
+	nfs4_sequence_done_free_slot(data->server, &data->res.seq_res,
+				     task->tk_status);
+
 	data->rpc_status = task->tk_status;
 	if (RPC_ASSASSINATED(task))
 		goto out;
-- 
cgit v1.2.3-59-g8ed1b


From a893693c151433a1375060ecb13419080426bc08 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:23 -0400
Subject: nfs41: locku sequence setup/done support

Separate nfs4_locku calls from nfs41: sequence setup/done support
Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index dc070819096a..d2cfa4f86537 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3533,7 +3533,7 @@ static struct nfs4_unlockdata *nfs4_alloc_unlockdata(struct file_lock *fl,
 	struct nfs4_unlockdata *p;
 	struct inode *inode = lsp->ls_state->inode;
 
-	p = kmalloc(sizeof(*p), GFP_KERNEL);
+	p = kzalloc(sizeof(*p), GFP_KERNEL);
 	if (p == NULL)
 		return NULL;
 	p->arg.fh = NFS_FH(inode);
@@ -3564,6 +3564,8 @@ static void nfs4_locku_done(struct rpc_task *task, void *data)
 {
 	struct nfs4_unlockdata *calldata = data;
 
+	nfs4_sequence_done(calldata->server, &calldata->res.seq_res,
+			   task->tk_status);
 	if (RPC_ASSASSINATED(task))
 		return;
 	switch (task->tk_status) {
@@ -3582,6 +3584,8 @@ static void nfs4_locku_done(struct rpc_task *task, void *data)
 			if (nfs4_async_handle_error(task, calldata->server, NULL) == -EAGAIN)
 				rpc_restart_call(task);
 	}
+	nfs4_sequence_free_slot(calldata->server->nfs_client,
+				&calldata->res.seq_res);
 }
 
 static void nfs4_locku_prepare(struct rpc_task *task, void *data)
@@ -3596,6 +3600,10 @@ static void nfs4_locku_prepare(struct rpc_task *task, void *data)
 		return;
 	}
 	calldata->timestamp = jiffies;
+	if (nfs4_setup_sequence(calldata->server->nfs_client,
+				&calldata->arg.seq_args,
+				&calldata->res.seq_res, 1, task))
+		return;
 	rpc_call_start(task);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 472cfbd9b97be210a9b61ac1c6a774cd456ea9d2 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:24 -0400
Subject: nfs41: unlink sequence setup/done support

Implement the rpc_call_prepare methods for
asynchronuos nfs rpcs, call nfs41_setup_sequence from
respective rpc_call_validate_args methods.

Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: separate free slot from sequence done]
[nfs41: sequence res use slotid]
[nfs41: remove SEQ4_STATUS_USE_TK_STATUS]
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c |  2 ++
 fs/nfs/unlink.c   | 17 +++++++++++++++++
 2 files changed, 19 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index d2cfa4f86537..97521e47e437 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2424,8 +2424,10 @@ static int nfs4_proc_unlink_done(struct rpc_task *task, struct inode *dir)
 {
 	struct nfs_removeres *res = task->tk_msg.rpc_resp;
 
+	nfs4_sequence_done(res->server, &res->seq_res, task->tk_status);
 	if (nfs4_async_handle_error(task, res->server, NULL) == -EAGAIN)
 		return 0;
+	nfs4_sequence_free_slot(res->server->nfs_client, &res->seq_res);
 	update_changeattr(dir, &res->cinfo);
 	nfs_post_op_update_inode(dir, &res->dir_attr);
 	return 1;
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index 83d0ce2600ab..089a21b4e3f9 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -15,6 +15,7 @@
 #include <linux/wait.h>
 
 #include "internal.h"
+#include "nfs4_fs.h"
 
 struct nfs_unlinkdata {
 	struct hlist_node list;
@@ -102,9 +103,25 @@ static void nfs_async_unlink_release(void *calldata)
 	nfs_sb_deactive(sb);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+void nfs_unlink_prepare(struct rpc_task *task, void *calldata)
+{
+	struct nfs_unlinkdata *data = calldata;
+	struct nfs_server *server = NFS_SERVER(data->dir);
+
+	if (nfs4_setup_sequence(server->nfs_client, &data->args.seq_args,
+				&data->res.seq_res, 1, task))
+		return;
+	rpc_call_start(task);
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 static const struct rpc_call_ops nfs_unlink_ops = {
 	.rpc_call_done = nfs_async_unlink_done,
 	.rpc_release = nfs_async_unlink_release,
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_unlink_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 };
 
 static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct nfs_unlinkdata *data)
-- 
cgit v1.2.3-59-g8ed1b


From f11c88af26453aee2823a1fd9120d0cd8dae7b9a Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:25 -0400
Subject: nfs41: read sequence setup/done support

Implement the read rpc_call_prepare method for
asynchronuos nfs rpcs, call nfs41_setup_sequence from
respective rpc_call_validate_args methods.

Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[move the nfs4_sequence_free_slot call in nfs_readpage_retry from]
[nfs41: separate free slot from sequence done]
[remove nfs_readargs.nfs_server, use calldata->inode instead]
Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Support sessions with O_DIRECT]
Signed-off-by: Dean Hildebrand <dhildeb@us.ibm.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/direct.c   |  3 +++
 fs/nfs/internal.h |  3 +++
 fs/nfs/nfs4proc.c |  5 +++++
 fs/nfs/read.c     | 30 ++++++++++++++++++++++++++++--
 4 files changed, 39 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 08f6b040d289..c8c53a03a585 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -259,6 +259,9 @@ static void nfs_direct_read_release(void *calldata)
 }
 
 static const struct rpc_call_ops nfs_read_direct_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_read_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_direct_read_result,
 	.rpc_release = nfs_direct_read_release,
 };
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 12f4c5e6fd39..988c8b9aa78b 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -206,6 +206,9 @@ extern int nfs4_path_walk(struct nfs_server *server,
 			  const char *path);
 #endif
 
+/* read.c */
+extern void nfs_read_prepare(struct rpc_task *task, void *calldata);
+
 /* nfs4proc.c */
 extern int _nfs4_call_sync(struct nfs_server *server,
 			   struct rpc_message *msg,
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 97521e47e437..6c0ac539d9c7 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2864,6 +2864,11 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data)
 {
 	struct nfs_server *server = NFS_SERVER(data->inode);
 
+	dprintk("--> %s\n", __func__);
+
+	/* nfs4_sequence_free_slot called in the read rpc_call_done */
+	nfs4_sequence_done(server, &data->res.seq_res, task->tk_status);
+
 	if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) {
 		rpc_restart_call(task);
 		return -EAGAIN;
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 70ba2b4cb9a4..d71f0d55ebde 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -22,6 +22,7 @@
 
 #include <asm/system.h>
 
+#include "nfs4_fs.h"
 #include "internal.h"
 #include "iostat.h"
 #include "fscache.h"
@@ -358,19 +359,25 @@ static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data
 	struct nfs_readres *resp = &data->res;
 
 	if (resp->eof || resp->count == argp->count)
-		return;
+		goto out;
 
 	/* This is a short read! */
 	nfs_inc_stats(data->inode, NFSIOS_SHORTREAD);
 	/* Has the server at least made some progress? */
 	if (resp->count == 0)
-		return;
+		goto out;
 
 	/* Yes, so retry the read at the end of the data */
 	argp->offset += resp->count;
 	argp->pgbase += resp->count;
 	argp->count -= resp->count;
 	rpc_restart_call(task);
+	return;
+out:
+	nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client,
+				&data->res.seq_res);
+	return;
+
 }
 
 /*
@@ -407,7 +414,23 @@ static void nfs_readpage_release_partial(void *calldata)
 	nfs_readdata_release(calldata);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+void nfs_read_prepare(struct rpc_task *task, void *calldata)
+{
+	struct nfs_read_data *data = calldata;
+
+	if (nfs4_setup_sequence(NFS_SERVER(data->inode)->nfs_client,
+				&data->args.seq_args, &data->res.seq_res,
+				0, task))
+		return;
+	rpc_call_start(task);
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 static const struct rpc_call_ops nfs_read_partial_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_read_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_readpage_result_partial,
 	.rpc_release = nfs_readpage_release_partial,
 };
@@ -471,6 +494,9 @@ static void nfs_readpage_release_full(void *calldata)
 }
 
 static const struct rpc_call_ops nfs_read_full_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_read_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_readpage_result_full,
 	.rpc_release = nfs_readpage_release_full,
 };
-- 
cgit v1.2.3-59-g8ed1b


From def6ed7ef45ed19c3d6ca765f3bfdff1fe4c6bba Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:26 -0400
Subject: nfs41 write sequence setup done support

Separate write calls from nfs41: sequence setup/done support

Implement the write rpc_call_prepare method for
asynchronuos nfs rpcs, call nfs41_setup_sequence from
respective rpc_call_validate_args methods.

Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[move the nfs4_sequence_free_slot call in nfs_readpage_retry from]
[nfs41: separate free slot from sequence done
Signed-off-by: Andy Adamson <andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Support sessions with O_DIRECT.]
Signed-off-by: Dean Hildebrand <dhildeb@us.ibm.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/direct.c   |  3 +++
 fs/nfs/internal.h |  3 +++
 fs/nfs/nfs4proc.c |  4 ++++
 fs/nfs/write.c    | 22 ++++++++++++++++++++++
 4 files changed, 32 insertions(+)

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index c8c53a03a585..1955bfeb3551 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -676,6 +676,9 @@ out_unlock:
 }
 
 static const struct rpc_call_ops nfs_write_direct_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_write_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_direct_write_result,
 	.rpc_release = nfs_direct_write_release,
 };
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 988c8b9aa78b..f62bc5226155 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -209,6 +209,9 @@ extern int nfs4_path_walk(struct nfs_server *server,
 /* read.c */
 extern void nfs_read_prepare(struct rpc_task *task, void *calldata);
 
+/* write.c */
+extern void nfs_write_prepare(struct rpc_task *task, void *calldata);
+
 /* nfs4proc.c */
 extern int _nfs4_call_sync(struct nfs_server *server,
 			   struct rpc_message *msg,
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 6c0ac539d9c7..5f71832436fd 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2890,6 +2890,10 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data)
 {
 	struct inode *inode = data->inode;
 	
+	/* slot is freed in nfs_writeback_done */
+	nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res,
+			   task->tk_status);
+
 	if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) {
 		rpc_restart_call(task);
 		return -EAGAIN;
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 035e6fb9f57e..195fe7667529 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -25,6 +25,7 @@
 #include "delegation.h"
 #include "internal.h"
 #include "iostat.h"
+#include "nfs4_fs.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PAGECACHE
 
@@ -1050,7 +1051,23 @@ out:
 	nfs_writedata_release(calldata);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+void nfs_write_prepare(struct rpc_task *task, void *calldata)
+{
+	struct nfs_write_data *data = calldata;
+	struct nfs_client *clp = (NFS_SERVER(data->inode))->nfs_client;
+
+	if (nfs4_setup_sequence(clp, &data->args.seq_args,
+				&data->res.seq_res, 1, task))
+		return;
+	rpc_call_start(task);
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 static const struct rpc_call_ops nfs_write_partial_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_write_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_writeback_done_partial,
 	.rpc_release = nfs_writeback_release_partial,
 };
@@ -1113,6 +1130,9 @@ remove_request:
 }
 
 static const struct rpc_call_ops nfs_write_full_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_write_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_writeback_done_full,
 	.rpc_release = nfs_writeback_release_full,
 };
@@ -1195,6 +1215,8 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 		/* Can't do anything about it except throw an error. */
 		task->tk_status = -EIO;
 	}
+	nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client,
+				&data->res.seq_res);
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 21d9a851aaa4161a9fddde720594659f3dae7fdd Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:27 -0400
Subject: nfs41 commit sequence setup done support

Separate commit calls from nfs41: sequence setup/done support

Implement the commit rpc_call_prepare method for
asynchronuos nfs rpcs, call nfs41_setup_sequence from
respective rpc_call_validate_args methods.

Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Support sessions with O_DIRECT.]
Signed-off-by: Dean Hildebrand <dhildeb@us.ibm.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: separate free slot from sequence done]
[nfs41: nfs4_sequence_free_slot use nfs_client for data server]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/direct.c   | 3 +++
 fs/nfs/nfs4proc.c | 4 ++++
 fs/nfs/write.c    | 3 +++
 3 files changed, 10 insertions(+)

diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 1955bfeb3551..489fc01a3204 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -538,6 +538,9 @@ static void nfs_direct_commit_release(void *calldata)
 }
 
 static const struct rpc_call_ops nfs_commit_direct_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_write_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_direct_commit_result,
 	.rpc_release = nfs_direct_commit_release,
 };
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 5f71832436fd..fbd6274f14f8 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2920,10 +2920,14 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data)
 {
 	struct inode *inode = data->inode;
 	
+	nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res,
+			   task->tk_status);
 	if (nfs4_async_handle_error(task, NFS_SERVER(inode), NULL) == -EAGAIN) {
 		rpc_restart_call(task);
 		return -EAGAIN;
 	}
+	nfs4_sequence_free_slot(NFS_SERVER(inode)->nfs_client,
+				&data->res.seq_res);
 	nfs_refresh_inode(inode, data->res.fattr);
 	return 0;
 }
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 195fe7667529..85a76409de13 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1373,6 +1373,9 @@ static void nfs_commit_release(void *calldata)
 }
 
 static const struct rpc_call_ops nfs_commit_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs_write_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs_commit_done,
 	.rpc_release = nfs_commit_release,
 };
-- 
cgit v1.2.3-59-g8ed1b


From 938e10109115a71cc69d475122f21cf75e5046cd Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:28 -0400
Subject: nfs41 delegreturn sequence setup done support

Separate delegreturn calls from nfs41: sequence setup/done support

Implement the delegreturn rpc_call_prepare method for
asynchronuos nfs rpcs, call nfs41_setup_sequence from
respective rpc_call_validate_args methods.

Call nfs4_sequence_done from respective rpc_call_done methods.

Note that we need to pass a pointer to the nfs_server in calls data
for passing on to nfs4_sequence_done.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[pnfs: client data server write validate and release]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 24 +++++++++++++++++++++++-
 1 file changed, 23 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index fbd6274f14f8..dc0feb5837b1 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3360,6 +3360,10 @@ struct nfs4_delegreturndata {
 static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
 {
 	struct nfs4_delegreturndata *data = calldata;
+
+	nfs4_sequence_done_free_slot(data->res.server, &data->res.seq_res,
+				     task->tk_status);
+
 	data->rpc_status = task->tk_status;
 	if (data->rpc_status == 0)
 		renew_lease(data->res.server, data->timestamp);
@@ -3370,7 +3374,25 @@ static void nfs4_delegreturn_release(void *calldata)
 	kfree(calldata);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+static void nfs4_delegreturn_prepare(struct rpc_task *task, void *data)
+{
+	struct nfs4_delegreturndata *d_data;
+
+	d_data = (struct nfs4_delegreturndata *)data;
+
+	if (nfs4_setup_sequence(d_data->res.server->nfs_client,
+				&d_data->args.seq_args,
+				&d_data->res.seq_res, 1, task))
+		return;
+	rpc_call_start(task);
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 static const struct rpc_call_ops nfs4_delegreturn_ops = {
+#if defined(CONFIG_NFS_V4_1)
+	.rpc_call_prepare = nfs4_delegreturn_prepare,
+#endif /* CONFIG_NFS_V4_1 */
 	.rpc_call_done = nfs4_delegreturn_done,
 	.rpc_release = nfs4_delegreturn_release,
 };
@@ -3392,7 +3414,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co
 	};
 	int status = 0;
 
-	data = kmalloc(sizeof(*data), GFP_KERNEL);
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
 	if (data == NULL)
 		return -ENOMEM;
 	data->args.fhandle = &data->fh;
-- 
cgit v1.2.3-59-g8ed1b


From 99fe60d062cfecf382c036065b3278b82b6c5eff Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:29 -0400
Subject: nfs41: exchange_id operation

Implement the exchange_id operation conforming to
http://tools.ietf.org/html/draft-ietf-nfsv4-minorversion1-26

Unlike NFSv4.0, NFSv4.1 requires machine credentials. RPC_AUTH_GSS machine
credentials will be passed into the kernel at mount time to be available for
the exchange_id operation.

RPC_AUTH_UNIX root mounts can use the UNIX root credential. Store the root
credential in the nfs_client struct.

Without a credential, NFSv4.1 state renewal fails.

[nfs41: establish clientid via exchange id only if cred != NULL]
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfsd41: move nfstime4 from under CONFIG_NFS_V4_1]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: do not wait a lease time in exchange id]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
[nfs41: Ignoring impid in decode_exchange_id is missing a READ_BUF]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: fix Xcode_exchange_id's xdr Xcoding pointer type]
[nfs41: get rid of unused struct nfs41_exchange_id_res members]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c          |  61 ++++++++++++++++++++
 fs/nfs/nfs4xdr.c           | 139 +++++++++++++++++++++++++++++++++++++++++++++
 include/linux/nfs4.h       |   1 +
 include/linux/nfs_fs_sb.h  |   6 ++
 include/linux/nfs_xdr.h    |  40 +++++++++++++
 include/linux/nfsd/state.h |   1 -
 6 files changed, 247 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index dc0feb5837b1..6f384e290753 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -48,6 +48,7 @@
 #include <linux/smp_lock.h>
 #include <linux/namei.h>
 #include <linux/mount.h>
+#include <linux/module.h>
 
 #include "nfs4_fs.h"
 #include "delegation.h"
@@ -433,11 +434,13 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 	spin_unlock(&tbl->slot_tbl_lock);
 
 	slot = tbl->slots + slotid;
+	args->sa_session = session;
 	args->sa_slotid = slotid;
 	args->sa_cache_this = cache_reply;
 
 	dprintk("<-- %s slotid=%d seqid=%d\n", __func__, slotid, slot->seq_nr);
 
+	res->sr_session = session;
 	res->sr_slotid = slotid;
 	res->sr_renewal_time = jiffies;
 	/*
@@ -4128,6 +4131,64 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 }
 
 #ifdef CONFIG_NFS_V4_1
+/*
+ * nfs4_proc_exchange_id()
+ *
+ * Since the clientid has expired, all compounds using sessions
+ * associated with the stale clientid will be returning
+ * NFS4ERR_BADSESSION in the sequence operation, and will therefore
+ * be in some phase of session reset.
+ */
+static int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
+{
+	nfs4_verifier verifier;
+	struct nfs41_exchange_id_args args = {
+		.client = clp,
+		.flags = clp->cl_exchange_flags,
+	};
+	struct nfs41_exchange_id_res res = {
+		.client = clp,
+	};
+	int status;
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_EXCHANGE_ID],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+		.rpc_cred = cred,
+	};
+	__be32 *p;
+
+	dprintk("--> %s\n", __func__);
+	BUG_ON(clp == NULL);
+	p = (u32 *)verifier.data;
+	*p++ = htonl((u32)clp->cl_boot_time.tv_sec);
+	*p = htonl((u32)clp->cl_boot_time.tv_nsec);
+	args.verifier = &verifier;
+
+	while (1) {
+		args.id_len = scnprintf(args.id, sizeof(args.id),
+					"%s/%s %u",
+					clp->cl_ipaddr,
+					rpc_peeraddr2str(clp->cl_rpcclient,
+							 RPC_DISPLAY_ADDR),
+					clp->cl_id_uniquifier);
+
+		status = rpc_call_sync(clp->cl_rpcclient, &msg, 0);
+
+		if (status != NFS4ERR_CLID_INUSE)
+			break;
+
+		if (signalled())
+			break;
+
+		if (++clp->cl_id_uniquifier == 0)
+			break;
+	}
+
+	dprintk("<-- %s status= %d\n", __func__, status);
+	return status;
+}
+
 /* Destroy the slot table */
 static void nfs4_destroy_slot_table(struct nfs4_session *session)
 {
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 5b944cd57218..783c4214dccd 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -246,6 +246,27 @@ static int nfs4_stat_to_errno(int);
 				(0)
 
 #if defined(CONFIG_NFS_V4_1)
+#define encode_exchange_id_maxsz (op_encode_hdr_maxsz + \
+				encode_verifier_maxsz + \
+				1 /* co_ownerid.len */ + \
+				XDR_QUADLEN(NFS4_EXCHANGE_ID_LEN) + \
+				1 /* flags */ + \
+				1 /* spa_how */ + \
+				0 /* SP4_NONE (for now) */ + \
+				1 /* zero implemetation id array */)
+#define decode_exchange_id_maxsz (op_decode_hdr_maxsz + \
+				2 /* eir_clientid */ + \
+				1 /* eir_sequenceid */ + \
+				1 /* eir_flags */ + \
+				1 /* spr_how */ + \
+				0 /* SP4_NONE (for now) */ + \
+				2 /* eir_server_owner.so_minor_id */ + \
+				/* eir_server_owner.so_major_id<> */ \
+				XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 + \
+				/* eir_server_scope<> */ \
+				XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 + \
+				1 /* eir_server_impl_id array length */ + \
+				0 /* ignored eir_server_impl_id contents */)
 #define encode_sequence_maxsz	0 /* stub */
 #define decode_sequence_maxsz	0 /* stub */
 #else /* CONFIG_NFS_V4_1 */
@@ -594,6 +615,14 @@ static int nfs4_stat_to_errno(int);
 				 decode_putfh_maxsz + \
 				 decode_lookup_maxsz + \
 				 decode_fs_locations_maxsz)
+#if defined(CONFIG_NFS_V4_1)
+#define NFS4_enc_exchange_id_sz \
+				(compound_encode_hdr_maxsz + \
+				 encode_exchange_id_maxsz)
+#define NFS4_dec_exchange_id_sz \
+				(compound_decode_hdr_maxsz + \
+				 decode_exchange_id_maxsz)
+#endif /* CONFIG_NFS_V4_1 */
 
 static const umode_t nfs_type2fmt[] = {
 	[NF4BAD] = 0,
@@ -1455,7 +1484,29 @@ static void encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *state
 	hdr->replen += decode_delegreturn_maxsz;
 }
 
+#if defined(CONFIG_NFS_V4_1)
 /* NFSv4.1 operations */
+static void encode_exchange_id(struct xdr_stream *xdr,
+			       struct nfs41_exchange_id_args *args,
+			       struct compound_hdr *hdr)
+{
+	__be32 *p;
+
+	RESERVE_SPACE(4 + sizeof(args->verifier->data));
+	WRITE32(OP_EXCHANGE_ID);
+	WRITEMEM(args->verifier->data, sizeof(args->verifier->data));
+
+	encode_string(xdr, args->id_len, args->id);
+
+	RESERVE_SPACE(12);
+	WRITE32(args->flags);
+	WRITE32(0);	/* zero length state_protect4_a */
+	WRITE32(0);	/* zero length implementation id array */
+	hdr->nops++;
+	hdr->replen += decode_exchange_id_maxsz;
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 static void encode_sequence(struct xdr_stream *xdr,
 			    const struct nfs4_sequence_args *args,
 			    struct compound_hdr *hdr)
@@ -2162,6 +2213,26 @@ static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs
 	return 0;
 }
 
+#if defined(CONFIG_NFS_V4_1)
+/*
+ * EXCHANGE_ID request
+ */
+static int nfs4_xdr_enc_exchange_id(struct rpc_rqst *req, uint32_t *p,
+				    struct nfs41_exchange_id_args *args)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr = {
+		.minorversion = args->client->cl_minorversion,
+	};
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, req, &hdr);
+	encode_exchange_id(&xdr, args, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 /*
  * START OF "GENERIC" DECODE ROUTINES.
  *   These may look a little ugly since they are imported from a "generic"
@@ -3877,6 +3948,52 @@ static int decode_delegreturn(struct xdr_stream *xdr)
 	return decode_op_hdr(xdr, OP_DELEGRETURN);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+static int decode_exchange_id(struct xdr_stream *xdr,
+			      struct nfs41_exchange_id_res *res)
+{
+	__be32 *p;
+	uint32_t dummy;
+	int status;
+	struct nfs_client *clp = res->client;
+
+	status = decode_op_hdr(xdr, OP_EXCHANGE_ID);
+	if (status)
+		return status;
+
+	READ_BUF(8);
+	READ64(clp->cl_ex_clid);
+	READ_BUF(12);
+	READ32(clp->cl_seqid);
+	READ32(clp->cl_exchange_flags);
+
+	/* We ask for SP4_NONE */
+	READ32(dummy);
+	if (dummy != SP4_NONE)
+		return -EIO;
+
+	/* Throw away minor_id */
+	READ_BUF(8);
+
+	/* Throw away Major id */
+	READ_BUF(4);
+	READ32(dummy);
+	READ_BUF(dummy);
+
+	/* Throw away server_scope */
+	READ_BUF(4);
+	READ32(dummy);
+	READ_BUF(dummy);
+
+	/* Throw away Implementation id array */
+	READ_BUF(4);
+	READ32(dummy);
+	READ_BUF(dummy);
+
+	return 0;
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 static int decode_sequence(struct xdr_stream *xdr,
 			   struct nfs4_sequence_res *res,
 			   struct rpc_rqst *rqstp)
@@ -4774,6 +4891,25 @@ out:
 	return status;
 }
 
+#if defined(CONFIG_NFS_V4_1)
+/*
+ * EXCHANGE_ID request
+ */
+static int nfs4_xdr_dec_exchange_id(struct rpc_rqst *rqstp, uint32_t *p,
+				    void *res)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_exchange_id(&xdr, res);
+	return status;
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus)
 {
 	uint32_t bitmap[2] = {0};
@@ -4943,6 +5079,9 @@ struct rpc_procinfo	nfs4_procedures[] = {
   PROC(GETACL,		enc_getacl,	dec_getacl),
   PROC(SETACL,		enc_setacl,	dec_setacl),
   PROC(FS_LOCATIONS,	enc_fs_locations, dec_fs_locations),
+#if defined(CONFIG_NFS_V4_1)
+  PROC(EXCHANGE_ID,	enc_exchange_id,	dec_exchange_id),
+#endif /* CONFIG_NFS_V4_1 */
 };
 
 struct rpc_version		nfs_version4 = {
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index 7c36fcf2dfb7..ad65709ed8d3 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -21,6 +21,7 @@
 #define NFS4_FHSIZE		128
 #define NFS4_MAXPATHLEN		PATH_MAX
 #define NFS4_MAXNAMLEN		NAME_MAX
+#define NFS4_OPAQUE_LIMIT	1024
 #define NFS4_MAX_SESSIONID_LEN	16
 
 #define NFS4_ACCESS_READ        0x0001
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 206485e5082f..435ed556efb5 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -78,6 +78,12 @@ struct nfs_client {
 #endif /* CONFIG_NFS_V4 */
 
 #ifdef CONFIG_NFS_V4_1
+	/* clientid returned from EXCHANGE_ID, used by session operations */
+	u64			cl_ex_clid;
+	/* The sequence id to use for the next CREATE_SESSION */
+	u32			cl_seqid;
+	/* The flags used for obtaining the clientid during EXCHANGE_ID */
+	u32			cl_exchange_flags;
 	struct nfs4_session	*cl_session; 	/* sharred session */
 #endif /* CONFIG_NFS_V4_1 */
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 4ac14b40efc9..5d70b924af5e 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -879,6 +879,46 @@ struct nfs4_fs_locations_res {
 
 #endif /* CONFIG_NFS_V4 */
 
+struct nfstime4 {
+	u64	seconds;
+	u32	nseconds;
+};
+
+#ifdef CONFIG_NFS_V4_1
+struct nfs_impl_id4 {
+	u32		domain_len;
+	char		*domain;
+	u32		name_len;
+	char		*name;
+	struct nfstime4	date;
+};
+
+#define NFS4_EXCHANGE_ID_LEN	(48)
+struct nfs41_exchange_id_args {
+	struct nfs_client		*client;
+	nfs4_verifier			*verifier;
+	unsigned int 			id_len;
+	char 				id[NFS4_EXCHANGE_ID_LEN];
+	u32				flags;
+};
+
+struct server_owner {
+	uint64_t			minor_id;
+	uint32_t			major_id_sz;
+	char				major_id[NFS4_OPAQUE_LIMIT];
+};
+
+struct server_scope {
+	uint32_t			server_scope_sz;
+	char 				server_scope[NFS4_OPAQUE_LIMIT];
+};
+
+struct nfs41_exchange_id_res {
+	struct nfs_client		*client;
+	u32				flags;
+};
+#endif /* CONFIG_NFS_V4_1 */
+
 struct nfs_page;
 
 #define NFS_PAGEVEC_SIZE	(8U)
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 4d61c873feed..7ef4b7ad1214 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -41,7 +41,6 @@
 #include <linux/kref.h>
 #include <linux/sunrpc/clnt.h>
 
-#define NFS4_OPAQUE_LIMIT 1024
 typedef struct {
 	u32             cl_boot;
 	u32             cl_id;
-- 
cgit v1.2.3-59-g8ed1b


From 2050f0cc0703aab7cee798b3cb47037754f368bc Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:30 -0400
Subject: nfs41: get_lease_time

get_lease_time uses the FSINFO rpc operation to
get the lease time attribute.

nfs4_get_lease_time() is only called from the state manager on session setup
so don't recover from clientid or sequence level errors.

We do need to recover from NFS4ERR_DELAY or NFS4ERR_GRACE.
Use NFS4_POLL_RETRY_MIN - the Linux server returns NFS4ERR_DELAY when an
upcall is needed to resolve an uncached export referenced by a file handle.

[nfs41: sequence res use slotid]
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove extraneous rpc_clnt pointer]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: have get_lease_time work on nfs_client]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: get_lease_time recover from NFS4ERR_DELAY]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
[define nfs4_get_lease_time_{args,res}]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c       | 94 +++++++++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/nfs4xdr.c        | 51 +++++++++++++++++++++++++++
 include/linux/nfs_xdr.h |  9 +++++
 3 files changed, 154 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 6f384e290753..eafc99afd356 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4189,6 +4189,100 @@ static int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 	return status;
 }
 
+struct nfs4_get_lease_time_data {
+	struct nfs4_get_lease_time_args *args;
+	struct nfs4_get_lease_time_res *res;
+	struct nfs_client *clp;
+};
+
+static void nfs4_get_lease_time_prepare(struct rpc_task *task,
+					void *calldata)
+{
+	int ret;
+	struct nfs4_get_lease_time_data *data =
+			(struct nfs4_get_lease_time_data *)calldata;
+
+	dprintk("--> %s\n", __func__);
+	/* just setup sequence, do not trigger session recovery
+	   since we're invoked within one */
+	ret = nfs41_setup_sequence(data->clp->cl_session,
+					&data->args->la_seq_args,
+					&data->res->lr_seq_res, 0, task);
+
+	BUG_ON(ret == -EAGAIN);
+	rpc_call_start(task);
+	dprintk("<-- %s\n", __func__);
+}
+
+/*
+ * Called from nfs4_state_manager thread for session setup, so don't recover
+ * from sequence operation or clientid errors.
+ */
+static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata)
+{
+	struct nfs4_get_lease_time_data *data =
+			(struct nfs4_get_lease_time_data *)calldata;
+
+	dprintk("--> %s\n", __func__);
+	nfs41_sequence_done(data->clp, &data->res->lr_seq_res, task->tk_status);
+	switch (task->tk_status) {
+	case -NFS4ERR_DELAY:
+	case -NFS4ERR_GRACE:
+		dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status);
+		rpc_delay(task, NFS4_POLL_RETRY_MIN);
+		task->tk_status = 0;
+		rpc_restart_call(task);
+		return;
+	}
+	nfs41_sequence_free_slot(data->clp, &data->res->lr_seq_res);
+	dprintk("<-- %s\n", __func__);
+}
+
+struct rpc_call_ops nfs4_get_lease_time_ops = {
+	.rpc_call_prepare = nfs4_get_lease_time_prepare,
+	.rpc_call_done = nfs4_get_lease_time_done,
+};
+
+int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo)
+{
+	struct rpc_task *task;
+	struct nfs4_get_lease_time_args args;
+	struct nfs4_get_lease_time_res res = {
+		.lr_fsinfo = fsinfo,
+	};
+	struct nfs4_get_lease_time_data data = {
+		.args = &args,
+		.res = &res,
+		.clp = clp,
+	};
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_GET_LEASE_TIME],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+	};
+	struct rpc_task_setup task_setup = {
+		.rpc_client = clp->cl_rpcclient,
+		.rpc_message = &msg,
+		.callback_ops = &nfs4_get_lease_time_ops,
+		.callback_data = &data
+	};
+	int status;
+
+	res.lr_seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
+	dprintk("--> %s\n", __func__);
+	task = rpc_run_task(&task_setup);
+
+	if (IS_ERR(task))
+		status = PTR_ERR(task);
+	else {
+		status = task->tk_status;
+		rpc_put_task(task);
+	}
+	dprintk("<-- %s return %d\n", __func__, status);
+
+	return status;
+}
+
 /* Destroy the slot table */
 static void nfs4_destroy_slot_table(struct nfs4_session *session)
 {
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 783c4214dccd..85ee1d17a461 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -622,6 +622,14 @@ static int nfs4_stat_to_errno(int);
 #define NFS4_dec_exchange_id_sz \
 				(compound_decode_hdr_maxsz + \
 				 decode_exchange_id_maxsz)
+#define NFS4_enc_get_lease_time_sz	(compound_encode_hdr_maxsz + \
+					 encode_sequence_maxsz + \
+					 encode_putrootfh_maxsz + \
+					 encode_fsinfo_maxsz)
+#define NFS4_dec_get_lease_time_sz	(compound_decode_hdr_maxsz + \
+					 decode_sequence_maxsz + \
+					 decode_putrootfh_maxsz + \
+					 decode_fsinfo_maxsz)
 #endif /* CONFIG_NFS_V4_1 */
 
 static const umode_t nfs_type2fmt[] = {
@@ -2231,6 +2239,27 @@ static int nfs4_xdr_enc_exchange_id(struct rpc_rqst *req, uint32_t *p,
 	encode_nops(&hdr);
 	return 0;
 }
+
+/*
+ * a GET_LEASE_TIME request
+ */
+static int nfs4_xdr_enc_get_lease_time(struct rpc_rqst *req, uint32_t *p,
+				       struct nfs4_get_lease_time_args *args)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(&args->la_seq_args),
+	};
+	const u32 lease_bitmap[2] = { FATTR4_WORD0_LEASE_TIME, 0 };
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->la_seq_args, &hdr);
+	encode_putrootfh(&xdr, &hdr);
+	encode_fsinfo(&xdr, lease_bitmap, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 /*
@@ -4908,6 +4937,27 @@ static int nfs4_xdr_dec_exchange_id(struct rpc_rqst *rqstp, uint32_t *p,
 		status = decode_exchange_id(&xdr, res);
 	return status;
 }
+
+/*
+ * a GET_LEASE_TIME request
+ */
+static int nfs4_xdr_dec_get_lease_time(struct rpc_rqst *rqstp, uint32_t *p,
+				       struct nfs4_get_lease_time_res *res)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_sequence(&xdr, &res->lr_seq_res, rqstp);
+	if (!status)
+		status = decode_putrootfh(&xdr);
+	if (!status)
+		status = decode_fsinfo(&xdr, res->lr_fsinfo);
+	return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus)
@@ -5081,6 +5131,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
   PROC(FS_LOCATIONS,	enc_fs_locations, dec_fs_locations),
 #if defined(CONFIG_NFS_V4_1)
   PROC(EXCHANGE_ID,	enc_exchange_id,	dec_exchange_id),
+  PROC(GET_LEASE_TIME,	enc_get_lease_time,	dec_get_lease_time),
 #endif /* CONFIG_NFS_V4_1 */
 };
 
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 5d70b924af5e..ca643aa87d46 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -174,6 +174,15 @@ struct nfs4_sequence_res {
 	int			sr_status;	/* sequence operation status */
 };
 
+struct nfs4_get_lease_time_args {
+	struct nfs4_sequence_args	la_seq_args;
+};
+
+struct nfs4_get_lease_time_res {
+	struct nfs_fsinfo	       *lr_fsinfo;
+	struct nfs4_sequence_res	lr_seq_res;
+};
+
 /*
  * Arguments to the open call.
  */
-- 
cgit v1.2.3-59-g8ed1b


From fc931582c260e53ca5ca23bd70ccc9b2265cca9f Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:31 -0400
Subject: nfs41: create_session operation

Implement the create_session operation conforming to
http://tools.ietf.org/html/draft-ietf-nfsv4-minorversion1-26

Set the real fore channel max operations to preserve server resources.
Note: If the server returns < NFS4_MAX_OPS, the client will very soon
get an NFS4ERR_TOO_MANY_OPS. A later patch will handle this.

Set the max_rqst_sz and max_resp_sz to PAGE_SIZE - we preallocate the buffers.

Set the back channel max_resp_sz_cached to zero to force the client to
always set csa_cachethis to FALSE because the current implementation
of the back channel DRC only supports caching the CB_SEQUENCE operation.

The client back channel server supports one slot, and desires 2 operations
per compound.

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove extraneous rpc_clnt pointer]
Use the struct nfs_client cl_rpcclient.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_init_channel_attrs, just use nfs41_create_session_args]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use rsize and wsize for session channel attributes]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: set channel max operations]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: set back channel attributes]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: obliterate nfs4_adjust_channel_attrs]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: have create_session work on nfs_client]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: move CONFIG_NFS_V4_1 endif]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
[moved nfs4_init_slot_table definition here]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use kcalloc to allocate slot table]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
[nfs41: fix Xcode_create_session's xdr Xcoding pointer type]
[nfs41: refactor decoding of channel attributes]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c       | 172 ++++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/nfs4xdr.c        | 185 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/nfs4.h    |  10 +++
 include/linux/nfs_xdr.h |  12 ++++
 4 files changed, 379 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index eafc99afd356..7d81d6e57533 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -54,6 +54,7 @@
 #include "delegation.h"
 #include "internal.h"
 #include "iostat.h"
+#include "callback.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PROC
 
@@ -4283,6 +4284,50 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo)
 	return status;
 }
 
+/*
+ * Initialize slot table
+ */
+static int nfs4_init_slot_table(struct nfs4_session *session)
+{
+	struct nfs4_slot_table *tbl = &session->fc_slot_table;
+	int i, max_slots = session->fc_attrs.max_reqs;
+	struct nfs4_slot *slot;
+	int ret = -ENOMEM;
+
+	BUG_ON(max_slots > NFS4_MAX_SLOT_TABLE);
+
+	dprintk("--> %s: max_reqs=%u\n", __func__,
+		session->fc_attrs.max_reqs);
+
+	slot = kcalloc(max_slots, sizeof(struct nfs4_slot), GFP_KERNEL);
+	if (!slot)
+		goto out;
+	for (i = 0; i < max_slots; ++i)
+		slot[i].seq_nr = 1;
+	ret = 0;
+
+	spin_lock(&tbl->slot_tbl_lock);
+	if (tbl->slots != NULL) {
+		spin_unlock(&tbl->slot_tbl_lock);
+		dprintk("%s: slot table already initialized. tbl=%p slots=%p\n",
+			__func__, tbl, tbl->slots);
+		WARN_ON(1);
+		goto out_free;
+	}
+	tbl->max_slots = max_slots;
+	tbl->slots = slot;
+	tbl->highest_used_slotid = -1;  /* no slot is currently used */
+	spin_unlock(&tbl->slot_tbl_lock);
+	dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__,
+		tbl, tbl->slots, tbl->max_slots);
+out:
+	dprintk("<-- %s: return %d\n", __func__, ret);
+	return ret;
+out_free:
+	kfree(slot);
+	goto out;
+}
+
 /* Destroy the slot table */
 static void nfs4_destroy_slot_table(struct nfs4_session *session)
 {
@@ -4314,6 +4359,133 @@ void nfs4_destroy_session(struct nfs4_session *session)
 	kfree(session);
 }
 
+/*
+ * Initialize the values to be used by the client in CREATE_SESSION
+ * If nfs4_init_session set the fore channel request and response sizes,
+ * use them.
+ *
+ * Set the back channel max_resp_sz_cached to zero to force the client to
+ * always set csa_cachethis to FALSE because the current implementation
+ * of the back channel DRC only supports caching the CB_SEQUENCE operation.
+ */
+static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args)
+{
+	struct nfs4_session *session = args->client->cl_session;
+	unsigned int mxrqst_sz = session->fc_attrs.max_rqst_sz,
+		     mxresp_sz = session->fc_attrs.max_resp_sz;
+
+	if (mxrqst_sz == 0)
+		mxrqst_sz = NFS_MAX_FILE_IO_SIZE;
+	if (mxresp_sz == 0)
+		mxresp_sz = NFS_MAX_FILE_IO_SIZE;
+	/* Fore channel attributes */
+	args->fc_attrs.headerpadsz = 0;
+	args->fc_attrs.max_rqst_sz = mxrqst_sz;
+	args->fc_attrs.max_resp_sz = mxresp_sz;
+	args->fc_attrs.max_resp_sz_cached = mxresp_sz;
+	args->fc_attrs.max_ops = NFS4_MAX_OPS;
+	args->fc_attrs.max_reqs = session->clp->cl_rpcclient->cl_xprt->max_reqs;
+
+	dprintk("%s: Fore Channel : max_rqst_sz=%u max_resp_sz=%u "
+		"max_resp_sz_cached=%u max_ops=%u max_reqs=%u\n",
+		__func__,
+		args->fc_attrs.max_rqst_sz, args->fc_attrs.max_resp_sz,
+		args->fc_attrs.max_resp_sz_cached, args->fc_attrs.max_ops,
+		args->fc_attrs.max_reqs);
+
+	/* Back channel attributes */
+	args->bc_attrs.headerpadsz = 0;
+	args->bc_attrs.max_rqst_sz = PAGE_SIZE;
+	args->bc_attrs.max_resp_sz = PAGE_SIZE;
+	args->bc_attrs.max_resp_sz_cached = 0;
+	args->bc_attrs.max_ops = NFS4_MAX_BACK_CHANNEL_OPS;
+	args->bc_attrs.max_reqs = 1;
+
+	dprintk("%s: Back Channel : max_rqst_sz=%u max_resp_sz=%u "
+		"max_resp_sz_cached=%u max_ops=%u max_reqs=%u\n",
+		__func__,
+		args->bc_attrs.max_rqst_sz, args->bc_attrs.max_resp_sz,
+		args->bc_attrs.max_resp_sz_cached, args->bc_attrs.max_ops,
+		args->bc_attrs.max_reqs);
+}
+
+static int _nfs4_proc_create_session(struct nfs_client *clp)
+{
+	struct nfs4_session *session = clp->cl_session;
+	struct nfs41_create_session_args args = {
+		.client = clp,
+		.cb_program = NFS4_CALLBACK,
+	};
+	struct nfs41_create_session_res res = {
+		.client = clp,
+	};
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CREATE_SESSION],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+	};
+	int status;
+
+	nfs4_init_channel_attrs(&args);
+	args.flags = (SESSION4_PERSIST);
+
+	status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0);
+
+	/* Set the negotiated values in the session's channel_attrs struct */
+
+	if (!status) {
+		/* Increment the clientid slot sequence id */
+		clp->cl_seqid++;
+	}
+
+	return status;
+}
+
+/*
+ * Issues a CREATE_SESSION operation to the server.
+ * It is the responsibility of the caller to verify the session is
+ * expired before calling this routine.
+ */
+int nfs4_proc_create_session(struct nfs_client *clp)
+{
+	int status;
+	unsigned *ptr;
+	struct nfs_fsinfo fsinfo;
+	struct nfs4_session *session = clp->cl_session;
+
+	dprintk("--> %s clp=%p session=%p\n", __func__, clp, session);
+
+	status = _nfs4_proc_create_session(clp);
+	if (status)
+		goto out;
+
+	/* Init the fore channel */
+	status = nfs4_init_slot_table(session);
+	dprintk("fore channel slot table initialization returned %d\n", status);
+	if (status)
+		goto out;
+
+	ptr = (unsigned *)&session->sess_id.data[0];
+	dprintk("%s client>seqid %d sessionid %u:%u:%u:%u\n", __func__,
+		clp->cl_seqid, ptr[0], ptr[1], ptr[2], ptr[3]);
+
+	/* Get the lease time */
+	status = nfs4_proc_get_lease_time(clp, &fsinfo);
+	if (status == 0) {
+		/* Update lease time and schedule renewal */
+		spin_lock(&clp->cl_lock);
+		clp->cl_lease_time = fsinfo.lease_time * HZ;
+		clp->cl_last_renewal = jiffies;
+		clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+		spin_unlock(&clp->cl_lock);
+
+		nfs4_schedule_state_renewal(clp);
+	}
+out:
+	dprintk("<-- %s\n", __func__);
+	return status;
+}
+
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = {
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 85ee1d17a461..7a243a2cf0be 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -246,6 +246,8 @@ static int nfs4_stat_to_errno(int);
 				(0)
 
 #if defined(CONFIG_NFS_V4_1)
+#define NFS4_MAX_MACHINE_NAME_LEN (64)
+
 #define encode_exchange_id_maxsz (op_encode_hdr_maxsz + \
 				encode_verifier_maxsz + \
 				1 /* co_ownerid.len */ + \
@@ -267,6 +269,31 @@ static int nfs4_stat_to_errno(int);
 				XDR_QUADLEN(NFS4_OPAQUE_LIMIT) + 1 + \
 				1 /* eir_server_impl_id array length */ + \
 				0 /* ignored eir_server_impl_id contents */)
+#define encode_channel_attrs_maxsz  (6 + 1 /* ca_rdma_ird.len (0) */)
+#define decode_channel_attrs_maxsz  (6 + \
+				     1 /* ca_rdma_ird.len */ + \
+				     1 /* ca_rdma_ird */)
+#define encode_create_session_maxsz  (op_encode_hdr_maxsz + \
+				     2 /* csa_clientid */ + \
+				     1 /* csa_sequence */ + \
+				     1 /* csa_flags */ + \
+				     encode_channel_attrs_maxsz + \
+				     encode_channel_attrs_maxsz + \
+				     1 /* csa_cb_program */ + \
+				     1 /* csa_sec_parms.len (1) */ + \
+				     1 /* cb_secflavor (AUTH_SYS) */ + \
+				     1 /* stamp */ + \
+				     1 /* machinename.len */ + \
+				     XDR_QUADLEN(NFS4_MAX_MACHINE_NAME_LEN) + \
+				     1 /* uid */ + \
+				     1 /* gid */ + \
+				     1 /* gids.len (0) */)
+#define decode_create_session_maxsz  (op_decode_hdr_maxsz +	\
+				     XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + \
+				     1 /* csr_sequence */ + \
+				     1 /* csr_flags */ + \
+				     decode_channel_attrs_maxsz + \
+				     decode_channel_attrs_maxsz)
 #define encode_sequence_maxsz	0 /* stub */
 #define decode_sequence_maxsz	0 /* stub */
 #else /* CONFIG_NFS_V4_1 */
@@ -622,6 +649,12 @@ static int nfs4_stat_to_errno(int);
 #define NFS4_dec_exchange_id_sz \
 				(compound_decode_hdr_maxsz + \
 				 decode_exchange_id_maxsz)
+#define NFS4_enc_create_session_sz \
+				(compound_encode_hdr_maxsz + \
+				 encode_create_session_maxsz)
+#define NFS4_dec_create_session_sz \
+				(compound_decode_hdr_maxsz + \
+				 decode_create_session_maxsz)
 #define NFS4_enc_get_lease_time_sz	(compound_encode_hdr_maxsz + \
 					 encode_sequence_maxsz + \
 					 encode_putrootfh_maxsz + \
@@ -712,6 +745,7 @@ static void encode_compound_hdr(struct xdr_stream *xdr,
 
 static void encode_nops(struct compound_hdr *hdr)
 {
+	BUG_ON(hdr->nops > NFS4_MAX_OPS);
 	*hdr->nops_p = htonl(hdr->nops);
 }
 
@@ -1513,6 +1547,68 @@ static void encode_exchange_id(struct xdr_stream *xdr,
 	hdr->nops++;
 	hdr->replen += decode_exchange_id_maxsz;
 }
+
+static void encode_create_session(struct xdr_stream *xdr,
+				  struct nfs41_create_session_args *args,
+				  struct compound_hdr *hdr)
+{
+	__be32 *p;
+	char machine_name[NFS4_MAX_MACHINE_NAME_LEN];
+	uint32_t len;
+	struct nfs_client *clp = args->client;
+
+	RESERVE_SPACE(4);
+	WRITE32(OP_CREATE_SESSION);
+
+	RESERVE_SPACE(8);
+	WRITE64(clp->cl_ex_clid);
+
+	RESERVE_SPACE(8);
+	WRITE32(clp->cl_seqid);			/*Sequence id */
+	WRITE32(args->flags);			/*flags */
+
+	RESERVE_SPACE(2*28);			/* 2 channel_attrs */
+	/* Fore Channel */
+	WRITE32(args->fc_attrs.headerpadsz);	/* header padding size */
+	WRITE32(args->fc_attrs.max_rqst_sz);	/* max req size */
+	WRITE32(args->fc_attrs.max_resp_sz);	/* max resp size */
+	WRITE32(args->fc_attrs.max_resp_sz_cached);	/* Max resp sz cached */
+	WRITE32(args->fc_attrs.max_ops);	/* max operations */
+	WRITE32(args->fc_attrs.max_reqs);	/* max requests */
+	WRITE32(0);				/* rdmachannel_attrs */
+
+	/* Back Channel */
+	WRITE32(args->fc_attrs.headerpadsz);	/* header padding size */
+	WRITE32(args->bc_attrs.max_rqst_sz);	/* max req size */
+	WRITE32(args->bc_attrs.max_resp_sz);	/* max resp size */
+	WRITE32(args->bc_attrs.max_resp_sz_cached);	/* Max resp sz cached */
+	WRITE32(args->bc_attrs.max_ops);	/* max operations */
+	WRITE32(args->bc_attrs.max_reqs);	/* max requests */
+	WRITE32(0);				/* rdmachannel_attrs */
+
+	RESERVE_SPACE(4);
+	WRITE32(args->cb_program);		/* cb_program */
+
+	RESERVE_SPACE(4);			/* # of security flavors */
+	WRITE32(1);
+
+	RESERVE_SPACE(4);
+	WRITE32(RPC_AUTH_UNIX);			/* auth_sys */
+
+	/* authsys_parms rfc1831 */
+	RESERVE_SPACE(4);
+	WRITE32((u32)clp->cl_boot_time.tv_nsec);	/* stamp */
+	len = scnprintf(machine_name, sizeof(machine_name), "%s",
+			clp->cl_ipaddr);
+	RESERVE_SPACE(16 + len);
+	WRITE32(len);
+	WRITEMEM(machine_name, len);
+	WRITE32(0);				/* UID */
+	WRITE32(0);				/* GID */
+	WRITE32(0);				/* No more gids */
+	hdr->nops++;
+	hdr->replen += decode_create_session_maxsz;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static void encode_sequence(struct xdr_stream *xdr,
@@ -2240,6 +2336,24 @@ static int nfs4_xdr_enc_exchange_id(struct rpc_rqst *req, uint32_t *p,
 	return 0;
 }
 
+/*
+ * a CREATE_SESSION request
+ */
+static int nfs4_xdr_enc_create_session(struct rpc_rqst *req, uint32_t *p,
+				       struct nfs41_create_session_args *args)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr = {
+		.minorversion = args->client->cl_minorversion,
+	};
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, req, &hdr);
+	encode_create_session(&xdr, args, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
+
 /*
  * a GET_LEASE_TIME request
  */
@@ -4021,6 +4135,59 @@ static int decode_exchange_id(struct xdr_stream *xdr,
 
 	return 0;
 }
+
+static int decode_chan_attrs(struct xdr_stream *xdr,
+			     struct nfs4_channel_attrs *attrs)
+{
+	__be32 *p;
+	u32 nr_attrs;
+
+	READ_BUF(28);
+	READ32(attrs->headerpadsz);
+	READ32(attrs->max_rqst_sz);
+	READ32(attrs->max_resp_sz);
+	READ32(attrs->max_resp_sz_cached);
+	READ32(attrs->max_ops);
+	READ32(attrs->max_reqs);
+	READ32(nr_attrs);
+	if (unlikely(nr_attrs > 1)) {
+		printk(KERN_WARNING "%s: Invalid rdma channel attrs count %u\n",
+			__func__, nr_attrs);
+		return -EINVAL;
+	}
+	if (nr_attrs == 1)
+		READ_BUF(4); /* skip rdma_attrs */
+	return 0;
+}
+
+static int decode_create_session(struct xdr_stream *xdr,
+				 struct nfs41_create_session_res *res)
+{
+	__be32 *p;
+	int status;
+	struct nfs_client *clp = res->client;
+	struct nfs4_session *session = clp->cl_session;
+
+	status = decode_op_hdr(xdr, OP_CREATE_SESSION);
+
+	if (status)
+		return status;
+
+	/* sessionid */
+	READ_BUF(NFS4_MAX_SESSIONID_LEN);
+	COPYMEM(&session->sess_id, NFS4_MAX_SESSIONID_LEN);
+
+	/* seqid, flags */
+	READ_BUF(8);
+	READ32(clp->cl_seqid);
+	READ32(session->flags);
+
+	/* Channel attributes */
+	status = decode_chan_attrs(xdr, &session->fc_attrs);
+	if (!status)
+		status = decode_chan_attrs(xdr, &session->bc_attrs);
+	return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static int decode_sequence(struct xdr_stream *xdr,
@@ -4938,6 +5105,23 @@ static int nfs4_xdr_dec_exchange_id(struct rpc_rqst *rqstp, uint32_t *p,
 	return status;
 }
 
+/*
+ * a CREATE_SESSION request
+ */
+static int nfs4_xdr_dec_create_session(struct rpc_rqst *rqstp, uint32_t *p,
+				       struct nfs41_create_session_res *res)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_create_session(&xdr, res);
+	return status;
+}
+
 /*
  * a GET_LEASE_TIME request
  */
@@ -5131,6 +5315,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
   PROC(FS_LOCATIONS,	enc_fs_locations, dec_fs_locations),
 #if defined(CONFIG_NFS_V4_1)
   PROC(EXCHANGE_ID,	enc_exchange_id,	dec_exchange_id),
+  PROC(CREATE_SESSION,	enc_create_session,	dec_create_session),
   PROC(GET_LEASE_TIME,	enc_get_lease_time,	dec_get_lease_time),
 #endif /* CONFIG_NFS_V4_1 */
 };
diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h
index ad65709ed8d3..bd2eba530667 100644
--- a/include/linux/nfs4.h
+++ b/include/linux/nfs4.h
@@ -131,6 +131,16 @@
 
 #define NFS4_MAX_UINT64	(~(u64)0)
 
+/* An NFS4 sessions server must support at least NFS4_MAX_OPS operations.
+ * If a compound requires more operations, adjust NFS4_MAX_OPS accordingly.
+ */
+#define NFS4_MAX_OPS   8
+
+/* Our NFS4 client back channel server only wants the cb_sequene and the
+ * actual operation per compound
+ */
+#define NFS4_MAX_BACK_CHANNEL_OPS 2
+
 enum nfs4_acl_whotype {
 	NFS4_ACL_WHO_NAMED = 0,
 	NFS4_ACL_WHO_OWNER,
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index ca643aa87d46..62f63fb0c4c8 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -926,6 +926,18 @@ struct nfs41_exchange_id_res {
 	struct nfs_client		*client;
 	u32				flags;
 };
+
+struct nfs41_create_session_args {
+	struct nfs_client	       *client;
+	uint32_t			flags;
+	uint32_t			cb_program;
+	struct nfs4_channel_attrs	fc_attrs;	/* Fore Channel */
+	struct nfs4_channel_attrs	bc_attrs;	/* Back Channel */
+};
+
+struct nfs41_create_session_res {
+	struct nfs_client	       *client;
+};
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs_page;
-- 
cgit v1.2.3-59-g8ed1b


From 8d35301d7db4165f2430c428865f7b0add47615d Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:32 -0400
Subject: nfs41: verify session channel attribues

Invalidate the session if the server returns invalid fore or back channel
attributes.

Use a KERN_WARNING to report the fatal session estabishment error.

Signed-off-by: Andy Adamson <andros@netapp.com>
[refactor nfs4_verify_channel_attrs]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 48 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 7d81d6e57533..4d7a8b9e1291 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4409,6 +4409,51 @@ static void nfs4_init_channel_attrs(struct nfs41_create_session_args *args)
 		args->bc_attrs.max_reqs);
 }
 
+static int _verify_channel_attr(char *chan, char *attr_name, u32 sent, u32 rcvd)
+{
+	if (rcvd <= sent)
+		return 0;
+	printk(KERN_WARNING "%s: Session INVALID: %s channel %s increased. "
+		"sent=%u rcvd=%u\n", __func__, chan, attr_name, sent, rcvd);
+	return -EINVAL;
+}
+
+#define _verify_fore_channel_attr(_name_) \
+	_verify_channel_attr("fore", #_name_, \
+			     args->fc_attrs._name_, \
+			     session->fc_attrs._name_)
+
+#define _verify_back_channel_attr(_name_) \
+	_verify_channel_attr("back", #_name_, \
+			     args->bc_attrs._name_, \
+			     session->bc_attrs._name_)
+
+/*
+ * The server is not allowed to increase the fore channel header pad size,
+ * maximum response size, or maximum number of operations.
+ *
+ * The back channel attributes are only negotiatied down: We send what the
+ * (back channel) server insists upon.
+ */
+static int nfs4_verify_channel_attrs(struct nfs41_create_session_args *args,
+				     struct nfs4_session *session)
+{
+	int ret = 0;
+
+	ret |= _verify_fore_channel_attr(headerpadsz);
+	ret |= _verify_fore_channel_attr(max_resp_sz);
+	ret |= _verify_fore_channel_attr(max_ops);
+
+	ret |= _verify_back_channel_attr(headerpadsz);
+	ret |= _verify_back_channel_attr(max_rqst_sz);
+	ret |= _verify_back_channel_attr(max_resp_sz);
+	ret |= _verify_back_channel_attr(max_resp_sz_cached);
+	ret |= _verify_back_channel_attr(max_ops);
+	ret |= _verify_back_channel_attr(max_reqs);
+
+	return ret;
+}
+
 static int _nfs4_proc_create_session(struct nfs_client *clp)
 {
 	struct nfs4_session *session = clp->cl_session;
@@ -4431,8 +4476,9 @@ static int _nfs4_proc_create_session(struct nfs_client *clp)
 
 	status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0);
 
-	/* Set the negotiated values in the session's channel_attrs struct */
-
+	if (!status)
+		/* Verify the session's negotiated channel_attrs values */
+		status = nfs4_verify_channel_attrs(&args, session);
 	if (!status) {
 		/* Increment the clientid slot sequence id */
 		clp->cl_seqid++;
-- 
cgit v1.2.3-59-g8ed1b


From 96b09e024fedf0a6604c8c688a3994d5ed991434 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:33 -0400
Subject: nfs41: use session attributes for rsize and wsize

Set the mount points rsize and wsize to the negotiated session fore channel
maximum response and requeset size. These values will be bound checked in
nfs_server_set_fsinfo.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[move nfs4_session_set_rwsize into CONFIG_NFS_V4]
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index a9828baaa445..bb7432d83b5a 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -1206,6 +1206,21 @@ static void nfs4_init_session(struct nfs_client *clp,
 #endif /* CONFIG_NFS_V4_1 */
 }
 
+/*
+ * Session has been established, and the client marked ready.
+ * Set the mount rsize and wsize with negotiated fore channel
+ * attributes which will be bound checked in nfs_server_set_fsinfo.
+ */
+static void nfs4_session_set_rwsize(struct nfs_server *server)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (!nfs4_has_session(server->nfs_client))
+		return;
+	server->rsize = server->nfs_client->cl_session->fc_attrs.max_resp_sz;
+	server->wsize = server->nfs_client->cl_session->fc_attrs.max_rqst_sz;
+#endif /* CONFIG_NFS_V4_1 */
+}
+
 /*
  * Create a version 4 volume record
  */
@@ -1296,6 +1311,8 @@ struct nfs_server *nfs4_create_server(const struct nfs_parsed_mount_data *data,
 		(unsigned long long) server->fsid.minor);
 	dprintk("Mount FH: %d\n", mntfh->size);
 
+	nfs4_session_set_rwsize(server);
+
 	error = nfs_probe_fsinfo(server, mntfh, &fattr);
 	if (error < 0)
 		goto error;
-- 
cgit v1.2.3-59-g8ed1b


From 0f3e66c6a6cae479028d31198288e524fe5ff90d Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:34 -0400
Subject: nfs41: destroy_session operation

Implement the destroy_session operation conforming to
http://tools.ietf.org/html/draft-ietf-nfsv4-minorversion1-26

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Andy Adamson<andros@umich.edu>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove extraneous rpc_clnt pointer]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41; NFS_CS_READY required for DESTROY_SESSION]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
[nfs41: fix encode_destroy_session's xdr Xcoding pointer type]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c | 30 ++++++++++++++++++++++++++++
 fs/nfs/nfs4xdr.c  | 59 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 fs/nfs/super.c    |  2 ++
 3 files changed, 91 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 4d7a8b9e1291..59e8bf5b10bb 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4532,6 +4532,36 @@ out:
 	return status;
 }
 
+/*
+ * Issue the over-the-wire RPC DESTROY_SESSION.
+ * The caller must serialize access to this routine.
+ */
+int nfs4_proc_destroy_session(struct nfs4_session *session)
+{
+	int status = 0;
+	struct rpc_message msg;
+
+	dprintk("--> nfs4_proc_destroy_session\n");
+
+	/* session is still being setup */
+	if (session->clp->cl_cons_state != NFS_CS_READY)
+		return status;
+
+	msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_DESTROY_SESSION];
+	msg.rpc_argp = session;
+	msg.rpc_resp = NULL;
+	msg.rpc_cred = NULL;
+	status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0);
+
+	if (status)
+		printk(KERN_WARNING
+			"Got error %d from the server on DESTROY_SESSION. "
+			"Session has been destroyed regardless...\n", status);
+
+	dprintk("<-- nfs4_proc_destroy_session\n");
+	return status;
+}
+
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = {
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 7a243a2cf0be..a81e49648ccb 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -294,6 +294,8 @@ static int nfs4_stat_to_errno(int);
 				     1 /* csr_flags */ + \
 				     decode_channel_attrs_maxsz + \
 				     decode_channel_attrs_maxsz)
+#define encode_destroy_session_maxsz    (op_encode_hdr_maxsz + 4)
+#define decode_destroy_session_maxsz    (op_decode_hdr_maxsz)
 #define encode_sequence_maxsz	0 /* stub */
 #define decode_sequence_maxsz	0 /* stub */
 #else /* CONFIG_NFS_V4_1 */
@@ -655,6 +657,10 @@ static int nfs4_stat_to_errno(int);
 #define NFS4_dec_create_session_sz \
 				(compound_decode_hdr_maxsz + \
 				 decode_create_session_maxsz)
+#define NFS4_enc_destroy_session_sz	(compound_encode_hdr_maxsz + \
+					 encode_destroy_session_maxsz)
+#define NFS4_dec_destroy_session_sz	(compound_decode_hdr_maxsz + \
+					 decode_destroy_session_maxsz)
 #define NFS4_enc_get_lease_time_sz	(compound_encode_hdr_maxsz + \
 					 encode_sequence_maxsz + \
 					 encode_putrootfh_maxsz + \
@@ -1609,6 +1615,18 @@ static void encode_create_session(struct xdr_stream *xdr,
 	hdr->nops++;
 	hdr->replen += decode_create_session_maxsz;
 }
+
+static void encode_destroy_session(struct xdr_stream *xdr,
+				   struct nfs4_session *session,
+				   struct compound_hdr *hdr)
+{
+	__be32 *p;
+	RESERVE_SPACE(4 + NFS4_MAX_SESSIONID_LEN);
+	WRITE32(OP_DESTROY_SESSION);
+	WRITEMEM(session->sess_id.data, NFS4_MAX_SESSIONID_LEN);
+	hdr->nops++;
+	hdr->replen += decode_destroy_session_maxsz;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static void encode_sequence(struct xdr_stream *xdr,
@@ -2354,6 +2372,24 @@ static int nfs4_xdr_enc_create_session(struct rpc_rqst *req, uint32_t *p,
 	return 0;
 }
 
+/*
+ * a DESTROY_SESSION request
+ */
+static int nfs4_xdr_enc_destroy_session(struct rpc_rqst *req, uint32_t *p,
+					struct nfs4_session *session)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr = {
+		.minorversion = session->clp->cl_minorversion,
+	};
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, req, &hdr);
+	encode_destroy_session(&xdr, session, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
+
 /*
  * a GET_LEASE_TIME request
  */
@@ -4188,6 +4224,11 @@ static int decode_create_session(struct xdr_stream *xdr,
 		status = decode_chan_attrs(xdr, &session->bc_attrs);
 	return status;
 }
+
+static int decode_destroy_session(struct xdr_stream *xdr, void *dummy)
+{
+	return decode_op_hdr(xdr, OP_DESTROY_SESSION);
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static int decode_sequence(struct xdr_stream *xdr,
@@ -5122,6 +5163,23 @@ static int nfs4_xdr_dec_create_session(struct rpc_rqst *rqstp, uint32_t *p,
 	return status;
 }
 
+/*
+ * a DESTROY_SESSION request
+ */
+static int nfs4_xdr_dec_destroy_session(struct rpc_rqst *rqstp, uint32_t *p,
+					void *dummy)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_destroy_session(&xdr, dummy);
+	return status;
+}
+
 /*
  * a GET_LEASE_TIME request
  */
@@ -5316,6 +5374,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
 #if defined(CONFIG_NFS_V4_1)
   PROC(EXCHANGE_ID,	enc_exchange_id,	dec_exchange_id),
   PROC(CREATE_SESSION,	enc_create_session,	dec_create_session),
+  PROC(DESTROY_SESSION,	enc_destroy_session,	dec_destroy_session),
   PROC(GET_LEASE_TIME,	enc_get_lease_time,	dec_get_lease_time),
 #endif /* CONFIG_NFS_V4_1 */
 };
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 5a8fdc791cc1..73db6f8c145a 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -2485,12 +2485,14 @@ static void nfs4_kill_super(struct super_block *sb)
 {
 	struct nfs_server *server = NFS_SB(sb);
 
+	dprintk("--> %s\n", __func__);
 	nfs_super_return_all_delegations(sb);
 	kill_anon_super(sb);
 
 	nfs4_renewd_prepare_shutdown(server);
 	nfs_fscache_release_super_cookie(sb);
 	nfs_free_server(server);
+	dprintk("<-- %s\n", __func__);
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 8328d59f380e26477b9c1ede99e33d021365bcd1 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:35 -0400
Subject: nfs41: enable nfs_client only nfs4_async_handle_error

The session is per struct nfs_client, not per nfs_server. Allow the handler
to be called with no nfs_server which simplifies the nfs4_proc_async_sequence session renewal call and will let it be used by pnfs file layout data servers.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 59e8bf5b10bb..5fe208b486d4 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3221,10 +3221,8 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen
 }
 
 static int
-nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, struct nfs4_state *state)
+_nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, struct nfs_client *clp, struct nfs4_state *state)
 {
-	struct nfs_client *clp = server->nfs_client;
-
 	if (!clp || task->tk_status >= 0)
 		return 0;
 	switch(task->tk_status) {
@@ -3244,7 +3242,8 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 			task->tk_status = 0;
 			return -EAGAIN;
 		case -NFS4ERR_DELAY:
-			nfs_inc_server_stats(server, NFSIOS_DELAY);
+			if (server)
+				nfs_inc_server_stats(server, NFSIOS_DELAY);
 		case -NFS4ERR_GRACE:
 			rpc_delay(task, NFS4_POLL_RETRY_MAX);
 			task->tk_status = 0;
@@ -3257,6 +3256,12 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 	return 0;
 }
 
+static int
+nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, struct nfs4_state *state)
+{
+	return _nfs4_async_handle_error(task, server, server->nfs_client, state);
+}
+
 int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short port, struct rpc_cred *cred)
 {
 	nfs4_verifier sc_verifier;
-- 
cgit v1.2.3-59-g8ed1b


From fc01cea963a246742ff15e118ce5259e3091352c Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:36 -0400
Subject: nfs41: sequence operation

Implement the sequence operation conforming to
http://tools.ietf.org/html/draft-ietf-nfsv4-minorversion1-26

Check returned sessionid, slotid and slot sequenceid in decode_sequence.

If the server returns different values for sessionID, slotID or slot sequence
number than what was sent, the server is looney tunes.

Pass the sequence operation status to nfs41_sequence_done in order to
determine when to increment the slot sequence ID.

Free slot is separated from sequence done.

Signed-off-by: Rahul Iyer <iyer@netapp.com>
Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Andy Adamson<andros@umich.edu>
[nfs41: sequence res use slotid]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: deref slot table in decode_sequence only for minorversion!=0]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: nfs4_call_sync]
[nfs41: remove SEQ4_STATUS_USE_TK_STATUS]
[nfs41: return ESERVERFAULT in decode_sequence]
[no sr_session, no sr_flags]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use nfs4_call_sync_sequence to renew session lease]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove nfs4_call_sync_sequence forward definition]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: use struct nfs_client for nfs41_proc_async_sequence]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: pass *session in seq_args and seq_res]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41 nfs41_sequence_call_done update error checking]
[nfs41 nfs41_sequence_done update error checking]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: remove switch on error from nfs41_sequence_call_done]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c |  92 +++++++++++++++++++++++++++++++++++++++
 fs/nfs/nfs4xdr.c  | 127 +++++++++++++++++++++++++++++++++++++++++++++++++++---
 2 files changed, 214 insertions(+), 5 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 5fe208b486d4..17768095f0cc 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4567,6 +4567,98 @@ int nfs4_proc_destroy_session(struct nfs4_session *session)
 	return status;
 }
 
+/*
+ * Renew the cl_session lease.
+ */
+static int nfs4_proc_sequence(struct nfs_client *clp, struct rpc_cred *cred)
+{
+	struct nfs4_sequence_args args;
+	struct nfs4_sequence_res res;
+
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SEQUENCE],
+		.rpc_argp = &args,
+		.rpc_resp = &res,
+		.rpc_cred = cred,
+	};
+
+	args.sa_cache_this = 0;
+
+	return nfs4_call_sync_sequence(clp, clp->cl_rpcclient, &msg, &args,
+				       &res, 0);
+}
+
+void nfs41_sequence_call_done(struct rpc_task *task, void *data)
+{
+	struct nfs_client *clp = (struct nfs_client *)data;
+
+	nfs41_sequence_done(clp, task->tk_msg.rpc_resp, task->tk_status);
+
+	if (task->tk_status < 0) {
+		dprintk("%s ERROR %d\n", __func__, task->tk_status);
+
+		if (_nfs4_async_handle_error(task, NULL, clp, NULL)
+								== -EAGAIN) {
+			rpc_restart_call(task);
+			return;
+		}
+	}
+	nfs41_sequence_free_slot(clp, task->tk_msg.rpc_resp);
+	dprintk("%s rpc_cred %p\n", __func__, task->tk_msg.rpc_cred);
+
+	put_rpccred(task->tk_msg.rpc_cred);
+	kfree(task->tk_msg.rpc_argp);
+	kfree(task->tk_msg.rpc_resp);
+
+	dprintk("<-- %s\n", __func__);
+}
+
+static void nfs41_sequence_prepare(struct rpc_task *task, void *data)
+{
+	struct nfs_client *clp;
+	struct nfs4_sequence_args *args;
+	struct nfs4_sequence_res *res;
+
+	clp = (struct nfs_client *)data;
+	args = task->tk_msg.rpc_argp;
+	res = task->tk_msg.rpc_resp;
+
+	if (nfs4_setup_sequence(clp, args, res, 0, task))
+		return;
+	rpc_call_start(task);
+}
+
+static const struct rpc_call_ops nfs41_sequence_ops = {
+	.rpc_call_done = nfs41_sequence_call_done,
+	.rpc_call_prepare = nfs41_sequence_prepare,
+};
+
+static int nfs41_proc_async_sequence(struct nfs_client *clp,
+				     struct rpc_cred *cred)
+{
+	struct nfs4_sequence_args *args;
+	struct nfs4_sequence_res *res;
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SEQUENCE],
+		.rpc_cred = cred,
+	};
+
+	args = kzalloc(sizeof(*args), GFP_KERNEL);
+	if (!args)
+		return -ENOMEM;
+	res = kzalloc(sizeof(*res), GFP_KERNEL);
+	if (!res) {
+		kfree(args);
+		return -ENOMEM;
+	}
+	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
+	msg.rpc_argp = args;
+	msg.rpc_resp = res;
+
+	return rpc_call_async(clp->cl_rpcclient, &msg, RPC_TASK_SOFT,
+			      &nfs41_sequence_ops, (void *)clp);
+}
+
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = {
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index a81e49648ccb..617273e7d47f 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -296,8 +296,10 @@ static int nfs4_stat_to_errno(int);
 				     decode_channel_attrs_maxsz)
 #define encode_destroy_session_maxsz    (op_encode_hdr_maxsz + 4)
 #define decode_destroy_session_maxsz    (op_decode_hdr_maxsz)
-#define encode_sequence_maxsz	0 /* stub */
-#define decode_sequence_maxsz	0 /* stub */
+#define encode_sequence_maxsz	(op_encode_hdr_maxsz + \
+				XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 4)
+#define decode_sequence_maxsz	(op_decode_hdr_maxsz + \
+				XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 5)
 #else /* CONFIG_NFS_V4_1 */
 #define encode_sequence_maxsz	0
 #define decode_sequence_maxsz	0
@@ -661,6 +663,12 @@ static int nfs4_stat_to_errno(int);
 					 encode_destroy_session_maxsz)
 #define NFS4_dec_destroy_session_sz	(compound_decode_hdr_maxsz + \
 					 decode_destroy_session_maxsz)
+#define NFS4_enc_sequence_sz \
+				(compound_decode_hdr_maxsz + \
+				 encode_sequence_maxsz)
+#define NFS4_dec_sequence_sz \
+				(compound_decode_hdr_maxsz + \
+				 decode_sequence_maxsz)
 #define NFS4_enc_get_lease_time_sz	(compound_encode_hdr_maxsz + \
 					 encode_sequence_maxsz + \
 					 encode_putrootfh_maxsz + \
@@ -1635,11 +1643,39 @@ static void encode_sequence(struct xdr_stream *xdr,
 {
 #if defined(CONFIG_NFS_V4_1)
 	struct nfs4_session *session = args->sa_session;
+	struct nfs4_slot_table *tp;
+	struct nfs4_slot *slot;
+	__be32 *p;
 
 	if (!session)
 		return;
 
-	/* stub */
+	tp = &session->fc_slot_table;
+
+	WARN_ON(args->sa_slotid == NFS4_MAX_SLOT_TABLE);
+	slot = tp->slots + args->sa_slotid;
+
+	RESERVE_SPACE(4);
+	WRITE32(OP_SEQUENCE);
+
+	/*
+	 * Sessionid + seqid + slotid + max slotid + cache_this
+	 */
+	dprintk("%s: sessionid=%u:%u:%u:%u seqid=%d slotid=%d "
+		"max_slotid=%d cache_this=%d\n",
+		__func__,
+		((u32 *)session->sess_id.data)[0],
+		((u32 *)session->sess_id.data)[1],
+		((u32 *)session->sess_id.data)[2],
+		((u32 *)session->sess_id.data)[3],
+		slot->seq_nr, args->sa_slotid,
+		tp->highest_used_slotid, args->sa_cache_this);
+	RESERVE_SPACE(NFS4_MAX_SESSIONID_LEN + 16);
+	WRITEMEM(session->sess_id.data, NFS4_MAX_SESSIONID_LEN);
+	WRITE32(slot->seq_nr);
+	WRITE32(args->sa_slotid);
+	WRITE32(tp->highest_used_slotid);
+	WRITE32(args->sa_cache_this);
 	hdr->nops++;
 	hdr->replen += decode_sequence_maxsz;
 #endif /* CONFIG_NFS_V4_1 */
@@ -2390,6 +2426,24 @@ static int nfs4_xdr_enc_destroy_session(struct rpc_rqst *req, uint32_t *p,
 	return 0;
 }
 
+/*
+ * a SEQUENCE request
+ */
+static int nfs4_xdr_enc_sequence(struct rpc_rqst *req, uint32_t *p,
+				 struct nfs4_sequence_args *args)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(args),
+	};
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, args, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
+
 /*
  * a GET_LEASE_TIME request
  */
@@ -4236,13 +4290,58 @@ static int decode_sequence(struct xdr_stream *xdr,
 			   struct rpc_rqst *rqstp)
 {
 #if defined(CONFIG_NFS_V4_1)
+	struct nfs4_slot *slot;
+	struct nfs4_sessionid id;
+	u32 dummy;
+	int status;
+	__be32 *p;
+
 	if (!res->sr_session)
 		return 0;
 
-	/* stub */
-#endif /* CONFIG_NFS_V4_1 */
+	status = decode_op_hdr(xdr, OP_SEQUENCE);
+	if (status)
+		goto out_err;
 
+	/*
+	 * If the server returns different values for sessionID, slotID or
+	 * sequence number, the server is looney tunes.
+	 */
+	status = -ESERVERFAULT;
+
+	slot = &res->sr_session->fc_slot_table.slots[res->sr_slotid];
+	READ_BUF(NFS4_MAX_SESSIONID_LEN + 20);
+	COPYMEM(id.data, NFS4_MAX_SESSIONID_LEN);
+	if (memcmp(id.data, res->sr_session->sess_id.data,
+		   NFS4_MAX_SESSIONID_LEN)) {
+		dprintk("%s Invalid session id\n", __func__);
+		goto out_err;
+	}
+	/* seqid */
+	READ32(dummy);
+	if (dummy != slot->seq_nr) {
+		dprintk("%s Invalid sequence number\n", __func__);
+		goto out_err;
+	}
+	/* slot id */
+	READ32(dummy);
+	if (dummy != res->sr_slotid) {
+		dprintk("%s Invalid slot id\n", __func__);
+		goto out_err;
+	}
+	/* highest slot id - currently not processed */
+	READ32(dummy);
+	/* target highest slot id - currently not processed */
+	READ32(dummy);
+	/* result flags - currently not processed */
+	READ32(dummy);
+	status = 0;
+out_err:
+	res->sr_status = status;
+	return status;
+#else  /* CONFIG_NFS_V4_1 */
 	return 0;
+#endif /* CONFIG_NFS_V4_1 */
 }
 
 /*
@@ -5180,6 +5279,23 @@ static int nfs4_xdr_dec_destroy_session(struct rpc_rqst *rqstp, uint32_t *p,
 	return status;
 }
 
+/*
+ * a SEQUENCE request
+ */
+static int nfs4_xdr_dec_sequence(struct rpc_rqst *rqstp, uint32_t *p,
+				 struct nfs4_sequence_res *res)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_sequence(&xdr, res, rqstp);
+	return status;
+}
+
 /*
  * a GET_LEASE_TIME request
  */
@@ -5375,6 +5491,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
   PROC(EXCHANGE_ID,	enc_exchange_id,	dec_exchange_id),
   PROC(CREATE_SESSION,	enc_create_session,	dec_create_session),
   PROC(DESTROY_SESSION,	enc_destroy_session,	dec_destroy_session),
+  PROC(SEQUENCE,	enc_sequence,	dec_sequence),
   PROC(GET_LEASE_TIME,	enc_get_lease_time,	dec_get_lease_time),
 #endif /* CONFIG_NFS_V4_1 */
 };
-- 
cgit v1.2.3-59-g8ed1b


From ac72b7b3b3263ce64d55094eac1d1bde5f34e64a Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:37 -0400
Subject: nfs41: reset the session slot table

Separated from nfs41: schedule async session reset

Do not kfree the session slot table upon session reset, just re-initialize it.
Add a boolean to nfs4_proc_create_session to inidicate if this is a
session reset or a session initialization.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 44 +++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 41 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 17768095f0cc..0b1214740248 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4289,6 +4289,37 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo)
 	return status;
 }
 
+/* Reset a slot table */
+static int nfs4_reset_slot_table(struct nfs4_session *session)
+{
+	struct nfs4_slot_table *tbl = &session->fc_slot_table;
+	int i, max_slots = session->fc_attrs.max_reqs;
+	int old_max_slots = session->fc_slot_table.max_slots;
+	int ret = 0;
+
+	dprintk("--> %s: max_reqs=%u, tbl %p\n", __func__,
+		session->fc_attrs.max_reqs, tbl);
+
+	/* Until we have dynamic slot table adjustment, insist
+	 * upon the same slot table size */
+	if (max_slots != old_max_slots) {
+		dprintk("%s reset slot table does't match old\n",
+			__func__);
+		ret = -EINVAL; /*XXX NFS4ERR_REQ_TOO_BIG ? */
+		goto out;
+	}
+	spin_lock(&tbl->slot_tbl_lock);
+	for (i = 0; i < max_slots; ++i)
+		tbl->slots[i].seq_nr = 1;
+	tbl->highest_used_slotid = -1;
+	spin_unlock(&tbl->slot_tbl_lock);
+	dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__,
+		tbl, tbl->slots, tbl->max_slots);
+out:
+	dprintk("<-- %s: return %d\n", __func__, ret);
+	return ret;
+}
+
 /*
  * Initialize slot table
  */
@@ -4497,7 +4528,7 @@ static int _nfs4_proc_create_session(struct nfs_client *clp)
  * It is the responsibility of the caller to verify the session is
  * expired before calling this routine.
  */
-int nfs4_proc_create_session(struct nfs_client *clp)
+int nfs4_proc_create_session(struct nfs_client *clp, int reset)
 {
 	int status;
 	unsigned *ptr;
@@ -4510,8 +4541,11 @@ int nfs4_proc_create_session(struct nfs_client *clp)
 	if (status)
 		goto out;
 
-	/* Init the fore channel */
-	status = nfs4_init_slot_table(session);
+	/* Init or reset the fore channel */
+	if (reset)
+		status = nfs4_reset_slot_table(session);
+	else
+		status = nfs4_init_slot_table(session);
 	dprintk("fore channel slot table initialization returned %d\n", status);
 	if (status)
 		goto out;
@@ -4520,6 +4554,10 @@ int nfs4_proc_create_session(struct nfs_client *clp)
 	dprintk("%s client>seqid %d sessionid %u:%u:%u:%u\n", __func__,
 		clp->cl_seqid, ptr[0], ptr[1], ptr[2], ptr[3]);
 
+	if (reset)
+		/* Lease time is aleady set */
+		goto out;
+
 	/* Get the lease time */
 	status = nfs4_proc_get_lease_time(clp, &fsinfo);
 	if (status == 0) {
-- 
cgit v1.2.3-59-g8ed1b


From 76db6d9500caeaa774a3e32a997eba30bbdc176b Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:38 -0400
Subject: nfs41: add session setup to the state manager

At mount, nfs_alloc_client sets the cl_state NFS4CLNT_LEASE_EXPIRED bit
and nfs4_alloc_session sets the NFS4CLNT_SESSION_SETUP bit, so both bits are
set when nfs4_lookup_root calls nfs4_recover_expired_lease which schedules
the nfs4_state_manager and waits for it to complete.

Place the session setup after the clientid establishment in nfs4_state_manager
so that the session is setup right after the clientid has been established
without rescheduling the state manager.

Unlike nfsv4.0, the nfs_client struct is not ready to use until the session
has been established.  Postpone marking the nfs_client struct to NFS_CS_READY
until after a successful CREATE_SESSION call so that other threads cannot use
the client until the session is established.

If the EXCHANGE_ID call fails and the session has not been setup (the
NFS4CLNT_SESSION_SETUP bit is set), mark the client with the error and return.

If the session setup CREATE_SESSION call fails with NFS4ERR_STALE_CLIENTID
which could occur due to server reboot or network partition inbetween the
EXCHANGE_ID and CREATE_SESSION call, reset the NFS4CLNT_LEASE_EXPIRED and
NFS4CLNT_SESSION_SETUP bits and try again.

If the CREATE_SESSION call fails with other errors, mark the client with
the error and return.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>

[nfs41: NFS_CS_SESSION_SETUP cl_cons_state for back channel setup]
  On session setup, the CREATE_SESSION reply races with the server back channel
  probe which needs to succeed to setup the back channel. Set a new
  cl_cons_state NFS_CS_SESSION_SETUP just prior to the CREATE_SESSION call
  and add it as a valid state to nfs_find_client so that the client back channel
  can find the nfs_client struct and won't drop the server backchannel probe.
  Use a new cl_cons_state so that NFSv4.0 back channel behaviour which only
  sets NFS_CS_READY is unchanged.
  Adjust waiting on the nfs_client_active_wq accordingly.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>

[nfs41: rename NFS_CS_SESSION_SETUP to NFS_CS_SESSION_INITING]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: set NFS_CL_SESSION_INITING in alloc_session]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: move session setup into a function]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[moved nfs4_proc_create_session declaration here]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c           | 10 ++++++----
 fs/nfs/internal.h         |  1 +
 fs/nfs/nfs4_fs.h          |  2 ++
 fs/nfs/nfs4proc.c         | 10 ++++++++++
 fs/nfs/nfs4state.c        | 35 ++++++++++++++++++++++++++++++++++-
 include/linux/nfs_fs_sb.h |  1 +
 6 files changed, 54 insertions(+), 5 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index bb7432d83b5a..d28a987f569e 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -366,7 +366,8 @@ struct nfs_client *nfs_find_client(const struct sockaddr *addr, u32 nfsversion)
 		struct sockaddr *clap = (struct sockaddr *)&clp->cl_addr;
 
 		/* Don't match clients that failed to initialise properly */
-		if (clp->cl_cons_state != NFS_CS_READY)
+		if (!(clp->cl_cons_state == NFS_CS_READY ||
+		      clp->cl_cons_state == NFS_CS_SESSION_INITING))
 			continue;
 
 		/* Different NFS versions cannot share the same nfs_client */
@@ -499,7 +500,7 @@ found_client:
 		nfs_free_client(new);
 
 	error = wait_event_killable(nfs_client_active_wq,
-				clp->cl_cons_state != NFS_CS_INITING);
+				clp->cl_cons_state < NFS_CS_INITING);
 	if (error < 0) {
 		nfs_put_client(clp);
 		return ERR_PTR(-ERESTARTSYS);
@@ -520,7 +521,7 @@ found_client:
 /*
  * Mark a server as ready or failed
  */
-static void nfs_mark_client_ready(struct nfs_client *clp, int state)
+void nfs_mark_client_ready(struct nfs_client *clp, int state)
 {
 	clp->cl_cons_state = state;
 	wake_up_all(&nfs_client_active_wq);
@@ -1135,7 +1136,8 @@ static int nfs4_init_client(struct nfs_client *clp,
 	if (error < 0)
 		goto error;
 
-	nfs_mark_client_ready(clp, NFS_CS_READY);
+	if (!nfs4_has_session(clp))
+		nfs_mark_client_ready(clp, NFS_CS_READY);
 	return 0;
 
 error:
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index f62bc5226155..f3b310e8ea03 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -100,6 +100,7 @@ extern void nfs_free_server(struct nfs_server *server);
 extern struct nfs_server *nfs_clone_server(struct nfs_server *,
 					   struct nfs_fh *,
 					   struct nfs_fattr *);
+extern void nfs_mark_client_ready(struct nfs_client *clp, int state);
 #ifdef CONFIG_PROC_FS
 extern int __init nfs_fs_proc_init(void);
 extern void nfs_fs_proc_exit(void);
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index eccf4e93e7d7..288717abaddc 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -44,6 +44,7 @@ enum nfs4_client_state {
 	NFS4CLNT_RECLAIM_REBOOT,
 	NFS4CLNT_RECLAIM_NOGRACE,
 	NFS4CLNT_DELEGRETURN,
+	NFS4CLNT_SESSION_SETUP,
 };
 
 /*
@@ -208,6 +209,7 @@ extern int nfs4_setup_sequence(struct nfs_client *clp,
 		int cache_reply, struct rpc_task *task);
 extern void nfs4_destroy_session(struct nfs4_session *session);
 extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp);
+extern int nfs4_proc_create_session(struct nfs_client *, int reset);
 #else /* CONFIG_NFS_v4_1 */
 static inline int nfs4_setup_sequence(struct nfs_client *clp,
 		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 0b1214740248..7fc0c9c8f5e3 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4382,6 +4382,16 @@ struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
 	session = kzalloc(sizeof(struct nfs4_session), GFP_KERNEL);
 	if (!session)
 		return NULL;
+
+	set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+	/*
+	 * The create session reply races with the server back
+	 * channel probe. Mark the client NFS_CS_SESSION_INITING
+	 * so that the client back channel can find the
+	 * nfs_client struct
+	 */
+	clp->cl_cons_state = NFS_CS_SESSION_INITING;
+
 	tbl = &session->fc_slot_table;
 	spin_lock_init(&tbl->slot_tbl_lock);
 	rpc_init_wait_queue(&tbl->slot_tbl_waitq, "Slot table");
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index bc683ed477e1..df5b4807daa7 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1113,6 +1113,27 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 	return status;
 }
 
+#ifdef CONFIG_NFS_V4_1
+
+static int nfs4_initialize_session(struct nfs_client *clp)
+{
+	int status;
+
+	status = nfs4_proc_create_session(clp, 0);
+	if (!status) {
+		nfs_mark_client_ready(clp, NFS_CS_READY);
+	} else if (status == -NFS4ERR_STALE_CLIENTID) {
+		set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+		set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+	} else {
+		nfs_mark_client_ready(clp, status);
+	}
+	return status;
+}
+#else /* CONFIG_NFS_V4_1 */
+static int nfs4_initialize_session(struct nfs_client *clp) { return 0; }
+#endif /* CONFIG_NFS_V4_1 */
+
 static void nfs4_state_manager(struct nfs_client *clp)
 {
 	int status = 0;
@@ -1126,6 +1147,9 @@ static void nfs4_state_manager(struct nfs_client *clp)
 				set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
 				if (status == -EAGAIN)
 					continue;
+				if (clp->cl_cons_state ==
+							NFS_CS_SESSION_INITING)
+					nfs_mark_client_ready(clp, status);
 				goto out_error;
 			}
 			clear_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state);
@@ -1136,7 +1160,16 @@ static void nfs4_state_manager(struct nfs_client *clp)
 			if (status != 0)
 				continue;
 		}
-
+		/* Setup the session */
+		if (nfs4_has_session(clp) &&
+		   test_and_clear_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)) {
+			status = nfs4_initialize_session(clp);
+			if (status) {
+				if (status == -NFS4ERR_STALE_CLIENTID)
+					continue;
+				goto out_error;
+			}
+		}
 		/* First recover reboot state... */
 		if (test_and_clear_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state)) {
 			status = nfs4_do_reclaim(clp, &nfs4_reboot_recovery_ops);
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index 435ed556efb5..d0902ccec9ce 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -24,6 +24,7 @@ struct nfs_client {
 	int			cl_cons_state;	/* current construction state (-ve: init error) */
 #define NFS_CS_READY		0		/* ready to be used */
 #define NFS_CS_INITING		1		/* busy initialising */
+#define NFS_CS_SESSION_INITING	2		/* busy initialising  session */
 	unsigned long		cl_res_state;	/* NFS resources state */
 #define NFS_CS_CALLBACK		1		/* - callback started */
 #define NFS_CS_IDMAP		2		/* - idmap started */
-- 
cgit v1.2.3-59-g8ed1b


From c3fad1b1aaf850bf692642642ace7cd0d64af0a3 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:39 -0400
Subject: nfs41: add session reset to state manager

Move the code to reset a session from the session_reclaimer to the
nfs4_state_manager.  Destroy the session, and create a new one. Treat
NFS4ERR_BADSESSION and NFS4ERR_DEADSESSION as a successful
nfs4_proc_destroy_session. Signal nfs4_proc_create_session that this is a
session reset so that the session slot table is re-used.

If the clientid is stale, set both NFS4CLNT_LEASE_EXPIRED and
NFS4CLNT_SESSION_SETUP bits and retry.

Use a switch statement in nfs4_session_recovery_handle_error for future
patche which will add handling for other errors.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>

[nfs41: session reset in nfs4_recovery_handle_error]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: reset session on nfs4_do_reclaim session reset error]
    If nfs4_do_reclaim gets a session reset error, nfs4_recovery_handle_error
    will set the NFS4CLNT_SESSION_SETUP bit, and the state manager should
    continue processing to reset the session.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[move nfs4_proc_destroy_session declaration here]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4_fs.h   |  1 +
 fs/nfs/nfs4state.c | 51 +++++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 50 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 288717abaddc..7cc1cddfd9ba 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -210,6 +210,7 @@ extern int nfs4_setup_sequence(struct nfs_client *clp,
 extern void nfs4_destroy_session(struct nfs4_session *session);
 extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp);
 extern int nfs4_proc_create_session(struct nfs_client *, int reset);
+extern int nfs4_proc_destroy_session(struct nfs4_session *);
 #else /* CONFIG_NFS_v4_1 */
 static inline int nfs4_setup_sequence(struct nfs_client *clp,
 		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index df5b4807daa7..8b7f007adc42 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1042,6 +1042,14 @@ static void nfs4_recovery_handle_error(struct nfs_client *clp, int error)
 		case -NFS4ERR_EXPIRED:
 			set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
 			nfs4_state_start_reclaim_nograce(clp);
+		case -NFS4ERR_BADSESSION:
+		case -NFS4ERR_BADSLOT:
+		case -NFS4ERR_BAD_HIGH_SLOT:
+		case -NFS4ERR_DEADSESSION:
+		case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+		case -NFS4ERR_SEQ_FALSE_RETRY:
+		case -NFS4ERR_SEQ_MISORDERED:
+			set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
 	}
 }
 
@@ -1114,6 +1122,36 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 }
 
 #ifdef CONFIG_NFS_V4_1
+static void nfs4_session_recovery_handle_error(struct nfs_client *clp, int err)
+{
+	switch (err) {
+	case -NFS4ERR_STALE_CLIENTID:
+		set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+		set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+	}
+}
+
+static int nfs4_reset_session(struct nfs_client *clp)
+{
+	int status;
+
+	status = nfs4_proc_destroy_session(clp->cl_session);
+	if (status && status != -NFS4ERR_BADSESSION &&
+	    status != -NFS4ERR_DEADSESSION) {
+		nfs4_session_recovery_handle_error(clp, status);
+		goto out;
+	}
+
+	memset(clp->cl_session->sess_id.data, 0, NFS4_MAX_SESSIONID_LEN);
+	status = nfs4_proc_create_session(clp, 1);
+	if (status)
+		nfs4_session_recovery_handle_error(clp, status);
+		/* fall through*/
+out:
+	/* Wake up the next rpc task even on error */
+	rpc_wake_up_next(&clp->cl_session->fc_slot_table.slot_tbl_waitq);
+	return status;
+}
 
 static int nfs4_initialize_session(struct nfs_client *clp)
 {
@@ -1131,6 +1169,7 @@ static int nfs4_initialize_session(struct nfs_client *clp)
 	return status;
 }
 #else /* CONFIG_NFS_V4_1 */
+static int nfs4_reset_session(struct nfs_client *clp) { return 0; }
 static int nfs4_initialize_session(struct nfs_client *clp) { return 0; }
 #endif /* CONFIG_NFS_V4_1 */
 
@@ -1160,10 +1199,13 @@ static void nfs4_state_manager(struct nfs_client *clp)
 			if (status != 0)
 				continue;
 		}
-		/* Setup the session */
+		/* Initialize or reset the session */
 		if (nfs4_has_session(clp) &&
 		   test_and_clear_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)) {
-			status = nfs4_initialize_session(clp);
+			if (clp->cl_cons_state == NFS_CS_SESSION_INITING)
+				status = nfs4_initialize_session(clp);
+			else
+				status = nfs4_reset_session(clp);
 			if (status) {
 				if (status == -NFS4ERR_STALE_CLIENTID)
 					continue;
@@ -1175,6 +1217,8 @@ static void nfs4_state_manager(struct nfs_client *clp)
 			status = nfs4_do_reclaim(clp, &nfs4_reboot_recovery_ops);
 			if (status == -NFS4ERR_STALE_CLIENTID)
 				continue;
+			if (test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state))
+				continue;
 			nfs4_state_end_reclaim_reboot(clp);
 			continue;
 		}
@@ -1188,6 +1232,9 @@ static void nfs4_state_manager(struct nfs_client *clp)
 					continue;
 				if (status == -NFS4ERR_EXPIRED)
 					continue;
+				if (test_bit(NFS4CLNT_SESSION_SETUP,
+								&clp->cl_state))
+					continue;
 				goto out_error;
 			} else
 				nfs4_state_end_reclaim_nograce(clp);
-- 
cgit v1.2.3-59-g8ed1b


From aae2006e9b0c294114915c13022fa348e1a88023 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:40 -0400
Subject: nfs41: sunrpc: Export the call prepare state for session reset

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 include/linux/sunrpc/clnt.h  |  1 +
 include/linux/sunrpc/sched.h |  1 +
 net/sunrpc/clnt.c            | 13 +++++++++++++
 net/sunrpc/sched.c           |  2 +-
 4 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h
index c39a21040dcb..37881f1a0bd7 100644
--- a/include/linux/sunrpc/clnt.h
+++ b/include/linux/sunrpc/clnt.h
@@ -143,6 +143,7 @@ int		rpc_call_sync(struct rpc_clnt *clnt,
 			      const struct rpc_message *msg, int flags);
 struct rpc_task *rpc_call_null(struct rpc_clnt *clnt, struct rpc_cred *cred,
 			       int flags);
+void		rpc_restart_call_prepare(struct rpc_task *);
 void		rpc_restart_call(struct rpc_task *);
 void		rpc_setbufsize(struct rpc_clnt *, unsigned int, unsigned int);
 size_t		rpc_max_payload(struct rpc_clnt *);
diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index 64981a2f1cae..177376880fab 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -237,6 +237,7 @@ void		rpc_show_tasks(void);
 int		rpc_init_mempool(void);
 void		rpc_destroy_mempool(void);
 extern struct workqueue_struct *rpciod_workqueue;
+void		rpc_prepare_task(struct rpc_task *task);
 
 static inline void rpc_exit(struct rpc_task *task, int status)
 {
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 5abab094441f..d00e8135f866 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -694,6 +694,19 @@ void rpc_force_rebind(struct rpc_clnt *clnt)
 }
 EXPORT_SYMBOL_GPL(rpc_force_rebind);
 
+/*
+ * Restart an (async) RPC call from the call_prepare state.
+ * Usually called from within the exit handler.
+ */
+void
+rpc_restart_call_prepare(struct rpc_task *task)
+{
+	if (RPC_ASSASSINATED(task))
+		return;
+	task->tk_action = rpc_prepare_task;
+}
+EXPORT_SYMBOL_GPL(rpc_restart_call_prepare);
+
 /*
  * Restart an (async) RPC call. Usually called from within the
  * exit handler.
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index ff50a0546865..1102ce1251f7 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -569,7 +569,7 @@ EXPORT_SYMBOL_GPL(rpc_delay);
 /*
  * Helper to call task->tk_ops->rpc_call_prepare
  */
-static void rpc_prepare_task(struct rpc_task *task)
+void rpc_prepare_task(struct rpc_task *task)
 {
 	task->tk_ops->rpc_call_prepare(task, task->tk_calldata);
 }
-- 
cgit v1.2.3-59-g8ed1b


From eedc020e718b8ce45381383ec66030f09eb02a1e Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:41 -0400
Subject: nfs41: use rpc prepare call state for session reset

[nfs41: change nfs4_restart_rpc argument]
[nfs41: check for session not minorversion]
[nfs41: trigger the state manager for session reset]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[always define nfs4_restart_rpc]
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/internal.h | 39 +++++++++++++++++++++++++++------------
 fs/nfs/nfs4proc.c | 15 ++++++++-------
 fs/nfs/read.c     |  2 +-
 fs/nfs/unlink.c   |  2 +-
 fs/nfs/write.c    |  8 ++++----
 5 files changed, 41 insertions(+), 25 deletions(-)

diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index f3b310e8ea03..78508f29c03a 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -2,6 +2,7 @@
  * NFS internal definitions
  */
 
+#include "nfs4_fs.h"
 #include <linux/mount.h>
 #include <linux/security.h>
 
@@ -17,6 +18,18 @@ struct nfs_string;
  */
 #define NFS_MAX_READAHEAD	(RPC_DEF_SLOT_TABLE - 1)
 
+/*
+ * Determine if sessions are in use.
+ */
+static inline int nfs4_has_session(const struct nfs_client *clp)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (clp->cl_session)
+		return 1;
+#endif /* CONFIG_NFS_V4_1 */
+	return 0;
+}
+
 struct nfs_clone_mount {
 	const struct super_block *sb;
 	const struct dentry *dentry;
@@ -148,6 +161,20 @@ extern __be32 * nfs_decode_dirent(__be32 *, struct nfs_entry *, int);
 extern struct rpc_procinfo nfs3_procedures[];
 extern __be32 *nfs3_decode_dirent(__be32 *, struct nfs_entry *, int);
 
+/* nfs4proc.c */
+static inline void nfs4_restart_rpc(struct rpc_task *task,
+				    const struct nfs_client *clp)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (nfs4_has_session(clp) &&
+	    test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)) {
+		rpc_restart_call_prepare(task);
+		return;
+	}
+#endif /* CONFIG_NFS_V4_1 */
+	rpc_restart_call(task);
+}
+
 /* nfs4xdr.c */
 #ifdef CONFIG_NFS_V4
 extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus);
@@ -225,18 +252,6 @@ extern int _nfs4_call_sync_session(struct nfs_server *server,
 				   struct nfs4_sequence_res *res,
 				   int cache_reply);
 
-/*
- * Determine if sessions are in use.
- */
-static inline int nfs4_has_session(const struct nfs_client *clp)
-{
-#ifdef CONFIG_NFS_V4_1
-	if (clp->cl_session)
-		return 1;
-#endif /* CONFIG_NFS_V4_1 */
-	return 0;
-}
-
 #ifdef CONFIG_NFS_V4_1
 extern void nfs41_sequence_free_slot(const struct nfs_client *,
 				     struct nfs4_sequence_res *res);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 7fc0c9c8f5e3..2d45606a8085 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1661,7 +1661,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 				break;
 		default:
 			if (nfs4_async_handle_error(task, server, state) == -EAGAIN) {
-				rpc_restart_call(task);
+				nfs4_restart_rpc(task, server->nfs_client);
 				return;
 			}
 	}
@@ -2874,7 +2874,7 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data)
 	nfs4_sequence_done(server, &data->res.seq_res, task->tk_status);
 
 	if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) {
-		rpc_restart_call(task);
+		nfs4_restart_rpc(task, server->nfs_client);
 		return -EAGAIN;
 	}
 
@@ -2899,7 +2899,7 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data)
 			   task->tk_status);
 
 	if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) {
-		rpc_restart_call(task);
+		nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
 		return -EAGAIN;
 	}
 	if (task->tk_status >= 0) {
@@ -2927,7 +2927,7 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data)
 	nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res,
 			   task->tk_status);
 	if (nfs4_async_handle_error(task, NFS_SERVER(inode), NULL) == -EAGAIN) {
-		rpc_restart_call(task);
+		nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
 		return -EAGAIN;
 	}
 	nfs4_sequence_free_slot(NFS_SERVER(inode)->nfs_client,
@@ -3628,7 +3628,8 @@ static void nfs4_locku_done(struct rpc_task *task, void *data)
 			break;
 		default:
 			if (nfs4_async_handle_error(task, calldata->server, NULL) == -EAGAIN)
-				rpc_restart_call(task);
+				nfs4_restart_rpc(task,
+						calldata->server->nfs_client);
 	}
 	nfs4_sequence_free_slot(calldata->server->nfs_client,
 				&calldata->res.seq_res);
@@ -4237,7 +4238,7 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata)
 		dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status);
 		rpc_delay(task, NFS4_POLL_RETRY_MIN);
 		task->tk_status = 0;
-		rpc_restart_call(task);
+		nfs4_restart_rpc(task, data->clp);
 		return;
 	}
 	nfs41_sequence_free_slot(data->clp, &data->res->lr_seq_res);
@@ -4647,7 +4648,7 @@ void nfs41_sequence_call_done(struct rpc_task *task, void *data)
 
 		if (_nfs4_async_handle_error(task, NULL, clp, NULL)
 								== -EAGAIN) {
-			rpc_restart_call(task);
+			nfs4_restart_rpc(task, clp);
 			return;
 		}
 	}
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index d71f0d55ebde..96c4ebfa46f4 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -371,7 +371,7 @@ static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data
 	argp->offset += resp->count;
 	argp->pgbase += resp->count;
 	argp->count -= resp->count;
-	rpc_restart_call(task);
+	nfs4_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client);
 	return;
 out:
 	nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client,
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index 089a21b4e3f9..1064c91ae810 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -83,7 +83,7 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata)
 	struct inode *dir = data->dir;
 
 	if (!NFS_PROTO(dir)->unlink_done(task, dir))
-		rpc_restart_call(task);
+		nfs4_restart_rpc(task, NFS_SERVER(dir)->nfs_client);
 }
 
 /**
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 85a76409de13..ce728829f79a 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1145,6 +1145,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 {
 	struct nfs_writeargs	*argp = &data->args;
 	struct nfs_writeres	*resp = &data->res;
+	struct nfs_server	*server = NFS_SERVER(data->inode);
 	int status;
 
 	dprintk("NFS: %5u nfs_writeback_done (status %d)\n",
@@ -1177,7 +1178,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 		if (time_before(complain, jiffies)) {
 			dprintk("NFS:       faulty NFS server %s:"
 				" (committed = %d) != (stable = %d)\n",
-				NFS_SERVER(data->inode)->nfs_client->cl_hostname,
+				server->nfs_client->cl_hostname,
 				resp->verf->committed, argp->stable);
 			complain = jiffies + 300 * HZ;
 		}
@@ -1203,7 +1204,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 				 */
 				argp->stable = NFS_FILE_SYNC;
 			}
-			rpc_restart_call(task);
+			nfs4_restart_rpc(task, server->nfs_client);
 			return -EAGAIN;
 		}
 		if (time_before(complain, jiffies)) {
@@ -1215,8 +1216,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 		/* Can't do anything about it except throw an error. */
 		task->tk_status = -EIO;
 	}
-	nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client,
-				&data->res.seq_res);
+	nfs4_sequence_free_slot(server->nfs_client, &data->res.seq_res);
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 4745e3154bf33048d970d1519703089493988f65 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:42 -0400
Subject: nfs41: kick start nfs41 session recovery when handling errors

Remove checking for any errors that the SEQUENCE operation does not return.
-NFS4ERR_STALE_CLIENTID, NFS4ERR_EXPIRED, NFS4ERR_CB_PATH_DOWN, NFS4ERR_BACK_CHAN_BUSY, NFS4ERR_OP_NOT_IN_SESSION.

SEQUENCE operation error recovery is very primative, we only reset the session.

Remove checking for any errors that are returned by the SEQUENCE operation, but
that resetting the session won't address.
NFS4ERR_RETRY_UNCACHED_REP, NFS4ERR_SEQUENCE_POS,NFS4ERR_TOO_MANY_OPS.

Add error checking for missing SEQUENCE errors that a session reset will
address.
NFS4ERR_BAD_HIGH_SLOT, NFS4ERR_DEADSESSION, NFS4ERR_SEQ_FALSE_RETRY.

A reset of the session is currently our only response to a SEQUENCE operation
error. Don't reset the session on errors where a new session won't help.

Don't reset the session on errors where a new session won't help.

[nfs41: nfs4_async_handle_error update error checking]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: trigger the state manager for session reset]
    Replace session state bit with nfs_client state bit.  Set the
    NFS4CLNT_SESSION_SETUP bit upon a session related error in the sync/async
    error handlers.
[nfs41: _nfs4_async_handle_error fix session reset error list]
Sequence operation errors that session reset could help.
NFS4ERR_BADSESSION
NFS4ERR_BADSLOT
NFS4ERR_BAD_HIGH_SLOT
NFS4ERR_DEADSESSION
NFS4ERR_CONN_NOT_BOUND_TO_SESSION
NFS4ERR_SEQ_FALSE_RETRY
NFS4ERR_SEQ_MISORDERED

Sequence operation errors that a session reset would not help

NFS4ERR_BADXDR
NFS4ERR_DELAY
NFS4ERR_REP_TOO_BIG
NFS4ERR_REP_TOO_BIG_TO_CACHE
NFS4ERR_REQ_TOO_BIG
NFS4ERR_RETRY_UNCACHED_REP
NFS4ERR_SEQUENCE_POS
NFS4ERR_TOO_MANY_OPS

Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41 nfs4_handle_exception fix session reset error list]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[moved nfs41_sequece_call_done code to nfs41: sequence operation]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 32 ++++++++++++++++++++++++++++++++
 1 file changed, 32 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 2d45606a8085..ed6c8899806d 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -249,7 +249,25 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode,
 			ret = nfs4_wait_clnt_recover(clp);
 			if (ret == 0)
 				exception->retry = 1;
+#if !defined(CONFIG_NFS_V4_1)
 			break;
+#else /* !defined(CONFIG_NFS_V4_1) */
+			if (!nfs4_has_session(server->nfs_client))
+				break;
+			/* FALLTHROUGH */
+		case -NFS4ERR_BADSESSION:
+		case -NFS4ERR_BADSLOT:
+		case -NFS4ERR_BAD_HIGH_SLOT:
+		case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+		case -NFS4ERR_DEADSESSION:
+		case -NFS4ERR_SEQ_FALSE_RETRY:
+		case -NFS4ERR_SEQ_MISORDERED:
+			dprintk("%s ERROR: %d Reset session\n", __func__,
+				errorcode);
+			set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+			exception->retry = 1;
+			/* FALLTHROUGH */
+#endif /* !defined(CONFIG_NFS_V4_1) */
 		case -NFS4ERR_FILE_OPEN:
 		case -NFS4ERR_GRACE:
 		case -NFS4ERR_DELAY:
@@ -3241,6 +3259,20 @@ _nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 				rpc_wake_up_queued_task(&clp->cl_rpcwaitq, task);
 			task->tk_status = 0;
 			return -EAGAIN;
+#if defined(CONFIG_NFS_V4_1)
+		case -NFS4ERR_BADSESSION:
+		case -NFS4ERR_BADSLOT:
+		case -NFS4ERR_BAD_HIGH_SLOT:
+		case -NFS4ERR_DEADSESSION:
+		case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+		case -NFS4ERR_SEQ_FALSE_RETRY:
+		case -NFS4ERR_SEQ_MISORDERED:
+			dprintk("%s ERROR %d, Reset session\n", __func__,
+				task->tk_status);
+			set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+			task->tk_status = 0;
+			return -EAGAIN;
+#endif /* CONFIG_NFS_V4_1 */
 		case -NFS4ERR_DELAY:
 			if (server)
 				nfs_inc_server_stats(server, NFSIOS_DELAY);
-- 
cgit v1.2.3-59-g8ed1b


From b069d94af7785750228287c35f9ce5c463515726 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:43 -0400
Subject: nfs41: schedule async session reset

Define a new session reset state which is set upon a sequence operation error
in both the sync and async error handlers.

Place all new requests and all but the last outstanding rpc on the
slot_tbl_waitq. Spawn the recovery thread when the last slot is free.
Call nfs4_proc_destroy_session, reinitialize the session, call
nfs4_proc_create_session, clear the session reset state, and wake up the next
task on the slot_tbl_waitq.

Return the nfs4_proc_destroy_session status to the session reclaimer and
check for NFS4ERR_BADSESSION and NFS4ERR_DEADSESSION. Other destroy session
errors should be handled in nfs4_proc_destroy_session where the call can
be retried with adjusted arguments.

Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
nfs41: make nfs4_wait_bit_killable public]
    nfs4_wait_bit_killable to be used by NFSv4.1 session recover logic.
Signed-off-by: Rahul Iyer <iyer@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: have create_session work on nfs_client]
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: trigger the state manager for session reset]
    Replace the session reset state with the NFS4CLNT_SESSION_SETUP cl_state.
    Place all rpc tasks to sleep on the slot table waitqueue until the slot
    table is drained, then schedule state recovery and wait for it to complete.
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: remove nfs41_session_recovery [ch]
Replaced by using the nfs4_state_manager.
Signed-off-by: Andy Adamson <andros@netapp.com>
[nfs41: nfs4_wait_bit_killable only used locally]
[nfs41: keep nfs4_wait_bit_killable static]
[nfs41: keep const nfs_server in nfs4_handle_exception]
[nfs41: remove session parameter from nfs4_find_slot]
Signed-off-by: Andy Adamson <andros@netapp.com
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: resset the session from nfs41_setup_sequence]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 34 ++++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index ed6c8899806d..a2855391b079 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -423,6 +423,22 @@ out:
 	return ret_id;
 }
 
+static int nfs4_recover_session(struct nfs4_session *session)
+{
+	struct nfs_client *clp = session->clp;
+	int ret;
+
+	for (;;) {
+		ret = nfs4_wait_clnt_recover(clp);
+		if (ret != 0)
+				return ret;
+		if (!test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state))
+			break;
+		nfs4_schedule_state_manager(clp);
+	}
+	return 0;
+}
+
 static int nfs41_setup_sequence(struct nfs4_session *session,
 				struct nfs4_sequence_args *args,
 				struct nfs4_sequence_res *res,
@@ -431,6 +447,7 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 {
 	struct nfs4_slot *slot;
 	struct nfs4_slot_table *tbl;
+	int status = 0;
 	u8 slotid;
 
 	dprintk("--> %s\n", __func__);
@@ -443,6 +460,23 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 	tbl = &session->fc_slot_table;
 
 	spin_lock(&tbl->slot_tbl_lock);
+	if (test_bit(NFS4CLNT_SESSION_SETUP, &session->clp->cl_state)) {
+		if (tbl->highest_used_slotid != -1) {
+			rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
+			spin_unlock(&tbl->slot_tbl_lock);
+			dprintk("<-- %s: Session reset: draining\n", __func__);
+			return -EAGAIN;
+		}
+
+		/* The slot table is empty; start the reset thread */
+		dprintk("%s Session Reset\n", __func__);
+		spin_unlock(&tbl->slot_tbl_lock);
+		status = nfs4_recover_session(session);
+		if (status)
+			return status;
+		spin_lock(&tbl->slot_tbl_lock);
+	}
+
 	slotid = nfs4_find_slot(tbl, task);
 	if (slotid == NFS4_MAX_SLOT_TABLE) {
 		rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
-- 
cgit v1.2.3-59-g8ed1b


From 29fba38b793225798104a1ac870e6807ebd078dd Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:44 -0400
Subject: nfs41: lease renewal

Send a NFSv4.1 SEQUENCE op rather than RENEW that was deprecated in
minorversion 1.
Use the nfs_client minorversion to select reboot_recover/
network_partition_recovery/state_renewal ops.

Note: we use reclaimer to create the nfs41 session before there are any
cl_superblocks for the nfs_client.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: check for session not minorversion]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[revamped patch for new nfs4_state_manager design]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: obliterate nfs4_state_recovery_ops.renew_lease method]
    moved to nfs4_state_maintenance_ops
[also undid per-minorversion nfs4_state_recovery_ops here]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4_fs.h    |  6 ++++++
 fs/nfs/nfs4proc.c   | 24 ++++++++++++++++++++++++
 fs/nfs/nfs4renewd.c |  4 +++-
 3 files changed, 33 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 7cc1cddfd9ba..ae25cc21fe70 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -180,6 +180,10 @@ struct nfs4_state_recovery_ops {
 	int (*recover_lock)(struct nfs4_state *, struct file_lock *);
 };
 
+struct nfs4_state_maintenance_ops {
+	int (*sched_state_renewal)(struct nfs_client *, struct rpc_cred *);
+};
+
 extern const struct dentry_operations nfs4_dentry_operations;
 extern const struct inode_operations nfs4_dir_inode_operations;
 
@@ -220,6 +224,8 @@ static inline int nfs4_setup_sequence(struct nfs_client *clp,
 }
 #endif /* CONFIG_NFS_V4_1 */
 
+extern struct nfs4_state_maintenance_ops *nfs4_state_renewal_ops[];
+
 extern const u32 nfs4_fattr_bitmap[2];
 extern const u32 nfs4_statfs_bitmap[2];
 extern const u32 nfs4_pathconf_bitmap[2];
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index a2855391b079..101ba6aed101 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3016,6 +3016,9 @@ static void nfs4_renew_done(struct rpc_task *task, void *data)
 	if (time_before(clp->cl_last_renewal,timestamp))
 		clp->cl_last_renewal = timestamp;
 	spin_unlock(&clp->cl_lock);
+	dprintk("%s calling put_rpccred on rpc_cred %p\n", __func__,
+				task->tk_msg.rpc_cred);
+	put_rpccred(task->tk_msg.rpc_cred);
 }
 
 static const struct rpc_call_ops nfs4_renew_ops = {
@@ -4790,6 +4793,27 @@ struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops = {
 	.recover_lock	= nfs4_lock_expired,
 };
 
+struct nfs4_state_maintenance_ops nfs40_state_renewal_ops = {
+	.sched_state_renewal = nfs4_proc_async_renew,
+};
+
+#if defined(CONFIG_NFS_V4_1)
+struct nfs4_state_maintenance_ops nfs41_state_renewal_ops = {
+	.sched_state_renewal = nfs41_proc_async_sequence,
+};
+#endif
+
+/*
+ * Per minor version reboot and network partition recovery ops
+ */
+
+struct nfs4_state_maintenance_ops *nfs4_state_renewal_ops[] = {
+	&nfs40_state_renewal_ops,
+#if defined(CONFIG_NFS_V4_1)
+	&nfs41_state_renewal_ops,
+#endif
+};
+
 static const struct inode_operations nfs4_file_inode_operations = {
 	.permission	= nfs_permission,
 	.getattr	= nfs_getattr,
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index f524e932ff7b..92b43e87267c 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -59,12 +59,14 @@
 void
 nfs4_renew_state(struct work_struct *work)
 {
+	struct nfs4_state_maintenance_ops *ops;
 	struct nfs_client *clp =
 		container_of(work, struct nfs_client, cl_renewd.work);
 	struct rpc_cred *cred;
 	long lease, timeout;
 	unsigned long last, now;
 
+	ops = nfs4_state_renewal_ops[clp->cl_minorversion];
 	dprintk("%s: start\n", __func__);
 	/* Are there any active superblocks? */
 	if (list_empty(&clp->cl_superblocks))
@@ -86,7 +88,7 @@ nfs4_renew_state(struct work_struct *work)
 			nfs_expire_all_delegations(clp);
 		} else {
 			/* Queue an asynchronous RENEW. */
-			nfs4_proc_async_renew(clp, cred);
+			ops->sched_state_renewal(clp, cred);
 			put_rpccred(cred);
 		}
 		timeout = (2 * lease) / 3;
-- 
cgit v1.2.3-59-g8ed1b


From 8e69514f2981d85108c2bd220ff8e188c0c27cdb Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:45 -0400
Subject: nfs41: support minorversion 1 for nfs4_check_lease

[moved nfs4_get_renew_cred related changes to
 "nfs41: introduce get_state_renewal_cred"]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4_fs.h   | 1 +
 fs/nfs/nfs4proc.c  | 2 ++
 fs/nfs/nfs4state.c | 4 +++-
 3 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index ae25cc21fe70..095bf4690e84 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -182,6 +182,7 @@ struct nfs4_state_recovery_ops {
 
 struct nfs4_state_maintenance_ops {
 	int (*sched_state_renewal)(struct nfs_client *, struct rpc_cred *);
+	int (*renew_lease)(struct nfs_client *, struct rpc_cred *);
 };
 
 extern const struct dentry_operations nfs4_dentry_operations;
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 101ba6aed101..da481bda44f7 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4795,11 +4795,13 @@ struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops = {
 
 struct nfs4_state_maintenance_ops nfs40_state_renewal_ops = {
 	.sched_state_renewal = nfs4_proc_async_renew,
+	.renew_lease = nfs4_proc_renew,
 };
 
 #if defined(CONFIG_NFS_V4_1)
 struct nfs4_state_maintenance_ops nfs41_state_renewal_ops = {
 	.sched_state_renewal = nfs41_proc_async_sequence,
+	.renew_lease = nfs4_proc_sequence,
 };
 #endif
 
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 8b7f007adc42..dc5599c7f4d9 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1083,6 +1083,8 @@ restart:
 static int nfs4_check_lease(struct nfs_client *clp)
 {
 	struct rpc_cred *cred;
+	struct nfs4_state_maintenance_ops *ops =
+		nfs4_state_renewal_ops[clp->cl_minorversion];
 	int status = -NFS4ERR_EXPIRED;
 
 	/* Is the client already known to have an expired lease? */
@@ -1094,7 +1096,7 @@ static int nfs4_check_lease(struct nfs_client *clp)
 		if (cred == NULL)
 			goto out;
 	}
-	status = nfs4_proc_renew(clp, cred);
+	status = ops->renew_lease(clp, cred);
 	put_rpccred(cred);
 out:
 	nfs4_recovery_handle_error(clp, status);
-- 
cgit v1.2.3-59-g8ed1b


From a7b721037f898b29a8083da59b1dccd3da385b07 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:46 -0400
Subject: nfs41: introduce get_state_renewal_cred

Use the machine cred for sending SEQUENCE to renew
the client's lease.

[revamp patch for new state management design starting 2.6.29]
[nfs41: support minorversion 1 for nfs4_check_lease]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: get cred in exchange_id when cred arg is NULL]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: use cl_machined_cred instead of cl_ex_cred]
    Since EXCHANGE_ID insists on using the machine credential, cl_ex_cred is
    not needed. nfs4_proc_exchange_id() is only called if the machine credential
    is available. Remove the credential logic from nfs4_proc_exchange_id.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4_fs.h    |  5 +++++
 fs/nfs/nfs4proc.c   |  3 +++
 fs/nfs/nfs4renewd.c |  2 +-
 fs/nfs/nfs4state.c  | 18 +++++-------------
 4 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 095bf4690e84..2b141c5758ec 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -182,6 +182,7 @@ struct nfs4_state_recovery_ops {
 
 struct nfs4_state_maintenance_ops {
 	int (*sched_state_renewal)(struct nfs_client *, struct rpc_cred *);
+	struct rpc_cred * (*get_state_renewal_cred_locked)(struct nfs_client *);
 	int (*renew_lease)(struct nfs_client *, struct rpc_cred *);
 };
 
@@ -240,7 +241,11 @@ extern void nfs4_kill_renewd(struct nfs_client *);
 extern void nfs4_renew_state(struct work_struct *);
 
 /* nfs4state.c */
+struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp);
 struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp);
+#if defined(CONFIG_NFS_V4_1)
+struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp);
+#endif /* CONFIG_NFS_V4_1 */
 
 extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *);
 extern void nfs4_put_state_owner(struct nfs4_state_owner *);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index da481bda44f7..f862a66e5fb9 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4236,6 +4236,7 @@ static int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 
 	dprintk("--> %s\n", __func__);
 	BUG_ON(clp == NULL);
+
 	p = (u32 *)verifier.data;
 	*p++ = htonl((u32)clp->cl_boot_time.tv_sec);
 	*p = htonl((u32)clp->cl_boot_time.tv_nsec);
@@ -4795,12 +4796,14 @@ struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops = {
 
 struct nfs4_state_maintenance_ops nfs40_state_renewal_ops = {
 	.sched_state_renewal = nfs4_proc_async_renew,
+	.get_state_renewal_cred_locked = nfs4_get_renew_cred_locked,
 	.renew_lease = nfs4_proc_renew,
 };
 
 #if defined(CONFIG_NFS_V4_1)
 struct nfs4_state_maintenance_ops nfs41_state_renewal_ops = {
 	.sched_state_renewal = nfs41_proc_async_sequence,
+	.get_state_renewal_cred_locked = nfs4_get_machine_cred_locked,
 	.renew_lease = nfs4_proc_sequence,
 };
 #endif
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index 92b43e87267c..e27c6cef18f2 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -78,7 +78,7 @@ nfs4_renew_state(struct work_struct *work)
 	timeout = (2 * lease) / 3 + (long)last - (long)now;
 	/* Are we close to a lease timeout? */
 	if (time_after(now, last + lease/3)) {
-		cred = nfs4_get_renew_cred_locked(clp);
+		cred = ops->get_state_renewal_cred_locked(clp);
 		spin_unlock(&clp->cl_lock);
 		if (cred == NULL) {
 			if (list_empty(&clp->cl_delegations)) {
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index dc5599c7f4d9..da940abcfaa5 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -77,7 +77,7 @@ static int nfs4_init_client(struct nfs_client *clp, struct rpc_cred *cred)
 	return status;
 }
 
-static struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp)
+struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp)
 {
 	struct rpc_cred *cred = NULL;
 
@@ -114,17 +114,7 @@ struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp)
 	return cred;
 }
 
-static struct rpc_cred *nfs4_get_renew_cred(struct nfs_client *clp)
-{
-	struct rpc_cred *cred;
-
-	spin_lock(&clp->cl_lock);
-	cred = nfs4_get_renew_cred_locked(clp);
-	spin_unlock(&clp->cl_lock);
-	return cred;
-}
-
-static struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp)
+struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp)
 {
 	struct nfs4_state_owner *sp;
 	struct rb_node *pos;
@@ -1090,7 +1080,9 @@ static int nfs4_check_lease(struct nfs_client *clp)
 	/* Is the client already known to have an expired lease? */
 	if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state))
 		return 0;
-	cred = nfs4_get_renew_cred(clp);
+	spin_lock(&clp->cl_lock);
+	cred = ops->get_state_renewal_cred_locked(clp);
+	spin_unlock(&clp->cl_lock);
 	if (cred == NULL) {
 		cred = nfs4_get_setclientid_cred(clp);
 		if (cred == NULL)
-- 
cgit v1.2.3-59-g8ed1b


From 591d71cbde186cc498c0d9718dc17f2fadf7c643 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:47 -0400
Subject: nfs41: establish sessions-based clientid

nfsv4.1 clientid is established via EXCHANGE_ID rather than
SETCLIENTID{,_CONFIRM}

This is implemented using a new establish_clid method in
nfs4_state_recovery_ops.

nfs41: establish clientid via exchange id only if cred != NULL

>From 2.6.26 reclaimer() uses machine cred for setting up the client id
therefore it is never expected to be NULL.

Signed-off-by: Rahul Iyer <iyer@netapp.com>
[removed dprintk]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: lease renewal]
[revamped patch for new nfs4_state_manager design]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4_fs.h   |  6 ++++--
 fs/nfs/nfs4proc.c  | 40 ++++++++++++++++++++++++++++++++++++++--
 fs/nfs/nfs4state.c | 12 ++++++++----
 3 files changed, 50 insertions(+), 8 deletions(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 2b141c5758ec..df9e36319499 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -178,6 +178,7 @@ struct nfs4_state_recovery_ops {
 	int state_flag_bit;
 	int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *);
 	int (*recover_lock)(struct nfs4_state *, struct file_lock *);
+	int (*establish_clid)(struct nfs_client *, struct rpc_cred *);
 };
 
 struct nfs4_state_maintenance_ops {
@@ -200,6 +201,7 @@ extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struc
 extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_proc_async_renew(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_proc_renew(struct nfs_client *, struct rpc_cred *);
+extern int nfs4_init_clientid(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait);
 extern struct dentry *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *);
 extern int nfs4_open_revalidate(struct inode *, struct dentry *, int, struct nameidata *);
@@ -207,8 +209,8 @@ extern int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fh
 extern int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
 		struct nfs4_fs_locations *fs_locations, struct page *page);
 
-extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops;
-extern struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops;
+extern struct nfs4_state_recovery_ops *nfs4_reboot_recovery_ops[];
+extern struct nfs4_state_recovery_ops *nfs4_nograce_recovery_ops[];
 #if defined(CONFIG_NFS_V4_1)
 extern int nfs4_setup_sequence(struct nfs_client *clp,
 		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index f862a66e5fb9..dd46339095ce 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4780,19 +4780,41 @@ static int nfs41_proc_async_sequence(struct nfs_client *clp,
 
 #endif /* CONFIG_NFS_V4_1 */
 
-struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops = {
+struct nfs4_state_recovery_ops nfs40_reboot_recovery_ops = {
 	.owner_flag_bit = NFS_OWNER_RECLAIM_REBOOT,
 	.state_flag_bit	= NFS_STATE_RECLAIM_REBOOT,
 	.recover_open	= nfs4_open_reclaim,
 	.recover_lock	= nfs4_lock_reclaim,
+	.establish_clid = nfs4_init_clientid,
 };
 
-struct nfs4_state_recovery_ops nfs4_nograce_recovery_ops = {
+#if defined(CONFIG_NFS_V4_1)
+struct nfs4_state_recovery_ops nfs41_reboot_recovery_ops = {
+	.owner_flag_bit = NFS_OWNER_RECLAIM_REBOOT,
+	.state_flag_bit	= NFS_STATE_RECLAIM_REBOOT,
+	.recover_open	= nfs4_open_reclaim,
+	.recover_lock	= nfs4_lock_reclaim,
+	.establish_clid = nfs4_proc_exchange_id,
+};
+#endif /* CONFIG_NFS_V4_1 */
+
+struct nfs4_state_recovery_ops nfs40_nograce_recovery_ops = {
+	.owner_flag_bit = NFS_OWNER_RECLAIM_NOGRACE,
+	.state_flag_bit	= NFS_STATE_RECLAIM_NOGRACE,
+	.recover_open	= nfs4_open_expired,
+	.recover_lock	= nfs4_lock_expired,
+	.establish_clid = nfs4_init_clientid,
+};
+
+#if defined(CONFIG_NFS_V4_1)
+struct nfs4_state_recovery_ops nfs41_nograce_recovery_ops = {
 	.owner_flag_bit = NFS_OWNER_RECLAIM_NOGRACE,
 	.state_flag_bit	= NFS_STATE_RECLAIM_NOGRACE,
 	.recover_open	= nfs4_open_expired,
 	.recover_lock	= nfs4_lock_expired,
+	.establish_clid = nfs4_proc_exchange_id,
 };
+#endif /* CONFIG_NFS_V4_1 */
 
 struct nfs4_state_maintenance_ops nfs40_state_renewal_ops = {
 	.sched_state_renewal = nfs4_proc_async_renew,
@@ -4812,6 +4834,20 @@ struct nfs4_state_maintenance_ops nfs41_state_renewal_ops = {
  * Per minor version reboot and network partition recovery ops
  */
 
+struct nfs4_state_recovery_ops *nfs4_reboot_recovery_ops[] = {
+	&nfs40_reboot_recovery_ops,
+#if defined(CONFIG_NFS_V4_1)
+	&nfs41_reboot_recovery_ops,
+#endif
+};
+
+struct nfs4_state_recovery_ops *nfs4_nograce_recovery_ops[] = {
+	&nfs40_nograce_recovery_ops,
+#if defined(CONFIG_NFS_V4_1)
+	&nfs41_nograce_recovery_ops,
+#endif
+};
+
 struct nfs4_state_maintenance_ops *nfs4_state_renewal_ops[] = {
 	&nfs40_state_renewal_ops,
 #if defined(CONFIG_NFS_V4_1)
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index da940abcfaa5..e17bd4412174 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -60,7 +60,7 @@ const nfs4_stateid zero_stateid;
 
 static LIST_HEAD(nfs4_clientid_list);
 
-static int nfs4_init_client(struct nfs_client *clp, struct rpc_cred *cred)
+int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	unsigned short port;
 	int status;
@@ -1098,11 +1098,13 @@ out:
 static int nfs4_reclaim_lease(struct nfs_client *clp)
 {
 	struct rpc_cred *cred;
+	struct nfs4_state_recovery_ops *ops =
+		nfs4_reboot_recovery_ops[clp->cl_minorversion];
 	int status = -ENOENT;
 
 	cred = nfs4_get_setclientid_cred(clp);
 	if (cred != NULL) {
-		status = nfs4_init_client(clp, cred);
+		status = ops->establish_clid(clp, cred);
 		put_rpccred(cred);
 		/* Handle case where the user hasn't set up machine creds */
 		if (status == -EACCES && cred == clp->cl_machine_cred) {
@@ -1208,7 +1210,8 @@ static void nfs4_state_manager(struct nfs_client *clp)
 		}
 		/* First recover reboot state... */
 		if (test_and_clear_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state)) {
-			status = nfs4_do_reclaim(clp, &nfs4_reboot_recovery_ops);
+			status = nfs4_do_reclaim(clp,
+				nfs4_reboot_recovery_ops[clp->cl_minorversion]);
 			if (status == -NFS4ERR_STALE_CLIENTID)
 				continue;
 			if (test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state))
@@ -1219,7 +1222,8 @@ static void nfs4_state_manager(struct nfs_client *clp)
 
 		/* Now recover expired state... */
 		if (test_and_clear_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state)) {
-			status = nfs4_do_reclaim(clp, &nfs4_nograce_recovery_ops);
+			status = nfs4_do_reclaim(clp,
+				nfs4_nograce_recovery_ops[clp->cl_minorversion]);
 			if (status < 0) {
 				set_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state);
 				if (status == -NFS4ERR_STALE_CLIENTID)
-- 
cgit v1.2.3-59-g8ed1b


From 90a16617ee6a052c3a1aac00eb67136324cf4dd0 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:48 -0400
Subject: nfs41: add a get_clid_cred function to nfs4_state_recovery_ops

EXCHANGE_ID has different credential requirements than SETCLIENTID.
Prepare for a separate credential function.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4_fs.h   | 1 +
 fs/nfs/nfs4proc.c  | 2 ++
 fs/nfs/nfs4state.c | 2 +-
 3 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index df9e36319499..1da3e354de22 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -179,6 +179,7 @@ struct nfs4_state_recovery_ops {
 	int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *);
 	int (*recover_lock)(struct nfs4_state *, struct file_lock *);
 	int (*establish_clid)(struct nfs_client *, struct rpc_cred *);
+	struct rpc_cred * (*get_clid_cred)(struct nfs_client *);
 };
 
 struct nfs4_state_maintenance_ops {
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index dd46339095ce..b4e5442efa6c 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4786,6 +4786,7 @@ struct nfs4_state_recovery_ops nfs40_reboot_recovery_ops = {
 	.recover_open	= nfs4_open_reclaim,
 	.recover_lock	= nfs4_lock_reclaim,
 	.establish_clid = nfs4_init_clientid,
+	.get_clid_cred	= nfs4_get_setclientid_cred,
 };
 
 #if defined(CONFIG_NFS_V4_1)
@@ -4804,6 +4805,7 @@ struct nfs4_state_recovery_ops nfs40_nograce_recovery_ops = {
 	.recover_open	= nfs4_open_expired,
 	.recover_lock	= nfs4_lock_expired,
 	.establish_clid = nfs4_init_clientid,
+	.get_clid_cred	= nfs4_get_setclientid_cred,
 };
 
 #if defined(CONFIG_NFS_V4_1)
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index e17bd4412174..90e56072844b 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1102,7 +1102,7 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 		nfs4_reboot_recovery_ops[clp->cl_minorversion];
 	int status = -ENOENT;
 
-	cred = nfs4_get_setclientid_cred(clp);
+	cred = ops->get_clid_cred(clp);
 	if (cred != NULL) {
 		status = ops->establish_clid(clp, cred);
 		put_rpccred(cred);
-- 
cgit v1.2.3-59-g8ed1b


From b4b82607ffcf09b57301846d154f2c09c0b807c0 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:22:49 -0400
Subject: nfs41: get_clid_cred for EXCHANGE_ID

Unlike SETCLIENTID, EXCHANGE_ID requires a machine credential. Do not search
for credentials other than the machine credential.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4_fs.h   |  1 +
 fs/nfs/nfs4proc.c  |  2 ++
 fs/nfs/nfs4state.c | 14 ++++++++++++++
 3 files changed, 17 insertions(+)

diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 1da3e354de22..61bc3a32e1e2 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -248,6 +248,7 @@ struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp);
 struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp);
 #if defined(CONFIG_NFS_V4_1)
 struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp);
+struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp);
 #endif /* CONFIG_NFS_V4_1 */
 
 extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index b4e5442efa6c..64d611658d1f 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4796,6 +4796,7 @@ struct nfs4_state_recovery_ops nfs41_reboot_recovery_ops = {
 	.recover_open	= nfs4_open_reclaim,
 	.recover_lock	= nfs4_lock_reclaim,
 	.establish_clid = nfs4_proc_exchange_id,
+	.get_clid_cred	= nfs4_get_exchange_id_cred,
 };
 #endif /* CONFIG_NFS_V4_1 */
 
@@ -4815,6 +4816,7 @@ struct nfs4_state_recovery_ops nfs41_nograce_recovery_ops = {
 	.recover_open	= nfs4_open_expired,
 	.recover_lock	= nfs4_lock_expired,
 	.establish_clid = nfs4_proc_exchange_id,
+	.get_clid_cred	= nfs4_get_exchange_id_cred,
 };
 #endif /* CONFIG_NFS_V4_1 */
 
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 90e56072844b..7dc971335ec2 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -114,6 +114,20 @@ struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp)
 	return cred;
 }
 
+#if defined(CONFIG_NFS_V4_1)
+
+struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp)
+{
+	struct rpc_cred *cred;
+
+	spin_lock(&clp->cl_lock);
+	cred = nfs4_get_machine_cred_locked(clp);
+	spin_unlock(&clp->cl_lock);
+	return cred;
+}
+
+#endif /* CONFIG_NFS_V4_1 */
+
 struct rpc_cred *nfs4_get_setclientid_cred(struct nfs_client *clp)
 {
 	struct nfs4_state_owner *sp;
-- 
cgit v1.2.3-59-g8ed1b


From 008f55d0e019943323c20a03493a2ba5672a4cc8 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:50 -0400
Subject: nfs41: recover lease in _nfs4_lookup_root

This creates the nfsv4.1 session on mount.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c   | 16 ++++++++++++++++
 fs/nfs/internal.h |  1 +
 fs/nfs/nfs4proc.c |  9 ++++++++-
 3 files changed, 25 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index d28a987f569e..3e232bf56dfb 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -527,6 +527,22 @@ void nfs_mark_client_ready(struct nfs_client *clp, int state)
 	wake_up_all(&nfs_client_active_wq);
 }
 
+/*
+ * With sessions, the client is not marked ready until after a
+ * successful EXCHANGE_ID and CREATE_SESSION.
+ *
+ * Map errors cl_cons_state errors to EPROTONOSUPPORT to indicate
+ * other versions of NFS can be tried.
+ */
+int nfs4_check_client_ready(struct nfs_client *clp)
+{
+	if (!nfs4_has_session(clp))
+		return 0;
+	if (clp->cl_cons_state < NFS_CS_READY)
+		return -EPROTONOSUPPORT;
+	return 0;
+}
+
 /*
  * Initialise the timeout values for a connection
  */
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 78508f29c03a..acee3274d275 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -114,6 +114,7 @@ extern struct nfs_server *nfs_clone_server(struct nfs_server *,
 					   struct nfs_fh *,
 					   struct nfs_fattr *);
 extern void nfs_mark_client_ready(struct nfs_client *clp, int state);
+extern int nfs4_check_client_ready(struct nfs_client *clp);
 #ifdef CONFIG_PROC_FS
 extern int __init nfs_fs_proc_init(void);
 extern void nfs_fs_proc_exit(void);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 64d611658d1f..31ce758d53d5 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2024,8 +2024,15 @@ static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
 		.rpc_argp = &args,
 		.rpc_resp = &res,
 	};
+	int status;
+
 	nfs_fattr_init(info->fattr);
-	return nfs4_call_sync(server, &msg, &args, &res, 0);
+	status = nfs4_recover_expired_lease(server);
+	if (!status)
+		status = nfs4_check_client_ready(server->nfs_client);
+	if (!status)
+		status = nfs4_call_sync(server, &msg, &args, &res, 0);
+	return status;
 }
 
 static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
-- 
cgit v1.2.3-59-g8ed1b


From 78722e9c9208a312695178f5331511badd190598 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Tue, 16 Jun 2009 14:43:15 -0400
Subject: nfs41: only retry EXCHANGE_ID on recoverable errors

Stops an infinite loop of EXCHANGE_ID.

Signed-off-by: Andy Adamson <andros@netapp.com>
[fixed checkpatch warnings]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4state.c | 23 ++++++++++++++++++++++-
 1 file changed, 22 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 7dc971335ec2..2cc0aca96eb2 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1183,6 +1183,27 @@ static int nfs4_reset_session(struct nfs_client *clp) { return 0; }
 static int nfs4_initialize_session(struct nfs_client *clp) { return 0; }
 #endif /* CONFIG_NFS_V4_1 */
 
+/* Set NFS4CLNT_LEASE_EXPIRED for all v4.0 errors and for recoverable errors
+ * on EXCHANGE_ID for v4.1
+ */
+static void nfs4_set_lease_expired(struct nfs_client *clp, int status)
+{
+	if (nfs4_has_session(clp)) {
+		switch (status) {
+		case -NFS4ERR_DELAY:
+		case -NFS4ERR_CLID_INUSE:
+		case -EAGAIN:
+			break;
+
+		case -NFS4ERR_NOT_SAME: /* FixMe: implement recovery
+					 * in nfs4_exchange_id */
+		default:
+			return;
+		}
+	}
+	set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+}
+
 static void nfs4_state_manager(struct nfs_client *clp)
 {
 	int status = 0;
@@ -1193,7 +1214,7 @@ static void nfs4_state_manager(struct nfs_client *clp)
 			/* We're going to have to re-establish a clientid */
 			status = nfs4_reclaim_lease(clp);
 			if (status) {
-				set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+				nfs4_set_lease_expired(clp, status);
 				if (status == -EAGAIN)
 					continue;
 				if (clp->cl_cons_state ==
-- 
cgit v1.2.3-59-g8ed1b


From 34dc1ad752ad3f55b2a6e6cd8cfcf3504682fec7 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:52 -0400
Subject: nfs41: increment_{open,lock}_seqid

Unlike minorversion0, in nfsv4.1 the open and lock seqids need
not be incremented by the client and should always be set to zero.

This is implemented using a new nfs_rpc_ops methods -
increment_open_seqid and increment_lock_seqid

Signed-off-by: Rahul Iyer <iyer@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: check for session not minorversion]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c  |  2 ++
 fs/nfs/nfs4state.c | 12 +++++++-----
 2 files changed, 9 insertions(+), 5 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 31ce758d53d5..f1f087b483ce 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -1808,6 +1808,8 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 	calldata->state = state;
 	calldata->arg.fh = NFS_FH(state->inode);
 	calldata->arg.stateid = &state->open_stateid;
+	if (nfs4_has_session(server->nfs_client))
+		memset(calldata->arg.stateid->data, 0, 4);    /* clear seqid */
 	/* Serialization for the sequence id */
 	calldata->arg.seqid = nfs_alloc_seqid(&state->owner->so_seqid);
 	if (calldata->arg.seqid == NULL)
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 2cc0aca96eb2..2cfca9929c9a 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -742,12 +742,14 @@ static void nfs_increment_seqid(int status, struct nfs_seqid *seqid)
 
 void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid)
 {
-	if (status == -NFS4ERR_BAD_SEQID) {
-		struct nfs4_state_owner *sp = container_of(seqid->sequence,
-				struct nfs4_state_owner, so_seqid);
+	struct nfs4_state_owner *sp = container_of(seqid->sequence,
+					struct nfs4_state_owner, so_seqid);
+	struct nfs_server *server = sp->so_server;
+
+	if (status == -NFS4ERR_BAD_SEQID)
 		nfs4_drop_state_owner(sp);
-	}
-	nfs_increment_seqid(status, seqid);
+	if (!nfs4_has_session(server->nfs_client))
+		nfs_increment_seqid(status, seqid);
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 18dca02aeb3c49dfce87c76be643b139d05cf647 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:22:53 -0400
Subject: nfs41: Add ability to read RPC call direction on TCP stream.

NFSv4.1 callbacks can arrive over an existing connection. This patch adds
the logic to read the RPC call direction (call or reply). It does this by
updating the state machine to look for the call direction invoking
xs_tcp_read_calldir(...) after reading the XID.

[nfs41: Keep track of RPC call/reply direction with a flag]

As per 11/14/08 review of RFC 53/85.

Add a new flag to track whether the incoming message is an RPC call or an
RPC reply.  TCP_RPC_REPLY is set in the 'struct sock_xprt' tcp_flags in
xs_tcp_read_calldir() if the message is an RPC reply sent on the forechannel.
It is cleared if the message is an RPC request sent on the back channel.

Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 net/sunrpc/xprtsock.c | 52 ++++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 49 insertions(+), 3 deletions(-)

diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index e18596146013..8975c10591c3 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -270,6 +270,12 @@ struct sock_xprt {
 #define TCP_RCV_COPY_FRAGHDR	(1UL << 1)
 #define TCP_RCV_COPY_XID	(1UL << 2)
 #define TCP_RCV_COPY_DATA	(1UL << 3)
+#define TCP_RCV_COPY_CALLDIR	(1UL << 4)
+
+/*
+ * TCP RPC flags
+ */
+#define TCP_RPC_REPLY		(1UL << 5)
 
 static inline struct sockaddr *xs_addr(struct rpc_xprt *xprt)
 {
@@ -956,7 +962,7 @@ static inline void xs_tcp_read_fraghdr(struct rpc_xprt *xprt, struct xdr_skb_rea
 	transport->tcp_offset = 0;
 
 	/* Sanity check of the record length */
-	if (unlikely(transport->tcp_reclen < 4)) {
+	if (unlikely(transport->tcp_reclen < 8)) {
 		dprintk("RPC:       invalid TCP record fragment length\n");
 		xprt_force_disconnect(xprt);
 		return;
@@ -991,13 +997,48 @@ static inline void xs_tcp_read_xid(struct sock_xprt *transport, struct xdr_skb_r
 	if (used != len)
 		return;
 	transport->tcp_flags &= ~TCP_RCV_COPY_XID;
-	transport->tcp_flags |= TCP_RCV_COPY_DATA;
+	transport->tcp_flags |= TCP_RCV_COPY_CALLDIR;
 	transport->tcp_copied = 4;
-	dprintk("RPC:       reading reply for XID %08x\n",
+	dprintk("RPC:       reading %s XID %08x\n",
+			(transport->tcp_flags & TCP_RPC_REPLY) ? "reply for"
+							      : "request with",
 			ntohl(transport->tcp_xid));
 	xs_tcp_check_fraghdr(transport);
 }
 
+static inline void xs_tcp_read_calldir(struct sock_xprt *transport,
+				       struct xdr_skb_reader *desc)
+{
+	size_t len, used;
+	u32 offset;
+	__be32	calldir;
+
+	/*
+	 * We want transport->tcp_offset to be 8 at the end of this routine
+	 * (4 bytes for the xid and 4 bytes for the call/reply flag).
+	 * When this function is called for the first time,
+	 * transport->tcp_offset is 4 (after having already read the xid).
+	 */
+	offset = transport->tcp_offset - sizeof(transport->tcp_xid);
+	len = sizeof(calldir) - offset;
+	dprintk("RPC:       reading CALL/REPLY flag (%Zu bytes)\n", len);
+	used = xdr_skb_read_bits(desc, &calldir, len);
+	transport->tcp_offset += used;
+	if (used != len)
+		return;
+	transport->tcp_flags &= ~TCP_RCV_COPY_CALLDIR;
+	transport->tcp_flags |= TCP_RCV_COPY_DATA;
+	transport->tcp_copied += 4;
+	if (ntohl(calldir) == RPC_REPLY)
+		transport->tcp_flags |= TCP_RPC_REPLY;
+	else
+		transport->tcp_flags &= ~TCP_RPC_REPLY;
+	dprintk("RPC:       reading %s CALL/REPLY flag %08x\n",
+			(transport->tcp_flags & TCP_RPC_REPLY) ?
+				"reply for" : "request with", calldir);
+	xs_tcp_check_fraghdr(transport);
+}
+
 static inline void xs_tcp_read_request(struct rpc_xprt *xprt, struct xdr_skb_reader *desc)
 {
 	struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
@@ -1114,6 +1155,11 @@ static int xs_tcp_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb, uns
 			xs_tcp_read_xid(transport, &desc);
 			continue;
 		}
+		/* Read in the call/reply flag */
+		if (transport->tcp_flags & TCP_RCV_COPY_CALLDIR) {
+			xs_tcp_read_calldir(transport, &desc);
+			continue;
+		}
 		/* Read in the request data */
 		if (transport->tcp_flags & TCP_RCV_COPY_DATA) {
 			xs_tcp_read_request(xprt, &desc);
-- 
cgit v1.2.3-59-g8ed1b


From f4a2e418bfd03a1f25f515e8a92ecd584d96cfc1 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:22:54 -0400
Subject: nfs41: Process the RPC call direction

Reading and storing the RPC direction is a three step process.

1. xs_tcp_read_calldir() reads the RPC direction, but it will not store it
in the XDR buffer since the 'struct rpc_rqst' is not yet available.

2. The 'struct rpc_rqst' is obtained during the TCP_RCV_COPY_DATA state.
This state need not necessarily be preceeded by the TCP_RCV_READ_CALLDIR.
For example, we may be reading a continuation packet to a large reply.
Therefore, we can't simply obtain the 'struct rpc_rqst' during the
TCP_RCV_READ_CALLDIR state and assume it's available during TCP_RCV_COPY_DATA.

This patch adds a new TCP_RCV_READ_CALLDIR flag to indicate the need to
read the RPC direction.  It then uses TCP_RCV_COPY_CALLDIR to indicate the
RPC direction needs to be saved after the 'struct rpc_rqst' has been allocated.

3. The 'struct rpc_rqst' is obtained by the xs_tcp_read_data() helper
functions.  xs_tcp_read_common() then saves the RPC direction in the XDR
buffer if TCP_RCV_COPY_CALLDIR is set.  This will happen when we're reading
the data immediately after the direction was read.  xs_tcp_read_common()
then clears this flag.

[was nfs41: Skip past the RPC call direction]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: sunrpc: Add RPC direction back into the XDR buffer]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: sunrpc: Don't skip past the RPC call direction]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 net/sunrpc/clnt.c     |  5 +++--
 net/sunrpc/xprtsock.c | 31 +++++++++++++++++++++++++------
 2 files changed, 28 insertions(+), 8 deletions(-)

diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index d00e8135f866..aca3ab6fc140 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -1390,13 +1390,14 @@ rpc_verify_header(struct rpc_task *task)
 	}
 	if ((len -= 3) < 0)
 		goto out_overflow;
-	p += 1;	/* skip XID */
 
+	p += 1; /* skip XID */
 	if ((n = ntohl(*p++)) != RPC_REPLY) {
 		dprintk("RPC: %5u %s: not an RPC reply: %x\n",
-				task->tk_pid, __func__, n);
+			task->tk_pid, __func__, n);
 		goto out_garbage;
 	}
+
 	if ((n = ntohl(*p++)) != RPC_MSG_ACCEPTED) {
 		if (--len < 0)
 			goto out_overflow;
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 8975c10591c3..a48df1449ece 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -270,12 +270,13 @@ struct sock_xprt {
 #define TCP_RCV_COPY_FRAGHDR	(1UL << 1)
 #define TCP_RCV_COPY_XID	(1UL << 2)
 #define TCP_RCV_COPY_DATA	(1UL << 3)
-#define TCP_RCV_COPY_CALLDIR	(1UL << 4)
+#define TCP_RCV_READ_CALLDIR	(1UL << 4)
+#define TCP_RCV_COPY_CALLDIR	(1UL << 5)
 
 /*
  * TCP RPC flags
  */
-#define TCP_RPC_REPLY		(1UL << 5)
+#define TCP_RPC_REPLY		(1UL << 6)
 
 static inline struct sockaddr *xs_addr(struct rpc_xprt *xprt)
 {
@@ -997,7 +998,7 @@ static inline void xs_tcp_read_xid(struct sock_xprt *transport, struct xdr_skb_r
 	if (used != len)
 		return;
 	transport->tcp_flags &= ~TCP_RCV_COPY_XID;
-	transport->tcp_flags |= TCP_RCV_COPY_CALLDIR;
+	transport->tcp_flags |= TCP_RCV_READ_CALLDIR;
 	transport->tcp_copied = 4;
 	dprintk("RPC:       reading %s XID %08x\n",
 			(transport->tcp_flags & TCP_RPC_REPLY) ? "reply for"
@@ -1026,9 +1027,13 @@ static inline void xs_tcp_read_calldir(struct sock_xprt *transport,
 	transport->tcp_offset += used;
 	if (used != len)
 		return;
-	transport->tcp_flags &= ~TCP_RCV_COPY_CALLDIR;
+	transport->tcp_flags &= ~TCP_RCV_READ_CALLDIR;
+	transport->tcp_flags |= TCP_RCV_COPY_CALLDIR;
 	transport->tcp_flags |= TCP_RCV_COPY_DATA;
-	transport->tcp_copied += 4;
+	/*
+	 * We don't yet have the XDR buffer, so we will write the calldir
+	 * out after we get the buffer from the 'struct rpc_rqst'
+	 */
 	if (ntohl(calldir) == RPC_REPLY)
 		transport->tcp_flags |= TCP_RPC_REPLY;
 	else
@@ -1059,6 +1064,20 @@ static inline void xs_tcp_read_request(struct rpc_xprt *xprt, struct xdr_skb_rea
 	}
 
 	rcvbuf = &req->rq_private_buf;
+
+	if (transport->tcp_flags & TCP_RCV_COPY_CALLDIR) {
+		/*
+		 * Save the RPC direction in the XDR buffer
+		 */
+		__be32	calldir = transport->tcp_flags & TCP_RPC_REPLY ?
+					htonl(RPC_REPLY) : 0;
+
+		memcpy(rcvbuf->head[0].iov_base + transport->tcp_copied,
+			&calldir, sizeof(calldir));
+		transport->tcp_copied += sizeof(calldir);
+		transport->tcp_flags &= ~TCP_RCV_COPY_CALLDIR;
+	}
+
 	len = desc->count;
 	if (len > transport->tcp_reclen - transport->tcp_offset) {
 		struct xdr_skb_reader my_desc;
@@ -1156,7 +1175,7 @@ static int xs_tcp_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb, uns
 			continue;
 		}
 		/* Read in the call/reply flag */
-		if (transport->tcp_flags & TCP_RCV_COPY_CALLDIR) {
+		if (transport->tcp_flags & TCP_RCV_READ_CALLDIR) {
 			xs_tcp_read_calldir(transport, &desc);
 			continue;
 		}
-- 
cgit v1.2.3-59-g8ed1b


From 9bdaa86d2a4cbf9a71de3048c0c0a874e7ebc8ad Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:55 -0400
Subject: nfs41: Refactor nfs4_{init,destroy}_callback for nfs4.0

Refactor-out code to bring the callback service up and down.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/client.c | 46 +++++++++++++++++++++++++++++++++++++---------
 1 file changed, 37 insertions(+), 9 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 3e232bf56dfb..d9657d455730 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -47,6 +47,9 @@
 #include "internal.h"
 #include "fscache.h"
 
+static int nfs4_init_callback(struct nfs_client *);
+static void nfs4_destroy_callback(struct nfs_client *);
+
 #define NFSDBG_FACILITY		NFSDBG_CLIENT
 
 static DEFINE_SPINLOCK(nfs_client_lock);
@@ -121,11 +124,8 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 
 	clp->rpc_ops = cl_init->rpc_ops;
 
-	if (cl_init->rpc_ops->version == 4) {
-		if (nfs_callback_up() < 0)
-			goto error_2;
-		__set_bit(NFS_CS_CALLBACK, &clp->cl_res_state);
-	}
+	if (nfs4_init_callback(clp) < 0)
+		goto error_2;
 
 	atomic_set(&clp->cl_count, 1);
 	clp->cl_cons_state = NFS_CS_INITING;
@@ -162,8 +162,7 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 	return clp;
 
 error_3:
-	if (__test_and_clear_bit(NFS_CS_CALLBACK, &clp->cl_res_state))
-		nfs_callback_down();
+	nfs4_destroy_callback(clp);
 error_2:
 	kfree(clp);
 error_0:
@@ -183,6 +182,17 @@ static void nfs4_shutdown_client(struct nfs_client *clp)
 #endif
 }
 
+/*
+ * Destroy the NFS4 callback service
+ */
+static void nfs4_destroy_callback(struct nfs_client *clp)
+{
+#ifdef CONFIG_NFS_V4
+	if (__test_and_clear_bit(NFS_CS_CALLBACK, &clp->cl_res_state))
+		nfs_callback_down();
+#endif /* CONFIG_NFS_V4 */
+}
+
 /*
  * Clears/puts all minor version specific parts from an nfs_client struct
  * reverting it to minorversion 0.
@@ -215,8 +225,7 @@ static void nfs_free_client(struct nfs_client *clp)
 	if (!IS_ERR(clp->cl_rpcclient))
 		rpc_shutdown_client(clp->cl_rpcclient);
 
-	if (__test_and_clear_bit(NFS_CS_CALLBACK, &clp->cl_res_state))
-		nfs_callback_down();
+	nfs4_destroy_callback(clp);
 
 	if (clp->cl_machine_cred != NULL)
 		put_rpccred(clp->cl_machine_cred);
@@ -1087,6 +1096,25 @@ error:
 }
 
 #ifdef CONFIG_NFS_V4
+/*
+ * Initialize the NFS4 callback service
+ */
+static int nfs4_init_callback(struct nfs_client *clp)
+{
+	int error;
+
+	if (clp->rpc_ops->version == 4) {
+		error = nfs_callback_up();
+		if (error < 0) {
+			dprintk("%s: failed to start callback. Error = %d\n",
+				__func__, error);
+			return error;
+		}
+		__set_bit(NFS_CS_CALLBACK, &clp->cl_res_state);
+	}
+	return 0;
+}
+
 /*
  * Initialize the minor version specific parts of an NFS4 client record
  */
-- 
cgit v1.2.3-59-g8ed1b


From 7146851376861fe55c7a48ac8fc1354a5fff6cd0 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:22:56 -0400
Subject: nfs41: minorversion support for nfs4_{init,destroy}_callback

move nfs4_init_callback into nfs4_init_client_minor_version
and nfs4_destroy_callback into nfs4_clear_client_minor_version

as these need to happen also when auto-negotiating the minorversion
once the callback service for nfs41 becomes different than for nfs4.0

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Fix checkpatch warning]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Type check arguments of nfs_callback_up]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Backchannel: Remove FIXME comment]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback.c | 65 +++++++++++++++++++++++++++++++++++++++----------------
 fs/nfs/callback.h |  2 +-
 fs/nfs/client.c   | 21 ++++++------------
 3 files changed, 54 insertions(+), 34 deletions(-)

diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index a886e692ddd0..4b1313eda6f5 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -59,7 +59,7 @@ module_param_call(callback_tcpport, param_set_port, param_get_int,
  * This is the callback kernel thread.
  */
 static int
-nfs_callback_svc(void *vrqstp)
+nfs4_callback_svc(void *vrqstp)
 {
 	int err, preverr = 0;
 	struct svc_rqst *rqstp = vrqstp;
@@ -97,20 +97,12 @@ nfs_callback_svc(void *vrqstp)
 }
 
 /*
- * Bring up the callback thread if it is not already up.
+ * Prepare to bring up the NFSv4 callback service
  */
-int nfs_callback_up(void)
+struct svc_rqst *
+nfs4_callback_up(struct svc_serv *serv)
 {
-	struct svc_serv *serv = NULL;
-	int ret = 0;
-
-	mutex_lock(&nfs_callback_mutex);
-	if (nfs_callback_info.users++ || nfs_callback_info.task != NULL)
-		goto out;
-	serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, NULL);
-	ret = -ENOMEM;
-	if (!serv)
-		goto out_err;
+	int ret;
 
 	ret = svc_create_xprt(serv, "tcp", PF_INET,
 				nfs_callback_set_tcpport, SVC_SOCK_ANONYMOUS);
@@ -131,18 +123,53 @@ int nfs_callback_up(void)
 		goto out_err;
 #endif	/* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */
 
-	nfs_callback_info.rqst = svc_prepare_thread(serv, &serv->sv_pools[0]);
-	if (IS_ERR(nfs_callback_info.rqst)) {
-		ret = PTR_ERR(nfs_callback_info.rqst);
-		nfs_callback_info.rqst = NULL;
+	return svc_prepare_thread(serv, &serv->sv_pools[0]);
+
+out_err:
+	if (ret == 0)
+		ret = -ENOMEM;
+	return ERR_PTR(ret);
+}
+
+/*
+ * Bring up the callback thread if it is not already up.
+ */
+int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt)
+{
+	struct svc_serv *serv = NULL;
+	struct svc_rqst *rqstp;
+	int (*callback_svc)(void *vrqstp);
+	char svc_name[12];
+	int ret = 0;
+
+	mutex_lock(&nfs_callback_mutex);
+	if (nfs_callback_info.users++ || nfs_callback_info.task != NULL)
+		goto out;
+	serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, NULL);
+	if (!serv) {
+		ret = -ENOMEM;
+		goto out_err;
+	}
+
+	if (!minorversion) {
+		rqstp = nfs4_callback_up(serv);
+		callback_svc = nfs4_callback_svc;
+	} else {
+		BUG();	/* for now */
+	}
+
+	if (IS_ERR(rqstp)) {
+		ret = PTR_ERR(rqstp);
 		goto out_err;
 	}
 
 	svc_sock_update_bufs(serv);
 
-	nfs_callback_info.task = kthread_run(nfs_callback_svc,
+	sprintf(svc_name, "nfsv4.%u-svc", minorversion);
+	nfs_callback_info.rqst = rqstp;
+	nfs_callback_info.task = kthread_run(callback_svc,
 					     nfs_callback_info.rqst,
-					     "nfsv4-svc");
+					     svc_name);
 	if (IS_ERR(nfs_callback_info.task)) {
 		ret = PTR_ERR(nfs_callback_info.task);
 		svc_exit_thread(nfs_callback_info.rqst);
diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index e110e286a262..9b907f408b8d 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -63,7 +63,7 @@ extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getat
 extern __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy);
 
 #ifdef CONFIG_NFS_V4
-extern int nfs_callback_up(void);
+extern int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt);
 extern void nfs_callback_down(void);
 #else
 #define nfs_callback_up()	(0)
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index d9657d455730..df2b40d1883d 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -47,9 +47,6 @@
 #include "internal.h"
 #include "fscache.h"
 
-static int nfs4_init_callback(struct nfs_client *);
-static void nfs4_destroy_callback(struct nfs_client *);
-
 #define NFSDBG_FACILITY		NFSDBG_CLIENT
 
 static DEFINE_SPINLOCK(nfs_client_lock);
@@ -124,9 +121,6 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 
 	clp->rpc_ops = cl_init->rpc_ops;
 
-	if (nfs4_init_callback(clp) < 0)
-		goto error_2;
-
 	atomic_set(&clp->cl_count, 1);
 	clp->cl_cons_state = NFS_CS_INITING;
 
@@ -136,7 +130,7 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 	if (cl_init->hostname) {
 		clp->cl_hostname = kstrdup(cl_init->hostname, GFP_KERNEL);
 		if (!clp->cl_hostname)
-			goto error_3;
+			goto error_cleanup;
 	}
 
 	INIT_LIST_HEAD(&clp->cl_superblocks);
@@ -161,9 +155,7 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 
 	return clp;
 
-error_3:
-	nfs4_destroy_callback(clp);
-error_2:
+error_cleanup:
 	kfree(clp);
 error_0:
 	return NULL;
@@ -207,6 +199,8 @@ static void nfs4_clear_client_minor_version(struct nfs_client *clp)
 
 	clp->cl_call_sync = _nfs4_call_sync;
 #endif /* CONFIG_NFS_V4_1 */
+
+	nfs4_destroy_callback(clp);
 }
 
 /*
@@ -225,8 +219,6 @@ static void nfs_free_client(struct nfs_client *clp)
 	if (!IS_ERR(clp->cl_rpcclient))
 		rpc_shutdown_client(clp->cl_rpcclient);
 
-	nfs4_destroy_callback(clp);
-
 	if (clp->cl_machine_cred != NULL)
 		put_rpccred(clp->cl_machine_cred);
 
@@ -1104,7 +1096,8 @@ static int nfs4_init_callback(struct nfs_client *clp)
 	int error;
 
 	if (clp->rpc_ops->version == 4) {
-		error = nfs_callback_up();
+		error = nfs_callback_up(clp->cl_minorversion,
+					clp->cl_rpcclient->cl_xprt);
 		if (error < 0) {
 			dprintk("%s: failed to start callback. Error = %d\n",
 				__func__, error);
@@ -1139,7 +1132,7 @@ static int nfs4_init_client_minor_version(struct nfs_client *clp)
 	}
 #endif /* CONFIG_NFS_V4_1 */
 
-	return 0;
+	return nfs4_init_callback(clp);
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 56632b5bff5af10eb12d7e9499b5ffcadcb7a7b2 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:22:58 -0400
Subject: nfs41: client callback structures

Adds new list of rpc_xprt structures, and a readers/writers lock to
protect the list.  The list is used to preallocate resources for
the backchannel during backchannel requests.  Callbacks are not
expected to cause significant latency, so only one callback will
be allowed at this time.

It also adds a pointer to the NFS callback service so that
requests can be directed to it for processing.

New callback members added to svc_serv. The NFSv4.1 callback service will
sleep on the svc_serv->svc_cb_waitq until new callback requests arrive.
The request will be queued in svc_serv->svc_cb_list. This patch adds this
list, the sleep queue and spinlock to svc_serv.

[nfs41: NFSv4.1 callback support]
Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/sunrpc/svc.h  |  8 ++++++++
 include/linux/sunrpc/xprt.h | 22 ++++++++++++++++++++++
 2 files changed, 30 insertions(+)

diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index 2a30775959e9..4a8afbd62007 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -96,6 +96,14 @@ struct svc_serv {
 	svc_thread_fn		sv_function;	/* main function for threads */
 	unsigned int		sv_drc_max_pages; /* Total pages for DRC */
 	unsigned int		sv_drc_pages_used;/* DRC pages used */
+#if defined(CONFIG_NFS_V4_1)
+	struct list_head	sv_cb_list;	/* queue for callback requests
+						 * that arrive over the same
+						 * connection */
+	spinlock_t		sv_cb_lock;	/* protects the svc_cb_list */
+	wait_queue_head_t	sv_cb_waitq;	/* sleep here if there are no
+						 * entries in the svc_cb_list */
+#endif /* CONFIG_NFS_V4_1 */
 };
 
 /*
diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 08afe43118f4..703af7ebf6cf 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -97,6 +97,12 @@ struct rpc_rqst {
 
 	unsigned long		rq_xtime;	/* when transmitted */
 	int			rq_ntrans;
+
+#if defined(CONFIG_NFS_V4_1)
+	struct list_head	rq_bc_list;	/* Callback service list */
+	unsigned long		rq_bc_pa_state;	/* Backchannel prealloc state */
+	struct list_head	rq_bc_pa_list;	/* Backchannel prealloc list */
+#endif /* CONFIG_NFS_V4_1 */
 };
 #define rq_svec			rq_snd_buf.head
 #define rq_slen			rq_snd_buf.len
@@ -174,6 +180,14 @@ struct rpc_xprt {
 	spinlock_t		reserve_lock;	/* lock slot table */
 	u32			xid;		/* Next XID value to use */
 	struct rpc_task *	snd_task;	/* Task blocked in send */
+#if defined(CONFIG_NFS_V4_1)
+	struct svc_serv		*bc_serv;       /* The RPC service which will */
+						/* process the callback */
+	spinlock_t		bc_pa_lock;	/* Protects the preallocated
+						 * items */
+	struct list_head	bc_pa_list;	/* List of preallocated
+						 * backchannel rpc_rqst's */
+#endif /* CONFIG_NFS_V4_1 */
 	struct list_head	recv;
 
 	struct {
@@ -192,6 +206,14 @@ struct rpc_xprt {
 	const char		*address_strings[RPC_DISPLAY_MAX];
 };
 
+#if defined(CONFIG_NFS_V4_1)
+/*
+ * Backchannel flags
+ */
+#define	RPC_BC_PA_IN_USE	0x0001		/* Preallocated backchannel */
+						/* buffer in use */
+#endif /* CONFIG_NFS_V4_1 */
+
 struct xprt_create {
 	int			ident;		/* XPRT_TRANSPORT identifier */
 	struct sockaddr *	srcaddr;	/* optional local address */
-- 
cgit v1.2.3-59-g8ed1b


From f9acac1a4710ce88871f1ae323fc91c1cb6e9d52 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:22:59 -0400
Subject: nfs41: Initialize new rpc_xprt callback related fields

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 net/sunrpc/xprt.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 06ca058572f2..52739f82df1e 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -1049,6 +1049,11 @@ found:
 
 	INIT_LIST_HEAD(&xprt->free);
 	INIT_LIST_HEAD(&xprt->recv);
+#if defined(CONFIG_NFS_V4_1)
+	spin_lock_init(&xprt->bc_pa_lock);
+	INIT_LIST_HEAD(&xprt->bc_pa_list);
+#endif /* CONFIG_NFS_V4_1 */
+
 	INIT_WORK(&xprt->task_cleanup, xprt_autoclose);
 	setup_timer(&xprt->timer, xprt_init_autodisconnect,
 			(unsigned long)xprt);
-- 
cgit v1.2.3-59-g8ed1b


From fb7a0b9addbdbbb13b7bc02abf55ee524ea19ce1 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:00 -0400
Subject: nfs41: New backchannel helper routines

This patch introduces support to setup the callback xprt on the client side.
It allocates/ destroys the preallocated memory structures used to process
backchannel requests.

At setup time, xprt_setup_backchannel() is invoked to allocate one or
more rpc_rqst structures and substructures.  This ensures that they
are available when an RPC callback arrives.  The rpc_rqst structures
are maintained in a linked list attached to the rpc_xprt structure.
We keep track of the number of allocations so that they can be correctly
removed when the channel is destroyed.

When an RPC callback arrives, xprt_alloc_bc_request() is invoked to
obtain a preallocated rpc_rqst structure.  An rpc_xprt structure is
returned, and its RPC_BC_PREALLOC_IN_USE bit is set in
rpc_xprt->bc_flags.  The structure is removed from the the list
since it is now in use, and it will be later added back when its
user is done with it.

After the RPC callback replies, the rpc_rqst structure is returned
by invoking xprt_free_bc_request().  This clears the
RPC_BC_PREALLOC_IN_USE bit and adds it back to the list, allowing it
to be reused by a subsequent RPC callback request.

To be consistent with the reception of RPC messages, the backchannel requests
should be placed into the 'struct rpc_rqst' rq_rcv_buf, which is then in turn
copied to the 'struct rpc_rqst' rq_private_buf.

[nfs41: Preallocate rpc_rqst receive buffer for handling callbacks]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Update copyright notice and explain page allocation]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/sunrpc/xprt.h   |   1 +
 net/sunrpc/Makefile           |   1 +
 net/sunrpc/backchannel_rqst.c | 278 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 280 insertions(+)
 create mode 100644 net/sunrpc/backchannel_rqst.c

diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 703af7ebf6cf..beae030e80b5 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -183,6 +183,7 @@ struct rpc_xprt {
 #if defined(CONFIG_NFS_V4_1)
 	struct svc_serv		*bc_serv;       /* The RPC service which will */
 						/* process the callback */
+	unsigned int		bc_alloc_count;	/* Total number of preallocs */
 	spinlock_t		bc_pa_lock;	/* Protects the preallocated
 						 * items */
 	struct list_head	bc_pa_list;	/* List of preallocated
diff --git a/net/sunrpc/Makefile b/net/sunrpc/Makefile
index 5369aa369b35..4a01f9684b85 100644
--- a/net/sunrpc/Makefile
+++ b/net/sunrpc/Makefile
@@ -13,5 +13,6 @@ sunrpc-y := clnt.o xprt.o socklib.o xprtsock.o sched.o \
 	    rpcb_clnt.o timer.o xdr.o \
 	    sunrpc_syms.o cache.o rpc_pipe.o \
 	    svc_xprt.o
+sunrpc-$(CONFIG_NFS_V4_1) += backchannel_rqst.o
 sunrpc-$(CONFIG_PROC_FS) += stats.o
 sunrpc-$(CONFIG_SYSCTL) += sysctl.o
diff --git a/net/sunrpc/backchannel_rqst.c b/net/sunrpc/backchannel_rqst.c
new file mode 100644
index 000000000000..f56e18a23498
--- /dev/null
+++ b/net/sunrpc/backchannel_rqst.c
@@ -0,0 +1,278 @@
+/******************************************************************************
+
+(c) 2007 Network Appliance, Inc.  All Rights Reserved.
+(c) 2009 NetApp.  All Rights Reserved.
+
+NetApp provides this source code under the GPL v2 License.
+The GPL v2 license is available at
+http://opensource.org/licenses/gpl-license.php.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
+CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
+PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+******************************************************************************/
+
+#include <linux/tcp.h>
+#include <linux/sunrpc/xprt.h>
+
+#ifdef RPC_DEBUG
+#define RPCDBG_FACILITY	RPCDBG_TRANS
+#endif
+
+#if defined(CONFIG_NFS_V4_1)
+
+/*
+ * Helper routines that track the number of preallocation elements
+ * on the transport.
+ */
+static inline int xprt_need_to_requeue(struct rpc_xprt *xprt)
+{
+	return xprt->bc_alloc_count > 0;
+}
+
+static inline void xprt_inc_alloc_count(struct rpc_xprt *xprt, unsigned int n)
+{
+	xprt->bc_alloc_count += n;
+}
+
+static inline int xprt_dec_alloc_count(struct rpc_xprt *xprt, unsigned int n)
+{
+	return xprt->bc_alloc_count -= n;
+}
+
+/*
+ * Free the preallocated rpc_rqst structure and the memory
+ * buffers hanging off of it.
+ */
+static void xprt_free_allocation(struct rpc_rqst *req)
+{
+	struct xdr_buf *xbufp;
+
+	dprintk("RPC:        free allocations for req= %p\n", req);
+	BUG_ON(test_bit(RPC_BC_PA_IN_USE, &req->rq_bc_pa_state));
+	xbufp = &req->rq_private_buf;
+	free_page((unsigned long)xbufp->head[0].iov_base);
+	xbufp = &req->rq_snd_buf;
+	free_page((unsigned long)xbufp->head[0].iov_base);
+	list_del(&req->rq_bc_pa_list);
+	kfree(req);
+}
+
+/*
+ * Preallocate up to min_reqs structures and related buffers for use
+ * by the backchannel.  This function can be called multiple times
+ * when creating new sessions that use the same rpc_xprt.  The
+ * preallocated buffers are added to the pool of resources used by
+ * the rpc_xprt.  Anyone of these resources may be used used by an
+ * incoming callback request.  It's up to the higher levels in the
+ * stack to enforce that the maximum number of session slots is not
+ * being exceeded.
+ *
+ * Some callback arguments can be large.  For example, a pNFS server
+ * using multiple deviceids.  The list can be unbound, but the client
+ * has the ability to tell the server the maximum size of the callback
+ * requests.  Each deviceID is 16 bytes, so allocate one page
+ * for the arguments to have enough room to receive a number of these
+ * deviceIDs.  The NFS client indicates to the pNFS server that its
+ * callback requests can be up to 4096 bytes in size.
+ */
+int xprt_setup_backchannel(struct rpc_xprt *xprt, unsigned int min_reqs)
+{
+	struct page *page_rcv = NULL, *page_snd = NULL;
+	struct xdr_buf *xbufp = NULL;
+	struct rpc_rqst *req, *tmp;
+	struct list_head tmp_list;
+	int i;
+
+	dprintk("RPC:       setup backchannel transport\n");
+
+	/*
+	 * We use a temporary list to keep track of the preallocated
+	 * buffers.  Once we're done building the list we splice it
+	 * into the backchannel preallocation list off of the rpc_xprt
+	 * struct.  This helps minimize the amount of time the list
+	 * lock is held on the rpc_xprt struct.  It also makes cleanup
+	 * easier in case of memory allocation errors.
+	 */
+	INIT_LIST_HEAD(&tmp_list);
+	for (i = 0; i < min_reqs; i++) {
+		/* Pre-allocate one backchannel rpc_rqst */
+		req = kzalloc(sizeof(struct rpc_rqst), GFP_KERNEL);
+		if (req == NULL) {
+			printk(KERN_ERR "Failed to create bc rpc_rqst\n");
+			goto out_free;
+		}
+
+		/* Add the allocated buffer to the tmp list */
+		dprintk("RPC:       adding req= %p\n", req);
+		list_add(&req->rq_bc_pa_list, &tmp_list);
+
+		req->rq_xprt = xprt;
+		INIT_LIST_HEAD(&req->rq_list);
+		INIT_LIST_HEAD(&req->rq_bc_list);
+
+		/* Preallocate one XDR receive buffer */
+		page_rcv = alloc_page(GFP_KERNEL);
+		if (page_rcv == NULL) {
+			printk(KERN_ERR "Failed to create bc receive xbuf\n");
+			goto out_free;
+		}
+		xbufp = &req->rq_rcv_buf;
+		xbufp->head[0].iov_base = page_address(page_rcv);
+		xbufp->head[0].iov_len = PAGE_SIZE;
+		xbufp->tail[0].iov_base = NULL;
+		xbufp->tail[0].iov_len = 0;
+		xbufp->page_len = 0;
+		xbufp->len = PAGE_SIZE;
+		xbufp->buflen = PAGE_SIZE;
+
+		/* Preallocate one XDR send buffer */
+		page_snd = alloc_page(GFP_KERNEL);
+		if (page_snd == NULL) {
+			printk(KERN_ERR "Failed to create bc snd xbuf\n");
+			goto out_free;
+		}
+
+		xbufp = &req->rq_snd_buf;
+		xbufp->head[0].iov_base = page_address(page_snd);
+		xbufp->head[0].iov_len = 0;
+		xbufp->tail[0].iov_base = NULL;
+		xbufp->tail[0].iov_len = 0;
+		xbufp->page_len = 0;
+		xbufp->len = 0;
+		xbufp->buflen = PAGE_SIZE;
+	}
+
+	/*
+	 * Add the temporary list to the backchannel preallocation list
+	 */
+	spin_lock_bh(&xprt->bc_pa_lock);
+	list_splice(&tmp_list, &xprt->bc_pa_list);
+	xprt_inc_alloc_count(xprt, min_reqs);
+	spin_unlock_bh(&xprt->bc_pa_lock);
+
+	dprintk("RPC:       setup backchannel transport done\n");
+	return 0;
+
+out_free:
+	/*
+	 * Memory allocation failed, free the temporary list
+	 */
+	list_for_each_entry_safe(req, tmp, &tmp_list, rq_bc_pa_list)
+		xprt_free_allocation(req);
+
+	dprintk("RPC:       setup backchannel transport failed\n");
+	return -1;
+}
+EXPORT_SYMBOL(xprt_setup_backchannel);
+
+/*
+ * Destroys the backchannel preallocated structures.
+ * Since these structures may have been allocated by multiple calls
+ * to xprt_setup_backchannel, we only destroy up to the maximum number
+ * of reqs specified by the caller.
+ * @xprt:	the transport holding the preallocated strucures
+ * @max_reqs	the maximum number of preallocated structures to destroy
+ */
+void xprt_destroy_backchannel(struct rpc_xprt *xprt, unsigned int max_reqs)
+{
+	struct rpc_rqst *req = NULL, *tmp = NULL;
+
+	dprintk("RPC:        destroy backchannel transport\n");
+
+	BUG_ON(max_reqs == 0);
+	spin_lock_bh(&xprt->bc_pa_lock);
+	xprt_dec_alloc_count(xprt, max_reqs);
+	list_for_each_entry_safe(req, tmp, &xprt->bc_pa_list, rq_bc_pa_list) {
+		dprintk("RPC:        req=%p\n", req);
+		xprt_free_allocation(req);
+		if (--max_reqs == 0)
+			break;
+	}
+	spin_unlock_bh(&xprt->bc_pa_lock);
+
+	dprintk("RPC:        backchannel list empty= %s\n",
+		list_empty(&xprt->bc_pa_list) ? "true" : "false");
+}
+EXPORT_SYMBOL(xprt_destroy_backchannel);
+
+/*
+ * One or more rpc_rqst structure have been preallocated during the
+ * backchannel setup.  Buffer space for the send and private XDR buffers
+ * has been preallocated as well.  Use xprt_alloc_bc_request to allocate
+ * to this request.  Use xprt_free_bc_request to return it.
+ *
+ * Return an available rpc_rqst, otherwise NULL if non are available.
+ */
+struct rpc_rqst *xprt_alloc_bc_request(struct rpc_xprt *xprt)
+{
+	struct rpc_rqst *req;
+
+	dprintk("RPC:       allocate a backchannel request\n");
+	spin_lock_bh(&xprt->bc_pa_lock);
+	if (!list_empty(&xprt->bc_pa_list)) {
+		req = list_first_entry(&xprt->bc_pa_list, struct rpc_rqst,
+				rq_bc_pa_list);
+		list_del(&req->rq_bc_pa_list);
+	} else {
+		req = NULL;
+	}
+	spin_unlock_bh(&xprt->bc_pa_lock);
+
+	if (req != NULL) {
+		set_bit(RPC_BC_PA_IN_USE, &req->rq_bc_pa_state);
+		req->rq_received = 0;
+		req->rq_bytes_sent = 0;
+		memcpy(&req->rq_private_buf, &req->rq_rcv_buf,
+			sizeof(req->rq_private_buf));
+	}
+	dprintk("RPC:       backchannel req=%p\n", req);
+	return req;
+}
+
+/*
+ * Return the preallocated rpc_rqst structure and XDR buffers
+ * associated with this rpc_task.
+ */
+void xprt_free_bc_request(struct rpc_rqst *req)
+{
+	struct rpc_xprt *xprt = req->rq_xprt;
+
+	dprintk("RPC:       free backchannel req=%p\n", req);
+
+	smp_mb__before_clear_bit();
+	BUG_ON(!test_bit(RPC_BC_PA_IN_USE, &req->rq_bc_pa_state));
+	clear_bit(RPC_BC_PA_IN_USE, &req->rq_bc_pa_state);
+	smp_mb__after_clear_bit();
+
+	if (!xprt_need_to_requeue(xprt)) {
+		/*
+		 * The last remaining session was destroyed while this
+		 * entry was in use.  Free the entry and don't attempt
+		 * to add back to the list because there is no need to
+		 * have anymore preallocated entries.
+		 */
+		dprintk("RPC:       Last session removed req=%p\n", req);
+		xprt_free_allocation(req);
+		return;
+	}
+
+	/*
+	 * Return it to the list of preallocations so that it
+	 * may be reused by a new callback request.
+	 */
+	spin_lock_bh(&xprt->bc_pa_lock);
+	list_add(&req->rq_bc_pa_list, &xprt->bc_pa_list);
+	spin_unlock_bh(&xprt->bc_pa_lock);
+}
+
+#endif /* CONFIG_NFS_V4_1 */
-- 
cgit v1.2.3-59-g8ed1b


From 4a8d70bfef01f8e6b27785e2625e88e9a80924a5 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:01 -0400
Subject: nfs41: New include/linux/sunrpc/bc_xprt.h

Contains prototype for backchannel helper routines.

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: xprt_setup_backchannel v4.0 only inline]
    Fix compile error when CONFIG_NFS_V4_1 is not set.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Update Copyright notice and fix formatting]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/sunrpc/bc_xprt.h | 46 ++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 46 insertions(+)
 create mode 100644 include/linux/sunrpc/bc_xprt.h

diff --git a/include/linux/sunrpc/bc_xprt.h b/include/linux/sunrpc/bc_xprt.h
new file mode 100644
index 000000000000..5965ae4f902d
--- /dev/null
+++ b/include/linux/sunrpc/bc_xprt.h
@@ -0,0 +1,46 @@
+/******************************************************************************
+
+(c) 2008 NetApp.  All Rights Reserved.
+
+NetApp provides this source code under the GPL v2 License.
+The GPL v2 license is available at
+http://opensource.org/licenses/gpl-license.php.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
+CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
+PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+******************************************************************************/
+
+/*
+ * Functions to create and manage the backchannel
+ */
+
+#ifndef _LINUX_SUNRPC_BC_XPRT_H
+#define _LINUX_SUNRPC_BC_XPRT_H
+
+#include <linux/sunrpc/svcsock.h>
+#include <linux/sunrpc/xprt.h>
+
+#ifdef CONFIG_NFS_V4_1
+struct rpc_rqst *xprt_alloc_bc_request(struct rpc_xprt *xprt);
+void xprt_free_bc_request(struct rpc_rqst *req);
+int xprt_setup_backchannel(struct rpc_xprt *, unsigned int min_reqs);
+void xprt_destroy_backchannel(struct rpc_xprt *, int max_reqs);
+#else /* CONFIG_NFS_V4_1 */
+static inline int xprt_setup_backchannel(struct rpc_xprt *xprt,
+					 unsigned int min_reqs)
+{
+	return 0;
+}
+#endif /* CONFIG_NFS_V4_1 */
+#endif /* _LINUX_SUNRPC_BC_XPRT_H */
+
-- 
cgit v1.2.3-59-g8ed1b


From 44b98efdd0a205bdca2cb63493350d06ff6804b1 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:02 -0400
Subject: nfs41: New xs_tcp_read_data()

Handles RPC replies and backchannel callbacks.  Traditionally the NFS
client has expected only RPC replies on its open connections.  With
NFSv4.1, callbacks can arrive over an existing open connection.

This patch refactors the old xs_tcp_read_request() into an RPC reply handler:
xs_tcp_read_reply(), a new backchannel callback handler: xs_tcp_read_callback(),
and a common routine to read the data off the transport: xs_tcp_read_common().
The new xs_tcp_read_callback() queues callback requests onto a queue where
the callback service (a separate thread) is listening for the processing.

This patch incorporates work and suggestions from Rahul Iyer (iyer@netapp.com)
and Benny Halevy (bhalevy@panasas.com).

xs_tcp_read_callback() drops the connection when the number of expected
callbacks is exceeded.  Use xprt_force_disconnect(), ensuring tasks on
the pending queue are awaken on disconnect.

[nfs41: Keep track of RPC call/reply direction with a flag]
[nfs41: Preallocate rpc_rqst receive buffer for handling callbacks]
Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: sunrpc: xs_tcp_read_callback() should use xprt_force_disconnect()]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Moves embedded #ifdefs into #ifdef function blocks]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 net/sunrpc/xprtsock.c | 144 +++++++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 126 insertions(+), 18 deletions(-)

diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index a48df1449ece..e3e3a57116fb 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -34,6 +34,9 @@
 #include <linux/sunrpc/sched.h>
 #include <linux/sunrpc/xprtsock.h>
 #include <linux/file.h>
+#ifdef CONFIG_NFS_V4_1
+#include <linux/sunrpc/bc_xprt.h>
+#endif
 
 #include <net/sock.h>
 #include <net/checksum.h>
@@ -1044,25 +1047,16 @@ static inline void xs_tcp_read_calldir(struct sock_xprt *transport,
 	xs_tcp_check_fraghdr(transport);
 }
 
-static inline void xs_tcp_read_request(struct rpc_xprt *xprt, struct xdr_skb_reader *desc)
+static inline void xs_tcp_read_common(struct rpc_xprt *xprt,
+				     struct xdr_skb_reader *desc,
+				     struct rpc_rqst *req)
 {
-	struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
-	struct rpc_rqst *req;
+	struct sock_xprt *transport =
+				container_of(xprt, struct sock_xprt, xprt);
 	struct xdr_buf *rcvbuf;
 	size_t len;
 	ssize_t r;
 
-	/* Find and lock the request corresponding to this xid */
-	spin_lock(&xprt->transport_lock);
-	req = xprt_lookup_rqst(xprt, transport->tcp_xid);
-	if (!req) {
-		transport->tcp_flags &= ~TCP_RCV_COPY_DATA;
-		dprintk("RPC:       XID %08x request not found!\n",
-				ntohl(transport->tcp_xid));
-		spin_unlock(&xprt->transport_lock);
-		return;
-	}
-
 	rcvbuf = &req->rq_private_buf;
 
 	if (transport->tcp_flags & TCP_RCV_COPY_CALLDIR) {
@@ -1114,7 +1108,7 @@ static inline void xs_tcp_read_request(struct rpc_xprt *xprt, struct xdr_skb_rea
 				"tcp_offset = %u, tcp_reclen = %u\n",
 				xprt, transport->tcp_copied,
 				transport->tcp_offset, transport->tcp_reclen);
-		goto out;
+		return;
 	}
 
 	dprintk("RPC:       XID %08x read %Zd bytes\n",
@@ -1130,11 +1124,125 @@ static inline void xs_tcp_read_request(struct rpc_xprt *xprt, struct xdr_skb_rea
 			transport->tcp_flags &= ~TCP_RCV_COPY_DATA;
 	}
 
-out:
+	return;
+}
+
+/*
+ * Finds the request corresponding to the RPC xid and invokes the common
+ * tcp read code to read the data.
+ */
+static inline int xs_tcp_read_reply(struct rpc_xprt *xprt,
+				    struct xdr_skb_reader *desc)
+{
+	struct sock_xprt *transport =
+				container_of(xprt, struct sock_xprt, xprt);
+	struct rpc_rqst *req;
+
+	dprintk("RPC:       read reply XID %08x\n", ntohl(transport->tcp_xid));
+
+	/* Find and lock the request corresponding to this xid */
+	spin_lock(&xprt->transport_lock);
+	req = xprt_lookup_rqst(xprt, transport->tcp_xid);
+	if (!req) {
+		dprintk("RPC:       XID %08x request not found!\n",
+				ntohl(transport->tcp_xid));
+		spin_unlock(&xprt->transport_lock);
+		return -1;
+	}
+
+	xs_tcp_read_common(xprt, desc, req);
+
 	if (!(transport->tcp_flags & TCP_RCV_COPY_DATA))
 		xprt_complete_rqst(req->rq_task, transport->tcp_copied);
+
 	spin_unlock(&xprt->transport_lock);
-	xs_tcp_check_fraghdr(transport);
+	return 0;
+}
+
+#if defined(CONFIG_NFS_V4_1)
+/*
+ * Obtains an rpc_rqst previously allocated and invokes the common
+ * tcp read code to read the data.  The result is placed in the callback
+ * queue.
+ * If we're unable to obtain the rpc_rqst we schedule the closing of the
+ * connection and return -1.
+ */
+static inline int xs_tcp_read_callback(struct rpc_xprt *xprt,
+				       struct xdr_skb_reader *desc)
+{
+	struct sock_xprt *transport =
+				container_of(xprt, struct sock_xprt, xprt);
+	struct rpc_rqst *req;
+
+	req = xprt_alloc_bc_request(xprt);
+	if (req == NULL) {
+		printk(KERN_WARNING "Callback slot table overflowed\n");
+		xprt_force_disconnect(xprt);
+		return -1;
+	}
+
+	req->rq_xid = transport->tcp_xid;
+	dprintk("RPC:       read callback  XID %08x\n", ntohl(req->rq_xid));
+	xs_tcp_read_common(xprt, desc, req);
+
+	if (!(transport->tcp_flags & TCP_RCV_COPY_DATA)) {
+		struct svc_serv *bc_serv = xprt->bc_serv;
+
+		/*
+		 * Add callback request to callback list.  The callback
+		 * service sleeps on the sv_cb_waitq waiting for new
+		 * requests.  Wake it up after adding enqueing the
+		 * request.
+		 */
+		dprintk("RPC:       add callback request to list\n");
+		spin_lock(&bc_serv->sv_cb_lock);
+		list_add(&req->rq_bc_list, &bc_serv->sv_cb_list);
+		spin_unlock(&bc_serv->sv_cb_lock);
+		wake_up(&bc_serv->sv_cb_waitq);
+	}
+
+	req->rq_private_buf.len = transport->tcp_copied;
+
+	return 0;
+}
+
+static inline int _xs_tcp_read_data(struct rpc_xprt *xprt,
+					struct xdr_skb_reader *desc)
+{
+	struct sock_xprt *transport =
+				container_of(xprt, struct sock_xprt, xprt);
+
+	return (transport->tcp_flags & TCP_RPC_REPLY) ?
+		xs_tcp_read_reply(xprt, desc) :
+		xs_tcp_read_callback(xprt, desc);
+}
+#else
+static inline int _xs_tcp_read_data(struct rpc_xprt *xprt,
+					struct xdr_skb_reader *desc)
+{
+	return xs_tcp_read_reply(xprt, desc);
+}
+#endif /* CONFIG_NFS_V4_1 */
+
+/*
+ * Read data off the transport.  This can be either an RPC_CALL or an
+ * RPC_REPLY.  Relay the processing to helper functions.
+ */
+static void xs_tcp_read_data(struct rpc_xprt *xprt,
+				    struct xdr_skb_reader *desc)
+{
+	struct sock_xprt *transport =
+				container_of(xprt, struct sock_xprt, xprt);
+
+	if (_xs_tcp_read_data(xprt, desc) == 0)
+		xs_tcp_check_fraghdr(transport);
+	else {
+		/*
+		 * The transport_lock protects the request handling.
+		 * There's no need to hold it to update the tcp_flags.
+		 */
+		transport->tcp_flags &= ~TCP_RCV_COPY_DATA;
+	}
 }
 
 static inline void xs_tcp_read_discard(struct sock_xprt *transport, struct xdr_skb_reader *desc)
@@ -1181,7 +1289,7 @@ static int xs_tcp_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb, uns
 		}
 		/* Read in the request data */
 		if (transport->tcp_flags & TCP_RCV_COPY_DATA) {
-			xs_tcp_read_request(xprt, &desc);
+			xs_tcp_read_data(xprt, &desc);
 			continue;
 		}
 		/* Skip over any trailing bytes on short reads */
-- 
cgit v1.2.3-59-g8ed1b


From 88b5ed73bcd0f21e008b6e303a02c8b7cb1199f4 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 17 Jun 2009 13:22:57 -0700
Subject: SUNRPC: Fix a missing "break" option in xs_tcp_setup_socket()

In the case of -EADDRNOTAVAIL and/or unhandled connection errors, we want
to get rid of the existing socket and retry immediately, just as the
comment says. Currently we end up sleeping for a minute, due to the missing
"break" statement.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 net/sunrpc/xprtsock.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 6c2d61586551..f7f3dfd211ea 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -1792,6 +1792,7 @@ static void xs_tcp_setup_socket(struct rpc_xprt *xprt,
 		 */
 		set_bit(XPRT_CONNECTION_CLOSE, &xprt->state);
 		xprt_force_disconnect(xprt);
+		break;
 	case -ECONNREFUSED:
 	case -ECONNRESET:
 	case -ENETUNREACH:
-- 
cgit v1.2.3-59-g8ed1b


From 01c3f05228ce7fc19baa103e4e4bf6c1b5062a53 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 17 Jun 2009 13:22:58 -0700
Subject: NFSv4: Fix the 'nolock' option regression

NFSv4 should just ignore the 'nolock' option. It is an NFSv2/v3 thing...
This fixes the Oops in http://bugzilla.kernel.org/show_bug.cgi?id=13330

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/super.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 26127b69a275..98b47d17740a 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -2240,6 +2240,11 @@ static void nfs4_fill_super(struct super_block *sb)
 	nfs_initialise_sb(sb);
 }
 
+static void nfs4_validate_mount_flags(struct nfs_parsed_mount_data *args)
+{
+	args->flags &= ~(NFS_MOUNT_NONLM|NFS_MOUNT_NOACL|NFS_MOUNT_VER3);
+}
+
 /*
  * Validate NFSv4 mount options
  */
@@ -2336,6 +2341,8 @@ static int nfs4_validate_mount_data(void *options,
 
 		nfs_validate_transport_protocol(args);
 
+		nfs4_validate_mount_flags(args);
+
 		if (args->auth_flavor_len > 1)
 			goto out_inval_auth;
 
-- 
cgit v1.2.3-59-g8ed1b


From d5122201a7f90b2aa73092f158b84d1d74f1134d Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 17 Jun 2009 13:22:58 -0700
Subject: NFSv4: Move error handling out of the delegation generic code

The NFSv4 delegation recovery code is required by the protocol to handle
more errors. Rather than add NFSv4.0 specific errors into 'generic'
delegation code, we should move the error handling into the NFSv4 layer.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/delegation.c | 18 +++---------------
 fs/nfs/nfs4proc.c   | 15 +++++++++++++--
 2 files changed, 16 insertions(+), 17 deletions(-)

diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index 968225a88015..d4f669f0683e 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -68,7 +68,7 @@ static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_
 {
 	struct inode *inode = state->inode;
 	struct file_lock *fl;
-	int status;
+	int status = 0;
 
 	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
 		if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK)))
@@ -76,21 +76,9 @@ static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_
 		if (nfs_file_open_context(fl->fl_file) != ctx)
 			continue;
 		status = nfs4_lock_delegation_recall(state, fl);
-		if (status >= 0)
-			continue;
-		switch (status) {
-			default:
-				printk(KERN_ERR "%s: unhandled error %d.\n",
-						__func__, status);
-			case -NFS4ERR_EXPIRED:
-				/* kill_proc(fl->fl_pid, SIGLOST, 1); */
-			case -NFS4ERR_STALE_CLIENTID:
-				nfs4_schedule_state_recovery(NFS_SERVER(inode)->nfs_client);
-				goto out_err;
-		}
+		if (status < 0)
+			break;
 	}
-	return 0;
-out_err:
 	return status;
 }
 
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 4674f8092da8..d326193c9785 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3630,8 +3630,19 @@ int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl)
 		goto out;
 	do {
 		err = _nfs4_do_setlk(state, F_SETLK, fl, 0);
-		if (err != -NFS4ERR_DELAY)
-			break;
+		switch (err) {
+			default:
+				printk(KERN_ERR "%s: unhandled error %d.\n",
+						__func__, err);
+			case 0:
+				goto out;
+			case -NFS4ERR_EXPIRED:
+			case -NFS4ERR_STALE_CLIENTID:
+				nfs4_schedule_state_recovery(server->nfs_client);
+				goto out;
+			case -NFS4ERR_DELAY:
+				break;
+		}
 		err = nfs4_handle_exception(server, err, &exception);
 	} while (exception.retry);
 out:
-- 
cgit v1.2.3-59-g8ed1b


From 965b5d679146c9f69bc0325388bb9ed357863c4f Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 17 Jun 2009 13:22:59 -0700
Subject: NFSv4: Handle more errors when recovering open file and locking state

It is possible for servers to return NFS4ERR_BAD_STATEID when
the state management code is recovering locks or is reclaiming state when
returning a delegation. Ensure that we handle that case.
While we're at it, add in handlers for NFS4ERR_STALE,
NFS4ERR_ADMIN_REVOKED, NFS4ERR_OPENMODE, NFS4ERR_DENIED and
NFS4ERR_STALE_STATEID, since the protocol appears to allow for them too.

Also handle ENOMEM...

Finally, rather than add new NFSv4.0-specific errors and error handling into
the generic delegation code, move that open file and locking state error
handling into the NFSv4 layer.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c  | 44 ++++++++++++++++++++++++++++++++++++++------
 fs/nfs/nfs4state.c | 28 ++++++++++++++++++----------
 2 files changed, 56 insertions(+), 16 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index d326193c9785..55314e721632 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -804,16 +804,30 @@ int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state
 		err = _nfs4_open_delegation_recall(ctx, state, stateid);
 		switch (err) {
 			case 0:
-				return err;
+			case -ENOENT:
+			case -ESTALE:
+				goto out;
 			case -NFS4ERR_STALE_CLIENTID:
 			case -NFS4ERR_STALE_STATEID:
 			case -NFS4ERR_EXPIRED:
 				/* Don't recall a delegation if it was lost */
 				nfs4_schedule_state_recovery(server->nfs_client);
-				return err;
+				goto out;
+			case -ERESTARTSYS:
+				/*
+				 * The show must go on: exit, but mark the
+				 * stateid as needing recovery.
+				 */
+			case -NFS4ERR_ADMIN_REVOKED:
+			case -NFS4ERR_BAD_STATEID:
+				nfs4_state_mark_reclaim_nograce(server->nfs_client, state);
+			case -ENOMEM:
+				err = 0;
+				goto out;
 		}
 		err = nfs4_handle_exception(server, err, &exception);
 	} while (exception.retry);
+out:
 	return err;
 }
 
@@ -3487,8 +3501,6 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 	ret = nfs4_wait_for_completion_rpc_task(task);
 	if (ret == 0) {
 		ret = data->rpc_status;
-		if (ret == -NFS4ERR_DENIED)
-			ret = -EAGAIN;
 	} else
 		data->cancelled = 1;
 	rpc_put_task(task);
@@ -3576,9 +3588,11 @@ static int nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock *
 	int err;
 
 	do {
+		err = _nfs4_proc_setlk(state, cmd, request);
+		if (err == -NFS4ERR_DENIED)
+			err = -EAGAIN;
 		err = nfs4_handle_exception(NFS_SERVER(state->inode),
-				_nfs4_proc_setlk(state, cmd, request),
-				&exception);
+				err, &exception);
 	} while (exception.retry);
 	return err;
 }
@@ -3635,11 +3649,29 @@ int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl)
 				printk(KERN_ERR "%s: unhandled error %d.\n",
 						__func__, err);
 			case 0:
+			case -ESTALE:
 				goto out;
 			case -NFS4ERR_EXPIRED:
 			case -NFS4ERR_STALE_CLIENTID:
+			case -NFS4ERR_STALE_STATEID:
 				nfs4_schedule_state_recovery(server->nfs_client);
 				goto out;
+			case -ERESTARTSYS:
+				/*
+				 * The show must go on: exit, but mark the
+				 * stateid as needing recovery.
+				 */
+			case -NFS4ERR_ADMIN_REVOKED:
+			case -NFS4ERR_BAD_STATEID:
+			case -NFS4ERR_OPENMODE:
+				nfs4_state_mark_reclaim_nograce(server->nfs_client, state);
+				err = 0;
+				goto out;
+			case -ENOMEM:
+			case -NFS4ERR_DENIED:
+				/* kill_proc(fl->fl_pid, SIGLOST, 1); */
+				err = 0;
+				goto out;
 			case -NFS4ERR_DELAY:
 				break;
 		}
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 0298e909559f..6e094e95cf0f 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -854,25 +854,29 @@ static int nfs4_reclaim_locks(struct nfs4_state *state, const struct nfs4_state_
 		if (nfs_file_open_context(fl->fl_file)->state != state)
 			continue;
 		status = ops->recover_lock(state, fl);
-		if (status >= 0)
-			continue;
 		switch (status) {
+			case 0:
+				break;
+			case -ESTALE:
+			case -NFS4ERR_ADMIN_REVOKED:
+			case -NFS4ERR_STALE_STATEID:
+			case -NFS4ERR_BAD_STATEID:
+			case -NFS4ERR_EXPIRED:
+			case -NFS4ERR_NO_GRACE:
+			case -NFS4ERR_STALE_CLIENTID:
+				goto out;
 			default:
 				printk(KERN_ERR "%s: unhandled error %d. Zeroing state\n",
 						__func__, status);
-			case -NFS4ERR_EXPIRED:
-			case -NFS4ERR_NO_GRACE:
+			case -ENOMEM:
+			case -NFS4ERR_DENIED:
 			case -NFS4ERR_RECLAIM_BAD:
 			case -NFS4ERR_RECLAIM_CONFLICT:
 				/* kill_proc(fl->fl_pid, SIGLOST, 1); */
-				break;
-			case -NFS4ERR_STALE_CLIENTID:
-				goto out_err;
+				status = 0;
 		}
 	}
-	up_write(&nfsi->rwsem);
-	return 0;
-out_err:
+out:
 	up_write(&nfsi->rwsem);
 	return status;
 }
@@ -918,6 +922,7 @@ restart:
 				printk(KERN_ERR "%s: unhandled error %d. Zeroing state\n",
 						__func__, status);
 			case -ENOENT:
+			case -ENOMEM:
 			case -ESTALE:
 				/*
 				 * Open state on this file cannot be recovered
@@ -928,6 +933,9 @@ restart:
 				/* Mark the file as being 'closed' */
 				state->state = 0;
 				break;
+			case -NFS4ERR_ADMIN_REVOKED:
+			case -NFS4ERR_STALE_STATEID:
+			case -NFS4ERR_BAD_STATEID:
 			case -NFS4ERR_RECLAIM_BAD:
 			case -NFS4ERR_RECLAIM_CONFLICT:
 				nfs4_state_mark_reclaim_nograce(sp->so_client, state);
-- 
cgit v1.2.3-59-g8ed1b


From 3f09df70e3a33590ae5a97b8a15486d3711c7065 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 17 Jun 2009 13:23:00 -0700
Subject: NFS: Ensure we always hold the BKL when dereferencing inode->i_flock

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/delegation.c | 16 ++++++++++++++--
 fs/nfs/nfs4state.c  |  9 +++++++++
 2 files changed, 23 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index d4f669f0683e..af05b918cb5b 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -70,15 +70,24 @@ static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_
 	struct file_lock *fl;
 	int status = 0;
 
+	if (inode->i_flock == NULL)
+		goto out;
+
+	/* Protect inode->i_flock using the BKL */
+	lock_kernel();
 	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
 		if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK)))
 			continue;
 		if (nfs_file_open_context(fl->fl_file) != ctx)
 			continue;
+		unlock_kernel();
 		status = nfs4_lock_delegation_recall(state, fl);
 		if (status < 0)
-			break;
+			goto out;
+		lock_kernel();
 	}
+	unlock_kernel();
+out:
 	return status;
 }
 
@@ -256,7 +265,10 @@ static int __nfs_inode_return_delegation(struct inode *inode, struct nfs_delegat
 	struct nfs_inode *nfsi = NFS_I(inode);
 
 	nfs_msync_inode(inode);
-	/* Guard against new delegated open calls */
+	/*
+	 * Guard against new delegated open/lock/unlock calls and against
+	 * state recovery
+	 */
 	down_write(&nfsi->rwsem);
 	nfs_delegation_claim_opens(inode, &delegation->stateid);
 	up_write(&nfsi->rwsem);
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 6e094e95cf0f..0b2a19bfe5a0 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -847,12 +847,19 @@ static int nfs4_reclaim_locks(struct nfs4_state *state, const struct nfs4_state_
 	struct file_lock *fl;
 	int status = 0;
 
+	if (inode->i_flock == NULL)
+		return 0;
+
+	/* Guard against delegation returns and new lock/unlock calls */
 	down_write(&nfsi->rwsem);
+	/* Protect inode->i_flock using the BKL */
+	lock_kernel();
 	for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) {
 		if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK)))
 			continue;
 		if (nfs_file_open_context(fl->fl_file)->state != state)
 			continue;
+		unlock_kernel();
 		status = ops->recover_lock(state, fl);
 		switch (status) {
 			case 0:
@@ -875,7 +882,9 @@ static int nfs4_reclaim_locks(struct nfs4_state *state, const struct nfs4_state_
 				/* kill_proc(fl->fl_pid, SIGLOST, 1); */
 				status = 0;
 		}
+		lock_kernel();
 	}
+	unlock_kernel();
 out:
 	up_write(&nfsi->rwsem);
 	return status;
-- 
cgit v1.2.3-59-g8ed1b


From 5cd973c44a92f4abf8f7084c804089b3eaa7b4bf Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Wed, 17 Jun 2009 13:23:01 -0700
Subject: NFSv4/NLM: Push file locking BKL dependencies down into the NLM layer

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/lockd/clntproc.c | 2 ++
 fs/nfs/file.c       | 6 ------
 2 files changed, 2 insertions(+), 6 deletions(-)

diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c
index dd7957064a8c..273e229353f3 100644
--- a/fs/lockd/clntproc.c
+++ b/fs/lockd/clntproc.c
@@ -165,6 +165,7 @@ int nlmclnt_proc(struct nlm_host *host, int cmd, struct file_lock *fl)
 	/* Set up the argument struct */
 	nlmclnt_setlockargs(call, fl);
 
+	lock_kernel();
 	if (IS_SETLK(cmd) || IS_SETLKW(cmd)) {
 		if (fl->fl_type != F_UNLCK) {
 			call->a_args.block = IS_SETLKW(cmd) ? 1 : 0;
@@ -178,6 +179,7 @@ int nlmclnt_proc(struct nlm_host *host, int cmd, struct file_lock *fl)
 
 	fl->fl_ops->fl_release_private(fl);
 	fl->fl_ops = NULL;
+	unlock_kernel();
 
 	dprintk("lockd: clnt proc returns %d\n", status);
 	return status;
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index ec7e27d00bc6..52ed56762583 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -592,7 +592,6 @@ static int do_getlk(struct file *filp, int cmd, struct file_lock *fl)
 	struct inode *inode = filp->f_mapping->host;
 	int status = 0;
 
-	lock_kernel();
 	/* Try local locking first */
 	posix_test_lock(filp, fl);
 	if (fl->fl_type != F_UNLCK) {
@@ -608,7 +607,6 @@ static int do_getlk(struct file *filp, int cmd, struct file_lock *fl)
 
 	status = NFS_PROTO(inode)->lock(filp, cmd, fl);
 out:
-	unlock_kernel();
 	return status;
 out_noconflict:
 	fl->fl_type = F_UNLCK;
@@ -650,13 +648,11 @@ static int do_unlk(struct file *filp, int cmd, struct file_lock *fl)
 	 * 	If we're signalled while cleaning up locks on process exit, we
 	 * 	still need to complete the unlock.
 	 */
-	lock_kernel();
 	/* Use local locking if mounted with "-onolock" */
 	if (!(NFS_SERVER(inode)->flags & NFS_MOUNT_NONLM))
 		status = NFS_PROTO(inode)->lock(filp, cmd, fl);
 	else
 		status = do_vfs_lock(filp, fl);
-	unlock_kernel();
 	return status;
 }
 
@@ -673,13 +669,11 @@ static int do_setlk(struct file *filp, int cmd, struct file_lock *fl)
 	if (status != 0)
 		goto out;
 
-	lock_kernel();
 	/* Use local locking if mounted with "-onolock" */
 	if (!(NFS_SERVER(inode)->flags & NFS_MOUNT_NONLM))
 		status = NFS_PROTO(inode)->lock(filp, cmd, fl);
 	else
 		status = do_vfs_lock(filp, fl);
-	unlock_kernel();
 	if (status < 0)
 		goto out;
 	/*
-- 
cgit v1.2.3-59-g8ed1b


From 55ae1aabfb108106dd095de2578ceef1c755a8b8 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:03 -0400
Subject: nfs41: Add backchannel processing support to RPC state machine

Adds rpc_run_bc_task() which is called by the NFS callback service to
process backchannel requests.  It performs similar work to rpc_run_task()
though "schedules" the backchannel task to be executed starting at the
call_trasmit state in the RPC state machine.

It also introduces some miscellaneous updates to the argument validation,
call_transmit, and transport cleanup functions to take into account
that there are now forechannel and backchannel tasks.

Backchannel requests do not carry an RPC message structure, since the
payload has already been XDR encoded using the existing NFSv4 callback
mechanism.

Introduce a new transmit state for the client to reply on to backchannel
requests.  This new state simply reserves the transport and issues the
reply.  In case of a connection related error, disconnects the transport and
drops the reply.  It requires the forechannel to re-establish the connection
and the server to retransmit the request, as stated in NFSv4.1 section
2.9.2 "Client and Server Transport Behavior".

Note: There is no need to loop attempting to reserve the transport.  If EAGAIN
is returned by xprt_prepare_transmit(), return with tk_status == 0,
setting tk_action to call_bc_transmit.  rpc_execute() will invoke it again
after the task is taken off the sleep queue.

[nfs41: rpc_run_bc_task() need not be exported outside RPC module]
[nfs41: New call_bc_transmit RPC state]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Backchannel: No need to loop in call_bc_transmit()]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[rpc_count_iostats incorrectly exits early]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Convert rpc_reply_expected() to inline function]
[Remove unnecessary BUG_ON()]
[Rename variable]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/sunrpc/sched.h |   2 +
 include/linux/sunrpc/xprt.h  |  12 +++++
 net/sunrpc/clnt.c            | 117 ++++++++++++++++++++++++++++++++++++++++++-
 net/sunrpc/stats.c           |   6 ++-
 net/sunrpc/sunrpc.h          |  37 ++++++++++++++
 net/sunrpc/xprt.c            |  38 +++++++++++---
 6 files changed, 203 insertions(+), 9 deletions(-)
 create mode 100644 net/sunrpc/sunrpc.h

diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h
index 177376880fab..401097781fc0 100644
--- a/include/linux/sunrpc/sched.h
+++ b/include/linux/sunrpc/sched.h
@@ -210,6 +210,8 @@ struct rpc_wait_queue {
  */
 struct rpc_task *rpc_new_task(const struct rpc_task_setup *);
 struct rpc_task *rpc_run_task(const struct rpc_task_setup *);
+struct rpc_task *rpc_run_bc_task(struct rpc_rqst *req,
+				const struct rpc_call_ops *ops);
 void		rpc_put_task(struct rpc_task *);
 void		rpc_exit_task(struct rpc_task *);
 void		rpc_release_calldata(const struct rpc_call_ops *, void *);
diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index beae030e80b5..55c6c37e249e 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -215,6 +215,18 @@ struct rpc_xprt {
 						/* buffer in use */
 #endif /* CONFIG_NFS_V4_1 */
 
+#if defined(CONFIG_NFS_V4_1)
+static inline int bc_prealloc(struct rpc_rqst *req)
+{
+	return test_bit(RPC_BC_PA_IN_USE, &req->rq_bc_pa_state);
+}
+#else
+static inline int bc_prealloc(struct rpc_rqst *req)
+{
+	return 0;
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 struct xprt_create {
 	int			ident;		/* XPRT_TRANSPORT identifier */
 	struct sockaddr *	srcaddr;	/* optional local address */
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index aca3ab6fc140..f3e93b8eb90f 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -36,7 +36,9 @@
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/rpc_pipe_fs.h>
 #include <linux/sunrpc/metrics.h>
+#include <linux/sunrpc/bc_xprt.h>
 
+#include "sunrpc.h"
 
 #ifdef RPC_DEBUG
 # define RPCDBG_FACILITY	RPCDBG_CALL
@@ -63,6 +65,9 @@ static void	call_decode(struct rpc_task *task);
 static void	call_bind(struct rpc_task *task);
 static void	call_bind_status(struct rpc_task *task);
 static void	call_transmit(struct rpc_task *task);
+#if defined(CONFIG_NFS_V4_1)
+static void	call_bc_transmit(struct rpc_task *task);
+#endif /* CONFIG_NFS_V4_1 */
 static void	call_status(struct rpc_task *task);
 static void	call_transmit_status(struct rpc_task *task);
 static void	call_refresh(struct rpc_task *task);
@@ -613,6 +618,50 @@ rpc_call_async(struct rpc_clnt *clnt, const struct rpc_message *msg, int flags,
 }
 EXPORT_SYMBOL_GPL(rpc_call_async);
 
+#if defined(CONFIG_NFS_V4_1)
+/**
+ * rpc_run_bc_task - Allocate a new RPC task for backchannel use, then run
+ * rpc_execute against it
+ * @ops: RPC call ops
+ */
+struct rpc_task *rpc_run_bc_task(struct rpc_rqst *req,
+					const struct rpc_call_ops *tk_ops)
+{
+	struct rpc_task *task;
+	struct xdr_buf *xbufp = &req->rq_snd_buf;
+	struct rpc_task_setup task_setup_data = {
+		.callback_ops = tk_ops,
+	};
+
+	dprintk("RPC: rpc_run_bc_task req= %p\n", req);
+	/*
+	 * Create an rpc_task to send the data
+	 */
+	task = rpc_new_task(&task_setup_data);
+	if (!task) {
+		xprt_free_bc_request(req);
+		goto out;
+	}
+	task->tk_rqstp = req;
+
+	/*
+	 * Set up the xdr_buf length.
+	 * This also indicates that the buffer is XDR encoded already.
+	 */
+	xbufp->len = xbufp->head[0].iov_len + xbufp->page_len +
+			xbufp->tail[0].iov_len;
+
+	task->tk_action = call_bc_transmit;
+	atomic_inc(&task->tk_count);
+	BUG_ON(atomic_read(&task->tk_count) != 2);
+	rpc_execute(task);
+
+out:
+	dprintk("RPC: rpc_run_bc_task: task= %p\n", task);
+	return task;
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 void
 rpc_call_start(struct rpc_task *task)
 {
@@ -1098,7 +1147,7 @@ call_transmit(struct rpc_task *task)
 	 * in order to allow access to the socket to other RPC requests.
 	 */
 	call_transmit_status(task);
-	if (task->tk_msg.rpc_proc->p_decode != NULL)
+	if (rpc_reply_expected(task))
 		return;
 	task->tk_action = rpc_exit_task;
 	rpc_wake_up_queued_task(&task->tk_xprt->pending, task);
@@ -1133,6 +1182,72 @@ call_transmit_status(struct rpc_task *task)
 	}
 }
 
+#if defined(CONFIG_NFS_V4_1)
+/*
+ * 5b.	Send the backchannel RPC reply.  On error, drop the reply.  In
+ * addition, disconnect on connectivity errors.
+ */
+static void
+call_bc_transmit(struct rpc_task *task)
+{
+	struct rpc_rqst *req = task->tk_rqstp;
+
+	BUG_ON(task->tk_status != 0);
+	task->tk_status = xprt_prepare_transmit(task);
+	if (task->tk_status == -EAGAIN) {
+		/*
+		 * Could not reserve the transport. Try again after the
+		 * transport is released.
+		 */
+		task->tk_status = 0;
+		task->tk_action = call_bc_transmit;
+		return;
+	}
+
+	task->tk_action = rpc_exit_task;
+	if (task->tk_status < 0) {
+		printk(KERN_NOTICE "RPC: Could not send backchannel reply "
+			"error: %d\n", task->tk_status);
+		return;
+	}
+
+	xprt_transmit(task);
+	xprt_end_transmit(task);
+	dprint_status(task);
+	switch (task->tk_status) {
+	case 0:
+		/* Success */
+		break;
+	case -EHOSTDOWN:
+	case -EHOSTUNREACH:
+	case -ENETUNREACH:
+	case -ETIMEDOUT:
+		/*
+		 * Problem reaching the server.  Disconnect and let the
+		 * forechannel reestablish the connection.  The server will
+		 * have to retransmit the backchannel request and we'll
+		 * reprocess it.  Since these ops are idempotent, there's no
+		 * need to cache our reply at this time.
+		 */
+		printk(KERN_NOTICE "RPC: Could not send backchannel reply "
+			"error: %d\n", task->tk_status);
+		xprt_conditional_disconnect(task->tk_xprt,
+			req->rq_connect_cookie);
+		break;
+	default:
+		/*
+		 * We were unable to reply and will have to drop the
+		 * request.  The server should reconnect and retransmit.
+		 */
+		BUG_ON(task->tk_status == -EAGAIN);
+		printk(KERN_NOTICE "RPC: Could not send backchannel reply "
+			"error: %d\n", task->tk_status);
+		break;
+	}
+	rpc_wake_up_queued_task(&req->rq_xprt->pending, task);
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 /*
  * 6.	Sort out the RPC call status
  */
diff --git a/net/sunrpc/stats.c b/net/sunrpc/stats.c
index 1ef6e46d9da2..8487aa0f1f5a 100644
--- a/net/sunrpc/stats.c
+++ b/net/sunrpc/stats.c
@@ -141,12 +141,14 @@ EXPORT_SYMBOL_GPL(rpc_free_iostats);
 void rpc_count_iostats(struct rpc_task *task)
 {
 	struct rpc_rqst *req = task->tk_rqstp;
-	struct rpc_iostats *stats = task->tk_client->cl_metrics;
+	struct rpc_iostats *stats;
 	struct rpc_iostats *op_metrics;
 	long rtt, execute, queue;
 
-	if (!stats || !req)
+	if (!task->tk_client || !task->tk_client->cl_metrics || !req)
 		return;
+
+	stats = task->tk_client->cl_metrics;
 	op_metrics = &stats[task->tk_msg.rpc_proc->p_statidx];
 
 	op_metrics->om_ops++;
diff --git a/net/sunrpc/sunrpc.h b/net/sunrpc/sunrpc.h
new file mode 100644
index 000000000000..5d9dd742264b
--- /dev/null
+++ b/net/sunrpc/sunrpc.h
@@ -0,0 +1,37 @@
+/******************************************************************************
+
+(c) 2008 NetApp.  All Rights Reserved.
+
+NetApp provides this source code under the GPL v2 License.
+The GPL v2 license is available at
+http://opensource.org/licenses/gpl-license.php.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
+CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
+PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+******************************************************************************/
+
+/*
+ * Functions and macros used internally by RPC
+ */
+
+#ifndef _NET_SUNRPC_SUNRPC_H
+#define _NET_SUNRPC_SUNRPC_H
+
+static inline int rpc_reply_expected(struct rpc_task *task)
+{
+	return (task->tk_msg.rpc_proc != NULL) &&
+		(task->tk_msg.rpc_proc->p_decode != NULL);
+}
+
+#endif /* _NET_SUNRPC_SUNRPC_H */
+
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 52739f82df1e..0eea2bfe111b 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -12,8 +12,9 @@
  *  -	Next, the caller puts together the RPC message, stuffs it into
  *	the request struct, and calls xprt_transmit().
  *  -	xprt_transmit sends the message and installs the caller on the
- *	transport's wait list. At the same time, it installs a timer that
- *	is run after the packet's timeout has expired.
+ *	transport's wait list. At the same time, if a reply is expected,
+ *	it installs a timer that is run after the packet's timeout has
+ *	expired.
  *  -	When a packet arrives, the data_ready handler walks the list of
  *	pending requests for that transport. If a matching XID is found, the
  *	caller is woken up, and the timer removed.
@@ -46,6 +47,8 @@
 #include <linux/sunrpc/clnt.h>
 #include <linux/sunrpc/metrics.h>
 
+#include "sunrpc.h"
+
 /*
  * Local variables
  */
@@ -873,7 +876,10 @@ void xprt_transmit(struct rpc_task *task)
 	dprintk("RPC: %5u xprt_transmit(%u)\n", task->tk_pid, req->rq_slen);
 
 	if (!req->rq_received) {
-		if (list_empty(&req->rq_list)) {
+		if (list_empty(&req->rq_list) && rpc_reply_expected(task)) {
+			/*
+			 * Add to the list only if we're expecting a reply
+			 */
 			spin_lock_bh(&xprt->transport_lock);
 			/* Update the softirq receive buffer */
 			memcpy(&req->rq_private_buf, &req->rq_rcv_buf,
@@ -908,8 +914,13 @@ void xprt_transmit(struct rpc_task *task)
 	/* Don't race with disconnect */
 	if (!xprt_connected(xprt))
 		task->tk_status = -ENOTCONN;
-	else if (!req->rq_received)
+	else if (!req->rq_received && rpc_reply_expected(task)) {
+		/*
+		 * Sleep on the pending queue since
+		 * we're expecting a reply.
+		 */
 		rpc_sleep_on(&xprt->pending, task, xprt_timer);
+	}
 	spin_unlock_bh(&xprt->transport_lock);
 }
 
@@ -982,11 +993,17 @@ static void xprt_request_init(struct rpc_task *task, struct rpc_xprt *xprt)
  */
 void xprt_release(struct rpc_task *task)
 {
-	struct rpc_xprt	*xprt = task->tk_xprt;
+	struct rpc_xprt	*xprt;
 	struct rpc_rqst	*req;
+	int is_bc_request;
 
 	if (!(req = task->tk_rqstp))
 		return;
+
+	/* Preallocated backchannel request? */
+	is_bc_request = bc_prealloc(req);
+
+	xprt = req->rq_xprt;
 	rpc_count_iostats(task);
 	spin_lock_bh(&xprt->transport_lock);
 	xprt->ops->release_xprt(xprt, task);
@@ -999,10 +1016,19 @@ void xprt_release(struct rpc_task *task)
 		mod_timer(&xprt->timer,
 				xprt->last_used + xprt->idle_timeout);
 	spin_unlock_bh(&xprt->transport_lock);
-	xprt->ops->buf_free(req->rq_buffer);
+	if (!bc_prealloc(req))
+		xprt->ops->buf_free(req->rq_buffer);
 	task->tk_rqstp = NULL;
 	if (req->rq_release_snd_buf)
 		req->rq_release_snd_buf(req);
+
+	/*
+	 * Early exit if this is a backchannel preallocated request.
+	 * There is no need to have it added to the RPC slot list.
+	 */
+	if (is_bc_request)
+		return;
+
 	memset(req, 0, sizeof(*req));	/* mark unused */
 
 	dprintk("RPC: %5u release request %p\n", task->tk_pid, req);
-- 
cgit v1.2.3-59-g8ed1b


From 0d90ba1cd416525c4825c111db862d8b15a02e9b Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:04 -0400
Subject: nfs41: Backchannel callback service helper routines

Executes the backchannel task on the RPC state machine using
the existing open connection previously established by the client.

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>

nfs41: Add bc_svc.o to sunrpc Makefile.

[nfs41: bc_send() does not need to be exported outside RPC module]
[nfs41: xprt_free_bc_request() need not be exported outside RPC module]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Update copyright]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/sunrpc/bc_xprt.h |  3 ++
 net/sunrpc/Makefile            |  2 +-
 net/sunrpc/bc_svc.c            | 81 ++++++++++++++++++++++++++++++++++++++++++
 net/sunrpc/xprtsock.c          |  3 ++
 4 files changed, 88 insertions(+), 1 deletion(-)
 create mode 100644 net/sunrpc/bc_svc.c

diff --git a/include/linux/sunrpc/bc_xprt.h b/include/linux/sunrpc/bc_xprt.h
index 5965ae4f902d..6508f0dc0eff 100644
--- a/include/linux/sunrpc/bc_xprt.h
+++ b/include/linux/sunrpc/bc_xprt.h
@@ -29,12 +29,15 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
 #include <linux/sunrpc/svcsock.h>
 #include <linux/sunrpc/xprt.h>
+#include <linux/sunrpc/sched.h>
 
 #ifdef CONFIG_NFS_V4_1
 struct rpc_rqst *xprt_alloc_bc_request(struct rpc_xprt *xprt);
 void xprt_free_bc_request(struct rpc_rqst *req);
 int xprt_setup_backchannel(struct rpc_xprt *, unsigned int min_reqs);
 void xprt_destroy_backchannel(struct rpc_xprt *, int max_reqs);
+void bc_release_request(struct rpc_task *);
+int bc_send(struct rpc_rqst *req);
 #else /* CONFIG_NFS_V4_1 */
 static inline int xprt_setup_backchannel(struct rpc_xprt *xprt,
 					 unsigned int min_reqs)
diff --git a/net/sunrpc/Makefile b/net/sunrpc/Makefile
index 4a01f9684b85..db73fd2a3f0e 100644
--- a/net/sunrpc/Makefile
+++ b/net/sunrpc/Makefile
@@ -13,6 +13,6 @@ sunrpc-y := clnt.o xprt.o socklib.o xprtsock.o sched.o \
 	    rpcb_clnt.o timer.o xdr.o \
 	    sunrpc_syms.o cache.o rpc_pipe.o \
 	    svc_xprt.o
-sunrpc-$(CONFIG_NFS_V4_1) += backchannel_rqst.o
+sunrpc-$(CONFIG_NFS_V4_1) += backchannel_rqst.o bc_svc.o
 sunrpc-$(CONFIG_PROC_FS) += stats.o
 sunrpc-$(CONFIG_SYSCTL) += sysctl.o
diff --git a/net/sunrpc/bc_svc.c b/net/sunrpc/bc_svc.c
new file mode 100644
index 000000000000..13f214f53120
--- /dev/null
+++ b/net/sunrpc/bc_svc.c
@@ -0,0 +1,81 @@
+/******************************************************************************
+
+(c) 2007 Network Appliance, Inc.  All Rights Reserved.
+(c) 2009 NetApp.  All Rights Reserved.
+
+NetApp provides this source code under the GPL v2 License.
+The GPL v2 license is available at
+http://opensource.org/licenses/gpl-license.php.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
+CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
+EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
+PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
+PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+******************************************************************************/
+
+/*
+ * The NFSv4.1 callback service helper routines.
+ * They implement the transport level processing required to send the
+ * reply over an existing open connection previously established by the client.
+ */
+
+#if defined(CONFIG_NFS_V4_1)
+
+#include <linux/module.h>
+
+#include <linux/sunrpc/xprt.h>
+#include <linux/sunrpc/sched.h>
+#include <linux/sunrpc/bc_xprt.h>
+
+#define RPCDBG_FACILITY	RPCDBG_SVCDSP
+
+void bc_release_request(struct rpc_task *task)
+{
+	struct rpc_rqst *req = task->tk_rqstp;
+
+	dprintk("RPC:       bc_release_request: task= %p\n", task);
+
+	/*
+	 * Release this request only if it's a backchannel
+	 * preallocated request
+	 */
+	if (!bc_prealloc(req))
+		return;
+	xprt_free_bc_request(req);
+}
+
+/* Empty callback ops */
+static const struct rpc_call_ops nfs41_callback_ops = {
+};
+
+
+/*
+ * Send the callback reply
+ */
+int bc_send(struct rpc_rqst *req)
+{
+	struct rpc_task *task;
+	int ret;
+
+	dprintk("RPC:       bc_send req= %p\n", req);
+	task = rpc_run_bc_task(req, &nfs41_callback_ops);
+	if (IS_ERR(task))
+		ret = PTR_ERR(task);
+	else {
+		BUG_ON(atomic_read(&task->tk_count) != 1);
+		ret = task->tk_status;
+		rpc_put_task(task);
+	}
+	return ret;
+	dprintk("RPC:       bc_send ret= %d \n", ret);
+}
+
+#endif /* CONFIG_NFS_V4_1 */
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index e3e3a57116fb..8a721867b601 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -2183,6 +2183,9 @@ static struct rpc_xprt_ops xs_tcp_ops = {
 	.buf_free		= rpc_free,
 	.send_request		= xs_tcp_send_request,
 	.set_retrans_timeout	= xprt_set_retrans_timeout_def,
+#if defined(CONFIG_NFS_V4_1)
+	.release_request	= bc_release_request,
+#endif /* CONFIG_NFS_V4_1 */
 	.close			= xs_tcp_close,
 	.destroy		= xs_destroy,
 	.print_stats		= xs_tcp_print_stats,
-- 
cgit v1.2.3-59-g8ed1b


From 1cad7ea6fe98dc414bd3df55275c147bd15ebf97 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:06 -0400
Subject: nfs41: Refactor svc_process()

net/sunrpc/svc.c:svc_process() is used by the NFSv4 callback service
to process RPC requests arriving over connections initiated by the
server.  NFSv4.1 supports callbacks over the backchannel on connections
initiated by the client.  This patch refactors svc_process() so that
common code can also be used by the backchannel.

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 net/sunrpc/svc.c | 80 +++++++++++++++++++++++++++++++++++---------------------
 1 file changed, 50 insertions(+), 30 deletions(-)

diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index 8847add6ca16..bfda66db2f4f 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -970,20 +970,18 @@ svc_printk(struct svc_rqst *rqstp, const char *fmt, ...)
 }
 
 /*
- * Process the RPC request.
+ * Common routine for processing the RPC request.
  */
-int
-svc_process(struct svc_rqst *rqstp)
+static int
+svc_process_common(struct svc_rqst *rqstp, struct kvec *argv, struct kvec *resv)
 {
 	struct svc_program	*progp;
 	struct svc_version	*versp = NULL;	/* compiler food */
 	struct svc_procedure	*procp = NULL;
-	struct kvec *		argv = &rqstp->rq_arg.head[0];
-	struct kvec *		resv = &rqstp->rq_res.head[0];
 	struct svc_serv		*serv = rqstp->rq_server;
 	kxdrproc_t		xdr;
 	__be32			*statp;
-	u32			dir, prog, vers, proc;
+	u32			prog, vers, proc;
 	__be32			auth_stat, rpc_stat;
 	int			auth_res;
 	__be32			*reply_statp;
@@ -993,19 +991,6 @@ svc_process(struct svc_rqst *rqstp)
 	if (argv->iov_len < 6*4)
 		goto err_short_len;
 
-	/* setup response xdr_buf.
-	 * Initially it has just one page
-	 */
-	rqstp->rq_resused = 1;
-	resv->iov_base = page_address(rqstp->rq_respages[0]);
-	resv->iov_len = 0;
-	rqstp->rq_res.pages = rqstp->rq_respages + 1;
-	rqstp->rq_res.len = 0;
-	rqstp->rq_res.page_base = 0;
-	rqstp->rq_res.page_len = 0;
-	rqstp->rq_res.buflen = PAGE_SIZE;
-	rqstp->rq_res.tail[0].iov_base = NULL;
-	rqstp->rq_res.tail[0].iov_len = 0;
 	/* Will be turned off only in gss privacy case: */
 	rqstp->rq_splice_ok = 1;
 	/* Will be turned off only when NFSv4 Sessions are used */
@@ -1014,17 +999,13 @@ svc_process(struct svc_rqst *rqstp)
 	/* Setup reply header */
 	rqstp->rq_xprt->xpt_ops->xpo_prep_reply_hdr(rqstp);
 
-	rqstp->rq_xid = svc_getu32(argv);
 	svc_putu32(resv, rqstp->rq_xid);
 
-	dir  = svc_getnl(argv);
 	vers = svc_getnl(argv);
 
 	/* First words of reply: */
 	svc_putnl(resv, 1);		/* REPLY */
 
-	if (dir != 0)		/* direction != CALL */
-		goto err_bad_dir;
 	if (vers != 2)		/* RPC version number */
 		goto err_bad_rpc;
 
@@ -1147,7 +1128,7 @@ svc_process(struct svc_rqst *rqstp)
  sendit:
 	if (svc_authorise(rqstp))
 		goto dropit;
-	return svc_send(rqstp);
+	return 1;		/* Caller can now send it */
 
  dropit:
 	svc_authorise(rqstp);	/* doesn't hurt to call this twice */
@@ -1161,12 +1142,6 @@ err_short_len:
 
 	goto dropit;			/* drop request */
 
-err_bad_dir:
-	svc_printk(rqstp, "bad direction %d, dropping request\n", dir);
-
-	serv->sv_stats->rpcbadfmt++;
-	goto dropit;			/* drop request */
-
 err_bad_rpc:
 	serv->sv_stats->rpcbadfmt++;
 	svc_putnl(resv, 1);	/* REJECT */
@@ -1219,6 +1194,51 @@ err_bad:
 }
 EXPORT_SYMBOL_GPL(svc_process);
 
+/*
+ * Process the RPC request.
+ */
+int
+svc_process(struct svc_rqst *rqstp)
+{
+	struct kvec		*argv = &rqstp->rq_arg.head[0];
+	struct kvec		*resv = &rqstp->rq_res.head[0];
+	struct svc_serv		*serv = rqstp->rq_server;
+	u32			dir;
+	int			error;
+
+	/*
+	 * Setup response xdr_buf.
+	 * Initially it has just one page
+	 */
+	rqstp->rq_resused = 1;
+	resv->iov_base = page_address(rqstp->rq_respages[0]);
+	resv->iov_len = 0;
+	rqstp->rq_res.pages = rqstp->rq_respages + 1;
+	rqstp->rq_res.len = 0;
+	rqstp->rq_res.page_base = 0;
+	rqstp->rq_res.page_len = 0;
+	rqstp->rq_res.buflen = PAGE_SIZE;
+	rqstp->rq_res.tail[0].iov_base = NULL;
+	rqstp->rq_res.tail[0].iov_len = 0;
+
+	rqstp->rq_xid = svc_getu32(argv);
+
+	dir  = svc_getnl(argv);
+	if (dir != 0) {
+		/* direction != CALL */
+		svc_printk(rqstp, "bad direction %d, dropping request\n", dir);
+		serv->sv_stats->rpcbadfmt++;
+		svc_drop(rqstp);
+		return 0;
+	}
+
+	error = svc_process_common(rqstp, argv, resv);
+	if (error <= 0)
+		return error;
+
+	return svc_send(rqstp);
+}
+
 /*
  * Return (transport-specific) limit on the rpc payload.
  */
-- 
cgit v1.2.3-59-g8ed1b


From 4d6bbb6233c9cf23822a2f66f8470c9f40854b77 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:07 -0400
Subject: nfs41: Backchannel bc_svc_process()

Implement the NFSv4.1 backchannel service.  Invokes the common callback
processing logic svc_process_common() to authenticate the call and
dispatch the appropriate NFSv4.1 XDR decoder and operation procedure.
It then invokes bc_send() to send the reply over the same connection.
bc_send() is implemented in a separate patch.

At this time there is no slot validation or reply cache handling.

[nfs41: Preallocate rpc_rqst receive buffer for handling callbacks]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Move bc_svc_process() declaration to correct patch]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/sunrpc/svc.h |  2 ++
 net/sunrpc/svc.c           | 49 ++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 51 insertions(+)

diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index 4a8afbd62007..16043c4a8bf4 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -419,6 +419,8 @@ int		   svc_set_num_threads(struct svc_serv *, struct svc_pool *, int);
 int		   svc_pool_stats_open(struct svc_serv *serv, struct file *file);
 void		   svc_destroy(struct svc_serv *);
 int		   svc_process(struct svc_rqst *);
+int		   bc_svc_process(struct svc_serv *, struct rpc_rqst *,
+			struct svc_rqst *);
 int		   svc_register(const struct svc_serv *, const int,
 				const unsigned short, const unsigned short);
 
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index bfda66db2f4f..06b52e465f47 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -25,6 +25,7 @@
 #include <linux/sunrpc/stats.h>
 #include <linux/sunrpc/svcsock.h>
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/bc_xprt.h>
 
 #define RPCDBG_FACILITY	RPCDBG_SVCDSP
 
@@ -1239,6 +1240,54 @@ svc_process(struct svc_rqst *rqstp)
 	return svc_send(rqstp);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+/*
+ * Process a backchannel RPC request that arrived over an existing
+ * outbound connection
+ */
+int
+bc_svc_process(struct svc_serv *serv, struct rpc_rqst *req,
+	       struct svc_rqst *rqstp)
+{
+	struct kvec	*argv = &rqstp->rq_arg.head[0];
+	struct kvec	*resv = &rqstp->rq_res.head[0];
+	int 		error;
+
+	/* Build the svc_rqst used by the common processing routine */
+	rqstp->rq_xid = req->rq_xid;
+	rqstp->rq_prot = req->rq_xprt->prot;
+	rqstp->rq_server = serv;
+
+	rqstp->rq_addrlen = sizeof(req->rq_xprt->addr);
+	memcpy(&rqstp->rq_addr, &req->rq_xprt->addr, rqstp->rq_addrlen);
+	memcpy(&rqstp->rq_arg, &req->rq_rcv_buf, sizeof(rqstp->rq_arg));
+	memcpy(&rqstp->rq_res, &req->rq_snd_buf, sizeof(rqstp->rq_res));
+
+	/* reset result send buffer "put" position */
+	resv->iov_len = 0;
+
+	if (rqstp->rq_prot != IPPROTO_TCP) {
+		printk(KERN_ERR "No support for Non-TCP transports!\n");
+		BUG();
+	}
+
+	/*
+	 * Skip the next two words because they've already been
+	 * processed in the trasport
+	 */
+	svc_getu32(argv);	/* XID */
+	svc_getnl(argv);	/* CALLDIR */
+
+	error = svc_process_common(rqstp, argv, resv);
+	if (error <= 0)
+		return error;
+
+	memcpy(&req->rq_snd_buf, &rqstp->rq_res, sizeof(req->rq_snd_buf));
+	return bc_send(req);
+}
+EXPORT_SYMBOL(bc_svc_process);
+#endif /* CONFIG_NFS_V4_1 */
+
 /*
  * Return (transport-specific) limit on the rpc payload.
  */
-- 
cgit v1.2.3-59-g8ed1b


From a43cde94feded0f65fce36330614691c650ae8fe Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:08 -0400
Subject: nfs41: Implement NFSv4.1 callback service process.

nfs41_callback_up() initializes the necessary queues and creates the new
nfs41_callback_svc thread.  This thread executes the callback service which
waits for requests to arrive on the svc_serv->sv_cb_list.

NFS41_BC_MIN_CALLBACKS is set to 1 because we expect callbacks to not
cause substantial latency.

The actual processing of the callback will be implemented as a separate patch.

There is only one NFSv4.1 callback service.  The first caller of
nfs4_callback_up() creates the service, subsequent callers increment a
reference count on the service.  The service is destroyed when the last
caller invokes nfs_callback_down().

The transport needs to hold a reference to the callback service in order
to invoke it during callback processing.  Currently this reference is only
obtained when the service is first created.  This is incorrect, since
subsequent registrations for other transports will leave the xprt->serv
pointer uninitialized, leading to an oops when a callback arrives on
the "unreferenced" transport.

This patch fixes the problem by ensuring that a reference to the service
is saved in xprt->serv, either because the service is created by this
invocation to nfs4_callback_up() or by a prior invocation.

Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Add a reference to svc_serv during callback service bring up]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Type check arguments of nfs_callback_up]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: save svc_serv in nfs_callback_info]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[Removal of ugly #ifdefs]
[nfs41: Update to removal of ugly #ifdefs]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback.c | 112 ++++++++++++++++++++++++++++++++++++++++++++++++++++--
 fs/nfs/callback.h |   7 ++++
 2 files changed, 115 insertions(+), 4 deletions(-)

diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index 4b1313eda6f5..470928898063 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -17,6 +17,9 @@
 #include <linux/freezer.h>
 #include <linux/kthread.h>
 #include <linux/sunrpc/svcauth_gss.h>
+#if defined(CONFIG_NFS_V4_1)
+#include <linux/sunrpc/bc_xprt.h>
+#endif
 
 #include <net/inet_sock.h>
 
@@ -28,6 +31,7 @@
 
 struct nfs_callback_data {
 	unsigned int users;
+	struct svc_serv *serv;
 	struct svc_rqst *rqst;
 	struct task_struct *task;
 };
@@ -131,6 +135,99 @@ out_err:
 	return ERR_PTR(ret);
 }
 
+#if defined(CONFIG_NFS_V4_1)
+/*
+ * The callback service for NFSv4.1 callbacks
+ */
+static int
+nfs41_callback_svc(void *vrqstp)
+{
+	struct svc_rqst *rqstp = vrqstp;
+	struct svc_serv *serv = rqstp->rq_server;
+	struct rpc_rqst *req;
+	int error;
+	DEFINE_WAIT(wq);
+
+	set_freezable();
+
+	/*
+	 * FIXME: do we really need to run this under the BKL? If so, please
+	 * add a comment about what it's intended to protect.
+	 */
+	lock_kernel();
+	while (!kthread_should_stop()) {
+		prepare_to_wait(&serv->sv_cb_waitq, &wq, TASK_INTERRUPTIBLE);
+		spin_lock_bh(&serv->sv_cb_lock);
+		if (!list_empty(&serv->sv_cb_list)) {
+			req = list_first_entry(&serv->sv_cb_list,
+					struct rpc_rqst, rq_bc_list);
+			list_del(&req->rq_bc_list);
+			spin_unlock_bh(&serv->sv_cb_lock);
+			dprintk("Invoking bc_svc_process()\n");
+			error = bc_svc_process(serv, req, rqstp);
+			dprintk("bc_svc_process() returned w/ error code= %d\n",
+				error);
+		} else {
+			spin_unlock_bh(&serv->sv_cb_lock);
+			schedule();
+		}
+		finish_wait(&serv->sv_cb_waitq, &wq);
+	}
+	unlock_kernel();
+	nfs_callback_info.task = NULL;
+	svc_exit_thread(rqstp);
+	return 0;
+}
+
+/*
+ * Bring up the NFSv4.1 callback service
+ */
+struct svc_rqst *
+nfs41_callback_up(struct svc_serv *serv, struct rpc_xprt *xprt)
+{
+	/*
+	 * Save the svc_serv in the transport so that it can
+	 * be referenced when the session backchannel is initialized
+	 */
+	xprt->bc_serv = serv;
+
+	INIT_LIST_HEAD(&serv->sv_cb_list);
+	spin_lock_init(&serv->sv_cb_lock);
+	init_waitqueue_head(&serv->sv_cb_waitq);
+	return svc_prepare_thread(serv, &serv->sv_pools[0]);
+}
+
+static inline int nfs_minorversion_callback_svc_setup(u32 minorversion,
+		struct svc_serv *serv, struct rpc_xprt *xprt,
+		struct svc_rqst **rqstpp, int (**callback_svc)(void *vrqstp))
+{
+	if (minorversion) {
+		*rqstpp = nfs41_callback_up(serv, xprt);
+		*callback_svc = nfs41_callback_svc;
+	}
+	return minorversion;
+}
+
+static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt,
+		struct nfs_callback_data *cb_info)
+{
+	if (minorversion)
+		xprt->bc_serv = cb_info->serv;
+}
+#else
+static inline int nfs_minorversion_callback_svc_setup(u32 minorversion,
+		struct svc_serv *serv, struct rpc_xprt *xprt,
+		struct svc_rqst **rqstpp, int (**callback_svc)(void *vrqstp))
+{
+	return 0;
+}
+
+static inline void nfs_callback_bc_serv(u32 minorversion, struct rpc_xprt *xprt,
+		struct nfs_callback_data *cb_info)
+{
+}
+#endif /* CONFIG_NFS_V4_1 */
+
 /*
  * Bring up the callback thread if it is not already up.
  */
@@ -141,21 +238,25 @@ int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt)
 	int (*callback_svc)(void *vrqstp);
 	char svc_name[12];
 	int ret = 0;
+	int minorversion_setup;
 
 	mutex_lock(&nfs_callback_mutex);
-	if (nfs_callback_info.users++ || nfs_callback_info.task != NULL)
+	if (nfs_callback_info.users++ || nfs_callback_info.task != NULL) {
+		nfs_callback_bc_serv(minorversion, xprt, &nfs_callback_info);
 		goto out;
+	}
 	serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, NULL);
 	if (!serv) {
 		ret = -ENOMEM;
 		goto out_err;
 	}
 
-	if (!minorversion) {
+	minorversion_setup =  nfs_minorversion_callback_svc_setup(minorversion,
+					serv, xprt, &rqstp, &callback_svc);
+	if (!minorversion_setup) {
+		/* v4.0 callback setup */
 		rqstp = nfs4_callback_up(serv);
 		callback_svc = nfs4_callback_svc;
-	} else {
-		BUG();	/* for now */
 	}
 
 	if (IS_ERR(rqstp)) {
@@ -166,6 +267,7 @@ int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt)
 	svc_sock_update_bufs(serv);
 
 	sprintf(svc_name, "nfsv4.%u-svc", minorversion);
+	nfs_callback_info.serv = serv;
 	nfs_callback_info.rqst = rqstp;
 	nfs_callback_info.task = kthread_run(callback_svc,
 					     nfs_callback_info.rqst,
@@ -173,6 +275,7 @@ int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt)
 	if (IS_ERR(nfs_callback_info.task)) {
 		ret = PTR_ERR(nfs_callback_info.task);
 		svc_exit_thread(nfs_callback_info.rqst);
+		nfs_callback_info.serv = NULL;
 		nfs_callback_info.rqst = NULL;
 		nfs_callback_info.task = NULL;
 		goto out_err;
@@ -205,6 +308,7 @@ void nfs_callback_down(void)
 	if (nfs_callback_info.users == 0 && nfs_callback_info.task != NULL) {
 		kthread_stop(nfs_callback_info.task);
 		svc_exit_thread(nfs_callback_info.rqst);
+		nfs_callback_info.serv = NULL;
 		nfs_callback_info.rqst = NULL;
 		nfs_callback_info.task = NULL;
 	}
diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 9b907f408b8d..29123b5604f2 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -70,6 +70,13 @@ extern void nfs_callback_down(void);
 #define nfs_callback_down()	do {} while(0)
 #endif
 
+/*
+ * nfs41: Callbacks are expected to not cause substantial latency,
+ * so we limit their concurrency to 1 by setting up the maximum number
+ * of slots for the backchannel.
+ */
+#define NFS41_BC_MIN_CALLBACKS 1
+
 extern unsigned int nfs_callback_set_tcpport;
 extern unsigned short nfs_callback_tcpport;
 extern unsigned short nfs_callback_tcpport6;
-- 
cgit v1.2.3-59-g8ed1b


From 7652e5a09ba319241607b22d9055ce93fd5b8039 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:23:09 -0400
Subject: nfs41: sunrpc: provide functions to create and destroy a svc_xprt for
 backchannel use

For nfs41 callbacks we need an svc_xprt to process requests coming up the
backchannel socket as rpc_rqst's that are transformed into svc_rqst's that
need a rq_xprt to be processed.

The svc_{udp,tcp}_create methods are too heavy for this job as svc_create_socket
creates an actual socket to listen on while for nfs41 we're "reusing" the
fore channel's socket.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/sunrpc/svcsock.h |  2 ++
 net/sunrpc/svcsock.c           | 39 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 41 insertions(+)

diff --git a/include/linux/sunrpc/svcsock.h b/include/linux/sunrpc/svcsock.h
index 483e10380aae..6bb1ec4ae310 100644
--- a/include/linux/sunrpc/svcsock.h
+++ b/include/linux/sunrpc/svcsock.h
@@ -42,6 +42,8 @@ int		svc_sock_names(char *buf, struct svc_serv *serv, char *toclose);
 int		svc_addsock(struct svc_serv *serv, int fd, char *name_return);
 void		svc_init_xprt_sock(void);
 void		svc_cleanup_xprt_sock(void);
+struct svc_xprt *svc_sock_create(struct svc_serv *serv, int prot);
+void		svc_sock_destroy(struct svc_xprt *);
 
 /*
  * svc_makesock socket characteristics
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 9d504234af4a..a2a03e500533 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -1327,3 +1327,42 @@ static void svc_sock_free(struct svc_xprt *xprt)
 		sock_release(svsk->sk_sock);
 	kfree(svsk);
 }
+
+/*
+ * Create a svc_xprt.
+ *
+ * For internal use only (e.g. nfsv4.1 backchannel).
+ * Callers should typically use the xpo_create() method.
+ */
+struct svc_xprt *svc_sock_create(struct svc_serv *serv, int prot)
+{
+	struct svc_sock *svsk;
+	struct svc_xprt *xprt = NULL;
+
+	dprintk("svc: %s\n", __func__);
+	svsk = kzalloc(sizeof(*svsk), GFP_KERNEL);
+	if (!svsk)
+		goto out;
+
+	xprt = &svsk->sk_xprt;
+	if (prot == IPPROTO_TCP)
+		svc_xprt_init(&svc_tcp_class, xprt, serv);
+	else if (prot == IPPROTO_UDP)
+		svc_xprt_init(&svc_udp_class, xprt, serv);
+	else
+		BUG();
+out:
+	dprintk("svc: %s return %p\n", __func__, xprt);
+	return xprt;
+}
+EXPORT_SYMBOL_GPL(svc_sock_create);
+
+/*
+ * Destroy a svc_sock.
+ */
+void svc_sock_destroy(struct svc_xprt *xprt)
+{
+	if (xprt)
+		kfree(container_of(xprt, struct svc_sock, sk_xprt));
+}
+EXPORT_SYMBOL_GPL(svc_sock_destroy);
-- 
cgit v1.2.3-59-g8ed1b


From 9c9f3f5fa62cc4959e4d4d1cf1ec74f2d6ac1197 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:23:10 -0400
Subject: nfs41: sunrpc: add a struct svc_xprt pointer to struct svc_serv for
 backchannel use

This svc_xprt is passed on to the callback service thread to be later used
to processes incoming svc_rqst's

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/sunrpc/svc.h | 1 +
 net/sunrpc/svc.c           | 4 ++++
 2 files changed, 5 insertions(+)

diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h
index 16043c4a8bf4..ea8009695c69 100644
--- a/include/linux/sunrpc/svc.h
+++ b/include/linux/sunrpc/svc.h
@@ -103,6 +103,7 @@ struct svc_serv {
 	spinlock_t		sv_cb_lock;	/* protects the svc_cb_list */
 	wait_queue_head_t	sv_cb_waitq;	/* sleep here if there are no
 						 * entries in the svc_cb_list */
+	struct svc_xprt		*bc_xprt;
 #endif /* CONFIG_NFS_V4_1 */
 };
 
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index 06b52e465f47..b35048fabe22 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -487,6 +487,10 @@ svc_destroy(struct svc_serv *serv)
 	if (svc_serv_is_pooled(serv))
 		svc_pool_map_put();
 
+#if defined(CONFIG_NFS_V4_1)
+	svc_sock_destroy(serv->bc_xprt);
+#endif /* CONFIG_NFS_V4_1 */
+
 	svc_unregister(serv);
 	kfree(serv->sv_pools);
 	kfree(serv);
-- 
cgit v1.2.3-59-g8ed1b


From 8f975242352e92898dc641ebff0d24808f39848a Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:23:11 -0400
Subject: nfs41: create a svc_xprt for nfs41 callback thread and use for
 incoming callbacks

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback.c | 17 ++++++++++++++++-
 net/sunrpc/svc.c  |  1 +
 2 files changed, 17 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index 470928898063..37815f3216aa 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -185,16 +185,31 @@ nfs41_callback_svc(void *vrqstp)
 struct svc_rqst *
 nfs41_callback_up(struct svc_serv *serv, struct rpc_xprt *xprt)
 {
+	struct svc_xprt *bc_xprt;
+	struct svc_rqst *rqstp = ERR_PTR(-ENOMEM);
+
+	dprintk("--> %s\n", __func__);
+	/* Create a svc_sock for the service */
+	bc_xprt = svc_sock_create(serv, xprt->prot);
+	if (!bc_xprt)
+		goto out;
+
 	/*
 	 * Save the svc_serv in the transport so that it can
 	 * be referenced when the session backchannel is initialized
 	 */
+	serv->bc_xprt = bc_xprt;
 	xprt->bc_serv = serv;
 
 	INIT_LIST_HEAD(&serv->sv_cb_list);
 	spin_lock_init(&serv->sv_cb_lock);
 	init_waitqueue_head(&serv->sv_cb_waitq);
-	return svc_prepare_thread(serv, &serv->sv_pools[0]);
+	rqstp = svc_prepare_thread(serv, &serv->sv_pools[0]);
+	if (IS_ERR(rqstp))
+		svc_sock_destroy(bc_xprt);
+out:
+	dprintk("--> %s return %p\n", __func__, rqstp);
+	return rqstp;
 }
 
 static inline int nfs_minorversion_callback_svc_setup(u32 minorversion,
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index b35048fabe22..6b90ce439c00 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -1258,6 +1258,7 @@ bc_svc_process(struct svc_serv *serv, struct rpc_rqst *req,
 	int 		error;
 
 	/* Build the svc_rqst used by the common processing routine */
+	rqstp->rq_xprt = serv->bc_xprt;
 	rqstp->rq_xid = req->rq_xid;
 	rqstp->rq_prot = req->rq_xprt->prot;
 	rqstp->rq_server = serv;
-- 
cgit v1.2.3-59-g8ed1b


From e82dc22dac6525a2f365a1d53c0483252d4aa38e Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:23:14 -0400
Subject: nfs41: Allow NFSv4 and NFSv4.1 callback services to coexist

Tracks the nfs_callback_info for both versions, enabling the callback
service for v4 and v4.1 to run concurrently and be stopped independently
of each other.

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback.c | 50 ++++++++++++++++++++++++--------------------------
 fs/nfs/callback.h |  7 ++-----
 fs/nfs/client.c   |  2 +-
 3 files changed, 27 insertions(+), 32 deletions(-)

diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index 37815f3216aa..e69b8f61189e 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -36,7 +36,7 @@ struct nfs_callback_data {
 	struct task_struct *task;
 };
 
-static struct nfs_callback_data nfs_callback_info;
+static struct nfs_callback_data nfs_callback_info[NFS4_MAX_MINOR_VERSION + 1];
 static DEFINE_MUTEX(nfs_callback_mutex);
 static struct svc_program nfs4_callback_program;
 
@@ -60,7 +60,7 @@ module_param_call(callback_tcpport, param_set_port, param_get_int,
 		 &nfs_callback_set_tcpport, 0644);
 
 /*
- * This is the callback kernel thread.
+ * This is the NFSv4 callback kernel thread.
  */
 static int
 nfs4_callback_svc(void *vrqstp)
@@ -174,8 +174,6 @@ nfs41_callback_svc(void *vrqstp)
 		finish_wait(&serv->sv_cb_waitq, &wq);
 	}
 	unlock_kernel();
-	nfs_callback_info.task = NULL;
-	svc_exit_thread(rqstp);
 	return 0;
 }
 
@@ -251,13 +249,14 @@ int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt)
 	struct svc_serv *serv = NULL;
 	struct svc_rqst *rqstp;
 	int (*callback_svc)(void *vrqstp);
+	struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion];
 	char svc_name[12];
 	int ret = 0;
 	int minorversion_setup;
 
 	mutex_lock(&nfs_callback_mutex);
-	if (nfs_callback_info.users++ || nfs_callback_info.task != NULL) {
-		nfs_callback_bc_serv(minorversion, xprt, &nfs_callback_info);
+	if (cb_info->users++ || cb_info->task != NULL) {
+		nfs_callback_bc_serv(minorversion, xprt, cb_info);
 		goto out;
 	}
 	serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, NULL);
@@ -282,17 +281,14 @@ int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt)
 	svc_sock_update_bufs(serv);
 
 	sprintf(svc_name, "nfsv4.%u-svc", minorversion);
-	nfs_callback_info.serv = serv;
-	nfs_callback_info.rqst = rqstp;
-	nfs_callback_info.task = kthread_run(callback_svc,
-					     nfs_callback_info.rqst,
-					     svc_name);
-	if (IS_ERR(nfs_callback_info.task)) {
-		ret = PTR_ERR(nfs_callback_info.task);
-		svc_exit_thread(nfs_callback_info.rqst);
-		nfs_callback_info.serv = NULL;
-		nfs_callback_info.rqst = NULL;
-		nfs_callback_info.task = NULL;
+	cb_info->serv = serv;
+	cb_info->rqst = rqstp;
+	cb_info->task = kthread_run(callback_svc, cb_info->rqst, svc_name);
+	if (IS_ERR(cb_info->task)) {
+		ret = PTR_ERR(cb_info->task);
+		svc_exit_thread(cb_info->rqst);
+		cb_info->rqst = NULL;
+		cb_info->task = NULL;
 		goto out_err;
 	}
 out:
@@ -309,23 +305,25 @@ out:
 out_err:
 	dprintk("NFS: Couldn't create callback socket or server thread; "
 		"err = %d\n", ret);
-	nfs_callback_info.users--;
+	cb_info->users--;
 	goto out;
 }
 
 /*
  * Kill the callback thread if it's no longer being used.
  */
-void nfs_callback_down(void)
+void nfs_callback_down(int minorversion)
 {
+	struct nfs_callback_data *cb_info = &nfs_callback_info[minorversion];
+
 	mutex_lock(&nfs_callback_mutex);
-	nfs_callback_info.users--;
-	if (nfs_callback_info.users == 0 && nfs_callback_info.task != NULL) {
-		kthread_stop(nfs_callback_info.task);
-		svc_exit_thread(nfs_callback_info.rqst);
-		nfs_callback_info.serv = NULL;
-		nfs_callback_info.rqst = NULL;
-		nfs_callback_info.task = NULL;
+	cb_info->users--;
+	if (cb_info->users == 0 && cb_info->task != NULL) {
+		kthread_stop(cb_info->task);
+		svc_exit_thread(cb_info->rqst);
+		cb_info->serv = NULL;
+		cb_info->rqst = NULL;
+		cb_info->task = NULL;
 	}
 	mutex_unlock(&nfs_callback_mutex);
 }
diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 29123b5604f2..88d2e1135b5a 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -64,11 +64,8 @@ extern __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy);
 
 #ifdef CONFIG_NFS_V4
 extern int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt);
-extern void nfs_callback_down(void);
-#else
-#define nfs_callback_up()	(0)
-#define nfs_callback_down()	do {} while(0)
-#endif
+extern void nfs_callback_down(int minorversion);
+#endif /* CONFIG_NFS_V4 */
 
 /*
  * nfs41: Callbacks are expected to not cause substantial latency,
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index df2b40d1883d..ac6575c4da31 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -181,7 +181,7 @@ static void nfs4_destroy_callback(struct nfs_client *clp)
 {
 #ifdef CONFIG_NFS_V4
 	if (__test_and_clear_bit(NFS_CS_CALLBACK, &clp->cl_res_state))
-		nfs_callback_down();
+		nfs_callback_down(clp->cl_minorversion);
 #endif /* CONFIG_NFS_V4 */
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 0b5b7ae0a853c91015bb3b1729166ca65f693322 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:23:15 -0400
Subject: nfs41: Setup the backchannel

The NFS v4.1 callback service has already been setup, and
rpc_xprt->serv points to the svc_serv structure describing it.
Invoke the xprt_setup_backchannel() initialization to pre-
allocate the necessary backchannel structures.

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: change nfs4_put_session(nfs4_session**) to nfs4_destroy_session(nfs_session*)]
Signed-off-by: Alexandros Batsakis <Alexandros.Batsakis@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[moved xprt_setup_backchannel from nfs4_init_session to nfs4_init_backchannel]
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/client.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index ac6575c4da31..4f75ec593be8 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -37,6 +37,7 @@
 #include <linux/in6.h>
 #include <net/ipv6.h>
 #include <linux/nfs_xdr.h>
+#include <linux/sunrpc/bc_xprt.h>
 
 #include <asm/system.h>
 
@@ -1096,6 +1097,14 @@ static int nfs4_init_callback(struct nfs_client *clp)
 	int error;
 
 	if (clp->rpc_ops->version == 4) {
+		if (nfs4_has_session(clp)) {
+			error = xprt_setup_backchannel(
+						clp->cl_rpcclient->cl_xprt,
+						NFS41_BC_MIN_CALLBACKS);
+			if (error < 0)
+				return error;
+		}
+
 		error = nfs_callback_up(clp->cl_minorversion,
 					clp->cl_rpcclient->cl_xprt);
 		if (error < 0) {
-- 
cgit v1.2.3-59-g8ed1b


From 0f91421e8ea986c990c01bf091ceff1d59c165ff Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:23:16 -0400
Subject: nfs41: Client indicates presence of NFSv4.1 callback channel.

Set the SESSION4_BACK_CHAN flag to indicate the client supports a backchannel.

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index f1f087b483ce..4213c8814997 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4598,7 +4598,7 @@ static int _nfs4_proc_create_session(struct nfs_client *clp)
 	int status;
 
 	nfs4_init_channel_attrs(&args);
-	args.flags = (SESSION4_PERSIST);
+	args.flags = (SESSION4_PERSIST | SESSION4_BACK_CHAN);
 
 	status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0);
 
-- 
cgit v1.2.3-59-g8ed1b


From 343952fa5aac888934ffc203abed26a823400eb6 Mon Sep 17 00:00:00 2001
From: Rahul Iyer <iyer@netapp.com>
Date: Wed, 1 Apr 2009 09:23:17 -0400
Subject: nfs41: Get the rpc_xprt * from the rpc_rqst instead of the rpc_clnt.

Obtain the rpc_xprt from the rpc_rqst so that calls and callback replies
can both use the same code path.  A client needs the rpc_xprt in order
to reply to a callback.

Signed-off-by: Rahul Iyer <iyer@netapp.com>
Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 net/sunrpc/xprt.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 0eea2bfe111b..c144611223fc 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -195,8 +195,8 @@ EXPORT_SYMBOL_GPL(xprt_load_transport);
  */
 int xprt_reserve_xprt(struct rpc_task *task)
 {
-	struct rpc_xprt	*xprt = task->tk_xprt;
 	struct rpc_rqst *req = task->tk_rqstp;
+	struct rpc_xprt	*xprt = req->rq_xprt;
 
 	if (test_and_set_bit(XPRT_LOCKED, &xprt->state)) {
 		if (task == xprt->snd_task)
@@ -858,7 +858,7 @@ out_unlock:
 
 void xprt_end_transmit(struct rpc_task *task)
 {
-	xprt_release_write(task->tk_xprt, task);
+	xprt_release_write(task->tk_rqstp->rq_xprt, task);
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 5a0ffe544c54f62be99751e369f4d0f44bd5ee19 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Wed, 1 Apr 2009 09:23:18 -0400
Subject: nfs41: Release backchannel resources associated with session

Frees the preallocated backchannel resources that are associated with
this session when the session is destroyed.

A backchannel is currently created once per session. Destroy the backchannel
only when the session is destroyed.

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Andy Adamson<andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c | 6 ++++++
 fs/nfs/super.c    | 1 -
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 4213c8814997..5da939d577d5 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -49,6 +49,7 @@
 #include <linux/namei.h>
 #include <linux/mount.h>
 #include <linux/module.h>
+#include <linux/sunrpc/bc_xprt.h>
 
 #include "nfs4_fs.h"
 #include "delegation.h"
@@ -4481,6 +4482,11 @@ struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
 
 void nfs4_destroy_session(struct nfs4_session *session)
 {
+	nfs4_proc_destroy_session(session);
+	dprintk("%s Destroy backchannel for xprt %p\n",
+		__func__, session->clp->cl_rpcclient->cl_xprt);
+	xprt_destroy_backchannel(session->clp->cl_rpcclient->cl_xprt,
+				NFS41_BC_MIN_CALLBACKS);
 	nfs4_destroy_slot_table(session);
 	kfree(session);
 }
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 73db6f8c145a..69804a8245f6 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -2488,7 +2488,6 @@ static void nfs4_kill_super(struct super_block *sb)
 	dprintk("--> %s\n", __func__);
 	nfs_super_return_all_delegations(sb);
 	kill_anon_super(sb);
-
 	nfs4_renewd_prepare_shutdown(server);
 	nfs_fscache_release_super_cookie(sb);
 	nfs_free_server(server);
-- 
cgit v1.2.3-59-g8ed1b


From b8f2ef84b079ceb22b42d6d353609db7eb8efa93 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:23:19 -0400
Subject: nfs41: store minorversion in cb_compound_hdr_arg

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback.h     |  1 +
 fs/nfs/callback_xdr.c | 12 +++++++-----
 2 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 88d2e1135b5a..51b15c2d3a56 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -26,6 +26,7 @@ enum nfs4_callback_opnum {
 struct cb_compound_hdr_arg {
 	unsigned int taglen;
 	const char *tag;
+	unsigned int minorversion;
 	unsigned int callback_ident;
 	unsigned nops;
 };
diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index dd0ef34b5845..91f6f74ffea7 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -132,7 +132,6 @@ static __be32 decode_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid)
 static __be32 decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound_hdr_arg *hdr)
 {
 	__be32 *p;
-	unsigned int minor_version;
 	__be32 status;
 
 	status = decode_string(xdr, &hdr->taglen, &hdr->tag);
@@ -147,15 +146,18 @@ static __be32 decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound
 	p = read_buf(xdr, 12);
 	if (unlikely(p == NULL))
 		return htonl(NFS4ERR_RESOURCE);
-	minor_version = ntohl(*p++);
+	hdr->minorversion = ntohl(*p++);
 	/* Check minor version is zero. */
-	if (minor_version != 0) {
-		printk(KERN_WARNING "%s: NFSv4 server callback with illegal minor version %u!\n",
-				__func__, minor_version);
+	if (hdr->minorversion != 0) {
+		printk(KERN_WARNING "%s: NFSv4 server callback with "
+			"illegal minor version %u!\n",
+			__func__, hdr->minorversion);
 		return htonl(NFS4ERR_MINOR_VERS_MISMATCH);
 	}
 	hdr->callback_ident = ntohl(*p++);
 	hdr->nops = ntohl(*p);
+	dprintk("%s: minorversion %d nops %d\n", __func__,
+		hdr->minorversion, hdr->nops);
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 48a9e2d2289ed4b5053b3986d99709e9b07a0923 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:23:20 -0400
Subject: nfs41: decode minorversion 1 cb_compound header

decode cb_compound header conforming to
http://tools.ietf.org/html/draft-ietf-nfsv4-minorversion1-26

Get rid of cb_compound_hdr_arg.callback_ident

callback_ident is not used anywhere so we shouldn't waste any memory to
store it.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: no need to break read_buf in decode_compound_hdr_arg]
See http://linux-nfs.org/pipermail/pnfs/2009-June/007844.html
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback.h     | 1 -
 fs/nfs/callback_xdr.c | 7 ++++---
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 51b15c2d3a56..a8ffa9d6ea2f 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -27,7 +27,6 @@ struct cb_compound_hdr_arg {
 	unsigned int taglen;
 	const char *tag;
 	unsigned int minorversion;
-	unsigned int callback_ident;
 	unsigned nops;
 };
 
diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index 91f6f74ffea7..f6cc79b1d1f3 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -147,14 +147,15 @@ static __be32 decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound
 	if (unlikely(p == NULL))
 		return htonl(NFS4ERR_RESOURCE);
 	hdr->minorversion = ntohl(*p++);
-	/* Check minor version is zero. */
-	if (hdr->minorversion != 0) {
+	/* Check minor version is zero or one. */
+	if (hdr->minorversion <= 1) {
+		p++;	/* skip callback_ident */
+	} else {
 		printk(KERN_WARNING "%s: NFSv4 server callback with "
 			"illegal minor version %u!\n",
 			__func__, hdr->minorversion);
 		return htonl(NFS4ERR_MINOR_VERS_MISMATCH);
 	}
-	hdr->callback_ident = ntohl(*p++);
 	hdr->nops = ntohl(*p);
 	dprintk("%s: minorversion %d nops %d\n", __func__,
 		hdr->minorversion, hdr->nops);
-- 
cgit v1.2.3-59-g8ed1b


From 45377b94edea18f53dd3ba4d46d94de4bb7c00b5 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:23:21 -0400
Subject: nfs41: callback numbers definitions

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index a8ffa9d6ea2f..7ba42b0d16c9 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -20,6 +20,16 @@ enum nfs4_callback_procnum {
 enum nfs4_callback_opnum {
 	OP_CB_GETATTR = 3,
 	OP_CB_RECALL  = 4,
+/* Callback operations new to NFSv4.1 */
+	OP_CB_LAYOUTRECALL  = 5,
+	OP_CB_NOTIFY        = 6,
+	OP_CB_PUSH_DELEG    = 7,
+	OP_CB_RECALL_ANY    = 8,
+	OP_CB_RECALLABLE_OBJ_AVAIL = 9,
+	OP_CB_RECALL_SLOT   = 10,
+	OP_CB_SEQUENCE      = 11,
+	OP_CB_WANTS_CANCELLED = 12,
+	OP_CB_NOTIFY_LOCK   = 13,
 	OP_CB_ILLEGAL = 10044,
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 34bc47c941a074f91c2455b4b08503d02c74b878 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:23:22 -0400
Subject: nfs41: consider minorversion in callback_xdr:process_op

Note that this patch changes the nfsv4.0 behavior also when
CONFIG_NFS_V4_1 is not defined where NFS4ERR_MINOR_VERS_MISMATCH
will be returned if the client received a CB_COMPOUND
with minorversion != 0.  Previously, it would have
returned NFS4ERR_OP_ILLEGAL for CB_SEQUENCE.
(or if the server is broken and sent OP_CB_GETATTR or OP_CB_RECALL
with minorversion!=0, they would have been processed normally.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: refactor op preprocessing out of process_op]
See http://linux-nfs.org/pipermail/pnfs/2009-June/007845.html
[nfs41: define CB_NOTIFY_DEVICEID as not supported]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback.h     |  1 +
 fs/nfs/callback_xdr.c | 85 ++++++++++++++++++++++++++++++++++++++++++---------
 2 files changed, 72 insertions(+), 14 deletions(-)

diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 7ba42b0d16c9..80fd8a82964f 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -30,6 +30,7 @@ enum nfs4_callback_opnum {
 	OP_CB_SEQUENCE      = 11,
 	OP_CB_WANTS_CANCELLED = 12,
 	OP_CB_NOTIFY_LOCK   = 13,
+	OP_CB_NOTIFY_DEVICEID = 14,
 	OP_CB_ILLEGAL = 10044,
 };
 
diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index f6cc79b1d1f3..41c5be1c1be5 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -356,31 +356,87 @@ out:
 	return status;
 }
 
-static __be32 process_op(struct svc_rqst *rqstp,
+#if defined(CONFIG_NFS_V4_1)
+
+static __be32
+preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op)
+{
+	switch (op_nr) {
+	case OP_CB_GETATTR:
+	case OP_CB_RECALL:
+		*op = &callback_ops[op_nr];
+		break;
+
+	case OP_CB_LAYOUTRECALL:
+	case OP_CB_NOTIFY_DEVICEID:
+	case OP_CB_NOTIFY:
+	case OP_CB_PUSH_DELEG:
+	case OP_CB_RECALL_ANY:
+	case OP_CB_RECALLABLE_OBJ_AVAIL:
+	case OP_CB_RECALL_SLOT:
+	case OP_CB_SEQUENCE:
+	case OP_CB_WANTS_CANCELLED:
+	case OP_CB_NOTIFY_LOCK:
+		return htonl(NFS4ERR_NOTSUPP);
+
+	default:
+		return htonl(NFS4ERR_OP_ILLEGAL);
+	}
+
+	return htonl(NFS_OK);
+}
+
+#else /* CONFIG_NFS_V4_1 */
+
+static __be32
+preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op)
+{
+	return htonl(NFS4ERR_MINOR_VERS_MISMATCH);
+}
+
+#endif /* CONFIG_NFS_V4_1 */
+
+static __be32
+preprocess_nfs4_op(unsigned int op_nr, struct callback_op **op)
+{
+	switch (op_nr) {
+	case OP_CB_GETATTR:
+	case OP_CB_RECALL:
+		*op = &callback_ops[op_nr];
+		break;
+	default:
+		return htonl(NFS4ERR_OP_ILLEGAL);
+	}
+
+	return htonl(NFS_OK);
+}
+
+static __be32 process_op(uint32_t minorversion, int nop,
+		struct svc_rqst *rqstp,
 		struct xdr_stream *xdr_in, void *argp,
 		struct xdr_stream *xdr_out, void *resp)
 {
 	struct callback_op *op = &callback_ops[0];
 	unsigned int op_nr = OP_CB_ILLEGAL;
-	__be32 status = 0;
+	__be32 status;
 	long maxlen;
 	__be32 res;
 
 	dprintk("%s: start\n", __func__);
 	status = decode_op_hdr(xdr_in, &op_nr);
-	if (likely(status == 0)) {
-		switch (op_nr) {
-			case OP_CB_GETATTR:
-			case OP_CB_RECALL:
-				op = &callback_ops[op_nr];
-				break;
-			default:
-				op_nr = OP_CB_ILLEGAL;
-				op = &callback_ops[0];
-				status = htonl(NFS4ERR_OP_ILLEGAL);
-		}
+	if (unlikely(status)) {
+		status = htonl(NFS4ERR_OP_ILLEGAL);
+		goto out;
 	}
 
+	dprintk("%s: minorversion=%d nop=%d op_nr=%u\n",
+		__func__, minorversion, nop, op_nr);
+
+	status = minorversion ? preprocess_nfs41_op(nop, op_nr, &op) :
+				preprocess_nfs4_op(op_nr, &op);
+	if (status == htonl(NFS4ERR_OP_ILLEGAL))
+		op_nr = OP_CB_ILLEGAL;
+out:
 	maxlen = xdr_out->end - xdr_out->p;
 	if (maxlen > 0 && maxlen < PAGE_SIZE) {
 		if (likely(status == 0 && op->decode_args != NULL))
@@ -428,7 +484,8 @@ static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *r
 		return rpc_system_err;
 
 	while (status == 0 && nops != hdr_arg.nops) {
-		status = process_op(rqstp, &xdr_in, argp, &xdr_out, resp);
+		status = process_op(hdr_arg.minorversion, nops,
+				    rqstp, &xdr_in, argp, &xdr_out, resp);
 		nops++;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 2d9b9ec344b19b7b65c732b7000114df57684140 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:23:24 -0400
Subject: nfs41: cb_sequence protocol level data structures

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback.h | 35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 80fd8a82964f..2b933ce064dd 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -70,6 +70,41 @@ struct cb_recallargs {
 	uint32_t truncate;
 };
 
+#if defined(CONFIG_NFS_V4_1)
+
+struct referring_call {
+	uint32_t			rc_sequenceid;
+	uint32_t			rc_slotid;
+};
+
+struct referring_call_list {
+	struct nfs4_sessionid		rcl_sessionid;
+	uint32_t			rcl_nrefcalls;
+	struct referring_call 		*rcl_refcalls;
+};
+
+struct cb_sequenceargs {
+	struct sockaddr_in		*csa_addr;
+	struct nfs4_sessionid		csa_sessionid;
+	uint32_t			csa_sequenceid;
+	uint32_t			csa_slotid;
+	uint32_t			csa_highestslotid;
+	uint32_t			csa_cachethis;
+	uint32_t			csa_nrclists;
+	struct referring_call_list	*csa_rclists;
+};
+
+struct cb_sequenceres {
+	uint32_t			csr_status;
+	struct nfs4_sessionid		csr_sessionid;
+	uint32_t			csr_sequenceid;
+	uint32_t			csr_slotid;
+	uint32_t			csr_highestslotid;
+	uint32_t			csr_target_highestslotid;
+};
+
+#endif /* CONFIG_NFS_V4_1 */
+
 extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res);
 extern __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy);
 
-- 
cgit v1.2.3-59-g8ed1b


From d49433e1e3bb144a5752ce2a8ba1139dc519df1a Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:23:25 -0400
Subject: nfs41: cb_sequence proc implementation

Currently, just free up any referring calls information.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: fix csr_{,target}highestslotid]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback.h      |  4 ++++
 fs/nfs/callback_proc.c | 28 ++++++++++++++++++++++++++++
 2 files changed, 32 insertions(+)

diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 2b933ce064dd..576f51f4dc52 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -103,6 +103,9 @@ struct cb_sequenceres {
 	uint32_t			csr_target_highestslotid;
 };
 
+extern unsigned nfs4_callback_sequence(struct cb_sequenceargs *args,
+				       struct cb_sequenceres *res);
+
 #endif /* CONFIG_NFS_V4_1 */
 
 extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res);
@@ -119,6 +122,7 @@ extern void nfs_callback_down(int minorversion);
  * of slots for the backchannel.
  */
 #define NFS41_BC_MIN_CALLBACKS 1
+#define NFS41_BC_MAX_CALLBACKS 1
 
 extern unsigned int nfs_callback_set_tcpport;
 extern unsigned short nfs_callback_tcpport;
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index f7e83e23cf9f..f731bbe7ac6a 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -101,3 +101,31 @@ out:
 	dprintk("%s: exit with status = %d\n", __func__, ntohl(res));
 	return res;
 }
+
+#if defined(CONFIG_NFS_V4_1)
+
+/* FIXME: validate args->cbs_{sequence,slot}id */
+/* FIXME: referring calls should be processed */
+unsigned nfs4_callback_sequence(struct cb_sequenceargs *args,
+				struct cb_sequenceres *res)
+{
+	int i;
+	unsigned status = 0;
+
+	for (i = 0; i < args->csa_nrclists; i++)
+		kfree(args->csa_rclists[i].rcl_refcalls);
+	kfree(args->csa_rclists);
+
+	memcpy(&res->csr_sessionid, &args->csa_sessionid,
+	       sizeof(res->csr_sessionid));
+	res->csr_sequenceid = args->csa_sequenceid;
+	res->csr_slotid = args->csa_slotid;
+	res->csr_highestslotid = NFS41_BC_MAX_CALLBACKS - 1;
+	res->csr_target_highestslotid = NFS41_BC_MAX_CALLBACKS - 1;
+
+	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
+	res->csr_status = status;
+	return status;
+}
+
+#endif /* CONFIG_NFS_V4_1 */
-- 
cgit v1.2.3-59-g8ed1b


From 4aece6a19cf7f474f15eb861ba74db4479884ce3 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:23:26 -0400
Subject: nfs41: cb_sequence xdr implementation

[nfs41: get rid of READMEM and COPYMEM for callback_xdr.c]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: get rid of READ64 in callback_xdr.c]
See http://linux-nfs.org/pipermail/pnfs/2009-June/007846.html
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback_xdr.c | 172 +++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 170 insertions(+), 2 deletions(-)

diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index 41c5be1c1be5..56a3cc510107 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -20,6 +20,11 @@
 				2 + 2 + 3 + 3)
 #define CB_OP_RECALL_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ)
 
+#if defined(CONFIG_NFS_V4_1)
+#define CB_OP_SEQUENCE_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ + \
+					4 + 1 + 3)
+#endif /* CONFIG_NFS_V4_1 */
+
 #define NFSDBG_FACILITY NFSDBG_CALLBACK
 
 typedef __be32 (*callback_process_op_t)(void *, void *);
@@ -207,6 +212,122 @@ out:
 	return status;
 }
 
+#if defined(CONFIG_NFS_V4_1)
+
+static unsigned decode_sessionid(struct xdr_stream *xdr,
+				 struct nfs4_sessionid *sid)
+{
+	uint32_t *p;
+	int len = NFS4_MAX_SESSIONID_LEN;
+
+	p = read_buf(xdr, len);
+	if (unlikely(p == NULL))
+		return htonl(NFS4ERR_RESOURCE);;
+
+	memcpy(sid->data, p, len);
+	return 0;
+}
+
+static unsigned decode_rc_list(struct xdr_stream *xdr,
+			       struct referring_call_list *rc_list)
+{
+	uint32_t *p;
+	int i;
+	unsigned status;
+
+	status = decode_sessionid(xdr, &rc_list->rcl_sessionid);
+	if (status)
+		goto out;
+
+	status = htonl(NFS4ERR_RESOURCE);
+	p = read_buf(xdr, sizeof(uint32_t));
+	if (unlikely(p == NULL))
+		goto out;
+
+	rc_list->rcl_nrefcalls = ntohl(*p++);
+	if (rc_list->rcl_nrefcalls) {
+		p = read_buf(xdr,
+			     rc_list->rcl_nrefcalls * 2 * sizeof(uint32_t));
+		if (unlikely(p == NULL))
+			goto out;
+		rc_list->rcl_refcalls = kmalloc(rc_list->rcl_nrefcalls *
+						sizeof(*rc_list->rcl_refcalls),
+						GFP_KERNEL);
+		if (unlikely(rc_list->rcl_refcalls == NULL))
+			goto out;
+		for (i = 0; i < rc_list->rcl_nrefcalls; i++) {
+			rc_list->rcl_refcalls[i].rc_sequenceid = ntohl(*p++);
+			rc_list->rcl_refcalls[i].rc_slotid = ntohl(*p++);
+		}
+	}
+	status = 0;
+
+out:
+	return status;
+}
+
+static unsigned decode_cb_sequence_args(struct svc_rqst *rqstp,
+					struct xdr_stream *xdr,
+					struct cb_sequenceargs *args)
+{
+	uint32_t *p;
+	int i;
+	unsigned status;
+
+	status = decode_sessionid(xdr, &args->csa_sessionid);
+	if (status)
+		goto out;
+
+	status = htonl(NFS4ERR_RESOURCE);
+	p = read_buf(xdr, 5 * sizeof(uint32_t));
+	if (unlikely(p == NULL))
+		goto out;
+
+	args->csa_addr = svc_addr_in(rqstp);
+	args->csa_sequenceid = ntohl(*p++);
+	args->csa_slotid = ntohl(*p++);
+	args->csa_highestslotid = ntohl(*p++);
+	args->csa_cachethis = ntohl(*p++);
+	args->csa_nrclists = ntohl(*p++);
+	args->csa_rclists = NULL;
+	if (args->csa_nrclists) {
+		args->csa_rclists = kmalloc(args->csa_nrclists *
+					    sizeof(*args->csa_rclists),
+					    GFP_KERNEL);
+		if (unlikely(args->csa_rclists == NULL))
+			goto out;
+
+		for (i = 0; i < args->csa_nrclists; i++) {
+			status = decode_rc_list(xdr, &args->csa_rclists[i]);
+			if (status)
+				goto out_free;
+		}
+	}
+	status = 0;
+
+	dprintk("%s: sessionid %x:%x:%x:%x sequenceid %u slotid %u "
+		"highestslotid %u cachethis %d nrclists %u\n",
+		__func__,
+		((u32 *)&args->csa_sessionid)[0],
+		((u32 *)&args->csa_sessionid)[1],
+		((u32 *)&args->csa_sessionid)[2],
+		((u32 *)&args->csa_sessionid)[3],
+		args->csa_sequenceid, args->csa_slotid,
+		args->csa_highestslotid, args->csa_cachethis,
+		args->csa_nrclists);
+out:
+	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
+	return status;
+
+out_free:
+	for (i = 0; i < args->csa_nrclists; i++)
+		kfree(args->csa_rclists[i].rcl_refcalls);
+	kfree(args->csa_rclists);
+	goto out;
+}
+
+#endif /* CONFIG_NFS_V4_1 */
+
 static __be32 encode_string(struct xdr_stream *xdr, unsigned int len, const char *str)
 {
 	__be32 *p;
@@ -358,12 +479,52 @@ out:
 
 #if defined(CONFIG_NFS_V4_1)
 
+static unsigned encode_sessionid(struct xdr_stream *xdr,
+				 const struct nfs4_sessionid *sid)
+{
+	uint32_t *p;
+	int len = NFS4_MAX_SESSIONID_LEN;
+
+	p = xdr_reserve_space(xdr, len);
+	if (unlikely(p == NULL))
+		return htonl(NFS4ERR_RESOURCE);
+
+	memcpy(p, sid, len);
+	return 0;
+}
+
+static unsigned encode_cb_sequence_res(struct svc_rqst *rqstp,
+				       struct xdr_stream *xdr,
+				       const struct cb_sequenceres *res)
+{
+	uint32_t *p;
+	unsigned status = res->csr_status;
+
+	if (unlikely(status != 0))
+		goto out;
+
+	encode_sessionid(xdr, &res->csr_sessionid);
+
+	p = xdr_reserve_space(xdr, 4 * sizeof(uint32_t));
+	if (unlikely(p == NULL))
+		return htonl(NFS4ERR_RESOURCE);
+
+	*p++ = htonl(res->csr_sequenceid);
+	*p++ = htonl(res->csr_slotid);
+	*p++ = htonl(res->csr_highestslotid);
+	*p++ = htonl(res->csr_target_highestslotid);
+out:
+	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
+	return status;
+}
+
 static __be32
 preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op)
 {
 	switch (op_nr) {
 	case OP_CB_GETATTR:
 	case OP_CB_RECALL:
+	case OP_CB_SEQUENCE:
 		*op = &callback_ops[op_nr];
 		break;
 
@@ -374,7 +535,6 @@ preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op)
 	case OP_CB_RECALL_ANY:
 	case OP_CB_RECALLABLE_OBJ_AVAIL:
 	case OP_CB_RECALL_SLOT:
-	case OP_CB_SEQUENCE:
 	case OP_CB_WANTS_CANCELLED:
 	case OP_CB_NOTIFY_LOCK:
 		return htonl(NFS4ERR_NOTSUPP);
@@ -512,7 +672,15 @@ static struct callback_op callback_ops[] = {
 		.process_op = (callback_process_op_t)nfs4_callback_recall,
 		.decode_args = (callback_decode_arg_t)decode_recall_args,
 		.res_maxsize = CB_OP_RECALL_RES_MAXSZ,
-	}
+	},
+#if defined(CONFIG_NFS_V4_1)
+	[OP_CB_SEQUENCE] = {
+		.process_op = (callback_process_op_t)nfs4_callback_sequence,
+		.decode_args = (callback_decode_arg_t)decode_cb_sequence_args,
+		.encode_res = (callback_encode_res_t)encode_cb_sequence_res,
+		.res_maxsize = CB_OP_SEQUENCE_RES_MAXSZ,
+	},
+#endif /* CONFIG_NFS_V4_1 */
 };
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 281fe15dc1d6ad46992f18b7a6644269ec5f7138 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Wed, 1 Apr 2009 09:23:27 -0400
Subject: nfs41: verify CB_SEQUENCE position in callback compound

CB_SEQUENCE must appear first in the callback compound RPC.
If it is not the first operation NFS4ERR_SEQUENCE_POS must be returned.
If the first operation ni the CB_COMPOUND is not CB_SEQUENCE then
NFS4ERR_OP_NOT_IN_SESSION must be returned.

Signed-off-by: Ricardo Labiaga <ricardo.labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: refactor op preprocessing out of process_op]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback_xdr.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index 56a3cc510107..537f21da6e5f 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -521,6 +521,14 @@ out:
 static __be32
 preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op)
 {
+	if (op_nr == OP_CB_SEQUENCE) {
+		if (nop != 0)
+			return htonl(NFS4ERR_SEQUENCE_POS);
+	} else {
+		if (nop == 0)
+			return htonl(NFS4ERR_OP_NOT_IN_SESSION);
+	}
+
 	switch (op_nr) {
 	case OP_CB_GETATTR:
 	case OP_CB_RECALL:
-- 
cgit v1.2.3-59-g8ed1b


From dd2b63d049480979016b959abc2d141cdddb1389 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:28 -0400
Subject: nfs41: Rename rq_received to rq_reply_bytes_recvd

The 'rq_received' member of 'struct rpc_rqst' is used to track when we
have received a reply to our request.  With v4.1, the backchannel
can now accept callback requests over the existing connection.  Rename
this field to make it clear that it is only used for tracking reply bytes
and not all bytes received on the connection.

Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 include/linux/sunrpc/xprt.h   |  3 ++-
 net/sunrpc/backchannel_rqst.c |  2 +-
 net/sunrpc/clnt.c             |  8 ++++----
 net/sunrpc/stats.c            |  2 +-
 net/sunrpc/xprt.c             | 15 ++++++++-------
 5 files changed, 16 insertions(+), 14 deletions(-)

diff --git a/include/linux/sunrpc/xprt.h b/include/linux/sunrpc/xprt.h
index 55c6c37e249e..1175d58efc2e 100644
--- a/include/linux/sunrpc/xprt.h
+++ b/include/linux/sunrpc/xprt.h
@@ -67,7 +67,8 @@ struct rpc_rqst {
 	struct rpc_task *	rq_task;	/* RPC task data */
 	__be32			rq_xid;		/* request XID */
 	int			rq_cong;	/* has incremented xprt->cong */
-	int			rq_received;	/* receive completed */
+	int			rq_reply_bytes_recvd;	/* number of reply */
+							/* bytes received */
 	u32			rq_seqno;	/* gss seq no. used on req. */
 	int			rq_enc_pages_num;
 	struct page		**rq_enc_pages;	/* scratch pages for use by
diff --git a/net/sunrpc/backchannel_rqst.c b/net/sunrpc/backchannel_rqst.c
index f56e18a23498..5a7d342e3087 100644
--- a/net/sunrpc/backchannel_rqst.c
+++ b/net/sunrpc/backchannel_rqst.c
@@ -230,7 +230,7 @@ struct rpc_rqst *xprt_alloc_bc_request(struct rpc_xprt *xprt)
 
 	if (req != NULL) {
 		set_bit(RPC_BC_PA_IN_USE, &req->rq_bc_pa_state);
-		req->rq_received = 0;
+		req->rq_reply_bytes_recvd = 0;
 		req->rq_bytes_sent = 0;
 		memcpy(&req->rq_private_buf, &req->rq_rcv_buf,
 			sizeof(req->rq_private_buf));
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index f3e93b8eb90f..5bc2f45bddf0 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -1258,8 +1258,8 @@ call_status(struct rpc_task *task)
 	struct rpc_rqst	*req = task->tk_rqstp;
 	int		status;
 
-	if (req->rq_received > 0 && !req->rq_bytes_sent)
-		task->tk_status = req->rq_received;
+	if (req->rq_reply_bytes_recvd > 0 && !req->rq_bytes_sent)
+		task->tk_status = req->rq_reply_bytes_recvd;
 
 	dprint_status(task);
 
@@ -1376,7 +1376,7 @@ call_decode(struct rpc_task *task)
 
 	/*
 	 * Ensure that we see all writes made by xprt_complete_rqst()
-	 * before it changed req->rq_received.
+	 * before it changed req->rq_reply_bytes_recvd.
 	 */
 	smp_rmb();
 	req->rq_rcv_buf.len = req->rq_private_buf.len;
@@ -1417,7 +1417,7 @@ out_retry:
 	task->tk_status = 0;
 	/* Note: rpc_verify_header() may have freed the RPC slot */
 	if (task->tk_rqstp == req) {
-		req->rq_received = req->rq_rcv_buf.len = 0;
+		req->rq_reply_bytes_recvd = req->rq_rcv_buf.len = 0;
 		if (task->tk_client->cl_discrtry)
 			xprt_conditional_disconnect(task->tk_xprt,
 					req->rq_connect_cookie);
diff --git a/net/sunrpc/stats.c b/net/sunrpc/stats.c
index 8487aa0f1f5a..1b4e6791ecf3 100644
--- a/net/sunrpc/stats.c
+++ b/net/sunrpc/stats.c
@@ -156,7 +156,7 @@ void rpc_count_iostats(struct rpc_task *task)
 	op_metrics->om_timeouts += task->tk_timeouts;
 
 	op_metrics->om_bytes_sent += task->tk_bytes_sent;
-	op_metrics->om_bytes_recv += req->rq_received;
+	op_metrics->om_bytes_recv += req->rq_reply_bytes_recvd;
 
 	queue = (long)req->rq_xtime - task->tk_start;
 	if (queue < 0)
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index c144611223fc..f412a852bc73 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -806,9 +806,10 @@ void xprt_complete_rqst(struct rpc_task *task, int copied)
 
 	list_del_init(&req->rq_list);
 	req->rq_private_buf.len = copied;
-	/* Ensure all writes are done before we update req->rq_received */
+	/* Ensure all writes are done before we update */
+	/* req->rq_reply_bytes_recvd */
 	smp_wmb();
-	req->rq_received = copied;
+	req->rq_reply_bytes_recvd = copied;
 	rpc_wake_up_queued_task(&xprt->pending, task);
 }
 EXPORT_SYMBOL_GPL(xprt_complete_rqst);
@@ -823,7 +824,7 @@ static void xprt_timer(struct rpc_task *task)
 	dprintk("RPC: %5u xprt_timer\n", task->tk_pid);
 
 	spin_lock_bh(&xprt->transport_lock);
-	if (!req->rq_received) {
+	if (!req->rq_reply_bytes_recvd) {
 		if (xprt->ops->timer)
 			xprt->ops->timer(task);
 	} else
@@ -845,8 +846,8 @@ int xprt_prepare_transmit(struct rpc_task *task)
 	dprintk("RPC: %5u xprt_prepare_transmit\n", task->tk_pid);
 
 	spin_lock_bh(&xprt->transport_lock);
-	if (req->rq_received && !req->rq_bytes_sent) {
-		err = req->rq_received;
+	if (req->rq_reply_bytes_recvd && !req->rq_bytes_sent) {
+		err = req->rq_reply_bytes_recvd;
 		goto out_unlock;
 	}
 	if (!xprt->ops->reserve_xprt(task))
@@ -875,7 +876,7 @@ void xprt_transmit(struct rpc_task *task)
 
 	dprintk("RPC: %5u xprt_transmit(%u)\n", task->tk_pid, req->rq_slen);
 
-	if (!req->rq_received) {
+	if (!req->rq_reply_bytes_recvd) {
 		if (list_empty(&req->rq_list) && rpc_reply_expected(task)) {
 			/*
 			 * Add to the list only if we're expecting a reply
@@ -914,7 +915,7 @@ void xprt_transmit(struct rpc_task *task)
 	/* Don't race with disconnect */
 	if (!xprt_connected(xprt))
 		task->tk_status = -ENOTCONN;
-	else if (!req->rq_received && rpc_reply_expected(task)) {
+	else if (!req->rq_reply_bytes_recvd && rpc_reply_expected(task)) {
 		/*
 		 * Sleep on the pending queue since
 		 * we're expecting a reply.
-- 
cgit v1.2.3-59-g8ed1b


From 65fc64e547c794764a441e16e95bb76c0e256bd7 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:30 -0400
Subject: nfs41: Backchannel: update cb_sequence args and results

Change the type of cs_addr and csr_status to 'struct sockaddr' and
'__be32' since the cb_sequence processing function will use existing
functionality that expects these types.

Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback.h     | 4 ++--
 fs/nfs/callback_xdr.c | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 576f51f4dc52..07baa8254ca1 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -84,7 +84,7 @@ struct referring_call_list {
 };
 
 struct cb_sequenceargs {
-	struct sockaddr_in		*csa_addr;
+	struct sockaddr			*csa_addr;
 	struct nfs4_sessionid		csa_sessionid;
 	uint32_t			csa_sequenceid;
 	uint32_t			csa_slotid;
@@ -95,7 +95,7 @@ struct cb_sequenceargs {
 };
 
 struct cb_sequenceres {
-	uint32_t			csr_status;
+	__be32				csr_status;
 	struct nfs4_sessionid		csr_sessionid;
 	uint32_t			csr_sequenceid;
 	uint32_t			csr_slotid;
diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index 537f21da6e5f..e5a2dac5f715 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -283,7 +283,7 @@ static unsigned decode_cb_sequence_args(struct svc_rqst *rqstp,
 	if (unlikely(p == NULL))
 		goto out;
 
-	args->csa_addr = svc_addr_in(rqstp);
+	args->csa_addr = svc_addr(rqstp);
 	args->csa_sequenceid = ntohl(*p++);
 	args->csa_slotid = ntohl(*p++);
 	args->csa_highestslotid = ntohl(*p++);
-- 
cgit v1.2.3-59-g8ed1b


From b73dafa7ac94ca8387f65c57cb63a7ffac91bf2c Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:31 -0400
Subject: nfs41: Backchannel: Refactor nfs4_reset_slot_table()

Generalize nfs4_reset_slot_table() so it can be used to reset the
backchannel slot table in addition to the forechannel slot table.

Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c | 38 +++++++++++++++++++++++++++-----------
 1 file changed, 27 insertions(+), 11 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 5da939d577d5..a7b1d6c228cd 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4370,19 +4370,21 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo)
 	return status;
 }
 
-/* Reset a slot table */
-static int nfs4_reset_slot_table(struct nfs4_session *session)
+/*
+ * Reset a slot table
+ */
+static int nfs4_reset_slot_table(struct nfs4_slot_table *tbl, int max_slots,
+		int old_max_slots, int ivalue)
 {
-	struct nfs4_slot_table *tbl = &session->fc_slot_table;
-	int i, max_slots = session->fc_attrs.max_reqs;
-	int old_max_slots = session->fc_slot_table.max_slots;
+	int i;
 	int ret = 0;
 
-	dprintk("--> %s: max_reqs=%u, tbl %p\n", __func__,
-		session->fc_attrs.max_reqs, tbl);
+	dprintk("--> %s: max_reqs=%u, tbl %p\n", __func__, max_slots, tbl);
 
-	/* Until we have dynamic slot table adjustment, insist
-	 * upon the same slot table size */
+	/*
+	 * Until we have dynamic slot table adjustment, insist
+	 * upon the same slot table size
+	 */
 	if (max_slots != old_max_slots) {
 		dprintk("%s reset slot table does't match old\n",
 			__func__);
@@ -4391,7 +4393,7 @@ static int nfs4_reset_slot_table(struct nfs4_session *session)
 	}
 	spin_lock(&tbl->slot_tbl_lock);
 	for (i = 0; i < max_slots; ++i)
-		tbl->slots[i].seq_nr = 1;
+		tbl->slots[i].seq_nr = ivalue;
 	tbl->highest_used_slotid = -1;
 	spin_unlock(&tbl->slot_tbl_lock);
 	dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__,
@@ -4401,6 +4403,20 @@ out:
 	return ret;
 }
 
+/*
+ * Reset the forechannel and backchannel slot tables
+ */
+static int nfs4_reset_slot_tables(struct nfs4_session *session)
+{
+	int status;
+
+	status = nfs4_reset_slot_table(&session->fc_slot_table,
+			session->fc_attrs.max_reqs,
+			session->fc_slot_table.max_slots,
+			1);
+	return status;
+}
+
 /*
  * Initialize slot table
  */
@@ -4639,7 +4655,7 @@ int nfs4_proc_create_session(struct nfs_client *clp, int reset)
 
 	/* Init or reset the fore channel */
 	if (reset)
-		status = nfs4_reset_slot_table(session);
+		status = nfs4_reset_slot_tables(session);
 	else
 		status = nfs4_init_slot_table(session);
 	dprintk("fore channel slot table initialization returned %d\n", status);
-- 
cgit v1.2.3-59-g8ed1b


From 050047ce71bcf60867d2af7a9dc965a9c6f15cb8 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:32 -0400
Subject: nfs41: Backchannel: Refactor nfs4_init_slot_table()

Generalize nfs4_init_slot_table() so it can be used to initialize the
backchannel slot table in addition to the forechannel slot table.

Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c | 26 +++++++++++++++++++-------
 1 file changed, 19 insertions(+), 7 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index a7b1d6c228cd..c3019ad85893 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4420,23 +4420,22 @@ static int nfs4_reset_slot_tables(struct nfs4_session *session)
 /*
  * Initialize slot table
  */
-static int nfs4_init_slot_table(struct nfs4_session *session)
+static int nfs4_init_slot_table(struct nfs4_slot_table *tbl,
+		int max_slots, int ivalue)
 {
-	struct nfs4_slot_table *tbl = &session->fc_slot_table;
-	int i, max_slots = session->fc_attrs.max_reqs;
+	int i;
 	struct nfs4_slot *slot;
 	int ret = -ENOMEM;
 
 	BUG_ON(max_slots > NFS4_MAX_SLOT_TABLE);
 
-	dprintk("--> %s: max_reqs=%u\n", __func__,
-		session->fc_attrs.max_reqs);
+	dprintk("--> %s: max_reqs=%u\n", __func__, max_slots);
 
 	slot = kcalloc(max_slots, sizeof(struct nfs4_slot), GFP_KERNEL);
 	if (!slot)
 		goto out;
 	for (i = 0; i < max_slots; ++i)
-		slot[i].seq_nr = 1;
+		slot[i].seq_nr = ivalue;
 	ret = 0;
 
 	spin_lock(&tbl->slot_tbl_lock);
@@ -4456,11 +4455,24 @@ static int nfs4_init_slot_table(struct nfs4_session *session)
 out:
 	dprintk("<-- %s: return %d\n", __func__, ret);
 	return ret;
+
 out_free:
 	kfree(slot);
 	goto out;
 }
 
+/*
+ * Initialize the forechannel and backchannel tables
+ */
+static int nfs4_init_slot_tables(struct nfs4_session *session)
+{
+	int status;
+
+	status = nfs4_init_slot_table(&session->fc_slot_table,
+			session->fc_attrs.max_reqs, 1);
+	return status;
+}
+
 /* Destroy the slot table */
 static void nfs4_destroy_slot_table(struct nfs4_session *session)
 {
@@ -4657,7 +4669,7 @@ int nfs4_proc_create_session(struct nfs_client *clp, int reset)
 	if (reset)
 		status = nfs4_reset_slot_tables(session);
 	else
-		status = nfs4_init_slot_table(session);
+		status = nfs4_init_slot_tables(session);
 	dprintk("fore channel slot table initialization returned %d\n", status);
 	if (status)
 		goto out;
-- 
cgit v1.2.3-59-g8ed1b


From f8625a6a4bb76207302be58453603d8e324df490 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:33 -0400
Subject: nfs41: Backchannel: Add a backchannel slot table to the session

Defines a new 'struct nfs4_slot_table' in the 'struct nfs4_session'
for use by the backchannel.  Initializes, resets, and destroys the backchannel
slot table in the same manner the forechannel slot table is initialized,
reset, and destroyed.

The sequenceid for each slot in the backchannel slot table is initialized
to 0, whereas the forechannel slotid's sequenceid is set to 1.

Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/nfs4proc.c         | 48 +++++++++++++++++++++++++++++++++++------------
 include/linux/nfs_fs_sb.h |  2 +-
 2 files changed, 37 insertions(+), 13 deletions(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index c3019ad85893..57dabb8a048e 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -4414,9 +4414,30 @@ static int nfs4_reset_slot_tables(struct nfs4_session *session)
 			session->fc_attrs.max_reqs,
 			session->fc_slot_table.max_slots,
 			1);
+	if (status)
+		return status;
+
+	status = nfs4_reset_slot_table(&session->bc_slot_table,
+			session->bc_attrs.max_reqs,
+			session->bc_slot_table.max_slots,
+			0);
 	return status;
 }
 
+/* Destroy the slot table */
+static void nfs4_destroy_slot_tables(struct nfs4_session *session)
+{
+	if (session->fc_slot_table.slots != NULL) {
+		kfree(session->fc_slot_table.slots);
+		session->fc_slot_table.slots = NULL;
+	}
+	if (session->bc_slot_table.slots != NULL) {
+		kfree(session->bc_slot_table.slots);
+		session->bc_slot_table.slots = NULL;
+	}
+	return;
+}
+
 /*
  * Initialize slot table
  */
@@ -4470,17 +4491,15 @@ static int nfs4_init_slot_tables(struct nfs4_session *session)
 
 	status = nfs4_init_slot_table(&session->fc_slot_table,
 			session->fc_attrs.max_reqs, 1);
-	return status;
-}
+	if (status)
+		return status;
 
-/* Destroy the slot table */
-static void nfs4_destroy_slot_table(struct nfs4_session *session)
-{
-	if (session->fc_slot_table.slots == NULL)
-		return;
-	kfree(session->fc_slot_table.slots);
-	session->fc_slot_table.slots = NULL;
-	return;
+	status = nfs4_init_slot_table(&session->bc_slot_table,
+			session->bc_attrs.max_reqs, 0);
+	if (status)
+		nfs4_destroy_slot_tables(session);
+
+	return status;
 }
 
 struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
@@ -4503,7 +4522,12 @@ struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
 
 	tbl = &session->fc_slot_table;
 	spin_lock_init(&tbl->slot_tbl_lock);
-	rpc_init_wait_queue(&tbl->slot_tbl_waitq, "Slot table");
+	rpc_init_wait_queue(&tbl->slot_tbl_waitq, "ForeChannel Slot table");
+
+	tbl = &session->bc_slot_table;
+	spin_lock_init(&tbl->slot_tbl_lock);
+	rpc_init_wait_queue(&tbl->slot_tbl_waitq, "BackChannel Slot table");
+
 	session->clp = clp;
 	return session;
 }
@@ -4515,7 +4539,7 @@ void nfs4_destroy_session(struct nfs4_session *session)
 		__func__, session->clp->cl_rpcclient->cl_xprt);
 	xprt_destroy_backchannel(session->clp->cl_rpcclient->cl_xprt,
 				NFS41_BC_MIN_CALLBACKS);
-	nfs4_destroy_slot_table(session);
+	nfs4_destroy_slot_tables(session);
 	kfree(session);
 }
 
diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h
index d0902ccec9ce..19fe15d12042 100644
--- a/include/linux/nfs_fs_sb.h
+++ b/include/linux/nfs_fs_sb.h
@@ -205,7 +205,7 @@ struct nfs4_session {
 	struct nfs4_channel_attrs	fc_attrs;
 	struct nfs4_slot_table		fc_slot_table;
 	struct nfs4_channel_attrs	bc_attrs;
-					/* back channel has one slot */
+	struct nfs4_slot_table		bc_slot_table;
 	struct nfs_client		*clp;
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 963891ac43ecf9974d82f4c178752e11e007cf87 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:34 -0400
Subject: nfs41: Backchannel: New find_client_with_session()

Finds the 'struct nfs_client' that matches the server's address, major
version number, and session ID.

Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback_proc.c | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index f731bbe7ac6a..6b342e82ebb3 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -104,6 +104,42 @@ out:
 
 #if defined(CONFIG_NFS_V4_1)
 
+/*
+ * Returns a pointer to a held 'struct nfs_client' that matches the server's
+ * address, major version number, and session ID.  It is the caller's
+ * responsibility to release the returned reference.
+ *
+ * Returns NULL if there are no connections with sessions, or if no session
+ * matches the one of interest.
+ */
+ static struct nfs_client *find_client_with_session(
+	const struct sockaddr *addr, u32 nfsversion,
+	struct nfs4_sessionid *sessionid)
+{
+	struct nfs_client *clp;
+
+	clp = nfs_find_client(addr, 4);
+	if (clp == NULL)
+		return NULL;
+
+	do {
+		struct nfs_client *prev = clp;
+
+		if (clp->cl_session != NULL) {
+			if (memcmp(clp->cl_session->sess_id.data,
+					sessionid->data,
+					NFS4_MAX_SESSIONID_LEN) == 0) {
+				/* Returns a held reference to clp */
+				return clp;
+			}
+		}
+		clp = nfs_find_client_next(prev);
+		nfs_put_client(prev);
+	} while (clp != NULL);
+
+	return NULL;
+}
+
 /* FIXME: validate args->cbs_{sequence,slot}id */
 /* FIXME: referring calls should be processed */
 unsigned nfs4_callback_sequence(struct cb_sequenceargs *args,
-- 
cgit v1.2.3-59-g8ed1b


From 68f3f90133d56e0c38f04f991e662c2b21592b31 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Wed, 1 Apr 2009 09:23:35 -0400
Subject: nfs41: Backchannel: CB_SEQUENCE validation

Validates the callback's sessionID, the slot number, and the sequence ID.
Increments the slot's sequence.

Detects replays, but simply prints a debug message (if debugging is enabled
since we don't yet implement a duplicate request cache for the backchannel.
This should not present a problem, since only idempotent callbacks are
currently implemented.

Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
[nfs41: Backchannel: Be more obvious about the return value]
[nfs41: Backchannel: dprink in host order]
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
---
 fs/nfs/callback_proc.c | 71 +++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 67 insertions(+), 4 deletions(-)

diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index 6b342e82ebb3..b7da1f54da68 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -104,6 +104,57 @@ out:
 
 #if defined(CONFIG_NFS_V4_1)
 
+/*
+ * Validate the sequenceID sent by the server.
+ * Return success if the sequenceID is one more than what we last saw on
+ * this slot, accounting for wraparound.  Increments the slot's sequence.
+ *
+ * We don't yet implement a duplicate request cache, so at this time
+ * we will log replays, and process them as if we had not seen them before,
+ * but we don't bump the sequence in the slot.  Not too worried about it,
+ * since we only currently implement idempotent callbacks anyway.
+ *
+ * We have a single slot backchannel at this time, so we don't bother
+ * checking the used_slots bit array on the table.  The lower layer guarantees
+ * a single outstanding callback request at a time.
+ */
+static int
+validate_seqid(struct nfs4_slot_table *tbl, u32 slotid, u32 seqid)
+{
+	struct nfs4_slot *slot;
+
+	dprintk("%s enter. slotid %d seqid %d\n",
+		__func__, slotid, seqid);
+
+	if (slotid > NFS41_BC_MAX_CALLBACKS)
+		return htonl(NFS4ERR_BADSLOT);
+
+	slot = tbl->slots + slotid;
+	dprintk("%s slot table seqid: %d\n", __func__, slot->seq_nr);
+
+	/* Normal */
+	if (likely(seqid == slot->seq_nr + 1)) {
+		slot->seq_nr++;
+		return htonl(NFS4_OK);
+	}
+
+	/* Replay */
+	if (seqid == slot->seq_nr) {
+		dprintk("%s seqid %d is a replay - no DRC available\n",
+			__func__, seqid);
+		return htonl(NFS4_OK);
+	}
+
+	/* Wraparound */
+	if (seqid == 1 && (slot->seq_nr + 1) == 0) {
+		slot->seq_nr = 1;
+		return htonl(NFS4_OK);
+	}
+
+	/* Misordered request */
+	return htonl(NFS4ERR_SEQ_MISORDERED);
+}
+
 /*
  * Returns a pointer to a held 'struct nfs_client' that matches the server's
  * address, major version number, and session ID.  It is the caller's
@@ -140,18 +191,27 @@ out:
 	return NULL;
 }
 
-/* FIXME: validate args->cbs_{sequence,slot}id */
 /* FIXME: referring calls should be processed */
 unsigned nfs4_callback_sequence(struct cb_sequenceargs *args,
 				struct cb_sequenceres *res)
 {
-	int i;
-	unsigned status = 0;
+	struct nfs_client *clp;
+	int i, status;
 
 	for (i = 0; i < args->csa_nrclists; i++)
 		kfree(args->csa_rclists[i].rcl_refcalls);
 	kfree(args->csa_rclists);
 
+	status = htonl(NFS4ERR_BADSESSION);
+	clp = find_client_with_session(args->csa_addr, 4, &args->csa_sessionid);
+	if (clp == NULL)
+		goto out;
+
+	status = validate_seqid(&clp->cl_session->bc_slot_table,
+				args->csa_slotid, args->csa_sequenceid);
+	if (status)
+		goto out_putclient;
+
 	memcpy(&res->csr_sessionid, &args->csa_sessionid,
 	       sizeof(res->csr_sessionid));
 	res->csr_sequenceid = args->csa_sequenceid;
@@ -159,9 +219,12 @@ unsigned nfs4_callback_sequence(struct cb_sequenceargs *args,
 	res->csr_highestslotid = NFS41_BC_MAX_CALLBACKS - 1;
 	res->csr_target_highestslotid = NFS41_BC_MAX_CALLBACKS - 1;
 
+out_putclient:
+	nfs_put_client(clp);
+out:
 	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
 	res->csr_status = status;
-	return status;
+	return res->csr_status;
 }
 
 #endif /* CONFIG_NFS_V4_1 */
-- 
cgit v1.2.3-59-g8ed1b


From bf40d3435caf49369058b1ed6bbc92f8e2bd92f1 Mon Sep 17 00:00:00 2001
From: Suresh Jayaraman <sjayaraman@suse.de>
Date: Wed, 17 Jun 2009 18:02:09 -0700
Subject: NFS: add support for splice writes

Adds support for splice writes. It effectively calls
generic_file_splice_write() to do the writes.

We need not worry about O_APPEND case as the combination of splice()
writes and O_APPEND is disallowed. This patch propagates NFS write
errors back to the caller. The number of bytes written via splice are
being added to NFSIO_NORMALWRITTENBYTES as these are effectively
cached writes.

Signed-off-by: Suresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/file.c | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 52ed56762583..0055b813ec2c 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -48,6 +48,9 @@ static ssize_t nfs_file_splice_read(struct file *filp, loff_t *ppos,
 					size_t count, unsigned int flags);
 static ssize_t nfs_file_read(struct kiocb *, const struct iovec *iov,
 				unsigned long nr_segs, loff_t pos);
+static ssize_t nfs_file_splice_write(struct pipe_inode_info *pipe,
+					struct file *filp, loff_t *ppos,
+					size_t count, unsigned int flags);
 static ssize_t nfs_file_write(struct kiocb *, const struct iovec *iov,
 				unsigned long nr_segs, loff_t pos);
 static int  nfs_file_flush(struct file *, fl_owner_t id);
@@ -73,6 +76,7 @@ const struct file_operations nfs_file_operations = {
 	.lock		= nfs_lock,
 	.flock		= nfs_flock,
 	.splice_read	= nfs_file_splice_read,
+	.splice_write	= nfs_file_splice_write,
 	.check_flags	= nfs_check_flags,
 	.setlease	= nfs_setlease,
 };
@@ -587,6 +591,33 @@ out_swapfile:
 	goto out;
 }
 
+static ssize_t nfs_file_splice_write(struct pipe_inode_info *pipe,
+				     struct file *filp, loff_t *ppos,
+				     size_t count, unsigned int flags)
+{
+	struct dentry *dentry = filp->f_path.dentry;
+	struct inode *inode = dentry->d_inode;
+	ssize_t ret;
+
+	dprintk("NFS splice_write(%s/%s, %lu@%llu)\n",
+		dentry->d_parent->d_name.name, dentry->d_name.name,
+		(unsigned long) count, (unsigned long long) *ppos);
+
+	/*
+	 * The combination of splice and an O_APPEND destination is disallowed.
+	 */
+
+	nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, count);
+
+	ret = generic_file_splice_write(pipe, filp, ppos, count, flags);
+	if (ret >= 0 && nfs_need_sync_write(filp, inode)) {
+		int err = nfs_do_fsync(nfs_file_open_context(filp), inode);
+		if (err < 0)
+			ret = err;
+	}
+	return ret;
+}
+
 static int do_getlk(struct file *filp, int cmd, struct file_lock *fl)
 {
 	struct inode *inode = filp->f_mapping->host;
-- 
cgit v1.2.3-59-g8ed1b


From c381ad2cf2d5dcd3991bcc8a18fddd9d5c66ccaa Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:09 -0700
Subject: NFS: Do not display the setting of the "intr" mount option

The "intr" mount option has been deprecated for a while, but
/proc/mounts continues to display "nointr" whether "intr" or "nointr"
has been specified for a mount point.

Since these options do not have any effect, simply do not display
them.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/super.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 98b47d17740a..539a61a0887b 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -514,7 +514,6 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss,
 		const char *nostr;
 	} nfs_info[] = {
 		{ NFS_MOUNT_SOFT, ",soft", ",hard" },
-		{ NFS_MOUNT_INTR, ",intr", ",nointr" },
 		{ NFS_MOUNT_POSIX, ",posix", "" },
 		{ NFS_MOUNT_NOCTO, ",nocto", "" },
 		{ NFS_MOUNT_NOAC, ",noac", "" },
-- 
cgit v1.2.3-59-g8ed1b


From a21bdd9b960ccce421b63aa0e3efda4fcdc26f10 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:10 -0700
Subject: NFS: Return error code from nfs_callback_up() to user space

If the kernel cannot start the NFSv4 callback service during a mount
request, it returns -ENOMEM to user space, resulting in this message:

   mount.nfs4: Cannot allocate memory

Adjust nfs_alloc_client() and nfs_get_client() to pass NFSv4 callback
start-up errors back to user space so a less mysterious error message
can be displayed by the mount command.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/client.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 75c9cd2aa119..0e9a1f9743b0 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -114,6 +114,7 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 {
 	struct nfs_client *clp;
 	struct rpc_cred *cred;
+	int err = -ENOMEM;
 
 	if ((clp = kzalloc(sizeof(*clp), GFP_KERNEL)) == NULL)
 		goto error_0;
@@ -121,7 +122,8 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 	clp->rpc_ops = cl_init->rpc_ops;
 
 	if (cl_init->rpc_ops->version == 4) {
-		if (nfs_callback_up() < 0)
+		err = nfs_callback_up();
+		if (err < 0)
 			goto error_2;
 		__set_bit(NFS_CS_CALLBACK, &clp->cl_res_state);
 	}
@@ -133,6 +135,7 @@ static struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_
 	clp->cl_addrlen = cl_init->addrlen;
 
 	if (cl_init->hostname) {
+		err = -ENOMEM;
 		clp->cl_hostname = kstrdup(cl_init->hostname, GFP_KERNEL);
 		if (!clp->cl_hostname)
 			goto error_3;
@@ -165,7 +168,7 @@ error_3:
 error_2:
 	kfree(clp);
 error_0:
-	return NULL;
+	return ERR_PTR(err);
 }
 
 static void nfs4_shutdown_client(struct nfs_client *clp)
@@ -456,9 +459,10 @@ static struct nfs_client *nfs_get_client(const struct nfs_client_initdata *cl_in
 		spin_unlock(&nfs_client_lock);
 
 		new = nfs_alloc_client(cl_init);
-	} while (new);
+	} while (!IS_ERR(new));
 
-	return ERR_PTR(-ENOMEM);
+	dprintk("--> nfs_get_client() = %ld [failed]\n", PTR_ERR(new));
+	return new;
 
 	/* install a new client and return with it unready */
 install_client:
-- 
cgit v1.2.3-59-g8ed1b


From 18fc31641925867c871bc75270ce642c039188d3 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:10 -0700
Subject: NFS: Fix false error return from nfs_callback_up() if ipv6.ko is not
 available

Clear "ret" if the error return from svc_create_xprt(AF_INET6) was
-EAFNOSUPORT.  Otherwise, callback start-up will succeed, but
nfs_callback_up() will return -EAFNOSUPPORT anyway, and the first
NFSv4 mount attempt after a reboot will fail.

Bug introduced by commit f738f517 in 2.6.30-rc1.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/callback.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index a886e692ddd0..6e3bd42c21cd 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -127,7 +127,9 @@ int nfs_callback_up(void)
 		nfs_callback_tcpport6 = ret;
 		dprintk("NFS: Callback listener port = %u (af %u)\n",
 				nfs_callback_tcpport6, PF_INET6);
-	} else if (ret != -EAFNOSUPPORT)
+	} else if (ret == -EAFNOSUPPORT)
+		ret = 0;
+	else
 		goto out_err;
 #endif	/* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */
 
-- 
cgit v1.2.3-59-g8ed1b


From 6c9dc4255108bab4ef5c177d369b99c3c23492a7 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:10 -0700
Subject: lockd: Update NSM state from SM_MON replies

When rpc.statd starts up in user space at boot time, it attempts to
write the latest NSM local state number into
/proc/sys/fs/nfs/nsm_local_state.

If lockd.ko isn't loaded yet (as is the case in most configurations),
that file doesn't exist, thus the kernel's NSM state remains set to
its initial value of zero during lockd operation.

This is a problem because rpc.statd and lockd use the NSM state number
to prevent repeated lock recovery on rebooted hosts.  If lockd sends
a zero NSM state, but then a delayed SM_NOTIFY with a real NSM state
number is received, there is no way for lockd or rpc.statd to
distinguish that stale SM_NOTIFY from an actual reboot.  Thus lock
recovery could be performed after the rebooted host has already
started reclaiming locks, and those locks will be lost.

We could change /etc/init.d/nfslock so it always modprobes lockd.ko
before starting rpc.statd.  However, if lockd.ko is ever unloaded
and reloaded, we are back at square one, since the NSM state is not
preserved across an unload/reload cycle.  This may happen frequently
on clients that use automounter.  A period of NFS inactivity causes
lockd.ko to be unloaded, and the kernel loses its NSM state setting.

Instead, let's use the fact that rpc.statd plants the local system's
NSM state in every SM_MON (and SM_UNMON) reply.  lockd performs a
synchronous SM_MON upcall to the local rpc.statd _before_ sending its
first NLM request to a new remote.  This would permit rpc.statd to
provide the current NSM state to lockd, even after lockd.ko had been
unloaded and reloaded.

Note that NLMPROC_LOCK arguments are constructed before the
nsm_monitor() call, so we have to rearrange argument construction very
slightly to make this all work out.

And, the kernel appears to treat NSM state as a u32 (see struct
nlm_args and nsm_res).  Make nsm_local_state a u32 as well, to ensure
we don't get bogus comparison results.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/lockd/clntproc.c         |  2 +-
 fs/lockd/mon.c              | 18 ++++++++++++------
 include/linux/lockd/lockd.h |  2 +-
 3 files changed, 14 insertions(+), 8 deletions(-)

diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c
index 273e229353f3..f2fdcbce143e 100644
--- a/fs/lockd/clntproc.c
+++ b/fs/lockd/clntproc.c
@@ -126,7 +126,6 @@ static void nlmclnt_setlockargs(struct nlm_rqst *req, struct file_lock *fl)
 	struct nlm_lock	*lock = &argp->lock;
 
 	nlmclnt_next_cookie(&argp->cookie);
-	argp->state   = nsm_local_state;
 	memcpy(&lock->fh, NFS_FH(fl->fl_file->f_path.dentry->d_inode), sizeof(struct nfs_fh));
 	lock->caller  = utsname()->nodename;
 	lock->oh.data = req->a_owner;
@@ -521,6 +520,7 @@ nlmclnt_lock(struct nlm_rqst *req, struct file_lock *fl)
 
 	if (nsm_monitor(host) < 0)
 		goto out;
+	req->a_args.state = nsm_local_state;
 
 	fl->fl_flags |= FL_ACCESS;
 	status = do_vfs_lock(fl);
diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 6d5d4a4169e5..38385336614c 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -53,7 +53,7 @@ static				DEFINE_SPINLOCK(nsm_lock);
 /*
  * Local NSM state
  */
-int	__read_mostly		nsm_local_state;
+u32	__read_mostly		nsm_local_state;
 int	__read_mostly		nsm_use_hostnames;
 
 static inline struct sockaddr *nsm_addr(const struct nsm_handle *nsm)
@@ -184,13 +184,19 @@ int nsm_monitor(const struct nlm_host *host)
 	nsm->sm_mon_name = nsm_use_hostnames ? nsm->sm_name : nsm->sm_addrbuf;
 
 	status = nsm_mon_unmon(nsm, NSMPROC_MON, &res);
-	if (res.status != 0)
+	if (unlikely(res.status != 0))
 		status = -EIO;
-	if (status < 0)
+	if (unlikely(status < 0)) {
 		printk(KERN_NOTICE "lockd: cannot monitor %s\n", nsm->sm_name);
-	else
-		nsm->sm_monitored = 1;
-	return status;
+		return status;
+	}
+
+	nsm->sm_monitored = 1;
+	if (unlikely(nsm_local_state != res.state)) {
+		nsm_local_state = res.state;
+		dprintk("lockd: NSM state changed to %d\n", nsm_local_state);
+	}
+	return 0;
 }
 
 /**
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 51855dfd8adb..c325b187966b 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -195,7 +195,7 @@ extern struct svc_procedure	nlmsvc_procedures4[];
 extern int			nlmsvc_grace_period;
 extern unsigned long		nlmsvc_timeout;
 extern int			nsm_use_hostnames;
-extern int			nsm_local_state;
+extern u32			nsm_local_state;
 
 /*
  * Lockd client functions
-- 
cgit v1.2.3-59-g8ed1b


From 0e5c2632e1c9182f0dadc31bec68d6f42e7905ea Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:11 -0700
Subject: lockd: Don't bother with RPC ping for NSM upcalls

Cut NSM upcall RPC traffic in half -- don't do a NULL call first.
The cases where a ping would be helpful are rare.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/lockd/mon.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c
index 38385336614c..7fce1b525849 100644
--- a/fs/lockd/mon.c
+++ b/fs/lockd/mon.c
@@ -112,6 +112,7 @@ static struct rpc_clnt *nsm_create(void)
 		.program		= &nsm_program,
 		.version		= NSM_VERSION,
 		.authflavor		= RPC_AUTH_NULL,
+		.flags			= RPC_CLNT_CREATE_NOPING,
 	};
 
 	return rpc_create(&args);
-- 
cgit v1.2.3-59-g8ed1b


From 2ad780978b7c0c3e7877949f098cbd06e7c73839 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:11 -0700
Subject: NFS: Clean up MNT program definitions

Clean up:  Relocate MNT program procedure number definitions to the
only file that uses them.  Relocate the version number definitions,
which are shared, to nfs.h.  Remove duplicate program number
definitions.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/mount_clnt.c  | 32 ++++++++++++++++++++++++++++----
 fs/nfs/nfsroot.c     |  3 +++
 include/linux/nfs.h  |  5 +++--
 include/linux/nfs2.h |  7 -------
 include/linux/nfs3.h |  5 -----
 5 files changed, 34 insertions(+), 18 deletions(-)

diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index ca905a5bb1ba..af45a374d56f 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -20,6 +20,30 @@
 # define NFSDBG_FACILITY	NFSDBG_MOUNT
 #endif
 
+/*
+ * Defined by RFC 1094, section A.5
+ */
+enum {
+	MOUNTPROC_NULL		= 0,
+	MOUNTPROC_MNT		= 1,
+	MOUNTPROC_DUMP		= 2,
+	MOUNTPROC_UMNT		= 3,
+	MOUNTPROC_UMNTALL	= 4,
+	MOUNTPROC_EXPORT	= 5,
+};
+
+/*
+ * Defined by RFC 1813, section 5.2
+ */
+enum {
+	MOUNTPROC3_NULL		= 0,
+	MOUNTPROC3_MNT		= 1,
+	MOUNTPROC3_DUMP		= 2,
+	MOUNTPROC3_UMNT		= 3,
+	MOUNTPROC3_UMNTALL	= 4,
+	MOUNTPROC3_EXPORT	= 5,
+};
+
 static struct rpc_program	mnt_program;
 
 struct mnt_fhstatus {
@@ -68,7 +92,7 @@ int nfs_mount(struct nfs_mount_request *info)
 	if (info->version == NFS_MNT3_VERSION)
 		msg.rpc_proc = &mnt_clnt->cl_procinfo[MOUNTPROC3_MNT];
 	else
-		msg.rpc_proc = &mnt_clnt->cl_procinfo[MNTPROC_MNT];
+		msg.rpc_proc = &mnt_clnt->cl_procinfo[MOUNTPROC_MNT];
 
 	status = rpc_call_sync(mnt_clnt, &msg, 0);
 	rpc_shutdown_client(mnt_clnt);
@@ -145,13 +169,13 @@ static int xdr_decode_fhstatus3(struct rpc_rqst *req, __be32 *p,
 #define MNT_fhstatus3_sz	(1 + 16)
 
 static struct rpc_procinfo mnt_procedures[] = {
-	[MNTPROC_MNT] = {
-		.p_proc		= MNTPROC_MNT,
+	[MOUNTPROC_MNT] = {
+		.p_proc		= MOUNTPROC_MNT,
 		.p_encode	= (kxdrproc_t) xdr_encode_dirpath,
 		.p_decode	= (kxdrproc_t) xdr_decode_fhstatus,
 		.p_arglen	= MNT_dirpath_sz,
 		.p_replen	= MNT_fhstatus_sz,
-		.p_statidx	= MNTPROC_MNT,
+		.p_statidx	= MOUNTPROC_MNT,
 		.p_name		= "MOUNT",
 	},
 };
diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c
index e3ed5908820b..24c1b93874c4 100644
--- a/fs/nfs/nfsroot.c
+++ b/fs/nfs/nfsroot.c
@@ -92,6 +92,9 @@
 #undef NFSROOT_DEBUG
 #define NFSDBG_FACILITY NFSDBG_ROOT
 
+/* Default port to use if server is not running a portmapper */
+#define NFS_MNT_PORT	627
+
 /* Default path we try to mount. "%s" gets replaced by our IP address */
 #define NFS_ROOT		"/tftpboot/%s"
 
diff --git a/include/linux/nfs.h b/include/linux/nfs.h
index 214d499718f7..f387919bbc59 100644
--- a/include/linux/nfs.h
+++ b/include/linux/nfs.h
@@ -25,8 +25,9 @@
 #define NFSMODE_SOCK	0140000
 #define NFSMODE_FIFO	0010000
 
-#define NFS_MNT_PROGRAM	100005
-#define NFS_MNT_PORT	627
+#define NFS_MNT_PROGRAM		100005
+#define NFS_MNT_VERSION		1
+#define NFS_MNT3_VERSION	3
 
 /*
  * NFS stats. The good thing with these values is that NFSv3 errors are
diff --git a/include/linux/nfs2.h b/include/linux/nfs2.h
index 0ed9517138fc..fde24b30cc9e 100644
--- a/include/linux/nfs2.h
+++ b/include/linux/nfs2.h
@@ -64,11 +64,4 @@ struct nfs2_fh {
 #define NFSPROC_READDIR		16
 #define NFSPROC_STATFS		17
 
-#define NFS_MNT_PROGRAM		100005
-#define NFS_MNT_VERSION		1
-#define MNTPROC_NULL		0
-#define MNTPROC_MNT		1
-#define MNTPROC_UMNT		3
-#define MNTPROC_UMNTALL		4
-
 #endif /* _LINUX_NFS2_H */
diff --git a/include/linux/nfs3.h b/include/linux/nfs3.h
index 539f3b550eab..ac33806ec7f9 100644
--- a/include/linux/nfs3.h
+++ b/include/linux/nfs3.h
@@ -88,12 +88,7 @@ struct nfs3_fh {
 #define NFS3PROC_PATHCONF	20
 #define NFS3PROC_COMMIT		21
 
-#define NFS_MNT3_PROGRAM	100005
 #define NFS_MNT3_VERSION	3
-#define MOUNTPROC3_NULL		0
-#define MOUNTPROC3_MNT		1
-#define MOUNTPROC3_UMNT		3
-#define MOUNTPROC3_UMNTALL	4
  
 
 #if defined(__KERNEL__)
-- 
cgit v1.2.3-59-g8ed1b


From 29a1bd6bf8c7a9ff511aaaf7e40fd6ca5198babe Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:11 -0700
Subject: NFS: Use xdr_stream-based XDR encoder for MNT's dirpath argument

Check the length of the supplied dirpath, and see that it fits
properly in the RPC buffer.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/mount_clnt.c | 49 ++++++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 44 insertions(+), 5 deletions(-)

diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index af45a374d56f..93361afcd5da 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -20,6 +20,21 @@
 # define NFSDBG_FACILITY	NFSDBG_MOUNT
 #endif
 
+/*
+ * Defined by RFC 1094, section A.3; and RFC 1813, section 5.1.4
+ */
+#define MNTPATHLEN		(1024)
+
+/*
+ * XDR data type sizes
+ */
+#define encode_dirpath_sz	(1 + XDR_QUADLEN(MNTPATHLEN))
+
+/*
+ * XDR argument and result sizes
+ */
+#define MNT_enc_dirpath_sz	encode_dirpath_sz
+
 /*
  * Defined by RFC 1094, section A.5
  */
@@ -135,6 +150,31 @@ static int xdr_encode_dirpath(struct rpc_rqst *req, __be32 *p,
 	return 0;
 }
 
+static int encode_mntdirpath(struct xdr_stream *xdr, const char *pathname)
+{
+	const u32 pathname_len = strlen(pathname);
+	__be32 *p;
+
+	if (unlikely(pathname_len > MNTPATHLEN))
+		return -EIO;
+
+	p = xdr_reserve_space(xdr, sizeof(u32) + pathname_len);
+	if (unlikely(p == NULL))
+		return -EIO;
+	xdr_encode_opaque(p, pathname, pathname_len);
+
+	return 0;
+}
+
+static int mnt_enc_dirpath(struct rpc_rqst *req, __be32 *p,
+			   const char *dirpath)
+{
+	struct xdr_stream xdr;
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	return encode_mntdirpath(&xdr, dirpath);
+}
+
 static int xdr_decode_fhstatus(struct rpc_rqst *req, __be32 *p,
 			       struct mnt_fhstatus *res)
 {
@@ -164,16 +204,15 @@ static int xdr_decode_fhstatus3(struct rpc_rqst *req, __be32 *p,
 	return 0;
 }
 
-#define MNT_dirpath_sz		(1 + 256)
 #define MNT_fhstatus_sz		(1 + 8)
 #define MNT_fhstatus3_sz	(1 + 16)
 
 static struct rpc_procinfo mnt_procedures[] = {
 	[MOUNTPROC_MNT] = {
 		.p_proc		= MOUNTPROC_MNT,
-		.p_encode	= (kxdrproc_t) xdr_encode_dirpath,
+		.p_encode	= (kxdrproc_t)mnt_enc_dirpath,
 		.p_decode	= (kxdrproc_t) xdr_decode_fhstatus,
-		.p_arglen	= MNT_dirpath_sz,
+		.p_arglen	= MNT_enc_dirpath_sz,
 		.p_replen	= MNT_fhstatus_sz,
 		.p_statidx	= MOUNTPROC_MNT,
 		.p_name		= "MOUNT",
@@ -183,9 +222,9 @@ static struct rpc_procinfo mnt_procedures[] = {
 static struct rpc_procinfo mnt3_procedures[] = {
 	[MOUNTPROC3_MNT] = {
 		.p_proc		= MOUNTPROC3_MNT,
-		.p_encode	= (kxdrproc_t) xdr_encode_dirpath,
+		.p_encode	= (kxdrproc_t)mnt_enc_dirpath,
 		.p_decode	= (kxdrproc_t) xdr_decode_fhstatus3,
-		.p_arglen	= MNT_dirpath_sz,
+		.p_arglen	= MNT_enc_dirpath_sz,
 		.p_replen	= MNT_fhstatus3_sz,
 		.p_statidx	= MOUNTPROC3_MNT,
 		.p_name		= "MOUNT",
-- 
cgit v1.2.3-59-g8ed1b


From 99835db430904e90c0640ebc6b91cd2a90a118f7 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:11 -0700
Subject: NFS: remove unused function in fs/nfs/mount_clnt.c

Clean up: remove xdr_encode_dirpath() now that it has been replaced.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/mount_clnt.c | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index 93361afcd5da..79b5954b8a1b 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -141,14 +141,6 @@ out_mnt_err:
 /*
  * XDR encode/decode functions for MOUNT
  */
-static int xdr_encode_dirpath(struct rpc_rqst *req, __be32 *p,
-			      const char *path)
-{
-	p = xdr_encode_string(p, path);
-
-	req->rq_slen = xdr_adjust_iovec(req->rq_svec, p);
-	return 0;
-}
 
 static int encode_mntdirpath(struct xdr_stream *xdr, const char *pathname)
 {
-- 
cgit v1.2.3-59-g8ed1b


From fb12529577541aa02f9c3d9e325329f9568dfb58 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:12 -0700
Subject: NFS: Add separate mountd status code decoders for each mountd version

Introduce data structures and xdr_stream-based decoding functions for
unmarshalling mountd status codes properly.

Mountd version 3 uses specific standard error return codes that are
not errno values and not NFS3ERR_ values.  These have a well-defined
standard mapping to local errno values.  Introduce data structures
and a decoder function that map these status codes to local errno
values properly.  This is new functionality (but not used yet).

Version 1 mountd status values are defined by RFC 1094 as UNIX error
values (errno values).  Errno values on heterogeneous systems do not
necessarily match each other.  To avoid exposing possibly incorrect
errno values to upper layers, the current XDR decoder converts all
non-zero MNT version 1 status codes to -EACCES.

The OpenGroup XNFS standard provides a mapping similar to but smaller
than the version 3 error codes.  Implement a decoder that uses the XNFS
error codes, replacing the current decoder.

For both mountd protocol versions, map unrecognized errors to -EACCES.

Finally we introduce a replacement data structure for mnt_fhstatus
at this time, which is used by the new XDR decoders.  In addition to
documenting that the status value returned by the XDR decoders is
always an errno, this new structure will be expanded in subsequent
patches.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/mount_clnt.c | 116 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 116 insertions(+)

diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index 79b5954b8a1b..8429885bc729 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -29,6 +29,8 @@
  * XDR data type sizes
  */
 #define encode_dirpath_sz	(1 + XDR_QUADLEN(MNTPATHLEN))
+#define MNT_status_sz		(1)
+#define MNT_fhs_status_sz	(1)
 
 /*
  * XDR argument and result sizes
@@ -61,6 +63,65 @@ enum {
 
 static struct rpc_program	mnt_program;
 
+/*
+ * Defined by OpenGroup XNFS Version 3W, chapter 8
+ */
+enum mountstat {
+	MNT_OK			= 0,
+	MNT_EPERM		= 1,
+	MNT_ENOENT		= 2,
+	MNT_EACCES		= 13,
+	MNT_EINVAL		= 22,
+};
+
+static struct {
+	u32 status;
+	int errno;
+} mnt_errtbl[] = {
+	{ .status = MNT_OK,			.errno = 0,		},
+	{ .status = MNT_EPERM,			.errno = -EPERM,	},
+	{ .status = MNT_ENOENT,			.errno = -ENOENT,	},
+	{ .status = MNT_EACCES,			.errno = -EACCES,	},
+	{ .status = MNT_EINVAL,			.errno = -EINVAL,	},
+};
+
+/*
+ * Defined by RFC 1813, section 5.1.5
+ */
+enum mountstat3 {
+	MNT3_OK			= 0,		/* no error */
+	MNT3ERR_PERM		= 1,		/* Not owner */
+	MNT3ERR_NOENT		= 2,		/* No such file or directory */
+	MNT3ERR_IO		= 5,		/* I/O error */
+	MNT3ERR_ACCES		= 13,		/* Permission denied */
+	MNT3ERR_NOTDIR		= 20,		/* Not a directory */
+	MNT3ERR_INVAL		= 22,		/* Invalid argument */
+	MNT3ERR_NAMETOOLONG	= 63,		/* Filename too long */
+	MNT3ERR_NOTSUPP		= 10004,	/* Operation not supported */
+	MNT3ERR_SERVERFAULT	= 10006,	/* A failure on the server */
+};
+
+static struct {
+	u32 status;
+	int errno;
+} mnt3_errtbl[] = {
+	{ .status = MNT3_OK,			.errno = 0,		},
+	{ .status = MNT3ERR_PERM,		.errno = -EPERM,	},
+	{ .status = MNT3ERR_NOENT,		.errno = -ENOENT,	},
+	{ .status = MNT3ERR_IO,			.errno = -EIO,		},
+	{ .status = MNT3ERR_ACCES,		.errno = -EACCES,	},
+	{ .status = MNT3ERR_NOTDIR,		.errno = -ENOTDIR,	},
+	{ .status = MNT3ERR_INVAL,		.errno = -EINVAL,	},
+	{ .status = MNT3ERR_NAMETOOLONG,	.errno = -ENAMETOOLONG,	},
+	{ .status = MNT3ERR_NOTSUPP,		.errno = -ENOTSUPP,	},
+	{ .status = MNT3ERR_SERVERFAULT,	.errno = -ESERVERFAULT,	},
+};
+
+struct mountres {
+	int errno;
+	struct nfs_fh *fh;
+};
+
 struct mnt_fhstatus {
 	u32 status;
 	struct nfs_fh *fh;
@@ -179,6 +240,61 @@ static int xdr_decode_fhstatus(struct rpc_rqst *req, __be32 *p,
 	return 0;
 }
 
+/*
+ * RFC 1094: "A non-zero status indicates some sort of error.  In this
+ * case, the status is a UNIX error number."  This can be problematic
+ * if the server and client use different errno values for the same
+ * error.
+ *
+ * However, the OpenGroup XNFS spec provides a simple mapping that is
+ * independent of local errno values on the server and the client.
+ */
+static int decode_status(struct xdr_stream *xdr, struct mountres *res)
+{
+	unsigned int i;
+	u32 status;
+	__be32 *p;
+
+	p = xdr_inline_decode(xdr, sizeof(status));
+	if (unlikely(p == NULL))
+		return -EIO;
+	status = ntohl(*p);
+
+	for (i = 0; i <= ARRAY_SIZE(mnt_errtbl); i++) {
+		if (mnt_errtbl[i].status == status) {
+			res->errno = mnt_errtbl[i].errno;
+			return 0;
+		}
+	}
+
+	dprintk("NFS: unrecognized MNT status code: %u\n", status);
+	res->errno = -EACCES;
+	return 0;
+}
+
+static int decode_fhs_status(struct xdr_stream *xdr, struct mountres *res)
+{
+	unsigned int i;
+	u32 status;
+	__be32 *p;
+
+	p = xdr_inline_decode(xdr, sizeof(status));
+	if (unlikely(p == NULL))
+		return -EIO;
+	status = ntohl(*p);
+
+	for (i = 0; i <= ARRAY_SIZE(mnt3_errtbl); i++) {
+		if (mnt3_errtbl[i].status == status) {
+			res->errno = mnt3_errtbl[i].errno;
+			return 0;
+		}
+	}
+
+	dprintk("NFS: unrecognized MNT3 status code: %u\n", status);
+	res->errno = -EACCES;
+	return 0;
+}
+
 static int xdr_decode_fhstatus3(struct rpc_rqst *req, __be32 *p,
 				struct mnt_fhstatus *res)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 4fdcd9966d8469be26a6f12122ac21ffce19fc20 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:12 -0700
Subject: NFS: add new file handle decoders to in-kernel mountd client

Introduce xdr_stream-based XDR file handle decoders to the in-kernel
mountd client.  These are more careful than the existing decoder
functions about buffer overflows and data type and range checking.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/mount_clnt.c | 39 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 39 insertions(+)

diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index 8429885bc729..dd8c6f448eaa 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -31,6 +31,8 @@
 #define encode_dirpath_sz	(1 + XDR_QUADLEN(MNTPATHLEN))
 #define MNT_status_sz		(1)
 #define MNT_fhs_status_sz	(1)
+#define MNT_fhandle_sz		XDR_QUADLEN(NFS2_FHSIZE)
+#define MNT_fhandle3_sz		(1 + XDR_QUADLEN(NFS3_FHSIZE))
 
 /*
  * XDR argument and result sizes
@@ -272,6 +274,20 @@ static int decode_status(struct xdr_stream *xdr, struct mountres *res)
 	return 0;
 }
 
+static int decode_fhandle(struct xdr_stream *xdr, struct mountres *res)
+{
+	struct nfs_fh *fh = res->fh;
+	__be32 *p;
+
+	p = xdr_inline_decode(xdr, NFS2_FHSIZE);
+	if (unlikely(p == NULL))
+		return -EIO;
+
+	fh->size = NFS2_FHSIZE;
+	memcpy(fh->data, p, NFS2_FHSIZE);
+	return 0;
+}
+
 static int decode_fhs_status(struct xdr_stream *xdr, struct mountres *res)
 {
 	unsigned int i;
@@ -295,6 +311,29 @@ static int decode_fhs_status(struct xdr_stream *xdr, struct mountres *res)
 	return 0;
 }
 
+static int decode_fhandle3(struct xdr_stream *xdr, struct mountres *res)
+{
+	struct nfs_fh *fh = res->fh;
+	u32 size;
+	__be32 *p;
+
+	p = xdr_inline_decode(xdr, sizeof(size));
+	if (unlikely(p == NULL))
+		return -EIO;
+
+	size = ntohl(*p++);
+	if (size > NFS3_FHSIZE || size == 0)
+		return -EIO;
+
+	p = xdr_inline_decode(xdr, size);
+	if (unlikely(p == NULL))
+		return -EIO;
+
+	fh->size = size;
+	memcpy(fh->data, p, size);
+	return 0;
+}
+
 static int xdr_decode_fhstatus3(struct rpc_rqst *req, __be32 *p,
 				struct mnt_fhstatus *res)
 {
-- 
cgit v1.2.3-59-g8ed1b


From a14017db2852f9393a401a0f64053c331003babf Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:12 -0700
Subject: NFS: add XDR decoder for mountd version 3 auth-flavor lists

Introduce an xdr_stream-based XDR decoder that can unpack the auth-
flavor list returned in a MNT3 reply.

The nfs_mount() function's caller allocates an array, and passes the
size and a pointer to it.  The decoder decodes all the flavors it can
into the array, and returns the number of decoded flavors.

If the caller is not interested in the auth flavors, it can pass a
value of zero as the size of the pre-allocated array.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/internal.h   |  6 ++++++
 fs/nfs/mount_clnt.c | 37 +++++++++++++++++++++++++++++++++++++
 2 files changed, 43 insertions(+)

diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index e4d6a8348adf..0207758de44a 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -29,6 +29,12 @@ struct nfs_clone_mount {
 	rpc_authflavor_t authflavor;
 };
 
+/*
+ * Note: RFC 1813 doesn't limit the number of auth flavors that
+ * a server can return, so make something up.
+ */
+#define NFS_MAX_SECFLAVORS	(12)
+
 /*
  * In-kernel mount arguments
  */
diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index dd8c6f448eaa..ed0a27ed538d 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -33,6 +33,7 @@
 #define MNT_fhs_status_sz	(1)
 #define MNT_fhandle_sz		XDR_QUADLEN(NFS2_FHSIZE)
 #define MNT_fhandle3_sz		(1 + XDR_QUADLEN(NFS3_FHSIZE))
+#define MNT_authflav3_sz	(1 + NFS_MAX_SECFLAVORS)
 
 /*
  * XDR argument and result sizes
@@ -122,6 +123,8 @@ static struct {
 struct mountres {
 	int errno;
 	struct nfs_fh *fh;
+	unsigned int *auth_count;
+	rpc_authflavor_t *auth_flavors;
 };
 
 struct mnt_fhstatus {
@@ -334,6 +337,40 @@ static int decode_fhandle3(struct xdr_stream *xdr, struct mountres *res)
 	return 0;
 }
 
+static int decode_auth_flavors(struct xdr_stream *xdr, struct mountres *res)
+{
+	rpc_authflavor_t *flavors = res->auth_flavors;
+	unsigned int *count = res->auth_count;
+	u32 entries, i;
+	__be32 *p;
+
+	if (*count == 0)
+		return 0;
+
+	p = xdr_inline_decode(xdr, sizeof(entries));
+	if (unlikely(p == NULL))
+		return -EIO;
+	entries = ntohl(*p);
+	dprintk("NFS: received %u auth flavors\n", entries);
+	if (entries > NFS_MAX_SECFLAVORS)
+		entries = NFS_MAX_SECFLAVORS;
+
+	p = xdr_inline_decode(xdr, sizeof(u32) * entries);
+	if (unlikely(p == NULL))
+		return -EIO;
+
+	if (entries > *count)
+		entries = *count;
+
+	for (i = 0; i < entries; i++) {
+		flavors[i] = ntohl(*p++);
+		dprintk("NFS:\tflavor %u: %d\n", i, flavors[i]);
+	}
+	*count = i;
+
+	return 0;
+}
+
 static int xdr_decode_fhstatus3(struct rpc_rqst *req, __be32 *p,
 				struct mnt_fhstatus *res)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 8e02f6b9aae9b265064f929c6df15222b9baf256 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:13 -0700
Subject: NFS: Update MNT and MNT3 reply decoding functions

Solder xdr_stream-based XDR decoding functions into the in-kernel mountd
client that are more careful about checking data types and watching for
buffer overflows.  The new MNT3 decoder includes support for auth-flavor
list decoding.

The "_sz" macro for MNT3 replies was missing the size of the file handle.
I've added this back, and included the size of the auth flavor array.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/internal.h   |  2 ++
 fs/nfs/mount_clnt.c | 63 +++++++++++++++++++++++++++++++++++++++++------------
 fs/nfs/nfsroot.c    |  2 ++
 fs/nfs/super.c      |  2 ++
 4 files changed, 55 insertions(+), 14 deletions(-)

diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 0207758de44a..10cf1110df48 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -83,6 +83,8 @@ struct nfs_mount_request {
 	unsigned short		protocol;
 	struct nfs_fh		*fh;
 	int			noresvport;
+	unsigned int		*auth_flav_len;
+	rpc_authflavor_t	*auth_flavs;
 };
 
 extern int nfs_mount(struct nfs_mount_request *info);
diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index ed0a27ed538d..618d815c3093 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -39,6 +39,9 @@
  * XDR argument and result sizes
  */
 #define MNT_enc_dirpath_sz	encode_dirpath_sz
+#define MNT_dec_mountres_sz	(MNT_status_sz + MNT_fhandle_sz)
+#define MNT_dec_mountres3_sz	(MNT_status_sz + MNT_fhandle_sz + \
+				 MNT_authflav3_sz)
 
 /*
  * Defined by RFC 1094, section A.5
@@ -140,8 +143,10 @@ struct mnt_fhstatus {
  */
 int nfs_mount(struct nfs_mount_request *info)
 {
-	struct mnt_fhstatus	result = {
-		.fh		= info->fh
+	struct mountres	result = {
+		.fh		= info->fh,
+		.auth_count	= info->auth_flav_len,
+		.auth_flavors	= info->auth_flavs,
 	};
 	struct rpc_message msg	= {
 		.rpc_argp	= info->dirpath,
@@ -180,7 +185,7 @@ int nfs_mount(struct nfs_mount_request *info)
 
 	if (status < 0)
 		goto out_call_err;
-	if (result.status != 0)
+	if (result.errno != 0)
 		goto out_mnt_err;
 
 	dprintk("NFS: MNT request succeeded\n");
@@ -191,16 +196,16 @@ out:
 
 out_clnt_err:
 	status = PTR_ERR(mnt_clnt);
-	dprintk("NFS: failed to create RPC client, status=%d\n", status);
+	dprintk("NFS: failed to create MNT RPC client, status=%d\n", status);
 	goto out;
 
 out_call_err:
-	dprintk("NFS: failed to start MNT request, status=%d\n", status);
+	dprintk("NFS: MNT request failed, status=%d\n", status);
 	goto out;
 
 out_mnt_err:
-	dprintk("NFS: MNT server returned result %d\n", result.status);
-	status = nfs_stat_to_errno(result.status);
+	dprintk("NFS: MNT server returned result %d\n", result.errno);
+	status = result.errno;
 	goto out;
 }
 
@@ -291,6 +296,20 @@ static int decode_fhandle(struct xdr_stream *xdr, struct mountres *res)
 	return 0;
 }
 
+static int mnt_dec_mountres(struct rpc_rqst *req, __be32 *p,
+			    struct mountres *res)
+{
+	struct xdr_stream xdr;
+	int status;
+
+	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
+
+	status = decode_status(&xdr, res);
+	if (unlikely(status != 0 || res->errno != 0))
+		return status;
+	return decode_fhandle(&xdr, res);
+}
+
 static int decode_fhs_status(struct xdr_stream *xdr, struct mountres *res)
 {
 	unsigned int i;
@@ -371,6 +390,25 @@ static int decode_auth_flavors(struct xdr_stream *xdr, struct mountres *res)
 	return 0;
 }
 
+static int mnt_dec_mountres3(struct rpc_rqst *req, __be32 *p,
+			     struct mountres *res)
+{
+	struct xdr_stream xdr;
+	int status;
+
+	xdr_init_decode(&xdr, &req->rq_rcv_buf, p);
+
+	status = decode_fhs_status(&xdr, res);
+	if (unlikely(status != 0 || res->errno != 0))
+		return status;
+	status = decode_fhandle3(&xdr, res);
+	if (unlikely(status != 0)) {
+		res->errno = -EBADHANDLE;
+		return 0;
+	}
+	return decode_auth_flavors(&xdr, res);
+}
+
 static int xdr_decode_fhstatus3(struct rpc_rqst *req, __be32 *p,
 				struct mnt_fhstatus *res)
 {
@@ -388,16 +426,13 @@ static int xdr_decode_fhstatus3(struct rpc_rqst *req, __be32 *p,
 	return 0;
 }
 
-#define MNT_fhstatus_sz		(1 + 8)
-#define MNT_fhstatus3_sz	(1 + 16)
-
 static struct rpc_procinfo mnt_procedures[] = {
 	[MOUNTPROC_MNT] = {
 		.p_proc		= MOUNTPROC_MNT,
 		.p_encode	= (kxdrproc_t)mnt_enc_dirpath,
-		.p_decode	= (kxdrproc_t) xdr_decode_fhstatus,
+		.p_decode	= (kxdrproc_t)mnt_dec_mountres,
 		.p_arglen	= MNT_enc_dirpath_sz,
-		.p_replen	= MNT_fhstatus_sz,
+		.p_replen	= MNT_dec_mountres_sz,
 		.p_statidx	= MOUNTPROC_MNT,
 		.p_name		= "MOUNT",
 	},
@@ -407,9 +442,9 @@ static struct rpc_procinfo mnt3_procedures[] = {
 	[MOUNTPROC3_MNT] = {
 		.p_proc		= MOUNTPROC3_MNT,
 		.p_encode	= (kxdrproc_t)mnt_enc_dirpath,
-		.p_decode	= (kxdrproc_t) xdr_decode_fhstatus3,
+		.p_decode	= (kxdrproc_t)mnt_dec_mountres3,
 		.p_arglen	= MNT_enc_dirpath_sz,
-		.p_replen	= MNT_fhstatus3_sz,
+		.p_replen	= MNT_dec_mountres3_sz,
 		.p_statidx	= MOUNTPROC3_MNT,
 		.p_name		= "MOUNT",
 	},
diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c
index 24c1b93874c4..8c55b27c0de4 100644
--- a/fs/nfs/nfsroot.c
+++ b/fs/nfs/nfsroot.c
@@ -490,6 +490,7 @@ static int __init root_nfs_get_handle(void)
 {
 	struct nfs_fh fh;
 	struct sockaddr_in sin;
+	unsigned int auth_flav_len = 0;
 	struct nfs_mount_request request = {
 		.sap		= (struct sockaddr *)&sin,
 		.salen		= sizeof(sin),
@@ -499,6 +500,7 @@ static int __init root_nfs_get_handle(void)
 		.protocol	= (nfs_data.flags & NFS_MOUNT_TCP) ?
 					XPRT_TRANSPORT_TCP : XPRT_TRANSPORT_UDP,
 		.fh		= &fh,
+		.auth_flav_len	= &auth_flav_len,
 	};
 	int status;
 
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 539a61a0887b..0b72357cdc41 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -1380,6 +1380,7 @@ out_security_failure:
 static int nfs_try_mount(struct nfs_parsed_mount_data *args,
 			 struct nfs_fh *root_fh)
 {
+	unsigned int auth_flavor_len = 0;
 	struct nfs_mount_request request = {
 		.sap		= (struct sockaddr *)
 						&args->mount_server.address,
@@ -1387,6 +1388,7 @@ static int nfs_try_mount(struct nfs_parsed_mount_data *args,
 		.protocol	= args->mount_server.protocol,
 		.fh		= root_fh,
 		.noresvport	= args->flags & NFS_MOUNT_NORESVPORT,
+		.auth_flav_len	= &auth_flavor_len,
 	};
 	int status;
 
-- 
cgit v1.2.3-59-g8ed1b


From 065015e5efff60884ad600a3e9a5127dbb684429 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:13 -0700
Subject: NFS: Remove unused XDR decoder functions

Clean up: Remove xdr_decode_fhstatus() and xdr_decode_fhstatus3(), now
that they are unused.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/mount_clnt.c | 29 -----------------------------
 1 file changed, 29 deletions(-)

diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c
index 618d815c3093..38ef9eaec407 100644
--- a/fs/nfs/mount_clnt.c
+++ b/fs/nfs/mount_clnt.c
@@ -238,18 +238,6 @@ static int mnt_enc_dirpath(struct rpc_rqst *req, __be32 *p,
 	return encode_mntdirpath(&xdr, dirpath);
 }
 
-static int xdr_decode_fhstatus(struct rpc_rqst *req, __be32 *p,
-			       struct mnt_fhstatus *res)
-{
-	struct nfs_fh *fh = res->fh;
-
-	if ((res->status = ntohl(*p++)) == 0) {
-		fh->size = NFS2_FHSIZE;
-		memcpy(fh->data, p, NFS2_FHSIZE);
-	}
-	return 0;
-}
-
 /*
  * RFC 1094: "A non-zero status indicates some sort of error.  In this
  * case, the status is a UNIX error number."  This can be problematic
@@ -409,23 +397,6 @@ static int mnt_dec_mountres3(struct rpc_rqst *req, __be32 *p,
 	return decode_auth_flavors(&xdr, res);
 }
 
-static int xdr_decode_fhstatus3(struct rpc_rqst *req, __be32 *p,
-				struct mnt_fhstatus *res)
-{
-	struct nfs_fh *fh = res->fh;
-	unsigned size;
-
-	if ((res->status = ntohl(*p++)) == 0) {
-		size = ntohl(*p++);
-		if (size <= NFS3_FHSIZE && size != 0) {
-			fh->size = size;
-			memcpy(fh->data, p, size);
-		} else
-			res->status = -EBADHANDLE;
-	}
-	return 0;
-}
-
 static struct rpc_procinfo mnt_procedures[] = {
 	[MOUNTPROC_MNT] = {
 		.p_proc		= MOUNTPROC_MNT,
-- 
cgit v1.2.3-59-g8ed1b


From d23c45fd84f79a3b84899dac053dcafe9d43ebc9 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:13 -0700
Subject: NFS: Invalid mount option values should always fail, even with
 "sloppy"

Ian Kent reports:

"I've noticed a couple of other regressions with the options vers
and proto option of mount.nfs(8).

The commands:

mount -t nfs -o vers=<invalid version> <server>:/<path> /<mountpoint>
mount -t nfs -o proto=<invalid proto> <server>:/<path> /<mountpoint>

both immediately fail.

But if the "-s" option is also used they both succeed with the
mount falling back to defaults (by the look of it).

In the past these failed even when the sloppy option was given, as
I think they should. I believe the sloppy option is meant to allow
the mount command to still function for mount options (for example
in shared autofs maps) that exist on other Unix implementations but
aren't present in the Linux mount.nfs(8). So, an invalid value
specified for a known mount option is different to an unknown mount
option and should fail appropriately."

See RH bugzilla 486266.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/super.c | 157 ++++++++++++++++++++++-----------------------------------
 1 file changed, 61 insertions(+), 96 deletions(-)

diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 0b72357cdc41..a2b2805caf9d 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -942,11 +942,6 @@ static int nfs_parse_security_flavors(char *value,
 	return 1;
 }
 
-static void nfs_parse_invalid_value(const char *option)
-{
-	dfprintk(MOUNT, "NFS:   bad value specified for %s option\n", option);
-}
-
 /*
  * Error-check and convert a string of mount options from user space into
  * a data structure.  The whole mount string is processed; bad options are
@@ -957,7 +952,7 @@ static int nfs_parse_mount_options(char *raw,
 				   struct nfs_parsed_mount_data *mnt)
 {
 	char *p, *string, *secdata;
-	int rc, sloppy = 0, errors = 0;
+	int rc, sloppy = 0, invalid_option = 0;
 
 	if (!raw) {
 		dfprintk(MOUNT, "NFS: mount options string was NULL.\n");
@@ -1091,113 +1086,82 @@ static int nfs_parse_mount_options(char *raw,
 		 */
 		case Opt_port:
 			if (match_int(args, &option) ||
-			    option < 0 || option > USHORT_MAX) {
-				errors++;
-				nfs_parse_invalid_value("port");
-			} else
-				mnt->nfs_server.port = option;
+			    option < 0 || option > USHORT_MAX)
+				goto out_invalid_value;
+			mnt->nfs_server.port = option;
 			break;
 		case Opt_rsize:
-			if (match_int(args, &option) || option < 0) {
-				errors++;
-				nfs_parse_invalid_value("rsize");
-			} else
-				mnt->rsize = option;
+			if (match_int(args, &option) || option < 0)
+				goto out_invalid_value;
+			mnt->rsize = option;
 			break;
 		case Opt_wsize:
-			if (match_int(args, &option) || option < 0) {
-				errors++;
-				nfs_parse_invalid_value("wsize");
-			} else
-				mnt->wsize = option;
+			if (match_int(args, &option) || option < 0)
+				goto out_invalid_value;
+			mnt->wsize = option;
 			break;
 		case Opt_bsize:
-			if (match_int(args, &option) || option < 0) {
-				errors++;
-				nfs_parse_invalid_value("bsize");
-			} else
-				mnt->bsize = option;
+			if (match_int(args, &option) || option < 0)
+				goto out_invalid_value;
+			mnt->bsize = option;
 			break;
 		case Opt_timeo:
-			if (match_int(args, &option) || option <= 0) {
-				errors++;
-				nfs_parse_invalid_value("timeo");
-			} else
-				mnt->timeo = option;
+			if (match_int(args, &option) || option <= 0)
+				goto out_invalid_value;
+			mnt->timeo = option;
 			break;
 		case Opt_retrans:
-			if (match_int(args, &option) || option <= 0) {
-				errors++;
-				nfs_parse_invalid_value("retrans");
-			} else
-				mnt->retrans = option;
+			if (match_int(args, &option) || option <= 0)
+				goto out_invalid_value;
+			mnt->retrans = option;
 			break;
 		case Opt_acregmin:
-			if (match_int(args, &option) || option < 0) {
-				errors++;
-				nfs_parse_invalid_value("acregmin");
-			} else
-				mnt->acregmin = option;
+			if (match_int(args, &option) || option < 0)
+				goto out_invalid_value;
+			mnt->acregmin = option;
 			break;
 		case Opt_acregmax:
-			if (match_int(args, &option) || option < 0) {
-				errors++;
-				nfs_parse_invalid_value("acregmax");
-			} else
-				mnt->acregmax = option;
+			if (match_int(args, &option) || option < 0)
+				goto out_invalid_value;
+			mnt->acregmax = option;
 			break;
 		case Opt_acdirmin:
-			if (match_int(args, &option) || option < 0) {
-				errors++;
-				nfs_parse_invalid_value("acdirmin");
-			} else
-				mnt->acdirmin = option;
+			if (match_int(args, &option) || option < 0)
+				goto out_invalid_value;
+			mnt->acdirmin = option;
 			break;
 		case Opt_acdirmax:
-			if (match_int(args, &option) || option < 0) {
-				errors++;
-				nfs_parse_invalid_value("acdirmax");
-			} else
-				mnt->acdirmax = option;
+			if (match_int(args, &option) || option < 0)
+				goto out_invalid_value;
+			mnt->acdirmax = option;
 			break;
 		case Opt_actimeo:
-			if (match_int(args, &option) || option < 0) {
-				errors++;
-				nfs_parse_invalid_value("actimeo");
-			} else
-				mnt->acregmin = mnt->acregmax =
-				mnt->acdirmin = mnt->acdirmax = option;
+			if (match_int(args, &option) || option < 0)
+				goto out_invalid_value;
+			mnt->acregmin = mnt->acregmax =
+			mnt->acdirmin = mnt->acdirmax = option;
 			break;
 		case Opt_namelen:
-			if (match_int(args, &option) || option < 0) {
-				errors++;
-				nfs_parse_invalid_value("namlen");
-			} else
-				mnt->namlen = option;
+			if (match_int(args, &option) || option < 0)
+				goto out_invalid_value;
+			mnt->namlen = option;
 			break;
 		case Opt_mountport:
 			if (match_int(args, &option) ||
-			    option < 0 || option > USHORT_MAX) {
-				errors++;
-				nfs_parse_invalid_value("mountport");
-			} else
-				mnt->mount_server.port = option;
+			    option < 0 || option > USHORT_MAX)
+				goto out_invalid_value;
+			mnt->mount_server.port = option;
 			break;
 		case Opt_mountvers:
 			if (match_int(args, &option) ||
 			    option < NFS_MNT_VERSION ||
-			    option > NFS_MNT3_VERSION) {
-				errors++;
-				nfs_parse_invalid_value("mountvers");
-			} else
-				mnt->mount_server.version = option;
+			    option > NFS_MNT3_VERSION)
+				goto out_invalid_value;
+			mnt->mount_server.version = option;
 			break;
 		case Opt_nfsvers:
-			if (match_int(args, &option)) {
-				errors++;
-				nfs_parse_invalid_value("nfsvers");
-				break;
-			}
+			if (match_int(args, &option))
+				goto out_invalid_value;
 			switch (option) {
 			case NFS2_VERSION:
 				mnt->flags &= ~NFS_MOUNT_VER3;
@@ -1206,8 +1170,7 @@ static int nfs_parse_mount_options(char *raw,
 				mnt->flags |= NFS_MOUNT_VER3;
 				break;
 			default:
-				errors++;
-				nfs_parse_invalid_value("nfsvers");
+				goto out_invalid_value;
 			}
 			break;
 
@@ -1221,9 +1184,9 @@ static int nfs_parse_mount_options(char *raw,
 			rc = nfs_parse_security_flavors(string, mnt);
 			kfree(string);
 			if (!rc) {
-				errors++;
 				dfprintk(MOUNT, "NFS:   unrecognized "
 						"security flavor\n");
+				return 0;
 			}
 			break;
 		case Opt_proto:
@@ -1237,23 +1200,25 @@ static int nfs_parse_mount_options(char *raw,
 			case Opt_xprt_udp:
 				mnt->flags &= ~NFS_MOUNT_TCP;
 				mnt->nfs_server.protocol = XPRT_TRANSPORT_UDP;
+				kfree(string);
 				break;
 			case Opt_xprt_tcp:
 				mnt->flags |= NFS_MOUNT_TCP;
 				mnt->nfs_server.protocol = XPRT_TRANSPORT_TCP;
+				kfree(string);
 				break;
 			case Opt_xprt_rdma:
 				/* vector side protocols to TCP */
 				mnt->flags |= NFS_MOUNT_TCP;
 				mnt->nfs_server.protocol = XPRT_TRANSPORT_RDMA;
 				xprt_load_transport(string);
+				kfree(string);
 				break;
 			default:
-				errors++;
 				dfprintk(MOUNT, "NFS:   unrecognized "
 						"transport protocol\n");
+				return 0;
 			}
-			kfree(string);
 			break;
 		case Opt_mountproto:
 			string = match_strdup(args);
@@ -1272,9 +1237,9 @@ static int nfs_parse_mount_options(char *raw,
 				break;
 			case Opt_xprt_rdma: /* not used for side protocols */
 			default:
-				errors++;
 				dfprintk(MOUNT, "NFS:   unrecognized "
 						"transport protocol\n");
+				return 0;
 			}
 			break;
 		case Opt_addr:
@@ -1330,9 +1295,9 @@ static int nfs_parse_mount_options(char *raw,
 					mnt->flags |= NFS_MOUNT_LOOKUP_CACHE_NONEG|NFS_MOUNT_LOOKUP_CACHE_NONE;
 					break;
 				default:
-					errors++;
 					dfprintk(MOUNT, "NFS:   invalid "
 							"lookupcache argument\n");
+					return 0;
 			};
 			break;
 
@@ -1350,20 +1315,20 @@ static int nfs_parse_mount_options(char *raw,
 			break;
 
 		default:
-			errors++;
+			invalid_option = 1;
 			dfprintk(MOUNT, "NFS:   unrecognized mount option "
 					"'%s'\n", p);
 		}
 	}
 
-	if (errors > 0) {
-		dfprintk(MOUNT, "NFS: parsing encountered %d error%s\n",
-				errors, (errors == 1 ? "" : "s"));
-		if (!sloppy)
-			return 0;
-	}
+	if (!sloppy && invalid_option)
+		return 0;
+
 	return 1;
 
+out_invalid_value:
+	printk(KERN_INFO "NFS: bad mount option value specified: %s \n", p);
+	return 0;
 out_nomem:
 	printk(KERN_INFO "NFS: not enough memory to parse option\n");
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From a5a16bae707cd5d2bc97d7bd1a30079f18113a77 Mon Sep 17 00:00:00 2001
From: Chuck Lever <chuck.lever@oracle.com>
Date: Wed, 17 Jun 2009 18:02:14 -0700
Subject: NFS: More "sloppy" parsing problems

Specifying "port=-5" with the kernel's current mount option parser
generates "unrecognized mount option".  If "sloppy" is set, this
causes the mount to succeed and use the default values; the desired
behavior is that, since this is a valid option with an invalid value,
the mount should fail, even with "sloppy."

To properly handle "sloppy" parsing, we need to distinguish between
correct options with invalid values, and incorrect options.  We will
need to parse integer values by hand, therefore, and not rely on
match_token().

For instance, these must all fail with "invalid value":

	port=12345678
	port=-5
	port=samuel

and not with "unrecognized option," as they do currently.

Thus, for the sake of match_token() we need to treat the values for
these options as strings, and do the conversion to integers using
strict_strtol().

This is basically the same solution we used for the earlier "retry="
fix (commit ecbb3845), except in this case the kernel actually has to
parse the value, rather than ignore it.

Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/super.c | 142 +++++++++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 108 insertions(+), 34 deletions(-)

diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index a2b2805caf9d..b798ed1bd36d 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -139,22 +139,22 @@ static const match_table_t nfs_mount_option_tokens = {
 	{ Opt_fscache_uniq, "fsc=%s" },
 	{ Opt_nofscache, "nofsc" },
 
-	{ Opt_port, "port=%u" },
-	{ Opt_rsize, "rsize=%u" },
-	{ Opt_wsize, "wsize=%u" },
-	{ Opt_bsize, "bsize=%u" },
-	{ Opt_timeo, "timeo=%u" },
-	{ Opt_retrans, "retrans=%u" },
-	{ Opt_acregmin, "acregmin=%u" },
-	{ Opt_acregmax, "acregmax=%u" },
-	{ Opt_acdirmin, "acdirmin=%u" },
-	{ Opt_acdirmax, "acdirmax=%u" },
-	{ Opt_actimeo, "actimeo=%u" },
-	{ Opt_namelen, "namlen=%u" },
-	{ Opt_mountport, "mountport=%u" },
-	{ Opt_mountvers, "mountvers=%u" },
-	{ Opt_nfsvers, "nfsvers=%u" },
-	{ Opt_nfsvers, "vers=%u" },
+	{ Opt_port, "port=%s" },
+	{ Opt_rsize, "rsize=%s" },
+	{ Opt_wsize, "wsize=%s" },
+	{ Opt_bsize, "bsize=%s" },
+	{ Opt_timeo, "timeo=%s" },
+	{ Opt_retrans, "retrans=%s" },
+	{ Opt_acregmin, "acregmin=%s" },
+	{ Opt_acregmax, "acregmax=%s" },
+	{ Opt_acdirmin, "acdirmin=%s" },
+	{ Opt_acdirmax, "acdirmax=%s" },
+	{ Opt_actimeo, "actimeo=%s" },
+	{ Opt_namelen, "namlen=%s" },
+	{ Opt_mountport, "mountport=%s" },
+	{ Opt_mountvers, "mountvers=%s" },
+	{ Opt_nfsvers, "nfsvers=%s" },
+	{ Opt_nfsvers, "vers=%s" },
 
 	{ Opt_sec, "sec=%s" },
 	{ Opt_proto, "proto=%s" },
@@ -976,7 +976,8 @@ static int nfs_parse_mount_options(char *raw,
 
 	while ((p = strsep(&raw, ",")) != NULL) {
 		substring_t args[MAX_OPT_ARGS];
-		int option, token;
+		unsigned long option;
+		int token;
 
 		if (!*p)
 			continue;
@@ -1085,82 +1086,155 @@ static int nfs_parse_mount_options(char *raw,
 		 * options that take numeric values
 		 */
 		case Opt_port:
-			if (match_int(args, &option) ||
-			    option < 0 || option > USHORT_MAX)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0 || option > USHORT_MAX)
 				goto out_invalid_value;
 			mnt->nfs_server.port = option;
 			break;
 		case Opt_rsize:
-			if (match_int(args, &option) || option < 0)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0)
 				goto out_invalid_value;
 			mnt->rsize = option;
 			break;
 		case Opt_wsize:
-			if (match_int(args, &option) || option < 0)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0)
 				goto out_invalid_value;
 			mnt->wsize = option;
 			break;
 		case Opt_bsize:
-			if (match_int(args, &option) || option < 0)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0)
 				goto out_invalid_value;
 			mnt->bsize = option;
 			break;
 		case Opt_timeo:
-			if (match_int(args, &option) || option <= 0)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0 || option == 0)
 				goto out_invalid_value;
 			mnt->timeo = option;
 			break;
 		case Opt_retrans:
-			if (match_int(args, &option) || option <= 0)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0 || option == 0)
 				goto out_invalid_value;
 			mnt->retrans = option;
 			break;
 		case Opt_acregmin:
-			if (match_int(args, &option) || option < 0)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0)
 				goto out_invalid_value;
 			mnt->acregmin = option;
 			break;
 		case Opt_acregmax:
-			if (match_int(args, &option) || option < 0)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0)
 				goto out_invalid_value;
 			mnt->acregmax = option;
 			break;
 		case Opt_acdirmin:
-			if (match_int(args, &option) || option < 0)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0)
 				goto out_invalid_value;
 			mnt->acdirmin = option;
 			break;
 		case Opt_acdirmax:
-			if (match_int(args, &option) || option < 0)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0)
 				goto out_invalid_value;
 			mnt->acdirmax = option;
 			break;
 		case Opt_actimeo:
-			if (match_int(args, &option) || option < 0)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0)
 				goto out_invalid_value;
 			mnt->acregmin = mnt->acregmax =
 			mnt->acdirmin = mnt->acdirmax = option;
 			break;
 		case Opt_namelen:
-			if (match_int(args, &option) || option < 0)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0)
 				goto out_invalid_value;
 			mnt->namlen = option;
 			break;
 		case Opt_mountport:
-			if (match_int(args, &option) ||
-			    option < 0 || option > USHORT_MAX)
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0 || option > USHORT_MAX)
 				goto out_invalid_value;
 			mnt->mount_server.port = option;
 			break;
 		case Opt_mountvers:
-			if (match_int(args, &option) ||
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0 ||
 			    option < NFS_MNT_VERSION ||
 			    option > NFS_MNT3_VERSION)
 				goto out_invalid_value;
 			mnt->mount_server.version = option;
 			break;
 		case Opt_nfsvers:
-			if (match_int(args, &option))
+			string = match_strdup(args);
+			if (string == NULL)
+				goto out_nomem;
+			rc = strict_strtoul(string, 10, &option);
+			kfree(string);
+			if (rc != 0)
 				goto out_invalid_value;
 			switch (option) {
 			case NFS2_VERSION:
-- 
cgit v1.2.3-59-g8ed1b


From 4bf259e3ae5015e73282ba66716c4a917e1264ac Mon Sep 17 00:00:00 2001
From: James Morris <jmorris@namei.org>
Date: Wed, 17 Jun 2009 18:02:14 -0700
Subject: nfs: remove unnecessary NFS_INO_INVALID_ACL checks

Unless I'm mistaken, NFS_INO_INVALID_ACL is being checked twice during
getacl calls (i.e. first via nfs_revalidate_inode() and then by each all
site).

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs3acl.c  | 2 --
 fs/nfs/nfs4proc.c | 2 --
 2 files changed, 4 deletions(-)

diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c
index 6bbf0e6daad2..bac60515a4b3 100644
--- a/fs/nfs/nfs3acl.c
+++ b/fs/nfs/nfs3acl.c
@@ -207,8 +207,6 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type)
 	status = nfs_revalidate_inode(server, inode);
 	if (status < 0)
 		return ERR_PTR(status);
-	if (NFS_I(inode)->cache_validity & NFS_INO_INVALID_ACL)
-		nfs_zap_acl_cache(inode);
 	acl = nfs3_get_cached_acl(inode, type);
 	if (acl != ERR_PTR(-EAGAIN))
 		return acl;
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 55314e721632..42e6a3b05498 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2824,8 +2824,6 @@ static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen)
 	ret = nfs_revalidate_inode(server, inode);
 	if (ret < 0)
 		return ret;
-	if (NFS_I(inode)->cache_validity & NFS_INO_INVALID_ACL)
-		nfs_zap_acl_cache(inode);
 	ret = nfs4_read_cached_acl(inode, buf, buflen);
 	if (ret != -ENOENT)
 		return ret;
-- 
cgit v1.2.3-59-g8ed1b


From 050df107c408a3df048524b3783a5fc6d4dccfdb Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Sat, 30 May 2009 13:25:05 -0300
Subject: thinkpad-acpi: store fw version with strict checking

Extend the thinkpad model and firmware identification data with the
release serial number for the BIOS and firmware (when available), as
that is easier to parse and compare than the version strings.

We're going to greatly extend the use of the ThinkPad DMI data through
quirk lists, so it is best to be quite strict and make sure what we
get from DMI is exactly what we expect, otherwise quirk matching may
result in quite insane things.

IBM (and Lenovo, at least for the ThinkPad line) uses this schema for
firmware versioning and model:

Firmware model: Two digits, [0-9A-Z]

Firmware version: AABBCCDD, where
  AA = firmware model, see above
  BB = "ET" for BIOS, "HT" for EC
  CC = release version, two digits, [0-9A-Z],
       "00" < "09" < "0A" < "10" < "A0" < "ZZ"
  DD = "WW"

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/thinkpad_acpi.c | 46 ++++++++++++++++++++++++++++++++----
 1 file changed, 42 insertions(+), 4 deletions(-)

diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c
index 912be65b6261..d2a0ef83fbb5 100644
--- a/drivers/platform/x86/thinkpad_acpi.c
+++ b/drivers/platform/x86/thinkpad_acpi.c
@@ -284,8 +284,10 @@ struct thinkpad_id_data {
 	char *bios_version_str;	/* Something like 1ZET51WW (1.03z) */
 	char *ec_version_str;	/* Something like 1ZHT51WW-1.04a */
 
-	u16 bios_model;		/* Big Endian, TP-1Y = 0x5931, 0 = unknown */
+	u16 bios_model;		/* 1Y = 0x5931, 0 = unknown */
 	u16 ec_model;
+	u16 bios_release;	/* 1ZETK1WW = 0x314b, 0 = unknown */
+	u16 ec_release;
 
 	char *model_str;	/* ThinkPad T43 */
 	char *nummodel_str;	/* 9384A9C for a 9384-A9C model */
@@ -7357,6 +7359,24 @@ err_out:
 
 /* Probing */
 
+static bool __pure __init tpacpi_is_fw_digit(const char c)
+{
+	return (c >= '0' && c <= '9') || (c >= 'A' && c <= 'Z');
+}
+
+/* Most models: xxyTkkWW (#.##c); Ancient 570/600 and -SL lacks (#.##c) */
+static bool __pure __init tpacpi_is_valid_fw_id(const char* const s,
+						const char t)
+{
+	return s && strlen(s) >= 8 &&
+		tpacpi_is_fw_digit(s[0]) &&
+		tpacpi_is_fw_digit(s[1]) &&
+		s[2] == t && s[3] == 'T' &&
+		tpacpi_is_fw_digit(s[4]) &&
+		tpacpi_is_fw_digit(s[5]) &&
+		s[6] == 'W' && s[7] == 'W';
+}
+
 /* returns 0 - probe ok, or < 0 - probe error.
  * Probe ok doesn't mean thinkpad found.
  * On error, kfree() cleanup on tp->* is not performed, caller must do it */
@@ -7383,10 +7403,15 @@ static int __must_check __init get_thinkpad_model_data(
 	tp->bios_version_str = kstrdup(s, GFP_KERNEL);
 	if (s && !tp->bios_version_str)
 		return -ENOMEM;
-	if (!tp->bios_version_str)
+
+	/* Really ancient ThinkPad 240X will fail this, which is fine */
+	if (!tpacpi_is_valid_fw_id(tp->bios_version_str, 'E'))
 		return 0;
+
 	tp->bios_model = tp->bios_version_str[0]
 			 | (tp->bios_version_str[1] << 8);
+	tp->bios_release = (tp->bios_version_str[4] << 8)
+			 | tp->bios_version_str[5];
 
 	/*
 	 * ThinkPad T23 or newer, A31 or newer, R50e or newer,
@@ -7405,8 +7430,21 @@ static int __must_check __init get_thinkpad_model_data(
 			tp->ec_version_str = kstrdup(ec_fw_string, GFP_KERNEL);
 			if (!tp->ec_version_str)
 				return -ENOMEM;
-			tp->ec_model = ec_fw_string[0]
-					| (ec_fw_string[1] << 8);
+
+			if (tpacpi_is_valid_fw_id(ec_fw_string, 'H')) {
+				tp->ec_model = ec_fw_string[0]
+						| (ec_fw_string[1] << 8);
+				tp->ec_release = (ec_fw_string[4] << 8)
+						| ec_fw_string[5];
+			} else {
+				printk(TPACPI_NOTICE
+					"ThinkPad firmware release %s "
+					"doesn't match the known patterns\n",
+					ec_fw_string);
+				printk(TPACPI_NOTICE
+					"please report this to %s\n",
+					TPACPI_MAIL);
+			}
 			break;
 		}
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 7d95a3d564901e88ed42810f054e579874151999 Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Sat, 30 May 2009 13:25:06 -0300
Subject: thinkpad-acpi: add quirklist engine

Add a quirklist engine suitable for matching ThinkPad firmware,
and change the code to use it.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/thinkpad_acpi.c | 119 ++++++++++++++++++++++++++++-------
 1 file changed, 98 insertions(+), 21 deletions(-)

diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c
index d2a0ef83fbb5..3981b060b7d5 100644
--- a/drivers/platform/x86/thinkpad_acpi.c
+++ b/drivers/platform/x86/thinkpad_acpi.c
@@ -364,6 +364,73 @@ static void tpacpi_log_usertask(const char * const what)
 		} \
 	} while (0)
 
+/*
+ * Quirk handling helpers
+ *
+ * ThinkPad IDs and versions seen in the field so far
+ * are two-characters from the set [0-9A-Z], i.e. base 36.
+ *
+ * We use values well outside that range as specials.
+ */
+
+#define TPACPI_MATCH_ANY		0xffffU
+#define TPACPI_MATCH_UNKNOWN		0U
+
+/* TPID('1', 'Y') == 0x5931 */
+#define TPID(__c1, __c2) (((__c2) << 8) | (__c1))
+
+#define TPACPI_Q_IBM(__id1, __id2, __quirk)	\
+	{ .vendor = PCI_VENDOR_ID_IBM,		\
+	  .bios = TPID(__id1, __id2),		\
+	  .ec = TPACPI_MATCH_ANY,		\
+	  .quirks = (__quirk) }
+
+#define TPACPI_Q_LNV(__id1, __id2, __quirk)	\
+	{ .vendor = PCI_VENDOR_ID_LENOVO,	\
+	  .bios = TPID(__id1, __id2),		\
+	  .ec = TPACPI_MATCH_ANY,		\
+	  .quirks = (__quirk) }
+
+struct tpacpi_quirk {
+	unsigned int vendor;
+	u16 bios;
+	u16 ec;
+	unsigned long quirks;
+};
+
+/**
+ * tpacpi_check_quirks() - search BIOS/EC version on a list
+ * @qlist:		array of &struct tpacpi_quirk
+ * @qlist_size:		number of elements in @qlist
+ *
+ * Iterates over a quirks list until one is found that matches the
+ * ThinkPad's vendor, BIOS and EC model.
+ *
+ * Returns 0 if nothing matches, otherwise returns the quirks field of
+ * the matching &struct tpacpi_quirk entry.
+ *
+ * The match criteria is: vendor, ec and bios much match.
+ */
+static unsigned long __init tpacpi_check_quirks(
+			const struct tpacpi_quirk *qlist,
+			unsigned int qlist_size)
+{
+	while (qlist_size) {
+		if ((qlist->vendor == thinkpad_id.vendor ||
+				qlist->vendor == TPACPI_MATCH_ANY) &&
+		    (qlist->bios == thinkpad_id.bios_model ||
+				qlist->bios == TPACPI_MATCH_ANY) &&
+		    (qlist->ec == thinkpad_id.ec_model ||
+				qlist->ec == TPACPI_MATCH_ANY))
+			return qlist->quirks;
+
+		qlist_size--;
+		qlist++;
+	}
+	return 0;
+}
+
+
 /****************************************************************************
  ****************************************************************************
  *
@@ -6223,30 +6290,18 @@ TPACPI_HANDLE(sfan, ec, "SFAN",	/* 570 */
  * We assume 0x07 really means auto mode while this quirk is active,
  * as this is far more likely than the ThinkPad being in level 7,
  * which is only used by the firmware during thermal emergencies.
+ *
+ * Enable for TP-1Y (T43), TP-78 (R51e), TP-76 (R52),
+ * TP-70 (T43, R52), which are known to be buggy.
  */
 
-static void fan_quirk1_detect(void)
+static void fan_quirk1_setup(void)
 {
-	/* In some ThinkPads, neither the EC nor the ACPI
-	 * DSDT initialize the HFSP register, and it ends up
-	 * being initially set to 0x07 when it *could* be
-	 * either 0x07 or 0x80.
-	 *
-	 * Enable for TP-1Y (T43), TP-78 (R51e),
-	 * TP-76 (R52), TP-70 (T43, R52), which are known
-	 * to be buggy. */
 	if (fan_control_initial_status == 0x07) {
-		switch (thinkpad_id.ec_model) {
-		case 0x5931: /* TP-1Y */
-		case 0x3837: /* TP-78 */
-		case 0x3637: /* TP-76 */
-		case 0x3037: /* TP-70 */
-			printk(TPACPI_NOTICE
-			       "fan_init: initial fan status is unknown, "
-			       "assuming it is in auto mode\n");
-			tp_features.fan_ctrl_status_undef = 1;
-			;;
-		}
+		printk(TPACPI_NOTICE
+		       "fan_init: initial fan status is unknown, "
+		       "assuming it is in auto mode\n");
+		tp_features.fan_ctrl_status_undef = 1;
 	}
 }
 
@@ -6804,9 +6859,27 @@ static const struct attribute_group fan_attr_group = {
 	.attrs = fan_attributes,
 };
 
+#define	TPACPI_FAN_Q1	0x0001
+
+#define TPACPI_FAN_QI(__id1, __id2, __quirks)	\
+	{ .vendor = PCI_VENDOR_ID_IBM,		\
+	  .bios = TPACPI_MATCH_ANY,		\
+	  .ec = TPID(__id1, __id2),		\
+	  .quirks = __quirks }
+
+static const struct tpacpi_quirk fan_quirk_table[] __initconst = {
+	TPACPI_FAN_QI('1', 'Y', TPACPI_FAN_Q1),
+	TPACPI_FAN_QI('7', '8', TPACPI_FAN_Q1),
+	TPACPI_FAN_QI('7', '6', TPACPI_FAN_Q1),
+	TPACPI_FAN_QI('7', '0', TPACPI_FAN_Q1),
+};
+
+#undef TPACPI_FAN_QI
+
 static int __init fan_init(struct ibm_init_struct *iibm)
 {
 	int rc;
+	unsigned long quirks;
 
 	vdbg_printk(TPACPI_DBG_INIT | TPACPI_DBG_FAN,
 			"initializing fan subdriver\n");
@@ -6823,6 +6896,9 @@ static int __init fan_init(struct ibm_init_struct *iibm)
 	TPACPI_ACPIHANDLE_INIT(gfan);
 	TPACPI_ACPIHANDLE_INIT(sfan);
 
+	quirks = tpacpi_check_quirks(fan_quirk_table,
+				     ARRAY_SIZE(fan_quirk_table));
+
 	if (gfan_handle) {
 		/* 570, 600e/x, 770e, 770x */
 		fan_status_access_mode = TPACPI_FAN_RD_ACPI_GFAN;
@@ -6832,7 +6908,8 @@ static int __init fan_init(struct ibm_init_struct *iibm)
 		if (likely(acpi_ec_read(fan_status_offset,
 					&fan_control_initial_status))) {
 			fan_status_access_mode = TPACPI_FAN_RD_TPEC;
-			fan_quirk1_detect();
+			if (quirks & TPACPI_FAN_Q1)
+				fan_quirk1_setup();
 		} else {
 			printk(TPACPI_ERR
 			       "ThinkPad ACPI EC access misbehaving, "
-- 
cgit v1.2.3-59-g8ed1b


From 60201732f03c1231742e5872abe55a3bf59849a5 Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Sat, 30 May 2009 13:25:07 -0300
Subject: thinkpad-acpi: fix BEEP ACPI handler warnings

Some ThinkPads want two arguments for BEEP, while others want just
one, causing ACPICA to log warnings like this:

ACPI Warning (nseval-0177): Excess arguments - method [BEEP] needs 1,
found 2 [20080926]

Deal with it.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/thinkpad_acpi.c | 26 ++++++++++++++++++++++++--
 1 file changed, 24 insertions(+), 2 deletions(-)

diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c
index 3981b060b7d5..da739d5c9210 100644
--- a/drivers/platform/x86/thinkpad_acpi.c
+++ b/drivers/platform/x86/thinkpad_acpi.c
@@ -264,6 +264,7 @@ static struct {
 	u32 wan:1;
 	u32 uwb:1;
 	u32 fan_ctrl_status_undef:1;
+	u32 beep_needs_two_args:1;
 	u32 input_device_registered:1;
 	u32 platform_drv_registered:1;
 	u32 platform_drv_attrs_registered:1;
@@ -5142,8 +5143,17 @@ static struct ibm_struct led_driver_data = {
 
 TPACPI_HANDLE(beep, ec, "BEEP");	/* all except R30, R31 */
 
+#define TPACPI_BEEP_Q1 0x0001
+
+static const struct tpacpi_quirk beep_quirk_table[] __initconst = {
+	TPACPI_Q_IBM('I', 'M', TPACPI_BEEP_Q1), /* 570 */
+	TPACPI_Q_IBM('I', 'U', TPACPI_BEEP_Q1), /* 570E - unverified */
+};
+
 static int __init beep_init(struct ibm_init_struct *iibm)
 {
+	unsigned long quirks;
+
 	vdbg_printk(TPACPI_DBG_INIT, "initializing beep subdriver\n");
 
 	TPACPI_ACPIHANDLE_INIT(beep);
@@ -5151,6 +5161,11 @@ static int __init beep_init(struct ibm_init_struct *iibm)
 	vdbg_printk(TPACPI_DBG_INIT, "beep is %s\n",
 		str_supported(beep_handle != NULL));
 
+	quirks = tpacpi_check_quirks(beep_quirk_table,
+				     ARRAY_SIZE(beep_quirk_table));
+
+	tp_features.beep_needs_two_args = !!(quirks & TPACPI_BEEP_Q1);
+
 	return (beep_handle)? 0 : 1;
 }
 
@@ -5182,8 +5197,15 @@ static int beep_write(char *buf)
 			/* beep_cmd set */
 		} else
 			return -EINVAL;
-		if (!acpi_evalf(beep_handle, NULL, NULL, "vdd", beep_cmd, 0))
-			return -EIO;
+		if (tp_features.beep_needs_two_args) {
+			if (!acpi_evalf(beep_handle, NULL, NULL, "vdd",
+					beep_cmd, 0))
+				return -EIO;
+		} else {
+			if (!acpi_evalf(beep_handle, NULL, NULL, "vd",
+					beep_cmd))
+				return -EIO;
+		}
 	}
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From f21179a47ff8d1046a61c1cf5920244997a4a7bb Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Sat, 30 May 2009 13:25:08 -0300
Subject: thinkpad-acpi: enhance led support

Add support for extra LEDs on recent ThinkPads, and avoid registering
with the led class the LEDs which are not available for a given
ThinkPad model.

All non-restricted LEDs are always available through the procfs
interface, as the firmware doesn't care if an attempt is made to
access an invalid LED.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 Documentation/laptops/thinkpad-acpi.txt | 23 +++++++--
 drivers/platform/x86/thinkpad_acpi.c    | 88 ++++++++++++++++++++++++++++-----
 2 files changed, 96 insertions(+), 15 deletions(-)

diff --git a/Documentation/laptops/thinkpad-acpi.txt b/Documentation/laptops/thinkpad-acpi.txt
index e7e9a69069e1..88fc0661de56 100644
--- a/Documentation/laptops/thinkpad-acpi.txt
+++ b/Documentation/laptops/thinkpad-acpi.txt
@@ -920,7 +920,7 @@ The available commands are:
 	echo '<LED number> off' >/proc/acpi/ibm/led
 	echo '<LED number> blink' >/proc/acpi/ibm/led
 
-The <LED number> range is 0 to 7. The set of LEDs that can be
+The <LED number> range is 0 to 15. The set of LEDs that can be
 controlled varies from model to model. Here is the common ThinkPad
 mapping:
 
@@ -932,6 +932,11 @@ mapping:
 	5 - UltraBase battery slot
 	6 - (unknown)
 	7 - standby
+	8 - dock status 1
+	9 - dock status 2
+	10, 11 - (unknown)
+	12 - thinkvantage
+	13, 14, 15 - (unknown)
 
 All of the above can be turned on and off and can be made to blink.
 
@@ -940,10 +945,12 @@ sysfs notes:
 The ThinkPad LED sysfs interface is described in detail by the LED class
 documentation, in Documentation/leds-class.txt.
 
-The leds are named (in LED ID order, from 0 to 7):
+The LEDs are named (in LED ID order, from 0 to 12):
 "tpacpi::power", "tpacpi:orange:batt", "tpacpi:green:batt",
 "tpacpi::dock_active", "tpacpi::bay_active", "tpacpi::dock_batt",
-"tpacpi::unknown_led", "tpacpi::standby".
+"tpacpi::unknown_led", "tpacpi::standby", "tpacpi::dock_status1",
+"tpacpi::dock_status2", "tpacpi::unknown_led2", "tpacpi::unknown_led3",
+"tpacpi::thinkvantage".
 
 Due to limitations in the sysfs LED class, if the status of the LED
 indicators cannot be read due to an error, thinkpad-acpi will report it as
@@ -958,6 +965,12 @@ ThinkPad indicator LED should blink in hardware accelerated mode, use the
 "timer" trigger, and leave the delay_on and delay_off parameters set to
 zero (to request hardware acceleration autodetection).
 
+LEDs that are known not to exist in a given ThinkPad model are not
+made available through the sysfs interface.  If you have a dock and you
+notice there are LEDs listed for your ThinkPad that do not exist (and
+are not in the dock), or if you notice that there are missing LEDs,
+a report to ibm-acpi-devel@lists.sourceforge.net is appreciated.
+
 
 ACPI sounds -- /proc/acpi/ibm/beep
 ----------------------------------
@@ -1555,3 +1568,7 @@ Sysfs interface changelog:
 0x020300:	hotkey enable/disable support removed, attributes
 		hotkey_bios_enabled and hotkey_enable deprecated and
 		marked for removal.
+
+0x020400:	Marker for 16 LEDs support.  Also, LEDs that are known
+		to not exist in a given model are not registered with
+		the LED sysfs class anymore.
diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c
index da739d5c9210..06c7c03c8f2f 100644
--- a/drivers/platform/x86/thinkpad_acpi.c
+++ b/drivers/platform/x86/thinkpad_acpi.c
@@ -22,7 +22,7 @@
  */
 
 #define TPACPI_VERSION "0.23"
-#define TPACPI_SYSFS_VERSION 0x020300
+#define TPACPI_SYSFS_VERSION 0x020400
 
 /*
  *  Changelog:
@@ -4815,7 +4815,7 @@ TPACPI_HANDLE(led, ec, "SLED",	/* 570 */
 	   "LED",		/* all others */
 	   );			/* R30, R31 */
 
-#define TPACPI_LED_NUMLEDS 8
+#define TPACPI_LED_NUMLEDS 16
 static struct tpacpi_led_classdev *tpacpi_leds;
 static enum led_status_t tpacpi_led_state_cache[TPACPI_LED_NUMLEDS];
 static const char * const tpacpi_led_names[TPACPI_LED_NUMLEDS] = {
@@ -4828,15 +4828,20 @@ static const char * const tpacpi_led_names[TPACPI_LED_NUMLEDS] = {
 	"tpacpi::dock_batt",
 	"tpacpi::unknown_led",
 	"tpacpi::standby",
+	"tpacpi::dock_status1",
+	"tpacpi::dock_status2",
+	"tpacpi::unknown_led2",
+	"tpacpi::unknown_led3",
+	"tpacpi::thinkvantage",
 };
-#define TPACPI_SAFE_LEDS	0x0081U
+#define TPACPI_SAFE_LEDS	0x1081U
 
 static inline bool tpacpi_is_led_restricted(const unsigned int led)
 {
 #ifdef CONFIG_THINKPAD_ACPI_UNSAFE_LEDS
 	return false;
 #else
-	return (TPACPI_SAFE_LEDS & (1 << led)) == 0;
+	return (1U & (TPACPI_SAFE_LEDS >> led)) == 0;
 #endif
 }
 
@@ -4998,6 +5003,10 @@ static int __init tpacpi_init_led(unsigned int led)
 
 	tpacpi_leds[led].led = led;
 
+	/* LEDs with no name don't get registered */
+	if (!tpacpi_led_names[led])
+		return 0;
+
 	tpacpi_leds[led].led_classdev.brightness_set = &led_sysfs_set;
 	tpacpi_leds[led].led_classdev.blink_set = &led_sysfs_blink_set;
 	if (led_supported == TPACPI_LED_570)
@@ -5016,10 +5025,59 @@ static int __init tpacpi_init_led(unsigned int led)
 	return rc;
 }
 
+static const struct tpacpi_quirk led_useful_qtable[] __initconst = {
+	TPACPI_Q_IBM('1', 'E', 0x009f), /* A30 */
+	TPACPI_Q_IBM('1', 'N', 0x009f), /* A31 */
+	TPACPI_Q_IBM('1', 'G', 0x009f), /* A31 */
+
+	TPACPI_Q_IBM('1', 'I', 0x0097), /* T30 */
+	TPACPI_Q_IBM('1', 'R', 0x0097), /* T40, T41, T42, R50, R51 */
+	TPACPI_Q_IBM('7', '0', 0x0097), /* T43, R52 */
+	TPACPI_Q_IBM('1', 'Y', 0x0097), /* T43 */
+	TPACPI_Q_IBM('1', 'W', 0x0097), /* R50e */
+	TPACPI_Q_IBM('1', 'V', 0x0097), /* R51 */
+	TPACPI_Q_IBM('7', '8', 0x0097), /* R51e */
+	TPACPI_Q_IBM('7', '6', 0x0097), /* R52 */
+
+	TPACPI_Q_IBM('1', 'K', 0x00bf), /* X30 */
+	TPACPI_Q_IBM('1', 'Q', 0x00bf), /* X31, X32 */
+	TPACPI_Q_IBM('1', 'U', 0x00bf), /* X40 */
+	TPACPI_Q_IBM('7', '4', 0x00bf), /* X41 */
+	TPACPI_Q_IBM('7', '5', 0x00bf), /* X41t */
+
+	TPACPI_Q_IBM('7', '9', 0x1f97), /* T60 (1) */
+	TPACPI_Q_IBM('7', '7', 0x1f97), /* Z60* (1) */
+	TPACPI_Q_IBM('7', 'F', 0x1f97), /* Z61* (1) */
+	TPACPI_Q_IBM('7', 'B', 0x1fb7), /* X60 (1) */
+
+	/* (1) - may have excess leds enabled on MSB */
+
+	/* Defaults (order matters, keep last, don't reorder!) */
+	{ /* Lenovo */
+	  .vendor = PCI_VENDOR_ID_LENOVO,
+	  .bios = TPACPI_MATCH_ANY, .ec = TPACPI_MATCH_ANY,
+	  .quirks = 0x1fffU,
+	},
+	{ /* IBM ThinkPads with no EC version string */
+	  .vendor = PCI_VENDOR_ID_IBM,
+	  .bios = TPACPI_MATCH_ANY, .ec = TPACPI_MATCH_UNKNOWN,
+	  .quirks = 0x00ffU,
+	},
+	{ /* IBM ThinkPads with EC version string */
+	  .vendor = PCI_VENDOR_ID_IBM,
+	  .bios = TPACPI_MATCH_ANY, .ec = TPACPI_MATCH_ANY,
+	  .quirks = 0x00bfU,
+	},
+};
+
+#undef TPACPI_LEDQ_IBM
+#undef TPACPI_LEDQ_LNV
+
 static int __init led_init(struct ibm_init_struct *iibm)
 {
 	unsigned int i;
 	int rc;
+	unsigned long useful_leds;
 
 	vdbg_printk(TPACPI_DBG_INIT, "initializing LED subdriver\n");
 
@@ -5041,6 +5099,9 @@ static int __init led_init(struct ibm_init_struct *iibm)
 	vdbg_printk(TPACPI_DBG_INIT, "LED commands are %s, mode %d\n",
 		str_supported(led_supported), led_supported);
 
+	if (led_supported == TPACPI_LED_NONE)
+		return 1;
+
 	tpacpi_leds = kzalloc(sizeof(*tpacpi_leds) * TPACPI_LED_NUMLEDS,
 			      GFP_KERNEL);
 	if (!tpacpi_leds) {
@@ -5048,8 +5109,12 @@ static int __init led_init(struct ibm_init_struct *iibm)
 		return -ENOMEM;
 	}
 
+	useful_leds = tpacpi_check_quirks(led_useful_qtable,
+					  ARRAY_SIZE(led_useful_qtable));
+
 	for (i = 0; i < TPACPI_LED_NUMLEDS; i++) {
-		if (!tpacpi_is_led_restricted(i)) {
+		if (!tpacpi_is_led_restricted(i) &&
+		    test_bit(i, &useful_leds)) {
 			rc = tpacpi_init_led(i);
 			if (rc < 0) {
 				led_exit();
@@ -5059,12 +5124,11 @@ static int __init led_init(struct ibm_init_struct *iibm)
 	}
 
 #ifdef CONFIG_THINKPAD_ACPI_UNSAFE_LEDS
-	if (led_supported != TPACPI_LED_NONE)
-		printk(TPACPI_NOTICE
-			"warning: userspace override of important "
-			"firmware LEDs is enabled\n");
+	printk(TPACPI_NOTICE
+		"warning: userspace override of important "
+		"firmware LEDs is enabled\n");
 #endif
-	return (led_supported != TPACPI_LED_NONE)? 0 : 1;
+	return 0;
 }
 
 #define str_led_status(s) \
@@ -5094,7 +5158,7 @@ static int led_read(char *p)
 	}
 
 	len += sprintf(p + len, "commands:\t"
-		       "<led> on, <led> off, <led> blink (<led> is 0-7)\n");
+		       "<led> on, <led> off, <led> blink (<led> is 0-15)\n");
 
 	return len;
 }
@@ -5109,7 +5173,7 @@ static int led_write(char *buf)
 		return -ENODEV;
 
 	while ((cmd = next_cmd(&buf))) {
-		if (sscanf(cmd, "%d", &led) != 1 || led < 0 || led > 7)
+		if (sscanf(cmd, "%d", &led) != 1 || led < 0 || led > 15)
 			return -EINVAL;
 
 		if (strstr(cmd, "off")) {
-- 
cgit v1.2.3-59-g8ed1b


From 8bf3d4c535c2b9689c2979b281c24e9f59c2f4ad Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Sat, 30 May 2009 13:25:09 -0300
Subject: thinkpad-acpi: silence bogus warning when ACPI video is disabled

Make use of acpi_video_backlight_support() also in hotkey_init, to make
sure this doesn't happen:

thinkpad_acpi: This ThinkPad has standard ACPI backlight brightness
               control, supported by the ACPI video driver
thinkpad_acpi: Disabling thinkpad-acpi brightness events by default...
thinkpad_acpi: Standard ACPI backlight interface not available,
               thinkpad_acpi native brightness control enabled
thinkpad_acpi: detected a 16-level brightness capable ThinkPad

Note that this is purely cosmetic, there is absolutely _no_ change in
behaviour.  Those events are sometimes enabled at runtime by userspace, but
the driver never enables them by itself unless someone messed with the
default keymaps.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Reported-by: Jochen Schulz <jrschulz@well-adjusted.de>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/thinkpad_acpi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c
index 06c7c03c8f2f..5a22a064222c 100644
--- a/drivers/platform/x86/thinkpad_acpi.c
+++ b/drivers/platform/x86/thinkpad_acpi.c
@@ -2655,7 +2655,7 @@ static int __init hotkey_init(struct ibm_init_struct *iibm)
 		/* update bright_acpimode... */
 		tpacpi_check_std_acpi_brightness_support();
 
-	if (tp_features.bright_acpimode) {
+	if (tp_features.bright_acpimode && acpi_video_backlight_support()) {
 		printk(TPACPI_INFO
 		       "This ThinkPad has standard ACPI backlight "
 		       "brightness control, supported by the ACPI "
-- 
cgit v1.2.3-59-g8ed1b


From 8b12b922ed5b9b6bfc345d3d6c6de56b2982af7f Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Thu, 14 May 2009 08:31:32 -0600
Subject: ACPI: acpi_device_register() should call device_register()

There is no apparent reason for acpi_device_register() to manually
register a new device in two steps (initialize then add).

Just call device_register() directly.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/scan.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index 8ff510b91d88..b8f5c005fbb5 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -530,11 +530,10 @@ static int acpi_device_register(struct acpi_device *device,
 	if (device->parent)
 		device->dev.parent = &parent->dev;
 	device->dev.bus = &acpi_bus_type;
-	device_initialize(&device->dev);
 	device->dev.release = &acpi_device_release;
-	result = device_add(&device->dev);
+	result = device_register(&device->dev);
 	if(result) {
-		dev_err(&device->dev, "Error adding device\n");
+		dev_err(&device->dev, "Error registering device\n");
 		goto end;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 0c526d96a5bd86c70507b7d9372e6a26a1e3ea43 Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Thu, 14 May 2009 08:31:37 -0600
Subject: ACPI: clean up whitespace in drivers/acpi/scan.c

Align labels in column 0, adjust spacing in 'if' statements, eliminate
trailing and superfluous whitespaces.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/scan.c | 44 ++++++++++++++++++++------------------------
 1 file changed, 20 insertions(+), 24 deletions(-)

diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index b8f5c005fbb5..c40515e86187 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -198,12 +198,12 @@ acpi_device_path_show(struct device *dev, struct device_attribute *attr, char *b
 	int result;
 
 	result = acpi_get_name(acpi_dev->handle, ACPI_FULL_PATHNAME, &path);
-	if(result)
+	if (result)
 		goto end;
 
 	result = sprintf(buf, "%s\n", (char*)path.pointer);
 	kfree(path.pointer);
-  end:
+end:
 	return result;
 }
 static DEVICE_ATTR(path, 0444, acpi_device_path_show, NULL);
@@ -217,21 +217,21 @@ static int acpi_device_setup_files(struct acpi_device *dev)
 	/*
 	 * Devices gotten from FADT don't have a "path" attribute
 	 */
-	if(dev->handle) {
+	if (dev->handle) {
 		result = device_create_file(&dev->dev, &dev_attr_path);
-		if(result)
+		if (result)
 			goto end;
 	}
 
-	if(dev->flags.hardware_id) {
+	if (dev->flags.hardware_id) {
 		result = device_create_file(&dev->dev, &dev_attr_hid);
-		if(result)
+		if (result)
 			goto end;
 	}
 
-	if (dev->flags.hardware_id || dev->flags.compatible_ids){
+	if (dev->flags.hardware_id || dev->flags.compatible_ids) {
 		result = device_create_file(&dev->dev, &dev_attr_modalias);
-		if(result)
+		if (result)
 			goto end;
 	}
 
@@ -242,7 +242,7 @@ static int acpi_device_setup_files(struct acpi_device *dev)
 	status = acpi_get_handle(dev->handle, "_EJ0", &temp);
 	if (ACPI_SUCCESS(status))
 		result = device_create_file(&dev->dev, &dev_attr_eject);
-  end:
+end:
 	return result;
 }
 
@@ -262,9 +262,9 @@ static void acpi_device_remove_files(struct acpi_device *dev)
 	if (dev->flags.hardware_id || dev->flags.compatible_ids)
 		device_remove_file(&dev->dev, &dev_attr_modalias);
 
-	if(dev->flags.hardware_id)
+	if (dev->flags.hardware_id)
 		device_remove_file(&dev->dev, &dev_attr_hid);
-	if(dev->handle)
+	if (dev->handle)
 		device_remove_file(&dev->dev, &dev_attr_path);
 }
 /* --------------------------------------------------------------------------
@@ -512,7 +512,7 @@ static int acpi_device_register(struct acpi_device *device,
 			break;
 		}
 	}
-	if(!found) {
+	if (!found) {
 		acpi_device_bus_id = new_bus_id;
 		strcpy(acpi_device_bus_id->bus_id, device->flags.hardware_id ? device->pnp.hardware_id : "device");
 		acpi_device_bus_id->instance_no = 0;
@@ -532,19 +532,19 @@ static int acpi_device_register(struct acpi_device *device,
 	device->dev.bus = &acpi_bus_type;
 	device->dev.release = &acpi_device_release;
 	result = device_register(&device->dev);
-	if(result) {
+	if (result) {
 		dev_err(&device->dev, "Error registering device\n");
 		goto end;
 	}
 
 	result = acpi_device_setup_files(device);
-	if(result)
+	if (result)
 		printk(KERN_ERR PREFIX "Error creating sysfs interface for device %s\n",
 		       dev_name(&device->dev));
 
 	device->removal_type = ACPI_BUS_REMOVAL_NORMAL;
 	return 0;
-  end:
+end:
 	mutex_lock(&acpi_device_lock);
 	if (device->parent)
 		list_del(&device->node);
@@ -576,7 +576,7 @@ static void acpi_device_unregister(struct acpi_device *device, int type)
  * @device: the device to add and initialize
  * @driver: driver for the device
  *
- * Used to initialize a device via its device driver.  Called whenever a 
+ * Used to initialize a device via its device driver.  Called whenever a
  * driver is bound to a device.  Invokes the driver's add() ops.
  */
 static int
@@ -584,7 +584,6 @@ acpi_bus_driver_init(struct acpi_device *device, struct acpi_driver *driver)
 {
 	int result = 0;
 
-
 	if (!device || !driver)
 		return -EINVAL;
 
@@ -801,7 +800,7 @@ static int acpi_bus_get_wakeup_device_flags(struct acpi_device *device)
 	if (!acpi_match_device_ids(device, button_device_ids))
 		device->wakeup.flags.run_wake = 1;
 
-      end:
+end:
 	if (ACPI_FAILURE(status))
 		device->flags.wake_capable = 0;
 	return 0;
@@ -1069,7 +1068,7 @@ static void acpi_device_set_id(struct acpi_device *device,
 		break;
 	}
 
-	/* 
+	/*
 	 * \_SB
 	 * ----
 	 * Fix for the system root bus device -- the only root-level device.
@@ -1319,7 +1318,7 @@ acpi_add_single_object(struct acpi_device **child,
 			device->parent->ops.bind(device);
 	}
 
-      end:
+end:
 	if (!result)
 		*child = device;
 	else {
@@ -1463,7 +1462,6 @@ acpi_bus_add(struct acpi_device **child,
 
 	return result;
 }
-
 EXPORT_SYMBOL(acpi_bus_add);
 
 int acpi_bus_start(struct acpi_device *device)
@@ -1483,7 +1481,6 @@ int acpi_bus_start(struct acpi_device *device)
 	}
 	return result;
 }
-
 EXPORT_SYMBOL(acpi_bus_start);
 
 int acpi_bus_trim(struct acpi_device *start, int rmdevice)
@@ -1541,7 +1538,6 @@ int acpi_bus_trim(struct acpi_device *start, int rmdevice)
 }
 EXPORT_SYMBOL_GPL(acpi_bus_trim);
 
-
 static int acpi_bus_scan_fixed(struct acpi_device *root)
 {
 	int result = 0;
@@ -1609,6 +1605,6 @@ int __init acpi_scan_init(void)
 	if (result)
 		acpi_device_unregister(acpi_root, ACPI_BUS_REMOVAL_NORMAL);
 
-      Done:
+Done:
 	return result;
 }
-- 
cgit v1.2.3-59-g8ed1b


From ce597bb42aa84bc73db80509b7c37e7fbc0b75c4 Mon Sep 17 00:00:00 2001
From: Alexander Chiang <achiang@hp.com>
Date: Wed, 10 Jun 2009 19:55:09 +0000
Subject: ACPI: make acpi_pci_bind() static

acpi_pci_root_add() explicitly assigns device->ops.bind, and later
calls acpi_pci_bind_root(), which also does the same thing.

We don't need to repeat ourselves; removing the explicit assignment
allows us to make acpi_pci_bind() static.

Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_bind.c     | 3 ++-
 drivers/acpi/pci_root.c     | 2 --
 include/acpi/acpi_drivers.h | 1 -
 3 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/acpi/pci_bind.c b/drivers/acpi/pci_bind.c
index bc46de3d967f..236765c6017b 100644
--- a/drivers/acpi/pci_bind.c
+++ b/drivers/acpi/pci_bind.c
@@ -44,6 +44,7 @@ struct acpi_pci_data {
 	struct pci_dev *dev;
 };
 
+static int acpi_pci_bind(struct acpi_device *device);
 static int acpi_pci_unbind(struct acpi_device *device);
 
 static void acpi_pci_data_handler(acpi_handle handle, u32 function,
@@ -108,7 +109,7 @@ acpi_status acpi_get_pci_id(acpi_handle handle, struct acpi_pci_id *id)
 
 EXPORT_SYMBOL(acpi_get_pci_id);
 
-int acpi_pci_bind(struct acpi_device *device)
+static int acpi_pci_bind(struct acpi_device *device)
 {
 	int result = 0;
 	acpi_status status;
diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 196f97d00956..ca8dba3b40b9 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -386,8 +386,6 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	strcpy(acpi_device_class(device), ACPI_PCI_ROOT_CLASS);
 	device->driver_data = root;
 
-	device->ops.bind = acpi_pci_bind;
-
 	/*
 	 * All supported architectures that use ACPI have support for
 	 * PCI domains, so we indicate this in _OSC support capabilities.
diff --git a/include/acpi/acpi_drivers.h b/include/acpi/acpi_drivers.h
index 0352c8f0b05b..7a2ce53146aa 100644
--- a/include/acpi/acpi_drivers.h
+++ b/include/acpi/acpi_drivers.h
@@ -99,7 +99,6 @@ void acpi_pci_irq_del_prt(int segment, int bus);
 struct pci_bus;
 
 acpi_status acpi_get_pci_id(acpi_handle handle, struct acpi_pci_id *id);
-int acpi_pci_bind(struct acpi_device *device);
 int acpi_pci_bind_root(struct acpi_device *device, struct acpi_pci_id *id,
 		       struct pci_bus *bus);
 
-- 
cgit v1.2.3-59-g8ed1b


From 275582031f9b3597a1b973f3ff617adfe639faa2 Mon Sep 17 00:00:00 2001
From: Alexander Chiang <achiang@hp.com>
Date: Wed, 10 Jun 2009 19:55:14 +0000
Subject: ACPI: Introduce acpi_is_root_bridge()

Returns whether an ACPI CA node is a PCI root bridge or not.

This API is generically useful, and shouldn't just be a hotplug function.

The implementation becomes much simpler as well.

Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_root.c            | 24 +++++++++++++++++++++++
 drivers/pci/hotplug/acpi_pcihp.c   | 40 ++------------------------------------
 drivers/pci/hotplug/acpiphp_glue.c |  2 +-
 include/acpi/acpi_bus.h            |  1 +
 include/linux/pci_hotplug.h        |  1 -
 5 files changed, 28 insertions(+), 40 deletions(-)

diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index ca8dba3b40b9..888cb9f5c5fb 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -142,6 +142,30 @@ acpi_handle acpi_get_pci_rootbridge_handle(unsigned int seg, unsigned int bus)
 
 EXPORT_SYMBOL_GPL(acpi_get_pci_rootbridge_handle);
 
+/**
+ * acpi_is_root_bridge - determine whether an ACPI CA node is a PCI root bridge
+ * @handle - the ACPI CA node in question.
+ *
+ * Note: we could make this API take a struct acpi_device * instead, but
+ * for now, it's more convenient to operate on an acpi_handle.
+ */
+int acpi_is_root_bridge(acpi_handle handle)
+{
+	int ret;
+	struct acpi_device *device;
+
+	ret = acpi_bus_get_device(handle, &device);
+	if (ret)
+		return 0;
+
+	ret = acpi_match_device_ids(device, root_device_ids);
+	if (ret)
+		return 0;
+	else
+		return 1;
+}
+EXPORT_SYMBOL_GPL(acpi_is_root_bridge);
+
 static acpi_status
 get_root_bridge_busnr_callback(struct acpi_resource *resource, void *data)
 {
diff --git a/drivers/pci/hotplug/acpi_pcihp.c b/drivers/pci/hotplug/acpi_pcihp.c
index fbc63d5e459f..eb159587d0bf 100644
--- a/drivers/pci/hotplug/acpi_pcihp.c
+++ b/drivers/pci/hotplug/acpi_pcihp.c
@@ -354,7 +354,7 @@ acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus,
 		status = acpi_run_hpp(handle, hpp);
 		if (ACPI_SUCCESS(status))
 			break;
-		if (acpi_root_bridge(handle))
+		if (acpi_is_root_bridge(handle))
 			break;
 		status = acpi_get_parent(handle, &phandle);
 		if (ACPI_FAILURE(status))
@@ -428,7 +428,7 @@ int acpi_get_hp_hw_control_from_firmware(struct pci_dev *pdev, u32 flags)
 		status = acpi_run_oshp(handle);
 		if (ACPI_SUCCESS(status))
 			goto got_one;
-		if (acpi_root_bridge(handle))
+		if (acpi_is_root_bridge(handle))
 			break;
 		chandle = handle;
 		status = acpi_get_parent(chandle, &handle);
@@ -449,42 +449,6 @@ got_one:
 }
 EXPORT_SYMBOL(acpi_get_hp_hw_control_from_firmware);
 
-/* acpi_root_bridge - check to see if this acpi object is a root bridge
- *
- * @handle - the acpi object in question.
- */
-int acpi_root_bridge(acpi_handle handle)
-{
-	acpi_status status;
-	struct acpi_device_info *info;
-	struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL};
-	int i;
-
-	status = acpi_get_object_info(handle, &buffer);
-	if (ACPI_SUCCESS(status)) {
-		info = buffer.pointer;
-		if ((info->valid & ACPI_VALID_HID) &&
-			!strcmp(PCI_ROOT_HID_STRING,
-					info->hardware_id.value)) {
-			kfree(buffer.pointer);
-			return 1;
-		}
-		if (info->valid & ACPI_VALID_CID) {
-			for (i=0; i < info->compatibility_id.count; i++) {
-				if (!strcmp(PCI_ROOT_HID_STRING,
-					info->compatibility_id.id[i].value)) {
-					kfree(buffer.pointer);
-					return 1;
-				}
-			}
-		}
-		kfree(buffer.pointer);
-	}
-	return 0;
-}
-EXPORT_SYMBOL_GPL(acpi_root_bridge);
-
-
 static int is_ejectable(acpi_handle handle)
 {
 	acpi_status status;
diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index 3a6064bce561..fc6636e3300b 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -1631,7 +1631,7 @@ find_root_bridges(acpi_handle handle, u32 lvl, void *context, void **rv)
 {
 	int *count = (int *)context;
 
-	if (acpi_root_bridge(handle)) {
+	if (acpi_is_root_bridge(handle)) {
 		acpi_install_notify_handler(handle, ACPI_SYSTEM_NOTIFY,
 				handle_hotplug_event_bridge, NULL);
 			(*count)++;
diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index c34b11022908..96d593ee4859 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -369,6 +369,7 @@ struct device *acpi_get_physical_pci_device(acpi_handle);
 
 /* helper */
 acpi_handle acpi_get_child(acpi_handle, acpi_integer);
+int acpi_is_root_bridge(acpi_handle);
 acpi_handle acpi_get_pci_rootbridge_handle(unsigned int, unsigned int);
 #define DEVICE_ACPI_HANDLE(dev) ((acpi_handle)((dev)->archdata.acpi_handle))
 
diff --git a/include/linux/pci_hotplug.h b/include/linux/pci_hotplug.h
index 20998746518e..a3576ef9fc74 100644
--- a/include/linux/pci_hotplug.h
+++ b/include/linux/pci_hotplug.h
@@ -226,7 +226,6 @@ struct hotplug_params {
 extern acpi_status acpi_get_hp_params_from_firmware(struct pci_bus *bus,
 				struct hotplug_params *hpp);
 int acpi_get_hp_hw_control_from_firmware(struct pci_dev *dev, u32 flags);
-int acpi_root_bridge(acpi_handle handle);
 int acpi_pci_check_ejectable(struct pci_bus *pbus, acpi_handle handle);
 int acpi_pci_detect_ejectable(struct pci_bus *pbus);
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 2f7bbceb5b6aa938024bb4dad93c410fa59ed3b9 Mon Sep 17 00:00:00 2001
From: Alexander Chiang <achiang@hp.com>
Date: Wed, 10 Jun 2009 19:55:20 +0000
Subject: ACPI: Introduce acpi_get_pci_dev()

Convert an ACPI CA handle to a struct pci_dev.

Performing this lookup dynamically allows us to get rid of the
ACPI-PCI binding code, which:

	- eliminates struct acpi_device vs struct pci_dev lifetime issues
	- lays more groundwork for eliminating .start from acpi_device_ops
	  and thus simplifying ACPI drivers
	- whacks out a lot of code

This change lays the groundwork for eliminating much of pci_bind.c.

Although pci_root.c may not be the most logical place for this
change, putting it here saves us from having to export acpi_pci_find_root.

Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_root.c     | 81 +++++++++++++++++++++++++++++++++++++++++++++
 include/acpi/acpi_drivers.h |  1 +
 2 files changed, 82 insertions(+)

diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 888cb9f5c5fb..e5099919e574 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -329,6 +329,87 @@ static struct acpi_pci_root *acpi_pci_find_root(acpi_handle handle)
 	return NULL;
 }
 
+struct acpi_handle_node {
+	struct list_head node;
+	acpi_handle handle;
+};
+
+/**
+ * acpi_get_pci_dev - convert ACPI CA handle to struct pci_dev
+ * @handle: the handle in question
+ *
+ * Given an ACPI CA handle, the desired PCI device is located in the
+ * list of PCI devices.
+ *
+ * If the device is found, its reference count is increased and this
+ * function returns a pointer to its data structure.  The caller must
+ * decrement the reference count by calling pci_dev_put().
+ * If no device is found, %NULL is returned.
+ */
+struct pci_dev *acpi_get_pci_dev(acpi_handle handle)
+{
+	int dev, fn;
+	unsigned long long adr;
+	acpi_status status;
+	acpi_handle phandle;
+	struct pci_bus *pbus;
+	struct pci_dev *pdev = NULL;
+	struct acpi_handle_node *node, *tmp;
+	struct acpi_pci_root *root;
+	LIST_HEAD(device_list);
+
+	/*
+	 * Walk up the ACPI CA namespace until we reach a PCI root bridge.
+	 */
+	phandle = handle;
+	while (!acpi_is_root_bridge(phandle)) {
+		node = kzalloc(sizeof(struct acpi_handle_node), GFP_KERNEL);
+		if (!node)
+			goto out;
+
+		INIT_LIST_HEAD(&node->node);
+		node->handle = phandle;
+		list_add(&node->node, &device_list);
+
+		status = acpi_get_parent(phandle, &phandle);
+		if (ACPI_FAILURE(status))
+			goto out;
+	}
+
+	root = acpi_pci_find_root(phandle);
+	if (!root)
+		goto out;
+
+	pbus = root->bus;
+
+	/*
+	 * Now, walk back down the PCI device tree until we return to our
+	 * original handle. Assumes that everything between the PCI root
+	 * bridge and the device we're looking for must be a P2P bridge.
+	 */
+	list_for_each_entry(node, &device_list, node) {
+		acpi_handle hnd = node->handle;
+		status = acpi_evaluate_integer(hnd, "_ADR", NULL, &adr);
+		if (ACPI_FAILURE(status))
+			goto out;
+		dev = (adr >> 16) & 0xffff;
+		fn  = adr & 0xffff;
+
+		pdev = pci_get_slot(pbus, PCI_DEVFN(dev, fn));
+		if (hnd == handle)
+			break;
+
+		pbus = pdev->subordinate;
+		pci_dev_put(pdev);
+	}
+out:
+	list_for_each_entry_safe(node, tmp, &device_list, node)
+		kfree(node);
+
+	return pdev;
+}
+EXPORT_SYMBOL_GPL(acpi_get_pci_dev);
+
 /**
  * acpi_pci_osc_control_set - commit requested control to Firmware
  * @handle: acpi_handle for the target ACPI object
diff --git a/include/acpi/acpi_drivers.h b/include/acpi/acpi_drivers.h
index 7a2ce53146aa..dbe3989952ee 100644
--- a/include/acpi/acpi_drivers.h
+++ b/include/acpi/acpi_drivers.h
@@ -98,6 +98,7 @@ void acpi_pci_irq_del_prt(int segment, int bus);
 
 struct pci_bus;
 
+struct pci_dev *acpi_get_pci_dev(acpi_handle);
 acpi_status acpi_get_pci_id(acpi_handle handle, struct acpi_pci_id *id);
 int acpi_pci_bind_root(struct acpi_device *device, struct acpi_pci_id *id,
 		       struct pci_bus *bus);
-- 
cgit v1.2.3-59-g8ed1b


From c22d7f5a389dad15de448b142f44e4000b3426f0 Mon Sep 17 00:00:00 2001
From: Alexander Chiang <achiang@hp.com>
Date: Wed, 10 Jun 2009 19:55:25 +0000
Subject: ACPI: rearrange acpi_pci_bind/acpi_pci_unbind in pci_bind.c

This is a pure code movement patch that does $subject in order
to make the following patch easier to read and review.

No functional change.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_bind.c | 90 ++++++++++++++++++++++++-------------------------
 1 file changed, 45 insertions(+), 45 deletions(-)

diff --git a/drivers/acpi/pci_bind.c b/drivers/acpi/pci_bind.c
index 236765c6017b..c283c29662a7 100644
--- a/drivers/acpi/pci_bind.c
+++ b/drivers/acpi/pci_bind.c
@@ -109,6 +109,51 @@ acpi_status acpi_get_pci_id(acpi_handle handle, struct acpi_pci_id *id)
 
 EXPORT_SYMBOL(acpi_get_pci_id);
 
+static int acpi_pci_unbind(struct acpi_device *device)
+{
+	int result = 0;
+	acpi_status status;
+	struct acpi_pci_data *data;
+	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
+
+
+	if (!device || !device->parent)
+		return -EINVAL;
+
+	status = acpi_get_name(device->handle, ACPI_FULL_PATHNAME, &buffer);
+	if (ACPI_FAILURE(status))
+		return -ENODEV;
+
+	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Unbinding PCI device [%s]...\n",
+			  (char *) buffer.pointer));
+	kfree(buffer.pointer);
+
+	status =
+	    acpi_get_data(device->handle, acpi_pci_data_handler,
+			  (void **)&data);
+	if (ACPI_FAILURE(status)) {
+		result = -ENODEV;
+		goto end;
+	}
+
+	status = acpi_detach_data(device->handle, acpi_pci_data_handler);
+	if (ACPI_FAILURE(status)) {
+		ACPI_EXCEPTION((AE_INFO, status,
+				"Unable to detach data from device %s",
+				acpi_device_bid(device)));
+		result = -ENODEV;
+		goto end;
+	}
+	if (data->dev->subordinate) {
+		acpi_pci_irq_del_prt(data->id.segment, data->bus->number);
+	}
+	pci_dev_put(data->dev);
+	kfree(data);
+
+      end:
+	return result;
+}
+
 static int acpi_pci_bind(struct acpi_device *device)
 {
 	int result = 0;
@@ -253,51 +298,6 @@ static int acpi_pci_bind(struct acpi_device *device)
 	return result;
 }
 
-static int acpi_pci_unbind(struct acpi_device *device)
-{
-	int result = 0;
-	acpi_status status;
-	struct acpi_pci_data *data;
-	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
-
-
-	if (!device || !device->parent)
-		return -EINVAL;
-
-	status = acpi_get_name(device->handle, ACPI_FULL_PATHNAME, &buffer);
-	if (ACPI_FAILURE(status))
-		return -ENODEV;
-
-	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Unbinding PCI device [%s]...\n",
-			  (char *) buffer.pointer));
-	kfree(buffer.pointer);
-
-	status =
-	    acpi_get_data(device->handle, acpi_pci_data_handler,
-			  (void **)&data);
-	if (ACPI_FAILURE(status)) {
-		result = -ENODEV;
-		goto end;
-	}
-
-	status = acpi_detach_data(device->handle, acpi_pci_data_handler);
-	if (ACPI_FAILURE(status)) {
-		ACPI_EXCEPTION((AE_INFO, status,
-				"Unable to detach data from device %s",
-				acpi_device_bid(device)));
-		result = -ENODEV;
-		goto end;
-	}
-	if (data->dev->subordinate) {
-		acpi_pci_irq_del_prt(data->id.segment, data->bus->number);
-	}
-	pci_dev_put(data->dev);
-	kfree(data);
-
-      end:
-	return result;
-}
-
 int
 acpi_pci_bind_root(struct acpi_device *device,
 		   struct acpi_pci_id *id, struct pci_bus *bus)
-- 
cgit v1.2.3-59-g8ed1b


From 499650de6906722184b639989b47227a362b62f8 Mon Sep 17 00:00:00 2001
From: Alexander Chiang <achiang@hp.com>
Date: Wed, 10 Jun 2009 19:55:30 +0000
Subject: ACPI: eviscerate pci_bind.c

Now that we can dynamically convert an ACPI CA handle to a
struct pci_dev at runtime, there's no need to statically bind
them during boot.

acpi_pci_bind/unbind are vastly simplified, and are only used
to evaluate _PRT methods on P2P bridges and non-bridge children.

This patch also changes the time-space tradeoff ever so slightly.

Looking up the ACPI-PCI binding is never in the performance path, and by
eliminating this caching, we save 24 bytes for each _ADR device in the
ACPI namespace.

This patch lays further groundwork to eventually eliminate
the acpi_driver_ops.bind callback.

Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_bind.c     | 245 +++++++-------------------------------------
 drivers/acpi/pci_root.c     |   2 +-
 include/acpi/acpi_drivers.h |   3 +-
 3 files changed, 39 insertions(+), 211 deletions(-)

diff --git a/drivers/acpi/pci_bind.c b/drivers/acpi/pci_bind.c
index c283c29662a7..703d2a3e8012 100644
--- a/drivers/acpi/pci_bind.c
+++ b/drivers/acpi/pci_bind.c
@@ -24,12 +24,7 @@
  */
 
 #include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/init.h>
 #include <linux/types.h>
-#include <linux/proc_fs.h>
-#include <linux/spinlock.h>
-#include <linux/pm.h>
 #include <linux/pci.h>
 #include <linux/acpi.h>
 #include <acpi/acpi_bus.h>
@@ -111,238 +106,72 @@ EXPORT_SYMBOL(acpi_get_pci_id);
 
 static int acpi_pci_unbind(struct acpi_device *device)
 {
-	int result = 0;
-	acpi_status status;
-	struct acpi_pci_data *data;
-	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
-
-
-	if (!device || !device->parent)
-		return -EINVAL;
-
-	status = acpi_get_name(device->handle, ACPI_FULL_PATHNAME, &buffer);
-	if (ACPI_FAILURE(status))
-		return -ENODEV;
-
-	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Unbinding PCI device [%s]...\n",
-			  (char *) buffer.pointer));
-	kfree(buffer.pointer);
+	struct pci_dev *dev;
 
-	status =
-	    acpi_get_data(device->handle, acpi_pci_data_handler,
-			  (void **)&data);
-	if (ACPI_FAILURE(status)) {
-		result = -ENODEV;
-		goto end;
-	}
+	dev = acpi_get_pci_dev(device->handle);
+	if (!dev)
+		return 0;
 
-	status = acpi_detach_data(device->handle, acpi_pci_data_handler);
-	if (ACPI_FAILURE(status)) {
-		ACPI_EXCEPTION((AE_INFO, status,
-				"Unable to detach data from device %s",
-				acpi_device_bid(device)));
-		result = -ENODEV;
-		goto end;
-	}
-	if (data->dev->subordinate) {
-		acpi_pci_irq_del_prt(data->id.segment, data->bus->number);
-	}
-	pci_dev_put(data->dev);
-	kfree(data);
+	if (dev->subordinate)
+		acpi_pci_irq_del_prt(pci_domain_nr(dev->bus),
+				     dev->subordinate->number);
 
-      end:
-	return result;
+	pci_dev_put(dev);
+	return 0;
 }
 
 static int acpi_pci_bind(struct acpi_device *device)
 {
-	int result = 0;
 	acpi_status status;
-	struct acpi_pci_data *data;
-	struct acpi_pci_data *pdata;
-	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
 	acpi_handle handle;
+	unsigned char bus;
+	struct pci_dev *dev;
 
-	if (!device || !device->parent)
-		return -EINVAL;
-
-	data = kzalloc(sizeof(struct acpi_pci_data), GFP_KERNEL);
-	if (!data)
-		return -ENOMEM;
-
-	status = acpi_get_name(device->handle, ACPI_FULL_PATHNAME, &buffer);
-	if (ACPI_FAILURE(status)) {
-		kfree(data);
-		return -ENODEV;
-	}
-
-	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Binding PCI device [%s]...\n",
-			  (char *)buffer.pointer));
-
-	/* 
-	 * Segment & Bus
-	 * -------------
-	 * These are obtained via the parent device's ACPI-PCI context.
-	 */
-	status = acpi_get_data(device->parent->handle, acpi_pci_data_handler,
-			       (void **)&pdata);
-	if (ACPI_FAILURE(status) || !pdata || !pdata->bus) {
-		ACPI_EXCEPTION((AE_INFO, status,
-				"Invalid ACPI-PCI context for parent device %s",
-				acpi_device_bid(device->parent)));
-		result = -ENODEV;
-		goto end;
-	}
-	data->id.segment = pdata->id.segment;
-	data->id.bus = pdata->bus->number;
-
-	/*
-	 * Device & Function
-	 * -----------------
-	 * These are simply obtained from the device's _ADR method.  Note
-	 * that a value of zero is valid.
-	 */
-	data->id.device = device->pnp.bus_address >> 16;
-	data->id.function = device->pnp.bus_address & 0xFFFF;
-
-	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "...to %04x:%02x:%02x.%d\n",
-			  data->id.segment, data->id.bus, data->id.device,
-			  data->id.function));
-
-	/*
-	 * TBD: Support slot devices (e.g. function=0xFFFF).
-	 */
-
-	/* 
-	 * Locate PCI Device
-	 * -----------------
-	 * Locate matching device in PCI namespace.  If it doesn't exist
-	 * this typically means that the device isn't currently inserted
-	 * (e.g. docking station, port replicator, etc.).
-	 */
-	data->dev = pci_get_slot(pdata->bus,
-				PCI_DEVFN(data->id.device, data->id.function));
-	if (!data->dev) {
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-				  "Device %04x:%02x:%02x.%d not present in PCI namespace\n",
-				  data->id.segment, data->id.bus,
-				  data->id.device, data->id.function));
-		result = -ENODEV;
-		goto end;
-	}
-	if (!data->dev->bus) {
-		printk(KERN_ERR PREFIX
-			    "Device %04x:%02x:%02x.%d has invalid 'bus' field\n",
-			    data->id.segment, data->id.bus,
-			    data->id.device, data->id.function);
-		result = -ENODEV;
-		goto end;
-	}
+	dev = acpi_get_pci_dev(device->handle);
+	if (!dev)
+		return 0;
 
 	/*
-	 * PCI Bridge?
-	 * -----------
-	 * If so, set the 'bus' field and install the 'bind' function to 
-	 * facilitate callbacks for all of its children.
+	 * Install the 'bind' function to facilitate callbacks for
+	 * children of the P2P bridge.
 	 */
-	if (data->dev->subordinate) {
+	if (dev->subordinate) {
 		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
 				  "Device %04x:%02x:%02x.%d is a PCI bridge\n",
-				  data->id.segment, data->id.bus,
-				  data->id.device, data->id.function));
-		data->bus = data->dev->subordinate;
+				  pci_domain_nr(dev->bus), dev->bus->number,
+				  PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn)));
 		device->ops.bind = acpi_pci_bind;
 		device->ops.unbind = acpi_pci_unbind;
 	}
 
 	/*
-	 * Attach ACPI-PCI Context
-	 * -----------------------
-	 * Thus binding the ACPI and PCI devices.
-	 */
-	status = acpi_attach_data(device->handle, acpi_pci_data_handler, data);
-	if (ACPI_FAILURE(status)) {
-		ACPI_EXCEPTION((AE_INFO, status,
-				"Unable to attach ACPI-PCI context to device %s",
-				acpi_device_bid(device)));
-		result = -ENODEV;
-		goto end;
-	}
-
-	/*
-	 * PCI Routing Table
-	 * -----------------
-	 * Evaluate and parse _PRT, if exists.  This code is independent of 
-	 * PCI bridges (above) to allow parsing of _PRT objects within the
-	 * scope of non-bridge devices.  Note that _PRTs within the scope of
-	 * a PCI bridge assume the bridge's subordinate bus number.
+	 * Evaluate and parse _PRT, if exists.  This code allows parsing of
+	 * _PRT objects within the scope of non-bridge devices.  Note that
+	 * _PRTs within the scope of a PCI bridge assume the bridge's
+	 * subordinate bus number.
 	 *
 	 * TBD: Can _PRTs exist within the scope of non-bridge PCI devices?
 	 */
 	status = acpi_get_handle(device->handle, METHOD_NAME__PRT, &handle);
-	if (ACPI_SUCCESS(status)) {
-		if (data->bus)	/* PCI-PCI bridge */
-			acpi_pci_irq_add_prt(device->handle, data->id.segment,
-					     data->bus->number);
-		else		/* non-bridge PCI device */
-			acpi_pci_irq_add_prt(device->handle, data->id.segment,
-					     data->id.bus);
-	}
-
-      end:
-	kfree(buffer.pointer);
-	if (result) {
-		pci_dev_put(data->dev);
-		kfree(data);
-	}
-	return result;
-}
+	if (ACPI_FAILURE(status))
+		goto out;
 
-int
-acpi_pci_bind_root(struct acpi_device *device,
-		   struct acpi_pci_id *id, struct pci_bus *bus)
-{
-	int result = 0;
-	acpi_status status;
-	struct acpi_pci_data *data = NULL;
-	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
+	if (dev->subordinate)
+		bus = dev->subordinate->number;
+	else
+		bus = dev->bus->number;
 
-	if (!device || !id || !bus) {
-		return -EINVAL;
-	}
+	acpi_pci_irq_add_prt(device->handle, pci_domain_nr(dev->bus), bus);
 
-	data = kzalloc(sizeof(struct acpi_pci_data), GFP_KERNEL);
-	if (!data)
-		return -ENOMEM;
+out:
+	pci_dev_put(dev);
+	return 0;
+}
 
-	data->id = *id;
-	data->bus = bus;
+int acpi_pci_bind_root(struct acpi_device *device)
+{
 	device->ops.bind = acpi_pci_bind;
 	device->ops.unbind = acpi_pci_unbind;
 
-	status = acpi_get_name(device->handle, ACPI_FULL_PATHNAME, &buffer);
-	if (ACPI_FAILURE(status)) {
-		kfree (data);
-		return -ENODEV;
-	}
-
-	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Binding PCI root bridge [%s] to "
-			"%04x:%02x\n", (char *)buffer.pointer,
-			id->segment, id->bus));
-
-	status = acpi_attach_data(device->handle, acpi_pci_data_handler, data);
-	if (ACPI_FAILURE(status)) {
-		ACPI_EXCEPTION((AE_INFO, status,
-				"Unable to attach ACPI-PCI context to device %s",
-				(char *)buffer.pointer));
-		result = -ENODEV;
-		goto end;
-	}
-
-      end:
-	kfree(buffer.pointer);
-	if (result != 0)
-		kfree(data);
-
-	return result;
+	return 0;
 }
diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index e5099919e574..f23fcc5c9674 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -603,7 +603,7 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	 * -----------------------
 	 * Thus binding the ACPI and PCI devices.
 	 */
-	result = acpi_pci_bind_root(device, &root->id, root->bus);
+	result = acpi_pci_bind_root(device);
 	if (result)
 		goto end;
 
diff --git a/include/acpi/acpi_drivers.h b/include/acpi/acpi_drivers.h
index dbe3989952ee..2740a2894837 100644
--- a/include/acpi/acpi_drivers.h
+++ b/include/acpi/acpi_drivers.h
@@ -100,8 +100,7 @@ struct pci_bus;
 
 struct pci_dev *acpi_get_pci_dev(acpi_handle);
 acpi_status acpi_get_pci_id(acpi_handle handle, struct acpi_pci_id *id);
-int acpi_pci_bind_root(struct acpi_device *device, struct acpi_pci_id *id,
-		       struct pci_bus *bus);
+int acpi_pci_bind_root(struct acpi_device *device);
 
 /* Arch-defined function to add a bus to the system */
 
-- 
cgit v1.2.3-59-g8ed1b


From 859a3f86ca83346f4097e956d0b27d96aa7a1cff Mon Sep 17 00:00:00 2001
From: Alexander Chiang <achiang@hp.com>
Date: Wed, 10 Jun 2009 19:55:35 +0000
Subject: ACPI: simplify acpi_pci_irq_add_prt() API

A PCI domain cannot change as you descend down subordinate buses, which
makes the 'segment' argument to acpi_pci_irq_add_prt() useless.

Change the interface to take a struct pci_bus *, from whence we can derive
the bus number and segment. Reducing the number of arguments makes life
simpler for callers.

Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_bind.c     |  8 ++++----
 drivers/acpi/pci_irq.c      | 10 +++++-----
 drivers/acpi/pci_root.c     |  3 +--
 include/acpi/acpi_drivers.h |  2 +-
 4 files changed, 11 insertions(+), 12 deletions(-)

diff --git a/drivers/acpi/pci_bind.c b/drivers/acpi/pci_bind.c
index 703d2a3e8012..6eb58ef366ef 100644
--- a/drivers/acpi/pci_bind.c
+++ b/drivers/acpi/pci_bind.c
@@ -124,7 +124,7 @@ static int acpi_pci_bind(struct acpi_device *device)
 {
 	acpi_status status;
 	acpi_handle handle;
-	unsigned char bus;
+	struct pci_bus *bus;
 	struct pci_dev *dev;
 
 	dev = acpi_get_pci_dev(device->handle);
@@ -157,11 +157,11 @@ static int acpi_pci_bind(struct acpi_device *device)
 		goto out;
 
 	if (dev->subordinate)
-		bus = dev->subordinate->number;
+		bus = dev->subordinate;
 	else
-		bus = dev->bus->number;
+		bus = dev->bus;
 
-	acpi_pci_irq_add_prt(device->handle, pci_domain_nr(dev->bus), bus);
+	acpi_pci_irq_add_prt(device->handle, bus);
 
 out:
 	pci_dev_put(dev);
diff --git a/drivers/acpi/pci_irq.c b/drivers/acpi/pci_irq.c
index 51b9f8280f88..3ed944cefb36 100644
--- a/drivers/acpi/pci_irq.c
+++ b/drivers/acpi/pci_irq.c
@@ -182,7 +182,7 @@ static void do_prt_fixups(struct acpi_prt_entry *entry,
 	}
 }
 
-static int acpi_pci_irq_add_entry(acpi_handle handle, int segment, int bus,
+static int acpi_pci_irq_add_entry(acpi_handle handle, struct pci_bus *bus,
 				  struct acpi_pci_routing_table *prt)
 {
 	struct acpi_prt_entry *entry;
@@ -196,8 +196,8 @@ static int acpi_pci_irq_add_entry(acpi_handle handle, int segment, int bus,
 	 * 1=INTA, 2=INTB.  We use the PCI encoding throughout, so convert
 	 * it here.
 	 */
-	entry->id.segment = segment;
-	entry->id.bus = bus;
+	entry->id.segment = pci_domain_nr(bus);
+	entry->id.bus = bus->number;
 	entry->id.device = (prt->address >> 16) & 0xFFFF;
 	entry->pin = prt->pin + 1;
 
@@ -242,7 +242,7 @@ static int acpi_pci_irq_add_entry(acpi_handle handle, int segment, int bus,
 	return 0;
 }
 
-int acpi_pci_irq_add_prt(acpi_handle handle, int segment, int bus)
+int acpi_pci_irq_add_prt(acpi_handle handle, struct pci_bus *bus)
 {
 	acpi_status status;
 	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
@@ -271,7 +271,7 @@ int acpi_pci_irq_add_prt(acpi_handle handle, int segment, int bus)
 
 	entry = buffer.pointer;
 	while (entry && (entry->length > 0)) {
-		acpi_pci_irq_add_entry(handle, segment, bus, entry);
+		acpi_pci_irq_add_entry(handle, bus, entry);
 		entry = (struct acpi_pci_routing_table *)
 		    ((unsigned long)entry + entry->length);
 	}
diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index f23fcc5c9674..f341b0756c9e 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -614,8 +614,7 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	 */
 	status = acpi_get_handle(device->handle, METHOD_NAME__PRT, &handle);
 	if (ACPI_SUCCESS(status))
-		result = acpi_pci_irq_add_prt(device->handle, root->id.segment,
-					      root->id.bus);
+		result = acpi_pci_irq_add_prt(device->handle, root->bus);
 
 	/*
 	 * Scan and bind all _ADR-Based Devices
diff --git a/include/acpi/acpi_drivers.h b/include/acpi/acpi_drivers.h
index 2740a2894837..686a960bde39 100644
--- a/include/acpi/acpi_drivers.h
+++ b/include/acpi/acpi_drivers.h
@@ -91,7 +91,7 @@ int acpi_pci_link_free_irq(acpi_handle handle);
 
 /* ACPI PCI Interrupt Routing (pci_irq.c) */
 
-int acpi_pci_irq_add_prt(acpi_handle handle, int segment, int bus);
+int acpi_pci_irq_add_prt(acpi_handle handle, struct pci_bus *bus);
 void acpi_pci_irq_del_prt(int segment, int bus);
 
 /* ACPI PCI Device Binding (pci_bind.c) */
-- 
cgit v1.2.3-59-g8ed1b


From d9efae3688addb15994c9ad9761dada6f988bc14 Mon Sep 17 00:00:00 2001
From: Alexander Chiang <achiang@hp.com>
Date: Wed, 10 Jun 2009 19:55:40 +0000
Subject: ACPI: simplify acpi_pci_irq_del_prt() API

There is no need to pass a segment/bus tuple to this API, as the callsite
always has a struct pci_bus. We can derive segment/bus from the
struct pci_bus, so let's take this opportunit to simplify the API and
make life easier for the callers.

Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_bind.c     | 3 +--
 drivers/acpi/pci_irq.c      | 7 ++++---
 include/acpi/acpi_drivers.h | 2 +-
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/acpi/pci_bind.c b/drivers/acpi/pci_bind.c
index 6eb58ef366ef..62cb383222f8 100644
--- a/drivers/acpi/pci_bind.c
+++ b/drivers/acpi/pci_bind.c
@@ -113,8 +113,7 @@ static int acpi_pci_unbind(struct acpi_device *device)
 		return 0;
 
 	if (dev->subordinate)
-		acpi_pci_irq_del_prt(pci_domain_nr(dev->bus),
-				     dev->subordinate->number);
+		acpi_pci_irq_del_prt(dev->subordinate);
 
 	pci_dev_put(dev);
 	return 0;
diff --git a/drivers/acpi/pci_irq.c b/drivers/acpi/pci_irq.c
index 3ed944cefb36..ef9509e33191 100644
--- a/drivers/acpi/pci_irq.c
+++ b/drivers/acpi/pci_irq.c
@@ -280,16 +280,17 @@ int acpi_pci_irq_add_prt(acpi_handle handle, struct pci_bus *bus)
 	return 0;
 }
 
-void acpi_pci_irq_del_prt(int segment, int bus)
+void acpi_pci_irq_del_prt(struct pci_bus *bus)
 {
 	struct acpi_prt_entry *entry, *tmp;
 
 	printk(KERN_DEBUG
 	       "ACPI: Delete PCI Interrupt Routing Table for %04x:%02x\n",
-	       segment, bus);
+	       pci_domain_nr(bus), bus->number);
 	spin_lock(&acpi_prt_lock);
 	list_for_each_entry_safe(entry, tmp, &acpi_prt_list, list) {
-		if (segment == entry->id.segment && bus == entry->id.bus) {
+		if (pci_domain_nr(bus) == entry->id.segment
+			&& bus->number == entry->id.bus) {
 			list_del(&entry->list);
 			kfree(entry);
 		}
diff --git a/include/acpi/acpi_drivers.h b/include/acpi/acpi_drivers.h
index 686a960bde39..98ebaaedc07f 100644
--- a/include/acpi/acpi_drivers.h
+++ b/include/acpi/acpi_drivers.h
@@ -92,7 +92,7 @@ int acpi_pci_link_free_irq(acpi_handle handle);
 /* ACPI PCI Interrupt Routing (pci_irq.c) */
 
 int acpi_pci_irq_add_prt(acpi_handle handle, struct pci_bus *bus);
-void acpi_pci_irq_del_prt(int segment, int bus);
+void acpi_pci_irq_del_prt(struct pci_bus *bus);
 
 /* ACPI PCI Device Binding (pci_bind.c) */
 
-- 
cgit v1.2.3-59-g8ed1b


From 97719a8726fe8d3ea12a85fbf4f514a915ba30ec Mon Sep 17 00:00:00 2001
From: Alexander Chiang <achiang@hp.com>
Date: Wed, 10 Jun 2009 19:55:45 +0000
Subject: ACPI: acpi_pci_unbind should clean up properly after acpi_pci_bind

In acpi_pci_bind, we set device->ops.bind and device->ops.unbind, but
never clear them out.

Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_bind.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/acpi/pci_bind.c b/drivers/acpi/pci_bind.c
index 62cb383222f8..a205769f1d00 100644
--- a/drivers/acpi/pci_bind.c
+++ b/drivers/acpi/pci_bind.c
@@ -109,12 +109,15 @@ static int acpi_pci_unbind(struct acpi_device *device)
 	struct pci_dev *dev;
 
 	dev = acpi_get_pci_dev(device->handle);
-	if (!dev)
-		return 0;
+	if (!dev || !dev->subordinate)
+		goto out;
 
-	if (dev->subordinate)
-		acpi_pci_irq_del_prt(dev->subordinate);
+	acpi_pci_irq_del_prt(dev->subordinate);
+
+	device->ops.bind = NULL;
+	device->ops.unbind = NULL;
 
+out:
 	pci_dev_put(dev);
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From d6aa484c1c0cd39ff3a42f4050b55d2a5b285ef5 Mon Sep 17 00:00:00 2001
From: Alexander Chiang <achiang@hp.com>
Date: Wed, 10 Jun 2009 19:55:50 +0000
Subject: PCI Hotplug: acpiphp: convert to acpi_get_pci_dev

Now that acpi_get_pci_dev is available, let's use it instead of
acpi_get_pci_id.

Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/pci/hotplug/acpiphp_glue.c | 25 +++++++------------------
 1 file changed, 7 insertions(+), 18 deletions(-)

diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index fc6636e3300b..0cb0f830a993 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -678,18 +678,9 @@ static void remove_bridge(acpi_handle handle)
 
 static struct pci_dev * get_apic_pci_info(acpi_handle handle)
 {
-	struct acpi_pci_id id;
-	struct pci_bus *bus;
 	struct pci_dev *dev;
 
-	if (ACPI_FAILURE(acpi_get_pci_id(handle, &id)))
-		return NULL;
-
-	bus = pci_find_bus(id.segment, id.bus);
-	if (!bus)
-		return NULL;
-
-	dev = pci_get_slot(bus, PCI_DEVFN(id.device, id.function));
+	dev = acpi_get_pci_dev(handle);
 	if (!dev)
 		return NULL;
 
@@ -1396,19 +1387,16 @@ static void acpiphp_sanitize_bus(struct pci_bus *bus)
 /* Program resources in newly inserted bridge */
 static int acpiphp_configure_bridge (acpi_handle handle)
 {
-	struct acpi_pci_id pci_id;
+	struct pci_dev *dev;
 	struct pci_bus *bus;
 
-	if (ACPI_FAILURE(acpi_get_pci_id(handle, &pci_id))) {
+	dev = acpi_get_pci_dev(handle);
+	if (!dev) {
 		err("cannot get PCI domain and bus number for bridge\n");
 		return -EINVAL;
 	}
-	bus = pci_find_bus(pci_id.segment, pci_id.bus);
-	if (!bus) {
-		err("cannot find bus %d:%d\n",
-				pci_id.segment, pci_id.bus);
-		return -EINVAL;
-	}
+
+	bus = dev->bus;
 
 	pci_bus_size_bridges(bus);
 	pci_bus_assign_resources(bus);
@@ -1416,6 +1404,7 @@ static int acpiphp_configure_bridge (acpi_handle handle)
 	acpiphp_set_hpp_values(handle, bus);
 	pci_enable_bridges(bus);
 	acpiphp_configure_ioapics(handle);
+	pci_dev_put(dev);
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 80ffdedf6020a77adcd06c01cfe6c488312b28f8 Mon Sep 17 00:00:00 2001
From: Alexander Chiang <achiang@hp.com>
Date: Wed, 10 Jun 2009 19:55:55 +0000
Subject: ACPI: kill acpi_get_pci_id

acpi_get_pci_dev() is better, and all callers have been converted, so
eliminate acpi_get_pci_id().

Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_bind.c     | 71 ---------------------------------------------
 include/acpi/acpi_drivers.h |  1 -
 2 files changed, 72 deletions(-)

diff --git a/drivers/acpi/pci_bind.c b/drivers/acpi/pci_bind.c
index a205769f1d00..a5a77b78a723 100644
--- a/drivers/acpi/pci_bind.c
+++ b/drivers/acpi/pci_bind.c
@@ -33,77 +33,6 @@
 #define _COMPONENT		ACPI_PCI_COMPONENT
 ACPI_MODULE_NAME("pci_bind");
 
-struct acpi_pci_data {
-	struct acpi_pci_id id;
-	struct pci_bus *bus;
-	struct pci_dev *dev;
-};
-
-static int acpi_pci_bind(struct acpi_device *device);
-static int acpi_pci_unbind(struct acpi_device *device);
-
-static void acpi_pci_data_handler(acpi_handle handle, u32 function,
-				  void *context)
-{
-
-	/* TBD: Anything we need to do here? */
-
-	return;
-}
-
-/**
- * acpi_get_pci_id
- * ------------------
- * This function is used by the ACPI Interpreter (a.k.a. Core Subsystem)
- * to resolve PCI information for ACPI-PCI devices defined in the namespace.
- * This typically occurs when resolving PCI operation region information.
- */
-acpi_status acpi_get_pci_id(acpi_handle handle, struct acpi_pci_id *id)
-{
-	int result = 0;
-	acpi_status status = AE_OK;
-	struct acpi_device *device = NULL;
-	struct acpi_pci_data *data = NULL;
-
-
-	if (!id)
-		return AE_BAD_PARAMETER;
-
-	result = acpi_bus_get_device(handle, &device);
-	if (result) {
-		printk(KERN_ERR PREFIX
-			    "Invalid ACPI Bus context for device %s\n",
-			    acpi_device_bid(device));
-		return AE_NOT_EXIST;
-	}
-
-	status = acpi_get_data(handle, acpi_pci_data_handler, (void **)&data);
-	if (ACPI_FAILURE(status) || !data) {
-		ACPI_EXCEPTION((AE_INFO, status,
-				"Invalid ACPI-PCI context for device %s",
-				acpi_device_bid(device)));
-		return status;
-	}
-
-	*id = data->id;
-
-	/*
-	   id->segment = data->id.segment;
-	   id->bus = data->id.bus;
-	   id->device = data->id.device;
-	   id->function = data->id.function;
-	 */
-
-	ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-			  "Device %s has PCI address %04x:%02x:%02x.%d\n",
-			  acpi_device_bid(device), id->segment, id->bus,
-			  id->device, id->function));
-
-	return AE_OK;
-}
-
-EXPORT_SYMBOL(acpi_get_pci_id);
-
 static int acpi_pci_unbind(struct acpi_device *device)
 {
 	struct pci_dev *dev;
diff --git a/include/acpi/acpi_drivers.h b/include/acpi/acpi_drivers.h
index 98ebaaedc07f..b69285773177 100644
--- a/include/acpi/acpi_drivers.h
+++ b/include/acpi/acpi_drivers.h
@@ -99,7 +99,6 @@ void acpi_pci_irq_del_prt(struct pci_bus *bus);
 struct pci_bus;
 
 struct pci_dev *acpi_get_pci_dev(acpi_handle);
-acpi_status acpi_get_pci_id(acpi_handle handle, struct acpi_pci_id *id);
 int acpi_pci_bind_root(struct acpi_device *device);
 
 /* Arch-defined function to add a bus to the system */
-- 
cgit v1.2.3-59-g8ed1b


From 1e4cffe78e1decd937c7b78410eec87da6b87954 Mon Sep 17 00:00:00 2001
From: Alexander Chiang <achiang@hp.com>
Date: Wed, 10 Jun 2009 19:56:00 +0000
Subject: ACPI: video: convert to acpi_get_pci_dev

Now that acpi_get_pci_dev is available, let's use it instead of
acpi_get_physical_pci_device()

Cc: Thomas Renninger <trenn@suse.de>
Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/video.c        | 6 +++---
 drivers/acpi/video_detect.c | 9 +++++----
 2 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c
index 1bdfb37377e3..5adbf9361e60 100644
--- a/drivers/acpi/video.c
+++ b/drivers/acpi/video.c
@@ -1054,15 +1054,15 @@ static void acpi_video_bus_find_cap(struct acpi_video_bus *video)
 static int acpi_video_bus_check(struct acpi_video_bus *video)
 {
 	acpi_status status = -ENOENT;
-	struct device *dev;
+	struct pci_dev *dev;
 
 	if (!video)
 		return -EINVAL;
 
-	dev = acpi_get_physical_pci_device(video->device->handle);
+	dev = acpi_get_pci_dev(video->device->handle);
 	if (!dev)
 		return -ENODEV;
-	put_device(dev);
+	pci_dev_put(dev);
 
 	/* Since there is no HID, CID and so on for VGA driver, we have
 	 * to check well known required nodes.
diff --git a/drivers/acpi/video_detect.c b/drivers/acpi/video_detect.c
index 09737275e25f..7cd2b63435ea 100644
--- a/drivers/acpi/video_detect.c
+++ b/drivers/acpi/video_detect.c
@@ -10,7 +10,7 @@
  * assinged
  *
  * After PCI devices are glued with ACPI devices
- * acpi_get_physical_pci_device() can be called to identify ACPI graphics
+ * acpi_get_pci_dev() can be called to identify ACPI graphics
  * devices for which a real graphics card is plugged in
  *
  * Now acpi_video_get_capabilities() can be called to check which
@@ -36,6 +36,7 @@
 
 #include <linux/acpi.h>
 #include <linux/dmi.h>
+#include <linux/pci.h>
 
 ACPI_MODULE_NAME("video");
 #define _COMPONENT		ACPI_VIDEO_COMPONENT
@@ -109,7 +110,7 @@ static acpi_status
 find_video(acpi_handle handle, u32 lvl, void *context, void **rv)
 {
 	long *cap = context;
-	struct device *dev;
+	struct pci_dev *dev;
 	struct acpi_device *acpi_dev;
 
 	const struct acpi_device_id video_ids[] = {
@@ -120,10 +121,10 @@ find_video(acpi_handle handle, u32 lvl, void *context, void **rv)
 		return AE_OK;
 
 	if (!acpi_match_device_ids(acpi_dev, video_ids)) {
-		dev = acpi_get_physical_pci_device(handle);
+		dev = acpi_get_pci_dev(handle);
 		if (!dev)
 			return AE_OK;
-		put_device(dev);
+		pci_dev_put(dev);
 		*cap |= acpi_is_video_device(acpi_dev);
 	}
 	return AE_OK;
-- 
cgit v1.2.3-59-g8ed1b


From 7fe2a6c275a5bcec52fb3ef643daaf8265b7af0d Mon Sep 17 00:00:00 2001
From: Alexander Chiang <achiang@hp.com>
Date: Wed, 10 Jun 2009 19:56:05 +0000
Subject: ACPI: kill acpi_get_physical_pci_device()

acpi_get_pci_dev() is (hopefully) better, and all callers have been
converted, so let's get rid of this duplicated functionality.

Cc: Thomas Renninger <trenn@suse.de>
Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/glue.c     | 40 ----------------------------------------
 include/acpi/acpi_bus.h |  1 -
 2 files changed, 41 deletions(-)

diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c
index 8bd2c2a6884d..a8a5c29958c8 100644
--- a/drivers/acpi/glue.c
+++ b/drivers/acpi/glue.c
@@ -140,46 +140,6 @@ struct device *acpi_get_physical_device(acpi_handle handle)
 
 EXPORT_SYMBOL(acpi_get_physical_device);
 
-/* ToDo: When a PCI bridge is found, return the PCI device behind the bridge
- *       This should work in general, but did not on a Lenovo T61 for the
- *	 graphics card. But this must be fixed when the PCI device is
- *       bound and the kernel device struct is attached to the acpi device
- * Note: A success call will increase reference count by one
- *       Do call put_device(dev) on the returned device then
- */
-struct device *acpi_get_physical_pci_device(acpi_handle handle)
-{
-	struct device *dev;
-	long long device_id;
-	acpi_status status;
-
-	status =
-		acpi_evaluate_integer(handle, "_ADR", NULL, &device_id);
-
-	if (ACPI_FAILURE(status))
-		return NULL;
-
-	/* We need to attempt to determine whether the _ADR refers to a
-	   PCI device or not. There's no terribly good way to do this,
-	   so the best we can hope for is to assume that there'll never
-	   be a device in the host bridge */
-	if (device_id >= 0x10000) {
-		/* It looks like a PCI device. Does it exist? */
-		dev = acpi_get_physical_device(handle);
-	} else {
-		/* It doesn't look like a PCI device. Does its parent
-		   exist? */
-		acpi_handle phandle;
-		if (acpi_get_parent(handle, &phandle))
-			return NULL;
-		dev = acpi_get_physical_device(phandle);
-	}
-	if (!dev)
-		return NULL;
-	return dev;
-}
-EXPORT_SYMBOL(acpi_get_physical_pci_device);
-
 static int acpi_bind_one(struct device *dev, acpi_handle handle)
 {
 	struct acpi_device *acpi_dev;
diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index 96d593ee4859..700b263c898a 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -365,7 +365,6 @@ struct acpi_bus_type {
 int register_acpi_bus_type(struct acpi_bus_type *);
 int unregister_acpi_bus_type(struct acpi_bus_type *);
 struct device *acpi_get_physical_device(acpi_handle);
-struct device *acpi_get_physical_pci_device(acpi_handle);
 
 /* helper */
 acpi_handle acpi_get_child(acpi_handle, acpi_integer);
-- 
cgit v1.2.3-59-g8ed1b


From 75d71c40dde5a9474c09ee291df22d50a1215bef Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario_limonciello@dell.com>
Date: Wed, 10 Jun 2009 19:40:46 +0000
Subject: dell-wmi: mask off upper bytes of event response

In debugging with some future machines that actually contain BIOS level
support for dell-wmi, I've determined that the upper half of the data that
comes back from wmi_get_event_data may sometimes contain extra information
that isn't currently relevant when pulling scan codes out of the data.
This causes dell-wmi to improperly respond to these events.

Signed-off-by: Mario Limonciello <mario_limonciello@dell.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/dell-wmi.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/platform/x86/dell-wmi.c b/drivers/platform/x86/dell-wmi.c
index 2fab94162147..8a0d39ee9217 100644
--- a/drivers/platform/x86/dell-wmi.c
+++ b/drivers/platform/x86/dell-wmi.c
@@ -122,7 +122,12 @@ static void dell_wmi_notify(u32 value, void *context)
 
 	if (obj && obj->type == ACPI_TYPE_BUFFER) {
 		int *buffer = (int *)obj->buffer.pointer;
-		key = dell_wmi_get_entry_by_scancode(buffer[1]);
+		/*
+		 *  The upper bytes of the event may contain
+		 *  additional information, so mask them off for the
+		 *  scancode lookup
+		 */
+		key = dell_wmi_get_entry_by_scancode(buffer[1] & 0xFFFF);
 		if (key) {
 			input_report_key(dell_wmi_input_dev, key->keycode, 1);
 			input_sync(dell_wmi_input_dev);
-- 
cgit v1.2.3-59-g8ed1b


From 5cab0098171712a9fd51399b06181c8dfdebe9c9 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario_limonciello@dell.com>
Date: Wed, 10 Jun 2009 19:40:47 +0000
Subject: dell-wmi: add additional keyboard events

Upcoming Dell hardware will send more keyboard events via WMI.  Add
support for them.

Signed-off-by: Mario Limonciello <mario_limonciello@dell.com>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/dell-wmi.c | 45 ++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 44 insertions(+), 1 deletion(-)

diff --git a/drivers/platform/x86/dell-wmi.c b/drivers/platform/x86/dell-wmi.c
index 8a0d39ee9217..9f345dc2d1b1 100644
--- a/drivers/platform/x86/dell-wmi.c
+++ b/drivers/platform/x86/dell-wmi.c
@@ -46,10 +46,53 @@ struct key_entry {
 	u16 keycode;
 };
 
-enum { KE_KEY, KE_SW, KE_END };
+enum { KE_KEY, KE_SW, KE_IGNORE, KE_END };
+
+/*
+ * Certain keys are flagged as KE_IGNORE. All of these are either
+ * notifications (rather than requests for change) or are also sent
+ * via the keyboard controller so should not be sent again.
+ */
 
 static struct key_entry dell_wmi_keymap[] = {
 	{KE_KEY, 0xe045, KEY_PROG1},
+	{KE_KEY, 0xe009, KEY_EJECTCD},
+
+	/* These also contain the brightness level at offset 6 */
+	{KE_KEY, 0xe006, KEY_BRIGHTNESSUP},
+	{KE_KEY, 0xe005, KEY_BRIGHTNESSDOWN},
+
+	/* Battery health status button */
+	{KE_KEY, 0xe007, KEY_BATTERY},
+
+	/* This is actually for all radios. Although physically a
+	 * switch, the notification does not provide an indication of
+	 * state and so it should be reported as a key */
+	{KE_KEY, 0xe008, KEY_WLAN},
+
+	/* The next device is at offset 6, the active devices are at
+	   offset 8 and the attached devices at offset 10 */
+	{KE_KEY, 0xe00b, KEY_DISPLAYTOGGLE},
+
+	{KE_IGNORE, 0xe00c, KEY_KBDILLUMTOGGLE},
+
+	/* BIOS error detected */
+	{KE_IGNORE, 0xe00d, KEY_RESERVED},
+
+	/* Wifi Catcher */
+	{KE_KEY, 0xe011, KEY_PROG2},
+
+	/* Ambient light sensor toggle */
+	{KE_IGNORE, 0xe013, KEY_RESERVED},
+
+	{KE_IGNORE, 0xe020, KEY_MUTE},
+	{KE_IGNORE, 0xe02e, KEY_VOLUMEDOWN},
+	{KE_IGNORE, 0xe030, KEY_VOLUMEUP},
+	{KE_IGNORE, 0xe033, KEY_KBDILLUMUP},
+	{KE_IGNORE, 0xe034, KEY_KBDILLUMDOWN},
+	{KE_IGNORE, 0xe03a, KEY_CAPSLOCK},
+	{KE_IGNORE, 0xe045, KEY_NUMLOCK},
+	{KE_IGNORE, 0xe046, KEY_SCROLLLOCK},
 	{KE_END, 0}
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From db18b040af6571a7eeed9e1adc2e92c9c87e4b1a Mon Sep 17 00:00:00 2001
From: Matthew Garrett <mjg@redhat.com>
Date: Wed, 10 Jun 2009 19:40:48 +0000
Subject: dell-wmi: don't generate errors on empty messages

There's no point in generating kernel messages if we didn't receive a
parsable keyboard event - only do so if there appeared to be a scancode.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/dell-wmi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/platform/x86/dell-wmi.c b/drivers/platform/x86/dell-wmi.c
index 9f345dc2d1b1..0f900cc9fa7a 100644
--- a/drivers/platform/x86/dell-wmi.c
+++ b/drivers/platform/x86/dell-wmi.c
@@ -176,9 +176,9 @@ static void dell_wmi_notify(u32 value, void *context)
 			input_sync(dell_wmi_input_dev);
 			input_report_key(dell_wmi_input_dev, key->keycode, 0);
 			input_sync(dell_wmi_input_dev);
-		} else
+		} else if (buffer[1] & 0xFFFF)
 			printk(KERN_INFO "dell-wmi: Unknown key %x pressed\n",
-			       buffer[1]);
+			       buffer[1] & 0xFFFF);
 	}
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 871043bc463e7d191e7b5b00436a8852921dd833 Mon Sep 17 00:00:00 2001
From: Matthew Garrett <mjg@redhat.com>
Date: Mon, 1 Jun 2009 15:25:45 +0100
Subject: hp-wmi: Add support for reporting tablet state

HP tablets send a WMI event when a tablet state change occurs, but use the
same method as is used for reporting docking and undocking. The same query
is used to obtain the state of the hardware. Bit 0 indicates the docking
state, while bit 2 indicates the tablet state. This patch breaks these out
and sends separate input events for tablet and dock state changes. An
additional sysfs file is added to report the tablet state.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/hp-wmi.c | 87 +++++++++++++++++++++++++++----------------
 1 file changed, 55 insertions(+), 32 deletions(-)

diff --git a/drivers/platform/x86/hp-wmi.c b/drivers/platform/x86/hp-wmi.c
index 50d9019de2be..46a7a6e8ed91 100644
--- a/drivers/platform/x86/hp-wmi.c
+++ b/drivers/platform/x86/hp-wmi.c
@@ -47,7 +47,7 @@ MODULE_ALIAS("wmi:5FB7F034-2C63-45e9-BE91-3D44E2C707E4");
 #define HPWMI_DISPLAY_QUERY 0x1
 #define HPWMI_HDDTEMP_QUERY 0x2
 #define HPWMI_ALS_QUERY 0x3
-#define HPWMI_DOCK_QUERY 0x4
+#define HPWMI_HARDWARE_QUERY 0x4
 #define HPWMI_WIRELESS_QUERY 0x5
 #define HPWMI_HOTKEY_QUERY 0xc
 
@@ -75,10 +75,9 @@ struct key_entry {
 	u16 keycode;
 };
 
-enum { KE_KEY, KE_SW, KE_END };
+enum { KE_KEY, KE_END };
 
 static struct key_entry hp_wmi_keymap[] = {
-	{KE_SW, 0x01, SW_DOCK},
 	{KE_KEY, 0x02, KEY_BRIGHTNESSUP},
 	{KE_KEY, 0x03, KEY_BRIGHTNESSDOWN},
 	{KE_KEY, 0x20e6, KEY_PROG1},
@@ -151,7 +150,22 @@ static int hp_wmi_als_state(void)
 
 static int hp_wmi_dock_state(void)
 {
-	return hp_wmi_perform_query(HPWMI_DOCK_QUERY, 0, 0);
+	int ret = hp_wmi_perform_query(HPWMI_HARDWARE_QUERY, 0, 0);
+
+	if (ret < 0)
+		return ret;
+
+	return ret & 0x1;
+}
+
+static int hp_wmi_tablet_state(void)
+{
+	int ret = hp_wmi_perform_query(HPWMI_HARDWARE_QUERY, 0, 0);
+
+	if (ret < 0)
+		return ret;
+
+	return (ret & 0x4) ? 1 : 0;
 }
 
 static int hp_wmi_wifi_set(void *data, enum rfkill_state state)
@@ -244,6 +258,15 @@ static ssize_t show_dock(struct device *dev, struct device_attribute *attr,
 	return sprintf(buf, "%d\n", value);
 }
 
+static ssize_t show_tablet(struct device *dev, struct device_attribute *attr,
+			 char *buf)
+{
+	int value = hp_wmi_tablet_state();
+	if (value < 0)
+		return -EINVAL;
+	return sprintf(buf, "%d\n", value);
+}
+
 static ssize_t set_als(struct device *dev, struct device_attribute *attr,
 		       const char *buf, size_t count)
 {
@@ -256,6 +279,7 @@ static DEVICE_ATTR(display, S_IRUGO, show_display, NULL);
 static DEVICE_ATTR(hddtemp, S_IRUGO, show_hddtemp, NULL);
 static DEVICE_ATTR(als, S_IRUGO | S_IWUSR, show_als, set_als);
 static DEVICE_ATTR(dock, S_IRUGO, show_dock, NULL);
+static DEVICE_ATTR(tablet, S_IRUGO, show_tablet, NULL);
 
 static struct key_entry *hp_wmi_get_entry_by_scancode(int code)
 {
@@ -338,13 +362,13 @@ static void hp_wmi_notify(u32 value, void *context)
 						 key->keycode, 0);
 				input_sync(hp_wmi_input_dev);
 				break;
-			case KE_SW:
-				input_report_switch(hp_wmi_input_dev,
-						    key->keycode,
-						    hp_wmi_dock_state());
-				input_sync(hp_wmi_input_dev);
-				break;
 			}
+		} else if (eventcode == 0x1) {
+			input_report_switch(hp_wmi_input_dev, SW_DOCK,
+					    hp_wmi_dock_state());
+			input_report_switch(hp_wmi_input_dev, SW_TABLET_MODE,
+					    hp_wmi_tablet_state());
+			input_sync(hp_wmi_input_dev);
 		} else if (eventcode == 0x5) {
 			if (wifi_rfkill)
 				rfkill_force_state(wifi_rfkill,
@@ -381,18 +405,19 @@ static int __init hp_wmi_input_setup(void)
 			set_bit(EV_KEY, hp_wmi_input_dev->evbit);
 			set_bit(key->keycode, hp_wmi_input_dev->keybit);
 			break;
-		case KE_SW:
-			set_bit(EV_SW, hp_wmi_input_dev->evbit);
-			set_bit(key->keycode, hp_wmi_input_dev->swbit);
-
-			/* Set initial dock state */
-			input_report_switch(hp_wmi_input_dev, key->keycode,
-					    hp_wmi_dock_state());
-			input_sync(hp_wmi_input_dev);
-			break;
 		}
 	}
 
+	set_bit(EV_SW, hp_wmi_input_dev->evbit);
+	set_bit(SW_DOCK, hp_wmi_input_dev->swbit);
+	set_bit(SW_TABLET_MODE, hp_wmi_input_dev->swbit);
+
+	/* Set initial hardware state */
+	input_report_switch(hp_wmi_input_dev, SW_DOCK, hp_wmi_dock_state());
+	input_report_switch(hp_wmi_input_dev, SW_TABLET_MODE,
+			    hp_wmi_tablet_state());
+	input_sync(hp_wmi_input_dev);
+
 	err = input_register_device(hp_wmi_input_dev);
 
 	if (err) {
@@ -409,6 +434,7 @@ static void cleanup_sysfs(struct platform_device *device)
 	device_remove_file(&device->dev, &dev_attr_hddtemp);
 	device_remove_file(&device->dev, &dev_attr_als);
 	device_remove_file(&device->dev, &dev_attr_dock);
+	device_remove_file(&device->dev, &dev_attr_tablet);
 }
 
 static int __init hp_wmi_bios_setup(struct platform_device *device)
@@ -426,6 +452,9 @@ static int __init hp_wmi_bios_setup(struct platform_device *device)
 	if (err)
 		goto add_sysfs_error;
 	err = device_create_file(&device->dev, &dev_attr_dock);
+	if (err)
+		goto add_sysfs_error;
+	err = device_create_file(&device->dev, &dev_attr_tablet);
 	if (err)
 		goto add_sysfs_error;
 
@@ -491,23 +520,17 @@ static int __exit hp_wmi_bios_remove(struct platform_device *device)
 
 static int hp_wmi_resume_handler(struct platform_device *device)
 {
-	struct key_entry *key;
-
 	/*
-	 * Docking state may have changed while suspended, so trigger
-	 * an input event for the current state. As this is a switch,
+	 * Hardware state may have changed while suspended, so trigger
+	 * input events for the current state. As this is a switch,
 	 * the input layer will only actually pass it on if the state
 	 * changed.
 	 */
-	for (key = hp_wmi_keymap; key->type != KE_END; key++) {
-		switch (key->type) {
-		case KE_SW:
-			input_report_switch(hp_wmi_input_dev, key->keycode,
-					    hp_wmi_dock_state());
-			input_sync(hp_wmi_input_dev);
-			break;
-		}
-	}
+
+	input_report_switch(hp_wmi_input_dev, SW_DOCK, hp_wmi_dock_state());
+	input_report_switch(hp_wmi_input_dev, SW_TABLET_MODE,
+			    hp_wmi_tablet_state());
+	input_sync(hp_wmi_input_dev);
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 6d2781310036a8d3fa2b590a6f83a298010fd64a Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 30 Apr 2009 09:35:37 -0600
Subject: ACPI: allow drivers to request both device and system notify events

System notify events (0x00-0x7f) are common across all device types
and should be handled in Linux/ACPI, not in drivers.  However, some
BIOSes use system notify events in device-specific ways that require
the driver to be involved.

This patch adds a ACPI_DRIVER_ALL_NOTIFY_EVENTS driver flag.  When a
driver sets this flag and supplies a .notify method, Linux/ACPI calls
the .notify method for ALL notify events on the device, not just the
device-specific (0x80-0xff) events.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/bus.c      | 6 +++++-
 include/acpi/acpi_bus.h | 3 +++
 2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c
index ae862f1798dc..cdfecc0a2ac6 100644
--- a/drivers/acpi/bus.c
+++ b/drivers/acpi/bus.c
@@ -549,6 +549,7 @@ static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 {
 	int result = 0;
 	struct acpi_device *device = NULL;
+	struct acpi_driver *driver;
 
 	blocking_notifier_call_chain(&acpi_bus_notify_list,
 		type, (void *)handle);
@@ -629,7 +630,10 @@ static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 		break;
 	}
 
-	return;
+	driver = device->driver;
+	if (driver && driver->ops.notify &&
+	    (driver->flags & ACPI_DRIVER_ALL_NOTIFY_EVENTS))
+		driver->ops.notify(device, type);
 }
 
 /* --------------------------------------------------------------------------
diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h
index c34b11022908..84e35d5646a1 100644
--- a/include/acpi/acpi_bus.h
+++ b/include/acpi/acpi_bus.h
@@ -114,10 +114,13 @@ struct acpi_device_ops {
 	acpi_op_notify notify;
 };
 
+#define ACPI_DRIVER_ALL_NOTIFY_EVENTS	0x1	/* system AND device events */
+
 struct acpi_driver {
 	char name[80];
 	char class[80];
 	const struct acpi_device_id *ids; /* Supported Hardware IDs */
+	unsigned int flags;
 	struct acpi_device_ops ops;
 	struct device_driver drv;
 	struct module *owner;
-- 
cgit v1.2.3-59-g8ed1b


From 48fe112744d1ff2e899a6491633ac58a3229aabf Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 30 Apr 2009 09:35:42 -0600
Subject: ACPI: ac: use .notify method instead of installing handler directly

This patch adds a .notify() method.  The presence of .notify() causes
Linux/ACPI to manage event handlers and notify handlers on our behalf,
so we don't have to install and remove them ourselves.

This driver apparently relies on seeing ALL notify events, not just
device-specific ones (because it used ACPI_ALL_NOTIFY).  We use the
ACPI_DRIVER_ALL_NOTIFY_EVENTS driver flag to request all events.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
CC: Alexey Starikovskiy <alexey.y.starikovskiy@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/ac.c | 20 +++++---------------
 1 file changed, 5 insertions(+), 15 deletions(-)

diff --git a/drivers/acpi/ac.c b/drivers/acpi/ac.c
index 88e42abf5d88..0df8fcb687d6 100644
--- a/drivers/acpi/ac.c
+++ b/drivers/acpi/ac.c
@@ -61,6 +61,7 @@ static int acpi_ac_open_fs(struct inode *inode, struct file *file);
 static int acpi_ac_add(struct acpi_device *device);
 static int acpi_ac_remove(struct acpi_device *device, int type);
 static int acpi_ac_resume(struct acpi_device *device);
+static void acpi_ac_notify(struct acpi_device *device, u32 event);
 
 static const struct acpi_device_id ac_device_ids[] = {
 	{"ACPI0003", 0},
@@ -72,10 +73,12 @@ static struct acpi_driver acpi_ac_driver = {
 	.name = "ac",
 	.class = ACPI_AC_CLASS,
 	.ids = ac_device_ids,
+	.flags = ACPI_DRIVER_ALL_NOTIFY_EVENTS,
 	.ops = {
 		.add = acpi_ac_add,
 		.remove = acpi_ac_remove,
 		.resume = acpi_ac_resume,
+		.notify = acpi_ac_notify,
 		},
 };
 
@@ -220,16 +223,14 @@ static int acpi_ac_remove_fs(struct acpi_device *device)
                                    Driver Model
    -------------------------------------------------------------------------- */
 
-static void acpi_ac_notify(acpi_handle handle, u32 event, void *data)
+static void acpi_ac_notify(struct acpi_device *device, u32 event)
 {
-	struct acpi_ac *ac = data;
-	struct acpi_device *device = NULL;
+	struct acpi_ac *ac = acpi_driver_data(device);
 
 
 	if (!ac)
 		return;
 
-	device = ac->device;
 	switch (event) {
 	default:
 		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
@@ -253,7 +254,6 @@ static void acpi_ac_notify(acpi_handle handle, u32 event, void *data)
 static int acpi_ac_add(struct acpi_device *device)
 {
 	int result = 0;
-	acpi_status status = AE_OK;
 	struct acpi_ac *ac = NULL;
 
 
@@ -286,13 +286,6 @@ static int acpi_ac_add(struct acpi_device *device)
 	ac->charger.get_property = get_ac_property;
 	power_supply_register(&ac->device->dev, &ac->charger);
 #endif
-	status = acpi_install_notify_handler(device->handle,
-					     ACPI_ALL_NOTIFY, acpi_ac_notify,
-					     ac);
-	if (ACPI_FAILURE(status)) {
-		result = -ENODEV;
-		goto end;
-	}
 
 	printk(KERN_INFO PREFIX "%s [%s] (%s)\n",
 	       acpi_device_name(device), acpi_device_bid(device),
@@ -328,7 +321,6 @@ static int acpi_ac_resume(struct acpi_device *device)
 
 static int acpi_ac_remove(struct acpi_device *device, int type)
 {
-	acpi_status status = AE_OK;
 	struct acpi_ac *ac = NULL;
 
 
@@ -337,8 +329,6 @@ static int acpi_ac_remove(struct acpi_device *device, int type)
 
 	ac = acpi_driver_data(device);
 
-	status = acpi_remove_notify_handler(device->handle,
-					    ACPI_ALL_NOTIFY, acpi_ac_notify);
 #ifdef CONFIG_ACPI_SYSFS_POWER
 	if (ac->charger.dev)
 		power_supply_unregister(&ac->charger);
-- 
cgit v1.2.3-59-g8ed1b


From d94066910943837558d2a461c6766da981260bf0 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 30 Apr 2009 09:35:47 -0600
Subject: ACPI: battery: use .notify method instead of installing handler
 directly

This patch adds a .notify() method.  The presence of .notify() causes
Linux/ACPI to manage event handlers and notify handlers on our behalf,
so we don't have to install and remove them ourselves.

This driver apparently relies on seeing ALL notify events, not just
device-specific ones (because it used ACPI_ALL_NOTIFY).  We use the
ACPI_DRIVER_ALL_NOTIFY_EVENTS driver flag to request all events.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
CC: Alexey Starikovskiy <alexey.y.starikovskiy@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/battery.c | 22 +++++-----------------
 1 file changed, 5 insertions(+), 17 deletions(-)

diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c
index b0de6312919a..eb00c4e3747a 100644
--- a/drivers/acpi/battery.c
+++ b/drivers/acpi/battery.c
@@ -796,13 +796,12 @@ static void acpi_battery_remove_fs(struct acpi_device *device)
                                  Driver Interface
    -------------------------------------------------------------------------- */
 
-static void acpi_battery_notify(acpi_handle handle, u32 event, void *data)
+static void acpi_battery_notify(struct acpi_device *device, u32 event)
 {
-	struct acpi_battery *battery = data;
-	struct acpi_device *device;
+	struct acpi_battery *battery = acpi_driver_data(device);
+
 	if (!battery)
 		return;
-	device = battery->device;
 	acpi_battery_update(battery);
 	acpi_bus_generate_proc_event(device, event,
 				     acpi_battery_present(battery));
@@ -819,7 +818,6 @@ static void acpi_battery_notify(acpi_handle handle, u32 event, void *data)
 static int acpi_battery_add(struct acpi_device *device)
 {
 	int result = 0;
-	acpi_status status = 0;
 	struct acpi_battery *battery = NULL;
 	if (!device)
 		return -EINVAL;
@@ -837,14 +835,6 @@ static int acpi_battery_add(struct acpi_device *device)
 	if (result)
 		goto end;
 #endif
-	status = acpi_install_notify_handler(device->handle,
-					     ACPI_ALL_NOTIFY,
-					     acpi_battery_notify, battery);
-	if (ACPI_FAILURE(status)) {
-		ACPI_EXCEPTION((AE_INFO, status, "Installing notify handler"));
-		result = -ENODEV;
-		goto end;
-	}
 	printk(KERN_INFO PREFIX "%s Slot [%s] (battery %s)\n",
 	       ACPI_BATTERY_DEVICE_NAME, acpi_device_bid(device),
 	       device->status.battery_present ? "present" : "absent");
@@ -860,15 +850,11 @@ static int acpi_battery_add(struct acpi_device *device)
 
 static int acpi_battery_remove(struct acpi_device *device, int type)
 {
-	acpi_status status = 0;
 	struct acpi_battery *battery = NULL;
 
 	if (!device || !acpi_driver_data(device))
 		return -EINVAL;
 	battery = acpi_driver_data(device);
-	status = acpi_remove_notify_handler(device->handle,
-					    ACPI_ALL_NOTIFY,
-					    acpi_battery_notify);
 #ifdef CONFIG_ACPI_PROCFS_POWER
 	acpi_battery_remove_fs(device);
 #endif
@@ -896,10 +882,12 @@ static struct acpi_driver acpi_battery_driver = {
 	.name = "battery",
 	.class = ACPI_BATTERY_CLASS,
 	.ids = battery_device_ids,
+	.flags = ACPI_DRIVER_ALL_NOTIFY_EVENTS,
 	.ops = {
 		.add = acpi_battery_add,
 		.resume = acpi_battery_resume,
 		.remove = acpi_battery_remove,
+		.notify = acpi_battery_notify,
 		},
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 586ed1604fd6137cae1e8ede8143c3b8897306fd Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 30 Apr 2009 09:35:53 -0600
Subject: ACPI: asus-laptop: use .notify method instead of installing handler
 directly

This patch adds a .notify() method.  The presence of .notify() causes
Linux/ACPI to manage event handlers and notify handlers on our behalf,
so we don't have to install and remove them ourselves.

This driver apparently relies on seeing ALL notify events, not just
device-specific ones (because it used ACPI_ALL_NOTIFY).  We use the
ACPI_DRIVER_ALL_NOTIFY_EVENTS driver flag to request all events.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
CC: Corentin Chary <corentincj@iksaif.net>
CC: acpi4asus-user@lists.sourceforge.net
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/asus-laptop.c | 23 +++++------------------
 1 file changed, 5 insertions(+), 18 deletions(-)

diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c
index bfc1a8892a32..eaffe732653a 100644
--- a/drivers/platform/x86/asus-laptop.c
+++ b/drivers/platform/x86/asus-laptop.c
@@ -207,13 +207,17 @@ MODULE_DEVICE_TABLE(acpi, asus_device_ids);
 
 static int asus_hotk_add(struct acpi_device *device);
 static int asus_hotk_remove(struct acpi_device *device, int type);
+static void asus_hotk_notify(struct acpi_device *device, u32 event);
+
 static struct acpi_driver asus_hotk_driver = {
 	.name = ASUS_HOTK_NAME,
 	.class = ASUS_HOTK_CLASS,
 	.ids = asus_device_ids,
+	.flags = ACPI_DRIVER_ALL_NOTIFY_EVENTS,
 	.ops = {
 		.add = asus_hotk_add,
 		.remove = asus_hotk_remove,
+		.notify = asus_hotk_notify,
 		},
 };
 
@@ -812,7 +816,7 @@ static int asus_setkeycode(struct input_dev *dev, int scancode, int keycode)
 	return -EINVAL;
 }
 
-static void asus_hotk_notify(acpi_handle handle, u32 event, void *data)
+static void asus_hotk_notify(struct acpi_device *device, u32 event)
 {
 	static struct key_entry *key;
 	u16 count;
@@ -1124,7 +1128,6 @@ static int asus_hotk_found;
 
 static int asus_hotk_add(struct acpi_device *device)
 {
-	acpi_status status = AE_OK;
 	int result;
 
 	if (!device)
@@ -1149,15 +1152,6 @@ static int asus_hotk_add(struct acpi_device *device)
 
 	asus_hotk_add_fs();
 
-	/*
-	 * We install the handler, it will receive the hotk in parameter, so, we
-	 * could add other data to the hotk struct
-	 */
-	status = acpi_install_notify_handler(hotk->handle, ACPI_ALL_NOTIFY,
-					     asus_hotk_notify, hotk);
-	if (ACPI_FAILURE(status))
-		printk(ASUS_ERR "Error installing notify handler\n");
-
 	asus_hotk_found = 1;
 
 	/* WLED and BLED are on by default */
@@ -1198,16 +1192,9 @@ end:
 
 static int asus_hotk_remove(struct acpi_device *device, int type)
 {
-	acpi_status status = 0;
-
 	if (!device || !acpi_driver_data(device))
 		return -EINVAL;
 
-	status = acpi_remove_notify_handler(hotk->handle, ACPI_ALL_NOTIFY,
-					    asus_hotk_notify);
-	if (ACPI_FAILURE(status))
-		printk(ASUS_ERR "Error removing notify handler\n");
-
 	kfree(hotk->name);
 	kfree(hotk);
 
-- 
cgit v1.2.3-59-g8ed1b


From 352fa202c3320ac4844cd38fa72c7a91d7c4cfea Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 30 Apr 2009 09:35:58 -0600
Subject: ACPI: asus-acpi: use .notify method instead of installing handler
 directly

This patch adds a .notify() method.  The presence of .notify() causes
Linux/ACPI to manage event handlers and notify handlers on our behalf,
so we don't have to install and remove them ourselves.

This driver relies on seeing system notify events, not device-specific
ones (because it used ACPI_SYSTEM_NOTIFY).  We use the
ACPI_DRIVER_ALL_NOTIFY_EVENTS driver flag to request all events, then
just ignore any device events we get.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
CC: Corentin Chary <corentincj@iksaif.net>
CC: Karol Kozimor <sziwan@users.sourceforge.net>
CC: acpi4asus-user@lists.sourceforge.net
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/asus_acpi.c | 30 +++++++++++++-----------------
 1 file changed, 13 insertions(+), 17 deletions(-)

diff --git a/drivers/platform/x86/asus_acpi.c b/drivers/platform/x86/asus_acpi.c
index ba1f7497e4b9..ddf5240ade8c 100644
--- a/drivers/platform/x86/asus_acpi.c
+++ b/drivers/platform/x86/asus_acpi.c
@@ -455,6 +455,8 @@ static struct asus_hotk *hotk;
  */
 static int asus_hotk_add(struct acpi_device *device);
 static int asus_hotk_remove(struct acpi_device *device, int type);
+static void asus_hotk_notify(struct acpi_device *device, u32 event);
+
 static const struct acpi_device_id asus_device_ids[] = {
 	{"ATK0100", 0},
 	{"", 0},
@@ -465,9 +467,11 @@ static struct acpi_driver asus_hotk_driver = {
 	.name = "asus_acpi",
 	.class = ACPI_HOTK_CLASS,
 	.ids = asus_device_ids,
+	.flags = ACPI_DRIVER_ALL_NOTIFY_EVENTS,
 	.ops = {
 		.add = asus_hotk_add,
 		.remove = asus_hotk_remove,
+		.notify = asus_hotk_notify,
 		},
 };
 
@@ -1101,12 +1105,20 @@ static int asus_hotk_remove_fs(struct acpi_device *device)
 	return 0;
 }
 
-static void asus_hotk_notify(acpi_handle handle, u32 event, void *data)
+static void asus_hotk_notify(struct acpi_device *device, u32 event)
 {
 	/* TODO Find a better way to handle events count. */
 	if (!hotk)
 		return;
 
+	/*
+	 * The BIOS *should* be sending us device events, but apparently
+	 * Asus uses system events instead, so just ignore any device
+	 * events we get.
+	 */
+	if (event > ACPI_MAX_SYS_NOTIFY)
+		return;
+
 	if ((event & ~((u32) BR_UP)) < 16)
 		hotk->brightness = (event & ~((u32) BR_UP));
 	else if ((event & ~((u32) BR_DOWN)) < 16)
@@ -1346,15 +1358,6 @@ static int asus_hotk_add(struct acpi_device *device)
 	if (result)
 		goto end;
 
-	/*
-	 * We install the handler, it will receive the hotk in parameter, so, we
-	 * could add other data to the hotk struct
-	 */
-	status = acpi_install_notify_handler(hotk->handle, ACPI_SYSTEM_NOTIFY,
-					     asus_hotk_notify, hotk);
-	if (ACPI_FAILURE(status))
-		printk(KERN_ERR "  Error installing notify handler\n");
-
 	/* For laptops without GPLV: init the hotk->brightness value */
 	if ((!hotk->methods->brightness_get)
 	    && (!hotk->methods->brightness_status)
@@ -1389,16 +1392,9 @@ end:
 
 static int asus_hotk_remove(struct acpi_device *device, int type)
 {
-	acpi_status status = 0;
-
 	if (!device || !acpi_driver_data(device))
 		return -EINVAL;
 
-	status = acpi_remove_notify_handler(hotk->handle, ACPI_SYSTEM_NOTIFY,
-					    asus_hotk_notify);
-	if (ACPI_FAILURE(status))
-		printk(KERN_ERR "Asus ACPI: Error removing notify handler\n");
-
 	asus_hotk_remove_fs(device);
 
 	kfree(hotk);
-- 
cgit v1.2.3-59-g8ed1b


From d9b9bd7b4a579ff0340d29c2547b952a920639e6 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 30 Apr 2009 09:36:03 -0600
Subject: ACPI: eeepc-laptop: use .notify method instead of installing handler
 directly

This patch adds a .notify() method.  The presence of .notify() causes
Linux/ACPI to manage event handlers and notify handlers on our behalf,
so we don't have to install and remove them ourselves.

This driver relies on seeing system notify events, not device-specific
ones (because it used ACPI_SYSTEM_NOTIFY).  We use the
ACPI_DRIVER_ALL_NOTIFY_EVENTS driver flag to request all events, then
just ignore any device events we get.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
CC: Corentin Chary <corentincj@iksaif.net>
CC: acpi4asus-user@lists.sourceforge.net
CC: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/eeepc-laptop.c | 18 ++++++------------
 1 file changed, 6 insertions(+), 12 deletions(-)

diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index 353a898c3693..1e28413060b2 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -180,6 +180,7 @@ static struct key_entry eeepc_keymap[] = {
  */
 static int eeepc_hotk_add(struct acpi_device *device);
 static int eeepc_hotk_remove(struct acpi_device *device, int type);
+static void eeepc_hotk_notify(struct acpi_device *device, u32 event);
 
 static const struct acpi_device_id eeepc_device_ids[] = {
 	{EEEPC_HOTK_HID, 0},
@@ -191,9 +192,11 @@ static struct acpi_driver eeepc_hotk_driver = {
 	.name = EEEPC_HOTK_NAME,
 	.class = EEEPC_HOTK_CLASS,
 	.ids = eeepc_device_ids,
+	.flags = ACPI_DRIVER_ALL_NOTIFY_EVENTS,
 	.ops = {
 		.add = eeepc_hotk_add,
 		.remove = eeepc_hotk_remove,
+		.notify = eeepc_hotk_notify,
 	},
 };
 
@@ -569,7 +572,7 @@ static void eeepc_rfkill_notify(acpi_handle handle, u32 event, void *data)
 	rfkill_force_state(ehotk->eeepc_wlan_rfkill, state);
 }
 
-static void eeepc_hotk_notify(acpi_handle handle, u32 event, void *data)
+static void eeepc_hotk_notify(struct acpi_device *device, u32 event)
 {
 	static struct key_entry *key;
 	u16 count;
@@ -577,6 +580,8 @@ static void eeepc_hotk_notify(acpi_handle handle, u32 event, void *data)
 
 	if (!ehotk)
 		return;
+	if (event > ACPI_MAX_SYS_NOTIFY)
+		return;
 	if (event >= NOTIFY_BRN_MIN && event <= NOTIFY_BRN_MAX)
 		brn = notify_brn();
 	count = ehotk->event_count[event % 128]++;
@@ -657,7 +662,6 @@ static void eeepc_unregister_rfkill_notifier(char *node)
 
 static int eeepc_hotk_add(struct acpi_device *device)
 {
-	acpi_status status = AE_OK;
 	int result;
 
 	if (!device)
@@ -675,10 +679,6 @@ static int eeepc_hotk_add(struct acpi_device *device)
 	result = eeepc_hotk_check();
 	if (result)
 		goto ehotk_fail;
-	status = acpi_install_notify_handler(ehotk->handle, ACPI_SYSTEM_NOTIFY,
-					     eeepc_hotk_notify, ehotk);
-	if (ACPI_FAILURE(status))
-		printk(EEEPC_ERR "Error installing notify handler\n");
 
 	eeepc_register_rfkill_notifier("\\_SB.PCI0.P0P6");
 	eeepc_register_rfkill_notifier("\\_SB.PCI0.P0P7");
@@ -759,14 +759,8 @@ static int eeepc_hotk_add(struct acpi_device *device)
 
 static int eeepc_hotk_remove(struct acpi_device *device, int type)
 {
-	acpi_status status = 0;
-
 	if (!device || !acpi_driver_data(device))
 		 return -EINVAL;
-	status = acpi_remove_notify_handler(ehotk->handle, ACPI_SYSTEM_NOTIFY,
-					    eeepc_hotk_notify);
-	if (ACPI_FAILURE(status))
-		printk(EEEPC_ERR "Error removing notify handler\n");
 
 	eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P6");
 	eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P7");
-- 
cgit v1.2.3-59-g8ed1b


From 02c37bd8d0737c31caaed9a65bd7cb80aefb4c9a Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Fri, 22 May 2009 11:43:41 -0600
Subject: ACPI: simplify notification debug messages

This replaces several messages that depend on the acpi_device struct
with a single message that uses just the acpi_handle.  We should be
able to deal with notifications to objects that do not yet have an
acpi_device struct.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/bus.c | 27 +++------------------------
 1 file changed, 3 insertions(+), 24 deletions(-)

diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c
index cdfecc0a2ac6..eb986385c57a 100644
--- a/drivers/acpi/bus.c
+++ b/drivers/acpi/bus.c
@@ -551,6 +551,9 @@ static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 	struct acpi_device *device = NULL;
 	struct acpi_driver *driver;
 
+	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Notification %#02x to handle %p\n",
+			  type, handle));
+
 	blocking_notifier_call_chain(&acpi_bus_notify_list,
 		type, (void *)handle);
 
@@ -560,9 +563,6 @@ static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 	switch (type) {
 
 	case ACPI_NOTIFY_BUS_CHECK:
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-				  "Received BUS CHECK notification for device [%s]\n",
-				  device->pnp.bus_id));
 		result = acpi_bus_check_scope(device);
 		/*
 		 * TBD: We'll need to outsource certain events to non-ACPI
@@ -571,9 +571,6 @@ static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 		break;
 
 	case ACPI_NOTIFY_DEVICE_CHECK:
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-				  "Received DEVICE CHECK notification for device [%s]\n",
-				  device->pnp.bus_id));
 		result = acpi_bus_check_device(device, NULL);
 		/*
 		 * TBD: We'll need to outsource certain events to non-ACPI
@@ -582,44 +579,26 @@ static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 		break;
 
 	case ACPI_NOTIFY_DEVICE_WAKE:
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-				  "Received DEVICE WAKE notification for device [%s]\n",
-				  device->pnp.bus_id));
 		/* TBD */
 		break;
 
 	case ACPI_NOTIFY_EJECT_REQUEST:
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-				  "Received EJECT REQUEST notification for device [%s]\n",
-				  device->pnp.bus_id));
 		/* TBD */
 		break;
 
 	case ACPI_NOTIFY_DEVICE_CHECK_LIGHT:
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-				  "Received DEVICE CHECK LIGHT notification for device [%s]\n",
-				  device->pnp.bus_id));
 		/* TBD: Exactly what does 'light' mean? */
 		break;
 
 	case ACPI_NOTIFY_FREQUENCY_MISMATCH:
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-				  "Received FREQUENCY MISMATCH notification for device [%s]\n",
-				  device->pnp.bus_id));
 		/* TBD */
 		break;
 
 	case ACPI_NOTIFY_BUS_MODE_MISMATCH:
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-				  "Received BUS MODE MISMATCH notification for device [%s]\n",
-				  device->pnp.bus_id));
 		/* TBD */
 		break;
 
 	case ACPI_NOTIFY_POWER_FAULT:
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-				  "Received POWER FAULT notification for device [%s]\n",
-				  device->pnp.bus_id));
 		/* TBD */
 		break;
 
-- 
cgit v1.2.3-59-g8ed1b


From aa8a149c0cc822e3886eb85b95cb2f7d67e5b7e6 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Fri, 22 May 2009 11:43:46 -0600
Subject: ACPI: remove unused "status_changed" return value from Check Device
 handling

Remove "status_changed" return from acpi_bus_check_device().  Nobody
does anything useful based on its value.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/bus.c | 22 +++-------------------
 1 file changed, 3 insertions(+), 19 deletions(-)

diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c
index eb986385c57a..19e78fb0a8d1 100644
--- a/drivers/acpi/bus.c
+++ b/drivers/acpi/bus.c
@@ -450,8 +450,7 @@ int acpi_bus_receive_event(struct acpi_bus_event *event)
                              Notification Handling
    -------------------------------------------------------------------------- */
 
-static int
-acpi_bus_check_device(struct acpi_device *device, int *status_changed)
+static int acpi_bus_check_device(struct acpi_device *device)
 {
 	acpi_status status = 0;
 	struct acpi_device_status old_status;
@@ -460,9 +459,6 @@ acpi_bus_check_device(struct acpi_device *device, int *status_changed)
 	if (!device)
 		return -EINVAL;
 
-	if (status_changed)
-		*status_changed = 0;
-
 	old_status = device->status;
 
 	/*
@@ -471,10 +467,6 @@ acpi_bus_check_device(struct acpi_device *device, int *status_changed)
 	 */
 	if (device->parent && !device->parent->status.present) {
 		device->status = device->parent->status;
-		if (STRUCT_TO_INT(old_status) != STRUCT_TO_INT(device->status)) {
-			if (status_changed)
-				*status_changed = 1;
-		}
 		return 0;
 	}
 
@@ -485,9 +477,6 @@ acpi_bus_check_device(struct acpi_device *device, int *status_changed)
 	if (STRUCT_TO_INT(old_status) == STRUCT_TO_INT(device->status))
 		return 0;
 
-	if (status_changed)
-		*status_changed = 1;
-
 	/*
 	 * Device Insertion/Removal
 	 */
@@ -505,20 +494,15 @@ acpi_bus_check_device(struct acpi_device *device, int *status_changed)
 static int acpi_bus_check_scope(struct acpi_device *device)
 {
 	int result = 0;
-	int status_changed = 0;
-
 
 	if (!device)
 		return -EINVAL;
 
 	/* Status Change? */
-	result = acpi_bus_check_device(device, &status_changed);
+	result = acpi_bus_check_device(device);
 	if (result)
 		return result;
 
-	if (!status_changed)
-		return 0;
-
 	/*
 	 * TBD: Enumerate child devices within this device's scope and
 	 *       run acpi_bus_check_device()'s on them.
@@ -571,7 +555,7 @@ static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 		break;
 
 	case ACPI_NOTIFY_DEVICE_CHECK:
-		result = acpi_bus_check_device(device, NULL);
+		result = acpi_bus_check_device(device);
 		/*
 		 * TBD: We'll need to outsource certain events to non-ACPI
 		 *      drivers via the device manager (device.c).
-- 
cgit v1.2.3-59-g8ed1b


From cdd5b8ca122cc4239375dee7fcdc658315c119e4 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Fri, 22 May 2009 11:43:51 -0600
Subject: ACPI: remove unused return values from Bus Check & Device Check
 handling

Remove return values from acpi_bus_check_device() and acpi_bus_check_scope()
since nobody looks at them.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/bus.c | 32 +++++++++++---------------------
 1 file changed, 11 insertions(+), 21 deletions(-)

diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c
index 19e78fb0a8d1..2b08c3dc79da 100644
--- a/drivers/acpi/bus.c
+++ b/drivers/acpi/bus.c
@@ -450,14 +450,13 @@ int acpi_bus_receive_event(struct acpi_bus_event *event)
                              Notification Handling
    -------------------------------------------------------------------------- */
 
-static int acpi_bus_check_device(struct acpi_device *device)
+static void acpi_bus_check_device(struct acpi_device *device)
 {
-	acpi_status status = 0;
+	acpi_status status;
 	struct acpi_device_status old_status;
 
-
 	if (!device)
-		return -EINVAL;
+		return;
 
 	old_status = device->status;
 
@@ -467,15 +466,15 @@ static int acpi_bus_check_device(struct acpi_device *device)
 	 */
 	if (device->parent && !device->parent->status.present) {
 		device->status = device->parent->status;
-		return 0;
+		return;
 	}
 
 	status = acpi_bus_get_status(device);
 	if (ACPI_FAILURE(status))
-		return -ENODEV;
+		return;
 
 	if (STRUCT_TO_INT(old_status) == STRUCT_TO_INT(device->status))
-		return 0;
+		return;
 
 	/*
 	 * Device Insertion/Removal
@@ -487,28 +486,20 @@ static int acpi_bus_check_device(struct acpi_device *device)
 		ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Device removal detected\n"));
 		/* TBD: Handle device removal */
 	}
-
-	return 0;
 }
 
-static int acpi_bus_check_scope(struct acpi_device *device)
+static void acpi_bus_check_scope(struct acpi_device *device)
 {
-	int result = 0;
-
 	if (!device)
-		return -EINVAL;
+		return;
 
 	/* Status Change? */
-	result = acpi_bus_check_device(device);
-	if (result)
-		return result;
+	acpi_bus_check_device(device);
 
 	/*
 	 * TBD: Enumerate child devices within this device's scope and
 	 *       run acpi_bus_check_device()'s on them.
 	 */
-
-	return 0;
 }
 
 static BLOCKING_NOTIFIER_HEAD(acpi_bus_notify_list);
@@ -531,7 +522,6 @@ EXPORT_SYMBOL_GPL(unregister_acpi_bus_notifier);
  */
 static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 {
-	int result = 0;
 	struct acpi_device *device = NULL;
 	struct acpi_driver *driver;
 
@@ -547,7 +537,7 @@ static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 	switch (type) {
 
 	case ACPI_NOTIFY_BUS_CHECK:
-		result = acpi_bus_check_scope(device);
+		acpi_bus_check_scope(device);
 		/*
 		 * TBD: We'll need to outsource certain events to non-ACPI
 		 *      drivers via the device manager (device.c).
@@ -555,7 +545,7 @@ static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 		break;
 
 	case ACPI_NOTIFY_DEVICE_CHECK:
-		result = acpi_bus_check_device(device);
+		acpi_bus_check_device(device);
 		/*
 		 * TBD: We'll need to outsource certain events to non-ACPI
 		 *      drivers via the device manager (device.c).
-- 
cgit v1.2.3-59-g8ed1b


From ff754e2e85557ed7244385f0f2053c80e8ac9948 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Fri, 22 May 2009 11:43:56 -0600
Subject: ACPI: use handle, not device, in system notification path

This patch changes the global system notification path so it uses the
acpi_handle, not the acpi_device.

System notifications often deal with device presence and status change.
In these cases, we may not have an acpi_device.  For example, we may
get a Device Check notification on an object that previously was not
present.  Since the object was not present, we would not have had an
acpi_device for it.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/bus.c | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c
index 2b08c3dc79da..2876fc70c3a9 100644
--- a/drivers/acpi/bus.c
+++ b/drivers/acpi/bus.c
@@ -450,11 +450,14 @@ int acpi_bus_receive_event(struct acpi_bus_event *event)
                              Notification Handling
    -------------------------------------------------------------------------- */
 
-static void acpi_bus_check_device(struct acpi_device *device)
+static void acpi_bus_check_device(acpi_handle handle)
 {
+	struct acpi_device *device;
 	acpi_status status;
 	struct acpi_device_status old_status;
 
+	if (acpi_bus_get_device(handle, &device))
+		return;
 	if (!device)
 		return;
 
@@ -488,13 +491,10 @@ static void acpi_bus_check_device(struct acpi_device *device)
 	}
 }
 
-static void acpi_bus_check_scope(struct acpi_device *device)
+static void acpi_bus_check_scope(acpi_handle handle)
 {
-	if (!device)
-		return;
-
 	/* Status Change? */
-	acpi_bus_check_device(device);
+	acpi_bus_check_device(handle);
 
 	/*
 	 * TBD: Enumerate child devices within this device's scope and
@@ -531,13 +531,10 @@ static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 	blocking_notifier_call_chain(&acpi_bus_notify_list,
 		type, (void *)handle);
 
-	if (acpi_bus_get_device(handle, &device))
-		return;
-
 	switch (type) {
 
 	case ACPI_NOTIFY_BUS_CHECK:
-		acpi_bus_check_scope(device);
+		acpi_bus_check_scope(handle);
 		/*
 		 * TBD: We'll need to outsource certain events to non-ACPI
 		 *      drivers via the device manager (device.c).
@@ -545,7 +542,7 @@ static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 		break;
 
 	case ACPI_NOTIFY_DEVICE_CHECK:
-		acpi_bus_check_device(device);
+		acpi_bus_check_device(handle);
 		/*
 		 * TBD: We'll need to outsource certain events to non-ACPI
 		 *      drivers via the device manager (device.c).
@@ -583,10 +580,13 @@ static void acpi_bus_notify(acpi_handle handle, u32 type, void *data)
 		break;
 	}
 
-	driver = device->driver;
-	if (driver && driver->ops.notify &&
-	    (driver->flags & ACPI_DRIVER_ALL_NOTIFY_EVENTS))
-		driver->ops.notify(device, type);
+	acpi_bus_get_device(handle, &device);
+	if (device) {
+		driver = device->driver;
+		if (driver && driver->ops.notify &&
+		    (driver->flags & ACPI_DRIVER_ALL_NOTIFY_EVENTS))
+			driver->ops.notify(device, type);
+	}
 }
 
 /* --------------------------------------------------------------------------
-- 
cgit v1.2.3-59-g8ed1b


From 586caae36cece718ff46b3a59b88af79e9f7a2e0 Mon Sep 17 00:00:00 2001
From: Len Brown <len.brown@intel.com>
Date: Thu, 18 Jun 2009 00:38:27 -0400
Subject: ACPI: battery: fix CONFIG_ACPI_PROCFS_POWER=n build warning
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drivers/acpi/battery.c:841: warning: label ‘end’ defined but not used

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/battery.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c
index eb00c4e3747a..58b4517ce712 100644
--- a/drivers/acpi/battery.c
+++ b/drivers/acpi/battery.c
@@ -832,14 +832,12 @@ static int acpi_battery_add(struct acpi_device *device)
 	acpi_battery_update(battery);
 #ifdef CONFIG_ACPI_PROCFS_POWER
 	result = acpi_battery_add_fs(device);
-	if (result)
-		goto end;
 #endif
-	printk(KERN_INFO PREFIX "%s Slot [%s] (battery %s)\n",
-	       ACPI_BATTERY_DEVICE_NAME, acpi_device_bid(device),
-	       device->status.battery_present ? "present" : "absent");
-      end:
-	if (result) {
+	if (!result) {
+		printk(KERN_INFO PREFIX "%s Slot [%s] (battery %s)\n",
+			ACPI_BATTERY_DEVICE_NAME, acpi_device_bid(device),
+			device->status.battery_present ? "present" : "absent");
+	} else {
 #ifdef CONFIG_ACPI_PROCFS_POWER
 		acpi_battery_remove_fs(device);
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 7e275cc4e8e20f82740bf40ae2f5695e9e35ff09 Mon Sep 17 00:00:00 2001
From: Len Brown <len.brown@intel.com>
Date: Fri, 15 May 2009 02:08:44 -0400
Subject: ACPI: idle: rename lapic timer workaround routines

cosmetic only.  The lapic_timer workaround routines
are specific to the lapic_timer, and are not acpi-generic.

old:

acpi_timer_check_state()
acpi_propagate_timer_broadcast()
acpi_state_timer_broadcast()

new:

lapic_timer_check_state()
lapic_timer_propagate_broadcast()
lapic_timer_state_broadcast()

also, simplify the code in acpi_processor_power_verify()
so that lapic_timer_check_state() is simply called
from one place for all valid C-states, including C1.

Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/processor_idle.c | 40 ++++++++++++++++++----------------------
 1 file changed, 18 insertions(+), 22 deletions(-)

diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index 10a2d913635a..1f60ccbd4c39 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -139,7 +139,7 @@ static void acpi_safe_halt(void)
  * are affected too. We pick the most conservative approach: we assume
  * that the local APIC stops in both C2 and C3.
  */
-static void acpi_timer_check_state(int state, struct acpi_processor *pr,
+static void lapic_timer_check_state(int state, struct acpi_processor *pr,
 				   struct acpi_processor_cx *cx)
 {
 	struct acpi_processor_power *pwr = &pr->power;
@@ -162,7 +162,7 @@ static void acpi_timer_check_state(int state, struct acpi_processor *pr,
 		pr->power.timer_broadcast_on_state = state;
 }
 
-static void acpi_propagate_timer_broadcast(struct acpi_processor *pr)
+static void lapic_timer_propagate_broadcast(struct acpi_processor *pr)
 {
 	unsigned long reason;
 
@@ -173,7 +173,7 @@ static void acpi_propagate_timer_broadcast(struct acpi_processor *pr)
 }
 
 /* Power(C) State timer broadcast control */
-static void acpi_state_timer_broadcast(struct acpi_processor *pr,
+static void lapic_timer_state_broadcast(struct acpi_processor *pr,
 				       struct acpi_processor_cx *cx,
 				       int broadcast)
 {
@@ -190,10 +190,10 @@ static void acpi_state_timer_broadcast(struct acpi_processor *pr,
 
 #else
 
-static void acpi_timer_check_state(int state, struct acpi_processor *pr,
+static void lapic_timer_check_state(int state, struct acpi_processor *pr,
 				   struct acpi_processor_cx *cstate) { }
-static void acpi_propagate_timer_broadcast(struct acpi_processor *pr) { }
-static void acpi_state_timer_broadcast(struct acpi_processor *pr,
+static void lapic_timer_propagate_broadcast(struct acpi_processor *pr) { }
+static void lapic_timer_state_broadcast(struct acpi_processor *pr,
 				       struct acpi_processor_cx *cx,
 				       int broadcast)
 {
@@ -614,29 +614,25 @@ static int acpi_processor_power_verify(struct acpi_processor *pr)
 		switch (cx->type) {
 		case ACPI_STATE_C1:
 			cx->valid = 1;
-			acpi_timer_check_state(i, pr, cx);
 			break;
 
 		case ACPI_STATE_C2:
 			acpi_processor_power_verify_c2(cx);
-			if (cx->valid)
-				acpi_timer_check_state(i, pr, cx);
 			break;
 
 		case ACPI_STATE_C3:
 			acpi_processor_power_verify_c3(pr, cx);
-			if (cx->valid)
-				acpi_timer_check_state(i, pr, cx);
 			break;
 		}
-		if (cx->valid)
-			tsc_check_state(cx->type);
+		if (!cx->valid)
+			continue;
 
-		if (cx->valid)
-			working++;
+		lapic_timer_check_state(i, pr, cx);
+		tsc_check_state(cx->type);
+		working++;
 	}
 
-	acpi_propagate_timer_broadcast(pr);
+	lapic_timer_propagate_broadcast(pr);
 
 	return (working);
 }
@@ -839,7 +835,7 @@ static int acpi_idle_enter_c1(struct cpuidle_device *dev,
 		return 0;
 	}
 
-	acpi_state_timer_broadcast(pr, cx, 1);
+	lapic_timer_state_broadcast(pr, cx, 1);
 	kt1 = ktime_get_real();
 	acpi_idle_do_entry(cx);
 	kt2 = ktime_get_real();
@@ -847,7 +843,7 @@ static int acpi_idle_enter_c1(struct cpuidle_device *dev,
 
 	local_irq_enable();
 	cx->usage++;
-	acpi_state_timer_broadcast(pr, cx, 0);
+	lapic_timer_state_broadcast(pr, cx, 0);
 
 	return idle_time;
 }
@@ -892,7 +888,7 @@ static int acpi_idle_enter_simple(struct cpuidle_device *dev,
 	 * Must be done before busmaster disable as we might need to
 	 * access HPET !
 	 */
-	acpi_state_timer_broadcast(pr, cx, 1);
+	lapic_timer_state_broadcast(pr, cx, 1);
 
 	if (cx->type == ACPI_STATE_C3)
 		ACPI_FLUSH_CPU_CACHE();
@@ -914,7 +910,7 @@ static int acpi_idle_enter_simple(struct cpuidle_device *dev,
 
 	cx->usage++;
 
-	acpi_state_timer_broadcast(pr, cx, 0);
+	lapic_timer_state_broadcast(pr, cx, 0);
 	cx->time += sleep_ticks;
 	return idle_time;
 }
@@ -981,7 +977,7 @@ static int acpi_idle_enter_bm(struct cpuidle_device *dev,
 	 * Must be done before busmaster disable as we might need to
 	 * access HPET !
 	 */
-	acpi_state_timer_broadcast(pr, cx, 1);
+	lapic_timer_state_broadcast(pr, cx, 1);
 
 	kt1 = ktime_get_real();
 	/*
@@ -1026,7 +1022,7 @@ static int acpi_idle_enter_bm(struct cpuidle_device *dev,
 
 	cx->usage++;
 
-	acpi_state_timer_broadcast(pr, cx, 0);
+	lapic_timer_state_broadcast(pr, cx, 0);
 	cx->time += sleep_ticks;
 	return idle_time;
 }
-- 
cgit v1.2.3-59-g8ed1b


From d7880f10c5d42ba182a97c1fd41d41d0b8837097 Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Thu, 18 Jun 2009 00:40:16 -0300
Subject: thinkpad-acpi: forbid the use of HBRV on Lenovo ThinkPads

Forcing thinkpad-acpi to do EC-based brightness control (HBRV) on a
X61 has very... interesting effects, instead of doing nothing (since
it doesn't have EC-based backlight control), it causes "weirdness" in
the fan tachometer readings, for example.

This means the EC register that used to be HBRV has been reused by
Lenovo for something else, but they didn't remove it from the DSDT.

Make sure the documentation reflects this data, and forbid the user
from forcing the driver to access HBRV on Lenovo ThinkPads.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 Documentation/laptops/thinkpad-acpi.txt | 14 ++++++++------
 drivers/platform/x86/thinkpad_acpi.c    | 10 ++++++++++
 2 files changed, 18 insertions(+), 6 deletions(-)

diff --git a/Documentation/laptops/thinkpad-acpi.txt b/Documentation/laptops/thinkpad-acpi.txt
index 88fc0661de56..f2ff638cce8d 100644
--- a/Documentation/laptops/thinkpad-acpi.txt
+++ b/Documentation/laptops/thinkpad-acpi.txt
@@ -1169,17 +1169,19 @@ may not be distinct.  Later Lenovo models that implement the ACPI
 display backlight brightness control methods have 16 levels, ranging
 from 0 to 15.
 
-There are two interfaces to the firmware for direct brightness control,
-EC and UCMS (or CMOS).  To select which one should be used, use the
-brightness_mode module parameter: brightness_mode=1 selects EC mode,
-brightness_mode=2 selects UCMS mode, brightness_mode=3 selects EC
-mode with NVRAM backing (so that brightness changes are remembered
-across shutdown/reboot).
+For IBM ThinkPads, there are two interfaces to the firmware for direct
+brightness control, EC and UCMS (or CMOS).  To select which one should be
+used, use the brightness_mode module parameter: brightness_mode=1 selects
+EC mode, brightness_mode=2 selects UCMS mode, brightness_mode=3 selects EC
+mode with NVRAM backing (so that brightness changes are remembered across
+shutdown/reboot).
 
 The driver tries to select which interface to use from a table of
 defaults for each ThinkPad model.  If it makes a wrong choice, please
 report this as a bug, so that we can fix it.
 
+Lenovo ThinkPads only support brightness_mode=2 (UCMS).
+
 When display backlight brightness controls are available through the
 standard ACPI interface, it is best to use it instead of this direct
 ThinkPad-specific interface.  The driver will disable its native
diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c
index 5a22a064222c..c8d74dbacbbd 100644
--- a/drivers/platform/x86/thinkpad_acpi.c
+++ b/drivers/platform/x86/thinkpad_acpi.c
@@ -5696,6 +5696,10 @@ static struct ibm_struct ecdump_driver_data = {
  *   Bit 3-0: backlight brightness level
  *
  * brightness_get_raw returns status data in the HBRV layout
+ *
+ * WARNING: The X61 has been verified to use HBRV for something else, so
+ * this should be used _only_ on IBM ThinkPads, and maybe with some careful
+ * testing on the very early *60 Lenovo models...
  */
 
 enum {
@@ -5996,6 +6000,12 @@ static int __init brightness_init(struct ibm_init_struct *iibm)
 			   brightness_mode);
 	}
 
+	/* Safety */
+	if (thinkpad_id.vendor != PCI_VENDOR_ID_IBM &&
+	    (brightness_mode == TPACPI_BRGHT_MODE_ECNVRAM ||
+	     brightness_mode == TPACPI_BRGHT_MODE_EC))
+		return -EINVAL;
+
 	if (tpacpi_brightness_get_raw(&b) < 0)
 		return 1;
 
-- 
cgit v1.2.3-59-g8ed1b


From d73772474f6ebbacbe820c31c0fa1cffa7160246 Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Thu, 18 Jun 2009 00:40:17 -0300
Subject: thinkpad-acpi: support the second fan on the X61

Support reading the tachometer of the auxiliary fan of a X60/X61.

It was found out by sheer luck, that bit 0 of EC register 0x31
(formely HBRV) selects which fan is active for tachometer readings
through EC 0x84/0x085: 0 for fan1, 1 for fan2.

Many thanks to Christoph Kl??nter, to Whoopie, and to weasel, who
helped confirm that behaviour.

Fan control through EC HFSP applies to both fans equally, regardless
of the state of bit 0 of EC 0x31.  That matches the way the DSDT uses
HFSP.

In order to better support the secondary fan, export a second
tachometer over hwmon, and add defensive measures to make sure we are
reading the correct tachometer.

Support for the second fan is whitelist-based, as I have not found
anything obvious to look for in the DSDT to detect the presence of
the second fan.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 Documentation/laptops/thinkpad-acpi.txt |  10 ++-
 drivers/platform/x86/thinkpad_acpi.c    | 122 +++++++++++++++++++++++++++++++-
 2 files changed, 130 insertions(+), 2 deletions(-)

diff --git a/Documentation/laptops/thinkpad-acpi.txt b/Documentation/laptops/thinkpad-acpi.txt
index f2ff638cce8d..0d5e91379ae8 100644
--- a/Documentation/laptops/thinkpad-acpi.txt
+++ b/Documentation/laptops/thinkpad-acpi.txt
@@ -1269,7 +1269,7 @@ Fan control and monitoring: fan speed, fan enable/disable
 
 procfs: /proc/acpi/ibm/fan
 sysfs device attributes: (hwmon "thinkpad") fan1_input, pwm1,
-			  pwm1_enable
+			  pwm1_enable, fan2_input
 sysfs hwmon driver attributes: fan_watchdog
 
 NOTE NOTE NOTE: fan control operations are disabled by default for
@@ -1282,6 +1282,9 @@ from the hardware registers of the embedded controller.  This is known
 to work on later R, T, X and Z series ThinkPads but may show a bogus
 value on other models.
 
+Some Lenovo ThinkPads support a secondary fan.  This fan cannot be
+controlled separately, it shares the main fan control.
+
 Fan levels:
 
 Most ThinkPad fans work in "levels" at the firmware interface.  Level 0
@@ -1412,6 +1415,11 @@ hwmon device attribute fan1_input:
 	which can take up to two minutes.  May return rubbish on older
 	ThinkPads.
 
+hwmon device attribute fan2_input:
+	Fan tachometer reading, in RPM, for the secondary fan.
+	Available only on some ThinkPads.  If the secondary fan is
+	not installed, will always read 0.
+
 hwmon driver attribute fan_watchdog:
 	Fan safety watchdog timer interval, in seconds.  Minimum is
 	1 second, maximum is 120 seconds.  0 disables the watchdog.
diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c
index c8d74dbacbbd..27ca676a7092 100644
--- a/drivers/platform/x86/thinkpad_acpi.c
+++ b/drivers/platform/x86/thinkpad_acpi.c
@@ -264,6 +264,7 @@ static struct {
 	u32 wan:1;
 	u32 uwb:1;
 	u32 fan_ctrl_status_undef:1;
+	u32 second_fan:1;
 	u32 beep_needs_two_args:1;
 	u32 input_device_registered:1;
 	u32 platform_drv_registered:1;
@@ -6298,6 +6299,21 @@ static struct ibm_struct volume_driver_data = {
  *	For firmware bugs, refer to:
  *	http://thinkwiki.org/wiki/Embedded_Controller_Firmware#Firmware_Issues
  *
+ *	----
+ *
+ *	ThinkPad EC register 0x31 bit 0 (only on select models)
+ *
+ *	When bit 0 of EC register 0x31 is zero, the tachometer registers
+ *	show the speed of the main fan.  When bit 0 of EC register 0x31
+ *	is one, the tachometer registers show the speed of the auxiliary
+ *	fan.
+ *
+ *	Fan control seems to affect both fans, regardless of the state
+ *	of this bit.
+ *
+ *	So far, only the firmware for the X60/X61 non-tablet versions
+ *	seem to support this (firmware TP-7M).
+ *
  * TPACPI_FAN_WR_ACPI_FANS:
  *	ThinkPad X31, X40, X41.  Not available in the X60.
  *
@@ -6324,6 +6340,8 @@ enum {					/* Fan control constants */
 	fan_status_offset = 0x2f,	/* EC register 0x2f */
 	fan_rpm_offset = 0x84,		/* EC register 0x84: LSB, 0x85 MSB (RPM)
 					 * 0x84 must be read before 0x85 */
+	fan_select_offset = 0x31,	/* EC register 0x31 (Firmware 7M)
+					   bit 0 selects which fan is active */
 
 	TP_EC_FAN_FULLSPEED = 0x40,	/* EC fan mode: full speed */
 	TP_EC_FAN_AUTO	    = 0x80,	/* EC fan mode: auto fan control */
@@ -6417,6 +6435,38 @@ static void fan_quirk1_handle(u8 *fan_status)
 	}
 }
 
+/* Select main fan on X60/X61, NOOP on others */
+static bool fan_select_fan1(void)
+{
+	if (tp_features.second_fan) {
+		u8 val;
+
+		if (ec_read(fan_select_offset, &val) < 0)
+			return false;
+		val &= 0xFEU;
+		if (ec_write(fan_select_offset, val) < 0)
+			return false;
+	}
+	return true;
+}
+
+/* Select secondary fan on X60/X61 */
+static bool fan_select_fan2(void)
+{
+	u8 val;
+
+	if (!tp_features.second_fan)
+		return false;
+
+	if (ec_read(fan_select_offset, &val) < 0)
+		return false;
+	val |= 0x01U;
+	if (ec_write(fan_select_offset, val) < 0)
+		return false;
+
+	return true;
+}
+
 /*
  * Call with fan_mutex held
  */
@@ -6494,6 +6544,8 @@ static int fan_get_speed(unsigned int *speed)
 	switch (fan_status_access_mode) {
 	case TPACPI_FAN_RD_TPEC:
 		/* all except 570, 600e/x, 770e, 770x */
+		if (unlikely(!fan_select_fan1()))
+			return -EIO;
 		if (unlikely(!acpi_ec_read(fan_rpm_offset, &lo) ||
 			     !acpi_ec_read(fan_rpm_offset + 1, &hi)))
 			return -EIO;
@@ -6510,6 +6562,34 @@ static int fan_get_speed(unsigned int *speed)
 	return 0;
 }
 
+static int fan2_get_speed(unsigned int *speed)
+{
+	u8 hi, lo;
+	bool rc;
+
+	switch (fan_status_access_mode) {
+	case TPACPI_FAN_RD_TPEC:
+		/* all except 570, 600e/x, 770e, 770x */
+		if (unlikely(!fan_select_fan2()))
+			return -EIO;
+		rc = !acpi_ec_read(fan_rpm_offset, &lo) ||
+			     !acpi_ec_read(fan_rpm_offset + 1, &hi);
+		fan_select_fan1(); /* play it safe */
+		if (rc)
+			return -EIO;
+
+		if (likely(speed))
+			*speed = (hi << 8) | lo;
+
+		break;
+
+	default:
+		return -ENXIO;
+	}
+
+	return 0;
+}
+
 static int fan_set_level(int level)
 {
 	if (!fan_control_allowed)
@@ -6915,6 +6995,25 @@ static struct device_attribute dev_attr_fan_fan1_input =
 	__ATTR(fan1_input, S_IRUGO,
 		fan_fan1_input_show, NULL);
 
+/* sysfs fan fan2_input ------------------------------------------------ */
+static ssize_t fan_fan2_input_show(struct device *dev,
+			   struct device_attribute *attr,
+			   char *buf)
+{
+	int res;
+	unsigned int speed;
+
+	res = fan2_get_speed(&speed);
+	if (res < 0)
+		return res;
+
+	return snprintf(buf, PAGE_SIZE, "%u\n", speed);
+}
+
+static struct device_attribute dev_attr_fan_fan2_input =
+	__ATTR(fan2_input, S_IRUGO,
+		fan_fan2_input_show, NULL);
+
 /* sysfs fan fan_watchdog (hwmon driver) ------------------------------- */
 static ssize_t fan_fan_watchdog_show(struct device_driver *drv,
 				     char *buf)
@@ -6948,6 +7047,7 @@ static DRIVER_ATTR(fan_watchdog, S_IWUSR | S_IRUGO,
 static struct attribute *fan_attributes[] = {
 	&dev_attr_fan_pwm1_enable.attr, &dev_attr_fan_pwm1.attr,
 	&dev_attr_fan_fan1_input.attr,
+	NULL, /* for fan2_input */
 	NULL
 };
 
@@ -6955,7 +7055,8 @@ static const struct attribute_group fan_attr_group = {
 	.attrs = fan_attributes,
 };
 
-#define	TPACPI_FAN_Q1	0x0001
+#define	TPACPI_FAN_Q1	0x0001		/* Unitialized HFSP */
+#define TPACPI_FAN_2FAN	0x0002		/* EC 0x31 bit 0 selects fan2 */
 
 #define TPACPI_FAN_QI(__id1, __id2, __quirks)	\
 	{ .vendor = PCI_VENDOR_ID_IBM,		\
@@ -6963,13 +7064,21 @@ static const struct attribute_group fan_attr_group = {
 	  .ec = TPID(__id1, __id2),		\
 	  .quirks = __quirks }
 
+#define TPACPI_FAN_QL(__id1, __id2, __quirks)	\
+	{ .vendor = PCI_VENDOR_ID_LENOVO,	\
+	  .bios = TPACPI_MATCH_ANY,		\
+	  .ec = TPID(__id1, __id2),		\
+	  .quirks = __quirks }
+
 static const struct tpacpi_quirk fan_quirk_table[] __initconst = {
 	TPACPI_FAN_QI('1', 'Y', TPACPI_FAN_Q1),
 	TPACPI_FAN_QI('7', '8', TPACPI_FAN_Q1),
 	TPACPI_FAN_QI('7', '6', TPACPI_FAN_Q1),
 	TPACPI_FAN_QI('7', '0', TPACPI_FAN_Q1),
+	TPACPI_FAN_QL('7', 'M', TPACPI_FAN_2FAN),
 };
 
+#undef TPACPI_FAN_QL
 #undef TPACPI_FAN_QI
 
 static int __init fan_init(struct ibm_init_struct *iibm)
@@ -6986,6 +7095,7 @@ static int __init fan_init(struct ibm_init_struct *iibm)
 	fan_control_commands = 0;
 	fan_watchdog_maxinterval = 0;
 	tp_features.fan_ctrl_status_undef = 0;
+	tp_features.second_fan = 0;
 	fan_control_desired_level = 7;
 
 	TPACPI_ACPIHANDLE_INIT(fans);
@@ -7006,6 +7116,11 @@ static int __init fan_init(struct ibm_init_struct *iibm)
 			fan_status_access_mode = TPACPI_FAN_RD_TPEC;
 			if (quirks & TPACPI_FAN_Q1)
 				fan_quirk1_setup();
+			if (quirks & TPACPI_FAN_2FAN) {
+				tp_features.second_fan = 1;
+				dbg_printk(TPACPI_DBG_INIT | TPACPI_DBG_FAN,
+					"secondary fan support enabled\n");
+			}
 		} else {
 			printk(TPACPI_ERR
 			       "ThinkPad ACPI EC access misbehaving, "
@@ -7061,6 +7176,11 @@ static int __init fan_init(struct ibm_init_struct *iibm)
 
 	if (fan_status_access_mode != TPACPI_FAN_NONE ||
 	    fan_control_access_mode != TPACPI_FAN_WR_NONE) {
+		if (tp_features.second_fan) {
+			/* attach second fan tachometer */
+			fan_attributes[ARRAY_SIZE(fan_attributes)-2] =
+					&dev_attr_fan_fan2_input.attr;
+		}
 		rc = sysfs_create_group(&tpacpi_sensors_pdev->dev.kobj,
 					 &fan_attr_group);
 		if (rc < 0)
-- 
cgit v1.2.3-59-g8ed1b


From 24a5d59f3477bcff4c069ff4d0ca9a3e037d0235 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Thu, 18 Jun 2009 12:33:16 +0200
Subject: udf: Use device size when drive reported bogus number of written
 blocks

Some drives report 0 as the number of written blocks when there are some blocks
recorded. Use device size in such case so that we can automagically mount such
media.

Signed-off-by: Jan Kara <jack@suse.cz>
---
 fs/udf/lowlevel.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/fs/udf/lowlevel.c b/fs/udf/lowlevel.c
index 703843f30ffd..1b88fd5df05d 100644
--- a/fs/udf/lowlevel.c
+++ b/fs/udf/lowlevel.c
@@ -56,7 +56,12 @@ unsigned long udf_get_last_block(struct super_block *sb)
 	struct block_device *bdev = sb->s_bdev;
 	unsigned long lblock = 0;
 
-	if (ioctl_by_bdev(bdev, CDROM_LAST_WRITTEN, (unsigned long) &lblock))
+	/*
+	 * ioctl failed or returned obviously bogus value?
+	 * Try using the device size...
+	 */
+	if (ioctl_by_bdev(bdev, CDROM_LAST_WRITTEN, (unsigned long) &lblock) ||
+	    lblock == 0)
 		lblock = bdev->bd_inode->i_size >> sb->s_blocksize_bits;
 
 	if (lblock)
-- 
cgit v1.2.3-59-g8ed1b


From 82e12fe9244ff653f703722a8937b595e10e71f4 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Tue, 16 Jun 2009 11:03:07 +1000
Subject: nfsd: don't take nfsd_mutex twice when setting number of threads.

Currently when we write a number to 'threads' in nfsdfs,
we take the nfsd_mutex, update the number of threads, then take the
mutex again to read the number of threads.

Mostly this isn't a big deal.  However if we are write '0', and
portmap happens to be dead, then we can get unpredictable behaviour.
If the nfsd threads all got killed quickly and the last thread is
waiting for portmap to respond, then the second time we take the mutex
we will block waiting for the last thread.
However if the nfsd threads didn't die quite that fast, then there
will be no contention when we try to take the mutex again.

Unpredictability isn't fun, and waiting for the last thread to exit is
pointless, so avoid taking the lock twice.
To achieve this, get nfsd_svc return a non-negative number of active
threads when not returning a negative error.

Signed-off-by: NeilBrown <neilb@suse.de>
---
 fs/nfsd/nfsctl.c | 14 +++++++++-----
 fs/nfsd/nfssvc.c |  6 ++++++
 2 files changed, 15 insertions(+), 5 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 877e713a0fd6..1250fb978ac1 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -207,10 +207,14 @@ static struct file_operations pool_stats_operations = {
 static ssize_t write_svc(struct file *file, char *buf, size_t size)
 {
 	struct nfsctl_svc *data;
+	int err;
 	if (size < sizeof(*data))
 		return -EINVAL;
 	data = (struct nfsctl_svc*) buf;
-	return nfsd_svc(data->svc_port, data->svc_nthreads);
+	err = nfsd_svc(data->svc_port, data->svc_nthreads);
+	if (err < 0)
+		return err;
+	return 0;
 }
 
 /**
@@ -692,12 +696,12 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size)
 		if (newthreads < 0)
 			return -EINVAL;
 		rv = nfsd_svc(NFS_PORT, newthreads);
-		if (rv)
+		if (rv < 0)
 			return rv;
-	}
+	} else
+		rv = nfsd_nrthreads();
 
-	return scnprintf(buf, SIMPLE_TRANSACTION_LIMIT, "%d\n",
-							nfsd_nrthreads());
+	return scnprintf(buf, SIMPLE_TRANSACTION_LIMIT, "%d\n", rv);
 }
 
 /**
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index cbba4a935786..209eaa0885d1 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -413,6 +413,12 @@ nfsd_svc(unsigned short port, int nrservs)
 		goto failure;
 
 	error = svc_set_num_threads(nfsd_serv, NULL, nrservs);
+	if (error == 0)
+		/* We are holding a reference to nfsd_serv which
+		 * we don't want to count in the return value,
+		 * so subtract 1
+		 */
+		error = nfsd_serv->sv_nrthreads - 1;
  failure:
 	svc_destroy(nfsd_serv);		/* Release server */
  out:
-- 
cgit v1.2.3-59-g8ed1b


From 671e1fcf63fd115eabcb693b06cbc2e4a3d1a3a3 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Tue, 16 Jun 2009 11:03:20 +1000
Subject: nfsd: optimise the starting of zero threads when none are running.

Currently, if we ask to set then number of nfsd threads to zero when
there are none running, we set up all the sockets and register the
service, and then tear it all down again.
This is pointless.

So detect that case and exit promptly.
(also remove an assignment to 'error' which was never used.

Signed-off-by: NeilBrown <neilb@suse.de>
Acked-by: Jeff Layton <jlayton@redhat.com>
---
 fs/nfsd/nfssvc.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 209eaa0885d1..d4c9884cd54b 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -390,12 +390,14 @@ nfsd_svc(unsigned short port, int nrservs)
 
 	mutex_lock(&nfsd_mutex);
 	dprintk("nfsd: creating service\n");
-	error = -EINVAL;
 	if (nrservs <= 0)
 		nrservs = 0;
 	if (nrservs > NFSD_MAXSERVS)
 		nrservs = NFSD_MAXSERVS;
-	
+	error = 0;
+	if (nrservs == 0 && nfsd_serv == NULL)
+		goto out;
+
 	/* Readahead param cache - will no-op if it already exists */
 	error =	nfsd_racache_init(2*nrservs);
 	if (error<0)
-- 
cgit v1.2.3-59-g8ed1b


From 110828c9cdce6e8ec68479ced4ca0bdc1135bb91 Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Tue, 16 Jun 2009 06:31:45 -0600
Subject: PCI: remove redundant __msi_set_enable()

We have the 'pos' of the MSI capability at all locations which call
msi_set_enable(), so pass it to msi_set_enable() instead of making it
find the capability every time.

Reviewed-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/msi.c | 38 ++++++++++++++++++--------------------
 1 file changed, 18 insertions(+), 20 deletions(-)

diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index f48f7550b4a7..79e56906c7f1 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -75,22 +75,17 @@ void arch_teardown_msi_irqs(struct pci_dev *dev)
 }
 #endif
 
-static void __msi_set_enable(struct pci_dev *dev, int pos, int enable)
+static void msi_set_enable(struct pci_dev *dev, int pos, int enable)
 {
 	u16 control;
 
-	if (pos) {
-		pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &control);
-		control &= ~PCI_MSI_FLAGS_ENABLE;
-		if (enable)
-			control |= PCI_MSI_FLAGS_ENABLE;
-		pci_write_config_word(dev, pos + PCI_MSI_FLAGS, control);
-	}
-}
+	BUG_ON(!pos);
 
-static void msi_set_enable(struct pci_dev *dev, int enable)
-{
-	__msi_set_enable(dev, pci_find_capability(dev, PCI_CAP_ID_MSI), enable);
+	pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &control);
+	control &= ~PCI_MSI_FLAGS_ENABLE;
+	if (enable)
+		control |= PCI_MSI_FLAGS_ENABLE;
+	pci_write_config_word(dev, pos + PCI_MSI_FLAGS, control);
 }
 
 static void msix_set_enable(struct pci_dev *dev, int enable)
@@ -300,7 +295,7 @@ static void __pci_restore_msi_state(struct pci_dev *dev)
 	pos = entry->msi_attrib.pos;
 
 	pci_intx_for_msi(dev, 0);
-	msi_set_enable(dev, 0);
+	msi_set_enable(dev, pos, 0);
 	write_msi_msg(dev->irq, &entry->msg);
 
 	pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &control);
@@ -362,9 +357,9 @@ static int msi_capability_init(struct pci_dev *dev, int nvec)
 	u16 control;
 	unsigned mask;
 
-	msi_set_enable(dev, 0);	/* Ensure msi is disabled as I set it up */
-
    	pos = pci_find_capability(dev, PCI_CAP_ID_MSI);
+	msi_set_enable(dev, pos, 0);	/* Disable MSI during set up */
+
 	pci_read_config_word(dev, msi_control_reg(pos), &control);
 	/* MSI Entry Initialization */
 	entry = alloc_msi_entry(dev);
@@ -396,7 +391,7 @@ static int msi_capability_init(struct pci_dev *dev, int nvec)
 
 	/* Set MSI enabled bits	 */
 	pci_intx_for_msi(dev, 0);
-	msi_set_enable(dev, 1);
+	msi_set_enable(dev, pos, 1);
 	dev->msi_enabled = 1;
 
 	dev->irq = entry->irq;
@@ -593,17 +588,20 @@ void pci_msi_shutdown(struct pci_dev *dev)
 	struct msi_desc *desc;
 	u32 mask;
 	u16 ctrl;
+	unsigned pos;
 
 	if (!pci_msi_enable || !dev || !dev->msi_enabled)
 		return;
 
-	msi_set_enable(dev, 0);
+	BUG_ON(list_empty(&dev->msi_list));
+	desc = list_first_entry(&dev->msi_list, struct msi_desc, list);
+	pos = desc->msi_attrib.pos;
+
+	msi_set_enable(dev, pos, 0);
 	pci_intx_for_msi(dev, 1);
 	dev->msi_enabled = 0;
 
-	BUG_ON(list_empty(&dev->msi_list));
-	desc = list_first_entry(&dev->msi_list, struct msi_desc, list);
-	pci_read_config_word(dev, desc->msi_attrib.pos + PCI_MSI_FLAGS, &ctrl);
+	pci_read_config_word(dev, pos + PCI_MSI_FLAGS, &ctrl);
 	mask = msi_capable_mask(ctrl);
 	msi_mask_irq(desc, mask, ~mask);
 
-- 
cgit v1.2.3-59-g8ed1b


From 268a03a42d3377d5fb41e6e7cbdec4e0b65cab2e Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Wed, 17 Jun 2009 19:03:57 -0600
Subject: PCI: drivers/pci/slot.c should depend on CONFIG_SYSFS

There is no way to interact with a physical PCI slot without
sysfs, so encode the dependency and prevent this build error:

	drivers/pci/slot.c: In function 'pci_hp_create_module_link':
	drivers/pci/slot.c:327: error: 'module_kset' undeclared

This patch _should_ make pci-sysfs.o depend on CONFIG_SYSFS too,
but we cannot (yet) because the PCI core merrily assumes the
existence of sysfs:

	drivers/built-in.o: In function `pci_bus_add_device':
	drivers/pci/bus.c:89: undefined reference to `pci_create_sysfs_dev_files'
	drivers/built-in.o: In function `pci_stop_dev':
	drivers/pci/remove.c:24: undefined reference to `pci_remove_sysfs_dev_files'

So do the minimal bit for now and figure out how to untangle it
later.

Reported-by: Randy Dunlap <randy.dunlap@oracle.com>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Fix-suggested-by: Matthew Wilcox <matthew@wil.cx>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/acpi/Kconfig        | 1 +
 drivers/pci/Makefile        | 3 ++-
 drivers/pci/hotplug/Kconfig | 2 +-
 3 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig
index 431f8b439553..7ec7d88c5999 100644
--- a/drivers/acpi/Kconfig
+++ b/drivers/acpi/Kconfig
@@ -266,6 +266,7 @@ config ACPI_DEBUG_FUNC_TRACE
 
 config ACPI_PCI_SLOT
 	tristate "PCI slot detection driver"
+	depends on SYSFS
 	default n
 	help
 	  This driver creates entries in /sys/bus/pci/slots/ for all PCI
diff --git a/drivers/pci/Makefile b/drivers/pci/Makefile
index ba6af162fd39..ed32f678247a 100644
--- a/drivers/pci/Makefile
+++ b/drivers/pci/Makefile
@@ -2,10 +2,11 @@
 # Makefile for the PCI bus specific drivers.
 #
 
-obj-y		+= access.o bus.o probe.o remove.o pci.o quirks.o slot.o \
+obj-y		+= access.o bus.o probe.o remove.o pci.o quirks.o \
 			pci-driver.o search.o pci-sysfs.o rom.o setup-res.o \
 			irq.o
 obj-$(CONFIG_PROC_FS) += proc.o
+obj-$(CONFIG_SYSFS) += slot.o
 
 # Build PCI Express stuff if needed
 obj-$(CONFIG_PCIEPORTBUS) += pcie/
diff --git a/drivers/pci/hotplug/Kconfig b/drivers/pci/hotplug/Kconfig
index ac888ccfa161..66f29bc00be4 100644
--- a/drivers/pci/hotplug/Kconfig
+++ b/drivers/pci/hotplug/Kconfig
@@ -4,7 +4,7 @@
 
 menuconfig HOTPLUG_PCI
 	tristate "Support for PCI Hotplug"
-	depends on PCI && HOTPLUG
+	depends on PCI && HOTPLUG && SYSFS
 	---help---
 	  Say Y here if you have a motherboard with a PCI Hotplug controller.
 	  This allows you to add and remove PCI cards while the machine is
-- 
cgit v1.2.3-59-g8ed1b


From dc64cd1131a3b5762e26bd8b01dc79030dd0c555 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:11:33 +0900
Subject: PCI ASPM: fix typo in struct pcie_link_state

Fix a typo in struct pcie_link_state.

The "sibiling" field in the struct pcie_link_state should be
"sibling".

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 777b2c76caf5..419f1f3697c0 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -32,7 +32,7 @@ struct endpoint_state {
 };
 
 struct pcie_link_state {
-	struct list_head sibiling;
+	struct list_head sibling;
 	struct pci_dev *pdev;
 	bool downstream_has_switch;
 
@@ -541,7 +541,7 @@ static void __pcie_aspm_configure_link_state(struct pci_dev *pdev,
 	state &= PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1;
 
 	/* check all links who have specific root port link */
-	list_for_each_entry(leaf, &link_list, sibiling) {
+	list_for_each_entry(leaf, &link_list, sibling) {
 		if (!list_empty(&leaf->children) ||
 			get_root_port_link(leaf) != root_port_link)
 			continue;
@@ -558,12 +558,12 @@ static void __pcie_aspm_configure_link_state(struct pci_dev *pdev,
 	 * __pcie_aspm_config_link for the order
 	 **/
 	if (state & PCIE_LINK_STATE_L1) {
-		list_for_each_entry(leaf, &link_list, sibiling) {
+		list_for_each_entry(leaf, &link_list, sibling) {
 			if (get_root_port_link(leaf) == root_port_link)
 				__pcie_aspm_config_link(leaf->pdev, state);
 		}
 	} else {
-		list_for_each_entry_reverse(leaf, &link_list, sibiling) {
+		list_for_each_entry_reverse(leaf, &link_list, sibling) {
 			if (get_root_port_link(leaf) == root_port_link)
 				__pcie_aspm_config_link(leaf->pdev, state);
 		}
@@ -682,7 +682,7 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
 	}
 
 	link_state->pdev = pdev;
-	list_add(&link_state->sibiling, &link_list);
+	list_add(&link_state->sibling, &link_list);
 
 	if (link_state->downstream_has_switch) {
 		/*
@@ -729,7 +729,7 @@ void pcie_aspm_exit_link_state(struct pci_dev *pdev)
 
 	/* All functions are removed, so just disable ASPM for the link */
 	__pcie_aspm_config_one_dev(parent, 0);
-	list_del(&link_state->sibiling);
+	list_del(&link_state->sibling);
 	list_del(&link_state->link);
 	/* Clock PM is for endpoint device */
 
@@ -806,7 +806,7 @@ static int pcie_aspm_set_policy(const char *val, struct kernel_param *kp)
 	down_read(&pci_bus_sem);
 	mutex_lock(&aspm_lock);
 	aspm_policy = i;
-	list_for_each_entry(link_state, &link_list, sibiling) {
+	list_for_each_entry(link_state, &link_list, sibling) {
 		pdev = link_state->pdev;
 		__pcie_aspm_configure_link_state(pdev,
 			policy_to_aspm_state(pdev));
-- 
cgit v1.2.3-59-g8ed1b


From 80bfdbe370d56a1448c7078cd6d286b89241a72e Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:12:43 +0900
Subject: PCI ASPM: cleanup aspm state field in struct pcie_link_state

The "support_state", "enabled_state" and "bios_aspm_state" fields in
the struct pcie_link_state take 2-bit value. So those fields don't
need to be defined as unsigned int. This patch makes those fields
2-bit, and cleans up some related code.

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 71 ++++++++++++++++++++++++++-----------------------
 1 file changed, 38 insertions(+), 33 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 419f1f3697c0..fdb3b7da7383 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -41,9 +41,10 @@ struct pcie_link_state {
 	struct list_head link;
 
 	/* ASPM state */
-	unsigned int support_state;
-	unsigned int enabled_state;
-	unsigned int bios_aspm_state;
+	u32 aspm_support:2;		/* Supported ASPM state */
+	u32 aspm_enabled:2;		/* Enabled ASPM state */
+	u32 aspm_default:2;		/* Default ASPM state by BIOS */
+
 	/* upstream component */
 	unsigned int l0s_upper_latency;
 	unsigned int l1_upper_latency;
@@ -88,9 +89,9 @@ static int policy_to_aspm_state(struct pci_dev *pdev)
 		return 0;
 	case POLICY_POWERSAVE:
 		/* Enable ASPM L0s/L1 */
-		return PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1;
+		return PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1;
 	case POLICY_DEFAULT:
-		return link_state->bios_aspm_state;
+		return link_state->aspm_default;
 	}
 	return 0;
 }
@@ -311,6 +312,7 @@ static void pcie_aspm_get_cap_device(struct pci_dev *pdev, u32 *state,
 	u32 reg32;
 	unsigned int latency;
 
+	*l0s = *l1 = *enabled = 0;
 	pos = pci_find_capability(pdev, PCI_CAP_ID_EXP);
 	pci_read_config_dword(pdev, pos + PCI_EXP_LNKCAP, &reg32);
 	*state = (reg32 & PCI_EXP_LNKCAP_ASPMS) >> 10;
@@ -333,26 +335,29 @@ static void pcie_aspm_get_cap_device(struct pci_dev *pdev, u32 *state,
 static void pcie_aspm_cap_init(struct pci_dev *pdev)
 {
 	struct pci_dev *child_dev;
-	u32 state, tmp;
+	u32 support, l0s, l1, enabled;
 	struct pcie_link_state *link_state = pdev->link_state;
 
 	/* upstream component states */
-	pcie_aspm_get_cap_device(pdev, &link_state->support_state,
-		&link_state->l0s_upper_latency,
-		&link_state->l1_upper_latency,
-		&link_state->enabled_state);
+	pcie_aspm_get_cap_device(pdev, &support, &l0s, &l1, &enabled);
+	link_state->aspm_support = support;
+	link_state->l0s_upper_latency = l0s;
+	link_state->l1_upper_latency = l1;
+	link_state->aspm_enabled = enabled;
+
 	/* downstream component states, all functions have the same setting */
 	child_dev = list_entry(pdev->subordinate->devices.next, struct pci_dev,
 		bus_list);
-	pcie_aspm_get_cap_device(child_dev, &state,
-		&link_state->l0s_down_latency,
-		&link_state->l1_down_latency,
-		&tmp);
-	link_state->support_state &= state;
-	if (!link_state->support_state)
+	pcie_aspm_get_cap_device(child_dev, &support, &l0s, &l1, &enabled);
+	link_state->aspm_support &= support;
+	link_state->l0s_down_latency = l0s;
+	link_state->l1_down_latency = l1;
+
+	if (!link_state->aspm_support)
 		return;
-	link_state->enabled_state &= link_state->support_state;
-	link_state->bios_aspm_state = link_state->enabled_state;
+
+	link_state->aspm_enabled &= link_state->aspm_support;
+	link_state->aspm_default = link_state->aspm_enabled;
 
 	/* ENDPOINT states*/
 	list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) {
@@ -371,7 +376,7 @@ static void pcie_aspm_cap_init(struct pci_dev *pdev)
 		latency = (reg32 & PCI_EXP_DEVCAP_L0S) >> 6;
 		latency = calc_L0S_latency(latency, 1);
 		ep_state->l0s_acceptable_latency = latency;
-		if (link_state->support_state & PCIE_LINK_STATE_L1) {
+		if (link_state->aspm_support & PCIE_LINK_STATE_L1) {
 			latency = (reg32 & PCI_EXP_DEVCAP_L1) >> 9;
 			latency = calc_L1_latency(latency, 1);
 			ep_state->l1_acceptable_latency = latency;
@@ -389,7 +394,7 @@ static unsigned int __pcie_aspm_check_state_one(struct pci_dev *pdev,
 
 	parent_dev = pdev->bus->self;
 	link_state = parent_dev->link_state;
-	state &= link_state->support_state;
+	state &= link_state->aspm_support;
 	if (state == 0)
 		return 0;
 	ep_state = &link_state->endpoints[PCI_FUNC(pdev->devfn)];
@@ -519,7 +524,7 @@ static void __pcie_aspm_config_link(struct pci_dev *pdev, unsigned int state)
 	if (!(state & PCIE_LINK_STATE_L1))
 		__pcie_aspm_config_one_dev(pdev, state);
 
-	link_state->enabled_state = state;
+	link_state->aspm_enabled = state;
 }
 
 static struct pcie_link_state *get_root_port_link(struct pcie_link_state *link)
@@ -550,7 +555,7 @@ static void __pcie_aspm_configure_link_state(struct pci_dev *pdev,
 	/* check root port link too in case it hasn't children */
 	state = pcie_aspm_check_state(root_port_link->pdev, state);
 
-	if (link_state->enabled_state == state)
+	if (link_state->aspm_enabled == state)
 		return;
 
 	/*
@@ -675,10 +680,11 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
 		pcie_aspm_configure_common_clock(pdev);
 		pcie_aspm_cap_init(pdev);
 	} else {
-		link_state->enabled_state = PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1;
-		link_state->bios_aspm_state = 0;
+		link_state->aspm_enabled =
+			(PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1);
+		link_state->aspm_default = 0;
 		/* Set support state to 0, so we will disable ASPM later */
-		link_state->support_state = 0;
+		link_state->aspm_support = 0;
 	}
 
 	link_state->pdev = pdev;
@@ -690,7 +696,7 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
 		 * initialization will config the whole hierarchy. but we must
 		 * make sure BIOS doesn't set unsupported link state
 		 **/
-		state = pcie_aspm_check_state(pdev, link_state->bios_aspm_state);
+		state = pcie_aspm_check_state(pdev, link_state->aspm_default);
 		__pcie_aspm_config_link(pdev, state);
 	} else
 		__pcie_aspm_configure_link_state(pdev,
@@ -753,7 +759,7 @@ void pcie_aspm_pm_state_change(struct pci_dev *pdev)
 	 * devices changed PM state, we should recheck if latency meets all
 	 * functions' requirement
 	 */
-	pcie_aspm_configure_link_state(pdev, link_state->enabled_state);
+	pcie_aspm_configure_link_state(pdev, link_state->aspm_enabled);
 }
 
 /*
@@ -776,12 +782,11 @@ void pci_disable_link_state(struct pci_dev *pdev, int state)
 	down_read(&pci_bus_sem);
 	mutex_lock(&aspm_lock);
 	link_state = parent->link_state;
-	link_state->support_state &=
-		~(state & (PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1));
+	link_state->aspm_support &= ~state;
 	if (state & PCIE_LINK_STATE_CLKPM)
 		link_state->clk_pm_capable = 0;
 
-	__pcie_aspm_configure_link_state(parent, link_state->enabled_state);
+	__pcie_aspm_configure_link_state(parent, link_state->aspm_enabled);
 	if (!link_state->clk_pm_capable && link_state->clk_pm_enabled)
 		pcie_set_clock_pm(parent, 0);
 	mutex_unlock(&aspm_lock);
@@ -842,7 +847,7 @@ static ssize_t link_state_show(struct device *dev,
 	struct pci_dev *pci_device = to_pci_dev(dev);
 	struct pcie_link_state *link_state = pci_device->link_state;
 
-	return sprintf(buf, "%d\n", link_state->enabled_state);
+	return sprintf(buf, "%d\n", link_state->aspm_enabled);
 }
 
 static ssize_t link_state_store(struct device *dev,
@@ -908,7 +913,7 @@ void pcie_aspm_create_sysfs_dev_files(struct pci_dev *pdev)
 		pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM) || !link_state)
 		return;
 
-	if (link_state->support_state)
+	if (link_state->aspm_support)
 		sysfs_add_file_to_group(&pdev->dev.kobj,
 			&dev_attr_link_state.attr, power_group);
 	if (link_state->clk_pm_capable)
@@ -924,7 +929,7 @@ void pcie_aspm_remove_sysfs_dev_files(struct pci_dev *pdev)
 		pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM) || !link_state)
 		return;
 
-	if (link_state->support_state)
+	if (link_state->aspm_support)
 		sysfs_remove_file_from_group(&pdev->dev.kobj,
 			&dev_attr_link_state.attr, power_group);
 	if (link_state->clk_pm_capable)
-- 
cgit v1.2.3-59-g8ed1b


From b6c2e54d3ea27719b920faf15db92dfe0260f0d2 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:14:58 +0900
Subject: PCI ASPM: cleanup latency field in struct pcie_link_state

Clean up latency related data structures for ASPM.

- Introduce struct acpi_latency for exit latency and acceptable
  latency management. With this change, struct endpoint_state is no
  longer needed.

- We don't need to hold both upstream latency and downstream latency
  in the current implementation.

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 66 +++++++++++++++++++++----------------------------
 1 file changed, 28 insertions(+), 38 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index fdb3b7da7383..88351c242a48 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -26,9 +26,9 @@
 #endif
 #define MODULE_PARAM_PREFIX "pcie_aspm."
 
-struct endpoint_state {
-	unsigned int l0s_acceptable_latency;
-	unsigned int l1_acceptable_latency;
+struct aspm_latency {
+	u32 l0s;			/* L0s latency (nsec) */
+	u32 l1;				/* L1 latency (nsec) */
 };
 
 struct pcie_link_state {
@@ -45,22 +45,19 @@ struct pcie_link_state {
 	u32 aspm_enabled:2;		/* Enabled ASPM state */
 	u32 aspm_default:2;		/* Default ASPM state by BIOS */
 
-	/* upstream component */
-	unsigned int l0s_upper_latency;
-	unsigned int l1_upper_latency;
-	/* downstream component */
-	unsigned int l0s_down_latency;
-	unsigned int l1_down_latency;
+	/* Latencies */
+	struct aspm_latency latency;	/* Exit latency */
+
 	/* Clock PM state*/
 	unsigned int clk_pm_capable;
 	unsigned int clk_pm_enabled;
 	unsigned int bios_clk_state;
 
 	/*
-	 * A pcie downstream port only has one slot under it, so at most there
-	 * are 8 functions
+	 * Endpoint acceptable latencies. A pcie downstream port only
+	 * has one slot under it, so at most there are 8 functions.
 	 */
-	struct endpoint_state endpoints[8];
+	struct aspm_latency acceptable[8];
 };
 
 static int aspm_disabled, aspm_force;
@@ -341,8 +338,8 @@ static void pcie_aspm_cap_init(struct pci_dev *pdev)
 	/* upstream component states */
 	pcie_aspm_get_cap_device(pdev, &support, &l0s, &l1, &enabled);
 	link_state->aspm_support = support;
-	link_state->l0s_upper_latency = l0s;
-	link_state->l1_upper_latency = l1;
+	link_state->latency.l0s = l0s;
+	link_state->latency.l1 = l1;
 	link_state->aspm_enabled = enabled;
 
 	/* downstream component states, all functions have the same setting */
@@ -350,8 +347,8 @@ static void pcie_aspm_cap_init(struct pci_dev *pdev)
 		bus_list);
 	pcie_aspm_get_cap_device(child_dev, &support, &l0s, &l1, &enabled);
 	link_state->aspm_support &= support;
-	link_state->l0s_down_latency = l0s;
-	link_state->l1_down_latency = l1;
+	link_state->latency.l0s = max_t(u32, link_state->latency.l0s, l0s);
+	link_state->latency.l1 = max_t(u32, link_state->latency.l1, l1);
 
 	if (!link_state->aspm_support)
 		return;
@@ -364,8 +361,8 @@ static void pcie_aspm_cap_init(struct pci_dev *pdev)
 		int pos;
 		u32 reg32;
 		unsigned int latency;
-		struct endpoint_state *ep_state =
-			&link_state->endpoints[PCI_FUNC(child_dev->devfn)];
+		struct aspm_latency *acceptable =
+			&link_state->acceptable[PCI_FUNC(child_dev->devfn)];
 
 		if (child_dev->pcie_type != PCI_EXP_TYPE_ENDPOINT &&
 			child_dev->pcie_type != PCI_EXP_TYPE_LEG_END)
@@ -375,11 +372,11 @@ static void pcie_aspm_cap_init(struct pci_dev *pdev)
 		pci_read_config_dword(child_dev, pos + PCI_EXP_DEVCAP, &reg32);
 		latency = (reg32 & PCI_EXP_DEVCAP_L0S) >> 6;
 		latency = calc_L0S_latency(latency, 1);
-		ep_state->l0s_acceptable_latency = latency;
+		acceptable->l0s = latency;
 		if (link_state->aspm_support & PCIE_LINK_STATE_L1) {
 			latency = (reg32 & PCI_EXP_DEVCAP_L1) >> 9;
 			latency = calc_L1_latency(latency, 1);
-			ep_state->l1_acceptable_latency = latency;
+			acceptable->l1 = latency;
 		}
 	}
 }
@@ -388,16 +385,16 @@ static unsigned int __pcie_aspm_check_state_one(struct pci_dev *pdev,
 	unsigned int state)
 {
 	struct pci_dev *parent_dev, *tmp_dev;
-	unsigned int latency, l1_latency = 0;
+	unsigned int l1_latency = 0;
 	struct pcie_link_state *link_state;
-	struct endpoint_state *ep_state;
+	struct aspm_latency *acceptable;
 
 	parent_dev = pdev->bus->self;
 	link_state = parent_dev->link_state;
 	state &= link_state->aspm_support;
 	if (state == 0)
 		return 0;
-	ep_state = &link_state->endpoints[PCI_FUNC(pdev->devfn)];
+	acceptable = &link_state->acceptable[PCI_FUNC(pdev->devfn)];
 
 	/*
 	 * Check latency for endpoint device.
@@ -411,21 +408,14 @@ static unsigned int __pcie_aspm_check_state_one(struct pci_dev *pdev,
 	while (state & (PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1)) {
 		parent_dev = tmp_dev->bus->self;
 		link_state = parent_dev->link_state;
-		if (state & PCIE_LINK_STATE_L0S) {
-			latency = max_t(unsigned int,
-					link_state->l0s_upper_latency,
-					link_state->l0s_down_latency);
-			if (latency > ep_state->l0s_acceptable_latency)
-				state &= ~PCIE_LINK_STATE_L0S;
-		}
-		if (state & PCIE_LINK_STATE_L1) {
-			latency = max_t(unsigned int,
-					link_state->l1_upper_latency,
-					link_state->l1_down_latency);
-			if (latency + l1_latency >
-					ep_state->l1_acceptable_latency)
-				state &= ~PCIE_LINK_STATE_L1;
-		}
+		if ((state & PCIE_LINK_STATE_L0S) &&
+		    (link_state->latency.l0s > acceptable->l0s))
+			state &= ~PCIE_LINK_STATE_L0S;
+
+		if ((state & PCIE_LINK_STATE_L1) &&
+		    (link_state->latency.l1 + l1_latency > acceptable->l1))
+			state &= ~PCIE_LINK_STATE_L1;
+
 		if (!parent_dev->bus->self) /* parent_dev is a root port */
 			break;
 		else {
-- 
cgit v1.2.3-59-g8ed1b


From 4d246e458918d469ad645fd5f937ac22982e0466 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:15:38 +0900
Subject: PCI ASPM: cleanup clkpm state in struct pcie_link_state

The "clk_pm_capable", "clk_pm_enable" and "bios_clk_state" fields in
the struct pcie_link_state only take 1-bit value. So those fields
don't need to be defined as unsigned int. This patch makes those
fields 1-bit, and cleans up some related code.

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 37 ++++++++++++++++++-------------------
 1 file changed, 18 insertions(+), 19 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 88351c242a48..2d7e695b7589 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -45,14 +45,13 @@ struct pcie_link_state {
 	u32 aspm_enabled:2;		/* Enabled ASPM state */
 	u32 aspm_default:2;		/* Default ASPM state by BIOS */
 
+	/* Clock PM state */
+	u32 clkpm_capable:1;		/* Clock PM capable? */
+	u32 clkpm_enabled:1;		/* Current Clock PM state */
+	u32 clkpm_default:1;		/* Default Clock PM state by BIOS */
+
 	/* Latencies */
 	struct aspm_latency latency;	/* Exit latency */
-
-	/* Clock PM state*/
-	unsigned int clk_pm_capable;
-	unsigned int clk_pm_enabled;
-	unsigned int bios_clk_state;
-
 	/*
 	 * Endpoint acceptable latencies. A pcie downstream port only
 	 * has one slot under it, so at most there are 8 functions.
@@ -105,7 +104,7 @@ static int policy_to_clkpm_state(struct pci_dev *pdev)
 		/* Disable Clock PM */
 		return 1;
 	case POLICY_DEFAULT:
-		return link_state->bios_clk_state;
+		return link_state->clkpm_default;
 	}
 	return 0;
 }
@@ -128,7 +127,7 @@ static void pcie_set_clock_pm(struct pci_dev *pdev, int enable)
 			reg16 &= ~PCI_EXP_LNKCTL_CLKREQ_EN;
 		pci_write_config_word(child_dev, pos + PCI_EXP_LNKCTL, reg16);
 	}
-	link_state->clk_pm_enabled = !!enable;
+	link_state->clkpm_enabled = !!enable;
 }
 
 static void pcie_check_clock_pm(struct pci_dev *pdev, int blacklist)
@@ -155,13 +154,13 @@ static void pcie_check_clock_pm(struct pci_dev *pdev, int blacklist)
 		if (!(reg16 & PCI_EXP_LNKCTL_CLKREQ_EN))
 			enabled = 0;
 	}
-	link_state->clk_pm_enabled = enabled;
-	link_state->bios_clk_state = enabled;
+	link_state->clkpm_enabled = enabled;
+	link_state->clkpm_default = enabled;
 	if (!blacklist) {
-		link_state->clk_pm_capable = capable;
+		link_state->clkpm_capable = capable;
 		pcie_set_clock_pm(pdev, policy_to_clkpm_state(pdev));
 	} else {
-		link_state->clk_pm_capable = 0;
+		link_state->clkpm_capable = 0;
 		pcie_set_clock_pm(pdev, 0);
 	}
 }
@@ -774,10 +773,10 @@ void pci_disable_link_state(struct pci_dev *pdev, int state)
 	link_state = parent->link_state;
 	link_state->aspm_support &= ~state;
 	if (state & PCIE_LINK_STATE_CLKPM)
-		link_state->clk_pm_capable = 0;
+		link_state->clkpm_capable = 0;
 
 	__pcie_aspm_configure_link_state(parent, link_state->aspm_enabled);
-	if (!link_state->clk_pm_capable && link_state->clk_pm_enabled)
+	if (!link_state->clkpm_capable && link_state->clkpm_enabled)
 		pcie_set_clock_pm(parent, 0);
 	mutex_unlock(&aspm_lock);
 	up_read(&pci_bus_sem);
@@ -805,8 +804,8 @@ static int pcie_aspm_set_policy(const char *val, struct kernel_param *kp)
 		pdev = link_state->pdev;
 		__pcie_aspm_configure_link_state(pdev,
 			policy_to_aspm_state(pdev));
-		if (link_state->clk_pm_capable &&
-		    link_state->clk_pm_enabled != policy_to_clkpm_state(pdev))
+		if (link_state->clkpm_capable &&
+		    link_state->clkpm_enabled != policy_to_clkpm_state(pdev))
 			pcie_set_clock_pm(pdev, policy_to_clkpm_state(pdev));
 
 	}
@@ -867,7 +866,7 @@ static ssize_t clk_ctl_show(struct device *dev,
 	struct pci_dev *pci_device = to_pci_dev(dev);
 	struct pcie_link_state *link_state = pci_device->link_state;
 
-	return sprintf(buf, "%d\n", link_state->clk_pm_enabled);
+	return sprintf(buf, "%d\n", link_state->clkpm_enabled);
 }
 
 static ssize_t clk_ctl_store(struct device *dev,
@@ -906,7 +905,7 @@ void pcie_aspm_create_sysfs_dev_files(struct pci_dev *pdev)
 	if (link_state->aspm_support)
 		sysfs_add_file_to_group(&pdev->dev.kobj,
 			&dev_attr_link_state.attr, power_group);
-	if (link_state->clk_pm_capable)
+	if (link_state->clkpm_capable)
 		sysfs_add_file_to_group(&pdev->dev.kobj,
 			&dev_attr_clk_ctl.attr, power_group);
 }
@@ -922,7 +921,7 @@ void pcie_aspm_remove_sysfs_dev_files(struct pci_dev *pdev)
 	if (link_state->aspm_support)
 		sysfs_remove_file_from_group(&pdev->dev.kobj,
 			&dev_attr_link_state.attr, power_group);
-	if (link_state->clk_pm_capable)
+	if (link_state->clkpm_capable)
 		sysfs_remove_file_from_group(&pdev->dev.kobj,
 			&dev_attr_clk_ctl.attr, power_group);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 5cde89d80172a393e49077d2450545b97ac8d972 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:17:04 +0900
Subject: PCI ASPM: cleanup misc in struct pcie_link_state

Cleanup for some fields in pcie_link_state.

- Add comments.
- make "downstream_has_switch" field 1-bit.

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 2d7e695b7589..a1ae9b6f3995 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -32,13 +32,11 @@ struct aspm_latency {
 };
 
 struct pcie_link_state {
-	struct list_head sibling;
-	struct pci_dev *pdev;
-	bool downstream_has_switch;
-
-	struct pcie_link_state *parent;
-	struct list_head children;
-	struct list_head link;
+	struct pci_dev *pdev;		/* Upstream component of the Link */
+	struct pcie_link_state *parent;	/* pointer to the parent Link state */
+	struct list_head sibling;	/* node in link_list */
+	struct list_head children;	/* list of child link states */
+	struct list_head link;		/* node in parent's children list */
 
 	/* ASPM state */
 	u32 aspm_support:2;		/* Supported ASPM state */
@@ -50,6 +48,8 @@ struct pcie_link_state {
 	u32 clkpm_enabled:1;		/* Current Clock PM state */
 	u32 clkpm_default:1;		/* Default Clock PM state by BIOS */
 
+	u32 has_switch:1;		/* Downstream has switches? */
+
 	/* Latencies */
 	struct aspm_latency latency;	/* Exit latency */
 	/*
@@ -648,7 +648,7 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
 	if (!link_state)
 		goto unlock_out;
 
-	link_state->downstream_has_switch = pcie_aspm_downstream_has_switch(pdev);
+	link_state->has_switch = pcie_aspm_downstream_has_switch(pdev);
 	INIT_LIST_HEAD(&link_state->children);
 	INIT_LIST_HEAD(&link_state->link);
 	if (pdev->bus->self) {/* this is a switch */
@@ -679,7 +679,7 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
 	link_state->pdev = pdev;
 	list_add(&link_state->sibling, &link_list);
 
-	if (link_state->downstream_has_switch) {
+	if (link_state->has_switch) {
 		/*
 		 * If link has switch, delay the link config. The leaf link
 		 * initialization will config the whole hierarchy. but we must
-- 
cgit v1.2.3-59-g8ed1b


From 5aa63583cbec27482c6f1d761a0509f59b7969a8 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:17:44 +0900
Subject: PCI ASPM: cleanup change input argument of aspm functions

In the current ASPM implementation, there are many functions that
take a pointer to struct pci_dev corresponding to the upstream component
of the link as a parameter. But, since those functions handle PCI
express link state, a pointer to struct pcie_link_state is more
suitable than a pointer to struct pci_dev. Changing a parameter to a
pointer to struct pcie_link_state makes ASPM code much simpler and
easier to read. This patch also contains some minor cleanups. This patch
doesn't have any functional change.

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 372 ++++++++++++++++++++++--------------------------
 1 file changed, 173 insertions(+), 199 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index a1ae9b6f3995..9eaaf95f65a2 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -75,10 +75,8 @@ static const char *policy_str[] = {
 
 #define LINK_RETRAIN_TIMEOUT HZ
 
-static int policy_to_aspm_state(struct pci_dev *pdev)
+static int policy_to_aspm_state(struct pcie_link_state *link)
 {
-	struct pcie_link_state *link_state = pdev->link_state;
-
 	switch (aspm_policy) {
 	case POLICY_PERFORMANCE:
 		/* Disable ASPM and Clock PM */
@@ -87,15 +85,13 @@ static int policy_to_aspm_state(struct pci_dev *pdev)
 		/* Enable ASPM L0s/L1 */
 		return PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1;
 	case POLICY_DEFAULT:
-		return link_state->aspm_default;
+		return link->aspm_default;
 	}
 	return 0;
 }
 
-static int policy_to_clkpm_state(struct pci_dev *pdev)
+static int policy_to_clkpm_state(struct pcie_link_state *link)
 {
-	struct pcie_link_state *link_state = pdev->link_state;
-
 	switch (aspm_policy) {
 	case POLICY_PERFORMANCE:
 		/* Disable ASPM and Clock PM */
@@ -104,73 +100,73 @@ static int policy_to_clkpm_state(struct pci_dev *pdev)
 		/* Disable Clock PM */
 		return 1;
 	case POLICY_DEFAULT:
-		return link_state->clkpm_default;
+		return link->clkpm_default;
 	}
 	return 0;
 }
 
-static void pcie_set_clock_pm(struct pci_dev *pdev, int enable)
+static void pcie_set_clock_pm(struct pcie_link_state *link, int enable)
 {
-	struct pci_dev *child_dev;
 	int pos;
 	u16 reg16;
-	struct pcie_link_state *link_state = pdev->link_state;
+	struct pci_dev *child;
+	struct pci_bus *linkbus = link->pdev->subordinate;
 
-	list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) {
-		pos = pci_find_capability(child_dev, PCI_CAP_ID_EXP);
+	list_for_each_entry(child, &linkbus->devices, bus_list) {
+		pos = pci_find_capability(child, PCI_CAP_ID_EXP);
 		if (!pos)
 			return;
-		pci_read_config_word(child_dev, pos + PCI_EXP_LNKCTL, &reg16);
+		pci_read_config_word(child, pos + PCI_EXP_LNKCTL, &reg16);
 		if (enable)
 			reg16 |= PCI_EXP_LNKCTL_CLKREQ_EN;
 		else
 			reg16 &= ~PCI_EXP_LNKCTL_CLKREQ_EN;
-		pci_write_config_word(child_dev, pos + PCI_EXP_LNKCTL, reg16);
+		pci_write_config_word(child, pos + PCI_EXP_LNKCTL, reg16);
 	}
-	link_state->clkpm_enabled = !!enable;
+	link->clkpm_enabled = !!enable;
 }
 
-static void pcie_check_clock_pm(struct pci_dev *pdev, int blacklist)
+static void pcie_check_clock_pm(struct pcie_link_state *link, int blacklist)
 {
-	int pos;
+	int pos, capable = 1, enabled = 1;
 	u32 reg32;
 	u16 reg16;
-	int capable = 1, enabled = 1;
-	struct pci_dev *child_dev;
-	struct pcie_link_state *link_state = pdev->link_state;
+	struct pci_dev *child;
+	struct pci_bus *linkbus = link->pdev->subordinate;
 
 	/* All functions should have the same cap and state, take the worst */
-	list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) {
-		pos = pci_find_capability(child_dev, PCI_CAP_ID_EXP);
+	list_for_each_entry(child, &linkbus->devices, bus_list) {
+		pos = pci_find_capability(child, PCI_CAP_ID_EXP);
 		if (!pos)
 			return;
-		pci_read_config_dword(child_dev, pos + PCI_EXP_LNKCAP, &reg32);
+		pci_read_config_dword(child, pos + PCI_EXP_LNKCAP, &reg32);
 		if (!(reg32 & PCI_EXP_LNKCAP_CLKPM)) {
 			capable = 0;
 			enabled = 0;
 			break;
 		}
-		pci_read_config_word(child_dev, pos + PCI_EXP_LNKCTL, &reg16);
+		pci_read_config_word(child, pos + PCI_EXP_LNKCTL, &reg16);
 		if (!(reg16 & PCI_EXP_LNKCTL_CLKREQ_EN))
 			enabled = 0;
 	}
-	link_state->clkpm_enabled = enabled;
-	link_state->clkpm_default = enabled;
+	link->clkpm_enabled = enabled;
+	link->clkpm_default = enabled;
 	if (!blacklist) {
-		link_state->clkpm_capable = capable;
-		pcie_set_clock_pm(pdev, policy_to_clkpm_state(pdev));
+		link->clkpm_capable = capable;
+		pcie_set_clock_pm(link, policy_to_clkpm_state(link));
 	} else {
-		link_state->clkpm_capable = 0;
-		pcie_set_clock_pm(pdev, 0);
+		link->clkpm_capable = 0;
+		pcie_set_clock_pm(link, 0);
 	}
 }
 
-static bool pcie_aspm_downstream_has_switch(struct pci_dev *pdev)
+static bool pcie_aspm_downstream_has_switch(struct pcie_link_state *link)
 {
-	struct pci_dev *child_dev;
+	struct pci_dev *child;
+	struct pci_bus *linkbus = link->pdev->subordinate;
 
-	list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) {
-		if (child_dev->pcie_type == PCI_EXP_TYPE_UPSTREAM)
+	list_for_each_entry(child, &linkbus->devices, bus_list) {
+		if (child->pcie_type == PCI_EXP_TYPE_UPSTREAM)
 			return true;
 	}
 	return false;
@@ -181,89 +177,79 @@ static bool pcie_aspm_downstream_has_switch(struct pci_dev *pdev)
  *   could use common clock. If they are, configure them to use the
  *   common clock. That will reduce the ASPM state exit latency.
  */
-static void pcie_aspm_configure_common_clock(struct pci_dev *pdev)
+static void pcie_aspm_configure_common_clock(struct pcie_link_state *link)
 {
-	int pos, child_pos, i = 0;
-	u16 reg16 = 0;
-	struct pci_dev *child_dev;
-	int same_clock = 1;
+	int ppos, cpos, same_clock = 1;
+	u16 reg16, parent_reg, child_reg[8];
 	unsigned long start_jiffies;
-	u16 child_regs[8], parent_reg;
+	struct pci_dev *child, *parent = link->pdev;
+	struct pci_bus *linkbus = parent->subordinate;
 	/*
-	 * all functions of a slot should have the same Slot Clock
+	 * All functions of a slot should have the same Slot Clock
 	 * Configuration, so just check one function
-	 * */
-	child_dev = list_entry(pdev->subordinate->devices.next, struct pci_dev,
-		bus_list);
-	BUG_ON(!child_dev->is_pcie);
+	 */
+	child = list_entry(linkbus->devices.next, struct pci_dev, bus_list);
+	BUG_ON(!child->is_pcie);
 
 	/* Check downstream component if bit Slot Clock Configuration is 1 */
-	child_pos = pci_find_capability(child_dev, PCI_CAP_ID_EXP);
-	pci_read_config_word(child_dev, child_pos + PCI_EXP_LNKSTA, &reg16);
+	cpos = pci_find_capability(child, PCI_CAP_ID_EXP);
+	pci_read_config_word(child, cpos + PCI_EXP_LNKSTA, &reg16);
 	if (!(reg16 & PCI_EXP_LNKSTA_SLC))
 		same_clock = 0;
 
 	/* Check upstream component if bit Slot Clock Configuration is 1 */
-	pos = pci_find_capability(pdev, PCI_CAP_ID_EXP);
-	pci_read_config_word(pdev, pos + PCI_EXP_LNKSTA, &reg16);
+	ppos = pci_find_capability(parent, PCI_CAP_ID_EXP);
+	pci_read_config_word(parent, ppos + PCI_EXP_LNKSTA, &reg16);
 	if (!(reg16 & PCI_EXP_LNKSTA_SLC))
 		same_clock = 0;
 
 	/* Configure downstream component, all functions */
-	list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) {
-		child_pos = pci_find_capability(child_dev, PCI_CAP_ID_EXP);
-		pci_read_config_word(child_dev, child_pos + PCI_EXP_LNKCTL,
-			&reg16);
-		child_regs[i] = reg16;
+	list_for_each_entry(child, &linkbus->devices, bus_list) {
+		cpos = pci_find_capability(child, PCI_CAP_ID_EXP);
+		pci_read_config_word(child, cpos + PCI_EXP_LNKCTL, &reg16);
+		child_reg[PCI_FUNC(child->devfn)] = reg16;
 		if (same_clock)
 			reg16 |= PCI_EXP_LNKCTL_CCC;
 		else
 			reg16 &= ~PCI_EXP_LNKCTL_CCC;
-		pci_write_config_word(child_dev, child_pos + PCI_EXP_LNKCTL,
-			reg16);
-		i++;
+		pci_write_config_word(child, cpos + PCI_EXP_LNKCTL, reg16);
 	}
 
 	/* Configure upstream component */
-	pci_read_config_word(pdev, pos + PCI_EXP_LNKCTL, &reg16);
+	pci_read_config_word(parent, ppos + PCI_EXP_LNKCTL, &reg16);
 	parent_reg = reg16;
 	if (same_clock)
 		reg16 |= PCI_EXP_LNKCTL_CCC;
 	else
 		reg16 &= ~PCI_EXP_LNKCTL_CCC;
-	pci_write_config_word(pdev, pos + PCI_EXP_LNKCTL, reg16);
+	pci_write_config_word(parent, ppos + PCI_EXP_LNKCTL, reg16);
 
-	/* retrain link */
+	/* Retrain link */
 	reg16 |= PCI_EXP_LNKCTL_RL;
-	pci_write_config_word(pdev, pos + PCI_EXP_LNKCTL, reg16);
+	pci_write_config_word(parent, ppos + PCI_EXP_LNKCTL, reg16);
 
-	/* Wait for link training end */
-	/* break out after waiting for timeout */
+	/* Wait for link training end. Break out after waiting for timeout */
 	start_jiffies = jiffies;
 	for (;;) {
-		pci_read_config_word(pdev, pos + PCI_EXP_LNKSTA, &reg16);
+		pci_read_config_word(parent, ppos + PCI_EXP_LNKSTA, &reg16);
 		if (!(reg16 & PCI_EXP_LNKSTA_LT))
 			break;
 		if (time_after(jiffies, start_jiffies + LINK_RETRAIN_TIMEOUT))
 			break;
 		msleep(1);
 	}
-	/* training failed -> recover */
-	if (reg16 & PCI_EXP_LNKSTA_LT) {
-		dev_printk (KERN_ERR, &pdev->dev, "ASPM: Could not configure"
-			    " common clock\n");
-		i = 0;
-		list_for_each_entry(child_dev, &pdev->subordinate->devices,
-				    bus_list) {
-			child_pos = pci_find_capability(child_dev,
-							PCI_CAP_ID_EXP);
-			pci_write_config_word(child_dev,
-					      child_pos + PCI_EXP_LNKCTL,
-					      child_regs[i]);
-			i++;
-		}
-		pci_write_config_word(pdev, pos + PCI_EXP_LNKCTL, parent_reg);
+	if (!(reg16 & PCI_EXP_LNKSTA_LT))
+		return;
+
+	/* Training failed. Restore common clock configurations */
+	dev_printk(KERN_ERR, &parent->dev,
+		   "ASPM: Could not configure common clock\n");
+	list_for_each_entry(child, &linkbus->devices, bus_list) {
+		cpos = pci_find_capability(child, PCI_CAP_ID_EXP);
+		pci_write_config_word(child, cpos + PCI_EXP_LNKCTL,
+				      child_reg[PCI_FUNC(child->devfn)]);
 	}
+	pci_write_config_word(parent, ppos + PCI_EXP_LNKCTL, parent_reg);
 }
 
 /*
@@ -328,51 +314,50 @@ static void pcie_aspm_get_cap_device(struct pci_dev *pdev, u32 *state,
 	*enabled = reg16 & (PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1);
 }
 
-static void pcie_aspm_cap_init(struct pci_dev *pdev)
+static void pcie_aspm_cap_init(struct pcie_link_state *link)
 {
-	struct pci_dev *child_dev;
 	u32 support, l0s, l1, enabled;
-	struct pcie_link_state *link_state = pdev->link_state;
+	struct pci_dev *child, *parent = link->pdev;
+	struct pci_bus *linkbus = parent->subordinate;
 
 	/* upstream component states */
-	pcie_aspm_get_cap_device(pdev, &support, &l0s, &l1, &enabled);
-	link_state->aspm_support = support;
-	link_state->latency.l0s = l0s;
-	link_state->latency.l1 = l1;
-	link_state->aspm_enabled = enabled;
+	pcie_aspm_get_cap_device(parent, &support, &l0s, &l1, &enabled);
+	link->aspm_support = support;
+	link->latency.l0s = l0s;
+	link->latency.l1 = l1;
+	link->aspm_enabled = enabled;
 
 	/* downstream component states, all functions have the same setting */
-	child_dev = list_entry(pdev->subordinate->devices.next, struct pci_dev,
-		bus_list);
-	pcie_aspm_get_cap_device(child_dev, &support, &l0s, &l1, &enabled);
-	link_state->aspm_support &= support;
-	link_state->latency.l0s = max_t(u32, link_state->latency.l0s, l0s);
-	link_state->latency.l1 = max_t(u32, link_state->latency.l1, l1);
-
-	if (!link_state->aspm_support)
+	child = list_entry(linkbus->devices.next, struct pci_dev, bus_list);
+	pcie_aspm_get_cap_device(child, &support, &l0s, &l1, &enabled);
+	link->aspm_support &= support;
+	link->latency.l0s = max_t(u32, link->latency.l0s, l0s);
+	link->latency.l1 = max_t(u32, link->latency.l1, l1);
+
+	if (!link->aspm_support)
 		return;
 
-	link_state->aspm_enabled &= link_state->aspm_support;
-	link_state->aspm_default = link_state->aspm_enabled;
+	link->aspm_enabled &= link->aspm_support;
+	link->aspm_default = link->aspm_enabled;
 
 	/* ENDPOINT states*/
-	list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) {
+	list_for_each_entry(child, &linkbus->devices, bus_list) {
 		int pos;
 		u32 reg32;
 		unsigned int latency;
 		struct aspm_latency *acceptable =
-			&link_state->acceptable[PCI_FUNC(child_dev->devfn)];
+			&link->acceptable[PCI_FUNC(child->devfn)];
 
-		if (child_dev->pcie_type != PCI_EXP_TYPE_ENDPOINT &&
-			child_dev->pcie_type != PCI_EXP_TYPE_LEG_END)
+		if (child->pcie_type != PCI_EXP_TYPE_ENDPOINT &&
+		    child->pcie_type != PCI_EXP_TYPE_LEG_END)
 			continue;
 
-		pos = pci_find_capability(child_dev, PCI_CAP_ID_EXP);
-		pci_read_config_dword(child_dev, pos + PCI_EXP_DEVCAP, &reg32);
+		pos = pci_find_capability(child, PCI_CAP_ID_EXP);
+		pci_read_config_dword(child, pos + PCI_EXP_DEVCAP, &reg32);
 		latency = (reg32 & PCI_EXP_DEVCAP_L0S) >> 6;
 		latency = calc_L0S_latency(latency, 1);
 		acceptable->l0s = latency;
-		if (link_state->aspm_support & PCIE_LINK_STATE_L1) {
+		if (link->aspm_support & PCIE_LINK_STATE_L1) {
 			latency = (reg32 & PCI_EXP_DEVCAP_L1) >> 9;
 			latency = calc_L1_latency(latency, 1);
 			acceptable->l1 = latency;
@@ -434,33 +419,33 @@ static unsigned int __pcie_aspm_check_state_one(struct pci_dev *pdev,
 	return state;
 }
 
-static unsigned int pcie_aspm_check_state(struct pci_dev *pdev,
-	unsigned int state)
+static u32 pcie_aspm_check_state(struct pcie_link_state *link, u32 state)
 {
-	struct pci_dev *child_dev;
+	pci_power_t power_state;
+	struct pci_dev *child;
+	struct pci_bus *linkbus = link->pdev->subordinate;
 
 	/* If no child, ignore the link */
-	if (list_empty(&pdev->subordinate->devices))
+	if (list_empty(&linkbus->devices))
 		return state;
-	list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) {
-		if (child_dev->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE) {
-			/*
-			 * If downstream component of a link is pci bridge, we
-			 * disable ASPM for now for the link
-			 * */
-			state = 0;
-			break;
-		}
-		if ((child_dev->pcie_type != PCI_EXP_TYPE_ENDPOINT &&
-			child_dev->pcie_type != PCI_EXP_TYPE_LEG_END))
+
+	list_for_each_entry(child, &linkbus->devices, bus_list) {
+		/*
+		 * If downstream component of a link is pci bridge, we
+		 * disable ASPM for now for the link
+		 */
+		if (child->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE)
+			return 0;
+
+		if ((child->pcie_type != PCI_EXP_TYPE_ENDPOINT &&
+		     child->pcie_type != PCI_EXP_TYPE_LEG_END))
 			continue;
 		/* Device not in D0 doesn't need check latency */
-		if (child_dev->current_state == PCI_D1 ||
-			child_dev->current_state == PCI_D2 ||
-			child_dev->current_state == PCI_D3hot ||
-			child_dev->current_state == PCI_D3cold)
+		power_state = child->current_state;
+		if (power_state == PCI_D1 || power_state == PCI_D2 ||
+		    power_state == PCI_D3hot || power_state == PCI_D3cold)
 			continue;
-		state = __pcie_aspm_check_state_one(child_dev, state);
+		state = __pcie_aspm_check_state_one(child, state);
 	}
 	return state;
 }
@@ -476,44 +461,38 @@ static void __pcie_aspm_config_one_dev(struct pci_dev *pdev, unsigned int state)
 	pci_write_config_word(pdev, pos + PCI_EXP_LNKCTL, reg16);
 }
 
-static void __pcie_aspm_config_link(struct pci_dev *pdev, unsigned int state)
+static void __pcie_aspm_config_link(struct pcie_link_state *link, u32 state)
 {
-	struct pci_dev *child_dev;
-	int valid = 1;
-	struct pcie_link_state *link_state = pdev->link_state;
+	struct pci_dev *child, *parent = link->pdev;
+	struct pci_bus *linkbus = parent->subordinate;
 
 	/* If no child, disable the link */
-	if (list_empty(&pdev->subordinate->devices))
+	if (list_empty(&linkbus->devices))
 		state = 0;
 	/*
-	 * if the downstream component has pci bridge function, don't do ASPM
-	 * now
+	 * If the downstream component has pci bridge function, don't
+	 * do ASPM now.
 	 */
-	list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) {
-		if (child_dev->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE) {
-			valid = 0;
-			break;
-		}
+	list_for_each_entry(child, &linkbus->devices, bus_list) {
+		if (child->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE)
+			return;
 	}
-	if (!valid)
-		return;
-
 	/*
-	 * spec 2.0 suggests all functions should be configured the same
-	 * setting for ASPM. Enabling ASPM L1 should be done in upstream
-	 * component first and then downstream, and vice versa for disabling
-	 * ASPM L1. Spec doesn't mention L0S.
+	 * Spec 2.0 suggests all functions should be configured the
+	 * same setting for ASPM. Enabling ASPM L1 should be done in
+	 * upstream component first and then downstream, and vice
+	 * versa for disabling ASPM L1. Spec doesn't mention L0S.
 	 */
 	if (state & PCIE_LINK_STATE_L1)
-		__pcie_aspm_config_one_dev(pdev, state);
+		__pcie_aspm_config_one_dev(parent, state);
 
-	list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list)
-		__pcie_aspm_config_one_dev(child_dev, state);
+	list_for_each_entry(child, &linkbus->devices, bus_list)
+		__pcie_aspm_config_one_dev(child, state);
 
 	if (!(state & PCIE_LINK_STATE_L1))
-		__pcie_aspm_config_one_dev(pdev, state);
+		__pcie_aspm_config_one_dev(parent, state);
 
-	link_state->aspm_enabled = state;
+	link->aspm_enabled = state;
 }
 
 static struct pcie_link_state *get_root_port_link(struct pcie_link_state *link)
@@ -524,42 +503,38 @@ static struct pcie_link_state *get_root_port_link(struct pcie_link_state *link)
 	return root_port_link;
 }
 
-/* check the whole hierarchy, and configure each link in the hierarchy */
-static void __pcie_aspm_configure_link_state(struct pci_dev *pdev,
-	unsigned int state)
+/* Check the whole hierarchy, and configure each link in the hierarchy */
+static void __pcie_aspm_configure_link_state(struct pcie_link_state *link,
+					     u32 state)
 {
-	struct pcie_link_state *link_state = pdev->link_state;
-	struct pcie_link_state *root_port_link = get_root_port_link(link_state);
-	struct pcie_link_state *leaf;
+	struct pcie_link_state *leaf, *root = get_root_port_link(link);
 
-	state &= PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1;
+	state &= (PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1);
 
-	/* check all links who have specific root port link */
+	/* Check all links who have specific root port link */
 	list_for_each_entry(leaf, &link_list, sibling) {
 		if (!list_empty(&leaf->children) ||
-			get_root_port_link(leaf) != root_port_link)
+		    get_root_port_link(leaf) != root)
 			continue;
-		state = pcie_aspm_check_state(leaf->pdev, state);
+		state = pcie_aspm_check_state(leaf, state);
 	}
-	/* check root port link too in case it hasn't children */
-	state = pcie_aspm_check_state(root_port_link->pdev, state);
-
-	if (link_state->aspm_enabled == state)
+	/* Check root port link too in case it hasn't children */
+	state = pcie_aspm_check_state(root, state);
+	if (link->aspm_enabled == state)
 		return;
-
 	/*
-	 * we must change the hierarchy. See comments in
+	 * We must change the hierarchy. See comments in
 	 * __pcie_aspm_config_link for the order
 	 **/
 	if (state & PCIE_LINK_STATE_L1) {
 		list_for_each_entry(leaf, &link_list, sibling) {
-			if (get_root_port_link(leaf) == root_port_link)
-				__pcie_aspm_config_link(leaf->pdev, state);
+			if (get_root_port_link(leaf) == root)
+				__pcie_aspm_config_link(leaf, state);
 		}
 	} else {
 		list_for_each_entry_reverse(leaf, &link_list, sibling) {
-			if (get_root_port_link(leaf) == root_port_link)
-				__pcie_aspm_config_link(leaf->pdev, state);
+			if (get_root_port_link(leaf) == root)
+				__pcie_aspm_config_link(leaf, state);
 		}
 	}
 }
@@ -568,20 +543,20 @@ static void __pcie_aspm_configure_link_state(struct pci_dev *pdev,
  * pcie_aspm_configure_link_state: enable/disable PCI express link state
  * @pdev: the root port or switch downstream port
  */
-static void pcie_aspm_configure_link_state(struct pci_dev *pdev,
-	unsigned int state)
+static void pcie_aspm_configure_link_state(struct pcie_link_state *link,
+					   u32 state)
 {
 	down_read(&pci_bus_sem);
 	mutex_lock(&aspm_lock);
-	__pcie_aspm_configure_link_state(pdev, state);
+	__pcie_aspm_configure_link_state(link, state);
 	mutex_unlock(&aspm_lock);
 	up_read(&pci_bus_sem);
 }
 
-static void free_link_state(struct pci_dev *pdev)
+static void free_link_state(struct pcie_link_state *link)
 {
-	kfree(pdev->link_state);
-	pdev->link_state = NULL;
+	link->pdev->link_state = NULL;
+	kfree(link);
 }
 
 static int pcie_aspm_sanity_check(struct pci_dev *pdev)
@@ -648,7 +623,6 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
 	if (!link_state)
 		goto unlock_out;
 
-	link_state->has_switch = pcie_aspm_downstream_has_switch(pdev);
 	INIT_LIST_HEAD(&link_state->children);
 	INIT_LIST_HEAD(&link_state->link);
 	if (pdev->bus->self) {/* this is a switch */
@@ -662,12 +636,13 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
 		list_add(&link_state->link, &parent_link_state->children);
 		link_state->parent = parent_link_state;
 	}
-
+	link_state->pdev = pdev;
+	link_state->has_switch = pcie_aspm_downstream_has_switch(link_state);
 	pdev->link_state = link_state;
 
 	if (!blacklist) {
-		pcie_aspm_configure_common_clock(pdev);
-		pcie_aspm_cap_init(pdev);
+		pcie_aspm_configure_common_clock(link_state);
+		pcie_aspm_cap_init(link_state);
 	} else {
 		link_state->aspm_enabled =
 			(PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1);
@@ -676,7 +651,6 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
 		link_state->aspm_support = 0;
 	}
 
-	link_state->pdev = pdev;
 	list_add(&link_state->sibling, &link_list);
 
 	if (link_state->has_switch) {
@@ -685,17 +659,18 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
 		 * initialization will config the whole hierarchy. but we must
 		 * make sure BIOS doesn't set unsupported link state
 		 **/
-		state = pcie_aspm_check_state(pdev, link_state->aspm_default);
-		__pcie_aspm_config_link(pdev, state);
+		state = pcie_aspm_check_state(link_state,
+					      link_state->aspm_default);
+		__pcie_aspm_config_link(link_state, state);
 	} else
-		__pcie_aspm_configure_link_state(pdev,
-			policy_to_aspm_state(pdev));
+		__pcie_aspm_configure_link_state(link_state,
+			policy_to_aspm_state(link_state));
 
-	pcie_check_clock_pm(pdev, blacklist);
+	pcie_check_clock_pm(link_state, blacklist);
 
 unlock_out:
 	if (error)
-		free_link_state(pdev);
+		free_link_state(link_state);
 	mutex_unlock(&aspm_lock);
 out:
 	up_read(&pci_bus_sem);
@@ -728,7 +703,7 @@ void pcie_aspm_exit_link_state(struct pci_dev *pdev)
 	list_del(&link_state->link);
 	/* Clock PM is for endpoint device */
 
-	free_link_state(parent);
+	free_link_state(link_state);
 out:
 	mutex_unlock(&aspm_lock);
 	up_read(&pci_bus_sem);
@@ -748,7 +723,7 @@ void pcie_aspm_pm_state_change(struct pci_dev *pdev)
 	 * devices changed PM state, we should recheck if latency meets all
 	 * functions' requirement
 	 */
-	pcie_aspm_configure_link_state(pdev, link_state->aspm_enabled);
+	pcie_aspm_configure_link_state(link_state, link_state->aspm_enabled);
 }
 
 /*
@@ -775,9 +750,9 @@ void pci_disable_link_state(struct pci_dev *pdev, int state)
 	if (state & PCIE_LINK_STATE_CLKPM)
 		link_state->clkpm_capable = 0;
 
-	__pcie_aspm_configure_link_state(parent, link_state->aspm_enabled);
+	__pcie_aspm_configure_link_state(link_state, link_state->aspm_enabled);
 	if (!link_state->clkpm_capable && link_state->clkpm_enabled)
-		pcie_set_clock_pm(parent, 0);
+		pcie_set_clock_pm(link_state, 0);
 	mutex_unlock(&aspm_lock);
 	up_read(&pci_bus_sem);
 }
@@ -786,7 +761,6 @@ EXPORT_SYMBOL(pci_disable_link_state);
 static int pcie_aspm_set_policy(const char *val, struct kernel_param *kp)
 {
 	int i;
-	struct pci_dev *pdev;
 	struct pcie_link_state *link_state;
 
 	for (i = 0; i < ARRAY_SIZE(policy_str); i++)
@@ -801,12 +775,12 @@ static int pcie_aspm_set_policy(const char *val, struct kernel_param *kp)
 	mutex_lock(&aspm_lock);
 	aspm_policy = i;
 	list_for_each_entry(link_state, &link_list, sibling) {
-		pdev = link_state->pdev;
-		__pcie_aspm_configure_link_state(pdev,
-			policy_to_aspm_state(pdev));
+		__pcie_aspm_configure_link_state(link_state,
+			policy_to_aspm_state(link_state));
 		if (link_state->clkpm_capable &&
-		    link_state->clkpm_enabled != policy_to_clkpm_state(pdev))
-			pcie_set_clock_pm(pdev, policy_to_clkpm_state(pdev));
+		    link_state->clkpm_enabled != policy_to_clkpm_state(link_state))
+			pcie_set_clock_pm(link_state,
+					  policy_to_clkpm_state(link_state));
 
 	}
 	mutex_unlock(&aspm_lock);
@@ -844,7 +818,7 @@ static ssize_t link_state_store(struct device *dev,
 		const char *buf,
 		size_t n)
 {
-	struct pci_dev *pci_device = to_pci_dev(dev);
+	struct pci_dev *pdev = to_pci_dev(dev);
 	int state;
 
 	if (n < 1)
@@ -852,7 +826,7 @@ static ssize_t link_state_store(struct device *dev,
 	state = buf[0]-'0';
 	if (state >= 0 && state <= 3) {
 		/* setup link aspm state */
-		pcie_aspm_configure_link_state(pci_device, state);
+		pcie_aspm_configure_link_state(pdev->link_state, state);
 		return n;
 	}
 
@@ -883,7 +857,7 @@ static ssize_t clk_ctl_store(struct device *dev,
 
 	down_read(&pci_bus_sem);
 	mutex_lock(&aspm_lock);
-	pcie_set_clock_pm(pci_device, !!state);
+	pcie_set_clock_pm(pci_device->link_state, !!state);
 	mutex_unlock(&aspm_lock);
 	up_read(&pci_bus_sem);
 
-- 
cgit v1.2.3-59-g8ed1b


From 8d349ace9a5c2a8404bcf4a371fe170480ffbebb Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:18:22 +0900
Subject: PCI ASPM: cleanup initialization

Clean up ASPM initialization by refactoring some functionality, renaming
functions, and moving things around.

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 142 +++++++++++++++++++++++++-----------------------
 1 file changed, 75 insertions(+), 67 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 9eaaf95f65a2..68a4d4b15f9f 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -126,7 +126,7 @@ static void pcie_set_clock_pm(struct pcie_link_state *link, int enable)
 	link->clkpm_enabled = !!enable;
 }
 
-static void pcie_check_clock_pm(struct pcie_link_state *link, int blacklist)
+static void pcie_clkpm_cap_init(struct pcie_link_state *link, int blacklist)
 {
 	int pos, capable = 1, enabled = 1;
 	u32 reg32;
@@ -151,13 +151,7 @@ static void pcie_check_clock_pm(struct pcie_link_state *link, int blacklist)
 	}
 	link->clkpm_enabled = enabled;
 	link->clkpm_default = enabled;
-	if (!blacklist) {
-		link->clkpm_capable = capable;
-		pcie_set_clock_pm(link, policy_to_clkpm_state(link));
-	} else {
-		link->clkpm_capable = 0;
-		pcie_set_clock_pm(link, 0);
-	}
+	link->clkpm_capable = (blacklist) ? 0 : capable;
 }
 
 static bool pcie_aspm_downstream_has_switch(struct pcie_link_state *link)
@@ -314,12 +308,23 @@ static void pcie_aspm_get_cap_device(struct pci_dev *pdev, u32 *state,
 	*enabled = reg16 & (PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1);
 }
 
-static void pcie_aspm_cap_init(struct pcie_link_state *link)
+static void pcie_aspm_cap_init(struct pcie_link_state *link, int blacklist)
 {
 	u32 support, l0s, l1, enabled;
 	struct pci_dev *child, *parent = link->pdev;
 	struct pci_bus *linkbus = parent->subordinate;
 
+	if (blacklist) {
+		/* Set support state to 0, so we will disable ASPM later */
+		link->aspm_support = 0;
+		link->aspm_default = 0;
+		link->aspm_enabled = PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1;
+		return;
+	}
+
+	/* Configure common clock before checking latencies */
+	pcie_aspm_configure_common_clock(link);
+
 	/* upstream component states */
 	pcie_aspm_get_cap_device(parent, &support, &l0s, &l1, &enabled);
 	link->aspm_support = support;
@@ -590,6 +595,42 @@ static int pcie_aspm_sanity_check(struct pci_dev *pdev)
 	return 0;
 }
 
+static struct pcie_link_state *pcie_aspm_setup_link_state(struct pci_dev *pdev)
+{
+	struct pcie_link_state *link;
+	int blacklist = !!pcie_aspm_sanity_check(pdev);
+
+	link = kzalloc(sizeof(*link), GFP_KERNEL);
+	if (!link)
+		return NULL;
+	INIT_LIST_HEAD(&link->sibling);
+	INIT_LIST_HEAD(&link->children);
+	INIT_LIST_HEAD(&link->link);
+	link->pdev = pdev;
+	link->has_switch = pcie_aspm_downstream_has_switch(link);
+	if (pdev->pcie_type == PCI_EXP_TYPE_DOWNSTREAM) {
+		struct pcie_link_state *parent;
+		parent = pdev->bus->parent->self->link_state;
+		if (!parent) {
+			kfree(link);
+			return NULL;
+		}
+		link->parent = parent;
+		list_add(&link->link, &parent->children);
+	}
+	list_add(&link->sibling, &link_list);
+
+	pdev->link_state = link;
+
+	/* Check ASPM capability */
+	pcie_aspm_cap_init(link, blacklist);
+
+	/* Check Clock PM capability */
+	pcie_clkpm_cap_init(link, blacklist);
+
+	return link;
+}
+
 /*
  * pcie_aspm_init_link_state: Initiate PCI express link state.
  * It is called after the pcie and its children devices are scaned.
@@ -597,80 +638,47 @@ static int pcie_aspm_sanity_check(struct pci_dev *pdev)
  */
 void pcie_aspm_init_link_state(struct pci_dev *pdev)
 {
-	unsigned int state;
-	struct pcie_link_state *link_state;
-	int error = 0;
-	int blacklist;
+	u32 state;
+	struct pcie_link_state *link;
 
 	if (aspm_disabled || !pdev->is_pcie || pdev->link_state)
 		return;
 	if (pdev->pcie_type != PCI_EXP_TYPE_ROOT_PORT &&
-		pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM)
+	    pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM)
 		return;
+
 	/* VIA has a strange chipset, root port is under a bridge */
 	if (pdev->pcie_type == PCI_EXP_TYPE_ROOT_PORT &&
-		pdev->bus->self)
+	    pdev->bus->self)
 		return;
+
 	down_read(&pci_bus_sem);
 	if (list_empty(&pdev->subordinate->devices))
 		goto out;
 
-	blacklist = !!pcie_aspm_sanity_check(pdev);
-
 	mutex_lock(&aspm_lock);
-
-	link_state = kzalloc(sizeof(*link_state), GFP_KERNEL);
-	if (!link_state)
-		goto unlock_out;
-
-	INIT_LIST_HEAD(&link_state->children);
-	INIT_LIST_HEAD(&link_state->link);
-	if (pdev->bus->self) {/* this is a switch */
-		struct pcie_link_state *parent_link_state;
-
-		parent_link_state = pdev->bus->parent->self->link_state;
-		if (!parent_link_state) {
-			kfree(link_state);
-			goto unlock_out;
-		}
-		list_add(&link_state->link, &parent_link_state->children);
-		link_state->parent = parent_link_state;
-	}
-	link_state->pdev = pdev;
-	link_state->has_switch = pcie_aspm_downstream_has_switch(link_state);
-	pdev->link_state = link_state;
-
-	if (!blacklist) {
-		pcie_aspm_configure_common_clock(link_state);
-		pcie_aspm_cap_init(link_state);
+	link = pcie_aspm_setup_link_state(pdev);
+	if (!link)
+		goto unlock;
+	/*
+	 * Setup initial ASPM state
+	 *
+	 * If link has switch, delay the link config. The leaf link
+	 * initialization will config the whole hierarchy. But we must
+	 * make sure BIOS doesn't set unsupported link state.
+	 */
+	if (link->has_switch) {
+		state = pcie_aspm_check_state(link, link->aspm_default);
+		__pcie_aspm_config_link(link, state);
 	} else {
-		link_state->aspm_enabled =
-			(PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1);
-		link_state->aspm_default = 0;
-		/* Set support state to 0, so we will disable ASPM later */
-		link_state->aspm_support = 0;
+		state = policy_to_aspm_state(link);
+		__pcie_aspm_configure_link_state(link, state);
 	}
 
-	list_add(&link_state->sibling, &link_list);
-
-	if (link_state->has_switch) {
-		/*
-		 * If link has switch, delay the link config. The leaf link
-		 * initialization will config the whole hierarchy. but we must
-		 * make sure BIOS doesn't set unsupported link state
-		 **/
-		state = pcie_aspm_check_state(link_state,
-					      link_state->aspm_default);
-		__pcie_aspm_config_link(link_state, state);
-	} else
-		__pcie_aspm_configure_link_state(link_state,
-			policy_to_aspm_state(link_state));
-
-	pcie_check_clock_pm(link_state, blacklist);
-
-unlock_out:
-	if (error)
-		free_link_state(link_state);
+	/* Setup initial Clock PM state */
+	state = (link->clkpm_capable) ? policy_to_clkpm_state(link) : 0;
+	pcie_set_clock_pm(link, state);
+unlock:
 	mutex_unlock(&aspm_lock);
 out:
 	up_read(&pci_bus_sem);
-- 
cgit v1.2.3-59-g8ed1b


From f7ea3d7fc03753b08e267fece19c56383e6b856f Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:19:00 +0900
Subject: PCI ASPM: cleanup __pcie_aspm_check_state_one

Clean up and simplify __pcie_aspm_check_state_one().

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 67 ++++++++++++++++++-------------------------------
 1 file changed, 25 insertions(+), 42 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 68a4d4b15f9f..694ba5657758 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -370,56 +370,39 @@ static void pcie_aspm_cap_init(struct pcie_link_state *link, int blacklist)
 	}
 }
 
-static unsigned int __pcie_aspm_check_state_one(struct pci_dev *pdev,
-	unsigned int state)
+/**
+ * __pcie_aspm_check_state_one - check latency for endpoint device.
+ * @endpoint: pointer to the struct pci_dev of endpoint device
+ *
+ * TBD: The latency from the endpoint to root complex vary per switch's
+ * upstream link state above the device. Here we just do a simple check
+ * which assumes all links above the device can be in L1 state, that
+ * is we just consider the worst case. If switch's upstream link can't
+ * be put into L0S/L1, then our check is too strictly.
+ */
+static u32 __pcie_aspm_check_state_one(struct pci_dev *endpoint, u32 state)
 {
-	struct pci_dev *parent_dev, *tmp_dev;
-	unsigned int l1_latency = 0;
-	struct pcie_link_state *link_state;
+	u32 l1_switch_latency = 0;
 	struct aspm_latency *acceptable;
+	struct pcie_link_state *link;
 
-	parent_dev = pdev->bus->self;
-	link_state = parent_dev->link_state;
-	state &= link_state->aspm_support;
-	if (state == 0)
-		return 0;
-	acceptable = &link_state->acceptable[PCI_FUNC(pdev->devfn)];
+	link = endpoint->bus->self->link_state;
+	state &= link->aspm_support;
+	acceptable = &link->acceptable[PCI_FUNC(endpoint->devfn)];
 
-	/*
-	 * Check latency for endpoint device.
-	 * TBD: The latency from the endpoint to root complex vary per
-	 * switch's upstream link state above the device. Here we just do a
-	 * simple check which assumes all links above the device can be in L1
-	 * state, that is we just consider the worst case. If switch's upstream
-	 * link can't be put into L0S/L1, then our check is too strictly.
-	 */
-	tmp_dev = pdev;
-	while (state & (PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1)) {
-		parent_dev = tmp_dev->bus->self;
-		link_state = parent_dev->link_state;
+	while (link && state) {
 		if ((state & PCIE_LINK_STATE_L0S) &&
-		    (link_state->latency.l0s > acceptable->l0s))
+		    (link->latency.l0s > acceptable->l0s))
 			state &= ~PCIE_LINK_STATE_L0S;
-
 		if ((state & PCIE_LINK_STATE_L1) &&
-		    (link_state->latency.l1 + l1_latency > acceptable->l1))
+		    (link->latency.l1 + l1_switch_latency > acceptable->l1))
 			state &= ~PCIE_LINK_STATE_L1;
-
-		if (!parent_dev->bus->self) /* parent_dev is a root port */
-			break;
-		else {
-			/*
-			 * parent_dev is the downstream port of a switch, make
-			 * tmp_dev the upstream port of the switch
-			 */
-			tmp_dev = parent_dev->bus->self;
-			/*
-			 * every switch on the path to root complex need 1 more
-			 * microsecond for L1. Spec doesn't mention L0S.
-			 */
-			if (state & PCIE_LINK_STATE_L1)
-				l1_latency += 1000;
-		}
+		link = link->parent;
+		/*
+		 * Every switch on the path to root complex need 1
+		 * more microsecond for L1. Spec doesn't mention L0s.
+		 */
+		l1_switch_latency += 1000;
 	}
 	return state;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 430842e29d396928989c0a45e05025e988004d79 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:20:10 +0900
Subject: PCI ASPM: cleanup clkpm checks

In the current ASPM implementation, callers of pcie_set_clock_pm() check
Clock PM capability of the link or current Clock PM state of the link.
This check should be done in pcie_set_clock_pm() itself.

This patch moves those checks into pcie_set_clock_pm(). It also
introduces pcie_set_clkpm_nocheck() that is equivalent to old
pcie_set_clock_pm(), for the caller who wants to change Clocl PM state
regardless of the Clock PM capability or current Clock PM state. In
addition, this patch changes the function name from
pcie_set_clock_pm() to pcie_set_clkpm() for consistency.

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 34 ++++++++++++++++++++--------------
 1 file changed, 20 insertions(+), 14 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 694ba5657758..0d0d3e630925 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -105,7 +105,7 @@ static int policy_to_clkpm_state(struct pcie_link_state *link)
 	return 0;
 }
 
-static void pcie_set_clock_pm(struct pcie_link_state *link, int enable)
+static void pcie_set_clkpm_nocheck(struct pcie_link_state *link, int enable)
 {
 	int pos;
 	u16 reg16;
@@ -126,6 +126,17 @@ static void pcie_set_clock_pm(struct pcie_link_state *link, int enable)
 	link->clkpm_enabled = !!enable;
 }
 
+static void pcie_set_clkpm(struct pcie_link_state *link, int enable)
+{
+	/* Don't enable Clock PM if the link is not Clock PM capable */
+	if (!link->clkpm_capable && enable)
+		return;
+	/* Need nothing if the specified equals to current state */
+	if (link->clkpm_enabled == enable)
+		return;
+	pcie_set_clkpm_nocheck(link, enable);
+}
+
 static void pcie_clkpm_cap_init(struct pcie_link_state *link, int blacklist)
 {
 	int pos, capable = 1, enabled = 1;
@@ -660,7 +671,7 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
 
 	/* Setup initial Clock PM state */
 	state = (link->clkpm_capable) ? policy_to_clkpm_state(link) : 0;
-	pcie_set_clock_pm(link, state);
+	pcie_set_clkpm(link, state);
 unlock:
 	mutex_unlock(&aspm_lock);
 out:
@@ -738,12 +749,11 @@ void pci_disable_link_state(struct pci_dev *pdev, int state)
 	mutex_lock(&aspm_lock);
 	link_state = parent->link_state;
 	link_state->aspm_support &= ~state;
-	if (state & PCIE_LINK_STATE_CLKPM)
-		link_state->clkpm_capable = 0;
-
 	__pcie_aspm_configure_link_state(link_state, link_state->aspm_enabled);
-	if (!link_state->clkpm_capable && link_state->clkpm_enabled)
-		pcie_set_clock_pm(link_state, 0);
+	if (state & PCIE_LINK_STATE_CLKPM) {
+		link_state->clkpm_capable = 0;
+		pcie_set_clkpm(link_state, 0);
+	}
 	mutex_unlock(&aspm_lock);
 	up_read(&pci_bus_sem);
 }
@@ -768,11 +778,7 @@ static int pcie_aspm_set_policy(const char *val, struct kernel_param *kp)
 	list_for_each_entry(link_state, &link_list, sibling) {
 		__pcie_aspm_configure_link_state(link_state,
 			policy_to_aspm_state(link_state));
-		if (link_state->clkpm_capable &&
-		    link_state->clkpm_enabled != policy_to_clkpm_state(link_state))
-			pcie_set_clock_pm(link_state,
-					  policy_to_clkpm_state(link_state));
-
+		pcie_set_clkpm(link_state, policy_to_clkpm_state(link_state));
 	}
 	mutex_unlock(&aspm_lock);
 	up_read(&pci_bus_sem);
@@ -839,7 +845,7 @@ static ssize_t clk_ctl_store(struct device *dev,
 		const char *buf,
 		size_t n)
 {
-	struct pci_dev *pci_device = to_pci_dev(dev);
+	struct pci_dev *pdev = to_pci_dev(dev);
 	int state;
 
 	if (n < 1)
@@ -848,7 +854,7 @@ static ssize_t clk_ctl_store(struct device *dev,
 
 	down_read(&pci_bus_sem);
 	mutex_lock(&aspm_lock);
-	pcie_set_clock_pm(pci_device->link_state, !!state);
+	pcie_set_clkpm_nocheck(pdev->link_state, !!state);
 	mutex_unlock(&aspm_lock);
 	up_read(&pci_bus_sem);
 
-- 
cgit v1.2.3-59-g8ed1b


From 7ab709910323a8af20722c066267516b3e7680a2 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:20:48 +0900
Subject: PCI ASPM: cleanup pcie_aspm_get_cap_device

Minor cleanup for pcie_aspm_get_cap_device().

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 0d0d3e630925..d85f77ff150c 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -292,19 +292,18 @@ static unsigned int calc_L1_latency(unsigned int latency_encoding, int ac)
 }
 
 static void pcie_aspm_get_cap_device(struct pci_dev *pdev, u32 *state,
-	unsigned int *l0s, unsigned int *l1, unsigned int *enabled)
+				     u32 *l0s, u32 *l1, u32 *enabled)
 {
 	int pos;
 	u16 reg16;
-	u32 reg32;
-	unsigned int latency;
+	u32 reg32, latency;
 
 	*l0s = *l1 = *enabled = 0;
 	pos = pci_find_capability(pdev, PCI_CAP_ID_EXP);
 	pci_read_config_dword(pdev, pos + PCI_EXP_LNKCAP, &reg32);
 	*state = (reg32 & PCI_EXP_LNKCAP_ASPMS) >> 10;
 	if (*state != PCIE_LINK_STATE_L0S &&
-		*state != (PCIE_LINK_STATE_L1|PCIE_LINK_STATE_L0S))
+	    *state != (PCIE_LINK_STATE_L1 | PCIE_LINK_STATE_L0S))
 		*state = 0;
 	if (*state == 0)
 		return;
@@ -316,7 +315,7 @@ static void pcie_aspm_get_cap_device(struct pci_dev *pdev, u32 *state,
 		*l1 = calc_L1_latency(latency, 0);
 	}
 	pci_read_config_word(pdev, pos + PCI_EXP_LNKCTL, &reg16);
-	*enabled = reg16 & (PCIE_LINK_STATE_L0S|PCIE_LINK_STATE_L1);
+	*enabled = reg16 & (PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1);
 }
 
 static void pcie_aspm_cap_init(struct pcie_link_state *link, int blacklist)
-- 
cgit v1.2.3-59-g8ed1b


From 5e0eaa7d3679c3ef8618803bc9311270e5816641 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:21:48 +0900
Subject: PCI ASPM: cleanup calc_Lx_latency

Cleanup for calc_L0S_latency() and calc_L1_latency().

  - Separate exit latency and acceptable latency calculation.
  - Some minor cleanups.

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 73 +++++++++++++++++++++++--------------------------
 1 file changed, 34 insertions(+), 39 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index d85f77ff150c..23fabb303e82 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -257,38 +257,36 @@ static void pcie_aspm_configure_common_clock(struct pcie_link_state *link)
 	pci_write_config_word(parent, ppos + PCI_EXP_LNKCTL, parent_reg);
 }
 
-/*
- * calc_L0S_latency: Convert L0s latency encoding to ns
- */
-static unsigned int calc_L0S_latency(unsigned int latency_encoding, int ac)
+/* Convert L0s latency encoding to ns */
+static u32 calc_l0s_latency(u32 encoding)
 {
-	unsigned int ns = 64;
+	if (encoding == 0x7)
+		return (5 * 1000);	/* > 4us */
+	return (64 << encoding);
+}
 
-	if (latency_encoding == 0x7) {
-		if (ac)
-			ns = -1U;
-		else
-			ns = 5*1000; /* > 4us */
-	} else
-		ns *= (1 << latency_encoding);
-	return ns;
+/* Convert L0s acceptable latency encoding to ns */
+static u32 calc_l0s_acceptable(u32 encoding)
+{
+	if (encoding == 0x7)
+		return -1U;
+	return (64 << encoding);
 }
 
-/*
- * calc_L1_latency: Convert L1 latency encoding to ns
- */
-static unsigned int calc_L1_latency(unsigned int latency_encoding, int ac)
+/* Convert L1 latency encoding to ns */
+static u32 calc_l1_latency(u32 encoding)
 {
-	unsigned int ns = 1000;
+	if (encoding == 0x7)
+		return (65 * 1000);	/* > 64us */
+	return (1000 << encoding);
+}
 
-	if (latency_encoding == 0x7) {
-		if (ac)
-			ns = -1U;
-		else
-			ns = 65*1000; /* > 64us */
-	} else
-		ns *= (1 << latency_encoding);
-	return ns;
+/* Convert L1 acceptable latency encoding to ns */
+static u32 calc_l1_acceptable(u32 encoding)
+{
+	if (encoding == 0x7)
+		return -1U;
+	return (1000 << encoding);
 }
 
 static void pcie_aspm_get_cap_device(struct pci_dev *pdev, u32 *state,
@@ -296,7 +294,7 @@ static void pcie_aspm_get_cap_device(struct pci_dev *pdev, u32 *state,
 {
 	int pos;
 	u16 reg16;
-	u32 reg32, latency;
+	u32 reg32, encoding;
 
 	*l0s = *l1 = *enabled = 0;
 	pos = pci_find_capability(pdev, PCI_CAP_ID_EXP);
@@ -308,11 +306,11 @@ static void pcie_aspm_get_cap_device(struct pci_dev *pdev, u32 *state,
 	if (*state == 0)
 		return;
 
-	latency = (reg32 & PCI_EXP_LNKCAP_L0SEL) >> 12;
-	*l0s = calc_L0S_latency(latency, 0);
+	encoding = (reg32 & PCI_EXP_LNKCAP_L0SEL) >> 12;
+	*l0s = calc_l0s_latency(encoding);
 	if (*state & PCIE_LINK_STATE_L1) {
-		latency = (reg32 & PCI_EXP_LNKCAP_L1EL) >> 15;
-		*l1 = calc_L1_latency(latency, 0);
+		encoding = (reg32 & PCI_EXP_LNKCAP_L1EL) >> 15;
+		*l1 = calc_l1_latency(encoding);
 	}
 	pci_read_config_word(pdev, pos + PCI_EXP_LNKCTL, &reg16);
 	*enabled = reg16 & (PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1);
@@ -358,8 +356,7 @@ static void pcie_aspm_cap_init(struct pcie_link_state *link, int blacklist)
 	/* ENDPOINT states*/
 	list_for_each_entry(child, &linkbus->devices, bus_list) {
 		int pos;
-		u32 reg32;
-		unsigned int latency;
+		u32 reg32, encoding;
 		struct aspm_latency *acceptable =
 			&link->acceptable[PCI_FUNC(child->devfn)];
 
@@ -369,13 +366,11 @@ static void pcie_aspm_cap_init(struct pcie_link_state *link, int blacklist)
 
 		pos = pci_find_capability(child, PCI_CAP_ID_EXP);
 		pci_read_config_dword(child, pos + PCI_EXP_DEVCAP, &reg32);
-		latency = (reg32 & PCI_EXP_DEVCAP_L0S) >> 6;
-		latency = calc_L0S_latency(latency, 1);
-		acceptable->l0s = latency;
+		encoding = (reg32 & PCI_EXP_DEVCAP_L0S) >> 6;
+		acceptable->l0s = calc_l0s_acceptable(encoding);
 		if (link->aspm_support & PCIE_LINK_STATE_L1) {
-			latency = (reg32 & PCI_EXP_DEVCAP_L1) >> 9;
-			latency = calc_L1_latency(latency, 1);
-			acceptable->l1 = latency;
+			encoding = (reg32 & PCI_EXP_DEVCAP_L1) >> 9;
+			acceptable->l1 = calc_l1_acceptable(encoding);
 		}
 	}
 }
-- 
cgit v1.2.3-59-g8ed1b


From efdf8288819df67d608a186f9d17a7d4051f3c1f Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:22:26 +0900
Subject: PCI ASPM: remove has_switch field

We don't need the 'has_switch' field in the struct pcie_link_state.

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 23fabb303e82..26fd39caebc5 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -48,8 +48,6 @@ struct pcie_link_state {
 	u32 clkpm_enabled:1;		/* Current Clock PM state */
 	u32 clkpm_default:1;		/* Default Clock PM state by BIOS */
 
-	u32 has_switch:1;		/* Downstream has switches? */
-
 	/* Latencies */
 	struct aspm_latency latency;	/* Exit latency */
 	/*
@@ -595,7 +593,6 @@ static struct pcie_link_state *pcie_aspm_setup_link_state(struct pci_dev *pdev)
 	INIT_LIST_HEAD(&link->children);
 	INIT_LIST_HEAD(&link->link);
 	link->pdev = pdev;
-	link->has_switch = pcie_aspm_downstream_has_switch(link);
 	if (pdev->pcie_type == PCI_EXP_TYPE_DOWNSTREAM) {
 		struct pcie_link_state *parent;
 		parent = pdev->bus->parent->self->link_state;
@@ -655,7 +652,7 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
 	 * initialization will config the whole hierarchy. But we must
 	 * make sure BIOS doesn't set unsupported link state.
 	 */
-	if (link->has_switch) {
+	if (pcie_aspm_downstream_has_switch(link)) {
 		state = pcie_aspm_check_state(link, link->aspm_default);
 		__pcie_aspm_config_link(link, state);
 	} else {
-- 
cgit v1.2.3-59-g8ed1b


From 3647584d9ef35c9ec4abefdbea29959c26c54f13 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:23:09 +0900
Subject: PCI ASPM: cleanup pcie_aspm_sanity_check

Minor cleanup for pcie_aspm_sanity_check().

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 21 +++++++++------------
 1 file changed, 9 insertions(+), 12 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 26fd39caebc5..04b6a3098505 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -552,27 +552,24 @@ static void free_link_state(struct pcie_link_state *link)
 
 static int pcie_aspm_sanity_check(struct pci_dev *pdev)
 {
-	struct pci_dev *child_dev;
-	int child_pos;
+	struct pci_dev *child;
+	int pos;
 	u32 reg32;
-
 	/*
-	 * Some functions in a slot might not all be PCIE functions, very
-	 * strange. Disable ASPM for the whole slot
+	 * Some functions in a slot might not all be PCIE functions,
+	 * very strange. Disable ASPM for the whole slot
 	 */
-	list_for_each_entry(child_dev, &pdev->subordinate->devices, bus_list) {
-		child_pos = pci_find_capability(child_dev, PCI_CAP_ID_EXP);
-		if (!child_pos)
+	list_for_each_entry(child, &pdev->subordinate->devices, bus_list) {
+		pos = pci_find_capability(child, PCI_CAP_ID_EXP);
+		if (!pos)
 			return -EINVAL;
-
 		/*
 		 * Disable ASPM for pre-1.1 PCIe device, we follow MS to use
 		 * RBER bit to determine if a function is 1.1 version device
 		 */
-		pci_read_config_dword(child_dev, child_pos + PCI_EXP_DEVCAP,
-			&reg32);
+		pci_read_config_dword(child, pos + PCI_EXP_DEVCAP, &reg32);
 		if (!(reg32 & PCI_EXP_DEVCAP_RBER) && !aspm_force) {
-			dev_printk(KERN_INFO, &child_dev->dev, "disabling ASPM"
+			dev_printk(KERN_INFO, &child->dev, "disabling ASPM"
 				" on pre-1.1 PCIe device.  You can enable it"
 				" with 'pcie_aspm=force'\n");
 			return -EINVAL;
-- 
cgit v1.2.3-59-g8ed1b


From 5c92ffb1ecc7f13267cdef5dda8a838937912c93 Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Wed, 13 May 2009 12:23:57 +0900
Subject: PCI ASPM: remove get_root_port_link

By having a pointer to the root port link, we can remove loops in
get_root_port_link() to search the root port link.

Acked-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pcie/aspm.c | 24 +++++++++++-------------
 1 file changed, 11 insertions(+), 13 deletions(-)

diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 04b6a3098505..3d27c97e0486 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -33,6 +33,7 @@ struct aspm_latency {
 
 struct pcie_link_state {
 	struct pci_dev *pdev;		/* Upstream component of the Link */
+	struct pcie_link_state *root;	/* pointer to the root port link */
 	struct pcie_link_state *parent;	/* pointer to the parent Link state */
 	struct list_head sibling;	/* node in link_list */
 	struct list_head children;	/* list of child link states */
@@ -486,26 +487,17 @@ static void __pcie_aspm_config_link(struct pcie_link_state *link, u32 state)
 	link->aspm_enabled = state;
 }
 
-static struct pcie_link_state *get_root_port_link(struct pcie_link_state *link)
-{
-	struct pcie_link_state *root_port_link = link;
-	while (root_port_link->parent)
-		root_port_link = root_port_link->parent;
-	return root_port_link;
-}
-
 /* Check the whole hierarchy, and configure each link in the hierarchy */
 static void __pcie_aspm_configure_link_state(struct pcie_link_state *link,
 					     u32 state)
 {
-	struct pcie_link_state *leaf, *root = get_root_port_link(link);
+	struct pcie_link_state *leaf, *root = link->root;
 
 	state &= (PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1);
 
 	/* Check all links who have specific root port link */
 	list_for_each_entry(leaf, &link_list, sibling) {
-		if (!list_empty(&leaf->children) ||
-		    get_root_port_link(leaf) != root)
+		if (!list_empty(&leaf->children) || (leaf->root != root))
 			continue;
 		state = pcie_aspm_check_state(leaf, state);
 	}
@@ -519,12 +511,12 @@ static void __pcie_aspm_configure_link_state(struct pcie_link_state *link,
 	 **/
 	if (state & PCIE_LINK_STATE_L1) {
 		list_for_each_entry(leaf, &link_list, sibling) {
-			if (get_root_port_link(leaf) == root)
+			if (leaf->root == root)
 				__pcie_aspm_config_link(leaf, state);
 		}
 	} else {
 		list_for_each_entry_reverse(leaf, &link_list, sibling) {
-			if (get_root_port_link(leaf) == root)
+			if (leaf->root == root)
 				__pcie_aspm_config_link(leaf, state);
 		}
 	}
@@ -600,6 +592,12 @@ static struct pcie_link_state *pcie_aspm_setup_link_state(struct pci_dev *pdev)
 		link->parent = parent;
 		list_add(&link->link, &parent->children);
 	}
+	/* Setup a pointer to the root port link */
+	if (!link->parent)
+		link->root = link;
+	else
+		link->root = link->parent->root;
+
 	list_add(&link->sibling, &link_list);
 
 	pdev->link_state = link;
-- 
cgit v1.2.3-59-g8ed1b


From aa93d632c496184e5b779dbcf961bf1c6ececf0b Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Tue, 5 May 2009 09:52:46 -0700
Subject: drm/i915: Require digital monitor on HDMI ports for detect

HDMI and DVI both require DDC/EDID on monitors, so use
that to know when a monitor is connected as the hot-plug
pins are shared with SDVO and DisplayPort

Signed-off-by: Keith Packard <keithp@keithp.com>
---
 drivers/gpu/drm/i915/intel_hdmi.c | 30 ++++++++++++++++--------------
 1 file changed, 16 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 4ea2a651b92c..2495359ea8de 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -31,6 +31,7 @@
 #include "drmP.h"
 #include "drm.h"
 #include "drm_crtc.h"
+#include "drm_edid.h"
 #include "intel_drv.h"
 #include "i915_drm.h"
 #include "i915_drv.h"
@@ -129,20 +130,26 @@ static bool intel_hdmi_mode_fixup(struct drm_encoder *encoder,
 	return true;
 }
 
-static void
-intel_hdmi_sink_detect(struct drm_connector *connector)
+static enum drm_connector_status
+intel_hdmi_edid_detect(struct drm_connector *connector)
 {
 	struct intel_output *intel_output = to_intel_output(connector);
 	struct intel_hdmi_priv *hdmi_priv = intel_output->dev_priv;
 	struct edid *edid = NULL;
+	enum drm_connector_status status = connector_status_disconnected;
 
 	edid = drm_get_edid(&intel_output->base,
 			    &intel_output->ddc_bus->adapter);
-	if (edid != NULL) {
-		hdmi_priv->has_hdmi_sink = drm_detect_hdmi_monitor(edid);
-		kfree(edid);
+	hdmi_priv->has_hdmi_sink = false;
+	if (edid) {
+		if (edid->digital) {
+			status = connector_status_connected;
+			hdmi_priv->has_hdmi_sink = drm_detect_hdmi_monitor(edid);
+		}
 		intel_output->base.display_info.raw_edid = NULL;
+		kfree(edid);
 	}
+	return status;
 }
 
 static enum drm_connector_status
@@ -154,11 +161,7 @@ igdng_hdmi_detect(struct drm_connector *connector)
 	/* FIXME hotplug detect */
 
 	hdmi_priv->has_hdmi_sink = false;
-	intel_hdmi_sink_detect(connector);
-	if (hdmi_priv->has_hdmi_sink)
-		return connector_status_connected;
-	else
-		return connector_status_disconnected;
+	return intel_hdmi_edid_detect(connector);
 }
 
 static enum drm_connector_status
@@ -201,10 +204,9 @@ intel_hdmi_detect(struct drm_connector *connector)
 		return connector_status_unknown;
 	}
 
-	if ((I915_READ(PORT_HOTPLUG_STAT) & bit) != 0) {
-		intel_hdmi_sink_detect(connector);
-		return connector_status_connected;
-	} else
+	if ((I915_READ(PORT_HOTPLUG_STAT) & bit) != 0)
+		return intel_hdmi_edid_detect(connector);
+	else
 		return connector_status_disconnected;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 98acd46f356e560c371c0e416d92e8e56be31804 Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Sun, 14 Jun 2009 12:31:58 -0700
Subject: drm/i915: Apple DMI info has inconsistent SYS_VENDOR information

Some machines say 'Apple Inc.' while others say 'Apple Computer, Inc'.
Switch the test to just look for 'Apple' instead.

Signed-off-by: Keith Packard <keithp@keithp.com>
---
 drivers/gpu/drm/i915/intel_lvds.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index f073ed8432e8..345e5055f1c0 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -456,7 +456,7 @@ static const struct dmi_system_id intel_no_lvds[] = {
 		.callback = intel_no_lvds_dmi_callback,
 		.ident = "Apple Mac Mini (Core series)",
 		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
+			DMI_MATCH(DMI_SYS_VENDOR, "Apple"),
 			DMI_MATCH(DMI_PRODUCT_NAME, "Macmini1,1"),
 		},
 	},
@@ -464,7 +464,7 @@ static const struct dmi_system_id intel_no_lvds[] = {
 		.callback = intel_no_lvds_dmi_callback,
 		.ident = "Apple Mac Mini (Core 2 series)",
 		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
+			DMI_MATCH(DMI_SYS_VENDOR, "Apple"),
 			DMI_MATCH(DMI_PRODUCT_NAME, "Macmini2,1"),
 		},
 	},
-- 
cgit v1.2.3-59-g8ed1b


From b99e228d354cc1e7f19fb8b5f1297d493e309186 Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Wed, 10 Jun 2009 19:08:16 -0700
Subject: drm/i915: check for CONFIG_PNP before using pnp function

Signed-off-by: Keith Packard <keithp@keithp.com>
---
 drivers/gpu/drm/i915/i915_gem_tiling.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c
index 5c1ceec49f5b..daeae62e1c28 100644
--- a/drivers/gpu/drm/i915/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
@@ -114,11 +114,13 @@ intel_alloc_mchbar_resource(struct drm_device *dev)
 	mchbar_addr = ((u64)temp_hi << 32) | temp_lo;
 
 	/* If ACPI doesn't have it, assume we need to allocate it ourselves */
+#ifdef CONFIG_PNP
 	if (mchbar_addr &&
 	    pnp_range_reserved(mchbar_addr, mchbar_addr + MCHBAR_SIZE)) {
 		ret = 0;
 		goto out_put;
 	}
+#endif
 
 	/* Get some space for it */
 	ret = pci_bus_alloc_resource(bridge_dev->bus, &dev_priv->mch_res,
-- 
cgit v1.2.3-59-g8ed1b


From f9c10a9b96a31b4a82a4fa807400c04f00284068 Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Sat, 30 May 2009 12:16:25 -0700
Subject: drm/i915: Change I2C api to pass around i2c_adapters

The existing API passed around intel_i2c_chan pointers, which are dependent
on the i2c bit-banging algo. This precluded the driver from using outputs
which use a different algo. Switching to the more general i2c_adpater allows
the driver to support non bit-banging DDC.

This also required moving the slave address into the output private
structures.

Signed-off-by: Keith Packard <keithp@keithp.com>
---
 drivers/gpu/drm/i915/dvo.h         |  4 +--
 drivers/gpu/drm/i915/dvo_ch7017.c  | 20 ++++++++-------
 drivers/gpu/drm/i915/dvo_ch7xxx.c  | 25 +++++++++---------
 drivers/gpu/drm/i915/dvo_ivch.c    | 21 +++++++--------
 drivers/gpu/drm/i915/dvo_sil164.c  | 25 +++++++++---------
 drivers/gpu/drm/i915/dvo_tfp410.c  | 25 +++++++++---------
 drivers/gpu/drm/i915/intel_drv.h   | 11 ++++----
 drivers/gpu/drm/i915/intel_dvo.c   | 16 +++++-------
 drivers/gpu/drm/i915/intel_hdmi.c  |  2 +-
 drivers/gpu/drm/i915/intel_i2c.c   | 16 ++++++++----
 drivers/gpu/drm/i915/intel_modes.c | 14 +++++-----
 drivers/gpu/drm/i915/intel_sdvo.c  | 52 +++++++++++++++++++-------------------
 12 files changed, 118 insertions(+), 113 deletions(-)

diff --git a/drivers/gpu/drm/i915/dvo.h b/drivers/gpu/drm/i915/dvo.h
index e747ac42fe3a..288fc50627e2 100644
--- a/drivers/gpu/drm/i915/dvo.h
+++ b/drivers/gpu/drm/i915/dvo.h
@@ -37,7 +37,7 @@ struct intel_dvo_device {
 	/* GPIO register used for i2c bus to control this device */
 	u32 gpio;
 	int slave_addr;
-	struct intel_i2c_chan *i2c_bus;
+	struct i2c_adapter *i2c_bus;
 
 	const struct intel_dvo_dev_ops *dev_ops;
 	void *dev_priv;
@@ -52,7 +52,7 @@ struct intel_dvo_dev_ops {
 	 * Returns NULL if the device does not exist.
 	 */
 	bool (*init)(struct intel_dvo_device *dvo,
-		     struct intel_i2c_chan *i2cbus);
+		     struct i2c_adapter *i2cbus);
 
 	/*
 	 * Called to allow the output a chance to create properties after the
diff --git a/drivers/gpu/drm/i915/dvo_ch7017.c b/drivers/gpu/drm/i915/dvo_ch7017.c
index 03d4b4973b02..621815b531db 100644
--- a/drivers/gpu/drm/i915/dvo_ch7017.c
+++ b/drivers/gpu/drm/i915/dvo_ch7017.c
@@ -176,19 +176,20 @@ static void ch7017_dpms(struct intel_dvo_device *dvo, int mode);
 
 static bool ch7017_read(struct intel_dvo_device *dvo, int addr, uint8_t *val)
 {
-	struct intel_i2c_chan *i2cbus = dvo->i2c_bus;
+	struct i2c_adapter *adapter = dvo->i2c_bus;
+	struct intel_i2c_chan *i2cbus = container_of(adapter, struct intel_i2c_chan, adapter);
 	u8 out_buf[2];
 	u8 in_buf[2];
 
 	struct i2c_msg msgs[] = {
 		{
-			.addr = i2cbus->slave_addr,
+			.addr = dvo->slave_addr,
 			.flags = 0,
 			.len = 1,
 			.buf = out_buf,
 		},
 		{
-			.addr = i2cbus->slave_addr,
+			.addr = dvo->slave_addr,
 			.flags = I2C_M_RD,
 			.len = 1,
 			.buf = in_buf,
@@ -208,10 +209,11 @@ static bool ch7017_read(struct intel_dvo_device *dvo, int addr, uint8_t *val)
 
 static bool ch7017_write(struct intel_dvo_device *dvo, int addr, uint8_t val)
 {
-	struct intel_i2c_chan *i2cbus = dvo->i2c_bus;
+	struct i2c_adapter *adapter = dvo->i2c_bus;
+	struct intel_i2c_chan *i2cbus = container_of(adapter, struct intel_i2c_chan, adapter);
 	uint8_t out_buf[2];
 	struct i2c_msg msg = {
-		.addr = i2cbus->slave_addr,
+		.addr = dvo->slave_addr,
 		.flags = 0,
 		.len = 2,
 		.buf = out_buf,
@@ -228,8 +230,9 @@ static bool ch7017_write(struct intel_dvo_device *dvo, int addr, uint8_t val)
 
 /** Probes for a CH7017 on the given bus and slave address. */
 static bool ch7017_init(struct intel_dvo_device *dvo,
-			struct intel_i2c_chan *i2cbus)
+			struct i2c_adapter *adapter)
 {
+	struct intel_i2c_chan *i2cbus = container_of(adapter, struct intel_i2c_chan, adapter);
 	struct ch7017_priv *priv;
 	uint8_t val;
 
@@ -237,8 +240,7 @@ static bool ch7017_init(struct intel_dvo_device *dvo,
 	if (priv == NULL)
 		return false;
 
-	dvo->i2c_bus = i2cbus;
-	dvo->i2c_bus->slave_addr = dvo->slave_addr;
+	dvo->i2c_bus = adapter;
 	dvo->dev_priv = priv;
 
 	if (!ch7017_read(dvo, CH7017_DEVICE_ID, &val))
@@ -248,7 +250,7 @@ static bool ch7017_init(struct intel_dvo_device *dvo,
 	    val != CH7018_DEVICE_ID_VALUE &&
 	    val != CH7019_DEVICE_ID_VALUE) {
 		DRM_DEBUG("ch701x not detected, got %d: from %s Slave %d.\n",
-			  val, i2cbus->adapter.name,i2cbus->slave_addr);
+			  val, i2cbus->adapter.name,dvo->slave_addr);
 		goto fail;
 	}
 
diff --git a/drivers/gpu/drm/i915/dvo_ch7xxx.c b/drivers/gpu/drm/i915/dvo_ch7xxx.c
index d2fd95dbd034..a9b896289680 100644
--- a/drivers/gpu/drm/i915/dvo_ch7xxx.c
+++ b/drivers/gpu/drm/i915/dvo_ch7xxx.c
@@ -123,19 +123,20 @@ static char *ch7xxx_get_id(uint8_t vid)
 static bool ch7xxx_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 {
 	struct ch7xxx_priv *ch7xxx= dvo->dev_priv;
-	struct intel_i2c_chan *i2cbus = dvo->i2c_bus;
+	struct i2c_adapter *adapter = dvo->i2c_bus;
+	struct intel_i2c_chan *i2cbus = container_of(adapter, struct intel_i2c_chan, adapter);
 	u8 out_buf[2];
 	u8 in_buf[2];
 
 	struct i2c_msg msgs[] = {
 		{
-			.addr = i2cbus->slave_addr,
+			.addr = dvo->slave_addr,
 			.flags = 0,
 			.len = 1,
 			.buf = out_buf,
 		},
 		{
-			.addr = i2cbus->slave_addr,
+			.addr = dvo->slave_addr,
 			.flags = I2C_M_RD,
 			.len = 1,
 			.buf = in_buf,
@@ -152,7 +153,7 @@ static bool ch7xxx_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 
 	if (!ch7xxx->quiet) {
 		DRM_DEBUG("Unable to read register 0x%02x from %s:%02x.\n",
-			  addr, i2cbus->adapter.name, i2cbus->slave_addr);
+			  addr, i2cbus->adapter.name, dvo->slave_addr);
 	}
 	return false;
 }
@@ -161,10 +162,11 @@ static bool ch7xxx_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 static bool ch7xxx_writeb(struct intel_dvo_device *dvo, int addr, uint8_t ch)
 {
 	struct ch7xxx_priv *ch7xxx = dvo->dev_priv;
-	struct intel_i2c_chan *i2cbus = dvo->i2c_bus;
+	struct i2c_adapter *adapter = dvo->i2c_bus;
+	struct intel_i2c_chan *i2cbus = container_of(adapter, struct intel_i2c_chan, adapter);
 	uint8_t out_buf[2];
 	struct i2c_msg msg = {
-		.addr = i2cbus->slave_addr,
+		.addr = dvo->slave_addr,
 		.flags = 0,
 		.len = 2,
 		.buf = out_buf,
@@ -178,14 +180,14 @@ static bool ch7xxx_writeb(struct intel_dvo_device *dvo, int addr, uint8_t ch)
 
 	if (!ch7xxx->quiet) {
 		DRM_DEBUG("Unable to write register 0x%02x to %s:%d.\n",
-			  addr, i2cbus->adapter.name, i2cbus->slave_addr);
+			  addr, i2cbus->adapter.name, dvo->slave_addr);
 	}
 
 	return false;
 }
 
 static bool ch7xxx_init(struct intel_dvo_device *dvo,
-			struct intel_i2c_chan *i2cbus)
+			struct i2c_adapter *adapter)
 {
 	/* this will detect the CH7xxx chip on the specified i2c bus */
 	struct ch7xxx_priv *ch7xxx;
@@ -196,8 +198,7 @@ static bool ch7xxx_init(struct intel_dvo_device *dvo,
 	if (ch7xxx == NULL)
 		return false;
 
-	dvo->i2c_bus = i2cbus;
-	dvo->i2c_bus->slave_addr = dvo->slave_addr;
+	dvo->i2c_bus = adapter;
 	dvo->dev_priv = ch7xxx;
 	ch7xxx->quiet = true;
 
@@ -207,7 +208,7 @@ static bool ch7xxx_init(struct intel_dvo_device *dvo,
 	name = ch7xxx_get_id(vendor);
 	if (!name) {
 		DRM_DEBUG("ch7xxx not detected; got 0x%02x from %s slave %d.\n",
-			  vendor, i2cbus->adapter.name, i2cbus->slave_addr);
+			  vendor, adapter->name, dvo->slave_addr);
 		goto out;
 	}
 
@@ -217,7 +218,7 @@ static bool ch7xxx_init(struct intel_dvo_device *dvo,
 
 	if (device != CH7xxx_DID) {
 		DRM_DEBUG("ch7xxx not detected; got 0x%02x from %s slave %d.\n",
-			  vendor, i2cbus->adapter.name, i2cbus->slave_addr);
+			  vendor, adapter->name, dvo->slave_addr);
 		goto out;
 	}
 
diff --git a/drivers/gpu/drm/i915/dvo_ivch.c b/drivers/gpu/drm/i915/dvo_ivch.c
index 0c8d375e8e37..aa176f9921fe 100644
--- a/drivers/gpu/drm/i915/dvo_ivch.c
+++ b/drivers/gpu/drm/i915/dvo_ivch.c
@@ -169,13 +169,14 @@ static void ivch_dump_regs(struct intel_dvo_device *dvo);
 static bool ivch_read(struct intel_dvo_device *dvo, int addr, uint16_t *data)
 {
 	struct ivch_priv *priv = dvo->dev_priv;
-	struct intel_i2c_chan *i2cbus = dvo->i2c_bus;
+	struct i2c_adapter *adapter = dvo->i2c_bus;
+	struct intel_i2c_chan *i2cbus = container_of(adapter, struct intel_i2c_chan, adapter);
 	u8 out_buf[1];
 	u8 in_buf[2];
 
 	struct i2c_msg msgs[] = {
 		{
-			.addr = i2cbus->slave_addr,
+			.addr = dvo->slave_addr,
 			.flags = I2C_M_RD,
 			.len = 0,
 		},
@@ -186,7 +187,7 @@ static bool ivch_read(struct intel_dvo_device *dvo, int addr, uint16_t *data)
 			.buf = out_buf,
 		},
 		{
-			.addr = i2cbus->slave_addr,
+			.addr = dvo->slave_addr,
 			.flags = I2C_M_RD | I2C_M_NOSTART,
 			.len = 2,
 			.buf = in_buf,
@@ -202,7 +203,7 @@ static bool ivch_read(struct intel_dvo_device *dvo, int addr, uint16_t *data)
 
 	if (!priv->quiet) {
 		DRM_DEBUG("Unable to read register 0x%02x from %s:%02x.\n",
-			  addr, i2cbus->adapter.name, i2cbus->slave_addr);
+			  addr, i2cbus->adapter.name, dvo->slave_addr);
 	}
 	return false;
 }
@@ -211,10 +212,11 @@ static bool ivch_read(struct intel_dvo_device *dvo, int addr, uint16_t *data)
 static bool ivch_write(struct intel_dvo_device *dvo, int addr, uint16_t data)
 {
 	struct ivch_priv *priv = dvo->dev_priv;
-	struct intel_i2c_chan *i2cbus = dvo->i2c_bus;
+	struct i2c_adapter *adapter = dvo->i2c_bus;
+	struct intel_i2c_chan *i2cbus = container_of(adapter, struct intel_i2c_chan, adapter);
 	u8 out_buf[3];
 	struct i2c_msg msg = {
-		.addr = i2cbus->slave_addr,
+		.addr = dvo->slave_addr,
 		.flags = 0,
 		.len = 3,
 		.buf = out_buf,
@@ -229,7 +231,7 @@ static bool ivch_write(struct intel_dvo_device *dvo, int addr, uint16_t data)
 
 	if (!priv->quiet) {
 		DRM_DEBUG("Unable to write register 0x%02x to %s:%d.\n",
-			  addr, i2cbus->adapter.name, i2cbus->slave_addr);
+			  addr, i2cbus->adapter.name, dvo->slave_addr);
 	}
 
 	return false;
@@ -237,7 +239,7 @@ static bool ivch_write(struct intel_dvo_device *dvo, int addr, uint16_t data)
 
 /** Probes the given bus and slave address for an ivch */
 static bool ivch_init(struct intel_dvo_device *dvo,
-		      struct intel_i2c_chan *i2cbus)
+		      struct i2c_adapter *adapter)
 {
 	struct ivch_priv *priv;
 	uint16_t temp;
@@ -246,8 +248,7 @@ static bool ivch_init(struct intel_dvo_device *dvo,
 	if (priv == NULL)
 		return false;
 
-	dvo->i2c_bus = i2cbus;
-	dvo->i2c_bus->slave_addr = dvo->slave_addr;
+	dvo->i2c_bus = adapter;
 	dvo->dev_priv = priv;
 	priv->quiet = true;
 
diff --git a/drivers/gpu/drm/i915/dvo_sil164.c b/drivers/gpu/drm/i915/dvo_sil164.c
index 033a4bb070b2..e1c1f7341e5c 100644
--- a/drivers/gpu/drm/i915/dvo_sil164.c
+++ b/drivers/gpu/drm/i915/dvo_sil164.c
@@ -76,19 +76,20 @@ struct sil164_priv {
 static bool sil164_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 {
 	struct sil164_priv *sil = dvo->dev_priv;
-	struct intel_i2c_chan *i2cbus = dvo->i2c_bus;
+	struct i2c_adapter *adapter = dvo->i2c_bus;
+	struct intel_i2c_chan *i2cbus = container_of(adapter, struct intel_i2c_chan, adapter);
 	u8 out_buf[2];
 	u8 in_buf[2];
 
 	struct i2c_msg msgs[] = {
 		{
-			.addr = i2cbus->slave_addr,
+			.addr = dvo->slave_addr,
 			.flags = 0,
 			.len = 1,
 			.buf = out_buf,
 		},
 		{
-			.addr = i2cbus->slave_addr,
+			.addr = dvo->slave_addr,
 			.flags = I2C_M_RD,
 			.len = 1,
 			.buf = in_buf,
@@ -105,7 +106,7 @@ static bool sil164_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 
 	if (!sil->quiet) {
 		DRM_DEBUG("Unable to read register 0x%02x from %s:%02x.\n",
-			  addr, i2cbus->adapter.name, i2cbus->slave_addr);
+			  addr, i2cbus->adapter.name, dvo->slave_addr);
 	}
 	return false;
 }
@@ -113,10 +114,11 @@ static bool sil164_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 static bool sil164_writeb(struct intel_dvo_device *dvo, int addr, uint8_t ch)
 {
 	struct sil164_priv *sil= dvo->dev_priv;
-	struct intel_i2c_chan *i2cbus = dvo->i2c_bus;
+	struct i2c_adapter *adapter = dvo->i2c_bus;
+	struct intel_i2c_chan *i2cbus = container_of(adapter, struct intel_i2c_chan, adapter);
 	uint8_t out_buf[2];
 	struct i2c_msg msg = {
-		.addr = i2cbus->slave_addr,
+		.addr = dvo->slave_addr,
 		.flags = 0,
 		.len = 2,
 		.buf = out_buf,
@@ -130,7 +132,7 @@ static bool sil164_writeb(struct intel_dvo_device *dvo, int addr, uint8_t ch)
 
 	if (!sil->quiet) {
 		DRM_DEBUG("Unable to write register 0x%02x to %s:%d.\n",
-			  addr, i2cbus->adapter.name, i2cbus->slave_addr);
+			  addr, i2cbus->adapter.name, dvo->slave_addr);
 	}
 
 	return false;
@@ -138,7 +140,7 @@ static bool sil164_writeb(struct intel_dvo_device *dvo, int addr, uint8_t ch)
 
 /* Silicon Image 164 driver for chip on i2c bus */
 static bool sil164_init(struct intel_dvo_device *dvo,
-			struct intel_i2c_chan *i2cbus)
+			struct i2c_adapter *adapter)
 {
 	/* this will detect the SIL164 chip on the specified i2c bus */
 	struct sil164_priv *sil;
@@ -148,8 +150,7 @@ static bool sil164_init(struct intel_dvo_device *dvo,
 	if (sil == NULL)
 		return false;
 
-	dvo->i2c_bus = i2cbus;
-	dvo->i2c_bus->slave_addr = dvo->slave_addr;
+	dvo->i2c_bus = adapter;
 	dvo->dev_priv = sil;
 	sil->quiet = true;
 
@@ -158,7 +159,7 @@ static bool sil164_init(struct intel_dvo_device *dvo,
 
 	if (ch != (SIL164_VID & 0xff)) {
 		DRM_DEBUG("sil164 not detected got %d: from %s Slave %d.\n",
-			  ch, i2cbus->adapter.name, i2cbus->slave_addr);
+			  ch, adapter->name, dvo->slave_addr);
 		goto out;
 	}
 
@@ -167,7 +168,7 @@ static bool sil164_init(struct intel_dvo_device *dvo,
 
 	if (ch != (SIL164_DID & 0xff)) {
 		DRM_DEBUG("sil164 not detected got %d: from %s Slave %d.\n",
-			  ch, i2cbus->adapter.name, i2cbus->slave_addr);
+			  ch, adapter->name, dvo->slave_addr);
 		goto out;
 	}
 	sil->quiet = false;
diff --git a/drivers/gpu/drm/i915/dvo_tfp410.c b/drivers/gpu/drm/i915/dvo_tfp410.c
index 207fda806ebf..9ecc907384ec 100644
--- a/drivers/gpu/drm/i915/dvo_tfp410.c
+++ b/drivers/gpu/drm/i915/dvo_tfp410.c
@@ -101,19 +101,20 @@ struct tfp410_priv {
 static bool tfp410_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 {
 	struct tfp410_priv *tfp = dvo->dev_priv;
-	struct intel_i2c_chan *i2cbus = dvo->i2c_bus;
+	struct i2c_adapter *adapter = dvo->i2c_bus;
+	struct intel_i2c_chan *i2cbus = container_of(adapter, struct intel_i2c_chan, adapter);
 	u8 out_buf[2];
 	u8 in_buf[2];
 
 	struct i2c_msg msgs[] = {
 		{
-			.addr = i2cbus->slave_addr,
+			.addr = dvo->slave_addr,
 			.flags = 0,
 			.len = 1,
 			.buf = out_buf,
 		},
 		{
-			.addr = i2cbus->slave_addr,
+			.addr = dvo->slave_addr,
 			.flags = I2C_M_RD,
 			.len = 1,
 			.buf = in_buf,
@@ -130,7 +131,7 @@ static bool tfp410_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 
 	if (!tfp->quiet) {
 		DRM_DEBUG("Unable to read register 0x%02x from %s:%02x.\n",
-			  addr, i2cbus->adapter.name, i2cbus->slave_addr);
+			  addr, i2cbus->adapter.name, dvo->slave_addr);
 	}
 	return false;
 }
@@ -138,10 +139,11 @@ static bool tfp410_readb(struct intel_dvo_device *dvo, int addr, uint8_t *ch)
 static bool tfp410_writeb(struct intel_dvo_device *dvo, int addr, uint8_t ch)
 {
 	struct tfp410_priv *tfp = dvo->dev_priv;
-	struct intel_i2c_chan *i2cbus = dvo->i2c_bus;
+	struct i2c_adapter *adapter = dvo->i2c_bus;
+	struct intel_i2c_chan *i2cbus = container_of(adapter, struct intel_i2c_chan, adapter);
 	uint8_t out_buf[2];
 	struct i2c_msg msg = {
-		.addr = i2cbus->slave_addr,
+		.addr = dvo->slave_addr,
 		.flags = 0,
 		.len = 2,
 		.buf = out_buf,
@@ -155,7 +157,7 @@ static bool tfp410_writeb(struct intel_dvo_device *dvo, int addr, uint8_t ch)
 
 	if (!tfp->quiet) {
 		DRM_DEBUG("Unable to write register 0x%02x to %s:%d.\n",
-			  addr, i2cbus->adapter.name, i2cbus->slave_addr);
+			  addr, i2cbus->adapter.name, dvo->slave_addr);
 	}
 
 	return false;
@@ -174,7 +176,7 @@ static int tfp410_getid(struct intel_dvo_device *dvo, int addr)
 
 /* Ti TFP410 driver for chip on i2c bus */
 static bool tfp410_init(struct intel_dvo_device *dvo,
-			struct intel_i2c_chan *i2cbus)
+			struct i2c_adapter *adapter)
 {
 	/* this will detect the tfp410 chip on the specified i2c bus */
 	struct tfp410_priv *tfp;
@@ -184,20 +186,19 @@ static bool tfp410_init(struct intel_dvo_device *dvo,
 	if (tfp == NULL)
 		return false;
 
-	dvo->i2c_bus = i2cbus;
-	dvo->i2c_bus->slave_addr = dvo->slave_addr;
+	dvo->i2c_bus = adapter;
 	dvo->dev_priv = tfp;
 	tfp->quiet = true;
 
 	if ((id = tfp410_getid(dvo, TFP410_VID_LO)) != TFP410_VID) {
 		DRM_DEBUG("tfp410 not detected got VID %X: from %s Slave %d.\n",
-			  id, i2cbus->adapter.name, i2cbus->slave_addr);
+			  id, adapter->name, dvo->slave_addr);
 		goto out;
 	}
 
 	if ((id = tfp410_getid(dvo, TFP410_DID_LO)) != TFP410_DID) {
 		DRM_DEBUG("tfp410 not detected got DID %X: from %s Slave %d.\n",
-			  id, i2cbus->adapter.name, i2cbus->slave_addr);
+			  id, adapter->name, dvo->slave_addr);
 		goto out;
 	}
 	tfp->quiet = false;
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index cd4b9c5f715e..d89a2fed35af 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -65,7 +65,6 @@ struct intel_i2c_chan {
 	u32 reg; /* GPIO reg */
 	struct i2c_adapter adapter;
 	struct i2c_algo_bit_data algo;
-        u8 slave_addr;
 };
 
 struct intel_framebuffer {
@@ -79,8 +78,8 @@ struct intel_output {
 
 	struct drm_encoder enc;
 	int type;
-	struct intel_i2c_chan *i2c_bus; /* for control functions */
-	struct intel_i2c_chan *ddc_bus; /* for DDC only stuff */
+	struct i2c_adapter *i2c_bus;
+	struct i2c_adapter *ddc_bus;
 	bool load_detect_temp;
 	bool needs_tv_clock;
 	void *dev_priv;
@@ -104,9 +103,9 @@ struct intel_crtc {
 #define enc_to_intel_output(x) container_of(x, struct intel_output, enc)
 #define to_intel_framebuffer(x) container_of(x, struct intel_framebuffer, base)
 
-struct intel_i2c_chan *intel_i2c_create(struct drm_device *dev, const u32 reg,
-					const char *name);
-void intel_i2c_destroy(struct intel_i2c_chan *chan);
+struct i2c_adapter *intel_i2c_create(struct drm_device *dev, const u32 reg,
+				     const char *name);
+void intel_i2c_destroy(struct i2c_adapter *adapter);
 int intel_ddc_get_modes(struct intel_output *intel_output);
 extern bool intel_ddc_probe(struct intel_output *intel_output);
 void intel_i2c_quirk_set(struct drm_device *dev, bool enable);
diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index 1ee3007d6ec0..13bff20930e8 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -384,10 +384,9 @@ void intel_dvo_init(struct drm_device *dev)
 {
 	struct intel_output *intel_output;
 	struct intel_dvo_device *dvo;
-	struct intel_i2c_chan *i2cbus = NULL;
+	struct i2c_adapter *i2cbus = NULL;
 	int ret = 0;
 	int i;
-	int gpio_inited = 0;
 	int encoder_type = DRM_MODE_ENCODER_NONE;
 	intel_output = kzalloc (sizeof(struct intel_output), GFP_KERNEL);
 	if (!intel_output)
@@ -420,14 +419,11 @@ void intel_dvo_init(struct drm_device *dev)
 		 * It appears that everything is on GPIOE except for panels
 		 * on i830 laptops, which are on GPIOB (DVOA).
 		 */
-		if (gpio_inited != gpio) {
-			if (i2cbus != NULL)
-				intel_i2c_destroy(i2cbus);
-			if (!(i2cbus = intel_i2c_create(dev, gpio,
-				gpio == GPIOB ? "DVOI2C_B" : "DVOI2C_E"))) {
-				continue;
-			}
-			gpio_inited = gpio;
+		if (i2cbus != NULL)
+			intel_i2c_destroy(i2cbus);
+		if (!(i2cbus = intel_i2c_create(dev, gpio,
+			gpio == GPIOB ? "DVOI2C_B" : "DVOI2C_E"))) {
+			continue;
 		}
 
 		if (dvo->dev_ops!= NULL)
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 2495359ea8de..fbe96005fa1e 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -139,7 +139,7 @@ intel_hdmi_edid_detect(struct drm_connector *connector)
 	enum drm_connector_status status = connector_status_disconnected;
 
 	edid = drm_get_edid(&intel_output->base,
-			    &intel_output->ddc_bus->adapter);
+			    intel_output->ddc_bus);
 	hdmi_priv->has_hdmi_sink = false;
 	if (edid) {
 		if (edid->digital) {
diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c
index f7061f68d050..62b8bead7652 100644
--- a/drivers/gpu/drm/i915/intel_i2c.c
+++ b/drivers/gpu/drm/i915/intel_i2c.c
@@ -124,6 +124,7 @@ static void set_data(void *data, int state_high)
  * @output: driver specific output device
  * @reg: GPIO reg to use
  * @name: name for this bus
+ * @slave_addr: slave address (if fixed)
  *
  * Creates and registers a new i2c bus with the Linux i2c layer, for use
  * in output probing and control (e.g. DDC or SDVO control functions).
@@ -139,8 +140,8 @@ static void set_data(void *data, int state_high)
  *   %GPIOH
  * see PRM for details on how these different busses are used.
  */
-struct intel_i2c_chan *intel_i2c_create(struct drm_device *dev, const u32 reg,
-					const char *name)
+struct i2c_adapter *intel_i2c_create(struct drm_device *dev, const u32 reg,
+				     const char *name)
 {
 	struct intel_i2c_chan *chan;
 
@@ -174,7 +175,7 @@ struct intel_i2c_chan *intel_i2c_create(struct drm_device *dev, const u32 reg,
 	intel_i2c_quirk_set(dev, false);
 	udelay(20);
 
-	return chan;
+	return &chan->adapter;
 
 out_free:
 	kfree(chan);
@@ -187,11 +188,16 @@ out_free:
  *
  * Unregister the adapter from the i2c layer, then free the structure.
  */
-void intel_i2c_destroy(struct intel_i2c_chan *chan)
+void intel_i2c_destroy(struct i2c_adapter *adapter)
 {
-	if (!chan)
+	struct intel_i2c_chan *chan;
+
+	if (!adapter)
 		return;
 
+	chan = container_of(adapter,
+			    struct intel_i2c_chan,
+			    adapter);
 	i2c_del_adapter(&chan->adapter);
 	kfree(chan);
 }
diff --git a/drivers/gpu/drm/i915/intel_modes.c b/drivers/gpu/drm/i915/intel_modes.c
index e0910fefce87..67e2f4632a24 100644
--- a/drivers/gpu/drm/i915/intel_modes.c
+++ b/drivers/gpu/drm/i915/intel_modes.c
@@ -53,10 +53,9 @@ bool intel_ddc_probe(struct intel_output *intel_output)
 		}
 	};
 
-	intel_i2c_quirk_set(intel_output->ddc_bus->drm_dev, true);
-	ret = i2c_transfer(&intel_output->ddc_bus->adapter, msgs, 2);
-	intel_i2c_quirk_set(intel_output->ddc_bus->drm_dev, false);
-
+	intel_i2c_quirk_set(intel_output->base.dev, true);
+	ret = i2c_transfer(intel_output->ddc_bus, msgs, 2);
+	intel_i2c_quirk_set(intel_output->base.dev, false);
 	if (ret == 2)
 		return true;
 
@@ -74,10 +73,9 @@ int intel_ddc_get_modes(struct intel_output *intel_output)
 	struct edid *edid;
 	int ret = 0;
 
-	intel_i2c_quirk_set(intel_output->ddc_bus->drm_dev, true);
-	edid = drm_get_edid(&intel_output->base,
-			    &intel_output->ddc_bus->adapter);
-	intel_i2c_quirk_set(intel_output->ddc_bus->drm_dev, false);
+	intel_i2c_quirk_set(intel_output->base.dev, true);
+	edid = drm_get_edid(&intel_output->base, intel_output->ddc_bus);
+	intel_i2c_quirk_set(intel_output->base.dev, false);
 	if (edid) {
 		drm_mode_connector_update_edid_property(&intel_output->base,
 							edid);
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 9a00adb3a508..13c39c827ebf 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -38,8 +38,8 @@
 #undef SDVO_DEBUG
 #define I915_SDVO	"i915_sdvo"
 struct intel_sdvo_priv {
-	struct intel_i2c_chan *i2c_bus;
-	int slaveaddr;
+	struct i2c_adapter *i2c_bus;
+	u8 slave_addr;
 
 	/* Register for the SDVO device: SDVOB or SDVOC */
 	int output_device;
@@ -146,13 +146,13 @@ static bool intel_sdvo_read_byte(struct intel_output *intel_output, u8 addr,
 
 	struct i2c_msg msgs[] = {
 		{
-			.addr = sdvo_priv->i2c_bus->slave_addr,
+			.addr = sdvo_priv->slave_addr >> 1,
 			.flags = 0,
 			.len = 1,
 			.buf = out_buf,
 		},
 		{
-			.addr = sdvo_priv->i2c_bus->slave_addr,
+			.addr = sdvo_priv->slave_addr >> 1,
 			.flags = I2C_M_RD,
 			.len = 1,
 			.buf = buf,
@@ -162,7 +162,7 @@ static bool intel_sdvo_read_byte(struct intel_output *intel_output, u8 addr,
 	out_buf[0] = addr;
 	out_buf[1] = 0;
 
-	if ((ret = i2c_transfer(&sdvo_priv->i2c_bus->adapter, msgs, 2)) == 2)
+	if ((ret = i2c_transfer(sdvo_priv->i2c_bus, msgs, 2)) == 2)
 	{
 		*ch = buf[0];
 		return true;
@@ -175,10 +175,11 @@ static bool intel_sdvo_read_byte(struct intel_output *intel_output, u8 addr,
 static bool intel_sdvo_write_byte(struct intel_output *intel_output, int addr,
 				  u8 ch)
 {
+	struct intel_sdvo_priv *sdvo_priv = intel_output->dev_priv;
 	u8 out_buf[2];
 	struct i2c_msg msgs[] = {
 		{
-			.addr = intel_output->i2c_bus->slave_addr,
+			.addr = sdvo_priv->slave_addr >> 1,
 			.flags = 0,
 			.len = 2,
 			.buf = out_buf,
@@ -188,7 +189,7 @@ static bool intel_sdvo_write_byte(struct intel_output *intel_output, int addr,
 	out_buf[0] = addr;
 	out_buf[1] = ch;
 
-	if (i2c_transfer(&intel_output->i2c_bus->adapter, msgs, 1) == 1)
+	if (i2c_transfer(intel_output->i2c_bus, msgs, 1) == 1)
 	{
 		return true;
 	}
@@ -1371,7 +1372,7 @@ intel_sdvo_hdmi_sink_detect(struct drm_connector *connector)
 
 	intel_sdvo_set_control_bus_switch(intel_output, sdvo_priv->ddc_bus);
 	edid = drm_get_edid(&intel_output->base,
-			    &intel_output->ddc_bus->adapter);
+			    intel_output->ddc_bus);
 	if (edid != NULL) {
 		sdvo_priv->is_hdmi = drm_detect_hdmi_monitor(edid);
 		kfree(edid);
@@ -1709,7 +1710,7 @@ intel_sdvo_chan_to_intel_output(struct intel_i2c_chan *chan)
 
 	list_for_each_entry(connector,
 			&dev->mode_config.connector_list, head) {
-		if (to_intel_output(connector)->ddc_bus == chan) {
+		if (to_intel_output(connector)->ddc_bus == &chan->adapter) {
 			intel_output = to_intel_output(connector);
 			break;
 		}
@@ -1723,7 +1724,7 @@ static int intel_sdvo_master_xfer(struct i2c_adapter *i2c_adap,
 	struct intel_output *intel_output;
 	struct intel_sdvo_priv *sdvo_priv;
 	struct i2c_algo_bit_data *algo_data;
-	struct i2c_algorithm *algo;
+	const struct i2c_algorithm *algo;
 
 	algo_data = (struct i2c_algo_bit_data *)i2c_adap->algo_data;
 	intel_output =
@@ -1733,7 +1734,7 @@ static int intel_sdvo_master_xfer(struct i2c_adapter *i2c_adap,
 		return -EINVAL;
 
 	sdvo_priv = intel_output->dev_priv;
-	algo = (struct i2c_algorithm *)intel_output->i2c_bus->adapter.algo;
+	algo = intel_output->i2c_bus->algo;
 
 	intel_sdvo_set_control_bus_switch(intel_output, sdvo_priv->ddc_bus);
 	return algo->master_xfer(i2c_adap, msgs, num);
@@ -1785,12 +1786,13 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 	struct drm_connector *connector;
 	struct intel_output *intel_output;
 	struct intel_sdvo_priv *sdvo_priv;
-	struct intel_i2c_chan *i2cbus = NULL;
-	struct intel_i2c_chan *ddcbus = NULL;
+	struct i2c_adapter *i2cbus = NULL;
+	struct i2c_adapter *ddcbus = NULL;
+
 	int connector_type;
 	u8 ch[0x40];
 	int i;
-	int encoder_type, output_id;
+	int encoder_type;
 	u8 slave_addr;
 
 	intel_output = kcalloc(sizeof(struct intel_output)+sizeof(struct intel_sdvo_priv), 1, GFP_KERNEL);
@@ -1802,25 +1804,23 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 	intel_output->type = INTEL_OUTPUT_SDVO;
 
 	/* setup the DDC bus. */
-	if (output_device == SDVOB)
+	if (output_device == SDVOB) {
 		i2cbus = intel_i2c_create(dev, GPIOE, "SDVOCTRL_E for SDVOB");
-	else
+		slave_addr = 0x38;
+	} else {
 		i2cbus = intel_i2c_create(dev, GPIOE, "SDVOCTRL_E for SDVOC");
-
+		slave_addr = 0x39;
+	}
+		
 	if (!i2cbus)
 		goto err_inteloutput;
 
 	slave_addr = intel_sdvo_get_slave_addr(dev, output_device);
 	sdvo_priv->i2c_bus = i2cbus;
+	sdvo_priv->slave_addr = slave_addr;
 
-	if (output_device == SDVOB) {
-		output_id = 1;
-	} else {
-		output_id = 2;
-	}
-	sdvo_priv->i2c_bus->slave_addr = slave_addr >> 1;
 	sdvo_priv->output_device = output_device;
-	intel_output->i2c_bus = i2cbus;
+	intel_output->i2c_bus = sdvo_priv->i2c_bus;
 	intel_output->dev_priv = sdvo_priv;
 
 	/* Read the regs to test if we can talk to the device */
@@ -1843,8 +1843,8 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 		goto err_i2c;
 
 	intel_sdvo_i2c_bit_algo.functionality =
-		intel_output->i2c_bus->adapter.algo->functionality;
-	ddcbus->adapter.algo = &intel_sdvo_i2c_bit_algo;
+		intel_output->i2c_bus->algo->functionality;
+	ddcbus->algo = &intel_sdvo_i2c_bit_algo;
 	intel_output->ddc_bus = ddcbus;
 
 	/* In defaut case sdvo lvds is false */
-- 
cgit v1.2.3-59-g8ed1b


From 308cd3a2e505b0d15f2852e8db5d648b60a6313b Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Sun, 14 Jun 2009 11:56:18 -0700
Subject: drm/i915: Clean up SDVO i2c handling

Eliminate the copy of i2c_bus in sdvo_priv.
Eliminate local copies of i2c_bus and ddcbus.
Eliminate unused settings of slave_addr.

Signed-off-by: Keith Packard <keithp@keithp.com>
---
 drivers/gpu/drm/i915/intel_sdvo.c | 54 +++++++++++++++------------------------
 1 file changed, 21 insertions(+), 33 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 13c39c827ebf..f03473779feb 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -38,7 +38,6 @@
 #undef SDVO_DEBUG
 #define I915_SDVO	"i915_sdvo"
 struct intel_sdvo_priv {
-	struct i2c_adapter *i2c_bus;
 	u8 slave_addr;
 
 	/* Register for the SDVO device: SDVOB or SDVOC */
@@ -162,7 +161,7 @@ static bool intel_sdvo_read_byte(struct intel_output *intel_output, u8 addr,
 	out_buf[0] = addr;
 	out_buf[1] = 0;
 
-	if ((ret = i2c_transfer(sdvo_priv->i2c_bus, msgs, 2)) == 2)
+	if ((ret = i2c_transfer(intel_output->i2c_bus, msgs, 2)) == 2)
 	{
 		*ch = buf[0];
 		return true;
@@ -1370,7 +1369,6 @@ intel_sdvo_hdmi_sink_detect(struct drm_connector *connector)
 	struct intel_sdvo_priv *sdvo_priv = intel_output->dev_priv;
 	struct edid *edid = NULL;
 
-	intel_sdvo_set_control_bus_switch(intel_output, sdvo_priv->ddc_bus);
 	edid = drm_get_edid(&intel_output->base,
 			    intel_output->ddc_bus);
 	if (edid != NULL) {
@@ -1550,7 +1548,6 @@ static void intel_sdvo_get_tv_modes(struct drm_connector *connector)
 static void intel_sdvo_get_lvds_modes(struct drm_connector *connector)
 {
 	struct intel_output *intel_output = to_intel_output(connector);
-	struct intel_sdvo_priv *sdvo_priv = intel_output->dev_priv;
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
 
 	/*
@@ -1558,8 +1555,6 @@ static void intel_sdvo_get_lvds_modes(struct drm_connector *connector)
 	 * Assume that the preferred modes are
 	 * arranged in priority order.
 	 */
-	/* set the bus switch and get the modes */
-	intel_sdvo_set_control_bus_switch(intel_output, sdvo_priv->ddc_bus);
 	intel_ddc_get_modes(intel_output);
 	if (list_empty(&connector->probed_modes) == false)
 		return;
@@ -1786,14 +1781,11 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 	struct drm_connector *connector;
 	struct intel_output *intel_output;
 	struct intel_sdvo_priv *sdvo_priv;
-	struct i2c_adapter *i2cbus = NULL;
-	struct i2c_adapter *ddcbus = NULL;
 
 	int connector_type;
 	u8 ch[0x40];
 	int i;
 	int encoder_type;
-	u8 slave_addr;
 
 	intel_output = kcalloc(sizeof(struct intel_output)+sizeof(struct intel_sdvo_priv), 1, GFP_KERNEL);
 	if (!intel_output) {
@@ -1801,27 +1793,24 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 	}
 
 	sdvo_priv = (struct intel_sdvo_priv *)(intel_output + 1);
+	sdvo_priv->output_device = output_device;
+
+	intel_output->dev_priv = sdvo_priv;
 	intel_output->type = INTEL_OUTPUT_SDVO;
 
 	/* setup the DDC bus. */
-	if (output_device == SDVOB) {
-		i2cbus = intel_i2c_create(dev, GPIOE, "SDVOCTRL_E for SDVOB");
-		slave_addr = 0x38;
-	} else {
-		i2cbus = intel_i2c_create(dev, GPIOE, "SDVOCTRL_E for SDVOC");
-		slave_addr = 0x39;
-	}
-		
-	if (!i2cbus)
+	if (output_device == SDVOB)
+		intel_output->i2c_bus = intel_i2c_create(dev, GPIOE, "SDVOCTRL_E for SDVOB");
+	else
+		intel_output->i2c_bus = intel_i2c_create(dev, GPIOE, "SDVOCTRL_E for SDVOC");
+
+	if (!intel_output->i2c_bus)
 		goto err_inteloutput;
 
-	slave_addr = intel_sdvo_get_slave_addr(dev, output_device);
-	sdvo_priv->i2c_bus = i2cbus;
-	sdvo_priv->slave_addr = slave_addr;
+	sdvo_priv->slave_addr = intel_sdvo_get_slave_addr(dev, output_device);
 
-	sdvo_priv->output_device = output_device;
-	intel_output->i2c_bus = sdvo_priv->i2c_bus;
-	intel_output->dev_priv = sdvo_priv;
+	/* Save the bit-banging i2c functionality for use by the DDC wrapper */
+	intel_sdvo_i2c_bit_algo.functionality = intel_output->i2c_bus->algo->functionality;
 
 	/* Read the regs to test if we can talk to the device */
 	for (i = 0; i < 0x40; i++) {
@@ -1835,17 +1824,15 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 
 	/* setup the DDC bus. */
 	if (output_device == SDVOB)
-		ddcbus = intel_i2c_create(dev, GPIOE, "SDVOB DDC BUS");
+		intel_output->ddc_bus = intel_i2c_create(dev, GPIOE, "SDVOB DDC BUS");
 	else
-		ddcbus = intel_i2c_create(dev, GPIOE, "SDVOC DDC BUS");
+		intel_output->ddc_bus = intel_i2c_create(dev, GPIOE, "SDVOC DDC BUS");
 
-	if (ddcbus == NULL)
+	if (intel_output->ddc_bus == NULL)
 		goto err_i2c;
 
-	intel_sdvo_i2c_bit_algo.functionality =
-		intel_output->i2c_bus->algo->functionality;
-	ddcbus->algo = &intel_sdvo_i2c_bit_algo;
-	intel_output->ddc_bus = ddcbus;
+	/* Wrap with our custom algo which switches to DDC mode */
+	intel_output->ddc_bus->algo = &intel_sdvo_i2c_bit_algo;
 
 	/* In defaut case sdvo lvds is false */
 	sdvo_priv->is_lvds = false;
@@ -1965,9 +1952,10 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 	return true;
 
 err_i2c:
-	if (ddcbus != NULL)
+	if (intel_output->ddc_bus != NULL)
 		intel_i2c_destroy(intel_output->ddc_bus);
-	intel_i2c_destroy(intel_output->i2c_bus);
+	if (intel_output->i2c_bus != NULL)
+		intel_i2c_destroy(intel_output->i2c_bus);
 err_inteloutput:
 	kfree(intel_output);
 
-- 
cgit v1.2.3-59-g8ed1b


From c31c4ba3437d98efa19710e30d694a1cfdf87aa5 Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Wed, 6 May 2009 11:48:58 -0700
Subject: drm/i915: add per-output hotplug callback for KMS

This allows each output to deal with plug/unplug events as needed.

Signed-off-by: Keith Packard <keithp@keithp.com>
---
 drivers/gpu/drm/i915/i915_irq.c  | 12 +++++++++++-
 drivers/gpu/drm/i915/intel_drv.h |  1 +
 2 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index b86b7b7130c6..228546f6eaa4 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -232,7 +232,17 @@ static void i915_hotplug_work_func(struct work_struct *work)
 	drm_i915_private_t *dev_priv = container_of(work, drm_i915_private_t,
 						    hotplug_work);
 	struct drm_device *dev = dev_priv->dev;
-
+	struct drm_mode_config *mode_config = &dev->mode_config;
+	struct drm_connector *connector;
+
+	if (mode_config->num_connector) {
+		list_for_each_entry(connector, &mode_config->connector_list, head) {
+			struct intel_output *intel_output = to_intel_output(connector);
+	
+			if (intel_output->hot_plug)
+				(*intel_output->hot_plug) (intel_output);
+		}
+	}
 	/* Just fire off a uevent and let userspace tell us what to do */
 	drm_sysfs_hotplug_event(dev);
 }
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index d89a2fed35af..c5858792c806 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -83,6 +83,7 @@ struct intel_output {
 	bool load_detect_temp;
 	bool needs_tv_clock;
 	void *dev_priv;
+	void (*hot_plug)(struct intel_output *);
 };
 
 struct intel_crtc {
-- 
cgit v1.2.3-59-g8ed1b


From a4fc5ed69817c73e32571ad7837bb707f9890009 Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Tue, 7 Apr 2009 16:16:42 -0700
Subject: drm/i915: Add Display Port support

Signed-off-by: Keith Packard <keithp@keithp.com>
---
 drivers/gpu/drm/i915/Makefile        |    2 +
 drivers/gpu/drm/i915/i915_drv.h      |   12 +
 drivers/gpu/drm/i915/i915_suspend.c  |   34 +-
 drivers/gpu/drm/i915/intel_display.c |  107 +++-
 drivers/gpu/drm/i915/intel_dp.c      | 1098 ++++++++++++++++++++++++++++++++++
 drivers/gpu/drm/i915/intel_dp.h      |  144 +++++
 drivers/gpu/drm/i915/intel_dp_i2c.c  |  272 +++++++++
 drivers/gpu/drm/i915/intel_drv.h     |    5 +
 8 files changed, 1668 insertions(+), 6 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/intel_dp.c
 create mode 100644 drivers/gpu/drm/i915/intel_dp.h
 create mode 100644 drivers/gpu/drm/i915/intel_dp_i2c.c

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 51c5a050aa73..30d6b99fb302 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -13,6 +13,8 @@ i915-y := i915_drv.o i915_dma.o i915_irq.o i915_mem.o \
 	  intel_crt.o \
 	  intel_lvds.o \
 	  intel_bios.o \
+	  intel_dp.o \
+	  intel_dp_i2c.o \
 	  intel_hdmi.o \
 	  intel_sdvo.o \
 	  intel_modes.o \
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 7a84f04e8439..bb4c2d387b6c 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -306,6 +306,17 @@ typedef struct drm_i915_private {
 	u32 saveCURBPOS;
 	u32 saveCURBBASE;
 	u32 saveCURSIZE;
+	u32 saveDP_B;
+	u32 saveDP_C;
+	u32 saveDP_D;
+	u32 savePIPEA_GMCH_DATA_M;
+	u32 savePIPEB_GMCH_DATA_M;
+	u32 savePIPEA_GMCH_DATA_N;
+	u32 savePIPEB_GMCH_DATA_N;
+	u32 savePIPEA_DP_LINK_M;
+	u32 savePIPEB_DP_LINK_M;
+	u32 savePIPEA_DP_LINK_N;
+	u32 savePIPEB_DP_LINK_N;
 
 	struct {
 		struct drm_mm gtt_space;
@@ -857,6 +868,7 @@ extern int i915_wait_ring(struct drm_device * dev, int n, const char *caller);
 #define HAS_128_BYTE_Y_TILING(dev) (IS_I9XX(dev) && !(IS_I915G(dev) || \
 						      IS_I915GM(dev)))
 #define SUPPORTS_INTEGRATED_HDMI(dev)	(IS_G4X(dev) || IS_IGDNG(dev))
+#define SUPPORTS_INTEGRATED_DP(dev)	(IS_G4X(dev) || IS_IGDNG(dev))
 #define I915_HAS_HOTPLUG(dev) (IS_I945G(dev) || IS_I945GM(dev) || IS_I965G(dev))
 
 #define PRIMARY_RINGBUFFER_SIZE         (128*1024)
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index a98e2831ed31..8d8e083d14ab 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -322,6 +322,20 @@ int i915_save_state(struct drm_device *dev)
 	dev_priv->savePP_OFF_DELAYS = I915_READ(PP_OFF_DELAYS);
 	dev_priv->savePP_DIVISOR = I915_READ(PP_DIVISOR);
 
+	/* Display Port state */
+	if (SUPPORTS_INTEGRATED_DP(dev)) {
+		dev_priv->saveDP_B = I915_READ(DP_B);
+		dev_priv->saveDP_C = I915_READ(DP_C);
+		dev_priv->saveDP_D = I915_READ(DP_D);
+		dev_priv->savePIPEA_GMCH_DATA_M = I915_READ(PIPEA_GMCH_DATA_M);
+		dev_priv->savePIPEB_GMCH_DATA_M = I915_READ(PIPEB_GMCH_DATA_M);
+		dev_priv->savePIPEA_GMCH_DATA_N = I915_READ(PIPEA_GMCH_DATA_N);
+		dev_priv->savePIPEB_GMCH_DATA_N = I915_READ(PIPEB_GMCH_DATA_N);
+		dev_priv->savePIPEA_DP_LINK_M = I915_READ(PIPEA_DP_LINK_M);
+		dev_priv->savePIPEB_DP_LINK_M = I915_READ(PIPEB_DP_LINK_M);
+		dev_priv->savePIPEA_DP_LINK_N = I915_READ(PIPEA_DP_LINK_N);
+		dev_priv->savePIPEB_DP_LINK_N = I915_READ(PIPEB_DP_LINK_N);
+	}
 	/* FIXME: save TV & SDVO state */
 
 	/* FBC state */
@@ -404,7 +418,19 @@ int i915_restore_state(struct drm_device *dev)
 			for (i = 0; i < 8; i++)
 				I915_WRITE(FENCE_REG_945_8 + (i * 4), dev_priv->saveFENCE[i+8]);
 	}
-
+	
+	/* Display port ratios (must be done before clock is set) */
+	if (SUPPORTS_INTEGRATED_DP(dev)) {
+		I915_WRITE(PIPEA_GMCH_DATA_M, dev_priv->savePIPEA_GMCH_DATA_M);
+		I915_WRITE(PIPEB_GMCH_DATA_M, dev_priv->savePIPEB_GMCH_DATA_M);
+		I915_WRITE(PIPEA_GMCH_DATA_N, dev_priv->savePIPEA_GMCH_DATA_N);
+		I915_WRITE(PIPEB_GMCH_DATA_N, dev_priv->savePIPEB_GMCH_DATA_N);
+		I915_WRITE(PIPEA_DP_LINK_M, dev_priv->savePIPEA_DP_LINK_M);
+		I915_WRITE(PIPEB_DP_LINK_M, dev_priv->savePIPEB_DP_LINK_M);
+		I915_WRITE(PIPEA_DP_LINK_N, dev_priv->savePIPEA_DP_LINK_N);
+		I915_WRITE(PIPEB_DP_LINK_N, dev_priv->savePIPEB_DP_LINK_N);
+	}
+	
 	/* Pipe & plane A info */
 	/* Prime the clock */
 	if (dev_priv->saveDPLL_A & DPLL_VCO_ENABLE) {
@@ -518,6 +544,12 @@ int i915_restore_state(struct drm_device *dev)
 	I915_WRITE(PP_DIVISOR, dev_priv->savePP_DIVISOR);
 	I915_WRITE(PP_CONTROL, dev_priv->savePP_CONTROL);
 
+	/* Display Port state */
+	if (SUPPORTS_INTEGRATED_DP(dev)) {
+		I915_WRITE(DP_B, dev_priv->saveDP_B);
+		I915_WRITE(DP_C, dev_priv->saveDP_C);
+		I915_WRITE(DP_D, dev_priv->saveDP_D);
+	}
 	/* FIXME: restore TV & SDVO state */
 
 	/* FBC info */
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 3e1c78162119..5af55aa0d7a6 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -29,6 +29,7 @@
 #include "intel_drv.h"
 #include "i915_drm.h"
 #include "i915_drv.h"
+#include "intel_dp.h"
 
 #include "drm_crtc_helper.h"
 
@@ -135,10 +136,11 @@ struct intel_limit {
 #define INTEL_LIMIT_G4X_HDMI_DAC   5
 #define INTEL_LIMIT_G4X_SINGLE_CHANNEL_LVDS   6
 #define INTEL_LIMIT_G4X_DUAL_CHANNEL_LVDS   7
-#define INTEL_LIMIT_IGD_SDVO_DAC    8
-#define INTEL_LIMIT_IGD_LVDS	    9
-#define INTEL_LIMIT_IGDNG_SDVO_DAC  10
-#define INTEL_LIMIT_IGDNG_LVDS	    11
+#define INTEL_LIMIT_G4X_DISPLAY_PORT	8
+#define INTEL_LIMIT_IGD_SDVO_DAC    9
+#define INTEL_LIMIT_IGD_LVDS	    10
+#define INTEL_LIMIT_IGDNG_SDVO_DAC  11
+#define INTEL_LIMIT_IGDNG_LVDS	    12
 
 /*The parameter is for SDVO on G4x platform*/
 #define G4X_DOT_SDVO_MIN           25000
@@ -218,6 +220,25 @@ struct intel_limit {
 #define G4X_P2_DUAL_CHANNEL_LVDS_FAST           7
 #define G4X_P2_DUAL_CHANNEL_LVDS_LIMIT          0
 
+/*The parameter is for DISPLAY PORT on G4x platform*/
+#define G4X_DOT_DISPLAY_PORT_MIN           161670
+#define G4X_DOT_DISPLAY_PORT_MAX           227000
+#define G4X_N_DISPLAY_PORT_MIN             1
+#define G4X_N_DISPLAY_PORT_MAX             2
+#define G4X_M_DISPLAY_PORT_MIN             97
+#define G4X_M_DISPLAY_PORT_MAX             108
+#define G4X_M1_DISPLAY_PORT_MIN            0x10
+#define G4X_M1_DISPLAY_PORT_MAX            0x12
+#define G4X_M2_DISPLAY_PORT_MIN            0x05
+#define G4X_M2_DISPLAY_PORT_MAX            0x06
+#define G4X_P_DISPLAY_PORT_MIN             10
+#define G4X_P_DISPLAY_PORT_MAX             20
+#define G4X_P1_DISPLAY_PORT_MIN            1
+#define G4X_P1_DISPLAY_PORT_MAX            2
+#define G4X_P2_DISPLAY_PORT_SLOW           10
+#define G4X_P2_DISPLAY_PORT_FAST           10
+#define G4X_P2_DISPLAY_PORT_LIMIT          0
+
 /* IGDNG */
 /* as we calculate clock using (register_value + 2) for
    N/M1/M2, so here the range value for them is (actual_value-2).
@@ -256,6 +277,10 @@ static bool
 intel_igdng_find_best_PLL(const intel_limit_t *limit, struct drm_crtc *crtc,
 			int target, int refclk, intel_clock_t *best_clock);
 
+static bool
+intel_find_pll_g4x_dp(const intel_limit_t *, struct drm_crtc *crtc,
+		      int target, int refclk, intel_clock_t *best_clock);
+
 static const intel_limit_t intel_limits[] = {
     { /* INTEL_LIMIT_I8XX_DVO_DAC */
         .dot = { .min = I8XX_DOT_MIN,		.max = I8XX_DOT_MAX },
@@ -389,6 +414,28 @@ static const intel_limit_t intel_limits[] = {
 	},
 	.find_pll = intel_g4x_find_best_PLL,
     },
+    {   /* INTEL_LIMIT_G4X_DISPLAY_PORT */
+        .dot = { .min = G4X_DOT_DISPLAY_PORT_MIN,
+                 .max = G4X_DOT_DISPLAY_PORT_MAX },
+        .vco = { .min = G4X_VCO_MIN,
+                 .max = G4X_VCO_MAX},
+        .n   = { .min = G4X_N_DISPLAY_PORT_MIN,
+                 .max = G4X_N_DISPLAY_PORT_MAX },
+        .m   = { .min = G4X_M_DISPLAY_PORT_MIN,
+                 .max = G4X_M_DISPLAY_PORT_MAX },
+        .m1  = { .min = G4X_M1_DISPLAY_PORT_MIN,
+                 .max = G4X_M1_DISPLAY_PORT_MAX },
+        .m2  = { .min = G4X_M2_DISPLAY_PORT_MIN,
+                 .max = G4X_M2_DISPLAY_PORT_MAX },
+        .p   = { .min = G4X_P_DISPLAY_PORT_MIN,
+                 .max = G4X_P_DISPLAY_PORT_MAX },
+        .p1  = { .min = G4X_P1_DISPLAY_PORT_MIN,
+                 .max = G4X_P1_DISPLAY_PORT_MAX},
+        .p2  = { .dot_limit = G4X_P2_DISPLAY_PORT_LIMIT,
+                 .p2_slow = G4X_P2_DISPLAY_PORT_SLOW,
+                 .p2_fast = G4X_P2_DISPLAY_PORT_FAST },
+        .find_pll = intel_find_pll_g4x_dp,
+    },
     { /* INTEL_LIMIT_IGD_SDVO */
         .dot = { .min = I9XX_DOT_MIN,		.max = I9XX_DOT_MAX},
         .vco = { .min = IGD_VCO_MIN,		.max = IGD_VCO_MAX },
@@ -478,6 +525,8 @@ static const intel_limit_t *intel_g4x_limit(struct drm_crtc *crtc)
 		limit = &intel_limits[INTEL_LIMIT_G4X_HDMI_DAC];
 	} else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_SDVO)) {
 		limit = &intel_limits[INTEL_LIMIT_G4X_SDVO];
+	} else if (intel_pipe_has_type (crtc, INTEL_OUTPUT_DISPLAYPORT)) {
+		limit = &intel_limits[INTEL_LIMIT_G4X_DISPLAY_PORT];
 	} else /* The option is for other outputs */
 		limit = &intel_limits[INTEL_LIMIT_I9XX_SDVO_DAC];
 
@@ -764,6 +813,35 @@ out:
 	return found;
 }
 
+/* DisplayPort has only two frequencies, 162MHz and 270MHz */
+static bool
+intel_find_pll_g4x_dp(const intel_limit_t *limit, struct drm_crtc *crtc,
+		      int target, int refclk, intel_clock_t *best_clock)
+{
+    intel_clock_t clock;
+    if (target < 200000) {
+	clock.dot = 161670;
+	clock.p = 20;
+	clock.p1 = 2;
+	clock.p2 = 10;
+	clock.n = 0x01;
+	clock.m = 97;
+	clock.m1 = 0x10;
+	clock.m2 = 0x05;
+    } else {
+	clock.dot = 270000;
+	clock.p = 10;
+	clock.p1 = 1;
+	clock.p2 = 10;
+	clock.n = 0x02;
+	clock.m = 108;
+	clock.m1 = 0x12;
+	clock.m2 = 0x06;
+    }
+    memcpy(best_clock, &clock, sizeof(intel_clock_t));
+    return true;
+}
+
 void
 intel_wait_for_vblank(struct drm_device *dev)
 {
@@ -1541,7 +1619,7 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc,
 	intel_clock_t clock;
 	u32 dpll = 0, fp = 0, dspcntr, pipeconf;
 	bool ok, is_sdvo = false, is_dvo = false;
-	bool is_crt = false, is_lvds = false, is_tv = false;
+	bool is_crt = false, is_lvds = false, is_tv = false, is_dp = false;
 	struct drm_mode_config *mode_config = &dev->mode_config;
 	struct drm_connector *connector;
 	const intel_limit_t *limit;
@@ -1585,6 +1663,9 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc,
 		case INTEL_OUTPUT_ANALOG:
 			is_crt = true;
 			break;
+		case INTEL_OUTPUT_DISPLAYPORT:
+			is_dp = true;
+			break;
 		}
 
 		num_outputs++;
@@ -1600,6 +1681,7 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc,
 	} else {
 		refclk = 48000;
 	}
+	
 
 	/*
 	 * Returns a set of divisors for the desired target clock with the given
@@ -1662,6 +1744,8 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc,
 			else if (IS_IGDNG(dev))
 				dpll |= (sdvo_pixel_multiply - 1) << PLL_REF_SDVO_HDMI_MULTIPLIER_SHIFT;
 		}
+		if (is_dp)
+			dpll |= DPLL_DVO_HIGH_SPEED;
 
 		/* compute bitmask from p1 value */
 		if (IS_IGD(dev))
@@ -1809,6 +1893,8 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc,
 		I915_WRITE(lvds_reg, lvds);
 		I915_READ(lvds_reg);
 	}
+	if (is_dp)
+		intel_dp_set_m_n(crtc, mode, adjusted_mode);
 
 	I915_WRITE(fp_reg, fp);
 	I915_WRITE(dpll_reg, dpll);
@@ -2475,6 +2561,8 @@ static void intel_setup_outputs(struct drm_device *dev)
 			found = intel_sdvo_init(dev, SDVOB);
 			if (!found && SUPPORTS_INTEGRATED_HDMI(dev))
 				intel_hdmi_init(dev, SDVOB);
+			if (!found && SUPPORTS_INTEGRATED_DP(dev))
+				intel_dp_init(dev, DP_B);
 		}
 
 		/* Before G4X SDVOC doesn't have its own detect register */
@@ -2487,7 +2575,11 @@ static void intel_setup_outputs(struct drm_device *dev)
 			found = intel_sdvo_init(dev, SDVOC);
 			if (!found && SUPPORTS_INTEGRATED_HDMI(dev))
 				intel_hdmi_init(dev, SDVOC);
+			if (!found && SUPPORTS_INTEGRATED_DP(dev))
+				intel_dp_init(dev, DP_C);
 		}
+		if (SUPPORTS_INTEGRATED_DP(dev) && (I915_READ(DP_D) & DP_DETECTED))
+			intel_dp_init(dev, DP_D);
 	} else
 		intel_dvo_init(dev);
 
@@ -2530,6 +2622,11 @@ static void intel_setup_outputs(struct drm_device *dev)
 				     (1 << 1));
 			clone_mask = (1 << INTEL_OUTPUT_TVOUT);
 			break;
+		case INTEL_OUTPUT_DISPLAYPORT:
+			crtc_mask = ((1 << 0) |
+				     (1 << 1));
+			clone_mask = (1 << INTEL_OUTPUT_DISPLAYPORT);
+			break;
 		}
 		encoder->possible_crtcs = crtc_mask;
 		encoder->possible_clones = intel_connector_clones(dev, clone_mask);
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
new file mode 100644
index 000000000000..c57cdab4f4a6
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -0,0 +1,1098 @@
+/*
+ * Copyright © 2008 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Authors:
+ *    Keith Packard <keithp@keithp.com>
+ *
+ */
+
+#include <linux/i2c.h>
+#include "drmP.h"
+#include "drm.h"
+#include "drm_crtc.h"
+#include "drm_crtc_helper.h"
+#include "intel_drv.h"
+#include "i915_drm.h"
+#include "i915_drv.h"
+#include "intel_dp.h"
+
+#define DP_LINK_STATUS_SIZE	6
+#define DP_LINK_CHECK_TIMEOUT	(10 * 1000)
+
+#define DP_LINK_CONFIGURATION_SIZE	9
+
+struct intel_dp_priv {
+	uint32_t output_reg;
+	uint32_t DP;
+	uint8_t  link_configuration[DP_LINK_CONFIGURATION_SIZE];
+	uint32_t save_DP;
+	uint8_t  save_link_configuration[DP_LINK_CONFIGURATION_SIZE];
+	bool has_audio;
+	uint8_t link_bw;
+	uint8_t lane_count;
+	uint8_t dpcd[4];
+	struct intel_output *intel_output;
+	struct i2c_adapter adapter;
+	struct i2c_algo_dp_aux_data algo;
+};
+
+static void
+intel_dp_link_train(struct intel_output *intel_output, uint32_t DP,
+		    uint8_t link_configuration[DP_LINK_CONFIGURATION_SIZE]);
+
+static void
+intel_dp_link_down(struct intel_output *intel_output, uint32_t DP);
+
+static int
+intel_dp_max_lane_count(struct intel_output *intel_output)
+{
+	struct intel_dp_priv   *dp_priv = intel_output->dev_priv;
+	int max_lane_count = 4;
+
+	if (dp_priv->dpcd[0] >= 0x11) {
+		max_lane_count = dp_priv->dpcd[2] & 0x1f;
+		switch (max_lane_count) {
+		case 1: case 2: case 4:
+			break;
+		default:
+			max_lane_count = 4;
+		}
+	}
+	return max_lane_count;
+}
+
+static int
+intel_dp_max_link_bw(struct intel_output *intel_output)
+{
+	struct intel_dp_priv   *dp_priv = intel_output->dev_priv;
+	int max_link_bw = dp_priv->dpcd[1];
+
+	switch (max_link_bw) {
+	case DP_LINK_BW_1_62:
+	case DP_LINK_BW_2_7:
+		break;
+	default:
+		max_link_bw = DP_LINK_BW_1_62;
+		break;
+	}
+	return max_link_bw;
+}
+
+static int
+intel_dp_link_clock(uint8_t link_bw)
+{
+	if (link_bw == DP_LINK_BW_2_7)
+		return 270000;
+	else
+		return 162000;
+}
+
+/* I think this is a fiction */
+static int
+intel_dp_link_required(int pixel_clock)
+{
+	return pixel_clock * 3;
+}
+
+static int
+intel_dp_mode_valid(struct drm_connector *connector,
+		    struct drm_display_mode *mode)
+{
+	struct intel_output *intel_output = to_intel_output(connector);
+	int max_link_clock = intel_dp_link_clock(intel_dp_max_link_bw(intel_output));
+	int max_lanes = intel_dp_max_lane_count(intel_output);
+
+	if (intel_dp_link_required(mode->clock) > max_link_clock * max_lanes)
+		return MODE_CLOCK_HIGH;
+
+	if (mode->clock < 10000)
+		return MODE_CLOCK_LOW;
+
+	return MODE_OK;
+}
+
+static uint32_t
+pack_aux(uint8_t *src, int src_bytes)
+{
+	int	i;
+	uint32_t v = 0;
+
+	if (src_bytes > 4)
+		src_bytes = 4;
+	for (i = 0; i < src_bytes; i++)
+		v |= ((uint32_t) src[i]) << ((3-i) * 8);
+	return v;
+}
+
+static void
+unpack_aux(uint32_t src, uint8_t *dst, int dst_bytes)
+{
+	int i;
+	if (dst_bytes > 4)
+		dst_bytes = 4;
+	for (i = 0; i < dst_bytes; i++)
+		dst[i] = src >> ((3-i) * 8);
+}
+
+static int
+intel_dp_aux_ch(struct intel_output *intel_output,
+		uint8_t *send, int send_bytes,
+		uint8_t *recv, int recv_size)
+{
+	struct intel_dp_priv *dp_priv = intel_output->dev_priv;
+	uint32_t output_reg = dp_priv->output_reg;
+	struct drm_device *dev = intel_output->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t ch_ctl = output_reg + 0x10;
+	uint32_t ch_data = ch_ctl + 4;
+	int i;
+	int recv_bytes;
+	uint32_t ctl;
+	uint32_t status;
+
+	/* Load the send data into the aux channel data registers */
+	for (i = 0; i < send_bytes; i += 4) {
+		uint32_t    d = pack_aux(send + i, send_bytes - i);;
+
+		I915_WRITE(ch_data + i, d);
+	}
+
+	/* The clock divider is based off the hrawclk,
+	 * and would like to run at 2MHz. The 133 below assumes
+	 * a 266MHz hrawclk; need to figure out how we're supposed
+	 * to know what hrawclk is...
+	 */
+	ctl = (DP_AUX_CH_CTL_SEND_BUSY |
+	       DP_AUX_CH_CTL_TIME_OUT_1600us |
+	       (send_bytes << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) |
+	       (5 << DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT) |
+	       (133 << DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT) |
+	       DP_AUX_CH_CTL_TIME_OUT_ERROR |
+	       DP_AUX_CH_CTL_RECEIVE_ERROR);
+
+	/* Send the command and wait for it to complete */
+	I915_WRITE(ch_ctl, ctl);
+	(void) I915_READ(ch_ctl);
+	for (;;) {
+		udelay(100);
+		status = I915_READ(ch_ctl);
+		if ((status & DP_AUX_CH_CTL_SEND_BUSY) == 0)
+			break;
+	}
+
+	/* Clear done status and any errors */
+	I915_WRITE(ch_ctl, (ctl |
+			DP_AUX_CH_CTL_DONE |
+			DP_AUX_CH_CTL_TIME_OUT_ERROR |
+			DP_AUX_CH_CTL_RECEIVE_ERROR));
+	(void) I915_READ(ch_ctl);
+
+	if ((status & DP_AUX_CH_CTL_DONE) == 0) {
+		printk(KERN_ERR "dp_aux_ch not done status 0x%08x\n", status);
+		return -1;
+	}
+
+	/* Check for timeout or receive error.
+	 * Timeouts occur when the sink is not connected
+	 */
+	if (status & (DP_AUX_CH_CTL_TIME_OUT_ERROR | DP_AUX_CH_CTL_RECEIVE_ERROR)) {
+		printk(KERN_ERR "dp_aux_ch error status 0x%08x\n", status);
+		return -1;
+	}
+
+	/* Unload any bytes sent back from the other side */
+	recv_bytes = ((status & DP_AUX_CH_CTL_MESSAGE_SIZE_MASK) >>
+		      DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT);
+
+	if (recv_bytes > recv_size)
+		recv_bytes = recv_size;
+	
+	for (i = 0; i < recv_bytes; i += 4) {
+		uint32_t    d = I915_READ(ch_data + i);
+
+		unpack_aux(d, recv + i, recv_bytes - i);
+	}
+
+	return recv_bytes;
+}
+
+/* Write data to the aux channel in native mode */
+static int
+intel_dp_aux_native_write(struct intel_output *intel_output,
+			  uint16_t address, uint8_t *send, int send_bytes)
+{
+	int ret;
+	uint8_t	msg[20];
+	int msg_bytes;
+	uint8_t	ack;
+
+	if (send_bytes > 16)
+		return -1;
+	msg[0] = AUX_NATIVE_WRITE << 4;
+	msg[1] = address >> 8;
+	msg[2] = address;
+	msg[3] = send_bytes - 1;
+	memcpy(&msg[4], send, send_bytes);
+	msg_bytes = send_bytes + 4;
+	for (;;) {
+		ret = intel_dp_aux_ch(intel_output, msg, msg_bytes, &ack, 1);
+		if (ret < 0)
+			return ret;
+		if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_ACK)
+			break;
+		else if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_DEFER)
+			udelay(100);
+		else
+			return -1;
+	}
+	return send_bytes;
+}
+
+/* Write a single byte to the aux channel in native mode */
+static int
+intel_dp_aux_native_write_1(struct intel_output *intel_output,
+			    uint16_t address, uint8_t byte)
+{
+	return intel_dp_aux_native_write(intel_output, address, &byte, 1);
+}
+
+/* read bytes from a native aux channel */
+static int
+intel_dp_aux_native_read(struct intel_output *intel_output,
+			 uint16_t address, uint8_t *recv, int recv_bytes)
+{
+	uint8_t msg[4];
+	int msg_bytes;
+	uint8_t reply[20];
+	int reply_bytes;
+	uint8_t ack;
+	int ret;
+
+	msg[0] = AUX_NATIVE_READ << 4;
+	msg[1] = address >> 8;
+	msg[2] = address & 0xff;
+	msg[3] = recv_bytes - 1;
+
+	msg_bytes = 4;
+	reply_bytes = recv_bytes + 1;
+
+	for (;;) {
+		ret = intel_dp_aux_ch(intel_output, msg, msg_bytes,
+				      reply, reply_bytes);
+		if (ret <= 0)
+			return ret;
+		ack = reply[0];
+		if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_ACK) {
+			memcpy(recv, reply + 1, ret - 1);
+			return ret - 1;
+		}
+		else if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_DEFER)
+			udelay(100);
+		else
+			return -1;
+	}
+}
+
+static int
+intel_dp_i2c_aux_ch(struct i2c_adapter *adapter,
+		    uint8_t *send, int send_bytes,
+		    uint8_t *recv, int recv_bytes)
+{
+	struct intel_dp_priv *dp_priv = container_of(adapter,
+						     struct intel_dp_priv,
+						     adapter);
+	struct intel_output *intel_output = dp_priv->intel_output;
+
+	return intel_dp_aux_ch(intel_output,
+			       send, send_bytes, recv, recv_bytes);
+}
+
+static int
+intel_dp_i2c_init(struct intel_output *intel_output, const char *name)
+{
+	struct intel_dp_priv   *dp_priv = intel_output->dev_priv;
+
+	DRM_ERROR("i2c_init %s\n", name);
+	dp_priv->algo.running = false;
+	dp_priv->algo.address = 0;
+	dp_priv->algo.aux_ch = intel_dp_i2c_aux_ch;
+
+	memset(&dp_priv->adapter, '\0', sizeof (dp_priv->adapter));
+	dp_priv->adapter.owner = THIS_MODULE;
+	dp_priv->adapter.class = I2C_CLASS_DDC;
+	strncpy (dp_priv->adapter.name, name, sizeof dp_priv->adapter.name - 1);
+	dp_priv->adapter.name[sizeof dp_priv->adapter.name - 1] = '\0';
+	dp_priv->adapter.algo_data = &dp_priv->algo;
+	dp_priv->adapter.dev.parent = &intel_output->base.kdev;
+	
+	return i2c_dp_aux_add_bus(&dp_priv->adapter);
+}
+
+static bool
+intel_dp_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode,
+		    struct drm_display_mode *adjusted_mode)
+{
+	struct intel_output *intel_output = enc_to_intel_output(encoder);
+	struct intel_dp_priv   *dp_priv = intel_output->dev_priv;
+	int lane_count, clock;
+	int max_lane_count = intel_dp_max_lane_count(intel_output);
+	int max_clock = intel_dp_max_link_bw(intel_output) == DP_LINK_BW_2_7 ? 1 : 0;
+	static int bws[2] = { DP_LINK_BW_1_62, DP_LINK_BW_2_7 };
+
+	for (lane_count = 1; lane_count <= max_lane_count; lane_count <<= 1) {
+		for (clock = 0; clock <= max_clock; clock++) {
+			int link_avail = intel_dp_link_clock(bws[clock]) * lane_count;
+
+			if (intel_dp_link_required(mode->clock) <= link_avail) {
+				dp_priv->link_bw = bws[clock];
+				dp_priv->lane_count = lane_count;
+				adjusted_mode->clock = intel_dp_link_clock(dp_priv->link_bw);
+				printk(KERN_ERR "link bw %02x lane count %d clock %d\n",
+				       dp_priv->link_bw, dp_priv->lane_count,
+				       adjusted_mode->clock);
+				return true;
+			}
+		}
+	}
+	return false;
+}
+
+struct intel_dp_m_n {
+	uint32_t	tu;
+	uint32_t	gmch_m;
+	uint32_t	gmch_n;
+	uint32_t	link_m;
+	uint32_t	link_n;
+};
+
+static void
+intel_reduce_ratio(uint32_t *num, uint32_t *den)
+{
+	while (*num > 0xffffff || *den > 0xffffff) {
+		*num >>= 1;
+		*den >>= 1;
+	}
+}
+
+static void
+intel_dp_compute_m_n(int bytes_per_pixel,
+		     int nlanes,
+		     int pixel_clock,
+		     int link_clock,
+		     struct intel_dp_m_n *m_n)
+{
+	m_n->tu = 64;
+	m_n->gmch_m = pixel_clock * bytes_per_pixel;
+	m_n->gmch_n = link_clock * nlanes;
+	intel_reduce_ratio(&m_n->gmch_m, &m_n->gmch_n);
+	m_n->link_m = pixel_clock;
+	m_n->link_n = link_clock;
+	intel_reduce_ratio(&m_n->link_m, &m_n->link_n);
+}
+
+void
+intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
+		 struct drm_display_mode *adjusted_mode)
+{
+	struct drm_device *dev = crtc->dev;
+	struct drm_mode_config *mode_config = &dev->mode_config;
+	struct drm_connector *connector;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	int lane_count = 4;
+	struct intel_dp_m_n m_n;
+
+	/*
+	 * Find the lane count in the intel_output private
+	 */
+	list_for_each_entry(connector, &mode_config->connector_list, head) {
+		struct intel_output *intel_output = to_intel_output(connector);
+		struct intel_dp_priv *dp_priv = intel_output->dev_priv;
+
+		if (!connector->encoder || connector->encoder->crtc != crtc)
+			continue;
+
+		if (intel_output->type == INTEL_OUTPUT_DISPLAYPORT) {
+			lane_count = dp_priv->lane_count;
+			break;
+		}
+	}
+
+	/*
+	 * Compute the GMCH and Link ratios. The '3' here is
+	 * the number of bytes_per_pixel post-LUT, which we always
+	 * set up for 8-bits of R/G/B, or 3 bytes total.
+	 */
+	intel_dp_compute_m_n(3, lane_count,
+			     mode->clock, adjusted_mode->clock, &m_n);
+
+	if (intel_crtc->pipe == 0) {
+		I915_WRITE(PIPEA_GMCH_DATA_M,
+		       ((m_n.tu - 1) << PIPE_GMCH_DATA_M_TU_SIZE_SHIFT) |
+		       m_n.gmch_m);
+		I915_WRITE(PIPEA_GMCH_DATA_N,
+		       m_n.gmch_n);
+		I915_WRITE(PIPEA_DP_LINK_M, m_n.link_m);
+		I915_WRITE(PIPEA_DP_LINK_N, m_n.link_n);
+	} else {
+		I915_WRITE(PIPEB_GMCH_DATA_M,
+		       ((m_n.tu - 1) << PIPE_GMCH_DATA_M_TU_SIZE_SHIFT) |
+		       m_n.gmch_m);
+		I915_WRITE(PIPEB_GMCH_DATA_N,
+		       m_n.gmch_n);
+		I915_WRITE(PIPEB_DP_LINK_M, m_n.link_m);
+		I915_WRITE(PIPEB_DP_LINK_N, m_n.link_n);
+	}
+}
+
+static void
+intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode,
+		  struct drm_display_mode *adjusted_mode)
+{
+	struct intel_output *intel_output = enc_to_intel_output(encoder);
+	struct intel_dp_priv *dp_priv = intel_output->dev_priv;
+	struct drm_crtc *crtc = intel_output->enc.crtc;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+
+	dp_priv->DP = (DP_LINK_TRAIN_OFF |
+			DP_VOLTAGE_0_4 |
+			DP_PRE_EMPHASIS_0 |
+			DP_SYNC_VS_HIGH |
+			DP_SYNC_HS_HIGH);
+
+	switch (dp_priv->lane_count) {
+	case 1:
+		dp_priv->DP |= DP_PORT_WIDTH_1;
+		break;
+	case 2:
+		dp_priv->DP |= DP_PORT_WIDTH_2;
+		break;
+	case 4:
+		dp_priv->DP |= DP_PORT_WIDTH_4;
+		break;
+	}
+	if (dp_priv->has_audio)
+		dp_priv->DP |= DP_AUDIO_OUTPUT_ENABLE;
+
+	memset(dp_priv->link_configuration, 0, DP_LINK_CONFIGURATION_SIZE);
+	dp_priv->link_configuration[0] = dp_priv->link_bw;
+	dp_priv->link_configuration[1] = dp_priv->lane_count;
+
+	/*
+	 * Check for DPCD version > 1.1,
+	 * enable enahanced frame stuff in that case
+	 */
+	if (dp_priv->dpcd[0] >= 0x11) {
+		dp_priv->link_configuration[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
+		dp_priv->DP |= DP_ENHANCED_FRAMING;
+	}
+
+	if (intel_crtc->pipe == 1)
+		dp_priv->DP |= DP_PIPEB_SELECT;
+}
+
+
+static void
+intel_dp_dpms(struct drm_encoder *encoder, int mode)
+{
+	struct intel_output *intel_output = enc_to_intel_output(encoder);
+	struct intel_dp_priv *dp_priv = intel_output->dev_priv;
+	struct drm_device *dev = intel_output->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t dp_reg = I915_READ(dp_priv->output_reg);
+
+	if (mode != DRM_MODE_DPMS_ON) {
+		if (dp_reg & DP_PORT_EN)
+			intel_dp_link_down(intel_output, dp_priv->DP);
+	} else {
+		if (!(dp_reg & DP_PORT_EN))
+			intel_dp_link_train(intel_output, dp_priv->DP, dp_priv->link_configuration);
+	}
+}
+
+/*
+ * Fetch AUX CH registers 0x202 - 0x207 which contain
+ * link status information
+ */
+static bool
+intel_dp_get_link_status(struct intel_output *intel_output,
+			 uint8_t link_status[DP_LINK_STATUS_SIZE])
+{
+	int ret;
+
+	ret = intel_dp_aux_native_read(intel_output,
+				       DP_LANE0_1_STATUS,
+				       link_status, DP_LINK_STATUS_SIZE);
+	if (ret != DP_LINK_STATUS_SIZE)
+		return false;
+	return true;
+}
+
+static uint8_t
+intel_dp_link_status(uint8_t link_status[DP_LINK_STATUS_SIZE],
+		     int r)
+{
+	return link_status[r - DP_LANE0_1_STATUS];
+}
+
+static void
+intel_dp_save(struct drm_connector *connector)
+{
+	struct intel_output *intel_output = to_intel_output(connector);
+	struct drm_device *dev = intel_output->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_dp_priv *dp_priv = intel_output->dev_priv;
+
+	dp_priv->save_DP = I915_READ(dp_priv->output_reg);
+	intel_dp_aux_native_read(intel_output, DP_LINK_BW_SET,
+				 dp_priv->save_link_configuration,
+				 sizeof (dp_priv->save_link_configuration));
+}
+
+static uint8_t
+intel_get_adjust_request_voltage(uint8_t link_status[DP_LINK_STATUS_SIZE],
+				 int lane)
+{
+	int	    i = DP_ADJUST_REQUEST_LANE0_1 + (lane >> 1);
+	int	    s = ((lane & 1) ?
+			 DP_ADJUST_VOLTAGE_SWING_LANE1_SHIFT :
+			 DP_ADJUST_VOLTAGE_SWING_LANE0_SHIFT);
+	uint8_t l = intel_dp_link_status(link_status, i);
+
+	return ((l >> s) & 3) << DP_TRAIN_VOLTAGE_SWING_SHIFT;
+}
+
+static uint8_t
+intel_get_adjust_request_pre_emphasis(uint8_t link_status[DP_LINK_STATUS_SIZE],
+				      int lane)
+{
+	int	    i = DP_ADJUST_REQUEST_LANE0_1 + (lane >> 1);
+	int	    s = ((lane & 1) ?
+			 DP_ADJUST_PRE_EMPHASIS_LANE1_SHIFT :
+			 DP_ADJUST_PRE_EMPHASIS_LANE0_SHIFT);
+	uint8_t l = intel_dp_link_status(link_status, i);
+
+	return ((l >> s) & 3) << DP_TRAIN_PRE_EMPHASIS_SHIFT;
+}
+
+
+#if 0
+static char	*voltage_names[] = {
+	"0.4V", "0.6V", "0.8V", "1.2V"
+};
+static char	*pre_emph_names[] = {
+	"0dB", "3.5dB", "6dB", "9.5dB"
+};
+static char	*link_train_names[] = {
+	"pattern 1", "pattern 2", "idle", "off"
+};
+#endif
+
+/*
+ * These are source-specific values; current Intel hardware supports
+ * a maximum voltage of 800mV and a maximum pre-emphasis of 6dB
+ */
+#define I830_DP_VOLTAGE_MAX	    DP_TRAIN_VOLTAGE_SWING_800
+
+static uint8_t
+intel_dp_pre_emphasis_max(uint8_t voltage_swing)
+{
+	switch (voltage_swing & DP_TRAIN_VOLTAGE_SWING_MASK) {
+	case DP_TRAIN_VOLTAGE_SWING_400:
+		return DP_TRAIN_PRE_EMPHASIS_6;
+	case DP_TRAIN_VOLTAGE_SWING_600:
+		return DP_TRAIN_PRE_EMPHASIS_6;
+	case DP_TRAIN_VOLTAGE_SWING_800:
+		return DP_TRAIN_PRE_EMPHASIS_3_5;
+	case DP_TRAIN_VOLTAGE_SWING_1200:
+	default:
+		return DP_TRAIN_PRE_EMPHASIS_0;
+	}
+}
+
+static void
+intel_get_adjust_train(struct intel_output *intel_output,
+		       uint8_t link_status[DP_LINK_STATUS_SIZE],
+		       int lane_count,
+		       uint8_t train_set[4])
+{
+	uint8_t v = 0;
+	uint8_t p = 0;
+	int lane;
+
+	for (lane = 0; lane < lane_count; lane++) {
+		uint8_t this_v = intel_get_adjust_request_voltage(link_status, lane);
+		uint8_t this_p = intel_get_adjust_request_pre_emphasis(link_status, lane);
+
+		if (this_v > v)
+			v = this_v;
+		if (this_p > p)
+			p = this_p;
+	}
+
+	if (v >= I830_DP_VOLTAGE_MAX)
+		v = I830_DP_VOLTAGE_MAX | DP_TRAIN_MAX_SWING_REACHED;
+
+	if (p >= intel_dp_pre_emphasis_max(v))
+		p = intel_dp_pre_emphasis_max(v) | DP_TRAIN_MAX_PRE_EMPHASIS_REACHED;
+
+	for (lane = 0; lane < 4; lane++)
+		train_set[lane] = v | p;
+}
+
+static uint32_t
+intel_dp_signal_levels(uint8_t train_set, int lane_count)
+{
+	uint32_t	signal_levels = 0;
+
+	switch (train_set & DP_TRAIN_VOLTAGE_SWING_MASK) {
+	case DP_TRAIN_VOLTAGE_SWING_400:
+	default:
+		signal_levels |= DP_VOLTAGE_0_4;
+		break;
+	case DP_TRAIN_VOLTAGE_SWING_600:
+		signal_levels |= DP_VOLTAGE_0_6;
+		break;
+	case DP_TRAIN_VOLTAGE_SWING_800:
+		signal_levels |= DP_VOLTAGE_0_8;
+		break;
+	case DP_TRAIN_VOLTAGE_SWING_1200:
+		signal_levels |= DP_VOLTAGE_1_2;
+		break;
+	}
+	switch (train_set & DP_TRAIN_PRE_EMPHASIS_MASK) {
+	case DP_TRAIN_PRE_EMPHASIS_0:
+	default:
+		signal_levels |= DP_PRE_EMPHASIS_0;
+		break;
+	case DP_TRAIN_PRE_EMPHASIS_3_5:
+		signal_levels |= DP_PRE_EMPHASIS_3_5;
+		break;
+	case DP_TRAIN_PRE_EMPHASIS_6:
+		signal_levels |= DP_PRE_EMPHASIS_6;
+		break;
+	case DP_TRAIN_PRE_EMPHASIS_9_5:
+		signal_levels |= DP_PRE_EMPHASIS_9_5;
+		break;
+	}
+	return signal_levels;
+}
+
+static uint8_t
+intel_get_lane_status(uint8_t link_status[DP_LINK_STATUS_SIZE],
+		      int lane)
+{
+	int i = DP_LANE0_1_STATUS + (lane >> 1);
+	int s = (lane & 1) * 4;
+	uint8_t l = intel_dp_link_status(link_status, i);
+
+	return (l >> s) & 0xf;
+}
+
+/* Check for clock recovery is done on all channels */
+static bool
+intel_clock_recovery_ok(uint8_t link_status[DP_LINK_STATUS_SIZE], int lane_count)
+{
+	int lane;
+	uint8_t lane_status;
+
+	for (lane = 0; lane < lane_count; lane++) {
+		lane_status = intel_get_lane_status(link_status, lane);
+		if ((lane_status & DP_LANE_CR_DONE) == 0)
+			return false;
+	}
+	return true;
+}
+
+/* Check to see if channel eq is done on all channels */
+#define CHANNEL_EQ_BITS (DP_LANE_CR_DONE|\
+			 DP_LANE_CHANNEL_EQ_DONE|\
+			 DP_LANE_SYMBOL_LOCKED)
+static bool
+intel_channel_eq_ok(uint8_t link_status[DP_LINK_STATUS_SIZE], int lane_count)
+{
+	uint8_t lane_align;
+	uint8_t lane_status;
+	int lane;
+
+	lane_align = intel_dp_link_status(link_status,
+					  DP_LANE_ALIGN_STATUS_UPDATED);
+	if ((lane_align & DP_INTERLANE_ALIGN_DONE) == 0)
+		return false;
+	for (lane = 0; lane < lane_count; lane++) {
+		lane_status = intel_get_lane_status(link_status, lane);
+		if ((lane_status & CHANNEL_EQ_BITS) != CHANNEL_EQ_BITS)
+			return false;
+	}
+	return true;
+}
+
+static bool
+intel_dp_set_link_train(struct intel_output *intel_output,
+			uint32_t dp_reg_value,
+			uint8_t dp_train_pat,
+			uint8_t train_set[4],
+			bool first)
+{
+	struct drm_device *dev = intel_output->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_dp_priv *dp_priv = intel_output->dev_priv;
+	int ret;
+
+	I915_WRITE(dp_priv->output_reg, dp_reg_value);
+	POSTING_READ(dp_priv->output_reg);
+	if (first)
+		intel_wait_for_vblank(dev);
+
+	intel_dp_aux_native_write_1(intel_output,
+				    DP_TRAINING_PATTERN_SET,
+				    dp_train_pat);
+
+	ret = intel_dp_aux_native_write(intel_output,
+					DP_TRAINING_LANE0_SET, train_set, 4);
+	if (ret != 4)
+		return false;
+
+	return true;
+}
+
+static void
+intel_dp_link_train(struct intel_output *intel_output, uint32_t DP,
+		    uint8_t link_configuration[DP_LINK_CONFIGURATION_SIZE])
+{
+	struct drm_device *dev = intel_output->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_dp_priv *dp_priv = intel_output->dev_priv;
+	uint8_t	train_set[4];
+	uint8_t link_status[DP_LINK_STATUS_SIZE];
+	int i;
+	uint8_t voltage;
+	bool clock_recovery = false;
+	bool channel_eq = false;
+	bool first = true;
+	int tries;
+
+	/* Write the link configuration data */
+	intel_dp_aux_native_write(intel_output, 0x100,
+				  link_configuration, DP_LINK_CONFIGURATION_SIZE);
+
+	DP |= DP_PORT_EN;
+	DP &= ~DP_LINK_TRAIN_MASK;
+	memset(train_set, 0, 4);
+	voltage = 0xff;
+	tries = 0;
+	clock_recovery = false;
+	for (;;) {
+		/* Use train_set[0] to set the voltage and pre emphasis values */
+		uint32_t    signal_levels = intel_dp_signal_levels(train_set[0], dp_priv->lane_count);
+		DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels;
+
+		if (!intel_dp_set_link_train(intel_output, DP | DP_LINK_TRAIN_PAT_1,
+					     DP_TRAINING_PATTERN_1, train_set, first))
+			break;
+		first = false;
+		/* Set training pattern 1 */
+
+		udelay(100);
+		if (!intel_dp_get_link_status(intel_output, link_status))
+			break;
+
+		if (intel_clock_recovery_ok(link_status, dp_priv->lane_count)) {
+			clock_recovery = true;
+			break;
+		}
+
+		/* Check to see if we've tried the max voltage */
+		for (i = 0; i < dp_priv->lane_count; i++)
+			if ((train_set[i] & DP_TRAIN_MAX_SWING_REACHED) == 0)
+				break;
+		if (i == dp_priv->lane_count)
+			break;
+
+		/* Check to see if we've tried the same voltage 5 times */
+		if ((train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK) == voltage) {
+			++tries;
+			if (tries == 5)
+				break;
+		} else
+			tries = 0;
+		voltage = train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK;
+
+		/* Compute new train_set as requested by target */
+		intel_get_adjust_train(intel_output, link_status, dp_priv->lane_count, train_set);
+	}
+
+	/* channel equalization */
+	tries = 0;
+	channel_eq = false;
+	for (;;) {
+		/* Use train_set[0] to set the voltage and pre emphasis values */
+		uint32_t    signal_levels = intel_dp_signal_levels(train_set[0], dp_priv->lane_count);
+		DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels;
+
+		/* channel eq pattern */
+		if (!intel_dp_set_link_train(intel_output, DP | DP_LINK_TRAIN_PAT_2,
+					     DP_TRAINING_PATTERN_2, train_set,
+					     false))
+			break;
+
+		udelay(400);
+		if (!intel_dp_get_link_status(intel_output, link_status))
+			break;
+
+		if (intel_channel_eq_ok(link_status, dp_priv->lane_count)) {
+			channel_eq = true;
+			break;
+		}
+
+		/* Try 5 times */
+		if (tries > 5)
+			break;
+
+		/* Compute new train_set as requested by target */
+		intel_get_adjust_train(intel_output, link_status, dp_priv->lane_count, train_set);
+		++tries;
+	}
+
+	I915_WRITE(dp_priv->output_reg, DP | DP_LINK_TRAIN_OFF);
+	POSTING_READ(dp_priv->output_reg);
+	intel_dp_aux_native_write_1(intel_output,
+				    DP_TRAINING_PATTERN_SET, DP_TRAINING_PATTERN_DISABLE);
+}
+
+static void
+intel_dp_link_down(struct intel_output *intel_output, uint32_t DP)
+{
+	struct drm_device *dev = intel_output->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_dp_priv *dp_priv = intel_output->dev_priv;
+
+	I915_WRITE(dp_priv->output_reg, DP & ~DP_PORT_EN);
+	POSTING_READ(dp_priv->output_reg);
+}
+
+static void
+intel_dp_restore(struct drm_connector *connector)
+{
+	struct intel_output *intel_output = to_intel_output(connector);
+	struct intel_dp_priv *dp_priv = intel_output->dev_priv;
+
+	if (dp_priv->save_DP & DP_PORT_EN)
+		intel_dp_link_train(intel_output, dp_priv->save_DP, dp_priv->save_link_configuration);
+	else
+		intel_dp_link_down(intel_output,  dp_priv->save_DP);
+}
+
+#if 0
+/*
+ * According to DP spec
+ * 5.1.2:
+ *  1. Read DPCD
+ *  2. Configure link according to Receiver Capabilities
+ *  3. Use Link Training from 2.5.3.3 and 3.5.1.3
+ *  4. Check link status on receipt of hot-plug interrupt
+ */
+
+static void
+intel_dp_check_link_status(struct intel_output *intel_output)
+{
+	struct intel_dp_priv *dp_priv = intel_output->dev_priv;
+	uint8_t link_status[DP_LINK_STATUS_SIZE];
+
+	if (!intel_output->enc.crtc)
+		return;
+
+	if (!intel_dp_get_link_status(intel_output, link_status)) {
+		intel_dp_link_down(intel_output, dp_priv->DP);
+		return;
+	}
+
+	if (!intel_channel_eq_ok(link_status, dp_priv->lane_count))
+		intel_dp_link_train(intel_output, dp_priv->DP, dp_priv->link_configuration);
+}
+#endif
+
+/**
+ * Uses CRT_HOTPLUG_EN and CRT_HOTPLUG_STAT to detect DP connection.
+ *
+ * \return true if DP port is connected.
+ * \return false if DP port is disconnected.
+ */
+static enum drm_connector_status
+intel_dp_detect(struct drm_connector *connector)
+{
+	struct intel_output *intel_output = to_intel_output(connector);
+	struct drm_device *dev = intel_output->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_dp_priv *dp_priv = intel_output->dev_priv;
+	uint32_t temp, bit;
+	enum drm_connector_status status;
+
+	dp_priv->has_audio = false;
+
+	temp = I915_READ(PORT_HOTPLUG_EN);
+
+	I915_WRITE(PORT_HOTPLUG_EN,
+	       temp |
+	       DPB_HOTPLUG_INT_EN |
+	       DPC_HOTPLUG_INT_EN |
+	       DPD_HOTPLUG_INT_EN);
+
+	POSTING_READ(PORT_HOTPLUG_EN);
+
+	switch (dp_priv->output_reg) {
+	case DP_B:
+		bit = DPB_HOTPLUG_INT_STATUS;
+		break;
+	case DP_C:
+		bit = DPC_HOTPLUG_INT_STATUS;
+		break;
+	case DP_D:
+		bit = DPD_HOTPLUG_INT_STATUS;
+		break;
+	default:
+		return connector_status_unknown;
+	}
+
+	temp = I915_READ(PORT_HOTPLUG_STAT);
+
+	if ((temp & bit) == 0)
+		return connector_status_disconnected;
+
+	status = connector_status_disconnected;
+	if (intel_dp_aux_native_read(intel_output,
+				     0x000, dp_priv->dpcd,
+				     sizeof (dp_priv->dpcd)) == sizeof (dp_priv->dpcd))
+	{
+		if (dp_priv->dpcd[0] != 0)
+			status = connector_status_connected;
+	}
+	return status;
+}
+
+static int intel_dp_get_modes(struct drm_connector *connector)
+{
+	struct intel_output *intel_output = to_intel_output(connector);
+
+	/* We should parse the EDID data and find out if it has an audio sink
+	 */
+
+	return intel_ddc_get_modes(intel_output);
+}
+
+static void
+intel_dp_destroy (struct drm_connector *connector)
+{
+	struct intel_output *intel_output = to_intel_output(connector);
+
+	if (intel_output->i2c_bus)
+		intel_i2c_destroy(intel_output->i2c_bus);
+	drm_sysfs_connector_remove(connector);
+	drm_connector_cleanup(connector);
+	kfree(intel_output);
+}
+
+static const struct drm_encoder_helper_funcs intel_dp_helper_funcs = {
+	.dpms = intel_dp_dpms,
+	.mode_fixup = intel_dp_mode_fixup,
+	.prepare = intel_encoder_prepare,
+	.mode_set = intel_dp_mode_set,
+	.commit = intel_encoder_commit,
+};
+
+static const struct drm_connector_funcs intel_dp_connector_funcs = {
+	.dpms = drm_helper_connector_dpms,
+	.save = intel_dp_save,
+	.restore = intel_dp_restore,
+	.detect = intel_dp_detect,
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.destroy = intel_dp_destroy,
+};
+
+static const struct drm_connector_helper_funcs intel_dp_connector_helper_funcs = {
+	.get_modes = intel_dp_get_modes,
+	.mode_valid = intel_dp_mode_valid,
+	.best_encoder = intel_best_encoder,
+};
+
+static void intel_dp_enc_destroy(struct drm_encoder *encoder)
+{
+	drm_encoder_cleanup(encoder);
+}
+
+static const struct drm_encoder_funcs intel_dp_enc_funcs = {
+	.destroy = intel_dp_enc_destroy,
+};
+
+void
+intel_dp_init(struct drm_device *dev, int output_reg)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_connector *connector;
+	struct intel_output *intel_output;
+	struct intel_dp_priv *dp_priv;
+
+	intel_output = kcalloc(sizeof(struct intel_output) + 
+			       sizeof(struct intel_dp_priv), 1, GFP_KERNEL);
+	if (!intel_output)
+		return;
+
+	dp_priv = (struct intel_dp_priv *)(intel_output + 1);
+
+	connector = &intel_output->base;
+	drm_connector_init(dev, connector, &intel_dp_connector_funcs,
+			   DRM_MODE_CONNECTOR_DisplayPort);
+	drm_connector_helper_add(connector, &intel_dp_connector_helper_funcs);
+
+	intel_output->type = INTEL_OUTPUT_DISPLAYPORT;
+
+	connector->interlace_allowed = true;
+	connector->doublescan_allowed = 0;
+
+	dp_priv->intel_output = intel_output;
+	dp_priv->output_reg = output_reg;
+	dp_priv->has_audio = false;
+	intel_output->dev_priv = dp_priv;
+
+	drm_encoder_init(dev, &intel_output->enc, &intel_dp_enc_funcs,
+			 DRM_MODE_ENCODER_TMDS);
+	drm_encoder_helper_add(&intel_output->enc, &intel_dp_helper_funcs);
+
+	drm_mode_connector_attach_encoder(&intel_output->base,
+					  &intel_output->enc);
+	drm_sysfs_connector_add(connector);
+
+	/* Set up the DDC bus. */
+	intel_dp_i2c_init(intel_output,
+			  (output_reg == DP_B) ? "DPDDC-B" :
+			  (output_reg == DP_C) ? "DPDDC-C" : "DPDDC-D");
+	intel_output->ddc_bus = &dp_priv->adapter;
+
+	/* For G4X desktop chip, PEG_BAND_GAP_DATA 3:0 must first be written
+	 * 0xd.  Failure to do so will result in spurious interrupts being
+	 * generated on the port when a cable is not attached.
+	 */
+	if (IS_G4X(dev) && !IS_GM45(dev)) {
+		u32 temp = I915_READ(PEG_BAND_GAP_DATA);
+		I915_WRITE(PEG_BAND_GAP_DATA, (temp & ~0xf) | 0xd);
+	}
+}
diff --git a/drivers/gpu/drm/i915/intel_dp.h b/drivers/gpu/drm/i915/intel_dp.h
new file mode 100644
index 000000000000..2b38054d3b6d
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dp.h
@@ -0,0 +1,144 @@
+/*
+ * Copyright © 2008 Keith Packard
+ *
+ * Permission to use, copy, modify, distribute, and sell this software and its
+ * documentation for any purpose is hereby granted without fee, provided that
+ * the above copyright notice appear in all copies and that both that copyright
+ * notice and this permission notice appear in supporting documentation, and
+ * that the name of the copyright holders not be used in advertising or
+ * publicity pertaining to distribution of the software without specific,
+ * written prior permission.  The copyright holders make no representations
+ * about the suitability of this software for any purpose.  It is provided "as
+ * is" without express or implied warranty.
+ *
+ * THE COPYRIGHT HOLDERS DISCLAIM ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
+ * INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO
+ * EVENT SHALL THE COPYRIGHT HOLDERS BE LIABLE FOR ANY SPECIAL, INDIRECT OR
+ * CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE,
+ * DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
+ * TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE
+ * OF THIS SOFTWARE.
+ */
+
+#ifndef _INTEL_DP_H_
+#define _INTEL_DP_H_
+
+/* From the VESA DisplayPort spec */
+
+#define AUX_NATIVE_WRITE	0x8
+#define AUX_NATIVE_READ		0x9
+#define AUX_I2C_WRITE		0x0
+#define AUX_I2C_READ		0x1
+#define AUX_I2C_STATUS		0x2
+#define AUX_I2C_MOT		0x4
+
+#define AUX_NATIVE_REPLY_ACK	(0x0 << 4)
+#define AUX_NATIVE_REPLY_NACK	(0x1 << 4)
+#define AUX_NATIVE_REPLY_DEFER	(0x2 << 4)
+#define AUX_NATIVE_REPLY_MASK	(0x3 << 4)
+
+#define AUX_I2C_REPLY_ACK	(0x0 << 6)
+#define AUX_I2C_REPLY_NACK	(0x1 << 6)
+#define AUX_I2C_REPLY_DEFER	(0x2 << 6)
+#define AUX_I2C_REPLY_MASK	(0x3 << 6)
+
+/* AUX CH addresses */
+#define	DP_LINK_BW_SET		0x100
+# define DP_LINK_BW_1_62		    0x06
+# define DP_LINK_BW_2_7			    0x0a
+
+#define DP_LANE_COUNT_SET	0x101
+# define DP_LANE_COUNT_MASK		    0x0f
+# define DP_LANE_COUNT_ENHANCED_FRAME_EN    (1 << 7)
+
+#define DP_TRAINING_PATTERN_SET	0x102
+
+# define DP_TRAINING_PATTERN_DISABLE	    0
+# define DP_TRAINING_PATTERN_1		    1
+# define DP_TRAINING_PATTERN_2		    2
+# define DP_TRAINING_PATTERN_MASK	    0x3
+
+# define DP_LINK_QUAL_PATTERN_DISABLE	    (0 << 2)
+# define DP_LINK_QUAL_PATTERN_D10_2	    (1 << 2)
+# define DP_LINK_QUAL_PATTERN_ERROR_RATE    (2 << 2)
+# define DP_LINK_QUAL_PATTERN_PRBS7	    (3 << 2)
+# define DP_LINK_QUAL_PATTERN_MASK	    (3 << 2)
+
+# define DP_RECOVERED_CLOCK_OUT_EN	    (1 << 4)
+# define DP_LINK_SCRAMBLING_DISABLE	    (1 << 5)
+
+# define DP_SYMBOL_ERROR_COUNT_BOTH	    (0 << 6)
+# define DP_SYMBOL_ERROR_COUNT_DISPARITY    (1 << 6)
+# define DP_SYMBOL_ERROR_COUNT_SYMBOL	    (2 << 6)
+# define DP_SYMBOL_ERROR_COUNT_MASK	    (3 << 6)
+
+#define DP_TRAINING_LANE0_SET		    0x103
+#define DP_TRAINING_LANE1_SET		    0x104
+#define DP_TRAINING_LANE2_SET		    0x105
+#define DP_TRAINING_LANE3_SET		    0x106
+
+# define DP_TRAIN_VOLTAGE_SWING_MASK	    0x3
+# define DP_TRAIN_VOLTAGE_SWING_SHIFT	    0
+# define DP_TRAIN_MAX_SWING_REACHED	    (1 << 2)
+# define DP_TRAIN_VOLTAGE_SWING_400	    (0 << 0)
+# define DP_TRAIN_VOLTAGE_SWING_600	    (1 << 0)
+# define DP_TRAIN_VOLTAGE_SWING_800	    (2 << 0)
+# define DP_TRAIN_VOLTAGE_SWING_1200	    (3 << 0)
+
+# define DP_TRAIN_PRE_EMPHASIS_MASK	    (3 << 3)
+# define DP_TRAIN_PRE_EMPHASIS_0	    (0 << 3)
+# define DP_TRAIN_PRE_EMPHASIS_3_5	    (1 << 3)
+# define DP_TRAIN_PRE_EMPHASIS_6	    (2 << 3)
+# define DP_TRAIN_PRE_EMPHASIS_9_5	    (3 << 3)
+
+# define DP_TRAIN_PRE_EMPHASIS_SHIFT	    3
+# define DP_TRAIN_MAX_PRE_EMPHASIS_REACHED  (1 << 5)
+
+#define DP_DOWNSPREAD_CTRL		    0x107
+# define DP_SPREAD_AMP_0_5		    (1 << 4)
+
+#define DP_MAIN_LINK_CHANNEL_CODING_SET	    0x108
+# define DP_SET_ANSI_8B10B		    (1 << 0)
+
+#define DP_LANE0_1_STATUS		    0x202
+#define DP_LANE2_3_STATUS		    0x203
+
+# define DP_LANE_CR_DONE		    (1 << 0)
+# define DP_LANE_CHANNEL_EQ_DONE	    (1 << 1)
+# define DP_LANE_SYMBOL_LOCKED		    (1 << 2)
+
+#define DP_LANE_ALIGN_STATUS_UPDATED	    0x204
+
+#define DP_INTERLANE_ALIGN_DONE		    (1 << 0)
+#define DP_DOWNSTREAM_PORT_STATUS_CHANGED   (1 << 6)
+#define DP_LINK_STATUS_UPDATED		    (1 << 7)
+
+#define DP_SINK_STATUS			    0x205
+
+#define DP_RECEIVE_PORT_0_STATUS	    (1 << 0)
+#define DP_RECEIVE_PORT_1_STATUS	    (1 << 1)
+
+#define DP_ADJUST_REQUEST_LANE0_1	    0x206
+#define DP_ADJUST_REQUEST_LANE2_3	    0x207
+
+#define DP_ADJUST_VOLTAGE_SWING_LANE0_MASK  0x03
+#define DP_ADJUST_VOLTAGE_SWING_LANE0_SHIFT 0
+#define DP_ADJUST_PRE_EMPHASIS_LANE0_MASK   0x0c
+#define DP_ADJUST_PRE_EMPHASIS_LANE0_SHIFT  2
+#define DP_ADJUST_VOLTAGE_SWING_LANE1_MASK  0x30
+#define DP_ADJUST_VOLTAGE_SWING_LANE1_SHIFT 4
+#define DP_ADJUST_PRE_EMPHASIS_LANE1_MASK   0xc0
+#define DP_ADJUST_PRE_EMPHASIS_LANE1_SHIFT  6
+
+struct i2c_algo_dp_aux_data {
+	bool running;
+	u16 address;
+	int (*aux_ch) (struct i2c_adapter *adapter,
+		       uint8_t *send, int send_bytes,
+		       uint8_t *recv, int recv_bytes);
+};
+
+int
+i2c_dp_aux_add_bus(struct i2c_adapter *adapter);
+
+#endif /* _INTEL_DP_H_ */
diff --git a/drivers/gpu/drm/i915/intel_dp_i2c.c b/drivers/gpu/drm/i915/intel_dp_i2c.c
new file mode 100644
index 000000000000..4e60f14b1a6d
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_dp_i2c.c
@@ -0,0 +1,272 @@
+/*
+ * Copyright © 2009 Keith Packard
+ *
+ * Permission to use, copy, modify, distribute, and sell this software and its
+ * documentation for any purpose is hereby granted without fee, provided that
+ * the above copyright notice appear in all copies and that both that copyright
+ * notice and this permission notice appear in supporting documentation, and
+ * that the name of the copyright holders not be used in advertising or
+ * publicity pertaining to distribution of the software without specific,
+ * written prior permission.  The copyright holders make no representations
+ * about the suitability of this software for any purpose.  It is provided "as
+ * is" without express or implied warranty.
+ *
+ * THE COPYRIGHT HOLDERS DISCLAIM ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
+ * INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO
+ * EVENT SHALL THE COPYRIGHT HOLDERS BE LIABLE FOR ANY SPECIAL, INDIRECT OR
+ * CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE,
+ * DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
+ * TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE
+ * OF THIS SOFTWARE.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/slab.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/i2c.h>
+#include "intel_dp.h"
+
+/* Run a single AUX_CH I2C transaction, writing/reading data as necessary */
+
+#define MODE_I2C_START	1
+#define MODE_I2C_WRITE	2
+#define MODE_I2C_READ	4
+#define MODE_I2C_STOP	8
+
+static int
+i2c_algo_dp_aux_transaction(struct i2c_adapter *adapter, int mode,
+			    uint8_t write_byte, uint8_t *read_byte)
+{
+	struct i2c_algo_dp_aux_data *algo_data = adapter->algo_data;
+	uint16_t address = algo_data->address;
+	uint8_t msg[5];
+	uint8_t reply[2];
+	int msg_bytes;
+	int reply_bytes;
+	int ret;
+
+	/* Set up the command byte */
+	if (mode & MODE_I2C_READ)
+		msg[0] = AUX_I2C_READ << 4;
+	else
+		msg[0] = AUX_I2C_WRITE << 4;
+
+	if (!(mode & MODE_I2C_STOP))
+		msg[0] |= AUX_I2C_MOT << 4;
+
+	msg[1] = address >> 8;
+	msg[2] = address;
+
+	switch (mode) {
+	case MODE_I2C_WRITE:
+		msg[3] = 0;
+		msg[4] = write_byte;
+		msg_bytes = 5;
+		reply_bytes = 1;
+		break;
+	case MODE_I2C_READ:
+		msg[3] = 0;
+		msg_bytes = 4;
+		reply_bytes = 2;
+		break;
+	default:
+		msg_bytes = 3;
+		reply_bytes = 1;
+		break;
+	}
+
+	for (;;) {
+		ret = (*algo_data->aux_ch)(adapter,
+					   msg, msg_bytes,
+					   reply, reply_bytes);
+		if (ret < 0) {
+			printk(KERN_ERR "aux_ch failed %d\n", ret);
+			return ret;
+		}
+		switch (reply[0] & AUX_I2C_REPLY_MASK) {
+		case AUX_I2C_REPLY_ACK:
+			if (mode == MODE_I2C_READ) {
+				*read_byte = reply[1];
+			}
+			return reply_bytes - 1;
+		case AUX_I2C_REPLY_NACK:
+			printk(KERN_ERR "aux_ch nack\n");
+			return -EREMOTEIO;
+		case AUX_I2C_REPLY_DEFER:
+			printk(KERN_ERR "aux_ch defer\n");
+			udelay(100);
+			break;
+		default:
+			printk(KERN_ERR "aux_ch invalid reply 0x%02x\n", reply[0]);
+			return -EREMOTEIO;
+		}
+	}
+}
+
+/*
+ * I2C over AUX CH
+ */
+
+/*
+ * Send the address. If the I2C link is running, this 'restarts'
+ * the connection with the new address, this is used for doing
+ * a write followed by a read (as needed for DDC)
+ */
+static int
+i2c_algo_dp_aux_address(struct i2c_adapter *adapter, u16 address, bool reading)
+{
+	struct i2c_algo_dp_aux_data *algo_data = adapter->algo_data;
+	int mode = MODE_I2C_START;
+	int ret;
+
+	if (reading)
+		mode |= MODE_I2C_READ;
+	else
+		mode |= MODE_I2C_WRITE;
+	algo_data->address = address;
+	algo_data->running = true;
+	ret = i2c_algo_dp_aux_transaction(adapter, mode, 0, NULL);
+	return ret;
+}
+
+/*
+ * Stop the I2C transaction. This closes out the link, sending
+ * a bare address packet with the MOT bit turned off
+ */
+static void
+i2c_algo_dp_aux_stop(struct i2c_adapter *adapter, bool reading)
+{
+	struct i2c_algo_dp_aux_data *algo_data = adapter->algo_data;
+	int mode = MODE_I2C_STOP;
+
+	if (reading)
+		mode |= MODE_I2C_READ;
+	else
+		mode |= MODE_I2C_WRITE;
+	if (algo_data->running) {
+		(void) i2c_algo_dp_aux_transaction(adapter, mode, 0, NULL);
+		algo_data->running = false;
+	}
+}
+
+/*
+ * Write a single byte to the current I2C address, the
+ * the I2C link must be running or this returns -EIO
+ */
+static int
+i2c_algo_dp_aux_put_byte(struct i2c_adapter *adapter, u8 byte)
+{
+	struct i2c_algo_dp_aux_data *algo_data = adapter->algo_data;
+	int ret;
+
+	if (!algo_data->running)
+		return -EIO;
+
+	ret = i2c_algo_dp_aux_transaction(adapter, MODE_I2C_WRITE, byte, NULL);
+	return ret;
+}
+
+/*
+ * Read a single byte from the current I2C address, the
+ * I2C link must be running or this returns -EIO
+ */
+static int
+i2c_algo_dp_aux_get_byte(struct i2c_adapter *adapter, u8 *byte_ret)
+{
+	struct i2c_algo_dp_aux_data *algo_data = adapter->algo_data;
+	int ret;
+
+	if (!algo_data->running)
+		return -EIO;
+
+	ret = i2c_algo_dp_aux_transaction(adapter, MODE_I2C_READ, 0, byte_ret);
+	return ret;
+}
+
+static int
+i2c_algo_dp_aux_xfer(struct i2c_adapter *adapter,
+		     struct i2c_msg *msgs,
+		     int num)
+{
+	int ret = 0;
+	bool reading = false;
+	int m;
+	int b;
+
+	for (m = 0; m < num; m++) {
+		u16 len = msgs[m].len;
+		u8 *buf = msgs[m].buf;
+		reading = (msgs[m].flags & I2C_M_RD) != 0;
+		ret = i2c_algo_dp_aux_address(adapter, msgs[m].addr, reading);
+		if (ret < 0)
+			break;
+		if (reading) {
+			for (b = 0; b < len; b++) {
+				ret = i2c_algo_dp_aux_get_byte(adapter, &buf[b]);
+				if (ret < 0)
+					break;
+			}
+		} else {
+			for (b = 0; b < len; b++) {
+				ret = i2c_algo_dp_aux_put_byte(adapter, buf[b]);
+				if (ret < 0)
+					break;
+			}
+		}
+		if (ret < 0)
+			break;
+	}
+	if (ret >= 0)
+		ret = num;
+	i2c_algo_dp_aux_stop(adapter, reading);
+	printk(KERN_ERR "dp_aux_xfer return %d\n", ret);
+	return ret;
+}
+
+static u32
+i2c_algo_dp_aux_functionality(struct i2c_adapter *adapter)
+{
+	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL |
+	       I2C_FUNC_SMBUS_READ_BLOCK_DATA |
+	       I2C_FUNC_SMBUS_BLOCK_PROC_CALL |
+	       I2C_FUNC_10BIT_ADDR;
+}
+
+static const struct i2c_algorithm i2c_dp_aux_algo = {
+	.master_xfer	= i2c_algo_dp_aux_xfer,
+	.functionality	= i2c_algo_dp_aux_functionality,
+};
+
+static void
+i2c_dp_aux_reset_bus(struct i2c_adapter *adapter)
+{
+	(void) i2c_algo_dp_aux_address(adapter, 0, false);
+	(void) i2c_algo_dp_aux_stop(adapter, false);
+					   
+}
+
+static int
+i2c_dp_aux_prepare_bus(struct i2c_adapter *adapter)
+{
+	adapter->algo = &i2c_dp_aux_algo;
+	adapter->retries = 3;
+	i2c_dp_aux_reset_bus(adapter);
+	return 0;
+}
+
+int
+i2c_dp_aux_add_bus(struct i2c_adapter *adapter)
+{
+	int error;
+	
+	error = i2c_dp_aux_prepare_bus(adapter);
+	if (error)
+		return error;
+	error = i2c_add_adapter(adapter);
+	return error;
+}
+EXPORT_SYMBOL(i2c_dp_aux_add_bus);
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index c5858792c806..004541c935a8 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -54,6 +54,7 @@
 #define INTEL_OUTPUT_LVDS 4
 #define INTEL_OUTPUT_TVOUT 5
 #define INTEL_OUTPUT_HDMI 6
+#define INTEL_OUTPUT_DISPLAYPORT 7
 
 #define INTEL_DVO_CHIP_NONE 0
 #define INTEL_DVO_CHIP_LVDS 1
@@ -116,6 +117,10 @@ extern bool intel_sdvo_init(struct drm_device *dev, int output_device);
 extern void intel_dvo_init(struct drm_device *dev);
 extern void intel_tv_init(struct drm_device *dev);
 extern void intel_lvds_init(struct drm_device *dev);
+extern void intel_dp_init(struct drm_device *dev, int dp_reg);
+void
+intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
+		 struct drm_display_mode *adjusted_mode);
 
 extern void intel_crtc_load_lut(struct drm_crtc *crtc);
 extern void intel_encoder_prepare (struct drm_encoder *encoder);
-- 
cgit v1.2.3-59-g8ed1b


From c8110e52b753f3d105604df84ac06cd6d1645409 Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Wed, 6 May 2009 11:51:10 -0700
Subject: drm/i915: Use hotplug callback to retrain DP link

When a DP monitor is plugged back in, it needs to be retrained if it was
active before.

Signed-off-by: Keith Packard <keithp@keithp.com>
---
 drivers/gpu/drm/i915/intel_dp.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index c57cdab4f4a6..3f8d7b449e70 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -47,6 +47,7 @@ struct intel_dp_priv {
 	uint32_t save_DP;
 	uint8_t  save_link_configuration[DP_LINK_CONFIGURATION_SIZE];
 	bool has_audio;
+	int dpms_mode;
 	uint8_t link_bw;
 	uint8_t lane_count;
 	uint8_t dpcd[4];
@@ -527,6 +528,7 @@ intel_dp_dpms(struct drm_encoder *encoder, int mode)
 		if (!(dp_reg & DP_PORT_EN))
 			intel_dp_link_train(intel_output, dp_priv->DP, dp_priv->link_configuration);
 	}
+	dp_priv->dpms_mode = mode;
 }
 
 /*
@@ -902,7 +904,6 @@ intel_dp_restore(struct drm_connector *connector)
 		intel_dp_link_down(intel_output,  dp_priv->save_DP);
 }
 
-#if 0
 /*
  * According to DP spec
  * 5.1.2:
@@ -929,7 +930,6 @@ intel_dp_check_link_status(struct intel_output *intel_output)
 	if (!intel_channel_eq_ok(link_status, dp_priv->lane_count))
 		intel_dp_link_train(intel_output, dp_priv->DP, dp_priv->link_configuration);
 }
-#endif
 
 /**
  * Uses CRT_HOTPLUG_EN and CRT_HOTPLUG_STAT to detect DP connection.
@@ -1043,6 +1043,15 @@ static const struct drm_encoder_funcs intel_dp_enc_funcs = {
 	.destroy = intel_dp_enc_destroy,
 };
 
+void
+intel_dp_hot_plug(struct intel_output *intel_output)
+{
+	struct intel_dp_priv *dp_priv = intel_output->dev_priv;
+
+	if (dp_priv->dpms_mode == DRM_MODE_DPMS_ON)
+		intel_dp_check_link_status(intel_output);
+}
+
 void
 intel_dp_init(struct drm_device *dev, int output_reg)
 {
@@ -1071,6 +1080,7 @@ intel_dp_init(struct drm_device *dev, int output_reg)
 	dp_priv->intel_output = intel_output;
 	dp_priv->output_reg = output_reg;
 	dp_priv->has_audio = false;
+	dp_priv->dpms_mode = DRM_MODE_DPMS_ON;
 	intel_output->dev_priv = dp_priv;
 
 	drm_encoder_init(dev, &intel_output->enc, &intel_dp_enc_funcs,
@@ -1086,6 +1096,7 @@ intel_dp_init(struct drm_device *dev, int output_reg)
 			  (output_reg == DP_B) ? "DPDDC-B" :
 			  (output_reg == DP_C) ? "DPDDC-C" : "DPDDC-D");
 	intel_output->ddc_bus = &dp_priv->adapter;
+	intel_output->hot_plug = intel_dp_hot_plug;
 
 	/* For G4X desktop chip, PEG_BAND_GAP_DATA 3:0 must first be written
 	 * 0xd.  Failure to do so will result in spurious interrupts being
-- 
cgit v1.2.3-59-g8ed1b


From e4b366996bc58a02b9dc35db3ef83f0454553f50 Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Fri, 5 Jun 2009 19:22:17 -0700
Subject: drm/i915: Split array of DAC limits into separate structures.

The array of DAC limits was only ever referenced with #defined constant
offsets, and keeping those #define values in sync with the array itself was a
nuisance. This will make future changes to the set of DAC limits less
error-prone.

Signed-off-by: Keith Packard <keithp@keithp.com>
---
 drivers/gpu/drm/i915/intel_display.c | 108 +++++++++++++++++------------------
 1 file changed, 51 insertions(+), 57 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 5af55aa0d7a6..73e7b9cecac8 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -128,20 +128,6 @@ struct intel_limit {
 #define I9XX_P2_LVDS_FAST		      7
 #define I9XX_P2_LVDS_SLOW_LIMIT		 112000
 
-#define INTEL_LIMIT_I8XX_DVO_DAC    0
-#define INTEL_LIMIT_I8XX_LVDS	    1
-#define INTEL_LIMIT_I9XX_SDVO_DAC   2
-#define INTEL_LIMIT_I9XX_LVDS	    3
-#define INTEL_LIMIT_G4X_SDVO	    4
-#define INTEL_LIMIT_G4X_HDMI_DAC   5
-#define INTEL_LIMIT_G4X_SINGLE_CHANNEL_LVDS   6
-#define INTEL_LIMIT_G4X_DUAL_CHANNEL_LVDS   7
-#define INTEL_LIMIT_G4X_DISPLAY_PORT	8
-#define INTEL_LIMIT_IGD_SDVO_DAC    9
-#define INTEL_LIMIT_IGD_LVDS	    10
-#define INTEL_LIMIT_IGDNG_SDVO_DAC  11
-#define INTEL_LIMIT_IGDNG_LVDS	    12
-
 /*The parameter is for SDVO on G4x platform*/
 #define G4X_DOT_SDVO_MIN           25000
 #define G4X_DOT_SDVO_MAX           270000
@@ -281,8 +267,7 @@ static bool
 intel_find_pll_g4x_dp(const intel_limit_t *, struct drm_crtc *crtc,
 		      int target, int refclk, intel_clock_t *best_clock);
 
-static const intel_limit_t intel_limits[] = {
-    { /* INTEL_LIMIT_I8XX_DVO_DAC */
+static const intel_limit_t intel_limits_i8xx_dvo = {
         .dot = { .min = I8XX_DOT_MIN,		.max = I8XX_DOT_MAX },
         .vco = { .min = I8XX_VCO_MIN,		.max = I8XX_VCO_MAX },
         .n   = { .min = I8XX_N_MIN,		.max = I8XX_N_MAX },
@@ -294,8 +279,9 @@ static const intel_limit_t intel_limits[] = {
 	.p2  = { .dot_limit = I8XX_P2_SLOW_LIMIT,
 		 .p2_slow = I8XX_P2_SLOW,	.p2_fast = I8XX_P2_FAST },
 	.find_pll = intel_find_best_PLL,
-    },
-    { /* INTEL_LIMIT_I8XX_LVDS */
+};
+
+static const intel_limit_t intel_limits_i8xx_lvds = {
         .dot = { .min = I8XX_DOT_MIN,		.max = I8XX_DOT_MAX },
         .vco = { .min = I8XX_VCO_MIN,		.max = I8XX_VCO_MAX },
         .n   = { .min = I8XX_N_MIN,		.max = I8XX_N_MAX },
@@ -307,8 +293,9 @@ static const intel_limit_t intel_limits[] = {
 	.p2  = { .dot_limit = I8XX_P2_SLOW_LIMIT,
 		 .p2_slow = I8XX_P2_LVDS_SLOW,	.p2_fast = I8XX_P2_LVDS_FAST },
 	.find_pll = intel_find_best_PLL,
-    },
-    { /* INTEL_LIMIT_I9XX_SDVO_DAC */
+};
+	
+static const intel_limit_t intel_limits_i9xx_sdvo = {
         .dot = { .min = I9XX_DOT_MIN,		.max = I9XX_DOT_MAX },
         .vco = { .min = I9XX_VCO_MIN,		.max = I9XX_VCO_MAX },
         .n   = { .min = I9XX_N_MIN,		.max = I9XX_N_MAX },
@@ -320,8 +307,9 @@ static const intel_limit_t intel_limits[] = {
 	.p2  = { .dot_limit = I9XX_P2_SDVO_DAC_SLOW_LIMIT,
 		 .p2_slow = I9XX_P2_SDVO_DAC_SLOW,	.p2_fast = I9XX_P2_SDVO_DAC_FAST },
 	.find_pll = intel_find_best_PLL,
-    },
-    { /* INTEL_LIMIT_I9XX_LVDS */
+};
+
+static const intel_limit_t intel_limits_i9xx_lvds = {
         .dot = { .min = I9XX_DOT_MIN,		.max = I9XX_DOT_MAX },
         .vco = { .min = I9XX_VCO_MIN,		.max = I9XX_VCO_MAX },
         .n   = { .min = I9XX_N_MIN,		.max = I9XX_N_MAX },
@@ -336,9 +324,10 @@ static const intel_limit_t intel_limits[] = {
 	.p2  = { .dot_limit = I9XX_P2_LVDS_SLOW_LIMIT,
 		 .p2_slow = I9XX_P2_LVDS_SLOW,	.p2_fast = I9XX_P2_LVDS_FAST },
 	.find_pll = intel_find_best_PLL,
-    },
+};
+
     /* below parameter and function is for G4X Chipset Family*/
-    { /* INTEL_LIMIT_G4X_SDVO */
+static const intel_limit_t intel_limits_g4x_sdvo = {
 	.dot = { .min = G4X_DOT_SDVO_MIN,	.max = G4X_DOT_SDVO_MAX },
 	.vco = { .min = G4X_VCO_MIN,	        .max = G4X_VCO_MAX},
 	.n   = { .min = G4X_N_SDVO_MIN,	        .max = G4X_N_SDVO_MAX },
@@ -352,8 +341,9 @@ static const intel_limit_t intel_limits[] = {
 		 .p2_fast = G4X_P2_SDVO_FAST
 	},
 	.find_pll = intel_g4x_find_best_PLL,
-    },
-    { /* INTEL_LIMIT_G4X_HDMI_DAC */
+};
+
+static const intel_limit_t intel_limits_g4x_hdmi = {
 	.dot = { .min = G4X_DOT_HDMI_DAC_MIN,	.max = G4X_DOT_HDMI_DAC_MAX },
 	.vco = { .min = G4X_VCO_MIN,	        .max = G4X_VCO_MAX},
 	.n   = { .min = G4X_N_HDMI_DAC_MIN,	.max = G4X_N_HDMI_DAC_MAX },
@@ -367,8 +357,9 @@ static const intel_limit_t intel_limits[] = {
 		 .p2_fast = G4X_P2_HDMI_DAC_FAST
 	},
 	.find_pll = intel_g4x_find_best_PLL,
-    },
-    { /* INTEL_LIMIT_G4X_SINGLE_CHANNEL_LVDS */
+};
+
+static const intel_limit_t intel_limits_g4x_single_channel_lvds = {
 	.dot = { .min = G4X_DOT_SINGLE_CHANNEL_LVDS_MIN,
 		 .max = G4X_DOT_SINGLE_CHANNEL_LVDS_MAX },
 	.vco = { .min = G4X_VCO_MIN,
@@ -390,8 +381,9 @@ static const intel_limit_t intel_limits[] = {
 		 .p2_fast = G4X_P2_SINGLE_CHANNEL_LVDS_FAST
 	},
 	.find_pll = intel_g4x_find_best_PLL,
-    },
-    { /* INTEL_LIMIT_G4X_DUAL_CHANNEL_LVDS */
+};
+
+static const intel_limit_t intel_limits_g4x_dual_channel_lvds = {
 	.dot = { .min = G4X_DOT_DUAL_CHANNEL_LVDS_MIN,
 		 .max = G4X_DOT_DUAL_CHANNEL_LVDS_MAX },
 	.vco = { .min = G4X_VCO_MIN,
@@ -413,8 +405,9 @@ static const intel_limit_t intel_limits[] = {
 		 .p2_fast = G4X_P2_DUAL_CHANNEL_LVDS_FAST
 	},
 	.find_pll = intel_g4x_find_best_PLL,
-    },
-    {   /* INTEL_LIMIT_G4X_DISPLAY_PORT */
+};
+
+static const intel_limit_t intel_limits_g4x_display_port = {
         .dot = { .min = G4X_DOT_DISPLAY_PORT_MIN,
                  .max = G4X_DOT_DISPLAY_PORT_MAX },
         .vco = { .min = G4X_VCO_MIN,
@@ -435,8 +428,9 @@ static const intel_limit_t intel_limits[] = {
                  .p2_slow = G4X_P2_DISPLAY_PORT_SLOW,
                  .p2_fast = G4X_P2_DISPLAY_PORT_FAST },
         .find_pll = intel_find_pll_g4x_dp,
-    },
-    { /* INTEL_LIMIT_IGD_SDVO */
+};
+
+static const intel_limit_t intel_limits_igd_sdvo = {
         .dot = { .min = I9XX_DOT_MIN,		.max = I9XX_DOT_MAX},
         .vco = { .min = IGD_VCO_MIN,		.max = IGD_VCO_MAX },
         .n   = { .min = IGD_N_MIN,		.max = IGD_N_MAX },
@@ -448,8 +442,9 @@ static const intel_limit_t intel_limits[] = {
 	.p2  = { .dot_limit = I9XX_P2_SDVO_DAC_SLOW_LIMIT,
 		 .p2_slow = I9XX_P2_SDVO_DAC_SLOW,	.p2_fast = I9XX_P2_SDVO_DAC_FAST },
 	.find_pll = intel_find_best_PLL,
-    },
-    { /* INTEL_LIMIT_IGD_LVDS */
+};
+
+static const intel_limit_t intel_limits_igd_lvds = {
         .dot = { .min = I9XX_DOT_MIN,		.max = I9XX_DOT_MAX },
         .vco = { .min = IGD_VCO_MIN,		.max = IGD_VCO_MAX },
         .n   = { .min = IGD_N_MIN,		.max = IGD_N_MAX },
@@ -462,8 +457,9 @@ static const intel_limit_t intel_limits[] = {
 	.p2  = { .dot_limit = I9XX_P2_LVDS_SLOW_LIMIT,
 		 .p2_slow = I9XX_P2_LVDS_SLOW,	.p2_fast = I9XX_P2_LVDS_SLOW },
 	.find_pll = intel_find_best_PLL,
-    },
-    { /* INTEL_LIMIT_IGDNG_SDVO_DAC */
+};
+
+static const intel_limit_t intel_limits_igdng_sdvo = {
 	.dot = { .min = IGDNG_DOT_MIN,          .max = IGDNG_DOT_MAX },
 	.vco = { .min = IGDNG_VCO_MIN,          .max = IGDNG_VCO_MAX },
 	.n   = { .min = IGDNG_N_MIN,            .max = IGDNG_N_MAX },
@@ -476,8 +472,9 @@ static const intel_limit_t intel_limits[] = {
 		 .p2_slow = IGDNG_P2_SDVO_DAC_SLOW,
 		 .p2_fast = IGDNG_P2_SDVO_DAC_FAST },
 	.find_pll = intel_igdng_find_best_PLL,
-    },
-    { /* INTEL_LIMIT_IGDNG_LVDS */
+};
+
+static const intel_limit_t intel_limits_igdng_lvds = {
 	.dot = { .min = IGDNG_DOT_MIN,          .max = IGDNG_DOT_MAX },
 	.vco = { .min = IGDNG_VCO_MIN,          .max = IGDNG_VCO_MAX },
 	.n   = { .min = IGDNG_N_MIN,            .max = IGDNG_N_MAX },
@@ -490,16 +487,15 @@ static const intel_limit_t intel_limits[] = {
 		 .p2_slow = IGDNG_P2_LVDS_SLOW,
 		 .p2_fast = IGDNG_P2_LVDS_FAST },
 	.find_pll = intel_igdng_find_best_PLL,
-    },
 };
 
 static const intel_limit_t *intel_igdng_limit(struct drm_crtc *crtc)
 {
 	const intel_limit_t *limit;
 	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS))
-		limit = &intel_limits[INTEL_LIMIT_IGDNG_LVDS];
+		limit = &intel_limits_igdng_lvds;
 	else
-		limit = &intel_limits[INTEL_LIMIT_IGDNG_SDVO_DAC];
+		limit = &intel_limits_igdng_sdvo;
 
 	return limit;
 }
@@ -514,21 +510,19 @@ static const intel_limit_t *intel_g4x_limit(struct drm_crtc *crtc)
 		if ((I915_READ(LVDS) & LVDS_CLKB_POWER_MASK) ==
 		    LVDS_CLKB_POWER_UP)
 			/* LVDS with dual channel */
-			limit = &intel_limits
-					[INTEL_LIMIT_G4X_DUAL_CHANNEL_LVDS];
+			limit = &intel_limits_g4x_dual_channel_lvds;
 		else
 			/* LVDS with dual channel */
-			limit = &intel_limits
-					[INTEL_LIMIT_G4X_SINGLE_CHANNEL_LVDS];
+			limit = &intel_limits_g4x_single_channel_lvds;
 	} else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_HDMI) ||
 		   intel_pipe_has_type(crtc, INTEL_OUTPUT_ANALOG)) {
-		limit = &intel_limits[INTEL_LIMIT_G4X_HDMI_DAC];
+		limit = &intel_limits_g4x_hdmi;
 	} else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_SDVO)) {
-		limit = &intel_limits[INTEL_LIMIT_G4X_SDVO];
+		limit = &intel_limits_g4x_sdvo;
 	} else if (intel_pipe_has_type (crtc, INTEL_OUTPUT_DISPLAYPORT)) {
-		limit = &intel_limits[INTEL_LIMIT_G4X_DISPLAY_PORT];
+		limit = &intel_limits_g4x_display_port;
 	} else /* The option is for other outputs */
-		limit = &intel_limits[INTEL_LIMIT_I9XX_SDVO_DAC];
+		limit = &intel_limits_i9xx_sdvo;
 
 	return limit;
 }
@@ -544,19 +538,19 @@ static const intel_limit_t *intel_limit(struct drm_crtc *crtc)
 		limit = intel_g4x_limit(crtc);
 	} else if (IS_I9XX(dev) && !IS_IGD(dev)) {
 		if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS))
-			limit = &intel_limits[INTEL_LIMIT_I9XX_LVDS];
+			limit = &intel_limits_i9xx_lvds;
 		else
-			limit = &intel_limits[INTEL_LIMIT_I9XX_SDVO_DAC];
+			limit = &intel_limits_i9xx_sdvo;
 	} else if (IS_IGD(dev)) {
 		if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS))
-			limit = &intel_limits[INTEL_LIMIT_IGD_LVDS];
+			limit = &intel_limits_igd_lvds;
 		else
-			limit = &intel_limits[INTEL_LIMIT_IGD_SDVO_DAC];
+			limit = &intel_limits_igd_sdvo;
 	} else {
 		if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS))
-			limit = &intel_limits[INTEL_LIMIT_I8XX_LVDS];
+			limit = &intel_limits_i8xx_lvds;
 		else
-			limit = &intel_limits[INTEL_LIMIT_I8XX_DVO_DAC];
+			limit = &intel_limits_i8xx_dvo;
 	}
 	return limit;
 }
-- 
cgit v1.2.3-59-g8ed1b


From b11248df4c0decb1e473d5025f237be32c0f67bb Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Thu, 11 Jun 2009 22:28:56 -0700
Subject: drm/i915: Add CLKCFG register definition

The CLKCFG register holds information about the GMCH plls and input clock
values.

Signed-off-by: Keith Packard <keithp@keithp.com>
---
 drivers/gpu/drm/i915/i915_reg.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index f6237a0b1133..544d5677a2fa 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -569,6 +569,19 @@
 #define C0DRB3			0x10206
 #define C1DRB3			0x10606
 
+/* Clocking configuration register */
+#define CLKCFG			0x10c00
+#define CLKCFG_FSB_400					(0 << 0)	/* hrawclk 100 */
+#define CLKCFG_FSB_533					(1 << 0)	/* hrawclk 133 */
+#define CLKCFG_FSB_667					(3 << 0)	/* hrawclk 166 */
+#define CLKCFG_FSB_800					(2 << 0)	/* hrawclk 200 */
+#define CLKCFG_FSB_1067					(6 << 0)	/* hrawclk 266 */
+#define CLKCFG_FSB_1333					(7 << 0)	/* hrawclk 333 */
+/* this is a guess, could be 5 as well */
+#define CLKCFG_FSB_1600					(4 << 0)	/* hrawclk 400 */
+#define CLKCFG_FSB_1600_ALT				(5 << 0)	/* hrawclk 400 */
+#define CLKCFG_FSB_MASK					(7 << 0)
+ 
 /** GM965 GM45 render standby register */
 #define MCHBAR_RENDER_STANDBY	0x111B8
 
-- 
cgit v1.2.3-59-g8ed1b


From a5b3da543d4882d57a2f3e05d37ad8e1e1453489 Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Thu, 11 Jun 2009 22:30:32 -0700
Subject: drm/i915: Clarify error returns from display port aux channel I/O

Use distinct error return values for each kind of aux channel I/O failure.

Signed-off-by: Keith Packard <keithp@keithp.com>
---
 drivers/gpu/drm/i915/intel_dp.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 3f8d7b449e70..818fe34f2b5c 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -209,15 +209,19 @@ intel_dp_aux_ch(struct intel_output *intel_output,
 
 	if ((status & DP_AUX_CH_CTL_DONE) == 0) {
 		printk(KERN_ERR "dp_aux_ch not done status 0x%08x\n", status);
-		return -1;
+		return -EBUSY;
 	}
 
 	/* Check for timeout or receive error.
 	 * Timeouts occur when the sink is not connected
 	 */
-	if (status & (DP_AUX_CH_CTL_TIME_OUT_ERROR | DP_AUX_CH_CTL_RECEIVE_ERROR)) {
-		printk(KERN_ERR "dp_aux_ch error status 0x%08x\n", status);
-		return -1;
+	if (status & DP_AUX_CH_CTL_RECEIVE_ERROR) {
+		printk(KERN_ERR "dp_aux_ch receive error status 0x%08x\n", status);
+		return -EIO;
+	}
+	if (status & DP_AUX_CH_CTL_TIME_OUT_ERROR) {
+		printk(KERN_ERR "dp_aux_ch timeout status 0x%08x\n", status);
+		return -ETIMEDOUT;
 	}
 
 	/* Unload any bytes sent back from the other side */
@@ -263,7 +267,7 @@ intel_dp_aux_native_write(struct intel_output *intel_output,
 		else if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_DEFER)
 			udelay(100);
 		else
-			return -1;
+			return -EIO;
 	}
 	return send_bytes;
 }
@@ -299,7 +303,9 @@ intel_dp_aux_native_read(struct intel_output *intel_output,
 	for (;;) {
 		ret = intel_dp_aux_ch(intel_output, msg, msg_bytes,
 				      reply, reply_bytes);
-		if (ret <= 0)
+		if (ret == 0)
+			return -EPROTO;
+		if (ret < 0)
 			return ret;
 		ack = reply[0];
 		if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_ACK) {
@@ -309,7 +315,7 @@ intel_dp_aux_native_read(struct intel_output *intel_output,
 		else if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_DEFER)
 			udelay(100);
 		else
-			return -1;
+			return -EIO;
 	}
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From fb0f8fbf97e8a25074c81c629500d94cafa9e366 Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Thu, 11 Jun 2009 22:31:31 -0700
Subject: drm/i915: Generate 2MHz clock for display port aux channel I/O. Retry
 I/O.

The display port aux channel clock is taken from the hrawclk value, which is
provided to the chip as the FSB frequency (as far as I can determine). The
strapping values for that are available in the CLKCFG register, now used to
select an appropriate divider to generate a 2MHz clock.

In addition, the DisplayPort spec requires that each aux channel I/O be
retried 'at least 3 times' in case the sink is idle when the first request
comes in.

Signed-off-by: Keith Packard <keithp@keithp.com>
---
 drivers/gpu/drm/i915/intel_dp.c | 102 +++++++++++++++++++++++++++-------------
 1 file changed, 70 insertions(+), 32 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 818fe34f2b5c..8f8d37d5663a 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -154,6 +154,36 @@ unpack_aux(uint32_t src, uint8_t *dst, int dst_bytes)
 		dst[i] = src >> ((3-i) * 8);
 }
 
+/* hrawclock is 1/4 the FSB frequency */
+static int
+intel_hrawclk(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t clkcfg;
+
+	clkcfg = I915_READ(CLKCFG);
+	switch (clkcfg & CLKCFG_FSB_MASK) {
+	case CLKCFG_FSB_400:
+		return 100;
+	case CLKCFG_FSB_533:
+		return 133;
+	case CLKCFG_FSB_667:
+		return 166;
+	case CLKCFG_FSB_800:
+		return 200;
+	case CLKCFG_FSB_1067:
+		return 266;
+	case CLKCFG_FSB_1333:
+		return 333;
+	/* these two are just a guess; one of them might be right */
+	case CLKCFG_FSB_1600:
+	case CLKCFG_FSB_1600_ALT:
+		return 400;
+	default:
+		return 133;
+	}
+}
+
 static int
 intel_dp_aux_ch(struct intel_output *intel_output,
 		uint8_t *send, int send_bytes,
@@ -169,44 +199,52 @@ intel_dp_aux_ch(struct intel_output *intel_output,
 	int recv_bytes;
 	uint32_t ctl;
 	uint32_t status;
-
-	/* Load the send data into the aux channel data registers */
-	for (i = 0; i < send_bytes; i += 4) {
-		uint32_t    d = pack_aux(send + i, send_bytes - i);;
-
-		I915_WRITE(ch_data + i, d);
-	}
+	uint32_t aux_clock_divider;
+	int try;
 
 	/* The clock divider is based off the hrawclk,
-	 * and would like to run at 2MHz. The 133 below assumes
-	 * a 266MHz hrawclk; need to figure out how we're supposed
-	 * to know what hrawclk is...
+	 * and would like to run at 2MHz. So, take the
+	 * hrawclk value and divide by 2 and use that
 	 */
-	ctl = (DP_AUX_CH_CTL_SEND_BUSY |
-	       DP_AUX_CH_CTL_TIME_OUT_1600us |
-	       (send_bytes << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) |
-	       (5 << DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT) |
-	       (133 << DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT) |
-	       DP_AUX_CH_CTL_TIME_OUT_ERROR |
-	       DP_AUX_CH_CTL_RECEIVE_ERROR);
-
-	/* Send the command and wait for it to complete */
-	I915_WRITE(ch_ctl, ctl);
-	(void) I915_READ(ch_ctl);
-	for (;;) {
-		udelay(100);
-		status = I915_READ(ch_ctl);
-		if ((status & DP_AUX_CH_CTL_SEND_BUSY) == 0)
+	aux_clock_divider = intel_hrawclk(dev) / 2;
+	/* Must try at least 3 times according to DP spec */
+	for (try = 0; try < 5; try++) {
+		/* Load the send data into the aux channel data registers */
+		for (i = 0; i < send_bytes; i += 4) {
+			uint32_t    d = pack_aux(send + i, send_bytes - i);;
+	
+			I915_WRITE(ch_data + i, d);
+		}
+	
+		ctl = (DP_AUX_CH_CTL_SEND_BUSY |
+		       DP_AUX_CH_CTL_TIME_OUT_400us |
+		       (send_bytes << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) |
+		       (5 << DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT) |
+		       (aux_clock_divider << DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT) |
+		       DP_AUX_CH_CTL_DONE |
+		       DP_AUX_CH_CTL_TIME_OUT_ERROR |
+		       DP_AUX_CH_CTL_RECEIVE_ERROR);
+	
+		/* Send the command and wait for it to complete */
+		I915_WRITE(ch_ctl, ctl);
+		(void) I915_READ(ch_ctl);
+		for (;;) {
+			udelay(100);
+			status = I915_READ(ch_ctl);
+			if ((status & DP_AUX_CH_CTL_SEND_BUSY) == 0)
+				break;
+		}
+	
+		/* Clear done status and any errors */
+		I915_WRITE(ch_ctl, (ctl |
+				DP_AUX_CH_CTL_DONE |
+				DP_AUX_CH_CTL_TIME_OUT_ERROR |
+				DP_AUX_CH_CTL_RECEIVE_ERROR));
+		(void) I915_READ(ch_ctl);
+		if ((status & DP_AUX_CH_CTL_TIME_OUT_ERROR) == 0)
 			break;
 	}
 
-	/* Clear done status and any errors */
-	I915_WRITE(ch_ctl, (ctl |
-			DP_AUX_CH_CTL_DONE |
-			DP_AUX_CH_CTL_TIME_OUT_ERROR |
-			DP_AUX_CH_CTL_RECEIVE_ERROR));
-	(void) I915_READ(ch_ctl);
-
 	if ((status & DP_AUX_CH_CTL_DONE) == 0) {
 		printk(KERN_ERR "dp_aux_ch not done status 0x%08x\n", status);
 		return -EBUSY;
-- 
cgit v1.2.3-59-g8ed1b


From 6ddbbbfe52f35301ef5a1b595f912d8d2b3ec143 Mon Sep 17 00:00:00 2001
From: Mike Sager <sager@netapp.com>
Date: Tue, 16 Jun 2009 04:20:47 +0300
Subject: nfsd41: Remove ip address collision detection case

Verified that cthon and pynfs exchange id tests pass (except for the
two expected fails: EID8 and EID50)

Signed-off-by: Mike Sager <sager@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4state.c | 18 ++++++------------
 1 file changed, 6 insertions(+), 12 deletions(-)

diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 99570c49add5..ef6944b19f06 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -1247,12 +1247,6 @@ nfsd4_exchange_id(struct svc_rqst *rqstp,
 			expire_client(conf);
 			goto out_new;
 		}
-		if (ip_addr != conf->cl_addr &&
-		    !(exid->flags & EXCHGID4_FLAG_UPD_CONFIRMED_REC_A)) {
-			/* Client collision. 18.35.4 case 3 */
-			status = nfserr_clid_inuse;
-			goto out;
-		}
 		/*
 		 * Set bit when the owner id and verifier map to an already
 		 * confirmed client id (18.35.3).
@@ -1266,12 +1260,12 @@ nfsd4_exchange_id(struct svc_rqst *rqstp,
 		copy_verf(conf, &verf);
 		new = conf;
 		goto out_copy;
-	} else {
-		/* 18.35.4 case 7 */
-		if (exid->flags & EXCHGID4_FLAG_UPD_CONFIRMED_REC_A) {
-			status = nfserr_noent;
-			goto out;
-		}
+	}
+
+	/* 18.35.4 case 7 */
+	if (exid->flags & EXCHGID4_FLAG_UPD_CONFIRMED_REC_A) {
+		status = nfserr_noent;
+		goto out;
 	}
 
 	unconf  = find_unconfirmed_client_by_str(dname, strhashval, true);
-- 
cgit v1.2.3-59-g8ed1b


From ef52bff8409bba78b042f1bcf33a0f49debc9774 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Tue, 16 Jun 2009 04:20:50 +0300
Subject: nfsd41: Backchannel: cleanup nfs4.0 callback encode routines

Mimic the client and prepare to share the back channel xdr with NFSv4.1.
Bump the number of operations in each encode routine, then backfill the
number of operations.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4callback.c | 24 ++++++++++++++++--------
 1 file changed, 16 insertions(+), 8 deletions(-)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index f4fab69a8c30..353eb4a0b847 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -140,8 +140,9 @@ struct nfs4_cb_compound_hdr {
 	int		status;
 	u32		ident;
 	u32		nops;
+	__be32		*nops_p;
 	u32		taglen;
-	char *		tag;
+	char		*tag;
 };
 
 static struct {
@@ -201,7 +202,7 @@ nfs_cb_stat_to_errno(int stat)
  * XDR encode
  */
 
-static int
+static void
 encode_cb_compound_hdr(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr)
 {
 	__be32 * p;
@@ -210,12 +211,18 @@ encode_cb_compound_hdr(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr)
 	WRITE32(0);            /* tag length is always 0 */
 	WRITE32(NFS4_MINOR_VERSION);
 	WRITE32(hdr->ident);
+	hdr->nops_p = p;
 	WRITE32(hdr->nops);
-	return 0;
 }
 
-static int
-encode_cb_recall(struct xdr_stream *xdr, struct nfs4_delegation *dp)
+static void encode_cb_nops(struct nfs4_cb_compound_hdr *hdr)
+{
+	*hdr->nops_p = htonl(hdr->nops);
+}
+
+static void
+encode_cb_recall(struct xdr_stream *xdr, struct nfs4_delegation *dp,
+		struct nfs4_cb_compound_hdr *hdr)
 {
 	__be32 *p;
 	int len = dp->dl_fh.fh_size;
@@ -227,7 +234,7 @@ encode_cb_recall(struct xdr_stream *xdr, struct nfs4_delegation *dp)
 	WRITE32(0); /* truncate optimization not implemented */
 	WRITE32(len);
 	WRITEMEM(&dp->dl_fh.fh_base, len);
-	return 0;
+	hdr->nops++;
 }
 
 static int
@@ -246,12 +253,13 @@ nfs4_xdr_enc_cb_recall(struct rpc_rqst *req, __be32 *p, struct nfs4_delegation *
 	struct xdr_stream xdr;
 	struct nfs4_cb_compound_hdr hdr = {
 		.ident = args->dl_ident,
-		.nops   = 1,
 	};
 
 	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
 	encode_cb_compound_hdr(&xdr, &hdr);
-	return (encode_cb_recall(&xdr, args));
+	encode_cb_recall(&xdr, args, &hdr);
+	encode_cb_nops(&hdr);
+	return 0;
 }
 
 
-- 
cgit v1.2.3-59-g8ed1b


From ab52ae6db035fa425f90146327ab7d2c5d3e5654 Mon Sep 17 00:00:00 2001
From: Andy Adamson <andros@netapp.com>
Date: Tue, 16 Jun 2009 04:20:53 +0300
Subject: nfsd41: Backchannel: minorversion support for the back channel

Prepare to share backchannel code with NFSv4.1.

Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
[nfsd41: use nfsd4_cb_sequence for callback minorversion]
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 fs/nfsd/nfs4callback.c     | 3 ++-
 fs/nfsd/nfs4state.c        | 1 +
 include/linux/nfsd/state.h | 3 ++-
 3 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 353eb4a0b847..3fd23f7aceca 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -141,6 +141,7 @@ struct nfs4_cb_compound_hdr {
 	u32		ident;
 	u32		nops;
 	__be32		*nops_p;
+	u32		minorversion;
 	u32		taglen;
 	char		*tag;
 };
@@ -209,7 +210,7 @@ encode_cb_compound_hdr(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr)
 
 	RESERVE_SPACE(16);
 	WRITE32(0);            /* tag length is always 0 */
-	WRITE32(NFS4_MINOR_VERSION);
+	WRITE32(hdr->minorversion);
 	WRITE32(hdr->ident);
 	hdr->nops_p = p;
 	WRITE32(hdr->nops);
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index ef6944b19f06..980a216a48c8 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -984,6 +984,7 @@ gen_callback(struct nfs4_client *clp, struct nfsd4_setclientid *se)
 	if ( !(parse_ipv4(se->se_callback_addr_len, se->se_callback_addr_val,
 	                 &cb->cb_addr, &cb->cb_port)))
 		goto out_err;
+	cb->cb_minorversion = 0;
 	cb->cb_prog = se->se_callback_prog;
 	cb->cb_ident = se->se_callback_ident;
 	return;
diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h
index 105cc100de05..f5a95fd34312 100644
--- a/include/linux/nfsd/state.h
+++ b/include/linux/nfsd/state.h
@@ -85,7 +85,8 @@ struct nfs4_cb_conn {
 	u32                     cb_addr;
 	unsigned short          cb_port;
 	u32                     cb_prog;
-	u32                     cb_ident;
+	u32			cb_minorversion;
+	u32                     cb_ident;	/* minorversion 0 only */
 	/* RPC client info */
 	atomic_t		cb_set;     /* successful CB_NULL call */
 	struct rpc_clnt *       cb_client;
-- 
cgit v1.2.3-59-g8ed1b


From 47fcb03fefee2501e79176932a4184fc24d6f8ec Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Mon, 18 May 2009 17:47:56 -0400
Subject: SUNRPC: Fix the TCP server's send buffer accounting

Currently, the sunrpc server is refusing to allow us to process new RPC
calls if the TCP send buffer is 2/3 full, even if we do actually have
enough free space to guarantee that we can send another request.
The following patch fixes svc_tcp_has_wspace() so that we only stop
processing requests if we know that the socket buffer cannot possibly fit
another reply.

It also fixes the tcp write_space() callback so that we only clear the
SOCK_NOSPACE flag when the TCP send buffer is less than 2/3 full.
This should ensure that the send window will grow as per the standard TCP
socket code.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 net/sunrpc/svcsock.c | 35 ++++++++++++++++++-----------------
 1 file changed, 18 insertions(+), 17 deletions(-)

diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 004a2f9dc432..b09c80c56ee3 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -380,6 +380,7 @@ static void svc_sock_setbufsize(struct socket *sock, unsigned int snd,
 	sock->sk->sk_sndbuf = snd * 2;
 	sock->sk->sk_rcvbuf = rcv * 2;
 	sock->sk->sk_userlocks |= SOCK_SNDBUF_LOCK|SOCK_RCVBUF_LOCK;
+	sock->sk->sk_write_space(sock->sk);
 	release_sock(sock->sk);
 #endif
 }
@@ -421,6 +422,15 @@ static void svc_write_space(struct sock *sk)
 	}
 }
 
+static void svc_tcp_write_space(struct sock *sk)
+{
+	struct socket *sock = sk->sk_socket;
+
+	if (sk_stream_wspace(sk) >= sk_stream_min_wspace(sk) && sock)
+		clear_bit(SOCK_NOSPACE, &sock->flags);
+	svc_write_space(sk);
+}
+
 /*
  * Copy the UDP datagram's destination address to the rqstp structure.
  * The 'destination' address in this case is the address to which the
@@ -1015,25 +1025,16 @@ static void svc_tcp_prep_reply_hdr(struct svc_rqst *rqstp)
 static int svc_tcp_has_wspace(struct svc_xprt *xprt)
 {
 	struct svc_sock *svsk =	container_of(xprt, struct svc_sock, sk_xprt);
-	struct svc_serv	*serv = svsk->sk_xprt.xpt_server;
+	struct svc_serv *serv = svsk->sk_xprt.xpt_server;
 	int required;
-	int wspace;
 
-	/*
-	 * Set the SOCK_NOSPACE flag before checking the available
-	 * sock space.
-	 */
+	if (test_bit(XPT_LISTENER, &xprt->xpt_flags))
+		return 1;
+	required = atomic_read(&xprt->xpt_reserved) + serv->sv_max_mesg;
+	if (sk_stream_wspace(svsk->sk_sk) >= required)
+		return 1;
 	set_bit(SOCK_NOSPACE, &svsk->sk_sock->flags);
-	required = atomic_read(&svsk->sk_xprt.xpt_reserved) + serv->sv_max_mesg;
-	wspace = sk_stream_wspace(svsk->sk_sk);
-
-	if (wspace < sk_stream_min_wspace(svsk->sk_sk))
-		return 0;
-	if (required * 2 > wspace)
-		return 0;
-
-	clear_bit(SOCK_NOSPACE, &svsk->sk_sock->flags);
-	return 1;
+	return 0;
 }
 
 static struct svc_xprt *svc_tcp_create(struct svc_serv *serv,
@@ -1089,7 +1090,7 @@ static void svc_tcp_init(struct svc_sock *svsk, struct svc_serv *serv)
 		dprintk("setting up TCP socket for reading\n");
 		sk->sk_state_change = svc_tcp_state_change;
 		sk->sk_data_ready = svc_tcp_data_ready;
-		sk->sk_write_space = svc_write_space;
+		sk->sk_write_space = svc_tcp_write_space;
 
 		svsk->sk_reclen = 0;
 		svsk->sk_tcplen = 0;
-- 
cgit v1.2.3-59-g8ed1b


From bb664f49f8be17d7b8bf9821144e8a53d7fcfe8a Mon Sep 17 00:00:00 2001
From: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Date: Wed, 17 Jun 2009 21:54:47 +0000
Subject: af_iucv: Change if condition in sendmsg() for more readability

Change the if condition to exit sendmsg() if the socket in not connected.

Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Ursula Braun <ursula.braun@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/iucv/af_iucv.c | 163 ++++++++++++++++++++++++++---------------------------
 1 file changed, 81 insertions(+), 82 deletions(-)

diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c
index a9b3a6f9ea95..42b7198a6883 100644
--- a/net/iucv/af_iucv.c
+++ b/net/iucv/af_iucv.c
@@ -747,108 +747,107 @@ static int iucv_sock_sendmsg(struct kiocb *iocb, struct socket *sock,
 		goto out;
 	}
 
-	if (sk->sk_state == IUCV_CONNECTED) {
-		/* initialize defaults */
-		cmsg_done   = 0;	/* check for duplicate headers */
-		txmsg.class = 0;
-
-		/* iterate over control messages */
-		for (cmsg = CMSG_FIRSTHDR(msg); cmsg;
-		     cmsg = CMSG_NXTHDR(msg, cmsg)) {
-
-			if (!CMSG_OK(msg, cmsg)) {
-				err = -EINVAL;
-				goto out;
-			}
+	/* Return if the socket is not in connected state */
+	if (sk->sk_state != IUCV_CONNECTED) {
+		err = -ENOTCONN;
+		goto out;
+	}
 
-			if (cmsg->cmsg_level != SOL_IUCV)
-				continue;
+	/* initialize defaults */
+	cmsg_done   = 0;	/* check for duplicate headers */
+	txmsg.class = 0;
 
-			if (cmsg->cmsg_type & cmsg_done) {
-				err = -EINVAL;
-				goto out;
-			}
-			cmsg_done |= cmsg->cmsg_type;
+	/* iterate over control messages */
+	for (cmsg = CMSG_FIRSTHDR(msg); cmsg;
+		cmsg = CMSG_NXTHDR(msg, cmsg)) {
 
-			switch (cmsg->cmsg_type) {
-			case SCM_IUCV_TRGCLS:
-				if (cmsg->cmsg_len != CMSG_LEN(TRGCLS_SIZE)) {
-					err = -EINVAL;
-					goto out;
-				}
+		if (!CMSG_OK(msg, cmsg)) {
+			err = -EINVAL;
+			goto out;
+		}
 
-				/* set iucv message target class */
-				memcpy(&txmsg.class,
-					(void *) CMSG_DATA(cmsg), TRGCLS_SIZE);
+		if (cmsg->cmsg_level != SOL_IUCV)
+			continue;
 
-				break;
+		if (cmsg->cmsg_type & cmsg_done) {
+			err = -EINVAL;
+			goto out;
+		}
+		cmsg_done |= cmsg->cmsg_type;
 
-			default:
+		switch (cmsg->cmsg_type) {
+		case SCM_IUCV_TRGCLS:
+			if (cmsg->cmsg_len != CMSG_LEN(TRGCLS_SIZE)) {
 				err = -EINVAL;
 				goto out;
-				break;
 			}
-		}
 
-		/* allocate one skb for each iucv message:
-		 * this is fine for SOCK_SEQPACKET (unless we want to support
-		 * segmented records using the MSG_EOR flag), but
-		 * for SOCK_STREAM we might want to improve it in future */
-		if (!(skb = sock_alloc_send_skb(sk, len,
-						msg->msg_flags & MSG_DONTWAIT,
-						&err)))
-			goto out;
+			/* set iucv message target class */
+			memcpy(&txmsg.class,
+				(void *) CMSG_DATA(cmsg), TRGCLS_SIZE);
 
-		if (memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len)) {
-			err = -EFAULT;
-			goto fail;
+			break;
+
+		default:
+			err = -EINVAL;
+			goto out;
+			break;
 		}
+	}
 
-		/* increment and save iucv message tag for msg_completion cbk */
-		txmsg.tag = iucv->send_tag++;
-		memcpy(CB_TAG(skb), &txmsg.tag, CB_TAG_LEN);
-		skb_queue_tail(&iucv->send_skb_q, skb);
+	/* allocate one skb for each iucv message:
+	 * this is fine for SOCK_SEQPACKET (unless we want to support
+	 * segmented records using the MSG_EOR flag), but
+	 * for SOCK_STREAM we might want to improve it in future */
+	skb = sock_alloc_send_skb(sk, len, msg->msg_flags & MSG_DONTWAIT,
+				  &err);
+	if (!skb)
+		goto out;
+	if (memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len)) {
+		err = -EFAULT;
+		goto fail;
+	}
 
-		if (((iucv->path->flags & IUCV_IPRMDATA) & iucv->flags)
-		    && skb->len <= 7) {
-			err = iucv_send_iprm(iucv->path, &txmsg, skb);
+	/* increment and save iucv message tag for msg_completion cbk */
+	txmsg.tag = iucv->send_tag++;
+	memcpy(CB_TAG(skb), &txmsg.tag, CB_TAG_LEN);
+	skb_queue_tail(&iucv->send_skb_q, skb);
 
-			/* on success: there is no message_complete callback
-			 * for an IPRMDATA msg; remove skb from send queue */
-			if (err == 0) {
-				skb_unlink(skb, &iucv->send_skb_q);
-				kfree_skb(skb);
-			}
+	if (((iucv->path->flags & IUCV_IPRMDATA) & iucv->flags)
+	      && skb->len <= 7) {
+		err = iucv_send_iprm(iucv->path, &txmsg, skb);
 
-			/* this error should never happen since the
-			 * IUCV_IPRMDATA path flag is set... sever path */
-			if (err == 0x15) {
-				iucv_path_sever(iucv->path, NULL);
-				skb_unlink(skb, &iucv->send_skb_q);
-				err = -EPIPE;
-				goto fail;
-			}
-		} else
-			err = iucv_message_send(iucv->path, &txmsg, 0, 0,
-						(void *) skb->data, skb->len);
-		if (err) {
-			if (err == 3) {
-				user_id[8] = 0;
-				memcpy(user_id, iucv->dst_user_id, 8);
-				appl_id[8] = 0;
-				memcpy(appl_id, iucv->dst_name, 8);
-				pr_err("Application %s on z/VM guest %s"
-				       " exceeds message limit\n",
-				       user_id, appl_id);
-			}
+		/* on success: there is no message_complete callback
+		 * for an IPRMDATA msg; remove skb from send queue */
+		if (err == 0) {
+			skb_unlink(skb, &iucv->send_skb_q);
+			kfree_skb(skb);
+		}
+
+		/* this error should never happen since the
+		 * IUCV_IPRMDATA path flag is set... sever path */
+		if (err == 0x15) {
+			iucv_path_sever(iucv->path, NULL);
 			skb_unlink(skb, &iucv->send_skb_q);
 			err = -EPIPE;
 			goto fail;
 		}
-
-	} else {
-		err = -ENOTCONN;
-		goto out;
+	} else
+		err = iucv_message_send(iucv->path, &txmsg, 0, 0,
+					(void *) skb->data, skb->len);
+	if (err) {
+		if (err == 3) {
+			user_id[8] = 0;
+			memcpy(user_id, iucv->dst_user_id, 8);
+			appl_id[8] = 0;
+			memcpy(appl_id, iucv->dst_name, 8);
+			pr_err("Application %s on z/VM guest %s"
+				" exceeds message limit\n",
+				appl_id, user_id);
+		}
+		skb_unlink(skb, &iucv->send_skb_q);
+		err = -EPIPE;
+		goto fail;
 	}
 
 	release_sock(sk);
-- 
cgit v1.2.3-59-g8ed1b


From 0ea920d211e0a870871965418923b08da2025b4a Mon Sep 17 00:00:00 2001
From: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Date: Wed, 17 Jun 2009 21:54:48 +0000
Subject: af_iucv: Return -EAGAIN if iucv msg limit is exceeded

If the iucv message limit for a communication path is exceeded,
sendmsg() returns -EAGAIN instead of -EPIPE.
The calling application can then handle this error situtation,
e.g. to try again after waiting some time.

For blocking sockets, sendmsg() waits up to the socket timeout
before returning -EAGAIN. For the new wait condition, a macro
has been introduced and the iucv_sock_wait_state() has been
refactored to this macro.

Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Ursula Braun <ursula.braun@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/iucv/af_iucv.h |   2 -
 net/iucv/af_iucv.c         | 144 ++++++++++++++++++++++++++++++++-------------
 2 files changed, 103 insertions(+), 43 deletions(-)

diff --git a/include/net/iucv/af_iucv.h b/include/net/iucv/af_iucv.h
index 21ee49ffcbaf..f82a1e877372 100644
--- a/include/net/iucv/af_iucv.h
+++ b/include/net/iucv/af_iucv.h
@@ -94,8 +94,6 @@ unsigned int iucv_sock_poll(struct file *file, struct socket *sock,
 			    poll_table *wait);
 void iucv_sock_link(struct iucv_sock_list *l, struct sock *s);
 void iucv_sock_unlink(struct iucv_sock_list *l, struct sock *s);
-int  iucv_sock_wait_state(struct sock *sk, int state, int state2,
-			  unsigned long timeo);
 int  iucv_sock_wait_cnt(struct sock *sk, unsigned long timeo);
 void iucv_accept_enqueue(struct sock *parent, struct sock *sk);
 void iucv_accept_unlink(struct sock *sk);
diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c
index 42b7198a6883..abadb4a846cf 100644
--- a/net/iucv/af_iucv.c
+++ b/net/iucv/af_iucv.c
@@ -53,6 +53,38 @@ static const u8 iprm_shutdown[8] =
 #define CB_TRGCLS(skb)	((skb)->cb + CB_TAG_LEN) /* iucv msg target class */
 #define CB_TRGCLS_LEN	(TRGCLS_SIZE)
 
+#define __iucv_sock_wait(sk, condition, timeo, ret)			\
+do {									\
+	DEFINE_WAIT(__wait);						\
+	long __timeo = timeo;						\
+	ret = 0;							\
+	while (!(condition)) {						\
+		prepare_to_wait(sk->sk_sleep, &__wait, TASK_INTERRUPTIBLE); \
+		if (!__timeo) {						\
+			ret = -EAGAIN;					\
+			break;						\
+		}							\
+		if (signal_pending(current)) {				\
+			ret = sock_intr_errno(__timeo);			\
+			break;						\
+		}							\
+		release_sock(sk);					\
+		__timeo = schedule_timeout(__timeo);			\
+		lock_sock(sk);						\
+		ret = sock_error(sk);					\
+		if (ret)						\
+			break;						\
+	}								\
+	finish_wait(sk->sk_sleep, &__wait);				\
+} while (0)
+
+#define iucv_sock_wait(sk, condition, timeo)				\
+({									\
+	int __ret = 0;							\
+	if (!(condition))						\
+		__iucv_sock_wait(sk, condition, timeo, __ret);		\
+	__ret;								\
+})
 
 static void iucv_sock_kill(struct sock *sk);
 static void iucv_sock_close(struct sock *sk);
@@ -121,6 +153,48 @@ static inline size_t iucv_msg_length(struct iucv_message *msg)
 	return msg->length;
 }
 
+/**
+ * iucv_sock_in_state() - check for specific states
+ * @sk:		sock structure
+ * @state:	first iucv sk state
+ * @state:	second iucv sk state
+ *
+ * Returns true if the socket in either in the first or second state.
+ */
+static int iucv_sock_in_state(struct sock *sk, int state, int state2)
+{
+	return (sk->sk_state == state || sk->sk_state == state2);
+}
+
+/**
+ * iucv_below_msglim() - function to check if messages can be sent
+ * @sk:		sock structure
+ *
+ * Returns true if the send queue length is lower than the message limit.
+ * Always returns true if the socket is not connected (no iucv path for
+ * checking the message limit).
+ */
+static inline int iucv_below_msglim(struct sock *sk)
+{
+	struct iucv_sock *iucv = iucv_sk(sk);
+
+	if (sk->sk_state != IUCV_CONNECTED)
+		return 1;
+	return (skb_queue_len(&iucv->send_skb_q) < iucv->path->msglim);
+}
+
+/**
+ * iucv_sock_wake_msglim() - Wake up thread waiting on msg limit
+ */
+static void iucv_sock_wake_msglim(struct sock *sk)
+{
+	read_lock(&sk->sk_callback_lock);
+	if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))
+		wake_up_interruptible_all(sk->sk_sleep);
+	sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT);
+	read_unlock(&sk->sk_callback_lock);
+}
+
 /* Timers */
 static void iucv_sock_timeout(unsigned long arg)
 {
@@ -212,7 +286,9 @@ static void iucv_sock_close(struct sock *sk)
 				timeo = sk->sk_lingertime;
 			else
 				timeo = IUCV_DISCONN_TIMEOUT;
-			err = iucv_sock_wait_state(sk, IUCV_CLOSED, 0, timeo);
+			err = iucv_sock_wait(sk,
+					iucv_sock_in_state(sk, IUCV_CLOSED, 0),
+					timeo);
 		}
 
 	case IUCV_CLOSING:   /* fall through */
@@ -393,39 +469,6 @@ struct sock *iucv_accept_dequeue(struct sock *parent, struct socket *newsock)
 	return NULL;
 }
 
-int iucv_sock_wait_state(struct sock *sk, int state, int state2,
-			 unsigned long timeo)
-{
-	DECLARE_WAITQUEUE(wait, current);
-	int err = 0;
-
-	add_wait_queue(sk->sk_sleep, &wait);
-	while (sk->sk_state != state && sk->sk_state != state2) {
-		set_current_state(TASK_INTERRUPTIBLE);
-
-		if (!timeo) {
-			err = -EAGAIN;
-			break;
-		}
-
-		if (signal_pending(current)) {
-			err = sock_intr_errno(timeo);
-			break;
-		}
-
-		release_sock(sk);
-		timeo = schedule_timeout(timeo);
-		lock_sock(sk);
-
-		err = sock_error(sk);
-		if (err)
-			break;
-	}
-	set_current_state(TASK_RUNNING);
-	remove_wait_queue(sk->sk_sleep, &wait);
-	return err;
-}
-
 /* Bind an unbound socket */
 static int iucv_sock_bind(struct socket *sock, struct sockaddr *addr,
 			  int addr_len)
@@ -570,8 +613,9 @@ static int iucv_sock_connect(struct socket *sock, struct sockaddr *addr,
 	}
 
 	if (sk->sk_state != IUCV_CONNECTED) {
-		err = iucv_sock_wait_state(sk, IUCV_CONNECTED, IUCV_DISCONN,
-				sock_sndtimeo(sk, flags & O_NONBLOCK));
+		err = iucv_sock_wait(sk, iucv_sock_in_state(sk, IUCV_CONNECTED,
+							    IUCV_DISCONN),
+				     sock_sndtimeo(sk, flags & O_NONBLOCK));
 	}
 
 	if (sk->sk_state == IUCV_DISCONN) {
@@ -725,9 +769,11 @@ static int iucv_sock_sendmsg(struct kiocb *iocb, struct socket *sock,
 	struct iucv_message txmsg;
 	struct cmsghdr *cmsg;
 	int cmsg_done;
+	long timeo;
 	char user_id[9];
 	char appl_id[9];
 	int err;
+	int noblock = msg->msg_flags & MSG_DONTWAIT;
 
 	err = sock_error(sk);
 	if (err)
@@ -799,8 +845,7 @@ static int iucv_sock_sendmsg(struct kiocb *iocb, struct socket *sock,
 	 * this is fine for SOCK_SEQPACKET (unless we want to support
 	 * segmented records using the MSG_EOR flag), but
 	 * for SOCK_STREAM we might want to improve it in future */
-	skb = sock_alloc_send_skb(sk, len, msg->msg_flags & MSG_DONTWAIT,
-				  &err);
+	skb = sock_alloc_send_skb(sk, len, noblock, &err);
 	if (!skb)
 		goto out;
 	if (memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len)) {
@@ -808,6 +853,18 @@ static int iucv_sock_sendmsg(struct kiocb *iocb, struct socket *sock,
 		goto fail;
 	}
 
+	/* wait if outstanding messages for iucv path has reached */
+	timeo = sock_sndtimeo(sk, noblock);
+	err = iucv_sock_wait(sk, iucv_below_msglim(sk), timeo);
+	if (err)
+		goto fail;
+
+	/* return -ECONNRESET if the socket is no longer connected */
+	if (sk->sk_state != IUCV_CONNECTED) {
+		err = -ECONNRESET;
+		goto fail;
+	}
+
 	/* increment and save iucv message tag for msg_completion cbk */
 	txmsg.tag = iucv->send_tag++;
 	memcpy(CB_TAG(skb), &txmsg.tag, CB_TAG_LEN);
@@ -844,9 +901,10 @@ static int iucv_sock_sendmsg(struct kiocb *iocb, struct socket *sock,
 			pr_err("Application %s on z/VM guest %s"
 				" exceeds message limit\n",
 				appl_id, user_id);
-		}
+			err = -EAGAIN;
+		} else
+			err = -EPIPE;
 		skb_unlink(skb, &iucv->send_skb_q);
-		err = -EPIPE;
 		goto fail;
 	}
 
@@ -1463,7 +1521,11 @@ static void iucv_callback_txdone(struct iucv_path *path,
 
 		spin_unlock_irqrestore(&list->lock, flags);
 
-		kfree_skb(this);
+		if (this) {
+			kfree_skb(this);
+			/* wake up any process waiting for sending */
+			iucv_sock_wake_msglim(sk);
+		}
 	}
 	BUG_ON(!this);
 
-- 
cgit v1.2.3-59-g8ed1b


From e3453f6342110d60edb37be92c4a4f668ca8b0c4 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Thu, 18 Jun 2009 07:03:47 +0000
Subject: pegasus usb-net: Fix endianness bugs

This fixes various endianness bugs. Some harmless and some real ones.
This is tested on a PowerPC-64 machine.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Cc: Stable <stable@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/usb/pegasus.c | 29 +++++++++++++++++------------
 1 file changed, 17 insertions(+), 12 deletions(-)

diff --git a/drivers/net/usb/pegasus.c b/drivers/net/usb/pegasus.c
index 2138535f2339..73acbd244aa1 100644
--- a/drivers/net/usb/pegasus.c
+++ b/drivers/net/usb/pegasus.c
@@ -297,7 +297,7 @@ static int update_eth_regs_async(pegasus_t * pegasus)
 
 	pegasus->dr.bRequestType = PEGASUS_REQT_WRITE;
 	pegasus->dr.bRequest = PEGASUS_REQ_SET_REGS;
-	pegasus->dr.wValue = 0;
+	pegasus->dr.wValue = cpu_to_le16(0);
 	pegasus->dr.wIndex = cpu_to_le16(EthCtrl0);
 	pegasus->dr.wLength = cpu_to_le16(3);
 	pegasus->ctrl_urb->transfer_buffer_length = 3;
@@ -446,11 +446,12 @@ static int write_eprom_word(pegasus_t * pegasus, __u8 index, __u16 data)
 	int i;
 	__u8 tmp, d[4] = { 0x3f, 0, 0, EPROM_WRITE };
 	int ret;
+	__le16 le_data = cpu_to_le16(data);
 
 	set_registers(pegasus, EpromOffset, 4, d);
 	enable_eprom_write(pegasus);
 	set_register(pegasus, EpromOffset, index);
-	set_registers(pegasus, EpromData, 2, &data);
+	set_registers(pegasus, EpromData, 2, &le_data);
 	set_register(pegasus, EpromCtrl, EPROM_WRITE);
 
 	for (i = 0; i < REG_TIMEOUT; i++) {
@@ -923,29 +924,32 @@ static struct net_device_stats *pegasus_netdev_stats(struct net_device *dev)
 
 static inline void disable_net_traffic(pegasus_t * pegasus)
 {
-	int tmp = 0;
+	__le16 tmp = cpu_to_le16(0);
 
-	set_registers(pegasus, EthCtrl0, 2, &tmp);
+	set_registers(pegasus, EthCtrl0, sizeof(tmp), &tmp);
 }
 
 static inline void get_interrupt_interval(pegasus_t * pegasus)
 {
-	__u8 data[2];
+	u16 data;
+	u8 interval;
 
-	read_eprom_word(pegasus, 4, (__u16 *) data);
+	read_eprom_word(pegasus, 4, &data);
+	interval = data >> 8;
 	if (pegasus->usb->speed != USB_SPEED_HIGH) {
-		if (data[1] < 0x80) {
+		if (interval < 0x80) {
 			if (netif_msg_timer(pegasus))
 				dev_info(&pegasus->intf->dev, "intr interval "
 					"changed from %ums to %ums\n",
-					data[1], 0x80);
-			data[1] = 0x80;
+					interval, 0x80);
+			interval = 0x80;
+			data = (data & 0x00FF) | ((u16)interval << 8);
 #ifdef PEGASUS_WRITE_EEPROM
-			write_eprom_word(pegasus, 4, *(__u16 *) data);
+			write_eprom_word(pegasus, 4, data);
 #endif
 		}
 	}
-	pegasus->intr_interval = data[1];
+	pegasus->intr_interval = interval;
 }
 
 static void set_carrier(struct net_device *net)
@@ -1299,7 +1303,8 @@ static int pegasus_blacklisted(struct usb_device *udev)
 	/* Special quirk to keep the driver from handling the Belkin Bluetooth
 	 * dongle which happens to have the same ID.
 	 */
-	if ((udd->idVendor == VENDOR_BELKIN && udd->idProduct == 0x0121) &&
+	if ((udd->idVendor == cpu_to_le16(VENDOR_BELKIN)) &&
+	    (udd->idProduct == cpu_to_le16(0x0121)) &&
 	    (udd->bDeviceClass == USB_CLASS_WIRELESS_CONTROLLER) &&
 	    (udd->bDeviceProtocol == 1))
 		return 1;
-- 
cgit v1.2.3-59-g8ed1b


From 5fb379ee67a7ec55ff65b467b472f3d69b60ba16 Mon Sep 17 00:00:00 2001
From: Sathya Perla <sathyap@serverengines.com>
Date: Thu, 18 Jun 2009 00:02:59 +0000
Subject: be2net: Add MCC queue mechanism for BE cmds

Currenlty all cmds use the blocking MCC mbox to post cmds. An mbox cmd is protected
via a spin_lock(cmd_lock) and not spin_lock_bh() as it is undesirable
to disable BHs while a blocking mbox cmd is in progress (and take long to finish.)
This can lockup a cmd in progress in process context. Instead cmds that may be
called in BH context must use the MCC queue to post cmds. The cmd completions
are rcvd in a separate completion queue and the events are placed in the tx-event
queue.

Signed-off-by: Sathya Perla <sathyap@serverengines.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/benet/be.h      |  95 ++++++++++------
 drivers/net/benet/be_cmds.c | 258 ++++++++++++++++++++++++++++++++++----------
 drivers/net/benet/be_cmds.h |  40 ++++++-
 drivers/net/benet/be_hw.h   |   8 +-
 drivers/net/benet/be_main.c | 215 ++++++++++++++++++++++++------------
 5 files changed, 455 insertions(+), 161 deletions(-)

diff --git a/drivers/net/benet/be.h b/drivers/net/benet/be.h
index b4bb06fdf307..ef5be133ce68 100644
--- a/drivers/net/benet/be.h
+++ b/drivers/net/benet/be.h
@@ -65,7 +65,7 @@ static inline char *nic_name(struct pci_dev *pdev)
 #define TX_CQ_LEN		1024
 #define RX_Q_LEN		1024	/* Does not support any other value */
 #define RX_CQ_LEN		1024
-#define MCC_Q_LEN		64	/* total size not to exceed 8 pages */
+#define MCC_Q_LEN		128	/* total size not to exceed 8 pages */
 #define MCC_CQ_LEN		256
 
 #define BE_NAPI_WEIGHT		64
@@ -91,6 +91,61 @@ struct be_queue_info {
 	atomic_t used;	/* Number of valid elements in the queue */
 };
 
+static inline u32 MODULO(u16 val, u16 limit)
+{
+	BUG_ON(limit & (limit - 1));
+	return val & (limit - 1);
+}
+
+static inline void index_adv(u16 *index, u16 val, u16 limit)
+{
+	*index = MODULO((*index + val), limit);
+}
+
+static inline void index_inc(u16 *index, u16 limit)
+{
+	*index = MODULO((*index + 1), limit);
+}
+
+static inline void *queue_head_node(struct be_queue_info *q)
+{
+	return q->dma_mem.va + q->head * q->entry_size;
+}
+
+static inline void *queue_tail_node(struct be_queue_info *q)
+{
+	return q->dma_mem.va + q->tail * q->entry_size;
+}
+
+static inline void queue_head_inc(struct be_queue_info *q)
+{
+	index_inc(&q->head, q->len);
+}
+
+static inline void queue_tail_inc(struct be_queue_info *q)
+{
+	index_inc(&q->tail, q->len);
+}
+
+
+struct be_eq_obj {
+	struct be_queue_info q;
+	char desc[32];
+
+	/* Adaptive interrupt coalescing (AIC) info */
+	bool enable_aic;
+	u16 min_eqd;		/* in usecs */
+	u16 max_eqd;		/* in usecs */
+	u16 cur_eqd;		/* in usecs */
+
+	struct napi_struct napi;
+};
+
+struct be_mcc_obj {
+	struct be_queue_info q;
+	struct be_queue_info cq;
+};
+
 struct be_ctrl_info {
 	u8 __iomem *csr;
 	u8 __iomem *db;		/* Door Bell */
@@ -98,11 +153,16 @@ struct be_ctrl_info {
 	int pci_func;
 
 	/* Mbox used for cmd request/response */
-	spinlock_t cmd_lock;	/* For serializing cmds to BE card */
+	spinlock_t mbox_lock;	/* For serializing mbox cmds to BE card */
 	struct be_dma_mem mbox_mem;
 	/* Mbox mem is adjusted to align to 16 bytes. The allocated addr
 	 * is stored for freeing purpose */
 	struct be_dma_mem mbox_mem_alloced;
+
+	/* MCC Rings */
+	struct be_mcc_obj mcc_obj;
+	spinlock_t mcc_lock;	/* For serializing mcc cmds to BE card */
+	spinlock_t mcc_cq_lock;
 };
 
 #include "be_cmds.h"
@@ -150,19 +210,6 @@ struct be_stats_obj {
 	struct be_dma_mem cmd;
 };
 
-struct be_eq_obj {
-	struct be_queue_info q;
-	char desc[32];
-
-	/* Adaptive interrupt coalescing (AIC) info */
-	bool enable_aic;
-	u16 min_eqd;		/* in usecs */
-	u16 max_eqd;		/* in usecs */
-	u16 cur_eqd;		/* in usecs */
-
-	struct napi_struct napi;
-};
-
 struct be_tx_obj {
 	struct be_queue_info q;
 	struct be_queue_info cq;
@@ -235,22 +282,6 @@ extern struct ethtool_ops be_ethtool_ops;
 
 #define BE_SET_NETDEV_OPS(netdev, ops)	(netdev->netdev_ops = ops)
 
-static inline u32 MODULO(u16 val, u16 limit)
-{
-	BUG_ON(limit & (limit - 1));
-	return val & (limit - 1);
-}
-
-static inline void index_adv(u16 *index, u16 val, u16 limit)
-{
-	*index = MODULO((*index + val), limit);
-}
-
-static inline void index_inc(u16 *index, u16 limit)
-{
-	*index = MODULO((*index + 1), limit);
-}
-
 #define PAGE_SHIFT_4K		12
 #define PAGE_SIZE_4K		(1 << PAGE_SHIFT_4K)
 
@@ -339,4 +370,6 @@ static inline u8 is_udp_pkt(struct sk_buff *skb)
 	return val;
 }
 
+extern void be_cq_notify(struct be_ctrl_info *ctrl, u16 qid, bool arm,
+		u16 num_popped);
 #endif				/* BE_H */
diff --git a/drivers/net/benet/be_cmds.c b/drivers/net/benet/be_cmds.c
index d444aed962bc..f1ec191f0c0d 100644
--- a/drivers/net/benet/be_cmds.c
+++ b/drivers/net/benet/be_cmds.c
@@ -17,6 +17,90 @@
 
 #include "be.h"
 
+void be_mcc_notify(struct be_ctrl_info *ctrl)
+{
+	struct be_queue_info *mccq = &ctrl->mcc_obj.q;
+	u32 val = 0;
+
+	val |= mccq->id & DB_MCCQ_RING_ID_MASK;
+	val |= 1 << DB_MCCQ_NUM_POSTED_SHIFT;
+	iowrite32(val, ctrl->db + DB_MCCQ_OFFSET);
+}
+
+/* To check if valid bit is set, check the entire word as we don't know
+ * the endianness of the data (old entry is host endian while a new entry is
+ * little endian) */
+static inline bool be_mcc_compl_is_new(struct be_mcc_cq_entry *compl)
+{
+	if (compl->flags != 0) {
+		compl->flags = le32_to_cpu(compl->flags);
+		BUG_ON((compl->flags & CQE_FLAGS_VALID_MASK) == 0);
+		return true;
+	} else {
+		return false;
+	}
+}
+
+/* Need to reset the entire word that houses the valid bit */
+static inline void be_mcc_compl_use(struct be_mcc_cq_entry *compl)
+{
+	compl->flags = 0;
+}
+
+static int be_mcc_compl_process(struct be_ctrl_info *ctrl,
+	struct be_mcc_cq_entry *compl)
+{
+	u16 compl_status, extd_status;
+
+	/* Just swap the status to host endian; mcc tag is opaquely copied
+	 * from mcc_wrb */
+	be_dws_le_to_cpu(compl, 4);
+
+	compl_status = (compl->status >> CQE_STATUS_COMPL_SHIFT) &
+				CQE_STATUS_COMPL_MASK;
+	if (compl_status != MCC_STATUS_SUCCESS) {
+		extd_status = (compl->status >> CQE_STATUS_EXTD_SHIFT) &
+				CQE_STATUS_EXTD_MASK;
+		printk(KERN_WARNING DRV_NAME
+			" error in cmd completion: status(compl/extd)=%d/%d\n",
+			compl_status, extd_status);
+		return -1;
+	}
+	return 0;
+}
+
+
+static struct be_mcc_cq_entry *be_mcc_compl_get(struct be_ctrl_info *ctrl)
+{
+	struct be_queue_info *mcc_cq = &ctrl->mcc_obj.cq;
+	struct be_mcc_cq_entry *compl = queue_tail_node(mcc_cq);
+
+	if (be_mcc_compl_is_new(compl)) {
+		queue_tail_inc(mcc_cq);
+		return compl;
+	}
+	return NULL;
+}
+
+void be_process_mcc(struct be_ctrl_info *ctrl)
+{
+	struct be_mcc_cq_entry *compl;
+	int num = 0;
+
+	spin_lock_bh(&ctrl->mcc_cq_lock);
+	while ((compl = be_mcc_compl_get(ctrl))) {
+		if (!(compl->flags & CQE_FLAGS_ASYNC_MASK)) {
+			be_mcc_compl_process(ctrl, compl);
+			atomic_dec(&ctrl->mcc_obj.q.used);
+		}
+		be_mcc_compl_use(compl);
+		num++;
+	}
+	if (num)
+		be_cq_notify(ctrl, ctrl->mcc_obj.cq.id, true, num);
+	spin_unlock_bh(&ctrl->mcc_cq_lock);
+}
+
 static int be_mbox_db_ready_wait(void __iomem *db)
 {
 	int cnt = 0, wait = 5;
@@ -44,11 +128,11 @@ static int be_mbox_db_ready_wait(void __iomem *db)
 
 /*
  * Insert the mailbox address into the doorbell in two steps
+ * Polls on the mbox doorbell till a command completion (or a timeout) occurs
  */
 static int be_mbox_db_ring(struct be_ctrl_info *ctrl)
 {
 	int status;
-	u16 compl_status, extd_status;
 	u32 val = 0;
 	void __iomem *db = ctrl->db + MPU_MAILBOX_DB_OFFSET;
 	struct be_dma_mem *mbox_mem = &ctrl->mbox_mem;
@@ -79,24 +163,17 @@ static int be_mbox_db_ring(struct be_ctrl_info *ctrl)
 	if (status != 0)
 		return status;
 
-	/* compl entry has been made now */
-	be_dws_le_to_cpu(cqe, sizeof(*cqe));
-	if (!(cqe->flags & CQE_FLAGS_VALID_MASK)) {
-		printk(KERN_WARNING DRV_NAME ": ERROR invalid mbox compl\n");
+	/* A cq entry has been made now */
+	if (be_mcc_compl_is_new(cqe)) {
+		status = be_mcc_compl_process(ctrl, &mbox->cqe);
+		be_mcc_compl_use(cqe);
+		if (status)
+			return status;
+	} else {
+		printk(KERN_WARNING DRV_NAME "invalid mailbox completion\n");
 		return -1;
 	}
-
-	compl_status = (cqe->status >> CQE_STATUS_COMPL_SHIFT) &
-				CQE_STATUS_COMPL_MASK;
-	if (compl_status != MCC_STATUS_SUCCESS) {
-		extd_status = (cqe->status >> CQE_STATUS_EXTD_SHIFT) &
-				CQE_STATUS_EXTD_MASK;
-		printk(KERN_WARNING DRV_NAME
-			": ERROR in cmd compl. status(compl/extd)=%d/%d\n",
-			compl_status, extd_status);
-	}
-
-	return compl_status;
+	return 0;
 }
 
 static int be_POST_stage_get(struct be_ctrl_info *ctrl, u16 *stage)
@@ -235,6 +312,18 @@ static inline struct be_mcc_wrb *wrb_from_mbox(struct be_dma_mem *mbox_mem)
 	return &((struct be_mcc_mailbox *)(mbox_mem->va))->wrb;
 }
 
+static inline struct be_mcc_wrb *wrb_from_mcc(struct be_queue_info *mccq)
+{
+	struct be_mcc_wrb *wrb = NULL;
+	if (atomic_read(&mccq->used) < mccq->len) {
+		wrb = queue_head_node(mccq);
+		queue_head_inc(mccq);
+		atomic_inc(&mccq->used);
+		memset(wrb, 0, sizeof(*wrb));
+	}
+	return wrb;
+}
+
 int be_cmd_eq_create(struct be_ctrl_info *ctrl,
 		struct be_queue_info *eq, int eq_delay)
 {
@@ -244,7 +333,7 @@ int be_cmd_eq_create(struct be_ctrl_info *ctrl,
 	struct be_dma_mem *q_mem = &eq->dma_mem;
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -272,7 +361,7 @@ int be_cmd_eq_create(struct be_ctrl_info *ctrl,
 		eq->id = le16_to_cpu(resp->eq_id);
 		eq->created = true;
 	}
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -284,7 +373,7 @@ int be_cmd_mac_addr_query(struct be_ctrl_info *ctrl, u8 *mac_addr,
 	struct be_cmd_resp_mac_query *resp = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -304,7 +393,7 @@ int be_cmd_mac_addr_query(struct be_ctrl_info *ctrl, u8 *mac_addr,
 	if (!status)
 		memcpy(mac_addr, resp->mac.addr, ETH_ALEN);
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -315,7 +404,7 @@ int be_cmd_pmac_add(struct be_ctrl_info *ctrl, u8 *mac_addr,
 	struct be_cmd_req_pmac_add *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -332,7 +421,7 @@ int be_cmd_pmac_add(struct be_ctrl_info *ctrl, u8 *mac_addr,
 		*pmac_id = le32_to_cpu(resp->pmac_id);
 	}
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -342,7 +431,7 @@ int be_cmd_pmac_del(struct be_ctrl_info *ctrl, u32 if_id, u32 pmac_id)
 	struct be_cmd_req_pmac_del *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -354,7 +443,7 @@ int be_cmd_pmac_del(struct be_ctrl_info *ctrl, u32 if_id, u32 pmac_id)
 	req->pmac_id = cpu_to_le32(pmac_id);
 
 	status = be_mbox_db_ring(ctrl);
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 
 	return status;
 }
@@ -370,7 +459,7 @@ int be_cmd_cq_create(struct be_ctrl_info *ctrl,
 	void *ctxt = &req->context;
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -388,7 +477,7 @@ int be_cmd_cq_create(struct be_ctrl_info *ctrl,
 	AMAP_SET_BITS(struct amap_cq_context, solevent, ctxt, sol_evts);
 	AMAP_SET_BITS(struct amap_cq_context, eventable, ctxt, 1);
 	AMAP_SET_BITS(struct amap_cq_context, eqid, ctxt, eq->id);
-	AMAP_SET_BITS(struct amap_cq_context, armed, ctxt, 0);
+	AMAP_SET_BITS(struct amap_cq_context, armed, ctxt, 1);
 	AMAP_SET_BITS(struct amap_cq_context, func, ctxt, ctrl->pci_func);
 	be_dws_cpu_to_le(ctxt, sizeof(req->context));
 
@@ -399,7 +488,56 @@ int be_cmd_cq_create(struct be_ctrl_info *ctrl,
 		cq->id = le16_to_cpu(resp->cq_id);
 		cq->created = true;
 	}
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
+
+	return status;
+}
+
+static u32 be_encoded_q_len(int q_len)
+{
+	u32 len_encoded = fls(q_len); /* log2(len) + 1 */
+	if (len_encoded == 16)
+		len_encoded = 0;
+	return len_encoded;
+}
+
+int be_cmd_mccq_create(struct be_ctrl_info *ctrl,
+			struct be_queue_info *mccq,
+			struct be_queue_info *cq)
+{
+	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
+	struct be_cmd_req_mcc_create *req = embedded_payload(wrb);
+	struct be_dma_mem *q_mem = &mccq->dma_mem;
+	void *ctxt = &req->context;
+	int status;
+
+	spin_lock(&ctrl->mbox_lock);
+	memset(wrb, 0, sizeof(*wrb));
+
+	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
+
+	be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+			OPCODE_COMMON_MCC_CREATE, sizeof(*req));
+
+	req->num_pages = PAGES_4K_SPANNED(q_mem->va, q_mem->size);
+
+	AMAP_SET_BITS(struct amap_mcc_context, fid, ctxt, ctrl->pci_func);
+	AMAP_SET_BITS(struct amap_mcc_context, valid, ctxt, 1);
+	AMAP_SET_BITS(struct amap_mcc_context, ring_size, ctxt,
+		be_encoded_q_len(mccq->len));
+	AMAP_SET_BITS(struct amap_mcc_context, cq_id, ctxt, cq->id);
+
+	be_dws_cpu_to_le(ctxt, sizeof(req->context));
+
+	be_cmd_page_addrs_prepare(req->pages, ARRAY_SIZE(req->pages), q_mem);
+
+	status = be_mbox_db_ring(ctrl);
+	if (!status) {
+		struct be_cmd_resp_mcc_create *resp = embedded_payload(wrb);
+		mccq->id = le16_to_cpu(resp->id);
+		mccq->created = true;
+	}
+	spin_unlock(&ctrl->mbox_lock);
 
 	return status;
 }
@@ -415,7 +553,7 @@ int be_cmd_txq_create(struct be_ctrl_info *ctrl,
 	int status;
 	u32 len_encoded;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -446,7 +584,7 @@ int be_cmd_txq_create(struct be_ctrl_info *ctrl,
 		txq->id = le16_to_cpu(resp->cid);
 		txq->created = true;
 	}
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 
 	return status;
 }
@@ -460,7 +598,7 @@ int be_cmd_rxq_create(struct be_ctrl_info *ctrl,
 	struct be_dma_mem *q_mem = &rxq->dma_mem;
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -482,7 +620,7 @@ int be_cmd_rxq_create(struct be_ctrl_info *ctrl,
 		rxq->id = le16_to_cpu(resp->id);
 		rxq->created = true;
 	}
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 
 	return status;
 }
@@ -496,7 +634,7 @@ int be_cmd_q_destroy(struct be_ctrl_info *ctrl, struct be_queue_info *q,
 	u8 subsys = 0, opcode = 0;
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 
 	memset(wrb, 0, sizeof(*wrb));
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -518,6 +656,10 @@ int be_cmd_q_destroy(struct be_ctrl_info *ctrl, struct be_queue_info *q,
 		subsys = CMD_SUBSYSTEM_ETH;
 		opcode = OPCODE_ETH_RX_DESTROY;
 		break;
+	case QTYPE_MCCQ:
+		subsys = CMD_SUBSYSTEM_COMMON;
+		opcode = OPCODE_COMMON_MCC_DESTROY;
+		break;
 	default:
 		printk(KERN_WARNING DRV_NAME ":bad Q type in Q destroy cmd\n");
 		status = -1;
@@ -528,7 +670,7 @@ int be_cmd_q_destroy(struct be_ctrl_info *ctrl, struct be_queue_info *q,
 
 	status = be_mbox_db_ring(ctrl);
 err:
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 
 	return status;
 }
@@ -541,7 +683,7 @@ int be_cmd_if_create(struct be_ctrl_info *ctrl, u32 flags, u8 *mac,
 	struct be_cmd_req_if_create *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -562,7 +704,7 @@ int be_cmd_if_create(struct be_ctrl_info *ctrl, u32 flags, u8 *mac,
 			*pmac_id = le32_to_cpu(resp->pmac_id);
 	}
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -572,7 +714,7 @@ int be_cmd_if_destroy(struct be_ctrl_info *ctrl, u32 interface_id)
 	struct be_cmd_req_if_destroy *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -583,7 +725,7 @@ int be_cmd_if_destroy(struct be_ctrl_info *ctrl, u32 interface_id)
 	req->interface_id = cpu_to_le32(interface_id);
 	status = be_mbox_db_ring(ctrl);
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 
 	return status;
 }
@@ -598,7 +740,7 @@ int be_cmd_get_stats(struct be_ctrl_info *ctrl, struct be_dma_mem *nonemb_cmd)
 	struct be_sge *sge = nonembedded_sgl(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	memset(req, 0, sizeof(*req));
@@ -617,7 +759,7 @@ int be_cmd_get_stats(struct be_ctrl_info *ctrl, struct be_dma_mem *nonemb_cmd)
 		be_dws_le_to_cpu(&resp->hw_stats, sizeof(resp->hw_stats));
 	}
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -628,7 +770,7 @@ int be_cmd_link_status_query(struct be_ctrl_info *ctrl,
 	struct be_cmd_req_link_status *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -646,7 +788,7 @@ int be_cmd_link_status_query(struct be_ctrl_info *ctrl,
 		link->speed = PHY_LINK_SPEED_ZERO;
 	}
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -656,7 +798,7 @@ int be_cmd_get_fw_ver(struct be_ctrl_info *ctrl, char *fw_ver)
 	struct be_cmd_req_get_fw_version *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -670,7 +812,7 @@ int be_cmd_get_fw_ver(struct be_ctrl_info *ctrl, char *fw_ver)
 		strncpy(fw_ver, resp->firmware_version_string, FW_VER_LEN);
 	}
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -681,7 +823,7 @@ int be_cmd_modify_eqd(struct be_ctrl_info *ctrl, u32 eq_id, u32 eqd)
 	struct be_cmd_req_modify_eq_delay *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -696,7 +838,7 @@ int be_cmd_modify_eqd(struct be_ctrl_info *ctrl, u32 eq_id, u32 eqd)
 
 	status = be_mbox_db_ring(ctrl);
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -707,7 +849,7 @@ int be_cmd_vlan_config(struct be_ctrl_info *ctrl, u32 if_id, u16 *vtag_array,
 	struct be_cmd_req_vlan_config *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -726,7 +868,7 @@ int be_cmd_vlan_config(struct be_ctrl_info *ctrl, u32 if_id, u16 *vtag_array,
 
 	status = be_mbox_db_ring(ctrl);
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -736,7 +878,7 @@ int be_cmd_promiscuous_config(struct be_ctrl_info *ctrl, u8 port_num, bool en)
 	struct be_cmd_req_promiscuous_config *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -751,7 +893,7 @@ int be_cmd_promiscuous_config(struct be_ctrl_info *ctrl, u8 port_num, bool en)
 
 	status = be_mbox_db_ring(ctrl);
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -762,7 +904,7 @@ int be_cmd_mcast_mac_set(struct be_ctrl_info *ctrl, u32 if_id, u8 *mac_table,
 	struct be_cmd_req_mcast_mac_config *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -780,7 +922,7 @@ int be_cmd_mcast_mac_set(struct be_ctrl_info *ctrl, u32 if_id, u8 *mac_table,
 
 	status = be_mbox_db_ring(ctrl);
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -790,7 +932,7 @@ int be_cmd_set_flow_control(struct be_ctrl_info *ctrl, u32 tx_fc, u32 rx_fc)
 	struct be_cmd_req_set_flow_control *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 
 	memset(wrb, 0, sizeof(*wrb));
 
@@ -804,7 +946,7 @@ int be_cmd_set_flow_control(struct be_ctrl_info *ctrl, u32 tx_fc, u32 rx_fc)
 
 	status = be_mbox_db_ring(ctrl);
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -814,7 +956,7 @@ int be_cmd_get_flow_control(struct be_ctrl_info *ctrl, u32 *tx_fc, u32 *rx_fc)
 	struct be_cmd_req_get_flow_control *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 
 	memset(wrb, 0, sizeof(*wrb));
 
@@ -831,7 +973,7 @@ int be_cmd_get_flow_control(struct be_ctrl_info *ctrl, u32 *tx_fc, u32 *rx_fc)
 		*rx_fc = le16_to_cpu(resp->rx_flow_control);
 	}
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
 
@@ -841,7 +983,7 @@ int be_cmd_query_fw_cfg(struct be_ctrl_info *ctrl, u32 *port_num)
 	struct be_cmd_req_query_fw_cfg *req = embedded_payload(wrb);
 	int status;
 
-	spin_lock(&ctrl->cmd_lock);
+	spin_lock(&ctrl->mbox_lock);
 
 	memset(wrb, 0, sizeof(*wrb));
 
@@ -856,6 +998,6 @@ int be_cmd_query_fw_cfg(struct be_ctrl_info *ctrl, u32 *port_num)
 		*port_num = le32_to_cpu(resp->phys_port);
 	}
 
-	spin_unlock(&ctrl->cmd_lock);
+	spin_unlock(&ctrl->mbox_lock);
 	return status;
 }
diff --git a/drivers/net/benet/be_cmds.h b/drivers/net/benet/be_cmds.h
index e499e2d5b8c3..0a9189defc2a 100644
--- a/drivers/net/benet/be_cmds.h
+++ b/drivers/net/benet/be_cmds.h
@@ -101,6 +101,7 @@ struct be_mcc_mailbox {
 #define OPCODE_COMMON_FIRMWARE_CONFIG			42
 #define OPCODE_COMMON_NTWK_INTERFACE_CREATE 		50
 #define OPCODE_COMMON_NTWK_INTERFACE_DESTROY 		51
+#define OPCODE_COMMON_MCC_DESTROY        		53
 #define OPCODE_COMMON_CQ_DESTROY        		54
 #define OPCODE_COMMON_EQ_DESTROY        		55
 #define OPCODE_COMMON_QUERY_FIRMWARE_CONFIG		58
@@ -269,6 +270,38 @@ struct be_cmd_resp_cq_create {
 	u16 rsvd0;
 } __packed;
 
+/******************** Create MCCQ ***************************/
+/* Pseudo amap definition in which each bit of the actual structure is defined
+ * as a byte: used to calculate offset/shift/mask of each field */
+struct amap_mcc_context {
+	u8 con_index[14];
+	u8 rsvd0[2];
+	u8 ring_size[4];
+	u8 fetch_wrb;
+	u8 fetch_r2t;
+	u8 cq_id[10];
+	u8 prod_index[14];
+	u8 fid[8];
+	u8 pdid[9];
+	u8 valid;
+	u8 rsvd1[32];
+	u8 rsvd2[32];
+} __packed;
+
+struct be_cmd_req_mcc_create {
+	struct be_cmd_req_hdr hdr;
+	u16 num_pages;
+	u16 rsvd0;
+	u8 context[sizeof(struct amap_mcc_context) / 8];
+	struct phys_addr pages[8];
+} __packed;
+
+struct be_cmd_resp_mcc_create {
+	struct be_cmd_resp_hdr hdr;
+	u16 id;
+	u16 rsvd0;
+} __packed;
+
 /******************** Create TxQ ***************************/
 #define BE_ETH_TX_RING_TYPE_STANDARD    	2
 #define BE_ULP1_NUM				1
@@ -341,7 +374,8 @@ enum {
 	QTYPE_EQ = 1,
 	QTYPE_CQ,
 	QTYPE_TXQ,
-	QTYPE_RXQ
+	QTYPE_RXQ,
+	QTYPE_MCCQ
 };
 
 struct be_cmd_req_q_destroy {
@@ -657,6 +691,9 @@ extern int be_cmd_cq_create(struct be_ctrl_info *ctrl,
 			struct be_queue_info *cq, struct be_queue_info *eq,
 			bool sol_evts, bool no_delay,
 			int num_cqe_dma_coalesce);
+extern int be_cmd_mccq_create(struct be_ctrl_info *ctrl,
+			struct be_queue_info *mccq,
+			struct be_queue_info *cq);
 extern int be_cmd_txq_create(struct be_ctrl_info *ctrl,
 			struct be_queue_info *txq,
 			struct be_queue_info *cq);
@@ -686,3 +723,4 @@ extern int be_cmd_set_flow_control(struct be_ctrl_info *ctrl,
 extern int be_cmd_get_flow_control(struct be_ctrl_info *ctrl,
 			u32 *tx_fc, u32 *rx_fc);
 extern int be_cmd_query_fw_cfg(struct be_ctrl_info *ctrl, u32 *port_num);
+extern void be_process_mcc(struct be_ctrl_info *ctrl);
diff --git a/drivers/net/benet/be_hw.h b/drivers/net/benet/be_hw.h
index b132aa4893ca..b02e805c1db3 100644
--- a/drivers/net/benet/be_hw.h
+++ b/drivers/net/benet/be_hw.h
@@ -61,7 +61,7 @@
 /* Clear the interrupt for this eq */
 #define DB_EQ_CLR_SHIFT			(9)	/* bit 9 */
 /* Must be 1 */
-#define DB_EQ_EVNT_SHIFT			(10)	/* bit 10 */
+#define DB_EQ_EVNT_SHIFT		(10)	/* bit 10 */
 /* Number of event entries processed */
 #define DB_EQ_NUM_POPPED_SHIFT		(16)	/* bits 16 - 28 */
 /* Rearm bit */
@@ -88,6 +88,12 @@
 /* Number of rx frags posted */
 #define DB_RQ_NUM_POSTED_SHIFT		(24)	/* bits 24 - 31 */
 
+/********** MCC door bell ************/
+#define DB_MCCQ_OFFSET 			0x140
+#define DB_MCCQ_RING_ID_MASK		0x7FF	/* bits 0 - 10 */
+/* Number of entries posted */
+#define DB_MCCQ_NUM_POSTED_SHIFT	(16)	/* bits 16 - 29 */
+
 /*
  * BE descriptors: host memory data structures whose formats
  * are hardwired in BE silicon.
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index 66bb56874d9b..a4ce80e776b6 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -60,26 +60,6 @@ static int be_queue_alloc(struct be_adapter *adapter, struct be_queue_info *q,
 	return 0;
 }
 
-static inline void *queue_head_node(struct be_queue_info *q)
-{
-	return q->dma_mem.va + q->head * q->entry_size;
-}
-
-static inline void *queue_tail_node(struct be_queue_info *q)
-{
-	return q->dma_mem.va + q->tail * q->entry_size;
-}
-
-static inline void queue_head_inc(struct be_queue_info *q)
-{
-	index_inc(&q->head, q->len);
-}
-
-static inline void queue_tail_inc(struct be_queue_info *q)
-{
-	index_inc(&q->tail, q->len);
-}
-
 static void be_intr_set(struct be_ctrl_info *ctrl, bool enable)
 {
 	u8 __iomem *addr = ctrl->pcicfg + PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET;
@@ -127,7 +107,7 @@ static void be_eq_notify(struct be_ctrl_info *ctrl, u16 qid,
 	iowrite32(val, ctrl->db + DB_EQ_OFFSET);
 }
 
-static void be_cq_notify(struct be_ctrl_info *ctrl, u16 qid,
+void be_cq_notify(struct be_ctrl_info *ctrl, u16 qid,
 		bool arm, u16 num_popped)
 {
 	u32 val = 0;
@@ -960,10 +940,8 @@ static void be_post_rx_frags(struct be_adapter *adapter)
 	return;
 }
 
-static struct be_eth_tx_compl *
-be_tx_compl_get(struct be_adapter *adapter)
+static struct be_eth_tx_compl *be_tx_compl_get(struct be_queue_info *tx_cq)
 {
-	struct be_queue_info *tx_cq = &adapter->tx_obj.cq;
 	struct be_eth_tx_compl *txcp = queue_tail_node(tx_cq);
 
 	if (txcp->dw[offsetof(struct amap_eth_tx_compl, valid) / 32] == 0)
@@ -1051,6 +1029,59 @@ static void be_tx_q_clean(struct be_adapter *adapter)
 	}
 }
 
+static void be_mcc_queues_destroy(struct be_adapter *adapter)
+{
+	struct be_queue_info *q;
+	struct be_ctrl_info *ctrl = &adapter->ctrl;
+
+	q = &ctrl->mcc_obj.q;
+	if (q->created)
+		be_cmd_q_destroy(ctrl, q, QTYPE_MCCQ);
+	be_queue_free(adapter, q);
+
+	q = &ctrl->mcc_obj.cq;
+	if (q->created)
+		be_cmd_q_destroy(ctrl, q, QTYPE_CQ);
+	be_queue_free(adapter, q);
+}
+
+/* Must be called only after TX qs are created as MCC shares TX EQ */
+static int be_mcc_queues_create(struct be_adapter *adapter)
+{
+	struct be_queue_info *q, *cq;
+	struct be_ctrl_info *ctrl = &adapter->ctrl;
+
+	/* Alloc MCC compl queue */
+	cq = &ctrl->mcc_obj.cq;
+	if (be_queue_alloc(adapter, cq, MCC_CQ_LEN,
+			sizeof(struct be_mcc_cq_entry)))
+		goto err;
+
+	/* Ask BE to create MCC compl queue; share TX's eq */
+	if (be_cmd_cq_create(ctrl, cq, &adapter->tx_eq.q, false, true, 0))
+		goto mcc_cq_free;
+
+	/* Alloc MCC queue */
+	q = &ctrl->mcc_obj.q;
+	if (be_queue_alloc(adapter, q, MCC_Q_LEN, sizeof(struct be_mcc_wrb)))
+		goto mcc_cq_destroy;
+
+	/* Ask BE to create MCC queue */
+	if (be_cmd_mccq_create(ctrl, q, cq))
+		goto mcc_q_free;
+
+	return 0;
+
+mcc_q_free:
+	be_queue_free(adapter, q);
+mcc_cq_destroy:
+	be_cmd_q_destroy(ctrl, cq, QTYPE_CQ);
+mcc_cq_free:
+	be_queue_free(adapter, cq);
+err:
+	return -1;
+}
+
 static void be_tx_queues_destroy(struct be_adapter *adapter)
 {
 	struct be_queue_info *q;
@@ -1263,7 +1294,7 @@ static irqreturn_t be_msix_rx(int irq, void *dev)
 	return IRQ_HANDLED;
 }
 
-static irqreturn_t be_msix_tx(int irq, void *dev)
+static irqreturn_t be_msix_tx_mcc(int irq, void *dev)
 {
 	struct be_adapter *adapter = dev;
 
@@ -1324,40 +1355,51 @@ int be_poll_rx(struct napi_struct *napi, int budget)
 	return work_done;
 }
 
-/* For TX we don't honour budget; consume everything */
-int be_poll_tx(struct napi_struct *napi, int budget)
+void be_process_tx(struct be_adapter *adapter)
 {
-	struct be_eq_obj *tx_eq = container_of(napi, struct be_eq_obj, napi);
-	struct be_adapter *adapter =
-		container_of(tx_eq, struct be_adapter, tx_eq);
-	struct be_tx_obj *tx_obj = &adapter->tx_obj;
-	struct be_queue_info *tx_cq = &tx_obj->cq;
-	struct be_queue_info *txq = &tx_obj->q;
+	struct be_queue_info *txq = &adapter->tx_obj.q;
+	struct be_queue_info *tx_cq = &adapter->tx_obj.cq;
 	struct be_eth_tx_compl *txcp;
 	u32 num_cmpl = 0;
 	u16 end_idx;
 
-	while ((txcp = be_tx_compl_get(adapter))) {
+	while ((txcp = be_tx_compl_get(tx_cq))) {
 		end_idx = AMAP_GET_BITS(struct amap_eth_tx_compl,
 					wrb_index, txcp);
 		be_tx_compl_process(adapter, end_idx);
 		num_cmpl++;
 	}
 
-	/* As Tx wrbs have been freed up, wake up netdev queue if
-	 * it was stopped due to lack of tx wrbs.
-	 */
-	if (netif_queue_stopped(adapter->netdev) &&
+	if (num_cmpl) {
+		be_cq_notify(&adapter->ctrl, tx_cq->id, true, num_cmpl);
+
+		/* As Tx wrbs have been freed up, wake up netdev queue if
+		 * it was stopped due to lack of tx wrbs.
+		 */
+		if (netif_queue_stopped(adapter->netdev) &&
 			atomic_read(&txq->used) < txq->len / 2) {
-		netif_wake_queue(adapter->netdev);
+			netif_wake_queue(adapter->netdev);
+		}
+
+		drvr_stats(adapter)->be_tx_events++;
+		drvr_stats(adapter)->be_tx_compl += num_cmpl;
 	}
+}
+
+/* As TX and MCC share the same EQ check for both TX and MCC completions.
+ * For TX/MCC we don't honour budget; consume everything
+ */
+static int be_poll_tx_mcc(struct napi_struct *napi, int budget)
+{
+	struct be_eq_obj *tx_eq = container_of(napi, struct be_eq_obj, napi);
+	struct be_adapter *adapter =
+		container_of(tx_eq, struct be_adapter, tx_eq);
 
 	napi_complete(napi);
 
-	be_cq_notify(&adapter->ctrl, tx_cq->id, true, num_cmpl);
+	be_process_tx(adapter);
 
-	drvr_stats(adapter)->be_tx_events++;
-	drvr_stats(adapter)->be_tx_compl += num_cmpl;
+	be_process_mcc(&adapter->ctrl);
 
 	return 1;
 }
@@ -1419,7 +1461,7 @@ static int be_msix_register(struct be_adapter *adapter)
 
 	sprintf(tx_eq->desc, "%s-tx", netdev->name);
 	vec = be_msix_vec_get(adapter, tx_eq->q.id);
-	status = request_irq(vec, be_msix_tx, 0, tx_eq->desc, adapter);
+	status = request_irq(vec, be_msix_tx_mcc, 0, tx_eq->desc, adapter);
 	if (status)
 		goto err;
 
@@ -1495,6 +1537,34 @@ static int be_open(struct net_device *netdev)
 	struct be_ctrl_info *ctrl = &adapter->ctrl;
 	struct be_eq_obj *rx_eq = &adapter->rx_eq;
 	struct be_eq_obj *tx_eq = &adapter->tx_eq;
+
+	/* First time posting */
+	be_post_rx_frags(adapter);
+
+	napi_enable(&rx_eq->napi);
+	napi_enable(&tx_eq->napi);
+
+	be_irq_register(adapter);
+
+	be_intr_set(ctrl, true);
+
+	/* The evt queues are created in unarmed state; arm them */
+	be_eq_notify(ctrl, rx_eq->q.id, true, false, 0);
+	be_eq_notify(ctrl, tx_eq->q.id, true, false, 0);
+
+	/* Rx compl queue may be in unarmed state; rearm it */
+	be_cq_notify(ctrl, adapter->rx_obj.cq.id, true, 0);
+
+	be_link_status_update(adapter);
+
+	schedule_delayed_work(&adapter->work, msecs_to_jiffies(100));
+	return 0;
+}
+
+static int be_setup(struct be_adapter *adapter)
+{
+	struct be_ctrl_info *ctrl = &adapter->ctrl;
+	struct net_device *netdev = adapter->netdev;
 	u32 if_flags;
 	int status;
 
@@ -1521,29 +1591,14 @@ static int be_open(struct net_device *netdev)
 	if (status != 0)
 		goto tx_qs_destroy;
 
-	/* First time posting */
-	be_post_rx_frags(adapter);
-
-	napi_enable(&rx_eq->napi);
-	napi_enable(&tx_eq->napi);
-
-	be_irq_register(adapter);
-
-	be_intr_set(ctrl, true);
-
-	/* The evt queues are created in the unarmed state; arm them */
-	be_eq_notify(ctrl, rx_eq->q.id, true, false, 0);
-	be_eq_notify(ctrl, tx_eq->q.id, true, false, 0);
-
-	/* The compl queues are created in the unarmed state; arm them */
-	be_cq_notify(ctrl, adapter->rx_obj.cq.id, true, 0);
-	be_cq_notify(ctrl, adapter->tx_obj.cq.id, true, 0);
-
-	be_link_status_update(adapter);
+	status = be_mcc_queues_create(adapter);
+	if (status != 0)
+		goto rx_qs_destroy;
 
-	schedule_delayed_work(&adapter->work, msecs_to_jiffies(100));
 	return 0;
 
+rx_qs_destroy:
+	be_rx_queues_destroy(adapter);
 tx_qs_destroy:
 	be_tx_queues_destroy(adapter);
 if_destroy:
@@ -1552,6 +1607,19 @@ do_none:
 	return status;
 }
 
+static int be_clear(struct be_adapter *adapter)
+{
+	struct be_ctrl_info *ctrl = &adapter->ctrl;
+
+	be_rx_queues_destroy(adapter);
+	be_tx_queues_destroy(adapter);
+
+	be_cmd_if_destroy(ctrl, adapter->if_handle);
+
+	be_mcc_queues_destroy(adapter);
+	return 0;
+}
+
 static int be_close(struct net_device *netdev)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
@@ -1581,10 +1649,6 @@ static int be_close(struct net_device *netdev)
 	napi_disable(&rx_eq->napi);
 	napi_disable(&tx_eq->napi);
 
-	be_rx_queues_destroy(adapter);
-	be_tx_queues_destroy(adapter);
-
-	be_cmd_if_destroy(ctrl, adapter->if_handle);
 	return 0;
 }
 
@@ -1673,7 +1737,7 @@ static void be_netdev_init(struct net_device *netdev)
 
 	netif_napi_add(netdev, &adapter->rx_eq.napi, be_poll_rx,
 		BE_NAPI_WEIGHT);
-	netif_napi_add(netdev, &adapter->tx_eq.napi, be_poll_tx,
+	netif_napi_add(netdev, &adapter->tx_eq.napi, be_poll_tx_mcc,
 		BE_NAPI_WEIGHT);
 
 	netif_carrier_off(netdev);
@@ -1755,7 +1819,9 @@ static int be_ctrl_init(struct be_adapter *adapter)
 	mbox_mem_align->va = PTR_ALIGN(mbox_mem_alloc->va, 16);
 	mbox_mem_align->dma = PTR_ALIGN(mbox_mem_alloc->dma, 16);
 	memset(mbox_mem_align->va, 0, sizeof(struct be_mcc_mailbox));
-	spin_lock_init(&ctrl->cmd_lock);
+	spin_lock_init(&ctrl->mbox_lock);
+	spin_lock_init(&ctrl->mcc_lock);
+	spin_lock_init(&ctrl->mcc_cq_lock);
 
 	val = ioread32(ctrl->pcicfg + PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET);
 	ctrl->pci_func = (val >> MEMBAR_CTRL_INT_CTRL_PFUNC_SHIFT) &
@@ -1793,6 +1859,8 @@ static void __devexit be_remove(struct pci_dev *pdev)
 
 	unregister_netdev(adapter->netdev);
 
+	be_clear(adapter);
+
 	be_stats_cleanup(adapter);
 
 	be_ctrl_cleanup(adapter);
@@ -1890,13 +1958,18 @@ static int __devinit be_probe(struct pci_dev *pdev,
 	be_netdev_init(netdev);
 	SET_NETDEV_DEV(netdev, &adapter->pdev->dev);
 
+	status = be_setup(adapter);
+	if (status)
+		goto stats_clean;
 	status = register_netdev(netdev);
 	if (status != 0)
-		goto stats_clean;
+		goto unsetup;
 
 	dev_info(&pdev->dev, "%s port %d\n", nic_name(pdev), adapter->port_num);
 	return 0;
 
+unsetup:
+	be_clear(adapter);
 stats_clean:
 	be_stats_cleanup(adapter);
 ctrl_clean:
@@ -1921,6 +1994,7 @@ static int be_suspend(struct pci_dev *pdev, pm_message_t state)
 	if (netif_running(netdev)) {
 		rtnl_lock();
 		be_close(netdev);
+		be_clear(adapter);
 		rtnl_unlock();
 	}
 
@@ -1947,6 +2021,7 @@ static int be_resume(struct pci_dev *pdev)
 
 	if (netif_running(netdev)) {
 		rtnl_lock();
+		be_setup(adapter);
 		be_open(netdev);
 		rtnl_unlock();
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 6ac7b687cb3acc437a586794949a43f5249956bb Mon Sep 17 00:00:00 2001
From: Sathya Perla <sathyap@serverengines.com>
Date: Thu, 18 Jun 2009 00:05:54 +0000
Subject: be2net: Use MCC queue for cmds that may be called in BH context

Currenlty multicast_set and promiscuous_config cmds -- that may be called in BH context --
use the blocking MCC mbox to post cmds.
An mbox cmd is protected via a spin_lock(cmd_lock) and not spin_lock_bh() as it is undesirable
to disable BHs while a blocking mbox cmd is in progress (and take long to finish.)
This can lockup a cmd in progress in process context.
So, these two cmds in BH context must use the MCC queue to post cmds.

Signed-off-by: Sathya Perla <sathyap@serverengines.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/benet/be_cmds.c | 69 ++++++++++++++++++++++++++++++++++-----------
 1 file changed, 52 insertions(+), 17 deletions(-)

diff --git a/drivers/net/benet/be_cmds.c b/drivers/net/benet/be_cmds.c
index f1ec191f0c0d..e4ad5e67fde7 100644
--- a/drivers/net/benet/be_cmds.c
+++ b/drivers/net/benet/be_cmds.c
@@ -17,7 +17,7 @@
 
 #include "be.h"
 
-void be_mcc_notify(struct be_ctrl_info *ctrl)
+static void be_mcc_notify(struct be_ctrl_info *ctrl)
 {
 	struct be_queue_info *mccq = &ctrl->mcc_obj.q;
 	u32 val = 0;
@@ -101,6 +101,28 @@ void be_process_mcc(struct be_ctrl_info *ctrl)
 	spin_unlock_bh(&ctrl->mcc_cq_lock);
 }
 
+/* Wait till no more pending mcc requests are present */
+static void be_mcc_wait_compl(struct be_ctrl_info *ctrl)
+{
+#define mcc_timeout		50000 /* 5s timeout */
+	int i;
+	for (i = 0; i < mcc_timeout; i++) {
+		be_process_mcc(ctrl);
+		if (atomic_read(&ctrl->mcc_obj.q.used) == 0)
+			break;
+		udelay(100);
+	}
+	if (i == mcc_timeout)
+		printk(KERN_WARNING DRV_NAME "mcc poll timed out\n");
+}
+
+/* Notify MCC requests and wait for completion */
+static void be_mcc_notify_wait(struct be_ctrl_info *ctrl)
+{
+	be_mcc_notify(ctrl);
+	be_mcc_wait_compl(ctrl);
+}
+
 static int be_mbox_db_ready_wait(void __iomem *db)
 {
 	int cnt = 0, wait = 5;
@@ -872,14 +894,18 @@ int be_cmd_vlan_config(struct be_ctrl_info *ctrl, u32 if_id, u16 *vtag_array,
 	return status;
 }
 
+/* Use MCC for this command as it may be called in BH context */
 int be_cmd_promiscuous_config(struct be_ctrl_info *ctrl, u8 port_num, bool en)
 {
-	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
-	struct be_cmd_req_promiscuous_config *req = embedded_payload(wrb);
-	int status;
+	struct be_mcc_wrb *wrb;
+	struct be_cmd_req_promiscuous_config *req;
 
-	spin_lock(&ctrl->mbox_lock);
-	memset(wrb, 0, sizeof(*wrb));
+	spin_lock_bh(&ctrl->mcc_lock);
+
+	wrb = wrb_from_mcc(&ctrl->mcc_obj.q);
+	BUG_ON(!wrb);
+
+	req = embedded_payload(wrb);
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
 
@@ -891,21 +917,29 @@ int be_cmd_promiscuous_config(struct be_ctrl_info *ctrl, u8 port_num, bool en)
 	else
 		req->port0_promiscuous = en;
 
-	status = be_mbox_db_ring(ctrl);
+	be_mcc_notify_wait(ctrl);
 
-	spin_unlock(&ctrl->mbox_lock);
-	return status;
+	spin_unlock_bh(&ctrl->mcc_lock);
+	return 0;
 }
 
+/*
+ * Use MCC for this command as it may be called in BH context
+ * (mc == NULL) => multicast promiscous
+ */
 int be_cmd_mcast_mac_set(struct be_ctrl_info *ctrl, u32 if_id, u8 *mac_table,
 			u32 num, bool promiscuous)
 {
-	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
-	struct be_cmd_req_mcast_mac_config *req = embedded_payload(wrb);
-	int status;
+#define BE_MAX_MC		32 /* set mcast promisc if > 32 */
+	struct be_mcc_wrb *wrb;
+	struct be_cmd_req_mcast_mac_config *req;
 
-	spin_lock(&ctrl->mbox_lock);
-	memset(wrb, 0, sizeof(*wrb));
+	spin_lock_bh(&ctrl->mcc_lock);
+
+	wrb = wrb_from_mcc(&ctrl->mcc_obj.q);
+	BUG_ON(!wrb);
+
+	req = embedded_payload(wrb);
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
 
@@ -920,10 +954,11 @@ int be_cmd_mcast_mac_set(struct be_ctrl_info *ctrl, u32 if_id, u8 *mac_table,
 			memcpy(req->mac, mac_table, ETH_ALEN * num);
 	}
 
-	status = be_mbox_db_ring(ctrl);
+	be_mcc_notify_wait(ctrl);
 
-	spin_unlock(&ctrl->mbox_lock);
-	return status;
+	spin_unlock_bh(&ctrl->mcc_lock);
+
+	return 0;
 }
 
 int be_cmd_set_flow_control(struct be_ctrl_info *ctrl, u32 tx_fc, u32 rx_fc)
-- 
cgit v1.2.3-59-g8ed1b


From 24307eef74bd38e3fc6a6df8f8a1bfc48967f9f6 Mon Sep 17 00:00:00 2001
From: Sathya Perla <sathyap@serverengines.com>
Date: Thu, 18 Jun 2009 00:09:25 +0000
Subject: be2net: cleanup multicast_set cmd to avoid mc_list copy

Cleanup multicast_set method to avoid an extra copy of mc_list
 and unwanted promiscuos sets to BE.

Signed-off-by: Sathya Perla <sathyap@serverengines.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/benet/be.h      |  1 +
 drivers/net/benet/be_cmds.c | 19 +++++++++++-------
 drivers/net/benet/be_cmds.h |  4 ++--
 drivers/net/benet/be_main.c | 49 ++++++++++++++++-----------------------------
 4 files changed, 32 insertions(+), 41 deletions(-)

diff --git a/drivers/net/benet/be.h b/drivers/net/benet/be.h
index ef5be133ce68..94b75cb072f7 100644
--- a/drivers/net/benet/be.h
+++ b/drivers/net/benet/be.h
@@ -274,6 +274,7 @@ struct be_adapter {
 
 	struct be_link_info link;
 	u32 port_num;
+	bool promiscuous;
 };
 
 extern struct ethtool_ops be_ethtool_ops;
diff --git a/drivers/net/benet/be_cmds.c b/drivers/net/benet/be_cmds.c
index e4ad5e67fde7..4a2e1f518f78 100644
--- a/drivers/net/benet/be_cmds.c
+++ b/drivers/net/benet/be_cmds.c
@@ -927,8 +927,8 @@ int be_cmd_promiscuous_config(struct be_ctrl_info *ctrl, u8 port_num, bool en)
  * Use MCC for this command as it may be called in BH context
  * (mc == NULL) => multicast promiscous
  */
-int be_cmd_mcast_mac_set(struct be_ctrl_info *ctrl, u32 if_id, u8 *mac_table,
-			u32 num, bool promiscuous)
+int be_cmd_multicast_set(struct be_ctrl_info *ctrl, u32 if_id,
+		struct dev_mc_list *mc_list, u32 mc_count)
 {
 #define BE_MAX_MC		32 /* set mcast promisc if > 32 */
 	struct be_mcc_wrb *wrb;
@@ -947,11 +947,16 @@ int be_cmd_mcast_mac_set(struct be_ctrl_info *ctrl, u32 if_id, u8 *mac_table,
 		OPCODE_COMMON_NTWK_MULTICAST_SET, sizeof(*req));
 
 	req->interface_id = if_id;
-	req->promiscuous = promiscuous;
-	if (!promiscuous) {
-		req->num_mac = cpu_to_le16(num);
-		if (num)
-			memcpy(req->mac, mac_table, ETH_ALEN * num);
+	if (mc_list && mc_count <= BE_MAX_MC) {
+		int i;
+		struct dev_mc_list *mc;
+
+		req->num_mac = cpu_to_le16(mc_count);
+
+		for (mc = mc_list, i = 0; mc; mc = mc->next, i++)
+			memcpy(req->mac[i].byte, mc->dmi_addr, ETH_ALEN);
+	} else {
+		req->promiscuous = 1;
 	}
 
 	be_mcc_notify_wait(ctrl);
diff --git a/drivers/net/benet/be_cmds.h b/drivers/net/benet/be_cmds.h
index 0a9189defc2a..a567aa437ec9 100644
--- a/drivers/net/benet/be_cmds.h
+++ b/drivers/net/benet/be_cmds.h
@@ -716,8 +716,8 @@ extern int be_cmd_vlan_config(struct be_ctrl_info *ctrl, u32 if_id,
 			bool promiscuous);
 extern int be_cmd_promiscuous_config(struct be_ctrl_info *ctrl,
 			u8 port_num, bool en);
-extern int be_cmd_mcast_mac_set(struct be_ctrl_info *ctrl, u32 if_id,
-			u8 *mac_table, u32 num, bool promiscuous);
+extern int be_cmd_multicast_set(struct be_ctrl_info *ctrl, u32 if_id,
+			struct dev_mc_list *mc_list, u32 mc_count);
 extern int be_cmd_set_flow_control(struct be_ctrl_info *ctrl,
 			u32 tx_fc, u32 rx_fc);
 extern int be_cmd_get_flow_control(struct be_ctrl_info *ctrl,
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index a4ce80e776b6..3dc68034c21d 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -549,47 +549,32 @@ static void be_vlan_rem_vid(struct net_device *netdev, u16 vid)
 	be_vid_config(netdev);
 }
 
-static void be_set_multicast_filter(struct net_device *netdev)
+static void be_set_multicast_list(struct net_device *netdev)
 {
 	struct be_adapter *adapter = netdev_priv(netdev);
-	struct dev_mc_list *mc_ptr;
-	u8 mac_addr[32][ETH_ALEN];
-	int i = 0;
+	struct be_ctrl_info *ctrl = &adapter->ctrl;
 
-	if (netdev->flags & IFF_ALLMULTI) {
-		/* set BE in Multicast promiscuous */
-		be_cmd_mcast_mac_set(&adapter->ctrl,
-					adapter->if_handle, NULL, 0, true);
-		return;
+	if (netdev->flags & IFF_PROMISC) {
+		be_cmd_promiscuous_config(ctrl, adapter->port_num, 1);
+		adapter->promiscuous = true;
+		goto done;
 	}
 
-	for (mc_ptr = netdev->mc_list; mc_ptr; mc_ptr = mc_ptr->next) {
-		memcpy(&mac_addr[i][0], mc_ptr->dmi_addr, ETH_ALEN);
-		if (++i >= 32) {
-			be_cmd_mcast_mac_set(&adapter->ctrl,
-				adapter->if_handle, &mac_addr[0][0], i, false);
-			i = 0;
-		}
-
+	/* BE was previously in promiscous mode; disable it */
+	if (adapter->promiscuous) {
+		adapter->promiscuous = false;
+		be_cmd_promiscuous_config(ctrl, adapter->port_num, 0);
 	}
 
-	if (i) {
-		/* reset the promiscuous mode also. */
-		be_cmd_mcast_mac_set(&adapter->ctrl,
-			adapter->if_handle, &mac_addr[0][0], i, false);
+	if (netdev->flags & IFF_ALLMULTI) {
+		be_cmd_multicast_set(ctrl, adapter->if_handle, NULL, 0);
+		goto done;
 	}
-}
-
-static void be_set_multicast_list(struct net_device *netdev)
-{
-	struct be_adapter *adapter = netdev_priv(netdev);
 
-	if (netdev->flags & IFF_PROMISC) {
-		be_cmd_promiscuous_config(&adapter->ctrl, adapter->port_num, 1);
-	} else {
-		be_cmd_promiscuous_config(&adapter->ctrl, adapter->port_num, 0);
-		be_set_multicast_filter(netdev);
-	}
+	be_cmd_multicast_set(ctrl, adapter->if_handle, netdev->mc_list,
+		netdev->mc_count);
+done:
+	return;
 }
 
 static void be_rx_rate_update(struct be_adapter *adapter)
-- 
cgit v1.2.3-59-g8ed1b


From a8f447bda3ee00e3a3ab080c48db40078ea65221 Mon Sep 17 00:00:00 2001
From: Sathya Perla <sathyap@serverengines.com>
Date: Thu, 18 Jun 2009 00:10:27 +0000
Subject: be2net: receive asynchronous link status notifications from BE

Rcv and process ansync link status notifications from BE instead of polling
 for link status in the be_worker thread.

Signed-off-by: Sathya Perla <sathyap@serverengines.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/benet/be.h      |  6 +++++-
 drivers/net/benet/be_cmds.c | 34 +++++++++++++++++++++++++++-------
 drivers/net/benet/be_cmds.h | 36 +++++++++++++++++++++++++++++-------
 drivers/net/benet/be_main.c | 37 +++++++++++++++++++------------------
 4 files changed, 80 insertions(+), 33 deletions(-)

diff --git a/drivers/net/benet/be.h b/drivers/net/benet/be.h
index 94b75cb072f7..f703758f0a6e 100644
--- a/drivers/net/benet/be.h
+++ b/drivers/net/benet/be.h
@@ -163,6 +163,10 @@ struct be_ctrl_info {
 	struct be_mcc_obj mcc_obj;
 	spinlock_t mcc_lock;	/* For serializing mcc cmds to BE card */
 	spinlock_t mcc_cq_lock;
+
+	/* MCC Async callback */
+	void (*async_cb)(void *adapter, bool link_up);
+	void *adapter_ctxt;
 };
 
 #include "be_cmds.h"
@@ -272,7 +276,7 @@ struct be_adapter {
 	u32 if_handle;		/* Used to configure filtering */
 	u32 pmac_id;		/* MAC addr handle used by BE card */
 
-	struct be_link_info link;
+	bool link_up;
 	u32 port_num;
 	bool promiscuous;
 };
diff --git a/drivers/net/benet/be_cmds.c b/drivers/net/benet/be_cmds.c
index 4a2e1f518f78..583517ed56f0 100644
--- a/drivers/net/benet/be_cmds.c
+++ b/drivers/net/benet/be_cmds.c
@@ -69,6 +69,20 @@ static int be_mcc_compl_process(struct be_ctrl_info *ctrl,
 	return 0;
 }
 
+/* Link state evt is a string of bytes; no need for endian swapping */
+static void be_async_link_state_process(struct be_ctrl_info *ctrl,
+		struct be_async_event_link_state *evt)
+{
+	ctrl->async_cb(ctrl->adapter_ctxt,
+		evt->port_link_status == ASYNC_EVENT_LINK_UP ? true : false);
+}
+
+static inline bool is_link_state_evt(u32 trailer)
+{
+	return (((trailer >> ASYNC_TRAILER_EVENT_CODE_SHIFT) &
+		ASYNC_TRAILER_EVENT_CODE_MASK) ==
+				ASYNC_EVENT_CODE_LINK_STATE);
+}
 
 static struct be_mcc_cq_entry *be_mcc_compl_get(struct be_ctrl_info *ctrl)
 {
@@ -89,7 +103,14 @@ void be_process_mcc(struct be_ctrl_info *ctrl)
 
 	spin_lock_bh(&ctrl->mcc_cq_lock);
 	while ((compl = be_mcc_compl_get(ctrl))) {
-		if (!(compl->flags & CQE_FLAGS_ASYNC_MASK)) {
+		if (compl->flags & CQE_FLAGS_ASYNC_MASK) {
+			/* Interpret flags as an async trailer */
+			BUG_ON(!is_link_state_evt(compl->flags));
+
+			/* Interpret compl as a async link evt */
+			be_async_link_state_process(ctrl,
+				(struct be_async_event_link_state *) compl);
+		} else {
 			be_mcc_compl_process(ctrl, compl);
 			atomic_dec(&ctrl->mcc_obj.q.used);
 		}
@@ -786,13 +807,15 @@ int be_cmd_get_stats(struct be_ctrl_info *ctrl, struct be_dma_mem *nonemb_cmd)
 }
 
 int be_cmd_link_status_query(struct be_ctrl_info *ctrl,
-			struct be_link_info *link)
+			bool *link_up)
 {
 	struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem);
 	struct be_cmd_req_link_status *req = embedded_payload(wrb);
 	int status;
 
 	spin_lock(&ctrl->mbox_lock);
+
+	*link_up = false;
 	memset(wrb, 0, sizeof(*wrb));
 
 	be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0);
@@ -803,11 +826,8 @@ int be_cmd_link_status_query(struct be_ctrl_info *ctrl,
 	status = be_mbox_db_ring(ctrl);
 	if (!status) {
 		struct be_cmd_resp_link_status *resp = embedded_payload(wrb);
-		link->speed = resp->mac_speed;
-		link->duplex = resp->mac_duplex;
-		link->fault = resp->mac_fault;
-	} else {
-		link->speed = PHY_LINK_SPEED_ZERO;
+		if (resp->mac_speed != PHY_LINK_SPEED_ZERO)
+			*link_up = true;
 	}
 
 	spin_unlock(&ctrl->mbox_lock);
diff --git a/drivers/net/benet/be_cmds.h b/drivers/net/benet/be_cmds.h
index a567aa437ec9..747626da7b4e 100644
--- a/drivers/net/benet/be_cmds.h
+++ b/drivers/net/benet/be_cmds.h
@@ -76,6 +76,34 @@ struct be_mcc_cq_entry {
 	u32 flags;		/* dword 3 */
 };
 
+/* When the async bit of mcc_compl is set, the last 4 bytes of
+ * mcc_compl is interpreted as follows:
+ */
+#define ASYNC_TRAILER_EVENT_CODE_SHIFT	8	/* bits 8 - 15 */
+#define ASYNC_TRAILER_EVENT_CODE_MASK	0xFF
+#define ASYNC_EVENT_CODE_LINK_STATE	0x1
+struct be_async_event_trailer {
+	u32 code;
+};
+
+enum {
+	ASYNC_EVENT_LINK_DOWN 	= 0x0,
+	ASYNC_EVENT_LINK_UP 	= 0x1
+};
+
+/* When the event code of an async trailer is link-state, the mcc_compl
+ * must be interpreted as follows
+ */
+struct be_async_event_link_state {
+	u8 physical_port;
+	u8 port_link_status;
+	u8 port_duplex;
+	u8 port_speed;
+	u8 port_fault;
+	u8 rsvd0[7];
+	struct be_async_event_trailer trailer;
+} __packed;
+
 struct be_mcc_mailbox {
 	struct be_mcc_wrb wrb;
 	struct be_mcc_cq_entry cqe;
@@ -580,12 +608,6 @@ struct be_cmd_req_link_status {
 	u32 rsvd;
 };
 
-struct be_link_info {
-	u8 duplex;
-	u8 speed;
-	u8 fault;
-};
-
 enum {
 	PHY_LINK_DUPLEX_NONE = 0x0,
 	PHY_LINK_DUPLEX_HALF = 0x1,
@@ -704,7 +726,7 @@ extern int be_cmd_rxq_create(struct be_ctrl_info *ctrl,
 extern int be_cmd_q_destroy(struct be_ctrl_info *ctrl, struct be_queue_info *q,
 			int type);
 extern int be_cmd_link_status_query(struct be_ctrl_info *ctrl,
-			struct be_link_info *link);
+			bool *link_up);
 extern int be_cmd_reset(struct be_ctrl_info *ctrl);
 extern int be_cmd_get_stats(struct be_ctrl_info *ctrl,
 			struct be_dma_mem *nonemb_cmd);
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index 3dc68034c21d..66c10c87f517 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -214,28 +214,24 @@ static void netdev_stats_update(struct be_adapter *adapter)
 	dev_stats->tx_window_errors = 0;
 }
 
-static void be_link_status_update(struct be_adapter *adapter)
+void be_link_status_update(void *ctxt, bool link_up)
 {
-	struct be_link_info *prev = &adapter->link;
-	struct be_link_info now = { 0 };
+	struct be_adapter *adapter = ctxt;
 	struct net_device *netdev = adapter->netdev;
 
-	be_cmd_link_status_query(&adapter->ctrl, &now);
-
 	/* If link came up or went down */
-	if (now.speed != prev->speed && (now.speed == PHY_LINK_SPEED_ZERO ||
-			prev->speed == PHY_LINK_SPEED_ZERO)) {
-		if (now.speed == PHY_LINK_SPEED_ZERO) {
-			netif_stop_queue(netdev);
-			netif_carrier_off(netdev);
-			printk(KERN_INFO "%s: Link down\n", netdev->name);
-		} else {
+	if (adapter->link_up != link_up) {
+		if (link_up) {
 			netif_start_queue(netdev);
 			netif_carrier_on(netdev);
 			printk(KERN_INFO "%s: Link up\n", netdev->name);
+		} else {
+			netif_stop_queue(netdev);
+			netif_carrier_off(netdev);
+			printk(KERN_INFO "%s: Link down\n", netdev->name);
 		}
+		adapter->link_up = link_up;
 	}
-	*prev = now;
 }
 
 /* Update the EQ delay n BE based on the RX frags consumed / sec */
@@ -1395,9 +1391,6 @@ static void be_worker(struct work_struct *work)
 		container_of(work, struct be_adapter, work.work);
 	int status;
 
-	/* Check link */
-	be_link_status_update(adapter);
-
 	/* Get Stats */
 	status = be_cmd_get_stats(&adapter->ctrl, &adapter->stats.cmd);
 	if (!status)
@@ -1522,6 +1515,8 @@ static int be_open(struct net_device *netdev)
 	struct be_ctrl_info *ctrl = &adapter->ctrl;
 	struct be_eq_obj *rx_eq = &adapter->rx_eq;
 	struct be_eq_obj *tx_eq = &adapter->tx_eq;
+	bool link_up;
+	int status;
 
 	/* First time posting */
 	be_post_rx_frags(adapter);
@@ -1540,7 +1535,10 @@ static int be_open(struct net_device *netdev)
 	/* Rx compl queue may be in unarmed state; rearm it */
 	be_cq_notify(ctrl, adapter->rx_obj.cq.id, true, 0);
 
-	be_link_status_update(adapter);
+	status = be_cmd_link_status_query(ctrl, &link_up);
+	if (status)
+		return status;
+	be_link_status_update(adapter, link_up);
 
 	schedule_delayed_work(&adapter->work, msecs_to_jiffies(100));
 	return 0;
@@ -1617,7 +1615,7 @@ static int be_close(struct net_device *netdev)
 
 	netif_stop_queue(netdev);
 	netif_carrier_off(netdev);
-	adapter->link.speed = PHY_LINK_SPEED_ZERO;
+	adapter->link_up = false;
 
 	be_intr_set(ctrl, false);
 
@@ -1808,6 +1806,9 @@ static int be_ctrl_init(struct be_adapter *adapter)
 	spin_lock_init(&ctrl->mcc_lock);
 	spin_lock_init(&ctrl->mcc_cq_lock);
 
+	ctrl->async_cb = be_link_status_update;
+	ctrl->adapter_ctxt = adapter;
+
 	val = ioread32(ctrl->pcicfg + PCICFG_MEMBAR_CTRL_INT_CTRL_OFFSET);
 	ctrl->pci_func = (val >> MEMBAR_CTRL_INT_CTRL_PFUNC_SHIFT) &
 					MEMBAR_CTRL_INT_CTRL_PFUNC_MASK;
-- 
cgit v1.2.3-59-g8ed1b


From 25502bda07b4cd4f1d9942cca2df446c4a0f167c Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dbaryshkov@gmail.com>
Date: Thu, 18 Jun 2009 04:16:46 +0000
Subject: ieee802154: use standard routine for printing dumps

Use print_hex_dump_bytes instead of self-written dumping function
for outputting packet dumps.

Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ieee802154/af_ieee802154.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/net/ieee802154/af_ieee802154.c b/net/ieee802154/af_ieee802154.c
index 882a927cefae..3bb6bdb1dac1 100644
--- a/net/ieee802154/af_ieee802154.c
+++ b/net/ieee802154/af_ieee802154.c
@@ -39,14 +39,6 @@
 
 #include "af802154.h"
 
-#define DBG_DUMP(data, len) { \
-	int i; \
-	pr_debug("function: %s: data: len %d:\n", __func__, len); \
-	for (i = 0; i < len; i++) {\
-		pr_debug("%02x: %02x\n", i, (data)[i]); \
-	} \
-}
-
 /*
  * Utility function for families
  */
@@ -302,10 +294,12 @@ static struct net_proto_family ieee802154_family_ops = {
 static int ieee802154_rcv(struct sk_buff *skb, struct net_device *dev,
 	struct packet_type *pt, struct net_device *orig_dev)
 {
-	DBG_DUMP(skb->data, skb->len);
 	if (!netif_running(dev))
 		return -ENODEV;
 	pr_debug("got frame, type %d, dev %p\n", dev->type, dev);
+#ifdef DEBUG
+	print_hex_dump_bytes("ieee802154_rcv ", DUMP_PREFIX_NONE, skb->data, skb->len);
+#endif
 
 	if (!net_eq(dev_net(dev), &init_net))
 		goto drop;
-- 
cgit v1.2.3-59-g8ed1b


From a060330e261cf71f5b3dd2f85bf3eeb9dba61a2e Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dbaryshkov@gmail.com>
Date: Thu, 18 Jun 2009 04:16:47 +0000
Subject: MAINTAINERS: fix address of IEEE 802.15.4 git tree

IEEE 802.15.4 git tree was moved from my private area to shared one.
Fix address accordingly.

Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 2cb7566904b1..7aa3081a8968 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2863,7 +2863,7 @@ P:	Sergey Lapin
 M:	slapin@ossfans.org
 L:	linux-zigbee-devel@lists.sourceforge.net
 W:	http://apps.sourceforge.net/trac/linux-zigbee
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/lumag/lowpan.git
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/lowpan/lowpan.git
 S:	Maintained
 F:	net/ieee802154/
 F:	drivers/ieee801254/
-- 
cgit v1.2.3-59-g8ed1b


From 68924920cb457ac44b14ca38343954bdcee046fc Mon Sep 17 00:00:00 2001
From: Jonas Sjöquist <jonas.sjoquist@ericsson.com>
Date: Thu, 18 Jun 2009 01:50:52 +0000
Subject: cdc_ether: additional PID's to the whitelist
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch adds five PID's to the whitelist set of devices.

Devices added to the whitelist:

Dell Wireless 5530 HSPA
Ericsson Mobile Broadband Module variants (F3507g, F3607gw and F3307)
Toshiba F3507g

Signed-off-by: Jonas Sjöquist <jonas.sjoquist@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/usb/Kconfig     |  4 +++-
 drivers/net/usb/cdc_ether.c | 25 +++++++++++++++++++++++++
 2 files changed, 28 insertions(+), 1 deletion(-)

diff --git a/drivers/net/usb/Kconfig b/drivers/net/usb/Kconfig
index 3717569828bf..a906d3998131 100644
--- a/drivers/net/usb/Kconfig
+++ b/drivers/net/usb/Kconfig
@@ -169,10 +169,12 @@ config USB_NET_CDCETHER
 	  The Linux-USB CDC Ethernet Gadget driver is an open implementation.
  	  This driver should work with at least the following devices:
 
+	    * Dell Wireless 5530 HSPA
  	    * Ericsson PipeRider (all variants)
+	    * Ericsson Mobile Broadband Module (all variants)
  	    * Motorola (DM100 and SB4100)
  	    * Broadcom Cable Modem (reference design)
- 	    * Toshiba PCX1100U
+	    * Toshiba (PCX1100U and F3507g)
 	    * ...
 
 	  This driver creates an interface named "ethX", where X depends on
diff --git a/drivers/net/usb/cdc_ether.c b/drivers/net/usb/cdc_ether.c
index 01fd528306ec..4a6aff579403 100644
--- a/drivers/net/usb/cdc_ether.c
+++ b/drivers/net/usb/cdc_ether.c
@@ -533,6 +533,31 @@ static const struct usb_device_id	products [] = {
 	USB_DEVICE_AND_INTERFACE_INFO(0x0bdb, 0x1900, USB_CLASS_COMM,
 			USB_CDC_SUBCLASS_MDLM, USB_CDC_PROTO_NONE),
 	.driver_info = (unsigned long) &cdc_info,
+}, {
+	/* Ericsson F3507g ver. 2 */
+	USB_DEVICE_AND_INTERFACE_INFO(0x0bdb, 0x1902, USB_CLASS_COMM,
+			USB_CDC_SUBCLASS_MDLM, USB_CDC_PROTO_NONE),
+	.driver_info = (unsigned long) &cdc_info,
+}, {
+	/* Ericsson F3607gw */
+	USB_DEVICE_AND_INTERFACE_INFO(0x0bdb, 0x1904, USB_CLASS_COMM,
+			USB_CDC_SUBCLASS_MDLM, USB_CDC_PROTO_NONE),
+	.driver_info = (unsigned long) &cdc_info,
+}, {
+	/* Ericsson F3307 */
+	USB_DEVICE_AND_INTERFACE_INFO(0x0bdb, 0x1906, USB_CLASS_COMM,
+			USB_CDC_SUBCLASS_MDLM, USB_CDC_PROTO_NONE),
+	.driver_info = (unsigned long) &cdc_info,
+}, {
+	/* Toshiba F3507g */
+	USB_DEVICE_AND_INTERFACE_INFO(0x0930, 0x130b, USB_CLASS_COMM,
+			USB_CDC_SUBCLASS_MDLM, USB_CDC_PROTO_NONE),
+	.driver_info = (unsigned long) &cdc_info,
+}, {
+	/* Dell F3507g */
+	USB_DEVICE_AND_INTERFACE_INFO(0x413c, 0x8147, USB_CLASS_COMM,
+			USB_CDC_SUBCLASS_MDLM, USB_CDC_PROTO_NONE),
+	.driver_info = (unsigned long) &cdc_info,
 },
 	{ },		// END
 };
-- 
cgit v1.2.3-59-g8ed1b


From 679e8a0f0ae3333e94b1d374d07775fce9066025 Mon Sep 17 00:00:00 2001
From: Andy Gospodarek <andy@greyhouse.net>
Date: Thu, 18 Jun 2009 11:57:37 +0000
Subject: e1000e: stop unnecessary polling when using msi-x

The last hunk of this commit:

    commit 12d04a3c12b420f23398b4d650127642469a60a6
    Author: Alexander Duyck <alexander.h.duyck@intel.com>
    Date:   Wed Mar 25 22:05:03 2009 +0000

        e1000e: commonize tx cleanup routine to match e1000 & igb

changed the logic for determining if we should call napi_complete or
not at then end of a napi poll.

If the NIC is using MSI-X with no work to do in ->poll, net_rx_action
can just spin indefinitely on older kernels and for 2 jiffies on newer
kernels since napi_complete is never called and budget isn't
decremented.

Discovered and verified while testing driver backport to an older
kernel.

Signed-off-by: Andy Gospodarek <andy@greyhouse.net>
Acked-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/e1000e/netdev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c
index 677f60490f67..679885a122b4 100644
--- a/drivers/net/e1000e/netdev.c
+++ b/drivers/net/e1000e/netdev.c
@@ -1997,7 +1997,7 @@ static int e1000_clean(struct napi_struct *napi, int budget)
 	struct e1000_adapter *adapter = container_of(napi, struct e1000_adapter, napi);
 	struct e1000_hw *hw = &adapter->hw;
 	struct net_device *poll_dev = adapter->netdev;
-	int tx_cleaned = 0, work_done = 0;
+	int tx_cleaned = 1, work_done = 0;
 
 	adapter = netdev_priv(poll_dev);
 
-- 
cgit v1.2.3-59-g8ed1b


From 40c27eeac42600b21d483087ff3885b31e6857c9 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Thu, 18 Jun 2009 03:49:49 +0000
Subject: r8169: remove unused variable

all references got removed by 865c652d6be9929927cabdc54b137b7541eb6612
(r8169: remove non-napi code).

Signed-off-by: Florian Westphal <fwestphal@astaro.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/r8169.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c
index 4e22462684c9..4b53b58d75fc 100644
--- a/drivers/net/r8169.c
+++ b/drivers/net/r8169.c
@@ -51,9 +51,6 @@
 #define TX_BUFFS_AVAIL(tp) \
 	(tp->dirty_tx + NUM_TX_DESC - tp->cur_tx - 1)
 
-/* Maximum events (Rx packets, etc.) to handle at each interrupt. */
-static const int max_interrupt_work = 20;
-
 /* Maximum number of multicast addresses to filter (vs. Rx-all-multicast).
    The RTL chips use a 64 element hash table based on the Ethernet CRC. */
 static const int multicast_filter_limit = 32;
-- 
cgit v1.2.3-59-g8ed1b


From 6877f54e6a3326c99aaf84b7bff6a3019da0b847 Mon Sep 17 00:00:00 2001
From: Prabhanjan Sarnaik <sarnaik@marvell.com>
Date: Thu, 18 Jun 2009 11:35:02 +0000
Subject: mv643xx_eth: fix unicast filter programming in promiscuous mode

The Unicast Promiscious Mode (UPM) bit in the mv643xx_eth port
configuration register doesn't do exactly what its name would suggest:
setting this bit merely enables reception of all unicast frames with a
destination address that differs from our local MAC address in bits
[47:4].  In particular, it doesn't have any effect on unicast frames
with a destination address that matches our MAC address in bits [47:4]
-- these will still be tested against the 16-entry unicast address
filter table.

Therefore, if the interface is set to promiscuous mode, just setting
the unicast promiscuous bit isn't enough -- we need to set all filter
bits in the unicast filter table to 1 as well.

Reported-by: Sachin Sanap <ssanap@marvell.com>
Signed-off-by: Prabhanjan Sarnaik <sarnaik@marvell.com>
Tested-by: Siddarth Gore <gores@marvell.com>
Tested-by: Mahavir Jain <mjain@marvell.com>
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Cc: stable@kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mv643xx_eth.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c
index 745ae8b4a2e8..0f32db3e92ad 100644
--- a/drivers/net/mv643xx_eth.c
+++ b/drivers/net/mv643xx_eth.c
@@ -1750,12 +1750,12 @@ static void mv643xx_eth_program_unicast_filter(struct net_device *dev)
 
 	uc_addr_set(mp, dev->dev_addr);
 
-	port_config = rdlp(mp, PORT_CONFIG);
+	port_config = rdlp(mp, PORT_CONFIG) & ~UNICAST_PROMISCUOUS_MODE;
+
 	nibbles = uc_addr_filter_mask(dev);
 	if (!nibbles) {
 		port_config |= UNICAST_PROMISCUOUS_MODE;
-		wrlp(mp, PORT_CONFIG, port_config);
-		return;
+		nibbles = 0xffff;
 	}
 
 	for (i = 0; i < 16; i += 4) {
@@ -1776,7 +1776,6 @@ static void mv643xx_eth_program_unicast_filter(struct net_device *dev)
 		wrl(mp, off, v);
 	}
 
-	port_config &= ~UNICAST_PROMISCUOUS_MODE;
 	wrlp(mp, PORT_CONFIG, port_config);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 9844813f226f6d07e1544e915529cb88f4fcb868 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Sun, 14 Jun 2009 02:00:02 -0400
Subject: asm-generic: uaccess: add missing access_ok() check to strnlen_user()

The strnlen_user() function was missing a access_ok() check on the pointer
given.  We've had cases on Blackfin systems where test programs caused
kernel crashes here because userspace passed up a NULL/-1 pointer and the
kernel gladly attempted to run strlen() on it.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 include/asm-generic/uaccess.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/asm-generic/uaccess.h b/include/asm-generic/uaccess.h
index 6d8cab22e294..5dd511b62ce9 100644
--- a/include/asm-generic/uaccess.h
+++ b/include/asm-generic/uaccess.h
@@ -291,6 +291,8 @@ strncpy_from_user(char *dst, const char __user *src, long count)
 #ifndef strnlen_user
 static inline long strnlen_user(const char __user *src, long n)
 {
+	if (!access_ok(VERIFY_READ, src, 1))
+		return 0;
 	return strlen((void * __force)src) + 1;
 }
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From a9ede5b355aabd667ed690f858c925a23927027b Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Sun, 14 Jun 2009 02:00:03 -0400
Subject: asm-generic: uaccess: fix up local access_ok() usage

There's no reason that I can see to use the short __access_ok() form
directly when the access_ok() is clearer in intent and for most people,
expands to the same C code (i.e. always specify the first field -- access
type).  Not all no-mmu systems lack memory protection, so the read/write
could feasibly be checked.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 include/asm-generic/uaccess.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/include/asm-generic/uaccess.h b/include/asm-generic/uaccess.h
index 5dd511b62ce9..b218b8513d04 100644
--- a/include/asm-generic/uaccess.h
+++ b/include/asm-generic/uaccess.h
@@ -163,7 +163,7 @@ static inline __must_check long __copy_to_user(void __user *to,
 #define put_user(x, ptr)					\
 ({								\
 	might_sleep();						\
-	__access_ok(ptr, sizeof (*ptr)) ?			\
+	access_ok(VERIFY_WRITE, ptr, sizeof(*ptr)) ?		\
 		__put_user(x, ptr) :				\
 		-EFAULT;					\
 })
@@ -219,7 +219,7 @@ extern int __put_user_bad(void) __attribute__((noreturn));
 #define get_user(x, ptr)					\
 ({								\
 	might_sleep();						\
-	__access_ok(ptr, sizeof (*ptr)) ?			\
+	access_ok(VERIFY_READ, ptr, sizeof(*ptr)) ?		\
 		__get_user(x, ptr) :				\
 		-EFAULT;					\
 })
@@ -244,7 +244,7 @@ static inline long copy_from_user(void *to,
 		const void __user * from, unsigned long n)
 {
 	might_sleep();
-	if (__access_ok(from, n))
+	if (access_ok(VERIFY_READ, from, n))
 		return __copy_from_user(to, from, n);
 	else
 		return n;
@@ -254,7 +254,7 @@ static inline long copy_to_user(void __user *to,
 		const void *from, unsigned long n)
 {
 	might_sleep();
-	if (__access_ok(to, n))
+	if (access_ok(VERIFY_WRITE, to, n))
 		return __copy_to_user(to, from, n);
 	else
 		return n;
@@ -278,7 +278,7 @@ __strncpy_from_user(char *dst, const char __user *src, long count)
 static inline long
 strncpy_from_user(char *dst, const char __user *src, long count)
 {
-	if (!__access_ok(src, 1))
+	if (!access_ok(VERIFY_READ, src, 1))
 		return -EFAULT;
 	return __strncpy_from_user(dst, src, count);
 }
@@ -318,7 +318,7 @@ static inline __must_check unsigned long
 clear_user(void __user *to, unsigned long n)
 {
 	might_sleep();
-	if (!__access_ok(to, n))
+	if (!access_ok(VERIFY_WRITE, to, n))
 		return n;
 
 	return __clear_user(to, n);
-- 
cgit v1.2.3-59-g8ed1b


From 804387a1af87f66a4b93eee230ba98f8b906b088 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Sun, 14 Jun 2009 22:38:11 +0200
Subject: asm-generic: drop HARDIRQ_BITS definition from hardirq.h

Architechtures normally don't need to set a HARDIRQ_BITS
unless they have hardcoded a specific value in assembly.
This drops the definition from asm-generic/hardirq.h, which
results in linux/hardirq.h setting its default of 10.

Both the old default of 8 and the linux/hardirq.h default
of 10 are sufficient because they only limit the number
of nested hardirqs, and we normally run out of stack space
much earlier than exceeding 256 or even 1024 nested interrupts.

Reported-by: Mike Frysinger <vapier@gentoo.org>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 include/asm-generic/hardirq.h | 13 -------------
 1 file changed, 13 deletions(-)

diff --git a/include/asm-generic/hardirq.h b/include/asm-generic/hardirq.h
index 3d5d2c906ab3..23bb4dad4962 100644
--- a/include/asm-generic/hardirq.h
+++ b/include/asm-generic/hardirq.h
@@ -11,19 +11,6 @@ typedef struct {
 
 #include <linux/irq_cpustat.h>	/* Standard mappings for irq_cpustat_t above */
 
-#ifndef HARDIRQ_BITS
-#define HARDIRQ_BITS	8
-#endif
-
-/*
- * The hardirq mask has to be large enough to have
- * space for potentially all IRQ sources in the system
- * nesting on a single CPU:
- */
-#if (1 << HARDIRQ_BITS) < NR_IRQS
-# error HARDIRQ_BITS is too low!
-#endif
-
 #ifndef ack_bad_irq
 static inline void ack_bad_irq(unsigned int irq)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 1527aab617af87f2d2121fa3ff5aa79d96e7ad2f Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Sun, 14 Jun 2009 22:46:16 +0200
Subject: asm-generic: list Arnd as asm-generic maintainer

I've modified about half the code in include/asm-generic now, and
people start sending me patches for it, so I should probably take
the formal responsibility for it.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 MAINTAINERS | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 035df9d26609..acf11052cb7d 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2453,6 +2453,14 @@ F:	drivers/net/wan/pc300too.c
 F:	drivers/net/wan/pci200syn.c
 F:	drivers/net/wan/wanxl*
 
+GENERIC INCLUDE/ASM HEADER FILES
+P:	Arnd Bergmann
+M:	arnd@arndb.de
+L:	linux-arch@vger.kernel.org
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic.git
+S:	Maintained
+F:	include/asm-generic
+
 GFS2 FILE SYSTEM
 P:	Steven Whitehouse
 M:	swhiteho@redhat.com
-- 
cgit v1.2.3-59-g8ed1b


From fcec9bf12442d0cd50d6cee125d168cfc3f37c5e Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Fri, 19 Jun 2009 10:30:29 +0200
Subject: asm-generic: hook up new system calls

sys_rt_tgsigqueueinfo and sys_perf_counter_open
have been added in 2.6.31, so hook them up in the
generic unistd.h file.

Since the file is now in the mainline kernel, we
are no longer reordering the numbers but just add
system calls at the end.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 include/asm-generic/unistd.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/include/asm-generic/unistd.h b/include/asm-generic/unistd.h
index 5b34b6233d6d..1125e5a1ee5d 100644
--- a/include/asm-generic/unistd.h
+++ b/include/asm-generic/unistd.h
@@ -618,8 +618,13 @@ __SYSCALL(__NR_migrate_pages, sys_migrate_pages)
 __SYSCALL(__NR_move_pages, sys_move_pages)
 #endif
 
+#define __NR_rt_tgsigqueueinfo 240
+__SYSCALL(__NR_rt_tgsigqueueinfo, sys_rt_tgsigqueueinfo)
+#define __NR_perf_counter_open 241
+__SYSCALL(__NR_perf_counter_open, sys_perf_counter_open)
+
 #undef __NR_syscalls
-#define __NR_syscalls 240
+#define __NR_syscalls 242
 
 /*
  * All syscalls below here should go away really,
-- 
cgit v1.2.3-59-g8ed1b


From 32a9ff9cc55b42543a84bca90b070df92af72671 Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Fri, 19 Jun 2009 10:41:19 +0200
Subject: lib/checksum.c: fix endianess bug

The new generic checksum code has a small dependency on endianess and
worked only on big-endian systems. I could not find a nice efficient
way to express this, so I added an #ifdef. Using
'result += le16_to_cpu(*buff);' would have worked as well, but
would be slightly less efficient on big-endian systems and IMHO
would not be clearer.

Also fix a bug that prevents this from working on 64-bit machines.
If you have a 64-bit CPU and want to use the generic checksum
code, you should probably do some more optimizations anyway, but
at least the code should not break.

Reported-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 lib/checksum.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/lib/checksum.c b/lib/checksum.c
index 12e5a1c91cda..b2e2fd468461 100644
--- a/lib/checksum.c
+++ b/lib/checksum.c
@@ -55,7 +55,11 @@ static unsigned int do_csum(const unsigned char *buff, int len)
 		goto out;
 	odd = 1 & (unsigned long) buff;
 	if (odd) {
+#ifdef __LITTLE_ENDIAN
 		result = *buff;
+#else
+		result += (*buff << 8);
+#endif
 		len--;
 		buff++;
 	}
@@ -71,7 +75,7 @@ static unsigned int do_csum(const unsigned char *buff, int len)
 		if (count) {
 			unsigned long carry = 0;
 			do {
-				unsigned long w = *(unsigned long *) buff;
+				unsigned long w = *(unsigned int *) buff;
 				count--;
 				buff += 4;
 				result += carry;
@@ -87,7 +91,11 @@ static unsigned int do_csum(const unsigned char *buff, int len)
 		}
 	}
 	if (len & 1)
+#ifdef __LITTLE_ENDIAN
+		result += *buff;
+#else
 		result += (*buff << 8);
+#endif
 	result = from32to16(result);
 	if (odd)
 		result = ((result >> 8) & 0xff) | ((result & 0xff) << 8);
-- 
cgit v1.2.3-59-g8ed1b


From c3c2174031def62eeb8433dee8ef7e8d9493b1cf Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 19 Jun 2009 11:28:48 +0100
Subject: [ARM] 5555/1: RealView: Include asm/smp_twd.h in realview-pbx.c

This header file is needed for twd_base.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-realview/realview_pbx.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm/mach-realview/realview_pbx.c b/arch/arm/mach-realview/realview_pbx.c
index 1fe294d0bf9d..ede2a57240a3 100644
--- a/arch/arm/mach-realview/realview_pbx.c
+++ b/arch/arm/mach-realview/realview_pbx.c
@@ -27,6 +27,7 @@
 #include <asm/irq.h>
 #include <asm/leds.h>
 #include <asm/mach-types.h>
+#include <asm/smp_twd.h>
 #include <asm/hardware/gic.h>
 #include <asm/hardware/cache-l2x0.h>
 
-- 
cgit v1.2.3-59-g8ed1b


From 41184f6a5ef0d88529904d54f06f88b67fb76f4a Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 19 Jun 2009 11:30:12 +0100
Subject: [ARM] 5556/1: Fix the irq_desc.cpu references

The cpu member of struct irq_desc was recently renamed to node. The
patch renames the ARM references to the old member.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/common/gic.c | 2 +-
 arch/arm/kernel/irq.c | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
index 664c7b8b1ba8..337741f734ac 100644
--- a/arch/arm/common/gic.c
+++ b/arch/arm/common/gic.c
@@ -117,7 +117,7 @@ static int gic_set_cpu(unsigned int irq, const struct cpumask *mask_val)
 	u32 val;
 
 	spin_lock(&irq_controller_lock);
-	irq_desc[irq].cpu = cpu;
+	irq_desc[irq].node = cpu;
 	val = readl(reg) & ~(0xff << shift);
 	val |= 1 << (cpu + shift);
 	writel(val, reg);
diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c
index 6874c7dca75a..096f600dc8d8 100644
--- a/arch/arm/kernel/irq.c
+++ b/arch/arm/kernel/irq.c
@@ -167,7 +167,7 @@ void __init init_IRQ(void)
 
 #ifdef CONFIG_SMP
 	cpumask_setall(bad_irq_desc.affinity);
-	bad_irq_desc.cpu = smp_processor_id();
+	bad_irq_desc.node = smp_processor_id();
 #endif
 	init_arch_irq();
 }
@@ -176,7 +176,7 @@ void __init init_IRQ(void)
 
 static void route_irq(struct irq_desc *desc, unsigned int irq, unsigned int cpu)
 {
-	pr_debug("IRQ%u: moving from cpu%u to cpu%u\n", irq, desc->cpu, cpu);
+	pr_debug("IRQ%u: moving from cpu%u to cpu%u\n", irq, desc->node, cpu);
 
 	spin_lock_irq(&desc->lock);
 	desc->chip->set_affinity(irq, cpumask_of(cpu));
@@ -195,7 +195,7 @@ void migrate_irqs(void)
 	for (i = 0; i < NR_IRQS; i++) {
 		struct irq_desc *desc = irq_desc + i;
 
-		if (desc->cpu == cpu) {
+		if (desc->node == cpu) {
 			unsigned int newcpu = cpumask_any_and(desc->affinity,
 							      cpu_online_mask);
 			if (newcpu >= nr_cpu_ids) {
-- 
cgit v1.2.3-59-g8ed1b


From c2860d43f5dfab599fc1308ab61b1d3e30801ceb Mon Sep 17 00:00:00 2001
From: "George G. Davis" <gdavis@mvista.com>
Date: Thu, 4 Jun 2009 17:16:04 +0100
Subject: [ARM] 5540/1: 32-bit Thumb-2 {ld,st}{m,rd} alignment fault fixup
 support

From: Min Zhang <mzhang@mvista.com>

Add alignment fault fixup support for 32-bit Thumb-2 LDM, LDRD, POP,
PUSH, STM and STRD instructions.  Alignment fault fixup support for
the remaining 32-bit Thumb-2 load/store instruction cases is not
included since ARMv6 and later processors include hardware support
for loads and stores of unaligned words and halfwords.

Signed-off-by: Min Zhang <mzhang@mvista.com>
Signed-off-by: George G. Davis <gdavis@mvista.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/alignment.c | 139 +++++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 119 insertions(+), 20 deletions(-)

diff --git a/arch/arm/mm/alignment.c b/arch/arm/mm/alignment.c
index 3a398befed41..03cd27d917b9 100644
--- a/arch/arm/mm/alignment.c
+++ b/arch/arm/mm/alignment.c
@@ -62,6 +62,12 @@
 #define SHIFT_ASR	0x40
 #define SHIFT_RORRRX	0x60
 
+#define BAD_INSTR 	0xdeadc0de
+
+/* Thumb-2 32 bit format per ARMv7 DDI0406A A6.3, either f800h,e800h,f800h */
+#define IS_T32(hi16) \
+	(((hi16) & 0xe000) == 0xe000 && ((hi16) & 0x1800))
+
 static unsigned long ai_user;
 static unsigned long ai_sys;
 static unsigned long ai_skipped;
@@ -332,38 +338,48 @@ do_alignment_ldrdstrd(unsigned long addr, unsigned long instr,
 		      struct pt_regs *regs)
 {
 	unsigned int rd = RD_BITS(instr);
-
-	if (((rd & 1) == 1) || (rd == 14))
+	unsigned int rd2;
+	int load;
+
+	if ((instr & 0xfe000000) == 0xe8000000) {
+		/* ARMv7 Thumb-2 32-bit LDRD/STRD */
+		rd2 = (instr >> 8) & 0xf;
+		load = !!(LDST_L_BIT(instr));
+	} else if (((rd & 1) == 1) || (rd == 14))
 		goto bad;
+	else {
+		load = ((instr & 0xf0) == 0xd0);
+		rd2 = rd + 1;
+	}
 
 	ai_dword += 1;
 
 	if (user_mode(regs))
 		goto user;
 
-	if ((instr & 0xf0) == 0xd0) {
+	if (load) {
 		unsigned long val;
 		get32_unaligned_check(val, addr);
 		regs->uregs[rd] = val;
 		get32_unaligned_check(val, addr + 4);
-		regs->uregs[rd + 1] = val;
+		regs->uregs[rd2] = val;
 	} else {
 		put32_unaligned_check(regs->uregs[rd], addr);
-		put32_unaligned_check(regs->uregs[rd + 1], addr + 4);
+		put32_unaligned_check(regs->uregs[rd2], addr + 4);
 	}
 
 	return TYPE_LDST;
 
  user:
-	if ((instr & 0xf0) == 0xd0) {
+	if (load) {
 		unsigned long val;
 		get32t_unaligned_check(val, addr);
 		regs->uregs[rd] = val;
 		get32t_unaligned_check(val, addr + 4);
-		regs->uregs[rd + 1] = val;
+		regs->uregs[rd2] = val;
 	} else {
 		put32t_unaligned_check(regs->uregs[rd], addr);
-		put32t_unaligned_check(regs->uregs[rd + 1], addr + 4);
+		put32t_unaligned_check(regs->uregs[rd2], addr + 4);
 	}
 
 	return TYPE_LDST;
@@ -616,8 +632,72 @@ thumb2arm(u16 tinstr)
 		/* Else fall through for illegal instruction case */
 
 	default:
-		return 0xdeadc0de;
+		return BAD_INSTR;
+	}
+}
+
+/*
+ * Convert Thumb-2 32 bit LDM, STM, LDRD, STRD to equivalent instruction
+ * handlable by ARM alignment handler, also find the corresponding handler,
+ * so that we can reuse ARM userland alignment fault fixups for Thumb.
+ *
+ * @pinstr: original Thumb-2 instruction; returns new handlable instruction
+ * @regs: register context.
+ * @poffset: return offset from faulted addr for later writeback
+ *
+ * NOTES:
+ * 1. Comments below refer to ARMv7 DDI0406A Thumb Instruction sections.
+ * 2. Register name Rt from ARMv7 is same as Rd from ARMv6 (Rd is Rt)
+ */
+static void *
+do_alignment_t32_to_handler(unsigned long *pinstr, struct pt_regs *regs,
+			    union offset_union *poffset)
+{
+	unsigned long instr = *pinstr;
+	u16 tinst1 = (instr >> 16) & 0xffff;
+	u16 tinst2 = instr & 0xffff;
+	poffset->un = 0;
+
+	switch (tinst1 & 0xffe0) {
+	/* A6.3.5 Load/Store multiple */
+	case 0xe880:		/* STM/STMIA/STMEA,LDM/LDMIA, PUSH/POP T2 */
+	case 0xe8a0:		/* ...above writeback version */
+	case 0xe900:		/* STMDB/STMFD, LDMDB/LDMEA */
+	case 0xe920:		/* ...above writeback version */
+		/* no need offset decision since handler calculates it */
+		return do_alignment_ldmstm;
+
+	case 0xf840:		/* POP/PUSH T3 (single register) */
+		if (RN_BITS(instr) == 13 && (tinst2 & 0x09ff) == 0x0904) {
+			u32 L = !!(LDST_L_BIT(instr));
+			const u32 subset[2] = {
+				0xe92d0000,	/* STMDB sp!,{registers} */
+				0xe8bd0000,	/* LDMIA sp!,{registers} */
+			};
+			*pinstr = subset[L] | (1<<RD_BITS(instr));
+			return do_alignment_ldmstm;
+		}
+		/* Else fall through for illegal instruction case */
+		break;
+
+	/* A6.3.6 Load/store double, STRD/LDRD(immed, lit, reg) */
+	case 0xe860:
+	case 0xe960:
+	case 0xe8e0:
+	case 0xe9e0:
+		poffset->un = (tinst2 & 0xff) << 2;
+	case 0xe940:
+	case 0xe9c0:
+		return do_alignment_ldrdstrd;
+
+	/*
+	 * No need to handle load/store instructions up to word size
+	 * since ARMv6 and later CPUs can perform unaligned accesses.
+	 */
+	default:
+		break;
 	}
+	return NULL;
 }
 
 static int
@@ -630,6 +710,8 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 	mm_segment_t fs;
 	unsigned int fault;
 	u16 tinstr = 0;
+	int isize = 4;
+	int thumb2_32b = 0;
 
 	instrptr = instruction_pointer(regs);
 
@@ -637,8 +719,19 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 	set_fs(KERNEL_DS);
 	if (thumb_mode(regs)) {
 		fault = __get_user(tinstr, (u16 *)(instrptr & ~1));
-		if (!(fault))
-			instr = thumb2arm(tinstr);
+		if (!fault) {
+			if (cpu_architecture() >= CPU_ARCH_ARMv7 &&
+			    IS_T32(tinstr)) {
+				/* Thumb-2 32-bit */
+				u16 tinst2 = 0;
+				fault = __get_user(tinst2, (u16 *)(instrptr+2));
+				instr = (tinstr << 16) | tinst2;
+				thumb2_32b = 1;
+			} else {
+				isize = 2;
+				instr = thumb2arm(tinstr);
+			}
+		}
 	} else
 		fault = __get_user(instr, (u32 *)instrptr);
 	set_fs(fs);
@@ -655,7 +748,7 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 
  fixup:
 
-	regs->ARM_pc += thumb_mode(regs) ? 2 : 4;
+	regs->ARM_pc += isize;
 
 	switch (CODING_BITS(instr)) {
 	case 0x00000000:	/* 3.13.4 load/store instruction extensions */
@@ -714,18 +807,25 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 		handler = do_alignment_ldrstr;
 		break;
 
-	case 0x08000000:	/* ldm or stm */
-		handler = do_alignment_ldmstm;
+	case 0x08000000:	/* ldm or stm, or thumb-2 32bit instruction */
+		if (thumb2_32b)
+			handler = do_alignment_t32_to_handler(&instr, regs, &offset);
+		else
+			handler = do_alignment_ldmstm;
 		break;
 
 	default:
 		goto bad;
 	}
 
+	if (!handler)
+		goto bad;
 	type = handler(addr, instr, regs);
 
-	if (type == TYPE_ERROR || type == TYPE_FAULT)
+	if (type == TYPE_ERROR || type == TYPE_FAULT) {
+		regs->ARM_pc -= isize;
 		goto bad_or_fault;
+	}
 
 	if (type == TYPE_LDST)
 		do_alignment_finish_ldst(addr, instr, regs, offset);
@@ -735,7 +835,6 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
  bad_or_fault:
 	if (type == TYPE_ERROR)
 		goto bad;
-	regs->ARM_pc -= thumb_mode(regs) ? 2 : 4;
 	/*
 	 * We got a fault - fix it up, or die.
 	 */
@@ -751,8 +850,8 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 	 */
 	printk(KERN_ERR "Alignment trap: not handling instruction "
 		"%0*lx at [<%08lx>]\n",
-		thumb_mode(regs) ? 4 : 8,
-		thumb_mode(regs) ? tinstr : instr, instrptr);
+		isize << 1,
+		isize == 2 ? tinstr : instr, instrptr);
 	ai_skipped += 1;
 	return 1;
 
@@ -763,8 +862,8 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 		printk("Alignment trap: %s (%d) PC=0x%08lx Instr=0x%0*lx "
 		       "Address=0x%08lx FSR 0x%03x\n", current->comm,
 			task_pid_nr(current), instrptr,
-		        thumb_mode(regs) ? 4 : 8,
-		        thumb_mode(regs) ? tinstr : instr,
+			isize << 1,
+			isize == 2 ? tinstr : instr,
 		        addr, fsr);
 
 	if (ai_usermode & UM_FIXUP)
-- 
cgit v1.2.3-59-g8ed1b


From 7436127ce9042f95a10bb5423f726fd63a61934d Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 19 Jun 2009 16:39:29 +0100
Subject: [ARM] 5557/1: Discard some ARM.ex*.*exit.text sections when !HOTPLUG
 or !HOTPLUG_CPU

Not discarding these sections when hotplug isn't available prevents the
kernel from building.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/vmlinux.lds.S | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S
index 6c0779792546..4340bf3d2c84 100644
--- a/arch/arm/kernel/vmlinux.lds.S
+++ b/arch/arm/kernel/vmlinux.lds.S
@@ -84,6 +84,14 @@ SECTIONS
 		*(.exitcall.exit)
 		*(.ARM.exidx.exit.text)
 		*(.ARM.extab.exit.text)
+#ifndef CONFIG_HOTPLUG_CPU
+		*(.ARM.exidx.cpuexit.text)
+		*(.ARM.extab.cpuexit.text)
+#endif
+#ifndef CONFIG_HOTPLUG
+		*(.ARM.exidx.devexit.text)
+		*(.ARM.extab.devexit.text)
+#endif
 #ifndef CONFIG_MMU
 		*(.fixup)
 		*(__ex_table)
-- 
cgit v1.2.3-59-g8ed1b


From c894ed6956f126d60d888e8efc5fb3a595ba89ae Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 19 Jun 2009 16:42:11 +0100
Subject: [ARM] 5558/1: Add extra checks to ARM unwinder to avoid tracing
 corrupt stacks

There are situations where the unwinder goes beyond stack boundaries and
unwinds random data. This patch moves the stack boundaries check after
the unwind_exec_insn() call and adds an extra check for possible
infinite loops (like "mov pc, lr" with pc == lr).

The patch also fixes a bug in the unwind instructions interpreter. The
0xb0 instruction can only set PC to LR if this wasn't already set by
a previous instruction (this is used on exceptions taken while in kernel
mode where svc_entry is annotated with ".save {r0 - pc}").

Tested-by: Tony Lindgren <tony@atomide.com>

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/unwind.c | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)

diff --git a/arch/arm/kernel/unwind.c b/arch/arm/kernel/unwind.c
index 1dedc2c7ff49..dd56e11f339a 100644
--- a/arch/arm/kernel/unwind.c
+++ b/arch/arm/kernel/unwind.c
@@ -212,7 +212,8 @@ static int unwind_exec_insn(struct unwind_ctrl_block *ctrl)
 			ctrl->vrs[14] = *vsp++;
 		ctrl->vrs[SP] = (unsigned long)vsp;
 	} else if (insn == 0xb0) {
-		ctrl->vrs[PC] = ctrl->vrs[LR];
+		if (ctrl->vrs[PC] == 0)
+			ctrl->vrs[PC] = ctrl->vrs[LR];
 		/* no further processing */
 		ctrl->entries = 0;
 	} else if (insn == 0xb1) {
@@ -309,18 +310,20 @@ int unwind_frame(struct stackframe *frame)
 	}
 
 	while (ctrl.entries > 0) {
-		int urc;
-
-		if (ctrl.vrs[SP] < low || ctrl.vrs[SP] >= high)
-			return -URC_FAILURE;
-		urc = unwind_exec_insn(&ctrl);
+		int urc = unwind_exec_insn(&ctrl);
 		if (urc < 0)
 			return urc;
+		if (ctrl.vrs[SP] < low || ctrl.vrs[SP] >= high)
+			return -URC_FAILURE;
 	}
 
 	if (ctrl.vrs[PC] == 0)
 		ctrl.vrs[PC] = ctrl.vrs[LR];
 
+	/* check for infinite loop */
+	if (frame->pc == ctrl.vrs[PC])
+		return -URC_FAILURE;
+
 	frame->fp = ctrl.vrs[FP];
 	frame->sp = ctrl.vrs[SP];
 	frame->lr = ctrl.vrs[LR];
@@ -332,7 +335,6 @@ int unwind_frame(struct stackframe *frame)
 void unwind_backtrace(struct pt_regs *regs, struct task_struct *tsk)
 {
 	struct stackframe frame;
-	unsigned long high, low;
 	register unsigned long current_sp asm ("sp");
 
 	pr_debug("%s(regs = %p tsk = %p)\n", __func__, regs, tsk);
@@ -362,9 +364,6 @@ void unwind_backtrace(struct pt_regs *regs, struct task_struct *tsk)
 		frame.pc = thread_saved_pc(tsk);
 	}
 
-	low = frame.sp & ~(THREAD_SIZE - 1);
-	high = low + THREAD_SIZE;
-
 	while (1) {
 		int urc;
 		unsigned long where = frame.pc;
-- 
cgit v1.2.3-59-g8ed1b


From feb97c3644a560ffdf9a17c65b1df807b5b4432f Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 19 Jun 2009 16:43:08 +0100
Subject: [ARM] 5559/1: Limit the stack unwinding caused by a kthread exit

When a kthread function returns, it branches to do_exit(). However, the
unwinding information isn't valid anymore and any stack trace caused by
do_exit() may be incorrect. This patch adds a kernel_thread_exit()
function and annotated with '.cantunwind' so that the unwinder stops
when reaching it.

Tested-by: Tony Lindgren <tony@atomide.com>

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/process.c | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index 1585423699ee..56820cce91a4 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -352,6 +352,23 @@ asm(	".section .text\n"
 "	.size	kernel_thread_helper, . - kernel_thread_helper\n"
 "	.previous");
 
+#ifdef CONFIG_ARM_UNWIND
+extern void kernel_thread_exit(long code);
+asm(	".section .text\n"
+"	.align\n"
+"	.type	kernel_thread_exit, #function\n"
+"kernel_thread_exit:\n"
+"	.fnstart\n"
+"	.cantunwind\n"
+"	bl	do_exit\n"
+"	nop\n"
+"	.fnend\n"
+"	.size	kernel_thread_exit, . - kernel_thread_exit\n"
+"	.previous");
+#else
+#define kernel_thread_exit	do_exit
+#endif
+
 /*
  * Create a kernel thread.
  */
@@ -363,7 +380,7 @@ pid_t kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
 
 	regs.ARM_r1 = (unsigned long)arg;
 	regs.ARM_r2 = (unsigned long)fn;
-	regs.ARM_r3 = (unsigned long)do_exit;
+	regs.ARM_r3 = (unsigned long)kernel_thread_exit;
 	regs.ARM_pc = (unsigned long)kernel_thread_helper;
 	regs.ARM_cpsr = SVC_MODE | PSR_ENDSTATE;
 
-- 
cgit v1.2.3-59-g8ed1b


From 8d96e7960b6b520eb52be6e1eb7c794da5db9555 Mon Sep 17 00:00:00 2001
From: Zhu Yi <yi.zhu@intel.com>
Date: Mon, 15 Jun 2009 21:36:13 +0200
Subject: iwmc3200wifi: check for iwm_priv_init error

We need to check for iwm_priv_init() errors and do proper cleanups.
Otherwise we may fail to catch the create_singlethread_workqueue()
error which will cause a kernel oops when destroy_workqueue() later.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwmc3200wifi/iwm.h    |  1 +
 drivers/net/wireless/iwmc3200wifi/main.c   | 10 ++++++++++
 drivers/net/wireless/iwmc3200wifi/netdev.c | 19 ++++++++++++-------
 3 files changed, 23 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/iwmc3200wifi/iwm.h b/drivers/net/wireless/iwmc3200wifi/iwm.h
index 635c16ee6186..2237448e0427 100644
--- a/drivers/net/wireless/iwmc3200wifi/iwm.h
+++ b/drivers/net/wireless/iwmc3200wifi/iwm.h
@@ -317,6 +317,7 @@ void *iwm_if_alloc(int sizeof_bus, struct device *dev,
 void iwm_if_free(struct iwm_priv *iwm);
 int iwm_mode_to_nl80211_iftype(int mode);
 int iwm_priv_init(struct iwm_priv *iwm);
+void iwm_priv_deinit(struct iwm_priv *iwm);
 void iwm_reset(struct iwm_priv *iwm);
 void iwm_tx_credit_init_pools(struct iwm_priv *iwm,
 			      struct iwm_umac_notif_alive *alive);
diff --git a/drivers/net/wireless/iwmc3200wifi/main.c b/drivers/net/wireless/iwmc3200wifi/main.c
index 6a2640f16b6d..4d3c423d8ffc 100644
--- a/drivers/net/wireless/iwmc3200wifi/main.c
+++ b/drivers/net/wireless/iwmc3200wifi/main.c
@@ -219,6 +219,16 @@ int iwm_priv_init(struct iwm_priv *iwm)
 	return 0;
 }
 
+void iwm_priv_deinit(struct iwm_priv *iwm)
+{
+	int i;
+
+	for (i = 0; i < IWM_TX_QUEUES; i++)
+		destroy_workqueue(iwm->txq[i].wq);
+
+	destroy_workqueue(iwm->rx_wq);
+}
+
 /*
  * We reset all the structures, and we reset the UMAC.
  * After calling this routine, you're expected to reload
diff --git a/drivers/net/wireless/iwmc3200wifi/netdev.c b/drivers/net/wireless/iwmc3200wifi/netdev.c
index 68e2c3b6c7a1..88dd82649b47 100644
--- a/drivers/net/wireless/iwmc3200wifi/netdev.c
+++ b/drivers/net/wireless/iwmc3200wifi/netdev.c
@@ -114,14 +114,20 @@ void *iwm_if_alloc(int sizeof_bus, struct device *dev,
 	iwm = wdev_to_iwm(wdev);
 	iwm->bus_ops = if_ops;
 	iwm->wdev = wdev;
-	iwm_priv_init(iwm);
+
+	ret = iwm_priv_init(iwm);
+	if (ret) {
+		dev_err(dev, "failed to init iwm_priv\n");
+		goto out_wdev;
+	}
+
 	wdev->iftype = iwm_mode_to_nl80211_iftype(iwm->conf.mode);
 
 	ndev = alloc_netdev_mq(0, "wlan%d", ether_setup,
 			       IWM_TX_QUEUES);
 	if (!ndev) {
 		dev_err(dev, "no memory for network device instance\n");
-		goto out_wdev;
+		goto out_priv;
 	}
 
 	ndev->netdev_ops = &iwm_netdev_ops;
@@ -141,6 +147,9 @@ void *iwm_if_alloc(int sizeof_bus, struct device *dev,
  out_ndev:
 	free_netdev(ndev);
 
+ out_priv:
+	iwm_priv_deinit(iwm);
+
  out_wdev:
 	iwm_wdev_free(iwm);
 	return ERR_PTR(ret);
@@ -148,15 +157,11 @@ void *iwm_if_alloc(int sizeof_bus, struct device *dev,
 
 void iwm_if_free(struct iwm_priv *iwm)
 {
-	int i;
-
 	if (!iwm_to_ndev(iwm))
 		return;
 
 	unregister_netdev(iwm_to_ndev(iwm));
 	free_netdev(iwm_to_ndev(iwm));
 	iwm_wdev_free(iwm);
-	destroy_workqueue(iwm->rx_wq);
-	for (i = 0; i < IWM_TX_QUEUES; i++)
-		destroy_workqueue(iwm->txq[i].wq);
+	iwm_priv_deinit(iwm);
 }
-- 
cgit v1.2.3-59-g8ed1b


From d7e057dca3f1b76ff44dd16fefcd493a52614aad Mon Sep 17 00:00:00 2001
From: Zhu Yi <yi.zhu@intel.com>
Date: Mon, 15 Jun 2009 21:36:14 +0200
Subject: iwmc3200wifi: add iwm_if_add and iwm_if_remove

We used to do alloc_netdev and register_netdev at the same time in
iwm_if_alloc. But some bus related structures will only be initialized
after iwm_priv is allocated. This caused a race condition that the
netdev might be registered earlier. The patch adds iwm_if_add and
iwm_if_remove so that the bus layer could register the device after
all initialization is done.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwmc3200wifi/iwm.h    |  2 ++
 drivers/net/wireless/iwmc3200wifi/netdev.c | 32 +++++++++++++++++++-----------
 drivers/net/wireless/iwmc3200wifi/sdio.c   |  9 +++++++++
 3 files changed, 31 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/iwmc3200wifi/iwm.h b/drivers/net/wireless/iwmc3200wifi/iwm.h
index 2237448e0427..4aa0ad1932ff 100644
--- a/drivers/net/wireless/iwmc3200wifi/iwm.h
+++ b/drivers/net/wireless/iwmc3200wifi/iwm.h
@@ -315,6 +315,8 @@ extern const struct iw_handler_def iwm_iw_handler_def;
 void *iwm_if_alloc(int sizeof_bus, struct device *dev,
 		   struct iwm_if_ops *if_ops);
 void iwm_if_free(struct iwm_priv *iwm);
+int iwm_if_add(struct iwm_priv *iwm);
+void iwm_if_remove(struct iwm_priv *iwm);
 int iwm_mode_to_nl80211_iftype(int mode);
 int iwm_priv_init(struct iwm_priv *iwm);
 void iwm_priv_deinit(struct iwm_priv *iwm);
diff --git a/drivers/net/wireless/iwmc3200wifi/netdev.c b/drivers/net/wireless/iwmc3200wifi/netdev.c
index 88dd82649b47..aaa20c6885c8 100644
--- a/drivers/net/wireless/iwmc3200wifi/netdev.c
+++ b/drivers/net/wireless/iwmc3200wifi/netdev.c
@@ -123,8 +123,7 @@ void *iwm_if_alloc(int sizeof_bus, struct device *dev,
 
 	wdev->iftype = iwm_mode_to_nl80211_iftype(iwm->conf.mode);
 
-	ndev = alloc_netdev_mq(0, "wlan%d", ether_setup,
-			       IWM_TX_QUEUES);
+	ndev = alloc_netdev_mq(0, "wlan%d", ether_setup, IWM_TX_QUEUES);
 	if (!ndev) {
 		dev_err(dev, "no memory for network device instance\n");
 		goto out_priv;
@@ -134,19 +133,10 @@ void *iwm_if_alloc(int sizeof_bus, struct device *dev,
 	ndev->wireless_handlers = &iwm_iw_handler_def;
 	ndev->ieee80211_ptr = wdev;
 	SET_NETDEV_DEV(ndev, wiphy_dev(wdev->wiphy));
-	ret = register_netdev(ndev);
-	if (ret < 0) {
-		dev_err(dev, "Failed to register netdev: %d\n", ret);
-		goto out_ndev;
-	}
-
 	wdev->netdev = ndev;
 
 	return iwm;
 
- out_ndev:
-	free_netdev(ndev);
-
  out_priv:
 	iwm_priv_deinit(iwm);
 
@@ -160,8 +150,26 @@ void iwm_if_free(struct iwm_priv *iwm)
 	if (!iwm_to_ndev(iwm))
 		return;
 
-	unregister_netdev(iwm_to_ndev(iwm));
 	free_netdev(iwm_to_ndev(iwm));
 	iwm_wdev_free(iwm);
 	iwm_priv_deinit(iwm);
 }
+
+int iwm_if_add(struct iwm_priv *iwm)
+{
+	struct net_device *ndev = iwm_to_ndev(iwm);
+	int ret;
+
+	ret = register_netdev(ndev);
+	if (ret < 0) {
+		dev_err(&ndev->dev, "Failed to register netdev: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+void iwm_if_remove(struct iwm_priv *iwm)
+{
+	unregister_netdev(iwm_to_ndev(iwm));
+}
diff --git a/drivers/net/wireless/iwmc3200wifi/sdio.c b/drivers/net/wireless/iwmc3200wifi/sdio.c
index b54da677b371..c0405715647a 100644
--- a/drivers/net/wireless/iwmc3200wifi/sdio.c
+++ b/drivers/net/wireless/iwmc3200wifi/sdio.c
@@ -454,10 +454,18 @@ static int iwm_sdio_probe(struct sdio_func *func,
 
 	INIT_WORK(&hw->isr_worker, iwm_sdio_isr_worker);
 
+	ret = iwm_if_add(iwm);
+	if (ret) {
+		dev_err(dev, "add SDIO interface failed\n");
+		goto destroy_wq;
+	}
+
 	dev_info(dev, "IWM SDIO probe\n");
 
 	return 0;
 
+ destroy_wq:
+	destroy_workqueue(hw->isr_wq);
  debugfs_exit:
 	iwm_debugfs_exit(iwm);
  if_free:
@@ -472,6 +480,7 @@ static void iwm_sdio_remove(struct sdio_func *func)
 	struct device *dev = &func->dev;
 
 	iwm_debugfs_exit(iwm);
+	iwm_if_remove(iwm);
 	iwm_if_free(iwm);
 	destroy_workqueue(hw->isr_wq);
 
-- 
cgit v1.2.3-59-g8ed1b


From 4e9aa52e36a7beb4c0163324a3de759d7cf2e442 Mon Sep 17 00:00:00 2001
From: Zhu Yi <yi.zhu@intel.com>
Date: Mon, 15 Jun 2009 21:59:48 +0200
Subject: iwmc3200wifi: fix potential kernel oops on module removal

The iwm_if_free() is called before destroy_workqueue for isr_wq on
device remove method. But if there is still some pending work in
the isr_wq, the required data structures are already freed at this
point. This leeds a kernel oops. The patch fixes this problem by
moving iwm_if_free after destroy_workqueue.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwmc3200wifi/sdio.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/iwmc3200wifi/sdio.c b/drivers/net/wireless/iwmc3200wifi/sdio.c
index c0405715647a..916681837fd2 100644
--- a/drivers/net/wireless/iwmc3200wifi/sdio.c
+++ b/drivers/net/wireless/iwmc3200wifi/sdio.c
@@ -479,10 +479,10 @@ static void iwm_sdio_remove(struct sdio_func *func)
 	struct iwm_priv *iwm = hw_to_iwm(hw);
 	struct device *dev = &func->dev;
 
-	iwm_debugfs_exit(iwm);
 	iwm_if_remove(iwm);
-	iwm_if_free(iwm);
 	destroy_workqueue(hw->isr_wq);
+	iwm_debugfs_exit(iwm);
+	iwm_if_free(iwm);
 
 	sdio_set_drvdata(func, NULL);
 
-- 
cgit v1.2.3-59-g8ed1b


From 68810c5dc5f6bbaa0bbf6acce7cb56d97a1c8fd0 Mon Sep 17 00:00:00 2001
From: Zhu Yi <yi.zhu@intel.com>
Date: Mon, 15 Jun 2009 21:59:49 +0200
Subject: iwmc3200wifi: add a mutex to protect iwm_reset_worker

The patch adds a mutex to protect the iwm_reset_worker against netdev
ndo_open and ndo_stop because all of them call iwm_up and iwm_down in
the implementation. Note the latter two are already protected by
rtnl. So if iwm_reset_worker is not required in the future, the mutex
can also be removed.

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/iwmc3200wifi/iwm.h  |  1 +
 drivers/net/wireless/iwmc3200wifi/main.c | 54 ++++++++++++++++++++++++++++----
 2 files changed, 49 insertions(+), 6 deletions(-)

diff --git a/drivers/net/wireless/iwmc3200wifi/iwm.h b/drivers/net/wireless/iwmc3200wifi/iwm.h
index 4aa0ad1932ff..77c339f8516c 100644
--- a/drivers/net/wireless/iwmc3200wifi/iwm.h
+++ b/drivers/net/wireless/iwmc3200wifi/iwm.h
@@ -288,6 +288,7 @@ struct iwm_priv {
 	u8 *eeprom;
 	struct timer_list watchdog;
 	struct work_struct reset_worker;
+	struct mutex mutex;
 	struct rfkill *rfkill;
 
 	char private[0] __attribute__((__aligned__(NETDEV_ALIGN)));
diff --git a/drivers/net/wireless/iwmc3200wifi/main.c b/drivers/net/wireless/iwmc3200wifi/main.c
index 4d3c423d8ffc..8be206d58222 100644
--- a/drivers/net/wireless/iwmc3200wifi/main.c
+++ b/drivers/net/wireless/iwmc3200wifi/main.c
@@ -112,6 +112,9 @@ static void iwm_statistics_request(struct work_struct *work)
 	iwm_send_umac_stats_req(iwm, 0);
 }
 
+int __iwm_up(struct iwm_priv *iwm);
+int __iwm_down(struct iwm_priv *iwm);
+
 static void iwm_reset_worker(struct work_struct *work)
 {
 	struct iwm_priv *iwm;
@@ -120,6 +123,19 @@ static void iwm_reset_worker(struct work_struct *work)
 
 	iwm = container_of(work, struct iwm_priv, reset_worker);
 
+	/*
+	 * XXX: The iwm->mutex is introduced purely for this reset work,
+	 * because the other users for iwm_up and iwm_down are only netdev
+	 * ndo_open and ndo_stop which are already protected by rtnl.
+	 * Please remove iwm->mutex together if iwm_reset_worker() is not
+	 * required in the future.
+	 */
+	if (!mutex_trylock(&iwm->mutex)) {
+		IWM_WARN(iwm, "We are in the middle of interface bringing "
+			 "UP/DOWN. Skip driver resetting.\n");
+		return;
+	}
+
 	if (iwm->umac_profile_active) {
 		profile = kmalloc(sizeof(struct iwm_umac_profile), GFP_KERNEL);
 		if (profile)
@@ -128,10 +144,10 @@ static void iwm_reset_worker(struct work_struct *work)
 			IWM_ERR(iwm, "Couldn't alloc memory for profile\n");
 	}
 
-	iwm_down(iwm);
+	__iwm_down(iwm);
 
 	while (retry++ < 3) {
-		ret = iwm_up(iwm);
+		ret = __iwm_up(iwm);
 		if (!ret)
 			break;
 
@@ -142,7 +158,7 @@ static void iwm_reset_worker(struct work_struct *work)
 		IWM_WARN(iwm, "iwm_up() failed: %d\n", ret);
 
 		kfree(profile);
-		return;
+		goto out;
 	}
 
 	if (profile) {
@@ -151,6 +167,9 @@ static void iwm_reset_worker(struct work_struct *work)
 		iwm_send_mlme_profile(iwm);
 		kfree(profile);
 	}
+
+ out:
+	mutex_unlock(&iwm->mutex);
 }
 
 static void iwm_watchdog(unsigned long data)
@@ -215,6 +234,7 @@ int iwm_priv_init(struct iwm_priv *iwm)
 	init_timer(&iwm->watchdog);
 	iwm->watchdog.function = iwm_watchdog;
 	iwm->watchdog.data = (unsigned long)iwm;
+	mutex_init(&iwm->mutex);
 
 	return 0;
 }
@@ -476,7 +496,7 @@ void iwm_link_off(struct iwm_priv *iwm)
 
 	iwm_rx_free(iwm);
 
-	cancel_delayed_work(&iwm->stats_request);
+	cancel_delayed_work_sync(&iwm->stats_request);
 	memset(wstats, 0, sizeof(struct iw_statistics));
 	wstats->qual.updated = IW_QUAL_ALL_INVALID;
 
@@ -521,7 +541,7 @@ static int iwm_channels_init(struct iwm_priv *iwm)
 	return 0;
 }
 
-int iwm_up(struct iwm_priv *iwm)
+int __iwm_up(struct iwm_priv *iwm)
 {
 	int ret;
 	struct iwm_notif *notif_reboot, *notif_ack = NULL;
@@ -657,7 +677,18 @@ int iwm_up(struct iwm_priv *iwm)
 	return -EIO;
 }
 
-int iwm_down(struct iwm_priv *iwm)
+int iwm_up(struct iwm_priv *iwm)
+{
+	int ret;
+
+	mutex_lock(&iwm->mutex);
+	ret = __iwm_up(iwm);
+	mutex_unlock(&iwm->mutex);
+
+	return ret;
+}
+
+int __iwm_down(struct iwm_priv *iwm)
 {
 	int ret;
 
@@ -688,3 +719,14 @@ int iwm_down(struct iwm_priv *iwm)
 
 	return 0;
 }
+
+int iwm_down(struct iwm_priv *iwm)
+{
+	int ret;
+
+	mutex_lock(&iwm->mutex);
+	ret = __iwm_down(iwm);
+	mutex_unlock(&iwm->mutex);
+
+	return ret;
+}
-- 
cgit v1.2.3-59-g8ed1b


From f72151fb6820e90ce12a15e2768aa41150c5186d Mon Sep 17 00:00:00 2001
From: Hin-Tak Leung <hintak.leung@gmail.com>
Date: Tue, 16 Jun 2009 03:15:56 +0100
Subject: zd1211rw: adding 083a:e503 as a ZD1211B device

Hans Pontar reported success on the sourceforge zd1211-devs mailing list.
The device is branded "Arcor Easy Stick A 50 WLAN" (device manufactured
by SMC for a German ISP - SMC model name: WN4501H-LF-IR). General
information and Windows driver are available under (German only):

http://www.arcor.de/hilfe/neu/index.php?sid=&aktion=anzeigen&rubrik=004018140&id=487

Device details:

USB-IDs: Vendor: 0x083A Device: 0xE503
Chip ID: zd1211b chip 083a:e503 v4810 high 00-1d-19 AL2230S_RF pa0 g--N-

Signed-off-by: Hin-Tak Leung <htl10@users.sourceforge.net>
Tested-by: Hans Pontar <pontar@gmx.de>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/zd1211rw/zd_usb.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/zd1211rw/zd_usb.c b/drivers/net/wireless/zd1211rw/zd_usb.c
index f0e5e943f6e3..14a19baff214 100644
--- a/drivers/net/wireless/zd1211rw/zd_usb.c
+++ b/drivers/net/wireless/zd1211rw/zd_usb.c
@@ -67,6 +67,7 @@ static struct usb_device_id usb_ids[] = {
 	{ USB_DEVICE(0x079b, 0x0062), .driver_info = DEVICE_ZD1211B },
 	{ USB_DEVICE(0x1582, 0x6003), .driver_info = DEVICE_ZD1211B },
 	{ USB_DEVICE(0x050d, 0x705c), .driver_info = DEVICE_ZD1211B },
+	{ USB_DEVICE(0x083a, 0xe503), .driver_info = DEVICE_ZD1211B },
 	{ USB_DEVICE(0x083a, 0xe506), .driver_info = DEVICE_ZD1211B },
 	{ USB_DEVICE(0x083a, 0x4505), .driver_info = DEVICE_ZD1211B },
 	{ USB_DEVICE(0x0471, 0x1236), .driver_info = DEVICE_ZD1211B },
-- 
cgit v1.2.3-59-g8ed1b


From f0214843ba23d9bf6dc6b8ad2c6ee27b60f0322e Mon Sep 17 00:00:00 2001
From: Jouni Malinen <jouni.malinen@atheros.com>
Date: Tue, 16 Jun 2009 11:59:23 +0300
Subject: ath9k: Fix PCI FATAL interrupts by restoring RETRY_TIMEOUT disabling

An earlier commit, 'ath9k: remove dummy PCI "retry timeout" fix', removed
code that was documented to disable RETRY_TIMEOUT register (PCI reg
0x41) since it was claimed to be a no-op. However, it turns out that
there are some combinations of hosts and ath9k-supported cards for
which this is not a no-op (reg 0x41 has value 0x80, not 0) and this
code (or something similar) is needed. In such cases, the driver may
be next to unusable due to very frequent PCI FATAL interrupts from the
card.

Reverting the earlier commit, i.e., restoring the RETRY_TIMEOUT
disabling, seems to resolve the issue. Since the removal of this code
was not based on any known issue and was purely a cleanup change, the
safest option here is to just revert that commit. Should there be
desire to clean this up in the future, the change will need to be
tested with a more complete coverage of cards and host systems.

http://bugzilla.kernel.org/show_bug.cgi?id=13483

Cc: stable@kernel.org
Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/ath/ath9k/pci.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/drivers/net/wireless/ath/ath9k/pci.c b/drivers/net/wireless/ath/ath9k/pci.c
index ccdf20a2e9be..170c5b32e49b 100644
--- a/drivers/net/wireless/ath/ath9k/pci.c
+++ b/drivers/net/wireless/ath/ath9k/pci.c
@@ -87,6 +87,7 @@ static int ath_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	struct ath_softc *sc;
 	struct ieee80211_hw *hw;
 	u8 csz;
+	u32 val;
 	int ret = 0;
 	struct ath_hw *ah;
 
@@ -133,6 +134,14 @@ static int ath_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	pci_set_master(pdev);
 
+	/*
+	 * Disable the RETRY_TIMEOUT register (0x41) to keep
+	 * PCI Tx retries from interfering with C3 CPU state.
+	 */
+	pci_read_config_dword(pdev, 0x40, &val);
+	if ((val & 0x0000ff00) != 0)
+		pci_write_config_dword(pdev, 0x40, val & 0xffff00ff);
+
 	ret = pci_request_region(pdev, 0, "ath9k");
 	if (ret) {
 		dev_err(&pdev->dev, "PCI memory region reserve error\n");
@@ -239,12 +248,21 @@ static int ath_pci_resume(struct pci_dev *pdev)
 	struct ieee80211_hw *hw = pci_get_drvdata(pdev);
 	struct ath_wiphy *aphy = hw->priv;
 	struct ath_softc *sc = aphy->sc;
+	u32 val;
 	int err;
 
 	err = pci_enable_device(pdev);
 	if (err)
 		return err;
 	pci_restore_state(pdev);
+	/*
+	 * Suspend/Resume resets the PCI configuration space, so we have to
+	 * re-disable the RETRY_TIMEOUT register (0x41) to keep
+	 * PCI Tx retries from interfering with C3 CPU state
+	 */
+	pci_read_config_dword(pdev, 0x40, &val);
+	if ((val & 0x0000ff00) != 0)
+		pci_write_config_dword(pdev, 0x40, val & 0xffff00ff);
 
 	/* Enable LED */
 	ath9k_hw_cfg_output(sc->sc_ah, ATH_LED_PIN,
-- 
cgit v1.2.3-59-g8ed1b


From 7fa20a7f60df0afceafbb8197b5d110507f42c72 Mon Sep 17 00:00:00 2001
From: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Date: Tue, 16 Jun 2009 14:53:24 +0100
Subject: rfkill: rfkill_set_block() when suspended nitpick

If we return after fiddling with the state, userspace will see the
wrong state and rfkill_set_sw_state() won't work until the next call to
rfkill_set_block().  At the moment rfkill_set_block() will always be
called from rfkill_resume(), but this will change in future.

Also, presumably the point of this test is to avoid bothering devices
which may be suspended.  If we don't want to call set_block(), we
probably don't want to call query() either :-).

Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 net/rfkill/core.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index 4e68ab439d5d..868d79f8ac1d 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -270,6 +270,9 @@ static void rfkill_set_block(struct rfkill *rfkill, bool blocked)
 	unsigned long flags;
 	int err;
 
+	if (unlikely(rfkill->dev.power.power_state.event & PM_EVENT_SLEEP))
+		return;
+
 	/*
 	 * Some platforms (...!) generate input events which affect the
 	 * _hard_ kill state -- whenever something tries to change the
@@ -292,9 +295,6 @@ static void rfkill_set_block(struct rfkill *rfkill, bool blocked)
 	rfkill->state |= RFKILL_BLOCK_SW_SETCALL;
 	spin_unlock_irqrestore(&rfkill->lock, flags);
 
-	if (unlikely(rfkill->dev.power.power_state.event & PM_EVENT_SLEEP))
-		return;
-
 	err = rfkill->ops->set_block(rfkill->data, blocked);
 
 	spin_lock_irqsave(&rfkill->lock, flags);
-- 
cgit v1.2.3-59-g8ed1b


From 06d5caf47ef4fbd9efdceae33293c42778cb7b0c Mon Sep 17 00:00:00 2001
From: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Date: Tue, 16 Jun 2009 15:39:51 +0100
Subject: rfkill: don't restore software blocked state on persistent devices

The setting of the "persistent" flag is also made more explicit using
a new rfkill_init_sw_state() function, instead of special-casing
rfkill_set_sw_state() when it is called before registration.

Suspend is a bit of a corner case so we try to get away without adding
another hack to rfkill-input - it's going to be removed soon.
If the state does change over suspend, users will simply have to prod
rfkill-input twice in order to toggle the state.

Userspace policy agents will be able to implement a more consistent user
experience.  For example, they can avoid the above problem if they
toggle devices individually.  Then there would be no "global state"
to get out of sync.

Currently there are only two rfkill drivers with persistent soft-blocked
state.  thinkpad-acpi already checks the software state on resume.
eeepc-laptop will require modification.

Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
CC: Marcel Holtmann <marcel@holtmann.org>
Acked-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/platform/x86/eeepc-laptop.c  |  8 ++++----
 drivers/platform/x86/thinkpad_acpi.c | 14 ++++++-------
 include/linux/rfkill.h               | 32 ++++++++++++++++++++++++-----
 net/rfkill/core.c                    | 40 ++++++++++++++++++++++--------------
 4 files changed, 63 insertions(+), 31 deletions(-)

diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index 03bf522bd7ab..01682eca4360 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -675,8 +675,8 @@ static int eeepc_hotk_add(struct acpi_device *device)
 		if (!ehotk->eeepc_wlan_rfkill)
 			goto wlan_fail;
 
-		rfkill_set_sw_state(ehotk->eeepc_wlan_rfkill,
-				    get_acpi(CM_ASL_WLAN) != 1);
+		rfkill_init_sw_state(ehotk->eeepc_wlan_rfkill,
+				     get_acpi(CM_ASL_WLAN) != 1);
 		result = rfkill_register(ehotk->eeepc_wlan_rfkill);
 		if (result)
 			goto wlan_fail;
@@ -693,8 +693,8 @@ static int eeepc_hotk_add(struct acpi_device *device)
 		if (!ehotk->eeepc_bluetooth_rfkill)
 			goto bluetooth_fail;
 
-		rfkill_set_sw_state(ehotk->eeepc_bluetooth_rfkill,
-				    get_acpi(CM_ASL_BLUETOOTH) != 1);
+		rfkill_init_sw_state(ehotk->eeepc_bluetooth_rfkill,
+				     get_acpi(CM_ASL_BLUETOOTH) != 1);
 		result = rfkill_register(ehotk->eeepc_bluetooth_rfkill);
 		if (result)
 			goto bluetooth_fail;
diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c
index 86e958539f46..40d64c03278c 100644
--- a/drivers/platform/x86/thinkpad_acpi.c
+++ b/drivers/platform/x86/thinkpad_acpi.c
@@ -1163,8 +1163,8 @@ static int __init tpacpi_new_rfkill(const enum tpacpi_rfk_id id,
 {
 	struct tpacpi_rfk *atp_rfk;
 	int res;
-	bool initial_sw_state = false;
-	int initial_sw_status;
+	bool sw_state = false;
+	int sw_status;
 
 	BUG_ON(id >= TPACPI_RFK_SW_MAX || tpacpi_rfkill_switches[id]);
 
@@ -1185,17 +1185,17 @@ static int __init tpacpi_new_rfkill(const enum tpacpi_rfk_id id,
 	atp_rfk->id = id;
 	atp_rfk->ops = tp_rfkops;
 
-	initial_sw_status = (tp_rfkops->get_status)();
-	if (initial_sw_status < 0) {
+	sw_status = (tp_rfkops->get_status)();
+	if (sw_status < 0) {
 		printk(TPACPI_ERR
 			"failed to read initial state for %s, error %d\n",
-			name, initial_sw_status);
+			name, sw_status);
 	} else {
-		initial_sw_state = (initial_sw_status == TPACPI_RFK_RADIO_OFF);
+		sw_state = (sw_status == TPACPI_RFK_RADIO_OFF);
 		if (set_default) {
 			/* try to keep the initial state, since we ask the
 			 * firmware to preserve it across S5 in NVRAM */
-			rfkill_set_sw_state(atp_rfk->rfkill, initial_sw_state);
+			rfkill_init_sw_state(atp_rfk->rfkill, sw_state);
 		}
 	}
 	rfkill_set_hw_state(atp_rfk->rfkill, tpacpi_rfk_check_hwblock_state());
diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index 16e39c7a67fc..dcac724340d8 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -160,8 +160,9 @@ struct rfkill * __must_check rfkill_alloc(const char *name,
  * the rfkill structure. Before calling this function the driver needs
  * to be ready to service method calls from rfkill.
  *
- * If the software blocked state is not set before registration,
- * set_block will be called to initialize it to a default value.
+ * If rfkill_init_sw_state() is not called before registration,
+ * set_block() will be called to initialize the software blocked state
+ * to a default value.
  *
  * If the hardware blocked state is not set before registration,
  * it is assumed to be unblocked.
@@ -234,9 +235,11 @@ bool __must_check rfkill_set_hw_state(struct rfkill *rfkill, bool blocked);
  * rfkill drivers that get events when the soft-blocked state changes
  * (yes, some platforms directly act on input but allow changing again)
  * use this function to notify the rfkill core (and through that also
- * userspace) of the current state.  It is not necessary to notify on
- * resume; since hibernation can always change the soft-blocked state,
- * the rfkill core will unconditionally restore the previous state.
+ * userspace) of the current state.
+ *
+ * Drivers should also call this function after resume if the state has
+ * been changed by the user.  This only makes sense for "persistent"
+ * devices (see rfkill_init_sw_state()).
  *
  * This function can be called in any context, even from within rfkill
  * callbacks.
@@ -246,6 +249,21 @@ bool __must_check rfkill_set_hw_state(struct rfkill *rfkill, bool blocked);
  */
 bool rfkill_set_sw_state(struct rfkill *rfkill, bool blocked);
 
+/**
+ * rfkill_init_sw_state - Initialize persistent software block state
+ * @rfkill: pointer to the rfkill class to modify.
+ * @state: the current software block state to set
+ *
+ * rfkill drivers that preserve their software block state over power off
+ * use this function to notify the rfkill core (and through that also
+ * userspace) of their initial state.  It should only be used before
+ * registration.
+ *
+ * In addition, it marks the device as "persistent".  Persistent devices
+ * are expected to preserve preserve their own state when suspended.
+ */
+void rfkill_init_sw_state(struct rfkill *rfkill, bool blocked);
+
 /**
  * rfkill_set_states - Set the internal rfkill block states
  * @rfkill: pointer to the rfkill class to modify.
@@ -307,6 +325,10 @@ static inline bool rfkill_set_sw_state(struct rfkill *rfkill, bool blocked)
 	return blocked;
 }
 
+static inline void rfkill_init_sw_state(struct rfkill *rfkill, bool blocked)
+{
+}
+
 static inline void rfkill_set_states(struct rfkill *rfkill, bool sw, bool hw)
 {
 }
diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index 868d79f8ac1d..dcf8df7c573c 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -56,7 +56,6 @@ struct rfkill {
 	u32			idx;
 
 	bool			registered;
-	bool			suspended;
 	bool			persistent;
 
 	const struct rfkill_ops	*ops;
@@ -224,7 +223,7 @@ static void rfkill_send_events(struct rfkill *rfkill, enum rfkill_operation op)
 
 static void rfkill_event(struct rfkill *rfkill)
 {
-	if (!rfkill->registered || rfkill->suspended)
+	if (!rfkill->registered)
 		return;
 
 	kobject_uevent(&rfkill->dev.kobj, KOBJ_CHANGE);
@@ -508,19 +507,32 @@ bool rfkill_set_sw_state(struct rfkill *rfkill, bool blocked)
 	blocked = blocked || hwblock;
 	spin_unlock_irqrestore(&rfkill->lock, flags);
 
-	if (!rfkill->registered) {
-		rfkill->persistent = true;
-	} else {
-		if (prev != blocked && !hwblock)
-			schedule_work(&rfkill->uevent_work);
+	if (!rfkill->registered)
+		return blocked;
 
-		rfkill_led_trigger_event(rfkill);
-	}
+	if (prev != blocked && !hwblock)
+		schedule_work(&rfkill->uevent_work);
+
+	rfkill_led_trigger_event(rfkill);
 
 	return blocked;
 }
 EXPORT_SYMBOL(rfkill_set_sw_state);
 
+void rfkill_init_sw_state(struct rfkill *rfkill, bool blocked)
+{
+	unsigned long flags;
+
+	BUG_ON(!rfkill);
+	BUG_ON(rfkill->registered);
+
+	spin_lock_irqsave(&rfkill->lock, flags);
+	__rfkill_set_sw_state(rfkill, blocked);
+	rfkill->persistent = true;
+	spin_unlock_irqrestore(&rfkill->lock, flags);
+}
+EXPORT_SYMBOL(rfkill_init_sw_state);
+
 void rfkill_set_states(struct rfkill *rfkill, bool sw, bool hw)
 {
 	unsigned long flags;
@@ -718,8 +730,6 @@ static int rfkill_suspend(struct device *dev, pm_message_t state)
 
 	rfkill_pause_polling(rfkill);
 
-	rfkill->suspended = true;
-
 	return 0;
 }
 
@@ -728,10 +738,10 @@ static int rfkill_resume(struct device *dev)
 	struct rfkill *rfkill = to_rfkill(dev);
 	bool cur;
 
-	cur = !!(rfkill->state & RFKILL_BLOCK_SW);
-	rfkill_set_block(rfkill, cur);
-
-	rfkill->suspended = false;
+	if (!rfkill->persistent) {
+		cur = !!(rfkill->state & RFKILL_BLOCK_SW);
+		rfkill_set_block(rfkill, cur);
+	}
 
 	rfkill_resume_polling(rfkill);
 
-- 
cgit v1.2.3-59-g8ed1b


From 96e9cfeb9692b0bc6e03f9b6f9cb3c67a40b76d1 Mon Sep 17 00:00:00 2001
From: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Date: Tue, 16 Jun 2009 14:53:52 +0100
Subject: eeepc-laptop: read rfkill soft-blocked state on resume

This will respect state changes over hibernation, e.g. if the user
disables the wireless in the BIOS setup screen.

It reveals an issue where ACPI silently kills the wireless on
suspend.  Normally, the BIOS restores the correct state from
non-volatile storage on boot.  But when hibernation is aborted,
the wireless would remain killed.  Fortunately we can work around
this in the resume handler by simply writing back the same value we
read from NVS.

Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/platform/x86/eeepc-laptop.c | 42 +++++++++++++++++++++++++++++++++----
 1 file changed, 38 insertions(+), 4 deletions(-)

diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index 01682eca4360..8153b3e59189 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -180,6 +180,7 @@ static struct key_entry eeepc_keymap[] = {
  */
 static int eeepc_hotk_add(struct acpi_device *device);
 static int eeepc_hotk_remove(struct acpi_device *device, int type);
+static int eeepc_hotk_resume(struct acpi_device *device);
 
 static const struct acpi_device_id eeepc_device_ids[] = {
 	{EEEPC_HOTK_HID, 0},
@@ -194,6 +195,7 @@ static struct acpi_driver eeepc_hotk_driver = {
 	.ops = {
 		.add = eeepc_hotk_add,
 		.remove = eeepc_hotk_remove,
+		.resume = eeepc_hotk_resume,
 	},
 };
 
@@ -512,15 +514,12 @@ static int notify_brn(void)
 	return -1;
 }
 
-static void eeepc_rfkill_notify(acpi_handle handle, u32 event, void *data)
+static void eeepc_rfkill_hotplug(void)
 {
 	struct pci_dev *dev;
 	struct pci_bus *bus = pci_find_bus(0, 1);
 	bool blocked;
 
-	if (event != ACPI_NOTIFY_BUS_CHECK)
-		return;
-
 	if (!bus) {
 		printk(EEEPC_WARNING "Unable to find PCI bus 1?\n");
 		return;
@@ -551,6 +550,14 @@ static void eeepc_rfkill_notify(acpi_handle handle, u32 event, void *data)
 	rfkill_set_sw_state(ehotk->eeepc_wlan_rfkill, blocked);
 }
 
+static void eeepc_rfkill_notify(acpi_handle handle, u32 event, void *data)
+{
+	if (event != ACPI_NOTIFY_BUS_CHECK)
+		return;
+
+	eeepc_rfkill_hotplug();
+}
+
 static void eeepc_hotk_notify(acpi_handle handle, u32 event, void *data)
 {
 	static struct key_entry *key;
@@ -734,6 +741,33 @@ static int eeepc_hotk_remove(struct acpi_device *device, int type)
 	return 0;
 }
 
+static int eeepc_hotk_resume(struct acpi_device *device)
+{
+	if (ehotk->eeepc_wlan_rfkill) {
+		bool wlan;
+
+		/* Workaround - it seems that _PTS disables the wireless
+		   without notification or changing the value read by WLAN.
+		   Normally this is fine because the correct value is restored
+		   from the non-volatile storage on resume, but we need to do
+		   it ourself if case suspend is aborted, or we lose wireless.
+		 */
+		wlan = get_acpi(CM_ASL_WLAN);
+		set_acpi(CM_ASL_WLAN, wlan);
+
+		rfkill_set_sw_state(ehotk->eeepc_wlan_rfkill,
+				    wlan != 1);
+
+		eeepc_rfkill_hotplug();
+	}
+
+	if (ehotk->eeepc_bluetooth_rfkill)
+		rfkill_set_sw_state(ehotk->eeepc_bluetooth_rfkill,
+				    get_acpi(CM_ASL_BLUETOOTH) != 1);
+
+	return 0;
+}
+
 /*
  * Hwmon
  */
-- 
cgit v1.2.3-59-g8ed1b


From 464902e812025792c9e33e19e1555c343672d5cf Mon Sep 17 00:00:00 2001
From: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Date: Tue, 16 Jun 2009 14:54:04 +0100
Subject: rfkill: export persistent attribute in sysfs

This information allows userspace to implement a hybrid policy where
it can store the rfkill soft-blocked state in platform non-volatile
storage if available, and if not then file-based storage can be used.

Some users prefer platform non-volatile storage because of the behaviour
when dual-booting multiple versions of Linux, or if the rfkill setting
is changed in the BIOS setting screens, or if the BIOS responds to
wireless-toggle hotkeys itself before the relevant platform driver has
been loaded.

Signed-off-by: Alan Jenkins <alan-jenkins@tuffmail.co.uk>
Acked-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 Documentation/rfkill.txt |  2 ++
 include/linux/rfkill.h   |  5 +++--
 net/rfkill/core.c        | 10 ++++++++++
 3 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/Documentation/rfkill.txt b/Documentation/rfkill.txt
index c8acd8659e91..b4860509c319 100644
--- a/Documentation/rfkill.txt
+++ b/Documentation/rfkill.txt
@@ -111,6 +111,8 @@ following attributes:
 
 	name: Name assigned by driver to this key (interface or driver name).
 	type: Driver type string ("wlan", "bluetooth", etc).
+	persistent: Whether the soft blocked state is initialised from
+	            non-volatile storage at startup.
 	state: Current state of the transmitter
 		0: RFKILL_STATE_SOFT_BLOCKED
 			transmitter is turned off by software
diff --git a/include/linux/rfkill.h b/include/linux/rfkill.h
index dcac724340d8..e73e2429a1b1 100644
--- a/include/linux/rfkill.h
+++ b/include/linux/rfkill.h
@@ -259,8 +259,9 @@ bool rfkill_set_sw_state(struct rfkill *rfkill, bool blocked);
  * userspace) of their initial state.  It should only be used before
  * registration.
  *
- * In addition, it marks the device as "persistent".  Persistent devices
- * are expected to preserve preserve their own state when suspended.
+ * In addition, it marks the device as "persistent", an attribute which
+ * can be read by userspace.  Persistent devices are expected to preserve
+ * their own state when suspended.
  */
 void rfkill_init_sw_state(struct rfkill *rfkill, bool blocked);
 
diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index dcf8df7c573c..79693fe2001e 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -610,6 +610,15 @@ static ssize_t rfkill_idx_show(struct device *dev,
 	return sprintf(buf, "%d\n", rfkill->idx);
 }
 
+static ssize_t rfkill_persistent_show(struct device *dev,
+			       struct device_attribute *attr,
+			       char *buf)
+{
+	struct rfkill *rfkill = to_rfkill(dev);
+
+	return sprintf(buf, "%d\n", rfkill->persistent);
+}
+
 static u8 user_state_from_blocked(unsigned long state)
 {
 	if (state & RFKILL_BLOCK_HW)
@@ -668,6 +677,7 @@ static struct device_attribute rfkill_dev_attrs[] = {
 	__ATTR(name, S_IRUGO, rfkill_name_show, NULL),
 	__ATTR(type, S_IRUGO, rfkill_type_show, NULL),
 	__ATTR(index, S_IRUGO, rfkill_idx_show, NULL),
+	__ATTR(persistent, S_IRUGO, rfkill_persistent_show, NULL),
 	__ATTR(state, S_IRUGO|S_IWUSR, rfkill_state_show, rfkill_state_store),
 	__ATTR(claim, S_IRUGO|S_IWUSR, rfkill_claim_show, rfkill_claim_store),
 	__ATTR_NULL
-- 
cgit v1.2.3-59-g8ed1b


From 8451d22dad40a66416b8d9c0952efa09ec5398c5 Mon Sep 17 00:00:00 2001
From: Jouni Malinen <jouni.malinen@atheros.com>
Date: Tue, 16 Jun 2009 11:59:23 +0300
Subject: ath5k: avoid PCI FATAL interrupts by restoring RETRY_TIMEOUT
 disabling

This reverts 'ath5k: remove dummy PCI "retry timeout" fix' on the
same theory as in 'ath9k: Fix PCI FATAL interrupts by restoring
RETRY_TIMEOUT disabling'.

Reported-by: Bob Copeland <me@bobcopeland.com>
Cc: stable@kernel.org
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/ath/ath5k/base.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/wireless/ath/ath5k/base.c b/drivers/net/wireless/ath/ath5k/base.c
index 55f7de09d134..fc5ba0e3a681 100644
--- a/drivers/net/wireless/ath/ath5k/base.c
+++ b/drivers/net/wireless/ath/ath5k/base.c
@@ -686,6 +686,13 @@ ath5k_pci_resume(struct pci_dev *pdev)
 	if (err)
 		return err;
 
+	/*
+	 * Suspend/Resume resets the PCI configuration space, so we have to
+	 * re-disable the RETRY_TIMEOUT register (0x41) to keep
+	 * PCI Tx retries from interfering with C3 CPU state
+	 */
+	pci_write_config_byte(pdev, 0x41, 0);
+
 	err = request_irq(pdev->irq, ath5k_intr, IRQF_SHARED, "ath", sc);
 	if (err) {
 		ATH5K_ERR(sc, "request_irq failed\n");
-- 
cgit v1.2.3-59-g8ed1b


From a878417cc576720d3c9ff5399522d06f226bad7d Mon Sep 17 00:00:00 2001
From: Troy Moure <twmoure@szypr.net>
Date: Wed, 17 Jun 2009 11:51:56 +0100
Subject: acer-wmi: fix rfkill conversion

"rfkill: rewrite" incorrectly reversed
the meaning of 'state' in acer_rfkill_update() when it changed
rfkill_force_state() to rfkill_set_sw_state().  Fix it.

Signed-off-by: Troy Moure <twmoure@szypr.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/platform/x86/acer-wmi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c
index 09a503e5da6a..be2fd6f91639 100644
--- a/drivers/platform/x86/acer-wmi.c
+++ b/drivers/platform/x86/acer-wmi.c
@@ -958,12 +958,12 @@ static void acer_rfkill_update(struct work_struct *ignored)
 
 	status = get_u32(&state, ACER_CAP_WIRELESS);
 	if (ACPI_SUCCESS(status))
-		rfkill_set_sw_state(wireless_rfkill, !!state);
+		rfkill_set_sw_state(wireless_rfkill, !state);
 
 	if (has_cap(ACER_CAP_BLUETOOTH)) {
 		status = get_u32(&state, ACER_CAP_BLUETOOTH);
 		if (ACPI_SUCCESS(status))
-			rfkill_set_sw_state(bluetooth_rfkill, !!state);
+			rfkill_set_sw_state(bluetooth_rfkill, !state);
 	}
 
 	schedule_delayed_work(&acer_rfkill_work, round_jiffies_relative(HZ));
-- 
cgit v1.2.3-59-g8ed1b


From 58f5fffdc3b8567b3fa8ed77d75699db87e2d1d4 Mon Sep 17 00:00:00 2001
From: Gabor Juhos <juhosg@openwrt.org>
Date: Wed, 17 Jun 2009 20:53:20 +0200
Subject: ath9k: wait for beacon frame along with CAB

Changes-licensed-under: ISC
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/ath/ath9k/recv.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/ath/ath9k/recv.c b/drivers/net/wireless/ath/ath9k/recv.c
index f99f3a76df3f..cece1c4c6bda 100644
--- a/drivers/net/wireless/ath/ath9k/recv.c
+++ b/drivers/net/wireless/ath/ath9k/recv.c
@@ -539,11 +539,14 @@ static void ath_rx_ps_beacon(struct ath_softc *sc, struct sk_buff *skb)
 	if (ath_beacon_dtim_pending_cab(skb)) {
 		/*
 		 * Remain awake waiting for buffered broadcast/multicast
-		 * frames.
+		 * frames. If the last broadcast/multicast frame is not
+		 * received properly, the next beacon frame will work as
+		 * a backup trigger for returning into NETWORK SLEEP state,
+		 * so we are waiting for it as well.
 		 */
 		DPRINTF(sc, ATH_DBG_PS, "Received DTIM beacon indicating "
 			"buffered broadcast/multicast frame(s)\n");
-		sc->sc_flags |= SC_OP_WAIT_FOR_CAB;
+		sc->sc_flags |= SC_OP_WAIT_FOR_CAB | SC_OP_WAIT_FOR_BEACON;
 		return;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 38ab422e64d991472ce21ffdd7a37e8a02279697 Mon Sep 17 00:00:00 2001
From: Gabor Juhos <juhosg@openwrt.org>
Date: Wed, 17 Jun 2009 20:53:21 +0200
Subject: ath9k: restore PS mode, before we put the chip into FULL SLEEP state.

We want to put the chip into FULL SLEEP state, when we are disabling the
radio, but the the current code always change it to AWAKE/NETWORK SLEEP.

Changes-licensed-under: ISC
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/ath/ath9k/main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c
index 9f49a3251d4d..66a6c1f5022a 100644
--- a/drivers/net/wireless/ath/ath9k/main.c
+++ b/drivers/net/wireless/ath/ath9k/main.c
@@ -1196,8 +1196,8 @@ void ath_radio_disable(struct ath_softc *sc)
 
 	ath9k_hw_phy_disable(ah);
 	ath9k_hw_configpcipowersave(ah, 1);
-	ath9k_hw_setpower(ah, ATH9K_PM_FULL_SLEEP);
 	ath9k_ps_restore(sc);
+	ath9k_hw_setpower(ah, ATH9K_PM_FULL_SLEEP);
 }
 
 /*******************/
-- 
cgit v1.2.3-59-g8ed1b


From 488f7e0cdd99e7dc75dbe32d78474c070deadaea Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Thu, 18 Jun 2009 11:08:01 -0700
Subject: MAINTAINERS: Fix Atheros pattern paths

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 MAINTAINERS | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 2cb7566904b1..d8209002fddc 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -940,7 +940,7 @@ M:	me@bobcopeland.com
 L:	linux-wireless@vger.kernel.org
 L:	ath5k-devel@lists.ath5k.org
 S:	Maintained
-F:	drivers/net/wireless/ath5k/
+F:	drivers/net/wireless/ath/ath5k/
 
 ATHEROS ATH9K WIRELESS DRIVER
 P:	Luis R. Rodriguez
@@ -956,7 +956,7 @@ M:	senthilkumar@atheros.com
 L:	linux-wireless@vger.kernel.org
 L:	ath9k-devel@lists.ath9k.org
 S:	Supported
-F:	drivers/net/wireless/ath9k/
+F:	drivers/net/wireless/ath/ath9k/
 
 ATHEROS AR9170 WIRELESS DRIVER
 P:	Christian Lamparter
@@ -964,7 +964,7 @@ M:	chunkeey@web.de
 L:	linux-wireless@vger.kernel.org
 W:	http://wireless.kernel.org/en/users/Drivers/ar9170
 S:	Maintained
-F:	drivers/net/wireless/ar9170/
+F:	drivers/net/wireless/ath/ar9170/
 
 ATI_REMOTE2 DRIVER
 P:	Ville Syrjala
-- 
cgit v1.2.3-59-g8ed1b


From eab0cd493c08632ef10624d0169849c973951c66 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Fri, 19 Jun 2009 01:06:45 +0200
Subject: ath5k: fix beacon_int handling

73ca5203366235f8a43e490767284ba8cfd8c479
(ath5k: remove conf->beacon_int usage)
removed bintval setting from ath5k_config. We need to init the
interval earlier and don't touch it in add_interface anymore.

Otherwise it will be set only once by upper layer through
bss_info_changed but not on second and further hostap executions.

We ended up having bintval 1000 which rendered the AP useless on
many clients.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: Nick Kossifidis <mickflemm@gmail.com>
Cc: Luis R. Rodriguez <lrodriguez@atheros.com>
Cc: Bob Copeland <me@bobcopeland.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/ath/ath5k/base.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/ath5k/base.c b/drivers/net/wireless/ath/ath5k/base.c
index fc5ba0e3a681..ea045151f953 100644
--- a/drivers/net/wireless/ath/ath5k/base.c
+++ b/drivers/net/wireless/ath/ath5k/base.c
@@ -538,6 +538,7 @@ ath5k_pci_probe(struct pci_dev *pdev,
 	sc->iobase = mem; /* So we can unmap it on detach */
 	sc->cachelsz = csz * sizeof(u32); /* convert to bytes */
 	sc->opmode = NL80211_IFTYPE_STATION;
+	sc->bintval = 1000;
 	mutex_init(&sc->lock);
 	spin_lock_init(&sc->rxbuflock);
 	spin_lock_init(&sc->txbuflock);
@@ -2755,9 +2756,6 @@ static int ath5k_add_interface(struct ieee80211_hw *hw,
 		goto end;
 	}
 
-	/* Set to a reasonable value. Note that this will
-	 * be set to mac80211's value at ath5k_config(). */
-	sc->bintval = 1000;
 	ath5k_hw_set_lladdr(sc->ah, conf->mac_addr);
 
 	ret = 0;
-- 
cgit v1.2.3-59-g8ed1b


From 155cc9e4b1d60161ee53ffaf2c15b9411f086fa7 Mon Sep 17 00:00:00 2001
From: Andrey Yurovsky <andrey@cozybit.com>
Date: Tue, 16 Jun 2009 11:31:04 -0700
Subject: cfg80211: allow adding/deleting stations on mesh

Commit b2a151a288 added a check that prevents adding or deleting
stations on non-AP interfaces.  Adding and deleting stations is
supported for Mesh Point interfaces, so add Mesh Point to that check as
well.

Signed-off-by: Andrey Yurovsky <andrey@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 net/wireless/nl80211.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 24168560ebae..2c55d25ed34d 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -1763,7 +1763,8 @@ static int nl80211_new_station(struct sk_buff *skb, struct genl_info *info)
 		goto out_rtnl;
 
 	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
-	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN) {
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) {
 		err = -EINVAL;
 		goto out;
 	}
@@ -1812,7 +1813,8 @@ static int nl80211_del_station(struct sk_buff *skb, struct genl_info *info)
 		goto out_rtnl;
 
 	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
-	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN) {
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) {
 		err = -EINVAL;
 		goto out;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 9a5e8bbc8fece7851a2a69a8676a6fd0507bc550 Mon Sep 17 00:00:00 2001
From: Andrey Yurovsky <andrey@cozybit.com>
Date: Tue, 16 Jun 2009 16:09:37 -0700
Subject: cfg80211: allow setting station parameters in mesh

Mesh Point interfaces can also set parameters, for example plink_open is
used to manually establish peer links from user-space (currently via
iw).  Add Mesh Point to the check in nl80211_set_station.

Signed-off-by: Andrey Yurovsky <andrey@cozybit.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 net/wireless/nl80211.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 2c55d25ed34d..304b3d568e07 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -1688,7 +1688,8 @@ static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info)
 		goto out_rtnl;
 
 	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
-	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN) {
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN &&
+	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) {
 		err = -EINVAL;
 		goto out;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From a97f4424fb4cddecf9b13c9b0e3f79924b624a7f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Thu, 18 Jun 2009 17:23:43 +0200
Subject: cfg80211: validate station settings

When I disallowed interfering with stations on non-AP interfaces,
I not only forget mesh but also managed interfaces which need
this for the authorized flag. Let's actually validate everything
properly.

This fixes an nl80211 regression introduced by the interfering,
under which wpa_supplicant -Dnl80211 could not properly connect.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 net/wireless/nl80211.c | 94 +++++++++++++++++++++++++++++++++++++++++---------
 1 file changed, 78 insertions(+), 16 deletions(-)

diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 304b3d568e07..241bddd0b4f1 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -1687,14 +1687,52 @@ static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info)
 	if (err)
 		goto out_rtnl;
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
-	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN &&
-	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) {
-		err = -EINVAL;
+	err = get_vlan(info->attrs[NL80211_ATTR_STA_VLAN], drv, &params.vlan);
+	if (err)
 		goto out;
+
+	/* validate settings */
+	err = 0;
+
+	switch (dev->ieee80211_ptr->iftype) {
+	case NL80211_IFTYPE_AP:
+	case NL80211_IFTYPE_AP_VLAN:
+		/* disallow mesh-specific things */
+		if (params.plink_action)
+			err = -EINVAL;
+		break;
+	case NL80211_IFTYPE_STATION:
+		/* disallow everything but AUTHORIZED flag */
+		if (params.plink_action)
+			err = -EINVAL;
+		if (params.vlan)
+			err = -EINVAL;
+		if (params.supported_rates)
+			err = -EINVAL;
+		if (params.ht_capa)
+			err = -EINVAL;
+		if (params.listen_interval >= 0)
+			err = -EINVAL;
+		if (params.sta_flags_mask & ~BIT(NL80211_STA_FLAG_AUTHORIZED))
+			err = -EINVAL;
+		break;
+	case NL80211_IFTYPE_MESH_POINT:
+		/* disallow things mesh doesn't support */
+		if (params.vlan)
+			err = -EINVAL;
+		if (params.ht_capa)
+			err = -EINVAL;
+		if (params.listen_interval >= 0)
+			err = -EINVAL;
+		if (params.supported_rates)
+			err = -EINVAL;
+		if (params.sta_flags_mask)
+			err = -EINVAL;
+		break;
+	default:
+		err = -EINVAL;
 	}
 
-	err = get_vlan(info->attrs[NL80211_ATTR_STA_VLAN], drv, &params.vlan);
 	if (err)
 		goto out;
 
@@ -1729,9 +1767,6 @@ static int nl80211_new_station(struct sk_buff *skb, struct genl_info *info)
 	if (!info->attrs[NL80211_ATTR_MAC])
 		return -EINVAL;
 
-	if (!info->attrs[NL80211_ATTR_STA_AID])
-		return -EINVAL;
-
 	if (!info->attrs[NL80211_ATTR_STA_LISTEN_INTERVAL])
 		return -EINVAL;
 
@@ -1746,9 +1781,11 @@ static int nl80211_new_station(struct sk_buff *skb, struct genl_info *info)
 	params.listen_interval =
 		nla_get_u16(info->attrs[NL80211_ATTR_STA_LISTEN_INTERVAL]);
 
-	params.aid = nla_get_u16(info->attrs[NL80211_ATTR_STA_AID]);
-	if (!params.aid || params.aid > IEEE80211_MAX_AID)
-		return -EINVAL;
+	if (info->attrs[NL80211_ATTR_STA_AID]) {
+		params.aid = nla_get_u16(info->attrs[NL80211_ATTR_STA_AID]);
+		if (!params.aid || params.aid > IEEE80211_MAX_AID)
+			return -EINVAL;
+	}
 
 	if (info->attrs[NL80211_ATTR_HT_CAPABILITY])
 		params.ht_capa =
@@ -1763,14 +1800,39 @@ static int nl80211_new_station(struct sk_buff *skb, struct genl_info *info)
 	if (err)
 		goto out_rtnl;
 
-	if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
-	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN &&
-	    dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT) {
-		err = -EINVAL;
+	err = get_vlan(info->attrs[NL80211_ATTR_STA_VLAN], drv, &params.vlan);
+	if (err)
 		goto out;
+
+	/* validate settings */
+	err = 0;
+
+	switch (dev->ieee80211_ptr->iftype) {
+	case NL80211_IFTYPE_AP:
+	case NL80211_IFTYPE_AP_VLAN:
+		/* all ok but must have AID */
+		if (!params.aid)
+			err = -EINVAL;
+		break;
+	case NL80211_IFTYPE_MESH_POINT:
+		/* disallow things mesh doesn't support */
+		if (params.vlan)
+			err = -EINVAL;
+		if (params.aid)
+			err = -EINVAL;
+		if (params.ht_capa)
+			err = -EINVAL;
+		if (params.listen_interval >= 0)
+			err = -EINVAL;
+		if (params.supported_rates)
+			err = -EINVAL;
+		if (params.sta_flags_mask)
+			err = -EINVAL;
+		break;
+	default:
+		err = -EINVAL;
 	}
 
-	err = get_vlan(info->attrs[NL80211_ATTR_STA_VLAN], drv, &params.vlan);
 	if (err)
 		goto out;
 
-- 
cgit v1.2.3-59-g8ed1b


From 5587931c30dcf778cf7071d1cbac8ea584706dd8 Mon Sep 17 00:00:00 2001
From: Joonyoung Shim <jy0922.shim@samsung.com>
Date: Tue, 16 Jun 2009 20:05:57 +0900
Subject: [ARM] Add old Feroceon support to compressed/head.S

This patch supports the cache handling for some old Feroceon cores for
which the CPU ID is like 0x41159260.  This is a complement to
commit ab6d15d50637fc25ee941710b23fed09ceb28db3.

Signed-off-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Nicolas Pitre <nico@marvell.com>
---
 arch/arm/boot/compressed/head.S | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/arch/arm/boot/compressed/head.S b/arch/arm/boot/compressed/head.S
index 01d49be3b2ca..4515728c5345 100644
--- a/arch/arm/boot/compressed/head.S
+++ b/arch/arm/boot/compressed/head.S
@@ -674,6 +674,15 @@ proc_types:
 		b	__armv4_mmu_cache_off
 		b	__armv5tej_mmu_cache_flush
 
+#ifdef CONFIG_CPU_FEROCEON_OLD_ID
+		/* this conflicts with the standard ARMv5TE entry */
+		.long	0x41009260		@ Old Feroceon
+		.long	0xff00fff0
+		b	__armv4_mmu_cache_on
+		b	__armv4_mmu_cache_off
+		b	__armv5tej_mmu_cache_flush
+#endif
+
 		.word	0x66015261		@ FA526
 		.word	0xff01fff1
 		b	__fa526_cache_on
-- 
cgit v1.2.3-59-g8ed1b


From f598282f5145036312d90875d0ed5c14b49fd8a7 Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <willy@linux.intel.com>
Date: Thu, 18 Jun 2009 19:15:59 -0700
Subject: PCI: Fix the NIU MSI-X problem in a better way

The previous MSI-X fix (8d181018532dd709ec1f789e374cda92d7b01ce1) had
three bugs.  First, it didn't move the write that disabled the vector.
This led to writing garbage to the MSI-X vector (spotted by Michael
Ellerman).  It didn't fix the PCI resume case, and it had a race window
where the device could generate an interrupt before the MSI-X registers
were programmed (leading to a DMA to random addresses).

Fortunately, the MSI-X capability has a bit to mask all the vectors.
By setting this bit instead of clearing the enable bit, we can ensure
the device will not generate spurious interrupts.  Since the capability
is now enabled, the NIU device will not have a problem with the reads
and writes to the MSI-X registers being in the original order in the code.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/msi.c | 46 ++++++++++++++++++++++++++++------------------
 1 file changed, 28 insertions(+), 18 deletions(-)

diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 79e56906c7f1..944e45e4a84f 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -313,22 +313,22 @@ static void __pci_restore_msix_state(struct pci_dev *dev)
 
 	if (!dev->msix_enabled)
 		return;
+	BUG_ON(list_empty(&dev->msi_list));
+	entry = list_entry(dev->msi_list.next, struct msi_desc, list);
+	pos = entry->msi_attrib.pos;
+	pci_read_config_word(dev, pos + PCI_MSIX_FLAGS, &control);
 
 	/* route the table */
 	pci_intx_for_msi(dev, 0);
-	msix_set_enable(dev, 0);
+	control |= PCI_MSIX_FLAGS_ENABLE | PCI_MSIX_FLAGS_MASKALL;
+	pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control);
 
 	list_for_each_entry(entry, &dev->msi_list, list) {
 		write_msi_msg(entry->irq, &entry->msg);
 		msix_mask_irq(entry, entry->masked);
 	}
 
-	BUG_ON(list_empty(&dev->msi_list));
-	entry = list_entry(dev->msi_list.next, struct msi_desc, list);
-	pos = entry->msi_attrib.pos;
-	pci_read_config_word(dev, pos + PCI_MSIX_FLAGS, &control);
 	control &= ~PCI_MSIX_FLAGS_MASKALL;
-	control |= PCI_MSIX_FLAGS_ENABLE;
 	pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control);
 }
 
@@ -419,11 +419,14 @@ static int msix_capability_init(struct pci_dev *dev,
 	u8 bir;
 	void __iomem *base;
 
-	msix_set_enable(dev, 0);/* Ensure msix is disabled as I set it up */
-
    	pos = pci_find_capability(dev, PCI_CAP_ID_MSIX);
+	pci_read_config_word(dev, pos + PCI_MSIX_FLAGS, &control);
+
+	/* Ensure MSI-X is disabled while it is set up */
+	control &= ~PCI_MSIX_FLAGS_ENABLE;
+	pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control);
+
 	/* Request & Map MSI-X table region */
- 	pci_read_config_word(dev, msi_control_reg(pos), &control);
 	nr_entries = multi_msix_capable(control);
 
  	pci_read_config_dword(dev, msix_table_offset_reg(pos), &table_offset);
@@ -434,7 +437,6 @@ static int msix_capability_init(struct pci_dev *dev,
 	if (base == NULL)
 		return -ENOMEM;
 
-	/* MSI-X Table Initialization */
 	for (i = 0; i < nvec; i++) {
 		entry = alloc_msi_entry(dev);
 		if (!entry)
@@ -447,7 +449,6 @@ static int msix_capability_init(struct pci_dev *dev,
 		entry->msi_attrib.default_irq = dev->irq;
 		entry->msi_attrib.pos = pos;
 		entry->mask_base = base;
-		msix_mask_irq(entry, 1);
 
 		list_add_tail(&entry->list, &dev->msi_list);
 	}
@@ -472,22 +473,31 @@ static int msix_capability_init(struct pci_dev *dev,
 		return ret;
 	}
 
+	/*
+	 * Some devices require MSI-X to be enabled before we can touch the
+	 * MSI-X registers.  We need to mask all the vectors to prevent
+	 * interrupts coming in before they're fully set up.
+	 */
+	control |= PCI_MSIX_FLAGS_MASKALL | PCI_MSIX_FLAGS_ENABLE;
+	pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control);
+
 	i = 0;
 	list_for_each_entry(entry, &dev->msi_list, list) {
 		entries[i].vector = entry->irq;
 		set_irq_msi(entry->irq, entry);
+		j = entries[i].entry;
+		entry->masked = readl(base + j * PCI_MSIX_ENTRY_SIZE +
+					PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET);
+		msix_mask_irq(entry, 1);
 		i++;
 	}
-	/* Set MSI-X enabled bits */
+
+	/* Set MSI-X enabled bits and unmask the function */
 	pci_intx_for_msi(dev, 0);
-	msix_set_enable(dev, 1);
 	dev->msix_enabled = 1;
 
-	list_for_each_entry(entry, &dev->msi_list, list) {
-		int vector = entry->msi_attrib.entry_nr;
-		entry->masked = readl(base + vector * PCI_MSIX_ENTRY_SIZE +
-					PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET);
-	}
+	control &= ~PCI_MSIX_FLAGS_MASKALL;
+	pci_write_config_word(dev, pos + PCI_MSIX_FLAGS, control);
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 2af5066f664cb011cf17d2e4414491fe24597e07 Mon Sep 17 00:00:00 2001
From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Date: Thu, 18 Jun 2009 19:20:26 -0700
Subject: PCI: make msi_free_irqs() to use msix_mask_irq() instead of open
 coded write

Use msix_mask_irq() instead of direct use of writel, so as not to clear
preserved bits in the Vector Control register [31:1].

Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/msi.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 944e45e4a84f..d9f06fbfa0bf 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -653,10 +653,7 @@ static int msi_free_irqs(struct pci_dev* dev)
 
 	list_for_each_entry_safe(entry, tmp, &dev->msi_list, list) {
 		if (entry->msi_attrib.is_msix) {
-			writel(1, entry->mask_base + entry->msi_attrib.entry_nr
-				  * PCI_MSIX_ENTRY_SIZE
-				  + PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET);
-
+			msix_mask_irq(entry, 1);
 			if (list_is_last(&entry->list, &dev->msi_list))
 				iounmap(entry->mask_base);
 		}
-- 
cgit v1.2.3-59-g8ed1b


From 6adb8f388ef2f23d4a81e1e42d15f22d62749a06 Mon Sep 17 00:00:00 2001
From: Paul Walmsley <paul@pwsan.com>
Date: Fri, 19 Jun 2009 19:08:24 -0600
Subject: OMAP3 clock: remove wait for DPLL3 M2 clock to stabilize

The original CDP kernel that this code comes from waited for 0x800
loops after switching the CORE DPLL M2 divider.  This does not appear
to be necessary.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
---
 arch/arm/mach-omap2/sram34xx.S | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/arch/arm/mach-omap2/sram34xx.S b/arch/arm/mach-omap2/sram34xx.S
index c080c82521e1..84781a6cd263 100644
--- a/arch/arm/mach-omap2/sram34xx.S
+++ b/arch/arm/mach-omap2/sram34xx.S
@@ -102,9 +102,6 @@ configure_core_dpll:
 	orr	r12, r12, r3, lsl #0x1B	@ r3 contains the M2 val
 	str	r12, [r11]
 	ldr	r12, [r11]		@ posted-write barrier for CM
-	mov 	r12, #0x800		@ wait for the clock to stabilise
-	cmp	r3, #2
-	bne	wait_clk_stable
 	bx	lr
 wait_clk_stable:
 	subs 	r12, r12, #1
-- 
cgit v1.2.3-59-g8ed1b


From 2f135eaf182761bb9a5cbd5138a447b0ad2a1fef Mon Sep 17 00:00:00 2001
From: Paul Walmsley <paul@pwsan.com>
Date: Fri, 19 Jun 2009 19:08:25 -0600
Subject: OMAP3 clock: initialize SDRC timings at kernel start

On the OMAP3, initialize SDRC timings when the kernel boots.  This ensures
that the kernel is running with known, optimized SDRC timings, rather than
whatever was configured by the bootloader.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
---
 arch/arm/mach-omap2/clock34xx.c |  3 ---
 arch/arm/mach-omap2/io.c        | 36 ++++++++++++++++++++++++++++++++++++
 2 files changed, 36 insertions(+), 3 deletions(-)

diff --git a/arch/arm/mach-omap2/clock34xx.c b/arch/arm/mach-omap2/clock34xx.c
index 9e43fe5209d3..5458ab3bf65a 100644
--- a/arch/arm/mach-omap2/clock34xx.c
+++ b/arch/arm/mach-omap2/clock34xx.c
@@ -718,9 +718,6 @@ static int omap3_core_dpll_m2_set_rate(struct clk *clk, unsigned long rate)
 	if (clk != &dpll3_m2_ck)
 		return -EINVAL;
 
-	if (rate == clk->rate)
-		return 0;
-
 	validrate = omap2_clksel_round_rate_div(clk, rate, &new_div);
 	if (validrate != rate)
 		return -EINVAL;
diff --git a/arch/arm/mach-omap2/io.c b/arch/arm/mach-omap2/io.c
index 32afd9448216..3a86b0f66031 100644
--- a/arch/arm/mach-omap2/io.c
+++ b/arch/arm/mach-omap2/io.c
@@ -21,6 +21,7 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/io.h>
+#include <linux/clk.h>
 
 #include <asm/tlb.h>
 
@@ -241,6 +242,40 @@ void __init omap2_map_common_io(void)
 	omapfb_reserve_sdram();
 }
 
+/*
+ * omap2_init_reprogram_sdrc - reprogram SDRC timing parameters
+ *
+ * Sets the CORE DPLL3 M2 divider to the same value that it's at
+ * currently.  This has the effect of setting the SDRC SDRAM AC timing
+ * registers to the values currently defined by the kernel.  Currently
+ * only defined for OMAP3; will return 0 if called on OMAP2.  Returns
+ * -EINVAL if the dpll3_m2_ck cannot be found, 0 if called on OMAP2,
+ * or passes along the return value of clk_set_rate().
+ */
+static int __init _omap2_init_reprogram_sdrc(void)
+{
+	struct clk *dpll3_m2_ck;
+	int v = -EINVAL;
+	long rate;
+
+	if (!cpu_is_omap34xx())
+		return 0;
+
+	dpll3_m2_ck = clk_get(NULL, "dpll3_m2_ck");
+	if (!dpll3_m2_ck)
+		return -EINVAL;
+
+	rate = clk_get_rate(dpll3_m2_ck);
+	pr_info("Reprogramming SDRC clock to %ld Hz\n", rate);
+	v = clk_set_rate(dpll3_m2_ck, rate);
+	if (v)
+		pr_err("dpll3_m2_clk rate change failed: %d\n", v);
+
+	clk_put(dpll3_m2_ck);
+
+	return v;
+}
+
 void __init omap2_init_common_hw(struct omap_sdrc_params *sp)
 {
 	omap2_mux_init();
@@ -249,6 +284,7 @@ void __init omap2_init_common_hw(struct omap_sdrc_params *sp)
 	clkdm_init(clockdomains_omap, clkdm_pwrdm_autodeps);
 	omap2_clk_init();
 	omap2_sdrc_init(sp);
+	_omap2_init_reprogram_sdrc();
 #endif
 	gpmc_init();
 }
-- 
cgit v1.2.3-59-g8ed1b


From c9812d042a21eb492a36cfabf9f41107f5ecee3d Mon Sep 17 00:00:00 2001
From: Paul Walmsley <paul@pwsan.com>
Date: Fri, 19 Jun 2009 19:08:26 -0600
Subject: OMAP3 clock: add a short delay when lowering CORE clk rate

When changing the SDRAM clock from 166MHz to 83MHz via the CORE DPLL M2
divider, add a short delay before returning to SDRAM to allow the SDRC
time to stabilize.  Without this delay, the system is prone to random
panics upon re-entering SDRAM.

This time delay varies based on MPU frequency.  At 500MHz MPU frequency at
room temperature, 64 loops seems to work okay; so add another 32 loops for
environmental and process variation.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
---
 arch/arm/mach-omap2/clock34xx.c        | 30 ++++++++++++++++++++++++++++--
 arch/arm/mach-omap2/sram34xx.S         | 20 +++++++++-----------
 arch/arm/plat-omap/include/mach/sram.h |  4 ++--
 arch/arm/plat-omap/sram.c              |  8 +++++---
 4 files changed, 44 insertions(+), 18 deletions(-)

diff --git a/arch/arm/mach-omap2/clock34xx.c b/arch/arm/mach-omap2/clock34xx.c
index 5458ab3bf65a..4bfa650bb34b 100644
--- a/arch/arm/mach-omap2/clock34xx.c
+++ b/arch/arm/mach-omap2/clock34xx.c
@@ -286,6 +286,20 @@ static struct omap_clk omap34xx_clks[] = {
 
 #define MIN_SDRC_DLL_LOCK_FREQ		83000000
 
+#define CYCLES_PER_MHZ			1000000
+
+/* Scale factor for fixed-point arith in omap3_core_dpll_m2_set_rate() */
+#define SDRC_MPURATE_SCALE		8
+
+/* 2^SDRC_MPURATE_BASE_SHIFT: MPU MHz that SDRC_MPURATE_LOOPS is defined for */
+#define SDRC_MPURATE_BASE_SHIFT		9
+
+/*
+ * SDRC_MPURATE_LOOPS: Number of MPU loops to execute at
+ * 2^MPURATE_BASE_SHIFT MHz for SDRC to stabilize
+ */
+#define SDRC_MPURATE_LOOPS		96
+
 /**
  * omap3_dpll_recalc - recalculate DPLL rate
  * @clk: DPLL struct clk
@@ -709,7 +723,8 @@ static int omap3_core_dpll_m2_set_rate(struct clk *clk, unsigned long rate)
 {
 	u32 new_div = 0;
 	u32 unlock_dll = 0;
-	unsigned long validrate, sdrcrate;
+	u32 c;
+	unsigned long validrate, sdrcrate, mpurate;
 	struct omap_sdrc_params *sp;
 
 	if (!clk || !rate)
@@ -737,6 +752,17 @@ static int omap3_core_dpll_m2_set_rate(struct clk *clk, unsigned long rate)
 		unlock_dll = 1;
 	}
 
+	/*
+	 * XXX This only needs to be done when the CPU frequency changes
+	 */
+	mpurate = arm_fck.rate / CYCLES_PER_MHZ;
+	c = (mpurate << SDRC_MPURATE_SCALE) >> SDRC_MPURATE_BASE_SHIFT;
+	c += 1;  /* for safety */
+	c *= SDRC_MPURATE_LOOPS;
+	c >>= SDRC_MPURATE_SCALE;
+	if (c == 0)
+		c = 1;
+
 	pr_debug("clock: changing CORE DPLL rate from %lu to %lu\n", clk->rate,
 		 validrate);
 	pr_debug("clock: SDRC timing params used: %08x %08x %08x\n",
@@ -747,7 +773,7 @@ static int omap3_core_dpll_m2_set_rate(struct clk *clk, unsigned long rate)
 
 	/* REVISIT: Add SDRC_MR changing to this code also */
 	omap3_configure_core_dpll(sp->rfr_ctrl, sp->actim_ctrla,
-				  sp->actim_ctrlb, new_div, unlock_dll);
+				  sp->actim_ctrlb, new_div, unlock_dll, c);
 
 	return 0;
 }
diff --git a/arch/arm/mach-omap2/sram34xx.S b/arch/arm/mach-omap2/sram34xx.S
index 84781a6cd263..8d4a88c30718 100644
--- a/arch/arm/mach-omap2/sram34xx.S
+++ b/arch/arm/mach-omap2/sram34xx.S
@@ -42,10 +42,14 @@
  * r0 = sdrc_rfr_ctrl r1 = sdrc_actim_ctrla r2 = sdrc_actim_ctrlb r3 = M2
  * r4 = Unlock SDRC DLL? (1 = yes, 0 = no) -- only unlock DLL for
  *      SDRC rates < 83MHz
+ * r5 = number of MPU cycles to wait for SDRC to stabilize after
+ *      reprogramming the SDRC when switching to a slower MPU speed
+ *
  */
 ENTRY(omap3_sram_configure_core_dpll)
 	stmfd	sp!, {r1-r12, lr}	@ store regs to stack
 	ldr	r4, [sp, #52]		@ pull extra args off the stack
+	ldr	r5, [sp, #56]		@ load extra args from the stack
 	dsb				@ flush buffered writes to interconnect
 	cmp	r3, #0x2
 	blne	configure_sdrc
@@ -59,7 +63,11 @@ ENTRY(omap3_sram_configure_core_dpll)
 	bleq	wait_dll_unlock
 	blne	wait_dll_lock
 	cmp	r3, #0x1
-	blne	configure_sdrc
+	beq	return_to_sdram
+	bl	configure_sdrc
+	mov	r12, r5			@ if slowing, wait for SDRC to stabilize
+	bl	wait_clk_stable
+return_to_sdram:
 	isb				@ prevent speculative exec past here
 	mov 	r0, #0 			@ return value
 	ldmfd	sp!, {r1-r12, pc}	@ restore regs and return
@@ -106,16 +114,6 @@ configure_core_dpll:
 wait_clk_stable:
 	subs 	r12, r12, #1
 	bne	wait_clk_stable
-	nop
-	nop
-	nop
-	nop
-	nop
-	nop
-	nop
-	nop
-	nop
-	nop
 	bx	lr
 enable_sdrc:
 	ldr 	r11, omap3_cm_iclken1_core
diff --git a/arch/arm/plat-omap/include/mach/sram.h b/arch/arm/plat-omap/include/mach/sram.h
index dca7c16ae903..c32fa0a220dc 100644
--- a/arch/arm/plat-omap/include/mach/sram.h
+++ b/arch/arm/plat-omap/include/mach/sram.h
@@ -24,7 +24,7 @@ extern u32 omap2_set_prcm(u32 dpll_ctrl_val, u32 sdrc_rfr_val, int bypass);
 extern u32 omap3_configure_core_dpll(u32 sdrc_rfr_ctrl,
 				     u32 sdrc_actim_ctrla,
 				     u32 sdrc_actim_ctrlb, u32 m2,
-				     u32 unlock_dll);
+				     u32 unlock_dll, u32 f);
 
 /* Do not use these */
 extern void omap1_sram_reprogram_clock(u32 ckctl, u32 dpllctl);
@@ -62,7 +62,7 @@ extern unsigned long omap243x_sram_reprogram_sdrc_sz;
 extern u32 omap3_sram_configure_core_dpll(u32 sdrc_rfr_ctrl,
 					  u32 sdrc_actim_ctrla,
 					  u32 sdrc_actim_ctrlb, u32 m2,
-					  u32 unlock_dll);
+					  u32 unlock_dll, u32 f);
 extern unsigned long omap3_sram_configure_core_dpll_sz;
 
 #endif
diff --git a/arch/arm/plat-omap/sram.c b/arch/arm/plat-omap/sram.c
index a5b9bcd6b108..79c0f0254426 100644
--- a/arch/arm/plat-omap/sram.c
+++ b/arch/arm/plat-omap/sram.c
@@ -371,15 +371,17 @@ static inline int omap243x_sram_init(void)
 static u32 (*_omap3_sram_configure_core_dpll)(u32 sdrc_rfr_ctrl,
 					      u32 sdrc_actim_ctrla,
 					      u32 sdrc_actim_ctrlb,
-					      u32 m2, u32 unlock_dll);
+					      u32 m2, u32 unlock_dll,
+					      u32 f);
 u32 omap3_configure_core_dpll(u32 sdrc_rfr_ctrl, u32 sdrc_actim_ctrla,
-			      u32 sdrc_actim_ctrlb, u32 m2, u32 unlock_dll)
+			      u32 sdrc_actim_ctrlb, u32 m2, u32 unlock_dll,
+			      u32 f)
 {
 	BUG_ON(!_omap3_sram_configure_core_dpll);
 	return _omap3_sram_configure_core_dpll(sdrc_rfr_ctrl,
 					       sdrc_actim_ctrla,
 					       sdrc_actim_ctrlb, m2,
-					       unlock_dll);
+					       unlock_dll, f);
 }
 
 /* REVISIT: Should this be same as omap34xx_sram_init() after off-idle? */
-- 
cgit v1.2.3-59-g8ed1b


From d0ba3922ae241a87d22a1c3ffad72b96fe993c9a Mon Sep 17 00:00:00 2001
From: Paul Walmsley <paul@pwsan.com>
Date: Fri, 19 Jun 2009 19:08:27 -0600
Subject: OMAP3 clock/SDRC: program SDRC_MR register during SDRC clock change

Program the SDRC_MR_0 register as well during SDRC clock changes.
This register allows selection of the memory CAS latency.  Some SDRAM
chips, such as the Qimonda HYB18M512160AF6, have a lower CAS latency
at lower clock rates.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
---
 arch/arm/mach-omap2/clock34xx.c        | 4 ++--
 arch/arm/mach-omap2/sram34xx.S         | 8 +++++++-
 arch/arm/plat-omap/include/mach/sram.h | 4 ++--
 arch/arm/plat-omap/sram.c              | 6 +++---
 4 files changed, 14 insertions(+), 8 deletions(-)

diff --git a/arch/arm/mach-omap2/clock34xx.c b/arch/arm/mach-omap2/clock34xx.c
index 4bfa650bb34b..cf41ab55fa97 100644
--- a/arch/arm/mach-omap2/clock34xx.c
+++ b/arch/arm/mach-omap2/clock34xx.c
@@ -771,9 +771,9 @@ static int omap3_core_dpll_m2_set_rate(struct clk *clk, unsigned long rate)
 	/* REVISIT: SRAM code doesn't support other M2 divisors yet */
 	WARN_ON(new_div != 1 && new_div != 2);
 
-	/* REVISIT: Add SDRC_MR changing to this code also */
 	omap3_configure_core_dpll(sp->rfr_ctrl, sp->actim_ctrla,
-				  sp->actim_ctrlb, new_div, unlock_dll, c);
+				  sp->actim_ctrlb, new_div, unlock_dll, c,
+				  sp->mr);
 
 	return 0;
 }
diff --git a/arch/arm/mach-omap2/sram34xx.S b/arch/arm/mach-omap2/sram34xx.S
index 8d4a88c30718..d13f1cc4bd58 100644
--- a/arch/arm/mach-omap2/sram34xx.S
+++ b/arch/arm/mach-omap2/sram34xx.S
@@ -44,12 +44,14 @@
  *      SDRC rates < 83MHz
  * r5 = number of MPU cycles to wait for SDRC to stabilize after
  *      reprogramming the SDRC when switching to a slower MPU speed
+ * r6 = SDRC_MR_0 register value
  *
  */
 ENTRY(omap3_sram_configure_core_dpll)
 	stmfd	sp!, {r1-r12, lr}	@ store regs to stack
 	ldr	r4, [sp, #52]		@ pull extra args off the stack
 	ldr	r5, [sp, #56]		@ load extra args from the stack
+	ldr	r6, [sp, #60]		@ load extra args from the stack
 	dsb				@ flush buffered writes to interconnect
 	cmp	r3, #0x2
 	blne	configure_sdrc
@@ -151,7 +153,9 @@ configure_sdrc:
 	str	r1, [r11]
 	ldr	r11, omap3_sdrc_actim_ctrlb
 	str	r2, [r11]
-	ldr	r2, [r11]		@ posted-write barrier for SDRC
+	ldr	r11, omap3_sdrc_mr_0
+	str	r6, [r11]
+	ldr	r6, [r11]		@ posted-write barrier for SDRC
 	bx	lr
 
 omap3_sdrc_power:
@@ -168,6 +172,8 @@ omap3_sdrc_actim_ctrla:
 	.word OMAP34XX_SDRC_REGADDR(SDRC_ACTIM_CTRL_A_0)
 omap3_sdrc_actim_ctrlb:
 	.word OMAP34XX_SDRC_REGADDR(SDRC_ACTIM_CTRL_B_0)
+omap3_sdrc_mr_0:
+	.word OMAP34XX_SDRC_REGADDR(SDRC_MR_0)
 omap3_sdrc_dlla_status:
 	.word OMAP34XX_SDRC_REGADDR(SDRC_DLLA_STATUS)
 omap3_sdrc_dlla_ctrl:
diff --git a/arch/arm/plat-omap/include/mach/sram.h b/arch/arm/plat-omap/include/mach/sram.h
index c32fa0a220dc..4f87056a3677 100644
--- a/arch/arm/plat-omap/include/mach/sram.h
+++ b/arch/arm/plat-omap/include/mach/sram.h
@@ -24,7 +24,7 @@ extern u32 omap2_set_prcm(u32 dpll_ctrl_val, u32 sdrc_rfr_val, int bypass);
 extern u32 omap3_configure_core_dpll(u32 sdrc_rfr_ctrl,
 				     u32 sdrc_actim_ctrla,
 				     u32 sdrc_actim_ctrlb, u32 m2,
-				     u32 unlock_dll, u32 f);
+				     u32 unlock_dll, u32 f, u32 sdrc_mr);
 
 /* Do not use these */
 extern void omap1_sram_reprogram_clock(u32 ckctl, u32 dpllctl);
@@ -62,7 +62,7 @@ extern unsigned long omap243x_sram_reprogram_sdrc_sz;
 extern u32 omap3_sram_configure_core_dpll(u32 sdrc_rfr_ctrl,
 					  u32 sdrc_actim_ctrla,
 					  u32 sdrc_actim_ctrlb, u32 m2,
-					  u32 unlock_dll, u32 f);
+					  u32 unlock_dll, u32 f, u32 sdrc_mr);
 extern unsigned long omap3_sram_configure_core_dpll_sz;
 
 #endif
diff --git a/arch/arm/plat-omap/sram.c b/arch/arm/plat-omap/sram.c
index 79c0f0254426..7dadf1a59fb5 100644
--- a/arch/arm/plat-omap/sram.c
+++ b/arch/arm/plat-omap/sram.c
@@ -372,16 +372,16 @@ static u32 (*_omap3_sram_configure_core_dpll)(u32 sdrc_rfr_ctrl,
 					      u32 sdrc_actim_ctrla,
 					      u32 sdrc_actim_ctrlb,
 					      u32 m2, u32 unlock_dll,
-					      u32 f);
+					      u32 f, u32 sdrc_mr);
 u32 omap3_configure_core_dpll(u32 sdrc_rfr_ctrl, u32 sdrc_actim_ctrla,
 			      u32 sdrc_actim_ctrlb, u32 m2, u32 unlock_dll,
-			      u32 f)
+			      u32 f, u32 sdrc_mr)
 {
 	BUG_ON(!_omap3_sram_configure_core_dpll);
 	return _omap3_sram_configure_core_dpll(sdrc_rfr_ctrl,
 					       sdrc_actim_ctrla,
 					       sdrc_actim_ctrlb, m2,
-					       unlock_dll, f);
+					       unlock_dll, f, sdrc_mr);
 }
 
 /* REVISIT: Should this be same as omap34xx_sram_init() after off-idle? */
-- 
cgit v1.2.3-59-g8ed1b


From 4267b5d15269ea6b26736a2ccd4c213e63e547ab Mon Sep 17 00:00:00 2001
From: Paul Walmsley <paul@pwsan.com>
Date: Fri, 19 Jun 2009 19:08:27 -0600
Subject: OMAP3 SRAM: add more comments on the SRAM code

Clean up comments and copyrights on the CORE DPLL3 M2 divider change code.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
---
 arch/arm/mach-omap2/sram34xx.S | 45 ++++++++++++++++++++++--------------------
 1 file changed, 24 insertions(+), 21 deletions(-)

diff --git a/arch/arm/mach-omap2/sram34xx.S b/arch/arm/mach-omap2/sram34xx.S
index d13f1cc4bd58..37a1e1fc0ab6 100644
--- a/arch/arm/mach-omap2/sram34xx.S
+++ b/arch/arm/mach-omap2/sram34xx.S
@@ -3,13 +3,12 @@
  *
  * Omap3 specific functions that need to be run in internal SRAM
  *
- * (C) Copyright 2007
- * Texas Instruments Inc.
- * Rajendra Nayak <rnayak@ti.com>
+ * Copyright (C) 2004, 2007, 2008 Texas Instruments, Inc.
+ * Copyright (C) 2008 Nokia Corporation
  *
- * (C) Copyright 2004
- * Texas Instruments, <www.ti.com>
+ * Rajendra Nayak <rnayak@ti.com>
  * Richard Woodruff <r-woodruff2@ti.com>
+ * Paul Walmsley
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License as
@@ -38,13 +37,16 @@
 	.text
 
 /*
- * Change frequency of core dpll
- * r0 = sdrc_rfr_ctrl r1 = sdrc_actim_ctrla r2 = sdrc_actim_ctrlb r3 = M2
- * r4 = Unlock SDRC DLL? (1 = yes, 0 = no) -- only unlock DLL for
+ * omap3_sram_configure_core_dpll - change DPLL3 M2 divider
+ * r0 = new SDRC_RFR_CTRL register contents
+ * r1 = new SDRC_ACTIM_CTRLA register contents
+ * r2 = new SDRC_ACTIM_CTRLB register contents
+ * r3 = new M2 divider setting (only 1 and 2 supported right now)
+ * r4 = unlock SDRC DLL? (1 = yes, 0 = no).  Only unlock DLL for
  *      SDRC rates < 83MHz
  * r5 = number of MPU cycles to wait for SDRC to stabilize after
  *      reprogramming the SDRC when switching to a slower MPU speed
- * r6 = SDRC_MR_0 register value
+ * r6 = new SDRC_MR_0 register value
  *
  */
 ENTRY(omap3_sram_configure_core_dpll)
@@ -53,22 +55,22 @@ ENTRY(omap3_sram_configure_core_dpll)
 	ldr	r5, [sp, #56]		@ load extra args from the stack
 	ldr	r6, [sp, #60]		@ load extra args from the stack
 	dsb				@ flush buffered writes to interconnect
-	cmp	r3, #0x2
-	blne	configure_sdrc
-	cmp	r4, #0x1
+	cmp	r3, #0x2		@ if increasing SDRC clk rate,
+	blne	configure_sdrc		@ program the SDRC regs early (for RFR)
+	cmp	r4, #0x1		@ set the intended DLL state
 	bleq	unlock_dll
 	blne	lock_dll
-	bl	sdram_in_selfrefresh	@ put the SDRAM in self refresh
-	bl 	configure_core_dpll
-	bl	enable_sdrc
-	cmp	r4, #0x1
+	bl	sdram_in_selfrefresh	@ put SDRAM in self refresh, idle SDRC
+	bl 	configure_core_dpll	@ change the DPLL3 M2 divider
+	bl	enable_sdrc		@ take SDRC out of idle
+	cmp	r4, #0x1		@ wait for DLL status to change
 	bleq	wait_dll_unlock
 	blne	wait_dll_lock
-	cmp	r3, #0x1
-	beq	return_to_sdram
-	bl	configure_sdrc
-	mov	r12, r5			@ if slowing, wait for SDRC to stabilize
-	bl	wait_clk_stable
+	cmp	r3, #0x1		@ if increasing SDRC clk rate,
+	beq	return_to_sdram		@ return to SDRAM code, otherwise,
+	bl	configure_sdrc		@ reprogram SDRC regs now
+	mov	r12, r5
+	bl	wait_clk_stable		@ wait for SDRC to stabilize
 return_to_sdram:
 	isb				@ prevent speculative exec past here
 	mov 	r0, #0 			@ return value
@@ -93,6 +95,7 @@ sdram_in_selfrefresh:
 	bic 	r12, r12, #0x4		@ clear PWDENA
 	str 	r12, [r11]		@ write back to SDRC_POWER register
 	ldr	r12, [r11]		@ posted-write barrier for SDRC
+idle_sdrc:
 	ldr	r11, omap3_cm_iclken1_core	@ read the CM_ICLKEN1_CORE reg
 	ldr	r12, [r11]
 	bic	r12, r12, #0x2		@ disable iclk bit for SDRC
-- 
cgit v1.2.3-59-g8ed1b


From df14e4747aa58126a508ae26661c73d83127c831 Mon Sep 17 00:00:00 2001
From: Paul Walmsley <paul@pwsan.com>
Date: Fri, 19 Jun 2009 19:08:28 -0600
Subject: OMAP3 SRAM: convert SRAM code to use macros rather than magic numbers

Convert omap3_sram_configure_core_dpll() to use macros rather than
magic numbers.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
---
 arch/arm/mach-omap2/sram34xx.S | 53 ++++++++++++++++++++++++++++++------------
 1 file changed, 38 insertions(+), 15 deletions(-)

diff --git a/arch/arm/mach-omap2/sram34xx.S b/arch/arm/mach-omap2/sram34xx.S
index 37a1e1fc0ab6..16eb4efa8b74 100644
--- a/arch/arm/mach-omap2/sram34xx.S
+++ b/arch/arm/mach-omap2/sram34xx.S
@@ -36,6 +36,29 @@
 
 	.text
 
+/* r4 parameters */
+#define SDRC_NO_UNLOCK_DLL		0x0
+#define SDRC_UNLOCK_DLL			0x1
+
+/* SDRC_DLLA_CTRL bit settings */
+#define DLLIDLE_MASK			0x4
+
+/* SDRC_DLLA_STATUS bit settings */
+#define LOCKSTATUS_MASK			0x4
+
+/* SDRC_POWER bit settings */
+#define SRFRONIDLEREQ_MASK		0x40
+#define PWDENA_MASK			0x4
+
+/* CM_IDLEST1_CORE bit settings */
+#define ST_SDRC_MASK			0x2
+
+/* CM_ICLKEN1_CORE bit settings */
+#define EN_SDRC_MASK			0x2
+
+/* CM_CLKSEL1_PLL bit settings */
+#define CORE_DPLL_CLKOUT_DIV_SHIFT	0x1b
+
 /*
  * omap3_sram_configure_core_dpll - change DPLL3 M2 divider
  * r0 = new SDRC_RFR_CTRL register contents
@@ -57,13 +80,13 @@ ENTRY(omap3_sram_configure_core_dpll)
 	dsb				@ flush buffered writes to interconnect
 	cmp	r3, #0x2		@ if increasing SDRC clk rate,
 	blne	configure_sdrc		@ program the SDRC regs early (for RFR)
-	cmp	r4, #0x1		@ set the intended DLL state
+	cmp	r4, #SDRC_UNLOCK_DLL	@ set the intended DLL state
 	bleq	unlock_dll
 	blne	lock_dll
 	bl	sdram_in_selfrefresh	@ put SDRAM in self refresh, idle SDRC
 	bl 	configure_core_dpll	@ change the DPLL3 M2 divider
 	bl	enable_sdrc		@ take SDRC out of idle
-	cmp	r4, #0x1		@ wait for DLL status to change
+	cmp	r4, #SDRC_UNLOCK_DLL	@ wait for DLL status to change
 	bleq	wait_dll_unlock
 	blne	wait_dll_lock
 	cmp	r3, #0x1		@ if increasing SDRC clk rate,
@@ -78,33 +101,33 @@ return_to_sdram:
 unlock_dll:
 	ldr	r11, omap3_sdrc_dlla_ctrl
 	ldr	r12, [r11]
-	orr	r12, r12, #0x4
+	orr	r12, r12, #DLLIDLE_MASK
 	str	r12, [r11]		@ (no OCP barrier needed)
 	bx	lr
 lock_dll:
 	ldr	r11, omap3_sdrc_dlla_ctrl
 	ldr	r12, [r11]
-	bic	r12, r12, #0x4
+	bic	r12, r12, #DLLIDLE_MASK
 	str	r12, [r11]		@ (no OCP barrier needed)
 	bx	lr
 sdram_in_selfrefresh:
 	ldr	r11, omap3_sdrc_power	@ read the SDRC_POWER register
 	ldr	r12, [r11]		@ read the contents of SDRC_POWER
 	mov	r9, r12			@ keep a copy of SDRC_POWER bits
-	orr 	r12, r12, #0x40		@ enable self refresh on idle req
-	bic 	r12, r12, #0x4		@ clear PWDENA
+	orr 	r12, r12, #SRFRONIDLEREQ_MASK	@ enable self refresh on idle
+	bic 	r12, r12, #PWDENA_MASK	@ clear PWDENA
 	str 	r12, [r11]		@ write back to SDRC_POWER register
 	ldr	r12, [r11]		@ posted-write barrier for SDRC
 idle_sdrc:
 	ldr	r11, omap3_cm_iclken1_core	@ read the CM_ICLKEN1_CORE reg
 	ldr	r12, [r11]
-	bic	r12, r12, #0x2		@ disable iclk bit for SDRC
+	bic	r12, r12, #EN_SDRC_MASK		@ disable iclk bit for SDRC
 	str 	r12, [r11]
 wait_sdrc_idle:
 	ldr 	r11, omap3_cm_idlest1_core
 	ldr 	r12, [r11]
-	and 	r12, r12, #0x2		@ check for SDRC idle
-	cmp 	r12, #2
+	and 	r12, r12, #ST_SDRC_MASK		@ check for SDRC idle
+	cmp 	r12, #ST_SDRC_MASK
 	bne 	wait_sdrc_idle
 	bx 	lr
 configure_core_dpll:
@@ -112,7 +135,7 @@ configure_core_dpll:
 	ldr	r12, [r11]
 	ldr	r10, core_m2_mask_val	@ modify m2 for core dpll
 	and	r12, r12, r10
-	orr	r12, r12, r3, lsl #0x1B	@ r3 contains the M2 val
+	orr	r12, r12, r3, lsl #CORE_DPLL_CLKOUT_DIV_SHIFT
 	str	r12, [r11]
 	ldr	r12, [r11]		@ posted-write barrier for CM
 	bx	lr
@@ -123,12 +146,12 @@ wait_clk_stable:
 enable_sdrc:
 	ldr 	r11, omap3_cm_iclken1_core
 	ldr	r12, [r11]
-	orr 	r12, r12, #0x2		@ enable iclk bit for SDRC
+	orr 	r12, r12, #EN_SDRC_MASK		@ enable iclk bit for SDRC
 	str 	r12, [r11]
 wait_sdrc_idle1:
 	ldr 	r11, omap3_cm_idlest1_core
 	ldr	r12, [r11]
-	and 	r12, r12, #0x2
+	and 	r12, r12, #ST_SDRC_MASK
 	cmp	r12, #0
 	bne	wait_sdrc_idle1
 restore_sdrc_power_val:
@@ -138,14 +161,14 @@ restore_sdrc_power_val:
 wait_dll_lock:
 	ldr	r11, omap3_sdrc_dlla_status
 	ldr	r12, [r11]
-	and 	r12, r12, #0x4
-	cmp	r12, #0x4
+	and 	r12, r12, #LOCKSTATUS_MASK
+	cmp	r12, #LOCKSTATUS_MASK
 	bne	wait_dll_lock
 	bx	lr
 wait_dll_unlock:
 	ldr	r11, omap3_sdrc_dlla_status
 	ldr	r12, [r11]
-	and	r12, r12, #0x4
+	and	r12, r12, #LOCKSTATUS_MASK
 	cmp	r12, #0x0
 	bne	wait_dll_unlock
 	bx	lr
-- 
cgit v1.2.3-59-g8ed1b


From 3afec6332e1e7cf2d74e0bf08160a68f43a59073 Mon Sep 17 00:00:00 2001
From: Tero Kristo <tero.kristo@nokia.com>
Date: Fri, 19 Jun 2009 19:08:29 -0600
Subject: OMAP3: Add support for DPLL3 divisor values higher than 2

Previously only 1 and 2 was supported. This is needed for DVFS VDD2 control.

Signed-off-by: Tero Kristo <tero.kristo@nokia.com>
---
 arch/arm/mach-omap2/clock34xx.c        | 9 +++------
 arch/arm/mach-omap2/sram34xx.S         | 8 +++++---
 arch/arm/plat-omap/include/mach/sram.h | 6 ++++--
 arch/arm/plat-omap/sram.c              | 6 +++---
 4 files changed, 15 insertions(+), 14 deletions(-)

diff --git a/arch/arm/mach-omap2/clock34xx.c b/arch/arm/mach-omap2/clock34xx.c
index cf41ab55fa97..045da923e75b 100644
--- a/arch/arm/mach-omap2/clock34xx.c
+++ b/arch/arm/mach-omap2/clock34xx.c
@@ -739,9 +739,9 @@ static int omap3_core_dpll_m2_set_rate(struct clk *clk, unsigned long rate)
 
 	sdrcrate = sdrc_ick.rate;
 	if (rate > clk->rate)
-		sdrcrate <<= ((rate / clk->rate) - 1);
+		sdrcrate <<= ((rate / clk->rate) >> 1);
 	else
-		sdrcrate >>= ((clk->rate / rate) - 1);
+		sdrcrate >>= ((clk->rate / rate) >> 1);
 
 	sp = omap2_sdrc_get_params(sdrcrate);
 	if (!sp)
@@ -768,12 +768,9 @@ static int omap3_core_dpll_m2_set_rate(struct clk *clk, unsigned long rate)
 	pr_debug("clock: SDRC timing params used: %08x %08x %08x\n",
 		 sp->rfr_ctrl, sp->actim_ctrla, sp->actim_ctrlb);
 
-	/* REVISIT: SRAM code doesn't support other M2 divisors yet */
-	WARN_ON(new_div != 1 && new_div != 2);
-
 	omap3_configure_core_dpll(sp->rfr_ctrl, sp->actim_ctrla,
 				  sp->actim_ctrlb, new_div, unlock_dll, c,
-				  sp->mr);
+				  sp->mr, rate > clk->rate);
 
 	return 0;
 }
diff --git a/arch/arm/mach-omap2/sram34xx.S b/arch/arm/mach-omap2/sram34xx.S
index 16eb4efa8b74..487fa8609cde 100644
--- a/arch/arm/mach-omap2/sram34xx.S
+++ b/arch/arm/mach-omap2/sram34xx.S
@@ -70,6 +70,7 @@
  * r5 = number of MPU cycles to wait for SDRC to stabilize after
  *      reprogramming the SDRC when switching to a slower MPU speed
  * r6 = new SDRC_MR_0 register value
+ * r7 = increasing SDRC rate? (1 = yes, 0 = no)
  *
  */
 ENTRY(omap3_sram_configure_core_dpll)
@@ -77,9 +78,10 @@ ENTRY(omap3_sram_configure_core_dpll)
 	ldr	r4, [sp, #52]		@ pull extra args off the stack
 	ldr	r5, [sp, #56]		@ load extra args from the stack
 	ldr	r6, [sp, #60]		@ load extra args from the stack
+	ldr	r7, [sp, #64]		@ load extra args from the stack
 	dsb				@ flush buffered writes to interconnect
-	cmp	r3, #0x2		@ if increasing SDRC clk rate,
-	blne	configure_sdrc		@ program the SDRC regs early (for RFR)
+	cmp	r7, #1			@ if increasing SDRC clk rate,
+	bleq	configure_sdrc		@ program the SDRC regs early (for RFR)
 	cmp	r4, #SDRC_UNLOCK_DLL	@ set the intended DLL state
 	bleq	unlock_dll
 	blne	lock_dll
@@ -89,7 +91,7 @@ ENTRY(omap3_sram_configure_core_dpll)
 	cmp	r4, #SDRC_UNLOCK_DLL	@ wait for DLL status to change
 	bleq	wait_dll_unlock
 	blne	wait_dll_lock
-	cmp	r3, #0x1		@ if increasing SDRC clk rate,
+	cmp	r7, #1			@ if increasing SDRC clk rate,
 	beq	return_to_sdram		@ return to SDRAM code, otherwise,
 	bl	configure_sdrc		@ reprogram SDRC regs now
 	mov	r12, r5
diff --git a/arch/arm/plat-omap/include/mach/sram.h b/arch/arm/plat-omap/include/mach/sram.h
index 4f87056a3677..4d53cc59d7a3 100644
--- a/arch/arm/plat-omap/include/mach/sram.h
+++ b/arch/arm/plat-omap/include/mach/sram.h
@@ -24,7 +24,8 @@ extern u32 omap2_set_prcm(u32 dpll_ctrl_val, u32 sdrc_rfr_val, int bypass);
 extern u32 omap3_configure_core_dpll(u32 sdrc_rfr_ctrl,
 				     u32 sdrc_actim_ctrla,
 				     u32 sdrc_actim_ctrlb, u32 m2,
-				     u32 unlock_dll, u32 f, u32 sdrc_mr);
+				     u32 unlock_dll, u32 f, u32 sdrc_mr,
+				     u32 inc);
 
 /* Do not use these */
 extern void omap1_sram_reprogram_clock(u32 ckctl, u32 dpllctl);
@@ -62,7 +63,8 @@ extern unsigned long omap243x_sram_reprogram_sdrc_sz;
 extern u32 omap3_sram_configure_core_dpll(u32 sdrc_rfr_ctrl,
 					  u32 sdrc_actim_ctrla,
 					  u32 sdrc_actim_ctrlb, u32 m2,
-					  u32 unlock_dll, u32 f, u32 sdrc_mr);
+					  u32 unlock_dll, u32 f, u32 sdrc_mr,
+					  u32 inc);
 extern unsigned long omap3_sram_configure_core_dpll_sz;
 
 #endif
diff --git a/arch/arm/plat-omap/sram.c b/arch/arm/plat-omap/sram.c
index 7dadf1a59fb5..65006df3f1b7 100644
--- a/arch/arm/plat-omap/sram.c
+++ b/arch/arm/plat-omap/sram.c
@@ -372,16 +372,16 @@ static u32 (*_omap3_sram_configure_core_dpll)(u32 sdrc_rfr_ctrl,
 					      u32 sdrc_actim_ctrla,
 					      u32 sdrc_actim_ctrlb,
 					      u32 m2, u32 unlock_dll,
-					      u32 f, u32 sdrc_mr);
+					      u32 f, u32 sdrc_mr, u32 inc);
 u32 omap3_configure_core_dpll(u32 sdrc_rfr_ctrl, u32 sdrc_actim_ctrla,
 			      u32 sdrc_actim_ctrlb, u32 m2, u32 unlock_dll,
-			      u32 f, u32 sdrc_mr)
+			      u32 f, u32 sdrc_mr, u32 inc)
 {
 	BUG_ON(!_omap3_sram_configure_core_dpll);
 	return _omap3_sram_configure_core_dpll(sdrc_rfr_ctrl,
 					       sdrc_actim_ctrla,
 					       sdrc_actim_ctrlb, m2,
-					       unlock_dll, f, sdrc_mr);
+					       unlock_dll, f, sdrc_mr, inc);
 }
 
 /* REVISIT: Should this be same as omap34xx_sram_init() after off-idle? */
-- 
cgit v1.2.3-59-g8ed1b


From 7b7bcefa35d62fec64c3615a6bef7866f34c7cc9 Mon Sep 17 00:00:00 2001
From: Paul Walmsley <paul@pwsan.com>
Date: Fri, 19 Jun 2009 19:08:29 -0600
Subject: OMAP3 SDRC: set FIXEDDELAY when disabling SDRC DLL

Correspondence with the TI OMAP hardware team indicates that
SDRC_DLLA_CTRL.FIXEDDELAY should be initialized to 0x0f.  This number
was apparently derived from process validation.  This is only used
when the SDRC DLL is unlocked (e.g., SDRC clock frequency less than
83MHz).

Signed-off-by: Paul Walmsley <paul@pwsan.com>
---
 arch/arm/mach-omap2/sram34xx.S | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/arch/arm/mach-omap2/sram34xx.S b/arch/arm/mach-omap2/sram34xx.S
index 487fa8609cde..f41f8d96ddba 100644
--- a/arch/arm/mach-omap2/sram34xx.S
+++ b/arch/arm/mach-omap2/sram34xx.S
@@ -41,8 +41,18 @@
 #define SDRC_UNLOCK_DLL			0x1
 
 /* SDRC_DLLA_CTRL bit settings */
+#define FIXEDDELAY_SHIFT		24
+#define FIXEDDELAY_MASK			(0xff << FIXEDDELAY_SHIFT)
 #define DLLIDLE_MASK			0x4
 
+/*
+ * SDRC_DLLA_CTRL default values: TI hardware team indicates that
+ * FIXEDDELAY should be initialized to 0xf.  This apparently was
+ * empirically determined during process testing, so no derivation
+ * was provided.
+ */
+#define FIXEDDELAY_DEFAULT		(0x0f << FIXEDDELAY_SHIFT)
+
 /* SDRC_DLLA_STATUS bit settings */
 #define LOCKSTATUS_MASK			0x4
 
@@ -103,6 +113,8 @@ return_to_sdram:
 unlock_dll:
 	ldr	r11, omap3_sdrc_dlla_ctrl
 	ldr	r12, [r11]
+	and	r12, r12, #FIXEDDELAY_MASK
+	orr	r12, r12, #FIXEDDELAY_DEFAULT
 	orr	r12, r12, #DLLIDLE_MASK
 	str	r12, [r11]		@ (no OCP barrier needed)
 	bx	lr
-- 
cgit v1.2.3-59-g8ed1b


From 2687069f3ac297b820c58de7222e4d16adbca498 Mon Sep 17 00:00:00 2001
From: Roel Kluin <roel.kluin@gmail.com>
Date: Fri, 19 Jun 2009 19:08:30 -0600
Subject: OMAP2 clock/powerdomain: off by 1 error in loop timeout comparisons

with while (i++ < MAX_CLOCK_ENABLE_WAIT); i can reach MAX_CLOCK_ENABLE_WAIT + 1
after the loop, so if (i == MAX_CLOCK_ENABLE_WAIT) that's still success.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Paul Walmsley <paul@pwsan.com>
---
 arch/arm/mach-omap2/clock.c       | 2 +-
 arch/arm/mach-omap2/powerdomain.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-omap2/clock.c b/arch/arm/mach-omap2/clock.c
index ba528f85749c..b0665f161c03 100644
--- a/arch/arm/mach-omap2/clock.c
+++ b/arch/arm/mach-omap2/clock.c
@@ -302,7 +302,7 @@ int omap2_wait_clock_ready(void __iomem *reg, u32 mask, const char *name)
 		udelay(1);
 	}
 
-	if (i < MAX_CLOCK_ENABLE_WAIT)
+	if (i <= MAX_CLOCK_ENABLE_WAIT)
 		pr_debug("Clock %s stable after %d loops\n", name, i);
 	else
 		printk(KERN_ERR "Clock %s didn't enable in %d tries\n",
diff --git a/arch/arm/mach-omap2/powerdomain.c b/arch/arm/mach-omap2/powerdomain.c
index 73e2971b1757..983f1cb676be 100644
--- a/arch/arm/mach-omap2/powerdomain.c
+++ b/arch/arm/mach-omap2/powerdomain.c
@@ -1099,7 +1099,7 @@ int pwrdm_wait_transition(struct powerdomain *pwrdm)
 	       (c++ < PWRDM_TRANSITION_BAILOUT))
 		udelay(1);
 
-	if (c >= PWRDM_TRANSITION_BAILOUT) {
+	if (c > PWRDM_TRANSITION_BAILOUT) {
 		printk(KERN_ERR "powerdomain: waited too long for "
 		       "powerdomain %s to complete transition\n", pwrdm->name);
 		return -EAGAIN;
-- 
cgit v1.2.3-59-g8ed1b


From fbe2b31b4b6dfa790cbc88e00631f3112c4fc54e Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 18 Jun 2009 14:46:47 -0600
Subject: ACPI: pci_root: check _CRS, then _BBN for downstream bus number

To find a host bridge's downstream bus number, we currently look at _BBN
first.  If _BBN returns a bus number we've already seen, we conclude that
_BBN was wrong and look for a bus number in _CRS.

However, the spec[1] (figure 5-5 and the example in sec 9.12.1) and an ACPI
FAQ[2] suggest that the OS should use _CRS to discover the bus number
range, and that _BBN is really intended to bootstrap _CRS methods that
reference PCI opregions.

This patch makes us always look at _CRS first.  If _CRS doesn't supply a
bus number, we look at _BBN.  If _BBN doesn't exist, we default to zero.
This makes the behavior consistent regardless of device discovery order.
Previously, if A and B had duplicate _BBNs and we found A first, we'd only
look at B's _CRS, whereas if we found B first, we'd only look at A's _CRS.

I'm told that Windows discovers host bridge bus numbers using _CRS, so
it should be fairly safe to rely on this BIOS functionality.

This patch also removes two misleading messages: we printed the "Wrong _BBN
value, reboot and use option 'pci=noacpi'" message before looking at _CRS,
so we would likely find the bus number in _CRS, the system would work fine,
and the user would be confused.  The "PCI _CRS %d overrides _BBN 0" message
incorrectly assumes _BBN was zero, and it's useless anyway because we
print the segment/bus number a few lines later.

References:
    [1] http://www.acpi.info/DOWNLOADS/ACPIspec30b.pdf
    [2] http://www.acpi.info/acpi_faq.htm _BBN/_CRS discussion
    http://download.microsoft.com/download/9/8/f/98f3fe47-dfc3-4e74-92a3-088782200fe7/TWAR05005_WinHEC05.ppt (slide 17)
    http://bugzilla.kernel.org/show_bug.cgi?id=1662 ASUS PR-DLS
    http://bugzilla.kernel.org/show_bug.cgi?id=1127 ASUS PR-DLSW
    http://bugzilla.kernel.org/show_bug.cgi?id=1741 ASUS PR-DLS533

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Reviewed-by: Alex Chiang <achiang@hp.com>
CC: Shaohua Li <shaohua.li@intel.com>
CC: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_root.c | 54 +++++++++++++++----------------------------------
 1 file changed, 16 insertions(+), 38 deletions(-)

diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 196f97d00956..a8b250783937 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -365,12 +365,12 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 {
 	int result = 0;
 	struct acpi_pci_root *root = NULL;
-	struct acpi_pci_root *tmp;
 	acpi_status status = AE_OK;
 	unsigned long long value = 0;
 	acpi_handle handle = NULL;
 	struct acpi_device *child;
 	u32 flags, base_flags;
+	int bus;
 
 
 	if (!device)
@@ -420,46 +420,24 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	/* 
 	 * Bus
 	 * ---
-	 * Obtained via _BBN, if exists, otherwise assumed to be zero (0).
+	 * Check _CRS first, then _BBN.  If no _BBN, default to zero.
 	 */
-	status = acpi_evaluate_integer(device->handle, METHOD_NAME__BBN, NULL,
-				       &value);
-	switch (status) {
-	case AE_OK:
-		root->id.bus = (u16) value;
-		break;
-	case AE_NOT_FOUND:
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Assuming bus 0 (no _BBN)\n"));
-		root->id.bus = 0;
-		break;
-	default:
-		ACPI_EXCEPTION((AE_INFO, status, "Evaluating _BBN"));
-		result = -ENODEV;
-		goto end;
-	}
-
-	/* Some systems have wrong _BBN */
-	list_for_each_entry(tmp, &acpi_pci_roots, node) {
-		if ((tmp->id.segment == root->id.segment)
-		    && (tmp->id.bus == root->id.bus)) {
-			int bus = 0;
-			acpi_status status;
-
-			printk(KERN_ERR PREFIX
-				    "Wrong _BBN value, reboot"
-				    " and use option 'pci=noacpi'\n");
-
-			status = try_get_root_bridge_busnr(device->handle, &bus);
-			if (ACPI_FAILURE(status))
-				break;
-			if (bus != root->id.bus) {
-				printk(KERN_INFO PREFIX
-				       "PCI _CRS %d overrides _BBN 0\n", bus);
-				root->id.bus = bus;
-			}
-			break;
+	status = try_get_root_bridge_busnr(device->handle, &bus);
+	if (ACPI_SUCCESS(status))
+		root->id.bus = bus;
+	else {
+		status = acpi_evaluate_integer(device->handle, METHOD_NAME__BBN,					       NULL, &value);
+		if (ACPI_SUCCESS(status))
+			root->id.bus = (u16) value;
+		else if (status == AE_NOT_FOUND)
+			root->id.bus = 0;
+		else {
+			ACPI_EXCEPTION((AE_INFO, status, "Evaluating _BBN"));
+			result = -ENODEV;
+			goto end;
 		}
 	}
+
 	/*
 	 * Device & Function
 	 * -----------------
-- 
cgit v1.2.3-59-g8ed1b


From f5eebbe119a861b5e4f5c67c886eab0937c686ed Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 18 Jun 2009 14:46:52 -0600
Subject: ACPI: pci_root: simplify acpi_pci_root_add() control flow

By looking up the segment & bus number earlier, we don't have to
worry about cleaning up if it fails.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_root.c | 100 ++++++++++++++++++------------------------------
 1 file changed, 38 insertions(+), 62 deletions(-)

diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index a8b250783937..0d69c0348c58 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -161,19 +161,22 @@ get_root_bridge_busnr_callback(struct acpi_resource *resource, void *data)
 	return AE_OK;
 }
 
-static acpi_status try_get_root_bridge_busnr(acpi_handle handle, int *busnum)
+static acpi_status try_get_root_bridge_busnr(acpi_handle handle,
+					     unsigned long long *bus)
 {
 	acpi_status status;
+	int busnum;
 
-	*busnum = -1;
+	busnum = -1;
 	status =
 	    acpi_walk_resources(handle, METHOD_NAME__CRS,
-				get_root_bridge_busnr_callback, busnum);
+				get_root_bridge_busnr_callback, &busnum);
 	if (ACPI_FAILURE(status))
 		return status;
 	/* Check if we really get a bus number from _CRS */
-	if (*busnum == -1)
+	if (busnum == -1)
 		return AE_ERROR;
+	*bus = busnum;
 	return AE_OK;
 }
 
@@ -363,24 +366,39 @@ EXPORT_SYMBOL(acpi_pci_osc_control_set);
 
 static int __devinit acpi_pci_root_add(struct acpi_device *device)
 {
-	int result = 0;
-	struct acpi_pci_root *root = NULL;
-	acpi_status status = AE_OK;
-	unsigned long long value = 0;
-	acpi_handle handle = NULL;
+	unsigned long long segment, bus;
+	acpi_status status;
+	int result;
+	struct acpi_pci_root *root;
+	acpi_handle handle;
 	struct acpi_device *child;
 	u32 flags, base_flags;
-	int bus;
 
+	segment = 0;
+	status = acpi_evaluate_integer(device->handle, METHOD_NAME__SEG, NULL,
+				       &segment);
+	if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) {
+		printk(KERN_ERR PREFIX "can't evaluate _SEG\n");
+		return -ENODEV;
+	}
 
-	if (!device)
-		return -EINVAL;
+	/* Check _CRS first, then _BBN.  If no _BBN, default to zero. */
+	bus = 0;
+	status = try_get_root_bridge_busnr(device->handle, &bus);
+	if (ACPI_FAILURE(status)) {
+		status = acpi_evaluate_integer(device->handle, METHOD_NAME__BBN,					       NULL, &bus);
+		if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) {
+			printk(KERN_ERR PREFIX
+			     "no bus number in _CRS and can't evaluate _BBN\n");
+			return -ENODEV;
+		}
+	}
 
 	root = kzalloc(sizeof(struct acpi_pci_root), GFP_KERNEL);
 	if (!root)
 		return -ENOMEM;
-	INIT_LIST_HEAD(&root->node);
 
+	INIT_LIST_HEAD(&root->node);
 	root->device = device;
 	strcpy(acpi_device_name(device), ACPI_PCI_ROOT_DEVICE_NAME);
 	strcpy(acpi_device_class(device), ACPI_PCI_ROOT_CLASS);
@@ -395,54 +413,13 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	flags = base_flags = OSC_PCI_SEGMENT_GROUPS_SUPPORT;
 	acpi_pci_osc_support(root, flags);
 
-	/* 
-	 * Segment
-	 * -------
-	 * Obtained via _SEG, if exists, otherwise assumed to be zero (0).
-	 */
-	status = acpi_evaluate_integer(device->handle, METHOD_NAME__SEG, NULL,
-				       &value);
-	switch (status) {
-	case AE_OK:
-		root->id.segment = (u16) value;
-		break;
-	case AE_NOT_FOUND:
-		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-				  "Assuming segment 0 (no _SEG)\n"));
-		root->id.segment = 0;
-		break;
-	default:
-		ACPI_EXCEPTION((AE_INFO, status, "Evaluating _SEG"));
-		result = -ENODEV;
-		goto end;
-	}
-
-	/* 
-	 * Bus
-	 * ---
-	 * Check _CRS first, then _BBN.  If no _BBN, default to zero.
-	 */
-	status = try_get_root_bridge_busnr(device->handle, &bus);
-	if (ACPI_SUCCESS(status))
-		root->id.bus = bus;
-	else {
-		status = acpi_evaluate_integer(device->handle, METHOD_NAME__BBN,					       NULL, &value);
-		if (ACPI_SUCCESS(status))
-			root->id.bus = (u16) value;
-		else if (status == AE_NOT_FOUND)
-			root->id.bus = 0;
-		else {
-			ACPI_EXCEPTION((AE_INFO, status, "Evaluating _BBN"));
-			result = -ENODEV;
-			goto end;
-		}
-	}
-
 	/*
 	 * Device & Function
 	 * -----------------
 	 * Obtained from _ADR (which has already been evaluated for us).
 	 */
+	root->id.segment = segment & 0xFFFF;
+	root->id.bus = bus & 0xFF;
 	root->id.device = device->pnp.bus_address >> 16;
 	root->id.function = device->pnp.bus_address & 0xFFFF;
 
@@ -509,13 +486,12 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	if (flags != base_flags)
 		acpi_pci_osc_support(root, flags);
 
-      end:
-	if (result) {
-		if (!list_empty(&root->node))
-			list_del(&root->node);
-		kfree(root);
-	}
+	return 0;
 
+end:
+	if (!list_empty(&root->node))
+		list_del(&root->node);
+	kfree(root);
 	return result;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From caf420c68afe01acd7c458ce40b85b3db5330ff5 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 18 Jun 2009 14:46:57 -0600
Subject: ACPI: pci_root: use driver data rather than list lookup

There's no need to search the list to find the acpi_pci_root
structure.  We saved it as device->driver_data when we added
the device.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_root.c | 21 ++++-----------------
 1 file changed, 4 insertions(+), 17 deletions(-)

diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 0d69c0348c58..7984e00540fa 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -497,30 +497,17 @@ end:
 
 static int acpi_pci_root_start(struct acpi_device *device)
 {
-	struct acpi_pci_root *root;
-
+	struct acpi_pci_root *root = acpi_driver_data(device);
 
-	list_for_each_entry(root, &acpi_pci_roots, node) {
-		if (root->device == device) {
-			pci_bus_add_devices(root->bus);
-			return 0;
-		}
-	}
-	return -ENODEV;
+	pci_bus_add_devices(root->bus);
+	return 0;
 }
 
 static int acpi_pci_root_remove(struct acpi_device *device, int type)
 {
-	struct acpi_pci_root *root = NULL;
-
-
-	if (!device || !acpi_driver_data(device))
-		return -EINVAL;
-
-	root = acpi_driver_data(device);
+	struct acpi_pci_root *root = acpi_driver_data(device);
 
 	kfree(root);
-
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From c1aec8341627dad5d63cc24aa6746dc077f5b706 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 18 Jun 2009 14:47:02 -0600
Subject: ACPI: pci_root: simplify list traversals

Using list_for_each_entry() makes traversing the root list easier.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Reviewed-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_root.c | 23 +++++++++--------------
 1 file changed, 9 insertions(+), 14 deletions(-)

diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 7984e00540fa..4fb747205418 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -82,7 +82,7 @@ static DEFINE_MUTEX(osc_lock);
 int acpi_pci_register_driver(struct acpi_pci_driver *driver)
 {
 	int n = 0;
-	struct list_head *entry;
+	struct acpi_pci_root *root;
 
 	struct acpi_pci_driver **pptr = &sub_driver;
 	while (*pptr)
@@ -92,9 +92,7 @@ int acpi_pci_register_driver(struct acpi_pci_driver *driver)
 	if (!driver->add)
 		return 0;
 
-	list_for_each(entry, &acpi_pci_roots) {
-		struct acpi_pci_root *root;
-		root = list_entry(entry, struct acpi_pci_root, node);
+	list_for_each_entry(root, &acpi_pci_roots, node) {
 		driver->add(root->device->handle);
 		n++;
 	}
@@ -106,7 +104,7 @@ EXPORT_SYMBOL(acpi_pci_register_driver);
 
 void acpi_pci_unregister_driver(struct acpi_pci_driver *driver)
 {
-	struct list_head *entry;
+	struct acpi_pci_root *root;
 
 	struct acpi_pci_driver **pptr = &sub_driver;
 	while (*pptr) {
@@ -120,23 +118,19 @@ void acpi_pci_unregister_driver(struct acpi_pci_driver *driver)
 	if (!driver->remove)
 		return;
 
-	list_for_each(entry, &acpi_pci_roots) {
-		struct acpi_pci_root *root;
-		root = list_entry(entry, struct acpi_pci_root, node);
+	list_for_each_entry(root, &acpi_pci_roots, node)
 		driver->remove(root->device->handle);
-	}
 }
 
 EXPORT_SYMBOL(acpi_pci_unregister_driver);
 
 acpi_handle acpi_get_pci_rootbridge_handle(unsigned int seg, unsigned int bus)
 {
-	struct acpi_pci_root *tmp;
+	struct acpi_pci_root *root;
 	
-	list_for_each_entry(tmp, &acpi_pci_roots, node) {
-		if ((tmp->id.segment == (u16) seg) && (tmp->id.bus == (u16) bus))
-			return tmp->device->handle;
-	}
+	list_for_each_entry(root, &acpi_pci_roots, node)
+		if ((root->id.segment == (u16) seg) && (root->id.bus == (u16) bus))
+			return root->device->handle;
 	return NULL;		
 }
 
@@ -301,6 +295,7 @@ static acpi_status acpi_pci_osc_support(struct acpi_pci_root *root, u32 flags)
 static struct acpi_pci_root *acpi_pci_find_root(acpi_handle handle)
 {
 	struct acpi_pci_root *root;
+
 	list_for_each_entry(root, &acpi_pci_roots, node) {
 		if (root->device->handle == handle)
 			return root;
-- 
cgit v1.2.3-59-g8ed1b


From 0705495d9010048e293013d9d129cf723363a0a8 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Thu, 18 Jun 2009 14:47:07 -0600
Subject: ACPI: pci_root: remove unused dev/fn information

We never use the PCI device & function number, so remove it to make
it clear that it's not needed.  Many PCI host bridges don't even
appear in config space, so it's meaningless to look at stuff from
_ADR, which doesn't exist in that case.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Reviewed-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_root.c | 25 +++++++++----------------
 1 file changed, 9 insertions(+), 16 deletions(-)

diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 4fb747205418..e95b5ac2e609 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -63,9 +63,10 @@ static struct acpi_driver acpi_pci_root_driver = {
 
 struct acpi_pci_root {
 	struct list_head node;
-	struct acpi_device * device;
-	struct acpi_pci_id id;
+	struct acpi_device *device;
 	struct pci_bus *bus;
+	u16 segment;
+	u8 bus_nr;
 
 	u32 osc_support_set;	/* _OSC state of support bits */
 	u32 osc_control_set;	/* _OSC state of control bits */
@@ -129,7 +130,7 @@ acpi_handle acpi_get_pci_rootbridge_handle(unsigned int seg, unsigned int bus)
 	struct acpi_pci_root *root;
 	
 	list_for_each_entry(root, &acpi_pci_roots, node)
-		if ((root->id.segment == (u16) seg) && (root->id.bus == (u16) bus))
+		if ((root->segment == (u16) seg) && (root->bus_nr == (u16) bus))
 			return root->device->handle;
 	return NULL;		
 }
@@ -395,6 +396,8 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 
 	INIT_LIST_HEAD(&root->node);
 	root->device = device;
+	root->segment = segment & 0xFFFF;
+	root->bus_nr = bus & 0xFF;
 	strcpy(acpi_device_name(device), ACPI_PCI_ROOT_DEVICE_NAME);
 	strcpy(acpi_device_class(device), ACPI_PCI_ROOT_CLASS);
 	device->driver_data = root;
@@ -408,16 +411,6 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	flags = base_flags = OSC_PCI_SEGMENT_GROUPS_SUPPORT;
 	acpi_pci_osc_support(root, flags);
 
-	/*
-	 * Device & Function
-	 * -----------------
-	 * Obtained from _ADR (which has already been evaluated for us).
-	 */
-	root->id.segment = segment & 0xFFFF;
-	root->id.bus = bus & 0xFF;
-	root->id.device = device->pnp.bus_address >> 16;
-	root->id.function = device->pnp.bus_address & 0xFFFF;
-
 	/*
 	 * TBD: Need PCI interface for enumeration/configuration of roots.
 	 */
@@ -427,7 +420,7 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 
 	printk(KERN_INFO PREFIX "%s [%s] (%04x:%02x)\n",
 	       acpi_device_name(device), acpi_device_bid(device),
-	       root->id.segment, root->id.bus);
+	       root->segment, root->bus_nr);
 
 	/*
 	 * Scan the Root Bridge
@@ -436,11 +429,11 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device)
 	 * PCI namespace does not get created until this call is made (and 
 	 * thus the root bridge's pci_dev does not exist).
 	 */
-	root->bus = pci_acpi_scan_root(device, root->id.segment, root->id.bus);
+	root->bus = pci_acpi_scan_root(device, segment, bus);
 	if (!root->bus) {
 		printk(KERN_ERR PREFIX
 			    "Bus %04x:%02x not present in PCI namespace\n",
-			    root->id.segment, root->id.bus);
+			    root->segment, root->bus_nr);
 		result = -ENODEV;
 		goto end;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 7b768f07dce463a054c9dd84862d15ccc3d2b712 Mon Sep 17 00:00:00 2001
From: "Pallipadi, Venkatesh" <venkatesh.pallipadi@intel.com>
Date: Fri, 19 Jun 2009 17:14:59 -0700
Subject: ACPI: pdc init related memory leak with physical CPU hotplug

arch_acpi_processor_cleanup_pdc() in x86 and ia64 results in memory allocated
for _PDC objects that is never freed and will cause memory leak in case of
physical CPU remove and add. Patch fixes the memory leak by freeing the
objects soon after _PDC is evaluated.

Reported-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 arch/ia64/kernel/acpi-processor.c | 12 ++++++++++++
 arch/x86/kernel/acpi/processor.c  | 13 +++++++++++++
 drivers/acpi/processor_core.c     |  2 ++
 include/acpi/processor.h          |  1 +
 4 files changed, 28 insertions(+)

diff --git a/arch/ia64/kernel/acpi-processor.c b/arch/ia64/kernel/acpi-processor.c
index cbe6cee5a550..dbda7bde6112 100644
--- a/arch/ia64/kernel/acpi-processor.c
+++ b/arch/ia64/kernel/acpi-processor.c
@@ -71,3 +71,15 @@ void arch_acpi_processor_init_pdc(struct acpi_processor *pr)
 }
 
 EXPORT_SYMBOL(arch_acpi_processor_init_pdc);
+
+void arch_acpi_processor_cleanup_pdc(struct acpi_processor *pr)
+{
+	if (pr->pdc) {
+		kfree(pr->pdc->pointer->buffer.pointer);
+		kfree(pr->pdc->pointer);
+		kfree(pr->pdc);
+		pr->pdc = NULL;
+	}
+}
+
+EXPORT_SYMBOL(arch_acpi_processor_cleanup_pdc);
diff --git a/arch/x86/kernel/acpi/processor.c b/arch/x86/kernel/acpi/processor.c
index 7c074eec39fb..d296f4a195c9 100644
--- a/arch/x86/kernel/acpi/processor.c
+++ b/arch/x86/kernel/acpi/processor.c
@@ -72,6 +72,7 @@ static void init_intel_pdc(struct acpi_processor *pr, struct cpuinfo_x86 *c)
 	return;
 }
 
+
 /* Initialize _PDC data based on the CPU vendor */
 void arch_acpi_processor_init_pdc(struct acpi_processor *pr)
 {
@@ -85,3 +86,15 @@ void arch_acpi_processor_init_pdc(struct acpi_processor *pr)
 }
 
 EXPORT_SYMBOL(arch_acpi_processor_init_pdc);
+
+void arch_acpi_processor_cleanup_pdc(struct acpi_processor *pr)
+{
+	if (pr->pdc) {
+		kfree(pr->pdc->pointer->buffer.pointer);
+		kfree(pr->pdc->pointer);
+		kfree(pr->pdc);
+		pr->pdc = NULL;
+	}
+}
+
+EXPORT_SYMBOL(arch_acpi_processor_cleanup_pdc);
diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c
index 23f0fb84f1c1..d40d45e904a5 100644
--- a/drivers/acpi/processor_core.c
+++ b/drivers/acpi/processor_core.c
@@ -731,6 +731,8 @@ static int __cpuinit acpi_processor_start(struct acpi_device *device)
 	/* _PDC call should be done before doing anything else (if reqd.). */
 	arch_acpi_processor_init_pdc(pr);
 	acpi_processor_set_pdc(pr);
+	arch_acpi_processor_cleanup_pdc(pr);
+
 #ifdef CONFIG_CPU_FREQ
 	acpi_processor_ppc_has_changed(pr);
 #endif
diff --git a/include/acpi/processor.h b/include/acpi/processor.h
index 4927c063347c..baf1e0a9a7ee 100644
--- a/include/acpi/processor.h
+++ b/include/acpi/processor.h
@@ -258,6 +258,7 @@ DECLARE_PER_CPU(struct acpi_processor *, processors);
 extern struct acpi_processor_errata errata;
 
 void arch_acpi_processor_init_pdc(struct acpi_processor *pr);
+void arch_acpi_processor_cleanup_pdc(struct acpi_processor *pr);
 
 #ifdef ARCH_HAS_POWER_INIT
 void acpi_processor_power_init_bm_check(struct acpi_processor_flags *flags,
-- 
cgit v1.2.3-59-g8ed1b


From 952e57ba3769d6fc6139b8a99c32ea2bb63f23e9 Mon Sep 17 00:00:00 2001
From: Tilman Schmidt <tilman@imap.cc>
Date: Sat, 20 Jun 2009 01:10:38 -0700
Subject: isdn: clean up documentation index

Remove duplicates, a stray merge conflict marker, and an entry for a file
which doesn't exist, and move one entry to its correct alphabetical place.

Signed-off-by: Tilman Schmidt <tilman@imap.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 Documentation/isdn/00-INDEX | 19 ++-----------------
 1 file changed, 2 insertions(+), 17 deletions(-)

diff --git a/Documentation/isdn/00-INDEX b/Documentation/isdn/00-INDEX
index f6010a536590..e87e336f590e 100644
--- a/Documentation/isdn/00-INDEX
+++ b/Documentation/isdn/00-INDEX
@@ -14,25 +14,14 @@ README
 	- general info on what you need and what to do for Linux ISDN.
 README.FAQ
 	- general info for FAQ.
-README.audio
-	- info for running audio over ISDN.
-README.fax
-	- info for using Fax over ISDN.
-README.gigaset
-	- info on the drivers for Siemens Gigaset ISDN adapters.
-README.icn
-	- info on the ICN-ISDN-card and its driver.
->>>>>>> 93af7aca44f0e82e67bda10a0fb73d383edcc8bd:Documentation/isdn/00-INDEX
 README.HiSax
 	- info on the HiSax driver which replaces the old teles.
+README.act2000
+	- info on driver for IBM ACT-2000 card.
 README.audio
 	- info for running audio over ISDN.
 README.avmb1
 	- info on driver for AVM-B1 ISDN card.
-README.act2000
-	- info on driver for IBM ACT-2000 card.
-README.eicon
-	- info on driver for Eicon active cards.
 README.concap
 	- info on "CONCAP" encapsulation protocol interface used for X.25.
 README.diversion
@@ -59,7 +48,3 @@ README.x25
 	- info for running X.25 over ISDN.
 syncPPP.FAQ
 	- frequently asked questions about running PPP over ISDN.
-README.hysdn
-	- info on driver for Hypercope active HYSDN cards
-README.mISDN
-	- info on the Modular ISDN subsystem (mISDN).
-- 
cgit v1.2.3-59-g8ed1b


From 73e42897e8e5619eacb787d2ce69be12f47cfc21 Mon Sep 17 00:00:00 2001
From: Neil Horman <nhorman@tuxdriver.com>
Date: Sat, 20 Jun 2009 01:15:16 -0700
Subject: ipv4: fix NULL pointer + success return in route lookup path

Don't drop route if we're not caching

	I recently got a report of an oops on a route lookup.  Maxime was
testing what would happen if route caching was turned off (doing so by setting
making rt_caching always return 0), and found that it triggered an oops.  I
looked at it and found that the problem stemmed from the fact that the route
lookup routines were returning success from their lookup paths (which is good),
but never set the **rp pointer to anything (which is bad).  This happens because
in rt_intern_hash, if rt_caching returns false, we call rt_drop and return 0.
This almost emulates slient success.  What we should be doing is assigning *rp =
rt and _not_ dropping the route.  This way, during slow path lookups, when we
create a new route cache entry, we don't immediately discard it, rather we just
don't add it into the cache hash table, but we let this one lookup use it for
the purpose of this route request.  Maxime has tested and reports it prevents
the oops.  There is still a subsequent routing issue that I'm looking into
further, but I'm confident that, even if its related to this same path, this
patch makes sense to take.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/route.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index cd76b3cb7092..65b3a8b11a6c 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1085,8 +1085,16 @@ restart:
 	now = jiffies;
 
 	if (!rt_caching(dev_net(rt->u.dst.dev))) {
-		rt_drop(rt);
-		return 0;
+		/*
+		 * If we're not caching, just tell the caller we
+		 * were successful and don't touch the route.  The
+		 * caller hold the sole reference to the cache entry, and
+		 * it will be released when the caller is done with it.
+		 * If we drop it here, the callers have no way to resolve routes
+		 * when we're not caching.  Instead, just point *rp at rt, so
+		 * the caller gets a single use out of the route
+		 */
+		goto report_and_exit;
 	}
 
 	rthp = &rt_hash_table[hash].chain;
@@ -1217,6 +1225,8 @@ restart:
 	rcu_assign_pointer(rt_hash_table[hash].chain, rt);
 
 	spin_unlock_bh(rt_hash_lock_addr(hash));
+
+report_and_exit:
 	if (rp)
 		*rp = rt;
 	else
-- 
cgit v1.2.3-59-g8ed1b


From 83b462c656813e002843ddb061c8cc99149cab14 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Sat, 20 Jun 2009 01:20:30 -0700
Subject: Net: qla3xxx, remove sleeping in atomic

We cannot sleep in ql_reset_work under spinlock, unlock before sleep,
relock after.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/qla3xxx.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/qla3xxx.c b/drivers/net/qla3xxx.c
index 8a823ecc99a9..bbc6d4d3cc94 100644
--- a/drivers/net/qla3xxx.c
+++ b/drivers/net/qla3xxx.c
@@ -3837,7 +3837,9 @@ static void ql_reset_work(struct work_struct *work)
 						      16) | ISP_CONTROL_RI));
 			}
 
+			spin_unlock_irqrestore(&qdev->hw_lock, hw_flags);
 			ssleep(1);
+			spin_lock_irqsave(&qdev->hw_lock, hw_flags);
 		} while (--max_wait_time);
 		spin_unlock_irqrestore(&qdev->hw_lock, hw_flags);
 
-- 
cgit v1.2.3-59-g8ed1b


From 6be832529a8129c9d90a1d3a78c5d503a710b6fc Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Sat, 20 Jun 2009 01:21:53 -0700
Subject: usbnet cdc_subset: fix issues talking to PXA gadgets

The host-side CDC subset driver is binding more specifically
than it should ... only to PXA 210/25x/26x Linux-USB gadgets.

Loosen that restriction to match the gadget driver driver.
This will various PXA 27x and PXA 3xx devices happier when
talking to Linux hosts, potentially others.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Tested-by: Aric D. Blumer <aric@sdgsystems.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/usb/cdc_subset.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/net/usb/cdc_subset.c b/drivers/net/usb/cdc_subset.c
index c66b9c324f54..ca39ace0b0eb 100644
--- a/drivers/net/usb/cdc_subset.c
+++ b/drivers/net/usb/cdc_subset.c
@@ -307,9 +307,10 @@ static const struct usb_device_id	products [] = {
 	USB_DEVICE (0x1286, 0x8001),    // "blob" bootloader
 	.driver_info =  (unsigned long) &blob_info,
 }, {
-	// Linux Ethernet/RNDIS gadget on pxa210/25x/26x, second config
-	// e.g. Gumstix, current OpenZaurus, ...
-	USB_DEVICE_VER (0x0525, 0xa4a2, 0x0203, 0x0203),
+	// Linux Ethernet/RNDIS gadget, mostly on PXA, second config
+	// e.g. Gumstix, current OpenZaurus, ... or anything else
+	// that just enables this gadget option.
+	USB_DEVICE (0x0525, 0xa4a2),
 	.driver_info =	(unsigned long) &linuxdev_info,
 },
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 33c050c586fec34dae36eb314bfc3a2c44654c05 Mon Sep 17 00:00:00 2001
From: Daniel Laird <daniel.j.laird@nxp.com>
Date: Wed, 5 Nov 2008 16:46:49 +0100
Subject: [WATCHDOG] Add pnx833x_wdt

Add support for PNX833x watchdog timer.

Signed-off-by: Daniel Laird <daniel.j.laird@nxp.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/watchdog/Kconfig       |   9 ++
 drivers/watchdog/Makefile      |   1 +
 drivers/watchdog/pnx833x_wdt.c | 282 +++++++++++++++++++++++++++++++++++++++++
 3 files changed, 292 insertions(+)
 create mode 100644 drivers/watchdog/pnx833x_wdt.c

diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index b166f2852a64..373c22fd24ea 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -729,6 +729,15 @@ config WDT_MTX1
 	  Hardware driver for the MTX-1 boards. This is a watchdog timer that
 	  will reboot the machine after a 100 seconds timer expired.
 
+config PNX833X_WDT
+	tristate "PNX833x Hardware Watchdog"
+	depends on SOC_PNX8335
+	help
+	  Hardware driver for the PNX833x's watchdog. This is a
+	  watchdog timer that will reboot the machine after a programable
+	  timer has expired and no process has written to /dev/watchdog during
+	  that time.
+
 config WDT_RM9K_GPI
 	tristate "RM9000/GPI hardware watchdog"
 	depends on CPU_RM9000
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index c3afa14d5be1..7231a1500860 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -101,6 +101,7 @@ obj-$(CONFIG_SBC_EPX_C3_WATCHDOG) += sbc_epx_c3.o
 obj-$(CONFIG_RC32434_WDT) += rc32434_wdt.o
 obj-$(CONFIG_INDYDOG) += indydog.o
 obj-$(CONFIG_WDT_MTX1) += mtx-1_wdt.o
+obj-$(CONFIG_PNX833X_WDT) += pnx833x_wdt.o
 obj-$(CONFIG_WDT_RM9K_GPI) += rm9k_wdt.o
 obj-$(CONFIG_SIBYTE_WDOG) += sb_wdog.o
 obj-$(CONFIG_AR7_WDT) += ar7_wdt.o
diff --git a/drivers/watchdog/pnx833x_wdt.c b/drivers/watchdog/pnx833x_wdt.c
new file mode 100644
index 000000000000..538ec2c05197
--- /dev/null
+++ b/drivers/watchdog/pnx833x_wdt.c
@@ -0,0 +1,282 @@
+/*
+ *  PNX833x Hardware Watchdog Driver
+ *  Copyright 2008 NXP Semiconductors
+ *  Daniel Laird <daniel.j.laird@nxp.com>
+ *  Andre McCurdy <andre.mccurdy@nxp.com>
+ *
+ *  Heavily based upon - IndyDog	0.3
+ *  A Hardware Watchdog Device for SGI IP22
+ *
+ * (c) Copyright 2002 Guido Guenther <agx@sigxcpu.org>, All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * based on softdog.c by Alan Cox <alan@redhat.com>
+ */
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/fs.h>
+#include <linux/mm.h>
+#include <linux/miscdevice.h>
+#include <linux/watchdog.h>
+#include <linux/notifier.h>
+#include <linux/reboot.h>
+#include <linux/init.h>
+#include <asm/mach-pnx833x/pnx833x.h>
+
+#define PFX "pnx833x: "
+#define WATCHDOG_TIMEOUT 30		/* 30 sec Maximum timeout */
+#define WATCHDOG_COUNT_FREQUENCY 68000000U /* Watchdog counts at 68MHZ. */
+
+/** CONFIG block */
+#define PNX833X_CONFIG                      (0x07000U)
+#define PNX833X_CONFIG_CPU_WATCHDOG         (0x54)
+#define PNX833X_CONFIG_CPU_WATCHDOG_COMPARE (0x58)
+#define PNX833X_CONFIG_CPU_COUNTERS_CONTROL (0x1c)
+
+/** RESET block */
+#define PNX833X_RESET                       (0x08000U)
+#define PNX833X_RESET_CONFIG                (0x08)
+
+static int pnx833x_wdt_alive;
+
+/* Set default timeout in MHZ.*/
+static int pnx833x_wdt_timeout = (WATCHDOG_TIMEOUT * WATCHDOG_COUNT_FREQUENCY);
+module_param(pnx833x_wdt_timeout, int, 0);
+MODULE_PARM_DESC(timeout, "Watchdog timeout in Mhz. (68Mhz clock), default="
+			__MODULE_STRING(pnx833x_wdt_timeout) "(30 seconds).");
+
+static int nowayout = WATCHDOG_NOWAYOUT;
+module_param(nowayout, int, 0);
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default="
+					__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
+
+static int start_enabled = 1;
+module_param(start_enabled, int, 0);
+MODULE_PARM_DESC(start_enabled, "Watchdog is started on module insertion "
+				"(default=" __MODULE_STRING(start_enabled) ")");
+
+static void pnx833x_wdt_start(void)
+{
+	/* Enable watchdog causing reset. */
+	PNX833X_REG(PNX833X_RESET + PNX833X_RESET_CONFIG) |= 0x1;
+	/* Set timeout.*/
+	PNX833X_REG(PNX833X_CONFIG +
+		PNX833X_CONFIG_CPU_WATCHDOG_COMPARE) = pnx833x_wdt_timeout;
+	/* Enable watchdog. */
+	PNX833X_REG(PNX833X_CONFIG +
+				PNX833X_CONFIG_CPU_COUNTERS_CONTROL) |= 0x1;
+
+	printk(KERN_INFO PFX "Started watchdog timer.\n");
+}
+
+static void pnx833x_wdt_stop(void)
+{
+	/* Disable watchdog causing reset. */
+	PNX833X_REG(PNX833X_RESET + PNX833X_CONFIG) &= 0xFFFFFFFE;
+	/* Disable watchdog.*/
+	PNX833X_REG(PNX833X_CONFIG +
+			PNX833X_CONFIG_CPU_COUNTERS_CONTROL) &= 0xFFFFFFFE;
+
+	printk(KERN_INFO PFX "Stopped watchdog timer.\n");
+}
+
+static void pnx833x_wdt_ping(void)
+{
+	PNX833X_REG(PNX833X_CONFIG +
+		PNX833X_CONFIG_CPU_WATCHDOG_COMPARE) = pnx833x_wdt_timeout;
+}
+
+/*
+ *	Allow only one person to hold it open
+ */
+static int pnx833x_wdt_open(struct inode *inode, struct file *file)
+{
+	if (test_and_set_bit(0, &pnx833x_wdt_alive))
+		return -EBUSY;
+
+	if (nowayout)
+		__module_get(THIS_MODULE);
+
+	/* Activate timer */
+	if (!start_enabled)
+		pnx833x_wdt_start();
+
+	pnx833x_wdt_ping();
+
+	printk(KERN_INFO "Started watchdog timer.\n");
+
+	return nonseekable_open(inode, file);
+}
+
+static int pnx833x_wdt_release(struct inode *inode, struct file *file)
+{
+	/* Shut off the timer.
+	 * Lock it in if it's a module and we defined ...NOWAYOUT */
+	if (!nowayout)
+		pnx833x_wdt_stop(); /* Turn the WDT off */
+
+	clear_bit(0, &pnx833x_wdt_alive);
+	return 0;
+}
+
+static ssize_t pnx833x_wdt_write(struct file *file, const char *data, size_t len, loff_t *ppos)
+{
+	/* Refresh the timer. */
+	if (len)
+		pnx833x_wdt_ping();
+
+	return len;
+}
+
+static long pnx833x_wdt_ioctl(struct file *file, unsigned int cmd,
+							unsigned long arg)
+{
+	int options, new_timeout = 0;
+	uint32_t timeout, timeout_left = 0;
+
+	static struct watchdog_info ident = {
+		.options = WDIOF_KEEPALIVEPING | WDIOF_SETTIMEOUT,
+		.firmware_version = 0,
+		.identity = "Hardware Watchdog for PNX833x",
+	};
+
+	switch (cmd) {
+	default:
+		return -ENOTTY;
+
+	case WDIOC_GETSUPPORT:
+		if (copy_to_user((struct watchdog_info *)arg,
+				 &ident, sizeof(ident)))
+			return -EFAULT;
+		return 0;
+
+	case WDIOC_GETSTATUS:
+	case WDIOC_GETBOOTSTATUS:
+		return put_user(0, (int *)arg);
+
+	case WDIOC_SETOPTIONS:
+		if (get_user(options, (int *)arg))
+			return -EFAULT;
+
+		if (options & WDIOS_DISABLECARD)
+			pnx833x_wdt_stop();
+
+		if (options & WDIOS_ENABLECARD)
+			pnx833x_wdt_start();
+
+		return 0;
+
+	case WDIOC_KEEPALIVE:
+		pnx833x_wdt_ping();
+		return 0;
+
+	case WDIOC_SETTIMEOUT:
+	{
+		if (get_user(new_timeout, (int *)arg))
+			return -EFAULT;
+
+		pnx833x_wdt_timeout = new_timeout;
+		PNX833X_REG(PNX833X_CONFIG +
+			PNX833X_CONFIG_CPU_WATCHDOG_COMPARE) = new_timeout;
+		return put_user(new_timeout, (int *)arg);
+	}
+
+	case WDIOC_GETTIMEOUT:
+		timeout = PNX833X_REG(PNX833X_CONFIG +
+					PNX833X_CONFIG_CPU_WATCHDOG_COMPARE);
+		return put_user(timeout, (int *)arg);
+
+	case WDIOC_GETTIMELEFT:
+		timeout_left = PNX833X_REG(PNX833X_CONFIG +
+						PNX833X_CONFIG_CPU_WATCHDOG);
+		return put_user(timeout_left, (int *)arg);
+
+	}
+}
+
+static int pnx833x_wdt_notify_sys(struct notifier_block *this,
+					unsigned long code, void *unused)
+{
+	if (code == SYS_DOWN || code == SYS_HALT)
+		pnx833x_wdt_stop(); /* Turn the WDT off */
+
+	return NOTIFY_DONE;
+}
+
+static const struct file_operations pnx833x_wdt_fops = {
+	.owner		= THIS_MODULE,
+	.llseek		= no_llseek,
+	.write		= pnx833x_wdt_write,
+	.unlocked_ioctl	= pnx833x_wdt_ioctl,
+	.open		= pnx833x_wdt_open,
+	.release	= pnx833x_wdt_release,
+};
+
+static struct miscdevice pnx833x_wdt_miscdev = {
+	.minor		= WATCHDOG_MINOR,
+	.name		= "watchdog",
+	.fops		= &pnx833x_wdt_fops,
+};
+
+static struct notifier_block pnx833x_wdt_notifier = {
+	.notifier_call = pnx833x_wdt_notify_sys,
+};
+
+static char banner[] __initdata =
+	KERN_INFO PFX "Hardware Watchdog Timer for PNX833x: Version 0.1\n";
+
+static int __init watchdog_init(void)
+{
+	int ret, cause;
+
+	/* Lets check the reason for the reset.*/
+	cause = PNX833X_REG(PNX833X_RESET);
+	/*If bit 31 is set then watchdog was cause of reset.*/
+	if (cause & 0x80000000) {
+		printk(KERN_INFO PFX "The system was previously reset due to "
+			"the watchdog firing - please investigate...\n");
+	}
+
+	ret = register_reboot_notifier(&pnx833x_wdt_notifier);
+	if (ret) {
+		printk(KERN_ERR PFX
+			"cannot register reboot notifier (err=%d)\n", ret);
+		return ret;
+	}
+
+	ret = misc_register(&pnx833x_wdt_miscdev);
+	if (ret) {
+		printk(KERN_ERR PFX
+			"cannot register miscdev on minor=%d (err=%d)\n",
+			WATCHDOG_MINOR, ret);
+		unregister_reboot_notifier(&pnx833x_wdt_notifier);
+		return ret;
+	}
+
+	printk(banner);
+	if (start_enabled)
+		pnx833x_wdt_start();
+
+	return 0;
+}
+
+static void __exit watchdog_exit(void)
+{
+	misc_deregister(&pnx833x_wdt_miscdev);
+	unregister_reboot_notifier(&pnx833x_wdt_notifier);
+}
+
+module_init(watchdog_init);
+module_exit(watchdog_exit);
+
+MODULE_AUTHOR("Daniel Laird/Andre McCurdy");
+MODULE_DESCRIPTION("Hardware Watchdog Device for PNX833x");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-- 
cgit v1.2.3-59-g8ed1b


From 01480701d5cef5b3b0f8406d2eab1eaff82f9d5c Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@stericsson.com>
Date: Wed, 6 May 2009 15:35:40 +0200
Subject: [WATCHDOG] U300 COH 901 327 watchdog driver

This patch adds support for the U300 COH 901 327 watchdog for the
U300 platform recently added to RMK:s ARM tree.

Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/watchdog/Kconfig         |  10 +
 drivers/watchdog/Makefile        |   1 +
 drivers/watchdog/coh901327_wdt.c | 537 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 548 insertions(+)
 create mode 100644 drivers/watchdog/coh901327_wdt.c

diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index 373c22fd24ea..631599b7e6a2 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -240,6 +240,16 @@ config ORION_WATCHDOG
 	  To compile this driver as a module, choose M here: the
 	  module will be called orion_wdt.
 
+config COH901327_WATCHDOG
+	bool "ST-Ericsson COH 901 327 watchdog"
+	depends on ARCH_U300
+	default y if MACH_U300
+	help
+	  Say Y here to include Watchdog timer support for the
+	  watchdog embedded into the ST-Ericsson U300 series platforms.
+	  This watchdog is used to reset the system and thus cannot be
+	  compiled as a module.
+
 # AVR32 Architecture
 
 config AT32AP700X_WDT
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index 7231a1500860..c3a22d3d5498 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -41,6 +41,7 @@ obj-$(CONFIG_PNX4008_WATCHDOG) += pnx4008_wdt.o
 obj-$(CONFIG_IOP_WATCHDOG) += iop_wdt.o
 obj-$(CONFIG_DAVINCI_WATCHDOG) += davinci_wdt.o
 obj-$(CONFIG_ORION_WATCHDOG) += orion_wdt.o
+obj-$(CONFIG_COH901327_WATCHDOG) += coh901327_wdt.o
 
 # AVR32 Architecture
 obj-$(CONFIG_AT32AP700X_WDT) += at32ap700x_wdt.o
diff --git a/drivers/watchdog/coh901327_wdt.c b/drivers/watchdog/coh901327_wdt.c
new file mode 100644
index 000000000000..fecb307d28e9
--- /dev/null
+++ b/drivers/watchdog/coh901327_wdt.c
@@ -0,0 +1,537 @@
+/*
+ * coh901327_wdt.c
+ *
+ * Copyright (C) 2008-2009 ST-Ericsson AB
+ * License terms: GNU General Public License (GPL) version 2
+ * Watchdog driver for the ST-Ericsson AB COH 901 327 IP core
+ * Author: Linus Walleij <linus.walleij@stericsson.com>
+ */
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/fs.h>
+#include <linux/miscdevice.h>
+#include <linux/watchdog.h>
+#include <linux/interrupt.h>
+#include <linux/pm.h>
+#include <linux/platform_device.h>
+#include <linux/io.h>
+#include <linux/bitops.h>
+#include <linux/uaccess.h>
+#include <linux/clk.h>
+
+#define DRV_NAME "WDOG COH 901 327"
+
+/*
+ * COH 901 327 register definitions
+ */
+
+/* WDOG_FEED Register 32bit (-/W) */
+#define U300_WDOG_FR							0x00
+#define U300_WDOG_FR_FEED_RESTART_TIMER					0xFEEDU
+/* WDOG_TIMEOUT Register 32bit (R/W) */
+#define U300_WDOG_TR							0x04
+#define U300_WDOG_TR_TIMEOUT_MASK					0x7FFFU
+/* WDOG_DISABLE1 Register 32bit (-/W) */
+#define U300_WDOG_D1R							0x08
+#define U300_WDOG_D1R_DISABLE1_DISABLE_TIMER				0x2BADU
+/* WDOG_DISABLE2 Register 32bit (R/W) */
+#define U300_WDOG_D2R							0x0C
+#define U300_WDOG_D2R_DISABLE2_DISABLE_TIMER				0xCAFEU
+#define U300_WDOG_D2R_DISABLE_STATUS_DISABLED				0xDABEU
+#define U300_WDOG_D2R_DISABLE_STATUS_ENABLED				0x0000U
+/* WDOG_STATUS Register 32bit (R/W) */
+#define U300_WDOG_SR							0x10
+#define U300_WDOG_SR_STATUS_TIMED_OUT					0xCFE8U
+#define U300_WDOG_SR_STATUS_NORMAL					0x0000U
+#define U300_WDOG_SR_RESET_STATUS_RESET					0xE8B4U
+/* WDOG_COUNT Register 32bit (R/-) */
+#define U300_WDOG_CR							0x14
+#define U300_WDOG_CR_VALID_IND						0x8000U
+#define U300_WDOG_CR_VALID_STABLE					0x0000U
+#define U300_WDOG_CR_COUNT_VALUE_MASK					0x7FFFU
+/* WDOG_JTAGOVR Register 32bit (R/W) */
+#define U300_WDOG_JOR							0x18
+#define U300_WDOG_JOR_JTAG_MODE_IND					0x0002U
+#define U300_WDOG_JOR_JTAG_WATCHDOG_ENABLE				0x0001U
+/* WDOG_RESTART Register 32bit (-/W) */
+#define U300_WDOG_RR							0x1C
+#define U300_WDOG_RR_RESTART_VALUE_RESUME				0xACEDU
+/* WDOG_IRQ_EVENT Register 32bit (R/W) */
+#define U300_WDOG_IER							0x20
+#define U300_WDOG_IER_WILL_BARK_IRQ_EVENT_IND				0x0001U
+#define U300_WDOG_IER_WILL_BARK_IRQ_ACK_ENABLE				0x0001U
+/* WDOG_IRQ_MASK Register 32bit (R/W) */
+#define U300_WDOG_IMR							0x24
+#define U300_WDOG_IMR_WILL_BARK_IRQ_ENABLE				0x0001U
+/* WDOG_IRQ_FORCE Register 32bit (R/W) */
+#define U300_WDOG_IFR							0x28
+#define U300_WDOG_IFR_WILL_BARK_IRQ_FORCE_ENABLE			0x0001U
+
+/* Default timeout in seconds = 1 minute */
+static int margin = 60;
+static resource_size_t phybase;
+static resource_size_t physize;
+static int irq;
+static void __iomem *virtbase;
+static unsigned long coh901327_users;
+static unsigned long boot_status;
+static u16 wdogenablestore;
+static u16 irqmaskstore;
+static struct device *parent;
+
+/*
+ * The watchdog block is of course always clocked, the
+ * clk_enable()/clk_disable() calls are mainly for performing reference
+ * counting higher up in the clock hierarchy.
+ */
+static struct clk *clk;
+
+/*
+ * Enabling and disabling functions.
+ */
+static void coh901327_enable(u16 timeout)
+{
+	u16 val;
+
+	clk_enable(clk);
+	/* Restart timer if it is disabled */
+	val = readw(virtbase + U300_WDOG_D2R);
+	if (val == U300_WDOG_D2R_DISABLE_STATUS_DISABLED)
+		writew(U300_WDOG_RR_RESTART_VALUE_RESUME,
+		       virtbase + U300_WDOG_RR);
+	/* Acknowledge any pending interrupt so it doesn't just fire off */
+	writew(U300_WDOG_IER_WILL_BARK_IRQ_ACK_ENABLE,
+	       virtbase + U300_WDOG_IER);
+	/* Enable the watchdog interrupt */
+	writew(U300_WDOG_IMR_WILL_BARK_IRQ_ENABLE, virtbase + U300_WDOG_IMR);
+	/* Activate the watchdog timer */
+	writew(timeout, virtbase + U300_WDOG_TR);
+	/* Start the watchdog timer */
+	writew(U300_WDOG_FR_FEED_RESTART_TIMER, virtbase + U300_WDOG_FR);
+	/*
+	 * Extra read so that this change propagate in the watchdog.
+	 */
+	(void) readw(virtbase + U300_WDOG_CR);
+	val = readw(virtbase + U300_WDOG_D2R);
+	clk_disable(clk);
+	if (val != U300_WDOG_D2R_DISABLE_STATUS_ENABLED)
+		dev_err(parent,
+			"%s(): watchdog not enabled! D2R value %04x\n",
+			__func__, val);
+}
+
+static void coh901327_disable(void)
+{
+	u16 val;
+
+	clk_enable(clk);
+	/* Disable the watchdog interrupt if it is active */
+	writew(0x0000U, virtbase + U300_WDOG_IMR);
+	/* If the watchdog is currently enabled, attempt to disable it */
+	val = readw(virtbase + U300_WDOG_D2R);
+	if (val != U300_WDOG_D2R_DISABLE_STATUS_DISABLED) {
+		writew(U300_WDOG_D1R_DISABLE1_DISABLE_TIMER,
+		       virtbase + U300_WDOG_D1R);
+		writew(U300_WDOG_D2R_DISABLE2_DISABLE_TIMER,
+		       virtbase + U300_WDOG_D2R);
+		/* Write this twice (else problems occur) */
+		writew(U300_WDOG_D2R_DISABLE2_DISABLE_TIMER,
+		       virtbase + U300_WDOG_D2R);
+	}
+	val = readw(virtbase + U300_WDOG_D2R);
+	clk_disable(clk);
+	if (val != U300_WDOG_D2R_DISABLE_STATUS_DISABLED)
+		dev_err(parent,
+			"%s(): watchdog not disabled! D2R value %04x\n",
+			__func__, val);
+}
+
+static void coh901327_start(void)
+{
+	coh901327_enable(margin * 100);
+}
+
+static void coh901327_keepalive(void)
+{
+	clk_enable(clk);
+	/* Feed the watchdog */
+	writew(U300_WDOG_FR_FEED_RESTART_TIMER,
+	       virtbase + U300_WDOG_FR);
+	clk_disable(clk);
+}
+
+static int coh901327_settimeout(int time)
+{
+	/*
+	 * Max margin is 327 since the 10ms
+	 * timeout register is max
+	 * 0x7FFF = 327670ms ~= 327s.
+	 */
+	if (time <= 0 || time > 327)
+		return -EINVAL;
+
+	margin = time;
+	clk_enable(clk);
+	/* Set new timeout value */
+	writew(margin * 100, virtbase + U300_WDOG_TR);
+	/* Feed the dog */
+	writew(U300_WDOG_FR_FEED_RESTART_TIMER,
+	       virtbase + U300_WDOG_FR);
+	clk_disable(clk);
+	return 0;
+}
+
+/*
+ * This interrupt occurs 10 ms before the watchdog WILL bark.
+ */
+static irqreturn_t coh901327_interrupt(int irq, void *data)
+{
+	u16 val;
+
+	/*
+	 * Ack IRQ? If this occurs we're FUBAR anyway, so
+	 * just acknowledge, disable the interrupt and await the imminent end.
+	 * If you at some point need a host of callbacks to be called
+	 * when the system is about to watchdog-reset, add them here!
+	 *
+	 * NOTE: on future versions of this IP-block, it will be possible
+	 * to prevent a watchdog reset by feeding the watchdog at this
+	 * point.
+	 */
+	clk_enable(clk);
+	val = readw(virtbase + U300_WDOG_IER);
+	if (val == U300_WDOG_IER_WILL_BARK_IRQ_EVENT_IND)
+		writew(U300_WDOG_IER_WILL_BARK_IRQ_ACK_ENABLE,
+		       virtbase + U300_WDOG_IER);
+	writew(0x0000U, virtbase + U300_WDOG_IMR);
+	clk_disable(clk);
+	dev_crit(parent, "watchdog is barking!\n");
+	return IRQ_HANDLED;
+}
+
+/*
+ * Allow only one user (daemon) to open the watchdog
+ */
+static int coh901327_open(struct inode *inode, struct file *file)
+{
+	if (test_and_set_bit(1, &coh901327_users))
+		return -EBUSY;
+	coh901327_start();
+	return nonseekable_open(inode, file);
+}
+
+static int coh901327_release(struct inode *inode, struct file *file)
+{
+	clear_bit(1, &coh901327_users);
+	coh901327_disable();
+	return 0;
+}
+
+static ssize_t coh901327_write(struct file *file, const char __user *data,
+			       size_t len, loff_t *ppos)
+{
+	if (len)
+		coh901327_keepalive();
+	return len;
+}
+
+static long coh901327_ioctl(struct file *file, unsigned int cmd,
+			    unsigned long arg)
+{
+	int ret = -ENOTTY;
+	u16 val;
+	int time;
+	int new_options;
+	union {
+		struct watchdog_info __user *ident;
+		int __user *i;
+	} uarg;
+	static struct watchdog_info ident = {
+		.options		= WDIOF_CARDRESET |
+					  WDIOF_SETTIMEOUT |
+					  WDIOF_KEEPALIVEPING,
+		.identity		= "COH 901 327 Watchdog",
+		.firmware_version	= 1,
+	};
+	uarg.i = (int __user *)arg;
+
+	switch (cmd) {
+	case WDIOC_GETSUPPORT:
+		ret = copy_to_user(uarg.ident, &ident,
+				   sizeof(ident)) ? -EFAULT : 0;
+		break;
+
+	case WDIOC_GETSTATUS:
+		ret = put_user(0, uarg.i);
+		break;
+
+	case WDIOC_GETBOOTSTATUS:
+		ret = put_user(boot_status, uarg.i);
+		break;
+
+	case WDIOC_SETOPTIONS:
+		ret = get_user(new_options, uarg.i);
+		if (ret)
+			break;
+		if (new_options & WDIOS_DISABLECARD)
+			coh901327_disable();
+		if (new_options & WDIOS_ENABLECARD)
+			coh901327_start();
+		ret = 0;
+		break;
+
+	case WDIOC_KEEPALIVE:
+		coh901327_keepalive();
+		ret = 0;
+		break;
+
+	case WDIOC_SETTIMEOUT:
+		ret = get_user(time, uarg.i);
+		if (ret)
+			break;
+
+		ret = coh901327_settimeout(time);
+		if (ret)
+			break;
+		/* Then fall through to return set value */
+
+	case WDIOC_GETTIMEOUT:
+		ret = put_user(margin, uarg.i);
+		break;
+
+	case WDIOC_GETTIMELEFT:
+		clk_enable(clk);
+		/* Read repeatedly until the value is stable! */
+		val = readw(virtbase + U300_WDOG_CR);
+		while (val & U300_WDOG_CR_VALID_IND)
+			val = readw(virtbase + U300_WDOG_CR);
+		val &= U300_WDOG_CR_COUNT_VALUE_MASK;
+		clk_disable(clk);
+		if (val != 0)
+			val /= 100;
+		ret = put_user(val, uarg.i);
+		break;
+	}
+	return ret;
+}
+
+static const struct file_operations coh901327_fops = {
+	.owner		= THIS_MODULE,
+	.llseek		= no_llseek,
+	.write		= coh901327_write,
+	.unlocked_ioctl	= coh901327_ioctl,
+	.open		= coh901327_open,
+	.release	= coh901327_release,
+};
+
+static struct miscdevice coh901327_miscdev = {
+	.minor		= WATCHDOG_MINOR,
+	.name		= "watchdog",
+	.fops		= &coh901327_fops,
+};
+
+static int __exit coh901327_remove(struct platform_device *pdev)
+{
+	misc_deregister(&coh901327_miscdev);
+	coh901327_disable();
+	free_irq(irq, pdev);
+	clk_put(clk);
+	iounmap(virtbase);
+	release_mem_region(phybase, physize);
+	return 0;
+}
+
+
+static int __init coh901327_probe(struct platform_device *pdev)
+{
+	int ret;
+	u16 val;
+	struct resource *res;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!res)
+		return -ENOENT;
+
+	parent = &pdev->dev;
+	physize = resource_size(res);
+	phybase = res->start;
+
+	if (request_mem_region(phybase, physize, DRV_NAME) == NULL) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	virtbase = ioremap(phybase, physize);
+	if (!virtbase) {
+		ret = -ENOMEM;
+		goto out_no_remap;
+	}
+
+	clk = clk_get(&pdev->dev, NULL);
+	if (IS_ERR(clk)) {
+		ret = PTR_ERR(clk);
+		dev_err(&pdev->dev, "could not get clock\n");
+		goto out_no_clk;
+	}
+	ret = clk_enable(clk);
+	if (ret) {
+		dev_err(&pdev->dev, "could not enable clock\n");
+		goto out_no_clk_enable;
+	}
+
+	val = readw(virtbase + U300_WDOG_SR);
+	switch (val) {
+	case U300_WDOG_SR_STATUS_TIMED_OUT:
+		dev_info(&pdev->dev,
+			"watchdog timed out since last chip reset!\n");
+		boot_status = WDIOF_CARDRESET;
+		/* Status will be cleared below */
+		break;
+	case U300_WDOG_SR_STATUS_NORMAL:
+		dev_info(&pdev->dev,
+			"in normal status, no timeouts have occurred.\n");
+		break;
+	default:
+		dev_info(&pdev->dev,
+			"contains an illegal status code (%08x)\n", val);
+		break;
+	}
+
+	val = readw(virtbase + U300_WDOG_D2R);
+	switch (val) {
+	case U300_WDOG_D2R_DISABLE_STATUS_DISABLED:
+		dev_info(&pdev->dev, "currently disabled.\n");
+		break;
+	case U300_WDOG_D2R_DISABLE_STATUS_ENABLED:
+		dev_info(&pdev->dev,
+			 "currently enabled! (disabling it now)\n");
+		coh901327_disable();
+		break;
+	default:
+		dev_err(&pdev->dev,
+			"contains an illegal enable/disable code (%08x)\n",
+			val);
+		break;
+	}
+
+	/* Reset the watchdog */
+	writew(U300_WDOG_SR_RESET_STATUS_RESET, virtbase + U300_WDOG_SR);
+
+	irq = platform_get_irq(pdev, 0);
+	if (request_irq(irq, coh901327_interrupt, IRQF_DISABLED,
+			DRV_NAME " Bark", pdev)) {
+		ret = -EIO;
+		goto out_no_irq;
+	}
+
+	clk_disable(clk);
+
+	ret = misc_register(&coh901327_miscdev);
+	if (ret == 0)
+		dev_info(&pdev->dev,
+			 "initialized. timer margin=%d sec\n", margin);
+	else
+		goto out_no_wdog;
+
+	return 0;
+
+out_no_wdog:
+	free_irq(irq, pdev);
+out_no_irq:
+	clk_disable(clk);
+out_no_clk_enable:
+	clk_put(clk);
+out_no_clk:
+	iounmap(virtbase);
+out_no_remap:
+	release_mem_region(phybase, SZ_4K);
+out:
+	return ret;
+}
+
+#ifdef CONFIG_PM
+static int coh901327_suspend(struct platform_device *pdev, pm_message_t state)
+{
+	irqmaskstore = readw(virtbase + U300_WDOG_IMR) & 0x0001U;
+	wdogenablestore = readw(virtbase + U300_WDOG_D2R);
+	/* If watchdog is on, disable it here and now */
+	if (wdogenablestore == U300_WDOG_D2R_DISABLE_STATUS_ENABLED)
+		coh901327_disable();
+	return 0;
+}
+
+static int coh901327_resume(struct platform_device *pdev)
+{
+	/* Restore the watchdog interrupt */
+	writew(irqmaskstore, virtbase + U300_WDOG_IMR);
+	if (wdogenablestore == U300_WDOG_D2R_DISABLE_STATUS_ENABLED) {
+		/* Restart the watchdog timer */
+		writew(U300_WDOG_RR_RESTART_VALUE_RESUME,
+		       virtbase + U300_WDOG_RR);
+		writew(U300_WDOG_FR_FEED_RESTART_TIMER,
+		       virtbase + U300_WDOG_FR);
+	}
+	return 0;
+}
+#else
+#define coh901327_suspend NULL
+#define coh901327_resume  NULL
+#endif
+
+/*
+ * Mistreating the watchdog is the only way to perform a software reset of the
+ * system on EMP platforms. So we implement this and export a symbol for it.
+ */
+void coh901327_watchdog_reset(void)
+{
+	/* Enable even if on JTAG too */
+	writew(U300_WDOG_JOR_JTAG_WATCHDOG_ENABLE,
+	       virtbase + U300_WDOG_JOR);
+	/*
+	 * Timeout = 5s, we have to wait for the watchdog reset to
+	 * actually take place: the watchdog will be reloaded with the
+	 * default value immediately, so we HAVE to reboot and get back
+	 * into the kernel in 30s, or the device will reboot again!
+	 * The boot loader will typically deactivate the watchdog, so we
+	 * need time enough for the boot loader to get to the point of
+	 * deactivating the watchdog before it is shut down by it.
+	 *
+	 * NOTE: on future versions of the watchdog, this restriction is
+	 * gone: the watchdog will be reloaded with a defaul value (1 min)
+	 * instead of last value, and you can conveniently set the watchdog
+	 * timeout to 10ms (value = 1) without any problems.
+	 */
+	coh901327_enable(500);
+	/* Return and await doom */
+}
+
+static struct platform_driver coh901327_driver = {
+	.driver = {
+		.owner	= THIS_MODULE,
+		.name	= "coh901327_wdog",
+	},
+	.remove		= __exit_p(coh901327_remove),
+	.suspend	= coh901327_suspend,
+	.resume		= coh901327_resume,
+};
+
+static int __init coh901327_init(void)
+{
+	return platform_driver_probe(&coh901327_driver, coh901327_probe);
+}
+module_init(coh901327_init);
+
+static void __exit coh901327_exit(void)
+{
+	platform_driver_unregister(&coh901327_driver);
+}
+module_exit(coh901327_exit);
+
+MODULE_AUTHOR("Linus Walleij <linus.walleij@stericsson.com>");
+MODULE_DESCRIPTION("COH 901 327 Watchdog");
+
+module_param(margin, int, 0);
+MODULE_PARM_DESC(margin, "Watchdog margin in seconds (default 60s)");
+
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-- 
cgit v1.2.3-59-g8ed1b


From 80e45b1e9edbca746618724d5b0a31500bdb6f39 Mon Sep 17 00:00:00 2001
From: Timo Kokkonen <timo.t.kokkonen@nokia.com>
Date: Fri, 27 Mar 2009 16:42:17 +0200
Subject: [WATCHDOG] twl4030 watchdog driver

Implementation of twl4030 watchdog driver.

Signed-off-by: Timo Kokkonen <timo.t.kokkonen@nokia.com>
Signed-off-by: Atal Shargorodsky <ext-atal.shargorodsky@nokia.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/mfd/twl4030-core.c     |  12 ++
 drivers/watchdog/Kconfig       |   7 ++
 drivers/watchdog/Makefile      |   1 +
 drivers/watchdog/twl4030_wdt.c | 272 +++++++++++++++++++++++++++++++++++++++++
 4 files changed, 292 insertions(+)
 create mode 100644 drivers/watchdog/twl4030_wdt.c

diff --git a/drivers/mfd/twl4030-core.c b/drivers/mfd/twl4030-core.c
index cd1008c19cd7..ca54996ffd0e 100644
--- a/drivers/mfd/twl4030-core.c
+++ b/drivers/mfd/twl4030-core.c
@@ -101,6 +101,12 @@
 #define twl_has_usb()	false
 #endif
 
+#if defined(CONFIG_TWL4030_WATCHDOG) || \
+	defined(CONFIG_TWL4030_WATCHDOG_MODULE)
+#define twl_has_watchdog()        true
+#else
+#define twl_has_watchdog()        false
+#endif
 
 /* Triton Core internal information (BEGIN) */
 
@@ -526,6 +532,12 @@ add_children(struct twl4030_platform_data *pdata, unsigned long features)
 		usb_transceiver = child;
 	}
 
+	if (twl_has_watchdog()) {
+		child = add_child(0, "twl4030_wdt", NULL, 0, false, 0, 0);
+		if (IS_ERR(child))
+			return PTR_ERR(child);
+	}
+
 	if (twl_has_regulator()) {
 		/*
 		child = add_regulator(TWL4030_REG_VPLL1, pdata->vpll1);
diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index 631599b7e6a2..c4c2206d6f3e 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -250,6 +250,13 @@ config COH901327_WATCHDOG
 	  This watchdog is used to reset the system and thus cannot be
 	  compiled as a module.
 
+config TWL4030_WATCHDOG
+	tristate "TWL4030 Watchdog"
+	depends on TWL4030_CORE
+	help
+	  Support for TI TWL4030 watchdog.  Say 'Y' here to enable the
+	  watchdog timer support for TWL4030 chips.
+
 # AVR32 Architecture
 
 config AT32AP700X_WDT
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index c3a22d3d5498..2c54c04858b2 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -28,6 +28,7 @@ obj-$(CONFIG_USBPCWATCHDOG) += pcwd_usb.o
 obj-$(CONFIG_AT91RM9200_WATCHDOG) += at91rm9200_wdt.o
 obj-$(CONFIG_AT91SAM9X_WATCHDOG) += at91sam9_wdt.o
 obj-$(CONFIG_OMAP_WATCHDOG) += omap_wdt.o
+obj-$(CONFIG_TWL4030_WATCHDOG) += twl4030_wdt.o
 obj-$(CONFIG_21285_WATCHDOG) += wdt285.o
 obj-$(CONFIG_977_WATCHDOG) += wdt977.o
 obj-$(CONFIG_IXP2000_WATCHDOG) += ixp2000_wdt.o
diff --git a/drivers/watchdog/twl4030_wdt.c b/drivers/watchdog/twl4030_wdt.c
new file mode 100644
index 000000000000..cb46556f2973
--- /dev/null
+++ b/drivers/watchdog/twl4030_wdt.c
@@ -0,0 +1,272 @@
+/*
+ * Copyright (C) Nokia Corporation
+ *
+ * Written by Timo Kokkonen <timo.t.kokkonen at nokia.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/fs.h>
+#include <linux/watchdog.h>
+#include <linux/platform_device.h>
+#include <linux/miscdevice.h>
+#include <linux/uaccess.h>
+#include <linux/i2c/twl4030.h>
+
+#define TWL4030_WATCHDOG_CFG_REG_OFFS	0x3
+
+#define TWL4030_WDT_STATE_OPEN		0x1
+#define TWL4030_WDT_STATE_ACTIVE	0x8
+
+static struct platform_device *twl4030_wdt_dev;
+
+struct twl4030_wdt {
+	struct miscdevice	miscdev;
+	int			timer_margin;
+	unsigned long		state;
+};
+
+static int nowayout = WATCHDOG_NOWAYOUT;
+module_param(nowayout, int, 0);
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started "
+	"(default=" __MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
+
+static int twl4030_wdt_write(unsigned char val)
+{
+	return twl4030_i2c_write_u8(TWL4030_MODULE_PM_RECEIVER, val,
+					TWL4030_WATCHDOG_CFG_REG_OFFS);
+}
+
+static int twl4030_wdt_enable(struct twl4030_wdt *wdt)
+{
+	return twl4030_wdt_write(wdt->timer_margin + 1);
+}
+
+static int twl4030_wdt_disable(struct twl4030_wdt *wdt)
+{
+	return twl4030_wdt_write(0);
+}
+
+static int twl4030_wdt_set_timeout(struct twl4030_wdt *wdt, int timeout)
+{
+	if (timeout < 0 || timeout > 30) {
+		dev_warn(wdt->miscdev.parent,
+			"Timeout can only be in the range [0-30] seconds");
+		return -EINVAL;
+	}
+	wdt->timer_margin = timeout;
+	return twl4030_wdt_enable(wdt);
+}
+
+static ssize_t twl4030_wdt_write_fop(struct file *file,
+		const char __user *data, size_t len, loff_t *ppos)
+{
+	struct twl4030_wdt *wdt = file->private_data;
+
+	if (len)
+		twl4030_wdt_enable(wdt);
+
+	return len;
+}
+
+static long twl4030_wdt_ioctl(struct file *file,
+		unsigned int cmd, unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+	int __user *p = argp;
+	int new_margin;
+	struct twl4030_wdt *wdt = file->private_data;
+
+	static const struct watchdog_info twl4030_wd_ident = {
+		.identity = "TWL4030 Watchdog",
+		.options = WDIOF_SETTIMEOUT,
+		.firmware_version = 0,
+	};
+
+	switch (cmd) {
+	case WDIOC_GETSUPPORT:
+		return copy_to_user(argp, &twl4030_wd_ident,
+				sizeof(twl4030_wd_ident)) ? -EFAULT : 0;
+
+	case WDIOC_GETSTATUS:
+	case WDIOC_GETBOOTSTATUS:
+		return put_user(0, p);
+
+	case WDIOC_KEEPALIVE:
+		twl4030_wdt_enable(wdt);
+		break;
+
+	case WDIOC_SETTIMEOUT:
+		if (get_user(new_margin, p))
+			return -EFAULT;
+		if (twl4030_wdt_set_timeout(wdt, new_margin))
+			return -EINVAL;
+		return put_user(wdt->timer_margin, p);
+
+	case WDIOC_GETTIMEOUT:
+		return put_user(wdt->timer_margin, p);
+
+	default:
+		return -ENOTTY;
+	}
+
+	return 0;
+}
+
+static int twl4030_wdt_open(struct inode *inode, struct file *file)
+{
+	struct twl4030_wdt *wdt = platform_get_drvdata(twl4030_wdt_dev);
+
+	/* /dev/watchdog can only be opened once */
+	if (test_and_set_bit(0, &wdt->state))
+		return -EBUSY;
+
+	wdt->state |= TWL4030_WDT_STATE_ACTIVE;
+	file->private_data = (void *) wdt;
+
+	twl4030_wdt_enable(wdt);
+	return nonseekable_open(inode, file);
+}
+
+static int twl4030_wdt_release(struct inode *inode, struct file *file)
+{
+	struct twl4030_wdt *wdt = file->private_data;
+	if (nowayout) {
+		dev_alert(wdt->miscdev.parent,
+		       "Unexpected close, watchdog still running!\n");
+		twl4030_wdt_enable(wdt);
+	} else {
+		if (twl4030_wdt_disable(wdt))
+			return -EFAULT;
+		wdt->state &= ~TWL4030_WDT_STATE_ACTIVE;
+	}
+
+	clear_bit(0, &wdt->state);
+	return 0;
+}
+
+static const struct file_operations twl4030_wdt_fops = {
+	.owner		= THIS_MODULE,
+	.llseek		= no_llseek,
+	.open		= twl4030_wdt_open,
+	.release	= twl4030_wdt_release,
+	.unlocked_ioctl	= twl4030_wdt_ioctl,
+	.write		= twl4030_wdt_write_fop,
+};
+
+static int __devinit twl4030_wdt_probe(struct platform_device *pdev)
+{
+	int ret = 0;
+	struct twl4030_wdt *wdt;
+
+	wdt = kzalloc(sizeof(struct twl4030_wdt), GFP_KERNEL);
+	if (!wdt)
+		return -ENOMEM;
+
+	wdt->state		= 0;
+	wdt->timer_margin	= 30;
+	wdt->miscdev.parent	= &pdev->dev;
+	wdt->miscdev.fops	= &twl4030_wdt_fops;
+	wdt->miscdev.minor	= WATCHDOG_MINOR;
+	wdt->miscdev.name	= "watchdog";
+
+	platform_set_drvdata(pdev, wdt);
+
+	twl4030_wdt_dev = pdev;
+
+	ret = misc_register(&wdt->miscdev);
+	if (ret) {
+		dev_err(wdt->miscdev.parent,
+			"Failed to register misc device\n");
+		platform_set_drvdata(pdev, NULL);
+		kfree(wdt);
+		twl4030_wdt_dev = NULL;
+		return ret;
+	}
+	return 0;
+}
+
+static int __devexit twl4030_wdt_remove(struct platform_device *pdev)
+{
+	struct twl4030_wdt *wdt = platform_get_drvdata(pdev);
+
+	if (wdt->state & TWL4030_WDT_STATE_ACTIVE)
+		if (twl4030_wdt_disable(wdt))
+			return -EFAULT;
+
+	wdt->state &= ~TWL4030_WDT_STATE_ACTIVE;
+	misc_deregister(&wdt->miscdev);
+
+	platform_set_drvdata(pdev, NULL);
+	kfree(wdt);
+	twl4030_wdt_dev = NULL;
+
+	return 0;
+}
+
+#ifdef CONFIG_PM
+static int twl4030_wdt_suspend(struct platform_device *pdev, pm_message_t state)
+{
+	struct twl4030_wdt *wdt = platform_get_drvdata(pdev);
+	if (wdt->state & TWL4030_WDT_STATE_ACTIVE)
+		return twl4030_wdt_disable(wdt);
+
+	return 0;
+}
+
+static int twl4030_wdt_resume(struct platform_device *pdev)
+{
+	struct twl4030_wdt *wdt = platform_get_drvdata(pdev);
+	if (wdt->state & TWL4030_WDT_STATE_ACTIVE)
+		return twl4030_wdt_enable(wdt);
+
+	return 0;
+}
+#else
+#define twl4030_wdt_suspend        NULL
+#define twl4030_wdt_resume         NULL
+#endif
+
+static struct platform_driver twl4030_wdt_driver = {
+	.probe		= twl4030_wdt_probe,
+	.remove		= __devexit_p(twl4030_wdt_remove),
+	.suspend	= twl4030_wdt_suspend,
+	.resume		= twl4030_wdt_resume,
+	.driver		= {
+		.owner	= THIS_MODULE,
+		.name	= "twl4030_wdt",
+	},
+};
+
+static int __devinit twl4030_wdt_init(void)
+{
+	return platform_driver_register(&twl4030_wdt_driver);
+}
+module_init(twl4030_wdt_init);
+
+static void __devexit twl4030_wdt_exit(void)
+{
+	platform_driver_unregister(&twl4030_wdt_driver);
+}
+module_exit(twl4030_wdt_exit);
+
+MODULE_AUTHOR("Nokia Corporation");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
+MODULE_ALIAS("platform:twl4030_wdt");
+
-- 
cgit v1.2.3-59-g8ed1b


From accde1684ff1ea607557fb7224d2dd57775117e1 Mon Sep 17 00:00:00 2001
From: dmitry pervushin <dpervushin@embeddedalley.com>
Date: Wed, 3 Jun 2009 20:21:21 +0400
Subject: [WATCHDOG] Freescale STMP: watchdog driver

Add watchdog timer support for Freescale STMP3xxx boards

Signed-off-by: Vitaly Wool <vital@embeddedalley.com>
Signed-off-by: Dmitry Pervushin <dpervushin@embeddedalley.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/watchdog/Kconfig        |   9 ++
 drivers/watchdog/Makefile       |   1 +
 drivers/watchdog/stmp3xxx_wdt.c | 296 ++++++++++++++++++++++++++++++++++++++++
 3 files changed, 306 insertions(+)
 create mode 100644 drivers/watchdog/stmp3xxx_wdt.c

diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index c4c2206d6f3e..70d7f8cc06b7 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -257,6 +257,15 @@ config TWL4030_WATCHDOG
 	  Support for TI TWL4030 watchdog.  Say 'Y' here to enable the
 	  watchdog timer support for TWL4030 chips.
 
+config STMP3XXX_WATCHDOG
+	tristate "Freescale STMP3XXX watchdog"
+	depends on ARCH_STMP3XXX
+	help
+	  Say Y here if to include support for the watchdog timer
+	  for the Sigmatel STMP37XX/378X SoC.
+	  To compile this driver as a module, choose M here: the
+	  module will be called stmp3xxx_wdt.
+
 # AVR32 Architecture
 
 config AT32AP700X_WDT
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index 2c54c04858b2..d437d5c2d22e 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -43,6 +43,7 @@ obj-$(CONFIG_IOP_WATCHDOG) += iop_wdt.o
 obj-$(CONFIG_DAVINCI_WATCHDOG) += davinci_wdt.o
 obj-$(CONFIG_ORION_WATCHDOG) += orion_wdt.o
 obj-$(CONFIG_COH901327_WATCHDOG) += coh901327_wdt.o
+obj-$(CONFIG_STMP3XXX_WATCHDOG) += stmp3xxx_wdt.o
 
 # AVR32 Architecture
 obj-$(CONFIG_AT32AP700X_WDT) += at32ap700x_wdt.o
diff --git a/drivers/watchdog/stmp3xxx_wdt.c b/drivers/watchdog/stmp3xxx_wdt.c
new file mode 100644
index 000000000000..5dd952681f32
--- /dev/null
+++ b/drivers/watchdog/stmp3xxx_wdt.c
@@ -0,0 +1,296 @@
+/*
+ * Watchdog driver for Freescale STMP37XX/STMP378X
+ *
+ * Author: Vitaly Wool <vital@embeddedalley.com>
+ *
+ * Copyright 2008 Freescale Semiconductor, Inc. All Rights Reserved.
+ * Copyright 2008 Embedded Alley Solutions, Inc All Rights Reserved.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/fs.h>
+#include <linux/miscdevice.h>
+#include <linux/watchdog.h>
+#include <linux/platform_device.h>
+#include <linux/spinlock.h>
+#include <linux/uaccess.h>
+
+#include <mach/platform.h>
+#include <mach/regs-rtc.h>
+
+#define DEFAULT_HEARTBEAT	19
+#define MAX_HEARTBEAT		(0x10000000 >> 6)
+
+/* missing bitmask in headers */
+#define BV_RTC_PERSISTENT1_GENERAL__RTC_FORCE_UPDATER     0x80000000
+
+#define WDT_IN_USE		0
+#define WDT_OK_TO_CLOSE		1
+
+#define WDOG_COUNTER_RATE	1000 /* 1 kHz clock */
+
+static DEFINE_SPINLOCK(stmp3xxx_wdt_io_lock);
+static unsigned long wdt_status;
+static const int nowayout = WATCHDOG_NOWAYOUT;
+static int heartbeat = DEFAULT_HEARTBEAT;
+static unsigned long boot_status;
+
+static void wdt_enable(u32 value)
+{
+	spin_lock(&stmp3xxx_wdt_io_lock);
+	__raw_writel(value, REGS_RTC_BASE + HW_RTC_WATCHDOG);
+	stmp3xxx_setl(BM_RTC_CTRL_WATCHDOGEN, REGS_RTC_BASE + HW_RTC_CTRL);
+	stmp3xxx_setl(BV_RTC_PERSISTENT1_GENERAL__RTC_FORCE_UPDATER,
+			REGS_RTC_BASE + HW_RTC_PERSISTENT1);
+	spin_unlock(&stmp3xxx_wdt_io_lock);
+}
+
+static void wdt_disable(void)
+{
+	spin_lock(&stmp3xxx_wdt_io_lock);
+	stmp3xxx_clearl(BV_RTC_PERSISTENT1_GENERAL__RTC_FORCE_UPDATER,
+			REGS_RTC_BASE + HW_RTC_PERSISTENT1);
+	stmp3xxx_clearl(BM_RTC_CTRL_WATCHDOGEN, REGS_RTC_BASE + HW_RTC_CTRL);
+	spin_unlock(&stmp3xxx_wdt_io_lock);
+}
+
+static void wdt_ping(void)
+{
+	wdt_enable(heartbeat * WDOG_COUNTER_RATE);
+}
+
+static int stmp3xxx_wdt_open(struct inode *inode, struct file *file)
+{
+	if (test_and_set_bit(WDT_IN_USE, &wdt_status))
+		return -EBUSY;
+
+	clear_bit(WDT_OK_TO_CLOSE, &wdt_status);
+	wdt_ping();
+
+	return nonseekable_open(inode, file);
+}
+
+static ssize_t stmp3xxx_wdt_write(struct file *file, const char __user *data,
+	size_t len, loff_t *ppos)
+{
+	if (len) {
+		if (!nowayout) {
+			size_t i;
+
+			clear_bit(WDT_OK_TO_CLOSE, &wdt_status);
+
+			for (i = 0; i != len; i++) {
+				char c;
+
+				if (get_user(c, data + i))
+					return -EFAULT;
+				if (c == 'V')
+					set_bit(WDT_OK_TO_CLOSE, &wdt_status);
+			}
+		}
+		wdt_ping();
+	}
+
+	return len;
+}
+
+static struct watchdog_info ident = {
+	.options	= WDIOF_CARDRESET |
+			  WDIOF_MAGICCLOSE |
+			  WDIOF_SETTIMEOUT |
+			  WDIOF_KEEPALIVEPING,
+	.identity	= "STMP3XXX Watchdog",
+};
+
+static long stmp3xxx_wdt_ioctl(struct file *file, unsigned int cmd,
+	unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+	int __user *p = argp;
+	int new_heartbeat, opts;
+	int ret = -ENOTTY;
+
+	switch (cmd) {
+	case WDIOC_GETSUPPORT:
+		ret = copy_to_user(argp, &ident, sizeof(ident)) ? -EFAULT : 0;
+		break;
+
+	case WDIOC_GETSTATUS:
+		ret = put_user(0, p);
+		break;
+
+	case WDIOC_GETBOOTSTATUS:
+		ret = put_user(boot_status, p);
+		break;
+
+	case WDIOC_SETOPTIONS:
+		if (get_user(opts, p)) {
+			ret = -EFAULT;
+			break;
+		}
+		if (opts & WDIOS_DISABLECARD)
+			wdt_disable();
+		else if (opts & WDIOS_ENABLECARD)
+			wdt_ping();
+		else {
+			pr_debug("%s: unknown option 0x%x\n", __func__, opts);
+			ret = -EINVAL;
+			break;
+		}
+		ret = 0;
+		break;
+
+	case WDIOC_KEEPALIVE:
+		wdt_ping();
+		ret = 0;
+		break;
+
+	case WDIOC_SETTIMEOUT:
+		if (get_user(new_heartbeat, p)) {
+			ret = -EFAULT;
+			break;
+		}
+		if (new_heartbeat <= 0 || new_heartbeat > MAX_HEARTBEAT) {
+			ret = -EINVAL;
+			break;
+		}
+
+		heartbeat = new_heartbeat;
+		wdt_ping();
+		/* Fall through */
+
+	case WDIOC_GETTIMEOUT:
+		ret = put_user(heartbeat, p);
+		break;
+	}
+	return ret;
+}
+
+static int stmp3xxx_wdt_release(struct inode *inode, struct file *file)
+{
+	int ret = 0;
+
+	if (!nowayout) {
+		if (!test_bit(WDT_OK_TO_CLOSE, &wdt_status)) {
+			wdt_ping();
+			pr_debug("%s: Device closed unexpectdly\n", __func__);
+			ret = -EINVAL;
+		} else {
+			wdt_disable();
+			clear_bit(WDT_OK_TO_CLOSE, &wdt_status);
+		}
+	}
+	clear_bit(WDT_IN_USE, &wdt_status);
+
+	return ret;
+}
+
+static const struct file_operations stmp3xxx_wdt_fops = {
+	.owner = THIS_MODULE,
+	.llseek = no_llseek,
+	.write = stmp3xxx_wdt_write,
+	.unlocked_ioctl = stmp3xxx_wdt_ioctl,
+	.open = stmp3xxx_wdt_open,
+	.release = stmp3xxx_wdt_release,
+};
+
+static struct miscdevice stmp3xxx_wdt_miscdev = {
+	.minor = WATCHDOG_MINOR,
+	.name = "watchdog",
+	.fops = &stmp3xxx_wdt_fops,
+};
+
+static int __devinit stmp3xxx_wdt_probe(struct platform_device *pdev)
+{
+	int ret = 0;
+
+	if (heartbeat < 1 || heartbeat > MAX_HEARTBEAT)
+		heartbeat = DEFAULT_HEARTBEAT;
+
+	boot_status = __raw_readl(REGS_RTC_BASE + HW_RTC_PERSISTENT1) &
+			BV_RTC_PERSISTENT1_GENERAL__RTC_FORCE_UPDATER;
+	boot_status = !!boot_status;
+	stmp3xxx_clearl(BV_RTC_PERSISTENT1_GENERAL__RTC_FORCE_UPDATER,
+			REGS_RTC_BASE + HW_RTC_PERSISTENT1);
+	wdt_disable();		/* disable for now */
+
+	ret = misc_register(&stmp3xxx_wdt_miscdev);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "cannot register misc device\n");
+		return ret;
+	}
+
+	printk(KERN_INFO "stmp3xxx watchdog: initialized, heartbeat %d sec\n",
+		heartbeat);
+
+	return ret;
+}
+
+static int __devexit stmp3xxx_wdt_remove(struct platform_device *pdev)
+{
+	misc_deregister(&stmp3xxx_wdt_miscdev);
+	return 0;
+}
+
+#ifdef CONFIG_PM
+static int wdt_suspended;
+static u32 wdt_saved_time;
+
+static int stmp3xxx_wdt_suspend(struct platform_device *pdev,
+				pm_message_t state)
+{
+	if (__raw_readl(REGS_RTC_BASE + HW_RTC_CTRL) &
+		BM_RTC_CTRL_WATCHDOGEN) {
+		wdt_suspended = 1;
+		wdt_saved_time = __raw_readl(REGS_RTC_BASE + HW_RTC_WATCHDOG);
+		wdt_disable();
+	}
+	return 0;
+}
+
+static int stmp3xxx_wdt_resume(struct platform_device *pdev)
+{
+	if (wdt_suspended) {
+		wdt_enable(wdt_saved_time);
+		wdt_suspended = 0;
+	}
+	return 0;
+}
+#else
+#define stmp3xxx_wdt_suspend	NULL
+#define stmp3xxx_wdt_resume	NULL
+#endif
+
+static struct platform_driver platform_wdt_driver = {
+	.driver = {
+		.name = "stmp3xxx_wdt",
+	},
+	.probe = stmp3xxx_wdt_probe,
+	.remove = __devexit_p(stmp3xxx_wdt_remove),
+	.suspend = stmp3xxx_wdt_suspend,
+	.resume = stmp3xxx_wdt_resume,
+};
+
+static int __init stmp3xxx_wdt_init(void)
+{
+	return platform_driver_register(&platform_wdt_driver);
+}
+
+static void __exit stmp3xxx_wdt_exit(void)
+{
+	return platform_driver_unregister(&platform_wdt_driver);
+}
+
+module_init(stmp3xxx_wdt_init);
+module_exit(stmp3xxx_wdt_exit);
+
+MODULE_DESCRIPTION("STMP3XXX Watchdog Driver");
+MODULE_LICENSE("GPL");
+
+module_param(heartbeat, int, 0);
+MODULE_PARM_DESC(heartbeat,
+		 "Watchdog heartbeat period in seconds from 1 to "
+		 __MODULE_STRING(MAX_HEARTBEAT) ", default "
+		 __MODULE_STRING(DEFAULT_HEARTBEAT));
+
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-- 
cgit v1.2.3-59-g8ed1b


From 90074dce5537e87b27125505bb89a373567a7ede Mon Sep 17 00:00:00 2001
From: matthieu castet <castet.matthieu@free.fr>
Date: Fri, 5 Jun 2009 18:57:18 +0200
Subject: [WATCHDOG] add bcm47xx watchdog driver

This add watchdog driver for broadcom 47xx device.
It uses the ssb subsytem to access embeded watchdog device.

Because the watchdog timeout is very short (about 2s), a soft timer is used
to increase the watchdog period.

Note : A patch for exporting the ssb_watchdog_timer_set will
be submitted on next linux-mips merge. Without this patch it can't
be build as a module.

Signed-off-by: Aleksandar Radovanovic <biblbroks@sezampro.rs>
Signed-off-by: Matthieu CASTET <castet.matthieu@free.fr>
Tested-by: Florian Fainelli <florian@openwrt.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 drivers/watchdog/Kconfig       |   6 +
 drivers/watchdog/Makefile      |   1 +
 drivers/watchdog/bcm47xx_wdt.c | 286 +++++++++++++++++++++++++++++++++++++++++
 3 files changed, 293 insertions(+)
 create mode 100644 drivers/watchdog/bcm47xx_wdt.c

diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index 70d7f8cc06b7..e8d45b6ccef8 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -729,6 +729,12 @@ config SBC_EPX_C3_WATCHDOG
 
 # MIPS Architecture
 
+config BCM47XX_WDT
+	tristate "Broadcom BCM47xx Watchdog Timer"
+	depends on BCM47XX
+	help
+	  Hardware driver for the Broadcom BCM47xx Watchog Timer.
+
 config RC32434_WDT
 	tristate "IDT RC32434 SoC Watchdog Timer"
 	depends on MIKROTIK_RB532
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index d437d5c2d22e..3d774294a2b7 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -101,6 +101,7 @@ obj-$(CONFIG_SBC_EPX_C3_WATCHDOG) += sbc_epx_c3.o
 # M68KNOMMU Architecture
 
 # MIPS Architecture
+obj-$(CONFIG_BCM47XX_WDT) += bcm47xx_wdt.o
 obj-$(CONFIG_RC32434_WDT) += rc32434_wdt.o
 obj-$(CONFIG_INDYDOG) += indydog.o
 obj-$(CONFIG_WDT_MTX1) += mtx-1_wdt.o
diff --git a/drivers/watchdog/bcm47xx_wdt.c b/drivers/watchdog/bcm47xx_wdt.c
new file mode 100644
index 000000000000..5c7011cda6a6
--- /dev/null
+++ b/drivers/watchdog/bcm47xx_wdt.c
@@ -0,0 +1,286 @@
+/*
+ *  Watchdog driver for Broadcom BCM47XX
+ *
+ *  Copyright (C) 2008 Aleksandar Radovanovic <biblbroks@sezampro.rs>
+ *  Copyright (C) 2009 Matthieu CASTET <castet.matthieu@free.fr>
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version
+ *  2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/bitops.h>
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/reboot.h>
+#include <linux/types.h>
+#include <linux/uaccess.h>
+#include <linux/watchdog.h>
+#include <linux/timer.h>
+#include <linux/jiffies.h>
+#include <linux/ssb/ssb_embedded.h>
+#include <asm/mach-bcm47xx/bcm47xx.h>
+
+#define DRV_NAME		"bcm47xx_wdt"
+
+#define WDT_DEFAULT_TIME	30	/* seconds */
+#define WDT_MAX_TIME		255	/* seconds */
+
+static int wdt_time = WDT_DEFAULT_TIME;
+static int nowayout = WATCHDOG_NOWAYOUT;
+
+module_param(wdt_time, int, 0);
+MODULE_PARM_DESC(wdt_time, "Watchdog time in seconds. (default="
+				__MODULE_STRING(WDT_DEFAULT_TIME) ")");
+
+#ifdef CONFIG_WATCHDOG_NOWAYOUT
+module_param(nowayout, int, 0);
+MODULE_PARM_DESC(nowayout,
+		"Watchdog cannot be stopped once started (default="
+				__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
+#endif
+
+static unsigned long bcm47xx_wdt_busy;
+static char expect_release;
+static struct timer_list wdt_timer;
+static atomic_t ticks;
+
+static inline void bcm47xx_wdt_hw_start(void)
+{
+	/* this is 2,5s on 100Mhz clock  and 2s on 133 Mhz */
+	ssb_watchdog_timer_set(&ssb_bcm47xx, 0xfffffff);
+}
+
+static inline int bcm47xx_wdt_hw_stop(void)
+{
+	return ssb_watchdog_timer_set(&ssb_bcm47xx, 0);
+}
+
+static void bcm47xx_timer_tick(unsigned long unused)
+{
+	if (!atomic_dec_and_test(&ticks)) {
+		bcm47xx_wdt_hw_start();
+		mod_timer(&wdt_timer, jiffies + HZ);
+	} else {
+		printk(KERN_CRIT DRV_NAME "Watchdog will fire soon!!!\n");
+	}
+}
+
+static inline void bcm47xx_wdt_pet(void)
+{
+	atomic_set(&ticks, wdt_time);
+}
+
+static void bcm47xx_wdt_start(void)
+{
+	bcm47xx_wdt_pet();
+	bcm47xx_timer_tick(0);
+}
+
+static void bcm47xx_wdt_pause(void)
+{
+	del_timer_sync(&wdt_timer);
+	bcm47xx_wdt_hw_stop();
+}
+
+static void bcm47xx_wdt_stop(void)
+{
+	bcm47xx_wdt_pause();
+}
+
+static int bcm47xx_wdt_settimeout(int new_time)
+{
+	if ((new_time <= 0) || (new_time > WDT_MAX_TIME))
+		return -EINVAL;
+
+	wdt_time = new_time;
+	return 0;
+}
+
+static int bcm47xx_wdt_open(struct inode *inode, struct file *file)
+{
+	if (test_and_set_bit(0, &bcm47xx_wdt_busy))
+		return -EBUSY;
+
+	bcm47xx_wdt_start();
+	return nonseekable_open(inode, file);
+}
+
+static int bcm47xx_wdt_release(struct inode *inode, struct file *file)
+{
+	if (expect_release == 42) {
+		bcm47xx_wdt_stop();
+	} else {
+		printk(KERN_CRIT DRV_NAME
+			": Unexpected close, not stopping watchdog!\n");
+		bcm47xx_wdt_start();
+	}
+
+	clear_bit(0, &bcm47xx_wdt_busy);
+	expect_release = 0;
+	return 0;
+}
+
+static ssize_t bcm47xx_wdt_write(struct file *file, const char __user *data,
+				size_t len, loff_t *ppos)
+{
+	if (len) {
+		if (!nowayout) {
+			size_t i;
+
+			expect_release = 0;
+
+			for (i = 0; i != len; i++) {
+				char c;
+				if (get_user(c, data + i))
+					return -EFAULT;
+				if (c == 'V')
+					expect_release = 42;
+			}
+		}
+		bcm47xx_wdt_pet();
+	}
+	return len;
+}
+
+static struct watchdog_info bcm47xx_wdt_info = {
+	.identity 	= DRV_NAME,
+	.options 	= WDIOF_SETTIMEOUT |
+				WDIOF_KEEPALIVEPING |
+				WDIOF_MAGICCLOSE,
+};
+
+static long bcm47xx_wdt_ioctl(struct file *file,
+					unsigned int cmd, unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+	int __user *p = argp;
+	int new_value, retval = -EINVAL;;
+
+	switch (cmd) {
+	case WDIOC_GETSUPPORT:
+		return copy_to_user(argp, &bcm47xx_wdt_info,
+				sizeof(bcm47xx_wdt_info)) ? -EFAULT : 0;
+
+	case WDIOC_GETSTATUS:
+	case WDIOC_GETBOOTSTATUS:
+		return put_user(0, p);
+
+	case WDIOC_SETOPTIONS:
+		if (get_user(new_value, p))
+			return -EFAULT;
+
+		if (new_value & WDIOS_DISABLECARD) {
+			bcm47xx_wdt_stop();
+			retval = 0;
+		}
+
+		if (new_value & WDIOS_ENABLECARD) {
+			bcm47xx_wdt_start();
+			retval = 0;
+		}
+
+		return retval;
+
+	case WDIOC_KEEPALIVE:
+		bcm47xx_wdt_pet();
+		return 0;
+
+	case WDIOC_SETTIMEOUT:
+		if (get_user(new_value, p))
+			return -EFAULT;
+
+		if (bcm47xx_wdt_settimeout(new_value))
+			return -EINVAL;
+
+		bcm47xx_wdt_pet();
+
+	case WDIOC_GETTIMEOUT:
+		return put_user(wdt_time, p);
+
+	default:
+		return -ENOTTY;
+	}
+}
+
+static int bcm47xx_wdt_notify_sys(struct notifier_block *this,
+	unsigned long code, void *unused)
+{
+	if (code == SYS_DOWN || code == SYS_HALT)
+		bcm47xx_wdt_stop();
+	return NOTIFY_DONE;
+}
+
+static const struct file_operations bcm47xx_wdt_fops = {
+	.owner		= THIS_MODULE,
+	.llseek		= no_llseek,
+	.unlocked_ioctl	= bcm47xx_wdt_ioctl,
+	.open		= bcm47xx_wdt_open,
+	.release	= bcm47xx_wdt_release,
+	.write		= bcm47xx_wdt_write,
+};
+
+static struct miscdevice bcm47xx_wdt_miscdev = {
+	.minor		= WATCHDOG_MINOR,
+	.name		= "watchdog",
+	.fops		= &bcm47xx_wdt_fops,
+};
+
+static struct notifier_block bcm47xx_wdt_notifier = {
+	.notifier_call = bcm47xx_wdt_notify_sys,
+};
+
+static int __init bcm47xx_wdt_init(void)
+{
+	int ret;
+
+	if (bcm47xx_wdt_hw_stop() < 0)
+		return -ENODEV;
+
+	setup_timer(&wdt_timer, bcm47xx_timer_tick, 0L);
+
+	if (bcm47xx_wdt_settimeout(wdt_time)) {
+		bcm47xx_wdt_settimeout(WDT_DEFAULT_TIME);
+		printk(KERN_INFO DRV_NAME ": "
+			"wdt_time value must be 0 < wdt_time < %d, using %d\n",
+			(WDT_MAX_TIME + 1), wdt_time);
+	}
+
+	ret = register_reboot_notifier(&bcm47xx_wdt_notifier);
+	if (ret)
+		return ret;
+
+	ret = misc_register(&bcm47xx_wdt_miscdev);
+	if (ret) {
+		unregister_reboot_notifier(&bcm47xx_wdt_notifier);
+		return ret;
+	}
+
+	printk(KERN_INFO "BCM47xx Watchdog Timer enabled (%d seconds%s)\n",
+				wdt_time, nowayout ? ", nowayout" : "");
+	return 0;
+}
+
+static void __exit bcm47xx_wdt_exit(void)
+{
+	if (!nowayout)
+		bcm47xx_wdt_stop();
+
+	misc_deregister(&bcm47xx_wdt_miscdev);
+
+	unregister_reboot_notifier(&bcm47xx_wdt_notifier);
+}
+
+module_init(bcm47xx_wdt_init);
+module_exit(bcm47xx_wdt_exit);
+
+MODULE_AUTHOR("Aleksandar Radovanovic");
+MODULE_DESCRIPTION("Watchdog driver for Broadcom BCM47xx");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-- 
cgit v1.2.3-59-g8ed1b


From 895c9c056591f05ba3b669cda6c12afb212c6253 Mon Sep 17 00:00:00 2001
From: Cliff Cai <cliff.cai@analog.com>
Date: Sat, 20 Jun 2009 11:29:05 -0400
Subject: ASoC: Blackfin: keep better track of SPORT configuration state

Do not let the SPORT be reconfigured until there are no more active
streams.  Then we can let the system reprogram the SPORT state.

Signed-off-by: Cliff Cai <cliff.cai@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
---
 sound/soc/blackfin/bf5xx-i2s.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/sound/soc/blackfin/bf5xx-i2s.c b/sound/soc/blackfin/bf5xx-i2s.c
index 964824419678..0e8af48650be 100644
--- a/sound/soc/blackfin/bf5xx-i2s.c
+++ b/sound/soc/blackfin/bf5xx-i2s.c
@@ -50,6 +50,7 @@ struct bf5xx_i2s_port {
 	u16 tcr2;
 	u16 rcr2;
 	int counter;
+	int configured;
 };
 
 static struct bf5xx_i2s_port bf5xx_i2s;
@@ -168,7 +169,7 @@ static int bf5xx_i2s_hw_params(struct snd_pcm_substream *substream,
 		break;
 	}
 
-	if (bf5xx_i2s.counter == 1) {
+	if (!bf5xx_i2s.configured) {
 		/*
 		 * TX and RX are not independent,they are enabled at the
 		 * same time, even if only one side is running. So, we
@@ -177,6 +178,7 @@ static int bf5xx_i2s_hw_params(struct snd_pcm_substream *substream,
 		 *
 		 * CPU DAI:slave mode.
 		 */
+		bf5xx_i2s.configured = 1;
 		ret = sport_config_rx(sport_handle, bf5xx_i2s.rcr1,
 				      bf5xx_i2s.rcr2, 0, 0);
 		if (ret) {
@@ -200,6 +202,9 @@ static void bf5xx_i2s_shutdown(struct snd_pcm_substream *substream,
 {
 	pr_debug("%s enter\n", __func__);
 	bf5xx_i2s.counter--;
+	/* No active stream, SPORT is allowed to be configured again. */
+	if (!bf5xx_i2s.counter)
+		bf5xx_i2s.configured = 0;
 }
 
 static int bf5xx_i2s_probe(struct platform_device *pdev,
-- 
cgit v1.2.3-59-g8ed1b


From 92a6ad34282f8403c4eb83025a2790af86cdb6bd Mon Sep 17 00:00:00 2001
From: Barry Song <barry.song@analog.com>
Date: Sat, 20 Jun 2009 11:29:57 -0400
Subject: ASoC: Blackfin: update the bf5xx_i2s_resume parameters

Latest ASoC only passes snd_soc_dai to the resume function.

Signed-off-by: Barry Song <barry.song@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
---
 sound/soc/blackfin/bf5xx-i2s.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/sound/soc/blackfin/bf5xx-i2s.c b/sound/soc/blackfin/bf5xx-i2s.c
index 0e8af48650be..af06904bab0f 100644
--- a/sound/soc/blackfin/bf5xx-i2s.c
+++ b/sound/soc/blackfin/bf5xx-i2s.c
@@ -249,8 +249,7 @@ static int bf5xx_i2s_suspend(struct snd_soc_dai *dai)
 	return 0;
 }
 
-static int bf5xx_i2s_resume(struct platform_device *pdev,
-			    struct snd_soc_dai *dai)
+static int bf5xx_i2s_resume(struct snd_soc_dai *dai)
 {
 	int ret;
 	struct sport_device *sport =
-- 
cgit v1.2.3-59-g8ed1b


From 578e4585685410cacd1a4ac86b7e3c12805be918 Mon Sep 17 00:00:00 2001
From: Benny Halevy <bhalevy@panasas.com>
Date: Thu, 18 Jun 2009 22:01:23 -0400
Subject: nfs41: Move initialization of nfs4_opendata seq_res to
 nfs4_init_opendata_res

nfs4_open_recover_helper clears opendata->o_res
before calling nfs4_init_opendata_res, thus causing
NFSv4.0 OPEN operations to be sent rather than nfsv4.1.

Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 fs/nfs/nfs4proc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 7a750f061c26..92ce43517814 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -680,6 +680,7 @@ static void nfs4_init_opendata_res(struct nfs4_opendata *p)
 	p->o_res.server = p->o_arg.server;
 	nfs_fattr_init(&p->f_attr);
 	nfs_fattr_init(&p->dir_attr);
+	p->o_res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 }
 
 static struct nfs4_opendata *nfs4_opendata_alloc(struct path *path,
@@ -711,7 +712,6 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct path *path,
 	p->o_arg.server = server;
 	p->o_arg.bitmask = server->attr_bitmask;
 	p->o_arg.claim = NFS4_OPEN_CLAIM_NULL;
-	p->o_res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
 	if (flags & O_EXCL) {
 		u32 *s = (u32 *) p->o_arg.u.verifier.data;
 		s[0] = jiffies;
-- 
cgit v1.2.3-59-g8ed1b


From e9f029855865e917821ef6034b31e340a4cfc815 Mon Sep 17 00:00:00 2001
From: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Date: Thu, 18 Jun 2009 22:01:24 -0400
Subject: nfs41: sunrpc: xprt_alloc_bc_request() should not use spin_lock_bh()

xprt_alloc_bc_request() is always called in soft interrupt context.
Grab the spin_lock instead of the bottom half spin_lock.  Softirqs
do not preempt other softirqs running on the same processor, so there
is no need to disable bottom halves.

Signed-off-by: Ricardo Labiaga <Ricardo.Labiaga@netapp.com>
Signed-off-by: Benny Halevy <bhalevy@panasas.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
---
 net/sunrpc/backchannel_rqst.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/net/sunrpc/backchannel_rqst.c b/net/sunrpc/backchannel_rqst.c
index 5a7d342e3087..553621fb2c41 100644
--- a/net/sunrpc/backchannel_rqst.c
+++ b/net/sunrpc/backchannel_rqst.c
@@ -211,6 +211,9 @@ EXPORT_SYMBOL(xprt_destroy_backchannel);
  * has been preallocated as well.  Use xprt_alloc_bc_request to allocate
  * to this request.  Use xprt_free_bc_request to return it.
  *
+ * We know that we're called in soft interrupt context, grab the spin_lock
+ * since there is no need to grab the bottom half spin_lock.
+ *
  * Return an available rpc_rqst, otherwise NULL if non are available.
  */
 struct rpc_rqst *xprt_alloc_bc_request(struct rpc_xprt *xprt)
@@ -218,7 +221,7 @@ struct rpc_rqst *xprt_alloc_bc_request(struct rpc_xprt *xprt)
 	struct rpc_rqst *req;
 
 	dprintk("RPC:       allocate a backchannel request\n");
-	spin_lock_bh(&xprt->bc_pa_lock);
+	spin_lock(&xprt->bc_pa_lock);
 	if (!list_empty(&xprt->bc_pa_list)) {
 		req = list_first_entry(&xprt->bc_pa_list, struct rpc_rqst,
 				rq_bc_pa_list);
@@ -226,7 +229,7 @@ struct rpc_rqst *xprt_alloc_bc_request(struct rpc_xprt *xprt)
 	} else {
 		req = NULL;
 	}
-	spin_unlock_bh(&xprt->bc_pa_lock);
+	spin_unlock(&xprt->bc_pa_lock);
 
 	if (req != NULL) {
 		set_bit(RPC_BC_PA_IN_USE, &req->rq_bc_pa_state);
-- 
cgit v1.2.3-59-g8ed1b


From e01916e3e7834cb51327e5e4983ff76bfce8a91f Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Sat, 20 Jun 2009 22:25:45 +0100
Subject: [ARM] wire up rt_tgsigqueueinfo and perf_counter_open

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/include/asm/unistd.h | 2 ++
 arch/arm/kernel/calls.S       | 2 ++
 2 files changed, 4 insertions(+)

diff --git a/arch/arm/include/asm/unistd.h b/arch/arm/include/asm/unistd.h
index 94cc58ef61ae..0e97b8cb77d5 100644
--- a/arch/arm/include/asm/unistd.h
+++ b/arch/arm/include/asm/unistd.h
@@ -389,6 +389,8 @@
 #define __NR_inotify_init1		(__NR_SYSCALL_BASE+360)
 #define __NR_preadv			(__NR_SYSCALL_BASE+361)
 #define __NR_pwritev			(__NR_SYSCALL_BASE+362)
+#define __NR_rt_tgsigqueueinfo		(__NR_SYSCALL_BASE+363)
+#define __NR_perf_counter_open		(__NR_SYSCALL_BASE+364)
 
 /*
  * The following SWIs are ARM private.
diff --git a/arch/arm/kernel/calls.S b/arch/arm/kernel/calls.S
index 1680e9e9c831..f776e72a4cb8 100644
--- a/arch/arm/kernel/calls.S
+++ b/arch/arm/kernel/calls.S
@@ -372,6 +372,8 @@
 /* 360 */	CALL(sys_inotify_init1)
 		CALL(sys_preadv)
 		CALL(sys_pwritev)
+		CALL(sys_rt_tgsigqueueinfo)
+		CALL(sys_perf_counter_open)
 #ifndef syscalls_counted
 .equ syscalls_padding, ((NR_syscalls + 3) & ~3) - NR_syscalls
 #define syscalls_counted
-- 
cgit v1.2.3-59-g8ed1b


From 3eadd3b21cec340dacdc24dd1f9735344290ca62 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Sat, 20 Jun 2009 22:28:41 +0100
Subject: [ARM] Update mach-types

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/tools/mach-types | 39 +++++++++++++++++++++++++++++++++++++--
 1 file changed, 37 insertions(+), 2 deletions(-)

diff --git a/arch/arm/tools/mach-types b/arch/arm/tools/mach-types
index fec64678a63a..33026eff2aa4 100644
--- a/arch/arm/tools/mach-types
+++ b/arch/arm/tools/mach-types
@@ -12,7 +12,7 @@
 #
 #   http://www.arm.linux.org.uk/developer/machines/?action=new
 #
-# Last update: Fri May 29 10:14:20 2009
+# Last update: Sat Jun 20 22:28:39 2009
 #
 # machine_is_xxx	CONFIG_xxxx		MACH_TYPE_xxx		number
 #
@@ -1455,7 +1455,7 @@ gba			MACH_GBA		GBA			1457
 h6044			MACH_H6044		H6044			1458
 app			MACH_APP		APP			1459
 tct_hammer		MACH_TCT_HAMMER		TCT_HAMMER		1460
-herald			MACH_HERMES		HERMES			1461
+herald			MACH_HERALD		HERALD			1461
 artemis			MACH_ARTEMIS		ARTEMIS			1462
 htctitan		MACH_HTCTITAN		HTCTITAN		1463
 qranium			MACH_QRANIUM		QRANIUM			1464
@@ -2245,3 +2245,38 @@ str9			MACH_STR9		STR9			2257
 omap3_wl_ff		MACH_OMAP3_WL_FF	OMAP3_WL_FF		2258
 simcom			MACH_SIMCOM		SIMCOM			2259
 mcwebio			MACH_MCWEBIO		MCWEBIO			2260
+omap3_phrazer		MACH_OMAP3_PHRAZER	OMAP3_PHRAZER		2261
+darwin			MACH_DARWIN		DARWIN			2262
+oratiscomu		MACH_ORATISCOMU		ORATISCOMU		2263
+rtsbc20			MACH_RTSBC20		RTSBC20			2264
+i780			MACH_I780		I780			2265
+gemini324		MACH_GEMINI324		GEMINI324		2266
+oratislan		MACH_ORATISLAN		ORATISLAN		2267
+oratisalog		MACH_ORATISALOG		ORATISALOG		2268
+oratismadi		MACH_ORATISMADI		ORATISMADI		2269
+oratisot16		MACH_ORATISOT16		ORATISOT16		2270
+oratisdesk		MACH_ORATISDESK		ORATISDESK		2271
+v2p_ca9			MACH_V2P_CA9		V2P_CA9			2272
+sintexo			MACH_SINTEXO		SINTEXO			2273
+cm3389			MACH_CM3389		CM3389			2274
+omap3_cio		MACH_OMAP3_CIO		OMAP3_CIO		2275
+sgh_i900		MACH_SGH_I900		SGH_I900		2276
+bst100			MACH_BST100		BST100			2277
+passion			MACH_PASSION		PASSION			2278
+indesign_at91sam	MACH_INDESIGN_AT91SAM	INDESIGN_AT91SAM	2279
+c4_badger		MACH_C4_BADGER		C4_BADGER		2280
+c4_viper		MACH_C4_VIPER		C4_VIPER		2281
+d2net			MACH_D2NET		D2NET			2282
+bigdisk			MACH_BIGDISK		BIGDISK			2283
+notalvision		MACH_NOTALVISION	NOTALVISION		2284
+omap3_kboc		MACH_OMAP3_KBOC		OMAP3_KBOC		2285
+cyclone			MACH_CYCLONE		CYCLONE			2286
+ninja			MACH_NINJA		NINJA			2287
+at91sam9g20ek_2mmc	MACH_AT91SAM9G20EK_2MMC	AT91SAM9G20EK_2MMC	2288
+bcmring			MACH_BCMRING		BCMRING			2289
+resol_dl2		MACH_RESOL_DL2		RESOL_DL2		2290
+ifosw			MACH_IFOSW		IFOSW			2291
+htcrhodium		MACH_HTCRHODIUM		HTCRHODIUM		2292
+htctopaz		MACH_HTCTOPAZ		HTCTOPAZ		2293
+matrix504		MACH_MATRIX504		MATRIX504		2294
+mrfsa			MACH_MRFSA		MRFSA			2295
-- 
cgit v1.2.3-59-g8ed1b


From 99bd0c0fc4b04da54cb311953ef9489931c19c63 Mon Sep 17 00:00:00 2001
From: Andreas Herrmann <andreas.herrmann3@amd.com>
Date: Fri, 19 Jun 2009 10:59:09 +0200
Subject: x86: Set cpu_llc_id on AMD CPUs

This counts when building sched domains in case NUMA information
is not available.

( See cpu_coregroup_mask() which uses llc_shared_map which in turn is
  created based on cpu_llc_id. )

Currently Linux builds domains as follows:
(example from a dual socket quad-core system)

 CPU0 attaching sched-domain:
  domain 0: span 0-7 level CPU
   groups: 0 1 2 3 4 5 6 7

  ...

 CPU7 attaching sched-domain:
  domain 0: span 0-7 level CPU
   groups: 7 0 1 2 3 4 5 6

Ever since that is borked for multi-core AMD CPU systems.
This patch fixes that and now we get a proper:

 CPU0 attaching sched-domain:
  domain 0: span 0-3 level MC
   groups: 0 1 2 3
   domain 1: span 0-7 level CPU
    groups: 0-3 4-7

  ...

 CPU7 attaching sched-domain:
  domain 0: span 4-7 level MC
   groups: 7 4 5 6
   domain 1: span 0-7 level CPU
    groups: 4-7 0-3

This allows scheduler to assign tasks to cores on different sockets
(i.e. that don't share last level cache) for performance reasons.

Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com>
LKML-Reference: <20090619085909.GJ5218@alberich.amd.com>
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/amd.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index e5b27d8f1b47..28e5f5956042 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -258,13 +258,15 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
 {
 #ifdef CONFIG_X86_HT
 	unsigned bits;
+	int cpu = smp_processor_id();
 
 	bits = c->x86_coreid_bits;
-
 	/* Low order bits define the core id (index of core in socket) */
 	c->cpu_core_id = c->initial_apicid & ((1 << bits)-1);
 	/* Convert the initial APIC ID into the socket ID */
 	c->phys_proc_id = c->initial_apicid >> bits;
+	/* use socket ID also for last level cache */
+	per_cpu(cpu_llc_id, cpu) = c->phys_proc_id;
 #endif
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 0c53decdd0a9f9c459ccabe0b5f79660bde5375b Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Tue, 16 Jun 2009 22:36:34 +0200
Subject: firewire: new stack is no longer experimental

The new stack is now recommended over the old one if used for industrial
video (IIDC/DCAM) or for storage devices (SBP-2) due to better
performance, improved compatibility, added features, and security.  It
should also be functionally on par with and is more secure than the old
ieee1394 stack in the use case of consumer video devices.

IP-over-1394 support for the new stack is currently emerging, and a
backend of the firedtv DVB driver to the new stack should be available
soon.

The one remaining area where the old stack is still required are audio
devices, as the new stack is not yet able to support the FFADO FireWire
audio framework.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/Kconfig | 52 ++++++++++++++++++++++--------------------------
 drivers/ieee1394/Kconfig | 19 ++++++------------
 2 files changed, 30 insertions(+), 41 deletions(-)

diff --git a/drivers/firewire/Kconfig b/drivers/firewire/Kconfig
index d6b1721e52ab..13efcd362072 100644
--- a/drivers/firewire/Kconfig
+++ b/drivers/firewire/Kconfig
@@ -1,28 +1,29 @@
-comment "A new alternative FireWire stack is available with EXPERIMENTAL=y"
-	depends on EXPERIMENTAL=n
-
-comment "Enable only one of the two stacks, unless you know what you are doing"
-	depends on EXPERIMENTAL
+comment "You can enable one or both FireWire driver stacks."
+comment "See the help texts for more information."
 
 config FIREWIRE
-	tristate "New FireWire stack, EXPERIMENTAL"
-	depends on EXPERIMENTAL
+	tristate "FireWire driver stack"
 	select CRC_ITU_T
 	help
-	  This is the "Juju" FireWire stack, a new alternative implementation
-	  designed for robustness and simplicity.  You can build either this
-	  stack, or the old stack (the ieee1394 driver, ohci1394 etc.) or both.
-	  Please read http://ieee1394.wiki.kernel.org/index.php/Juju_Migration
-	  before you enable the new stack.
+	  This is the new-generation IEEE 1394 (FireWire) driver stack
+	  a.k.a. Juju, a new implementation designed for robustness and
+	  simplicity.
+	  See http://ieee1394.wiki.kernel.org/index.php/Juju_Migration
+	  for information about migration from the older Linux 1394 stack
+	  to the new driver stack.
 
 	  To compile this driver as a module, say M here: the module will be
 	  called firewire-core.
 
 	  This module functionally replaces ieee1394, raw1394, and video1394.
 	  To access it from application programs, you generally need at least
-	  libraw1394 version 2.  IIDC/DCAM applications also need libdc1394
-	  version 2.  No libraries are required to access storage devices
-	  through the firewire-sbp2 driver.
+	  libraw1394 v2.  IIDC/DCAM applications need libdc1394 v2.
+	  No libraries are required to access storage devices through the
+	  firewire-sbp2 driver.
+
+	  NOTE:
+	  FireWire audio devices currently require the old drivers (ieee1394,
+	  ohci1394, raw1394).
 
 config FIREWIRE_OHCI
 	tristate "OHCI-1394 controllers"
@@ -37,11 +38,9 @@ config FIREWIRE_OHCI
 	  stack.
 
 	  NOTE:
-
-	  You should only build either firewire-ohci or the old ohci1394 driver,
-	  but not both.  If you nevertheless want to install both, you should
-	  configure them only as modules and blacklist the driver(s) which you
-	  don't want to have auto-loaded.  Add either
+	  If you want to install firewire-ohci and ohci1394 together, you
+	  should configure them only as modules and blacklist the driver(s)
+	  which you don't want to have auto-loaded.  Add either
 
 	      blacklist firewire-ohci
 	  or
@@ -50,12 +49,7 @@ config FIREWIRE_OHCI
 	      blacklist dv1394
 
 	  to /etc/modprobe.conf or /etc/modprobe.d/* and update modprobe.conf
-	  depending on your distribution.  The latter two modules should be
-	  blacklisted together with ohci1394 because they depend on ohci1394.
-
-	  If you have an old modprobe which doesn't implement the blacklist
-	  directive, use "install modulename /bin/true" for the modules to be
-	  blacklisted.
+	  depending on your distribution.
 
 config FIREWIRE_OHCI_DEBUG
 	bool
@@ -79,13 +73,15 @@ config FIREWIRE_SBP2
 	  configuration section.
 
 config FIREWIRE_NET
-	tristate "IP networking over 1394"
-	depends on FIREWIRE && INET
+	tristate "IP networking over 1394 (EXPERIMENTAL)"
+	depends on FIREWIRE && INET && EXPERIMENTAL
 	help
 	  This enables IPv4 over IEEE 1394, providing IP connectivity with
 	  other implementations of RFC 2734 as found on several operating
 	  systems.  Multicast support is currently limited.
 
+	  NOTE, this driver is not stable yet!
+
 	  To compile this driver as a module, say M here:  The module will be
 	  called firewire-net.  It replaces eth1394 of the classic IEEE 1394
 	  stack.
diff --git a/drivers/ieee1394/Kconfig b/drivers/ieee1394/Kconfig
index 584245881f4a..f102fcc7e52a 100644
--- a/drivers/ieee1394/Kconfig
+++ b/drivers/ieee1394/Kconfig
@@ -4,7 +4,7 @@ menu "IEEE 1394 (FireWire) support"
 source "drivers/firewire/Kconfig"
 
 config IEEE1394
-	tristate "Stable FireWire stack"
+	tristate "Legacy alternative FireWire driver stack"
 	depends on PCI || BROKEN
 	help
 	  IEEE 1394 describes a high performance serial bus, which is also
@@ -33,11 +33,9 @@ config IEEE1394_OHCI1394
 	  module will be called ohci1394.
 
 	  NOTE:
-
-	  You should only build either ohci1394 or the new firewire-ohci driver,
-	  but not both.  If you nevertheless want to install both, you should
-	  configure them only as modules and blacklist the driver(s) which you
-	  don't want to have auto-loaded.  Add either
+	  If you want to install firewire-ohci and ohci1394 together, you
+	  should configure them only as modules and blacklist the driver(s)
+	  which you don't want to have auto-loaded.  Add either
 
 	      blacklist firewire-ohci
 	  or
@@ -46,12 +44,7 @@ config IEEE1394_OHCI1394
 	      blacklist dv1394
 
 	  to /etc/modprobe.conf or /etc/modprobe.d/* and update modprobe.conf
-	  depending on your distribution.  The latter two modules should be
-	  blacklisted together with ohci1394 because they depend on ohci1394.
-
-	  If you have an old modprobe which doesn't implement the blacklist
-	  directive, use "install modulename /bin/true" for the modules to be
-	  blacklisted.
+	  depending on your distribution.
 
 comment "PCILynx controller requires I2C"
 	depends on IEEE1394 && I2C=n
@@ -105,7 +98,7 @@ config IEEE1394_ETH1394_ROM_ENTRY
 	default n
 
 config IEEE1394_ETH1394
-	tristate "IP networking over 1394"
+	tristate "IP networking over 1394 (experimental)"
 	depends on IEEE1394 && EXPERIMENTAL && INET
 	select IEEE1394_ETH1394_ROM_ENTRY
 	help
-- 
cgit v1.2.3-59-g8ed1b


From d9f2a5ecb2846d0fd368fb4c45182e43f38e4471 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 20 Jun 2009 13:19:25 +0530
Subject: perf_counter, x8: Fix L1-data-Cache-Store-Referencees for AMD

Fix AMD's Data Cache Refills from System event.

After this patch :

 ./tools/perf/perf stat -e l1d -e l1d-misses -e l1d-write -e l1d-prefetch -e l1d-prefetch-miss -e l1i -e l1i-misses -e l1i-prefetch -e l2 -e l2-misses -e l2-write -e dtlb -e dtlb-misses -e itlb -e itlb-misses -e bpu -e bpu-misses ls /dev/ > /dev/null

 Performance counter stats for 'ls /dev/':

        2499484  L1-data-Cache-Load-Referencees             (scaled from 3.97%)
          70347  L1-data-Cache-Load-Misses                  (scaled from 7.30%)
           9360  L1-data-Cache-Store-Referencees            (scaled from 8.64%)
          32804  L1-data-Cache-Prefetch-Referencees         (scaled from 17.72%)
           7693  L1-data-Cache-Prefetch-Misses              (scaled from 22.97%)
        2180945  L1-instruction-Cache-Load-Referencees      (scaled from 28.48%)
          14518  L1-instruction-Cache-Load-Misses           (scaled from 35.00%)
           2405  L1-instruction-Cache-Prefetch-Referencees  (scaled from 34.89%)
          71387  L2-Cache-Load-Referencees                  (scaled from 34.94%)
          18732  L2-Cache-Load-Misses                       (scaled from 34.92%)
          79918  L2-Cache-Store-Referencees                 (scaled from 36.02%)
        1295294  Data-TLB-Cache-Load-Referencees            (scaled from 35.99%)
          30896  Data-TLB-Cache-Load-Misses                 (scaled from 33.36%)
        1222030  Instruction-TLB-Cache-Load-Referencees     (scaled from 29.46%)
            357  Instruction-TLB-Cache-Load-Misses          (scaled from 20.46%)
         530888  Branch-Cache-Load-Referencees              (scaled from 11.48%)
           8638  Branch-Cache-Load-Misses                   (scaled from 5.09%)

    0.011295149  seconds time elapsed.

Earlier it always shows value 0.

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
LKML-Reference: <1245484165.3102.6.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/perf_counter.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c
index 76dfef23f789..22eb3a1d4f9c 100644
--- a/arch/x86/kernel/cpu/perf_counter.c
+++ b/arch/x86/kernel/cpu/perf_counter.c
@@ -401,7 +401,7 @@ static const u64 amd_hw_cache_event_ids
 		[ C(RESULT_MISS)   ] = 0x0041, /* Data Cache Misses          */
 	},
 	[ C(OP_WRITE) ] = {
-		[ C(RESULT_ACCESS) ] = 0x0042, /* Data Cache Refills from L2 */
+		[ C(RESULT_ACCESS) ] = 0x0142, /* Data Cache Refills :system */
 		[ C(RESULT_MISS)   ] = 0,
 	},
 	[ C(OP_PREFETCH) ] = {
-- 
cgit v1.2.3-59-g8ed1b


From c1f47b454ce759d7b13604137a233cad4617e1e8 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sun, 21 Jun 2009 13:58:51 +0200
Subject: perf_counter tools: Fix vmlinux fallback when running on a different
 kernel

Lucas De Marchi reported that perf report and perf annotate
displays mismatching profile if a perf.data is analyzed on
an older kernel - even if the correct vmlinux is specified
via the -k option.

The reason is the fallback path in util/symbol.c:dso__load_kernel():

int dso__load_kernel(struct dso *self, const char *vmlinux,
                     symbol_filter_t filter, int verbose)
{
        int err = -1;

        if (vmlinux)
                err = dso__load_vmlinux(self, vmlinux, filter, verbose);

        if (err)
                err = dso__load_kallsyms(self, filter, verbose);

        return err;
}

dso__load_vmlinux() returns negative on error, but on success it
returns the number of symbols loaded - which confuses the function
to load the kallsyms.

This is normally harmless, as reporting is usually performed on the
same kernel that is analyzed - but if there's a mismatch then we
load the wrong kallsyms and create a non-sensical symbol tree.

The fix is to only fall back to kallsyms on errors.

Reported-by: Lucas De Marchi <lucas.de.marchi@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/util/symbol.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index 86e14375e74e..01b62fa03996 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -629,7 +629,7 @@ int dso__load_kernel(struct dso *self, const char *vmlinux,
 	if (vmlinux)
 		err = dso__load_vmlinux(self, vmlinux, filter, verbose);
 
-	if (err)
+	if (err < 0)
 		err = dso__load_kallsyms(self, filter, verbose);
 
 	return err;
-- 
cgit v1.2.3-59-g8ed1b


From ea09bcc9c298d3057270abd78a973fc678d55c4c Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Sat, 23 May 2009 11:43:37 -0400
Subject: sd: Physical block size and alignment support

Extract physical block size and lowest aligned LBA from READ
CAPACITY(16) response and adjust queue parameters.

Report physical block size and alignment when applicable.

[jejb: fix up trailing whitespace]
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/sd.c | 23 +++++++++++++++++++++--
 drivers/scsi/sd.h |  1 +
 2 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 878b17a9af30..a86064b6458d 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -1307,6 +1307,7 @@ static int read_capacity_16(struct scsi_disk *sdkp, struct scsi_device *sdp,
 	int sense_valid = 0;
 	int the_result;
 	int retries = 3;
+	unsigned int alignment;
 	unsigned long long lba;
 	unsigned sector_size;
 
@@ -1358,6 +1359,16 @@ static int read_capacity_16(struct scsi_disk *sdkp, struct scsi_device *sdp,
 		return -EOVERFLOW;
 	}
 
+	/* Logical blocks per physical block exponent */
+	sdkp->hw_sector_size = (1 << (buffer[13] & 0xf)) * sector_size;
+
+	/* Lowest aligned logical block */
+	alignment = ((buffer[14] & 0x3f) << 8 | buffer[15]) * sector_size;
+	blk_queue_alignment_offset(sdp->request_queue, alignment);
+	if (alignment && sdkp->first_scan)
+		sd_printk(KERN_NOTICE, sdkp,
+			  "physical block alignment offset: %u\n", alignment);
+
 	sdkp->capacity = lba + 1;
 	return sector_size;
 }
@@ -1409,6 +1420,7 @@ static int read_capacity_10(struct scsi_disk *sdkp, struct scsi_device *sdp,
 	}
 
 	sdkp->capacity = lba + 1;
+	sdkp->hw_sector_size = sector_size;
 	return sector_size;
 }
 
@@ -1521,11 +1533,17 @@ got_data:
 		string_get_size(sz, STRING_UNITS_10, cap_str_10,
 				sizeof(cap_str_10));
 
-		if (sdkp->first_scan || old_capacity != sdkp->capacity)
+		if (sdkp->first_scan || old_capacity != sdkp->capacity) {
 			sd_printk(KERN_NOTICE, sdkp,
-				  "%llu %d-byte hardware sectors: (%s/%s)\n",
+				  "%llu %d-byte logical blocks: (%s/%s)\n",
 				  (unsigned long long)sdkp->capacity,
 				  sector_size, cap_str_10, cap_str_2);
+
+			if (sdkp->hw_sector_size != sector_size)
+				sd_printk(KERN_NOTICE, sdkp,
+					  "%u-byte physical blocks\n",
+					  sdkp->hw_sector_size);
+		}
 	}
 
 	/* Rescale capacity to 512-byte units */
@@ -1538,6 +1556,7 @@ got_data:
 	else if (sector_size == 256)
 		sdkp->capacity >>= 1;
 
+	blk_queue_physical_block_size(sdp->request_queue, sdkp->hw_sector_size);
 	sdkp->device->sector_size = sector_size;
 }
 
diff --git a/drivers/scsi/sd.h b/drivers/scsi/sd.h
index 708778cf5f06..8474b5bad3fe 100644
--- a/drivers/scsi/sd.h
+++ b/drivers/scsi/sd.h
@@ -45,6 +45,7 @@ struct scsi_disk {
 	unsigned int	openers;	/* protected by BKL for now, yuck */
 	sector_t	capacity;	/* size in 512-byte sectors */
 	u32		index;
+	unsigned short	hw_sector_size;
 	u8		media_present;
 	u8		write_prot;
 	u8		protection_type;/* Data Integrity Field */
-- 
cgit v1.2.3-59-g8ed1b


From 3821d768912a47ddbd6cab52943a8284df88003c Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Sat, 23 May 2009 11:43:38 -0400
Subject: sd: Detect non-rotational devices

Detect non-rotational devices and set the queue flag accordingly.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/sd.c | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index a86064b6458d..2148d659c281 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -1794,6 +1794,29 @@ void sd_read_app_tag_own(struct scsi_disk *sdkp, unsigned char *buffer)
 	return;
 }
 
+/**
+ * sd_read_block_characteristics - Query block dev. characteristics
+ * @disk: disk to query
+ */
+static void sd_read_block_characteristics(struct scsi_disk *sdkp)
+{
+	char *buffer;
+	u16 rot;
+
+	/* Block Device Characteristics VPD */
+	buffer = scsi_get_vpd_page(sdkp->device, 0xb1);
+
+	if (buffer == NULL)
+		return;
+
+	rot = get_unaligned_be16(&buffer[4]);
+
+	if (rot == 1)
+		queue_flag_set_unlocked(QUEUE_FLAG_NONROT, sdkp->disk->queue);
+
+	kfree(buffer);
+}
+
 /**
  *	sd_revalidate_disk - called the first time a new disk is seen,
  *	performs disk spin up, read_capacity, etc.
@@ -1831,6 +1854,7 @@ static int sd_revalidate_disk(struct gendisk *disk)
 	 */
 	if (sdkp->media_present) {
 		sd_read_capacity(sdkp, buffer);
+		sd_read_block_characteristics(sdkp);
 		sd_read_write_protect_flag(sdkp, buffer);
 		sd_read_cache_type(sdkp, buffer);
 		sd_read_app_tag_own(sdkp, buffer);
@@ -1953,6 +1977,8 @@ static void sd_probe_async(void *data, async_cookie_t cookie)
 	add_disk(gd);
 	sd_dif_config_host(sdkp);
 
+	sd_revalidate_disk(gd);
+
 	sd_printk(KERN_NOTICE, sdkp, "Attached SCSI %sdisk\n",
 		  sdp->removable ? "removable " : "");
 }
-- 
cgit v1.2.3-59-g8ed1b


From d11b6916961d6ec7d7215332cbbe9feec086721d Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Sat, 23 May 2009 11:43:39 -0400
Subject: sd: Block limits VPD support

Query the block limits VPD page and adjust queue minimum and optimal I/O
sizes.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/sd.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 2148d659c281..e4ef11af18a2 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -1794,6 +1794,29 @@ void sd_read_app_tag_own(struct scsi_disk *sdkp, unsigned char *buffer)
 	return;
 }
 
+/**
+ * sd_read_block_limits - Query disk device for preferred I/O sizes.
+ * @disk: disk to query
+ */
+static void sd_read_block_limits(struct scsi_disk *sdkp)
+{
+	unsigned int sector_sz = sdkp->device->sector_size;
+	char *buffer;
+
+	/* Block Limits VPD */
+	buffer = scsi_get_vpd_page(sdkp->device, 0xb0);
+
+	if (buffer == NULL)
+		return;
+
+	blk_queue_io_min(sdkp->disk->queue,
+			 get_unaligned_be16(&buffer[6]) * sector_sz);
+	blk_queue_io_opt(sdkp->disk->queue,
+			 get_unaligned_be32(&buffer[12]) * sector_sz);
+
+	kfree(buffer);
+}
+
 /**
  * sd_read_block_characteristics - Query block dev. characteristics
  * @disk: disk to query
@@ -1854,6 +1877,7 @@ static int sd_revalidate_disk(struct gendisk *disk)
 	 */
 	if (sdkp->media_present) {
 		sd_read_capacity(sdkp, buffer);
+		sd_read_block_limits(sdkp);
 		sd_read_block_characteristics(sdkp);
 		sd_read_write_protect_flag(sdkp, buffer);
 		sd_read_cache_type(sdkp, buffer);
-- 
cgit v1.2.3-59-g8ed1b


From 295ab1b54393aec064533fbc5b483844736ccbf0 Mon Sep 17 00:00:00 2001
From: Karen Xie <kxie@chelsio.com>
Date: Mon, 15 Jun 2009 11:15:16 -0700
Subject: cxgb3i: use kref to track ddp usage

The iscsi ddp functionality could be used by multiple iscsi entities,
add a refcnt to keep track of it, so we would not release it pre-maturely.

Signed-off-by: Karen Xie <kxie@chelsio.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/cxgb3i/cxgb3i_ddp.c | 54 ++++++++++++++++++++++++----------------
 drivers/scsi/cxgb3i/cxgb3i_ddp.h |  2 ++
 2 files changed, 35 insertions(+), 21 deletions(-)

diff --git a/drivers/scsi/cxgb3i/cxgb3i_ddp.c b/drivers/scsi/cxgb3i/cxgb3i_ddp.c
index 99c912547902..8eb2848403f4 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_ddp.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_ddp.c
@@ -598,30 +598,40 @@ int cxgb3i_adapter_ddp_info(struct t3cdev *tdev,
  * release all the resource held by the ddp pagepod manager for a given
  * adapter if needed
  */
-void cxgb3i_ddp_cleanup(struct t3cdev *tdev)
+
+static void ddp_cleanup(struct kref *kref)
 {
+	struct cxgb3i_ddp_info *ddp = container_of(kref,
+						struct cxgb3i_ddp_info,
+						refcnt);
 	int i = 0;
+
+	ddp_log_info("kref release ddp 0x%p, t3dev 0x%p.\n", ddp, ddp->tdev);
+
+	ddp->tdev->ulp_iscsi = NULL;
+	while (i < ddp->nppods) {
+		struct cxgb3i_gather_list *gl = ddp->gl_map[i];
+		if (gl) {
+			int npods = (gl->nelem + PPOD_PAGES_MAX - 1)
+					>> PPOD_PAGES_SHIFT;
+			ddp_log_info("t3dev 0x%p, ddp %d + %d.\n",
+					ddp->tdev, i, npods);
+			kfree(gl);
+			ddp_free_gl_skb(ddp, i, npods);
+			i += npods;
+		} else
+			i++;
+	}
+	cxgb3i_free_big_mem(ddp);
+}
+
+void cxgb3i_ddp_cleanup(struct t3cdev *tdev)
+{
 	struct cxgb3i_ddp_info *ddp = (struct cxgb3i_ddp_info *)tdev->ulp_iscsi;
 
 	ddp_log_info("t3dev 0x%p, release ddp 0x%p.\n", tdev, ddp);
-
-	if (ddp) {
-		tdev->ulp_iscsi = NULL;
-		while (i < ddp->nppods) {
-			struct cxgb3i_gather_list *gl = ddp->gl_map[i];
-			if (gl) {
-				int npods = (gl->nelem + PPOD_PAGES_MAX - 1)
-						>> PPOD_PAGES_SHIFT;
-				ddp_log_info("t3dev 0x%p, ddp %d + %d.\n",
-						tdev, i, npods);
-				kfree(gl);
-				ddp_free_gl_skb(ddp, i, npods);
-				i += npods;
-			} else
-				i++;
-		}
-		cxgb3i_free_big_mem(ddp);
-	}
+	if (ddp)
+		kref_put(&ddp->refcnt, ddp_cleanup);
 }
 
 /**
@@ -631,12 +641,13 @@ void cxgb3i_ddp_cleanup(struct t3cdev *tdev)
  */
 static void ddp_init(struct t3cdev *tdev)
 {
-	struct cxgb3i_ddp_info *ddp;
+	struct cxgb3i_ddp_info *ddp = tdev->ulp_iscsi;
 	struct ulp_iscsi_info uinfo;
 	unsigned int ppmax, bits;
 	int i, err;
 
-	if (tdev->ulp_iscsi) {
+	if (ddp) {
+		kref_get(&ddp->refcnt);
 		ddp_log_warn("t3dev 0x%p, ddp 0x%p already set up.\n",
 				tdev, tdev->ulp_iscsi);
 		return;
@@ -670,6 +681,7 @@ static void ddp_init(struct t3cdev *tdev)
 					  ppmax *
 					  sizeof(struct cxgb3i_gather_list *));
 	spin_lock_init(&ddp->map_lock);
+	kref_init(&ddp->refcnt);
 
 	ddp->tdev = tdev;
 	ddp->pdev = uinfo.pdev;
diff --git a/drivers/scsi/cxgb3i/cxgb3i_ddp.h b/drivers/scsi/cxgb3i/cxgb3i_ddp.h
index 0d296de7cf32..87dd56b422bf 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_ddp.h
+++ b/drivers/scsi/cxgb3i/cxgb3i_ddp.h
@@ -54,6 +54,7 @@ struct cxgb3i_gather_list {
  * struct cxgb3i_ddp_info - cxgb3i direct data placement for pdu payload
  *
  * @list:	list head to link elements
+ * @refcnt:	ref. count
  * @tdev:	pointer to t3cdev used by cxgb3 driver
  * @max_txsz:	max tx packet size for ddp
  * @max_rxsz:	max rx packet size for ddp
@@ -70,6 +71,7 @@ struct cxgb3i_gather_list {
  */
 struct cxgb3i_ddp_info {
 	struct list_head list;
+	struct kref refcnt;
 	struct t3cdev *tdev;
 	struct pci_dev *pdev;
 	unsigned int max_txsz;
-- 
cgit v1.2.3-59-g8ed1b


From 9194c6264040d71f851236437a392594b26e5b91 Mon Sep 17 00:00:00 2001
From: Karen Xie <kxie@chelsio.com>
Date: Mon, 15 Jun 2009 11:15:16 -0700
Subject: cxgb3i: suppot of different kernel page sizes

The default kernel pages supported are 4K, 8K, 16K, and 64K. Re-calculate
entries if PAGE_SIZE is not one of the defaults.

Signed-off-by: Karen Xie <kxie@chelsio.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/cxgb3i/cxgb3i_ddp.c | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/drivers/scsi/cxgb3i/cxgb3i_ddp.c b/drivers/scsi/cxgb3i/cxgb3i_ddp.c
index 8eb2848403f4..344fd53b9954 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_ddp.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_ddp.c
@@ -206,6 +206,31 @@ int cxgb3i_ddp_find_page_index(unsigned long pgsz)
 	return DDP_PGIDX_MAX;
 }
 
+/**
+ * cxgb3i_ddp_adjust_page_table - adjust page table with PAGE_SIZE
+ * return the ddp page index, if no match is found return DDP_PGIDX_MAX.
+ */
+int cxgb3i_ddp_adjust_page_table(void)
+{
+	int i;
+	unsigned int base_order, order;
+
+	if (PAGE_SIZE < (1UL << ddp_page_shift[0])) {
+		ddp_log_info("PAGE_SIZE 0x%lx too small, min. 0x%lx.\n",
+				PAGE_SIZE, 1UL << ddp_page_shift[0]);
+		return -EINVAL;
+	}
+
+	base_order = get_order(1UL << ddp_page_shift[0]);
+	order = get_order(1 << PAGE_SHIFT);
+	for (i = 0; i < DDP_PGIDX_MAX; i++) {
+		/* first is the kernel page size, then just doubling the size */
+		ddp_page_order[i] = order - base_order + i;
+		ddp_page_shift[i] = PAGE_SHIFT + i;
+	}
+	return 0;
+}
+
 static inline void ddp_gl_unmap(struct pci_dev *pdev,
 				struct cxgb3i_gather_list *gl)
 {
@@ -727,6 +752,17 @@ void cxgb3i_ddp_init(struct t3cdev *tdev)
 {
 	if (page_idx == DDP_PGIDX_MAX) {
 		page_idx = cxgb3i_ddp_find_page_index(PAGE_SIZE);
+
+		if (page_idx == DDP_PGIDX_MAX) {
+			ddp_log_info("system PAGE_SIZE %lu, update hw.\n",
+					PAGE_SIZE);
+			if (cxgb3i_ddp_adjust_page_table() < 0) {
+				ddp_log_info("PAGE_SIZE %lu, ddp disabled.\n",
+						PAGE_SIZE);
+				return;
+			}
+			page_idx = cxgb3i_ddp_find_page_index(PAGE_SIZE);
+		}
 		ddp_log_info("system PAGE_SIZE %lu, ddp idx %u.\n",
 				PAGE_SIZE, page_idx);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From d355e57d58193b89283b0c8153649f0427b0bdad Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Mon, 15 Jun 2009 22:11:08 -0500
Subject: libiscsi: don't run scsi eh if iscsi task is making progress

If we are sending or receiving data for the task successfully do
not run the scsi eh, because we know the task is making progress.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/libiscsi.c     | 62 +++++++++++++++++++++++++++++++++++----------
 drivers/scsi/libiscsi_tcp.c |  6 +++--
 include/scsi/libiscsi.h     |  4 +++
 3 files changed, 57 insertions(+), 15 deletions(-)

diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index 59908aead531..b55b7991d5fa 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -954,6 +954,7 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 		task = iscsi_itt_to_ctask(conn, hdr->itt);
 		if (!task)
 			return ISCSI_ERR_BAD_ITT;
+		task->last_xfer = jiffies;
 		break;
 	case ISCSI_OP_R2T:
 		/*
@@ -1192,10 +1193,12 @@ static int iscsi_xmit_task(struct iscsi_conn *conn)
 	spin_unlock_bh(&conn->session->lock);
 	rc = conn->session->tt->xmit_task(task);
 	spin_lock_bh(&conn->session->lock);
-	__iscsi_put_task(task);
-	if (!rc)
+	if (!rc) {
 		/* done with this task */
+		task->last_xfer = jiffies;
 		conn->task = NULL;
+	}
+	__iscsi_put_task(task);
 	return rc;
 }
 
@@ -1361,6 +1364,9 @@ static inline struct iscsi_task *iscsi_alloc_task(struct iscsi_conn *conn,
 	task->state = ISCSI_TASK_PENDING;
 	task->conn = conn;
 	task->sc = sc;
+	task->have_checked_conn = false;
+	task->last_timeout = jiffies;
+	task->last_xfer = jiffies;
 	INIT_LIST_HEAD(&task->running);
 	return task;
 }
@@ -1716,17 +1722,18 @@ static int iscsi_has_ping_timed_out(struct iscsi_conn *conn)
 		return 0;
 }
 
-static enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *scmd)
+static enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *sc)
 {
+	enum blk_eh_timer_return rc = BLK_EH_NOT_HANDLED;
+	struct iscsi_task *task = NULL;
 	struct iscsi_cls_session *cls_session;
 	struct iscsi_session *session;
 	struct iscsi_conn *conn;
-	enum blk_eh_timer_return rc = BLK_EH_NOT_HANDLED;
 
-	cls_session = starget_to_session(scsi_target(scmd->device));
+	cls_session = starget_to_session(scsi_target(sc->device));
 	session = cls_session->dd_data;
 
-	ISCSI_DBG_SESSION(session, "scsi cmd %p timedout\n", scmd);
+	ISCSI_DBG_SESSION(session, "scsi cmd %p timedout\n", sc);
 
 	spin_lock(&session->lock);
 	if (session->state != ISCSI_STATE_LOGGED_IN) {
@@ -1745,6 +1752,26 @@ static enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *scmd)
 		goto done;
 	}
 
+	task = (struct iscsi_task *)sc->SCp.ptr;
+	if (!task)
+		goto done;
+	/*
+	 * If we have sent (at least queued to the network layer) a pdu or
+	 * recvd one for the task since the last timeout ask for
+	 * more time. If on the next timeout we have not made progress
+	 * we can check if it is the task or connection when we send the
+	 * nop as a ping.
+	 */
+	if (time_after_eq(task->last_xfer, task->last_timeout)) {
+		ISCSI_DBG_CONN(conn, "Command making progress. Asking "
+			       "scsi-ml for more time to complete. "
+			       "Last data recv at %lu. Last timeout was at "
+			       "%lu\n.", task->last_xfer, task->last_timeout);
+		task->have_checked_conn = false;
+		rc = BLK_EH_RESET_TIMER;
+		goto done;
+	}
+
 	if (!conn->recv_timeout && !conn->ping_timeout)
 		goto done;
 	/*
@@ -1755,20 +1782,29 @@ static enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *scmd)
 		rc = BLK_EH_RESET_TIMER;
 		goto done;
 	}
+
+	/* Assumes nop timeout is shorter than scsi cmd timeout */
+	if (task->have_checked_conn)
+		goto done;
+
 	/*
-	 * if we are about to check the transport then give the command
-	 * more time
+	 * Checking the transport already or nop from a cmd timeout still
+	 * running
 	 */
-	if (time_before_eq(conn->last_recv + (conn->recv_timeout * HZ),
-			   jiffies)) {
+	if (conn->ping_task) {
+		task->have_checked_conn = true;
 		rc = BLK_EH_RESET_TIMER;
 		goto done;
 	}
 
-	/* if in the middle of checking the transport then give us more time */
-	if (conn->ping_task)
-		rc = BLK_EH_RESET_TIMER;
+	/* Make sure there is a transport check done */
+	iscsi_send_nopout(conn, NULL);
+	task->have_checked_conn = true;
+	rc = BLK_EH_RESET_TIMER;
+
 done:
+	if (task)
+		task->last_timeout = jiffies;
 	spin_unlock(&session->lock);
 	ISCSI_DBG_SESSION(session, "return %s\n", rc == BLK_EH_RESET_TIMER ?
 			  "timer reset" : "nh");
diff --git a/drivers/scsi/libiscsi_tcp.c b/drivers/scsi/libiscsi_tcp.c
index 2bc07090321d..2e0746d70303 100644
--- a/drivers/scsi/libiscsi_tcp.c
+++ b/drivers/scsi/libiscsi_tcp.c
@@ -686,6 +686,7 @@ iscsi_tcp_hdr_dissect(struct iscsi_conn *conn, struct iscsi_hdr *hdr)
 				     "offset=%d, datalen=%d)\n",
 				      tcp_task->data_offset,
 				      tcp_conn->in.datalen);
+			task->last_xfer = jiffies;
 			rc = iscsi_segment_seek_sg(&tcp_conn->in.segment,
 						   sdb->table.sgl,
 						   sdb->table.nents,
@@ -713,9 +714,10 @@ iscsi_tcp_hdr_dissect(struct iscsi_conn *conn, struct iscsi_hdr *hdr)
 			rc = ISCSI_ERR_BAD_ITT;
 		else if (ahslen)
 			rc = ISCSI_ERR_AHSLEN;
-		else if (task->sc->sc_data_direction == DMA_TO_DEVICE)
+		else if (task->sc->sc_data_direction == DMA_TO_DEVICE) {
+			task->last_xfer = jiffies;
 			rc = iscsi_tcp_r2t_rsp(conn, task);
-		else
+		} else
 			rc = ISCSI_ERR_PROTO;
 		spin_unlock(&conn->session->lock);
 		break;
diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h
index 196525cd402f..61afeb59a836 100644
--- a/include/scsi/libiscsi.h
+++ b/include/scsi/libiscsi.h
@@ -125,6 +125,10 @@ struct iscsi_task {
 	struct scsi_cmnd	*sc;		/* associated SCSI cmd*/
 	struct iscsi_conn	*conn;		/* used connection    */
 
+	/* data processing tracking */
+	unsigned long		last_xfer;
+	unsigned long		last_timeout;
+	bool			have_checked_conn;
 	/* state set/tested under session->lock */
 	int			state;
 	atomic_t		refcount;
-- 
cgit v1.2.3-59-g8ed1b


From 32382492eb18e8e20be382a1743d0c08469d1e84 Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Mon, 15 Jun 2009 22:11:09 -0500
Subject: iscsi_tcp: propogate EAGAIN from sendpage to libiscsi

The net layer might return -EAGAIN because it could not
get space/mem within the sock sndtimeo or becuase the tcp/ip
connection was down. For the latter we do not want to retry
because the conn/session should just be shutdown and restarted.
libiscsi knows the state of the session recovery so propogate
this error to that layer. It will either do iscsi recovery
or have us retry the operation. Right now if we have partially
sent a pdu we would always retry the IO xmit slowing down
recovery.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/iscsi_tcp.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index b7c092d63bbe..518dbd91df85 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -253,8 +253,6 @@ static int iscsi_sw_tcp_xmit_segment(struct iscsi_tcp_conn *tcp_conn,
 
 		if (r < 0) {
 			iscsi_tcp_segment_unmap(segment);
-			if (copied || r == -EAGAIN)
-				break;
 			return r;
 		}
 		copied += r;
@@ -275,11 +273,17 @@ static int iscsi_sw_tcp_xmit(struct iscsi_conn *conn)
 
 	while (1) {
 		rc = iscsi_sw_tcp_xmit_segment(tcp_conn, segment);
-		if (rc < 0) {
+		/*
+		 * We may not have been able to send data because the conn
+		 * is getting stopped. libiscsi will know so propogate err
+		 * for it to do the right thing.
+		 */
+		if (rc == -EAGAIN)
+			return rc;
+		else if (rc < 0) {
 			rc = ISCSI_ERR_XMIT_FAILED;
 			goto error;
-		}
-		if (rc == 0)
+		} else if (rc == 0)
 			break;
 
 		consumed += rc;
-- 
cgit v1.2.3-59-g8ed1b


From bd2199d417313a056d4d2b2bac852231e1b50a4e Mon Sep 17 00:00:00 2001
From: Erez Zilber <erezzi.list@gmail.com>
Date: Mon, 15 Jun 2009 22:11:10 -0500
Subject: libiscsi: add conn and scsi eh log debug flags

Allow the user to control the debug logs in libiscsi. We will now
have a module param for connection, session & error handling.

[Mike Christie - Fixed up to compile on current code and added
missing ISCSI_DBG_EH conversions]

Signed-off-by: Erez Zilber <erezzi.list@gmail.com>
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/libiscsi.c | 109 +++++++++++++++++++++++++++++-------------------
 1 file changed, 65 insertions(+), 44 deletions(-)

diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index b55b7991d5fa..716cc344c5df 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -38,15 +38,30 @@
 #include <scsi/scsi_transport_iscsi.h>
 #include <scsi/libiscsi.h>
 
-static int iscsi_dbg_lib;
-module_param_named(debug_libiscsi, iscsi_dbg_lib, int, S_IRUGO | S_IWUSR);
-MODULE_PARM_DESC(debug_libiscsi, "Turn on debugging for libiscsi module. "
-		 "Set to 1 to turn on, and zero to turn off. Default "
-		 "is off.");
+static int iscsi_dbg_lib_conn;
+module_param_named(debug_libiscsi_conn, iscsi_dbg_lib_conn, int,
+		   S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(debug_libiscsi_conn,
+		 "Turn on debugging for connections in libiscsi module. "
+		 "Set to 1 to turn on, and zero to turn off. Default is off.");
+
+static int iscsi_dbg_lib_session;
+module_param_named(debug_libiscsi_session, iscsi_dbg_lib_session, int,
+		   S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(debug_libiscsi_session,
+		 "Turn on debugging for sessions in libiscsi module. "
+		 "Set to 1 to turn on, and zero to turn off. Default is off.");
+
+static int iscsi_dbg_lib_eh;
+module_param_named(debug_libiscsi_eh, iscsi_dbg_lib_eh, int,
+		   S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(debug_libiscsi_eh,
+		 "Turn on debugging for error handling in libiscsi module. "
+		 "Set to 1 to turn on, and zero to turn off. Default is off.");
 
 #define ISCSI_DBG_CONN(_conn, dbg_fmt, arg...)			\
 	do {							\
-		if (iscsi_dbg_lib)				\
+		if (iscsi_dbg_lib_conn)				\
 			iscsi_conn_printk(KERN_INFO, _conn,	\
 					     "%s " dbg_fmt,	\
 					     __func__, ##arg);	\
@@ -54,7 +69,15 @@ MODULE_PARM_DESC(debug_libiscsi, "Turn on debugging for libiscsi module. "
 
 #define ISCSI_DBG_SESSION(_session, dbg_fmt, arg...)			\
 	do {								\
-		if (iscsi_dbg_lib)					\
+		if (iscsi_dbg_lib_session)				\
+			iscsi_session_printk(KERN_INFO, _session,	\
+					     "%s " dbg_fmt,		\
+					     __func__, ##arg);		\
+	} while (0);
+
+#define ISCSI_DBG_EH(_session, dbg_fmt, arg...)				\
+	do {								\
+		if (iscsi_dbg_lib_eh)					\
 			iscsi_session_printk(KERN_INFO, _session,	\
 					     "%s " dbg_fmt,		\
 					     __func__, ##arg);		\
@@ -1561,10 +1584,10 @@ int iscsi_eh_target_reset(struct scsi_cmnd *sc)
 	spin_lock_bh(&session->lock);
 	if (session->state == ISCSI_STATE_TERMINATE) {
 failed:
-		iscsi_session_printk(KERN_INFO, session,
-				     "failing target reset: Could not log "
-				     "back into target [age %d]\n",
-				     session->age);
+		ISCSI_DBG_EH(session,
+			     "failing target reset: Could not log back into "
+			     "target [age %d]\n",
+			     session->age);
 		spin_unlock_bh(&session->lock);
 		mutex_unlock(&session->eh_mutex);
 		return FAILED;
@@ -1578,7 +1601,7 @@ failed:
 	 */
 	iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
 
-	ISCSI_DBG_SESSION(session, "wait for relogin\n");
+	ISCSI_DBG_EH(session, "wait for relogin\n");
 	wait_event_interruptible(conn->ehwait,
 				 session->state == ISCSI_STATE_TERMINATE ||
 				 session->state == ISCSI_STATE_LOGGED_IN ||
@@ -1588,10 +1611,10 @@ failed:
 
 	mutex_lock(&session->eh_mutex);
 	spin_lock_bh(&session->lock);
-	if (session->state == ISCSI_STATE_LOGGED_IN)
-		iscsi_session_printk(KERN_INFO, session,
-				     "target reset succeeded\n");
-	else
+	if (session->state == ISCSI_STATE_LOGGED_IN) {
+		ISCSI_DBG_EH(session,
+			     "target reset succeeded\n");
+	} else
 		goto failed;
 	spin_unlock_bh(&session->lock);
 	mutex_unlock(&session->eh_mutex);
@@ -1607,7 +1630,7 @@ static void iscsi_tmf_timedout(unsigned long data)
 	spin_lock(&session->lock);
 	if (conn->tmf_state == TMF_QUEUED) {
 		conn->tmf_state = TMF_TIMEDOUT;
-		ISCSI_DBG_SESSION(session, "tmf timedout\n");
+		ISCSI_DBG_EH(session, "tmf timedout\n");
 		/* unblock eh_abort() */
 		wake_up(&conn->ehwait);
 	}
@@ -1627,7 +1650,7 @@ static int iscsi_exec_task_mgmt_fn(struct iscsi_conn *conn,
 		spin_unlock_bh(&session->lock);
 		iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
 		spin_lock_bh(&session->lock);
-		ISCSI_DBG_SESSION(session, "tmf exec failure\n");
+		ISCSI_DBG_EH(session, "tmf exec failure\n");
 		return -EPERM;
 	}
 	conn->tmfcmd_pdus_cnt++;
@@ -1635,7 +1658,7 @@ static int iscsi_exec_task_mgmt_fn(struct iscsi_conn *conn,
 	conn->tmf_timer.function = iscsi_tmf_timedout;
 	conn->tmf_timer.data = (unsigned long)conn;
 	add_timer(&conn->tmf_timer);
-	ISCSI_DBG_SESSION(session, "tmf set timeout\n");
+	ISCSI_DBG_EH(session, "tmf set timeout\n");
 
 	spin_unlock_bh(&session->lock);
 	mutex_unlock(&session->eh_mutex);
@@ -1733,7 +1756,7 @@ static enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *sc)
 	cls_session = starget_to_session(scsi_target(sc->device));
 	session = cls_session->dd_data;
 
-	ISCSI_DBG_SESSION(session, "scsi cmd %p timedout\n", sc);
+	ISCSI_DBG_EH(session, "scsi cmd %p timedout\n", sc);
 
 	spin_lock(&session->lock);
 	if (session->state != ISCSI_STATE_LOGGED_IN) {
@@ -1763,10 +1786,10 @@ static enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *sc)
 	 * nop as a ping.
 	 */
 	if (time_after_eq(task->last_xfer, task->last_timeout)) {
-		ISCSI_DBG_CONN(conn, "Command making progress. Asking "
-			       "scsi-ml for more time to complete. "
-			       "Last data recv at %lu. Last timeout was at "
-			       "%lu\n.", task->last_xfer, task->last_timeout);
+		ISCSI_DBG_EH(session, "Command making progress. Asking "
+			     "scsi-ml for more time to complete. "
+			     "Last data recv at %lu. Last timeout was at "
+			     "%lu\n.", task->last_xfer, task->last_timeout);
 		task->have_checked_conn = false;
 		rc = BLK_EH_RESET_TIMER;
 		goto done;
@@ -1806,8 +1829,8 @@ done:
 	if (task)
 		task->last_timeout = jiffies;
 	spin_unlock(&session->lock);
-	ISCSI_DBG_SESSION(session, "return %s\n", rc == BLK_EH_RESET_TIMER ?
-			  "timer reset" : "nh");
+	ISCSI_DBG_EH(session, "return %s\n", rc == BLK_EH_RESET_TIMER ?
+		     "timer reset" : "nh");
 	return rc;
 }
 
@@ -1877,7 +1900,7 @@ int iscsi_eh_abort(struct scsi_cmnd *sc)
 	cls_session = starget_to_session(scsi_target(sc->device));
 	session = cls_session->dd_data;
 
-	ISCSI_DBG_SESSION(session, "aborting sc %p\n", sc);
+	ISCSI_DBG_EH(session, "aborting sc %p\n", sc);
 
 	mutex_lock(&session->eh_mutex);
 	spin_lock_bh(&session->lock);
@@ -1886,8 +1909,8 @@ int iscsi_eh_abort(struct scsi_cmnd *sc)
 	 * got the command.
 	 */
 	if (!sc->SCp.ptr) {
-		ISCSI_DBG_SESSION(session, "sc never reached iscsi layer or "
-				  "it completed.\n");
+		ISCSI_DBG_EH(session, "sc never reached iscsi layer or "
+				      "it completed.\n");
 		spin_unlock_bh(&session->lock);
 		mutex_unlock(&session->eh_mutex);
 		return SUCCESS;
@@ -1901,7 +1924,7 @@ int iscsi_eh_abort(struct scsi_cmnd *sc)
 	    sc->SCp.phase != session->age) {
 		spin_unlock_bh(&session->lock);
 		mutex_unlock(&session->eh_mutex);
-		ISCSI_DBG_SESSION(session, "failing abort due to dropped "
+		ISCSI_DBG_EH(session, "failing abort due to dropped "
 				  "session.\n");
 		return FAILED;
 	}
@@ -1911,13 +1934,12 @@ int iscsi_eh_abort(struct scsi_cmnd *sc)
 	age = session->age;
 
 	task = (struct iscsi_task *)sc->SCp.ptr;
-	ISCSI_DBG_SESSION(session, "aborting [sc %p itt 0x%x]\n",
-			  sc, task->itt);
+	ISCSI_DBG_EH(session, "aborting [sc %p itt 0x%x]\n",
+		     sc, task->itt);
 
 	/* task completed before time out */
 	if (!task->sc) {
-		ISCSI_DBG_SESSION(session, "sc completed while abort in "
-				  "progress\n");
+		ISCSI_DBG_EH(session, "sc completed while abort in progress\n");
 		goto success;
 	}
 
@@ -1966,8 +1988,8 @@ int iscsi_eh_abort(struct scsi_cmnd *sc)
 		if (!sc->SCp.ptr) {
 			conn->tmf_state = TMF_INITIAL;
 			/* task completed before tmf abort response */
-			ISCSI_DBG_SESSION(session, "sc completed while abort "
-					  "in progress\n");
+			ISCSI_DBG_EH(session, "sc completed while abort	in "
+					      "progress\n");
 			goto success;
 		}
 		/* fall through */
@@ -1979,16 +2001,16 @@ int iscsi_eh_abort(struct scsi_cmnd *sc)
 success:
 	spin_unlock_bh(&session->lock);
 success_unlocked:
-	ISCSI_DBG_SESSION(session, "abort success [sc %p itt 0x%x]\n",
-			  sc, task->itt);
+	ISCSI_DBG_EH(session, "abort success [sc %p itt 0x%x]\n",
+		     sc, task->itt);
 	mutex_unlock(&session->eh_mutex);
 	return SUCCESS;
 
 failed:
 	spin_unlock_bh(&session->lock);
 failed_unlocked:
-	ISCSI_DBG_SESSION(session, "abort failed [sc %p itt 0x%x]\n", sc,
-			  task ? task->itt : 0);
+	ISCSI_DBG_EH(session, "abort failed [sc %p itt 0x%x]\n", sc,
+		     task ? task->itt : 0);
 	mutex_unlock(&session->eh_mutex);
 	return FAILED;
 }
@@ -2015,8 +2037,7 @@ int iscsi_eh_device_reset(struct scsi_cmnd *sc)
 	cls_session = starget_to_session(scsi_target(sc->device));
 	session = cls_session->dd_data;
 
-	ISCSI_DBG_SESSION(session, "LU Reset [sc %p lun %u]\n",
-			  sc, sc->device->lun);
+	ISCSI_DBG_EH(session, "LU Reset [sc %p lun %u]\n", sc, sc->device->lun);
 
 	mutex_lock(&session->eh_mutex);
 	spin_lock_bh(&session->lock);
@@ -2070,8 +2091,8 @@ int iscsi_eh_device_reset(struct scsi_cmnd *sc)
 unlock:
 	spin_unlock_bh(&session->lock);
 done:
-	ISCSI_DBG_SESSION(session, "dev reset result = %s\n",
-			 rc == SUCCESS ? "SUCCESS" : "FAILED");
+	ISCSI_DBG_EH(session, "dev reset result = %s\n",
+		     rc == SUCCESS ? "SUCCESS" : "FAILED");
 	mutex_unlock(&session->eh_mutex);
 	return rc;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 93bdcba5a7e55307e27671594c3cd8b4669b9e7a Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Wed, 17 Jun 2009 15:10:10 +0900
Subject: scsi_transport_sas: needs to call blk_end_request_all for SMP
 requests

We need to call blk_end_request_all to complete SMP requests properly.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_transport_sas.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/scsi_transport_sas.c b/drivers/scsi/scsi_transport_sas.c
index d606452297cf..0895d3c71b03 100644
--- a/drivers/scsi/scsi_transport_sas.c
+++ b/drivers/scsi/scsi_transport_sas.c
@@ -173,9 +173,9 @@ static void sas_smp_request(struct request_queue *q, struct Scsi_Host *shost,
 		ret = handler(shost, rphy, req);
 		req->errors = ret;
 
-		spin_lock_irq(q->queue_lock);
+		blk_end_request_all(req, ret);
 
-		req->end_io(req, ret);
+		spin_lock_irq(q->queue_lock);
 	}
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 111986593561fc4c94a1fba3f3cd84476fb40b22 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Wed, 17 Jun 2009 15:10:11 +0900
Subject: block: revert "bsg: setting rq->bio to NULL"

The SMP handler (sas_smp_request) was fixed to use the block API
properly, so we don't need this workaround to avoid blk_put_request()
warning.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 block/bsg.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/block/bsg.c b/block/bsg.c
index 54106f052f70..e7d475254248 100644
--- a/block/bsg.c
+++ b/block/bsg.c
@@ -315,7 +315,6 @@ out:
 	blk_put_request(rq);
 	if (next_rq) {
 		blk_rq_unmap_user(next_rq->bio);
-		next_rq->bio = NULL;
 		blk_put_request(next_rq);
 	}
 	return ERR_PTR(ret);
@@ -449,7 +448,6 @@ static int blk_complete_sgv4_hdr_rq(struct request *rq, struct sg_io_v4 *hdr,
 		hdr->dout_resid = rq->resid_len;
 		hdr->din_resid = rq->next_rq->resid_len;
 		blk_rq_unmap_user(bidi_bio);
-		rq->next_rq->bio = NULL;
 		blk_put_request(rq->next_rq);
 	} else if (rq_data_dir(rq) == READ)
 		hdr->din_resid = rq->resid_len;
@@ -468,7 +466,6 @@ static int blk_complete_sgv4_hdr_rq(struct request *rq, struct sg_io_v4 *hdr,
 	blk_rq_unmap_user(bio);
 	if (rq->cmd != rq->__cmd)
 		kfree(rq->cmd);
-	rq->bio = NULL;
 	blk_put_request(rq);
 
 	return ret;
-- 
cgit v1.2.3-59-g8ed1b


From 30c9afa6cc477f6f21f8a0b36f3b81080941a0c9 Mon Sep 17 00:00:00 2001
From: Joe Eykholt <jeykholt@cisco.com>
Date: Mon, 15 Jun 2009 23:22:14 -0700
Subject: fix race that can give duplicate host number

Just once, two fcoe instances got the same host number
from scsi_add_host().

Use atomic_t and atomic_inc_return() to get next host number.
Subtract 1, so that scsi_host still starts with 0.

[jejb: added comment about unusual subtraction]
Signed-off-by: Joe Eykholt <jeykholt@cisco.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/hosts.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c
index 89d41a424b33..5fd2da494d08 100644
--- a/drivers/scsi/hosts.c
+++ b/drivers/scsi/hosts.c
@@ -40,7 +40,7 @@
 #include "scsi_logging.h"
 
 
-static int scsi_host_next_hn;		/* host_no for next new host */
+static atomic_t scsi_host_next_hn;	/* host_no for next new host */
 
 
 static void scsi_host_cls_release(struct device *dev)
@@ -333,7 +333,11 @@ struct Scsi_Host *scsi_host_alloc(struct scsi_host_template *sht, int privsize)
 
 	mutex_init(&shost->scan_mutex);
 
-	shost->host_no = scsi_host_next_hn++; /* XXX(hch): still racy */
+	/*
+	 * subtract one because we increment first then return, but we need to
+	 * know what the next host number was before increment
+	 */
+	shost->host_no = atomic_inc_return(&scsi_host_next_hn) - 1;
 	shost->dma_channel = 0xff;
 
 	/* These three are default values which can be overridden */
-- 
cgit v1.2.3-59-g8ed1b


From 27dc9c5a3d652b0d55ab9ab396dcce9f13bc77c3 Mon Sep 17 00:00:00 2001
From: Anirban Chakraborty <anirban.chakraborty@qlogic.com>
Date: Wed, 17 Jun 2009 10:30:28 -0700
Subject: qla2xxx: Fixed a bug in number of response queue creation logic.

Cc: stable@kernel.org
Signed-off-by: Anirban Chakraborty <anirban.chakraborty@qlogic.com>
Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/qla2xxx/qla_os.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index dcf011679c8b..f0396e79b6fa 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -1663,7 +1663,7 @@ skip_pio:
 		/* queue 0 uses two msix vectors */
 		if (ql2xmultique_tag) {
 			cpus = num_online_cpus();
-			ha->max_rsp_queues = (ha->msix_count - 1 - cpus) ?
+			ha->max_rsp_queues = (ha->msix_count - 1 > cpus) ?
 				(cpus + 1) : (ha->msix_count - 1);
 			ha->max_req_queues = 2;
 		} else if (ql2xmaxqueues > 1) {
-- 
cgit v1.2.3-59-g8ed1b


From 1bb395485160d203a726a19e4fcb1a154748d804 Mon Sep 17 00:00:00 2001
From: Harish Zunjarrao <harish.zunjarrao@qlogic.com>
Date: Wed, 17 Jun 2009 10:30:29 -0700
Subject: qla2xxx: Correct iiDMA-update calling conventions.

* To set iiDMA speeds for ISP81XX, bits 5-0 are used whereas for
  other older ISPs bits 2-0 are used.
* Pass proper VP index

Signed-off-by: Harish Zunjarrao <harish.zunjarrao@qlogic.com>
Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/qla2xxx/qla_init.c |  2 +-
 drivers/scsi/qla2xxx/qla_mbx.c  | 13 +++++++------
 2 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
index 262026129325..f2ce8e3cc91b 100644
--- a/drivers/scsi/qla2xxx/qla_init.c
+++ b/drivers/scsi/qla2xxx/qla_init.c
@@ -2301,7 +2301,7 @@ qla2x00_iidma_fcport(scsi_qla_host_t *vha, fc_port_t *fcport)
 	static char *link_speeds[] = { "1", "2", "?", "4", "8", "10" };
 	char *link_speed;
 	int rval;
-	uint16_t mb[6];
+	uint16_t mb[4];
 	struct qla_hw_data *ha = vha->hw;
 
 	if (!IS_IIDMA_CAPABLE(ha))
diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c
index 451ece0760b0..779ce14e9a2a 100644
--- a/drivers/scsi/qla2xxx/qla_mbx.c
+++ b/drivers/scsi/qla2xxx/qla_mbx.c
@@ -2697,10 +2697,13 @@ qla2x00_set_idma_speed(scsi_qla_host_t *vha, uint16_t loop_id,
 	mcp->mb[0] = MBC_PORT_PARAMS;
 	mcp->mb[1] = loop_id;
 	mcp->mb[2] = BIT_0;
-	mcp->mb[3] = port_speed & (BIT_2|BIT_1|BIT_0);
-	mcp->mb[4] = mcp->mb[5] = 0;
-	mcp->out_mb = MBX_5|MBX_4|MBX_3|MBX_2|MBX_1|MBX_0;
-	mcp->in_mb = MBX_5|MBX_4|MBX_3|MBX_1|MBX_0;
+	if (IS_QLA81XX(vha->hw))
+		mcp->mb[3] = port_speed & (BIT_5|BIT_4|BIT_3|BIT_2|BIT_1|BIT_0);
+	else
+		mcp->mb[3] = port_speed & (BIT_2|BIT_1|BIT_0);
+	mcp->mb[9] = vha->vp_idx;
+	mcp->out_mb = MBX_9|MBX_3|MBX_2|MBX_1|MBX_0;
+	mcp->in_mb = MBX_3|MBX_1|MBX_0;
 	mcp->tov = MBX_TOV_SECONDS;
 	mcp->flags = 0;
 	rval = qla2x00_mailbox_command(vha, mcp);
@@ -2710,8 +2713,6 @@ qla2x00_set_idma_speed(scsi_qla_host_t *vha, uint16_t loop_id,
 		mb[0] = mcp->mb[0];
 		mb[1] = mcp->mb[1];
 		mb[3] = mcp->mb[3];
-		mb[4] = mcp->mb[4];
-		mb[5] = mcp->mb[5];
 	}
 
 	if (rval != QLA_SUCCESS) {
-- 
cgit v1.2.3-59-g8ed1b


From 9d2683c05ce57b4a0231e028927bf1197e8324a8 Mon Sep 17 00:00:00 2001
From: Andrew Vasquez <andrew.vasquez@qlogic.com>
Date: Wed, 17 Jun 2009 10:30:30 -0700
Subject: qla2xxx: Limit querying to supported mailbox-registers while reading
 FW state.

Pre-ISP24xx chips have dedicated uses for mailbox 4 and 5 which
software should typically not query nor update.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/qla2xxx/qla_mbx.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c
index 779ce14e9a2a..fe69f3057671 100644
--- a/drivers/scsi/qla2xxx/qla_mbx.c
+++ b/drivers/scsi/qla2xxx/qla_mbx.c
@@ -1267,17 +1267,22 @@ qla2x00_get_firmware_state(scsi_qla_host_t *vha, uint16_t *states)
 
 	mcp->mb[0] = MBC_GET_FIRMWARE_STATE;
 	mcp->out_mb = MBX_0;
-	mcp->in_mb = MBX_5|MBX_4|MBX_3|MBX_2|MBX_1|MBX_0;
+	if (IS_FWI2_CAPABLE(vha->hw))
+		mcp->in_mb = MBX_5|MBX_4|MBX_3|MBX_2|MBX_1|MBX_0;
+	else
+		mcp->in_mb = MBX_1|MBX_0;
 	mcp->tov = MBX_TOV_SECONDS;
 	mcp->flags = 0;
 	rval = qla2x00_mailbox_command(vha, mcp);
 
 	/* Return firmware states. */
 	states[0] = mcp->mb[1];
-	states[1] = mcp->mb[2];
-	states[2] = mcp->mb[3];
-	states[3] = mcp->mb[4];
-	states[4] = mcp->mb[5];
+	if (IS_FWI2_CAPABLE(vha->hw)) {
+		states[1] = mcp->mb[2];
+		states[2] = mcp->mb[3];
+		states[3] = mcp->mb[4];
+		states[4] = mcp->mb[5];
+	}
 
 	if (rval != QLA_SUCCESS) {
 		/*EMPTY*/
-- 
cgit v1.2.3-59-g8ed1b


From e18e963b7e533149676b5d387d0a56160df9f111 Mon Sep 17 00:00:00 2001
From: Andrew Vasquez <andrew.vasquez@qlogic.com>
Date: Wed, 17 Jun 2009 10:30:31 -0700
Subject: qla2xxx: Correct (again) overflow during dump-processing on
 large-memory ISP23xx parts.

Commit 7b867cf76fbcc8d77867cbec6f509f71dce8a98f ([SCSI] qla2xxx:
Refactor qla data structures) inadvertently reverted
e792121ec85672c1fa48f79d13986a3f4f56c590 ([SCSI] qla2xxx: Correct
overflow during dump-processing on large-memory ISP23xx parts.).

Cc: stable@kernel.org
Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/qla2xxx/qla_dbg.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/qla2xxx/qla_dbg.c b/drivers/scsi/qla2xxx/qla_dbg.c
index 4a990f4da4ea..cca8e4ab0372 100644
--- a/drivers/scsi/qla2xxx/qla_dbg.c
+++ b/drivers/scsi/qla2xxx/qla_dbg.c
@@ -216,7 +216,7 @@ qla24xx_soft_reset(struct qla_hw_data *ha)
 
 static int
 qla2xxx_dump_ram(struct qla_hw_data *ha, uint32_t addr, uint16_t *ram,
-    uint16_t ram_words, void **nxt)
+    uint32_t ram_words, void **nxt)
 {
 	int rval;
 	uint32_t cnt, stat, timer, words, idx;
-- 
cgit v1.2.3-59-g8ed1b


From e0420029dedb867e5c2c044a0737338a7e9cec0b Mon Sep 17 00:00:00 2001
From: Andrew Vasquez <andrew.vasquez@qlogic.com>
Date: Wed, 17 Jun 2009 10:30:32 -0700
Subject: qla2xxx: Update version number to 8.03.01-k4.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/qla2xxx/qla_version.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/qla2xxx/qla_version.h b/drivers/scsi/qla2xxx/qla_version.h
index b63feaf43126..84369705a9ad 100644
--- a/drivers/scsi/qla2xxx/qla_version.h
+++ b/drivers/scsi/qla2xxx/qla_version.h
@@ -7,7 +7,7 @@
 /*
  * Driver version
  */
-#define QLA2XXX_VERSION      "8.03.01-k3"
+#define QLA2XXX_VERSION      "8.03.01-k4"
 
 #define QLA_DRIVER_MAJOR_VER	8
 #define QLA_DRIVER_MINOR_VER	3
-- 
cgit v1.2.3-59-g8ed1b


From 598fa4b775d064d4656132c78d9a312eb1d2f91f Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Wed, 17 Jun 2009 15:01:58 -0400
Subject: enhance device info matching for multiple tables

The current scsi_devinfo.c matching routines use a single table for
the global blacklist.  However, we're developing a need to blacklist
from specific transports too (notably some tape drives using SPI which
don't respond well to high speed protocols).  Instead of developing
separate blacklist matching for each transport class needing it,
enhance the current list matching to permit multiple lists.

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_devinfo.c | 247 ++++++++++++++++++++++++++++++++++++++++----
 drivers/scsi/scsi_priv.h    |  15 +++
 2 files changed, 240 insertions(+), 22 deletions(-)

diff --git a/drivers/scsi/scsi_devinfo.c b/drivers/scsi/scsi_devinfo.c
index 8821df9a277b..93c2622cb969 100644
--- a/drivers/scsi/scsi_devinfo.c
+++ b/drivers/scsi/scsi_devinfo.c
@@ -24,6 +24,13 @@ struct scsi_dev_info_list {
 	unsigned compatible; /* for use with scsi_static_device_list entries */
 };
 
+struct scsi_dev_info_list_table {
+	struct list_head node;	/* our node for being on the master list */
+	struct list_head scsi_dev_info_list; /* head of dev info list */
+	const char *name;	/* name of list for /proc (NULL for global) */
+	int key;		/* unique numeric identifier */
+};
+
 
 static const char spaces[] = "                "; /* 16 of them */
 static unsigned scsi_default_dev_flags;
@@ -247,6 +254,22 @@ static struct {
 	{ NULL, NULL, NULL, 0 },
 };
 
+static struct scsi_dev_info_list_table *scsi_devinfo_lookup_by_key(int key)
+{
+	struct scsi_dev_info_list_table *devinfo_table;
+	int found = 0;
+
+	list_for_each_entry(devinfo_table, &scsi_dev_info_list, node)
+		if (devinfo_table->key == key) {
+			found = 1;
+			break;
+		}
+	if (!found)
+		return ERR_PTR(-EINVAL);
+
+	return devinfo_table;
+}
+
 /*
  * scsi_strcpy_devinfo: called from scsi_dev_info_list_add to copy into
  * devinfo vendor and model strings.
@@ -295,8 +318,39 @@ static void scsi_strcpy_devinfo(char *name, char *to, size_t to_length,
  **/
 static int scsi_dev_info_list_add(int compatible, char *vendor, char *model,
 			    char *strflags, int flags)
+{
+	return scsi_dev_info_list_add_keyed(compatible, vendor, model,
+					    strflags, flags,
+					    SCSI_DEVINFO_GLOBAL);
+}
+
+/**
+ * scsi_dev_info_list_add_keyed - add one dev_info list entry.
+ * @compatible: if true, null terminate short strings.  Otherwise space pad.
+ * @vendor:	vendor string
+ * @model:	model (product) string
+ * @strflags:	integer string
+ * @flags:	if strflags NULL, use this flag value
+ * @key:	specify list to use
+ *
+ * Description:
+ * 	Create and add one dev_info entry for @vendor, @model,
+ * 	@strflags or @flag in list specified by @key. If @compatible,
+ * 	add to the tail of the list, do not space pad, and set
+ * 	devinfo->compatible. The scsi_static_device_list entries are
+ * 	added with @compatible 1 and @clfags NULL.
+ *
+ * Returns: 0 OK, -error on failure.
+ **/
+int scsi_dev_info_list_add_keyed(int compatible, char *vendor, char *model,
+				 char *strflags, int flags, int key)
 {
 	struct scsi_dev_info_list *devinfo;
+	struct scsi_dev_info_list_table *devinfo_table =
+		scsi_devinfo_lookup_by_key(key);
+
+	if (IS_ERR(devinfo_table))
+		return PTR_ERR(devinfo_table);
 
 	devinfo = kmalloc(sizeof(*devinfo), GFP_KERNEL);
 	if (!devinfo) {
@@ -317,12 +371,15 @@ static int scsi_dev_info_list_add(int compatible, char *vendor, char *model,
 	devinfo->compatible = compatible;
 
 	if (compatible)
-		list_add_tail(&devinfo->dev_info_list, &scsi_dev_info_list);
+		list_add_tail(&devinfo->dev_info_list,
+			      &devinfo_table->scsi_dev_info_list);
 	else
-		list_add(&devinfo->dev_info_list, &scsi_dev_info_list);
+		list_add(&devinfo->dev_info_list,
+			 &devinfo_table->scsi_dev_info_list);
 
 	return 0;
 }
+EXPORT_SYMBOL(scsi_dev_info_list_add_keyed);
 
 /**
  * scsi_dev_info_list_add_str - parse dev_list and add to the scsi_dev_info_list.
@@ -382,22 +439,48 @@ static int scsi_dev_info_list_add_str(char *dev_list)
  * @model:	model name
  *
  * Description:
- *     Search the scsi_dev_info_list for an entry matching @vendor and
- *     @model, if found, return the matching flags value, else return
- *     the host or global default settings.  Called during scan time.
+ *     Search the global scsi_dev_info_list (specified by list zero)
+ *     for an entry matching @vendor and @model, if found, return the
+ *     matching flags value, else return the host or global default
+ *     settings.  Called during scan time.
  **/
 int scsi_get_device_flags(struct scsi_device *sdev,
 			  const unsigned char *vendor,
 			  const unsigned char *model)
+{
+	return scsi_get_device_flags_keyed(sdev, vendor, model,
+					   SCSI_DEVINFO_GLOBAL);
+}
+
+
+/**
+ * get_device_flags_keyed - get device specific flags from the dynamic device list.
+ * @sdev:       &scsi_device to get flags for
+ * @vendor:	vendor name
+ * @model:	model name
+ * @key:	list to look up
+ *
+ * Description:
+ *     Search the scsi_dev_info_list specified by @key for an entry
+ *     matching @vendor and @model, if found, return the matching
+ *     flags value, else return the host or global default settings.
+ *     Called during scan time.
+ **/
+int scsi_get_device_flags_keyed(struct scsi_device *sdev,
+				const unsigned char *vendor,
+				const unsigned char *model,
+				int key)
 {
 	struct scsi_dev_info_list *devinfo;
-	unsigned int bflags;
+	struct scsi_dev_info_list_table *devinfo_table;
+
+	devinfo_table = scsi_devinfo_lookup_by_key(key);
 
-	bflags = sdev->sdev_bflags;
-	if (!bflags)
-		bflags = scsi_default_dev_flags;
+	if (IS_ERR(devinfo_table))
+		return PTR_ERR(devinfo_table);
 
-	list_for_each_entry(devinfo, &scsi_dev_info_list, dev_info_list) {
+	list_for_each_entry(devinfo, &devinfo_table->scsi_dev_info_list,
+			    dev_info_list) {
 		if (devinfo->compatible) {
 			/*
 			 * Behave like the older version of get_device_flags.
@@ -447,32 +530,89 @@ int scsi_get_device_flags(struct scsi_device *sdev,
 				return devinfo->flags;
 		}
 	}
-	return bflags;
+	/* nothing found, return nothing */
+	if (key != SCSI_DEVINFO_GLOBAL)
+		return 0;
+
+	/* except for the global list, where we have an exception */
+	if (sdev->sdev_bflags)
+		return sdev->sdev_bflags;
+
+	return scsi_default_dev_flags;
 }
+EXPORT_SYMBOL(scsi_get_device_flags_keyed);
 
 #ifdef CONFIG_SCSI_PROC_FS
+struct double_list {
+	struct list_head *top;
+	struct list_head *bottom;
+};
+
 static int devinfo_seq_show(struct seq_file *m, void *v)
 {
+	struct double_list *dl = v;
+	struct scsi_dev_info_list_table *devinfo_table =
+		list_entry(dl->top, struct scsi_dev_info_list_table, node);
 	struct scsi_dev_info_list *devinfo =
-		list_entry(v, struct scsi_dev_info_list, dev_info_list);
+		list_entry(dl->bottom, struct scsi_dev_info_list,
+			   dev_info_list);
+
+	if (devinfo_table->scsi_dev_info_list.next == dl->bottom &&
+	    devinfo_table->name)
+		seq_printf(m, "[%s]:\n", devinfo_table->name);
 
 	seq_printf(m, "'%.8s' '%.16s' 0x%x\n",
-			devinfo->vendor, devinfo->model, devinfo->flags);
+		   devinfo->vendor, devinfo->model, devinfo->flags);
 	return 0;
 }
 
-static void * devinfo_seq_start(struct seq_file *m, loff_t *pos)
+static void *devinfo_seq_start(struct seq_file *m, loff_t *ppos)
 {
-	return seq_list_start(&scsi_dev_info_list, *pos);
+	struct double_list *dl = kmalloc(sizeof(*dl), GFP_KERNEL);
+	loff_t pos = *ppos;
+
+	if (!dl)
+		return NULL;
+
+	list_for_each(dl->top, &scsi_dev_info_list) {
+		struct scsi_dev_info_list_table *devinfo_table =
+			list_entry(dl->top, struct scsi_dev_info_list_table,
+				   node);
+		list_for_each(dl->bottom, &devinfo_table->scsi_dev_info_list)
+			if (pos-- == 0)
+				return dl;
+	}
+
+	kfree(dl);
+	return NULL;
 }
 
-static void * devinfo_seq_next(struct seq_file *m, void *v, loff_t *pos)
+static void *devinfo_seq_next(struct seq_file *m, void *v, loff_t *ppos)
 {
-	return seq_list_next(v, &scsi_dev_info_list, pos);
+	struct double_list *dl = v;
+	struct scsi_dev_info_list_table *devinfo_table =
+		list_entry(dl->top, struct scsi_dev_info_list_table, node);
+
+	++*ppos;
+	dl->bottom = dl->bottom->next;
+	while (&devinfo_table->scsi_dev_info_list == dl->bottom) {
+		dl->top = dl->top->next;
+		if (dl->top == &scsi_dev_info_list) {
+			kfree(dl);
+			return NULL;
+		}
+		devinfo_table = list_entry(dl->top,
+					   struct scsi_dev_info_list_table,
+					   node);
+		dl->bottom = devinfo_table->scsi_dev_info_list.next;
+	}
+
+	return dl;
 }
 
 static void devinfo_seq_stop(struct seq_file *m, void *v)
 {
+	kfree(v);
 }
 
 static const struct seq_operations scsi_devinfo_seq_ops = {
@@ -549,19 +689,78 @@ MODULE_PARM_DESC(default_dev_flags,
  **/
 void scsi_exit_devinfo(void)
 {
-	struct list_head *lh, *lh_next;
-	struct scsi_dev_info_list *devinfo;
-
 #ifdef CONFIG_SCSI_PROC_FS
 	remove_proc_entry("scsi/device_info", NULL);
 #endif
 
-	list_for_each_safe(lh, lh_next, &scsi_dev_info_list) {
+	scsi_dev_info_remove_list(SCSI_DEVINFO_GLOBAL);
+}
+
+/**
+ * scsi_dev_info_add_list - add a new devinfo list
+ * @key:	key of the list to add
+ * @name:	Name of the list to add (for /proc/scsi/device_info)
+ *
+ * Adds the requested list, returns zero on success, -EEXIST if the
+ * key is already registered to a list, or other error on failure.
+ */
+int scsi_dev_info_add_list(int key, const char *name)
+{
+	struct scsi_dev_info_list_table *devinfo_table =
+		scsi_devinfo_lookup_by_key(key);
+
+	if (!IS_ERR(devinfo_table))
+		/* list already exists */
+		return -EEXIST;
+
+	devinfo_table = kmalloc(sizeof(*devinfo_table), GFP_KERNEL);
+
+	if (!devinfo_table)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&devinfo_table->node);
+	INIT_LIST_HEAD(&devinfo_table->scsi_dev_info_list);
+	devinfo_table->name = name;
+	devinfo_table->key = key;
+	list_add_tail(&devinfo_table->node, &scsi_dev_info_list);
+
+	return 0;
+}
+EXPORT_SYMBOL(scsi_dev_info_add_list);
+
+/**
+ * scsi_dev_info_remove_list - destroy an added devinfo list
+ * @key: key of the list to destroy
+ *
+ * Iterates over the entire list first, freeing all the values, then
+ * frees the list itself.  Returns 0 on success or -EINVAL if the key
+ * can't be found.
+ */
+int scsi_dev_info_remove_list(int key)
+{
+	struct list_head *lh, *lh_next;
+	struct scsi_dev_info_list_table *devinfo_table =
+		scsi_devinfo_lookup_by_key(key);
+
+	if (IS_ERR(devinfo_table))
+		/* no such list */
+		return -EINVAL;
+
+	/* remove from the master list */
+	list_del(&devinfo_table->node);
+
+	list_for_each_safe(lh, lh_next, &devinfo_table->scsi_dev_info_list) {
+		struct scsi_dev_info_list *devinfo;
+
 		devinfo = list_entry(lh, struct scsi_dev_info_list,
 				     dev_info_list);
 		kfree(devinfo);
 	}
+	kfree(devinfo_table);
+
+	return 0;
 }
+EXPORT_SYMBOL(scsi_dev_info_remove_list);
 
 /**
  * scsi_init_devinfo - set up the dynamic device list.
@@ -577,10 +776,14 @@ int __init scsi_init_devinfo(void)
 #endif
 	int error, i;
 
-	error = scsi_dev_info_list_add_str(scsi_dev_flags);
+	error = scsi_dev_info_add_list(SCSI_DEVINFO_GLOBAL, NULL);
 	if (error)
 		return error;
 
+	error = scsi_dev_info_list_add_str(scsi_dev_flags);
+	if (error)
+		goto out;
+
 	for (i = 0; scsi_static_device_list[i].vendor; i++) {
 		error = scsi_dev_info_list_add(1 /* compatibile */,
 				scsi_static_device_list[i].vendor,
diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h
index fbc83bebdd8e..b4e49cd6e75d 100644
--- a/drivers/scsi/scsi_priv.h
+++ b/drivers/scsi/scsi_priv.h
@@ -39,9 +39,24 @@ static inline void scsi_log_completion(struct scsi_cmnd *cmd, int disposition)
 #endif
 
 /* scsi_devinfo.c */
+
+/* list of keys for the lists */
+enum {
+	SCSI_DEVINFO_GLOBAL = 0,
+};
+
 extern int scsi_get_device_flags(struct scsi_device *sdev,
 				 const unsigned char *vendor,
 				 const unsigned char *model);
+extern int scsi_get_device_flags_keyed(struct scsi_device *sdev,
+				       const unsigned char *vendor,
+				       const unsigned char *model, int key);
+extern int scsi_dev_info_list_add_keyed(int compatible, char *vendor,
+					char *model, char *strflags,
+					int flags, int key);
+extern int scsi_dev_info_add_list(int key, const char *name);
+extern int scsi_dev_info_remove_list(int key);
+
 extern int __init scsi_init_devinfo(void);
 extern void scsi_exit_devinfo(void);
 
-- 
cgit v1.2.3-59-g8ed1b


From 9872b81cf9f4b163e9c558d79e76b832c58a4814 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Wed, 17 Jun 2009 15:03:41 -0400
Subject: scsi_transport_spi: use spi target settings instead of inquiry data
 for DV

Right at the moment, we carefully set up the spi_support_xx in the
device configuration routines, but then we never actually use the
results: we rely on the inquiry strings.  If we're going to allow
overrides to the inquiry data, we have to rely on our own internal
settings.

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_transport_spi.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/scsi/scsi_transport_spi.c b/drivers/scsi/scsi_transport_spi.c
index 654a34fb04cb..00cfb40b5efa 100644
--- a/drivers/scsi/scsi_transport_spi.c
+++ b/drivers/scsi/scsi_transport_spi.c
@@ -833,7 +833,7 @@ spi_dv_device_internal(struct scsi_device *sdev, u8 *buffer)
 		return;
 	}
 
-	if (!scsi_device_wide(sdev)) {
+	if (!spi_support_wide(starget)) {
 		spi_max_width(starget) = 0;
 		max_width = 0;
 	}
@@ -860,7 +860,7 @@ spi_dv_device_internal(struct scsi_device *sdev, u8 *buffer)
 		return;
 
 	/* device can't handle synchronous */
-	if (!scsi_device_sync(sdev) && !scsi_device_dt(sdev))
+	if (!spi_support_sync(starget) && !spi_support_dt(starget))
 		return;
 
 	/* len == -1 is the signal that we need to ascertain the
@@ -876,13 +876,14 @@ spi_dv_device_internal(struct scsi_device *sdev, u8 *buffer)
 
 	/* try QAS requests; this should be harmless to set if the
 	 * target supports it */
-	if (scsi_device_qas(sdev) && spi_max_qas(starget)) {
+	if (spi_support_qas(starget) && spi_max_qas(starget)) {
 		DV_SET(qas, 1);
 	} else {
 		DV_SET(qas, 0);
 	}
 
-	if (scsi_device_ius(sdev) && spi_max_iu(starget) && min_period < 9) {
+	if (spi_support_ius(starget) && spi_max_iu(starget) &&
+	    min_period < 9) {
 		/* This u320 (or u640). Set IU transfers */
 		DV_SET(iu, 1);
 		/* Then set the optional parameters */
@@ -902,7 +903,7 @@ spi_dv_device_internal(struct scsi_device *sdev, u8 *buffer)
 		i->f->get_signalling(shost);
 	if (spi_signalling(shost) == SPI_SIGNAL_SE ||
 	    spi_signalling(shost) == SPI_SIGNAL_HVD ||
-	    !scsi_device_dt(sdev)) {
+	    !spi_support_dt(starget)) {
 		DV_SET(dt, 0);
 	} else {
 		DV_SET(dt, 1);
-- 
cgit v1.2.3-59-g8ed1b


From a9e0edb687151617fe89cc5ab0086ebfc73ffccb Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Wed, 17 Jun 2009 19:05:05 +0000
Subject: scsi_transport_spi: Blacklist Ultrium-3 tape for IU transfers

There have been several bug reports which identified the Ultrium-3
tape as just hanging up on the bus during certain types of IU
transfer.  The identified culpret is type 0x02 (MULTIPLE COMMAND)
transfers.  The only way to prevent this tape wedging is to prevent it
from using IU transfers at all.  So this patch uses the exported
blacklist matching technology to recognise the drive and force it not
to use IU transfers.

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_priv.h          |  1 +
 drivers/scsi/scsi_transport_spi.c | 40 ++++++++++++++++++++++++++++++++++++++-
 2 files changed, 40 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h
index b4e49cd6e75d..00264aab8c8a 100644
--- a/drivers/scsi/scsi_priv.h
+++ b/drivers/scsi/scsi_priv.h
@@ -43,6 +43,7 @@ static inline void scsi_log_completion(struct scsi_cmnd *cmd, int disposition)
 /* list of keys for the lists */
 enum {
 	SCSI_DEVINFO_GLOBAL = 0,
+	SCSI_DEVINFO_SPI,
 };
 
 extern int scsi_get_device_flags(struct scsi_device *sdev,
diff --git a/drivers/scsi/scsi_transport_spi.c b/drivers/scsi/scsi_transport_spi.c
index 00cfb40b5efa..c25bd9a34e02 100644
--- a/drivers/scsi/scsi_transport_spi.c
+++ b/drivers/scsi/scsi_transport_spi.c
@@ -46,6 +46,22 @@
 #define DV_RETRIES	3	/* should only need at most 
 				 * two cc/ua clears */
 
+/* Our blacklist flags */
+enum {
+	SPI_BLIST_NOIUS = 0x1,
+};
+
+/* blacklist table, modelled on scsi_devinfo.c */
+static struct {
+	char *vendor;
+	char *model;
+	unsigned flags;
+} spi_static_device_list[] __initdata = {
+	{"HP", "Ultrium 3-SCSI", SPI_BLIST_NOIUS },
+	{"IBM", "ULTRIUM-TD3", SPI_BLIST_NOIUS },
+	{NULL, NULL, 0}
+};
+
 /* Private data accessors (keep these out of the header file) */
 #define spi_dv_in_progress(x) (((struct spi_transport_attrs *)&(x)->starget_data)->dv_in_progress)
 #define spi_dv_mutex(x) (((struct spi_transport_attrs *)&(x)->starget_data)->dv_mutex)
@@ -207,6 +223,9 @@ static int spi_device_configure(struct transport_container *tc,
 {
 	struct scsi_device *sdev = to_scsi_device(dev);
 	struct scsi_target *starget = sdev->sdev_target;
+	unsigned bflags = scsi_get_device_flags_keyed(sdev, &sdev->inquiry[8],
+						      &sdev->inquiry[16],
+						      SCSI_DEVINFO_SPI);
 
 	/* Populate the target capability fields with the values
 	 * gleaned from the device inquiry */
@@ -216,6 +235,10 @@ static int spi_device_configure(struct transport_container *tc,
 	spi_support_dt(starget) = scsi_device_dt(sdev);
 	spi_support_dt_only(starget) = scsi_device_dt_only(sdev);
 	spi_support_ius(starget) = scsi_device_ius(sdev);
+	if (bflags & SPI_BLIST_NOIUS) {
+		dev_info(dev, "Information Units disabled by blacklist\n");
+		spi_support_ius(starget) = 0;
+	}
 	spi_support_qas(starget) = scsi_device_qas(sdev);
 
 	return 0;
@@ -1524,7 +1547,21 @@ EXPORT_SYMBOL(spi_release_transport);
 
 static __init int spi_transport_init(void)
 {
-	int error = transport_class_register(&spi_transport_class);
+	int error = scsi_dev_info_add_list(SCSI_DEVINFO_SPI,
+					   "SCSI Parallel Transport Class");
+	if (!error) {
+		int i;
+
+		for (i = 0; spi_static_device_list[i].vendor; i++)
+			scsi_dev_info_list_add_keyed(1,	/* compatible */
+						     spi_static_device_list[i].vendor,
+						     spi_static_device_list[i].model,
+						     NULL,
+						     spi_static_device_list[i].flags,
+						     SCSI_DEVINFO_SPI);
+	}
+
+	error = transport_class_register(&spi_transport_class);
 	if (error)
 		return error;
 	error = anon_transport_class_register(&spi_device_class);
@@ -1536,6 +1573,7 @@ static void __exit spi_transport_exit(void)
 	transport_class_unregister(&spi_transport_class);
 	anon_transport_class_unregister(&spi_device_class);
 	transport_class_unregister(&spi_host_class);
+	scsi_dev_info_remove_list(SCSI_DEVINFO_SPI);
 }
 
 MODULE_AUTHOR("Martin Hicks");
-- 
cgit v1.2.3-59-g8ed1b


From 95fecd90397ec1f85eb31ede955d846a86d2077b Mon Sep 17 00:00:00 2001
From: Wayne Boyer <wayneb@linux.vnet.ibm.com>
Date: Tue, 16 Jun 2009 15:13:28 -0700
Subject: ipr: add test for MSI interrupt support

The return value from pci_enable_msi() can not always be trusted.  This patch
adds code to generate an interrupt after MSI has been enabled and tests
whether or not we can receive and process it.  If the tests fails, then fall
back to LSI.

Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com>
Acked-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/ipr.c | 108 +++++++++++++++++++++++++++++++++++++++++++++++++----
 drivers/scsi/ipr.h |   6 ++-
 2 files changed, 105 insertions(+), 9 deletions(-)

diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index 0f8bc772b112..15ce8e51d5de 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -7367,6 +7367,7 @@ static void __devinit ipr_init_ioa_cfg(struct ipr_ioa_cfg *ioa_cfg,
 	INIT_LIST_HEAD(&ioa_cfg->used_res_q);
 	INIT_WORK(&ioa_cfg->work_q, ipr_worker_thread);
 	init_waitqueue_head(&ioa_cfg->reset_wait_q);
+	init_waitqueue_head(&ioa_cfg->msi_wait_q);
 	ioa_cfg->sdt_state = INACTIVE;
 	if (ipr_enable_cache)
 		ioa_cfg->cache_state = CACHE_ENABLED;
@@ -7416,6 +7417,89 @@ ipr_get_chip_cfg(const struct pci_device_id *dev_id)
 	return NULL;
 }
 
+/**
+ * ipr_test_intr - Handle the interrupt generated in ipr_test_msi().
+ * @pdev:		PCI device struct
+ *
+ * Description: Simply set the msi_received flag to 1 indicating that
+ * Message Signaled Interrupts are supported.
+ *
+ * Return value:
+ * 	0 on success / non-zero on failure
+ **/
+static irqreturn_t __devinit ipr_test_intr(int irq, void *devp)
+{
+	struct ipr_ioa_cfg *ioa_cfg = (struct ipr_ioa_cfg *)devp;
+	unsigned long lock_flags = 0;
+	irqreturn_t rc = IRQ_HANDLED;
+
+	spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
+
+	ioa_cfg->msi_received = 1;
+	wake_up(&ioa_cfg->msi_wait_q);
+
+	spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
+	return rc;
+}
+
+/**
+ * ipr_test_msi - Test for Message Signaled Interrupt (MSI) support.
+ * @pdev:		PCI device struct
+ *
+ * Description: The return value from pci_enable_msi() can not always be
+ * trusted.  This routine sets up and initiates a test interrupt to determine
+ * if the interrupt is received via the ipr_test_intr() service routine.
+ * If the tests fails, the driver will fall back to LSI.
+ *
+ * Return value:
+ * 	0 on success / non-zero on failure
+ **/
+static int __devinit ipr_test_msi(struct ipr_ioa_cfg *ioa_cfg,
+				  struct pci_dev *pdev)
+{
+	int rc;
+	volatile u32 int_reg;
+	unsigned long lock_flags = 0;
+
+	ENTER;
+
+	spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
+	init_waitqueue_head(&ioa_cfg->msi_wait_q);
+	ioa_cfg->msi_received = 0;
+	ipr_mask_and_clear_interrupts(ioa_cfg, ~IPR_PCII_IOA_TRANS_TO_OPER);
+	writel(IPR_PCII_IO_DEBUG_ACKNOWLEDGE, ioa_cfg->regs.clr_interrupt_mask_reg);
+	int_reg = readl(ioa_cfg->regs.sense_interrupt_mask_reg);
+	spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
+
+	rc = request_irq(pdev->irq, ipr_test_intr, 0, IPR_NAME, ioa_cfg);
+	if (rc) {
+		dev_err(&pdev->dev, "Can not assign irq %d\n", pdev->irq);
+		return rc;
+	} else if (ipr_debug)
+		dev_info(&pdev->dev, "IRQ assigned: %d\n", pdev->irq);
+
+	writel(IPR_PCII_IO_DEBUG_ACKNOWLEDGE, ioa_cfg->regs.sense_interrupt_reg);
+	int_reg = readl(ioa_cfg->regs.sense_interrupt_reg);
+	wait_event_timeout(ioa_cfg->msi_wait_q, ioa_cfg->msi_received, HZ);
+	ipr_mask_and_clear_interrupts(ioa_cfg, ~IPR_PCII_IOA_TRANS_TO_OPER);
+
+	spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags);
+	if (!ioa_cfg->msi_received) {
+		/* MSI test failed */
+		dev_info(&pdev->dev, "MSI test failed.  Falling back to LSI.\n");
+		rc = -EOPNOTSUPP;
+	} else if (ipr_debug)
+		dev_info(&pdev->dev, "MSI test succeeded.\n");
+
+	spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags);
+
+	free_irq(pdev->irq, ioa_cfg);
+
+	LEAVE;
+
+	return rc;
+}
+
 /**
  * ipr_probe_ioa - Allocates memory and does first stage of initialization
  * @pdev:		PCI device struct
@@ -7441,11 +7525,6 @@ static int __devinit ipr_probe_ioa(struct pci_dev *pdev,
 		goto out;
 	}
 
-	if (!(rc = pci_enable_msi(pdev)))
-		dev_info(&pdev->dev, "MSI enabled\n");
-	else if (ipr_debug)
-		dev_info(&pdev->dev, "Cannot enable MSI\n");
-
 	dev_info(&pdev->dev, "Found IOA with IRQ: %d\n", pdev->irq);
 
 	host = scsi_host_alloc(&driver_template, sizeof(*ioa_cfg));
@@ -7519,6 +7598,18 @@ static int __devinit ipr_probe_ioa(struct pci_dev *pdev,
 		goto cleanup_nomem;
 	}
 
+	/* Enable MSI style interrupts if they are supported. */
+	if (!(rc = pci_enable_msi(pdev))) {
+		rc = ipr_test_msi(ioa_cfg, pdev);
+		if (rc == -EOPNOTSUPP)
+			pci_disable_msi(pdev);
+		else if (rc)
+			goto out_msi_disable;
+		else
+			dev_info(&pdev->dev, "MSI enabled with IRQ: %d\n", pdev->irq);
+	} else if (ipr_debug)
+		dev_info(&pdev->dev, "Cannot enable MSI.\n");
+
 	/* Save away PCI config space for use following IOA reset */
 	rc = pci_save_state(pdev);
 
@@ -7556,7 +7647,9 @@ static int __devinit ipr_probe_ioa(struct pci_dev *pdev,
 		ioa_cfg->ioa_unit_checked = 1;
 
 	ipr_mask_and_clear_interrupts(ioa_cfg, ~IPR_PCII_IOA_TRANS_TO_OPER);
-	rc = request_irq(pdev->irq, ipr_isr, IRQF_SHARED, IPR_NAME, ioa_cfg);
+	rc = request_irq(pdev->irq, ipr_isr,
+			 ioa_cfg->msi_received ? 0 : IRQF_SHARED,
+			 IPR_NAME, ioa_cfg);
 
 	if (rc) {
 		dev_err(&pdev->dev, "Couldn't register IRQ %d! rc=%d\n",
@@ -7583,12 +7676,13 @@ cleanup_nolog:
 	ipr_free_mem(ioa_cfg);
 cleanup_nomem:
 	iounmap(ipr_regs);
+out_msi_disable:
+	pci_disable_msi(pdev);
 out_release_regions:
 	pci_release_regions(pdev);
 out_scsi_host_put:
 	scsi_host_put(host);
 out_disable:
-	pci_disable_msi(pdev);
 	pci_disable_device(pdev);
 	goto out;
 }
diff --git a/drivers/scsi/ipr.h b/drivers/scsi/ipr.h
index 79a3ae4fb2c7..2d9269b26f88 100644
--- a/drivers/scsi/ipr.h
+++ b/drivers/scsi/ipr.h
@@ -37,8 +37,8 @@
 /*
  * Literals
  */
-#define IPR_DRIVER_VERSION "2.4.2"
-#define IPR_DRIVER_DATE "(January 21, 2009)"
+#define IPR_DRIVER_VERSION "2.4.3"
+#define IPR_DRIVER_DATE "(June 10, 2009)"
 
 /*
  * IPR_MAX_CMD_PER_LUN: This defines the maximum number of outstanding
@@ -1094,6 +1094,7 @@ struct ipr_ioa_cfg {
 	u8 needs_hard_reset:1;
 	u8 dual_raid:1;
 	u8 needs_warm_reset:1;
+	u8 msi_received:1;
 
 	u8 revid;
 
@@ -1179,6 +1180,7 @@ struct ipr_ioa_cfg {
 	struct work_struct work_q;
 
 	wait_queue_head_t reset_wait_q;
+	wait_queue_head_t msi_wait_q;
 
 	struct ipr_dump *dump;
 	enum ipr_sdt_state sdt_state;
-- 
cgit v1.2.3-59-g8ed1b


From 1be7bd82bf4c5d9d3efd1de0e2ebe2c5b1db8340 Mon Sep 17 00:00:00 2001
From: Wayne Boyer <wayneb@linux.vnet.ibm.com>
Date: Wed, 17 Jun 2009 09:55:35 -0700
Subject: ipr: differentiate pci-x and pci-e based adapters

MSI has only been tested on and known to work with PCI-E based adapters.  This
patch adds a field to struct ipr_chip_t to indicate which type of interrupt to
use based on what is known about the chip.

Signed-off-by: Wayne Boyer <wayneb@linux.vnet.ibm.com>
Acked-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/ipr.c | 32 +++++++++++++++++---------------
 drivers/scsi/ipr.h |  4 ++++
 2 files changed, 21 insertions(+), 15 deletions(-)

diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index 15ce8e51d5de..5f045505a1f4 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -131,13 +131,13 @@ static const struct ipr_chip_cfg_t ipr_chip_cfg[] = {
 };
 
 static const struct ipr_chip_t ipr_chip[] = {
-	{ PCI_VENDOR_ID_MYLEX, PCI_DEVICE_ID_IBM_GEMSTONE, &ipr_chip_cfg[0] },
-	{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CITRINE, &ipr_chip_cfg[0] },
-	{ PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_OBSIDIAN, &ipr_chip_cfg[0] },
-	{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN, &ipr_chip_cfg[0] },
-	{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN_E, &ipr_chip_cfg[0] },
-	{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_SNIPE, &ipr_chip_cfg[1] },
-	{ PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_SCAMP, &ipr_chip_cfg[1] }
+	{ PCI_VENDOR_ID_MYLEX, PCI_DEVICE_ID_IBM_GEMSTONE, IPR_USE_LSI, &ipr_chip_cfg[0] },
+	{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_CITRINE, IPR_USE_LSI, &ipr_chip_cfg[0] },
+	{ PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_OBSIDIAN, IPR_USE_LSI, &ipr_chip_cfg[0] },
+	{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN, IPR_USE_LSI, &ipr_chip_cfg[0] },
+	{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_OBSIDIAN_E, IPR_USE_MSI, &ipr_chip_cfg[0] },
+	{ PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_IBM_SNIPE, IPR_USE_LSI, &ipr_chip_cfg[1] },
+	{ PCI_VENDOR_ID_ADAPTEC2, PCI_DEVICE_ID_ADAPTEC2_SCAMP, IPR_USE_LSI, &ipr_chip_cfg[1] }
 };
 
 static int ipr_max_bus_speeds [] = {
@@ -7399,21 +7399,21 @@ static void __devinit ipr_init_ioa_cfg(struct ipr_ioa_cfg *ioa_cfg,
 }
 
 /**
- * ipr_get_chip_cfg - Find adapter chip configuration
+ * ipr_get_chip_info - Find adapter chip information
  * @dev_id:		PCI device id struct
  *
  * Return value:
- * 	ptr to chip config on success / NULL on failure
+ * 	ptr to chip information on success / NULL on failure
  **/
-static const struct ipr_chip_cfg_t * __devinit
-ipr_get_chip_cfg(const struct pci_device_id *dev_id)
+static const struct ipr_chip_t * __devinit
+ipr_get_chip_info(const struct pci_device_id *dev_id)
 {
 	int i;
 
 	for (i = 0; i < ARRAY_SIZE(ipr_chip); i++)
 		if (ipr_chip[i].vendor == dev_id->vendor &&
 		    ipr_chip[i].device == dev_id->device)
-			return ipr_chip[i].cfg;
+			return &ipr_chip[i];
 	return NULL;
 }
 
@@ -7540,14 +7540,16 @@ static int __devinit ipr_probe_ioa(struct pci_dev *pdev,
 	ata_host_init(&ioa_cfg->ata_host, &pdev->dev,
 		      sata_port_info.flags, &ipr_sata_ops);
 
-	ioa_cfg->chip_cfg = ipr_get_chip_cfg(dev_id);
+	ioa_cfg->ipr_chip = ipr_get_chip_info(dev_id);
 
-	if (!ioa_cfg->chip_cfg) {
+	if (!ioa_cfg->ipr_chip) {
 		dev_err(&pdev->dev, "Unknown adapter chipset 0x%04X 0x%04X\n",
 			dev_id->vendor, dev_id->device);
 		goto out_scsi_host_put;
 	}
 
+	ioa_cfg->chip_cfg = ioa_cfg->ipr_chip->cfg;
+
 	if (ipr_transop_timeout)
 		ioa_cfg->transop_timeout = ipr_transop_timeout;
 	else if (dev_id->driver_data & IPR_USE_LONG_TRANSOP_TIMEOUT)
@@ -7599,7 +7601,7 @@ static int __devinit ipr_probe_ioa(struct pci_dev *pdev,
 	}
 
 	/* Enable MSI style interrupts if they are supported. */
-	if (!(rc = pci_enable_msi(pdev))) {
+	if (ioa_cfg->ipr_chip->intr_type == IPR_USE_MSI && !pci_enable_msi(pdev)) {
 		rc = ipr_test_msi(ioa_cfg, pdev);
 		if (rc == -EOPNOTSUPP)
 			pci_disable_msi(pdev);
diff --git a/drivers/scsi/ipr.h b/drivers/scsi/ipr.h
index 2d9269b26f88..4b63dd6b1c81 100644
--- a/drivers/scsi/ipr.h
+++ b/drivers/scsi/ipr.h
@@ -1025,6 +1025,9 @@ struct ipr_chip_cfg_t {
 struct ipr_chip_t {
 	u16 vendor;
 	u16 device;
+	u16 intr_type;
+#define IPR_USE_LSI			0x00
+#define IPR_USE_MSI			0x01
 	const struct ipr_chip_cfg_t *cfg;
 };
 
@@ -1160,6 +1163,7 @@ struct ipr_ioa_cfg {
 
 	unsigned int transop_timeout;
 	const struct ipr_chip_cfg_t *chip_cfg;
+	const struct ipr_chip_t *ipr_chip;
 
 	void __iomem *hdw_dma_regs;	/* iomapped PCI memory space */
 	unsigned long hdw_dma_regs_pci;	/* raw PCI memory space */
-- 
cgit v1.2.3-59-g8ed1b


From 7cbdca23c8a4e6d007b62c9136ba0e5f86e069d0 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Sat, 13 Jun 2009 15:51:08 -0500
Subject: Revert "[SCSI] cnic: fix error: implicit declaration of function
 ‘__symbol_get’"
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This reverts commit bc3bf8fd330ce981ce632a1a4a283eee46838f32.

All the commit did was add a second #include of <linux/module.h> which is
the wrong fix.

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/net/cnic.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/cnic.c b/drivers/net/cnic.c
index 44f77eb1180f..a9e2fd35bb41 100644
--- a/drivers/net/cnic.c
+++ b/drivers/net/cnic.c
@@ -25,8 +25,6 @@
 #include <linux/delay.h>
 #include <linux/ethtool.h>
 #include <linux/if_vlan.h>
-#include <linux/module.h>
-
 #if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
 #define BCM_VLAN 1
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From e2ee3616bc334ab51e68aad6905761ca97f35559 Mon Sep 17 00:00:00 2001
From: Michael Chan <mchan@broadcom.com>
Date: Sat, 13 Jun 2009 17:43:02 -0700
Subject: cnic: Fix __symbol_get() build error.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Ingo molnar <mingo@elte.hu> reported the error

drivers/net/cnic.c:2520: error: implicit declaration of function ‘__symbol_get’

when CONFIG_MODULES is not defined.  Fix by using symbol_get() instead.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/net/cnic.c    | 4 ++--
 drivers/net/cnic_if.h | 2 ++
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/cnic.c b/drivers/net/cnic.c
index a9e2fd35bb41..4d1515f45ba2 100644
--- a/drivers/net/cnic.c
+++ b/drivers/net/cnic.c
@@ -2519,9 +2519,9 @@ static struct cnic_dev *init_bnx2_cnic(struct net_device *dev)
 	struct cnic_dev *cdev;
 	struct cnic_local *cp;
 	struct cnic_eth_dev *ethdev = NULL;
-	struct cnic_eth_dev *(*probe)(void *) = NULL;
+	struct cnic_eth_dev *(*probe)(struct net_device *) = NULL;
 
-	probe = __symbol_get("bnx2_cnic_probe");
+	probe = symbol_get(bnx2_cnic_probe);
 	if (probe) {
 		ethdev = (*probe)(dev);
 		symbol_put_addr(probe);
diff --git a/drivers/net/cnic_if.h b/drivers/net/cnic_if.h
index 06380963a34e..d1bce27ee99e 100644
--- a/drivers/net/cnic_if.h
+++ b/drivers/net/cnic_if.h
@@ -296,4 +296,6 @@ extern int cnic_register_driver(int ulp_type, struct cnic_ulp_ops *ulp_ops);
 
 extern int cnic_unregister_driver(int ulp_type);
 
+extern struct cnic_eth_dev *bnx2_cnic_probe(struct net_device *dev);
+
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 895553824ed9d2c1409d4dc6e0db4d6dfd344679 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Thu, 18 Jun 2009 09:55:17 -0700
Subject: cnic: add NETDEV_1000 and NETDEVICES to Kconfig select

NETDEVICES + NETDEV_1000 need to be enabled so that kconfig will check
those branches for selects and enforce "select UIO" under CNIC.
Otherwise the build fails with:

ERROR: "uio_unregister_device" [drivers/net/cnic.ko] undefined!
ERROR: "uio_event_notify" [drivers/net/cnic.ko] undefined!
ERROR: "__uio_register_device" [drivers/net/cnic.ko] undefined!

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Acked-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/bnx2i/Kconfig | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/scsi/bnx2i/Kconfig b/drivers/scsi/bnx2i/Kconfig
index b62b482e55e7..1e9f7141102b 100644
--- a/drivers/scsi/bnx2i/Kconfig
+++ b/drivers/scsi/bnx2i/Kconfig
@@ -1,6 +1,8 @@
 config SCSI_BNX2_ISCSI
 	tristate "Broadcom NetXtreme II iSCSI support"
 	select SCSI_ISCSI_ATTRS
+	select NETDEVICES
+	select NETDEV_1000
 	select CNIC
 	depends on PCI
 	---help---
-- 
cgit v1.2.3-59-g8ed1b


From ea61fca58c1373a48c0741798f70364d4498d2af Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 15 May 2009 00:40:33 -0400
Subject: scsi_debug: Add support for physical block exponent and alignment

This patch adds support for setting the physical block exponent and
lowest aligned LBA in the READ CAPACITY(16) response.

The B0 VPD page is adjusted accordingly.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Acked-by: Douglas Gilbert <dgilbert@interlog.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_debug.c | 30 +++++++++++++++++++++++++++++-
 1 file changed, 29 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c
index 41a21772df12..fb9af207d61d 100644
--- a/drivers/scsi/scsi_debug.c
+++ b/drivers/scsi/scsi_debug.c
@@ -101,6 +101,8 @@ static const char * scsi_debug_version_date = "20070104";
 #define DEF_DIF 0
 #define DEF_GUARD 0
 #define DEF_ATO 1
+#define DEF_PHYSBLK_EXP 0
+#define DEF_LOWEST_ALIGNED 0
 
 /* bit mask values for scsi_debug_opts */
 #define SCSI_DEBUG_OPT_NOISE   1
@@ -156,6 +158,8 @@ static int scsi_debug_dix = DEF_DIX;
 static int scsi_debug_dif = DEF_DIF;
 static int scsi_debug_guard = DEF_GUARD;
 static int scsi_debug_ato = DEF_ATO;
+static int scsi_debug_physblk_exp = DEF_PHYSBLK_EXP;
+static int scsi_debug_lowest_aligned = DEF_LOWEST_ALIGNED;
 
 static int scsi_debug_cmnd_count = 0;
 
@@ -657,7 +661,12 @@ static unsigned char vpdb0_data[] = {
 
 static int inquiry_evpd_b0(unsigned char * arr)
 {
+	unsigned int gran;
+
 	memcpy(arr, vpdb0_data, sizeof(vpdb0_data));
+	gran = 1 << scsi_debug_physblk_exp;
+	arr[2] = (gran >> 8) & 0xff;
+	arr[3] = gran & 0xff;
 	if (sdebug_store_sectors > 0x400) {
 		arr[4] = (sdebug_store_sectors >> 24) & 0xff;
 		arr[5] = (sdebug_store_sectors >> 16) & 0xff;
@@ -945,6 +954,9 @@ static int resp_readcap16(struct scsi_cmnd * scp,
 	arr[9] = (scsi_debug_sector_size >> 16) & 0xff;
 	arr[10] = (scsi_debug_sector_size >> 8) & 0xff;
 	arr[11] = scsi_debug_sector_size & 0xff;
+	arr[13] = scsi_debug_physblk_exp & 0xf;
+	arr[14] = (scsi_debug_lowest_aligned >> 8) & 0x3f;
+	arr[15] = scsi_debug_lowest_aligned & 0xff;
 
 	if (scsi_debug_dif) {
 		arr[12] = (scsi_debug_dif - 1) << 1; /* P_TYPE */
@@ -2380,6 +2392,8 @@ module_param_named(dix, scsi_debug_dix, int, S_IRUGO);
 module_param_named(dif, scsi_debug_dif, int, S_IRUGO);
 module_param_named(guard, scsi_debug_guard, int, S_IRUGO);
 module_param_named(ato, scsi_debug_ato, int, S_IRUGO);
+module_param_named(physblk_exp, scsi_debug_physblk_exp, int, S_IRUGO);
+module_param_named(lowest_aligned, scsi_debug_lowest_aligned, int, S_IRUGO);
 
 MODULE_AUTHOR("Eric Youngdale + Douglas Gilbert");
 MODULE_DESCRIPTION("SCSI debug adapter driver");
@@ -2401,7 +2415,9 @@ MODULE_PARM_DESC(ptype, "SCSI peripheral type(def=0[disk])");
 MODULE_PARM_DESC(scsi_level, "SCSI level to simulate(def=5[SPC-3])");
 MODULE_PARM_DESC(virtual_gb, "virtual gigabyte size (def=0 -> use dev_size_mb)");
 MODULE_PARM_DESC(vpd_use_hostno, "0 -> dev ids ignore hostno (def=1 -> unique dev ids)");
-MODULE_PARM_DESC(sector_size, "hardware sector size in bytes (def=512)");
+MODULE_PARM_DESC(sector_size, "logical block size in bytes (def=512)");
+MODULE_PARM_DESC(physblk_exp, "physical block exponent (def=0)");
+MODULE_PARM_DESC(lowest_aligned, "lowest aligned lba (def=0)");
 MODULE_PARM_DESC(dix, "data integrity extensions mask (def=0)");
 MODULE_PARM_DESC(dif, "data integrity field type: 0-3 (def=0)");
 MODULE_PARM_DESC(guard, "protection checksum: 0=crc, 1=ip (def=0)");
@@ -2874,6 +2890,18 @@ static int __init scsi_debug_init(void)
 		return -EINVAL;
 	}
 
+	if (scsi_debug_physblk_exp > 15) {
+		printk(KERN_ERR "scsi_debug_init: invalid physblk_exp %u\n",
+		       scsi_debug_physblk_exp);
+		return -EINVAL;
+	}
+
+	if (scsi_debug_lowest_aligned > 0x3fff) {
+		printk(KERN_ERR "scsi_debug_init: lowest_aligned too big: %u\n",
+		       scsi_debug_lowest_aligned);
+		return -EINVAL;
+	}
+
 	if (scsi_debug_dev_size_mb < 1)
 		scsi_debug_dev_size_mb = 1;  /* force minimum 1 MB ramdisk */
 	sz = (unsigned long)scsi_debug_dev_size_mb * 1048576;
-- 
cgit v1.2.3-59-g8ed1b


From d5488eb9cd2b06f7dcca7053274edb337987c67c Mon Sep 17 00:00:00 2001
From: Robert Love <robert.w.love@intel.com>
Date: Wed, 10 Jun 2009 15:30:59 -0700
Subject: fcoe: Add runtime debug logging with module parameter debug_logging

This patch converts all FC_DBG statements to use new runtime tunable
debug macros. The fcoe.ko module now has a debug_logging module
parameter.

fcoe_debug_logging is an unsigned integer representing a bitmask of all
available logging levels. Currently only two logging levels are
supported-

   bit
LSB 0 = general fcoe logging
    1 = netdevice related logging

This patch also attempts to clean up some debug statement formatting
so it's more readable.

Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/fcoe/fcoe.c | 108 +++++++++++++++++++++++------------------------
 drivers/scsi/fcoe/fcoe.h |  24 +++++++++++
 2 files changed, 77 insertions(+), 55 deletions(-)

diff --git a/drivers/scsi/fcoe/fcoe.c b/drivers/scsi/fcoe/fcoe.c
index c15878e88157..0a5609bb5817 100644
--- a/drivers/scsi/fcoe/fcoe.c
+++ b/drivers/scsi/fcoe/fcoe.c
@@ -45,8 +45,6 @@
 
 #include "fcoe.h"
 
-static int debug_fcoe;
-
 MODULE_AUTHOR("Open-FCoE.org");
 MODULE_DESCRIPTION("FCoE");
 MODULE_LICENSE("GPL v2");
@@ -305,23 +303,22 @@ static int fcoe_netdev_config(struct fc_lport *lp, struct net_device *netdev)
 #ifdef NETIF_F_FCOE_CRC
 	if (netdev->features & NETIF_F_FCOE_CRC) {
 		lp->crc_offload = 1;
-		printk(KERN_DEBUG "fcoe:%s supports FCCRC offload\n",
-		       netdev->name);
+		FCOE_NETDEV_DBG(netdev, "Supports FCCRC offload\n");
 	}
 #endif
 #ifdef NETIF_F_FSO
 	if (netdev->features & NETIF_F_FSO) {
 		lp->seq_offload = 1;
 		lp->lso_max = netdev->gso_max_size;
-		printk(KERN_DEBUG "fcoe:%s supports LSO for max len 0x%x\n",
-		       netdev->name, lp->lso_max);
+		FCOE_NETDEV_DBG(netdev, "Supports LSO for max len 0x%x\n",
+				lp->lso_max);
 	}
 #endif
 	if (netdev->fcoe_ddp_xid) {
 		lp->lro_enabled = 1;
 		lp->lro_xid = netdev->fcoe_ddp_xid;
-		printk(KERN_DEBUG "fcoe:%s supports LRO for max xid 0x%x\n",
-		       netdev->name, lp->lro_xid);
+		FCOE_NETDEV_DBG(netdev, "Supports LRO for max xid 0x%x\n",
+				lp->lro_xid);
 	}
 	skb_queue_head_init(&fc->fcoe_pending_queue);
 	fc->fcoe_pending_queue_active = 0;
@@ -407,7 +404,8 @@ static int fcoe_shost_config(struct fc_lport *lp, struct Scsi_Host *shost,
 	/* add the new host to the SCSI-ml */
 	rc = scsi_add_host(lp->host, dev);
 	if (rc) {
-		FC_DBG("fcoe_shost_config:error on scsi_add_host\n");
+		FCOE_NETDEV_DBG(fcoe_netdev(lp), "fcoe_shost_config: "
+				"error on scsi_add_host\n");
 		return rc;
 	}
 	sprintf(fc_host_symbolic_name(lp->host), "%s v%s over %s",
@@ -448,8 +446,7 @@ static int fcoe_if_destroy(struct net_device *netdev)
 
 	BUG_ON(!netdev);
 
-	printk(KERN_DEBUG "fcoe_if_destroy:interface on %s\n",
-	       netdev->name);
+	FCOE_NETDEV_DBG(netdev, "Destroying interface\n");
 
 	lp = fcoe_hostlist_lookup(netdev);
 	if (!lp)
@@ -560,8 +557,7 @@ static int fcoe_if_create(struct net_device *netdev)
 
 	BUG_ON(!netdev);
 
-	printk(KERN_DEBUG "fcoe_if_create:interface on %s\n",
-	       netdev->name);
+	FCOE_NETDEV_DBG(netdev, "Create Interface\n");
 
 	lp = fcoe_hostlist_lookup(netdev);
 	if (lp)
@@ -570,7 +566,7 @@ static int fcoe_if_create(struct net_device *netdev)
 	shost = libfc_host_alloc(&fcoe_shost_template,
 				 sizeof(struct fcoe_softc));
 	if (!shost) {
-		FC_DBG("Could not allocate host structure\n");
+		FCOE_NETDEV_DBG(netdev, "Could not allocate host structure\n");
 		return -ENOMEM;
 	}
 	lp = shost_priv(shost);
@@ -579,7 +575,8 @@ static int fcoe_if_create(struct net_device *netdev)
 	/* configure fc_lport, e.g., em */
 	rc = fcoe_lport_config(lp);
 	if (rc) {
-		FC_DBG("Could not configure lport\n");
+		FCOE_NETDEV_DBG(netdev, "Could not configure lport for the "
+				"interface\n");
 		goto out_host_put;
 	}
 
@@ -593,28 +590,32 @@ static int fcoe_if_create(struct net_device *netdev)
 	/* configure lport network properties */
 	rc = fcoe_netdev_config(lp, netdev);
 	if (rc) {
-		FC_DBG("Could not configure netdev for the interface\n");
+		FCOE_NETDEV_DBG(netdev, "Could not configure netdev for the "
+				"interface\n");
 		goto out_netdev_cleanup;
 	}
 
 	/* configure lport scsi host properties */
 	rc = fcoe_shost_config(lp, shost, &netdev->dev);
 	if (rc) {
-		FC_DBG("Could not configure shost for lport\n");
+		FCOE_NETDEV_DBG(netdev, "Could not configure shost for the "
+				"interface\n");
 		goto out_netdev_cleanup;
 	}
 
 	/* lport exch manager allocation */
 	rc = fcoe_em_config(lp);
 	if (rc) {
-		FC_DBG("Could not configure em for lport\n");
+		FCOE_NETDEV_DBG(netdev, "Could not configure the EM for the "
+				"interface\n");
 		goto out_netdev_cleanup;
 	}
 
 	/* Initialize the library */
 	rc = fcoe_libfc_config(lp, &fcoe_libfc_fcn_templ);
 	if (rc) {
-		FC_DBG("Could not configure libfc for lport!\n");
+		FCOE_NETDEV_DBG(netdev, "Could not configure libfc for the "
+				"interface\n");
 		goto out_lp_destroy;
 	}
 
@@ -653,7 +654,7 @@ static int __init fcoe_if_init(void)
 		fc_attach_transport(&fcoe_transport_function);
 
 	if (!scsi_transport_fcoe_sw) {
-		printk(KERN_ERR "fcoe_init:fc_attach_transport() failed\n");
+		printk(KERN_ERR "fcoe: Failed to attach to the FC transport\n");
 		return -ENODEV;
 	}
 
@@ -714,7 +715,7 @@ static void fcoe_percpu_thread_destroy(unsigned int cpu)
 	unsigned targ_cpu = smp_processor_id();
 #endif /* CONFIG_SMP */
 
-	printk(KERN_DEBUG "fcoe: Destroying receive thread for CPU %d\n", cpu);
+	FCOE_DBG("Destroying receive thread for CPU %d\n", cpu);
 
 	/* Prevent any new skbs from being queued for this CPU. */
 	p = &per_cpu(fcoe_percpu, cpu);
@@ -736,8 +737,8 @@ static void fcoe_percpu_thread_destroy(unsigned int cpu)
 		p0 = &per_cpu(fcoe_percpu, targ_cpu);
 		spin_lock_bh(&p0->fcoe_rx_list.lock);
 		if (p0->thread) {
-			FC_DBG("Moving frames from CPU %d to CPU %d\n",
-			       cpu, targ_cpu);
+			FCOE_DBG("Moving frames from CPU %d to CPU %d\n",
+				 cpu, targ_cpu);
 
 			while ((skb = __skb_dequeue(&p->fcoe_rx_list)) != NULL)
 				__skb_queue_tail(&p0->fcoe_rx_list, skb);
@@ -803,12 +804,12 @@ static int fcoe_cpu_callback(struct notifier_block *nfb,
 	switch (action) {
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
-		FC_DBG("CPU %x online: Create Rx thread\n", cpu);
+		FCOE_DBG("CPU %x online: Create Rx thread\n", cpu);
 		fcoe_percpu_thread_create(cpu);
 		break;
 	case CPU_DEAD:
 	case CPU_DEAD_FROZEN:
-		FC_DBG("CPU %x offline: Remove Rx thread\n", cpu);
+		FCOE_DBG("CPU %x offline: Remove Rx thread\n", cpu);
 		fcoe_percpu_thread_destroy(cpu);
 		break;
 	default:
@@ -846,24 +847,21 @@ int fcoe_rcv(struct sk_buff *skb, struct net_device *dev,
 	fc = container_of(ptype, struct fcoe_softc, fcoe_packet_type);
 	lp = fc->ctlr.lp;
 	if (unlikely(lp == NULL)) {
-		FC_DBG("cannot find hba structure");
+		FCOE_NETDEV_DBG(dev, "Cannot find hba structure");
 		goto err2;
 	}
 	if (!lp->link_up)
 		goto err2;
 
-	if (unlikely(debug_fcoe)) {
-		FC_DBG("skb_info: len:%d data_len:%d head:%p data:%p tail:%p "
-		       "end:%p sum:%d dev:%s", skb->len, skb->data_len,
-		       skb->head, skb->data, skb_tail_pointer(skb),
-		       skb_end_pointer(skb), skb->csum,
-		       skb->dev ? skb->dev->name : "<NULL>");
-
-	}
+	FCOE_NETDEV_DBG(dev, "skb_info: len:%d data_len:%d head:%p "
+			"data:%p tail:%p end:%p sum:%d dev:%s",
+			skb->len, skb->data_len, skb->head, skb->data,
+			skb_tail_pointer(skb), skb_end_pointer(skb),
+			skb->csum, skb->dev ? skb->dev->name : "<NULL>");
 
 	/* check for FCOE packet type */
 	if (unlikely(eth_hdr(skb)->h_proto != htons(ETH_P_FCOE))) {
-		FC_DBG("wrong FC type frame");
+		FCOE_NETDEV_DBG(dev, "Wrong FC type frame");
 		goto err;
 	}
 
@@ -901,8 +899,9 @@ int fcoe_rcv(struct sk_buff *skb, struct net_device *dev,
 		 * the first CPU now. For non-SMP systems this
 		 * will check the same CPU twice.
 		 */
-		FC_DBG("CPU is online, but no receive thread ready "
-		       "for incoming skb- using first online CPU.\n");
+		FCOE_NETDEV_DBG(dev, "CPU is online, but no receive thread "
+				"ready for incoming skb- using first online "
+				"CPU.\n");
 
 		spin_unlock_bh(&fps->fcoe_rx_list.lock);
 		cpu = first_cpu(cpu_online_map);
@@ -1201,19 +1200,17 @@ int fcoe_percpu_receive_thread(void *arg)
 		fr = fcoe_dev_from_skb(skb);
 		lp = fr->fr_dev;
 		if (unlikely(lp == NULL)) {
-			FC_DBG("invalid HBA Structure");
+			FCOE_NETDEV_DBG(skb->dev, "Invalid HBA Structure");
 			kfree_skb(skb);
 			continue;
 		}
 
-		if (unlikely(debug_fcoe)) {
-			FC_DBG("skb_info: len:%d data_len:%d head:%p data:%p "
-			       "tail:%p end:%p sum:%d dev:%s",
-			       skb->len, skb->data_len,
-			       skb->head, skb->data, skb_tail_pointer(skb),
-			       skb_end_pointer(skb), skb->csum,
-			       skb->dev ? skb->dev->name : "<NULL>");
-		}
+		FCOE_NETDEV_DBG(skb->dev, "skb_info: len:%d data_len:%d "
+				"head:%p data:%p tail:%p end:%p sum:%d dev:%s",
+				skb->len, skb->data_len,
+				skb->head, skb->data, skb_tail_pointer(skb),
+				skb_end_pointer(skb), skb->csum,
+				skb->dev ? skb->dev->name : "<NULL>");
 
 		/*
 		 * Save source MAC address before discarding header.
@@ -1233,7 +1230,7 @@ int fcoe_percpu_receive_thread(void *arg)
 		stats = fc_lport_get_stats(lp);
 		if (unlikely(FC_FCOE_DECAPS_VER(hp) != FC_FCOE_VER)) {
 			if (stats->ErrorFrames < 5)
-				printk(KERN_WARNING "FCoE version "
+				printk(KERN_WARNING "fcoe: FCoE version "
 				       "mismatch: The frame has "
 				       "version %x, but the "
 				       "initiator supports version "
@@ -1286,7 +1283,7 @@ int fcoe_percpu_receive_thread(void *arg)
 		if (fr_flags(fp) & FCPHF_CRC_UNCHECKED) {
 			if (le32_to_cpu(fr_crc(fp)) !=
 			    ~crc32(~0, skb->data, fr_len)) {
-				if (debug_fcoe || stats->InvalidCRCCount < 5)
+				if (stats->InvalidCRCCount < 5)
 					printk(KERN_WARNING "fcoe: dropping "
 					       "frame with CRC error\n");
 				stats->InvalidCRCCount++;
@@ -1432,7 +1429,8 @@ static int fcoe_device_notification(struct notifier_block *notifier,
 	case NETDEV_REGISTER:
 		break;
 	default:
-		FC_DBG("Unknown event %ld from netdev netlink\n", event);
+		FCOE_NETDEV_DBG(real_dev, "Unknown event %ld "
+				"from netdev netlink\n", event);
 	}
 	if (link_possible && !fcoe_link_ok(lp))
 		fcoe_ctlr_link_up(&fc->ctlr);
@@ -1505,8 +1503,8 @@ static int fcoe_ethdrv_get(const struct net_device *netdev)
 
 	owner = fcoe_netdev_to_module_owner(netdev);
 	if (owner) {
-		printk(KERN_DEBUG "fcoe:hold driver module %s for %s\n",
-		       module_name(owner), netdev->name);
+		FCOE_NETDEV_DBG(netdev, "Hold driver module %s\n",
+				module_name(owner));
 		return  try_module_get(owner);
 	}
 	return -ENODEV;
@@ -1527,8 +1525,8 @@ static int fcoe_ethdrv_put(const struct net_device *netdev)
 
 	owner = fcoe_netdev_to_module_owner(netdev);
 	if (owner) {
-		printk(KERN_DEBUG "fcoe:release driver module %s for %s\n",
-		       module_name(owner), netdev->name);
+		FCOE_NETDEV_DBG(netdev, "Release driver module %s\n",
+				module_name(owner));
 		module_put(owner);
 		return 0;
 	}
@@ -1559,7 +1557,7 @@ static int fcoe_destroy(const char *buffer, struct kernel_param *kp)
 	}
 	rc = fcoe_if_destroy(netdev);
 	if (rc) {
-		printk(KERN_ERR "fcoe: fcoe_if_destroy(%s) failed\n",
+		printk(KERN_ERR "fcoe: Failed to destroy interface (%s)\n",
 		       netdev->name);
 		rc = -EIO;
 		goto out_putdev;
@@ -1598,7 +1596,7 @@ static int fcoe_create(const char *buffer, struct kernel_param *kp)
 
 	rc = fcoe_if_create(netdev);
 	if (rc) {
-		printk(KERN_ERR "fcoe: fcoe_if_create(%s) failed\n",
+		printk(KERN_ERR "fcoe: Failed to create interface (%s)\n",
 		       netdev->name);
 		fcoe_ethdrv_put(netdev);
 		rc = -EIO;
diff --git a/drivers/scsi/fcoe/fcoe.h b/drivers/scsi/fcoe/fcoe.h
index a1eb8c1988b0..0d724fa0898f 100644
--- a/drivers/scsi/fcoe/fcoe.h
+++ b/drivers/scsi/fcoe/fcoe.h
@@ -40,6 +40,30 @@
 #define FCOE_MIN_XID		0x0001	/* the min xid supported by fcoe_sw */
 #define FCOE_MAX_XID		0x07ef	/* the max xid supported by fcoe_sw */
 
+unsigned int fcoe_debug_logging;
+module_param_named(debug_logging, fcoe_debug_logging, int, S_IRUGO|S_IWUSR);
+MODULE_PARM_DESC(debug_logging, "a bit mask of logging levels");
+
+#define FCOE_LOGGING        0x01 /* General logging, not categorized */
+#define FCOE_NETDEV_LOGGING 0x02 /* Netdevice logging */
+
+#define FCOE_CHECK_LOGGING(LEVEL, CMD)					\
+do {                                                            	\
+	if (unlikely(fcoe_debug_logging & LEVEL))			\
+		do {							\
+			CMD;						\
+		} while (0);						\
+} while (0);
+
+#define FCOE_DBG(fmt, args...)						\
+	FCOE_CHECK_LOGGING(FCOE_LOGGING,				\
+			   printk(KERN_INFO "fcoe: " fmt, ##args);)
+
+#define FCOE_NETDEV_DBG(netdev, fmt, args...)			\
+	FCOE_CHECK_LOGGING(FCOE_NETDEV_LOGGING,			\
+			   printk(KERN_INFO "fcoe: %s" fmt,	\
+				  netdev->name, ##args);)
+
 /*
  * this percpu struct for fcoe
  */
-- 
cgit v1.2.3-59-g8ed1b


From 650bd12b9e31ec51d7ad0df3c4f94d863b827976 Mon Sep 17 00:00:00 2001
From: Robert Love <robert.w.love@intel.com>
Date: Wed, 10 Jun 2009 15:31:05 -0700
Subject: libfcoe: Add runtime debugging with module param debug_logging

This patch adds a 'debug_logging' module parameter to
libfcoe.ko. It is an unsigned int that represents a bitmask of
available debug logging levels, each of which can be tuned at
runtime. Currently there are only two logging levels for this
module-

   bit
LSB 0 = libfcoe general logging
    1 = FIP logging

Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/fcoe/libfcoe.c | 94 ++++++++++++++++++++++++++-------------------
 1 file changed, 54 insertions(+), 40 deletions(-)

diff --git a/drivers/scsi/fcoe/libfcoe.c b/drivers/scsi/fcoe/libfcoe.c
index 2f5bc7fd3fa9..f544340d318b 100644
--- a/drivers/scsi/fcoe/libfcoe.c
+++ b/drivers/scsi/fcoe/libfcoe.c
@@ -56,15 +56,28 @@ static void fcoe_ctlr_recv_work(struct work_struct *);
 
 static u8 fcoe_all_fcfs[ETH_ALEN] = FIP_ALL_FCF_MACS;
 
-static u32 fcoe_ctlr_debug;	/* 1 for basic, 2 for noisy debug */
+unsigned int libfcoe_debug_logging;
+module_param_named(debug_logging, libfcoe_debug_logging, int, S_IRUGO|S_IWUSR);
+MODULE_PARM_DESC(debug_logging, "a bit mask of logging levels");
 
-#define FIP_DBG_LVL(level, fmt, args...) 				\
+#define LIBFCOE_LOGGING     0x01 /* General logging, not categorized */
+#define LIBFCOE_FIP_LOGGING 0x02 /* FIP logging */
+
+#define LIBFCOE_CHECK_LOGGING(LEVEL, CMD)				\
+do {                                                            	\
+	if (unlikely(libfcoe_debug_logging & LEVEL))			\
 		do {							\
-			if (fcoe_ctlr_debug >= (level))			\
-				FC_DBG(fmt, ##args);			\
-		} while (0)
+			CMD;						\
+		} while (0);						\
+} while (0);
+
+#define LIBFCOE_DBG(fmt, args...)					\
+	LIBFCOE_CHECK_LOGGING(LIBFCOE_LOGGING,				\
+			      printk(KERN_INFO "libfcoe: " fmt, ##args);)
 
-#define FIP_DBG(fmt, args...)	FIP_DBG_LVL(1, fmt, ##args)
+#define LIBFCOE_FIP_DBG(fmt, args...)					\
+	LIBFCOE_CHECK_LOGGING(LIBFCOE_FIP_LOGGING,			\
+			      printk(KERN_INFO "fip: " fmt, ##args);)
 
 /*
  * Return non-zero if FCF fcoe_size has been validated.
@@ -243,7 +256,7 @@ void fcoe_ctlr_link_up(struct fcoe_ctlr *fip)
 		fip->last_link = 1;
 		fip->link = 1;
 		spin_unlock_bh(&fip->lock);
-		FIP_DBG("%s", "setting AUTO mode.\n");
+		LIBFCOE_FIP_DBG("%s", "setting AUTO mode.\n");
 		fc_linkup(fip->lp);
 		fcoe_ctlr_solicit(fip, NULL);
 	} else
@@ -614,7 +627,8 @@ static int fcoe_ctlr_parse_adv(struct sk_buff *skb, struct fcoe_fcf *fcf)
 			       ((struct fip_mac_desc *)desc)->fd_mac,
 			       ETH_ALEN);
 			if (!is_valid_ether_addr(fcf->fcf_mac)) {
-				FIP_DBG("invalid MAC addr in FIP adv\n");
+				LIBFCOE_FIP_DBG("Invalid MAC address "
+						"in FIP adv\n");
 				return -EINVAL;
 			}
 			break;
@@ -647,8 +661,8 @@ static int fcoe_ctlr_parse_adv(struct sk_buff *skb, struct fcoe_fcf *fcf)
 		case FIP_DT_LOGO:
 		case FIP_DT_ELP:
 		default:
-			FIP_DBG("unexpected descriptor type %x in FIP adv\n",
-				desc->fip_dtype);
+			LIBFCOE_FIP_DBG("unexpected descriptor type %x "
+					"in FIP adv\n", desc->fip_dtype);
 			/* standard says ignore unknown descriptors >= 128 */
 			if (desc->fip_dtype < FIP_DT_VENDOR_BASE)
 				return -EINVAL;
@@ -664,8 +678,8 @@ static int fcoe_ctlr_parse_adv(struct sk_buff *skb, struct fcoe_fcf *fcf)
 	return 0;
 
 len_err:
-	FIP_DBG("FIP length error in descriptor type %x len %zu\n",
-		desc->fip_dtype, dlen);
+	LIBFCOE_FIP_DBG("FIP length error in descriptor type %x len %zu\n",
+			desc->fip_dtype, dlen);
 	return -EINVAL;
 }
 
@@ -728,9 +742,10 @@ static void fcoe_ctlr_recv_adv(struct fcoe_ctlr *fip, struct sk_buff *skb)
 	}
 	mtu_valid = fcoe_ctlr_mtu_valid(fcf);
 	fcf->time = jiffies;
-	FIP_DBG_LVL(found ? 2 : 1, "%s FCF for fab %llx map %x val %d\n",
-		    found ? "old" : "new",
-		    fcf->fabric_name, fcf->fc_map, mtu_valid);
+	if (!found) {
+		LIBFCOE_FIP_DBG("New FCF for fab %llx map %x val %d\n",
+				fcf->fabric_name, fcf->fc_map, mtu_valid);
+	}
 
 	/*
 	 * If this advertisement is not solicited and our max receive size
@@ -807,7 +822,8 @@ static void fcoe_ctlr_recv_els(struct fcoe_ctlr *fip, struct sk_buff *skb)
 			       ((struct fip_mac_desc *)desc)->fd_mac,
 			       ETH_ALEN);
 			if (!is_valid_ether_addr(granted_mac)) {
-				FIP_DBG("invalid MAC addrs in FIP ELS\n");
+				LIBFCOE_FIP_DBG("Invalid MAC address "
+						"in FIP ELS\n");
 				goto drop;
 			}
 			break;
@@ -825,8 +841,8 @@ static void fcoe_ctlr_recv_els(struct fcoe_ctlr *fip, struct sk_buff *skb)
 			els_dtype = desc->fip_dtype;
 			break;
 		default:
-			FIP_DBG("unexpected descriptor type %x "
-				"in FIP adv\n", desc->fip_dtype);
+			LIBFCOE_FIP_DBG("unexpected descriptor type %x "
+					"in FIP adv\n", desc->fip_dtype);
 			/* standard says ignore unknown descriptors >= 128 */
 			if (desc->fip_dtype < FIP_DT_VENDOR_BASE)
 				goto drop;
@@ -867,8 +883,8 @@ static void fcoe_ctlr_recv_els(struct fcoe_ctlr *fip, struct sk_buff *skb)
 	return;
 
 len_err:
-	FIP_DBG("FIP length error in descriptor type %x len %zu\n",
-		desc->fip_dtype, dlen);
+	LIBFCOE_FIP_DBG("FIP length error in descriptor type %x len %zu\n",
+			desc->fip_dtype, dlen);
 drop:
 	kfree_skb(skb);
 }
@@ -894,7 +910,7 @@ static void fcoe_ctlr_recv_clr_vlink(struct fcoe_ctlr *fip,
 	struct fc_lport *lp = fip->lp;
 	u32	desc_mask;
 
-	FIP_DBG("Clear Virtual Link received\n");
+	LIBFCOE_FIP_DBG("Clear Virtual Link received\n");
 	if (!fcf)
 		return;
 	if (!fcf || !fc_host_port_id(lp->host))
@@ -952,9 +968,9 @@ static void fcoe_ctlr_recv_clr_vlink(struct fcoe_ctlr *fip,
 	 * reset only if all required descriptors were present and valid.
 	 */
 	if (desc_mask) {
-		FIP_DBG("missing descriptors mask %x\n", desc_mask);
+		LIBFCOE_FIP_DBG("missing descriptors mask %x\n", desc_mask);
 	} else {
-		FIP_DBG("performing Clear Virtual Link\n");
+		LIBFCOE_FIP_DBG("performing Clear Virtual Link\n");
 		fcoe_ctlr_reset(fip, FIP_ST_ENABLED);
 	}
 }
@@ -1002,10 +1018,6 @@ static int fcoe_ctlr_recv_handler(struct fcoe_ctlr *fip, struct sk_buff *skb)
 	op = ntohs(fiph->fip_op);
 	sub = fiph->fip_subcode;
 
-	FIP_DBG_LVL(2, "ver %x op %x/%x dl %x fl %x\n",
-		    FIP_VER_DECAPS(fiph->fip_ver), op, sub,
-		    ntohs(fiph->fip_dl_len), ntohs(fiph->fip_flags));
-
 	if (FIP_VER_DECAPS(fiph->fip_ver) != FIP_VER)
 		goto drop;
 	if (ntohs(fiph->fip_dl_len) * FIP_BPW + sizeof(*fiph) > skb->len)
@@ -1017,7 +1029,7 @@ static int fcoe_ctlr_recv_handler(struct fcoe_ctlr *fip, struct sk_buff *skb)
 		fip->map_dest = 0;
 		fip->state = FIP_ST_ENABLED;
 		state = FIP_ST_ENABLED;
-		FIP_DBG("using FIP mode\n");
+		LIBFCOE_FIP_DBG("Using FIP mode\n");
 	}
 	spin_unlock_bh(&fip->lock);
 	if (state != FIP_ST_ENABLED)
@@ -1052,14 +1064,15 @@ static void fcoe_ctlr_select(struct fcoe_ctlr *fip)
 	struct fcoe_fcf *best = NULL;
 
 	list_for_each_entry(fcf, &fip->fcfs, list) {
-		FIP_DBG("consider FCF for fab %llx VFID %d map %x val %d\n",
-			fcf->fabric_name, fcf->vfid,
-			fcf->fc_map, fcoe_ctlr_mtu_valid(fcf));
+		LIBFCOE_FIP_DBG("consider FCF for fab %llx VFID %d map %x "
+				"val %d\n", fcf->fabric_name, fcf->vfid,
+				fcf->fc_map, fcoe_ctlr_mtu_valid(fcf));
 		if (!fcoe_ctlr_fcf_usable(fcf)) {
-			FIP_DBG("FCF for fab %llx map %x %svalid %savailable\n",
-				fcf->fabric_name, fcf->fc_map,
-				(fcf->flags & FIP_FL_SOL) ? "" : "in",
-				(fcf->flags & FIP_FL_AVAIL) ? "" : "un");
+			LIBFCOE_FIP_DBG("FCF for fab %llx map %x %svalid "
+					"%savailable\n", fcf->fabric_name,
+					fcf->fc_map, (fcf->flags & FIP_FL_SOL)
+					? "" : "in", (fcf->flags & FIP_FL_AVAIL)
+					? "" : "un");
 			continue;
 		}
 		if (!best) {
@@ -1069,7 +1082,8 @@ static void fcoe_ctlr_select(struct fcoe_ctlr *fip)
 		if (fcf->fabric_name != best->fabric_name ||
 		    fcf->vfid != best->vfid ||
 		    fcf->fc_map != best->fc_map) {
-			FIP_DBG("conflicting fabric, VFID, or FC-MAP\n");
+			LIBFCOE_FIP_DBG("Conflicting fabric, VFID, "
+					"or FC-MAP\n");
 			return;
 		}
 		if (fcf->pri < best->pri)
@@ -1113,7 +1127,7 @@ static void fcoe_ctlr_timeout(unsigned long arg)
 	if (sel != fcf) {
 		fcf = sel;		/* the old FCF may have been freed */
 		if (sel) {
-			printk(KERN_INFO "host%d: FIP selected "
+			printk(KERN_INFO "libfcoe: host%d: FIP selected "
 			       "Fibre-Channel Forwarder MAC %s\n",
 			       fip->lp->host->host_no,
 			       print_mac(buf, sel->fcf_mac));
@@ -1123,7 +1137,7 @@ static void fcoe_ctlr_timeout(unsigned long arg)
 			fip->ctlr_ka_time = jiffies + sel->fka_period;
 			fip->link = 1;
 		} else {
-			printk(KERN_NOTICE "host%d: "
+			printk(KERN_NOTICE "libfcoe: host%d: "
 			       "FIP Fibre-Channel Forwarder timed out.  "
 			       "Starting FCF discovery.\n",
 			       fip->lp->host->host_no);
@@ -1247,7 +1261,7 @@ int fcoe_ctlr_recv_flogi(struct fcoe_ctlr *fip, struct fc_frame *fp, u8 *sa)
 			return -EINVAL;
 		}
 		fip->state = FIP_ST_NON_FIP;
-		FIP_DBG("received FLOGI LS_ACC using non-FIP mode\n");
+		LIBFCOE_FIP_DBG("received FLOGI LS_ACC using non-FIP mode\n");
 
 		/*
 		 * FLOGI accepted.
@@ -1276,7 +1290,7 @@ int fcoe_ctlr_recv_flogi(struct fcoe_ctlr *fip, struct fc_frame *fp, u8 *sa)
 			memcpy(fip->dest_addr, sa, ETH_ALEN);
 			fip->map_dest = 0;
 			if (fip->state == FIP_ST_NON_FIP)
-				FIP_DBG("received FLOGI REQ, "
+				LIBFCOE_FIP_DBG("received FLOGI REQ, "
 						"using non-FIP mode\n");
 			fip->state = FIP_ST_NON_FIP;
 		}
-- 
cgit v1.2.3-59-g8ed1b


From 7414705ea4aef9ce438e547f3138a680d2d1096c Mon Sep 17 00:00:00 2001
From: Robert Love <robert.w.love@intel.com>
Date: Wed, 10 Jun 2009 15:31:10 -0700
Subject: libfc: Add runtime debugging with debug_logging module parameter

This patch adds the /sys/module/libfc/parameters/debug_logging
file to sysfs as a module parameter. It accepts an integer
bitmask for logging. Currently it supports:

   bit
LSB 0 = general libfc debugging
    1 = lport debugging
    2 = disc debugging
    3 = rport debugging
    4 = fcp debugging
    5 = EM debugging
    6 = exch/seq debugging
    7 = scsi logging (mostly error handling)

the other bits are not used at this time.

The patch converts all of the libfc source files to use
these new macros and removes the old FC_DBG macro.

Signed-off-by: Robert Love <robert.w.love@intel.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/libfc/fc_disc.c  |  83 ++++++++++------------
 drivers/scsi/libfc/fc_exch.c  |  58 +++++++---------
 drivers/scsi/libfc/fc_fcp.c   |  97 +++++++++++++-------------
 drivers/scsi/libfc/fc_lport.c | 156 ++++++++++++++++++++----------------------
 drivers/scsi/libfc/fc_rport.c | 120 ++++++++++++++------------------
 include/scsi/fc_encode.h      |   2 -
 include/scsi/libfc.h          |  77 ++++++++++++++++++---
 7 files changed, 301 insertions(+), 292 deletions(-)

diff --git a/drivers/scsi/libfc/fc_disc.c b/drivers/scsi/libfc/fc_disc.c
index 4c880656990b..6fabf66972b9 100644
--- a/drivers/scsi/libfc/fc_disc.c
+++ b/drivers/scsi/libfc/fc_disc.c
@@ -45,14 +45,6 @@
 
 #define	FC_DISC_DELAY		3
 
-static int fc_disc_debug;
-
-#define FC_DEBUG_DISC(fmt...)			\
-	do {					\
-		if (fc_disc_debug)		\
-			FC_DBG(fmt);		\
-	} while (0)
-
 static void fc_disc_gpn_ft_req(struct fc_disc *);
 static void fc_disc_gpn_ft_resp(struct fc_seq *, struct fc_frame *, void *);
 static int fc_disc_new_target(struct fc_disc *, struct fc_rport *,
@@ -137,8 +129,8 @@ static void fc_disc_rport_callback(struct fc_lport *lport,
 	struct fc_rport_libfc_priv *rdata = rport->dd_data;
 	struct fc_disc *disc = &lport->disc;
 
-	FC_DEBUG_DISC("Received a %d event for port (%6x)\n", event,
-		      rport->port_id);
+	FC_DISC_DBG(disc, "Received a %d event for port (%6x)\n", event,
+		    rport->port_id);
 
 	switch (event) {
 	case RPORT_EV_CREATED:
@@ -191,8 +183,7 @@ static void fc_disc_recv_rscn_req(struct fc_seq *sp, struct fc_frame *fp,
 
 	lport = disc->lport;
 
-	FC_DEBUG_DISC("Received an RSCN event on port (%6x)\n",
-		      fc_host_port_id(lport->host));
+	FC_DISC_DBG(disc, "Received an RSCN event\n");
 
 	/* make sure the frame contains an RSCN message */
 	rp = fc_frame_payload_get(fp, sizeof(*rp));
@@ -225,8 +216,8 @@ static void fc_disc_recv_rscn_req(struct fc_seq *sp, struct fc_frame *fp,
 		 */
 		switch (fmt) {
 		case ELS_ADDR_FMT_PORT:
-			FC_DEBUG_DISC("Port address format for port (%6x)\n",
-				      ntoh24(pp->rscn_fid));
+			FC_DISC_DBG(disc, "Port address format for port "
+				    "(%6x)\n", ntoh24(pp->rscn_fid));
 			dp = kzalloc(sizeof(*dp), GFP_KERNEL);
 			if (!dp) {
 				redisc = 1;
@@ -243,19 +234,19 @@ static void fc_disc_recv_rscn_req(struct fc_seq *sp, struct fc_frame *fp,
 		case ELS_ADDR_FMT_DOM:
 		case ELS_ADDR_FMT_FAB:
 		default:
-			FC_DEBUG_DISC("Address format is (%d)\n", fmt);
+			FC_DISC_DBG(disc, "Address format is (%d)\n", fmt);
 			redisc = 1;
 			break;
 		}
 	}
 	lport->tt.seq_els_rsp_send(sp, ELS_LS_ACC, NULL);
 	if (redisc) {
-		FC_DEBUG_DISC("RSCN received: rediscovering\n");
+		FC_DISC_DBG(disc, "RSCN received: rediscovering\n");
 		fc_disc_restart(disc);
 	} else {
-		FC_DEBUG_DISC("RSCN received: not rediscovering. "
-			      "redisc %d state %d in_prog %d\n",
-			      redisc, lport->state, disc->pending);
+		FC_DISC_DBG(disc, "RSCN received: not rediscovering. "
+			    "redisc %d state %d in_prog %d\n",
+			    redisc, lport->state, disc->pending);
 		list_for_each_entry_safe(dp, next, &disc_ports, peers) {
 			list_del(&dp->peers);
 			rport = lport->tt.rport_lookup(lport, dp->ids.port_id);
@@ -270,7 +261,7 @@ static void fc_disc_recv_rscn_req(struct fc_seq *sp, struct fc_frame *fp,
 	fc_frame_free(fp);
 	return;
 reject:
-	FC_DEBUG_DISC("Received a bad RSCN frame\n");
+	FC_DISC_DBG(disc, "Received a bad RSCN frame\n");
 	rjt_data.fp = NULL;
 	rjt_data.reason = ELS_RJT_LOGIC;
 	rjt_data.explan = ELS_EXPL_NONE;
@@ -302,7 +293,8 @@ static void fc_disc_recv_req(struct fc_seq *sp, struct fc_frame *fp,
 		mutex_unlock(&disc->disc_mutex);
 		break;
 	default:
-		FC_DBG("Received an unsupported request. opcode (%x)\n", op);
+		FC_DISC_DBG(disc, "Received an unsupported request, "
+			    "the opcode is (%x)\n", op);
 		break;
 	}
 }
@@ -320,12 +312,10 @@ static void fc_disc_restart(struct fc_disc *disc)
 	struct fc_rport_libfc_priv *rdata, *next;
 	struct fc_lport *lport = disc->lport;
 
-	FC_DEBUG_DISC("Restarting discovery for port (%6x)\n",
-		      fc_host_port_id(lport->host));
+	FC_DISC_DBG(disc, "Restarting discovery\n");
 
 	list_for_each_entry_safe(rdata, next, &disc->rports, peers) {
 		rport = PRIV_TO_RPORT(rdata);
-		FC_DEBUG_DISC("list_del(%6x)\n", rport->port_id);
 		list_del(&rdata->peers);
 		lport->tt.rport_logoff(rport);
 	}
@@ -485,8 +475,7 @@ static void fc_disc_done(struct fc_disc *disc)
 	struct fc_lport *lport = disc->lport;
 	enum fc_disc_event event;
 
-	FC_DEBUG_DISC("Discovery complete for port (%6x)\n",
-		      fc_host_port_id(lport->host));
+	FC_DISC_DBG(disc, "Discovery complete\n");
 
 	event = disc->event;
 	disc->event = DISC_EV_NONE;
@@ -510,10 +499,10 @@ static void fc_disc_error(struct fc_disc *disc, struct fc_frame *fp)
 {
 	struct fc_lport *lport = disc->lport;
 	unsigned long delay = 0;
-	if (fc_disc_debug)
-		FC_DBG("Error %ld, retries %d/%d\n",
-		       PTR_ERR(fp), disc->retry_count,
-		       FC_DISC_RETRY_LIMIT);
+
+	FC_DISC_DBG(disc, "Error %ld, retries %d/%d\n",
+		    PTR_ERR(fp), disc->retry_count,
+		    FC_DISC_RETRY_LIMIT);
 
 	if (!fp || PTR_ERR(fp) == -FC_EX_TIMEOUT) {
 		/*
@@ -649,9 +638,9 @@ static int fc_disc_gpn_ft_parse(struct fc_disc *disc, void *buf, size_t len)
 					      &disc->rogue_rports);
 				lport->tt.rport_login(rport);
 			} else
-				FC_DBG("Failed to allocate memory for "
-				       "the newly discovered port (%6x)\n",
-				       dp.ids.port_id);
+				printk(KERN_WARNING "libfc: Failed to allocate "
+				       "memory for the newly discovered port "
+				       "(%6x)\n", dp.ids.port_id);
 		}
 
 		if (np->fp_flags & FC_NS_FID_LAST) {
@@ -671,9 +660,8 @@ static int fc_disc_gpn_ft_parse(struct fc_disc *disc, void *buf, size_t len)
 	 */
 	if (error == 0 && len > 0 && len < sizeof(*np)) {
 		if (np != &disc->partial_buf) {
-			FC_DEBUG_DISC("Partial buffer remains "
-				      "for discovery by (%6x)\n",
-				      fc_host_port_id(lport->host));
+			FC_DISC_DBG(disc, "Partial buffer remains "
+				    "for discovery\n");
 			memcpy(&disc->partial_buf, np, len);
 		}
 		disc->buf_len = (unsigned char) len;
@@ -721,8 +709,7 @@ static void fc_disc_gpn_ft_resp(struct fc_seq *sp, struct fc_frame *fp,
 	int error;
 
 	mutex_lock(&disc->disc_mutex);
-	FC_DEBUG_DISC("Received a GPN_FT response on port (%6x)\n",
-		      fc_host_port_id(disc->lport->host));
+	FC_DISC_DBG(disc, "Received a GPN_FT response\n");
 
 	if (IS_ERR(fp)) {
 		fc_disc_error(disc, fp);
@@ -738,30 +725,30 @@ static void fc_disc_gpn_ft_resp(struct fc_seq *sp, struct fc_frame *fp,
 	    disc->seq_count == 0) {
 		cp = fc_frame_payload_get(fp, sizeof(*cp));
 		if (!cp) {
-			FC_DBG("GPN_FT response too short, len %d\n",
-			       fr_len(fp));
+			FC_DISC_DBG(disc, "GPN_FT response too short, len %d\n",
+				    fr_len(fp));
 		} else if (ntohs(cp->ct_cmd) == FC_FS_ACC) {
 
 			/* Accepted, parse the response. */
 			buf = cp + 1;
 			len -= sizeof(*cp);
 		} else if (ntohs(cp->ct_cmd) == FC_FS_RJT) {
-			FC_DBG("GPN_FT rejected reason %x exp %x "
-			       "(check zoning)\n", cp->ct_reason,
-			       cp->ct_explan);
+			FC_DISC_DBG(disc, "GPN_FT rejected reason %x exp %x "
+				    "(check zoning)\n", cp->ct_reason,
+				    cp->ct_explan);
 			disc->event = DISC_EV_FAILED;
 			fc_disc_done(disc);
 		} else {
-			FC_DBG("GPN_FT unexpected response code %x\n",
-			       ntohs(cp->ct_cmd));
+			FC_DISC_DBG(disc, "GPN_FT unexpected response code "
+				    "%x\n", ntohs(cp->ct_cmd));
 		}
 	} else if (fr_sof(fp) == FC_SOF_N3 &&
 		   seq_cnt == disc->seq_count) {
 		buf = fh + 1;
 	} else {
-		FC_DBG("GPN_FT unexpected frame - out of sequence? "
-		       "seq_cnt %x expected %x sof %x eof %x\n",
-		       seq_cnt, disc->seq_count, fr_sof(fp), fr_eof(fp));
+		FC_DISC_DBG(disc, "GPN_FT unexpected frame - out of sequence? "
+			    "seq_cnt %x expected %x sof %x eof %x\n",
+			    seq_cnt, disc->seq_count, fr_sof(fp), fr_eof(fp));
 	}
 	if (buf) {
 		error = fc_disc_gpn_ft_parse(disc, buf, len);
diff --git a/drivers/scsi/libfc/fc_exch.c b/drivers/scsi/libfc/fc_exch.c
index 7af9bceb8aa9..2bc22be5f849 100644
--- a/drivers/scsi/libfc/fc_exch.c
+++ b/drivers/scsi/libfc/fc_exch.c
@@ -32,18 +32,7 @@
 #include <scsi/libfc.h>
 #include <scsi/fc_encode.h>
 
-/*
- * fc_exch_debug can be set in debugger or at compile time to get more logs.
- */
-static int fc_exch_debug;
-
-#define FC_DEBUG_EXCH(fmt...)			\
-	do {					\
-		if (fc_exch_debug)		\
-			FC_DBG(fmt);		\
-	} while (0)
-
-static struct kmem_cache *fc_em_cachep;	/* cache for exchanges */
+static struct kmem_cache *fc_em_cachep;        /* cache for exchanges */
 
 /*
  * Structure and function definitions for managing Fibre Channel Exchanges
@@ -333,8 +322,8 @@ static inline void fc_exch_timer_set_locked(struct fc_exch *ep,
 	if (ep->state & (FC_EX_RST_CLEANUP | FC_EX_DONE))
 		return;
 
-	FC_DEBUG_EXCH("Exchange (%4x) timed out, notifying the upper layer\n",
-		      ep->xid);
+	FC_EXCH_DBG(ep, "Exchange timed out, notifying the upper layer\n");
+
 	if (schedule_delayed_work(&ep->timeout_work,
 				  msecs_to_jiffies(timer_msec)))
 		fc_exch_hold(ep);		/* hold for timer */
@@ -545,7 +534,7 @@ struct fc_exch *fc_exch_alloc(struct fc_exch_mgr *mp,
 		/* alloc a new xid */
 		xid = fc_em_alloc_xid(mp, fp);
 		if (!xid) {
-			printk(KERN_ERR "fc_em_alloc_xid() failed\n");
+			printk(KERN_WARNING "libfc: Failed to allocate an exhange\n");
 			goto err;
 		}
 	}
@@ -820,8 +809,8 @@ static struct fc_seq *fc_seq_start_next_locked(struct fc_seq *sp)
 	struct fc_exch *ep = fc_seq_exch(sp);
 
 	sp = fc_seq_alloc(ep, ep->seq_id++);
-	FC_DEBUG_EXCH("exch %4x f_ctl %6x seq %2x\n",
-		      ep->xid, ep->f_ctl, sp->id);
+	FC_EXCH_DBG(ep, "f_ctl %6x seq %2x\n",
+		    ep->f_ctl, sp->id);
 	return sp;
 }
 /*
@@ -901,7 +890,7 @@ void fc_seq_els_rsp_send(struct fc_seq *sp, enum fc_els_cmd els_cmd,
 		fc_exch_els_rec(sp, els_data->fp);
 		break;
 	default:
-		FC_DBG("Invalid ELS CMD:%x\n", els_cmd);
+		FC_EXCH_DBG(fc_seq_exch(sp), "Invalid ELS CMD:%x\n", els_cmd);
 	}
 }
 EXPORT_SYMBOL(fc_seq_els_rsp_send);
@@ -1134,7 +1123,7 @@ static void fc_exch_recv_req(struct fc_lport *lp, struct fc_exch_mgr *mp,
 			lp->tt.lport_recv(lp, sp, fp);
 		fc_exch_release(ep);	/* release from lookup */
 	} else {
-		FC_DEBUG_EXCH("exch/seq lookup failed: reject %x\n", reject);
+		FC_EM_DBG(mp, "exch/seq lookup failed: reject %x\n", reject);
 		fc_frame_free(fp);
 	}
 }
@@ -1242,10 +1231,10 @@ static void fc_exch_recv_resp(struct fc_exch_mgr *mp, struct fc_frame *fp)
 	sp = fc_seq_lookup_orig(mp, fp);	/* doesn't hold sequence */
 	if (!sp) {
 		atomic_inc(&mp->stats.xid_not_found);
-		FC_DEBUG_EXCH("seq lookup failed\n");
+		FC_EM_DBG(mp, "seq lookup failed\n");
 	} else {
 		atomic_inc(&mp->stats.non_bls_resp);
-		FC_DEBUG_EXCH("non-BLS response to sequence");
+		FC_EM_DBG(mp, "non-BLS response to sequence");
 	}
 	fc_frame_free(fp);
 }
@@ -1266,8 +1255,8 @@ static void fc_exch_abts_resp(struct fc_exch *ep, struct fc_frame *fp)
 	int rc = 1, has_rec = 0;
 
 	fh = fc_frame_header_get(fp);
-	FC_DEBUG_EXCH("exch: BLS rctl %x - %s\n",
-		      fh->fh_r_ctl, fc_exch_rctl_name(fh->fh_r_ctl));
+	FC_EXCH_DBG(ep, "exch: BLS rctl %x - %s\n", fh->fh_r_ctl,
+		    fc_exch_rctl_name(fh->fh_r_ctl));
 
 	if (cancel_delayed_work_sync(&ep->timeout_work))
 		fc_exch_release(ep);	/* release from pending timer hold */
@@ -1359,9 +1348,9 @@ static void fc_exch_recv_bls(struct fc_exch_mgr *mp, struct fc_frame *fp)
 		case FC_RCTL_ACK_0:
 			break;
 		default:
-			FC_DEBUG_EXCH("BLS rctl %x - %s received",
-				      fh->fh_r_ctl,
-				      fc_exch_rctl_name(fh->fh_r_ctl));
+			FC_EXCH_DBG(ep, "BLS rctl %x - %s received",
+				    fh->fh_r_ctl,
+				    fc_exch_rctl_name(fh->fh_r_ctl));
 			break;
 		}
 		fc_frame_free(fp);
@@ -1599,7 +1588,8 @@ static void fc_exch_rrq_resp(struct fc_seq *sp, struct fc_frame *fp, void *arg)
 
 		if (err == -FC_EX_CLOSED || err == -FC_EX_TIMEOUT)
 			goto cleanup;
-		FC_DBG("Cannot process RRQ, because of frame error %d\n", err);
+		FC_EXCH_DBG(aborted_ep, "Cannot process RRQ, "
+			    "frame error %d\n", err);
 		return;
 	}
 
@@ -1608,12 +1598,13 @@ static void fc_exch_rrq_resp(struct fc_seq *sp, struct fc_frame *fp, void *arg)
 
 	switch (op) {
 	case ELS_LS_RJT:
-		FC_DBG("LS_RJT for RRQ");
+		FC_EXCH_DBG(aborted_ep, "LS_RJT for RRQ");
 		/* fall through */
 	case ELS_LS_ACC:
 		goto cleanup;
 	default:
-		FC_DBG("unexpected response op %x for RRQ", op);
+		FC_EXCH_DBG(aborted_ep, "unexpected response op %x "
+			    "for RRQ", op);
 		return;
 	}
 
@@ -1740,8 +1731,8 @@ struct fc_exch_mgr *fc_exch_mgr_alloc(struct fc_lport *lp,
 	size_t len;
 
 	if (max_xid <= min_xid || min_xid == 0 || max_xid == FC_XID_UNKNOWN) {
-		FC_DBG("Invalid min_xid 0x:%x and max_xid 0x:%x\n",
-		       min_xid, max_xid);
+		FC_LPORT_DBG(lp, "Invalid min_xid 0x:%x and max_xid 0x:%x\n",
+			     min_xid, max_xid);
 		return NULL;
 	}
 
@@ -1878,7 +1869,8 @@ void fc_exch_recv(struct fc_lport *lp, struct fc_exch_mgr *mp,
 
 	/* lport lock ? */
 	if (!lp || !mp || (lp->state == LPORT_ST_NONE)) {
-		FC_DBG("fc_lport or EM is not allocated and configured");
+		FC_LPORT_DBG(lp, "Receiving frames for an lport that "
+			     "has not been initialized correctly\n");
 		fc_frame_free(fp);
 		return;
 	}
@@ -1904,7 +1896,7 @@ void fc_exch_recv(struct fc_lport *lp, struct fc_exch_mgr *mp,
 			fc_exch_recv_req(lp, mp, fp);
 		break;
 	default:
-		FC_DBG("dropping invalid frame (eof %x)", fr_eof(fp));
+		FC_EM_DBG(mp, "dropping invalid frame (eof %x)", fr_eof(fp));
 		fc_frame_free(fp);
 		break;
 	}
diff --git a/drivers/scsi/libfc/fc_fcp.c b/drivers/scsi/libfc/fc_fcp.c
index ad8b747837b0..e303e0d12c4b 100644
--- a/drivers/scsi/libfc/fc_fcp.c
+++ b/drivers/scsi/libfc/fc_fcp.c
@@ -43,13 +43,9 @@ MODULE_AUTHOR("Open-FCoE.org");
 MODULE_DESCRIPTION("libfc");
 MODULE_LICENSE("GPL v2");
 
-static int fc_fcp_debug;
-
-#define FC_DEBUG_FCP(fmt...)			\
-	do {					\
-		if (fc_fcp_debug)		\
-			FC_DBG(fmt);		\
-	} while (0)
+unsigned int fc_debug_logging;
+module_param_named(debug_logging, fc_debug_logging, int, S_IRUGO|S_IWUSR);
+MODULE_PARM_DESC(debug_logging, "a bit mask of logging levels");
 
 static struct kmem_cache *scsi_pkt_cachep;
 
@@ -347,8 +343,8 @@ static void fc_fcp_recv_data(struct fc_fcp_pkt *fsp, struct fc_frame *fp)
 		if ((fr_flags(fp) & FCPHF_CRC_UNCHECKED) &&
 		    fc_frame_crc_check(fp))
 			goto crc_err;
-		FC_DEBUG_FCP("data received past end. len %zx offset %zx "
-			     "data_len %x\n", len, offset, fsp->data_len);
+		FC_FCP_DBG(fsp, "data received past end. len %zx offset %zx "
+			   "data_len %x\n", len, offset, fsp->data_len);
 		fc_fcp_retry_cmd(fsp);
 		return;
 	}
@@ -411,7 +407,8 @@ crc_err:
 			stats->ErrorFrames++;
 			/* FIXME - per cpu count, not total count! */
 			if (stats->InvalidCRCCount++ < 5)
-				printk(KERN_WARNING "CRC error on data frame for port (%6x)\n",
+				printk(KERN_WARNING "libfc: CRC error on data "
+				       "frame for port (%6x)\n",
 				       fc_host_port_id(lp->host));
 			/*
 			 * Assume the frame is total garbage.
@@ -475,14 +472,14 @@ static int fc_fcp_send_data(struct fc_fcp_pkt *fsp, struct fc_seq *seq,
 	WARN_ON(seq_blen <= 0);
 	if (unlikely(offset + seq_blen > fsp->data_len)) {
 		/* this should never happen */
-		FC_DEBUG_FCP("xfer-ready past end. seq_blen %zx offset %zx\n",
-			     seq_blen, offset);
+		FC_FCP_DBG(fsp, "xfer-ready past end. seq_blen %zx "
+			   "offset %zx\n", seq_blen, offset);
 		fc_fcp_send_abort(fsp);
 		return 0;
 	} else if (offset != fsp->xfer_len) {
 		/* Out of Order Data Request - no problem, but unexpected. */
-		FC_DEBUG_FCP("xfer-ready non-contiguous. "
-			     "seq_blen %zx offset %zx\n", seq_blen, offset);
+		FC_FCP_DBG(fsp, "xfer-ready non-contiguous. "
+			   "seq_blen %zx offset %zx\n", seq_blen, offset);
 	}
 
 	/*
@@ -493,7 +490,7 @@ static int fc_fcp_send_data(struct fc_fcp_pkt *fsp, struct fc_seq *seq,
 	t_blen = fsp->max_payload;
 	if (lp->seq_offload) {
 		t_blen = min(seq_blen, (size_t)lp->lso_max);
-		FC_DEBUG_FCP("fsp=%p:lso:blen=%zx lso_max=0x%x t_blen=%zx\n",
+		FC_FCP_DBG(fsp, "fsp=%p:lso:blen=%zx lso_max=0x%x t_blen=%zx\n",
 			   fsp, seq_blen, lp->lso_max, t_blen);
 	}
 
@@ -694,7 +691,7 @@ static void fc_fcp_reduce_can_queue(struct fc_lport *lp)
 	if (!can_queue)
 		can_queue = 1;
 	lp->host->can_queue = can_queue;
-	shost_printk(KERN_ERR, lp->host, "Could not allocate frame.\n"
+	shost_printk(KERN_ERR, lp->host, "libfc: Could not allocate frame.\n"
 		     "Reducing can_queue to %d.\n", can_queue);
 done:
 	spin_unlock_irqrestore(lp->host->host_lock, flags);
@@ -768,7 +765,7 @@ static void fc_fcp_recv(struct fc_seq *seq, struct fc_frame *fp, void *arg)
 
 		fc_fcp_resp(fsp, fp);
 	} else {
-		FC_DBG("unexpected frame.  r_ctl %x\n", r_ctl);
+		FC_FCP_DBG(fsp, "unexpected frame.  r_ctl %x\n", r_ctl);
 	}
 unlock:
 	fc_fcp_unlock_pkt(fsp);
@@ -877,17 +874,17 @@ static void fc_fcp_resp(struct fc_fcp_pkt *fsp, struct fc_frame *fp)
 			return;
 		}
 		fsp->status_code = FC_DATA_OVRRUN;
-		FC_DBG("tgt %6x xfer len %zx greater than expected len %x. "
-		       "data len %x\n",
-		       fsp->rport->port_id,
-		       fsp->xfer_len, expected_len, fsp->data_len);
+		FC_FCP_DBG(fsp, "tgt %6x xfer len %zx greater than expected, "
+			   "len %x, data len %x\n",
+			   fsp->rport->port_id,
+			   fsp->xfer_len, expected_len, fsp->data_len);
 	}
 	fc_fcp_complete_locked(fsp);
 	return;
 
 len_err:
-	FC_DBG("short FCP response. flags 0x%x len %u respl %u snsl %u\n",
-	       flags, fr_len(fp), respl, snsl);
+	FC_FCP_DBG(fsp, "short FCP response. flags 0x%x len %u respl %u "
+		   "snsl %u\n", flags, fr_len(fp), respl, snsl);
 err:
 	fsp->status_code = FC_ERROR;
 	fc_fcp_complete_locked(fsp);
@@ -1107,13 +1104,11 @@ static void fc_fcp_error(struct fc_fcp_pkt *fsp, struct fc_frame *fp)
 	if (fc_fcp_lock_pkt(fsp))
 		return;
 
-	switch (error) {
-	case -FC_EX_CLOSED:
+	if (error == -FC_EX_CLOSED) {
 		fc_fcp_retry_cmd(fsp);
 		goto unlock;
-	default:
-		FC_DBG("unknown error %ld\n", PTR_ERR(fp));
 	}
+
 	/*
 	 * clear abort pending, because the lower layer
 	 * decided to force completion.
@@ -1145,10 +1140,10 @@ static int fc_fcp_pkt_abort(struct fc_lport *lp, struct fc_fcp_pkt *fsp)
 	fsp->wait_for_comp = 0;
 
 	if (!rc) {
-		FC_DBG("target abort cmd  failed\n");
+		FC_FCP_DBG(fsp, "target abort cmd  failed\n");
 		rc = FAILED;
 	} else if (fsp->state & FC_SRB_ABORTED) {
-		FC_DBG("target abort cmd  passed\n");
+		FC_FCP_DBG(fsp, "target abort cmd  passed\n");
 		rc = SUCCESS;
 		fc_fcp_complete_locked(fsp);
 	}
@@ -1213,7 +1208,7 @@ static int fc_lun_reset(struct fc_lport *lp, struct fc_fcp_pkt *fsp,
 	spin_unlock_bh(&fsp->scsi_pkt_lock);
 
 	if (!rc) {
-		FC_DBG("lun reset failed\n");
+		FC_SCSI_DBG(lp, "lun reset failed\n");
 		return FAILED;
 	}
 
@@ -1221,7 +1216,7 @@ static int fc_lun_reset(struct fc_lport *lp, struct fc_fcp_pkt *fsp,
 	if (fsp->cdb_status != FCP_TMF_CMPL)
 		return FAILED;
 
-	FC_DBG("lun reset to lun %u completed\n", lun);
+	FC_SCSI_DBG(lp, "lun reset to lun %u completed\n", lun);
 	fc_fcp_cleanup_each_cmd(lp, id, lun, FC_CMD_ABORTED);
 	return SUCCESS;
 }
@@ -1388,13 +1383,13 @@ static void fc_fcp_rec_resp(struct fc_seq *seq, struct fc_frame *fp, void *arg)
 		rjt = fc_frame_payload_get(fp, sizeof(*rjt));
 		switch (rjt->er_reason) {
 		default:
-			FC_DEBUG_FCP("device %x unexpected REC reject "
-				     "reason %d expl %d\n",
-				     fsp->rport->port_id, rjt->er_reason,
-				     rjt->er_explan);
+			FC_FCP_DBG(fsp, "device %x unexpected REC reject "
+				   "reason %d expl %d\n",
+				   fsp->rport->port_id, rjt->er_reason,
+				   rjt->er_explan);
 			/* fall through */
 		case ELS_RJT_UNSUP:
-			FC_DEBUG_FCP("device does not support REC\n");
+			FC_FCP_DBG(fsp, "device does not support REC\n");
 			rp = fsp->rport->dd_data;
 			/*
 			 * if we do not spport RECs or got some bogus
@@ -1514,8 +1509,8 @@ static void fc_fcp_rec_error(struct fc_fcp_pkt *fsp, struct fc_frame *fp)
 		break;
 
 	default:
-		FC_DBG("REC %p fid %x error unexpected error %d\n",
-		       fsp, fsp->rport->port_id, error);
+		FC_FCP_DBG(fsp, "REC %p fid %x error unexpected error %d\n",
+			   fsp, fsp->rport->port_id, error);
 		fsp->status_code = FC_CMD_PLOGO;
 		/* fall through */
 
@@ -1524,9 +1519,9 @@ static void fc_fcp_rec_error(struct fc_fcp_pkt *fsp, struct fc_frame *fp)
 		 * Assume REC or LS_ACC was lost.
 		 * The exchange manager will have aborted REC, so retry.
 		 */
-		FC_DBG("REC fid %x error error %d retry %d/%d\n",
-		       fsp->rport->port_id, error, fsp->recov_retry,
-		       FC_MAX_RECOV_RETRY);
+		FC_FCP_DBG(fsp, "REC fid %x error error %d retry %d/%d\n",
+			   fsp->rport->port_id, error, fsp->recov_retry,
+			   FC_MAX_RECOV_RETRY);
 		if (fsp->recov_retry++ < FC_MAX_RECOV_RETRY)
 			fc_fcp_rec(fsp);
 		else
@@ -2011,9 +2006,11 @@ int fc_eh_device_reset(struct scsi_cmnd *sc_cmd)
 	if (lp->state != LPORT_ST_READY)
 		return rc;
 
+	FC_SCSI_DBG(lp, "Resetting rport (%6x)\n", rport->port_id);
+
 	fsp = fc_fcp_pkt_alloc(lp, GFP_NOIO);
 	if (fsp == NULL) {
-		FC_DBG("could not allocate scsi_pkt\n");
+		printk(KERN_WARNING "libfc: could not allocate scsi_pkt\n");
 		sc_cmd->result = DID_NO_CONNECT << 16;
 		goto out;
 	}
@@ -2048,17 +2045,21 @@ int fc_eh_host_reset(struct scsi_cmnd *sc_cmd)
 	struct fc_lport *lp = shost_priv(shost);
 	unsigned long wait_tmo;
 
+	FC_SCSI_DBG(lp, "Resetting host\n");
+
 	lp->tt.lport_reset(lp);
 	wait_tmo = jiffies + FC_HOST_RESET_TIMEOUT;
 	while (!fc_fcp_lport_queue_ready(lp) && time_before(jiffies, wait_tmo))
 		msleep(1000);
 
 	if (fc_fcp_lport_queue_ready(lp)) {
-		shost_printk(KERN_INFO, shost, "Host reset succeeded.\n");
+		shost_printk(KERN_INFO, shost, "libfc: Host reset succeeded "
+			     "on port (%6x)\n", fc_host_port_id(lp->host));
 		return SUCCESS;
 	} else {
-		shost_printk(KERN_INFO, shost, "Host reset failed. "
-			     "lport not ready.\n");
+		shost_printk(KERN_INFO, shost, "libfc: Host reset failed, "
+			     "port (%6x) is not ready.\n",
+			     fc_host_port_id(lp->host));
 		return FAILED;
 	}
 }
@@ -2117,7 +2118,8 @@ void fc_fcp_destroy(struct fc_lport *lp)
 	struct fc_fcp_internal *si = fc_get_scsi_internal(lp);
 
 	if (!list_empty(&si->scsi_pkt_queue))
-		printk(KERN_ERR "Leaked scsi packets.\n");
+		printk(KERN_ERR "libfc: Leaked SCSI packets when destroying "
+		       "port (%6x)\n", fc_host_port_id(lp->host));
 
 	mempool_destroy(si->scsi_pkt_pool);
 	kfree(si);
@@ -2166,7 +2168,8 @@ static int __init libfc_init(void)
 					    sizeof(struct fc_fcp_pkt),
 					    0, SLAB_HWCACHE_ALIGN, NULL);
 	if (scsi_pkt_cachep == NULL) {
-		FC_DBG("Unable to allocate SRB cache...module load failed!");
+		printk(KERN_ERR "libfc: Unable to allocate SRB cache, "
+		       "module load failed!");
 		return -ENOMEM;
 	}
 
diff --git a/drivers/scsi/libfc/fc_lport.c b/drivers/scsi/libfc/fc_lport.c
index e0c247724d2b..745fa5555d6a 100644
--- a/drivers/scsi/libfc/fc_lport.c
+++ b/drivers/scsi/libfc/fc_lport.c
@@ -101,14 +101,6 @@
 
 #define	DNS_DELAY	      3 /* Discovery delay after RSCN (in seconds)*/
 
-static int fc_lport_debug;
-
-#define FC_DEBUG_LPORT(fmt...)			\
-	do {					\
-		if (fc_lport_debug)		\
-			FC_DBG(fmt);		\
-	} while (0)
-
 static void fc_lport_error(struct fc_lport *, struct fc_frame *);
 
 static void fc_lport_enter_reset(struct fc_lport *);
@@ -151,8 +143,8 @@ static void fc_lport_rport_callback(struct fc_lport *lport,
 				    struct fc_rport *rport,
 				    enum fc_rport_event event)
 {
-	FC_DEBUG_LPORT("Received a %d event for port (%6x)\n", event,
-		       rport->port_id);
+	FC_LPORT_DBG(lport, "Received a %d event for port (%6x)\n", event,
+		     rport->port_id);
 
 	switch (event) {
 	case RPORT_EV_CREATED:
@@ -162,19 +154,19 @@ static void fc_lport_rport_callback(struct fc_lport *lport,
 				lport->dns_rp = rport;
 				fc_lport_enter_rpn_id(lport);
 			} else {
-				FC_DEBUG_LPORT("Received an CREATED event on "
-					       "port (%6x) for the directory "
-					       "server, but the lport is not "
-					       "in the DNS state, it's in the "
-					       "%d state", rport->port_id,
-					       lport->state);
+				FC_LPORT_DBG(lport, "Received an CREATED event "
+					     "on port (%6x) for the directory "
+					     "server, but the lport is not "
+					     "in the DNS state, it's in the "
+					     "%d state", rport->port_id,
+					     lport->state);
 				lport->tt.rport_logoff(rport);
 			}
 			mutex_unlock(&lport->lp_mutex);
 		} else
-			FC_DEBUG_LPORT("Received an event for port (%6x) "
-				       "which is not the directory server\n",
-				       rport->port_id);
+			FC_LPORT_DBG(lport, "Received an event for port (%6x) "
+				     "which is not the directory server\n",
+				     rport->port_id);
 		break;
 	case RPORT_EV_LOGO:
 	case RPORT_EV_FAILED:
@@ -185,9 +177,9 @@ static void fc_lport_rport_callback(struct fc_lport *lport,
 			mutex_unlock(&lport->lp_mutex);
 
 		} else
-			FC_DEBUG_LPORT("Received an event for port (%6x) "
-				       "which is not the directory server\n",
-				       rport->port_id);
+			FC_LPORT_DBG(lport, "Received an event for port (%6x) "
+				     "which is not the directory server\n",
+				     rport->port_id);
 		break;
 	case RPORT_EV_NONE:
 		break;
@@ -363,8 +355,8 @@ static void fc_lport_add_fc4_type(struct fc_lport *lport, enum fc_fh_type type)
 static void fc_lport_recv_rlir_req(struct fc_seq *sp, struct fc_frame *fp,
 				   struct fc_lport *lport)
 {
-	FC_DEBUG_LPORT("Received RLIR request while in state %s\n",
-		       fc_lport_state(lport));
+	FC_LPORT_DBG(lport, "Received RLIR request while in state %s\n",
+		     fc_lport_state(lport));
 
 	lport->tt.seq_els_rsp_send(sp, ELS_LS_ACC, NULL);
 	fc_frame_free(fp);
@@ -389,8 +381,8 @@ static void fc_lport_recv_echo_req(struct fc_seq *sp, struct fc_frame *in_fp,
 	void *dp;
 	u32 f_ctl;
 
-	FC_DEBUG_LPORT("Received RLIR request while in state %s\n",
-		       fc_lport_state(lport));
+	FC_LPORT_DBG(lport, "Received RLIR request while in state %s\n",
+		     fc_lport_state(lport));
 
 	len = fr_len(in_fp) - sizeof(struct fc_frame_header);
 	pp = fc_frame_payload_get(in_fp, len);
@@ -437,8 +429,8 @@ static void fc_lport_recv_rnid_req(struct fc_seq *sp, struct fc_frame *in_fp,
 	size_t len;
 	u32 f_ctl;
 
-	FC_DEBUG_LPORT("Received RNID request while in state %s\n",
-		       fc_lport_state(lport));
+	FC_LPORT_DBG(lport, "Received RNID request while in state %s\n",
+		     fc_lport_state(lport));
 
 	req = fc_frame_payload_get(in_fp, sizeof(*req));
 	if (!req) {
@@ -498,8 +490,8 @@ static void fc_lport_recv_adisc_req(struct fc_seq *sp, struct fc_frame *in_fp,
 	size_t len;
 	u32 f_ctl;
 
-	FC_DEBUG_LPORT("Received ADISC request while in state %s\n",
-		       fc_lport_state(lport));
+	FC_LPORT_DBG(lport, "Received ADISC request while in state %s\n",
+		     fc_lport_state(lport));
 
 	req = fc_frame_payload_get(in_fp, sizeof(*req));
 	if (!req) {
@@ -574,8 +566,8 @@ EXPORT_SYMBOL(fc_fabric_login);
  */
 void fc_linkup(struct fc_lport *lport)
 {
-	FC_DEBUG_LPORT("Link is up for port (%6x)\n",
-		       fc_host_port_id(lport->host));
+	printk(KERN_INFO "libfc: Link up on port (%6x)\n",
+	       fc_host_port_id(lport->host));
 
 	mutex_lock(&lport->lp_mutex);
 	if (!lport->link_up) {
@@ -595,8 +587,8 @@ EXPORT_SYMBOL(fc_linkup);
 void fc_linkdown(struct fc_lport *lport)
 {
 	mutex_lock(&lport->lp_mutex);
-	FC_DEBUG_LPORT("Link is down for port (%6x)\n",
-		       fc_host_port_id(lport->host));
+	printk(KERN_INFO "libfc: Link down on port (%6x)\n",
+	       fc_host_port_id(lport->host));
 
 	if (lport->link_up) {
 		lport->link_up = 0;
@@ -701,12 +693,11 @@ void fc_lport_disc_callback(struct fc_lport *lport, enum fc_disc_event event)
 {
 	switch (event) {
 	case DISC_EV_SUCCESS:
-		FC_DEBUG_LPORT("Got a SUCCESS event for port (%6x)\n",
-			       fc_host_port_id(lport->host));
+		FC_LPORT_DBG(lport, "Discovery succeeded\n");
 		break;
 	case DISC_EV_FAILED:
-		FC_DEBUG_LPORT("Got a FAILED event for port (%6x)\n",
-			       fc_host_port_id(lport->host));
+		printk(KERN_ERR "libfc: Discovery failed for port (%6x)\n",
+		       fc_host_port_id(lport->host));
 		mutex_lock(&lport->lp_mutex);
 		fc_lport_enter_reset(lport);
 		mutex_unlock(&lport->lp_mutex);
@@ -726,8 +717,8 @@ void fc_lport_disc_callback(struct fc_lport *lport, enum fc_disc_event event)
  */
 static void fc_lport_enter_ready(struct fc_lport *lport)
 {
-	FC_DEBUG_LPORT("Port (%6x) entered Ready from state %s\n",
-		       fc_host_port_id(lport->host), fc_lport_state(lport));
+	FC_LPORT_DBG(lport, "Entered READY from state %s\n",
+		     fc_lport_state(lport));
 
 	fc_lport_state_enter(lport, LPORT_ST_READY);
 
@@ -762,8 +753,8 @@ static void fc_lport_recv_flogi_req(struct fc_seq *sp_in,
 	u32 local_fid;
 	u32 f_ctl;
 
-	FC_DEBUG_LPORT("Received FLOGI request while in state %s\n",
-		       fc_lport_state(lport));
+	FC_LPORT_DBG(lport, "Received FLOGI request while in state %s\n",
+		     fc_lport_state(lport));
 
 	fh = fc_frame_header_get(rx_fp);
 	remote_fid = ntoh24(fh->fh_s_id);
@@ -772,12 +763,11 @@ static void fc_lport_recv_flogi_req(struct fc_seq *sp_in,
 		goto out;
 	remote_wwpn = get_unaligned_be64(&flp->fl_wwpn);
 	if (remote_wwpn == lport->wwpn) {
-		FC_DBG("FLOGI from port with same WWPN %llx "
-		       "possible configuration error\n",
-		       (unsigned long long)remote_wwpn);
+		printk(KERN_WARNING "libfc: Received FLOGI from port "
+		       "with same WWPN %llx\n", remote_wwpn);
 		goto out;
 	}
-	FC_DBG("FLOGI from port WWPN %llx\n", (unsigned long long)remote_wwpn);
+	FC_LPORT_DBG(lport, "FLOGI from port WWPN %llx\n", remote_wwpn);
 
 	/*
 	 * XXX what is the right thing to do for FIDs?
@@ -909,7 +899,8 @@ static void fc_lport_recv_req(struct fc_lport *lport, struct fc_seq *sp,
 			}
 		}
 	} else {
-		FC_DBG("dropping invalid frame (eof %x)\n", fr_eof(fp));
+		FC_LPORT_DBG(lport, "dropping invalid frame (eof %x)\n",
+			     fr_eof(fp));
 		fc_frame_free(fp);
 	}
 	mutex_unlock(&lport->lp_mutex);
@@ -947,8 +938,8 @@ EXPORT_SYMBOL(fc_lport_reset);
  */
 static void fc_lport_enter_reset(struct fc_lport *lport)
 {
-	FC_DEBUG_LPORT("Port (%6x) entered RESET state from %s state\n",
-		       fc_host_port_id(lport->host), fc_lport_state(lport));
+	FC_LPORT_DBG(lport, "Entered RESET state from %s state\n",
+		     fc_lport_state(lport));
 
 	fc_lport_state_enter(lport, LPORT_ST_RESET);
 
@@ -982,9 +973,9 @@ static void fc_lport_enter_reset(struct fc_lport *lport)
 static void fc_lport_error(struct fc_lport *lport, struct fc_frame *fp)
 {
 	unsigned long delay = 0;
-	FC_DEBUG_LPORT("Error %ld in state %s, retries %d\n",
-		       PTR_ERR(fp), fc_lport_state(lport),
-		       lport->retry_count);
+	FC_LPORT_DBG(lport, "Error %ld in state %s, retries %d\n",
+		     PTR_ERR(fp), fc_lport_state(lport),
+		     lport->retry_count);
 
 	if (!fp || PTR_ERR(fp) == -FC_EX_TIMEOUT) {
 		/*
@@ -1040,11 +1031,11 @@ static void fc_lport_rft_id_resp(struct fc_seq *sp, struct fc_frame *fp,
 
 	mutex_lock(&lport->lp_mutex);
 
-	FC_DEBUG_LPORT("Received a RFT_ID response\n");
+	FC_LPORT_DBG(lport, "Received a RFT_ID response\n");
 
 	if (lport->state != LPORT_ST_RFT_ID) {
-		FC_DBG("Received a RFT_ID response, but in state %s\n",
-		       fc_lport_state(lport));
+		FC_LPORT_DBG(lport, "Received a RFT_ID response, but in state "
+			     "%s\n", fc_lport_state(lport));
 		if (IS_ERR(fp))
 			goto err;
 		goto out;
@@ -1094,11 +1085,11 @@ static void fc_lport_rpn_id_resp(struct fc_seq *sp, struct fc_frame *fp,
 
 	mutex_lock(&lport->lp_mutex);
 
-	FC_DEBUG_LPORT("Received a RPN_ID response\n");
+	FC_LPORT_DBG(lport, "Received a RPN_ID response\n");
 
 	if (lport->state != LPORT_ST_RPN_ID) {
-		FC_DBG("Received a RPN_ID response, but in state %s\n",
-		       fc_lport_state(lport));
+		FC_LPORT_DBG(lport, "Received a RPN_ID response, but in state "
+			     "%s\n", fc_lport_state(lport));
 		if (IS_ERR(fp))
 			goto err;
 		goto out;
@@ -1146,11 +1137,11 @@ static void fc_lport_scr_resp(struct fc_seq *sp, struct fc_frame *fp,
 
 	mutex_lock(&lport->lp_mutex);
 
-	FC_DEBUG_LPORT("Received a SCR response\n");
+	FC_LPORT_DBG(lport, "Received a SCR response\n");
 
 	if (lport->state != LPORT_ST_SCR) {
-		FC_DBG("Received a SCR response, but in state %s\n",
-		       fc_lport_state(lport));
+		FC_LPORT_DBG(lport, "Received a SCR response, but in state "
+			     "%s\n", fc_lport_state(lport));
 		if (IS_ERR(fp))
 			goto err;
 		goto out;
@@ -1184,8 +1175,8 @@ static void fc_lport_enter_scr(struct fc_lport *lport)
 {
 	struct fc_frame *fp;
 
-	FC_DEBUG_LPORT("Port (%6x) entered SCR state from %s state\n",
-		       fc_host_port_id(lport->host), fc_lport_state(lport));
+	FC_LPORT_DBG(lport, "Entered SCR state from %s state\n",
+		     fc_lport_state(lport));
 
 	fc_lport_state_enter(lport, LPORT_ST_SCR);
 
@@ -1213,8 +1204,8 @@ static void fc_lport_enter_rft_id(struct fc_lport *lport)
 	struct fc_ns_fts *lps;
 	int i;
 
-	FC_DEBUG_LPORT("Port (%6x) entered RFT_ID state from %s state\n",
-		       fc_host_port_id(lport->host), fc_lport_state(lport));
+	FC_LPORT_DBG(lport, "Entered RFT_ID state from %s state\n",
+		     fc_lport_state(lport));
 
 	fc_lport_state_enter(lport, LPORT_ST_RFT_ID);
 
@@ -1253,8 +1244,8 @@ static void fc_lport_enter_rpn_id(struct fc_lport *lport)
 {
 	struct fc_frame *fp;
 
-	FC_DEBUG_LPORT("Port (%6x) entered RPN_ID state from %s state\n",
-		       fc_host_port_id(lport->host), fc_lport_state(lport));
+	FC_LPORT_DBG(lport, "Entered RPN_ID state from %s state\n",
+		     fc_lport_state(lport));
 
 	fc_lport_state_enter(lport, LPORT_ST_RPN_ID);
 
@@ -1294,8 +1285,8 @@ static void fc_lport_enter_dns(struct fc_lport *lport)
 	dp.ids.roles = FC_RPORT_ROLE_UNKNOWN;
 	dp.lp = lport;
 
-	FC_DEBUG_LPORT("Port (%6x) entered DNS state from %s state\n",
-		       fc_host_port_id(lport->host), fc_lport_state(lport));
+	FC_LPORT_DBG(lport, "Entered DNS state from %s state\n",
+		     fc_lport_state(lport));
 
 	fc_lport_state_enter(lport, LPORT_ST_DNS);
 
@@ -1374,11 +1365,11 @@ static void fc_lport_logo_resp(struct fc_seq *sp, struct fc_frame *fp,
 
 	mutex_lock(&lport->lp_mutex);
 
-	FC_DEBUG_LPORT("Received a LOGO response\n");
+	FC_LPORT_DBG(lport, "Received a LOGO response\n");
 
 	if (lport->state != LPORT_ST_LOGO) {
-		FC_DBG("Received a LOGO response, but in state %s\n",
-		       fc_lport_state(lport));
+		FC_LPORT_DBG(lport, "Received a LOGO response, but in state "
+			     "%s\n", fc_lport_state(lport));
 		if (IS_ERR(fp))
 			goto err;
 		goto out;
@@ -1413,8 +1404,8 @@ static void fc_lport_enter_logo(struct fc_lport *lport)
 	struct fc_frame *fp;
 	struct fc_els_logo *logo;
 
-	FC_DEBUG_LPORT("Port (%6x) entered LOGO state from %s state\n",
-		       fc_host_port_id(lport->host), fc_lport_state(lport));
+	FC_LPORT_DBG(lport, "Entered LOGO state from %s state\n",
+		     fc_lport_state(lport));
 
 	fc_lport_state_enter(lport, LPORT_ST_LOGO);
 
@@ -1456,11 +1447,11 @@ static void fc_lport_flogi_resp(struct fc_seq *sp, struct fc_frame *fp,
 
 	mutex_lock(&lport->lp_mutex);
 
-	FC_DEBUG_LPORT("Received a FLOGI response\n");
+	FC_LPORT_DBG(lport, "Received a FLOGI response\n");
 
 	if (lport->state != LPORT_ST_FLOGI) {
-		FC_DBG("Received a FLOGI response, but in state %s\n",
-		       fc_lport_state(lport));
+		FC_LPORT_DBG(lport, "Received a FLOGI response, but in state "
+			     "%s\n", fc_lport_state(lport));
 		if (IS_ERR(fp))
 			goto err;
 		goto out;
@@ -1475,7 +1466,8 @@ static void fc_lport_flogi_resp(struct fc_seq *sp, struct fc_frame *fp,
 	did = ntoh24(fh->fh_d_id);
 	if (fc_frame_payload_op(fp) == ELS_LS_ACC && did != 0) {
 
-		FC_DEBUG_LPORT("Assigned fid %x\n", did);
+		printk(KERN_INFO "libfc: Assigned FID (%6x) in FLOGI response\n",
+		       did);
 		fc_host_port_id(lport->host) = did;
 
 		flp = fc_frame_payload_get(fp, sizeof(*flp));
@@ -1494,7 +1486,8 @@ static void fc_lport_flogi_resp(struct fc_seq *sp, struct fc_frame *fp,
 				if (e_d_tov > lport->e_d_tov)
 					lport->e_d_tov = e_d_tov;
 				lport->r_a_tov = 2 * e_d_tov;
-				FC_DBG("Point-to-Point mode\n");
+				printk(KERN_INFO "libfc: Port (%6x) entered "
+				       "point to point mode\n", did);
 				fc_lport_ptp_setup(lport, ntoh24(fh->fh_s_id),
 						   get_unaligned_be64(
 							   &flp->fl_wwpn),
@@ -1517,7 +1510,7 @@ static void fc_lport_flogi_resp(struct fc_seq *sp, struct fc_frame *fp,
 			}
 		}
 	} else {
-		FC_DBG("bad FLOGI response\n");
+		FC_LPORT_DBG(lport, "Bad FLOGI response\n");
 	}
 
 out:
@@ -1537,7 +1530,8 @@ void fc_lport_enter_flogi(struct fc_lport *lport)
 {
 	struct fc_frame *fp;
 
-	FC_DEBUG_LPORT("Processing FLOGI state\n");
+	FC_LPORT_DBG(lport, "Entered FLOGI state from %s state\n",
+		     fc_lport_state(lport));
 
 	fc_lport_state_enter(lport, LPORT_ST_FLOGI);
 
diff --git a/drivers/scsi/libfc/fc_rport.c b/drivers/scsi/libfc/fc_rport.c
index 7bfbff7e0efb..7162385f52eb 100644
--- a/drivers/scsi/libfc/fc_rport.c
+++ b/drivers/scsi/libfc/fc_rport.c
@@ -55,14 +55,6 @@
 #include <scsi/libfc.h>
 #include <scsi/fc_encode.h>
 
-static int fc_rport_debug;
-
-#define FC_DEBUG_RPORT(fmt...)			\
-	do {					\
-		if (fc_rport_debug)		\
-			FC_DBG(fmt);		\
-	} while (0)
-
 struct workqueue_struct *rport_event_queue;
 
 static void fc_rport_enter_plogi(struct fc_rport *);
@@ -97,7 +89,7 @@ static const char *fc_rport_state_names[] = {
 static void fc_rport_rogue_destroy(struct device *dev)
 {
 	struct fc_rport *rport = dev_to_rport(dev);
-	FC_DEBUG_RPORT("Destroying rogue rport (%6x)\n", rport->port_id);
+	FC_RPORT_DBG(rport, "Destroying rogue rport\n");
 	kfree(rport);
 }
 
@@ -263,8 +255,8 @@ static void fc_rport_work(struct work_struct *work)
 
 			fc_rport_state_enter(new_rport, RPORT_ST_READY);
 		} else {
-			FC_DBG("Failed to create the rport for port "
-			       "(%6x).\n", ids.port_id);
+			printk(KERN_WARNING "libfc: Failed to allocate "
+			       " memory for rport (%6x)\n", ids.port_id);
 			event = RPORT_EV_FAILED;
 		}
 		if (rport->port_id != FC_FID_DIR_SERV)
@@ -309,7 +301,7 @@ int fc_rport_login(struct fc_rport *rport)
 
 	mutex_lock(&rdata->rp_mutex);
 
-	FC_DEBUG_RPORT("Login to port (%6x)\n", rport->port_id);
+	FC_RPORT_DBG(rport, "Login to port\n");
 
 	fc_rport_enter_plogi(rport);
 
@@ -329,16 +321,13 @@ int fc_rport_login(struct fc_rport *rport)
 int fc_rport_logoff(struct fc_rport *rport)
 {
 	struct fc_rport_libfc_priv *rdata = rport->dd_data;
-	struct fc_lport *lport = rdata->local_port;
 
 	mutex_lock(&rdata->rp_mutex);
 
-	FC_DEBUG_RPORT("Remove port (%6x)\n", rport->port_id);
+	FC_RPORT_DBG(rport, "Remove port\n");
 
 	if (rdata->rp_state == RPORT_ST_NONE) {
-		FC_DEBUG_RPORT("(%6x): Port (%6x) in NONE state,"
-			       " not removing", fc_host_port_id(lport->host),
-			       rport->port_id);
+		FC_RPORT_DBG(rport, "Port in NONE state, not removing\n");
 		mutex_unlock(&rdata->rp_mutex);
 		goto out;
 	}
@@ -379,7 +368,7 @@ static void fc_rport_enter_ready(struct fc_rport *rport)
 
 	fc_rport_state_enter(rport, RPORT_ST_READY);
 
-	FC_DEBUG_RPORT("Port (%6x) is Ready\n", rport->port_id);
+	FC_RPORT_DBG(rport, "Port is Ready\n");
 
 	rdata->event = RPORT_EV_CREATED;
 	queue_work(rport_event_queue, &rdata->event_work);
@@ -436,8 +425,8 @@ static void fc_rport_error(struct fc_rport *rport, struct fc_frame *fp)
 {
 	struct fc_rport_libfc_priv *rdata = rport->dd_data;
 
-	FC_DEBUG_RPORT("Error %ld in state %s, retries %d\n",
-		       PTR_ERR(fp), fc_rport_state(rport), rdata->retries);
+	FC_RPORT_DBG(rport, "Error %ld in state %s, retries %d\n",
+		     PTR_ERR(fp), fc_rport_state(rport), rdata->retries);
 
 	switch (rdata->rp_state) {
 	case RPORT_ST_PLOGI:
@@ -479,8 +468,8 @@ static void fc_rport_error_retry(struct fc_rport *rport, struct fc_frame *fp)
 		return fc_rport_error(rport, fp);
 
 	if (rdata->retries < rdata->local_port->max_rport_retry_count) {
-		FC_DEBUG_RPORT("Error %ld in state %s, retrying\n",
-			       PTR_ERR(fp), fc_rport_state(rport));
+		FC_RPORT_DBG(rport, "Error %ld in state %s, retrying\n",
+			     PTR_ERR(fp), fc_rport_state(rport));
 		rdata->retries++;
 		/* no additional delay on exchange timeouts */
 		if (PTR_ERR(fp) == -FC_EX_TIMEOUT)
@@ -517,12 +506,11 @@ static void fc_rport_plogi_resp(struct fc_seq *sp, struct fc_frame *fp,
 
 	mutex_lock(&rdata->rp_mutex);
 
-	FC_DEBUG_RPORT("Received a PLOGI response from port (%6x)\n",
-		       rport->port_id);
+	FC_RPORT_DBG(rport, "Received a PLOGI response\n");
 
 	if (rdata->rp_state != RPORT_ST_PLOGI) {
-		FC_DBG("Received a PLOGI response, but in state %s\n",
-		       fc_rport_state(rport));
+		FC_RPORT_DBG(rport, "Received a PLOGI response, but in state "
+			     "%s\n", fc_rport_state(rport));
 		if (IS_ERR(fp))
 			goto err;
 		goto out;
@@ -583,8 +571,8 @@ static void fc_rport_enter_plogi(struct fc_rport *rport)
 	struct fc_lport *lport = rdata->local_port;
 	struct fc_frame *fp;
 
-	FC_DEBUG_RPORT("Port (%6x) entered PLOGI state from %s state\n",
-		       rport->port_id, fc_rport_state(rport));
+	FC_RPORT_DBG(rport, "Port entered PLOGI state from %s state\n",
+		     fc_rport_state(rport));
 
 	fc_rport_state_enter(rport, RPORT_ST_PLOGI);
 
@@ -628,12 +616,11 @@ static void fc_rport_prli_resp(struct fc_seq *sp, struct fc_frame *fp,
 
 	mutex_lock(&rdata->rp_mutex);
 
-	FC_DEBUG_RPORT("Received a PRLI response from port (%6x)\n",
-		       rport->port_id);
+	FC_RPORT_DBG(rport, "Received a PRLI response\n");
 
 	if (rdata->rp_state != RPORT_ST_PRLI) {
-		FC_DBG("Received a PRLI response, but in state %s\n",
-		       fc_rport_state(rport));
+		FC_RPORT_DBG(rport, "Received a PRLI response, but in state "
+			     "%s\n", fc_rport_state(rport));
 		if (IS_ERR(fp))
 			goto err;
 		goto out;
@@ -663,7 +650,7 @@ static void fc_rport_prli_resp(struct fc_seq *sp, struct fc_frame *fp,
 		fc_rport_enter_rtv(rport);
 
 	} else {
-		FC_DBG("Bad ELS response\n");
+		FC_RPORT_DBG(rport, "Bad ELS response for PRLI command\n");
 		rdata->event = RPORT_EV_FAILED;
 		fc_rport_state_enter(rport, RPORT_ST_NONE);
 		queue_work(rport_event_queue, &rdata->event_work);
@@ -695,12 +682,11 @@ static void fc_rport_logo_resp(struct fc_seq *sp, struct fc_frame *fp,
 
 	mutex_lock(&rdata->rp_mutex);
 
-	FC_DEBUG_RPORT("Received a LOGO response from port (%6x)\n",
-		       rport->port_id);
+	FC_RPORT_DBG(rport, "Received a LOGO response\n");
 
 	if (rdata->rp_state != RPORT_ST_LOGO) {
-		FC_DEBUG_RPORT("Received a LOGO response, but in state %s\n",
-			       fc_rport_state(rport));
+		FC_RPORT_DBG(rport, "Received a LOGO response, but in state "
+			     "%s\n", fc_rport_state(rport));
 		if (IS_ERR(fp))
 			goto err;
 		goto out;
@@ -715,7 +701,7 @@ static void fc_rport_logo_resp(struct fc_seq *sp, struct fc_frame *fp,
 	if (op == ELS_LS_ACC) {
 		fc_rport_enter_rtv(rport);
 	} else {
-		FC_DBG("Bad ELS response\n");
+		FC_RPORT_DBG(rport, "Bad ELS response for LOGO command\n");
 		rdata->event = RPORT_EV_LOGO;
 		fc_rport_state_enter(rport, RPORT_ST_NONE);
 		queue_work(rport_event_queue, &rdata->event_work);
@@ -745,8 +731,8 @@ static void fc_rport_enter_prli(struct fc_rport *rport)
 	} *pp;
 	struct fc_frame *fp;
 
-	FC_DEBUG_RPORT("Port (%6x) entered PRLI state from %s state\n",
-		       rport->port_id, fc_rport_state(rport));
+	FC_RPORT_DBG(rport, "Port entered PRLI state from %s state\n",
+		     fc_rport_state(rport));
 
 	fc_rport_state_enter(rport, RPORT_ST_PRLI);
 
@@ -784,12 +770,11 @@ static void fc_rport_rtv_resp(struct fc_seq *sp, struct fc_frame *fp,
 
 	mutex_lock(&rdata->rp_mutex);
 
-	FC_DEBUG_RPORT("Received a RTV response from port (%6x)\n",
-		       rport->port_id);
+	FC_RPORT_DBG(rport, "Received a RTV response\n");
 
 	if (rdata->rp_state != RPORT_ST_RTV) {
-		FC_DBG("Received a RTV response, but in state %s\n",
-		       fc_rport_state(rport));
+		FC_RPORT_DBG(rport, "Received a RTV response, but in state "
+			     "%s\n", fc_rport_state(rport));
 		if (IS_ERR(fp))
 			goto err;
 		goto out;
@@ -844,8 +829,8 @@ static void fc_rport_enter_rtv(struct fc_rport *rport)
 	struct fc_rport_libfc_priv *rdata = rport->dd_data;
 	struct fc_lport *lport = rdata->local_port;
 
-	FC_DEBUG_RPORT("Port (%6x) entered RTV state from %s state\n",
-		       rport->port_id, fc_rport_state(rport));
+	FC_RPORT_DBG(rport, "Port entered RTV state from %s state\n",
+		     fc_rport_state(rport));
 
 	fc_rport_state_enter(rport, RPORT_ST_RTV);
 
@@ -875,8 +860,8 @@ static void fc_rport_enter_logo(struct fc_rport *rport)
 	struct fc_lport *lport = rdata->local_port;
 	struct fc_frame *fp;
 
-	FC_DEBUG_RPORT("Port (%6x) entered LOGO state from %s state\n",
-		       rport->port_id, fc_rport_state(rport));
+	FC_RPORT_DBG(rport, "Port entered LOGO state from %s state\n",
+		     fc_rport_state(rport));
 
 	fc_rport_state_enter(rport, RPORT_ST_LOGO);
 
@@ -983,14 +968,13 @@ static void fc_rport_recv_plogi_req(struct fc_rport *rport,
 
 	fh = fc_frame_header_get(fp);
 
-	FC_DEBUG_RPORT("Received PLOGI request from port (%6x) "
-		       "while in state %s\n", ntoh24(fh->fh_s_id),
-		       fc_rport_state(rport));
+	FC_RPORT_DBG(rport, "Received PLOGI request while in state %s\n",
+		     fc_rport_state(rport));
 
 	sid = ntoh24(fh->fh_s_id);
 	pl = fc_frame_payload_get(fp, sizeof(*pl));
 	if (!pl) {
-		FC_DBG("incoming PLOGI from %x too short\n", sid);
+		FC_RPORT_DBG(rport, "Received PLOGI too short\n");
 		WARN_ON(1);
 		/* XXX TBD: send reject? */
 		fc_frame_free(fp);
@@ -1012,26 +996,26 @@ static void fc_rport_recv_plogi_req(struct fc_rport *rport,
 	 */
 	switch (rdata->rp_state) {
 	case RPORT_ST_INIT:
-		FC_DEBUG_RPORT("incoming PLOGI from %6x wwpn %llx state INIT "
-			       "- reject\n", sid, (unsigned long long)wwpn);
+		FC_RPORT_DBG(rport, "Received PLOGI, wwpn %llx state INIT "
+			     "- reject\n", (unsigned long long)wwpn);
 		reject = ELS_RJT_UNSUP;
 		break;
 	case RPORT_ST_PLOGI:
-		FC_DEBUG_RPORT("incoming PLOGI from %x in PLOGI state %d\n",
-			       sid, rdata->rp_state);
+		FC_RPORT_DBG(rport, "Received PLOGI in PLOGI state %d\n",
+			     rdata->rp_state);
 		if (wwpn < lport->wwpn)
 			reject = ELS_RJT_INPROG;
 		break;
 	case RPORT_ST_PRLI:
 	case RPORT_ST_READY:
-		FC_DEBUG_RPORT("incoming PLOGI from %x in logged-in state %d "
-			       "- ignored for now\n", sid, rdata->rp_state);
+		FC_RPORT_DBG(rport, "Received PLOGI in logged-in state %d "
+			     "- ignored for now\n", rdata->rp_state);
 		/* XXX TBD - should reset */
 		break;
 	case RPORT_ST_NONE:
 	default:
-		FC_DEBUG_RPORT("incoming PLOGI from %x in unexpected "
-			       "state %d\n", sid, rdata->rp_state);
+		FC_RPORT_DBG(rport, "Received PLOGI in unexpected "
+			     "state %d\n", rdata->rp_state);
 		fc_frame_free(fp);
 		return;
 		break;
@@ -1115,9 +1099,8 @@ static void fc_rport_recv_prli_req(struct fc_rport *rport,
 
 	fh = fc_frame_header_get(rx_fp);
 
-	FC_DEBUG_RPORT("Received PRLI request from port (%6x) "
-		       "while in state %s\n", ntoh24(fh->fh_s_id),
-		       fc_rport_state(rport));
+	FC_RPORT_DBG(rport, "Received PRLI request while in state %s\n",
+		     fc_rport_state(rport));
 
 	switch (rdata->rp_state) {
 	case RPORT_ST_PRLI:
@@ -1252,9 +1235,8 @@ static void fc_rport_recv_prlo_req(struct fc_rport *rport, struct fc_seq *sp,
 
 	fh = fc_frame_header_get(fp);
 
-	FC_DEBUG_RPORT("Received PRLO request from port (%6x) "
-		       "while in state %s\n", ntoh24(fh->fh_s_id),
-		       fc_rport_state(rport));
+	FC_RPORT_DBG(rport, "Received PRLO request while in state %s\n",
+		     fc_rport_state(rport));
 
 	if (rdata->rp_state == RPORT_ST_NONE) {
 		fc_frame_free(fp);
@@ -1286,9 +1268,8 @@ static void fc_rport_recv_logo_req(struct fc_rport *rport, struct fc_seq *sp,
 
 	fh = fc_frame_header_get(fp);
 
-	FC_DEBUG_RPORT("Received LOGO request from port (%6x) "
-		       "while in state %s\n", ntoh24(fh->fh_s_id),
-		       fc_rport_state(rport));
+	FC_RPORT_DBG(rport, "Received LOGO request while in state %s\n",
+		     fc_rport_state(rport));
 
 	if (rdata->rp_state == RPORT_ST_NONE) {
 		fc_frame_free(fp);
@@ -1308,7 +1289,6 @@ static void fc_rport_flush_queue(void)
 	flush_workqueue(rport_event_queue);
 }
 
-
 int fc_rport_init(struct fc_lport *lport)
 {
 	if (!lport->tt.rport_create)
diff --git a/include/scsi/fc_encode.h b/include/scsi/fc_encode.h
index 6300f556bce5..a0ff61c3e935 100644
--- a/include/scsi/fc_encode.h
+++ b/include/scsi/fc_encode.h
@@ -107,7 +107,6 @@ static inline int fc_ct_fill(struct fc_lport *lport, struct fc_frame *fp,
 		break;
 
 	default:
-		FC_DBG("Invalid op code %x \n", op);
 		return -EINVAL;
 	}
 	*r_ctl = FC_RCTL_DD_UNSOL_CTL;
@@ -298,7 +297,6 @@ static inline int fc_els_fill(struct fc_lport *lport, struct fc_rport *rport,
 		break;
 
 	default:
-		FC_DBG("Invalid op code %x \n", op);
 		return -EINVAL;
 	}
 
diff --git a/include/scsi/libfc.h b/include/scsi/libfc.h
index ebdd9f4cf070..b92584a8843a 100644
--- a/include/scsi/libfc.h
+++ b/include/scsi/libfc.h
@@ -34,17 +34,72 @@
 
 #include <scsi/fc_frame.h>
 
-#define LIBFC_DEBUG
-
-#ifdef LIBFC_DEBUG
-/* Log messages */
-#define FC_DBG(fmt, args...)						\
-	do {								\
-		printk(KERN_INFO "%s " fmt, __func__, ##args);		\
-	} while (0)
-#else
-#define FC_DBG(fmt, args...)
-#endif
+#define FC_LIBFC_LOGGING 0x01 /* General logging, not categorized */
+#define FC_LPORT_LOGGING 0x02 /* lport layer logging */
+#define FC_DISC_LOGGING  0x04 /* discovery layer logging */
+#define FC_RPORT_LOGGING 0x08 /* rport layer logging */
+#define FC_FCP_LOGGING   0x10 /* I/O path logging */
+#define FC_EM_LOGGING    0x20 /* Exchange Manager logging */
+#define FC_EXCH_LOGGING  0x40 /* Exchange/Sequence logging */
+#define FC_SCSI_LOGGING  0x80 /* SCSI logging (mostly error handling) */
+
+extern unsigned int fc_debug_logging;
+
+#define FC_CHECK_LOGGING(LEVEL, CMD)				\
+do {								\
+	if (unlikely(fc_debug_logging & LEVEL))			\
+		do {						\
+			CMD;					\
+		} while (0);					\
+} while (0);
+
+#define FC_LIBFC_DBG(fmt, args...)					\
+	FC_CHECK_LOGGING(FC_LIBFC_LOGGING,				\
+			 printk(KERN_INFO "libfc: " fmt, ##args);)
+
+#define FC_LPORT_DBG(lport, fmt, args...)				\
+	FC_CHECK_LOGGING(FC_LPORT_LOGGING,				\
+			 printk(KERN_INFO "lport: %6x: " fmt,		\
+				fc_host_port_id(lport->host), ##args);)
+
+#define FC_DISC_DBG(disc, fmt, args...)					\
+	FC_CHECK_LOGGING(FC_DISC_LOGGING,				\
+			 printk(KERN_INFO "disc: %6x: " fmt,		\
+				fc_host_port_id(disc->lport->host),	\
+				##args);)
+
+#define FC_RPORT_DBG(rport, fmt, args...)				\
+do {									\
+	struct fc_rport_libfc_priv *rdata = rport->dd_data;		\
+	struct fc_lport *lport = rdata->local_port;			\
+	FC_CHECK_LOGGING(FC_RPORT_LOGGING,				\
+			 printk(KERN_INFO "rport: %6x: %6x: " fmt,	\
+				fc_host_port_id(lport->host),		\
+				rport->port_id, ##args);)		\
+} while (0);
+
+#define FC_FCP_DBG(pkt, fmt, args...)					\
+	FC_CHECK_LOGGING(FC_FCP_LOGGING,				\
+			 printk(KERN_INFO "fcp: %6x: %6x: " fmt,	\
+				fc_host_port_id(pkt->lp->host),		\
+				pkt->rport->port_id, ##args);)
+
+#define FC_EM_DBG(em, fmt, args...)					\
+	FC_CHECK_LOGGING(FC_EM_LOGGING,					\
+			 printk(KERN_INFO "em: %6x: " fmt,		\
+				fc_host_port_id(em->lp->host),		\
+				##args);)
+
+#define FC_EXCH_DBG(exch, fmt, args...)					\
+	FC_CHECK_LOGGING(FC_EXCH_LOGGING,				\
+			 printk(KERN_INFO "exch: %6x: %4x: " fmt,	\
+				fc_host_port_id(exch->lp->host),	\
+				exch->xid, ##args);)
+
+#define FC_SCSI_DBG(lport, fmt, args...)				\
+	FC_CHECK_LOGGING(FC_SCSI_LOGGING,                               \
+			 printk(KERN_INFO "scsi: %6x: " fmt,		\
+				fc_host_port_id(lport->host), ##args);)
 
 /*
  * libfc error codes
-- 
cgit v1.2.3-59-g8ed1b


From f1d7fb7a8ab55357b6c7d44a53f644a043680ed1 Mon Sep 17 00:00:00 2001
From: Brian King <brking@linux.vnet.ibm.com>
Date: Thu, 18 Jun 2009 09:06:52 -0500
Subject: ibmvfc: Process async events before command responses

Since async events could indicate changes to link status, or
events which could affect decisions made during discovery, we should
process async events prior to command completion responses.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/ibmvscsi/ibmvfc.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/scsi/ibmvscsi/ibmvfc.c b/drivers/scsi/ibmvscsi/ibmvfc.c
index b4b805e8d7db..497a4cc4d9ef 100644
--- a/drivers/scsi/ibmvscsi/ibmvfc.c
+++ b/drivers/scsi/ibmvscsi/ibmvfc.c
@@ -2783,27 +2783,27 @@ static void ibmvfc_tasklet(void *data)
 
 	spin_lock_irqsave(vhost->host->host_lock, flags);
 	while (!done) {
-		/* Pull all the valid messages off the CRQ */
-		while ((crq = ibmvfc_next_crq(vhost)) != NULL) {
-			ibmvfc_handle_crq(crq, vhost);
-			crq->valid = 0;
-		}
-
 		/* Pull all the valid messages off the async CRQ */
 		while ((async = ibmvfc_next_async_crq(vhost)) != NULL) {
 			ibmvfc_handle_async(async, vhost);
 			async->valid = 0;
 		}
 
-		vio_enable_interrupts(vdev);
-		if ((crq = ibmvfc_next_crq(vhost)) != NULL) {
-			vio_disable_interrupts(vdev);
+		/* Pull all the valid messages off the CRQ */
+		while ((crq = ibmvfc_next_crq(vhost)) != NULL) {
 			ibmvfc_handle_crq(crq, vhost);
 			crq->valid = 0;
-		} else if ((async = ibmvfc_next_async_crq(vhost)) != NULL) {
+		}
+
+		vio_enable_interrupts(vdev);
+		if ((async = ibmvfc_next_async_crq(vhost)) != NULL) {
 			vio_disable_interrupts(vdev);
 			ibmvfc_handle_async(async, vhost);
 			async->valid = 0;
+		} else if ((crq = ibmvfc_next_crq(vhost)) != NULL) {
+			vio_disable_interrupts(vdev);
+			ibmvfc_handle_crq(crq, vhost);
+			crq->valid = 0;
 		} else
 			done = 1;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 017b2ae33c0fc7d70320cc7f1cce0efb6ce8d929 Mon Sep 17 00:00:00 2001
From: Brian King <brking@linux.vnet.ibm.com>
Date: Thu, 18 Jun 2009 09:06:55 -0500
Subject: ibmvfc: Fix endless PRLI loop in discovery

Fixes a problem seen where sending a PRLI to a target
resulted in it sending a LOGO. This caused the ibmvfc
driver to go back through discovery again, which caused
another PRLI attempt, which caused another LOGO. Fix this
behavior by ignoring LOGO if we haven't even logged into
the target yet.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/ibmvscsi/ibmvfc.c | 16 ++++++++++++----
 drivers/scsi/ibmvscsi/ibmvfc.h |  1 +
 2 files changed, 13 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/ibmvscsi/ibmvfc.c b/drivers/scsi/ibmvscsi/ibmvfc.c
index 497a4cc4d9ef..166d96450a0e 100644
--- a/drivers/scsi/ibmvscsi/ibmvfc.c
+++ b/drivers/scsi/ibmvscsi/ibmvfc.c
@@ -2254,10 +2254,13 @@ static void ibmvfc_handle_async(struct ibmvfc_async_crq *crq,
 				continue;
 			if (crq->node_name && tgt->ids.node_name != crq->node_name)
 				continue;
-			ibmvfc_set_tgt_action(tgt, IBMVFC_TGT_ACTION_DEL_RPORT);
+			if (tgt->need_login && crq->event == IBMVFC_AE_ELS_LOGO)
+				tgt->logo_rcvd = 1;
+			if (!tgt->need_login || crq->event == IBMVFC_AE_ELS_PLOGI) {
+				ibmvfc_set_tgt_action(tgt, IBMVFC_TGT_ACTION_DEL_RPORT);
+				ibmvfc_reinit_host(vhost);
+			}
 		}
-
-		ibmvfc_reinit_host(vhost);
 		break;
 	case IBMVFC_AE_LINK_DOWN:
 	case IBMVFC_AE_ADAPTER_FAILED:
@@ -2927,7 +2930,11 @@ static void ibmvfc_tgt_prli_done(struct ibmvfc_event *evt)
 		break;
 	case IBMVFC_MAD_FAILED:
 	default:
-		if (ibmvfc_retry_cmd(rsp->status, rsp->error))
+		if ((rsp->status & IBMVFC_VIOS_FAILURE) && rsp->error == IBMVFC_PLOGI_REQUIRED)
+			level += ibmvfc_retry_tgt_init(tgt, ibmvfc_tgt_send_plogi);
+		else if (tgt->logo_rcvd)
+			level += ibmvfc_retry_tgt_init(tgt, ibmvfc_tgt_send_plogi);
+		else if (ibmvfc_retry_cmd(rsp->status, rsp->error))
 			level += ibmvfc_retry_tgt_init(tgt, ibmvfc_tgt_send_prli);
 		else
 			ibmvfc_set_tgt_action(tgt, IBMVFC_TGT_ACTION_DEL_RPORT);
@@ -3054,6 +3061,7 @@ static void ibmvfc_tgt_send_plogi(struct ibmvfc_target *tgt)
 		return;
 
 	kref_get(&tgt->kref);
+	tgt->logo_rcvd = 0;
 	evt = ibmvfc_get_event(vhost);
 	vhost->discovery_threads++;
 	ibmvfc_set_tgt_action(tgt, IBMVFC_TGT_ACTION_INIT_WAIT);
diff --git a/drivers/scsi/ibmvscsi/ibmvfc.h b/drivers/scsi/ibmvscsi/ibmvfc.h
index c2668d7d67f5..007fa1c9ef14 100644
--- a/drivers/scsi/ibmvscsi/ibmvfc.h
+++ b/drivers/scsi/ibmvscsi/ibmvfc.h
@@ -605,6 +605,7 @@ struct ibmvfc_target {
 	int need_login;
 	int add_rport;
 	int init_retries;
+	int logo_rcvd;
 	u32 cancel_key;
 	struct ibmvfc_service_parms service_parms;
 	struct ibmvfc_service_parms service_parms_change;
-- 
cgit v1.2.3-59-g8ed1b


From 5e2fb917920c62c5ad260962471aeb578b52ac40 Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Thu, 18 Jun 2009 21:03:23 +0200
Subject: explain the hidden scsi_wait_scan Kconfig variable

People keep sending patches to expose CONFIG_SCSI_WAIT_SCAN as a tunable
item.  These patches aren't accepted upstream, so let's stop the ongoing
irritation of people due to the unconditionally installed module and its
Kconfig symbol.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/Kconfig | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index 6a19ed9a1194..9c23122f755f 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -258,10 +258,21 @@ config SCSI_SCAN_ASYNC
 	  or async on the kernel's command line.
 
 config SCSI_WAIT_SCAN
-	tristate
+	tristate  # No prompt here, this is an invisible symbol.
 	default m
 	depends on SCSI
 	depends on MODULES
+# scsi_wait_scan is a loadable module which waits until all the async scans are
+# complete.  The idea is to use it in initrd/ initramfs scripts.  You modprobe
+# it after all the modprobes of the root SCSI drivers and it will wait until
+# they have all finished scanning their buses before allowing the boot to
+# proceed.  (This method is not applicable if targets boot independently in
+# parallel with the initiator, or with transports with non-deterministic target
+# discovery schemes, or if a transport driver does not support scsi_wait_scan.)
+#
+# This symbol is not exposed as a prompt because little is to be gained by
+# disabling it, whereas people who accidentally switch it off may wonder why
+# their mkinitrd gets into trouble.
 
 menu "SCSI Transports"
 	depends on SCSI
-- 
cgit v1.2.3-59-g8ed1b


From 75be63bcf73ebdd1fdc1d49f6bf2d1326a1ba7de Mon Sep 17 00:00:00 2001
From: John Stoffel <john@stoffel.org>
Date: Fri, 19 Jun 2009 16:08:58 -0400
Subject: sym53c8xx: ratelimit parity errors

This makes a huge difference when you have a serial console on bootup to limit
these messages to a sane number.

Signed-off-by: John Stoffel <john@stoffel.org>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/sym53c8xx_2/sym_hipd.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/sym53c8xx_2/sym_hipd.c b/drivers/scsi/sym53c8xx_2/sym_hipd.c
index 69ad4945c936..297deb817a5d 100644
--- a/drivers/scsi/sym53c8xx_2/sym_hipd.c
+++ b/drivers/scsi/sym53c8xx_2/sym_hipd.c
@@ -2321,8 +2321,9 @@ static void sym_int_par (struct sym_hcb *np, u_short sist)
 	int phase	= cmd & 7;
 	struct sym_ccb *cp	= sym_ccb_from_dsa(np, dsa);
 
-	printf("%s: SCSI parity error detected: SCR1=%d DBC=%x SBCL=%x\n",
-		sym_name(np), hsts, dbc, sbcl);
+	if (printk_ratelimit())
+		printf("%s: SCSI parity error detected: SCR1=%d DBC=%x SBCL=%x\n",
+			sym_name(np), hsts, dbc, sbcl);
 
 	/*
 	 *  Check that the chip is connected to the SCSI BUS.
-- 
cgit v1.2.3-59-g8ed1b


From b5c6f77680f4ff1775838fcedfdd6026bf5ad777 Mon Sep 17 00:00:00 2001
From: Giridhar Malavali <giridhar.malavali@qlogic.com>
Date: Fri, 19 Jun 2009 16:26:53 -0700
Subject: fc_transport: The softirq_done function registration for BSG request

Registered the softirq_done function, since this is requried iby an request
using block level request timeout functionality. This function will be called
by the block layer as part of time out clean process to release the BSG
request.

Moved some of the BSG request completion activities to softirq_done routine to
take care of both normal and timout completions.

Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Acked-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_transport_fc.c | 37 +++++++++++++++++++------------------
 1 file changed, 19 insertions(+), 18 deletions(-)

diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index 3f64d93b6c8b..453d9e658eb6 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -3397,7 +3397,6 @@ fc_destroy_bsgjob(struct fc_bsg_job *job)
 	kfree(job);
 }
 
-
 /**
  * fc_bsg_jobdone - completion routine for bsg requests that the LLD has
  *                  completed
@@ -3408,15 +3407,10 @@ fc_bsg_jobdone(struct fc_bsg_job *job)
 {
 	struct request *req = job->req;
 	struct request *rsp = req->next_rq;
-	unsigned long flags;
 	int err;
 
-	spin_lock_irqsave(&job->job_lock, flags);
-	job->state_flags |= FC_RQST_STATE_DONE;
-	job->ref_cnt--;
-	spin_unlock_irqrestore(&job->job_lock, flags);
-
 	err = job->req->errors = job->reply->result;
+
 	if (err < 0)
 		/* we're only returning the result field in the reply */
 		job->req->sense_len = sizeof(uint32_t);
@@ -3433,13 +3427,27 @@ fc_bsg_jobdone(struct fc_bsg_job *job)
 		rsp->resid_len -= min(job->reply->reply_payload_rcv_len,
 				      rsp->resid_len);
 	}
+	blk_complete_request(req);
+}
+
+/**
+ * fc_bsg_softirq_done - softirq done routine for destroying the bsg requests
+ * @req:        BSG request that holds the job to be destroyed
+ */
+static void fc_bsg_softirq_done(struct request *rq)
+{
+	struct fc_bsg_job *job = rq->special;
+	unsigned long flags;
 
-	blk_end_request_all(req, err);
+	spin_lock_irqsave(&job->job_lock, flags);
+	job->state_flags |= FC_RQST_STATE_DONE;
+	job->ref_cnt--;
+	spin_unlock_irqrestore(&job->job_lock, flags);
 
+	blk_end_request_all(rq, rq->errors);
 	fc_destroy_bsgjob(job);
 }
 
-
 /**
  * fc_bsg_job_timeout - handler for when a bsg request timesout
  * @req:	request that timed out
@@ -3471,19 +3479,10 @@ fc_bsg_job_timeout(struct request *req)
 				"abort failed with status %d\n", err);
 	}
 
-	if (!done) {
-		spin_lock_irqsave(&job->job_lock, flags);
-		job->ref_cnt--;
-		spin_unlock_irqrestore(&job->job_lock, flags);
-		fc_destroy_bsgjob(job);
-	}
-
 	/* the blk_end_sync_io() doesn't check the error */
 	return BLK_EH_HANDLED;
 }
 
-
-
 static int
 fc_bsg_map_buffer(struct fc_bsg_buffer *buf, struct request *req)
 {
@@ -3879,6 +3878,7 @@ fc_bsg_hostadd(struct Scsi_Host *shost, struct fc_host_attrs *fc_host)
 
 	q->queuedata = shost;
 	queue_flag_set_unlocked(QUEUE_FLAG_BIDI, q);
+	blk_queue_softirq_done(q, fc_bsg_softirq_done);
 	blk_queue_rq_timed_out(q, fc_bsg_job_timeout);
 	blk_queue_rq_timeout(q, FC_DEFAULT_BSG_TIMEOUT);
 
@@ -3924,6 +3924,7 @@ fc_bsg_rportadd(struct Scsi_Host *shost, struct fc_rport *rport)
 
 	q->queuedata = rport;
 	queue_flag_set_unlocked(QUEUE_FLAG_BIDI, q);
+	blk_queue_softirq_done(q, fc_bsg_softirq_done);
 	blk_queue_rq_timed_out(q, fc_bsg_job_timeout);
 	blk_queue_rq_timeout(q, BLK_DEFAULT_SG_TIMEOUT);
 
-- 
cgit v1.2.3-59-g8ed1b


From 47e7e89ed029780adf2cc0cf506fcd4c2d5ca1e2 Mon Sep 17 00:00:00 2001
From: Giridhar Malavali <giridhar.malavali@qlogic.com>
Date: Fri, 19 Jun 2009 16:26:54 -0700
Subject: fc_transport: Selective return value from BSG timeout function

The return value from BSG timout function should be based on the state of the
BSG job. This helps block layer to take selective actions to clean up BSG job.

Signed-off-by: Giridhar Malavali <giridhar.malavali@qlogic.com>
Acked-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_transport_fc.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index 453d9e658eb6..140c50c8a5d2 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -3480,7 +3480,10 @@ fc_bsg_job_timeout(struct request *req)
 	}
 
 	/* the blk_end_sync_io() doesn't check the error */
-	return BLK_EH_HANDLED;
+	if (done)
+		return BLK_EH_NOT_HANDLED;
+	else
+		return BLK_EH_HANDLED;
 }
 
 static int
-- 
cgit v1.2.3-59-g8ed1b


From 24add1c4326ce5ca22e7af8b84bb113cd48efac9 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 20 Jun 2009 13:29:21 +0530
Subject: scsi_transport_iscsi: return -EOVERFLOW for Too many iscsi targets
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

setting err as -EOVERFLOW for Too many iscsi targets.

Also fixes a spurious compiler warning for gcc 4.3.3 and gcc 4.4 :

  CC      drivers/scsi/scsi_transport_iscsi.o
drivers/scsi/scsi_transport_iscsi.c: In function ‘iscsi_add_session’:
drivers/scsi/scsi_transport_iscsi.c:678: warning: ‘err’ may be used uninitialized in this function

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Acked-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_transport_iscsi.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index f3e664628d7a..783e33c65eb7 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -692,6 +692,7 @@ int iscsi_add_session(struct iscsi_cls_session *session, unsigned int target_id)
 						 "Too many iscsi targets. Max "
 						 "number of targets is %d.\n",
 						 ISCSI_MAX_TARGET - 1);
+			err = -EOVERFLOW;
 			goto release_host;
 		}
 	}
-- 
cgit v1.2.3-59-g8ed1b


From b391277a56b9eaaff4474339c703e574ed7fab5b Mon Sep 17 00:00:00 2001
From: Hannes Reinecke <hare@suse.de>
Date: Thu, 18 Jun 2009 09:57:18 +0200
Subject: sd, sr: fix Driver 'sd' needs updating message

If a SCSI ULD driver sets blk_queue_prep_rq(), it should clean it
up itself on remove(), and not from the bus callbacks. This
removes the need to hook into bus->remove(), which should not
be used at the same time as driver->remove().

[jejb: fix sdkp initialisation problem due to mismerge]
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_lib.c    |  1 +
 drivers/scsi/scsi_priv.h   |  1 -
 drivers/scsi/scsi_sysfs.c  | 17 -----------------
 drivers/scsi/sd.c          |  1 +
 drivers/scsi/sr.c          |  1 +
 include/scsi/scsi_driver.h |  1 +
 6 files changed, 4 insertions(+), 18 deletions(-)

diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 30f3275e119e..f3c40898fc7d 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1207,6 +1207,7 @@ int scsi_prep_fn(struct request_queue *q, struct request *req)
 		ret = scsi_setup_blk_pc_cmnd(sdev, req);
 	return scsi_prep_return(q, req, ret);
 }
+EXPORT_SYMBOL(scsi_prep_fn);
 
 /*
  * scsi_dev_queue_ready: if we can send requests to sdev, return 1 else
diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h
index 00264aab8c8a..021e503c8c44 100644
--- a/drivers/scsi/scsi_priv.h
+++ b/drivers/scsi/scsi_priv.h
@@ -87,7 +87,6 @@ extern int scsi_init_queue(void);
 extern void scsi_exit_queue(void);
 struct request_queue;
 struct request;
-extern int scsi_prep_fn(struct request_queue *, struct request *);
 extern struct kmem_cache *scsi_sdb_cache;
 
 /* scsi_proc.c */
diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c
index fa4711d12744..91482f2dcc50 100644
--- a/drivers/scsi/scsi_sysfs.c
+++ b/drivers/scsi/scsi_sysfs.c
@@ -420,29 +420,12 @@ static int scsi_bus_resume(struct device * dev)
 	return err;
 }
 
-static int scsi_bus_remove(struct device *dev)
-{
-	struct device_driver *drv = dev->driver;
-	struct scsi_device *sdev = to_scsi_device(dev);
-	int err = 0;
-
-	/* reset the prep_fn back to the default since the
-	 * driver may have altered it and it's being removed */
-	blk_queue_prep_rq(sdev->request_queue, scsi_prep_fn);
-
-	if (drv && drv->remove)
-		err = drv->remove(dev);
-
-	return 0;
-}
-
 struct bus_type scsi_bus_type = {
         .name		= "scsi",
         .match		= scsi_bus_match,
 	.uevent		= scsi_bus_uevent,
 	.suspend	= scsi_bus_suspend,
 	.resume		= scsi_bus_resume,
-	.remove		= scsi_bus_remove,
 };
 EXPORT_SYMBOL_GPL(scsi_bus_type);
 
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index e4ef11af18a2..5616cd780ff3 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -2123,6 +2123,7 @@ static int sd_remove(struct device *dev)
 
 	async_synchronize_full();
 	sdkp = dev_get_drvdata(dev);
+	blk_queue_prep_rq(sdkp->device->request_queue, scsi_prep_fn);
 	device_del(&sdkp->dev);
 	del_gendisk(sdkp->disk);
 	sd_shutdown(dev);
diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
index cd350dfc1216..cce0fe4c8a3b 100644
--- a/drivers/scsi/sr.c
+++ b/drivers/scsi/sr.c
@@ -881,6 +881,7 @@ static int sr_remove(struct device *dev)
 {
 	struct scsi_cd *cd = dev_get_drvdata(dev);
 
+	blk_queue_prep_rq(cd->device->request_queue, scsi_prep_fn);
 	del_gendisk(cd->disk);
 
 	mutex_lock(&sr_ref_mutex);
diff --git a/include/scsi/scsi_driver.h b/include/scsi/scsi_driver.h
index 1f5ca7f62116..9fd6702f02e2 100644
--- a/include/scsi/scsi_driver.h
+++ b/include/scsi/scsi_driver.h
@@ -32,5 +32,6 @@ int scsi_setup_blk_pc_cmnd(struct scsi_device *sdev, struct request *req);
 int scsi_setup_fs_cmnd(struct scsi_device *sdev, struct request *req);
 int scsi_prep_state_check(struct scsi_device *sdev, struct request *req);
 int scsi_prep_return(struct request_queue *q, struct request *req, int ret);
+int scsi_prep_fn(struct request_queue *, struct request *);
 
 #endif /* _SCSI_SCSI_DRIVER_H */
-- 
cgit v1.2.3-59-g8ed1b


From 3c559ea8fd003962d9a28c64b2dd5c6d83ca6edb Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Sun, 21 Jun 2009 12:11:43 -0500
Subject: [SCSI] scsi_transport_fc: replace BUS_ID_SIZE by fixed count

BUS_ID_SIZE is being removed from the kernel.

Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_transport_fc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index 140c50c8a5d2..2eee9e6e4fe8 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -3861,7 +3861,7 @@ fc_bsg_hostadd(struct Scsi_Host *shost, struct fc_host_attrs *fc_host)
 	struct fc_internal *i = to_fc_internal(shost->transportt);
 	struct request_queue *q;
 	int err;
-	char bsg_name[BUS_ID_SIZE]; /*20*/
+	char bsg_name[20];
 
 	fc_host->rqst_q = NULL;
 
-- 
cgit v1.2.3-59-g8ed1b


From 920d44ed965b1fb842ae65fad1a51fc7570e1181 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Sun, 21 Jun 2009 16:11:33 -0700
Subject: ide: Take over as maintainer.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 1d4704300c1d..487aaea9d778 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2802,10 +2802,10 @@ S:	Supported
 F:	drivers/scsi/ips.*
 
 IDE SUBSYSTEM
-P:	Bartlomiej Zolnierkiewicz
-M:	bzolnier@gmail.com
+P:	David S. Miller
+M:	davem@davemloft.net
 L:	linux-ide@vger.kernel.org
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/bart/ide-2.6.git
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/davem/ide-2.6.git
 S:	Maintained
 F:	Documentation/ide/
 F:	drivers/ide/
-- 
cgit v1.2.3-59-g8ed1b


From ff0f242626313f3544254cb882039794b7b70e4b Mon Sep 17 00:00:00 2001
From: Tony Lindgren <tony@atomide.com>
Date: Wed, 17 Jun 2009 03:20:21 -0700
Subject: i2c-omap: Fix build breaking typo cpu_is_omap_2430

Hi Ben,

Can you please queue this fix?

Thanks,

Tony

>From ffe2b2cdf6283770b70a197e3748c6b40a1006be Mon Sep 17 00:00:00 2001
From: Tony Lindgren <tony@atomide.com>
Date: Wed, 17 Jun 2009 13:14:23 +0300
Subject: [PATCH] i2c-omap: Fix build breaking typo in cpu_is_omap_2430

Commit 84bf2c86 introduced a typo, it should be cpu_is_omap2430
instead. The typo was probably caused by a mismerge.

Without this patch all omaps fail to build with:
error: implicit declaration of function 'cpu_is_omap_2430'

Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 drivers/i2c/busses/i2c-omap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/i2c/busses/i2c-omap.c b/drivers/i2c/busses/i2c-omap.c
index b606db85525d..ad8d2010c921 100644
--- a/drivers/i2c/busses/i2c-omap.c
+++ b/drivers/i2c/busses/i2c-omap.c
@@ -339,7 +339,7 @@ static int omap_i2c_init(struct omap_i2c_dev *dev)
 		 * to get longer filter period for better noise suppression.
 		 * The filter is iclk (fclk for HS) period.
 		 */
-		if (dev->speed > 400 || cpu_is_omap_2430())
+		if (dev->speed > 400 || cpu_is_omap2430())
 			internal_clk = 19200;
 		else if (dev->speed > 100)
 			internal_clk = 9600;
-- 
cgit v1.2.3-59-g8ed1b


From dc1972d02747d2170fb1d78d114801f5ecb27506 Mon Sep 17 00:00:00 2001
From: Michael Trimarchi <michael@evidence.eu.com>
Date: Fri, 19 Jun 2009 14:50:02 +0200
Subject: i2c: Fix stuck transaction on cpm-i2c driver

When a process tries to read/write a disconnected i2c device, it receives a signal (e.g. ctrl-c) and the kernel gets stuck.

BUG: soft lockup - CPU#0 stuck for 61s! [I2CEEpromTest:392]
NIP: c01628f8 LR: c01628f0 CTR: c00177cc
REGS: c39abd70 TRAP: 0901   Not tainted  (2.6.25.7-alcore)
MSR: 00009032 <EE,ME,IR,DR>  CR: 42042048  XER: 20000000
TASK = c3889bd0[392] 'I2CEEpromTest' THREAD: c39aa000
GPR00: 00009000 c39abe20 c3889bd0 c39075c8 c39abe28 00000001 00000000 00000001
GPR08: c3889bd0 c39075c8 00009032 c39abe34 00002437
NIP [c01628f8] cpm_i2c_xfer+0x5fc/0x6d0
LR [c01628f0] cpm_i2c_xfer+0x5f4/0x6d0
Call Trace:
[c39abe20] [c0162924] cpm_i2c_xfer+0x628/0x6d0 (unreliable)
[c39abe90] [c015f6a0] i2c_transfer+0x88/0xb4
[c39abeb0] [c0160164] i2c_master_recv+0x48/0x6c
[c39abed0] [c01618dc] i2cdev_read+0x50/0xe4
[c39abef0] [c0068b24] vfs_read+0xc4/0x108
[c39abf10] [c0068f4c] sys_read+0x4c/0x90
[c39abf40] [c000d348] ret_from_syscall+0x0/0x38
Instruction dump:
3bc00064 92610010 3bf201c8 92810014 3b61

This happen because though the wait_event_interruptible_timeout takes the
signals into account, the driver does not handle them.
We propose to change the wait_event_interruptible_timeout with
wait_event_timeout, leaving the signals to be handled in other points
on the upper layers.

Signed-off-by: Bruno Morelli <bruno@evidence.eu.com>
Signed-off-by: Michael Trimarchi <michael@evidence.eu.com>
Acked-by: Jochen Friedrich <jochen@scram.de>
[ben-linux@fluff.org: fix title for patch]
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 drivers/i2c/busses/i2c-cpm.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/i2c/busses/i2c-cpm.c b/drivers/i2c/busses/i2c-cpm.c
index b5db8b883615..9c2e10082b79 100644
--- a/drivers/i2c/busses/i2c-cpm.c
+++ b/drivers/i2c/busses/i2c-cpm.c
@@ -140,7 +140,7 @@ static irqreturn_t cpm_i2c_interrupt(int irq, void *dev_id)
 
 	dev_dbg(&adap->dev, "Interrupt: %x\n", i);
 
-	wake_up_interruptible(&cpm->i2c_wait);
+	wake_up(&cpm->i2c_wait);
 
 	return i ? IRQ_HANDLED : IRQ_NONE;
 }
@@ -364,12 +364,12 @@ static int cpm_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num)
 		dev_dbg(&adap->dev, "test ready.\n");
 		pmsg = &msgs[tptr];
 		if (pmsg->flags & I2C_M_RD)
-			ret = wait_event_interruptible_timeout(cpm->i2c_wait,
+			ret = wait_event_timeout(cpm->i2c_wait,
 				(in_be16(&tbdf[tptr].cbd_sc) & BD_SC_NAK) ||
 				!(in_be16(&rbdf[rptr].cbd_sc) & BD_SC_EMPTY),
 				1 * HZ);
 		else
-			ret = wait_event_interruptible_timeout(cpm->i2c_wait,
+			ret = wait_event_timeout(cpm->i2c_wait,
 				!(in_be16(&tbdf[tptr].cbd_sc) & BD_SC_READY),
 				1 * HZ);
 		if (ret == 0) {
-- 
cgit v1.2.3-59-g8ed1b


From 7e23091347664bf357ca651545c93e99fafc7b40 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Sat, 20 Jun 2009 22:15:31 +0000
Subject: mlx4_en: Counting all the dropped packets on the TX side

Reporting the counter's value through 'ethtool -S'

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mlx4/en_tx.c | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/drivers/net/mlx4/en_tx.c b/drivers/net/mlx4/en_tx.c
index 5dc7466ad035..c0177c364bbf 100644
--- a/drivers/net/mlx4/en_tx.c
+++ b/drivers/net/mlx4/en_tx.c
@@ -515,14 +515,12 @@ static int get_real_size(struct sk_buff *skb, struct net_device *dev,
 			else {
 				if (netif_msg_tx_err(priv))
 					en_warn(priv, "Non-linear headers\n");
-				dev_kfree_skb_any(skb);
 				return 0;
 			}
 		}
 		if (unlikely(*lso_header_size > MAX_LSO_HDR_SIZE)) {
 			if (netif_msg_tx_err(priv))
 				en_warn(priv, "LSO header size too big\n");
-			dev_kfree_skb_any(skb);
 			return 0;
 		}
 	} else {
@@ -622,7 +620,7 @@ int mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 	}
 	real_size = get_real_size(skb, dev, &lso_header_size);
 	if (unlikely(!real_size))
-		return NETDEV_TX_OK;
+		goto tx_drop;
 
 	/* Allign descriptor to TXBB size */
 	desc_size = ALIGN(real_size, TXBB_SIZE);
@@ -630,8 +628,7 @@ int mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (unlikely(nr_txbb > MAX_DESC_TXBBS)) {
 		if (netif_msg_tx_err(priv))
 			en_warn(priv, "Oversized header or SG list\n");
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
+		goto tx_drop;
 	}
 
 	tx_ind = skb->queue_mapping;
@@ -657,8 +654,7 @@ int mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (unlikely(!priv->port_up)) {
 		if (netif_msg_tx_err(priv))
 			en_warn(priv, "xmit: port down!\n");
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
+		goto tx_drop;
 	}
 
 	/* Track current inflight packets for performance analysis */
@@ -785,5 +781,10 @@ int mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 	mlx4_en_xmit_poll(priv, tx_ind);
 
 	return 0;
+
+tx_drop:
+	dev_kfree_skb_any(skb);
+	priv->stats.tx_dropped++;
+	return NETDEV_TX_OK;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From d4ddbaa6a9a09c019fc1a7fed5a0fa403ac437b9 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Sat, 20 Jun 2009 22:15:39 +0000
Subject: mlx4_en: Removed redundant skb->len check

We don't need this check in the transmit function

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mlx4/en_tx.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/mlx4/en_tx.c b/drivers/net/mlx4/en_tx.c
index c0177c364bbf..e63132361a94 100644
--- a/drivers/net/mlx4/en_tx.c
+++ b/drivers/net/mlx4/en_tx.c
@@ -614,10 +614,6 @@ int mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 	int lso_header_size;
 	void *fragptr;
 
-	if (unlikely(!skb->len)) {
-		dev_kfree_skb_any(skb);
-		return NETDEV_TX_OK;
-	}
 	real_size = get_real_size(skb, dev, &lso_header_size);
 	if (unlikely(!real_size))
 		goto tx_drop;
-- 
cgit v1.2.3-59-g8ed1b


From a11faac79fdbf771ed1ab310f6ef44b389423fe7 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Sat, 20 Jun 2009 22:15:46 +0000
Subject: mlx4_en: using stop/start_all_queues

After we moved to be a multi queue device, need to stop/start
all of our transmit queues.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mlx4/en_netdev.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index e02bafdd3682..20a34cb6392e 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -668,7 +668,7 @@ int mlx4_en_start_port(struct net_device *dev)
 	queue_work(mdev->workqueue, &priv->mcast_task);
 
 	priv->port_up = true;
-	netif_start_queue(dev);
+	netif_tx_start_all_queues(dev);
 	return 0;
 
 mac_err:
@@ -700,7 +700,7 @@ void mlx4_en_stop_port(struct net_device *dev)
 		en_dbg(DRV, priv, "stop port called while port already down\n");
 		return;
 	}
-	netif_stop_queue(dev);
+	netif_tx_stop_all_queues(dev);
 
 	/* Synchronize with tx routine */
 	netif_tx_lock_bh(dev);
-- 
cgit v1.2.3-59-g8ed1b


From 3c05f5ef7c09291e51ae327e854bf43cb8e55a55 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Sat, 20 Jun 2009 22:15:52 +0000
Subject: mlx4_en: Cancel port_up check in transmit function

When closing the port, we stop all transmit queues under the transmit
lock. It ensures that we will not attempt to transmit new packets after
the physical port was closed.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mlx4/en_netdev.c | 4 ++--
 drivers/net/mlx4/en_tx.c     | 7 -------
 2 files changed, 2 insertions(+), 9 deletions(-)

diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index 20a34cb6392e..03a557cc3b7a 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -700,14 +700,14 @@ void mlx4_en_stop_port(struct net_device *dev)
 		en_dbg(DRV, priv, "stop port called while port already down\n");
 		return;
 	}
-	netif_tx_stop_all_queues(dev);
 
 	/* Synchronize with tx routine */
 	netif_tx_lock_bh(dev);
-	priv->port_up = false;
+	netif_tx_stop_all_queues(dev);
 	netif_tx_unlock_bh(dev);
 
 	/* close port*/
+	priv->port_up = false;
 	mlx4_CLOSE_PORT(mdev->dev, priv->port);
 
 	/* Unregister Mac address for the port */
diff --git a/drivers/net/mlx4/en_tx.c b/drivers/net/mlx4/en_tx.c
index e63132361a94..99a6a36dc27b 100644
--- a/drivers/net/mlx4/en_tx.c
+++ b/drivers/net/mlx4/en_tx.c
@@ -646,13 +646,6 @@ int mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
 		return NETDEV_TX_BUSY;
 	}
 
-	/* Now that we know what Tx ring to use */
-	if (unlikely(!priv->port_up)) {
-		if (netif_msg_tx_err(priv))
-			en_warn(priv, "xmit: port down!\n");
-		goto tx_drop;
-	}
-
 	/* Track current inflight packets for performance analysis */
 	AVG_PERF_COUNTER(priv->pstats.inflight_avg,
 			 (u32) (ring->prod - ring->cons - 1));
-- 
cgit v1.2.3-59-g8ed1b


From 7237b400554c9bb5ba0091b5e39f4620f3dd5637 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Sat, 20 Jun 2009 22:16:02 +0000
Subject: mlx4_en: Removed redundant check on lso header size

This check that verifies that the LSO header along with control
segment and first data segment do not cross 128 bytes is no longer
required.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mlx4/en_tx.c   | 5 -----
 drivers/net/mlx4/mlx4_en.h | 1 -
 2 files changed, 6 deletions(-)

diff --git a/drivers/net/mlx4/en_tx.c b/drivers/net/mlx4/en_tx.c
index 99a6a36dc27b..08c43f2ae72b 100644
--- a/drivers/net/mlx4/en_tx.c
+++ b/drivers/net/mlx4/en_tx.c
@@ -518,11 +518,6 @@ static int get_real_size(struct sk_buff *skb, struct net_device *dev,
 				return 0;
 			}
 		}
-		if (unlikely(*lso_header_size > MAX_LSO_HDR_SIZE)) {
-			if (netif_msg_tx_err(priv))
-				en_warn(priv, "LSO header size too big\n");
-			return 0;
-		}
 	} else {
 		*lso_header_size = 0;
 		if (!is_inline(skb, NULL))
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index d43a9e4c2aea..ad861db66f19 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -99,7 +99,6 @@
 #define RSS_FACTOR		2
 #define TXBB_SIZE		64
 #define HEADROOM		(2048 / TXBB_SIZE + 1)
-#define MAX_LSO_HDR_SIZE	92
 #define STAMP_STRIDE		64
 #define STAMP_DWORDS		(STAMP_STRIDE / 4)
 #define STAMP_SHIFT		31
-- 
cgit v1.2.3-59-g8ed1b


From 0314db69d7564859890ff75e3f71cb4079b29869 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Sat, 20 Jun 2009 22:16:10 +0000
Subject: mlx4_en: Remove redundant refill code on RX

Our RX rings are always full, there is no need to check whether
we need to fill them or not. If we fail to allocate a new socket
buffer, the incoming packet is dropped an the ring remains full.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mlx4/en_netdev.c |  2 -
 drivers/net/mlx4/en_rx.c     | 96 --------------------------------------------
 drivers/net/mlx4/mlx4_en.h   |  4 --
 3 files changed, 102 deletions(-)

diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index 03a557cc3b7a..93f4abd990a9 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -881,7 +881,6 @@ void mlx4_en_destroy_netdev(struct net_device *dev)
 		mlx4_free_hwq_res(mdev->dev, &priv->res, MLX4_EN_PAGE_SIZE);
 
 	cancel_delayed_work(&priv->stats_task);
-	cancel_delayed_work(&priv->refill_task);
 	/* flush any pending task for this netdev */
 	flush_workqueue(mdev->workqueue);
 
@@ -986,7 +985,6 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
 	spin_lock_init(&priv->stats_lock);
 	INIT_WORK(&priv->mcast_task, mlx4_en_do_set_multicast);
 	INIT_WORK(&priv->mac_task, mlx4_en_do_set_mac);
-	INIT_DELAYED_WORK(&priv->refill_task, mlx4_en_rx_refill);
 	INIT_WORK(&priv->watchdog_task, mlx4_en_restart);
 	INIT_WORK(&priv->linkstate_task, mlx4_en_linkstate);
 	INIT_DELAYED_WORK(&priv->stats_task, mlx4_en_do_get_stats);
diff --git a/drivers/net/mlx4/en_rx.c b/drivers/net/mlx4/en_rx.c
index 5a14899c1e25..91bdfdfd431f 100644
--- a/drivers/net/mlx4/en_rx.c
+++ b/drivers/net/mlx4/en_rx.c
@@ -269,31 +269,6 @@ reduce_rings:
 	return 0;
 }
 
-static int mlx4_en_fill_rx_buf(struct net_device *dev,
-			       struct mlx4_en_rx_ring *ring)
-{
-	struct mlx4_en_priv *priv = netdev_priv(dev);
-	int num = 0;
-	int err;
-
-	while ((u32) (ring->prod - ring->cons) < ring->actual_size) {
-		err = mlx4_en_prepare_rx_desc(priv, ring, ring->prod &
-					      ring->size_mask);
-		if (err) {
-			if (netif_msg_rx_err(priv))
-				en_warn(priv, "Failed preparing rx descriptor\n");
-			priv->port_stats.rx_alloc_failed++;
-			break;
-		}
-		++num;
-		++ring->prod;
-	}
-	if ((u32) (ring->prod - ring->cons) == ring->actual_size)
-		ring->full = 1;
-
-	return num;
-}
-
 static void mlx4_en_free_rx_buf(struct mlx4_en_priv *priv,
 				struct mlx4_en_rx_ring *ring)
 {
@@ -312,42 +287,6 @@ static void mlx4_en_free_rx_buf(struct mlx4_en_priv *priv,
 	}
 }
 
-
-void mlx4_en_rx_refill(struct work_struct *work)
-{
-	struct delayed_work *delay = to_delayed_work(work);
-	struct mlx4_en_priv *priv = container_of(delay, struct mlx4_en_priv,
-						 refill_task);
-	struct mlx4_en_dev *mdev = priv->mdev;
-	struct net_device *dev = priv->dev;
-	struct mlx4_en_rx_ring *ring;
-	int need_refill = 0;
-	int i;
-
-	mutex_lock(&mdev->state_lock);
-	if (!mdev->device_up || !priv->port_up)
-		goto out;
-
-	/* We only get here if there are no receive buffers, so we can't race
-	 * with Rx interrupts while filling buffers */
-	for (i = 0; i < priv->rx_ring_num; i++) {
-		ring = &priv->rx_ring[i];
-		if (ring->need_refill) {
-			if (mlx4_en_fill_rx_buf(dev, ring)) {
-				ring->need_refill = 0;
-				mlx4_en_update_rx_prod_db(ring);
-			} else
-				need_refill = 1;
-		}
-	}
-	if (need_refill)
-		queue_delayed_work(mdev->workqueue, &priv->refill_task, HZ);
-
-out:
-	mutex_unlock(&mdev->state_lock);
-}
-
-
 int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
 			   struct mlx4_en_rx_ring *ring, u32 size, u16 stride)
 {
@@ -457,9 +396,6 @@ int mlx4_en_activate_rx_rings(struct mlx4_en_priv *priv)
 			ring_ind--;
 			goto err_allocator;
 		}
-
-		/* Fill Rx buffers */
-		ring->full = 0;
 	}
 	err = mlx4_en_fill_rx_buffers(priv);
 	if (err)
@@ -647,33 +583,6 @@ static struct sk_buff *mlx4_en_rx_skb(struct mlx4_en_priv *priv,
 	return skb;
 }
 
-static void mlx4_en_copy_desc(struct mlx4_en_priv *priv,
-			      struct mlx4_en_rx_ring *ring,
-			      int from, int to, int num)
-{
-	struct skb_frag_struct *skb_frags_from;
-	struct skb_frag_struct *skb_frags_to;
-	struct mlx4_en_rx_desc *rx_desc_from;
-	struct mlx4_en_rx_desc *rx_desc_to;
-	int from_index, to_index;
-	int nr, i;
-
-	for (i = 0; i < num; i++) {
-		from_index = (from + i) & ring->size_mask;
-		to_index = (to + i) & ring->size_mask;
-		skb_frags_from = ring->rx_info + (from_index << priv->log_rx_info);
-		skb_frags_to = ring->rx_info + (to_index << priv->log_rx_info);
-		rx_desc_from = ring->buf + (from_index << ring->log_stride);
-		rx_desc_to = ring->buf + (to_index << ring->log_stride);
-
-		for (nr = 0; nr < priv->num_frags; nr++) {
-			skb_frags_to[nr].page = skb_frags_from[nr].page;
-			skb_frags_to[nr].page_offset = skb_frags_from[nr].page_offset;
-			rx_desc_to->data[nr].addr = rx_desc_from->data[nr].addr;
-		}
-	}
-}
-
 
 int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int budget)
 {
@@ -821,11 +730,6 @@ out:
 	wmb(); /* ensure HW sees CQ consumer before we post new buffers */
 	ring->cons = cq->mcq.cons_index;
 	ring->prod += polled; /* Polled descriptors were realocated in place */
-	if (unlikely(!ring->full)) {
-		mlx4_en_copy_desc(priv, ring, ring->cons - polled,
-				  ring->prod - polled, polled);
-		mlx4_en_fill_rx_buf(dev, ring);
-	}
 	mlx4_en_update_rx_prod_db(ring);
 	return polled;
 }
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index ad861db66f19..c7c5e86804ff 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -295,8 +295,6 @@ struct mlx4_en_rx_ring {
 	u32 prod;
 	u32 cons;
 	u32 buf_size;
-	int need_refill;
-	int full;
 	void *buf;
 	void *rx_info;
 	unsigned long bytes;
@@ -494,7 +492,6 @@ struct mlx4_en_priv {
 	struct mlx4_en_cq rx_cq[MAX_RX_RINGS];
 	struct work_struct mcast_task;
 	struct work_struct mac_task;
-	struct delayed_work refill_task;
 	struct work_struct watchdog_task;
 	struct work_struct linkstate_task;
 	struct delayed_work stats_task;
@@ -564,7 +561,6 @@ void mlx4_en_set_default_rss_map(struct mlx4_en_priv *priv,
 int mlx4_en_config_rss_steer(struct mlx4_en_priv *priv);
 void mlx4_en_release_rss_steer(struct mlx4_en_priv *priv);
 int mlx4_en_free_tx_buf(struct net_device *dev, struct mlx4_en_tx_ring *ring);
-void mlx4_en_rx_refill(struct work_struct *work);
 void mlx4_en_rx_irq(struct mlx4_cq *mcq);
 
 int mlx4_SET_MCAST_FLTR(struct mlx4_dev *dev, u8 port, u64 mac, u64 clear, u8 mode);
-- 
cgit v1.2.3-59-g8ed1b


From 8c52da503b7e4cf961807f11824e3258ef9f7f1c Mon Sep 17 00:00:00 2001
From: Eric Anholt <eric@anholt.net>
Date: Thu, 18 Jun 2009 20:22:19 -0700
Subject: drm/i915: Add missing dependency on Intel AGP support.

Users could accidentally enable AGP but not the Intel AGP support, and get
a DRM that doesn't probe as a result.

Bug #22358.

Signed-off-by: Eric Anholt <eric@anholt.net>
---
 drivers/gpu/drm/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index c961fe415aef..39b393d38bb3 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -81,6 +81,7 @@ config DRM_I830
 
 config DRM_I915
 	tristate "i915 driver"
+	depends on AGP_INTEL
 	select FB_CFB_FILLRECT
 	select FB_CFB_COPYAREA
 	select FB_CFB_IMAGEBLIT
-- 
cgit v1.2.3-59-g8ed1b


From a8f4310be59a2e7fc80fba945bcb32b18f4ad54f Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Mon, 22 Jun 2009 07:36:52 +0200
Subject: ALSA: ctxfi - Allow unknown PCI SSIDs

Allow unknown PCI SSIDs for emu20k1 and emu20k2 as "unknown" model.
Also, add a black-list check in case any device has to be listed
as "unsupported".  It has a negative value in the pci quirk entry.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/ctxfi/ctatc.c      | 25 +++++++++++++++++++------
 sound/pci/ctxfi/cthardware.h |  3 +++
 2 files changed, 22 insertions(+), 6 deletions(-)

diff --git a/sound/pci/ctxfi/ctatc.c b/sound/pci/ctxfi/ctatc.c
index b0adc8094009..32e3c26e969e 100644
--- a/sound/pci/ctxfi/ctatc.c
+++ b/sound/pci/ctxfi/ctatc.c
@@ -46,8 +46,6 @@ static struct snd_pci_quirk __devinitdata subsys_20k1_list[] = {
 	SND_PCI_QUIRK(PCI_VENDOR_ID_CREATIVE, 0x0031, "SB073x", CTSB073X),
 	SND_PCI_QUIRK_MASK(PCI_VENDOR_ID_CREATIVE, 0xf000, 0x6000,
 			   "UAA", CTUAA),
-	SND_PCI_QUIRK_VENDOR(PCI_VENDOR_ID_CREATIVE,
-			     "Unknown", CT20K1_UNKNOWN),
 	{ } /* terminator */
 };
 
@@ -67,13 +65,16 @@ static struct snd_pci_quirk __devinitdata subsys_20k2_list[] = {
 };
 
 static const char *ct_subsys_name[NUM_CTCARDS] = {
+	/* 20k1 models */
 	[CTSB055X]	= "SB055x",
 	[CTSB073X]	= "SB073x",
-	[CTSB0760]	= "SB076x",
 	[CTUAA]		= "UAA",
 	[CT20K1_UNKNOWN] = "Unknown",
+	/* 20k2 models */
+	[CTSB0760]	= "SB076x",
 	[CTHENDRIX]	= "Hendrix",
 	[CTSB0880]	= "SB0880",
+	[CT20K2_UNKNOWN] = "Unknown",
 };
 
 static struct {
@@ -1240,9 +1241,21 @@ static int __devinit atc_identify_card(struct ct_atc *atc)
 		return -ENOENT;
 	}
 	p = snd_pci_quirk_lookup(atc->pci, list);
-	if (!p)
-		return -ENOENT;
-	atc->model = p->value;
+	if (p) {
+		if (p->value < 0) {
+			printk(KERN_ERR "ctxfi: "
+			       "Device %04x:%04x is black-listed\n",
+			       atc->pci->subsystem_vendor,
+			       atc->pci->subsystem_device);
+			return -ENOENT;
+		}
+		atc->model = p->value;
+	} else {
+		if (atc->chip_type == ATC20K1)
+			atc->model = CT20K1_UNKNOWN;
+		else
+			atc->model = CT20K2_UNKNOWN;
+	}
 	atc->model_name = ct_subsys_name[atc->model];
 	snd_printd("ctxfi: chip %s model %s (%04x:%04x) is found\n",
 		   atc->chip_name, atc->model_name,
diff --git a/sound/pci/ctxfi/cthardware.h b/sound/pci/ctxfi/cthardware.h
index 4a8e04f090a4..a4c2cad80f3a 100644
--- a/sound/pci/ctxfi/cthardware.h
+++ b/sound/pci/ctxfi/cthardware.h
@@ -30,13 +30,16 @@ enum CHIPTYP {
 enum CTCARDS {
 	/* 20k1 models */
 	CTSB055X,
+	CT20K1_MODEL_FIRST = CTSB055X,
 	CTSB073X,
 	CTUAA,
 	CT20K1_UNKNOWN,
 	/* 20k2 models */
 	CTSB0760,
+	CT20K2_MODEL_FIRST = CTSB0760,
 	CTHENDRIX,
 	CTSB0880,
+	CT20K2_UNKNOWN,
 	NUM_CTCARDS		/* This should always be the last */
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From f6b24caaf933a466397915a08e30e885a32f905a Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Sun, 21 Jun 2009 22:42:30 -0700
Subject: via-velocity: Fix velocity driver unmapping incorrect size.

When a packet is greater than ETH_ZLEN, we end up assigning the
boolean result of a comparison to the size we unmap.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/via-velocity.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/via-velocity.c b/drivers/net/via-velocity.c
index b02f7adff5dc..3ba35956327a 100644
--- a/drivers/net/via-velocity.c
+++ b/drivers/net/via-velocity.c
@@ -1847,7 +1847,7 @@ static void velocity_free_tx_buf(struct velocity_info *vptr, struct velocity_td_
 	 */
 	if (tdinfo->skb_dma) {
 
-		pktlen = (skb->len > ETH_ZLEN ? : ETH_ZLEN);
+		pktlen = max_t(unsigned int, skb->len, ETH_ZLEN);
 		for (i = 0; i < tdinfo->nskb_dma; i++) {
 #ifdef VELOCITY_ZERO_COPY_SUPPORT
 			pci_unmap_single(vptr->pdev, tdinfo->skb_dma[i], le16_to_cpu(td->tdesc1.len), PCI_DMA_TODEVICE);
-- 
cgit v1.2.3-59-g8ed1b


From e01698aed04811b9a9c4f8d54b73cb182757063d Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Sun, 21 Jun 2009 22:48:03 -0700
Subject: ide cmd64x: Remove serialize setting.

This begins to fix regressions reported by Frans Pop on his Ultra-10.

There are still some funnies left that we are investigating.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/cmd64x.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/ide/cmd64x.c b/drivers/ide/cmd64x.c
index 03c86209446f..680e5975217f 100644
--- a/drivers/ide/cmd64x.c
+++ b/drivers/ide/cmd64x.c
@@ -389,8 +389,7 @@ static const struct ide_port_info cmd64x_chipsets[] __devinitdata = {
 		.init_chipset	= init_chipset_cmd64x,
 		.enablebits	= {{0x51,0x04,0x04}, {0x51,0x08,0x08}},
 		.port_ops	= &cmd648_port_ops,
-		.host_flags	= IDE_HFLAG_SERIALIZE |
-				  IDE_HFLAG_ABUSE_PREFETCH,
+		.host_flags	= IDE_HFLAG_ABUSE_PREFETCH,
 		.pio_mask	= ATA_PIO5,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA2,
-- 
cgit v1.2.3-59-g8ed1b


From 8bd9bca3c1a214350e2f2f1e2fd493ed24c06f7e Mon Sep 17 00:00:00 2001
From: Mariusz Kozlowski <m.kozlowski@tuxland.pl>
Date: Sun, 21 Jun 2009 20:26:59 +0200
Subject: sound: fix check for return value in snd_pcm_hw_refine

'params' is a pointer and looking at the code this probably should be a check
for ioctl return value.

Signed-off-by: Mariusz Kozlowski <m.kozlowski@tuxland.pl>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/core/pcm_native.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c
index 84da3ba17c86..ac2150e0670d 100644
--- a/sound/core/pcm_native.c
+++ b/sound/core/pcm_native.c
@@ -320,7 +320,7 @@ int snd_pcm_hw_refine(struct snd_pcm_substream *substream,
                     snd_mask_max(&params->masks[SNDRV_PCM_HW_PARAM_CHANNELS])) {
 			changed = substream->ops->ioctl(substream,
 					SNDRV_PCM_IOCTL1_FIFO_SIZE, params);
-			if (params < 0)
+			if (changed < 0)
 				return changed;
 		}
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 115551d98e4c05bb982bd940f08a489fa9b0a4a1 Mon Sep 17 00:00:00 2001
From: Simon Arlott <simon@fire.lp0.eu>
Date: Sun, 21 Jun 2009 19:50:48 +0100
Subject: ALSA: via82xx: add option to disable 500ms delay in
 snd_via82xx_codec_wait

There's a large 500ms delay in snd_via82xx_codec_wait() that, at least
on my hardware, appears to be unnecessary. The rest of the init of
the card works without logging any warnings or errors and both audio
and mixer settings work.

This adds an "nodelay" parameter to disable this (undocumented in the
code) large delay improving bootup time by 489-500ms.

[    1.034217] initcall alsa_card_via82xx_init+0x0/0x16 returned 0 after 505757 usecs
vs.
[    0.533136] initcall alsa_card_via82xx_init+0x0/0x16 returned 0 after 15915 usecs

Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/via82xx.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/sound/pci/via82xx.c b/sound/pci/via82xx.c
index 1ef58c51c213..949fcaf6b70e 100644
--- a/sound/pci/via82xx.c
+++ b/sound/pci/via82xx.c
@@ -85,6 +85,7 @@ static int joystick;
 static int ac97_clock = 48000;
 static char *ac97_quirk;
 static int dxs_support;
+static int nodelay;
 
 module_param(index, int, 0444);
 MODULE_PARM_DESC(index, "Index value for VIA 82xx bridge.");
@@ -102,6 +103,8 @@ module_param(ac97_quirk, charp, 0444);
 MODULE_PARM_DESC(ac97_quirk, "AC'97 workaround for strange hardware.");
 module_param(dxs_support, int, 0444);
 MODULE_PARM_DESC(dxs_support, "Support for DXS channels (0 = auto, 1 = enable, 2 = disable, 3 = 48k only, 4 = no VRA, 5 = enable any sample rate)");
+module_param(nodelay, int, 0444);
+MODULE_PARM_DESC(nodelay, "Disable 500ms init delay");
 
 /* just for backward compatibility */
 static int enable;
@@ -549,7 +552,8 @@ static void snd_via82xx_codec_wait(struct snd_ac97 *ac97)
 	int err;
 	err = snd_via82xx_codec_ready(chip, ac97->num);
 	/* here we need to wait fairly for long time.. */
-	msleep(500);
+	if (!nodelay)
+		msleep(500);
 }
 
 static void snd_via82xx_codec_write(struct snd_ac97 *ac97,
-- 
cgit v1.2.3-59-g8ed1b


From b8621516cc7dd22fe4ad4c2fd8f7d852a33212df Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Mon, 22 Jun 2009 08:00:10 +0200
Subject: ALSA: hda - Fix unsigned comparison in patch_sigmatel.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the comparison of unsigned int that causes a compile warning below
by changing to the right signed type:
  patch_sigmatel.c: In function ‘stac92xx_vref_set’:
  patch_sigmatel.c:658: warning: comparison of unsigned expression < 0 is always false

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/hda/patch_sigmatel.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c
index 93e47c96a38b..090957735e05 100644
--- a/sound/pci/hda/patch_sigmatel.c
+++ b/sound/pci/hda/patch_sigmatel.c
@@ -639,7 +639,7 @@ static int stac92xx_smux_enum_put(struct snd_kcontrol *kcontrol,
 static unsigned int stac92xx_vref_set(struct hda_codec *codec,
 					hda_nid_t nid, unsigned int new_vref)
 {
-	unsigned int error;
+	int error;
 	unsigned int pincfg;
 	pincfg = snd_hda_codec_read(codec, nid, 0,
 				AC_VERB_GET_PIN_WIDGET_CONTROL, 0);
@@ -2703,7 +2703,7 @@ static int stac92xx_dc_bias_put(struct snd_kcontrol *kcontrol,
 {
 	struct hda_codec *codec = snd_kcontrol_chip(kcontrol);
 	unsigned int new_vref = 0;
-	unsigned int error;
+	int error;
 	hda_nid_t nid = kcontrol->private_value;
 
 	if (ucontrol->value.enumerated.item[0] == 0)
-- 
cgit v1.2.3-59-g8ed1b


From ae14ef68e8e67ca5b8b29f0eb640f7c106617f4e Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Mon, 22 Jun 2009 08:16:56 +0200
Subject: ALSA: hda - Get back Input Source for ALC262 toshiba-s06 model

The commit f9e336f65b666b8f1764d17e9b7c21c90748a37e
    ALSA: hda - Unify capture mixer creation in realtek codes
removed the "Input Source" mixer element creation for toshiba-s06 model
because it contains a digital-mic input.

This patch take the control back.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Cc: <stable@kernel.org>
---
 sound/pci/hda/patch_realtek.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index bf4b78a74a8f..105fb23398e1 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -11539,6 +11539,7 @@ static struct alc_config_preset alc262_presets[] = {
 		.capsrc_nids = alc262_dmic_capsrc_nids,
 		.dac_nids = alc262_dac_nids,
 		.adc_nids = alc262_dmic_adc_nids, /* ADC0 */
+		.num_adc_nids = 1, /* single ADC */
 		.dig_out_nid = ALC262_DIGOUT_NID,
 		.num_channel_mode = ARRAY_SIZE(alc262_modes),
 		.channel_mode = alc262_modes,
-- 
cgit v1.2.3-59-g8ed1b


From 376b508ffde3b17e105265f89b83bdb044b1c1ae Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Mon, 22 Jun 2009 11:03:13 +0200
Subject: ALSA: hda - Add quirk for Sony VAIO Z21MN

It needs model=toshiba-s06 to work with the digital-mic.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Cc: <stable@kernel.org>
---
 sound/pci/hda/patch_realtek.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 105fb23398e1..d20168494dd7 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -11332,6 +11332,7 @@ static struct snd_pci_quirk alc262_cfg_tbl[] = {
 	SND_PCI_QUIRK(0x104d, 0x8203, "Sony UX-90", ALC262_HIPPO),
 	SND_PCI_QUIRK(0x104d, 0x820f, "Sony ASSAMD", ALC262_SONY_ASSAMD),
 	SND_PCI_QUIRK(0x104d, 0x9016, "Sony VAIO", ALC262_AUTO), /* dig-only */
+	SND_PCI_QUIRK(0x104d, 0x9025, "Sony VAIO Z21MN", ALC262_TOSHIBA_S06),
 	SND_PCI_QUIRK_MASK(0x104d, 0xff00, 0x9000, "Sony VAIO",
 			   ALC262_SONY_ASSAMD),
 	SND_PCI_QUIRK(0x1179, 0x0001, "Toshiba dynabook SS RX1",
-- 
cgit v1.2.3-59-g8ed1b


From 0169b6b33b2b4d9e66e98afc56272e5fbd350df4 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Mon, 22 Jun 2009 10:50:19 +0200
Subject: ALSA: hda - Fix check of input source type for realtek codecs

Fix the check of the input-source type by checking the widget type of
each capture-source item.  Since some codecs can have both the mixer
and selector types depending on the ADC, alc_mux_enum_put() needs to
check each widget.

With this change, spec->capture_style gets unneeded, so it's removed,
too.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/hda/patch_realtek.c | 21 +++------------------
 1 file changed, 3 insertions(+), 18 deletions(-)

diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index d20168494dd7..28a587353b11 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -250,13 +250,6 @@ enum {
 	ALC883_MODEL_LAST,
 };
 
-/* styles of capture selection */
-enum {
-	CAPT_MUX = 0,	/* only mux based */
-	CAPT_MIX,	/* only mixer based */
-	CAPT_1MUX_MIX,	/* first mux and other mixers */
-};
-
 /* for GPIO Poll */
 #define GPIO_MASK	0x03
 
@@ -306,7 +299,6 @@ struct alc_spec {
 	hda_nid_t *adc_nids;
 	hda_nid_t *capsrc_nids;
 	hda_nid_t dig_in_nid;		/* digital-in NID; optional */
-	int capture_style;		/* capture style (CAPT_*) */
 
 	/* capture source */
 	unsigned int num_mux_defs;
@@ -420,12 +412,13 @@ static int alc_mux_enum_put(struct snd_kcontrol *kcontrol,
 	unsigned int mux_idx;
 	hda_nid_t nid = spec->capsrc_nids ?
 		spec->capsrc_nids[adc_idx] : spec->adc_nids[adc_idx];
+	unsigned int type;
 
 	mux_idx = adc_idx >= spec->num_mux_defs ? 0 : adc_idx;
 	imux = &spec->input_mux[mux_idx];
 
-	if (spec->capture_style &&
-	    !(spec->capture_style == CAPT_1MUX_MIX && !adc_idx)) {
+	type = (get_wcaps(codec, nid) & AC_WCAP_TYPE) >> AC_WCAP_TYPE_SHIFT;
+	if (type == AC_WID_AUD_MIX) {
 		/* Matrix-mixer style (e.g. ALC882) */
 		unsigned int *cur_val = &spec->cur_mux[adc_idx];
 		unsigned int i, idx;
@@ -7557,7 +7550,6 @@ static int patch_alc882(struct hda_codec *codec)
 	spec->stream_digital_playback = &alc882_pcm_digital_playback;
 	spec->stream_digital_capture = &alc882_pcm_digital_capture;
 
-	spec->capture_style = CAPT_MIX; /* matrix-style capture */
 	if (!spec->adc_nids && spec->input_mux) {
 		/* check whether NID 0x07 is valid */
 		unsigned int wcap = get_wcaps(codec, 0x07);
@@ -9781,7 +9773,6 @@ static int patch_alc883(struct hda_codec *codec)
 		}
 		if (!spec->capsrc_nids)
 			spec->capsrc_nids = alc883_capsrc_nids;
-		spec->capture_style = CAPT_MIX; /* matrix-style capture */
 		spec->init_amp = ALC_INIT_DEFAULT; /* always initialize */
 		break;
 	case 0x10ec0889:
@@ -9791,8 +9782,6 @@ static int patch_alc883(struct hda_codec *codec)
 		}
 		if (!spec->capsrc_nids)
 			spec->capsrc_nids = alc889_capsrc_nids;
-		spec->capture_style = CAPT_1MUX_MIX; /* 1mux/Nmix-style
-							capture */
 		break;
 	default:
 		if (!spec->num_adc_nids) {
@@ -9801,7 +9790,6 @@ static int patch_alc883(struct hda_codec *codec)
 		}
 		if (!spec->capsrc_nids)
 			spec->capsrc_nids = alc883_capsrc_nids;
-		spec->capture_style = CAPT_MIX; /* matrix-style capture */
 		break;
 	}
 
@@ -11642,7 +11630,6 @@ static int patch_alc262(struct hda_codec *codec)
 	spec->stream_digital_playback = &alc262_pcm_digital_playback;
 	spec->stream_digital_capture = &alc262_pcm_digital_capture;
 
-	spec->capture_style = CAPT_MIX;
 	if (!spec->adc_nids && spec->input_mux) {
 		/* check whether NID 0x07 is valid */
 		unsigned int wcap = get_wcaps(codec, 0x07);
@@ -15556,7 +15543,6 @@ static int patch_alc861vd(struct hda_codec *codec)
 	spec->adc_nids = alc861vd_adc_nids;
 	spec->num_adc_nids = ARRAY_SIZE(alc861vd_adc_nids);
 	spec->capsrc_nids = alc861vd_capsrc_nids;
-	spec->capture_style = CAPT_MIX;
 
 	set_capture_mixer(spec);
 	set_beep_amp(spec, 0x0b, 0x05, HDA_INPUT);
@@ -17476,7 +17462,6 @@ static int patch_alc662(struct hda_codec *codec)
 	spec->adc_nids = alc662_adc_nids;
 	spec->num_adc_nids = ARRAY_SIZE(alc662_adc_nids);
 	spec->capsrc_nids = alc662_capsrc_nids;
-	spec->capture_style = CAPT_MIX;
 
 	if (!spec->cap_mixer)
 		set_capture_mixer(spec);
-- 
cgit v1.2.3-59-g8ed1b


From 8c927b4acf819ed0170bff991bc9eaec4c85deb8 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Mon, 22 Jun 2009 10:56:54 +0200
Subject: ALSA: hda - Add digital-mic support to ALC262 auto model

Add the digital-mic support with ALC262 auto model.
The new ALC262 models have the digital mic at NID 0x12.  This widget
isn't checked in the current alc262_auto_create_analog_input_ctls()
since it's under 0x18.  So, just reuse the routine for alc269 to fix
the behavior.

But, it doesn't suffice: the digital mic is supported only with the
ADC0, we have to exclude other ADCs when d-mic is detected.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/hda/patch_realtek.c | 86 +++++++++++++++++++++++++------------------
 1 file changed, 51 insertions(+), 35 deletions(-)

diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 28a587353b11..334533197425 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -10901,9 +10901,27 @@ static int alc262_auto_create_multi_out_ctls(struct alc_spec *spec,
 	return 0;
 }
 
-/* identical with ALC880 */
-#define alc262_auto_create_analog_input_ctls \
-	alc880_auto_create_analog_input_ctls
+static int alc262_auto_create_analog_input_ctls(struct alc_spec *spec,
+						const struct auto_pin_cfg *cfg)
+{
+	int err;
+
+	err = alc880_auto_create_analog_input_ctls(spec, cfg);
+	if (err < 0)
+		return err;
+	/* digital-mic input pin is excluded in alc880_auto_create..()
+	 * because it's under 0x18
+	 */
+	if (cfg->input_pins[AUTO_PIN_MIC] == 0x12 ||
+	    cfg->input_pins[AUTO_PIN_FRONT_MIC] == 0x12) {
+		struct hda_input_mux *imux = &spec->private_imux[0];
+		imux->items[imux->num_items].label = "Int Mic";
+		imux->items[imux->num_items].index = 0x09;
+		imux->num_items++;
+	}
+	return 0;
+}
+
 
 /*
  * generic initialization of ADC, input mixers and output mixers
@@ -11631,19 +11649,35 @@ static int patch_alc262(struct hda_codec *codec)
 	spec->stream_digital_capture = &alc262_pcm_digital_capture;
 
 	if (!spec->adc_nids && spec->input_mux) {
-		/* check whether NID 0x07 is valid */
-		unsigned int wcap = get_wcaps(codec, 0x07);
-
-		/* get type */
-		wcap = (wcap & AC_WCAP_TYPE) >> AC_WCAP_TYPE_SHIFT;
-		if (wcap != AC_WID_AUD_IN) {
-			spec->adc_nids = alc262_adc_nids_alt;
-			spec->num_adc_nids = ARRAY_SIZE(alc262_adc_nids_alt);
-			spec->capsrc_nids = alc262_capsrc_nids_alt;
+		int i;
+		/* check whether the digital-mic has to be supported */
+		for (i = 0; i < spec->input_mux->num_items; i++) {
+			if (spec->input_mux->items[i].index >= 9)
+				break;
+		}
+		if (i < spec->input_mux->num_items) {
+			/* use only ADC0 */
+			spec->adc_nids = alc262_dmic_adc_nids;
+			spec->num_adc_nids = 1;
+			spec->capsrc_nids = alc262_dmic_capsrc_nids;
 		} else {
-			spec->adc_nids = alc262_adc_nids;
-			spec->num_adc_nids = ARRAY_SIZE(alc262_adc_nids);
-			spec->capsrc_nids = alc262_capsrc_nids;
+			/* all analog inputs */
+			/* check whether NID 0x07 is valid */
+			unsigned int wcap = get_wcaps(codec, 0x07);
+
+			/* get type */
+			wcap = (wcap & AC_WCAP_TYPE) >> AC_WCAP_TYPE_SHIFT;
+			if (wcap != AC_WID_AUD_IN) {
+				spec->adc_nids = alc262_adc_nids_alt;
+				spec->num_adc_nids =
+					ARRAY_SIZE(alc262_adc_nids_alt);
+				spec->capsrc_nids = alc262_capsrc_nids_alt;
+			} else {
+				spec->adc_nids = alc262_adc_nids;
+				spec->num_adc_nids =
+					ARRAY_SIZE(alc262_adc_nids);
+				spec->capsrc_nids = alc262_capsrc_nids;
+			}
 		}
 	}
 	if (!spec->cap_mixer && !spec->no_analog)
@@ -13233,26 +13267,8 @@ static int alc269_auto_create_multi_out_ctls(struct alc_spec *spec,
 	return 0;
 }
 
-static int alc269_auto_create_analog_input_ctls(struct alc_spec *spec,
-						const struct auto_pin_cfg *cfg)
-{
-	int err;
-
-	err = alc880_auto_create_analog_input_ctls(spec, cfg);
-	if (err < 0)
-		return err;
-	/* digital-mic input pin is excluded in alc880_auto_create..()
-	 * because it's under 0x18
-	 */
-	if (cfg->input_pins[AUTO_PIN_MIC] == 0x12 ||
-	    cfg->input_pins[AUTO_PIN_FRONT_MIC] == 0x12) {
-		struct hda_input_mux *imux = &spec->private_imux[0];
-		imux->items[imux->num_items].label = "Int Mic";
-		imux->items[imux->num_items].index = 0x05;
-		imux->num_items++;
-	}
-	return 0;
-}
+#define alc269_auto_create_analog_input_ctls \
+	alc262_auto_create_analog_input_ctls
 
 #ifdef CONFIG_SND_HDA_POWER_SAVE
 #define alc269_loopbacks	alc880_loopbacks
-- 
cgit v1.2.3-59-g8ed1b


From 0e0497c0c017664994819f4602dc07fd95896c52 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:08:02 +0100
Subject: dm mpath: validate table argument count

The parser reads the argument count as a number but doesn't check that
sufficient arguments are supplied. This command triggers the bug:

dmsetup create mpath --table "0 `blockdev --getsize /dev/mapper/cr0`
    multipath 0 0 2 1 round-robin 1000 0 1 1 /dev/mapper/cr0
    round-robin 0 1 1 /dev/mapper/cr1 1000"
kernel BUG at drivers/md/dm-mpath.c:530!

Cc: stable@kernel.org
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-mpath.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index 6a386ab4f7eb..d880299334e9 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -553,6 +553,12 @@ static int parse_path_selector(struct arg_set *as, struct priority_group *pg,
 		return -EINVAL;
 	}
 
+	if (ps_argc > as->argc) {
+		dm_put_path_selector(pst);
+		ti->error = "not enough arguments for path selector";
+		return -EINVAL;
+	}
+
 	r = pst->create(&pg->ps, ps_argc, as->argv);
 	if (r) {
 		dm_put_path_selector(pst);
-- 
cgit v1.2.3-59-g8ed1b


From 6423f9ea8035138d70bae1a278d3b57b743f8b3e Mon Sep 17 00:00:00 2001
From: Clemens Ladisch <clemens@ladisch.de>
Date: Mon, 22 Jun 2009 10:01:59 +0200
Subject: sound: seq_midi_event: fix decoding of (N)RPN events

When decoding (N)RPN sequencer events into raw MIDI commands, the
extra_decode_xrpn() function had accidentally swapped the MSB and LSB
controller values of both the parameter number and the data value.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Cc: <stable@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/core/seq/seq_midi_event.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/sound/core/seq/seq_midi_event.c b/sound/core/seq/seq_midi_event.c
index 8284f176a342..b5d6ea4904c0 100644
--- a/sound/core/seq/seq_midi_event.c
+++ b/sound/core/seq/seq_midi_event.c
@@ -504,10 +504,10 @@ static int extra_decode_xrpn(struct snd_midi_event *dev, unsigned char *buf,
 	if (dev->nostat && count < 12)
 		return -ENOMEM;
 	cmd = MIDI_CMD_CONTROL|(ev->data.control.channel & 0x0f);
-	bytes[0] = ev->data.control.param & 0x007f;
-	bytes[1] = (ev->data.control.param & 0x3f80) >> 7;
-	bytes[2] = ev->data.control.value & 0x007f;
-	bytes[3] = (ev->data.control.value & 0x3f80) >> 7;
+	bytes[0] = (ev->data.control.param & 0x3f80) >> 7;
+	bytes[1] = ev->data.control.param & 0x007f;
+	bytes[2] = (ev->data.control.value & 0x3f80) >> 7;
+	bytes[3] = ev->data.control.value & 0x007f;
 	if (cmd != dev->lastcmd && !dev->nostat) {
 		if (count < 9)
 			return -ENOMEM;
-- 
cgit v1.2.3-59-g8ed1b


From e094f4f15f5169526c7200b9bde44b900548a81e Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:10 +0100
Subject: dm mpath: validate hw_handler argument count

Fix arg count parsing error in hw handlers.

Cc: stable@kernel.org
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-mpath.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index d880299334e9..f25bdebcb4ab 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -705,6 +705,11 @@ static int parse_hw_handler(struct arg_set *as, struct multipath *m)
 	if (!hw_argc)
 		return 0;
 
+	if (hw_argc > as->argc) {
+		ti->error = "not enough arguments for hardware handler";
+		return -EINVAL;
+	}
+
 	m->hw_handler_name = kstrdup(shift(as), GFP_KERNEL);
 	request_module("scsi_dh_%s", m->hw_handler_name);
 	if (scsi_dh_handler_exist(m->hw_handler_name) == 0) {
-- 
cgit v1.2.3-59-g8ed1b


From 4d89b7b4e4726893453d0fb4ddbb5b3e16353994 Mon Sep 17 00:00:00 2001
From: Milan Broz <mbroz@redhat.com>
Date: Mon, 22 Jun 2009 10:12:11 +0100
Subject: dm: sysfs skip output when device is being destroyed

Do not process sysfs attributes when device is being destroyed.

Otherwise code can cause
  BUG_ON(test_bit(DMF_FREEING, &md->flags));
in dm_put() call.

Cc: stable@kernel.org
Signed-off-by: Milan Broz <mbroz@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 48db308fae67..f1db689667ea 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1777,6 +1777,10 @@ struct mapped_device *dm_get_from_kobject(struct kobject *kobj)
 	if (&md->kobj != kobj)
 		return NULL;
 
+	if (test_bit(DMF_FREEING, &md->flags) ||
+	    test_bit(DMF_DELETING, &md->flags))
+		return NULL;
+
 	dm_get(md);
 	return md;
 }
-- 
cgit v1.2.3-59-g8ed1b


From a0cf7ea9549ec60988369f90e5c0f855f08abac9 Mon Sep 17 00:00:00 2001
From: Hannes Reinecke <hare@suse.de>
Date: Mon, 22 Jun 2009 10:12:11 +0100
Subject: dm mpath: change attached scsi_dh

When specifying a different hardware handler via multipath
features we should be able to override the built-in defaults.

The problem here is the hardware table from scsi_dh is compiled
in and cannot be changed from userland. The multipath.conf OTOH
is purely user-defined and, what's more, the user might have a valid
reason for modifying it.
(EG EMC Clariion can well be run in PNR mode even though ALUA is
active, or the user might want to try ALUA on any as-of-yet unknown
devices)

So _not_ allowing multipath to override the device handler setting
will just add to the confusion and makes error tracking even more
difficult.

Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-mpath.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index f25bdebcb4ab..545abcc25c42 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -597,9 +597,20 @@ static struct pgpath *parse_path(struct arg_set *as, struct path_selector *ps,
 	}
 
 	if (m->hw_handler_name) {
-		r = scsi_dh_attach(bdev_get_queue(p->path.dev->bdev),
-				   m->hw_handler_name);
+		struct request_queue *q = bdev_get_queue(p->path.dev->bdev);
+
+		r = scsi_dh_attach(q, m->hw_handler_name);
+		if (r == -EBUSY) {
+			/*
+			 * Already attached to different hw_handler,
+			 * try to reattach with correct one.
+			 */
+			scsi_dh_detach(q);
+			r = scsi_dh_attach(q, m->hw_handler_name);
+		}
+
 		if (r < 0) {
+			ti->error = "error attaching hardware handler";
 			dm_put_device(ti, p->path.dev);
 			goto bad;
 		}
-- 
cgit v1.2.3-59-g8ed1b


From e54f77ddda72781ec1c1696b21aabd6a30cbb7c6 Mon Sep 17 00:00:00 2001
From: Chandra Seetharaman <sekharan@us.ibm.com>
Date: Mon, 22 Jun 2009 10:12:12 +0100
Subject: dm mpath: call activate fn for each path in pg_init

Fixed a problem affecting reinstatement of passive paths.

Before we moved the hardware handler from dm to SCSI, it performed a pg_init
for a path group and didn't maintain any state about each path in hardware
handler code.

But in SCSI dh, such state is now maintained, as we want to fail I/O early on a
path if it is not the active path.

All the hardware handlers have a state now and set to active or some form of
inactive.  They have prep_fn() which uses this state to fail the I/O without
it ever being sent to the device.

So in effect when dm-multipath calls scsi_dh_activate(), activate is
sent to only one path and the "state" of that path is changed appropriately
to "active" while other paths in the same path group are never changed
as they never got an "activate".

In order make sure all the paths in a path group gets their state set
properly when a pg_init happens, we need to call scsi_dh_activate() on
all paths in a path group.

Doing this at the hardware handler layer is not a good option as we
want the multipath layer to define the relationship between path and path
groups and not the hardware handler.

Attached patch sends an "activate" on each path in a path group when a
path group is switched. It also sends an activate when a path is reinstated.

Signed-off-by: Chandra Seetharaman <sekharan@us.ibm.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-mpath.c | 63 +++++++++++++++++++++------------------------------
 1 file changed, 26 insertions(+), 37 deletions(-)

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index 545abcc25c42..838f01b1dd30 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -35,6 +35,7 @@ struct pgpath {
 
 	struct dm_path path;
 	struct work_struct deactivate_path;
+	struct work_struct activate_path;
 };
 
 #define path_to_pgpath(__pgp) container_of((__pgp), struct pgpath, path)
@@ -64,8 +65,6 @@ struct multipath {
 	spinlock_t lock;
 
 	const char *hw_handler_name;
-	struct work_struct activate_path;
-	struct pgpath *pgpath_to_activate;
 	unsigned nr_priority_groups;
 	struct list_head priority_groups;
 	unsigned pg_init_required;	/* pg_init needs calling? */
@@ -128,6 +127,7 @@ static struct pgpath *alloc_pgpath(void)
 	if (pgpath) {
 		pgpath->is_active = 1;
 		INIT_WORK(&pgpath->deactivate_path, deactivate_path);
+		INIT_WORK(&pgpath->activate_path, activate_path);
 	}
 
 	return pgpath;
@@ -160,7 +160,6 @@ static struct priority_group *alloc_priority_group(void)
 
 static void free_pgpaths(struct list_head *pgpaths, struct dm_target *ti)
 {
-	unsigned long flags;
 	struct pgpath *pgpath, *tmp;
 	struct multipath *m = ti->private;
 
@@ -169,10 +168,6 @@ static void free_pgpaths(struct list_head *pgpaths, struct dm_target *ti)
 		if (m->hw_handler_name)
 			scsi_dh_detach(bdev_get_queue(pgpath->path.dev->bdev));
 		dm_put_device(ti, pgpath->path.dev);
-		spin_lock_irqsave(&m->lock, flags);
-		if (m->pgpath_to_activate == pgpath)
-			m->pgpath_to_activate = NULL;
-		spin_unlock_irqrestore(&m->lock, flags);
 		free_pgpath(pgpath);
 	}
 }
@@ -202,7 +197,6 @@ static struct multipath *alloc_multipath(struct dm_target *ti)
 		m->queue_io = 1;
 		INIT_WORK(&m->process_queued_ios, process_queued_ios);
 		INIT_WORK(&m->trigger_event, trigger_event);
-		INIT_WORK(&m->activate_path, activate_path);
 		m->mpio_pool = mempool_create_slab_pool(MIN_IOS, _mpio_cache);
 		if (!m->mpio_pool) {
 			kfree(m);
@@ -427,8 +421,8 @@ static void process_queued_ios(struct work_struct *work)
 {
 	struct multipath *m =
 		container_of(work, struct multipath, process_queued_ios);
-	struct pgpath *pgpath = NULL;
-	unsigned init_required = 0, must_queue = 1;
+	struct pgpath *pgpath = NULL, *tmp;
+	unsigned must_queue = 1;
 	unsigned long flags;
 
 	spin_lock_irqsave(&m->lock, flags);
@@ -446,19 +440,15 @@ static void process_queued_ios(struct work_struct *work)
 		must_queue = 0;
 
 	if (m->pg_init_required && !m->pg_init_in_progress && pgpath) {
-		m->pgpath_to_activate = pgpath;
 		m->pg_init_count++;
 		m->pg_init_required = 0;
-		m->pg_init_in_progress = 1;
-		init_required = 1;
+		list_for_each_entry(tmp, &pgpath->pg->pgpaths, list) {
+			if (queue_work(kmpath_handlerd, &tmp->activate_path))
+				m->pg_init_in_progress++;
+		}
 	}
-
 out:
 	spin_unlock_irqrestore(&m->lock, flags);
-
-	if (init_required)
-		queue_work(kmpath_handlerd, &m->activate_path);
-
 	if (!must_queue)
 		dispatch_queued_ios(m);
 }
@@ -946,9 +936,13 @@ static int reinstate_path(struct pgpath *pgpath)
 
 	pgpath->is_active = 1;
 
-	m->current_pgpath = NULL;
-	if (!m->nr_valid_paths++ && m->queue_size)
+	if (!m->nr_valid_paths++ && m->queue_size) {
+		m->current_pgpath = NULL;
 		queue_work(kmultipathd, &m->process_queued_ios);
+	} else if (m->hw_handler_name && (m->current_pg == pgpath->pg)) {
+		if (queue_work(kmpath_handlerd, &pgpath->activate_path))
+			m->pg_init_in_progress++;
+	}
 
 	dm_path_uevent(DM_UEVENT_PATH_REINSTATED, m->ti,
 		      pgpath->path.dev->name, m->nr_valid_paths);
@@ -1124,35 +1118,30 @@ static void pg_init_done(struct dm_path *path, int errors)
 
 	spin_lock_irqsave(&m->lock, flags);
 	if (errors) {
-		DMERR("Could not failover device. Error %d.", errors);
-		m->current_pgpath = NULL;
-		m->current_pg = NULL;
+		if (pgpath == m->current_pgpath) {
+			DMERR("Could not failover device. Error %d.", errors);
+			m->current_pgpath = NULL;
+			m->current_pg = NULL;
+		}
 	} else if (!m->pg_init_required) {
 		m->queue_io = 0;
 		pg->bypassed = 0;
 	}
 
-	m->pg_init_in_progress = 0;
-	queue_work(kmultipathd, &m->process_queued_ios);
+	m->pg_init_in_progress--;
+	if (!m->pg_init_in_progress)
+		queue_work(kmultipathd, &m->process_queued_ios);
 	spin_unlock_irqrestore(&m->lock, flags);
 }
 
 static void activate_path(struct work_struct *work)
 {
 	int ret;
-	struct multipath *m =
-		container_of(work, struct multipath, activate_path);
-	struct dm_path *path;
-	unsigned long flags;
+	struct pgpath *pgpath =
+		container_of(work, struct pgpath, activate_path);
 
-	spin_lock_irqsave(&m->lock, flags);
-	path = &m->pgpath_to_activate->path;
-	m->pgpath_to_activate = NULL;
-	spin_unlock_irqrestore(&m->lock, flags);
-	if (!path)
-		return;
-	ret = scsi_dh_activate(bdev_get_queue(path->dev->bdev));
-	pg_init_done(path, ret);
+	ret = scsi_dh_activate(bdev_get_queue(pgpath->path.dev->bdev));
+	pg_init_done(&pgpath->path, ret);
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From a72986c562eeec3f7b992198c168f0f41606fe53 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:13 +0100
Subject: dm raid1: keep retrying alloc if mempool_alloc failed

If the code can't handle allocation failures, use __GFP_NOFAIL so that
in case of memory pressure the allocator will retry indefinitely and
won't return NULL which would cause a crash in the function.

This is still not a correct fix, it may cause a classic deadlock when
memory manager waits for I/O being done and I/O waits for some free memory.
I/O code shouldn't allocate any memory. But in this case it probably
doesn't matter much in practice, people usually do not swap on RAID.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-region-hash.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/md/dm-region-hash.c b/drivers/md/dm-region-hash.c
index 7b899be0b087..36dbe29f2fd6 100644
--- a/drivers/md/dm-region-hash.c
+++ b/drivers/md/dm-region-hash.c
@@ -283,7 +283,7 @@ static struct dm_region *__rh_alloc(struct dm_region_hash *rh, region_t region)
 
 	nreg = mempool_alloc(rh->region_pool, GFP_ATOMIC);
 	if (unlikely(!nreg))
-		nreg = kmalloc(sizeof(*nreg), GFP_NOIO);
+		nreg = kmalloc(sizeof(*nreg), GFP_NOIO | __GFP_NOFAIL);
 
 	nreg->state = rh->log->type->in_sync(rh->log, region, 1) ?
 		      DM_RH_CLEAN : DM_RH_NOSYNC;
-- 
cgit v1.2.3-59-g8ed1b


From 53b351f972a882ea8b6cdb19602535f1057c884a Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:13 +0100
Subject: dm mpath: flush keventd queue in destructor

The commit fe9cf30eb8186ef267d1868dc9f12f2d0f40835a moves dm table event
submission from kmultipath queue to kernel kevent queue to avoid a
deadlock.

There is a possibility of race condition because kevent queue is not flushed
in the multipath destructor. The scenario is:
- some event happens and is queued to keventd
- keventd thread is delayed due to scheuling latency or some other work
- multipath device is destroyed
- keventd now attempts to process work_struct that is residing in already
  released memory.

The patch flushes the keventd queue in multipath constructor.
I've already fixed similar bug in dm-raid1.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Cc: stable@kernel.org
---
 drivers/md/dm-mpath.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index 838f01b1dd30..92bdcbb7c935 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -848,6 +848,7 @@ static void multipath_dtr(struct dm_target *ti)
 
 	flush_workqueue(kmpath_handlerd);
 	flush_workqueue(kmultipathd);
+	flush_scheduled_work();
 	free_multipath(m);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 8cbeb67ad50f7d68e5e83be2cb2284de8f9c03b5 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:14 +0100
Subject: dm: avoid unsupported spanning of md stripe boundaries

A bio that has two or more vector entries, size less than or equal to
page size, that crosses a stripe boundary of an underlying md device is
accepted by device mapper (it conforms to all its limits) but not by the
underlying device.

The fix is: If device mapper selects the one-page maximum request size,
it also needs to set its own q->merge_bvec_fn to reject any bios with
multiple vector entries that span more pages.

The problem was discovered in the following scenario:
  * MD - RAID-0
  * LV on the top of it (raid1, snapshot or striped with chunk
size/stripe larger than RAID-0 stripe)
  * one of the logical volumes is exported to xen domU
  * inside xen domU it is partitioned, the key point is that the partition
must be unaligned on page boundary (fdisk normally aligns the partition to
63 sectors which will trigger it)
  * install the system on the partitioned disk in domU
This causes I/O failures in dom0.
Reference: https://bugzilla.redhat.com/show_bug.cgi?id=223947

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index f1db689667ea..59806e67a175 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -925,6 +925,16 @@ static int dm_merge_bvec(struct request_queue *q,
 	 */
 	if (max_size && ti->type->merge)
 		max_size = ti->type->merge(ti, bvm, biovec, max_size);
+	/*
+	 * If the target doesn't support merge method and some of the devices
+	 * provided their merge_bvec method (we know this by looking at
+	 * queue_max_hw_sectors), then we can't allow bios with multiple vector
+	 * entries.  So always set max_size to 0, and the code below allows
+	 * just one page.
+	 */
+	else if (queue_max_hw_sectors(q) <= PAGE_SIZE >> 9)
+
+		max_size = 0;
 
 out_table:
 	dm_table_put(map);
-- 
cgit v1.2.3-59-g8ed1b


From 5657e8fa45cf230df278040c420fb80e06309d8f Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:14 +0100
Subject: dm: use i_size_read

Use i_size_read() instead of reading i_size.

If someone changes the size of the device simultaneously, i_size_read
is guaranteed to return a valid value (either the old one or the new one).

i_size can return some intermediate invalid value (on 32-bit computers
with 64-bit i_size, the reads to both halves of i_size can be interleaved
with updates to i_size, resulting in garbage being returned).

Cc: stable@kernel.org
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-exception-store.h | 2 +-
 drivers/md/dm-log.c             | 2 +-
 drivers/md/dm-table.c           | 3 ++-
 3 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/md/dm-exception-store.h b/drivers/md/dm-exception-store.h
index c92701dc5001..2442c8c07898 100644
--- a/drivers/md/dm-exception-store.h
+++ b/drivers/md/dm-exception-store.h
@@ -156,7 +156,7 @@ static inline void dm_consecutive_chunk_count_inc(struct dm_snap_exception *e)
  */
 static inline sector_t get_dev_size(struct block_device *bdev)
 {
-	return bdev->bd_inode->i_size >> SECTOR_SHIFT;
+	return i_size_read(bdev->bd_inode) >> SECTOR_SHIFT;
 }
 
 static inline chunk_t sector_to_chunk(struct dm_exception_store *store,
diff --git a/drivers/md/dm-log.c b/drivers/md/dm-log.c
index 6fa8ccf91c70..6352a9ad4446 100644
--- a/drivers/md/dm-log.c
+++ b/drivers/md/dm-log.c
@@ -416,7 +416,7 @@ static int create_log_context(struct dm_dirty_log *log, struct dm_target *ti,
 				       bitset_size,
 				       ti->limits.logical_block_size);
 
-		if (buf_size > dev->bdev->bd_inode->i_size) {
+		if (buf_size > i_size_read(dev->bdev->bd_inode)) {
 			DMWARN("log device %s too small: need %llu bytes",
 				dev->name, (unsigned long long)buf_size);
 			kfree(lc);
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index e9a73bb242b0..0e2210c0c168 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -388,7 +388,8 @@ static void close_dev(struct dm_dev_internal *d, struct mapped_device *md)
 static int check_device_area(struct dm_dev_internal *dd, sector_t start,
 			     sector_t len)
 {
-	sector_t dev_size = dd->dm_dev.bdev->bd_inode->i_size >> SECTOR_SHIFT;
+	sector_t dev_size = i_size_read(dd->dm_dev.bdev->bd_inode) >>
+			    SECTOR_SHIFT;
 
 	if (!dev_size)
 		return 1;
-- 
cgit v1.2.3-59-g8ed1b


From f6bd4eb73cdf2a5bf954e497972842f39cabb7e3 Mon Sep 17 00:00:00 2001
From: Jonathan Brassow <jbrassow@redhat.com>
Date: Mon, 22 Jun 2009 10:12:15 +0100
Subject: dm exception store: fix exstore lookup to be case insensitive

When snapshots are created using 'p' instead of 'P' as the
exception store type, the device-mapper table loading fails.

This patch makes the code case insensitive as intended and fixes some
regressions reported with device-mapper snapshots.

Signed-off-by: Jonathan Brassow <jbrassow@redhat.com>
Cc: stable@kernel.org
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-exception-store.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/md/dm-exception-store.c b/drivers/md/dm-exception-store.c
index 75d8081a9041..c3ae51584b12 100644
--- a/drivers/md/dm-exception-store.c
+++ b/drivers/md/dm-exception-store.c
@@ -216,7 +216,7 @@ int dm_exception_store_create(struct dm_target *ti, int argc, char **argv,
 		return -EINVAL;
 	}
 
-	type = get_type(argv[1]);
+	type = get_type(&persistent);
 	if (!type) {
 		ti->error = "Exception store type not recognised";
 		r = -EINVAL;
-- 
cgit v1.2.3-59-g8ed1b


From db8fef4fabe4a546ce74f80bff64fd43776e5912 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:15 +0100
Subject: dm: rename suspended_bdev to bdev

Rename suspended_bdev to bdev.

This patch doesn't change any functionality, just renames the variable.
In the next patch, the variable will be used even for non-suspended device.

(Pre-requisite for the per-target barrier support patches.)

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c | 36 ++++++++++++++++++------------------
 1 file changed, 18 insertions(+), 18 deletions(-)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 59806e67a175..1cfd9b72403d 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -157,7 +157,7 @@ struct mapped_device {
 	 * freeze/thaw support require holding onto a super block
 	 */
 	struct super_block *frozen_sb;
-	struct block_device *suspended_bdev;
+	struct block_device *bdev;
 
 	/* forced geometry settings */
 	struct hd_geometry geometry;
@@ -1214,9 +1214,9 @@ static void free_dev(struct mapped_device *md)
 {
 	int minor = MINOR(disk_devt(md->disk));
 
-	if (md->suspended_bdev) {
+	if (md->bdev) {
 		unlock_fs(md);
-		bdput(md->suspended_bdev);
+		bdput(md->bdev);
 	}
 	destroy_workqueue(md->wq);
 	mempool_destroy(md->tio_pool);
@@ -1259,9 +1259,9 @@ static void __set_size(struct mapped_device *md, sector_t size)
 {
 	set_capacity(md->disk, size);
 
-	mutex_lock(&md->suspended_bdev->bd_inode->i_mutex);
-	i_size_write(md->suspended_bdev->bd_inode, (loff_t)size << SECTOR_SHIFT);
-	mutex_unlock(&md->suspended_bdev->bd_inode->i_mutex);
+	mutex_lock(&md->bdev->bd_inode->i_mutex);
+	i_size_write(md->bdev->bd_inode, (loff_t)size << SECTOR_SHIFT);
+	mutex_unlock(&md->bdev->bd_inode->i_mutex);
 }
 
 static int __bind(struct mapped_device *md, struct dm_table *t)
@@ -1277,7 +1277,7 @@ static int __bind(struct mapped_device *md, struct dm_table *t)
 	if (size != get_capacity(md->disk))
 		memset(&md->geometry, 0, sizeof(md->geometry));
 
-	if (md->suspended_bdev)
+	if (md->bdev)
 		__set_size(md, size);
 
 	if (!size) {
@@ -1521,7 +1521,7 @@ int dm_swap_table(struct mapped_device *md, struct dm_table *table)
 		goto out;
 
 	/* without bdev, the device size cannot be changed */
-	if (!md->suspended_bdev)
+	if (!md->bdev)
 		if (get_capacity(md->disk) != dm_table_get_size(table))
 			goto out;
 
@@ -1543,7 +1543,7 @@ static int lock_fs(struct mapped_device *md)
 
 	WARN_ON(md->frozen_sb);
 
-	md->frozen_sb = freeze_bdev(md->suspended_bdev);
+	md->frozen_sb = freeze_bdev(md->bdev);
 	if (IS_ERR(md->frozen_sb)) {
 		r = PTR_ERR(md->frozen_sb);
 		md->frozen_sb = NULL;
@@ -1563,7 +1563,7 @@ static void unlock_fs(struct mapped_device *md)
 	if (!test_bit(DMF_FROZEN, &md->flags))
 		return;
 
-	thaw_bdev(md->suspended_bdev, md->frozen_sb);
+	thaw_bdev(md->bdev, md->frozen_sb);
 	md->frozen_sb = NULL;
 	clear_bit(DMF_FROZEN, &md->flags);
 }
@@ -1603,8 +1603,8 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags)
 
 	/* bdget() can stall if the pending I/Os are not flushed */
 	if (!noflush) {
-		md->suspended_bdev = bdget_disk(md->disk, 0);
-		if (!md->suspended_bdev) {
+		md->bdev = bdget_disk(md->disk, 0);
+		if (!md->bdev) {
 			DMWARN("bdget failed in dm_suspend");
 			r = -ENOMEM;
 			goto out;
@@ -1675,9 +1675,9 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags)
 	set_bit(DMF_SUSPENDED, &md->flags);
 
 out:
-	if (r && md->suspended_bdev) {
-		bdput(md->suspended_bdev);
-		md->suspended_bdev = NULL;
+	if (r && md->bdev) {
+		bdput(md->bdev);
+		md->bdev = NULL;
 	}
 
 	dm_table_put(map);
@@ -1708,9 +1708,9 @@ int dm_resume(struct mapped_device *md)
 
 	unlock_fs(md);
 
-	if (md->suspended_bdev) {
-		bdput(md->suspended_bdev);
-		md->suspended_bdev = NULL;
+	if (md->bdev) {
+		bdput(md->bdev);
+		md->bdev = NULL;
 	}
 
 	clear_bit(DMF_SUSPENDED, &md->flags);
-- 
cgit v1.2.3-59-g8ed1b


From 32a926da5a16c01a8213331e5764472ce2f14a8d Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:17 +0100
Subject: dm: always hold bdev reference

Fix a potential deadlock when creating multiple snapshots by holding a
reference to struct block_device for the whole lifecycle of every dm
device instead of obtaining it independently at each point it is needed.

bdget_disk() was called while the device was being suspended, in
dm_suspend().  However there could be other devices already suspended,
for example when creating additional snapshots of a device. bdget_disk()
can wait for IO and allocate memory resulting in waiting for the
already-suspended device - deadlock.

This patch changes the code so that it gets the reference to struct
block_device when struct mapped_device is allocated and initialized in
alloc_dev() where it is always OK to allocate memory or wait for I/O.
It drops the reference when it is destroyed in free_dev().  Thus there
is no call to bdget_disk() while any device is suspended.

Previously unlock_fs() was called only if bdev was held.  Now it is
called unconditionally, but the superfluous calls are harmless because
it returns immediately if the filesystem was not previously frozen.

This patch also now allows the device size to be changed in a
noflush suspend because the bdev is held.  This has no adverse effect.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c | 57 ++++++++++++++++-----------------------------------------
 1 file changed, 16 insertions(+), 41 deletions(-)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 1cfd9b72403d..5e06f1e6234f 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1180,6 +1180,10 @@ static struct mapped_device *alloc_dev(int minor)
 	if (!md->wq)
 		goto bad_thread;
 
+	md->bdev = bdget_disk(md->disk, 0);
+	if (!md->bdev)
+		goto bad_bdev;
+
 	/* Populate the mapping, nobody knows we exist yet */
 	spin_lock(&_minor_lock);
 	old_md = idr_replace(&_minor_idr, md, minor);
@@ -1189,6 +1193,8 @@ static struct mapped_device *alloc_dev(int minor)
 
 	return md;
 
+bad_bdev:
+	destroy_workqueue(md->wq);
 bad_thread:
 	put_disk(md->disk);
 bad_disk:
@@ -1214,10 +1220,8 @@ static void free_dev(struct mapped_device *md)
 {
 	int minor = MINOR(disk_devt(md->disk));
 
-	if (md->bdev) {
-		unlock_fs(md);
-		bdput(md->bdev);
-	}
+	unlock_fs(md);
+	bdput(md->bdev);
 	destroy_workqueue(md->wq);
 	mempool_destroy(md->tio_pool);
 	mempool_destroy(md->io_pool);
@@ -1277,8 +1281,7 @@ static int __bind(struct mapped_device *md, struct dm_table *t)
 	if (size != get_capacity(md->disk))
 		memset(&md->geometry, 0, sizeof(md->geometry));
 
-	if (md->bdev)
-		__set_size(md, size);
+	__set_size(md, size);
 
 	if (!size) {
 		dm_table_destroy(t);
@@ -1520,11 +1523,6 @@ int dm_swap_table(struct mapped_device *md, struct dm_table *table)
 	if (!dm_suspended(md))
 		goto out;
 
-	/* without bdev, the device size cannot be changed */
-	if (!md->bdev)
-		if (get_capacity(md->disk) != dm_table_get_size(table))
-			goto out;
-
 	__unbind(md);
 	r = __bind(md, table);
 
@@ -1552,9 +1550,6 @@ static int lock_fs(struct mapped_device *md)
 
 	set_bit(DMF_FROZEN, &md->flags);
 
-	/* don't bdput right now, we don't want the bdev
-	 * to go away while it is locked.
-	 */
 	return 0;
 }
 
@@ -1601,24 +1596,14 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags)
 	/* This does not get reverted if there's an error later. */
 	dm_table_presuspend_targets(map);
 
-	/* bdget() can stall if the pending I/Os are not flushed */
-	if (!noflush) {
-		md->bdev = bdget_disk(md->disk, 0);
-		if (!md->bdev) {
-			DMWARN("bdget failed in dm_suspend");
-			r = -ENOMEM;
+	/*
+	 * Flush I/O to the device. noflush supersedes do_lockfs,
+	 * because lock_fs() needs to flush I/Os.
+	 */
+	if (!noflush && do_lockfs) {
+		r = lock_fs(md);
+		if (r)
 			goto out;
-		}
-
-		/*
-		 * Flush I/O to the device. noflush supersedes do_lockfs,
-		 * because lock_fs() needs to flush I/Os.
-		 */
-		if (do_lockfs) {
-			r = lock_fs(md);
-			if (r)
-				goto out;
-		}
 	}
 
 	/*
@@ -1675,11 +1660,6 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags)
 	set_bit(DMF_SUSPENDED, &md->flags);
 
 out:
-	if (r && md->bdev) {
-		bdput(md->bdev);
-		md->bdev = NULL;
-	}
-
 	dm_table_put(map);
 
 out_unlock:
@@ -1708,11 +1688,6 @@ int dm_resume(struct mapped_device *md)
 
 	unlock_fs(md);
 
-	if (md->bdev) {
-		bdput(md->bdev);
-		md->bdev = NULL;
-	}
-
 	clear_bit(DMF_SUSPENDED, &md->flags);
 
 	dm_table_unplug_all(map);
-- 
cgit v1.2.3-59-g8ed1b


From 531fe96364f30879753d46c1f52ab839e12d2e5d Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:17 +0100
Subject: dm: make dm_flush return void

Make dm_flush return void.

The first error during flush is stored in md->barrier_error instead.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c | 17 ++++-------------
 1 file changed, 4 insertions(+), 13 deletions(-)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 5e06f1e6234f..e34d694ddd04 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1439,34 +1439,25 @@ static int dm_wait_for_completion(struct mapped_device *md, int interruptible)
 	return r;
 }
 
-static int dm_flush(struct mapped_device *md)
+static void dm_flush(struct mapped_device *md)
 {
 	dm_wait_for_completion(md, TASK_UNINTERRUPTIBLE);
-	return 0;
 }
 
 static void process_barrier(struct mapped_device *md, struct bio *bio)
 {
-	int error = dm_flush(md);
+	dm_flush(md);
 
-	if (unlikely(error)) {
-		bio_endio(bio, error);
-		return;
-	}
 	if (bio_empty_barrier(bio)) {
 		bio_endio(bio, 0);
 		return;
 	}
 
 	__split_and_process_bio(md, bio);
-
-	error = dm_flush(md);
-
-	if (!error && md->barrier_error)
-		error = md->barrier_error;
+	dm_flush(md);
 
 	if (md->barrier_error != DM_ENDIO_REQUEUE)
-		bio_endio(bio, error);
+		bio_endio(bio, md->barrier_error);
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 2761e95fe40ca0d01864310fa4d488d7c5e34e18 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:18 +0100
Subject: dm: process requeue in dm_wq_work

If barrier request was returned with DM_ENDIO_REQUEUE,
requeue it in dm_wq_work instead of dec_pending.

This allows us to correctly handle a situation when some targets
are asking for a requeue and other targets signal an error.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index e34d694ddd04..910bce85f443 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -536,9 +536,11 @@ static void dec_pending(struct dm_io *io, int error)
 			 * Target requested pushing back the I/O.
 			 */
 			spin_lock_irqsave(&md->deferred_lock, flags);
-			if (__noflush_suspending(md))
-				bio_list_add_head(&md->deferred, io->bio);
-			else
+			if (__noflush_suspending(md)) {
+				if (!bio_barrier(io->bio))
+					bio_list_add_head(&md->deferred,
+							  io->bio);
+			} else
 				/* noflush suspend was interrupted. */
 				io->error = -EIO;
 			spin_unlock_irqrestore(&md->deferred_lock, flags);
@@ -1458,6 +1460,11 @@ static void process_barrier(struct mapped_device *md, struct bio *bio)
 
 	if (md->barrier_error != DM_ENDIO_REQUEUE)
 		bio_endio(bio, md->barrier_error);
+	else {
+		spin_lock_irq(&md->deferred_lock);
+		bio_list_add_head(&md->deferred, bio);
+		spin_unlock_irq(&md->deferred_lock);
+	}
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 5aa2781d964e9835c441932110484bc454b5c207 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:18 +0100
Subject: dm: store only first barrier error

With the following patches, more than one error can occur during
processing.  Change md->barrier_error so that only the first one is
recorded and returned to the caller.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 910bce85f443..77972090abe5 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -555,7 +555,8 @@ static void dec_pending(struct dm_io *io, int error)
 			 * a per-device variable for error reporting.
 			 * Note that you can't touch the bio after end_io_acct
 			 */
-			md->barrier_error = io_error;
+			if (!md->barrier_error)
+				md->barrier_error = io_error;
 			end_io_acct(io);
 		} else {
 			end_io_acct(io);
@@ -867,7 +868,8 @@ static void __split_and_process_bio(struct mapped_device *md, struct bio *bio)
 		if (!bio_barrier(bio))
 			bio_io_error(bio);
 		else
-			md->barrier_error = -EIO;
+			if (!md->barrier_error)
+				md->barrier_error = -EIO;
 		return;
 	}
 
@@ -1448,16 +1450,15 @@ static void dm_flush(struct mapped_device *md)
 
 static void process_barrier(struct mapped_device *md, struct bio *bio)
 {
+	md->barrier_error = 0;
+
 	dm_flush(md);
 
-	if (bio_empty_barrier(bio)) {
-		bio_endio(bio, 0);
-		return;
+	if (!bio_empty_barrier(bio)) {
+		__split_and_process_bio(md, bio);
+		dm_flush(md);
 	}
 
-	__split_and_process_bio(md, bio);
-	dm_flush(md);
-
 	if (md->barrier_error != DM_ENDIO_REQUEUE)
 		bio_endio(bio, md->barrier_error);
 	else {
-- 
cgit v1.2.3-59-g8ed1b


From fdb9572b73abd008b80931c3b1f157dac3888bb9 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:19 +0100
Subject: dm: remove EOPNOTSUPP for barriers

If the underlying device doesn't support barriers and dm receives a
barrier, it waits until all requests on that device drain so it no
longer needs to report -EOPNOTSUPP to the caller.

This patch deals with the confusing situation when moving a volume from
one physical device to another triggers an EOPNOTSUPP on a volume that
didn't report it before.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 77972090abe5..8498dc4ce1f0 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -555,7 +555,7 @@ static void dec_pending(struct dm_io *io, int error)
 			 * a per-device variable for error reporting.
 			 * Note that you can't touch the bio after end_io_acct
 			 */
-			if (!md->barrier_error)
+			if (!md->barrier_error && io_error != -EOPNOTSUPP)
 				md->barrier_error = io_error;
 			end_io_acct(io);
 		} else {
-- 
cgit v1.2.3-59-g8ed1b


From 27eaa14975d8b53f0bad422e53cdf8e5f6dd44ec Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:20 +0100
Subject: dm: remove check that prevents mapping empty bios

Remove the check that the size of the cloned bio is not zero because a
subsequent patch needs to send zero-sized barriers down this path.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 8498dc4ce1f0..7d9ca7094337 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -637,11 +637,6 @@ static void __map_bio(struct dm_target *ti, struct bio *clone,
 	sector_t sector;
 	struct mapped_device *md;
 
-	/*
-	 * Sanity checks.
-	 */
-	BUG_ON(!clone->bi_size);
-
 	clone->bi_end_io = clone_endio;
 	clone->bi_private = tio;
 
-- 
cgit v1.2.3-59-g8ed1b


From f9ab94cee313746573b2d693bc2afb807ebb0998 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:20 +0100
Subject: dm: introduce num_flush_requests

Introduce num_flush_requests for a target to set to say how many flush
instructions (empty barriers) it wants to receive.  These are sent by
__clone_and_map_empty_barrier with map_info->flush_request going from 0
to (num_flush_requests - 1).

Old targets without flush support won't receive any flush requests.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c               | 39 +++++++++++++++++++++++++++++++++++++++
 include/linux/device-mapper.h | 11 +++++++++++
 2 files changed, 50 insertions(+)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 7d9ca7094337..badb7519cccb 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -750,6 +750,40 @@ static struct bio *clone_bio(struct bio *bio, sector_t sector,
 	return clone;
 }
 
+static void __flush_target(struct clone_info *ci, struct dm_target *ti,
+			  unsigned flush_nr)
+{
+	struct dm_target_io *tio = alloc_tio(ci->md);
+	struct bio *clone;
+
+	tio->io = ci->io;
+	tio->ti = ti;
+
+	memset(&tio->info, 0, sizeof(tio->info));
+	tio->info.flush_request = flush_nr;
+
+	clone = bio_alloc_bioset(GFP_NOIO, 0, ci->md->bs);
+	__bio_clone(clone, ci->bio);
+	clone->bi_destructor = dm_bio_destructor;
+
+	__map_bio(ti, clone, tio);
+}
+
+static int __clone_and_map_empty_barrier(struct clone_info *ci)
+{
+	unsigned target_nr = 0, flush_nr;
+	struct dm_target *ti;
+
+	while ((ti = dm_table_get_target(ci->map, target_nr++)))
+		for (flush_nr = 0; flush_nr < ti->num_flush_requests;
+		     flush_nr++)
+			__flush_target(ci, ti, flush_nr);
+
+	ci->sector_count = 0;
+
+	return 0;
+}
+
 static int __clone_and_map(struct clone_info *ci)
 {
 	struct bio *clone, *bio = ci->bio;
@@ -757,6 +791,9 @@ static int __clone_and_map(struct clone_info *ci)
 	sector_t len = 0, max;
 	struct dm_target_io *tio;
 
+	if (unlikely(bio_empty_barrier(bio)))
+		return __clone_and_map_empty_barrier(ci);
+
 	ti = dm_table_find_target(ci->map, ci->sector);
 	if (!dm_target_is_valid(ti))
 		return -EIO;
@@ -877,6 +914,8 @@ static void __split_and_process_bio(struct mapped_device *md, struct bio *bio)
 	ci.io->md = md;
 	ci.sector = bio->bi_sector;
 	ci.sector_count = bio_sectors(bio);
+	if (unlikely(bio_empty_barrier(bio)))
+		ci.sector_count = 1;
 	ci.idx = bio->bi_idx;
 
 	start_io_acct(ci.io);
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index 49c2362977fd..fc36a4d07723 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -21,6 +21,7 @@ typedef enum { STATUSTYPE_INFO, STATUSTYPE_TABLE } status_type_t;
 union map_info {
 	void *ptr;
 	unsigned long long ll;
+	unsigned flush_request;
 };
 
 /*
@@ -167,6 +168,16 @@ struct dm_target {
 	/* Always a power of 2 */
 	sector_t split_io;
 
+	/*
+	 * A number of zero-length barrier requests that will be submitted
+	 * to the target for the purpose of flushing cache.
+	 *
+	 * The request number will be placed in union map_info->flush_request.
+	 * It is a responsibility of the target driver to remap these requests
+	 * to the real underlying devices.
+	 */
+	unsigned num_flush_requests;
+
 	/*
 	 * These are automatically filled in by
 	 * dm_table_get_device.
-- 
cgit v1.2.3-59-g8ed1b


From 9015df24a8008d7bea2bd3df881783ebe0dcb9af Mon Sep 17 00:00:00 2001
From: Alasdair G Kergon <agk@redhat.com>
Date: Mon, 22 Jun 2009 10:12:21 +0100
Subject: dm: initialise tio in alloc_tio

Move repeated dm_target_io initialisation inside alloc_tio().

Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c | 33 +++++++++++++++------------------
 1 file changed, 15 insertions(+), 18 deletions(-)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index badb7519cccb..edf9f2467691 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -391,11 +391,6 @@ static void free_io(struct mapped_device *md, struct dm_io *io)
 	mempool_free(io, md->io_pool);
 }
 
-static struct dm_target_io *alloc_tio(struct mapped_device *md)
-{
-	return mempool_alloc(md->tio_pool, GFP_NOIO);
-}
-
 static void free_tio(struct mapped_device *md, struct dm_target_io *tio)
 {
 	mempool_free(tio, md->tio_pool);
@@ -750,16 +745,24 @@ static struct bio *clone_bio(struct bio *bio, sector_t sector,
 	return clone;
 }
 
-static void __flush_target(struct clone_info *ci, struct dm_target *ti,
-			  unsigned flush_nr)
+static struct dm_target_io *alloc_tio(struct clone_info *ci,
+				      struct dm_target *ti)
 {
-	struct dm_target_io *tio = alloc_tio(ci->md);
-	struct bio *clone;
+	struct dm_target_io *tio = mempool_alloc(ci->md->tio_pool, GFP_NOIO);
 
 	tio->io = ci->io;
 	tio->ti = ti;
-
 	memset(&tio->info, 0, sizeof(tio->info));
+
+	return tio;
+}
+
+static void __flush_target(struct clone_info *ci, struct dm_target *ti,
+			  unsigned flush_nr)
+{
+	struct dm_target_io *tio = alloc_tio(ci, ti);
+	struct bio *clone;
+
 	tio->info.flush_request = flush_nr;
 
 	clone = bio_alloc_bioset(GFP_NOIO, 0, ci->md->bs);
@@ -803,10 +806,7 @@ static int __clone_and_map(struct clone_info *ci)
 	/*
 	 * Allocate a target io object.
 	 */
-	tio = alloc_tio(ci->md);
-	tio->io = ci->io;
-	tio->ti = ti;
-	memset(&tio->info, 0, sizeof(tio->info));
+	tio = alloc_tio(ci, ti);
 
 	if (ci->sector_count <= max) {
 		/*
@@ -862,10 +862,7 @@ static int __clone_and_map(struct clone_info *ci)
 
 				max = max_io_len(ci->md, ci->sector, ti);
 
-				tio = alloc_tio(ci->md);
-				tio->io = ci->io;
-				tio->ti = ti;
-				memset(&tio->info, 0, sizeof(tio->info));
+				tio = alloc_tio(ci, ti);
 			}
 
 			len = min(remaining, max);
-- 
cgit v1.2.3-59-g8ed1b


From 52b1fd5a27c625c78373e024bf570af3c9d44a79 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:21 +0100
Subject: dm: send empty barriers to targets in dm_flush

Pass empty barrier flushes to the targets in dm_flush().

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index edf9f2467691..36142e947ffc 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -164,6 +164,9 @@ struct mapped_device {
 
 	/* sysfs handle */
 	struct kobject kobj;
+
+	/* zero-length barrier that will be cloned and submitted to targets */
+	struct bio barrier_bio;
 };
 
 #define MIN_IOS 256
@@ -1477,6 +1480,13 @@ static int dm_wait_for_completion(struct mapped_device *md, int interruptible)
 static void dm_flush(struct mapped_device *md)
 {
 	dm_wait_for_completion(md, TASK_UNINTERRUPTIBLE);
+
+	bio_init(&md->barrier_bio);
+	md->barrier_bio.bi_bdev = md->bdev;
+	md->barrier_bio.bi_rw = WRITE_BARRIER;
+	__split_and_process_bio(md, &md->barrier_bio);
+
+	dm_wait_for_completion(md, TASK_UNINTERRUPTIBLE);
 }
 
 static void process_barrier(struct mapped_device *md, struct bio *bio)
-- 
cgit v1.2.3-59-g8ed1b


From 433bcac5645508b71eab2710b6817c3ef937eba8 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:22 +0100
Subject: dm: linear support flush

Flush support for the linear target.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-linear.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/md/dm-linear.c b/drivers/md/dm-linear.c
index 79fb53e51c70..ecbb17421da4 100644
--- a/drivers/md/dm-linear.c
+++ b/drivers/md/dm-linear.c
@@ -53,6 +53,7 @@ static int linear_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		goto bad;
 	}
 
+	ti->num_flush_requests = 1;
 	ti->private = lc;
 	return 0;
 
@@ -81,7 +82,8 @@ static void linear_map_bio(struct dm_target *ti, struct bio *bio)
 	struct linear_c *lc = ti->private;
 
 	bio->bi_bdev = lc->dev->bdev;
-	bio->bi_sector = linear_map_sector(ti, bio->bi_sector);
+	if (bio_sectors(bio))
+		bio->bi_sector = linear_map_sector(ti, bio->bi_sector);
 }
 
 static int linear_map(struct dm_target *ti, struct bio *bio,
-- 
cgit v1.2.3-59-g8ed1b


From 374bf7e7f6cc38b0483351a2029a97910eadde1b Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:22 +0100
Subject: dm: stripe support flush

Flush support for the stripe target.

This sets ti->num_flush_requests to the number of stripes and
remaps individual flush requests to the appropriate stripe devices.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-stripe.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/md/dm-stripe.c b/drivers/md/dm-stripe.c
index 41569bc60abc..c64fe827a5f1 100644
--- a/drivers/md/dm-stripe.c
+++ b/drivers/md/dm-stripe.c
@@ -167,6 +167,7 @@ static int stripe_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	sc->stripes = stripes;
 	sc->stripe_width = width;
 	ti->split_io = chunk_size;
+	ti->num_flush_requests = stripes;
 
 	sc->chunk_mask = ((sector_t) chunk_size) - 1;
 	for (sc->chunk_shift = 0; chunk_size; sc->chunk_shift++)
@@ -211,10 +212,18 @@ static int stripe_map(struct dm_target *ti, struct bio *bio,
 		      union map_info *map_context)
 {
 	struct stripe_c *sc = (struct stripe_c *) ti->private;
+	sector_t offset, chunk;
+	uint32_t stripe;
 
-	sector_t offset = bio->bi_sector - ti->begin;
-	sector_t chunk = offset >> sc->chunk_shift;
-	uint32_t stripe = sector_div(chunk, sc->stripes);
+	if (unlikely(bio_empty_barrier(bio))) {
+		BUG_ON(map_context->flush_request >= sc->stripes);
+		bio->bi_bdev = sc->stripe[map_context->flush_request].dev->bdev;
+		return DM_MAPIO_REMAPPED;
+	}
+
+	offset = bio->bi_sector - ti->begin;
+	chunk = offset >> sc->chunk_shift;
+	stripe = sector_div(chunk, sc->stripes);
 
 	bio->bi_bdev = sc->stripe[stripe].dev->bdev;
 	bio->bi_sector = sc->stripe[stripe].physical_start +
-- 
cgit v1.2.3-59-g8ed1b


From 647c7db14ef9cacc4ccb3683e206b61f0de6dc2b Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:23 +0100
Subject: dm crypt: support flush

Flush support for dm-crypt target.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-crypt.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 53394e863c74..04db6c4004a8 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -1132,6 +1132,7 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		goto bad_crypt_queue;
 	}
 
+	ti->num_flush_requests = 1;
 	ti->private = cc;
 	return 0;
 
@@ -1189,6 +1190,13 @@ static int crypt_map(struct dm_target *ti, struct bio *bio,
 		     union map_info *map_context)
 {
 	struct dm_crypt_io *io;
+	struct crypt_config *cc;
+
+	if (unlikely(bio_empty_barrier(bio))) {
+		cc = ti->private;
+		bio->bi_bdev = cc->dev->bdev;
+		return DM_MAPIO_REMAPPED;
+	}
 
 	io = crypt_io_alloc(ti, bio, bio->bi_sector - ti->begin);
 
-- 
cgit v1.2.3-59-g8ed1b


From c927259e34e518d913d86f51c71b786a513f94d6 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:23 +0100
Subject: dm delay: support barriers

Flush support for dm-delay target.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-delay.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/md/dm-delay.c b/drivers/md/dm-delay.c
index 559dbb52bc85..8ad8a9044bbf 100644
--- a/drivers/md/dm-delay.c
+++ b/drivers/md/dm-delay.c
@@ -197,6 +197,7 @@ out:
 	mutex_init(&dc->timer_lock);
 	atomic_set(&dc->may_delay, 1);
 
+	ti->num_flush_requests = 1;
 	ti->private = dc;
 	return 0;
 
@@ -278,8 +279,9 @@ static int delay_map(struct dm_target *ti, struct bio *bio,
 
 	if ((bio_data_dir(bio) == WRITE) && (dc->dev_write)) {
 		bio->bi_bdev = dc->dev_write->bdev;
-		bio->bi_sector = dc->start_write +
-				 (bio->bi_sector - ti->begin);
+		if (bio_sectors(bio))
+			bio->bi_sector = dc->start_write +
+					 (bio->bi_sector - ti->begin);
 
 		return delay_bio(dc, dc->write_delay, bio);
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 8627921fa2ef6d40fd9b787e163ba3a9ff8f471d Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:24 +0100
Subject: dm mpath: support barriers

Flush support for dm-multipath target.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-mpath.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index 92bdcbb7c935..f1cf8f7449d6 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -835,6 +835,8 @@ static int multipath_ctr(struct dm_target *ti, unsigned int argc,
 		goto bad;
 	}
 
+	ti->num_flush_requests = 1;
+
 	return 0;
 
  bad:
-- 
cgit v1.2.3-59-g8ed1b


From 494b3ee7d4f69210def80aecce28d08c3f0755d5 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:25 +0100
Subject: dm snapshot: support barriers

Flush support for dm-snapshot target.

This patch just forwards the flush request to either the origin or the snapshot
device.  (It doesn't flush exception store metadata.)

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-snap.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index d73f17fc7778..d573165cd2b7 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -678,6 +678,7 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 
 	ti->private = s;
 	ti->split_io = s->store->chunk_size;
+	ti->num_flush_requests = 1;
 
 	return 0;
 
@@ -1030,6 +1031,11 @@ static int snapshot_map(struct dm_target *ti, struct bio *bio,
 	chunk_t chunk;
 	struct dm_snap_pending_exception *pe = NULL;
 
+	if (unlikely(bio_empty_barrier(bio))) {
+		bio->bi_bdev = s->store->cow->bdev;
+		return DM_MAPIO_REMAPPED;
+	}
+
 	chunk = sector_to_chunk(s->store, bio->bi_sector);
 
 	/* Full snapshots are not usable */
@@ -1338,6 +1344,8 @@ static int origin_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	}
 
 	ti->private = dev;
+	ti->num_flush_requests = 1;
+
 	return 0;
 }
 
@@ -1353,6 +1361,9 @@ static int origin_map(struct dm_target *ti, struct bio *bio,
 	struct dm_dev *dev = ti->private;
 	bio->bi_bdev = dev->bdev;
 
+	if (unlikely(bio_empty_barrier(bio)))
+		return DM_MAPIO_REMAPPED;
+
 	/* Only tell snapshots if this is a write */
 	return (bio_rw(bio) == WRITE) ? do_origin(dev, bio) : DM_MAPIO_REMAPPED;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 5af443a7e1c0864100cc44525a9821aa2a2f4719 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:25 +0100
Subject: dm io: record eopnotsupp

Add another field, eopnotsupp_bits. It is subset of error_bits, representing
regions that returned -EOPNOTSUPP.  (The bit is set in both error_bits and
eopnotsupp_bits).

This value will be used in further patches.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-io.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c
index e73aabd61cd7..a89f41f00757 100644
--- a/drivers/md/dm-io.c
+++ b/drivers/md/dm-io.c
@@ -22,6 +22,7 @@ struct dm_io_client {
 /* FIXME: can we shrink this ? */
 struct io {
 	unsigned long error_bits;
+	unsigned long eopnotsupp_bits;
 	atomic_t count;
 	struct task_struct *sleeper;
 	struct dm_io_client *client;
@@ -107,8 +108,11 @@ static inline unsigned bio_get_region(struct bio *bio)
  *---------------------------------------------------------------*/
 static void dec_count(struct io *io, unsigned int region, int error)
 {
-	if (error)
+	if (error) {
 		set_bit(region, &io->error_bits);
+		if (error == -EOPNOTSUPP)
+			set_bit(region, &io->eopnotsupp_bits);
+	}
 
 	if (atomic_dec_and_test(&io->count)) {
 		if (io->sleeper)
@@ -361,6 +365,7 @@ static int sync_io(struct dm_io_client *client, unsigned int num_regions,
 	}
 
 	io.error_bits = 0;
+	io.eopnotsupp_bits = 0;
 	atomic_set(&io.count, 1); /* see dispatch_io() */
 	io.sleeper = current;
 	io.client = client;
@@ -397,6 +402,7 @@ static int async_io(struct dm_io_client *client, unsigned int num_regions,
 
 	io = mempool_alloc(client->pool, GFP_NOIO);
 	io->error_bits = 0;
+	io->eopnotsupp_bits = 0;
 	atomic_set(&io->count, 1); /* see dispatch_io() */
 	io->sleeper = NULL;
 	io->client = client;
-- 
cgit v1.2.3-59-g8ed1b


From 51aa322849581f1a73594e48ea0df63f914ee6a2 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:26 +0100
Subject: dm io: retry after barrier error

If -EOPNOTSUPP was returned and the request was a barrier request, retry it
without barrier.

Retry all regions for now. Barriers are submitted only for one-region requests,
so it doesn't matter.  (In the future, retries can be limited to the actual
regions that failed.)

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-io.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c
index a89f41f00757..3a2e6a2f8bdd 100644
--- a/drivers/md/dm-io.c
+++ b/drivers/md/dm-io.c
@@ -364,6 +364,7 @@ static int sync_io(struct dm_io_client *client, unsigned int num_regions,
 		return -EIO;
 	}
 
+retry:
 	io.error_bits = 0;
 	io.eopnotsupp_bits = 0;
 	atomic_set(&io.count, 1); /* see dispatch_io() */
@@ -382,6 +383,11 @@ static int sync_io(struct dm_io_client *client, unsigned int num_regions,
 	}
 	set_current_state(TASK_RUNNING);
 
+	if (io.eopnotsupp_bits && (rw & (1 << BIO_RW_BARRIER))) {
+		rw &= ~(1 << BIO_RW_BARRIER);
+		goto retry;
+	}
+
 	if (error_bits)
 		*error_bits = io.error_bits;
 
-- 
cgit v1.2.3-59-g8ed1b


From 2bd023452592e5f5cf90dd426cc39b7632b15b76 Mon Sep 17 00:00:00 2001
From: Mikulas Patocka <mpatocka@redhat.com>
Date: Mon, 22 Jun 2009 10:12:26 +0100
Subject: dm snapshot: use barrier when writing exception store

Send barrier requests when updating the exception area.

Exception area updates need to be ordered w.r.t. data writes, so that
the writes are not reordered in hardware disk cache.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-snap-persistent.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/md/dm-snap-persistent.c b/drivers/md/dm-snap-persistent.c
index 2662a41337e7..6e3fe4f14934 100644
--- a/drivers/md/dm-snap-persistent.c
+++ b/drivers/md/dm-snap-persistent.c
@@ -636,7 +636,7 @@ static void persistent_commit_exception(struct dm_exception_store *store,
 	/*
 	 * Commit exceptions to disk.
 	 */
-	if (ps->valid && area_io(ps, WRITE))
+	if (ps->valid && area_io(ps, WRITE_BARRIER))
 		ps->valid = 0;
 
 	/*
-- 
cgit v1.2.3-59-g8ed1b


From 02ab823fd1a27d193bda06b74fdad685a20a3e5e Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Mon, 22 Jun 2009 10:12:27 +0100
Subject: dm mpath: add start_io and nr_bytes to path selectors

This patch makes two additions to the dm path selector interface for
dynamic load balancers:
  o a new hook, start_io()
  o a new parameter 'nr_bytes' to select_path()/start_io()/end_io()
    to pass the size of the I/O

start_io() is called when a target driver actually submits I/O
to the selected path.
Path selectors can use it to start accounting of the I/O.
(e.g. counting the number of in-flight I/Os.)
The start_io hook is based on the patch posted by Stefan Bader:
https://www.redhat.com/archives/dm-devel/2005-October/msg00050.html

nr_bytes, the size of the I/O, is so path selectors can take the
size of the I/O into account when deciding which path to use.
dm-service-time uses it to estimate service time, for example.
(Added the nr_bytes member to dm_mpath_io instead of using existing
 details.bi_size, since request-based dm patch deletes it.)

Signed-off-by: Stefan Bader <stefan.bader@canonical.com>
Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-mpath.c         | 28 ++++++++++++++++++----------
 drivers/md/dm-path-selector.h |  8 ++++++--
 drivers/md/dm-round-robin.c   |  2 +-
 3 files changed, 25 insertions(+), 13 deletions(-)

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index f1cf8f7449d6..890c0e8ed13e 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -101,6 +101,7 @@ struct multipath {
 struct dm_mpath_io {
 	struct pgpath *pgpath;
 	struct dm_bio_details details;
+	size_t nr_bytes;
 };
 
 typedef int (*action_fn) (struct pgpath *pgpath);
@@ -244,11 +245,12 @@ static void __switch_pg(struct multipath *m, struct pgpath *pgpath)
 	m->pg_init_count = 0;
 }
 
-static int __choose_path_in_pg(struct multipath *m, struct priority_group *pg)
+static int __choose_path_in_pg(struct multipath *m, struct priority_group *pg,
+			       size_t nr_bytes)
 {
 	struct dm_path *path;
 
-	path = pg->ps.type->select_path(&pg->ps, &m->repeat_count);
+	path = pg->ps.type->select_path(&pg->ps, &m->repeat_count, nr_bytes);
 	if (!path)
 		return -ENXIO;
 
@@ -260,7 +262,7 @@ static int __choose_path_in_pg(struct multipath *m, struct priority_group *pg)
 	return 0;
 }
 
-static void __choose_pgpath(struct multipath *m)
+static void __choose_pgpath(struct multipath *m, size_t nr_bytes)
 {
 	struct priority_group *pg;
 	unsigned bypassed = 1;
@@ -272,12 +274,12 @@ static void __choose_pgpath(struct multipath *m)
 	if (m->next_pg) {
 		pg = m->next_pg;
 		m->next_pg = NULL;
-		if (!__choose_path_in_pg(m, pg))
+		if (!__choose_path_in_pg(m, pg, nr_bytes))
 			return;
 	}
 
 	/* Don't change PG until it has no remaining paths */
-	if (m->current_pg && !__choose_path_in_pg(m, m->current_pg))
+	if (m->current_pg && !__choose_path_in_pg(m, m->current_pg, nr_bytes))
 		return;
 
 	/*
@@ -289,7 +291,7 @@ static void __choose_pgpath(struct multipath *m)
 		list_for_each_entry(pg, &m->priority_groups, list) {
 			if (pg->bypassed == bypassed)
 				continue;
-			if (!__choose_path_in_pg(m, pg))
+			if (!__choose_path_in_pg(m, pg, nr_bytes))
 				return;
 		}
 	} while (bypassed--);
@@ -320,6 +322,7 @@ static int map_io(struct multipath *m, struct bio *bio,
 		  struct dm_mpath_io *mpio, unsigned was_queued)
 {
 	int r = DM_MAPIO_REMAPPED;
+	size_t nr_bytes = bio->bi_size;
 	unsigned long flags;
 	struct pgpath *pgpath;
 
@@ -328,7 +331,7 @@ static int map_io(struct multipath *m, struct bio *bio,
 	/* Do we need to select a new pgpath? */
 	if (!m->current_pgpath ||
 	    (!m->queue_io && (m->repeat_count && --m->repeat_count == 0)))
-		__choose_pgpath(m);
+		__choose_pgpath(m, nr_bytes);
 
 	pgpath = m->current_pgpath;
 
@@ -353,6 +356,11 @@ static int map_io(struct multipath *m, struct bio *bio,
 		r = -EIO;	/* Failed */
 
 	mpio->pgpath = pgpath;
+	mpio->nr_bytes = nr_bytes;
+
+	if (r == DM_MAPIO_REMAPPED && pgpath->pg->ps.type->start_io)
+		pgpath->pg->ps.type->start_io(&pgpath->pg->ps, &pgpath->path,
+					      nr_bytes);
 
 	spin_unlock_irqrestore(&m->lock, flags);
 
@@ -431,7 +439,7 @@ static void process_queued_ios(struct work_struct *work)
 		goto out;
 
 	if (!m->current_pgpath)
-		__choose_pgpath(m);
+		__choose_pgpath(m, 0);
 
 	pgpath = m->current_pgpath;
 
@@ -1209,7 +1217,7 @@ static int multipath_end_io(struct dm_target *ti, struct bio *bio,
 	if (pgpath) {
 		ps = &pgpath->pg->ps;
 		if (ps->type->end_io)
-			ps->type->end_io(ps, &pgpath->path);
+			ps->type->end_io(ps, &pgpath->path, mpio->nr_bytes);
 	}
 	if (r != DM_ENDIO_INCOMPLETE)
 		mempool_free(mpio, m->mpio_pool);
@@ -1425,7 +1433,7 @@ static int multipath_ioctl(struct dm_target *ti, unsigned int cmd,
 	spin_lock_irqsave(&m->lock, flags);
 
 	if (!m->current_pgpath)
-		__choose_pgpath(m);
+		__choose_pgpath(m, 0);
 
 	if (m->current_pgpath) {
 		bdev = m->current_pgpath->path.dev->bdev;
diff --git a/drivers/md/dm-path-selector.h b/drivers/md/dm-path-selector.h
index 27357b85d73d..e7d1fa8b0459 100644
--- a/drivers/md/dm-path-selector.h
+++ b/drivers/md/dm-path-selector.h
@@ -56,7 +56,8 @@ struct path_selector_type {
 	 * the path fails.
 	 */
 	struct dm_path *(*select_path) (struct path_selector *ps,
-				     unsigned *repeat_count);
+					unsigned *repeat_count,
+					size_t nr_bytes);
 
 	/*
 	 * Notify the selector that a path has failed.
@@ -75,7 +76,10 @@ struct path_selector_type {
 	int (*status) (struct path_selector *ps, struct dm_path *path,
 		       status_type_t type, char *result, unsigned int maxlen);
 
-	int (*end_io) (struct path_selector *ps, struct dm_path *path);
+	int (*start_io) (struct path_selector *ps, struct dm_path *path,
+			 size_t nr_bytes);
+	int (*end_io) (struct path_selector *ps, struct dm_path *path,
+		       size_t nr_bytes);
 };
 
 /* Register a path selector */
diff --git a/drivers/md/dm-round-robin.c b/drivers/md/dm-round-robin.c
index cdfbf65b28cb..24752f449bef 100644
--- a/drivers/md/dm-round-robin.c
+++ b/drivers/md/dm-round-robin.c
@@ -161,7 +161,7 @@ static int rr_reinstate_path(struct path_selector *ps, struct dm_path *p)
 }
 
 static struct dm_path *rr_select_path(struct path_selector *ps,
-				   unsigned *repeat_count)
+				      unsigned *repeat_count, size_t nr_bytes)
 {
 	struct selector *s = (struct selector *) ps->context;
 	struct path_info *pi = NULL;
-- 
cgit v1.2.3-59-g8ed1b


From fd5e033908b7b743b5650790f196761dd930f988 Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Mon, 22 Jun 2009 10:12:27 +0100
Subject: dm mpath: add queue length load balancer

This patch adds a dynamic load balancer, dm-queue-length, which
balances the number of in-flight I/Os across the paths.

The code is based on the patch posted by Stefan Bader:
https://www.redhat.com/archives/dm-devel/2005-October/msg00050.html

Signed-off-by: Stefan Bader <stefan.bader@canonical.com>
Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 Documentation/device-mapper/dm-queue-length.txt |  39 ++++
 drivers/md/Kconfig                              |   9 +
 drivers/md/Makefile                             |   1 +
 drivers/md/dm-queue-length.c                    | 263 ++++++++++++++++++++++++
 4 files changed, 312 insertions(+)
 create mode 100644 Documentation/device-mapper/dm-queue-length.txt
 create mode 100644 drivers/md/dm-queue-length.c

diff --git a/Documentation/device-mapper/dm-queue-length.txt b/Documentation/device-mapper/dm-queue-length.txt
new file mode 100644
index 000000000000..f4db2562175c
--- /dev/null
+++ b/Documentation/device-mapper/dm-queue-length.txt
@@ -0,0 +1,39 @@
+dm-queue-length
+===============
+
+dm-queue-length is a path selector module for device-mapper targets,
+which selects a path with the least number of in-flight I/Os.
+The path selector name is 'queue-length'.
+
+Table parameters for each path: [<repeat_count>]
+	<repeat_count>: The number of I/Os to dispatch using the selected
+			path before switching to the next path.
+			If not given, internal default is used. To check
+			the default value, see the activated table.
+
+Status for each path: <status> <fail-count> <in-flight>
+	<status>: 'A' if the path is active, 'F' if the path is failed.
+	<fail-count>: The number of path failures.
+	<in-flight>: The number of in-flight I/Os on the path.
+
+
+Algorithm
+=========
+
+dm-queue-length increments/decrements 'in-flight' when an I/O is
+dispatched/completed respectively.
+dm-queue-length selects a path with the minimum 'in-flight'.
+
+
+Examples
+========
+In case that 2 paths (sda and sdb) are used with repeat_count == 128.
+
+# echo "0 10 multipath 0 0 1 1 queue-length 0 2 1 8:0 128 8:16 128" \
+  dmsetup create test
+#
+# dmsetup table
+test: 0 10 multipath 0 0 1 1 queue-length 0 2 1 8:0 128 8:16 128
+#
+# dmsetup status
+test: 0 10 multipath 2 0 0 0 1 1 E 0 2 1 8:0 A 0 0 8:16 A 0 0
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig
index 36e0675be9f7..3b311d273346 100644
--- a/drivers/md/Kconfig
+++ b/drivers/md/Kconfig
@@ -249,6 +249,15 @@ config DM_MULTIPATH
 	---help---
 	  Allow volume managers to support multipath hardware.
 
+config DM_MULTIPATH_QL
+	tristate "I/O Path Selector based on the number of in-flight I/Os"
+	depends on DM_MULTIPATH
+	---help---
+	  This path selector is a dynamic load balancer which selects
+	  the path with the least number of in-flight I/Os.
+
+	  If unsure, say N.
+
 config DM_DELAY
 	tristate "I/O delaying target (EXPERIMENTAL)"
 	depends on BLK_DEV_DM && EXPERIMENTAL
diff --git a/drivers/md/Makefile b/drivers/md/Makefile
index 45cc5951d928..ff9f545dd516 100644
--- a/drivers/md/Makefile
+++ b/drivers/md/Makefile
@@ -36,6 +36,7 @@ obj-$(CONFIG_BLK_DEV_DM)	+= dm-mod.o
 obj-$(CONFIG_DM_CRYPT)		+= dm-crypt.o
 obj-$(CONFIG_DM_DELAY)		+= dm-delay.o
 obj-$(CONFIG_DM_MULTIPATH)	+= dm-multipath.o dm-round-robin.o
+obj-$(CONFIG_DM_MULTIPATH_QL)	+= dm-queue-length.o
 obj-$(CONFIG_DM_SNAPSHOT)	+= dm-snapshot.o
 obj-$(CONFIG_DM_MIRROR)		+= dm-mirror.o dm-log.o dm-region-hash.o
 obj-$(CONFIG_DM_ZERO)		+= dm-zero.o
diff --git a/drivers/md/dm-queue-length.c b/drivers/md/dm-queue-length.c
new file mode 100644
index 000000000000..f92b6cea9d9c
--- /dev/null
+++ b/drivers/md/dm-queue-length.c
@@ -0,0 +1,263 @@
+/*
+ * Copyright (C) 2004-2005 IBM Corp.  All Rights Reserved.
+ * Copyright (C) 2006-2009 NEC Corporation.
+ *
+ * dm-queue-length.c
+ *
+ * Module Author: Stefan Bader, IBM
+ * Modified by: Kiyoshi Ueda, NEC
+ *
+ * This file is released under the GPL.
+ *
+ * queue-length path selector - choose a path with the least number of
+ * in-flight I/Os.
+ */
+
+#include "dm.h"
+#include "dm-path-selector.h"
+
+#include <linux/slab.h>
+#include <linux/ctype.h>
+#include <linux/errno.h>
+#include <linux/module.h>
+#include <asm/atomic.h>
+
+#define DM_MSG_PREFIX	"multipath queue-length"
+#define QL_MIN_IO	128
+#define QL_VERSION	"0.1.0"
+
+struct selector {
+	struct list_head	valid_paths;
+	struct list_head	failed_paths;
+};
+
+struct path_info {
+	struct list_head	list;
+	struct dm_path		*path;
+	unsigned		repeat_count;
+	atomic_t		qlen;	/* the number of in-flight I/Os */
+};
+
+static struct selector *alloc_selector(void)
+{
+	struct selector *s = kmalloc(sizeof(*s), GFP_KERNEL);
+
+	if (s) {
+		INIT_LIST_HEAD(&s->valid_paths);
+		INIT_LIST_HEAD(&s->failed_paths);
+	}
+
+	return s;
+}
+
+static int ql_create(struct path_selector *ps, unsigned argc, char **argv)
+{
+	struct selector *s = alloc_selector();
+
+	if (!s)
+		return -ENOMEM;
+
+	ps->context = s;
+	return 0;
+}
+
+static void ql_free_paths(struct list_head *paths)
+{
+	struct path_info *pi, *next;
+
+	list_for_each_entry_safe(pi, next, paths, list) {
+		list_del(&pi->list);
+		kfree(pi);
+	}
+}
+
+static void ql_destroy(struct path_selector *ps)
+{
+	struct selector *s = ps->context;
+
+	ql_free_paths(&s->valid_paths);
+	ql_free_paths(&s->failed_paths);
+	kfree(s);
+	ps->context = NULL;
+}
+
+static int ql_status(struct path_selector *ps, struct dm_path *path,
+		     status_type_t type, char *result, unsigned maxlen)
+{
+	unsigned sz = 0;
+	struct path_info *pi;
+
+	/* When called with NULL path, return selector status/args. */
+	if (!path)
+		DMEMIT("0 ");
+	else {
+		pi = path->pscontext;
+
+		switch (type) {
+		case STATUSTYPE_INFO:
+			DMEMIT("%d ", atomic_read(&pi->qlen));
+			break;
+		case STATUSTYPE_TABLE:
+			DMEMIT("%u ", pi->repeat_count);
+			break;
+		}
+	}
+
+	return sz;
+}
+
+static int ql_add_path(struct path_selector *ps, struct dm_path *path,
+		       int argc, char **argv, char **error)
+{
+	struct selector *s = ps->context;
+	struct path_info *pi;
+	unsigned repeat_count = QL_MIN_IO;
+
+	/*
+	 * Arguments: [<repeat_count>]
+	 * 	<repeat_count>: The number of I/Os before switching path.
+	 * 			If not given, default (QL_MIN_IO) is used.
+	 */
+	if (argc > 1) {
+		*error = "queue-length ps: incorrect number of arguments";
+		return -EINVAL;
+	}
+
+	if ((argc == 1) && (sscanf(argv[0], "%u", &repeat_count) != 1)) {
+		*error = "queue-length ps: invalid repeat count";
+		return -EINVAL;
+	}
+
+	/* Allocate the path information structure */
+	pi = kmalloc(sizeof(*pi), GFP_KERNEL);
+	if (!pi) {
+		*error = "queue-length ps: Error allocating path information";
+		return -ENOMEM;
+	}
+
+	pi->path = path;
+	pi->repeat_count = repeat_count;
+	atomic_set(&pi->qlen, 0);
+
+	path->pscontext = pi;
+
+	list_add_tail(&pi->list, &s->valid_paths);
+
+	return 0;
+}
+
+static void ql_fail_path(struct path_selector *ps, struct dm_path *path)
+{
+	struct selector *s = ps->context;
+	struct path_info *pi = path->pscontext;
+
+	list_move(&pi->list, &s->failed_paths);
+}
+
+static int ql_reinstate_path(struct path_selector *ps, struct dm_path *path)
+{
+	struct selector *s = ps->context;
+	struct path_info *pi = path->pscontext;
+
+	list_move_tail(&pi->list, &s->valid_paths);
+
+	return 0;
+}
+
+/*
+ * Select a path having the minimum number of in-flight I/Os
+ */
+static struct dm_path *ql_select_path(struct path_selector *ps,
+				      unsigned *repeat_count, size_t nr_bytes)
+{
+	struct selector *s = ps->context;
+	struct path_info *pi = NULL, *best = NULL;
+
+	if (list_empty(&s->valid_paths))
+		return NULL;
+
+	/* Change preferred (first in list) path to evenly balance. */
+	list_move_tail(s->valid_paths.next, &s->valid_paths);
+
+	list_for_each_entry(pi, &s->valid_paths, list) {
+		if (!best ||
+		    (atomic_read(&pi->qlen) < atomic_read(&best->qlen)))
+			best = pi;
+
+		if (!atomic_read(&best->qlen))
+			break;
+	}
+
+	if (!best)
+		return NULL;
+
+	*repeat_count = best->repeat_count;
+
+	return best->path;
+}
+
+static int ql_start_io(struct path_selector *ps, struct dm_path *path,
+		       size_t nr_bytes)
+{
+	struct path_info *pi = path->pscontext;
+
+	atomic_inc(&pi->qlen);
+
+	return 0;
+}
+
+static int ql_end_io(struct path_selector *ps, struct dm_path *path,
+		     size_t nr_bytes)
+{
+	struct path_info *pi = path->pscontext;
+
+	atomic_dec(&pi->qlen);
+
+	return 0;
+}
+
+static struct path_selector_type ql_ps = {
+	.name		= "queue-length",
+	.module		= THIS_MODULE,
+	.table_args	= 1,
+	.info_args	= 1,
+	.create		= ql_create,
+	.destroy	= ql_destroy,
+	.status		= ql_status,
+	.add_path	= ql_add_path,
+	.fail_path	= ql_fail_path,
+	.reinstate_path	= ql_reinstate_path,
+	.select_path	= ql_select_path,
+	.start_io	= ql_start_io,
+	.end_io		= ql_end_io,
+};
+
+static int __init dm_ql_init(void)
+{
+	int r = dm_register_path_selector(&ql_ps);
+
+	if (r < 0)
+		DMERR("register failed %d", r);
+
+	DMINFO("version " QL_VERSION " loaded");
+
+	return r;
+}
+
+static void __exit dm_ql_exit(void)
+{
+	int r = dm_unregister_path_selector(&ql_ps);
+
+	if (r < 0)
+		DMERR("unregister failed %d", r);
+}
+
+module_init(dm_ql_init);
+module_exit(dm_ql_exit);
+
+MODULE_AUTHOR("Stefan Bader <Stefan.Bader at de.ibm.com>");
+MODULE_DESCRIPTION(
+	"(C) Copyright IBM Corp. 2004,2005   All Rights Reserved.\n"
+	DM_NAME " path selector to balance the number of in-flight I/Os"
+);
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-59-g8ed1b


From f392ba889b019602976082bfe7bf486c2594f85c Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Mon, 22 Jun 2009 10:12:28 +0100
Subject: dm mpath: add service time load balancer

This patch adds a service time oriented dynamic load balancer,
dm-service-time, which selects the path with the shortest estimated
service time for the incoming I/O.
The service time is estimated by dividing the in-flight I/O size
by a performance value of each path.

The performance value can be given as a table argument at the table
loading time.  If no performance value is given, all paths are
considered equal.

Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 Documentation/device-mapper/dm-service-time.txt |  91 +++++++
 drivers/md/Kconfig                              |  10 +
 drivers/md/Makefile                             |   1 +
 drivers/md/dm-service-time.c                    | 339 ++++++++++++++++++++++++
 4 files changed, 441 insertions(+)
 create mode 100644 Documentation/device-mapper/dm-service-time.txt
 create mode 100644 drivers/md/dm-service-time.c

diff --git a/Documentation/device-mapper/dm-service-time.txt b/Documentation/device-mapper/dm-service-time.txt
new file mode 100644
index 000000000000..7d00668e97bb
--- /dev/null
+++ b/Documentation/device-mapper/dm-service-time.txt
@@ -0,0 +1,91 @@
+dm-service-time
+===============
+
+dm-service-time is a path selector module for device-mapper targets,
+which selects a path with the shortest estimated service time for
+the incoming I/O.
+
+The service time for each path is estimated by dividing the total size
+of in-flight I/Os on a path with the performance value of the path.
+The performance value is a relative throughput value among all paths
+in a path-group, and it can be specified as a table argument.
+
+The path selector name is 'service-time'.
+
+Table parameters for each path: [<repeat_count> [<relative_throughput>]]
+	<repeat_count>: The number of I/Os to dispatch using the selected
+			path before switching to the next path.
+			If not given, internal default is used.  To check
+			the default value, see the activated table.
+	<relative_throughput>: The relative throughput value of the path
+			among all paths in the path-group.
+			The valid range is 0-100.
+			If not given, minimum value '1' is used.
+			If '0' is given, the path isn't selected while
+			other paths having a positive value are available.
+
+Status for each path: <status> <fail-count> <in-flight-size> \
+		      <relative_throughput>
+	<status>: 'A' if the path is active, 'F' if the path is failed.
+	<fail-count>: The number of path failures.
+	<in-flight-size>: The size of in-flight I/Os on the path.
+	<relative_throughput>: The relative throughput value of the path
+			among all paths in the path-group.
+
+
+Algorithm
+=========
+
+dm-service-time adds the I/O size to 'in-flight-size' when the I/O is
+dispatched and substracts when completed.
+Basically, dm-service-time selects a path having minimum service time
+which is calculated by:
+
+	('in-flight-size' + 'size-of-incoming-io') / 'relative_throughput'
+
+However, some optimizations below are used to reduce the calculation
+as much as possible.
+
+	1. If the paths have the same 'relative_throughput', skip
+	   the division and just compare the 'in-flight-size'.
+
+	2. If the paths have the same 'in-flight-size', skip the division
+	   and just compare the 'relative_throughput'.
+
+	3. If some paths have non-zero 'relative_throughput' and others
+	   have zero 'relative_throughput', ignore those paths with zero
+	   'relative_throughput'.
+
+If such optimizations can't be applied, calculate service time, and
+compare service time.
+If calculated service time is equal, the path having maximum
+'relative_throughput' may be better.  So compare 'relative_throughput'
+then.
+
+
+Examples
+========
+In case that 2 paths (sda and sdb) are used with repeat_count == 128
+and sda has an average throughput 1GB/s and sdb has 4GB/s,
+'relative_throughput' value may be '1' for sda and '4' for sdb.
+
+# echo "0 10 multipath 0 0 1 1 service-time 0 2 2 8:0 128 1 8:16 128 4" \
+  dmsetup create test
+#
+# dmsetup table
+test: 0 10 multipath 0 0 1 1 service-time 0 2 2 8:0 128 1 8:16 128 4
+#
+# dmsetup status
+test: 0 10 multipath 2 0 0 0 1 1 E 0 2 2 8:0 A 0 0 1 8:16 A 0 0 4
+
+
+Or '2' for sda and '8' for sdb would be also true.
+
+# echo "0 10 multipath 0 0 1 1 service-time 0 2 2 8:0 128 2 8:16 128 8" \
+  dmsetup create test
+#
+# dmsetup table
+test: 0 10 multipath 0 0 1 1 service-time 0 2 2 8:0 128 2 8:16 128 8
+#
+# dmsetup status
+test: 0 10 multipath 2 0 0 0 1 1 E 0 2 2 8:0 A 0 0 2 8:16 A 0 0 8
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig
index 3b311d273346..09f93fa68912 100644
--- a/drivers/md/Kconfig
+++ b/drivers/md/Kconfig
@@ -258,6 +258,16 @@ config DM_MULTIPATH_QL
 
 	  If unsure, say N.
 
+config DM_MULTIPATH_ST
+	tristate "I/O Path Selector based on the service time"
+	depends on DM_MULTIPATH
+	---help---
+	  This path selector is a dynamic load balancer which selects
+	  the path expected to complete the incoming I/O in the shortest
+	  time.
+
+	  If unsure, say N.
+
 config DM_DELAY
 	tristate "I/O delaying target (EXPERIMENTAL)"
 	depends on BLK_DEV_DM && EXPERIMENTAL
diff --git a/drivers/md/Makefile b/drivers/md/Makefile
index ff9f545dd516..dade52f60733 100644
--- a/drivers/md/Makefile
+++ b/drivers/md/Makefile
@@ -37,6 +37,7 @@ obj-$(CONFIG_DM_CRYPT)		+= dm-crypt.o
 obj-$(CONFIG_DM_DELAY)		+= dm-delay.o
 obj-$(CONFIG_DM_MULTIPATH)	+= dm-multipath.o dm-round-robin.o
 obj-$(CONFIG_DM_MULTIPATH_QL)	+= dm-queue-length.o
+obj-$(CONFIG_DM_MULTIPATH_ST)	+= dm-service-time.o
 obj-$(CONFIG_DM_SNAPSHOT)	+= dm-snapshot.o
 obj-$(CONFIG_DM_MIRROR)		+= dm-mirror.o dm-log.o dm-region-hash.o
 obj-$(CONFIG_DM_ZERO)		+= dm-zero.o
diff --git a/drivers/md/dm-service-time.c b/drivers/md/dm-service-time.c
new file mode 100644
index 000000000000..cfa668f46c40
--- /dev/null
+++ b/drivers/md/dm-service-time.c
@@ -0,0 +1,339 @@
+/*
+ * Copyright (C) 2007-2009 NEC Corporation.  All Rights Reserved.
+ *
+ * Module Author: Kiyoshi Ueda
+ *
+ * This file is released under the GPL.
+ *
+ * Throughput oriented path selector.
+ */
+
+#include "dm.h"
+#include "dm-path-selector.h"
+
+#define DM_MSG_PREFIX	"multipath service-time"
+#define ST_MIN_IO	1
+#define ST_MAX_RELATIVE_THROUGHPUT	100
+#define ST_MAX_RELATIVE_THROUGHPUT_SHIFT	7
+#define ST_MAX_INFLIGHT_SIZE	((size_t)-1 >> ST_MAX_RELATIVE_THROUGHPUT_SHIFT)
+#define ST_VERSION	"0.2.0"
+
+struct selector {
+	struct list_head valid_paths;
+	struct list_head failed_paths;
+};
+
+struct path_info {
+	struct list_head list;
+	struct dm_path *path;
+	unsigned repeat_count;
+	unsigned relative_throughput;
+	atomic_t in_flight_size;	/* Total size of in-flight I/Os */
+};
+
+static struct selector *alloc_selector(void)
+{
+	struct selector *s = kmalloc(sizeof(*s), GFP_KERNEL);
+
+	if (s) {
+		INIT_LIST_HEAD(&s->valid_paths);
+		INIT_LIST_HEAD(&s->failed_paths);
+	}
+
+	return s;
+}
+
+static int st_create(struct path_selector *ps, unsigned argc, char **argv)
+{
+	struct selector *s = alloc_selector();
+
+	if (!s)
+		return -ENOMEM;
+
+	ps->context = s;
+	return 0;
+}
+
+static void free_paths(struct list_head *paths)
+{
+	struct path_info *pi, *next;
+
+	list_for_each_entry_safe(pi, next, paths, list) {
+		list_del(&pi->list);
+		kfree(pi);
+	}
+}
+
+static void st_destroy(struct path_selector *ps)
+{
+	struct selector *s = ps->context;
+
+	free_paths(&s->valid_paths);
+	free_paths(&s->failed_paths);
+	kfree(s);
+	ps->context = NULL;
+}
+
+static int st_status(struct path_selector *ps, struct dm_path *path,
+		     status_type_t type, char *result, unsigned maxlen)
+{
+	unsigned sz = 0;
+	struct path_info *pi;
+
+	if (!path)
+		DMEMIT("0 ");
+	else {
+		pi = path->pscontext;
+
+		switch (type) {
+		case STATUSTYPE_INFO:
+			DMEMIT("%d %u ", atomic_read(&pi->in_flight_size),
+			       pi->relative_throughput);
+			break;
+		case STATUSTYPE_TABLE:
+			DMEMIT("%u %u ", pi->repeat_count,
+			       pi->relative_throughput);
+			break;
+		}
+	}
+
+	return sz;
+}
+
+static int st_add_path(struct path_selector *ps, struct dm_path *path,
+		       int argc, char **argv, char **error)
+{
+	struct selector *s = ps->context;
+	struct path_info *pi;
+	unsigned repeat_count = ST_MIN_IO;
+	unsigned relative_throughput = 1;
+
+	/*
+	 * Arguments: [<repeat_count> [<relative_throughput>]]
+	 * 	<repeat_count>: The number of I/Os before switching path.
+	 * 			If not given, default (ST_MIN_IO) is used.
+	 * 	<relative_throughput>: The relative throughput value of
+	 *			the path among all paths in the path-group.
+	 * 			The valid range: 0-<ST_MAX_RELATIVE_THROUGHPUT>
+	 *			If not given, minimum value '1' is used.
+	 *			If '0' is given, the path isn't selected while
+	 * 			other paths having a positive value are
+	 * 			available.
+	 */
+	if (argc > 2) {
+		*error = "service-time ps: incorrect number of arguments";
+		return -EINVAL;
+	}
+
+	if (argc && (sscanf(argv[0], "%u", &repeat_count) != 1)) {
+		*error = "service-time ps: invalid repeat count";
+		return -EINVAL;
+	}
+
+	if ((argc == 2) &&
+	    (sscanf(argv[1], "%u", &relative_throughput) != 1 ||
+	     relative_throughput > ST_MAX_RELATIVE_THROUGHPUT)) {
+		*error = "service-time ps: invalid relative_throughput value";
+		return -EINVAL;
+	}
+
+	/* allocate the path */
+	pi = kmalloc(sizeof(*pi), GFP_KERNEL);
+	if (!pi) {
+		*error = "service-time ps: Error allocating path context";
+		return -ENOMEM;
+	}
+
+	pi->path = path;
+	pi->repeat_count = repeat_count;
+	pi->relative_throughput = relative_throughput;
+	atomic_set(&pi->in_flight_size, 0);
+
+	path->pscontext = pi;
+
+	list_add_tail(&pi->list, &s->valid_paths);
+
+	return 0;
+}
+
+static void st_fail_path(struct path_selector *ps, struct dm_path *path)
+{
+	struct selector *s = ps->context;
+	struct path_info *pi = path->pscontext;
+
+	list_move(&pi->list, &s->failed_paths);
+}
+
+static int st_reinstate_path(struct path_selector *ps, struct dm_path *path)
+{
+	struct selector *s = ps->context;
+	struct path_info *pi = path->pscontext;
+
+	list_move_tail(&pi->list, &s->valid_paths);
+
+	return 0;
+}
+
+/*
+ * Compare the estimated service time of 2 paths, pi1 and pi2,
+ * for the incoming I/O.
+ *
+ * Returns:
+ * < 0 : pi1 is better
+ * 0   : no difference between pi1 and pi2
+ * > 0 : pi2 is better
+ *
+ * Description:
+ * Basically, the service time is estimated by:
+ *     ('pi->in-flight-size' + 'incoming') / 'pi->relative_throughput'
+ * To reduce the calculation, some optimizations are made.
+ * (See comments inline)
+ */
+static int st_compare_load(struct path_info *pi1, struct path_info *pi2,
+			   size_t incoming)
+{
+	size_t sz1, sz2, st1, st2;
+
+	sz1 = atomic_read(&pi1->in_flight_size);
+	sz2 = atomic_read(&pi2->in_flight_size);
+
+	/*
+	 * Case 1: Both have same throughput value. Choose less loaded path.
+	 */
+	if (pi1->relative_throughput == pi2->relative_throughput)
+		return sz1 - sz2;
+
+	/*
+	 * Case 2a: Both have same load. Choose higher throughput path.
+	 * Case 2b: One path has no throughput value. Choose the other one.
+	 */
+	if (sz1 == sz2 ||
+	    !pi1->relative_throughput || !pi2->relative_throughput)
+		return pi2->relative_throughput - pi1->relative_throughput;
+
+	/*
+	 * Case 3: Calculate service time. Choose faster path.
+	 *         Service time using pi1:
+	 *             st1 = (sz1 + incoming) / pi1->relative_throughput
+	 *         Service time using pi2:
+	 *             st2 = (sz2 + incoming) / pi2->relative_throughput
+	 *
+	 *         To avoid the division, transform the expression to use
+	 *         multiplication.
+	 *         Because ->relative_throughput > 0 here, if st1 < st2,
+	 *         the expressions below are the same meaning:
+	 *             (sz1 + incoming) / pi1->relative_throughput <
+	 *                 (sz2 + incoming) / pi2->relative_throughput
+	 *             (sz1 + incoming) * pi2->relative_throughput <
+	 *                 (sz2 + incoming) * pi1->relative_throughput
+	 *         So use the later one.
+	 */
+	sz1 += incoming;
+	sz2 += incoming;
+	if (unlikely(sz1 >= ST_MAX_INFLIGHT_SIZE ||
+		     sz2 >= ST_MAX_INFLIGHT_SIZE)) {
+		/*
+		 * Size may be too big for multiplying pi->relative_throughput
+		 * and overflow.
+		 * To avoid the overflow and mis-selection, shift down both.
+		 */
+		sz1 >>= ST_MAX_RELATIVE_THROUGHPUT_SHIFT;
+		sz2 >>= ST_MAX_RELATIVE_THROUGHPUT_SHIFT;
+	}
+	st1 = sz1 * pi2->relative_throughput;
+	st2 = sz2 * pi1->relative_throughput;
+	if (st1 != st2)
+		return st1 - st2;
+
+	/*
+	 * Case 4: Service time is equal. Choose higher throughput path.
+	 */
+	return pi2->relative_throughput - pi1->relative_throughput;
+}
+
+static struct dm_path *st_select_path(struct path_selector *ps,
+				      unsigned *repeat_count, size_t nr_bytes)
+{
+	struct selector *s = ps->context;
+	struct path_info *pi = NULL, *best = NULL;
+
+	if (list_empty(&s->valid_paths))
+		return NULL;
+
+	/* Change preferred (first in list) path to evenly balance. */
+	list_move_tail(s->valid_paths.next, &s->valid_paths);
+
+	list_for_each_entry(pi, &s->valid_paths, list)
+		if (!best || (st_compare_load(pi, best, nr_bytes) < 0))
+			best = pi;
+
+	if (!best)
+		return NULL;
+
+	*repeat_count = best->repeat_count;
+
+	return best->path;
+}
+
+static int st_start_io(struct path_selector *ps, struct dm_path *path,
+		       size_t nr_bytes)
+{
+	struct path_info *pi = path->pscontext;
+
+	atomic_add(nr_bytes, &pi->in_flight_size);
+
+	return 0;
+}
+
+static int st_end_io(struct path_selector *ps, struct dm_path *path,
+		     size_t nr_bytes)
+{
+	struct path_info *pi = path->pscontext;
+
+	atomic_sub(nr_bytes, &pi->in_flight_size);
+
+	return 0;
+}
+
+static struct path_selector_type st_ps = {
+	.name		= "service-time",
+	.module		= THIS_MODULE,
+	.table_args	= 2,
+	.info_args	= 2,
+	.create		= st_create,
+	.destroy	= st_destroy,
+	.status		= st_status,
+	.add_path	= st_add_path,
+	.fail_path	= st_fail_path,
+	.reinstate_path	= st_reinstate_path,
+	.select_path	= st_select_path,
+	.start_io	= st_start_io,
+	.end_io		= st_end_io,
+};
+
+static int __init dm_st_init(void)
+{
+	int r = dm_register_path_selector(&st_ps);
+
+	if (r < 0)
+		DMERR("register failed %d", r);
+
+	DMINFO("version " ST_VERSION " loaded");
+
+	return r;
+}
+
+static void __exit dm_st_exit(void)
+{
+	int r = dm_unregister_path_selector(&st_ps);
+
+	if (r < 0)
+		DMERR("unregister failed %d", r);
+}
+
+module_init(dm_st_init);
+module_exit(dm_st_exit);
+
+MODULE_DESCRIPTION(DM_NAME " throughput oriented path selector");
+MODULE_AUTHOR("Kiyoshi Ueda <k-ueda@ct.jp.nec.com>");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-59-g8ed1b


From 1b6da754594e6e26c24e6fbc1a34f9c03e4617a3 Mon Sep 17 00:00:00 2001
From: Jonthan Brassow <jbrassow@redhat.com>
Date: Mon, 22 Jun 2009 10:12:29 +0100
Subject: dm table: improve warning message when devices not freed before
 destruction

Report any devices forgotten to be freed before a table is destroyed.

Signed-off-by: Jonathan Brassow <jbrassow@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-table.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 0e2210c0c168..af1ceae2582a 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -267,6 +267,8 @@ static void free_devices(struct list_head *devices)
 	list_for_each_safe(tmp, next, devices) {
 		struct dm_dev_internal *dd =
 		    list_entry(tmp, struct dm_dev_internal, list);
+		DMWARN("dm_table_destroy: dm_put_device call missing for %s",
+		       dd->dm_dev.name);
 		kfree(dd);
 	}
 }
@@ -296,12 +298,8 @@ void dm_table_destroy(struct dm_table *t)
 	vfree(t->highs);
 
 	/* free the device list */
-	if (t->devices.next != &t->devices) {
-		DMWARN("devices still present during destroy: "
-		       "dm_table_remove_device calls missing");
-
+	if (t->devices.next != &t->devices)
 		free_devices(&t->devices);
-	}
 
 	kfree(t);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 486d220fe4909b5745c4faa67faddd30a707abe2 Mon Sep 17 00:00:00 2001
From: Peter Rajnoha <prajnoha@redhat.com>
Date: Mon, 22 Jun 2009 10:12:29 +0100
Subject: dm: sysfs add suspended attribute

Add a file named 'suspended' to each device-mapper device directory in
sysfs.  It holds the value 1 while the device is suspended.  Otherwise
it holds 0.

Signed-off-by: Peter Rajnoha <prajnoha@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-sysfs.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/md/dm-sysfs.c b/drivers/md/dm-sysfs.c
index a2a45e6c7c8b..4b045903a4e2 100644
--- a/drivers/md/dm-sysfs.c
+++ b/drivers/md/dm-sysfs.c
@@ -57,12 +57,21 @@ static ssize_t dm_attr_uuid_show(struct mapped_device *md, char *buf)
 	return strlen(buf);
 }
 
+static ssize_t dm_attr_suspended_show(struct mapped_device *md, char *buf)
+{
+	sprintf(buf, "%d\n", dm_suspended(md));
+
+	return strlen(buf);
+}
+
 static DM_ATTR_RO(name);
 static DM_ATTR_RO(uuid);
+static DM_ATTR_RO(suspended);
 
 static struct attribute *dm_attrs[] = {
 	&dm_attr_name.attr,
 	&dm_attr_uuid.attr,
+	&dm_attr_suspended.attr,
 	NULL,
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 60935eb21d3c5bac79618000f38f92c249d153c4 Mon Sep 17 00:00:00 2001
From: Milan Broz <mbroz@redhat.com>
Date: Mon, 22 Jun 2009 10:12:30 +0100
Subject: dm ioctl: support cookies for udev

Add support for passing a 32 bit "cookie" into the kernel with the
DM_SUSPEND, DM_DEV_RENAME and DM_DEV_REMOVE ioctls.  The (unsigned)
value of this cookie is returned to userspace alongside the uevents
issued by these ioctls in the variable DM_COOKIE.

This means the userspace process issuing these ioctls can be notified
by udev after udev has completed any actions triggered.

To minimise the interface extension, we pass the cookie into the
kernel in the event_nr field which is otherwise unused when calling
these ioctls.  Incrementing the version number allows userspace to
determine in advance whether or not the kernel supports the cookie.
If the kernel does support this but userspace does not, there should
be no impact as the new variable will just get ignored.

Signed-off-by: Milan Broz <mbroz@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-ioctl.c    | 14 ++++++++++----
 drivers/md/dm.c          | 25 +++++++++++++++++++------
 drivers/md/dm.h          |  3 ++-
 include/linux/dm-ioctl.h | 14 ++++++++++++--
 4 files changed, 43 insertions(+), 13 deletions(-)

diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c
index 1128d3fba797..1c871736f48c 100644
--- a/drivers/md/dm-ioctl.c
+++ b/drivers/md/dm-ioctl.c
@@ -276,7 +276,7 @@ retry:
 	up_write(&_hash_lock);
 }
 
-static int dm_hash_rename(const char *old, const char *new)
+static int dm_hash_rename(uint32_t cookie, const char *old, const char *new)
 {
 	char *new_name, *old_name;
 	struct hash_cell *hc;
@@ -333,7 +333,7 @@ static int dm_hash_rename(const char *old, const char *new)
 		dm_table_put(table);
 	}
 
-	dm_kobject_uevent(hc->md);
+	dm_kobject_uevent(hc->md, KOBJ_CHANGE, cookie);
 
 	dm_put(hc->md);
 	up_write(&_hash_lock);
@@ -680,6 +680,9 @@ static int dev_remove(struct dm_ioctl *param, size_t param_size)
 
 	__hash_remove(hc);
 	up_write(&_hash_lock);
+
+	dm_kobject_uevent(md, KOBJ_REMOVE, param->event_nr);
+
 	dm_put(md);
 	param->data_size = 0;
 	return 0;
@@ -715,7 +718,7 @@ static int dev_rename(struct dm_ioctl *param, size_t param_size)
 		return r;
 
 	param->data_size = 0;
-	return dm_hash_rename(param->name, new_name);
+	return dm_hash_rename(param->event_nr, param->name, new_name);
 }
 
 static int dev_set_geometry(struct dm_ioctl *param, size_t param_size)
@@ -842,8 +845,11 @@ static int do_resume(struct dm_ioctl *param)
 	if (dm_suspended(md))
 		r = dm_resume(md);
 
-	if (!r)
+
+	if (!r) {
+		dm_kobject_uevent(md, KOBJ_CHANGE, param->event_nr);
 		r = __dev_status(md, param);
+	}
 
 	dm_put(md);
 	return r;
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 36142e947ffc..a9210bb594e7 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -24,6 +24,13 @@
 
 #define DM_MSG_PREFIX "core"
 
+/*
+ * Cookies are numeric values sent with CHANGE and REMOVE
+ * uevents while resuming, removing or renaming the device.
+ */
+#define DM_COOKIE_ENV_VAR_NAME "DM_COOKIE"
+#define DM_COOKIE_LENGTH 24
+
 static const char *_name = DM_NAME;
 
 static unsigned int major = 0;
@@ -1731,11 +1738,7 @@ int dm_resume(struct mapped_device *md)
 	clear_bit(DMF_SUSPENDED, &md->flags);
 
 	dm_table_unplug_all(map);
-
-	dm_kobject_uevent(md);
-
 	r = 0;
-
 out:
 	dm_table_put(map);
 	mutex_unlock(&md->suspend_lock);
@@ -1746,9 +1749,19 @@ out:
 /*-----------------------------------------------------------------
  * Event notification.
  *---------------------------------------------------------------*/
-void dm_kobject_uevent(struct mapped_device *md)
+void dm_kobject_uevent(struct mapped_device *md, enum kobject_action action,
+		       unsigned cookie)
 {
-	kobject_uevent(&disk_to_dev(md->disk)->kobj, KOBJ_CHANGE);
+	char udev_cookie[DM_COOKIE_LENGTH];
+	char *envp[] = { udev_cookie, NULL };
+
+	if (!cookie)
+		kobject_uevent(&disk_to_dev(md->disk)->kobj, action);
+	else {
+		snprintf(udev_cookie, DM_COOKIE_LENGTH, "%s=%u",
+			 DM_COOKIE_ENV_VAR_NAME, cookie);
+		kobject_uevent_env(&disk_to_dev(md->disk)->kobj, action, envp);
+	}
 }
 
 uint32_t dm_next_uevent_seq(struct mapped_device *md)
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index a31506d93e91..b5935c610c44 100644
--- a/drivers/md/dm.h
+++ b/drivers/md/dm.h
@@ -92,7 +92,8 @@ void dm_stripe_exit(void);
 int dm_open_count(struct mapped_device *md);
 int dm_lock_for_deletion(struct mapped_device *md);
 
-void dm_kobject_uevent(struct mapped_device *md);
+void dm_kobject_uevent(struct mapped_device *md, enum kobject_action action,
+		       unsigned cookie);
 
 int dm_kcopyd_init(void);
 void dm_kcopyd_exit(void);
diff --git a/include/linux/dm-ioctl.h b/include/linux/dm-ioctl.h
index 48e44ee2b466..2ab84c83c31a 100644
--- a/include/linux/dm-ioctl.h
+++ b/include/linux/dm-ioctl.h
@@ -123,6 +123,16 @@ struct dm_ioctl {
 	__u32 target_count;	/* in/out */
 	__s32 open_count;	/* out */
 	__u32 flags;		/* in/out */
+
+	/*
+	 * event_nr holds either the event number (input and output) or the
+	 * udev cookie value (input only).
+	 * The DM_DEV_WAIT ioctl takes an event number as input.
+	 * The DM_SUSPEND, DM_DEV_REMOVE and DM_DEV_RENAME ioctls
+	 * use the field as a cookie to return in the DM_COOKIE
+	 * variable with the uevents they issue.
+	 * For output, the ioctls return the event number, not the cookie.
+	 */
 	__u32 event_nr;      	/* in/out */
 	__u32 padding;
 
@@ -256,9 +266,9 @@ enum {
 #define DM_DEV_SET_GEOMETRY	_IOWR(DM_IOCTL, DM_DEV_SET_GEOMETRY_CMD, struct dm_ioctl)
 
 #define DM_VERSION_MAJOR	4
-#define DM_VERSION_MINOR	14
+#define DM_VERSION_MINOR	15
 #define DM_VERSION_PATCHLEVEL	0
-#define DM_VERSION_EXTRA	"-ioctl (2008-04-23)"
+#define DM_VERSION_EXTRA	"-ioctl (2009-04-01)"
 
 /* Status bits */
 #define DM_READONLY_FLAG	(1 << 0) /* In/Out */
-- 
cgit v1.2.3-59-g8ed1b


From 02acc3a4fa0a6c2a5ccc4fb722b55fb710265882 Mon Sep 17 00:00:00 2001
From: Mike Snitzer <snitzer@redhat.com>
Date: Mon, 22 Jun 2009 10:12:30 +0100
Subject: dm table: ensure targets are aligned to logical_block_size

Ensure I/O is aligned to the logical block size of target devices.

Rename check_device_area() to device_area_is_valid() for clarity and
establish the device limits including the logical block size prior to
calling it.

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-table.c | 58 ++++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 44 insertions(+), 14 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index af1ceae2582a..535fdaf2473d 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -383,16 +383,45 @@ static void close_dev(struct dm_dev_internal *d, struct mapped_device *md)
 /*
  * If possible, this checks an area of a destination device is valid.
  */
-static int check_device_area(struct dm_dev_internal *dd, sector_t start,
-			     sector_t len)
+static int device_area_is_valid(struct dm_target *ti, struct block_device *bdev,
+			     sector_t start, sector_t len)
 {
-	sector_t dev_size = i_size_read(dd->dm_dev.bdev->bd_inode) >>
-			    SECTOR_SHIFT;
+	sector_t dev_size = i_size_read(bdev->bd_inode) >> SECTOR_SHIFT;
+	unsigned short logical_block_size_sectors =
+		ti->limits.logical_block_size >> SECTOR_SHIFT;
+	char b[BDEVNAME_SIZE];
 
 	if (!dev_size)
 		return 1;
 
-	return ((start < dev_size) && (len <= (dev_size - start)));
+	if ((start >= dev_size) || (start + len > dev_size)) {
+		DMWARN("%s: %s too small for target",
+		       dm_device_name(ti->table->md), bdevname(bdev, b));
+		return 0;
+	}
+
+	if (logical_block_size_sectors <= 1)
+		return 1;
+
+	if (start & (logical_block_size_sectors - 1)) {
+		DMWARN("%s: start=%llu not aligned to h/w "
+		       "logical block size %hu of %s",
+		       dm_device_name(ti->table->md),
+		       (unsigned long long)start,
+		       ti->limits.logical_block_size, bdevname(bdev, b));
+		return 0;
+	}
+
+	if (len & (logical_block_size_sectors - 1)) {
+		DMWARN("%s: len=%llu not aligned to h/w "
+		       "logical block size %hu of %s",
+		       dm_device_name(ti->table->md),
+		       (unsigned long long)len,
+		       ti->limits.logical_block_size, bdevname(bdev, b));
+		return 0;
+	}
+
+	return 1;
 }
 
 /*
@@ -478,14 +507,7 @@ static int __table_get_device(struct dm_table *t, struct dm_target *ti,
 	}
 	atomic_inc(&dd->count);
 
-	if (!check_device_area(dd, start, len)) {
-		DMWARN("device %s too small for target", path);
-		dm_put_device(ti, &dd->dm_dev);
-		return -EINVAL;
-	}
-
 	*result = &dd->dm_dev;
-
 	return 0;
 }
 
@@ -554,8 +576,16 @@ int dm_get_device(struct dm_target *ti, const char *path, sector_t start,
 	int r = __table_get_device(ti->table, ti, path,
 				   start, len, mode, result);
 
-	if (!r)
-		dm_set_device_limits(ti, (*result)->bdev);
+	if (r)
+		return r;
+
+	dm_set_device_limits(ti, (*result)->bdev);
+
+	if (!device_area_is_valid(ti, (*result)->bdev, start, len)) {
+		dm_put_device(ti, *result);
+		*result = NULL;
+		return -EINVAL;
+	}
 
 	return r;
 }
-- 
cgit v1.2.3-59-g8ed1b


From be6d4305db093ad1cc623f7dd3d2470b7bd73fa4 Mon Sep 17 00:00:00 2001
From: Mike Snitzer <snitzer@redhat.com>
Date: Mon, 22 Jun 2009 10:12:31 +0100
Subject: dm table: validate device logical_block_size

Impose necessary and sufficient conditions on a devices's table such
that any incoming bio which respects its logical_block_size can be
processed successfully.

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-table.c | 69 +++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 69 insertions(+)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 535fdaf2473d..e3bcfb8b15a1 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -724,6 +724,71 @@ static void check_for_valid_limits(struct io_restrictions *rs)
 		rs->bounce_pfn = -1;
 }
 
+/*
+ * Impose necessary and sufficient conditions on a devices's table such
+ * that any incoming bio which respects its logical_block_size can be
+ * processed successfully.  If it falls across the boundary between
+ * two or more targets, the size of each piece it gets split into must
+ * be compatible with the logical_block_size of the target processing it.
+ */
+static int validate_hardware_logical_block_alignment(struct dm_table *table)
+{
+	/*
+	 * This function uses arithmetic modulo the logical_block_size
+	 * (in units of 512-byte sectors).
+	 */
+	unsigned short device_logical_block_size_sects =
+		table->limits.logical_block_size >> SECTOR_SHIFT;
+
+	/*
+	 * Offset of the start of the next table entry, mod logical_block_size.
+	 */
+	unsigned short next_target_start = 0;
+
+	/*
+	 * Given an aligned bio that extends beyond the end of a
+	 * target, how many sectors must the next target handle?
+	 */
+	unsigned short remaining = 0;
+
+	struct dm_target *uninitialized_var(ti);
+	unsigned i = 0;
+
+	/*
+	 * Check each entry in the table in turn.
+	 */
+	while (i < dm_table_get_num_targets(table)) {
+		ti = dm_table_get_target(table, i++);
+
+		/*
+		 * If the remaining sectors fall entirely within this
+		 * table entry are they compatible with its logical_block_size?
+		 */
+		if (remaining < ti->len &&
+		    remaining & ((ti->limits.logical_block_size >>
+				  SECTOR_SHIFT) - 1))
+			break;	/* Error */
+
+		next_target_start =
+		    (unsigned short) ((next_target_start + ti->len) &
+				      (device_logical_block_size_sects - 1));
+		remaining = next_target_start ?
+		    device_logical_block_size_sects - next_target_start : 0;
+	}
+
+	if (remaining) {
+		DMWARN("%s: table line %u (start sect %llu len %llu) "
+		       "not aligned to hardware logical block size %hu",
+		       dm_device_name(table->md), i,
+		       (unsigned long long) ti->begin,
+		       (unsigned long long) ti->len,
+		       table->limits.logical_block_size);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 int dm_table_add_target(struct dm_table *t, const char *type,
 			sector_t start, sector_t len, char *params)
 {
@@ -823,6 +888,10 @@ int dm_table_complete(struct dm_table *t)
 
 	check_for_valid_limits(&t->limits);
 
+	r = validate_hardware_logical_block_alignment(t);
+	if (r)
+		return r;
+
 	/* how many indexes will the btree have ? */
 	leaf_nodes = dm_div_up(t->num_targets, KEYS_PER_NODE);
 	t->depth = 1 + int_log(leaf_nodes, CHILDREN_PER_NODE);
-- 
cgit v1.2.3-59-g8ed1b


From 5ab97588fb266187b88d1ad893251c94388f18ba Mon Sep 17 00:00:00 2001
From: Mike Snitzer <snitzer@redhat.com>
Date: Mon, 22 Jun 2009 10:12:32 +0100
Subject: dm table: replace struct io_restrictions with struct queue_limits

Use blk_stack_limits() to stack block limits (including topology) rather
than duplicate the equivalent within Device Mapper.

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-table.c         | 138 +++++++++++++-----------------------------
 include/linux/device-mapper.h |  16 +----
 2 files changed, 45 insertions(+), 109 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index e3bcfb8b15a1..41ec2bf9fbe9 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -66,7 +66,7 @@ struct dm_table {
 	 * These are optimistic limits taken from all the
 	 * targets, some targets will need smaller limits.
 	 */
-	struct io_restrictions limits;
+	struct queue_limits limits;
 
 	/* events get handed up using this callback */
 	void (*event_fn)(void *);
@@ -88,43 +88,6 @@ static unsigned int int_log(unsigned int n, unsigned int base)
 	return result;
 }
 
-/*
- * Returns the minimum that is _not_ zero, unless both are zero.
- */
-#define min_not_zero(l, r) (l == 0) ? r : ((r == 0) ? l : min(l, r))
-
-/*
- * Combine two io_restrictions, always taking the lower value.
- */
-static void combine_restrictions_low(struct io_restrictions *lhs,
-				     struct io_restrictions *rhs)
-{
-	lhs->max_sectors =
-		min_not_zero(lhs->max_sectors, rhs->max_sectors);
-
-	lhs->max_phys_segments =
-		min_not_zero(lhs->max_phys_segments, rhs->max_phys_segments);
-
-	lhs->max_hw_segments =
-		min_not_zero(lhs->max_hw_segments, rhs->max_hw_segments);
-
-	lhs->logical_block_size = max(lhs->logical_block_size,
-				      rhs->logical_block_size);
-
-	lhs->max_segment_size =
-		min_not_zero(lhs->max_segment_size, rhs->max_segment_size);
-
-	lhs->max_hw_sectors =
-		min_not_zero(lhs->max_hw_sectors, rhs->max_hw_sectors);
-
-	lhs->seg_boundary_mask =
-		min_not_zero(lhs->seg_boundary_mask, rhs->seg_boundary_mask);
-
-	lhs->bounce_pfn = min_not_zero(lhs->bounce_pfn, rhs->bounce_pfn);
-
-	lhs->no_cluster |= rhs->no_cluster;
-}
-
 /*
  * Calculate the index of the child node of the n'th node k'th key.
  */
@@ -511,10 +474,14 @@ static int __table_get_device(struct dm_table *t, struct dm_target *ti,
 	return 0;
 }
 
+/*
+ * Returns the minimum that is _not_ zero, unless both are zero.
+ */
+#define min_not_zero(l, r) (l == 0) ? r : ((r == 0) ? l : min(l, r))
+
 void dm_set_device_limits(struct dm_target *ti, struct block_device *bdev)
 {
 	struct request_queue *q = bdev_get_queue(bdev);
-	struct io_restrictions *rs = &ti->limits;
 	char b[BDEVNAME_SIZE];
 
 	if (unlikely(!q)) {
@@ -523,15 +490,9 @@ void dm_set_device_limits(struct dm_target *ti, struct block_device *bdev)
 		return;
 	}
 
-	/*
-	 * Combine the device limits low.
-	 *
-	 * FIXME: if we move an io_restriction struct
-	 *        into q this would just be a call to
-	 *        combine_restrictions_low()
-	 */
-	rs->max_sectors =
-		min_not_zero(rs->max_sectors, queue_max_sectors(q));
+	if (blk_stack_limits(&ti->limits, &q->limits, 0) < 0)
+		DMWARN("%s: target device %s is misaligned",
+		       dm_device_name(ti->table->md), bdevname(bdev, b));
 
 	/*
 	 * Check if merge fn is supported.
@@ -540,33 +501,9 @@ void dm_set_device_limits(struct dm_target *ti, struct block_device *bdev)
 	 */
 
 	if (q->merge_bvec_fn && !ti->type->merge)
-		rs->max_sectors =
-			min_not_zero(rs->max_sectors,
+		ti->limits.max_sectors =
+			min_not_zero(ti->limits.max_sectors,
 				     (unsigned int) (PAGE_SIZE >> 9));
-
-	rs->max_phys_segments =
-		min_not_zero(rs->max_phys_segments,
-			     queue_max_phys_segments(q));
-
-	rs->max_hw_segments =
-		min_not_zero(rs->max_hw_segments, queue_max_hw_segments(q));
-
-	rs->logical_block_size = max(rs->logical_block_size,
-				     queue_logical_block_size(q));
-
-	rs->max_segment_size =
-		min_not_zero(rs->max_segment_size, queue_max_segment_size(q));
-
-	rs->max_hw_sectors =
-		min_not_zero(rs->max_hw_sectors, queue_max_hw_sectors(q));
-
-	rs->seg_boundary_mask =
-		min_not_zero(rs->seg_boundary_mask,
-			     queue_segment_boundary(q));
-
-	rs->bounce_pfn = min_not_zero(rs->bounce_pfn, queue_bounce_pfn(q));
-
-	rs->no_cluster |= !test_bit(QUEUE_FLAG_CLUSTER, &q->queue_flags);
 }
 EXPORT_SYMBOL_GPL(dm_set_device_limits);
 
@@ -704,24 +641,32 @@ int dm_split_args(int *argc, char ***argvp, char *input)
 	return 0;
 }
 
-static void check_for_valid_limits(struct io_restrictions *rs)
+static void init_valid_queue_limits(struct queue_limits *limits)
 {
-	if (!rs->max_sectors)
-		rs->max_sectors = SAFE_MAX_SECTORS;
-	if (!rs->max_hw_sectors)
-		rs->max_hw_sectors = SAFE_MAX_SECTORS;
-	if (!rs->max_phys_segments)
-		rs->max_phys_segments = MAX_PHYS_SEGMENTS;
-	if (!rs->max_hw_segments)
-		rs->max_hw_segments = MAX_HW_SEGMENTS;
-	if (!rs->logical_block_size)
-		rs->logical_block_size = 1 << SECTOR_SHIFT;
-	if (!rs->max_segment_size)
-		rs->max_segment_size = MAX_SEGMENT_SIZE;
-	if (!rs->seg_boundary_mask)
-		rs->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
-	if (!rs->bounce_pfn)
-		rs->bounce_pfn = -1;
+	if (!limits->max_sectors)
+		limits->max_sectors = SAFE_MAX_SECTORS;
+	if (!limits->max_hw_sectors)
+		limits->max_hw_sectors = SAFE_MAX_SECTORS;
+	if (!limits->max_phys_segments)
+		limits->max_phys_segments = MAX_PHYS_SEGMENTS;
+	if (!limits->max_hw_segments)
+		limits->max_hw_segments = MAX_HW_SEGMENTS;
+	if (!limits->logical_block_size)
+		limits->logical_block_size = 1 << SECTOR_SHIFT;
+	if (!limits->physical_block_size)
+		limits->physical_block_size = 1 << SECTOR_SHIFT;
+	if (!limits->io_min)
+		limits->io_min = 1 << SECTOR_SHIFT;
+	if (!limits->max_segment_size)
+		limits->max_segment_size = MAX_SEGMENT_SIZE;
+	if (!limits->seg_boundary_mask)
+		limits->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
+	if (!limits->bounce_pfn)
+		limits->bounce_pfn = -1;
+	/*
+	 * The other fields (alignment_offset, io_opt, misaligned)
+	 * hold 0 from the kzalloc().
+	 */
 }
 
 /*
@@ -841,9 +786,12 @@ int dm_table_add_target(struct dm_table *t, const char *type,
 
 	t->highs[t->num_targets++] = tgt->begin + tgt->len - 1;
 
-	/* FIXME: the plan is to combine high here and then have
-	 * the merge fn apply the target level restrictions. */
-	combine_restrictions_low(&t->limits, &tgt->limits);
+	if (blk_stack_limits(&t->limits, &tgt->limits, 0) < 0)
+		DMWARN("%s: target device (start sect %llu len %llu) "
+		       "is misaligned",
+		       dm_device_name(t->md),
+		       (unsigned long long) tgt->begin,
+		       (unsigned long long) tgt->len);
 	return 0;
 
  bad:
@@ -886,7 +834,7 @@ int dm_table_complete(struct dm_table *t)
 	int r = 0;
 	unsigned int leaf_nodes;
 
-	check_for_valid_limits(&t->limits);
+	init_valid_queue_limits(&t->limits);
 
 	r = validate_hardware_logical_block_alignment(t);
 	if (r)
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index fc36a4d07723..236880c1dc3f 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -144,18 +144,6 @@ struct target_type {
 	struct list_head list;
 };
 
-struct io_restrictions {
-	unsigned long bounce_pfn;
-	unsigned long seg_boundary_mask;
-	unsigned max_hw_sectors;
-	unsigned max_sectors;
-	unsigned max_segment_size;
-	unsigned short logical_block_size;
-	unsigned short max_hw_segments;
-	unsigned short max_phys_segments;
-	unsigned char no_cluster; /* inverted so that 0 is default */
-};
-
 struct dm_target {
 	struct dm_table *table;
 	struct target_type *type;
@@ -164,7 +152,7 @@ struct dm_target {
 	sector_t begin;
 	sector_t len;
 
-	/* FIXME: turn this into a mask, and merge with io_restrictions */
+	/* FIXME: turn this into a mask, and merge with queue_limits */
 	/* Always a power of 2 */
 	sector_t split_io;
 
@@ -182,7 +170,7 @@ struct dm_target {
 	 * These are automatically filled in by
 	 * dm_table_get_device.
 	 */
-	struct io_restrictions limits;
+	struct queue_limits limits;
 
 	/* target specific data */
 	void *private;
-- 
cgit v1.2.3-59-g8ed1b


From 1197764e403d97231eb6da2b1e16f511a7fd3101 Mon Sep 17 00:00:00 2001
From: Mike Snitzer <snitzer@redhat.com>
Date: Mon, 22 Jun 2009 10:12:32 +0100
Subject: dm table: establish queue limits by copying table limits

Copy the table's queue_limits to the DM device's request_queue.  This
properly initializes the queue's topology limits and also avoids having
to track the evolution of 'struct queue_limits' in
dm_table_set_restrictions()

Also fixes a bug that was introduced in dm_table_set_restrictions() via
commit ae03bf639a5027d27270123f5f6e3ee6a412781d.  In addition to
establishing 'bounce_pfn' in the queue's limits blk_queue_bounce_limit()
also performs an allocation to setup the ISA DMA pool.  This allocation
resulted in "sleeping function called from invalid context" when called
from dm_table_set_restrictions().

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-table.c | 12 ++----------
 1 file changed, 2 insertions(+), 10 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 41ec2bf9fbe9..267817edc844 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -956,17 +956,9 @@ no_integrity:
 void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q)
 {
 	/*
-	 * Make sure we obey the optimistic sub devices
-	 * restrictions.
+	 * Copy table's limits to the DM device's request_queue
 	 */
-	blk_queue_max_sectors(q, t->limits.max_sectors);
-	blk_queue_max_phys_segments(q, t->limits.max_phys_segments);
-	blk_queue_max_hw_segments(q, t->limits.max_hw_segments);
-	blk_queue_logical_block_size(q, t->limits.logical_block_size);
-	blk_queue_max_segment_size(q, t->limits.max_segment_size);
-	blk_queue_max_hw_sectors(q, t->limits.max_hw_sectors);
-	blk_queue_segment_boundary(q, t->limits.seg_boundary_mask);
-	blk_queue_bounce_limit(q, t->limits.bounce_pfn);
+	q->limits = t->limits;
 
 	if (t->limits.no_cluster)
 		queue_flag_clear_unlocked(QUEUE_FLAG_CLUSTER, q);
-- 
cgit v1.2.3-59-g8ed1b


From af4874e03ed82f050d5872d8c39ce64bf16b5c38 Mon Sep 17 00:00:00 2001
From: Mike Snitzer <snitzer@redhat.com>
Date: Mon, 22 Jun 2009 10:12:33 +0100
Subject: dm target:s introduce iterate devices fn

Add .iterate_devices to 'struct target_type' to allow a function to be
called for all devices in a DM target.  Implemented it for all targets
except those in dm-snap.c (origin and snapshot).

(The raid1 version number jumps to 1.12 because we originally reserved
1.1 to 1.11 for 'block_on_error' but ended up using 'handle_errors'
instead.)

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Cc: martin.petersen@oracle.com
---
 drivers/md/dm-crypt.c         | 11 ++++++++++-
 drivers/md/dm-delay.c         | 20 +++++++++++++++++++-
 drivers/md/dm-linear.c        | 11 ++++++++++-
 drivers/md/dm-mpath.c         | 23 ++++++++++++++++++++++-
 drivers/md/dm-raid1.c         | 17 ++++++++++++++++-
 drivers/md/dm-stripe.c        | 18 +++++++++++++++++-
 include/linux/device-mapper.h | 11 +++++++++++
 7 files changed, 105 insertions(+), 6 deletions(-)

diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 04db6c4004a8..9933eb861c71 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -1313,9 +1313,17 @@ static int crypt_merge(struct dm_target *ti, struct bvec_merge_data *bvm,
 	return min(max_size, q->merge_bvec_fn(q, bvm, biovec));
 }
 
+static int crypt_iterate_devices(struct dm_target *ti,
+				 iterate_devices_callout_fn fn, void *data)
+{
+	struct crypt_config *cc = ti->private;
+
+	return fn(ti, cc->dev, cc->start, data);
+}
+
 static struct target_type crypt_target = {
 	.name   = "crypt",
-	.version= {1, 6, 0},
+	.version = {1, 7, 0},
 	.module = THIS_MODULE,
 	.ctr    = crypt_ctr,
 	.dtr    = crypt_dtr,
@@ -1326,6 +1334,7 @@ static struct target_type crypt_target = {
 	.resume = crypt_resume,
 	.message = crypt_message,
 	.merge  = crypt_merge,
+	.iterate_devices = crypt_iterate_devices,
 };
 
 static int __init dm_crypt_init(void)
diff --git a/drivers/md/dm-delay.c b/drivers/md/dm-delay.c
index 8ad8a9044bbf..4e5b843cd4d7 100644
--- a/drivers/md/dm-delay.c
+++ b/drivers/md/dm-delay.c
@@ -318,9 +318,26 @@ static int delay_status(struct dm_target *ti, status_type_t type,
 	return 0;
 }
 
+static int delay_iterate_devices(struct dm_target *ti,
+				 iterate_devices_callout_fn fn, void *data)
+{
+	struct delay_c *dc = ti->private;
+	int ret = 0;
+
+	ret = fn(ti, dc->dev_read, dc->start_read, data);
+	if (ret)
+		goto out;
+
+	if (dc->dev_write)
+		ret = fn(ti, dc->dev_write, dc->start_write, data);
+
+out:
+	return ret;
+}
+
 static struct target_type delay_target = {
 	.name	     = "delay",
-	.version     = {1, 0, 2},
+	.version     = {1, 1, 0},
 	.module      = THIS_MODULE,
 	.ctr	     = delay_ctr,
 	.dtr	     = delay_dtr,
@@ -328,6 +345,7 @@ static struct target_type delay_target = {
 	.presuspend  = delay_presuspend,
 	.resume	     = delay_resume,
 	.status	     = delay_status,
+	.iterate_devices = delay_iterate_devices,
 };
 
 static int __init dm_delay_init(void)
diff --git a/drivers/md/dm-linear.c b/drivers/md/dm-linear.c
index ecbb17421da4..9184b6deb868 100644
--- a/drivers/md/dm-linear.c
+++ b/drivers/md/dm-linear.c
@@ -134,9 +134,17 @@ static int linear_merge(struct dm_target *ti, struct bvec_merge_data *bvm,
 	return min(max_size, q->merge_bvec_fn(q, bvm, biovec));
 }
 
+static int linear_iterate_devices(struct dm_target *ti,
+				  iterate_devices_callout_fn fn, void *data)
+{
+	struct linear_c *lc = ti->private;
+
+	return fn(ti, lc->dev, lc->start, data);
+}
+
 static struct target_type linear_target = {
 	.name   = "linear",
-	.version= {1, 0, 3},
+	.version = {1, 1, 0},
 	.module = THIS_MODULE,
 	.ctr    = linear_ctr,
 	.dtr    = linear_dtr,
@@ -144,6 +152,7 @@ static struct target_type linear_target = {
 	.status = linear_status,
 	.ioctl  = linear_ioctl,
 	.merge  = linear_merge,
+	.iterate_devices = linear_iterate_devices,
 };
 
 int __init dm_linear_init(void)
diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index 890c0e8ed13e..f8aeaaa54afe 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -1450,12 +1450,32 @@ static int multipath_ioctl(struct dm_target *ti, unsigned int cmd,
 	return r ? : __blkdev_driver_ioctl(bdev, mode, cmd, arg);
 }
 
+static int multipath_iterate_devices(struct dm_target *ti,
+				     iterate_devices_callout_fn fn, void *data)
+{
+	struct multipath *m = ti->private;
+	struct priority_group *pg;
+	struct pgpath *p;
+	int ret = 0;
+
+	list_for_each_entry(pg, &m->priority_groups, list) {
+		list_for_each_entry(p, &pg->pgpaths, list) {
+			ret = fn(ti, p->path.dev, ti->begin, data);
+			if (ret)
+				goto out;
+		}
+	}
+
+out:
+	return ret;
+}
+
 /*-----------------------------------------------------------------
  * Module setup
  *---------------------------------------------------------------*/
 static struct target_type multipath_target = {
 	.name = "multipath",
-	.version = {1, 0, 5},
+	.version = {1, 1, 0},
 	.module = THIS_MODULE,
 	.ctr = multipath_ctr,
 	.dtr = multipath_dtr,
@@ -1466,6 +1486,7 @@ static struct target_type multipath_target = {
 	.status = multipath_status,
 	.message = multipath_message,
 	.ioctl  = multipath_ioctl,
+	.iterate_devices = multipath_iterate_devices,
 };
 
 static int __init dm_multipath_init(void)
diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c
index 076fbb4e967a..ce8868c768cc 100644
--- a/drivers/md/dm-raid1.c
+++ b/drivers/md/dm-raid1.c
@@ -1283,9 +1283,23 @@ static int mirror_status(struct dm_target *ti, status_type_t type,
 	return 0;
 }
 
+static int mirror_iterate_devices(struct dm_target *ti,
+				  iterate_devices_callout_fn fn, void *data)
+{
+	struct mirror_set *ms = ti->private;
+	int ret = 0;
+	unsigned i;
+
+	for (i = 0; !ret && i < ms->nr_mirrors; i++)
+		ret = fn(ti, ms->mirror[i].dev,
+			 ms->mirror[i].offset, data);
+
+	return ret;
+}
+
 static struct target_type mirror_target = {
 	.name	 = "mirror",
-	.version = {1, 0, 20},
+	.version = {1, 12, 0},
 	.module	 = THIS_MODULE,
 	.ctr	 = mirror_ctr,
 	.dtr	 = mirror_dtr,
@@ -1295,6 +1309,7 @@ static struct target_type mirror_target = {
 	.postsuspend = mirror_postsuspend,
 	.resume	 = mirror_resume,
 	.status	 = mirror_status,
+	.iterate_devices = mirror_iterate_devices,
 };
 
 static int __init dm_mirror_init(void)
diff --git a/drivers/md/dm-stripe.c b/drivers/md/dm-stripe.c
index c64fe827a5f1..b240e85ae39a 100644
--- a/drivers/md/dm-stripe.c
+++ b/drivers/md/dm-stripe.c
@@ -313,15 +313,31 @@ static int stripe_end_io(struct dm_target *ti, struct bio *bio,
 	return error;
 }
 
+static int stripe_iterate_devices(struct dm_target *ti,
+				  iterate_devices_callout_fn fn, void *data)
+{
+	struct stripe_c *sc = ti->private;
+	int ret = 0;
+	unsigned i = 0;
+
+	do
+		ret = fn(ti, sc->stripe[i].dev,
+			 sc->stripe[i].physical_start, data);
+	while (!ret && ++i < sc->stripes);
+
+	return ret;
+}
+
 static struct target_type stripe_target = {
 	.name   = "striped",
-	.version = {1, 1, 0},
+	.version = {1, 2, 0},
 	.module = THIS_MODULE,
 	.ctr    = stripe_ctr,
 	.dtr    = stripe_dtr,
 	.map    = stripe_map,
 	.end_io = stripe_end_io,
 	.status = stripe_status,
+	.iterate_devices = stripe_iterate_devices,
 };
 
 int __init dm_stripe_init(void)
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index 236880c1dc3f..deac3b4e5e18 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -11,6 +11,7 @@
 #include <linux/bio.h>
 #include <linux/blkdev.h>
 
+struct dm_dev;
 struct dm_target;
 struct dm_table;
 struct mapped_device;
@@ -81,6 +82,15 @@ typedef int (*dm_ioctl_fn) (struct dm_target *ti, unsigned int cmd,
 typedef int (*dm_merge_fn) (struct dm_target *ti, struct bvec_merge_data *bvm,
 			    struct bio_vec *biovec, int max_size);
 
+typedef int (*iterate_devices_callout_fn) (struct dm_target *ti,
+					   struct dm_dev *dev,
+					   sector_t physical_start,
+					   void *data);
+
+typedef int (*dm_iterate_devices_fn) (struct dm_target *ti,
+				      iterate_devices_callout_fn fn,
+				      void *data);
+
 /*
  * Returns:
  *    0: The target can handle the next I/O immediately.
@@ -139,6 +149,7 @@ struct target_type {
 	dm_ioctl_fn ioctl;
 	dm_merge_fn merge;
 	dm_busy_fn busy;
+	dm_iterate_devices_fn iterate_devices;
 
 	/* For internal device-mapper use. */
 	struct list_head list;
-- 
cgit v1.2.3-59-g8ed1b


From 18d8594dd93a1ae2fafd591ec026e87d743292bf Mon Sep 17 00:00:00 2001
From: Mike Snitzer <snitzer@redhat.com>
Date: Mon, 22 Jun 2009 10:12:33 +0100
Subject: dm log: fix create_log_context to use logical_block_size of log
 device

create_log_context() must use the logical_block_size from the log disk,
where the I/O happens, not the target's logical_block_size.

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-log.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/md/dm-log.c b/drivers/md/dm-log.c
index 6352a9ad4446..9443896ede07 100644
--- a/drivers/md/dm-log.c
+++ b/drivers/md/dm-log.c
@@ -412,9 +412,10 @@ static int create_log_context(struct dm_dirty_log *log, struct dm_target *ti,
 		/*
 		 * Buffer holds both header and bitset.
 		 */
-		buf_size = dm_round_up((LOG_OFFSET << SECTOR_SHIFT) +
-				       bitset_size,
-				       ti->limits.logical_block_size);
+		buf_size =
+		    dm_round_up((LOG_OFFSET << SECTOR_SHIFT) + bitset_size,
+				bdev_logical_block_size(lc->header_location.
+							    bdev));
 
 		if (buf_size > i_size_read(dev->bdev->bd_inode)) {
 			DMWARN("log device %s too small: need %llu bytes",
-- 
cgit v1.2.3-59-g8ed1b


From 754c5fc7ebb417b23601a6222a6005cc2e7f2913 Mon Sep 17 00:00:00 2001
From: Mike Snitzer <snitzer@redhat.com>
Date: Mon, 22 Jun 2009 10:12:34 +0100
Subject: dm: calculate queue limits during resume not load

Currently, device-mapper maintains a separate instance of 'struct
queue_limits' for each table of each device.  When the configuration of
a device is to be changed, first its table is loaded and this structure
is populated, then the device is 'resumed' and the calculated
queue_limits are applied.

This places restrictions on how userspace may process related devices,
where it is often advantageous to 'load' tables for several devices
at once before 'resuming' them together.  As the new queue_limits
only take effect after the 'resume', if they are changing and one
device uses another, the latter must be 'resumed' before the former
may be 'loaded'.

This patch moves the calculation of these queue_limits out of
the 'load' operation into 'resume'.  Since we are no longer
pre-calculating this struct, we no longer need to maintain copies
within our dm structs.

dm_set_device_limits() now passes the 'start' of the device's
data area (aka pe_start) as the 'offset' to blk_stack_limits().

init_valid_queue_limits() is replaced by blk_set_default_limits().

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Cc: martin.petersen@oracle.com
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-table.c         | 185 +++++++++++++++++++++++-------------------
 drivers/md/dm.c               |  12 ++-
 drivers/md/dm.h               |   5 +-
 include/linux/device-mapper.h |  10 +--
 4 files changed, 117 insertions(+), 95 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 267817edc844..09a57113955e 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -62,12 +62,6 @@ struct dm_table {
 	/* a list of devices used by this table */
 	struct list_head devices;
 
-	/*
-	 * These are optimistic limits taken from all the
-	 * targets, some targets will need smaller limits.
-	 */
-	struct queue_limits limits;
-
 	/* events get handed up using this callback */
 	void (*event_fn)(void *);
 	void *event_context;
@@ -346,18 +340,21 @@ static void close_dev(struct dm_dev_internal *d, struct mapped_device *md)
 /*
  * If possible, this checks an area of a destination device is valid.
  */
-static int device_area_is_valid(struct dm_target *ti, struct block_device *bdev,
-			     sector_t start, sector_t len)
+static int device_area_is_valid(struct dm_target *ti, struct dm_dev *dev,
+				sector_t start, void *data)
 {
-	sector_t dev_size = i_size_read(bdev->bd_inode) >> SECTOR_SHIFT;
+	struct queue_limits *limits = data;
+	struct block_device *bdev = dev->bdev;
+	sector_t dev_size =
+		i_size_read(bdev->bd_inode) >> SECTOR_SHIFT;
 	unsigned short logical_block_size_sectors =
-		ti->limits.logical_block_size >> SECTOR_SHIFT;
+		limits->logical_block_size >> SECTOR_SHIFT;
 	char b[BDEVNAME_SIZE];
 
 	if (!dev_size)
 		return 1;
 
-	if ((start >= dev_size) || (start + len > dev_size)) {
+	if ((start >= dev_size) || (start + ti->len > dev_size)) {
 		DMWARN("%s: %s too small for target",
 		       dm_device_name(ti->table->md), bdevname(bdev, b));
 		return 0;
@@ -371,16 +368,16 @@ static int device_area_is_valid(struct dm_target *ti, struct block_device *bdev,
 		       "logical block size %hu of %s",
 		       dm_device_name(ti->table->md),
 		       (unsigned long long)start,
-		       ti->limits.logical_block_size, bdevname(bdev, b));
+		       limits->logical_block_size, bdevname(bdev, b));
 		return 0;
 	}
 
-	if (len & (logical_block_size_sectors - 1)) {
+	if (ti->len & (logical_block_size_sectors - 1)) {
 		DMWARN("%s: len=%llu not aligned to h/w "
 		       "logical block size %hu of %s",
 		       dm_device_name(ti->table->md),
-		       (unsigned long long)len,
-		       ti->limits.logical_block_size, bdevname(bdev, b));
+		       (unsigned long long)ti->len,
+		       limits->logical_block_size, bdevname(bdev, b));
 		return 0;
 	}
 
@@ -479,18 +476,21 @@ static int __table_get_device(struct dm_table *t, struct dm_target *ti,
  */
 #define min_not_zero(l, r) (l == 0) ? r : ((r == 0) ? l : min(l, r))
 
-void dm_set_device_limits(struct dm_target *ti, struct block_device *bdev)
+int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
+			 sector_t start, void *data)
 {
+	struct queue_limits *limits = data;
+	struct block_device *bdev = dev->bdev;
 	struct request_queue *q = bdev_get_queue(bdev);
 	char b[BDEVNAME_SIZE];
 
 	if (unlikely(!q)) {
 		DMWARN("%s: Cannot set limits for nonexistent device %s",
 		       dm_device_name(ti->table->md), bdevname(bdev, b));
-		return;
+		return 0;
 	}
 
-	if (blk_stack_limits(&ti->limits, &q->limits, 0) < 0)
+	if (blk_stack_limits(limits, &q->limits, start) < 0)
 		DMWARN("%s: target device %s is misaligned",
 		       dm_device_name(ti->table->md), bdevname(bdev, b));
 
@@ -501,32 +501,21 @@ void dm_set_device_limits(struct dm_target *ti, struct block_device *bdev)
 	 */
 
 	if (q->merge_bvec_fn && !ti->type->merge)
-		ti->limits.max_sectors =
-			min_not_zero(ti->limits.max_sectors,
+		limits->max_sectors =
+			min_not_zero(limits->max_sectors,
 				     (unsigned int) (PAGE_SIZE >> 9));
+	return 0;
 }
 EXPORT_SYMBOL_GPL(dm_set_device_limits);
 
 int dm_get_device(struct dm_target *ti, const char *path, sector_t start,
 		  sector_t len, fmode_t mode, struct dm_dev **result)
 {
-	int r = __table_get_device(ti->table, ti, path,
-				   start, len, mode, result);
-
-	if (r)
-		return r;
-
-	dm_set_device_limits(ti, (*result)->bdev);
-
-	if (!device_area_is_valid(ti, (*result)->bdev, start, len)) {
-		dm_put_device(ti, *result);
-		*result = NULL;
-		return -EINVAL;
-	}
-
-	return r;
+	return __table_get_device(ti->table, ti, path,
+				  start, len, mode, result);
 }
 
+
 /*
  * Decrement a devices use count and remove it if necessary.
  */
@@ -641,34 +630,6 @@ int dm_split_args(int *argc, char ***argvp, char *input)
 	return 0;
 }
 
-static void init_valid_queue_limits(struct queue_limits *limits)
-{
-	if (!limits->max_sectors)
-		limits->max_sectors = SAFE_MAX_SECTORS;
-	if (!limits->max_hw_sectors)
-		limits->max_hw_sectors = SAFE_MAX_SECTORS;
-	if (!limits->max_phys_segments)
-		limits->max_phys_segments = MAX_PHYS_SEGMENTS;
-	if (!limits->max_hw_segments)
-		limits->max_hw_segments = MAX_HW_SEGMENTS;
-	if (!limits->logical_block_size)
-		limits->logical_block_size = 1 << SECTOR_SHIFT;
-	if (!limits->physical_block_size)
-		limits->physical_block_size = 1 << SECTOR_SHIFT;
-	if (!limits->io_min)
-		limits->io_min = 1 << SECTOR_SHIFT;
-	if (!limits->max_segment_size)
-		limits->max_segment_size = MAX_SEGMENT_SIZE;
-	if (!limits->seg_boundary_mask)
-		limits->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
-	if (!limits->bounce_pfn)
-		limits->bounce_pfn = -1;
-	/*
-	 * The other fields (alignment_offset, io_opt, misaligned)
-	 * hold 0 from the kzalloc().
-	 */
-}
-
 /*
  * Impose necessary and sufficient conditions on a devices's table such
  * that any incoming bio which respects its logical_block_size can be
@@ -676,14 +637,15 @@ static void init_valid_queue_limits(struct queue_limits *limits)
  * two or more targets, the size of each piece it gets split into must
  * be compatible with the logical_block_size of the target processing it.
  */
-static int validate_hardware_logical_block_alignment(struct dm_table *table)
+static int validate_hardware_logical_block_alignment(struct dm_table *table,
+						 struct queue_limits *limits)
 {
 	/*
 	 * This function uses arithmetic modulo the logical_block_size
 	 * (in units of 512-byte sectors).
 	 */
 	unsigned short device_logical_block_size_sects =
-		table->limits.logical_block_size >> SECTOR_SHIFT;
+		limits->logical_block_size >> SECTOR_SHIFT;
 
 	/*
 	 * Offset of the start of the next table entry, mod logical_block_size.
@@ -697,6 +659,7 @@ static int validate_hardware_logical_block_alignment(struct dm_table *table)
 	unsigned short remaining = 0;
 
 	struct dm_target *uninitialized_var(ti);
+	struct queue_limits ti_limits;
 	unsigned i = 0;
 
 	/*
@@ -705,12 +668,19 @@ static int validate_hardware_logical_block_alignment(struct dm_table *table)
 	while (i < dm_table_get_num_targets(table)) {
 		ti = dm_table_get_target(table, i++);
 
+		blk_set_default_limits(&ti_limits);
+
+		/* combine all target devices' limits */
+		if (ti->type->iterate_devices)
+			ti->type->iterate_devices(ti, dm_set_device_limits,
+						  &ti_limits);
+
 		/*
 		 * If the remaining sectors fall entirely within this
 		 * table entry are they compatible with its logical_block_size?
 		 */
 		if (remaining < ti->len &&
-		    remaining & ((ti->limits.logical_block_size >>
+		    remaining & ((ti_limits.logical_block_size >>
 				  SECTOR_SHIFT) - 1))
 			break;	/* Error */
 
@@ -723,11 +693,11 @@ static int validate_hardware_logical_block_alignment(struct dm_table *table)
 
 	if (remaining) {
 		DMWARN("%s: table line %u (start sect %llu len %llu) "
-		       "not aligned to hardware logical block size %hu",
+		       "not aligned to h/w logical block size %hu",
 		       dm_device_name(table->md), i,
 		       (unsigned long long) ti->begin,
 		       (unsigned long long) ti->len,
-		       table->limits.logical_block_size);
+		       limits->logical_block_size);
 		return -EINVAL;
 	}
 
@@ -786,12 +756,6 @@ int dm_table_add_target(struct dm_table *t, const char *type,
 
 	t->highs[t->num_targets++] = tgt->begin + tgt->len - 1;
 
-	if (blk_stack_limits(&t->limits, &tgt->limits, 0) < 0)
-		DMWARN("%s: target device (start sect %llu len %llu) "
-		       "is misaligned",
-		       dm_device_name(t->md),
-		       (unsigned long long) tgt->begin,
-		       (unsigned long long) tgt->len);
 	return 0;
 
  bad:
@@ -834,12 +798,6 @@ int dm_table_complete(struct dm_table *t)
 	int r = 0;
 	unsigned int leaf_nodes;
 
-	init_valid_queue_limits(&t->limits);
-
-	r = validate_hardware_logical_block_alignment(t);
-	if (r)
-		return r;
-
 	/* how many indexes will the btree have ? */
 	leaf_nodes = dm_div_up(t->num_targets, KEYS_PER_NODE);
 	t->depth = 1 + int_log(leaf_nodes, CHILDREN_PER_NODE);
@@ -914,6 +872,57 @@ struct dm_target *dm_table_find_target(struct dm_table *t, sector_t sector)
 	return &t->targets[(KEYS_PER_NODE * n) + k];
 }
 
+/*
+ * Establish the new table's queue_limits and validate them.
+ */
+int dm_calculate_queue_limits(struct dm_table *table,
+			      struct queue_limits *limits)
+{
+	struct dm_target *uninitialized_var(ti);
+	struct queue_limits ti_limits;
+	unsigned i = 0;
+
+	blk_set_default_limits(limits);
+
+	while (i < dm_table_get_num_targets(table)) {
+		blk_set_default_limits(&ti_limits);
+
+		ti = dm_table_get_target(table, i++);
+
+		if (!ti->type->iterate_devices)
+			goto combine_limits;
+
+		/*
+		 * Combine queue limits of all the devices this target uses.
+		 */
+		ti->type->iterate_devices(ti, dm_set_device_limits,
+					  &ti_limits);
+
+		/*
+		 * Check each device area is consistent with the target's
+		 * overall queue limits.
+		 */
+		if (!ti->type->iterate_devices(ti, device_area_is_valid,
+					       &ti_limits))
+			return -EINVAL;
+
+combine_limits:
+		/*
+		 * Merge this target's queue limits into the overall limits
+		 * for the table.
+		 */
+		if (blk_stack_limits(limits, &ti_limits, 0) < 0)
+			DMWARN("%s: target device "
+			       "(start sect %llu len %llu) "
+			       "is misaligned",
+			       dm_device_name(table->md),
+			       (unsigned long long) ti->begin,
+			       (unsigned long long) ti->len);
+	}
+
+	return validate_hardware_logical_block_alignment(table, limits);
+}
+
 /*
  * Set the integrity profile for this device if all devices used have
  * matching profiles.
@@ -953,14 +962,24 @@ no_integrity:
 	return;
 }
 
-void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q)
+void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
+			       struct queue_limits *limits)
 {
+	/*
+	 * Each target device in the table has a data area that should normally
+	 * be aligned such that the DM device's alignment_offset is 0.
+	 * FIXME: Propagate alignment_offsets up the stack and warn of
+	 *	  sub-optimal or inconsistent settings.
+	 */
+	limits->alignment_offset = 0;
+	limits->misaligned = 0;
+
 	/*
 	 * Copy table's limits to the DM device's request_queue
 	 */
-	q->limits = t->limits;
+	q->limits = *limits;
 
-	if (t->limits.no_cluster)
+	if (limits->no_cluster)
 		queue_flag_clear_unlocked(QUEUE_FLAG_CLUSTER, q);
 	else
 		queue_flag_set_unlocked(QUEUE_FLAG_CLUSTER, q);
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index a9210bb594e7..f609793a92d0 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1313,7 +1313,8 @@ static void __set_size(struct mapped_device *md, sector_t size)
 	mutex_unlock(&md->bdev->bd_inode->i_mutex);
 }
 
-static int __bind(struct mapped_device *md, struct dm_table *t)
+static int __bind(struct mapped_device *md, struct dm_table *t,
+		  struct queue_limits *limits)
 {
 	struct request_queue *q = md->queue;
 	sector_t size;
@@ -1337,7 +1338,7 @@ static int __bind(struct mapped_device *md, struct dm_table *t)
 
 	write_lock(&md->map_lock);
 	md->map = t;
-	dm_table_set_restrictions(t, q);
+	dm_table_set_restrictions(t, q, limits);
 	write_unlock(&md->map_lock);
 
 	return 0;
@@ -1562,6 +1563,7 @@ static void dm_queue_flush(struct mapped_device *md)
  */
 int dm_swap_table(struct mapped_device *md, struct dm_table *table)
 {
+	struct queue_limits limits;
 	int r = -EINVAL;
 
 	mutex_lock(&md->suspend_lock);
@@ -1570,8 +1572,12 @@ int dm_swap_table(struct mapped_device *md, struct dm_table *table)
 	if (!dm_suspended(md))
 		goto out;
 
+	r = dm_calculate_queue_limits(table, &limits);
+	if (r)
+		goto out;
+
 	__unbind(md);
-	r = __bind(md, table);
+	r = __bind(md, table, &limits);
 
 out:
 	mutex_unlock(&md->suspend_lock);
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index b5935c610c44..604e85caadf6 100644
--- a/drivers/md/dm.h
+++ b/drivers/md/dm.h
@@ -41,7 +41,10 @@ void dm_table_event_callback(struct dm_table *t,
 			     void (*fn)(void *), void *context);
 struct dm_target *dm_table_get_target(struct dm_table *t, unsigned int index);
 struct dm_target *dm_table_find_target(struct dm_table *t, sector_t sector);
-void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q);
+int dm_calculate_queue_limits(struct dm_table *table,
+			      struct queue_limits *limits);
+void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
+			       struct queue_limits *limits);
 struct list_head *dm_table_get_devices(struct dm_table *t);
 void dm_table_presuspend_targets(struct dm_table *t);
 void dm_table_postsuspend_targets(struct dm_table *t);
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index deac3b4e5e18..e6bf3b8c7bf2 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -103,7 +103,8 @@ void dm_error(const char *message);
 /*
  * Combine device limits.
  */
-void dm_set_device_limits(struct dm_target *ti, struct block_device *bdev);
+int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
+			 sector_t start, void *data);
 
 struct dm_dev {
 	struct block_device *bdev;
@@ -163,7 +164,6 @@ struct dm_target {
 	sector_t begin;
 	sector_t len;
 
-	/* FIXME: turn this into a mask, and merge with queue_limits */
 	/* Always a power of 2 */
 	sector_t split_io;
 
@@ -177,12 +177,6 @@ struct dm_target {
 	 */
 	unsigned num_flush_requests;
 
-	/*
-	 * These are automatically filled in by
-	 * dm_table_get_device.
-	 */
-	struct queue_limits limits;
-
 	/* target specific data */
 	void *private;
 
-- 
cgit v1.2.3-59-g8ed1b


From f5db4af466e2dca0fe822019812d586ca910b00c Mon Sep 17 00:00:00 2001
From: Jonthan Brassow <jbrassow@redhat.com>
Date: Mon, 22 Jun 2009 10:12:35 +0100
Subject: dm raid1: add userspace log

This patch contains a device-mapper mirror log module that forwards
requests to userspace for processing.

The structures used for communication between kernel and userspace are
located in include/linux/dm-log-userspace.h.  Due to the frequency,
diversity, and 2-way communication nature of the exchanges between
kernel and userspace, 'connector' was chosen as the interface for
communication.

The first log implementations written in userspace - "clustered-disk"
and "clustered-core" - support clustered shared storage.   A userspace
daemon (in the LVM2 source code repository) uses openAIS/corosync to
process requests in an ordered fashion with the rest of the nodes in the
cluster so as to prevent log state corruption.  Other implementations
with no association to LVM or openAIS/corosync, are certainly possible.

(Imagine if two machines are writing to the same region of a mirror.
They would both mark the region dirty, but you need a cluster-aware
entity that can handle properly marking the region clean when they are
done.  Otherwise, you might clear the region when the first machine is
done, not the second.)

Signed-off-by: Jonathan Brassow <jbrassow@redhat.com>
Cc: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 Documentation/device-mapper/dm-log.txt |  54 +++
 drivers/md/Kconfig                     |  11 +
 drivers/md/Makefile                    |   3 +
 drivers/md/dm-log-userspace-base.c     | 696 +++++++++++++++++++++++++++++++++
 drivers/md/dm-log-userspace-transfer.c | 276 +++++++++++++
 drivers/md/dm-log-userspace-transfer.h |  18 +
 include/linux/Kbuild                   |   1 +
 include/linux/connector.h              |   4 +-
 include/linux/dm-log-userspace.h       | 386 ++++++++++++++++++
 9 files changed, 1448 insertions(+), 1 deletion(-)
 create mode 100644 Documentation/device-mapper/dm-log.txt
 create mode 100644 drivers/md/dm-log-userspace-base.c
 create mode 100644 drivers/md/dm-log-userspace-transfer.c
 create mode 100644 drivers/md/dm-log-userspace-transfer.h
 create mode 100644 include/linux/dm-log-userspace.h

diff --git a/Documentation/device-mapper/dm-log.txt b/Documentation/device-mapper/dm-log.txt
new file mode 100644
index 000000000000..994dd75475a6
--- /dev/null
+++ b/Documentation/device-mapper/dm-log.txt
@@ -0,0 +1,54 @@
+Device-Mapper Logging
+=====================
+The device-mapper logging code is used by some of the device-mapper
+RAID targets to track regions of the disk that are not consistent.
+A region (or portion of the address space) of the disk may be
+inconsistent because a RAID stripe is currently being operated on or
+a machine died while the region was being altered.  In the case of
+mirrors, a region would be considered dirty/inconsistent while you
+are writing to it because the writes need to be replicated for all
+the legs of the mirror and may not reach the legs at the same time.
+Once all writes are complete, the region is considered clean again.
+
+There is a generic logging interface that the device-mapper RAID
+implementations use to perform logging operations (see
+dm_dirty_log_type in include/linux/dm-dirty-log.h).  Various different
+logging implementations are available and provide different
+capabilities.  The list includes:
+
+Type		Files
+====		=====
+disk		drivers/md/dm-log.c
+core		drivers/md/dm-log.c
+userspace	drivers/md/dm-log-userspace* include/linux/dm-log-userspace.h
+
+The "disk" log type
+-------------------
+This log implementation commits the log state to disk.  This way, the
+logging state survives reboots/crashes.
+
+The "core" log type
+-------------------
+This log implementation keeps the log state in memory.  The log state
+will not survive a reboot or crash, but there may be a small boost in
+performance.  This method can also be used if no storage device is
+available for storing log state.
+
+The "userspace" log type
+------------------------
+This log type simply provides a way to export the log API to userspace,
+so log implementations can be done there.  This is done by forwarding most
+logging requests to userspace, where a daemon receives and processes the
+request.
+
+The structure used for communication between kernel and userspace are
+located in include/linux/dm-log-userspace.h.  Due to the frequency,
+diversity, and 2-way communication nature of the exchanges between
+kernel and userspace, 'connector' is used as the interface for
+communication.
+
+There are currently two userspace log implementations that leverage this
+framework - "clustered_disk" and "clustered_core".  These implementations
+provide a cluster-coherent log for shared-storage.  Device-mapper mirroring
+can be used in a shared-storage environment when the cluster log implementations
+are employed.
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig
index 09f93fa68912..020f9573fd82 100644
--- a/drivers/md/Kconfig
+++ b/drivers/md/Kconfig
@@ -231,6 +231,17 @@ config DM_MIRROR
          Allow volume managers to mirror logical volumes, also
          needed for live data migration tools such as 'pvmove'.
 
+config DM_LOG_USERSPACE
+	tristate "Mirror userspace logging (EXPERIMENTAL)"
+	depends on DM_MIRROR && EXPERIMENTAL && NET
+	select CONNECTOR
+	---help---
+	  The userspace logging module provides a mechanism for
+	  relaying the dm-dirty-log API to userspace.  Log designs
+	  which are more suited to userspace implementation (e.g.
+	  shared storage logs) or experimental logs can be implemented
+	  by leveraging this framework.
+
 config DM_ZERO
 	tristate "Zero target"
 	depends on BLK_DEV_DM
diff --git a/drivers/md/Makefile b/drivers/md/Makefile
index dade52f60733..1dc4185bd781 100644
--- a/drivers/md/Makefile
+++ b/drivers/md/Makefile
@@ -8,6 +8,8 @@ dm-multipath-y	+= dm-path-selector.o dm-mpath.o
 dm-snapshot-y	+= dm-snap.o dm-exception-store.o dm-snap-transient.o \
 		    dm-snap-persistent.o
 dm-mirror-y	+= dm-raid1.o
+dm-log-userspace-y \
+		+= dm-log-userspace-base.o dm-log-userspace-transfer.o
 md-mod-y	+= md.o bitmap.o
 raid456-y	+= raid5.o
 raid6_pq-y	+= raid6algos.o raid6recov.o raid6tables.o \
@@ -40,6 +42,7 @@ obj-$(CONFIG_DM_MULTIPATH_QL)	+= dm-queue-length.o
 obj-$(CONFIG_DM_MULTIPATH_ST)	+= dm-service-time.o
 obj-$(CONFIG_DM_SNAPSHOT)	+= dm-snapshot.o
 obj-$(CONFIG_DM_MIRROR)		+= dm-mirror.o dm-log.o dm-region-hash.o
+obj-$(CONFIG_DM_LOG_USERSPACE)	+= dm-log-userspace.o
 obj-$(CONFIG_DM_ZERO)		+= dm-zero.o
 
 quiet_cmd_unroll = UNROLL  $@
diff --git a/drivers/md/dm-log-userspace-base.c b/drivers/md/dm-log-userspace-base.c
new file mode 100644
index 000000000000..e69b96560997
--- /dev/null
+++ b/drivers/md/dm-log-userspace-base.c
@@ -0,0 +1,696 @@
+/*
+ * Copyright (C) 2006-2009 Red Hat, Inc.
+ *
+ * This file is released under the LGPL.
+ */
+
+#include <linux/bio.h>
+#include <linux/dm-dirty-log.h>
+#include <linux/device-mapper.h>
+#include <linux/dm-log-userspace.h>
+
+#include "dm-log-userspace-transfer.h"
+
+struct flush_entry {
+	int type;
+	region_t region;
+	struct list_head list;
+};
+
+struct log_c {
+	struct dm_target *ti;
+	uint32_t region_size;
+	region_t region_count;
+	char uuid[DM_UUID_LEN];
+
+	char *usr_argv_str;
+	uint32_t usr_argc;
+
+	/*
+	 * in_sync_hint gets set when doing is_remote_recovering.  It
+	 * represents the first region that needs recovery.  IOW, the
+	 * first zero bit of sync_bits.  This can be useful for to limit
+	 * traffic for calls like is_remote_recovering and get_resync_work,
+	 * but be take care in its use for anything else.
+	 */
+	uint64_t in_sync_hint;
+
+	spinlock_t flush_lock;
+	struct list_head flush_list;  /* only for clear and mark requests */
+};
+
+static mempool_t *flush_entry_pool;
+
+static void *flush_entry_alloc(gfp_t gfp_mask, void *pool_data)
+{
+	return kmalloc(sizeof(struct flush_entry), gfp_mask);
+}
+
+static void flush_entry_free(void *element, void *pool_data)
+{
+	kfree(element);
+}
+
+static int userspace_do_request(struct log_c *lc, const char *uuid,
+				int request_type, char *data, size_t data_size,
+				char *rdata, size_t *rdata_size)
+{
+	int r;
+
+	/*
+	 * If the server isn't there, -ESRCH is returned,
+	 * and we must keep trying until the server is
+	 * restored.
+	 */
+retry:
+	r = dm_consult_userspace(uuid, request_type, data,
+				 data_size, rdata, rdata_size);
+
+	if (r != -ESRCH)
+		return r;
+
+	DMERR(" Userspace log server not found.");
+	while (1) {
+		set_current_state(TASK_INTERRUPTIBLE);
+		schedule_timeout(2*HZ);
+		DMWARN("Attempting to contact userspace log server...");
+		r = dm_consult_userspace(uuid, DM_ULOG_CTR, lc->usr_argv_str,
+					 strlen(lc->usr_argv_str) + 1,
+					 NULL, NULL);
+		if (!r)
+			break;
+	}
+	DMINFO("Reconnected to userspace log server... DM_ULOG_CTR complete");
+	r = dm_consult_userspace(uuid, DM_ULOG_RESUME, NULL,
+				 0, NULL, NULL);
+	if (!r)
+		goto retry;
+
+	DMERR("Error trying to resume userspace log: %d", r);
+
+	return -ESRCH;
+}
+
+static int build_constructor_string(struct dm_target *ti,
+				    unsigned argc, char **argv,
+				    char **ctr_str)
+{
+	int i, str_size;
+	char *str = NULL;
+
+	*ctr_str = NULL;
+
+	for (i = 0, str_size = 0; i < argc; i++)
+		str_size += strlen(argv[i]) + 1; /* +1 for space between args */
+
+	str_size += 20; /* Max number of chars in a printed u64 number */
+
+	str = kzalloc(str_size, GFP_KERNEL);
+	if (!str) {
+		DMWARN("Unable to allocate memory for constructor string");
+		return -ENOMEM;
+	}
+
+	for (i = 0, str_size = 0; i < argc; i++)
+		str_size += sprintf(str + str_size, "%s ", argv[i]);
+	str_size += sprintf(str + str_size, "%llu",
+			    (unsigned long long)ti->len);
+
+	*ctr_str = str;
+	return str_size;
+}
+
+/*
+ * userspace_ctr
+ *
+ * argv contains:
+ *	<UUID> <other args>
+ * Where 'other args' is the userspace implementation specific log
+ * arguments.  An example might be:
+ *	<UUID> clustered_disk <arg count> <log dev> <region_size> [[no]sync]
+ *
+ * So, this module will strip off the <UUID> for identification purposes
+ * when communicating with userspace about a log; but will pass on everything
+ * else.
+ */
+static int userspace_ctr(struct dm_dirty_log *log, struct dm_target *ti,
+			 unsigned argc, char **argv)
+{
+	int r = 0;
+	int str_size;
+	char *ctr_str = NULL;
+	struct log_c *lc = NULL;
+	uint64_t rdata;
+	size_t rdata_size = sizeof(rdata);
+
+	if (argc < 3) {
+		DMWARN("Too few arguments to userspace dirty log");
+		return -EINVAL;
+	}
+
+	lc = kmalloc(sizeof(*lc), GFP_KERNEL);
+	if (!lc) {
+		DMWARN("Unable to allocate userspace log context.");
+		return -ENOMEM;
+	}
+
+	lc->ti = ti;
+
+	if (strlen(argv[0]) > (DM_UUID_LEN - 1)) {
+		DMWARN("UUID argument too long.");
+		kfree(lc);
+		return -EINVAL;
+	}
+
+	strncpy(lc->uuid, argv[0], DM_UUID_LEN);
+	spin_lock_init(&lc->flush_lock);
+	INIT_LIST_HEAD(&lc->flush_list);
+
+	str_size = build_constructor_string(ti, argc - 1, argv + 1, &ctr_str);
+	if (str_size < 0) {
+		kfree(lc);
+		return str_size;
+	}
+
+	/* Send table string */
+	r = dm_consult_userspace(lc->uuid, DM_ULOG_CTR,
+				 ctr_str, str_size, NULL, NULL);
+
+	if (r == -ESRCH) {
+		DMERR("Userspace log server not found");
+		goto out;
+	}
+
+	/* Since the region size does not change, get it now */
+	rdata_size = sizeof(rdata);
+	r = dm_consult_userspace(lc->uuid, DM_ULOG_GET_REGION_SIZE,
+				 NULL, 0, (char *)&rdata, &rdata_size);
+
+	if (r) {
+		DMERR("Failed to get region size of dirty log");
+		goto out;
+	}
+
+	lc->region_size = (uint32_t)rdata;
+	lc->region_count = dm_sector_div_up(ti->len, lc->region_size);
+
+out:
+	if (r) {
+		kfree(lc);
+		kfree(ctr_str);
+	} else {
+		lc->usr_argv_str = ctr_str;
+		lc->usr_argc = argc;
+		log->context = lc;
+	}
+
+	return r;
+}
+
+static void userspace_dtr(struct dm_dirty_log *log)
+{
+	int r;
+	struct log_c *lc = log->context;
+
+	r = dm_consult_userspace(lc->uuid, DM_ULOG_DTR,
+				 NULL, 0,
+				 NULL, NULL);
+
+	kfree(lc->usr_argv_str);
+	kfree(lc);
+
+	return;
+}
+
+static int userspace_presuspend(struct dm_dirty_log *log)
+{
+	int r;
+	struct log_c *lc = log->context;
+
+	r = dm_consult_userspace(lc->uuid, DM_ULOG_PRESUSPEND,
+				 NULL, 0,
+				 NULL, NULL);
+
+	return r;
+}
+
+static int userspace_postsuspend(struct dm_dirty_log *log)
+{
+	int r;
+	struct log_c *lc = log->context;
+
+	r = dm_consult_userspace(lc->uuid, DM_ULOG_POSTSUSPEND,
+				 NULL, 0,
+				 NULL, NULL);
+
+	return r;
+}
+
+static int userspace_resume(struct dm_dirty_log *log)
+{
+	int r;
+	struct log_c *lc = log->context;
+
+	lc->in_sync_hint = 0;
+	r = dm_consult_userspace(lc->uuid, DM_ULOG_RESUME,
+				 NULL, 0,
+				 NULL, NULL);
+
+	return r;
+}
+
+static uint32_t userspace_get_region_size(struct dm_dirty_log *log)
+{
+	struct log_c *lc = log->context;
+
+	return lc->region_size;
+}
+
+/*
+ * userspace_is_clean
+ *
+ * Check whether a region is clean.  If there is any sort of
+ * failure when consulting the server, we return not clean.
+ *
+ * Returns: 1 if clean, 0 otherwise
+ */
+static int userspace_is_clean(struct dm_dirty_log *log, region_t region)
+{
+	int r;
+	uint64_t region64 = (uint64_t)region;
+	int64_t is_clean;
+	size_t rdata_size;
+	struct log_c *lc = log->context;
+
+	rdata_size = sizeof(is_clean);
+	r = userspace_do_request(lc, lc->uuid, DM_ULOG_IS_CLEAN,
+				 (char *)&region64, sizeof(region64),
+				 (char *)&is_clean, &rdata_size);
+
+	return (r) ? 0 : (int)is_clean;
+}
+
+/*
+ * userspace_in_sync
+ *
+ * Check if the region is in-sync.  If there is any sort
+ * of failure when consulting the server, we assume that
+ * the region is not in sync.
+ *
+ * If 'can_block' is set, return immediately
+ *
+ * Returns: 1 if in-sync, 0 if not-in-sync, -EWOULDBLOCK
+ */
+static int userspace_in_sync(struct dm_dirty_log *log, region_t region,
+			     int can_block)
+{
+	int r;
+	uint64_t region64 = region;
+	int64_t in_sync;
+	size_t rdata_size;
+	struct log_c *lc = log->context;
+
+	/*
+	 * We can never respond directly - even if in_sync_hint is
+	 * set.  This is because another machine could see a device
+	 * failure and mark the region out-of-sync.  If we don't go
+	 * to userspace to ask, we might think the region is in-sync
+	 * and allow a read to pick up data that is stale.  (This is
+	 * very unlikely if a device actually fails; but it is very
+	 * likely if a connection to one device from one machine fails.)
+	 *
+	 * There still might be a problem if the mirror caches the region
+	 * state as in-sync... but then this call would not be made.  So,
+	 * that is a mirror problem.
+	 */
+	if (!can_block)
+		return -EWOULDBLOCK;
+
+	rdata_size = sizeof(in_sync);
+	r = userspace_do_request(lc, lc->uuid, DM_ULOG_IN_SYNC,
+				 (char *)&region64, sizeof(region64),
+				 (char *)&in_sync, &rdata_size);
+	return (r) ? 0 : (int)in_sync;
+}
+
+/*
+ * userspace_flush
+ *
+ * This function is ok to block.
+ * The flush happens in two stages.  First, it sends all
+ * clear/mark requests that are on the list.  Then it
+ * tells the server to commit them.  This gives the
+ * server a chance to optimise the commit, instead of
+ * doing it for every request.
+ *
+ * Additionally, we could implement another thread that
+ * sends the requests up to the server - reducing the
+ * load on flush.  Then the flush would have less in
+ * the list and be responsible for the finishing commit.
+ *
+ * Returns: 0 on success, < 0 on failure
+ */
+static int userspace_flush(struct dm_dirty_log *log)
+{
+	int r = 0;
+	unsigned long flags;
+	struct log_c *lc = log->context;
+	LIST_HEAD(flush_list);
+	struct flush_entry *fe, *tmp_fe;
+
+	spin_lock_irqsave(&lc->flush_lock, flags);
+	list_splice_init(&lc->flush_list, &flush_list);
+	spin_unlock_irqrestore(&lc->flush_lock, flags);
+
+	if (list_empty(&flush_list))
+		return 0;
+
+	/*
+	 * FIXME: Count up requests, group request types,
+	 * allocate memory to stick all requests in and
+	 * send to server in one go.  Failing the allocation,
+	 * do it one by one.
+	 */
+
+	list_for_each_entry(fe, &flush_list, list) {
+		r = userspace_do_request(lc, lc->uuid, fe->type,
+					 (char *)&fe->region,
+					 sizeof(fe->region),
+					 NULL, NULL);
+		if (r)
+			goto fail;
+	}
+
+	r = userspace_do_request(lc, lc->uuid, DM_ULOG_FLUSH,
+				 NULL, 0, NULL, NULL);
+
+fail:
+	/*
+	 * We can safely remove these entries, even if failure.
+	 * Calling code will receive an error and will know that
+	 * the log facility has failed.
+	 */
+	list_for_each_entry_safe(fe, tmp_fe, &flush_list, list) {
+		list_del(&fe->list);
+		mempool_free(fe, flush_entry_pool);
+	}
+
+	if (r)
+		dm_table_event(lc->ti->table);
+
+	return r;
+}
+
+/*
+ * userspace_mark_region
+ *
+ * This function should avoid blocking unless absolutely required.
+ * (Memory allocation is valid for blocking.)
+ */
+static void userspace_mark_region(struct dm_dirty_log *log, region_t region)
+{
+	unsigned long flags;
+	struct log_c *lc = log->context;
+	struct flush_entry *fe;
+
+	/* Wait for an allocation, but _never_ fail */
+	fe = mempool_alloc(flush_entry_pool, GFP_NOIO);
+	BUG_ON(!fe);
+
+	spin_lock_irqsave(&lc->flush_lock, flags);
+	fe->type = DM_ULOG_MARK_REGION;
+	fe->region = region;
+	list_add(&fe->list, &lc->flush_list);
+	spin_unlock_irqrestore(&lc->flush_lock, flags);
+
+	return;
+}
+
+/*
+ * userspace_clear_region
+ *
+ * This function must not block.
+ * So, the alloc can't block.  In the worst case, it is ok to
+ * fail.  It would simply mean we can't clear the region.
+ * Does nothing to current sync context, but does mean
+ * the region will be re-sync'ed on a reload of the mirror
+ * even though it is in-sync.
+ */
+static void userspace_clear_region(struct dm_dirty_log *log, region_t region)
+{
+	unsigned long flags;
+	struct log_c *lc = log->context;
+	struct flush_entry *fe;
+
+	/*
+	 * If we fail to allocate, we skip the clearing of
+	 * the region.  This doesn't hurt us in any way, except
+	 * to cause the region to be resync'ed when the
+	 * device is activated next time.
+	 */
+	fe = mempool_alloc(flush_entry_pool, GFP_ATOMIC);
+	if (!fe) {
+		DMERR("Failed to allocate memory to clear region.");
+		return;
+	}
+
+	spin_lock_irqsave(&lc->flush_lock, flags);
+	fe->type = DM_ULOG_CLEAR_REGION;
+	fe->region = region;
+	list_add(&fe->list, &lc->flush_list);
+	spin_unlock_irqrestore(&lc->flush_lock, flags);
+
+	return;
+}
+
+/*
+ * userspace_get_resync_work
+ *
+ * Get a region that needs recovery.  It is valid to return
+ * an error for this function.
+ *
+ * Returns: 1 if region filled, 0 if no work, <0 on error
+ */
+static int userspace_get_resync_work(struct dm_dirty_log *log, region_t *region)
+{
+	int r;
+	size_t rdata_size;
+	struct log_c *lc = log->context;
+	struct {
+		int64_t i; /* 64-bit for mix arch compatibility */
+		region_t r;
+	} pkg;
+
+	if (lc->in_sync_hint >= lc->region_count)
+		return 0;
+
+	rdata_size = sizeof(pkg);
+	r = userspace_do_request(lc, lc->uuid, DM_ULOG_GET_RESYNC_WORK,
+				 NULL, 0,
+				 (char *)&pkg, &rdata_size);
+
+	*region = pkg.r;
+	return (r) ? r : (int)pkg.i;
+}
+
+/*
+ * userspace_set_region_sync
+ *
+ * Set the sync status of a given region.  This function
+ * must not fail.
+ */
+static void userspace_set_region_sync(struct dm_dirty_log *log,
+				      region_t region, int in_sync)
+{
+	int r;
+	struct log_c *lc = log->context;
+	struct {
+		region_t r;
+		int64_t i;
+	} pkg;
+
+	pkg.r = region;
+	pkg.i = (int64_t)in_sync;
+
+	r = userspace_do_request(lc, lc->uuid, DM_ULOG_SET_REGION_SYNC,
+				 (char *)&pkg, sizeof(pkg),
+				 NULL, NULL);
+
+	/*
+	 * It would be nice to be able to report failures.
+	 * However, it is easy emough to detect and resolve.
+	 */
+	return;
+}
+
+/*
+ * userspace_get_sync_count
+ *
+ * If there is any sort of failure when consulting the server,
+ * we assume that the sync count is zero.
+ *
+ * Returns: sync count on success, 0 on failure
+ */
+static region_t userspace_get_sync_count(struct dm_dirty_log *log)
+{
+	int r;
+	size_t rdata_size;
+	uint64_t sync_count;
+	struct log_c *lc = log->context;
+
+	rdata_size = sizeof(sync_count);
+	r = userspace_do_request(lc, lc->uuid, DM_ULOG_GET_SYNC_COUNT,
+				 NULL, 0,
+				 (char *)&sync_count, &rdata_size);
+
+	if (r)
+		return 0;
+
+	if (sync_count >= lc->region_count)
+		lc->in_sync_hint = lc->region_count;
+
+	return (region_t)sync_count;
+}
+
+/*
+ * userspace_status
+ *
+ * Returns: amount of space consumed
+ */
+static int userspace_status(struct dm_dirty_log *log, status_type_t status_type,
+			    char *result, unsigned maxlen)
+{
+	int r = 0;
+	size_t sz = (size_t)maxlen;
+	struct log_c *lc = log->context;
+
+	switch (status_type) {
+	case STATUSTYPE_INFO:
+		r = userspace_do_request(lc, lc->uuid, DM_ULOG_STATUS_INFO,
+					 NULL, 0,
+					 result, &sz);
+
+		if (r) {
+			sz = 0;
+			DMEMIT("%s 1 COM_FAILURE", log->type->name);
+		}
+		break;
+	case STATUSTYPE_TABLE:
+		sz = 0;
+		DMEMIT("%s %u %s %s", log->type->name, lc->usr_argc + 1,
+		       lc->uuid, lc->usr_argv_str);
+		break;
+	}
+	return (r) ? 0 : (int)sz;
+}
+
+/*
+ * userspace_is_remote_recovering
+ *
+ * Returns: 1 if region recovering, 0 otherwise
+ */
+static int userspace_is_remote_recovering(struct dm_dirty_log *log,
+					  region_t region)
+{
+	int r;
+	uint64_t region64 = region;
+	struct log_c *lc = log->context;
+	static unsigned long long limit;
+	struct {
+		int64_t is_recovering;
+		uint64_t in_sync_hint;
+	} pkg;
+	size_t rdata_size = sizeof(pkg);
+
+	/*
+	 * Once the mirror has been reported to be in-sync,
+	 * it will never again ask for recovery work.  So,
+	 * we can safely say there is not a remote machine
+	 * recovering if the device is in-sync.  (in_sync_hint
+	 * must be reset at resume time.)
+	 */
+	if (region < lc->in_sync_hint)
+		return 0;
+	else if (jiffies < limit)
+		return 1;
+
+	limit = jiffies + (HZ / 4);
+	r = userspace_do_request(lc, lc->uuid, DM_ULOG_IS_REMOTE_RECOVERING,
+				 (char *)&region64, sizeof(region64),
+				 (char *)&pkg, &rdata_size);
+	if (r)
+		return 1;
+
+	lc->in_sync_hint = pkg.in_sync_hint;
+
+	return (int)pkg.is_recovering;
+}
+
+static struct dm_dirty_log_type _userspace_type = {
+	.name = "userspace",
+	.module = THIS_MODULE,
+	.ctr = userspace_ctr,
+	.dtr = userspace_dtr,
+	.presuspend = userspace_presuspend,
+	.postsuspend = userspace_postsuspend,
+	.resume = userspace_resume,
+	.get_region_size = userspace_get_region_size,
+	.is_clean = userspace_is_clean,
+	.in_sync = userspace_in_sync,
+	.flush = userspace_flush,
+	.mark_region = userspace_mark_region,
+	.clear_region = userspace_clear_region,
+	.get_resync_work = userspace_get_resync_work,
+	.set_region_sync = userspace_set_region_sync,
+	.get_sync_count = userspace_get_sync_count,
+	.status = userspace_status,
+	.is_remote_recovering = userspace_is_remote_recovering,
+};
+
+static int __init userspace_dirty_log_init(void)
+{
+	int r = 0;
+
+	flush_entry_pool = mempool_create(100, flush_entry_alloc,
+					  flush_entry_free, NULL);
+
+	if (!flush_entry_pool) {
+		DMWARN("Unable to create flush_entry_pool:  No memory.");
+		return -ENOMEM;
+	}
+
+	r = dm_ulog_tfr_init();
+	if (r) {
+		DMWARN("Unable to initialize userspace log communications");
+		mempool_destroy(flush_entry_pool);
+		return r;
+	}
+
+	r = dm_dirty_log_type_register(&_userspace_type);
+	if (r) {
+		DMWARN("Couldn't register userspace dirty log type");
+		dm_ulog_tfr_exit();
+		mempool_destroy(flush_entry_pool);
+		return r;
+	}
+
+	DMINFO("version 1.0.0 loaded");
+	return 0;
+}
+
+static void __exit userspace_dirty_log_exit(void)
+{
+	dm_dirty_log_type_unregister(&_userspace_type);
+	dm_ulog_tfr_exit();
+	mempool_destroy(flush_entry_pool);
+
+	DMINFO("version 1.0.0 unloaded");
+	return;
+}
+
+module_init(userspace_dirty_log_init);
+module_exit(userspace_dirty_log_exit);
+
+MODULE_DESCRIPTION(DM_NAME " userspace dirty log link");
+MODULE_AUTHOR("Jonathan Brassow <dm-devel@redhat.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/md/dm-log-userspace-transfer.c b/drivers/md/dm-log-userspace-transfer.c
new file mode 100644
index 000000000000..0ca1ee768a1f
--- /dev/null
+++ b/drivers/md/dm-log-userspace-transfer.c
@@ -0,0 +1,276 @@
+/*
+ * Copyright (C) 2006-2009 Red Hat, Inc.
+ *
+ * This file is released under the LGPL.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <net/sock.h>
+#include <linux/workqueue.h>
+#include <linux/connector.h>
+#include <linux/device-mapper.h>
+#include <linux/dm-log-userspace.h>
+
+#include "dm-log-userspace-transfer.h"
+
+static uint32_t dm_ulog_seq;
+
+/*
+ * Netlink/Connector is an unreliable protocol.  How long should
+ * we wait for a response before assuming it was lost and retrying?
+ * (If we do receive a response after this time, it will be discarded
+ * and the response to the resent request will be waited for.
+ */
+#define DM_ULOG_RETRY_TIMEOUT (15 * HZ)
+
+/*
+ * Pre-allocated space for speed
+ */
+#define DM_ULOG_PREALLOCED_SIZE 512
+static struct cn_msg *prealloced_cn_msg;
+static struct dm_ulog_request *prealloced_ulog_tfr;
+
+static struct cb_id ulog_cn_id = {
+	.idx = CN_IDX_DM,
+	.val = CN_VAL_DM_USERSPACE_LOG
+};
+
+static DEFINE_MUTEX(dm_ulog_lock);
+
+struct receiving_pkg {
+	struct list_head list;
+	struct completion complete;
+
+	uint32_t seq;
+
+	int error;
+	size_t *data_size;
+	char *data;
+};
+
+static DEFINE_SPINLOCK(receiving_list_lock);
+static struct list_head receiving_list;
+
+static int dm_ulog_sendto_server(struct dm_ulog_request *tfr)
+{
+	int r;
+	struct cn_msg *msg = prealloced_cn_msg;
+
+	memset(msg, 0, sizeof(struct cn_msg));
+
+	msg->id.idx = ulog_cn_id.idx;
+	msg->id.val = ulog_cn_id.val;
+	msg->ack = 0;
+	msg->seq = tfr->seq;
+	msg->len = sizeof(struct dm_ulog_request) + tfr->data_size;
+
+	r = cn_netlink_send(msg, 0, gfp_any());
+
+	return r;
+}
+
+/*
+ * Parameters for this function can be either msg or tfr, but not
+ * both.  This function fills in the reply for a waiting request.
+ * If just msg is given, then the reply is simply an ACK from userspace
+ * that the request was received.
+ *
+ * Returns: 0 on success, -ENOENT on failure
+ */
+static int fill_pkg(struct cn_msg *msg, struct dm_ulog_request *tfr)
+{
+	uint32_t rtn_seq = (msg) ? msg->seq : (tfr) ? tfr->seq : 0;
+	struct receiving_pkg *pkg;
+
+	/*
+	 * The 'receiving_pkg' entries in this list are statically
+	 * allocated on the stack in 'dm_consult_userspace'.
+	 * Each process that is waiting for a reply from the user
+	 * space server will have an entry in this list.
+	 *
+	 * We are safe to do it this way because the stack space
+	 * is unique to each process, but still addressable by
+	 * other processes.
+	 */
+	list_for_each_entry(pkg, &receiving_list, list) {
+		if (rtn_seq != pkg->seq)
+			continue;
+
+		if (msg) {
+			pkg->error = -msg->ack;
+			/*
+			 * If we are trying again, we will need to know our
+			 * storage capacity.  Otherwise, along with the
+			 * error code, we make explicit that we have no data.
+			 */
+			if (pkg->error != -EAGAIN)
+				*(pkg->data_size) = 0;
+		} else if (tfr->data_size > *(pkg->data_size)) {
+			DMERR("Insufficient space to receive package [%u] "
+			      "(%u vs %lu)", tfr->request_type,
+			      tfr->data_size, *(pkg->data_size));
+
+			*(pkg->data_size) = 0;
+			pkg->error = -ENOSPC;
+		} else {
+			pkg->error = tfr->error;
+			memcpy(pkg->data, tfr->data, tfr->data_size);
+			*(pkg->data_size) = tfr->data_size;
+		}
+		complete(&pkg->complete);
+		return 0;
+	}
+
+	return -ENOENT;
+}
+
+/*
+ * This is the connector callback that delivers data
+ * that was sent from userspace.
+ */
+static void cn_ulog_callback(void *data)
+{
+	struct cn_msg *msg = (struct cn_msg *)data;
+	struct dm_ulog_request *tfr = (struct dm_ulog_request *)(msg + 1);
+
+	spin_lock(&receiving_list_lock);
+	if (msg->len == 0)
+		fill_pkg(msg, NULL);
+	else if (msg->len < sizeof(*tfr))
+		DMERR("Incomplete message received (expected %u, got %u): [%u]",
+		      (unsigned)sizeof(*tfr), msg->len, msg->seq);
+	else
+		fill_pkg(NULL, tfr);
+	spin_unlock(&receiving_list_lock);
+}
+
+/**
+ * dm_consult_userspace
+ * @uuid: log's uuid (must be DM_UUID_LEN in size)
+ * @request_type:  found in include/linux/dm-log-userspace.h
+ * @data: data to tx to the server
+ * @data_size: size of data in bytes
+ * @rdata: place to put return data from server
+ * @rdata_size: value-result (amount of space given/amount of space used)
+ *
+ * rdata_size is undefined on failure.
+ *
+ * Memory used to communicate with userspace is zero'ed
+ * before populating to ensure that no unwanted bits leak
+ * from kernel space to user-space.  All userspace log communications
+ * between kernel and user space go through this function.
+ *
+ * Returns: 0 on success, -EXXX on failure
+ **/
+int dm_consult_userspace(const char *uuid, int request_type,
+			 char *data, size_t data_size,
+			 char *rdata, size_t *rdata_size)
+{
+	int r = 0;
+	size_t dummy = 0;
+	int overhead_size =
+		sizeof(struct dm_ulog_request *) + sizeof(struct cn_msg);
+	struct dm_ulog_request *tfr = prealloced_ulog_tfr;
+	struct receiving_pkg pkg;
+
+	if (data_size > (DM_ULOG_PREALLOCED_SIZE - overhead_size)) {
+		DMINFO("Size of tfr exceeds preallocated size");
+		return -EINVAL;
+	}
+
+	if (!rdata_size)
+		rdata_size = &dummy;
+resend:
+	/*
+	 * We serialize the sending of requests so we can
+	 * use the preallocated space.
+	 */
+	mutex_lock(&dm_ulog_lock);
+
+	memset(tfr, 0, DM_ULOG_PREALLOCED_SIZE - overhead_size);
+	memcpy(tfr->uuid, uuid, DM_UUID_LEN);
+	tfr->seq = dm_ulog_seq++;
+
+	/*
+	 * Must be valid request type (all other bits set to
+	 * zero).  This reserves other bits for possible future
+	 * use.
+	 */
+	tfr->request_type = request_type & DM_ULOG_REQUEST_MASK;
+
+	tfr->data_size = data_size;
+	if (data && data_size)
+		memcpy(tfr->data, data, data_size);
+
+	memset(&pkg, 0, sizeof(pkg));
+	init_completion(&pkg.complete);
+	pkg.seq = tfr->seq;
+	pkg.data_size = rdata_size;
+	pkg.data = rdata;
+	spin_lock(&receiving_list_lock);
+	list_add(&(pkg.list), &receiving_list);
+	spin_unlock(&receiving_list_lock);
+
+	r = dm_ulog_sendto_server(tfr);
+
+	mutex_unlock(&dm_ulog_lock);
+
+	if (r) {
+		DMERR("Unable to send log request [%u] to userspace: %d",
+		      request_type, r);
+		spin_lock(&receiving_list_lock);
+		list_del_init(&(pkg.list));
+		spin_unlock(&receiving_list_lock);
+
+		goto out;
+	}
+
+	r = wait_for_completion_timeout(&(pkg.complete), DM_ULOG_RETRY_TIMEOUT);
+	spin_lock(&receiving_list_lock);
+	list_del_init(&(pkg.list));
+	spin_unlock(&receiving_list_lock);
+	if (!r) {
+		DMWARN("[%s] Request timed out: [%u/%u] - retrying",
+		       (strlen(uuid) > 8) ?
+		       (uuid + (strlen(uuid) - 8)) : (uuid),
+		       request_type, pkg.seq);
+		goto resend;
+	}
+
+	r = pkg.error;
+	if (r == -EAGAIN)
+		goto resend;
+
+out:
+	return r;
+}
+
+int dm_ulog_tfr_init(void)
+{
+	int r;
+	void *prealloced;
+
+	INIT_LIST_HEAD(&receiving_list);
+
+	prealloced = kmalloc(DM_ULOG_PREALLOCED_SIZE, GFP_KERNEL);
+	if (!prealloced)
+		return -ENOMEM;
+
+	prealloced_cn_msg = prealloced;
+	prealloced_ulog_tfr = prealloced + sizeof(struct cn_msg);
+
+	r = cn_add_callback(&ulog_cn_id, "dmlogusr", cn_ulog_callback);
+	if (r) {
+		cn_del_callback(&ulog_cn_id);
+		return r;
+	}
+
+	return 0;
+}
+
+void dm_ulog_tfr_exit(void)
+{
+	cn_del_callback(&ulog_cn_id);
+	kfree(prealloced_cn_msg);
+}
diff --git a/drivers/md/dm-log-userspace-transfer.h b/drivers/md/dm-log-userspace-transfer.h
new file mode 100644
index 000000000000..c26d8e4e2710
--- /dev/null
+++ b/drivers/md/dm-log-userspace-transfer.h
@@ -0,0 +1,18 @@
+/*
+ * Copyright (C) 2006-2009 Red Hat, Inc.
+ *
+ * This file is released under the LGPL.
+ */
+
+#ifndef __DM_LOG_USERSPACE_TRANSFER_H__
+#define __DM_LOG_USERSPACE_TRANSFER_H__
+
+#define DM_MSG_PREFIX "dm-log-userspace"
+
+int dm_ulog_tfr_init(void);
+void dm_ulog_tfr_exit(void);
+int dm_consult_userspace(const char *uuid, int request_type,
+			 char *data, size_t data_size,
+			 char *rdata, size_t *rdata_size);
+
+#endif /* __DM_LOG_USERSPACE_TRANSFER_H__ */
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 03f22076381f..334a3593cdfd 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -57,6 +57,7 @@ header-y += dlmconstants.h
 header-y += dlm_device.h
 header-y += dlm_netlink.h
 header-y += dm-ioctl.h
+header-y += dm-log-userspace.h
 header-y += dn.h
 header-y += dqblk_xfs.h
 header-y += efs_fs_sb.h
diff --git a/include/linux/connector.h b/include/linux/connector.h
index b9966e64604e..b68d27850d51 100644
--- a/include/linux/connector.h
+++ b/include/linux/connector.h
@@ -41,8 +41,10 @@
 #define CN_IDX_BB			0x5	/* BlackBoard, from the TSP GPL sampling framework */
 #define CN_DST_IDX			0x6
 #define CN_DST_VAL			0x1
+#define CN_IDX_DM			0x7	/* Device Mapper */
+#define CN_VAL_DM_USERSPACE_LOG		0x1
 
-#define CN_NETLINK_USERS		7
+#define CN_NETLINK_USERS		8
 
 /*
  * Maximum connector's message size.
diff --git a/include/linux/dm-log-userspace.h b/include/linux/dm-log-userspace.h
new file mode 100644
index 000000000000..642e3017b51f
--- /dev/null
+++ b/include/linux/dm-log-userspace.h
@@ -0,0 +1,386 @@
+/*
+ * Copyright (C) 2006-2009 Red Hat, Inc.
+ *
+ * This file is released under the LGPL.
+ */
+
+#ifndef __DM_LOG_USERSPACE_H__
+#define __DM_LOG_USERSPACE_H__
+
+#include <linux/dm-ioctl.h> /* For DM_UUID_LEN */
+
+/*
+ * The device-mapper userspace log module consists of a kernel component and
+ * a user-space component.  The kernel component implements the API defined
+ * in dm-dirty-log.h.  Its purpose is simply to pass the parameters and
+ * return values of those API functions between kernel and user-space.
+ *
+ * Below are defined the 'request_types' - DM_ULOG_CTR, DM_ULOG_DTR, etc.
+ * These request types represent the different functions in the device-mapper
+ * dirty log API.  Each of these is described in more detail below.
+ *
+ * The user-space program must listen for requests from the kernel (representing
+ * the various API functions) and process them.
+ *
+ * User-space begins by setting up the communication link (error checking
+ * removed for clarity):
+ *	fd = socket(PF_NETLINK, SOCK_DGRAM, NETLINK_CONNECTOR);
+ *	addr.nl_family = AF_NETLINK;
+ *	addr.nl_groups = CN_IDX_DM;
+ *	addr.nl_pid = 0;
+ *	r = bind(fd, (struct sockaddr *) &addr, sizeof(addr));
+ *	opt = addr.nl_groups;
+ *	setsockopt(fd, SOL_NETLINK, NETLINK_ADD_MEMBERSHIP, &opt, sizeof(opt));
+ *
+ * User-space will then wait to receive requests form the kernel, which it
+ * will process as described below.  The requests are received in the form,
+ * ((struct dm_ulog_request) + (additional data)).  Depending on the request
+ * type, there may or may not be 'additional data'.  In the descriptions below,
+ * you will see 'Payload-to-userspace' and 'Payload-to-kernel'.  The
+ * 'Payload-to-userspace' is what the kernel sends in 'additional data' as
+ * necessary parameters to complete the request.  The 'Payload-to-kernel' is
+ * the 'additional data' returned to the kernel that contains the necessary
+ * results of the request.  The 'data_size' field in the dm_ulog_request
+ * structure denotes the availability and amount of payload data.
+ */
+
+/*
+ * DM_ULOG_CTR corresponds to (found in dm-dirty-log.h):
+ * int (*ctr)(struct dm_dirty_log *log, struct dm_target *ti,
+ *	      unsigned argc, char **argv);
+ *
+ * Payload-to-userspace:
+ *	A single string containing all the argv arguments separated by ' 's
+ * Payload-to-kernel:
+ *	None.  ('data_size' in the dm_ulog_request struct should be 0.)
+ *
+ * The UUID contained in the dm_ulog_request structure is the reference that
+ * will be used by all request types to a specific log.  The constructor must
+ * record this assotiation with instance created.
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_CTR                    1
+
+/*
+ * DM_ULOG_DTR corresponds to (found in dm-dirty-log.h):
+ * void (*dtr)(struct dm_dirty_log *log);
+ *
+ * Payload-to-userspace:
+ *	A single string containing all the argv arguments separated by ' 's
+ * Payload-to-kernel:
+ *	None.  ('data_size' in the dm_ulog_request struct should be 0.)
+ *
+ * The UUID contained in the dm_ulog_request structure is all that is
+ * necessary to identify the log instance being destroyed.  There is no
+ * payload data.
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and clearing
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_DTR                    2
+
+/*
+ * DM_ULOG_PRESUSPEND corresponds to (found in dm-dirty-log.h):
+ * int (*presuspend)(struct dm_dirty_log *log);
+ *
+ * Payload-to-userspace:
+ *	None.
+ * Payload-to-kernel:
+ *	None.
+ *
+ * The UUID contained in the dm_ulog_request structure is all that is
+ * necessary to identify the log instance being presuspended.  There is no
+ * payload data.
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_PRESUSPEND             3
+
+/*
+ * DM_ULOG_POSTSUSPEND corresponds to (found in dm-dirty-log.h):
+ * int (*postsuspend)(struct dm_dirty_log *log);
+ *
+ * Payload-to-userspace:
+ *	None.
+ * Payload-to-kernel:
+ *	None.
+ *
+ * The UUID contained in the dm_ulog_request structure is all that is
+ * necessary to identify the log instance being postsuspended.  There is no
+ * payload data.
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_POSTSUSPEND            4
+
+/*
+ * DM_ULOG_RESUME corresponds to (found in dm-dirty-log.h):
+ * int (*resume)(struct dm_dirty_log *log);
+ *
+ * Payload-to-userspace:
+ *	None.
+ * Payload-to-kernel:
+ *	None.
+ *
+ * The UUID contained in the dm_ulog_request structure is all that is
+ * necessary to identify the log instance being resumed.  There is no
+ * payload data.
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_RESUME                 5
+
+/*
+ * DM_ULOG_GET_REGION_SIZE corresponds to (found in dm-dirty-log.h):
+ * uint32_t (*get_region_size)(struct dm_dirty_log *log);
+ *
+ * Payload-to-userspace:
+ *	None.
+ * Payload-to-kernel:
+ *	uint64_t - contains the region size
+ *
+ * The region size is something that was determined at constructor time.
+ * It is returned in the payload area and 'data_size' is set to
+ * reflect this.
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field appropriately.
+ */
+#define DM_ULOG_GET_REGION_SIZE        6
+
+/*
+ * DM_ULOG_IS_CLEAN corresponds to (found in dm-dirty-log.h):
+ * int (*is_clean)(struct dm_dirty_log *log, region_t region);
+ *
+ * Payload-to-userspace:
+ *	uint64_t - the region to get clean status on
+ * Payload-to-kernel:
+ *	int64_t  - 1 if clean, 0 otherwise
+ *
+ * Payload is sizeof(uint64_t) and contains the region for which the clean
+ * status is being made.
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - filling the payload with 0 (not clean) or
+ * 1 (clean), setting 'data_size' and 'error' appropriately.
+ */
+#define DM_ULOG_IS_CLEAN               7
+
+/*
+ * DM_ULOG_IN_SYNC corresponds to (found in dm-dirty-log.h):
+ * int (*in_sync)(struct dm_dirty_log *log, region_t region,
+ *		  int can_block);
+ *
+ * Payload-to-userspace:
+ *	uint64_t - the region to get sync status on
+ * Payload-to-kernel:
+ *	int64_t - 1 if in-sync, 0 otherwise
+ *
+ * Exactly the same as 'is_clean' above, except this time asking "has the
+ * region been recovered?" vs. "is the region not being modified?"
+ */
+#define DM_ULOG_IN_SYNC                8
+
+/*
+ * DM_ULOG_FLUSH corresponds to (found in dm-dirty-log.h):
+ * int (*flush)(struct dm_dirty_log *log);
+ *
+ * Payload-to-userspace:
+ *	None.
+ * Payload-to-kernel:
+ *	None.
+ *
+ * No incoming or outgoing payload.  Simply flush log state to disk.
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and clearing
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_FLUSH                  9
+
+/*
+ * DM_ULOG_MARK_REGION corresponds to (found in dm-dirty-log.h):
+ * void (*mark_region)(struct dm_dirty_log *log, region_t region);
+ *
+ * Payload-to-userspace:
+ *	uint64_t [] - region(s) to mark
+ * Payload-to-kernel:
+ *	None.
+ *
+ * Incoming payload contains the one or more regions to mark dirty.
+ * The number of regions contained in the payload can be determined from
+ * 'data_size/sizeof(uint64_t)'.
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and clearing
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_MARK_REGION           10
+
+/*
+ * DM_ULOG_CLEAR_REGION corresponds to (found in dm-dirty-log.h):
+ * void (*clear_region)(struct dm_dirty_log *log, region_t region);
+ *
+ * Payload-to-userspace:
+ *	uint64_t [] - region(s) to clear
+ * Payload-to-kernel:
+ *	None.
+ *
+ * Incoming payload contains the one or more regions to mark clean.
+ * The number of regions contained in the payload can be determined from
+ * 'data_size/sizeof(uint64_t)'.
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and clearing
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_CLEAR_REGION          11
+
+/*
+ * DM_ULOG_GET_RESYNC_WORK corresponds to (found in dm-dirty-log.h):
+ * int (*get_resync_work)(struct dm_dirty_log *log, region_t *region);
+ *
+ * Payload-to-userspace:
+ *	None.
+ * Payload-to-kernel:
+ *	{
+ *		int64_t i; -- 1 if recovery necessary, 0 otherwise
+ *		uint64_t r; -- The region to recover if i=1
+ *	}
+ * 'data_size' should be set appropriately.
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field appropriately.
+ */
+#define DM_ULOG_GET_RESYNC_WORK       12
+
+/*
+ * DM_ULOG_SET_REGION_SYNC corresponds to (found in dm-dirty-log.h):
+ * void (*set_region_sync)(struct dm_dirty_log *log,
+ *			   region_t region, int in_sync);
+ *
+ * Payload-to-userspace:
+ *	{
+ *		uint64_t - region to set sync state on
+ *		int64_t  - 0 if not-in-sync, 1 if in-sync
+ *	}
+ * Payload-to-kernel:
+ *	None.
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and clearing
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_SET_REGION_SYNC       13
+
+/*
+ * DM_ULOG_GET_SYNC_COUNT corresponds to (found in dm-dirty-log.h):
+ * region_t (*get_sync_count)(struct dm_dirty_log *log);
+ *
+ * Payload-to-userspace:
+ *	None.
+ * Payload-to-kernel:
+ *	uint64_t - the number of in-sync regions
+ *
+ * No incoming payload.  Kernel-bound payload contains the number of
+ * regions that are in-sync (in a size_t).
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_GET_SYNC_COUNT        14
+
+/*
+ * DM_ULOG_STATUS_INFO corresponds to (found in dm-dirty-log.h):
+ * int (*status)(struct dm_dirty_log *log, STATUSTYPE_INFO,
+ *		 char *result, unsigned maxlen);
+ *
+ * Payload-to-userspace:
+ *	None.
+ * Payload-to-kernel:
+ *	Character string containing STATUSTYPE_INFO
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_STATUS_INFO           15
+
+/*
+ * DM_ULOG_STATUS_TABLE corresponds to (found in dm-dirty-log.h):
+ * int (*status)(struct dm_dirty_log *log, STATUSTYPE_TABLE,
+ *		 char *result, unsigned maxlen);
+ *
+ * Payload-to-userspace:
+ *	None.
+ * Payload-to-kernel:
+ *	Character string containing STATUSTYPE_TABLE
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_STATUS_TABLE          16
+
+/*
+ * DM_ULOG_IS_REMOTE_RECOVERING corresponds to (found in dm-dirty-log.h):
+ * int (*is_remote_recovering)(struct dm_dirty_log *log, region_t region);
+ *
+ * Payload-to-userspace:
+ *	uint64_t - region to determine recovery status on
+ * Payload-to-kernel:
+ *	{
+ *		int64_t is_recovering;  -- 0 if no, 1 if yes
+ *		uint64_t in_sync_hint;  -- lowest region still needing resync
+ *	}
+ *
+ * When the request has been processed, user-space must return the
+ * dm_ulog_request to the kernel - setting the 'error' field and
+ * 'data_size' appropriately.
+ */
+#define DM_ULOG_IS_REMOTE_RECOVERING  17
+
+/*
+ * (DM_ULOG_REQUEST_MASK & request_type) to get the request type
+ *
+ * Payload-to-userspace:
+ *	A single string containing all the argv arguments separated by ' 's
+ * Payload-to-kernel:
+ *	None.  ('data_size' in the dm_ulog_request struct should be 0.)
+ *
+ * We are reserving 8 bits of the 32-bit 'request_type' field for the
+ * various request types above.  The remaining 24-bits are currently
+ * set to zero and are reserved for future use and compatibility concerns.
+ *
+ * User-space should always use DM_ULOG_REQUEST_TYPE to aquire the
+ * request type from the 'request_type' field to maintain forward compatibility.
+ */
+#define DM_ULOG_REQUEST_MASK 0xFF
+#define DM_ULOG_REQUEST_TYPE(request_type) \
+	(DM_ULOG_REQUEST_MASK & (request_type))
+
+struct dm_ulog_request {
+	char uuid[DM_UUID_LEN]; /* Ties a request to a specific mirror log */
+	char padding[7];        /* Padding because DM_UUID_LEN = 129 */
+
+	int32_t error;          /* Used to report back processing errors */
+
+	uint32_t seq;           /* Sequence number for request */
+	uint32_t request_type;  /* DM_ULOG_* defined above */
+	uint32_t data_size;     /* How much data (not including this struct) */
+
+	char data[0];
+};
+
+#endif /* __DM_LOG_USERSPACE_H__ */
-- 
cgit v1.2.3-59-g8ed1b


From cec47e3d4a861e1d942b3a580d0bbef2700d2bb2 Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Mon, 22 Jun 2009 10:12:35 +0100
Subject: dm: prepare for request based option

This patch adds core functions for request-based dm.

When struct mapped device (md) is initialized, md->queue has
an I/O scheduler and the following functions are used for
request-based dm as the queue functions:
    make_request_fn: dm_make_request()
    pref_fn:         dm_prep_fn()
    request_fn:      dm_request_fn()
    softirq_done_fn: dm_softirq_done()
    lld_busy_fn:     dm_lld_busy()
Actual initializations are done in another patch (PATCH 2).

Below is a brief summary of how request-based dm behaves, including:
  - making request from bio
  - cloning, mapping and dispatching request
  - completing request and bio
  - suspending md
  - resuming md

  bio to request
  ==============
  md->queue->make_request_fn() (dm_make_request()) calls __make_request()
  for a bio submitted to the md.
  Then, the bio is kept in the queue as a new request or merged into
  another request in the queue if possible.

  Cloning and Mapping
  ===================
  Cloning and mapping are done in md->queue->request_fn() (dm_request_fn()),
  when requests are dispatched after they are sorted by the I/O scheduler.

  dm_request_fn() checks busy state of underlying devices using
  target's busy() function and stops dispatching requests to keep them
  on the dm device's queue if busy.
  It helps better I/O merging, since no merge is done for a request
  once it is dispatched to underlying devices.

  Actual cloning and mapping are done in dm_prep_fn() and map_request()
  called from dm_request_fn().
  dm_prep_fn() clones not only request but also bios of the request
  so that dm can hold bio completion in error cases and prevent
  the bio submitter from noticing the error.
  (See the "Completion" section below for details.)

  After the cloning, the clone is mapped by target's map_rq() function
    and inserted to underlying device's queue using
    blk_insert_cloned_request().

  Completion
  ==========
  Request completion can be hooked by rq->end_io(), but then, all bios
  in the request will have been completed even error cases, and the bio
  submitter will have noticed the error.
  To prevent the bio completion in error cases, request-based dm clones
  both bio and request and hooks both bio->bi_end_io() and rq->end_io():
      bio->bi_end_io(): end_clone_bio()
      rq->end_io():     end_clone_request()

  Summary of the request completion flow is below:
  blk_end_request() for a clone request
    => blk_update_request()
       => bio->bi_end_io() == end_clone_bio() for each clone bio
          => Free the clone bio
          => Success: Complete the original bio (blk_update_request())
             Error:   Don't complete the original bio
    => blk_finish_request()
       => rq->end_io() == end_clone_request()
          => blk_complete_request()
             => dm_softirq_done()
                => Free the clone request
                => Success: Complete the original request (blk_end_request())
                   Error:   Requeue the original request

  end_clone_bio() completes the original request on the size of
  the original bio in successful cases.
  Even if all bios in the original request are completed by that
  completion, the original request must not be completed yet to keep
  the ordering of request completion for the stacking.
  So end_clone_bio() uses blk_update_request() instead of
  blk_end_request().
  In error cases, end_clone_bio() doesn't complete the original bio.
  It just frees the cloned bio and gives over the error handling to
  end_clone_request().

  end_clone_request(), which is called with queue lock held, completes
  the clone request and the original request in a softirq context
  (dm_softirq_done()), which has no queue lock, to avoid a deadlock
  issue on submission of another request during the completion:
      - The submitted request may be mapped to the same device
      - Request submission requires queue lock, but the queue lock
        has been held by itself and it doesn't know that

  The clone request has no clone bio when dm_softirq_done() is called.
  So target drivers can't resubmit it again even error cases.
  Instead, they can ask dm core for requeueing and remapping
  the original request in that cases.

  suspend
  =======
  Request-based dm uses stopping md->queue as suspend of the md.
  For noflush suspend, just stops md->queue.

  For flush suspend, inserts a marker request to the tail of md->queue.
  And dispatches all requests in md->queue until the marker comes to
  the front of md->queue.  Then, stops dispatching request and waits
  for the all dispatched requests to complete.
  After that, completes the marker request, stops md->queue and
  wake up the waiter on the suspend queue, md->wait.

  resume
  ======
  Starts md->queue.

Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-table.c         |  14 +
 drivers/md/dm.c               | 705 +++++++++++++++++++++++++++++++++++++++++-
 drivers/md/dm.h               |   1 +
 include/linux/device-mapper.h |   9 +
 4 files changed, 725 insertions(+), 4 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 09a57113955e..c5f784419f23 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -1080,6 +1080,20 @@ int dm_table_any_congested(struct dm_table *t, int bdi_bits)
 	return r;
 }
 
+int dm_table_any_busy_target(struct dm_table *t)
+{
+	unsigned i;
+	struct dm_target *ti;
+
+	for (i = 0; i < t->num_targets; i++) {
+		ti = t->targets + i;
+		if (ti->type->busy && ti->type->busy(ti))
+			return 1;
+	}
+
+	return 0;
+}
+
 void dm_table_unplug_all(struct dm_table *t)
 {
 	struct dm_dev_internal *dd;
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index f609793a92d0..be003e5fea3d 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -78,7 +78,7 @@ struct dm_rq_target_io {
  */
 struct dm_rq_clone_bio_info {
 	struct bio *orig;
-	struct request *rq;
+	struct dm_rq_target_io *tio;
 };
 
 union map_info *dm_get_mapinfo(struct bio *bio)
@@ -88,6 +88,14 @@ union map_info *dm_get_mapinfo(struct bio *bio)
 	return NULL;
 }
 
+union map_info *dm_get_rq_mapinfo(struct request *rq)
+{
+	if (rq && rq->end_io_data)
+		return &((struct dm_rq_target_io *)rq->end_io_data)->info;
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(dm_get_rq_mapinfo);
+
 #define MINOR_ALLOCED ((void *)-1)
 
 /*
@@ -169,6 +177,12 @@ struct mapped_device {
 	/* forced geometry settings */
 	struct hd_geometry geometry;
 
+	/* marker of flush suspend for request-based dm */
+	struct request suspend_rq;
+
+	/* For saving the address of __make_request for request based dm */
+	make_request_fn *saved_make_request_fn;
+
 	/* sysfs handle */
 	struct kobject kobj;
 
@@ -406,6 +420,26 @@ static void free_tio(struct mapped_device *md, struct dm_target_io *tio)
 	mempool_free(tio, md->tio_pool);
 }
 
+static struct dm_rq_target_io *alloc_rq_tio(struct mapped_device *md)
+{
+	return mempool_alloc(md->tio_pool, GFP_ATOMIC);
+}
+
+static void free_rq_tio(struct dm_rq_target_io *tio)
+{
+	mempool_free(tio, tio->md->tio_pool);
+}
+
+static struct dm_rq_clone_bio_info *alloc_bio_info(struct mapped_device *md)
+{
+	return mempool_alloc(md->io_pool, GFP_ATOMIC);
+}
+
+static void free_bio_info(struct dm_rq_clone_bio_info *info)
+{
+	mempool_free(info, info->tio->md->io_pool);
+}
+
 static void start_io_acct(struct dm_io *io)
 {
 	struct mapped_device *md = io->md;
@@ -615,6 +649,262 @@ static void clone_endio(struct bio *bio, int error)
 	dec_pending(io, error);
 }
 
+/*
+ * Partial completion handling for request-based dm
+ */
+static void end_clone_bio(struct bio *clone, int error)
+{
+	struct dm_rq_clone_bio_info *info = clone->bi_private;
+	struct dm_rq_target_io *tio = info->tio;
+	struct bio *bio = info->orig;
+	unsigned int nr_bytes = info->orig->bi_size;
+
+	bio_put(clone);
+
+	if (tio->error)
+		/*
+		 * An error has already been detected on the request.
+		 * Once error occurred, just let clone->end_io() handle
+		 * the remainder.
+		 */
+		return;
+	else if (error) {
+		/*
+		 * Don't notice the error to the upper layer yet.
+		 * The error handling decision is made by the target driver,
+		 * when the request is completed.
+		 */
+		tio->error = error;
+		return;
+	}
+
+	/*
+	 * I/O for the bio successfully completed.
+	 * Notice the data completion to the upper layer.
+	 */
+
+	/*
+	 * bios are processed from the head of the list.
+	 * So the completing bio should always be rq->bio.
+	 * If it's not, something wrong is happening.
+	 */
+	if (tio->orig->bio != bio)
+		DMERR("bio completion is going in the middle of the request");
+
+	/*
+	 * Update the original request.
+	 * Do not use blk_end_request() here, because it may complete
+	 * the original request before the clone, and break the ordering.
+	 */
+	blk_update_request(tio->orig, 0, nr_bytes);
+}
+
+/*
+ * Don't touch any member of the md after calling this function because
+ * the md may be freed in dm_put() at the end of this function.
+ * Or do dm_get() before calling this function and dm_put() later.
+ */
+static void rq_completed(struct mapped_device *md, int run_queue)
+{
+	int wakeup_waiters = 0;
+	struct request_queue *q = md->queue;
+	unsigned long flags;
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	if (!queue_in_flight(q))
+		wakeup_waiters = 1;
+	spin_unlock_irqrestore(q->queue_lock, flags);
+
+	/* nudge anyone waiting on suspend queue */
+	if (wakeup_waiters)
+		wake_up(&md->wait);
+
+	if (run_queue)
+		blk_run_queue(q);
+
+	/*
+	 * dm_put() must be at the end of this function. See the comment above
+	 */
+	dm_put(md);
+}
+
+static void dm_unprep_request(struct request *rq)
+{
+	struct request *clone = rq->special;
+	struct dm_rq_target_io *tio = clone->end_io_data;
+
+	rq->special = NULL;
+	rq->cmd_flags &= ~REQ_DONTPREP;
+
+	blk_rq_unprep_clone(clone);
+	free_rq_tio(tio);
+}
+
+/*
+ * Requeue the original request of a clone.
+ */
+void dm_requeue_unmapped_request(struct request *clone)
+{
+	struct dm_rq_target_io *tio = clone->end_io_data;
+	struct mapped_device *md = tio->md;
+	struct request *rq = tio->orig;
+	struct request_queue *q = rq->q;
+	unsigned long flags;
+
+	dm_unprep_request(rq);
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	if (elv_queue_empty(q))
+		blk_plug_device(q);
+	blk_requeue_request(q, rq);
+	spin_unlock_irqrestore(q->queue_lock, flags);
+
+	rq_completed(md, 0);
+}
+EXPORT_SYMBOL_GPL(dm_requeue_unmapped_request);
+
+static void __stop_queue(struct request_queue *q)
+{
+	blk_stop_queue(q);
+}
+
+static void stop_queue(struct request_queue *q)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	__stop_queue(q);
+	spin_unlock_irqrestore(q->queue_lock, flags);
+}
+
+static void __start_queue(struct request_queue *q)
+{
+	if (blk_queue_stopped(q))
+		blk_start_queue(q);
+}
+
+static void start_queue(struct request_queue *q)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	__start_queue(q);
+	spin_unlock_irqrestore(q->queue_lock, flags);
+}
+
+/*
+ * Complete the clone and the original request.
+ * Must be called without queue lock.
+ */
+static void dm_end_request(struct request *clone, int error)
+{
+	struct dm_rq_target_io *tio = clone->end_io_data;
+	struct mapped_device *md = tio->md;
+	struct request *rq = tio->orig;
+
+	if (blk_pc_request(rq)) {
+		rq->errors = clone->errors;
+		rq->resid_len = clone->resid_len;
+
+		if (rq->sense)
+			/*
+			 * We are using the sense buffer of the original
+			 * request.
+			 * So setting the length of the sense data is enough.
+			 */
+			rq->sense_len = clone->sense_len;
+	}
+
+	BUG_ON(clone->bio);
+	free_rq_tio(tio);
+
+	blk_end_request_all(rq, error);
+
+	rq_completed(md, 1);
+}
+
+/*
+ * Request completion handler for request-based dm
+ */
+static void dm_softirq_done(struct request *rq)
+{
+	struct request *clone = rq->completion_data;
+	struct dm_rq_target_io *tio = clone->end_io_data;
+	dm_request_endio_fn rq_end_io = tio->ti->type->rq_end_io;
+	int error = tio->error;
+
+	if (!(rq->cmd_flags & REQ_FAILED) && rq_end_io)
+		error = rq_end_io(tio->ti, clone, error, &tio->info);
+
+	if (error <= 0)
+		/* The target wants to complete the I/O */
+		dm_end_request(clone, error);
+	else if (error == DM_ENDIO_INCOMPLETE)
+		/* The target will handle the I/O */
+		return;
+	else if (error == DM_ENDIO_REQUEUE)
+		/* The target wants to requeue the I/O */
+		dm_requeue_unmapped_request(clone);
+	else {
+		DMWARN("unimplemented target endio return value: %d", error);
+		BUG();
+	}
+}
+
+/*
+ * Complete the clone and the original request with the error status
+ * through softirq context.
+ */
+static void dm_complete_request(struct request *clone, int error)
+{
+	struct dm_rq_target_io *tio = clone->end_io_data;
+	struct request *rq = tio->orig;
+
+	tio->error = error;
+	rq->completion_data = clone;
+	blk_complete_request(rq);
+}
+
+/*
+ * Complete the not-mapped clone and the original request with the error status
+ * through softirq context.
+ * Target's rq_end_io() function isn't called.
+ * This may be used when the target's map_rq() function fails.
+ */
+void dm_kill_unmapped_request(struct request *clone, int error)
+{
+	struct dm_rq_target_io *tio = clone->end_io_data;
+	struct request *rq = tio->orig;
+
+	rq->cmd_flags |= REQ_FAILED;
+	dm_complete_request(clone, error);
+}
+EXPORT_SYMBOL_GPL(dm_kill_unmapped_request);
+
+/*
+ * Called with the queue lock held
+ */
+static void end_clone_request(struct request *clone, int error)
+{
+	/*
+	 * For just cleaning up the information of the queue in which
+	 * the clone was dispatched.
+	 * The clone is *NOT* freed actually here because it is alloced from
+	 * dm own mempool and REQ_ALLOCED isn't set in clone->cmd_flags.
+	 */
+	__blk_put_request(clone->q, clone);
+
+	/*
+	 * Actual request completion is done in a softirq context which doesn't
+	 * hold the queue lock.  Otherwise, deadlock could occur because:
+	 *     - another request may be submitted by the upper level driver
+	 *       of the stacking during the completion
+	 *     - the submission which requires queue lock may be done
+	 *       against this queue
+	 */
+	dm_complete_request(clone, error);
+}
+
 static sector_t max_io_len(struct mapped_device *md,
 			   sector_t sector, struct dm_target *ti)
 {
@@ -998,7 +1288,7 @@ out:
  * The request function that just remaps the bio built up by
  * dm_merge_bvec.
  */
-static int dm_request(struct request_queue *q, struct bio *bio)
+static int _dm_request(struct request_queue *q, struct bio *bio)
 {
 	int rw = bio_data_dir(bio);
 	struct mapped_device *md = q->queuedata;
@@ -1035,12 +1325,274 @@ static int dm_request(struct request_queue *q, struct bio *bio)
 	return 0;
 }
 
+static int dm_make_request(struct request_queue *q, struct bio *bio)
+{
+	struct mapped_device *md = q->queuedata;
+
+	if (unlikely(bio_barrier(bio))) {
+		bio_endio(bio, -EOPNOTSUPP);
+		return 0;
+	}
+
+	return md->saved_make_request_fn(q, bio); /* call __make_request() */
+}
+
+static int dm_request_based(struct mapped_device *md)
+{
+	return blk_queue_stackable(md->queue);
+}
+
+static int dm_request(struct request_queue *q, struct bio *bio)
+{
+	struct mapped_device *md = q->queuedata;
+
+	if (dm_request_based(md))
+		return dm_make_request(q, bio);
+
+	return _dm_request(q, bio);
+}
+
+void dm_dispatch_request(struct request *rq)
+{
+	int r;
+
+	if (blk_queue_io_stat(rq->q))
+		rq->cmd_flags |= REQ_IO_STAT;
+
+	rq->start_time = jiffies;
+	r = blk_insert_cloned_request(rq->q, rq);
+	if (r)
+		dm_complete_request(rq, r);
+}
+EXPORT_SYMBOL_GPL(dm_dispatch_request);
+
+static void dm_rq_bio_destructor(struct bio *bio)
+{
+	struct dm_rq_clone_bio_info *info = bio->bi_private;
+	struct mapped_device *md = info->tio->md;
+
+	free_bio_info(info);
+	bio_free(bio, md->bs);
+}
+
+static int dm_rq_bio_constructor(struct bio *bio, struct bio *bio_orig,
+				 void *data)
+{
+	struct dm_rq_target_io *tio = data;
+	struct mapped_device *md = tio->md;
+	struct dm_rq_clone_bio_info *info = alloc_bio_info(md);
+
+	if (!info)
+		return -ENOMEM;
+
+	info->orig = bio_orig;
+	info->tio = tio;
+	bio->bi_end_io = end_clone_bio;
+	bio->bi_private = info;
+	bio->bi_destructor = dm_rq_bio_destructor;
+
+	return 0;
+}
+
+static int setup_clone(struct request *clone, struct request *rq,
+		       struct dm_rq_target_io *tio)
+{
+	int r = blk_rq_prep_clone(clone, rq, tio->md->bs, GFP_ATOMIC,
+				  dm_rq_bio_constructor, tio);
+
+	if (r)
+		return r;
+
+	clone->cmd = rq->cmd;
+	clone->cmd_len = rq->cmd_len;
+	clone->sense = rq->sense;
+	clone->buffer = rq->buffer;
+	clone->end_io = end_clone_request;
+	clone->end_io_data = tio;
+
+	return 0;
+}
+
+static int dm_rq_flush_suspending(struct mapped_device *md)
+{
+	return !md->suspend_rq.special;
+}
+
+/*
+ * Called with the queue lock held.
+ */
+static int dm_prep_fn(struct request_queue *q, struct request *rq)
+{
+	struct mapped_device *md = q->queuedata;
+	struct dm_rq_target_io *tio;
+	struct request *clone;
+
+	if (unlikely(rq == &md->suspend_rq)) {
+		if (dm_rq_flush_suspending(md))
+			return BLKPREP_OK;
+		else
+			/* The flush suspend was interrupted */
+			return BLKPREP_KILL;
+	}
+
+	if (unlikely(rq->special)) {
+		DMWARN("Already has something in rq->special.");
+		return BLKPREP_KILL;
+	}
+
+	tio = alloc_rq_tio(md); /* Only one for each original request */
+	if (!tio)
+		/* -ENOMEM */
+		return BLKPREP_DEFER;
+
+	tio->md = md;
+	tio->ti = NULL;
+	tio->orig = rq;
+	tio->error = 0;
+	memset(&tio->info, 0, sizeof(tio->info));
+
+	clone = &tio->clone;
+	if (setup_clone(clone, rq, tio)) {
+		/* -ENOMEM */
+		free_rq_tio(tio);
+		return BLKPREP_DEFER;
+	}
+
+	rq->special = clone;
+	rq->cmd_flags |= REQ_DONTPREP;
+
+	return BLKPREP_OK;
+}
+
+static void map_request(struct dm_target *ti, struct request *rq,
+			struct mapped_device *md)
+{
+	int r;
+	struct request *clone = rq->special;
+	struct dm_rq_target_io *tio = clone->end_io_data;
+
+	/*
+	 * Hold the md reference here for the in-flight I/O.
+	 * We can't rely on the reference count by device opener,
+	 * because the device may be closed during the request completion
+	 * when all bios are completed.
+	 * See the comment in rq_completed() too.
+	 */
+	dm_get(md);
+
+	tio->ti = ti;
+	r = ti->type->map_rq(ti, clone, &tio->info);
+	switch (r) {
+	case DM_MAPIO_SUBMITTED:
+		/* The target has taken the I/O to submit by itself later */
+		break;
+	case DM_MAPIO_REMAPPED:
+		/* The target has remapped the I/O so dispatch it */
+		dm_dispatch_request(clone);
+		break;
+	case DM_MAPIO_REQUEUE:
+		/* The target wants to requeue the I/O */
+		dm_requeue_unmapped_request(clone);
+		break;
+	default:
+		if (r > 0) {
+			DMWARN("unimplemented target map return value: %d", r);
+			BUG();
+		}
+
+		/* The target wants to complete the I/O */
+		dm_kill_unmapped_request(clone, r);
+		break;
+	}
+}
+
+/*
+ * q->request_fn for request-based dm.
+ * Called with the queue lock held.
+ */
+static void dm_request_fn(struct request_queue *q)
+{
+	struct mapped_device *md = q->queuedata;
+	struct dm_table *map = dm_get_table(md);
+	struct dm_target *ti;
+	struct request *rq;
+
+	/*
+	 * For noflush suspend, check blk_queue_stopped() to immediately
+	 * quit I/O dispatching.
+	 */
+	while (!blk_queue_plugged(q) && !blk_queue_stopped(q)) {
+		rq = blk_peek_request(q);
+		if (!rq)
+			goto plug_and_out;
+
+		if (unlikely(rq == &md->suspend_rq)) { /* Flush suspend maker */
+			if (queue_in_flight(q))
+				/* Not quiet yet.  Wait more */
+				goto plug_and_out;
+
+			/* This device should be quiet now */
+			__stop_queue(q);
+			blk_start_request(rq);
+			__blk_end_request_all(rq, 0);
+			wake_up(&md->wait);
+			goto out;
+		}
+
+		ti = dm_table_find_target(map, blk_rq_pos(rq));
+		if (ti->type->busy && ti->type->busy(ti))
+			goto plug_and_out;
+
+		blk_start_request(rq);
+		spin_unlock(q->queue_lock);
+		map_request(ti, rq, md);
+		spin_lock_irq(q->queue_lock);
+	}
+
+	goto out;
+
+plug_and_out:
+	if (!elv_queue_empty(q))
+		/* Some requests still remain, retry later */
+		blk_plug_device(q);
+
+out:
+	dm_table_put(map);
+
+	return;
+}
+
+int dm_underlying_device_busy(struct request_queue *q)
+{
+	return blk_lld_busy(q);
+}
+EXPORT_SYMBOL_GPL(dm_underlying_device_busy);
+
+static int dm_lld_busy(struct request_queue *q)
+{
+	int r;
+	struct mapped_device *md = q->queuedata;
+	struct dm_table *map = dm_get_table(md);
+
+	if (!map || test_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags))
+		r = 1;
+	else
+		r = dm_table_any_busy_target(map);
+
+	dm_table_put(map);
+
+	return r;
+}
+
 static void dm_unplug_all(struct request_queue *q)
 {
 	struct mapped_device *md = q->queuedata;
 	struct dm_table *map = dm_get_table(md);
 
 	if (map) {
+		if (dm_request_based(md))
+			generic_unplug_device(q);
+
 		dm_table_unplug_all(map);
 		dm_table_put(map);
 	}
@@ -1055,7 +1607,16 @@ static int dm_any_congested(void *congested_data, int bdi_bits)
 	if (!test_bit(DMF_BLOCK_IO_FOR_SUSPEND, &md->flags)) {
 		map = dm_get_table(md);
 		if (map) {
-			r = dm_table_any_congested(map, bdi_bits);
+			/*
+			 * Request-based dm cares about only own queue for
+			 * the query about congestion status of request_queue
+			 */
+			if (dm_request_based(md))
+				r = md->queue->backing_dev_info.state &
+				    bdi_bits;
+			else
+				r = dm_table_any_congested(map, bdi_bits);
+
 			dm_table_put(map);
 		}
 	}
@@ -1458,6 +2019,8 @@ static int dm_wait_for_completion(struct mapped_device *md, int interruptible)
 {
 	int r = 0;
 	DECLARE_WAITQUEUE(wait, current);
+	struct request_queue *q = md->queue;
+	unsigned long flags;
 
 	dm_unplug_all(md->queue);
 
@@ -1467,7 +2030,14 @@ static int dm_wait_for_completion(struct mapped_device *md, int interruptible)
 		set_current_state(interruptible);
 
 		smp_mb();
-		if (!atomic_read(&md->pending))
+		if (dm_request_based(md)) {
+			spin_lock_irqsave(q->queue_lock, flags);
+			if (!queue_in_flight(q) && blk_queue_stopped(q)) {
+				spin_unlock_irqrestore(q->queue_lock, flags);
+				break;
+			}
+			spin_unlock_irqrestore(q->queue_lock, flags);
+		} else if (!atomic_read(&md->pending))
 			break;
 
 		if (interruptible == TASK_INTERRUPTIBLE &&
@@ -1584,6 +2154,67 @@ out:
 	return r;
 }
 
+static void dm_rq_invalidate_suspend_marker(struct mapped_device *md)
+{
+	md->suspend_rq.special = (void *)0x1;
+}
+
+static void dm_rq_abort_suspend(struct mapped_device *md, int noflush)
+{
+	struct request_queue *q = md->queue;
+	unsigned long flags;
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	if (!noflush)
+		dm_rq_invalidate_suspend_marker(md);
+	__start_queue(q);
+	spin_unlock_irqrestore(q->queue_lock, flags);
+}
+
+static void dm_rq_start_suspend(struct mapped_device *md, int noflush)
+{
+	struct request *rq = &md->suspend_rq;
+	struct request_queue *q = md->queue;
+
+	if (noflush)
+		stop_queue(q);
+	else {
+		blk_rq_init(q, rq);
+		blk_insert_request(q, rq, 0, NULL);
+	}
+}
+
+static int dm_rq_suspend_available(struct mapped_device *md, int noflush)
+{
+	int r = 1;
+	struct request *rq = &md->suspend_rq;
+	struct request_queue *q = md->queue;
+	unsigned long flags;
+
+	if (noflush)
+		return r;
+
+	/* The marker must be protected by queue lock if it is in use */
+	spin_lock_irqsave(q->queue_lock, flags);
+	if (unlikely(rq->ref_count)) {
+		/*
+		 * This can happen, when the previous flush suspend was
+		 * interrupted, the marker is still in the queue and
+		 * this flush suspend has been invoked, because we don't
+		 * remove the marker at the time of suspend interruption.
+		 * We have only one marker per mapped_device, so we can't
+		 * start another flush suspend while it is in use.
+		 */
+		BUG_ON(!rq->special); /* The marker should be invalidated */
+		DMWARN("Invalidating the previous flush suspend is still in"
+		       " progress.  Please retry later.");
+		r = 0;
+	}
+	spin_unlock_irqrestore(q->queue_lock, flags);
+
+	return r;
+}
+
 /*
  * Functions to lock and unlock any filesystem running on the
  * device.
@@ -1623,6 +2254,53 @@ static void unlock_fs(struct mapped_device *md)
  * dm_bind_table, dm_suspend must be called to flush any in
  * flight bios and ensure that any further io gets deferred.
  */
+/*
+ * Suspend mechanism in request-based dm.
+ *
+ * After the suspend starts, further incoming requests are kept in
+ * the request_queue and deferred.
+ * Remaining requests in the request_queue at the start of suspend are flushed
+ * if it is flush suspend.
+ * The suspend completes when the following conditions have been satisfied,
+ * so wait for it:
+ *    1. q->in_flight is 0 (which means no in_flight request)
+ *    2. queue has been stopped (which means no request dispatching)
+ *
+ *
+ * Noflush suspend
+ * ---------------
+ * Noflush suspend doesn't need to dispatch remaining requests.
+ * So stop the queue immediately.  Then, wait for all in_flight requests
+ * to be completed or requeued.
+ *
+ * To abort noflush suspend, start the queue.
+ *
+ *
+ * Flush suspend
+ * -------------
+ * Flush suspend needs to dispatch remaining requests.  So stop the queue
+ * after the remaining requests are completed. (Requeued request must be also
+ * re-dispatched and completed.  Until then, we can't stop the queue.)
+ *
+ * During flushing the remaining requests, further incoming requests are also
+ * inserted to the same queue.  To distinguish which requests are to be
+ * flushed, we insert a marker request to the queue at the time of starting
+ * flush suspend, like a barrier.
+ * The dispatching is blocked when the marker is found on the top of the queue.
+ * And the queue is stopped when all in_flight requests are completed, since
+ * that means the remaining requests are completely flushed.
+ * Then, the marker is removed from the queue.
+ *
+ * To abort flush suspend, we also need to take care of the marker, not only
+ * starting the queue.
+ * We don't remove the marker forcibly from the queue since it's against
+ * the block-layer manner.  Instead, we put a invalidated mark on the marker.
+ * When the invalidated marker is found on the top of the queue, it is
+ * immediately removed from the queue, so it doesn't block dispatching.
+ * Because we have only one marker per mapped_device, we can't start another
+ * flush suspend until the invalidated marker is removed from the queue.
+ * So fail and return with -EBUSY in such a case.
+ */
 int dm_suspend(struct mapped_device *md, unsigned suspend_flags)
 {
 	struct dm_table *map = NULL;
@@ -1637,6 +2315,11 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags)
 		goto out_unlock;
 	}
 
+	if (dm_request_based(md) && !dm_rq_suspend_available(md, noflush)) {
+		r = -EBUSY;
+		goto out_unlock;
+	}
+
 	map = dm_get_table(md);
 
 	/*
@@ -1682,6 +2365,9 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags)
 
 	flush_workqueue(md->wq);
 
+	if (dm_request_based(md))
+		dm_rq_start_suspend(md, noflush);
+
 	/*
 	 * At this point no more requests are entering target request routines.
 	 * We call dm_wait_for_completion to wait for all existing requests
@@ -1698,6 +2384,9 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags)
 	if (r < 0) {
 		dm_queue_flush(md);
 
+		if (dm_request_based(md))
+			dm_rq_abort_suspend(md, noflush);
+
 		unlock_fs(md);
 		goto out; /* pushback list is already flushed, so skip flush */
 	}
@@ -1739,6 +2428,14 @@ int dm_resume(struct mapped_device *md)
 
 	dm_queue_flush(md);
 
+	/*
+	 * Flushing deferred I/Os must be done after targets are resumed
+	 * so that mapping of targets can work correctly.
+	 * Request-based dm is queueing the deferred I/Os in its request_queue.
+	 */
+	if (dm_request_based(md))
+		start_queue(md->queue);
+
 	unlock_fs(md);
 
 	clear_bit(DMF_SUSPENDED, &md->flags);
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index 604e85caadf6..8dcabb1caff1 100644
--- a/drivers/md/dm.h
+++ b/drivers/md/dm.h
@@ -50,6 +50,7 @@ void dm_table_presuspend_targets(struct dm_table *t);
 void dm_table_postsuspend_targets(struct dm_table *t);
 int dm_table_resume_targets(struct dm_table *t);
 int dm_table_any_congested(struct dm_table *t, int bdi_bits);
+int dm_table_any_busy_target(struct dm_table *t);
 
 /*
  * To check the return value from dm_table_find_target().
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index e6bf3b8c7bf2..0d6310657f32 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -234,6 +234,7 @@ struct gendisk *dm_disk(struct mapped_device *md);
 int dm_suspended(struct mapped_device *md);
 int dm_noflush_suspending(struct dm_target *ti);
 union map_info *dm_get_mapinfo(struct bio *bio);
+union map_info *dm_get_rq_mapinfo(struct request *rq);
 
 /*
  * Geometry functions.
@@ -396,4 +397,12 @@ static inline unsigned long to_bytes(sector_t n)
 	return (n << SECTOR_SHIFT);
 }
 
+/*-----------------------------------------------------------------
+ * Helper for block layer and dm core operations
+ *---------------------------------------------------------------*/
+void dm_dispatch_request(struct request *rq);
+void dm_requeue_unmapped_request(struct request *rq);
+void dm_kill_unmapped_request(struct request *rq, int error);
+int dm_underlying_device_busy(struct request_queue *q);
+
 #endif	/* _LINUX_DEVICE_MAPPER_H */
-- 
cgit v1.2.3-59-g8ed1b


From e6ee8c0b767540f59e20da3ced282601db8aa502 Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Mon, 22 Jun 2009 10:12:36 +0100
Subject: dm: enable request based option

This patch enables request-based dm.

o Request-based dm and bio-based dm coexist, since there are
  some target drivers which are more fitting to bio-based dm.
  Also, there are other bio-based devices in the kernel
  (e.g. md, loop).
  Since bio-based device can't receive struct request,
  there are some limitations on device stacking between
  bio-based and request-based.

                     type of underlying device
                   bio-based      request-based
   ----------------------------------------------
    bio-based         OK                OK
    request-based     --                OK

  The device type is recognized by the queue flag in the kernel,
  so dm follows that.

o The type of a dm device is decided at the first table binding time.
  Once the type of a dm device is decided, the type can't be changed.

o Mempool allocations are deferred to at the table loading time, since
  mempools for request-based dm are different from those for bio-based
  dm and needed mempool type is fixed by the type of table.

o Currently, request-based dm supports only tables that have a single
  target.  To support multiple targets, we need to support request
  splitting or prevent bio/request from spanning multiple targets.
  The former needs lots of changes in the block layer, and the latter
  needs that all target drivers support merge() function.
  Both will take a time.

Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-ioctl.c |  13 ++++
 drivers/md/dm-table.c | 111 ++++++++++++++++++++++++++++++++++
 drivers/md/dm.c       | 162 ++++++++++++++++++++++++++++++++++++++++++--------
 drivers/md/dm.h       |  25 ++++++++
 4 files changed, 285 insertions(+), 26 deletions(-)

diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c
index 1c871736f48c..7f77f18fcafa 100644
--- a/drivers/md/dm-ioctl.c
+++ b/drivers/md/dm-ioctl.c
@@ -1050,6 +1050,12 @@ static int populate_table(struct dm_table *table,
 		next = spec->next;
 	}
 
+	r = dm_table_set_type(table);
+	if (r) {
+		DMWARN("unable to set table type");
+		return r;
+	}
+
 	return dm_table_complete(table);
 }
 
@@ -1095,6 +1101,13 @@ static int table_load(struct dm_ioctl *param, size_t param_size)
 		goto out;
 	}
 
+	r = dm_table_alloc_md_mempools(t);
+	if (r) {
+		DMWARN("unable to allocate mempools for this table");
+		dm_table_destroy(t);
+		goto out;
+	}
+
 	down_write(&_hash_lock);
 	hc = dm_get_mdptr(md);
 	if (!hc || hc->md != md) {
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index c5f784419f23..aaeb82ed2852 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -41,6 +41,7 @@
 struct dm_table {
 	struct mapped_device *md;
 	atomic_t holders;
+	unsigned type;
 
 	/* btree table */
 	unsigned int depth;
@@ -65,6 +66,8 @@ struct dm_table {
 	/* events get handed up using this callback */
 	void (*event_fn)(void *);
 	void *event_context;
+
+	struct dm_md_mempools *mempools;
 };
 
 /*
@@ -258,6 +261,8 @@ void dm_table_destroy(struct dm_table *t)
 	if (t->devices.next != &t->devices)
 		free_devices(&t->devices);
 
+	dm_free_md_mempools(t->mempools);
+
 	kfree(t);
 }
 
@@ -764,6 +769,99 @@ int dm_table_add_target(struct dm_table *t, const char *type,
 	return r;
 }
 
+int dm_table_set_type(struct dm_table *t)
+{
+	unsigned i;
+	unsigned bio_based = 0, request_based = 0;
+	struct dm_target *tgt;
+	struct dm_dev_internal *dd;
+	struct list_head *devices;
+
+	for (i = 0; i < t->num_targets; i++) {
+		tgt = t->targets + i;
+		if (dm_target_request_based(tgt))
+			request_based = 1;
+		else
+			bio_based = 1;
+
+		if (bio_based && request_based) {
+			DMWARN("Inconsistent table: different target types"
+			       " can't be mixed up");
+			return -EINVAL;
+		}
+	}
+
+	if (bio_based) {
+		/* We must use this table as bio-based */
+		t->type = DM_TYPE_BIO_BASED;
+		return 0;
+	}
+
+	BUG_ON(!request_based); /* No targets in this table */
+
+	/* Non-request-stackable devices can't be used for request-based dm */
+	devices = dm_table_get_devices(t);
+	list_for_each_entry(dd, devices, list) {
+		if (!blk_queue_stackable(bdev_get_queue(dd->dm_dev.bdev))) {
+			DMWARN("table load rejected: including"
+			       " non-request-stackable devices");
+			return -EINVAL;
+		}
+	}
+
+	/*
+	 * Request-based dm supports only tables that have a single target now.
+	 * To support multiple targets, request splitting support is needed,
+	 * and that needs lots of changes in the block-layer.
+	 * (e.g. request completion process for partial completion.)
+	 */
+	if (t->num_targets > 1) {
+		DMWARN("Request-based dm doesn't support multiple targets yet");
+		return -EINVAL;
+	}
+
+	t->type = DM_TYPE_REQUEST_BASED;
+
+	return 0;
+}
+
+unsigned dm_table_get_type(struct dm_table *t)
+{
+	return t->type;
+}
+
+bool dm_table_request_based(struct dm_table *t)
+{
+	return dm_table_get_type(t) == DM_TYPE_REQUEST_BASED;
+}
+
+int dm_table_alloc_md_mempools(struct dm_table *t)
+{
+	unsigned type = dm_table_get_type(t);
+
+	if (unlikely(type == DM_TYPE_NONE)) {
+		DMWARN("no table type is set, can't allocate mempools");
+		return -EINVAL;
+	}
+
+	t->mempools = dm_alloc_md_mempools(type);
+	if (!t->mempools)
+		return -ENOMEM;
+
+	return 0;
+}
+
+void dm_table_free_md_mempools(struct dm_table *t)
+{
+	dm_free_md_mempools(t->mempools);
+	t->mempools = NULL;
+}
+
+struct dm_md_mempools *dm_table_get_md_mempools(struct dm_table *t)
+{
+	return t->mempools;
+}
+
 static int setup_indexes(struct dm_table *t)
 {
 	int i;
@@ -985,6 +1083,19 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 		queue_flag_set_unlocked(QUEUE_FLAG_CLUSTER, q);
 
 	dm_table_set_integrity(t);
+
+	/*
+	 * QUEUE_FLAG_STACKABLE must be set after all queue settings are
+	 * visible to other CPUs because, once the flag is set, incoming bios
+	 * are processed by request-based dm, which refers to the queue
+	 * settings.
+	 * Until the flag set, bios are passed to bio-based dm and queued to
+	 * md->deferred where queue settings are not needed yet.
+	 * Those bios are passed to request-based dm at the resume time.
+	 */
+	smp_mb();
+	if (dm_table_request_based(t))
+		queue_flag_set_unlocked(QUEUE_FLAG_STACKABLE, q);
 }
 
 unsigned int dm_table_get_num_targets(struct dm_table *t)
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index be003e5fea3d..5a843c1f4d64 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -190,6 +190,15 @@ struct mapped_device {
 	struct bio barrier_bio;
 };
 
+/*
+ * For mempools pre-allocation at the table loading time.
+ */
+struct dm_md_mempools {
+	mempool_t *io_pool;
+	mempool_t *tio_pool;
+	struct bio_set *bs;
+};
+
 #define MIN_IOS 256
 static struct kmem_cache *_io_cache;
 static struct kmem_cache *_tio_cache;
@@ -1739,10 +1748,22 @@ static struct mapped_device *alloc_dev(int minor)
 	INIT_LIST_HEAD(&md->uevent_list);
 	spin_lock_init(&md->uevent_lock);
 
-	md->queue = blk_alloc_queue(GFP_KERNEL);
+	md->queue = blk_init_queue(dm_request_fn, NULL);
 	if (!md->queue)
 		goto bad_queue;
 
+	/*
+	 * Request-based dm devices cannot be stacked on top of bio-based dm
+	 * devices.  The type of this dm device has not been decided yet,
+	 * although we initialized the queue using blk_init_queue().
+	 * The type is decided at the first table loading time.
+	 * To prevent problematic device stacking, clear the queue flag
+	 * for request stacking support until then.
+	 *
+	 * This queue is new, so no concurrency on the queue_flags.
+	 */
+	queue_flag_clear_unlocked(QUEUE_FLAG_STACKABLE, md->queue);
+	md->saved_make_request_fn = md->queue->make_request_fn;
 	md->queue->queuedata = md;
 	md->queue->backing_dev_info.congested_fn = dm_any_congested;
 	md->queue->backing_dev_info.congested_data = md;
@@ -1751,18 +1772,9 @@ static struct mapped_device *alloc_dev(int minor)
 	blk_queue_bounce_limit(md->queue, BLK_BOUNCE_ANY);
 	md->queue->unplug_fn = dm_unplug_all;
 	blk_queue_merge_bvec(md->queue, dm_merge_bvec);
-
-	md->io_pool = mempool_create_slab_pool(MIN_IOS, _io_cache);
-	if (!md->io_pool)
-		goto bad_io_pool;
-
-	md->tio_pool = mempool_create_slab_pool(MIN_IOS, _tio_cache);
-	if (!md->tio_pool)
-		goto bad_tio_pool;
-
-	md->bs = bioset_create(16, 0);
-	if (!md->bs)
-		goto bad_no_bioset;
+	blk_queue_softirq_done(md->queue, dm_softirq_done);
+	blk_queue_prep_rq(md->queue, dm_prep_fn);
+	blk_queue_lld_busy(md->queue, dm_lld_busy);
 
 	md->disk = alloc_disk(1);
 	if (!md->disk)
@@ -1804,12 +1816,6 @@ bad_bdev:
 bad_thread:
 	put_disk(md->disk);
 bad_disk:
-	bioset_free(md->bs);
-bad_no_bioset:
-	mempool_destroy(md->tio_pool);
-bad_tio_pool:
-	mempool_destroy(md->io_pool);
-bad_io_pool:
 	blk_cleanup_queue(md->queue);
 bad_queue:
 	free_minor(minor);
@@ -1829,9 +1835,12 @@ static void free_dev(struct mapped_device *md)
 	unlock_fs(md);
 	bdput(md->bdev);
 	destroy_workqueue(md->wq);
-	mempool_destroy(md->tio_pool);
-	mempool_destroy(md->io_pool);
-	bioset_free(md->bs);
+	if (md->tio_pool)
+		mempool_destroy(md->tio_pool);
+	if (md->io_pool)
+		mempool_destroy(md->io_pool);
+	if (md->bs)
+		bioset_free(md->bs);
 	blk_integrity_unregister(md->disk);
 	del_gendisk(md->disk);
 	free_minor(minor);
@@ -1846,6 +1855,29 @@ static void free_dev(struct mapped_device *md)
 	kfree(md);
 }
 
+static void __bind_mempools(struct mapped_device *md, struct dm_table *t)
+{
+	struct dm_md_mempools *p;
+
+	if (md->io_pool && md->tio_pool && md->bs)
+		/* the md already has necessary mempools */
+		goto out;
+
+	p = dm_table_get_md_mempools(t);
+	BUG_ON(!p || md->io_pool || md->tio_pool || md->bs);
+
+	md->io_pool = p->io_pool;
+	p->io_pool = NULL;
+	md->tio_pool = p->tio_pool;
+	p->tio_pool = NULL;
+	md->bs = p->bs;
+	p->bs = NULL;
+
+out:
+	/* mempool bind completed, now no need any mempools in the table */
+	dm_table_free_md_mempools(t);
+}
+
 /*
  * Bind a table to the device.
  */
@@ -1897,6 +1929,18 @@ static int __bind(struct mapped_device *md, struct dm_table *t,
 
 	dm_table_event_callback(t, event_callback, md);
 
+	/*
+	 * The queue hasn't been stopped yet, if the old table type wasn't
+	 * for request-based during suspension.  So stop it to prevent
+	 * I/O mapping before resume.
+	 * This must be done before setting the queue restrictions,
+	 * because request-based dm may be run just after the setting.
+	 */
+	if (dm_table_request_based(t) && !blk_queue_stopped(q))
+		stop_queue(q);
+
+	__bind_mempools(md, t);
+
 	write_lock(&md->map_lock);
 	md->map = t;
 	dm_table_set_restrictions(t, q, limits);
@@ -2110,10 +2154,14 @@ static void dm_wq_work(struct work_struct *work)
 
 		up_write(&md->io_lock);
 
-		if (bio_barrier(c))
-			process_barrier(md, c);
-		else
-			__split_and_process_bio(md, c);
+		if (dm_request_based(md))
+			generic_make_request(c);
+		else {
+			if (bio_barrier(c))
+				process_barrier(md, c);
+			else
+				__split_and_process_bio(md, c);
+		}
 
 		down_write(&md->io_lock);
 	}
@@ -2146,6 +2194,13 @@ int dm_swap_table(struct mapped_device *md, struct dm_table *table)
 	if (r)
 		goto out;
 
+	/* cannot change the device type, once a table is bound */
+	if (md->map &&
+	    (dm_table_get_type(md->map) != dm_table_get_type(table))) {
+		DMWARN("can't change the device type after a table is bound");
+		goto out;
+	}
+
 	__unbind(md);
 	r = __bind(md, table, &limits);
 
@@ -2542,6 +2597,61 @@ int dm_noflush_suspending(struct dm_target *ti)
 }
 EXPORT_SYMBOL_GPL(dm_noflush_suspending);
 
+struct dm_md_mempools *dm_alloc_md_mempools(unsigned type)
+{
+	struct dm_md_mempools *pools = kmalloc(sizeof(*pools), GFP_KERNEL);
+
+	if (!pools)
+		return NULL;
+
+	pools->io_pool = (type == DM_TYPE_BIO_BASED) ?
+			 mempool_create_slab_pool(MIN_IOS, _io_cache) :
+			 mempool_create_slab_pool(MIN_IOS, _rq_bio_info_cache);
+	if (!pools->io_pool)
+		goto free_pools_and_out;
+
+	pools->tio_pool = (type == DM_TYPE_BIO_BASED) ?
+			  mempool_create_slab_pool(MIN_IOS, _tio_cache) :
+			  mempool_create_slab_pool(MIN_IOS, _rq_tio_cache);
+	if (!pools->tio_pool)
+		goto free_io_pool_and_out;
+
+	pools->bs = (type == DM_TYPE_BIO_BASED) ?
+		    bioset_create(16, 0) : bioset_create(MIN_IOS, 0);
+	if (!pools->bs)
+		goto free_tio_pool_and_out;
+
+	return pools;
+
+free_tio_pool_and_out:
+	mempool_destroy(pools->tio_pool);
+
+free_io_pool_and_out:
+	mempool_destroy(pools->io_pool);
+
+free_pools_and_out:
+	kfree(pools);
+
+	return NULL;
+}
+
+void dm_free_md_mempools(struct dm_md_mempools *pools)
+{
+	if (!pools)
+		return;
+
+	if (pools->io_pool)
+		mempool_destroy(pools->io_pool);
+
+	if (pools->tio_pool)
+		mempool_destroy(pools->tio_pool);
+
+	if (pools->bs)
+		bioset_free(pools->bs);
+
+	kfree(pools);
+}
+
 static struct block_device_operations dm_blk_dops = {
 	.open = dm_blk_open,
 	.release = dm_blk_close,
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index 8dcabb1caff1..a7663eba17e2 100644
--- a/drivers/md/dm.h
+++ b/drivers/md/dm.h
@@ -22,6 +22,13 @@
 #define DM_SUSPEND_LOCKFS_FLAG		(1 << 0)
 #define DM_SUSPEND_NOFLUSH_FLAG		(1 << 1)
 
+/*
+ * Type of table and mapped_device's mempool
+ */
+#define DM_TYPE_NONE		0
+#define DM_TYPE_BIO_BASED	1
+#define DM_TYPE_REQUEST_BASED	2
+
 /*
  * List of devices that a metadevice uses and should open/close.
  */
@@ -32,6 +39,7 @@ struct dm_dev_internal {
 };
 
 struct dm_table;
+struct dm_md_mempools;
 
 /*-----------------------------------------------------------------
  * Internal table functions.
@@ -51,12 +59,23 @@ void dm_table_postsuspend_targets(struct dm_table *t);
 int dm_table_resume_targets(struct dm_table *t);
 int dm_table_any_congested(struct dm_table *t, int bdi_bits);
 int dm_table_any_busy_target(struct dm_table *t);
+int dm_table_set_type(struct dm_table *t);
+unsigned dm_table_get_type(struct dm_table *t);
+bool dm_table_request_based(struct dm_table *t);
+int dm_table_alloc_md_mempools(struct dm_table *t);
+void dm_table_free_md_mempools(struct dm_table *t);
+struct dm_md_mempools *dm_table_get_md_mempools(struct dm_table *t);
 
 /*
  * To check the return value from dm_table_find_target().
  */
 #define dm_target_is_valid(t) ((t)->table)
 
+/*
+ * To check whether the target type is request-based or not (bio-based).
+ */
+#define dm_target_request_based(t) ((t)->type->map_rq != NULL)
+
 /*-----------------------------------------------------------------
  * A registry of target types.
  *---------------------------------------------------------------*/
@@ -102,4 +121,10 @@ void dm_kobject_uevent(struct mapped_device *md, enum kobject_action action,
 int dm_kcopyd_init(void);
 void dm_kcopyd_exit(void);
 
+/*
+ * Mempool operations
+ */
+struct dm_md_mempools *dm_alloc_md_mempools(unsigned type);
+void dm_free_md_mempools(struct dm_md_mempools *pools);
+
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 5d67aa2366ccb8257d103d0b43df855605c3c086 Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Mon, 22 Jun 2009 10:12:36 +0100
Subject: dm: do not set QUEUE_ORDERED_DRAIN if request based

Request-based dm doesn't have barrier support yet.
So we need to set QUEUE_ORDERED_DRAIN only for bio-based dm.
Since the device type is decided at the first table loading time,
the flag set is deferred until then.

Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Acked-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-table.c |  5 +++++
 drivers/md/dm.c       | 11 ++++++++++-
 drivers/md/dm.h       |  1 +
 3 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index aaeb82ed2852..4899ebe767c8 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -830,6 +830,11 @@ unsigned dm_table_get_type(struct dm_table *t)
 	return t->type;
 }
 
+bool dm_table_bio_based(struct dm_table *t)
+{
+	return dm_table_get_type(t) == DM_TYPE_BIO_BASED;
+}
+
 bool dm_table_request_based(struct dm_table *t)
 {
 	return dm_table_get_type(t) == DM_TYPE_REQUEST_BASED;
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 5a843c1f4d64..00c768860818 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1768,7 +1768,6 @@ static struct mapped_device *alloc_dev(int minor)
 	md->queue->backing_dev_info.congested_fn = dm_any_congested;
 	md->queue->backing_dev_info.congested_data = md;
 	blk_queue_make_request(md->queue, dm_request);
-	blk_queue_ordered(md->queue, QUEUE_ORDERED_DRAIN, NULL);
 	blk_queue_bounce_limit(md->queue, BLK_BOUNCE_ANY);
 	md->queue->unplug_fn = dm_unplug_all;
 	blk_queue_merge_bvec(md->queue, dm_merge_bvec);
@@ -2201,6 +2200,16 @@ int dm_swap_table(struct mapped_device *md, struct dm_table *table)
 		goto out;
 	}
 
+	/*
+	 * It is enought that blk_queue_ordered() is called only once when
+	 * the first bio-based table is bound.
+	 *
+	 * This setting should be moved to alloc_dev() when request-based dm
+	 * supports barrier.
+	 */
+	if (!md->map && dm_table_bio_based(table))
+		blk_queue_ordered(md->queue, QUEUE_ORDERED_DRAIN, NULL);
+
 	__unbind(md);
 	r = __bind(md, table, &limits);
 
diff --git a/drivers/md/dm.h b/drivers/md/dm.h
index a7663eba17e2..23278ae80f08 100644
--- a/drivers/md/dm.h
+++ b/drivers/md/dm.h
@@ -61,6 +61,7 @@ int dm_table_any_congested(struct dm_table *t, int bdi_bits);
 int dm_table_any_busy_target(struct dm_table *t);
 int dm_table_set_type(struct dm_table *t);
 unsigned dm_table_get_type(struct dm_table *t);
+bool dm_table_bio_based(struct dm_table *t);
 bool dm_table_request_based(struct dm_table *t);
 int dm_table_alloc_md_mempools(struct dm_table *t);
 void dm_table_free_md_mempools(struct dm_table *t);
-- 
cgit v1.2.3-59-g8ed1b


From 523d9297d43cce3fa6de6474b7674329e98743b1 Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Mon, 22 Jun 2009 10:12:37 +0100
Subject: dm: disable interrupt when taking map_lock

This patch disables interrupt when taking map_lock to avoid
lockdep warnings in request-based dm.

request-based dm takes map_lock after taking queue_lock with
disabling interrupt:
  spin_lock_irqsave(queue_lock)
  q->request_fn() == dm_request_fn()
    => dm_get_table()
         => read_lock(map_lock)
while queue_lock could be (but isn't) taken in interrupt context.

Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Acked-by: Christof Schmitt <christof.schmitt@de.ibm.com>
Acked-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 00c768860818..3c6d4ee8921d 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -512,12 +512,13 @@ static void queue_io(struct mapped_device *md, struct bio *bio)
 struct dm_table *dm_get_table(struct mapped_device *md)
 {
 	struct dm_table *t;
+	unsigned long flags;
 
-	read_lock(&md->map_lock);
+	read_lock_irqsave(&md->map_lock, flags);
 	t = md->map;
 	if (t)
 		dm_table_get(t);
-	read_unlock(&md->map_lock);
+	read_unlock_irqrestore(&md->map_lock, flags);
 
 	return t;
 }
@@ -1910,6 +1911,7 @@ static int __bind(struct mapped_device *md, struct dm_table *t,
 {
 	struct request_queue *q = md->queue;
 	sector_t size;
+	unsigned long flags;
 
 	size = dm_table_get_size(t);
 
@@ -1940,10 +1942,10 @@ static int __bind(struct mapped_device *md, struct dm_table *t,
 
 	__bind_mempools(md, t);
 
-	write_lock(&md->map_lock);
+	write_lock_irqsave(&md->map_lock, flags);
 	md->map = t;
 	dm_table_set_restrictions(t, q, limits);
-	write_unlock(&md->map_lock);
+	write_unlock_irqrestore(&md->map_lock, flags);
 
 	return 0;
 }
@@ -1951,14 +1953,15 @@ static int __bind(struct mapped_device *md, struct dm_table *t,
 static void __unbind(struct mapped_device *md)
 {
 	struct dm_table *map = md->map;
+	unsigned long flags;
 
 	if (!map)
 		return;
 
 	dm_table_event_callback(map, NULL, NULL);
-	write_lock(&md->map_lock);
+	write_lock_irqsave(&md->map_lock, flags);
 	md->map = NULL;
-	write_unlock(&md->map_lock);
+	write_unlock_irqrestore(&md->map_lock, flags);
 	dm_table_destroy(map);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From f40c67f0f7e2767f80f7cbcbc1ab86c4113c202e Mon Sep 17 00:00:00 2001
From: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Date: Mon, 22 Jun 2009 10:12:37 +0100
Subject: dm mpath: change to be request based

This patch converts dm-multipath target to request-based from bio-based.

Basically, the patch just converts the I/O unit from struct bio
to struct request.
In the course of the conversion, it also changes the I/O queueing
mechanism.  The change in the I/O queueing is described in details
as follows.

I/O queueing mechanism change
-----------------------------
In I/O submission, map_io(), there is no mechanism change from
bio-based, since the clone request is ready for retry as it is.
However, in I/O complition, do_end_io(), there is a mechanism change
from bio-based, since the clone request is not ready for retry.

In do_end_io() of bio-based, the clone bio has all needed memory
for resubmission.  So the target driver can queue it and resubmit
it later without memory allocations.
The mechanism has almost no overhead.

On the other hand, in do_end_io() of request-based, the clone request
doesn't have clone bios, so the target driver can't resubmit it
as it is.  To resubmit the clone request, memory allocation for
clone bios is needed, and it takes some overheads.
To avoid the overheads just for queueing, the target driver doesn't
queue the clone request inside itself.
Instead, the target driver asks dm core for queueing and remapping
the original request of the clone request, since the overhead for
queueing is just a freeing memory for the clone request.

As a result, the target driver doesn't need to record/restore
the information of the original request for resubmitting
the clone request.  So dm_bio_details in dm_mpath_io is removed.

multipath_busy()
---------------------
The target driver returns "busy", only when the following case:
  o The target driver will map I/Os, if map() function is called
  and
  o The mapped I/Os will wait on underlying device's queue due to
    their congestions, if map() function is called now.

In other cases, the target driver doesn't return "busy".
Otherwise, dm core will keep the I/Os and the target driver can't
do what it wants.
(e.g. the target driver can't map I/Os now, so wants to kill I/Os.)

Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Acked-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-mpath.c | 193 +++++++++++++++++++++++++++++++++-----------------
 1 file changed, 128 insertions(+), 65 deletions(-)

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index f8aeaaa54afe..c70604a20897 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -8,7 +8,6 @@
 #include <linux/device-mapper.h>
 
 #include "dm-path-selector.h"
-#include "dm-bio-record.h"
 #include "dm-uevent.h"
 
 #include <linux/ctype.h>
@@ -83,7 +82,7 @@ struct multipath {
 	unsigned pg_init_count;		/* Number of times pg_init called */
 
 	struct work_struct process_queued_ios;
-	struct bio_list queued_ios;
+	struct list_head queued_ios;
 	unsigned queue_size;
 
 	struct work_struct trigger_event;
@@ -100,7 +99,6 @@ struct multipath {
  */
 struct dm_mpath_io {
 	struct pgpath *pgpath;
-	struct dm_bio_details details;
 	size_t nr_bytes;
 };
 
@@ -194,6 +192,7 @@ static struct multipath *alloc_multipath(struct dm_target *ti)
 	m = kzalloc(sizeof(*m), GFP_KERNEL);
 	if (m) {
 		INIT_LIST_HEAD(&m->priority_groups);
+		INIT_LIST_HEAD(&m->queued_ios);
 		spin_lock_init(&m->lock);
 		m->queue_io = 1;
 		INIT_WORK(&m->process_queued_ios, process_queued_ios);
@@ -318,13 +317,14 @@ static int __must_push_back(struct multipath *m)
 		dm_noflush_suspending(m->ti));
 }
 
-static int map_io(struct multipath *m, struct bio *bio,
+static int map_io(struct multipath *m, struct request *clone,
 		  struct dm_mpath_io *mpio, unsigned was_queued)
 {
 	int r = DM_MAPIO_REMAPPED;
-	size_t nr_bytes = bio->bi_size;
+	size_t nr_bytes = blk_rq_bytes(clone);
 	unsigned long flags;
 	struct pgpath *pgpath;
+	struct block_device *bdev;
 
 	spin_lock_irqsave(&m->lock, flags);
 
@@ -341,16 +341,18 @@ static int map_io(struct multipath *m, struct bio *bio,
 	if ((pgpath && m->queue_io) ||
 	    (!pgpath && m->queue_if_no_path)) {
 		/* Queue for the daemon to resubmit */
-		bio_list_add(&m->queued_ios, bio);
+		list_add_tail(&clone->queuelist, &m->queued_ios);
 		m->queue_size++;
 		if ((m->pg_init_required && !m->pg_init_in_progress) ||
 		    !m->queue_io)
 			queue_work(kmultipathd, &m->process_queued_ios);
 		pgpath = NULL;
 		r = DM_MAPIO_SUBMITTED;
-	} else if (pgpath)
-		bio->bi_bdev = pgpath->path.dev->bdev;
-	else if (__must_push_back(m))
+	} else if (pgpath) {
+		bdev = pgpath->path.dev->bdev;
+		clone->q = bdev_get_queue(bdev);
+		clone->rq_disk = bdev->bd_disk;
+	} else if (__must_push_back(m))
 		r = DM_MAPIO_REQUEUE;
 	else
 		r = -EIO;	/* Failed */
@@ -398,30 +400,31 @@ static void dispatch_queued_ios(struct multipath *m)
 {
 	int r;
 	unsigned long flags;
-	struct bio *bio = NULL, *next;
 	struct dm_mpath_io *mpio;
 	union map_info *info;
+	struct request *clone, *n;
+	LIST_HEAD(cl);
 
 	spin_lock_irqsave(&m->lock, flags);
-	bio = bio_list_get(&m->queued_ios);
+	list_splice_init(&m->queued_ios, &cl);
 	spin_unlock_irqrestore(&m->lock, flags);
 
-	while (bio) {
-		next = bio->bi_next;
-		bio->bi_next = NULL;
+	list_for_each_entry_safe(clone, n, &cl, queuelist) {
+		list_del_init(&clone->queuelist);
 
-		info = dm_get_mapinfo(bio);
+		info = dm_get_rq_mapinfo(clone);
 		mpio = info->ptr;
 
-		r = map_io(m, bio, mpio, 1);
-		if (r < 0)
-			bio_endio(bio, r);
-		else if (r == DM_MAPIO_REMAPPED)
-			generic_make_request(bio);
-		else if (r == DM_MAPIO_REQUEUE)
-			bio_endio(bio, -EIO);
-
-		bio = next;
+		r = map_io(m, clone, mpio, 1);
+		if (r < 0) {
+			mempool_free(mpio, m->mpio_pool);
+			dm_kill_unmapped_request(clone, r);
+		} else if (r == DM_MAPIO_REMAPPED)
+			dm_dispatch_request(clone);
+		else if (r == DM_MAPIO_REQUEUE) {
+			mempool_free(mpio, m->mpio_pool);
+			dm_requeue_unmapped_request(clone);
+		}
 	}
 }
 
@@ -863,21 +866,24 @@ static void multipath_dtr(struct dm_target *ti)
 }
 
 /*
- * Map bios, recording original fields for later in case we have to resubmit
+ * Map cloned requests
  */
-static int multipath_map(struct dm_target *ti, struct bio *bio,
+static int multipath_map(struct dm_target *ti, struct request *clone,
 			 union map_info *map_context)
 {
 	int r;
 	struct dm_mpath_io *mpio;
 	struct multipath *m = (struct multipath *) ti->private;
 
-	mpio = mempool_alloc(m->mpio_pool, GFP_NOIO);
-	dm_bio_record(&mpio->details, bio);
+	mpio = mempool_alloc(m->mpio_pool, GFP_ATOMIC);
+	if (!mpio)
+		/* ENOMEM, requeue */
+		return DM_MAPIO_REQUEUE;
+	memset(mpio, 0, sizeof(*mpio));
 
 	map_context->ptr = mpio;
-	bio->bi_rw |= (1 << BIO_RW_FAILFAST_TRANSPORT);
-	r = map_io(m, bio, mpio, 0);
+	clone->cmd_flags |= REQ_FAILFAST_TRANSPORT;
+	r = map_io(m, clone, mpio, 0);
 	if (r < 0 || r == DM_MAPIO_REQUEUE)
 		mempool_free(mpio, m->mpio_pool);
 
@@ -1158,53 +1164,41 @@ static void activate_path(struct work_struct *work)
 /*
  * end_io handling
  */
-static int do_end_io(struct multipath *m, struct bio *bio,
+static int do_end_io(struct multipath *m, struct request *clone,
 		     int error, struct dm_mpath_io *mpio)
 {
+	/*
+	 * We don't queue any clone request inside the multipath target
+	 * during end I/O handling, since those clone requests don't have
+	 * bio clones.  If we queue them inside the multipath target,
+	 * we need to make bio clones, that requires memory allocation.
+	 * (See drivers/md/dm.c:end_clone_bio() about why the clone requests
+	 *  don't have bio clones.)
+	 * Instead of queueing the clone request here, we queue the original
+	 * request into dm core, which will remake a clone request and
+	 * clone bios for it and resubmit it later.
+	 */
+	int r = DM_ENDIO_REQUEUE;
 	unsigned long flags;
 
-	if (!error)
+	if (!error && !clone->errors)
 		return 0;	/* I/O complete */
 
-	if ((error == -EWOULDBLOCK) && bio_rw_ahead(bio))
-		return error;
-
 	if (error == -EOPNOTSUPP)
 		return error;
 
-	spin_lock_irqsave(&m->lock, flags);
-	if (!m->nr_valid_paths) {
-		if (__must_push_back(m)) {
-			spin_unlock_irqrestore(&m->lock, flags);
-			return DM_ENDIO_REQUEUE;
-		} else if (!m->queue_if_no_path) {
-			spin_unlock_irqrestore(&m->lock, flags);
-			return -EIO;
-		} else {
-			spin_unlock_irqrestore(&m->lock, flags);
-			goto requeue;
-		}
-	}
-	spin_unlock_irqrestore(&m->lock, flags);
-
 	if (mpio->pgpath)
 		fail_path(mpio->pgpath);
 
-      requeue:
-	dm_bio_restore(&mpio->details, bio);
-
-	/* queue for the daemon to resubmit or fail */
 	spin_lock_irqsave(&m->lock, flags);
-	bio_list_add(&m->queued_ios, bio);
-	m->queue_size++;
-	if (!m->queue_io)
-		queue_work(kmultipathd, &m->process_queued_ios);
+	if (!m->nr_valid_paths && !m->queue_if_no_path && !__must_push_back(m))
+		r = -EIO;
 	spin_unlock_irqrestore(&m->lock, flags);
 
-	return DM_ENDIO_INCOMPLETE;	/* io not complete */
+	return r;
 }
 
-static int multipath_end_io(struct dm_target *ti, struct bio *bio,
+static int multipath_end_io(struct dm_target *ti, struct request *clone,
 			    int error, union map_info *map_context)
 {
 	struct multipath *m = ti->private;
@@ -1213,14 +1207,13 @@ static int multipath_end_io(struct dm_target *ti, struct bio *bio,
 	struct path_selector *ps;
 	int r;
 
-	r  = do_end_io(m, bio, error, mpio);
+	r  = do_end_io(m, clone, error, mpio);
 	if (pgpath) {
 		ps = &pgpath->pg->ps;
 		if (ps->type->end_io)
 			ps->type->end_io(ps, &pgpath->path, mpio->nr_bytes);
 	}
-	if (r != DM_ENDIO_INCOMPLETE)
-		mempool_free(mpio, m->mpio_pool);
+	mempool_free(mpio, m->mpio_pool);
 
 	return r;
 }
@@ -1470,6 +1463,75 @@ out:
 	return ret;
 }
 
+static int __pgpath_busy(struct pgpath *pgpath)
+{
+	struct request_queue *q = bdev_get_queue(pgpath->path.dev->bdev);
+
+	return dm_underlying_device_busy(q);
+}
+
+/*
+ * We return "busy", only when we can map I/Os but underlying devices
+ * are busy (so even if we map I/Os now, the I/Os will wait on
+ * the underlying queue).
+ * In other words, if we want to kill I/Os or queue them inside us
+ * due to map unavailability, we don't return "busy".  Otherwise,
+ * dm core won't give us the I/Os and we can't do what we want.
+ */
+static int multipath_busy(struct dm_target *ti)
+{
+	int busy = 0, has_active = 0;
+	struct multipath *m = ti->private;
+	struct priority_group *pg;
+	struct pgpath *pgpath;
+	unsigned long flags;
+
+	spin_lock_irqsave(&m->lock, flags);
+
+	/* Guess which priority_group will be used at next mapping time */
+	if (unlikely(!m->current_pgpath && m->next_pg))
+		pg = m->next_pg;
+	else if (likely(m->current_pg))
+		pg = m->current_pg;
+	else
+		/*
+		 * We don't know which pg will be used at next mapping time.
+		 * We don't call __choose_pgpath() here to avoid to trigger
+		 * pg_init just by busy checking.
+		 * So we don't know whether underlying devices we will be using
+		 * at next mapping time are busy or not. Just try mapping.
+		 */
+		goto out;
+
+	/*
+	 * If there is one non-busy active path at least, the path selector
+	 * will be able to select it. So we consider such a pg as not busy.
+	 */
+	busy = 1;
+	list_for_each_entry(pgpath, &pg->pgpaths, list)
+		if (pgpath->is_active) {
+			has_active = 1;
+
+			if (!__pgpath_busy(pgpath)) {
+				busy = 0;
+				break;
+			}
+		}
+
+	if (!has_active)
+		/*
+		 * No active path in this pg, so this pg won't be used and
+		 * the current_pg will be changed at next mapping time.
+		 * We need to try mapping to determine it.
+		 */
+		busy = 0;
+
+out:
+	spin_unlock_irqrestore(&m->lock, flags);
+
+	return busy;
+}
+
 /*-----------------------------------------------------------------
  * Module setup
  *---------------------------------------------------------------*/
@@ -1479,14 +1541,15 @@ static struct target_type multipath_target = {
 	.module = THIS_MODULE,
 	.ctr = multipath_ctr,
 	.dtr = multipath_dtr,
-	.map = multipath_map,
-	.end_io = multipath_end_io,
+	.map_rq = multipath_map,
+	.rq_end_io = multipath_end_io,
 	.presuspend = multipath_presuspend,
 	.resume = multipath_resume,
 	.status = multipath_status,
 	.message = multipath_message,
 	.ioctl  = multipath_ioctl,
 	.iterate_devices = multipath_iterate_devices,
+	.busy = multipath_busy,
 };
 
 static int __init dm_multipath_init(void)
-- 
cgit v1.2.3-59-g8ed1b


From bd974240c9a7c6c560504bf390cd8985a16b68f6 Mon Sep 17 00:00:00 2001
From: Oskar Schirmer <os@emlix.com>
Date: Wed, 10 Jun 2009 12:58:45 -0700
Subject: xtensa: cache inquiry and unaligned cache handling functions

The existing xtensa cache handling functions work on page-aligned
memory regions.

These functions are needed for the s6000 dma engine which can work on
a byte-granularity.

Signed-off-by: Oskar Schirmer <os@emlix.com>
Cc: Johannes Weiner <jw@emlix.com>
Cc: Daniel Glockner <dg@emlix.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Chris Zankel <chris@zankel.net>
---
 arch/xtensa/include/asm/cacheflush.h | 95 ++++++++++++++++++++++++++++++++++++
 1 file changed, 95 insertions(+)

diff --git a/arch/xtensa/include/asm/cacheflush.h b/arch/xtensa/include/asm/cacheflush.h
index 8fc1c0c8de07..b7b8fbe47c77 100644
--- a/arch/xtensa/include/asm/cacheflush.h
+++ b/arch/xtensa/include/asm/cacheflush.h
@@ -155,5 +155,100 @@ extern void copy_from_user_page(struct vm_area_struct*, struct page*,
 
 #endif
 
+#define XTENSA_CACHEBLK_LOG2	29
+#define XTENSA_CACHEBLK_SIZE	(1 << XTENSA_CACHEBLK_LOG2)
+#define XTENSA_CACHEBLK_MASK	(7 << XTENSA_CACHEBLK_LOG2)
+
+#if XCHAL_HAVE_CACHEATTR
+static inline u32 xtensa_get_cacheattr(void)
+{
+	u32 r;
+	asm volatile("	rsr %0, CACHEATTR" : "=a"(r));
+	return r;
+}
+
+static inline u32 xtensa_get_dtlb1(u32 addr)
+{
+	u32 r = addr & XTENSA_CACHEBLK_MASK;
+	return r | ((xtensa_get_cacheattr() >> (r >> (XTENSA_CACHEBLK_LOG2-2)))
+			& 0xF);
+}
+#else
+static inline u32 xtensa_get_dtlb1(u32 addr)
+{
+	u32 r;
+	asm volatile("	rdtlb1 %0, %1" : "=a"(r) : "a"(addr));
+	asm volatile("	dsync");
+	return r;
+}
+
+static inline u32 xtensa_get_cacheattr(void)
+{
+	u32 r = 0;
+	u32 a = 0;
+	do {
+		a -= XTENSA_CACHEBLK_SIZE;
+		r = (r << 4) | (xtensa_get_dtlb1(a) & 0xF);
+	} while (a);
+	return r;
+}
+#endif
+
+static inline int xtensa_need_flush_dma_source(u32 addr)
+{
+	return (xtensa_get_dtlb1(addr) & ((1 << XCHAL_CA_BITS) - 1)) >= 4;
+}
+
+static inline int xtensa_need_invalidate_dma_destination(u32 addr)
+{
+	return (xtensa_get_dtlb1(addr) & ((1 << XCHAL_CA_BITS) - 1)) != 2;
+}
+
+static inline void flush_dcache_unaligned(u32 addr, u32 size)
+{
+	u32 cnt;
+	if (size) {
+		cnt = (size + ((XCHAL_DCACHE_LINESIZE - 1) & addr)
+			+ XCHAL_DCACHE_LINESIZE - 1) / XCHAL_DCACHE_LINESIZE;
+		while (cnt--) {
+			asm volatile("	dhwb %0, 0" : : "a"(addr));
+			addr += XCHAL_DCACHE_LINESIZE;
+		}
+		asm volatile("	dsync");
+	}
+}
+
+static inline void invalidate_dcache_unaligned(u32 addr, u32 size)
+{
+	int cnt;
+	if (size) {
+		asm volatile("	dhwbi %0, 0 ;" : : "a"(addr));
+		cnt = (size + ((XCHAL_DCACHE_LINESIZE - 1) & addr)
+			- XCHAL_DCACHE_LINESIZE - 1) / XCHAL_DCACHE_LINESIZE;
+		while (cnt-- > 0) {
+			asm volatile("	dhi %0, %1" : : "a"(addr),
+						"n"(XCHAL_DCACHE_LINESIZE));
+			addr += XCHAL_DCACHE_LINESIZE;
+		}
+		asm volatile("	dhwbi %0, %1" : : "a"(addr),
+						"n"(XCHAL_DCACHE_LINESIZE));
+		asm volatile("	dsync");
+	}
+}
+
+static inline void flush_invalidate_dcache_unaligned(u32 addr, u32 size)
+{
+	u32 cnt;
+	if (size) {
+		cnt = (size + ((XCHAL_DCACHE_LINESIZE - 1) & addr)
+			+ XCHAL_DCACHE_LINESIZE - 1) / XCHAL_DCACHE_LINESIZE;
+		while (cnt--) {
+			asm volatile("	dhwbi %0, 0" : : "a"(addr));
+			addr += XCHAL_DCACHE_LINESIZE;
+		}
+		asm volatile("	dsync");
+	}
+}
+
 #endif /* __KERNEL__ */
 #endif /* _XTENSA_CACHEFLUSH_H */
-- 
cgit v1.2.3-59-g8ed1b


From 1beee21030ed3dc39a41c7b524dbc1a318b518bd Mon Sep 17 00:00:00 2001
From: Daniel Glöckner <dg@emlix.com>
Date: Tue, 5 May 2009 15:03:21 +0000
Subject: xtensa: allow variant to initialize own irq chips
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There was already a PLATFORM_NR_IRQS define, which is now accompanied
by a VARIANT_NR_IRQS. To be able to initialize these interrupts,
init_IRQ now calls a variant specific hook.

Changes compared to v1:
- adapted to new CONFIG_VARIANT_IRQ_EXT
- removed definition and call of platform_init_IRQ as there already
  is a platform_init_irq defined in asm/platform.h with a weak default
  in kernel/platform.c
- renamed variant_init_IRQ to variant_init_irq

Note that I could not find the call site of platform_init_irq although
it is stated in platform.h that it is called from init_IRQ.

Signed-off-by: Daniel Glöckner <dg@emlix.com>
Signed-off-by: Chris Zankel <chris@zankel.net>
---
 arch/xtensa/include/asm/irq.h | 12 +++++++++++-
 arch/xtensa/kernel/irq.c      |  2 ++
 2 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/arch/xtensa/include/asm/irq.h b/arch/xtensa/include/asm/irq.h
index dfac82dc52ad..4c0ccc9c4f4c 100644
--- a/arch/xtensa/include/asm/irq.h
+++ b/arch/xtensa/include/asm/irq.h
@@ -11,6 +11,7 @@
 #ifndef _XTENSA_IRQ_H
 #define _XTENSA_IRQ_H
 
+#include <linux/init.h>
 #include <platform/hardware.h>
 #include <variant/core.h>
 
@@ -21,11 +22,20 @@ static inline void variant_irq_enable(unsigned int irq) { }
 static inline void variant_irq_disable(unsigned int irq) { }
 #endif
 
+#ifndef VARIANT_NR_IRQS
+# define VARIANT_NR_IRQS 0
+#endif
 #ifndef PLATFORM_NR_IRQS
 # define PLATFORM_NR_IRQS 0
 #endif
 #define XTENSA_NR_IRQS XCHAL_NUM_INTERRUPTS
-#define NR_IRQS (XTENSA_NR_IRQS + PLATFORM_NR_IRQS)
+#define NR_IRQS (XTENSA_NR_IRQS + VARIANT_NR_IRQS + PLATFORM_NR_IRQS)
+
+#if VARIANT_NR_IRQS == 0
+static inline void variant_init_irq(void) { }
+#else
+void variant_init_irq(void) __init;
+#endif
 
 static __inline__ int irq_canonicalize(int irq)
 {
diff --git a/arch/xtensa/kernel/irq.c b/arch/xtensa/kernel/irq.c
index a36c85edd045..a1badb32fcda 100644
--- a/arch/xtensa/kernel/irq.c
+++ b/arch/xtensa/kernel/irq.c
@@ -197,4 +197,6 @@ void __init init_IRQ(void)
 	}
 
 	cached_irq_mask = 0;
+
+	variant_init_irq();
 }
-- 
cgit v1.2.3-59-g8ed1b


From f24e552c2dc3221dc7bd2296fd8a705283c4b2d7 Mon Sep 17 00:00:00 2001
From: Oskar Schirmer <os@emlix.com>
Date: Wed, 10 Jun 2009 12:58:45 -0700
Subject: xtensa: s6000 dma engine support

There are four slightly different dma engines on the s6000 family.
One for memory-memory transfers, the other three for memory-device.

This patch implements a platform-specific kernel-API to control these
engines.  It is needed for the network, video, audio peripherals on
s6000.

Signed-off-by: Oskar Schirmer <os@emlix.com>
Signed-off-by: Daniel Glockner <dg@emlix.com>
Signed-off-by: Fabian Godehardt <fg@emlix.com>
Cc: Daniel Glockner <dg@emlix.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Chris Zankel <chris@zankel.net>
---
 arch/xtensa/variants/s6000/Makefile               |   2 +-
 arch/xtensa/variants/s6000/dmac.c                 | 173 ++++++++++
 arch/xtensa/variants/s6000/include/variant/dmac.h | 387 ++++++++++++++++++++++
 3 files changed, 561 insertions(+), 1 deletion(-)
 create mode 100644 arch/xtensa/variants/s6000/dmac.c
 create mode 100644 arch/xtensa/variants/s6000/include/variant/dmac.h

diff --git a/arch/xtensa/variants/s6000/Makefile b/arch/xtensa/variants/s6000/Makefile
index d83f3805130c..3e7ef0a0c498 100644
--- a/arch/xtensa/variants/s6000/Makefile
+++ b/arch/xtensa/variants/s6000/Makefile
@@ -1,4 +1,4 @@
 # s6000 Makefile
 
-obj-y		+= irq.o gpio.o
+obj-y		+= irq.o gpio.o dmac.o
 obj-$(CONFIG_XTENSA_CALIBRATE_CCOUNT)	+= delay.o
diff --git a/arch/xtensa/variants/s6000/dmac.c b/arch/xtensa/variants/s6000/dmac.c
new file mode 100644
index 000000000000..dc7f7c573518
--- /dev/null
+++ b/arch/xtensa/variants/s6000/dmac.c
@@ -0,0 +1,173 @@
+/*
+ * Authors:	Oskar Schirmer <os@emlix.com>
+ *		Daniel Gloeckner <dg@emlix.com>
+ * (c) 2008 emlix GmbH http://www.emlix.com
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/io.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/spinlock.h>
+#include <asm/cacheflush.h>
+#include <variant/dmac.h>
+
+/* DMA engine lookup */
+
+struct s6dmac_ctrl s6dmac_ctrl[S6_DMAC_NB];
+
+
+/* DMA control, per engine */
+
+void s6dmac_put_fifo_cache(u32 dmac, int chan, u32 src, u32 dst, u32 size)
+{
+	if (xtensa_need_flush_dma_source(src)) {
+		u32 base = src;
+		u32 span = size;
+		u32 chunk = readl(DMA_CHNL(dmac, chan) + S6_DMA_CMONCHUNK);
+		if (chunk && (size > chunk)) {
+			s32 skip =
+				readl(DMA_CHNL(dmac, chan) + S6_DMA_SRCSKIP);
+			u32 gaps = (size+chunk-1)/chunk - 1;
+			if (skip >= 0) {
+				span += gaps * skip;
+			} else if (-skip > chunk) {
+				s32 decr = gaps * (chunk + skip);
+				base += decr;
+				span = chunk - decr;
+			} else {
+				span = max(span + gaps * skip,
+					(chunk + skip) * gaps - skip);
+			}
+		}
+		flush_dcache_unaligned(base, span);
+	}
+	if (xtensa_need_invalidate_dma_destination(dst)) {
+		u32 base = dst;
+		u32 span = size;
+		u32 chunk = readl(DMA_CHNL(dmac, chan) + S6_DMA_CMONCHUNK);
+		if (chunk && (size > chunk)) {
+			s32 skip =
+				readl(DMA_CHNL(dmac, chan) + S6_DMA_DSTSKIP);
+			u32 gaps = (size+chunk-1)/chunk - 1;
+			if (skip >= 0) {
+				span += gaps * skip;
+			} else if (-skip > chunk) {
+				s32 decr = gaps * (chunk + skip);
+				base += decr;
+				span = chunk - decr;
+			} else {
+				span = max(span + gaps * skip,
+					(chunk + skip) * gaps - skip);
+			}
+		}
+		invalidate_dcache_unaligned(base, span);
+	}
+	s6dmac_put_fifo(dmac, chan, src, dst, size);
+}
+
+void s6dmac_disable_error_irqs(u32 dmac, u32 mask)
+{
+	unsigned long flags;
+	spinlock_t *spinl = &s6dmac_ctrl[_dmac_addr_index(dmac)].lock;
+	spin_lock_irqsave(spinl, flags);
+	_s6dmac_disable_error_irqs(dmac, mask);
+	spin_unlock_irqrestore(spinl, flags);
+}
+
+u32 s6dmac_int_sources(u32 dmac, u32 channel)
+{
+	u32 mask, ret, tmp;
+	mask = 1 << channel;
+
+	tmp = readl(dmac + S6_DMA_TERMCNTIRQSTAT);
+	tmp &= mask;
+	writel(tmp, dmac + S6_DMA_TERMCNTIRQCLR);
+	ret = tmp >> channel;
+
+	tmp = readl(dmac + S6_DMA_PENDCNTIRQSTAT);
+	tmp &= mask;
+	writel(tmp, dmac + S6_DMA_PENDCNTIRQCLR);
+	ret |= (tmp >> channel) << 1;
+
+	tmp = readl(dmac + S6_DMA_LOWWMRKIRQSTAT);
+	tmp &= mask;
+	writel(tmp, dmac + S6_DMA_LOWWMRKIRQCLR);
+	ret |= (tmp >> channel) << 2;
+
+	tmp = readl(dmac + S6_DMA_INTRAW0);
+	tmp &= (mask << S6_DMA_INT0_OVER) | (mask << S6_DMA_INT0_UNDER);
+	writel(tmp, dmac + S6_DMA_INTCLEAR0);
+
+	if (tmp & (mask << S6_DMA_INT0_UNDER))
+		ret |= 1 << 3;
+	if (tmp & (mask << S6_DMA_INT0_OVER))
+		ret |= 1 << 4;
+
+	tmp = readl(dmac + S6_DMA_MASTERERRINFO);
+	mask <<= S6_DMA_INT1_CHANNEL;
+	if (((tmp >> S6_DMA_MASTERERR_CHAN(0)) & S6_DMA_MASTERERR_CHAN_MASK)
+			== channel)
+		mask |= 1 << S6_DMA_INT1_MASTER;
+	if (((tmp >> S6_DMA_MASTERERR_CHAN(1)) & S6_DMA_MASTERERR_CHAN_MASK)
+			== channel)
+		mask |= 1 << (S6_DMA_INT1_MASTER + 1);
+	if (((tmp >> S6_DMA_MASTERERR_CHAN(2)) & S6_DMA_MASTERERR_CHAN_MASK)
+			== channel)
+		mask |= 1 << (S6_DMA_INT1_MASTER + 2);
+
+	tmp = readl(dmac + S6_DMA_INTRAW1) & mask;
+	writel(tmp, dmac + S6_DMA_INTCLEAR1);
+	ret |= ((tmp >> channel) & 1) << 5;
+	ret |= ((tmp >> S6_DMA_INT1_MASTER) & S6_DMA_INT1_MASTER_MASK) << 6;
+
+	return ret;
+}
+
+void s6dmac_release_chan(u32 dmac, int chan)
+{
+	if (chan >= 0)
+		s6dmac_disable_chan(dmac, chan);
+}
+
+
+/* global init */
+
+static inline void __init dmac_init(u32 dmac, u8 chan_nb)
+{
+	s6dmac_ctrl[S6_DMAC_INDEX(dmac)].dmac = dmac;
+	spin_lock_init(&s6dmac_ctrl[S6_DMAC_INDEX(dmac)].lock);
+	s6dmac_ctrl[S6_DMAC_INDEX(dmac)].chan_nb = chan_nb;
+	writel(S6_DMA_INT1_MASTER_MASK << S6_DMA_INT1_MASTER,
+		dmac + S6_DMA_INTCLEAR1);
+}
+
+static inline void __init dmac_master(u32 dmac,
+	u32 m0start, u32 m0end, u32 m1start, u32 m1end)
+{
+	writel(m0start, dmac + S6_DMA_MASTER0START);
+	writel(m0end - 1, dmac + S6_DMA_MASTER0END);
+	writel(m1start, dmac + S6_DMA_MASTER1START);
+	writel(m1end - 1, dmac + S6_DMA_MASTER1END);
+}
+
+static void __init s6_dmac_init(void)
+{
+	dmac_init(S6_REG_LMSDMA, S6_LMSDMA_NB);
+	dmac_master(S6_REG_LMSDMA,
+		S6_MEM_DDR, S6_MEM_PCIE_APER, S6_MEM_EFI, S6_MEM_GMAC);
+	dmac_init(S6_REG_NIDMA, S6_NIDMA_NB);
+	dmac_init(S6_REG_DPDMA, S6_DPDMA_NB);
+	dmac_master(S6_REG_DPDMA,
+		S6_MEM_DDR, S6_MEM_PCIE_APER, S6_REG_DP, S6_REG_DPDMA);
+	dmac_init(S6_REG_HIFDMA, S6_HIFDMA_NB);
+	dmac_master(S6_REG_HIFDMA,
+		S6_MEM_GMAC, S6_MEM_PCIE_CFG, S6_MEM_PCIE_APER, S6_MEM_AUX);
+}
+
+arch_initcall(s6_dmac_init);
diff --git a/arch/xtensa/variants/s6000/include/variant/dmac.h b/arch/xtensa/variants/s6000/include/variant/dmac.h
new file mode 100644
index 000000000000..89ab9484fb71
--- /dev/null
+++ b/arch/xtensa/variants/s6000/include/variant/dmac.h
@@ -0,0 +1,387 @@
+/*
+ * include/asm-xtensa/variant-s6000/dmac.h
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2006 Tensilica Inc.
+ * Copyright (C) 2008 Emlix GmbH <info@emlix.com>
+ * Authors:	Fabian Godehardt <fg@emlix.com>
+ *		Oskar Schirmer <os@emlix.com>
+ *		Daniel Gloeckner <dg@emlix.com>
+ */
+
+#ifndef __ASM_XTENSA_S6000_DMAC_H
+#define __ASM_XTENSA_S6000_DMAC_H
+#include <linux/io.h>
+#include <variant/hardware.h>
+
+/* DMA global */
+
+#define S6_DMA_INTSTAT0		0x000
+#define S6_DMA_INTSTAT1		0x004
+#define S6_DMA_INTENABLE0	0x008
+#define S6_DMA_INTENABLE1	0x00C
+#define S6_DMA_INTRAW0		0x010
+#define S6_DMA_INTRAW1		0x014
+#define S6_DMA_INTCLEAR0	0x018
+#define S6_DMA_INTCLEAR1	0x01C
+#define S6_DMA_INTSET0		0x020
+#define S6_DMA_INTSET1		0x024
+#define S6_DMA_INT0_UNDER		0
+#define S6_DMA_INT0_OVER		16
+#define S6_DMA_INT1_CHANNEL		0
+#define S6_DMA_INT1_MASTER		16
+#define S6_DMA_INT1_MASTER_MASK			7
+#define S6_DMA_TERMCNTIRQSTAT	0x028
+#define S6_DMA_TERMCNTIRQCLR	0x02C
+#define S6_DMA_TERMCNTIRQSET	0x030
+#define S6_DMA_PENDCNTIRQSTAT	0x034
+#define S6_DMA_PENDCNTIRQCLR	0x038
+#define S6_DMA_PENDCNTIRQSET	0x03C
+#define S6_DMA_LOWWMRKIRQSTAT	0x040
+#define S6_DMA_LOWWMRKIRQCLR	0x044
+#define S6_DMA_LOWWMRKIRQSET	0x048
+#define S6_DMA_MASTERERRINFO	0x04C
+#define S6_DMA_MASTERERR_CHAN(n)	(4*(n))
+#define S6_DMA_MASTERERR_CHAN_MASK		0xF
+#define S6_DMA_DESCRFIFO0	0x050
+#define S6_DMA_DESCRFIFO1	0x054
+#define S6_DMA_DESCRFIFO2	0x058
+#define S6_DMA_DESCRFIFO2_AUTODISABLE	24
+#define S6_DMA_DESCRFIFO3	0x05C
+#define S6_DMA_MASTER0START	0x060
+#define S6_DMA_MASTER0END	0x064
+#define S6_DMA_MASTER1START	0x068
+#define S6_DMA_MASTER1END	0x06C
+#define S6_DMA_NEXTFREE		0x070
+#define S6_DMA_NEXTFREE_CHAN		0
+#define S6_DMA_NEXTFREE_CHAN_MASK	0x1F
+#define S6_DMA_NEXTFREE_ENA		16
+#define S6_DMA_NEXTFREE_ENA_MASK	((1 << 16) - 1)
+#define S6_DMA_DPORTCTRLGRP(p)	((p) * 4 + 0x074)
+#define S6_DMA_DPORTCTRLGRP_FRAMEREP	0
+#define S6_DMA_DPORTCTRLGRP_NRCHANS	1
+#define S6_DMA_DPORTCTRLGRP_NRCHANS_1		0
+#define S6_DMA_DPORTCTRLGRP_NRCHANS_3		1
+#define S6_DMA_DPORTCTRLGRP_NRCHANS_4		2
+#define S6_DMA_DPORTCTRLGRP_NRCHANS_2		3
+#define S6_DMA_DPORTCTRLGRP_ENA		31
+
+
+/* DMA per channel */
+
+#define DMA_CHNL(dmac, n)	((dmac) + 0x1000 + (n) * 0x100)
+#define DMA_INDEX_CHNL(addr)	(((addr) >> 8) & 0xF)
+#define DMA_MASK_DMAC(addr)	((addr) & 0xFFFF0000)
+#define S6_DMA_CHNCTRL		0x000
+#define S6_DMA_CHNCTRL_ENABLE		0
+#define S6_DMA_CHNCTRL_PAUSE		1
+#define S6_DMA_CHNCTRL_PRIO		2
+#define S6_DMA_CHNCTRL_PRIO_MASK		3
+#define S6_DMA_CHNCTRL_PERIPHXFER	4
+#define S6_DMA_CHNCTRL_PERIPHENA	5
+#define S6_DMA_CHNCTRL_SRCINC		6
+#define S6_DMA_CHNCTRL_DSTINC		7
+#define S6_DMA_CHNCTRL_BURSTLOG		8
+#define S6_DMA_CHNCTRL_BURSTLOG_MASK		7
+#define S6_DMA_CHNCTRL_DESCFIFODEPTH	12
+#define S6_DMA_CHNCTRL_DESCFIFODEPTH_MASK	0x1F
+#define S6_DMA_CHNCTRL_DESCFIFOFULL	17
+#define S6_DMA_CHNCTRL_BWCONSEL		18
+#define S6_DMA_CHNCTRL_BWCONENA		19
+#define S6_DMA_CHNCTRL_PENDGCNTSTAT	20
+#define S6_DMA_CHNCTRL_PENDGCNTSTAT_MASK	0x3F
+#define S6_DMA_CHNCTRL_LOWWMARK		26
+#define S6_DMA_CHNCTRL_LOWWMARK_MASK		0xF
+#define S6_DMA_CHNCTRL_TSTAMP		30
+#define S6_DMA_TERMCNTNB	0x004
+#define S6_DMA_TERMCNTNB_MASK			0xFFFF
+#define S6_DMA_TERMCNTTMO	0x008
+#define S6_DMA_TERMCNTSTAT	0x00C
+#define S6_DMA_TERMCNTSTAT_MASK		0xFF
+#define S6_DMA_CMONCHUNK	0x010
+#define S6_DMA_SRCSKIP		0x014
+#define S6_DMA_DSTSKIP		0x018
+#define S6_DMA_CUR_SRC		0x024
+#define S6_DMA_CUR_DST		0x028
+#define S6_DMA_TIMESTAMP	0x030
+
+/* DMA channel lists */
+
+#define S6_DPDMA_CHAN(stream, channel)	(4 * (stream) + (channel))
+#define S6_DPDMA_NB	16
+
+#define S6_HIFDMA_GMACTX	0
+#define S6_HIFDMA_GMACRX	1
+#define S6_HIFDMA_I2S0		2
+#define S6_HIFDMA_I2S1		3
+#define S6_HIFDMA_EGIB		4
+#define S6_HIFDMA_PCITX		5
+#define S6_HIFDMA_PCIRX		6
+#define S6_HIFDMA_NB	7
+
+#define S6_NIDMA_NB	4
+
+#define S6_LMSDMA_NB	12
+
+/* controller access */
+
+#define S6_DMAC_NB	4
+#define S6_DMAC_INDEX(dmac)	(((unsigned)(dmac) >> 18) % S6_DMAC_NB)
+
+struct s6dmac_ctrl {
+	u32 dmac;
+	spinlock_t lock;
+	u8 chan_nb;
+};
+
+extern struct s6dmac_ctrl s6dmac_ctrl[S6_DMAC_NB];
+
+
+/* DMA control, per channel */
+
+static inline int s6dmac_fifo_full(u32 dmac, int chan)
+{
+	return (readl(DMA_CHNL(dmac, chan) + S6_DMA_CHNCTRL)
+		& (1 << S6_DMA_CHNCTRL_DESCFIFOFULL)) && 1;
+}
+
+static inline int s6dmac_termcnt_irq(u32 dmac, int chan)
+{
+	u32 m = 1 << chan;
+	int r = (readl(dmac + S6_DMA_TERMCNTIRQSTAT) & m) && 1;
+	if (r)
+		writel(m, dmac + S6_DMA_TERMCNTIRQCLR);
+	return r;
+}
+
+static inline int s6dmac_pendcnt_irq(u32 dmac, int chan)
+{
+	u32 m = 1 << chan;
+	int r = (readl(dmac + S6_DMA_PENDCNTIRQSTAT) & m) && 1;
+	if (r)
+		writel(m, dmac + S6_DMA_PENDCNTIRQCLR);
+	return r;
+}
+
+static inline int s6dmac_lowwmark_irq(u32 dmac, int chan)
+{
+	int r = (readl(dmac + S6_DMA_LOWWMRKIRQSTAT) & (1 << chan)) ? 1 : 0;
+	if (r)
+		writel(1 << chan, dmac + S6_DMA_LOWWMRKIRQCLR);
+	return r;
+}
+
+static inline u32 s6dmac_pending_count(u32 dmac, int chan)
+{
+	return (readl(DMA_CHNL(dmac, chan) + S6_DMA_CHNCTRL)
+			>> S6_DMA_CHNCTRL_PENDGCNTSTAT)
+		& S6_DMA_CHNCTRL_PENDGCNTSTAT_MASK;
+}
+
+static inline void s6dmac_set_terminal_count(u32 dmac, int chan, u32 n)
+{
+	n &= S6_DMA_TERMCNTNB_MASK;
+	n |= readl(DMA_CHNL(dmac, chan) + S6_DMA_TERMCNTNB)
+	      & ~S6_DMA_TERMCNTNB_MASK;
+	writel(n, DMA_CHNL(dmac, chan) + S6_DMA_TERMCNTNB);
+}
+
+static inline u32 s6dmac_get_terminal_count(u32 dmac, int chan)
+{
+	return (readl(DMA_CHNL(dmac, chan) + S6_DMA_TERMCNTNB))
+		& S6_DMA_TERMCNTNB_MASK;
+}
+
+static inline u32 s6dmac_timestamp(u32 dmac, int chan)
+{
+	return readl(DMA_CHNL(dmac, chan) + S6_DMA_TIMESTAMP);
+}
+
+static inline u32 s6dmac_cur_src(u32 dmac, int chan)
+{
+	return readl(DMA_CHNL(dmac, chan) + S6_DMA_CUR_SRC);
+}
+
+static inline u32 s6dmac_cur_dst(u32 dmac, int chan)
+{
+	return readl(DMA_CHNL(dmac, chan) + S6_DMA_CUR_DST);
+}
+
+static inline void s6dmac_disable_chan(u32 dmac, int chan)
+{
+	u32 ctrl;
+	writel(readl(DMA_CHNL(dmac, chan) + S6_DMA_CHNCTRL)
+		& ~(1 << S6_DMA_CHNCTRL_ENABLE),
+		DMA_CHNL(dmac, chan) + S6_DMA_CHNCTRL);
+	do
+		ctrl = readl(DMA_CHNL(dmac, chan) + S6_DMA_CHNCTRL);
+	while (ctrl & (1 << S6_DMA_CHNCTRL_ENABLE));
+}
+
+static inline void s6dmac_set_stride_skip(u32 dmac, int chan,
+	int comchunk,		/* 0: disable scatter/gather */
+	int srcskip, int dstskip)
+{
+	writel(comchunk, DMA_CHNL(dmac, chan) + S6_DMA_CMONCHUNK);
+	writel(srcskip, DMA_CHNL(dmac, chan) + S6_DMA_SRCSKIP);
+	writel(dstskip, DMA_CHNL(dmac, chan) + S6_DMA_DSTSKIP);
+}
+
+static inline void s6dmac_enable_chan(u32 dmac, int chan,
+	int prio,               /* 0 (highest) .. 3 (lowest) */
+	int periphxfer,         /* <0: disable p.req.line, 0..1: mode */
+	int srcinc, int dstinc, /* 0: dont increment src/dst address */
+	int comchunk,		/* 0: disable scatter/gather */
+	int srcskip, int dstskip,
+	int burstsize,		/* 4 for I2S, 7 for everything else */
+	int bandwidthconserve,  /* <0: disable, 0..1: select */
+	int lowwmark,           /* 0..15 */
+	int timestamp,		/* 0: disable timestamp */
+	int enable)		/* 0: disable for now */
+{
+	writel(1, DMA_CHNL(dmac, chan) + S6_DMA_TERMCNTNB);
+	writel(0, DMA_CHNL(dmac, chan) + S6_DMA_TERMCNTTMO);
+	writel(lowwmark << S6_DMA_CHNCTRL_LOWWMARK,
+		DMA_CHNL(dmac, chan) + S6_DMA_CHNCTRL);
+	s6dmac_set_stride_skip(dmac, chan, comchunk, srcskip, dstskip);
+	writel(((enable ? 1 : 0) << S6_DMA_CHNCTRL_ENABLE) |
+		(prio << S6_DMA_CHNCTRL_PRIO) |
+		(((periphxfer > 0) ? 1 : 0) << S6_DMA_CHNCTRL_PERIPHXFER) |
+		(((periphxfer < 0) ? 0 : 1) << S6_DMA_CHNCTRL_PERIPHENA) |
+		((srcinc ? 1 : 0) << S6_DMA_CHNCTRL_SRCINC) |
+		((dstinc ? 1 : 0) << S6_DMA_CHNCTRL_DSTINC) |
+		(burstsize << S6_DMA_CHNCTRL_BURSTLOG) |
+		(((bandwidthconserve > 0) ? 1 : 0) << S6_DMA_CHNCTRL_BWCONSEL) |
+		(((bandwidthconserve < 0) ? 0 : 1) << S6_DMA_CHNCTRL_BWCONENA) |
+		(lowwmark << S6_DMA_CHNCTRL_LOWWMARK) |
+		((timestamp ? 1 : 0) << S6_DMA_CHNCTRL_TSTAMP),
+		DMA_CHNL(dmac, chan) + S6_DMA_CHNCTRL);
+}
+
+
+/* DMA control, per engine */
+
+static inline unsigned _dmac_addr_index(u32 dmac)
+{
+	unsigned i = S6_DMAC_INDEX(dmac);
+	if (s6dmac_ctrl[i].dmac != dmac)
+		BUG();
+	return i;
+}
+
+static inline void _s6dmac_disable_error_irqs(u32 dmac, u32 mask)
+{
+	writel(mask, dmac + S6_DMA_TERMCNTIRQCLR);
+	writel(mask, dmac + S6_DMA_PENDCNTIRQCLR);
+	writel(mask, dmac + S6_DMA_LOWWMRKIRQCLR);
+	writel(readl(dmac + S6_DMA_INTENABLE0)
+		& ~((mask << S6_DMA_INT0_UNDER) | (mask << S6_DMA_INT0_OVER)),
+		dmac + S6_DMA_INTENABLE0);
+	writel(readl(dmac + S6_DMA_INTENABLE1) & ~(mask << S6_DMA_INT1_CHANNEL),
+		dmac + S6_DMA_INTENABLE1);
+	writel((mask << S6_DMA_INT0_UNDER) | (mask << S6_DMA_INT0_OVER),
+		dmac + S6_DMA_INTCLEAR0);
+	writel(mask << S6_DMA_INT1_CHANNEL, dmac + S6_DMA_INTCLEAR1);
+}
+
+/*
+ * request channel from specified engine
+ * with chan<0, accept any channel
+ * further parameters see s6dmac_enable_chan
+ * returns < 0 upon error, channel nb otherwise
+ */
+static inline int s6dmac_request_chan(u32 dmac, int chan,
+	int prio,
+	int periphxfer,
+	int srcinc, int dstinc,
+	int comchunk,
+	int srcskip, int dstskip,
+	int burstsize,
+	int bandwidthconserve,
+	int lowwmark,
+	int timestamp,
+	int enable)
+{
+	int r = chan;
+	unsigned long flags;
+	spinlock_t *spinl = &s6dmac_ctrl[_dmac_addr_index(dmac)].lock;
+	spin_lock_irqsave(spinl, flags);
+	if (r < 0) {
+		r = (readl(dmac + S6_DMA_NEXTFREE) >> S6_DMA_NEXTFREE_CHAN)
+			& S6_DMA_NEXTFREE_CHAN_MASK;
+	}
+	if (r >= s6dmac_ctrl[_dmac_addr_index(dmac)].chan_nb) {
+		if (chan < 0)
+			r = -EBUSY;
+		else
+			r = -ENXIO;
+	} else if (((readl(dmac + S6_DMA_NEXTFREE) >> S6_DMA_NEXTFREE_ENA)
+			>> r) & 1) {
+		r = -EBUSY;
+	} else {
+		s6dmac_enable_chan(dmac, r, prio, periphxfer,
+			srcinc, dstinc, comchunk, srcskip, dstskip, burstsize,
+			bandwidthconserve, lowwmark, timestamp, enable);
+	}
+	spin_unlock_irqrestore(spinl, flags);
+	return r;
+}
+
+static inline void s6dmac_put_fifo(u32 dmac, int chan,
+	u32 src, u32 dst, u32 size)
+{
+	unsigned long flags;
+	spinlock_t *spinl = &s6dmac_ctrl[_dmac_addr_index(dmac)].lock;
+	spin_lock_irqsave(spinl, flags);
+	writel(src, dmac + S6_DMA_DESCRFIFO0);
+	writel(dst, dmac + S6_DMA_DESCRFIFO1);
+	writel(size, dmac + S6_DMA_DESCRFIFO2);
+	writel(chan, dmac + S6_DMA_DESCRFIFO3);
+	spin_unlock_irqrestore(spinl, flags);
+}
+
+static inline u32 s6dmac_channel_enabled(u32 dmac, int chan)
+{
+	return readl(DMA_CHNL(dmac, chan) + S6_DMA_CHNCTRL) &
+			(1 << S6_DMA_CHNCTRL_ENABLE);
+}
+
+/*
+ * group 1-4 data port channels
+ * with port=0..3, nrch=1-4 channels,
+ * frrep=0/1 (dis- or enable frame repeat)
+ */
+static inline void s6dmac_dp_setup_group(u32 dmac, int port,
+			int nrch, int frrep)
+{
+	const static u8 mask[4] = {0, 3, 1, 2};
+	BUG_ON(dmac != S6_REG_DPDMA);
+	if ((port < 0) || (port > 3) || (nrch < 1) || (nrch > 4))
+		return;
+	writel((mask[nrch - 1] << S6_DMA_DPORTCTRLGRP_NRCHANS)
+		| ((frrep ? 1 : 0) << S6_DMA_DPORTCTRLGRP_FRAMEREP),
+		dmac + S6_DMA_DPORTCTRLGRP(port));
+}
+
+static inline void s6dmac_dp_switch_group(u32 dmac, int port, int enable)
+{
+	u32 tmp;
+	BUG_ON(dmac != S6_REG_DPDMA);
+	tmp = readl(dmac + S6_DMA_DPORTCTRLGRP(port));
+	if (enable)
+		tmp |= (1 << S6_DMA_DPORTCTRLGRP_ENA);
+	else
+		tmp &= ~(1 << S6_DMA_DPORTCTRLGRP_ENA);
+	writel(tmp, dmac + S6_DMA_DPORTCTRLGRP(port));
+}
+
+extern void s6dmac_put_fifo_cache(u32 dmac, int chan,
+	u32 src, u32 dst, u32 size);
+extern void s6dmac_disable_error_irqs(u32 dmac, u32 mask);
+extern u32 s6dmac_int_sources(u32 dmac, u32 channel);
+extern void s6dmac_release_chan(u32 dmac, int chan);
+
+#endif /* __ASM_XTENSA_S6000_DMAC_H */
-- 
cgit v1.2.3-59-g8ed1b


From 0b3eb21b2f2222c4c1e3e21fc3cd427404d3991a Mon Sep 17 00:00:00 2001
From: Daniel Glöckner <dg@emlix.com>
Date: Tue, 5 May 2009 15:03:22 +0000
Subject: xtensa: support s6000 gpio irqs and alternate function selection
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Implement an irq chip to handle interrupts via gpio.  The GPIO chip
initialization function now takes a bitmask denoting pins that should
be configured for their alternate function.

changes compared to v1:
- fixed bug on edge interrupt configuration
- accommodated to function name change
- moved definition of VARIANT_NR_IRQS to this patch
- renamed __XTENSA_S6000_IRQ_H to _XTENSA_S6000_IRQ_H as requested

Signed-off-by: Daniel Glöckner <dg@emlix.com>
Signed-off-by: Johannes Weiner <jw@emlix.com>
Signed-off-by: Chris Zankel <chris@zankel.net>
---
 arch/xtensa/include/asm/gpio.h                    |   8 +-
 arch/xtensa/platforms/s6105/setup.c               |   2 +-
 arch/xtensa/variants/s6000/gpio.c                 | 163 +++++++++++++++++++++-
 arch/xtensa/variants/s6000/include/variant/gpio.h |   2 +-
 arch/xtensa/variants/s6000/include/variant/irq.h  |   6 +-
 5 files changed, 171 insertions(+), 10 deletions(-)

diff --git a/arch/xtensa/include/asm/gpio.h b/arch/xtensa/include/asm/gpio.h
index 0763b0763960..a8c9fc46c790 100644
--- a/arch/xtensa/include/asm/gpio.h
+++ b/arch/xtensa/include/asm/gpio.h
@@ -38,14 +38,14 @@ static inline int gpio_cansleep(unsigned int gpio)
 	return __gpio_cansleep(gpio);
 }
 
-/*
- * Not implemented, yet.
- */
 static inline int gpio_to_irq(unsigned int gpio)
 {
-	return -ENOSYS;
+	return __gpio_to_irq(gpio);
 }
 
+/*
+ * Not implemented, yet.
+ */
 static inline int irq_to_gpio(unsigned int irq)
 {
 	return -EINVAL;
diff --git a/arch/xtensa/platforms/s6105/setup.c b/arch/xtensa/platforms/s6105/setup.c
index 855ddeadc43d..95fa23c8e632 100644
--- a/arch/xtensa/platforms/s6105/setup.c
+++ b/arch/xtensa/platforms/s6105/setup.c
@@ -49,7 +49,7 @@ void __init platform_setup(char **cmdline)
 
 void __init platform_init(bp_tag_t *first)
 {
-	s6_gpio_init();
+	s6_gpio_init(0);
 	gpio_request(GPIO_LED1_NGREEN, "led1_green");
 	gpio_request(GPIO_LED1_RED, "led1_red");
 	gpio_direction_output(GPIO_LED1_NGREEN, 1);
diff --git a/arch/xtensa/variants/s6000/gpio.c b/arch/xtensa/variants/s6000/gpio.c
index 79317fdcf14c..380a70fff756 100644
--- a/arch/xtensa/variants/s6000/gpio.c
+++ b/arch/xtensa/variants/s6000/gpio.c
@@ -4,15 +4,20 @@
  * Copyright (c) 2009 emlix GmbH
  * Authors:	Oskar Schirmer <os@emlix.com>
  *		Johannes Weiner <jw@emlix.com>
+ *		Daniel Gloeckner <dg@emlix.com>
  */
+#include <linux/bitops.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/io.h>
+#include <linux/irq.h>
 #include <linux/gpio.h>
 
 #include <variant/hardware.h>
 
+#define IRQ_BASE XTENSA_NR_IRQS
+
 #define S6_GPIO_DATA		0x000
 #define S6_GPIO_IS		0x404
 #define S6_GPIO_IBE		0x408
@@ -52,19 +57,175 @@ static void set(struct gpio_chip *chip, unsigned int off, int val)
 	writeb(val ? ~0 : 0, S6_REG_GPIO + S6_GPIO_DATA + S6_GPIO_OFFSET(off));
 }
 
+static int to_irq(struct gpio_chip *chip, unsigned offset)
+{
+	if (offset < 8)
+		return offset + IRQ_BASE;
+	return -EINVAL;
+}
+
 static struct gpio_chip gpiochip = {
 	.owner = THIS_MODULE,
 	.direction_input = direction_input,
 	.get = get,
 	.direction_output = direction_output,
 	.set = set,
+	.to_irq = to_irq,
 	.base = 0,
 	.ngpio = 24,
 	.can_sleep = 0, /* no blocking io needed */
 	.exported = 0, /* no exporting to userspace */
 };
 
-int s6_gpio_init(void)
+int s6_gpio_init(u32 afsel)
 {
+	writeb(afsel, S6_REG_GPIO + S6_GPIO_BANK(0) + S6_GPIO_AFSEL);
+	writeb(afsel >> 8, S6_REG_GPIO + S6_GPIO_BANK(1) + S6_GPIO_AFSEL);
+	writeb(afsel >> 16, S6_REG_GPIO + S6_GPIO_BANK(2) + S6_GPIO_AFSEL);
 	return gpiochip_add(&gpiochip);
 }
+
+static void ack(unsigned int irq)
+{
+	writeb(1 << (irq - IRQ_BASE), S6_REG_GPIO + S6_GPIO_IC);
+}
+
+static void mask(unsigned int irq)
+{
+	u8 r = readb(S6_REG_GPIO + S6_GPIO_IE);
+	r &= ~(1 << (irq - IRQ_BASE));
+	writeb(r, S6_REG_GPIO + S6_GPIO_IE);
+}
+
+static void unmask(unsigned int irq)
+{
+	u8 m = readb(S6_REG_GPIO + S6_GPIO_IE);
+	m |= 1 << (irq - IRQ_BASE);
+	writeb(m, S6_REG_GPIO + S6_GPIO_IE);
+}
+
+static int set_type(unsigned int irq, unsigned int type)
+{
+	const u8 m = 1 << (irq - IRQ_BASE);
+	irq_flow_handler_t handler;
+	struct irq_desc *desc;
+	u8 reg;
+
+	if (type == IRQ_TYPE_PROBE) {
+		if ((readb(S6_REG_GPIO + S6_GPIO_BANK(0) + S6_GPIO_AFSEL) & m)
+		    || (readb(S6_REG_GPIO + S6_GPIO_BANK(0) + S6_GPIO_IE) & m)
+		    || readb(S6_REG_GPIO + S6_GPIO_BANK(0) + S6_GPIO_DIR
+			      + S6_GPIO_MASK(irq - IRQ_BASE)))
+			return 0;
+		type = IRQ_TYPE_EDGE_BOTH;
+	}
+
+	reg = readb(S6_REG_GPIO + S6_GPIO_BANK(0) + S6_GPIO_IS);
+	if (type & (IRQ_TYPE_LEVEL_LOW | IRQ_TYPE_LEVEL_HIGH)) {
+		reg |= m;
+		handler = handle_level_irq;
+	} else {
+		reg &= ~m;
+		handler = handle_edge_irq;
+	}
+	writeb(reg, S6_REG_GPIO + S6_GPIO_BANK(0) + S6_GPIO_IS);
+	desc = irq_to_desc(irq);
+	desc->handle_irq = handler;
+
+	reg = readb(S6_REG_GPIO + S6_GPIO_BANK(0) + S6_GPIO_IEV);
+	if (type & (IRQ_TYPE_LEVEL_HIGH | IRQ_TYPE_EDGE_RISING))
+		reg |= m;
+	else
+		reg &= ~m;
+	writeb(reg, S6_REG_GPIO + S6_GPIO_BANK(0) + S6_GPIO_IEV);
+
+	reg = readb(S6_REG_GPIO + S6_GPIO_BANK(0) + S6_GPIO_IBE);
+	if ((type & IRQ_TYPE_EDGE_BOTH) == IRQ_TYPE_EDGE_BOTH)
+		reg |= m;
+	else
+		reg &= ~m;
+	writeb(reg, S6_REG_GPIO + S6_GPIO_BANK(0) + S6_GPIO_IBE);
+	return 0;
+}
+
+static struct irq_chip gpioirqs = {
+	.name = "GPIO",
+	.ack = ack,
+	.mask = mask,
+	.unmask = unmask,
+	.set_type = set_type,
+};
+
+static u8 demux_masks[4];
+
+static void demux_irqs(unsigned int irq, struct irq_desc *desc)
+{
+	u8 *mask = get_irq_desc_data(desc);
+	u8 pending;
+	int cirq;
+
+	desc->chip->mask(irq);
+	desc->chip->ack(irq);
+	pending = readb(S6_REG_GPIO + S6_GPIO_BANK(0) + S6_GPIO_MIS) & *mask;
+	cirq = IRQ_BASE - 1;
+	while (pending) {
+		int n = ffs(pending);
+		cirq += n;
+		pending >>= n;
+		generic_handle_irq(cirq);
+	}
+	desc->chip->unmask(irq);
+}
+
+extern const signed char *platform_irq_mappings[XTENSA_NR_IRQS];
+
+void __init variant_init_irq(void)
+{
+	int irq, n;
+	writeb(0, S6_REG_GPIO + S6_GPIO_BANK(0) + S6_GPIO_IE);
+	for (irq = n = 0; irq < XTENSA_NR_IRQS; irq++) {
+		const signed char *mapping = platform_irq_mappings[irq];
+		int alone = 1;
+		u8 mask;
+		if (!mapping)
+			continue;
+		for(mask = 0; *mapping != -1; mapping++)
+			switch (*mapping) {
+			case S6_INTC_GPIO(0):
+				mask |= 1 << 0;
+				break;
+			case S6_INTC_GPIO(1):
+				mask |= 1 << 1;
+				break;
+			case S6_INTC_GPIO(2):
+				mask |= 1 << 2;
+				break;
+			case S6_INTC_GPIO(3):
+				mask |= 0x1f << 3;
+				break;
+			default:
+				alone = 0;
+			}
+		if (mask) {
+			int cirq, i;
+			if (!alone) {
+				printk(KERN_ERR "chained irq chips can't share"
+					" parent irq %i\n", irq);
+				continue;
+			}
+			demux_masks[n] = mask;
+			cirq = IRQ_BASE - 1;
+			do {
+				i = ffs(mask);
+				cirq += i;
+				mask >>= i;
+				set_irq_chip(cirq, &gpioirqs);
+				set_irq_type(irq, IRQ_TYPE_LEVEL_LOW);
+			} while (mask);
+			set_irq_data(irq, demux_masks + n);
+			set_irq_chained_handler(irq, demux_irqs);
+			if (++n == ARRAY_SIZE(demux_masks))
+				break;
+		}
+	}
+}
diff --git a/arch/xtensa/variants/s6000/include/variant/gpio.h b/arch/xtensa/variants/s6000/include/variant/gpio.h
index 8327f62167eb..8484ab0df461 100644
--- a/arch/xtensa/variants/s6000/include/variant/gpio.h
+++ b/arch/xtensa/variants/s6000/include/variant/gpio.h
@@ -1,6 +1,6 @@
 #ifndef _XTENSA_VARIANT_S6000_GPIO_H
 #define _XTENSA_VARIANT_S6000_GPIO_H
 
-extern int s6_gpio_init(void);
+extern int s6_gpio_init(u32 afsel);
 
 #endif /* _XTENSA_VARIANT_S6000_GPIO_H */
diff --git a/arch/xtensa/variants/s6000/include/variant/irq.h b/arch/xtensa/variants/s6000/include/variant/irq.h
index fa031cb0acc4..97d6fc48deff 100644
--- a/arch/xtensa/variants/s6000/include/variant/irq.h
+++ b/arch/xtensa/variants/s6000/include/variant/irq.h
@@ -1,9 +1,9 @@
-#ifndef __XTENSA_S6000_IRQ_H
-#define __XTENSA_S6000_IRQ_H
+#ifndef _XTENSA_S6000_IRQ_H
+#define _XTENSA_S6000_IRQ_H
 
 #define NO_IRQ		(-1)
+#define VARIANT_NR_IRQS 8 /* GPIO interrupts */
 
 extern void variant_irq_enable(unsigned int irq);
-extern void variant_irq_disable(unsigned int irq);
 
 #endif /* __XTENSA_S6000_IRQ_H */
-- 
cgit v1.2.3-59-g8ed1b


From 8b0215aa5b01eb3cb54ca57bfa36e94a0d039ed9 Mon Sep 17 00:00:00 2001
From: Oskar Schirmer <os@emlix.com>
Date: Wed, 10 Jun 2009 12:58:48 -0700
Subject: s6gmac: xtensa s6000 on-chip ethernet driver

The s6000 on-chip MAC supports 10/100/1000Mbit and is connected to an
external PHY via MII or RGMII interface.

[jw@emlix.com: don't use device->bus_id directly]
Signed-off-by: Oskar Schirmer <os@emlix.com>
Signed-off-by: Daniel Glockner <dg@emlix.com>
Acked-by: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Johannes Weiner <jw@emlix.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Chris Zankel <chris@zankel.net>
---
 drivers/net/Kconfig  |   11 +
 drivers/net/Makefile |    1 +
 drivers/net/s6gmac.c | 1073 ++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 1085 insertions(+)
 create mode 100644 drivers/net/s6gmac.c

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 892a9e4e275f..1dc721517e4c 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2443,6 +2443,17 @@ config JME
 	  To compile this driver as a module, choose M here. The module
 	  will be called jme.
 
+config S6GMAC
+	tristate "S6105 GMAC ethernet support"
+	depends on XTENSA_VARIANT_S6000
+	select PHYLIB
+	help
+	  This driver supports the on chip ethernet device on the
+	  S6105 xtensa processor.
+
+	  To compile this driver as a module, choose M here. The module
+	  will be called s6gmac.
+
 endif # NETDEV_1000
 
 #
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index d366fb2b40e9..4b58a59f211b 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -245,6 +245,7 @@ obj-$(CONFIG_XTENSA_XT2000_SONIC) += xtsonic.o
 
 obj-$(CONFIG_DNET) += dnet.o
 obj-$(CONFIG_MACB) += macb.o
+obj-$(CONFIG_S6GMAC) += s6gmac.o
 
 obj-$(CONFIG_ARM) += arm/
 obj-$(CONFIG_DEV_APPLETALK) += appletalk/
diff --git a/drivers/net/s6gmac.c b/drivers/net/s6gmac.c
new file mode 100644
index 000000000000..5345e47b35ac
--- /dev/null
+++ b/drivers/net/s6gmac.c
@@ -0,0 +1,1073 @@
+/*
+ * Ethernet driver for S6105 on chip network device
+ * (c)2008 emlix GmbH http://www.emlix.com
+ * Authors:	Oskar Schirmer <os@emlix.com>
+ *		Daniel Gloeckner <dg@emlix.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/types.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/if.h>
+#include <linux/stddef.h>
+#include <linux/mii.h>
+#include <linux/phy.h>
+#include <linux/platform_device.h>
+#include <variant/hardware.h>
+#include <variant/dmac.h>
+
+#define DRV_NAME "s6gmac"
+#define DRV_PRMT DRV_NAME ": "
+
+
+/* register declarations */
+
+#define S6_GMAC_MACCONF1	0x000
+#define S6_GMAC_MACCONF1_TXENA		0
+#define S6_GMAC_MACCONF1_SYNCTX		1
+#define S6_GMAC_MACCONF1_RXENA		2
+#define S6_GMAC_MACCONF1_SYNCRX		3
+#define S6_GMAC_MACCONF1_TXFLOWCTRL	4
+#define S6_GMAC_MACCONF1_RXFLOWCTRL	5
+#define S6_GMAC_MACCONF1_LOOPBACK	8
+#define S6_GMAC_MACCONF1_RESTXFUNC	16
+#define S6_GMAC_MACCONF1_RESRXFUNC	17
+#define S6_GMAC_MACCONF1_RESTXMACCTRL	18
+#define S6_GMAC_MACCONF1_RESRXMACCTRL	19
+#define S6_GMAC_MACCONF1_SIMULRES	30
+#define S6_GMAC_MACCONF1_SOFTRES	31
+#define S6_GMAC_MACCONF2	0x004
+#define S6_GMAC_MACCONF2_FULL		0
+#define S6_GMAC_MACCONF2_CRCENA		1
+#define S6_GMAC_MACCONF2_PADCRCENA	2
+#define S6_GMAC_MACCONF2_LENGTHFCHK	4
+#define S6_GMAC_MACCONF2_HUGEFRAMENA	5
+#define S6_GMAC_MACCONF2_IFMODE		8
+#define S6_GMAC_MACCONF2_IFMODE_NIBBLE		1
+#define S6_GMAC_MACCONF2_IFMODE_BYTE		2
+#define S6_GMAC_MACCONF2_IFMODE_MASK		3
+#define S6_GMAC_MACCONF2_PREAMBLELEN	12
+#define S6_GMAC_MACCONF2_PREAMBLELEN_MASK	0x0F
+#define S6_GMAC_MACIPGIFG	0x008
+#define S6_GMAC_MACIPGIFG_B2BINTERPGAP	0
+#define S6_GMAC_MACIPGIFG_B2BINTERPGAP_MASK	0x7F
+#define S6_GMAC_MACIPGIFG_MINIFGENFORCE	8
+#define S6_GMAC_MACIPGIFG_B2BINTERPGAP2	16
+#define S6_GMAC_MACIPGIFG_B2BINTERPGAP1	24
+#define S6_GMAC_MACHALFDUPLEX	0x00C
+#define S6_GMAC_MACHALFDUPLEX_COLLISWIN	0
+#define S6_GMAC_MACHALFDUPLEX_COLLISWIN_MASK	0x3F
+#define S6_GMAC_MACHALFDUPLEX_RETXMAX	12
+#define S6_GMAC_MACHALFDUPLEX_RETXMAX_MASK	0x0F
+#define S6_GMAC_MACHALFDUPLEX_EXCESSDEF	16
+#define S6_GMAC_MACHALFDUPLEX_NOBACKOFF	17
+#define S6_GMAC_MACHALFDUPLEX_BPNOBCKOF	18
+#define S6_GMAC_MACHALFDUPLEX_ALTBEBENA	19
+#define S6_GMAC_MACHALFDUPLEX_ALTBEBTRN	20
+#define S6_GMAC_MACHALFDUPLEX_ALTBEBTR_MASK	0x0F
+#define S6_GMAC_MACMAXFRAMELEN	0x010
+#define S6_GMAC_MACMIICONF	0x020
+#define S6_GMAC_MACMIICONF_CSEL		0
+#define S6_GMAC_MACMIICONF_CSEL_DIV10		0
+#define S6_GMAC_MACMIICONF_CSEL_DIV12		1
+#define S6_GMAC_MACMIICONF_CSEL_DIV14		2
+#define S6_GMAC_MACMIICONF_CSEL_DIV18		3
+#define S6_GMAC_MACMIICONF_CSEL_DIV24		4
+#define S6_GMAC_MACMIICONF_CSEL_DIV34		5
+#define S6_GMAC_MACMIICONF_CSEL_DIV68		6
+#define S6_GMAC_MACMIICONF_CSEL_DIV168		7
+#define S6_GMAC_MACMIICONF_CSEL_MASK		7
+#define S6_GMAC_MACMIICONF_PREAMBLESUPR	4
+#define S6_GMAC_MACMIICONF_SCANAUTOINCR	5
+#define S6_GMAC_MACMIICMD	0x024
+#define S6_GMAC_MACMIICMD_READ		0
+#define S6_GMAC_MACMIICMD_SCAN		1
+#define S6_GMAC_MACMIIADDR	0x028
+#define S6_GMAC_MACMIIADDR_REG		0
+#define S6_GMAC_MACMIIADDR_REG_MASK		0x1F
+#define S6_GMAC_MACMIIADDR_PHY		8
+#define S6_GMAC_MACMIIADDR_PHY_MASK		0x1F
+#define S6_GMAC_MACMIICTRL	0x02C
+#define S6_GMAC_MACMIISTAT	0x030
+#define S6_GMAC_MACMIIINDI	0x034
+#define S6_GMAC_MACMIIINDI_BUSY		0
+#define S6_GMAC_MACMIIINDI_SCAN		1
+#define S6_GMAC_MACMIIINDI_INVAL	2
+#define S6_GMAC_MACINTERFSTAT	0x03C
+#define S6_GMAC_MACINTERFSTAT_LINKFAIL	3
+#define S6_GMAC_MACINTERFSTAT_EXCESSDEF	9
+#define S6_GMAC_MACSTATADDR1	0x040
+#define S6_GMAC_MACSTATADDR2	0x044
+
+#define S6_GMAC_FIFOCONF0	0x048
+#define S6_GMAC_FIFOCONF0_HSTRSTWT	0
+#define S6_GMAC_FIFOCONF0_HSTRSTSR	1
+#define S6_GMAC_FIFOCONF0_HSTRSTFR	2
+#define S6_GMAC_FIFOCONF0_HSTRSTST	3
+#define S6_GMAC_FIFOCONF0_HSTRSTFT	4
+#define S6_GMAC_FIFOCONF0_WTMENREQ	8
+#define S6_GMAC_FIFOCONF0_SRFENREQ	9
+#define S6_GMAC_FIFOCONF0_FRFENREQ	10
+#define S6_GMAC_FIFOCONF0_STFENREQ	11
+#define S6_GMAC_FIFOCONF0_FTFENREQ	12
+#define S6_GMAC_FIFOCONF0_WTMENRPLY	16
+#define S6_GMAC_FIFOCONF0_SRFENRPLY	17
+#define S6_GMAC_FIFOCONF0_FRFENRPLY	18
+#define S6_GMAC_FIFOCONF0_STFENRPLY	19
+#define S6_GMAC_FIFOCONF0_FTFENRPLY	20
+#define S6_GMAC_FIFOCONF1	0x04C
+#define S6_GMAC_FIFOCONF2	0x050
+#define S6_GMAC_FIFOCONF2_CFGLWM	0
+#define S6_GMAC_FIFOCONF2_CFGHWM	16
+#define S6_GMAC_FIFOCONF3	0x054
+#define S6_GMAC_FIFOCONF3_CFGFTTH	0
+#define S6_GMAC_FIFOCONF3_CFGHWMFT	16
+#define S6_GMAC_FIFOCONF4	0x058
+#define S6_GMAC_FIFOCONF_RSV_PREVDROP	0
+#define S6_GMAC_FIFOCONF_RSV_RUNT	1
+#define S6_GMAC_FIFOCONF_RSV_FALSECAR	2
+#define S6_GMAC_FIFOCONF_RSV_CODEERR	3
+#define S6_GMAC_FIFOCONF_RSV_CRCERR	4
+#define S6_GMAC_FIFOCONF_RSV_LENGTHERR	5
+#define S6_GMAC_FIFOCONF_RSV_LENRANGE	6
+#define S6_GMAC_FIFOCONF_RSV_OK		7
+#define S6_GMAC_FIFOCONF_RSV_MULTICAST	8
+#define S6_GMAC_FIFOCONF_RSV_BROADCAST	9
+#define S6_GMAC_FIFOCONF_RSV_DRIBBLE	10
+#define S6_GMAC_FIFOCONF_RSV_CTRLFRAME	11
+#define S6_GMAC_FIFOCONF_RSV_PAUSECTRL	12
+#define S6_GMAC_FIFOCONF_RSV_UNOPCODE	13
+#define S6_GMAC_FIFOCONF_RSV_VLANTAG	14
+#define S6_GMAC_FIFOCONF_RSV_LONGEVENT	15
+#define S6_GMAC_FIFOCONF_RSV_TRUNCATED	16
+#define S6_GMAC_FIFOCONF_RSV_MASK		0x3FFFF
+#define S6_GMAC_FIFOCONF5	0x05C
+#define S6_GMAC_FIFOCONF5_DROPLT64	18
+#define S6_GMAC_FIFOCONF5_CFGBYTM	19
+#define S6_GMAC_FIFOCONF5_RXDROPSIZE	20
+#define S6_GMAC_FIFOCONF5_RXDROPSIZE_MASK	0xF
+
+#define S6_GMAC_STAT_REGS	0x080
+#define S6_GMAC_STAT_SIZE_MIN		12
+#define S6_GMAC_STATTR64	0x080
+#define S6_GMAC_STATTR64_SIZE		18
+#define S6_GMAC_STATTR127	0x084
+#define S6_GMAC_STATTR127_SIZE		18
+#define S6_GMAC_STATTR255	0x088
+#define S6_GMAC_STATTR255_SIZE		18
+#define S6_GMAC_STATTR511	0x08C
+#define S6_GMAC_STATTR511_SIZE		18
+#define S6_GMAC_STATTR1K	0x090
+#define S6_GMAC_STATTR1K_SIZE		18
+#define S6_GMAC_STATTRMAX	0x094
+#define S6_GMAC_STATTRMAX_SIZE		18
+#define S6_GMAC_STATTRMGV	0x098
+#define S6_GMAC_STATTRMGV_SIZE		18
+#define S6_GMAC_STATRBYT	0x09C
+#define S6_GMAC_STATRBYT_SIZE		24
+#define S6_GMAC_STATRPKT	0x0A0
+#define S6_GMAC_STATRPKT_SIZE		18
+#define S6_GMAC_STATRFCS	0x0A4
+#define S6_GMAC_STATRFCS_SIZE		12
+#define S6_GMAC_STATRMCA	0x0A8
+#define S6_GMAC_STATRMCA_SIZE		18
+#define S6_GMAC_STATRBCA	0x0AC
+#define S6_GMAC_STATRBCA_SIZE		22
+#define S6_GMAC_STATRXCF	0x0B0
+#define S6_GMAC_STATRXCF_SIZE		18
+#define S6_GMAC_STATRXPF	0x0B4
+#define S6_GMAC_STATRXPF_SIZE		12
+#define S6_GMAC_STATRXUO	0x0B8
+#define S6_GMAC_STATRXUO_SIZE		12
+#define S6_GMAC_STATRALN	0x0BC
+#define S6_GMAC_STATRALN_SIZE		12
+#define S6_GMAC_STATRFLR	0x0C0
+#define S6_GMAC_STATRFLR_SIZE		16
+#define S6_GMAC_STATRCDE	0x0C4
+#define S6_GMAC_STATRCDE_SIZE		12
+#define S6_GMAC_STATRCSE	0x0C8
+#define S6_GMAC_STATRCSE_SIZE		12
+#define S6_GMAC_STATRUND	0x0CC
+#define S6_GMAC_STATRUND_SIZE		12
+#define S6_GMAC_STATROVR	0x0D0
+#define S6_GMAC_STATROVR_SIZE		12
+#define S6_GMAC_STATRFRG	0x0D4
+#define S6_GMAC_STATRFRG_SIZE		12
+#define S6_GMAC_STATRJBR	0x0D8
+#define S6_GMAC_STATRJBR_SIZE		12
+#define S6_GMAC_STATRDRP	0x0DC
+#define S6_GMAC_STATRDRP_SIZE		12
+#define S6_GMAC_STATTBYT	0x0E0
+#define S6_GMAC_STATTBYT_SIZE		24
+#define S6_GMAC_STATTPKT	0x0E4
+#define S6_GMAC_STATTPKT_SIZE		18
+#define S6_GMAC_STATTMCA	0x0E8
+#define S6_GMAC_STATTMCA_SIZE		18
+#define S6_GMAC_STATTBCA	0x0EC
+#define S6_GMAC_STATTBCA_SIZE		18
+#define S6_GMAC_STATTXPF	0x0F0
+#define S6_GMAC_STATTXPF_SIZE		12
+#define S6_GMAC_STATTDFR	0x0F4
+#define S6_GMAC_STATTDFR_SIZE		12
+#define S6_GMAC_STATTEDF	0x0F8
+#define S6_GMAC_STATTEDF_SIZE		12
+#define S6_GMAC_STATTSCL	0x0FC
+#define S6_GMAC_STATTSCL_SIZE		12
+#define S6_GMAC_STATTMCL	0x100
+#define S6_GMAC_STATTMCL_SIZE		12
+#define S6_GMAC_STATTLCL	0x104
+#define S6_GMAC_STATTLCL_SIZE		12
+#define S6_GMAC_STATTXCL	0x108
+#define S6_GMAC_STATTXCL_SIZE		12
+#define S6_GMAC_STATTNCL	0x10C
+#define S6_GMAC_STATTNCL_SIZE		13
+#define S6_GMAC_STATTPFH	0x110
+#define S6_GMAC_STATTPFH_SIZE		12
+#define S6_GMAC_STATTDRP	0x114
+#define S6_GMAC_STATTDRP_SIZE		12
+#define S6_GMAC_STATTJBR	0x118
+#define S6_GMAC_STATTJBR_SIZE		12
+#define S6_GMAC_STATTFCS	0x11C
+#define S6_GMAC_STATTFCS_SIZE		12
+#define S6_GMAC_STATTXCF	0x120
+#define S6_GMAC_STATTXCF_SIZE		12
+#define S6_GMAC_STATTOVR	0x124
+#define S6_GMAC_STATTOVR_SIZE		12
+#define S6_GMAC_STATTUND	0x128
+#define S6_GMAC_STATTUND_SIZE		12
+#define S6_GMAC_STATTFRG	0x12C
+#define S6_GMAC_STATTFRG_SIZE		12
+#define S6_GMAC_STATCARRY(n)	(0x130 + 4*(n))
+#define S6_GMAC_STATCARRYMSK(n)	(0x138 + 4*(n))
+#define S6_GMAC_STATCARRY1_RDRP		0
+#define S6_GMAC_STATCARRY1_RJBR		1
+#define S6_GMAC_STATCARRY1_RFRG		2
+#define S6_GMAC_STATCARRY1_ROVR		3
+#define S6_GMAC_STATCARRY1_RUND		4
+#define S6_GMAC_STATCARRY1_RCSE		5
+#define S6_GMAC_STATCARRY1_RCDE		6
+#define S6_GMAC_STATCARRY1_RFLR		7
+#define S6_GMAC_STATCARRY1_RALN		8
+#define S6_GMAC_STATCARRY1_RXUO		9
+#define S6_GMAC_STATCARRY1_RXPF		10
+#define S6_GMAC_STATCARRY1_RXCF		11
+#define S6_GMAC_STATCARRY1_RBCA		12
+#define S6_GMAC_STATCARRY1_RMCA		13
+#define S6_GMAC_STATCARRY1_RFCS		14
+#define S6_GMAC_STATCARRY1_RPKT		15
+#define S6_GMAC_STATCARRY1_RBYT		16
+#define S6_GMAC_STATCARRY1_TRMGV	25
+#define S6_GMAC_STATCARRY1_TRMAX	26
+#define S6_GMAC_STATCARRY1_TR1K		27
+#define S6_GMAC_STATCARRY1_TR511	28
+#define S6_GMAC_STATCARRY1_TR255	29
+#define S6_GMAC_STATCARRY1_TR127	30
+#define S6_GMAC_STATCARRY1_TR64		31
+#define S6_GMAC_STATCARRY2_TDRP		0
+#define S6_GMAC_STATCARRY2_TPFH		1
+#define S6_GMAC_STATCARRY2_TNCL		2
+#define S6_GMAC_STATCARRY2_TXCL		3
+#define S6_GMAC_STATCARRY2_TLCL		4
+#define S6_GMAC_STATCARRY2_TMCL		5
+#define S6_GMAC_STATCARRY2_TSCL		6
+#define S6_GMAC_STATCARRY2_TEDF		7
+#define S6_GMAC_STATCARRY2_TDFR		8
+#define S6_GMAC_STATCARRY2_TXPF		9
+#define S6_GMAC_STATCARRY2_TBCA		10
+#define S6_GMAC_STATCARRY2_TMCA		11
+#define S6_GMAC_STATCARRY2_TPKT		12
+#define S6_GMAC_STATCARRY2_TBYT		13
+#define S6_GMAC_STATCARRY2_TFRG		14
+#define S6_GMAC_STATCARRY2_TUND		15
+#define S6_GMAC_STATCARRY2_TOVR		16
+#define S6_GMAC_STATCARRY2_TXCF		17
+#define S6_GMAC_STATCARRY2_TFCS		18
+#define S6_GMAC_STATCARRY2_TJBR		19
+
+#define S6_GMAC_HOST_PBLKCTRL	0x140
+#define S6_GMAC_HOST_PBLKCTRL_TXENA	0
+#define S6_GMAC_HOST_PBLKCTRL_RXENA	1
+#define S6_GMAC_HOST_PBLKCTRL_TXSRES	2
+#define S6_GMAC_HOST_PBLKCTRL_RXSRES	3
+#define S6_GMAC_HOST_PBLKCTRL_TXBSIZ	8
+#define S6_GMAC_HOST_PBLKCTRL_RXBSIZ	12
+#define S6_GMAC_HOST_PBLKCTRL_SIZ_16		4
+#define S6_GMAC_HOST_PBLKCTRL_SIZ_32		5
+#define S6_GMAC_HOST_PBLKCTRL_SIZ_64		6
+#define S6_GMAC_HOST_PBLKCTRL_SIZ_128		7
+#define S6_GMAC_HOST_PBLKCTRL_SIZ_MASK		0xF
+#define S6_GMAC_HOST_PBLKCTRL_STATENA	16
+#define S6_GMAC_HOST_PBLKCTRL_STATAUTOZ	17
+#define S6_GMAC_HOST_PBLKCTRL_STATCLEAR	18
+#define S6_GMAC_HOST_PBLKCTRL_RGMII	19
+#define S6_GMAC_HOST_INTMASK	0x144
+#define S6_GMAC_HOST_INTSTAT	0x148
+#define S6_GMAC_HOST_INT_TXBURSTOVER	3
+#define S6_GMAC_HOST_INT_TXPREWOVER	4
+#define S6_GMAC_HOST_INT_RXBURSTUNDER	5
+#define S6_GMAC_HOST_INT_RXPOSTRFULL	6
+#define S6_GMAC_HOST_INT_RXPOSTRUNDER	7
+#define S6_GMAC_HOST_RXFIFOHWM	0x14C
+#define S6_GMAC_HOST_CTRLFRAMXP	0x150
+#define S6_GMAC_HOST_DSTADDRLO(n) (0x160 + 8*(n))
+#define S6_GMAC_HOST_DSTADDRHI(n) (0x164 + 8*(n))
+#define S6_GMAC_HOST_DSTMASKLO(n) (0x180 + 8*(n))
+#define S6_GMAC_HOST_DSTMASKHI(n) (0x184 + 8*(n))
+
+#define S6_GMAC_BURST_PREWR	0x1B0
+#define S6_GMAC_BURST_PREWR_LEN		0
+#define S6_GMAC_BURST_PREWR_LEN_MASK		((1 << 20) - 1)
+#define S6_GMAC_BURST_PREWR_CFE		20
+#define S6_GMAC_BURST_PREWR_PPE		21
+#define S6_GMAC_BURST_PREWR_FCS		22
+#define S6_GMAC_BURST_PREWR_PAD		23
+#define S6_GMAC_BURST_POSTRD	0x1D0
+#define S6_GMAC_BURST_POSTRD_LEN	0
+#define S6_GMAC_BURST_POSTRD_LEN_MASK		((1 << 20) - 1)
+#define S6_GMAC_BURST_POSTRD_DROP	20
+
+
+/* data handling */
+
+#define S6_NUM_TX_SKB	8	/* must be larger than TX fifo size */
+#define S6_NUM_RX_SKB	16
+#define S6_MAX_FRLEN	1536
+
+struct s6gmac {
+	u32 reg;
+	u32 tx_dma;
+	u32 rx_dma;
+	u32 io;
+	u8 tx_chan;
+	u8 rx_chan;
+	spinlock_t lock;
+	u8 tx_skb_i, tx_skb_o;
+	u8 rx_skb_i, rx_skb_o;
+	struct sk_buff *tx_skb[S6_NUM_TX_SKB];
+	struct sk_buff *rx_skb[S6_NUM_RX_SKB];
+	unsigned long carry[sizeof(struct net_device_stats) / sizeof(long)];
+	unsigned long stats[sizeof(struct net_device_stats) / sizeof(long)];
+	struct phy_device *phydev;
+	struct {
+		struct mii_bus *bus;
+		int irq[PHY_MAX_ADDR];
+	} mii;
+	struct {
+		unsigned int mbit;
+		u8 giga;
+		u8 isup;
+		u8 full;
+	} link;
+};
+
+static void s6gmac_rx_fillfifo(struct s6gmac *pd)
+{
+	struct sk_buff *skb;
+	while ((((u8)(pd->rx_skb_i - pd->rx_skb_o)) < S6_NUM_RX_SKB)
+			&& (!s6dmac_fifo_full(pd->rx_dma, pd->rx_chan))
+			&& (skb = dev_alloc_skb(S6_MAX_FRLEN + 2))) {
+		pd->rx_skb[(pd->rx_skb_i++) % S6_NUM_RX_SKB] = skb;
+		s6dmac_put_fifo_cache(pd->rx_dma, pd->rx_chan,
+			pd->io, (u32)skb->data, S6_MAX_FRLEN);
+	}
+}
+
+static void s6gmac_rx_interrupt(struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	u32 pfx;
+	struct sk_buff *skb;
+	while (((u8)(pd->rx_skb_i - pd->rx_skb_o)) >
+			s6dmac_pending_count(pd->rx_dma, pd->rx_chan)) {
+		skb = pd->rx_skb[(pd->rx_skb_o++) % S6_NUM_RX_SKB];
+		pfx = readl(pd->reg + S6_GMAC_BURST_POSTRD);
+		if (pfx & (1 << S6_GMAC_BURST_POSTRD_DROP)) {
+			dev_kfree_skb_irq(skb);
+		} else {
+			skb_put(skb, (pfx >> S6_GMAC_BURST_POSTRD_LEN)
+				& S6_GMAC_BURST_POSTRD_LEN_MASK);
+			skb->dev = dev;
+			skb->protocol = eth_type_trans(skb, dev);
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+			netif_rx(skb);
+		}
+	}
+}
+
+static void s6gmac_tx_interrupt(struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	while (((u8)(pd->tx_skb_i - pd->tx_skb_o)) >
+			s6dmac_pending_count(pd->tx_dma, pd->tx_chan)) {
+		dev_kfree_skb_irq(pd->tx_skb[(pd->tx_skb_o++) % S6_NUM_TX_SKB]);
+	}
+	if (!s6dmac_fifo_full(pd->tx_dma, pd->tx_chan))
+		netif_wake_queue(dev);
+}
+
+struct s6gmac_statinf {
+	unsigned reg_size : 4; /* 0: unused */
+	unsigned reg_off : 6;
+	unsigned net_index : 6;
+};
+
+#define S6_STATS_B (8 * sizeof(u32))
+#define S6_STATS_C(b, r, f) [b] = { \
+	BUILD_BUG_ON_ZERO(r##_SIZE < S6_GMAC_STAT_SIZE_MIN) + \
+	BUILD_BUG_ON_ZERO((r##_SIZE - (S6_GMAC_STAT_SIZE_MIN - 1)) \
+			>= (1<<4)) + \
+	r##_SIZE - (S6_GMAC_STAT_SIZE_MIN - 1), \
+	BUILD_BUG_ON_ZERO(((unsigned)((r - S6_GMAC_STAT_REGS) / sizeof(u32))) \
+			>= ((1<<6)-1)) + \
+	(r - S6_GMAC_STAT_REGS) / sizeof(u32), \
+	BUILD_BUG_ON_ZERO((offsetof(struct net_device_stats, f)) \
+			% sizeof(unsigned long)) + \
+	BUILD_BUG_ON_ZERO((((unsigned)(offsetof(struct net_device_stats, f)) \
+			/ sizeof(unsigned long)) >= (1<<6))) + \
+	BUILD_BUG_ON_ZERO((sizeof(((struct net_device_stats *)0)->f) \
+			!= sizeof(unsigned long))) + \
+	(offsetof(struct net_device_stats, f)) / sizeof(unsigned long)},
+
+static const struct s6gmac_statinf statinf[2][S6_STATS_B] = { {
+	S6_STATS_C(S6_GMAC_STATCARRY1_RBYT, S6_GMAC_STATRBYT, rx_bytes)
+	S6_STATS_C(S6_GMAC_STATCARRY1_RPKT, S6_GMAC_STATRPKT, rx_packets)
+	S6_STATS_C(S6_GMAC_STATCARRY1_RFCS, S6_GMAC_STATRFCS, rx_crc_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY1_RMCA, S6_GMAC_STATRMCA, multicast)
+	S6_STATS_C(S6_GMAC_STATCARRY1_RALN, S6_GMAC_STATRALN, rx_frame_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY1_RFLR, S6_GMAC_STATRFLR, rx_length_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY1_RCDE, S6_GMAC_STATRCDE, rx_missed_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY1_RUND, S6_GMAC_STATRUND, rx_length_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY1_ROVR, S6_GMAC_STATROVR, rx_length_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY1_RFRG, S6_GMAC_STATRFRG, rx_crc_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY1_RJBR, S6_GMAC_STATRJBR, rx_crc_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY1_RDRP, S6_GMAC_STATRDRP, rx_dropped)
+}, {
+	S6_STATS_C(S6_GMAC_STATCARRY2_TBYT, S6_GMAC_STATTBYT, tx_bytes)
+	S6_STATS_C(S6_GMAC_STATCARRY2_TPKT, S6_GMAC_STATTPKT, tx_packets)
+	S6_STATS_C(S6_GMAC_STATCARRY2_TEDF, S6_GMAC_STATTEDF, tx_aborted_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY2_TXCL, S6_GMAC_STATTXCL, tx_aborted_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY2_TNCL, S6_GMAC_STATTNCL, collisions)
+	S6_STATS_C(S6_GMAC_STATCARRY2_TDRP, S6_GMAC_STATTDRP, tx_dropped)
+	S6_STATS_C(S6_GMAC_STATCARRY2_TJBR, S6_GMAC_STATTJBR, tx_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY2_TFCS, S6_GMAC_STATTFCS, tx_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY2_TOVR, S6_GMAC_STATTOVR, tx_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY2_TUND, S6_GMAC_STATTUND, tx_errors)
+	S6_STATS_C(S6_GMAC_STATCARRY2_TFRG, S6_GMAC_STATTFRG, tx_errors)
+} };
+
+static void s6gmac_stats_collect(struct s6gmac *pd,
+		const struct s6gmac_statinf *inf)
+{
+	int b;
+	for (b = 0; b < S6_STATS_B; b++) {
+		if (inf[b].reg_size) {
+			pd->stats[inf[b].net_index] +=
+				readl(pd->reg + S6_GMAC_STAT_REGS
+					+ sizeof(u32) * inf[b].reg_off);
+		}
+	}
+}
+
+static void s6gmac_stats_carry(struct s6gmac *pd,
+		const struct s6gmac_statinf *inf, u32 mask)
+{
+	int b;
+	while (mask) {
+		b = fls(mask) - 1;
+		mask &= ~(1 << b);
+		pd->carry[inf[b].net_index] += (1 << inf[b].reg_size);
+	}
+}
+
+static inline u32 s6gmac_stats_pending(struct s6gmac *pd, int carry)
+{
+	int r = readl(pd->reg + S6_GMAC_STATCARRY(carry)) &
+		~readl(pd->reg + S6_GMAC_STATCARRYMSK(carry));
+	return r;
+}
+
+static inline void s6gmac_stats_interrupt(struct s6gmac *pd, int carry)
+{
+	u32 mask;
+	mask = s6gmac_stats_pending(pd, carry);
+	if (mask) {
+		writel(mask, pd->reg + S6_GMAC_STATCARRY(carry));
+		s6gmac_stats_carry(pd, &statinf[carry][0], mask);
+	}
+}
+
+static irqreturn_t s6gmac_interrupt(int irq, void *dev_id)
+{
+	struct net_device *dev = (struct net_device *)dev_id;
+	struct s6gmac *pd = netdev_priv(dev);
+	if (!dev)
+		return IRQ_NONE;
+	spin_lock(&pd->lock);
+	if (s6dmac_termcnt_irq(pd->rx_dma, pd->rx_chan))
+		s6gmac_rx_interrupt(dev);
+	s6gmac_rx_fillfifo(pd);
+	if (s6dmac_termcnt_irq(pd->tx_dma, pd->tx_chan))
+		s6gmac_tx_interrupt(dev);
+	s6gmac_stats_interrupt(pd, 0);
+	s6gmac_stats_interrupt(pd, 1);
+	spin_unlock(&pd->lock);
+	return IRQ_HANDLED;
+}
+
+static inline void s6gmac_set_dstaddr(struct s6gmac *pd, int n,
+	u32 addrlo, u32 addrhi, u32 masklo, u32 maskhi)
+{
+	writel(addrlo, pd->reg + S6_GMAC_HOST_DSTADDRLO(n));
+	writel(addrhi, pd->reg + S6_GMAC_HOST_DSTADDRHI(n));
+	writel(masklo, pd->reg + S6_GMAC_HOST_DSTMASKLO(n));
+	writel(maskhi, pd->reg + S6_GMAC_HOST_DSTMASKHI(n));
+}
+
+static inline void s6gmac_stop_device(struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	writel(0, pd->reg + S6_GMAC_MACCONF1);
+}
+
+static inline void s6gmac_init_device(struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	int is_rgmii = !!(pd->phydev->supported
+		& (SUPPORTED_1000baseT_Full | SUPPORTED_1000baseT_Half));
+#if 0
+	writel(1 << S6_GMAC_MACCONF1_SYNCTX |
+		1 << S6_GMAC_MACCONF1_SYNCRX |
+		1 << S6_GMAC_MACCONF1_TXFLOWCTRL |
+		1 << S6_GMAC_MACCONF1_RXFLOWCTRL |
+		1 << S6_GMAC_MACCONF1_RESTXFUNC |
+		1 << S6_GMAC_MACCONF1_RESRXFUNC |
+		1 << S6_GMAC_MACCONF1_RESTXMACCTRL |
+		1 << S6_GMAC_MACCONF1_RESRXMACCTRL,
+		pd->reg + S6_GMAC_MACCONF1);
+#endif
+	writel(1 << S6_GMAC_MACCONF1_SOFTRES, pd->reg + S6_GMAC_MACCONF1);
+	udelay(1000);
+	writel(1 << S6_GMAC_MACCONF1_TXENA | 1 << S6_GMAC_MACCONF1_RXENA,
+		pd->reg + S6_GMAC_MACCONF1);
+	writel(1 << S6_GMAC_HOST_PBLKCTRL_TXSRES |
+		1 << S6_GMAC_HOST_PBLKCTRL_RXSRES,
+		pd->reg + S6_GMAC_HOST_PBLKCTRL);
+	writel(S6_GMAC_HOST_PBLKCTRL_SIZ_128 << S6_GMAC_HOST_PBLKCTRL_TXBSIZ |
+		S6_GMAC_HOST_PBLKCTRL_SIZ_128 << S6_GMAC_HOST_PBLKCTRL_RXBSIZ |
+		1 << S6_GMAC_HOST_PBLKCTRL_STATENA |
+		1 << S6_GMAC_HOST_PBLKCTRL_STATCLEAR |
+		is_rgmii << S6_GMAC_HOST_PBLKCTRL_RGMII,
+		pd->reg + S6_GMAC_HOST_PBLKCTRL);
+	writel(1 << S6_GMAC_MACCONF1_TXENA |
+		1 << S6_GMAC_MACCONF1_RXENA |
+		(dev->flags & IFF_LOOPBACK ? 1 : 0)
+			<< S6_GMAC_MACCONF1_LOOPBACK,
+		pd->reg + S6_GMAC_MACCONF1);
+	writel(dev->mtu && (dev->mtu < (S6_MAX_FRLEN - ETH_HLEN-ETH_FCS_LEN)) ?
+			dev->mtu+ETH_HLEN+ETH_FCS_LEN : S6_MAX_FRLEN,
+		pd->reg + S6_GMAC_MACMAXFRAMELEN);
+	writel((pd->link.full ? 1 : 0) << S6_GMAC_MACCONF2_FULL |
+		1 << S6_GMAC_MACCONF2_PADCRCENA |
+		1 << S6_GMAC_MACCONF2_LENGTHFCHK |
+		(pd->link.giga ?
+			S6_GMAC_MACCONF2_IFMODE_BYTE :
+			S6_GMAC_MACCONF2_IFMODE_NIBBLE)
+			<< S6_GMAC_MACCONF2_IFMODE |
+		7 << S6_GMAC_MACCONF2_PREAMBLELEN,
+		pd->reg + S6_GMAC_MACCONF2);
+	writel(0, pd->reg + S6_GMAC_MACSTATADDR1);
+	writel(0, pd->reg + S6_GMAC_MACSTATADDR2);
+	writel(1 << S6_GMAC_FIFOCONF0_WTMENREQ |
+		1 << S6_GMAC_FIFOCONF0_SRFENREQ |
+		1 << S6_GMAC_FIFOCONF0_FRFENREQ |
+		1 << S6_GMAC_FIFOCONF0_STFENREQ |
+		1 << S6_GMAC_FIFOCONF0_FTFENREQ,
+		pd->reg + S6_GMAC_FIFOCONF0);
+	writel(128 << S6_GMAC_FIFOCONF3_CFGFTTH |
+		128 << S6_GMAC_FIFOCONF3_CFGHWMFT,
+		pd->reg + S6_GMAC_FIFOCONF3);
+	writel((S6_GMAC_FIFOCONF_RSV_MASK & ~(
+			1 << S6_GMAC_FIFOCONF_RSV_RUNT |
+			1 << S6_GMAC_FIFOCONF_RSV_CRCERR |
+			1 << S6_GMAC_FIFOCONF_RSV_OK |
+			1 << S6_GMAC_FIFOCONF_RSV_DRIBBLE |
+			1 << S6_GMAC_FIFOCONF_RSV_CTRLFRAME |
+			1 << S6_GMAC_FIFOCONF_RSV_PAUSECTRL |
+			1 << S6_GMAC_FIFOCONF_RSV_UNOPCODE |
+			1 << S6_GMAC_FIFOCONF_RSV_TRUNCATED)) |
+		1 << S6_GMAC_FIFOCONF5_DROPLT64 |
+		pd->link.giga << S6_GMAC_FIFOCONF5_CFGBYTM |
+		1 << S6_GMAC_FIFOCONF5_RXDROPSIZE,
+		pd->reg + S6_GMAC_FIFOCONF5);
+	writel(1 << S6_GMAC_FIFOCONF_RSV_RUNT |
+		1 << S6_GMAC_FIFOCONF_RSV_CRCERR |
+		1 << S6_GMAC_FIFOCONF_RSV_DRIBBLE |
+		1 << S6_GMAC_FIFOCONF_RSV_CTRLFRAME |
+		1 << S6_GMAC_FIFOCONF_RSV_PAUSECTRL |
+		1 << S6_GMAC_FIFOCONF_RSV_UNOPCODE |
+		1 << S6_GMAC_FIFOCONF_RSV_TRUNCATED,
+		pd->reg + S6_GMAC_FIFOCONF4);
+	s6gmac_set_dstaddr(pd, 0,
+		0xFFFFFFFF, 0x0000FFFF, 0xFFFFFFFF, 0x0000FFFF);
+	s6gmac_set_dstaddr(pd, 1,
+		dev->dev_addr[5] |
+		dev->dev_addr[4] << 8 |
+		dev->dev_addr[3] << 16 |
+		dev->dev_addr[2] << 24,
+		dev->dev_addr[1] |
+		dev->dev_addr[0] << 8,
+		0xFFFFFFFF, 0x0000FFFF);
+	s6gmac_set_dstaddr(pd, 2,
+		0x00000000, 0x00000100, 0x00000000, 0x00000100);
+	s6gmac_set_dstaddr(pd, 3,
+		0x00000000, 0x00000000, 0x00000000, 0x00000000);
+	writel(1 << S6_GMAC_HOST_PBLKCTRL_TXENA |
+		1 << S6_GMAC_HOST_PBLKCTRL_RXENA |
+		S6_GMAC_HOST_PBLKCTRL_SIZ_128 << S6_GMAC_HOST_PBLKCTRL_TXBSIZ |
+		S6_GMAC_HOST_PBLKCTRL_SIZ_128 << S6_GMAC_HOST_PBLKCTRL_RXBSIZ |
+		1 << S6_GMAC_HOST_PBLKCTRL_STATENA |
+		1 << S6_GMAC_HOST_PBLKCTRL_STATCLEAR |
+		is_rgmii << S6_GMAC_HOST_PBLKCTRL_RGMII,
+		pd->reg + S6_GMAC_HOST_PBLKCTRL);
+}
+
+static void s6mii_enable(struct s6gmac *pd)
+{
+	writel(readl(pd->reg + S6_GMAC_MACCONF1) &
+		~(1 << S6_GMAC_MACCONF1_SOFTRES),
+		pd->reg + S6_GMAC_MACCONF1);
+	writel((readl(pd->reg + S6_GMAC_MACMIICONF)
+		& ~(S6_GMAC_MACMIICONF_CSEL_MASK << S6_GMAC_MACMIICONF_CSEL))
+		| (S6_GMAC_MACMIICONF_CSEL_DIV168 << S6_GMAC_MACMIICONF_CSEL),
+		pd->reg + S6_GMAC_MACMIICONF);
+}
+
+static int s6mii_busy(struct s6gmac *pd, int tmo)
+{
+	while (readl(pd->reg + S6_GMAC_MACMIIINDI)) {
+		if (--tmo == 0)
+			return -ETIME;
+		udelay(64);
+	}
+	return 0;
+}
+
+static int s6mii_read(struct mii_bus *bus, int phy_addr, int regnum)
+{
+	struct s6gmac *pd = bus->priv;
+	s6mii_enable(pd);
+	if (s6mii_busy(pd, 256))
+		return -ETIME;
+	writel(phy_addr << S6_GMAC_MACMIIADDR_PHY |
+		regnum << S6_GMAC_MACMIIADDR_REG,
+		pd->reg + S6_GMAC_MACMIIADDR);
+	writel(1 << S6_GMAC_MACMIICMD_READ, pd->reg + S6_GMAC_MACMIICMD);
+	writel(0, pd->reg + S6_GMAC_MACMIICMD);
+	if (s6mii_busy(pd, 256))
+		return -ETIME;
+	return (u16)readl(pd->reg + S6_GMAC_MACMIISTAT);
+}
+
+static int s6mii_write(struct mii_bus *bus, int phy_addr, int regnum, u16 value)
+{
+	struct s6gmac *pd = bus->priv;
+	s6mii_enable(pd);
+	if (s6mii_busy(pd, 256))
+		return -ETIME;
+	writel(phy_addr << S6_GMAC_MACMIIADDR_PHY |
+		regnum << S6_GMAC_MACMIIADDR_REG,
+		pd->reg + S6_GMAC_MACMIIADDR);
+	writel(value, pd->reg + S6_GMAC_MACMIICTRL);
+	if (s6mii_busy(pd, 256))
+		return -ETIME;
+	return 0;
+}
+
+static int s6mii_reset(struct mii_bus *bus)
+{
+	struct s6gmac *pd = bus->priv;
+	s6mii_enable(pd);
+	if (s6mii_busy(pd, PHY_INIT_TIMEOUT))
+		return -ETIME;
+	return 0;
+}
+
+static void s6gmac_set_rgmii_txclock(struct s6gmac *pd)
+{
+	u32 pllsel = readl(S6_REG_GREG1 + S6_GREG1_PLLSEL);
+	pllsel &= ~(S6_GREG1_PLLSEL_GMAC_MASK << S6_GREG1_PLLSEL_GMAC);
+	switch (pd->link.mbit) {
+	case 10:
+		pllsel |= S6_GREG1_PLLSEL_GMAC_2500KHZ << S6_GREG1_PLLSEL_GMAC;
+		break;
+	case 100:
+		pllsel |= S6_GREG1_PLLSEL_GMAC_25MHZ << S6_GREG1_PLLSEL_GMAC;
+		break;
+	case 1000:
+		pllsel |= S6_GREG1_PLLSEL_GMAC_125MHZ << S6_GREG1_PLLSEL_GMAC;
+		break;
+	default:
+		return;
+	}
+	writel(pllsel, S6_REG_GREG1 + S6_GREG1_PLLSEL);
+}
+
+static inline void s6gmac_linkisup(struct net_device *dev, int isup)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	struct phy_device *phydev = pd->phydev;
+
+	pd->link.full = phydev->duplex;
+	pd->link.giga = (phydev->speed == 1000);
+	if (pd->link.mbit != phydev->speed) {
+		pd->link.mbit = phydev->speed;
+		s6gmac_set_rgmii_txclock(pd);
+	}
+	pd->link.isup = isup;
+	if (isup)
+		netif_carrier_on(dev);
+	phy_print_status(phydev);
+}
+
+static void s6gmac_adjust_link(struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	struct phy_device *phydev = pd->phydev;
+	if (pd->link.isup &&
+			(!phydev->link ||
+			(pd->link.mbit != phydev->speed) ||
+			(pd->link.full != phydev->duplex))) {
+		pd->link.isup = 0;
+		netif_tx_disable(dev);
+		if (!phydev->link) {
+			netif_carrier_off(dev);
+			phy_print_status(phydev);
+		}
+	}
+	if (!pd->link.isup && phydev->link) {
+		if (pd->link.full != phydev->duplex) {
+			u32 maccfg = readl(pd->reg + S6_GMAC_MACCONF2);
+			if (phydev->duplex)
+				maccfg |= 1 << S6_GMAC_MACCONF2_FULL;
+			else
+				maccfg &= ~(1 << S6_GMAC_MACCONF2_FULL);
+			writel(maccfg, pd->reg + S6_GMAC_MACCONF2);
+		}
+
+		if (pd->link.giga != (phydev->speed == 1000)) {
+			u32 fifocfg = readl(pd->reg + S6_GMAC_FIFOCONF5);
+			u32 maccfg = readl(pd->reg + S6_GMAC_MACCONF2);
+			maccfg &= ~(S6_GMAC_MACCONF2_IFMODE_MASK
+				     << S6_GMAC_MACCONF2_IFMODE);
+			if (phydev->speed == 1000) {
+				fifocfg |= 1 << S6_GMAC_FIFOCONF5_CFGBYTM;
+				maccfg |= S6_GMAC_MACCONF2_IFMODE_BYTE
+					   << S6_GMAC_MACCONF2_IFMODE;
+			} else {
+				fifocfg &= ~(1 << S6_GMAC_FIFOCONF5_CFGBYTM);
+				maccfg |= S6_GMAC_MACCONF2_IFMODE_NIBBLE
+					   << S6_GMAC_MACCONF2_IFMODE;
+			}
+			writel(fifocfg, pd->reg + S6_GMAC_FIFOCONF5);
+			writel(maccfg, pd->reg + S6_GMAC_MACCONF2);
+		}
+
+		if (!s6dmac_fifo_full(pd->tx_dma, pd->tx_chan))
+			netif_wake_queue(dev);
+		s6gmac_linkisup(dev, 1);
+	}
+}
+
+static inline int s6gmac_phy_start(struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	int i = 0;
+	struct phy_device *p = NULL;
+	while ((!(p = pd->mii.bus->phy_map[i])) && (i < PHY_MAX_ADDR))
+		i++;
+	p = phy_connect(dev, dev_name(&p->dev), &s6gmac_adjust_link, 0,
+			PHY_INTERFACE_MODE_RGMII);
+	if (IS_ERR(p)) {
+		printk(KERN_ERR "%s: Could not attach to PHY\n", dev->name);
+		return PTR_ERR(p);
+	}
+	p->supported &= PHY_GBIT_FEATURES;
+	p->advertising = p->supported;
+	pd->phydev = p;
+	return 0;
+}
+
+static inline void s6gmac_init_stats(struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	u32 mask;
+	mask =	1 << S6_GMAC_STATCARRY1_RDRP |
+		1 << S6_GMAC_STATCARRY1_RJBR |
+		1 << S6_GMAC_STATCARRY1_RFRG |
+		1 << S6_GMAC_STATCARRY1_ROVR |
+		1 << S6_GMAC_STATCARRY1_RUND |
+		1 << S6_GMAC_STATCARRY1_RCDE |
+		1 << S6_GMAC_STATCARRY1_RFLR |
+		1 << S6_GMAC_STATCARRY1_RALN |
+		1 << S6_GMAC_STATCARRY1_RMCA |
+		1 << S6_GMAC_STATCARRY1_RFCS |
+		1 << S6_GMAC_STATCARRY1_RPKT |
+		1 << S6_GMAC_STATCARRY1_RBYT;
+	writel(mask, pd->reg + S6_GMAC_STATCARRY(0));
+	writel(~mask, pd->reg + S6_GMAC_STATCARRYMSK(0));
+	mask =	1 << S6_GMAC_STATCARRY2_TDRP |
+		1 << S6_GMAC_STATCARRY2_TNCL |
+		1 << S6_GMAC_STATCARRY2_TXCL |
+		1 << S6_GMAC_STATCARRY2_TEDF |
+		1 << S6_GMAC_STATCARRY2_TPKT |
+		1 << S6_GMAC_STATCARRY2_TBYT |
+		1 << S6_GMAC_STATCARRY2_TFRG |
+		1 << S6_GMAC_STATCARRY2_TUND |
+		1 << S6_GMAC_STATCARRY2_TOVR |
+		1 << S6_GMAC_STATCARRY2_TFCS |
+		1 << S6_GMAC_STATCARRY2_TJBR;
+	writel(mask, pd->reg + S6_GMAC_STATCARRY(1));
+	writel(~mask, pd->reg + S6_GMAC_STATCARRYMSK(1));
+}
+
+static inline void s6gmac_init_dmac(struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	s6dmac_disable_chan(pd->tx_dma, pd->tx_chan);
+	s6dmac_disable_chan(pd->rx_dma, pd->rx_chan);
+	s6dmac_disable_error_irqs(pd->tx_dma, 1 << S6_HIFDMA_GMACTX);
+	s6dmac_disable_error_irqs(pd->rx_dma, 1 << S6_HIFDMA_GMACRX);
+}
+
+static int s6gmac_tx(struct sk_buff *skb, struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	unsigned long flags;
+	spin_lock_irqsave(&pd->lock, flags);
+	dev->trans_start = jiffies;
+	writel(skb->len << S6_GMAC_BURST_PREWR_LEN |
+		0 << S6_GMAC_BURST_PREWR_CFE |
+		1 << S6_GMAC_BURST_PREWR_PPE |
+		1 << S6_GMAC_BURST_PREWR_FCS |
+		((skb->len < ETH_ZLEN) ? 1 : 0) << S6_GMAC_BURST_PREWR_PAD,
+		pd->reg + S6_GMAC_BURST_PREWR);
+	s6dmac_put_fifo_cache(pd->tx_dma, pd->tx_chan,
+		(u32)skb->data, pd->io, skb->len);
+	if (s6dmac_fifo_full(pd->tx_dma, pd->tx_chan))
+		netif_stop_queue(dev);
+	if (((u8)(pd->tx_skb_i - pd->tx_skb_o)) >= S6_NUM_TX_SKB) {
+		printk(KERN_ERR "GMAC BUG: skb tx ring overflow [%x, %x]\n",
+			pd->tx_skb_o, pd->tx_skb_i);
+		BUG();
+	}
+	pd->tx_skb[(pd->tx_skb_i++) % S6_NUM_TX_SKB] = skb;
+	spin_unlock_irqrestore(&pd->lock, flags);
+	return 0;
+}
+
+static void s6gmac_tx_timeout(struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	unsigned long flags;
+	spin_lock_irqsave(&pd->lock, flags);
+	s6gmac_tx_interrupt(dev);
+	spin_unlock_irqrestore(&pd->lock, flags);
+}
+
+static int s6gmac_open(struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	unsigned long flags;
+	phy_read_status(pd->phydev);
+	spin_lock_irqsave(&pd->lock, flags);
+	pd->link.mbit = 0;
+	s6gmac_linkisup(dev, pd->phydev->link);
+	s6gmac_init_device(dev);
+	s6gmac_init_stats(dev);
+	s6gmac_init_dmac(dev);
+	s6gmac_rx_fillfifo(pd);
+	s6dmac_enable_chan(pd->rx_dma, pd->rx_chan,
+		2, 1, 0, 1, 0, 0, 0, 7, -1, 2, 0, 1);
+	s6dmac_enable_chan(pd->tx_dma, pd->tx_chan,
+		2, 0, 1, 0, 0, 0, 0, 7, -1, 2, 0, 1);
+	writel(0 << S6_GMAC_HOST_INT_TXBURSTOVER |
+		0 << S6_GMAC_HOST_INT_TXPREWOVER |
+		0 << S6_GMAC_HOST_INT_RXBURSTUNDER |
+		0 << S6_GMAC_HOST_INT_RXPOSTRFULL |
+		0 << S6_GMAC_HOST_INT_RXPOSTRUNDER,
+		pd->reg + S6_GMAC_HOST_INTMASK);
+	spin_unlock_irqrestore(&pd->lock, flags);
+	phy_start(pd->phydev);
+	netif_start_queue(dev);
+	return 0;
+}
+
+static int s6gmac_stop(struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	unsigned long flags;
+	netif_stop_queue(dev);
+	phy_stop(pd->phydev);
+	spin_lock_irqsave(&pd->lock, flags);
+	s6gmac_init_dmac(dev);
+	s6gmac_stop_device(dev);
+	while (pd->tx_skb_i != pd->tx_skb_o)
+		dev_kfree_skb(pd->tx_skb[(pd->tx_skb_o++) % S6_NUM_TX_SKB]);
+	while (pd->rx_skb_i != pd->rx_skb_o)
+		dev_kfree_skb(pd->rx_skb[(pd->rx_skb_o++) % S6_NUM_RX_SKB]);
+	spin_unlock_irqrestore(&pd->lock, flags);
+	return 0;
+}
+
+static struct net_device_stats *s6gmac_stats(struct net_device *dev)
+{
+	struct s6gmac *pd = netdev_priv(dev);
+	struct net_device_stats *st = (struct net_device_stats *)&pd->stats;
+	int i;
+	do {
+		unsigned long flags;
+		spin_lock_irqsave(&pd->lock, flags);
+		for (i = 0; i < sizeof(pd->stats) / sizeof(unsigned long); i++)
+			pd->stats[i] =
+				pd->carry[i] << (S6_GMAC_STAT_SIZE_MIN - 1);
+		s6gmac_stats_collect(pd, &statinf[0][0]);
+		s6gmac_stats_collect(pd, &statinf[1][0]);
+		i = s6gmac_stats_pending(pd, 0) |
+			s6gmac_stats_pending(pd, 1);
+		spin_unlock_irqrestore(&pd->lock, flags);
+	} while (i);
+	st->rx_errors = st->rx_crc_errors +
+			st->rx_frame_errors +
+			st->rx_length_errors +
+			st->rx_missed_errors;
+	st->tx_errors += st->tx_aborted_errors;
+	return st;
+}
+
+static int __devinit s6gmac_probe(struct platform_device *pdev)
+{
+	struct net_device *dev;
+	struct s6gmac *pd;
+	int res;
+	unsigned long i;
+	struct mii_bus *mb;
+	dev = alloc_etherdev(sizeof(*pd));
+	if (!dev) {
+		printk(KERN_ERR DRV_PRMT "etherdev alloc failed, aborting.\n");
+		return -ENOMEM;
+	}
+	dev->open = s6gmac_open;
+	dev->stop = s6gmac_stop;
+	dev->hard_start_xmit = s6gmac_tx;
+	dev->tx_timeout = s6gmac_tx_timeout;
+	dev->watchdog_timeo = HZ;
+	dev->get_stats = s6gmac_stats;
+	dev->irq = platform_get_irq(pdev, 0);
+	pd = netdev_priv(dev);
+	memset(pd, 0, sizeof(*pd));
+	spin_lock_init(&pd->lock);
+	pd->reg = platform_get_resource(pdev, IORESOURCE_MEM, 0)->start;
+	i = platform_get_resource(pdev, IORESOURCE_DMA, 0)->start;
+	pd->tx_dma = DMA_MASK_DMAC(i);
+	pd->tx_chan = DMA_INDEX_CHNL(i);
+	i = platform_get_resource(pdev, IORESOURCE_DMA, 1)->start;
+	pd->rx_dma = DMA_MASK_DMAC(i);
+	pd->rx_chan = DMA_INDEX_CHNL(i);
+	pd->io = platform_get_resource(pdev, IORESOURCE_IO, 0)->start;
+	res = request_irq(dev->irq, &s6gmac_interrupt, 0, dev->name, dev);
+	if (res) {
+		printk(KERN_ERR DRV_PRMT "irq request failed: %d\n", dev->irq);
+		goto errirq;
+	}
+	res = register_netdev(dev);
+	if (res) {
+		printk(KERN_ERR DRV_PRMT "error registering device %s\n",
+			dev->name);
+		goto errdev;
+	}
+	mb = mdiobus_alloc();
+	if (!mb) {
+		printk(KERN_ERR DRV_PRMT "error allocating mii bus\n");
+		goto errmii;
+	}
+	mb->name = "s6gmac_mii";
+	mb->read = s6mii_read;
+	mb->write = s6mii_write;
+	mb->reset = s6mii_reset;
+	mb->priv = pd;
+	snprintf(mb->id, MII_BUS_ID_SIZE, "0");
+	mb->phy_mask = ~(1 << 0);
+	mb->irq = &pd->mii.irq[0];
+	for (i = 0; i < PHY_MAX_ADDR; i++) {
+		int n = platform_get_irq(pdev, i + 1);
+		if (n < 0)
+			n = PHY_POLL;
+		pd->mii.irq[i] = n;
+	}
+	mdiobus_register(mb);
+	pd->mii.bus = mb;
+	res = s6gmac_phy_start(dev);
+	if (res)
+		return res;
+	platform_set_drvdata(pdev, dev);
+	return 0;
+errmii:
+	unregister_netdev(dev);
+errdev:
+	free_irq(dev->irq, dev);
+errirq:
+	free_netdev(dev);
+	return res;
+}
+
+static int __devexit s6gmac_remove(struct platform_device *pdev)
+{
+	struct net_device *dev = platform_get_drvdata(pdev);
+	if (dev) {
+		struct s6gmac *pd = netdev_priv(dev);
+		mdiobus_unregister(pd->mii.bus);
+		unregister_netdev(dev);
+		free_irq(dev->irq, dev);
+		free_netdev(dev);
+		platform_set_drvdata(pdev, NULL);
+	}
+	return 0;
+}
+
+static struct platform_driver s6gmac_driver = {
+	.probe = s6gmac_probe,
+	.remove = __devexit_p(s6gmac_remove),
+	.driver = {
+		.name = "s6gmac",
+		.owner = THIS_MODULE,
+	},
+};
+
+static int __init s6gmac_init(void)
+{
+	printk(KERN_INFO DRV_PRMT "S6 GMAC ethernet driver\n");
+	return platform_driver_register(&s6gmac_driver);
+}
+
+
+static void __exit s6gmac_exit(void)
+{
+	platform_driver_unregister(&s6gmac_driver);
+}
+
+module_init(s6gmac_init);
+module_exit(s6gmac_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("S6105 on chip Ethernet driver");
+MODULE_AUTHOR("Oskar Schirmer <os@emlix.com>");
-- 
cgit v1.2.3-59-g8ed1b


From 059cafe6df5d3b078c4b1be42623c62ab32de6fd Mon Sep 17 00:00:00 2001
From: Oskar Schirmer <os@emlix.com>
Date: Wed, 10 Jun 2009 12:58:48 -0700
Subject: xtensa: s6105 specific configuration for s6gmac

Platform-specific configuration for the s6gmac driver, including the
PHY interrupt line.

Signed-off-by: Daniel Glockner <dg@emlix.com>
Signed-off-by: Oskar Schirmer <os@emlix.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Johannes Weiner <jw@emlix.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 arch/xtensa/platforms/s6105/device.c | 91 ++++++++++++++++++++++++++++++++++++
 arch/xtensa/platforms/s6105/setup.c  |  9 ++++
 2 files changed, 100 insertions(+)

diff --git a/arch/xtensa/platforms/s6105/device.c b/arch/xtensa/platforms/s6105/device.c
index 78b08be5a92d..963634a3463d 100644
--- a/arch/xtensa/platforms/s6105/device.c
+++ b/arch/xtensa/platforms/s6105/device.c
@@ -5,14 +5,27 @@
  */
 
 #include <linux/kernel.h>
+#include <linux/gpio.h>
 #include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/phy.h>
 #include <linux/platform_device.h>
 #include <linux/serial.h>
 #include <linux/serial_8250.h>
 
 #include <variant/hardware.h>
+#include <variant/dmac.h>
 
+#include <platform/gpio.h>
+
+#define GPIO3_INTNUM		3
 #define UART_INTNUM		4
+#define GMAC_INTNUM		5
+
+static const signed char gpio3_irq_mappings[] = {
+	S6_INTC_GPIO(3),
+	-1
+};
 
 static const signed char uart_irq_mappings[] = {
 	S6_INTC_UART(0),
@@ -20,8 +33,18 @@ static const signed char uart_irq_mappings[] = {
 	-1,
 };
 
+static const signed char gmac_irq_mappings[] = {
+	S6_INTC_GMAC_STAT,
+	S6_INTC_GMAC_ERR,
+	S6_INTC_DMA_HOSTTERMCNT(0),
+	S6_INTC_DMA_HOSTTERMCNT(1),
+	-1
+};
+
 const signed char *platform_irq_mappings[NR_IRQS] = {
+	[GPIO3_INTNUM] = gpio3_irq_mappings,
 	[UART_INTNUM] = uart_irq_mappings,
+	[GMAC_INTNUM] = gmac_irq_mappings,
 };
 
 static struct plat_serial8250_port serial_platform_data[] = {
@@ -46,6 +69,66 @@ static struct plat_serial8250_port serial_platform_data[] = {
 	{ },
 };
 
+static struct resource s6_gmac_resource[] = {
+	{
+		.name   = "mem",
+		.start  = (resource_size_t)S6_REG_GMAC,
+		.end    = (resource_size_t)S6_REG_GMAC + 0x10000 - 1,
+		.flags  = IORESOURCE_MEM,
+	},
+	{
+		.name   = "dma",
+		.start  = (resource_size_t)
+			DMA_CHNL(S6_REG_HIFDMA, S6_HIFDMA_GMACTX),
+		.end    = (resource_size_t)
+			DMA_CHNL(S6_REG_HIFDMA, S6_HIFDMA_GMACTX) + 0x100 - 1,
+		.flags  = IORESOURCE_DMA,
+	},
+	{
+		.name   = "dma",
+		.start  = (resource_size_t)
+			DMA_CHNL(S6_REG_HIFDMA, S6_HIFDMA_GMACRX),
+		.end    = (resource_size_t)
+			DMA_CHNL(S6_REG_HIFDMA, S6_HIFDMA_GMACRX) + 0x100 - 1,
+		.flags  = IORESOURCE_DMA,
+	},
+	{
+		.name   = "io",
+		.start  = (resource_size_t)S6_MEM_GMAC,
+		.end    = (resource_size_t)S6_MEM_GMAC + 0x2000000 - 1,
+		.flags  = IORESOURCE_IO,
+	},
+	{
+		.name   = "irq",
+		.start  = (resource_size_t)GMAC_INTNUM,
+		.flags  = IORESOURCE_IRQ,
+	},
+	{
+		.name   = "irq",
+		.start  = (resource_size_t)PHY_POLL,
+		.flags  = IORESOURCE_IRQ,
+	},
+};
+
+static int __init prepare_phy_irq(int pin)
+{
+	int irq;
+	if (gpio_request(pin, "s6gmac_phy") < 0)
+		goto fail;
+	if (gpio_direction_input(pin) < 0)
+		goto free;
+	irq = gpio_to_irq(pin);
+	if (irq < 0)
+		goto free;
+	if (set_irq_type(irq, IRQ_TYPE_LEVEL_LOW) < 0)
+		goto free;
+	return irq;
+free:
+	gpio_free(pin);
+fail:
+	return PHY_POLL;
+}
+
 static struct platform_device platform_devices[] = {
 	{
 		.name = "serial8250",
@@ -54,12 +137,20 @@ static struct platform_device platform_devices[] = {
 			.platform_data = serial_platform_data,
 		},
 	},
+	{
+		.name = "s6gmac",
+		.id = 0,
+		.resource = s6_gmac_resource,
+		.num_resources = ARRAY_SIZE(s6_gmac_resource),
+	},
 };
 
 static int __init device_init(void)
 {
 	int i;
 
+	s6_gmac_resource[5].start = prepare_phy_irq(GPIO_PHY_IRQ);
+
 	for (i = 0; i < ARRAY_SIZE(platform_devices); i++)
 		platform_device_register(&platform_devices[i]);
 	return 0;
diff --git a/arch/xtensa/platforms/s6105/setup.c b/arch/xtensa/platforms/s6105/setup.c
index 95fa23c8e632..86ce730f7913 100644
--- a/arch/xtensa/platforms/s6105/setup.c
+++ b/arch/xtensa/platforms/s6105/setup.c
@@ -35,12 +35,21 @@ void __init platform_setup(char **cmdline)
 {
 	unsigned long reg;
 
+	reg = readl(S6_REG_GREG1 + S6_GREG1_PLLSEL);
+	reg &= ~(S6_GREG1_PLLSEL_GMAC_MASK << S6_GREG1_PLLSEL_GMAC |
+		S6_GREG1_PLLSEL_GMII_MASK << S6_GREG1_PLLSEL_GMII);
+	reg |= S6_GREG1_PLLSEL_GMAC_125MHZ << S6_GREG1_PLLSEL_GMAC |
+		S6_GREG1_PLLSEL_GMII_125MHZ << S6_GREG1_PLLSEL_GMII;
+	writel(reg, S6_REG_GREG1 + S6_GREG1_PLLSEL);
+
 	reg = readl(S6_REG_GREG1 + S6_GREG1_CLKGATE);
 	reg &= ~(1 << S6_GREG1_BLOCK_SB);
+	reg &= ~(1 << S6_GREG1_BLOCK_GMAC);
 	writel(reg, S6_REG_GREG1 + S6_GREG1_CLKGATE);
 
 	reg = readl(S6_REG_GREG1 + S6_GREG1_BLOCKENA);
 	reg |= 1 << S6_GREG1_BLOCK_SB;
+	reg |= 1 << S6_GREG1_BLOCK_GMAC;
 	writel(reg, S6_REG_GREG1 + S6_GREG1_BLOCKENA);
 
 	printk(KERN_NOTICE "S6105 on Stretch S6000 - "
-- 
cgit v1.2.3-59-g8ed1b


From 607c2add0f3b2cc12567f658e9dfdb07df2b106e Mon Sep 17 00:00:00 2001
From: Daniel Glockner <dg@emlix.com>
Date: Wed, 10 Jun 2009 12:58:49 -0700
Subject: xtensa: enable s6gmac in s6105_defconfig

Signed-off-by: Daniel Glockner <dg@emlix.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Johannes Weiner <jw@emlix.com>
Cc: Oskar Schirmer <os@emlix.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 arch/xtensa/configs/s6105_defconfig | 49 ++++++++++++++++++++++++++++++++++++-
 1 file changed, 48 insertions(+), 1 deletion(-)

diff --git a/arch/xtensa/configs/s6105_defconfig b/arch/xtensa/configs/s6105_defconfig
index 768bee006037..a1eca01f8725 100644
--- a/arch/xtensa/configs/s6105_defconfig
+++ b/arch/xtensa/configs/s6105_defconfig
@@ -263,7 +263,54 @@ CONFIG_HAVE_IDE=y
 # CONFIG_SCSI_NETLINK is not set
 # CONFIG_ATA is not set
 # CONFIG_MD is not set
-# CONFIG_NETDEVICES is not set
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+# CONFIG_VETH is not set
+CONFIG_PHYLIB=y
+
+#
+# MII PHY device drivers
+#
+# CONFIG_MARVELL_PHY is not set
+# CONFIG_DAVICOM_PHY is not set
+# CONFIG_QSEMI_PHY is not set
+# CONFIG_LXT_PHY is not set
+# CONFIG_CICADA_PHY is not set
+# CONFIG_VITESSE_PHY is not set
+CONFIG_SMSC_PHY=y
+# CONFIG_BROADCOM_PHY is not set
+# CONFIG_ICPLUS_PHY is not set
+# CONFIG_REALTEK_PHY is not set
+# CONFIG_NATIONAL_PHY is not set
+# CONFIG_STE10XP is not set
+# CONFIG_LSI_ET1011C_PHY is not set
+# CONFIG_FIXED_PHY is not set
+# CONFIG_MDIO_BITBANG is not set
+# CONFIG_NET_ETHERNET is not set
+CONFIG_NETDEV_1000=y
+CONFIG_S6GMAC=y
+# CONFIG_NETDEV_10000 is not set
+
+#
+# Wireless LAN
+#
+# CONFIG_WLAN_PRE80211 is not set
+# CONFIG_WLAN_80211 is not set
+# CONFIG_IWLWIFI_LEDS is not set
+
+#
+# Enable WiMAX (Networking options) to see the WiMAX drivers
+#
+# CONFIG_WAN is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
 # CONFIG_ISDN is not set
 # CONFIG_PHONE is not set
 
-- 
cgit v1.2.3-59-g8ed1b


From 759c67e87bce54bd08e5fee1fdca975cebe5d3c2 Mon Sep 17 00:00:00 2001
From: Daniel Glockner <dg@emlix.com>
Date: Wed, 10 Jun 2009 12:58:50 -0700
Subject: xtensa: add m41t62 rtc to s6105 platform

Signed-off-by: Daniel Glockner <dg@emlix.com>
Cc: David Brownell <david-b@pacbell.net>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 arch/xtensa/platforms/s6105/device.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/xtensa/platforms/s6105/device.c b/arch/xtensa/platforms/s6105/device.c
index 963634a3463d..65333ffefb07 100644
--- a/arch/xtensa/platforms/s6105/device.c
+++ b/arch/xtensa/platforms/s6105/device.c
@@ -143,6 +143,9 @@ static struct platform_device platform_devices[] = {
 		.resource = s6_gmac_resource,
 		.num_resources = ARRAY_SIZE(s6_gmac_resource),
 	},
+	{
+		I2C_BOARD_INFO("m41t62", S6I2C_ADDR_M41T62),
+	},
 };
 
 static int __init device_init(void)
-- 
cgit v1.2.3-59-g8ed1b


From aafd1255d08fb26cab87d1b28ff35a15bdb2ed68 Mon Sep 17 00:00:00 2001
From: Daniel Glockner <dg@emlix.com>
Date: Wed, 10 Jun 2009 12:58:51 -0700
Subject: xtensa: enable m41t80 driver in s6105_defconfig

Signed-off-by: Daniel Glockner <dg@emlix.com>
Cc: David Brownell <david-b@pacbell.net>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 arch/xtensa/configs/s6105_defconfig | 56 +++++++++++++++++++++++++++++++++++--
 1 file changed, 53 insertions(+), 3 deletions(-)

diff --git a/arch/xtensa/configs/s6105_defconfig b/arch/xtensa/configs/s6105_defconfig
index a1eca01f8725..bb84fbc9921f 100644
--- a/arch/xtensa/configs/s6105_defconfig
+++ b/arch/xtensa/configs/s6105_defconfig
@@ -351,8 +351,6 @@ CONFIG_UNIX98_PTYS=y
 # CONFIG_LEGACY_PTYS is not set
 # CONFIG_IPMI_HANDLER is not set
 # CONFIG_HW_RANDOM is not set
-# CONFIG_RTC is not set
-# CONFIG_GEN_RTC is not set
 # CONFIG_R3964 is not set
 # CONFIG_RAW_DRIVER is not set
 # CONFIG_TCG_TPM is not set
@@ -434,7 +432,59 @@ CONFIG_SSB_POSSIBLE=y
 # CONFIG_MEMSTICK is not set
 # CONFIG_NEW_LEDS is not set
 # CONFIG_ACCESSIBILITY is not set
-# CONFIG_RTC_CLASS is not set
+CONFIG_RTC_LIB=y
+CONFIG_RTC_CLASS=y
+CONFIG_RTC_HCTOSYS=y
+CONFIG_RTC_HCTOSYS_DEVICE="rtc0"
+# CONFIG_RTC_DEBUG is not set
+
+#
+# RTC interfaces
+#
+# CONFIG_RTC_INTF_SYSFS is not set
+# CONFIG_RTC_INTF_PROC is not set
+# CONFIG_RTC_INTF_DEV is not set
+# CONFIG_RTC_DRV_TEST is not set
+
+#
+# I2C RTC drivers
+#
+# CONFIG_RTC_DRV_DS1307 is not set
+# CONFIG_RTC_DRV_DS1374 is not set
+# CONFIG_RTC_DRV_DS1672 is not set
+# CONFIG_RTC_DRV_MAX6900 is not set
+# CONFIG_RTC_DRV_RS5C372 is not set
+# CONFIG_RTC_DRV_ISL1208 is not set
+# CONFIG_RTC_DRV_X1205 is not set
+# CONFIG_RTC_DRV_PCF8563 is not set
+# CONFIG_RTC_DRV_PCF8583 is not set
+CONFIG_RTC_DRV_M41T80=y
+# CONFIG_RTC_DRV_M41T80_WDT is not set
+# CONFIG_RTC_DRV_S35390A is not set
+# CONFIG_RTC_DRV_FM3130 is not set
+# CONFIG_RTC_DRV_RX8581 is not set
+
+#
+# SPI RTC drivers
+#
+
+#
+# Platform RTC drivers
+#
+# CONFIG_RTC_DRV_DS1286 is not set
+# CONFIG_RTC_DRV_DS1511 is not set
+# CONFIG_RTC_DRV_DS1553 is not set
+# CONFIG_RTC_DRV_DS1742 is not set
+# CONFIG_RTC_DRV_STK17TA8 is not set
+# CONFIG_RTC_DRV_M48T86 is not set
+# CONFIG_RTC_DRV_M48T35 is not set
+# CONFIG_RTC_DRV_M48T59 is not set
+# CONFIG_RTC_DRV_BQ4802 is not set
+# CONFIG_RTC_DRV_V3020 is not set
+
+#
+# on-CPU RTC drivers
+#
 # CONFIG_DMADEVICES is not set
 # CONFIG_UIO is not set
 # CONFIG_STAGING is not set
-- 
cgit v1.2.3-59-g8ed1b


From d7d1104fa40f66dbe50840f05b34268144f8a17a Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:03 +0200
Subject: [S390] time: convert from bootmem to slab

The slab allocator is earlier available so convert the
bootmem allocations to slab/gfp allocations.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/time.c  | 16 ++++------------
 arch/s390/kernel/vtime.c |  5 +----
 2 files changed, 5 insertions(+), 16 deletions(-)

diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index 215330a2c128..d4c8e9c47c81 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -36,7 +36,6 @@
 #include <linux/notifier.h>
 #include <linux/clocksource.h>
 #include <linux/clockchips.h>
-#include <linux/bootmem.h>
 #include <asm/uaccess.h>
 #include <asm/delay.h>
 #include <asm/s390_ext.h>
@@ -62,9 +61,6 @@
 
 u64 sched_clock_base_cc = -1;	/* Force to data section. */
 
-static ext_int_info_t ext_int_info_cc;
-static ext_int_info_t ext_int_etr_cc;
-
 static DEFINE_PER_CPU(struct clock_event_device, comparators);
 
 /*
@@ -255,15 +251,11 @@ void __init time_init(void)
 	stp_reset();
 
 	/* request the clock comparator external interrupt */
-	if (register_early_external_interrupt(0x1004,
-					      clock_comparator_interrupt,
-					      &ext_int_info_cc) != 0)
+	if (register_external_interrupt(0x1004, clock_comparator_interrupt))
                 panic("Couldn't request external interrupt 0x1004");
 
 	/* request the timing alert external interrupt */
-	if (register_early_external_interrupt(0x1406,
-					      timing_alert_interrupt,
-					      &ext_int_etr_cc) != 0)
+	if (register_external_interrupt(0x1406, timing_alert_interrupt))
 		panic("Couldn't request external interrupt 0x1406");
 
 	if (clocksource_register(&clocksource_tod) != 0)
@@ -1445,14 +1437,14 @@ static void __init stp_reset(void)
 {
 	int rc;
 
-	stp_page = alloc_bootmem_pages(PAGE_SIZE);
+	stp_page = (void *) get_zeroed_page(GFP_ATOMIC);
 	rc = chsc_sstpc(stp_page, STP_OP_CTRL, 0x0000);
 	if (rc == 0)
 		set_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags);
 	else if (stp_online) {
 		pr_warning("The real or virtual hardware system does "
 			   "not provide an STP interface\n");
-		free_bootmem((unsigned long) stp_page, PAGE_SIZE);
+		free_page((unsigned long) stp_page);
 		stp_page = NULL;
 		stp_online = 0;
 	}
diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c
index c8eb7255332b..ade17e771f05 100644
--- a/arch/s390/kernel/vtime.c
+++ b/arch/s390/kernel/vtime.c
@@ -25,8 +25,6 @@
 #include <asm/irq_regs.h>
 #include <asm/cputime.h>
 
-static ext_int_info_t ext_int_info_timer;
-
 static DEFINE_PER_CPU(struct vtimer_queue, virt_cpu_timer);
 
 DEFINE_PER_CPU(struct s390_idle_data, s390_idle) = {
@@ -557,8 +555,7 @@ void init_cpu_vtimer(void)
 void __init vtime_init(void)
 {
 	/* request the cpu timer external interrupt */
-	if (register_early_external_interrupt(0x1005, do_cpu_timer_interrupt,
-					      &ext_int_info_timer) != 0)
+	if (register_external_interrupt(0x1005, do_cpu_timer_interrupt))
 		panic("Couldn't request external interrupt 0x1005");
 
 	/* Enable cpu timer interrupts on the boot cpu. */
-- 
cgit v1.2.3-59-g8ed1b


From 6d56eee2c016b0b131e444d02a66b0fef7df3ef0 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:04 +0200
Subject: [S390] 3215 console: convert from bootmem to slab

The slab allocator is earlier available so convert the
bootmem allocations to slab/gfp allocations.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/char/con3215.c | 18 +++++++-----------
 1 file changed, 7 insertions(+), 11 deletions(-)

diff --git a/drivers/s390/char/con3215.c b/drivers/s390/char/con3215.c
index 04dc734805c6..51e6379c5b93 100644
--- a/drivers/s390/char/con3215.c
+++ b/drivers/s390/char/con3215.c
@@ -20,10 +20,7 @@
 #include <linux/interrupt.h>
 #include <linux/err.h>
 #include <linux/reboot.h>
-
 #include <linux/slab.h>
-#include <linux/bootmem.h>
-
 #include <asm/ccwdev.h>
 #include <asm/cio.h>
 #include <asm/io.h>
@@ -883,7 +880,7 @@ static int __init con3215_init(void)
 	raw3215_freelist = NULL;
 	spin_lock_init(&raw3215_freelist_lock);
 	for (i = 0; i < NR_3215_REQ; i++) {
-		req = (struct raw3215_req *) alloc_bootmem_low(sizeof(struct raw3215_req));
+		req = kzalloc(sizeof(struct raw3215_req), GFP_KERNEL | GFP_DMA);
 		req->next = raw3215_freelist;
 		raw3215_freelist = req;
 	}
@@ -893,10 +890,9 @@ static int __init con3215_init(void)
 		return -ENODEV;
 
 	raw3215[0] = raw = (struct raw3215_info *)
-		alloc_bootmem_low(sizeof(struct raw3215_info));
-	memset(raw, 0, sizeof(struct raw3215_info));
-	raw->buffer = (char *) alloc_bootmem_low(RAW3215_BUFFER_SIZE);
-	raw->inbuf = (char *) alloc_bootmem_low(RAW3215_INBUF_SIZE);
+		kzalloc(sizeof(struct raw3215_info), GFP_KERNEL | GFP_DMA);
+	raw->buffer = kzalloc(RAW3215_BUFFER_SIZE, GFP_KERNEL | GFP_DMA);
+	raw->inbuf = kzalloc(RAW3215_INBUF_SIZE, GFP_KERNEL | GFP_DMA);
 	raw->cdev = cdev;
 	dev_set_drvdata(&cdev->dev, raw);
 	cdev->handler = raw3215_irq;
@@ -906,9 +902,9 @@ static int __init con3215_init(void)
 
 	/* Request the console irq */
 	if (raw3215_startup(raw) != 0) {
-		free_bootmem((unsigned long) raw->inbuf, RAW3215_INBUF_SIZE);
-		free_bootmem((unsigned long) raw->buffer, RAW3215_BUFFER_SIZE);
-		free_bootmem((unsigned long) raw, sizeof(struct raw3215_info));
+		kfree(raw->inbuf);
+		kfree(raw->buffer);
+		kfree(raw);
 		raw3215[0] = NULL;
 		return -ENODEV;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 33403dcfcdfd097d80213a715604eab2dca93b2e Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:05 +0200
Subject: [S390] 3270 console: convert from bootmem to slab

The slab allocator is earlier available so convert the
bootmem allocations to slab/gfp allocations.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/char/con3270.c | 13 +++++--------
 drivers/s390/char/raw3270.c | 32 ++------------------------------
 2 files changed, 7 insertions(+), 38 deletions(-)

diff --git a/drivers/s390/char/con3270.c b/drivers/s390/char/con3270.c
index 44d02e371c04..bb838bdf829d 100644
--- a/drivers/s390/char/con3270.c
+++ b/drivers/s390/char/con3270.c
@@ -7,7 +7,6 @@
  *     Copyright IBM Corp. 2003, 2009
  */
 
-#include <linux/bootmem.h>
 #include <linux/console.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
@@ -600,16 +599,14 @@ con3270_init(void)
 	if (IS_ERR(rp))
 		return PTR_ERR(rp);
 
-	condev = (struct con3270 *) alloc_bootmem_low(sizeof(struct con3270));
-	memset(condev, 0, sizeof(struct con3270));
+	condev = kzalloc(sizeof(struct con3270), GFP_KERNEL | GFP_DMA);
 	condev->view.dev = rp;
 
-	condev->read = raw3270_request_alloc_bootmem(0);
+	condev->read = raw3270_request_alloc(0);
 	condev->read->callback = con3270_read_callback;
 	condev->read->callback_data = condev;
-	condev->write = 
-		raw3270_request_alloc_bootmem(CON3270_OUTPUT_BUFFER_SIZE);
-	condev->kreset = raw3270_request_alloc_bootmem(1);
+	condev->write = raw3270_request_alloc(CON3270_OUTPUT_BUFFER_SIZE);
+	condev->kreset = raw3270_request_alloc(1);
 
 	INIT_LIST_HEAD(&condev->lines);
 	INIT_LIST_HEAD(&condev->update);
@@ -623,7 +620,7 @@ con3270_init(void)
 
 	INIT_LIST_HEAD(&condev->freemem);
 	for (i = 0; i < CON3270_STRING_PAGES; i++) {
-		cbuf = (void *) alloc_bootmem_low_pages(PAGE_SIZE);
+		cbuf = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
 		add_string_memory(&condev->freemem, cbuf, PAGE_SIZE);
 	}
 	condev->cline = alloc_string(&condev->freemem, condev->view.cols);
diff --git a/drivers/s390/char/raw3270.c b/drivers/s390/char/raw3270.c
index acab7b2dfe8a..9047b62294d0 100644
--- a/drivers/s390/char/raw3270.c
+++ b/drivers/s390/char/raw3270.c
@@ -7,7 +7,6 @@
  *     Copyright IBM Corp. 2003, 2009
  */
 
-#include <linux/bootmem.h>
 #include <linux/module.h>
 #include <linux/err.h>
 #include <linux/init.h>
@@ -143,33 +142,6 @@ raw3270_request_alloc(size_t size)
 	return rq;
 }
 
-#ifdef CONFIG_TN3270_CONSOLE
-/*
- * Allocate a new 3270 ccw request from bootmem. Only works very
- * early in the boot process. Only con3270.c should be using this.
- */
-struct raw3270_request __init *raw3270_request_alloc_bootmem(size_t size)
-{
-	struct raw3270_request *rq;
-
-	rq = alloc_bootmem_low(sizeof(struct raw3270));
-
-	/* alloc output buffer. */
-	if (size > 0)
-		rq->buffer = alloc_bootmem_low(size);
-	rq->size = size;
-	INIT_LIST_HEAD(&rq->list);
-
-	/*
-	 * Setup ccw.
-	 */
-	rq->ccw.cda = __pa(rq->buffer);
-	rq->ccw.flags = CCW_FLAG_SLI;
-
-	return rq;
-}
-#endif
-
 /*
  * Free 3270 ccw request
  */
@@ -846,8 +818,8 @@ struct raw3270 __init *raw3270_setup_console(struct ccw_device *cdev)
 	char *ascebc;
 	int rc;
 
-	rp = (struct raw3270 *) alloc_bootmem_low(sizeof(struct raw3270));
-	ascebc = (char *) alloc_bootmem(256);
+	rp = kzalloc(sizeof(struct raw3270), GFP_KERNEL | GFP_DMA);
+	ascebc = kzalloc(256, GFP_KERNEL);
 	rc = raw3270_setup_device(cdev, rp, ascebc);
 	if (rc)
 		return ERR_PTR(rc);
-- 
cgit v1.2.3-59-g8ed1b


From 4c8f4794b61e89dd68f96cfc23a9d9b6c25be420 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:06 +0200
Subject: [S390] sclp console: convert from bootmem to slab

The slab allocator is earlier available so convert the
bootmem allocations to slab/gfp allocations.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/char/sclp_con.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/s390/char/sclp_con.c b/drivers/s390/char/sclp_con.c
index 336811a77672..fb54e7e47e59 100644
--- a/drivers/s390/char/sclp_con.c
+++ b/drivers/s390/char/sclp_con.c
@@ -11,7 +11,6 @@
 #include <linux/init.h>
 #include <linux/timer.h>
 #include <linux/jiffies.h>
-#include <linux/bootmem.h>
 #include <linux/termios.h>
 #include <linux/err.h>
 #include <linux/reboot.h>
@@ -298,8 +297,8 @@ sclp_console_init(void)
 	/* Allocate pages for output buffering */
 	INIT_LIST_HEAD(&sclp_con_pages);
 	for (i = 0; i < MAX_CONSOLE_PAGES; i++) {
-		page = alloc_bootmem_low_pages(PAGE_SIZE);
-		list_add_tail((struct list_head *) page, &sclp_con_pages);
+		page = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+		list_add_tail(page, &sclp_con_pages);
 	}
 	INIT_LIST_HEAD(&sclp_con_outqueue);
 	spin_lock_init(&sclp_con_lock);
-- 
cgit v1.2.3-59-g8ed1b


From 5c0792f6924333290ec3ca31c02e6555d73dba04 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:07 +0200
Subject: [S390] vt220 console: convert from bootmem to slab

The slab allocator is earlier available so convert the
bootmem allocations to slab/gfp allocations.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/char/sclp_vt220.c | 18 +++++-------------
 1 file changed, 5 insertions(+), 13 deletions(-)

diff --git a/drivers/s390/char/sclp_vt220.c b/drivers/s390/char/sclp_vt220.c
index 5518e24946aa..178724f2a4c3 100644
--- a/drivers/s390/char/sclp_vt220.c
+++ b/drivers/s390/char/sclp_vt220.c
@@ -20,7 +20,6 @@
 #include <linux/major.h>
 #include <linux/console.h>
 #include <linux/kdev_t.h>
-#include <linux/bootmem.h>
 #include <linux/interrupt.h>
 #include <linux/init.h>
 #include <linux/reboot.h>
@@ -601,10 +600,7 @@ static void __init __sclp_vt220_free_pages(void)
 
 	list_for_each_safe(page, p, &sclp_vt220_empty) {
 		list_del(page);
-		if (slab_is_available())
-			free_page((unsigned long) page);
-		else
-			free_bootmem((unsigned long) page, PAGE_SIZE);
+		free_page((unsigned long) page);
 	}
 }
 
@@ -640,16 +636,12 @@ static int __init __sclp_vt220_init(int num_pages)
 	sclp_vt220_flush_later = 0;
 
 	/* Allocate pages for output buffering */
+	rc = -ENOMEM;
 	for (i = 0; i < num_pages; i++) {
-		if (slab_is_available())
-			page = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
-		else
-			page = alloc_bootmem_low_pages(PAGE_SIZE);
-		if (!page) {
-			rc = -ENOMEM;
+		page = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+		if (!page)
 			goto out;
-		}
-		list_add_tail((struct list_head *) page, &sclp_vt220_empty);
+		list_add_tail(page, &sclp_vt220_empty);
 	}
 	rc = sclp_register(&sclp_vt220_register);
 out:
-- 
cgit v1.2.3-59-g8ed1b


From 66d51f3e81b1067bdc836b3aba609eec957c693c Mon Sep 17 00:00:00 2001
From: Pekka Enberg <penberg@cs.helsinki.fi>
Date: Mon, 22 Jun 2009 12:08:08 +0200
Subject: [S390] s390: remove DEBUG_MALLOC

The kernel now has kmemleak and kmemtrace so there's no reason to keep
this ugly s390 hack around. I am not sure how it's supposed to work on
SMP anyway as it uses a global variable to temporarily store the return
value of all kmalloc() calls:

  void *b;

  #define kmalloc(x...) (PRINT_INFO(" kmalloc %p\n",b=kmalloc(x)),b)

Cc: <linux-s390@vger.kernel.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/debug.h | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/arch/s390/include/asm/debug.h b/arch/s390/include/asm/debug.h
index 9450ce6e32de..31ed5686a968 100644
--- a/arch/s390/include/asm/debug.h
+++ b/arch/s390/include/asm/debug.h
@@ -248,14 +248,5 @@ int debug_unregister_view(debug_info_t* id, struct debug_view* view);
 #define PRINT_FATAL(x...) printk ( KERN_DEBUG PRINTK_HEADER x )
 #endif				/* DASD_DEBUG */
 
-#undef DEBUG_MALLOC
-#ifdef DEBUG_MALLOC
-void *b;
-#define kmalloc(x...) (PRINT_INFO(" kmalloc %p\n",b=kmalloc(x)),b)
-#define kfree(x) PRINT_INFO(" kfree %p\n",x);kfree(x)
-#define get_zeroed_page(x...) (PRINT_INFO(" gfp %p\n",b=get_zeroed_page(x)),b)
-#define __get_free_pages(x...) (PRINT_INFO(" gfps %p\n",b=__get_free_pages(x)),b)
-#endif				/* DEBUG_MALLOC */
-
 #endif				/* __KERNEL__ */
 #endif				/* DEBUG_H */
-- 
cgit v1.2.3-59-g8ed1b


From f3dfa86caa4a54aceb2b235bf28a6f6ad73b2716 Mon Sep 17 00:00:00 2001
From: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Date: Mon, 22 Jun 2009 12:08:09 +0200
Subject: [S390] Use del_timer instead of del_timer_sync

When syncing the sclp console queue, we call del_timer_sync() while holding
the "sclp_con_lock" spinlock. This lock is also taken in the timer function
"sclp_console_timeout". Therefore the sync version of del_timer() cannot be
used here. Because the synchronous deletion of the timer is only needed
in the suspend callback and in that case only one CPU is remaining and
therefore it is not possible that the timer function is running in parallel,
we can safely use del_timer() instead of del_timer_sync().

Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/char/sclp_con.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/s390/char/sclp_con.c b/drivers/s390/char/sclp_con.c
index fb54e7e47e59..ad698d30cb3b 100644
--- a/drivers/s390/char/sclp_con.c
+++ b/drivers/s390/char/sclp_con.c
@@ -109,7 +109,7 @@ static void sclp_console_sync_queue(void)
 
 	spin_lock_irqsave(&sclp_con_lock, flags);
 	if (timer_pending(&sclp_con_timer))
-		del_timer_sync(&sclp_con_timer);
+		del_timer(&sclp_con_timer);
 	while (sclp_con_queue_running) {
 		spin_unlock_irqrestore(&sclp_con_lock, flags);
 		sclp_sync_wait();
-- 
cgit v1.2.3-59-g8ed1b


From 60b5df2f12f2ab54bfa7c1f0f0ce3f5953e73c0b Mon Sep 17 00:00:00 2001
From: Jan Glauber <jang@linux.vnet.ibm.com>
Date: Mon, 22 Jun 2009 12:08:10 +0200
Subject: [S390] qdio: move adapter interrupt tasklet code

Move the adapter interrupt tasklet function to the qdio main code
since all the functions used by the tasklet are located there.

Signed-off-by: Jan Glauber <jang@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/cio/qdio.h         | 11 +-----
 drivers/s390/cio/qdio_debug.c   |  3 +-
 drivers/s390/cio/qdio_main.c    | 84 +++++++++++++++++++++++++++++++++++------
 drivers/s390/cio/qdio_thinint.c | 57 ----------------------------
 4 files changed, 75 insertions(+), 80 deletions(-)

diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h
index 13bcb8114388..b1241f8fae88 100644
--- a/drivers/s390/cio/qdio.h
+++ b/drivers/s390/cio/qdio.h
@@ -351,15 +351,6 @@ static inline unsigned long long get_usecs(void)
 	((bufnr - dec) & QDIO_MAX_BUFFERS_MASK)
 
 /* prototypes for thin interrupt */
-void qdio_sync_after_thinint(struct qdio_q *q);
-int get_buf_state(struct qdio_q *q, unsigned int bufnr, unsigned char *state,
-		  int auto_ack);
-void qdio_check_outbound_after_thinint(struct qdio_q *q);
-int qdio_inbound_q_moved(struct qdio_q *q);
-void qdio_kick_handler(struct qdio_q *q);
-void qdio_stop_polling(struct qdio_q *q);
-int qdio_siga_sync_q(struct qdio_q *q);
-
 void qdio_setup_thinint(struct qdio_irq *irq_ptr);
 int qdio_establish_thinint(struct qdio_irq *irq_ptr);
 void qdio_shutdown_thinint(struct qdio_irq *irq_ptr);
@@ -392,4 +383,6 @@ void qdio_setup_destroy_sysfs(struct ccw_device *cdev);
 int qdio_setup_init(void);
 void qdio_setup_exit(void);
 
+int debug_get_buf_state(struct qdio_q *q, unsigned int bufnr,
+			unsigned char *state);
 #endif /* _CIO_QDIO_H */
diff --git a/drivers/s390/cio/qdio_debug.c b/drivers/s390/cio/qdio_debug.c
index e3434b34f86c..b8626d4df116 100644
--- a/drivers/s390/cio/qdio_debug.c
+++ b/drivers/s390/cio/qdio_debug.c
@@ -70,9 +70,8 @@ static int qstat_show(struct seq_file *m, void *v)
 	seq_printf(m, "slsb buffer states:\n");
 	seq_printf(m, "|0      |8      |16     |24     |32     |40     |48     |56  63|\n");
 
-	qdio_siga_sync_q(q);
 	for (i = 0; i < QDIO_MAX_BUFFERS_PER_Q; i++) {
-		get_buf_state(q, i, &state, 0);
+		debug_get_buf_state(q, i, &state);
 		switch (state) {
 		case SLSB_P_INPUT_NOT_INIT:
 		case SLSB_P_OUTPUT_NOT_INIT:
diff --git a/drivers/s390/cio/qdio_main.c b/drivers/s390/cio/qdio_main.c
index d79cf5bf0e62..377d881385cf 100644
--- a/drivers/s390/cio/qdio_main.c
+++ b/drivers/s390/cio/qdio_main.c
@@ -231,8 +231,8 @@ static inline int get_buf_states(struct qdio_q *q, unsigned int bufnr,
 	return i;
 }
 
-inline int get_buf_state(struct qdio_q *q, unsigned int bufnr,
-		  unsigned char *state, int auto_ack)
+static inline int get_buf_state(struct qdio_q *q, unsigned int bufnr,
+				unsigned char *state, int auto_ack)
 {
 	return get_buf_states(q, bufnr, state, 1, auto_ack);
 }
@@ -276,7 +276,7 @@ void qdio_init_buf_states(struct qdio_irq *irq_ptr)
 			       QDIO_MAX_BUFFERS_PER_Q);
 }
 
-static int qdio_siga_sync(struct qdio_q *q, unsigned int output,
+static inline int qdio_siga_sync(struct qdio_q *q, unsigned int output,
 			  unsigned int input)
 {
 	int cc;
@@ -293,7 +293,7 @@ static int qdio_siga_sync(struct qdio_q *q, unsigned int output,
 	return cc;
 }
 
-inline int qdio_siga_sync_q(struct qdio_q *q)
+static inline int qdio_siga_sync_q(struct qdio_q *q)
 {
 	if (q->is_input_q)
 		return qdio_siga_sync(q, 0, q->mask);
@@ -358,8 +358,7 @@ static inline int qdio_siga_input(struct qdio_q *q)
 	return cc;
 }
 
-/* called from thinint inbound handler */
-void qdio_sync_after_thinint(struct qdio_q *q)
+static inline void qdio_sync_after_thinint(struct qdio_q *q)
 {
 	if (pci_out_supported(q)) {
 		if (need_siga_sync_thinint(q))
@@ -370,7 +369,14 @@ void qdio_sync_after_thinint(struct qdio_q *q)
 		qdio_siga_sync_q(q);
 }
 
-inline void qdio_stop_polling(struct qdio_q *q)
+int debug_get_buf_state(struct qdio_q *q, unsigned int bufnr,
+			unsigned char *state)
+{
+	qdio_siga_sync_q(q);
+	return get_buf_states(q, bufnr, state, 1, 0);
+}
+
+static inline void qdio_stop_polling(struct qdio_q *q)
 {
 	if (!q->u.in.polling)
 		return;
@@ -516,7 +522,7 @@ out:
 	return q->first_to_check;
 }
 
-int qdio_inbound_q_moved(struct qdio_q *q)
+static int qdio_inbound_q_moved(struct qdio_q *q)
 {
 	int bufnr;
 
@@ -570,7 +576,23 @@ static int qdio_inbound_q_done(struct qdio_q *q)
 	}
 }
 
-void qdio_kick_handler(struct qdio_q *q)
+static inline int tiqdio_inbound_q_done(struct qdio_q *q)
+{
+	unsigned char state = 0;
+
+	if (!atomic_read(&q->nr_buf_used))
+		return 1;
+
+	qdio_siga_sync_q(q);
+	get_buf_state(q, q->first_to_check, &state, 0);
+
+	if (state == SLSB_P_INPUT_PRIMED)
+		/* more work coming */
+		return 0;
+	return 1;
+}
+
+static void qdio_kick_handler(struct qdio_q *q)
 {
 	int start = q->first_to_kick;
 	int end = q->first_to_check;
@@ -619,7 +641,6 @@ again:
 		goto again;
 }
 
-/* inbound tasklet */
 void qdio_inbound_processing(unsigned long data)
 {
 	struct qdio_q *q = (struct qdio_q *)data;
@@ -797,8 +818,7 @@ void qdio_outbound_timer(unsigned long data)
 	tasklet_schedule(&q->tasklet);
 }
 
-/* called from thinint inbound tasklet */
-void qdio_check_outbound_after_thinint(struct qdio_q *q)
+static inline void qdio_check_outbound_after_thinint(struct qdio_q *q)
 {
 	struct qdio_q *out;
 	int i;
@@ -811,6 +831,46 @@ void qdio_check_outbound_after_thinint(struct qdio_q *q)
 			tasklet_schedule(&out->tasklet);
 }
 
+static void __tiqdio_inbound_processing(struct qdio_q *q)
+{
+	qdio_perf_stat_inc(&perf_stats.thinint_inbound);
+	qdio_sync_after_thinint(q);
+
+	/*
+	 * The interrupt could be caused by a PCI request. Check the
+	 * PCI capable outbound queues.
+	 */
+	qdio_check_outbound_after_thinint(q);
+
+	if (!qdio_inbound_q_moved(q))
+		return;
+
+	qdio_kick_handler(q);
+
+	if (!tiqdio_inbound_q_done(q)) {
+		qdio_perf_stat_inc(&perf_stats.thinint_inbound_loop);
+		if (likely(q->irq_ptr->state != QDIO_IRQ_STATE_STOPPED))
+			tasklet_schedule(&q->tasklet);
+	}
+
+	qdio_stop_polling(q);
+	/*
+	 * We need to check again to not lose initiative after
+	 * resetting the ACK state.
+	 */
+	if (!tiqdio_inbound_q_done(q)) {
+		qdio_perf_stat_inc(&perf_stats.thinint_inbound_loop2);
+		if (likely(q->irq_ptr->state != QDIO_IRQ_STATE_STOPPED))
+			tasklet_schedule(&q->tasklet);
+	}
+}
+
+void tiqdio_inbound_processing(unsigned long data)
+{
+	struct qdio_q *q = (struct qdio_q *)data;
+	__tiqdio_inbound_processing(q);
+}
+
 static inline void qdio_set_state(struct qdio_irq *irq_ptr,
 				  enum qdio_irq_states state)
 {
diff --git a/drivers/s390/cio/qdio_thinint.c b/drivers/s390/cio/qdio_thinint.c
index c655d011a78d..e122f780f5ee 100644
--- a/drivers/s390/cio/qdio_thinint.c
+++ b/drivers/s390/cio/qdio_thinint.c
@@ -126,68 +126,11 @@ void tiqdio_remove_input_queues(struct qdio_irq *irq_ptr)
 	}
 }
 
-static inline int tiqdio_inbound_q_done(struct qdio_q *q)
-{
-	unsigned char state = 0;
-
-	if (!atomic_read(&q->nr_buf_used))
-		return 1;
-
-	qdio_siga_sync_q(q);
-	get_buf_state(q, q->first_to_check, &state, 0);
-
-	if (state == SLSB_P_INPUT_PRIMED)
-		/* more work coming */
-		return 0;
-	return 1;
-}
-
 static inline int shared_ind(struct qdio_irq *irq_ptr)
 {
 	return irq_ptr->dsci == &q_indicators[TIQDIO_SHARED_IND].ind;
 }
 
-static void __tiqdio_inbound_processing(struct qdio_q *q)
-{
-	qdio_perf_stat_inc(&perf_stats.thinint_inbound);
-	qdio_sync_after_thinint(q);
-
-	/*
-	 * Maybe we have work on our outbound queues... at least
-	 * we have to check the PCI capable queues.
-	 */
-	qdio_check_outbound_after_thinint(q);
-
-	if (!qdio_inbound_q_moved(q))
-		return;
-
-	qdio_kick_handler(q);
-
-	if (!tiqdio_inbound_q_done(q)) {
-		qdio_perf_stat_inc(&perf_stats.thinint_inbound_loop);
-		if (likely(q->irq_ptr->state != QDIO_IRQ_STATE_STOPPED))
-			tasklet_schedule(&q->tasklet);
-	}
-
-	qdio_stop_polling(q);
-	/*
-	 * We need to check again to not lose initiative after
-	 * resetting the ACK state.
-	 */
-	if (!tiqdio_inbound_q_done(q)) {
-		qdio_perf_stat_inc(&perf_stats.thinint_inbound_loop2);
-		if (likely(q->irq_ptr->state != QDIO_IRQ_STATE_STOPPED))
-			tasklet_schedule(&q->tasklet);
-	}
-}
-
-void tiqdio_inbound_processing(unsigned long data)
-{
-	struct qdio_q *q = (struct qdio_q *)data;
-
-	__tiqdio_inbound_processing(q);
-}
-
 /* check for work on all inbound thinint queues */
 static void tiqdio_tasklet_fn(unsigned long data)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 9a2c160a8cbd5b3253672b3bac462c64d0d2eef7 Mon Sep 17 00:00:00 2001
From: Jan Glauber <jang@linux.vnet.ibm.com>
Date: Mon, 22 Jun 2009 12:08:11 +0200
Subject: [S390] qdio: fix check for running under z/VM

The check whether qdio runs under z/VM was incorrect since SIGA-Sync is not
set if the device runs with QIOASSIST. Use MACHINE_IS_VM instead to prevent
polling under z/VM.

Merge qdio_inbound_q_done and tiqdio_is_inbound_q_done.

Signed-off-by: Jan Glauber <jang@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/cio/qdio_main.c | 48 ++++++++++++--------------------------------
 1 file changed, 13 insertions(+), 35 deletions(-)

diff --git a/drivers/s390/cio/qdio_main.c b/drivers/s390/cio/qdio_main.c
index 377d881385cf..127e78eef651 100644
--- a/drivers/s390/cio/qdio_main.c
+++ b/drivers/s390/cio/qdio_main.c
@@ -499,7 +499,7 @@ check_next:
 		/*
 		 * No siga-sync needed for non-qebsm here, as the inbound queue
 		 * will be synced on the next siga-r, resp.
-		 * tiqdio_is_inbound_q_done will do the siga-sync.
+		 * qdio_inbound_q_done will do the siga-sync.
 		 */
 		q->first_to_check = add_buf(q->first_to_check, count);
 		atomic_sub(count, &q->nr_buf_used);
@@ -530,35 +530,32 @@ static int qdio_inbound_q_moved(struct qdio_q *q)
 
 	if ((bufnr != q->last_move) || q->qdio_error) {
 		q->last_move = bufnr;
-		if (!need_siga_sync(q) && !pci_out_supported(q))
+		if (!is_thinint_irq(q->irq_ptr) && !MACHINE_IS_VM)
 			q->u.in.timestamp = get_usecs();
-
-		DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "in moved");
 		return 1;
 	} else
 		return 0;
 }
 
-static int qdio_inbound_q_done(struct qdio_q *q)
+static inline int qdio_inbound_q_done(struct qdio_q *q)
 {
 	unsigned char state = 0;
 
 	if (!atomic_read(&q->nr_buf_used))
 		return 1;
 
-	/*
-	 * We need that one for synchronization with the adapter, as it
-	 * does a kind of PCI avoidance.
-	 */
 	qdio_siga_sync_q(q);
-
 	get_buf_state(q, q->first_to_check, &state, 0);
+
 	if (state == SLSB_P_INPUT_PRIMED)
-		/* we got something to do */
+		/* more work coming */
 		return 0;
 
-	/* on VM, we don't poll, so the q is always done here */
-	if (need_siga_sync(q) || pci_out_supported(q))
+	if (is_thinint_irq(q->irq_ptr))
+		return 1;
+
+	/* don't poll under z/VM */
+	if (MACHINE_IS_VM)
 		return 1;
 
 	/*
@@ -569,27 +566,8 @@ static int qdio_inbound_q_done(struct qdio_q *q)
 		DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "in done:%3d",
 			      q->first_to_check);
 		return 1;
-	} else {
-		DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "in notd:%3d",
-			      q->first_to_check);
-		return 0;
-	}
-}
-
-static inline int tiqdio_inbound_q_done(struct qdio_q *q)
-{
-	unsigned char state = 0;
-
-	if (!atomic_read(&q->nr_buf_used))
-		return 1;
-
-	qdio_siga_sync_q(q);
-	get_buf_state(q, q->first_to_check, &state, 0);
-
-	if (state == SLSB_P_INPUT_PRIMED)
-		/* more work coming */
+	} else
 		return 0;
-	return 1;
 }
 
 static void qdio_kick_handler(struct qdio_q *q)
@@ -847,7 +825,7 @@ static void __tiqdio_inbound_processing(struct qdio_q *q)
 
 	qdio_kick_handler(q);
 
-	if (!tiqdio_inbound_q_done(q)) {
+	if (!qdio_inbound_q_done(q)) {
 		qdio_perf_stat_inc(&perf_stats.thinint_inbound_loop);
 		if (likely(q->irq_ptr->state != QDIO_IRQ_STATE_STOPPED))
 			tasklet_schedule(&q->tasklet);
@@ -858,7 +836,7 @@ static void __tiqdio_inbound_processing(struct qdio_q *q)
 	 * We need to check again to not lose initiative after
 	 * resetting the ACK state.
 	 */
-	if (!tiqdio_inbound_q_done(q)) {
+	if (!qdio_inbound_q_done(q)) {
 		qdio_perf_stat_inc(&perf_stats.thinint_inbound_loop2);
 		if (likely(q->irq_ptr->state != QDIO_IRQ_STATE_STOPPED))
 			tasklet_schedule(&q->tasklet);
-- 
cgit v1.2.3-59-g8ed1b


From 36e3e72120e27939233e4bd88a8d74b3a2377428 Mon Sep 17 00:00:00 2001
From: Jan Glauber <jang@linux.vnet.ibm.com>
Date: Mon, 22 Jun 2009 12:08:12 +0200
Subject: [S390] qdio: extract all primed SBALs at once

For devices without QIOASSIST primed SBALS were extracted in a loop.
Remove the loop since get_buf_states can already return more than
one primed SBAL.

Signed-off-by: Jan Glauber <jang@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/cio/qdio_main.c | 34 ++++++----------------------------
 1 file changed, 6 insertions(+), 28 deletions(-)

diff --git a/drivers/s390/cio/qdio_main.c b/drivers/s390/cio/qdio_main.c
index 127e78eef651..779b7741d495 100644
--- a/drivers/s390/cio/qdio_main.c
+++ b/drivers/s390/cio/qdio_main.c
@@ -476,19 +476,13 @@ static int get_inbound_buffer_frontier(struct qdio_q *q)
 	count = min(atomic_read(&q->nr_buf_used), QDIO_MAX_BUFFERS_MASK);
 	stop = add_buf(q->first_to_check, count);
 
-	/*
-	 * No siga sync here, as a PCI or we after a thin interrupt
-	 * will sync the queues.
-	 */
-
-	/* need to set count to 1 for non-qebsm */
-	if (!is_qebsm(q))
-		count = 1;
-
-check_next:
 	if (q->first_to_check == stop)
 		goto out;
 
+	/*
+	 * No siga sync here, as a PCI or we after a thin interrupt
+	 * already sync'ed the queues.
+	 */
 	count = get_buf_states(q, q->first_to_check, &state, count, 1);
 	if (!count)
 		goto out;
@@ -496,14 +490,9 @@ check_next:
 	switch (state) {
 	case SLSB_P_INPUT_PRIMED:
 		inbound_primed(q, count);
-		/*
-		 * No siga-sync needed for non-qebsm here, as the inbound queue
-		 * will be synced on the next siga-r, resp.
-		 * qdio_inbound_q_done will do the siga-sync.
-		 */
 		q->first_to_check = add_buf(q->first_to_check, count);
 		atomic_sub(count, &q->nr_buf_used);
-		goto check_next;
+		break;
 	case SLSB_P_INPUT_ERROR:
 		announce_buffer_error(q, count);
 		/* process the buffer, the upper layer will take care of it */
@@ -641,11 +630,6 @@ static int get_outbound_buffer_frontier(struct qdio_q *q)
 	count = min(atomic_read(&q->nr_buf_used), QDIO_MAX_BUFFERS_MASK);
 	stop = add_buf(q->first_to_check, count);
 
-	/* need to set count to 1 for non-qebsm */
-	if (!is_qebsm(q))
-		count = 1;
-
-check_next:
 	if (q->first_to_check == stop)
 		return q->first_to_check;
 
@@ -660,13 +644,7 @@ check_next:
 
 		atomic_sub(count, &q->nr_buf_used);
 		q->first_to_check = add_buf(q->first_to_check, count);
-		/*
-		 * We fetch all buffer states at once. get_buf_states may
-		 * return count < stop. For QEBSM we do not loop.
-		 */
-		if (is_qebsm(q))
-			break;
-		goto check_next;
+		break;
 	case SLSB_P_OUTPUT_ERROR:
 		announce_buffer_error(q, count);
 		/* process the buffer, the upper layer will take care of it */
-- 
cgit v1.2.3-59-g8ed1b


From cf9a031c2cc881e9873ab9ccf5e1f59f5b5167aa Mon Sep 17 00:00:00 2001
From: Jan Glauber <jang@linux.vnet.ibm.com>
Date: Mon, 22 Jun 2009 12:08:13 +0200
Subject: [S390] qdio: merge AI tasklet into interrupt handler

Since the adapter interrupt tasklet only schedules the queue tasklets
and contains no code that requires serialization in can be merged
with the adapter interrupt handler. That possibly safes some CPU
cycles.

Signed-off-by: Jan Glauber <jang@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/cio/qdio_thinint.c | 65 +++++++++++++----------------------------
 1 file changed, 21 insertions(+), 44 deletions(-)

diff --git a/drivers/s390/cio/qdio_thinint.c b/drivers/s390/cio/qdio_thinint.c
index e122f780f5ee..981a77ea7ee2 100644
--- a/drivers/s390/cio/qdio_thinint.c
+++ b/drivers/s390/cio/qdio_thinint.c
@@ -43,9 +43,6 @@ struct indicator_t {
 };
 static struct indicator_t *q_indicators;
 
-static void tiqdio_tasklet_fn(unsigned long data);
-static DECLARE_TASKLET(tiqdio_tasklet, tiqdio_tasklet_fn, 0);
-
 static int css_qdio_omit_svs;
 
 static inline unsigned long do_clear_global_summary(void)
@@ -103,11 +100,6 @@ void tiqdio_add_input_queues(struct qdio_irq *irq_ptr)
 	xchg(irq_ptr->dsci, 1);
 }
 
-/*
- * we cannot stop the tiqdio tasklet here since it is for all
- * thinint qdio devices and it must run as long as there is a
- * thinint device left
- */
 void tiqdio_remove_input_queues(struct qdio_irq *irq_ptr)
 {
 	struct qdio_q *q;
@@ -131,17 +123,34 @@ static inline int shared_ind(struct qdio_irq *irq_ptr)
 	return irq_ptr->dsci == &q_indicators[TIQDIO_SHARED_IND].ind;
 }
 
-/* check for work on all inbound thinint queues */
-static void tiqdio_tasklet_fn(unsigned long data)
+/**
+ * tiqdio_thinint_handler - thin interrupt handler for qdio
+ * @ind: pointer to adapter local summary indicator
+ * @drv_data: NULL
+ */
+static void tiqdio_thinint_handler(void *ind, void *drv_data)
 {
 	struct qdio_q *q;
 
-	qdio_perf_stat_inc(&perf_stats.tasklet_thinint);
-again:
+	qdio_perf_stat_inc(&perf_stats.thin_int);
+
+	/*
+	 * SVS only when needed: issue SVS to benefit from iqdio interrupt
+	 * avoidance (SVS clears adapter interrupt suppression overwrite)
+	 */
+	if (!css_qdio_omit_svs)
+		do_clear_global_summary();
+
+	/*
+	 * reset local summary indicator (tiqdio_alsi) to stop adapter
+	 * interrupts for now
+	 */
+	xchg((u8 *)ind, 0);
 
 	/* protect tiq_list entries, only changed in activate or shutdown */
 	rcu_read_lock();
 
+	/* check for work on all inbound thinint queues */
 	list_for_each_entry_rcu(q, &tiq_list, entry)
 		/* only process queues from changed sets */
 		if (*q->irq_ptr->dsci) {
@@ -169,37 +178,6 @@ again:
 		if (*tiqdio_alsi)
 			xchg(&q_indicators[TIQDIO_SHARED_IND].ind, 1);
 	}
-
-	/* check for more work */
-	if (*tiqdio_alsi) {
-		xchg(tiqdio_alsi, 0);
-		qdio_perf_stat_inc(&perf_stats.tasklet_thinint_loop);
-		goto again;
-	}
-}
-
-/**
- * tiqdio_thinint_handler - thin interrupt handler for qdio
- * @ind: pointer to adapter local summary indicator
- * @drv_data: NULL
- */
-static void tiqdio_thinint_handler(void *ind, void *drv_data)
-{
-	qdio_perf_stat_inc(&perf_stats.thin_int);
-
-	/*
-	 * SVS only when needed: issue SVS to benefit from iqdio interrupt
-	 * avoidance (SVS clears adapter interrupt suppression overwrite)
-	 */
-	if (!css_qdio_omit_svs)
-		do_clear_global_summary();
-
-	/*
-	 * reset local summary indicator (tiqdio_alsi) to stop adapter
-	 * interrupts for now, the tasklet will clean all dsci's
-	 */
-	xchg((u8 *)ind, 0);
-	tasklet_hi_schedule(&tiqdio_tasklet);
 }
 
 static int set_subchannel_ind(struct qdio_irq *irq_ptr, int reset)
@@ -319,5 +297,4 @@ void __exit tiqdio_unregister_thinints(void)
 		s390_unregister_adapter_interrupt(tiqdio_alsi, QDIO_AIRQ_ISC);
 		isc_unregister(QDIO_AIRQ_ISC);
 	}
-	tasklet_kill(&tiqdio_tasklet);
 }
-- 
cgit v1.2.3-59-g8ed1b


From f0a0b15e0f3aff0a25f21f58bef8e40e80b16dc6 Mon Sep 17 00:00:00 2001
From: Jan Glauber <jang@linux.vnet.ibm.com>
Date: Mon, 22 Jun 2009 12:08:14 +0200
Subject: [S390] qdio: leave inbound SBALs primed

It is not required to change the state of primed SBALs. Leaving them
primed saves a SQBS instruction under z/VM.

Signed-off-by: Jan Glauber <jang@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/cio/qdio_main.c | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/drivers/s390/cio/qdio_main.c b/drivers/s390/cio/qdio_main.c
index 779b7741d495..75b521963a4e 100644
--- a/drivers/s390/cio/qdio_main.c
+++ b/drivers/s390/cio/qdio_main.c
@@ -455,13 +455,6 @@ static inline void inbound_primed(struct qdio_q *q, int count)
 	count--;
 	if (!count)
 		return;
-
-	/*
-	 * Need to change all PRIMED buffers to NOT_INIT, otherwise
-	 * we're loosing initiative in the thinint code.
-	 */
-	set_buf_states(q, q->first_to_check, SLSB_P_INPUT_NOT_INIT,
-		       count);
 }
 
 static int get_inbound_buffer_frontier(struct qdio_q *q)
-- 
cgit v1.2.3-59-g8ed1b


From 6618241b47cd131503610d8df68dd6f4948e5c1a Mon Sep 17 00:00:00 2001
From: Jan Glauber <jang@linux.vnet.ibm.com>
Date: Mon, 22 Jun 2009 12:08:15 +0200
Subject: [S390] qdio: Sanitize do_QDIO sanity checks

Remove unneeded sanity checks from do_QDIO since this is the hot path.
Change the type of bufnr and count to unsigned int so the check for the
maximum value works.

Reported-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Jan Glauber <jang@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/qdio.h | 2 +-
 drivers/s390/cio/qdio_main.c | 9 ++-------
 2 files changed, 3 insertions(+), 8 deletions(-)

diff --git a/arch/s390/include/asm/qdio.h b/arch/s390/include/asm/qdio.h
index 402d6dcf0d26..79d849f014f0 100644
--- a/arch/s390/include/asm/qdio.h
+++ b/arch/s390/include/asm/qdio.h
@@ -380,7 +380,7 @@ extern int qdio_establish(struct qdio_initialize *);
 extern int qdio_activate(struct ccw_device *);
 
 extern int do_QDIO(struct ccw_device *cdev, unsigned int callflags,
-		   int q_nr, int bufnr, int count);
+		   int q_nr, unsigned int bufnr, unsigned int count);
 extern int qdio_cleanup(struct ccw_device*, int);
 extern int qdio_shutdown(struct ccw_device*, int);
 extern int qdio_free(struct ccw_device *);
diff --git a/drivers/s390/cio/qdio_main.c b/drivers/s390/cio/qdio_main.c
index 75b521963a4e..0038750ad945 100644
--- a/drivers/s390/cio/qdio_main.c
+++ b/drivers/s390/cio/qdio_main.c
@@ -1497,18 +1497,13 @@ out:
  * @count: how many buffers to process
  */
 int do_QDIO(struct ccw_device *cdev, unsigned int callflags,
-	    int q_nr, int bufnr, int count)
+	    int q_nr, unsigned int bufnr, unsigned int count)
 {
 	struct qdio_irq *irq_ptr;
 
-	if ((bufnr > QDIO_MAX_BUFFERS_PER_Q) ||
-	    (count > QDIO_MAX_BUFFERS_PER_Q) ||
-	    (q_nr >= QDIO_MAX_QUEUES_PER_IRQ))
+	if (bufnr >= QDIO_MAX_BUFFERS_PER_Q || count > QDIO_MAX_BUFFERS_PER_Q)
 		return -EINVAL;
 
-	if (!count)
-		return 0;
-
 	irq_ptr = cdev->private->qdio_data;
 	if (!irq_ptr)
 		return -ENODEV;
-- 
cgit v1.2.3-59-g8ed1b


From 772f54720ab82a6e88f0a8a84d76e7af15ca1f0c Mon Sep 17 00:00:00 2001
From: Felix Beck <felix.beck@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:16 +0200
Subject: [S390] ap/zcrypt: Suspend/Resume ap bus and zcrypt

Add Suspend/Resume support to ap bus and zcrypt. All enhancements are
done in the ap bus. No changes in the crypto card specific part are
necessary.

Signed-off-by: Felix Beck <felix.beck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/crypto/ap_bus.c | 85 ++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 83 insertions(+), 2 deletions(-)

diff --git a/drivers/s390/crypto/ap_bus.c b/drivers/s390/crypto/ap_bus.c
index 9c148406b980..727a809636d8 100644
--- a/drivers/s390/crypto/ap_bus.c
+++ b/drivers/s390/crypto/ap_bus.c
@@ -54,6 +54,12 @@ static int ap_poll_thread_start(void);
 static void ap_poll_thread_stop(void);
 static void ap_request_timeout(unsigned long);
 static inline void ap_schedule_poll_timer(void);
+static int __ap_poll_device(struct ap_device *ap_dev, unsigned long *flags);
+static int ap_device_remove(struct device *dev);
+static int ap_device_probe(struct device *dev);
+static void ap_interrupt_handler(void *unused1, void *unused2);
+static void ap_reset(struct ap_device *ap_dev);
+static void ap_config_timeout(unsigned long ptr);
 
 /*
  * Module description.
@@ -101,6 +107,10 @@ static struct hrtimer ap_poll_timer;
  * If z/VM change to 1500000 nanoseconds to adjust to z/VM polling.*/
 static unsigned long long poll_timeout = 250000;
 
+/* Suspend flag */
+static int ap_suspend_flag;
+static struct bus_type ap_bus_type;
+
 /**
  * ap_using_interrupts() - Returns non-zero if interrupt support is
  * available.
@@ -617,10 +627,79 @@ static int ap_uevent (struct device *dev, struct kobj_uevent_env *env)
 	return retval;
 }
 
+static int ap_bus_suspend(struct device *dev, pm_message_t state)
+{
+	struct ap_device *ap_dev = to_ap_dev(dev);
+	unsigned long flags;
+
+	if (!ap_suspend_flag) {
+		ap_suspend_flag = 1;
+
+		/* Disable scanning for devices, thus we do not want to scan
+		 * for them after removing.
+		 */
+		del_timer_sync(&ap_config_timer);
+		if (ap_work_queue != NULL) {
+			destroy_workqueue(ap_work_queue);
+			ap_work_queue = NULL;
+		}
+		tasklet_disable(&ap_tasklet);
+	}
+	/* Poll on the device until all requests are finished. */
+	do {
+		flags = 0;
+		__ap_poll_device(ap_dev, &flags);
+	} while ((flags & 1) || (flags & 2));
+
+	ap_device_remove(dev);
+	return 0;
+}
+
+static int ap_bus_resume(struct device *dev)
+{
+	int rc = 0;
+	struct ap_device *ap_dev = to_ap_dev(dev);
+
+	if (ap_suspend_flag) {
+		ap_suspend_flag = 0;
+		if (!ap_interrupts_available())
+			ap_interrupt_indicator = NULL;
+		ap_device_probe(dev);
+		ap_reset(ap_dev);
+		setup_timer(&ap_dev->timeout, ap_request_timeout,
+			    (unsigned long) ap_dev);
+		ap_scan_bus(NULL);
+		init_timer(&ap_config_timer);
+		ap_config_timer.function = ap_config_timeout;
+		ap_config_timer.data = 0;
+		ap_config_timer.expires = jiffies + ap_config_time * HZ;
+		add_timer(&ap_config_timer);
+		ap_work_queue = create_singlethread_workqueue("kapwork");
+		if (!ap_work_queue)
+			return -ENOMEM;
+		tasklet_enable(&ap_tasklet);
+		if (!ap_using_interrupts())
+			ap_schedule_poll_timer();
+		else
+			tasklet_schedule(&ap_tasklet);
+		if (ap_thread_flag)
+			rc = ap_poll_thread_start();
+	} else {
+		ap_device_probe(dev);
+		ap_reset(ap_dev);
+		setup_timer(&ap_dev->timeout, ap_request_timeout,
+			    (unsigned long) ap_dev);
+	}
+
+	return rc;
+}
+
 static struct bus_type ap_bus_type = {
 	.name = "ap",
 	.match = &ap_bus_match,
 	.uevent = &ap_uevent,
+	.suspend = ap_bus_suspend,
+	.resume = ap_bus_resume
 };
 
 static int ap_device_probe(struct device *dev)
@@ -1066,7 +1145,7 @@ ap_config_timeout(unsigned long ptr)
  */
 static inline void ap_schedule_poll_timer(void)
 {
-	if (ap_using_interrupts())
+	if (ap_using_interrupts() || ap_suspend_flag)
 		return;
 	if (hrtimer_is_queued(&ap_poll_timer))
 		return;
@@ -1384,6 +1463,8 @@ static int ap_poll_thread(void *data)
 
 	set_user_nice(current, 19);
 	while (1) {
+		if (ap_suspend_flag)
+			return 0;
 		if (need_resched()) {
 			schedule();
 			continue;
@@ -1414,7 +1495,7 @@ static int ap_poll_thread_start(void)
 {
 	int rc;
 
-	if (ap_using_interrupts())
+	if (ap_using_interrupts() || ap_suspend_flag)
 		return 0;
 	mutex_lock(&ap_poll_thread_mutex);
 	if (!ap_poll_kthread) {
-- 
cgit v1.2.3-59-g8ed1b


From e6125fba81e362d9b314d10893af1d9dc5658f33 Mon Sep 17 00:00:00 2001
From: Stefan Haberland <stefan.haberland@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:17 +0200
Subject: [S390] dasd_pm: fix stop flag handling

The stop flags are handled in the generic restore function so the
stop flag is removed also for FBA and DIAG devices.

Signed-off-by: Stefan Haberland <stefan.haberland@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/block/dasd.c      | 12 +++++++++++-
 drivers/s390/block/dasd_eckd.c | 10 +---------
 2 files changed, 12 insertions(+), 10 deletions(-)

diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index e5b84db0aa03..99e71536213a 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -2503,15 +2503,25 @@ int dasd_generic_restore_device(struct ccw_device *cdev)
 	if (IS_ERR(device))
 		return PTR_ERR(device);
 
+	/* allow new IO again */
+	device->stopped &= ~DASD_STOPPED_PM;
+	device->stopped &= ~DASD_UNRESUMED_PM;
+
 	dasd_schedule_device_bh(device);
 	if (device->block)
 		dasd_schedule_block_bh(device->block);
 
 	if (device->discipline->restore)
 		rc = device->discipline->restore(device);
+	if (rc)
+		/*
+		 * if the resume failed for the DASD we put it in
+		 * an UNRESUMED stop state
+		 */
+		device->stopped |= DASD_UNRESUMED_PM;
 
 	dasd_put_device(device);
-	return rc;
+	return 0;
 }
 EXPORT_SYMBOL_GPL(dasd_generic_restore_device);
 
diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c
index 1c28ec3e4ccb..f8b1f04f26b8 100644
--- a/drivers/s390/block/dasd_eckd.c
+++ b/drivers/s390/block/dasd_eckd.c
@@ -3243,9 +3243,6 @@ int dasd_eckd_restore_device(struct dasd_device *device)
 	int is_known, rc;
 	struct dasd_uid temp_uid;
 
-	/* allow new IO again */
-	device->stopped &= ~DASD_STOPPED_PM;
-
 	private = (struct dasd_eckd_private *) device->private;
 
 	/* Read Configuration Data */
@@ -3295,12 +3292,7 @@ int dasd_eckd_restore_device(struct dasd_device *device)
 	return 0;
 
 out_err:
-	/*
-	 * if the resume failed for the DASD we put it in
-	 * an UNRESUMED stop state
-	 */
-	device->stopped |= DASD_UNRESUMED_PM;
-	return 0;
+	return -1;
 }
 
 static struct ccw_driver dasd_eckd_driver = {
-- 
cgit v1.2.3-59-g8ed1b


From 4a9c75255e1fef4247cf960d3c3eb528c8cd8409 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:18 +0200
Subject: [S390] pm: fix build error for !SMP

Fix build error for !SMP:

arch/s390/power/built-in.o: In function `swsusp_arch_resume':
(.text+0x1b4): undefined reference to `smp_get_phys_cpu_id'
arch/s390/power/built-in.o: In function `swsusp_arch_resume':
(.text+0x288): undefined reference to `smp_switch_boot_cpu_in_resume'
make: *** [.tmp_vmlinux1] Error 1

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/power/swsusp_asm64.S | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/arch/s390/power/swsusp_asm64.S b/arch/s390/power/swsusp_asm64.S
index 3c74e7d827c9..76d688da32fa 100644
--- a/arch/s390/power/swsusp_asm64.S
+++ b/arch/s390/power/swsusp_asm64.S
@@ -109,10 +109,11 @@ swsusp_arch_resume:
 	aghi	%r15,-STACK_FRAME_OVERHEAD
 	stg	%r1,__SF_BACKCHAIN(%r15)
 
+#ifdef CONFIG_SMP
 	/* Save boot cpu number */
 	brasl	%r14,smp_get_phys_cpu_id
 	lgr	%r10,%r2
-
+#endif
 	/* Deactivate DAT */
 	stnsm	__SF_EMPTY(%r15),0xfb
 
@@ -177,11 +178,12 @@ swsusp_arch_resume:
 	/* Pointer to save arae */
 	lghi	%r13,0x1000
 
+#ifdef CONFIG_SMP
 	/* Switch CPUs */
 	lgr	%r2,%r10		/* get cpu id */
 	llgf	%r3,0x318(%r13)
 	brasl	%r14,smp_switch_boot_cpu_in_resume
-
+#endif
 	/* Restore prefix register */
 	spx	0x318(%r13)
 
-- 
cgit v1.2.3-59-g8ed1b


From 4f0076f77fb64889d4e5e425b63333e5764b446d Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:19 +0200
Subject: [S390] driver_data access

Replace the remaining direct accesses to the driver_data pointer
with calls to the dev_get_drvdata() and dev_set_drvdata() functions.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/char/con3215.c   | 4 ++--
 drivers/s390/char/monreader.c | 6 +++---
 drivers/s390/char/raw3270.c   | 4 ++--
 drivers/s390/char/tape_core.c | 2 +-
 drivers/s390/char/vmlogrdr.c  | 4 ++--
 drivers/s390/char/vmur.c      | 2 +-
 drivers/s390/net/netiucv.c    | 4 ++--
 7 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/drivers/s390/char/con3215.c b/drivers/s390/char/con3215.c
index 51e6379c5b93..21639d6c996f 100644
--- a/drivers/s390/char/con3215.c
+++ b/drivers/s390/char/con3215.c
@@ -732,7 +732,7 @@ static int raw3215_pm_stop(struct ccw_device *cdev)
 	unsigned long flags;
 
 	/* Empty the output buffer, then prevent new I/O. */
-	raw = cdev->dev.driver_data;
+	raw = dev_get_drvdata(&cdev->dev);
 	spin_lock_irqsave(get_ccwdev_lock(raw->cdev), flags);
 	raw3215_make_room(raw, RAW3215_BUFFER_SIZE);
 	raw->flags |= RAW3215_FROZEN;
@@ -746,7 +746,7 @@ static int raw3215_pm_start(struct ccw_device *cdev)
 	unsigned long flags;
 
 	/* Allow I/O again and flush output buffer. */
-	raw = cdev->dev.driver_data;
+	raw = dev_get_drvdata(&cdev->dev);
 	spin_lock_irqsave(get_ccwdev_lock(raw->cdev), flags);
 	raw->flags &= ~RAW3215_FROZEN;
 	raw->flags |= RAW3215_FLUSHING;
diff --git a/drivers/s390/char/monreader.c b/drivers/s390/char/monreader.c
index 75a8831eebbc..7892550d7932 100644
--- a/drivers/s390/char/monreader.c
+++ b/drivers/s390/char/monreader.c
@@ -320,7 +320,7 @@ static int mon_open(struct inode *inode, struct file *filp)
 		goto out_path;
 	}
 	filp->private_data = monpriv;
-	monreader_device->driver_data = monpriv;
+	dev_set_drvdata(&monreader_device, monpriv);
 	unlock_kernel();
 	return nonseekable_open(inode, filp);
 
@@ -463,7 +463,7 @@ static struct miscdevice mon_dev = {
  *****************************************************************************/
 static int monreader_freeze(struct device *dev)
 {
-	struct mon_private *monpriv = dev->driver_data;
+	struct mon_private *monpriv = dev_get_drvdata(&dev);
 	int rc;
 
 	if (!monpriv)
@@ -487,7 +487,7 @@ static int monreader_freeze(struct device *dev)
 
 static int monreader_thaw(struct device *dev)
 {
-	struct mon_private *monpriv = dev->driver_data;
+	struct mon_private *monpriv = dev_get_drvdata(dev);
 	int rc;
 
 	if (!monpriv)
diff --git a/drivers/s390/char/raw3270.c b/drivers/s390/char/raw3270.c
index 9047b62294d0..d6a022f55e92 100644
--- a/drivers/s390/char/raw3270.c
+++ b/drivers/s390/char/raw3270.c
@@ -1322,7 +1322,7 @@ static int raw3270_pm_stop(struct ccw_device *cdev)
 	struct raw3270_view *view;
 	unsigned long flags;
 
-	rp = cdev->dev.driver_data;
+	rp = dev_get_drvdata(&cdev->dev);
 	if (!rp)
 		return 0;
 	spin_lock_irqsave(get_ccwdev_lock(rp->cdev), flags);
@@ -1348,7 +1348,7 @@ static int raw3270_pm_start(struct ccw_device *cdev)
 	struct raw3270 *rp;
 	unsigned long flags;
 
-	rp = cdev->dev.driver_data;
+	rp = dev_get_drvdata(&cdev->dev);
 	if (!rp)
 		return 0;
 	spin_lock_irqsave(get_ccwdev_lock(rp->cdev), flags);
diff --git a/drivers/s390/char/tape_core.c b/drivers/s390/char/tape_core.c
index 595aa04cfd01..1d420d947596 100644
--- a/drivers/s390/char/tape_core.c
+++ b/drivers/s390/char/tape_core.c
@@ -396,7 +396,7 @@ int tape_generic_pm_suspend(struct ccw_device *cdev)
 {
 	struct tape_device *device;
 
-	device = cdev->dev.driver_data;
+	device = dev_get_drvdata(&cdev->dev);
 	if (!device) {
 		return -ENODEV;
 	}
diff --git a/drivers/s390/char/vmlogrdr.c b/drivers/s390/char/vmlogrdr.c
index 411cfa3c7719..c20a4fe6da51 100644
--- a/drivers/s390/char/vmlogrdr.c
+++ b/drivers/s390/char/vmlogrdr.c
@@ -663,7 +663,7 @@ static struct attribute *vmlogrdr_attrs[] = {
 static int vmlogrdr_pm_prepare(struct device *dev)
 {
 	int rc;
-	struct vmlogrdr_priv_t *priv = dev->driver_data;
+	struct vmlogrdr_priv_t *priv = dev_get_drvdata(dev);
 
 	rc = 0;
 	if (priv) {
@@ -753,7 +753,7 @@ static int vmlogrdr_register_device(struct vmlogrdr_priv_t *priv)
 		dev->bus = &iucv_bus;
 		dev->parent = iucv_root;
 		dev->driver = &vmlogrdr_driver;
-		dev->driver_data = priv;
+		dev_set_drvdata(dev, priv);
 		/*
 		 * The release function could be called after the
 		 * module has been unloaded. It's _only_ task is to
diff --git a/drivers/s390/char/vmur.c b/drivers/s390/char/vmur.c
index 7d9e67cb6471..31b902e94f7b 100644
--- a/drivers/s390/char/vmur.c
+++ b/drivers/s390/char/vmur.c
@@ -170,7 +170,7 @@ static void urdev_put(struct urdev *urd)
  */
 static int ur_pm_suspend(struct ccw_device *cdev)
 {
-	struct urdev *urd = cdev->dev.driver_data;
+	struct urdev *urd = dev_get_drvdata(&cdev->dev);
 
 	TRACE("ur_pm_suspend: cdev=%p\n", cdev);
 	if (urd->open_flag) {
diff --git a/drivers/s390/net/netiucv.c b/drivers/s390/net/netiucv.c
index 52574ce797b2..8c36eafcfbfe 100644
--- a/drivers/s390/net/netiucv.c
+++ b/drivers/s390/net/netiucv.c
@@ -1307,7 +1307,7 @@ static void netiucv_pm_complete(struct device *dev)
  */
 static int netiucv_pm_freeze(struct device *dev)
 {
-	struct netiucv_priv *priv = dev->driver_data;
+	struct netiucv_priv *priv = dev_get_drvdata(dev);
 	struct net_device *ndev = NULL;
 	int rc = 0;
 
@@ -1331,7 +1331,7 @@ out:
  */
 static int netiucv_pm_restore_thaw(struct device *dev)
 {
-	struct netiucv_priv *priv = dev->driver_data;
+	struct netiucv_priv *priv = dev_get_drvdata(dev);
 	struct net_device *ndev = NULL;
 	int rc = 0;
 
-- 
cgit v1.2.3-59-g8ed1b


From e98bbaafcd1c47d30f3245517fb585f1aaaca4db Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:20 +0200
Subject: [S390] lockless idle time accounting

Replace the spinlock used in the idle time accounting with a sequence
counter mechanism analog to seqlock.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/cputime.h |  2 +-
 arch/s390/kernel/smp.c          | 28 +++++++++++++++++++---------
 arch/s390/kernel/vtime.c        | 22 +++++++++++++++-------
 3 files changed, 35 insertions(+), 17 deletions(-)

diff --git a/arch/s390/include/asm/cputime.h b/arch/s390/include/asm/cputime.h
index ec917d42ee6d..7a3817a656df 100644
--- a/arch/s390/include/asm/cputime.h
+++ b/arch/s390/include/asm/cputime.h
@@ -178,7 +178,7 @@ cputime64_to_clock_t(cputime64_t cputime)
 }
 
 struct s390_idle_data {
-	spinlock_t lock;
+	unsigned int sequence;
 	unsigned long long idle_count;
 	unsigned long long idle_enter;
 	unsigned long long idle_time;
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index fd8e3111a4e8..2270730f5354 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -856,13 +856,20 @@ static ssize_t show_idle_count(struct sys_device *dev,
 {
 	struct s390_idle_data *idle;
 	unsigned long long idle_count;
+	unsigned int sequence;
 
 	idle = &per_cpu(s390_idle, dev->id);
-	spin_lock(&idle->lock);
+repeat:
+	sequence = idle->sequence;
+	smp_rmb();
+	if (sequence & 1)
+		goto repeat;
 	idle_count = idle->idle_count;
 	if (idle->idle_enter)
 		idle_count++;
-	spin_unlock(&idle->lock);
+	smp_rmb();
+	if (idle->sequence != sequence)
+		goto repeat;
 	return sprintf(buf, "%llu\n", idle_count);
 }
 static SYSDEV_ATTR(idle_count, 0444, show_idle_count, NULL);
@@ -872,15 +879,22 @@ static ssize_t show_idle_time(struct sys_device *dev,
 {
 	struct s390_idle_data *idle;
 	unsigned long long now, idle_time, idle_enter;
+	unsigned int sequence;
 
 	idle = &per_cpu(s390_idle, dev->id);
-	spin_lock(&idle->lock);
 	now = get_clock();
+repeat:
+	sequence = idle->sequence;
+	smp_rmb();
+	if (sequence & 1)
+		goto repeat;
 	idle_time = idle->idle_time;
 	idle_enter = idle->idle_enter;
 	if (idle_enter != 0ULL && idle_enter < now)
 		idle_time += now - idle_enter;
-	spin_unlock(&idle->lock);
+	smp_rmb();
+	if (idle->sequence != sequence)
+		goto repeat;
 	return sprintf(buf, "%llu\n", idle_time >> 12);
 }
 static SYSDEV_ATTR(idle_time_us, 0444, show_idle_time, NULL);
@@ -908,11 +922,7 @@ static int __cpuinit smp_cpu_notify(struct notifier_block *self,
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
 		idle = &per_cpu(s390_idle, cpu);
-		spin_lock_irq(&idle->lock);
-		idle->idle_enter = 0;
-		idle->idle_time = 0;
-		idle->idle_count = 0;
-		spin_unlock_irq(&idle->lock);
+		memset(idle, 0, sizeof(struct s390_idle_data));
 		if (sysfs_create_group(&s->kobj, &cpu_online_attr_group))
 			return NOTIFY_BAD;
 		break;
diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c
index ade17e771f05..c41bb0d416e1 100644
--- a/arch/s390/kernel/vtime.c
+++ b/arch/s390/kernel/vtime.c
@@ -27,9 +27,7 @@
 
 static DEFINE_PER_CPU(struct vtimer_queue, virt_cpu_timer);
 
-DEFINE_PER_CPU(struct s390_idle_data, s390_idle) = {
-	.lock = __SPIN_LOCK_UNLOCKED(s390_idle.lock)
-};
+DEFINE_PER_CPU(struct s390_idle_data, s390_idle);
 
 static inline __u64 get_vtimer(void)
 {
@@ -151,11 +149,13 @@ void vtime_start_cpu(void)
 		vq->elapsed -= vq->idle - S390_lowcore.async_enter_timer;
 	}
 
-	spin_lock(&idle->lock);
+	idle->sequence++;
+	smp_wmb();
 	idle->idle_time += idle_time;
 	idle->idle_enter = 0ULL;
 	idle->idle_count++;
-	spin_unlock(&idle->lock);
+	smp_wmb();
+	idle->sequence++;
 }
 
 void vtime_stop_cpu(void)
@@ -242,15 +242,23 @@ cputime64_t s390_get_idle_time(int cpu)
 {
 	struct s390_idle_data *idle;
 	unsigned long long now, idle_time, idle_enter;
+	unsigned int sequence;
 
 	idle = &per_cpu(s390_idle, cpu);
-	spin_lock(&idle->lock);
+
 	now = get_clock();
+repeat:
+	sequence = idle->sequence;
+	smp_rmb();
+	if (sequence & 1)
+		goto repeat;
 	idle_time = 0;
 	idle_enter = idle->idle_enter;
 	if (idle_enter != 0ULL && idle_enter < now)
 		idle_time = now - idle_enter;
-	spin_unlock(&idle->lock);
+	smp_rmb();
+	if (idle->sequence != sequence)
+		goto repeat;
 	return idle_time;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 181d95229b0931ee2ce6aad7348079cbc10e8d05 Mon Sep 17 00:00:00 2001
From: Sebastian Ott <sebott@linux.vnet.ibm.com>
Date: Mon, 22 Jun 2009 12:08:21 +0200
Subject: [S390] dasd: fix refcounting in dasd_change_state

To set a dasd online dasd_change_state is called twice. The first
cycle will schedule initial analysis of the device, set the rc to
-EAGAIN and will not touch the device state any more.
The initial analysis will in turn call dasd_change_state to increase
the state to the final DASD_STATE_ONLINE.

If the dasd_change_state on the second thread outruns the other one
both finish with the state set to DASD_STATE_ONLINE and the device
refcount will be decreased by 2.

Fix this by leaving dasd_change_state on rc == -EAGAIN so that the
refcount will always be decreased by 1.

Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 drivers/s390/block/dasd.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index 99e71536213a..749836668655 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -470,7 +470,7 @@ static int dasd_decrease_state(struct dasd_device *device)
  */
 static void dasd_change_state(struct dasd_device *device)
 {
-        int rc;
+	int rc;
 
 	if (device->state == device->target)
 		/* Already where we want to go today... */
@@ -479,8 +479,10 @@ static void dasd_change_state(struct dasd_device *device)
 		rc = dasd_increase_state(device);
 	else
 		rc = dasd_decrease_state(device);
-        if (rc && rc != -EAGAIN)
-                device->target = device->state;
+	if (rc == -EAGAIN)
+		return;
+	if (rc)
+		device->target = device->state;
 
 	if (device->state == device->target) {
 		wake_up(&dasd_init_waitq);
-- 
cgit v1.2.3-59-g8ed1b


From 12310e9c1b9a53896e4df0459039dd125f62aa9b Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:22 +0200
Subject: [S390] Enable tick based perf_counter on s390.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/Kconfig                    | 1 +
 arch/s390/include/asm/perf_counter.h | 8 ++++++++
 tools/perf/perf.h                    | 6 ++++++
 3 files changed, 15 insertions(+)
 create mode 100644 arch/s390/include/asm/perf_counter.h

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index a14dba0e4d67..e577839f3073 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -94,6 +94,7 @@ config S390
 	select HAVE_KVM if 64BIT
 	select HAVE_ARCH_TRACEHOOK
 	select INIT_ALL_POSSIBLE
+	select HAVE_PERF_COUNTERS
 
 source "init/Kconfig"
 
diff --git a/arch/s390/include/asm/perf_counter.h b/arch/s390/include/asm/perf_counter.h
new file mode 100644
index 000000000000..a7205a3828cb
--- /dev/null
+++ b/arch/s390/include/asm/perf_counter.h
@@ -0,0 +1,8 @@
+/*
+ * Performance counter support - s390 specific definitions.
+ *
+ * Copyright 2009 Martin Schwidefsky, IBM Corporation.
+ */
+
+static inline void set_perf_counter_pending(void) {}
+static inline void clear_perf_counter_pending(void) {}
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index bccb529dac08..ceb68aa51f7f 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -13,6 +13,12 @@
 #define cpu_relax()	asm volatile ("" ::: "memory");
 #endif
 
+#ifdef __s390__
+#include "../../arch/s390/include/asm/unistd.h"
+#define rmb()		asm volatile("bcr 15,0" ::: "memory")
+#define cpu_relax()	asm volatile("" ::: "memory");
+#endif
+
 #include <time.h>
 #include <unistd.h>
 #include <sys/types.h>
-- 
cgit v1.2.3-59-g8ed1b


From acf018004f76617dbab36ef4b5480d4351f9cdff Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:23 +0200
Subject: [S390] kprobes: defer setting of ctlblk state

get_krobe_ctlblk returns a per cpu kprobe control block which holds
the state of the current cpu wrt to kprobe.
When inserting/removing a kprobe the state of the cpu which replaces
the code is changed to KPROBE_SWAP_INST. This however is done when
preemption is still enabled. So the state of the current cpu doesn't
necessarily reflect the real state.
To fix this move the code that changes the state to non-preemptible
context.

Reported-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/kprobes.c | 18 +++++++-----------
 1 file changed, 7 insertions(+), 11 deletions(-)

diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c
index 9bb2f6241d9f..86783efa24ee 100644
--- a/arch/s390/kernel/kprobes.c
+++ b/arch/s390/kernel/kprobes.c
@@ -154,39 +154,35 @@ void __kprobes get_instruction_type(struct arch_specific_insn *ainsn)
 
 static int __kprobes swap_instruction(void *aref)
 {
+	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+	unsigned long status = kcb->kprobe_status;
 	struct ins_replace_args *args = aref;
+	int rc;
 
-	return probe_kernel_write(args->ptr, &args->new, sizeof(args->new));
+	kcb->kprobe_status = KPROBE_SWAP_INST;
+	rc = probe_kernel_write(args->ptr, &args->new, sizeof(args->new));
+	kcb->kprobe_status = status;
+	return rc;
 }
 
 void __kprobes arch_arm_kprobe(struct kprobe *p)
 {
-	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-	unsigned long status = kcb->kprobe_status;
 	struct ins_replace_args args;
 
 	args.ptr = p->addr;
 	args.old = p->opcode;
 	args.new = BREAKPOINT_INSTRUCTION;
-
-	kcb->kprobe_status = KPROBE_SWAP_INST;
 	stop_machine(swap_instruction, &args, NULL);
-	kcb->kprobe_status = status;
 }
 
 void __kprobes arch_disarm_kprobe(struct kprobe *p)
 {
-	struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
-	unsigned long status = kcb->kprobe_status;
 	struct ins_replace_args args;
 
 	args.ptr = p->addr;
 	args.old = BREAKPOINT_INSTRUCTION;
 	args.new = p->opcode;
-
-	kcb->kprobe_status = KPROBE_SWAP_INST;
 	stop_machine(swap_instruction, &args, NULL);
-	kcb->kprobe_status = status;
 }
 
 void __kprobes arch_remove_kprobe(struct kprobe *p)
-- 
cgit v1.2.3-59-g8ed1b


From da6330fccc251db73945ee3eb6248985cf2574de Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Mon, 22 Jun 2009 12:08:24 +0200
Subject: [S390] Update default configuration.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/defconfig | 72 +++++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 54 insertions(+), 18 deletions(-)

diff --git a/arch/s390/defconfig b/arch/s390/defconfig
index d401d56c255f..fcba206529f3 100644
--- a/arch/s390/defconfig
+++ b/arch/s390/defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.30-rc3
-# Thu Apr 23 09:29:52 2009
+# Linux kernel version: 2.6.30
+# Mon Jun 22 11:08:16 2009
 #
 CONFIG_SCHED_MC=y
 CONFIG_MMU=y
@@ -25,6 +25,7 @@ CONFIG_VIRT_CPU_ACCOUNTING=y
 CONFIG_ARCH_SUPPORTS_DEBUG_PAGEALLOC=y
 CONFIG_S390=y
 CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+CONFIG_CONSTRUCTORS=y
 
 #
 # General setup
@@ -90,7 +91,6 @@ CONFIG_SYSCTL_SYSCALL=y
 CONFIG_KALLSYMS=y
 # CONFIG_KALLSYMS_ALL is not set
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
-# CONFIG_STRIP_ASM_SYMS is not set
 CONFIG_HOTPLUG=y
 CONFIG_PRINTK=y
 CONFIG_BUG=y
@@ -103,7 +103,14 @@ CONFIG_TIMERFD=y
 CONFIG_EVENTFD=y
 CONFIG_SHMEM=y
 CONFIG_AIO=y
+CONFIG_HAVE_PERF_COUNTERS=y
+
+#
+# Performance Counters
+#
+# CONFIG_PERF_COUNTERS is not set
 CONFIG_VM_EVENT_COUNTERS=y
+# CONFIG_STRIP_ASM_SYMS is not set
 # CONFIG_COMPAT_BRK is not set
 CONFIG_SLAB=y
 # CONFIG_SLUB is not set
@@ -119,6 +126,11 @@ CONFIG_HAVE_KRETPROBES=y
 CONFIG_HAVE_ARCH_TRACEHOOK=y
 CONFIG_USE_GENERIC_SMP_HELPERS=y
 CONFIG_HAVE_DEFAULT_NO_SPIN_MUTEXES=y
+
+#
+# GCOV-based kernel profiling
+#
+# CONFIG_GCOV_KERNEL is not set
 # CONFIG_SLOW_WORK is not set
 # CONFIG_HAVE_GENERIC_DMA_COHERENT is not set
 CONFIG_SLABINFO=y
@@ -150,7 +162,7 @@ CONFIG_DEFAULT_DEADLINE=y
 # CONFIG_DEFAULT_NOOP is not set
 CONFIG_DEFAULT_IOSCHED="deadline"
 CONFIG_PREEMPT_NOTIFIERS=y
-# CONFIG_FREEZER is not set
+CONFIG_FREEZER=y
 
 #
 # Base setup
@@ -199,6 +211,7 @@ CONFIG_ARCH_SPARSEMEM_DEFAULT=y
 CONFIG_ARCH_SELECT_MEMORY_MODEL=y
 CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y
 CONFIG_ARCH_ENABLE_MEMORY_HOTREMOVE=y
+CONFIG_ARCH_HIBERNATION_POSSIBLE=y
 CONFIG_SELECT_MEMORY_MODEL=y
 # CONFIG_FLATMEM_MANUAL is not set
 # CONFIG_DISCONTIGMEM_MANUAL is not set
@@ -218,9 +231,9 @@ CONFIG_PHYS_ADDR_T_64BIT=y
 CONFIG_ZONE_DMA_FLAG=1
 CONFIG_BOUNCE=y
 CONFIG_VIRT_TO_BUS=y
-CONFIG_UNEVICTABLE_LRU=y
 CONFIG_HAVE_MLOCK=y
 CONFIG_HAVE_MLOCKED_PAGE_BIT=y
+CONFIG_DEFAULT_MMAP_MIN_ADDR=4096
 
 #
 # I/O subsystem configuration
@@ -257,6 +270,16 @@ CONFIG_KEXEC=y
 # CONFIG_ZFCPDUMP is not set
 CONFIG_S390_GUEST=y
 CONFIG_SECCOMP=y
+
+#
+# Power Management
+#
+CONFIG_PM=y
+# CONFIG_PM_DEBUG is not set
+CONFIG_PM_SLEEP_SMP=y
+CONFIG_PM_SLEEP=y
+CONFIG_HIBERNATION=y
+CONFIG_PM_STD_PARTITION=""
 CONFIG_NET=y
 
 #
@@ -384,6 +407,7 @@ CONFIG_SCTP_HMAC_MD5=y
 # CONFIG_ECONET is not set
 # CONFIG_WAN_ROUTER is not set
 # CONFIG_PHONET is not set
+# CONFIG_IEEE802154 is not set
 CONFIG_NET_SCHED=y
 
 #
@@ -446,6 +470,7 @@ CONFIG_CAN_BCM=m
 # CAN Device Drivers
 #
 CONFIG_CAN_VCAN=m
+# CONFIG_CAN_DEV is not set
 # CONFIG_CAN_DEBUG_DEVICES is not set
 # CONFIG_AF_RXRPC is not set
 # CONFIG_WIMAX is not set
@@ -524,10 +549,6 @@ CONFIG_BLK_DEV_SR=y
 CONFIG_BLK_DEV_SR_VENDOR=y
 CONFIG_CHR_DEV_SG=y
 # CONFIG_CHR_DEV_SCH is not set
-
-#
-# Some SCSI devices (e.g. CD jukebox) support multiple LUNs
-#
 CONFIG_SCSI_MULTI_LUN=y
 CONFIG_SCSI_CONSTANTS=y
 CONFIG_SCSI_LOGGING=y
@@ -578,7 +599,6 @@ CONFIG_DM_MULTIPATH=m
 # CONFIG_DM_DELAY is not set
 # CONFIG_DM_UEVENT is not set
 CONFIG_NETDEVICES=y
-CONFIG_COMPAT_NET_DEV_OPS=y
 # CONFIG_IFB is not set
 CONFIG_DUMMY=m
 CONFIG_BONDING=m
@@ -595,6 +615,7 @@ CONFIG_NET_ETHERNET=y
 # CONFIG_IBM_NEW_EMAC_NO_FLOW_CTRL is not set
 # CONFIG_IBM_NEW_EMAC_MAL_CLR_ICINTSTAT is not set
 # CONFIG_IBM_NEW_EMAC_MAL_COMMON_ERR is not set
+# CONFIG_KS8842 is not set
 CONFIG_NETDEV_1000=y
 CONFIG_NETDEV_10000=y
 # CONFIG_TR is not set
@@ -674,6 +695,11 @@ CONFIG_S390_TAPE_34XX=m
 # CONFIG_MONREADER is not set
 CONFIG_MONWRITER=m
 CONFIG_S390_VMUR=m
+
+#
+# PPS support
+#
+# CONFIG_PPS is not set
 # CONFIG_POWER_SUPPLY is not set
 # CONFIG_THERMAL is not set
 # CONFIG_THERMAL_HWMON is not set
@@ -683,6 +709,10 @@ CONFIG_S390_VMUR=m
 # CONFIG_NEW_LEDS is not set
 CONFIG_ACCESSIBILITY=y
 # CONFIG_AUXDISPLAY is not set
+
+#
+# TI VLYNQ
+#
 # CONFIG_STAGING is not set
 
 #
@@ -703,11 +733,12 @@ CONFIG_FS_MBCACHE=y
 # CONFIG_REISERFS_FS is not set
 # CONFIG_JFS_FS is not set
 CONFIG_FS_POSIX_ACL=y
-CONFIG_FILE_LOCKING=y
 # CONFIG_XFS_FS is not set
 # CONFIG_GFS2_FS is not set
 # CONFIG_OCFS2_FS is not set
 # CONFIG_BTRFS_FS is not set
+CONFIG_FILE_LOCKING=y
+CONFIG_FSNOTIFY=y
 CONFIG_DNOTIFY=y
 CONFIG_INOTIFY=y
 CONFIG_INOTIFY_USER=y
@@ -865,19 +896,23 @@ CONFIG_DEBUG_MEMORY_INIT=y
 CONFIG_SYSCTL_SYSCALL_CHECK=y
 # CONFIG_DEBUG_PAGEALLOC is not set
 CONFIG_HAVE_FUNCTION_TRACER=y
+CONFIG_HAVE_FUNCTION_GRAPH_TRACER=y
+CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST=y
+CONFIG_HAVE_DYNAMIC_FTRACE=y
+CONFIG_HAVE_FTRACE_MCOUNT_RECORD=y
+CONFIG_HAVE_FTRACE_SYSCALLS=y
 CONFIG_TRACING_SUPPORT=y
-
-#
-# Tracers
-#
+CONFIG_FTRACE=y
 # CONFIG_FUNCTION_TRACER is not set
 # CONFIG_IRQSOFF_TRACER is not set
 # CONFIG_PREEMPT_TRACER is not set
 # CONFIG_SCHED_TRACER is not set
-# CONFIG_CONTEXT_SWITCH_TRACER is not set
-# CONFIG_EVENT_TRACER is not set
+# CONFIG_ENABLE_DEFAULT_TRACERS is not set
+# CONFIG_FTRACE_SYSCALLS is not set
 # CONFIG_BOOT_TRACER is not set
-# CONFIG_TRACE_BRANCH_PROFILING is not set
+CONFIG_BRANCH_PROFILE_NONE=y
+# CONFIG_PROFILE_ANNOTATED_BRANCHES is not set
+# CONFIG_PROFILE_ALL_BRANCHES is not set
 # CONFIG_STACK_TRACER is not set
 # CONFIG_KMEMTRACE is not set
 # CONFIG_WORKQUEUE_TRACER is not set
@@ -886,6 +921,7 @@ CONFIG_TRACING_SUPPORT=y
 CONFIG_SAMPLES=y
 # CONFIG_SAMPLE_KOBJECT is not set
 # CONFIG_SAMPLE_KPROBES is not set
+# CONFIG_KMEMCHECK is not set
 
 #
 # Security options
-- 
cgit v1.2.3-59-g8ed1b


From 51e268423151fc7bb41945bde7843160b6a14c32 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Mon, 22 Jun 2009 16:43:14 +0530
Subject: perf_counter tools: Define separate declarations for H/W and S/W
 events

Define separate declarations for H/W and S/W events to:

 1. Shorten name to save some space so that we can add more members
 2. Fix alignment
 3. Avoid declaring HARDWARE/SOFTWARE again and again.

Removed unused CR(x, y)

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <1245669194.17153.6.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/util/parse-events.c | 44 +++++++++++++++++++++---------------------
 1 file changed, 22 insertions(+), 22 deletions(-)

diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 35d04da38d6a..12abab3a0d63 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -18,30 +18,30 @@ struct event_symbol {
 	char	*symbol;
 };
 
-#define C(x, y) .type = PERF_TYPE_##x, .config = PERF_COUNT_##y
-#define CR(x, y) .type = PERF_TYPE_##x, .config = y
+#define CHW(x) .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_##x
+#define CSW(x) .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_##x
 
 static struct event_symbol event_symbols[] = {
-  { C(HARDWARE, HW_CPU_CYCLES),		"cpu-cycles",		},
-  { C(HARDWARE, HW_CPU_CYCLES),		"cycles",		},
-  { C(HARDWARE, HW_INSTRUCTIONS),	"instructions",		},
-  { C(HARDWARE, HW_CACHE_REFERENCES),	"cache-references",	},
-  { C(HARDWARE, HW_CACHE_MISSES),	"cache-misses",		},
-  { C(HARDWARE, HW_BRANCH_INSTRUCTIONS),"branch-instructions",	},
-  { C(HARDWARE, HW_BRANCH_INSTRUCTIONS),"branches",		},
-  { C(HARDWARE, HW_BRANCH_MISSES),	"branch-misses",	},
-  { C(HARDWARE, HW_BUS_CYCLES),		"bus-cycles",		},
-
-  { C(SOFTWARE, SW_CPU_CLOCK),		"cpu-clock",		},
-  { C(SOFTWARE, SW_TASK_CLOCK),		"task-clock",		},
-  { C(SOFTWARE, SW_PAGE_FAULTS),	"page-faults",		},
-  { C(SOFTWARE, SW_PAGE_FAULTS),	"faults",		},
-  { C(SOFTWARE, SW_PAGE_FAULTS_MIN),	"minor-faults",		},
-  { C(SOFTWARE, SW_PAGE_FAULTS_MAJ),	"major-faults",		},
-  { C(SOFTWARE, SW_CONTEXT_SWITCHES),	"context-switches",	},
-  { C(SOFTWARE, SW_CONTEXT_SWITCHES),	"cs",			},
-  { C(SOFTWARE, SW_CPU_MIGRATIONS),	"cpu-migrations",	},
-  { C(SOFTWARE, SW_CPU_MIGRATIONS),	"migrations",		},
+  { CHW(CPU_CYCLES),		"cpu-cycles",		},
+  { CHW(CPU_CYCLES),		"cycles",		},
+  { CHW(INSTRUCTIONS),		"instructions",		},
+  { CHW(CACHE_REFERENCES),	"cache-references",	},
+  { CHW(CACHE_MISSES),		"cache-misses",		},
+  { CHW(BRANCH_INSTRUCTIONS),	"branch-instructions",	},
+  { CHW(BRANCH_INSTRUCTIONS),	"branches",		},
+  { CHW(BRANCH_MISSES),		"branch-misses",	},
+  { CHW(BUS_CYCLES),		"bus-cycles",		},
+
+  { CSW(CPU_CLOCK),		"cpu-clock",		},
+  { CSW(TASK_CLOCK),		"task-clock",		},
+  { CSW(PAGE_FAULTS),		"page-faults",		},
+  { CSW(PAGE_FAULTS),		"faults",		},
+  { CSW(PAGE_FAULTS_MIN),	"minor-faults",		},
+  { CSW(PAGE_FAULTS_MAJ),	"major-faults",		},
+  { CSW(CONTEXT_SWITCHES),	"context-switches",	},
+  { CSW(CONTEXT_SWITCHES),	"cs",			},
+  { CSW(CPU_MIGRATIONS),	"cpu-migrations",	},
+  { CSW(CPU_MIGRATIONS),	"migrations",		},
 };
 
 #define __PERF_COUNTER_FIELD(config, name) \
-- 
cgit v1.2.3-59-g8ed1b


From 74d5b5889ea71a95d8924c08f8a7c6e2bdcbc0ba Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Mon, 22 Jun 2009 16:44:28 +0530
Subject: perf_counter tools: Introduce alias member in event_symbol

By introducing alias member in event_symbol :

1. duplicate lines are removed, like:
   cpu-cycles and cycles
   branch-instructions and branches
   context-switches and cs
   cpu-migrations and migrations

2. We can also add alias for another events.

Now ./perf list looks like :

List of pre-defined events (to be used in -e):

  cpu-cycles OR cycles                     [Hardware event]
  instructions                             [Hardware event]
  cache-references                         [Hardware event]
  cache-misses                             [Hardware event]
  branch-instructions OR branches          [Hardware event]
  branch-misses                            [Hardware event]
  bus-cycles                               [Hardware event]

  cpu-clock                                [Software event]
  task-clock                               [Software event]
  page-faults                              [Software event]
  faults                                   [Software event]
  minor-faults                             [Software event]
  major-faults                             [Software event]
  context-switches OR cs                   [Software event]
  cpu-migrations OR migrations             [Software event]

  rNNN                                     [raw hardware event descriptor]

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <1245669268.17153.8.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/util/parse-events.c | 63 +++++++++++++++++++++++++-----------------
 1 file changed, 38 insertions(+), 25 deletions(-)

diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 12abab3a0d63..f5695486ad3f 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -16,32 +16,29 @@ struct event_symbol {
 	u8	type;
 	u64	config;
 	char	*symbol;
+	char	*alias;
 };
 
 #define CHW(x) .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_##x
 #define CSW(x) .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_##x
 
 static struct event_symbol event_symbols[] = {
-  { CHW(CPU_CYCLES),		"cpu-cycles",		},
-  { CHW(CPU_CYCLES),		"cycles",		},
-  { CHW(INSTRUCTIONS),		"instructions",		},
-  { CHW(CACHE_REFERENCES),	"cache-references",	},
-  { CHW(CACHE_MISSES),		"cache-misses",		},
-  { CHW(BRANCH_INSTRUCTIONS),	"branch-instructions",	},
-  { CHW(BRANCH_INSTRUCTIONS),	"branches",		},
-  { CHW(BRANCH_MISSES),		"branch-misses",	},
-  { CHW(BUS_CYCLES),		"bus-cycles",		},
-
-  { CSW(CPU_CLOCK),		"cpu-clock",		},
-  { CSW(TASK_CLOCK),		"task-clock",		},
-  { CSW(PAGE_FAULTS),		"page-faults",		},
-  { CSW(PAGE_FAULTS),		"faults",		},
-  { CSW(PAGE_FAULTS_MIN),	"minor-faults",		},
-  { CSW(PAGE_FAULTS_MAJ),	"major-faults",		},
-  { CSW(CONTEXT_SWITCHES),	"context-switches",	},
-  { CSW(CONTEXT_SWITCHES),	"cs",			},
-  { CSW(CPU_MIGRATIONS),	"cpu-migrations",	},
-  { CSW(CPU_MIGRATIONS),	"migrations",		},
+  { CHW(CPU_CYCLES),		"cpu-cycles",		"cycles"	},
+  { CHW(INSTRUCTIONS),		"instructions",		""		},
+  { CHW(CACHE_REFERENCES),	"cache-references",	""		},
+  { CHW(CACHE_MISSES),		"cache-misses",		""		},
+  { CHW(BRANCH_INSTRUCTIONS),	"branch-instructions",	"branches"	},
+  { CHW(BRANCH_MISSES),		"branch-misses",	""		},
+  { CHW(BUS_CYCLES),		"bus-cycles",		""		},
+
+  { CSW(CPU_CLOCK),		"cpu-clock",		""		},
+  { CSW(TASK_CLOCK),		"task-clock",		""		},
+  { CSW(PAGE_FAULTS),		"page-faults",		""		},
+  { CSW(PAGE_FAULTS),		"faults",		""		},
+  { CSW(PAGE_FAULTS_MIN),	"minor-faults",		""		},
+  { CSW(PAGE_FAULTS_MAJ),	"major-faults",		""		},
+  { CSW(CONTEXT_SWITCHES),	"context-switches",	"cs"		},
+  { CSW(CPU_MIGRATIONS),	"cpu-migrations",	"migrations"	},
 };
 
 #define __PERF_COUNTER_FIELD(config, name) \
@@ -196,6 +193,19 @@ static int parse_generic_hw_symbols(const char *str, struct perf_counter_attr *a
 	return 0;
 }
 
+static int check_events(const char *str, unsigned int i)
+{
+	if (!strncmp(str, event_symbols[i].symbol,
+		     strlen(event_symbols[i].symbol)))
+		return 1;
+
+	if (strlen(event_symbols[i].alias))
+		if (!strncmp(str, event_symbols[i].alias,
+		      strlen(event_symbols[i].alias)))
+			return 1;
+	return 0;
+}
+
 /*
  * Each event can have multiple symbolic names.
  * Symbolic names are (almost) exactly matched.
@@ -235,9 +245,7 @@ static int parse_event_symbols(const char *str, struct perf_counter_attr *attr)
 	}
 
 	for (i = 0; i < ARRAY_SIZE(event_symbols); i++) {
-		if (!strncmp(str, event_symbols[i].symbol,
-			     strlen(event_symbols[i].symbol))) {
-
+		if (check_events(str, i)) {
 			attr->type = event_symbols[i].type;
 			attr->config = event_symbols[i].config;
 
@@ -289,6 +297,7 @@ void print_events(void)
 {
 	struct event_symbol *syms = event_symbols;
 	unsigned int i, type, prev_type = -1;
+	char name[40];
 
 	fprintf(stderr, "\n");
 	fprintf(stderr, "List of pre-defined events (to be used in -e):\n");
@@ -301,14 +310,18 @@ void print_events(void)
 		if (type != prev_type)
 			fprintf(stderr, "\n");
 
-		fprintf(stderr, "  %-30s [%s]\n", syms->symbol,
+		if (strlen(syms->alias))
+			sprintf(name, "%s OR %s", syms->symbol, syms->alias);
+		else
+			strcpy(name, syms->symbol);
+		fprintf(stderr, "  %-40s [%s]\n", name,
 			event_type_descriptors[type]);
 
 		prev_type = type;
 	}
 
 	fprintf(stderr, "\n");
-	fprintf(stderr, "  %-30s [raw hardware event descriptor]\n",
+	fprintf(stderr, "  %-40s [raw hardware event descriptor]\n",
 		"rNNN");
 	fprintf(stderr, "\n");
 
-- 
cgit v1.2.3-59-g8ed1b


From 8cc20198cfccd06cef705c14fd50bde603e2e306 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Mon, 22 Jun 2009 14:13:55 +0200
Subject: netfilter: nf_conntrack: death_by_timeout() fix

death_by_timeout() might delete a conntrack from hash list
and insert it in dying list.

 nf_ct_delete_from_lists(ct);
 nf_ct_insert_dying_list(ct);

I believe a (lockless) reader could *catch* ct while doing a lookup
and miss the end of its chain.
(nulls lookup algo must check the null value at the end of lookup and
should restart if the null value is not the expected one.
cf Documentation/RCU/rculist_nulls.txt for details)

We need to change nf_conntrack_init_net() and use a different "null" value,
guaranteed not being used in regular lists. Choose very large values, since
hash table uses [0..size-1] null values.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 net/netfilter/nf_conntrack_core.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 5f72b94b4918..5276a2dd56fe 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -1267,13 +1267,19 @@ err_cache:
 	return ret;
 }
 
+/*
+ * We need to use special "null" values, not used in hash table
+ */
+#define UNCONFIRMED_NULLS_VAL	((1<<30)+0)
+#define DYING_NULLS_VAL		((1<<30)+1)
+
 static int nf_conntrack_init_net(struct net *net)
 {
 	int ret;
 
 	atomic_set(&net->ct.count, 0);
-	INIT_HLIST_NULLS_HEAD(&net->ct.unconfirmed, 0);
-	INIT_HLIST_NULLS_HEAD(&net->ct.dying, 0);
+	INIT_HLIST_NULLS_HEAD(&net->ct.unconfirmed, UNCONFIRMED_NULLS_VAL);
+	INIT_HLIST_NULLS_HEAD(&net->ct.dying, DYING_NULLS_VAL);
 	net->ct.stat = alloc_percpu(struct ip_conntrack_stat);
 	if (!net->ct.stat) {
 		ret = -ENOMEM;
-- 
cgit v1.2.3-59-g8ed1b


From 5c8ec910e789a92229978d8fd1fce7b62e8ac711 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 22 Jun 2009 14:14:16 +0200
Subject: netfilter: nf_conntrack: fix confirmation race condition

New connection tracking entries are inserted into the hash before they
are fully set up, namely the CONFIRMED bit is not set and the timer not
started yet. This can theoretically lead to a race with timer, which
would set the timeout value to a relative value, most likely already in
the past.

Perform hash insertion as the final step to fix this.

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 net/netfilter/nf_conntrack_core.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 5276a2dd56fe..b0b06c7a9483 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -425,7 +425,6 @@ __nf_conntrack_confirm(struct sk_buff *skb)
 	/* Remove from unconfirmed list */
 	hlist_nulls_del_rcu(&ct->tuplehash[IP_CT_DIR_ORIGINAL].hnnode);
 
-	__nf_conntrack_hash_insert(ct, hash, repl_hash);
 	/* Timer relative to confirmation time, not original
 	   setting time, otherwise we'd get timer wrap in
 	   weird delay cases. */
@@ -433,8 +432,16 @@ __nf_conntrack_confirm(struct sk_buff *skb)
 	add_timer(&ct->timeout);
 	atomic_inc(&ct->ct_general.use);
 	set_bit(IPS_CONFIRMED_BIT, &ct->status);
+
+	/* Since the lookup is lockless, hash insertion must be done after
+	 * starting the timer and setting the CONFIRMED bit. The RCU barriers
+	 * guarantee that no other CPU can find the conntrack before the above
+	 * stores are visible.
+	 */
+	__nf_conntrack_hash_insert(ct, hash, repl_hash);
 	NF_CT_STAT_INC(net, insert);
 	spin_unlock_bh(&nf_conntrack_lock);
+
 	help = nfct_help(ct);
 	if (help && help->helper)
 		nf_conntrack_event_cache(IPCT_HELPER, ct);
-- 
cgit v1.2.3-59-g8ed1b


From 8d8890b7751387f58ce0a6428773de2fbc0fd596 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 22 Jun 2009 14:14:41 +0200
Subject: netfilter: nf_conntrack: fix conntrack lookup race

The RCU protected conntrack hash lookup only checks whether the entry
has a refcount of zero to decide whether it is stale. This is not
sufficient, entries are explicitly removed while there is at least
one reference left, possibly more. Explicitly check whether the entry
has been marked as dying to fix this.

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 net/netfilter/nf_conntrack_core.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index b0b06c7a9483..7508f11c5b39 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -335,7 +335,8 @@ begin:
 	h = __nf_conntrack_find(net, tuple);
 	if (h) {
 		ct = nf_ct_tuplehash_to_ctrack(h);
-		if (unlikely(!atomic_inc_not_zero(&ct->ct_general.use)))
+		if (unlikely(nf_ct_is_dying(ct) ||
+			     !atomic_inc_not_zero(&ct->ct_general.use)))
 			h = NULL;
 		else {
 			if (unlikely(!nf_ct_tuple_equal(tuple, &h->tuple))) {
@@ -510,7 +511,8 @@ static noinline int early_drop(struct net *net, unsigned int hash)
 			cnt++;
 		}
 
-		if (ct && unlikely(!atomic_inc_not_zero(&ct->ct_general.use)))
+		if (ct && unlikely(nf_ct_is_dying(ct) ||
+				   !atomic_inc_not_zero(&ct->ct_general.use)))
 			ct = NULL;
 		if (ct || cnt >= NF_CT_EVICTION_RANGE)
 			break;
-- 
cgit v1.2.3-59-g8ed1b


From f9ffc31251c2caa11962c9b74ce650e2167fa8d1 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 22 Jun 2009 14:15:02 +0200
Subject: netfilter: fix some sparse endianess warnings

net/netfilter/xt_NFQUEUE.c:46:9: warning: incorrect type in assignment (different base types)
net/netfilter/xt_NFQUEUE.c:46:9:    expected unsigned int [unsigned] [usertype] ipaddr
net/netfilter/xt_NFQUEUE.c:46:9:    got restricted unsigned int
net/netfilter/xt_NFQUEUE.c:68:10: warning: incorrect type in assignment (different base types)
net/netfilter/xt_NFQUEUE.c:68:10:    expected unsigned int [unsigned] <noident>
net/netfilter/xt_NFQUEUE.c:68:10:    got restricted unsigned int
net/netfilter/xt_NFQUEUE.c:69:10: warning: incorrect type in assignment (different base types)
net/netfilter/xt_NFQUEUE.c:69:10:    expected unsigned int [unsigned] <noident>
net/netfilter/xt_NFQUEUE.c:69:10:    got restricted unsigned int
net/netfilter/xt_NFQUEUE.c:70:10: warning: incorrect type in assignment (different base types)
net/netfilter/xt_NFQUEUE.c:70:10:    expected unsigned int [unsigned] <noident>
net/netfilter/xt_NFQUEUE.c:70:10:    got restricted unsigned int
net/netfilter/xt_NFQUEUE.c:71:10: warning: incorrect type in assignment (different base types)
net/netfilter/xt_NFQUEUE.c:71:10:    expected unsigned int [unsigned] <noident>
net/netfilter/xt_NFQUEUE.c:71:10:    got restricted unsigned int

net/netfilter/xt_cluster.c:20:55: warning: incorrect type in return expression (different base types)
net/netfilter/xt_cluster.c:20:55:    expected unsigned int
net/netfilter/xt_cluster.c:20:55:    got restricted unsigned int const [usertype] ip
net/netfilter/xt_cluster.c:20:55: warning: incorrect type in return expression (different base types)
net/netfilter/xt_cluster.c:20:55:    expected unsigned int
net/netfilter/xt_cluster.c:20:55:    got restricted unsigned int const [usertype] ip

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 net/netfilter/xt_NFQUEUE.c | 8 ++++----
 net/netfilter/xt_cluster.c | 8 ++++----
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/net/netfilter/xt_NFQUEUE.c b/net/netfilter/xt_NFQUEUE.c
index 498b45101df7..f28f6a5fc02d 100644
--- a/net/netfilter/xt_NFQUEUE.c
+++ b/net/netfilter/xt_NFQUEUE.c
@@ -40,12 +40,12 @@ nfqueue_tg(struct sk_buff *skb, const struct xt_target_param *par)
 static u32 hash_v4(const struct sk_buff *skb)
 {
 	const struct iphdr *iph = ip_hdr(skb);
-	u32 ipaddr;
+	__be32 ipaddr;
 
 	/* packets in either direction go into same queue */
 	ipaddr = iph->saddr ^ iph->daddr;
 
-	return jhash_2words(ipaddr, iph->protocol, jhash_initval);
+	return jhash_2words((__force u32)ipaddr, iph->protocol, jhash_initval);
 }
 
 static unsigned int
@@ -63,14 +63,14 @@ nfqueue_tg4_v1(struct sk_buff *skb, const struct xt_target_param *par)
 static u32 hash_v6(const struct sk_buff *skb)
 {
 	const struct ipv6hdr *ip6h = ipv6_hdr(skb);
-	u32 addr[4];
+	__be32 addr[4];
 
 	addr[0] = ip6h->saddr.s6_addr32[0] ^ ip6h->daddr.s6_addr32[0];
 	addr[1] = ip6h->saddr.s6_addr32[1] ^ ip6h->daddr.s6_addr32[1];
 	addr[2] = ip6h->saddr.s6_addr32[2] ^ ip6h->daddr.s6_addr32[2];
 	addr[3] = ip6h->saddr.s6_addr32[3] ^ ip6h->daddr.s6_addr32[3];
 
-	return jhash2(addr, ARRAY_SIZE(addr), jhash_initval);
+	return jhash2((__force u32 *)addr, ARRAY_SIZE(addr), jhash_initval);
 }
 
 static unsigned int
diff --git a/net/netfilter/xt_cluster.c b/net/netfilter/xt_cluster.c
index 69a639f35403..225ee3ecd69d 100644
--- a/net/netfilter/xt_cluster.c
+++ b/net/netfilter/xt_cluster.c
@@ -15,14 +15,14 @@
 #include <net/netfilter/nf_conntrack.h>
 #include <linux/netfilter/xt_cluster.h>
 
-static inline u_int32_t nf_ct_orig_ipv4_src(const struct nf_conn *ct)
+static inline u32 nf_ct_orig_ipv4_src(const struct nf_conn *ct)
 {
-	return ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip;
+	return (__force u32)ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip;
 }
 
-static inline const void *nf_ct_orig_ipv6_src(const struct nf_conn *ct)
+static inline const u32 *nf_ct_orig_ipv6_src(const struct nf_conn *ct)
 {
-	return ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip6;
+	return (__force u32 *)ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip6;
 }
 
 static inline u_int32_t
-- 
cgit v1.2.3-59-g8ed1b


From 249556192859490b6280552d4b877064f9f5ee48 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 22 Jun 2009 14:15:30 +0200
Subject: netfilter: nf_log: fix direct userspace memory access in proc handler

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 net/netfilter/nf_log.c | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c
index 2fefe147750a..4e620305f28c 100644
--- a/net/netfilter/nf_log.c
+++ b/net/netfilter/nf_log.c
@@ -47,7 +47,6 @@ int nf_log_register(u_int8_t pf, struct nf_logger *logger)
 	mutex_lock(&nf_log_mutex);
 
 	if (pf == NFPROTO_UNSPEC) {
-		int i;
 		for (i = NFPROTO_UNSPEC; i < NFPROTO_NUMPROTO; i++)
 			list_add_tail(&(logger->list[i]), &(nf_loggers_l[i]));
 	} else {
@@ -216,7 +215,7 @@ static const struct file_operations nflog_file_ops = {
 #endif /* PROC_FS */
 
 #ifdef CONFIG_SYSCTL
-struct ctl_path nf_log_sysctl_path[] = {
+static struct ctl_path nf_log_sysctl_path[] = {
 	{ .procname = "net", .ctl_name = CTL_NET, },
 	{ .procname = "netfilter", .ctl_name = NET_NETFILTER, },
 	{ .procname = "nf_log", .ctl_name = CTL_UNNUMBERED, },
@@ -228,19 +227,26 @@ static struct ctl_table nf_log_sysctl_table[NFPROTO_NUMPROTO+1];
 static struct ctl_table_header *nf_log_dir_header;
 
 static int nf_log_proc_dostring(ctl_table *table, int write, struct file *filp,
-			 void *buffer, size_t *lenp, loff_t *ppos)
+			 void __user *buffer, size_t *lenp, loff_t *ppos)
 {
 	const struct nf_logger *logger;
+	char buf[NFLOGGER_NAME_LEN];
+	size_t size = *lenp;
 	int r = 0;
 	int tindex = (unsigned long)table->extra1;
 
 	if (write) {
-		if (!strcmp(buffer, "NONE")) {
+		if (size > sizeof(buf))
+			size = sizeof(buf);
+		if (copy_from_user(buf, buffer, size))
+			return -EFAULT;
+
+		if (!strcmp(buf, "NONE")) {
 			nf_log_unbind_pf(tindex);
 			return 0;
 		}
 		mutex_lock(&nf_log_mutex);
-		logger = __find_logger(tindex, buffer);
+		logger = __find_logger(tindex, buf);
 		if (logger == NULL) {
 			mutex_unlock(&nf_log_mutex);
 			return -ENOENT;
-- 
cgit v1.2.3-59-g8ed1b


From 6d62182fea6cc6bbc8d82a691ad0608d68a54aeb Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@medozas.de>
Date: Mon, 22 Jun 2009 14:16:45 +0200
Subject: netfilter: xt_quota: fix incomplete initialization

Commit v2.6.29-rc5-872-gacc738f ("xtables: avoid pointer to self")
forgot to copy the initial quota value supplied by iptables into the
private structure, thus counting from whatever was in the memory
kmalloc returned.

Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 net/netfilter/xt_quota.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/netfilter/xt_quota.c b/net/netfilter/xt_quota.c
index 01dd07b764ec..98fc190e8f0e 100644
--- a/net/netfilter/xt_quota.c
+++ b/net/netfilter/xt_quota.c
@@ -54,6 +54,7 @@ static bool quota_mt_check(const struct xt_mtchk_param *par)
 	if (q->master == NULL)
 		return -ENOMEM;
 
+	q->master->quota = q->quota;
 	return true;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 4d900f9df5f0569c2dc536701e2c11b6d50ebebf Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 22 Jun 2009 14:17:12 +0200
Subject: netfilter: xt_rateest: fix comparison with self
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

As noticed by Török Edwin <edwintorok@gmail.com>:

Compiling the kernel with clang has shown this warning:

net/netfilter/xt_rateest.c:69:16: warning: self-comparison always results in a
constant value
                        ret &= pps2 == pps2;
                                    ^
Looking at the code:
if (info->flags & XT_RATEEST_MATCH_BPS)
            ret &= bps1 == bps2;
        if (info->flags & XT_RATEEST_MATCH_PPS)
            ret &= pps2 == pps2;

Judging from the MATCH_BPS case it seems to be a typo, with the intention of
comparing pps1 with pps2.

http://bugzilla.kernel.org/show_bug.cgi?id=13535

Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 net/netfilter/xt_rateest.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/netfilter/xt_rateest.c b/net/netfilter/xt_rateest.c
index 220a1d588ee0..4fc6a917f6de 100644
--- a/net/netfilter/xt_rateest.c
+++ b/net/netfilter/xt_rateest.c
@@ -66,7 +66,7 @@ xt_rateest_mt(const struct sk_buff *skb, const struct xt_match_param *par)
 		if (info->flags & XT_RATEEST_MATCH_BPS)
 			ret &= bps1 == bps2;
 		if (info->flags & XT_RATEEST_MATCH_PPS)
-			ret &= pps2 == pps2;
+			ret &= pps1 == pps2;
 		break;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 29959a09cc1aabd2d5f4f03afc0305de6bd29248 Mon Sep 17 00:00:00 2001
From: Wai Yew CHAY <wychay@ctl.creative.com>
Date: Mon, 22 Jun 2009 14:52:34 +0200
Subject: ALSA: ctxfi - Add PM support

Added the suspend/resume support to ctxfi driver.

The team tested on the following seems ok:
  AMD Athlon 64 3500+ / ASUS A8N-E / 512MB DDR ATI / Radeon X1300
  20k1 & 20k2 cards

Signed-off-by: Wai Yew CHAY <wychay@ctl.creative.com>
Singed-off-by: Ryan RICHARDS <ryan_richards@creativelabs.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/ctxfi/ctatc.c      | 181 +++++++++++++++++++++++++++++++------------
 sound/pci/ctxfi/ctatc.h      |   7 ++
 sound/pci/ctxfi/cthardware.h |   4 +
 sound/pci/ctxfi/cthw20k1.c   |  83 ++++++++++++++++----
 sound/pci/ctxfi/cthw20k2.c   |  65 ++++++++++++----
 sound/pci/ctxfi/ctmixer.c    |  92 +++++++++++++++-------
 sound/pci/ctxfi/ctmixer.h    |   3 +
 sound/pci/ctxfi/ctpcm.c      |   4 +
 sound/pci/ctxfi/xfi.c        |  22 ++++++
 9 files changed, 354 insertions(+), 107 deletions(-)

diff --git a/sound/pci/ctxfi/ctatc.c b/sound/pci/ctxfi/ctatc.c
index 32e3c26e969e..a49c76647307 100644
--- a/sound/pci/ctxfi/ctatc.c
+++ b/sound/pci/ctxfi/ctatc.c
@@ -261,13 +261,8 @@ static int atc_pcm_playback_prepare(struct ct_atc *atc, struct ct_atc_pcm *apcm)
 	int device = apcm->substream->pcm->device;
 	unsigned int pitch;
 
-	if (NULL != apcm->src) {
-		/* Prepared pcm playback */
-		return 0;
-	}
-
 	/* first release old resources */
-	atc->pcm_release_resources(atc, apcm);
+	atc_pcm_release_resources(atc, apcm);
 
 	/* Get SRC resource */
 	desc.multi = apcm->substream->runtime->channels;
@@ -661,10 +656,7 @@ static int atc_pcm_capture_prepare(struct ct_atc *atc, struct ct_atc_pcm *apcm)
 	unsigned int pitch;
 	int mix_base = 0, imp_base = 0;
 
-	if (NULL != apcm->src) {
-		/* Prepared pcm capture */
-		return 0;
-	}
+	atc_pcm_release_resources(atc, apcm);
 
 	/* Get needed resources. */
 	err = atc_pcm_capture_get_resources(atc, apcm);
@@ -867,7 +859,7 @@ spdif_passthru_playback_setup(struct ct_atc *atc, struct ct_atc_pcm *apcm)
 	struct dao *dao = container_of(atc->daios[SPDIFOO], struct dao, daio);
 	unsigned int rate = apcm->substream->runtime->rate;
 	unsigned int status;
-	int err;
+	int err = 0;
 	unsigned char iec958_con_fs;
 
 	switch (rate) {
@@ -908,8 +900,7 @@ spdif_passthru_playback_prepare(struct ct_atc *atc, struct ct_atc_pcm *apcm)
 	int err;
 	int i;
 
-	if (NULL != apcm->src)
-		return 0;
+	atc_pcm_release_resources(atc, apcm);
 
 	/* Configure SPDIFOO and PLL to passthrough mode;
 	 * determine pll_rate. */
@@ -1116,32 +1107,20 @@ static int atc_spdif_out_passthru(struct ct_atc *atc, unsigned char state)
 	return err;
 }
 
-static int ct_atc_destroy(struct ct_atc *atc)
+static int atc_release_resources(struct ct_atc *atc)
 {
-	struct daio_mgr *daio_mgr;
-	struct dao *dao;
-	struct dai *dai;
-	struct daio *daio;
-	struct sum_mgr *sum_mgr;
-	struct src_mgr *src_mgr;
-	struct srcimp_mgr *srcimp_mgr;
-	struct srcimp *srcimp;
-	struct ct_mixer *mixer;
-	int i = 0;
-
-	if (NULL == atc)
-		return 0;
-
-	if (atc->timer) {
-		ct_timer_free(atc->timer);
-		atc->timer = NULL;
-	}
-
-	/* Stop hardware and disable all interrupts */
-	if (NULL != atc->hw)
-		((struct hw *)atc->hw)->card_stop(atc->hw);
-
-	/* Destroy internal mixer objects */
+	int i;
+	struct daio_mgr *daio_mgr = NULL;
+	struct dao *dao = NULL;
+	struct dai *dai = NULL;
+	struct daio *daio = NULL;
+	struct sum_mgr *sum_mgr = NULL;
+	struct src_mgr *src_mgr = NULL;
+	struct srcimp_mgr *srcimp_mgr = NULL;
+	struct srcimp *srcimp = NULL;
+	struct ct_mixer *mixer = NULL;
+
+	/* disconnect internal mixer objects */
 	if (NULL != atc->mixer) {
 		mixer = atc->mixer;
 		mixer->set_input_left(mixer, MIX_LINE_IN, NULL);
@@ -1150,7 +1129,6 @@ static int ct_atc_destroy(struct ct_atc *atc)
 		mixer->set_input_right(mixer, MIX_MIC_IN, NULL);
 		mixer->set_input_left(mixer, MIX_SPDIF_IN, NULL);
 		mixer->set_input_right(mixer, MIX_SPDIF_IN, NULL);
-		ct_mixer_destroy(atc->mixer);
 	}
 
 	if (NULL != atc->daios) {
@@ -1168,6 +1146,7 @@ static int ct_atc_destroy(struct ct_atc *atc)
 			daio_mgr->put_daio(daio_mgr, daio);
 		}
 		kfree(atc->daios);
+		atc->daios = NULL;
 	}
 
 	if (NULL != atc->pcm) {
@@ -1176,6 +1155,7 @@ static int ct_atc_destroy(struct ct_atc *atc)
 			sum_mgr->put_sum(sum_mgr, atc->pcm[i]);
 
 		kfree(atc->pcm);
+		atc->pcm = NULL;
 	}
 
 	if (NULL != atc->srcs) {
@@ -1184,6 +1164,7 @@ static int ct_atc_destroy(struct ct_atc *atc)
 			src_mgr->put_src(src_mgr, atc->srcs[i]);
 
 		kfree(atc->srcs);
+		atc->srcs = NULL;
 	}
 
 	if (NULL != atc->srcimps) {
@@ -1194,8 +1175,30 @@ static int ct_atc_destroy(struct ct_atc *atc)
 			srcimp_mgr->put_srcimp(srcimp_mgr, atc->srcimps[i]);
 		}
 		kfree(atc->srcimps);
+		atc->srcimps = NULL;
+	}
+
+	return 0;
+}
+
+static int ct_atc_destroy(struct ct_atc *atc)
+{
+	int i = 0;
+
+	if (NULL == atc)
+		return 0;
+
+	if (atc->timer) {
+		ct_timer_free(atc->timer);
+		atc->timer = NULL;
 	}
 
+	atc_release_resources(atc);
+
+	/* Destroy internal mixer objects */
+	if (NULL != atc->mixer)
+		ct_mixer_destroy(atc->mixer);
+
 	for (i = 0; i < NUM_RSCTYP; i++) {
 		if ((NULL != rsc_mgr_funcs[i].destroy) &&
 		    (NULL != atc->rsc_mgrs[i]))
@@ -1323,7 +1326,7 @@ static int __devinit atc_create_hw_devs(struct ct_atc *atc)
 	return 0;
 }
 
-static int __devinit atc_get_resources(struct ct_atc *atc)
+static int atc_get_resources(struct ct_atc *atc)
 {
 	struct daio_desc da_desc = {0};
 	struct daio_mgr *daio_mgr;
@@ -1420,16 +1423,10 @@ static int __devinit atc_get_resources(struct ct_atc *atc)
 		atc->n_pcm++;
 	}
 
-	err = ct_mixer_create(atc, (struct ct_mixer **)&atc->mixer);
-	if (err) {
-		printk(KERN_ERR "ctxfi: Failed to create mixer obj!!!\n");
-		return err;
-	}
-
 	return 0;
 }
 
-static void __devinit
+static void
 atc_connect_dai(struct src_mgr *src_mgr, struct dai *dai,
 		struct src **srcs, struct srcimp **srcimps)
 {
@@ -1468,7 +1465,7 @@ atc_connect_dai(struct src_mgr *src_mgr, struct dai *dai,
 	src_mgr->commit_write(src_mgr); /* Synchronously enable SRCs */
 }
 
-static void __devinit atc_connect_resources(struct ct_atc *atc)
+static void atc_connect_resources(struct ct_atc *atc)
 {
 	struct dai *dai;
 	struct dao *dao;
@@ -1514,6 +1511,84 @@ static void __devinit atc_connect_resources(struct ct_atc *atc)
 	}
 }
 
+#ifdef CONFIG_PM
+static int atc_suspend(struct ct_atc *atc, pm_message_t state)
+{
+	int i;
+	struct hw *hw = atc->hw;
+
+	snd_power_change_state(atc->card, SNDRV_CTL_POWER_D3hot);
+
+	for (i = FRONT; i < NUM_PCMS; i++) {
+		if (!atc->pcms[i])
+			continue;
+
+		snd_pcm_suspend_all(atc->pcms[i]);
+	}
+
+	atc_release_resources(atc);
+
+	hw->suspend(hw, state);
+
+	return 0;
+}
+
+static int atc_hw_resume(struct ct_atc *atc)
+{
+	struct hw *hw = atc->hw;
+	struct card_conf info = {0};
+
+	/* Re-initialize card hardware. */
+	info.rsr = atc->rsr;
+	info.msr = atc->msr;
+	info.vm_pgt_phys = atc_get_ptp_phys(atc, 0);
+	return hw->resume(hw, &info);
+}
+
+static int atc_resources_resume(struct ct_atc *atc)
+{
+	struct ct_mixer *mixer;
+	int err = 0;
+
+	/* Get resources */
+	err = atc_get_resources(atc);
+	if (err < 0) {
+		atc_release_resources(atc);
+		return err;
+	}
+
+	/* Build topology */
+	atc_connect_resources(atc);
+
+	mixer = atc->mixer;
+	mixer->resume(mixer);
+
+	return 0;
+}
+
+static int atc_resume(struct ct_atc *atc)
+{
+	int err = 0;
+
+	/* Do hardware resume. */
+	err = atc_hw_resume(atc);
+	if (err < 0) {
+		printk(KERN_ERR "ctxfi: pci_enable_device failed, "
+		       "disabling device\n");
+		snd_card_disconnect(atc->card);
+		return err;
+	}
+
+	err = atc_resources_resume(atc);
+	if (err < 0)
+		return err;
+
+	snd_power_change_state(atc->card, SNDRV_CTL_POWER_D0);
+
+	return 0;
+}
+#endif
+
 static struct ct_atc atc_preset __devinitdata = {
 	.map_audio_buffer = ct_map_audio_buffer,
 	.unmap_audio_buffer = ct_unmap_audio_buffer,
@@ -1542,6 +1617,10 @@ static struct ct_atc atc_preset __devinitdata = {
 	.spdif_out_set_status = atc_spdif_out_set_status,
 	.spdif_out_passthru = atc_spdif_out_passthru,
 	.have_digit_io_switch = atc_have_digit_io_switch,
+#ifdef CONFIG_PM
+	.suspend = atc_suspend,
+	.resume = atc_resume,
+#endif
 };
 
 /**
@@ -1600,6 +1679,12 @@ int __devinit ct_atc_create(struct snd_card *card, struct pci_dev *pci,
 	if (err < 0)
 		goto error1;
 
+	err = ct_mixer_create(atc, (struct ct_mixer **)&atc->mixer);
+	if (err) {
+		printk(KERN_ERR "ctxfi: Failed to create mixer obj!!!\n");
+		goto error1;
+	}
+
 	/* Get resources */
 	err = atc_get_resources(atc);
 	if (err < 0)
diff --git a/sound/pci/ctxfi/ctatc.h b/sound/pci/ctxfi/ctatc.h
index 9fe620ea5f3f..9fd8a5708943 100644
--- a/sound/pci/ctxfi/ctatc.h
+++ b/sound/pci/ctxfi/ctatc.h
@@ -136,6 +136,13 @@ struct ct_atc {
 	unsigned char n_pcm;
 
 	struct ct_timer *timer;
+
+#ifdef CONFIG_PM
+	int (*suspend)(struct ct_atc *atc, pm_message_t state);
+	int (*resume)(struct ct_atc *atc);
+#define NUM_PCMS (NUM_CTALSADEVS - 1)
+	struct snd_pcm *pcms[NUM_PCMS];
+#endif
 };
 
 
diff --git a/sound/pci/ctxfi/cthardware.h b/sound/pci/ctxfi/cthardware.h
index a4c2cad80f3a..af55405f5dec 100644
--- a/sound/pci/ctxfi/cthardware.h
+++ b/sound/pci/ctxfi/cthardware.h
@@ -64,6 +64,10 @@ struct hw {
 	int (*card_init)(struct hw *hw, struct card_conf *info);
 	int (*card_stop)(struct hw *hw);
 	int (*pll_init)(struct hw *hw, unsigned int rsr);
+#ifdef CONFIG_PM
+	int (*suspend)(struct hw *hw, pm_message_t state);
+	int (*resume)(struct hw *hw, struct card_conf *info);
+#endif
 	int (*is_adc_source_selected)(struct hw *hw, enum ADCSRC source);
 	int (*select_adc_source)(struct hw *hw, enum ADCSRC source);
 	int (*have_digit_io_switch)(struct hw *hw);
diff --git a/sound/pci/ctxfi/cthw20k1.c b/sound/pci/ctxfi/cthw20k1.c
index cb69d9ddfbe3..ad3e1d144464 100644
--- a/sound/pci/ctxfi/cthw20k1.c
+++ b/sound/pci/ctxfi/cthw20k1.c
@@ -1911,9 +1911,17 @@ static int hw_card_start(struct hw *hw)
 		goto error1;
 	}
 
-	err = pci_request_regions(pci, "XFi");
-	if (err < 0)
-		goto error1;
+	if (!hw->io_base) {
+		err = pci_request_regions(pci, "XFi");
+		if (err < 0)
+			goto error1;
+
+		if (hw->model == CTUAA)
+			hw->io_base = pci_resource_start(pci, 5);
+		else
+			hw->io_base = pci_resource_start(pci, 0);
+
+	}
 
 	/* Switch to X-Fi mode from UAA mode if neeeded */
 	if (hw->model == CTUAA) {
@@ -1921,18 +1929,17 @@ static int hw_card_start(struct hw *hw)
 		if (err)
 			goto error2;
 
-		hw->io_base = pci_resource_start(pci, 5);
-	} else {
-		hw->io_base = pci_resource_start(pci, 0);
 	}
 
-	err = request_irq(pci->irq, ct_20k1_interrupt, IRQF_SHARED,
-			  "ctxfi", hw);
-	if (err < 0) {
-		printk(KERN_ERR "XFi: Cannot get irq %d\n", pci->irq);
-		goto error2;
+	if (hw->irq < 0) {
+		err = request_irq(pci->irq, ct_20k1_interrupt, IRQF_SHARED,
+				  "ctxfi", hw);
+		if (err < 0) {
+			printk(KERN_ERR "XFi: Cannot get irq %d\n", pci->irq);
+			goto error2;
+		}
+		hw->irq = pci->irq;
 	}
-	hw->irq = pci->irq;
 
 	pci_set_master(pci);
 
@@ -1948,6 +1955,15 @@ error1:
 
 static int hw_card_stop(struct hw *hw)
 {
+	unsigned int data;
+
+	/* disable transport bus master and queueing of request */
+	hw_write_20kx(hw, TRNCTL, 0x00);
+
+	/* disable pll */
+	data = hw_read_20kx(hw, PLLCTL);
+	hw_write_20kx(hw, PLLCTL, (data & (~(0x0F<<12))));
+
 	/* TODO: Disable interrupt and so on... */
 	if (hw->irq >= 0)
 		synchronize_irq(hw->irq);
@@ -1987,11 +2003,9 @@ static int hw_card_init(struct hw *hw, struct card_conf *info)
 	struct trn_conf trn_info = {0};
 
 	/* Get PCI io port base address and do Hendrix switch if needed. */
-	if (!hw->io_base) {
-		err = hw_card_start(hw);
-		if (err)
-			return err;
-	}
+	err = hw_card_start(hw);
+	if (err)
+		return err;
 
 	/* PLL init */
 	err = hw_pll_init(hw, info->rsr);
@@ -2064,6 +2078,37 @@ static int hw_card_init(struct hw *hw, struct card_conf *info)
 	return 0;
 }
 
+#ifdef CONFIG_PM
+static int hw_suspend(struct hw *hw, pm_message_t state)
+{
+	struct pci_dev *pci = hw->pci;
+
+	hw_card_stop(hw);
+
+	if (hw->model == CTUAA) {
+		/* Switch to UAA config space. */
+		pci_write_config_dword(pci, UAA_CFG_SPACE_FLAG, 0x0);
+	}
+
+	pci_disable_device(pci);
+	pci_save_state(pci);
+	pci_set_power_state(pci, pci_choose_state(pci, state));
+
+	return 0;
+}
+
+static int hw_resume(struct hw *hw, struct card_conf *info)
+{
+	struct pci_dev *pci = hw->pci;
+
+	pci_set_power_state(pci, PCI_D0);
+	pci_restore_state(pci);
+
+	/* Re-initialize card hardware. */
+	return hw_card_init(hw, info);
+}
+#endif
+
 static u32 hw_read_20kx(struct hw *hw, u32 reg)
 {
 	u32 value;
@@ -2128,6 +2173,10 @@ static struct hw ct20k1_preset __devinitdata = {
 	.is_adc_source_selected = hw_is_adc_input_selected,
 	.select_adc_source = hw_adc_input_select,
 	.have_digit_io_switch = hw_have_digit_io_switch,
+#ifdef CONFIG_PM
+	.suspend = hw_suspend,
+	.resume = hw_resume,
+#endif
 
 	.src_rsc_get_ctrl_blk = src_get_rsc_ctrl_blk,
 	.src_rsc_put_ctrl_blk = src_put_rsc_ctrl_blk,
diff --git a/sound/pci/ctxfi/cthw20k2.c b/sound/pci/ctxfi/cthw20k2.c
index 4493a51c6b01..dec46d04b041 100644
--- a/sound/pci/ctxfi/cthw20k2.c
+++ b/sound/pci/ctxfi/cthw20k2.c
@@ -1860,16 +1860,18 @@ static int hw_card_start(struct hw *hw)
 		goto error1;
 	}
 
-	err = pci_request_regions(pci, "XFi");
-	if (err < 0)
-		goto error1;
+	if (!hw->io_base) {
+		err = pci_request_regions(pci, "XFi");
+		if (err < 0)
+			goto error1;
 
-	hw->io_base = pci_resource_start(hw->pci, 2);
-	hw->mem_base = (unsigned long)ioremap(hw->io_base,
+		hw->io_base = pci_resource_start(hw->pci, 2);
+		hw->mem_base = (unsigned long)ioremap(hw->io_base,
 					pci_resource_len(hw->pci, 2));
-	if (NULL == (void *)hw->mem_base) {
-		err = -ENOENT;
-		goto error2;
+		if (NULL == (void *)hw->mem_base) {
+			err = -ENOENT;
+			goto error2;
+		}
 	}
 
 	/* Switch to 20k2 mode from UAA mode. */
@@ -1901,6 +1903,15 @@ error1:
 
 static int hw_card_stop(struct hw *hw)
 {
+	unsigned int data;
+
+	/* disable transport bus master and queueing of request */
+	hw_write_20kx(hw, TRANSPORT_CTL, 0x00);
+
+	/* disable pll */
+	data = hw_read_20kx(hw, PLL_ENB);
+	hw_write_20kx(hw, PLL_ENB, (data & (~0x07)));
+
 	/* TODO: Disable interrupt and so on... */
 	return 0;
 }
@@ -1939,11 +1950,9 @@ static int hw_card_init(struct hw *hw, struct card_conf *info)
 
 	/* Get PCI io port/memory base address and
 	 * do 20kx core switch if needed. */
-	if (!hw->io_base) {
-		err = hw_card_start(hw);
-		if (err)
-			return err;
-	}
+	err = hw_card_start(hw);
+	if (err)
+		return err;
 
 	/* PLL init */
 	err = hw_pll_init(hw, info->rsr);
@@ -2006,6 +2015,32 @@ static int hw_card_init(struct hw *hw, struct card_conf *info)
 	return 0;
 }
 
+#ifdef CONFIG_PM
+static int hw_suspend(struct hw *hw, pm_message_t state)
+{
+	struct pci_dev *pci = hw->pci;
+
+	hw_card_stop(hw);
+
+	pci_disable_device(pci);
+	pci_save_state(pci);
+	pci_set_power_state(pci, pci_choose_state(pci, state));
+
+	return 0;
+}
+
+static int hw_resume(struct hw *hw, struct card_conf *info)
+{
+	struct pci_dev *pci = hw->pci;
+
+	pci_set_power_state(pci, PCI_D0);
+	pci_restore_state(pci);
+
+	/* Re-initialize card hardware. */
+	return hw_card_init(hw, info);
+}
+#endif
+
 static u32 hw_read_20kx(struct hw *hw, u32 reg)
 {
 	return readl((void *)(hw->mem_base + reg));
@@ -2025,6 +2060,10 @@ static struct hw ct20k2_preset __devinitdata = {
 	.is_adc_source_selected = hw_is_adc_input_selected,
 	.select_adc_source = hw_adc_input_select,
 	.have_digit_io_switch = hw_have_digit_io_switch,
+#ifdef CONFIG_PM
+	.suspend = hw_suspend,
+	.resume = hw_resume,
+#endif
 
 	.src_rsc_get_ctrl_blk = src_get_rsc_ctrl_blk,
 	.src_rsc_put_ctrl_blk = src_put_rsc_ctrl_blk,
diff --git a/sound/pci/ctxfi/ctmixer.c b/sound/pci/ctxfi/ctmixer.c
index 666722d9de41..f26d7cd9db9f 100644
--- a/sound/pci/ctxfi/ctmixer.c
+++ b/sound/pci/ctxfi/ctmixer.c
@@ -462,6 +462,43 @@ do_digit_io_switch(struct ct_atc *atc, int state)
 	return;
 }
 
+static void do_switch(struct ct_atc *atc, enum CTALSA_MIXER_CTL type, int state)
+{
+	struct ct_mixer *mixer = atc->mixer;
+
+	/* Do changes in mixer. */
+	if ((SWH_CAPTURE_START <= type) && (SWH_CAPTURE_END >= type)) {
+		if (state) {
+			ct_mixer_recording_select(mixer,
+						  get_amixer_index(type));
+		} else {
+			ct_mixer_recording_unselect(mixer,
+						    get_amixer_index(type));
+		}
+	}
+	/* Do changes out of mixer. */
+	if (state && (MIXER_LINEIN_C_S == type || MIXER_MIC_C_S == type))
+		do_line_mic_switch(atc, type);
+	else if (MIXER_WAVEF_P_S == type)
+		atc->line_front_unmute(atc, state);
+	else if (MIXER_WAVES_P_S == type)
+		atc->line_surround_unmute(atc, state);
+	else if (MIXER_WAVEC_P_S == type)
+		atc->line_clfe_unmute(atc, state);
+	else if (MIXER_WAVER_P_S == type)
+		atc->line_rear_unmute(atc, state);
+	else if (MIXER_LINEIN_P_S == type)
+		atc->line_in_unmute(atc, state);
+	else if (MIXER_SPDIFO_P_S == type)
+		atc->spdif_out_unmute(atc, state);
+	else if (MIXER_SPDIFI_P_S == type)
+		atc->spdif_in_unmute(atc, state);
+	else if (MIXER_DIGITAL_IO_S == type)
+		do_digit_io_switch(atc, state);
+
+	return;
+}
+
 static int ct_alsa_mix_switch_info(struct snd_kcontrol *kcontrol,
 				   struct snd_ctl_elem_info *uinfo)
 {
@@ -498,35 +535,7 @@ static int ct_alsa_mix_switch_put(struct snd_kcontrol *kcontrol,
 		return 0;
 
 	set_switch_state(mixer, type, state);
-	/* Do changes in mixer. */
-	if ((SWH_CAPTURE_START <= type) && (SWH_CAPTURE_END >= type)) {
-		if (state) {
-			ct_mixer_recording_select(mixer,
-						  get_amixer_index(type));
-		} else {
-			ct_mixer_recording_unselect(mixer,
-						    get_amixer_index(type));
-		}
-	}
-	/* Do changes out of mixer. */
-	if (state && (MIXER_LINEIN_C_S == type || MIXER_MIC_C_S == type))
-		do_line_mic_switch(atc, type);
-	else if (MIXER_WAVEF_P_S == type)
-		atc->line_front_unmute(atc, state);
-	else if (MIXER_WAVES_P_S == type)
-		atc->line_surround_unmute(atc, state);
-	else if (MIXER_WAVEC_P_S == type)
-		atc->line_clfe_unmute(atc, state);
-	else if (MIXER_WAVER_P_S == type)
-		atc->line_rear_unmute(atc, state);
-	else if (MIXER_LINEIN_P_S == type)
-		atc->line_in_unmute(atc, state);
-	else if (MIXER_SPDIFO_P_S == type)
-		atc->spdif_out_unmute(atc, state);
-	else if (MIXER_SPDIFI_P_S == type)
-		atc->spdif_in_unmute(atc, state);
-	else if (MIXER_DIGITAL_IO_S == type)
-		do_digit_io_switch(atc, state);
+	do_switch(atc, type, state);
 
 	return 1;
 }
@@ -1039,6 +1048,28 @@ mixer_set_input_right(struct ct_mixer *mixer,
 	return 0;
 }
 
+#ifdef CONFIG_PM
+static int mixer_resume(struct ct_mixer *mixer)
+{
+	int i, state;
+	struct amixer *amixer;
+
+	/* resume topology and volume gain. */
+	for (i = 0; i < NUM_CT_AMIXERS*CHN_NUM; i++) {
+		amixer = mixer->amixers[i];
+		amixer->ops->commit_write(amixer);
+	}
+
+	/* resume switch state. */
+	for (i = SWH_MIXER_START; i <= SWH_MIXER_END; i++) {
+		state = get_switch_state(mixer, i);
+		do_switch(mixer->atc, i, state);
+	}
+
+	return 0;
+}
+#endif
+
 int ct_mixer_destroy(struct ct_mixer *mixer)
 {
 	struct sum_mgr *sum_mgr = (struct sum_mgr *)mixer->atc->rsc_mgrs[SUM];
@@ -1087,6 +1118,9 @@ int ct_mixer_create(struct ct_atc *atc, struct ct_mixer **rmixer)
 	mixer->get_output_ports = mixer_get_output_ports;
 	mixer->set_input_left = mixer_set_input_left;
 	mixer->set_input_right = mixer_set_input_right;
+#ifdef CONFIG_PM
+	mixer->resume = mixer_resume;
+#endif
 
 	/* Allocate chip resources for mixer obj */
 	err = ct_mixer_get_resources(mixer);
diff --git a/sound/pci/ctxfi/ctmixer.h b/sound/pci/ctxfi/ctmixer.h
index e2d96ebde746..b009e989e77d 100644
--- a/sound/pci/ctxfi/ctmixer.h
+++ b/sound/pci/ctxfi/ctmixer.h
@@ -56,6 +56,9 @@ struct ct_mixer {
 			      enum MIXER_PORT_T type, struct rsc *rsc);
 	int (*set_input_right)(struct ct_mixer *mixer,
 			       enum MIXER_PORT_T type, struct rsc *rsc);
+#ifdef CONFIG_PM
+	int (*resume)(struct ct_mixer *mixer);
+#endif
 };
 
 int ct_alsa_mix_create(struct ct_atc *atc,
diff --git a/sound/pci/ctxfi/ctpcm.c b/sound/pci/ctxfi/ctpcm.c
index 9e5c0c4da726..60ea23180acb 100644
--- a/sound/pci/ctxfi/ctpcm.c
+++ b/sound/pci/ctxfi/ctpcm.c
@@ -422,5 +422,9 @@ int ct_alsa_pcm_create(struct ct_atc *atc,
 	snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV_SG,
 			snd_dma_pci_data(atc->pci), 128*1024, 128*1024);
 
+#ifdef CONFIG_PM
+	atc->pcms[device] = pcm;
+#endif
+
 	return 0;
 }
diff --git a/sound/pci/ctxfi/xfi.c b/sound/pci/ctxfi/xfi.c
index 2d3dd89af151..76541748e7bc 100644
--- a/sound/pci/ctxfi/xfi.c
+++ b/sound/pci/ctxfi/xfi.c
@@ -121,11 +121,33 @@ static void __devexit ct_card_remove(struct pci_dev *pci)
 	pci_set_drvdata(pci, NULL);
 }
 
+#ifdef CONFIG_PM
+static int ct_card_suspend(struct pci_dev *pci, pm_message_t state)
+{
+	struct snd_card *card = pci_get_drvdata(pci);
+	struct ct_atc *atc = card->private_data;
+
+	return atc->suspend(atc, state);
+}
+
+static int ct_card_resume(struct pci_dev *pci)
+{
+	struct snd_card *card = pci_get_drvdata(pci);
+	struct ct_atc *atc = card->private_data;
+
+	return atc->resume(atc);
+}
+#endif
+
 static struct pci_driver ct_driver = {
 	.name = "SB-XFi",
 	.id_table = ct_pci_dev_ids,
 	.probe = ct_card_probe,
 	.remove = __devexit_p(ct_card_remove),
+#ifdef CONFIG_PM
+	.suspend = ct_card_suspend,
+	.resume = ct_card_resume,
+#endif
 };
 
 static int __init ct_card_init(void)
-- 
cgit v1.2.3-59-g8ed1b


From 6ade7fa7bb4484b4dc78e55700d0f62bacd491e9 Mon Sep 17 00:00:00 2001
From: Peter Korsgaard <jacmet@sunsite.dk>
Date: Thu, 18 Jun 2009 23:54:43 +0200
Subject: [ARM] S3C64XX: fix HCLK gate defines

A few typos seems to have sneaked into the HCLK gate defines, causing the
usb host clock to not get enabled. Fix them according to the reference
manual and throw in the 3d accel bit for good measure.

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c64xx/include/plat/regs-clock.h | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/arch/arm/plat-s3c64xx/include/plat/regs-clock.h b/arch/arm/plat-s3c64xx/include/plat/regs-clock.h
index 52836d41e333..a8777a755dfa 100644
--- a/arch/arm/plat-s3c64xx/include/plat/regs-clock.h
+++ b/arch/arm/plat-s3c64xx/include/plat/regs-clock.h
@@ -88,11 +88,11 @@
 #define S3C6400_CLKDIV2_SPI0_SHIFT	(0)
 
 /* HCLK GATE Registers */
-#define S3C_CLKCON_HCLK_BUS	(1<<30)
-#define S3C_CLKCON_HCLK_SECUR	(1<<29)
-#define S3C_CLKCON_HCLK_SDMA1	(1<<28)
-#define S3C_CLKCON_HCLK_SDMA2	(1<<27)
-#define S3C_CLKCON_HCLK_UHOST	(1<<26)
+#define S3C_CLKCON_HCLK_3DSE	(1<<31)
+#define S3C_CLKCON_HCLK_UHOST	(1<<29)
+#define S3C_CLKCON_HCLK_SECUR	(1<<28)
+#define S3C_CLKCON_HCLK_SDMA1	(1<<27)
+#define S3C_CLKCON_HCLK_SDMA0	(1<<26)
 #define S3C_CLKCON_HCLK_IROM	(1<<25)
 #define S3C_CLKCON_HCLK_DDR1	(1<<24)
 #define S3C_CLKCON_HCLK_DDR0	(1<<23)
-- 
cgit v1.2.3-59-g8ed1b


From 386f43517f8bf026320fd4b9e077140681d7e595 Mon Sep 17 00:00:00 2001
From: Peter Korsgaard <jacmet@sunsite.dk>
Date: Thu, 18 Jun 2009 23:54:44 +0200
Subject: [ARM] S3C64XX: clock.c: fix typo in usb-host clock ctrlbit

The usb-host clock was using the wrong define (the SCLK enable for the
usb-host-bus) to change the HCLK register instead of the HCLK_UHOST bit.

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c64xx/clock.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/plat-s3c64xx/clock.c b/arch/arm/plat-s3c64xx/clock.c
index 0bc2fa1dfc40..7a36e899360d 100644
--- a/arch/arm/plat-s3c64xx/clock.c
+++ b/arch/arm/plat-s3c64xx/clock.c
@@ -191,7 +191,7 @@ static struct clk init_clocks[] = {
 		.id		= -1,
 		.parent		= &clk_h,
 		.enable		= s3c64xx_hclk_ctrl,
-		.ctrlbit	= S3C_CLKCON_SCLK_UHOST,
+		.ctrlbit	= S3C_CLKCON_HCLK_UHOST,
 	}, {
 		.name		= "hsmmc",
 		.id		= 0,
-- 
cgit v1.2.3-59-g8ed1b


From 0b495737f311cb8eb5e66404309ea16e4d7ded8a Mon Sep 17 00:00:00 2001
From: Marek Szyprowski <m.szyprowski@samsung.com>
Date: Thu, 18 Jun 2009 13:30:27 +0200
Subject: [ARM] S3C64XX: add to_irq() support for EINT() GPIO N group

Add to_irq() function to onvert gpio to irq for external interrupt
group (GPN).

Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c64xx/gpiolib.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/arch/arm/plat-s3c64xx/gpiolib.c b/arch/arm/plat-s3c64xx/gpiolib.c
index da7b60ee5e67..92859290ea33 100644
--- a/arch/arm/plat-s3c64xx/gpiolib.c
+++ b/arch/arm/plat-s3c64xx/gpiolib.c
@@ -321,6 +321,11 @@ static struct s3c_gpio_cfg gpio_2bit_cfg_eint11 = {
 	.get_pull	= s3c_gpio_getpull_updown,
 };
 
+int s3c64xx_gpio2int_gpn(struct gpio_chip *chip, unsigned pin)
+{
+	return IRQ_EINT(0) + pin;
+}
+
 static struct s3c_gpio_chip gpio_2bit[] = {
 	{
 		.base	= S3C64XX_GPF_BASE,
@@ -353,6 +358,7 @@ static struct s3c_gpio_chip gpio_2bit[] = {
 			.base	= S3C64XX_GPN(0),
 			.ngpio	= S3C64XX_GPIO_N_NR,
 			.label	= "GPN",
+			.to_irq = s3c64xx_gpio2int_gpn,
 		},
 	}, {
 		.base	= S3C64XX_GPO_BASE,
-- 
cgit v1.2.3-59-g8ed1b


From 49fb88af23f3344ba53d6dbe34ac0b1426d81006 Mon Sep 17 00:00:00 2001
From: Marek Szyprowski <m.szyprowski@samsung.com>
Date: Thu, 18 Jun 2009 13:30:16 +0200
Subject: [ARM] S3C: Fix gpio-config off-by-one bug

Fix gpio-config off-by-one bug. Without this patch, touching GPA0 pin on
S3C64XX platform causes kernel oops.

Reviewed-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c/gpio-config.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/plat-s3c/gpio-config.c b/arch/arm/plat-s3c/gpio-config.c
index 08044dec9731..456969b6fa0d 100644
--- a/arch/arm/plat-s3c/gpio-config.c
+++ b/arch/arm/plat-s3c/gpio-config.c
@@ -119,7 +119,7 @@ int s3c_gpio_setcfg_s3c64xx_4bit(struct s3c_gpio_chip *chip,
 	unsigned int shift = (off & 7) * 4;
 	u32 con;
 
-	if (off < 8 && chip->chip.ngpio >= 8)
+	if (off < 8 && chip->chip.ngpio > 8)
 		reg -= 4;
 
 	if (s3c_gpio_is_cfg_special(cfg)) {
-- 
cgit v1.2.3-59-g8ed1b


From 129dd98194747a3b8ac1ff876d8d1f2440660d01 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Sun, 21 Jun 2009 23:59:01 +0200
Subject: fusion: mptsas, fix lock imbalance

Fix two typos in mptsas_not_responding_devices. It was mutex_lock instead
of unlock.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Acked-by: "Desai, Kashyap" <Kashyap.Desai@lsi.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/message/fusion/mptsas.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/message/fusion/mptsas.c b/drivers/message/fusion/mptsas.c
index 20e0b447e8e8..55ff25244af4 100644
--- a/drivers/message/fusion/mptsas.c
+++ b/drivers/message/fusion/mptsas.c
@@ -3518,7 +3518,7 @@ retry_page:
 		} else
 			mptsas_volume_delete(ioc, sas_info->fw.id);
 	}
-	mutex_lock(&ioc->sas_device_info_mutex);
+	mutex_unlock(&ioc->sas_device_info_mutex);
 
 	/* expanders */
 	mutex_lock(&ioc->sas_topology_mutex);
@@ -3549,7 +3549,7 @@ retry_page:
 			goto redo_expander_scan;
 		}
 	}
-	mutex_lock(&ioc->sas_topology_mutex);
+	mutex_unlock(&ioc->sas_topology_mutex);
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 8c8145b8734028f6deb487f7d64748da4c6c39ac Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Mon, 22 Jun 2009 17:00:38 +0200
Subject: ALSA: hda - Make jack-plug notification selectable

Make the jack-plug notification via input layer selectable via Kconfig.
This is often unnecessary, and the similr function will be provided
using the ALSA control API in near future anyway.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/hda/Kconfig          | 9 ++++++++-
 sound/pci/hda/patch_conexant.c | 4 ++--
 sound/pci/hda/patch_sigmatel.c | 6 +++---
 3 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/sound/pci/hda/Kconfig b/sound/pci/hda/Kconfig
index c710150d5065..04438f1d682d 100644
--- a/sound/pci/hda/Kconfig
+++ b/sound/pci/hda/Kconfig
@@ -2,7 +2,6 @@ menuconfig SND_HDA_INTEL
 	tristate "Intel HD Audio"
 	select SND_PCM
 	select SND_VMASTER
-	select SND_JACK if INPUT=y || INPUT=SND
 	help
 	  Say Y here to include support for Intel "High Definition
 	  Audio" (Azalia) and its compatible devices.
@@ -39,6 +38,14 @@ config SND_HDA_INPUT_BEEP
 	  Say Y here to build a digital beep interface for HD-audio
 	  driver. This interface is used to generate digital beeps.
 
+config SND_HDA_INPUT_JACK
+	bool "Support jack plugging notification via input layer"
+	depends on INPUT=y || INPUT=SND_HDA_INTEL
+	select SND_JACK
+	help
+	  Say Y here to enable the jack plugging notification via
+	  input layer.
+
 config SND_HDA_CODEC_REALTEK
 	bool "Build Realtek HD-audio codec support"
 	default y
diff --git a/sound/pci/hda/patch_conexant.c b/sound/pci/hda/patch_conexant.c
index 4fcbe21829ab..ac868c59f9e3 100644
--- a/sound/pci/hda/patch_conexant.c
+++ b/sound/pci/hda/patch_conexant.c
@@ -349,7 +349,7 @@ static int conexant_mux_enum_put(struct snd_kcontrol *kcontrol,
 				     &spec->cur_mux[adc_idx]);
 }
 
-#ifdef CONFIG_SND_JACK
+#ifdef CONFIG_SND_HDA_INPUT_JACK
 static void conexant_free_jack_priv(struct snd_jack *jack)
 {
 	struct conexant_jack *jacks = jack->private_data;
@@ -463,7 +463,7 @@ static int conexant_init(struct hda_codec *codec)
 
 static void conexant_free(struct hda_codec *codec)
 {
-#ifdef CONFIG_SND_JACK
+#ifdef CONFIG_SND_HDA_INPUT_JACK
 	struct conexant_spec *spec = codec->spec;
 	if (spec->jacks.list) {
 		struct conexant_jack *jacks = spec->jacks.list;
diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c
index 090957735e05..14f3c3e0f62d 100644
--- a/sound/pci/hda/patch_sigmatel.c
+++ b/sound/pci/hda/patch_sigmatel.c
@@ -4035,7 +4035,7 @@ static void stac_gpio_set(struct hda_codec *codec, unsigned int mask,
 			   AC_VERB_SET_GPIO_DATA, gpiostate); /* sync */
 }
 
-#ifdef CONFIG_SND_JACK
+#ifdef CONFIG_SND_HDA_INPUT_JACK
 static void stac92xx_free_jack_priv(struct snd_jack *jack)
 {
 	struct sigmatel_jack *jacks = jack->private_data;
@@ -4047,7 +4047,7 @@ static void stac92xx_free_jack_priv(struct snd_jack *jack)
 static int stac92xx_add_jack(struct hda_codec *codec,
 		hda_nid_t nid, int type)
 {
-#ifdef CONFIG_SND_JACK
+#ifdef CONFIG_SND_HDA_INPUT_JACK
 	struct sigmatel_spec *spec = codec->spec;
 	struct sigmatel_jack *jack;
 	int def_conf = snd_hda_codec_get_pincfg(codec, nid);
@@ -4336,7 +4336,7 @@ static int stac92xx_init(struct hda_codec *codec)
 
 static void stac92xx_free_jacks(struct hda_codec *codec)
 {
-#ifdef CONFIG_SND_JACK
+#ifdef CONFIG_SND_HDA_INPUT_JACK
 	/* free jack instances manually when clearing/reconfiguring */
 	struct sigmatel_spec *spec = codec->spec;
 	if (!codec->bus->shutdown && spec->jacks.list) {
-- 
cgit v1.2.3-59-g8ed1b


From 520f2c346af463fa00924b236e092da482b344cc Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <peterz@infradead.org>
Date: Mon, 22 Jun 2009 16:52:51 +0200
Subject: perf report: Output more symbol related debug data

Print more symbol relocation related info under -vv.

Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-report.c | 5 +++--
 tools/perf/util/symbol.c    | 4 ++++
 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 5eb5566f0c95..ec230a0146e9 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -797,7 +797,7 @@ resolve_symbol(struct thread *thread, struct map **mapp,
 {
 	struct dso *dso = dsop ? *dsop : NULL;
 	struct map *map = mapp ? *mapp : NULL;
-	uint64_t ip = *ipp;
+	u64 ip = *ipp;
 
 	if (!thread)
 		return NULL;
@@ -814,7 +814,6 @@ resolve_symbol(struct thread *thread, struct map **mapp,
 			*mapp = map;
 got_map:
 		ip = map->map_ip(map, ip);
-		*ipp  = ip;
 
 		dso = map->dso;
 	} else {
@@ -828,6 +827,8 @@ got_map:
 		dso = kernel_dso;
 	}
 	dprintf(" ...... dso: %s\n", dso ? dso->name : "<not found>");
+	dprintf(" ...... map: %Lx -> %Lx\n", *ipp, ip);
+	*ipp  = ip;
 
 	if (dsop)
 		*dsop = dso;
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index 01b62fa03996..9c659ef6aec2 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -535,6 +535,10 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
 		gelf_getshdr(sec, &shdr);
 		obj_start = sym.st_value;
 
+		if (verbose >= 2)
+			printf("adjusting symbol: st_value: %Lx sh_addr: %Lx sh_offset: %Lx\n",
+				(u64)sym.st_value, (u64)shdr.sh_addr, (u64)shdr.sh_offset);
+
 		sym.st_value -= shdr.sh_addr - shdr.sh_offset;
 
 		f = symbol__new(sym.st_value, sym.st_size,
-- 
cgit v1.2.3-59-g8ed1b


From c0c22dbfa8ba3c5045eeb9c76d2822ffc44fefc3 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Mon, 22 Jun 2009 20:47:26 +0530
Subject: perf_counter tools: Set alias for page-faults

"faults" should be alias for "page-faults"

Also fixed alignment and 80 characters issue

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <1245683846.12092.1.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/util/parse-events.c | 36 ++++++++++++++++++------------------
 1 file changed, 18 insertions(+), 18 deletions(-)

diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index f5695486ad3f..06af2fadcd87 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -33,8 +33,7 @@ static struct event_symbol event_symbols[] = {
 
   { CSW(CPU_CLOCK),		"cpu-clock",		""		},
   { CSW(TASK_CLOCK),		"task-clock",		""		},
-  { CSW(PAGE_FAULTS),		"page-faults",		""		},
-  { CSW(PAGE_FAULTS),		"faults",		""		},
+  { CSW(PAGE_FAULTS),		"page-faults",		"faults"	},
   { CSW(PAGE_FAULTS_MIN),	"minor-faults",		""		},
   { CSW(PAGE_FAULTS_MAJ),	"major-faults",		""		},
   { CSW(CONTEXT_SWITCHES),	"context-switches",	"cs"		},
@@ -71,24 +70,24 @@ static char *sw_event_names[] = {
 
 #define MAX_ALIASES 8
 
-static char *hw_cache [][MAX_ALIASES] = {
-	{ "L1-data"		, "l1-d", "l1d"					},
-	{ "L1-instruction"	, "l1-i", "l1i"					},
-	{ "L2"			, "l2"						},
-	{ "Data-TLB"		, "dtlb", "d-tlb"				},
-	{ "Instruction-TLB"	, "itlb", "i-tlb"				},
-	{ "Branch"		, "bpu" , "btb", "bpc"				},
+static char *hw_cache[][MAX_ALIASES] = {
+	{ "L1-data",		"l1-d",		"l1d"			},
+	{ "L1-instruction",	"l1-i",		"l1i"			},
+	{ "L2",			"l2"					},
+	{ "Data-TLB",		"dtlb",		"d-tlb"			},
+	{ "Instruction-TLB",	"itlb",		"i-tlb"			},
+	{ "Branch",		"bpu" ,		"btb",		"bpc"	},
 };
 
-static char *hw_cache_op [][MAX_ALIASES] = {
-	{ "Load"		, "read"					},
-	{ "Store"		, "write"					},
-	{ "Prefetch"		, "speculative-read", "speculative-load"	},
+static char *hw_cache_op[][MAX_ALIASES] = {
+	{ "Load",		"read"					},
+	{ "Store",		"write"					},
+	{ "Prefetch",		"speculative-read", "speculative-load"	},
 };
 
-static char *hw_cache_result [][MAX_ALIASES] = {
-	{ "Reference"		, "ops", "access"				},
-	{ "Miss"								},
+static char *hw_cache_result[][MAX_ALIASES] = {
+	{ "Reference",		"ops",		"access"		},
+	{ "Miss"							},
 };
 
 char *event_name(int counter)
@@ -160,7 +159,8 @@ static int parse_aliases(const char *str, char *names[][MAX_ALIASES], int size)
 	return -1;
 }
 
-static int parse_generic_hw_symbols(const char *str, struct perf_counter_attr *attr)
+static int
+parse_generic_hw_symbols(const char *str, struct perf_counter_attr *attr)
 {
 	int cache_type = -1, cache_op = 0, cache_result = 0;
 
@@ -201,7 +201,7 @@ static int check_events(const char *str, unsigned int i)
 
 	if (strlen(event_symbols[i].alias))
 		if (!strncmp(str, event_symbols[i].alias,
-		      strlen(event_symbols[i].alias)))
+			     strlen(event_symbols[i].alias)))
 			return 1;
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 752a4787511bf7515f99609ff4ae52341b5bfcde Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Mon, 22 Jun 2009 11:24:43 -0700
Subject: Revert "char: moxa, prevent opening unavailable ports"

This reverts commit a90b037583d5f1ae3e54e9c687c79df82d1d34a4, which
already got fixed as commit f0e8527726b9e56649b9eafde3bc0fbc4dd2dd47:
the same patch (trivial differences) got applied twice.

Requested-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/moxa.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/char/moxa.c b/drivers/char/moxa.c
index 6799588b0099..65b6ff2442c6 100644
--- a/drivers/char/moxa.c
+++ b/drivers/char/moxa.c
@@ -1189,11 +1189,6 @@ static int moxa_open(struct tty_struct *tty, struct file *filp)
 		return -ENODEV;
 	}
 
-	if (port % MAX_PORTS_PER_BOARD >= brd->numPorts) {
-		retval = -ENODEV;
-		goto out_unlock;
-	}
-
 	ch = &brd->ports[port % MAX_PORTS_PER_BOARD];
 	ch->port.count++;
 	tty->driver_data = ch;
@@ -1218,8 +1213,8 @@ static int moxa_open(struct tty_struct *tty, struct file *filp)
 				moxa_close_port(tty);
 	} else
 		ch->port.flags |= ASYNC_NORMAL_ACTIVE;
-out_unlock:
 	mutex_unlock(&moxa_openlock);
+
 	return retval;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From e2434dc1c19412639dd047a4d4eff8ed0e5d0d50 Mon Sep 17 00:00:00 2001
From: Jens Rottmann <JRottmann@LiPPERTEmbedded.de>
Date: Mon, 22 Jun 2009 16:51:49 +0100
Subject: parport_pc: after superio probing restore original register values

CONFIG_PARPORT_PC_SUPERIO probes for various superio chips by writing
byte sequences to a set of different potential I/O ranges.  But the
probed ranges are not exclusive to parallel ports.  Some of our boards
just happen to have a watchdog in one of them.  Took us almost a week
to figure out why some distros reboot without warning after running
flawlessly for 3 hours.  For exactly 170 = 0xAA minutes, that is ...

Fixed by restoring original values after probing.  Also fixed too small
request_region() in detect_and_report_it87().

Signed-off-by: Jens Rottmann <JRottmann@LiPPERTEmbedded.de>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Cc: <stable@kernel.org>
Acked-by: Jeff Garzik <jgarzik@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/parport/parport_pc.c | 31 +++++++++++++++++++++++++------
 1 file changed, 25 insertions(+), 6 deletions(-)

diff --git a/drivers/parport/parport_pc.c b/drivers/parport/parport_pc.c
index 151bf5bc8afe..7f1cca701c11 100644
--- a/drivers/parport/parport_pc.c
+++ b/drivers/parport/parport_pc.c
@@ -1471,11 +1471,13 @@ static void __devinit decode_smsc(int efer, int key, int devid, int devrev)
 
 static void __devinit winbond_check(int io, int key)
 {
-	int devid, devrev, oldid, x_devid, x_devrev, x_oldid;
+	int origval, devid, devrev, oldid, x_devid, x_devrev, x_oldid;
 
 	if (!request_region(io, 3, __func__))
 		return;
 
+	origval = inb(io); /* Save original value */
+
 	/* First probe without key */
 	outb(0x20, io);
 	x_devid = inb(io + 1);
@@ -1495,6 +1497,8 @@ static void __devinit winbond_check(int io, int key)
 	oldid = inb(io + 1);
 	outb(0xaa, io);    /* Magic Seal */
 
+	outb(origval, io); /* in case we poked some entirely different hardware */
+
 	if ((x_devid == devid) && (x_devrev == devrev) && (x_oldid == oldid))
 		goto out; /* protection against false positives */
 
@@ -1505,11 +1509,15 @@ out:
 
 static void __devinit winbond_check2(int io, int key)
 {
-	int devid, devrev, oldid, x_devid, x_devrev, x_oldid;
+	int origval[3], devid, devrev, oldid, x_devid, x_devrev, x_oldid;
 
 	if (!request_region(io, 3, __func__))
 		return;
 
+	origval[0] = inb(io); /* Save original values */
+	origval[1] = inb(io + 1);
+	origval[2] = inb(io + 2);
+
 	/* First probe without the key */
 	outb(0x20, io + 2);
 	x_devid = inb(io + 2);
@@ -1528,6 +1536,10 @@ static void __devinit winbond_check2(int io, int key)
 	oldid = inb(io + 2);
 	outb(0xaa, io);    /* Magic Seal */
 
+	outb(origval[0], io); /* in case we poked some entirely different hardware */
+	outb(origval[1], io + 1);
+	outb(origval[2], io + 2);
+
 	if (x_devid == devid && x_devrev == devrev && x_oldid == oldid)
 		goto out; /* protection against false positives */
 
@@ -1538,11 +1550,13 @@ out:
 
 static void __devinit smsc_check(int io, int key)
 {
-	int id, rev, oldid, oldrev, x_id, x_rev, x_oldid, x_oldrev;
+	int origval, id, rev, oldid, oldrev, x_id, x_rev, x_oldid, x_oldrev;
 
 	if (!request_region(io, 3, __func__))
 		return;
 
+	origval = inb(io); /* Save original value */
+
 	/* First probe without the key */
 	outb(0x0d, io);
 	x_oldid = inb(io + 1);
@@ -1566,6 +1580,8 @@ static void __devinit smsc_check(int io, int key)
 	rev = inb(io + 1);
 	outb(0xaa, io);    /* Magic Seal */
 
+	outb(origval, io); /* in case we poked some entirely different hardware */
+
 	if (x_id == id && x_oldrev == oldrev &&
 	    x_oldid == oldid && x_rev == rev)
 		goto out; /* protection against false positives */
@@ -1602,11 +1618,12 @@ static void __devinit detect_and_report_smsc(void)
 static void __devinit detect_and_report_it87(void)
 {
 	u16 dev;
-	u8 r;
+	u8 origval, r;
 	if (verbose_probing)
 		printk(KERN_DEBUG "IT8705 Super-IO detection, now testing port 2E ...\n");
-	if (!request_region(0x2e, 1, __func__))
+	if (!request_region(0x2e, 2, __func__))
 		return;
+	origval = inb(0x2e);		/* Save original value */
 	outb(0x87, 0x2e);
 	outb(0x01, 0x2e);
 	outb(0x55, 0x2e);
@@ -1626,8 +1643,10 @@ static void __devinit detect_and_report_it87(void)
 		outb(r | 8, 0x2F);
 		outb(0x02, 0x2E);	/* Lock */
 		outb(0x02, 0x2F);
+	} else {
+		outb(origval, 0x2e);	/* Oops, sorry to disturb */
 	}
-	release_region(0x2e, 1);
+	release_region(0x2e, 2);
 }
 #endif /* CONFIG_PARPORT_PC_SUPERIO */
 
-- 
cgit v1.2.3-59-g8ed1b


From dfa7c4d869b7d3d37b70f1de856f2901b6ebfcf0 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Mon, 22 Jun 2009 16:54:27 +0100
Subject: parport_pc: set properly the dma_mask for parport_pc device

parport_pc_probe_port() creates the own 'parport_pc' device if the
device argument is NULL. Then parport_pc_probe_port() doesn't
initialize the dma_mask and coherent_dma_mask of the device and calls
dma_alloc_coherent with it. dma_alloc_coherent fails because
dma_alloc_coherent() doesn't accept the uninitialized dma_mask:

http://lkml.org/lkml/2009/6/16/150

Long ago, X86_32 and X86_64 had the own dma_alloc_coherent
implementations; X86_32 accepted a device having dma_mask that is not
initialized however X86_64 didn't. When we merged them, we chose to
prohibit a device having dma_mask that is not initialized. I think
that it's good to require drivers to set up dma_mask (and
coherent_dma_mask) properly if the drivers want DMA.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Reported-by: Malcom Blaney <malcolm.blaney@maptek.com.au>
Tested-by: Malcom Blaney <malcolm.blaney@maptek.com.au>
Cc: stable@kernel.org
Signed-off-by: Alan Cox <alan@linux.intel.com>
Acked-by: Jeff Garzik <jgarzik@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/parport/parport_pc.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/parport/parport_pc.c b/drivers/parport/parport_pc.c
index 7f1cca701c11..1032d5fdbd42 100644
--- a/drivers/parport/parport_pc.c
+++ b/drivers/parport/parport_pc.c
@@ -2290,6 +2290,9 @@ struct parport *parport_pc_probe_port(unsigned long int base,
 		if (IS_ERR(pdev))
 			return NULL;
 		dev = &pdev->dev;
+
+		dev->coherent_dma_mask = DMA_BIT_MASK(24);
+		dev->dma_mask = &dev->coherent_dma_mask;
 	}
 
 	ops = kmalloc(sizeof(struct parport_operations), GFP_KERNEL);
-- 
cgit v1.2.3-59-g8ed1b


From 56578abfd16a1a7554f64000d5fc0a377d4dda6a Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Mon, 22 Jun 2009 18:31:10 +0100
Subject: bfin_jtag_comm: clean up printk usage

The original patch garned some feedback and a v2 was posted, but that
version seems to have been missed when merging the driver.

At any rate, this cleans up the printk usage as suggested by Jiri Slaby.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/bfin_jtag_comm.c | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/drivers/char/bfin_jtag_comm.c b/drivers/char/bfin_jtag_comm.c
index 44c113d56045..1d7c34c73b20 100644
--- a/drivers/char/bfin_jtag_comm.c
+++ b/drivers/char/bfin_jtag_comm.c
@@ -8,6 +8,10 @@
  * Licensed under the GPL-2 or later.
  */
 
+#define DRV_NAME "bfin-jtag-comm"
+#define DEV_NAME "ttyBFJC"
+#define pr_fmt(fmt) DRV_NAME ": " fmt
+
 #include <linux/circ_buf.h>
 #include <linux/console.h>
 #include <linux/delay.h>
@@ -22,18 +26,14 @@
 #include <linux/tty_flip.h>
 #include <asm/atomic.h>
 
+#define pr_init(fmt, args...) ({ static const __initconst char __fmt[] = fmt; printk(__fmt, ## args); })
+
 /* See the Debug/Emulation chapter in the HRM */
 #define EMUDOF   0x00000001	/* EMUDAT_OUT full & valid */
 #define EMUDIF   0x00000002	/* EMUDAT_IN full & valid */
 #define EMUDOOVF 0x00000004	/* EMUDAT_OUT overflow */
 #define EMUDIOVF 0x00000008	/* EMUDAT_IN overflow */
 
-#define DRV_NAME "bfin-jtag-comm"
-#define DEV_NAME "ttyBFJC"
-
-#define pr_init(fmt, args...) ({ static const __initdata char __fmt[] = fmt; printk(__fmt, ## args); })
-#define debug(fmt, args...) pr_debug(DRV_NAME ": " fmt, ## args)
-
 static inline uint32_t bfin_write_emudat(uint32_t emudat)
 {
 	__asm__ __volatile__("emudat = %0;" : : "d"(emudat));
@@ -74,7 +74,7 @@ bfin_jc_emudat_manager(void *arg)
 	while (!kthread_should_stop()) {
 		/* no one left to give data to, so sleep */
 		if (bfin_jc_tty == NULL && circ_empty(&bfin_jc_write_buf)) {
-			debug("waiting for readers\n");
+			pr_debug("waiting for readers\n");
 			__set_current_state(TASK_UNINTERRUPTIBLE);
 			schedule();
 			__set_current_state(TASK_RUNNING);
@@ -82,7 +82,7 @@ bfin_jc_emudat_manager(void *arg)
 
 		/* no data available, so just chill */
 		if (!(bfin_read_DBGSTAT() & EMUDIF) && circ_empty(&bfin_jc_write_buf)) {
-			debug("waiting for data (in_len = %i) (circ: %i %i)\n",
+			pr_debug("waiting for data (in_len = %i) (circ: %i %i)\n",
 				inbound_len, bfin_jc_write_buf.tail, bfin_jc_write_buf.head);
 			if (inbound_len)
 				schedule();
@@ -99,11 +99,11 @@ bfin_jc_emudat_manager(void *arg)
 			if (tty != NULL) {
 				uint32_t emudat = bfin_read_emudat();
 				if (inbound_len == 0) {
-					debug("incoming length: 0x%08x\n", emudat);
+					pr_debug("incoming length: 0x%08x\n", emudat);
 					inbound_len = emudat;
 				} else {
 					size_t num_chars = (4 <= inbound_len ? 4 : inbound_len);
-					debug("  incoming data: 0x%08x (pushing %zu)\n", emudat, num_chars);
+					pr_debug("  incoming data: 0x%08x (pushing %zu)\n", emudat, num_chars);
 					inbound_len -= num_chars;
 					tty_insert_flip_string(tty, (unsigned char *)&emudat, num_chars);
 					tty_flip_buffer_push(tty);
@@ -117,7 +117,7 @@ bfin_jc_emudat_manager(void *arg)
 			if (outbound_len == 0) {
 				outbound_len = circ_cnt(&bfin_jc_write_buf);
 				bfin_write_emudat(outbound_len);
-				debug("outgoing length: 0x%08x\n", outbound_len);
+				pr_debug("outgoing length: 0x%08x\n", outbound_len);
 			} else {
 				struct tty_struct *tty;
 				int tail = bfin_jc_write_buf.tail;
@@ -136,7 +136,7 @@ bfin_jc_emudat_manager(void *arg)
 				if (tty)
 					tty_wakeup(tty);
 				mutex_unlock(&bfin_jc_tty_mutex);
-				debug("  outgoing data: 0x%08x (pushing %zu)\n", emudat, ate);
+				pr_debug("  outgoing data: 0x%08x (pushing %zu)\n", emudat, ate);
 			}
 		}
 	}
@@ -149,7 +149,7 @@ static int
 bfin_jc_open(struct tty_struct *tty, struct file *filp)
 {
 	mutex_lock(&bfin_jc_tty_mutex);
-	debug("open %lu\n", bfin_jc_count);
+	pr_debug("open %lu\n", bfin_jc_count);
 	++bfin_jc_count;
 	bfin_jc_tty = tty;
 	wake_up_process(bfin_jc_kthread);
@@ -161,7 +161,7 @@ static void
 bfin_jc_close(struct tty_struct *tty, struct file *filp)
 {
 	mutex_lock(&bfin_jc_tty_mutex);
-	debug("close %lu\n", bfin_jc_count);
+	pr_debug("close %lu\n", bfin_jc_count);
 	if (--bfin_jc_count == 0)
 		bfin_jc_tty = NULL;
 	wake_up_process(bfin_jc_kthread);
@@ -174,7 +174,7 @@ bfin_jc_circ_write(const unsigned char *buf, int count)
 {
 	int i;
 	count = min(count, circ_free(&bfin_jc_write_buf));
-	debug("going to write chunk of %i bytes\n", count);
+	pr_debug("going to write chunk of %i bytes\n", count);
 	for (i = 0; i < count; ++i)
 		circ_byte(&bfin_jc_write_buf, bfin_jc_write_buf.head + i) = buf[i];
 	bfin_jc_write_buf.head += i;
-- 
cgit v1.2.3-59-g8ed1b


From 9c529a3d76dffae943868ebad07b042d15764712 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Mon, 22 Jun 2009 18:37:24 +0100
Subject: serial: bfin_5xx: add missing spin_lock init

The Blackfin serial driver never initialized the spin_lock that is part of
the serial core structure, but we never noticed because spin_lock's are
rarely enabled on UP systems.  Yeah lockdep and friends.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/bfin_5xx.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/serial/bfin_5xx.c b/drivers/serial/bfin_5xx.c
index e2f6b1bfac98..d7fcca18223b 100644
--- a/drivers/serial/bfin_5xx.c
+++ b/drivers/serial/bfin_5xx.c
@@ -1110,6 +1110,7 @@ static void __init bfin_serial_init_ports(void)
 	bfin_serial_hw_init();
 
 	for (i = 0; i < nr_active_ports; i++) {
+		spin_lock_init(&bfin_serial_ports[i].port.lock);
 		bfin_serial_ports[i].port.uartclk   = get_sclk();
 		bfin_serial_ports[i].port.fifosize  = BFIN_UART_TX_FIFO_SIZE;
 		bfin_serial_ports[i].port.ops       = &bfin_serial_pops;
-- 
cgit v1.2.3-59-g8ed1b


From 607c268ef9a4675287e77f732071e426e62c2d86 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Mon, 22 Jun 2009 18:41:47 +0100
Subject: serial: bfin_5xx: fix building as module when early printk is enabled

Since early printk only makes sense/works when the serial driver is built
into the kernel, disable the option for this driver when it is going to be
built as a module.  Otherwise we get build failures due to the ifdef
handling.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/bfin_5xx.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/serial/bfin_5xx.c b/drivers/serial/bfin_5xx.c
index d7fcca18223b..b4a7650af696 100644
--- a/drivers/serial/bfin_5xx.c
+++ b/drivers/serial/bfin_5xx.c
@@ -38,6 +38,10 @@
 #include <asm/cacheflush.h>
 #endif
 
+#ifdef CONFIG_SERIAL_BFIN_MODULE
+# undef CONFIG_EARLY_PRINTK
+#endif
+
 /* UART name and device definitions */
 #define BFIN_SERIAL_NAME	"ttyBF"
 #define BFIN_SERIAL_MAJOR	204
-- 
cgit v1.2.3-59-g8ed1b


From 52e3632ea603ef92757d5d0dedcd9fc8643445e3 Mon Sep 17 00:00:00 2001
From: Roel Kluin <roel.kluin@gmail.com>
Date: Mon, 22 Jun 2009 18:41:56 +0100
Subject: serial: fix off by one errors

In zs_console_putchar() occurs:

	if (zs_transmit_drain(zport, irq))
		write_zsdata(zport, ch);

However if in zs_transmit_drain() no empty Tx Buffer occurs, limit reaches
-1 => true, and the write still occurs.

This patch changes postfix to prefix decrements in this and similar
functions to prevent similar mistakes in the future.  This decreases the
iterations with one but the chosen loop count was arbitrary anyway.

In sunhv limit reaches -1, not 0, so the test is off by one.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Acked-by: David S. Miller <davem@davemloft.net>
Acked-by: Maciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/sb1250-duart.c | 6 +++---
 drivers/serial/sunhv.c        | 2 +-
 drivers/serial/zs.c           | 6 +++---
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/serial/sb1250-duart.c b/drivers/serial/sb1250-duart.c
index a4fb343a08da..319e8b83f6be 100644
--- a/drivers/serial/sb1250-duart.c
+++ b/drivers/serial/sb1250-duart.c
@@ -204,7 +204,7 @@ static int sbd_receive_drain(struct sbd_port *sport)
 {
 	int loops = 10000;
 
-	while (sbd_receive_ready(sport) && loops--)
+	while (sbd_receive_ready(sport) && --loops)
 		read_sbdchn(sport, R_DUART_RX_HOLD);
 	return loops;
 }
@@ -218,7 +218,7 @@ static int __maybe_unused sbd_transmit_drain(struct sbd_port *sport)
 {
 	int loops = 10000;
 
-	while (!sbd_transmit_ready(sport) && loops--)
+	while (!sbd_transmit_ready(sport) && --loops)
 		udelay(2);
 	return loops;
 }
@@ -232,7 +232,7 @@ static int sbd_line_drain(struct sbd_port *sport)
 {
 	int loops = 10000;
 
-	while (!sbd_transmit_empty(sport) && loops--)
+	while (!sbd_transmit_empty(sport) && --loops)
 		udelay(2);
 	return loops;
 }
diff --git a/drivers/serial/sunhv.c b/drivers/serial/sunhv.c
index a94a2ab4b571..1df5325faab2 100644
--- a/drivers/serial/sunhv.c
+++ b/drivers/serial/sunhv.c
@@ -461,7 +461,7 @@ static void sunhv_console_write_paged(struct console *con, const char *s, unsign
 					break;
 				udelay(1);
 			}
-			if (limit <= 0)
+			if (limit < 0)
 				break;
 			page_bytes -= written;
 			ra += written;
diff --git a/drivers/serial/zs.c b/drivers/serial/zs.c
index 9e6a873f8203..d8c2809b1ab6 100644
--- a/drivers/serial/zs.c
+++ b/drivers/serial/zs.c
@@ -231,7 +231,7 @@ static int zs_receive_drain(struct zs_port *zport)
 {
 	int loops = 10000;
 
-	while ((read_zsreg(zport, R0) & Rx_CH_AV) && loops--)
+	while ((read_zsreg(zport, R0) & Rx_CH_AV) && --loops)
 		read_zsdata(zport);
 	return loops;
 }
@@ -241,7 +241,7 @@ static int zs_transmit_drain(struct zs_port *zport, int irq)
 	struct zs_scc *scc = zport->scc;
 	int loops = 10000;
 
-	while (!(read_zsreg(zport, R0) & Tx_BUF_EMP) && loops--) {
+	while (!(read_zsreg(zport, R0) & Tx_BUF_EMP) && --loops) {
 		zs_spin_unlock_cond_irq(&scc->zlock, irq);
 		udelay(2);
 		zs_spin_lock_cond_irq(&scc->zlock, irq);
@@ -254,7 +254,7 @@ static int zs_line_drain(struct zs_port *zport, int irq)
 	struct zs_scc *scc = zport->scc;
 	int loops = 10000;
 
-	while (!(read_zsreg(zport, R1) & ALL_SNT) && loops--) {
+	while (!(read_zsreg(zport, R1) & ALL_SNT) && --loops) {
 		zs_spin_unlock_cond_irq(&scc->zlock, irq);
 		udelay(2);
 		zs_spin_lock_cond_irq(&scc->zlock, irq);
-- 
cgit v1.2.3-59-g8ed1b


From eca41044268887838fa122aa24475df8f23d614c Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Mon, 22 Jun 2009 18:42:03 +0100
Subject: n_r3964: fix lock imbalance

There is omitted BKunL in r3964_read.

Centralize the paths to one point with one unlock.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/n_r3964.c | 26 ++++++++++++++------------
 1 file changed, 14 insertions(+), 12 deletions(-)

diff --git a/drivers/char/n_r3964.c b/drivers/char/n_r3964.c
index d2e93e343226..2e99158ebb8a 100644
--- a/drivers/char/n_r3964.c
+++ b/drivers/char/n_r3964.c
@@ -1062,7 +1062,7 @@ static ssize_t r3964_read(struct tty_struct *tty, struct file *file,
 	struct r3964_client_info *pClient;
 	struct r3964_message *pMsg;
 	struct r3964_client_message theMsg;
-	int count;
+	int ret;
 
 	TRACE_L("read()");
 
@@ -1074,8 +1074,8 @@ static ssize_t r3964_read(struct tty_struct *tty, struct file *file,
 		if (pMsg == NULL) {
 			/* no messages available. */
 			if (file->f_flags & O_NONBLOCK) {
-				unlock_kernel();
-				return -EAGAIN;
+				ret = -EAGAIN;
+				goto unlock;
 			}
 			/* block until there is a message: */
 			wait_event_interruptible(pInfo->read_wait,
@@ -1085,29 +1085,31 @@ static ssize_t r3964_read(struct tty_struct *tty, struct file *file,
 		/* If we still haven't got a message, we must have been signalled */
 
 		if (!pMsg) {
-			unlock_kernel();
-			return -EINTR;
+			ret = -EINTR;
+			goto unlock;
 		}
 
 		/* deliver msg to client process: */
 		theMsg.msg_id = pMsg->msg_id;
 		theMsg.arg = pMsg->arg;
 		theMsg.error_code = pMsg->error_code;
-		count = sizeof(struct r3964_client_message);
+		ret = sizeof(struct r3964_client_message);
 
 		kfree(pMsg);
 		TRACE_M("r3964_read - msg kfree %p", pMsg);
 
-		if (copy_to_user(buf, &theMsg, count)) {
-			unlock_kernel();
-			return -EFAULT;
+		if (copy_to_user(buf, &theMsg, ret)) {
+			ret = -EFAULT;
+			goto unlock;
 		}
 
-		TRACE_PS("read - return %d", count);
-		return count;
+		TRACE_PS("read - return %d", ret);
+		goto unlock;
 	}
+	ret = -EPERM;
+unlock:
 	unlock_kernel();
-	return -EPERM;
+	return ret;
 }
 
 static ssize_t r3964_write(struct tty_struct *tty, struct file *file,
-- 
cgit v1.2.3-59-g8ed1b


From 69ae59d7d8df14413cf0a97b3e372d7dc8352563 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Mon, 22 Jun 2009 18:42:10 +0100
Subject: pcmcia/cm4000: fix lock imbalance

Don't return from switch/case, break instead, so that we unlock BKL.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/pcmcia/cm4000_cs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/char/pcmcia/cm4000_cs.c b/drivers/char/pcmcia/cm4000_cs.c
index dbb912574569..881934c068c8 100644
--- a/drivers/char/pcmcia/cm4000_cs.c
+++ b/drivers/char/pcmcia/cm4000_cs.c
@@ -1575,7 +1575,8 @@ static long cmm_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 		clear_bit(LOCK_IO, &dev->flags);
 		wake_up_interruptible(&dev->ioq);
 
-		return 0;
+		rc = 0;
+		break;
 	case CM_IOCSPTS:
 		{
 			struct ptsreq krnptsreq;
-- 
cgit v1.2.3-59-g8ed1b


From a115902f67ef51fbbe83e214fb761aaa9734c1ce Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Mon, 22 Jun 2009 18:42:18 +0100
Subject: vt_ioctl: fix lock imbalance

Don't return from switch/case directly in vt_ioctl. Set ret and break
instead so that we unlock BKL.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/vt_ioctl.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/char/vt_ioctl.c b/drivers/char/vt_ioctl.c
index e6ce632a393e..7539bed0f7e0 100644
--- a/drivers/char/vt_ioctl.c
+++ b/drivers/char/vt_ioctl.c
@@ -396,7 +396,8 @@ int vt_ioctl(struct tty_struct *tty, struct file * file,
 	kbd = kbd_table + console;
 	switch (cmd) {
 	case TIOCLINUX:
-		return tioclinux(tty, arg);
+		ret = tioclinux(tty, arg);
+		break;
 	case KIOCSOUND:
 		if (!perm)
 			goto eperm;
-- 
cgit v1.2.3-59-g8ed1b


From a6540f731d506d9e82444cf0020e716613d4c46c Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@linux.intel.com>
Date: Mon, 22 Jun 2009 18:42:29 +0100
Subject: ppp: Fix throttling bugs

The ppp layer goes around calling the unthrottle method from non sleeping
paths. This isn't safe because the unthrottle methods in the tty layer need
to be able to sleep (consider a USB dongle).

Until now this didn't show up because the ppp layer never actually throttled
a port so the unthrottle was always a no-op. Currently it's a mutex taking
path so warnings are spewed if the unthrottle occurs via certain paths.

Fix this by removing the unneccessary unthrottle calls.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/net/ppp_async.c   | 1 -
 drivers/net/ppp_synctty.c | 1 -
 2 files changed, 2 deletions(-)

diff --git a/drivers/net/ppp_async.c b/drivers/net/ppp_async.c
index 6de8399d6dd9..17c116bb332c 100644
--- a/drivers/net/ppp_async.c
+++ b/drivers/net/ppp_async.c
@@ -356,7 +356,6 @@ ppp_asynctty_receive(struct tty_struct *tty, const unsigned char *buf,
 	if (!skb_queue_empty(&ap->rqueue))
 		tasklet_schedule(&ap->tsk);
 	ap_put(ap);
-	tty_unthrottle(tty);
 }
 
 static void
diff --git a/drivers/net/ppp_synctty.c b/drivers/net/ppp_synctty.c
index d2fa2db13586..aa3d39f38e22 100644
--- a/drivers/net/ppp_synctty.c
+++ b/drivers/net/ppp_synctty.c
@@ -397,7 +397,6 @@ ppp_sync_receive(struct tty_struct *tty, const unsigned char *buf,
 	if (!skb_queue_empty(&ap->rqueue))
 		tasklet_schedule(&ap->tsk);
 	sp_put(ap);
-	tty_unthrottle(tty);
 }
 
 static void
-- 
cgit v1.2.3-59-g8ed1b


From 94362fd7fbad653c9517efa4aa7cd8fdadd527b1 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@linux.intel.com>
Date: Mon, 22 Jun 2009 18:42:36 +0100
Subject: tty: fix some bogns in the serqt_usb2 driver

Remove the replicated urban legends from the comments and fix a couple of
other silly calls

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/staging/serqt_usb2/serqt_usb2.c | 29 ++++++++++-------------------
 1 file changed, 10 insertions(+), 19 deletions(-)

diff --git a/drivers/staging/serqt_usb2/serqt_usb2.c b/drivers/staging/serqt_usb2/serqt_usb2.c
index 581232b719fd..90b29b564631 100644
--- a/drivers/staging/serqt_usb2/serqt_usb2.c
+++ b/drivers/staging/serqt_usb2/serqt_usb2.c
@@ -284,21 +284,12 @@ static void ProcessModemStatus(struct quatech_port *qt_port,
 	return;
 }
 
-static void ProcessRxChar(struct usb_serial_port *port, unsigned char Data)
+static void ProcessRxChar(struct tty_struct *tty, struct usb_serial_port *port,
+						unsigned char data)
 {
-	struct tty_struct *tty;
 	struct urb *urb = port->read_urb;
-	tty = tty_port_tty_get(&port->port);
-
-	/* if we insert more than TTY_FLIPBUF_SIZE characters, we drop them. */
-
-	if (tty && urb->actual_length) {
-		tty_buffer_request_room(tty, 1);
-		tty_insert_flip_string(tty, &Data, 1);
-		/* tty_flip_buffer_push(tty); */
-	}
-
-	return;
+	if (urb->actual_length)
+		tty_insert_flip_char(tty, data, TTY_NORMAL);
 }
 
 static void qt_write_bulk_callback(struct urb *urb)
@@ -435,8 +426,10 @@ static void qt_read_bulk_callback(struct urb *urb)
 				case 0xff:
 					dbg("No status sequence. \n");
 
-					ProcessRxChar(port, data[i]);
-					ProcessRxChar(port, data[i + 1]);
+					if (tty) {
+						ProcessRxChar(tty, port, data[i]);
+						ProcessRxChar(tty, port, data[i + 1]);
+					}
 					i += 2;
 					break;
 				}
@@ -444,10 +437,8 @@ static void qt_read_bulk_callback(struct urb *urb)
 					continue;
 			}
 
-			if (tty && urb->actual_length) {
-				tty_buffer_request_room(tty, 1);
-				tty_insert_flip_string(tty, (data + i), 1);
-			}
+			if (tty && urb->actual_length)
+				tty_insert_flip_char(tty, data[i], TTY_NORMAL);
 
 		}
 		tty_flip_buffer_push(tty);
-- 
cgit v1.2.3-59-g8ed1b


From 099d5270897606473d63091afcc63f53ee1894bc Mon Sep 17 00:00:00 2001
From: Kevin Hilman <khilman@deeprootsystems.com>
Date: Mon, 22 Jun 2009 18:42:42 +0100
Subject: serial@ add OMAP wakeup-enable register

Add the wakeup enable register to the list of OMAP-specific UART
registers.  This is to support forthcoming OMAP PM enhancements which
use the wakeup feature of the OMAP's 8250-based UART.

Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/serial_reg.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/linux/serial_reg.h b/include/linux/serial_reg.h
index 96c0d93fc2ca..850db2e80510 100644
--- a/include/linux/serial_reg.h
+++ b/include/linux/serial_reg.h
@@ -323,6 +323,7 @@
 #define UART_OMAP_MVER		0x14	/* Module version register */
 #define UART_OMAP_SYSC		0x15	/* System configuration register */
 #define UART_OMAP_SYSS		0x16	/* System status register */
+#define UART_OMAP_WER		0x17	/* Wake-up enable register */
 
 #endif /* _LINUX_SERIAL_REG_H */
 
-- 
cgit v1.2.3-59-g8ed1b


From 90ceb9644d7cdec00a90255473359a7e2bb537a9 Mon Sep 17 00:00:00 2001
From: Peter Korsgaard <jacmet@sunsite.dk>
Date: Mon, 22 Jun 2009 18:42:49 +0100
Subject: serial: samsung.c: mark s3c24xx_serial_remove as __devexit

Mark the remove function as __devexit so it gets eliminated in
CONFIG_HOTPLUG=n builds.  Saves ~100 bytes.

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Acked-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/s3c2400.c | 2 +-
 drivers/serial/s3c2410.c | 2 +-
 drivers/serial/s3c2412.c | 2 +-
 drivers/serial/s3c2440.c | 2 +-
 drivers/serial/s3c24a0.c | 2 +-
 drivers/serial/s3c6400.c | 2 +-
 drivers/serial/samsung.c | 2 +-
 drivers/serial/samsung.h | 2 +-
 8 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/serial/s3c2400.c b/drivers/serial/s3c2400.c
index 4873f2978bd2..fb00ed5296e6 100644
--- a/drivers/serial/s3c2400.c
+++ b/drivers/serial/s3c2400.c
@@ -78,7 +78,7 @@ static int s3c2400_serial_probe(struct platform_device *dev)
 
 static struct platform_driver s3c2400_serial_drv = {
 	.probe		= s3c2400_serial_probe,
-	.remove		= s3c24xx_serial_remove,
+	.remove		= __devexit_p(s3c24xx_serial_remove),
 	.driver		= {
 		.name	= "s3c2400-uart",
 		.owner	= THIS_MODULE,
diff --git a/drivers/serial/s3c2410.c b/drivers/serial/s3c2410.c
index 87c182ef71b8..b5d7cbcba2ae 100644
--- a/drivers/serial/s3c2410.c
+++ b/drivers/serial/s3c2410.c
@@ -90,7 +90,7 @@ static int s3c2410_serial_probe(struct platform_device *dev)
 
 static struct platform_driver s3c2410_serial_drv = {
 	.probe		= s3c2410_serial_probe,
-	.remove		= s3c24xx_serial_remove,
+	.remove		= __devexit_p(s3c24xx_serial_remove),
 	.driver		= {
 		.name	= "s3c2410-uart",
 		.owner	= THIS_MODULE,
diff --git a/drivers/serial/s3c2412.c b/drivers/serial/s3c2412.c
index fd017b375568..11dcb90bdfef 100644
--- a/drivers/serial/s3c2412.c
+++ b/drivers/serial/s3c2412.c
@@ -123,7 +123,7 @@ static int s3c2412_serial_probe(struct platform_device *dev)
 
 static struct platform_driver s3c2412_serial_drv = {
 	.probe		= s3c2412_serial_probe,
-	.remove		= s3c24xx_serial_remove,
+	.remove		= __devexit_p(s3c24xx_serial_remove),
 	.driver		= {
 		.name	= "s3c2412-uart",
 		.owner	= THIS_MODULE,
diff --git a/drivers/serial/s3c2440.c b/drivers/serial/s3c2440.c
index 29cbb0afef8e..06c5b0cc47a3 100644
--- a/drivers/serial/s3c2440.c
+++ b/drivers/serial/s3c2440.c
@@ -153,7 +153,7 @@ static int s3c2440_serial_probe(struct platform_device *dev)
 
 static struct platform_driver s3c2440_serial_drv = {
 	.probe		= s3c2440_serial_probe,
-	.remove		= s3c24xx_serial_remove,
+	.remove		= __devexit_p(s3c24xx_serial_remove),
 	.driver		= {
 		.name	= "s3c2440-uart",
 		.owner	= THIS_MODULE,
diff --git a/drivers/serial/s3c24a0.c b/drivers/serial/s3c24a0.c
index ebf2fd3c8f7d..786a067d62ac 100644
--- a/drivers/serial/s3c24a0.c
+++ b/drivers/serial/s3c24a0.c
@@ -94,7 +94,7 @@ static int s3c24a0_serial_probe(struct platform_device *dev)
 
 static struct platform_driver s3c24a0_serial_drv = {
 	.probe		= s3c24a0_serial_probe,
-	.remove		= s3c24xx_serial_remove,
+	.remove		= __devexit_p(s3c24xx_serial_remove),
 	.driver		= {
 		.name	= "s3c24a0-uart",
 		.owner	= THIS_MODULE,
diff --git a/drivers/serial/s3c6400.c b/drivers/serial/s3c6400.c
index 3e3785233682..48f1a3781f0d 100644
--- a/drivers/serial/s3c6400.c
+++ b/drivers/serial/s3c6400.c
@@ -124,7 +124,7 @@ static int s3c6400_serial_probe(struct platform_device *dev)
 
 static struct platform_driver s3c6400_serial_drv = {
 	.probe		= s3c6400_serial_probe,
-	.remove		= s3c24xx_serial_remove,
+	.remove		= __devexit_p(s3c24xx_serial_remove),
 	.driver		= {
 		.name	= "s3c6400-uart",
 		.owner	= THIS_MODULE,
diff --git a/drivers/serial/samsung.c b/drivers/serial/samsung.c
index 93b5d75db126..c8851a0db63a 100644
--- a/drivers/serial/samsung.c
+++ b/drivers/serial/samsung.c
@@ -1174,7 +1174,7 @@ int s3c24xx_serial_probe(struct platform_device *dev,
 
 EXPORT_SYMBOL_GPL(s3c24xx_serial_probe);
 
-int s3c24xx_serial_remove(struct platform_device *dev)
+int __devexit s3c24xx_serial_remove(struct platform_device *dev)
 {
 	struct uart_port *port = s3c24xx_dev_to_port(&dev->dev);
 
diff --git a/drivers/serial/samsung.h b/drivers/serial/samsung.h
index 7afb94843a08..d3fe315969f6 100644
--- a/drivers/serial/samsung.h
+++ b/drivers/serial/samsung.h
@@ -72,7 +72,7 @@ struct s3c24xx_uart_port {
 extern int s3c24xx_serial_probe(struct platform_device *dev,
 				struct s3c24xx_uart_info *uart);
 
-extern int s3c24xx_serial_remove(struct platform_device *dev);
+extern int __devexit s3c24xx_serial_remove(struct platform_device *dev);
 
 extern int s3c24xx_serial_initconsole(struct platform_driver *drv,
 				      struct s3c24xx_uart_info *uart);
-- 
cgit v1.2.3-59-g8ed1b


From be10eb7589337e5defbe214dae038a53dd21add8 Mon Sep 17 00:00:00 2001
From: Paul Fulghum <paulkf@microgate.com>
Date: Mon, 22 Jun 2009 18:42:56 +0100
Subject: tty: n_hdlc add buffer flushing

Add flush_buffer tty callback to flush rx buffers.
Add TCFLSH ioctl processing to flush tx buffers.
Increase default tx buffers from 1 to 3.
Remove unneeded flush_buffer call in open callback.
Remove vendor specific CVS version string.

Signed-off-by: Paul Fulghum <paulkf@microgate.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/n_hdlc.c | 46 +++++++++++++++++++++++++++++++++++++---------
 1 file changed, 37 insertions(+), 9 deletions(-)

diff --git a/drivers/char/n_hdlc.c b/drivers/char/n_hdlc.c
index 461ece591a5b..1c43c8cdee25 100644
--- a/drivers/char/n_hdlc.c
+++ b/drivers/char/n_hdlc.c
@@ -10,7 +10,6 @@
  *	Paul Mackerras <Paul.Mackerras@cs.anu.edu.au>
  *
  * Original release 01/11/99
- * $Id: n_hdlc.c,v 4.8 2003/05/06 21:18:51 paulkf Exp $
  *
  * This code is released under the GNU General Public License (GPL)
  *
@@ -79,7 +78,6 @@
  */
 
 #define HDLC_MAGIC 0x239e
-#define HDLC_VERSION "$Revision: 4.8 $"
 
 #include <linux/module.h>
 #include <linux/init.h>
@@ -114,7 +112,7 @@
 #define MAX_HDLC_FRAME_SIZE 65535 
 #define DEFAULT_RX_BUF_COUNT 10
 #define MAX_RX_BUF_COUNT 60
-#define DEFAULT_TX_BUF_COUNT 1
+#define DEFAULT_TX_BUF_COUNT 3
 
 struct n_hdlc_buf {
 	struct n_hdlc_buf *link;
@@ -199,6 +197,31 @@ static void n_hdlc_tty_wakeup(struct tty_struct *tty);
 #define tty2n_hdlc(tty)	((struct n_hdlc *) ((tty)->disc_data))
 #define n_hdlc2tty(n_hdlc)	((n_hdlc)->tty)
 
+static void flush_rx_queue(struct tty_struct *tty)
+{
+	struct n_hdlc *n_hdlc = tty2n_hdlc(tty);
+	struct n_hdlc_buf *buf;
+
+	while ((buf = n_hdlc_buf_get(&n_hdlc->rx_buf_list)))
+		n_hdlc_buf_put(&n_hdlc->rx_free_buf_list, buf);
+}
+
+static void flush_tx_queue(struct tty_struct *tty)
+{
+	struct n_hdlc *n_hdlc = tty2n_hdlc(tty);
+	struct n_hdlc_buf *buf;
+	unsigned long flags;
+
+	while ((buf = n_hdlc_buf_get(&n_hdlc->tx_buf_list)))
+		n_hdlc_buf_put(&n_hdlc->tx_free_buf_list, buf);
+ 	spin_lock_irqsave(&n_hdlc->tx_buf_list.spinlock, flags);
+	if (n_hdlc->tbuf) {
+		n_hdlc_buf_put(&n_hdlc->tx_free_buf_list, n_hdlc->tbuf);
+		n_hdlc->tbuf = NULL;
+	}
+	spin_unlock_irqrestore(&n_hdlc->tx_buf_list.spinlock, flags);
+}
+
 static struct tty_ldisc_ops n_hdlc_ldisc = {
 	.owner		= THIS_MODULE,
 	.magic		= TTY_LDISC_MAGIC,
@@ -211,6 +234,7 @@ static struct tty_ldisc_ops n_hdlc_ldisc = {
 	.poll		= n_hdlc_tty_poll,
 	.receive_buf	= n_hdlc_tty_receive,
 	.write_wakeup	= n_hdlc_tty_wakeup,
+	.flush_buffer   = flush_rx_queue,
 };
 
 /**
@@ -341,10 +365,7 @@ static int n_hdlc_tty_open (struct tty_struct *tty)
 	set_bit(TTY_NO_WRITE_SPLIT,&tty->flags);
 #endif
 	
-	/* Flush any pending characters in the driver and discipline. */
-	if (tty->ldisc->ops->flush_buffer)
-		tty->ldisc->ops->flush_buffer(tty);
-
+	/* flush receive data from driver */
 	tty_driver_flush_buffer(tty);
 		
 	if (debuglevel >= DEBUG_LEVEL_INFO)	
@@ -763,6 +784,14 @@ static int n_hdlc_tty_ioctl(struct tty_struct *tty, struct file *file,
 		error = put_user(count, (int __user *)arg);
 		break;
 
+	case TCFLSH:
+		switch (arg) {
+		case TCIOFLUSH:
+		case TCOFLUSH:
+			flush_tx_queue(tty);
+		}
+		/* fall through to default */
+
 	default:
 		error = n_tty_ioctl_helper(tty, file, cmd, arg);
 		break;
@@ -919,8 +948,7 @@ static struct n_hdlc_buf* n_hdlc_buf_get(struct n_hdlc_buf_list *list)
 }	/* end of n_hdlc_buf_get() */
 
 static char hdlc_banner[] __initdata =
-	KERN_INFO "HDLC line discipline: version " HDLC_VERSION
-	", maxframe=%u\n";
+	KERN_INFO "HDLC line discipline maxframe=%u\n";
 static char hdlc_register_ok[] __initdata =
 	KERN_INFO "N_HDLC line discipline registered.\n";
 static char hdlc_register_fail[] __initdata =
-- 
cgit v1.2.3-59-g8ed1b


From 2421c48bd74debb537de94c1bd15cbabab272aa1 Mon Sep 17 00:00:00 2001
From: Richard Röjfors <richard.rojfors.ext@mocean-labs.com>
Date: Mon, 22 Jun 2009 18:43:03 +0100
Subject: timbuart: Fix for tx_empty
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Hardware updated to support TX FIFO empty.

Signed-off-by: Richard Röjfors <richard.rojfors.ext@mocean-labs.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/timbuart.c | 50 +++++++++++++++++++++++------------------------
 1 file changed, 25 insertions(+), 25 deletions(-)

diff --git a/drivers/serial/timbuart.c b/drivers/serial/timbuart.c
index ac9e5d5f742e..063a313b755c 100644
--- a/drivers/serial/timbuart.c
+++ b/drivers/serial/timbuart.c
@@ -33,29 +33,29 @@ struct timbuart_port {
 	struct uart_port	port;
 	struct tasklet_struct	tasklet;
 	int			usedma;
-	u8			last_ier;
+	u32			last_ier;
 	struct platform_device  *dev;
 };
 
 static int baudrates[] = {9600, 19200, 38400, 57600, 115200, 230400, 460800,
 	921600, 1843200, 3250000};
 
-static void timbuart_mctrl_check(struct uart_port *port, u8 isr, u8 *ier);
+static void timbuart_mctrl_check(struct uart_port *port, u32 isr, u32 *ier);
 
 static irqreturn_t timbuart_handleinterrupt(int irq, void *devid);
 
 static void timbuart_stop_rx(struct uart_port *port)
 {
 	/* spin lock held by upper layer, disable all RX interrupts */
-	u8 ier = ioread8(port->membase + TIMBUART_IER) & ~RXFLAGS;
-	iowrite8(ier, port->membase + TIMBUART_IER);
+	u32 ier = ioread32(port->membase + TIMBUART_IER) & ~RXFLAGS;
+	iowrite32(ier, port->membase + TIMBUART_IER);
 }
 
 static void timbuart_stop_tx(struct uart_port *port)
 {
 	/* spinlock held by upper layer, disable TX interrupt */
-	u8 ier = ioread8(port->membase + TIMBUART_IER) & ~TXBAE;
-	iowrite8(ier, port->membase + TIMBUART_IER);
+	u32 ier = ioread32(port->membase + TIMBUART_IER) & ~TXBAE;
+	iowrite32(ier, port->membase + TIMBUART_IER);
 }
 
 static void timbuart_start_tx(struct uart_port *port)
@@ -72,14 +72,14 @@ static void timbuart_flush_buffer(struct uart_port *port)
 	u8 ctl = ioread8(port->membase + TIMBUART_CTRL) | TIMBUART_CTRL_FLSHTX;
 
 	iowrite8(ctl, port->membase + TIMBUART_CTRL);
-	iowrite8(TXBF, port->membase + TIMBUART_ISR);
+	iowrite32(TXBF, port->membase + TIMBUART_ISR);
 }
 
 static void timbuart_rx_chars(struct uart_port *port)
 {
 	struct tty_struct *tty = port->info->port.tty;
 
-	while (ioread8(port->membase + TIMBUART_ISR) & RXDP) {
+	while (ioread32(port->membase + TIMBUART_ISR) & RXDP) {
 		u8 ch = ioread8(port->membase + TIMBUART_RXFIFO);
 		port->icount.rx++;
 		tty_insert_flip_char(tty, ch, TTY_NORMAL);
@@ -97,7 +97,7 @@ static void timbuart_tx_chars(struct uart_port *port)
 {
 	struct circ_buf *xmit = &port->info->xmit;
 
-	while (!(ioread8(port->membase + TIMBUART_ISR) & TXBF) &&
+	while (!(ioread32(port->membase + TIMBUART_ISR) & TXBF) &&
 		!uart_circ_empty(xmit)) {
 		iowrite8(xmit->buf[xmit->tail],
 			port->membase + TIMBUART_TXFIFO);
@@ -114,7 +114,7 @@ static void timbuart_tx_chars(struct uart_port *port)
 		ioread8(port->membase + TIMBUART_BAUDRATE));
 }
 
-static void timbuart_handle_tx_port(struct uart_port *port, u8 isr, u8 *ier)
+static void timbuart_handle_tx_port(struct uart_port *port, u32 isr, u32 *ier)
 {
 	struct timbuart_port *uart =
 		container_of(port, struct timbuart_port, port);
@@ -129,7 +129,7 @@ static void timbuart_handle_tx_port(struct uart_port *port, u8 isr, u8 *ier)
 	if (isr & TXFLAGS) {
 		timbuart_tx_chars(port);
 		/* clear all TX interrupts */
-		iowrite8(TXFLAGS, port->membase + TIMBUART_ISR);
+		iowrite32(TXFLAGS, port->membase + TIMBUART_ISR);
 
 		if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
 			uart_write_wakeup(port);
@@ -148,7 +148,7 @@ static void timbuart_handle_tx_port(struct uart_port *port, u8 isr, u8 *ier)
 	dev_dbg(port->dev, "%s - leaving\n", __func__);
 }
 
-void timbuart_handle_rx_port(struct uart_port *port, u8 isr, u8 *ier)
+void timbuart_handle_rx_port(struct uart_port *port, u32 isr, u32 *ier)
 {
 	if (isr & RXFLAGS) {
 		/* Some RX status is set */
@@ -161,7 +161,7 @@ void timbuart_handle_rx_port(struct uart_port *port, u8 isr, u8 *ier)
 			timbuart_rx_chars(port);
 
 		/* ack all RX interrupts */
-		iowrite8(RXFLAGS, port->membase + TIMBUART_ISR);
+		iowrite32(RXFLAGS, port->membase + TIMBUART_ISR);
 	}
 
 	/* always have the RX interrupts enabled */
@@ -173,11 +173,11 @@ void timbuart_handle_rx_port(struct uart_port *port, u8 isr, u8 *ier)
 void timbuart_tasklet(unsigned long arg)
 {
 	struct timbuart_port *uart = (struct timbuart_port *)arg;
-	u8 isr, ier = 0;
+	u32 isr, ier = 0;
 
 	spin_lock(&uart->port.lock);
 
-	isr = ioread8(uart->port.membase + TIMBUART_ISR);
+	isr = ioread32(uart->port.membase + TIMBUART_ISR);
 	dev_dbg(uart->port.dev, "%s ISR: %x\n", __func__, isr);
 
 	if (!uart->usedma)
@@ -188,7 +188,7 @@ void timbuart_tasklet(unsigned long arg)
 	if (!uart->usedma)
 		timbuart_handle_rx_port(&uart->port, isr, &ier);
 
-	iowrite8(ier, uart->port.membase + TIMBUART_IER);
+	iowrite32(ier, uart->port.membase + TIMBUART_IER);
 
 	spin_unlock(&uart->port.lock);
 	dev_dbg(uart->port.dev, "%s leaving\n", __func__);
@@ -196,9 +196,9 @@ void timbuart_tasklet(unsigned long arg)
 
 static unsigned int timbuart_tx_empty(struct uart_port *port)
 {
-	u8 isr = ioread8(port->membase + TIMBUART_ISR);
+	u32 isr = ioread32(port->membase + TIMBUART_ISR);
 
-	return (isr & TXBAE) ? TIOCSER_TEMT : 0;
+	return (isr & TXBE) ? TIOCSER_TEMT : 0;
 }
 
 static unsigned int timbuart_get_mctrl(struct uart_port *port)
@@ -222,13 +222,13 @@ static void timbuart_set_mctrl(struct uart_port *port, unsigned int mctrl)
 		iowrite8(TIMBUART_CTRL_RTS, port->membase + TIMBUART_CTRL);
 }
 
-static void timbuart_mctrl_check(struct uart_port *port, u8 isr, u8 *ier)
+static void timbuart_mctrl_check(struct uart_port *port, u32 isr, u32 *ier)
 {
 	unsigned int cts;
 
 	if (isr & CTS_DELTA) {
 		/* ack */
-		iowrite8(CTS_DELTA, port->membase + TIMBUART_ISR);
+		iowrite32(CTS_DELTA, port->membase + TIMBUART_ISR);
 		cts = timbuart_get_mctrl(port);
 		uart_handle_cts_change(port, cts & TIOCM_CTS);
 		wake_up_interruptible(&port->info->delta_msr_wait);
@@ -255,9 +255,9 @@ static int timbuart_startup(struct uart_port *port)
 	dev_dbg(port->dev, "%s\n", __func__);
 
 	iowrite8(TIMBUART_CTRL_FLSHRX, port->membase + TIMBUART_CTRL);
-	iowrite8(0xff, port->membase + TIMBUART_ISR);
+	iowrite32(0x1ff, port->membase + TIMBUART_ISR);
 	/* Enable all but TX interrupts */
-	iowrite8(RXBAF | RXBF | RXTT | CTS_DELTA,
+	iowrite32(RXBAF | RXBF | RXTT | CTS_DELTA,
 		port->membase + TIMBUART_IER);
 
 	return request_irq(port->irq, timbuart_handleinterrupt, IRQF_SHARED,
@@ -270,7 +270,7 @@ static void timbuart_shutdown(struct uart_port *port)
 		container_of(port, struct timbuart_port, port);
 	dev_dbg(port->dev, "%s\n", __func__);
 	free_irq(port->irq, uart);
-	iowrite8(0, port->membase + TIMBUART_IER);
+	iowrite32(0, port->membase + TIMBUART_IER);
 }
 
 static int get_bindex(int baud)
@@ -359,10 +359,10 @@ static irqreturn_t timbuart_handleinterrupt(int irq, void *devid)
 	struct timbuart_port *uart = (struct timbuart_port *)devid;
 
 	if (ioread8(uart->port.membase + TIMBUART_IPR)) {
-		uart->last_ier = ioread8(uart->port.membase + TIMBUART_IER);
+		uart->last_ier = ioread32(uart->port.membase + TIMBUART_IER);
 
 		/* disable interrupts, the tasklet enables them again */
-		iowrite8(0, uart->port.membase + TIMBUART_IER);
+		iowrite32(0, uart->port.membase + TIMBUART_IER);
 
 		/* fire off bottom half */
 		tasklet_schedule(&uart->tasklet);
-- 
cgit v1.2.3-59-g8ed1b


From 04896a77a97b87e1611dedd61be88264ef4ac96c Mon Sep 17 00:00:00 2001
From: Robert Love <rlove@google.com>
Date: Mon, 22 Jun 2009 18:43:11 +0100
Subject: msm_serial: serial driver for MSM7K onboard serial peripheral.

Signed-off-by: Brian Swetland <swetland@google.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/Kconfig      |  10 +
 drivers/serial/Makefile     |   1 +
 drivers/serial/msm_serial.c | 767 ++++++++++++++++++++++++++++++++++++++++++++
 drivers/serial/msm_serial.h | 117 +++++++
 include/linux/serial_core.h |   3 +
 5 files changed, 898 insertions(+)
 create mode 100644 drivers/serial/msm_serial.c
 create mode 100644 drivers/serial/msm_serial.h

diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
index 1132c5cae7ab..037c1e0b7c4c 100644
--- a/drivers/serial/Kconfig
+++ b/drivers/serial/Kconfig
@@ -1320,6 +1320,16 @@ config SERIAL_SGI_IOC3
 	  If you have an SGI Altix with an IOC3 serial card,
 	  say Y or M.  Otherwise, say N.
 
+config SERIAL_MSM
+	bool "MSM on-chip serial port support"
+	depends on ARM && ARCH_MSM
+	select SERIAL_CORE
+
+config SERIAL_MSM_CONSOLE
+	bool "MSM serial console support"
+	depends on SERIAL_MSM=y
+	select SERIAL_CORE_CONSOLE
+
 config SERIAL_NETX
 	tristate "NetX serial port support"
 	depends on ARM && ARCH_NETX
diff --git a/drivers/serial/Makefile b/drivers/serial/Makefile
index 45a8658f54d5..d5a29981c6c4 100644
--- a/drivers/serial/Makefile
+++ b/drivers/serial/Makefile
@@ -71,6 +71,7 @@ obj-$(CONFIG_SERIAL_SGI_IOC4) += ioc4_serial.o
 obj-$(CONFIG_SERIAL_SGI_IOC3) += ioc3_serial.o
 obj-$(CONFIG_SERIAL_ATMEL) += atmel_serial.o
 obj-$(CONFIG_SERIAL_UARTLITE) += uartlite.o
+obj-$(CONFIG_SERIAL_MSM) += msm_serial.o
 obj-$(CONFIG_SERIAL_NETX) += netx-serial.o
 obj-$(CONFIG_SERIAL_OF_PLATFORM) += of_serial.o
 obj-$(CONFIG_SERIAL_OF_PLATFORM_NWPSERIAL) += nwpserial.o
diff --git a/drivers/serial/msm_serial.c b/drivers/serial/msm_serial.c
new file mode 100644
index 000000000000..1a7c856f76f8
--- /dev/null
+++ b/drivers/serial/msm_serial.c
@@ -0,0 +1,767 @@
+/*
+ * drivers/serial/msm_serial.c - driver for msm7k serial device and console
+ *
+ * Copyright (C) 2007 Google, Inc.
+ * Author: Robert Love <rlove@google.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#if defined(CONFIG_SERIAL_MSM_CONSOLE) && defined(CONFIG_MAGIC_SYSRQ)
+# define SUPPORT_SYSRQ
+#endif
+
+#include <linux/hrtimer.h>
+#include <linux/module.h>
+#include <linux/io.h>
+#include <linux/ioport.h>
+#include <linux/irq.h>
+#include <linux/init.h>
+#include <linux/console.h>
+#include <linux/tty.h>
+#include <linux/tty_flip.h>
+#include <linux/serial_core.h>
+#include <linux/serial.h>
+#include <linux/clk.h>
+#include <linux/platform_device.h>
+
+#include "msm_serial.h"
+
+struct msm_port {
+	struct uart_port	uart;
+	char			name[16];
+	struct clk		*clk;
+	unsigned int		imr;
+};
+
+#define UART_TO_MSM(uart_port)	((struct msm_port *) uart_port)
+
+static inline void msm_write(struct uart_port *port, unsigned int val,
+			     unsigned int off)
+{
+	__raw_writel(val, port->membase + off);
+}
+
+static inline unsigned int msm_read(struct uart_port *port, unsigned int off)
+{
+	return __raw_readl(port->membase + off);
+}
+
+static void msm_stop_tx(struct uart_port *port)
+{
+	struct msm_port *msm_port = UART_TO_MSM(port);
+
+	msm_port->imr &= ~UART_IMR_TXLEV;
+	msm_write(port, msm_port->imr, UART_IMR);
+}
+
+static void msm_start_tx(struct uart_port *port)
+{
+	struct msm_port *msm_port = UART_TO_MSM(port);
+
+	msm_port->imr |= UART_IMR_TXLEV;
+	msm_write(port, msm_port->imr, UART_IMR);
+}
+
+static void msm_stop_rx(struct uart_port *port)
+{
+	struct msm_port *msm_port = UART_TO_MSM(port);
+
+	msm_port->imr &= ~(UART_IMR_RXLEV | UART_IMR_RXSTALE);
+	msm_write(port, msm_port->imr, UART_IMR);
+}
+
+static void msm_enable_ms(struct uart_port *port)
+{
+	struct msm_port *msm_port = UART_TO_MSM(port);
+
+	msm_port->imr |= UART_IMR_DELTA_CTS;
+	msm_write(port, msm_port->imr, UART_IMR);
+}
+
+static void handle_rx(struct uart_port *port)
+{
+	struct tty_struct *tty = port->info->port.tty;
+	unsigned int sr;
+
+	/*
+	 * Handle overrun. My understanding of the hardware is that overrun
+	 * is not tied to the RX buffer, so we handle the case out of band.
+	 */
+	if ((msm_read(port, UART_SR) & UART_SR_OVERRUN)) {
+		port->icount.overrun++;
+		tty_insert_flip_char(tty, 0, TTY_OVERRUN);
+		msm_write(port, UART_CR_CMD_RESET_ERR, UART_CR);
+	}
+
+	/* and now the main RX loop */
+	while ((sr = msm_read(port, UART_SR)) & UART_SR_RX_READY) {
+		unsigned int c;
+		char flag = TTY_NORMAL;
+
+		c = msm_read(port, UART_RF);
+
+		if (sr & UART_SR_RX_BREAK) {
+			port->icount.brk++;
+			if (uart_handle_break(port))
+				continue;
+		} else if (sr & UART_SR_PAR_FRAME_ERR) {
+			port->icount.frame++;
+		} else {
+			port->icount.rx++;
+		}
+
+		/* Mask conditions we're ignorning. */
+		sr &= port->read_status_mask;
+
+		if (sr & UART_SR_RX_BREAK) {
+			flag = TTY_BREAK;
+		} else if (sr & UART_SR_PAR_FRAME_ERR) {
+			flag = TTY_FRAME;
+		}
+
+		if (!uart_handle_sysrq_char(port, c))
+			tty_insert_flip_char(tty, c, flag);
+	}
+
+	tty_flip_buffer_push(tty);
+}
+
+static void handle_tx(struct uart_port *port)
+{
+	struct circ_buf *xmit = &port->info->xmit;
+	struct msm_port *msm_port = UART_TO_MSM(port);
+	int sent_tx;
+
+	if (port->x_char) {
+		msm_write(port, port->x_char, UART_TF);
+		port->icount.tx++;
+		port->x_char = 0;
+	}
+
+	while (msm_read(port, UART_SR) & UART_SR_TX_READY) {
+		if (uart_circ_empty(xmit)) {
+			/* disable tx interrupts */
+			msm_port->imr &= ~UART_IMR_TXLEV;
+			msm_write(port, msm_port->imr, UART_IMR);
+			break;
+		}
+
+		msm_write(port, xmit->buf[xmit->tail], UART_TF);
+
+		xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
+		port->icount.tx++;
+		sent_tx = 1;
+	}
+
+	if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
+		uart_write_wakeup(port);
+}
+
+static void handle_delta_cts(struct uart_port *port)
+{
+	msm_write(port, UART_CR_CMD_RESET_CTS, UART_CR);
+	port->icount.cts++;
+	wake_up_interruptible(&port->info->delta_msr_wait);
+}
+
+static irqreturn_t msm_irq(int irq, void *dev_id)
+{
+	struct uart_port *port = dev_id;
+	struct msm_port *msm_port = UART_TO_MSM(port);
+	unsigned int misr;
+
+	spin_lock(&port->lock);
+	misr = msm_read(port, UART_MISR);
+	msm_write(port, 0, UART_IMR); /* disable interrupt */
+
+	if (misr & (UART_IMR_RXLEV | UART_IMR_RXSTALE))
+		handle_rx(port);
+	if (misr & UART_IMR_TXLEV)
+		handle_tx(port);
+	if (misr & UART_IMR_DELTA_CTS)
+		handle_delta_cts(port);
+
+	msm_write(port, msm_port->imr, UART_IMR); /* restore interrupt */
+	spin_unlock(&port->lock);
+
+	return IRQ_HANDLED;
+}
+
+static unsigned int msm_tx_empty(struct uart_port *port)
+{
+	return (msm_read(port, UART_SR) & UART_SR_TX_EMPTY) ? TIOCSER_TEMT : 0;
+}
+
+static unsigned int msm_get_mctrl(struct uart_port *port)
+{
+	return TIOCM_CAR | TIOCM_CTS | TIOCM_DSR | TIOCM_RTS;
+}
+
+static void msm_set_mctrl(struct uart_port *port, unsigned int mctrl)
+{
+	unsigned int mr;
+
+	mr = msm_read(port, UART_MR1);
+
+	if (!(mctrl & TIOCM_RTS)) {
+		mr &= ~UART_MR1_RX_RDY_CTL;
+		msm_write(port, mr, UART_MR1);
+		msm_write(port, UART_CR_CMD_RESET_RFR, UART_CR);
+	} else {
+		mr |= UART_MR1_RX_RDY_CTL;
+		msm_write(port, mr, UART_MR1);
+	}
+}
+
+static void msm_break_ctl(struct uart_port *port, int break_ctl)
+{
+	if (break_ctl)
+		msm_write(port, UART_CR_CMD_START_BREAK, UART_CR);
+	else
+		msm_write(port, UART_CR_CMD_STOP_BREAK, UART_CR);
+}
+
+static void msm_set_baud_rate(struct uart_port *port, unsigned int baud)
+{
+	unsigned int baud_code, rxstale, watermark;
+
+	switch (baud) {
+	case 300:
+		baud_code = UART_CSR_300;
+		rxstale = 1;
+		break;
+	case 600:
+		baud_code = UART_CSR_600;
+		rxstale = 1;
+		break;
+	case 1200:
+		baud_code = UART_CSR_1200;
+		rxstale = 1;
+		break;
+	case 2400:
+		baud_code = UART_CSR_2400;
+		rxstale = 1;
+		break;
+	case 4800:
+		baud_code = UART_CSR_4800;
+		rxstale = 1;
+		break;
+	case 9600:
+		baud_code = UART_CSR_9600;
+		rxstale = 2;
+		break;
+	case 14400:
+		baud_code = UART_CSR_14400;
+		rxstale = 3;
+		break;
+	case 19200:
+		baud_code = UART_CSR_19200;
+		rxstale = 4;
+		break;
+	case 28800:
+		baud_code = UART_CSR_28800;
+		rxstale = 6;
+		break;
+	case 38400:
+		baud_code = UART_CSR_38400;
+		rxstale = 8;
+		break;
+	case 57600:
+		baud_code = UART_CSR_57600;
+		rxstale = 16;
+		break;
+	case 115200:
+	default:
+		baud_code = UART_CSR_115200;
+		rxstale = 31;
+		break;
+	}
+
+	msm_write(port, baud_code, UART_CSR);
+
+	/* RX stale watermark */
+	watermark = UART_IPR_STALE_LSB & rxstale;
+	watermark |= UART_IPR_RXSTALE_LAST;
+	watermark |= UART_IPR_STALE_TIMEOUT_MSB & (rxstale << 2);
+	msm_write(port, watermark, UART_IPR);
+
+	/* set RX watermark */
+	watermark = (port->fifosize * 3) / 4;
+	msm_write(port, watermark, UART_RFWR);
+
+	/* set TX watermark */
+	msm_write(port, 10, UART_TFWR);
+}
+
+static void msm_reset(struct uart_port *port)
+{
+	/* reset everything */
+	msm_write(port, UART_CR_CMD_RESET_RX, UART_CR);
+	msm_write(port, UART_CR_CMD_RESET_TX, UART_CR);
+	msm_write(port, UART_CR_CMD_RESET_ERR, UART_CR);
+	msm_write(port, UART_CR_CMD_RESET_BREAK_INT, UART_CR);
+	msm_write(port, UART_CR_CMD_RESET_CTS, UART_CR);
+	msm_write(port, UART_CR_CMD_SET_RFR, UART_CR);
+}
+
+static void msm_init_clock(struct uart_port *port)
+{
+	struct msm_port *msm_port = UART_TO_MSM(port);
+
+	clk_enable(msm_port->clk);
+
+	msm_write(port, 0xC0, UART_MREG);
+	msm_write(port, 0xB2, UART_NREG);
+	msm_write(port, 0x7D, UART_DREG);
+	msm_write(port, 0x1C, UART_MNDREG);
+}
+
+static int msm_startup(struct uart_port *port)
+{
+	struct msm_port *msm_port = UART_TO_MSM(port);
+	unsigned int data, rfr_level;
+	int ret;
+
+	snprintf(msm_port->name, sizeof(msm_port->name),
+		 "msm_serial%d", port->line);
+
+	ret = request_irq(port->irq, msm_irq, IRQF_TRIGGER_HIGH,
+			  msm_port->name, port);
+	if (unlikely(ret))
+		return ret;
+
+	msm_init_clock(port);
+
+	if (likely(port->fifosize > 12))
+		rfr_level = port->fifosize - 12;
+	else
+		rfr_level = port->fifosize;
+
+	/* set automatic RFR level */
+	data = msm_read(port, UART_MR1);
+	data &= ~UART_MR1_AUTO_RFR_LEVEL1;
+	data &= ~UART_MR1_AUTO_RFR_LEVEL0;
+	data |= UART_MR1_AUTO_RFR_LEVEL1 & (rfr_level << 2);
+	data |= UART_MR1_AUTO_RFR_LEVEL0 & rfr_level;
+	msm_write(port, data, UART_MR1);
+
+	/* make sure that RXSTALE count is non-zero */
+	data = msm_read(port, UART_IPR);
+	if (unlikely(!data)) {
+		data |= UART_IPR_RXSTALE_LAST;
+		data |= UART_IPR_STALE_LSB;
+		msm_write(port, data, UART_IPR);
+	}
+
+	msm_reset(port);
+
+	msm_write(port, 0x05, UART_CR);	/* enable TX & RX */
+
+	/* turn on RX and CTS interrupts */
+	msm_port->imr = UART_IMR_RXLEV | UART_IMR_RXSTALE |
+			UART_IMR_CURRENT_CTS;
+	msm_write(port, msm_port->imr, UART_IMR);
+
+	return 0;
+}
+
+static void msm_shutdown(struct uart_port *port)
+{
+	struct msm_port *msm_port = UART_TO_MSM(port);
+
+	msm_port->imr = 0;
+	msm_write(port, 0, UART_IMR); /* disable interrupts */
+
+	clk_disable(msm_port->clk);
+
+	free_irq(port->irq, port);
+}
+
+static void msm_set_termios(struct uart_port *port, struct ktermios *termios,
+			    struct ktermios *old)
+{
+	unsigned long flags;
+	unsigned int baud, mr;
+
+	spin_lock_irqsave(&port->lock, flags);
+
+	/* calculate and set baud rate */
+	baud = uart_get_baud_rate(port, termios, old, 300, 115200);
+	msm_set_baud_rate(port, baud);
+
+	/* calculate parity */
+	mr = msm_read(port, UART_MR2);
+	mr &= ~UART_MR2_PARITY_MODE;
+	if (termios->c_cflag & PARENB) {
+		if (termios->c_cflag & PARODD)
+			mr |= UART_MR2_PARITY_MODE_ODD;
+		else if (termios->c_cflag & CMSPAR)
+			mr |= UART_MR2_PARITY_MODE_SPACE;
+		else
+			mr |= UART_MR2_PARITY_MODE_EVEN;
+	}
+
+	/* calculate bits per char */
+	mr &= ~UART_MR2_BITS_PER_CHAR;
+	switch (termios->c_cflag & CSIZE) {
+	case CS5:
+		mr |= UART_MR2_BITS_PER_CHAR_5;
+		break;
+	case CS6:
+		mr |= UART_MR2_BITS_PER_CHAR_6;
+		break;
+	case CS7:
+		mr |= UART_MR2_BITS_PER_CHAR_7;
+		break;
+	case CS8:
+	default:
+		mr |= UART_MR2_BITS_PER_CHAR_8;
+		break;
+	}
+
+	/* calculate stop bits */
+	mr &= ~(UART_MR2_STOP_BIT_LEN_ONE | UART_MR2_STOP_BIT_LEN_TWO);
+	if (termios->c_cflag & CSTOPB)
+		mr |= UART_MR2_STOP_BIT_LEN_TWO;
+	else
+		mr |= UART_MR2_STOP_BIT_LEN_ONE;
+
+	/* set parity, bits per char, and stop bit */
+	msm_write(port, mr, UART_MR2);
+
+	/* calculate and set hardware flow control */
+	mr = msm_read(port, UART_MR1);
+	mr &= ~(UART_MR1_CTS_CTL | UART_MR1_RX_RDY_CTL);
+	if (termios->c_cflag & CRTSCTS) {
+		mr |= UART_MR1_CTS_CTL;
+		mr |= UART_MR1_RX_RDY_CTL;
+	}
+	msm_write(port, mr, UART_MR1);
+
+	/* Configure status bits to ignore based on termio flags. */
+	port->read_status_mask = 0;
+	if (termios->c_iflag & INPCK)
+		port->read_status_mask |= UART_SR_PAR_FRAME_ERR;
+	if (termios->c_iflag & (BRKINT | PARMRK))
+		port->read_status_mask |= UART_SR_RX_BREAK;
+
+	uart_update_timeout(port, termios->c_cflag, baud);
+
+	spin_unlock_irqrestore(&port->lock, flags);
+}
+
+static const char *msm_type(struct uart_port *port)
+{
+	return "MSM";
+}
+
+static void msm_release_port(struct uart_port *port)
+{
+	struct platform_device *pdev = to_platform_device(port->dev);
+	struct resource *resource;
+	resource_size_t size;
+
+	resource = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (unlikely(!resource))
+		return;
+	size = resource->end - resource->start + 1;
+
+	release_mem_region(port->mapbase, size);
+	iounmap(port->membase);
+	port->membase = NULL;
+}
+
+static int msm_request_port(struct uart_port *port)
+{
+	struct platform_device *pdev = to_platform_device(port->dev);
+	struct resource *resource;
+	resource_size_t size;
+
+	resource = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (unlikely(!resource))
+		return -ENXIO;
+	size = resource->end - resource->start + 1;
+
+	if (unlikely(!request_mem_region(port->mapbase, size, "msm_serial")))
+		return -EBUSY;
+
+	port->membase = ioremap(port->mapbase, size);
+	if (!port->membase) {
+		release_mem_region(port->mapbase, size);
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
+static void msm_config_port(struct uart_port *port, int flags)
+{
+	if (flags & UART_CONFIG_TYPE) {
+		port->type = PORT_MSM;
+		msm_request_port(port);
+	}
+}
+
+static int msm_verify_port(struct uart_port *port, struct serial_struct *ser)
+{
+	if (unlikely(ser->type != PORT_UNKNOWN && ser->type != PORT_MSM))
+		return -EINVAL;
+	if (unlikely(port->irq != ser->irq))
+		return -EINVAL;
+	return 0;
+}
+
+static void msm_power(struct uart_port *port, unsigned int state,
+		      unsigned int oldstate)
+{
+	struct msm_port *msm_port = UART_TO_MSM(port);
+
+	switch (state) {
+	case 0:
+		clk_enable(msm_port->clk);
+		break;
+	case 3:
+		clk_disable(msm_port->clk);
+		break;
+	default:
+		printk(KERN_ERR "msm_serial: Unknown PM state %d\n", state);
+	}
+}
+
+static struct uart_ops msm_uart_pops = {
+	.tx_empty = msm_tx_empty,
+	.set_mctrl = msm_set_mctrl,
+	.get_mctrl = msm_get_mctrl,
+	.stop_tx = msm_stop_tx,
+	.start_tx = msm_start_tx,
+	.stop_rx = msm_stop_rx,
+	.enable_ms = msm_enable_ms,
+	.break_ctl = msm_break_ctl,
+	.startup = msm_startup,
+	.shutdown = msm_shutdown,
+	.set_termios = msm_set_termios,
+	.type = msm_type,
+	.release_port = msm_release_port,
+	.request_port = msm_request_port,
+	.config_port = msm_config_port,
+	.verify_port = msm_verify_port,
+	.pm = msm_power,
+};
+
+static struct msm_port msm_uart_ports[] = {
+	{
+		.uart = {
+			.iotype = UPIO_MEM,
+			.ops = &msm_uart_pops,
+			.flags = UPF_BOOT_AUTOCONF,
+			.fifosize = 512,
+			.line = 0,
+		},
+	},
+	{
+		.uart = {
+			.iotype = UPIO_MEM,
+			.ops = &msm_uart_pops,
+			.flags = UPF_BOOT_AUTOCONF,
+			.fifosize = 512,
+			.line = 1,
+		},
+	},
+	{
+		.uart = {
+			.iotype = UPIO_MEM,
+			.ops = &msm_uart_pops,
+			.flags = UPF_BOOT_AUTOCONF,
+			.fifosize = 64,
+			.line = 2,
+		},
+	},
+};
+
+#define UART_NR	ARRAY_SIZE(msm_uart_ports)
+
+static inline struct uart_port *get_port_from_line(unsigned int line)
+{
+	return &msm_uart_ports[line].uart;
+}
+
+#ifdef CONFIG_SERIAL_MSM_CONSOLE
+
+static void msm_console_putchar(struct uart_port *port, int c)
+{
+	while (!(msm_read(port, UART_SR) & UART_SR_TX_READY))
+		;
+	msm_write(port, c, UART_TF);
+}
+
+static void msm_console_write(struct console *co, const char *s,
+			      unsigned int count)
+{
+	struct uart_port *port;
+	struct msm_port *msm_port;
+
+	BUG_ON(co->index < 0 || co->index >= UART_NR);
+
+	port = get_port_from_line(co->index);
+	msm_port = UART_TO_MSM(port);
+
+	spin_lock(&port->lock);
+	uart_console_write(port, s, count, msm_console_putchar);
+	spin_unlock(&port->lock);
+}
+
+static int __init msm_console_setup(struct console *co, char *options)
+{
+	struct uart_port *port;
+	int baud, flow, bits, parity;
+
+	if (unlikely(co->index >= UART_NR || co->index < 0))
+		return -ENXIO;
+
+	port = get_port_from_line(co->index);
+
+	if (unlikely(!port->membase))
+		return -ENXIO;
+
+	port->cons = co;
+
+	msm_init_clock(port);
+
+	if (options)
+		uart_parse_options(options, &baud, &parity, &bits, &flow);
+
+	bits = 8;
+	parity = 'n';
+	flow = 'n';
+	msm_write(port, UART_MR2_BITS_PER_CHAR_8 | UART_MR2_STOP_BIT_LEN_ONE,
+		  UART_MR2);	/* 8N1 */
+
+	if (baud < 300 || baud > 115200)
+		baud = 115200;
+	msm_set_baud_rate(port, baud);
+
+	msm_reset(port);
+
+	printk(KERN_INFO "msm_serial: console setup on port #%d\n", port->line);
+
+	return uart_set_options(port, co, baud, parity, bits, flow);
+}
+
+static struct uart_driver msm_uart_driver;
+
+static struct console msm_console = {
+	.name = "ttyMSM",
+	.write = msm_console_write,
+	.device = uart_console_device,
+	.setup = msm_console_setup,
+	.flags = CON_PRINTBUFFER,
+	.index = -1,
+	.data = &msm_uart_driver,
+};
+
+#define MSM_CONSOLE	(&msm_console)
+
+#else
+#define MSM_CONSOLE	NULL
+#endif
+
+static struct uart_driver msm_uart_driver = {
+	.owner = THIS_MODULE,
+	.driver_name = "msm_serial",
+	.dev_name = "ttyMSM",
+	.nr = UART_NR,
+	.cons = MSM_CONSOLE,
+};
+
+static int __init msm_serial_probe(struct platform_device *pdev)
+{
+	struct msm_port *msm_port;
+	struct resource *resource;
+	struct uart_port *port;
+
+	if (unlikely(pdev->id < 0 || pdev->id >= UART_NR))
+		return -ENXIO;
+
+	printk(KERN_INFO "msm_serial: detected port #%d\n", pdev->id);
+
+	port = get_port_from_line(pdev->id);
+	port->dev = &pdev->dev;
+	msm_port = UART_TO_MSM(port);
+
+	msm_port->clk = clk_get(&pdev->dev, "uart_clk");
+	if (unlikely(IS_ERR(msm_port->clk)))
+		return PTR_ERR(msm_port->clk);
+	port->uartclk = clk_get_rate(msm_port->clk);
+
+	resource = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (unlikely(!resource))
+		return -ENXIO;
+	port->mapbase = resource->start;
+
+	port->irq = platform_get_irq(pdev, 0);
+	if (unlikely(port->irq < 0))
+		return -ENXIO;
+
+	platform_set_drvdata(pdev, port);
+
+	return uart_add_one_port(&msm_uart_driver, port);
+}
+
+static int __devexit msm_serial_remove(struct platform_device *pdev)
+{
+	struct msm_port *msm_port = platform_get_drvdata(pdev);
+
+	clk_put(msm_port->clk);
+
+	return 0;
+}
+
+static struct platform_driver msm_platform_driver = {
+	.probe = msm_serial_probe,
+	.remove = msm_serial_remove,
+	.driver = {
+		.name = "msm_serial",
+		.owner = THIS_MODULE,
+	},
+};
+
+static int __init msm_serial_init(void)
+{
+	int ret;
+
+	ret = uart_register_driver(&msm_uart_driver);
+	if (unlikely(ret))
+		return ret;
+
+	ret = platform_driver_probe(&msm_platform_driver, msm_serial_probe);
+	if (unlikely(ret))
+		uart_unregister_driver(&msm_uart_driver);
+
+	printk(KERN_INFO "msm_serial: driver initialized\n");
+
+	return ret;
+}
+
+static void __exit msm_serial_exit(void)
+{
+#ifdef CONFIG_SERIAL_MSM_CONSOLE
+	unregister_console(&msm_console);
+#endif
+	platform_driver_unregister(&msm_platform_driver);
+	uart_unregister_driver(&msm_uart_driver);
+}
+
+module_init(msm_serial_init);
+module_exit(msm_serial_exit);
+
+MODULE_AUTHOR("Robert Love <rlove@google.com>");
+MODULE_DESCRIPTION("Driver for msm7x serial device");
+MODULE_LICENSE("GPL");
diff --git a/drivers/serial/msm_serial.h b/drivers/serial/msm_serial.h
new file mode 100644
index 000000000000..689f1fa0e84e
--- /dev/null
+++ b/drivers/serial/msm_serial.h
@@ -0,0 +1,117 @@
+/*
+ * drivers/serial/msm_serial.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ * Author: Robert Love <rlove@google.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef __DRIVERS_SERIAL_MSM_SERIAL_H
+#define __DRIVERS_SERIAL_MSM_SERIAL_H
+
+#define UART_MR1			0x0000
+
+#define UART_MR1_AUTO_RFR_LEVEL0	0x3F
+#define UART_MR1_AUTO_RFR_LEVEL1	0x3FF00
+#define UART_MR1_RX_RDY_CTL    		(1 << 7)
+#define UART_MR1_CTS_CTL       		(1 << 6)
+
+#define UART_MR2			0x0004
+#define UART_MR2_ERROR_MODE		(1 << 6)
+#define UART_MR2_BITS_PER_CHAR		0x30
+#define UART_MR2_BITS_PER_CHAR_5	(0x0 << 4)
+#define UART_MR2_BITS_PER_CHAR_6	(0x1 << 4)
+#define UART_MR2_BITS_PER_CHAR_7	(0x2 << 4)
+#define UART_MR2_BITS_PER_CHAR_8	(0x3 << 4)
+#define UART_MR2_STOP_BIT_LEN_ONE	(0x1 << 2)
+#define UART_MR2_STOP_BIT_LEN_TWO	(0x3 << 2)
+#define UART_MR2_PARITY_MODE_NONE	0x0
+#define UART_MR2_PARITY_MODE_ODD	0x1
+#define UART_MR2_PARITY_MODE_EVEN	0x2
+#define UART_MR2_PARITY_MODE_SPACE	0x3
+#define UART_MR2_PARITY_MODE		0x3
+
+#define UART_CSR	0x0008
+#define UART_CSR_115200	0xFF
+#define UART_CSR_57600	0xEE
+#define UART_CSR_38400	0xDD
+#define UART_CSR_28800	0xCC
+#define UART_CSR_19200	0xBB
+#define UART_CSR_14400	0xAA
+#define UART_CSR_9600	0x99
+#define UART_CSR_4800	0x77
+#define UART_CSR_2400	0x55
+#define UART_CSR_1200	0x44
+#define UART_CSR_600	0x33
+#define UART_CSR_300	0x22
+
+#define UART_TF		0x000C
+
+#define UART_CR				0x0010
+#define UART_CR_CMD_NULL		(0 << 4)
+#define UART_CR_CMD_RESET_RX		(1 << 4)
+#define UART_CR_CMD_RESET_TX		(2 << 4)
+#define UART_CR_CMD_RESET_ERR		(3 << 4)
+#define UART_CR_CMD_RESET_BREAK_INT	(4 << 4)
+#define UART_CR_CMD_START_BREAK		(5 << 4)
+#define UART_CR_CMD_STOP_BREAK		(6 << 4)
+#define UART_CR_CMD_RESET_CTS		(7 << 4)
+#define UART_CR_CMD_PACKET_MODE		(9 << 4)
+#define UART_CR_CMD_MODE_RESET		(12 << 4)
+#define UART_CR_CMD_SET_RFR		(13 << 4)
+#define UART_CR_CMD_RESET_RFR		(14 << 4)
+#define UART_CR_TX_DISABLE		(1 << 3)
+#define UART_CR_TX_ENABLE		(1 << 3)
+#define UART_CR_RX_DISABLE		(1 << 3)
+#define UART_CR_RX_ENABLE		(1 << 3)
+
+#define UART_IMR		0x0014
+#define UART_IMR_TXLEV		(1 << 0)
+#define UART_IMR_RXSTALE	(1 << 3)
+#define UART_IMR_RXLEV		(1 << 4)
+#define UART_IMR_DELTA_CTS	(1 << 5)
+#define UART_IMR_CURRENT_CTS	(1 << 6)
+
+#define UART_IPR_RXSTALE_LAST		0x20
+#define UART_IPR_STALE_LSB		0x1F
+#define UART_IPR_STALE_TIMEOUT_MSB	0x3FF80
+
+#define UART_IPR	0x0018
+#define UART_TFWR	0x001C
+#define UART_RFWR	0x0020
+#define UART_HCR	0x0024
+
+#define UART_MREG		0x0028
+#define UART_NREG		0x002C
+#define UART_DREG		0x0030
+#define UART_MNDREG		0x0034
+#define UART_IRDA		0x0038
+#define UART_MISR_MODE		0x0040
+#define UART_MISR_RESET		0x0044
+#define UART_MISR_EXPORT	0x0048
+#define UART_MISR_VAL		0x004C
+#define UART_TEST_CTRL		0x0050
+
+#define UART_SR			0x0008
+#define UART_SR_HUNT_CHAR	(1 << 7)
+#define UART_SR_RX_BREAK	(1 << 6)
+#define UART_SR_PAR_FRAME_ERR	(1 << 5)
+#define UART_SR_OVERRUN		(1 << 4)
+#define UART_SR_TX_EMPTY	(1 << 3)
+#define UART_SR_TX_READY	(1 << 2)
+#define UART_SR_RX_FULL		(1 << 1)
+#define UART_SR_RX_READY	(1 << 0)
+
+#define UART_RF		0x000C
+#define UART_MISR	0x0010
+#define UART_ISR	0x0014
+
+#endif	/* __DRIVERS_SERIAL_MSM_SERIAL_H */
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index 6fd80c4243f1..23d2fb051f97 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -171,6 +171,9 @@
 /* Timberdale UART */
 #define PORT_TIMBUART	87
 
+/* Qualcomm MSM SoCs */
+#define PORT_MSM	88
+
 #ifdef __KERNEL__
 
 #include <linux/compiler.h>
-- 
cgit v1.2.3-59-g8ed1b


From 44da59e4006fbf7c4cc9b54485a37a40726091ee Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@linux.intel.com>
Date: Mon, 22 Jun 2009 18:43:18 +0100
Subject: msm: fixups to match current code

The tty layer is now a bit more fussy about reporting the right baud rate
back. Make the msm driver match the current state of affairs.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/msm_serial.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/serial/msm_serial.c b/drivers/serial/msm_serial.c
index 1a7c856f76f8..698048f64f5e 100644
--- a/drivers/serial/msm_serial.c
+++ b/drivers/serial/msm_serial.c
@@ -229,7 +229,7 @@ static void msm_break_ctl(struct uart_port *port, int break_ctl)
 		msm_write(port, UART_CR_CMD_STOP_BREAK, UART_CR);
 }
 
-static void msm_set_baud_rate(struct uart_port *port, unsigned int baud)
+static int msm_set_baud_rate(struct uart_port *port, unsigned int baud)
 {
 	unsigned int baud_code, rxstale, watermark;
 
@@ -281,6 +281,7 @@ static void msm_set_baud_rate(struct uart_port *port, unsigned int baud)
 	case 115200:
 	default:
 		baud_code = UART_CSR_115200;
+		baud = 115200;
 		rxstale = 31;
 		break;
 	}
@@ -299,6 +300,8 @@ static void msm_set_baud_rate(struct uart_port *port, unsigned int baud)
 
 	/* set TX watermark */
 	msm_write(port, 10, UART_TFWR);
+
+	return baud;
 }
 
 static void msm_reset(struct uart_port *port)
@@ -395,8 +398,10 @@ static void msm_set_termios(struct uart_port *port, struct ktermios *termios,
 
 	/* calculate and set baud rate */
 	baud = uart_get_baud_rate(port, termios, old, 300, 115200);
-	msm_set_baud_rate(port, baud);
-
+	baud = msm_set_baud_rate(port, baud);
+	if (tty_termios_baud_rate(termios))
+		tty_termios_encode_baud_rate(termios, baud, baud);
+	
 	/* calculate parity */
 	mr = msm_read(port, UART_MR2);
 	mr &= ~UART_MR2_PARITY_MODE;
-- 
cgit v1.2.3-59-g8ed1b


From e38be994b9cad09b0d8d78a1875d7e8a2e115d29 Mon Sep 17 00:00:00 2001
From: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Date: Mon, 22 Jun 2009 11:11:20 +0900
Subject: Making fs/minix/minix.h double including safe

I happened to find that fs/minix/minix.h doesn't guard double include.

Yes, I know this never cause something destructive because this is
self-evidence that no source file includes minix.h twice, but I think
fixing this is better than disregarding it.

Signed-off-by: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/minix/minix.h | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/fs/minix/minix.h b/fs/minix/minix.h
index cb7fdd11f9a5..9dcf95b42116 100644
--- a/fs/minix/minix.h
+++ b/fs/minix/minix.h
@@ -1,3 +1,6 @@
+#ifndef FS_MINIX_H
+#define FS_MINIX_H
+
 #include <linux/fs.h>
 #include <linux/pagemap.h>
 #include <linux/minix_fs.h>
@@ -86,3 +89,5 @@ static inline struct minix_inode_info *minix_i(struct inode *inode)
 {
 	return list_entry(inode, struct minix_inode_info, vfs_inode);
 }
+
+#endif /* FS_MINIX_H */
-- 
cgit v1.2.3-59-g8ed1b


From eadfe21989d728b5af936487627b4e288bd805f8 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 22 Jun 2009 15:32:31 +0100
Subject: LDSCRIPT: Name INIT_RAM_FS consistently

In asm-generic/vmlinux.lds.h, name INIT_RAM_FS consistently, no matter the
setting of CONFIG_BLK_DEV_INITRD.  This corrects:

	commit ef53dae8658cf0e93d380983824a661067948d87
	Author: Sam Ravnborg <sam@ravnborg.org>
	Date:   Sun Jun 7 20:46:37 2009 +0200
	Subject: Improve vmlinux.lds.h support for arch specific linker scripts

Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-generic/vmlinux.lds.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 55413e568f07..92b73b6140ff 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -625,7 +625,7 @@
 	*(.init.ramfs)							\
 	VMLINUX_SYMBOL(__initramfs_end) = .;
 #else
-#define INITRAMFS
+#define INIT_RAM_FS
 #endif
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 2e8b5a09ebf1f98f02c1988a48415e89d4c25168 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 22 Jun 2009 15:32:36 +0100
Subject: MN10300: Fix the vmlinux ldscript

Fix the MN10300 vmlinux ldscript.  It needs to use various macros from
asm-generic/vmlinux.lds.h to correctly include all that it needs to.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/mn10300/kernel/vmlinux.lds.S | 60 +++++++++++----------------------------
 1 file changed, 16 insertions(+), 44 deletions(-)

diff --git a/arch/mn10300/kernel/vmlinux.lds.S b/arch/mn10300/kernel/vmlinux.lds.S
index 24de6b90f401..bcebcefb4ad7 100644
--- a/arch/mn10300/kernel/vmlinux.lds.S
+++ b/arch/mn10300/kernel/vmlinux.lds.S
@@ -38,14 +38,10 @@ SECTIONS
 
   _etext = .;			/* End of text section */
 
-  . = ALIGN(16);		/* Exception table */
-  __start___ex_table = .;
-  __ex_table : { *(__ex_table) }
-  __stop___ex_table = .;
-
+  EXCEPTION_TABLE(16)
   BUG_TABLE
 
-  RODATA
+  RO_DATA(PAGE_SIZE)
 
   /* writeable */
   .data : {			/* Data */
@@ -53,27 +49,19 @@ SECTIONS
 	CONSTRUCTORS
 	}
 
-  . = ALIGN(PAGE_SIZE);
-  __nosave_begin = .;
-  .data_nosave : { *(.data.nosave) }
-  . = ALIGN(PAGE_SIZE);
-  __nosave_end = .;
-
-  . = ALIGN(PAGE_SIZE);
-  .data.page_aligned : { *(.data.idt) }
+  .data_nosave : { NOSAVE_DATA; }
 
-  . = ALIGN(32);
-  .data.cacheline_aligned : { *(.data.cacheline_aligned) }
+  .data.page_aligned : { PAGE_ALIGNED_DATA(PAGE_SIZE); }
+  .data.cacheline_aligned : { CACHELINE_ALIGNED_DATA(32); }
 
   /* rarely changed data like cpu maps */
   . = ALIGN(32);
   .data.read_mostly : AT(ADDR(.data.read_mostly)) {
-	*(.data.read_mostly)
+	READ_MOSTLY_DATA(32);
 	_edata = .;		/* End of data section */
   }
 
-  . = ALIGN(THREAD_SIZE);	/* init_task */
-  .data.init_task : { *(.data.init_task) }
+  .data.init_task : { INIT_TASK(THREAD_SIZE); }
 
   /* might get freed after init */
   . = ALIGN(PAGE_SIZE);
@@ -88,23 +76,18 @@ SECTIONS
   __init_begin = .;
   .init.text : {
 	_sinittext = .;
-	*(.init.text)
+	INIT_TEXT;
 	_einittext = .;
   }
-  .init.data : { *(.init.data) }
-  . = ALIGN(16);
-  __setup_start = .;
-  .setup.init : { KEEP(*(.init.setup)) }
-  __setup_end = .;
+  .init.data : { INIT_DATA; }
+  .setup.init : { INIT_SETUP(16); }
 
   __initcall_start = .;
   .initcall.init : {
 	INITCALLS
   }
   __initcall_end = .;
-  __con_initcall_start = .;
-  .con_initcall.init : { *(.con_initcall.init) }
-  __con_initcall_end = .;
+  .con_initcall.init : { CON_INITCALL; }
 
   SECURITY_INIT
   . = ALIGN(4);
@@ -114,28 +97,17 @@ SECTIONS
  .altinstr_replacement : { *(.altinstr_replacement) }
   /* .exit.text is discard at runtime, not link time, to deal with references
      from .altinstructions and .eh_frame */
-  .exit.text : { *(.exit.text) }
-  .exit.data : { *(.exit.data) }
+	.exit.text : { EXIT_TEXT; }
+	.exit.data : { EXIT_DATA; }
 
-#ifdef CONFIG_BLK_DEV_INITRD
-  . = ALIGN(PAGE_SIZE);
-  __initramfs_start = .;
-  .init.ramfs : { *(.init.ramfs) }
-  __initramfs_end = .;
-#endif
+  .init.ramfs : { INIT_RAM_FS; }
 
   PERCPU(32)
   . = ALIGN(PAGE_SIZE);
   __init_end = .;
   /* freed after init ends here */
 
-  __bss_start = .;		/* BSS */
-  .bss : {
-	*(.bss.page_aligned)
-	*(.bss)
-  }
-  . = ALIGN(4);
-  __bss_stop = .;
+  BSS(4)
 
   _end = . ;
 
@@ -145,7 +117,7 @@ SECTIONS
 
   /* Sections to be discarded */
   /DISCARD/ : {
-	*(.exitcall.exit)
+	EXIT_CALL
 	}
 
   STABS_DEBUG
-- 
cgit v1.2.3-59-g8ed1b


From 1c520dfbf391e1617ef61553f815b8006a066c44 Mon Sep 17 00:00:00 2001
From: Joel Becker <joel.becker@oracle.com>
Date: Fri, 19 Jun 2009 15:14:13 -0700
Subject: ocfs2: Provide the ocfs2_dlm_lvb_valid() stack API.

The Lock Value Block (LVB) of a DLM lock can be lost when nodes die and
the DLM cannot reconstruct its state.  Clients of the DLM need to know
this.

ocfs2's internal DLM, o2dlm, explicitly zeroes out the LVB when it loses
track of the state.  This is not a standard behavior, but ocfs2 has
always relied on it.  Thus, an o2dlm LVB is always "valid".

ocfs2 now supports both o2dlm and fs/dlm via the stack glue.  When
fs/dlm loses track of an LVBs state, it sets a flag
(DLM_SBF_VALNOTVALID) on the Lock Status Block (LKSB).  The contents of
the LVB may be garbage or merely stale.

ocfs2 doesn't want to try to guess at the validity of the stale LVB.
Instead, it should be checking the VALNOTVALID flag.  As this is the
'standard' way of treating LVBs, we will promote this behavior.

We add a stack glue API ocfs2_dlm_lvb_valid().  It returns non-zero when
the LVB is valid.  o2dlm will always return valid, while fs/dlm will
check VALNOTVALID.

Signed-off-by: Joel Becker <joel.becker@oracle.com>
Acked-by: Mark Fasheh <mfasheh@suse.com>
---
 fs/ocfs2/dlmglue.c    |  9 ++++++---
 fs/ocfs2/stack_o2cb.c | 11 +++++++++++
 fs/ocfs2/stack_user.c |  8 ++++++++
 fs/ocfs2/stackglue.c  | 13 +++++++------
 fs/ocfs2/stackglue.h  |  6 ++++++
 5 files changed, 38 insertions(+), 9 deletions(-)

diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 6cdeaa76f27f..83d2ddb27186 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -1989,7 +1989,8 @@ static inline int ocfs2_meta_lvb_is_trustable(struct inode *inode,
 {
 	struct ocfs2_meta_lvb *lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
 
-	if (lvb->lvb_version == OCFS2_LVB_VERSION
+	if (ocfs2_dlm_lvb_valid(&lockres->l_lksb)
+	    && lvb->lvb_version == OCFS2_LVB_VERSION
 	    && be32_to_cpu(lvb->lvb_igeneration) == inode->i_generation)
 		return 1;
 	return 0;
@@ -2382,7 +2383,8 @@ int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno, int ex)
 		return status;
 
 	lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
-	if (lvb->lvb_version == OCFS2_ORPHAN_LVB_VERSION)
+	if (ocfs2_dlm_lvb_valid(&lockres->l_lksb) &&
+	    lvb->lvb_version == OCFS2_ORPHAN_LVB_VERSION)
 		*seqno = be32_to_cpu(lvb->lvb_os_seqno);
 	return status;
 }
@@ -3627,7 +3629,8 @@ static int ocfs2_refresh_qinfo(struct ocfs2_mem_dqinfo *oinfo)
 	struct ocfs2_global_disk_dqinfo *gdinfo;
 	int status = 0;
 
-	if (lvb->lvb_version == OCFS2_QINFO_LVB_VERSION) {
+	if (ocfs2_dlm_lvb_valid(&lockres->l_lksb) &&
+	    lvb->lvb_version == OCFS2_QINFO_LVB_VERSION) {
 		info->dqi_bgrace = be32_to_cpu(lvb->lvb_bgrace);
 		info->dqi_igrace = be32_to_cpu(lvb->lvb_igrace);
 		oinfo->dqi_syncms = be32_to_cpu(lvb->lvb_syncms);
diff --git a/fs/ocfs2/stack_o2cb.c b/fs/ocfs2/stack_o2cb.c
index fcd120f1493a..3f661376a2de 100644
--- a/fs/ocfs2/stack_o2cb.c
+++ b/fs/ocfs2/stack_o2cb.c
@@ -236,6 +236,16 @@ static int o2cb_dlm_lock_status(union ocfs2_dlm_lksb *lksb)
 	return dlm_status_to_errno(lksb->lksb_o2dlm.status);
 }
 
+/*
+ * o2dlm aways has a "valid" LVB. If the dlm loses track of the LVB
+ * contents, it will zero out the LVB.  Thus the caller can always trust
+ * the contents.
+ */
+static int o2cb_dlm_lvb_valid(union ocfs2_dlm_lksb *lksb)
+{
+	return 1;
+}
+
 static void *o2cb_dlm_lvb(union ocfs2_dlm_lksb *lksb)
 {
 	return (void *)(lksb->lksb_o2dlm.lvb);
@@ -354,6 +364,7 @@ static struct ocfs2_stack_operations o2cb_stack_ops = {
 	.dlm_lock	= o2cb_dlm_lock,
 	.dlm_unlock	= o2cb_dlm_unlock,
 	.lock_status	= o2cb_dlm_lock_status,
+	.lvb_valid	= o2cb_dlm_lvb_valid,
 	.lock_lvb	= o2cb_dlm_lvb,
 	.dump_lksb	= o2cb_dump_lksb,
 };
diff --git a/fs/ocfs2/stack_user.c b/fs/ocfs2/stack_user.c
index 9b76d41a8ac6..ff4c798a5635 100644
--- a/fs/ocfs2/stack_user.c
+++ b/fs/ocfs2/stack_user.c
@@ -738,6 +738,13 @@ static int user_dlm_lock_status(union ocfs2_dlm_lksb *lksb)
 	return lksb->lksb_fsdlm.sb_status;
 }
 
+static int user_dlm_lvb_valid(union ocfs2_dlm_lksb *lksb)
+{
+	int invalid = lksb->lksb_fsdlm.sb_flags & DLM_SBF_VALNOTVALID;
+
+	return !invalid;
+}
+
 static void *user_dlm_lvb(union ocfs2_dlm_lksb *lksb)
 {
 	if (!lksb->lksb_fsdlm.sb_lvbptr)
@@ -873,6 +880,7 @@ static struct ocfs2_stack_operations ocfs2_user_plugin_ops = {
 	.dlm_lock	= user_dlm_lock,
 	.dlm_unlock	= user_dlm_unlock,
 	.lock_status	= user_dlm_lock_status,
+	.lvb_valid	= user_dlm_lvb_valid,
 	.lock_lvb	= user_dlm_lvb,
 	.plock		= user_plock,
 	.dump_lksb	= user_dlm_dump_lksb,
diff --git a/fs/ocfs2/stackglue.c b/fs/ocfs2/stackglue.c
index 68b668b0e60a..3f2f1c45b7b6 100644
--- a/fs/ocfs2/stackglue.c
+++ b/fs/ocfs2/stackglue.c
@@ -6,7 +6,7 @@
  * Code which implements an OCFS2 specific interface to underlying
  * cluster stacks.
  *
- * Copyright (C) 2007 Oracle.  All rights reserved.
+ * Copyright (C) 2007, 2009 Oracle.  All rights reserved.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public
@@ -271,11 +271,12 @@ int ocfs2_dlm_lock_status(union ocfs2_dlm_lksb *lksb)
 }
 EXPORT_SYMBOL_GPL(ocfs2_dlm_lock_status);
 
-/*
- * Why don't we cast to ocfs2_meta_lvb?  The "clean" answer is that we
- * don't cast at the glue level.  The real answer is that the header
- * ordering is nigh impossible.
- */
+int ocfs2_dlm_lvb_valid(union ocfs2_dlm_lksb *lksb)
+{
+	return active_stack->sp_ops->lvb_valid(lksb);
+}
+EXPORT_SYMBOL_GPL(ocfs2_dlm_lvb_valid);
+
 void *ocfs2_dlm_lvb(union ocfs2_dlm_lksb *lksb)
 {
 	return active_stack->sp_ops->lock_lvb(lksb);
diff --git a/fs/ocfs2/stackglue.h b/fs/ocfs2/stackglue.h
index c571af375ef8..03a44d60eac9 100644
--- a/fs/ocfs2/stackglue.h
+++ b/fs/ocfs2/stackglue.h
@@ -185,6 +185,11 @@ struct ocfs2_stack_operations {
 	 */
 	int (*lock_status)(union ocfs2_dlm_lksb *lksb);
 
+	/*
+	 * Return non-zero if the LVB is valid.
+	 */
+	int (*lvb_valid)(union ocfs2_dlm_lksb *lksb);
+
 	/*
 	 * Pull the lvb pointer off of the stack-specific lksb.
 	 */
@@ -252,6 +257,7 @@ int ocfs2_dlm_unlock(struct ocfs2_cluster_connection *conn,
 		     struct ocfs2_lock_res *astarg);
 
 int ocfs2_dlm_lock_status(union ocfs2_dlm_lksb *lksb);
+int ocfs2_dlm_lvb_valid(union ocfs2_dlm_lksb *lksb);
 void *ocfs2_dlm_lvb(union ocfs2_dlm_lksb *lksb);
 void ocfs2_dlm_dump_lksb(union ocfs2_dlm_lksb *lksb);
 
-- 
cgit v1.2.3-59-g8ed1b


From 1962f39abbb2d5643a7d59169422661a2d58793d Mon Sep 17 00:00:00 2001
From: Tao Ma <tao.ma@oracle.com>
Date: Fri, 19 Jun 2009 15:36:52 +0800
Subject: ocfs2: Update atime in splice read if necessary.

We should call ocfs2_inode_lock_atime instead of ocfs2_inode_lock
in ocfs2_file_splice_read like we do in ocfs2_file_aio_read so
that we can update atime in splice read if necessary.

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/file.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 07267e0da909..62442e413a00 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -2026,7 +2026,7 @@ static ssize_t ocfs2_file_splice_read(struct file *in,
 				      size_t len,
 				      unsigned int flags)
 {
-	int ret = 0;
+	int ret = 0, lock_level = 0;
 	struct inode *inode = in->f_path.dentry->d_inode;
 
 	mlog_entry("(0x%p, 0x%p, %u, '%.*s')\n", in, pipe,
@@ -2037,12 +2037,12 @@ static ssize_t ocfs2_file_splice_read(struct file *in,
 	/*
 	 * See the comment in ocfs2_file_aio_read()
 	 */
-	ret = ocfs2_inode_lock(inode, NULL, 0);
+	ret = ocfs2_inode_lock_atime(inode, in->f_vfsmnt, &lock_level);
 	if (ret < 0) {
 		mlog_errno(ret);
 		goto bail;
 	}
-	ocfs2_inode_unlock(inode, 0);
+	ocfs2_inode_unlock(inode, lock_level);
 
 	ret = generic_file_splice_read(in, ppos, pipe, len, flags);
 
-- 
cgit v1.2.3-59-g8ed1b


From 94e41ecfe0f202df948fdbb19a53308a58cf2184 Mon Sep 17 00:00:00 2001
From: Sunil Mushran <sunil.mushran@oracle.com>
Date: Fri, 19 Jun 2009 14:45:54 -0700
Subject: ocfs2: Pin journal head before accessing jh->b_committed_data

This patch adds jbd_lock_bh_state() and jbd_unlock_bh_state() around accessses
to jh->b_committed_data.

Fixes oss bugzilla#1131
http://oss.oracle.com/bugzilla/show_bug.cgi?id=1131

Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/suballoc.c | 28 ++++++++++++++++++++++++----
 1 file changed, 24 insertions(+), 4 deletions(-)

diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c
index 8439f6b324b9..73a16d4666dc 100644
--- a/fs/ocfs2/suballoc.c
+++ b/fs/ocfs2/suballoc.c
@@ -923,14 +923,23 @@ static int ocfs2_test_bg_bit_allocatable(struct buffer_head *bg_bh,
 					 int nr)
 {
 	struct ocfs2_group_desc *bg = (struct ocfs2_group_desc *) bg_bh->b_data;
+	int ret;
 
 	if (ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap))
 		return 0;
-	if (!buffer_jbd(bg_bh) || !bh2jh(bg_bh)->b_committed_data)
+
+	if (!buffer_jbd(bg_bh))
 		return 1;
 
+	jbd_lock_bh_state(bg_bh);
 	bg = (struct ocfs2_group_desc *) bh2jh(bg_bh)->b_committed_data;
-	return !ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap);
+	if (bg)
+		ret = !ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap);
+	else
+		ret = 1;
+	jbd_unlock_bh_state(bg_bh);
+
+	return ret;
 }
 
 static int ocfs2_block_group_find_clear_bits(struct ocfs2_super *osb,
@@ -1885,6 +1894,7 @@ static inline int ocfs2_block_group_clear_bits(handle_t *handle,
 	unsigned int tmp;
 	int journal_type = OCFS2_JOURNAL_ACCESS_WRITE;
 	struct ocfs2_group_desc *undo_bg = NULL;
+	int cluster_bitmap = 0;
 
 	mlog_entry_void();
 
@@ -1905,18 +1915,28 @@ static inline int ocfs2_block_group_clear_bits(handle_t *handle,
 	}
 
 	if (ocfs2_is_cluster_bitmap(alloc_inode))
-		undo_bg = (struct ocfs2_group_desc *) bh2jh(group_bh)->b_committed_data;
+		cluster_bitmap = 1;
+
+	if (cluster_bitmap) {
+		jbd_lock_bh_state(group_bh);
+		undo_bg = (struct ocfs2_group_desc *)
+					bh2jh(group_bh)->b_committed_data;
+		BUG_ON(!undo_bg);
+	}
 
 	tmp = num_bits;
 	while(tmp--) {
 		ocfs2_clear_bit((bit_off + tmp),
 				(unsigned long *) bg->bg_bitmap);
-		if (ocfs2_is_cluster_bitmap(alloc_inode))
+		if (cluster_bitmap)
 			ocfs2_set_bit(bit_off + tmp,
 				      (unsigned long *) undo_bg->bg_bitmap);
 	}
 	le16_add_cpu(&bg->bg_free_bits_count, num_bits);
 
+	if (cluster_bitmap)
+		jbd_unlock_bh_state(group_bh);
+
 	status = ocfs2_journal_dirty(handle, group_bh);
 	if (status < 0)
 		mlog_errno(status);
-- 
cgit v1.2.3-59-g8ed1b


From c3d38840abaa45c1c5a5fabbb8ffc9a0d1a764d1 Mon Sep 17 00:00:00 2001
From: Sunil Mushran <sunil.mushran@oracle.com>
Date: Fri, 19 Jun 2009 14:45:55 -0700
Subject: ocfs2: Fix ocfs2_osb_dump()

Skip printing information that is not valid for local mounts.

Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/super.c | 32 ++++++++++++++++++--------------
 1 file changed, 18 insertions(+), 14 deletions(-)

diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index d33767f17ba3..d64739b593e9 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -234,20 +234,24 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 			"%10s => Opts: 0x%lX  AtimeQuanta: %u\n", "Mount",
 			osb->s_mount_opt, osb->s_atime_quantum);
 
-	out += snprintf(buf + out, len - out,
-			"%10s => Stack: %s  Name: %*s  Version: %d.%d\n",
-			"Cluster",
-			(*osb->osb_cluster_stack == '\0' ?
-			 "o2cb" : osb->osb_cluster_stack),
-			cconn->cc_namelen, cconn->cc_name,
-			cconn->cc_version.pv_major, cconn->cc_version.pv_minor);
+	if (cconn) {
+		out += snprintf(buf + out, len - out,
+				"%10s => Stack: %s  Name: %*s  "
+				"Version: %d.%d\n", "Cluster",
+				(*osb->osb_cluster_stack == '\0' ?
+				 "o2cb" : osb->osb_cluster_stack),
+				cconn->cc_namelen, cconn->cc_name,
+				cconn->cc_version.pv_major,
+				cconn->cc_version.pv_minor);
+	}
 
 	spin_lock(&osb->dc_task_lock);
 	out += snprintf(buf + out, len - out,
 			"%10s => Pid: %d  Count: %lu  WakeSeq: %lu  "
 			"WorkSeq: %lu\n", "DownCnvt",
-			task_pid_nr(osb->dc_task), osb->blocked_lock_count,
-			osb->dc_wake_sequence, osb->dc_work_sequence);
+			(osb->dc_task ?  task_pid_nr(osb->dc_task) : -1),
+			osb->blocked_lock_count, osb->dc_wake_sequence,
+			osb->dc_work_sequence);
 	spin_unlock(&osb->dc_task_lock);
 
 	spin_lock(&osb->osb_lock);
@@ -267,14 +271,15 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 
 	out += snprintf(buf + out, len - out,
 			"%10s => Pid: %d  Interval: %lu  Needs: %d\n", "Commit",
-			task_pid_nr(osb->commit_task), osb->osb_commit_interval,
+			(osb->commit_task ? task_pid_nr(osb->commit_task) : -1),
+			osb->osb_commit_interval,
 			atomic_read(&osb->needs_checkpoint));
 
 	out += snprintf(buf + out, len - out,
-			"%10s => State: %d  NumTxns: %d  TxnId: %lu\n",
+			"%10s => State: %d  TxnId: %lu  NumTxns: %d\n",
 			"Journal", osb->journal->j_state,
-			atomic_read(&osb->journal->j_num_trans),
-			osb->journal->j_trans_id);
+			osb->journal->j_trans_id,
+			atomic_read(&osb->journal->j_num_trans));
 
 	out += snprintf(buf + out, len - out,
 			"%10s => GlobalAllocs: %d  LocalAllocs: %d  "
@@ -302,7 +307,6 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 
 	out += snprintf(buf + out, len - out, "%10s => %3s  %10s\n",
 			"Slots", "Num", "RecoGen");
-
 	for (i = 0; i < osb->max_slots; ++i) {
 		out += snprintf(buf + out, len - out,
 				"%10s  %c %3d  %10d\n",
-- 
cgit v1.2.3-59-g8ed1b


From 692684e19e317a374c18e70a44d6413e51f71c11 Mon Sep 17 00:00:00 2001
From: Sunil Mushran <sunil.mushran@oracle.com>
Date: Fri, 19 Jun 2009 16:53:17 -0700
Subject: ocfs2: Stop orphan scan as early as possible during umount

Currently if the orphan scan fires a tick before the user issues the umount,
the umount will wait for the queued orphan scan tasks to complete.

This patch makes the umount stop the orphan scan as early as possible so as
to reduce the probability of the queued tasks slowing down the umount.

Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/journal.c | 14 ++++++++++++--
 fs/ocfs2/ocfs2.h   |  6 ++++++
 fs/ocfs2/super.c   |  5 +++--
 3 files changed, 21 insertions(+), 4 deletions(-)

diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 4a3b9e6b31ad..70215a21fb20 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1880,6 +1880,9 @@ void ocfs2_queue_orphan_scan(struct ocfs2_super *osb)
 
 	os = &osb->osb_orphan_scan;
 
+	if (atomic_read(&os->os_state) == ORPHAN_SCAN_INACTIVE)
+		goto out;
+
 	status = ocfs2_orphan_scan_lock(osb, &seqno, DLM_LOCK_EX);
 	if (status < 0) {
 		if (status != -EAGAIN)
@@ -1887,6 +1890,10 @@ void ocfs2_queue_orphan_scan(struct ocfs2_super *osb)
 		goto out;
 	}
 
+	/* Do no queue the tasks if the volume is being umounted */
+	if (atomic_read(&os->os_state) == ORPHAN_SCAN_INACTIVE)
+		goto unlock;
+
 	if (os->os_seqno != seqno) {
 		os->os_seqno = seqno;
 		goto unlock;
@@ -1920,8 +1927,9 @@ void ocfs2_orphan_scan_work(struct work_struct *work)
 
 	mutex_lock(&os->os_lock);
 	ocfs2_queue_orphan_scan(osb);
-	schedule_delayed_work(&os->os_orphan_scan_work,
-			      ocfs2_orphan_scan_timeout());
+	if (atomic_read(&os->os_state) == ORPHAN_SCAN_ACTIVE)
+		schedule_delayed_work(&os->os_orphan_scan_work,
+				      ocfs2_orphan_scan_timeout());
 	mutex_unlock(&os->os_lock);
 }
 
@@ -1930,6 +1938,7 @@ void ocfs2_orphan_scan_stop(struct ocfs2_super *osb)
 	struct ocfs2_orphan_scan *os;
 
 	os = &osb->osb_orphan_scan;
+	atomic_set(&os->os_state, ORPHAN_SCAN_INACTIVE);
 	mutex_lock(&os->os_lock);
 	cancel_delayed_work(&os->os_orphan_scan_work);
 	mutex_unlock(&os->os_lock);
@@ -1940,6 +1949,7 @@ int ocfs2_orphan_scan_init(struct ocfs2_super *osb)
 	struct ocfs2_orphan_scan *os;
 
 	os = &osb->osb_orphan_scan;
+	atomic_set(&os->os_state, ORPHAN_SCAN_ACTIVE);
 	os->os_osb = osb;
 	os->os_count = 0;
 	os->os_scantime = CURRENT_TIME;
diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h
index 18c1d9ec1c93..60e89503ce5a 100644
--- a/fs/ocfs2/ocfs2.h
+++ b/fs/ocfs2/ocfs2.h
@@ -154,6 +154,11 @@ struct ocfs2_lock_res {
 #endif
 };
 
+enum ocfs2_orphan_scan_state {
+	ORPHAN_SCAN_ACTIVE,
+	ORPHAN_SCAN_INACTIVE
+};
+
 struct ocfs2_orphan_scan {
 	struct mutex 		os_lock;
 	struct ocfs2_super 	*os_osb;
@@ -162,6 +167,7 @@ struct ocfs2_orphan_scan {
 	struct timespec		os_scantime;  /* time this node ran the scan */
 	u32			os_count;      /* tracks node specific scans */
 	u32  			os_seqno;       /* tracks cluster wide scans */
+	atomic_t		os_state;              /* ACTIVE or INACTIVE */
 };
 
 struct ocfs2_dlm_debug {
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index d64739b593e9..3e8a68b103ab 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -1814,14 +1814,15 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err)
 
 	debugfs_remove(osb->osb_ctxt);
 
+	/* Orphan scan should be stopped as early as possible */
+	ocfs2_orphan_scan_stop(osb);
+
 	ocfs2_disable_quotas(osb);
 
 	ocfs2_shutdown_local_alloc(osb);
 
 	ocfs2_truncate_log_shutdown(osb);
 
-	ocfs2_orphan_scan_stop(osb);
-
 	/* This will disable recovery and flush any recovery work. */
 	ocfs2_recovery_exit(osb);
 
-- 
cgit v1.2.3-59-g8ed1b


From 3211949f8998dde71d9fe2e063de045ece5e0473 Mon Sep 17 00:00:00 2001
From: Sunil Mushran <sunil.mushran@oracle.com>
Date: Fri, 19 Jun 2009 16:53:18 -0700
Subject: ocfs2: Do not initialize lvb in ocfs2_orphan_scan_lock_res_init()

We don't access the LVB in our ocfs2_*_lock_res_init() functions.

Since the LVB can become invalid during some cluster recovery
operations, the dlmglue must be able to handle an uninitialized
LVB.

For the orphan scan lock, we initialized an uninitialzed LVB with our
scan sequence number plus one.  This starts a normal orphan scan
cycle.

Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/dlmglue.c | 7 +++----
 fs/ocfs2/journal.c | 1 +
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 83d2ddb27186..d3d1f9372f7e 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -644,14 +644,10 @@ static void ocfs2_nfs_sync_lock_res_init(struct ocfs2_lock_res *res,
 static void ocfs2_orphan_scan_lock_res_init(struct ocfs2_lock_res *res,
 					    struct ocfs2_super *osb)
 {
-	struct ocfs2_orphan_scan_lvb *lvb;
-
 	ocfs2_lock_res_init_once(res);
 	ocfs2_build_lock_name(OCFS2_LOCK_TYPE_ORPHAN_SCAN, 0, 0, res->l_name);
 	ocfs2_lock_res_init_common(osb, res, OCFS2_LOCK_TYPE_ORPHAN_SCAN,
 				   &ocfs2_orphan_scan_lops, osb);
-	lvb = ocfs2_dlm_lvb(&res->l_lksb);
-	lvb->lvb_version = OCFS2_ORPHAN_LVB_VERSION;
 }
 
 void ocfs2_file_lock_res_init(struct ocfs2_lock_res *lockres,
@@ -2386,6 +2382,9 @@ int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno, int ex)
 	if (ocfs2_dlm_lvb_valid(&lockres->l_lksb) &&
 	    lvb->lvb_version == OCFS2_ORPHAN_LVB_VERSION)
 		*seqno = be32_to_cpu(lvb->lvb_os_seqno);
+	else
+		*seqno = osb->osb_orphan_scan.os_seqno + 1;
+
 	return status;
 }
 
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 70215a21fb20..0b2c27a9485e 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1952,6 +1952,7 @@ int ocfs2_orphan_scan_init(struct ocfs2_super *osb)
 	atomic_set(&os->os_state, ORPHAN_SCAN_ACTIVE);
 	os->os_osb = osb;
 	os->os_count = 0;
+	os->os_seqno = 0;
 	os->os_scantime = CURRENT_TIME;
 	mutex_init(&os->os_lock);
 
-- 
cgit v1.2.3-59-g8ed1b


From df152c241df9e9d2b9a65d37bd02961abe7f591a Mon Sep 17 00:00:00 2001
From: Sunil Mushran <sunil.mushran@oracle.com>
Date: Mon, 22 Jun 2009 11:40:07 -0700
Subject: ocfs2: Disable orphan scanning for local and hard-ro mounts

Local and Hard-RO mounts do not need orphan scanning.

Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/dlmglue.c | 26 ++++++++++++++++----------
 fs/ocfs2/dlmglue.h |  4 ++--
 fs/ocfs2/journal.c | 30 +++++++++++++++++-------------
 fs/ocfs2/journal.h |  2 +-
 fs/ocfs2/super.c   | 32 +++++++++++++++-----------------
 5 files changed, 51 insertions(+), 43 deletions(-)

diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index d3d1f9372f7e..1841bbb49cb6 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -2366,15 +2366,20 @@ void ocfs2_inode_unlock(struct inode *inode,
 	mlog_exit_void();
 }
 
-int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno, int ex)
+int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno)
 {
 	struct ocfs2_lock_res *lockres;
 	struct ocfs2_orphan_scan_lvb *lvb;
-	int level = ex ? DLM_LOCK_EX : DLM_LOCK_PR;
 	int status = 0;
 
+	if (ocfs2_is_hard_readonly(osb))
+		return -EROFS;
+
+	if (ocfs2_mount_local(osb))
+		return 0;
+
 	lockres = &osb->osb_orphan_scan.os_lockres;
-	status = ocfs2_cluster_lock(osb, lockres, level, 0, 0);
+	status = ocfs2_cluster_lock(osb, lockres, DLM_LOCK_EX, 0, 0);
 	if (status < 0)
 		return status;
 
@@ -2388,17 +2393,18 @@ int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno, int ex)
 	return status;
 }
 
-void ocfs2_orphan_scan_unlock(struct ocfs2_super *osb, u32 seqno, int ex)
+void ocfs2_orphan_scan_unlock(struct ocfs2_super *osb, u32 seqno)
 {
 	struct ocfs2_lock_res *lockres;
 	struct ocfs2_orphan_scan_lvb *lvb;
-	int level = ex ? DLM_LOCK_EX : DLM_LOCK_PR;
 
-	lockres = &osb->osb_orphan_scan.os_lockres;
-	lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
-	lvb->lvb_version = OCFS2_ORPHAN_LVB_VERSION;
-	lvb->lvb_os_seqno = cpu_to_be32(seqno);
-	ocfs2_cluster_unlock(osb, lockres, level);
+	if (!ocfs2_is_hard_readonly(osb) && !ocfs2_mount_local(osb)) {
+		lockres = &osb->osb_orphan_scan.os_lockres;
+		lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
+		lvb->lvb_version = OCFS2_ORPHAN_LVB_VERSION;
+		lvb->lvb_os_seqno = cpu_to_be32(seqno);
+		ocfs2_cluster_unlock(osb, lockres, DLM_LOCK_EX);
+	}
 }
 
 int ocfs2_super_lock(struct ocfs2_super *osb,
diff --git a/fs/ocfs2/dlmglue.h b/fs/ocfs2/dlmglue.h
index 31b90d7b8f51..30f683107f1e 100644
--- a/fs/ocfs2/dlmglue.h
+++ b/fs/ocfs2/dlmglue.h
@@ -121,8 +121,8 @@ int ocfs2_super_lock(struct ocfs2_super *osb,
 		     int ex);
 void ocfs2_super_unlock(struct ocfs2_super *osb,
 			int ex);
-int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno, int ex);
-void ocfs2_orphan_scan_unlock(struct ocfs2_super *osb, u32 seqno, int ex);
+int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno);
+void ocfs2_orphan_scan_unlock(struct ocfs2_super *osb, u32 seqno);
 
 int ocfs2_rename_lock(struct ocfs2_super *osb);
 void ocfs2_rename_unlock(struct ocfs2_super *osb);
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 0b2c27a9485e..f033760ecbea 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1883,7 +1883,7 @@ void ocfs2_queue_orphan_scan(struct ocfs2_super *osb)
 	if (atomic_read(&os->os_state) == ORPHAN_SCAN_INACTIVE)
 		goto out;
 
-	status = ocfs2_orphan_scan_lock(osb, &seqno, DLM_LOCK_EX);
+	status = ocfs2_orphan_scan_lock(osb, &seqno);
 	if (status < 0) {
 		if (status != -EAGAIN)
 			mlog_errno(status);
@@ -1910,7 +1910,7 @@ void ocfs2_queue_orphan_scan(struct ocfs2_super *osb)
 	os->os_count++;
 	os->os_scantime = CURRENT_TIME;
 unlock:
-	ocfs2_orphan_scan_unlock(osb, seqno, DLM_LOCK_EX);
+	ocfs2_orphan_scan_unlock(osb, seqno);
 out:
 	return;
 }
@@ -1938,29 +1938,33 @@ void ocfs2_orphan_scan_stop(struct ocfs2_super *osb)
 	struct ocfs2_orphan_scan *os;
 
 	os = &osb->osb_orphan_scan;
-	atomic_set(&os->os_state, ORPHAN_SCAN_INACTIVE);
-	mutex_lock(&os->os_lock);
-	cancel_delayed_work(&os->os_orphan_scan_work);
-	mutex_unlock(&os->os_lock);
+	if (atomic_read(&os->os_state) == ORPHAN_SCAN_ACTIVE) {
+		atomic_set(&os->os_state, ORPHAN_SCAN_INACTIVE);
+		mutex_lock(&os->os_lock);
+		cancel_delayed_work(&os->os_orphan_scan_work);
+		mutex_unlock(&os->os_lock);
+	}
 }
 
-int ocfs2_orphan_scan_init(struct ocfs2_super *osb)
+void ocfs2_orphan_scan_init(struct ocfs2_super *osb)
 {
 	struct ocfs2_orphan_scan *os;
 
 	os = &osb->osb_orphan_scan;
-	atomic_set(&os->os_state, ORPHAN_SCAN_ACTIVE);
 	os->os_osb = osb;
 	os->os_count = 0;
 	os->os_seqno = 0;
 	os->os_scantime = CURRENT_TIME;
 	mutex_init(&os->os_lock);
+	INIT_DELAYED_WORK(&os->os_orphan_scan_work, ocfs2_orphan_scan_work);
 
-	INIT_DELAYED_WORK(&os->os_orphan_scan_work,
-			  ocfs2_orphan_scan_work);
-	schedule_delayed_work(&os->os_orphan_scan_work,
-			      ocfs2_orphan_scan_timeout());
-	return 0;
+	if (ocfs2_is_hard_readonly(osb) || ocfs2_mount_local(osb))
+		atomic_set(&os->os_state, ORPHAN_SCAN_INACTIVE);
+	else {
+		atomic_set(&os->os_state, ORPHAN_SCAN_ACTIVE);
+		schedule_delayed_work(&os->os_orphan_scan_work,
+				      ocfs2_orphan_scan_timeout());
+	}
 }
 
 struct ocfs2_orphan_filldir_priv {
diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h
index 61045eeb3f6e..5432c7f79cc6 100644
--- a/fs/ocfs2/journal.h
+++ b/fs/ocfs2/journal.h
@@ -144,7 +144,7 @@ static inline void ocfs2_inode_set_new(struct ocfs2_super *osb,
 }
 
 /* Exported only for the journal struct init code in super.c. Do not call. */
-int ocfs2_orphan_scan_init(struct ocfs2_super *osb);
+void ocfs2_orphan_scan_init(struct ocfs2_super *osb);
 void ocfs2_orphan_scan_stop(struct ocfs2_super *osb);
 void ocfs2_orphan_scan_exit(struct ocfs2_super *osb);
 
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index 3e8a68b103ab..0746e1abdfc7 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -205,11 +205,10 @@ static const match_table_t tokens = {
 #ifdef CONFIG_DEBUG_FS
 static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 {
-	int out = 0;
-	int i;
 	struct ocfs2_cluster_connection *cconn = osb->cconn;
 	struct ocfs2_recovery_map *rm = osb->recovery_map;
-	struct ocfs2_orphan_scan *os;
+	struct ocfs2_orphan_scan *os = &osb->osb_orphan_scan;
+	int i, out = 0;
 
 	out += snprintf(buf + out, len - out,
 			"%10s => Id: %-s  Uuid: %-s  Gen: 0x%X  Label: %-s\n",
@@ -305,6 +304,16 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 			atomic_read(&osb->s_num_inodes_stolen));
 	spin_unlock(&osb->osb_lock);
 
+	out += snprintf(buf + out, len - out, "OrphanScan => ");
+	out += snprintf(buf + out, len - out, "Local: %u  Global: %u ",
+			os->os_count, os->os_seqno);
+	out += snprintf(buf + out, len - out, " Last Scan: ");
+	if (atomic_read(&os->os_state) == ORPHAN_SCAN_INACTIVE)
+		out += snprintf(buf + out, len - out, "Disabled\n");
+	else
+		out += snprintf(buf + out, len - out, "%lu seconds ago\n",
+				(get_seconds() - os->os_scantime.tv_sec));
+
 	out += snprintf(buf + out, len - out, "%10s => %3s  %10s\n",
 			"Slots", "Num", "RecoGen");
 	for (i = 0; i < osb->max_slots; ++i) {
@@ -315,13 +324,6 @@ static int ocfs2_osb_dump(struct ocfs2_super *osb, char *buf, int len)
 				i, osb->slot_recovery_generations[i]);
 	}
 
-	os = &osb->osb_orphan_scan;
-	out += snprintf(buf + out, len - out, "Orphan Scan=> ");
-	out += snprintf(buf + out, len - out, "Local: %u  Global: %u ",
-			os->os_count, os->os_seqno);
-	out += snprintf(buf + out, len - out, " Last Scan: %lu seconds ago\n",
-			(get_seconds() - os->os_scantime.tv_sec));
-
 	return out;
 }
 
@@ -1179,6 +1181,9 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
 	atomic_set(&osb->vol_state, VOLUME_MOUNTED_QUOTAS);
 	wake_up(&osb->osb_mount_event);
 
+	/* Start this when the mount is almost sure of being successful */
+	ocfs2_orphan_scan_init(osb);
+
 	mlog_exit(status);
 	return status;
 
@@ -1983,13 +1988,6 @@ static int ocfs2_initialize_super(struct super_block *sb,
 		goto bail;
 	}
 
-	status = ocfs2_orphan_scan_init(osb);
-	if (status) {
-		mlog(ML_ERROR, "Unable to initialize delayed orphan scan\n");
-		mlog_errno(status);
-		goto bail;
-	}
-
 	init_waitqueue_head(&osb->checkpoint_event);
 	atomic_set(&osb->needs_checkpoint, 0);
 
-- 
cgit v1.2.3-59-g8ed1b


From 9a7aa12f3911853a3574d47d567b81a2a5df7208 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Thu, 4 Jun 2009 15:26:49 +0200
Subject: vfs: Set special lockdep map for dirs only if not set by fs

Some filesystems need to set lockdep map for i_mutex differently for
different directories. For example OCFS2 has system directories (for
orphan inode tracking and for gathering all system files like journal
or quota files into a single place) which have different locking
locking rules than standard directories. For a filesystem setting
lockdep map is naturaly done when the inode is read but we have to
modify unlock_new_inode() not to overwrite the lockdep map the filesystem
has set.

Acked-by: peterz@infradead.org
CC: mingo@redhat.com
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/inode.c              | 17 +++++++++++------
 include/linux/lockdep.h | 15 +++++++++++++++
 2 files changed, 26 insertions(+), 6 deletions(-)

diff --git a/fs/inode.c b/fs/inode.c
index f643be565df8..04c785bb63c3 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -665,12 +665,17 @@ void unlock_new_inode(struct inode *inode)
 	if (inode->i_mode & S_IFDIR) {
 		struct file_system_type *type = inode->i_sb->s_type;
 
-		/*
-		 * ensure nobody is actually holding i_mutex
-		 */
-		mutex_destroy(&inode->i_mutex);
-		mutex_init(&inode->i_mutex);
-		lockdep_set_class(&inode->i_mutex, &type->i_mutex_dir_key);
+		/* Set new key only if filesystem hasn't already changed it */
+		if (!lockdep_match_class(&inode->i_mutex,
+		    &type->i_mutex_key)) {
+			/*
+			 * ensure nobody is actually holding i_mutex
+			 */
+			mutex_destroy(&inode->i_mutex);
+			mutex_init(&inode->i_mutex);
+			lockdep_set_class(&inode->i_mutex,
+					  &type->i_mutex_dir_key);
+		}
 	}
 #endif
 	/*
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index da5a5a1f4cd2..b25d1b53df0d 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -258,6 +258,16 @@ extern void lockdep_init_map(struct lockdep_map *lock, const char *name,
 #define lockdep_set_subclass(lock, sub)	\
 		lockdep_init_map(&(lock)->dep_map, #lock, \
 				 (lock)->dep_map.key, sub)
+/*
+ * Compare locking classes
+ */
+#define lockdep_match_class(lock, key) lockdep_match_key(&(lock)->dep_map, key)
+
+static inline int lockdep_match_key(struct lockdep_map *lock,
+				    struct lock_class_key *key)
+{
+	return lock->key == key;
+}
 
 /*
  * Acquire a lock.
@@ -326,6 +336,11 @@ static inline void lockdep_on(void)
 #define lockdep_set_class_and_subclass(lock, key, sub) \
 		do { (void)(key); } while (0)
 #define lockdep_set_subclass(lock, sub)		do { } while (0)
+/*
+ * We don't define lockdep_match_class() and lockdep_match_key() for !LOCKDEP
+ * case since the result is not well defined and the caller should rather
+ * #ifdef the call himself.
+ */
 
 # define INIT_LOCKDEP
 # define lockdep_reset()		do { debug_locks = 1; } while (0)
-- 
cgit v1.2.3-59-g8ed1b


From cb25797d451dc774d9dbc402a65f16a0e32199fe Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Thu, 4 Jun 2009 15:26:50 +0200
Subject: ocfs2: Add lockdep annotations

Add lockdep support to OCFS2. The support also covers all of the cluster
locks except for open locks, journal locks, and local quotafile locks. These
are special because they are acquired for a node, not for a particular process
and lockdep cannot deal with such type of locking.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/dlmglue.c | 81 +++++++++++++++++++++++++++++++++++++++++++-----------
 fs/ocfs2/dlmglue.h | 20 ++++++++++++--
 fs/ocfs2/inode.c   | 11 ++++++++
 fs/ocfs2/namei.c   | 15 ++++++----
 fs/ocfs2/ocfs2.h   |  4 +++
 fs/ocfs2/sysfile.c | 17 ++++++++++++
 6 files changed, 123 insertions(+), 25 deletions(-)

diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 1841bbb49cb6..110bb57c46ab 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -92,6 +92,9 @@ struct ocfs2_unblock_ctl {
 	enum ocfs2_unblock_action unblock_action;
 };
 
+/* Lockdep class keys */
+struct lock_class_key lockdep_keys[OCFS2_NUM_LOCK_TYPES];
+
 static int ocfs2_check_meta_downconvert(struct ocfs2_lock_res *lockres,
 					int new_level);
 static void ocfs2_set_meta_lvb(struct ocfs2_lock_res *lockres);
@@ -317,9 +320,16 @@ static int ocfs2_lock_create(struct ocfs2_super *osb,
 			     u32 dlm_flags);
 static inline int ocfs2_may_continue_on_blocked_lock(struct ocfs2_lock_res *lockres,
 						     int wanted);
-static void ocfs2_cluster_unlock(struct ocfs2_super *osb,
-				 struct ocfs2_lock_res *lockres,
-				 int level);
+static void __ocfs2_cluster_unlock(struct ocfs2_super *osb,
+				   struct ocfs2_lock_res *lockres,
+				   int level, unsigned long caller_ip);
+static inline void ocfs2_cluster_unlock(struct ocfs2_super *osb,
+					struct ocfs2_lock_res *lockres,
+					int level)
+{
+	__ocfs2_cluster_unlock(osb, lockres, level, _RET_IP_);
+}
+
 static inline void ocfs2_generic_handle_downconvert_action(struct ocfs2_lock_res *lockres);
 static inline void ocfs2_generic_handle_convert_action(struct ocfs2_lock_res *lockres);
 static inline void ocfs2_generic_handle_attach_action(struct ocfs2_lock_res *lockres);
@@ -489,6 +499,13 @@ static void ocfs2_lock_res_init_common(struct ocfs2_super *osb,
 	ocfs2_add_lockres_tracking(res, osb->osb_dlm_debug);
 
 	ocfs2_init_lock_stats(res);
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	if (type != OCFS2_LOCK_TYPE_OPEN)
+		lockdep_init_map(&res->l_lockdep_map, ocfs2_lock_type_strings[type],
+				 &lockdep_keys[type], 0);
+	else
+		res->l_lockdep_map.key = NULL;
+#endif
 }
 
 void ocfs2_lock_res_init_once(struct ocfs2_lock_res *res)
@@ -1252,11 +1269,13 @@ static int ocfs2_wait_for_mask_interruptible(struct ocfs2_mask_waiter *mw,
 	return ret;
 }
 
-static int ocfs2_cluster_lock(struct ocfs2_super *osb,
-			      struct ocfs2_lock_res *lockres,
-			      int level,
-			      u32 lkm_flags,
-			      int arg_flags)
+static int __ocfs2_cluster_lock(struct ocfs2_super *osb,
+				struct ocfs2_lock_res *lockres,
+				int level,
+				u32 lkm_flags,
+				int arg_flags,
+				int l_subclass,
+				unsigned long caller_ip)
 {
 	struct ocfs2_mask_waiter mw;
 	int wait, catch_signals = !(osb->s_mount_opt & OCFS2_MOUNT_NOINTR);
@@ -1399,13 +1418,37 @@ out:
 	}
 	ocfs2_update_lock_stats(lockres, level, &mw, ret);
 
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	if (!ret && lockres->l_lockdep_map.key != NULL) {
+		if (level == DLM_LOCK_PR)
+			rwsem_acquire_read(&lockres->l_lockdep_map, l_subclass,
+				!!(arg_flags & OCFS2_META_LOCK_NOQUEUE),
+				caller_ip);
+		else
+			rwsem_acquire(&lockres->l_lockdep_map, l_subclass,
+				!!(arg_flags & OCFS2_META_LOCK_NOQUEUE),
+				caller_ip);
+	}
+#endif
 	mlog_exit(ret);
 	return ret;
 }
 
-static void ocfs2_cluster_unlock(struct ocfs2_super *osb,
-				 struct ocfs2_lock_res *lockres,
-				 int level)
+static inline int ocfs2_cluster_lock(struct ocfs2_super *osb,
+				     struct ocfs2_lock_res *lockres,
+				     int level,
+				     u32 lkm_flags,
+				     int arg_flags)
+{
+	return __ocfs2_cluster_lock(osb, lockres, level, lkm_flags, arg_flags,
+				    0, _RET_IP_);
+}
+
+
+static void __ocfs2_cluster_unlock(struct ocfs2_super *osb,
+				   struct ocfs2_lock_res *lockres,
+				   int level,
+				   unsigned long caller_ip)
 {
 	unsigned long flags;
 
@@ -1414,6 +1457,10 @@ static void ocfs2_cluster_unlock(struct ocfs2_super *osb,
 	ocfs2_dec_holders(lockres, level);
 	ocfs2_downconvert_on_unlock(osb, lockres);
 	spin_unlock_irqrestore(&lockres->l_lock, flags);
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	if (lockres->l_lockdep_map.key != NULL)
+		rwsem_release(&lockres->l_lockdep_map, 1, caller_ip);
+#endif
 	mlog_exit_void();
 }
 
@@ -2159,10 +2206,11 @@ static int ocfs2_assign_bh(struct inode *inode,
  * returns < 0 error if the callback will never be called, otherwise
  * the result of the lock will be communicated via the callback.
  */
-int ocfs2_inode_lock_full(struct inode *inode,
-			 struct buffer_head **ret_bh,
-			 int ex,
-			 int arg_flags)
+int ocfs2_inode_lock_full_nested(struct inode *inode,
+				 struct buffer_head **ret_bh,
+				 int ex,
+				 int arg_flags,
+				 int subclass)
 {
 	int status, level, acquired;
 	u32 dlm_flags;
@@ -2200,7 +2248,8 @@ int ocfs2_inode_lock_full(struct inode *inode,
 	if (arg_flags & OCFS2_META_LOCK_NOQUEUE)
 		dlm_flags |= DLM_LKF_NOQUEUE;
 
-	status = ocfs2_cluster_lock(osb, lockres, level, dlm_flags, arg_flags);
+	status = __ocfs2_cluster_lock(osb, lockres, level, dlm_flags,
+				      arg_flags, subclass, _RET_IP_);
 	if (status < 0) {
 		if (status != -EAGAIN && status != -EIOCBRETRY)
 			mlog_errno(status);
diff --git a/fs/ocfs2/dlmglue.h b/fs/ocfs2/dlmglue.h
index 30f683107f1e..7553836931de 100644
--- a/fs/ocfs2/dlmglue.h
+++ b/fs/ocfs2/dlmglue.h
@@ -78,6 +78,14 @@ struct ocfs2_orphan_scan_lvb {
 /* don't block waiting for the downconvert thread, instead return -EAGAIN */
 #define OCFS2_LOCK_NONBLOCK		(0x04)
 
+/* Locking subclasses of inode cluster lock */
+enum {
+	OI_LS_NORMAL = 0,
+	OI_LS_PARENT,
+	OI_LS_RENAME1,
+	OI_LS_RENAME2,
+};
+
 int ocfs2_dlm_init(struct ocfs2_super *osb);
 void ocfs2_dlm_shutdown(struct ocfs2_super *osb, int hangup_pending);
 void ocfs2_lock_res_init_once(struct ocfs2_lock_res *res);
@@ -104,17 +112,23 @@ void ocfs2_open_unlock(struct inode *inode);
 int ocfs2_inode_lock_atime(struct inode *inode,
 			  struct vfsmount *vfsmnt,
 			  int *level);
-int ocfs2_inode_lock_full(struct inode *inode,
+int ocfs2_inode_lock_full_nested(struct inode *inode,
 			 struct buffer_head **ret_bh,
 			 int ex,
-			 int arg_flags);
+			 int arg_flags,
+			 int subclass);
 int ocfs2_inode_lock_with_page(struct inode *inode,
 			      struct buffer_head **ret_bh,
 			      int ex,
 			      struct page *page);
+/* Variants without special locking class or flags */
+#define ocfs2_inode_lock_full(i, r, e, f)\
+		ocfs2_inode_lock_full_nested(i, r, e, f, OI_LS_NORMAL)
+#define ocfs2_inode_lock_nested(i, b, e, s)\
+		ocfs2_inode_lock_full_nested(i, b, e, 0, s)
 /* 99% of the time we don't want to supply any additional flags --
  * those are for very specific cases only. */
-#define ocfs2_inode_lock(i, b, e) ocfs2_inode_lock_full(i, b, e, 0)
+#define ocfs2_inode_lock(i, b, e) ocfs2_inode_lock_full_nested(i, b, e, 0, OI_LS_NORMAL)
 void ocfs2_inode_unlock(struct inode *inode,
 		       int ex);
 int ocfs2_super_lock(struct ocfs2_super *osb,
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 10e1fa87396a..4dc8890ba316 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -215,6 +215,8 @@ bail:
 static int ocfs2_init_locked_inode(struct inode *inode, void *opaque)
 {
 	struct ocfs2_find_inode_args *args = opaque;
+	static struct lock_class_key ocfs2_quota_ip_alloc_sem_key,
+				     ocfs2_file_ip_alloc_sem_key;
 
 	mlog_entry("inode = %p, opaque = %p\n", inode, opaque);
 
@@ -223,6 +225,15 @@ static int ocfs2_init_locked_inode(struct inode *inode, void *opaque)
 	if (args->fi_sysfile_type != 0)
 		lockdep_set_class(&inode->i_mutex,
 			&ocfs2_sysfile_lock_key[args->fi_sysfile_type]);
+	if (args->fi_sysfile_type == USER_QUOTA_SYSTEM_INODE ||
+	    args->fi_sysfile_type == GROUP_QUOTA_SYSTEM_INODE ||
+	    args->fi_sysfile_type == LOCAL_USER_QUOTA_SYSTEM_INODE ||
+	    args->fi_sysfile_type == LOCAL_GROUP_QUOTA_SYSTEM_INODE)
+		lockdep_set_class(&OCFS2_I(inode)->ip_alloc_sem,
+				  &ocfs2_quota_ip_alloc_sem_key);
+	else
+		lockdep_set_class(&OCFS2_I(inode)->ip_alloc_sem,
+				  &ocfs2_file_ip_alloc_sem_key);
 
 	mlog_exit(0);
 	return 0;
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 33464c6b60a2..8601f934010b 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -118,7 +118,7 @@ static struct dentry *ocfs2_lookup(struct inode *dir, struct dentry *dentry,
 	mlog(0, "find name %.*s in directory %llu\n", dentry->d_name.len,
 	     dentry->d_name.name, (unsigned long long)OCFS2_I(dir)->ip_blkno);
 
-	status = ocfs2_inode_lock(dir, NULL, 0);
+	status = ocfs2_inode_lock_nested(dir, NULL, 0, OI_LS_PARENT);
 	if (status < 0) {
 		if (status != -ENOENT)
 			mlog_errno(status);
@@ -636,7 +636,7 @@ static int ocfs2_link(struct dentry *old_dentry,
 	if (S_ISDIR(inode->i_mode))
 		return -EPERM;
 
-	err = ocfs2_inode_lock(dir, &parent_fe_bh, 1);
+	err = ocfs2_inode_lock_nested(dir, &parent_fe_bh, 1, OI_LS_PARENT);
 	if (err < 0) {
 		if (err != -ENOENT)
 			mlog_errno(err);
@@ -800,7 +800,8 @@ static int ocfs2_unlink(struct inode *dir,
 		return -EPERM;
 	}
 
-	status = ocfs2_inode_lock(dir, &parent_node_bh, 1);
+	status = ocfs2_inode_lock_nested(dir, &parent_node_bh, 1,
+					 OI_LS_PARENT);
 	if (status < 0) {
 		if (status != -ENOENT)
 			mlog_errno(status);
@@ -978,7 +979,8 @@ static int ocfs2_double_lock(struct ocfs2_super *osb,
 			inode1 = tmpinode;
 		}
 		/* lock id2 */
-		status = ocfs2_inode_lock(inode2, bh2, 1);
+		status = ocfs2_inode_lock_nested(inode2, bh2, 1,
+						 OI_LS_RENAME1);
 		if (status < 0) {
 			if (status != -ENOENT)
 				mlog_errno(status);
@@ -987,7 +989,7 @@ static int ocfs2_double_lock(struct ocfs2_super *osb,
 	}
 
 	/* lock id1 */
-	status = ocfs2_inode_lock(inode1, bh1, 1);
+	status = ocfs2_inode_lock_nested(inode1, bh1, 1, OI_LS_RENAME2);
 	if (status < 0) {
 		/*
 		 * An error return must mean that no cluster locks
@@ -1103,7 +1105,8 @@ static int ocfs2_rename(struct inode *old_dir,
 	 * won't have to concurrently downconvert the inode and the
 	 * dentry locks.
 	 */
-	status = ocfs2_inode_lock(old_inode, &old_inode_bh, 1);
+	status = ocfs2_inode_lock_nested(old_inode, &old_inode_bh, 1,
+					 OI_LS_PARENT);
 	if (status < 0) {
 		if (status != -ENOENT)
 			mlog_errno(status);
diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h
index 60e89503ce5a..c9345ebb8493 100644
--- a/fs/ocfs2/ocfs2.h
+++ b/fs/ocfs2/ocfs2.h
@@ -34,6 +34,7 @@
 #include <linux/workqueue.h>
 #include <linux/kref.h>
 #include <linux/mutex.h>
+#include <linux/lockdep.h>
 #ifndef CONFIG_OCFS2_COMPAT_JBD
 # include <linux/jbd2.h>
 #else
@@ -152,6 +153,9 @@ struct ocfs2_lock_res {
 	unsigned int		 l_lock_max_exmode; 	   /* Max wait for EX */
 	unsigned int		 l_lock_refresh;	   /* Disk refreshes */
 #endif
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	struct lockdep_map	 l_lockdep_map;
+#endif
 };
 
 enum ocfs2_orphan_scan_state {
diff --git a/fs/ocfs2/sysfile.c b/fs/ocfs2/sysfile.c
index ab713ebdd546..6f53f5e7256a 100644
--- a/fs/ocfs2/sysfile.c
+++ b/fs/ocfs2/sysfile.c
@@ -50,6 +50,8 @@ static inline int is_in_system_inode_array(struct ocfs2_super *osb,
 					   int type,
 					   u32 slot);
 
+static struct lock_class_key ocfs2_sysfile_cluster_lock_key[NUM_SYSTEM_INODES];
+
 static inline int is_global_system_inode(int type)
 {
 	return type >= OCFS2_FIRST_ONLINE_SYSTEM_INODE &&
@@ -118,6 +120,21 @@ static struct inode * _ocfs2_get_system_file_inode(struct ocfs2_super *osb,
 		inode = NULL;
 		goto bail;
 	}
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	if (type == LOCAL_USER_QUOTA_SYSTEM_INODE ||
+	    type == LOCAL_GROUP_QUOTA_SYSTEM_INODE ||
+	    type == JOURNAL_SYSTEM_INODE) {
+		/* Ignore inode lock on these inodes as the lock does not
+		 * really belong to any process and lockdep cannot handle
+		 * that */
+		OCFS2_I(inode)->ip_inode_lockres.l_lockdep_map.key = NULL;
+	} else {
+		lockdep_init_map(&OCFS2_I(inode)->ip_inode_lockres.
+								l_lockdep_map,
+				 ocfs2_system_inodes[type].si_name,
+				 &ocfs2_sysfile_cluster_lock_key[type], 0);
+	}
+#endif
 bail:
 
 	return inode;
-- 
cgit v1.2.3-59-g8ed1b


From d246ab307d1d003c80fe279897dea22bf52b6e41 Mon Sep 17 00:00:00 2001
From: Tao Ma <tao.ma@oracle.com>
Date: Thu, 18 Jun 2009 13:12:06 +0800
Subject: ocfs2/trivial: Wrap ocfs2_sysfile_cluster_lock_key within define.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Actually ocfs2_sysfile_cluster_lock_key is only used if we enable
CONFIG_DEBUG_LOCK_ALLOC. Wrap it so that we can avoid a building
warning.
fs/ocfs2/sysfile.c:53: warning: ‘ocfs2_sysfile_cluster_lock_key’
defined but not used

Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
---
 fs/ocfs2/sysfile.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/fs/ocfs2/sysfile.c b/fs/ocfs2/sysfile.c
index 6f53f5e7256a..40e53702948c 100644
--- a/fs/ocfs2/sysfile.c
+++ b/fs/ocfs2/sysfile.c
@@ -50,7 +50,9 @@ static inline int is_in_system_inode_array(struct ocfs2_super *osb,
 					   int type,
 					   u32 slot);
 
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
 static struct lock_class_key ocfs2_sysfile_cluster_lock_key[NUM_SYSTEM_INODES];
+#endif
 
 static inline int is_global_system_inode(int type)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 9ccdac3662dbf3c75e8f8851a214bdf7d365a4bd Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 22 Jun 2009 22:34:55 +0100
Subject: [ARM] idle: clean up pm_idle calling, obey hlt_counter

pm_idle is used by infrastructure (eg, cpuidle) which expects architectures
to call it in a certain way.  Arrange for ARM to follow x86's lead on this
and call pm_idle() with interrupts already disabled.  However, we expect
pm_idle() to enable interrupts before it returns.

Also, OMAP wants to be able to disable hlt-ing, so allow hlt_counter to
prevent all calls to pm_idle.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/process.c | 58 ++++++++++++++++++++++++++---------------------
 1 file changed, 32 insertions(+), 26 deletions(-)

diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index 56820cce91a4..39196dff478c 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -114,9 +114,6 @@ void arm_machine_restart(char mode, const char *cmd)
 /*
  * Function pointers to optional machine specific functions
  */
-void (*pm_idle)(void);
-EXPORT_SYMBOL(pm_idle);
-
 void (*pm_power_off)(void);
 EXPORT_SYMBOL(pm_power_off);
 
@@ -130,20 +127,19 @@ EXPORT_SYMBOL_GPL(arm_pm_restart);
  */
 static void default_idle(void)
 {
-	if (hlt_counter)
-		cpu_relax();
-	else {
-		local_irq_disable();
-		if (!need_resched())
-			arch_idle();
-		local_irq_enable();
-	}
+	if (!need_resched())
+		arch_idle();
+	local_irq_enable();
 }
 
+void (*pm_idle)(void) = default_idle;
+EXPORT_SYMBOL(pm_idle);
+
 /*
- * The idle thread.  We try to conserve power, while trying to keep
- * overall latency low.  The architecture specific idle is passed
- * a value to indicate the level of "idleness" of the system.
+ * The idle thread, has rather strange semantics for calling pm_idle,
+ * but this is what x86 does and we need to do the same, so that
+ * things like cpuidle get called in the same way.  The only difference
+ * is that we always respect 'hlt_counter' to prevent low power idle.
  */
 void cpu_idle(void)
 {
@@ -151,21 +147,31 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		void (*idle)(void) = pm_idle;
-
+		tick_nohz_stop_sched_tick(1);
+		leds_event(led_idle_start);
+		while (!need_resched()) {
 #ifdef CONFIG_HOTPLUG_CPU
-		if (cpu_is_offline(smp_processor_id())) {
-			leds_event(led_idle_start);
-			cpu_die();
-		}
+			if (cpu_is_offline(smp_processor_id()))
+				cpu_die();
 #endif
 
-		if (!idle)
-			idle = default_idle;
-		leds_event(led_idle_start);
-		tick_nohz_stop_sched_tick(1);
-		while (!need_resched())
-			idle();
+			local_irq_disable();
+			if (hlt_counter) {
+				local_irq_enable();
+				cpu_relax();
+			} else {
+				stop_critical_timings();
+				pm_idle();
+				start_critical_timings();
+				/*
+				 * This will eventually be removed - pm_idle
+				 * functions should always return with IRQs
+				 * enabled.
+				 */
+				WARN_ON(irqs_disabled());
+				local_irq_enable();
+			}
+		}
 		leds_event(led_idle_end);
 		tick_nohz_restart_sched_tick();
 		preempt_enable_no_resched();
-- 
cgit v1.2.3-59-g8ed1b


From 4839641333d4593bfc4fb29aa3af10d36f607d5b Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Tue, 23 Jun 2009 01:34:19 +0100
Subject: jffs2: fix another potential leak on error path in scan.c

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 fs/jffs2/scan.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/jffs2/scan.c b/fs/jffs2/scan.c
index 7515e73e2bfb..696686cc206e 100644
--- a/fs/jffs2/scan.c
+++ b/fs/jffs2/scan.c
@@ -130,9 +130,9 @@ int jffs2_scan_medium(struct jffs2_sb_info *c)
 	if (jffs2_sum_active()) {
 		s = kzalloc(sizeof(struct jffs2_summary), GFP_KERNEL);
 		if (!s) {
-			kfree(flashbuf);
 			JFFS2_WARNING("Can't allocate memory for summary\n");
-			return -ENOMEM;
+			ret = -ENOMEM;
+			goto out;
 		}
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From e56e03b0cfeb997a4be9ad874c193824364942e0 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Sun, 7 Jun 2009 16:31:52 -0400
Subject: Blackfin: unify memory region checks between kgdb and traps

The kgdb (in multiple places) and traps code developed pretty much
identical checks for how to access different regions of the Blackfin
memory map, but each wasn't 100%, so unify them to avoid duplication,
bitrot, and bugs with edge cases.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/include/asm/uaccess.h |  22 +++
 arch/blackfin/kernel/kgdb.c         | 297 ++++++++++++------------------------
 arch/blackfin/kernel/process.c      | 151 ++++++++++++++----
 arch/blackfin/kernel/traps.c        |  60 +++-----
 4 files changed, 256 insertions(+), 274 deletions(-)

diff --git a/arch/blackfin/include/asm/uaccess.h b/arch/blackfin/include/asm/uaccess.h
index 8894e9ffbb57..2f469a1f80fb 100644
--- a/arch/blackfin/include/asm/uaccess.h
+++ b/arch/blackfin/include/asm/uaccess.h
@@ -265,4 +265,26 @@ __clear_user(void *to, unsigned long n)
 
 #define clear_user(to, n) __clear_user(to, n)
 
+/* How to interpret these return values:
+ *	CORE:      can be accessed by core load or dma memcpy
+ *	CORE_ONLY: can only be accessed by core load
+ *	DMA:       can only be accessed by dma memcpy
+ *	IDMA:      can only be accessed by interprocessor dma memcpy (BF561)
+ *	ITEST:     can be accessed by isram memcpy or dma memcpy
+ */
+enum {
+	BFIN_MEM_ACCESS_CORE = 0,
+	BFIN_MEM_ACCESS_CORE_ONLY,
+	BFIN_MEM_ACCESS_DMA,
+	BFIN_MEM_ACCESS_IDMA,
+	BFIN_MEM_ACCESS_ITEST,
+};
+/**
+ *	bfin_mem_access_type() - what kind of memory access is required
+ *	@addr:   the address to check
+ *	@size:   number of bytes needed
+ *	@return: <0 is error, >=0 is BFIN_MEM_ACCESS_xxx enum (see above)
+ */
+int bfin_mem_access_type(unsigned long addr, unsigned long size);
+
 #endif				/* _BLACKFIN_UACCESS_H */
diff --git a/arch/blackfin/kernel/kgdb.c b/arch/blackfin/kernel/kgdb.c
index da28f796ad78..cce79d05b90b 100644
--- a/arch/blackfin/kernel/kgdb.c
+++ b/arch/blackfin/kernel/kgdb.c
@@ -34,15 +34,6 @@ int gdb_bfin_vector = -1;
 #error change the definition of slavecpulocks
 #endif
 
-#define IN_MEM(addr, size, l1_addr, l1_size) \
-({ \
-	unsigned long __addr = (unsigned long)(addr); \
-	(l1_size && __addr >= l1_addr && __addr + (size) <= l1_addr + l1_size); \
-})
-#define ASYNC_BANK_SIZE \
-	(ASYNC_BANK0_SIZE + ASYNC_BANK1_SIZE + \
-	 ASYNC_BANK2_SIZE + ASYNC_BANK3_SIZE)
-
 void pt_regs_to_gdb_regs(unsigned long *gdb_regs, struct pt_regs *regs)
 {
 	gdb_regs[BFIN_R0] = regs->r0;
@@ -463,41 +454,88 @@ static int hex(char ch)
 
 static int validate_memory_access_address(unsigned long addr, int size)
 {
-	int cpu = raw_smp_processor_id();
-
-	if (size < 0)
+	if (size < 0 || addr == 0)
 		return -EFAULT;
-	if (addr >= 0x1000 && (addr + size) <= physical_mem_end)
-		return 0;
-	if (addr >= SYSMMR_BASE)
-		return 0;
-	if (IN_MEM(addr, size, ASYNC_BANK0_BASE, ASYNC_BANK_SIZE))
-		return 0;
-	if (cpu == 0) {
-		if (IN_MEM(addr, size, L1_SCRATCH_START, L1_SCRATCH_LENGTH))
-			return 0;
-		if (IN_MEM(addr, size, L1_CODE_START, L1_CODE_LENGTH))
-			return 0;
-		if (IN_MEM(addr, size, L1_DATA_A_START, L1_DATA_A_LENGTH))
-			return 0;
-		if (IN_MEM(addr, size, L1_DATA_B_START, L1_DATA_B_LENGTH))
-			return 0;
-#ifdef CONFIG_SMP
-	} else if (cpu == 1) {
-		if (IN_MEM(addr, size, COREB_L1_SCRATCH_START, L1_SCRATCH_LENGTH))
+	return bfin_mem_access_type(addr, size);
+}
+
+static int bfin_probe_kernel_read(char *dst, char *src, int size)
+{
+	unsigned long lsrc = (unsigned long)src;
+	int mem_type;
+
+	mem_type = validate_memory_access_address(lsrc, size);
+	if (mem_type < 0)
+		return mem_type;
+
+	if (lsrc >= SYSMMR_BASE) {
+		if (size == 2 && lsrc % 2 == 0) {
+			u16 mmr = bfin_read16(src);
+			memcpy(dst, &mmr, sizeof(mmr));
 			return 0;
-		if (IN_MEM(addr, size, COREB_L1_CODE_START, L1_CODE_LENGTH))
+		} else if (size == 4 && lsrc % 4 == 0) {
+			u32 mmr = bfin_read32(src);
+			memcpy(dst, &mmr, sizeof(mmr));
 			return 0;
-		if (IN_MEM(addr, size, COREB_L1_DATA_A_START, L1_DATA_A_LENGTH))
+		}
+	} else {
+		switch (mem_type) {
+			case BFIN_MEM_ACCESS_CORE:
+			case BFIN_MEM_ACCESS_CORE_ONLY:
+				return probe_kernel_read(dst, src, size);
+			/* XXX: should support IDMA here with SMP */
+			case BFIN_MEM_ACCESS_DMA:
+				if (dma_memcpy(dst, src, size))
+					return 0;
+				break;
+			case BFIN_MEM_ACCESS_ITEST:
+				if (isram_memcpy(dst, src, size))
+					return 0;
+				break;
+		}
+	}
+
+	return -EFAULT;
+}
+
+static int bfin_probe_kernel_write(char *dst, char *src, int size)
+{
+	unsigned long ldst = (unsigned long)dst;
+	int mem_type;
+
+	mem_type = validate_memory_access_address(ldst, size);
+	if (mem_type < 0)
+		return mem_type;
+
+	if (ldst >= SYSMMR_BASE) {
+		if (size == 2 && ldst % 2 == 0) {
+			u16 mmr;
+			memcpy(&mmr, src, sizeof(mmr));
+			bfin_write16(dst, mmr);
 			return 0;
-		if (IN_MEM(addr, size, COREB_L1_DATA_B_START, L1_DATA_B_LENGTH))
+		} else if (size == 4 && ldst % 4 == 0) {
+			u32 mmr;
+			memcpy(&mmr, src, sizeof(mmr));
+			bfin_write32(dst, mmr);
 			return 0;
-#endif
+		}
+	} else {
+		switch (mem_type) {
+			case BFIN_MEM_ACCESS_CORE:
+			case BFIN_MEM_ACCESS_CORE_ONLY:
+				return probe_kernel_write(dst, src, size);
+			/* XXX: should support IDMA here with SMP */
+			case BFIN_MEM_ACCESS_DMA:
+				if (dma_memcpy(dst, src, size))
+					return 0;
+				break;
+			case BFIN_MEM_ACCESS_ITEST:
+				if (isram_memcpy(dst, src, size))
+					return 0;
+				break;
+		}
 	}
 
-	if (IN_MEM(addr, size, L2_START, L2_LENGTH))
-		return 0;
-
 	return -EFAULT;
 }
 
@@ -509,14 +547,6 @@ int kgdb_mem2hex(char *mem, char *buf, int count)
 {
 	char *tmp;
 	int err;
-	unsigned char *pch;
-	unsigned short mmr16;
-	unsigned long mmr32;
-	int cpu = raw_smp_processor_id();
-
-	err = validate_memory_access_address((unsigned long)mem, count);
-	if (err)
-		return err;
 
 	/*
 	 * We use the upper half of buf as an intermediate buffer for the
@@ -524,44 +554,7 @@ int kgdb_mem2hex(char *mem, char *buf, int count)
 	 */
 	tmp = buf + count;
 
-	if ((unsigned int)mem >= SYSMMR_BASE) { /*access MMR registers*/
-		switch (count) {
-		case 2:
-			if ((unsigned int)mem % 2 == 0) {
-				mmr16 = *(unsigned short *)mem;
-				pch = (unsigned char *)&mmr16;
-				*tmp++ = *pch++;
-				*tmp++ = *pch++;
-				tmp -= 2;
-			} else
-				err = -EFAULT;
-			break;
-		case 4:
-			if ((unsigned int)mem % 4 == 0) {
-				mmr32 = *(unsigned long *)mem;
-				pch = (unsigned char *)&mmr32;
-				*tmp++ = *pch++;
-				*tmp++ = *pch++;
-				*tmp++ = *pch++;
-				*tmp++ = *pch++;
-				tmp -= 4;
-			} else
-				err = -EFAULT;
-			break;
-		default:
-			err = -EFAULT;
-		}
-	} else if ((cpu == 0 && IN_MEM(mem, count, L1_CODE_START, L1_CODE_LENGTH))
-#ifdef CONFIG_SMP
-		|| (cpu == 1 && IN_MEM(mem, count, COREB_L1_CODE_START, L1_CODE_LENGTH))
-#endif
-		) {
-		/* access L1 instruction SRAM*/
-		if (dma_memcpy(tmp, mem, count) == NULL)
-			err = -EFAULT;
-	} else
-		err = probe_kernel_read(tmp, mem, count);
-
+	err = bfin_probe_kernel_read(tmp, mem, count);
 	if (!err) {
 		while (count > 0) {
 			buf = pack_hex_byte(buf, *tmp);
@@ -582,13 +575,8 @@ int kgdb_mem2hex(char *mem, char *buf, int count)
  */
 int kgdb_ebin2mem(char *buf, char *mem, int count)
 {
-	char *tmp_old;
-	char *tmp_new;
-	unsigned short *mmr16;
-	unsigned long *mmr32;
-	int err;
+	char *tmp_old, *tmp_new;
 	int size;
-	int cpu = raw_smp_processor_id();
 
 	tmp_old = tmp_new = buf;
 
@@ -601,41 +589,7 @@ int kgdb_ebin2mem(char *buf, char *mem, int count)
 		tmp_old++;
 	}
 
-	err = validate_memory_access_address((unsigned long)mem, size);
-	if (err)
-		return err;
-
-	if ((unsigned int)mem >= SYSMMR_BASE) { /*access MMR registers*/
-		switch (size) {
-		case 2:
-			if ((unsigned int)mem % 2 == 0) {
-				mmr16 = (unsigned short *)buf;
-				*(unsigned short *)mem = *mmr16;
-			} else
-				err = -EFAULT;
-			break;
-		case 4:
-			if ((unsigned int)mem % 4 == 0) {
-				mmr32 = (unsigned long *)buf;
-				*(unsigned long *)mem = *mmr32;
-			} else
-				err = -EFAULT;
-			break;
-		default:
-			err = -EFAULT;
-		}
-	} else if ((cpu == 0 && IN_MEM(mem, count, L1_CODE_START, L1_CODE_LENGTH))
-#ifdef CONFIG_SMP
-		|| (cpu == 1 && IN_MEM(mem, count, COREB_L1_CODE_START, L1_CODE_LENGTH))
-#endif
-		) {
-		/* access L1 instruction SRAM */
-		if (dma_memcpy(mem, buf, size) == NULL)
-			err = -EFAULT;
-	} else
-		err = probe_kernel_write(mem, buf, size);
-
-	return err;
+	return bfin_probe_kernel_write(mem, buf, count);
 }
 
 /*
@@ -645,16 +599,7 @@ int kgdb_ebin2mem(char *buf, char *mem, int count)
  */
 int kgdb_hex2mem(char *buf, char *mem, int count)
 {
-	char *tmp_raw;
-	char *tmp_hex;
-	unsigned short *mmr16;
-	unsigned long *mmr32;
-	int err;
-	int cpu = raw_smp_processor_id();
-
-	err = validate_memory_access_address((unsigned long)mem, count);
-	if (err)
-		return err;
+	char *tmp_raw, *tmp_hex;
 
 	/*
 	 * We use the upper half of buf as an intermediate buffer for the
@@ -669,39 +614,18 @@ int kgdb_hex2mem(char *buf, char *mem, int count)
 		*tmp_raw |= hex(*tmp_hex--) << 4;
 	}
 
-	if ((unsigned int)mem >= SYSMMR_BASE) { /*access MMR registers*/
-		switch (count) {
-		case 2:
-			if ((unsigned int)mem % 2 == 0) {
-				mmr16 = (unsigned short *)tmp_raw;
-				*(unsigned short *)mem = *mmr16;
-			} else
-				err = -EFAULT;
-			break;
-		case 4:
-			if ((unsigned int)mem % 4 == 0) {
-				mmr32 = (unsigned long *)tmp_raw;
-				*(unsigned long *)mem = *mmr32;
-			} else
-				err = -EFAULT;
-			break;
-		default:
-			err = -EFAULT;
-		}
-	} else if ((cpu == 0 && IN_MEM(mem, count, L1_CODE_START, L1_CODE_LENGTH))
-#ifdef CONFIG_SMP
-		|| (cpu == 1 && IN_MEM(mem, count, COREB_L1_CODE_START, L1_CODE_LENGTH))
-#endif
-		) {
-		/* access L1 instruction SRAM */
-		if (dma_memcpy(mem, tmp_raw, count) == NULL)
-			err = -EFAULT;
-	} else
-		err = probe_kernel_write(mem, tmp_raw, count);
-
-	return err;
+	return bfin_probe_kernel_write(mem, tmp_raw, count);
 }
 
+#define IN_MEM(addr, size, l1_addr, l1_size) \
+({ \
+	unsigned long __addr = (unsigned long)(addr); \
+	(l1_size && __addr >= l1_addr && __addr + (size) <= l1_addr + l1_size); \
+})
+#define ASYNC_BANK_SIZE \
+	(ASYNC_BANK0_SIZE + ASYNC_BANK1_SIZE + \
+	 ASYNC_BANK2_SIZE + ASYNC_BANK3_SIZE)
+
 int kgdb_validate_break_address(unsigned long addr)
 {
 	int cpu = raw_smp_processor_id();
@@ -724,46 +648,17 @@ int kgdb_validate_break_address(unsigned long addr)
 
 int kgdb_arch_set_breakpoint(unsigned long addr, char *saved_instr)
 {
-	int err;
-	int cpu = raw_smp_processor_id();
-
-	if ((cpu == 0 && IN_MEM(addr, BREAK_INSTR_SIZE, L1_CODE_START, L1_CODE_LENGTH))
-#ifdef CONFIG_SMP
-		|| (cpu == 1 && IN_MEM(addr, BREAK_INSTR_SIZE, COREB_L1_CODE_START, L1_CODE_LENGTH))
-#endif
-		) {
-		/* access L1 instruction SRAM */
-		if (dma_memcpy(saved_instr, (void *)addr, BREAK_INSTR_SIZE)
-			== NULL)
-			return -EFAULT;
-
-		if (dma_memcpy((void *)addr, arch_kgdb_ops.gdb_bpt_instr,
-			BREAK_INSTR_SIZE) == NULL)
-			return -EFAULT;
-
-		return 0;
-	} else {
-		err = probe_kernel_read(saved_instr, (char *)addr,
-			BREAK_INSTR_SIZE);
-		if (err)
-			return err;
-
-		return probe_kernel_write((char *)addr,
-			arch_kgdb_ops.gdb_bpt_instr, BREAK_INSTR_SIZE);
-	}
+	int err = bfin_probe_kernel_read(saved_instr, (char *)addr,
+	                                 BREAK_INSTR_SIZE);
+	if (err)
+		return err;
+	return bfin_probe_kernel_write((char *)addr, arch_kgdb_ops.gdb_bpt_instr,
+	                               BREAK_INSTR_SIZE);
 }
 
 int kgdb_arch_remove_breakpoint(unsigned long addr, char *bundle)
 {
-	if (IN_MEM(addr, BREAK_INSTR_SIZE, L1_CODE_START, L1_CODE_LENGTH)) {
-		/* access L1 instruction SRAM */
-		if (dma_memcpy((void *)addr, bundle, BREAK_INSTR_SIZE) == NULL)
-			return -EFAULT;
-
-		return 0;
-	} else
-		return probe_kernel_write((char *)addr,
-				(char *)bundle, BREAK_INSTR_SIZE);
+	return bfin_probe_kernel_write((char *)addr, bundle, BREAK_INSTR_SIZE);
 }
 
 int kgdb_arch_init(void)
diff --git a/arch/blackfin/kernel/process.c b/arch/blackfin/kernel/process.c
index 3e1d86e456f6..79cad0ac5892 100644
--- a/arch/blackfin/kernel/process.c
+++ b/arch/blackfin/kernel/process.c
@@ -344,6 +344,87 @@ void finish_atomic_sections (struct pt_regs *regs)
 	}
 }
 
+static inline
+int in_mem(unsigned long addr, unsigned long size,
+           unsigned long start, unsigned long end)
+{
+	return addr >= start && addr + size <= end;
+}
+static inline
+int in_mem_const_off(unsigned long addr, unsigned long size, unsigned long off,
+                     unsigned long const_addr, unsigned long const_size)
+{
+	return const_size &&
+	       in_mem(addr, size, const_addr + off, const_addr + const_size);
+}
+static inline
+int in_mem_const(unsigned long addr, unsigned long size,
+                 unsigned long const_addr, unsigned long const_size)
+{
+	return in_mem_const_off(addr, 0, size, const_addr, const_size);
+}
+#define IN_ASYNC(bnum, bctlnum) \
+({ \
+	(bfin_read_EBIU_AMGCTL() & 0xe) < ((bnum + 1) << 1) ? -EFAULT : \
+	bfin_read_EBIU_AMBCTL##bctlnum() & B##bnum##RDYEN ? -EFAULT : \
+	BFIN_MEM_ACCESS_CORE; \
+})
+
+int bfin_mem_access_type(unsigned long addr, unsigned long size)
+{
+	int cpu = raw_smp_processor_id();
+
+	/* Check that things do not wrap around */
+	if (addr > ULONG_MAX - size)
+		return -EFAULT;
+
+	if (in_mem(addr, size, FIXED_CODE_START, physical_mem_end))
+		return BFIN_MEM_ACCESS_CORE;
+
+	if (in_mem_const(addr, size, L1_CODE_START, L1_CODE_LENGTH))
+		return cpu == 0 ? BFIN_MEM_ACCESS_ITEST : BFIN_MEM_ACCESS_IDMA;
+	if (in_mem_const(addr, size, L1_SCRATCH_START, L1_SCRATCH_LENGTH))
+		return cpu == 0 ? BFIN_MEM_ACCESS_CORE_ONLY : -EFAULT;
+	if (in_mem_const(addr, size, L1_DATA_A_START, L1_DATA_A_LENGTH))
+		return cpu == 0 ? BFIN_MEM_ACCESS_CORE : BFIN_MEM_ACCESS_IDMA;
+	if (in_mem_const(addr, size, L1_DATA_B_START, L1_DATA_B_LENGTH))
+		return cpu == 0 ? BFIN_MEM_ACCESS_CORE : BFIN_MEM_ACCESS_IDMA;
+#ifdef COREB_L1_CODE_START
+	if (in_mem_const(addr, size, COREB_L1_CODE_START, L1_CODE_LENGTH))
+		return cpu == 1 ? BFIN_MEM_ACCESS_ITEST : BFIN_MEM_ACCESS_IDMA;
+	if (in_mem_const(addr, size, COREB_L1_SCRATCH_START, L1_SCRATCH_LENGTH))
+		return cpu == 1 ? BFIN_MEM_ACCESS_CORE_ONLY : -EFAULT;
+	if (in_mem_const(addr, size, COREB_L1_DATA_A_START, L1_DATA_A_LENGTH))
+		return cpu == 1 ? BFIN_MEM_ACCESS_CORE : BFIN_MEM_ACCESS_IDMA;
+	if (in_mem_const(addr, size, COREB_L1_DATA_B_START, L1_DATA_B_LENGTH))
+		return cpu == 1 ? BFIN_MEM_ACCESS_CORE : BFIN_MEM_ACCESS_IDMA;
+#endif
+	if (in_mem_const(addr, size, L2_START, L2_LENGTH))
+		return BFIN_MEM_ACCESS_CORE;
+
+	if (addr >= SYSMMR_BASE)
+		return BFIN_MEM_ACCESS_CORE_ONLY;
+
+	/* We can't read EBIU banks that aren't enabled or we end up hanging
+	 * on the access to the async space.
+	 */
+	if (in_mem_const(addr, size, ASYNC_BANK0_BASE, ASYNC_BANK0_SIZE))
+		return IN_ASYNC(0, 0);
+	if (in_mem_const(addr, size, ASYNC_BANK1_BASE, ASYNC_BANK1_SIZE))
+		return IN_ASYNC(1, 0);
+	if (in_mem_const(addr, size, ASYNC_BANK2_BASE, ASYNC_BANK2_SIZE))
+		return IN_ASYNC(2, 1);
+	if (in_mem_const(addr, size, ASYNC_BANK3_BASE, ASYNC_BANK3_SIZE))
+		return IN_ASYNC(3, 1);
+
+	if (in_mem_const(addr, size, BOOT_ROM_START, BOOT_ROM_LENGTH))
+		return BFIN_MEM_ACCESS_CORE;
+	if (in_mem_const(addr, size, L1_ROM_START, L1_ROM_LENGTH))
+		return BFIN_MEM_ACCESS_DMA;
+
+	return -EFAULT;
+}
+
 #if defined(CONFIG_ACCESS_CHECK)
 #ifdef CONFIG_ACCESS_OK_L1
 __attribute__((l1_text))
@@ -353,51 +434,61 @@ int _access_ok(unsigned long addr, unsigned long size)
 {
 	if (size == 0)
 		return 1;
-	if (addr > (addr + size))
+	/* Check that things do not wrap around */
+	if (addr > ULONG_MAX - size)
 		return 0;
 	if (segment_eq(get_fs(), KERNEL_DS))
 		return 1;
 #ifdef CONFIG_MTD_UCLINUX
-	if (addr >= memory_start && (addr + size) <= memory_end)
-		return 1;
-	if (addr >= memory_mtd_end && (addr + size) <= physical_mem_end)
+	if (1)
+#else
+	if (0)
+#endif
+	{
+		if (in_mem(addr, size, memory_start, memory_end))
+			return 1;
+		if (in_mem(addr, size, memory_mtd_end, physical_mem_end))
+			return 1;
+# ifndef CONFIG_ROMFS_ON_MTD
+		if (0)
+# endif
+			/* For XIP, allow user space to use pointers within the ROMFS.  */
+			if (in_mem(addr, size, memory_mtd_start, memory_mtd_end))
+				return 1;
+	} else {
+		if (in_mem(addr, size, memory_start, physical_mem_end))
+			return 1;
+	}
+
+	if (in_mem(addr, size, (unsigned long)__init_begin, (unsigned long)__init_end))
 		return 1;
 
-#ifdef CONFIG_ROMFS_ON_MTD
-	/* For XIP, allow user space to use pointers within the ROMFS.  */
-	if (addr >= memory_mtd_start && (addr + size) <= memory_mtd_end)
+	if (in_mem_const(addr, size, L1_CODE_START, L1_CODE_LENGTH))
 		return 1;
-#endif
-#else
-	if (addr >= memory_start && (addr + size) <= physical_mem_end)
+	if (in_mem_const_off(addr, size, _etext_l1 - _stext_l1, L1_CODE_START, L1_CODE_LENGTH))
 		return 1;
-#endif
-	if (addr >= (unsigned long)__init_begin &&
-	    addr + size <= (unsigned long)__init_end)
+	if (in_mem_const_off(addr, size, _ebss_l1 - _sdata_l1, L1_DATA_A_START, L1_DATA_A_LENGTH))
 		return 1;
-	if (addr >= get_l1_scratch_start()
-	    && addr + size <= get_l1_scratch_start() + L1_SCRATCH_LENGTH)
+	if (in_mem_const_off(addr, size, _ebss_b_l1 - _sdata_b_l1, L1_DATA_B_START, L1_DATA_B_LENGTH))
 		return 1;
-#if L1_CODE_LENGTH != 0
-	if (addr >= get_l1_code_start() + (_etext_l1 - _stext_l1)
-	    && addr + size <= get_l1_code_start() + L1_CODE_LENGTH)
+#ifdef COREB_L1_CODE_START
+	if (in_mem_const(addr, size, COREB_L1_CODE_START, L1_CODE_LENGTH))
 		return 1;
-#endif
-#if L1_DATA_A_LENGTH != 0
-	if (addr >= get_l1_data_a_start() + (_ebss_l1 - _sdata_l1)
-	    && addr + size <= get_l1_data_a_start() + L1_DATA_A_LENGTH)
+	if (in_mem_const(addr, size, COREB_L1_SCRATCH_START, L1_SCRATCH_LENGTH))
 		return 1;
-#endif
-#if L1_DATA_B_LENGTH != 0
-	if (addr >= get_l1_data_b_start() + (_ebss_b_l1 - _sdata_b_l1)
-	    && addr + size <= get_l1_data_b_start() + L1_DATA_B_LENGTH)
+	if (in_mem_const(addr, size, COREB_L1_DATA_A_START, L1_DATA_A_LENGTH))
 		return 1;
-#endif
-#if L2_LENGTH != 0
-	if (addr >= L2_START + (_ebss_l2 - _stext_l2)
-	    && addr + size <= L2_START + L2_LENGTH)
+	if (in_mem_const(addr, size, COREB_L1_DATA_B_START, L1_DATA_B_LENGTH))
 		return 1;
 #endif
+	if (in_mem_const_off(addr, size, _ebss_l2 - _stext_l2, L2_START, L2_LENGTH))
+		return 1;
+
+	if (in_mem_const(addr, size, BOOT_ROM_START, BOOT_ROM_LENGTH))
+		return 1;
+	if (in_mem_const(addr, size, L1_ROM_START, L1_ROM_LENGTH))
+		return 1;
+
 	return 0;
 }
 EXPORT_SYMBOL(_access_ok);
diff --git a/arch/blackfin/kernel/traps.c b/arch/blackfin/kernel/traps.c
index d279552fe9b0..8eeb457ce5d5 100644
--- a/arch/blackfin/kernel/traps.c
+++ b/arch/blackfin/kernel/traps.c
@@ -37,6 +37,7 @@
 #include <asm/traps.h>
 #include <asm/cacheflush.h>
 #include <asm/cplb.h>
+#include <asm/dma.h>
 #include <asm/blackfin.h>
 #include <asm/irq_handler.h>
 #include <linux/irq.h>
@@ -636,57 +637,30 @@ asmlinkage void trap_c(struct pt_regs *fp)
  */
 static bool get_instruction(unsigned short *val, unsigned short *address)
 {
-
-	unsigned long addr;
-
-	addr = (unsigned long)address;
+	unsigned long addr = (unsigned long)address;
 
 	/* Check for odd addresses */
 	if (addr & 0x1)
 		return false;
 
-	/* Check that things do not wrap around */
-	if (addr > (addr + 2))
+	/* MMR region will never have instructions */
+	if (addr >= SYSMMR_BASE)
 		return false;
 
-	/*
-	 * Since we are in exception context, we need to do a little address checking
-	 * We need to make sure we are only accessing valid memory, and
-	 * we don't read something in the async space that can hang forever
-	 */
-	if ((addr >= FIXED_CODE_START && (addr + 2) <= physical_mem_end) ||
-#if L2_LENGTH != 0
-	    (addr >= L2_START && (addr + 2) <= (L2_START + L2_LENGTH)) ||
-#endif
-	    (addr >= BOOT_ROM_START && (addr + 2) <= (BOOT_ROM_START + BOOT_ROM_LENGTH)) ||
-#if L1_DATA_A_LENGTH != 0
-	    (addr >= L1_DATA_A_START && (addr + 2) <= (L1_DATA_A_START + L1_DATA_A_LENGTH)) ||
-#endif
-#if L1_DATA_B_LENGTH != 0
-	    (addr >= L1_DATA_B_START && (addr + 2) <= (L1_DATA_B_START + L1_DATA_B_LENGTH)) ||
-#endif
-	    (addr >= L1_SCRATCH_START && (addr + 2) <= (L1_SCRATCH_START + L1_SCRATCH_LENGTH)) ||
-	    (!(bfin_read_EBIU_AMBCTL0() & B0RDYEN) &&
-	       addr >= ASYNC_BANK0_BASE && (addr + 2) <= (ASYNC_BANK0_BASE + ASYNC_BANK0_SIZE)) ||
-	    (!(bfin_read_EBIU_AMBCTL0() & B1RDYEN) &&
-	       addr >= ASYNC_BANK1_BASE && (addr + 2) <= (ASYNC_BANK1_BASE + ASYNC_BANK1_SIZE)) ||
-	    (!(bfin_read_EBIU_AMBCTL1() & B2RDYEN) &&
-	       addr >= ASYNC_BANK2_BASE && (addr + 2) <= (ASYNC_BANK2_BASE + ASYNC_BANK1_SIZE)) ||
-	    (!(bfin_read_EBIU_AMBCTL1() & B3RDYEN) &&
-	      addr >= ASYNC_BANK3_BASE && (addr + 2) <= (ASYNC_BANK3_BASE + ASYNC_BANK1_SIZE))) {
-		*val = *address;
-		return true;
+	switch (bfin_mem_access_type(addr, 2)) {
+		case BFIN_MEM_ACCESS_CORE:
+		case BFIN_MEM_ACCESS_CORE_ONLY:
+			*val = *address;
+			return true;
+		case BFIN_MEM_ACCESS_DMA:
+			dma_memcpy(val, address, 2);
+			return true;
+		case BFIN_MEM_ACCESS_ITEST:
+			isram_memcpy(val, address, 2);
+			return true;
+		default: /* invalid access */
+			return false;
 	}
-
-#if L1_CODE_LENGTH != 0
-	if (addr >= L1_CODE_START && (addr + 2) <= (L1_CODE_START + L1_CODE_LENGTH)) {
-		isram_memcpy(val, address, 2);
-		return true;
-	}
-#endif
-
-
-	return false;
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 2780cd64346782a6116e316c559d70a7655ab6e5 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Thu, 11 Jun 2009 09:22:02 -0400
Subject: Blackfin: bf518f-ezbrd: update DSA resources

The common DSA code changed structure layout, so update the BF518F-EZBRD
resources accordingly.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/mach-bf518/boards/ezbrd.c | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

diff --git a/arch/blackfin/mach-bf518/boards/ezbrd.c b/arch/blackfin/mach-bf518/boards/ezbrd.c
index 1382f0382359..d9791106be9f 100644
--- a/arch/blackfin/mach-bf518/boards/ezbrd.c
+++ b/arch/blackfin/mach-bf518/boards/ezbrd.c
@@ -119,13 +119,19 @@ static struct platform_device bfin_mac_device = {
 };
 
 #if defined(CONFIG_NET_DSA_KSZ8893M) || defined(CONFIG_NET_DSA_KSZ8893M_MODULE)
-static struct dsa_platform_data ksz8893m_switch_data = {
+static struct dsa_chip_data ksz8893m_switch_chip_data = {
 	.mii_bus = &bfin_mii_bus.dev,
+	.port_names = {
+		NULL,
+		"eth%d",
+		"eth%d",
+		"cpu",
+	},
+};
+static struct dsa_platform_data ksz8893m_switch_data = {
+	.nr_chips = 1,
 	.netdev = &bfin_mac_device.dev,
-	.port_names[0]	= NULL,
-	.port_names[1]	= "eth%d",
-	.port_names[2]	= "eth%d",
-	.port_names[3]	= "cpu",
+	.chip = &ksz8893m_switch_chip_data,
 };
 
 static struct platform_device ksz8893m_switch_device = {
-- 
cgit v1.2.3-59-g8ed1b


From 4d5e6fd42c137dad3b1aced073c6fcb494a8e507 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Sat, 13 Jun 2009 06:34:49 -0400
Subject: Blackfin: bf533-ezkit: add resources for FISP devices

The BF533-EZKIT has two Flash In-System Programming devices hooked up to
the async memory bus, so add resources for the primary flashes and the
SRAMs on the devices.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/mach-bf533/boards/ezkit.c | 106 ++++++++++++++++++++++++++++++++
 1 file changed, 106 insertions(+)

diff --git a/arch/blackfin/mach-bf533/boards/ezkit.c b/arch/blackfin/mach-bf533/boards/ezkit.c
index 89a5ec4ca048..4e3e511bf146 100644
--- a/arch/blackfin/mach-bf533/boards/ezkit.c
+++ b/arch/blackfin/mach-bf533/boards/ezkit.c
@@ -32,6 +32,7 @@
 #include <linux/platform_device.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/partitions.h>
+#include <linux/mtd/plat-ram.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/flash.h>
 #if defined(CONFIG_USB_ISP1362_HCD) || defined(CONFIG_USB_ISP1362_HCD_MODULE)
@@ -86,6 +87,101 @@ static struct platform_device smc91x_device = {
 };
 #endif
 
+#if defined(CONFIG_MTD_PSD4256G) || defined(CONFIG_MTD_PSD4256G_MODULE)
+static const char *map_probes[] = {
+	"stm_flash",
+	NULL,
+};
+
+static struct platdata_mtd_ram stm_pri_data_a = {
+	.mapname    = "Flash A Primary",
+	.map_probes = map_probes,
+	.bankwidth  = 2,
+};
+
+static struct resource stm_pri_resource_a = {
+	.start = 0x20000000,
+	.end   = 0x200fffff,
+	.flags = IORESOURCE_MEM,
+};
+
+static struct platform_device stm_pri_device_a = {
+	.name          = "mtd-ram",
+	.id            = 0,
+	.dev = {
+		.platform_data = &stm_pri_data_a,
+	},
+	.num_resources = 1,
+	.resource      = &stm_pri_resource_a,
+};
+
+static struct platdata_mtd_ram stm_pri_data_b = {
+	.mapname    = "Flash B Primary",
+	.map_probes = map_probes,
+	.bankwidth  = 2,
+};
+
+static struct resource stm_pri_resource_b = {
+	.start = 0x20100000,
+	.end   = 0x201fffff,
+	.flags = IORESOURCE_MEM,
+};
+
+static struct platform_device stm_pri_device_b = {
+	.name          = "mtd-ram",
+	.id            = 4,
+	.dev = {
+		.platform_data = &stm_pri_data_b,
+	},
+	.num_resources = 1,
+	.resource      = &stm_pri_resource_b,
+};
+#endif
+
+#if defined(CONFIG_MTD_PLATRAM) || defined(CONFIG_MTD_PLATRAM_MODULE)
+static struct platdata_mtd_ram sram_data_a = {
+	.mapname   = "Flash A SRAM",
+	.bankwidth = 2,
+};
+
+static struct resource sram_resource_a = {
+	.start = 0x20240000,
+	.end   = 0x2024ffff,
+	.flags = IORESOURCE_MEM,
+};
+
+static struct platform_device sram_device_a = {
+	.name          = "mtd-ram",
+	.id            = 8,
+	.dev = {
+		.platform_data = &sram_data_a,
+	},
+	.num_resources = 1,
+	.resource      = &sram_resource_a,
+};
+
+static struct platdata_mtd_ram sram_data_b = {
+	.mapname   = "Flash B SRAM",
+	.bankwidth = 2,
+};
+
+static struct resource sram_resource_b = {
+	.start = 0x202c0000,
+	.end   = 0x202cffff,
+	.flags = IORESOURCE_MEM,
+};
+
+static struct platform_device sram_device_b = {
+	.name          = "mtd-ram",
+	.id            = 9,
+	.dev = {
+		.platform_data = &sram_data_b,
+	},
+	.num_resources = 1,
+	.resource      = &sram_resource_b,
+};
+#endif
+
 #if defined(CONFIG_MTD_M25P80) || defined(CONFIG_MTD_M25P80_MODULE)
 static struct mtd_partition bfin_spi_flash_partitions[] = {
 	{
@@ -357,6 +453,16 @@ static struct platform_device *ezkit_devices[] __initdata = {
 
 	&bfin_dpmc,
 
+#if defined(CONFIG_MTD_PSD4256G) || defined(CONFIG_MTD_PSD4256G_MODULE)
+	&stm_pri_device_a,
+	&stm_pri_device_b,
+#endif
+
+#if defined(CONFIG_MTD_PLATRAM) || defined(CONFIG_MTD_PLATRAM_MODULE)
+	&sram_device_a,
+	&sram_device_b,
+#endif
+
 #if defined(CONFIG_SMC91X) || defined(CONFIG_SMC91X_MODULE)
 	&smc91x_device,
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From a200ad22bb15fe01cf222fa631687876baad5e01 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Sat, 13 Jun 2009 06:37:14 -0400
Subject: Blackfin: update anomaly lists

Update anomaly headers to match latest released anomaly sheets.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/mach-bf518/include/mach/anomaly.h | 37 ++++++----
 arch/blackfin/mach-bf527/include/mach/anomaly.h | 15 ++++-
 arch/blackfin/mach-bf533/include/mach/anomaly.h | 77 +++++++++++----------
 arch/blackfin/mach-bf537/include/mach/anomaly.h | 41 +++++++-----
 arch/blackfin/mach-bf538/include/mach/anomaly.h | 24 ++++---
 arch/blackfin/mach-bf548/include/mach/anomaly.h | 20 ++++--
 arch/blackfin/mach-bf561/include/mach/anomaly.h | 89 +++++++++++++------------
 7 files changed, 178 insertions(+), 125 deletions(-)

diff --git a/arch/blackfin/mach-bf518/include/mach/anomaly.h b/arch/blackfin/mach-bf518/include/mach/anomaly.h
index b69bd9af38dd..426e064062a0 100644
--- a/arch/blackfin/mach-bf518/include/mach/anomaly.h
+++ b/arch/blackfin/mach-bf518/include/mach/anomaly.h
@@ -7,7 +7,7 @@
  */
 
 /* This file should be up to date with:
- *  - Revision B, 02/03/2009; ADSP-BF512/BF514/BF516/BF518 Blackfin Processor Anomaly List
+ *  - Revision C, 06/12/2009; ADSP-BF512/BF514/BF516/BF518 Blackfin Processor Anomaly List
  */
 
 /* We plan on not supporting 0.0 silicon, but 0.1 isn't out yet - sorry */
@@ -18,7 +18,7 @@
 #ifndef _MACH_ANOMALY_H_
 #define _MACH_ANOMALY_H_
 
-/* Multi-issue instruction with dsp32shiftimm in slot1 and P-reg store in slot 2 not supported */
+/* Multi-Issue Instruction with dsp32shiftimm in slot1 and P-reg Store in slot2 Not Supported */
 #define ANOMALY_05000074 (1)
 /* Rx.H Cannot Be Used to Access 16-bit System MMR Registers */
 #define ANOMALY_05000122 (1)
@@ -45,29 +45,31 @@
 /* Speculative Fetches of Indirect-Pointer Instructions Can Cause False Hardware Errors */
 #define ANOMALY_05000426 (1)
 /* Software System Reset Corrupts PLL_LOCKCNT Register */
-#define ANOMALY_05000430 (1)
+#define ANOMALY_05000430 (__SILICON_REVISION__ < 1)
 /* Incorrect Use of Stack in Lockbox Firmware During Authentication */
 #define ANOMALY_05000431 (1)
 /* Certain SIC Registers are not Reset After Soft or Core Double Fault Reset */
-#define ANOMALY_05000435 (1)
+#define ANOMALY_05000435 (__SILICON_REVISION__ < 1)
 /* PORTx_DRIVE and PORTx_HYSTERESIS Registers Read Back Incorrect Values */
-#define ANOMALY_05000438 (1)
+#define ANOMALY_05000438 (__SILICON_REVISION__ < 1)
 /* Preboot Cannot be Used to Alter the PLL_DIV Register */
-#define ANOMALY_05000439 (1)
+#define ANOMALY_05000439 (__SILICON_REVISION__ < 1)
 /* bfrom_SysControl() Cannot be Used to Write the PLL_DIV Register */
-#define ANOMALY_05000440 (1)
+#define ANOMALY_05000440 (__SILICON_REVISION__ < 1)
 /* IFLUSH Instruction at End of Hardware Loop Causes Infinite Stall */
 #define ANOMALY_05000443 (1)
 /* Incorrect L1 Instruction Bank B Memory Map Location */
-#define ANOMALY_05000444 (1)
+#define ANOMALY_05000444 (__SILICON_REVISION__ < 1)
 /* Incorrect Default Hysteresis Setting for RESET, NMI, and BMODE Signals */
-#define ANOMALY_05000452 (1)
+#define ANOMALY_05000452 (__SILICON_REVISION__ < 1)
 /* PWM_TRIPB Signal Not Available on PG10 */
-#define ANOMALY_05000453 (1)
+#define ANOMALY_05000453 (__SILICON_REVISION__ < 1)
 /* PPI_FS3 is Driven One Half Cycle Later Than PPI Data */
-#define ANOMALY_05000455 (1)
-/* False Hardware Error when RETI points to invalid memory */
+#define ANOMALY_05000455 (__SILICON_REVISION__ < 1)
+/* False Hardware Error when RETI Points to Invalid Memory */
 #define ANOMALY_05000461 (1)
+/* Synchronization Problem at Startup May Cause SPORT Transmit Channels to Misalign */
+#define ANOMALY_05000462 (1)
 
 /* Anomalies that don't exist on this proc */
 #define ANOMALY_05000099 (0)
@@ -78,24 +80,30 @@
 #define ANOMALY_05000158 (0)
 #define ANOMALY_05000171 (0)
 #define ANOMALY_05000179 (0)
+#define ANOMALY_05000182 (0)
 #define ANOMALY_05000183 (0)
 #define ANOMALY_05000198 (0)
+#define ANOMALY_05000202 (0)
 #define ANOMALY_05000215 (0)
 #define ANOMALY_05000220 (0)
 #define ANOMALY_05000227 (0)
 #define ANOMALY_05000230 (0)
 #define ANOMALY_05000231 (0)
 #define ANOMALY_05000233 (0)
+#define ANOMALY_05000234 (0)
 #define ANOMALY_05000242 (0)
 #define ANOMALY_05000244 (0)
 #define ANOMALY_05000248 (0)
 #define ANOMALY_05000250 (0)
+#define ANOMALY_05000257 (0)
 #define ANOMALY_05000261 (0)
 #define ANOMALY_05000263 (0)
 #define ANOMALY_05000266 (0)
 #define ANOMALY_05000273 (0)
 #define ANOMALY_05000274 (0)
 #define ANOMALY_05000278 (0)
+#define ANOMALY_05000281 (0)
+#define ANOMALY_05000283 (0)
 #define ANOMALY_05000285 (0)
 #define ANOMALY_05000287 (0)
 #define ANOMALY_05000301 (0)
@@ -103,10 +111,13 @@
 #define ANOMALY_05000307 (0)
 #define ANOMALY_05000311 (0)
 #define ANOMALY_05000312 (0)
+#define ANOMALY_05000315 (0)
 #define ANOMALY_05000323 (0)
 #define ANOMALY_05000353 (0)
+#define ANOMALY_05000357 (0)
 #define ANOMALY_05000362 (1)
 #define ANOMALY_05000363 (0)
+#define ANOMALY_05000371 (0)
 #define ANOMALY_05000380 (0)
 #define ANOMALY_05000386 (0)
 #define ANOMALY_05000389 (0)
@@ -117,5 +128,7 @@
 #define ANOMALY_05000448 (0)
 #define ANOMALY_05000456 (0)
 #define ANOMALY_05000450 (0)
+#define ANOMALY_05000465 (0)
+#define ANOMALY_05000467 (0)
 
 #endif
diff --git a/arch/blackfin/mach-bf527/include/mach/anomaly.h b/arch/blackfin/mach-bf527/include/mach/anomaly.h
index c84ddea95749..0d63f7406168 100644
--- a/arch/blackfin/mach-bf527/include/mach/anomaly.h
+++ b/arch/blackfin/mach-bf527/include/mach/anomaly.h
@@ -34,7 +34,7 @@
 #define _ANOMALY_BF527(rev527) (ANOMALY_BF527 && __SILICON_REVISION__ rev527)
 #define _ANOMALY_BF526_BF527(rev526, rev527) (_ANOMALY_BF526(rev526) || _ANOMALY_BF527(rev527))
 
-/* Multi-issue instruction with dsp32shiftimm in slot1 and P-reg store in slot 2 not supported */
+/* Multi-Issue Instruction with dsp32shiftimm in slot1 and P-reg Store in slot2 Not Supported */
 #define ANOMALY_05000074 (1)
 /* DMA_RUN Bit Is Not Valid after a Peripheral Receive Channel DMA Stops */
 #define ANOMALY_05000119 (1)	/* note: brokenness is noted in documentation, not anomaly sheet */
@@ -184,8 +184,12 @@
 #define ANOMALY_05000456 (1)
 /* Host DMA Port Responds to Certain Bus Activity Without HOST_CE Assertion */
 #define ANOMALY_05000457 (1)
-/* False Hardware Error when RETI points to invalid memory */
+/* False Hardware Error when RETI Points to Invalid Memory */
 #define ANOMALY_05000461 (1)
+/* USB Rx DMA hang */
+#define ANOMALY_05000465 (1)
+/* Possible RX data corruption when control & data EP FIFOs are accessed via the core */
+#define ANOMALY_05000467 (1)
 
 /* Anomalies that don't exist on this proc */
 #define ANOMALY_05000099 (0)
@@ -195,24 +199,30 @@
 #define ANOMALY_05000158 (0)
 #define ANOMALY_05000171 (0)
 #define ANOMALY_05000179 (0)
+#define ANOMALY_05000182 (0)
 #define ANOMALY_05000183 (0)
 #define ANOMALY_05000198 (0)
+#define ANOMALY_05000202 (0)
 #define ANOMALY_05000215 (0)
 #define ANOMALY_05000220 (0)
 #define ANOMALY_05000227 (0)
 #define ANOMALY_05000230 (0)
 #define ANOMALY_05000231 (0)
 #define ANOMALY_05000233 (0)
+#define ANOMALY_05000234 (0)
 #define ANOMALY_05000242 (0)
 #define ANOMALY_05000244 (0)
 #define ANOMALY_05000248 (0)
 #define ANOMALY_05000250 (0)
+#define ANOMALY_05000257 (0)
 #define ANOMALY_05000261 (0)
 #define ANOMALY_05000263 (0)
 #define ANOMALY_05000266 (0)
 #define ANOMALY_05000273 (0)
 #define ANOMALY_05000274 (0)
 #define ANOMALY_05000278 (0)
+#define ANOMALY_05000281 (0)
+#define ANOMALY_05000283 (0)
 #define ANOMALY_05000285 (0)
 #define ANOMALY_05000287 (0)
 #define ANOMALY_05000301 (0)
@@ -220,6 +230,7 @@
 #define ANOMALY_05000307 (0)
 #define ANOMALY_05000311 (0)
 #define ANOMALY_05000312 (0)
+#define ANOMALY_05000315 (0)
 #define ANOMALY_05000323 (0)
 #define ANOMALY_05000362 (1)
 #define ANOMALY_05000363 (0)
diff --git a/arch/blackfin/mach-bf533/include/mach/anomaly.h b/arch/blackfin/mach-bf533/include/mach/anomaly.h
index 31145b509e20..70a0ad69c610 100644
--- a/arch/blackfin/mach-bf533/include/mach/anomaly.h
+++ b/arch/blackfin/mach-bf533/include/mach/anomaly.h
@@ -34,7 +34,7 @@
 # define ANOMALY_BF533 0
 #endif
 
-/* Multi-issue instruction with dsp32shiftimm in slot1 and P-reg store in slot 2 not supported */
+/* Multi-Issue Instruction with dsp32shiftimm in slot1 and P-reg Store in slot2 Not Supported */
 #define ANOMALY_05000074 (1)
 /* UART Line Status Register (UART_LSR) Bits Are Not Updated at the Same Time */
 #define ANOMALY_05000099 (__SILICON_REVISION__ < 5)
@@ -46,7 +46,7 @@
 #define ANOMALY_05000122 (1)
 /* Instruction DMA Can Cause Data Cache Fills to Fail (Boot Implications) */
 #define ANOMALY_05000158 (__SILICON_REVISION__ < 5)
-/* PPI Data Lengths Between 8 and 16 Do Not Zero Out Upper Bits */
+/* PPI Data Lengths between 8 and 16 Do Not Zero Out Upper Bits */
 #define ANOMALY_05000166 (1)
 /* Turning SPORTs on while External Frame Sync Is Active May Corrupt Data */
 #define ANOMALY_05000167 (1)
@@ -56,13 +56,13 @@
 #define ANOMALY_05000180 (1)
 /* Timer Pin Limitations for PPI TX Modes with External Frame Syncs */
 #define ANOMALY_05000183 (__SILICON_REVISION__ < 4)
-/* False Protection Exceptions */
+/* False Protection Exceptions when Speculative Fetch Is Cancelled */
 #define ANOMALY_05000189 (__SILICON_REVISION__ < 4)
 /* False I/O Pin Interrupts on Edge-Sensitive Inputs When Polarity Setting Is Changed */
 #define ANOMALY_05000193 (__SILICON_REVISION__ < 4)
 /* Restarting SPORT in Specific Modes May Cause Data Corruption */
 #define ANOMALY_05000194 (__SILICON_REVISION__ < 4)
-/* Failing MMR Accesses When Stalled by Preceding Memory Read */
+/* Failing MMR Accesses when Preceding Memory Read Stalls */
 #define ANOMALY_05000198 (__SILICON_REVISION__ < 5)
 /* Current DMA Address Shows Wrong Value During Carry Fix */
 #define ANOMALY_05000199 (__SILICON_REVISION__ < 4)
@@ -74,7 +74,7 @@
 #define ANOMALY_05000202 (__SILICON_REVISION__ < 5)
 /* Specific Sequence That Can Cause DMA Error or DMA Stopping */
 #define ANOMALY_05000203 (__SILICON_REVISION__ < 4)
-/* Incorrect data read with write-through cache and allocate cache lines on reads only mode */
+/* Incorrect Data Read with Writethrough "Allocate Cache Lines on Reads Only" Cache Mode */
 #define ANOMALY_05000204 (__SILICON_REVISION__ < 4 && ANOMALY_BF533)
 /* Recovery from "Brown-Out" Condition */
 #define ANOMALY_05000207 (__SILICON_REVISION__ < 4)
@@ -106,7 +106,7 @@
 #define ANOMALY_05000244 (__SILICON_REVISION__ < 5)
 /* False Hardware Error from an Access in the Shadow of a Conditional Branch */
 #define ANOMALY_05000245 (1)
-/* Data CPLBs Should Prevent Spurious Hardware Errors */
+/* Data CPLBs Should Prevent False Hardware Errors */
 #define ANOMALY_05000246 (__SILICON_REVISION__ < 5)
 /* Incorrect Bit Shift of Data Word in Multichannel (TDM) Mode in Certain Conditions */
 #define ANOMALY_05000250 (__SILICON_REVISION__ == 4)
@@ -148,21 +148,21 @@
 #define ANOMALY_05000277 (__SILICON_REVISION__ < 6)
 /* Disabling Peripherals with DMA Running May Cause DMA System Instability */
 #define ANOMALY_05000278 (__SILICON_REVISION__ < 6)
-/* False Hardware Error Exception When ISR Context Is Not Restored */
+/* False Hardware Error Exception when ISR Context Is Not Restored */
 #define ANOMALY_05000281 (__SILICON_REVISION__ < 6)
 /* Memory DMA Corruption with 32-Bit Data and Traffic Control */
 #define ANOMALY_05000282 (__SILICON_REVISION__ < 6)
-/* System MMR Write Is Stalled Indefinitely When Killed in a Particular Stage */
+/* System MMR Write Is Stalled Indefinitely when Killed in a Particular Stage */
 #define ANOMALY_05000283 (__SILICON_REVISION__ < 6)
 /* SPORTs May Receive Bad Data If FIFOs Fill Up */
 #define ANOMALY_05000288 (__SILICON_REVISION__ < 6)
 /* Memory-To-Memory DMA Source/Destination Descriptors Must Be in Same Memory Space */
 #define ANOMALY_05000301 (__SILICON_REVISION__ < 6)
-/* SSYNCs After Writes To DMA MMR Registers May Not Be Handled Correctly */
+/* SSYNCs after Writes to DMA MMR Registers May Not Be Handled Correctly */
 #define ANOMALY_05000302 (__SILICON_REVISION__ < 5)
 /* SPORT_HYS Bit in PLL_CTL Register Is Not Functional */
 #define ANOMALY_05000305 (__SILICON_REVISION__ < 5)
-/* New Feature: Additional PPI Frame Sync Sampling Options (Not Available On Older Silicon) */
+/* ALT_TIMING Bit in PPI_CONTROL Register Is Not Functional */
 #define ANOMALY_05000306 (__SILICON_REVISION__ < 5)
 /* SCKELOW Bit Does Not Maintain State Through Hibernate */
 #define ANOMALY_05000307 (1)	/* note: brokenness is noted in documentation, not anomaly sheet */
@@ -170,11 +170,11 @@
 #define ANOMALY_05000310 (1)
 /* Erroneous Flag (GPIO) Pin Operations under Specific Sequences */
 #define ANOMALY_05000311 (__SILICON_REVISION__ < 6)
-/* Errors When SSYNC, CSYNC, or Loads to LT, LB and LC Registers Are Interrupted */
+/* Errors when SSYNC, CSYNC, or Loads to LT, LB and LC Registers Are Interrupted */
 #define ANOMALY_05000312 (__SILICON_REVISION__ < 6)
 /* PPI Is Level-Sensitive on First Transfer In Single Frame Sync Modes */
 #define ANOMALY_05000313 (__SILICON_REVISION__ < 6)
-/* Killed System MMR Write Completes Erroneously On Next System MMR Access */
+/* Killed System MMR Write Completes Erroneously on Next System MMR Access */
 #define ANOMALY_05000315 (__SILICON_REVISION__ < 6)
 /* Internal Voltage Regulator Values of 1.05V, 1.10V and 1.15V Not Allowed for LQFP Packages */
 #define ANOMALY_05000319 ((ANOMALY_BF531 || ANOMALY_BF532) && __SILICON_REVISION__ < 6)
@@ -200,7 +200,7 @@
 #define ANOMALY_05000426 (1)
 /* IFLUSH Instruction at End of Hardware Loop Causes Infinite Stall */
 #define ANOMALY_05000443 (1)
-/* False Hardware Error when RETI points to invalid memory */
+/* False Hardware Error when RETI Points to Invalid Memory */
 #define ANOMALY_05000461 (1)
 
 /* These anomalies have been "phased" out of analog.com anomaly sheets and are
@@ -215,17 +215,17 @@
 #define ANOMALY_05000070 (__SILICON_REVISION__ < 2)
 /* Writing FIO_DIR can corrupt a programmable flag's data */
 #define ANOMALY_05000079 (__SILICON_REVISION__ < 2)
-/* Timer Auto-Baud Mode requires the UART clock to be enabled */
+/* Timer Auto-Baud Mode requires the UART clock to be enabled. */
 #define ANOMALY_05000086 (__SILICON_REVISION__ < 2)
 /* Internal Clocking Modes on SPORT0 not supported */
 #define ANOMALY_05000088 (__SILICON_REVISION__ < 2)
 /* Internal voltage regulator does not wake up from an RTC wakeup */
 #define ANOMALY_05000092 (__SILICON_REVISION__ < 2)
-/* The IFLUSH instruction must be preceded by a CSYNC instruction */
+/* The IFLUSH Instruction Must Be Preceded by a CSYNC Instruction */
 #define ANOMALY_05000093 (__SILICON_REVISION__ < 2)
-/* Vectoring to an instruction that is presently being filled into the instruction cache may cause erroneous behavior */
+/* Vectoring to instruction that is being filled into the i-cache may cause erroneous behavior */
 #define ANOMALY_05000095 (__SILICON_REVISION__ < 2)
-/* PREFETCH, FLUSH, and FLUSHINV must be followed by a CSYNC */
+/* PREFETCH, FLUSH, and FLUSHINV Instructions Must Be Followed by a CSYNC Instruction */
 #define ANOMALY_05000096 (__SILICON_REVISION__ < 2)
 /* Performance Monitor 0 and 1 are swapped when monitoring memory events */
 #define ANOMALY_05000097 (__SILICON_REVISION__ < 2)
@@ -235,45 +235,45 @@
 #define ANOMALY_05000100 (__SILICON_REVISION__ < 2)
 /* Reading X_MODIFY or Y_MODIFY while DMA channel is active */
 #define ANOMALY_05000101 (__SILICON_REVISION__ < 2)
-/* Descriptor-based MemDMA may lock up with 32-bit transfers or if transfers span 64KB buffers */
+/* Descriptor MemDMA may lock up with 32-bit transfers or if transfers span 64KB buffers */
 #define ANOMALY_05000102 (__SILICON_REVISION__ < 2)
-/* Incorrect value written to the cycle counters */
+/* Incorrect Value Written to the Cycle Counters */
 #define ANOMALY_05000103 (__SILICON_REVISION__ < 2)
-/* Stores to L1 Data memory incorrect when a specific sequence is followed */
+/* Stores to L1 Data Memory Incorrect when a Specific Sequence Is Followed */
 #define ANOMALY_05000104 (__SILICON_REVISION__ < 2)
 /* Programmable Flag (PF3) functionality not supported in all PPI modes */
 #define ANOMALY_05000106 (__SILICON_REVISION__ < 2)
 /* Data store can be lost when targeting a cache line fill */
 #define ANOMALY_05000107 (__SILICON_REVISION__ < 2)
-/* Reserved bits in SYSCFG register not set at power on */
+/* Reserved Bits in SYSCFG Register Not Set at Power-On */
 #define ANOMALY_05000109 (__SILICON_REVISION__ < 3)
 /* Infinite Core Stall */
 #define ANOMALY_05000114 (__SILICON_REVISION__ < 2)
-/* PPI_FSx may glitch when generated by the on chip Timers */
+/* PPI_FSx may glitch when generated by the on chip Timers. */
 #define ANOMALY_05000115 (__SILICON_REVISION__ < 2)
-/* Trace Buffers may record discontinuities into emulation mode and/or exception, NMI, reset handlers */
+/* Trace Buffers May Contain Errors in Emulation Mode and/or Exception, NMI, Reset Handlers */
 #define ANOMALY_05000116 (__SILICON_REVISION__ < 3)
 /* DTEST registers allow access to Data Cache when DTEST_COMMAND< 14 >= 0 */
 #define ANOMALY_05000117 (__SILICON_REVISION__ < 2)
 /* Booting from an 8-bit or 24-bit Addressable SPI device is not supported */
 #define ANOMALY_05000118 (__SILICON_REVISION__ < 2)
-/* DTEST_COMMAND initiated memory access may be incorrect if data cache or DMA is active */
+/* DTEST_COMMAND Initiated Memory Access May Be Incorrect If Data Cache or DMA Is Active */
 #define ANOMALY_05000123 (__SILICON_REVISION__ < 3)
 /* DMA Lock-up at CCLK to SCLK ratios of 4:1, 2:1, or 1:1 */
 #define ANOMALY_05000124 (__SILICON_REVISION__ < 3)
-/* Erroneous exception when enabling cache */
+/* Erroneous Exception when Enabling Cache */
 #define ANOMALY_05000125 (__SILICON_REVISION__ < 3)
 /* SPI clock polarity and phase bits incorrect during booting */
 #define ANOMALY_05000126 (__SILICON_REVISION__ < 3)
-/* DMEM_CONTROL is not set on Reset */
+/* DMEM_CONTROL<12> Is Not Set on Reset */
 #define ANOMALY_05000137 (__SILICON_REVISION__ < 3)
 /* SPI boot will not complete if there is a zero fill block in the loader file */
 #define ANOMALY_05000138 (__SILICON_REVISION__ == 2)
-/* Timerx_Config must be set for using the PPI in GP output mode with internal Frame Syncs */
+/* TIMERx_CONFIG[5] must be set for PPI in GP output mode with internal Frame Syncs */
 #define ANOMALY_05000139 (__SILICON_REVISION__ < 2)
 /* Allowing the SPORT RX FIFO to fill will cause an overflow */
 #define ANOMALY_05000140 (__SILICON_REVISION__ < 3)
-/* An Infinite Stall occurs with a particular sequence of consecutive dual dag events */
+/* Infinite Stall may occur with a particular sequence of consecutive dual dag events */
 #define ANOMALY_05000141 (__SILICON_REVISION__ < 3)
 /* Interrupts may be lost when a programmable input flag is configured to be edge sensitive */
 #define ANOMALY_05000142 (__SILICON_REVISION__ < 3)
@@ -287,7 +287,7 @@
 #define ANOMALY_05000146 (__SILICON_REVISION__ < 3)
 /* Source MDMA descriptor may stop with a DMA Error near beginning of descriptor fetch */
 #define ANOMALY_05000147 (__SILICON_REVISION__ < 3)
-/* When booting from a 16-bit asynchronous memory device, the upper 8-bits of each word must be 0x00 */
+/* When booting from 16-bit asynchronous memory, the upper 8 bits of each word must be 0x00 */
 #define ANOMALY_05000148 (__SILICON_REVISION__ < 3)
 /* Frame Delay in SPORT Multichannel Mode */
 #define ANOMALY_05000153 (__SILICON_REVISION__ < 3)
@@ -295,13 +295,13 @@
 #define ANOMALY_05000154 (__SILICON_REVISION__ < 3)
 /* Timer1 can not be used for PWMOUT mode when a certain PPI mode is in use */
 #define ANOMALY_05000155 (__SILICON_REVISION__ < 3)
-/* Killed 32-bit MMR write leads to next system MMR access thinking it should be 32-bit */
+/* Killed 32-Bit MMR Write Leads to Next System MMR Access Thinking It Should Be 32-Bit */
 #define ANOMALY_05000157 (__SILICON_REVISION__ < 3)
-/* SPORT transmit data is not gated by external frame sync in certain conditions */
+/* SPORT Transmit Data Is Not Gated by External Frame Sync in Certain Conditions */
 #define ANOMALY_05000163 (__SILICON_REVISION__ < 3)
-/* SDRAM auto-refresh and subsequent Power Ups */
+/* Undefined Behavior when Power-Up Sequence Is Issued to SDRAM during Auto-Refresh */
 #define ANOMALY_05000168 (__SILICON_REVISION__ < 3)
-/* DATA CPLB page miss can result in lost write-through cache data writes */
+/* DATA CPLB Page Miss Can Result in Lost Write-Through Data Cache Writes */
 #define ANOMALY_05000169 (__SILICON_REVISION__ < 3)
 /* DMA vs Core accesses to external memory */
 #define ANOMALY_05000173 (__SILICON_REVISION__ < 3)
@@ -309,15 +309,15 @@
 #define ANOMALY_05000174 (__SILICON_REVISION__ < 3)
 /* Overlapping Sequencer and Memory Stalls */
 #define ANOMALY_05000175 (__SILICON_REVISION__ < 3)
-/* Multiplication of (-1) by (-1) followed by an accumulator saturation */
+/* Overflow Bit Asserted when Multiplication of -1 by -1 Followed by Accumulator Saturation */
 #define ANOMALY_05000176 (__SILICON_REVISION__ < 3)
-/* Disabling the PPI resets the PPI configuration registers */
+/* Disabling the PPI Resets the PPI Configuration Registers */
 #define ANOMALY_05000181 (__SILICON_REVISION__ < 3)
-/* PPI TX Mode with 2 External Frame Syncs */
+/* Early PPI Transmit when FS1 Asserts before FS2 in TX Mode with 2 External Frame Syncs */
 #define ANOMALY_05000185 (__SILICON_REVISION__ < 3)
 /* PPI does not invert the Driving PPICLK edge in Transmit Modes */
 #define ANOMALY_05000191 (__SILICON_REVISION__ < 3)
-/* In PPI Transmit Modes with External Frame Syncs POLC */
+/* In PPI Transmit Modes with External Frame Syncs POLC bit must be set to 1 */
 #define ANOMALY_05000192 (__SILICON_REVISION__ < 3)
 /* Internal Voltage Regulator may not start up */
 #define ANOMALY_05000206 (__SILICON_REVISION__ < 3)
@@ -326,6 +326,7 @@
 #define ANOMALY_05000120 (0)
 #define ANOMALY_05000149 (0)
 #define ANOMALY_05000171 (0)
+#define ANOMALY_05000182 (0)
 #define ANOMALY_05000220 (0)
 #define ANOMALY_05000248 (0)
 #define ANOMALY_05000266 (0)
@@ -345,5 +346,7 @@
 #define ANOMALY_05000448 (0)
 #define ANOMALY_05000456 (0)
 #define ANOMALY_05000450 (0)
+#define ANOMALY_05000465 (0)
+#define ANOMALY_05000467 (0)
 
 #endif
diff --git a/arch/blackfin/mach-bf537/include/mach/anomaly.h b/arch/blackfin/mach-bf537/include/mach/anomaly.h
index fc9663425465..57c128cc3b64 100644
--- a/arch/blackfin/mach-bf537/include/mach/anomaly.h
+++ b/arch/blackfin/mach-bf537/include/mach/anomaly.h
@@ -34,13 +34,13 @@
 # define ANOMALY_BF537 0
 #endif
 
-/* Multi-issue instruction with dsp32shiftimm in slot1 and P-reg store in slot 2 not supported */
+/* Multi-Issue Instruction with dsp32shiftimm in slot1 and P-reg Store in slot2 Not Supported */
 #define ANOMALY_05000074 (1)
 /* DMA_RUN Bit Is Not Valid after a Peripheral Receive Channel DMA Stops */
 #define ANOMALY_05000119 (1)
 /* Rx.H Cannot Be Used to Access 16-bit System MMR Registers */
 #define ANOMALY_05000122 (1)
-/* Killed 32-bit MMR write leads to next system MMR access thinking it should be 32-bit */
+/* Killed 32-Bit MMR Write Leads to Next System MMR Access Thinking It Should Be 32-Bit */
 #define ANOMALY_05000157 (__SILICON_REVISION__ < 2)
 /* PPI_DELAY Not Functional in PPI Modes with 0 Frame Syncs */
 #define ANOMALY_05000180 (1)
@@ -50,11 +50,11 @@
 #define ANOMALY_05000244 (__SILICON_REVISION__ < 3)
 /* False Hardware Error from an Access in the Shadow of a Conditional Branch */
 #define ANOMALY_05000245 (1)
-/* CLKIN Buffer Output Enable Reset Behavior Is Changed */
+/* Buffered CLKIN Output Is Disabled by Default */
 #define ANOMALY_05000247 (1)
 /* Incorrect Bit Shift of Data Word in Multichannel (TDM) Mode in Certain Conditions */
 #define ANOMALY_05000250 (__SILICON_REVISION__ < 3)
-/* EMAC Tx DMA error after an early frame abort */
+/* EMAC TX DMA Error After an Early Frame Abort */
 #define ANOMALY_05000252 (__SILICON_REVISION__ < 3)
 /* Maximum External Clock Speed for Timers */
 #define ANOMALY_05000253 (__SILICON_REVISION__ < 3)
@@ -62,7 +62,7 @@
 #define ANOMALY_05000254 (__SILICON_REVISION__ > 2)
 /* Entering Hibernate State with RTC Seconds Interrupt Not Functional */
 #define ANOMALY_05000255 (__SILICON_REVISION__ < 3)
-/* EMAC MDIO input latched on wrong MDC edge */
+/* EMAC MDIO Input Latched on Wrong MDC Edge */
 #define ANOMALY_05000256 (__SILICON_REVISION__ < 3)
 /* Interrupt/Exception During Short Hardware Loop May Cause Bad Instruction Fetches */
 #define ANOMALY_05000257 (__SILICON_REVISION__ < 3)
@@ -80,7 +80,7 @@
 #define ANOMALY_05000264 (__SILICON_REVISION__ < 3)
 /* Sensitivity To Noise with Slow Input Edge Rates on External SPORT TX and RX Clocks */
 #define ANOMALY_05000265 (1)
-/* Memory DMA error when peripheral DMA is running with non-zero DEB_TRAFFIC_PERIOD */
+/* Memory DMA Error when Peripheral DMA Is Running with Non-Zero DEB_TRAFFIC_PERIOD */
 #define ANOMALY_05000268 (__SILICON_REVISION__ < 3)
 /* High I/O Activity Causes Output Voltage of Internal Voltage Regulator (Vddint) to Decrease */
 #define ANOMALY_05000270 (__SILICON_REVISION__ < 3)
@@ -92,15 +92,15 @@
 #define ANOMALY_05000277 (__SILICON_REVISION__ < 3)
 /* Disabling Peripherals with DMA Running May Cause DMA System Instability */
 #define ANOMALY_05000278 (((ANOMALY_BF536 || ANOMALY_BF537) && __SILICON_REVISION__ < 3) || (ANOMALY_BF534 && __SILICON_REVISION__ < 2))
-/* SPI Master boot mode does not work well with Atmel Data flash devices */
+/* SPI Master Boot Mode Does Not Work Well with Atmel Data Flash Devices */
 #define ANOMALY_05000280 (1)
-/* False Hardware Error Exception When ISR Context Is Not Restored */
+/* False Hardware Error Exception when ISR Context Is Not Restored */
 #define ANOMALY_05000281 (__SILICON_REVISION__ < 3)
 /* Memory DMA Corruption with 32-Bit Data and Traffic Control */
 #define ANOMALY_05000282 (__SILICON_REVISION__ < 3)
-/* System MMR Write Is Stalled Indefinitely When Killed in a Particular Stage */
+/* System MMR Write Is Stalled Indefinitely when Killed in a Particular Stage */
 #define ANOMALY_05000283 (__SILICON_REVISION__ < 3)
-/* New Feature: EMAC TX DMA Word Alignment (Not Available On Older Silicon) */
+/* TXDWA Bit in EMAC_SYSCTL Register Is Not Functional */
 #define ANOMALY_05000285 (__SILICON_REVISION__ < 3)
 /* SPORTs May Receive Bad Data If FIFOs Fill Up */
 #define ANOMALY_05000288 (__SILICON_REVISION__ < 3)
@@ -112,25 +112,25 @@
 #define ANOMALY_05000305 (__SILICON_REVISION__ < 3)
 /* SCKELOW Bit Does Not Maintain State Through Hibernate */
 #define ANOMALY_05000307 (__SILICON_REVISION__ < 3)
-/* Writing UART_THR while UART clock is disabled sends erroneous start bit */
+/* Writing UART_THR While UART Clock Is Disabled Sends Erroneous Start Bit */
 #define ANOMALY_05000309 (__SILICON_REVISION__ < 3)
 /* False Hardware Errors Caused by Fetches at the Boundary of Reserved Memory */
 #define ANOMALY_05000310 (1)
-/* Errors When SSYNC, CSYNC, or Loads to LT, LB and LC Registers Are Interrupted */
+/* Errors when SSYNC, CSYNC, or Loads to LT, LB and LC Registers Are Interrupted */
 #define ANOMALY_05000312 (1)
 /* PPI Is Level-Sensitive on First Transfer In Single Frame Sync Modes */
 #define ANOMALY_05000313 (1)
-/* Killed System MMR Write Completes Erroneously On Next System MMR Access */
+/* Killed System MMR Write Completes Erroneously on Next System MMR Access */
 #define ANOMALY_05000315 (__SILICON_REVISION__ < 3)
-/* EMAC RMII mode: collisions occur in Full Duplex mode */
+/* EMAC RMII Mode: Collisions Occur in Full Duplex Mode */
 #define ANOMALY_05000316 (__SILICON_REVISION__ < 3)
-/* EMAC RMII mode: TX frames in half duplex fail with status No Carrier */
+/* EMAC RMII Mode: TX Frames in Half Duplex Fail with Status "No Carrier" */
 #define ANOMALY_05000321 (__SILICON_REVISION__ < 3)
-/* EMAC RMII mode at 10-Base-T speed: RX frames not received properly */
+/* EMAC RMII Mode at 10-Base-T Speed: RX Frames Not Received Properly */
 #define ANOMALY_05000322 (1)
 /* Ethernet MAC MDIO Reads Do Not Meet IEEE Specification */
 #define ANOMALY_05000341 (__SILICON_REVISION__ >= 3)
-/* New Feature: UART Remains Enabled after UART Boot */
+/* UART Gets Disabled after UART Boot */
 #define ANOMALY_05000350 (__SILICON_REVISION__ >= 3)
 /* Regulator Programming Blocked when Hibernate Wakeup Source Remains Active */
 #define ANOMALY_05000355 (1)
@@ -154,7 +154,7 @@
 #define ANOMALY_05000426 (1)
 /* IFLUSH Instruction at End of Hardware Loop Causes Infinite Stall */
 #define ANOMALY_05000443 (1)
-/* False Hardware Error when RETI points to invalid memory */
+/* False Hardware Error when RETI Points to Invalid Memory */
 #define ANOMALY_05000461 (1)
 
 /* Anomalies that don't exist on this proc */
@@ -165,14 +165,17 @@
 #define ANOMALY_05000158 (0)
 #define ANOMALY_05000171 (0)
 #define ANOMALY_05000179 (0)
+#define ANOMALY_05000182 (0)
 #define ANOMALY_05000183 (0)
 #define ANOMALY_05000198 (0)
+#define ANOMALY_05000202 (0)
 #define ANOMALY_05000215 (0)
 #define ANOMALY_05000220 (0)
 #define ANOMALY_05000227 (0)
 #define ANOMALY_05000230 (0)
 #define ANOMALY_05000231 (0)
 #define ANOMALY_05000233 (0)
+#define ANOMALY_05000234 (0)
 #define ANOMALY_05000242 (0)
 #define ANOMALY_05000248 (0)
 #define ANOMALY_05000266 (0)
@@ -195,5 +198,7 @@
 #define ANOMALY_05000448 (0)
 #define ANOMALY_05000456 (0)
 #define ANOMALY_05000450 (0)
+#define ANOMALY_05000465 (0)
+#define ANOMALY_05000467 (0)
 
 #endif
diff --git a/arch/blackfin/mach-bf538/include/mach/anomaly.h b/arch/blackfin/mach-bf538/include/mach/anomaly.h
index 175ca9ef7232..c97acdf85cd3 100644
--- a/arch/blackfin/mach-bf538/include/mach/anomaly.h
+++ b/arch/blackfin/mach-bf538/include/mach/anomaly.h
@@ -30,13 +30,13 @@
 # define ANOMALY_BF539 0
 #endif
 
-/* Multi-issue instruction with dsp32shiftimm in slot1 and P-reg store in slot 2 not supported */
+/* Multi-Issue Instruction with dsp32shiftimm in slot1 and P-reg Store in slot2 Not Supported */
 #define ANOMALY_05000074 (1)
 /* DMA_RUN Bit Is Not Valid after a Peripheral Receive Channel DMA Stops */
 #define ANOMALY_05000119 (1)
 /* Rx.H Cannot Be Used to Access 16-bit System MMR Registers */
 #define ANOMALY_05000122 (1)
-/* PPI Data Lengths Between 8 and 16 Do Not Zero Out Upper Bits */
+/* PPI Data Lengths between 8 and 16 Do Not Zero Out Upper Bits */
 #define ANOMALY_05000166 (1)
 /* PPI_COUNT Cannot Be Programmed to 0 in General Purpose TX or RX Modes */
 #define ANOMALY_05000179 (1)
@@ -70,11 +70,11 @@
 #define ANOMALY_05000277 (__SILICON_REVISION__ < 4)
 /* Disabling Peripherals with DMA Running May Cause DMA System Instability */
 #define ANOMALY_05000278 (__SILICON_REVISION__ < 4)
-/* False Hardware Error Exception When ISR Context Is Not Restored */
+/* False Hardware Error Exception when ISR Context Is Not Restored */
 #define ANOMALY_05000281 (__SILICON_REVISION__ < 4)
 /* Memory DMA Corruption with 32-Bit Data and Traffic Control */
 #define ANOMALY_05000282 (__SILICON_REVISION__ < 4)
-/* System MMR Write Is Stalled Indefinitely When Killed in a Particular Stage */
+/* System MMR Write Is Stalled Indefinitely when Killed in a Particular Stage */
 #define ANOMALY_05000283 (__SILICON_REVISION__ < 4)
 /* SPORTs May Receive Bad Data If FIFOs Fill Up */
 #define ANOMALY_05000288 (__SILICON_REVISION__ < 4)
@@ -92,11 +92,11 @@
 #define ANOMALY_05000307 (__SILICON_REVISION__ < 4)
 /* False Hardware Errors Caused by Fetches at the Boundary of Reserved Memory */
 #define ANOMALY_05000310 (1)
-/* Errors When SSYNC, CSYNC, or Loads to LT, LB and LC Registers Are Interrupted */
+/* Errors when SSYNC, CSYNC, or Loads to LT, LB and LC Registers Are Interrupted */
 #define ANOMALY_05000312 (__SILICON_REVISION__ < 5)
 /* PPI Is Level-Sensitive on First Transfer In Single Frame Sync Modes */
 #define ANOMALY_05000313 (__SILICON_REVISION__ < 4)
-/* Killed System MMR Write Completes Erroneously On Next System MMR Access */
+/* Killed System MMR Write Completes Erroneously on Next System MMR Access */
 #define ANOMALY_05000315 (__SILICON_REVISION__ < 4)
 /* PFx Glitch on Write to FIO_FLAG_D or FIO_FLAG_T */
 #define ANOMALY_05000318 (ANOMALY_BF539 && __SILICON_REVISION__ < 4)
@@ -110,7 +110,7 @@
 #define ANOMALY_05000371 (__SILICON_REVISION__ < 5)
 /* Entering Hibernate State with Peripheral Wakeups Enabled Draws Excess Current */
 #define ANOMALY_05000374 (__SILICON_REVISION__ == 4)
-/* New Feature: Open-Drain GPIO Outputs on PC1 and PC4 (Not Available on Older Silicon) */
+/* GPIO Pins PC1 and PC4 Can Function as Normal Outputs */
 #define ANOMALY_05000375 (__SILICON_REVISION__ < 4)
 /* SSYNC Stalls Processor when Executed from Non-Cacheable Memory */
 #define ANOMALY_05000402 (__SILICON_REVISION__ < 4)
@@ -126,26 +126,32 @@
 #define ANOMALY_05000436 (__SILICON_REVISION__ > 3)
 /* IFLUSH Instruction at End of Hardware Loop Causes Infinite Stall */
 #define ANOMALY_05000443 (1)
-/* False Hardware Error when RETI points to invalid memory */
+/* False Hardware Error when RETI Points to Invalid Memory */
 #define ANOMALY_05000461 (1)
 
 /* Anomalies that don't exist on this proc */
 #define ANOMALY_05000099 (0)
 #define ANOMALY_05000120 (0)
+#define ANOMALY_05000125 (0)
 #define ANOMALY_05000149 (0)
 #define ANOMALY_05000158 (0)
 #define ANOMALY_05000171 (0)
+#define ANOMALY_05000182 (0)
 #define ANOMALY_05000198 (0)
+#define ANOMALY_05000202 (0)
 #define ANOMALY_05000215 (0)
 #define ANOMALY_05000220 (0)
 #define ANOMALY_05000227 (0)
 #define ANOMALY_05000230 (0)
 #define ANOMALY_05000231 (0)
+#define ANOMALY_05000234 (0)
 #define ANOMALY_05000242 (0)
 #define ANOMALY_05000248 (0)
 #define ANOMALY_05000250 (0)
 #define ANOMALY_05000254 (0)
+#define ANOMALY_05000257 (0)
 #define ANOMALY_05000263 (0)
+#define ANOMALY_05000266 (0)
 #define ANOMALY_05000274 (0)
 #define ANOMALY_05000287 (0)
 #define ANOMALY_05000305 (0)
@@ -166,5 +172,7 @@
 #define ANOMALY_05000448 (0)
 #define ANOMALY_05000456 (0)
 #define ANOMALY_05000450 (0)
+#define ANOMALY_05000465 (0)
+#define ANOMALY_05000467 (0)
 
 #endif
diff --git a/arch/blackfin/mach-bf548/include/mach/anomaly.h b/arch/blackfin/mach-bf548/include/mach/anomaly.h
index c510ae688e28..18a4cd24f673 100644
--- a/arch/blackfin/mach-bf548/include/mach/anomaly.h
+++ b/arch/blackfin/mach-bf548/include/mach/anomaly.h
@@ -18,7 +18,7 @@
 # error will not work on BF548 silicon version 0.0, or 0.1
 #endif
 
-/* Multi-issue instruction with dsp32shiftimm in slot1 and P-reg store in slot 2 not supported */
+/* Multi-Issue Instruction with dsp32shiftimm in slot1 and P-reg Store in slot2 Not Supported */
 #define ANOMALY_05000074 (1)
 /* DMA_RUN Bit Is Not Valid after a Peripheral Receive Channel DMA Stops */
 #define ANOMALY_05000119 (1)
@@ -30,17 +30,17 @@
 #define ANOMALY_05000265 (1)
 /* Certain Data Cache Writethrough Modes Fail for Vddint <= 0.9V */
 #define ANOMALY_05000272 (1)
-/* False Hardware Error Exception When ISR Context Is Not Restored */
+/* False Hardware Error Exception when ISR Context Is Not Restored */
 #define ANOMALY_05000281 (__SILICON_REVISION__ < 1)
 /* SSYNCs After Writes To CAN/DMA MMR Registers Are Not Always Handled Correctly */
 #define ANOMALY_05000304 (__SILICON_REVISION__ < 1)
 /* False Hardware Errors Caused by Fetches at the Boundary of Reserved Memory */
 #define ANOMALY_05000310 (1)
-/* Errors When SSYNC, CSYNC, or Loads to LT, LB and LC Registers Are Interrupted */
+/* Errors when SSYNC, CSYNC, or Loads to LT, LB and LC Registers Are Interrupted */
 #define ANOMALY_05000312 (__SILICON_REVISION__ < 1)
 /* TWI Slave Boot Mode Is Not Functional */
 #define ANOMALY_05000324 (__SILICON_REVISION__ < 1)
-/* External FIFO Boot Mode Is Not Functional */
+/* FIFO Boot Mode Not Functional */
 #define ANOMALY_05000325 (__SILICON_REVISION__ < 2)
 /* Data Lost When Core and DMA Accesses Are Made to the USB FIFO Simultaneously */
 #define ANOMALY_05000327 (__SILICON_REVISION__ < 1)
@@ -178,8 +178,12 @@
 #define ANOMALY_05000450 (1)
 /* USB Receive Interrupt Is Not Generated in DMA Mode 1 */
 #define ANOMALY_05000456 (__SILICON_REVISION__ < 3)
-/* False Hardware Error when RETI points to invalid memory */
+/* False Hardware Error when RETI Points to Invalid Memory */
 #define ANOMALY_05000461 (1)
+/* USB Rx DMA hang */
+#define ANOMALY_05000465 (1)
+/* Possible RX data corruption when control & data EP FIFOs are accessed via the core */
+#define ANOMALY_05000467 (1)
 
 /* Anomalies that don't exist on this proc */
 #define ANOMALY_05000099 (0)
@@ -189,30 +193,36 @@
 #define ANOMALY_05000158 (0)
 #define ANOMALY_05000171 (0)
 #define ANOMALY_05000179 (0)
+#define ANOMALY_05000182 (0)
 #define ANOMALY_05000183 (0)
 #define ANOMALY_05000198 (0)
+#define ANOMALY_05000202 (0)
 #define ANOMALY_05000215 (0)
 #define ANOMALY_05000220 (0)
 #define ANOMALY_05000227 (0)
 #define ANOMALY_05000230 (0)
 #define ANOMALY_05000231 (0)
 #define ANOMALY_05000233 (0)
+#define ANOMALY_05000234 (0)
 #define ANOMALY_05000242 (0)
 #define ANOMALY_05000244 (0)
 #define ANOMALY_05000248 (0)
 #define ANOMALY_05000250 (0)
 #define ANOMALY_05000254 (0)
+#define ANOMALY_05000257 (0)
 #define ANOMALY_05000261 (0)
 #define ANOMALY_05000263 (0)
 #define ANOMALY_05000266 (0)
 #define ANOMALY_05000273 (0)
 #define ANOMALY_05000274 (0)
 #define ANOMALY_05000278 (0)
+#define ANOMALY_05000283 (0)
 #define ANOMALY_05000287 (0)
 #define ANOMALY_05000301 (0)
 #define ANOMALY_05000305 (0)
 #define ANOMALY_05000307 (0)
 #define ANOMALY_05000311 (0)
+#define ANOMALY_05000315 (0)
 #define ANOMALY_05000323 (0)
 #define ANOMALY_05000362 (1)
 #define ANOMALY_05000363 (0)
diff --git a/arch/blackfin/mach-bf561/include/mach/anomaly.h b/arch/blackfin/mach-bf561/include/mach/anomaly.h
index dccd396cd931..94b8e277f09d 100644
--- a/arch/blackfin/mach-bf561/include/mach/anomaly.h
+++ b/arch/blackfin/mach-bf561/include/mach/anomaly.h
@@ -18,19 +18,19 @@
 # error will not work on BF561 silicon version 0.0, 0.1, 0.2, or 0.4
 #endif
 
-/* Multi-issue instruction with dsp32shiftimm in slot1 and P-reg store in slot 2 not supported */
+/* Multi-Issue Instruction with dsp32shiftimm in slot1 and P-reg Store in slot2 Not Supported */
 #define ANOMALY_05000074 (1)
 /* UART Line Status Register (UART_LSR) Bits Are Not Updated at the Same Time */
 #define ANOMALY_05000099 (__SILICON_REVISION__ < 5)
-/* Trace Buffers may record discontinuities into emulation mode and/or exception, NMI, reset handlers */
+/* Trace Buffers May Contain Errors in Emulation Mode and/or Exception, NMI, Reset Handlers */
 #define ANOMALY_05000116 (__SILICON_REVISION__ < 3)
-/* Testset instructions restricted to 32-bit aligned memory locations */
+/* TESTSET Instructions Restricted to 32-Bit Aligned Memory Locations */
 #define ANOMALY_05000120 (1)
 /* Rx.H Cannot Be Used to Access 16-bit System MMR Registers */
 #define ANOMALY_05000122 (1)
-/* Erroneous exception when enabling cache */
+/* Erroneous Exception when Enabling Cache */
 #define ANOMALY_05000125 (__SILICON_REVISION__ < 3)
-/* Signbits instruction not functional under certain conditions */
+/* SIGNBITS Instruction Not Functional under Certain Conditions */
 #define ANOMALY_05000127 (1)
 /* Two bits in the Watchpoint Status Register (WPSTAT) are swapped */
 #define ANOMALY_05000134 (__SILICON_REVISION__ < 3)
@@ -40,7 +40,7 @@
 #define ANOMALY_05000136 (__SILICON_REVISION__ < 3)
 /* Allowing the SPORT RX FIFO to fill will cause an overflow */
 #define ANOMALY_05000140 (__SILICON_REVISION__ < 3)
-/* An Infinite Stall occurs with a particular sequence of consecutive dual dag events */
+/* Infinite Stall may occur with a particular sequence of consecutive dual dag events */
 #define ANOMALY_05000141 (__SILICON_REVISION__ < 3)
 /* Interrupts may be lost when a programmable input flag is configured to be edge sensitive */
 #define ANOMALY_05000142 (__SILICON_REVISION__ < 3)
@@ -52,7 +52,7 @@
 #define ANOMALY_05000146 (__SILICON_REVISION__ < 3)
 /* Source MDMA descriptor may stop with a DMA Error near beginning of descriptor fetch */
 #define ANOMALY_05000147 (__SILICON_REVISION__ < 3)
-/* IMDMA S1/D1 channel may stall */
+/* IMDMA S1/D1 Channel May Stall */
 #define ANOMALY_05000149 (1)
 /* DMA engine may lose data due to incorrect handshaking */
 #define ANOMALY_05000150 (__SILICON_REVISION__ < 3)
@@ -66,7 +66,7 @@
 #define ANOMALY_05000154 (__SILICON_REVISION__ < 3)
 /* Timers in PWM-Out Mode with PPI GP Receive (Input) Mode with 0 Frame Syncs */
 #define ANOMALY_05000156 (__SILICON_REVISION__ < 4)
-/* Killed 32-bit MMR write leads to next system MMR access thinking it should be 32-bit */
+/* Killed 32-Bit MMR Write Leads to Next System MMR Access Thinking It Should Be 32-Bit */
 #define ANOMALY_05000157 (__SILICON_REVISION__ < 3)
 /* DMA Lock-up at CCLK to SCLK ratios of 4:1, 2:1, or 1:1 */
 #define ANOMALY_05000159 (__SILICON_REVISION__ < 3)
@@ -76,17 +76,17 @@
 #define ANOMALY_05000161 (__SILICON_REVISION__ < 3)
 /* DMEM_CONTROL<12> is not set on Reset */
 #define ANOMALY_05000162 (__SILICON_REVISION__ < 3)
-/* SPORT transmit data is not gated by external frame sync in certain conditions */
+/* SPORT Transmit Data Is Not Gated by External Frame Sync in Certain Conditions */
 #define ANOMALY_05000163 (__SILICON_REVISION__ < 3)
-/* PPI Data Lengths Between 8 and 16 Do Not Zero Out Upper Bits */
+/* PPI Data Lengths between 8 and 16 Do Not Zero Out Upper Bits */
 #define ANOMALY_05000166 (1)
 /* Turning SPORTs on while External Frame Sync Is Active May Corrupt Data */
 #define ANOMALY_05000167 (1)
-/* SDRAM auto-refresh and subsequent Power Ups */
+/* Undefined Behavior when Power-Up Sequence Is Issued to SDRAM during Auto-Refresh */
 #define ANOMALY_05000168 (__SILICON_REVISION__ < 5)
-/* DATA CPLB page miss can result in lost write-through cache data writes */
+/* DATA CPLB Page Miss Can Result in Lost Write-Through Data Cache Writes */
 #define ANOMALY_05000169 (__SILICON_REVISION__ < 5)
-/* Boot-ROM code modifies SICA_IWRx wakeup registers */
+/* Boot-ROM Modifies SICA_IWRx Wakeup Registers */
 #define ANOMALY_05000171 (__SILICON_REVISION__ < 5)
 /* DSPID register values incorrect */
 #define ANOMALY_05000172 (__SILICON_REVISION__ < 3)
@@ -96,29 +96,29 @@
 #define ANOMALY_05000174 (__SILICON_REVISION__ < 5)
 /* Overlapping Sequencer and Memory Stalls */
 #define ANOMALY_05000175 (__SILICON_REVISION__ < 5)
-/* Multiplication of (-1) by (-1) followed by an accumulator saturation */
+/* Overflow Bit Asserted when Multiplication of -1 by -1 Followed by Accumulator Saturation */
 #define ANOMALY_05000176 (__SILICON_REVISION__ < 5)
 /* PPI_COUNT Cannot Be Programmed to 0 in General Purpose TX or RX Modes */
 #define ANOMALY_05000179 (__SILICON_REVISION__ < 5)
 /* PPI_DELAY Not Functional in PPI Modes with 0 Frame Syncs */
 #define ANOMALY_05000180 (1)
-/* Disabling the PPI resets the PPI configuration registers */
+/* Disabling the PPI Resets the PPI Configuration Registers */
 #define ANOMALY_05000181 (__SILICON_REVISION__ < 5)
-/* IMDMA does not operate to full speed for 600MHz and higher devices */
+/* Internal Memory DMA Does Not Operate at Full Speed */
 #define ANOMALY_05000182 (1)
-/* Timer Pin limitations for PPI TX Modes with External Frame Syncs */
+/* Timer Pin Limitations for PPI TX Modes with External Frame Syncs */
 #define ANOMALY_05000184 (__SILICON_REVISION__ < 5)
-/* PPI TX Mode with 2 External Frame Syncs */
+/* Early PPI Transmit when FS1 Asserts before FS2 in TX Mode with 2 External Frame Syncs */
 #define ANOMALY_05000185 (__SILICON_REVISION__ < 5)
-/* PPI packing with Data Length greater than 8 bits (not a meaningful mode) */
+/* Upper PPI Pins Driven when PPI Packing Enabled and Data Length >8 Bits */
 #define ANOMALY_05000186 (__SILICON_REVISION__ < 5)
 /* IMDMA Corrupted Data after a Halt */
 #define ANOMALY_05000187 (1)
 /* IMDMA Restrictions on Descriptor and Buffer Placement in Memory */
 #define ANOMALY_05000188 (__SILICON_REVISION__ < 5)
-/* False Protection Exceptions */
+/* False Protection Exceptions when Speculative Fetch Is Cancelled */
 #define ANOMALY_05000189 (__SILICON_REVISION__ < 5)
-/* PPI not functional at core voltage < 1Volt */
+/* PPI Not Functional at Core Voltage < 1Volt */
 #define ANOMALY_05000190 (1)
 /* PPI does not invert the Driving PPICLK edge in Transmit Modes */
 #define ANOMALY_05000191 (__SILICON_REVISION__ < 3)
@@ -126,7 +126,7 @@
 #define ANOMALY_05000193 (__SILICON_REVISION__ < 5)
 /* Restarting SPORT in Specific Modes May Cause Data Corruption */
 #define ANOMALY_05000194 (__SILICON_REVISION__ < 5)
-/* Failing MMR Accesses When Stalled by Preceding Memory Read */
+/* Failing MMR Accesses when Preceding Memory Read Stalls */
 #define ANOMALY_05000198 (__SILICON_REVISION__ < 5)
 /* Current DMA Address Shows Wrong Value During Carry Fix */
 #define ANOMALY_05000199 (__SILICON_REVISION__ < 5)
@@ -134,9 +134,9 @@
 #define ANOMALY_05000200 (__SILICON_REVISION__ < 5)
 /* Possible Infinite Stall with Specific Dual-DAG Situation */
 #define ANOMALY_05000202 (__SILICON_REVISION__ < 5)
-/* Incorrect data read with write-through cache and allocate cache lines on reads only mode */
+/* Incorrect Data Read with Writethrough "Allocate Cache Lines on Reads Only" Cache Mode */
 #define ANOMALY_05000204 (__SILICON_REVISION__ < 5)
-/* Specific sequence that can cause DMA error or DMA stopping */
+/* Specific Sequence that Can Cause DMA Error or DMA Stopping */
 #define ANOMALY_05000205 (__SILICON_REVISION__ < 5)
 /* Recovery from "Brown-Out" Condition */
 #define ANOMALY_05000207 (__SILICON_REVISION__ < 5)
@@ -158,7 +158,7 @@
 #define ANOMALY_05000230 (__SILICON_REVISION__ < 5)
 /* UART STB Bit Incorrectly Affects Receiver Setting */
 #define ANOMALY_05000231 (__SILICON_REVISION__ < 5)
-/* SPORT data transmit lines are incorrectly driven in multichannel mode */
+/* SPORT Data Transmit Lines Are Incorrectly Driven in Multichannel Mode */
 #define ANOMALY_05000232 (__SILICON_REVISION__ < 5)
 /* DF Bit in PLL_CTL Register Does Not Respond to Hardware Reset */
 #define ANOMALY_05000242 (__SILICON_REVISION__ < 5)
@@ -166,7 +166,7 @@
 #define ANOMALY_05000244 (__SILICON_REVISION__ < 5)
 /* False Hardware Error from an Access in the Shadow of a Conditional Branch */
 #define ANOMALY_05000245 (__SILICON_REVISION__ < 5)
-/* TESTSET operation forces stall on the other core */
+/* TESTSET Operation Forces Stall on the Other Core */
 #define ANOMALY_05000248 (__SILICON_REVISION__ < 5)
 /* Incorrect Bit Shift of Data Word in Multichannel (TDM) Mode in Certain Conditions */
 #define ANOMALY_05000250 (__SILICON_REVISION__ > 2 && __SILICON_REVISION__ < 5)
@@ -192,9 +192,9 @@
 #define ANOMALY_05000264 (__SILICON_REVISION__ < 5)
 /* Sensitivity To Noise with Slow Input Edge Rates on External SPORT TX and RX Clocks */
 #define ANOMALY_05000265 (__SILICON_REVISION__ < 5)
-/* IMDMA destination IRQ status must be read prior to using IMDMA */
+/* IMDMA Destination IRQ Status Must Be Read Prior to Using IMDMA */
 #define ANOMALY_05000266 (__SILICON_REVISION__ > 3)
-/* IMDMA may corrupt data under certain conditions */
+/* IMDMA May Corrupt Data under Certain Conditions */
 #define ANOMALY_05000267 (1)
 /* High I/O Activity Causes Output Voltage of Internal Voltage Regulator (Vddint) to Increase */
 #define ANOMALY_05000269 (1)
@@ -202,7 +202,7 @@
 #define ANOMALY_05000270 (1)
 /* Certain Data Cache Writethrough Modes Fail for Vddint <= 0.9V */
 #define ANOMALY_05000272 (1)
-/* Data cache write back to external synchronous memory may be lost */
+/* Data Cache Write Back to External Synchronous Memory May Be Lost */
 #define ANOMALY_05000274 (1)
 /* PPI Timing and Sampling Information Updates */
 #define ANOMALY_05000275 (__SILICON_REVISION__ > 2)
@@ -212,17 +212,17 @@
 #define ANOMALY_05000277 (__SILICON_REVISION__ < 3)
 /* Disabling Peripherals with DMA Running May Cause DMA System Instability */
 #define ANOMALY_05000278 (__SILICON_REVISION__ < 5)
-/* False Hardware Error Exception When ISR Context Is Not Restored */
+/* False Hardware Error Exception when ISR Context Is Not Restored */
 #define ANOMALY_05000281 (__SILICON_REVISION__ < 5)
-/* System MMR Write Is Stalled Indefinitely When Killed in a Particular Stage */
+/* System MMR Write Is Stalled Indefinitely when Killed in a Particular Stage */
 #define ANOMALY_05000283 (1)
-/* A read will receive incorrect data under certain conditions */
+/* Reads Will Receive Incorrect Data under Certain Conditions */
 #define ANOMALY_05000287 (__SILICON_REVISION__ < 5)
 /* SPORTs May Receive Bad Data If FIFOs Fill Up */
 #define ANOMALY_05000288 (__SILICON_REVISION__ < 5)
 /* Memory-To-Memory DMA Source/Destination Descriptors Must Be in Same Memory Space */
 #define ANOMALY_05000301 (1)
-/* SSYNCs After Writes To DMA MMR Registers May Not Be Handled Correctly */
+/* SSYNCs after Writes to DMA MMR Registers May Not Be Handled Correctly */
 #define ANOMALY_05000302 (1)
 /* SPORT_HYS Bit in PLL_CTL Register Is Not Functional */
 #define ANOMALY_05000305 (__SILICON_REVISION__ < 5)
@@ -230,25 +230,25 @@
 #define ANOMALY_05000307 (__SILICON_REVISION__ < 5)
 /* False Hardware Errors Caused by Fetches at the Boundary of Reserved Memory */
 #define ANOMALY_05000310 (1)
-/* Errors When SSYNC, CSYNC, or Loads to LT, LB and LC Registers Are Interrupted */
+/* Errors when SSYNC, CSYNC, or Loads to LT, LB and LC Registers Are Interrupted */
 #define ANOMALY_05000312 (1)
 /* PPI Is Level-Sensitive on First Transfer In Single Frame Sync Modes */
 #define ANOMALY_05000313 (1)
-/* Killed System MMR Write Completes Erroneously On Next System MMR Access */
+/* Killed System MMR Write Completes Erroneously on Next System MMR Access */
 #define ANOMALY_05000315 (1)
-/* PF2 Output Remains Asserted After SPI Master Boot */
+/* PF2 Output Remains Asserted after SPI Master Boot */
 #define ANOMALY_05000320 (__SILICON_REVISION__ > 3)
-/* Erroneous GPIO Flag Pin Operations Under Specific Sequences */
+/* Erroneous GPIO Flag Pin Operations under Specific Sequences */
 #define ANOMALY_05000323 (1)
-/* SPORT Secondary Receive Channel Not Functional When Word Length Exceeds 16 Bits */
+/* SPORT Secondary Receive Channel Not Functional when Word Length >16 Bits */
 #define ANOMALY_05000326 (__SILICON_REVISION__ > 3)
-/* New Feature: 24-Bit SPI Boot Mode Support (Not Available On Older Silicon) */
+/* 24-Bit SPI Boot Mode Is Not Functional */
 #define ANOMALY_05000331 (__SILICON_REVISION__ < 5)
-/* New Feature: Slave SPI Boot Mode Supported (Not Available On Older Silicon) */
+/* Slave SPI Boot Mode Is Not Functional */
 #define ANOMALY_05000332 (__SILICON_REVISION__ < 5)
-/* Flag Data Register Writes One SCLK Cycle After Edge Is Detected May Clear Interrupt Status */
+/* Flag Data Register Writes One SCLK Cycle after Edge Is Detected May Clear Interrupt Status */
 #define ANOMALY_05000333 (__SILICON_REVISION__ < 5)
-/* New Feature: Additional PPI Frame Sync Sampling Options (Not Available on Older Silicon) */
+/* ALT_TIMING Bit in PLL_CTL Register Is Not Functional */
 #define ANOMALY_05000339 (__SILICON_REVISION__ < 5)
 /* Memory DMA FIFO Causes Throughput Degradation on Writes to External Memory */
 #define ANOMALY_05000343 (__SILICON_REVISION__ < 5)
@@ -276,7 +276,7 @@
 #define ANOMALY_05000428 (__SILICON_REVISION__ > 3)
 /* IFLUSH Instruction at End of Hardware Loop Causes Infinite Stall */
 #define ANOMALY_05000443 (1)
-/* False Hardware Error when RETI points to invalid memory */
+/* False Hardware Error when RETI Points to Invalid Memory */
 #define ANOMALY_05000461 (1)
 
 /* Anomalies that don't exist on this proc */
@@ -284,6 +284,7 @@
 #define ANOMALY_05000158 (0)
 #define ANOMALY_05000183 (0)
 #define ANOMALY_05000233 (0)
+#define ANOMALY_05000234 (0)
 #define ANOMALY_05000273 (0)
 #define ANOMALY_05000311 (0)
 #define ANOMALY_05000353 (1)
@@ -298,5 +299,7 @@
 #define ANOMALY_05000448 (0)
 #define ANOMALY_05000456 (0)
 #define ANOMALY_05000450 (0)
+#define ANOMALY_05000465 (0)
+#define ANOMALY_05000467 (0)
 
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 26579216f3cdf1ae05f0af8412b444870a167510 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Mon, 15 Jun 2009 06:10:03 -0400
Subject: Blackfin: redo handling of bad irqs

With the common IRQ code initializing much more of the irq_desc state, we
can't blindly initialize it ourselves to the local bad_irq state.  If we
do, we end up wrongly clobbering many fields.  So punt most of the bad irq
code as the common layers will handle the default state, and simply call
handle_bad_irq() directly when the IRQ we are processing is invalid.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/kernel/irqchip.c | 53 +++++++++---------------------------------
 1 file changed, 11 insertions(+), 42 deletions(-)

diff --git a/arch/blackfin/kernel/irqchip.c b/arch/blackfin/kernel/irqchip.c
index 6e31e935bb31..36bba3027735 100644
--- a/arch/blackfin/kernel/irqchip.c
+++ b/arch/blackfin/kernel/irqchip.c
@@ -38,38 +38,15 @@
 #include <asm/pda.h>
 
 static atomic_t irq_err_count;
-static spinlock_t irq_controller_lock;
-
-/*
- * Dummy mask/unmask handler
- */
-void dummy_mask_unmask_irq(unsigned int irq)
-{
-}
-
 void ack_bad_irq(unsigned int irq)
 {
 	atomic_inc(&irq_err_count);
 	printk(KERN_ERR "IRQ: spurious interrupt %d\n", irq);
 }
 
-static struct irq_chip bad_chip = {
-	.ack = dummy_mask_unmask_irq,
-	.mask = dummy_mask_unmask_irq,
-	.unmask = dummy_mask_unmask_irq,
-};
-
-static int bad_stats;
 static struct irq_desc bad_irq_desc = {
-	.status = IRQ_DISABLED,
-	.chip = &bad_chip,
 	.handle_irq = handle_bad_irq,
-	.depth = 1,
 	.lock = __SPIN_LOCK_UNLOCKED(irq_desc->lock),
-	.kstat_irqs = &bad_stats,
-#ifdef CONFIG_SMP
-	.affinity = CPU_MASK_ALL
-#endif
 };
 
 #ifdef CONFIG_CPUMASK_OFFSTACK
@@ -119,21 +96,13 @@ __attribute__((l1_text))
 #endif
 asmlinkage void asm_do_IRQ(unsigned int irq, struct pt_regs *regs)
 {
-	struct pt_regs *old_regs;
-	struct irq_desc *desc = irq_desc + irq;
 #ifndef CONFIG_IPIPE
 	unsigned short pending, other_ints;
 #endif
-	old_regs = set_irq_regs(regs);
-
-	/*
-	 * Some hardware gives randomly wrong interrupts.  Rather
-	 * than crashing, do something sensible.
-	 */
-	if (irq >= NR_IRQS)
-		desc = &bad_irq_desc;
+	struct pt_regs *old_regs = set_irq_regs(regs);
 
 	irq_enter();
+
 #ifdef CONFIG_DEBUG_STACKOVERFLOW
 	/* Debugging check for stack overflow: is there less than STACK_WARN free? */
 	{
@@ -149,7 +118,15 @@ asmlinkage void asm_do_IRQ(unsigned int irq, struct pt_regs *regs)
 		}
 	}
 #endif
-	generic_handle_irq(irq);
+
+	/*
+	 * Some hardware gives randomly wrong interrupts.  Rather
+	 * than crashing, do something sensible.
+	 */
+	if (irq >= NR_IRQS)
+		handle_bad_irq(irq, &bad_irq_desc);
+	else
+		generic_handle_irq(irq);
 
 #ifndef CONFIG_IPIPE
 	/*
@@ -173,14 +150,6 @@ asmlinkage void asm_do_IRQ(unsigned int irq, struct pt_regs *regs)
 
 void __init init_IRQ(void)
 {
-	struct irq_desc *desc;
-	int irq;
-
-	spin_lock_init(&irq_controller_lock);
-	for (irq = 0, desc = irq_desc; irq < NR_IRQS; irq++, desc++) {
-		*desc = bad_irq_desc;
-	}
-
 	init_arch_irq();
 
 #ifdef CONFIG_DEBUG_BFIN_HWTRACE_EXPAND
-- 
cgit v1.2.3-59-g8ed1b


From 46f288a0f983401ebadb918751d342cbf819cde5 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Mon, 15 Jun 2009 06:13:58 -0400
Subject: Blackfin: only build show_interrupts() when procfs is enabled

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/kernel/irqchip.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/blackfin/kernel/irqchip.c b/arch/blackfin/kernel/irqchip.c
index 36bba3027735..8abb642283b4 100644
--- a/arch/blackfin/kernel/irqchip.c
+++ b/arch/blackfin/kernel/irqchip.c
@@ -54,6 +54,7 @@ static struct irq_desc bad_irq_desc = {
 #error "Blackfin architecture does not support CONFIG_CPUMASK_OFFSTACK."
 #endif
 
+#ifdef CONFIG_PROC_FS
 int show_interrupts(struct seq_file *p, void *v)
 {
 	int i = *(loff_t *) v, j;
@@ -85,6 +86,7 @@ int show_interrupts(struct seq_file *p, void *v)
 	}
 	return 0;
 }
+#endif
 
 /*
  * do_IRQ handles all hardware IRQs.  Decoded IRQs should not
-- 
cgit v1.2.3-59-g8ed1b


From 6f10fdabdce356aac3c948e659f39b6f1e2f7382 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Mon, 15 Jun 2009 06:18:38 -0400
Subject: Blackfin: simplify irq stack overflow checking

Take a page from x86 and abstract the stack checking out of the
asm_do_IRQ() function so that the result is easier to digest.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/kernel/irqchip.c | 32 +++++++++++++++++---------------
 1 file changed, 17 insertions(+), 15 deletions(-)

diff --git a/arch/blackfin/kernel/irqchip.c b/arch/blackfin/kernel/irqchip.c
index 8abb642283b4..7378440792a4 100644
--- a/arch/blackfin/kernel/irqchip.c
+++ b/arch/blackfin/kernel/irqchip.c
@@ -88,6 +88,22 @@ int show_interrupts(struct seq_file *p, void *v)
 }
 #endif
 
+#ifdef CONFIG_DEBUG_STACKOVERFLOW
+static void check_stack_overflow(int irq)
+{
+	/* Debugging check for stack overflow: is there less than STACK_WARN free? */
+	long sp = __get_SP() & (THREAD_SIZE - 1);
+
+	if (unlikely(sp < (sizeof(struct thread_info) + STACK_WARN))) {
+		dump_stack();
+		pr_emerg("irq%i: possible stack overflow only %ld bytes free\n",
+			irq, sp - sizeof(struct thread_info));
+	}
+}
+#else
+static inline void check_stack_overflow(int irq) { }
+#endif
+
 /*
  * do_IRQ handles all hardware IRQs.  Decoded IRQs should not
  * come via this function.  Instead, they should provide their
@@ -105,21 +121,7 @@ asmlinkage void asm_do_IRQ(unsigned int irq, struct pt_regs *regs)
 
 	irq_enter();
 
-#ifdef CONFIG_DEBUG_STACKOVERFLOW
-	/* Debugging check for stack overflow: is there less than STACK_WARN free? */
-	{
-		long sp;
-
-		sp = __get_SP() & (THREAD_SIZE-1);
-
-		if (unlikely(sp < (sizeof(struct thread_info) + STACK_WARN))) {
-			dump_stack();
-			printk(KERN_EMERG "%s: possible stack overflow while handling irq %i "
-					" only %ld bytes free\n",
-				__func__, irq, sp - sizeof(struct thread_info));
-		}
-	}
-#endif
+	check_stack_overflow(irq);
 
 	/*
 	 * Some hardware gives randomly wrong interrupts.  Rather
-- 
cgit v1.2.3-59-g8ed1b


From 81b79c213d0200fdd16951a9fb18748fd511d810 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Mon, 15 Jun 2009 06:22:08 -0400
Subject: Blackfin: abstract irq14 lowering in do_irq

Split out the optional IRQ14 lowering code to further simplify the
asm_do_IRQ() function and keep the ifdef nest under control.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/kernel/irqchip.c | 43 ++++++++++++++++++++++++------------------
 1 file changed, 25 insertions(+), 18 deletions(-)

diff --git a/arch/blackfin/kernel/irqchip.c b/arch/blackfin/kernel/irqchip.c
index 7378440792a4..4b5fd36187d9 100644
--- a/arch/blackfin/kernel/irqchip.c
+++ b/arch/blackfin/kernel/irqchip.c
@@ -104,6 +104,29 @@ static void check_stack_overflow(int irq)
 static inline void check_stack_overflow(int irq) { }
 #endif
 
+#ifndef CONFIG_IPIPE
+static void maybe_lower_to_irq14(void)
+{
+	unsigned short pending, other_ints;
+
+	/*
+	 * If we're the only interrupt running (ignoring IRQ15 which
+	 * is for syscalls), lower our priority to IRQ14 so that
+	 * softirqs run at that level.  If there's another,
+	 * lower-level interrupt, irq_exit will defer softirqs to
+	 * that. If the interrupt pipeline is enabled, we are already
+	 * running at IRQ14 priority, so we don't need this code.
+	 */
+	CSYNC();
+	pending = bfin_read_IPEND() & ~0x8000;
+	other_ints = pending & (pending - 1);
+	if (other_ints == 0)
+		lower_to_irq14();
+}
+#else
+static inline void maybe_lower_to_irq14(void) { }
+#endif
+
 /*
  * do_IRQ handles all hardware IRQs.  Decoded IRQs should not
  * come via this function.  Instead, they should provide their
@@ -114,9 +137,6 @@ __attribute__((l1_text))
 #endif
 asmlinkage void asm_do_IRQ(unsigned int irq, struct pt_regs *regs)
 {
-#ifndef CONFIG_IPIPE
-	unsigned short pending, other_ints;
-#endif
 	struct pt_regs *old_regs = set_irq_regs(regs);
 
 	irq_enter();
@@ -132,21 +152,8 @@ asmlinkage void asm_do_IRQ(unsigned int irq, struct pt_regs *regs)
 	else
 		generic_handle_irq(irq);
 
-#ifndef CONFIG_IPIPE
-	/*
-	 * If we're the only interrupt running (ignoring IRQ15 which
-	 * is for syscalls), lower our priority to IRQ14 so that
-	 * softirqs run at that level.  If there's another,
-	 * lower-level interrupt, irq_exit will defer softirqs to
-	 * that. If the interrupt pipeline is enabled, we are already
-	 * running at IRQ14 priority, so we don't need this code.
-	 */
-	CSYNC();
-	pending = bfin_read_IPEND() & ~0x8000;
-	other_ints = pending & (pending - 1);
-	if (other_ints == 0)
-		lower_to_irq14();
-#endif /* !CONFIG_IPIPE */
+	maybe_lower_to_irq14();
+
 	irq_exit();
 
 	set_irq_regs(old_regs);
-- 
cgit v1.2.3-59-g8ed1b


From 0de4adfb8c9674fa1572b0ff1371acc94b0be901 Mon Sep 17 00:00:00 2001
From: Sonic Zhang <sonic.zhang@analog.com>
Date: Mon, 15 Jun 2009 07:39:19 +0000
Subject: Blackfin: fix accidental reset in some boot modes

We read the SWRST (Software Reset) register to get at the last reset
state, and then we may configure the DOUBLE_FAULT bit to control behavior
when a double fault occurs.  But if the lower bits of the register is
already set (like UART boot mode on a BF54x), we inadvertently make the
system reset by writing to the SYSTEM_RESET field at the same time.  So
make sure the lower 4 bits are always cleared.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/kernel/setup.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c
index 6454babdfaff..b2782eae31e1 100644
--- a/arch/blackfin/kernel/setup.c
+++ b/arch/blackfin/kernel/setup.c
@@ -837,7 +837,8 @@ void __init setup_arch(char **cmdline_p)
     defined(CONFIG_BF538) || defined(CONFIG_BF539)
 	_bfin_swrst = bfin_read_SWRST();
 #else
-	_bfin_swrst = bfin_read_SYSCR();
+	/* Clear boot mode field */
+	_bfin_swrst = bfin_read_SYSCR() & ~0xf;
 #endif
 
 #ifdef CONFIG_DEBUG_DOUBLEFAULT_PRINT
-- 
cgit v1.2.3-59-g8ed1b


From dc2c46bb702629d20a3786e10b540c7dcf2c017f Mon Sep 17 00:00:00 2001
From: Graf Yang <graf.yang@analog.com>
Date: Mon, 15 Jun 2009 08:23:41 +0000
Subject: Blackfin: bf526-ezbrd: set SPI flash resources to SST device

The BF526-EZBRD has a SST SPI flash on it, not a ST Micro.

Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/mach-bf527/boards/ezbrd.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/blackfin/mach-bf527/boards/ezbrd.c b/arch/blackfin/mach-bf527/boards/ezbrd.c
index 9f9c0005dcf1..b2f30f06b73e 100644
--- a/arch/blackfin/mach-bf527/boards/ezbrd.c
+++ b/arch/blackfin/mach-bf527/boards/ezbrd.c
@@ -237,10 +237,10 @@ static struct flash_platform_data bfin_spi_flash_data = {
 	.name = "m25p80",
 	.parts = bfin_spi_flash_partitions,
 	.nr_parts = ARRAY_SIZE(bfin_spi_flash_partitions),
-	.type = "m25p16",
+	.type = "sst25wf040",
 };
 
-/* SPI flash chip (m25p64) */
+/* SPI flash chip (sst25wf040) */
 static struct bfin5xx_spi_chip spi_flash_chip_info = {
 	.enable_dma = 0,         /* use dma transfer with this chip*/
 	.bits_per_word = 8,
-- 
cgit v1.2.3-59-g8ed1b


From 3d15f302d089d0583463745cbece077c1e8294b1 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Mon, 15 Jun 2009 16:21:44 +0000
Subject: Blackfin: allow people to select BF51x-0.1 silicon rev

Now that 0.1 of the BF51x is coming out, allow people to build for it.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index 8ea0d942cdea..c2c4a62b6762 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -274,7 +274,7 @@ config BF_REV_0_0
 
 config BF_REV_0_1
 	bool "0.1"
-	depends on (BF52x || (BF54x && !BF54xM))
+	depends on (BF51x || BF52x || (BF54x && !BF54xM))
 
 config BF_REV_0_2
 	bool "0.2"
-- 
cgit v1.2.3-59-g8ed1b


From 06ecc190f3928850cb77c498f745fc8e9a7e2fd7 Mon Sep 17 00:00:00 2001
From: Philippe Gerum <rpm@xenomai.org>
Date: Tue, 16 Jun 2009 05:25:37 +0200
Subject: Blackfin: convert interrupt pipeline to irqflags

Signed-off-by: Philippe Gerum <rpm@xenomai.org>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/include/asm/ipipe_base.h |  30 +++---
 arch/blackfin/include/asm/irq.h        |   7 --
 arch/blackfin/include/asm/irqflags.h   | 164 +++++++++++++++++++++++++++++++--
 arch/blackfin/include/asm/system.h     |   4 +-
 arch/blackfin/kernel/ipipe.c           |   2 +-
 5 files changed, 177 insertions(+), 30 deletions(-)

diff --git a/arch/blackfin/include/asm/ipipe_base.h b/arch/blackfin/include/asm/ipipe_base.h
index 3e8acbd1a3be..490098f532a7 100644
--- a/arch/blackfin/include/asm/ipipe_base.h
+++ b/arch/blackfin/include/asm/ipipe_base.h
@@ -51,23 +51,23 @@
 
 extern unsigned long __ipipe_root_status; /* Alias to ipipe_root_cpudom_var(status) */
 
-static inline void __ipipe_stall_root(void)
-{
-	volatile unsigned long *p = &__ipipe_root_status;
-	set_bit(0, p);
-}
+#define __ipipe_stall_root()						\
+	do {								\
+		volatile unsigned long *p = &__ipipe_root_status;	\
+		set_bit(0, p);						\
+	} while (0)
 
-static inline unsigned long __ipipe_test_and_stall_root(void)
-{
-	volatile unsigned long *p = &__ipipe_root_status;
-	return test_and_set_bit(0, p);
-}
+#define __ipipe_test_and_stall_root()					\
+	({								\
+		volatile unsigned long *p = &__ipipe_root_status;	\
+		test_and_set_bit(0, p);					\
+	})
 
-static inline unsigned long __ipipe_test_root(void)
-{
-	const unsigned long *p = &__ipipe_root_status;
-	return test_bit(0, p);
-}
+#define __ipipe_test_root()					\
+	({							\
+		const unsigned long *p = &__ipipe_root_status;	\
+		test_bit(0, p);					\
+	})
 
 #endif /* !__ASSEMBLY__ */
 
diff --git a/arch/blackfin/include/asm/irq.h b/arch/blackfin/include/asm/irq.h
index 9a7f63a83c47..42a15f5ce0d0 100644
--- a/arch/blackfin/include/asm/irq.h
+++ b/arch/blackfin/include/asm/irq.h
@@ -22,13 +22,6 @@
 /* SYS_IRQS and NR_IRQS are defined in <mach-bf5xx/irq.h> */
 #include <mach/irq.h>
 
-/* Xenomai IPIPE helpers */
-#define local_irq_restore_hw(x) local_irq_restore(x)
-#define local_irq_save_hw(x)    local_irq_save(x)
-#define local_irq_enable_hw(x)  local_irq_enable(x)
-#define local_irq_disable_hw(x) local_irq_disable(x)
-#define irqs_disabled_hw(x)     irqs_disabled(x)
-
 #if ANOMALY_05000244 && defined(CONFIG_BFIN_ICACHE)
 # define NOP_PAD_ANOMALY_05000244 "nop; nop;"
 #else
diff --git a/arch/blackfin/include/asm/irqflags.h b/arch/blackfin/include/asm/irqflags.h
index 139cba4651b1..9b19a19d9ae9 100644
--- a/arch/blackfin/include/asm/irqflags.h
+++ b/arch/blackfin/include/asm/irqflags.h
@@ -31,6 +31,150 @@ static inline unsigned long bfin_cli(void)
 	return flags;
 }
 
+#ifdef CONFIG_IPIPE
+
+#include <linux/ipipe_base.h>
+#include <linux/ipipe_trace.h>
+
+#ifdef CONFIG_DEBUG_HWERR
+# define bfin_no_irqs 0x3f
+#else
+# define bfin_no_irqs 0x1f
+#endif
+
+#define raw_local_irq_disable()				\
+	do {						\
+		ipipe_check_context(ipipe_root_domain);	\
+		__ipipe_stall_root();			\
+		barrier();				\
+	} while (0)
+
+static inline void raw_local_irq_enable(void)
+{
+	barrier();
+	ipipe_check_context(ipipe_root_domain);
+	__ipipe_unstall_root();
+}
+
+#define raw_local_save_flags_ptr(x)					\
+	do {								\
+		*(x) = __ipipe_test_root() ? bfin_no_irqs : bfin_irq_flags; \
+	} while (0)
+
+#define raw_local_save_flags(x)		raw_local_save_flags_ptr(&(x))
+
+#define raw_irqs_disabled_flags(x)	((x) == bfin_no_irqs)
+
+#define raw_local_irq_save_ptr(x)					\
+	do {								\
+		*(x) = __ipipe_test_and_stall_root() ? bfin_no_irqs : bfin_irq_flags; \
+		barrier();						\
+	} while (0)
+
+#define raw_local_irq_save(x)				\
+	do {						\
+		ipipe_check_context(ipipe_root_domain);	\
+		raw_local_irq_save_ptr(&(x));		\
+	} while (0)
+
+static inline unsigned long raw_mangle_irq_bits(int virt, unsigned long real)
+{
+	/*
+	 * Merge virtual and real interrupt mask bits into a single
+	 * 32bit word.
+	 */
+	return (real & ~(1 << 31)) | ((virt != 0) << 31);
+}
+
+static inline int raw_demangle_irq_bits(unsigned long *x)
+{
+	int virt = (*x & (1 << 31)) != 0;
+	*x &= ~(1L << 31);
+	return virt;
+}
+
+static inline void local_irq_disable_hw_notrace(void)
+{
+	bfin_cli();
+}
+
+static inline void local_irq_enable_hw_notrace(void)
+{
+	bfin_sti(bfin_irq_flags);
+}
+
+#define local_save_flags_hw(flags)			\
+	do {						\
+		(flags) = bfin_read_IMASK();		\
+	} while (0)
+
+#define irqs_disabled_flags_hw(flags) (((flags) & ~0x3f) == 0)
+
+#define irqs_disabled_hw()			\
+	({					\
+	unsigned long flags;			\
+	local_save_flags_hw(flags);		\
+	irqs_disabled_flags_hw(flags);		\
+	})
+
+static inline void local_irq_save_ptr_hw(unsigned long *flags)
+{
+	*flags = bfin_cli();
+#ifdef CONFIG_DEBUG_HWERR
+	bfin_sti(0x3f);
+#endif
+}
+
+#define local_irq_save_hw_notrace(flags)		\
+	do {						\
+		local_irq_save_ptr_hw(&(flags));	\
+	} while (0)
+
+static inline void local_irq_restore_hw_notrace(unsigned long flags)
+{
+	if (!irqs_disabled_flags_hw(flags))
+		local_irq_enable_hw_notrace();
+}
+
+#ifdef CONFIG_IPIPE_TRACE_IRQSOFF
+# define local_irq_disable_hw()				\
+	do {						\
+		if (!irqs_disabled_hw()) {		\
+			local_irq_disable_hw_notrace();	\
+			ipipe_trace_begin(0x80000000);	\
+		}					\
+	} while (0)
+# define local_irq_enable_hw()				\
+	do {						\
+		if (irqs_disabled_hw()) {		\
+			ipipe_trace_end(0x80000000);	\
+			local_irq_enable_hw_notrace();	\
+		}					\
+	} while (0)
+# define local_irq_save_hw(flags)			\
+	do {						\
+		local_save_flags_hw(flags);		\
+		if (!irqs_disabled_flags_hw(flags)) {	\
+			local_irq_disable_hw_notrace();	\
+			ipipe_trace_begin(0x80000001);	\
+		}					\
+	} while (0)
+# define local_irq_restore_hw(flags)			\
+	do {						\
+		if (!irqs_disabled_flags_hw(flags)) {	\
+			ipipe_trace_end(0x80000001);	\
+			local_irq_enable_hw_notrace();	\
+		}					\
+	} while (0)
+#else /* !CONFIG_IPIPE_TRACE_IRQSOFF */
+# define local_irq_disable_hw()		local_irq_disable_hw_notrace()
+# define local_irq_enable_hw()		local_irq_enable_hw_notrace()
+# define local_irq_save_hw(flags)	local_irq_save_hw_notrace(flags)
+# define local_irq_restore_hw(flags)	local_irq_restore_hw_notrace(flags)
+#endif /* !CONFIG_IPIPE_TRACE_IRQSOFF */
+
+#else /* CONFIG_IPIPE */
+
 static inline void raw_local_irq_disable(void)
 {
 	bfin_cli();
@@ -44,12 +188,6 @@ static inline void raw_local_irq_enable(void)
 
 #define raw_irqs_disabled_flags(flags) (((flags) & ~0x3f) == 0)
 
-static inline void raw_local_irq_restore(unsigned long flags)
-{
-	if (!raw_irqs_disabled_flags(flags))
-		raw_local_irq_enable();
-}
-
 static inline unsigned long __raw_local_irq_save(void)
 {
 	unsigned long flags = bfin_cli();
@@ -60,4 +198,18 @@ static inline unsigned long __raw_local_irq_save(void)
 }
 #define raw_local_irq_save(flags) do { (flags) = __raw_local_irq_save(); } while (0)
 
+#define local_irq_save_hw(flags)	raw_local_irq_save(flags)
+#define local_irq_restore_hw(flags)	raw_local_irq_restore(flags)
+#define local_irq_enable_hw()		raw_local_irq_enable()
+#define local_irq_disable_hw()		raw_local_irq_disable()
+#define irqs_disabled_hw()		irqs_disabled()
+
+#endif /* !CONFIG_IPIPE */
+
+static inline void raw_local_irq_restore(unsigned long flags)
+{
+	if (!raw_irqs_disabled_flags(flags))
+		raw_local_irq_enable();
+}
+
 #endif
diff --git a/arch/blackfin/include/asm/system.h b/arch/blackfin/include/asm/system.h
index 294dbda24164..85e8f16cf8c2 100644
--- a/arch/blackfin/include/asm/system.h
+++ b/arch/blackfin/include/asm/system.h
@@ -135,11 +135,13 @@ struct __xchg_dummy {
 };
 #define __xg(x) ((volatile struct __xchg_dummy *)(x))
 
+#include <mach/blackfin.h>
+
 static inline unsigned long __xchg(unsigned long x, volatile void *ptr,
 				   int size)
 {
 	unsigned long tmp = 0;
-	unsigned long flags = 0;
+	unsigned long flags;
 
 	local_irq_save_hw(flags);
 
diff --git a/arch/blackfin/kernel/ipipe.c b/arch/blackfin/kernel/ipipe.c
index d8cde1fc5cb9..2b979edee140 100644
--- a/arch/blackfin/kernel/ipipe.c
+++ b/arch/blackfin/kernel/ipipe.c
@@ -52,7 +52,7 @@ EXPORT_SYMBOL(__ipipe_freq_scale);
 
 atomic_t __ipipe_irq_lvdepth[IVG15 + 1];
 
-unsigned long __ipipe_irq_lvmask = __all_masked_irq_flags;
+unsigned long __ipipe_irq_lvmask = bfin_no_irqs;
 EXPORT_SYMBOL(__ipipe_irq_lvmask);
 
 static void __ipipe_ack_irq(unsigned irq, struct irq_desc *desc)
-- 
cgit v1.2.3-59-g8ed1b


From a40494a62a11dbaf326397aa94b2018ead09884d Mon Sep 17 00:00:00 2001
From: Philippe Gerum <rpm@xenomai.org>
Date: Tue, 16 Jun 2009 05:25:42 +0200
Subject: Blackfin: allow CONFIG_TICKSOURCE_GPTMR0 with interrupt pipeline

Signed-off-by: Philippe Gerum <rpm@xenomai.org>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/Kconfig                     |  1 -
 arch/blackfin/include/asm/ipipe.h         |  7 ++++-
 arch/blackfin/mach-common/ints-priority.c | 47 +++++++++++++++----------------
 3 files changed, 28 insertions(+), 27 deletions(-)

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index c2c4a62b6762..8140a2fc8bbd 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -623,7 +623,6 @@ choice
 config TICKSOURCE_GPTMR0
 	bool "Gptimer0 (SCLK domain)"
 	select BFIN_GPTIMERS
-	depends on !IPIPE
 
 config TICKSOURCE_CORETMR
 	bool "Core timer (CCLK domain)"
diff --git a/arch/blackfin/include/asm/ipipe.h b/arch/blackfin/include/asm/ipipe.h
index bbe1c3726b69..ffa5e60faa9f 100644
--- a/arch/blackfin/include/asm/ipipe.h
+++ b/arch/blackfin/include/asm/ipipe.h
@@ -207,7 +207,7 @@ void ipipe_init_irq_threads(void);
 
 int ipipe_start_irq_thread(unsigned irq, struct irq_desc *desc);
 
-#ifdef CONFIG_GENERIC_CLOCKEVENTS
+#ifdef CONFIG_TICKSOURCE_CORETMR
 #define IRQ_SYSTMR		IRQ_CORETMR
 #define IRQ_PRIOTMR		IRQ_CORETMR
 #else
@@ -240,8 +240,13 @@ int ipipe_start_irq_thread(unsigned irq, struct irq_desc *desc);
 #define ipipe_init_irq_threads()		do { } while (0)
 #define ipipe_start_irq_thread(irq, desc)	0
 
+#ifndef CONFIG_TICKSOURCE_GPTMR0
 #define IRQ_SYSTMR		IRQ_CORETMR
 #define IRQ_PRIOTMR		IRQ_CORETMR
+#else
+#define IRQ_SYSTMR		IRQ_TIMER0
+#define IRQ_PRIOTMR		CONFIG_IRQ_TIMER0
+#endif
 
 #define __ipipe_root_tick_p(regs)	1
 
diff --git a/arch/blackfin/mach-common/ints-priority.c b/arch/blackfin/mach-common/ints-priority.c
index af70f09acd55..b42150190d0e 100644
--- a/arch/blackfin/mach-common/ints-priority.c
+++ b/arch/blackfin/mach-common/ints-priority.c
@@ -1052,35 +1052,34 @@ int __init init_arch_irq(void)
 			set_irq_chained_handler(irq, bfin_demux_error_irq);
 			break;
 #endif
-#if defined(CONFIG_TICKSOURCE_GPTMR0)
-		case IRQ_TIMER0:
-			set_irq_handler(irq, handle_percpu_irq);
-			break;
-#endif
 #ifdef CONFIG_SMP
 		case IRQ_SUPPLE_0:
 		case IRQ_SUPPLE_1:
 			set_irq_handler(irq, handle_percpu_irq);
 			break;
 #endif
-		default:
 #ifdef CONFIG_IPIPE
-			/*
-			 * We want internal interrupt sources to be
-			 * masked, because ISRs may trigger interrupts
-			 * recursively (e.g. DMA), but interrupts are
-			 * _not_ masked at CPU level. So let's handle
-			 * most of them as level interrupts, except
-			 * the timer interrupt which is special.
-			 */
-			if (irq == IRQ_SYSTMR || irq == IRQ_CORETMR)
-				set_irq_handler(irq, handle_simple_irq);
-			else
-				set_irq_handler(irq, handle_level_irq);
+#ifndef CONFIG_TICKSOURCE_CORETMR
+		case IRQ_TIMER0:
+			set_irq_handler(irq, handle_simple_irq);
+			break;
+#endif /* !CONFIG_TICKSOURCE_CORETMR */
+		case IRQ_CORETMR:
+			set_irq_handler(irq, handle_simple_irq);
+			break;
+		default:
+			set_irq_handler(irq, handle_level_irq);
+			break;
 #else /* !CONFIG_IPIPE */
+#ifdef CONFIG_TICKSOURCE_GPTMR0
+		case IRQ_TIMER0:
+			set_irq_handler(irq, handle_percpu_irq);
+			break;
+#endif /* CONFIG_TICKSOURCE_GPTMR0 */
+		default:
 			set_irq_handler(irq, handle_simple_irq);
-#endif /* !CONFIG_IPIPE */
 			break;
+#endif	/* !CONFIG_IPIPE */
 		}
 	}
 
@@ -1224,15 +1223,14 @@ __attribute__((l1_text))
 asmlinkage int __ipipe_grab_irq(int vec, struct pt_regs *regs)
 {
 	struct ipipe_percpu_domain_data *p = ipipe_root_cpudom_ptr();
-	struct ipipe_domain *this_domain = ipipe_current_domain;
+	struct ipipe_domain *this_domain = __ipipe_current_domain;
 	struct ivgx *ivg_stop = ivg7_13[vec-IVG7].istop;
 	struct ivgx *ivg = ivg7_13[vec-IVG7].ifirst;
 	int irq, s;
 
-	if (likely(vec == EVT_IVTMR_P)) {
+	if (likely(vec == EVT_IVTMR_P))
 		irq = IRQ_CORETMR;
-
-	} else {
+	else {
 #if defined(SIC_ISR0) || defined(SICA_ISR0)
 		unsigned long sic_status[3];
 
@@ -1262,12 +1260,11 @@ asmlinkage int __ipipe_grab_irq(int vec, struct pt_regs *regs)
 				break;
 		}
 #endif
-
 		irq = ivg->irqno;
 	}
 
 	if (irq == IRQ_SYSTMR) {
-#ifndef CONFIG_GENERIC_CLOCKEVENTS
+#if !defined(CONFIG_GENERIC_CLOCKEVENTS) || defined(CONFIG_TICKSOURCE_GPTMR0)
 		bfin_write_TIMER_STATUS(1); /* Latch TIMIL0 */
 #endif
 		/* This is basically what we need from the register frame. */
-- 
cgit v1.2.3-59-g8ed1b


From 5ba3b249c9e08100b0822f17276348b3447d6ee3 Mon Sep 17 00:00:00 2001
From: Philippe Gerum <rpm@xenomai.org>
Date: Tue, 16 Jun 2009 05:25:50 +0200
Subject: Blackfin: remove obsolete mcount support from I-pipe code

Signed-off-by: Philippe Gerum <rpm@xenomai.org>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/kernel/Makefile |  1 -
 arch/blackfin/kernel/ipipe.c  |  5 ----
 arch/blackfin/kernel/mcount.S | 70 -------------------------------------------
 3 files changed, 76 deletions(-)
 delete mode 100644 arch/blackfin/kernel/mcount.S

diff --git a/arch/blackfin/kernel/Makefile b/arch/blackfin/kernel/Makefile
index 3731088e181b..141d9281e4b0 100644
--- a/arch/blackfin/kernel/Makefile
+++ b/arch/blackfin/kernel/Makefile
@@ -20,7 +20,6 @@ obj-$(CONFIG_FUNCTION_GRAPH_TRACER)  += ftrace.o
 CFLAGS_REMOVE_ftrace.o = -pg
 
 obj-$(CONFIG_IPIPE)                  += ipipe.o
-obj-$(CONFIG_IPIPE_TRACE_MCOUNT)     += mcount.o
 obj-$(CONFIG_BFIN_GPTIMERS)          += gptimers.o
 obj-$(CONFIG_CPLB_INFO)              += cplbinfo.o
 obj-$(CONFIG_MODULES)                += module.o
diff --git a/arch/blackfin/kernel/ipipe.c b/arch/blackfin/kernel/ipipe.c
index 2b979edee140..b8d22034b9a6 100644
--- a/arch/blackfin/kernel/ipipe.c
+++ b/arch/blackfin/kernel/ipipe.c
@@ -342,8 +342,3 @@ void ___ipipe_sync_pipeline(unsigned long syncmask)
 }
 
 EXPORT_SYMBOL(show_stack);
-
-#ifdef CONFIG_IPIPE_TRACE_MCOUNT
-void notrace _mcount(void);
-EXPORT_SYMBOL(_mcount);
-#endif /* CONFIG_IPIPE_TRACE_MCOUNT */
diff --git a/arch/blackfin/kernel/mcount.S b/arch/blackfin/kernel/mcount.S
deleted file mode 100644
index edcfb3865f46..000000000000
--- a/arch/blackfin/kernel/mcount.S
+++ /dev/null
@@ -1,70 +0,0 @@
-/*
- * linux/arch/blackfin/mcount.S
- *
- * Copyright (C) 2006 Analog Devices Inc.
- *
- * 2007/04/12 Save index, length, modify and base registers. --rpm
- */
-
-#include <linux/linkage.h>
-#include <asm/blackfin.h>
-
-.text
-
-.align 4 	/* just in case */
-
-ENTRY(__mcount)
-	[--sp] = i0;
-	[--sp] = i1;
-	[--sp] = i2;
-	[--sp] = i3;
-	[--sp] = l0;
-	[--sp] = l1;
-	[--sp] = l2;
-	[--sp] = l3;
-	[--sp] = m0;
-	[--sp] = m1;
-	[--sp] = m2;
-	[--sp] = m3;
-	[--sp] = b0;
-	[--sp] = b1;
-	[--sp] = b2;
-	[--sp] = b3;
-	[--sp] = ( r7:0, p5:0 );
-	[--sp] = ASTAT;
-
-	p1.L = _ipipe_trace_enable;
-	p1.H = _ipipe_trace_enable;
-	r7 = [p1];
-	CC = r7 == 0;
-	if CC jump out;
-	link 0x10;
-	r0 = 0x0;
-	[sp + 0xc] = r0; /* v */
-	r0 = 0x0;	/* type: IPIPE_TRACE_FN */
-	r1 = rets;
-	p0 = [fp];	/* p0: Prior FP */
-	r2 = [p0 + 4];	/* r2: Prior RETS */
-	call ___ipipe_trace;
-	unlink;
-out:
-	ASTAT = [sp++];
-	( r7:0, p5:0 ) = [sp++];
-	b3 = [sp++];
-	b2 = [sp++];
-	b1 = [sp++];
-	b0 = [sp++];
-	m3 = [sp++];
-	m2 = [sp++];
-	m1 = [sp++];
-	m0 = [sp++];
-	l3 = [sp++];
-	l2 = [sp++];
-	l1 = [sp++];
-	l0 = [sp++];
-	i3 = [sp++];
-	i2 = [sp++];
-	i1 = [sp++];
-	i0 = [sp++];
-	rts;
-ENDPROC(__mcount)
-- 
cgit v1.2.3-59-g8ed1b


From 7c039a90f02c3fdcab8d3ca170c05ad37014189e Mon Sep 17 00:00:00 2001
From: Philippe Gerum <rpm@xenomai.org>
Date: Tue, 16 Jun 2009 05:25:54 +0200
Subject: Blackfin: update I-pipe patch level

Signed-off-by: Philippe Gerum <rpm@xenomai.org>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/include/asm/ipipe.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/blackfin/include/asm/ipipe.h b/arch/blackfin/include/asm/ipipe.h
index ffa5e60faa9f..87ba9ad399cb 100644
--- a/arch/blackfin/include/asm/ipipe.h
+++ b/arch/blackfin/include/asm/ipipe.h
@@ -35,9 +35,9 @@
 #include <asm/atomic.h>
 #include <asm/traps.h>
 
-#define IPIPE_ARCH_STRING     "1.10-00"
+#define IPIPE_ARCH_STRING     "1.11-00"
 #define IPIPE_MAJOR_NUMBER    1
-#define IPIPE_MINOR_NUMBER    10
+#define IPIPE_MINOR_NUMBER    11
 #define IPIPE_PATCH_NUMBER    0
 
 #ifdef CONFIG_SMP
-- 
cgit v1.2.3-59-g8ed1b


From 41ba653f24a39a0e6a4afe9b2763a95a57e042c2 Mon Sep 17 00:00:00 2001
From: Jie Zhang <jie.zhang@analog.com>
Date: Tue, 16 Jun 2009 09:48:33 +0000
Subject: Blackfin: decouple unrelated cache settings to get exact behavior

The current cache options don't really represent the hardware features.
They end up setting different aspects of the hardware so that the end
result is to turn on/off the cache.  Unfortunately, when we hit cache
problems with the hardware, it's difficult to test different settings to
root cause the problem.  The current settings also don't cleanly allow for
different caching behaviors with different regions of memory.

So split the configure options such that they properly reflect the settings
that are applied to the hardware.

Signed-off-by: Jie Zhang <jie.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/Kconfig                    | 57 +++++++++++++------
 arch/blackfin/include/asm/cache.h        |  4 +-
 arch/blackfin/include/asm/cacheflush.h   | 10 ++--
 arch/blackfin/include/asm/cplb.h         | 32 ++++++-----
 arch/blackfin/kernel/cplb-mpu/cplbinit.c | 10 ++--
 arch/blackfin/kernel/cplb-mpu/cplbmgr.c  | 36 ++++++++----
 arch/blackfin/kernel/setup.c             | 96 ++++++++++++++++++++++++++------
 arch/blackfin/mach-common/arch_checks.c  |  4 +-
 arch/blackfin/mach-common/cpufreq.c      |  2 +-
 arch/blackfin/mach-common/pm.c           |  4 +-
 arch/blackfin/mm/init.c                  |  2 +-
 11 files changed, 176 insertions(+), 81 deletions(-)

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index 8140a2fc8bbd..220635a1ebdd 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -907,23 +907,41 @@ endchoice
 
 
 comment "Cache Support"
+
 config BFIN_ICACHE
 	bool "Enable ICACHE"
+	default y
+config BFIN_ICACHE_LOCK
+	bool "Enable Instruction Cache Locking"
+	depends on BFIN_ICACHE
+	default n
+config BFIN_EXTMEM_ICACHEABLE
+	bool "Enable ICACHE for external memory"
+	depends on BFIN_ICACHE
+	default y
+config BFIN_L2_ICACHEABLE
+	bool "Enable ICACHE for L2 SRAM"
+	depends on BFIN_ICACHE
+	depends on BF54x || BF561
+	default n
+
 config BFIN_DCACHE
 	bool "Enable DCACHE"
+	default y
 config BFIN_DCACHE_BANKA
 	bool "Enable only 16k BankA DCACHE - BankB is SRAM"
 	depends on BFIN_DCACHE && !BF531
 	default n
-config BFIN_ICACHE_LOCK
-	bool "Enable Instruction Cache Locking"
-
-choice
-	prompt "External memory cache policy"
+config BFIN_EXTMEM_DCACHEABLE
+	bool "Enable DCACHE for external memory"
 	depends on BFIN_DCACHE
-	default BFIN_WB if !SMP
-	default BFIN_WT if SMP
-config BFIN_WB
+	default y
+choice
+	prompt "External memory DCACHE policy"
+	depends on BFIN_EXTMEM_DCACHEABLE
+	default BFIN_EXTMEM_WRITEBACK if !SMP
+	default BFIN_EXTMEM_WRITETHROUGH if SMP
+config BFIN_EXTMEM_WRITEBACK
 	bool "Write back"
 	depends on !SMP
 	help
@@ -941,7 +959,7 @@ config BFIN_WB
 	  If you are unsure of the options and you want to be safe,
 	  then go with Write Through.
 
-config BFIN_WT
+config BFIN_EXTMEM_WRITETHROUGH
 	bool "Write through"
 	help
 	  Write Back Policy:
@@ -960,23 +978,26 @@ config BFIN_WT
 
 endchoice
 
+config BFIN_L2_DCACHEABLE
+	bool "Enable DCACHE for L2 SRAM"
+	depends on BFIN_DCACHE
+	depends on BF54x || BF561
+	default n
 choice
-	prompt "L2 SRAM cache policy"
-	depends on (BF54x || BF561)
-	default BFIN_L2_WT
-config BFIN_L2_WB
+	prompt "L2 SRAM DCACHE policy"
+	depends on BFIN_L2_DCACHEABLE
+	default BFIN_L2_WRITEBACK
+config BFIN_L2_WRITEBACK
 	bool "Write back"
 	depends on !SMP
 
-config BFIN_L2_WT
+config BFIN_L2_WRITETHROUGH
 	bool "Write through"
 	depends on !SMP
-
-config BFIN_L2_NOT_CACHED
-	bool "Not cached"
-
 endchoice
 
+
+comment "Memory Protection Unit"
 config MPU
 	bool "Enable the memory protection unit (EXPERIMENTAL)"
 	default n
diff --git a/arch/blackfin/include/asm/cache.h b/arch/blackfin/include/asm/cache.h
index 2ef669ed9222..477050ad5c53 100644
--- a/arch/blackfin/include/asm/cache.h
+++ b/arch/blackfin/include/asm/cache.h
@@ -35,10 +35,10 @@
 
 #if defined(CONFIG_SMP) && \
     !defined(CONFIG_BFIN_CACHE_COHERENT)
-# if defined(CONFIG_BFIN_ICACHE)
+# if defined(CONFIG_BFIN_ICACHEABLE) || defined(CONFIG_BFIN_L2_ICACHEABLE)
 # define __ARCH_SYNC_CORE_ICACHE
 # endif
-# if defined(CONFIG_BFIN_DCACHE)
+# if defined(CONFIG_BFIN_DCACHEABLE) || defined(CONFIG_BFIN_L2_DCACHEABLE)
 # define __ARCH_SYNC_CORE_DCACHE
 # endif
 #ifndef __ASSEMBLY__
diff --git a/arch/blackfin/include/asm/cacheflush.h b/arch/blackfin/include/asm/cacheflush.h
index 5c17dee53b5d..7e55549e180f 100644
--- a/arch/blackfin/include/asm/cacheflush.h
+++ b/arch/blackfin/include/asm/cacheflush.h
@@ -56,7 +56,7 @@ extern void blackfin_invalidate_entire_icache(void);
 
 static inline void flush_icache_range(unsigned start, unsigned end)
 {
-#if defined(CONFIG_BFIN_WB)
+#if defined(CONFIG_BFIN_EXTMEM_WRITEBACK) || defined(CONFIG_BFIN_L2_WRITEBACK)
 	blackfin_dcache_flush_range(start, end);
 #endif
 
@@ -87,9 +87,9 @@ do { memcpy(dst, src, len);						\
 #else
 # define invalidate_dcache_range(start,end)	do { } while (0)
 #endif
-#if defined(CONFIG_BFIN_DCACHE) && defined(CONFIG_BFIN_WB)
+#if defined(CONFIG_BFIN_EXTMEM_WRITEBACK) || defined(CONFIG_BFIN_L2_WRITEBACK)
 # define flush_dcache_range(start,end)		blackfin_dcache_flush_range((start), (end))
-# define flush_dcache_page(page)			blackfin_dflush_page(page_address(page))
+# define flush_dcache_page(page)		blackfin_dflush_page(page_address(page))
 #else
 # define flush_dcache_range(start,end)		do { } while (0)
 # define flush_dcache_page(page)		do { } while (0)
@@ -100,7 +100,7 @@ extern unsigned long reserved_mem_icache_on;
 
 static inline int bfin_addr_dcacheable(unsigned long addr)
 {
-#ifdef CONFIG_BFIN_DCACHE
+#ifdef CONFIG_BFIN_EXTMEM_DCACHEABLE
 	if (addr < (_ramend - DMA_UNCACHED_REGION))
 		return 1;
 #endif
@@ -109,7 +109,7 @@ static inline int bfin_addr_dcacheable(unsigned long addr)
 		addr >= _ramend && addr < physical_mem_end)
 		return 1;
 
-#ifndef CONFIG_BFIN_L2_NOT_CACHED
+#ifdef CONFIG_BFIN_L2_DCACHEABLE
 	if (addr >= L2_START && addr < L2_START + L2_LENGTH)
 		return 1;
 #endif
diff --git a/arch/blackfin/include/asm/cplb.h b/arch/blackfin/include/asm/cplb.h
index a75a6a9f0949..c5dacf8f8cf9 100644
--- a/arch/blackfin/include/asm/cplb.h
+++ b/arch/blackfin/include/asm/cplb.h
@@ -37,8 +37,6 @@
 #define L1_IMEMORY        (               CPLB_USER_RD | CPLB_VALID | CPLB_LOCK)
 #define SDRAM_INON_CHBL   (               CPLB_USER_RD | CPLB_VALID)
 
-/*Use the menuconfig cache policy here - CONFIG_BFIN_WT/CONFIG_BFIN_WB*/
-
 #if ANOMALY_05000158
 #define ANOMALY_05000158_WORKAROUND             0x200
 #else
@@ -47,10 +45,12 @@
 
 #define CPLB_COMMON	(CPLB_DIRTY | CPLB_SUPV_WR | CPLB_USER_WR | CPLB_USER_RD | CPLB_VALID | ANOMALY_05000158_WORKAROUND)
 
-#ifdef CONFIG_BFIN_WB         /*Write Back Policy */
+#ifdef CONFIG_BFIN_EXTMEM_WRITEBACK
 #define SDRAM_DGENERIC   (CPLB_L1_CHBL | CPLB_COMMON)
-#else                           /*Write Through */
+#elif defined(CONFIG_BFIN_EXTMEM_WRITETHROUGH)
 #define SDRAM_DGENERIC   (CPLB_L1_CHBL | CPLB_WT | CPLB_L1_AOW  | CPLB_COMMON)
+#else
+#define SDRAM_DGENERIC   (CPLB_COMMON)
 #endif
 
 #define SDRAM_DNON_CHBL  (CPLB_COMMON)
@@ -61,21 +61,23 @@
 
 #ifdef CONFIG_SMP
 #define L2_ATTR          (INITIAL_T | I_CPLB | D_CPLB)
-#define L2_IMEMORY       (CPLB_COMMON)
-#define L2_DMEMORY       (CPLB_LOCK | CPLB_COMMON)
+#define L2_IMEMORY       (CPLB_COMMON | PAGE_SIZE_1MB)
+#define L2_DMEMORY       (CPLB_LOCK | CPLB_COMMON | PAGE_SIZE_1MB)
 
 #else
 #define L2_ATTR          (INITIAL_T | SWITCH_T | I_CPLB | D_CPLB)
-#define L2_IMEMORY       (SDRAM_IGENERIC)
-
-# if defined(CONFIG_BFIN_L2_WB)
-# define L2_DMEMORY      (CPLB_L1_CHBL | CPLB_COMMON)
-# elif defined(CONFIG_BFIN_L2_WT)
-# define L2_DMEMORY      (CPLB_L1_CHBL | CPLB_WT | CPLB_L1_AOW  | CPLB_COMMON)
-# elif defined(CONFIG_BFIN_L2_NOT_CACHED)
-# define L2_DMEMORY      (CPLB_COMMON)
+# if defined(CONFIG_BFIN_L2_ICACHEABLE)
+# define L2_IMEMORY      (CPLB_L1_CHBL | CPLB_USER_RD | CPLB_VALID | PAGE_SIZE_1MB)
+# else
+# define L2_IMEMORY      (               CPLB_USER_RD | CPLB_VALID | PAGE_SIZE_1MB)
+# endif
+
+# if defined(CONFIG_BFIN_L2_WRITEBACK)
+# define L2_DMEMORY      (CPLB_L1_CHBL | CPLB_COMMON | PAGE_SIZE_1MB)
+# elif defined(CONFIG_BFIN_L2_WRITETHROUGH)
+# define L2_DMEMORY      (CPLB_L1_CHBL | CPLB_WT | CPLB_L1_AOW | CPLB_COMMON | PAGE_SIZE_1MB)
 # else
-# define L2_DMEMORY      (0)
+# define L2_DMEMORY      (CPLB_COMMON | PAGE_SIZE_1MB)
 # endif
 #endif /* CONFIG_SMP */
 
diff --git a/arch/blackfin/kernel/cplb-mpu/cplbinit.c b/arch/blackfin/kernel/cplb-mpu/cplbinit.c
index c006a44527bf..36193eed9a1f 100644
--- a/arch/blackfin/kernel/cplb-mpu/cplbinit.c
+++ b/arch/blackfin/kernel/cplb-mpu/cplbinit.c
@@ -46,13 +46,13 @@ void __init generate_cplb_tables_cpu(unsigned int cpu)
 
 	printk(KERN_INFO "MPU: setting up cplb tables with memory protection\n");
 
-#ifdef CONFIG_BFIN_ICACHE
+#ifdef CONFIG_BFIN_EXTMEM_ICACHEABLE
 	i_cache = CPLB_L1_CHBL | ANOMALY_05000158_WORKAROUND;
 #endif
 
-#ifdef CONFIG_BFIN_DCACHE
+#ifdef CONFIG_BFIN_EXTMEM_DCACHEABLE
 	d_cache = CPLB_L1_CHBL;
-#ifdef CONFIG_BFIN_WT
+#ifdef CONFIG_BFIN_EXTMEM_WRITETROUGH
 	d_cache |= CPLB_L1_AOW | CPLB_WT;
 #endif
 #endif
@@ -91,9 +91,9 @@ void __init generate_cplb_tables_cpu(unsigned int cpu)
 	/* Cover L2 memory */
 #if L2_LENGTH > 0
 	dcplb_tbl[cpu][i_d].addr = L2_START;
-	dcplb_tbl[cpu][i_d++].data = L2_DMEMORY | PAGE_SIZE_1MB;
+	dcplb_tbl[cpu][i_d++].data = L2_DMEMORY;
 	icplb_tbl[cpu][i_i].addr = L2_START;
-	icplb_tbl[cpu][i_i++].data = L2_IMEMORY | PAGE_SIZE_1MB;
+	icplb_tbl[cpu][i_i++].data = L2_IMEMORY;
 #endif
 
 	first_mask_dcplb = i_d;
diff --git a/arch/blackfin/kernel/cplb-mpu/cplbmgr.c b/arch/blackfin/kernel/cplb-mpu/cplbmgr.c
index 784923e52a9a..bcdfe9b0b71f 100644
--- a/arch/blackfin/kernel/cplb-mpu/cplbmgr.c
+++ b/arch/blackfin/kernel/cplb-mpu/cplbmgr.c
@@ -150,15 +150,19 @@ static noinline int dcplb_miss(unsigned int cpu)
 	nr_dcplb_miss[cpu]++;
 
 	d_data = CPLB_SUPV_WR | CPLB_VALID | CPLB_DIRTY | PAGE_SIZE_4KB;
-#ifdef CONFIG_BFIN_DCACHE
+#ifdef CONFIG_BFIN_EXTMEM_DCACHEABLE
 	if (bfin_addr_dcacheable(addr)) {
 		d_data |= CPLB_L1_CHBL | ANOMALY_05000158_WORKAROUND;
-#ifdef CONFIG_BFIN_WT
+# ifdef CONFIG_BFIN_EXTMEM_WRITETHROUGH
 		d_data |= CPLB_L1_AOW | CPLB_WT;
-#endif
+# endif
 	}
 #endif
-	if (addr >= physical_mem_end) {
+
+	if (L2_LENGTH && addr >= L2_START && addr < L2_START + L2_LENGTH) {
+		addr = L2_START;
+		d_data = L2_DMEMORY;
+	} else if (addr >= physical_mem_end) {
 		if (addr >= ASYNC_BANK0_BASE && addr < ASYNC_BANK3_BASE + ASYNC_BANK3_SIZE
 		    && (status & FAULT_USERSUPV)) {
 			addr &= ~0x3fffff;
@@ -235,7 +239,7 @@ static noinline int icplb_miss(unsigned int cpu)
 
 	i_data = CPLB_VALID | CPLB_PORTPRIO | PAGE_SIZE_4KB;
 
-#ifdef CONFIG_BFIN_ICACHE
+#ifdef CONFIG_BFIN_EXTMEM_ICACHEABLE
 	/*
 	 * Normal RAM, and possibly the reserved memory area, are
 	 * cacheable.
@@ -245,7 +249,10 @@ static noinline int icplb_miss(unsigned int cpu)
 		i_data |= CPLB_L1_CHBL | ANOMALY_05000158_WORKAROUND;
 #endif
 
-	if (addr >= physical_mem_end) {
+	if (L2_LENGTH && addr >= L2_START && addr < L2_START + L2_LENGTH) {
+		addr = L2_START;
+		i_data = L2_IMEMORY;
+	} else if (addr >= physical_mem_end) {
 		if (addr >= BOOT_ROM_START && addr < BOOT_ROM_START + BOOT_ROM_LENGTH
 		    && (status & FAULT_USERSUPV)) {
 			addr &= ~(1 * 1024 * 1024 - 1);
@@ -365,13 +372,18 @@ void set_mask_dcplbs(unsigned long *masks, unsigned int cpu)
 	local_irq_save_hw(flags);
 	current_rwx_mask[cpu] = masks;
 
-	d_data = CPLB_SUPV_WR | CPLB_VALID | CPLB_DIRTY | PAGE_SIZE_4KB;
-#ifdef CONFIG_BFIN_DCACHE
-	d_data |= CPLB_L1_CHBL;
-#ifdef CONFIG_BFIN_WT
-	d_data |= CPLB_L1_AOW | CPLB_WT;
-#endif
+	if (L2_LENGTH && addr >= L2_START && addr < L2_START + L2_LENGTH) {
+		addr = L2_START;
+		d_data = L2_DMEMORY;
+	} else {
+		d_data = CPLB_SUPV_WR | CPLB_VALID | CPLB_DIRTY | PAGE_SIZE_4KB;
+#ifdef CONFIG_BFIN_EXTMEM_DCACHEABLE
+		d_data |= CPLB_L1_CHBL;
+# ifdef CONFIG_BFIN_EXTMEM_WRITETHROUGH
+		d_data |= CPLB_L1_AOW | CPLB_WT;
+# endif
 #endif
+	}
 
 	disable_dcplb();
 	for (i = first_mask_dcplb; i < first_switched_dcplb; i++) {
diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c
index b2782eae31e1..8d7892820130 100644
--- a/arch/blackfin/kernel/setup.c
+++ b/arch/blackfin/kernel/setup.c
@@ -117,15 +117,49 @@ void __cpuinit bfin_setup_caches(unsigned int cpu)
 	 */
 #ifdef CONFIG_BFIN_ICACHE
 	printk(KERN_INFO "Instruction Cache Enabled for CPU%u\n", cpu);
+	printk(KERN_INFO "  External memory:"
+# ifdef CONFIG_BFIN_EXTMEM_ICACHEABLE
+	       " cacheable"
+# else
+	       " uncacheable"
+# endif
+	       " in instruction cache\n");
+	if (L2_LENGTH)
+		printk(KERN_INFO "  L2 SRAM        :"
+# ifdef CONFIG_BFIN_L2_ICACHEABLE
+		       " cacheable"
+# else
+		       " uncacheable"
+# endif
+		       " in instruction cache\n");
+
+#else
+	printk(KERN_INFO "Instruction Cache Disabled for CPU%u\n", cpu);
 #endif
+
 #ifdef CONFIG_BFIN_DCACHE
-	printk(KERN_INFO "Data Cache Enabled for CPU%u"
-# if defined CONFIG_BFIN_WB
-		" (write-back)"
-# elif defined CONFIG_BFIN_WT
-		" (write-through)"
+	printk(KERN_INFO "Data Cache Enabled for CPU%u\n", cpu);
+	printk(KERN_INFO "  External memory:"
+# if defined CONFIG_BFIN_EXTMEM_WRITEBACK
+	       " cacheable (write-back)"
+# elif defined CONFIG_BFIN_EXTMEM_WRITETHROUGH
+	       " cacheable (write-through)"
+# else
+	       " uncacheable"
+# endif
+	       " in data cache\n");
+	if (L2_LENGTH)
+		printk(KERN_INFO "  L2 SRAM        :"
+# if defined CONFIG_BFIN_L2_WRITEBACK
+		       " cacheable (write-back)"
+# elif defined CONFIG_BFIN_L2_WRITETHROUGH
+		       " cacheable (write-through)"
+# else
+		       " uncacheable"
 # endif
-		"\n", cpu);
+		       " in data cache\n");
+#else
+	printk(KERN_INFO "Data Cache Disabled for CPU%u\n", cpu);
 #endif
 }
 
@@ -516,7 +550,7 @@ static __init void memory_setup(void)
 	    && ((unsigned long *)mtd_phys)[1] == ROMSB_WORD1)
 		mtd_size =
 		    PAGE_ALIGN(be32_to_cpu(((unsigned long *)mtd_phys)[2]));
-#  if (defined(CONFIG_BFIN_ICACHE) && ANOMALY_05000263)
+#  if (defined(CONFIG_BFIN_EXTMEM_ICACHEABLE) && ANOMALY_05000263)
 	/* Due to a Hardware Anomaly we need to limit the size of usable
 	 * instruction memory to max 60MB, 56 if HUNT_FOR_ZERO is on
 	 * 05000263 - Hardware loop corrupted when taking an ICPLB exception
@@ -544,7 +578,7 @@ static __init void memory_setup(void)
 	dma_memcpy((void *)uclinux_ram_map.phys, _end, uclinux_ram_map.size);
 #endif				/* CONFIG_MTD_UCLINUX */
 
-#if (defined(CONFIG_BFIN_ICACHE) && ANOMALY_05000263)
+#if (defined(CONFIG_BFIN_EXTMEM_ICACHEABLE) && ANOMALY_05000263)
 	/* Due to a Hardware Anomaly we need to limit the size of usable
 	 * instruction memory to max 60MB, 56 if HUNT_FOR_ZERO is on
 	 * 05000263 - Hardware loop corrupted when taking an ICPLB exception
@@ -1158,16 +1192,25 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 		icache_size = 0;
 
 	seq_printf(m, "cache size\t: %d KB(L1 icache) "
-		"%d KB(L1 dcache%s) %d KB(L2 cache)\n",
-		icache_size, dcache_size,
-#if defined CONFIG_BFIN_WB
-		"-wb"
-#elif defined CONFIG_BFIN_WT
-		"-wt"
-#endif
-		"", 0);
-
+		"%d KB(L1 dcache) %d KB(L2 cache)\n",
+		icache_size, dcache_size, 0);
 	seq_printf(m, "%s\n", cache);
+	seq_printf(m, "external memory\t: "
+#if defined(CONFIG_BFIN_EXTMEM_ICACHEABLE)
+		   "cacheable"
+#else
+		   "uncacheable"
+#endif
+		   " in instruction cache\n");
+	seq_printf(m, "external memory\t: "
+#if defined(CONFIG_BFIN_EXTMEM_WRITEBACK)
+		      "cacheable (write-back)"
+#elif defined(CONFIG_BFIN_EXTMEM_WRITETHROUGH)
+		      "cacheable (write-through)"
+#else
+		      "uncacheable"
+#endif
+		      " in data cache\n");
 
 	if (icache_size)
 		seq_printf(m, "icache setup\t: %d Sub-banks/%d Ways, %d Lines/Way\n",
@@ -1240,8 +1283,25 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 	if (cpu_num != num_possible_cpus() - 1)
 		return 0;
 
-	if (L2_LENGTH)
+	if (L2_LENGTH) {
 		seq_printf(m, "L2 SRAM\t\t: %dKB\n", L2_LENGTH/0x400);
+		seq_printf(m, "L2 SRAM\t\t: "
+#if defined(CONFIG_BFIN_L2_ICACHEABLE)
+			      "cacheable"
+#else
+			      "uncacheable"
+#endif
+			      " in instruction cache\n");
+		seq_printf(m, "L2 SRAM\t\t: "
+#if defined(CONFIG_BFIN_L2_WRITEBACK)
+			      "cacheable (write-back)"
+#elif defined(CONFIG_BFIN_L2_WRITETHROUGH)
+			      "cacheable (write-through)"
+#else
+			      "uncacheable"
+#endif
+			      " in data cache\n");
+	}
 	seq_printf(m, "board name\t: %s\n", bfin_board_name);
 	seq_printf(m, "board memory\t: %ld kB (0x%p -> 0x%p)\n",
 		 physical_mem_end >> 10, (void *)0, (void *)physical_mem_end);
diff --git a/arch/blackfin/mach-common/arch_checks.c b/arch/blackfin/mach-common/arch_checks.c
index da93d9207165..5998d8632a73 100644
--- a/arch/blackfin/mach-common/arch_checks.c
+++ b/arch/blackfin/mach-common/arch_checks.c
@@ -74,7 +74,7 @@
 
 /* if 220 exists, can not set External Memory WB and L2 not_cached, either External Memory not_cached and L2 WB */
 #if ANOMALY_05000220 && \
-	((defined(CONFIG_BFIN_WB) && defined(CONFIG_BFIN_L2_NOT_CACHED)) || \
-	 (!defined(CONFIG_BFIN_DCACHE) && defined(CONFIG_BFIN_L2_WB)))
+	((defined(CONFIG_BFIN_EXTMEM_WRITEBACK) && !defined(CONFIG_BFIN_L2_DCACHEABLE)) || \
+	 (!defined(CONFIG_BFIN_EXTMEM_DCACHEABLE) && defined(CONFIG_BFIN_L2_WRITEBACK)))
 # error You are exposing Anomaly 220 in this config, either config L2 as Write Through, or make External Memory WB.
 #endif
diff --git a/arch/blackfin/mach-common/cpufreq.c b/arch/blackfin/mach-common/cpufreq.c
index 70e3411f558c..85c658083279 100644
--- a/arch/blackfin/mach-common/cpufreq.c
+++ b/arch/blackfin/mach-common/cpufreq.c
@@ -141,7 +141,7 @@ static int __init __bfin_cpu_init(struct cpufreq_policy *policy)
 	sclk = get_sclk() / 1000;
 
 #if ANOMALY_05000273 || ANOMALY_05000274 || \
-	(!defined(CONFIG_BF54x) && defined(CONFIG_BFIN_DCACHE))
+	(!defined(CONFIG_BF54x) && defined(CONFIG_BFIN_EXTMEM_DCACHEABLE))
 	min_cclk = sclk * 2;
 #else
 	min_cclk = sclk;
diff --git a/arch/blackfin/mach-common/pm.c b/arch/blackfin/mach-common/pm.c
index bce5a84be49f..9e7e27b7fc8d 100644
--- a/arch/blackfin/mach-common/pm.c
+++ b/arch/blackfin/mach-common/pm.c
@@ -132,7 +132,7 @@ int bf53x_resume_l1_mem(unsigned char *memptr)
 	return 0;
 }
 
-#ifdef CONFIG_BFIN_WB
+#if defined(CONFIG_BFIN_EXTMEM_WRITEBACK) || defined(CONFIG_BFIN_L2_WRITEBACK)
 static void flushinv_all_dcache(void)
 {
 	u32 way, bank, subbank, set;
@@ -175,7 +175,7 @@ static inline void dcache_disable(void)
 #ifdef CONFIG_BFIN_DCACHE
 	unsigned long ctrl;
 
-#ifdef CONFIG_BFIN_WB
+#if defined(CONFIG_BFIN_EXTMEM_WRITEBACK) || defined(CONFIG_BFIN_L2_WRITEBACK)
 	flushinv_all_dcache();
 #endif
 	SSYNC();
diff --git a/arch/blackfin/mm/init.c b/arch/blackfin/mm/init.c
index 014a55abd09a..68bd0bd680cd 100644
--- a/arch/blackfin/mm/init.c
+++ b/arch/blackfin/mm/init.c
@@ -160,7 +160,7 @@ void __init mem_init(void)
 
 	/* do not count in kernel image between _rambase and _ramstart */
 	reservedpages -= (_ramstart - _rambase) >> PAGE_SHIFT;
-#if (defined(CONFIG_BFIN_ICACHE) && ANOMALY_05000263)
+#if (defined(CONFIG_BFIN_EXTMEM_ICACHEABLE) && ANOMALY_05000263)
 	reservedpages += (_ramend - memory_end - DMA_UNCACHED_REGION) >> PAGE_SHIFT;
 #endif
 
-- 
cgit v1.2.3-59-g8ed1b


From 841a534367c2cfdc325a11958c51406da17686c7 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Wed, 17 Jun 2009 07:11:42 +0000
Subject: Blackfin: update defconfigs

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/configs/BF518F-EZBRD_defconfig | 24 +++++++++++++++---------
 arch/blackfin/configs/BF526-EZBRD_defconfig  | 22 ++++++++++++----------
 arch/blackfin/configs/BF527-EZKIT_defconfig  | 22 ++++++++++++++--------
 arch/blackfin/configs/BF533-EZKIT_defconfig  | 25 +++++++++++++++++--------
 arch/blackfin/configs/BF533-STAMP_defconfig  | 22 ++++++++++++++--------
 arch/blackfin/configs/BF537-STAMP_defconfig  | 22 ++++++++++++++--------
 arch/blackfin/configs/BF538-EZKIT_defconfig  | 22 ++++++++++++++--------
 arch/blackfin/configs/BF548-EZKIT_defconfig  | 27 ++++++++++++++++-----------
 arch/blackfin/configs/BF561-EZKIT_defconfig  | 27 ++++++++++++++++-----------
 arch/blackfin/configs/BlackStamp_defconfig   | 12 +++++++++---
 arch/blackfin/configs/CM-BF527_defconfig     | 22 ++++++++++++++--------
 arch/blackfin/configs/CM-BF533_defconfig     | 23 ++++++++++++++---------
 arch/blackfin/configs/CM-BF537E_defconfig    | 22 ++++++++++++++--------
 arch/blackfin/configs/CM-BF537U_defconfig    | 23 ++++++++++++++---------
 arch/blackfin/configs/CM-BF548_defconfig     | 25 ++++++++++++++++---------
 arch/blackfin/configs/CM-BF561_defconfig     | 25 ++++++++++++++++---------
 arch/blackfin/configs/H8606_defconfig        | 14 ++++++++++----
 arch/blackfin/configs/PNAV-10_defconfig      | 20 +++++++++++++-------
 arch/blackfin/configs/SRV1_defconfig         | 14 ++++++++++----
 arch/blackfin/configs/TCM-BF537_defconfig    | 14 ++++++++++----
 20 files changed, 272 insertions(+), 155 deletions(-)

diff --git a/arch/blackfin/configs/BF518F-EZBRD_defconfig b/arch/blackfin/configs/BF518F-EZBRD_defconfig
index baec1337f282..dcfb4889559a 100644
--- a/arch/blackfin/configs/BF518F-EZBRD_defconfig
+++ b/arch/blackfin/configs/BF518F-EZBRD_defconfig
@@ -326,11 +326,17 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -413,11 +419,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -916,7 +922,7 @@ CONFIG_MMC_BLOCK_BOUNCE=y
 # CONFIG_MMC_SDHCI is not set
 CONFIG_SDH_BFIN=m
 CONFIG_SDH_BFIN_MISSING_CMD_PULLUP_WORKAROUND=y
-CONFIG_SDH_BFIN_ENABLE_SDIO_IRQ=y
+# CONFIG_SDH_BFIN_ENABLE_SDIO_IRQ is not set
 # CONFIG_MMC_SPI is not set
 # CONFIG_MEMSTICK is not set
 # CONFIG_NEW_LEDS is not set
@@ -1147,7 +1153,7 @@ CONFIG_SCHED_DEBUG=y
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
 # CONFIG_DEBUG_KOBJECT is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 CONFIG_DEBUG_INFO=y
 # CONFIG_DEBUG_VM is not set
 # CONFIG_DEBUG_WRITECOUNT is not set
diff --git a/arch/blackfin/configs/BF526-EZBRD_defconfig b/arch/blackfin/configs/BF526-EZBRD_defconfig
index c06262e41f7c..48a3a7a9099c 100644
--- a/arch/blackfin/configs/BF526-EZBRD_defconfig
+++ b/arch/blackfin/configs/BF526-EZBRD_defconfig
@@ -331,16 +331,18 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
-# CONFIG_MPU is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
 
 #
-# Asynchonous Memory Configuration
+# Memory Protection Unit
 #
+# CONFIG_MPU is not set
 
 #
 # EBIU_AMGCTL Global Control
@@ -418,11 +420,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -1424,7 +1426,7 @@ CONFIG_SCHED_DEBUG=y
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
 # CONFIG_DEBUG_KOBJECT is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 CONFIG_DEBUG_INFO=y
 # CONFIG_DEBUG_VM is not set
 # CONFIG_DEBUG_WRITECOUNT is not set
diff --git a/arch/blackfin/configs/BF527-EZKIT_defconfig b/arch/blackfin/configs/BF527-EZKIT_defconfig
index e9175c608aa7..dd8352791daf 100644
--- a/arch/blackfin/configs/BF527-EZKIT_defconfig
+++ b/arch/blackfin/configs/BF527-EZKIT_defconfig
@@ -331,11 +331,17 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -418,11 +424,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -1505,7 +1511,7 @@ CONFIG_SCHED_DEBUG=y
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
 # CONFIG_DEBUG_KOBJECT is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 CONFIG_DEBUG_INFO=y
 # CONFIG_DEBUG_VM is not set
 # CONFIG_DEBUG_WRITECOUNT is not set
diff --git a/arch/blackfin/configs/BF533-EZKIT_defconfig b/arch/blackfin/configs/BF533-EZKIT_defconfig
index 5aa63bafdd62..4c044805cb5c 100644
--- a/arch/blackfin/configs/BF533-EZKIT_defconfig
+++ b/arch/blackfin/configs/BF533-EZKIT_defconfig
@@ -289,15 +289,24 @@ CONFIG_BFIN_GPTIMERS=m
 CONFIG_DMA_UNCACHED_1M=y
 # CONFIG_DMA_UNCACHED_NONE is not set
 
+#
+# Cache Support
+#
 #
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -391,11 +400,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -1052,7 +1061,7 @@ CONFIG_SCHED_DEBUG=y
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
 # CONFIG_DEBUG_KOBJECT is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 CONFIG_DEBUG_INFO=y
 # CONFIG_DEBUG_VM is not set
 # CONFIG_DEBUG_WRITECOUNT is not set
diff --git a/arch/blackfin/configs/BF533-STAMP_defconfig b/arch/blackfin/configs/BF533-STAMP_defconfig
index fed25329e13c..c99bbcd09a68 100644
--- a/arch/blackfin/configs/BF533-STAMP_defconfig
+++ b/arch/blackfin/configs/BF533-STAMP_defconfig
@@ -293,11 +293,17 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -391,11 +397,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -1216,7 +1222,7 @@ CONFIG_SCHED_DEBUG=y
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
 # CONFIG_DEBUG_KOBJECT is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 CONFIG_DEBUG_INFO=y
 # CONFIG_DEBUG_VM is not set
 # CONFIG_DEBUG_WRITECOUNT is not set
diff --git a/arch/blackfin/configs/BF537-STAMP_defconfig b/arch/blackfin/configs/BF537-STAMP_defconfig
index f9ac20d55799..092ffda80e68 100644
--- a/arch/blackfin/configs/BF537-STAMP_defconfig
+++ b/arch/blackfin/configs/BF537-STAMP_defconfig
@@ -300,11 +300,17 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -399,11 +405,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -1269,7 +1275,7 @@ CONFIG_SCHED_DEBUG=y
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
 # CONFIG_DEBUG_KOBJECT is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 CONFIG_DEBUG_INFO=y
 # CONFIG_DEBUG_VM is not set
 # CONFIG_DEBUG_WRITECOUNT is not set
diff --git a/arch/blackfin/configs/BF538-EZKIT_defconfig b/arch/blackfin/configs/BF538-EZKIT_defconfig
index ee98e227b887..fa698a89f6fe 100644
--- a/arch/blackfin/configs/BF538-EZKIT_defconfig
+++ b/arch/blackfin/configs/BF538-EZKIT_defconfig
@@ -311,11 +311,17 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -398,11 +404,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -1203,7 +1209,7 @@ CONFIG_SCHED_DEBUG=y
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
 # CONFIG_DEBUG_KOBJECT is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 CONFIG_DEBUG_INFO=y
 # CONFIG_DEBUG_VM is not set
 # CONFIG_DEBUG_WRITECOUNT is not set
diff --git a/arch/blackfin/configs/BF548-EZKIT_defconfig b/arch/blackfin/configs/BF548-EZKIT_defconfig
index deeabef8ab80..b3d3cab81cfe 100644
--- a/arch/blackfin/configs/BF548-EZKIT_defconfig
+++ b/arch/blackfin/configs/BF548-EZKIT_defconfig
@@ -366,14 +366,19 @@ CONFIG_DMA_UNCACHED_2M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
-# CONFIG_BFIN_L2_WB is not set
-CONFIG_BFIN_L2_WT=y
-# CONFIG_BFIN_L2_NOT_CACHED is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+# CONFIG_BFIN_L2_ICACHEABLE is not set
+# CONFIG_BFIN_L2_DCACHEABLE is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -459,11 +464,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -1606,7 +1611,7 @@ CONFIG_SCHED_DEBUG=y
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
 # CONFIG_DEBUG_KOBJECT is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 CONFIG_DEBUG_INFO=y
 # CONFIG_DEBUG_VM is not set
 # CONFIG_DEBUG_WRITECOUNT is not set
diff --git a/arch/blackfin/configs/BF561-EZKIT_defconfig b/arch/blackfin/configs/BF561-EZKIT_defconfig
index dcfbe2e2931e..0313cd1d9824 100644
--- a/arch/blackfin/configs/BF561-EZKIT_defconfig
+++ b/arch/blackfin/configs/BF561-EZKIT_defconfig
@@ -331,14 +331,19 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
-# CONFIG_BFIN_L2_WB is not set
-CONFIG_BFIN_L2_WT=y
-# CONFIG_BFIN_L2_NOT_CACHED is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+# CONFIG_BFIN_L2_ICACHEABLE is not set
+# CONFIG_BFIN_L2_DCACHEABLE is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -425,11 +430,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -1044,7 +1049,7 @@ CONFIG_SCHED_DEBUG=y
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
 # CONFIG_DEBUG_KOBJECT is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 CONFIG_DEBUG_INFO=y
 # CONFIG_DEBUG_VM is not set
 # CONFIG_DEBUG_WRITECOUNT is not set
diff --git a/arch/blackfin/configs/BlackStamp_defconfig b/arch/blackfin/configs/BlackStamp_defconfig
index 174c578b8ec4..5d944ffd4ab0 100644
--- a/arch/blackfin/configs/BlackStamp_defconfig
+++ b/arch/blackfin/configs/BlackStamp_defconfig
@@ -285,11 +285,17 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
diff --git a/arch/blackfin/configs/CM-BF527_defconfig b/arch/blackfin/configs/CM-BF527_defconfig
index e17875e8abe8..648a31d01bf4 100644
--- a/arch/blackfin/configs/CM-BF527_defconfig
+++ b/arch/blackfin/configs/CM-BF527_defconfig
@@ -329,11 +329,17 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -417,11 +423,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -1246,7 +1252,7 @@ CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC_VALUE=0
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
 # CONFIG_DEBUG_KOBJECT is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 # CONFIG_DEBUG_INFO is not set
 # CONFIG_DEBUG_VM is not set
 # CONFIG_DEBUG_WRITECOUNT is not set
diff --git a/arch/blackfin/configs/CM-BF533_defconfig b/arch/blackfin/configs/CM-BF533_defconfig
index fafd95e84b28..ae665b93b875 100644
--- a/arch/blackfin/configs/CM-BF533_defconfig
+++ b/arch/blackfin/configs/CM-BF533_defconfig
@@ -262,12 +262,17 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
-CONFIG_L1_MAX_PIECE=16
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -353,10 +358,10 @@ CONFIG_IP_FIB_HASH=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
-CONFIG_INET_DIAG=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -873,7 +878,7 @@ CONFIG_ENABLE_MUST_CHECK=y
 CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 CONFIG_DEBUG_MMRS=y
 CONFIG_DEBUG_HUNT_FOR_ZERO=y
 CONFIG_DEBUG_BFIN_HWTRACE_ON=y
diff --git a/arch/blackfin/configs/CM-BF537E_defconfig b/arch/blackfin/configs/CM-BF537E_defconfig
index e73aa5af58b9..d74b6f4db35d 100644
--- a/arch/blackfin/configs/CM-BF537E_defconfig
+++ b/arch/blackfin/configs/CM-BF537E_defconfig
@@ -297,11 +297,17 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -383,11 +389,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -861,7 +867,7 @@ CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 CONFIG_DEBUG_SECTION_MISMATCH=y
 # CONFIG_DEBUG_KERNEL is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 # CONFIG_DEBUG_MEMORY_INIT is not set
 # CONFIG_RCU_CPU_STALL_DETECTOR is not set
 
diff --git a/arch/blackfin/configs/CM-BF537U_defconfig b/arch/blackfin/configs/CM-BF537U_defconfig
index 80211303f6b9..7fc8dfa1719f 100644
--- a/arch/blackfin/configs/CM-BF537U_defconfig
+++ b/arch/blackfin/configs/CM-BF537U_defconfig
@@ -270,12 +270,17 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
-CONFIG_L1_MAX_PIECE=16
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -361,10 +366,10 @@ CONFIG_IP_FIB_HASH=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
-CONFIG_INET_DIAG=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -901,7 +906,7 @@ CONFIG_ENABLE_MUST_CHECK=y
 CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 CONFIG_DEBUG_MMRS=y
 CONFIG_DEBUG_HUNT_FOR_ZERO=y
 CONFIG_DEBUG_BFIN_HWTRACE_ON=y
diff --git a/arch/blackfin/configs/CM-BF548_defconfig b/arch/blackfin/configs/CM-BF548_defconfig
index dd815f0d1517..acca4e51a45a 100644
--- a/arch/blackfin/configs/CM-BF548_defconfig
+++ b/arch/blackfin/configs/CM-BF548_defconfig
@@ -333,12 +333,19 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
-CONFIG_L1_MAX_PIECE=16
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+# CONFIG_BFIN_L2_ICACHEABLE is not set
+# CONFIG_BFIN_L2_DCACHEABLE is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -428,11 +435,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -1334,7 +1341,7 @@ CONFIG_ENABLE_MUST_CHECK=y
 CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 # CONFIG_SAMPLES is not set
 CONFIG_DEBUG_MMRS=y
 CONFIG_DEBUG_HUNT_FOR_ZERO=y
diff --git a/arch/blackfin/configs/CM-BF561_defconfig b/arch/blackfin/configs/CM-BF561_defconfig
index 16c198bd40c5..bae4ee6e68bb 100644
--- a/arch/blackfin/configs/CM-BF561_defconfig
+++ b/arch/blackfin/configs/CM-BF561_defconfig
@@ -308,12 +308,19 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
-CONFIG_L1_MAX_PIECE=16
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+# CONFIG_BFIN_L2_ICACHEABLE is not set
+# CONFIG_BFIN_L2_DCACHEABLE is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -395,11 +402,11 @@ CONFIG_IP_FIB_HASH=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
@@ -837,7 +844,7 @@ CONFIG_ENABLE_MUST_CHECK=y
 CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 # CONFIG_SAMPLES is not set
 CONFIG_DEBUG_MMRS=y
 CONFIG_DEBUG_HUNT_FOR_ZERO=y
diff --git a/arch/blackfin/configs/H8606_defconfig b/arch/blackfin/configs/H8606_defconfig
index 6b4c1a982383..a6a7c8ede705 100644
--- a/arch/blackfin/configs/H8606_defconfig
+++ b/arch/blackfin/configs/H8606_defconfig
@@ -258,12 +258,18 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-CONFIG_BFIN_ICACHE_LOCK=y
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
-CONFIG_L1_MAX_PIECE=16
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
+# CONFIG_MPU is not set
 
 #
 # Asynchonous Memory Configuration
diff --git a/arch/blackfin/configs/PNAV-10_defconfig b/arch/blackfin/configs/PNAV-10_defconfig
index 09701f907e9b..ff377fae061b 100644
--- a/arch/blackfin/configs/PNAV-10_defconfig
+++ b/arch/blackfin/configs/PNAV-10_defconfig
@@ -295,11 +295,17 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -382,11 +388,11 @@ CONFIG_IP_PNP=y
 # CONFIG_INET_IPCOMP is not set
 # CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=y
-CONFIG_INET_XFRM_MODE_TUNNEL=y
-CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_INET_LRO is not set
-CONFIG_INET_DIAG=y
+# CONFIG_INET_DIAG is not set
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
 CONFIG_TCP_CONG_CUBIC=y
diff --git a/arch/blackfin/configs/SRV1_defconfig b/arch/blackfin/configs/SRV1_defconfig
index ec84a53daae9..814f9cacf407 100644
--- a/arch/blackfin/configs/SRV1_defconfig
+++ b/arch/blackfin/configs/SRV1_defconfig
@@ -279,12 +279,18 @@ CONFIG_DMA_UNCACHED_2M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
-CONFIG_L1_MAX_PIECE=16
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
+# CONFIG_MPU is not set
 
 #
 # Asynchonous Memory Configuration
diff --git a/arch/blackfin/configs/TCM-BF537_defconfig b/arch/blackfin/configs/TCM-BF537_defconfig
index 6e2796240fdc..375e75a27abc 100644
--- a/arch/blackfin/configs/TCM-BF537_defconfig
+++ b/arch/blackfin/configs/TCM-BF537_defconfig
@@ -287,11 +287,17 @@ CONFIG_DMA_UNCACHED_1M=y
 # Cache Support
 #
 CONFIG_BFIN_ICACHE=y
+# CONFIG_BFIN_ICACHE_LOCK is not set
 CONFIG_BFIN_DCACHE=y
 # CONFIG_BFIN_DCACHE_BANKA is not set
-# CONFIG_BFIN_ICACHE_LOCK is not set
-CONFIG_BFIN_WB=y
-# CONFIG_BFIN_WT is not set
+CONFIG_BFIN_EXTMEM_ICACHEABLE=y
+CONFIG_BFIN_EXTMEM_DCACHEABLE=y
+CONFIG_BFIN_EXTMEM_WRITEBACK=y
+# CONFIG_BFIN_EXTMEM_WRITETHROUGH is not set
+
+#
+# Memory Protection Unit
+#
 # CONFIG_MPU is not set
 
 #
@@ -709,7 +715,7 @@ CONFIG_FRAME_WARN=1024
 CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
-# CONFIG_DEBUG_BUGVERBOSE is not set
+CONFIG_DEBUG_BUGVERBOSE=y
 # CONFIG_DEBUG_MEMORY_INIT is not set
 # CONFIG_RCU_CPU_STALL_DETECTOR is not set
 # CONFIG_SYSCTL_SYSCALL_CHECK is not set
-- 
cgit v1.2.3-59-g8ed1b


From 985895bd8d1e41079b41da32cdc57876a4a74126 Mon Sep 17 00:00:00 2001
From: Robin Getz <rgetz@blackfin.uclinux.org>
Date: Wed, 17 Jun 2009 08:12:54 -0400
Subject: Blackfin: stick the CPU name into boot image name

Rather than use "Linux" in the boot image name (as this is redundant --
the image type is already set to "linux"), use the CPU name.  This makes
it fairly obvious when a wrong image is accidentally booted.  Otherwise
there is no kernel output and you waste time scratching your head
wondering wtf just happened.

Signed-off-by: Robin Getz <rgetz@blackfin.uclinux.org>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/boot/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/blackfin/boot/Makefile b/arch/blackfin/boot/Makefile
index 3ab6f23561dd..fd9ccc5fea10 100644
--- a/arch/blackfin/boot/Makefile
+++ b/arch/blackfin/boot/Makefile
@@ -13,7 +13,7 @@ extra-y += vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 vmlinux.bin.lzma
 
 quiet_cmd_uimage = UIMAGE  $@
       cmd_uimage = $(CONFIG_SHELL) $(MKIMAGE) -A $(ARCH) -O linux -T kernel \
-                   -C $(2) -n 'Linux-$(KERNELRELEASE)' -a $(CONFIG_BOOT_LOAD) \
+                   -C $(2) -n '$(MACHINE)-$(KERNELRELEASE)' -a $(CONFIG_BOOT_LOAD) \
                    -e $(shell $(NM) vmlinux | awk '$$NF == "__start" {print $$1}') \
                    -d $< $@
 
-- 
cgit v1.2.3-59-g8ed1b


From fa48f84a8cc722ca48b32fa0c338b6c3b358717d Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Wed, 17 Jun 2009 11:25:06 -0400
Subject: Blackfin: unify memory map headers

Many aspects of the Blackfin memory map is exactly the same across all
variants.  Rather than copy and paste all of these duplicated values in
each header, unify all of these into the common Blackfin memory map header
file.  In the process, push down BF561 SMP specific stuff to the BF561
specific header to keep the noise down.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/include/asm/blackfin.h             |  1 +
 arch/blackfin/include/asm/mem_map.h              | 97 ++++++++++++------------
 arch/blackfin/mach-bf518/include/mach/blackfin.h |  1 -
 arch/blackfin/mach-bf518/include/mach/mem_map.h  | 56 +++-----------
 arch/blackfin/mach-bf527/include/mach/blackfin.h |  1 -
 arch/blackfin/mach-bf527/include/mach/mem_map.h  | 56 +++-----------
 arch/blackfin/mach-bf533/include/mach/blackfin.h |  1 -
 arch/blackfin/mach-bf533/include/mach/mem_map.h  | 56 +++-----------
 arch/blackfin/mach-bf537/include/mach/blackfin.h |  1 -
 arch/blackfin/mach-bf537/include/mach/mem_map.h  | 56 +++-----------
 arch/blackfin/mach-bf538/include/mach/blackfin.h |  1 -
 arch/blackfin/mach-bf538/include/mach/mem_map.h  | 57 +++-----------
 arch/blackfin/mach-bf548/include/mach/blackfin.h |  1 -
 arch/blackfin/mach-bf548/include/mach/mem_map.h  | 51 +++----------
 arch/blackfin/mach-bf561/include/mach/blackfin.h |  1 -
 arch/blackfin/mach-bf561/include/mach/mem_map.h  | 58 ++++++++++----
 16 files changed, 145 insertions(+), 350 deletions(-)

diff --git a/arch/blackfin/include/asm/blackfin.h b/arch/blackfin/include/asm/blackfin.h
index 8bb2cb139756..4d4439583396 100644
--- a/arch/blackfin/include/asm/blackfin.h
+++ b/arch/blackfin/include/asm/blackfin.h
@@ -86,6 +86,7 @@ static inline void CSYNC(void)
 
 #endif /* __ASSEMBLY__ */
 
+#include <asm/mem_map.h>
 #include <mach/blackfin.h>
 #include <asm/bfin-global.h>
 
diff --git a/arch/blackfin/include/asm/mem_map.h b/arch/blackfin/include/asm/mem_map.h
index e92b31051bb7..5e21627c9ba2 100644
--- a/arch/blackfin/include/asm/mem_map.h
+++ b/arch/blackfin/include/asm/mem_map.h
@@ -1,87 +1,84 @@
 /*
- * mem_map.h
- * Common header file for blackfin family of processors.
+ * Common Blackfin memory map
  *
+ * Copyright 2004-2009 Analog Devices Inc.
+ * Licensed under the GPL-2 or later.
  */
 
-#ifndef _MEM_MAP_H_
-#define _MEM_MAP_H_
+#ifndef __BFIN_MEM_MAP_H__
+#define __BFIN_MEM_MAP_H__
 
 #include <mach/mem_map.h>
 
-#ifndef __ASSEMBLY__
+/* Every Blackfin so far has MMRs like this */
+#ifndef COREMMR_BASE
+# define COREMMR_BASE 0xFFE00000
+#endif
+#ifndef SYSMMR_BASE
+# define SYSMMR_BASE  0xFFC00000
+#endif
 
-#ifdef CONFIG_SMP
-static inline ulong get_l1_scratch_start_cpu(int cpu)
-{
-	return (cpu) ? COREB_L1_SCRATCH_START : COREA_L1_SCRATCH_START;
-}
-static inline ulong get_l1_code_start_cpu(int cpu)
-{
-	return (cpu) ? COREB_L1_CODE_START : COREA_L1_CODE_START;
-}
-static inline ulong get_l1_data_a_start_cpu(int cpu)
-{
-	return (cpu) ? COREB_L1_DATA_A_START : COREA_L1_DATA_A_START;
-}
-static inline ulong get_l1_data_b_start_cpu(int cpu)
-{
-	return (cpu) ? COREB_L1_DATA_B_START : COREA_L1_DATA_B_START;
-}
+/* Every Blackfin so far has on-chip Scratch Pad SRAM like this */
+#ifndef L1_SCRATCH_START
+# define L1_SCRATCH_START  0xFFB00000
+# define L1_SCRATCH_LENGTH 0x1000
+#endif
 
-static inline ulong get_l1_scratch_start(void)
-{
-	return get_l1_scratch_start_cpu(blackfin_core_id());
-}
-static inline ulong get_l1_code_start(void)
-{
-	return get_l1_code_start_cpu(blackfin_core_id());
-}
-static inline ulong get_l1_data_a_start(void)
-{
-	return get_l1_data_a_start_cpu(blackfin_core_id());
-}
-static inline ulong get_l1_data_b_start(void)
-{
-	return get_l1_data_b_start_cpu(blackfin_core_id());
-}
+/* Most parts lack on-chip L2 SRAM */
+#ifndef L2_START
+# define L2_START  0
+# define L2_LENGTH 0
+#endif
+
+/* Most parts lack on-chip L1 ROM */
+#ifndef L1_ROM_START
+# define L1_ROM_START  0
+# define L1_ROM_LENGTH 0
+#endif
+
+/* Allow wonky SMP ports to override this */
+#ifndef GET_PDA_SAFE
+# define GET_PDA_SAFE(preg) \
+	preg.l = _cpu_pda; \
+	preg.h = _cpu_pda;
+# define GET_PDA(preg, dreg) GET_PDA_SAFE(preg)
 
-#else /* !CONFIG_SMP */
+# ifndef __ASSEMBLY__
 
-static inline ulong get_l1_scratch_start_cpu(int cpu)
+static inline unsigned long get_l1_scratch_start_cpu(int cpu)
 {
 	return L1_SCRATCH_START;
 }
-static inline ulong get_l1_code_start_cpu(int cpu)
+static inline unsigned long get_l1_code_start_cpu(int cpu)
 {
 	return L1_CODE_START;
 }
-static inline ulong get_l1_data_a_start_cpu(int cpu)
+static inline unsigned long get_l1_data_a_start_cpu(int cpu)
 {
 	return L1_DATA_A_START;
 }
-static inline ulong get_l1_data_b_start_cpu(int cpu)
+static inline unsigned long get_l1_data_b_start_cpu(int cpu)
 {
 	return L1_DATA_B_START;
 }
-static inline ulong get_l1_scratch_start(void)
+static inline unsigned long get_l1_scratch_start(void)
 {
 	return get_l1_scratch_start_cpu(0);
 }
-static inline ulong get_l1_code_start(void)
+static inline unsigned long get_l1_code_start(void)
 {
 	return  get_l1_code_start_cpu(0);
 }
-static inline ulong get_l1_data_a_start(void)
+static inline unsigned long get_l1_data_a_start(void)
 {
 	return get_l1_data_a_start_cpu(0);
 }
-static inline ulong get_l1_data_b_start(void)
+static inline unsigned long get_l1_data_b_start(void)
 {
 	return get_l1_data_b_start_cpu(0);
 }
 
-#endif /* CONFIG_SMP */
-#endif /* __ASSEMBLY__ */
+# endif /* __ASSEMBLY__ */
+#endif /* !GET_PDA_SAFE */
 
-#endif				/* _MEM_MAP_H_ */
+#endif
diff --git a/arch/blackfin/mach-bf518/include/mach/blackfin.h b/arch/blackfin/mach-bf518/include/mach/blackfin.h
index 267bb7c8bfb5..e8e14c2769ed 100644
--- a/arch/blackfin/mach-bf518/include/mach/blackfin.h
+++ b/arch/blackfin/mach-bf518/include/mach/blackfin.h
@@ -33,7 +33,6 @@
 #define _MACH_BLACKFIN_H_
 
 #include "bf518.h"
-#include "mem_map.h"
 #include "defBF512.h"
 #include "anomaly.h"
 
diff --git a/arch/blackfin/mach-bf518/include/mach/mem_map.h b/arch/blackfin/mach-bf518/include/mach/mem_map.h
index 62bcc781bfaa..3c6777cb3532 100644
--- a/arch/blackfin/mach-bf518/include/mach/mem_map.h
+++ b/arch/blackfin/mach-bf518/include/mach/mem_map.h
@@ -1,38 +1,16 @@
 /*
- * file:         include/asm-blackfin/mach-bf518/mem_map.h
- * based on:	include/asm-blackfin/mach-bf527/mem_map.h
- * author:	Bryan Wu <cooloney@kernel.org>
+ * BF51x memory map
  *
- * created:
- * description:
- *	Memory MAP Common header file for blackfin BF518/6/4/2 of processors.
- * rev:
- *
- * modified:
- *
- * bugs:         enter bugs at http://blackfin.uclinux.org/
- *
- * this program is free software; you can redistribute it and/or modify
- * it under the terms of the gnu general public license as published by
- * the free software foundation; either version 2, or (at your option)
- * any later version.
- *
- * this program is distributed in the hope that it will be useful,
- * but without any warranty; without even the implied warranty of
- * merchantability or fitness for a particular purpose.  see the
- * gnu general public license for more details.
- *
- * you should have received a copy of the gnu general public license
- * along with this program; see the file copying.
- * if not, write to the free software foundation,
- * 59 temple place - suite 330, boston, ma 02111-1307, usa.
+ * Copyright 2004-2009 Analog Devices Inc.
+ * Licensed under the GPL-2 or later.
  */
 
-#ifndef _MEM_MAP_518_H_
-#define _MEM_MAP_518_H_
+#ifndef __BFIN_MACH_MEM_MAP_H__
+#define __BFIN_MACH_MEM_MAP_H__
 
-#define COREMMR_BASE           0xFFE00000	/* Core MMRs */
-#define SYSMMR_BASE            0xFFC00000	/* System MMRs */
+#ifndef __BFIN_MEM_MAP_H__
+# error "do not include mach/mem_map.h directly -- use asm/mem_map.h"
+#endif
 
 /* Async Memory Banks */
 #define ASYNC_BANK3_BASE	0x20300000	/* Async Bank 3 */
@@ -89,20 +67,4 @@
 #define BFIN_DSUPBANKS		0
 #endif				/*CONFIG_BFIN_DCACHE */
 
-/* Level 2 Memory - none */
-
-#define L2_START	0
-#define L2_LENGTH	0
-
-/* Scratch Pad Memory */
-
-#define L1_SCRATCH_START	0xFFB00000
-#define L1_SCRATCH_LENGTH	0x1000
-
-#define GET_PDA_SAFE(preg)		\
-	preg.l = _cpu_pda;		\
-	preg.h = _cpu_pda;
-
-#define GET_PDA(preg, dreg)	GET_PDA_SAFE(preg)
-
-#endif				/* _MEM_MAP_518_H_ */
+#endif
diff --git a/arch/blackfin/mach-bf527/include/mach/blackfin.h b/arch/blackfin/mach-bf527/include/mach/blackfin.h
index 417abcd61f4d..03665a8e16be 100644
--- a/arch/blackfin/mach-bf527/include/mach/blackfin.h
+++ b/arch/blackfin/mach-bf527/include/mach/blackfin.h
@@ -33,7 +33,6 @@
 #define _MACH_BLACKFIN_H_
 
 #include "bf527.h"
-#include "mem_map.h"
 #include "defBF522.h"
 #include "anomaly.h"
 
diff --git a/arch/blackfin/mach-bf527/include/mach/mem_map.h b/arch/blackfin/mach-bf527/include/mach/mem_map.h
index 019e0017ad81..d96e894afd2c 100644
--- a/arch/blackfin/mach-bf527/include/mach/mem_map.h
+++ b/arch/blackfin/mach-bf527/include/mach/mem_map.h
@@ -1,38 +1,16 @@
 /*
- * file:         include/asm-blackfin/mach-bf527/mem_map.h
- * based on:	include/asm-blackfin/mach-bf537/mem_map.h
- * author:	Michael Hennerich (michael.hennerich@analog.com)
+ * BF52x memory map
  *
- * created:
- * description:
- *	Memory MAP Common header file for blackfin BF527/5/2 of processors.
- * rev:
- *
- * modified:
- *
- * bugs:         enter bugs at http://blackfin.uclinux.org/
- *
- * this program is free software; you can redistribute it and/or modify
- * it under the terms of the gnu general public license as published by
- * the free software foundation; either version 2, or (at your option)
- * any later version.
- *
- * this program is distributed in the hope that it will be useful,
- * but without any warranty; without even the implied warranty of
- * merchantability or fitness for a particular purpose.  see the
- * gnu general public license for more details.
- *
- * you should have received a copy of the gnu general public license
- * along with this program; see the file copying.
- * if not, write to the free software foundation,
- * 59 temple place - suite 330, boston, ma 02111-1307, usa.
+ * Copyright 2004-2009 Analog Devices Inc.
+ * Licensed under the GPL-2 or later.
  */
 
-#ifndef _MEM_MAP_527_H_
-#define _MEM_MAP_527_H_
+#ifndef __BFIN_MACH_MEM_MAP_H__
+#define __BFIN_MACH_MEM_MAP_H__
 
-#define COREMMR_BASE           0xFFE00000	/* Core MMRs */
-#define SYSMMR_BASE            0xFFC00000	/* System MMRs */
+#ifndef __BFIN_MEM_MAP_H__
+# error "do not include mach/mem_map.h directly -- use asm/mem_map.h"
+#endif
 
 /* Async Memory Banks */
 #define ASYNC_BANK3_BASE	0x20300000	/* Async Bank 3 */
@@ -89,20 +67,4 @@
 #define BFIN_DSUPBANKS	0
 #endif				/*CONFIG_BFIN_DCACHE */
 
-/* Level 2 Memory - none */
-
-#define L2_START	0
-#define L2_LENGTH	0
-
-/* Scratch Pad Memory */
-
-#define L1_SCRATCH_START	0xFFB00000
-#define L1_SCRATCH_LENGTH	0x1000
-
-#define GET_PDA_SAFE(preg)		\
-	preg.l = _cpu_pda;		\
-	preg.h = _cpu_pda;
-
-#define GET_PDA(preg, dreg)	GET_PDA_SAFE(preg)
-
-#endif				/* _MEM_MAP_527_H_ */
+#endif
diff --git a/arch/blackfin/mach-bf533/include/mach/blackfin.h b/arch/blackfin/mach-bf533/include/mach/blackfin.h
index 045184f81a29..39aa175f19f5 100644
--- a/arch/blackfin/mach-bf533/include/mach/blackfin.h
+++ b/arch/blackfin/mach-bf533/include/mach/blackfin.h
@@ -34,7 +34,6 @@
 #define BF533_FAMILY
 
 #include "bf533.h"
-#include "mem_map.h"
 #include "defBF532.h"
 #include "anomaly.h"
 
diff --git a/arch/blackfin/mach-bf533/include/mach/mem_map.h b/arch/blackfin/mach-bf533/include/mach/mem_map.h
index fc33b7cb9937..197af1a398ac 100644
--- a/arch/blackfin/mach-bf533/include/mach/mem_map.h
+++ b/arch/blackfin/mach-bf533/include/mach/mem_map.h
@@ -1,38 +1,16 @@
 /*
- * File:         include/asm-blackfin/mach-bf533/mem_map.h
- * Based on:
- * Author:
+ * BF533 memory map
  *
- * Created:
- * Description:
- *
- * Rev:
- *
- * Modified:
- *
- * Bugs:         Enter bugs at http://blackfin.uclinux.org/
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; see the file COPYING.
- * If not, write to the Free Software Foundation,
- * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ * Copyright 2004-2009 Analog Devices Inc.
+ * Licensed under the GPL-2 or later.
  */
 
-#ifndef _MEM_MAP_533_H_
-#define _MEM_MAP_533_H_
+#ifndef __BFIN_MACH_MEM_MAP_H__
+#define __BFIN_MACH_MEM_MAP_H__
 
-#define COREMMR_BASE           0xFFE00000	 /* Core MMRs */
-#define SYSMMR_BASE            0xFFC00000	 /* System MMRs */
+#ifndef __BFIN_MEM_MAP_H__
+# error "do not include mach/mem_map.h directly -- use asm/mem_map.h"
+#endif
 
 /* Async Memory Banks */
 #define ASYNC_BANK3_BASE	0x20300000	 /* Async Bank 3 */
@@ -158,20 +136,4 @@
 
 #endif
 
-/* Level 2 Memory - none */
-
-#define L2_START	0
-#define L2_LENGTH	0
-
-/* Scratch Pad Memory */
-
-#define L1_SCRATCH_START	0xFFB00000
-#define L1_SCRATCH_LENGTH	0x1000
-
-#define GET_PDA_SAFE(preg)		\
-	preg.l = _cpu_pda;		\
-	preg.h = _cpu_pda;
-
-#define GET_PDA(preg, dreg)	GET_PDA_SAFE(preg)
-
-#endif				/* _MEM_MAP_533_H_ */
+#endif
diff --git a/arch/blackfin/mach-bf537/include/mach/blackfin.h b/arch/blackfin/mach-bf537/include/mach/blackfin.h
index 7d6069c886f1..f5e5015ad831 100644
--- a/arch/blackfin/mach-bf537/include/mach/blackfin.h
+++ b/arch/blackfin/mach-bf537/include/mach/blackfin.h
@@ -35,7 +35,6 @@
 #define BF537_FAMILY
 
 #include "bf537.h"
-#include "mem_map.h"
 #include "defBF534.h"
 #include "anomaly.h"
 
diff --git a/arch/blackfin/mach-bf537/include/mach/mem_map.h b/arch/blackfin/mach-bf537/include/mach/mem_map.h
index f9010c4b4bf3..942f08de306b 100644
--- a/arch/blackfin/mach-bf537/include/mach/mem_map.h
+++ b/arch/blackfin/mach-bf537/include/mach/mem_map.h
@@ -1,38 +1,16 @@
 /*
- * file:         include/asm-blackfin/mach-bf537/mem_map.h
- * based on:
- * author:
+ * BF537 memory map
  *
- * created:
- * description:
- *	Memory MAP Common header file for blackfin BF537/6/4 of processors.
- * rev:
- *
- * modified:
- *
- * bugs:         enter bugs at http://blackfin.uclinux.org/
- *
- * this program is free software; you can redistribute it and/or modify
- * it under the terms of the gnu general public license as published by
- * the free software foundation; either version 2, or (at your option)
- * any later version.
- *
- * this program is distributed in the hope that it will be useful,
- * but without any warranty; without even the implied warranty of
- * merchantability or fitness for a particular purpose.  see the
- * gnu general public license for more details.
- *
- * you should have received a copy of the gnu general public license
- * along with this program; see the file copying.
- * if not, write to the free software foundation,
- * 59 temple place - suite 330, boston, ma 02111-1307, usa.
+ * Copyright 2004-2009 Analog Devices Inc.
+ * Licensed under the GPL-2 or later.
  */
 
-#ifndef _MEM_MAP_537_H_
-#define _MEM_MAP_537_H_
+#ifndef __BFIN_MACH_MEM_MAP_H__
+#define __BFIN_MACH_MEM_MAP_H__
 
-#define COREMMR_BASE           0xFFE00000	 /* Core MMRs */
-#define SYSMMR_BASE            0xFFC00000	 /* System MMRs */
+#ifndef __BFIN_MEM_MAP_H__
+# error "do not include mach/mem_map.h directly -- use asm/mem_map.h"
+#endif
 
 /* Async Memory Banks */
 #define ASYNC_BANK3_BASE	0x20300000	 /* Async Bank 3 */
@@ -166,20 +144,4 @@
 
 #endif
 
-/* Level 2 Memory - none */
-
-#define L2_START	0
-#define L2_LENGTH	0
-
-/* Scratch Pad Memory */
-
-#define L1_SCRATCH_START	0xFFB00000
-#define L1_SCRATCH_LENGTH	0x1000
-
-#define GET_PDA_SAFE(preg)		\
-	preg.l = _cpu_pda;		\
-	preg.h = _cpu_pda;
-
-#define GET_PDA(preg, dreg)	GET_PDA_SAFE(preg)
-
-#endif				/* _MEM_MAP_537_H_ */
+#endif
diff --git a/arch/blackfin/mach-bf538/include/mach/blackfin.h b/arch/blackfin/mach-bf538/include/mach/blackfin.h
index 6f628353dde3..9496196ac164 100644
--- a/arch/blackfin/mach-bf538/include/mach/blackfin.h
+++ b/arch/blackfin/mach-bf538/include/mach/blackfin.h
@@ -35,7 +35,6 @@
 #define BF538_FAMILY
 
 #include "bf538.h"
-#include "mem_map.h"
 #include "defBF539.h"
 #include "anomaly.h"
 
diff --git a/arch/blackfin/mach-bf538/include/mach/mem_map.h b/arch/blackfin/mach-bf538/include/mach/mem_map.h
index 76811966690e..aff00f453e9e 100644
--- a/arch/blackfin/mach-bf538/include/mach/mem_map.h
+++ b/arch/blackfin/mach-bf538/include/mach/mem_map.h
@@ -1,38 +1,16 @@
 /*
- * File:         include/asm-blackfin/mach-bf538/mem_map.h
- * Based on:
- * Author:
+ * BF538 memory map
  *
- * Created:
- * Description:
- *
- * Rev:
- *
- * Modified:
- *
- * Bugs:         Enter bugs at http://blackfin.uclinux.org/
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2, or (at your option)
- * any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; see the file COPYING.
- * If not, write to the Free Software Foundation,
- * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ * Copyright 2004-2009 Analog Devices Inc.
+ * Licensed under the GPL-2 or later.
  */
 
-#ifndef _MEM_MAP_538_H_
-#define _MEM_MAP_538_H_
+#ifndef __BFIN_MACH_MEM_MAP_H__
+#define __BFIN_MACH_MEM_MAP_H__
 
-#define COREMMR_BASE           0xFFE00000	 /* Core MMRs */
-#define SYSMMR_BASE            0xFFC00000	 /* System MMRs */
+#ifndef __BFIN_MEM_MAP_H__
+# error "do not include mach/mem_map.h directly -- use asm/mem_map.h"
+#endif
 
 /* Async Memory Banks */
 #define ASYNC_BANK3_BASE	0x20300000	 /* Async Bank 3 */
@@ -93,21 +71,4 @@
 #define BFIN_DSUPBANKS	0
 #endif /*CONFIG_BFIN_DCACHE*/
 
-
-/* Level 2 Memory - none */
-
-#define L2_START	0
-#define L2_LENGTH	0
-
-/* Scratch Pad Memory */
-
-#define L1_SCRATCH_START	0xFFB00000
-#define L1_SCRATCH_LENGTH	0x1000
-
-#define GET_PDA_SAFE(preg)		\
-	preg.l = _cpu_pda;		\
-	preg.h = _cpu_pda;
-
-#define GET_PDA(preg, dreg)	GET_PDA_SAFE(preg)
-
-#endif				/* _MEM_MAP_538_H_ */
+#endif
diff --git a/arch/blackfin/mach-bf548/include/mach/blackfin.h b/arch/blackfin/mach-bf548/include/mach/blackfin.h
index cf6c1500222a..6b97396d817f 100644
--- a/arch/blackfin/mach-bf548/include/mach/blackfin.h
+++ b/arch/blackfin/mach-bf548/include/mach/blackfin.h
@@ -33,7 +33,6 @@
 #define _MACH_BLACKFIN_H_
 
 #include "bf548.h"
-#include "mem_map.h"
 #include "anomaly.h"
 
 #ifdef CONFIG_BF542
diff --git a/arch/blackfin/mach-bf548/include/mach/mem_map.h b/arch/blackfin/mach-bf548/include/mach/mem_map.h
index 70b9c1194024..caac2dfb41eb 100644
--- a/arch/blackfin/mach-bf548/include/mach/mem_map.h
+++ b/arch/blackfin/mach-bf548/include/mach/mem_map.h
@@ -1,38 +1,16 @@
 /*
- * file:         include/asm-blackfin/mach-bf548/mem_map.h
- * based on:
- * author:
+ * BF548 memory map
  *
- * created:
- * description:
- *	Memory MAP Common header file for blackfin BF537/6/4 of processors.
- * rev:
- *
- * modified:
- *
- * bugs:         enter bugs at http://blackfin.uclinux.org/
- *
- * this program is free software; you can redistribute it and/or modify
- * it under the terms of the gnu general public license as published by
- * the free software foundation; either version 2, or (at your option)
- * any later version.
- *
- * this program is distributed in the hope that it will be useful,
- * but without any warranty; without even the implied warranty of
- * merchantability or fitness for a particular purpose.  see the
- * gnu general public license for more details.
- *
- * you should have received a copy of the gnu general public license
- * along with this program; see the file copying.
- * if not, write to the free software foundation,
- * 59 temple place - suite 330, boston, ma 02111-1307, usa.
+ * Copyright 2004-2009 Analog Devices Inc.
+ * Licensed under the GPL-2 or later.
  */
 
-#ifndef _MEM_MAP_548_H_
-#define _MEM_MAP_548_H_
+#ifndef __BFIN_MACH_MEM_MAP_H__
+#define __BFIN_MACH_MEM_MAP_H__
 
-#define COREMMR_BASE           0xFFE00000	 /* Core MMRs */
-#define SYSMMR_BASE            0xFFC00000	 /* System MMRs */
+#ifndef __BFIN_MEM_MAP_H__
+# error "do not include mach/mem_map.h directly -- use asm/mem_map.h"
+#endif
 
 /* Async Memory Banks */
 #define ASYNC_BANK3_BASE	0x2C000000	 /* Async Bank 3 */
@@ -103,15 +81,4 @@
 # define L2_LENGTH          0x20000
 #endif
 
-/* Scratch Pad Memory */
-
-#define L1_SCRATCH_START	0xFFB00000
-#define L1_SCRATCH_LENGTH	0x1000
-
-#define GET_PDA_SAFE(preg)		\
-	preg.l = _cpu_pda;		\
-	preg.h = _cpu_pda;
-
-#define GET_PDA(preg, dreg)	GET_PDA_SAFE(preg)
-
-#endif/* _MEM_MAP_548_H_ */
+#endif
diff --git a/arch/blackfin/mach-bf561/include/mach/blackfin.h b/arch/blackfin/mach-bf561/include/mach/blackfin.h
index f79f6626b7ec..8be31358ef88 100644
--- a/arch/blackfin/mach-bf561/include/mach/blackfin.h
+++ b/arch/blackfin/mach-bf561/include/mach/blackfin.h
@@ -34,7 +34,6 @@
 #define BF561_FAMILY
 
 #include "bf561.h"
-#include "mem_map.h"
 #include "defBF561.h"
 #include "anomaly.h"
 
diff --git a/arch/blackfin/mach-bf561/include/mach/mem_map.h b/arch/blackfin/mach-bf561/include/mach/mem_map.h
index 419dffdc96eb..a63e15c86d90 100644
--- a/arch/blackfin/mach-bf561/include/mach/mem_map.h
+++ b/arch/blackfin/mach-bf561/include/mach/mem_map.h
@@ -1,13 +1,16 @@
 /*
- * Memory MAP
- * Common header file for blackfin BF561 of processors.
+ * BF561 memory map
+ *
+ * Copyright 2004-2009 Analog Devices Inc.
+ * Licensed under the GPL-2 or later.
  */
 
-#ifndef _MEM_MAP_561_H_
-#define _MEM_MAP_561_H_
+#ifndef __BFIN_MACH_MEM_MAP_H__
+#define __BFIN_MACH_MEM_MAP_H__
 
-#define COREMMR_BASE           0xFFE00000	 /* Core MMRs */
-#define SYSMMR_BASE            0xFFC00000	 /* System MMRs */
+#ifndef __BFIN_MEM_MAP_H__
+# error "do not include mach/mem_map.h directly -- use asm/mem_map.h"
+#endif
 
 /* Async Memory Banks */
 #define ASYNC_BANK3_BASE	0x2C000000	 /* Async Bank 3 */
@@ -82,9 +85,6 @@
 #define COREA_L1_SCRATCH_START	0xFFB00000
 #define COREB_L1_SCRATCH_START	0xFF700000
 
-#define L1_SCRATCH_START	COREA_L1_SCRATCH_START
-#define L1_SCRATCH_LENGTH	0x1000
-
 #ifdef __ASSEMBLY__
 
 /*
@@ -155,14 +155,42 @@
 	dreg = ROT dreg BY -1;		\
 	dreg = CC;
 
-#else
-#define GET_PDA_SAFE(preg)		\
-	preg.l = _cpu_pda;		\
-	preg.h = _cpu_pda;
+static inline unsigned long get_l1_scratch_start_cpu(int cpu)
+{
+	return cpu ? COREB_L1_SCRATCH_START : COREA_L1_SCRATCH_START;
+}
+static inline unsigned long get_l1_code_start_cpu(int cpu)
+{
+	return cpu ? COREB_L1_CODE_START : COREA_L1_CODE_START;
+}
+static inline unsigned long get_l1_data_a_start_cpu(int cpu)
+{
+	return cpu ? COREB_L1_DATA_A_START : COREA_L1_DATA_A_START;
+}
+static inline unsigned long get_l1_data_b_start_cpu(int cpu)
+{
+	return cpu ? COREB_L1_DATA_B_START : COREA_L1_DATA_B_START;
+}
+
+static inline unsigned long get_l1_scratch_start(void)
+{
+	return get_l1_scratch_start_cpu(blackfin_core_id());
+}
+static inline unsigned long get_l1_code_start(void)
+{
+	return get_l1_code_start_cpu(blackfin_core_id());
+}
+static inline unsigned long get_l1_data_a_start(void)
+{
+	return get_l1_data_a_start_cpu(blackfin_core_id());
+}
+static inline unsigned long get_l1_data_b_start(void)
+{
+	return get_l1_data_b_start_cpu(blackfin_core_id());
+}
 
-#define GET_PDA(preg, dreg)	GET_PDA_SAFE(preg)
 #endif /* CONFIG_SMP */
 
 #endif /* __ASSEMBLY__ */
 
-#endif				/* _MEM_MAP_533_H_ */
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From 8f580f7c82ed9edeb3629568aabcde2caff3f236 Mon Sep 17 00:00:00 2001
From: Graf Yang <graf.yang@analog.com>
Date: Thu, 18 Jun 2009 04:21:39 +0000
Subject: Blackfin: fix typo in TRAS define in mem_init.h header

We defined SDRAM_tRAS to TRAS_4, but then wrongly defined SDRAM_tRAS_num
to 3.

Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/include/asm/mem_init.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/blackfin/include/asm/mem_init.h b/arch/blackfin/include/asm/mem_init.h
index 61f7487fbf12..fc164b35a9c5 100644
--- a/arch/blackfin/include/asm/mem_init.h
+++ b/arch/blackfin/include/asm/mem_init.h
@@ -59,7 +59,7 @@
 #define SDRAM_tRP       TRP_1
 #define SDRAM_tRP_num   1
 #define SDRAM_tRAS      TRAS_4
-#define SDRAM_tRAS_num  3
+#define SDRAM_tRAS_num  4
 #define SDRAM_tRCD      TRCD_1
 #define SDRAM_tWR       TWR_2
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From ee48efb5dc45aeb9786dea6469d3e1bea5105036 Mon Sep 17 00:00:00 2001
From: Graf Yang <graf.yang@analog.com>
Date: Thu, 18 Jun 2009 04:32:04 +0000
Subject: Blackfin: bf526-ezbrd: handle different SDRAM chips

The BF526-EZBRD changed SDRAM chips between board revisions, so create a
timing table that can accommodate both.

Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/Kconfig                |  7 ++-
 arch/blackfin/include/asm/mem_init.h | 86 ++++++++++++++++++++++++++++++++++++
 2 files changed, 92 insertions(+), 1 deletion(-)

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index 220635a1ebdd..384f7cd6b41e 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -358,7 +358,7 @@ config MEM_MT48LC8M32B2B5_7
 
 config MEM_MT48LC32M16A2TG_75
 	bool
-	depends on (BFIN527_EZKIT || BFIN532_IP0X || BLACKSTAMP || BFIN526_EZBRD)
+	depends on (BFIN527_EZKIT || BFIN532_IP0X || BLACKSTAMP)
 	default y
 
 config MEM_MT48LC32M8A2_75
@@ -366,6 +366,11 @@ config MEM_MT48LC32M8A2_75
 	depends on (BFIN518F_EZBRD)
 	default y
 
+config MEM_MT48H32M16LFCJ_75
+	bool
+	depends on (BFIN526_EZBRD)
+	default y
+
 source "arch/blackfin/mach-bf518/Kconfig"
 source "arch/blackfin/mach-bf527/Kconfig"
 source "arch/blackfin/mach-bf533/Kconfig"
diff --git a/arch/blackfin/include/asm/mem_init.h b/arch/blackfin/include/asm/mem_init.h
index fc164b35a9c5..4179e329b9c9 100644
--- a/arch/blackfin/include/asm/mem_init.h
+++ b/arch/blackfin/include/asm/mem_init.h
@@ -89,6 +89,85 @@
 #endif
 #endif
 
+/*
+ * The BF526-EZ-Board changed SDRAM chips between revisions,
+ * so we use below timings to accommodate both.
+ */
+#if defined(CONFIG_MEM_MT48H32M16LFCJ_75)
+#if (CONFIG_SCLK_HZ > 119402985)
+#define SDRAM_tRP       TRP_2
+#define SDRAM_tRP_num   2
+#define SDRAM_tRAS      TRAS_8
+#define SDRAM_tRAS_num  8
+#define SDRAM_tRCD      TRCD_2
+#define SDRAM_tWR       TWR_2
+#endif
+#if (CONFIG_SCLK_HZ > 104477612) && (CONFIG_SCLK_HZ <= 119402985)
+#define SDRAM_tRP       TRP_2
+#define SDRAM_tRP_num   2
+#define SDRAM_tRAS      TRAS_7
+#define SDRAM_tRAS_num  7
+#define SDRAM_tRCD      TRCD_2
+#define SDRAM_tWR       TWR_2
+#endif
+#if (CONFIG_SCLK_HZ > 89552239) && (CONFIG_SCLK_HZ <= 104477612)
+#define SDRAM_tRP       TRP_2
+#define SDRAM_tRP_num   2
+#define SDRAM_tRAS      TRAS_6
+#define SDRAM_tRAS_num  6
+#define SDRAM_tRCD      TRCD_2
+#define SDRAM_tWR       TWR_2
+#endif
+#if (CONFIG_SCLK_HZ > 74626866) && (CONFIG_SCLK_HZ <= 89552239)
+#define SDRAM_tRP       TRP_2
+#define SDRAM_tRP_num   2
+#define SDRAM_tRAS      TRAS_5
+#define SDRAM_tRAS_num  5
+#define SDRAM_tRCD      TRCD_2
+#define SDRAM_tWR       TWR_2
+#endif
+#if (CONFIG_SCLK_HZ > 66666667) && (CONFIG_SCLK_HZ <= 74626866)
+#define SDRAM_tRP       TRP_2
+#define SDRAM_tRP_num   2
+#define SDRAM_tRAS      TRAS_4
+#define SDRAM_tRAS_num  4
+#define SDRAM_tRCD      TRCD_2
+#define SDRAM_tWR       TWR_2
+#endif
+#if (CONFIG_SCLK_HZ > 59701493) && (CONFIG_SCLK_HZ <= 66666667)
+#define SDRAM_tRP       TRP_2
+#define SDRAM_tRP_num   2
+#define SDRAM_tRAS      TRAS_4
+#define SDRAM_tRAS_num  4
+#define SDRAM_tRCD      TRCD_1
+#define SDRAM_tWR       TWR_2
+#endif
+#if (CONFIG_SCLK_HZ > 44776119) && (CONFIG_SCLK_HZ <= 59701493)
+#define SDRAM_tRP       TRP_2
+#define SDRAM_tRP_num   2
+#define SDRAM_tRAS      TRAS_3
+#define SDRAM_tRAS_num  3
+#define SDRAM_tRCD      TRCD_1
+#define SDRAM_tWR       TWR_2
+#endif
+#if (CONFIG_SCLK_HZ > 29850746) && (CONFIG_SCLK_HZ <= 44776119)
+#define SDRAM_tRP       TRP_1
+#define SDRAM_tRP_num   1
+#define SDRAM_tRAS      TRAS_3
+#define SDRAM_tRAS_num  3
+#define SDRAM_tRCD      TRCD_1
+#define SDRAM_tWR       TWR_2
+#endif
+#if (CONFIG_SCLK_HZ <= 29850746)
+#define SDRAM_tRP       TRP_1
+#define SDRAM_tRP_num   1
+#define SDRAM_tRAS      TRAS_2
+#define SDRAM_tRAS_num  2
+#define SDRAM_tRCD      TRCD_1
+#define SDRAM_tWR       TWR_2
+#endif
+#endif
+
 #if defined(CONFIG_MEM_MT48LC16M8A2TG_75) || \
     defined(CONFIG_MEM_MT48LC8M32B2B5_7)
   /*SDRAM INFORMATION: */
@@ -109,6 +188,13 @@
 #define SDRAM_CL    CL_3
 #endif
 
+#if defined(CONFIG_MEM_MT48H32M16LFCJ_75)
+  /*SDRAM INFORMATION: */
+#define SDRAM_Tref  64		/* Refresh period in milliseconds   */
+#define SDRAM_NRA   8192	/* Number of row addresses in SDRAM */
+#define SDRAM_CL    CL_2
+#endif
+
 
 #ifdef CONFIG_BFIN_KERNEL_CLOCK_MEMINIT_CALC
 /* Equation from section 17 (p17-46) of BF533 HRM */
-- 
cgit v1.2.3-59-g8ed1b


From 3a7f5b1605570f5259e71d1846be27588a00bbaf Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Thu, 18 Jun 2009 19:13:03 +0000
Subject: Blackfin: drop unused ISP1760 port1_disable from board resources

The port1 disable stuff was dropped from the USB ISP1760, so update the
Blackfin boards accordingly.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/mach-bf527/boards/cm_bf527.c | 1 -
 arch/blackfin/mach-bf527/boards/ezkit.c    | 1 -
 arch/blackfin/mach-bf537/boards/stamp.c    | 1 -
 arch/blackfin/mach-bf548/boards/ezkit.c    | 1 -
 arch/blackfin/mach-bf561/boards/ezkit.c    | 1 -
 5 files changed, 5 deletions(-)

diff --git a/arch/blackfin/mach-bf527/boards/cm_bf527.c b/arch/blackfin/mach-bf527/boards/cm_bf527.c
index 1eaf27ff722e..f4867ce0c618 100644
--- a/arch/blackfin/mach-bf527/boards/cm_bf527.c
+++ b/arch/blackfin/mach-bf527/boards/cm_bf527.c
@@ -78,7 +78,6 @@ static struct resource bfin_isp1760_resources[] = {
 
 static struct isp1760_platform_data isp1760_priv = {
 	.is_isp1761 = 0,
-	.port1_disable = 0,
 	.bus_width_16 = 1,
 	.port1_otg = 0,
 	.analog_oc = 0,
diff --git a/arch/blackfin/mach-bf527/boards/ezkit.c b/arch/blackfin/mach-bf527/boards/ezkit.c
index 3e5b7db6b065..799a1d1fa890 100644
--- a/arch/blackfin/mach-bf527/boards/ezkit.c
+++ b/arch/blackfin/mach-bf527/boards/ezkit.c
@@ -77,7 +77,6 @@ static struct resource bfin_isp1760_resources[] = {
 
 static struct isp1760_platform_data isp1760_priv = {
 	.is_isp1761 = 0,
-	.port1_disable = 0,
 	.bus_width_16 = 1,
 	.port1_otg = 0,
 	.analog_oc = 0,
diff --git a/arch/blackfin/mach-bf537/boards/stamp.c b/arch/blackfin/mach-bf537/boards/stamp.c
index ff7228caa7da..c1f76dd2c4ed 100644
--- a/arch/blackfin/mach-bf537/boards/stamp.c
+++ b/arch/blackfin/mach-bf537/boards/stamp.c
@@ -79,7 +79,6 @@ static struct resource bfin_isp1760_resources[] = {
 
 static struct isp1760_platform_data isp1760_priv = {
 	.is_isp1761 = 0,
-	.port1_disable = 0,
 	.bus_width_16 = 1,
 	.port1_otg = 0,
 	.analog_oc = 0,
diff --git a/arch/blackfin/mach-bf548/boards/ezkit.c b/arch/blackfin/mach-bf548/boards/ezkit.c
index 805a57b5e650..81f5b95cc361 100644
--- a/arch/blackfin/mach-bf548/boards/ezkit.c
+++ b/arch/blackfin/mach-bf548/boards/ezkit.c
@@ -76,7 +76,6 @@ static struct resource bfin_isp1760_resources[] = {
 
 static struct isp1760_platform_data isp1760_priv = {
 	.is_isp1761 = 0,
-	.port1_disable = 0,
 	.bus_width_16 = 1,
 	.port1_otg = 0,
 	.analog_oc = 0,
diff --git a/arch/blackfin/mach-bf561/boards/ezkit.c b/arch/blackfin/mach-bf561/boards/ezkit.c
index b5ef7ff7b7bd..4df904f9e90a 100644
--- a/arch/blackfin/mach-bf561/boards/ezkit.c
+++ b/arch/blackfin/mach-bf561/boards/ezkit.c
@@ -62,7 +62,6 @@ static struct resource bfin_isp1760_resources[] = {
 
 static struct isp1760_platform_data isp1760_priv = {
 	.is_isp1761 = 0,
-	.port1_disable = 0,
 	.bus_width_16 = 1,
 	.port1_otg = 0,
 	.analog_oc = 0,
-- 
cgit v1.2.3-59-g8ed1b


From 3aca47c02097a78a566f67e7ec5fa3e0f2583a73 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Thu, 18 Jun 2009 19:40:47 +0000
Subject: Blackfin: fix GPTMR0_CLOCKSOURCE dependency on BFIN_GPTIMERS

The GPTMR0_CLOCKSOURCE Kconfig option requires the gptimers framework, so
make sure it is selected when this option is enabled.

Reported-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index 384f7cd6b41e..7faa2f554ab1 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -648,6 +648,7 @@ config CYCLES_CLOCKSOURCE
 
 config GPTMR0_CLOCKSOURCE
 	bool "Use GPTimer0 as a clocksource (higher rating)"
+	select BFIN_GPTIMERS
 	depends on GENERIC_CLOCKEVENTS
 	depends on !TICKSOURCE_GPTMR0
 
-- 
cgit v1.2.3-59-g8ed1b


From bd854c077e660b5f44b5049219645042bcba61ac Mon Sep 17 00:00:00 2001
From: Robin Getz <robin.getz@analog.com>
Date: Thu, 18 Jun 2009 22:53:43 +0000
Subject: Blackfin: fix early crash when booting on wrong cpu

Make sure we process the kernel command line before poking the hardware,
so that we can process early printk.  This helps ensure that if you boot
a kernel configured for a different processor, something will be left in
the log buffer.

Signed-off-by: Robin Getz <robin.getz@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/kernel/setup.c | 23 +++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)

diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c
index 8d7892820130..298f023bcc09 100644
--- a/arch/blackfin/kernel/setup.c
+++ b/arch/blackfin/kernel/setup.c
@@ -477,9 +477,11 @@ static __init void parse_cmdline_early(char *cmdline_p)
 			} else if (!memcmp(to, "clkin_hz=", 9)) {
 				to += 9;
 				early_init_clkin_hz(to);
+#ifdef CONFIG_EARLY_PRINTK
 			} else if (!memcmp(to, "earlyprintk=", 12)) {
 				to += 12;
 				setup_early_printk(to);
+#endif
 			} else if (!memcmp(to, "memmap=", 7)) {
 				to += 7;
 				parse_memmap(to);
@@ -798,6 +800,11 @@ void __init setup_arch(char **cmdline_p)
 {
 	unsigned long sclk, cclk;
 
+	/* Check to make sure we are running on the right processor */
+	if (unlikely(CPUID != bfin_cpuid()))
+		printk(KERN_ERR "ERROR: Not running on ADSP-%s: unknown CPUID 0x%04x Rev 0.%d\n",
+			CPU, bfin_cpuid(), bfin_revid());
+
 #ifdef CONFIG_DUMMY_CONSOLE
 	conswitchp = &dummy_con;
 #endif
@@ -812,14 +819,17 @@ void __init setup_arch(char **cmdline_p)
 	memcpy(boot_command_line, command_line, COMMAND_LINE_SIZE);
 	boot_command_line[COMMAND_LINE_SIZE - 1] = '\0';
 
-	/* setup memory defaults from the user config */
-	physical_mem_end = 0;
-	_ramend = get_mem_size() * 1024 * 1024;
-
 	memset(&bfin_memmap, 0, sizeof(bfin_memmap));
 
+	/* If the user does not specify things on the command line, use
+	 * what the bootloader set things up as
+	 */
+	physical_mem_end = 0;
 	parse_cmdline_early(&command_line[0]);
 
+	if (_ramend == 0)
+		_ramend = get_mem_size() * 1024 * 1024;
+
 	if (physical_mem_end == 0)
 		physical_mem_end = _ramend;
 
@@ -910,10 +920,7 @@ void __init setup_arch(char **cmdline_p)
 	else
 		printk(KERN_INFO "Compiled for ADSP-%s Rev 0.%d\n", CPU, bfin_compiled_revid());
 
-	if (unlikely(CPUID != bfin_cpuid()))
-		printk(KERN_ERR "ERROR: Not running on ADSP-%s: unknown CPUID 0x%04x Rev 0.%d\n",
-			CPU, bfin_cpuid(), bfin_revid());
-	else {
+	if (likely(CPUID == bfin_cpuid())) {
 		if (bfin_revid() != bfin_compiled_revid()) {
 			if (bfin_compiled_revid() == -1)
 				printk(KERN_ERR "Warning: Compiled for Rev none, but running on Rev %d\n",
-- 
cgit v1.2.3-59-g8ed1b


From 986d6c1e05642edac81cb8cc99f36a26d16ef220 Mon Sep 17 00:00:00 2001
From: Yi Li <yi.li@analog.com>
Date: Fri, 19 Jun 2009 08:51:11 +0000
Subject: Blackfin: drop BF535-specific text for exception 0x2A (unaligned
 instruction)

We don't support the BF535 at all, and the exception 0x2A text specific to
it is pretty verbose and confusing (since the behavior is simply odd), so
punt it to keep the noise down.

Signed-off-by: Yi Li <yi.li@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/include/asm/traps.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/arch/blackfin/include/asm/traps.h b/arch/blackfin/include/asm/traps.h
index 34f7295fb070..3cdc454cde23 100644
--- a/arch/blackfin/include/asm/traps.h
+++ b/arch/blackfin/include/asm/traps.h
@@ -111,9 +111,7 @@
 	level "   bits in the Watchpoint Instruction Address Control register (WPIACTL) is set.\n"
 #define EXC_0x2A(level) \
 	"Instruction fetch misaligned address violation\n" \
-	level " - Attempted misaligned instruction cache fetch. On a misaligned instruction fetch\n" \
-	level "   exception, the return address provided in RETX is the destination address which is\n" \
-	level "   misaligned, rather than the address of the offending instruction.\n"
+	level " - Attempted misaligned instruction cache fetch.\n"
 #define EXC_0x2B(level) \
 	"CPLB protection violation\n" \
 	level " - Illegal instruction fetch access (memory protection violation).\n"
-- 
cgit v1.2.3-59-g8ed1b


From 5ecf3e03cd513e8dba080d389b56bac11a2b0d8a Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Fri, 19 Jun 2009 18:56:57 -0400
Subject: Blackfin: hook up new perf_counter_open syscall

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/include/asm/unistd.h | 3 ++-
 arch/blackfin/mach-common/entry.S  | 1 +
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/blackfin/include/asm/unistd.h b/arch/blackfin/include/asm/unistd.h
index da35133c171d..c8e7ee4768cd 100644
--- a/arch/blackfin/include/asm/unistd.h
+++ b/arch/blackfin/include/asm/unistd.h
@@ -381,8 +381,9 @@
 #define __NR_preadv		366
 #define __NR_pwritev		367
 #define __NR_rt_tgsigqueueinfo	368
+#define __NR_perf_counter_open	369
 
-#define __NR_syscall		369
+#define __NR_syscall		370
 #define NR_syscalls		__NR_syscall
 
 /* Old optional stuff no one actually uses */
diff --git a/arch/blackfin/mach-common/entry.S b/arch/blackfin/mach-common/entry.S
index 31fa313e81cf..5a4e7c7fd92c 100644
--- a/arch/blackfin/mach-common/entry.S
+++ b/arch/blackfin/mach-common/entry.S
@@ -1609,6 +1609,7 @@ ENTRY(_sys_call_table)
 	.long _sys_preadv
 	.long _sys_pwritev
 	.long _sys_rt_tgsigqueueinfo
+	.long _sys_perf_counter_open
 
 	.rept NR_syscalls-(.-_sys_call_table)/4
 	.long _sys_ni_syscall
-- 
cgit v1.2.3-59-g8ed1b


From 9e06dd39f2b6d7e35981e0d7aded618686b32ccb Mon Sep 17 00:00:00 2001
From: Jesse Barnes <jbarnes@virtuousgeek.org>
Date: Mon, 22 Jun 2009 18:05:12 -0700
Subject: drm/i915: correct suspend/resume ordering

We need to save register state *after* idling GEM, clearing the ring,
and uninstalling the IRQ handler, or we might end up saving bogus
fence regs, for one.  Our restore ordering should already be correct,
since we do GEM, ring and IRQ init after restoring the last register
state, which prevents us from clobbering things.

I put this together to potentially address a bug, but I haven't heard
back if it fixes it yet.  However I think it stands on its own, so I'm
sending it in.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
---
 drivers/gpu/drm/i915/i915_drv.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 98560e1e899a..e3cb4025e323 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -67,8 +67,6 @@ static int i915_suspend(struct drm_device *dev, pm_message_t state)
 
 	pci_save_state(dev->pdev);
 
-	i915_save_state(dev);
-
 	/* If KMS is active, we do the leavevt stuff here */
 	if (drm_core_check_feature(dev, DRIVER_MODESET)) {
 		if (i915_gem_idle(dev))
@@ -77,6 +75,8 @@ static int i915_suspend(struct drm_device *dev, pm_message_t state)
 		drm_irq_uninstall(dev);
 	}
 
+	i915_save_state(dev);
+
 	intel_opregion_free(dev, 1);
 
 	if (state.event == PM_EVENT_SUSPEND) {
-- 
cgit v1.2.3-59-g8ed1b


From 42b86e06c7db365f1947dda9b75317cbb3c9fb5b Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Mon, 22 Jun 2009 21:48:37 -0400
Subject: Blackfin: fix dma-mapping build errors

The recent deprecation of dma_sync_{sg,single} ironically broke Blackfin
systems.  This is because we don't define dma_sync_sg_for_cpu at all, so
until the DMA asm-generic conversion/cleanup is done after the next
release, simply stub out the dma_sync_sg_for_{cpu,device} functions.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
---
 arch/blackfin/include/asm/dma-mapping.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/arch/blackfin/include/asm/dma-mapping.h b/arch/blackfin/include/asm/dma-mapping.h
index d7d9148e433c..ed6b1f3cccce 100644
--- a/arch/blackfin/include/asm/dma-mapping.h
+++ b/arch/blackfin/include/asm/dma-mapping.h
@@ -95,4 +95,17 @@ static inline void dma_sync_single_for_device(struct device *dev,
 					enum dma_data_direction dir)
 {
 }
+
+static inline void dma_sync_sg_for_cpu(struct device *dev,
+					struct scatterlist *sg,
+					int nents, enum dma_data_direction dir)
+{
+}
+
+static inline void dma_sync_sg_for_device(struct device *dev,
+					struct scatterlist *sg,
+					int nents, enum dma_data_direction dir)
+{
+}
+
 #endif				/* _BLACKFIN_DMA_MAPPING_H */
-- 
cgit v1.2.3-59-g8ed1b


From 3fbe18d65d66054667aaee849bed74674bb50062 Mon Sep 17 00:00:00 2001
From: Zhao Yakui <yakui.zhao@intel.com>
Date: Mon, 22 Jun 2009 15:31:25 +0800
Subject: drm/i915: Add support for changing LVDS panel fitting using an output
 property.

Previously the driver would always scale the chosen video mode to fill the
panel.  This adds 1:1 and maintain-aspect-ratio scaling modes.

v2: the drm_calloc/drm_free is replaced by kzalloc/kfree based
on Eric's suggestion.

Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
---
 drivers/gpu/drm/i915/i915_reg.h   |  16 +++
 drivers/gpu/drm/i915/intel_lvds.c | 285 +++++++++++++++++++++++++++++++++++---
 2 files changed, 280 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 544d5677a2fa..88bf7521405f 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -847,9 +847,25 @@
 #define   HORIZ_INTERP_MASK	(3 << 6)
 #define   HORIZ_AUTO_SCALE	(1 << 5)
 #define   PANEL_8TO6_DITHER_ENABLE (1 << 3)
+#define   PFIT_FILTER_FUZZY	(0 << 24)
+#define   PFIT_SCALING_AUTO	(0 << 26)
+#define   PFIT_SCALING_PROGRAMMED (1 << 26)
+#define   PFIT_SCALING_PILLAR	(2 << 26)
+#define   PFIT_SCALING_LETTER	(3 << 26)
 #define PFIT_PGM_RATIOS	0x61234
 #define   PFIT_VERT_SCALE_MASK			0xfff00000
 #define   PFIT_HORIZ_SCALE_MASK			0x0000fff0
+/* Pre-965 */
+#define		PFIT_VERT_SCALE_SHIFT		20
+#define		PFIT_VERT_SCALE_MASK		0xfff00000
+#define		PFIT_HORIZ_SCALE_SHIFT		4
+#define		PFIT_HORIZ_SCALE_MASK		0x0000fff0
+/* 965+ */
+#define		PFIT_VERT_SCALE_SHIFT_965	16
+#define		PFIT_VERT_SCALE_MASK_965	0x1fff0000
+#define		PFIT_HORIZ_SCALE_SHIFT_965	0
+#define		PFIT_HORIZ_SCALE_MASK_965	0x00001fff
+
 #define PFIT_AUTO_RATIOS 0x61238
 
 /* Backlight control */
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 345e5055f1c0..f416ead71204 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -39,6 +39,21 @@
 
 #define I915_LVDS "i915_lvds"
 
+/*
+ * the following four scaling options are defined.
+ * #define DRM_MODE_SCALE_NON_GPU	0
+ * #define DRM_MODE_SCALE_FULLSCREEN	1
+ * #define DRM_MODE_SCALE_NO_SCALE	2
+ * #define DRM_MODE_SCALE_ASPECT	3
+ */
+
+/* Private structure for the integrated LVDS support */
+struct intel_lvds_priv {
+	int fitting_mode;
+	u32 pfit_control;
+	u32 pfit_pgm_ratios;
+};
+
 /**
  * Sets the backlight level.
  *
@@ -213,10 +228,24 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder,
 				  struct drm_display_mode *mode,
 				  struct drm_display_mode *adjusted_mode)
 {
+	/*
+	 * float point operation is not supported . So the PANEL_RATIO_FACTOR
+	 * is defined, which can avoid the float point computation when
+	 * calculating the panel ratio.
+	 */
+#define PANEL_RATIO_FACTOR 8192
 	struct drm_device *dev = encoder->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
 	struct drm_encoder *tmp_encoder;
+	struct intel_output *intel_output = enc_to_intel_output(encoder);
+	struct intel_lvds_priv *lvds_priv = intel_output->dev_priv;
+	u32 pfit_control = 0, pfit_pgm_ratios = 0;
+	int left_border = 0, right_border = 0, top_border = 0;
+	int bottom_border = 0;
+	bool border = 0;
+	int panel_ratio, desired_ratio, vert_scale, horiz_scale;
+	int horiz_ratio, vert_ratio;
 
 	/* Should never happen!! */
 	if (!IS_I965G(dev) && intel_crtc->pipe == 0) {
@@ -232,7 +261,9 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder,
 			return false;
 		}
 	}
-
+	/* If we don't have a panel mode, there is nothing we can do */
+	if (dev_priv->panel_fixed_mode == NULL)
+		return true;
 	/*
 	 * If we have timings from the BIOS for the panel, put them in
 	 * to the adjusted mode.  The CRTC will be set up for this mode,
@@ -256,6 +287,191 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder,
 		drm_mode_set_crtcinfo(adjusted_mode, CRTC_INTERLACE_HALVE_V);
 	}
 
+	/* Make sure pre-965s set dither correctly */
+	if (!IS_I965G(dev)) {
+		if (dev_priv->panel_wants_dither || dev_priv->lvds_dither)
+			pfit_control |= PANEL_8TO6_DITHER_ENABLE;
+	}
+
+	/* Native modes don't need fitting */
+	if (adjusted_mode->hdisplay == mode->hdisplay &&
+			adjusted_mode->vdisplay == mode->vdisplay) {
+		pfit_pgm_ratios = 0;
+		border = 0;
+		goto out;
+	}
+
+	/* 965+ wants fuzzy fitting */
+	if (IS_I965G(dev))
+		pfit_control |= (intel_crtc->pipe << PFIT_PIPE_SHIFT) |
+					PFIT_FILTER_FUZZY;
+
+	/*
+	 * Deal with panel fitting options. Figure out how to stretch the
+	 * image based on its aspect ratio & the current panel fitting mode.
+	 */
+	panel_ratio = adjusted_mode->hdisplay * PANEL_RATIO_FACTOR /
+				adjusted_mode->vdisplay;
+	desired_ratio = mode->hdisplay * PANEL_RATIO_FACTOR /
+				mode->vdisplay;
+	/*
+	 * Enable automatic panel scaling for non-native modes so that they fill
+	 * the screen.  Should be enabled before the pipe is enabled, according
+	 * to register description and PRM.
+	 * Change the value here to see the borders for debugging
+	 */
+	I915_WRITE(BCLRPAT_A, 0);
+	I915_WRITE(BCLRPAT_B, 0);
+
+	switch (lvds_priv->fitting_mode) {
+	case DRM_MODE_SCALE_NO_SCALE:
+		/*
+		 * For centered modes, we have to calculate border widths &
+		 * heights and modify the values programmed into the CRTC.
+		 */
+		left_border = (adjusted_mode->hdisplay - mode->hdisplay) / 2;
+		right_border = left_border;
+		if (mode->hdisplay & 1)
+			right_border++;
+		top_border = (adjusted_mode->vdisplay - mode->vdisplay) / 2;
+		bottom_border = top_border;
+		if (mode->vdisplay & 1)
+			bottom_border++;
+		/* Set active & border values */
+		adjusted_mode->crtc_hdisplay = mode->hdisplay;
+		adjusted_mode->crtc_hblank_start = mode->hdisplay +
+						right_border - 1;
+		adjusted_mode->crtc_hblank_end = adjusted_mode->crtc_htotal -
+						left_border - 1;
+		adjusted_mode->crtc_hsync_start =
+				adjusted_mode->crtc_hblank_start;
+		adjusted_mode->crtc_hsync_end =
+				adjusted_mode->crtc_hblank_end;
+		adjusted_mode->crtc_vdisplay = mode->vdisplay;
+		adjusted_mode->crtc_vblank_start = mode->vdisplay +
+						bottom_border - 1;
+		adjusted_mode->crtc_vblank_end = adjusted_mode->crtc_vtotal -
+						top_border - 1;
+		adjusted_mode->crtc_vsync_start =
+				adjusted_mode->crtc_vblank_start;
+		adjusted_mode->crtc_vsync_end =
+				adjusted_mode->crtc_vblank_end;
+		border = 1;
+		break;
+	case DRM_MODE_SCALE_ASPECT:
+		/* Scale but preserve the spect ratio */
+		pfit_control |= PFIT_ENABLE;
+		if (IS_I965G(dev)) {
+			/* 965+ is easy, it does everything in hw */
+			if (panel_ratio > desired_ratio)
+				pfit_control |= PFIT_SCALING_PILLAR;
+			else if (panel_ratio < desired_ratio)
+				pfit_control |= PFIT_SCALING_LETTER;
+			else
+				pfit_control |= PFIT_SCALING_AUTO;
+		} else {
+			/*
+			 * For earlier chips we have to calculate the scaling
+			 * ratio by hand and program it into the
+			 * PFIT_PGM_RATIO register
+			 */
+			u32 horiz_bits, vert_bits, bits = 12;
+			horiz_ratio = mode->hdisplay * PANEL_RATIO_FACTOR/
+						adjusted_mode->hdisplay;
+			vert_ratio = mode->vdisplay * PANEL_RATIO_FACTOR/
+						adjusted_mode->vdisplay;
+			horiz_scale = adjusted_mode->hdisplay *
+					PANEL_RATIO_FACTOR / mode->hdisplay;
+			vert_scale = adjusted_mode->vdisplay *
+					PANEL_RATIO_FACTOR / mode->vdisplay;
+
+			/* retain aspect ratio */
+			if (panel_ratio > desired_ratio) { /* Pillar */
+				u32 scaled_width;
+				scaled_width = mode->hdisplay * vert_scale /
+						PANEL_RATIO_FACTOR;
+				horiz_ratio = vert_ratio;
+				pfit_control |= (VERT_AUTO_SCALE |
+						 VERT_INTERP_BILINEAR |
+						 HORIZ_INTERP_BILINEAR);
+				/* Pillar will have left/right borders */
+				left_border = (adjusted_mode->hdisplay -
+						scaled_width) / 2;
+				right_border = left_border;
+				if (mode->hdisplay & 1) /* odd resolutions */
+					right_border++;
+				adjusted_mode->crtc_hdisplay = scaled_width;
+				adjusted_mode->crtc_hblank_start =
+					scaled_width + right_border - 1;
+				adjusted_mode->crtc_hblank_end =
+				 adjusted_mode->crtc_htotal - left_border - 1;
+				adjusted_mode->crtc_hsync_start =
+					adjusted_mode->crtc_hblank_start;
+				adjusted_mode->crtc_hsync_end =
+					adjusted_mode->crtc_hblank_end;
+				border = 1;
+			} else if (panel_ratio < desired_ratio) { /* letter */
+				u32 scaled_height = mode->vdisplay *
+					horiz_scale / PANEL_RATIO_FACTOR;
+				vert_ratio = horiz_ratio;
+				pfit_control |= (HORIZ_AUTO_SCALE |
+						 VERT_INTERP_BILINEAR |
+						 HORIZ_INTERP_BILINEAR);
+				/* Letterbox will have top/bottom border */
+				top_border = (adjusted_mode->vdisplay -
+					scaled_height) / 2;
+				bottom_border = top_border;
+				if (mode->vdisplay & 1)
+					bottom_border++;
+				adjusted_mode->crtc_vdisplay = scaled_height;
+				adjusted_mode->crtc_vblank_start =
+					scaled_height + bottom_border - 1;
+				adjusted_mode->crtc_vblank_end =
+				 adjusted_mode->crtc_vtotal - top_border - 1;
+				adjusted_mode->crtc_vsync_start =
+					adjusted_mode->crtc_vblank_start;
+				adjusted_mode->crtc_vsync_end =
+					adjusted_mode->crtc_vblank_end;
+				border = 1;
+			} else {
+			/* Aspects match, Let hw scale both directions */
+				pfit_control |= (VERT_AUTO_SCALE |
+						 HORIZ_AUTO_SCALE |
+						 VERT_INTERP_BILINEAR |
+						 HORIZ_INTERP_BILINEAR);
+			}
+			horiz_bits = (1 << bits) * horiz_ratio /
+					PANEL_RATIO_FACTOR;
+			vert_bits = (1 << bits) * vert_ratio /
+					PANEL_RATIO_FACTOR;
+			pfit_pgm_ratios =
+				((vert_bits << PFIT_VERT_SCALE_SHIFT) &
+						PFIT_VERT_SCALE_MASK) |
+				((horiz_bits << PFIT_HORIZ_SCALE_SHIFT) &
+						PFIT_HORIZ_SCALE_MASK);
+		}
+		break;
+
+	case DRM_MODE_SCALE_FULLSCREEN:
+		/*
+		 * Full scaling, even if it changes the aspect ratio.
+		 * Fortunately this is all done for us in hw.
+		 */
+		pfit_control |= PFIT_ENABLE;
+		if (IS_I965G(dev))
+			pfit_control |= PFIT_SCALING_AUTO;
+		else
+			pfit_control |= (VERT_AUTO_SCALE | HORIZ_AUTO_SCALE |
+					 VERT_INTERP_BILINEAR |
+					 HORIZ_INTERP_BILINEAR);
+		break;
+	default:
+		break;
+	}
+
+out:
+	lvds_priv->pfit_control = pfit_control;
+	lvds_priv->pfit_pgm_ratios = pfit_pgm_ratios;
 	/*
 	 * XXX: It would be nice to support lower refresh rates on the
 	 * panels to reduce power consumption, and perhaps match the
@@ -301,8 +517,8 @@ static void intel_lvds_mode_set(struct drm_encoder *encoder,
 {
 	struct drm_device *dev = encoder->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
-	u32 pfit_control;
+	struct intel_output *intel_output = enc_to_intel_output(encoder);
+	struct intel_lvds_priv *lvds_priv = intel_output->dev_priv;
 
 	/*
 	 * The LVDS pin pair will already have been turned on in the
@@ -319,22 +535,8 @@ static void intel_lvds_mode_set(struct drm_encoder *encoder,
 	 * screen.  Should be enabled before the pipe is enabled, according to
 	 * register description and PRM.
 	 */
-	if (mode->hdisplay != adjusted_mode->hdisplay ||
-	    mode->vdisplay != adjusted_mode->vdisplay)
-		pfit_control = (PFIT_ENABLE | VERT_AUTO_SCALE |
-				HORIZ_AUTO_SCALE | VERT_INTERP_BILINEAR |
-				HORIZ_INTERP_BILINEAR);
-	else
-		pfit_control = 0;
-
-	if (!IS_I965G(dev)) {
-		if (dev_priv->panel_wants_dither || dev_priv->lvds_dither)
-			pfit_control |= PANEL_8TO6_DITHER_ENABLE;
-	}
-	else
-		pfit_control |= intel_crtc->pipe << PFIT_PIPE_SHIFT;
-
-	I915_WRITE(PFIT_CONTROL, pfit_control);
+	I915_WRITE(PFIT_PGM_RATIOS, lvds_priv->pfit_pgm_ratios);
+	I915_WRITE(PFIT_CONTROL, lvds_priv->pfit_control);
 }
 
 /**
@@ -406,6 +608,34 @@ static int intel_lvds_set_property(struct drm_connector *connector,
 				   struct drm_property *property,
 				   uint64_t value)
 {
+	struct drm_device *dev = connector->dev;
+	struct intel_output *intel_output =
+			to_intel_output(connector);
+
+	if (property == dev->mode_config.scaling_mode_property &&
+				connector->encoder) {
+		struct drm_crtc *crtc = connector->encoder->crtc;
+		struct intel_lvds_priv *lvds_priv = intel_output->dev_priv;
+		if (value == DRM_MODE_SCALE_NON_GPU) {
+			DRM_DEBUG_KMS(I915_LVDS,
+					"non_GPU property is unsupported\n");
+			return 0;
+		}
+		if (lvds_priv->fitting_mode == value) {
+			/* the LVDS scaling property is not changed */
+			return 0;
+		}
+		lvds_priv->fitting_mode = value;
+		if (crtc && crtc->enabled) {
+			/*
+			 * If the CRTC is enabled, the display will be changed
+			 * according to the new panel fitting mode.
+			 */
+			drm_crtc_helper_set_mode(crtc, &crtc->mode,
+				crtc->x, crtc->y, crtc->fb);
+		}
+	}
+
 	return 0;
 }
 
@@ -518,6 +748,7 @@ void intel_lvds_init(struct drm_device *dev)
 	struct drm_encoder *encoder;
 	struct drm_display_mode *scan; /* *modes, *bios_mode; */
 	struct drm_crtc *crtc;
+	struct intel_lvds_priv *lvds_priv;
 	u32 lvds;
 	int pipe, gpio = GPIOC;
 
@@ -531,7 +762,8 @@ void intel_lvds_init(struct drm_device *dev)
 		gpio = PCH_GPIOC;
 	}
 
-	intel_output = kzalloc(sizeof(struct intel_output), GFP_KERNEL);
+	intel_output = kzalloc(sizeof(struct intel_output) +
+				sizeof(struct intel_lvds_priv), GFP_KERNEL);
 	if (!intel_output) {
 		return;
 	}
@@ -553,7 +785,18 @@ void intel_lvds_init(struct drm_device *dev)
 	connector->interlace_allowed = false;
 	connector->doublescan_allowed = false;
 
+	lvds_priv = (struct intel_lvds_priv *)(intel_output + 1);
+	intel_output->dev_priv = lvds_priv;
+	/* create the scaling mode property */
+	drm_mode_create_scaling_mode_property(dev);
+	/*
+	 * the initial panel fitting mode will be FULL_SCREEN.
+	 */
 
+	drm_connector_attach_property(&intel_output->base,
+				      dev->mode_config.scaling_mode_property,
+				      DRM_MODE_SCALE_FULLSCREEN);
+	lvds_priv->fitting_mode = DRM_MODE_SCALE_FULLSCREEN;
 	/*
 	 * LVDS discovery:
 	 * 1) check for EDID on DDC
@@ -649,5 +892,5 @@ failed:
 	if (intel_output->ddc_bus)
 		intel_i2c_destroy(intel_output->ddc_bus);
 	drm_connector_cleanup(connector);
-	kfree(connector);
+	kfree(intel_output);
 }
-- 
cgit v1.2.3-59-g8ed1b


From aa0261f230105b86409e29bbe851b09830d93d50 Mon Sep 17 00:00:00 2001
From: Zhao Yakui <yakui.zhao@intel.com>
Date: Mon, 22 Jun 2009 15:31:26 +0800
Subject: drm/i915: Don't change the blank/sync width when calculating scaled
 modes

Also, use the border instead of border minus one.

At the same time, make sure the horizontal border and hsync are even for
the LVDS that works in dual-channel mode. So both horizontal border and hsync
start are also changed to be even, even for the LVDS in single-channel mode.

https://bugs.freedesktop.org/show_bug.cgi?id=20951

Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
---
 drivers/gpu/drm/i915/intel_lvds.c | 91 +++++++++++++++++++++++++++++++--------
 1 file changed, 73 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index f416ead71204..9564ca44a977 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -246,6 +246,9 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder,
 	bool border = 0;
 	int panel_ratio, desired_ratio, vert_scale, horiz_scale;
 	int horiz_ratio, vert_ratio;
+	u32 hsync_width, vsync_width;
+	u32 hblank_width, vblank_width;
+	u32 hsync_pos, vsync_pos;
 
 	/* Should never happen!! */
 	if (!IS_I965G(dev) && intel_crtc->pipe == 0) {
@@ -306,6 +309,14 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder,
 		pfit_control |= (intel_crtc->pipe << PFIT_PIPE_SHIFT) |
 					PFIT_FILTER_FUZZY;
 
+	hsync_width = adjusted_mode->crtc_hsync_end -
+					adjusted_mode->crtc_hsync_start;
+	vsync_width = adjusted_mode->crtc_vsync_end -
+					adjusted_mode->crtc_vsync_start;
+	hblank_width = adjusted_mode->crtc_hblank_end -
+					adjusted_mode->crtc_hblank_start;
+	vblank_width = adjusted_mode->crtc_vblank_end -
+					adjusted_mode->crtc_vblank_start;
 	/*
 	 * Deal with panel fitting options. Figure out how to stretch the
 	 * image based on its aspect ratio & the current panel fitting mode.
@@ -339,23 +350,39 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder,
 			bottom_border++;
 		/* Set active & border values */
 		adjusted_mode->crtc_hdisplay = mode->hdisplay;
+		/* Keep the boder be even */
+		if (right_border & 1)
+			right_border++;
+		/* use the border directly instead of border minuse one */
 		adjusted_mode->crtc_hblank_start = mode->hdisplay +
-						right_border - 1;
-		adjusted_mode->crtc_hblank_end = adjusted_mode->crtc_htotal -
-						left_border - 1;
+						right_border;
+		/* keep the blank width constant */
+		adjusted_mode->crtc_hblank_end =
+			adjusted_mode->crtc_hblank_start + hblank_width;
+		/* get the hsync pos relative to hblank start */
+		hsync_pos = (hblank_width - hsync_width) / 2;
+		/* keep the hsync pos be even */
+		if (hsync_pos & 1)
+			hsync_pos++;
 		adjusted_mode->crtc_hsync_start =
-				adjusted_mode->crtc_hblank_start;
+				adjusted_mode->crtc_hblank_start + hsync_pos;
+		/* keep the hsync width constant */
 		adjusted_mode->crtc_hsync_end =
-				adjusted_mode->crtc_hblank_end;
+				adjusted_mode->crtc_hsync_start + hsync_width;
 		adjusted_mode->crtc_vdisplay = mode->vdisplay;
+		/* use the border instead of border minus one */
 		adjusted_mode->crtc_vblank_start = mode->vdisplay +
-						bottom_border - 1;
-		adjusted_mode->crtc_vblank_end = adjusted_mode->crtc_vtotal -
-						top_border - 1;
+						bottom_border;
+		/* keep the vblank width constant */
+		adjusted_mode->crtc_vblank_end =
+				adjusted_mode->crtc_vblank_start + vblank_width;
+		/* get the vsync start postion relative to vblank start */
+		vsync_pos = (vblank_width - vsync_width) / 2;
 		adjusted_mode->crtc_vsync_start =
-				adjusted_mode->crtc_vblank_start;
+				adjusted_mode->crtc_vblank_start + vsync_pos;
+		/* keep the vsync width constant */
 		adjusted_mode->crtc_vsync_end =
-				adjusted_mode->crtc_vblank_end;
+				adjusted_mode->crtc_vblank_start + vsync_width;
 		border = 1;
 		break;
 	case DRM_MODE_SCALE_ASPECT:
@@ -400,15 +427,32 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder,
 				right_border = left_border;
 				if (mode->hdisplay & 1) /* odd resolutions */
 					right_border++;
+				/* keep the border be even */
+				if (right_border & 1)
+					right_border++;
 				adjusted_mode->crtc_hdisplay = scaled_width;
+				/* use border instead of border minus one */
 				adjusted_mode->crtc_hblank_start =
-					scaled_width + right_border - 1;
+					scaled_width + right_border;
+				/* keep the hblank width constant */
 				adjusted_mode->crtc_hblank_end =
-				 adjusted_mode->crtc_htotal - left_border - 1;
+					adjusted_mode->crtc_hblank_start +
+							hblank_width;
+				/*
+				 * get the hsync start pos relative to
+				 * hblank start
+				 */
+				hsync_pos = (hblank_width - hsync_width) / 2;
+				/* keep the hsync_pos be even */
+				if (hsync_pos & 1)
+					hsync_pos++;
 				adjusted_mode->crtc_hsync_start =
-					adjusted_mode->crtc_hblank_start;
+					adjusted_mode->crtc_hblank_start +
+							hsync_pos;
+				/* keept hsync width constant */
 				adjusted_mode->crtc_hsync_end =
-					adjusted_mode->crtc_hblank_end;
+					adjusted_mode->crtc_hsync_start +
+							hsync_width;
 				border = 1;
 			} else if (panel_ratio < desired_ratio) { /* letter */
 				u32 scaled_height = mode->vdisplay *
@@ -424,14 +468,25 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder,
 				if (mode->vdisplay & 1)
 					bottom_border++;
 				adjusted_mode->crtc_vdisplay = scaled_height;
+				/* use border instead of border minus one */
 				adjusted_mode->crtc_vblank_start =
-					scaled_height + bottom_border - 1;
+					scaled_height + bottom_border;
+				/* keep the vblank width constant */
 				adjusted_mode->crtc_vblank_end =
-				 adjusted_mode->crtc_vtotal - top_border - 1;
+					adjusted_mode->crtc_vblank_start +
+							vblank_width;
+				/*
+				 * get the vsync start pos relative to
+				 * vblank start
+				 */
+				vsync_pos = (vblank_width - vsync_width) / 2;
 				adjusted_mode->crtc_vsync_start =
-					adjusted_mode->crtc_vblank_start;
+					adjusted_mode->crtc_vblank_start +
+							vsync_pos;
+				/* keep the vsync width constant */
 				adjusted_mode->crtc_vsync_end =
-					adjusted_mode->crtc_vblank_end;
+					adjusted_mode->crtc_vsync_start +
+							vsync_width;
 				border = 1;
 			} else {
 			/* Aspects match, Let hw scale both directions */
-- 
cgit v1.2.3-59-g8ed1b


From cfd43c025ddef0b1c723bb9811d2bde52b285710 Mon Sep 17 00:00:00 2001
From: Krzysztof Halasa <khc@pm.waw.pl>
Date: Sat, 20 Jun 2009 00:31:28 +0200
Subject: drm/i915: Fix size_t handling in off-by-default debug printfs

Signed-off-by: Krzysztof Halasa <khc@pm.waw.pl>
Signed-off-by: Eric Anholt <eric@anholt.net>
---
 drivers/gpu/drm/i915/i915_gem.c       | 6 +++---
 drivers/gpu/drm/i915/i915_gem_debug.c | 2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index fd2b8bdffe3f..8660b2144b27 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1006,7 +1006,7 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 
 	mutex_lock(&dev->struct_mutex);
 #if WATCH_BUF
-	DRM_INFO("set_domain_ioctl %p(%d), %08x %08x\n",
+	DRM_INFO("set_domain_ioctl %p(%zd), %08x %08x\n",
 		 obj, obj->size, read_domains, write_domain);
 #endif
 	if (read_domains & I915_GEM_DOMAIN_GTT) {
@@ -1050,7 +1050,7 @@ i915_gem_sw_finish_ioctl(struct drm_device *dev, void *data,
 	}
 
 #if WATCH_BUF
-	DRM_INFO("%s: sw_finish %d (%p %d)\n",
+	DRM_INFO("%s: sw_finish %d (%p %zd)\n",
 		 __func__, args->handle, obj, obj->size);
 #endif
 	obj_priv = obj->driver_private;
@@ -2423,7 +2423,7 @@ i915_gem_object_bind_to_gtt(struct drm_gem_object *obj, unsigned alignment)
 	}
 
 #if WATCH_BUF
-	DRM_INFO("Binding object of size %d at 0x%08x\n",
+	DRM_INFO("Binding object of size %zd at 0x%08x\n",
 		 obj->size, obj_priv->gtt_offset);
 #endif
 	ret = i915_gem_object_get_pages(obj);
diff --git a/drivers/gpu/drm/i915/i915_gem_debug.c b/drivers/gpu/drm/i915/i915_gem_debug.c
index 8d0b943e2c5a..f94b5985f734 100644
--- a/drivers/gpu/drm/i915/i915_gem_debug.c
+++ b/drivers/gpu/drm/i915/i915_gem_debug.c
@@ -143,7 +143,7 @@ i915_gem_object_check_coherency(struct drm_gem_object *obj, int handle)
 	uint32_t *backing_map = NULL;
 	int bad_count = 0;
 
-	DRM_INFO("%s: checking coherency of object %p@0x%08x (%d, %dkb):\n",
+	DRM_INFO("%s: checking coherency of object %p@0x%08x (%d, %zdkb):\n",
 		 __func__, obj, obj_priv->gtt_offset, handle,
 		 obj->size / 1024);
 
-- 
cgit v1.2.3-59-g8ed1b


From 921809a5831821eaf86e799c4b3d7c666ee352b1 Mon Sep 17 00:00:00 2001
From: Krzysztof Halasa <khc@pm.waw.pl>
Date: Fri, 19 Jun 2009 22:35:09 +0200
Subject: drm/i915: Catch up to obj_priv->page_list rename in disabled debug
 code.

Signed-off-by:  Krzysztof Halasa <khc@pm.waw.pl>
Signed-off-by: Eric Anholt <eric@anholt.net>
---
 drivers/gpu/drm/i915/i915_gem_debug.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gem_debug.c b/drivers/gpu/drm/i915/i915_gem_debug.c
index f94b5985f734..e602614bd3f8 100644
--- a/drivers/gpu/drm/i915/i915_gem_debug.c
+++ b/drivers/gpu/drm/i915/i915_gem_debug.c
@@ -87,7 +87,7 @@ i915_gem_dump_object(struct drm_gem_object *obj, int len,
 			chunk_len = page_len - chunk;
 			if (chunk_len > 128)
 				chunk_len = 128;
-			i915_gem_dump_page(obj_priv->page_list[page],
+			i915_gem_dump_page(obj_priv->pages[page],
 					   chunk, chunk + chunk_len,
 					   obj_priv->gtt_offset +
 					   page * PAGE_SIZE,
@@ -157,7 +157,7 @@ i915_gem_object_check_coherency(struct drm_gem_object *obj, int handle)
 	for (page = 0; page < obj->size / PAGE_SIZE; page++) {
 		int i;
 
-		backing_map = kmap_atomic(obj_priv->page_list[page], KM_USER0);
+		backing_map = kmap_atomic(obj_priv->pages[page], KM_USER0);
 
 		if (backing_map == NULL) {
 			DRM_ERROR("failed to map backing page\n");
-- 
cgit v1.2.3-59-g8ed1b


From 8ed9a5bc9c9425ef93a1b03b418300a5e18b2361 Mon Sep 17 00:00:00 2001
From: "ling.ma@intel.com" <ling.ma@intel.com>
Date: Mon, 22 Jun 2009 22:08:35 +0800
Subject: drm/i915: set TV detection mode when tv is already connected

We used load_detect_temp flag to determine whether to set tv to the test
mode. However if the TV already has a mode set, we still need to set the
test mode to determine connection.  This results in blinking, but there is
no other reliable way to determine TV connection.

freedesktop.org bug #22035

Signed-off-by: Ma Ling <ling.ma@intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
---
 drivers/gpu/drm/i915/intel_tv.c | 53 +++++++++++++++++++----------------------
 1 file changed, 25 insertions(+), 28 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index ea68992e4416..a43c98e3f077 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -1383,34 +1383,31 @@ intel_tv_detect_type (struct drm_crtc *crtc, struct intel_output *intel_output)
 	/*
 	 * Detect TV by polling)
 	 */
-	if (intel_output->load_detect_temp) {
-		/* TV not currently running, prod it with destructive detect */
-		save_tv_dac = tv_dac;
-		tv_ctl = I915_READ(TV_CTL);
-		save_tv_ctl = tv_ctl;
-		tv_ctl &= ~TV_ENC_ENABLE;
-		tv_ctl &= ~TV_TEST_MODE_MASK;
-		tv_ctl |= TV_TEST_MODE_MONITOR_DETECT;
-		tv_dac &= ~TVDAC_SENSE_MASK;
-		tv_dac &= ~DAC_A_MASK;
-		tv_dac &= ~DAC_B_MASK;
-		tv_dac &= ~DAC_C_MASK;
-		tv_dac |= (TVDAC_STATE_CHG_EN |
-			   TVDAC_A_SENSE_CTL |
-			   TVDAC_B_SENSE_CTL |
-			   TVDAC_C_SENSE_CTL |
-			   DAC_CTL_OVERRIDE |
-			   DAC_A_0_7_V |
-			   DAC_B_0_7_V |
-			   DAC_C_0_7_V);
-		I915_WRITE(TV_CTL, tv_ctl);
-		I915_WRITE(TV_DAC, tv_dac);
-		intel_wait_for_vblank(dev);
-		tv_dac = I915_READ(TV_DAC);
-		I915_WRITE(TV_DAC, save_tv_dac);
-		I915_WRITE(TV_CTL, save_tv_ctl);
-		intel_wait_for_vblank(dev);
-	}
+	save_tv_dac = tv_dac;
+	tv_ctl = I915_READ(TV_CTL);
+	save_tv_ctl = tv_ctl;
+	tv_ctl &= ~TV_ENC_ENABLE;
+	tv_ctl &= ~TV_TEST_MODE_MASK;
+	tv_ctl |= TV_TEST_MODE_MONITOR_DETECT;
+	tv_dac &= ~TVDAC_SENSE_MASK;
+	tv_dac &= ~DAC_A_MASK;
+	tv_dac &= ~DAC_B_MASK;
+	tv_dac &= ~DAC_C_MASK;
+	tv_dac |= (TVDAC_STATE_CHG_EN |
+		   TVDAC_A_SENSE_CTL |
+		   TVDAC_B_SENSE_CTL |
+		   TVDAC_C_SENSE_CTL |
+		   DAC_CTL_OVERRIDE |
+		   DAC_A_0_7_V |
+		   DAC_B_0_7_V |
+		   DAC_C_0_7_V);
+	I915_WRITE(TV_CTL, tv_ctl);
+	I915_WRITE(TV_DAC, tv_dac);
+	intel_wait_for_vblank(dev);
+	tv_dac = I915_READ(TV_DAC);
+	I915_WRITE(TV_DAC, save_tv_dac);
+	I915_WRITE(TV_CTL, save_tv_ctl);
+	intel_wait_for_vblank(dev);
 	/*
 	 *  A B C
 	 *  0 1 1 Composite
-- 
cgit v1.2.3-59-g8ed1b


From 1b16de0b070dc6fa29b7a99980eabe3325ee5983 Mon Sep 17 00:00:00 2001
From: Jesse Barnes <jbarnes@virtuousgeek.org>
Date: Mon, 22 Jun 2009 11:30:30 -0700
Subject: drm/i915: fix LFP data fetch

Apparently the proper way to do this is to use the LFP data pointer
block to figure out the LFP data block entry size, then use that plus
the panel index to calculate an offset into the LFP data block array.

Similar fix has already been pushed to the 2D driver to fix fdo bug
applied to the VBIOS reader, and things look sane).

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Eric Anholt <eric@anholt.net>
---
 drivers/gpu/drm/i915/intel_bios.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index cdd126d068a7..716409a57244 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -99,9 +99,11 @@ parse_lfp_panel_data(struct drm_i915_private *dev_priv,
 {
 	struct bdb_lvds_options *lvds_options;
 	struct bdb_lvds_lfp_data *lvds_lfp_data;
+	struct bdb_lvds_lfp_data_ptrs *lvds_lfp_data_ptrs;
 	struct bdb_lvds_lfp_data_entry *entry;
 	struct lvds_dvo_timing *dvo_timing;
 	struct drm_display_mode *panel_fixed_mode;
+	int lfp_data_size;
 
 	/* Defaults if we can't find VBT info */
 	dev_priv->lvds_dither = 0;
@@ -119,9 +121,17 @@ parse_lfp_panel_data(struct drm_i915_private *dev_priv,
 	if (!lvds_lfp_data)
 		return;
 
+	lvds_lfp_data_ptrs = find_section(bdb, BDB_LVDS_LFP_DATA_PTRS);
+	if (!lvds_lfp_data_ptrs)
+		return;
+
 	dev_priv->lvds_vbt = 1;
 
-	entry = &lvds_lfp_data->data[lvds_options->panel_type];
+	lfp_data_size = lvds_lfp_data_ptrs->ptr[1].dvo_timing_offset -
+		lvds_lfp_data_ptrs->ptr[0].dvo_timing_offset;
+	entry = (struct bdb_lvds_lfp_data_entry *)
+		((uint8_t *)lvds_lfp_data->data + (lfp_data_size *
+						   lvds_options->panel_type));
 	dvo_timing = &entry->dvo_timing;
 
 	panel_fixed_mode = kzalloc(sizeof(*panel_fixed_mode), GFP_KERNEL);
-- 
cgit v1.2.3-59-g8ed1b


From 56d21b07d44e0a33ab846f4f08e9e33bd87e5d4b Mon Sep 17 00:00:00 2001
From: Zhenyu Wang <zhenyuw@linux.intel.com>
Date: Wed, 17 Jun 2009 09:43:25 +0800
Subject: drm/i915: Fix HDMI regression introduced in new chipset support

Remove wrongly added NULL_PACKETS_DURING_VSYNC setting for HDMI.

Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Signed-off-by: Eric Anholt <eric@anholt.net>
---
 drivers/gpu/drm/i915/intel_hdmi.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 3955476eb64f..9e30daae37dc 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -57,8 +57,7 @@ static void intel_hdmi_mode_set(struct drm_encoder *encoder,
 	sdvox = SDVO_ENCODING_HDMI |
 		SDVO_BORDER_ENABLE |
 		SDVO_VSYNC_ACTIVE_HIGH |
-		SDVO_HSYNC_ACTIVE_HIGH |
-		SDVO_NULL_PACKETS_DURING_VSYNC;
+		SDVO_HSYNC_ACTIVE_HIGH;
 
 	if (hdmi_priv->has_hdmi_sink)
 		sdvox |= SDVO_AUDIO_ENABLE;
-- 
cgit v1.2.3-59-g8ed1b


From 31950eb66ff47c946fd9c65c2f8c94b6b7ba13fc Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Mon, 22 Jun 2009 21:18:12 -0700
Subject: mm/init: cpu_hotplug_init() must be initialized before SLAB

SLAB uses get/put_online_cpus() which use a mutex which is itself only
initialized when cpu_hotplug_init() is called.  Currently we hang suring
boot in SLAB due to doing that too late.

Reported by James Bottomley and Sachin Sant (and possibly others).
Debugged by Benjamin Herrenschmidt.

This just removes the dynamic initialization of the data structures, and
replaces it with a static one, avoiding this dependency entirely, and
removing one unnecessary special initcall.

Tested-by: Sachin Sant <sachinp@in.ibm.com>
Tested-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Tested-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cpu.h |  5 -----
 init/main.c         |  1 -
 kernel/cpu.c        | 13 +++++--------
 3 files changed, 5 insertions(+), 14 deletions(-)

diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index 2643d848df90..4d668e05d458 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -69,7 +69,6 @@ static inline void unregister_cpu_notifier(struct notifier_block *nb)
 
 int cpu_up(unsigned int cpu);
 void notify_cpu_starting(unsigned int cpu);
-extern void cpu_hotplug_init(void);
 extern void cpu_maps_update_begin(void);
 extern void cpu_maps_update_done(void);
 
@@ -84,10 +83,6 @@ static inline void unregister_cpu_notifier(struct notifier_block *nb)
 {
 }
 
-static inline void cpu_hotplug_init(void)
-{
-}
-
 static inline void cpu_maps_update_begin(void)
 {
 }
diff --git a/init/main.c b/init/main.c
index 09131ec090c1..4870dfeb9ee5 100644
--- a/init/main.c
+++ b/init/main.c
@@ -678,7 +678,6 @@ asmlinkage void __init start_kernel(void)
 #endif
 	page_cgroup_init();
 	enable_debug_pagealloc();
-	cpu_hotplug_init();
 	kmemtrace_init();
 	kmemleak_init();
 	debug_objects_mem_init();
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 395b6974dc8d..8ce10043e4ac 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -34,14 +34,11 @@ static struct {
 	 * an ongoing cpu hotplug operation.
 	 */
 	int refcount;
-} cpu_hotplug;
-
-void __init cpu_hotplug_init(void)
-{
-	cpu_hotplug.active_writer = NULL;
-	mutex_init(&cpu_hotplug.lock);
-	cpu_hotplug.refcount = 0;
-}
+} cpu_hotplug = {
+	.active_writer = NULL,
+	.lock = __MUTEX_INITIALIZER(cpu_hotplug.lock),
+	.refcount = 0,
+};
 
 #ifdef CONFIG_HOTPLUG_CPU
 
-- 
cgit v1.2.3-59-g8ed1b


From 616511d039af402670de8500d0e24495113a9cab Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Mon, 22 Jun 2009 15:09:13 -0400
Subject: VFS: Uninline the function put_mnt_ns()

In order to allow modules to use it without having to export vfsmount_lock.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/namespace.c                | 8 ++++++--
 include/linux/mnt_namespace.h | 9 +--------
 2 files changed, 7 insertions(+), 10 deletions(-)

diff --git a/fs/namespace.c b/fs/namespace.c
index 2dd333b0fe7f..6645846f2056 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -2246,10 +2246,14 @@ void __init mnt_init(void)
 	init_mount_tree();
 }
 
-void __put_mnt_ns(struct mnt_namespace *ns)
+void put_mnt_ns(struct mnt_namespace *ns)
 {
-	struct vfsmount *root = ns->root;
+	struct vfsmount *root;
 	LIST_HEAD(umount_list);
+
+	if (!atomic_dec_and_lock(&ns->count, &vfsmount_lock))
+		return;
+	root = ns->root;
 	ns->root = NULL;
 	spin_unlock(&vfsmount_lock);
 	down_write(&namespace_sem);
diff --git a/include/linux/mnt_namespace.h b/include/linux/mnt_namespace.h
index 3a059298cc19..299d11af5f79 100644
--- a/include/linux/mnt_namespace.h
+++ b/include/linux/mnt_namespace.h
@@ -26,14 +26,7 @@ struct fs_struct;
 
 extern struct mnt_namespace *copy_mnt_ns(unsigned long, struct mnt_namespace *,
 		struct fs_struct *);
-extern void __put_mnt_ns(struct mnt_namespace *ns);
-
-static inline void put_mnt_ns(struct mnt_namespace *ns)
-{
-	if (atomic_dec_and_lock(&ns->count, &vfsmount_lock))
-		/* releases vfsmount_lock */
-		__put_mnt_ns(ns);
-}
+extern void put_mnt_ns(struct mnt_namespace *ns);
 
 static inline void exit_mnt_ns(struct task_struct *p)
 {
-- 
cgit v1.2.3-59-g8ed1b


From cf8d2c11cb77f129675478792122f50827e5b0ae Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Mon, 22 Jun 2009 15:09:13 -0400
Subject: VFS: Add VFS helper functions for setting up private namespaces

The purpose of this patch is to improve the remote mount path lookup
support for distributed filesystems such as the NFSv4 client.

When given a mount command of the form "mount server:/foo/bar /mnt", the
NFSv4 client is required to look up the filehandle for "server:/", and
then look up each component of the remote mount path "foo/bar" in order
to find the directory that is actually going to be mounted on /mnt.
Following that remote mount path may involve following symlinks,
crossing server-side mount points and even following referrals to
filesystem volumes on other servers.

Since the standard VFS path lookup code already supports walking paths
that contain all these features (using in-kernel automounts for
following referrals) we would like to be able to reuse that rather than
duplicate the full path traversal functionality in the NFSv4 client code.

This patch therefore defines a VFS helper function create_mnt_ns(), that
sets up a temporary filesystem namespace and attaches a root filesystem to
it. It exports the create_mnt_ns() and put_mnt_ns() function for use by
filesystem modules.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/namespace.c                | 45 +++++++++++++++++++++++++++++++++++--------
 include/linux/mnt_namespace.h |  1 +
 2 files changed, 38 insertions(+), 8 deletions(-)

diff --git a/fs/namespace.c b/fs/namespace.c
index 6645846f2056..a7bea8c8bd46 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1937,6 +1937,21 @@ dput_out:
 	return retval;
 }
 
+static struct mnt_namespace *alloc_mnt_ns(void)
+{
+	struct mnt_namespace *new_ns;
+
+	new_ns = kmalloc(sizeof(struct mnt_namespace), GFP_KERNEL);
+	if (!new_ns)
+		return ERR_PTR(-ENOMEM);
+	atomic_set(&new_ns->count, 1);
+	new_ns->root = NULL;
+	INIT_LIST_HEAD(&new_ns->list);
+	init_waitqueue_head(&new_ns->poll);
+	new_ns->event = 0;
+	return new_ns;
+}
+
 /*
  * Allocate a new namespace structure and populate it with contents
  * copied from the namespace of the passed in task structure.
@@ -1948,14 +1963,9 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns,
 	struct vfsmount *rootmnt = NULL, *pwdmnt = NULL;
 	struct vfsmount *p, *q;
 
-	new_ns = kmalloc(sizeof(struct mnt_namespace), GFP_KERNEL);
-	if (!new_ns)
-		return ERR_PTR(-ENOMEM);
-
-	atomic_set(&new_ns->count, 1);
-	INIT_LIST_HEAD(&new_ns->list);
-	init_waitqueue_head(&new_ns->poll);
-	new_ns->event = 0;
+	new_ns = alloc_mnt_ns();
+	if (IS_ERR(new_ns))
+		return new_ns;
 
 	down_write(&namespace_sem);
 	/* First pass: copy the tree topology */
@@ -2019,6 +2029,24 @@ struct mnt_namespace *copy_mnt_ns(unsigned long flags, struct mnt_namespace *ns,
 	return new_ns;
 }
 
+/**
+ * create_mnt_ns - creates a private namespace and adds a root filesystem
+ * @mnt: pointer to the new root filesystem mountpoint
+ */
+struct mnt_namespace *create_mnt_ns(struct vfsmount *mnt)
+{
+	struct mnt_namespace *new_ns;
+
+	new_ns = alloc_mnt_ns();
+	if (!IS_ERR(new_ns)) {
+		mnt->mnt_ns = new_ns;
+		new_ns->root = mnt;
+		list_add(&new_ns->list, &new_ns->root->mnt_list);
+	}
+	return new_ns;
+}
+EXPORT_SYMBOL(create_mnt_ns);
+
 SYSCALL_DEFINE5(mount, char __user *, dev_name, char __user *, dir_name,
 		char __user *, type, unsigned long, flags, void __user *, data)
 {
@@ -2264,3 +2292,4 @@ void put_mnt_ns(struct mnt_namespace *ns)
 	release_mounts(&umount_list);
 	kfree(ns);
 }
+EXPORT_SYMBOL(put_mnt_ns);
diff --git a/include/linux/mnt_namespace.h b/include/linux/mnt_namespace.h
index 299d11af5f79..3beb2592b03f 100644
--- a/include/linux/mnt_namespace.h
+++ b/include/linux/mnt_namespace.h
@@ -24,6 +24,7 @@ struct proc_mounts {
 
 struct fs_struct;
 
+extern struct mnt_namespace *create_mnt_ns(struct vfsmount *mnt);
 extern struct mnt_namespace *copy_mnt_ns(unsigned long, struct mnt_namespace *,
 		struct fs_struct *);
 extern void put_mnt_ns(struct mnt_namespace *ns);
-- 
cgit v1.2.3-59-g8ed1b


From c02d7adf8c5429727a98bad1d039bccad4c61c50 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Mon, 22 Jun 2009 15:09:14 -0400
Subject: NFSv4: Replace nfs4_path_walk() with VFS path lookup in a private
 namespace

As noted in the previous patch, the NFSv4 client mount code currently
has several limitations. If the mount path contains symlinks, or
referrals, or even if it just contains a '..', then the client code in
nfs4_path_walk() will fail with an error.

This patch replaces the nfs4_path_walk()-based lookup with a helper
function that sets up a private namespace to represent the namespace on the
server, then uses the ordinary VFS and NFS path lookup code to walk down the
mount path in that namespace.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/nfs/super.c | 178 ++++++++++++++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 157 insertions(+), 21 deletions(-)

diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 3d460527daab..8e0673a0d6aa 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -42,6 +42,8 @@
 #include <linux/smp_lock.h>
 #include <linux/seq_file.h>
 #include <linux/mount.h>
+#include <linux/mnt_namespace.h>
+#include <linux/namei.h>
 #include <linux/nfs_idmap.h>
 #include <linux/vfs.h>
 #include <linux/inet.h>
@@ -272,10 +274,14 @@ static const struct super_operations nfs_sops = {
 #ifdef CONFIG_NFS_V4
 static int nfs4_get_sb(struct file_system_type *fs_type,
 	int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
+static int nfs4_remote_get_sb(struct file_system_type *fs_type,
+	int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
 static int nfs4_xdev_get_sb(struct file_system_type *fs_type,
 	int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
 static int nfs4_referral_get_sb(struct file_system_type *fs_type,
 	int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
+static int nfs4_remote_referral_get_sb(struct file_system_type *fs_type,
+	int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt);
 static void nfs4_kill_super(struct super_block *sb);
 
 static struct file_system_type nfs4_fs_type = {
@@ -286,6 +292,14 @@ static struct file_system_type nfs4_fs_type = {
 	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
 
+static struct file_system_type nfs4_remote_fs_type = {
+	.owner		= THIS_MODULE,
+	.name		= "nfs4",
+	.get_sb		= nfs4_remote_get_sb,
+	.kill_sb	= nfs4_kill_super,
+	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
+};
+
 struct file_system_type nfs4_xdev_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "nfs4",
@@ -294,6 +308,14 @@ struct file_system_type nfs4_xdev_fs_type = {
 	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
 };
 
+static struct file_system_type nfs4_remote_referral_fs_type = {
+	.owner		= THIS_MODULE,
+	.name		= "nfs4",
+	.get_sb		= nfs4_remote_referral_get_sb,
+	.kill_sb	= nfs4_kill_super,
+	.fs_flags	= FS_RENAME_DOES_D_MOVE|FS_REVAL_DOT|FS_BINARY_MOUNTDATA,
+};
+
 struct file_system_type nfs4_referral_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "nfs4",
@@ -2433,12 +2455,12 @@ out_no_client_address:
 }
 
 /*
- * Get the superblock for an NFS4 mountpoint
+ * Get the superblock for the NFS4 root partition
  */
-static int nfs4_get_sb(struct file_system_type *fs_type,
+static int nfs4_remote_get_sb(struct file_system_type *fs_type,
 	int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt)
 {
-	struct nfs_parsed_mount_data *data;
+	struct nfs_parsed_mount_data *data = raw_data;
 	struct super_block *s;
 	struct nfs_server *server;
 	struct nfs_fh *mntfh;
@@ -2449,18 +2471,12 @@ static int nfs4_get_sb(struct file_system_type *fs_type,
 	};
 	int error = -ENOMEM;
 
-	data = kzalloc(sizeof(*data), GFP_KERNEL);
 	mntfh = kzalloc(sizeof(*mntfh), GFP_KERNEL);
 	if (data == NULL || mntfh == NULL)
 		goto out_free_fh;
 
 	security_init_mnt_opts(&data->lsm_opts);
 
-	/* Validate the mount data */
-	error = nfs4_validate_mount_data(raw_data, data, dev_name);
-	if (error < 0)
-		goto out;
-
 	/* Get a volume representation */
 	server = nfs4_create_server(data, mntfh);
 	if (IS_ERR(server)) {
@@ -2473,7 +2489,7 @@ static int nfs4_get_sb(struct file_system_type *fs_type,
 		compare_super = NULL;
 
 	/* Get a superblock - note that we may end up sharing one that already exists */
-	s = sget(fs_type, compare_super, nfs_set_super, &sb_mntdata);
+	s = sget(&nfs4_fs_type, compare_super, nfs_set_super, &sb_mntdata);
 	if (IS_ERR(s)) {
 		error = PTR_ERR(s);
 		goto out_free;
@@ -2510,14 +2526,9 @@ static int nfs4_get_sb(struct file_system_type *fs_type,
 	error = 0;
 
 out:
-	kfree(data->client_address);
-	kfree(data->nfs_server.export_path);
-	kfree(data->nfs_server.hostname);
-	kfree(data->fscache_uniq);
 	security_free_mnt_opts(&data->lsm_opts);
 out_free_fh:
 	kfree(mntfh);
-	kfree(data);
 	return error;
 
 out_free:
@@ -2531,6 +2542,102 @@ error_splat_super:
 	goto out;
 }
 
+static struct vfsmount *nfs_do_root_mount(struct file_system_type *fs_type,
+		int flags, void *data, const char *hostname)
+{
+	struct vfsmount *root_mnt;
+	char *root_devname;
+	size_t len;
+
+	len = strlen(hostname) + 3;
+	root_devname = kmalloc(len, GFP_KERNEL);
+	if (root_devname == NULL)
+		return ERR_PTR(-ENOMEM);
+	snprintf(root_devname, len, "%s:/", hostname);
+	root_mnt = vfs_kern_mount(fs_type, flags, root_devname, data);
+	kfree(root_devname);
+	return root_mnt;
+}
+
+static int nfs_follow_remote_path(struct vfsmount *root_mnt,
+		const char *export_path, struct vfsmount *mnt_target)
+{
+	struct mnt_namespace *ns_private;
+	struct nameidata nd;
+	struct super_block *s;
+	int ret;
+
+	ns_private = create_mnt_ns(root_mnt);
+	ret = PTR_ERR(ns_private);
+	if (IS_ERR(ns_private))
+		goto out_mntput;
+
+	ret = vfs_path_lookup(root_mnt->mnt_root, root_mnt,
+			export_path, LOOKUP_FOLLOW, &nd);
+
+	put_mnt_ns(ns_private);
+
+	if (ret != 0)
+		goto out_err;
+
+	s = nd.path.mnt->mnt_sb;
+	atomic_inc(&s->s_active);
+	mnt_target->mnt_sb = s;
+	mnt_target->mnt_root = dget(nd.path.dentry);
+
+	path_put(&nd.path);
+	down_write(&s->s_umount);
+	return 0;
+out_mntput:
+	mntput(root_mnt);
+out_err:
+	return ret;
+}
+
+/*
+ * Get the superblock for an NFS4 mountpoint
+ */
+static int nfs4_get_sb(struct file_system_type *fs_type,
+	int flags, const char *dev_name, void *raw_data, struct vfsmount *mnt)
+{
+	struct nfs_parsed_mount_data *data;
+	char *export_path;
+	struct vfsmount *root_mnt;
+	int error = -ENOMEM;
+
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
+	if (data == NULL)
+		goto out_free_data;
+
+	/* Validate the mount data */
+	error = nfs4_validate_mount_data(raw_data, data, dev_name);
+	if (error < 0)
+		goto out;
+
+	export_path = data->nfs_server.export_path;
+	data->nfs_server.export_path = "/";
+	root_mnt = nfs_do_root_mount(&nfs4_remote_fs_type, flags, data,
+			data->nfs_server.hostname);
+	data->nfs_server.export_path = export_path;
+
+	error = PTR_ERR(root_mnt);
+	if (IS_ERR(root_mnt))
+		goto out;
+
+	error = nfs_follow_remote_path(root_mnt, export_path, mnt);
+
+out:
+	kfree(data->client_address);
+	kfree(data->nfs_server.export_path);
+	kfree(data->nfs_server.hostname);
+	kfree(data->fscache_uniq);
+out_free_data:
+	kfree(data);
+	dprintk("<-- nfs4_get_sb() = %d%s\n", error,
+			error != 0 ? " [error]" : "");
+	return error;
+}
+
 static void nfs4_kill_super(struct super_block *sb)
 {
 	struct nfs_server *server = NFS_SB(sb);
@@ -2627,12 +2734,9 @@ error_splat_super:
 	return error;
 }
 
-/*
- * Create an NFS4 server record on referral traversal
- */
-static int nfs4_referral_get_sb(struct file_system_type *fs_type, int flags,
-				const char *dev_name, void *raw_data,
-				struct vfsmount *mnt)
+static int nfs4_remote_referral_get_sb(struct file_system_type *fs_type,
+		int flags, const char *dev_name, void *raw_data,
+		struct vfsmount *mnt)
 {
 	struct nfs_clone_mount *data = raw_data;
 	struct super_block *s;
@@ -2711,4 +2815,36 @@ error_splat_super:
 	return error;
 }
 
+/*
+ * Create an NFS4 server record on referral traversal
+ */
+static int nfs4_referral_get_sb(struct file_system_type *fs_type,
+		int flags, const char *dev_name, void *raw_data,
+		struct vfsmount *mnt)
+{
+	struct nfs_clone_mount *data = raw_data;
+	char *export_path;
+	struct vfsmount *root_mnt;
+	int error;
+
+	dprintk("--> nfs4_referral_get_sb()\n");
+
+	export_path = data->mnt_path;
+	data->mnt_path = "/";
+
+	root_mnt = nfs_do_root_mount(&nfs4_remote_referral_fs_type,
+			flags, data, data->hostname);
+	data->mnt_path = export_path;
+
+	error = PTR_ERR(root_mnt);
+	if (IS_ERR(root_mnt))
+		goto out;
+
+	error = nfs_follow_remote_path(root_mnt, export_path, mnt);
+out:
+	dprintk("<-- nfs4_referral_get_sb() = %d%s\n", error,
+			error != 0 ? " [error]" : "");
+	return error;
+}
+
 #endif /* CONFIG_NFS_V4 */
-- 
cgit v1.2.3-59-g8ed1b


From 0b75b35c7cad33e7613f5adf28fa10fe8b09b1c3 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Mon, 22 Jun 2009 15:09:14 -0400
Subject: NFS: Fix nfs_path() to always return a '/' at the beginning of the
 path

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/nfs/namespace.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index f01caec84463..40c766782891 100644
--- a/fs/nfs/namespace.c
+++ b/fs/nfs/namespace.c
@@ -65,6 +65,11 @@ char *nfs_path(const char *base,
 		dentry = dentry->d_parent;
 	}
 	spin_unlock(&dcache_lock);
+	if (*end != '/') {
+		if (--buflen < 0)
+			goto Elong;
+		*--end = '/';
+	}
 	namelen = strlen(base);
 	/* Strip off excess slashes in base string */
 	while (namelen > 0 && base[namelen - 1] == '/')
-- 
cgit v1.2.3-59-g8ed1b


From b88f8a546f5dba213938fdfc11e66bc5c2421623 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Mon, 22 Jun 2009 15:09:14 -0400
Subject: NFS: Correct the NFS mount path when following a referral

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/nfs/super.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 8e0673a0d6aa..0b4cbdc60abd 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -2559,6 +2559,27 @@ static struct vfsmount *nfs_do_root_mount(struct file_system_type *fs_type,
 	return root_mnt;
 }
 
+static void nfs_fix_devname(const struct path *path, struct vfsmount *mnt)
+{
+	char *page = (char *) __get_free_page(GFP_KERNEL);
+	char *devname, *tmp;
+
+	if (page == NULL)
+		return;
+	devname = nfs_path(path->mnt->mnt_devname,
+			path->mnt->mnt_root, path->dentry,
+			page, PAGE_SIZE);
+	if (devname == NULL)
+		goto out_freepage;
+	tmp = kstrdup(devname, GFP_KERNEL);
+	if (tmp == NULL)
+		goto out_freepage;
+	kfree(mnt->mnt_devname);
+	mnt->mnt_devname = tmp;
+out_freepage:
+	free_page((unsigned long)page);
+}
+
 static int nfs_follow_remote_path(struct vfsmount *root_mnt,
 		const char *export_path, struct vfsmount *mnt_target)
 {
@@ -2585,6 +2606,9 @@ static int nfs_follow_remote_path(struct vfsmount *root_mnt,
 	mnt_target->mnt_sb = s;
 	mnt_target->mnt_root = dget(nd.path.dentry);
 
+	/* Correct the device pathname */
+	nfs_fix_devname(&nd.path, mnt_target);
+
 	path_put(&nd.path);
 	down_write(&s->s_umount);
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 0cf89dcdbc53f2b43e4ce7419b6ff47f4309c2eb Mon Sep 17 00:00:00 2001
From: Hannes Hering <hering2@de.ibm.com>
Date: Mon, 22 Jun 2009 22:18:51 -0700
Subject: IB/ehca: Tolerate dynamic memory operations before driver load

Implement toleration of dynamic memory operations and 16 GB gigantic
pages, where "toleration" means that the driver can cope with dynamic
memory operations that happen before the driver is loaded.  While the
ehca driver is loaded, dynamic memory operations are still prohibited
by returning NOTIFY_BAD from the memory notifier.

On module load the driver walks through available system memory,
checks for available memory ranges and then registers the kernel
internal memory region accordingly.  The translation of address ranges
is implemented via a 3-level busmap.

Signed-off-by: Hannes Hering <hering2@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/ehca/ehca_main.c |  18 +-
 drivers/infiniband/hw/ehca/ehca_mrmw.c | 508 ++++++++++++++++++++++++++++++++-
 drivers/infiniband/hw/ehca/ehca_mrmw.h |  13 +-
 3 files changed, 522 insertions(+), 17 deletions(-)

diff --git a/drivers/infiniband/hw/ehca/ehca_main.c b/drivers/infiniband/hw/ehca/ehca_main.c
index ce4e6eff4792..14a18b7be245 100644
--- a/drivers/infiniband/hw/ehca/ehca_main.c
+++ b/drivers/infiniband/hw/ehca/ehca_main.c
@@ -506,6 +506,7 @@ static int ehca_init_device(struct ehca_shca *shca)
 	shca->ib_device.detach_mcast	    = ehca_detach_mcast;
 	shca->ib_device.process_mad	    = ehca_process_mad;
 	shca->ib_device.mmap		    = ehca_mmap;
+	shca->ib_device.dma_ops		    = &ehca_dma_mapping_ops;
 
 	if (EHCA_BMASK_GET(HCA_CAP_SRQ, shca->hca_cap)) {
 		shca->ib_device.uverbs_cmd_mask |=
@@ -1028,17 +1029,23 @@ static int __init ehca_module_init(void)
 		goto module_init1;
 	}
 
+	ret = ehca_create_busmap();
+	if (ret) {
+		ehca_gen_err("Cannot create busmap.");
+		goto module_init2;
+	}
+
 	ret = ibmebus_register_driver(&ehca_driver);
 	if (ret) {
 		ehca_gen_err("Cannot register eHCA device driver");
 		ret = -EINVAL;
-		goto module_init2;
+		goto module_init3;
 	}
 
 	ret = register_memory_notifier(&ehca_mem_nb);
 	if (ret) {
 		ehca_gen_err("Failed registering memory add/remove notifier");
-		goto module_init3;
+		goto module_init4;
 	}
 
 	if (ehca_poll_all_eqs != 1) {
@@ -1053,9 +1060,12 @@ static int __init ehca_module_init(void)
 
 	return 0;
 
-module_init3:
+module_init4:
 	ibmebus_unregister_driver(&ehca_driver);
 
+module_init3:
+	ehca_destroy_busmap();
+
 module_init2:
 	ehca_destroy_slab_caches();
 
@@ -1073,6 +1083,8 @@ static void __exit ehca_module_exit(void)
 
 	unregister_memory_notifier(&ehca_mem_nb);
 
+	ehca_destroy_busmap();
+
 	ehca_destroy_slab_caches();
 
 	ehca_destroy_comp_pool();
diff --git a/drivers/infiniband/hw/ehca/ehca_mrmw.c b/drivers/infiniband/hw/ehca/ehca_mrmw.c
index 72f83f7df614..7663a2a9f130 100644
--- a/drivers/infiniband/hw/ehca/ehca_mrmw.c
+++ b/drivers/infiniband/hw/ehca/ehca_mrmw.c
@@ -53,6 +53,38 @@
 /* max number of rpages (per hcall register_rpages) */
 #define MAX_RPAGES 512
 
+/* DMEM toleration management */
+#define EHCA_SECTSHIFT        SECTION_SIZE_BITS
+#define EHCA_SECTSIZE          (1UL << EHCA_SECTSHIFT)
+#define EHCA_HUGEPAGESHIFT     34
+#define EHCA_HUGEPAGE_SIZE     (1UL << EHCA_HUGEPAGESHIFT)
+#define EHCA_HUGEPAGE_PFN_MASK ((EHCA_HUGEPAGE_SIZE - 1) >> PAGE_SHIFT)
+#define EHCA_INVAL_ADDR        0xFFFFFFFFFFFFFFFFULL
+#define EHCA_DIR_INDEX_SHIFT 13                   /* 8k Entries in 64k block */
+#define EHCA_TOP_INDEX_SHIFT (EHCA_DIR_INDEX_SHIFT * 2)
+#define EHCA_MAP_ENTRIES (1 << EHCA_DIR_INDEX_SHIFT)
+#define EHCA_TOP_MAP_SIZE (0x10000)               /* currently fixed map size */
+#define EHCA_DIR_MAP_SIZE (0x10000)
+#define EHCA_ENT_MAP_SIZE (0x10000)
+#define EHCA_INDEX_MASK (EHCA_MAP_ENTRIES - 1)
+
+static unsigned long ehca_mr_len;
+
+/*
+ * Memory map data structures
+ */
+struct ehca_dir_bmap {
+	u64 ent[EHCA_MAP_ENTRIES];
+};
+struct ehca_top_bmap {
+	struct ehca_dir_bmap *dir[EHCA_MAP_ENTRIES];
+};
+struct ehca_bmap {
+	struct ehca_top_bmap *top[EHCA_MAP_ENTRIES];
+};
+
+static struct ehca_bmap *ehca_bmap;
+
 static struct kmem_cache *mr_cache;
 static struct kmem_cache *mw_cache;
 
@@ -68,6 +100,8 @@ enum ehca_mr_pgsize {
 #define EHCA_MR_PGSHIFT1M  20
 #define EHCA_MR_PGSHIFT16M 24
 
+static u64 ehca_map_vaddr(void *caddr);
+
 static u32 ehca_encode_hwpage_size(u32 pgsize)
 {
 	int log = ilog2(pgsize);
@@ -135,7 +169,8 @@ struct ib_mr *ehca_get_dma_mr(struct ib_pd *pd, int mr_access_flags)
 			goto get_dma_mr_exit0;
 		}
 
-		ret = ehca_reg_maxmr(shca, e_maxmr, (u64 *)KERNELBASE,
+		ret = ehca_reg_maxmr(shca, e_maxmr,
+				     (void *)ehca_map_vaddr((void *)KERNELBASE),
 				     mr_access_flags, e_pd,
 				     &e_maxmr->ib.ib_mr.lkey,
 				     &e_maxmr->ib.ib_mr.rkey);
@@ -251,7 +286,7 @@ struct ib_mr *ehca_reg_phys_mr(struct ib_pd *pd,
 
 		ret = ehca_reg_mr(shca, e_mr, iova_start, size, mr_access_flags,
 				  e_pd, &pginfo, &e_mr->ib.ib_mr.lkey,
-				  &e_mr->ib.ib_mr.rkey);
+				  &e_mr->ib.ib_mr.rkey, EHCA_REG_MR);
 		if (ret) {
 			ib_mr = ERR_PTR(ret);
 			goto reg_phys_mr_exit1;
@@ -370,7 +405,7 @@ reg_user_mr_fallback:
 
 	ret = ehca_reg_mr(shca, e_mr, (u64 *)virt, length, mr_access_flags,
 			  e_pd, &pginfo, &e_mr->ib.ib_mr.lkey,
-			  &e_mr->ib.ib_mr.rkey);
+			  &e_mr->ib.ib_mr.rkey, EHCA_REG_MR);
 	if (ret == -EINVAL && pginfo.hwpage_size > PAGE_SIZE) {
 		ehca_warn(pd->device, "failed to register mr "
 			  "with hwpage_size=%llx", hwpage_size);
@@ -794,7 +829,7 @@ struct ib_fmr *ehca_alloc_fmr(struct ib_pd *pd,
 	ret = ehca_reg_mr(shca, e_fmr, NULL,
 			  fmr_attr->max_pages * (1 << fmr_attr->page_shift),
 			  mr_access_flags, e_pd, &pginfo,
-			  &tmp_lkey, &tmp_rkey);
+			  &tmp_lkey, &tmp_rkey, EHCA_REG_MR);
 	if (ret) {
 		ib_fmr = ERR_PTR(ret);
 		goto alloc_fmr_exit1;
@@ -983,6 +1018,10 @@ free_fmr_exit0:
 
 /*----------------------------------------------------------------------*/
 
+static int ehca_reg_bmap_mr_rpages(struct ehca_shca *shca,
+				   struct ehca_mr *e_mr,
+				   struct ehca_mr_pginfo *pginfo);
+
 int ehca_reg_mr(struct ehca_shca *shca,
 		struct ehca_mr *e_mr,
 		u64 *iova_start,
@@ -991,7 +1030,8 @@ int ehca_reg_mr(struct ehca_shca *shca,
 		struct ehca_pd *e_pd,
 		struct ehca_mr_pginfo *pginfo,
 		u32 *lkey, /*OUT*/
-		u32 *rkey) /*OUT*/
+		u32 *rkey, /*OUT*/
+		enum ehca_reg_type reg_type)
 {
 	int ret;
 	u64 h_ret;
@@ -1015,7 +1055,13 @@ int ehca_reg_mr(struct ehca_shca *shca,
 
 	e_mr->ipz_mr_handle = hipzout.handle;
 
-	ret = ehca_reg_mr_rpages(shca, e_mr, pginfo);
+	if (reg_type == EHCA_REG_BUSMAP_MR)
+		ret = ehca_reg_bmap_mr_rpages(shca, e_mr, pginfo);
+	else if (reg_type == EHCA_REG_MR)
+		ret = ehca_reg_mr_rpages(shca, e_mr, pginfo);
+	else
+		ret = -EINVAL;
+
 	if (ret)
 		goto ehca_reg_mr_exit1;
 
@@ -1316,7 +1362,7 @@ int ehca_rereg_mr(struct ehca_shca *shca,
 		e_mr->fmr_map_cnt = save_mr.fmr_map_cnt;
 
 		ret = ehca_reg_mr(shca, e_mr, iova_start, size, acl,
-				  e_pd, pginfo, lkey, rkey);
+				  e_pd, pginfo, lkey, rkey, EHCA_REG_MR);
 		if (ret) {
 			u32 offset = (u64)(&e_mr->flags) - (u64)e_mr;
 			memcpy(&e_mr->flags, &(save_mr.flags),
@@ -1409,7 +1455,7 @@ int ehca_unmap_one_fmr(struct ehca_shca *shca,
 	ret = ehca_reg_mr(shca, e_fmr, NULL,
 			  (e_fmr->fmr_max_pages * e_fmr->fmr_page_size),
 			  e_fmr->acl, e_pd, &pginfo, &tmp_lkey,
-			  &tmp_rkey);
+			  &tmp_rkey, EHCA_REG_MR);
 	if (ret) {
 		u32 offset = (u64)(&e_fmr->flags) - (u64)e_fmr;
 		memcpy(&e_fmr->flags, &(save_mr.flags),
@@ -1478,6 +1524,90 @@ ehca_reg_smr_exit0:
 } /* end ehca_reg_smr() */
 
 /*----------------------------------------------------------------------*/
+static inline void *ehca_calc_sectbase(int top, int dir, int idx)
+{
+	unsigned long ret = idx;
+	ret |= dir << EHCA_DIR_INDEX_SHIFT;
+	ret |= top << EHCA_TOP_INDEX_SHIFT;
+	return abs_to_virt(ret << SECTION_SIZE_BITS);
+}
+
+#define ehca_bmap_valid(entry) \
+	((u64)entry != (u64)EHCA_INVAL_ADDR)
+
+static u64 ehca_reg_mr_section(int top, int dir, int idx, u64 *kpage,
+			       struct ehca_shca *shca, struct ehca_mr *mr,
+			       struct ehca_mr_pginfo *pginfo)
+{
+	u64 h_ret = 0;
+	unsigned long page = 0;
+	u64 rpage = virt_to_abs(kpage);
+	int page_count;
+
+	void *sectbase = ehca_calc_sectbase(top, dir, idx);
+	if ((unsigned long)sectbase & (pginfo->hwpage_size - 1)) {
+		ehca_err(&shca->ib_device, "reg_mr_section will probably fail:"
+					   "hwpage_size does not fit to "
+					   "section start address");
+	}
+	page_count = EHCA_SECTSIZE / pginfo->hwpage_size;
+
+	while (page < page_count) {
+		u64 rnum;
+		for (rnum = 0; (rnum < MAX_RPAGES) && (page < page_count);
+		     rnum++) {
+			void *pg = sectbase + ((page++) * pginfo->hwpage_size);
+			kpage[rnum] = virt_to_abs(pg);
+		}
+
+		h_ret = hipz_h_register_rpage_mr(shca->ipz_hca_handle, mr,
+			ehca_encode_hwpage_size(pginfo->hwpage_size),
+			0, rpage, rnum);
+
+		if ((h_ret != H_SUCCESS) && (h_ret != H_PAGE_REGISTERED)) {
+			ehca_err(&shca->ib_device, "register_rpage_mr failed");
+			return h_ret;
+		}
+	}
+	return h_ret;
+}
+
+static u64 ehca_reg_mr_sections(int top, int dir, u64 *kpage,
+				struct ehca_shca *shca, struct ehca_mr *mr,
+				struct ehca_mr_pginfo *pginfo)
+{
+	u64 hret = H_SUCCESS;
+	int idx;
+
+	for (idx = 0; idx < EHCA_MAP_ENTRIES; idx++) {
+		if (!ehca_bmap_valid(ehca_bmap->top[top]->dir[dir]->ent[idx]))
+			continue;
+
+		hret = ehca_reg_mr_section(top, dir, idx, kpage, shca, mr,
+					   pginfo);
+		if ((hret != H_SUCCESS) && (hret != H_PAGE_REGISTERED))
+				return hret;
+	}
+	return hret;
+}
+
+static u64 ehca_reg_mr_dir_sections(int top, u64 *kpage, struct ehca_shca *shca,
+				    struct ehca_mr *mr,
+				    struct ehca_mr_pginfo *pginfo)
+{
+	u64 hret = H_SUCCESS;
+	int dir;
+
+	for (dir = 0; dir < EHCA_MAP_ENTRIES; dir++) {
+		if (!ehca_bmap_valid(ehca_bmap->top[top]->dir[dir]))
+			continue;
+
+		hret = ehca_reg_mr_sections(top, dir, kpage, shca, mr, pginfo);
+		if ((hret != H_SUCCESS) && (hret != H_PAGE_REGISTERED))
+				return hret;
+	}
+	return hret;
+}
 
 /* register internal max-MR to internal SHCA */
 int ehca_reg_internal_maxmr(
@@ -1495,6 +1625,11 @@ int ehca_reg_internal_maxmr(
 	u32 num_hwpages;
 	u64 hw_pgsize;
 
+	if (!ehca_bmap) {
+		ret = -EFAULT;
+		goto ehca_reg_internal_maxmr_exit0;
+	}
+
 	e_mr = ehca_mr_new();
 	if (!e_mr) {
 		ehca_err(&shca->ib_device, "out of memory");
@@ -1504,8 +1639,8 @@ int ehca_reg_internal_maxmr(
 	e_mr->flags |= EHCA_MR_FLAG_MAXMR;
 
 	/* register internal max-MR on HCA */
-	size_maxmr = (u64)high_memory - PAGE_OFFSET;
-	iova_start = (u64 *)KERNELBASE;
+	size_maxmr = ehca_mr_len;
+	iova_start = (u64 *)ehca_map_vaddr((void *)KERNELBASE);
 	ib_pbuf.addr = 0;
 	ib_pbuf.size = size_maxmr;
 	num_kpages = NUM_CHUNKS(((u64)iova_start % PAGE_SIZE) + size_maxmr,
@@ -1524,7 +1659,7 @@ int ehca_reg_internal_maxmr(
 
 	ret = ehca_reg_mr(shca, e_mr, iova_start, size_maxmr, 0, e_pd,
 			  &pginfo, &e_mr->ib.ib_mr.lkey,
-			  &e_mr->ib.ib_mr.rkey);
+			  &e_mr->ib.ib_mr.rkey, EHCA_REG_BUSMAP_MR);
 	if (ret) {
 		ehca_err(&shca->ib_device, "reg of internal max MR failed, "
 			 "e_mr=%p iova_start=%p size_maxmr=%llx num_kpages=%x "
@@ -2077,8 +2212,8 @@ int ehca_mr_is_maxmr(u64 size,
 		     u64 *iova_start)
 {
 	/* a MR is treated as max-MR only if it fits following: */
-	if ((size == ((u64)high_memory - PAGE_OFFSET)) &&
-	    (iova_start == (void *)KERNELBASE)) {
+	if ((size == ehca_mr_len) &&
+	    (iova_start == (void *)ehca_map_vaddr((void *)KERNELBASE))) {
 		ehca_gen_dbg("this is a max-MR");
 		return 1;
 	} else
@@ -2184,3 +2319,350 @@ void ehca_cleanup_mrmw_cache(void)
 	if (mw_cache)
 		kmem_cache_destroy(mw_cache);
 }
+
+static inline int ehca_init_top_bmap(struct ehca_top_bmap *ehca_top_bmap,
+				     int dir)
+{
+	if (!ehca_bmap_valid(ehca_top_bmap->dir[dir])) {
+		ehca_top_bmap->dir[dir] =
+			kmalloc(sizeof(struct ehca_dir_bmap), GFP_KERNEL);
+		if (!ehca_top_bmap->dir[dir])
+			return -ENOMEM;
+		/* Set map block to 0xFF according to EHCA_INVAL_ADDR */
+		memset(ehca_top_bmap->dir[dir], 0xFF, EHCA_ENT_MAP_SIZE);
+	}
+	return 0;
+}
+
+static inline int ehca_init_bmap(struct ehca_bmap *ehca_bmap, int top, int dir)
+{
+	if (!ehca_bmap_valid(ehca_bmap->top[top])) {
+		ehca_bmap->top[top] =
+			kmalloc(sizeof(struct ehca_top_bmap), GFP_KERNEL);
+		if (!ehca_bmap->top[top])
+			return -ENOMEM;
+		/* Set map block to 0xFF according to EHCA_INVAL_ADDR */
+		memset(ehca_bmap->top[top], 0xFF, EHCA_DIR_MAP_SIZE);
+	}
+	return ehca_init_top_bmap(ehca_bmap->top[top], dir);
+}
+
+static inline int ehca_calc_index(unsigned long i, unsigned long s)
+{
+	return (i >> s) & EHCA_INDEX_MASK;
+}
+
+void ehca_destroy_busmap(void)
+{
+	int top, dir;
+
+	if (!ehca_bmap)
+		return;
+
+	for (top = 0; top < EHCA_MAP_ENTRIES; top++) {
+		if (!ehca_bmap_valid(ehca_bmap->top[top]))
+			continue;
+		for (dir = 0; dir < EHCA_MAP_ENTRIES; dir++) {
+			if (!ehca_bmap_valid(ehca_bmap->top[top]->dir[dir]))
+				continue;
+
+			kfree(ehca_bmap->top[top]->dir[dir]);
+		}
+
+		kfree(ehca_bmap->top[top]);
+	}
+
+	kfree(ehca_bmap);
+	ehca_bmap = NULL;
+}
+
+static int ehca_update_busmap(unsigned long pfn, unsigned long nr_pages)
+{
+	unsigned long i, start_section, end_section;
+	int top, dir, idx;
+
+	if (!nr_pages)
+		return 0;
+
+	if (!ehca_bmap) {
+		ehca_bmap = kmalloc(sizeof(struct ehca_bmap), GFP_KERNEL);
+		if (!ehca_bmap)
+			return -ENOMEM;
+		/* Set map block to 0xFF according to EHCA_INVAL_ADDR */
+		memset(ehca_bmap, 0xFF, EHCA_TOP_MAP_SIZE);
+	}
+
+	start_section = phys_to_abs(pfn * PAGE_SIZE) / EHCA_SECTSIZE;
+	end_section = phys_to_abs((pfn + nr_pages) * PAGE_SIZE) / EHCA_SECTSIZE;
+	for (i = start_section; i < end_section; i++) {
+		int ret;
+		top = ehca_calc_index(i, EHCA_TOP_INDEX_SHIFT);
+		dir = ehca_calc_index(i, EHCA_DIR_INDEX_SHIFT);
+		idx = i & EHCA_INDEX_MASK;
+
+		ret = ehca_init_bmap(ehca_bmap, top, dir);
+		if (ret) {
+			ehca_destroy_busmap();
+			return ret;
+		}
+		ehca_bmap->top[top]->dir[dir]->ent[idx] = ehca_mr_len;
+		ehca_mr_len += EHCA_SECTSIZE;
+	}
+	return 0;
+}
+
+static int ehca_is_hugepage(unsigned long pfn)
+{
+	int page_order;
+
+	if (pfn & EHCA_HUGEPAGE_PFN_MASK)
+		return 0;
+
+	page_order = compound_order(pfn_to_page(pfn));
+	if (page_order + PAGE_SHIFT != EHCA_HUGEPAGESHIFT)
+		return 0;
+
+	return 1;
+}
+
+static int ehca_create_busmap_callback(unsigned long initial_pfn,
+				       unsigned long total_nr_pages, void *arg)
+{
+	int ret;
+	unsigned long pfn, start_pfn, end_pfn, nr_pages;
+
+	if ((total_nr_pages * PAGE_SIZE) < EHCA_HUGEPAGE_SIZE)
+		return ehca_update_busmap(initial_pfn, total_nr_pages);
+
+	/* Given chunk is >= 16GB -> check for hugepages */
+	start_pfn = initial_pfn;
+	end_pfn = initial_pfn + total_nr_pages;
+	pfn = start_pfn;
+
+	while (pfn < end_pfn) {
+		if (ehca_is_hugepage(pfn)) {
+			/* Add mem found in front of the hugepage */
+			nr_pages = pfn - start_pfn;
+			ret = ehca_update_busmap(start_pfn, nr_pages);
+			if (ret)
+				return ret;
+			/* Skip the hugepage */
+			pfn += (EHCA_HUGEPAGE_SIZE / PAGE_SIZE);
+			start_pfn = pfn;
+		} else
+			pfn += (EHCA_SECTSIZE / PAGE_SIZE);
+	}
+
+	/* Add mem found behind the hugepage(s)  */
+	nr_pages = pfn - start_pfn;
+	return ehca_update_busmap(start_pfn, nr_pages);
+}
+
+int ehca_create_busmap(void)
+{
+	int ret;
+
+	ehca_mr_len = 0;
+	ret = walk_memory_resource(0, 1ULL << MAX_PHYSMEM_BITS, NULL,
+				   ehca_create_busmap_callback);
+	return ret;
+}
+
+static int ehca_reg_bmap_mr_rpages(struct ehca_shca *shca,
+				   struct ehca_mr *e_mr,
+				   struct ehca_mr_pginfo *pginfo)
+{
+	int top;
+	u64 hret, *kpage;
+
+	kpage = ehca_alloc_fw_ctrlblock(GFP_KERNEL);
+	if (!kpage) {
+		ehca_err(&shca->ib_device, "kpage alloc failed");
+		return -ENOMEM;
+	}
+	for (top = 0; top < EHCA_MAP_ENTRIES; top++) {
+		if (!ehca_bmap_valid(ehca_bmap->top[top]))
+			continue;
+		hret = ehca_reg_mr_dir_sections(top, kpage, shca, e_mr, pginfo);
+		if ((hret != H_PAGE_REGISTERED) && (hret != H_SUCCESS))
+			break;
+	}
+
+	ehca_free_fw_ctrlblock(kpage);
+
+	if (hret == H_SUCCESS)
+		return 0; /* Everything is fine */
+	else {
+		ehca_err(&shca->ib_device, "ehca_reg_bmap_mr_rpages failed, "
+				 "h_ret=%lli e_mr=%p top=%x lkey=%x "
+				 "hca_hndl=%llx mr_hndl=%llx", hret, e_mr, top,
+				 e_mr->ib.ib_mr.lkey,
+				 shca->ipz_hca_handle.handle,
+				 e_mr->ipz_mr_handle.handle);
+		return ehca2ib_return_code(hret);
+	}
+}
+
+static u64 ehca_map_vaddr(void *caddr)
+{
+	int top, dir, idx;
+	unsigned long abs_addr, offset;
+	u64 entry;
+
+	if (!ehca_bmap)
+		return EHCA_INVAL_ADDR;
+
+	abs_addr = virt_to_abs(caddr);
+	top = ehca_calc_index(abs_addr, EHCA_TOP_INDEX_SHIFT + EHCA_SECTSHIFT);
+	if (!ehca_bmap_valid(ehca_bmap->top[top]))
+		return EHCA_INVAL_ADDR;
+
+	dir = ehca_calc_index(abs_addr, EHCA_DIR_INDEX_SHIFT + EHCA_SECTSHIFT);
+	if (!ehca_bmap_valid(ehca_bmap->top[top]->dir[dir]))
+		return EHCA_INVAL_ADDR;
+
+	idx = ehca_calc_index(abs_addr, EHCA_SECTSHIFT);
+
+	entry = ehca_bmap->top[top]->dir[dir]->ent[idx];
+	if (ehca_bmap_valid(entry)) {
+		offset = (unsigned long)caddr & (EHCA_SECTSIZE - 1);
+		return entry | offset;
+	} else
+		return EHCA_INVAL_ADDR;
+}
+
+static int ehca_dma_mapping_error(struct ib_device *dev, u64 dma_addr)
+{
+	return dma_addr == EHCA_INVAL_ADDR;
+}
+
+static u64 ehca_dma_map_single(struct ib_device *dev, void *cpu_addr,
+			       size_t size, enum dma_data_direction direction)
+{
+	if (cpu_addr)
+		return ehca_map_vaddr(cpu_addr);
+	else
+		return EHCA_INVAL_ADDR;
+}
+
+static void ehca_dma_unmap_single(struct ib_device *dev, u64 addr, size_t size,
+				  enum dma_data_direction direction)
+{
+	/* This is only a stub; nothing to be done here */
+}
+
+static u64 ehca_dma_map_page(struct ib_device *dev, struct page *page,
+			     unsigned long offset, size_t size,
+			     enum dma_data_direction direction)
+{
+	u64 addr;
+
+	if (offset + size > PAGE_SIZE)
+		return EHCA_INVAL_ADDR;
+
+	addr = ehca_map_vaddr(page_address(page));
+	if (!ehca_dma_mapping_error(dev, addr))
+		addr += offset;
+
+	return addr;
+}
+
+static void ehca_dma_unmap_page(struct ib_device *dev, u64 addr, size_t size,
+				enum dma_data_direction direction)
+{
+	/* This is only a stub; nothing to be done here */
+}
+
+static int ehca_dma_map_sg(struct ib_device *dev, struct scatterlist *sgl,
+			   int nents, enum dma_data_direction direction)
+{
+	struct scatterlist *sg;
+	int i;
+
+	for_each_sg(sgl, sg, nents, i) {
+		u64 addr;
+		addr = ehca_map_vaddr(sg_virt(sg));
+		if (ehca_dma_mapping_error(dev, addr))
+			return 0;
+
+		sg->dma_address = addr;
+		sg->dma_length = sg->length;
+	}
+	return nents;
+}
+
+static void ehca_dma_unmap_sg(struct ib_device *dev, struct scatterlist *sg,
+			      int nents, enum dma_data_direction direction)
+{
+	/* This is only a stub; nothing to be done here */
+}
+
+static u64 ehca_dma_address(struct ib_device *dev, struct scatterlist *sg)
+{
+	return sg->dma_address;
+}
+
+static unsigned int ehca_dma_len(struct ib_device *dev, struct scatterlist *sg)
+{
+	return sg->length;
+}
+
+static void ehca_dma_sync_single_for_cpu(struct ib_device *dev, u64 addr,
+					 size_t size,
+					 enum dma_data_direction dir)
+{
+	dma_sync_single_for_cpu(dev->dma_device, addr, size, dir);
+}
+
+static void ehca_dma_sync_single_for_device(struct ib_device *dev, u64 addr,
+					    size_t size,
+					    enum dma_data_direction dir)
+{
+	dma_sync_single_for_device(dev->dma_device, addr, size, dir);
+}
+
+static void *ehca_dma_alloc_coherent(struct ib_device *dev, size_t size,
+				     u64 *dma_handle, gfp_t flag)
+{
+	struct page *p;
+	void *addr = NULL;
+	u64 dma_addr;
+
+	p = alloc_pages(flag, get_order(size));
+	if (p) {
+		addr = page_address(p);
+		dma_addr = ehca_map_vaddr(addr);
+		if (ehca_dma_mapping_error(dev, dma_addr)) {
+			free_pages((unsigned long)addr,	get_order(size));
+			return NULL;
+		}
+		if (dma_handle)
+			*dma_handle = dma_addr;
+		return addr;
+	}
+	return NULL;
+}
+
+static void ehca_dma_free_coherent(struct ib_device *dev, size_t size,
+				   void *cpu_addr, u64 dma_handle)
+{
+	if (cpu_addr && size)
+		free_pages((unsigned long)cpu_addr, get_order(size));
+}
+
+
+struct ib_dma_mapping_ops ehca_dma_mapping_ops = {
+	.mapping_error          = ehca_dma_mapping_error,
+	.map_single             = ehca_dma_map_single,
+	.unmap_single           = ehca_dma_unmap_single,
+	.map_page               = ehca_dma_map_page,
+	.unmap_page             = ehca_dma_unmap_page,
+	.map_sg                 = ehca_dma_map_sg,
+	.unmap_sg               = ehca_dma_unmap_sg,
+	.dma_address            = ehca_dma_address,
+	.dma_len                = ehca_dma_len,
+	.sync_single_for_cpu    = ehca_dma_sync_single_for_cpu,
+	.sync_single_for_device = ehca_dma_sync_single_for_device,
+	.alloc_coherent         = ehca_dma_alloc_coherent,
+	.free_coherent          = ehca_dma_free_coherent,
+};
diff --git a/drivers/infiniband/hw/ehca/ehca_mrmw.h b/drivers/infiniband/hw/ehca/ehca_mrmw.h
index bc8f4e31c123..50d8b51306dd 100644
--- a/drivers/infiniband/hw/ehca/ehca_mrmw.h
+++ b/drivers/infiniband/hw/ehca/ehca_mrmw.h
@@ -42,6 +42,11 @@
 #ifndef _EHCA_MRMW_H_
 #define _EHCA_MRMW_H_
 
+enum ehca_reg_type {
+	EHCA_REG_MR,
+	EHCA_REG_BUSMAP_MR
+};
+
 int ehca_reg_mr(struct ehca_shca *shca,
 		struct ehca_mr *e_mr,
 		u64 *iova_start,
@@ -50,7 +55,8 @@ int ehca_reg_mr(struct ehca_shca *shca,
 		struct ehca_pd *e_pd,
 		struct ehca_mr_pginfo *pginfo,
 		u32 *lkey,
-		u32 *rkey);
+		u32 *rkey,
+		enum ehca_reg_type reg_type);
 
 int ehca_reg_mr_rpages(struct ehca_shca *shca,
 		       struct ehca_mr *e_mr,
@@ -118,4 +124,9 @@ void ehca_mrmw_reverse_map_acl(const u32 *hipz_acl,
 
 void ehca_mr_deletenew(struct ehca_mr *mr);
 
+int ehca_create_busmap(void);
+
+void ehca_destroy_busmap(void);
+
+extern struct ib_dma_mapping_ops ehca_dma_mapping_ops;
 #endif  /*_EHCA_MRMW_H_*/
-- 
cgit v1.2.3-59-g8ed1b


From af04662b4d80de5797a595bc9855d09ef4fe55cc Mon Sep 17 00:00:00 2001
From: Roel Kluin <roel.kluin@gmail.com>
Date: Mon, 22 Jun 2009 22:23:48 -0700
Subject: IB/ehca: Ensure that guid_entry index is not negative

This prevents the memcpy() of a guid_entries element using a negative index.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/ehca/ehca_hca.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/ehca/ehca_hca.c b/drivers/infiniband/hw/ehca/ehca_hca.c
index 9209c5332dfe..8b92f85d4dd0 100644
--- a/drivers/infiniband/hw/ehca/ehca_hca.c
+++ b/drivers/infiniband/hw/ehca/ehca_hca.c
@@ -319,7 +319,7 @@ int ehca_query_gid(struct ib_device *ibdev, u8 port,
 					      ib_device);
 	struct hipz_query_port *rblock;
 
-	if (index > 255) {
+	if (index < 0 || index > 255) {
 		ehca_err(&shca->ib_device, "Invalid index: %x.", index);
 		return -EINVAL;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 66388d67a0d7bf39735650de54e42064d1af8b62 Mon Sep 17 00:00:00 2001
From: Faisal Latif <faisal.latif@intel.com>
Date: Mon, 22 Jun 2009 22:52:30 -0700
Subject: RDMA/nes: Fix max_qp_init_rd_atom returned from query device

In nes_query_device(), max_qp_init_rd_atom is incorrectly set to
max_qp_wr.  This was found when a test application had a dapl async
event error.

Signed-off-by: Faisal Latif <faisal.latif@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/nes/nes_verbs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c
index 64d5cfd8f380..21e0fd336cf7 100644
--- a/drivers/infiniband/hw/nes/nes_verbs.c
+++ b/drivers/infiniband/hw/nes/nes_verbs.c
@@ -654,7 +654,7 @@ static int nes_query_device(struct ib_device *ibdev, struct ib_device_attr *prop
 		default:
 			props->max_qp_rd_atom = 0;
 	}
-	props->max_qp_init_rd_atom = props->max_qp_wr;
+	props->max_qp_init_rd_atom = props->max_qp_rd_atom;
 	props->atomic_cap = IB_ATOMIC_NONE;
 	props->max_map_per_fmr = 1;
 
-- 
cgit v1.2.3-59-g8ed1b


From 68237a0ff84503270373c39229be83e865ea08d4 Mon Sep 17 00:00:00 2001
From: Faisal Latif <faisal.latif@intel.com>
Date: Mon, 22 Jun 2009 22:53:28 -0700
Subject: RDMA/nes: Fix FIN state handling under error conditions

During cluster testing, one QP was not closed, as FIN is not handled
properly when its rexmit count expires or in some cases when RST is is
received after sending FIN.  The reason is that the cm_id does not get
decremented under these conditions.

Signed-off-by: Faisal Latif <faisal.latif@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/nes/nes_cm.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c
index 11c7d6642014..114b802771ad 100644
--- a/drivers/infiniband/hw/nes/nes_cm.c
+++ b/drivers/infiniband/hw/nes/nes_cm.c
@@ -472,6 +472,7 @@ int schedule_nes_timer(struct nes_cm_node *cm_node, struct sk_buff *skb,
 
 static void nes_retrans_expired(struct nes_cm_node *cm_node)
 {
+	struct iw_cm_id *cm_id = cm_node->cm_id;
 	switch (cm_node->state) {
 	case NES_CM_STATE_SYN_RCVD:
 	case NES_CM_STATE_CLOSING:
@@ -479,7 +480,9 @@ static void nes_retrans_expired(struct nes_cm_node *cm_node)
 		break;
 	case NES_CM_STATE_LAST_ACK:
 	case NES_CM_STATE_FIN_WAIT1:
-	case NES_CM_STATE_MPAREJ_RCVD:
+		if (cm_node->cm_id)
+			cm_id->rem_ref(cm_id);
+		cm_node->state = NES_CM_STATE_CLOSED;
 		send_reset(cm_node, NULL);
 		break;
 	default:
@@ -1406,6 +1409,7 @@ static void handle_rst_pkt(struct nes_cm_node *cm_node, struct sk_buff *skb,
 	case NES_CM_STATE_CLOSED:
 		drop_packet(skb);
 		break;
+	case NES_CM_STATE_FIN_WAIT1:
 	case NES_CM_STATE_LAST_ACK:
 		cm_node->cm_id->rem_ref(cm_node->cm_id);
 	case NES_CM_STATE_TIME_WAIT:
@@ -1413,8 +1417,6 @@ static void handle_rst_pkt(struct nes_cm_node *cm_node, struct sk_buff *skb,
 		rem_ref_cm_node(cm_node->cm_core, cm_node);
 		drop_packet(skb);
 		break;
-	case NES_CM_STATE_FIN_WAIT1:
-		nes_debug(NES_DBG_CM, "Bad state %s[%u]\n", __func__, __LINE__);
 	default:
 		drop_packet(skb);
 		break;
-- 
cgit v1.2.3-59-g8ed1b


From ad5d8eac91bb04533e49bbb6434791758300711d Mon Sep 17 00:00:00 2001
From: Evgeni Golov <sargentd@die-welt.net>
Date: Mon, 22 Jun 2009 09:46:20 +0200
Subject: [libata] beautify module parameters

1. add defaults to description where possible
2. add value definition (off=0, on=1) where missing

v2: reformatted as per request by Jeff Garzik <jgarzik@pobox.com>
    "Enable foo (0=off, 1=on [default])"

Signed-off-by: Evgeni Golov <sargentd@die-welt.net>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index ca4d208ddf3b..1d894c9d73d0 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -125,19 +125,19 @@ MODULE_PARM_DESC(force, "Force ATA configurations including cable type, link spe
 
 static int atapi_enabled = 1;
 module_param(atapi_enabled, int, 0444);
-MODULE_PARM_DESC(atapi_enabled, "Enable discovery of ATAPI devices (0=off, 1=on)");
+MODULE_PARM_DESC(atapi_enabled, "Enable discovery of ATAPI devices (0=off, 1=on [default])");
 
 static int atapi_dmadir = 0;
 module_param(atapi_dmadir, int, 0444);
-MODULE_PARM_DESC(atapi_dmadir, "Enable ATAPI DMADIR bridge support (0=off, 1=on)");
+MODULE_PARM_DESC(atapi_dmadir, "Enable ATAPI DMADIR bridge support (0=off [default], 1=on)");
 
 int atapi_passthru16 = 1;
 module_param(atapi_passthru16, int, 0444);
-MODULE_PARM_DESC(atapi_passthru16, "Enable ATA_16 passthru for ATAPI devices; on by default (0=off, 1=on)");
+MODULE_PARM_DESC(atapi_passthru16, "Enable ATA_16 passthru for ATAPI devices (0=off, 1=on [default])");
 
 int libata_fua = 0;
 module_param_named(fua, libata_fua, int, 0444);
-MODULE_PARM_DESC(fua, "FUA support (0=off, 1=on)");
+MODULE_PARM_DESC(fua, "FUA support (0=off [default], 1=on)");
 
 static int ata_ignore_hpa;
 module_param_named(ignore_hpa, ata_ignore_hpa, int, 0644);
@@ -153,11 +153,11 @@ MODULE_PARM_DESC(ata_probe_timeout, "Set ATA probing timeout (seconds)");
 
 int libata_noacpi = 0;
 module_param_named(noacpi, libata_noacpi, int, 0444);
-MODULE_PARM_DESC(noacpi, "Disables the use of ACPI in probe/suspend/resume when set");
+MODULE_PARM_DESC(noacpi, "Disable the use of ACPI in probe/suspend/resume (0=off [default], 1=on)");
 
 int libata_allow_tpm = 0;
 module_param_named(allow_tpm, libata_allow_tpm, int, 0444);
-MODULE_PARM_DESC(allow_tpm, "Permit the use of TPM commands");
+MODULE_PARM_DESC(allow_tpm, "Permit the use of TPM commands (0=off [default], 1=on)");
 
 MODULE_AUTHOR("Jeff Garzik");
 MODULE_DESCRIPTION("Library module for ATA devices");
-- 
cgit v1.2.3-59-g8ed1b


From 918d7b7c330f8afe18cb1b8692fc5f45a798634e Mon Sep 17 00:00:00 2001
From: Sergey Matyukevich <geomatsi@gmail.com>
Date: Fri, 19 Jun 2009 08:27:40 +0400
Subject: [libata] PATA driver for CF interface on AT91SAM9260 SoC

This patch provides PATA driver for CompactFlash interface in True IDE
mode on AT91SAM9260 SoC.

Signed-off-by: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/Kconfig     |   8 ++
 drivers/ata/Makefile    |   1 +
 drivers/ata/pata_at91.c | 361 ++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 370 insertions(+)
 create mode 100644 drivers/ata/pata_at91.c

diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
index 2aa1908e5ce0..b17c57f85032 100644
--- a/drivers/ata/Kconfig
+++ b/drivers/ata/Kconfig
@@ -679,6 +679,14 @@ config PATA_PLATFORM
 
 	  If unsure, say N.
 
+config PATA_AT91
+	tristate "PATA support for AT91SAM9260"
+	depends on ARM && ARCH_AT91
+	help
+	  This option enables support for IDE devices on the Atmel AT91SAM9260 SoC.
+
+	  If unsure, say N.
+
 config PATA_OF_PLATFORM
 	tristate "OpenFirmware platform device PATA support"
 	depends on PATA_PLATFORM && PPC_OF
diff --git a/drivers/ata/Makefile b/drivers/ata/Makefile
index 1558059874f0..38906f9bbb4e 100644
--- a/drivers/ata/Makefile
+++ b/drivers/ata/Makefile
@@ -72,6 +72,7 @@ obj-$(CONFIG_PATA_SCH)		+= pata_sch.o
 obj-$(CONFIG_PATA_BF54X)	+= pata_bf54x.o
 obj-$(CONFIG_PATA_OCTEON_CF)	+= pata_octeon_cf.o
 obj-$(CONFIG_PATA_PLATFORM)	+= pata_platform.o
+obj-$(CONFIG_PATA_AT91)	+= pata_at91.o
 obj-$(CONFIG_PATA_OF_PLATFORM)	+= pata_of_platform.o
 obj-$(CONFIG_PATA_ICSIDE)	+= pata_icside.o
 # Should be last but two libata driver
diff --git a/drivers/ata/pata_at91.c b/drivers/ata/pata_at91.c
new file mode 100644
index 000000000000..4b27617be26d
--- /dev/null
+++ b/drivers/ata/pata_at91.c
@@ -0,0 +1,361 @@
+/*
+ * PATA driver for AT91SAM9260 Static Memory Controller
+ * with CompactFlash interface in True IDE mode
+ *
+ * Copyright (C) 2009 Matyukevich Sergey
+ *
+ * Based on:
+ *      * generic platform driver by Paul Mundt: drivers/ata/pata_platform.c
+ *      * pata_at32 driver by Kristoffer Nyborg Gregertsen
+ *      * at91_ide driver by Stanislaw Gruszka
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/blkdev.h>
+#include <scsi/scsi_host.h>
+#include <linux/ata.h>
+#include <linux/clk.h>
+#include <linux/libata.h>
+#include <linux/platform_device.h>
+#include <linux/ata_platform.h>
+
+#include <mach/at91sam9260_matrix.h>
+#include <mach/at91sam9_smc.h>
+#include <mach/at91sam9260.h>
+#include <mach/board.h>
+#include <mach/gpio.h>
+
+
+#define DRV_NAME "pata_at91"
+#define DRV_VERSION "0.1"
+
+#define CF_IDE_OFFSET	    0x00c00000
+#define CF_ALT_IDE_OFFSET   0x00e00000
+#define CF_IDE_RES_SIZE     0x08
+
+struct at91_ide_info {
+	unsigned long mode;
+	unsigned int cs;
+
+	void __iomem *ide_addr;
+	void __iomem *alt_addr;
+};
+
+const struct ata_timing initial_timing =
+	{XFER_PIO_0, 70, 290, 240, 600, 165, 150, 600, 0};
+
+static unsigned int calc_mck_cycles(unsigned int ns, unsigned int mck_hz)
+{
+	unsigned long mul;
+
+    /*
+     * cycles = x [nsec] * f [Hz] / 10^9 [ns in sec] =
+     *     x * (f / 1_000_000_000) =
+     *     x * ((f * 65536) / 1_000_000_000) / 65536 =
+     *     x * (((f / 10_000) * 65536) / 100_000) / 65536 =
+     */
+
+    mul = (mck_hz / 10000) << 16;
+    mul /= 100000;
+
+    return (ns * mul + 65536) >> 16;    /* rounding */
+}
+
+static void set_smc_mode(struct at91_ide_info *info)
+{
+    at91_sys_write(AT91_SMC_MODE(info->cs), info->mode);
+    return;
+}
+
+static void set_smc_timing(struct device *dev,
+		struct at91_ide_info *info, const struct ata_timing *ata)
+{
+	int read_cycle, write_cycle, active, recover;
+	int nrd_setup, nrd_pulse, nrd_recover;
+	int nwe_setup, nwe_pulse;
+
+	int ncs_write_setup, ncs_write_pulse;
+	int ncs_read_setup, ncs_read_pulse;
+
+	unsigned int mck_hz;
+	struct clk *mck;
+
+	read_cycle  = ata->cyc8b;
+	nrd_setup   = ata->setup;
+	nrd_pulse   = ata->act8b;
+	nrd_recover = ata->rec8b;
+
+	mck = clk_get(NULL, "mck");
+	BUG_ON(IS_ERR(mck));
+	mck_hz = clk_get_rate(mck);
+
+	read_cycle  = calc_mck_cycles(read_cycle, mck_hz);
+	nrd_setup   = calc_mck_cycles(nrd_setup, mck_hz);
+	nrd_pulse   = calc_mck_cycles(nrd_pulse, mck_hz);
+	nrd_recover = calc_mck_cycles(nrd_recover, mck_hz);
+
+	clk_put(mck);
+
+	active  = nrd_setup + nrd_pulse;
+	recover = read_cycle - active;
+
+	/* Need at least two cycles recovery */
+	if (recover < 2)
+		read_cycle = active + 2;
+
+	/* (CS0, CS1, DIR, OE) <= (CFCE1, CFCE2, CFRNW, NCSX) timings */
+	ncs_read_setup = 1;
+	ncs_read_pulse = read_cycle - 2;
+
+	/* Write timings same as read timings */
+	write_cycle = read_cycle;
+	nwe_setup = nrd_setup;
+	nwe_pulse = nrd_pulse;
+	ncs_write_setup = ncs_read_setup;
+	ncs_write_pulse = ncs_read_pulse;
+
+	dev_dbg(dev, "ATA timings: nrd_setup = %d nrd_pulse = %d nrd_cycle = %d\n",
+			nrd_setup, nrd_pulse, read_cycle);
+	dev_dbg(dev, "ATA timings: nwe_setup = %d nwe_pulse = %d nwe_cycle = %d\n",
+			nwe_setup, nwe_pulse, write_cycle);
+	dev_dbg(dev, "ATA timings: ncs_read_setup = %d ncs_read_pulse = %d\n",
+			ncs_read_setup, ncs_read_pulse);
+	dev_dbg(dev, "ATA timings: ncs_write_setup = %d ncs_write_pulse = %d\n",
+			ncs_write_setup, ncs_write_pulse);
+
+	at91_sys_write(AT91_SMC_SETUP(info->cs),
+			AT91_SMC_NWESETUP_(nwe_setup) |
+			AT91_SMC_NRDSETUP_(nrd_setup) |
+			AT91_SMC_NCS_WRSETUP_(ncs_write_setup) |
+			AT91_SMC_NCS_RDSETUP_(ncs_read_setup));
+
+	at91_sys_write(AT91_SMC_PULSE(info->cs),
+			AT91_SMC_NWEPULSE_(nwe_pulse) |
+			AT91_SMC_NRDPULSE_(nrd_pulse) |
+			AT91_SMC_NCS_WRPULSE_(ncs_write_pulse) |
+			AT91_SMC_NCS_RDPULSE_(ncs_read_pulse));
+
+	at91_sys_write(AT91_SMC_CYCLE(info->cs),
+			AT91_SMC_NWECYCLE_(write_cycle) |
+			AT91_SMC_NRDCYCLE_(read_cycle));
+
+	return;
+}
+
+static void pata_at91_set_piomode(struct ata_port *ap, struct ata_device *adev)
+{
+	struct at91_ide_info *info = ap->host->private_data;
+	struct ata_timing timing;
+	int ret;
+
+	/* Compute ATA timing and set it to SMC */
+	ret = ata_timing_compute(adev, adev->pio_mode, &timing, 1000, 0);
+	if (ret) {
+		dev_warn(ap->dev, "Failed to compute ATA timing %d, \
+				set PIO_0 timing\n", ret);
+		set_smc_timing(ap->dev, info, &initial_timing);
+	} else {
+		set_smc_timing(ap->dev, info, &timing);
+	}
+
+	/* Setup SMC mode */
+	set_smc_mode(info);
+
+	return;
+}
+
+static unsigned int pata_at91_data_xfer_noirq(struct ata_device *dev,
+		unsigned char *buf, unsigned int buflen, int rw)
+{
+	struct at91_ide_info *info = dev->link->ap->host->private_data;
+	unsigned int consumed;
+	unsigned long flags;
+	unsigned int mode;
+
+	local_irq_save(flags);
+	mode = at91_sys_read(AT91_SMC_MODE(info->cs));
+
+	/* set 16bit mode before writing data */
+	at91_sys_write(AT91_SMC_MODE(info->cs),
+			(mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_16);
+
+	consumed = ata_sff_data_xfer(dev, buf, buflen, rw);
+
+	/* restore 8bit mode after data is written */
+	at91_sys_write(AT91_SMC_MODE(info->cs),
+			(mode & ~AT91_SMC_DBW) | AT91_SMC_DBW_8);
+
+	local_irq_restore(flags);
+	return consumed;
+}
+
+static struct scsi_host_template pata_at91_sht = {
+	ATA_PIO_SHT(DRV_NAME),
+};
+
+static struct ata_port_operations pata_at91_port_ops = {
+	.inherits	= &ata_sff_port_ops,
+
+	.sff_data_xfer	= pata_at91_data_xfer_noirq,
+	.set_piomode	= pata_at91_set_piomode,
+	.cable_detect	= ata_cable_40wire,
+	.port_start	= ATA_OP_NULL,
+};
+
+static int __devinit pata_at91_probe(struct platform_device *pdev)
+{
+	struct at91_cf_data *board = pdev->dev.platform_data;
+	struct device *dev = &pdev->dev;
+	struct at91_ide_info *info;
+	struct resource *mem_res;
+	struct ata_host *host;
+	struct ata_port *ap;
+	int irq_flags = 0;
+	int irq = 0;
+	int ret;
+
+	/*  get platform resources: IO/CTL memories and irq/rst pins */
+
+	if (pdev->num_resources != 1) {
+		dev_err(&pdev->dev, "invalid number of resources\n");
+		return -EINVAL;
+	}
+
+	mem_res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+
+	if (!mem_res) {
+		dev_err(dev, "failed to get mem resource\n");
+		return -EINVAL;
+	}
+
+	irq = board->irq_pin;
+
+	/* init ata host */
+
+	host = ata_host_alloc(dev, 1);
+
+	if (!host)
+		return -ENOMEM;
+
+	ap = host->ports[0];
+	ap->ops = &pata_at91_port_ops;
+	ap->flags |= ATA_FLAG_SLAVE_POSS;
+	ap->pio_mask = ATA_PIO4;
+
+	if (!irq) {
+		ap->flags |= ATA_FLAG_PIO_POLLING;
+		ata_port_desc(ap, "no IRQ, using PIO polling");
+	}
+
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+
+	if (!info) {
+		dev_err(dev, "failed to allocate memory for private data\n");
+		return -ENOMEM;
+	}
+
+	info->cs    = board->chipselect;
+	info->mode  = AT91_SMC_READMODE | AT91_SMC_WRITEMODE |
+		AT91_SMC_EXNWMODE_READY | AT91_SMC_BAT_SELECT |
+		AT91_SMC_DBW_8 | AT91_SMC_TDF_(0);
+
+	info->ide_addr = devm_ioremap(dev,
+			mem_res->start + CF_IDE_OFFSET, CF_IDE_RES_SIZE);
+
+	if (!info->ide_addr) {
+		dev_err(dev, "failed to map IO base\n");
+		ret = -ENOMEM;
+		goto err_ide_ioremap;
+	}
+
+	info->alt_addr = devm_ioremap(dev,
+			mem_res->start + CF_ALT_IDE_OFFSET, CF_IDE_RES_SIZE);
+
+	if (!info->alt_addr) {
+		dev_err(dev, "failed to map CTL base\n");
+		ret = -ENOMEM;
+		goto err_alt_ioremap;
+	}
+
+	ap->ioaddr.cmd_addr = info->ide_addr;
+	ap->ioaddr.ctl_addr = info->alt_addr + 0x06;
+	ap->ioaddr.altstatus_addr = ap->ioaddr.ctl_addr;
+
+	ata_sff_std_ports(&ap->ioaddr);
+
+	ata_port_desc(ap, "mmio cmd 0x%llx ctl 0x%llx",
+			(unsigned long long)mem_res->start + CF_IDE_OFFSET,
+			(unsigned long long)mem_res->start + CF_ALT_IDE_OFFSET);
+
+	host->private_data = info;
+
+	return ata_host_activate(host, irq ? gpio_to_irq(irq) : 0,
+			irq ? ata_sff_interrupt : NULL,
+			irq_flags, &pata_at91_sht);
+
+err_alt_ioremap:
+	devm_iounmap(dev, info->ide_addr);
+
+err_ide_ioremap:
+	kfree(info);
+
+	return ret;
+}
+
+static int __devexit pata_at91_remove(struct platform_device *pdev)
+{
+	struct ata_host *host = dev_get_drvdata(&pdev->dev);
+	struct at91_ide_info *info = host->private_data;
+	struct device *dev = &pdev->dev;
+
+	if (!host)
+		return 0;
+
+	ata_host_detach(host);
+
+	if (!info)
+		return 0;
+
+	devm_iounmap(dev, info->ide_addr);
+	devm_iounmap(dev, info->alt_addr);
+
+	kfree(info);
+	return 0;
+}
+
+static struct platform_driver pata_at91_driver = {
+	.probe		= pata_at91_probe,
+	.remove		= __devexit_p(pata_at91_remove),
+	.driver 	= {
+		.name		= DRV_NAME,
+		.owner		= THIS_MODULE,
+	},
+};
+
+static int __init pata_at91_init(void)
+{
+	return platform_driver_register(&pata_at91_driver);
+}
+
+static void __exit pata_at91_exit(void)
+{
+	platform_driver_unregister(&pata_at91_driver);
+}
+
+
+module_init(pata_at91_init);
+module_exit(pata_at91_exit);
+
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Driver for CF in True IDE mode on AT91SAM9260 SoC");
+MODULE_AUTHOR("Matyukevich Sergey");
+MODULE_VERSION(DRV_VERSION);
+
-- 
cgit v1.2.3-59-g8ed1b


From dc77ad4c8727d3a1c23eadcb287501dab480d634 Mon Sep 17 00:00:00 2001
From: Dave Liu <daveliu@freescale.com>
Date: Wed, 10 Jun 2009 22:53:37 -0500
Subject: sata_fsl: Add power mgmt support

Signed-off-by: Dave Liu <daveliu@freescale.com>
Signed-off-by: Liu Yu <yu.liu@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/sata_fsl.c | 35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index 36b8629203be..94eaa432c40a 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -1378,6 +1378,37 @@ static int sata_fsl_remove(struct of_device *ofdev)
 	return 0;
 }
 
+#ifdef CONFIG_PM
+static int sata_fsl_suspend(struct of_device *op, pm_message_t state)
+{
+	struct ata_host *host = dev_get_drvdata(&op->dev);
+	return ata_host_suspend(host, state);
+}
+
+static int sata_fsl_resume(struct of_device *op)
+{
+	struct ata_host *host = dev_get_drvdata(&op->dev);
+	struct sata_fsl_host_priv *host_priv = host->private_data;
+	int ret;
+	void __iomem *hcr_base = host_priv->hcr_base;
+	struct ata_port *ap = host->ports[0];
+	struct sata_fsl_port_priv *pp = ap->private_data;
+
+	ret = sata_fsl_init_controller(host);
+	if (ret) {
+		dev_printk(KERN_ERR, &op->dev,
+			"Error initialize hardware\n");
+		return ret;
+	}
+
+	/* Recovery the CHBA register in host controller cmd register set */
+	iowrite32(pp->cmdslot_paddr & 0xffffffff, hcr_base + CHBA);
+
+	ata_host_resume(host);
+	return 0;
+}
+#endif
+
 static struct of_device_id fsl_sata_match[] = {
 	{
 		.compatible = "fsl,pq-sata",
@@ -1392,6 +1423,10 @@ static struct of_platform_driver fsl_sata_driver = {
 	.match_table	= fsl_sata_match,
 	.probe		= sata_fsl_probe,
 	.remove		= sata_fsl_remove,
+#ifdef CONFIG_PM
+	.suspend	= sata_fsl_suspend,
+	.resume		= sata_fsl_resume,
+#endif
 };
 
 static int __init sata_fsl_init(void)
-- 
cgit v1.2.3-59-g8ed1b


From 0d9e6659a1bde3733cfd0072adbb3514b579e383 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 11 Jun 2009 11:04:45 +0900
Subject: libata: don't set IORDY for reset

Before issuing reset, libata configures xfermode to PIO0 which makes
some drivers turn on IORDY which may cause the controller to lock up
if the port is not occupied.  IORDY isn't necessary at this point
anyway.  Make ata_pio_need_iordy() return zero if it's being called
for reset.

This fixes bko#11703.  Reported and tracked down by Daniel Gnoutcheff
and Constantine Gavrilov.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Daniel Gnoutcheff <gnoutchd@union.edu>
Cc: Constantine Gavrilov <constantine.gavrilov@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 1d894c9d73d0..045a486a09ea 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -1993,11 +1993,17 @@ unsigned int ata_do_simple_cmd(struct ata_device *dev, u8 cmd)
  *	Check if the current speed of the device requires IORDY. Used
  *	by various controllers for chip configuration.
  */
-
 unsigned int ata_pio_need_iordy(const struct ata_device *adev)
 {
-	/* Controller doesn't support  IORDY. Probably a pointless check
-	   as the caller should know this */
+	/* Don't set IORDY if we're preparing for reset.  IORDY may
+	 * lead to controller lock up on certain controllers if the
+	 * port is not occupied.  See bko#11703 for details.
+	 */
+	if (adev->link->ap->pflags & ATA_PFLAG_RESETTING)
+		return 0;
+	/* Controller doesn't support IORDY.  Probably a pointless
+	 * check as the caller should know this.
+	 */
 	if (adev->link->ap->flags & ATA_FLAG_NO_IORDY)
 		return 0;
 	/* CF spec. r4.1 Table 22 says no iordy on PIO5 and PIO6.  */
@@ -2020,7 +2026,6 @@ unsigned int ata_pio_need_iordy(const struct ata_device *adev)
  *	Compute the highest mode possible if we are not using iordy. Return
  *	-1 if no iordy mode is available.
  */
-
 static u32 ata_pio_mask_no_iordy(const struct ata_device *adev)
 {
 	/* If we have no drive specific rule, then PIO 2 is non IORDY */
-- 
cgit v1.2.3-59-g8ed1b


From 99987bea474ceca8ec6fb05f81d7d188634cdffd Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Mon, 22 Jun 2009 23:04:13 -0700
Subject: IB/mthca: Replace dma_sync_single() use with proper functions

dma_sync_single() is deprecated now, and the use in mthca is wrong:
there should be a dma_sync_single_for_cpu() before touching the memory
from the CPU, and a dma_sync_single_for_device() afterwards.  Fix
this, prompted by a kick in the pants from a patch from FUJITA
Tomonori <fujita.tomonori@lab.ntt.co.jp>.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mthca/mthca_mr.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/infiniband/hw/mthca/mthca_mr.c b/drivers/infiniband/hw/mthca/mthca_mr.c
index d606edf10858..065b20899876 100644
--- a/drivers/infiniband/hw/mthca/mthca_mr.c
+++ b/drivers/infiniband/hw/mthca/mthca_mr.c
@@ -352,10 +352,14 @@ static void mthca_arbel_write_mtt_seg(struct mthca_dev *dev,
 
 	BUG_ON(!mtts);
 
+	dma_sync_single_for_cpu(&dev->pdev->dev, dma_handle,
+				list_len * sizeof (u64), DMA_TO_DEVICE);
+
 	for (i = 0; i < list_len; ++i)
 		mtts[i] = cpu_to_be64(buffer_list[i] | MTHCA_MTT_FLAG_PRESENT);
 
-	dma_sync_single(&dev->pdev->dev, dma_handle, list_len * sizeof (u64), DMA_TO_DEVICE);
+	dma_sync_single_for_device(&dev->pdev->dev, dma_handle,
+				   list_len * sizeof (u64), DMA_TO_DEVICE);
 }
 
 int mthca_write_mtt(struct mthca_dev *dev, struct mthca_mtt *mtt,
@@ -803,12 +807,15 @@ int mthca_arbel_map_phys_fmr(struct ib_fmr *ibfmr, u64 *page_list,
 
 	wmb();
 
+	dma_sync_single_for_cpu(&dev->pdev->dev, fmr->mem.arbel.dma_handle,
+				list_len * sizeof(u64), DMA_TO_DEVICE);
+
 	for (i = 0; i < list_len; ++i)
 		fmr->mem.arbel.mtts[i] = cpu_to_be64(page_list[i] |
 						     MTHCA_MTT_FLAG_PRESENT);
 
-	dma_sync_single(&dev->pdev->dev, fmr->mem.arbel.dma_handle,
-			list_len * sizeof(u64), DMA_TO_DEVICE);
+	dma_sync_single_for_device(&dev->pdev->dev, fmr->mem.arbel.dma_handle,
+				   list_len * sizeof(u64), DMA_TO_DEVICE);
 
 	fmr->mem.arbel.mpt->key    = cpu_to_be32(key);
 	fmr->mem.arbel.mpt->lkey   = cpu_to_be32(key);
-- 
cgit v1.2.3-59-g8ed1b


From e727f5cde90d5a8b92a1ffa49c636a3790301469 Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Mon, 22 Jun 2009 23:07:56 -0700
Subject: mlx4_core: Fix dma_sync_single_for_cpu() with matching for_device()
 calls

Commit 5d23a1d2 ("net: replace dma_sync_single with
dma_sync_single_for_cpu") replaced uses of the deprectated function
dma_sync_single() with calls to dma_sync_single_for_cpu().  However,
to be correct, the code should do a sync for_cpu() before touching the
memory and for_device() after it's done.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/net/mlx4/mr.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/net/mlx4/mr.c b/drivers/net/mlx4/mr.c
index 5887e4764d22..f96948be0a44 100644
--- a/drivers/net/mlx4/mr.c
+++ b/drivers/net/mlx4/mr.c
@@ -399,11 +399,14 @@ static int mlx4_write_mtt_chunk(struct mlx4_dev *dev, struct mlx4_mtt *mtt,
 	if (!mtts)
 		return -ENOMEM;
 
+	dma_sync_single_for_cpu(&dev->pdev->dev, dma_handle,
+				npages * sizeof (u64), DMA_TO_DEVICE);
+
 	for (i = 0; i < npages; ++i)
 		mtts[i] = cpu_to_be64(page_list[i] | MLX4_MTT_FLAG_PRESENT);
 
-	dma_sync_single_for_cpu(&dev->pdev->dev, dma_handle,
-				npages * sizeof (u64), DMA_TO_DEVICE);
+	dma_sync_single_for_device(&dev->pdev->dev, dma_handle,
+				   npages * sizeof (u64), DMA_TO_DEVICE);
 
 	return 0;
 }
@@ -547,11 +550,14 @@ int mlx4_map_phys_fmr(struct mlx4_dev *dev, struct mlx4_fmr *fmr, u64 *page_list
 	/* Make sure MPT status is visible before writing MTT entries */
 	wmb();
 
+	dma_sync_single_for_cpu(&dev->pdev->dev, fmr->dma_handle,
+				npages * sizeof(u64), DMA_TO_DEVICE);
+
 	for (i = 0; i < npages; ++i)
 		fmr->mtts[i] = cpu_to_be64(page_list[i] | MLX4_MTT_FLAG_PRESENT);
 
-	dma_sync_single_for_cpu(&dev->pdev->dev, fmr->dma_handle,
-				npages * sizeof(u64), DMA_TO_DEVICE);
+	dma_sync_single_for_device(&dev->pdev->dev, fmr->dma_handle,
+				   npages * sizeof(u64), DMA_TO_DEVICE);
 
 	fmr->mpt->key    = cpu_to_be32(key);
 	fmr->mpt->lkey   = cpu_to_be32(key);
-- 
cgit v1.2.3-59-g8ed1b


From 14422f9dd8515bfbe6fdbde37eadf59e2980f104 Mon Sep 17 00:00:00 2001
From: Mauro Carvalho Chehab <mchehab@redhat.com>
Date: Tue, 16 Jun 2009 23:55:44 -0300
Subject: V4L/DVB (12010): cx88: Properly support Leadtek TV2000 XP Global

Fix Leadtek TV2000 XP Global entries and add missing PCI ID's.

Thanks to Terry Wu <terrywu2009@gmail.com> for pointing us for the proper settings.

Cc: Terry Wu <terrywu2009@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 Documentation/video4linux/CARDLIST.cx88 |  6 +--
 drivers/media/video/cx88/cx88-cards.c   | 94 ++++++++++++++++++++++++++-------
 2 files changed, 78 insertions(+), 22 deletions(-)

diff --git a/Documentation/video4linux/CARDLIST.cx88 b/Documentation/video4linux/CARDLIST.cx88
index 89093f531727..0736518b2f88 100644
--- a/Documentation/video4linux/CARDLIST.cx88
+++ b/Documentation/video4linux/CARDLIST.cx88
@@ -6,8 +6,8 @@
   5 -> Leadtek Winfast 2000XP Expert                       [107d:6611,107d:6613]
   6 -> AverTV Studio 303 (M126)                            [1461:000b]
   7 -> MSI TV-@nywhere Master                              [1462:8606]
-  8 -> Leadtek Winfast DV2000                              [107d:6620]
-  9 -> Leadtek PVR 2000                                    [107d:663b,107d:663c,107d:6632]
+  8 -> Leadtek Winfast DV2000                              [107d:6620,107d:6621]
+  9 -> Leadtek PVR 2000                                    [107d:663b,107d:663c,107d:6632,107d:6630,107d:6638,107d:6631,107d:6637,107d:663d]
  10 -> IODATA GV-VCP3/PCI                                  [10fc:d003]
  11 -> Prolink PlayTV PVR
  12 -> ASUS PVR-416                                        [1043:4823,1461:c111]
@@ -59,7 +59,7 @@
  58 -> Pinnacle PCTV HD 800i                               [11bd:0051]
  59 -> DViCO FusionHDTV 5 PCI nano                         [18ac:d530]
  60 -> Pinnacle Hybrid PCTV                                [12ab:1788]
- 61 -> Winfast TV2000 XP Global                            [107d:6f18]
+ 61 -> Leadtek TV2000 XP Global                            [107d:6f18,107d:6618]
  62 -> PowerColor RA330                                    [14f1:ea3d]
  63 -> Geniatech X8000-MT DVBT                             [14f1:8852]
  64 -> DViCO FusionHDTV DVB-T PRO                          [18ac:db30]
diff --git a/drivers/media/video/cx88/cx88-cards.c b/drivers/media/video/cx88/cx88-cards.c
index 94b7a52629d0..a5cc1c1fc2d6 100644
--- a/drivers/media/video/cx88/cx88-cards.c
+++ b/drivers/media/video/cx88/cx88-cards.c
@@ -1524,33 +1524,45 @@ static const struct cx88_board cx88_boards[] = {
 		},
 		.mpeg           = CX88_MPEG_DVB,
 	},
+	/* Terry Wu <terrywu2009@gmail.com> */
+	/* TV Audio :      set GPIO 2, 18, 19 value to 0, 1, 0 */
+	/* FM Audio :      set GPIO 2, 18, 19 value to 0, 0, 0 */
+	/* Line-in Audio : set GPIO 2, 18, 19 value to 0, 1, 1 */
+	/* Mute Audio :    set GPIO 2 value to 1               */
 	[CX88_BOARD_WINFAST_TV2000_XP_GLOBAL] = {
-		.name           = "Winfast TV2000 XP Global",
+		.name           = "Leadtek TV2000 XP Global",
 		.tuner_type     = TUNER_XC2028,
 		.tuner_addr     = 0x61,
+		.radio_type     = TUNER_XC2028,
+		.radio_addr     = 0x61,
 		.input          = { {
 			.type   = CX88_VMUX_TELEVISION,
 			.vmux   = 0,
-			.gpio0  = 0x0400, /* pin 2:mute = 0 (off?) */
+			.gpio0  = 0x0400,       /* pin 2 = 0 */
 			.gpio1  = 0x0000,
-			.gpio2  = 0x0800, /* pin 19:audio = 0 (tv) */
-
+			.gpio2  = 0x0C04,       /* pin 18 = 1, pin 19 = 0 */
+			.gpio3  = 0x0000,
 		}, {
 			.type   = CX88_VMUX_COMPOSITE1,
 			.vmux   = 1,
-			.gpio0  = 0x0400, /* probably?  or 0x0404 to turn mute on */
+			.gpio0  = 0x0400,       /* pin 2 = 0 */
 			.gpio1  = 0x0000,
-			.gpio2  = 0x0808, /* pin 19:audio = 1 (line) */
-
+			.gpio2  = 0x0C0C,       /* pin 18 = 1, pin 19 = 1 */
+			.gpio3  = 0x0000,
 		}, {
 			.type   = CX88_VMUX_SVIDEO,
 			.vmux   = 2,
+			.gpio0  = 0x0400,       /* pin 2 = 0 */
+			.gpio1  = 0x0000,
+			.gpio2  = 0x0C0C,       /* pin 18 = 1, pin 19 = 1 */
+			.gpio3  = 0x0000,
 		} },
 		.radio = {
 			.type   = CX88_RADIO,
-			.gpio0  = 0x004ff,
-			.gpio1  = 0x010ff,
-			.gpio2  = 0x0ff,
+			.gpio0  = 0x0400,        /* pin 2 = 0 */
+			.gpio1  = 0x0000,
+			.gpio2  = 0x0C00,       /* pin 18 = 0, pin 19 = 0 */
+			.gpio3  = 0x0000,
 		},
 	},
 	[CX88_BOARD_POWERCOLOR_REAL_ANGEL] = {
@@ -2438,6 +2450,41 @@ static const struct cx88_subid cx88_subids[] = {
 		.subvendor = 0x107d,
 		.subdevice = 0x6654,
 		.card      = CX88_BOARD_WINFAST_DTV1800H,
+	}, {
+		/* PVR2000 PAL Model [107d:6630] */
+		.subvendor = 0x107d,
+		.subdevice = 0x6630,
+		.card      = CX88_BOARD_LEADTEK_PVR2000,
+	}, {
+		/* PVR2000 PAL Model [107d:6638] */
+		.subvendor = 0x107d,
+		.subdevice = 0x6638,
+		.card      = CX88_BOARD_LEADTEK_PVR2000,
+	}, {
+		/* PVR2000 NTSC Model [107d:6631] */
+		.subvendor = 0x107d,
+		.subdevice = 0x6631,
+		.card      = CX88_BOARD_LEADTEK_PVR2000,
+	}, {
+		/* PVR2000 NTSC Model [107d:6637] */
+		.subvendor = 0x107d,
+		.subdevice = 0x6637,
+		.card      = CX88_BOARD_LEADTEK_PVR2000,
+	}, {
+		/* PVR2000 NTSC Model [107d:663d] */
+		.subvendor = 0x107d,
+		.subdevice = 0x663d,
+		.card      = CX88_BOARD_LEADTEK_PVR2000,
+	}, {
+		/* DV2000 NTSC Model [107d:6621] */
+		.subvendor = 0x107d,
+		.subdevice = 0x6621,
+		.card      = CX88_BOARD_WINFAST_DV2000,
+	}, {
+		/* TV2000 XP Global [107d:6618]  */
+		.subvendor = 0x107d,
+		.subdevice = 0x6618,
+		.card      = CX88_BOARD_WINFAST_TV2000_XP_GLOBAL,
 	},
 };
 
@@ -2446,12 +2493,6 @@ static const struct cx88_subid cx88_subids[] = {
 
 static void leadtek_eeprom(struct cx88_core *core, u8 *eeprom_data)
 {
-	/* This is just for the "Winfast 2000XP Expert" board ATM; I don't have data on
-	 * any others.
-	 *
-	 * Byte 0 is 1 on the NTSC board.
-	 */
-
 	if (eeprom_data[4] != 0x7d ||
 	    eeprom_data[5] != 0x10 ||
 	    eeprom_data[7] != 0x66) {
@@ -2459,8 +2500,19 @@ static void leadtek_eeprom(struct cx88_core *core, u8 *eeprom_data)
 		return;
 	}
 
-	core->board.tuner_type = (eeprom_data[6] == 0x13) ?
-		TUNER_PHILIPS_FM1236_MK3 : TUNER_PHILIPS_FM1216ME_MK3;
+	/* Terry Wu <terrywu2009@gmail.com> */
+	switch (eeprom_data[6]) {
+	case 0x13: /* SSID 6613 for TV2000 XP Expert NTSC Model */
+	case 0x21: /* SSID 6621 for DV2000 NTSC Model */
+	case 0x31: /* SSID 6631 for PVR2000 NTSC Model */
+	case 0x37: /* SSID 6637 for PVR2000 NTSC Model */
+	case 0x3d: /* SSID 6637 for PVR2000 NTSC Model */
+		core->board.tuner_type = TUNER_PHILIPS_FM1236_MK3;
+		break;
+	default:
+		core->board.tuner_type = TUNER_PHILIPS_FM1216ME_MK3;
+		break;
+	}
 
 	info_printk(core, "Leadtek Winfast 2000XP Expert config: "
 		    "tuner=%d, eeprom[0]=0x%02x\n",
@@ -2713,7 +2765,6 @@ static int cx88_xc2028_tuner_callback(struct cx88_core *core,
 {
 	/* Board-specific callbacks */
 	switch (core->boardnr) {
-	case CX88_BOARD_WINFAST_TV2000_XP_GLOBAL:
 	case CX88_BOARD_POWERCOLOR_REAL_ANGEL:
 	case CX88_BOARD_GENIATECH_X8000_MT:
 	case CX88_BOARD_KWORLD_ATSC_120:
@@ -2725,6 +2776,7 @@ static int cx88_xc2028_tuner_callback(struct cx88_core *core,
 	case CX88_BOARD_DVICO_FUSIONHDTV_DVB_T_PRO:
 	case CX88_BOARD_DVICO_FUSIONHDTV_5_PCI_NANO:
 		return cx88_dvico_xc2028_callback(core, command, arg);
+	case CX88_BOARD_WINFAST_TV2000_XP_GLOBAL:
 	case CX88_BOARD_WINFAST_DTV1800H:
 		return cx88_xc3028_winfast1800h_callback(core, command, arg);
 	}
@@ -2914,6 +2966,7 @@ static void cx88_card_setup_pre_i2c(struct cx88_core *core)
 		udelay(1000);
 		break;
 
+	case CX88_BOARD_WINFAST_TV2000_XP_GLOBAL:
 	case CX88_BOARD_WINFAST_DTV1800H:
 		/* GPIO 12 (xc3028 tuner reset) */
 		cx_set(MO_GP1_IO, 0x1010);
@@ -2950,6 +3003,7 @@ void cx88_setup_xc3028(struct cx88_core *core, struct xc2028_ctrl *ctl)
 	case CX88_BOARD_DVICO_FUSIONHDTV_5_PCI_NANO:
 		ctl->demod = XC3028_FE_OREN538;
 		break;
+	case CX88_BOARD_WINFAST_TV2000_XP_GLOBAL:
 	case CX88_BOARD_PROLINK_PV_GLOBAL_XTREME:
 	case CX88_BOARD_PROLINK_PV_8000GT:
 		/*
@@ -2993,6 +3047,8 @@ static void cx88_card_setup(struct cx88_core *core)
 		if (0 == core->i2c_rc)
 			gdi_eeprom(core, eeprom);
 		break;
+	case CX88_BOARD_LEADTEK_PVR2000:
+	case CX88_BOARD_WINFAST_DV2000:
 	case CX88_BOARD_WINFAST2000XP_EXPERT:
 		if (0 == core->i2c_rc)
 			leadtek_eeprom(core, eeprom);
-- 
cgit v1.2.3-59-g8ed1b


From b8bfb5fb348d939a96fc8f71996a2e5e48b4544b Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Sat, 13 Jun 2009 18:56:22 -0300
Subject: V4L/DVB (12071): gspca: fix NULL pointer deref in query_ctrl

gspca: fix NULL pointer deref in query_ctrl

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/gspca.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/media/video/gspca/gspca.c b/drivers/media/video/gspca/gspca.c
index f7e0355ad644..1e89600986c8 100644
--- a/drivers/media/video/gspca/gspca.c
+++ b/drivers/media/video/gspca/gspca.c
@@ -1042,13 +1042,11 @@ static int vidioc_queryctrl(struct file *file, void *priv,
 		for (i = 0; i < gspca_dev->sd_desc->nctrls; i++) {
 			if (gspca_dev->ctrl_dis & (1 << i))
 				continue;
-			if (ctrls->qctrl.id < id)
+			if (gspca_dev->sd_desc->ctrls[i].qctrl.id < id)
 				continue;
-			if (ctrls != NULL) {
-				if (gspca_dev->sd_desc->ctrls[i].qctrl.id
+			if (ctrls && gspca_dev->sd_desc->ctrls[i].qctrl.id
 					    > ctrls->qctrl.id)
-					continue;
-			}
+				continue;
 			ctrls = &gspca_dev->sd_desc->ctrls[i];
 		}
 	} else {
-- 
cgit v1.2.3-59-g8ed1b


From 02ab18b0f497bed623814677577b76cc97234085 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Sun, 14 Jun 2009 04:32:04 -0300
Subject: V4L/DVB (12072): gspca-ov519: add extra controls

This patch adds autobrightness (so that it can
be turned off to make the already present brightness
control work) and light frequency filtering controls.

The lightfreq control needed 2 different entries
in the ctrls array, as the number of options differs
depending on the sensor. Always one of the 2 entires is
disabled ofcourse.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/ov519.c | 214 ++++++++++++++++++++++++++++++++++++--
 include/linux/videodev2.h         |   3 +-
 2 files changed, 207 insertions(+), 10 deletions(-)

diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c
index 188866ac6cef..baa488dd33d8 100644
--- a/drivers/media/video/gspca/ov519.c
+++ b/drivers/media/video/gspca/ov519.c
@@ -65,6 +65,8 @@ struct sd {
 	__u8 colors;
 	__u8 hflip;
 	__u8 vflip;
+	__u8 autobrightness;
+	__u8 freq;
 
 	__u8 stopped;		/* Streaming is temporarily paused */
 
@@ -94,11 +96,17 @@ static int sd_sethflip(struct gspca_dev *gspca_dev, __s32 val);
 static int sd_gethflip(struct gspca_dev *gspca_dev, __s32 *val);
 static int sd_setvflip(struct gspca_dev *gspca_dev, __s32 val);
 static int sd_getvflip(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setautobrightness(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getautobrightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setfreq(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getfreq(struct gspca_dev *gspca_dev, __s32 *val);
 static void setbrightness(struct gspca_dev *gspca_dev);
 static void setcontrast(struct gspca_dev *gspca_dev);
 static void setcolors(struct gspca_dev *gspca_dev);
+static void setautobrightness(struct sd *sd);
+static void setfreq(struct sd *sd);
 
-static struct ctrl sd_ctrls[] = {
+static const struct ctrl sd_ctrls[] = {
 	{
 	    {
 		.id      = V4L2_CID_BRIGHTNESS,
@@ -141,7 +149,7 @@ static struct ctrl sd_ctrls[] = {
 	    .set = sd_setcolors,
 	    .get = sd_getcolors,
 	},
-/* next controls work with ov7670 only */
+/* The flip controls work with ov7670 only */
 #define HFLIP_IDX 3
 	{
 	    {
@@ -172,6 +180,51 @@ static struct ctrl sd_ctrls[] = {
 	    .set = sd_setvflip,
 	    .get = sd_getvflip,
 	},
+#define AUTOBRIGHT_IDX 5
+	{
+	    {
+		.id      = V4L2_CID_AUTOBRIGHTNESS,
+		.type    = V4L2_CTRL_TYPE_BOOLEAN,
+		.name    = "Auto Brightness",
+		.minimum = 0,
+		.maximum = 1,
+		.step    = 1,
+#define AUTOBRIGHT_DEF 1
+		.default_value = AUTOBRIGHT_DEF,
+	    },
+	    .set = sd_setautobrightness,
+	    .get = sd_getautobrightness,
+	},
+#define FREQ_IDX 6
+	{
+	    {
+		.id	 = V4L2_CID_POWER_LINE_FREQUENCY,
+		.type    = V4L2_CTRL_TYPE_MENU,
+		.name    = "Light frequency filter",
+		.minimum = 0,
+		.maximum = 2,	/* 0: 0, 1: 50Hz, 2:60Hz */
+		.step    = 1,
+#define FREQ_DEF 0
+		.default_value = FREQ_DEF,
+	    },
+	    .set = sd_setfreq,
+	    .get = sd_getfreq,
+	},
+#define OV7670_FREQ_IDX 7
+	{
+	    {
+		.id	 = V4L2_CID_POWER_LINE_FREQUENCY,
+		.type    = V4L2_CTRL_TYPE_MENU,
+		.name    = "Light frequency filter",
+		.minimum = 0,
+		.maximum = 3,	/* 0: 0, 1: 50Hz, 2:60Hz 3: Auto Hz */
+		.step    = 1,
+#define OV7670_FREQ_DEF 3
+		.default_value = OV7670_FREQ_DEF,
+	    },
+	    .set = sd_setfreq,
+	    .get = sd_getfreq,
+	},
 };
 
 static const struct v4l2_pix_format ov519_vga_mode[] = {
@@ -416,7 +469,7 @@ static const struct ov_i2c_regvals norm_6x30[] = {
 	{ 0x07, 0x2d }, /* Sharpness */
 	{ 0x0c, 0x20 },
 	{ 0x0d, 0x20 },
-	{ 0x0e, 0x20 },
+	{ 0x0e, 0xa0 }, /* Was 0x20, bit7 enables a 2x gain which we need */
 	{ 0x0f, 0x05 },
 	{ 0x10, 0x9a },
 	{ 0x11, 0x00 }, /* Pixel clock = fastest */
@@ -1659,9 +1712,21 @@ static int sd_config(struct gspca_dev *gspca_dev,
 	sd->colors = COLOR_DEF;
 	sd->hflip = HFLIP_DEF;
 	sd->vflip = VFLIP_DEF;
-	if (sd->sensor != SEN_OV7670)
-		gspca_dev->ctrl_dis = (1 << HFLIP_IDX)
-					| (1 << VFLIP_IDX);
+	sd->autobrightness = AUTOBRIGHT_DEF;
+	if (sd->sensor == SEN_OV7670) {
+		sd->freq = OV7670_FREQ_DEF;
+		gspca_dev->ctrl_dis = 1 << FREQ_IDX;
+	} else {
+		sd->freq = FREQ_DEF;
+		gspca_dev->ctrl_dis = (1 << HFLIP_IDX) | (1 << VFLIP_IDX) |
+				      (1 << OV7670_FREQ_IDX);
+	}
+	if (sd->sensor == SEN_OV7640 || sd->sensor == SEN_OV7670)
+		gspca_dev->ctrl_dis |= 1 << AUTOBRIGHT_IDX;
+	/* OV8610 Frequency filter control should work but needs testing */
+	if (sd->sensor == SEN_OV8610)
+		gspca_dev->ctrl_dis |= 1 << FREQ_IDX;
+
 	return 0;
 error:
 	PDEBUG(D_ERR, "OV519 Config failed");
@@ -2233,7 +2298,6 @@ static int set_ov_sensor_window(struct sd *sd)
 		msleep(10);	/* need to sleep between read and write to
 				 * same reg! */
 		i2c_w(sd, OV7670_REG_VREF, v);
-		sethvflip(sd);
 	} else {
 		i2c_w(sd, 0x17, hwsbase);
 		i2c_w(sd, 0x18, hwebase + (sd->gspca_dev.width >> hwscale));
@@ -2268,6 +2332,9 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	setcontrast(gspca_dev);
 	setbrightness(gspca_dev);
 	setcolors(gspca_dev);
+	sethvflip(sd);
+	setautobrightness(sd);
+	setfreq(sd);
 
 	ret = ov51x_restart(sd);
 	if (ret < 0)
@@ -2394,8 +2461,7 @@ static void setbrightness(struct gspca_dev *gspca_dev)
 		break;
 	case SEN_OV7620:
 		/* 7620 doesn't like manual changes when in auto mode */
-/*fixme
- *		if (!sd->auto_brt) */
+		if (!sd->autobrightness)
 			i2c_w(sd, OV7610_REG_BRT, val);
 		break;
 	case SEN_OV7670:
@@ -2482,6 +2548,70 @@ static void setcolors(struct gspca_dev *gspca_dev)
 	}
 }
 
+static void setautobrightness(struct sd *sd)
+{
+	if (sd->sensor == SEN_OV7640 || sd->sensor == SEN_OV7670)
+		return;
+
+	i2c_w_mask(sd, 0x2d, sd->autobrightness ? 0x10 : 0x00, 0x10);
+}
+
+static void setfreq(struct sd *sd)
+{
+	if (sd->sensor == SEN_OV7670) {
+		switch (sd->freq) {
+		case 0: /* Banding filter disabled */
+			i2c_w_mask(sd, OV7670_REG_COM8, 0, OV7670_COM8_BFILT);
+			break;
+		case 1: /* 50 hz */
+			i2c_w_mask(sd, OV7670_REG_COM8, OV7670_COM8_BFILT,
+				   OV7670_COM8_BFILT);
+			i2c_w_mask(sd, OV7670_REG_COM11, 0x08, 0x18);
+			break;
+		case 2: /* 60 hz */
+			i2c_w_mask(sd, OV7670_REG_COM8, OV7670_COM8_BFILT,
+				   OV7670_COM8_BFILT);
+			i2c_w_mask(sd, OV7670_REG_COM11, 0x00, 0x18);
+			break;
+		case 3: /* Auto hz */
+			i2c_w_mask(sd, OV7670_REG_COM8, OV7670_COM8_BFILT,
+				   OV7670_COM8_BFILT);
+			i2c_w_mask(sd, OV7670_REG_COM11, OV7670_COM11_HZAUTO,
+				   0x18);
+			break;
+		}
+	} else {
+		switch (sd->freq) {
+		case 0: /* Banding filter disabled */
+			i2c_w_mask(sd, 0x2d, 0x00, 0x04);
+			i2c_w_mask(sd, 0x2a, 0x00, 0x80);
+			break;
+		case 1: /* 50 hz (filter on and framerate adj) */
+			i2c_w_mask(sd, 0x2d, 0x04, 0x04);
+			i2c_w_mask(sd, 0x2a, 0x80, 0x80);
+			/* 20 fps -> 16.667 fps */
+			if (sd->sensor == SEN_OV6620 ||
+			    sd->sensor == SEN_OV6630)
+				i2c_w(sd, 0x2b, 0x5e);
+			else
+				i2c_w(sd, 0x2b, 0xac);
+			break;
+		case 2: /* 60 hz (filter on, ...) */
+			i2c_w_mask(sd, 0x2d, 0x04, 0x04);
+			if (sd->sensor == SEN_OV6620 ||
+			    sd->sensor == SEN_OV6630) {
+				/* 20 fps -> 15 fps */
+				i2c_w_mask(sd, 0x2a, 0x80, 0x80);
+				i2c_w(sd, 0x2b, 0xa8);
+			} else {
+				/* no framerate adj. */
+				i2c_w_mask(sd, 0x2a, 0x00, 0x80);
+			}
+			break;
+		}
+	}
+}
+
 static int sd_setbrightness(struct gspca_dev *gspca_dev, __s32 val)
 {
 	struct sd *sd = (struct sd *) gspca_dev;
@@ -2572,6 +2702,71 @@ static int sd_getvflip(struct gspca_dev *gspca_dev, __s32 *val)
 	return 0;
 }
 
+static int sd_setautobrightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->autobrightness = val;
+	if (gspca_dev->streaming)
+		setautobrightness(sd);
+	return 0;
+}
+
+static int sd_getautobrightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->autobrightness;
+	return 0;
+}
+
+static int sd_setfreq(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->freq = val;
+	if (gspca_dev->streaming)
+		setfreq(sd);
+	return 0;
+}
+
+static int sd_getfreq(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->freq;
+	return 0;
+}
+
+static int sd_querymenu(struct gspca_dev *gspca_dev,
+			struct v4l2_querymenu *menu)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	switch (menu->id) {
+	case V4L2_CID_POWER_LINE_FREQUENCY:
+		switch (menu->index) {
+		case 0:		/* V4L2_CID_POWER_LINE_FREQUENCY_DISABLED */
+			strcpy((char *) menu->name, "NoFliker");
+			return 0;
+		case 1:		/* V4L2_CID_POWER_LINE_FREQUENCY_50HZ */
+			strcpy((char *) menu->name, "50 Hz");
+			return 0;
+		case 2:		/* V4L2_CID_POWER_LINE_FREQUENCY_60HZ */
+			strcpy((char *) menu->name, "60 Hz");
+			return 0;
+		case 3:
+			if (sd->sensor != SEN_OV7670)
+				return -EINVAL;
+
+			strcpy((char *) menu->name, "Automatic");
+			return 0;
+		}
+		break;
+	}
+	return -EINVAL;
+}
+
 /* sub-driver description */
 static const struct sd_desc sd_desc = {
 	.name = MODULE_NAME,
@@ -2582,6 +2777,7 @@ static const struct sd_desc sd_desc = {
 	.start = sd_start,
 	.stopN = sd_stopN,
 	.pkt_scan = sd_pkt_scan,
+	.querymenu = sd_querymenu,
 };
 
 /* -- module initialisation -- */
diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h
index f24eceecc5a6..772d226cb5ca 100644
--- a/include/linux/videodev2.h
+++ b/include/linux/videodev2.h
@@ -894,9 +894,10 @@ enum v4l2_colorfx {
 	V4L2_COLORFX_BW		= 1,
 	V4L2_COLORFX_SEPIA	= 2,
 };
+#define V4L2_CID_AUTOBRIGHTNESS			(V4L2_CID_BASE+32)
 
 /* last CID + 1 */
-#define V4L2_CID_LASTP1                         (V4L2_CID_BASE+32)
+#define V4L2_CID_LASTP1                         (V4L2_CID_BASE+33)
 
 /*  MPEG-class control IDs defined by V4L2 */
 #define V4L2_CID_MPEG_BASE 			(V4L2_CTRL_CLASS_MPEG | 0x900)
-- 
cgit v1.2.3-59-g8ed1b


From 7d9713735d7537baf2b00be806a8de08a5c9f11b Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Sun, 14 Jun 2009 05:28:17 -0300
Subject: V4L/DVB (12073): gspca_ov519: limit ov6630 qvif uv swap fix to
 ov66308AF

The fix for the UV swapping in qcif mode with the ov6630, which I did
to fix this issue on a ov518 cam with an ov66308AF, causes UV swapping in
qcif with another cam of mine with the ov518 and an ov66308AE, so this
patch changes the code to differentiate between the ov66308AF and other
ov6630 versions, and restricts the UV swap fix to the ov66308AF.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/ov519.c | 38 ++++++++++++++++++++++++--------------
 1 file changed, 24 insertions(+), 14 deletions(-)

diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c
index baa488dd33d8..2d5d95f02779 100644
--- a/drivers/media/video/gspca/ov519.c
+++ b/drivers/media/video/gspca/ov519.c
@@ -77,12 +77,13 @@ struct sd {
 #define SEN_UNKNOWN 0
 #define SEN_OV6620 1
 #define SEN_OV6630 2
-#define SEN_OV7610 3
-#define SEN_OV7620 4
-#define SEN_OV7640 5
-#define SEN_OV7670 6
-#define SEN_OV76BE 7
-#define SEN_OV8610 8
+#define SEN_OV66308AF 3
+#define SEN_OV7610 4
+#define SEN_OV7620 5
+#define SEN_OV7640 6
+#define SEN_OV7670 7
+#define SEN_OV76BE 8
+#define SEN_OV8610 9
 };
 
 /* V4L2 controls supported by the driver */
@@ -1415,13 +1416,14 @@ static int ov6xx0_configure(struct sd *sd)
 		break;
 	case 0x01:
 		sd->sensor = SEN_OV6620;
+		PDEBUG(D_PROBE, "Sensor is an OV6620");
 		break;
 	case 0x02:
 		sd->sensor = SEN_OV6630;
 		PDEBUG(D_PROBE, "Sensor is an OV66308AE");
 		break;
 	case 0x03:
-		sd->sensor = SEN_OV6630;
+		sd->sensor = SEN_OV66308AF;
 		PDEBUG(D_PROBE, "Sensor is an OV66308AF");
 		break;
 	case 0x90:
@@ -1745,6 +1747,7 @@ static int sd_init(struct gspca_dev *gspca_dev)
 			return -EIO;
 		break;
 	case SEN_OV6630:
+	case SEN_OV66308AF:
 		if (write_i2c_regvals(sd, norm_6x30, ARRAY_SIZE(norm_6x30)))
 			return -EIO;
 		break;
@@ -2081,6 +2084,7 @@ static int mode_init_ov_sensor_regs(struct sd *sd)
 		break;
 	case SEN_OV6620:
 	case SEN_OV6630:
+	case SEN_OV66308AF:
 		i2c_w_mask(sd, 0x14, qvga ? 0x20 : 0x00, 0x20);
 		break;
 	default:
@@ -2101,7 +2105,8 @@ static int mode_init_ov_sensor_regs(struct sd *sd)
 
 	/* OV7640 is 8-bit only */
 
-	if (sd->sensor != SEN_OV6630 && sd->sensor != SEN_OV7640)
+	if (sd->sensor != SEN_OV6630 && sd->sensor != SEN_OV66308AF &&
+					sd->sensor != SEN_OV7640)
 		i2c_w_mask(sd, 0x13, 0x00, 0x20);
 
 	/******** Clock programming ********/
@@ -2188,15 +2193,14 @@ static int set_ov_sensor_window(struct sd *sd)
 		break;
 	case SEN_OV6620:
 	case SEN_OV6630:
+	case SEN_OV66308AF:
 		hwsbase = 0x38;
 		hwebase = 0x3a;
 		vwsbase = 0x05;
 		vwebase = 0x06;
-		if (qvga) {
+		if (sd->sensor == SEN_OV66308AF && qvga)
 			/* HDG: this fixes U and V getting swapped */
-			hwsbase--;
-			vwsbase--;
-		}
+			hwsbase++;
 		break;
 	case SEN_OV7620:
 		hwsbase = 0x2f;		/* From 7620.SET (spec is wrong) */
@@ -2220,6 +2224,7 @@ static int set_ov_sensor_window(struct sd *sd)
 	switch (sd->sensor) {
 	case SEN_OV6620:
 	case SEN_OV6630:
+	case SEN_OV66308AF:
 		if (qvga) {		/* QCIF */
 			hwscale = 0;
 			vwscale = 0;
@@ -2456,6 +2461,7 @@ static void setbrightness(struct gspca_dev *gspca_dev)
 	case SEN_OV76BE:
 	case SEN_OV6620:
 	case SEN_OV6630:
+	case SEN_OV66308AF:
 	case SEN_OV7640:
 		i2c_w(sd, OV7610_REG_BRT, val);
 		break;
@@ -2484,6 +2490,7 @@ static void setcontrast(struct gspca_dev *gspca_dev)
 		i2c_w(sd, OV7610_REG_CNT, val);
 		break;
 	case SEN_OV6630:
+	case SEN_OV66308AF:
 		i2c_w_mask(sd, OV7610_REG_CNT, val >> 4, 0x0f);
 		break;
 	case SEN_OV8610: {
@@ -2528,6 +2535,7 @@ static void setcolors(struct gspca_dev *gspca_dev)
 	case SEN_OV76BE:
 	case SEN_OV6620:
 	case SEN_OV6630:
+	case SEN_OV66308AF:
 		i2c_w(sd, OV7610_REG_SAT, val);
 		break;
 	case SEN_OV7620:
@@ -2591,7 +2599,8 @@ static void setfreq(struct sd *sd)
 			i2c_w_mask(sd, 0x2a, 0x80, 0x80);
 			/* 20 fps -> 16.667 fps */
 			if (sd->sensor == SEN_OV6620 ||
-			    sd->sensor == SEN_OV6630)
+			    sd->sensor == SEN_OV6630 ||
+			    sd->sensor == SEN_OV66308AF)
 				i2c_w(sd, 0x2b, 0x5e);
 			else
 				i2c_w(sd, 0x2b, 0xac);
@@ -2599,7 +2608,8 @@ static void setfreq(struct sd *sd)
 		case 2: /* 60 hz (filter on, ...) */
 			i2c_w_mask(sd, 0x2d, 0x04, 0x04);
 			if (sd->sensor == SEN_OV6620 ||
-			    sd->sensor == SEN_OV6630) {
+			    sd->sensor == SEN_OV6630 ||
+			    sd->sensor == SEN_OV66308AF) {
 				/* 20 fps -> 15 fps */
 				i2c_w_mask(sd, 0x2a, 0x80, 0x80);
 				i2c_w(sd, 0x2b, 0xa8);
-- 
cgit v1.2.3-59-g8ed1b


From 124cc9c0c8acc77ac2f1114ee7eea961334020ba Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Sun, 14 Jun 2009 05:48:00 -0300
Subject: V4L/DVB (12074): gspca_ov519: Add 320x240 and 160x120 support for cif
 sensor cams

gspca_ov519: Add 320x240 and 160x120 support for cif sensor cams

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/ov519.c | 33 ++++++++++++++++++++++++++++++---
 1 file changed, 30 insertions(+), 3 deletions(-)

diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c
index 2d5d95f02779..55db32c95bea 100644
--- a/drivers/media/video/gspca/ov519.c
+++ b/drivers/media/video/gspca/ov519.c
@@ -241,11 +241,21 @@ static const struct v4l2_pix_format ov519_vga_mode[] = {
 		.priv = 0},
 };
 static const struct v4l2_pix_format ov519_sif_mode[] = {
+	{160, 120, V4L2_PIX_FMT_JPEG, V4L2_FIELD_NONE,
+		.bytesperline = 160,
+		.sizeimage = 160 * 120 * 3 / 8 + 590,
+		.colorspace = V4L2_COLORSPACE_JPEG,
+		.priv = 3},
 	{176, 144, V4L2_PIX_FMT_JPEG, V4L2_FIELD_NONE,
 		.bytesperline = 176,
 		.sizeimage = 176 * 144 * 3 / 8 + 590,
 		.colorspace = V4L2_COLORSPACE_JPEG,
 		.priv = 1},
+	{320, 240, V4L2_PIX_FMT_JPEG, V4L2_FIELD_NONE,
+		.bytesperline = 320,
+		.sizeimage = 320 * 240 * 3 / 8 + 590,
+		.colorspace = V4L2_COLORSPACE_JPEG,
+		.priv = 2},
 	{352, 288, V4L2_PIX_FMT_JPEG, V4L2_FIELD_NONE,
 		.bytesperline = 352,
 		.sizeimage = 352 * 288 * 3 / 8 + 590,
@@ -266,11 +276,21 @@ static const struct v4l2_pix_format ov518_vga_mode[] = {
 		.priv = 0},
 };
 static const struct v4l2_pix_format ov518_sif_mode[] = {
+	{160, 120, V4L2_PIX_FMT_OV518, V4L2_FIELD_NONE,
+		.bytesperline = 160,
+		.sizeimage = 40000,
+		.colorspace = V4L2_COLORSPACE_JPEG,
+		.priv = 3},
 	{176, 144, V4L2_PIX_FMT_OV518, V4L2_FIELD_NONE,
 		.bytesperline = 176,
 		.sizeimage = 40000,
 		.colorspace = V4L2_COLORSPACE_JPEG,
 		.priv = 1},
+	{320, 240, V4L2_PIX_FMT_OV518, V4L2_FIELD_NONE,
+		.bytesperline = 320,
+		.sizeimage = 320 * 240 * 3 / 8 + 590,
+		.colorspace = V4L2_COLORSPACE_JPEG,
+		.priv = 2},
 	{352, 288, V4L2_PIX_FMT_OV518, V4L2_FIELD_NONE,
 		.bytesperline = 352,
 		.sizeimage = 352 * 288 * 3 / 8 + 590,
@@ -2039,7 +2059,7 @@ static int mode_init_ov_sensor_regs(struct sd *sd)
 	int qvga;
 
 	gspca_dev = &sd->gspca_dev;
-	qvga = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].priv;
+	qvga = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].priv & 1;
 
 	/******** Mode (VGA/QVGA) and sensor specific regs ********/
 	switch (sd->sensor) {
@@ -2168,13 +2188,14 @@ static void sethvflip(struct sd *sd)
 static int set_ov_sensor_window(struct sd *sd)
 {
 	struct gspca_dev *gspca_dev;
-	int qvga;
+	int qvga, crop;
 	int hwsbase, hwebase, vwsbase, vwebase, hwscale, vwscale;
 	int ret, hstart, hstop, vstop, vstart;
 	__u8 v;
 
 	gspca_dev = &sd->gspca_dev;
-	qvga = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].priv;
+	qvga = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].priv & 1;
+	crop = gspca_dev->cam.cam_mode[(int) gspca_dev->curr_mode].priv & 2;
 
 	/* The different sensor ICs handle setting up of window differently.
 	 * IF YOU SET IT WRONG, YOU WILL GET ALL ZERO ISOC DATA FROM OV51x!! */
@@ -2201,6 +2222,12 @@ static int set_ov_sensor_window(struct sd *sd)
 		if (sd->sensor == SEN_OV66308AF && qvga)
 			/* HDG: this fixes U and V getting swapped */
 			hwsbase++;
+		if (crop) {
+			hwsbase += 8;
+			hwebase += 8;
+			vwsbase += 11;
+			vwebase += 11;
+		}
 		break;
 	case SEN_OV7620:
 		hwsbase = 0x2f;		/* From 7620.SET (spec is wrong) */
-- 
cgit v1.2.3-59-g8ed1b


From 92918a53ee74bb326430aaa958caa0cf111b54b1 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Sun, 14 Jun 2009 06:21:35 -0300
Subject: V4L/DVB (12075): gspca_ov519: check ov518 packet numbers

Check ov518 packet numbers to detect dropped packets.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/ov519.c | 29 +++++++++++++++++++++++------
 1 file changed, 23 insertions(+), 6 deletions(-)

diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c
index 55db32c95bea..c2982137dc21 100644
--- a/drivers/media/video/gspca/ov519.c
+++ b/drivers/media/video/gspca/ov519.c
@@ -50,6 +50,8 @@ static int i2c_detect_tries = 10;
 struct sd {
 	struct gspca_dev gspca_dev;		/* !! must be the first item */
 
+	__u8 packet_nr;
+
 	char bridge;
 #define BRIDGE_OV511		0
 #define BRIDGE_OV511PLUS	1
@@ -2391,18 +2393,33 @@ static void ov518_pkt_scan(struct gspca_dev *gspca_dev,
 			__u8 *data,			/* isoc packet */
 			int len)			/* iso packet length */
 {
-	PDEBUG(D_STREAM, "ov518_pkt_scan: %d bytes", len);
-
-	if (len & 7) {
-		len--;
-		PDEBUG(D_STREAM, "packet number: %d\n", (int)data[len]);
-	}
+	struct sd *sd = (struct sd *) gspca_dev;
 
 	/* A false positive here is likely, until OVT gives me
 	 * the definitive SOF/EOF format */
 	if ((!(data[0] | data[1] | data[2] | data[3] | data[5])) && data[6]) {
 		gspca_frame_add(gspca_dev, LAST_PACKET, frame, data, 0);
 		gspca_frame_add(gspca_dev, FIRST_PACKET, frame, data, 0);
+		sd->packet_nr = 0;
+	}
+
+	if (gspca_dev->last_packet_type == DISCARD_PACKET)
+		return;
+
+	/* Does this device use packet numbers ? */
+	if (len & 7) {
+		len--;
+		if (sd->packet_nr == data[len])
+			sd->packet_nr++;
+		/* The last few packets of the frame (which are all 0's
+		   except that they may contain part of the footer), are
+		   numbered 0 */
+		else if (sd->packet_nr == 0 || data[len]) {
+			PDEBUG(D_ERR, "Invalid packet nr: %d (expect: %d)",
+				(int)data[len], (int)sd->packet_nr);
+			gspca_dev->last_packet_type = DISCARD_PACKET;
+			return;
+		}
 	}
 
 	/* intermediate packet */
-- 
cgit v1.2.3-59-g8ed1b


From 9e4d82588174e68abe8e3568202f0b530415661f Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Sun, 14 Jun 2009 06:25:06 -0300
Subject: V4L/DVB (12076): gspca_ov519: Fix led inversion with some cams

My ov519 cam has it led inverted, the same has been
reported on the ov51x-jpeg list for another
creative cam. This patch fixes this without changing
the behaviour for other cams.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/ov519.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c
index c2982137dc21..c2c087222f17 100644
--- a/drivers/media/video/gspca/ov519.c
+++ b/drivers/media/video/gspca/ov519.c
@@ -58,6 +58,10 @@ struct sd {
 #define BRIDGE_OV518		2
 #define BRIDGE_OV518PLUS	3
 #define BRIDGE_OV519		4
+#define BRIDGE_MASK		7
+
+	char invert_led;
+#define BRIDGE_INVERT_LED	8
 
 	/* Determined by sensor type */
 	__u8 sif;
@@ -1468,6 +1472,9 @@ static int ov6xx0_configure(struct sd *sd)
 /* Turns on or off the LED. Only has an effect with OV511+/OV518(+)/OV519 */
 static void ov51x_led_control(struct sd *sd, int on)
 {
+	if (sd->invert_led)
+		on = !on;
+
 	switch (sd->bridge) {
 	/* OV511 has no LED control */
 	case BRIDGE_OV511PLUS:
@@ -1650,7 +1657,8 @@ static int sd_config(struct gspca_dev *gspca_dev,
 	struct cam *cam;
 	int ret = 0;
 
-	sd->bridge = id->driver_info;
+	sd->bridge = id->driver_info & BRIDGE_MASK;
+	sd->invert_led = id->driver_info & BRIDGE_INVERT_LED;
 
 	switch (sd->bridge) {
 	case BRIDGE_OV518:
@@ -2840,8 +2848,10 @@ static const __devinitdata struct usb_device_id device_table[] = {
 	{USB_DEVICE(0x041e, 0x405f), .driver_info = BRIDGE_OV519 },
 	{USB_DEVICE(0x041e, 0x4060), .driver_info = BRIDGE_OV519 },
 	{USB_DEVICE(0x041e, 0x4061), .driver_info = BRIDGE_OV519 },
-	{USB_DEVICE(0x041e, 0x4064), .driver_info = BRIDGE_OV519 },
-	{USB_DEVICE(0x041e, 0x4068), .driver_info = BRIDGE_OV519 },
+	{USB_DEVICE(0x041e, 0x4064),
+	 .driver_info = BRIDGE_OV519 | BRIDGE_INVERT_LED },
+	{USB_DEVICE(0x041e, 0x4068),
+	 .driver_info = BRIDGE_OV519 | BRIDGE_INVERT_LED },
 	{USB_DEVICE(0x045e, 0x028c), .driver_info = BRIDGE_OV519 },
 	{USB_DEVICE(0x054c, 0x0154), .driver_info = BRIDGE_OV519 },
 	{USB_DEVICE(0x054c, 0x0155), .driver_info = BRIDGE_OV519 },
-- 
cgit v1.2.3-59-g8ed1b


From 80142efa715581c06d01b37f299a240309699ff4 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Sun, 14 Jun 2009 06:26:49 -0300
Subject: V4L/DVB (12077): gspca_ov519: Fix 320x240 with ov7660 sensor

As reported on the ov51x-jpeg list, and as I can confirm with my own cam
the ov7670 in 320x240 has a number of broken columns of pixels
at the left of the picture. This was not present in the old
driver as it always used 640x480 and did software
downscaling (took me a while to figure that one out).
The fix adds a sensor specific if in so far sensor
neutral code :( But this is the only way to fix this,
this cannot be fixed by only changing sensor registers.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/ov519.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c
index c2c087222f17..c7e88cb0095a 100644
--- a/drivers/media/video/gspca/ov519.c
+++ b/drivers/media/video/gspca/ov519.c
@@ -1977,7 +1977,11 @@ static int ov519_mode_init_regs(struct sd *sd)
 
 	reg_w(sd, OV519_R10_H_SIZE,	sd->gspca_dev.width >> 4);
 	reg_w(sd, OV519_R11_V_SIZE,	sd->gspca_dev.height >> 3);
-	reg_w(sd, OV519_R12_X_OFFSETL,	0x00);
+	if (sd->sensor == SEN_OV7670 &&
+	    sd->gspca_dev.cam.cam_mode[sd->gspca_dev.curr_mode].priv)
+		reg_w(sd, OV519_R12_X_OFFSETL, 0x04);
+	else
+		reg_w(sd, OV519_R12_X_OFFSETL, 0x00);
 	reg_w(sd, OV519_R13_X_OFFSETH,	0x00);
 	reg_w(sd, OV519_R14_Y_OFFSETL,	0x00);
 	reg_w(sd, OV519_R15_Y_OFFSETH,	0x00);
@@ -2314,7 +2318,7 @@ static int set_ov_sensor_window(struct sd *sd)
 		if (qvga) {		/* QVGA from ov7670.c by
 					 * Jonathan Corbet */
 			hstart = 164;
-			hstop = 20;
+			hstop = 28;
 			vstart = 14;
 			vstop = 494;
 		} else {		/* VGA */
-- 
cgit v1.2.3-59-g8ed1b


From f5cee95c2e4c56b50cdb8edd33cf04902946cd25 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Sun, 14 Jun 2009 06:32:52 -0300
Subject: V4L/DVB (12078): gspca_ov519: Better default contrast for ov6630

Hmm, another one with an extra if (life sucks) the
default contrast really is no good for the ov6630, it
isn't even high enough in full daylight, this gives
the ov6630 a different initial value for a better out
of the box experience.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/ov519.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c
index c7e88cb0095a..9d4b69dbf966 100644
--- a/drivers/media/video/gspca/ov519.c
+++ b/drivers/media/video/gspca/ov519.c
@@ -1740,7 +1740,10 @@ static int sd_config(struct gspca_dev *gspca_dev,
 		break;
 	}
 	sd->brightness = BRIGHTNESS_DEF;
-	sd->contrast = CONTRAST_DEF;
+	if (sd->sensor == SEN_OV6630 || sd->sensor == SEN_OV66308AF)
+		sd->contrast = 200; /* The default is too low for the ov6630 */
+	else
+		sd->contrast = CONTRAST_DEF;
 	sd->colors = COLOR_DEF;
 	sd->hflip = HFLIP_DEF;
 	sd->vflip = VFLIP_DEF;
-- 
cgit v1.2.3-59-g8ed1b


From 1876bb923c98c605eca69f0bfe295f7b5f5eba28 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Sun, 14 Jun 2009 06:45:50 -0300
Subject: V4L/DVB (12079): gspca_ov519: add support for the ov511 bridge

gspca_ov519: add support for the ov511 bridge

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/ov519.c | 533 +++++++++++++++++++++++++++++++++++++-
 include/linux/videodev2.h         |   1 +
 2 files changed, 521 insertions(+), 13 deletions(-)

diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c
index 9d4b69dbf966..1f8e2613ecc5 100644
--- a/drivers/media/video/gspca/ov519.c
+++ b/drivers/media/video/gspca/ov519.c
@@ -76,8 +76,8 @@ struct sd {
 
 	__u8 stopped;		/* Streaming is temporarily paused */
 
-	__u8 frame_rate;	/* current Framerate (OV519 only) */
-	__u8 clockdiv;		/* clockdiv override for OV519 only */
+	__u8 frame_rate;	/* current Framerate */
+	__u8 clockdiv;		/* clockdiv override */
 
 	char sensor;		/* Type of image sensor chip (SEN_*) */
 #define SEN_UNKNOWN 0
@@ -304,17 +304,77 @@ static const struct v4l2_pix_format ov518_sif_mode[] = {
 		.priv = 0},
 };
 
+static const struct v4l2_pix_format ov511_vga_mode[] = {
+	{320, 240, V4L2_PIX_FMT_OV511, V4L2_FIELD_NONE,
+		.bytesperline = 320,
+		.sizeimage = 320 * 240 * 3,
+		.colorspace = V4L2_COLORSPACE_JPEG,
+		.priv = 1},
+	{640, 480, V4L2_PIX_FMT_OV511, V4L2_FIELD_NONE,
+		.bytesperline = 640,
+		.sizeimage = 640 * 480 * 2,
+		.colorspace = V4L2_COLORSPACE_JPEG,
+		.priv = 0},
+};
+static const struct v4l2_pix_format ov511_sif_mode[] = {
+	{160, 120, V4L2_PIX_FMT_OV511, V4L2_FIELD_NONE,
+		.bytesperline = 160,
+		.sizeimage = 40000,
+		.colorspace = V4L2_COLORSPACE_JPEG,
+		.priv = 3},
+	{176, 144, V4L2_PIX_FMT_OV511, V4L2_FIELD_NONE,
+		.bytesperline = 176,
+		.sizeimage = 40000,
+		.colorspace = V4L2_COLORSPACE_JPEG,
+		.priv = 1},
+	{320, 240, V4L2_PIX_FMT_OV511, V4L2_FIELD_NONE,
+		.bytesperline = 320,
+		.sizeimage = 320 * 240 * 3,
+		.colorspace = V4L2_COLORSPACE_JPEG,
+		.priv = 2},
+	{352, 288, V4L2_PIX_FMT_OV511, V4L2_FIELD_NONE,
+		.bytesperline = 352,
+		.sizeimage = 352 * 288 * 3,
+		.colorspace = V4L2_COLORSPACE_JPEG,
+		.priv = 0},
+};
 
 /* Registers common to OV511 / OV518 */
+#define R51x_FIFO_PSIZE			0x30	/* 2 bytes wide w/ OV518(+) */
 #define R51x_SYS_RESET          	0x50
+	/* Reset type flags */
+	#define	OV511_RESET_OMNICE	0x08
 #define R51x_SYS_INIT         		0x53
 #define R51x_SYS_SNAP			0x52
 #define R51x_SYS_CUST_ID		0x5F
 #define R51x_COMP_LUT_BEGIN		0x80
 
 /* OV511 Camera interface register numbers */
+#define R511_CAM_DELAY			0x10
+#define R511_CAM_EDGE			0x11
+#define R511_CAM_PXCNT			0x12
+#define R511_CAM_LNCNT			0x13
+#define R511_CAM_PXDIV			0x14
+#define R511_CAM_LNDIV			0x15
+#define R511_CAM_UV_EN			0x16
+#define R511_CAM_LINE_MODE		0x17
+#define R511_CAM_OPTS			0x18
+
+#define R511_SNAP_FRAME			0x19
+#define R511_SNAP_PXCNT			0x1A
+#define R511_SNAP_LNCNT			0x1B
+#define R511_SNAP_PXDIV			0x1C
+#define R511_SNAP_LNDIV			0x1D
+#define R511_SNAP_UV_EN			0x1E
+#define R511_SNAP_UV_EN			0x1E
+#define R511_SNAP_OPTS			0x1F
+
+#define R511_DRAM_FLOW_CTL		0x20
+#define R511_FIFO_OPTS			0x31
+#define R511_I2C_CTL			0x40
 #define R511_SYS_LED_CTL		0x55	/* OV511+ only */
-#define	OV511_RESET_NOREGS		0x3F	/* All but OV511 & regs */
+#define R511_COMP_EN			0x78
+#define R511_COMP_LUT_EN		0x79
 
 /* OV518 Camera interface register numbers */
 #define R518_GPIO_OUT			0x56	/* OV518(+) only */
@@ -1079,13 +1139,128 @@ static int ov518_reg_w32(struct sd *sd, __u16 index, u32 value, int n)
 	return ret;
 }
 
+static int ov511_i2c_w(struct sd *sd, __u8 reg, __u8 value)
+{
+	int rc, retries;
+
+	PDEBUG(D_USBO, "i2c 0x%02x -> [0x%02x]", value, reg);
+
+	/* Three byte write cycle */
+	for (retries = 6; ; ) {
+		/* Select camera register */
+		rc = reg_w(sd, R51x_I2C_SADDR_3, reg);
+		if (rc < 0)
+			return rc;
+
+		/* Write "value" to I2C data port of OV511 */
+		rc = reg_w(sd, R51x_I2C_DATA, value);
+		if (rc < 0)
+			return rc;
+
+		/* Initiate 3-byte write cycle */
+		rc = reg_w(sd, R511_I2C_CTL, 0x01);
+		if (rc < 0)
+			return rc;
+
+		do
+			rc = reg_r(sd, R511_I2C_CTL);
+		while (rc > 0 && ((rc & 1) == 0)); /* Retry until idle */
+
+		if (rc < 0)
+			return rc;
+
+		if ((rc & 2) == 0) /* Ack? */
+			break;
+		if (--retries < 0) {
+			PDEBUG(D_USBO, "i2c write retries exhausted");
+			return -1;
+		}
+	}
+
+	return 0;
+}
+
+static int ov511_i2c_r(struct sd *sd, __u8 reg)
+{
+	int rc, value, retries;
+
+	/* Two byte write cycle */
+	for (retries = 6; ; ) {
+		/* Select camera register */
+		rc = reg_w(sd, R51x_I2C_SADDR_2, reg);
+		if (rc < 0)
+			return rc;
+
+		/* Initiate 2-byte write cycle */
+		rc = reg_w(sd, R511_I2C_CTL, 0x03);
+		if (rc < 0)
+			return rc;
+
+		do
+			rc = reg_r(sd, R511_I2C_CTL);
+		while (rc > 0 && ((rc & 1) == 0)); /* Retry until idle */
+
+		if (rc < 0)
+			return rc;
+
+		if ((rc & 2) == 0) /* Ack? */
+			break;
+
+		/* I2C abort */
+		reg_w(sd, R511_I2C_CTL, 0x10);
+
+		if (--retries < 0) {
+			PDEBUG(D_USBI, "i2c write retries exhausted");
+			return -1;
+		}
+	}
+
+	/* Two byte read cycle */
+	for (retries = 6; ; ) {
+		/* Initiate 2-byte read cycle */
+		rc = reg_w(sd, R511_I2C_CTL, 0x05);
+		if (rc < 0)
+			return rc;
+
+		do
+			rc = reg_r(sd, R511_I2C_CTL);
+		while (rc > 0 && ((rc & 1) == 0)); /* Retry until idle */
+
+		if (rc < 0)
+			return rc;
+
+		if ((rc & 2) == 0) /* Ack? */
+			break;
+
+		/* I2C abort */
+		rc = reg_w(sd, R511_I2C_CTL, 0x10);
+		if (rc < 0)
+			return rc;
+
+		if (--retries < 0) {
+			PDEBUG(D_USBI, "i2c read retries exhausted");
+			return -1;
+		}
+	}
+
+	value = reg_r(sd, R51x_I2C_DATA);
+
+	PDEBUG(D_USBI, "i2c [0x%02X] -> 0x%02X", reg, value);
+
+	/* This is needed to make i2c_w() work */
+	rc = reg_w(sd, R511_I2C_CTL, 0x05);
+	if (rc < 0)
+		return rc;
+
+	return value;
+}
 
 /*
  * The OV518 I2C I/O procedure is different, hence, this function.
  * This is normally only called from i2c_w(). Note that this function
  * always succeeds regardless of whether the sensor is present and working.
  */
-static int i2c_w(struct sd *sd,
+static int ov518_i2c_w(struct sd *sd,
 		__u8 reg,
 		__u8 value)
 {
@@ -1120,7 +1295,7 @@ static int i2c_w(struct sd *sd,
  * This is normally only called from i2c_r(). Note that this function
  * always succeeds regardless of whether the sensor is present and working.
  */
-static int i2c_r(struct sd *sd, __u8 reg)
+static int ov518_i2c_r(struct sd *sd, __u8 reg)
 {
 	int rc, value;
 
@@ -1143,6 +1318,34 @@ static int i2c_r(struct sd *sd, __u8 reg)
 	return value;
 }
 
+static int i2c_w(struct sd *sd, __u8 reg, __u8 value)
+{
+	switch (sd->bridge) {
+	case BRIDGE_OV511:
+	case BRIDGE_OV511PLUS:
+		return ov511_i2c_w(sd, reg, value);
+	case BRIDGE_OV518:
+	case BRIDGE_OV518PLUS:
+	case BRIDGE_OV519:
+		return ov518_i2c_w(sd, reg, value);
+	}
+	return -1; /* Should never happen */
+}
+
+static int i2c_r(struct sd *sd, __u8 reg)
+{
+	switch (sd->bridge) {
+	case BRIDGE_OV511:
+	case BRIDGE_OV511PLUS:
+		return ov511_i2c_r(sd, reg);
+	case BRIDGE_OV518:
+	case BRIDGE_OV518PLUS:
+	case BRIDGE_OV519:
+		return ov518_i2c_r(sd, reg);
+	}
+	return -1; /* Should never happen */
+}
+
 /* Writes bits at positions specified by mask to an I2C reg. Bits that are in
  * the same position as 1's in "mask" are cleared and set to "value". Bits
  * that are in the same position as 0's in "mask" are preserved, regardless
@@ -1490,9 +1693,31 @@ static void ov51x_led_control(struct sd *sd, int on)
 	}
 }
 
-/* OV518 quantization tables are 8x4 (instead of 8x8) */
-static int ov518_upload_quan_tables(struct sd *sd)
+static int ov51x_upload_quan_tables(struct sd *sd)
 {
+	const unsigned char yQuanTable511[] = {
+		0, 1, 1, 2, 2, 3, 3, 4,
+		1, 1, 1, 2, 2, 3, 4, 4,
+		1, 1, 2, 2, 3, 4, 4, 4,
+		2, 2, 2, 3, 4, 4, 4, 4,
+		2, 2, 3, 4, 4, 5, 5, 5,
+		3, 3, 4, 4, 5, 5, 5, 5,
+		3, 4, 4, 4, 5, 5, 5, 5,
+		4, 4, 4, 4, 5, 5, 5, 5
+	};
+
+	const unsigned char uvQuanTable511[] = {
+		0, 2, 2, 3, 4, 4, 4, 4,
+		2, 2, 2, 4, 4, 4, 4, 4,
+		2, 2, 3, 4, 4, 4, 4, 4,
+		3, 4, 4, 4, 4, 4, 4, 4,
+		4, 4, 4, 4, 4, 4, 4, 4,
+		4, 4, 4, 4, 4, 4, 4, 4,
+		4, 4, 4, 4, 4, 4, 4, 4,
+		4, 4, 4, 4, 4, 4, 4, 4
+	};
+
+	/* OV518 quantization tables are 8x4 (instead of 8x8) */
 	const unsigned char yQuanTable518[] = {
 		5, 4, 5, 6, 6, 7, 7, 7,
 		5, 5, 5, 5, 6, 7, 7, 7,
@@ -1507,14 +1732,23 @@ static int ov518_upload_quan_tables(struct sd *sd)
 		7, 7, 7, 7, 7, 7, 8, 8
 	};
 
-	const unsigned char *pYTable = yQuanTable518;
-	const unsigned char *pUVTable = uvQuanTable518;
+	const unsigned char *pYTable, *pUVTable;
 	unsigned char val0, val1;
-	int i, rc, reg = R51x_COMP_LUT_BEGIN;
+	int i, size, rc, reg = R51x_COMP_LUT_BEGIN;
 
 	PDEBUG(D_PROBE, "Uploading quantization tables");
 
-	for (i = 0; i < 16; i++) {
+	if (sd->bridge == BRIDGE_OV511 || sd->bridge == BRIDGE_OV511PLUS) {
+		pYTable = yQuanTable511;
+		pUVTable = uvQuanTable511;
+		size  = 32;
+	} else {
+		pYTable = yQuanTable518;
+		pUVTable = uvQuanTable518;
+		size  = 16;
+	}
+
+	for (i = 0; i < size; i++) {
 		val0 = *pYTable++;
 		val1 = *pYTable++;
 		val0 &= 0x0f;
@@ -1529,7 +1763,7 @@ static int ov518_upload_quan_tables(struct sd *sd)
 		val0 &= 0x0f;
 		val1 &= 0x0f;
 		val0 |= val1 << 4;
-		rc = reg_w(sd, reg + 16, val0);
+		rc = reg_w(sd, reg + size, val0);
 		if (rc < 0)
 			return rc;
 
@@ -1539,6 +1773,87 @@ static int ov518_upload_quan_tables(struct sd *sd)
 	return 0;
 }
 
+/* This initializes the OV511/OV511+ and the sensor */
+static int ov511_configure(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	int rc;
+
+	/* For 511 and 511+ */
+	const struct ov_regvals init_511[] = {
+		{ R51x_SYS_RESET,	0x7f },
+		{ R51x_SYS_INIT,	0x01 },
+		{ R51x_SYS_RESET,	0x7f },
+		{ R51x_SYS_INIT,	0x01 },
+		{ R51x_SYS_RESET,	0x3f },
+		{ R51x_SYS_INIT,	0x01 },
+		{ R51x_SYS_RESET,	0x3d },
+	};
+
+	const struct ov_regvals norm_511[] = {
+		{ R511_DRAM_FLOW_CTL, 	0x01 },
+		{ R51x_SYS_SNAP,	0x00 },
+		{ R51x_SYS_SNAP,	0x02 },
+		{ R51x_SYS_SNAP,	0x00 },
+		{ R511_FIFO_OPTS,	0x1f },
+		{ R511_COMP_EN,		0x00 },
+		{ R511_COMP_LUT_EN,	0x03 },
+	};
+
+	const struct ov_regvals norm_511_p[] = {
+		{ R511_DRAM_FLOW_CTL,	0xff },
+		{ R51x_SYS_SNAP,	0x00 },
+		{ R51x_SYS_SNAP,	0x02 },
+		{ R51x_SYS_SNAP,	0x00 },
+		{ R511_FIFO_OPTS,	0xff },
+		{ R511_COMP_EN,		0x00 },
+		{ R511_COMP_LUT_EN,	0x03 },
+	};
+
+	const struct ov_regvals compress_511[] = {
+		{ 0x70, 0x1f },
+		{ 0x71, 0x05 },
+		{ 0x72, 0x06 },
+		{ 0x73, 0x06 },
+		{ 0x74, 0x14 },
+		{ 0x75, 0x03 },
+		{ 0x76, 0x04 },
+		{ 0x77, 0x04 },
+	};
+
+	PDEBUG(D_PROBE, "Device custom id %x", reg_r(sd, R51x_SYS_CUST_ID));
+
+	rc = write_regvals(sd, init_511, ARRAY_SIZE(init_511));
+	if (rc < 0)
+		return rc;
+
+	switch (sd->bridge) {
+	case BRIDGE_OV511:
+		rc = write_regvals(sd, norm_511, ARRAY_SIZE(norm_511));
+		if (rc < 0)
+			return rc;
+		break;
+	case BRIDGE_OV511PLUS:
+		rc = write_regvals(sd, norm_511_p, ARRAY_SIZE(norm_511_p));
+		if (rc < 0)
+			return rc;
+		break;
+	}
+
+	/* Init compression */
+	rc = write_regvals(sd, compress_511, ARRAY_SIZE(compress_511));
+	if (rc < 0)
+		return rc;
+
+	rc = ov51x_upload_quan_tables(sd);
+	if (rc < 0) {
+		PDEBUG(D_ERR, "Error uploading quantization tables");
+		return rc;
+	}
+
+	return 0;
+}
+
 /* This initializes the OV518/OV518+ and the sensor */
 static int ov518_configure(struct gspca_dev *gspca_dev)
 {
@@ -1615,7 +1930,7 @@ static int ov518_configure(struct gspca_dev *gspca_dev)
 		break;
 	}
 
-	rc = ov518_upload_quan_tables(sd);
+	rc = ov51x_upload_quan_tables(sd);
 	if (rc < 0) {
 		PDEBUG(D_ERR, "Error uploading quantization tables");
 		return rc;
@@ -1661,6 +1976,10 @@ static int sd_config(struct gspca_dev *gspca_dev,
 	sd->invert_led = id->driver_info & BRIDGE_INVERT_LED;
 
 	switch (sd->bridge) {
+	case BRIDGE_OV511:
+	case BRIDGE_OV511PLUS:
+		ret = ov511_configure(gspca_dev);
+		break;
 	case BRIDGE_OV518:
 	case BRIDGE_OV518PLUS:
 		ret = ov518_configure(gspca_dev);
@@ -1719,6 +2038,16 @@ static int sd_config(struct gspca_dev *gspca_dev,
 
 	cam = &gspca_dev->cam;
 	switch (sd->bridge) {
+	case BRIDGE_OV511:
+	case BRIDGE_OV511PLUS:
+		if (!sd->sif) {
+			cam->cam_mode = ov511_vga_mode;
+			cam->nmodes = ARRAY_SIZE(ov511_vga_mode);
+		} else {
+			cam->cam_mode = ov511_sif_mode;
+			cam->nmodes = ARRAY_SIZE(ov511_sif_mode);
+		}
+		break;
 	case BRIDGE_OV518:
 	case BRIDGE_OV518PLUS:
 		if (!sd->sif) {
@@ -1810,6 +2139,126 @@ static int sd_init(struct gspca_dev *gspca_dev)
 	return 0;
 }
 
+/* Set up the OV511/OV511+ with the given image parameters.
+ *
+ * Do not put any sensor-specific code in here (including I2C I/O functions)
+ */
+static int ov511_mode_init_regs(struct sd *sd)
+{
+	int hsegs, vsegs, packet_size, fps, needed;
+	int interlaced = 0;
+	struct usb_host_interface *alt;
+	struct usb_interface *intf;
+
+	intf = usb_ifnum_to_if(sd->gspca_dev.dev, sd->gspca_dev.iface);
+	alt = usb_altnum_to_altsetting(intf, sd->gspca_dev.alt);
+	if (!alt) {
+		PDEBUG(D_ERR, "Couldn't get altsetting");
+		return -EIO;
+	}
+
+	packet_size = le16_to_cpu(alt->endpoint[0].desc.wMaxPacketSize);
+	reg_w(sd, R51x_FIFO_PSIZE, packet_size >> 5);
+
+	reg_w(sd, R511_CAM_UV_EN, 0x01);
+	reg_w(sd, R511_SNAP_UV_EN, 0x01);
+	reg_w(sd, R511_SNAP_OPTS, 0x03);
+
+	/* Here I'm assuming that snapshot size == image size.
+	 * I hope that's always true. --claudio
+	 */
+	hsegs = (sd->gspca_dev.width >> 3) - 1;
+	vsegs = (sd->gspca_dev.height >> 3) - 1;
+
+	reg_w(sd, R511_CAM_PXCNT, hsegs);
+	reg_w(sd, R511_CAM_LNCNT, vsegs);
+	reg_w(sd, R511_CAM_PXDIV, 0x00);
+	reg_w(sd, R511_CAM_LNDIV, 0x00);
+
+	/* YUV420, low pass filter on */
+	reg_w(sd, R511_CAM_OPTS, 0x03);
+
+	/* Snapshot additions */
+	reg_w(sd, R511_SNAP_PXCNT, hsegs);
+	reg_w(sd, R511_SNAP_LNCNT, vsegs);
+	reg_w(sd, R511_SNAP_PXDIV, 0x00);
+	reg_w(sd, R511_SNAP_LNDIV, 0x00);
+
+	/******** Set the framerate ********/
+	if (frame_rate > 0)
+		sd->frame_rate = frame_rate;
+
+	switch (sd->sensor) {
+	case SEN_OV6620:
+		/* No framerate control, doesn't like higher rates yet */
+		sd->clockdiv = 3;
+		break;
+
+	/* Note once the FIXME's in mode_init_ov_sensor_regs() are fixed
+	   for more sensors we need to do this for them too */
+	case SEN_OV7620:
+	case SEN_OV7640:
+		if (sd->gspca_dev.width == 320)
+			interlaced = 1;
+		/* Fall through */
+	case SEN_OV6630:
+	case SEN_OV76BE:
+	case SEN_OV7610:
+	case SEN_OV7670:
+		switch (sd->frame_rate) {
+		case 30:
+		case 25:
+			/* Not enough bandwidth to do 640x480 @ 30 fps */
+			if (sd->gspca_dev.width != 640) {
+				sd->clockdiv = 0;
+				break;
+			}
+			/* Fall through for 640x480 case */
+		default:
+/*		case 20: */
+/*		case 15: */
+			sd->clockdiv = 1;
+			break;
+		case 10:
+			sd->clockdiv = 2;
+			break;
+		case 5:
+			sd->clockdiv = 5;
+			break;
+		}
+		if (interlaced) {
+			sd->clockdiv = (sd->clockdiv + 1) * 2 - 1;
+			/* Higher then 10 does not work */
+			if (sd->clockdiv > 10)
+				sd->clockdiv = 10;
+		}
+		break;
+
+	case SEN_OV8610:
+		/* No framerate control ?? */
+		sd->clockdiv = 0;
+		break;
+	}
+
+	/* Check if we have enough bandwidth to disable compression */
+	fps = (interlaced ? 60 : 30) / (sd->clockdiv + 1) + 1;
+	needed = fps * sd->gspca_dev.width * sd->gspca_dev.height * 3 / 2;
+	/* 1400 is a conservative estimate of the max nr of isoc packets/sec */
+	if (needed > 1400 * packet_size) {
+		/* Enable Y and UV quantization and compression */
+		reg_w(sd, R511_COMP_EN, 0x07);
+		reg_w(sd, R511_COMP_LUT_EN, 0x03);
+	} else {
+		reg_w(sd, R511_COMP_EN, 0x06);
+		reg_w(sd, R511_COMP_LUT_EN, 0x00);
+	}
+
+	reg_w(sd, R51x_SYS_RESET, OV511_RESET_OMNICE);
+	reg_w(sd, R51x_SYS_RESET, 0);
+
+	return 0;
+}
+
 /* Sets up the OV518/OV518+ with the given image parameters
  *
  * OV518 needs a completely different approach, until we can figure out what
@@ -2363,6 +2812,10 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	int ret = 0;
 
 	switch (sd->bridge) {
+	case BRIDGE_OV511:
+	case BRIDGE_OV511PLUS:
+		ret = ov511_mode_init_regs(sd);
+		break;
 	case BRIDGE_OV518:
 	case BRIDGE_OV518PLUS:
 		ret = ov518_mode_init_regs(sd);
@@ -2403,6 +2856,56 @@ static void sd_stopN(struct gspca_dev *gspca_dev)
 	ov51x_led_control(sd, 0);
 }
 
+static void ov511_pkt_scan(struct gspca_dev *gspca_dev,
+			struct gspca_frame *frame,	/* target */
+			__u8 *in,			/* isoc packet */
+			int len)			/* iso packet length */
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	/* SOF/EOF packets have 1st to 8th bytes zeroed and the 9th
+	 * byte non-zero. The EOF packet has image width/height in the
+	 * 10th and 11th bytes. The 9th byte is given as follows:
+	 *
+	 * bit 7: EOF
+	 *     6: compression enabled
+	 *     5: 422/420/400 modes
+	 *     4: 422/420/400 modes
+	 *     3: 1
+	 *     2: snapshot button on
+	 *     1: snapshot frame
+	 *     0: even/odd field
+	 */
+	if (!(in[0] | in[1] | in[2] | in[3] | in[4] | in[5] | in[6] | in[7]) &&
+	    (in[8] & 0x08)) {
+		if (in[8] & 0x80) {
+			/* Frame end */
+			if ((in[9] + 1) * 8 != gspca_dev->width ||
+			    (in[10] + 1) * 8 != gspca_dev->height) {
+				PDEBUG(D_ERR, "Invalid frame size, got: %dx%d,"
+					" requested: %dx%d\n",
+					(in[9] + 1) * 8, (in[10] + 1) * 8,
+					gspca_dev->width, gspca_dev->height);
+				gspca_dev->last_packet_type = DISCARD_PACKET;
+				return;
+			}
+			/* Add 11 byte footer to frame, might be usefull */
+			gspca_frame_add(gspca_dev, LAST_PACKET, frame, in, 11);
+			return;
+		} else {
+			/* Frame start */
+			gspca_frame_add(gspca_dev, FIRST_PACKET, frame, in, 0);
+			sd->packet_nr = 0;
+		}
+	}
+
+	/* Ignore the packet number */
+	len--;
+
+	/* intermediate packet */
+	gspca_frame_add(gspca_dev, INTER_PACKET, frame, in, len);
+}
+
 static void ov518_pkt_scan(struct gspca_dev *gspca_dev,
 			struct gspca_frame *frame,	/* target */
 			__u8 *data,			/* isoc packet */
@@ -2495,6 +2998,7 @@ static void sd_pkt_scan(struct gspca_dev *gspca_dev,
 	switch (sd->bridge) {
 	case BRIDGE_OV511:
 	case BRIDGE_OV511PLUS:
+		ov511_pkt_scan(gspca_dev, frame, data, len);
 		break;
 	case BRIDGE_OV518:
 	case BRIDGE_OV518PLUS:
@@ -2862,12 +3366,15 @@ static const __devinitdata struct usb_device_id device_table[] = {
 	{USB_DEVICE(0x045e, 0x028c), .driver_info = BRIDGE_OV519 },
 	{USB_DEVICE(0x054c, 0x0154), .driver_info = BRIDGE_OV519 },
 	{USB_DEVICE(0x054c, 0x0155), .driver_info = BRIDGE_OV519 },
+	{USB_DEVICE(0x05a9, 0x0511), .driver_info = BRIDGE_OV511 },
 	{USB_DEVICE(0x05a9, 0x0518), .driver_info = BRIDGE_OV518 },
 	{USB_DEVICE(0x05a9, 0x0519), .driver_info = BRIDGE_OV519 },
 	{USB_DEVICE(0x05a9, 0x0530), .driver_info = BRIDGE_OV519 },
 	{USB_DEVICE(0x05a9, 0x4519), .driver_info = BRIDGE_OV519 },
 	{USB_DEVICE(0x05a9, 0x8519), .driver_info = BRIDGE_OV519 },
+	{USB_DEVICE(0x05a9, 0xa511), .driver_info = BRIDGE_OV511PLUS },
 	{USB_DEVICE(0x05a9, 0xa518), .driver_info = BRIDGE_OV518PLUS },
+	{USB_DEVICE(0x0813, 0x0002), .driver_info = BRIDGE_OV511PLUS },
 	{}
 };
 
diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h
index 772d226cb5ca..8a025d510904 100644
--- a/include/linux/videodev2.h
+++ b/include/linux/videodev2.h
@@ -348,6 +348,7 @@ struct v4l2_pix_format {
 #define V4L2_PIX_FMT_SQ905C   v4l2_fourcc('9', '0', '5', 'C') /* compressed RGGB bayer */
 #define V4L2_PIX_FMT_PJPG     v4l2_fourcc('P', 'J', 'P', 'G') /* Pixart 73xx JPEG */
 #define V4L2_PIX_FMT_YVYU     v4l2_fourcc('Y', 'V', 'Y', 'U') /* 16 YVU 4:2:2 */
+#define V4L2_PIX_FMT_OV511    v4l2_fourcc('O', '5', '1', '1') /* ov511 JPEG */
 #define V4L2_PIX_FMT_OV518    v4l2_fourcc('O', '5', '1', '8') /* ov518 JPEG */
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From b282d87332f5b3c2ac2e289f772b33067e4be77b Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Sun, 14 Jun 2009 19:10:40 -0300
Subject: V4L/DVB (12080): gspca_ov519: Fix ov518+ with OV7620AE (Trust
 spacecam 320)

gspca_ov519: Fix ov518+ with OV7620AE (Trust spacecam 320)

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/ov519.c | 91 +++++++++++++++++++++++----------------
 1 file changed, 55 insertions(+), 36 deletions(-)

diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c
index 1f8e2613ecc5..3aebc744368d 100644
--- a/drivers/media/video/gspca/ov519.c
+++ b/drivers/media/video/gspca/ov519.c
@@ -269,37 +269,43 @@ static const struct v4l2_pix_format ov519_sif_mode[] = {
 		.priv = 0},
 };
 
+/* Note some of the sizeimage values for the ov511 / ov518 may seem
+   larger then necessary, however they need to be this big as the ov511 /
+   ov518 always fills the entire isoc frame, using 0 padding bytes when
+   it doesn't have any data. So with low framerates the amount of data
+   transfered can become quite large (libv4l will remove all the 0 padding
+   in userspace). */
 static const struct v4l2_pix_format ov518_vga_mode[] = {
 	{320, 240, V4L2_PIX_FMT_OV518, V4L2_FIELD_NONE,
 		.bytesperline = 320,
-		.sizeimage = 320 * 240 * 3 / 8 + 590,
+		.sizeimage = 320 * 240 * 3,
 		.colorspace = V4L2_COLORSPACE_JPEG,
 		.priv = 1},
 	{640, 480, V4L2_PIX_FMT_OV518, V4L2_FIELD_NONE,
 		.bytesperline = 640,
-		.sizeimage = 640 * 480 * 3 / 8 + 590,
+		.sizeimage = 640 * 480 * 2,
 		.colorspace = V4L2_COLORSPACE_JPEG,
 		.priv = 0},
 };
 static const struct v4l2_pix_format ov518_sif_mode[] = {
 	{160, 120, V4L2_PIX_FMT_OV518, V4L2_FIELD_NONE,
 		.bytesperline = 160,
-		.sizeimage = 40000,
+		.sizeimage = 70000,
 		.colorspace = V4L2_COLORSPACE_JPEG,
 		.priv = 3},
 	{176, 144, V4L2_PIX_FMT_OV518, V4L2_FIELD_NONE,
 		.bytesperline = 176,
-		.sizeimage = 40000,
+		.sizeimage = 70000,
 		.colorspace = V4L2_COLORSPACE_JPEG,
 		.priv = 1},
 	{320, 240, V4L2_PIX_FMT_OV518, V4L2_FIELD_NONE,
 		.bytesperline = 320,
-		.sizeimage = 320 * 240 * 3 / 8 + 590,
+		.sizeimage = 320 * 240 * 3,
 		.colorspace = V4L2_COLORSPACE_JPEG,
 		.priv = 2},
 	{352, 288, V4L2_PIX_FMT_OV518, V4L2_FIELD_NONE,
 		.bytesperline = 352,
-		.sizeimage = 352 * 288 * 3 / 8 + 590,
+		.sizeimage = 352 * 288 * 3,
 		.colorspace = V4L2_COLORSPACE_JPEG,
 		.priv = 0},
 };
@@ -319,12 +325,12 @@ static const struct v4l2_pix_format ov511_vga_mode[] = {
 static const struct v4l2_pix_format ov511_sif_mode[] = {
 	{160, 120, V4L2_PIX_FMT_OV511, V4L2_FIELD_NONE,
 		.bytesperline = 160,
-		.sizeimage = 40000,
+		.sizeimage = 70000,
 		.colorspace = V4L2_COLORSPACE_JPEG,
 		.priv = 3},
 	{176, 144, V4L2_PIX_FMT_OV511, V4L2_FIELD_NONE,
 		.bytesperline = 176,
-		.sizeimage = 40000,
+		.sizeimage = 70000,
 		.colorspace = V4L2_COLORSPACE_JPEG,
 		.priv = 1},
 	{320, 240, V4L2_PIX_FMT_OV511, V4L2_FIELD_NONE,
@@ -698,7 +704,7 @@ static const struct ov_i2c_regvals norm_7620[] = {
 	{ 0x23, 0x00 },
 	{ 0x26, 0xa2 },
 	{ 0x27, 0xea },
-	{ 0x28, 0x20 },
+	{ 0x28, 0x22 }, /* Was 0x20, bit1 enables a 2x gain which we need */
 	{ 0x29, 0x00 },
 	{ 0x2a, 0x10 },
 	{ 0x2b, 0x00 },
@@ -1525,7 +1531,6 @@ static int ov8xx0_configure(struct sd *sd)
 	}
 
 	/* Set sensor-specific vars */
-/*	sd->sif = 0;		already done */
 	return 0;
 }
 
@@ -1562,15 +1567,13 @@ static int ov7xx0_configure(struct sd *sd)
 		}
 	} else if ((rc & 3) == 1) {
 		/* I don't know what's different about the 76BE yet. */
-		if (i2c_r(sd, 0x15) & 1)
+		if (i2c_r(sd, 0x15) & 1) {
 			PDEBUG(D_PROBE, "Sensor is an OV7620AE");
-		else
+			sd->sensor = SEN_OV7620;
+		} else {
 			PDEBUG(D_PROBE, "Sensor is an OV76BE");
-
-		/* OV511+ will return all zero isoc data unless we
-		 * configure the sensor as a 7620. Someone needs to
-		 * find the exact reg. setting that causes this. */
-		sd->sensor = SEN_OV76BE;
+			sd->sensor = SEN_OV76BE;
+		}
 	} else if ((rc & 3) == 0) {
 		/* try to read product id registers */
 		high = i2c_r(sd, 0x0a);
@@ -1616,7 +1619,6 @@ static int ov7xx0_configure(struct sd *sd)
 	}
 
 	/* Set sensor-specific vars */
-/*	sd->sif = 0;		already done */
 	return 0;
 }
 
@@ -2198,11 +2200,11 @@ static int ov511_mode_init_regs(struct sd *sd)
 	   for more sensors we need to do this for them too */
 	case SEN_OV7620:
 	case SEN_OV7640:
+	case SEN_OV76BE:
 		if (sd->gspca_dev.width == 320)
 			interlaced = 1;
 		/* Fall through */
 	case SEN_OV6630:
-	case SEN_OV76BE:
 	case SEN_OV7610:
 	case SEN_OV7670:
 		switch (sd->frame_rate) {
@@ -2268,7 +2270,19 @@ static int ov511_mode_init_regs(struct sd *sd)
  */
 static int ov518_mode_init_regs(struct sd *sd)
 {
-	int hsegs, vsegs;
+	int hsegs, vsegs, packet_size;
+	struct usb_host_interface *alt;
+	struct usb_interface *intf;
+
+	intf = usb_ifnum_to_if(sd->gspca_dev.dev, sd->gspca_dev.iface);
+	alt = usb_altnum_to_altsetting(intf, sd->gspca_dev.alt);
+	if (!alt) {
+		PDEBUG(D_ERR, "Couldn't get altsetting");
+		return -EIO;
+	}
+
+	packet_size = le16_to_cpu(alt->endpoint[0].desc.wMaxPacketSize);
+	ov518_reg_w32(sd, R51x_FIFO_PSIZE, packet_size & ~7, 2);
 
 	/******** Set the mode ********/
 
@@ -2305,20 +2319,30 @@ static int ov518_mode_init_regs(struct sd *sd)
 	/* Windows driver does this here; who knows why */
 	reg_w(sd, 0x2f, 0x80);
 
-	/******** Set the framerate (to 30 FPS) ********/
-	if (sd->bridge == BRIDGE_OV518PLUS)
-		sd->clockdiv = 1;
-	else
-		sd->clockdiv = 0;
+	/******** Set the framerate  ********/
+	sd->clockdiv = 1;
 
 	/* Mode independent, but framerate dependent, regs */
-	reg_w(sd, 0x51, 0x04);	/* Clock divider; lower==faster */
+	/* 0x51: Clock divider; Only works on some cams which use 2 crystals */
+	reg_w(sd, 0x51, 0x04);
 	reg_w(sd, 0x22, 0x18);
 	reg_w(sd, 0x23, 0xff);
 
-	if (sd->bridge == BRIDGE_OV518PLUS)
-		reg_w(sd, 0x21, 0x19);
-	else
+	if (sd->bridge == BRIDGE_OV518PLUS) {
+		switch (sd->sensor) {
+		case SEN_OV7620:
+			if (sd->gspca_dev.width == 320) {
+				reg_w(sd, 0x20, 0x00);
+				reg_w(sd, 0x21, 0x19);
+			} else {
+				reg_w(sd, 0x20, 0x60);
+				reg_w(sd, 0x21, 0x1f);
+			}
+			break;
+		default:
+			reg_w(sd, 0x21, 0x19);
+		}
+	} else
 		reg_w(sd, 0x71, 0x17);	/* Compression-related? */
 
 	/* FIXME: Sensor-specific */
@@ -2537,21 +2561,16 @@ static int mode_init_ov_sensor_regs(struct sd *sd)
 		i2c_w_mask(sd, 0x14, qvga ? 0x20 : 0x00, 0x20);
 		break;
 	case SEN_OV7620:
-/*		i2c_w(sd, 0x2b, 0x00); */
+	case SEN_OV76BE:
 		i2c_w_mask(sd, 0x14, qvga ? 0x20 : 0x00, 0x20);
 		i2c_w_mask(sd, 0x28, qvga ? 0x00 : 0x20, 0x20);
 		i2c_w(sd, 0x24, qvga ? 0x20 : 0x3a);
 		i2c_w(sd, 0x25, qvga ? 0x30 : 0x60);
 		i2c_w_mask(sd, 0x2d, qvga ? 0x40 : 0x00, 0x40);
-		i2c_w_mask(sd, 0x67, qvga ? 0xf0 : 0x90, 0xf0);
+		i2c_w_mask(sd, 0x67, qvga ? 0xb0 : 0x90, 0xf0);
 		i2c_w_mask(sd, 0x74, qvga ? 0x20 : 0x00, 0x20);
 		break;
-	case SEN_OV76BE:
-/*		i2c_w(sd, 0x2b, 0x00); */
-		i2c_w_mask(sd, 0x14, qvga ? 0x20 : 0x00, 0x20);
-		break;
 	case SEN_OV7640:
-/*		i2c_w(sd, 0x2b, 0x00); */
 		i2c_w_mask(sd, 0x14, qvga ? 0x20 : 0x00, 0x20);
 		i2c_w_mask(sd, 0x28, qvga ? 0x00 : 0x20, 0x20);
 /*		i2c_w(sd, 0x24, qvga ? 0x20 : 0x3a); */
-- 
cgit v1.2.3-59-g8ed1b


From ae49c40461d8981b232e3fec28234d492067f0e1 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Sun, 14 Jun 2009 19:15:07 -0300
Subject: V4L/DVB (12081): gspca_ov519: Cleanup some sensor special cases

gspca_ov519: Cleanup some sensor special cases

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/ov519.c | 24 ++++--------------------
 1 file changed, 4 insertions(+), 20 deletions(-)

diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c
index 3aebc744368d..cb5f3c786db4 100644
--- a/drivers/media/video/gspca/ov519.c
+++ b/drivers/media/video/gspca/ov519.c
@@ -529,7 +529,7 @@ static const struct ov_i2c_regvals norm_6x20[] = {
 	{ 0x28, 0x05 },
 	{ 0x2a, 0x04 }, /* Disable framerate adjust */
 /*	{ 0x2b, 0xac },  * Framerate; Set 2a[7] first */
-	{ 0x2d, 0x99 },
+	{ 0x2d, 0x85 },
 	{ 0x33, 0xa0 }, /* Color Processing Parameter */
 	{ 0x34, 0xd2 }, /* Max A/D range */
 	{ 0x38, 0x8b },
@@ -2120,6 +2120,8 @@ static int sd_init(struct gspca_dev *gspca_dev)
 /*	case SEN_OV76BE: */
 		if (write_i2c_regvals(sd, norm_7610, ARRAY_SIZE(norm_7610)))
 			return -EIO;
+		if (i2c_w_mask(sd, 0x0e, 0x00, 0x40))
+			return -EIO;
 		break;
 	case SEN_OV7620:
 		if (write_i2c_regvals(sd, norm_7620, ARRAY_SIZE(norm_7620)))
@@ -2597,10 +2599,6 @@ static int mode_init_ov_sensor_regs(struct sd *sd)
 	}
 
 	/******** Palette-specific regs ********/
-	if (sd->sensor == SEN_OV7610 || sd->sensor == SEN_OV76BE) {
-		/* not valid on the OV6620/OV7620/6630? */
-		i2c_w_mask(sd, 0x0e, 0x00, 0x40);
-	}
 
 	/* The OV518 needs special treatment. Although both the OV518
 	 * and the OV6630 support a 16-bit video bus, only the 8 bit Y
@@ -2615,21 +2613,7 @@ static int mode_init_ov_sensor_regs(struct sd *sd)
 		i2c_w_mask(sd, 0x13, 0x00, 0x20);
 
 	/******** Clock programming ********/
-	/* The OV6620 needs special handling. This prevents the
-	 * severe banding that normally occurs */
-	if (sd->sensor == SEN_OV6620) {
-
-		/* Clock down */
-		i2c_w(sd, 0x2a, 0x04);
-		i2c_w(sd, 0x11, sd->clockdiv);
-		i2c_w(sd, 0x2a, 0x84);
-		/* This next setting is critical. It seems to improve
-		 * the gain or the contrast. The "reserved" bits seem
-		 * to have some effect in this case. */
-		i2c_w(sd, 0x2d, 0x85);
-	} else {
-		i2c_w(sd, 0x11, sd->clockdiv);
-	}
+	i2c_w(sd, 0x11, sd->clockdiv);
 
 	/******** Special Features ********/
 /* no evidence this is possible with OV7670, either */
-- 
cgit v1.2.3-59-g8ed1b


From 8668d504d72c384fbfb6ab6f5d02a9fe4d813554 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Wed, 17 Jun 2009 18:37:57 -0300
Subject: V4L/DVB (12082): gspca_stv06xx: Add support for st6422 bridge and
 sensor

Add support for st6422 bridge and sensor to the stv06xx gspca sub driver,
tested with:
Logitech QuickCam Messenger     046d:08f0       ST6422  integrated
Logitech QuickCam Mess. Plus    046d:08f6       ST6422  integrated

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/stv06xx/Makefile         |   3 +-
 drivers/media/video/gspca/stv06xx/stv06xx.c        |  53 ++-
 drivers/media/video/gspca/stv06xx/stv06xx.h        |  11 +
 drivers/media/video/gspca/stv06xx/stv06xx_hdcs.c   |  10 +-
 drivers/media/video/gspca/stv06xx/stv06xx_sensor.h |   3 +-
 drivers/media/video/gspca/stv06xx/stv06xx_st6422.c | 453 +++++++++++++++++++++
 drivers/media/video/gspca/stv06xx/stv06xx_st6422.h |  59 +++
 7 files changed, 577 insertions(+), 15 deletions(-)
 create mode 100644 drivers/media/video/gspca/stv06xx/stv06xx_st6422.c
 create mode 100644 drivers/media/video/gspca/stv06xx/stv06xx_st6422.h

diff --git a/drivers/media/video/gspca/stv06xx/Makefile b/drivers/media/video/gspca/stv06xx/Makefile
index feeaa94ab588..2f3c3a606ce4 100644
--- a/drivers/media/video/gspca/stv06xx/Makefile
+++ b/drivers/media/video/gspca/stv06xx/Makefile
@@ -3,7 +3,8 @@ obj-$(CONFIG_USB_STV06XX) += gspca_stv06xx.o
 gspca_stv06xx-objs := stv06xx.o \
 		      stv06xx_vv6410.o \
 		      stv06xx_hdcs.o \
-		      stv06xx_pb0100.o
+		      stv06xx_pb0100.o \
+		      stv06xx_st6422.o
 
 EXTRA_CFLAGS += -Idrivers/media/video/gspca
 
diff --git a/drivers/media/video/gspca/stv06xx/stv06xx.c b/drivers/media/video/gspca/stv06xx/stv06xx.c
index e573c3406324..0da8e0de0456 100644
--- a/drivers/media/video/gspca/stv06xx/stv06xx.c
+++ b/drivers/media/video/gspca/stv06xx/stv06xx.c
@@ -92,11 +92,10 @@ static int stv06xx_write_sensor_finish(struct sd *sd)
 {
 	int err = 0;
 
-	if (IS_850(sd)) {
+	if (sd->bridge == BRIDGE_STV610) {
 		struct usb_device *udev = sd->gspca_dev.dev;
 		__u8 *buf = sd->gspca_dev.usb_buf;
 
-		/* Quickam Web needs an extra packet */
 		buf[0] = 0;
 		err = usb_control_msg(udev, usb_sndctrlpipe(udev, 0),
 				      0x04, 0x40, 0x1704, 0, buf, 1,
@@ -253,7 +252,7 @@ static int stv06xx_init(struct gspca_dev *gspca_dev)
 
 	err = sd->sensor->init(sd);
 
-	if (dump_sensor)
+	if (dump_sensor && sd->sensor->dump)
 		sd->sensor->dump(sd);
 
 	return (err < 0) ? err : 0;
@@ -318,6 +317,8 @@ static void stv06xx_pkt_scan(struct gspca_dev *gspca_dev,
 			__u8 *data,			/* isoc packet */
 			int len)			/* iso packet length */
 {
+	struct sd *sd = (struct sd *) gspca_dev;
+
 	PDEBUG(D_PACK, "Packet of length %d arrived", len);
 
 	/* A packet may contain several frames
@@ -343,14 +344,29 @@ static void stv06xx_pkt_scan(struct gspca_dev *gspca_dev,
 		if (len < chunk_len) {
 			PDEBUG(D_ERR, "URB packet length is smaller"
 				" than the specified chunk length");
+			gspca_dev->last_packet_type = DISCARD_PACKET;
 			return;
 		}
 
+		/* First byte seem to be 02=data 2nd byte is unknown??? */
+		if (sd->bridge == BRIDGE_ST6422 && (id & 0xFF00) == 0x0200)
+			goto frame_data;
+
 		switch (id) {
 		case 0x0200:
 		case 0x4200:
+frame_data:
 			PDEBUG(D_PACK, "Frame data packet detected");
 
+			if (sd->to_skip) {
+				int skip = (sd->to_skip < chunk_len) ?
+					    sd->to_skip : chunk_len;
+				data += skip;
+				len -= skip;
+				chunk_len -= skip;
+				sd->to_skip -= skip;
+			}
+
 			gspca_frame_add(gspca_dev, INTER_PACKET, frame,
 					data, chunk_len);
 			break;
@@ -365,6 +381,9 @@ static void stv06xx_pkt_scan(struct gspca_dev *gspca_dev,
 			gspca_frame_add(gspca_dev, FIRST_PACKET,
 					frame, data, 0);
 
+			if (sd->bridge == BRIDGE_ST6422)
+				sd->to_skip = gspca_dev->width * 4;
+
 			if (chunk_len)
 				PDEBUG(D_ERR, "Chunk length is "
 					      "non-zero on a SOF");
@@ -395,8 +414,12 @@ static void stv06xx_pkt_scan(struct gspca_dev *gspca_dev,
 			/* Unknown chunk with 2 bytes of data,
 			   occurs 2-3 times per USB interrupt */
 			break;
+		case 0x42ff:
+			PDEBUG(D_PACK, "Chunk 0x42ff detected");
+			/* Special chunk seen sometimes on the ST6422 */
+			break;
 		default:
-			PDEBUG(D_PACK, "Unknown chunk %d detected", id);
+			PDEBUG(D_PACK, "Unknown chunk 0x%04x detected", id);
 			/* Unknown chunk */
 		}
 		data    += chunk_len;
@@ -428,11 +451,16 @@ static int stv06xx_config(struct gspca_dev *gspca_dev,
 
 	cam = &gspca_dev->cam;
 	sd->desc = sd_desc;
+	sd->bridge = id->driver_info;
 	gspca_dev->sd_desc = &sd->desc;
 
 	if (dump_bridge)
 		stv06xx_dump_bridge(sd);
 
+	sd->sensor = &stv06xx_sensor_st6422;
+	if (!sd->sensor->probe(sd))
+		return 0;
+
 	sd->sensor = &stv06xx_sensor_vv6410;
 	if (!sd->sensor->probe(sd))
 		return 0;
@@ -457,9 +485,20 @@ static int stv06xx_config(struct gspca_dev *gspca_dev,
 
 /* -- module initialisation -- */
 static const __devinitdata struct usb_device_id device_table[] = {
-	{USB_DEVICE(0x046d, 0x0840)}, /* QuickCam Express */
-	{USB_DEVICE(0x046d, 0x0850)}, /* LEGO cam / QuickCam Web */
-	{USB_DEVICE(0x046d, 0x0870)}, /* Dexxa WebCam USB */
+	/* QuickCam Express */
+	{USB_DEVICE(0x046d, 0x0840), .driver_info = BRIDGE_STV600 },
+	/* LEGO cam / QuickCam Web */
+	{USB_DEVICE(0x046d, 0x0850), .driver_info = BRIDGE_STV610 },
+	/* Dexxa WebCam USB */
+	{USB_DEVICE(0x046d, 0x0870), .driver_info = BRIDGE_STV602 },
+	/* QuickCam Messenger */
+	{USB_DEVICE(0x046D, 0x08F0), .driver_info = BRIDGE_ST6422 },
+	/* QuickCam Communicate */
+	{USB_DEVICE(0x046D, 0x08F5), .driver_info = BRIDGE_ST6422 },
+	/* QuickCam Messenger (new) */
+	{USB_DEVICE(0x046D, 0x08F6), .driver_info = BRIDGE_ST6422 },
+	/* QuickCam Messenger (new) */
+	{USB_DEVICE(0x046D, 0x08DA), .driver_info = BRIDGE_ST6422 },
 	{}
 };
 MODULE_DEVICE_TABLE(usb, device_table);
diff --git a/drivers/media/video/gspca/stv06xx/stv06xx.h b/drivers/media/video/gspca/stv06xx/stv06xx.h
index 1207e7d17f14..9df7137fe67e 100644
--- a/drivers/media/video/gspca/stv06xx/stv06xx.h
+++ b/drivers/media/video/gspca/stv06xx/stv06xx.h
@@ -93,6 +93,17 @@ struct sd {
 
 	/* Sensor private data */
 	void *sensor_priv;
+
+	/* The first 4 lines produced by the stv6422 are no good, this keeps
+	   track of how many bytes we still need to skip during a frame */
+	int to_skip;
+
+	/* Bridge / Camera type */
+	u8 bridge;
+	#define BRIDGE_STV600 0
+	#define BRIDGE_STV602 1
+	#define BRIDGE_STV610 2
+	#define BRIDGE_ST6422 3 /* With integrated sensor */
 };
 
 int stv06xx_write_bridge(struct sd *sd, u16 address, u16 i2c_data);
diff --git a/drivers/media/video/gspca/stv06xx/stv06xx_hdcs.c b/drivers/media/video/gspca/stv06xx/stv06xx_hdcs.c
index b16903814203..3039ec208f3a 100644
--- a/drivers/media/video/gspca/stv06xx/stv06xx_hdcs.c
+++ b/drivers/media/video/gspca/stv06xx/stv06xx_hdcs.c
@@ -434,7 +434,7 @@ static int hdcs_probe_1x00(struct sd *sd)
 	hdcs->exp.er = 100;
 
 	/*
-	 * Frame rate on HDCS-1000 0x46D:0x840 depends on PSMP:
+	 * Frame rate on HDCS-1000 with STV600 depends on PSMP:
 	 *  4 = doesn't work at all
 	 *  5 = 7.8 fps,
 	 *  6 = 6.9 fps,
@@ -443,7 +443,7 @@ static int hdcs_probe_1x00(struct sd *sd)
 	 * 15 = 4.4 fps,
 	 * 31 = 2.8 fps
 	 *
-	 * Frame rate on HDCS-1000 0x46D:0x870 depends on PSMP:
+	 * Frame rate on HDCS-1000 with STV602 depends on PSMP:
 	 * 15 = doesn't work at all
 	 * 18 = doesn't work at all
 	 * 19 = 7.3 fps
@@ -453,7 +453,7 @@ static int hdcs_probe_1x00(struct sd *sd)
 	 * 24 = 6.3 fps
 	 * 30 = 5.4 fps
 	 */
-	hdcs->psmp = IS_870(sd) ? 20 : 5;
+	hdcs->psmp = (sd->bridge == BRIDGE_STV602) ? 20 : 5;
 
 	sd->sensor_priv = hdcs;
 
@@ -530,7 +530,7 @@ static int hdcs_init(struct sd *sd)
 	int i, err = 0;
 
 	/* Set the STV0602AA in STV0600 emulation mode */
-	if (IS_870(sd))
+	if (sd->bridge == BRIDGE_STV602)
 		stv06xx_write_bridge(sd, STV_STV0600_EMULATION, 1);
 
 	/* Execute the bridge init */
@@ -558,7 +558,7 @@ static int hdcs_init(struct sd *sd)
 		return err;
 
 	/* Set PGA sample duration
-	(was 0x7E for IS_870, but caused slow framerate with HDCS-1020) */
+	(was 0x7E for the STV602, but caused slow framerate with HDCS-1020) */
 	if (IS_1020(sd))
 		err = stv06xx_write_sensor(sd, HDCS_TCTRL,
 				(HDCS_ADC_START_SIG_DUR << 6) | hdcs->psmp);
diff --git a/drivers/media/video/gspca/stv06xx/stv06xx_sensor.h b/drivers/media/video/gspca/stv06xx/stv06xx_sensor.h
index e88c42f7d2f8..934b9cebc1ab 100644
--- a/drivers/media/video/gspca/stv06xx/stv06xx_sensor.h
+++ b/drivers/media/video/gspca/stv06xx/stv06xx_sensor.h
@@ -32,14 +32,13 @@
 
 #include "stv06xx.h"
 
-#define IS_850(sd)	((sd)->gspca_dev.dev->descriptor.idProduct == 0x850)
-#define IS_870(sd)	((sd)->gspca_dev.dev->descriptor.idProduct == 0x870)
 #define IS_1020(sd)	((sd)->sensor == &stv06xx_sensor_hdcs1020)
 
 extern const struct stv06xx_sensor stv06xx_sensor_vv6410;
 extern const struct stv06xx_sensor stv06xx_sensor_hdcs1x00;
 extern const struct stv06xx_sensor stv06xx_sensor_hdcs1020;
 extern const struct stv06xx_sensor stv06xx_sensor_pb0100;
+extern const struct stv06xx_sensor stv06xx_sensor_st6422;
 
 struct stv06xx_sensor {
 	/* Defines the name of a sensor */
diff --git a/drivers/media/video/gspca/stv06xx/stv06xx_st6422.c b/drivers/media/video/gspca/stv06xx/stv06xx_st6422.c
new file mode 100644
index 000000000000..87cb5b9ddfa7
--- /dev/null
+++ b/drivers/media/video/gspca/stv06xx/stv06xx_st6422.c
@@ -0,0 +1,453 @@
+/*
+ * Support for the sensor part which is integrated (I think) into the
+ * st6422 stv06xx alike bridge, as its integrated there are no i2c writes
+ * but instead direct bridge writes.
+ *
+ * Copyright (c) 2009 Hans de Goede <hdegoede@redhat.com>
+ *
+ * Strongly based on qc-usb-messenger, which is:
+ * Copyright (c) 2001 Jean-Fredric Clere, Nikolas Zimmermann, Georg Acher
+ *		      Mark Cave-Ayland, Carlo E Prelz, Dick Streefland
+ * Copyright (c) 2002, 2003 Tuukka Toivonen
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include "stv06xx_st6422.h"
+
+static struct v4l2_pix_format st6422_mode[] = {
+	/* Note we actually get 124 lines of data, of which we skip the 4st
+	   4 as they are garbage */
+	{
+		162,
+		120,
+		V4L2_PIX_FMT_SGRBG8,
+		V4L2_FIELD_NONE,
+		.sizeimage = 162 * 120,
+		.bytesperline = 162,
+		.colorspace = V4L2_COLORSPACE_SRGB,
+		.priv = 1
+	},
+	/* Note we actually get 248 lines of data, of which we skip the 4st
+	   4 as they are garbage, and we tell the app it only gets the
+	   first 240 of the 244 lines it actually gets, so that it ignores
+	   the last 4. */
+	{
+		324,
+		240,
+		V4L2_PIX_FMT_SGRBG8,
+		V4L2_FIELD_NONE,
+		.sizeimage = 324 * 244,
+		.bytesperline = 324,
+		.colorspace = V4L2_COLORSPACE_SRGB,
+		.priv = 0
+	},
+};
+
+static const struct ctrl st6422_ctrl[] = {
+#define BRIGHTNESS_IDX 0
+	{
+		{
+			.id		= V4L2_CID_BRIGHTNESS,
+			.type		= V4L2_CTRL_TYPE_INTEGER,
+			.name		= "Brightness",
+			.minimum	= 0,
+			.maximum	= 31,
+			.step		= 1,
+			.default_value  = 3
+		},
+		.set = st6422_set_brightness,
+		.get = st6422_get_brightness
+	},
+#define CONTRAST_IDX 1
+	{
+		{
+			.id		= V4L2_CID_CONTRAST,
+			.type		= V4L2_CTRL_TYPE_INTEGER,
+			.name		= "Contrast",
+			.minimum	= 0,
+			.maximum	= 15,
+			.step		= 1,
+			.default_value  = 11
+		},
+		.set = st6422_set_contrast,
+		.get = st6422_get_contrast
+	},
+#define GAIN_IDX 2
+	{
+		{
+			.id		= V4L2_CID_GAIN,
+			.type		= V4L2_CTRL_TYPE_INTEGER,
+			.name		= "Gain",
+			.minimum	= 0,
+			.maximum	= 255,
+			.step		= 1,
+			.default_value  = 64
+		},
+		.set = st6422_set_gain,
+		.get = st6422_get_gain
+	},
+#define EXPOSURE_IDX 3
+	{
+		{
+			.id		= V4L2_CID_EXPOSURE,
+			.type		= V4L2_CTRL_TYPE_INTEGER,
+			.name		= "Exposure",
+			.minimum	= 0,
+			.maximum	= 1023,
+			.step		= 1,
+			.default_value  = 256
+		},
+		.set = st6422_set_exposure,
+		.get = st6422_get_exposure
+	},
+};
+
+static int st6422_probe(struct sd *sd)
+{
+	int i;
+	s32 *sensor_settings;
+
+	if (sd->bridge != BRIDGE_ST6422)
+		return -ENODEV;
+
+	info("st6422 sensor detected");
+
+	sensor_settings = kmalloc(ARRAY_SIZE(st6422_ctrl) * sizeof(s32),
+				  GFP_KERNEL);
+	if (!sensor_settings)
+		return -ENOMEM;
+
+	sd->gspca_dev.cam.cam_mode = st6422_mode;
+	sd->gspca_dev.cam.nmodes = ARRAY_SIZE(st6422_mode);
+	sd->desc.ctrls = st6422_ctrl;
+	sd->desc.nctrls = ARRAY_SIZE(st6422_ctrl);
+	sd->sensor_priv = sensor_settings;
+
+	for (i = 0; i < sd->desc.nctrls; i++)
+		sensor_settings[i] = st6422_ctrl[i].qctrl.default_value;
+
+	return 0;
+}
+
+static int st6422_init(struct sd *sd)
+{
+	int err = 0, i;
+
+	const u16 st6422_bridge_init[][2] = {
+		{ STV_ISO_ENABLE, 0x00 }, /* disable capture */
+		{ 0x1436, 0x00 },
+		{ 0x1432, 0x03 },	/* 0x00-0x1F brightness */
+		{ 0x143a, 0xF9 },	/* 0x00-0x0F contrast */
+		{ 0x0509, 0x38 },	/* R */
+		{ 0x050a, 0x38 },	/* G */
+		{ 0x050b, 0x38 },	/* B */
+		{ 0x050c, 0x2A },
+		{ 0x050d, 0x01 },
+
+
+		{ 0x1431, 0x00 },	/* 0x00-0x07 ??? */
+		{ 0x1433, 0x34 },	/* 160x120, 0x00-0x01 night filter */
+		{ 0x1438, 0x18 },	/* 640x480 */
+/* 18 bayes */
+/* 10 compressed? */
+
+		{ 0x1439, 0x00 },
+/* antiflimmer??  0xa2 ger perfekt bild mot monitor */
+
+		{ 0x143b, 0x05 },
+		{ 0x143c, 0x00 },	/* 0x00-0x01 - ??? */
+
+
+/* shutter time 0x0000-0x03FF */
+/* low value  give good picures on moving objects (but requires much light) */
+/* high value gives good picures in darkness (but tends to be overexposed) */
+		{ 0x143e, 0x01 },
+		{ 0x143d, 0x00 },
+
+		{ 0x1442, 0xe2 },
+/* write: 1x1x xxxx */
+/* read:  1x1x xxxx */
+/*        bit 5 == button pressed and hold if 0 */
+/* write 0xe2,0xea */
+
+/* 0x144a */
+/* 0x00 init */
+/* bit 7 == button has been pressed, but not handled */
+
+/* interrupt */
+/* if(urb->iso_frame_desc[i].status == 0x80) { */
+/* if(urb->iso_frame_desc[i].status == 0x88) { */
+
+		{ 0x1500, 0xd0 },
+		{ 0x1500, 0xd0 },
+		{ 0x1500, 0x50 },	/* 0x00 - 0xFF  0x80 == compr ? */
+
+		{ 0x1501, 0xaf },
+/* high val-> ljus area blir morkare. */
+/* low val -> ljus area blir ljusare. */
+		{ 0x1502, 0xc2 },
+/* high val-> ljus area blir morkare. */
+/* low val -> ljus area blir ljusare. */
+		{ 0x1503, 0x45 },
+/* high val-> ljus area blir morkare. */
+/* low val -> ljus area blir ljusare. */
+
+		{ 0x1505, 0x02 },
+/* 2  : 324x248  80352 bytes */
+/* 7  : 248x162  40176 bytes */
+/* c+f: 162*124  20088 bytes */
+
+		{ 0x150e, 0x8e },
+		{ 0x150f, 0x37 },
+		{ 0x15c0, 0x00 },
+		{ 0x15c1, 1023 }, /* 160x120, ISOC_PACKET_SIZE */
+		{ 0x15c3, 0x08 },	/* 0x04/0x14 ... test pictures ??? */
+
+
+		{ 0x143f, 0x01 },	/* commit settings */
+
+	};
+
+	for (i = 0; i < ARRAY_SIZE(st6422_bridge_init) && !err; i++) {
+		err = stv06xx_write_bridge(sd, st6422_bridge_init[i][0],
+					       st6422_bridge_init[i][1]);
+	}
+
+	return err;
+}
+
+static void st6422_disconnect(struct sd *sd)
+{
+	sd->sensor = NULL;
+	kfree(sd->sensor_priv);
+}
+
+static int st6422_start(struct sd *sd)
+{
+	int err, packet_size;
+	struct cam *cam = &sd->gspca_dev.cam;
+	s32 *sensor_settings = sd->sensor_priv;
+	struct usb_host_interface *alt;
+	struct usb_interface *intf;
+
+	intf = usb_ifnum_to_if(sd->gspca_dev.dev, sd->gspca_dev.iface);
+	alt = usb_altnum_to_altsetting(intf, sd->gspca_dev.alt);
+	if (!alt) {
+		PDEBUG(D_ERR, "Couldn't get altsetting");
+		return -EIO;
+	}
+
+	packet_size = le16_to_cpu(alt->endpoint[0].desc.wMaxPacketSize);
+	err = stv06xx_write_bridge(sd, 0x15c1, packet_size);
+	if (err < 0)
+		return err;
+
+	if (cam->cam_mode[sd->gspca_dev.curr_mode].priv)
+		err = stv06xx_write_bridge(sd, 0x1505, 0x0f);
+	else
+		err = stv06xx_write_bridge(sd, 0x1505, 0x02);
+	if (err < 0)
+		return err;
+
+	err = st6422_set_brightness(&sd->gspca_dev,
+				    sensor_settings[BRIGHTNESS_IDX]);
+	if (err < 0)
+		return err;
+
+	err = st6422_set_contrast(&sd->gspca_dev,
+				  sensor_settings[CONTRAST_IDX]);
+	if (err < 0)
+		return err;
+
+	err = st6422_set_exposure(&sd->gspca_dev,
+				  sensor_settings[EXPOSURE_IDX]);
+	if (err < 0)
+		return err;
+
+	err = st6422_set_gain(&sd->gspca_dev,
+			      sensor_settings[GAIN_IDX]);
+	if (err < 0)
+		return err;
+
+	PDEBUG(D_STREAM, "Starting stream");
+
+	return 0;
+}
+
+static int st6422_stop(struct sd *sd)
+{
+	PDEBUG(D_STREAM, "Halting stream");
+
+	return 0;
+}
+
+static int st6422_get_brightness(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	s32 *sensor_settings = sd->sensor_priv;
+
+	*val = sensor_settings[BRIGHTNESS_IDX];
+
+	PDEBUG(D_V4L2, "Read brightness %d", *val);
+
+	return 0;
+}
+
+static int st6422_set_brightness(struct gspca_dev *gspca_dev, __s32 val)
+{
+	int err;
+	struct sd *sd = (struct sd *) gspca_dev;
+	s32 *sensor_settings = sd->sensor_priv;
+
+	sensor_settings[BRIGHTNESS_IDX] = val;
+
+	if (!gspca_dev->streaming)
+		return 0;
+
+	/* val goes from 0 -> 31 */
+	PDEBUG(D_V4L2, "Set brightness to %d", val);
+	err = stv06xx_write_bridge(sd, 0x1432, val);
+	if (err < 0)
+		return err;
+
+	/* commit settings */
+	err = stv06xx_write_bridge(sd, 0x143f, 0x01);
+	return (err < 0) ? err : 0;
+}
+
+static int st6422_get_contrast(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	s32 *sensor_settings = sd->sensor_priv;
+
+	*val = sensor_settings[CONTRAST_IDX];
+
+	PDEBUG(D_V4L2, "Read contrast %d", *val);
+
+	return 0;
+}
+
+static int st6422_set_contrast(struct gspca_dev *gspca_dev, __s32 val)
+{
+	int err;
+	struct sd *sd = (struct sd *) gspca_dev;
+	s32 *sensor_settings = sd->sensor_priv;
+
+	sensor_settings[CONTRAST_IDX] = val;
+
+	if (!gspca_dev->streaming)
+		return 0;
+
+	/* Val goes from 0 -> 15 */
+	PDEBUG(D_V4L2, "Set contrast to %d\n", val);
+	err = stv06xx_write_bridge(sd, 0x143a, 0xf0 | val);
+	if (err < 0)
+		return err;
+
+	/* commit settings */
+	err = stv06xx_write_bridge(sd, 0x143f, 0x01);
+	return (err < 0) ? err : 0;
+}
+
+static int st6422_get_gain(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	s32 *sensor_settings = sd->sensor_priv;
+
+	*val = sensor_settings[GAIN_IDX];
+
+	PDEBUG(D_V4L2, "Read gain %d", *val);
+
+	return 0;
+}
+
+static int st6422_set_gain(struct gspca_dev *gspca_dev, __s32 val)
+{
+	int err;
+	struct sd *sd = (struct sd *) gspca_dev;
+	s32 *sensor_settings = sd->sensor_priv;
+
+	sensor_settings[GAIN_IDX] = val;
+
+	if (!gspca_dev->streaming)
+		return 0;
+
+	PDEBUG(D_V4L2, "Set gain to %d", val);
+
+	/* Set red, green, blue, gain */
+	err = stv06xx_write_bridge(sd, 0x0509, val);
+	if (err < 0)
+		return err;
+
+	err = stv06xx_write_bridge(sd, 0x050a, val);
+	if (err < 0)
+		return err;
+
+	err = stv06xx_write_bridge(sd, 0x050b, val);
+	if (err < 0)
+		return err;
+
+	/* 2 mystery writes */
+	err = stv06xx_write_bridge(sd, 0x050c, 0x2a);
+	if (err < 0)
+		return err;
+
+	err = stv06xx_write_bridge(sd, 0x050d, 0x01);
+	if (err < 0)
+		return err;
+
+	/* commit settings */
+	err = stv06xx_write_bridge(sd, 0x143f, 0x01);
+	return (err < 0) ? err : 0;
+}
+
+static int st6422_get_exposure(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+	s32 *sensor_settings = sd->sensor_priv;
+
+	*val = sensor_settings[EXPOSURE_IDX];
+
+	PDEBUG(D_V4L2, "Read exposure %d", *val);
+
+	return 0;
+}
+
+static int st6422_set_exposure(struct gspca_dev *gspca_dev, __s32 val)
+{
+	int err;
+	struct sd *sd = (struct sd *) gspca_dev;
+	s32 *sensor_settings = sd->sensor_priv;
+
+	sensor_settings[EXPOSURE_IDX] = val;
+
+	if (!gspca_dev->streaming)
+		return 0;
+
+	PDEBUG(D_V4L2, "Set exposure to %d\n", val);
+	err = stv06xx_write_bridge(sd, 0x143d, val & 0xff);
+	if (err < 0)
+		return err;
+
+	err = stv06xx_write_bridge(sd, 0x143e, val >> 8);
+	if (err < 0)
+		return err;
+
+	/* commit settings */
+	err = stv06xx_write_bridge(sd, 0x143f, 0x01);
+	return (err < 0) ? err : 0;
+}
diff --git a/drivers/media/video/gspca/stv06xx/stv06xx_st6422.h b/drivers/media/video/gspca/stv06xx/stv06xx_st6422.h
new file mode 100644
index 000000000000..b2d45fe50522
--- /dev/null
+++ b/drivers/media/video/gspca/stv06xx/stv06xx_st6422.h
@@ -0,0 +1,59 @@
+/*
+ * Support for the sensor part which is integrated (I think) into the
+ * st6422 stv06xx alike bridge, as its integrated there are no i2c writes
+ * but instead direct bridge writes.
+ *
+ * Copyright (c) 2009 Hans de Goede <hdegoede@redhat.com>
+ *
+ * Strongly based on qc-usb-messenger, which is:
+ * Copyright (c) 2001 Jean-Fredric Clere, Nikolas Zimmermann, Georg Acher
+ *		      Mark Cave-Ayland, Carlo E Prelz, Dick Streefland
+ * Copyright (c) 2002, 2003 Tuukka Toivonen
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#ifndef STV06XX_ST6422_H_
+#define STV06XX_ST6422_H_
+
+#include "stv06xx_sensor.h"
+
+static int st6422_probe(struct sd *sd);
+static int st6422_start(struct sd *sd);
+static int st6422_init(struct sd *sd);
+static int st6422_stop(struct sd *sd);
+static void st6422_disconnect(struct sd *sd);
+
+/* V4L2 controls supported by the driver */
+static int st6422_get_brightness(struct gspca_dev *gspca_dev, __s32 *val);
+static int st6422_set_brightness(struct gspca_dev *gspca_dev, __s32 val);
+static int st6422_get_contrast(struct gspca_dev *gspca_dev, __s32 *val);
+static int st6422_set_contrast(struct gspca_dev *gspca_dev, __s32 val);
+static int st6422_get_gain(struct gspca_dev *gspca_dev, __s32 *val);
+static int st6422_set_gain(struct gspca_dev *gspca_dev, __s32 val);
+static int st6422_get_exposure(struct gspca_dev *gspca_dev, __s32 *val);
+static int st6422_set_exposure(struct gspca_dev *gspca_dev, __s32 val);
+
+const struct stv06xx_sensor stv06xx_sensor_st6422 = {
+	.name = "ST6422",
+	.init = st6422_init,
+	.probe = st6422_probe,
+	.start = st6422_start,
+	.stop = st6422_stop,
+	.disconnect = st6422_disconnect,
+};
+
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From 98b1e9be882eff0f00bb5770ced9d9b24eb0238c Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Wed, 17 Jun 2009 18:41:01 -0300
Subject: V4L/DVB (12083): ov511: remove ov518 usb id's from the driver

ov511: remove ov518 usb id's from the driver, as they have not been working
ever since the decompression code got removed from the kernel, and they
are no supported by the gspca_ov519 module.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/ov511.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/media/video/ov511.c b/drivers/media/video/ov511.c
index 08cfd3e4ae8a..0bc2cf573c76 100644
--- a/drivers/media/video/ov511.c
+++ b/drivers/media/video/ov511.c
@@ -211,8 +211,6 @@ static const int i2c_detect_tries = 5;
 static struct usb_device_id device_table [] = {
 	{ USB_DEVICE(VEND_OMNIVISION, PROD_OV511) },
 	{ USB_DEVICE(VEND_OMNIVISION, PROD_OV511PLUS) },
-	{ USB_DEVICE(VEND_OMNIVISION, PROD_OV518) },
-	{ USB_DEVICE(VEND_OMNIVISION, PROD_OV518PLUS) },
 	{ USB_DEVICE(VEND_MATTEL, PROD_ME2CAM) },
 	{ }  /* Terminating entry */
 };
-- 
cgit v1.2.3-59-g8ed1b


From 0220f8870e66628f19c36bad813e881ebfaae7a6 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Wed, 17 Jun 2009 18:50:10 -0300
Subject: V4L/DVB (12084): ov511: mark as deprecated

Mark the v4l1 ov511 as deprecated as we now have ov511 support in
the gspca ov519 driver. Note we should really also keep track of this
in Documentation/feature-removal-schedule.txt, but that is not
part of the v4l-dvb tree.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/Kconfig | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig
index 94f440535c64..061e147f6f26 100644
--- a/drivers/media/video/Kconfig
+++ b/drivers/media/video/Kconfig
@@ -866,9 +866,13 @@ config USB_W9968CF
 	  module will be called w9968cf.
 
 config USB_OV511
-	tristate "USB OV511 Camera support"
+	tristate "USB OV511 Camera support (DEPRECATED)"
 	depends on VIDEO_V4L1
 	---help---
+	  This driver is DEPRECATED please use the gspca ov519 module
+	  instead. Note that for the ov511 / ov518 support of the gspca module
+	  you need atleast version 0.6.0 of libv4l.
+
 	  Say Y here if you want to connect this type of camera to your
 	  computer's USB port. See <file:Documentation/video4linux/ov511.txt>
 	  for more information and for a list of supported cameras.
-- 
cgit v1.2.3-59-g8ed1b


From e080fcd9298d544f3233d8c45304990be1920b3d Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Thu, 18 Jun 2009 05:03:16 -0300
Subject: V4L/DVB (12085): gspca_ov519: constify ov518 inititial register value
 tables

gspca_ov519: constify ov518 inititial register value tables

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/ov519.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/media/video/gspca/ov519.c b/drivers/media/video/gspca/ov519.c
index cb5f3c786db4..2f6e135d94bc 100644
--- a/drivers/media/video/gspca/ov519.c
+++ b/drivers/media/video/gspca/ov519.c
@@ -1863,7 +1863,7 @@ static int ov518_configure(struct gspca_dev *gspca_dev)
 	int rc;
 
 	/* For 518 and 518+ */
-	static struct ov_regvals init_518[] = {
+	const struct ov_regvals init_518[] = {
 		{ R51x_SYS_RESET,	0x40 },
 		{ R51x_SYS_INIT,	0xe1 },
 		{ R51x_SYS_RESET,	0x3e },
@@ -1874,7 +1874,7 @@ static int ov518_configure(struct gspca_dev *gspca_dev)
 		{ 0x5d,			0x03 },
 	};
 
-	static struct ov_regvals norm_518[] = {
+	const struct ov_regvals norm_518[] = {
 		{ R51x_SYS_SNAP,	0x02 }, /* Reset */
 		{ R51x_SYS_SNAP,	0x01 }, /* Enable */
 		{ 0x31, 		0x0f },
@@ -1887,7 +1887,7 @@ static int ov518_configure(struct gspca_dev *gspca_dev)
 		{ 0x2f,			0x80 },
 	};
 
-	static struct ov_regvals norm_518_p[] = {
+	const struct ov_regvals norm_518_p[] = {
 		{ R51x_SYS_SNAP,	0x02 }, /* Reset */
 		{ R51x_SYS_SNAP,	0x01 }, /* Enable */
 		{ 0x31, 		0x0f },
-- 
cgit v1.2.3-59-g8ed1b


From 9764398bdeef49414b37ef8bd35abfec1f44bd3e Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Thu, 18 Jun 2009 05:08:11 -0300
Subject: V4L/DVB (12086): gspca_sonixj: Fix control index numbering

The control index defines for the gspca_sonixj driver were numbered
wrong, causing us to disable the wrong controls on various sensors

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/sonixj.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/media/video/gspca/sonixj.c b/drivers/media/video/gspca/sonixj.c
index dc6a6f11354a..9e31d1bbc678 100644
--- a/drivers/media/video/gspca/sonixj.c
+++ b/drivers/media/video/gspca/sonixj.c
@@ -98,6 +98,7 @@ static int sd_setinfrared(struct gspca_dev *gspca_dev, __s32 val);
 static int sd_getinfrared(struct gspca_dev *gspca_dev, __s32 *val);
 
 static struct ctrl sd_ctrls[] = {
+#define BRIGHTNESS_IDX 0
 	{
 	    {
 		.id      = V4L2_CID_BRIGHTNESS,
@@ -113,6 +114,7 @@ static struct ctrl sd_ctrls[] = {
 	    .set = sd_setbrightness,
 	    .get = sd_getbrightness,
 	},
+#define CONTRAST_IDX 1
 	{
 	    {
 		.id      = V4L2_CID_CONTRAST,
@@ -128,6 +130,7 @@ static struct ctrl sd_ctrls[] = {
 	    .set = sd_setcontrast,
 	    .get = sd_getcontrast,
 	},
+#define COLOR_IDX 2
 	{
 	    {
 		.id      = V4L2_CID_SATURATION,
@@ -142,6 +145,7 @@ static struct ctrl sd_ctrls[] = {
 	    .set = sd_setcolors,
 	    .get = sd_getcolors,
 	},
+#define BLUE_BALANCE_IDX 3
 	{
 	    {
 		.id      = V4L2_CID_BLUE_BALANCE,
@@ -156,6 +160,7 @@ static struct ctrl sd_ctrls[] = {
 	    .set = sd_setblue_balance,
 	    .get = sd_getblue_balance,
 	},
+#define RED_BALANCE_IDX 4
 	{
 	    {
 		.id      = V4L2_CID_RED_BALANCE,
@@ -170,6 +175,7 @@ static struct ctrl sd_ctrls[] = {
 	    .set = sd_setred_balance,
 	    .get = sd_getred_balance,
 	},
+#define GAMMA_IDX 5
 	{
 	    {
 		.id      = V4L2_CID_GAMMA,
@@ -184,7 +190,7 @@ static struct ctrl sd_ctrls[] = {
 	    .set = sd_setgamma,
 	    .get = sd_getgamma,
 	},
-#define AUTOGAIN_IDX 5
+#define AUTOGAIN_IDX 6
 	{
 	    {
 		.id      = V4L2_CID_AUTOGAIN,
@@ -200,7 +206,7 @@ static struct ctrl sd_ctrls[] = {
 	    .get = sd_getautogain,
 	},
 /* ov7630/ov7648 only */
-#define VFLIP_IDX 6
+#define VFLIP_IDX 7
 	{
 	    {
 		.id      = V4L2_CID_VFLIP,
@@ -216,7 +222,7 @@ static struct ctrl sd_ctrls[] = {
 	    .get = sd_getvflip,
 	},
 /* mt9v111 only */
-#define INFRARED_IDX 7
+#define INFRARED_IDX 8
 	{
 	    {
 		.id      = V4L2_CID_INFRARED,
-- 
cgit v1.2.3-59-g8ed1b


From cc7b5b573feb5edfe68c028bc1ea383dab37dde2 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Thu, 18 Jun 2009 05:14:42 -0300
Subject: V4L/DVB (12087): gspca_sonixj: enable support for 0c45:613e camera

gspca_sonixj: enable support for 0c45:613e camera, and slightly tweak
the ov7630 register init values for a much better picture.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/sonixj.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/media/video/gspca/sonixj.c b/drivers/media/video/gspca/sonixj.c
index 9e31d1bbc678..184bb924f456 100644
--- a/drivers/media/video/gspca/sonixj.c
+++ b/drivers/media/video/gspca/sonixj.c
@@ -610,7 +610,9 @@ static const u8 ov7630_sensor_init[][8] = {
 /* win: i2c_r from 00 to 80 */
 	{0xd1, 0x21, 0x03, 0x80, 0x10, 0x20, 0x80, 0x10},
 	{0xb1, 0x21, 0x0c, 0x20, 0x20, 0x00, 0x00, 0x10},
-	{0xd1, 0x21, 0x11, 0x00, 0x48, 0xc0, 0x00, 0x10},
+/* HDG: 0x11 was 0x00 change to 0x01 for better exposure (15 fps instead of 30)
+	0x13 was 0xc0 change to 0xc3 for auto gain and exposure */
+	{0xd1, 0x21, 0x11, 0x01, 0x48, 0xc3, 0x00, 0x10},
 	{0xb1, 0x21, 0x15, 0x80, 0x03, 0x00, 0x00, 0x10},
 	{0xd1, 0x21, 0x17, 0x1b, 0xbd, 0x05, 0xf6, 0x10},
 	{0xa1, 0x21, 0x1b, 0x04, 0x00, 0x00, 0x00, 0x10},
@@ -644,7 +646,6 @@ static const u8 ov7630_sensor_init[][8] = {
 	{0xa1, 0x21, 0x00, 0x00, 0x00, 0x00, 0x00, 0x10},
 	{0xb1, 0x21, 0x01, 0x80, 0x80, 0x00, 0x00, 0x10},
 /* */
-	{0xa1, 0x21, 0x11, 0x00, 0x00, 0x00, 0x00, 0x10},
 	{0xa1, 0x21, 0x2a, 0x88, 0x00, 0x00, 0x00, 0x10},
 	{0xa1, 0x21, 0x2b, 0x34, 0x00, 0x00, 0x00, 0x10},
 /* */
@@ -2239,7 +2240,7 @@ static const __devinitdata struct usb_device_id device_table[] = {
 	{USB_DEVICE(0x0c45, 0x613b), BSI(SN9C120, OV7660, 0x21)},
 #endif
 	{USB_DEVICE(0x0c45, 0x613c), BSI(SN9C120, HV7131R, 0x11)},
-/*	{USB_DEVICE(0x0c45, 0x613e), BSI(SN9C120, OV7630, 0x??)}, */
+	{USB_DEVICE(0x0c45, 0x613e), BSI(SN9C120, OV7630, 0x21)},
 	{USB_DEVICE(0x0c45, 0x6143), BSI(SN9C120, SP80708, 0x18)},
 	{}
 };
-- 
cgit v1.2.3-59-g8ed1b


From 119893b2dfb18515bfdcc5edb83422e6aa126a86 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Thu, 18 Jun 2009 05:20:51 -0300
Subject: V4L/DVB (12088): Mark the v4l1 uvcvideo quickcam messenger driver as
 deprecated

Mark the v4l1 uvcvideo quickcam messenger driver as deprecated, the one
cam it supports, is now also supported by the v4l2 gspca stv06xx driver.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/usbvideo/Kconfig | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/media/video/usbvideo/Kconfig b/drivers/media/video/usbvideo/Kconfig
index e4cb99c1f94b..adb1c044ad7d 100644
--- a/drivers/media/video/usbvideo/Kconfig
+++ b/drivers/media/video/usbvideo/Kconfig
@@ -38,10 +38,13 @@ config USB_KONICAWC
 	  module will be called konicawc.
 
 config USB_QUICKCAM_MESSENGER
-	tristate "USB Logitech Quickcam Messenger"
+	tristate "USB Logitech Quickcam Messenger (DEPRECATED)"
 	depends on VIDEO_V4L1
 	select VIDEO_USBVIDEO
 	---help---
+	  This driver is DEPRECATED please use the gspca stv06xx module
+	  instead.
+
 	  Say Y or M here to enable support for the USB Logitech Quickcam
 	  Messenger webcam.
 
-- 
cgit v1.2.3-59-g8ed1b


From a5d1cc39fee739cf4fc2a1f43da812c50de9d3d6 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Thu, 18 Jun 2009 06:03:20 -0300
Subject: V4L/DVB (12089): gspca_sonixj: increase 640x480 frame-buffersize

gspca_sonixj: increase 640x480 frame-buffersize, as I was getting buffer
overflows during my testing of a "Premier" 0c45:613e cam

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/sonixj.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/media/video/gspca/sonixj.c b/drivers/media/video/gspca/sonixj.c
index 184bb924f456..46988120ec65 100644
--- a/drivers/media/video/gspca/sonixj.c
+++ b/drivers/media/video/gspca/sonixj.c
@@ -274,7 +274,8 @@ static const struct v4l2_pix_format vga_mode[] = {
 		.priv = 1},
 	{640, 480, V4L2_PIX_FMT_JPEG, V4L2_FIELD_NONE,
 		.bytesperline = 640,
-		.sizeimage = 640 * 480 * 3 / 8 + 590,
+		/* Note 3 / 8 is not large enough, not even 5 / 8 is ?! */
+		.sizeimage = 640 * 480 * 3 / 4 + 590,
 		.colorspace = V4L2_COLORSPACE_JPEG,
 		.priv = 0},
 };
-- 
cgit v1.2.3-59-g8ed1b


From 1fec747cd389b4812a9932a1416d76e8a53596b2 Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Thu, 18 Jun 2009 06:05:07 -0300
Subject: V4L/DVB (12090): gspca_sonixj: enable autogain control for the ov7620

gspca_sonixj: enable autogain control for the ov7620, and not only
make it enable autogain but also auto exposure (and do the
same for the ov7648).

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/sonixj.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/media/video/gspca/sonixj.c b/drivers/media/video/gspca/sonixj.c
index 46988120ec65..6f475b471fb9 100644
--- a/drivers/media/video/gspca/sonixj.c
+++ b/drivers/media/video/gspca/sonixj.c
@@ -251,7 +251,7 @@ static __u32 ctrl_dis[] = {
 						/* SENSOR_MT9V111 3 */
 	(1 << INFRARED_IDX) | (1 << VFLIP_IDX),
 						/* SENSOR_OM6802 4 */
-	(1 << AUTOGAIN_IDX) | (1 << INFRARED_IDX),
+	(1 << INFRARED_IDX),
 						/* SENSOR_OV7630 5 */
 	(1 << INFRARED_IDX),
 						/* SENSOR_OV7648 6 */
@@ -1577,7 +1577,7 @@ static void setautogain(struct gspca_dev *gspca_dev)
 		else
 			comb = 0xa0;
 		if (sd->autogain)
-			comb |= 0x02;
+			comb |= 0x03;
 		i2c_w1(&sd->gspca_dev, 0x13, comb);
 		return;
 	    }
-- 
cgit v1.2.3-59-g8ed1b


From 37c6dbe290c05023b47f52528e30ce51336b93eb Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Thu, 18 Jun 2009 07:35:36 -0300
Subject: V4L/DVB (12091): gspca_sonixj: Add light frequency control

gspca_sonixj: Add light frequency control

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/sonixj.c | 133 ++++++++++++++++++++++++++++++++++---
 1 file changed, 123 insertions(+), 10 deletions(-)

diff --git a/drivers/media/video/gspca/sonixj.c b/drivers/media/video/gspca/sonixj.c
index 6f475b471fb9..7daa60299c0e 100644
--- a/drivers/media/video/gspca/sonixj.c
+++ b/drivers/media/video/gspca/sonixj.c
@@ -46,6 +46,7 @@ struct sd {
 	u8 gamma;
 	u8 vflip;			/* ov7630/ov7648 only */
 	u8 infrared;			/* mt9v111 only */
+	u8 freq;			/* ov76xx only */
 	u8 quality;			/* image quality */
 #define QUALITY_MIN 60
 #define QUALITY_MAX 95
@@ -96,6 +97,8 @@ static int sd_setvflip(struct gspca_dev *gspca_dev, __s32 val);
 static int sd_getvflip(struct gspca_dev *gspca_dev, __s32 *val);
 static int sd_setinfrared(struct gspca_dev *gspca_dev, __s32 val);
 static int sd_getinfrared(struct gspca_dev *gspca_dev, __s32 *val);
+static int sd_setfreq(struct gspca_dev *gspca_dev, __s32 val);
+static int sd_getfreq(struct gspca_dev *gspca_dev, __s32 *val);
 
 static struct ctrl sd_ctrls[] = {
 #define BRIGHTNESS_IDX 0
@@ -237,19 +240,35 @@ static struct ctrl sd_ctrls[] = {
 	    .set = sd_setinfrared,
 	    .get = sd_getinfrared,
 	},
+/* ov7630/ov7648/ov7660 only */
+#define FREQ_IDX 9
+	{
+	    {
+		.id	 = V4L2_CID_POWER_LINE_FREQUENCY,
+		.type    = V4L2_CTRL_TYPE_MENU,
+		.name    = "Light frequency filter",
+		.minimum = 0,
+		.maximum = 2,	/* 0: 0, 1: 50Hz, 2:60Hz */
+		.step    = 1,
+#define FREQ_DEF 2
+		.default_value = FREQ_DEF,
+	    },
+	    .set = sd_setfreq,
+	    .get = sd_getfreq,
+	},
 };
 
 /* table of the disabled controls */
 static __u32 ctrl_dis[] = {
-	(1 << INFRARED_IDX) | (1 << VFLIP_IDX),
+	(1 << INFRARED_IDX) | (1 << VFLIP_IDX) | (1 << FREQ_IDX),
 						/* SENSOR_HV7131R 0 */
-	(1 << INFRARED_IDX) | (1 << VFLIP_IDX),
+	(1 << INFRARED_IDX) | (1 << VFLIP_IDX) | (1 << FREQ_IDX),
 						/* SENSOR_MI0360 1 */
-	(1 << INFRARED_IDX) | (1 << VFLIP_IDX),
+	(1 << INFRARED_IDX) | (1 << VFLIP_IDX) | (1 << FREQ_IDX),
 						/* SENSOR_MO4000 2 */
-	(1 << VFLIP_IDX),
+	(1 << VFLIP_IDX) | (1 << FREQ_IDX),
 						/* SENSOR_MT9V111 3 */
-	(1 << INFRARED_IDX) | (1 << VFLIP_IDX),
+	(1 << INFRARED_IDX) | (1 << VFLIP_IDX) | (1 << FREQ_IDX),
 						/* SENSOR_OM6802 4 */
 	(1 << INFRARED_IDX),
 						/* SENSOR_OV7630 5 */
@@ -257,8 +276,8 @@ static __u32 ctrl_dis[] = {
 						/* SENSOR_OV7648 6 */
 	(1 << AUTOGAIN_IDX) | (1 << INFRARED_IDX) | (1 << VFLIP_IDX),
 						/* SENSOR_OV7660 7 */
-	(1 << AUTOGAIN_IDX) | (1 << INFRARED_IDX) | (1 << VFLIP_IDX),
-						/* SENSOR_SP80708 8 */
+	(1 << AUTOGAIN_IDX) | (1 << INFRARED_IDX) | (1 << VFLIP_IDX) |
+			      (1 << FREQ_IDX),	/* SENSOR_SP80708 8 */
 };
 
 static const struct v4l2_pix_format vga_mode[] = {
@@ -647,8 +666,8 @@ static const u8 ov7630_sensor_init[][8] = {
 	{0xa1, 0x21, 0x00, 0x00, 0x00, 0x00, 0x00, 0x10},
 	{0xb1, 0x21, 0x01, 0x80, 0x80, 0x00, 0x00, 0x10},
 /* */
-	{0xa1, 0x21, 0x2a, 0x88, 0x00, 0x00, 0x00, 0x10},
-	{0xa1, 0x21, 0x2b, 0x34, 0x00, 0x00, 0x00, 0x10},
+/*	{0xa1, 0x21, 0x2a, 0x88, 0x00, 0x00, 0x00, 0x10}, * set by setfreq */
+/*	{0xa1, 0x21, 0x2b, 0x34, 0x00, 0x00, 0x00, 0x10}, * set by setfreq */
 /* */
 	{0xa1, 0x21, 0x10, 0x83, 0x00, 0x00, 0x00, 0x10},
 /*	{0xb1, 0x21, 0x01, 0x88, 0x70, 0x00, 0x00, 0x10}, */
@@ -681,7 +700,7 @@ static const u8 ov7648_sensor_init[][8] = {
 	{0xd1, 0x21, 0x21, 0x86, 0x00, 0xde, 0xa0, 0x10},
 /*	{0xd1, 0x21, 0x25, 0x80, 0x32, 0xfe, 0xa0, 0x10}, jfm done */
 /*	{0xd1, 0x21, 0x29, 0x00, 0x91, 0x00, 0x88, 0x10}, jfm done */
-	{0xb1, 0x21, 0x2d, 0x85, 0x00, 0x00, 0x00, 0x10},
+/*	{0xb1, 0x21, 0x2d, 0x85, 0x00, 0x00, 0x00, 0x10}, set by setfreq */
 /*...*/
 /*	{0xa1, 0x21, 0x12, 0x08, 0x00, 0x00, 0x00, 0x10}, jfm done */
 /*	{0xa1, 0x21, 0x75, 0x06, 0x00, 0x00, 0x00, 0x10},   * COMN
@@ -1307,6 +1326,7 @@ static int sd_config(struct gspca_dev *gspca_dev,
 	else
 		sd->vflip = 1;
 	sd->infrared = INFRARED_DEF;
+	sd->freq = FREQ_DEF;
 	sd->quality = QUALITY_DEF;
 	sd->jpegqual = 80;
 
@@ -1610,6 +1630,58 @@ static void setinfrared(struct sd *sd)
 		sd->infrared ? 0x66 : 0x64);
 }
 
+static void setfreq(struct gspca_dev *gspca_dev)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	if (sd->sensor == SENSOR_OV7660) {
+		switch (sd->freq) {
+		case 0: /* Banding filter disabled */
+			i2c_w1(gspca_dev, 0x13, 0xdf);
+			break;
+		case 1: /* 50 hz */
+			i2c_w1(gspca_dev, 0x13, 0xff);
+			i2c_w1(gspca_dev, 0x3b, 0x0a);
+			break;
+		case 2: /* 60 hz */
+			i2c_w1(gspca_dev, 0x13, 0xff);
+			i2c_w1(gspca_dev, 0x3b, 0x02);
+			break;
+		}
+	} else {
+		u8 reg2a = 0, reg2b = 0, reg2d = 0;
+
+		/* Get reg2a / reg2d base values */
+		switch (sd->sensor) {
+		case SENSOR_OV7630:
+			reg2a = 0x08;
+			reg2d = 0x01;
+			break;
+		case SENSOR_OV7648:
+			reg2a = 0x11;
+			reg2d = 0x81;
+			break;
+		}
+
+		switch (sd->freq) {
+		case 0: /* Banding filter disabled */
+			break;
+		case 1: /* 50 hz (filter on and framerate adj) */
+			reg2a |= 0x80;
+			reg2b = 0xac;
+			reg2d |= 0x04;
+			break;
+		case 2: /* 60 hz (filter on, no framerate adj) */
+			reg2a |= 0x80;
+			reg2d |= 0x04;
+			break;
+		}
+		i2c_w1(gspca_dev, 0x2a, reg2a);
+		i2c_w1(gspca_dev, 0x2b, reg2b);
+		i2c_w1(gspca_dev, 0x2d, reg2d);
+	}
+}
+
 static void setjpegqual(struct gspca_dev *gspca_dev)
 {
 	struct sd *sd = (struct sd *) gspca_dev;
@@ -1836,6 +1908,7 @@ static int sd_start(struct gspca_dev *gspca_dev)
 	setbrightness(gspca_dev);
 	setcontrast(gspca_dev);
 	setautogain(gspca_dev);
+	setfreq(gspca_dev);
 	return 0;
 }
 
@@ -2139,6 +2212,24 @@ static int sd_getinfrared(struct gspca_dev *gspca_dev, __s32 *val)
 	return 0;
 }
 
+static int sd_setfreq(struct gspca_dev *gspca_dev, __s32 val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	sd->freq = val;
+	if (gspca_dev->streaming)
+		setfreq(gspca_dev);
+	return 0;
+}
+
+static int sd_getfreq(struct gspca_dev *gspca_dev, __s32 *val)
+{
+	struct sd *sd = (struct sd *) gspca_dev;
+
+	*val = sd->freq;
+	return 0;
+}
+
 static int sd_set_jcomp(struct gspca_dev *gspca_dev,
 			struct v4l2_jpegcompression *jcomp)
 {
@@ -2167,6 +2258,27 @@ static int sd_get_jcomp(struct gspca_dev *gspca_dev,
 	return 0;
 }
 
+static int sd_querymenu(struct gspca_dev *gspca_dev,
+			struct v4l2_querymenu *menu)
+{
+	switch (menu->id) {
+	case V4L2_CID_POWER_LINE_FREQUENCY:
+		switch (menu->index) {
+		case 0:		/* V4L2_CID_POWER_LINE_FREQUENCY_DISABLED */
+			strcpy((char *) menu->name, "NoFliker");
+			return 0;
+		case 1:		/* V4L2_CID_POWER_LINE_FREQUENCY_50HZ */
+			strcpy((char *) menu->name, "50 Hz");
+			return 0;
+		case 2:		/* V4L2_CID_POWER_LINE_FREQUENCY_60HZ */
+			strcpy((char *) menu->name, "60 Hz");
+			return 0;
+		}
+		break;
+	}
+	return -EINVAL;
+}
+
 /* sub-driver description */
 static const struct sd_desc sd_desc = {
 	.name = MODULE_NAME,
@@ -2181,6 +2293,7 @@ static const struct sd_desc sd_desc = {
 	.dq_callback = do_autogain,
 	.get_jcomp = sd_get_jcomp,
 	.set_jcomp = sd_set_jcomp,
+	.querymenu = sd_querymenu,
 };
 
 /* -- module initialisation -- */
-- 
cgit v1.2.3-59-g8ed1b


From f800952c21157f11a5510d9cf700c9a7ba30800d Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Thu, 18 Jun 2009 14:29:20 -0300
Subject: V4L/DVB (12092): gspca_sonixj + ov7630: invert vflip control instead
 of changing default

gspca_sonixj + ov7630 had the default value for flip enabled, as otherwise
the picture is upside down. It is better to instead invert the meaning
of the control in the set function, and have the default be no vflip,
as one would expect vflip enabled to be upside down.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/sonixj.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/media/video/gspca/sonixj.c b/drivers/media/video/gspca/sonixj.c
index 7daa60299c0e..46ec24e40458 100644
--- a/drivers/media/video/gspca/sonixj.c
+++ b/drivers/media/video/gspca/sonixj.c
@@ -218,7 +218,7 @@ static struct ctrl sd_ctrls[] = {
 		.minimum = 0,
 		.maximum = 1,
 		.step    = 1,
-#define VFLIP_DEF 0			/* vflip def = 1 for ov7630 */
+#define VFLIP_DEF 0
 		.default_value = VFLIP_DEF,
 	    },
 	    .set = sd_setvflip,
@@ -1321,10 +1321,7 @@ static int sd_config(struct gspca_dev *gspca_dev,
 	sd->gamma = GAMMA_DEF;
 	sd->autogain = AUTOGAIN_DEF;
 	sd->ag_cnt = -1;
-	if (sd->sensor != SENSOR_OV7630)
-		sd->vflip = 0;
-	else
-		sd->vflip = 1;
+	sd->vflip = VFLIP_DEF;
 	sd->infrared = INFRARED_DEF;
 	sd->freq = FREQ_DEF;
 	sd->quality = QUALITY_DEF;
@@ -1613,12 +1610,15 @@ static void setvflip(struct sd *sd)
 {
 	u8 comn;
 
-	if (sd->sensor == SENSOR_OV7630)
+	if (sd->sensor == SENSOR_OV7630) {
 		comn = 0x02;
-	else
+		if (!sd->vflip)
+			comn |= 0x80;
+	} else {
 		comn = 0x06;
-	if (sd->vflip)
-		comn |= 0x80;
+		if (sd->vflip)
+			comn |= 0x80;
+	}
 	i2c_w1(&sd->gspca_dev, 0x75, comn);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 3fb4a57b494e05dba4d1305e2347c6633b76c20e Mon Sep 17 00:00:00 2001
From: Hans de Goede <hdegoede@redhat.com>
Date: Thu, 18 Jun 2009 14:31:36 -0300
Subject: V4L/DVB (12093): gspca_sonixj: Name saturation control saturation,
 not color

Name saturation control saturation, not color and make the default
less saturated (the old default was overdoing it).

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/gspca/sonixj.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/media/video/gspca/sonixj.c b/drivers/media/video/gspca/sonixj.c
index 46ec24e40458..0d02f41fa7d0 100644
--- a/drivers/media/video/gspca/sonixj.c
+++ b/drivers/media/video/gspca/sonixj.c
@@ -138,11 +138,11 @@ static struct ctrl sd_ctrls[] = {
 	    {
 		.id      = V4L2_CID_SATURATION,
 		.type    = V4L2_CTRL_TYPE_INTEGER,
-		.name    = "Color",
+		.name    = "Saturation",
 		.minimum = 0,
 		.maximum = 40,
 		.step    = 1,
-#define COLOR_DEF 32
+#define COLOR_DEF 25
 		.default_value = COLOR_DEF,
 	    },
 	    .set = sd_setcolors,
-- 
cgit v1.2.3-59-g8ed1b


From 0cde9b2533d6fe79307173f24209228aaf34bc98 Mon Sep 17 00:00:00 2001
From: "Igor M. Liplianin" <liplianin@netup.ru>
Date: Sun, 14 Jun 2009 13:17:15 -0300
Subject: V4L/DVB (12095): Change lnbh24 configure bits for NetUP card.

Signed-off-by: Igor M. Liplianin <liplianin@netup.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/cx23885/cx23885-dvb.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/media/video/cx23885/cx23885-dvb.c b/drivers/media/video/cx23885/cx23885-dvb.c
index e236df23370e..236eea0c41e5 100644
--- a/drivers/media/video/cx23885/cx23885-dvb.c
+++ b/drivers/media/video/cx23885/cx23885-dvb.c
@@ -736,7 +736,8 @@ static int dvb_register(struct cx23885_tsport *port)
 					if (!dvb_attach(lnbh24_attach,
 							fe0->dvb.frontend,
 							&i2c_bus->i2c_adap,
-							LNBH24_PCL, 0, 0x09))
+							LNBH24_PCL,
+							LNBH24_TTX, 0x09))
 						printk(KERN_ERR
 							"No LNBH24 found!\n");
 
@@ -756,7 +757,8 @@ static int dvb_register(struct cx23885_tsport *port)
 					if (!dvb_attach(lnbh24_attach,
 							fe0->dvb.frontend,
 							&i2c_bus->i2c_adap,
-							LNBH24_PCL, 0, 0x0a))
+							LNBH24_PCL,
+							LNBH24_TTX, 0x0a))
 						printk(KERN_ERR
 							"No LNBH24 found!\n");
 
-- 
cgit v1.2.3-59-g8ed1b


From 68191edeb50773993f4a05651b0a085bd110fbeb Mon Sep 17 00:00:00 2001
From: Abylay Ospan <aospan@netup.ru>
Date: Sun, 14 Jun 2009 14:10:05 -0300
Subject: V4L/DVB (12096): Bug fix: stv0900 register read must using i2c in one
 transaction

Signed-off-by: Abylay Ospan <aospan@netup.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/dvb/frontends/stv0900_core.c | 40 +++++++++++++++---------------
 1 file changed, 20 insertions(+), 20 deletions(-)

diff --git a/drivers/media/dvb/frontends/stv0900_core.c b/drivers/media/dvb/frontends/stv0900_core.c
index 8499bcf7f251..4daec8ad92c3 100644
--- a/drivers/media/dvb/frontends/stv0900_core.c
+++ b/drivers/media/dvb/frontends/stv0900_core.c
@@ -149,31 +149,31 @@ void stv0900_write_reg(struct stv0900_internal *i_params, u16 reg_addr,
 		dprintk(KERN_ERR "%s: i2c error %d\n", __func__, ret);
 }
 
-u8 stv0900_read_reg(struct stv0900_internal *i_params, u16 reg_addr)
+u8 stv0900_read_reg(struct stv0900_internal *i_params, u16 reg)
 {
-	u8 data[2];
 	int ret;
-	struct i2c_msg i2cmsg = {
-		.addr  = i_params->i2c_addr,
-		.flags = 0,
-		.len   = 2,
-		.buf   = data,
+	u8 b0[] = { MSB(reg), LSB(reg) };
+	u8 buf = 0;
+	struct i2c_msg msg[] = {
+		{
+			.addr	= i_params->i2c_addr,
+			.flags	= 0,
+			.buf = b0,
+			.len = 2,
+		}, {
+			.addr	= i_params->i2c_addr,
+			.flags	= I2C_M_RD,
+			.buf = &buf,
+			.len = 1,
+		},
 	};
 
-	data[0] = MSB(reg_addr);
-	data[1] = LSB(reg_addr);
-
-	ret = i2c_transfer(i_params->i2c_adap, &i2cmsg, 1);
-	if (ret != 1)
-		dprintk(KERN_ERR "%s: i2c error %d\n", __func__, ret);
-
-	i2cmsg.flags = I2C_M_RD;
-	i2cmsg.len = 1;
-	ret = i2c_transfer(i_params->i2c_adap, &i2cmsg, 1);
-	if (ret != 1)
-		dprintk(KERN_ERR "%s: i2c error %d\n", __func__, ret);
+	ret = i2c_transfer(i_params->i2c_adap, msg, 2);
+	if (ret != 2)
+		dprintk(KERN_ERR "%s: i2c error %d, reg[0x%02x]\n",
+				__func__, ret, reg);
 
-	return data[0];
+	return buf;
 }
 
 void extract_mask_pos(u32 label, u8 *mask, u8 *pos)
-- 
cgit v1.2.3-59-g8ed1b


From ee1ebcfea6ee16491f88e8023554dd214e1ba85c Mon Sep 17 00:00:00 2001
From: Abylay Ospan <aospan@netup.ru>
Date: Mon, 8 Jun 2009 04:31:26 -0300
Subject: V4L/DVB (12097): Implement reading uncorrected blocks for stv0900

Signed-off-by: Abylay Ospan <aospan@netup.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/dvb/frontends/stv0900_core.c | 39 ++++++++++++++++++++++++++++++
 1 file changed, 39 insertions(+)

diff --git a/drivers/media/dvb/frontends/stv0900_core.c b/drivers/media/dvb/frontends/stv0900_core.c
index 4daec8ad92c3..9ab4f3014673 100644
--- a/drivers/media/dvb/frontends/stv0900_core.c
+++ b/drivers/media/dvb/frontends/stv0900_core.c
@@ -712,6 +712,44 @@ static s32 stv0900_carr_get_quality(struct dvb_frontend *fe,
 	return c_n;
 }
 
+static int stv0900_read_ucblocks(struct dvb_frontend *fe, u32 * ucblocks)
+{
+	struct stv0900_state *state = fe->demodulator_priv;
+	struct stv0900_internal *i_params = state->internal;
+	enum fe_stv0900_demod_num demod = state->demod;
+	u8 err_val1, err_val0;
+	s32 err_field1, err_field0;
+	u32 header_err_val = 0;
+
+	*ucblocks = 0x0;
+	if (stv0900_get_standard(fe, demod) == STV0900_DVBS2_STANDARD) {
+		/* DVB-S2 delineator errors count */
+
+		/* retreiving number for errnous headers */
+		dmd_reg(err_field0, R0900_P1_BBFCRCKO0,
+					R0900_P2_BBFCRCKO0);
+		dmd_reg(err_field1, R0900_P1_BBFCRCKO1,
+					R0900_P2_BBFCRCKO1);
+
+		err_val1 = stv0900_read_reg(i_params, err_field1);
+		err_val0 = stv0900_read_reg(i_params, err_field0);
+		header_err_val = (err_val1<<8) | err_val0;
+
+		/* retreiving number for errnous packets */
+		dmd_reg(err_field0, R0900_P1_UPCRCKO0,
+					R0900_P2_UPCRCKO0);
+		dmd_reg(err_field1, R0900_P1_UPCRCKO1,
+					R0900_P2_UPCRCKO1);
+
+		err_val1 = stv0900_read_reg(i_params, err_field1);
+		err_val0 = stv0900_read_reg(i_params, err_field0);
+		*ucblocks = (err_val1<<8) | err_val0;
+		*ucblocks += header_err_val;
+	}
+
+	return 0;
+}
+
 static int stv0900_read_snr(struct dvb_frontend *fe, u16 *snr)
 {
 	*snr = stv0900_carr_get_quality(fe,
@@ -1882,6 +1920,7 @@ static struct dvb_frontend_ops stv0900_ops = {
 	.read_ber			= stv0900_read_ber,
 	.read_signal_strength		= stv0900_read_signal_strength,
 	.read_snr			= stv0900_read_snr,
+	.read_ucblocks                  = stv0900_read_ucblocks,
 };
 
 struct dvb_frontend *stv0900_attach(const struct stv0900_config *config,
-- 
cgit v1.2.3-59-g8ed1b


From f867c3f4eab1d5006df4f3734fab1134feffbeba Mon Sep 17 00:00:00 2001
From: "Igor M. Liplianin" <liplianin@netup.ru>
Date: Fri, 19 Jun 2009 05:45:23 -0300
Subject: V4L/DVB (12098): Create table for customize stv0900 ts registers.

Signed-off-by: Igor M. Liplianin <liplianin@netup.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/dvb/frontends/stv0900.h      |  7 ++++++-
 drivers/media/dvb/frontends/stv0900_core.c | 21 +++++++++++++++++++--
 drivers/media/dvb/frontends/stv0900_priv.h |  2 ++
 drivers/media/video/cx23885/cx23885-dvb.c  | 17 +++++++++++++++--
 4 files changed, 42 insertions(+), 5 deletions(-)

diff --git a/drivers/media/dvb/frontends/stv0900.h b/drivers/media/dvb/frontends/stv0900.h
index 8a1332c2031d..bf4e9b633044 100644
--- a/drivers/media/dvb/frontends/stv0900.h
+++ b/drivers/media/dvb/frontends/stv0900.h
@@ -29,6 +29,11 @@
 #include <linux/dvb/frontend.h>
 #include "dvb_frontend.h"
 
+struct stv0900_reg {
+	u16 addr;
+	u8  val;
+};
+
 struct stv0900_config {
 	u8 demod_address;
 	u32 xtal;
@@ -38,7 +43,7 @@ struct stv0900_config {
 
 	u8 path1_mode;
 	u8 path2_mode;
-
+	struct stv0900_reg *ts_config_regs;
 	u8 tun1_maddress;/* 0, 1, 2, 3 for 0xc0, 0xc2, 0xc4, 0xc6 */
 	u8 tun2_maddress;
 	u8 tun1_adc;/* 1 for stv6110, 2 for stb6100 */
diff --git a/drivers/media/dvb/frontends/stv0900_core.c b/drivers/media/dvb/frontends/stv0900_core.c
index 9ab4f3014673..1da045fbb4ef 100644
--- a/drivers/media/dvb/frontends/stv0900_core.c
+++ b/drivers/media/dvb/frontends/stv0900_core.c
@@ -1393,7 +1393,7 @@ static enum fe_stv0900_error stv0900_init_internal(struct dvb_frontend *fe,
 	struct stv0900_state *state = fe->demodulator_priv;
 	enum fe_stv0900_error error = STV0900_NO_ERROR;
 	enum fe_stv0900_error demodError = STV0900_NO_ERROR;
-	int selosci;
+	int selosci, i;
 
 	struct stv0900_inode *temp_int = find_inode(state->i2c_adap,
 						state->config->demod_address);
@@ -1440,7 +1440,23 @@ static enum fe_stv0900_error stv0900_init_internal(struct dvb_frontend *fe,
 				stv0900_write_bits(state->internal, F0900_P1_ROLLOFF_CONTROL, p_init->rolloff);
 				stv0900_write_bits(state->internal, F0900_P2_ROLLOFF_CONTROL, p_init->rolloff);
 
-				stv0900_set_ts_parallel_serial(state->internal, p_init->path1_ts_clock, p_init->path2_ts_clock);
+				state->internal->ts_config = p_init->ts_config;
+				if (state->internal->ts_config == NULL)
+					stv0900_set_ts_parallel_serial(state->internal,
+							p_init->path1_ts_clock,
+							p_init->path2_ts_clock);
+				else {
+					for (i = 0; state->internal->ts_config[i].addr != 0xffff; i++)
+						stv0900_write_reg(state->internal,
+								state->internal->ts_config[i].addr,
+								state->internal->ts_config[i].val);
+
+					stv0900_write_bits(state->internal, F0900_P2_RST_HWARE, 1);
+					stv0900_write_bits(state->internal, F0900_P2_RST_HWARE, 0);
+					stv0900_write_bits(state->internal, F0900_P1_RST_HWARE, 1);
+					stv0900_write_bits(state->internal, F0900_P1_RST_HWARE, 0);
+				}
+
 				stv0900_write_bits(state->internal, F0900_P1_TUN_MADDRESS, p_init->tun1_maddress);
 				switch (p_init->tuner1_adc) {
 				case 1:
@@ -1954,6 +1970,7 @@ struct dvb_frontend *stv0900_attach(const struct stv0900_config *config,
 		init_params.tun1_iq_inversion	= STV0900_IQ_NORMAL;
 		init_params.tuner1_adc		= config->tun1_adc;
 		init_params.path2_ts_clock	= config->path2_mode;
+		init_params.ts_config		= config->ts_config_regs;
 		init_params.tun2_maddress	= config->tun2_maddress;
 		init_params.tuner2_adc		= config->tun2_adc;
 		init_params.tun2_iq_inversion	= STV0900_IQ_SWAPPED;
diff --git a/drivers/media/dvb/frontends/stv0900_priv.h b/drivers/media/dvb/frontends/stv0900_priv.h
index 67dc8ec634e2..5ed7a145c7d3 100644
--- a/drivers/media/dvb/frontends/stv0900_priv.h
+++ b/drivers/media/dvb/frontends/stv0900_priv.h
@@ -271,6 +271,7 @@ struct stv0900_init_params{
 
 	/* IQ from the tuner2 to the demod */
 	enum stv0900_iq_inversion	tun2_iq_inversion;
+	struct stv0900_reg		*ts_config;
 };
 
 struct stv0900_search_params {
@@ -363,6 +364,7 @@ struct stv0900_internal{
 	u8			i2c_addr;
 	u8			clkmode;/* 0 for CLKI, 2 for XTALI */
 	u8			chip_id;
+	struct stv0900_reg	*ts_config;
 	enum fe_stv0900_error	errs;
 	int dmds_used;
 };
diff --git a/drivers/media/video/cx23885/cx23885-dvb.c b/drivers/media/video/cx23885/cx23885-dvb.c
index 236eea0c41e5..48de57b61f63 100644
--- a/drivers/media/video/cx23885/cx23885-dvb.c
+++ b/drivers/media/video/cx23885/cx23885-dvb.c
@@ -45,6 +45,7 @@
 #include "dibx000_common.h"
 #include "zl10353.h"
 #include "stv0900.h"
+#include "stv0900_reg.h"
 #include "stv6110.h"
 #include "lnbh24.h"
 #include "cx24116.h"
@@ -370,13 +371,25 @@ static struct zl10353_config dvico_fusionhdtv_xc3028 = {
 	.disable_i2c_gate_ctrl = 1,
 };
 
+static struct stv0900_reg stv0900_ts_regs[] = {
+	{ R0900_TSGENERAL, 0x00 },
+	{ R0900_P1_TSSPEED, 0x40 },
+	{ R0900_P2_TSSPEED, 0x40 },
+	{ R0900_P1_TSCFGM, 0xc0 },
+	{ R0900_P2_TSCFGM, 0xc0 },
+	{ R0900_P1_TSCFGH, 0xe0 },
+	{ R0900_P2_TSCFGH, 0xe0 },
+	{ R0900_P1_TSCFGL, 0x20 },
+	{ R0900_P2_TSCFGL, 0x20 },
+	{ 0xffff, 0xff }, /* terminate */
+};
+
 static struct stv0900_config netup_stv0900_config = {
 	.demod_address = 0x68,
 	.xtal = 27000000,
 	.clkmode = 3,/* 0-CLKI, 2-XTALI, else AUTO */
 	.diseqc_mode = 2,/* 2/3 PWM */
-	.path1_mode = 2,/*Serial continues clock */
-	.path2_mode = 2,/*Serial continues clock */
+	.ts_config_regs = stv0900_ts_regs,
 	.tun1_maddress = 0,/* 0x60 */
 	.tun2_maddress = 3,/* 0x63 */
 	.tun1_adc = 1,/* 1 Vpp */
-- 
cgit v1.2.3-59-g8ed1b


From cdf7bfa8926fb26d5900103ae09eb5f3eddb95cc Mon Sep 17 00:00:00 2001
From: Devin Heitmueller <dheitmueller@kernellabs.com>
Date: Fri, 19 Jun 2009 00:20:28 -0300
Subject: V4L/DVB (12100): em28xx: make sure the analog GPIOs are set if we
 used a card hint

In cases where the board had a default USB ID, we would not indentify the
board until after the call to em28xx_set_mode().  As a result, for those
boards the analog GPIOs were not being set before probing the i2c bus for
devices (the probe would occur with the GPIOs being all high).

Make a call to em28xx_set_mode() so that the GPIOs are set properly before
probing the i2c bus for devices.

This problem was detected with the EVGA inDtube, where the tvp5150 is not
powered on unless GPIO1 is pulled low.

Signed-off-by: Devin Heitmueller <dheitmueller@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/em28xx/em28xx-cards.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/media/video/em28xx/em28xx-cards.c b/drivers/media/video/em28xx/em28xx-cards.c
index 00cc791a9e44..0b6e5c7c3466 100644
--- a/drivers/media/video/em28xx/em28xx-cards.c
+++ b/drivers/media/video/em28xx/em28xx-cards.c
@@ -2101,6 +2101,12 @@ void em28xx_card_setup(struct em28xx *dev)
 	case EM2880_BOARD_MSI_DIGIVOX_AD:
 		if (!em28xx_hint_board(dev))
 			em28xx_set_model(dev);
+
+		/* In cases where we had to use a board hint, the call to
+		   em28xx_set_mode() in em28xx_pre_card_setup() was a no-op,
+		   so make the call now so the analog GPIOs are set properly
+		   before probing the i2c bus. */
+		em28xx_set_mode(dev, EM28XX_ANALOG_MODE);
 		break;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 19859229d7d98bc2d582ff45045dd7f73d649383 Mon Sep 17 00:00:00 2001
From: Devin Heitmueller <dheitmueller@kernellabs.com>
Date: Fri, 19 Jun 2009 00:33:54 -0300
Subject: V4L/DVB (12101): em28xx: add support for EVGA inDtube

Add support for the EVGA inDtube.  Both ATSC and analog side validated as
fully functional.

Thanks to Jake Crimmins from EVGA for providing the correct GPIO info.
Thanks to Alan Hagge for doing all the device testing.
Thanks to Greg Williamson for providing hardware for testing.

Cc: Jake Crimmins <jcrimmins@evga.com>
Cc: Alan Hagge <ahagge@gmail.com>
Cc: Greg Williamson <cheeseboy16@gmail.com>
Signed-off-by: Devin Heitmueller <dheitmueller@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 Documentation/video4linux/CARDLIST.em28xx |  1 +
 drivers/media/video/em28xx/em28xx-cards.c | 48 +++++++++++++++++++++++++++++++
 drivers/media/video/em28xx/em28xx-dvb.c   |  1 +
 drivers/media/video/em28xx/em28xx.h       |  1 +
 4 files changed, 51 insertions(+)

diff --git a/Documentation/video4linux/CARDLIST.em28xx b/Documentation/video4linux/CARDLIST.em28xx
index a98a688c11b8..873630e7e53e 100644
--- a/Documentation/video4linux/CARDLIST.em28xx
+++ b/Documentation/video4linux/CARDLIST.em28xx
@@ -65,3 +65,4 @@
  67 -> Terratec Grabby                          (em2860)        [0ccd:0096]
  68 -> Terratec AV350                           (em2860)        [0ccd:0084]
  69 -> KWorld ATSC 315U HDTV TV Box             (em2882)        [eb1a:a313]
+ 70 -> Evga inDtube                             (em2882)
diff --git a/drivers/media/video/em28xx/em28xx-cards.c b/drivers/media/video/em28xx/em28xx-cards.c
index 0b6e5c7c3466..972c4addc406 100644
--- a/drivers/media/video/em28xx/em28xx-cards.c
+++ b/drivers/media/video/em28xx/em28xx-cards.c
@@ -139,6 +139,24 @@ static struct em28xx_reg_seq kworld_330u_digital[] = {
 	{ -1,			-1,	-1,		-1},
 };
 
+/* Evga inDtube
+   GPIO0 - Enable digital power (s5h1409) - low to enable
+   GPIO1 - Enable analog power (tvp5150/emp202) - low to enable
+   GPIO4 - xc3028 reset
+   GOP3  - s5h1409 reset
+ */
+static struct em28xx_reg_seq evga_indtube_analog[] = {
+	{EM28XX_R08_GPIO,	0x79,   0xff,		60},
+	{	-1,		-1,	-1,		-1},
+};
+
+static struct em28xx_reg_seq evga_indtube_digital[] = {
+	{EM28XX_R08_GPIO,	0x7a,	0xff,		 1},
+	{EM2880_R04_GPO,	0x04,	0xff,		10},
+	{EM2880_R04_GPO,	0x0c,	0xff,		 1},
+	{ -1,			-1,	-1,		-1},
+};
+
 /* Callback for the most boards */
 static struct em28xx_reg_seq default_tuner_gpio[] = {
 	{EM28XX_R08_GPIO,	EM_GPIO_4,	EM_GPIO_4,	10},
@@ -1449,6 +1467,31 @@ struct em28xx_board em28xx_boards[] = {
 			.gpio     = terratec_av350_unmute_gpio,
 		} },
 	},
+	[EM2882_BOARD_EVGA_INDTUBE] = {
+		.name         = "Evga inDtube",
+		.tuner_type   = TUNER_XC2028,
+		.tuner_gpio   = default_tuner_gpio,
+		.decoder      = EM28XX_TVP5150,
+		.mts_firmware = 1,
+		.has_dvb      = 1,
+		.dvb_gpio     = evga_indtube_digital,
+		.input        = { {
+			.type     = EM28XX_VMUX_TELEVISION,
+			.vmux     = TVP5150_COMPOSITE0,
+			.amux     = EM28XX_AMUX_VIDEO,
+			.gpio     = evga_indtube_analog,
+		}, {
+			.type     = EM28XX_VMUX_COMPOSITE1,
+			.vmux     = TVP5150_COMPOSITE1,
+			.amux     = EM28XX_AMUX_LINE_IN,
+			.gpio     = evga_indtube_analog,
+		}, {
+			.type     = EM28XX_VMUX_SVIDEO,
+			.vmux     = TVP5150_SVIDEO,
+			.amux     = EM28XX_AMUX_LINE_IN,
+			.gpio     = evga_indtube_analog,
+		} },
+	},
 };
 const unsigned int em28xx_bcount = ARRAY_SIZE(em28xx_boards);
 
@@ -1571,6 +1614,7 @@ static struct em28xx_hash_table em28xx_eeprom_hash[] = {
 	{0x72cc5a8b, EM2820_BOARD_PROLINK_PLAYTV_BOX4_USB2, TUNER_YMEC_TVF_5533MF},
 	{0x966a0441, EM2880_BOARD_KWORLD_DVB_310U, TUNER_XC2028},
 	{0x9567eb1a, EM2880_BOARD_EMPIRE_DUAL_TV, TUNER_XC2028},
+	{0xcee44a99, EM2882_BOARD_EVGA_INDTUBE, TUNER_XC2028},
 };
 
 /* I2C devicelist hash table for devices with generic USB IDs */
@@ -1834,6 +1878,10 @@ static void em28xx_setup_xc3028(struct em28xx *dev, struct xc2028_ctrl *ctl)
 		ctl->demod = XC3028_FE_CHINA;
 		ctl->fname = XC2028_DEFAULT_FIRMWARE;
 		break;
+	case EM2882_BOARD_EVGA_INDTUBE:
+		ctl->demod = XC3028_FE_CHINA;
+		ctl->fname = XC3028L_DEFAULT_FIRMWARE;
+		break;
 	default:
 		ctl->demod = XC3028_FE_OREN538;
 	}
diff --git a/drivers/media/video/em28xx/em28xx-dvb.c b/drivers/media/video/em28xx/em28xx-dvb.c
index 563dd2b1c8e9..e7b47c8da8f3 100644
--- a/drivers/media/video/em28xx/em28xx-dvb.c
+++ b/drivers/media/video/em28xx/em28xx-dvb.c
@@ -445,6 +445,7 @@ static int dvb_init(struct em28xx *dev)
 		}
 		break;
 	case EM2883_BOARD_KWORLD_HYBRID_330U:
+	case EM2882_BOARD_EVGA_INDTUBE:
 		dvb->frontend = dvb_attach(s5h1409_attach,
 					   &em28xx_s5h1409_with_xc3028,
 					   &dev->i2c_adap);
diff --git a/drivers/media/video/em28xx/em28xx.h b/drivers/media/video/em28xx/em28xx.h
index 8bf81be1da61..813ce45c2f99 100644
--- a/drivers/media/video/em28xx/em28xx.h
+++ b/drivers/media/video/em28xx/em28xx.h
@@ -106,6 +106,7 @@
 #define EM2860_BOARD_TERRATEC_GRABBY		  67
 #define EM2860_BOARD_TERRATEC_AV350		  68
 #define EM2882_BOARD_KWORLD_ATSC_315U		  69
+#define EM2882_BOARD_EVGA_INDTUBE		  70
 
 /* Limits minimum and default number of buffers */
 #define EM28XX_MIN_BUF 4
-- 
cgit v1.2.3-59-g8ed1b


From a4c473033b6a100773a4fd8b7ba1e45baeb1e692 Mon Sep 17 00:00:00 2001
From: Devin Heitmueller <dheitmueller@kernellabs.com>
Date: Sat, 20 Jun 2009 21:34:42 -0300
Subject: V4L/DVB (12102): em28xx: add Remote control support for EVGA inDtube

Add an IR profile for the EVGA inDtube remote control (which is an NEC type
remote)

Signed-off-by: Devin Heitmueller <dheitmueller@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/common/ir-keymaps.c         | 23 +++++++++++++++++++++++
 drivers/media/video/em28xx/em28xx-cards.c |  2 ++
 include/media/ir-common.h                 |  2 ++
 3 files changed, 27 insertions(+)

diff --git a/drivers/media/common/ir-keymaps.c b/drivers/media/common/ir-keymaps.c
index 3fe158ac7bbf..4216328552f6 100644
--- a/drivers/media/common/ir-keymaps.c
+++ b/drivers/media/common/ir-keymaps.c
@@ -2750,3 +2750,26 @@ IR_KEYTAB_TYPE ir_codes_dm1105_nec[IR_KEYTAB_SIZE] = {
 	[0x1b] = KEY_B,		/*recall*/
 };
 EXPORT_SYMBOL_GPL(ir_codes_dm1105_nec);
+
+/* EVGA inDtube
+   Devin Heitmueller <devin.heitmueller@gmail.com>
+ */
+IR_KEYTAB_TYPE ir_codes_evga_indtube[IR_KEYTAB_SIZE] = {
+	[0x12] = KEY_POWER,
+	[0x02] = KEY_MODE,	/* TV */
+	[0x14] = KEY_MUTE,
+	[0x1a] = KEY_CHANNELUP,
+	[0x16] = KEY_TV2,	/* PIP */
+	[0x1d] = KEY_VOLUMEUP,
+	[0x05] = KEY_CHANNELDOWN,
+	[0x0f] = KEY_PLAYPAUSE,
+	[0x19] = KEY_VOLUMEDOWN,
+	[0x1c] = KEY_REWIND,
+	[0x0d] = KEY_RECORD,
+	[0x18] = KEY_FORWARD,
+	[0x1e] = KEY_PREVIOUS,
+	[0x1b] = KEY_STOP,
+	[0x1f] = KEY_NEXT,
+	[0x13] = KEY_CAMERA,
+};
+EXPORT_SYMBOL_GPL(ir_codes_evga_indtube);
diff --git a/drivers/media/video/em28xx/em28xx-cards.c b/drivers/media/video/em28xx/em28xx-cards.c
index 972c4addc406..c43fdb9bc888 100644
--- a/drivers/media/video/em28xx/em28xx-cards.c
+++ b/drivers/media/video/em28xx/em28xx-cards.c
@@ -1472,9 +1472,11 @@ struct em28xx_board em28xx_boards[] = {
 		.tuner_type   = TUNER_XC2028,
 		.tuner_gpio   = default_tuner_gpio,
 		.decoder      = EM28XX_TVP5150,
+		.xclk         = EM28XX_XCLK_FREQUENCY_12MHZ, /* NEC IR */
 		.mts_firmware = 1,
 		.has_dvb      = 1,
 		.dvb_gpio     = evga_indtube_digital,
+		.ir_codes     = ir_codes_evga_indtube,
 		.input        = { {
 			.type     = EM28XX_VMUX_TELEVISION,
 			.vmux     = TVP5150_COMPOSITE0,
diff --git a/include/media/ir-common.h b/include/media/ir-common.h
index 7b5b91f60425..9dcb632f6083 100644
--- a/include/media/ir-common.h
+++ b/include/media/ir-common.h
@@ -162,6 +162,8 @@ extern IR_KEYTAB_TYPE ir_codes_ati_tv_wonder_hd_600[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_kworld_plus_tv_analog[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_kaiomy[IR_KEYTAB_SIZE];
 extern IR_KEYTAB_TYPE ir_codes_dm1105_nec[IR_KEYTAB_SIZE];
+extern IR_KEYTAB_TYPE ir_codes_evga_indtube[IR_KEYTAB_SIZE];
+
 #endif
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From c6711c3e6d4976716633047c0f6bbd953d6831fb Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Sun, 14 Jun 2009 05:20:21 -0300
Subject: V4L/DVB (12104): ivtv/cx18: fix regression: class controls are no
 longer seen

A previous change (v4l2-common: remove v4l2_ctrl_query_fill_std) broke
the handling of class controls in VIDIOC_QUERYCTRL. The MPEG class control
was broken for all drivers that use the cx2341x module and the USER class
control was broken for ivtv and cx18.

This change adds back proper class control support.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/cx18/cx18-controls.c | 2 ++
 drivers/media/video/cx2341x.c            | 2 ++
 drivers/media/video/ivtv/ivtv-controls.c | 2 ++
 3 files changed, 6 insertions(+)

diff --git a/drivers/media/video/cx18/cx18-controls.c b/drivers/media/video/cx18/cx18-controls.c
index 8e35c3aed544..5136df198338 100644
--- a/drivers/media/video/cx18/cx18-controls.c
+++ b/drivers/media/video/cx18/cx18-controls.c
@@ -61,6 +61,8 @@ int cx18_queryctrl(struct file *file, void *fh, struct v4l2_queryctrl *qctrl)
 
 	switch (qctrl->id) {
 	/* Standard V4L2 controls */
+	case V4L2_CID_USER_CLASS:
+		return v4l2_ctrl_query_fill(qctrl, 0, 0, 0, 0);
 	case V4L2_CID_BRIGHTNESS:
 	case V4L2_CID_HUE:
 	case V4L2_CID_SATURATION:
diff --git a/drivers/media/video/cx2341x.c b/drivers/media/video/cx2341x.c
index 8ded52946334..4c8e95853fa3 100644
--- a/drivers/media/video/cx2341x.c
+++ b/drivers/media/video/cx2341x.c
@@ -500,6 +500,8 @@ int cx2341x_ctrl_query(const struct cx2341x_mpeg_params *params,
 	int err;
 
 	switch (qctrl->id) {
+	case V4L2_CID_MPEG_CLASS:
+		return v4l2_ctrl_query_fill(qctrl, 0, 0, 0, 0);
 	case V4L2_CID_MPEG_STREAM_TYPE:
 		return v4l2_ctrl_query_fill(qctrl,
 				V4L2_MPEG_STREAM_TYPE_MPEG2_PS,
diff --git a/drivers/media/video/ivtv/ivtv-controls.c b/drivers/media/video/ivtv/ivtv-controls.c
index 84995bcf4a75..a3b77ed3f089 100644
--- a/drivers/media/video/ivtv/ivtv-controls.c
+++ b/drivers/media/video/ivtv/ivtv-controls.c
@@ -60,6 +60,8 @@ int ivtv_queryctrl(struct file *file, void *fh, struct v4l2_queryctrl *qctrl)
 
 	switch (qctrl->id) {
 	/* Standard V4L2 controls */
+	case V4L2_CID_USER_CLASS:
+		return v4l2_ctrl_query_fill(qctrl, 0, 0, 0, 0);
 	case V4L2_CID_BRIGHTNESS:
 	case V4L2_CID_HUE:
 	case V4L2_CID_SATURATION:
-- 
cgit v1.2.3-59-g8ed1b


From be5daa9bd220d384c7010aee6d3886279a61a183 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Sun, 14 Jun 2009 06:47:35 -0300
Subject: V4L/DVB (12107): smscoreapi: fix compile warning

gcc 4.3.1 generates this warning:

v4l/smscoreapi.c: In function 'smscore_gpio_configure':
v4l/smscoreapi.c:1481: warning: 'GroupNum' may be used uninitialized in this function
v4l/smscoreapi.c:1480: warning: 'TranslatedPinNum' may be used uninitialized in this function

While in practice this will not happen, it is something that the compiler
can't determine. Initializing these two local variables to 0 suppresses
this warning.

Cc: Udi Atar <udi.linuxtv@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/dvb/siano/smscoreapi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/media/dvb/siano/smscoreapi.c b/drivers/media/dvb/siano/smscoreapi.c
index 32be382f0e97..a246903c3341 100644
--- a/drivers/media/dvb/siano/smscoreapi.c
+++ b/drivers/media/dvb/siano/smscoreapi.c
@@ -1422,8 +1422,8 @@ int smscore_gpio_configure(struct smscore_device_t *coredev, u8 PinNum,
 		struct smscore_gpio_config *pGpioConfig) {
 
 	u32 totalLen;
-	u32 TranslatedPinNum;
-	u32 GroupNum;
+	u32 TranslatedPinNum = 0;
+	u32 GroupNum = 0;
 	u32 ElectricChar;
 	u32 groupCfg;
 	void *buffer;
-- 
cgit v1.2.3-59-g8ed1b


From 719cd4ab9695059e00f5248d1dceb534381fccb3 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Sun, 14 Jun 2009 07:12:11 -0300
Subject: V4L/DVB (12108): v4l2-i2c-drv.h: add comment describing when not to
 use this header.

Make it very clear that this header should not be used for i2c drivers that
do not need to be compiled for pre-2.6.26 kernels.

As soon as the minimum supported kernel in the v4l-dvb repository becomes
2.6.26 or up, then this header should be removed entirely.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 include/media/v4l2-i2c-drv.h | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/include/media/v4l2-i2c-drv.h b/include/media/v4l2-i2c-drv.h
index 10a2882c3cbf..74bf741d1a9b 100644
--- a/include/media/v4l2-i2c-drv.h
+++ b/include/media/v4l2-i2c-drv.h
@@ -22,7 +22,7 @@
  */
 
 /* NOTE: the full version of this header is in the v4l-dvb repository
- * and allows v4l i2c drivers to be compiled on older kernels as well.
+ * and allows v4l i2c drivers to be compiled on pre-2.6.26 kernels.
  * The version of this header as it appears in the kernel is a stripped
  * version (without all the backwards compatibility stuff) and so it
  * looks a bit odd.
@@ -30,6 +30,9 @@
  * If you look at the full version then you will understand the reason
  * for introducing this header since you really don't want to have all
  * the tricky backwards compatibility code in each and every i2c driver.
+ *
+ * If the i2c driver will never be compiled for pre-2.6.26 kernels, then
+ * DO NOT USE this header! Just write it as a regular i2c driver.
  */
 
 #ifndef __V4L2_I2C_DRV_H__
-- 
cgit v1.2.3-59-g8ed1b


From 5543e2b4c495714e5f1b2dbbe250df15239af1cd Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Sat, 20 Jun 2009 06:29:12 -0300
Subject: V4L/DVB (12109): radio-tea5764: fix incorrect rxsubchans value

rxsubchans was only set when stereo was detected, otherwise it was
left to 0 instead of setting it to mono.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/radio/radio-tea5764.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/media/radio/radio-tea5764.c b/drivers/media/radio/radio-tea5764.c
index 393623818ade..3cd76dddb6aa 100644
--- a/drivers/media/radio/radio-tea5764.c
+++ b/drivers/media/radio/radio-tea5764.c
@@ -322,7 +322,9 @@ static int vidioc_g_tuner(struct file *file, void *priv,
 	v->rangehigh  = FREQ_MAX * FREQ_MUL;
 	v->capability = V4L2_TUNER_CAP_LOW | V4L2_TUNER_CAP_STEREO;
 	if (r->tunchk & TEA5764_TUNCHK_STEREO)
-			v->rxsubchans = V4L2_TUNER_SUB_STEREO;
+		v->rxsubchans = V4L2_TUNER_SUB_STEREO;
+	else
+		v->rxsubchans = V4L2_TUNER_SUB_MONO;
 	v->audmode = tea5764_get_audout_mode(radio);
 	v->signal = TEA5764_TUNCHK_LEVEL(r->tunchk) * 0xffff / 0xf;
 	v->afc = TEA5764_TUNCHK_IFCNT(r->tunchk);
-- 
cgit v1.2.3-59-g8ed1b


From 54bb501c069bbe34cf8becf0a9985fc6873d6b21 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Sat, 20 Jun 2009 09:18:34 -0300
Subject: V4L/DVB (12111): tcm825x: remove incorrect __exit_p wrapper

tcm825x_remove is not necessarily called on module exit, it can also be
called when the i2c_adapter is removed. While the i2c adapter might never
be removed on an embedded system, in practice this sensor driver can also
be used in e.g. a USB webcam where this is a perfectly acceptable thing
to do.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/tcm825x.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/media/video/tcm825x.c b/drivers/media/video/tcm825x.c
index b30c49248217..b90e9da3167d 100644
--- a/drivers/media/video/tcm825x.c
+++ b/drivers/media/video/tcm825x.c
@@ -878,7 +878,7 @@ static int tcm825x_probe(struct i2c_client *client,
 	return rval;
 }
 
-static int __exit tcm825x_remove(struct i2c_client *client)
+static int tcm825x_remove(struct i2c_client *client)
 {
 	struct tcm825x_sensor *sensor = i2c_get_clientdata(client);
 
@@ -902,7 +902,7 @@ static struct i2c_driver tcm825x_i2c_driver = {
 		.name = TCM825X_NAME,
 	},
 	.probe	= tcm825x_probe,
-	.remove	= __exit_p(tcm825x_remove),
+	.remove	= tcm825x_remove,
 	.id_table = tcm825x_id,
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From aad40d3d0cd9b679e83f6a902ad1e2b8f7b4c9bb Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Sat, 20 Jun 2009 09:21:37 -0300
Subject: V4L/DVB (12112): cx231xx: fix uninitialized variable.

The variable 'rc' could be used uninitialized in the cx231xx_capture_start
function. Sri informed me that it should be initialized to -1.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/cx231xx/cx231xx-avcore.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/media/video/cx231xx/cx231xx-avcore.c b/drivers/media/video/cx231xx/cx231xx-avcore.c
index 6a9464079b4c..96f07d1473fb 100644
--- a/drivers/media/video/cx231xx/cx231xx-avcore.c
+++ b/drivers/media/video/cx231xx/cx231xx-avcore.c
@@ -2055,7 +2055,7 @@ int cx231xx_initialize_stream_xfer(struct cx231xx *dev, u32 media_type)
 
 int cx231xx_capture_start(struct cx231xx *dev, int start, u8 media_type)
 {
-	int rc;
+	int rc = -1;
 	u32 ep_mask = -1;
 	struct pcb_config *pcb_config;
 
-- 
cgit v1.2.3-59-g8ed1b


From 9d68fc0ad40b852470026ee58a07e1d662571d04 Mon Sep 17 00:00:00 2001
From: Michael Krufky <mkrufky@kernellabs.com>
Date: Fri, 19 Jun 2009 16:21:37 -0300
Subject: V4L/DVB (12115): tda10048: add missing entry to pll_tab for 3.8 MHz
 IF

Thanks for Terry Wu for pointing out the missing entry.

Cc: Terry Wu <terrywu2009@gmail.com>
Signed-off-by: Michael Krufky <mkrufky@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/dvb/frontends/tda10048.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/media/dvb/frontends/tda10048.c b/drivers/media/dvb/frontends/tda10048.c
index 4302c563a6b8..cc8862ce4aae 100644
--- a/drivers/media/dvb/frontends/tda10048.c
+++ b/drivers/media/dvb/frontends/tda10048.c
@@ -210,6 +210,7 @@ static struct pll_tab {
 	{ TDA10048_CLK_4000,  TDA10048_IF_36130, 10, 0, 0 },
 	{ TDA10048_CLK_16000, TDA10048_IF_3300,  10, 3, 0 },
 	{ TDA10048_CLK_16000, TDA10048_IF_3500,  10, 3, 0 },
+	{ TDA10048_CLK_16000, TDA10048_IF_3800,  10, 3, 0 },
 	{ TDA10048_CLK_16000, TDA10048_IF_4000,  10, 3, 0 },
 	{ TDA10048_CLK_16000, TDA10048_IF_4300,  10, 3, 0 },
 	{ TDA10048_CLK_16000, TDA10048_IF_36130, 10, 3, 0 },
-- 
cgit v1.2.3-59-g8ed1b


From b34cdc36c4aad10cf4eaadacf067835d6a622f1b Mon Sep 17 00:00:00 2001
From: Michael Krufky <mkrufky@kernellabs.com>
Date: Thu, 21 May 2009 12:49:28 -0300
Subject: V4L/DVB (12116): cx23885: ensure correct IF freq is used on HVR1200 &
 HVR1700

Ensure that we're programming the tda18271 tuner with the correct
IF frequencies to match the programming of the TDA10048 DVB-T demod
for the HVR1200 and HVR1700 products.

Signed-off-by: Michael Krufky <mkrufky@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/cx23885/cx23885-dvb.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/media/video/cx23885/cx23885-dvb.c b/drivers/media/video/cx23885/cx23885-dvb.c
index 48de57b61f63..48a975134ac5 100644
--- a/drivers/media/video/cx23885/cx23885-dvb.c
+++ b/drivers/media/video/cx23885/cx23885-dvb.c
@@ -243,12 +243,22 @@ static struct tda18271_std_map hauppauge_tda18271_std_map = {
 		      .if_lvl = 6, .rfagc_top = 0x37 },
 };
 
+static struct tda18271_std_map hauppauge_hvr1200_tda18271_std_map = {
+	.dvbt_6   = { .if_freq = 3300, .agc_mode = 3, .std = 4,
+		      .if_lvl = 1, .rfagc_top = 0x37, },
+	.dvbt_7   = { .if_freq = 3800, .agc_mode = 3, .std = 5,
+		      .if_lvl = 1, .rfagc_top = 0x37, },
+	.dvbt_8   = { .if_freq = 4300, .agc_mode = 3, .std = 6,
+		      .if_lvl = 1, .rfagc_top = 0x37, },
+};
+
 static struct tda18271_config hauppauge_tda18271_config = {
 	.std_map = &hauppauge_tda18271_std_map,
 	.gate    = TDA18271_GATE_ANALOG,
 };
 
 static struct tda18271_config hauppauge_hvr1200_tuner_config = {
+	.std_map = &hauppauge_hvr1200_tda18271_std_map,
 	.gate    = TDA18271_GATE_ANALOG,
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From e17d787c513f41f59969247062561fff6340f211 Mon Sep 17 00:00:00 2001
From: Mike Isely <isely@pobox.com>
Date: Sat, 20 Jun 2009 14:45:52 -0300
Subject: V4L/DVB (12118): pvrusb2: Fix hardware scaling when used with cx25840

The cx25840 module requires that its VBI initialization entry point be
called in order for hardware-scaled video capture to work properly -
even if we don't care about VBI.  Making this behavior even more
subtle is that if the capture resolution is set to 720x480 - which is
the default that the pvrusb2 driver sets up - then the cx25840
bypasses the hardware scaler.  Therefore this problem does not
manifest itself until some other resolution, e.g. 640x480, is tried.
MythTV typically defaults to 640x480 or 480x480, which means that
things break whenever the driver is used with MythTV.

This all has been known for a while (since at least Nov 2006), but
recent changes in the pvrusb2 driver (specifically in regards to
sub-device support) caused this to break again.  VBI initialization
must happen *after* the chip's firmware is loaded, not before.  With
this fix, 24xxx devices work correctly again.

A related fix that is part of this changeset is that now we
re-initialize VBI any time after we issue a reset to the cx25840
driver.  Issuing a chip reset erases the state that the VBI setup
previously did.  Until the HVR-1950 came along this subtlety went
unnoticed, because the pvrusb2 driver previously never issued such a
reset.  But with the HVR-1950 we have to do that reset in order to
correctly transition from digital back to analog mode - and since the
HVR-1950 always starts in digital mode (required for the DVB side to
initialize correctly) then this device has never had a chance to work
correctly in analog mode!  Analog capture on the HVR-1950 has been
broken this *ENTIRE* time.  I had missed it until now because I've
usually been testing at the default 720x480 resolution which does not
require scaling...  What fun.  By re-initializing VBI after a cx25840
chip reset, correct behavior is restored.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/pvrusb2/pvrusb2-hdw.c | 55 +++++++++++++++++--------------
 1 file changed, 31 insertions(+), 24 deletions(-)

diff --git a/drivers/media/video/pvrusb2/pvrusb2-hdw.c b/drivers/media/video/pvrusb2/pvrusb2-hdw.c
index 0c745b142fb7..e97f8e0e466c 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-hdw.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-hdw.c
@@ -1987,6 +1987,34 @@ static unsigned int pvr2_copy_i2c_addr_list(
 }
 
 
+static void pvr2_hdw_cx25840_vbi_hack(struct pvr2_hdw *hdw)
+{
+	/*
+	  Mike Isely <isely@pobox.com> 19-Nov-2006 - This bit of nuttiness
+	  for cx25840 causes that module to correctly set up its video
+	  scaling.  This is really a problem in the cx25840 module itself,
+	  but we work around it here.  The problem has not been seen in
+	  ivtv because there VBI is supported and set up.  We don't do VBI
+	  here (at least not yet) and thus we never attempted to even set
+	  it up.
+	*/
+	struct v4l2_format fmt;
+	if (hdw->decoder_client_id != PVR2_CLIENT_ID_CX25840) {
+		/* We're not using a cx25840 so don't enable the hack */
+		return;
+	}
+
+	pvr2_trace(PVR2_TRACE_INIT,
+		   "Module ID %u:"
+		   " Executing cx25840 VBI hack",
+		   hdw->decoder_client_id);
+	memset(&fmt, 0, sizeof(fmt));
+	fmt.type = V4L2_BUF_TYPE_SLICED_VBI_CAPTURE;
+	v4l2_device_call_all(&hdw->v4l2_dev, hdw->decoder_client_id,
+			     video, s_fmt, &fmt);
+}
+
+
 static int pvr2_hdw_load_subdev(struct pvr2_hdw *hdw,
 				const struct pvr2_device_client_desc *cd)
 {
@@ -2078,30 +2106,6 @@ static int pvr2_hdw_load_subdev(struct pvr2_hdw *hdw,
 	/* client-specific setup... */
 	switch (mid) {
 	case PVR2_CLIENT_ID_CX25840:
-		hdw->decoder_client_id = mid;
-		{
-			/*
-			  Mike Isely <isely@pobox.com> 19-Nov-2006 - This
-			  bit of nuttiness for cx25840 causes that module
-			  to correctly set up its video scaling.  This is
-			  really a problem in the cx25840 module itself,
-			  but we work around it here.  The problem has not
-			  been seen in ivtv because there VBI is supported
-			  and set up.  We don't do VBI here (at least not
-			  yet) and thus we never attempted to even set it
-			  up.
-			*/
-			struct v4l2_format fmt;
-			pvr2_trace(PVR2_TRACE_INIT,
-				   "Module ID %u:"
-				   " Executing cx25840 VBI hack",
-				   mid);
-			memset(&fmt, 0, sizeof(fmt));
-			fmt.type = V4L2_BUF_TYPE_SLICED_VBI_CAPTURE;
-			v4l2_device_call_all(&hdw->v4l2_dev, mid,
-					     video, s_fmt, &fmt);
-		}
-		break;
 	case PVR2_CLIENT_ID_SAA7115:
 		hdw->decoder_client_id = mid;
 		break;
@@ -2202,6 +2206,8 @@ static void pvr2_hdw_setup_low(struct pvr2_hdw *hdw)
 		cptr->info->set_value(cptr,~0,cptr->info->default_value);
 	}
 
+	pvr2_hdw_cx25840_vbi_hack(hdw);
+
 	/* Set up special default values for the television and radio
 	   frequencies here.  It's not really important what these defaults
 	   are, but I set them to something usable in the Chicago area just
@@ -4076,6 +4082,7 @@ int pvr2_hdw_cmd_decoder_reset(struct pvr2_hdw *hdw)
 	if (hdw->decoder_client_id) {
 		v4l2_device_call_all(&hdw->v4l2_dev, hdw->decoder_client_id,
 				     core, reset, 0);
+		pvr2_hdw_cx25840_vbi_hack(hdw);
 		return 0;
 	}
 	pvr2_trace(PVR2_TRACE_INIT,
-- 
cgit v1.2.3-59-g8ed1b


From a6862da2f3c7ce3ec6644958bc8937b630b9e2c1 Mon Sep 17 00:00:00 2001
From: Mike Isely <isely@pobox.com>
Date: Sat, 20 Jun 2009 14:50:14 -0300
Subject: V4L/DVB (12119): pvrusb2: Re-fix hardware scaling on video standard
 change

The cx25840 module's VBI initialization logic uses the current video
standard as part of its internal algorithm.  This therefore means that
we probably need to make sure that the correct video standard has been
set before initializing VBI.  (Normally we would not care about VBI,
but as described in an earlier changeset, VBI must be initialized
correctly on the cx25840 in order for the chip's hardware scaler to
operate correctly.)

It's kind of messy to force the video standard to be set before
initializing VBI (mainly because we can't know what the app really
wants that early in the initialization process).  So this patch does
the next best thing: VBI is re-initialized after any point where the
video standard has been set.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/pvrusb2/pvrusb2-hdw.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/media/video/pvrusb2/pvrusb2-hdw.c b/drivers/media/video/pvrusb2/pvrusb2-hdw.c
index e97f8e0e466c..44da3f9b5b6d 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-hdw.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-hdw.c
@@ -2960,6 +2960,7 @@ static void pvr2_subdev_update(struct pvr2_hdw *hdw)
 			vs = hdw->std_mask_cur;
 			v4l2_device_call_all(&hdw->v4l2_dev, 0,
 					     core, s_std, vs);
+			pvr2_hdw_cx25840_vbi_hack(hdw);
 		}
 		hdw->tuner_signal_stale = !0;
 		hdw->cropcap_stale = !0;
-- 
cgit v1.2.3-59-g8ed1b


From 6f441ed78e28ea02940e58ffa89fbbc734ab6da3 Mon Sep 17 00:00:00 2001
From: Mike Isely <isely@pobox.com>
Date: Sat, 20 Jun 2009 14:51:29 -0300
Subject: V4L/DVB (12120): pvrusb2: Change initial default frequency setting

Change default frequency to be US Broadcast channel 3 - with the
transition to d igital the previous value has now become useless.
This change is PURELY to help with my testing (I need to set some kind
of default so it might as well be some thing usable).

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/pvrusb2/pvrusb2-hdw.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/media/video/pvrusb2/pvrusb2-hdw.c b/drivers/media/video/pvrusb2/pvrusb2-hdw.c
index 44da3f9b5b6d..cbc388729d77 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-hdw.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-hdw.c
@@ -85,8 +85,8 @@ MODULE_PARM_DESC(video_std,"specify initial video standard");
 module_param_array(tolerance,    int, NULL, 0444);
 MODULE_PARM_DESC(tolerance,"specify stream error tolerance");
 
-/* US Broadcast channel 7 (175.25 MHz) */
-static int default_tv_freq    = 175250000L;
+/* US Broadcast channel 3 (61.25 MHz), to help with testing */
+static int default_tv_freq    = 61250000L;
 /* 104.3 MHz, a usable FM station for my area */
 static int default_radio_freq = 104300000L;
 
-- 
cgit v1.2.3-59-g8ed1b


From 81e804c9c2e38431c1c01165d06076776c6fcbd6 Mon Sep 17 00:00:00 2001
From: Mike Isely <isely@pobox.com>
Date: Sat, 20 Jun 2009 14:55:31 -0300
Subject: V4L/DVB (12121): pvrusb2: Improve handling of routing schemes

The pvrusb2 driver has a concept of "routing scheme" which defines
which physical inputs should be connected based on application's
choice of logical input.  The correct "routing scheme" depends on the
specific device since different devices might wire up their muxes

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/pvrusb2/pvrusb2-audio.c       | 14 +++++++------
 drivers/media/video/pvrusb2/pvrusb2-cs53l32a.c    | 14 +++++++------
 drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c | 24 +++++++++++++----------
 drivers/media/video/pvrusb2/pvrusb2-video-v4l.c   | 24 +++++++++++++----------
 4 files changed, 44 insertions(+), 32 deletions(-)

diff --git a/drivers/media/video/pvrusb2/pvrusb2-audio.c b/drivers/media/video/pvrusb2/pvrusb2-audio.c
index 10ef1a2c13ea..416933ca607d 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-audio.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-audio.c
@@ -48,11 +48,13 @@ static const int routing_scheme0[] = {
 						MSP_DSP_IN_SCART),
 };
 
-static const struct routing_scheme routing_schemes[] = {
-	[PVR2_ROUTING_SCHEME_HAUPPAUGE] = {
-		.def = routing_scheme0,
-		.cnt = ARRAY_SIZE(routing_scheme0),
-	},
+static const struct routing_scheme routing_def0 = {
+	.def = routing_scheme0,
+	.cnt = ARRAY_SIZE(routing_scheme0),
+};
+
+static const struct routing_scheme *routing_schemes[] = {
+	[PVR2_ROUTING_SCHEME_HAUPPAUGE] = &routing_def0,
 };
 
 void pvr2_msp3400_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
@@ -65,7 +67,7 @@ void pvr2_msp3400_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
 		pvr2_trace(PVR2_TRACE_CHIPS, "subdev msp3400 v4l2 set_stereo");
 
 		if ((sid < ARRAY_SIZE(routing_schemes)) &&
-		    ((sp = routing_schemes + sid) != NULL) &&
+		    ((sp = routing_schemes[sid]) != NULL) &&
 		    (hdw->input_val >= 0) &&
 		    (hdw->input_val < sp->cnt)) {
 			input = sp->def[hdw->input_val];
diff --git a/drivers/media/video/pvrusb2/pvrusb2-cs53l32a.c b/drivers/media/video/pvrusb2/pvrusb2-cs53l32a.c
index 9023adf3fdcc..41f6e009d5ef 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-cs53l32a.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-cs53l32a.c
@@ -49,11 +49,13 @@ static const int routing_scheme1[] = {
 	[PVR2_CVAL_INPUT_SVIDEO] =  0,
 };
 
-static const struct routing_scheme routing_schemes[] = {
-	[PVR2_ROUTING_SCHEME_ONAIR] = {
-		.def = routing_scheme1,
-		.cnt = ARRAY_SIZE(routing_scheme1),
-	},
+static const struct routing_scheme routing_def1 = {
+	.def = routing_scheme1,
+	.cnt = ARRAY_SIZE(routing_scheme1),
+};
+
+static const struct routing_scheme *routing_schemes[] = {
+	[PVR2_ROUTING_SCHEME_ONAIR] = &routing_def1,
 };
 
 
@@ -66,7 +68,7 @@ void pvr2_cs53l32a_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
 		pvr2_trace(PVR2_TRACE_CHIPS, "subdev v4l2 set_input(%d)",
 			   hdw->input_val);
 		if ((sid < ARRAY_SIZE(routing_schemes)) &&
-		    ((sp = routing_schemes + sid) != NULL) &&
+		    ((sp = routing_schemes[sid]) != NULL) &&
 		    (hdw->input_val >= 0) &&
 		    (hdw->input_val < sp->cnt)) {
 			input = sp->def[hdw->input_val];
diff --git a/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c b/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c
index 05e52358ae49..8710c6218aa8 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c
@@ -68,6 +68,11 @@ static const struct routing_scheme_item routing_scheme0[] = {
 	},
 };
 
+static const struct routing_scheme routing_def0 = {
+	.def = routing_scheme0,
+	.cnt = ARRAY_SIZE(routing_scheme0),
+};
+
 /* Specific to gotview device */
 static const struct routing_scheme_item routing_schemegv[] = {
 	[PVR2_CVAL_INPUT_TV] = {
@@ -90,15 +95,14 @@ static const struct routing_scheme_item routing_schemegv[] = {
 	},
 };
 
-static const struct routing_scheme routing_schemes[] = {
-	[PVR2_ROUTING_SCHEME_HAUPPAUGE] = {
-		.def = routing_scheme0,
-		.cnt = ARRAY_SIZE(routing_scheme0),
-	},
-	[PVR2_ROUTING_SCHEME_GOTVIEW] = {
-		.def = routing_schemegv,
-		.cnt = ARRAY_SIZE(routing_schemegv),
-	},
+static const struct routing_scheme routing_defgv = {
+	.def = routing_schemegv,
+	.cnt = ARRAY_SIZE(routing_schemegv),
+};
+
+static const struct routing_scheme *routing_schemes[] = {
+	[PVR2_ROUTING_SCHEME_HAUPPAUGE] = &routing_def0,
+	[PVR2_ROUTING_SCHEME_GOTVIEW] = &routing_defgv,
 };
 
 void pvr2_cx25840_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
@@ -111,7 +115,7 @@ void pvr2_cx25840_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
 		unsigned int sid = hdw->hdw_desc->signal_routing_scheme;
 
 		if ((sid < ARRAY_SIZE(routing_schemes)) &&
-		    ((sp = routing_schemes + sid) != NULL) &&
+		    ((sp = routing_schemes[sid]) != NULL) &&
 		    (hdw->input_val >= 0) &&
 		    (hdw->input_val < sp->cnt)) {
 			vid_input = sp->def[hdw->input_val].vid;
diff --git a/drivers/media/video/pvrusb2/pvrusb2-video-v4l.c b/drivers/media/video/pvrusb2/pvrusb2-video-v4l.c
index d2fe7c8f2c3a..8c32288eebed 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-video-v4l.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-video-v4l.c
@@ -54,6 +54,11 @@ static const int routing_scheme0[] = {
 	[PVR2_CVAL_INPUT_SVIDEO] =  SAA7115_SVIDEO2,
 };
 
+static const struct routing_scheme routing_def0 = {
+	.def = routing_scheme0,
+	.cnt = ARRAY_SIZE(routing_scheme0),
+};
+
 static const int routing_scheme1[] = {
 	[PVR2_CVAL_INPUT_TV] = SAA7115_COMPOSITE4,
 	[PVR2_CVAL_INPUT_RADIO] = SAA7115_COMPOSITE5,
@@ -61,15 +66,14 @@ static const int routing_scheme1[] = {
 	[PVR2_CVAL_INPUT_SVIDEO] =  SAA7115_SVIDEO2, /* or SVIDEO0, it seems */
 };
 
-static const struct routing_scheme routing_schemes[] = {
-	[PVR2_ROUTING_SCHEME_HAUPPAUGE] = {
-		.def = routing_scheme0,
-		.cnt = ARRAY_SIZE(routing_scheme0),
-	},
-	[PVR2_ROUTING_SCHEME_ONAIR] = {
-		.def = routing_scheme1,
-		.cnt = ARRAY_SIZE(routing_scheme1),
-	},
+static const struct routing_scheme routing_def1 = {
+	.def = routing_scheme1,
+	.cnt = ARRAY_SIZE(routing_scheme1),
+};
+
+static const struct routing_scheme *routing_schemes[] = {
+	[PVR2_ROUTING_SCHEME_HAUPPAUGE] = &routing_def0,
+	[PVR2_ROUTING_SCHEME_ONAIR] = &routing_def1,
 };
 
 void pvr2_saa7115_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
@@ -82,7 +86,7 @@ void pvr2_saa7115_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
 		pvr2_trace(PVR2_TRACE_CHIPS, "subdev v4l2 set_input(%d)",
 			   hdw->input_val);
 		if ((sid < ARRAY_SIZE(routing_schemes)) &&
-		    ((sp = routing_schemes + sid) != NULL) &&
+		    ((sp = routing_schemes[sid]) != NULL) &&
 		    (hdw->input_val >= 0) &&
 		    (hdw->input_val < sp->cnt)) {
 			input = sp->def[hdw->input_val];
-- 
cgit v1.2.3-59-g8ed1b


From 90135c96869fa0ef3182282b2a661b57fcdb7230 Mon Sep 17 00:00:00 2001
From: Mike Isely <isely@pobox.com>
Date: Sat, 20 Jun 2009 14:57:24 -0300
Subject: V4L/DVB (12122): pvrusb2: De-obfuscate code which handles routing
 schemes

This change does not change any outward behavior; it merely chops down
some large if-conditions with embedded assignments into something a
little more maintainable for others (I of course never had a problem
with this...).

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/pvrusb2/pvrusb2-cs53l32a.c    | 12 ++++++------
 drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c | 15 +++++++--------
 drivers/media/video/pvrusb2/pvrusb2-video-v4l.c   | 13 +++++++------
 3 files changed, 20 insertions(+), 20 deletions(-)

diff --git a/drivers/media/video/pvrusb2/pvrusb2-cs53l32a.c b/drivers/media/video/pvrusb2/pvrusb2-cs53l32a.c
index 41f6e009d5ef..68980e19409f 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-cs53l32a.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-cs53l32a.c
@@ -67,12 +67,11 @@ void pvr2_cs53l32a_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
 		u32 input;
 		pvr2_trace(PVR2_TRACE_CHIPS, "subdev v4l2 set_input(%d)",
 			   hdw->input_val);
-		if ((sid < ARRAY_SIZE(routing_schemes)) &&
-		    ((sp = routing_schemes[sid]) != NULL) &&
-		    (hdw->input_val >= 0) &&
-		    (hdw->input_val < sp->cnt)) {
-			input = sp->def[hdw->input_val];
-		} else {
+		sp = (sid < ARRAY_SIZE(routing_schemes)) ?
+			routing_schemes[sid] : NULL;
+		if ((sp == NULL) ||
+		    (hdw->input_val < 0) ||
+		    (hdw->input_val >= sp->cnt)) {
 			pvr2_trace(PVR2_TRACE_ERROR_LEGS,
 				   "*** WARNING *** subdev v4l2 set_input:"
 				   " Invalid routing scheme (%u)"
@@ -80,6 +79,7 @@ void pvr2_cs53l32a_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
 				   sid, hdw->input_val);
 			return;
 		}
+		input = sp->def[hdw->input_val];
 		sd->ops->audio->s_routing(sd, input, 0, 0);
 	}
 }
diff --git a/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c b/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c
index 8710c6218aa8..82c135835753 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-cx2584x-v4l.c
@@ -114,13 +114,11 @@ void pvr2_cx25840_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
 		const struct routing_scheme *sp;
 		unsigned int sid = hdw->hdw_desc->signal_routing_scheme;
 
-		if ((sid < ARRAY_SIZE(routing_schemes)) &&
-		    ((sp = routing_schemes[sid]) != NULL) &&
-		    (hdw->input_val >= 0) &&
-		    (hdw->input_val < sp->cnt)) {
-			vid_input = sp->def[hdw->input_val].vid;
-			aud_input = sp->def[hdw->input_val].aud;
-		} else {
+		sp = (sid < ARRAY_SIZE(routing_schemes)) ?
+			routing_schemes[sid] : NULL;
+		if ((sp == NULL) ||
+		    (hdw->input_val < 0) ||
+		    (hdw->input_val >= sp->cnt)) {
 			pvr2_trace(PVR2_TRACE_ERROR_LEGS,
 				   "*** WARNING *** subdev cx2584x set_input:"
 				   " Invalid routing scheme (%u)"
@@ -128,7 +126,8 @@ void pvr2_cx25840_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
 				   sid, hdw->input_val);
 			return;
 		}
-
+		vid_input = sp->def[hdw->input_val].vid;
+		aud_input = sp->def[hdw->input_val].aud;
 		pvr2_trace(PVR2_TRACE_CHIPS,
 			   "subdev cx2584x set_input vid=0x%x aud=0x%x",
 			   vid_input, aud_input);
diff --git a/drivers/media/video/pvrusb2/pvrusb2-video-v4l.c b/drivers/media/video/pvrusb2/pvrusb2-video-v4l.c
index 8c32288eebed..4c96cf48c796 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-video-v4l.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-video-v4l.c
@@ -85,12 +85,12 @@ void pvr2_saa7115_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
 
 		pvr2_trace(PVR2_TRACE_CHIPS, "subdev v4l2 set_input(%d)",
 			   hdw->input_val);
-		if ((sid < ARRAY_SIZE(routing_schemes)) &&
-		    ((sp = routing_schemes[sid]) != NULL) &&
-		    (hdw->input_val >= 0) &&
-		    (hdw->input_val < sp->cnt)) {
-			input = sp->def[hdw->input_val];
-		} else {
+
+		sp = (sid < ARRAY_SIZE(routing_schemes)) ?
+			routing_schemes[sid] : NULL;
+		if ((sp == NULL) ||
+		    (hdw->input_val < 0) ||
+		    (hdw->input_val >= sp->cnt)) {
 			pvr2_trace(PVR2_TRACE_ERROR_LEGS,
 				   "*** WARNING *** subdev v4l2 set_input:"
 				   " Invalid routing scheme (%u)"
@@ -98,6 +98,7 @@ void pvr2_saa7115_subdev_update(struct pvr2_hdw *hdw, struct v4l2_subdev *sd)
 				   sid, hdw->input_val);
 			return;
 		}
+		input = sp->def[hdw->input_val];
 		sd->ops->video->s_routing(sd, input, 0, 0);
 	}
 }
-- 
cgit v1.2.3-59-g8ed1b


From f0222c7d860f09a61bec5e500539f28db0184b38 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Tue, 9 Jun 2009 17:12:33 -0300
Subject: V4L/DVB (12125): v4l2: add new s_config subdev ops and
 v4l2_i2c_new_subdev_cfg/board calls

Add a new s_config core ops call: this is called with the irq and platform
data to be used to initialize the subdev.

Added new v4l2_i2c_new_subdev_cfg and v4l2_i2c_new_subdev_board calls
that allows you to pass these new arguments.

The existing v4l2_i2c_new_subdev functions were modified to also call
s_config.

In the future the existing v4l2_i2c_new_subdev functions will be replaced
by a single v4l2_i2c_new_subdev function similar to v4l2_i2c_new_subdev_cfg
but without the irq and platform_data arguments.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/v4l2-common.c | 105 ++++++++++++++++++++++++++++++++++++++
 include/media/v4l2-common.h       |  16 ++++++
 include/media/v4l2-subdev.h       |   7 ++-
 3 files changed, 127 insertions(+), 1 deletion(-)

diff --git a/drivers/media/video/v4l2-common.c b/drivers/media/video/v4l2-common.c
index f96475626da7..e7b443c116f0 100644
--- a/drivers/media/video/v4l2-common.c
+++ b/drivers/media/video/v4l2-common.c
@@ -802,6 +802,17 @@ struct v4l2_subdev *v4l2_i2c_new_subdev(struct v4l2_device *v4l2_dev,
 	/* Decrease the module use count to match the first try_module_get. */
 	module_put(client->driver->driver.owner);
 
+	if (sd) {
+		/* We return errors from v4l2_subdev_call only if we have the
+		   callback as the .s_config is not mandatory */
+		int err = v4l2_subdev_call(sd, core, s_config, 0, NULL);
+
+		if (err && err != -ENOIOCTLCMD) {
+			v4l2_device_unregister_subdev(sd);
+			sd = NULL;
+		}
+	}
+
 error:
 	/* If we have a client but no subdev, then something went wrong and
 	   we must unregister the client. */
@@ -852,6 +863,17 @@ struct v4l2_subdev *v4l2_i2c_new_probed_subdev(struct v4l2_device *v4l2_dev,
 	/* Decrease the module use count to match the first try_module_get. */
 	module_put(client->driver->driver.owner);
 
+	if (sd) {
+		/* We return errors from v4l2_subdev_call only if we have the
+		   callback as the .s_config is not mandatory */
+		int err = v4l2_subdev_call(sd, core, s_config, 0, NULL);
+
+		if (err && err != -ENOIOCTLCMD) {
+			v4l2_device_unregister_subdev(sd);
+			sd = NULL;
+		}
+	}
+
 error:
 	/* If we have a client but no subdev, then something went wrong and
 	   we must unregister the client. */
@@ -872,6 +894,89 @@ struct v4l2_subdev *v4l2_i2c_new_probed_subdev_addr(struct v4l2_device *v4l2_dev
 }
 EXPORT_SYMBOL_GPL(v4l2_i2c_new_probed_subdev_addr);
 
+/* Load an i2c sub-device. */
+struct v4l2_subdev *v4l2_i2c_new_subdev_board(struct v4l2_device *v4l2_dev,
+		struct i2c_adapter *adapter, const char *module_name,
+		struct i2c_board_info *info, const unsigned short *probe_addrs)
+{
+	struct v4l2_subdev *sd = NULL;
+	struct i2c_client *client;
+
+	BUG_ON(!v4l2_dev);
+
+	if (module_name)
+		request_module(module_name);
+
+	/* Create the i2c client */
+	if (info->addr == 0 && probe_addrs)
+		client = i2c_new_probed_device(adapter, info, probe_addrs);
+	else
+		client = i2c_new_device(adapter, info);
+
+	/* Note: by loading the module first we are certain that c->driver
+	   will be set if the driver was found. If the module was not loaded
+	   first, then the i2c core tries to delay-load the module for us,
+	   and then c->driver is still NULL until the module is finally
+	   loaded. This delay-load mechanism doesn't work if other drivers
+	   want to use the i2c device, so explicitly loading the module
+	   is the best alternative. */
+	if (client == NULL || client->driver == NULL)
+		goto error;
+
+	/* Lock the module so we can safely get the v4l2_subdev pointer */
+	if (!try_module_get(client->driver->driver.owner))
+		goto error;
+	sd = i2c_get_clientdata(client);
+
+	/* Register with the v4l2_device which increases the module's
+	   use count as well. */
+	if (v4l2_device_register_subdev(v4l2_dev, sd))
+		sd = NULL;
+	/* Decrease the module use count to match the first try_module_get. */
+	module_put(client->driver->driver.owner);
+
+	if (sd) {
+		/* We return errors from v4l2_subdev_call only if we have the
+		   callback as the .s_config is not mandatory */
+		int err = v4l2_subdev_call(sd, core, s_config,
+				info->irq, info->platform_data);
+
+		if (err && err != -ENOIOCTLCMD) {
+			v4l2_device_unregister_subdev(sd);
+			sd = NULL;
+		}
+	}
+
+error:
+	/* If we have a client but no subdev, then something went wrong and
+	   we must unregister the client. */
+	if (client && sd == NULL)
+		i2c_unregister_device(client);
+	return sd;
+}
+EXPORT_SYMBOL_GPL(v4l2_i2c_new_subdev_board);
+
+struct v4l2_subdev *v4l2_i2c_new_subdev_cfg(struct v4l2_device *v4l2_dev,
+		struct i2c_adapter *adapter,
+		const char *module_name, const char *client_type,
+		int irq, void *platform_data,
+		u8 addr, const unsigned short *probe_addrs)
+{
+	struct i2c_board_info info;
+
+	/* Setup the i2c board info with the device type and
+	   the device address. */
+	memset(&info, 0, sizeof(info));
+	strlcpy(info.type, client_type, sizeof(info.type));
+	info.addr = addr;
+	info.irq = irq;
+	info.platform_data = platform_data;
+
+	return v4l2_i2c_new_subdev_board(v4l2_dev, adapter, module_name,
+			&info, probe_addrs);
+}
+EXPORT_SYMBOL_GPL(v4l2_i2c_new_subdev_cfg);
+
 /* Return i2c client address of v4l2_subdev. */
 unsigned short v4l2_i2c_subdev_addr(struct v4l2_subdev *sd)
 {
diff --git a/include/media/v4l2-common.h b/include/media/v4l2-common.h
index c48c24e4d0fa..95f4364322eb 100644
--- a/include/media/v4l2-common.h
+++ b/include/media/v4l2-common.h
@@ -153,6 +153,22 @@ struct v4l2_subdev *v4l2_i2c_new_probed_subdev(struct v4l2_device *v4l2_dev,
 struct v4l2_subdev *v4l2_i2c_new_probed_subdev_addr(struct v4l2_device *v4l2_dev,
 		struct i2c_adapter *adapter,
 		const char *module_name, const char *client_type, u8 addr);
+
+/* Load an i2c module and return an initialized v4l2_subdev struct.
+   Only call request_module if module_name != NULL.
+   The client_type argument is the name of the chip that's on the adapter. */
+struct v4l2_subdev *v4l2_i2c_new_subdev_cfg(struct v4l2_device *v4l2_dev,
+		struct i2c_adapter *adapter,
+		const char *module_name, const char *client_type,
+		int irq, void *platform_data,
+		u8 addr, const unsigned short *probe_addrs);
+
+struct i2c_board_info;
+
+struct v4l2_subdev *v4l2_i2c_new_subdev_board(struct v4l2_device *v4l2_dev,
+		struct i2c_adapter *adapter, const char *module_name,
+		struct i2c_board_info *info, const unsigned short *probe_addrs);
+
 /* Initialize an v4l2_subdev with data from an i2c_client struct */
 void v4l2_i2c_subdev_init(struct v4l2_subdev *sd, struct i2c_client *client,
 		const struct v4l2_subdev_ops *ops);
diff --git a/include/media/v4l2-subdev.h b/include/media/v4l2-subdev.h
index a503e1cee78b..5dcb36785529 100644
--- a/include/media/v4l2-subdev.h
+++ b/include/media/v4l2-subdev.h
@@ -79,7 +79,11 @@ struct v4l2_decode_vbi_line {
    not yet implemented) since ops provide proper type-checking.
  */
 
-/* init: initialize the sensor registors to some sort of reasonable default
+/* s_config: if set, then it is always called by the v4l2_i2c_new_subdev*
+	functions after the v4l2_subdev was registered. It is used to pass
+	platform data to the subdev which can be used during initialization.
+
+   init: initialize the sensor registors to some sort of reasonable default
 	values. Do not use for new drivers and should be removed in existing
 	drivers.
 
@@ -96,6 +100,7 @@ struct v4l2_decode_vbi_line {
 struct v4l2_subdev_core_ops {
 	int (*g_chip_ident)(struct v4l2_subdev *sd, struct v4l2_dbg_chip_ident *chip);
 	int (*log_status)(struct v4l2_subdev *sd);
+	int (*s_config)(struct v4l2_subdev *sd, int irq, void *platform_data);
 	int (*init)(struct v4l2_subdev *sd, u32 val);
 	int (*load_fw)(struct v4l2_subdev *sd);
 	int (*reset)(struct v4l2_subdev *sd, u32 val);
-- 
cgit v1.2.3-59-g8ed1b


From 2c0b19ac3b73199fe7b3fbff884051046554c048 Mon Sep 17 00:00:00 2001
From: Hans Verkuil <hverkuil@xs4all.nl>
Date: Tue, 9 Jun 2009 17:29:29 -0300
Subject: V4L/DVB (12128): v4l2: update framework documentation.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 Documentation/video4linux/v4l2-framework.txt | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/Documentation/video4linux/v4l2-framework.txt b/Documentation/video4linux/v4l2-framework.txt
index d54c1e4c6a9c..ba4706afc5fb 100644
--- a/Documentation/video4linux/v4l2-framework.txt
+++ b/Documentation/video4linux/v4l2-framework.txt
@@ -390,6 +390,30 @@ later date. It differs between i2c drivers and as such can be confusing.
 To see which chip variants are supported you can look in the i2c driver code
 for the i2c_device_id table. This lists all the possibilities.
 
+There are two more helper functions:
+
+v4l2_i2c_new_subdev_cfg: this function adds new irq and platform_data
+arguments and has both 'addr' and 'probed_addrs' arguments: if addr is not
+0 then that will be used (non-probing variant), otherwise the probed_addrs
+are probed.
+
+For example: this will probe for address 0x10:
+
+struct v4l2_subdev *sd = v4l2_i2c_new_subdev_cfg(v4l2_dev, adapter,
+	       "module_foo", "chipid", 0, NULL, 0, I2C_ADDRS(0x10));
+
+v4l2_i2c_new_subdev_board uses an i2c_board_info struct which is passed
+to the i2c driver and replaces the irq, platform_data and addr arguments.
+
+If the subdev supports the s_config core ops, then that op is called with
+the irq and platform_data arguments after the subdev was setup. The older
+v4l2_i2c_new_(probed_)subdev functions will call s_config as well, but with
+irq set to 0 and platform_data set to NULL.
+
+Note that in the next kernel release the functions v4l2_i2c_new_subdev,
+v4l2_i2c_new_probed_subdev and v4l2_i2c_new_probed_subdev_addr will all be
+replaced by a single v4l2_i2c_new_subdev that is identical to
+v4l2_i2c_new_subdev_cfg but without the irq and platform_data arguments.
 
 struct video_device
 -------------------
-- 
cgit v1.2.3-59-g8ed1b


From b0d3159be9a36fd8b7b1cf88b812d951add53d11 Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11901): v4l2: Create helper function for bounding and
 aligning images

Most hardware has limits on minimum and maximum image dimensions and also
requirements about alignment.  For example, image width must be even or a
multiple of four.  Some hardware has requirements that the total image size
(width * height) be a multiple of some power of two.

v4l_bound_align_image() will enforce min and max width and height, power of
two alignment on width and height, and power of two alignment on total
image size.

It uses an efficient algorithm that will try to find the "closest" image
size that meets the requirements.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/v4l2-common.c | 75 ++++++++++++++++++++++++++++++++++++++-
 include/media/v4l2-common.h       | 10 ++++++
 2 files changed, 84 insertions(+), 1 deletion(-)

diff --git a/drivers/media/video/v4l2-common.c b/drivers/media/video/v4l2-common.c
index e7b443c116f0..1ebbe738019f 100644
--- a/drivers/media/video/v4l2-common.c
+++ b/drivers/media/video/v4l2-common.c
@@ -1021,4 +1021,77 @@ const unsigned short *v4l2_i2c_tuner_addrs(enum v4l2_i2c_tuner_type type)
 }
 EXPORT_SYMBOL_GPL(v4l2_i2c_tuner_addrs);
 
-#endif
+/* Clamp x to be between min and max, aligned to a multiple of 2^align.  min
+ * and max don't have to be aligned, but there must be at least one valid
+ * value.  E.g., min=17,max=31,align=4 is not allowed as there are no multiples
+ * of 16 between 17 and 31.  */
+static unsigned int clamp_align(unsigned int x, unsigned int min,
+				unsigned int max, unsigned int align)
+{
+	/* Bits that must be zero to be aligned */
+	unsigned int mask = ~((1 << align) - 1);
+
+	/* Round to nearest aligned value */
+	if (align)
+		x = (x + (1 << (align - 1))) & mask;
+
+	/* Clamp to aligned value of min and max */
+	if (x < min)
+		x = (min + ~mask) & mask;
+	else if (x > max)
+		x = max & mask;
+
+	return x;
+}
+
+/* Bound an image to have a width between wmin and wmax, and height between
+ * hmin and hmax, inclusive.  Additionally, the width will be a multiple of
+ * 2^walign, the height will be a multiple of 2^halign, and the overall size
+ * (width*height) will be a multiple of 2^salign.  The image may be shrunk
+ * or enlarged to fit the alignment constraints.
+ *
+ * The width or height maximum must not be smaller than the corresponding
+ * minimum.  The alignments must not be so high there are no possible image
+ * sizes within the allowed bounds.  wmin and hmin must be at least 1
+ * (don't use 0).  If you don't care about a certain alignment, specify 0,
+ * as 2^0 is 1 and one byte alignment is equivalent to no alignment.  If
+ * you only want to adjust downward, specify a maximum that's the same as
+ * the initial value.
+ */
+void v4l_bound_align_image(u32 *w, unsigned int wmin, unsigned int wmax,
+			   unsigned int walign,
+			   u32 *h, unsigned int hmin, unsigned int hmax,
+			   unsigned int halign, unsigned int salign)
+{
+	*w = clamp_align(*w, wmin, wmax, walign);
+	*h = clamp_align(*h, hmin, hmax, halign);
+
+	/* Usually we don't need to align the size and are done now. */
+	if (!salign)
+		return;
+
+	/* How much alignment do we have? */
+	walign = __ffs(*w);
+	halign = __ffs(*h);
+	/* Enough to satisfy the image alignment? */
+	if (walign + halign < salign) {
+		/* Max walign where there is still a valid width */
+		unsigned int wmaxa = __fls(wmax ^ (wmin - 1));
+		/* Max halign where there is still a valid height */
+		unsigned int hmaxa = __fls(hmax ^ (hmin - 1));
+
+		/* up the smaller alignment until we have enough */
+		do {
+			if (halign >= hmaxa ||
+			    (walign <= halign && walign < wmaxa)) {
+				*w = clamp_align(*w, wmin, wmax, walign + 1);
+				walign = __ffs(*w);
+			} else {
+				*h = clamp_align(*h, hmin, hmax, halign + 1);
+				halign = __ffs(*h);
+			}
+		} while (halign + walign < salign);
+	}
+}
+EXPORT_SYMBOL_GPL(v4l_bound_align_image);
+#endif /* defined(CONFIG_I2C) */
diff --git a/include/media/v4l2-common.h b/include/media/v4l2-common.h
index 95f4364322eb..33a18426ab9b 100644
--- a/include/media/v4l2-common.h
+++ b/include/media/v4l2-common.h
@@ -209,4 +209,14 @@ struct v4l2_routing {
 	u32 output;
 };
 
+/* ------------------------------------------------------------------------- */
+
+/* Miscellaneous helper functions */
+
+void v4l_bound_align_image(unsigned int *w, unsigned int wmin,
+			   unsigned int wmax, unsigned int walign,
+			   unsigned int *h, unsigned int hmin,
+			   unsigned int hmax, unsigned int halign,
+			   unsigned int salign);
+
 #endif /* V4L2_COMMON_H_ */
-- 
cgit v1.2.3-59-g8ed1b


From 4a6b8df2133c1f218a503e0432a9e6cc3d461a30 Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11902): pxa-camera: Use v4l bounding/alignment function

The v4l function has a better algorithm for aligning image size.

For instance the old code would change 159x243 into 156x240 to meet the
alignment requirements.  The new function will use 160x243, which is a lot
closer to what was asked for originally.

Cc: Robert Jarzmik <robert.jarzmik@free.fr>
Cc: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/pxa_camera.c | 34 +++++++---------------------------
 1 file changed, 7 insertions(+), 27 deletions(-)

diff --git a/drivers/media/video/pxa_camera.c b/drivers/media/video/pxa_camera.c
index f60de40fd21f..46e0d8ad880f 100644
--- a/drivers/media/video/pxa_camera.c
+++ b/drivers/media/video/pxa_camera.c
@@ -162,13 +162,6 @@
 			CICR0_PERRM | CICR0_QDM | CICR0_CDM | CICR0_SOFM | \
 			CICR0_EOFM | CICR0_FOM)
 
-/*
- * YUV422P picture size should be a multiple of 16, so the heuristic aligns
- * height, width on 4 byte boundaries to reach the 16 multiple for the size.
- */
-#define YUV422P_X_Y_ALIGN 4
-#define YUV422P_SIZE_ALIGN YUV422P_X_Y_ALIGN * YUV422P_X_Y_ALIGN
-
 /*
  * Structures
  */
@@ -1398,28 +1391,15 @@ static int pxa_camera_try_fmt(struct soc_camera_device *icd,
 		return -EINVAL;
 	}
 
-	/* limit to pxa hardware capabilities */
-	if (pix->height < 32)
-		pix->height = 32;
-	if (pix->height > 2048)
-		pix->height = 2048;
-	if (pix->width < 48)
-		pix->width = 48;
-	if (pix->width > 2048)
-		pix->width = 2048;
-	pix->width &= ~0x01;
-
 	/*
-	 * YUV422P planar format requires images size to be a 16 bytes
-	 * multiple. If not, zeros will be inserted between Y and U planes, and
-	 * U and V planes, and YUV422P standard would be violated.
+	 * Limit to pxa hardware capabilities.  YUV422P planar format requires
+	 * images size to be a multiple of 16 bytes.  If not, zeros will be
+	 * inserted between Y and U planes, and U and V planes, which violates
+	 * the YUV422P standard.
 	 */
-	if (xlate->host_fmt->fourcc == V4L2_PIX_FMT_YUV422P) {
-		if (!IS_ALIGNED(pix->width * pix->height, YUV422P_SIZE_ALIGN))
-			pix->height = ALIGN(pix->height, YUV422P_X_Y_ALIGN);
-		if (!IS_ALIGNED(pix->width * pix->height, YUV422P_SIZE_ALIGN))
-			pix->width = ALIGN(pix->width, YUV422P_X_Y_ALIGN);
-	}
+	v4l_bound_align_image(&pix->width, 48, 2048, 1,
+			      &pix->height, 32, 2048, 0,
+			      xlate->host_fmt->fourcc == V4L2_PIX_FMT_YUV422P ? 4 : 0);
 
 	pix->bytesperline = pix->width *
 		DIV_ROUND_UP(xlate->host_fmt->depth, 8);
-- 
cgit v1.2.3-59-g8ed1b


From bc44fc061ea1f2b7918ec0bb55013b8054c81752 Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11903): sh_mobile_ceu_camera: Use v4l bounding/alignment
 function

The v4l function has a better algorithm for aligning image size.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/sh_mobile_ceu_camera.c | 12 ++----------
 1 file changed, 2 insertions(+), 10 deletions(-)

diff --git a/drivers/media/video/sh_mobile_ceu_camera.c b/drivers/media/video/sh_mobile_ceu_camera.c
index d369e8409ab8..0db88a53d92c 100644
--- a/drivers/media/video/sh_mobile_ceu_camera.c
+++ b/drivers/media/video/sh_mobile_ceu_camera.c
@@ -689,16 +689,8 @@ static int sh_mobile_ceu_try_fmt(struct soc_camera_device *icd,
 
 	/* FIXME: calculate using depth and bus width */
 
-	if (f->fmt.pix.height < 4)
-		f->fmt.pix.height = 4;
-	if (f->fmt.pix.height > 1920)
-		f->fmt.pix.height = 1920;
-	if (f->fmt.pix.width < 2)
-		f->fmt.pix.width = 2;
-	if (f->fmt.pix.width > 2560)
-		f->fmt.pix.width = 2560;
-	f->fmt.pix.width &= ~0x01;
-	f->fmt.pix.height &= ~0x03;
+	v4l_bound_align_image(&f->fmt.pix.width, 2, 2560, 1,
+			      &f->fmt.pix.height, 4, 1920, 2, 0);
 
 	f->fmt.pix.bytesperline = f->fmt.pix.width *
 		DIV_ROUND_UP(xlate->host_fmt->depth, 8);
-- 
cgit v1.2.3-59-g8ed1b


From 728f5b93f48cbfebd8e939bec2be1252fce7dae1 Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11904): zoran: Use v4l bounding/alignment functiob

The v4l function has a better algorithm for aligning image size.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/zoran/zoran_driver.c | 14 ++++----------
 1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/drivers/media/video/zoran/zoran_driver.c b/drivers/media/video/zoran/zoran_driver.c
index 643cccaa1aab..3d7df32a3d87 100644
--- a/drivers/media/video/zoran/zoran_driver.c
+++ b/drivers/media/video/zoran/zoran_driver.c
@@ -2088,16 +2088,10 @@ static int zoran_try_fmt_vid_cap(struct file *file, void *__fh,
 		return -EINVAL;
 	}
 
-	bpp = (zoran_formats[i].depth + 7) / 8;
-	fmt->fmt.pix.width &= ~((bpp == 2) ? 1 : 3);
-	if (fmt->fmt.pix.width > BUZ_MAX_WIDTH)
-		fmt->fmt.pix.width = BUZ_MAX_WIDTH;
-	if (fmt->fmt.pix.width < BUZ_MIN_WIDTH)
-		fmt->fmt.pix.width = BUZ_MIN_WIDTH;
-	if (fmt->fmt.pix.height > BUZ_MAX_HEIGHT)
-		fmt->fmt.pix.height = BUZ_MAX_HEIGHT;
-	if (fmt->fmt.pix.height < BUZ_MIN_HEIGHT)
-		fmt->fmt.pix.height = BUZ_MIN_HEIGHT;
+	bpp = DIV_ROUND_UP(zoran_formats[i].depth, 8);
+	v4l_bound_align_image(
+		&fmt->fmt.pix.width, BUZ_MIN_WIDTH, BUZ_MAX_WIDTH, bpp == 2 ? 1 : 2,
+		&fmt->fmt.pix.height, BUZ_MIN_HEIGHT, BUZ_MAX_HEIGHT, 0, 0);
 	mutex_unlock(&zr->resource_lock);
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 3adbbb8e2a87d58401466c825e9ff191e3b5a7b6 Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11905): vivi: Use v4l bounding/alignment function

The v4l function has a better algorithm for aligning image size.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/vivi.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/drivers/media/video/vivi.c b/drivers/media/video/vivi.c
index fbfefae7886f..cd7266858462 100644
--- a/drivers/media/video/vivi.c
+++ b/drivers/media/video/vivi.c
@@ -883,15 +883,8 @@ static int vidioc_try_fmt_vid_cap(struct file *file, void *priv,
 	maxh  = norm_maxh();
 
 	f->fmt.pix.field = field;
-	if (f->fmt.pix.height < 32)
-		f->fmt.pix.height = 32;
-	if (f->fmt.pix.height > maxh)
-		f->fmt.pix.height = maxh;
-	if (f->fmt.pix.width < 48)
-		f->fmt.pix.width = 48;
-	if (f->fmt.pix.width > maxw)
-		f->fmt.pix.width = maxw;
-	f->fmt.pix.width &= ~0x03;
+	v4l_bound_align_image(&f->fmt.pix.width, 48, maxw, 2,
+			      &f->fmt.pix.height, 32, maxh, 0, 0);
 	f->fmt.pix.bytesperline =
 		(f->fmt.pix.width * fmt->depth) >> 3;
 	f->fmt.pix.sizeimage =
-- 
cgit v1.2.3-59-g8ed1b


From bc52d6eb44de8f19934768d4d10d19fdbdc99950 Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11906): saa7134: Use v4l bounding/alignment function

The v4l function has a better algorithm for aligning image size.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/saa7134/saa7134-video.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/drivers/media/video/saa7134/saa7134-video.c b/drivers/media/video/saa7134/saa7134-video.c
index e305c1674cee..ba87128542e0 100644
--- a/drivers/media/video/saa7134/saa7134-video.c
+++ b/drivers/media/video/saa7134/saa7134-video.c
@@ -1640,15 +1640,8 @@ static int saa7134_try_fmt_vid_cap(struct file *file, void *priv,
 	}
 
 	f->fmt.pix.field = field;
-	if (f->fmt.pix.width  < 48)
-		f->fmt.pix.width  = 48;
-	if (f->fmt.pix.height < 32)
-		f->fmt.pix.height = 32;
-	if (f->fmt.pix.width > maxw)
-		f->fmt.pix.width = maxw;
-	if (f->fmt.pix.height > maxh)
-		f->fmt.pix.height = maxh;
-	f->fmt.pix.width &= ~0x03;
+	v4l_bound_align_image(&f->fmt.pix.width, 48, maxw, 2,
+			      &f->fmt.pix.height, 32, maxh, 0, 0);
 	f->fmt.pix.bytesperline =
 		(f->fmt.pix.width * fmt->depth) >> 3;
 	f->fmt.pix.sizeimage =
-- 
cgit v1.2.3-59-g8ed1b


From 4b89945e590f94e82a6e7f33e21cbd0d83774b9e Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11907): cx88: Use v4l bounding/alignment function

The v4l function has a better algorithm for aligning image size.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/cx88/cx88-video.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/drivers/media/video/cx88/cx88-video.c b/drivers/media/video/cx88/cx88-video.c
index 0ccac702bea4..b12770848c00 100644
--- a/drivers/media/video/cx88/cx88-video.c
+++ b/drivers/media/video/cx88/cx88-video.c
@@ -1111,15 +1111,8 @@ static int vidioc_try_fmt_vid_cap(struct file *file, void *priv,
 	}
 
 	f->fmt.pix.field = field;
-	if (f->fmt.pix.height < 32)
-		f->fmt.pix.height = 32;
-	if (f->fmt.pix.height > maxh)
-		f->fmt.pix.height = maxh;
-	if (f->fmt.pix.width < 48)
-		f->fmt.pix.width = 48;
-	if (f->fmt.pix.width > maxw)
-		f->fmt.pix.width = maxw;
-	f->fmt.pix.width &= ~0x03;
+	v4l_bound_align_image(&f->fmt.pix.width, 48, maxw, 2,
+			      &f->fmt.pix.height, 32, maxh, 0, 0);
 	f->fmt.pix.bytesperline =
 		(f->fmt.pix.width * fmt->depth) >> 3;
 	f->fmt.pix.sizeimage =
-- 
cgit v1.2.3-59-g8ed1b


From 1c657a99fd655c0daa7450854a914d21c1da805c Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11908): w8968cf: Use v4l bounding/alignment function

The v4l function has a better algorithm for aligning image size.

The existing code was casting pointers to u32 and to unsigned int into
pointers to u16.  This could mess up if someone passed in an image size
greater than 65,535 and on big-endian platforms it won't work at all.

The existing bounding code would shrink an image if it was too big, but
returned ERANGE if it was too small.  The code will not shrink or expand as
necessary.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/w9968cf.c | 35 ++++++++++++++---------------------
 1 file changed, 14 insertions(+), 21 deletions(-)

diff --git a/drivers/media/video/w9968cf.c b/drivers/media/video/w9968cf.c
index f59b2bd07e89..6c3f23e31b5c 100644
--- a/drivers/media/video/w9968cf.c
+++ b/drivers/media/video/w9968cf.c
@@ -460,7 +460,7 @@ static int w9968cf_set_picture(struct w9968cf_device*, struct video_picture);
 static int w9968cf_set_window(struct w9968cf_device*, struct video_window);
 static int w9968cf_postprocess_frame(struct w9968cf_device*,
 				     struct w9968cf_frame_t*);
-static int w9968cf_adjust_window_size(struct w9968cf_device*, u16* w, u16* h);
+static int w9968cf_adjust_window_size(struct w9968cf_device*, u32 *w, u32 *h);
 static void w9968cf_init_framelist(struct w9968cf_device*);
 static void w9968cf_push_frame(struct w9968cf_device*, u8 f_num);
 static void w9968cf_pop_frame(struct w9968cf_device*,struct w9968cf_frame_t**);
@@ -1763,8 +1763,7 @@ w9968cf_set_window(struct w9968cf_device* cam, struct video_window win)
 	#define UNSC(x) ((x) >> 10)
 
 	/* Make sure we are using a supported resolution */
-	if ((err = w9968cf_adjust_window_size(cam, (u16*)&win.width,
-					      (u16*)&win.height)))
+	if ((err = w9968cf_adjust_window_size(cam, &win.width, &win.height)))
 		goto error;
 
 	/* Scaling factors */
@@ -1914,12 +1913,9 @@ error:
   Return 0 on success, -1 otherwise.
   --------------------------------------------------------------------------*/
 static int
-w9968cf_adjust_window_size(struct w9968cf_device* cam, u16* width, u16* height)
+w9968cf_adjust_window_size(struct w9968cf_device *cam, u32 *width, u32 *height)
 {
-	u16 maxw, maxh;
-
-	if ((*width < cam->minwidth) || (*height < cam->minheight))
-		return -ERANGE;
+	unsigned int maxw, maxh, align;
 
 	maxw = cam->upscaling && !(cam->vpp_flag & VPP_DECOMPRESSION) &&
 	       w9968cf_vpp ? max((u16)W9968CF_MAX_WIDTH, cam->maxwidth)
@@ -1927,16 +1923,10 @@ w9968cf_adjust_window_size(struct w9968cf_device* cam, u16* width, u16* height)
 	maxh = cam->upscaling && !(cam->vpp_flag & VPP_DECOMPRESSION) &&
 	       w9968cf_vpp ? max((u16)W9968CF_MAX_HEIGHT, cam->maxheight)
 			   : cam->maxheight;
+	align = (cam->vpp_flag & VPP_DECOMPRESSION) ? 4 : 0;
 
-	if (*width > maxw)
-		*width = maxw;
-	if (*height > maxh)
-		*height = maxh;
-
-	if (cam->vpp_flag & VPP_DECOMPRESSION) {
-		*width  &= ~15L; /* multiple of 16 */
-		*height &= ~15L;
-	}
+	v4l_bound_align_image(width, cam->minwidth, maxw, align,
+			      height, cam->minheight, maxh, align, 0);
 
 	PDBGG("Window size adjusted w=%u, h=%u ", *width, *height)
 
@@ -3043,8 +3033,8 @@ static long w9968cf_v4l_ioctl(struct file *filp,
 		if (win.clipcount != 0 || win.flags != 0)
 			return -EINVAL;
 
-		if ((err = w9968cf_adjust_window_size(cam, (u16*)&win.width,
-						      (u16*)&win.height))) {
+		if ((err = w9968cf_adjust_window_size(cam, &win.width,
+						      &win.height))) {
 			DBG(4, "Resolution not supported (%ux%u). "
 			       "VIDIOCSWIN failed", win.width, win.height)
 			return err;
@@ -3116,6 +3106,7 @@ static long w9968cf_v4l_ioctl(struct file *filp,
 	{
 		struct video_mmap mmap;
 		struct w9968cf_frame_t* fr;
+		u32 w, h;
 		int err = 0;
 
 		if (copy_from_user(&mmap, arg, sizeof(mmap)))
@@ -3164,8 +3155,10 @@ static long w9968cf_v4l_ioctl(struct file *filp,
 		   }
 		}
 
-		if ((err = w9968cf_adjust_window_size(cam, (u16*)&mmap.width,
-						      (u16*)&mmap.height))) {
+		w = mmap.width; h = mmap.height;
+		err = w9968cf_adjust_window_size(cam, &w, &h);
+		mmap.width = w; mmap.height = h;
+		if (err) {
 			DBG(4, "Resolution not supported (%dx%d). "
 			       "VIDIOCMCAPTURE failed",
 			    mmap.width, mmap.height)
-- 
cgit v1.2.3-59-g8ed1b


From 2449afcbcc654dbaa9dabeda9daecb69719b0aaa Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11909): cx23885: Use v4l bounding/alignment function

The v4l function has a better algorithm for aligning image size.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/cx23885/cx23885-video.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/drivers/media/video/cx23885/cx23885-video.c b/drivers/media/video/cx23885/cx23885-video.c
index 66bbd2e71105..70836af3ab48 100644
--- a/drivers/media/video/cx23885/cx23885-video.c
+++ b/drivers/media/video/cx23885/cx23885-video.c
@@ -963,15 +963,8 @@ static int vidioc_try_fmt_vid_cap(struct file *file, void *priv,
 	}
 
 	f->fmt.pix.field = field;
-	if (f->fmt.pix.height < 32)
-		f->fmt.pix.height = 32;
-	if (f->fmt.pix.height > maxh)
-		f->fmt.pix.height = maxh;
-	if (f->fmt.pix.width < 48)
-		f->fmt.pix.width = 48;
-	if (f->fmt.pix.width > maxw)
-		f->fmt.pix.width = maxw;
-	f->fmt.pix.width &= ~0x03;
+	v4l_bound_align_image(&f->fmt.pix.width, 48, maxw, 2,
+			      &f->fmt.pix.height, 32, maxh, 0, 0);
 	f->fmt.pix.bytesperline =
 		(f->fmt.pix.width * fmt->depth) >> 3;
 	f->fmt.pix.sizeimage =
-- 
cgit v1.2.3-59-g8ed1b


From 653dc59b6468c2ba51f3b4aee609daa8f67d3e3a Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11910): mt9: Use v4l bounding/alignment function

The v4l function has a better algorithm for aligning image size.

Cc: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/mt9m001.c | 12 +++---------
 drivers/media/video/mt9t031.c | 14 +++-----------
 drivers/media/video/mt9v022.c | 12 +++---------
 3 files changed, 9 insertions(+), 29 deletions(-)

diff --git a/drivers/media/video/mt9m001.c b/drivers/media/video/mt9m001.c
index 459c04cbf69d..4d794b42d6cd 100644
--- a/drivers/media/video/mt9m001.c
+++ b/drivers/media/video/mt9m001.c
@@ -280,15 +280,9 @@ static int mt9m001_try_fmt(struct soc_camera_device *icd,
 {
 	struct v4l2_pix_format *pix = &f->fmt.pix;
 
-	if (pix->height < 32 + icd->y_skip_top)
-		pix->height = 32 + icd->y_skip_top;
-	if (pix->height > 1024 + icd->y_skip_top)
-		pix->height = 1024 + icd->y_skip_top;
-	if (pix->width < 48)
-		pix->width = 48;
-	if (pix->width > 1280)
-		pix->width = 1280;
-	pix->width &= ~0x01; /* has to be even, unsure why was ~3 */
+	v4l_bound_align_image(&pix->width, 48, 1280, 1,
+			      &pix->height, 32 + icd->y_skip_top,
+			      1024 + icd->y_skip_top, 0, 0);
 
 	return 0;
 }
diff --git a/drivers/media/video/mt9t031.c b/drivers/media/video/mt9t031.c
index f72aeb7c4deb..4207fb342670 100644
--- a/drivers/media/video/mt9t031.c
+++ b/drivers/media/video/mt9t031.c
@@ -385,17 +385,9 @@ static int mt9t031_try_fmt(struct soc_camera_device *icd,
 {
 	struct v4l2_pix_format *pix = &f->fmt.pix;
 
-	if (pix->height < MT9T031_MIN_HEIGHT)
-		pix->height = MT9T031_MIN_HEIGHT;
-	if (pix->height > MT9T031_MAX_HEIGHT)
-		pix->height = MT9T031_MAX_HEIGHT;
-	if (pix->width < MT9T031_MIN_WIDTH)
-		pix->width = MT9T031_MIN_WIDTH;
-	if (pix->width > MT9T031_MAX_WIDTH)
-		pix->width = MT9T031_MAX_WIDTH;
-
-	pix->width &= ~0x01; /* has to be even */
-	pix->height &= ~0x01; /* has to be even */
+	v4l_bound_align_image(
+		&pix->width, MT9T031_MIN_WIDTH, MT9T031_MAX_WIDTH, 1,
+		&pix->height, MT9T031_MIN_HEIGHT, MT9T031_MAX_HEIGHT, 1, 0);
 
 	return 0;
 }
diff --git a/drivers/media/video/mt9v022.c b/drivers/media/video/mt9v022.c
index be20d312b1dc..dbdcc86ae50d 100644
--- a/drivers/media/video/mt9v022.c
+++ b/drivers/media/video/mt9v022.c
@@ -364,15 +364,9 @@ static int mt9v022_try_fmt(struct soc_camera_device *icd,
 {
 	struct v4l2_pix_format *pix = &f->fmt.pix;
 
-	if (pix->height < 32 + icd->y_skip_top)
-		pix->height = 32 + icd->y_skip_top;
-	if (pix->height > 480 + icd->y_skip_top)
-		pix->height = 480 + icd->y_skip_top;
-	if (pix->width < 48)
-		pix->width = 48;
-	if (pix->width > 752)
-		pix->width = 752;
-	pix->width &= ~0x03; /* ? */
+	v4l_bound_align_image(&pix->width, 48, 752, 2 /* ? */,
+			      &pix->height, 32 + icd->y_skip_top,
+			      480 + icd->y_skip_top, 0, 0);
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 9bd0e8d7d1bf0dc586bad905c7878b611da3acdc Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11911): cx231xx: Use v4l bounding/alignment function

The v4l function has a better algorithm for aligning image size.

Cc: Srinivasa Deevi <srinivasa.deevi@conexant.com>
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/cx231xx/cx231xx-video.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/drivers/media/video/cx231xx/cx231xx-video.c b/drivers/media/video/cx231xx/cx231xx-video.c
index a23ae73fe634..6a524d847110 100644
--- a/drivers/media/video/cx231xx/cx231xx-video.c
+++ b/drivers/media/video/cx231xx/cx231xx-video.c
@@ -955,8 +955,8 @@ static int vidioc_try_fmt_vid_cap(struct file *file, void *priv,
 {
 	struct cx231xx_fh *fh = priv;
 	struct cx231xx *dev = fh->dev;
-	int width = f->fmt.pix.width;
-	int height = f->fmt.pix.height;
+	unsigned int width = f->fmt.pix.width;
+	unsigned int height = f->fmt.pix.height;
 	unsigned int maxw = norm_maxw(dev);
 	unsigned int maxh = norm_maxh(dev);
 	unsigned int hscale, vscale;
@@ -971,17 +971,7 @@ static int vidioc_try_fmt_vid_cap(struct file *file, void *priv,
 
 	/* width must even because of the YUYV format
 	   height must be even because of interlacing */
-	height &= 0xfffe;
-	width &= 0xfffe;
-
-	if (unlikely(height < 32))
-		height = 32;
-	if (unlikely(height > maxh))
-		height = maxh;
-	if (unlikely(width < 48))
-		width = 48;
-	if (unlikely(width > maxw))
-		width = maxw;
+	v4l_bound_align_image(&width, 48, maxw, 1, &height, 32, maxh, 1, 0);
 
 	get_scale(dev, width, height, &hscale, &vscale);
 
-- 
cgit v1.2.3-59-g8ed1b


From ccb83408b258f7e9f9fe763f9a7d06ebcc21134f Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11912): em28xx: Use v4l bounding/alignment function

The v4l function has a better algorithm for aligning image size.

It appears that the em2800 can only scale by 50% or 100%, i.e. the only
heights supported might be 240 and 480.  In that case the old code would
set any height other than 240 to 480.  Request 240 get 240, but request 239
and then you get 480.  Change it to round to the nearest supported value.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/em28xx/em28xx-video.c | 38 ++++++++++---------------------
 1 file changed, 12 insertions(+), 26 deletions(-)

diff --git a/drivers/media/video/em28xx/em28xx-video.c b/drivers/media/video/em28xx/em28xx-video.c
index 882796e84dbc..8fe1beecfffa 100644
--- a/drivers/media/video/em28xx/em28xx-video.c
+++ b/drivers/media/video/em28xx/em28xx-video.c
@@ -687,8 +687,8 @@ static int vidioc_try_fmt_vid_cap(struct file *file, void *priv,
 {
 	struct em28xx_fh      *fh    = priv;
 	struct em28xx         *dev   = fh->dev;
-	int                   width  = f->fmt.pix.width;
-	int                   height = f->fmt.pix.height;
+	unsigned int          width  = f->fmt.pix.width;
+	unsigned int          height = f->fmt.pix.height;
 	unsigned int          maxw   = norm_maxw(dev);
 	unsigned int          maxh   = norm_maxh(dev);
 	unsigned int          hscale, vscale;
@@ -701,34 +701,20 @@ static int vidioc_try_fmt_vid_cap(struct file *file, void *priv,
 		return -EINVAL;
 	}
 
-	/* width must even because of the YUYV format
-	   height must be even because of interlacing */
-	height &= 0xfffe;
-	width  &= 0xfffe;
-
-	if (unlikely(height < 32))
-		height = 32;
-	if (unlikely(height > maxh))
-		height = maxh;
-	if (unlikely(width < 48))
-		width = 48;
-	if (unlikely(width > maxw))
-		width = maxw;
-
 	if (dev->board.is_em2800) {
 		/* the em2800 can only scale down to 50% */
-		if (height % (maxh / 2))
-			height = maxh;
-		if (width % (maxw / 2))
-			width = maxw;
-		/* according to empiatech support */
-		/* the MaxPacketSize is to small to support */
-		/* framesizes larger than 640x480 @ 30 fps */
-		/* or 640x576 @ 25 fps. As this would cut */
-		/* of a part of the image we prefer */
-		/* 360x576 or 360x480 for now */
+		height = height > (3 * maxh / 4) ? maxh : maxh / 2;
+		width = width > (3 * maxw / 4) ? maxw : maxw / 2;
+		/* According to empiatech support the MaxPacketSize is too small
+		 * to support framesizes larger than 640x480 @ 30 fps or 640x576
+		 * @ 25 fps.  As this would cut of a part of the image we prefer
+		 * 360x576 or 360x480 for now */
 		if (width == maxw && height == maxh)
 			width /= 2;
+	} else {
+		/* width must even because of the YUYV format
+		   height must be even because of interlacing */
+		v4l_bound_align_image(&width, 48, maxw, 1, &height, 32, maxh, 1, 0);
 	}
 
 	get_scale(dev, width, height, &hscale, &vscale);
-- 
cgit v1.2.3-59-g8ed1b


From 1ca27379f3673b40edbd2fec53b93c993fdb4f0c Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Sat, 30 May 2009 21:45:46 -0300
Subject: V4L/DVB (11913): cx231xx: TRY_FMT should not actually set anything

In the TRY_FMT handler the function get_scale() is called to find what the
scaler hardware will produce for a requested size.

The problem is that get_scale(struct cx231xx *dev, ..., unsigned int *vscale,
unsigned int *hscale) saves the calculated scale values into both the
pointer arguments and into dev's hscale and vscale fields.  TRY_FMT shouldn't
actually change anything in the device state.

The code to in get_scale() that writes to dev->[hv]scale can just be
deleted.  In all cases when dev's fields should be modified, get_scale()
was called with get_scale(dev, ..., &dev->hscale, &dev->vscale), so dev was
getting updated anyway.

This didn't actually cause a problem because nothing ever actually made use
of the hscale and vscale fields.  I changed cx231xx_resolution_set() to use
those fields rather than re-calculate them with a call to get_scale().

Updating [hv]scale in cx231xx_resolution_set() isn't necessary because
every call of cx231xx_resolution_set() was already preceded by a call to
get_scale() or setting the [hv]scale fields, so they will be always be
up-to-date w.r.t. width and height.

Removing the call to get_scale() from cx231xx_resolution_set() allowed
making get_scale() a static function, which is a good thing for something
with such a short name.  There is already another function with the same
name in the em28xx driver, but that one is static.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/cx231xx/cx231xx-avcore.c | 17 ++++-------------
 drivers/media/video/cx231xx/cx231xx-video.c  | 10 +++-------
 drivers/media/video/cx231xx/cx231xx.h        |  3 ---
 3 files changed, 7 insertions(+), 23 deletions(-)

diff --git a/drivers/media/video/cx231xx/cx231xx-avcore.c b/drivers/media/video/cx231xx/cx231xx-avcore.c
index 96f07d1473fb..28f48f41f218 100644
--- a/drivers/media/video/cx231xx/cx231xx-avcore.c
+++ b/drivers/media/video/cx231xx/cx231xx-avcore.c
@@ -1052,22 +1052,13 @@ int cx231xx_set_audio_decoder_input(struct cx231xx *dev,
 /* Set resolution of the video */
 int cx231xx_resolution_set(struct cx231xx *dev)
 {
-	int width, height;
-	u32 hscale, vscale;
-	int status = 0;
-
-	width = dev->width;
-	height = dev->height;
-
-	get_scale(dev, width, height, &hscale, &vscale);
-
 	/* set horzontal scale */
-	status = vid_blk_write_word(dev, HSCALE_CTRL, hscale);
+	int status = vid_blk_write_word(dev, HSCALE_CTRL, dev->hscale);
+	if (status)
+		return status;
 
 	/* set vertical scale */
-	status = vid_blk_write_word(dev, VSCALE_CTRL, vscale);
-
-	return status;
+	return vid_blk_write_word(dev, VSCALE_CTRL, dev->vscale);
 }
 
 /******************************************************************************
diff --git a/drivers/media/video/cx231xx/cx231xx-video.c b/drivers/media/video/cx231xx/cx231xx-video.c
index 6a524d847110..609bae6098d3 100644
--- a/drivers/media/video/cx231xx/cx231xx-video.c
+++ b/drivers/media/video/cx231xx/cx231xx-video.c
@@ -893,9 +893,9 @@ static int check_dev(struct cx231xx *dev)
 	return 0;
 }
 
-void get_scale(struct cx231xx *dev,
-	       unsigned int width, unsigned int height,
-	       unsigned int *hscale, unsigned int *vscale)
+static void get_scale(struct cx231xx *dev,
+		      unsigned int width, unsigned int height,
+		      unsigned int *hscale, unsigned int *vscale)
 {
 	unsigned int maxw = norm_maxw(dev);
 	unsigned int maxh = norm_maxh(dev);
@@ -907,10 +907,6 @@ void get_scale(struct cx231xx *dev,
 	*vscale = (((unsigned long)maxh) << 12) / height - 4096L;
 	if (*vscale >= 0x4000)
 		*vscale = 0x3fff;
-
-	dev->hscale = *hscale;
-	dev->vscale = *vscale;
-
 }
 
 /* ------------------------------------------------------------------
diff --git a/drivers/media/video/cx231xx/cx231xx.h b/drivers/media/video/cx231xx/cx231xx.h
index e38eb2d425f7..a0f823ac6b8d 100644
--- a/drivers/media/video/cx231xx/cx231xx.h
+++ b/drivers/media/video/cx231xx/cx231xx.h
@@ -722,9 +722,6 @@ int cx231xx_set_video_input_mux(struct cx231xx *dev, u8 input);
 int cx231xx_set_decoder_video_input(struct cx231xx *dev, u8 pin_type, u8 input);
 int cx231xx_do_mode_ctrl_overrides(struct cx231xx *dev);
 int cx231xx_set_audio_input(struct cx231xx *dev, u8 input);
-void get_scale(struct cx231xx *dev,
-	       unsigned int width, unsigned int height,
-	       unsigned int *hscale, unsigned int *vscale);
 
 /* Provided by cx231xx-video.c */
 int cx231xx_register_extension(struct cx231xx_ops *dev);
-- 
cgit v1.2.3-59-g8ed1b


From d8b2996607d492ffa99628bafc80da14d3a5482d Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Fri, 12 Jun 2009 16:31:29 -0300
Subject: V4L/DVB (12003): v4l2: Move bounding code outside I2C ifdef block

On Fri, 12 Jun 2009, Randy Dunlap wrote:
> From: Randy Dunlap <randy.dunlap@oracle.com>
>
> Move v4l_bound_align_image() outside of an #ifdef CONFIG_I2C block
> so that it is always built.  Fixes a build error:

clamp_align() should be moved as well, since it's only used by
v4l_bound_align_image().  I'm attaching an alternate version that fixes
this.  Labeled the endif too.

Reported-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/video/v4l2-common.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/media/video/v4l2-common.c b/drivers/media/video/v4l2-common.c
index 1ebbe738019f..b91d66a767d7 100644
--- a/drivers/media/video/v4l2-common.c
+++ b/drivers/media/video/v4l2-common.c
@@ -1021,6 +1021,8 @@ const unsigned short *v4l2_i2c_tuner_addrs(enum v4l2_i2c_tuner_type type)
 }
 EXPORT_SYMBOL_GPL(v4l2_i2c_tuner_addrs);
 
+#endif /* defined(CONFIG_I2C) */
+
 /* Clamp x to be between min and max, aligned to a multiple of 2^align.  min
  * and max don't have to be aligned, but there must be at least one valid
  * value.  E.g., min=17,max=31,align=4 is not allowed as there are no multiples
@@ -1094,4 +1096,3 @@ void v4l_bound_align_image(u32 *w, unsigned int wmin, unsigned int wmax,
 	}
 }
 EXPORT_SYMBOL_GPL(v4l_bound_align_image);
-#endif /* defined(CONFIG_I2C) */
-- 
cgit v1.2.3-59-g8ed1b


From 0a5ded56fd3f4096681f8e6a249fb058485f4e46 Mon Sep 17 00:00:00 2001
From: Manu Abraham <abraham.manu@gmail.com>
Date: Wed, 17 Jun 2009 16:48:17 -0300
Subject: V4L/DVB (12130): Fix a redundant compiler warning
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drivers/media/dvb/frontends/stv090x.c: In function ‘stv090x_optimize_carloop_short’:
drivers/media/dvb/frontends/stv090x.c:2677: warning: ‘short_crl’ may be used uninitialized in this function

Signed-off-by: Manu Abraham <manu@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/dvb/frontends/stv090x.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/media/dvb/frontends/stv090x.c b/drivers/media/dvb/frontends/stv090x.c
index 96ef745a2e4e..4e89bb02bff3 100644
--- a/drivers/media/dvb/frontends/stv090x.c
+++ b/drivers/media/dvb/frontends/stv090x.c
@@ -2674,7 +2674,7 @@ static u8 stv090x_optimize_carloop(struct stv090x_state *state, enum stv090x_mod
 
 static u8 stv090x_optimize_carloop_short(struct stv090x_state *state)
 {
-	struct stv090x_short_frame_crloop *short_crl;
+	struct stv090x_short_frame_crloop *short_crl = NULL;
 	s32 index = 0;
 	u8 aclc = 0x0b;
 
-- 
cgit v1.2.3-59-g8ed1b


From eebf8d86acf0db974dfaad8e8285f4e12ca488e2 Mon Sep 17 00:00:00 2001
From: Manu Abraham <abraham.manu@gmail.com>
Date: Thu, 18 Jun 2009 04:50:53 -0300
Subject: V4L/DVB (12131): BUGFIX: An incorrect Carrier Recovery Loop
 optimization table was being

loaded for a given chip version. This would cause the optimization in
tuning not to be applied and thus a failed expectation, in tuning speed
increment. The patch swaps the tables in use. It also fixes a possible
one in a million condition where state->dev_ver implies an older Cut
(Cut < 2.0, eventhough the driver doesn't attach to any Cut older than
2.0) or even negative (due to a bad I2C bus master driver) for the card
combination.

Thanks to Mauro Carvalho Chehab <mchehab@infradead.org> for pointing
out the issue at large.

Signed-off-by: Manu Abraham <manu@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
---
 drivers/media/dvb/frontends/stv090x.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/media/dvb/frontends/stv090x.c b/drivers/media/dvb/frontends/stv090x.c
index 4e89bb02bff3..488bdfb34fb3 100644
--- a/drivers/media/dvb/frontends/stv090x.c
+++ b/drivers/media/dvb/frontends/stv090x.c
@@ -2694,10 +2694,13 @@ static u8 stv090x_optimize_carloop_short(struct stv090x_state *state)
 		break;
 	}
 
-	if (state->dev_ver >= 0x30)
-		short_crl = stv090x_s2_short_crl_cut20;
-	else if (state->dev_ver >= 0x20)
+	if (state->dev_ver >= 0x30) {
+		/* Cut 3.0 and up */
 		short_crl = stv090x_s2_short_crl_cut30;
+	} else {
+		/* Cut 2.0 and up: we don't support cuts older than 2.0 */
+		short_crl = stv090x_s2_short_crl_cut20;
+	}
 
 	if (state->srate <= 3000000)
 		aclc = short_crl[index].crl_2;
-- 
cgit v1.2.3-59-g8ed1b


From 789cd4702bf830416d2e1794495407be42fe95ad Mon Sep 17 00:00:00 2001
From: Ulrik Bech Hald <ubh@ti.com>
Date: Fri, 12 Jun 2009 16:18:32 -0500
Subject: [WATCHDOG] OMAP fixes: enable clock in probe, trigger timer reload

This patch contains two fixes:

1)In omap_wdt_probe() the watchdog is reset and disabled. This
requires register access and the clks needs to be enabled temporarily

2)In omap_wdt_open() the timer register needs to be reloaded
to trigger a new timer value (the default of 60s)

Tested on OMAP34xx platform (Zoom1)

Reviewed-by: Kevin Hilman <khilman@ti.deeprootsystems.com>
Signed-off-by: Ulrik Bech Hald <ubh@ti.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/watchdog/omap_wdt.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/watchdog/omap_wdt.c b/drivers/watchdog/omap_wdt.c
index f2713851aaab..3ed571a2ab18 100644
--- a/drivers/watchdog/omap_wdt.c
+++ b/drivers/watchdog/omap_wdt.c
@@ -159,6 +159,7 @@ static int omap_wdt_open(struct inode *inode, struct file *file)
 	file->private_data = (void *) wdev;
 
 	omap_wdt_set_timeout(wdev);
+	omap_wdt_ping(wdev); /* trigger loading of new timeout value */
 	omap_wdt_enable(wdev);
 
 	return nonseekable_open(inode, file);
@@ -313,6 +314,9 @@ static int __devinit omap_wdt_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, wdev);
 
+	clk_enable(wdev->ick);
+	clk_enable(wdev->fck);
+
 	omap_wdt_disable(wdev);
 	omap_wdt_adjust_timeout(timer_margin);
 
@@ -332,6 +336,9 @@ static int __devinit omap_wdt_probe(struct platform_device *pdev)
 	/* autogate OCP interface clock */
 	__raw_writel(0x01, wdev->base + OMAP_WATCHDOG_SYS_CONFIG);
 
+	clk_disable(wdev->ick);
+	clk_disable(wdev->fck);
+
 	omap_wdt_dev = pdev;
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 44df75353bc8f32e26e049284053a61d4f1047d6 Mon Sep 17 00:00:00 2001
From: Tom Mingarelli <thomas.mingarelli@hp.com>
Date: Thu, 18 Jun 2009 23:28:57 +0000
Subject: [WATCHDOG] hpwdt: Add NMI priority option

Add a priority option so that the user can choose if we do the NMI
first or last.

Signed-off-by: Thomas Mingarelli <thomas.mingarelli@hp.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 Documentation/watchdog/hpwdt.txt | 19 +++++++++++++++----
 drivers/watchdog/hpwdt.c         | 26 +++++++++++++++++++++-----
 2 files changed, 36 insertions(+), 9 deletions(-)

diff --git a/Documentation/watchdog/hpwdt.txt b/Documentation/watchdog/hpwdt.txt
index 127839e53043..9c24d5ffbb06 100644
--- a/Documentation/watchdog/hpwdt.txt
+++ b/Documentation/watchdog/hpwdt.txt
@@ -19,30 +19,41 @@ Last reviewed: 06/02/2009
  not be updated in a timely fashion and a hardware system reset (also known as
  an Automatic Server Recovery (ASR)) event will occur.
 
- The hpwdt driver also has three (3) module parameters. They are the following:
+ The hpwdt driver also has four (4) module parameters. They are the following:
 
  soft_margin - allows the user to set the watchdog timer value
  allow_kdump - allows the user to save off a kernel dump image after an NMI
  nowayout    - basic watchdog parameter that does not allow the timer to
                be restarted or an impending ASR to be escaped.
+ priority    - determines whether or not the hpwdt driver is first on the
+               die_notify list to handle NMIs or last. The default value
+               for this module parameter is 0 or LAST. If the user wants to
+               enable NMI sourcing then reload the hpwdt driver with
+               priority=1 (and boot with nmi_watchdog=0).
 
  NOTE: More information about watchdog drivers in general, including the ioctl
        interface to /dev/watchdog can be found in
        Documentation/watchdog/watchdog-api.txt and Documentation/IPMI.txt.
 
- The NMI sourcing capability is disabled when the driver discovers that the
- nmi_watchdog is turned on (nmi_watchdog = 1). This is due to the inability to
+ The priority parameter was introduced due to other kernel software that relied
+ on handling NMIs (like oprofile). Keeping hpwdt's priority at 0 (or LAST)
+ enables the users of NMIs for non critical events to be work as expected.
+
+ The NMI sourcing capability is disabled by default due to the inability to
  distinguish between "NMI Watchdog Ticks" and "HW generated NMI events" in the
  Linux kernel. What this means is that the hpwdt nmi handler code is called
  each time the NMI signal fires off. This could amount to several thousands of
  NMIs in a matter of seconds. If a user sees the Linux kernel's "dazed and
  confused" message in the logs or if the system gets into a hung state, then
- the user should reboot with nmi_watchdog=0.
+ the hpwdt driver can be reloaded with the "priority" module parameter set
+ (priority=1).
 
  1. If the kernel has not been booted with nmi_watchdog turned off then
     edit /boot/grub/menu.lst and place the nmi_watchdog=0 at the end of the
     currently booting kernel line.
  2. reboot the sever
+ 3. Once the system comes up perform a rmmod hpwdt
+ 4. insmod /lib/modules/`uname -r`/kernel/drivers/char/watchdog/hpwdt.ko priority=1
 
  Now, the hpwdt can successfully receive and source the NMI and provide a log
  message that details the reason for the NMI (as determined by the HP BIOS).
diff --git a/drivers/watchdog/hpwdt.c b/drivers/watchdog/hpwdt.c
index c0b9169ba5d5..a6c5674c78e6 100644
--- a/drivers/watchdog/hpwdt.c
+++ b/drivers/watchdog/hpwdt.c
@@ -120,7 +120,8 @@ static int nowayout = WATCHDOG_NOWAYOUT;
 static char expect_release;
 static unsigned long hpwdt_is_open;
 static unsigned int allow_kdump;
-static int hpwdt_nmi_sourcing;
+static unsigned int hpwdt_nmi_sourcing;
+static unsigned int priority;		/* hpwdt at end of die_notify list */
 
 static void __iomem *pci_mem_addr;		/* the PCI-memory address */
 static unsigned long __iomem *hpwdt_timer_reg;
@@ -623,7 +624,7 @@ static struct miscdevice hpwdt_miscdev = {
 
 static struct notifier_block die_notifier = {
 	.notifier_call = hpwdt_pretimeout,
-	.priority = 0x7FFFFFFF,
+	.priority = 0,
 };
 
 /*
@@ -641,7 +642,8 @@ static void __devinit hpwdt_check_nmi_sourcing(struct pci_dev *dev)
 		hpwdt_nmi_sourcing = 1;
 	else
 		dev_warn(&dev->dev, "NMI sourcing is disabled. To enable this "
-			"functionality you must reboot with nmi_watchdog=0.\n");
+			"functionality you must reboot with nmi_watchdog=0 "
+			"and load the hpwdt driver with priority=1.\n");
 }
 #else
 static void __devinit hpwdt_check_nmi_sourcing(struct pci_dev *dev)
@@ -714,6 +716,14 @@ static int __devinit hpwdt_init_one(struct pci_dev *dev,
 	cmn_regs.u1.rah = 0x0D;
 	cmn_regs.u1.ral = 0x02;
 
+	/*
+	 * If the priority is set to 1, then we will be put first on the
+	 * die notify list to handle a critical NMI. The default is to
+	 * be last so other users of the NMI signal can function.
+	 */
+	if (priority)
+		die_notifier.priority = 0x7FFFFFFF;
+
 	retval = register_die_notifier(&die_notifier);
 	if (retval != 0) {
 		dev_warn(&dev->dev,
@@ -733,9 +743,11 @@ static int __devinit hpwdt_init_one(struct pci_dev *dev,
 	printk(KERN_INFO
 		"hp Watchdog Timer Driver: %s"
 		", timer margin: %d seconds (nowayout=%d)"
-		", allow kernel dump: %s (default = 0/OFF).\n",
+		", allow kernel dump: %s (default = 0/OFF)"
+		", priority: %s (default = 0/LAST).\n",
 		HPWDT_VERSION, soft_margin, nowayout,
-		(allow_kdump == 0) ? "OFF" : "ON");
+		(allow_kdump == 0) ? "OFF" : "ON",
+		(priority == 0) ? "LAST" : "FIRST");
 
 	return 0;
 
@@ -798,5 +810,9 @@ module_param(nowayout, int, 0);
 MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default="
 		__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
 
+module_param(priority, int, 0);
+MODULE_PARM_DESC(priority, "The hpwdt driver handles NMIs first or last"
+		" (default = 0/Last)\n");
+
 module_init(hpwdt_init);
 module_exit(hpwdt_cleanup);
-- 
cgit v1.2.3-59-g8ed1b


From 9b901ee0cb007eb4e2ee056e5b1c5c2837d53bdb Mon Sep 17 00:00:00 2001
From: Wim Van Sebroeck <wim@iguana.be>
Date: Fri, 19 Jun 2009 09:32:57 +0000
Subject: [WATCHDOG] wdt_pci.c: remove #ifdef CONFIG_WDT_501_PCI

Change the wdt_pci.c watchdog driver so that the code is the same for
both the PCI-WDT500 as the PCI-WDT501 card. The selection of the card
is now being done via the module parameter: 'type' instead of the
config option CONFIG_WDT_501_PCI.

Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/watchdog/Kconfig   |  18 ++-----
 drivers/watchdog/wdt_pci.c | 122 ++++++++++++++++++++++-----------------------
 2 files changed, 66 insertions(+), 74 deletions(-)

diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index e8d45b6ccef8..b1ccc04f3c9a 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -1007,24 +1007,16 @@ config WDTPCI
 	---help---
 	  If you have a PCI-WDT500/501 watchdog board, say Y here, otherwise N.
 
-	  To compile this driver as a module, choose M here: the
-	  module will be called wdt_pci.
-
-config WDT_501_PCI
-	bool "PCI-WDT501 features"
-	depends on WDTPCI
-	help
-	  Saying Y here and creating a character special file /dev/temperature
-	  with major number 10 and minor number 131 ("man mknod") will give
-	  you a thermometer inside your computer: reading from
-	  /dev/temperature yields one byte, the temperature in degrees
-	  Fahrenheit. This works only if you have a PCI-WDT501 watchdog board
-	  installed.
+	  If you have a PCI-WDT501 watchdog board then you can enable the
+	  temperature sensor by setting the type parameter to 501.
 
 	  If you want to enable the Fan Tachometer on the PCI-WDT501, then you
 	  can do this via the tachometer parameter. Only do this if you have a
 	  fan tachometer actually set up.
 
+	  To compile this driver as a module, choose M here: the
+	  module will be called wdt_pci.
+
 #
 # USB-based Watchdog Cards
 #
diff --git a/drivers/watchdog/wdt_pci.c b/drivers/watchdog/wdt_pci.c
index c45839a4a34d..7a1bdc7c95a9 100644
--- a/drivers/watchdog/wdt_pci.c
+++ b/drivers/watchdog/wdt_pci.c
@@ -2,7 +2,7 @@
  *	Industrial Computer Source PCI-WDT500/501 driver
  *
  *	(c) Copyright 1996-1997 Alan Cox <alan@lxorguk.ukuu.org.uk>,
- *						 All Rights Reserved.
+ *						All Rights Reserved.
  *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License
@@ -99,14 +99,16 @@ MODULE_PARM_DESC(nowayout,
 		"Watchdog cannot be stopped once started (default="
 				__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
 
-#ifdef CONFIG_WDT_501_PCI
 /* Support for the Fan Tachometer on the PCI-WDT501 */
 static int tachometer;
-
 module_param(tachometer, int, 0);
 MODULE_PARM_DESC(tachometer,
-	"PCI-WDT501 Fan Tachometer support (0=disable, default=0)");
-#endif /* CONFIG_WDT_501_PCI */
+		"PCI-WDT501 Fan Tachometer support (0=disable, default=0)");
+
+static int type = 500;
+module_param(type, int, 0);
+MODULE_PARM_DESC(type,
+		"PCI-WDT501 Card type (500 or 501 , default=500)");
 
 /*
  *	Programming support
@@ -266,22 +268,21 @@ static int wdtpci_get_status(int *status)
 		*status |= WDIOF_EXTERN1;
 	if (new_status & WDC_SR_ISII1)
 		*status |= WDIOF_EXTERN2;
-#ifdef CONFIG_WDT_501_PCI
-	if (!(new_status & WDC_SR_TGOOD))
-		*status |= WDIOF_OVERHEAT;
-	if (!(new_status & WDC_SR_PSUOVER))
-		*status |= WDIOF_POWEROVER;
-	if (!(new_status & WDC_SR_PSUUNDR))
-		*status |= WDIOF_POWERUNDER;
-	if (tachometer) {
-		if (!(new_status & WDC_SR_FANGOOD))
-			*status |= WDIOF_FANFAULT;
+	if (type == 501) {
+		if (!(new_status & WDC_SR_TGOOD))
+			*status |= WDIOF_OVERHEAT;
+		if (!(new_status & WDC_SR_PSUOVER))
+			*status |= WDIOF_POWEROVER;
+		if (!(new_status & WDC_SR_PSUUNDR))
+			*status |= WDIOF_POWERUNDER;
+		if (tachometer) {
+			if (!(new_status & WDC_SR_FANGOOD))
+				*status |= WDIOF_FANFAULT;
+		}
 	}
-#endif /* CONFIG_WDT_501_PCI */
 	return 0;
 }
 
-#ifdef CONFIG_WDT_501_PCI
 /**
  *	wdtpci_get_temperature:
  *
@@ -300,7 +301,6 @@ static int wdtpci_get_temperature(int *temperature)
 	*temperature = (c * 11 / 15) + 7;
 	return 0;
 }
-#endif /* CONFIG_WDT_501_PCI */
 
 /**
  *	wdtpci_interrupt:
@@ -327,22 +327,22 @@ static irqreturn_t wdtpci_interrupt(int irq, void *dev_id)
 
 	printk(KERN_CRIT PFX "status %d\n", status);
 
-#ifdef CONFIG_WDT_501_PCI
-	if (!(status & WDC_SR_TGOOD)) {
-		u8 alarm = inb(WDT_RT);
-		printk(KERN_CRIT PFX "Overheat alarm.(%d)\n", alarm);
-		udelay(8);
-	}
-	if (!(status & WDC_SR_PSUOVER))
-		printk(KERN_CRIT PFX "PSU over voltage.\n");
-	if (!(status & WDC_SR_PSUUNDR))
-		printk(KERN_CRIT PFX "PSU under voltage.\n");
-	if (tachometer) {
-		if (!(status & WDC_SR_FANGOOD))
-			printk(KERN_CRIT PFX "Possible fan fault.\n");
+	if (type == 501) {
+		if (!(status & WDC_SR_TGOOD)) {
+			printk(KERN_CRIT PFX "Overheat alarm.(%d)\n",
+								inb(WDT_RT));
+			udelay(8);
+		}
+		if (!(status & WDC_SR_PSUOVER))
+			printk(KERN_CRIT PFX "PSU over voltage.\n");
+		if (!(status & WDC_SR_PSUUNDR))
+			printk(KERN_CRIT PFX "PSU under voltage.\n");
+		if (tachometer) {
+			if (!(status & WDC_SR_FANGOOD))
+				printk(KERN_CRIT PFX "Possible fan fault.\n");
+		}
 	}
-#endif /* CONFIG_WDT_501_PCI */
-	if (!(status&WDC_SR_WCCR)) {
+	if (!(status & WDC_SR_WCCR)) {
 #ifdef SOFTWARE_REBOOT
 #ifdef ONLY_TESTING
 		printk(KERN_CRIT PFX "Would Reboot.\n");
@@ -371,12 +371,13 @@ static irqreturn_t wdtpci_interrupt(int irq, void *dev_id)
  */
 
 static ssize_t wdtpci_write(struct file *file, const char __user *buf,
-					size_t count, loff_t *ppos)
+						size_t count, loff_t *ppos)
 {
 	if (count) {
 		if (!nowayout) {
 			size_t i;
 
+			/* In case it was set long ago */
 			expect_close = 0;
 
 			for (i = 0; i != count; i++) {
@@ -406,10 +407,10 @@ static ssize_t wdtpci_write(struct file *file, const char __user *buf,
 static long wdtpci_ioctl(struct file *file, unsigned int cmd,
 							unsigned long arg)
 {
-	int new_heartbeat;
-	int status;
 	void __user *argp = (void __user *)arg;
 	int __user *p = argp;
+	int new_heartbeat;
+	int status;
 
 	static struct watchdog_info ident = {
 		.options =		WDIOF_SETTIMEOUT|
@@ -421,11 +422,12 @@ static long wdtpci_ioctl(struct file *file, unsigned int cmd,
 
 	/* Add options according to the card we have */
 	ident.options |= (WDIOF_EXTERN1|WDIOF_EXTERN2);
-#ifdef CONFIG_WDT_501_PCI
-	ident.options |= (WDIOF_OVERHEAT|WDIOF_POWERUNDER|WDIOF_POWEROVER);
-	if (tachometer)
-		ident.options |= WDIOF_FANFAULT;
-#endif /* CONFIG_WDT_501_PCI */
+	if (type == 501) {
+		ident.options |= (WDIOF_OVERHEAT|WDIOF_POWERUNDER|
+							WDIOF_POWEROVER);
+		if (tachometer)
+			ident.options |= WDIOF_FANFAULT;
+	}
 
 	switch (cmd) {
 	case WDIOC_GETSUPPORT:
@@ -503,7 +505,6 @@ static int wdtpci_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-#ifdef CONFIG_WDT_501_PCI
 /**
  *	wdtpci_temp_read:
  *	@file: file handle to the watchdog board
@@ -554,7 +555,6 @@ static int wdtpci_temp_release(struct inode *inode, struct file *file)
 {
 	return 0;
 }
-#endif /* CONFIG_WDT_501_PCI */
 
 /**
  *	notify_sys:
@@ -596,7 +596,6 @@ static struct miscdevice wdtpci_miscdev = {
 	.fops	= &wdtpci_fops,
 };
 
-#ifdef CONFIG_WDT_501_PCI
 static const struct file_operations wdtpci_temp_fops = {
 	.owner		= THIS_MODULE,
 	.llseek		= no_llseek,
@@ -610,7 +609,6 @@ static struct miscdevice temp_miscdev = {
 	.name	= "temperature",
 	.fops	= &wdtpci_temp_fops,
 };
-#endif /* CONFIG_WDT_501_PCI */
 
 /*
  *	The WDT card needs to learn about soft shutdowns in order to
@@ -633,6 +631,11 @@ static int __devinit wdtpci_init_one(struct pci_dev *dev,
 		return -ENODEV;
 	}
 
+	if (type != 500 && type != 501) {
+		printk(KERN_ERR PFX "unknown card type '%d'.\n", type);
+		return -ENODEV;
+	}
+
 	if (pci_enable_device(dev)) {
 		printk(KERN_ERR PFX "Not possible to enable PCI Device\n");
 		return -ENODEV;
@@ -678,15 +681,15 @@ static int __devinit wdtpci_init_one(struct pci_dev *dev,
 		goto out_irq;
 	}
 
-#ifdef CONFIG_WDT_501_PCI
-	ret = misc_register(&temp_miscdev);
-	if (ret) {
-		printk(KERN_ERR PFX
+	if (type == 501) {
+		ret = misc_register(&temp_miscdev);
+		if (ret) {
+			printk(KERN_ERR PFX
 			"cannot register miscdev on minor=%d (err=%d)\n",
-					TEMP_MINOR, ret);
-		goto out_rbt;
+							TEMP_MINOR, ret);
+			goto out_rbt;
+		}
 	}
-#endif /* CONFIG_WDT_501_PCI */
 
 	ret = misc_register(&wdtpci_miscdev);
 	if (ret) {
@@ -698,20 +701,18 @@ static int __devinit wdtpci_init_one(struct pci_dev *dev,
 
 	printk(KERN_INFO PFX "initialized. heartbeat=%d sec (nowayout=%d)\n",
 		heartbeat, nowayout);
-#ifdef CONFIG_WDT_501_PCI
-	printk(KERN_INFO "wdt: Fan Tachometer is %s\n",
+	if (type == 501)
+		printk(KERN_INFO "wdt: Fan Tachometer is %s\n",
 				(tachometer ? "Enabled" : "Disabled"));
-#endif /* CONFIG_WDT_501_PCI */
 
 	ret = 0;
 out:
 	return ret;
 
 out_misc:
-#ifdef CONFIG_WDT_501_PCI
-	misc_deregister(&temp_miscdev);
+	if (type == 501)
+		misc_deregister(&temp_miscdev);
 out_rbt:
-#endif /* CONFIG_WDT_501_PCI */
 	unregister_reboot_notifier(&wdtpci_notifier);
 out_irq:
 	free_irq(irq, &wdtpci_miscdev);
@@ -728,9 +729,8 @@ static void __devexit wdtpci_remove_one(struct pci_dev *pdev)
 	/* here we assume only one device will ever have
 	 * been picked up and registered by probe function */
 	misc_deregister(&wdtpci_miscdev);
-#ifdef CONFIG_WDT_501_PCI
-	misc_deregister(&temp_miscdev);
-#endif /* CONFIG_WDT_501_PCI */
+	if (type == 501)
+		misc_deregister(&temp_miscdev);
 	unregister_reboot_notifier(&wdtpci_notifier);
 	free_irq(irq, &wdtpci_miscdev);
 	release_region(io, 16);
-- 
cgit v1.2.3-59-g8ed1b


From 7b994836f09fc3ce7d073ad6f8259a1a84003e02 Mon Sep 17 00:00:00 2001
From: Roel Kluin <[mailto:roel.kluin@gmail.com]>
Date: Tue, 23 Jun 2009 10:04:14 +0200
Subject: CRISv10: remove redundant tests on unsigned

Since dmanr is unsigned, negatives are wrapped and caught by the other test.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Acked-by: Jesper Nilsson <jesper.nilsson@axis.com>
---
 arch/cris/arch-v10/kernel/dma.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/cris/arch-v10/kernel/dma.c b/arch/cris/arch-v10/kernel/dma.c
index 929e68666299..d31504b4a19e 100644
--- a/arch/cris/arch-v10/kernel/dma.c
+++ b/arch/cris/arch-v10/kernel/dma.c
@@ -24,7 +24,7 @@ int cris_request_dma(unsigned int dmanr, const char * device_id,
 	unsigned long int gens;
 	int fail = -EINVAL;
 
-	if ((dmanr < 0) || (dmanr >= MAX_DMA_CHANNELS)) {
+	if (dmanr >= MAX_DMA_CHANNELS) {
 		printk(KERN_CRIT "cris_request_dma: invalid DMA channel %u\n", dmanr);
 		return -EINVAL;
 	}
@@ -213,7 +213,7 @@ int cris_request_dma(unsigned int dmanr, const char * device_id,
 void cris_free_dma(unsigned int dmanr, const char * device_id)
 {
 	unsigned long flags;
-	if ((dmanr < 0) || (dmanr >= MAX_DMA_CHANNELS)) {
+	if (dmanr >= MAX_DMA_CHANNELS) {
 		printk(KERN_CRIT "cris_free_dma: invalid DMA channel %u\n", dmanr);
 		return;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From dee412066aeb16c43cf31599948c1a1de385df56 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Tue, 23 Jun 2009 02:22:39 +0530
Subject: perf stat: Fix command option / manpage

-l is not supported, it should be -S for scale.

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <1245703959.6167.16.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/Documentation/perf-stat.txt | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
index c368a72721d7..0d74346d21ab 100644
--- a/tools/perf/Documentation/perf-stat.txt
+++ b/tools/perf/Documentation/perf-stat.txt
@@ -8,8 +8,8 @@ perf-stat - Run a command and gather performance counter statistics
 SYNOPSIS
 --------
 [verse]
-'perf stat' [-e <EVENT> | --event=EVENT] [-l] [-a] <command>
-'perf stat' [-e <EVENT> | --event=EVENT] [-l] [-a] -- <command> [<options>]
+'perf stat' [-e <EVENT> | --event=EVENT] [-S] [-a] <command>
+'perf stat' [-e <EVENT> | --event=EVENT] [-S] [-a] -- <command> [<options>]
 
 DESCRIPTION
 -----------
@@ -40,7 +40,7 @@ OPTIONS
 -a::
         system-wide collection
 
--l::
+-S::
         scale counter values
 
 EXAMPLES
-- 
cgit v1.2.3-59-g8ed1b


From 2e046b9487dcc60707cac77fb8f744ec830209cd Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 19 Jun 2009 14:40:51 +0900
Subject: sh: Provide cpu_idle_wait() to fix up cpuidle/SMP build.

Crib the x86 cpu_idle_wait() implementation and shove it in with the
idle code, subsequently enabling ARCH_HAS_CPU_IDLE_WAIT.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig              |  3 +++
 arch/sh/include/asm/system.h |  1 +
 arch/sh/kernel/idle.c        | 23 ++++++++++++++++++++++-
 3 files changed, 26 insertions(+), 1 deletion(-)

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index ac1c620d1c7d..09c0aef31d03 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -151,6 +151,9 @@ config ARCH_NO_VIRT_TO_BUS
 config ARCH_HAS_DEFAULT_IDLE
 	def_bool y
 
+config ARCH_HAS_CPU_IDLE_WAIT
+	def_bool y
+
 config IO_TRAPPED
 	bool
 
diff --git a/arch/sh/include/asm/system.h b/arch/sh/include/asm/system.h
index a88895e6dcb0..ab79e1f4fbe0 100644
--- a/arch/sh/include/asm/system.h
+++ b/arch/sh/include/asm/system.h
@@ -154,6 +154,7 @@ extern struct dentry *sh_debugfs_root;
 
 void per_cpu_trap_init(void);
 void default_idle(void);
+void cpu_idle_wait(void);
 
 asmlinkage void break_point_trap(void);
 
diff --git a/arch/sh/kernel/idle.c b/arch/sh/kernel/idle.c
index f35ed0348850..27ff2dc093c7 100644
--- a/arch/sh/kernel/idle.c
+++ b/arch/sh/kernel/idle.c
@@ -1,7 +1,7 @@
 /*
  * The idle loop for all SuperH platforms.
  *
- *  Copyright (C) 2002 - 2008  Paul Mundt
+ *  Copyright (C) 2002 - 2009  Paul Mundt
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
@@ -15,6 +15,7 @@
 #include <linux/preempt.h>
 #include <linux/thread_info.h>
 #include <linux/irqflags.h>
+#include <linux/smp.h>
 #include <asm/pgalloc.h>
 #include <asm/system.h>
 #include <asm/atomic.h>
@@ -79,3 +80,23 @@ void cpu_idle(void)
 		check_pgt_cache();
 	}
 }
+
+static void do_nothing(void *unused)
+{
+}
+
+/*
+ * cpu_idle_wait - Used to ensure that all the CPUs discard old value of
+ * pm_idle and update to new pm_idle value. Required while changing pm_idle
+ * handler on SMP systems.
+ *
+ * Caller must have changed pm_idle to the new value before the call. Old
+ * pm_idle value will not be used by any CPU after the return of this function.
+ */
+void cpu_idle_wait(void)
+{
+	smp_mb();
+	/* kick all the CPUs so that they exit out of pm_idle */
+	smp_call_function(do_nothing, NULL, 1);
+}
+EXPORT_SYMBOL_GPL(cpu_idle_wait);
-- 
cgit v1.2.3-59-g8ed1b


From 997d0030934f7722d81e5b1cdd7af85ac7ed960b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 19 Jun 2009 15:37:11 +0900
Subject: sh: Use local TLB flush in set_pte_phys().

set_pte_phys() presently uses the global flush_tlb_one(), which locks on
SMP trying to do the IPI. As we have not even initialized the other CPUs
at this point, switch to the local_ variant so the flush happens on the
boot CPU.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/init.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index ee8e6bbe882c..d1be04204219 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -70,7 +70,7 @@ static void set_pte_phys(unsigned long addr, unsigned long phys, pgprot_t prot)
 	}
 
 	set_pte(pte, pfn_pte(phys >> PAGE_SHIFT, prot));
-	flush_tlb_one(get_asid(), addr);
+	local_flush_tlb_one(get_asid(), addr);
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From b29fa1fbc210a51103c7a16af44df42cadce0361 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 19 Jun 2009 15:39:47 +0900
Subject: sh: Wire up the uncached fixmap on sh64 as well.

Now that sh64 also can use the uncached section, wire up the fixmap for
it as well.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/init.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index d1be04204219..fe532aeaa16d 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -177,10 +177,8 @@ void __init paging_init(void)
 
 	free_area_init_nodes(max_zone_pfns);
 
-#ifdef CONFIG_SUPERH32
 	/* Set up the uncached fixmap */
 	set_fixmap_nocache(FIX_UNCACHED, __pa(&__uncached_start));
-#endif
 }
 
 static struct kcore_list kcore_mem, kcore_vmalloc;
-- 
cgit v1.2.3-59-g8ed1b


From 2eb2a4368273e123d2716a704c86f3130aa14c22 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 19 Jun 2009 16:00:00 +0900
Subject: sh: SH7786 SMP support.

SH7786 is roughly identical to SH-X3 proto SMP, though there are only 2
CPUs. This just wraps in to the existing SH-X3 SMP code with some minor
changes for SH7786, including wiring up the IPIs properly, enabling
IRQ_PER_CPU, and so forth.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                        |  7 ++++++-
 arch/sh/kernel/cpu/sh4a/Makefile       |  6 +++---
 arch/sh/kernel/cpu/sh4a/setup-sh7786.c | 29 +++++++++--------------------
 3 files changed, 18 insertions(+), 24 deletions(-)

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 09c0aef31d03..eff0f2352c93 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -71,6 +71,9 @@ config GENERIC_HARDIRQS_NO__DO_IRQ
 config GENERIC_IRQ_PROBE
 	def_bool y
 
+config IRQ_PER_CPU
+	def_bool y
+
 config GENERIC_GPIO
 	def_bool n
 
@@ -414,6 +417,8 @@ config CPU_SUBTYPE_SH7786
 	select CPU_HAS_PTEAEX
 	select ARCH_SPARSEMEM_ENABLE
 	select SYS_SUPPORTS_NUMA
+	select SYS_SUPPORTS_SMP
+	select GENERIC_CLOCKEVENTS_BROADCAST if SMP
 
 config CPU_SUBTYPE_SHX3
 	bool "Support SH-X3 processor"
@@ -651,7 +656,7 @@ config NR_CPUS
 	int "Maximum number of CPUs (2-32)"
 	range 2 32
 	depends on SMP
-	default "4" if CPU_SHX3
+	default "4" if CPU_SUBTYPE_SHX3
 	default "2"
 	help
 	  This allows you to specify the maximum number of CPUs which this
diff --git a/arch/sh/kernel/cpu/sh4a/Makefile b/arch/sh/kernel/cpu/sh4a/Makefile
index 96ea09ca8cc1..ebdd391d5f42 100644
--- a/arch/sh/kernel/cpu/sh4a/Makefile
+++ b/arch/sh/kernel/cpu/sh4a/Makefile
@@ -16,7 +16,7 @@ obj-$(CONFIG_CPU_SUBTYPE_SH7366)	+= setup-sh7366.o
 obj-$(CONFIG_CPU_SUBTYPE_SHX3)		+= setup-shx3.o
 
 # SMP setup
-smp-$(CONFIG_CPU_SUBTYPE_SHX3)		:= smp-shx3.o
+smp-$(CONFIG_CPU_SHX3)			:= smp-shx3.o
 
 # Primary on-chip clocks (common)
 clock-$(CONFIG_CPU_SUBTYPE_SH7763)	:= clock-sh7763.o
@@ -38,6 +38,6 @@ pinmux-$(CONFIG_CPU_SUBTYPE_SH7724)	:= pinmux-sh7724.o
 pinmux-$(CONFIG_CPU_SUBTYPE_SH7785)	:= pinmux-sh7785.o
 pinmux-$(CONFIG_CPU_SUBTYPE_SH7786)	:= pinmux-sh7786.o
 
-obj-y			+= $(clock-y)
-obj-$(CONFIG_SMP)	+= $(smp-y)
+obj-y				+= $(clock-y)
+obj-$(CONFIG_SMP)		+= $(smp-y)
 obj-$(CONFIG_GENERIC_GPIO)	+= $(pinmux-y)
diff --git a/arch/sh/kernel/cpu/sh4a/setup-sh7786.c b/arch/sh/kernel/cpu/sh4a/setup-sh7786.c
index 93e0d2c017e8..b70049470a0b 100644
--- a/arch/sh/kernel/cpu/sh4a/setup-sh7786.c
+++ b/arch/sh/kernel/cpu/sh4a/setup-sh7786.c
@@ -595,9 +595,8 @@ enum {
 	HSPI,
 	GPIO0, GPIO1,
 	Thermal,
-	INTC0, INTC1, INTC2, INTC3, INTC4, INTC5, INTC6, INTC7,
-
-	/* interrupt groups */
+	INTICI0, INTICI1, INTICI2, INTICI3,
+	INTICI4, INTICI5, INTICI6, INTICI7,
 };
 
 static struct intc_vect vectors[] __initdata = {
@@ -638,10 +637,12 @@ static struct intc_vect vectors[] __initdata = {
 	INTC_VECT(HSPI, 0xe80),
 	INTC_VECT(GPIO0, 0xea0), INTC_VECT(GPIO1, 0xec0),
 	INTC_VECT(Thermal, 0xee0),
+	INTC_VECT(INTICI0, 0xf00), INTC_VECT(INTICI1, 0xf20),
+	INTC_VECT(INTICI2, 0xf40), INTC_VECT(INTICI3, 0xf60),
+	INTC_VECT(INTICI4, 0xf80), INTC_VECT(INTICI5, 0xfa0),
+	INTC_VECT(INTICI6, 0xfc0), INTC_VECT(INTICI7, 0xfe0),
 };
 
-/* FIXME: Main CPU support only now */
-#if 1 /* Main CPU */
 #define CnINTMSK0	0xfe410030
 #define CnINTMSK1	0xfe410040
 #define CnINTMSKCLR0	0xfe410050
@@ -654,21 +655,6 @@ static struct intc_vect vectors[] __initdata = {
 #define CnINT2MSKCR1	0xfe410a34
 #define CnINT2MSKCR2	0xfe410a38
 #define CnINT2MSKCR3	0xfe410a3c
-#else /* Sub CPU */
-#define CnINTMSK0	0xfe410034
-#define CnINTMSK1	0xfe410044
-#define CnINTMSKCLR0	0xfe410054
-#define CnINTMSKCLR1	0xfe410064
-#define CnINT2MSKR0	0xfe410b20
-#define CnINT2MSKR1	0xfe410b24
-#define CnINT2MSKR2	0xfe410b28
-#define CnINT2MSKR3	0xfe410b2c
-#define CnINT2MSKCR0	0xfe410b30
-#define CnINT2MSKCR1	0xfe410b34
-#define CnINT2MSKCR2	0xfe410b38
-#define CnINT2MSKCR3	0xfe410b3c
-#endif
-
 #define INTMSK2		0xfe410068
 #define INTMSKCLR2	0xfe41006c
 
@@ -753,6 +739,9 @@ static struct intc_prio_reg prio_registers[] __initdata = {
 						  GPIO1, Thermal } },
 	{ 0xfe41085c, 0, 32, 8, /* INT2PRI23 */ { 0, 0, 0, 0 } },
 	{ 0xfe410860, 0, 32, 8, /* INT2PRI24 */ { 0, 0, 0, 0 } },
+	{ 0xfe410090, 0xfe4100a0, 32, 4, /* CnICIPRI / CnICIPRICLR */
+	  { INTICI7, INTICI6, INTICI5, INTICI4,
+	    INTICI3, INTICI2, INTICI1, INTICI0 }, INTC_SMP(4, 2) },
 };
 
 static DECLARE_INTC_DESC(intc_desc, "sh7786", vectors, NULL,
-- 
cgit v1.2.3-59-g8ed1b


From 854c879f5abf309ebd378bea1ee41acf4ddf7194 Mon Sep 17 00:00:00 2001
From: Pekka J Enberg <penberg@cs.helsinki.fi>
Date: Mon, 22 Jun 2009 17:39:41 +0300
Subject: x86: Move init_gbpages() to setup_arch()

The init_gbpages() function is conditionally called from
init_memory_mapping() function. There are two call-sites where
this 'after_bootmem' condition can be true: setup_arch() and
mem_init() via pci_iommu_alloc().

Therefore, it's safe to move the call to init_gbpages() to
setup_arch() as it's always called before mem_init().

This removes an after_bootmem use - paving the way to remove
all uses of that state variable.

Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Acked-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <Pine.LNX.4.64.0906221731210.19474@melkki.cs.Helsinki.FI>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/setup.c | 16 ++++++++++++++++
 arch/x86/mm/init.c      | 17 -----------------
 2 files changed, 16 insertions(+), 17 deletions(-)

diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index be5ae80f897f..de2cab132844 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -289,6 +289,20 @@ void * __init extend_brk(size_t size, size_t align)
 	return ret;
 }
 
+#ifdef CONFIG_X86_64
+static void __init init_gbpages(void)
+{
+	if (direct_gbpages && cpu_has_gbpages)
+		printk(KERN_INFO "Using GB pages for direct mapping\n");
+	else
+		direct_gbpages = 0;
+}
+#else
+static inline void init_gbpages(void)
+{
+}
+#endif
+
 static void __init reserve_brk(void)
 {
 	if (_brk_end > _brk_start)
@@ -871,6 +885,8 @@ void __init setup_arch(char **cmdline_p)
 
 	reserve_brk();
 
+	init_gbpages();
+
 	/* max_pfn_mapped is updated here */
 	max_low_pfn_mapped = init_memory_mapping(0, max_low_pfn<<PAGE_SHIFT);
 	max_pfn_mapped = max_low_pfn_mapped;
diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c
index f53b57e4086f..47ce9a2ce5e7 100644
--- a/arch/x86/mm/init.c
+++ b/arch/x86/mm/init.c
@@ -177,20 +177,6 @@ static int __meminit save_mr(struct map_range *mr, int nr_range,
 	return nr_range;
 }
 
-#ifdef CONFIG_X86_64
-static void __init init_gbpages(void)
-{
-	if (direct_gbpages && cpu_has_gbpages)
-		printk(KERN_INFO "Using GB pages for direct mapping\n");
-	else
-		direct_gbpages = 0;
-}
-#else
-static inline void init_gbpages(void)
-{
-}
-#endif
-
 /*
  * Setup the direct mapping of the physical memory at PAGE_OFFSET.
  * This runs before bootmem is initialized and gets pages directly from
@@ -210,9 +196,6 @@ unsigned long __init_refok init_memory_mapping(unsigned long start,
 
 	printk(KERN_INFO "init_memory_mapping: %016lx-%016lx\n", start, end);
 
-	if (!after_bootmem)
-		init_gbpages();
-
 #if defined(CONFIG_DEBUG_PAGEALLOC) || defined(CONFIG_KMEMCHECK)
 	/*
 	 * For CONFIG_DEBUG_PAGEALLOC, identity mapping will use small pages.
-- 
cgit v1.2.3-59-g8ed1b


From 971121f27179508beb18b86b84e1f8399d7b3875 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 23 Jun 2009 17:36:23 +0900
Subject: sh: Fix up more dma-mapping fallout.

commit dbe6f1869188b6e04e38aa861dd198befb08bcd7
("dma-mapping: mark dma_sync_single and dma_sync_sg as deprecated"
conveniently broke every single SH build.

In the future it would be great if people could at least bother
figuring out how to use grep.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/include/asm/dma-mapping.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/arch/sh/include/asm/dma-mapping.h b/arch/sh/include/asm/dma-mapping.h
index ea9d4f41c9d2..69d56dd4c968 100644
--- a/arch/sh/include/asm/dma-mapping.h
+++ b/arch/sh/include/asm/dma-mapping.h
@@ -97,7 +97,7 @@ static inline void dma_unmap_page(struct device *dev, dma_addr_t dma_address,
 	dma_unmap_single(dev, dma_address, size, dir);
 }
 
-static inline void dma_sync_single(struct device *dev, dma_addr_t dma_handle,
+static inline void __dma_sync_single(struct device *dev, dma_addr_t dma_handle,
 				   size_t size, enum dma_data_direction dir)
 {
 #if defined(CONFIG_PCI) && !defined(CONFIG_SH_PCIDMA_NONCOHERENT)
@@ -119,7 +119,7 @@ static inline void dma_sync_single_range(struct device *dev,
 	dma_cache_sync(dev, phys_to_virt(dma_handle) + offset, size, dir);
 }
 
-static inline void dma_sync_sg(struct device *dev, struct scatterlist *sg,
+static inline void __dma_sync_sg(struct device *dev, struct scatterlist *sg,
 			       int nelems, enum dma_data_direction dir)
 {
 	int i;
@@ -137,7 +137,7 @@ static inline void dma_sync_single_for_cpu(struct device *dev,
 					   dma_addr_t dma_handle, size_t size,
 					   enum dma_data_direction dir)
 {
-	dma_sync_single(dev, dma_handle, size, dir);
+	__dma_sync_single(dev, dma_handle, size, dir);
 	debug_dma_sync_single_for_cpu(dev, dma_handle, size, dir);
 }
 
@@ -146,7 +146,7 @@ static inline void dma_sync_single_for_device(struct device *dev,
 					      size_t size,
 					      enum dma_data_direction dir)
 {
-	dma_sync_single(dev, dma_handle, size, dir);
+	__dma_sync_single(dev, dma_handle, size, dir);
 	debug_dma_sync_single_for_device(dev, dma_handle, size, dir);
 }
 
@@ -177,7 +177,7 @@ static inline void dma_sync_sg_for_cpu(struct device *dev,
 				       struct scatterlist *sg, int nelems,
 				       enum dma_data_direction dir)
 {
-	dma_sync_sg(dev, sg, nelems, dir);
+	__dma_sync_sg(dev, sg, nelems, dir);
 	debug_dma_sync_sg_for_cpu(dev, sg, nelems, dir);
 }
 
@@ -185,7 +185,7 @@ static inline void dma_sync_sg_for_device(struct device *dev,
 					  struct scatterlist *sg, int nelems,
 					  enum dma_data_direction dir)
 {
-	dma_sync_sg(dev, sg, nelems, dir);
+	__dma_sync_sg(dev, sg, nelems, dir);
 	debug_dma_sync_sg_for_device(dev, sg, nelems, dir);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From bfdb4d9f0f611687d71cf6a460efc9e755f4a462 Mon Sep 17 00:00:00 2001
From: Arun R Bharadwaj <arun@linux.vnet.ibm.com>
Date: Tue, 23 Jun 2009 10:00:58 +0530
Subject: timers: Fix timer_migration interface which accepts any number as
 input

Poornima Nayek reported:

| Timer migration interface /proc/sys/kernel/timer_migration in
| 2.6.30-git9 accepts any numerical value as input.
|
| Steps to reproduce:
| 1. echo -6666666 > /proc/sys/kernel/timer_migration
| 2. cat /proc/sys/kernel/timer_migration
| -6666666
|
| 1. echo 44444444444444444444444444444444444444444444444444444444444 > /proc/sys/kernel/timer_migration
| 2. cat /proc/sys/kernel/timer_migration
| -1357789412
|
| Expected behavior: Should 'echo: write error: Invalid argument' while
| setting any value other then 0 & 1

Restrict valid values to 0 and 1.

Reported-by: Poornima Nayak <mpnayak@linux.vnet.ibm.com>
Tested-by: Poornima Nayak <mpnayak@linux.vnet.ibm.com>
Signed-off-by: Arun R Bharadwaj <arun@linux.vnet.ibm.com>
Cc: poornima nayak <mpnayak@linux.vnet.ibm.com>
Cc: Arun Bharadwaj <arun@linux.vnet.ibm.com>
LKML-Reference: <20090623043058.GA3249@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sysctl.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 62e4ff9968b5..c428ba161db1 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -335,7 +335,10 @@ static struct ctl_table kern_table[] = {
 		.data		= &sysctl_timer_migration,
 		.maxlen		= sizeof(unsigned int),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec,
+		.proc_handler	= &proc_dointvec_minmax,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &zero,
+		.extra2		= &one,
 	},
 #endif
 	{
-- 
cgit v1.2.3-59-g8ed1b


From 3d906ef10a539ff336010afab8f6f9c4fe379695 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Tue, 23 Jun 2009 11:23:07 +0200
Subject: perf_counter tools: Handle overlapping MMAP events

Martin Schwidefsky reported "perf report" symbol resolution
problems on S390.

Since we only report MMAP, not MUNMAP, we have to deal with
overlapping maps.

We used to simply throw out the old map on the assumption whole
maps got unmapped. This obviously doesn't deal with partial
unmaps. However it appears some dynamic linkers do fancy
partial unmaps (s390), so do something more elaborate and
truncate the old maps, only removing them when they've been
fully covered.

This resolves (part of) the S390 symbol resolution problems.

Reported-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Tested-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-report.c | 24 +++++++++++++++++++++---
 1 file changed, 21 insertions(+), 3 deletions(-)

diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index ec230a0146e9..b4e76f75ba87 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -400,9 +400,27 @@ static void thread__insert_map(struct thread *self, struct map *map)
 
 	list_for_each_entry_safe(pos, tmp, &self->maps, node) {
 		if (map__overlap(pos, map)) {
-			list_del_init(&pos->node);
-			/* XXX leaks dsos */
-			free(pos);
+			if (verbose >= 2) {
+				printf("overlapping maps:\n");
+				map__fprintf(map, stdout);
+				map__fprintf(pos, stdout);
+			}
+
+			if (map->start <= pos->start && map->end > pos->start)
+				pos->start = map->end;
+
+			if (map->end >= pos->end && map->start < pos->end)
+				pos->end = map->start;
+
+			if (verbose >= 2) {
+				printf("after collision:\n");
+				map__fprintf(pos, stdout);
+			}
+
+			if (pos->start >= pos->end) {
+				list_del_init(&pos->node);
+				free(pos);
+			}
 		}
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From f29ac756a40d0f1bb07d682ea521e7b666ff06d5 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 19 Jun 2009 18:27:26 +0200
Subject: perf_counter: Optimize perf_swcounter_event()

Similar to tracepoints, use an enable variable to reduce
overhead when unused.

Only look for a counter of a particular event type when we know
there is at least one in the system.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_counter.h | 11 ++++++++++-
 kernel/perf_counter.c        | 18 +++++++++++++++---
 2 files changed, 25 insertions(+), 4 deletions(-)

diff --git a/include/linux/perf_counter.h b/include/linux/perf_counter.h
index 89698d8aba5c..e7213e46cf9c 100644
--- a/include/linux/perf_counter.h
+++ b/include/linux/perf_counter.h
@@ -669,7 +669,16 @@ static inline int is_software_counter(struct perf_counter *counter)
 		(counter->attr.type != PERF_TYPE_HW_CACHE);
 }
 
-extern void perf_swcounter_event(u32, u64, int, struct pt_regs *, u64);
+extern atomic_t perf_swcounter_enabled[PERF_COUNT_SW_MAX];
+
+extern void __perf_swcounter_event(u32, u64, int, struct pt_regs *, u64);
+
+static inline void
+perf_swcounter_event(u32 event, u64 nr, int nmi, struct pt_regs *regs, u64 addr)
+{
+	if (atomic_read(&perf_swcounter_enabled[event]))
+		__perf_swcounter_event(event, nr, nmi, regs, addr);
+}
 
 extern void __perf_counter_mmap(struct vm_area_struct *vma);
 
diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index 1a933a221ea4..7515c7695428 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -3317,8 +3317,8 @@ out:
 	put_cpu_var(perf_cpu_context);
 }
 
-void
-perf_swcounter_event(u32 event, u64 nr, int nmi, struct pt_regs *regs, u64 addr)
+void __perf_swcounter_event(u32 event, u64 nr, int nmi,
+			    struct pt_regs *regs, u64 addr)
 {
 	struct perf_sample_data data = {
 		.regs = regs,
@@ -3509,9 +3509,19 @@ static const struct pmu *tp_perf_counter_init(struct perf_counter *counter)
 }
 #endif
 
+atomic_t perf_swcounter_enabled[PERF_COUNT_SW_MAX];
+
+static void sw_perf_counter_destroy(struct perf_counter *counter)
+{
+	u64 event = counter->attr.config;
+
+	atomic_dec(&perf_swcounter_enabled[event]);
+}
+
 static const struct pmu *sw_perf_counter_init(struct perf_counter *counter)
 {
 	const struct pmu *pmu = NULL;
+	u64 event = counter->attr.config;
 
 	/*
 	 * Software counters (currently) can't in general distinguish
@@ -3520,7 +3530,7 @@ static const struct pmu *sw_perf_counter_init(struct perf_counter *counter)
 	 * to be kernel events, and page faults are never hypervisor
 	 * events.
 	 */
-	switch (counter->attr.config) {
+	switch (event) {
 	case PERF_COUNT_SW_CPU_CLOCK:
 		pmu = &perf_ops_cpu_clock;
 
@@ -3541,6 +3551,8 @@ static const struct pmu *sw_perf_counter_init(struct perf_counter *counter)
 	case PERF_COUNT_SW_PAGE_FAULTS_MAJ:
 	case PERF_COUNT_SW_CONTEXT_SWITCHES:
 	case PERF_COUNT_SW_CPU_MIGRATIONS:
+		atomic_inc(&perf_swcounter_enabled[event]);
+		counter->destroy = sw_perf_counter_destroy;
 		pmu = &perf_ops_generic;
 		break;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From b84fbc9fb1d943e2c5f4efe52ed0e3c93a4bdb6a Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 22 Jun 2009 13:57:40 +0200
Subject: perf_counter: Push inherit into perf_counter_alloc()

Teach perf_counter_alloc() about inheritance so that we can
optimize the inherit path in the next patch.

Remove the child_counter->atrr.inherit = 1 line because the
only way to get there is if parent_counter->attr.inherit == 1
and we copy the attrs.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/perf_counter.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index 7515c7695428..0a45490f4029 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -3568,6 +3568,7 @@ perf_counter_alloc(struct perf_counter_attr *attr,
 		   int cpu,
 		   struct perf_counter_context *ctx,
 		   struct perf_counter *group_leader,
+		   struct perf_counter *parent_counter,
 		   gfp_t gfpflags)
 {
 	const struct pmu *pmu;
@@ -3603,6 +3604,8 @@ perf_counter_alloc(struct perf_counter_attr *attr,
 	counter->ctx		= ctx;
 	counter->oncpu		= -1;
 
+	counter->parent		= parent_counter;
+
 	counter->ns		= get_pid_ns(current->nsproxy->pid_ns);
 	counter->id		= atomic64_inc_return(&perf_counter_id);
 
@@ -3827,7 +3830,7 @@ SYSCALL_DEFINE5(perf_counter_open,
 	}
 
 	counter = perf_counter_alloc(&attr, cpu, ctx, group_leader,
-				     GFP_KERNEL);
+				     NULL, GFP_KERNEL);
 	ret = PTR_ERR(counter);
 	if (IS_ERR(counter))
 		goto err_put_context;
@@ -3893,7 +3896,8 @@ inherit_counter(struct perf_counter *parent_counter,
 
 	child_counter = perf_counter_alloc(&parent_counter->attr,
 					   parent_counter->cpu, child_ctx,
-					   group_leader, GFP_KERNEL);
+					   group_leader, parent_counter,
+					   GFP_KERNEL);
 	if (IS_ERR(child_counter))
 		return child_counter;
 	get_ctx(child_ctx);
@@ -3916,12 +3920,6 @@ inherit_counter(struct perf_counter *parent_counter,
 	 */
 	add_counter_to_ctx(child_counter, child_ctx);
 
-	child_counter->parent = parent_counter;
-	/*
-	 * inherit into child's child as well:
-	 */
-	child_counter->attr.inherit = 1;
-
 	/*
 	 * Get a reference to the parent filp - we will fput it
 	 * when the child counter exits. This is safe to do because
-- 
cgit v1.2.3-59-g8ed1b


From f344011ccb85469445369153c3d27c4ee4bc2ac8 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 22 Jun 2009 13:58:35 +0200
Subject: perf_counter: Optimize perf_counter_alloc()'s inherit case

We don't need to add usage counts for swcounter and attr usage
models for inherited counters since the parent counter will
always have one, which suffices to generate the needed output.

This avoids up to 3 global atomic increments per inherited
counter.

LKML-Reference: <new-submission>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/perf_counter.c | 32 ++++++++++++++++++++------------
 1 file changed, 20 insertions(+), 12 deletions(-)

diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index 0a45490f4029..c2b19c111718 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -1508,11 +1508,13 @@ static void free_counter(struct perf_counter *counter)
 {
 	perf_pending_sync(counter);
 
-	atomic_dec(&nr_counters);
-	if (counter->attr.mmap)
-		atomic_dec(&nr_mmap_counters);
-	if (counter->attr.comm)
-		atomic_dec(&nr_comm_counters);
+	if (!counter->parent) {
+		atomic_dec(&nr_counters);
+		if (counter->attr.mmap)
+			atomic_dec(&nr_mmap_counters);
+		if (counter->attr.comm)
+			atomic_dec(&nr_comm_counters);
+	}
 
 	if (counter->destroy)
 		counter->destroy(counter);
@@ -3515,6 +3517,8 @@ static void sw_perf_counter_destroy(struct perf_counter *counter)
 {
 	u64 event = counter->attr.config;
 
+	WARN_ON(counter->parent);
+
 	atomic_dec(&perf_swcounter_enabled[event]);
 }
 
@@ -3551,8 +3555,10 @@ static const struct pmu *sw_perf_counter_init(struct perf_counter *counter)
 	case PERF_COUNT_SW_PAGE_FAULTS_MAJ:
 	case PERF_COUNT_SW_CONTEXT_SWITCHES:
 	case PERF_COUNT_SW_CPU_MIGRATIONS:
-		atomic_inc(&perf_swcounter_enabled[event]);
-		counter->destroy = sw_perf_counter_destroy;
+		if (!counter->parent) {
+			atomic_inc(&perf_swcounter_enabled[event]);
+			counter->destroy = sw_perf_counter_destroy;
+		}
 		pmu = &perf_ops_generic;
 		break;
 	}
@@ -3663,11 +3669,13 @@ done:
 
 	counter->pmu = pmu;
 
-	atomic_inc(&nr_counters);
-	if (counter->attr.mmap)
-		atomic_inc(&nr_mmap_counters);
-	if (counter->attr.comm)
-		atomic_inc(&nr_comm_counters);
+	if (!counter->parent) {
+		atomic_inc(&nr_counters);
+		if (counter->attr.mmap)
+			atomic_inc(&nr_mmap_counters);
+		if (counter->attr.comm)
+			atomic_inc(&nr_comm_counters);
+	}
 
 	return counter;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 0c405b3346ea08098a82a1ee82912b018dfa9f96 Mon Sep 17 00:00:00 2001
From: Jonathan McDowell <noodles@earth.li>
Date: Tue, 23 Jun 2009 13:30:21 +0300
Subject: OMAP1: Fix compilation of arch/arm/mach-omap1/mailbox.c

This fixes the positioning of " in MODULE_AUTHOR, which is currently
causing a build failure on latest git with CONFIG_OMAP_MBOX_FWK=m; the
original breakage appears to date from the end of last year in
a5abbbe52b7e89a7633319c5417bd4331f7ac8ed

Signed-Off-By: Jonathan McDowell <noodles@earth.li>
Acked-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap1/mailbox.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/mach-omap1/mailbox.c b/arch/arm/mach-omap1/mailbox.c
index 0af4d6c85b47..6810b4aeb02c 100644
--- a/arch/arm/mach-omap1/mailbox.c
+++ b/arch/arm/mach-omap1/mailbox.c
@@ -203,5 +203,5 @@ module_exit(omap1_mbox_exit);
 
 MODULE_LICENSE("GPL v2");
 MODULE_DESCRIPTION("omap mailbox: omap1 architecture specific functions");
-MODULE_AUTHOR("Hiroshi DOYU" <Hiroshi.DOYU@nokia.com>);
+MODULE_AUTHOR("Hiroshi DOYU <Hiroshi.DOYU@nokia.com>");
 MODULE_ALIAS("platform:omap1-mailbox");
-- 
cgit v1.2.3-59-g8ed1b


From e4d24ec39c98cdc9cd97c26fdd426bbab0034fbe Mon Sep 17 00:00:00 2001
From: Andrew de Quincey <adq_dvb@lidskialf.net>
Date: Tue, 23 Jun 2009 13:30:21 +0300
Subject: OMAP1: Fix N770 MMC support

Some of the N770's MMC configuration options seem to have been
dropped. This patch adds them back in again.

Note that only the .ocr_mask change was /critical/, but I've added the
.max_freq setting back as well, as the original sources had it. Can
anyone confirm if this is unnecessary?

Secondly, there is support in the original code for a 4wire/higher
speed mode. As I don't have the requisite N770 hardware (I think it
was a rev2 N770?) to test this, I can't really add it back.

Signed-off-by: Andrew de Quincey <adq@lidskialf.net>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap1/board-nokia770.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/arm/mach-omap1/board-nokia770.c b/arch/arm/mach-omap1/board-nokia770.c
index e70fc7c66bbb..d9ebba01ffce 100644
--- a/arch/arm/mach-omap1/board-nokia770.c
+++ b/arch/arm/mach-omap1/board-nokia770.c
@@ -205,9 +205,11 @@ static int nokia770_mmc_get_cover_state(struct device *dev, int slot)
 static struct omap_mmc_platform_data nokia770_mmc2_data = {
 	.nr_slots                       = 1,
 	.dma_mask			= 0xffffffff,
+	.max_freq                       = 12000000,
 	.slots[0]       = {
 		.set_power		= nokia770_mmc_set_power,
 		.get_cover_state	= nokia770_mmc_get_cover_state,
+		.ocr_mask               = MMC_VDD_32_33|MMC_VDD_33_34,
 		.name                   = "mmcblk",
 	},
 };
-- 
cgit v1.2.3-59-g8ed1b


From d376f89701b0aa5b45d25fbfbeb1a0040399ad30 Mon Sep 17 00:00:00 2001
From: Huang Weiyi <weiyi.huang@gmail.com>
Date: Tue, 23 Jun 2009 13:30:22 +0300
Subject: OMAP1: remove duplicated #include

Remove duplicated #include in arch/arm/mach-omap1/board-nokia770.c.

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap1/board-nokia770.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/arm/mach-omap1/board-nokia770.c b/arch/arm/mach-omap1/board-nokia770.c
index d9ebba01ffce..ed2a48a9ce74 100644
--- a/arch/arm/mach-omap1/board-nokia770.c
+++ b/arch/arm/mach-omap1/board-nokia770.c
@@ -36,7 +36,6 @@
 #include <mach/hwa742.h>
 #include <mach/lcd_mipid.h>
 #include <mach/mmc.h>
-#include <mach/usb.h>
 #include <mach/clock.h>
 
 #define ADS7846_PENDOWN_GPIO	15
-- 
cgit v1.2.3-59-g8ed1b


From 762ad3a476baa1831f732488e80960f4aa024393 Mon Sep 17 00:00:00 2001
From: Grazvydas Ignotas <notasas@gmail.com>
Date: Tue, 23 Jun 2009 13:30:22 +0300
Subject: OMAP2/3: mmc-twl4030: use correct controller in twl_mmc23_set_power

twl_mmc23_set_power() has MMC2 twl_mmc_controller hardcoded in it, which
breaks MMC3. Find the right controller to use instead.

Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap2/mmc-twl4030.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/arch/arm/mach-omap2/mmc-twl4030.c b/arch/arm/mach-omap2/mmc-twl4030.c
index 9756a878fd90..1541fd4c8d0f 100644
--- a/arch/arm/mach-omap2/mmc-twl4030.c
+++ b/arch/arm/mach-omap2/mmc-twl4030.c
@@ -263,8 +263,19 @@ static int twl_mmc1_set_power(struct device *dev, int slot, int power_on,
 static int twl_mmc23_set_power(struct device *dev, int slot, int power_on, int vdd)
 {
 	int ret = 0;
-	struct twl_mmc_controller *c = &hsmmc[1];
+	struct twl_mmc_controller *c = NULL;
 	struct omap_mmc_platform_data *mmc = dev->platform_data;
+	int i;
+
+	for (i = 1; i < ARRAY_SIZE(hsmmc); i++) {
+		if (mmc == hsmmc[i].mmc) {
+			c = &hsmmc[i];
+			break;
+		}
+	}
+
+	if (c == NULL)
+		return -ENODEV;
 
 	/* If we don't see a Vcc regulator, assume it's a fixed
 	 * voltage always-on regulator.
-- 
cgit v1.2.3-59-g8ed1b


From 091a58af0ba1765d80b1e74382c7572baceb1bdc Mon Sep 17 00:00:00 2001
From: Roel Kluin <roel.kluin@gmail.com>
Date: Tue, 23 Jun 2009 13:30:22 +0300
Subject: OMAP2/3: omap mailbox: platform_get_irq() error ignored

platform_get_irq may return -ENXIO. but struct omap_mbox mbox_dsp_info.irq
is unsigned, so the error was not noticed.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap2/mailbox.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/arm/mach-omap2/mailbox.c b/arch/arm/mach-omap2/mailbox.c
index fd5b8a5925cc..6f71f3730c97 100644
--- a/arch/arm/mach-omap2/mailbox.c
+++ b/arch/arm/mach-omap2/mailbox.c
@@ -282,12 +282,12 @@ static int __devinit omap2_mbox_probe(struct platform_device *pdev)
 		return -ENOMEM;
 
 	/* DSP or IVA2 IRQ */
-	mbox_dsp_info.irq = platform_get_irq(pdev, 0);
-	if (mbox_dsp_info.irq < 0) {
+	ret = platform_get_irq(pdev, 0);
+	if (ret < 0) {
 		dev_err(&pdev->dev, "invalid irq resource\n");
-		ret = -ENODEV;
 		goto err_dsp;
 	}
+	mbox_dsp_info.irq = ret;
 
 	ret = omap_mbox_register(&pdev->dev, &mbox_dsp_info);
 	if (ret)
-- 
cgit v1.2.3-59-g8ed1b


From 8e25ad964aac0bf6b30dd013303750089f819679 Mon Sep 17 00:00:00 2001
From: Kevin Hilman <khilman@deeprootsystems.com>
Date: Tue, 23 Jun 2009 13:30:23 +0300
Subject: OMAP2/3: Add omap_type() for determining GP/EMU/HS

The omap_type() function is added and returns the DEVICETYPE field of
the CONTROL_STATUS register.  The result can be used for conditional
code based on whether device is GP (general purpose), EMU or
HS (high security). Also move the type defines so omap1 code
compile does not require ifdefs for sections using these defines.

This code is needed for the following fix to set the SRAM
size correctly for HS omaps.  Also at least PM and watchdog
code will need this function.

Signed-off-by: Kevin Hilman <khilman@ti.deeprootsystems.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap2/id.c              | 22 ++++++++++++++++++++++
 arch/arm/plat-omap/include/mach/cpu.h | 22 +++++++++++-----------
 2 files changed, 33 insertions(+), 11 deletions(-)

diff --git a/arch/arm/mach-omap2/id.c b/arch/arm/mach-omap2/id.c
index 458990e20c60..a98201cc265c 100644
--- a/arch/arm/mach-omap2/id.c
+++ b/arch/arm/mach-omap2/id.c
@@ -48,6 +48,28 @@ int omap_chip_is(struct omap_chip_id oci)
 }
 EXPORT_SYMBOL(omap_chip_is);
 
+int omap_type(void)
+{
+	u32 val = 0;
+
+	if (cpu_is_omap24xx())
+		val = omap_ctrl_readl(OMAP24XX_CONTROL_STATUS);
+	else if (cpu_is_omap34xx())
+		val = omap_ctrl_readl(OMAP343X_CONTROL_STATUS);
+	else {
+		pr_err("Cannot detect omap type!\n");
+		goto out;
+	}
+
+	val &= OMAP2_DEVICETYPE_MASK;
+	val >>= 8;
+
+out:
+	return val;
+}
+EXPORT_SYMBOL(omap_type);
+
+
 /*----------------------------------------------------------------------------*/
 
 #define OMAP_TAP_IDCODE		0x0204
diff --git a/arch/arm/plat-omap/include/mach/cpu.h b/arch/arm/plat-omap/include/mach/cpu.h
index fc60c4ebcc28..285eaa3a8275 100644
--- a/arch/arm/plat-omap/include/mach/cpu.h
+++ b/arch/arm/plat-omap/include/mach/cpu.h
@@ -30,6 +30,17 @@
 #ifndef __ASM_ARCH_OMAP_CPU_H
 #define __ASM_ARCH_OMAP_CPU_H
 
+/*
+ * Omap device type i.e. EMU/HS/TST/GP/BAD
+ */
+#define OMAP2_DEVICE_TYPE_TEST		0
+#define OMAP2_DEVICE_TYPE_EMU		1
+#define OMAP2_DEVICE_TYPE_SEC		2
+#define OMAP2_DEVICE_TYPE_GP		3
+#define OMAP2_DEVICE_TYPE_BAD		4
+
+int omap_type(void);
+
 struct omap_chip_id {
 	u8 oc;
 	u8 type;
@@ -424,17 +435,6 @@ IS_OMAP_TYPE(3430, 0x3430)
 
 
 int omap_chip_is(struct omap_chip_id oci);
-int omap_type(void);
-
-/*
- * Macro to detect device type i.e. EMU/HS/TST/GP/BAD
- */
-#define OMAP2_DEVICE_TYPE_TEST		0
-#define OMAP2_DEVICE_TYPE_EMU		1
-#define OMAP2_DEVICE_TYPE_SEC		2
-#define OMAP2_DEVICE_TYPE_GP		3
-#define OMAP2_DEVICE_TYPE_BAD		4
-
 void omap2_check_revision(void);
 
 #endif    /* defined(CONFIG_ARCH_OMAP2) || defined(CONFIG_ARCH_OMAP3) */
-- 
cgit v1.2.3-59-g8ed1b


From 5b0acc59d1bc5c310dfd6976555664f9dcf4dacd Mon Sep 17 00:00:00 2001
From: Tero Kristo <tero.kristo@nokia.com>
Date: Tue, 23 Jun 2009 13:30:23 +0300
Subject: OMAP3: SRAM size fix for HS/EMU devices

SRAM size fix for HS/EMU devices

Signed-off-by: Tero Kristo <tero.kristo@nokia.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/plat-omap/sram.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/arch/arm/plat-omap/sram.c b/arch/arm/plat-omap/sram.c
index 65006df3f1b7..4ea73804d21e 100644
--- a/arch/arm/plat-omap/sram.c
+++ b/arch/arm/plat-omap/sram.c
@@ -133,7 +133,12 @@ void __init omap_detect_sram(void)
 			if (cpu_is_omap34xx()) {
 				omap_sram_base = OMAP3_SRAM_PUB_VA;
 				omap_sram_start = OMAP3_SRAM_PUB_PA;
-				omap_sram_size = 0x8000; /* 32K */
+				if ((omap_type() == OMAP2_DEVICE_TYPE_EMU) ||
+				    (omap_type() == OMAP2_DEVICE_TYPE_SEC)) {
+					omap_sram_size = 0x7000; /* 28K */
+				} else {
+					omap_sram_size = 0x8000; /* 32K */
+				}
 			} else {
 				omap_sram_base = OMAP2_SRAM_PUB_VA;
 				omap_sram_start = OMAP2_SRAM_PUB_PA;
-- 
cgit v1.2.3-59-g8ed1b


From aecedb94b366d6fb5e2a17ca18a5dc78e593198e Mon Sep 17 00:00:00 2001
From: Kalle Jokiniemi <ext-kalle.jokiniemi@nokia.com>
Date: Tue, 23 Jun 2009 13:30:24 +0300
Subject: OMAP3: DMA: Enable idlemodes for DMA OCP

This patch enables MStandby smart-idle mode, autoidle smartidle mode,
and the autoidle bit for DMA4_OCP_SYSCONFIG.

Signed-off-by: Kalle Jokiniemi <ext-kalle.jokiniemi@nokia.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Kevin Hilman <khilman@ti.deeprootsystems.com>
---
 arch/arm/plat-omap/dma.c              | 13 +++++++++++++
 arch/arm/plat-omap/include/mach/dma.h | 15 +++++++++++++++
 2 files changed, 28 insertions(+)

diff --git a/arch/arm/plat-omap/dma.c b/arch/arm/plat-omap/dma.c
index def14ec265b3..7677a4a1cef2 100644
--- a/arch/arm/plat-omap/dma.c
+++ b/arch/arm/plat-omap/dma.c
@@ -2457,6 +2457,19 @@ static int __init omap_init_dma(void)
 		setup_irq(irq, &omap24xx_dma_irq);
 	}
 
+	/* Enable smartidle idlemodes and autoidle */
+	if (cpu_is_omap34xx()) {
+		u32 v = dma_read(OCP_SYSCONFIG);
+		v &= ~(DMA_SYSCONFIG_MIDLEMODE_MASK |
+				DMA_SYSCONFIG_SIDLEMODE_MASK |
+				DMA_SYSCONFIG_AUTOIDLE);
+		v |= (DMA_SYSCONFIG_MIDLEMODE(DMA_IDLEMODE_SMARTIDLE) |
+			DMA_SYSCONFIG_SIDLEMODE(DMA_IDLEMODE_SMARTIDLE) |
+			DMA_SYSCONFIG_AUTOIDLE);
+		dma_write(v , OCP_SYSCONFIG);
+	}
+
+
 	/* FIXME: Update LCD DMA to work on 24xx */
 	if (cpu_class_is_omap1()) {
 		r = request_irq(INT_DMA_LCD, lcd_dma_irq_handler, 0,
diff --git a/arch/arm/plat-omap/include/mach/dma.h b/arch/arm/plat-omap/include/mach/dma.h
index 8c1eae88737e..7b939cc01962 100644
--- a/arch/arm/plat-omap/include/mach/dma.h
+++ b/arch/arm/plat-omap/include/mach/dma.h
@@ -389,6 +389,21 @@
 #define DMA_THREAD_FIFO_25		(0x02 << 14)
 #define DMA_THREAD_FIFO_50		(0x03 << 14)
 
+/* DMA4_OCP_SYSCONFIG bits */
+#define DMA_SYSCONFIG_MIDLEMODE_MASK		(3 << 12)
+#define DMA_SYSCONFIG_CLOCKACTIVITY_MASK	(3 << 8)
+#define DMA_SYSCONFIG_EMUFREE			(1 << 5)
+#define DMA_SYSCONFIG_SIDLEMODE_MASK		(3 << 3)
+#define DMA_SYSCONFIG_SOFTRESET			(1 << 2)
+#define DMA_SYSCONFIG_AUTOIDLE			(1 << 0)
+
+#define DMA_SYSCONFIG_MIDLEMODE(n)		((n) << 12)
+#define DMA_SYSCONFIG_SIDLEMODE(n)		((n) << 3)
+
+#define DMA_IDLEMODE_SMARTIDLE			0x2
+#define DMA_IDLEMODE_NO_IDLE			0x1
+#define DMA_IDLEMODE_FORCE_IDLE			0x0
+
 /* Chaining modes*/
 #ifndef CONFIG_ARCH_OMAP1
 #define OMAP_DMA_STATIC_CHAIN		0x1
-- 
cgit v1.2.3-59-g8ed1b


From 6d453e84b587f38e4197bb2c6a37296c4a80cbac Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Tue, 23 Jun 2009 13:30:24 +0300
Subject: OMAP2/3: gpmc-onenand: correct use of async timings

Use async timings when sync timings are not requested.

Also ensure that OneNAND is in async mode when async
timings are used.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap2/gpmc-onenand.c | 21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-omap2/gpmc-onenand.c b/arch/arm/mach-omap2/gpmc-onenand.c
index 2fd22f9c5f0e..54fec53a48e7 100644
--- a/arch/arm/mach-omap2/gpmc-onenand.c
+++ b/arch/arm/mach-omap2/gpmc-onenand.c
@@ -31,6 +31,8 @@ static struct platform_device gpmc_onenand_device = {
 static int omap2_onenand_set_async_mode(int cs, void __iomem *onenand_base)
 {
 	struct gpmc_timings t;
+	u32 reg;
+	int err;
 
 	const int t_cer = 15;
 	const int t_avdp = 12;
@@ -43,6 +45,11 @@ static int omap2_onenand_set_async_mode(int cs, void __iomem *onenand_base)
 	const int t_wpl = 40;
 	const int t_wph = 30;
 
+	/* Ensure sync read and sync write are disabled */
+	reg = readw(onenand_base + ONENAND_REG_SYS_CFG1);
+	reg &= ~ONENAND_SYS_CFG1_SYNC_READ & ~ONENAND_SYS_CFG1_SYNC_WRITE;
+	writew(reg, onenand_base + ONENAND_REG_SYS_CFG1);
+
 	memset(&t, 0, sizeof(t));
 	t.sync_clk = 0;
 	t.cs_on = 0;
@@ -74,7 +81,16 @@ static int omap2_onenand_set_async_mode(int cs, void __iomem *onenand_base)
 			  GPMC_CONFIG1_DEVICESIZE_16 |
 			  GPMC_CONFIG1_MUXADDDATA);
 
-	return gpmc_cs_set_timings(cs, &t);
+	err = gpmc_cs_set_timings(cs, &t);
+	if (err)
+		return err;
+
+	/* Ensure sync read and sync write are disabled */
+	reg = readw(onenand_base + ONENAND_REG_SYS_CFG1);
+	reg &= ~ONENAND_SYS_CFG1_SYNC_READ & ~ONENAND_SYS_CFG1_SYNC_WRITE;
+	writew(reg, onenand_base + ONENAND_REG_SYS_CFG1);
+
+	return 0;
 }
 
 static void set_onenand_cfg(void __iomem *onenand_base, int latency,
@@ -124,7 +140,8 @@ static int omap2_onenand_set_sync_mode(struct omap_onenand_platform_data *cfg,
 	} else if (cfg->flags & ONENAND_SYNC_READWRITE) {
 		sync_read = 1;
 		sync_write = 1;
-	}
+	} else
+		return omap2_onenand_set_async_mode(cs, onenand_base);
 
 	if (!freq) {
 		/* Very first call freq is not known */
-- 
cgit v1.2.3-59-g8ed1b


From c8e6488f7b56d82453fc7d526118e9f1c2df133a Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Tue, 23 Jun 2009 13:30:25 +0300
Subject: OMAP3: RX51: Use OneNAND sync read / write

Use OneNAND sync read / write

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/mach-omap2/board-rx51-peripherals.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm/mach-omap2/board-rx51-peripherals.c b/arch/arm/mach-omap2/board-rx51-peripherals.c
index da93b86234ed..9a0bf6744a05 100644
--- a/arch/arm/mach-omap2/board-rx51-peripherals.c
+++ b/arch/arm/mach-omap2/board-rx51-peripherals.c
@@ -362,6 +362,7 @@ static struct omap_onenand_platform_data board_onenand_data = {
 	.gpio_irq	= 65,
 	.parts		= onenand_partitions,
 	.nr_parts	= ARRAY_SIZE(onenand_partitions),
+	.flags		= ONENAND_SYNC_READWRITE,
 };
 
 static void __init board_onenand_init(void)
-- 
cgit v1.2.3-59-g8ed1b


From f48ef99ca14577f3ea0a48c0e05ed7f5f6d211e9 Mon Sep 17 00:00:00 2001
From: Fernando Guzman Lugo <x0095840@ti.com>
Date: Tue, 23 Jun 2009 13:30:25 +0300
Subject: OMAP: IOMMU: function flush_iotlb_page is not flushing correct entry

The function flush_iotlb_page is not loading the CAM register with
the correct entry to be flushed, so it is flushing other entry

Signed-off-by: Fernando Guzman Lugo <x0095840@ti.com>
Signed-off-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/plat-omap/iommu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/plat-omap/iommu.c b/arch/arm/plat-omap/iommu.c
index 4cf449fa2cb5..4a0301399013 100644
--- a/arch/arm/plat-omap/iommu.c
+++ b/arch/arm/plat-omap/iommu.c
@@ -298,7 +298,7 @@ void flush_iotlb_page(struct iommu *obj, u32 da)
 		if ((start <= da) && (da < start + bytes)) {
 			dev_dbg(obj->dev, "%s: %08x<=%08x(%x)\n",
 				__func__, start, da, bytes);
-
+			iotlb_load_cr(obj, &cr);
 			iommu_write_reg(obj, 1, MMU_FLUSH_ENTRY);
 		}
 	}
-- 
cgit v1.2.3-59-g8ed1b


From cb5793db5ecf108594d8006ae838e47996a76a19 Mon Sep 17 00:00:00 2001
From: janboe <janboe.ye@gmail.com>
Date: Tue, 23 Jun 2009 13:30:25 +0300
Subject: OMAP2/3: Initialize gpio debounce register

Some bootloader may initialize debounce register and this will make
dbclk not consist with the debounce register after linux kernel boot
up.

Signed-off-by: janboe <janboe.ye@gmail.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/plat-omap/gpio.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm/plat-omap/gpio.c b/arch/arm/plat-omap/gpio.c
index 7fd89ba8d3b5..26b387c12423 100644
--- a/arch/arm/plat-omap/gpio.c
+++ b/arch/arm/plat-omap/gpio.c
@@ -1585,6 +1585,7 @@ static int __init _omap_gpio_init(void)
 			__raw_writel(0x00000000, bank->base + OMAP24XX_GPIO_IRQENABLE1);
 			__raw_writel(0xffffffff, bank->base + OMAP24XX_GPIO_IRQSTATUS1);
 			__raw_writew(0x0015, bank->base + OMAP24XX_GPIO_SYSCONFIG);
+			__raw_writel(0x00000000, bank->base + OMAP24XX_GPIO_DEBOUNCE_EN);
 
 			/* Initialize interface clock ungated, module enabled */
 			__raw_writel(0, bank->base + OMAP24XX_GPIO_CTRL);
-- 
cgit v1.2.3-59-g8ed1b


From 01542cd1bbf995f951e2c2383d7911e96b12bec6 Mon Sep 17 00:00:00 2001
From: Dhananjay Phadke <dhananjay@netxen.com>
Date: Mon, 22 Jun 2009 20:26:20 +0000
Subject: netxen: fix build with without CONFIG_PM

wrap pci suspend() and resume() with CONFIG_PM check.

Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/netxen/netxen_nic_main.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/netxen/netxen_nic_main.c b/drivers/net/netxen/netxen_nic_main.c
index 71daa3d5f114..43ac333898bf 100644
--- a/drivers/net/netxen/netxen_nic_main.c
+++ b/drivers/net/netxen/netxen_nic_main.c
@@ -1178,6 +1178,7 @@ static void __devexit netxen_nic_remove(struct pci_dev *pdev)
 	free_netdev(netdev);
 }
 
+#ifdef CONFIG_PM
 static int
 netxen_nic_suspend(struct pci_dev *pdev, pm_message_t state)
 {
@@ -1242,6 +1243,7 @@ netxen_nic_resume(struct pci_dev *pdev)
 
 	return 0;
 }
+#endif
 
 static int netxen_nic_open(struct net_device *netdev)
 {
@@ -1771,8 +1773,10 @@ static struct pci_driver netxen_driver = {
 	.id_table = netxen_pci_tbl,
 	.probe = netxen_nic_probe,
 	.remove = __devexit_p(netxen_nic_remove),
+#ifdef CONFIG_PM
 	.suspend = netxen_nic_suspend,
 	.resume = netxen_nic_resume
+#endif
 };
 
 /* Driver Registration on NetXen card    */
-- 
cgit v1.2.3-59-g8ed1b


From 96f2ebd2e10417da151202c750d8664767a2194b Mon Sep 17 00:00:00 2001
From: Dhananjay Phadke <dhananjay@netxen.com>
Date: Mon, 22 Jun 2009 20:26:21 +0000
Subject: netxen: fix firmware init handshake

Make sure all functions run firmware init handshake.

If PCI function 0 fails to initialize firmware, mark the
state failed so that other functions on the same board
bail out quickly instead of waiting 30s for firmware
handshake.

Signed-off-by: Dhananjay Phadke <dhananjay@netxen.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/netxen/netxen_nic_init.c | 37 +++++++++++++++++++-----------------
 drivers/net/netxen/netxen_nic_main.c |  3 ++-
 2 files changed, 22 insertions(+), 18 deletions(-)

diff --git a/drivers/net/netxen/netxen_nic_init.c b/drivers/net/netxen/netxen_nic_init.c
index bdb143d2b5c7..055bb61d6e77 100644
--- a/drivers/net/netxen/netxen_nic_init.c
+++ b/drivers/net/netxen/netxen_nic_init.c
@@ -944,28 +944,31 @@ int netxen_phantom_init(struct netxen_adapter *adapter, int pegtune_val)
 	u32 val = 0;
 	int retries = 60;
 
-	if (!pegtune_val) {
-		do {
-			val = NXRD32(adapter, CRB_CMDPEG_STATE);
+	if (pegtune_val)
+		return 0;
 
-			if (val == PHAN_INITIALIZE_COMPLETE ||
-				val == PHAN_INITIALIZE_ACK)
-				return 0;
+	do {
+		val = NXRD32(adapter, CRB_CMDPEG_STATE);
 
-			msleep(500);
+		switch (val) {
+		case PHAN_INITIALIZE_COMPLETE:
+		case PHAN_INITIALIZE_ACK:
+			return 0;
+		case PHAN_INITIALIZE_FAILED:
+			goto out_err;
+		default:
+			break;
+		}
 
-		} while (--retries);
+		msleep(500);
 
-		if (!retries) {
-			pegtune_val = NXRD32(adapter,
-					NETXEN_ROMUSB_GLB_PEGTUNE_DONE);
-			printk(KERN_WARNING "netxen_phantom_init: init failed, "
-					"pegtune_val=%x\n", pegtune_val);
-			return -1;
-		}
-	}
+	} while (--retries);
 
-	return 0;
+	NXWR32(adapter, CRB_CMDPEG_STATE, PHAN_INITIALIZE_FAILED);
+
+out_err:
+	dev_warn(&adapter->pdev->dev, "firmware init failed\n");
+	return -EIO;
 }
 
 static int
diff --git a/drivers/net/netxen/netxen_nic_main.c b/drivers/net/netxen/netxen_nic_main.c
index 43ac333898bf..2919a2d12bf4 100644
--- a/drivers/net/netxen/netxen_nic_main.c
+++ b/drivers/net/netxen/netxen_nic_main.c
@@ -705,7 +705,7 @@ netxen_start_firmware(struct netxen_adapter *adapter, int request_fw)
 		first_driver = (adapter->ahw.pci_func == 0);
 
 	if (!first_driver)
-		return 0;
+		goto wait_init;
 
 	first_boot = NXRD32(adapter, NETXEN_CAM_RAM(0x1fc));
 
@@ -752,6 +752,7 @@ netxen_start_firmware(struct netxen_adapter *adapter, int request_fw)
 		| (_NETXEN_NIC_LINUX_SUBVERSION);
 	NXWR32(adapter, CRB_DRIVER_VERSION, val);
 
+wait_init:
 	/* Handshake with the card before we register the devices. */
 	err = netxen_phantom_init(adapter, NETXEN_NIC_PEG_TUNE);
 	if (err) {
-- 
cgit v1.2.3-59-g8ed1b


From fec37ab56f5b86b413f71258f36b181f57180d9c Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 22 Jun 2009 21:31:20 +0000
Subject: can: let SJA1000 driver depend on HAS_IOMEM

Fixes this compile error on s390:

drivers/net/can/sja1000/sja1000_platform.c: In function 'sp_read_reg':
drivers/net/can/sja1000/sja1000_platform.c:42: error: implicit declaration of function 'ioread8'
drivers/net/can/sja1000/sja1000_platform.c: In function 'sp_write_reg':
drivers/net/can/sja1000/sja1000_platform.c:47: error: implicit declaration of function 'iowrite8'
drivers/net/can/sja1000/sja1000_platform.c: In function 'sp_probe':
drivers/net/can/sja1000/sja1000_platform.c:79: error: implicit declaration of function 'ioremap_nocache'

Cc: Wolfgang Grandegger <wg@grandegger.com>
Cc: Oliver Hartkopp <oliver.hartkopp@volkswagen.de>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/can/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/can/Kconfig b/drivers/net/can/Kconfig
index d5e18812bf49..33821a81cbf8 100644
--- a/drivers/net/can/Kconfig
+++ b/drivers/net/can/Kconfig
@@ -36,7 +36,7 @@ config CAN_CALC_BITTIMING
 	  If unsure, say Y.
 
 config CAN_SJA1000
-	depends on CAN_DEV
+	depends on CAN_DEV && HAS_IOMEM
 	tristate "Philips SJA1000"
 	---help---
 	  Driver for the SJA1000 CAN controllers from Philips or NXP
-- 
cgit v1.2.3-59-g8ed1b


From 0cf08dcb78e8d61b6d4b2eb5cdb296d969971626 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 22 Jun 2009 21:32:18 +0000
Subject: net: let KS8842 driver depend on HAS_IOMEM

Fixes this compile error on s390:

  CC      drivers/net/ks8842.o
drivers/net/ks8842.c: In function 'ks8842_select_bank':
drivers/net/ks8842.c:124: error: implicit declaration of function 'iowrite16'
drivers/net/ks8842.c: In function 'ks8842_write8':
drivers/net/ks8842.c:131: error: implicit declaration of function 'iowrite8'

Cc: Richard Rojfors <richard.rojfors.ext@mocean-labs.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 01f282cd0989..dd145c1714a4 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -1725,6 +1725,7 @@ config TLAN
 
 config KS8842
 	tristate "Micrel KSZ8842"
+	depends on HAS_IOMEM
 	help
 	  This platform driver is for Micrel KSZ8842 chip.
 
-- 
cgit v1.2.3-59-g8ed1b


From 291e99a112f9bf34c027031de7ef8b94a2692937 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 4 May 2009 09:30:51 +0100
Subject: [ARM] S3C24XX: Fix use of CONFIG_S3C24XX_PWM

CONFIG_S3C24XX_PWM was defined in arch/arm/plat-s3c24xx/Kconfig but
not used anywhere else as the corresponding makefile used
CONFIG_HAVE_PWM (selected by CONFIG_S3C24XX_PWM) to compile the PWM
driver.

Change the makefile to use CONFIG_S3C24XX_PWM to compile this driver
to ensure it is only build when needed.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c24xx/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/plat-s3c24xx/Makefile b/arch/arm/plat-s3c24xx/Makefile
index 636cb12711df..579a165c2827 100644
--- a/arch/arm/plat-s3c24xx/Makefile
+++ b/arch/arm/plat-s3c24xx/Makefile
@@ -29,7 +29,7 @@ obj-$(CONFIG_PM_SIMTEC)		+= pm-simtec.o
 obj-$(CONFIG_PM)		+= pm.o
 obj-$(CONFIG_PM)		+= irq-pm.o
 obj-$(CONFIG_PM)		+= sleep.o
-obj-$(CONFIG_HAVE_PWM)		+= pwm.o
+obj-$(CONFIG_S3C24XX_PWM)	+= pwm.o
 obj-$(CONFIG_S3C2410_CLOCK)	+= s3c2410-clock.o
 obj-$(CONFIG_S3C2410_DMA)	+= dma.o
 obj-$(CONFIG_S3C24XX_ADC)	+= adc.o
-- 
cgit v1.2.3-59-g8ed1b


From a18327f35a879a6467c3e901da7f68944b191732 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 4 May 2009 09:53:12 +0100
Subject: [ARM] S3C: Remove unused CONFIG_DEBUG_S3C_PORT

Remove the unused CONFIG_DEBUG_S3C_PORT as we currently only have
support for using the S3C UARTs via the low-level debug code.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/Kconfig.debug                | 8 --------
 arch/arm/configs/s3c2410_defconfig    | 1 -
 arch/arm/configs/s3c6400_defconfig    | 1 -
 arch/arm/configs/tct_hammer_defconfig | 1 -
 4 files changed, 11 deletions(-)

diff --git a/arch/arm/Kconfig.debug b/arch/arm/Kconfig.debug
index a71fd941ade7..a89e4734b8f0 100644
--- a/arch/arm/Kconfig.debug
+++ b/arch/arm/Kconfig.debug
@@ -99,14 +99,6 @@ config DEBUG_CLPS711X_UART2
 	  output to the second serial port on these devices.  Saying N will
 	  cause the debug messages to appear on the first serial port.
 
-config DEBUG_S3C_PORT
-	depends on DEBUG_LL && PLAT_S3C
-	bool "Kernel low-level debugging messages via S3C UART"
-	help
-	  Say Y here if you want debug print routines to go to one of the
-	  S3C internal UARTs. The chosen UART must have been configured
-	  before it is used.
-
 config DEBUG_S3C_UART
 	depends on PLAT_S3C
 	int "S3C UART to use for low-level debug"
diff --git a/arch/arm/configs/s3c2410_defconfig b/arch/arm/configs/s3c2410_defconfig
index 2d58b8fe59be..f4f1899f3c88 100644
--- a/arch/arm/configs/s3c2410_defconfig
+++ b/arch/arm/configs/s3c2410_defconfig
@@ -2298,7 +2298,6 @@ CONFIG_DEBUG_ERRORS=y
 # CONFIG_DEBUG_STACK_USAGE is not set
 CONFIG_DEBUG_LL=y
 # CONFIG_DEBUG_ICEDCC is not set
-CONFIG_DEBUG_S3C_PORT=y
 CONFIG_DEBUG_S3C_UART=0
 
 #
diff --git a/arch/arm/configs/s3c6400_defconfig b/arch/arm/configs/s3c6400_defconfig
index 2e8fa50e9a09..32860609e057 100644
--- a/arch/arm/configs/s3c6400_defconfig
+++ b/arch/arm/configs/s3c6400_defconfig
@@ -816,7 +816,6 @@ CONFIG_DEBUG_ERRORS=y
 # CONFIG_DEBUG_STACK_USAGE is not set
 CONFIG_DEBUG_LL=y
 # CONFIG_DEBUG_ICEDCC is not set
-CONFIG_DEBUG_S3C_PORT=y
 CONFIG_DEBUG_S3C_UART=0
 
 #
diff --git a/arch/arm/configs/tct_hammer_defconfig b/arch/arm/configs/tct_hammer_defconfig
index 07dfb98df4f0..9d32faef05f6 100644
--- a/arch/arm/configs/tct_hammer_defconfig
+++ b/arch/arm/configs/tct_hammer_defconfig
@@ -857,7 +857,6 @@ CONFIG_DEBUG_ERRORS=y
 # CONFIG_DEBUG_STACK_USAGE is not set
 CONFIG_DEBUG_LL=y
 # CONFIG_DEBUG_ICEDCC is not set
-# CONFIG_DEBUG_S3C_PORT is not set
 CONFIG_DEBUG_S3C_UART=0
 
 #
-- 
cgit v1.2.3-59-g8ed1b


From d5fdd6babcfc2b0e6a8da1acf492a69fb54b4c47 Mon Sep 17 00:00:00 2001
From: Brian Haley <brian.haley@hp.com>
Date: Tue, 23 Jun 2009 04:31:07 -0700
Subject: ipv6: Use correct data types for ICMPv6 type and code

Change all the code that deals directly with ICMPv6 type and code
values to use u8 instead of a signed int as that's the actual data
type.

Signed-off-by: Brian Haley <brian.haley@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/icmpv6.h  |  6 +++---
 include/net/protocol.h  |  2 +-
 include/net/rawv6.h     |  2 +-
 include/net/xfrm.h      |  2 +-
 net/dccp/ipv6.c         |  2 +-
 net/ipv6/ah6.c          |  2 +-
 net/ipv6/esp6.c         |  2 +-
 net/ipv6/icmp.c         | 12 ++++++------
 net/ipv6/ip6_tunnel.c   | 18 +++++++++---------
 net/ipv6/ipcomp6.c      |  2 +-
 net/ipv6/mip6.c         |  2 +-
 net/ipv6/raw.c          |  4 ++--
 net/ipv6/route.c        |  2 +-
 net/ipv6/tcp_ipv6.c     |  2 +-
 net/ipv6/tunnel6.c      |  2 +-
 net/ipv6/udp.c          |  6 +++---
 net/ipv6/udp_impl.h     |  2 +-
 net/ipv6/udplite.c      |  2 +-
 net/ipv6/xfrm6_tunnel.c |  2 +-
 net/sctp/ipv6.c         |  2 +-
 20 files changed, 38 insertions(+), 38 deletions(-)

diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h
index 10d701eec484..b6a85183c333 100644
--- a/include/linux/icmpv6.h
+++ b/include/linux/icmpv6.h
@@ -175,16 +175,16 @@ struct icmp6_filter {
 
 
 extern void				icmpv6_send(struct sk_buff *skb,
-						    int type, int code,
+						    u8 type, u8 code,
 						    __u32 info, 
 						    struct net_device *dev);
 
 extern int				icmpv6_init(void);
-extern int				icmpv6_err_convert(int type, int code,
+extern int				icmpv6_err_convert(u8 type, u8 code,
 							   int *err);
 extern void				icmpv6_cleanup(void);
 extern void				icmpv6_param_prob(struct sk_buff *skb,
-							  int code, int pos);
+							  u8 code, int pos);
 
 struct flowi;
 struct in6_addr;
diff --git a/include/net/protocol.h b/include/net/protocol.h
index ffa5b8b1f1df..1089d5aabd49 100644
--- a/include/net/protocol.h
+++ b/include/net/protocol.h
@@ -53,7 +53,7 @@ struct inet6_protocol
 
 	void	(*err_handler)(struct sk_buff *skb,
 			       struct inet6_skb_parm *opt,
-			       int type, int code, int offset,
+			       u8 type, u8 code, int offset,
 			       __be32 info);
 
 	int	(*gso_send_check)(struct sk_buff *skb);
diff --git a/include/net/rawv6.h b/include/net/rawv6.h
index 8a22599f26ba..f6b9b830df8c 100644
--- a/include/net/rawv6.h
+++ b/include/net/rawv6.h
@@ -6,7 +6,7 @@
 #include <net/protocol.h>
 
 void raw6_icmp_error(struct sk_buff *, int nexthdr,
-		int type, int code, int inner_offset, __be32);
+		u8 type, u8 code, int inner_offset, __be32);
 int raw6_local_deliver(struct sk_buff *, int);
 
 extern int			rawv6_rcv(struct sock *sk,
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 736bca450886..9e3a3f4c1f60 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -1274,7 +1274,7 @@ struct xfrm_tunnel {
 struct xfrm6_tunnel {
 	int (*handler)(struct sk_buff *skb);
 	int (*err_handler)(struct sk_buff *skb, struct inet6_skb_parm *opt,
-			   int type, int code, int offset, __be32 info);
+			   u8 type, u8 code, int offset, __be32 info);
 	struct xfrm6_tunnel *next;
 	int priority;
 };
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 05ea7440d9e5..3e70faab2989 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -85,7 +85,7 @@ static inline __u32 dccp_v6_init_sequence(struct sk_buff *skb)
 }
 
 static void dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
-			int type, int code, int offset, __be32 info)
+			u8 type, u8 code, int offset, __be32 info)
 {
 	struct ipv6hdr *hdr = (struct ipv6hdr *)skb->data;
 	const struct dccp_hdr *dh = (struct dccp_hdr *)(skb->data + offset);
diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c
index 52449f7a1b71..86f42a288c4b 100644
--- a/net/ipv6/ah6.c
+++ b/net/ipv6/ah6.c
@@ -405,7 +405,7 @@ out:
 }
 
 static void ah6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
-		    int type, int code, int offset, __be32 info)
+		    u8 type, u8 code, int offset, __be32 info)
 {
 	struct net *net = dev_net(skb->dev);
 	struct ipv6hdr *iph = (struct ipv6hdr*)skb->data;
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index c2f250150db1..678bb95b1525 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -354,7 +354,7 @@ static u32 esp6_get_mtu(struct xfrm_state *x, int mtu)
 }
 
 static void esp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
-		     int type, int code, int offset, __be32 info)
+		     u8 type, u8 code, int offset, __be32 info)
 {
 	struct net *net = dev_net(skb->dev);
 	struct ipv6hdr *iph = (struct ipv6hdr*)skb->data;
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 36dff8807183..eab62a7a8f06 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -117,7 +117,7 @@ static __inline__ void icmpv6_xmit_unlock(struct sock *sk)
 /*
  * Slightly more convenient version of icmpv6_send.
  */
-void icmpv6_param_prob(struct sk_buff *skb, int code, int pos)
+void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos)
 {
 	icmpv6_send(skb, ICMPV6_PARAMPROB, code, pos, skb->dev);
 	kfree_skb(skb);
@@ -161,7 +161,7 @@ static int is_ineligible(struct sk_buff *skb)
 /*
  * Check the ICMP output rate limit
  */
-static inline int icmpv6_xrlim_allow(struct sock *sk, int type,
+static inline int icmpv6_xrlim_allow(struct sock *sk, u8 type,
 				     struct flowi *fl)
 {
 	struct dst_entry *dst;
@@ -305,7 +305,7 @@ static inline void mip6_addr_swap(struct sk_buff *skb) {}
 /*
  *	Send an ICMP message in response to a packet in error
  */
-void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info,
+void icmpv6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info,
 		 struct net_device *dev)
 {
 	struct net *net = dev_net(skb->dev);
@@ -590,7 +590,7 @@ out:
 	icmpv6_xmit_unlock(sk);
 }
 
-static void icmpv6_notify(struct sk_buff *skb, int type, int code, __be32 info)
+static void icmpv6_notify(struct sk_buff *skb, u8 type, u8 code, __be32 info)
 {
 	struct inet6_protocol *ipprot;
 	int inner_offset;
@@ -643,7 +643,7 @@ static int icmpv6_rcv(struct sk_buff *skb)
 	struct in6_addr *saddr, *daddr;
 	struct ipv6hdr *orig_hdr;
 	struct icmp6hdr *hdr;
-	int type;
+	u8 type;
 
 	if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) {
 		struct sec_path *sp = skb_sec_path(skb);
@@ -914,7 +914,7 @@ static const struct icmp6_err {
 	},
 };
 
-int icmpv6_err_convert(int type, int code, int *err)
+int icmpv6_err_convert(u8 type, u8 code, int *err)
 {
 	int fatal = 0;
 
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 404d16a97d5c..51f410e7775a 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -394,13 +394,13 @@ parse_tlv_tnl_enc_lim(struct sk_buff *skb, __u8 * raw)
 
 static int
 ip6_tnl_err(struct sk_buff *skb, __u8 ipproto, struct inet6_skb_parm *opt,
-	    int *type, int *code, int *msg, __u32 *info, int offset)
+	    u8 *type, u8 *code, int *msg, __u32 *info, int offset)
 {
 	struct ipv6hdr *ipv6h = (struct ipv6hdr *) skb->data;
 	struct ip6_tnl *t;
 	int rel_msg = 0;
-	int rel_type = ICMPV6_DEST_UNREACH;
-	int rel_code = ICMPV6_ADDR_UNREACH;
+	u8 rel_type = ICMPV6_DEST_UNREACH;
+	u8 rel_code = ICMPV6_ADDR_UNREACH;
 	__u32 rel_info = 0;
 	__u16 len;
 	int err = -ENOENT;
@@ -488,11 +488,11 @@ out:
 
 static int
 ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
-	   int type, int code, int offset, __be32 info)
+	   u8 type, u8 code, int offset, __be32 info)
 {
 	int rel_msg = 0;
-	int rel_type = type;
-	int rel_code = code;
+	u8 rel_type = type;
+	u8 rel_code = code;
 	__u32 rel_info = ntohl(info);
 	int err;
 	struct sk_buff *skb2;
@@ -586,11 +586,11 @@ out:
 
 static int
 ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
-	   int type, int code, int offset, __be32 info)
+	   u8 type, u8 code, int offset, __be32 info)
 {
 	int rel_msg = 0;
-	int rel_type = type;
-	int rel_code = code;
+	u8 rel_type = type;
+	u8 rel_code = code;
 	__u32 rel_info = ntohl(info);
 	int err;
 
diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c
index 3a0b3be7ece5..79c172f1ff01 100644
--- a/net/ipv6/ipcomp6.c
+++ b/net/ipv6/ipcomp6.c
@@ -51,7 +51,7 @@
 #include <linux/mutex.h>
 
 static void ipcomp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
-				int type, int code, int offset, __be32 info)
+				u8 type, u8 code, int offset, __be32 info)
 {
 	__be32 spi;
 	struct ipv6hdr *iph = (struct ipv6hdr*)skb->data;
diff --git a/net/ipv6/mip6.c b/net/ipv6/mip6.c
index f995e19c87a9..f797e8c6f3b3 100644
--- a/net/ipv6/mip6.c
+++ b/net/ipv6/mip6.c
@@ -54,7 +54,7 @@ static inline void *mip6_padn(__u8 *data, __u8 padlen)
 	return data + padlen;
 }
 
-static inline void mip6_param_prob(struct sk_buff *skb, int code, int pos)
+static inline void mip6_param_prob(struct sk_buff *skb, u8 code, int pos)
 {
 	icmpv6_send(skb, ICMPV6_PARAMPROB, code, pos, skb->dev);
 }
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 8b0b6f948063..d6c3c1c34b2d 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -310,7 +310,7 @@ out:
 
 static void rawv6_err(struct sock *sk, struct sk_buff *skb,
 	       struct inet6_skb_parm *opt,
-	       int type, int code, int offset, __be32 info)
+	       u8 type, u8 code, int offset, __be32 info)
 {
 	struct inet_sock *inet = inet_sk(sk);
 	struct ipv6_pinfo *np = inet6_sk(sk);
@@ -343,7 +343,7 @@ static void rawv6_err(struct sock *sk, struct sk_buff *skb,
 }
 
 void raw6_icmp_error(struct sk_buff *skb, int nexthdr,
-		int type, int code, int inner_offset, __be32 info)
+		u8 type, u8 code, int inner_offset, __be32 info)
 {
 	struct sock *sk;
 	int hash;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 658293ea05ba..1473ee0a1f51 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1865,7 +1865,7 @@ int ipv6_route_ioctl(struct net *net, unsigned int cmd, void __user *arg)
  *	Drop the packet on the floor
  */
 
-static int ip6_pkt_drop(struct sk_buff *skb, int code, int ipstats_mib_noroutes)
+static int ip6_pkt_drop(struct sk_buff *skb, u8 code, int ipstats_mib_noroutes)
 {
 	int type;
 	struct dst_entry *dst = skb_dst(skb);
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 53b6a4192b16..58810c65b635 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -317,7 +317,7 @@ failure:
 }
 
 static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
-		int type, int code, int offset, __be32 info)
+		u8 type, u8 code, int offset, __be32 info)
 {
 	struct ipv6hdr *hdr = (struct ipv6hdr*)skb->data;
 	const struct tcphdr *th = (struct tcphdr *)(skb->data+offset);
diff --git a/net/ipv6/tunnel6.c b/net/ipv6/tunnel6.c
index 669f280989c3..633ad789effc 100644
--- a/net/ipv6/tunnel6.c
+++ b/net/ipv6/tunnel6.c
@@ -124,7 +124,7 @@ drop:
 }
 
 static void tunnel6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
-			int type, int code, int offset, __be32 info)
+			u8 type, u8 code, int offset, __be32 info)
 {
 	struct xfrm6_tunnel *handler;
 
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 023beda6b224..33b59bd92c4d 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -312,7 +312,7 @@ csum_copy_err:
 }
 
 void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
-		    int type, int code, int offset, __be32 info,
+		    u8 type, u8 code, int offset, __be32 info,
 		    struct udp_table *udptable)
 {
 	struct ipv6_pinfo *np;
@@ -346,8 +346,8 @@ out:
 }
 
 static __inline__ void udpv6_err(struct sk_buff *skb,
-				 struct inet6_skb_parm *opt, int type,
-				 int code, int offset, __be32 info     )
+				 struct inet6_skb_parm *opt, u8 type,
+				 u8 code, int offset, __be32 info     )
 {
 	__udp6_lib_err(skb, opt, type, code, offset, info, &udp_table);
 }
diff --git a/net/ipv6/udp_impl.h b/net/ipv6/udp_impl.h
index 23779208c334..6bb303471e20 100644
--- a/net/ipv6/udp_impl.h
+++ b/net/ipv6/udp_impl.h
@@ -9,7 +9,7 @@
 
 extern int  	__udp6_lib_rcv(struct sk_buff *, struct udp_table *, int );
 extern void 	__udp6_lib_err(struct sk_buff *, struct inet6_skb_parm *,
-			       int , int , int , __be32 , struct udp_table *);
+			       u8 , u8 , int , __be32 , struct udp_table *);
 
 extern int	udp_v6_get_port(struct sock *sk, unsigned short snum);
 
diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c
index ba162a824585..4818c48688f2 100644
--- a/net/ipv6/udplite.c
+++ b/net/ipv6/udplite.c
@@ -20,7 +20,7 @@ static int udplitev6_rcv(struct sk_buff *skb)
 
 static void udplitev6_err(struct sk_buff *skb,
 			  struct inet6_skb_parm *opt,
-			  int type, int code, int offset, __be32 info)
+			  u8 type, u8 code, int offset, __be32 info)
 {
 	__udp6_lib_err(skb, opt, type, code, offset, info, &udplite_table);
 }
diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c
index 80193db224d9..81a95c00e503 100644
--- a/net/ipv6/xfrm6_tunnel.c
+++ b/net/ipv6/xfrm6_tunnel.c
@@ -262,7 +262,7 @@ static int xfrm6_tunnel_rcv(struct sk_buff *skb)
 }
 
 static int xfrm6_tunnel_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
-			    int type, int code, int offset, __be32 info)
+			    u8 type, u8 code, int offset, __be32 info)
 {
 	/* xfrm6_tunnel native err handling */
 	switch (type) {
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index a63de3f7f185..6a4b19094143 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -133,7 +133,7 @@ static struct notifier_block sctp_inet6addr_notifier = {
 
 /* ICMP error handler. */
 SCTP_STATIC void sctp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
-			     int type, int code, int offset, __be32 info)
+			     u8 type, u8 code, int offset, __be32 info)
 {
 	struct inet6_dev *idev;
 	struct sock *sk;
-- 
cgit v1.2.3-59-g8ed1b


From 0634a632f5dea8281ae7c9a96800582ff9eb1475 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Tue, 23 Jun 2009 13:51:19 +0200
Subject: asm-generic: add dummy pgprot_noncached()

Most architectures now provide a pgprot_noncached(), the
remaining ones can simply use an dummy default implementation,
except for cris and xtensa, which should override the
default appropriately.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Cc: Chris Zankel <chris@zankel.net>
Cc: Magnus Damm <magnus.damm@gmail.com>
---
 include/asm-generic/pgtable.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index e410f602cab1..e2bd73e8f9c0 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -129,6 +129,10 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addres
 #define move_pte(pte, prot, old_addr, new_addr)	(pte)
 #endif
 
+#ifndef pgprot_noncached
+#define pgprot_noncached(prot)	(prot)
+#endif
+
 #ifndef pgprot_writecombine
 #define pgprot_writecombine pgprot_noncached
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From f49156ea1bf3bccf45a01351cf3db2b5f6a8597e Mon Sep 17 00:00:00 2001
From: Timur Tabi <timur@freescale.com>
Date: Tue, 26 May 2009 10:21:42 -0500
Subject: powerpc/qe: add polling timeout to qe_issue_cmd()

The qe_issue_cmd() function (Freescale PowerPC QUICC Engine library) polls
on a register until a status bit changes, but does not include a timeout
to handle the situation if the bit never changes.  Change the code to use
the new spin_event_timeout() macro, which simplifies polling on a register
without a timeout.

Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/sysdev/qe_lib/qe.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/arch/powerpc/sysdev/qe_lib/qe.c b/arch/powerpc/sysdev/qe_lib/qe.c
index b28b0e512d67..237e3654f48c 100644
--- a/arch/powerpc/sysdev/qe_lib/qe.c
+++ b/arch/powerpc/sysdev/qe_lib/qe.c
@@ -112,6 +112,7 @@ int qe_issue_cmd(u32 cmd, u32 device, u8 mcn_protocol, u32 cmd_input)
 {
 	unsigned long flags;
 	u8 mcn_shift = 0, dev_shift = 0;
+	u32 ret;
 
 	spin_lock_irqsave(&qe_lock, flags);
 	if (cmd == QE_RESET) {
@@ -139,11 +140,13 @@ int qe_issue_cmd(u32 cmd, u32 device, u8 mcn_protocol, u32 cmd_input)
 	}
 
 	/* wait for the QE_CR_FLG to clear */
-	while(in_be32(&qe_immr->cp.cecr) & QE_CR_FLG)
-		cpu_relax();
+	ret = spin_event_timeout((in_be32(&qe_immr->cp.cecr) & QE_CR_FLG) == 0,
+			   100, 0);
+	/* On timeout (e.g. failure), the expression will be false (ret == 0),
+	   otherwise it will be true (ret == 1). */
 	spin_unlock_irqrestore(&qe_lock, flags);
 
-	return 0;
+	return ret == 1;
 }
 EXPORT_SYMBOL(qe_issue_cmd);
 
-- 
cgit v1.2.3-59-g8ed1b


From 5e10cf587a7b1aa34099cbfc244e6f04e96dc835 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@kernel.crashing.org>
Date: Tue, 19 May 2009 00:52:20 -0500
Subject: powerpc/cpm1: Remove IMAP_ADDR

We no longer user IMAP_ADDR for anything so kill it off.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/include/asm/cpm1.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/arch/powerpc/include/asm/cpm1.h b/arch/powerpc/include/asm/cpm1.h
index 2ff798744c1d..7685ffde8821 100644
--- a/arch/powerpc/include/asm/cpm1.h
+++ b/arch/powerpc/include/asm/cpm1.h
@@ -598,8 +598,6 @@ typedef struct risc_timer_pram {
 #define CICR_IEN		((uint)0x00000080)	/* Int. enable */
 #define CICR_SPS		((uint)0x00000001)	/* SCC Spread */
 
-#define IMAP_ADDR		(get_immrbase())
-
 #define CPM_PIN_INPUT     0
 #define CPM_PIN_OUTPUT    1
 #define CPM_PIN_PRIMARY   0
-- 
cgit v1.2.3-59-g8ed1b


From a44a23ed4d144e4fb83d256b32021b732bda4787 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@kernel.crashing.org>
Date: Thu, 18 Jun 2009 23:28:34 -0500
Subject: powerpc/85xx: Stop using ppc_md.init on socrates

Match what other 85xx platforms do for of_platform_bus_probe and use
machine_device_initcall.  This is one small step in killing of
ppc_md.init.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/platforms/85xx/socrates.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/powerpc/platforms/85xx/socrates.c b/arch/powerpc/platforms/85xx/socrates.c
index d0e8443b12c6..747d8fb3ab82 100644
--- a/arch/powerpc/platforms/85xx/socrates.c
+++ b/arch/powerpc/platforms/85xx/socrates.c
@@ -102,10 +102,11 @@ static struct of_device_id __initdata socrates_of_bus_ids[] = {
 	{},
 };
 
-static void __init socrates_init(void)
+static int __init socrates_publish_devices(void)
 {
-	of_platform_bus_probe(NULL, socrates_of_bus_ids, NULL);
+	return of_platform_bus_probe(NULL, socrates_of_bus_ids, NULL);
 }
+machine_device_initcall(socrates, socrates_publish_devices);
 
 /*
  * Called very early, device-tree isn't unflattened
@@ -124,7 +125,6 @@ define_machine(socrates) {
 	.name			= "Socrates",
 	.probe			= socrates_probe,
 	.setup_arch		= socrates_setup_arch,
-	.init			= socrates_init,
 	.init_IRQ		= socrates_pic_init,
 	.get_irq		= mpic_get_irq,
 	.restart		= fsl_rstcr_restart,
-- 
cgit v1.2.3-59-g8ed1b


From fa874618c3155e4f255387987d630a203578b5ae Mon Sep 17 00:00:00 2001
From: Randy Vinson <rvinson@mvista.com>
Date: Fri, 19 Jun 2009 03:22:08 +0400
Subject: powerpc/85xx: Fix FSL RapidIO probing on MDS boards

FSL RapidIO won't probe without a proper compatible entry. This
patch fixes the issue by adding fsl,rapidio-delta compatible to
mpc85xx_ids.

Signed-off-by: Randy Vinson <rvinson@mvista.com>
Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/platforms/85xx/mpc85xx_mds.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/powerpc/platforms/85xx/mpc85xx_mds.c b/arch/powerpc/platforms/85xx/mpc85xx_mds.c
index 77f90b356356..60ed9c067b1d 100644
--- a/arch/powerpc/platforms/85xx/mpc85xx_mds.c
+++ b/arch/powerpc/platforms/85xx/mpc85xx_mds.c
@@ -285,6 +285,7 @@ static struct of_device_id mpc85xx_ids[] = {
 	{ .type = "qe", },
 	{ .compatible = "fsl,qe", },
 	{ .compatible = "gianfar", },
+	{ .compatible = "fsl,rapidio-delta", },
 	{},
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From cb1ffb6204712b04396ae0a9f3d1bf93cd8df8fb Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@kernel.crashing.org>
Date: Fri, 19 Jun 2009 03:30:42 -0500
Subject: powerpc/85xx: Fix issue found by lockdep trace in smp_85xx_kick_cpu

lockdep trace found the following:

------------[ cut here ]------------
Badness at c007baf0 [verbose debug info unavailable]
NIP: c007baf0 LR: c007bad8 CTR: 00000000
REGS: ef855e00 TRAP: 0700   Tainted: G        W
(2.6.30-06736-g12a31df-dirty)
MSR: 00021000 <ME,CE>  CR: 24044022  XER: 20000000
TASK = ef858000[1] 'swapper' THREAD: ef854000 CPU: 0
GPR00: 00000000 ef855eb0 ef858000 00000001 000000d0 f1000000 ffbc8000 ffffffff
GPR08: 000000d0 c0760000 c0710000 00000007 2fffffff 1004a388 7ffd9400 00000000
GPR16: 00000000 7ffcd100 7ffcd100 7ffcd100 c059cd78 c075c498 c057da7c ffffffff
GPR24: ffbc8000 f1000000 00000001 c00bf8b0 c07595d4 000000d0 00021000 000000d0
NIP [c007baf0] lockdep_trace_alloc+0xc0/0xf0
LR [c007bad8] lockdep_trace_alloc+0xa8/0xf0
Call Trace:
[ef855eb0] [c007ba60] lockdep_trace_alloc+0x30/0xf0 (unreliable)
[ef855ec0] [c00cb3ac] kmem_cache_alloc+0x2c/0xf0
[ef855ee0] [c00bf8b0] __get_vm_area_node+0x80/0x1c0
[ef855f10] [c0017580] __ioremap_caller+0x1d0/0x1e0
[ef855f40] [c057da7c] smp_85xx_kick_cpu+0x64/0x124
[ef855f60] [c0599180] __cpu_up+0xd0/0x1a4
[ef855f80] [c05997c4] cpu_up+0x14c/0x1e0
[ef855fc0] [c05732a0] kernel_init+0x100/0x1c4
[ef855ff0] [c0011524] kernel_thread+0x4c/0x68
Instruction dump:
8009c174 2f800000 409e0048 73c08000 40820040 4818980d 2f830000 419effa0
3d20c076 8009c388 2f800000 409eff90 <0fe00000> 4bffff88 60000000 60000000

We were calling ioremap after we local_irq_restore(flags).  A simple
reorder fixes the problem.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/platforms/85xx/smp.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/arch/powerpc/platforms/85xx/smp.c b/arch/powerpc/platforms/85xx/smp.c
index cc0b0db8a6f3..62c592ede641 100644
--- a/arch/powerpc/platforms/85xx/smp.c
+++ b/arch/powerpc/platforms/85xx/smp.c
@@ -52,20 +52,19 @@ smp_85xx_kick_cpu(int nr)
 
 	pr_debug("smp_85xx_kick_cpu: kick CPU #%d\n", nr);
 
-	local_irq_save(flags);
-
 	np = of_get_cpu_node(nr, NULL);
 	cpu_rel_addr = of_get_property(np, "cpu-release-addr", NULL);
 
 	if (cpu_rel_addr == NULL) {
 		printk(KERN_ERR "No cpu-release-addr for cpu %d\n", nr);
-		local_irq_restore(flags);
 		return;
 	}
 
 	/* Map the spin table */
 	bptr_vaddr = ioremap(*cpu_rel_addr, SIZE_BOOT_ENTRY);
 
+	local_irq_save(flags);
+
 	out_be32(bptr_vaddr + BOOT_ENTRY_PIR, nr);
 	out_be32(bptr_vaddr + BOOT_ENTRY_ADDR_LOWER, __pa(__early_start));
 
@@ -73,10 +72,10 @@ smp_85xx_kick_cpu(int nr)
 	while ((__secondary_hold_acknowledge != nr) && (++n < 1000))
 		mdelay(1);
 
-	iounmap(bptr_vaddr);
-
 	local_irq_restore(flags);
 
+	iounmap(bptr_vaddr);
+
 	pr_debug("waited %d msecs for CPU #%d.\n", n, nr);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From b053dc5a722eade28514f2cc922caf7a4baad987 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@kernel.crashing.org>
Date: Fri, 19 Jun 2009 08:31:05 -0500
Subject: powerpc: Refactor device tree binding

Split device tree binding out of booting-without-of.txt and put them
into their own files per binding.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 Documentation/powerpc/booting-without-of.txt     | 1168 +---------------------
 Documentation/powerpc/dts-bindings/4xx/emac.txt  |  148 +++
 Documentation/powerpc/dts-bindings/gpio/gpio.txt |   50 +
 Documentation/powerpc/dts-bindings/gpio/mdio.txt |   19 +
 Documentation/powerpc/dts-bindings/marvell.txt   |  521 ++++++++++
 Documentation/powerpc/dts-bindings/phy.txt       |   25 +
 Documentation/powerpc/dts-bindings/spi-bus.txt   |   57 ++
 Documentation/powerpc/dts-bindings/usb-ehci.txt  |   25 +
 Documentation/powerpc/dts-bindings/xilinx.txt    |  295 ++++++
 9 files changed, 1142 insertions(+), 1166 deletions(-)
 create mode 100644 Documentation/powerpc/dts-bindings/4xx/emac.txt
 create mode 100644 Documentation/powerpc/dts-bindings/gpio/gpio.txt
 create mode 100644 Documentation/powerpc/dts-bindings/gpio/mdio.txt
 create mode 100644 Documentation/powerpc/dts-bindings/marvell.txt
 create mode 100644 Documentation/powerpc/dts-bindings/phy.txt
 create mode 100644 Documentation/powerpc/dts-bindings/spi-bus.txt
 create mode 100644 Documentation/powerpc/dts-bindings/usb-ehci.txt
 create mode 100644 Documentation/powerpc/dts-bindings/xilinx.txt

diff --git a/Documentation/powerpc/booting-without-of.txt b/Documentation/powerpc/booting-without-of.txt
index 8d999d862d0e..79f533f38c61 100644
--- a/Documentation/powerpc/booting-without-of.txt
+++ b/Documentation/powerpc/booting-without-of.txt
@@ -1238,1122 +1238,7 @@ descriptions for the SOC devices for which new nodes have been
 defined; this list will expand as more and more SOC-containing
 platforms are moved over to use the flattened-device-tree model.
 
-   a) PHY nodes
-
-   Required properties:
-
-    - device_type : Should be "ethernet-phy"
-    - interrupts : <a b> where a is the interrupt number and b is a
-      field that represents an encoding of the sense and level
-      information for the interrupt.  This should be encoded based on
-      the information in section 2) depending on the type of interrupt
-      controller you have.
-    - interrupt-parent : the phandle for the interrupt controller that
-      services interrupts for this device.
-    - reg : The ID number for the phy, usually a small integer
-    - linux,phandle :  phandle for this node; likely referenced by an
-      ethernet controller node.
-
-
-   Example:
-
-	ethernet-phy@0 {
-		linux,phandle = <2452000>
-		interrupt-parent = <40000>;
-		interrupts = <35 1>;
-		reg = <0>;
-		device_type = "ethernet-phy";
-	};
-
-
-   b) Interrupt controllers
-
-   Some SOC devices contain interrupt controllers that are different
-   from the standard Open PIC specification.  The SOC device nodes for
-   these types of controllers should be specified just like a standard
-   OpenPIC controller.  Sense and level information should be encoded
-   as specified in section 2) of this chapter for each device that
-   specifies an interrupt.
-
-   Example :
-
-	pic@40000 {
-		linux,phandle = <40000>;
-		interrupt-controller;
-		#address-cells = <0>;
-		reg = <40000 40000>;
-		compatible = "chrp,open-pic";
-		device_type = "open-pic";
-	};
-
-    c) 4xx/Axon EMAC ethernet nodes
-
-    The EMAC ethernet controller in IBM and AMCC 4xx chips, and also
-    the Axon bridge.  To operate this needs to interact with a ths
-    special McMAL DMA controller, and sometimes an RGMII or ZMII
-    interface.  In addition to the nodes and properties described
-    below, the node for the OPB bus on which the EMAC sits must have a
-    correct clock-frequency property.
-
-      i) The EMAC node itself
-
-    Required properties:
-    - device_type       : "network"
-
-    - compatible        : compatible list, contains 2 entries, first is
-			  "ibm,emac-CHIP" where CHIP is the host ASIC (440gx,
-			  405gp, Axon) and second is either "ibm,emac" or
-			  "ibm,emac4".  For Axon, thus, we have: "ibm,emac-axon",
-			  "ibm,emac4"
-    - interrupts        : <interrupt mapping for EMAC IRQ and WOL IRQ>
-    - interrupt-parent  : optional, if needed for interrupt mapping
-    - reg               : <registers mapping>
-    - local-mac-address : 6 bytes, MAC address
-    - mal-device        : phandle of the associated McMAL node
-    - mal-tx-channel    : 1 cell, index of the tx channel on McMAL associated
-			  with this EMAC
-    - mal-rx-channel    : 1 cell, index of the rx channel on McMAL associated
-			  with this EMAC
-    - cell-index        : 1 cell, hardware index of the EMAC cell on a given
-			  ASIC (typically 0x0 and 0x1 for EMAC0 and EMAC1 on
-			  each Axon chip)
-    - max-frame-size    : 1 cell, maximum frame size supported in bytes
-    - rx-fifo-size      : 1 cell, Rx fifo size in bytes for 10 and 100 Mb/sec
-			  operations.
-			  For Axon, 2048
-    - tx-fifo-size      : 1 cell, Tx fifo size in bytes for 10 and 100 Mb/sec
-			  operations.
-			  For Axon, 2048.
-    - fifo-entry-size   : 1 cell, size of a fifo entry (used to calculate
-			  thresholds).
-			  For Axon, 0x00000010
-    - mal-burst-size    : 1 cell, MAL burst size (used to calculate thresholds)
-			  in bytes.
-			  For Axon, 0x00000100 (I think ...)
-    - phy-mode          : string, mode of operations of the PHY interface.
-			  Supported values are: "mii", "rmii", "smii", "rgmii",
-			  "tbi", "gmii", rtbi", "sgmii".
-			  For Axon on CAB, it is "rgmii"
-    - mdio-device       : 1 cell, required iff using shared MDIO registers
-			  (440EP).  phandle of the EMAC to use to drive the
-			  MDIO lines for the PHY used by this EMAC.
-    - zmii-device       : 1 cell, required iff connected to a ZMII.  phandle of
-			  the ZMII device node
-    - zmii-channel      : 1 cell, required iff connected to a ZMII.  Which ZMII
-			  channel or 0xffffffff if ZMII is only used for MDIO.
-    - rgmii-device      : 1 cell, required iff connected to an RGMII. phandle
-			  of the RGMII device node.
-			  For Axon: phandle of plb5/plb4/opb/rgmii
-    - rgmii-channel     : 1 cell, required iff connected to an RGMII.  Which
-			  RGMII channel is used by this EMAC.
-			  Fox Axon: present, whatever value is appropriate for each
-			  EMAC, that is the content of the current (bogus) "phy-port"
-			  property.
-
-    Optional properties:
-    - phy-address       : 1 cell, optional, MDIO address of the PHY. If absent,
-			  a search is performed.
-    - phy-map           : 1 cell, optional, bitmap of addresses to probe the PHY
-			  for, used if phy-address is absent. bit 0x00000001 is
-			  MDIO address 0.
-			  For Axon it can be absent, though my current driver
-			  doesn't handle phy-address yet so for now, keep
-			  0x00ffffff in it.
-    - rx-fifo-size-gige : 1 cell, Rx fifo size in bytes for 1000 Mb/sec
-			  operations (if absent the value is the same as
-			  rx-fifo-size).  For Axon, either absent or 2048.
-    - tx-fifo-size-gige : 1 cell, Tx fifo size in bytes for 1000 Mb/sec
-			  operations (if absent the value is the same as
-			  tx-fifo-size). For Axon, either absent or 2048.
-    - tah-device        : 1 cell, optional. If connected to a TAH engine for
-			  offload, phandle of the TAH device node.
-    - tah-channel       : 1 cell, optional. If appropriate, channel used on the
-			  TAH engine.
-
-    Example:
-
-	EMAC0: ethernet@40000800 {
-		device_type = "network";
-		compatible = "ibm,emac-440gp", "ibm,emac";
-		interrupt-parent = <&UIC1>;
-		interrupts = <1c 4 1d 4>;
-		reg = <40000800 70>;
-		local-mac-address = [00 04 AC E3 1B 1E];
-		mal-device = <&MAL0>;
-		mal-tx-channel = <0 1>;
-		mal-rx-channel = <0>;
-		cell-index = <0>;
-		max-frame-size = <5dc>;
-		rx-fifo-size = <1000>;
-		tx-fifo-size = <800>;
-		phy-mode = "rmii";
-		phy-map = <00000001>;
-		zmii-device = <&ZMII0>;
-		zmii-channel = <0>;
-	};
-
-      ii) McMAL node
-
-    Required properties:
-    - device_type        : "dma-controller"
-    - compatible         : compatible list, containing 2 entries, first is
-			   "ibm,mcmal-CHIP" where CHIP is the host ASIC (like
-			   emac) and the second is either "ibm,mcmal" or
-			   "ibm,mcmal2".
-			   For Axon, "ibm,mcmal-axon","ibm,mcmal2"
-    - interrupts         : <interrupt mapping for the MAL interrupts sources:
-                           5 sources: tx_eob, rx_eob, serr, txde, rxde>.
-                           For Axon: This is _different_ from the current
-			   firmware.  We use the "delayed" interrupts for txeob
-			   and rxeob. Thus we end up with mapping those 5 MPIC
-			   interrupts, all level positive sensitive: 10, 11, 32,
-			   33, 34 (in decimal)
-    - dcr-reg            : < DCR registers range >
-    - dcr-parent         : if needed for dcr-reg
-    - num-tx-chans       : 1 cell, number of Tx channels
-    - num-rx-chans       : 1 cell, number of Rx channels
-
-      iii) ZMII node
-
-    Required properties:
-    - compatible         : compatible list, containing 2 entries, first is
-			   "ibm,zmii-CHIP" where CHIP is the host ASIC (like
-			   EMAC) and the second is "ibm,zmii".
-			   For Axon, there is no ZMII node.
-    - reg                : <registers mapping>
-
-      iv) RGMII node
-
-    Required properties:
-    - compatible         : compatible list, containing 2 entries, first is
-			   "ibm,rgmii-CHIP" where CHIP is the host ASIC (like
-			   EMAC) and the second is "ibm,rgmii".
-                           For Axon, "ibm,rgmii-axon","ibm,rgmii"
-    - reg                : <registers mapping>
-    - revision           : as provided by the RGMII new version register if
-			   available.
-			   For Axon: 0x0000012a
-
-   d) Xilinx IP cores
-
-   The Xilinx EDK toolchain ships with a set of IP cores (devices) for use
-   in Xilinx Spartan and Virtex FPGAs.  The devices cover the whole range
-   of standard device types (network, serial, etc.) and miscellaneous
-   devices (gpio, LCD, spi, etc).  Also, since these devices are
-   implemented within the fpga fabric every instance of the device can be
-   synthesised with different options that change the behaviour.
-
-   Each IP-core has a set of parameters which the FPGA designer can use to
-   control how the core is synthesized.  Historically, the EDK tool would
-   extract the device parameters relevant to device drivers and copy them
-   into an 'xparameters.h' in the form of #define symbols.  This tells the
-   device drivers how the IP cores are configured, but it requres the kernel
-   to be recompiled every time the FPGA bitstream is resynthesized.
-
-   The new approach is to export the parameters into the device tree and
-   generate a new device tree each time the FPGA bitstream changes.  The
-   parameters which used to be exported as #defines will now become
-   properties of the device node.  In general, device nodes for IP-cores
-   will take the following form:
-
-	(name): (generic-name)@(base-address) {
-		compatible = "xlnx,(ip-core-name)-(HW_VER)"
-			     [, (list of compatible devices), ...];
-		reg = <(baseaddr) (size)>;
-		interrupt-parent = <&interrupt-controller-phandle>;
-		interrupts = < ... >;
-		xlnx,(parameter1) = "(string-value)";
-		xlnx,(parameter2) = <(int-value)>;
-	};
-
-	(generic-name):   an open firmware-style name that describes the
-			generic class of device.  Preferably, this is one word, such
-			as 'serial' or 'ethernet'.
-	(ip-core-name):	the name of the ip block (given after the BEGIN
-			directive in system.mhs).  Should be in lowercase
-			and all underscores '_' converted to dashes '-'.
-	(name):		is derived from the "PARAMETER INSTANCE" value.
-	(parameter#):	C_* parameters from system.mhs.  The C_ prefix is
-			dropped from the parameter name, the name is converted
-			to lowercase and all underscore '_' characters are
-			converted to dashes '-'.
-	(baseaddr):	the baseaddr parameter value (often named C_BASEADDR).
-	(HW_VER):	from the HW_VER parameter.
-	(size):		the address range size (often C_HIGHADDR - C_BASEADDR + 1).
-
-   Typically, the compatible list will include the exact IP core version
-   followed by an older IP core version which implements the same
-   interface or any other device with the same interface.
-
-   'reg', 'interrupt-parent' and 'interrupts' are all optional properties.
-
-   For example, the following block from system.mhs:
-
-	BEGIN opb_uartlite
-		PARAMETER INSTANCE = opb_uartlite_0
-		PARAMETER HW_VER = 1.00.b
-		PARAMETER C_BAUDRATE = 115200
-		PARAMETER C_DATA_BITS = 8
-		PARAMETER C_ODD_PARITY = 0
-		PARAMETER C_USE_PARITY = 0
-		PARAMETER C_CLK_FREQ = 50000000
-		PARAMETER C_BASEADDR = 0xEC100000
-		PARAMETER C_HIGHADDR = 0xEC10FFFF
-		BUS_INTERFACE SOPB = opb_7
-		PORT OPB_Clk = CLK_50MHz
-		PORT Interrupt = opb_uartlite_0_Interrupt
-		PORT RX = opb_uartlite_0_RX
-		PORT TX = opb_uartlite_0_TX
-		PORT OPB_Rst = sys_bus_reset_0
-	END
-
-   becomes the following device tree node:
-
-	opb_uartlite_0: serial@ec100000 {
-		device_type = "serial";
-		compatible = "xlnx,opb-uartlite-1.00.b";
-		reg = <ec100000 10000>;
-		interrupt-parent = <&opb_intc_0>;
-		interrupts = <1 0>; // got this from the opb_intc parameters
-		current-speed = <d#115200>;	// standard serial device prop
-		clock-frequency = <d#50000000>;	// standard serial device prop
-		xlnx,data-bits = <8>;
-		xlnx,odd-parity = <0>;
-		xlnx,use-parity = <0>;
-	};
-
-   Some IP cores actually implement 2 or more logical devices.  In
-   this case, the device should still describe the whole IP core with
-   a single node and add a child node for each logical device.  The
-   ranges property can be used to translate from parent IP-core to the
-   registers of each device.  In addition, the parent node should be
-   compatible with the bus type 'xlnx,compound', and should contain
-   #address-cells and #size-cells, as with any other bus.  (Note: this
-   makes the assumption that both logical devices have the same bus
-   binding.  If this is not true, then separate nodes should be used
-   for each logical device).  The 'cell-index' property can be used to
-   enumerate logical devices within an IP core.  For example, the
-   following is the system.mhs entry for the dual ps2 controller found
-   on the ml403 reference design.
-
-	BEGIN opb_ps2_dual_ref
-		PARAMETER INSTANCE = opb_ps2_dual_ref_0
-		PARAMETER HW_VER = 1.00.a
-		PARAMETER C_BASEADDR = 0xA9000000
-		PARAMETER C_HIGHADDR = 0xA9001FFF
-		BUS_INTERFACE SOPB = opb_v20_0
-		PORT Sys_Intr1 = ps2_1_intr
-		PORT Sys_Intr2 = ps2_2_intr
-		PORT Clkin1 = ps2_clk_rx_1
-		PORT Clkin2 = ps2_clk_rx_2
-		PORT Clkpd1 = ps2_clk_tx_1
-		PORT Clkpd2 = ps2_clk_tx_2
-		PORT Rx1 = ps2_d_rx_1
-		PORT Rx2 = ps2_d_rx_2
-		PORT Txpd1 = ps2_d_tx_1
-		PORT Txpd2 = ps2_d_tx_2
-	END
-
-   It would result in the following device tree nodes:
-
-	opb_ps2_dual_ref_0: opb-ps2-dual-ref@a9000000 {
-		#address-cells = <1>;
-		#size-cells = <1>;
-		compatible = "xlnx,compound";
-		ranges = <0 a9000000 2000>;
-		// If this device had extra parameters, then they would
-		// go here.
-		ps2@0 {
-			compatible = "xlnx,opb-ps2-dual-ref-1.00.a";
-			reg = <0 40>;
-			interrupt-parent = <&opb_intc_0>;
-			interrupts = <3 0>;
-			cell-index = <0>;
-		};
-		ps2@1000 {
-			compatible = "xlnx,opb-ps2-dual-ref-1.00.a";
-			reg = <1000 40>;
-			interrupt-parent = <&opb_intc_0>;
-			interrupts = <3 0>;
-			cell-index = <0>;
-		};
-	};
-
-   Also, the system.mhs file defines bus attachments from the processor
-   to the devices.  The device tree structure should reflect the bus
-   attachments.  Again an example; this system.mhs fragment:
-
-	BEGIN ppc405_virtex4
-		PARAMETER INSTANCE = ppc405_0
-		PARAMETER HW_VER = 1.01.a
-		BUS_INTERFACE DPLB = plb_v34_0
-		BUS_INTERFACE IPLB = plb_v34_0
-	END
-
-	BEGIN opb_intc
-		PARAMETER INSTANCE = opb_intc_0
-		PARAMETER HW_VER = 1.00.c
-		PARAMETER C_BASEADDR = 0xD1000FC0
-		PARAMETER C_HIGHADDR = 0xD1000FDF
-		BUS_INTERFACE SOPB = opb_v20_0
-	END
-
-	BEGIN opb_uart16550
-		PARAMETER INSTANCE = opb_uart16550_0
-		PARAMETER HW_VER = 1.00.d
-		PARAMETER C_BASEADDR = 0xa0000000
-		PARAMETER C_HIGHADDR = 0xa0001FFF
-		BUS_INTERFACE SOPB = opb_v20_0
-	END
-
-	BEGIN plb_v34
-		PARAMETER INSTANCE = plb_v34_0
-		PARAMETER HW_VER = 1.02.a
-	END
-
-	BEGIN plb_bram_if_cntlr
-		PARAMETER INSTANCE = plb_bram_if_cntlr_0
-		PARAMETER HW_VER = 1.00.b
-		PARAMETER C_BASEADDR = 0xFFFF0000
-		PARAMETER C_HIGHADDR = 0xFFFFFFFF
-		BUS_INTERFACE SPLB = plb_v34_0
-	END
-
-	BEGIN plb2opb_bridge
-		PARAMETER INSTANCE = plb2opb_bridge_0
-		PARAMETER HW_VER = 1.01.a
-		PARAMETER C_RNG0_BASEADDR = 0x20000000
-		PARAMETER C_RNG0_HIGHADDR = 0x3FFFFFFF
-		PARAMETER C_RNG1_BASEADDR = 0x60000000
-		PARAMETER C_RNG1_HIGHADDR = 0x7FFFFFFF
-		PARAMETER C_RNG2_BASEADDR = 0x80000000
-		PARAMETER C_RNG2_HIGHADDR = 0xBFFFFFFF
-		PARAMETER C_RNG3_BASEADDR = 0xC0000000
-		PARAMETER C_RNG3_HIGHADDR = 0xDFFFFFFF
-		BUS_INTERFACE SPLB = plb_v34_0
-		BUS_INTERFACE MOPB = opb_v20_0
-	END
-
-   Gives this device tree (some properties removed for clarity):
-
-	plb@0 {
-		#address-cells = <1>;
-		#size-cells = <1>;
-		compatible = "xlnx,plb-v34-1.02.a";
-		device_type = "ibm,plb";
-		ranges; // 1:1 translation
-
-		plb_bram_if_cntrl_0: bram@ffff0000 {
-			reg = <ffff0000 10000>;
-		}
-
-		opb@20000000 {
-			#address-cells = <1>;
-			#size-cells = <1>;
-			ranges = <20000000 20000000 20000000
-				  60000000 60000000 20000000
-				  80000000 80000000 40000000
-				  c0000000 c0000000 20000000>;
-
-			opb_uart16550_0: serial@a0000000 {
-				reg = <a00000000 2000>;
-			};
-
-			opb_intc_0: interrupt-controller@d1000fc0 {
-				reg = <d1000fc0 20>;
-			};
-		};
-	};
-
-   That covers the general approach to binding xilinx IP cores into the
-   device tree.  The following are bindings for specific devices:
-
-      i) Xilinx ML300 Framebuffer
-
-      Simple framebuffer device from the ML300 reference design (also on the
-      ML403 reference design as well as others).
-
-      Optional properties:
-       - resolution = <xres yres> : pixel resolution of framebuffer.  Some
-                                    implementations use a different resolution.
-                                    Default is <d#640 d#480>
-       - virt-resolution = <xvirt yvirt> : Size of framebuffer in memory.
-                                           Default is <d#1024 d#480>.
-       - rotate-display (empty) : rotate display 180 degrees.
-
-      ii) Xilinx SystemACE
-
-      The Xilinx SystemACE device is used to program FPGAs from an FPGA
-      bitstream stored on a CF card.  It can also be used as a generic CF
-      interface device.
-
-      Optional properties:
-       - 8-bit (empty) : Set this property for SystemACE in 8 bit mode
-
-      iii) Xilinx EMAC and Xilinx TEMAC
-
-      Xilinx Ethernet devices.  In addition to general xilinx properties
-      listed above, nodes for these devices should include a phy-handle
-      property, and may include other common network device properties
-      like local-mac-address.
-
-      iv) Xilinx Uartlite
-
-      Xilinx uartlite devices are simple fixed speed serial ports.
-
-      Required properties:
-       - current-speed : Baud rate of uartlite
-
-      v) Xilinx hwicap
-
-		Xilinx hwicap devices provide access to the configuration logic
-		of the FPGA through the Internal Configuration Access Port
-		(ICAP).  The ICAP enables partial reconfiguration of the FPGA,
-		readback of the configuration information, and some control over
-		'warm boots' of the FPGA fabric.
-
-		Required properties:
-		- xlnx,family : The family of the FPGA, necessary since the
-                      capabilities of the underlying ICAP hardware
-                      differ between different families.  May be
-                      'virtex2p', 'virtex4', or 'virtex5'.
-
-      vi) Xilinx Uart 16550
-
-      Xilinx UART 16550 devices are very similar to the NS16550 but with
-      different register spacing and an offset from the base address.
-
-      Required properties:
-       - clock-frequency : Frequency of the clock input
-       - reg-offset : A value of 3 is required
-       - reg-shift : A value of 2 is required
-
-    e) USB EHCI controllers
-
-    Required properties:
-      - compatible : should be "usb-ehci".
-      - reg : should contain at least address and length of the standard EHCI
-        register set for the device. Optional platform-dependent registers
-        (debug-port or other) can be also specified here, but only after
-        definition of standard EHCI registers.
-      - interrupts : one EHCI interrupt should be described here.
-    If device registers are implemented in big endian mode, the device
-    node should have "big-endian-regs" property.
-    If controller implementation operates with big endian descriptors,
-    "big-endian-desc" property should be specified.
-    If both big endian registers and descriptors are used by the controller
-    implementation, "big-endian" property can be specified instead of having
-    both "big-endian-regs" and "big-endian-desc".
-
-     Example (Sequoia 440EPx):
-	    ehci@e0000300 {
-		   compatible = "ibm,usb-ehci-440epx", "usb-ehci";
-		   interrupt-parent = <&UIC0>;
-		   interrupts = <1a 4>;
-		   reg = <0 e0000300 90 0 e0000390 70>;
-		   big-endian;
-	   };
-
-   f) MDIO on GPIOs
-
-   Currently defined compatibles:
-   - virtual,gpio-mdio
-
-   MDC and MDIO lines connected to GPIO controllers are listed in the
-   gpios property as described in section VIII.1 in the following order:
-
-   MDC, MDIO.
-
-   Example:
-
-	mdio {
-		compatible = "virtual,mdio-gpio";
-		#address-cells = <1>;
-		#size-cells = <0>;
-		gpios = <&qe_pio_a 11
-			 &qe_pio_c 6>;
-	};
-
-    g) SPI (Serial Peripheral Interface) busses
-
-    SPI busses can be described with a node for the SPI master device
-    and a set of child nodes for each SPI slave on the bus.  For this
-    discussion, it is assumed that the system's SPI controller is in
-    SPI master mode.  This binding does not describe SPI controllers
-    in slave mode.
-
-    The SPI master node requires the following properties:
-    - #address-cells  - number of cells required to define a chip select
-			address on the SPI bus.
-    - #size-cells     - should be zero.
-    - compatible      - name of SPI bus controller following generic names
-			recommended practice.
-    No other properties are required in the SPI bus node.  It is assumed
-    that a driver for an SPI bus device will understand that it is an SPI bus.
-    However, the binding does not attempt to define the specific method for
-    assigning chip select numbers.  Since SPI chip select configuration is
-    flexible and non-standardized, it is left out of this binding with the
-    assumption that board specific platform code will be used to manage
-    chip selects.  Individual drivers can define additional properties to
-    support describing the chip select layout.
-
-    SPI slave nodes must be children of the SPI master node and can
-    contain the following properties.
-    - reg             - (required) chip select address of device.
-    - compatible      - (required) name of SPI device following generic names
-			recommended practice
-    - spi-max-frequency - (required) Maximum SPI clocking speed of device in Hz
-    - spi-cpol        - (optional) Empty property indicating device requires
-			inverse clock polarity (CPOL) mode
-    - spi-cpha        - (optional) Empty property indicating device requires
-			shifted clock phase (CPHA) mode
-    - spi-cs-high     - (optional) Empty property indicating device requires
-			chip select active high
-
-    SPI example for an MPC5200 SPI bus:
-		spi@f00 {
-			#address-cells = <1>;
-			#size-cells = <0>;
-			compatible = "fsl,mpc5200b-spi","fsl,mpc5200-spi";
-			reg = <0xf00 0x20>;
-			interrupts = <2 13 0 2 14 0>;
-			interrupt-parent = <&mpc5200_pic>;
-
-			ethernet-switch@0 {
-				compatible = "micrel,ks8995m";
-				spi-max-frequency = <1000000>;
-				reg = <0>;
-			};
-
-			codec@1 {
-				compatible = "ti,tlv320aic26";
-				spi-max-frequency = <100000>;
-				reg = <1>;
-			};
-		};
-
-VII - Marvell Discovery mv64[345]6x System Controller chips
-===========================================================
-
-The Marvell mv64[345]60 series of system controller chips contain
-many of the peripherals needed to implement a complete computer
-system.  In this section, we define device tree nodes to describe
-the system controller chip itself and each of the peripherals
-which it contains.  Compatible string values for each node are
-prefixed with the string "marvell,", for Marvell Technology Group Ltd.
-
-1) The /system-controller node
-
-  This node is used to represent the system-controller and must be
-  present when the system uses a system controller chip. The top-level
-  system-controller node contains information that is global to all
-  devices within the system controller chip. The node name begins
-  with "system-controller" followed by the unit address, which is
-  the base address of the memory-mapped register set for the system
-  controller chip.
-
-  Required properties:
-
-    - ranges : Describes the translation of system controller addresses
-      for memory mapped registers.
-    - clock-frequency: Contains the main clock frequency for the system
-      controller chip.
-    - reg : This property defines the address and size of the
-      memory-mapped registers contained within the system controller
-      chip.  The address specified in the "reg" property should match
-      the unit address of the system-controller node.
-    - #address-cells : Address representation for system controller
-      devices.  This field represents the number of cells needed to
-      represent the address of the memory-mapped registers of devices
-      within the system controller chip.
-    - #size-cells : Size representation for for the memory-mapped
-      registers within the system controller chip.
-    - #interrupt-cells : Defines the width of cells used to represent
-      interrupts.
-
-  Optional properties:
-
-    - model : The specific model of the system controller chip.  Such
-      as, "mv64360", "mv64460", or "mv64560".
-    - compatible : A string identifying the compatibility identifiers
-      of the system controller chip.
-
-  The system-controller node contains child nodes for each system
-  controller device that the platform uses.  Nodes should not be created
-  for devices which exist on the system controller chip but are not used
-
-  Example Marvell Discovery mv64360 system-controller node:
-
-    system-controller@f1000000 { /* Marvell Discovery mv64360 */
-	    #address-cells = <1>;
-	    #size-cells = <1>;
-	    model = "mv64360";                      /* Default */
-	    compatible = "marvell,mv64360";
-	    clock-frequency = <133333333>;
-	    reg = <0xf1000000 0x10000>;
-	    virtual-reg = <0xf1000000>;
-	    ranges = <0x88000000 0x88000000 0x1000000 /* PCI 0 I/O Space */
-		    0x80000000 0x80000000 0x8000000 /* PCI 0 MEM Space */
-		    0xa0000000 0xa0000000 0x4000000 /* User FLASH */
-		    0x00000000 0xf1000000 0x0010000 /* Bridge's regs */
-		    0xf2000000 0xf2000000 0x0040000>;/* Integrated SRAM */
-
-	    [ child node definitions... ]
-    }
-
-2) Child nodes of /system-controller
-
-   a) Marvell Discovery MDIO bus
-
-   The MDIO is a bus to which the PHY devices are connected.  For each
-   device that exists on this bus, a child node should be created.  See
-   the definition of the PHY node below for an example of how to define
-   a PHY.
-
-   Required properties:
-     - #address-cells : Should be <1>
-     - #size-cells : Should be <0>
-     - device_type : Should be "mdio"
-     - compatible : Should be "marvell,mv64360-mdio"
-
-   Example:
-
-     mdio {
-	     #address-cells = <1>;
-	     #size-cells = <0>;
-	     device_type = "mdio";
-	     compatible = "marvell,mv64360-mdio";
-
-	     ethernet-phy@0 {
-		     ......
-	     };
-     };
-
-
-   b) Marvell Discovery ethernet controller
-
-   The Discover ethernet controller is described with two levels
-   of nodes.  The first level describes an ethernet silicon block
-   and the second level describes up to 3 ethernet nodes within
-   that block.  The reason for the multiple levels is that the
-   registers for the node are interleaved within a single set
-   of registers.  The "ethernet-block" level describes the
-   shared register set, and the "ethernet" nodes describe ethernet
-   port-specific properties.
-
-   Ethernet block node
-
-   Required properties:
-     - #address-cells : <1>
-     - #size-cells : <0>
-     - compatible : "marvell,mv64360-eth-block"
-     - reg : Offset and length of the register set for this block
-
-   Example Discovery Ethernet block node:
-     ethernet-block@2000 {
-	     #address-cells = <1>;
-	     #size-cells = <0>;
-	     compatible = "marvell,mv64360-eth-block";
-	     reg = <0x2000 0x2000>;
-	     ethernet@0 {
-		     .......
-	     };
-     };
-
-   Ethernet port node
-
-   Required properties:
-     - device_type : Should be "network".
-     - compatible : Should be "marvell,mv64360-eth".
-     - reg : Should be <0>, <1>, or <2>, according to which registers
-       within the silicon block the device uses.
-     - interrupts : <a> where a is the interrupt number for the port.
-     - interrupt-parent : the phandle for the interrupt controller
-       that services interrupts for this device.
-     - phy : the phandle for the PHY connected to this ethernet
-       controller.
-     - local-mac-address : 6 bytes, MAC address
-
-   Example Discovery Ethernet port node:
-     ethernet@0 {
-	     device_type = "network";
-	     compatible = "marvell,mv64360-eth";
-	     reg = <0>;
-	     interrupts = <32>;
-	     interrupt-parent = <&PIC>;
-	     phy = <&PHY0>;
-	     local-mac-address = [ 00 00 00 00 00 00 ];
-     };
-
-
-
-   c) Marvell Discovery PHY nodes
-
-   Required properties:
-     - device_type : Should be "ethernet-phy"
-     - interrupts : <a> where a is the interrupt number for this phy.
-     - interrupt-parent : the phandle for the interrupt controller that
-       services interrupts for this device.
-     - reg : The ID number for the phy, usually a small integer
-
-   Example Discovery PHY node:
-     ethernet-phy@1 {
-	     device_type = "ethernet-phy";
-	     compatible = "broadcom,bcm5421";
-	     interrupts = <76>;      /* GPP 12 */
-	     interrupt-parent = <&PIC>;
-	     reg = <1>;
-     };
-
-
-   d) Marvell Discovery SDMA nodes
-
-   Represent DMA hardware associated with the MPSC (multiprotocol
-   serial controllers).
-
-   Required properties:
-     - compatible : "marvell,mv64360-sdma"
-     - reg : Offset and length of the register set for this device
-     - interrupts : <a> where a is the interrupt number for the DMA
-       device.
-     - interrupt-parent : the phandle for the interrupt controller
-       that services interrupts for this device.
-
-   Example Discovery SDMA node:
-     sdma@4000 {
-	     compatible = "marvell,mv64360-sdma";
-	     reg = <0x4000 0xc18>;
-	     virtual-reg = <0xf1004000>;
-	     interrupts = <36>;
-	     interrupt-parent = <&PIC>;
-     };
-
-
-   e) Marvell Discovery BRG nodes
-
-   Represent baud rate generator hardware associated with the MPSC
-   (multiprotocol serial controllers).
-
-   Required properties:
-     - compatible : "marvell,mv64360-brg"
-     - reg : Offset and length of the register set for this device
-     - clock-src : A value from 0 to 15 which selects the clock
-       source for the baud rate generator.  This value corresponds
-       to the CLKS value in the BRGx configuration register.  See
-       the mv64x60 User's Manual.
-     - clock-frequence : The frequency (in Hz) of the baud rate
-       generator's input clock.
-     - current-speed : The current speed setting (presumably by
-       firmware) of the baud rate generator.
-
-   Example Discovery BRG node:
-     brg@b200 {
-	     compatible = "marvell,mv64360-brg";
-	     reg = <0xb200 0x8>;
-	     clock-src = <8>;
-	     clock-frequency = <133333333>;
-	     current-speed = <9600>;
-     };
-
-
-   f) Marvell Discovery CUNIT nodes
-
-   Represent the Serial Communications Unit device hardware.
-
-   Required properties:
-     - reg : Offset and length of the register set for this device
-
-   Example Discovery CUNIT node:
-     cunit@f200 {
-	     reg = <0xf200 0x200>;
-     };
-
-
-   g) Marvell Discovery MPSCROUTING nodes
-
-   Represent the Discovery's MPSC routing hardware
-
-   Required properties:
-     - reg : Offset and length of the register set for this device
-
-   Example Discovery CUNIT node:
-     mpscrouting@b500 {
-	     reg = <0xb400 0xc>;
-     };
-
-
-   h) Marvell Discovery MPSCINTR nodes
-
-   Represent the Discovery's MPSC DMA interrupt hardware registers
-   (SDMA cause and mask registers).
-
-   Required properties:
-     - reg : Offset and length of the register set for this device
-
-   Example Discovery MPSCINTR node:
-     mpsintr@b800 {
-	     reg = <0xb800 0x100>;
-     };
-
-
-   i) Marvell Discovery MPSC nodes
-
-   Represent the Discovery's MPSC (Multiprotocol Serial Controller)
-   serial port.
-
-   Required properties:
-     - device_type : "serial"
-     - compatible : "marvell,mv64360-mpsc"
-     - reg : Offset and length of the register set for this device
-     - sdma : the phandle for the SDMA node used by this port
-     - brg : the phandle for the BRG node used by this port
-     - cunit : the phandle for the CUNIT node used by this port
-     - mpscrouting : the phandle for the MPSCROUTING node used by this port
-     - mpscintr : the phandle for the MPSCINTR node used by this port
-     - cell-index : the hardware index of this cell in the MPSC core
-     - max_idle : value needed for MPSC CHR3 (Maximum Frame Length)
-       register
-     - interrupts : <a> where a is the interrupt number for the MPSC.
-     - interrupt-parent : the phandle for the interrupt controller
-       that services interrupts for this device.
-
-   Example Discovery MPSCINTR node:
-     mpsc@8000 {
-	     device_type = "serial";
-	     compatible = "marvell,mv64360-mpsc";
-	     reg = <0x8000 0x38>;
-	     virtual-reg = <0xf1008000>;
-	     sdma = <&SDMA0>;
-	     brg = <&BRG0>;
-	     cunit = <&CUNIT>;
-	     mpscrouting = <&MPSCROUTING>;
-	     mpscintr = <&MPSCINTR>;
-	     cell-index = <0>;
-	     max_idle = <40>;
-	     interrupts = <40>;
-	     interrupt-parent = <&PIC>;
-     };
-
-
-   j) Marvell Discovery Watch Dog Timer nodes
-
-   Represent the Discovery's watchdog timer hardware
-
-   Required properties:
-     - compatible : "marvell,mv64360-wdt"
-     - reg : Offset and length of the register set for this device
-
-   Example Discovery Watch Dog Timer node:
-     wdt@b410 {
-	     compatible = "marvell,mv64360-wdt";
-	     reg = <0xb410 0x8>;
-     };
-
-
-   k) Marvell Discovery I2C nodes
-
-   Represent the Discovery's I2C hardware
-
-   Required properties:
-     - device_type : "i2c"
-     - compatible : "marvell,mv64360-i2c"
-     - reg : Offset and length of the register set for this device
-     - interrupts : <a> where a is the interrupt number for the I2C.
-     - interrupt-parent : the phandle for the interrupt controller
-       that services interrupts for this device.
-
-   Example Discovery I2C node:
-	     compatible = "marvell,mv64360-i2c";
-	     reg = <0xc000 0x20>;
-	     virtual-reg = <0xf100c000>;
-	     interrupts = <37>;
-	     interrupt-parent = <&PIC>;
-     };
-
-
-   l) Marvell Discovery PIC (Programmable Interrupt Controller) nodes
-
-   Represent the Discovery's PIC hardware
-
-   Required properties:
-     - #interrupt-cells : <1>
-     - #address-cells : <0>
-     - compatible : "marvell,mv64360-pic"
-     - reg : Offset and length of the register set for this device
-     - interrupt-controller
-
-   Example Discovery PIC node:
-     pic {
-	     #interrupt-cells = <1>;
-	     #address-cells = <0>;
-	     compatible = "marvell,mv64360-pic";
-	     reg = <0x0 0x88>;
-	     interrupt-controller;
-     };
-
-
-   m) Marvell Discovery MPP (Multipurpose Pins) multiplexing nodes
-
-   Represent the Discovery's MPP hardware
-
-   Required properties:
-     - compatible : "marvell,mv64360-mpp"
-     - reg : Offset and length of the register set for this device
-
-   Example Discovery MPP node:
-     mpp@f000 {
-	     compatible = "marvell,mv64360-mpp";
-	     reg = <0xf000 0x10>;
-     };
-
-
-   n) Marvell Discovery GPP (General Purpose Pins) nodes
-
-   Represent the Discovery's GPP hardware
-
-   Required properties:
-     - compatible : "marvell,mv64360-gpp"
-     - reg : Offset and length of the register set for this device
-
-   Example Discovery GPP node:
-     gpp@f000 {
-	     compatible = "marvell,mv64360-gpp";
-	     reg = <0xf100 0x20>;
-     };
-
-
-   o) Marvell Discovery PCI host bridge node
-
-   Represents the Discovery's PCI host bridge device.  The properties
-   for this node conform to Rev 2.1 of the PCI Bus Binding to IEEE
-   1275-1994.  A typical value for the compatible property is
-   "marvell,mv64360-pci".
-
-   Example Discovery PCI host bridge node
-     pci@80000000 {
-	     #address-cells = <3>;
-	     #size-cells = <2>;
-	     #interrupt-cells = <1>;
-	     device_type = "pci";
-	     compatible = "marvell,mv64360-pci";
-	     reg = <0xcf8 0x8>;
-	     ranges = <0x01000000 0x0        0x0
-			     0x88000000 0x0 0x01000000
-		       0x02000000 0x0 0x80000000
-			     0x80000000 0x0 0x08000000>;
-	     bus-range = <0 255>;
-	     clock-frequency = <66000000>;
-	     interrupt-parent = <&PIC>;
-	     interrupt-map-mask = <0xf800 0x0 0x0 0x7>;
-	     interrupt-map = <
-		     /* IDSEL 0x0a */
-		     0x5000 0 0 1 &PIC 80
-		     0x5000 0 0 2 &PIC 81
-		     0x5000 0 0 3 &PIC 91
-		     0x5000 0 0 4 &PIC 93
-
-		     /* IDSEL 0x0b */
-		     0x5800 0 0 1 &PIC 91
-		     0x5800 0 0 2 &PIC 93
-		     0x5800 0 0 3 &PIC 80
-		     0x5800 0 0 4 &PIC 81
-
-		     /* IDSEL 0x0c */
-		     0x6000 0 0 1 &PIC 91
-		     0x6000 0 0 2 &PIC 93
-		     0x6000 0 0 3 &PIC 80
-		     0x6000 0 0 4 &PIC 81
-
-		     /* IDSEL 0x0d */
-		     0x6800 0 0 1 &PIC 93
-		     0x6800 0 0 2 &PIC 80
-		     0x6800 0 0 3 &PIC 81
-		     0x6800 0 0 4 &PIC 91
-	     >;
-     };
-
-
-   p) Marvell Discovery CPU Error nodes
-
-   Represent the Discovery's CPU error handler device.
-
-   Required properties:
-     - compatible : "marvell,mv64360-cpu-error"
-     - reg : Offset and length of the register set for this device
-     - interrupts : the interrupt number for this device
-     - interrupt-parent : the phandle for the interrupt controller
-       that services interrupts for this device.
-
-   Example Discovery CPU Error node:
-     cpu-error@0070 {
-	     compatible = "marvell,mv64360-cpu-error";
-	     reg = <0x70 0x10 0x128 0x28>;
-	     interrupts = <3>;
-	     interrupt-parent = <&PIC>;
-     };
-
-
-   q) Marvell Discovery SRAM Controller nodes
-
-   Represent the Discovery's SRAM controller device.
-
-   Required properties:
-     - compatible : "marvell,mv64360-sram-ctrl"
-     - reg : Offset and length of the register set for this device
-     - interrupts : the interrupt number for this device
-     - interrupt-parent : the phandle for the interrupt controller
-       that services interrupts for this device.
-
-   Example Discovery SRAM Controller node:
-     sram-ctrl@0380 {
-	     compatible = "marvell,mv64360-sram-ctrl";
-	     reg = <0x380 0x80>;
-	     interrupts = <13>;
-	     interrupt-parent = <&PIC>;
-     };
-
-
-   r) Marvell Discovery PCI Error Handler nodes
-
-   Represent the Discovery's PCI error handler device.
-
-   Required properties:
-     - compatible : "marvell,mv64360-pci-error"
-     - reg : Offset and length of the register set for this device
-     - interrupts : the interrupt number for this device
-     - interrupt-parent : the phandle for the interrupt controller
-       that services interrupts for this device.
-
-   Example Discovery PCI Error Handler node:
-     pci-error@1d40 {
-	     compatible = "marvell,mv64360-pci-error";
-	     reg = <0x1d40 0x40 0xc28 0x4>;
-	     interrupts = <12>;
-	     interrupt-parent = <&PIC>;
-     };
-
-
-   s) Marvell Discovery Memory Controller nodes
-
-   Represent the Discovery's memory controller device.
-
-   Required properties:
-     - compatible : "marvell,mv64360-mem-ctrl"
-     - reg : Offset and length of the register set for this device
-     - interrupts : the interrupt number for this device
-     - interrupt-parent : the phandle for the interrupt controller
-       that services interrupts for this device.
-
-   Example Discovery Memory Controller node:
-     mem-ctrl@1400 {
-	     compatible = "marvell,mv64360-mem-ctrl";
-	     reg = <0x1400 0x60>;
-	     interrupts = <17>;
-	     interrupt-parent = <&PIC>;
-     };
-
-
-VIII - Specifying interrupt information for devices
+VII - Specifying interrupt information for devices
 ===================================================
 
 The device tree represents the busses and devices of a hardware
@@ -2439,56 +1324,7 @@ encodings listed below:
 	2 =  high to low edge sensitive type enabled
 	3 =  low to high edge sensitive type enabled
 
-IX - Specifying GPIO information for devices
-============================================
-
-1) gpios property
------------------
-
-Nodes that makes use of GPIOs should define them using `gpios' property,
-format of which is: <&gpio-controller1-phandle gpio1-specifier
-		     &gpio-controller2-phandle gpio2-specifier
-		     0 /* holes are permitted, means no GPIO 3 */
-		     &gpio-controller4-phandle gpio4-specifier
-		     ...>;
-
-Note that gpio-specifier length is controller dependent.
-
-gpio-specifier may encode: bank, pin position inside the bank,
-whether pin is open-drain and whether pin is logically inverted.
-
-Example of the node using GPIOs:
-
-	node {
-		gpios = <&qe_pio_e 18 0>;
-	};
-
-In this example gpio-specifier is "18 0" and encodes GPIO pin number,
-and empty GPIO flags as accepted by the "qe_pio_e" gpio-controller.
-
-2) gpio-controller nodes
-------------------------
-
-Every GPIO controller node must have #gpio-cells property defined,
-this information will be used to translate gpio-specifiers.
-
-Example of two SOC GPIO banks defined as gpio-controller nodes:
-
-	qe_pio_a: gpio-controller@1400 {
-		#gpio-cells = <2>;
-		compatible = "fsl,qe-pario-bank-a", "fsl,qe-pario-bank";
-		reg = <0x1400 0x18>;
-		gpio-controller;
-	};
-
-	qe_pio_e: gpio-controller@1460 {
-		#gpio-cells = <2>;
-		compatible = "fsl,qe-pario-bank-e", "fsl,qe-pario-bank";
-		reg = <0x1460 0x18>;
-		gpio-controller;
-	};
-
-X - Specifying Device Power Management Information (sleep property)
+VIII - Specifying Device Power Management Information (sleep property)
 ===================================================================
 
 Devices on SOCs often have mechanisms for placing devices into low-power
diff --git a/Documentation/powerpc/dts-bindings/4xx/emac.txt b/Documentation/powerpc/dts-bindings/4xx/emac.txt
new file mode 100644
index 000000000000..2161334a7ca5
--- /dev/null
+++ b/Documentation/powerpc/dts-bindings/4xx/emac.txt
@@ -0,0 +1,148 @@
+    4xx/Axon EMAC ethernet nodes
+
+    The EMAC ethernet controller in IBM and AMCC 4xx chips, and also
+    the Axon bridge.  To operate this needs to interact with a ths
+    special McMAL DMA controller, and sometimes an RGMII or ZMII
+    interface.  In addition to the nodes and properties described
+    below, the node for the OPB bus on which the EMAC sits must have a
+    correct clock-frequency property.
+
+      i) The EMAC node itself
+
+    Required properties:
+    - device_type       : "network"
+
+    - compatible        : compatible list, contains 2 entries, first is
+			  "ibm,emac-CHIP" where CHIP is the host ASIC (440gx,
+			  405gp, Axon) and second is either "ibm,emac" or
+			  "ibm,emac4".  For Axon, thus, we have: "ibm,emac-axon",
+			  "ibm,emac4"
+    - interrupts        : <interrupt mapping for EMAC IRQ and WOL IRQ>
+    - interrupt-parent  : optional, if needed for interrupt mapping
+    - reg               : <registers mapping>
+    - local-mac-address : 6 bytes, MAC address
+    - mal-device        : phandle of the associated McMAL node
+    - mal-tx-channel    : 1 cell, index of the tx channel on McMAL associated
+			  with this EMAC
+    - mal-rx-channel    : 1 cell, index of the rx channel on McMAL associated
+			  with this EMAC
+    - cell-index        : 1 cell, hardware index of the EMAC cell on a given
+			  ASIC (typically 0x0 and 0x1 for EMAC0 and EMAC1 on
+			  each Axon chip)
+    - max-frame-size    : 1 cell, maximum frame size supported in bytes
+    - rx-fifo-size      : 1 cell, Rx fifo size in bytes for 10 and 100 Mb/sec
+			  operations.
+			  For Axon, 2048
+    - tx-fifo-size      : 1 cell, Tx fifo size in bytes for 10 and 100 Mb/sec
+			  operations.
+			  For Axon, 2048.
+    - fifo-entry-size   : 1 cell, size of a fifo entry (used to calculate
+			  thresholds).
+			  For Axon, 0x00000010
+    - mal-burst-size    : 1 cell, MAL burst size (used to calculate thresholds)
+			  in bytes.
+			  For Axon, 0x00000100 (I think ...)
+    - phy-mode          : string, mode of operations of the PHY interface.
+			  Supported values are: "mii", "rmii", "smii", "rgmii",
+			  "tbi", "gmii", rtbi", "sgmii".
+			  For Axon on CAB, it is "rgmii"
+    - mdio-device       : 1 cell, required iff using shared MDIO registers
+			  (440EP).  phandle of the EMAC to use to drive the
+			  MDIO lines for the PHY used by this EMAC.
+    - zmii-device       : 1 cell, required iff connected to a ZMII.  phandle of
+			  the ZMII device node
+    - zmii-channel      : 1 cell, required iff connected to a ZMII.  Which ZMII
+			  channel or 0xffffffff if ZMII is only used for MDIO.
+    - rgmii-device      : 1 cell, required iff connected to an RGMII. phandle
+			  of the RGMII device node.
+			  For Axon: phandle of plb5/plb4/opb/rgmii
+    - rgmii-channel     : 1 cell, required iff connected to an RGMII.  Which
+			  RGMII channel is used by this EMAC.
+			  Fox Axon: present, whatever value is appropriate for each
+			  EMAC, that is the content of the current (bogus) "phy-port"
+			  property.
+
+    Optional properties:
+    - phy-address       : 1 cell, optional, MDIO address of the PHY. If absent,
+			  a search is performed.
+    - phy-map           : 1 cell, optional, bitmap of addresses to probe the PHY
+			  for, used if phy-address is absent. bit 0x00000001 is
+			  MDIO address 0.
+			  For Axon it can be absent, though my current driver
+			  doesn't handle phy-address yet so for now, keep
+			  0x00ffffff in it.
+    - rx-fifo-size-gige : 1 cell, Rx fifo size in bytes for 1000 Mb/sec
+			  operations (if absent the value is the same as
+			  rx-fifo-size).  For Axon, either absent or 2048.
+    - tx-fifo-size-gige : 1 cell, Tx fifo size in bytes for 1000 Mb/sec
+			  operations (if absent the value is the same as
+			  tx-fifo-size). For Axon, either absent or 2048.
+    - tah-device        : 1 cell, optional. If connected to a TAH engine for
+			  offload, phandle of the TAH device node.
+    - tah-channel       : 1 cell, optional. If appropriate, channel used on the
+			  TAH engine.
+
+    Example:
+
+	EMAC0: ethernet@40000800 {
+		device_type = "network";
+		compatible = "ibm,emac-440gp", "ibm,emac";
+		interrupt-parent = <&UIC1>;
+		interrupts = <1c 4 1d 4>;
+		reg = <40000800 70>;
+		local-mac-address = [00 04 AC E3 1B 1E];
+		mal-device = <&MAL0>;
+		mal-tx-channel = <0 1>;
+		mal-rx-channel = <0>;
+		cell-index = <0>;
+		max-frame-size = <5dc>;
+		rx-fifo-size = <1000>;
+		tx-fifo-size = <800>;
+		phy-mode = "rmii";
+		phy-map = <00000001>;
+		zmii-device = <&ZMII0>;
+		zmii-channel = <0>;
+	};
+
+      ii) McMAL node
+
+    Required properties:
+    - device_type        : "dma-controller"
+    - compatible         : compatible list, containing 2 entries, first is
+			   "ibm,mcmal-CHIP" where CHIP is the host ASIC (like
+			   emac) and the second is either "ibm,mcmal" or
+			   "ibm,mcmal2".
+			   For Axon, "ibm,mcmal-axon","ibm,mcmal2"
+    - interrupts         : <interrupt mapping for the MAL interrupts sources:
+                           5 sources: tx_eob, rx_eob, serr, txde, rxde>.
+                           For Axon: This is _different_ from the current
+			   firmware.  We use the "delayed" interrupts for txeob
+			   and rxeob. Thus we end up with mapping those 5 MPIC
+			   interrupts, all level positive sensitive: 10, 11, 32,
+			   33, 34 (in decimal)
+    - dcr-reg            : < DCR registers range >
+    - dcr-parent         : if needed for dcr-reg
+    - num-tx-chans       : 1 cell, number of Tx channels
+    - num-rx-chans       : 1 cell, number of Rx channels
+
+      iii) ZMII node
+
+    Required properties:
+    - compatible         : compatible list, containing 2 entries, first is
+			   "ibm,zmii-CHIP" where CHIP is the host ASIC (like
+			   EMAC) and the second is "ibm,zmii".
+			   For Axon, there is no ZMII node.
+    - reg                : <registers mapping>
+
+      iv) RGMII node
+
+    Required properties:
+    - compatible         : compatible list, containing 2 entries, first is
+			   "ibm,rgmii-CHIP" where CHIP is the host ASIC (like
+			   EMAC) and the second is "ibm,rgmii".
+                           For Axon, "ibm,rgmii-axon","ibm,rgmii"
+    - reg                : <registers mapping>
+    - revision           : as provided by the RGMII new version register if
+			   available.
+			   For Axon: 0x0000012a
+
diff --git a/Documentation/powerpc/dts-bindings/gpio/gpio.txt b/Documentation/powerpc/dts-bindings/gpio/gpio.txt
new file mode 100644
index 000000000000..edaa84d288a1
--- /dev/null
+++ b/Documentation/powerpc/dts-bindings/gpio/gpio.txt
@@ -0,0 +1,50 @@
+Specifying GPIO information for devices
+============================================
+
+1) gpios property
+-----------------
+
+Nodes that makes use of GPIOs should define them using `gpios' property,
+format of which is: <&gpio-controller1-phandle gpio1-specifier
+		     &gpio-controller2-phandle gpio2-specifier
+		     0 /* holes are permitted, means no GPIO 3 */
+		     &gpio-controller4-phandle gpio4-specifier
+		     ...>;
+
+Note that gpio-specifier length is controller dependent.
+
+gpio-specifier may encode: bank, pin position inside the bank,
+whether pin is open-drain and whether pin is logically inverted.
+
+Example of the node using GPIOs:
+
+	node {
+		gpios = <&qe_pio_e 18 0>;
+	};
+
+In this example gpio-specifier is "18 0" and encodes GPIO pin number,
+and empty GPIO flags as accepted by the "qe_pio_e" gpio-controller.
+
+2) gpio-controller nodes
+------------------------
+
+Every GPIO controller node must have #gpio-cells property defined,
+this information will be used to translate gpio-specifiers.
+
+Example of two SOC GPIO banks defined as gpio-controller nodes:
+
+	qe_pio_a: gpio-controller@1400 {
+		#gpio-cells = <2>;
+		compatible = "fsl,qe-pario-bank-a", "fsl,qe-pario-bank";
+		reg = <0x1400 0x18>;
+		gpio-controller;
+	};
+
+	qe_pio_e: gpio-controller@1460 {
+		#gpio-cells = <2>;
+		compatible = "fsl,qe-pario-bank-e", "fsl,qe-pario-bank";
+		reg = <0x1460 0x18>;
+		gpio-controller;
+	};
+
+
diff --git a/Documentation/powerpc/dts-bindings/gpio/mdio.txt b/Documentation/powerpc/dts-bindings/gpio/mdio.txt
new file mode 100644
index 000000000000..bc9549529014
--- /dev/null
+++ b/Documentation/powerpc/dts-bindings/gpio/mdio.txt
@@ -0,0 +1,19 @@
+MDIO on GPIOs
+
+Currently defined compatibles:
+- virtual,gpio-mdio
+
+MDC and MDIO lines connected to GPIO controllers are listed in the
+gpios property as described in section VIII.1 in the following order:
+
+MDC, MDIO.
+
+Example:
+
+mdio {
+	compatible = "virtual,mdio-gpio";
+	#address-cells = <1>;
+	#size-cells = <0>;
+	gpios = <&qe_pio_a 11
+		 &qe_pio_c 6>;
+};
diff --git a/Documentation/powerpc/dts-bindings/marvell.txt b/Documentation/powerpc/dts-bindings/marvell.txt
new file mode 100644
index 000000000000..3708a2fd4747
--- /dev/null
+++ b/Documentation/powerpc/dts-bindings/marvell.txt
@@ -0,0 +1,521 @@
+Marvell Discovery mv64[345]6x System Controller chips
+===========================================================
+
+The Marvell mv64[345]60 series of system controller chips contain
+many of the peripherals needed to implement a complete computer
+system.  In this section, we define device tree nodes to describe
+the system controller chip itself and each of the peripherals
+which it contains.  Compatible string values for each node are
+prefixed with the string "marvell,", for Marvell Technology Group Ltd.
+
+1) The /system-controller node
+
+  This node is used to represent the system-controller and must be
+  present when the system uses a system controller chip. The top-level
+  system-controller node contains information that is global to all
+  devices within the system controller chip. The node name begins
+  with "system-controller" followed by the unit address, which is
+  the base address of the memory-mapped register set for the system
+  controller chip.
+
+  Required properties:
+
+    - ranges : Describes the translation of system controller addresses
+      for memory mapped registers.
+    - clock-frequency: Contains the main clock frequency for the system
+      controller chip.
+    - reg : This property defines the address and size of the
+      memory-mapped registers contained within the system controller
+      chip.  The address specified in the "reg" property should match
+      the unit address of the system-controller node.
+    - #address-cells : Address representation for system controller
+      devices.  This field represents the number of cells needed to
+      represent the address of the memory-mapped registers of devices
+      within the system controller chip.
+    - #size-cells : Size representation for for the memory-mapped
+      registers within the system controller chip.
+    - #interrupt-cells : Defines the width of cells used to represent
+      interrupts.
+
+  Optional properties:
+
+    - model : The specific model of the system controller chip.  Such
+      as, "mv64360", "mv64460", or "mv64560".
+    - compatible : A string identifying the compatibility identifiers
+      of the system controller chip.
+
+  The system-controller node contains child nodes for each system
+  controller device that the platform uses.  Nodes should not be created
+  for devices which exist on the system controller chip but are not used
+
+  Example Marvell Discovery mv64360 system-controller node:
+
+    system-controller@f1000000 { /* Marvell Discovery mv64360 */
+	    #address-cells = <1>;
+	    #size-cells = <1>;
+	    model = "mv64360";                      /* Default */
+	    compatible = "marvell,mv64360";
+	    clock-frequency = <133333333>;
+	    reg = <0xf1000000 0x10000>;
+	    virtual-reg = <0xf1000000>;
+	    ranges = <0x88000000 0x88000000 0x1000000 /* PCI 0 I/O Space */
+		    0x80000000 0x80000000 0x8000000 /* PCI 0 MEM Space */
+		    0xa0000000 0xa0000000 0x4000000 /* User FLASH */
+		    0x00000000 0xf1000000 0x0010000 /* Bridge's regs */
+		    0xf2000000 0xf2000000 0x0040000>;/* Integrated SRAM */
+
+	    [ child node definitions... ]
+    }
+
+2) Child nodes of /system-controller
+
+   a) Marvell Discovery MDIO bus
+
+   The MDIO is a bus to which the PHY devices are connected.  For each
+   device that exists on this bus, a child node should be created.  See
+   the definition of the PHY node below for an example of how to define
+   a PHY.
+
+   Required properties:
+     - #address-cells : Should be <1>
+     - #size-cells : Should be <0>
+     - device_type : Should be "mdio"
+     - compatible : Should be "marvell,mv64360-mdio"
+
+   Example:
+
+     mdio {
+	     #address-cells = <1>;
+	     #size-cells = <0>;
+	     device_type = "mdio";
+	     compatible = "marvell,mv64360-mdio";
+
+	     ethernet-phy@0 {
+		     ......
+	     };
+     };
+
+
+   b) Marvell Discovery ethernet controller
+
+   The Discover ethernet controller is described with two levels
+   of nodes.  The first level describes an ethernet silicon block
+   and the second level describes up to 3 ethernet nodes within
+   that block.  The reason for the multiple levels is that the
+   registers for the node are interleaved within a single set
+   of registers.  The "ethernet-block" level describes the
+   shared register set, and the "ethernet" nodes describe ethernet
+   port-specific properties.
+
+   Ethernet block node
+
+   Required properties:
+     - #address-cells : <1>
+     - #size-cells : <0>
+     - compatible : "marvell,mv64360-eth-block"
+     - reg : Offset and length of the register set for this block
+
+   Example Discovery Ethernet block node:
+     ethernet-block@2000 {
+	     #address-cells = <1>;
+	     #size-cells = <0>;
+	     compatible = "marvell,mv64360-eth-block";
+	     reg = <0x2000 0x2000>;
+	     ethernet@0 {
+		     .......
+	     };
+     };
+
+   Ethernet port node
+
+   Required properties:
+     - device_type : Should be "network".
+     - compatible : Should be "marvell,mv64360-eth".
+     - reg : Should be <0>, <1>, or <2>, according to which registers
+       within the silicon block the device uses.
+     - interrupts : <a> where a is the interrupt number for the port.
+     - interrupt-parent : the phandle for the interrupt controller
+       that services interrupts for this device.
+     - phy : the phandle for the PHY connected to this ethernet
+       controller.
+     - local-mac-address : 6 bytes, MAC address
+
+   Example Discovery Ethernet port node:
+     ethernet@0 {
+	     device_type = "network";
+	     compatible = "marvell,mv64360-eth";
+	     reg = <0>;
+	     interrupts = <32>;
+	     interrupt-parent = <&PIC>;
+	     phy = <&PHY0>;
+	     local-mac-address = [ 00 00 00 00 00 00 ];
+     };
+
+
+
+   c) Marvell Discovery PHY nodes
+
+   Required properties:
+     - device_type : Should be "ethernet-phy"
+     - interrupts : <a> where a is the interrupt number for this phy.
+     - interrupt-parent : the phandle for the interrupt controller that
+       services interrupts for this device.
+     - reg : The ID number for the phy, usually a small integer
+
+   Example Discovery PHY node:
+     ethernet-phy@1 {
+	     device_type = "ethernet-phy";
+	     compatible = "broadcom,bcm5421";
+	     interrupts = <76>;      /* GPP 12 */
+	     interrupt-parent = <&PIC>;
+	     reg = <1>;
+     };
+
+
+   d) Marvell Discovery SDMA nodes
+
+   Represent DMA hardware associated with the MPSC (multiprotocol
+   serial controllers).
+
+   Required properties:
+     - compatible : "marvell,mv64360-sdma"
+     - reg : Offset and length of the register set for this device
+     - interrupts : <a> where a is the interrupt number for the DMA
+       device.
+     - interrupt-parent : the phandle for the interrupt controller
+       that services interrupts for this device.
+
+   Example Discovery SDMA node:
+     sdma@4000 {
+	     compatible = "marvell,mv64360-sdma";
+	     reg = <0x4000 0xc18>;
+	     virtual-reg = <0xf1004000>;
+	     interrupts = <36>;
+	     interrupt-parent = <&PIC>;
+     };
+
+
+   e) Marvell Discovery BRG nodes
+
+   Represent baud rate generator hardware associated with the MPSC
+   (multiprotocol serial controllers).
+
+   Required properties:
+     - compatible : "marvell,mv64360-brg"
+     - reg : Offset and length of the register set for this device
+     - clock-src : A value from 0 to 15 which selects the clock
+       source for the baud rate generator.  This value corresponds
+       to the CLKS value in the BRGx configuration register.  See
+       the mv64x60 User's Manual.
+     - clock-frequence : The frequency (in Hz) of the baud rate
+       generator's input clock.
+     - current-speed : The current speed setting (presumably by
+       firmware) of the baud rate generator.
+
+   Example Discovery BRG node:
+     brg@b200 {
+	     compatible = "marvell,mv64360-brg";
+	     reg = <0xb200 0x8>;
+	     clock-src = <8>;
+	     clock-frequency = <133333333>;
+	     current-speed = <9600>;
+     };
+
+
+   f) Marvell Discovery CUNIT nodes
+
+   Represent the Serial Communications Unit device hardware.
+
+   Required properties:
+     - reg : Offset and length of the register set for this device
+
+   Example Discovery CUNIT node:
+     cunit@f200 {
+	     reg = <0xf200 0x200>;
+     };
+
+
+   g) Marvell Discovery MPSCROUTING nodes
+
+   Represent the Discovery's MPSC routing hardware
+
+   Required properties:
+     - reg : Offset and length of the register set for this device
+
+   Example Discovery CUNIT node:
+     mpscrouting@b500 {
+	     reg = <0xb400 0xc>;
+     };
+
+
+   h) Marvell Discovery MPSCINTR nodes
+
+   Represent the Discovery's MPSC DMA interrupt hardware registers
+   (SDMA cause and mask registers).
+
+   Required properties:
+     - reg : Offset and length of the register set for this device
+
+   Example Discovery MPSCINTR node:
+     mpsintr@b800 {
+	     reg = <0xb800 0x100>;
+     };
+
+
+   i) Marvell Discovery MPSC nodes
+
+   Represent the Discovery's MPSC (Multiprotocol Serial Controller)
+   serial port.
+
+   Required properties:
+     - device_type : "serial"
+     - compatible : "marvell,mv64360-mpsc"
+     - reg : Offset and length of the register set for this device
+     - sdma : the phandle for the SDMA node used by this port
+     - brg : the phandle for the BRG node used by this port
+     - cunit : the phandle for the CUNIT node used by this port
+     - mpscrouting : the phandle for the MPSCROUTING node used by this port
+     - mpscintr : the phandle for the MPSCINTR node used by this port
+     - cell-index : the hardware index of this cell in the MPSC core
+     - max_idle : value needed for MPSC CHR3 (Maximum Frame Length)
+       register
+     - interrupts : <a> where a is the interrupt number for the MPSC.
+     - interrupt-parent : the phandle for the interrupt controller
+       that services interrupts for this device.
+
+   Example Discovery MPSCINTR node:
+     mpsc@8000 {
+	     device_type = "serial";
+	     compatible = "marvell,mv64360-mpsc";
+	     reg = <0x8000 0x38>;
+	     virtual-reg = <0xf1008000>;
+	     sdma = <&SDMA0>;
+	     brg = <&BRG0>;
+	     cunit = <&CUNIT>;
+	     mpscrouting = <&MPSCROUTING>;
+	     mpscintr = <&MPSCINTR>;
+	     cell-index = <0>;
+	     max_idle = <40>;
+	     interrupts = <40>;
+	     interrupt-parent = <&PIC>;
+     };
+
+
+   j) Marvell Discovery Watch Dog Timer nodes
+
+   Represent the Discovery's watchdog timer hardware
+
+   Required properties:
+     - compatible : "marvell,mv64360-wdt"
+     - reg : Offset and length of the register set for this device
+
+   Example Discovery Watch Dog Timer node:
+     wdt@b410 {
+	     compatible = "marvell,mv64360-wdt";
+	     reg = <0xb410 0x8>;
+     };
+
+
+   k) Marvell Discovery I2C nodes
+
+   Represent the Discovery's I2C hardware
+
+   Required properties:
+     - device_type : "i2c"
+     - compatible : "marvell,mv64360-i2c"
+     - reg : Offset and length of the register set for this device
+     - interrupts : <a> where a is the interrupt number for the I2C.
+     - interrupt-parent : the phandle for the interrupt controller
+       that services interrupts for this device.
+
+   Example Discovery I2C node:
+	     compatible = "marvell,mv64360-i2c";
+	     reg = <0xc000 0x20>;
+	     virtual-reg = <0xf100c000>;
+	     interrupts = <37>;
+	     interrupt-parent = <&PIC>;
+     };
+
+
+   l) Marvell Discovery PIC (Programmable Interrupt Controller) nodes
+
+   Represent the Discovery's PIC hardware
+
+   Required properties:
+     - #interrupt-cells : <1>
+     - #address-cells : <0>
+     - compatible : "marvell,mv64360-pic"
+     - reg : Offset and length of the register set for this device
+     - interrupt-controller
+
+   Example Discovery PIC node:
+     pic {
+	     #interrupt-cells = <1>;
+	     #address-cells = <0>;
+	     compatible = "marvell,mv64360-pic";
+	     reg = <0x0 0x88>;
+	     interrupt-controller;
+     };
+
+
+   m) Marvell Discovery MPP (Multipurpose Pins) multiplexing nodes
+
+   Represent the Discovery's MPP hardware
+
+   Required properties:
+     - compatible : "marvell,mv64360-mpp"
+     - reg : Offset and length of the register set for this device
+
+   Example Discovery MPP node:
+     mpp@f000 {
+	     compatible = "marvell,mv64360-mpp";
+	     reg = <0xf000 0x10>;
+     };
+
+
+   n) Marvell Discovery GPP (General Purpose Pins) nodes
+
+   Represent the Discovery's GPP hardware
+
+   Required properties:
+     - compatible : "marvell,mv64360-gpp"
+     - reg : Offset and length of the register set for this device
+
+   Example Discovery GPP node:
+     gpp@f000 {
+	     compatible = "marvell,mv64360-gpp";
+	     reg = <0xf100 0x20>;
+     };
+
+
+   o) Marvell Discovery PCI host bridge node
+
+   Represents the Discovery's PCI host bridge device.  The properties
+   for this node conform to Rev 2.1 of the PCI Bus Binding to IEEE
+   1275-1994.  A typical value for the compatible property is
+   "marvell,mv64360-pci".
+
+   Example Discovery PCI host bridge node
+     pci@80000000 {
+	     #address-cells = <3>;
+	     #size-cells = <2>;
+	     #interrupt-cells = <1>;
+	     device_type = "pci";
+	     compatible = "marvell,mv64360-pci";
+	     reg = <0xcf8 0x8>;
+	     ranges = <0x01000000 0x0        0x0
+			     0x88000000 0x0 0x01000000
+		       0x02000000 0x0 0x80000000
+			     0x80000000 0x0 0x08000000>;
+	     bus-range = <0 255>;
+	     clock-frequency = <66000000>;
+	     interrupt-parent = <&PIC>;
+	     interrupt-map-mask = <0xf800 0x0 0x0 0x7>;
+	     interrupt-map = <
+		     /* IDSEL 0x0a */
+		     0x5000 0 0 1 &PIC 80
+		     0x5000 0 0 2 &PIC 81
+		     0x5000 0 0 3 &PIC 91
+		     0x5000 0 0 4 &PIC 93
+
+		     /* IDSEL 0x0b */
+		     0x5800 0 0 1 &PIC 91
+		     0x5800 0 0 2 &PIC 93
+		     0x5800 0 0 3 &PIC 80
+		     0x5800 0 0 4 &PIC 81
+
+		     /* IDSEL 0x0c */
+		     0x6000 0 0 1 &PIC 91
+		     0x6000 0 0 2 &PIC 93
+		     0x6000 0 0 3 &PIC 80
+		     0x6000 0 0 4 &PIC 81
+
+		     /* IDSEL 0x0d */
+		     0x6800 0 0 1 &PIC 93
+		     0x6800 0 0 2 &PIC 80
+		     0x6800 0 0 3 &PIC 81
+		     0x6800 0 0 4 &PIC 91
+	     >;
+     };
+
+
+   p) Marvell Discovery CPU Error nodes
+
+   Represent the Discovery's CPU error handler device.
+
+   Required properties:
+     - compatible : "marvell,mv64360-cpu-error"
+     - reg : Offset and length of the register set for this device
+     - interrupts : the interrupt number for this device
+     - interrupt-parent : the phandle for the interrupt controller
+       that services interrupts for this device.
+
+   Example Discovery CPU Error node:
+     cpu-error@0070 {
+	     compatible = "marvell,mv64360-cpu-error";
+	     reg = <0x70 0x10 0x128 0x28>;
+	     interrupts = <3>;
+	     interrupt-parent = <&PIC>;
+     };
+
+
+   q) Marvell Discovery SRAM Controller nodes
+
+   Represent the Discovery's SRAM controller device.
+
+   Required properties:
+     - compatible : "marvell,mv64360-sram-ctrl"
+     - reg : Offset and length of the register set for this device
+     - interrupts : the interrupt number for this device
+     - interrupt-parent : the phandle for the interrupt controller
+       that services interrupts for this device.
+
+   Example Discovery SRAM Controller node:
+     sram-ctrl@0380 {
+	     compatible = "marvell,mv64360-sram-ctrl";
+	     reg = <0x380 0x80>;
+	     interrupts = <13>;
+	     interrupt-parent = <&PIC>;
+     };
+
+
+   r) Marvell Discovery PCI Error Handler nodes
+
+   Represent the Discovery's PCI error handler device.
+
+   Required properties:
+     - compatible : "marvell,mv64360-pci-error"
+     - reg : Offset and length of the register set for this device
+     - interrupts : the interrupt number for this device
+     - interrupt-parent : the phandle for the interrupt controller
+       that services interrupts for this device.
+
+   Example Discovery PCI Error Handler node:
+     pci-error@1d40 {
+	     compatible = "marvell,mv64360-pci-error";
+	     reg = <0x1d40 0x40 0xc28 0x4>;
+	     interrupts = <12>;
+	     interrupt-parent = <&PIC>;
+     };
+
+
+   s) Marvell Discovery Memory Controller nodes
+
+   Represent the Discovery's memory controller device.
+
+   Required properties:
+     - compatible : "marvell,mv64360-mem-ctrl"
+     - reg : Offset and length of the register set for this device
+     - interrupts : the interrupt number for this device
+     - interrupt-parent : the phandle for the interrupt controller
+       that services interrupts for this device.
+
+   Example Discovery Memory Controller node:
+     mem-ctrl@1400 {
+	     compatible = "marvell,mv64360-mem-ctrl";
+	     reg = <0x1400 0x60>;
+	     interrupts = <17>;
+	     interrupt-parent = <&PIC>;
+     };
+
+
diff --git a/Documentation/powerpc/dts-bindings/phy.txt b/Documentation/powerpc/dts-bindings/phy.txt
new file mode 100644
index 000000000000..bb8c742eb8c5
--- /dev/null
+++ b/Documentation/powerpc/dts-bindings/phy.txt
@@ -0,0 +1,25 @@
+PHY nodes
+
+Required properties:
+
+ - device_type : Should be "ethernet-phy"
+ - interrupts : <a b> where a is the interrupt number and b is a
+   field that represents an encoding of the sense and level
+   information for the interrupt.  This should be encoded based on
+   the information in section 2) depending on the type of interrupt
+   controller you have.
+ - interrupt-parent : the phandle for the interrupt controller that
+   services interrupts for this device.
+ - reg : The ID number for the phy, usually a small integer
+ - linux,phandle :  phandle for this node; likely referenced by an
+   ethernet controller node.
+
+Example:
+
+ethernet-phy@0 {
+	linux,phandle = <2452000>
+	interrupt-parent = <40000>;
+	interrupts = <35 1>;
+	reg = <0>;
+	device_type = "ethernet-phy";
+};
diff --git a/Documentation/powerpc/dts-bindings/spi-bus.txt b/Documentation/powerpc/dts-bindings/spi-bus.txt
new file mode 100644
index 000000000000..e782add2e457
--- /dev/null
+++ b/Documentation/powerpc/dts-bindings/spi-bus.txt
@@ -0,0 +1,57 @@
+SPI (Serial Peripheral Interface) busses
+
+SPI busses can be described with a node for the SPI master device
+and a set of child nodes for each SPI slave on the bus.  For this
+discussion, it is assumed that the system's SPI controller is in
+SPI master mode.  This binding does not describe SPI controllers
+in slave mode.
+
+The SPI master node requires the following properties:
+- #address-cells  - number of cells required to define a chip select
+    		address on the SPI bus.
+- #size-cells     - should be zero.
+- compatible      - name of SPI bus controller following generic names
+    		recommended practice.
+No other properties are required in the SPI bus node.  It is assumed
+that a driver for an SPI bus device will understand that it is an SPI bus.
+However, the binding does not attempt to define the specific method for
+assigning chip select numbers.  Since SPI chip select configuration is
+flexible and non-standardized, it is left out of this binding with the
+assumption that board specific platform code will be used to manage
+chip selects.  Individual drivers can define additional properties to
+support describing the chip select layout.
+
+SPI slave nodes must be children of the SPI master node and can
+contain the following properties.
+- reg             - (required) chip select address of device.
+- compatible      - (required) name of SPI device following generic names
+    		recommended practice
+- spi-max-frequency - (required) Maximum SPI clocking speed of device in Hz
+- spi-cpol        - (optional) Empty property indicating device requires
+    		inverse clock polarity (CPOL) mode
+- spi-cpha        - (optional) Empty property indicating device requires
+    		shifted clock phase (CPHA) mode
+- spi-cs-high     - (optional) Empty property indicating device requires
+    		chip select active high
+
+SPI example for an MPC5200 SPI bus:
+	spi@f00 {
+		#address-cells = <1>;
+		#size-cells = <0>;
+		compatible = "fsl,mpc5200b-spi","fsl,mpc5200-spi";
+		reg = <0xf00 0x20>;
+		interrupts = <2 13 0 2 14 0>;
+		interrupt-parent = <&mpc5200_pic>;
+
+		ethernet-switch@0 {
+			compatible = "micrel,ks8995m";
+			spi-max-frequency = <1000000>;
+			reg = <0>;
+		};
+
+		codec@1 {
+			compatible = "ti,tlv320aic26";
+			spi-max-frequency = <100000>;
+			reg = <1>;
+		};
+	};
diff --git a/Documentation/powerpc/dts-bindings/usb-ehci.txt b/Documentation/powerpc/dts-bindings/usb-ehci.txt
new file mode 100644
index 000000000000..fa18612f757b
--- /dev/null
+++ b/Documentation/powerpc/dts-bindings/usb-ehci.txt
@@ -0,0 +1,25 @@
+USB EHCI controllers
+
+Required properties:
+  - compatible : should be "usb-ehci".
+  - reg : should contain at least address and length of the standard EHCI
+    register set for the device. Optional platform-dependent registers
+    (debug-port or other) can be also specified here, but only after
+    definition of standard EHCI registers.
+  - interrupts : one EHCI interrupt should be described here.
+If device registers are implemented in big endian mode, the device
+node should have "big-endian-regs" property.
+If controller implementation operates with big endian descriptors,
+"big-endian-desc" property should be specified.
+If both big endian registers and descriptors are used by the controller
+implementation, "big-endian" property can be specified instead of having
+both "big-endian-regs" and "big-endian-desc".
+
+Example (Sequoia 440EPx):
+    ehci@e0000300 {
+	   compatible = "ibm,usb-ehci-440epx", "usb-ehci";
+	   interrupt-parent = <&UIC0>;
+	   interrupts = <1a 4>;
+	   reg = <0 e0000300 90 0 e0000390 70>;
+	   big-endian;
+   };
diff --git a/Documentation/powerpc/dts-bindings/xilinx.txt b/Documentation/powerpc/dts-bindings/xilinx.txt
new file mode 100644
index 000000000000..80339fe4300b
--- /dev/null
+++ b/Documentation/powerpc/dts-bindings/xilinx.txt
@@ -0,0 +1,295 @@
+   d) Xilinx IP cores
+
+   The Xilinx EDK toolchain ships with a set of IP cores (devices) for use
+   in Xilinx Spartan and Virtex FPGAs.  The devices cover the whole range
+   of standard device types (network, serial, etc.) and miscellaneous
+   devices (gpio, LCD, spi, etc).  Also, since these devices are
+   implemented within the fpga fabric every instance of the device can be
+   synthesised with different options that change the behaviour.
+
+   Each IP-core has a set of parameters which the FPGA designer can use to
+   control how the core is synthesized.  Historically, the EDK tool would
+   extract the device parameters relevant to device drivers and copy them
+   into an 'xparameters.h' in the form of #define symbols.  This tells the
+   device drivers how the IP cores are configured, but it requres the kernel
+   to be recompiled every time the FPGA bitstream is resynthesized.
+
+   The new approach is to export the parameters into the device tree and
+   generate a new device tree each time the FPGA bitstream changes.  The
+   parameters which used to be exported as #defines will now become
+   properties of the device node.  In general, device nodes for IP-cores
+   will take the following form:
+
+	(name): (generic-name)@(base-address) {
+		compatible = "xlnx,(ip-core-name)-(HW_VER)"
+			     [, (list of compatible devices), ...];
+		reg = <(baseaddr) (size)>;
+		interrupt-parent = <&interrupt-controller-phandle>;
+		interrupts = < ... >;
+		xlnx,(parameter1) = "(string-value)";
+		xlnx,(parameter2) = <(int-value)>;
+	};
+
+	(generic-name):   an open firmware-style name that describes the
+			generic class of device.  Preferably, this is one word, such
+			as 'serial' or 'ethernet'.
+	(ip-core-name):	the name of the ip block (given after the BEGIN
+			directive in system.mhs).  Should be in lowercase
+			and all underscores '_' converted to dashes '-'.
+	(name):		is derived from the "PARAMETER INSTANCE" value.
+	(parameter#):	C_* parameters from system.mhs.  The C_ prefix is
+			dropped from the parameter name, the name is converted
+			to lowercase and all underscore '_' characters are
+			converted to dashes '-'.
+	(baseaddr):	the baseaddr parameter value (often named C_BASEADDR).
+	(HW_VER):	from the HW_VER parameter.
+	(size):		the address range size (often C_HIGHADDR - C_BASEADDR + 1).
+
+   Typically, the compatible list will include the exact IP core version
+   followed by an older IP core version which implements the same
+   interface or any other device with the same interface.
+
+   'reg', 'interrupt-parent' and 'interrupts' are all optional properties.
+
+   For example, the following block from system.mhs:
+
+	BEGIN opb_uartlite
+		PARAMETER INSTANCE = opb_uartlite_0
+		PARAMETER HW_VER = 1.00.b
+		PARAMETER C_BAUDRATE = 115200
+		PARAMETER C_DATA_BITS = 8
+		PARAMETER C_ODD_PARITY = 0
+		PARAMETER C_USE_PARITY = 0
+		PARAMETER C_CLK_FREQ = 50000000
+		PARAMETER C_BASEADDR = 0xEC100000
+		PARAMETER C_HIGHADDR = 0xEC10FFFF
+		BUS_INTERFACE SOPB = opb_7
+		PORT OPB_Clk = CLK_50MHz
+		PORT Interrupt = opb_uartlite_0_Interrupt
+		PORT RX = opb_uartlite_0_RX
+		PORT TX = opb_uartlite_0_TX
+		PORT OPB_Rst = sys_bus_reset_0
+	END
+
+   becomes the following device tree node:
+
+	opb_uartlite_0: serial@ec100000 {
+		device_type = "serial";
+		compatible = "xlnx,opb-uartlite-1.00.b";
+		reg = <ec100000 10000>;
+		interrupt-parent = <&opb_intc_0>;
+		interrupts = <1 0>; // got this from the opb_intc parameters
+		current-speed = <d#115200>;	// standard serial device prop
+		clock-frequency = <d#50000000>;	// standard serial device prop
+		xlnx,data-bits = <8>;
+		xlnx,odd-parity = <0>;
+		xlnx,use-parity = <0>;
+	};
+
+   Some IP cores actually implement 2 or more logical devices.  In
+   this case, the device should still describe the whole IP core with
+   a single node and add a child node for each logical device.  The
+   ranges property can be used to translate from parent IP-core to the
+   registers of each device.  In addition, the parent node should be
+   compatible with the bus type 'xlnx,compound', and should contain
+   #address-cells and #size-cells, as with any other bus.  (Note: this
+   makes the assumption that both logical devices have the same bus
+   binding.  If this is not true, then separate nodes should be used
+   for each logical device).  The 'cell-index' property can be used to
+   enumerate logical devices within an IP core.  For example, the
+   following is the system.mhs entry for the dual ps2 controller found
+   on the ml403 reference design.
+
+	BEGIN opb_ps2_dual_ref
+		PARAMETER INSTANCE = opb_ps2_dual_ref_0
+		PARAMETER HW_VER = 1.00.a
+		PARAMETER C_BASEADDR = 0xA9000000
+		PARAMETER C_HIGHADDR = 0xA9001FFF
+		BUS_INTERFACE SOPB = opb_v20_0
+		PORT Sys_Intr1 = ps2_1_intr
+		PORT Sys_Intr2 = ps2_2_intr
+		PORT Clkin1 = ps2_clk_rx_1
+		PORT Clkin2 = ps2_clk_rx_2
+		PORT Clkpd1 = ps2_clk_tx_1
+		PORT Clkpd2 = ps2_clk_tx_2
+		PORT Rx1 = ps2_d_rx_1
+		PORT Rx2 = ps2_d_rx_2
+		PORT Txpd1 = ps2_d_tx_1
+		PORT Txpd2 = ps2_d_tx_2
+	END
+
+   It would result in the following device tree nodes:
+
+	opb_ps2_dual_ref_0: opb-ps2-dual-ref@a9000000 {
+		#address-cells = <1>;
+		#size-cells = <1>;
+		compatible = "xlnx,compound";
+		ranges = <0 a9000000 2000>;
+		// If this device had extra parameters, then they would
+		// go here.
+		ps2@0 {
+			compatible = "xlnx,opb-ps2-dual-ref-1.00.a";
+			reg = <0 40>;
+			interrupt-parent = <&opb_intc_0>;
+			interrupts = <3 0>;
+			cell-index = <0>;
+		};
+		ps2@1000 {
+			compatible = "xlnx,opb-ps2-dual-ref-1.00.a";
+			reg = <1000 40>;
+			interrupt-parent = <&opb_intc_0>;
+			interrupts = <3 0>;
+			cell-index = <0>;
+		};
+	};
+
+   Also, the system.mhs file defines bus attachments from the processor
+   to the devices.  The device tree structure should reflect the bus
+   attachments.  Again an example; this system.mhs fragment:
+
+	BEGIN ppc405_virtex4
+		PARAMETER INSTANCE = ppc405_0
+		PARAMETER HW_VER = 1.01.a
+		BUS_INTERFACE DPLB = plb_v34_0
+		BUS_INTERFACE IPLB = plb_v34_0
+	END
+
+	BEGIN opb_intc
+		PARAMETER INSTANCE = opb_intc_0
+		PARAMETER HW_VER = 1.00.c
+		PARAMETER C_BASEADDR = 0xD1000FC0
+		PARAMETER C_HIGHADDR = 0xD1000FDF
+		BUS_INTERFACE SOPB = opb_v20_0
+	END
+
+	BEGIN opb_uart16550
+		PARAMETER INSTANCE = opb_uart16550_0
+		PARAMETER HW_VER = 1.00.d
+		PARAMETER C_BASEADDR = 0xa0000000
+		PARAMETER C_HIGHADDR = 0xa0001FFF
+		BUS_INTERFACE SOPB = opb_v20_0
+	END
+
+	BEGIN plb_v34
+		PARAMETER INSTANCE = plb_v34_0
+		PARAMETER HW_VER = 1.02.a
+	END
+
+	BEGIN plb_bram_if_cntlr
+		PARAMETER INSTANCE = plb_bram_if_cntlr_0
+		PARAMETER HW_VER = 1.00.b
+		PARAMETER C_BASEADDR = 0xFFFF0000
+		PARAMETER C_HIGHADDR = 0xFFFFFFFF
+		BUS_INTERFACE SPLB = plb_v34_0
+	END
+
+	BEGIN plb2opb_bridge
+		PARAMETER INSTANCE = plb2opb_bridge_0
+		PARAMETER HW_VER = 1.01.a
+		PARAMETER C_RNG0_BASEADDR = 0x20000000
+		PARAMETER C_RNG0_HIGHADDR = 0x3FFFFFFF
+		PARAMETER C_RNG1_BASEADDR = 0x60000000
+		PARAMETER C_RNG1_HIGHADDR = 0x7FFFFFFF
+		PARAMETER C_RNG2_BASEADDR = 0x80000000
+		PARAMETER C_RNG2_HIGHADDR = 0xBFFFFFFF
+		PARAMETER C_RNG3_BASEADDR = 0xC0000000
+		PARAMETER C_RNG3_HIGHADDR = 0xDFFFFFFF
+		BUS_INTERFACE SPLB = plb_v34_0
+		BUS_INTERFACE MOPB = opb_v20_0
+	END
+
+   Gives this device tree (some properties removed for clarity):
+
+	plb@0 {
+		#address-cells = <1>;
+		#size-cells = <1>;
+		compatible = "xlnx,plb-v34-1.02.a";
+		device_type = "ibm,plb";
+		ranges; // 1:1 translation
+
+		plb_bram_if_cntrl_0: bram@ffff0000 {
+			reg = <ffff0000 10000>;
+		}
+
+		opb@20000000 {
+			#address-cells = <1>;
+			#size-cells = <1>;
+			ranges = <20000000 20000000 20000000
+				  60000000 60000000 20000000
+				  80000000 80000000 40000000
+				  c0000000 c0000000 20000000>;
+
+			opb_uart16550_0: serial@a0000000 {
+				reg = <a00000000 2000>;
+			};
+
+			opb_intc_0: interrupt-controller@d1000fc0 {
+				reg = <d1000fc0 20>;
+			};
+		};
+	};
+
+   That covers the general approach to binding xilinx IP cores into the
+   device tree.  The following are bindings for specific devices:
+
+      i) Xilinx ML300 Framebuffer
+
+      Simple framebuffer device from the ML300 reference design (also on the
+      ML403 reference design as well as others).
+
+      Optional properties:
+       - resolution = <xres yres> : pixel resolution of framebuffer.  Some
+                                    implementations use a different resolution.
+                                    Default is <d#640 d#480>
+       - virt-resolution = <xvirt yvirt> : Size of framebuffer in memory.
+                                           Default is <d#1024 d#480>.
+       - rotate-display (empty) : rotate display 180 degrees.
+
+      ii) Xilinx SystemACE
+
+      The Xilinx SystemACE device is used to program FPGAs from an FPGA
+      bitstream stored on a CF card.  It can also be used as a generic CF
+      interface device.
+
+      Optional properties:
+       - 8-bit (empty) : Set this property for SystemACE in 8 bit mode
+
+      iii) Xilinx EMAC and Xilinx TEMAC
+
+      Xilinx Ethernet devices.  In addition to general xilinx properties
+      listed above, nodes for these devices should include a phy-handle
+      property, and may include other common network device properties
+      like local-mac-address.
+
+      iv) Xilinx Uartlite
+
+      Xilinx uartlite devices are simple fixed speed serial ports.
+
+      Required properties:
+       - current-speed : Baud rate of uartlite
+
+      v) Xilinx hwicap
+
+		Xilinx hwicap devices provide access to the configuration logic
+		of the FPGA through the Internal Configuration Access Port
+		(ICAP).  The ICAP enables partial reconfiguration of the FPGA,
+		readback of the configuration information, and some control over
+		'warm boots' of the FPGA fabric.
+
+		Required properties:
+		- xlnx,family : The family of the FPGA, necessary since the
+                      capabilities of the underlying ICAP hardware
+                      differ between different families.  May be
+                      'virtex2p', 'virtex4', or 'virtex5'.
+
+      vi) Xilinx Uart 16550
+
+      Xilinx UART 16550 devices are very similar to the NS16550 but with
+      different register spacing and an offset from the base address.
+
+      Required properties:
+       - clock-frequency : Frequency of the clock input
+       - reg-offset : A value of 3 is required
+       - reg-shift : A value of 2 is required
+
+
-- 
cgit v1.2.3-59-g8ed1b


From 6cc7959f385abc77abdeda5462f96407a556265a Mon Sep 17 00:00:00 2001
From: Huang Weiyi <weiyi.huang@gmail.com>
Date: Sat, 20 Jun 2009 19:16:59 +0800
Subject: powerpc/85xx: remove duplicated #include

Remove duplicated #include in arch/powerpc/platforms/85xx/xes_mpc85xx.c.

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/platforms/85xx/xes_mpc85xx.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/powerpc/platforms/85xx/xes_mpc85xx.c b/arch/powerpc/platforms/85xx/xes_mpc85xx.c
index ee01532786e4..1b426050a2f9 100644
--- a/arch/powerpc/platforms/85xx/xes_mpc85xx.c
+++ b/arch/powerpc/platforms/85xx/xes_mpc85xx.c
@@ -32,7 +32,6 @@
 
 #include <sysdev/fsl_soc.h>
 #include <sysdev/fsl_pci.h>
-#include <linux/of_platform.h>
 
 /* A few bit definitions needed for fixups on some boards */
 #define MPC85xx_L2CTL_L2E		0x80000000 /* L2 enable */
-- 
cgit v1.2.3-59-g8ed1b


From 66c6b856d8738278a4a6e558d25c90e9950aa112 Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Fri, 19 Jun 2009 03:37:52 +0400
Subject: powerpc/85xx: Make eSDHC 1-bit only transfer mode default for
 MPC8569E-MDS

For yet unknown reason 4-bit mode doesn't work on MPC8569E-MDS boards,
so make 1-bit mode default. When we resolve the issue, u-boot will
remove sdhci,1-bit-only property from the device tree, while SDHCI
will still work with older u-boots.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/boot/dts/mpc8569mds.dts | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/powerpc/boot/dts/mpc8569mds.dts b/arch/powerpc/boot/dts/mpc8569mds.dts
index a8dcb018c4a5..a680165292f2 100644
--- a/arch/powerpc/boot/dts/mpc8569mds.dts
+++ b/arch/powerpc/boot/dts/mpc8569mds.dts
@@ -253,6 +253,7 @@
 			/* Filled in by U-Boot */
 			clock-frequency = <0>;
 			status = "disabled";
+			sdhci,1-bit-only;
 		};
 
 		crypto@30000 {
-- 
cgit v1.2.3-59-g8ed1b


From ae281064be164342554b34f4ca5c4af33dce3de1 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Tue, 23 Jun 2009 14:40:26 +0100
Subject: kmemleak: use pr_fmt

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 mm/kmemleak.c | 52 ++++++++++++++++++++++++----------------------------
 1 file changed, 24 insertions(+), 28 deletions(-)

diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index ec759b60077a..c96f2c8700aa 100644
--- a/mm/kmemleak.c
+++ b/mm/kmemleak.c
@@ -61,6 +61,8 @@
  * structure.
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/list.h>
@@ -311,7 +313,7 @@ static int unreferenced_object(struct kmemleak_object *object)
 
 static void print_referenced(struct kmemleak_object *object)
 {
-	pr_info("kmemleak: referenced object 0x%08lx (size %zu)\n",
+	pr_info("referenced object 0x%08lx (size %zu)\n",
 		object->pointer, object->size);
 }
 
@@ -320,7 +322,7 @@ static void print_unreferenced(struct seq_file *seq,
 {
 	int i;
 
-	print_helper(seq, "kmemleak: unreferenced object 0x%08lx (size %zu):\n",
+	print_helper(seq, "unreferenced object 0x%08lx (size %zu):\n",
 		     object->pointer, object->size);
 	print_helper(seq, "  comm \"%s\", pid %d, jiffies %lu\n",
 		     object->comm, object->pid, object->jiffies);
@@ -344,7 +346,7 @@ static void dump_object_info(struct kmemleak_object *object)
 	trace.nr_entries = object->trace_len;
 	trace.entries = object->trace;
 
-	pr_notice("kmemleak: Object 0x%08lx (size %zu):\n",
+	pr_notice("Object 0x%08lx (size %zu):\n",
 		  object->tree_node.start, object->size);
 	pr_notice("  comm \"%s\", pid %d, jiffies %lu\n",
 		  object->comm, object->pid, object->jiffies);
@@ -372,7 +374,7 @@ static struct kmemleak_object *lookup_object(unsigned long ptr, int alias)
 		object = prio_tree_entry(node, struct kmemleak_object,
 					 tree_node);
 		if (!alias && object->pointer != ptr) {
-			kmemleak_warn("kmemleak: Found object by alias");
+			kmemleak_warn("Found object by alias");
 			object = NULL;
 		}
 	} else
@@ -467,8 +469,7 @@ static void create_object(unsigned long ptr, size_t size, int min_count,
 
 	object = kmem_cache_alloc(object_cache, gfp & GFP_KMEMLEAK_MASK);
 	if (!object) {
-		kmemleak_stop("kmemleak: Cannot allocate a kmemleak_object "
-			      "structure\n");
+		kmemleak_stop("Cannot allocate a kmemleak_object structure\n");
 		return;
 	}
 
@@ -527,8 +528,8 @@ static void create_object(unsigned long ptr, size_t size, int min_count,
 	if (node != &object->tree_node) {
 		unsigned long flags;
 
-		kmemleak_stop("kmemleak: Cannot insert 0x%lx into the object "
-			      "search tree (already existing)\n", ptr);
+		kmemleak_stop("Cannot insert 0x%lx into the object search tree "
+			      "(already existing)\n", ptr);
 		object = lookup_object(ptr, 1);
 		spin_lock_irqsave(&object->lock, flags);
 		dump_object_info(object);
@@ -553,7 +554,7 @@ static void delete_object(unsigned long ptr)
 	write_lock_irqsave(&kmemleak_lock, flags);
 	object = lookup_object(ptr, 0);
 	if (!object) {
-		kmemleak_warn("kmemleak: Freeing unknown object at 0x%08lx\n",
+		kmemleak_warn("Freeing unknown object at 0x%08lx\n",
 			      ptr);
 		write_unlock_irqrestore(&kmemleak_lock, flags);
 		return;
@@ -588,8 +589,7 @@ static void make_gray_object(unsigned long ptr)
 
 	object = find_and_get_object(ptr, 0);
 	if (!object) {
-		kmemleak_warn("kmemleak: Graying unknown object at 0x%08lx\n",
-			      ptr);
+		kmemleak_warn("Graying unknown object at 0x%08lx\n", ptr);
 		return;
 	}
 
@@ -610,8 +610,7 @@ static void make_black_object(unsigned long ptr)
 
 	object = find_and_get_object(ptr, 0);
 	if (!object) {
-		kmemleak_warn("kmemleak: Blacking unknown object at 0x%08lx\n",
-			      ptr);
+		kmemleak_warn("Blacking unknown object at 0x%08lx\n", ptr);
 		return;
 	}
 
@@ -634,21 +633,20 @@ static void add_scan_area(unsigned long ptr, unsigned long offset,
 
 	object = find_and_get_object(ptr, 0);
 	if (!object) {
-		kmemleak_warn("kmemleak: Adding scan area to unknown "
-			      "object at 0x%08lx\n", ptr);
+		kmemleak_warn("Adding scan area to unknown object at 0x%08lx\n",
+			      ptr);
 		return;
 	}
 
 	area = kmem_cache_alloc(scan_area_cache, gfp & GFP_KMEMLEAK_MASK);
 	if (!area) {
-		kmemleak_warn("kmemleak: Cannot allocate a scan area\n");
+		kmemleak_warn("Cannot allocate a scan area\n");
 		goto out;
 	}
 
 	spin_lock_irqsave(&object->lock, flags);
 	if (offset + length > object->size) {
-		kmemleak_warn("kmemleak: Scan area larger than object "
-			      "0x%08lx\n", ptr);
+		kmemleak_warn("Scan area larger than object 0x%08lx\n", ptr);
 		dump_object_info(object);
 		kmem_cache_free(scan_area_cache, area);
 		goto out_unlock;
@@ -677,8 +675,7 @@ static void object_no_scan(unsigned long ptr)
 
 	object = find_and_get_object(ptr, 0);
 	if (!object) {
-		kmemleak_warn("kmemleak: Not scanning unknown object at "
-			      "0x%08lx\n", ptr);
+		kmemleak_warn("Not scanning unknown object at 0x%08lx\n", ptr);
 		return;
 	}
 
@@ -699,7 +696,7 @@ static void log_early(int op_type, const void *ptr, size_t size,
 	struct early_log *log;
 
 	if (crt_early_log >= ARRAY_SIZE(early_log)) {
-		kmemleak_stop("kmemleak: Early log buffer exceeded\n");
+		kmemleak_stop("Early log buffer exceeded\n");
 		return;
 	}
 
@@ -966,7 +963,7 @@ static void kmemleak_scan(void)
 		 * 1 reference to any object at this point.
 		 */
 		if (atomic_read(&object->use_count) > 1) {
-			pr_debug("kmemleak: object->use_count = %d\n",
+			pr_debug("object->use_count = %d\n",
 				 atomic_read(&object->use_count));
 			dump_object_info(object);
 		}
@@ -1062,7 +1059,7 @@ static int kmemleak_scan_thread(void *arg)
 {
 	static int first_run = 1;
 
-	pr_info("kmemleak: Automatic memory scanning thread started\n");
+	pr_info("Automatic memory scanning thread started\n");
 
 	/*
 	 * Wait before the first scan to allow the system to fully initialize.
@@ -1108,7 +1105,7 @@ static int kmemleak_scan_thread(void *arg)
 			timeout = schedule_timeout_interruptible(timeout);
 	}
 
-	pr_info("kmemleak: Automatic memory scanning thread ended\n");
+	pr_info("Automatic memory scanning thread ended\n");
 
 	return 0;
 }
@@ -1123,7 +1120,7 @@ void start_scan_thread(void)
 		return;
 	scan_thread = kthread_run(kmemleak_scan_thread, NULL, "kmemleak");
 	if (IS_ERR(scan_thread)) {
-		pr_warning("kmemleak: Failed to create the scan thread\n");
+		pr_warning("Failed to create the scan thread\n");
 		scan_thread = NULL;
 	}
 }
@@ -1367,7 +1364,7 @@ static void kmemleak_cleanup(void)
 	cleanup_thread = kthread_run(kmemleak_cleanup_thread, NULL,
 				     "kmemleak-clean");
 	if (IS_ERR(cleanup_thread))
-		pr_warning("kmemleak: Failed to create the clean-up thread\n");
+		pr_warning("Failed to create the clean-up thread\n");
 }
 
 /*
@@ -1488,8 +1485,7 @@ static int __init kmemleak_late_init(void)
 	dentry = debugfs_create_file("kmemleak", S_IRUGO, NULL, NULL,
 				     &kmemleak_fops);
 	if (!dentry)
-		pr_warning("kmemleak: Failed to create the debugfs kmemleak "
-			   "file\n");
+		pr_warning("Failed to create the debugfs kmemleak file\n");
 	mutex_lock(&kmemleak_mutex);
 	start_scan_thread();
 	mutex_unlock(&kmemleak_mutex);
-- 
cgit v1.2.3-59-g8ed1b


From bf96d1e3e7a35a17cea255048ffb3243bd9c8123 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Tue, 23 Jun 2009 14:40:27 +0100
Subject: kmemleak: Do not force the slab debugging Kconfig options

Selecting DEBUG_SLAB or SLUB_DEBUG by the KMEMLEAK menu entry may cause
issues with other dependencies (KMEMCHECK). These configuration options
aren't strictly needed by kmemleak but they may increase the chances of
finding leaks. This patch also updates the KMEMLEAK config entry help
text.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
---
 lib/Kconfig.debug | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 23067ab1a73c..4c32b1a1a06e 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -340,8 +340,6 @@ config DEBUG_KMEMLEAK
 	bool "Kernel memory leak detector"
 	depends on DEBUG_KERNEL && EXPERIMENTAL && (X86 || ARM) && \
 		!MEMORY_HOTPLUG
-	select DEBUG_SLAB if SLAB
-	select SLUB_DEBUG if SLUB
 	select DEBUG_FS if SYSFS
 	select STACKTRACE if STACKTRACE_SUPPORT
 	select KALLSYMS
@@ -355,6 +353,9 @@ config DEBUG_KMEMLEAK
 	  allocations. See Documentation/kmemleak.txt for more
 	  details.
 
+	  Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances
+	  of finding leaks due to the slab objects poisoning.
+
 	  In order to access the kmemleak file, debugfs needs to be
 	  mounted (usually at /sys/kernel/debug).
 
-- 
cgit v1.2.3-59-g8ed1b


From 503dcbeba50fd3545283594bc391b4a400fa6c48 Mon Sep 17 00:00:00 2001
From: Tony Lindgren <tony@atomide.com>
Date: Tue, 23 Jun 2009 16:55:30 +0300
Subject: OMAP: Fix IOMEM macro for assembly

Otherwise IOMEM calculations can fail.

Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 arch/arm/plat-omap/include/mach/io.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/plat-omap/include/mach/io.h b/arch/arm/plat-omap/include/mach/io.h
index 3b2814720569..73f483d56ca6 100644
--- a/arch/arm/plat-omap/include/mach/io.h
+++ b/arch/arm/plat-omap/include/mach/io.h
@@ -201,7 +201,7 @@
 #define OMAP2_IO_ADDRESS(pa)	IOMEM(__OMAP2_IO_ADDRESS(pa))
 
 #ifdef __ASSEMBLER__
-#define IOMEM(x)		x
+#define IOMEM(x)		(x)
 #else
 #define IOMEM(x)		((void __force __iomem *)(x))
 
-- 
cgit v1.2.3-59-g8ed1b


From b0a28589b2fc9bee8ed83dee006a497d1ce93841 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Tue, 23 Jun 2009 16:39:53 +0200
Subject: perf report: Fix help text typo

Reported-by: Brice Goglin <Brice.Goglin@inria.fr>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/Documentation/perf-report.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
index 52d3fc6846a9..40c1db83a16d 100644
--- a/tools/perf/Documentation/perf-report.txt
+++ b/tools/perf/Documentation/perf-report.txt
@@ -13,7 +13,7 @@ SYNOPSIS
 DESCRIPTION
 -----------
 This command displays the performance counter profile information recorded
-via perf report.
+via perf record.
 
 OPTIONS
 -------
-- 
cgit v1.2.3-59-g8ed1b


From 927dbcd668ede8d2210cc59bea548ca9ff45b240 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben@simtec.co.uk>
Date: Tue, 23 Jun 2009 16:15:38 +0100
Subject: [ARM] S3C24XX: Fix spi-bus configuration build errors

The commit ec976d6eb021dc8f2994248c310a41540f4756bd
removed a number of gpio definitions from <mach/hardware.h>
but misssed updating these two files:

Fix the following build errors by including <linux/gpio.h>:
arch/arm/plat-s3c24xx/spi-bus1-gpg5_6_7.c: In function 's3c24xx_spi_gpiocfg_bus1_gpg5_6_7':
arch/arm/plat-s3c24xx/spi-bus1-gpg5_6_7.c:25: error: implicit declaration of function 's3c2410_gpio_cfgpin'
arch/arm/plat-s3c24xx/spi-bus1-gpg5_6_7.c:28: error: implicit declaration of function 's3c2410_gpio_pullup'
arch/arm/plat-s3c24xx/spi-bus0-gpe11_12_13.c: In function 's3c24xx_spi_gpiocfg_bus0_gpe11_12_13':
arch/arm/plat-s3c24xx/spi-bus0-gpe11_12_13.c:25: error: implicit declaration of function 's3c2410_gpio_cfgpin'
arch/arm/plat-s3c24xx/spi-bus0-gpe11_12_13.c:28: error: implicit declaration of function 's3c2410_gpio_pullup'

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c24xx/spi-bus0-gpe11_12_13.c | 3 +--
 arch/arm/plat-s3c24xx/spi-bus1-gpg5_6_7.c    | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/arch/arm/plat-s3c24xx/spi-bus0-gpe11_12_13.c b/arch/arm/plat-s3c24xx/spi-bus0-gpe11_12_13.c
index 9edf7894eedd..da7a61728c18 100644
--- a/arch/arm/plat-s3c24xx/spi-bus0-gpe11_12_13.c
+++ b/arch/arm/plat-s3c24xx/spi-bus0-gpe11_12_13.c
@@ -12,8 +12,7 @@
 */
 
 #include <linux/kernel.h>
-
-#include <mach/hardware.h>
+#include <linux/gpio.h>
 
 #include <mach/spi.h>
 #include <mach/regs-gpio.h>
diff --git a/arch/arm/plat-s3c24xx/spi-bus1-gpg5_6_7.c b/arch/arm/plat-s3c24xx/spi-bus1-gpg5_6_7.c
index f34d0fc69ad8..86b9edc67413 100644
--- a/arch/arm/plat-s3c24xx/spi-bus1-gpg5_6_7.c
+++ b/arch/arm/plat-s3c24xx/spi-bus1-gpg5_6_7.c
@@ -12,8 +12,7 @@
 */
 
 #include <linux/kernel.h>
-
-#include <mach/hardware.h>
+#include <linux/gpio.h>
 
 #include <mach/spi.h>
 #include <mach/regs-gpio.h>
-- 
cgit v1.2.3-59-g8ed1b


From ffd14417bdf2a1650bcb16d37c7e658535d1681a Mon Sep 17 00:00:00 2001
From: Huang Weiyi <weiyi.huang@gmail.com>
Date: Tue, 23 Jun 2009 20:58:57 +0800
Subject: [ARM] MINI2440: remove duplicated #include

Remove duplicated #include('s) in
  arch/arm/mach-s3c2440/mach-mini2440.c

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/mach-s3c2440/mach-mini2440.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/arch/arm/mach-s3c2440/mach-mini2440.c b/arch/arm/mach-s3c2440/mach-mini2440.c
index 6a5bc3021bdb..e714bfabbc09 100644
--- a/arch/arm/mach-s3c2440/mach-mini2440.c
+++ b/arch/arm/mach-s3c2440/mach-mini2440.c
@@ -48,8 +48,6 @@
 #include <plat/mci.h>
 #include <plat/udc.h>
 
-#include <plat/regs-serial.h>
-
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/nand.h>
 #include <linux/mtd/nand_ecc.h>
-- 
cgit v1.2.3-59-g8ed1b


From a3c79901170aba0d3717c2602326bab639eb1344 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Tue, 23 Jun 2009 16:30:02 +0100
Subject: [ARM] S3C24XX: Fix missing s3c_iis_device.

Commit 52da219e9664e537a745877b0efa7cf2b1ff2996 removed
the s3c_device_iis, but didn't replace it with anything
so a number of s3c24xx machines are currently failing
to build.

As a temporary fix, re-instate s3c_device_iis until a
proper replacement can be done for it.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c/include/plat/devs.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm/plat-s3c/include/plat/devs.h b/arch/arm/plat-s3c/include/plat/devs.h
index b5b9c4d46e9a..2e170827e0b0 100644
--- a/arch/arm/plat-s3c/include/plat/devs.h
+++ b/arch/arm/plat-s3c/include/plat/devs.h
@@ -37,6 +37,7 @@ extern struct platform_device s3c_device_i2c1;
 extern struct platform_device s3c_device_rtc;
 extern struct platform_device s3c_device_adc;
 extern struct platform_device s3c_device_sdi;
+extern struct platform_device s3c_device_iis;
 extern struct platform_device s3c_device_hwmon;
 extern struct platform_device s3c_device_hsmmc0;
 extern struct platform_device s3c_device_hsmmc1;
-- 
cgit v1.2.3-59-g8ed1b


From b647712f669f4fadf428a14ab45ab4c03558cdf5 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Tue, 23 Jun 2009 16:34:35 +0100
Subject: [ARM] S3C: Fix S3C24XX build to not include s3c64xx IIS devices

Commit 52da219e9664e537a745877b0efa7cf2b1ff2996 added IIS
platform devices, but these do not build on s3c24xx systems
and the file depends on SND_S3C24XX_SOC, which is selected
for all S3C64XX/S3C24XX systems.

As a quick fix, make the dev-audio.o file depends on
SND_S3C64XX_SOC_I2S instead.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/plat-s3c/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/plat-s3c/Makefile b/arch/arm/plat-s3c/Makefile
index 74bb7cb5da49..0761766b1833 100644
--- a/arch/arm/plat-s3c/Makefile
+++ b/arch/arm/plat-s3c/Makefile
@@ -34,7 +34,7 @@ obj-$(CONFIG_S3C_DEV_HSMMC)	+= dev-hsmmc.o
 obj-$(CONFIG_S3C_DEV_HSMMC1)	+= dev-hsmmc1.o
 obj-y				+= dev-i2c0.o
 obj-$(CONFIG_S3C_DEV_I2C1)	+= dev-i2c1.o
-obj-$(CONFIG_SND_S3C24XX_SOC)	+= dev-audio.o
+obj-$(CONFIG_SND_S3C64XX_SOC_I2S)	+= dev-audio.o
 obj-$(CONFIG_S3C_DEV_FB)	+= dev-fb.o
 obj-$(CONFIG_S3C_DEV_USB_HOST)	+= dev-usb.o
 obj-$(CONFIG_S3C_DEV_USB_HSOTG)	+= dev-usb-hsotg.o
-- 
cgit v1.2.3-59-g8ed1b


From b5aa8a0fc132dd512c33e7c2621d075e3b77a65e Mon Sep 17 00:00:00 2001
From: Grégoire Henry <Gregoire.Henry@pps.jussieu.fr>
Date: Tue, 23 Jun 2009 15:41:02 +0200
Subject: drm/i915: initialize fence registers to zero when loading GEM
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Unitialized fence register could leads to corrupted display. Problem
encountered on MacBooks (revision 1 and 2), directly booting from EFI
or through BIOS emulation.

(bug #21710 at freedestop.org)

Signed-off-by: Grégoire Henry <henry@pps.jussieu.fr>
Signed-off-by: Eric Anholt <eric@anholt.net>
---
 drivers/gpu/drm/i915/i915_gem.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 8660b2144b27..876b65cb7629 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -4227,6 +4227,7 @@ i915_gem_lastclose(struct drm_device *dev)
 void
 i915_gem_load(struct drm_device *dev)
 {
+	int i;
 	drm_i915_private_t *dev_priv = dev->dev_private;
 
 	spin_lock_init(&dev_priv->mm.active_list_lock);
@@ -4246,6 +4247,18 @@ i915_gem_load(struct drm_device *dev)
 	else
 		dev_priv->num_fence_regs = 8;
 
+	/* Initialize fence registers to zero */
+	if (IS_I965G(dev)) {
+		for (i = 0; i < 16; i++)
+			I915_WRITE64(FENCE_REG_965_0 + (i * 8), 0);
+	} else {
+		for (i = 0; i < 8; i++)
+			I915_WRITE(FENCE_REG_830_0 + (i * 4), 0);
+		if (IS_I945G(dev) || IS_I945GM(dev) || IS_G33(dev))
+			for (i = 0; i < 8; i++)
+				I915_WRITE(FENCE_REG_945_8 + (i * 4), 0);
+	}
+
 	i915_gem_detect_bit_6_swizzle(dev);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From bb38c222e07cbe252cfbd99cd6a9c834b0a2639a Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 24 Jun 2009 01:41:05 +0900
Subject: sh: Fix up HAVE_PERF_COUNTERS typo.

That's HAVE_PERF_COUNTERS not HAVE_PERF_COUNTER. This was right
initially but I seem to have screwed it up while re-typing it out
by hand on another machine when I checked it in. Hmph.

Reported-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index eff0f2352c93..e2bdd7b94fd9 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -15,7 +15,7 @@ config SUPERH
 	select HAVE_IOREMAP_PROT if MMU
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_DMA_API_DEBUG
-	select HAVE_PERF_COUNTER
+	select HAVE_PERF_COUNTERS
 	select RTC_LIB
 	select GENERIC_ATOMIC64
 	help
-- 
cgit v1.2.3-59-g8ed1b


From cb4cbcf6b3cf79f80c157afdc8dd8221643d8481 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Tue, 23 Jun 2009 08:57:55 +0900
Subject: mm: fix incorrect page removal from LRU

The isolated page is "cursor_page" not "page".

This could cause LRU list corruption under memory pressure, caught by
CONFIG_DEBUG_LIST.

Reported-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Tested-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/vmscan.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index e8fa2d9eb212..54155268dfca 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -932,7 +932,7 @@ static unsigned long isolate_lru_pages(unsigned long nr_to_scan,
 				continue;
 			if (__isolate_lru_page(cursor_page, mode, file) == 0) {
 				list_move(&cursor_page->lru, dst);
-				mem_cgroup_del_lru(page);
+				mem_cgroup_del_lru(cursor_page);
 				nr_taken++;
 				scan++;
 			}
-- 
cgit v1.2.3-59-g8ed1b


From 622a8f5f7bf160507861cf05309020049f42976d Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Tue, 23 Jun 2009 18:20:10 +0100
Subject: [ARM] s3c2410_defconfig: add MINI2440 machine to build

Add the MINI2440 to the list of machines built by the
central defconfig.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/configs/s3c2410_defconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm/configs/s3c2410_defconfig b/arch/arm/configs/s3c2410_defconfig
index f4f1899f3c88..b49810461e41 100644
--- a/arch/arm/configs/s3c2410_defconfig
+++ b/arch/arm/configs/s3c2410_defconfig
@@ -260,6 +260,7 @@ CONFIG_MACH_NEXCODER_2440=y
 CONFIG_SMDK2440_CPU2440=y
 CONFIG_MACH_AT2440EVB=y
 CONFIG_CPU_S3C2442=y
+CONFIG_MACH_MINI2440=y
 
 #
 # S3C2442 Machines
-- 
cgit v1.2.3-59-g8ed1b


From 51af243c631d9d03376d1454a24cf0a96e6f17cb Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Tue, 23 Jun 2009 18:23:27 +0100
Subject: [ARM] MINI2440: Add missing flash_bbt flat to NAND

The commit 9db41f9edcb87ae050fcb171c44be7f212728d54 added
the .flash_bbt flag to the nand set, so add this back into
the mach-mini2440.c file (taken out on initial commit to
allow build).

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/mach-s3c2440/mach-mini2440.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm/mach-s3c2440/mach-mini2440.c b/arch/arm/mach-s3c2440/mach-mini2440.c
index e714bfabbc09..ec71a6965786 100644
--- a/arch/arm/mach-s3c2440/mach-mini2440.c
+++ b/arch/arm/mach-s3c2440/mach-mini2440.c
@@ -273,6 +273,7 @@ static struct s3c2410_nand_set mini2440_nand_sets[] __initdata = {
 		.nr_chips	= 1,
 		.nr_partitions	= ARRAY_SIZE(mini2440_default_nand_part),
 		.partitions	= mini2440_default_nand_part,
+		.flash_bbt 	= 1, /* we use u-boot to create a BBT */
 	},
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 1d4d6da535be97b710e87a33c4828c97c36eee21 Mon Sep 17 00:00:00 2001
From: Alexander Schmidt <alexs@linux.vnet.ibm.com>
Date: Tue, 23 Jun 2009 10:30:04 -0700
Subject: IB/ehca: Bump version number

Increment version number for DMEM toleration.

Signed-off-by: Alexander Schmidt <alexs@linux.vnet.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/ehca/ehca_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/ehca/ehca_main.c b/drivers/infiniband/hw/ehca/ehca_main.c
index 14a18b7be245..fab18a2c74a8 100644
--- a/drivers/infiniband/hw/ehca/ehca_main.c
+++ b/drivers/infiniband/hw/ehca/ehca_main.c
@@ -52,7 +52,7 @@
 #include "ehca_tools.h"
 #include "hcp_if.h"
 
-#define HCAD_VERSION "0027"
+#define HCAD_VERSION "0028"
 
 MODULE_LICENSE("Dual BSD/GPL");
 MODULE_AUTHOR("Christoph Raisch <raisch@de.ibm.com>");
-- 
cgit v1.2.3-59-g8ed1b


From 716abb1fdf3274ac81dc404f3659cc05d8cdf606 Mon Sep 17 00:00:00 2001
From: Peter Huewe <peterhuewe@gmx.de>
Date: Tue, 23 Jun 2009 10:38:42 -0700
Subject: RDMA: Add __init/__exit macros to addr.c and cma.c

Add __init and __exit annotations to the module_init/module_exit
functions from drivers/infiniband/core/addr.c and cma.c.

Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Acked-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/addr.c | 4 ++--
 drivers/infiniband/core/cma.c  | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c
index ce511d8748ce..5be1bd4fc7ed 100644
--- a/drivers/infiniband/core/addr.c
+++ b/drivers/infiniband/core/addr.c
@@ -514,7 +514,7 @@ static struct notifier_block nb = {
 	.notifier_call = netevent_callback
 };
 
-static int addr_init(void)
+static int __init addr_init(void)
 {
 	addr_wq = create_singlethread_workqueue("ib_addr");
 	if (!addr_wq)
@@ -524,7 +524,7 @@ static int addr_init(void)
 	return 0;
 }
 
-static void addr_cleanup(void)
+static void __exit addr_cleanup(void)
 {
 	unregister_netevent_notifier(&nb);
 	destroy_workqueue(addr_wq);
diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index 851de83ff455..075317884b53 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -2960,7 +2960,7 @@ static void cma_remove_one(struct ib_device *device)
 	kfree(cma_dev);
 }
 
-static int cma_init(void)
+static int __init cma_init(void)
 {
 	int ret, low, high, remaining;
 
@@ -2990,7 +2990,7 @@ err:
 	return ret;
 }
 
-static void cma_cleanup(void)
+static void __exit cma_cleanup(void)
 {
 	ib_unregister_client(&cma_client);
 	unregister_netdevice_notifier(&cma_nb);
-- 
cgit v1.2.3-59-g8ed1b


From 788c7df451467df71638dd79a2d63d78c6e13b9c Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Date: Tue, 23 Jun 2009 13:49:05 +0100
Subject: hugetlb: fault flags instead of write_access

handle_mm_fault() is now passing fault flags rather than write_access
down to hugetlb_fault(), so better recognize that in hugetlb_fault(),
and in hugetlb_no_page().

Signed-off-by: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Acked-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/hugetlb.h |  4 ++--
 mm/hugetlb.c            | 17 +++++++++--------
 2 files changed, 11 insertions(+), 10 deletions(-)

diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index a05a5ef33391..2723513a5651 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -33,7 +33,7 @@ void hugetlb_report_meminfo(struct seq_file *);
 int hugetlb_report_node_meminfo(int, char *);
 unsigned long hugetlb_total_pages(void);
 int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
-			unsigned long address, int write_access);
+			unsigned long address, unsigned int flags);
 int hugetlb_reserve_pages(struct inode *inode, long from, long to,
 						struct vm_area_struct *vma,
 						int acctflags);
@@ -98,7 +98,7 @@ static inline void hugetlb_report_meminfo(struct seq_file *m)
 #define pud_huge(x)	0
 #define is_hugepage_only_range(mm, addr, len)	0
 #define hugetlb_free_pgd_range(tlb, addr, end, floor, ceiling) ({BUG(); 0; })
-#define hugetlb_fault(mm, vma, addr, write)	({ BUG(); 0; })
+#define hugetlb_fault(mm, vma, addr, flags)	({ BUG(); 0; })
 
 #define hugetlb_change_protection(vma, address, end, newprot)
 
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index a56e6f3ce979..d0351e31f474 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1985,7 +1985,7 @@ static struct page *hugetlbfs_pagecache_page(struct hstate *h,
 }
 
 static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
-			unsigned long address, pte_t *ptep, int write_access)
+			unsigned long address, pte_t *ptep, unsigned int flags)
 {
 	struct hstate *h = hstate_vma(vma);
 	int ret = VM_FAULT_SIGBUS;
@@ -2053,7 +2053,7 @@ retry:
 	 * any allocations necessary to record that reservation occur outside
 	 * the spinlock.
 	 */
-	if (write_access && !(vma->vm_flags & VM_SHARED))
+	if ((flags & FAULT_FLAG_WRITE) && !(vma->vm_flags & VM_SHARED))
 		if (vma_needs_reservation(h, vma, address) < 0) {
 			ret = VM_FAULT_OOM;
 			goto backout_unlocked;
@@ -2072,7 +2072,7 @@ retry:
 				&& (vma->vm_flags & VM_SHARED)));
 	set_huge_pte_at(mm, address, ptep, new_pte);
 
-	if (write_access && !(vma->vm_flags & VM_SHARED)) {
+	if ((flags & FAULT_FLAG_WRITE) && !(vma->vm_flags & VM_SHARED)) {
 		/* Optimization, do the COW without a second fault */
 		ret = hugetlb_cow(mm, vma, address, ptep, new_pte, page);
 	}
@@ -2091,7 +2091,7 @@ backout_unlocked:
 }
 
 int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
-			unsigned long address, int write_access)
+			unsigned long address, unsigned int flags)
 {
 	pte_t *ptep;
 	pte_t entry;
@@ -2112,7 +2112,7 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	mutex_lock(&hugetlb_instantiation_mutex);
 	entry = huge_ptep_get(ptep);
 	if (huge_pte_none(entry)) {
-		ret = hugetlb_no_page(mm, vma, address, ptep, write_access);
+		ret = hugetlb_no_page(mm, vma, address, ptep, flags);
 		goto out_mutex;
 	}
 
@@ -2126,7 +2126,7 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	 * page now as it is used to determine if a reservation has been
 	 * consumed.
 	 */
-	if (write_access && !pte_write(entry)) {
+	if ((flags & FAULT_FLAG_WRITE) && !pte_write(entry)) {
 		if (vma_needs_reservation(h, vma, address) < 0) {
 			ret = VM_FAULT_OOM;
 			goto out_mutex;
@@ -2143,7 +2143,7 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 		goto out_page_table_lock;
 
 
-	if (write_access) {
+	if (flags & FAULT_FLAG_WRITE) {
 		if (!pte_write(entry)) {
 			ret = hugetlb_cow(mm, vma, address, ptep, entry,
 							pagecache_page);
@@ -2152,7 +2152,8 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 		entry = pte_mkdirty(entry);
 	}
 	entry = pte_mkyoung(entry);
-	if (huge_ptep_set_access_flags(vma, address, ptep, entry, write_access))
+	if (huge_ptep_set_access_flags(vma, address, ptep, entry,
+						flags & FAULT_FLAG_WRITE))
 		update_mmu_cache(vma, address, entry);
 
 out_page_table_lock:
-- 
cgit v1.2.3-59-g8ed1b


From d26ed650d9947a786bbda8de9cd914dbeebc1a68 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Date: Tue, 23 Jun 2009 13:52:49 +0100
Subject: mm: don't rely on flags coincidence

Indeed FOLL_WRITE matches FAULT_FLAG_WRITE, matches GUP_FLAGS_WRITE,
and it's tempting to devise a set of Grand Unified Paging flags;
but not today.  So until then, let's rely upon the compiler to spot
the coincidence, "rather than have that subtle dependency and a
comment for it" - as you remarked in another context yesterday.

Signed-off-by: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Acked-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/memory.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/mm/memory.c b/mm/memory.c
index 98bcb90d5957..50da9511aa77 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1311,8 +1311,10 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 			while (!(page = follow_page(vma, start, foll_flags))) {
 				int ret;
 
-				/* FOLL_WRITE matches FAULT_FLAG_WRITE! */
-				ret = handle_mm_fault(mm, vma, start, foll_flags & FOLL_WRITE);
+				ret = handle_mm_fault(mm, vma, start,
+					(foll_flags & FOLL_WRITE) ?
+					FAULT_FLAG_WRITE : 0);
+
 				if (ret & VM_FAULT_ERROR) {
 					if (ret & VM_FAULT_OOM)
 						return i ? i : -ENOMEM;
-- 
cgit v1.2.3-59-g8ed1b


From b8389018212e8c4e03ede4df5405796100ef4390 Mon Sep 17 00:00:00 2001
From: Kim Kyuwon <chammoru@gmail.com>
Date: Wed, 10 Jun 2009 12:48:48 -0700
Subject: leds: fix led-bd2802 errors while resuming

LED_CORE_SUSPENDRESUME flag is not needed in the bd2802 driver, because
all works for suspend/resume is done in bd2802_suspend and bd2802_suspend
functions.  And this patch allows bd2802 to be configured again when it
resumes from suspend.

Signed-off-by: Kim Kyuwon <q1.kim@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 drivers/leds/leds-bd2802.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/drivers/leds/leds-bd2802.c b/drivers/leds/leds-bd2802.c
index 4149ecb3a9b2..6832b9fd0422 100644
--- a/drivers/leds/leds-bd2802.c
+++ b/drivers/leds/leds-bd2802.c
@@ -538,7 +538,6 @@ static int bd2802_register_led_classdev(struct bd2802_led *led)
 	led->cdev_led1r.brightness = LED_OFF;
 	led->cdev_led1r.brightness_set = bd2802_set_led1r_brightness;
 	led->cdev_led1r.blink_set = bd2802_set_led1r_blink;
-	led->cdev_led1r.flags |= LED_CORE_SUSPENDRESUME;
 
 	ret = led_classdev_register(&led->client->dev, &led->cdev_led1r);
 	if (ret < 0) {
@@ -551,7 +550,6 @@ static int bd2802_register_led_classdev(struct bd2802_led *led)
 	led->cdev_led1g.brightness = LED_OFF;
 	led->cdev_led1g.brightness_set = bd2802_set_led1g_brightness;
 	led->cdev_led1g.blink_set = bd2802_set_led1g_blink;
-	led->cdev_led1g.flags |= LED_CORE_SUSPENDRESUME;
 
 	ret = led_classdev_register(&led->client->dev, &led->cdev_led1g);
 	if (ret < 0) {
@@ -564,7 +562,6 @@ static int bd2802_register_led_classdev(struct bd2802_led *led)
 	led->cdev_led1b.brightness = LED_OFF;
 	led->cdev_led1b.brightness_set = bd2802_set_led1b_brightness;
 	led->cdev_led1b.blink_set = bd2802_set_led1b_blink;
-	led->cdev_led1b.flags |= LED_CORE_SUSPENDRESUME;
 
 	ret = led_classdev_register(&led->client->dev, &led->cdev_led1b);
 	if (ret < 0) {
@@ -577,7 +574,6 @@ static int bd2802_register_led_classdev(struct bd2802_led *led)
 	led->cdev_led2r.brightness = LED_OFF;
 	led->cdev_led2r.brightness_set = bd2802_set_led2r_brightness;
 	led->cdev_led2r.blink_set = bd2802_set_led2r_blink;
-	led->cdev_led2r.flags |= LED_CORE_SUSPENDRESUME;
 
 	ret = led_classdev_register(&led->client->dev, &led->cdev_led2r);
 	if (ret < 0) {
@@ -590,7 +586,6 @@ static int bd2802_register_led_classdev(struct bd2802_led *led)
 	led->cdev_led2g.brightness = LED_OFF;
 	led->cdev_led2g.brightness_set = bd2802_set_led2g_brightness;
 	led->cdev_led2g.blink_set = bd2802_set_led2g_blink;
-	led->cdev_led2g.flags |= LED_CORE_SUSPENDRESUME;
 
 	ret = led_classdev_register(&led->client->dev, &led->cdev_led2g);
 	if (ret < 0) {
@@ -723,8 +718,7 @@ static int bd2802_resume(struct i2c_client *client)
 	struct bd2802_led *led = i2c_get_clientdata(client);
 
 	if (!bd2802_is_all_off(led) || led->adf_on) {
-		gpio_set_value(led->pdata->reset_gpio, 1);
-		udelay(100);
+		bd2802_reset_cancel(led);
 		bd2802_restore_state(led);
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 1b18cf413f63ff6de5ba3e5028e869c21322a4df Mon Sep 17 00:00:00 2001
From: Kim Kyuwon <chammoru@gmail.com>
Date: Wed, 10 Jun 2009 12:48:50 -0700
Subject: leds: change the license information

Change the license to 'GPL v2'

Signed-off-by: Kim Kyuwon <q1.kim@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 drivers/leds/leds-bd2802.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/leds/leds-bd2802.c b/drivers/leds/leds-bd2802.c
index 6832b9fd0422..7a03efd54f69 100644
--- a/drivers/leds/leds-bd2802.c
+++ b/drivers/leds/leds-bd2802.c
@@ -756,4 +756,4 @@ module_exit(bd2802_exit);
 
 MODULE_AUTHOR("Kim Kyuwon <q1.kim@samsung.com>");
 MODULE_DESCRIPTION("BD2802 LED driver");
-MODULE_LICENSE("GPL");
+MODULE_LICENSE("GPL v2");
-- 
cgit v1.2.3-59-g8ed1b


From 8792f7cf4368f9bc337eee65851d8e7abbbf946c Mon Sep 17 00:00:00 2001
From: Kim Kyuwon <chammoru@gmail.com>
Date: Wed, 10 Jun 2009 12:48:50 -0700
Subject: leds: add the sysfs interface into the leds-bd2802 driver for
 changing wave pattern and led current.

Allow the user application to change the wave pattern and led current by
'wave_pattern' and 'rgb_current' sysfs files.

Signed-off-by: Kim Kyuwon <q1.kim@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 drivers/leds/leds-bd2802.c | 86 +++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 73 insertions(+), 13 deletions(-)

diff --git a/drivers/leds/leds-bd2802.c b/drivers/leds/leds-bd2802.c
index 7a03efd54f69..779d7f262c04 100644
--- a/drivers/leds/leds-bd2802.c
+++ b/drivers/leds/leds-bd2802.c
@@ -97,6 +97,10 @@ struct bd2802_led {
 	enum led_ids			led_id;
 	enum led_colors			color;
 	enum led_bits			state;
+
+	/* General attributes of RGB LEDs */
+	int				wave_pattern;
+	int				rgb_current;
 };
 
 
@@ -254,7 +258,7 @@ static void bd2802_set_on(struct bd2802_led *led, enum led_ids id,
 		bd2802_reset_cancel(led);
 
 	reg = bd2802_get_reg_addr(id, color, BD2802_REG_CURRENT1SETUP);
-	bd2802_write_byte(led->client, reg, BD2802_CURRENT_032);
+	bd2802_write_byte(led->client, reg, led->rgb_current);
 	reg = bd2802_get_reg_addr(id, color, BD2802_REG_CURRENT2SETUP);
 	bd2802_write_byte(led->client, reg, BD2802_CURRENT_000);
 	reg = bd2802_get_reg_addr(id, color, BD2802_REG_WAVEPATTERN);
@@ -275,9 +279,9 @@ static void bd2802_set_blink(struct bd2802_led *led, enum led_ids id,
 	reg = bd2802_get_reg_addr(id, color, BD2802_REG_CURRENT1SETUP);
 	bd2802_write_byte(led->client, reg, BD2802_CURRENT_000);
 	reg = bd2802_get_reg_addr(id, color, BD2802_REG_CURRENT2SETUP);
-	bd2802_write_byte(led->client, reg, BD2802_CURRENT_032);
+	bd2802_write_byte(led->client, reg, led->rgb_current);
 	reg = bd2802_get_reg_addr(id, color, BD2802_REG_WAVEPATTERN);
-	bd2802_write_byte(led->client, reg, BD2802_PATTERN_HALF);
+	bd2802_write_byte(led->client, reg, led->wave_pattern);
 
 	bd2802_enable(led, id);
 	bd2802_update_state(led, id, color, BD2802_BLINK);
@@ -406,7 +410,7 @@ static void bd2802_enable_adv_conf(struct bd2802_led *led)
 		ret = device_create_file(&led->client->dev,
 						bd2802_addr_attributes[i]);
 		if (ret) {
-			dev_err(&led->client->dev, "failed to sysfs file %s\n",
+			dev_err(&led->client->dev, "failed: sysfs file %s\n",
 					bd2802_addr_attributes[i]->attr.name);
 			goto failed_remove_files;
 		}
@@ -483,6 +487,52 @@ static struct device_attribute bd2802_adv_conf_attr = {
 	.store = bd2802_store_adv_conf,
 };
 
+#define BD2802_CONTROL_ATTR(attr_name, name_str)			\
+static ssize_t bd2802_show_##attr_name(struct device *dev,		\
+	struct device_attribute *attr, char *buf)			\
+{									\
+	struct bd2802_led *led = i2c_get_clientdata(to_i2c_client(dev));\
+	ssize_t ret;							\
+	down_read(&led->rwsem);						\
+	ret = sprintf(buf, "0x%02x\n", led->attr_name);			\
+	up_read(&led->rwsem);						\
+	return ret;							\
+}									\
+static ssize_t bd2802_store_##attr_name(struct device *dev,		\
+	struct device_attribute *attr, const char *buf, size_t count)	\
+{									\
+	struct bd2802_led *led = i2c_get_clientdata(to_i2c_client(dev));\
+	unsigned long val;						\
+	int ret;							\
+	if (!count)							\
+		return -EINVAL;						\
+	ret = strict_strtoul(buf, 16, &val);				\
+	if (ret)							\
+		return ret;						\
+	down_write(&led->rwsem);					\
+	led->attr_name = val;						\
+	up_write(&led->rwsem);						\
+	return count;							\
+}									\
+static struct device_attribute bd2802_##attr_name##_attr = {		\
+	.attr = {							\
+		.name = name_str,					\
+		.mode = 0644,						\
+		.owner = THIS_MODULE					\
+	},								\
+	.show = bd2802_show_##attr_name,				\
+	.store = bd2802_store_##attr_name,				\
+};
+
+BD2802_CONTROL_ATTR(wave_pattern, "wave_pattern");
+BD2802_CONTROL_ATTR(rgb_current, "rgb_current");
+
+static struct device_attribute *bd2802_attributes[] = {
+	&bd2802_adv_conf_attr,
+	&bd2802_wave_pattern_attr,
+	&bd2802_rgb_current_attr,
+};
+
 static void bd2802_led_work(struct work_struct *work)
 {
 	struct bd2802_led *led = container_of(work, struct bd2802_led, work);
@@ -635,7 +685,7 @@ static int __devinit bd2802_probe(struct i2c_client *client,
 {
 	struct bd2802_led *led;
 	struct bd2802_led_platform_data *pdata;
-	int ret;
+	int ret, i;
 
 	led = kzalloc(sizeof(struct bd2802_led), GFP_KERNEL);
 	if (!led) {
@@ -665,13 +715,20 @@ static int __devinit bd2802_probe(struct i2c_client *client,
 	/* To save the power, reset BD2802 after detecting */
 	gpio_set_value(led->pdata->reset_gpio, 0);
 
+	/* Default attributes */
+	led->wave_pattern = BD2802_PATTERN_HALF;
+	led->rgb_current = BD2802_CURRENT_032;
+
 	init_rwsem(&led->rwsem);
 
-	ret = device_create_file(&client->dev, &bd2802_adv_conf_attr);
-	if (ret) {
-		dev_err(&client->dev, "failed to create sysfs file %s\n",
-					bd2802_adv_conf_attr.attr.name);
-		goto failed_free;
+	for (i = 0; i < ARRAY_SIZE(bd2802_attributes); i++) {
+		ret = device_create_file(&led->client->dev,
+						bd2802_attributes[i]);
+		if (ret) {
+			dev_err(&led->client->dev, "failed: sysfs file %s\n",
+					bd2802_attributes[i]->attr.name);
+			goto failed_unregister_dev_file;
+		}
 	}
 
 	ret = bd2802_register_led_classdev(led);
@@ -681,7 +738,8 @@ static int __devinit bd2802_probe(struct i2c_client *client,
 	return 0;
 
 failed_unregister_dev_file:
-	device_remove_file(&client->dev, &bd2802_adv_conf_attr);
+	for (i--; i >= 0; i--)
+		device_remove_file(&led->client->dev, bd2802_attributes[i]);
 failed_free:
 	i2c_set_clientdata(client, NULL);
 	kfree(led);
@@ -692,12 +750,14 @@ failed_free:
 static int __exit bd2802_remove(struct i2c_client *client)
 {
 	struct bd2802_led *led = i2c_get_clientdata(client);
+	int i;
 
-	bd2802_unregister_led_classdev(led);
 	gpio_set_value(led->pdata->reset_gpio, 0);
+	bd2802_unregister_led_classdev(led);
 	if (led->adf_on)
 		bd2802_disable_adv_conf(led);
-	device_remove_file(&client->dev, &bd2802_adv_conf_attr);
+	for (i = 0; i < ARRAY_SIZE(bd2802_attributes); i++)
+		device_remove_file(&led->client->dev, bd2802_attributes[i]);
 	i2c_set_clientdata(client, NULL);
 	kfree(led);
 
-- 
cgit v1.2.3-59-g8ed1b


From 7fd02170e25b3b60fc21cd7b64bf1ed42e6a7cbe Mon Sep 17 00:00:00 2001
From: Zhenwen Xu <helight.xu@gmail.com>
Date: Wed, 10 Jun 2009 12:48:51 -0700
Subject: leds: leds-gpio - fix a section mismatch

WARNING: drivers/leds/leds-gpio.o(.text+0x153): Section mismatch in reference from the function gpio_led_probe() to the function .devinit.text:create_gpio_led()

The function gpio_led_probe() references the function __devinit
create_gpio_led().  This is often because gpio_led_probe lacks a __devinit
annotation or the annotation of create_gpio_led is wrong.

Signed-off-by: Zhenwen Xu <helight.xu@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 drivers/leds/leds-gpio.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/leds/leds-gpio.c b/drivers/leds/leds-gpio.c
index d2109054de85..76895e691042 100644
--- a/drivers/leds/leds-gpio.c
+++ b/drivers/leds/leds-gpio.c
@@ -129,7 +129,7 @@ static void delete_gpio_led(struct gpio_led_data *led)
 }
 
 #ifdef CONFIG_LEDS_GPIO_PLATFORM
-static int gpio_led_probe(struct platform_device *pdev)
+static int __devinit gpio_led_probe(struct platform_device *pdev)
 {
 	struct gpio_led_platform_data *pdata = pdev->dev.platform_data;
 	struct gpio_led_data *leds_data;
-- 
cgit v1.2.3-59-g8ed1b


From 2216c6e83ccbc9d34f541621ff23f510cd8a256f Mon Sep 17 00:00:00 2001
From: Tobias Mueller <Tobias_Mueller@twam.info>
Date: Wed, 10 Jun 2009 12:48:52 -0700
Subject: leds: alix-leds2 fixed for Award BIOS

Add initialisation of GPIO ports for compatibility with boards with Award
BIOS (e.g.  ALIX.3D3).

Signed-off-by: Tobias Mueller <Tobias_Mueller@twam.info>
Reviewed-by: Constantin Baranov <const@mimas.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 drivers/leds/Kconfig      | 1 +
 drivers/leds/leds-alix2.c | 7 ++++++-
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/leds/Kconfig b/drivers/leds/Kconfig
index 9b60b6b684d9..37e91184756d 100644
--- a/drivers/leds/Kconfig
+++ b/drivers/leds/Kconfig
@@ -75,6 +75,7 @@ config LEDS_ALIX2
 	depends on LEDS_CLASS && X86 && EXPERIMENTAL
 	help
 	  This option enables support for the PCEngines ALIX.2 and ALIX.3 LEDs.
+	  You have to set leds-alix2.force=1 for boards with Award BIOS.
 
 config LEDS_H1940
 	tristate "LED Support for iPAQ H1940 device"
diff --git a/drivers/leds/leds-alix2.c b/drivers/leds/leds-alix2.c
index ddbd7730dfc8..731d4eef3425 100644
--- a/drivers/leds/leds-alix2.c
+++ b/drivers/leds/leds-alix2.c
@@ -14,7 +14,7 @@
 
 static int force = 0;
 module_param(force, bool, 0444);
-MODULE_PARM_DESC(force, "Assume system has ALIX.2 style LEDs");
+MODULE_PARM_DESC(force, "Assume system has ALIX.2/ALIX.3 style LEDs");
 
 struct alix_led {
 	struct led_classdev cdev;
@@ -155,6 +155,11 @@ static int __init alix_led_init(void)
 		goto out;
 	}
 
+	/* enable output on GPIO for LED 1,2,3 */
+	outl(1 << 6, 0x6104);
+	outl(1 << 9, 0x6184);
+	outl(1 << 11, 0x6184);
+
 	pdev = platform_device_register_simple(KBUILD_MODNAME, -1, NULL, 0);
 	if (!IS_ERR(pdev)) {
 		ret = platform_driver_probe(&alix_led_driver, alix_led_probe);
-- 
cgit v1.2.3-59-g8ed1b


From 92722b1bb1ebcba767f9c6ee499992ee33367268 Mon Sep 17 00:00:00 2001
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
Date: Thu, 11 Jun 2009 14:17:48 +0100
Subject: leds: Further document parameters for blink_set()

The documentation for the parameters of blink_set() was a bit hard
to find so put some where I'd expected to find it.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 include/linux/leds.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/include/linux/leds.h b/include/linux/leds.h
index 376fe07732ea..c7f0b148df06 100644
--- a/include/linux/leds.h
+++ b/include/linux/leds.h
@@ -45,7 +45,9 @@ struct led_classdev {
 	/* Get LED brightness level */
 	enum led_brightness (*brightness_get)(struct led_classdev *led_cdev);
 
-	/* Activate hardware accelerated blink */
+	/* Activate hardware accelerated blink, delays are in
+	 * miliseconds and if none is provided then a sensible default
+	 * should be chosen. */
 	int		(*blink_set)(struct led_classdev *led_cdev,
 				     unsigned long *delay_on,
 				     unsigned long *delay_off);
-- 
cgit v1.2.3-59-g8ed1b


From 34abdf252699ebc549fad54c1db481612f22a826 Mon Sep 17 00:00:00 2001
From: Richard Purdie <rpurdie@linux.intel.com>
Date: Wed, 17 Jun 2009 13:05:27 +0100
Subject: leds: Remove an orphan Kconfig entry

Remove an orphan Kconfig entry (LEDS_LP5521)

Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 drivers/leds/Kconfig | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/drivers/leds/Kconfig b/drivers/leds/Kconfig
index 37e91184756d..cfcd6bf831c9 100644
--- a/drivers/leds/Kconfig
+++ b/drivers/leds/Kconfig
@@ -146,16 +146,6 @@ config LEDS_GPIO_OF
 	  of_platform devices.  For instance, LEDs which are listed in a "dts"
 	  file.
 
-config LEDS_LP5521
-	tristate "LED Support for the LP5521 LEDs"
-	depends on LEDS_CLASS && I2C
-	help
-	  If you say 'Y' here you get support for the National Semiconductor
-	  LP5521 LED driver used in n8x0 boards.
-
-	  This driver can be built as a module by choosing 'M'. The module
-	  will be called leds-lp5521.
-
 config LEDS_CLEVO_MAIL
 	tristate "Mail LED on Clevo notebook"
 	depends on LEDS_CLASS && X86 && SERIO_I8042 && DMI
-- 
cgit v1.2.3-59-g8ed1b


From 07172d2bfa339d6c150d8cdd7c02128177feffbb Mon Sep 17 00:00:00 2001
From: Antonio Ospite <ospite@studenti.unina.it>
Date: Fri, 19 Jun 2009 13:53:07 +0200
Subject: leds: pca9532 - Indent using tabs, not spaces.

Indent using tabs, not spaces.

Signed-off-by: Antonio Ospite <ospite@studenti.unina.it>
Acked-by: Riku Voipio <riku.voipio@iki.fi>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 drivers/leds/leds-pca9532.c | 58 ++++++++++++++++++++++-----------------------
 1 file changed, 29 insertions(+), 29 deletions(-)

diff --git a/drivers/leds/leds-pca9532.c b/drivers/leds/leds-pca9532.c
index 3937244fdcab..dba8921240f2 100644
--- a/drivers/leds/leds-pca9532.c
+++ b/drivers/leds/leds-pca9532.c
@@ -35,7 +35,7 @@ struct pca9532_data {
 	struct pca9532_led leds[16];
 	struct mutex update_lock;
 	struct input_dev    *idev;
-       struct work_struct work;
+	struct work_struct work;
 	u8 pwm[2];
 	u8 psc[2];
 };
@@ -87,14 +87,14 @@ static int pca9532_calcpwm(struct i2c_client *client, int pwm, int blink,
 	if (b > 0xFF)
 		return -EINVAL;
 	data->pwm[pwm] = b;
-       data->psc[pwm] = blink;
-       return 0;
+	data->psc[pwm] = blink;
+	return 0;
 }
 
 static int pca9532_setpwm(struct i2c_client *client, int pwm)
 {
-       struct pca9532_data *data = i2c_get_clientdata(client);
-       mutex_lock(&data->update_lock);
+	struct pca9532_data *data = i2c_get_clientdata(client);
+	mutex_lock(&data->update_lock);
 	i2c_smbus_write_byte_data(client, PCA9532_REG_PWM(pwm),
 		data->pwm[pwm]);
 	i2c_smbus_write_byte_data(client, PCA9532_REG_PSC(pwm),
@@ -132,11 +132,11 @@ static void pca9532_set_brightness(struct led_classdev *led_cdev,
 		led->state = PCA9532_ON;
 	else {
 		led->state = PCA9532_PWM0; /* Thecus: hardcode one pwm */
-               err = pca9532_calcpwm(led->client, 0, 0, value);
+		err = pca9532_calcpwm(led->client, 0, 0, value);
 		if (err)
 			return; /* XXX: led api doesn't allow error code? */
 	}
-       schedule_work(&led->work);
+	schedule_work(&led->work);
 }
 
 static int pca9532_set_blink(struct led_classdev *led_cdev,
@@ -145,7 +145,7 @@ static int pca9532_set_blink(struct led_classdev *led_cdev,
 	struct pca9532_led *led = ldev_to_led(led_cdev);
 	struct i2c_client *client = led->client;
 	int psc;
-       int err = 0;
+	int err = 0;
 
 	if (*delay_on == 0 && *delay_off == 0) {
 	/* led subsystem ask us for a blink rate */
@@ -157,11 +157,11 @@ static int pca9532_set_blink(struct led_classdev *led_cdev,
 
 	/* Thecus specific: only use PSC/PWM 0 */
 	psc = (*delay_on * 152-1)/1000;
-       err = pca9532_calcpwm(client, 0, psc, led_cdev->brightness);
-       if (err)
-               return err;
-       schedule_work(&led->work);
-       return 0;
+	err = pca9532_calcpwm(client, 0, psc, led_cdev->brightness);
+	if (err)
+		return err;
+	schedule_work(&led->work);
+	return 0;
 }
 
 static int pca9532_event(struct input_dev *dev, unsigned int type,
@@ -178,15 +178,15 @@ static int pca9532_event(struct input_dev *dev, unsigned int type,
 	else
 		data->pwm[1] = 0;
 
-       schedule_work(&data->work);
+	schedule_work(&data->work);
 
-       return 0;
+	return 0;
 }
 
 static void pca9532_input_work(struct work_struct *work)
 {
-       struct pca9532_data *data;
-       data = container_of(work, struct pca9532_data, work);
+	struct pca9532_data *data;
+	data = container_of(work, struct pca9532_data, work);
 	mutex_lock(&data->update_lock);
 	i2c_smbus_write_byte_data(data->client, PCA9532_REG_PWM(1),
 		data->pwm[1]);
@@ -195,11 +195,11 @@ static void pca9532_input_work(struct work_struct *work)
 
 static void pca9532_led_work(struct work_struct *work)
 {
-       struct pca9532_led *led;
-       led = container_of(work, struct pca9532_led, work);
-       if (led->state == PCA9532_PWM0)
-               pca9532_setpwm(led->client, 0);
-       pca9532_setled(led);
+	struct pca9532_led *led;
+	led = container_of(work, struct pca9532_led, work);
+	if (led->state == PCA9532_PWM0)
+		pca9532_setpwm(led->client, 0);
+	pca9532_setled(led);
 }
 
 static int pca9532_configure(struct i2c_client *client,
@@ -232,7 +232,7 @@ static int pca9532_configure(struct i2c_client *client,
 			led->ldev.brightness = LED_OFF;
 			led->ldev.brightness_set = pca9532_set_brightness;
 			led->ldev.blink_set = pca9532_set_blink;
-                       INIT_WORK(&led->work, pca9532_led_work);
+			INIT_WORK(&led->work, pca9532_led_work);
 			err = led_classdev_register(&client->dev, &led->ldev);
 			if (err < 0) {
 				dev_err(&client->dev,
@@ -262,11 +262,11 @@ static int pca9532_configure(struct i2c_client *client,
 						BIT_MASK(SND_TONE);
 			data->idev->event = pca9532_event;
 			input_set_drvdata(data->idev, data);
-                       INIT_WORK(&data->work, pca9532_input_work);
+			INIT_WORK(&data->work, pca9532_input_work);
 			err = input_register_device(data->idev);
 			if (err) {
 				input_free_device(data->idev);
-                               cancel_work_sync(&data->work);
+				cancel_work_sync(&data->work);
 				data->idev = NULL;
 				goto exit;
 			}
@@ -283,13 +283,13 @@ exit:
 				break;
 			case PCA9532_TYPE_LED:
 				led_classdev_unregister(&data->leds[i].ldev);
-                               cancel_work_sync(&data->leds[i].work);
+				cancel_work_sync(&data->leds[i].work);
 				break;
 			case PCA9532_TYPE_N2100_BEEP:
 				if (data->idev != NULL) {
 					input_unregister_device(data->idev);
 					input_free_device(data->idev);
-                                       cancel_work_sync(&data->work);
+					cancel_work_sync(&data->work);
 					data->idev = NULL;
 				}
 				break;
@@ -340,13 +340,13 @@ static int pca9532_remove(struct i2c_client *client)
 			break;
 		case PCA9532_TYPE_LED:
 			led_classdev_unregister(&data->leds[i].ldev);
-                       cancel_work_sync(&data->leds[i].work);
+			cancel_work_sync(&data->leds[i].work);
 			break;
 		case PCA9532_TYPE_N2100_BEEP:
 			if (data->idev != NULL) {
 				input_unregister_device(data->idev);
 				input_free_device(data->idev);
-                               cancel_work_sync(&data->work);
+				cancel_work_sync(&data->work);
 				data->idev = NULL;
 			}
 			break;
-- 
cgit v1.2.3-59-g8ed1b


From 5054d39e327f76df022163a2ebd02e444c5d65f9 Mon Sep 17 00:00:00 2001
From: Antonio Ospite <ospite@studenti.unina.it>
Date: Fri, 19 Jun 2009 13:55:42 +0200
Subject: leds: LED driver for National Semiconductor LP3944 Funlight Chip

LEDs driver for National Semiconductor LP3944 Funlight Chip
http://www.national.com/pf/LP/LP3944.html

This helper chip can drive up to 8 leds, with two programmable DIM
modes; it could even be used as a gpio expander but this driver assumes
it is used as a led controller.

The DIM modes are used to set _blink_ patterns for leds, the pattern is
specified supplying two parameters:
  - period: from 0s to 1.6s
  - duty cycle: percentage of the period the led is on, from 0 to 100

LP3944 can be found on Motorola A910 smartphone, where it drives the rgb
leds, the camera flash light and the displays backlights.

Signed-off-by: Antonio Ospite <ospite@studenti.unina.it>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 Documentation/leds-lp3944.txt |  50 +++++
 drivers/leds/Kconfig          |  11 +
 drivers/leds/Makefile         |   1 +
 drivers/leds/leds-lp3944.c    | 466 ++++++++++++++++++++++++++++++++++++++++++
 include/linux/leds-lp3944.h   |  53 +++++
 5 files changed, 581 insertions(+)
 create mode 100644 Documentation/leds-lp3944.txt
 create mode 100644 drivers/leds/leds-lp3944.c
 create mode 100644 include/linux/leds-lp3944.h

diff --git a/Documentation/leds-lp3944.txt b/Documentation/leds-lp3944.txt
new file mode 100644
index 000000000000..c6eda18b15ef
--- /dev/null
+++ b/Documentation/leds-lp3944.txt
@@ -0,0 +1,50 @@
+Kernel driver lp3944
+====================
+
+  * National Semiconductor LP3944 Fun-light Chip
+    Prefix: 'lp3944'
+    Addresses scanned: None (see the Notes section below)
+    Datasheet: Publicly available at the National Semiconductor website
+               http://www.national.com/pf/LP/LP3944.html
+
+Authors:
+        Antonio Ospite <ospite@studenti.unina.it>
+
+
+Description
+-----------
+The LP3944 is a helper chip that can drive up to 8 leds, with two programmable
+DIM modes; it could even be used as a gpio expander but this driver assumes it
+is used as a led controller.
+
+The DIM modes are used to set _blink_ patterns for leds, the pattern is
+specified supplying two parameters:
+  - period: from 0s to 1.6s
+  - duty cycle: percentage of the period the led is on, from 0 to 100
+
+Setting a led in DIM0 or DIM1 mode makes it blink according to the pattern.
+See the datasheet for details.
+
+LP3944 can be found on Motorola A910 smartphone, where it drives the rgb
+leds, the camera flash light and the lcds power.
+
+
+Notes
+-----
+The chip is used mainly in embedded contexts, so this driver expects it is
+registered using the i2c_board_info mechanism.
+
+To register the chip at address 0x60 on adapter 0, set the platform data
+according to include/linux/leds-lp3944.h, set the i2c board info:
+
+	static struct i2c_board_info __initdata a910_i2c_board_info[] = {
+		{
+			I2C_BOARD_INFO("lp3944", 0x60),
+			.platform_data = &a910_lp3944_leds,
+		},
+	};
+
+and register it in the platform init function
+
+	i2c_register_board_info(0, a910_i2c_board_info,
+			ARRAY_SIZE(a910_i2c_board_info));
diff --git a/drivers/leds/Kconfig b/drivers/leds/Kconfig
index cfcd6bf831c9..7c8e7122aaa9 100644
--- a/drivers/leds/Kconfig
+++ b/drivers/leds/Kconfig
@@ -146,6 +146,17 @@ config LEDS_GPIO_OF
 	  of_platform devices.  For instance, LEDs which are listed in a "dts"
 	  file.
 
+config LEDS_LP3944
+	tristate "LED Support for N.S. LP3944 (Fun Light) I2C chip"
+	depends on LEDS_CLASS && I2C
+	help
+    This option enables support for LEDs connected to the National
+    Semiconductor LP3944 Lighting Management Unit (LMU) also known as
+    Fun Light Chip.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called leds-lp3944.
+
 config LEDS_CLEVO_MAIL
 	tristate "Mail LED on Clevo notebook"
 	depends on LEDS_CLASS && X86 && SERIO_I8042 && DMI
diff --git a/drivers/leds/Makefile b/drivers/leds/Makefile
index 2d41c4dcf92f..e8cdcf77a4c3 100644
--- a/drivers/leds/Makefile
+++ b/drivers/leds/Makefile
@@ -20,6 +20,7 @@ obj-$(CONFIG_LEDS_COBALT_RAQ)		+= leds-cobalt-raq.o
 obj-$(CONFIG_LEDS_SUNFIRE)		+= leds-sunfire.o
 obj-$(CONFIG_LEDS_PCA9532)		+= leds-pca9532.o
 obj-$(CONFIG_LEDS_GPIO)			+= leds-gpio.o
+obj-$(CONFIG_LEDS_LP3944)		+= leds-lp3944.o
 obj-$(CONFIG_LEDS_CLEVO_MAIL)		+= leds-clevo-mail.o
 obj-$(CONFIG_LEDS_HP6XX)		+= leds-hp6xx.o
 obj-$(CONFIG_LEDS_FSG)			+= leds-fsg.o
diff --git a/drivers/leds/leds-lp3944.c b/drivers/leds/leds-lp3944.c
new file mode 100644
index 000000000000..5946208ba26e
--- /dev/null
+++ b/drivers/leds/leds-lp3944.c
@@ -0,0 +1,466 @@
+/*
+ * leds-lp3944.c - driver for National Semiconductor LP3944 Funlight Chip
+ *
+ * Copyright (C) 2009 Antonio Ospite <ospite@studenti.unina.it>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+/*
+ * I2C driver for National Semiconductor LP3944 Funlight Chip
+ * http://www.national.com/pf/LP/LP3944.html
+ *
+ * This helper chip can drive up to 8 leds, with two programmable DIM modes;
+ * it could even be used as a gpio expander but this driver assumes it is used
+ * as a led controller.
+ *
+ * The DIM modes are used to set _blink_ patterns for leds, the pattern is
+ * specified supplying two parameters:
+ *   - period: from 0s to 1.6s
+ *   - duty cycle: percentage of the period the led is on, from 0 to 100
+ *
+ * LP3944 can be found on Motorola A910 smartphone, where it drives the rgb
+ * leds, the camera flash light and the displays backlights.
+ */
+
+#include <linux/module.h>
+#include <linux/i2c.h>
+#include <linux/leds.h>
+#include <linux/mutex.h>
+#include <linux/workqueue.h>
+#include <linux/leds-lp3944.h>
+
+/* Read Only Registers */
+#define LP3944_REG_INPUT1     0x00 /* LEDs 0-7 InputRegister (Read Only) */
+#define LP3944_REG_REGISTER1  0x01 /* None (Read Only) */
+
+#define LP3944_REG_PSC0       0x02 /* Frequency Prescaler 0 (R/W) */
+#define LP3944_REG_PWM0       0x03 /* PWM Register 0 (R/W) */
+#define LP3944_REG_PSC1       0x04 /* Frequency Prescaler 1 (R/W) */
+#define LP3944_REG_PWM1       0x05 /* PWM Register 1 (R/W) */
+#define LP3944_REG_LS0        0x06 /* LEDs 0-3 Selector (R/W) */
+#define LP3944_REG_LS1        0x07 /* LEDs 4-7 Selector (R/W) */
+
+/* These registers are not used to control leds in LP3944, they can store
+ * arbitrary values which the chip will ignore.
+ */
+#define LP3944_REG_REGISTER8  0x08
+#define LP3944_REG_REGISTER9  0x09
+
+#define LP3944_DIM0 0
+#define LP3944_DIM1 1
+
+/* period in ms */
+#define LP3944_PERIOD_MIN 0
+#define LP3944_PERIOD_MAX 1600
+
+/* duty cycle is a percentage */
+#define LP3944_DUTY_CYCLE_MIN 0
+#define LP3944_DUTY_CYCLE_MAX 100
+
+#define ldev_to_led(c)       container_of(c, struct lp3944_led_data, ldev)
+
+/* Saved data */
+struct lp3944_led_data {
+	u8 id;
+	enum lp3944_type type;
+	enum lp3944_status status;
+	struct led_classdev ldev;
+	struct i2c_client *client;
+	struct work_struct work;
+};
+
+struct lp3944_data {
+	struct mutex lock;
+	struct i2c_client *client;
+	struct lp3944_led_data leds[LP3944_LEDS_MAX];
+};
+
+static int lp3944_reg_read(struct i2c_client *client, u8 reg, u8 *value)
+{
+	int tmp;
+
+	tmp = i2c_smbus_read_byte_data(client, reg);
+	if (tmp < 0)
+		return -EINVAL;
+
+	*value = tmp;
+
+	return 0;
+}
+
+static int lp3944_reg_write(struct i2c_client *client, u8 reg, u8 value)
+{
+	return i2c_smbus_write_byte_data(client, reg, value);
+}
+
+/**
+ * Set the period for DIM status
+ *
+ * @client: the i2c client
+ * @dim: either LP3944_DIM0 or LP3944_DIM1
+ * @period: period of a blink, that is a on/off cycle, expressed in ms.
+ */
+static int lp3944_dim_set_period(struct i2c_client *client, u8 dim, u16 period)
+{
+	u8 psc_reg;
+	u8 psc_value;
+	int err;
+
+	if (dim == LP3944_DIM0)
+		psc_reg = LP3944_REG_PSC0;
+	else if (dim == LP3944_DIM1)
+		psc_reg = LP3944_REG_PSC1;
+	else
+		return -EINVAL;
+
+	/* Convert period to Prescaler value */
+	if (period > LP3944_PERIOD_MAX)
+		return -EINVAL;
+
+	psc_value = (period * 255) / LP3944_PERIOD_MAX;
+
+	err = lp3944_reg_write(client, psc_reg, psc_value);
+
+	return err;
+}
+
+/**
+ * Set the duty cycle for DIM status
+ *
+ * @client: the i2c client
+ * @dim: either LP3944_DIM0 or LP3944_DIM1
+ * @duty_cycle: percentage of a period during which a led is ON
+ */
+static int lp3944_dim_set_dutycycle(struct i2c_client *client, u8 dim,
+				    u8 duty_cycle)
+{
+	u8 pwm_reg;
+	u8 pwm_value;
+	int err;
+
+	if (dim == LP3944_DIM0)
+		pwm_reg = LP3944_REG_PWM0;
+	else if (dim == LP3944_DIM1)
+		pwm_reg = LP3944_REG_PWM1;
+	else
+		return -EINVAL;
+
+	/* Convert duty cycle to PWM value */
+	if (duty_cycle > LP3944_DUTY_CYCLE_MAX)
+		return -EINVAL;
+
+	pwm_value = (duty_cycle * 255) / LP3944_DUTY_CYCLE_MAX;
+
+	err = lp3944_reg_write(client, pwm_reg, pwm_value);
+
+	return err;
+}
+
+/**
+ * Set the led status
+ *
+ * @led: a lp3944_led_data structure
+ * @status: one of LP3944_LED_STATUS_OFF
+ *                 LP3944_LED_STATUS_ON
+ *                 LP3944_LED_STATUS_DIM0
+ *                 LP3944_LED_STATUS_DIM1
+ */
+static int lp3944_led_set(struct lp3944_led_data *led, u8 status)
+{
+	struct lp3944_data *data = i2c_get_clientdata(led->client);
+	u8 id = led->id;
+	u8 reg;
+	u8 val = 0;
+	int err;
+
+	dev_dbg(&led->client->dev, "%s: %s, status before normalization:%d\n",
+		__func__, led->ldev.name, status);
+
+	switch (id) {
+	case LP3944_LED0:
+	case LP3944_LED1:
+	case LP3944_LED2:
+	case LP3944_LED3:
+		reg = LP3944_REG_LS0;
+		break;
+	case LP3944_LED4:
+	case LP3944_LED5:
+	case LP3944_LED6:
+	case LP3944_LED7:
+		id -= LP3944_LED4;
+		reg = LP3944_REG_LS1;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (status > LP3944_LED_STATUS_DIM1)
+		return -EINVAL;
+
+	/* invert only 0 and 1, leave unchanged the other values,
+	 * remember we are abusing status to set blink patterns
+	 */
+	if (led->type == LP3944_LED_TYPE_LED_INVERTED && status < 2)
+		status = 1 - status;
+
+	mutex_lock(&data->lock);
+	lp3944_reg_read(led->client, reg, &val);
+
+	val &= ~(LP3944_LED_STATUS_MASK << (id << 1));
+	val |= (status << (id << 1));
+
+	dev_dbg(&led->client->dev, "%s: %s, reg:%d id:%d status:%d val:%#x\n",
+		__func__, led->ldev.name, reg, id, status, val);
+
+	/* set led status */
+	err = lp3944_reg_write(led->client, reg, val);
+	mutex_unlock(&data->lock);
+
+	return err;
+}
+
+static int lp3944_led_set_blink(struct led_classdev *led_cdev,
+				unsigned long *delay_on,
+				unsigned long *delay_off)
+{
+	struct lp3944_led_data *led = ldev_to_led(led_cdev);
+	u16 period;
+	u8 duty_cycle;
+	int err;
+
+	/* units are in ms */
+	if (*delay_on + *delay_off > LP3944_PERIOD_MAX)
+		return -EINVAL;
+
+	if (*delay_on == 0 && *delay_off == 0) {
+		/* Special case: the leds subsystem requires a default user
+		 * friendly blink pattern for the LED.  Let's blink the led
+		 * slowly (1Hz).
+		 */
+		*delay_on = 500;
+		*delay_off = 500;
+	}
+
+	period = (*delay_on) + (*delay_off);
+
+	/* duty_cycle is the percentage of period during which the led is ON */
+	duty_cycle = 100 * (*delay_on) / period;
+
+	/* invert duty cycle for inverted leds, this has the same effect of
+	 * swapping delay_on and delay_off
+	 */
+	if (led->type == LP3944_LED_TYPE_LED_INVERTED)
+		duty_cycle = 100 - duty_cycle;
+
+	/* NOTE: using always the first DIM mode, this means that all leds
+	 * will have the same blinking pattern.
+	 *
+	 * We could find a way later to have two leds blinking in hardware
+	 * with different patterns at the same time, falling back to software
+	 * control for the other ones.
+	 */
+	err = lp3944_dim_set_period(led->client, LP3944_DIM0, period);
+	if (err)
+		return err;
+
+	err = lp3944_dim_set_dutycycle(led->client, LP3944_DIM0, duty_cycle);
+	if (err)
+		return err;
+
+	dev_dbg(&led->client->dev, "%s: OK hardware accelerated blink!\n",
+		__func__);
+
+	led->status = LP3944_LED_STATUS_DIM0;
+	schedule_work(&led->work);
+
+	return 0;
+}
+
+static void lp3944_led_set_brightness(struct led_classdev *led_cdev,
+				      enum led_brightness brightness)
+{
+	struct lp3944_led_data *led = ldev_to_led(led_cdev);
+
+	dev_dbg(&led->client->dev, "%s: %s, %d\n",
+		__func__, led_cdev->name, brightness);
+
+	led->status = brightness;
+	schedule_work(&led->work);
+}
+
+static void lp3944_led_work(struct work_struct *work)
+{
+	struct lp3944_led_data *led;
+
+	led = container_of(work, struct lp3944_led_data, work);
+	lp3944_led_set(led, led->status);
+}
+
+static int lp3944_configure(struct i2c_client *client,
+			    struct lp3944_data *data,
+			    struct lp3944_platform_data *pdata)
+{
+	int i, err = 0;
+
+	for (i = 0; i < pdata->leds_size; i++) {
+		struct lp3944_led *pled = &pdata->leds[i];
+		struct lp3944_led_data *led = &data->leds[i];
+		led->client = client;
+		led->id = i;
+
+		switch (pled->type) {
+
+		case LP3944_LED_TYPE_LED:
+		case LP3944_LED_TYPE_LED_INVERTED:
+			led->type = pled->type;
+			led->status = pled->status;
+			led->ldev.name = pled->name;
+			led->ldev.max_brightness = 1;
+			led->ldev.brightness_set = lp3944_led_set_brightness;
+			led->ldev.blink_set = lp3944_led_set_blink;
+			led->ldev.flags = LED_CORE_SUSPENDRESUME;
+
+			INIT_WORK(&led->work, lp3944_led_work);
+			err = led_classdev_register(&client->dev, &led->ldev);
+			if (err < 0) {
+				dev_err(&client->dev,
+					"couldn't register LED %s\n",
+					led->ldev.name);
+				goto exit;
+			}
+
+			/* to expose the default value to userspace */
+			led->ldev.brightness = led->status;
+
+			/* Set the default led status */
+			err = lp3944_led_set(led, led->status);
+			if (err < 0) {
+				dev_err(&client->dev,
+					"%s couldn't set STATUS %d\n",
+					led->ldev.name, led->status);
+				goto exit;
+			}
+			break;
+
+		case LP3944_LED_TYPE_NONE:
+		default:
+			break;
+
+		}
+	}
+	return 0;
+
+exit:
+	if (i > 0)
+		for (i = i - 1; i >= 0; i--)
+			switch (pdata->leds[i].type) {
+
+			case LP3944_LED_TYPE_LED:
+			case LP3944_LED_TYPE_LED_INVERTED:
+				led_classdev_unregister(&data->leds[i].ldev);
+				cancel_work_sync(&data->leds[i].work);
+				break;
+
+			case LP3944_LED_TYPE_NONE:
+			default:
+				break;
+			}
+
+	return err;
+}
+
+static int __devinit lp3944_probe(struct i2c_client *client,
+				  const struct i2c_device_id *id)
+{
+	struct lp3944_platform_data *lp3944_pdata = client->dev.platform_data;
+	struct lp3944_data *data;
+
+	if (lp3944_pdata == NULL) {
+		dev_err(&client->dev, "no platform data\n");
+		return -EINVAL;
+	}
+
+	/* Let's see whether this adapter can support what we need. */
+	if (!i2c_check_functionality(client->adapter,
+				I2C_FUNC_SMBUS_BYTE_DATA)) {
+		dev_err(&client->dev, "insufficient functionality!\n");
+		return -ENODEV;
+	}
+
+	data = kzalloc(sizeof(struct lp3944_data), GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	data->client = client;
+	i2c_set_clientdata(client, data);
+
+	mutex_init(&data->lock);
+
+	dev_info(&client->dev, "lp3944 enabled\n");
+
+	lp3944_configure(client, data, lp3944_pdata);
+	return 0;
+}
+
+static int __devexit lp3944_remove(struct i2c_client *client)
+{
+	struct lp3944_platform_data *pdata = client->dev.platform_data;
+	struct lp3944_data *data = i2c_get_clientdata(client);
+	int i;
+
+	for (i = 0; i < pdata->leds_size; i++)
+		switch (data->leds[i].type) {
+		case LP3944_LED_TYPE_LED:
+		case LP3944_LED_TYPE_LED_INVERTED:
+			led_classdev_unregister(&data->leds[i].ldev);
+			cancel_work_sync(&data->leds[i].work);
+			break;
+
+		case LP3944_LED_TYPE_NONE:
+		default:
+			break;
+		}
+
+	kfree(data);
+	i2c_set_clientdata(client, NULL);
+
+	return 0;
+}
+
+/* lp3944 i2c driver struct */
+static const struct i2c_device_id lp3944_id[] = {
+	{"lp3944", 0},
+	{}
+};
+
+MODULE_DEVICE_TABLE(i2c, lp3944_id);
+
+static struct i2c_driver lp3944_driver = {
+	.driver   = {
+		   .name = "lp3944",
+	},
+	.probe    = lp3944_probe,
+	.remove   = __devexit_p(lp3944_remove),
+	.id_table = lp3944_id,
+};
+
+static int __init lp3944_module_init(void)
+{
+	return i2c_add_driver(&lp3944_driver);
+}
+
+static void __exit lp3944_module_exit(void)
+{
+	i2c_del_driver(&lp3944_driver);
+}
+
+module_init(lp3944_module_init);
+module_exit(lp3944_module_exit);
+
+MODULE_AUTHOR("Antonio Ospite <ospite@studenti.unina.it>");
+MODULE_DESCRIPTION("LP3944 Fun Light Chip");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/leds-lp3944.h b/include/linux/leds-lp3944.h
new file mode 100644
index 000000000000..afc9f9fd70f5
--- /dev/null
+++ b/include/linux/leds-lp3944.h
@@ -0,0 +1,53 @@
+/*
+ * leds-lp3944.h - platform data structure for lp3944 led controller
+ *
+ * Copyright (C) 2009 Antonio Ospite <ospite@studenti.unina.it>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#ifndef __LINUX_LEDS_LP3944_H
+#define __LINUX_LEDS_LP3944_H
+
+#include <linux/leds.h>
+#include <linux/workqueue.h>
+
+#define LP3944_LED0 0
+#define LP3944_LED1 1
+#define LP3944_LED2 2
+#define LP3944_LED3 3
+#define LP3944_LED4 4
+#define LP3944_LED5 5
+#define LP3944_LED6 6
+#define LP3944_LED7 7
+#define LP3944_LEDS_MAX 8
+
+#define LP3944_LED_STATUS_MASK	0x03
+enum lp3944_status {
+	LP3944_LED_STATUS_OFF  = 0x0,
+	LP3944_LED_STATUS_ON   = 0x1,
+	LP3944_LED_STATUS_DIM0 = 0x2,
+	LP3944_LED_STATUS_DIM1 = 0x3
+};
+
+enum lp3944_type {
+	LP3944_LED_TYPE_NONE,
+	LP3944_LED_TYPE_LED,
+	LP3944_LED_TYPE_LED_INVERTED,
+};
+
+struct lp3944_led {
+	char *name;
+	enum lp3944_type type;
+	enum lp3944_status status;
+};
+
+struct lp3944_platform_data {
+	struct lp3944_led leds[LP3944_LEDS_MAX];
+	u8 leds_size;
+};
+
+#endif /* __LINUX_LEDS_LP3944_H */
-- 
cgit v1.2.3-59-g8ed1b


From ed88bae6918fa990cbfe47316bd0f790121aaf00 Mon Sep 17 00:00:00 2001
From: Trent Piepho <xyzzy@speakeasy.org>
Date: Tue, 12 May 2009 15:33:12 -0700
Subject: leds: Add options to have GPIO LEDs start on or keep their state

There already is a "default-on" trigger but there are problems with it.

For one, it's a inefficient way to do it and requires led trigger support
to be compiled in.

But the real reason is that is produces a glitch on the LED.  The GPIO is
allocate with the LED *off*, then *later* when the trigger runs it is
turned back on.  If the LED was already on via the GPIO's reset default or
action of the firmware, this produces a glitch where the LED goes from on
to off to on.  While normally this is fast enough that it wouldn't be
noticeable to a human observer, there are still serious problems.

One is that there may be something else on the GPIO line, like a hardware
alarm or watchdog, that is fast enough to notice the glitch.

Another is that the kernel may panic before the LED is turned back on, thus
hanging with the LED in the wrong state.  This is not just speculation, but
actually happened to me with an embedded system that has an LED which
should turn off when the kernel finishes booting, which was left in the
incorrect state due to a bug in the OF LED binding code.

We also let GPIO LEDs get their initial value from whatever the current
state of the GPIO line is.  On some systems the LEDs are put into some
state by the firmware or hardware before Linux boots, and it is desired to
have them keep this state which is otherwise unknown to Linux.

This requires that the underlying GPIO driver support reading the value of
output GPIOs.  Some drivers support this and some do not.

The platform device binding gains a field in the platform data
"default_state" that controls this.  There are three constants defined to
select from on, off, or keeping the current state.  The OpenFirmware
binding uses a property named "default-state" that can be set to "on",
"off", or "keep".  The default if the property isn't present is off.

Signed-off-by: Trent Piepho <xyzzy@speakeasy.org>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Acked-by: Wolfram Sang <w.sang@pengutronix.de>
Acked-by: Sean MacLennan <smaclennan@pikatech.com>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 Documentation/powerpc/dts-bindings/gpio/led.txt | 17 ++++++++++++++++-
 drivers/leds/leds-gpio.c                        | 20 +++++++++++++++++---
 include/linux/leds.h                            |  9 +++++++--
 3 files changed, 40 insertions(+), 6 deletions(-)

diff --git a/Documentation/powerpc/dts-bindings/gpio/led.txt b/Documentation/powerpc/dts-bindings/gpio/led.txt
index 4fe14deedc0a..064db928c3c1 100644
--- a/Documentation/powerpc/dts-bindings/gpio/led.txt
+++ b/Documentation/powerpc/dts-bindings/gpio/led.txt
@@ -16,10 +16,17 @@ LED sub-node properties:
   string defining the trigger assigned to the LED.  Current triggers are:
     "backlight" - LED will act as a back-light, controlled by the framebuffer
 		  system
-    "default-on" - LED will turn on
+    "default-on" - LED will turn on, but see "default-state" below
     "heartbeat" - LED "double" flashes at a load average based rate
     "ide-disk" - LED indicates disk activity
     "timer" - LED flashes at a fixed, configurable rate
+- default-state:  (optional) The initial state of the LED.  Valid
+  values are "on", "off", and "keep".  If the LED is already on or off
+  and the default-state property is set the to same value, then no
+  glitch should be produced where the LED momentarily turns off (or
+  on).  The "keep" setting will keep the LED at whatever its current
+  state is, without producing a glitch.  The default is off if this
+  property is not present.
 
 Examples:
 
@@ -30,14 +37,22 @@ leds {
 		gpios = <&mcu_pio 0 1>; /* Active low */
 		linux,default-trigger = "ide-disk";
 	};
+
+	fault {
+		gpios = <&mcu_pio 1 0>;
+		/* Keep LED on if BIOS detected hardware fault */
+		default-state = "keep";
+	};
 };
 
 run-control {
 	compatible = "gpio-leds";
 	red {
 		gpios = <&mpc8572 6 0>;
+		default-state = "off";
 	};
 	green {
 		gpios = <&mpc8572 7 0>;
+		default-state = "on";
 	};
 }
diff --git a/drivers/leds/leds-gpio.c b/drivers/leds/leds-gpio.c
index 76895e691042..6b06638eb5b4 100644
--- a/drivers/leds/leds-gpio.c
+++ b/drivers/leds/leds-gpio.c
@@ -76,7 +76,7 @@ static int __devinit create_gpio_led(const struct gpio_led *template,
 	struct gpio_led_data *led_dat, struct device *parent,
 	int (*blink_set)(unsigned, unsigned long *, unsigned long *))
 {
-	int ret;
+	int ret, state;
 
 	/* skip leds that aren't available */
 	if (!gpio_is_valid(template->gpio)) {
@@ -99,11 +99,15 @@ static int __devinit create_gpio_led(const struct gpio_led *template,
 		led_dat->cdev.blink_set = gpio_blink_set;
 	}
 	led_dat->cdev.brightness_set = gpio_led_set;
-	led_dat->cdev.brightness = LED_OFF;
+	if (template->default_state == LEDS_GPIO_DEFSTATE_KEEP)
+		state = !!gpio_get_value(led_dat->gpio) ^ led_dat->active_low;
+	else
+		state = (template->default_state == LEDS_GPIO_DEFSTATE_ON);
+	led_dat->cdev.brightness = state ? LED_FULL : LED_OFF;
 	if (!template->retain_state_suspended)
 		led_dat->cdev.flags |= LED_CORE_SUSPENDRESUME;
 
-	ret = gpio_direction_output(led_dat->gpio, led_dat->active_low);
+	ret = gpio_direction_output(led_dat->gpio, led_dat->active_low ^ state);
 	if (ret < 0)
 		goto err;
 
@@ -223,12 +227,22 @@ static int __devinit of_gpio_leds_probe(struct of_device *ofdev,
 	memset(&led, 0, sizeof(led));
 	for_each_child_of_node(np, child) {
 		enum of_gpio_flags flags;
+		const char *state;
 
 		led.gpio = of_get_gpio_flags(child, 0, &flags);
 		led.active_low = flags & OF_GPIO_ACTIVE_LOW;
 		led.name = of_get_property(child, "label", NULL) ? : child->name;
 		led.default_trigger =
 			of_get_property(child, "linux,default-trigger", NULL);
+		state = of_get_property(child, "default-state", NULL);
+		if (state) {
+			if (!strcmp(state, "keep"))
+				led.default_state = LEDS_GPIO_DEFSTATE_KEEP;
+			else if(!strcmp(state, "on"))
+				led.default_state = LEDS_GPIO_DEFSTATE_ON;
+			else
+				led.default_state = LEDS_GPIO_DEFSTATE_OFF;
+		}
 
 		ret = create_gpio_led(&led, &pdata->led_data[pdata->num_leds++],
 				      &ofdev->dev, NULL);
diff --git a/include/linux/leds.h b/include/linux/leds.h
index c7f0b148df06..62af62915cf7 100644
--- a/include/linux/leds.h
+++ b/include/linux/leds.h
@@ -143,9 +143,14 @@ struct gpio_led {
 	const char *name;
 	const char *default_trigger;
 	unsigned 	gpio;
-	u8 		active_low : 1;
-	u8		retain_state_suspended : 1;
+	unsigned	active_low : 1;
+	unsigned	retain_state_suspended : 1;
+	unsigned	default_state : 2;
+	/* default_state should be one of LEDS_GPIO_DEFSTATE_(ON|OFF|KEEP) */
 };
+#define LEDS_GPIO_DEFSTATE_OFF	0
+#define LEDS_GPIO_DEFSTATE_ON	1
+#define LEDS_GPIO_DEFSTATE_KEEP	2
 
 struct gpio_led_platform_data {
 	int 		num_leds;
-- 
cgit v1.2.3-59-g8ed1b


From a1dd8c617217322614f0465ae347895c4b58e1ab Mon Sep 17 00:00:00 2001
From: Richard Purdie <rpurdie@linux.intel.com>
Date: Mon, 22 Jun 2009 14:54:13 +0100
Subject: leds: Futher document blink_set

Futher document blink_set function pointer

Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 include/linux/leds.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/linux/leds.h b/include/linux/leds.h
index 62af62915cf7..d8bf9665e70c 100644
--- a/include/linux/leds.h
+++ b/include/linux/leds.h
@@ -47,7 +47,8 @@ struct led_classdev {
 
 	/* Activate hardware accelerated blink, delays are in
 	 * miliseconds and if none is provided then a sensible default
-	 * should be chosen. */
+	 * should be chosen. The call can adjust the timings if it can't
+	 * match the values specified exactly. */
 	int		(*blink_set)(struct led_classdev *led_cdev,
 				     unsigned long *delay_on,
 				     unsigned long *delay_off);
-- 
cgit v1.2.3-59-g8ed1b


From 1d469c6c38c9deaa1836d2c1955330944719e4ef Mon Sep 17 00:00:00 2001
From: Aviv Laufer <aviv.laufer@gmail.com>
Date: Tue, 23 Jun 2009 16:28:36 +0300
Subject: backlight: Fix tdo24m crash on kmalloc

There is  a crash in tdo24m module caused by a call to kmalloc with
the second parameter sizeof(flag) instead of flag.

Signed-off-by: Aviv Laufer <aviv.laufer@gmail.com>
Signed-off-by: Richard Purdie <rpurdie@linux.intel.com>
---
 drivers/video/backlight/tdo24m.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/video/backlight/tdo24m.c b/drivers/video/backlight/tdo24m.c
index 1dae7f8f3c6b..51422fc4f606 100644
--- a/drivers/video/backlight/tdo24m.c
+++ b/drivers/video/backlight/tdo24m.c
@@ -356,7 +356,7 @@ static int __devinit tdo24m_probe(struct spi_device *spi)
 	lcd->power = FB_BLANK_POWERDOWN;
 	lcd->mode = MODE_VGA;	/* default to VGA */
 
-	lcd->buf = kmalloc(TDO24M_SPI_BUFF_SIZE, sizeof(GFP_KERNEL));
+	lcd->buf = kmalloc(TDO24M_SPI_BUFF_SIZE, GFP_KERNEL);
 	if (lcd->buf == NULL) {
 		kfree(lcd);
 		return -ENOMEM;
-- 
cgit v1.2.3-59-g8ed1b


From a5c9b696ec109bb54d547fdb437a7a0c2d514670 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Date: Tue, 23 Jun 2009 12:36:58 -0700
Subject: mm: pass mm to grab_swap_token

If a kthread happens to use get_user_pages() on an mm (as KSM does),
there's a chance that it will end up trying to read in a swap page, then
oops in grab_swap_token() because the kthread has no mm: GUP passes down
the right mm, so grab_swap_token() ought to be using it.

We have not identified a stronger case than KSM's daemon (not yet in
mainline), but the issue must have come up before, since RHEL has included
a fix for this for years (though a different fix, they just back out of
grab_swap_token if current->mm is unset: which is what we first proposed,
but using the right mm here seems more correct).

Reported-by: Izik Eidus <ieidus@redhat.com>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h | 12 ++++++------
 mm/memory.c          |  2 +-
 mm/thrash.c          | 32 +++++++++++++++-----------------
 3 files changed, 22 insertions(+), 24 deletions(-)

diff --git a/include/linux/swap.h b/include/linux/swap.h
index c88b36665f79..7c15334f3ff2 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -298,8 +298,8 @@ extern int try_to_free_swap(struct page *);
 struct backing_dev_info;
 
 /* linux/mm/thrash.c */
-extern struct mm_struct * swap_token_mm;
-extern void grab_swap_token(void);
+extern struct mm_struct *swap_token_mm;
+extern void grab_swap_token(struct mm_struct *);
 extern void __put_swap_token(struct mm_struct *);
 
 static inline int has_swap_token(struct mm_struct *mm)
@@ -419,10 +419,10 @@ static inline swp_entry_t get_swap_page(void)
 }
 
 /* linux/mm/thrash.c */
-#define put_swap_token(x) do { } while(0)
-#define grab_swap_token()  do { } while(0)
-#define has_swap_token(x) 0
-#define disable_swap_token() do { } while(0)
+#define put_swap_token(mm)	do { } while (0)
+#define grab_swap_token(mm)	do { } while (0)
+#define has_swap_token(mm)	0
+#define disable_swap_token()	do { } while (0)
 
 static inline void
 mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent)
diff --git a/mm/memory.c b/mm/memory.c
index 50da9511aa77..f46ac18ba231 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2519,7 +2519,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	delayacct_set_flag(DELAYACCT_PF_SWAPIN);
 	page = lookup_swap_cache(entry);
 	if (!page) {
-		grab_swap_token(); /* Contend for token _before_ read-in */
+		grab_swap_token(mm); /* Contend for token _before_ read-in */
 		page = swapin_readahead(entry,
 					GFP_HIGHUSER_MOVABLE, vma, address);
 		if (!page) {
diff --git a/mm/thrash.c b/mm/thrash.c
index c4c5205a9c35..2372d4ed5dd8 100644
--- a/mm/thrash.c
+++ b/mm/thrash.c
@@ -26,47 +26,45 @@ static DEFINE_SPINLOCK(swap_token_lock);
 struct mm_struct *swap_token_mm;
 static unsigned int global_faults;
 
-void grab_swap_token(void)
+void grab_swap_token(struct mm_struct *mm)
 {
 	int current_interval;
 
 	global_faults++;
 
-	current_interval = global_faults - current->mm->faultstamp;
+	current_interval = global_faults - mm->faultstamp;
 
 	if (!spin_trylock(&swap_token_lock))
 		return;
 
 	/* First come first served */
 	if (swap_token_mm == NULL) {
-		current->mm->token_priority = current->mm->token_priority + 2;
-		swap_token_mm = current->mm;
+		mm->token_priority = mm->token_priority + 2;
+		swap_token_mm = mm;
 		goto out;
 	}
 
-	if (current->mm != swap_token_mm) {
-		if (current_interval < current->mm->last_interval)
-			current->mm->token_priority++;
+	if (mm != swap_token_mm) {
+		if (current_interval < mm->last_interval)
+			mm->token_priority++;
 		else {
-			if (likely(current->mm->token_priority > 0))
-				current->mm->token_priority--;
+			if (likely(mm->token_priority > 0))
+				mm->token_priority--;
 		}
 		/* Check if we deserve the token */
-		if (current->mm->token_priority >
-				swap_token_mm->token_priority) {
-			current->mm->token_priority += 2;
-			swap_token_mm = current->mm;
+		if (mm->token_priority > swap_token_mm->token_priority) {
+			mm->token_priority += 2;
+			swap_token_mm = mm;
 		}
 	} else {
 		/* Token holder came in again! */
-		current->mm->token_priority += 2;
+		mm->token_priority += 2;
 	}
 
 out:
-	current->mm->faultstamp = global_faults;
-	current->mm->last_interval = current_interval;
+	mm->faultstamp = global_faults;
+	mm->last_interval = current_interval;
 	spin_unlock(&swap_token_lock);
-return;
 }
 
 /* Called on process exit. */
-- 
cgit v1.2.3-59-g8ed1b


From 01ff53f416757da416413bc32229770a8448b6ef Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Tue, 23 Jun 2009 12:37:01 -0700
Subject: rmap: fixup page_referenced() for nommu systems

After the recent changes that went into mm/vmscan.c to overhaul stuff, we
ended up with these warnings on no-mmu systems:

  mm/vmscan.c: In function `shrink_page_list':
  mm/vmscan.c:580: warning: unused variable `vm_flags'
  mm/vmscan.c: In function `shrink_active_list':
  mm/vmscan.c:1294: warning: `vm_flags' may be used uninitialized in this function
  mm/vmscan.c:1242: note: `vm_flags' was declared here

This is because the no-mmu function defines page_referenced() to work on
the first argument only (the page).  It does not clear the vm_flags given
to it because for no-mmu systems, they never actually get utilized.  Since
that is no longer strictly true, we need to set vm_flags to 0 like
everyone else so gcc can do proper dead code elimination without annoying
us with unused warnings.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Cc: David Howells <dhowells@redhat.com>
Acked-by: David McCullough <davidm@snapgear.com>
Cc: Greg Ungerer <gerg@uclinux.org>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/rmap.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index 216d024f830d..bf116d0dbf23 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -118,7 +118,14 @@ int try_to_munlock(struct page *);
 #define anon_vma_prepare(vma)	(0)
 #define anon_vma_link(vma)	do {} while (0)
 
-#define page_referenced(page, locked, cnt, flags) TestClearPageReferenced(page)
+static inline int page_referenced(struct page *page, int is_locked,
+				  struct mem_cgroup *cnt,
+				  unsigned long *vm_flags)
+{
+	*vm_flags = 0;
+	return TestClearPageReferenced(page);
+}
+
 #define try_to_unmap(page, refs) SWAP_FAIL
 
 static inline int page_mkclean(struct page *page)
-- 
cgit v1.2.3-59-g8ed1b


From 364df0ebfbbb1330bfc6ca159f4d6020efc15a12 Mon Sep 17 00:00:00 2001
From: Dimitri Sivanich <sivanich@sgi.com>
Date: Tue, 23 Jun 2009 12:37:04 -0700
Subject: mm: fix handling of pagesets for downed cpus

After downing/upping a cpu, an attempt to set
/proc/sys/vm/percpu_pagelist_fraction results in an oops in
percpu_pagelist_fraction_sysctl_handler().

If a processor is downed then we need to set the pageset pointer back to
the boot pageset.

Updates of the high water marks should not access pagesets of unpopulated
zones (those pointer go to the boot pagesets which would be no longer
functional if their size would be increased beyond zero).

Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>
Signed-off-by: Christoph Lameter <cl@linux-foundation.org>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/page_alloc.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 30d5093a099d..aecc9cdfdfce 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3026,7 +3026,7 @@ bad:
 		if (dzone == zone)
 			break;
 		kfree(zone_pcp(dzone, cpu));
-		zone_pcp(dzone, cpu) = NULL;
+		zone_pcp(dzone, cpu) = &boot_pageset[cpu];
 	}
 	return -ENOMEM;
 }
@@ -3041,7 +3041,7 @@ static inline void free_zone_pagesets(int cpu)
 		/* Free per_cpu_pageset if it is slab allocated */
 		if (pset != &boot_pageset[cpu])
 			kfree(pset);
-		zone_pcp(zone, cpu) = NULL;
+		zone_pcp(zone, cpu) = &boot_pageset[cpu];
 	}
 }
 
@@ -4659,7 +4659,7 @@ int percpu_pagelist_fraction_sysctl_handler(ctl_table *table, int write,
 	ret = proc_dointvec_minmax(table, write, file, buffer, length, ppos);
 	if (!write || (ret == -EINVAL))
 		return ret;
-	for_each_zone(zone) {
+	for_each_populated_zone(zone) {
 		for_each_online_cpu(cpu) {
 			unsigned long  high;
 			high = zone->present_pages / percpu_pagelist_fraction;
-- 
cgit v1.2.3-59-g8ed1b


From 840c516f9c395c64240ad35f858f37ea16270afa Mon Sep 17 00:00:00 2001
From: "Robert P. J. Day" <rpjday@crashcourse.ca>
Date: Tue, 23 Jun 2009 12:37:06 -0700
Subject: h8/300: fix incorrect "select" directives in arch/h8300/Kconfig.cpu.

Fix the incorrect "select" directives by dropping the "CONFIG_"
prefixes, and correcting the typo "H8S2768" to "H8S2678".

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Acked-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/h8300/Kconfig.cpu | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/arch/h8300/Kconfig.cpu b/arch/h8300/Kconfig.cpu
index b65dcfe51d9c..6e2ecff199c5 100644
--- a/arch/h8300/Kconfig.cpu
+++ b/arch/h8300/Kconfig.cpu
@@ -13,7 +13,7 @@ config H8300H_GENERIC
 
 config H8300H_AKI3068NET
 	bool "AE-3068/69"
-	select CONFIG_H83068
+	select H83068
 	help
 	  AKI-H8/3068F / AKI-H8/3069F Flashmicom LAN Board Support
 	  More Information. (Japanese Only)
@@ -24,7 +24,7 @@ config H8300H_AKI3068NET
 
 config H8300H_H8MAX
 	bool "H8MAX"
-	select CONFIG_H83068
+	select H83068
 	help
 	  H8MAX Evaluation Board Support
 	  More Information. (Japanese Only)
@@ -32,7 +32,7 @@ config H8300H_H8MAX
 
 config H8300H_SIM
 	bool "H8/300H Simulator"
-	select CONFIG_H83007
+	select H83007
 	help
 	  GDB Simulator Support
 	  More Information.
@@ -45,7 +45,7 @@ config H8S_GENERIC
 
 config H8S_EDOSK2674
 	bool "EDOSK-2674"
-	select CONFIG_H8S2768
+	select H8S2678
 	help
 	  Renesas EDOSK-2674 Evaluation Board Support
 	  More Information.
-- 
cgit v1.2.3-59-g8ed1b


From 4e8a2372f9255a1464ef488ed925455f53fbdaa1 Mon Sep 17 00:00:00 2001
From: Mikael Pettersson <mikpe@it.uu.se>
Date: Tue, 23 Jun 2009 12:37:07 -0700
Subject: nvidiafb: fix boot-time printk string

On bootup nvidiafb prints the following on my Apple G5:

	nvidiafb: CRTC 1appears to have a CRT attached

There should be a space between the '1' and the 'appears'.  Add it.

Signed-off-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/nvidia/nv_setup.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/video/nvidia/nv_setup.c b/drivers/video/nvidia/nv_setup.c
index 135ae18bfce8..eef2bb298d9f 100644
--- a/drivers/video/nvidia/nv_setup.c
+++ b/drivers/video/nvidia/nv_setup.c
@@ -543,8 +543,7 @@ int NVCommonSetup(struct fb_info *info)
 		} else if (analog_on_B) {
 			CRTCnumber = outputBfromCRTC;
 			FlatPanel = 0;
-			printk("nvidiafb: CRTC %i"
-			       "appears to have a "
+			printk("nvidiafb: CRTC %i appears to have a "
 			       "CRT attached\n", CRTCnumber);
 		} else if (slaved_on_A) {
 			CRTCnumber = 0;
-- 
cgit v1.2.3-59-g8ed1b


From cca03c0aeb18a975abec28df518a2b64ae3e6964 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Tue, 23 Jun 2009 17:12:49 +0530
Subject: perf stat: Fix verbose for perf stat

Error message should use stderr for verbose (-v), otherwise
message will be lost for:

 $ ./perf stat -v <cmd>  > /dev/null

For example on AMD bus-cycles event is not available so now
it looks like:

 $ ./perf stat -v -e bus-cycles ls > /dev/null
Error: counter 0, sys_perf_counter_open() syscall returned with -1 (Invalid argument)

 Performance counter stats for 'ls':

  <not counted>  bus-cycles

    0.006765877  seconds time elapsed.

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <1245757369.3776.1.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-stat.c | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 6d3eeac1ea25..5e04fcc8d077 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -109,6 +109,10 @@ static u64			walltime_nsecs_noise;
 static u64			runtime_cycles_avg;
 static u64			runtime_cycles_noise;
 
+
+#define ERR_PERF_OPEN \
+"Error: counter %d, sys_perf_counter_open() syscall returned with %d (%s)\n"
+
 static void create_perf_stat_counter(int counter)
 {
 	struct perf_counter_attr *attr = attrs + counter;
@@ -119,20 +123,20 @@ static void create_perf_stat_counter(int counter)
 
 	if (system_wide) {
 		int cpu;
-		for (cpu = 0; cpu < nr_cpus; cpu ++) {
+		for (cpu = 0; cpu < nr_cpus; cpu++) {
 			fd[cpu][counter] = sys_perf_counter_open(attr, -1, cpu, -1, 0);
-			if (fd[cpu][counter] < 0 && verbose) {
-				printf("Error: counter %d, sys_perf_counter_open() syscall returned with %d (%s)\n", counter, fd[cpu][counter], strerror(errno));
-			}
+			if (fd[cpu][counter] < 0 && verbose)
+				fprintf(stderr, ERR_PERF_OPEN, counter,
+					fd[cpu][counter], strerror(errno));
 		}
 	} else {
 		attr->inherit	= inherit;
 		attr->disabled	= 1;
 
 		fd[0][counter] = sys_perf_counter_open(attr, 0, -1, -1, 0);
-		if (fd[0][counter] < 0 && verbose) {
-			printf("Error: counter %d, sys_perf_counter_open() syscall returned with %d (%s)\n", counter, fd[0][counter], strerror(errno));
-		}
+		if (fd[0][counter] < 0 && verbose)
+			fprintf(stderr, ERR_PERF_OPEN, counter,
+				fd[0][counter], strerror(errno));
 	}
 }
 
@@ -168,7 +172,7 @@ static void read_counter(int counter)
 	count[0] = count[1] = count[2] = 0;
 
 	nv = scale ? 3 : 1;
-	for (cpu = 0; cpu < nr_cpus; cpu ++) {
+	for (cpu = 0; cpu < nr_cpus; cpu++) {
 		if (fd[cpu][counter] < 0)
 			continue;
 
-- 
cgit v1.2.3-59-g8ed1b


From 7262b6e4a4cc18d0f67df145d032c843e4bc382b Mon Sep 17 00:00:00 2001
From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Date: Tue, 23 Jun 2009 12:40:54 +0900
Subject: x86, mce: Fix mce resume on 32bit

Calling mcheck_init() on resume is required only with
CONFIG_X86_OLD_MCE=y.

Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Acked-by: Andi Kleen <andi@firstfloor.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/power/cpu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/power/cpu.c b/arch/x86/power/cpu.c
index d277ef1eea51..b3d20b9cac63 100644
--- a/arch/x86/power/cpu.c
+++ b/arch/x86/power/cpu.c
@@ -244,7 +244,7 @@ static void __restore_processor_state(struct saved_context *ctxt)
 	do_fpu_end();
 	mtrr_ap_init();
 
-#ifdef CONFIG_X86_32
+#ifdef CONFIG_X86_OLD_MCE
 	mcheck_init(&boot_cpu_data);
 #endif
 }
-- 
cgit v1.2.3-59-g8ed1b


From 2c2e2c389d03bb16b8cdf9db3ac615385fac100f Mon Sep 17 00:00:00 2001
From: Fenghua Yu <fenghua.yu@intel.com>
Date: Fri, 19 Jun 2009 13:47:29 -0700
Subject: IOMMU Identity Mapping Support (drivers/pci/intel_iommu.c)

Identity mapping for IOMMU defines a single domain to 1:1 map all PCI
devices to all usable memory.

This reduces map/unmap overhead in DMA API's and improve IOMMU
performance. On 10Gb network cards, Netperf shows no performance
degradation compared to non-IOMMU performance.

This method may lose some of DMA remapping benefits like isolation.

The patch sets up identity mapping for all PCI devices to all usable
memory. In the DMA API, there is no overhead to maintain page tables,
invalidate iotlb, flush cache etc.

32 bit DMA devices don't use identity mapping domain, in order to access
memory beyond 4GiB.

When kernel option iommu=pt, pass through is first tried. If pass
through succeeds, IOMMU goes to pass through. If pass through is not
supported in hw or fail for whatever reason, IOMMU goes to identity
mapping.

Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 314 +++++++++++++++++++++++++++++++++++++---------
 1 file changed, 254 insertions(+), 60 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 178853a07440..e53eacd75c8d 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -39,6 +39,7 @@
 #include <linux/sysdev.h>
 #include <asm/cacheflush.h>
 #include <asm/iommu.h>
+#include <asm/e820.h>
 #include "pci.h"
 
 #define ROOT_SIZE		VTD_PAGE_SIZE
@@ -217,6 +218,14 @@ static inline bool dma_pte_present(struct dma_pte *pte)
 	return (pte->val & 3) != 0;
 }
 
+/*
+ * This domain is a statically identity mapping domain.
+ *	1. This domain creats a static 1:1 mapping to all usable memory.
+ * 	2. It maps to each iommu if successful.
+ *	3. Each iommu mapps to this domain if successful.
+ */
+struct dmar_domain *si_domain;
+
 /* devices under the same p2p bridge are owned in one domain */
 #define DOMAIN_FLAG_P2P_MULTIPLE_DEVICES (1 << 0)
 
@@ -225,6 +234,9 @@ static inline bool dma_pte_present(struct dma_pte *pte)
  */
 #define DOMAIN_FLAG_VIRTUAL_MACHINE	(1 << 1)
 
+/* si_domain contains mulitple devices */
+#define DOMAIN_FLAG_STATIC_IDENTITY	(1 << 2)
+
 struct dmar_domain {
 	int	id;			/* domain id */
 	unsigned long iommu_bmp;	/* bitmap of iommus this domain uses*/
@@ -435,12 +447,14 @@ int iommu_calculate_agaw(struct intel_iommu *iommu)
 	return __iommu_calculate_agaw(iommu, DEFAULT_DOMAIN_ADDRESS_WIDTH);
 }
 
-/* in native case, each domain is related to only one iommu */
+/* This functionin only returns single iommu in a domain */
 static struct intel_iommu *domain_get_iommu(struct dmar_domain *domain)
 {
 	int iommu_id;
 
+	/* si_domain and vm domain should not get here. */
 	BUG_ON(domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE);
+	BUG_ON(domain->flags & DOMAIN_FLAG_STATIC_IDENTITY);
 
 	iommu_id = find_first_bit(&domain->iommu_bmp, g_num_of_iommus);
 	if (iommu_id < 0 || iommu_id >= g_num_of_iommus)
@@ -1189,48 +1203,71 @@ void free_dmar_iommu(struct intel_iommu *iommu)
 	free_context_table(iommu);
 }
 
-static struct dmar_domain * iommu_alloc_domain(struct intel_iommu *iommu)
+static struct dmar_domain *alloc_domain(void)
 {
-	unsigned long num;
-	unsigned long ndomains;
 	struct dmar_domain *domain;
-	unsigned long flags;
 
 	domain = alloc_domain_mem();
 	if (!domain)
 		return NULL;
 
+	memset(&domain->iommu_bmp, 0, sizeof(unsigned long));
+	domain->flags = 0;
+
+	return domain;
+}
+
+static int iommu_attach_domain(struct dmar_domain *domain,
+			       struct intel_iommu *iommu)
+{
+	int num;
+	unsigned long ndomains;
+	unsigned long flags;
+
 	ndomains = cap_ndoms(iommu->cap);
 
 	spin_lock_irqsave(&iommu->lock, flags);
+
 	num = find_first_zero_bit(iommu->domain_ids, ndomains);
 	if (num >= ndomains) {
 		spin_unlock_irqrestore(&iommu->lock, flags);
-		free_domain_mem(domain);
 		printk(KERN_ERR "IOMMU: no free domain ids\n");
-		return NULL;
+		return -ENOMEM;
 	}
 
-	set_bit(num, iommu->domain_ids);
 	domain->id = num;
-	memset(&domain->iommu_bmp, 0, sizeof(unsigned long));
+	set_bit(num, iommu->domain_ids);
 	set_bit(iommu->seq_id, &domain->iommu_bmp);
-	domain->flags = 0;
 	iommu->domains[num] = domain;
 	spin_unlock_irqrestore(&iommu->lock, flags);
 
-	return domain;
+	return 0;
 }
 
-static void iommu_free_domain(struct dmar_domain *domain)
+static void iommu_detach_domain(struct dmar_domain *domain,
+				struct intel_iommu *iommu)
 {
 	unsigned long flags;
-	struct intel_iommu *iommu;
-
-	iommu = domain_get_iommu(domain);
+	int num, ndomains;
+	int found = 0;
 
 	spin_lock_irqsave(&iommu->lock, flags);
-	clear_bit(domain->id, iommu->domain_ids);
+	ndomains = cap_ndoms(iommu->cap);
+	num = find_first_bit(iommu->domain_ids, ndomains);
+	for (; num < ndomains; ) {
+		if (iommu->domains[num] == domain) {
+			found = 1;
+			break;
+		}
+		num = find_next_bit(iommu->domain_ids,
+				    cap_ndoms(iommu->cap), num+1);
+	}
+
+	if (found) {
+		clear_bit(num, iommu->domain_ids);
+		clear_bit(iommu->seq_id, &domain->iommu_bmp);
+		iommu->domains[num] = NULL;
+	}
 	spin_unlock_irqrestore(&iommu->lock, flags);
 }
 
@@ -1350,6 +1387,8 @@ static int domain_init(struct dmar_domain *domain, int guest_width)
 
 static void domain_exit(struct dmar_domain *domain)
 {
+	struct dmar_drhd_unit *drhd;
+	struct intel_iommu *iommu;
 	u64 end;
 
 	/* Domain 0 is reserved, so dont process it */
@@ -1368,7 +1407,10 @@ static void domain_exit(struct dmar_domain *domain)
 	/* free page tables */
 	dma_pte_free_pagetable(domain, 0, end);
 
-	iommu_free_domain(domain);
+	for_each_active_iommu(iommu, drhd)
+		if (test_bit(iommu->seq_id, &domain->iommu_bmp))
+			iommu_detach_domain(domain, iommu);
+
 	free_domain_mem(domain);
 }
 
@@ -1408,7 +1450,8 @@ static int domain_context_mapping_one(struct dmar_domain *domain, int segment,
 	id = domain->id;
 	pgd = domain->pgd;
 
-	if (domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE) {
+	if (domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE ||
+	    domain->flags & DOMAIN_FLAG_STATIC_IDENTITY) {
 		int found = 0;
 
 		/* find an available domain id for this device in iommu */
@@ -1433,6 +1476,7 @@ static int domain_context_mapping_one(struct dmar_domain *domain, int segment,
 			}
 
 			set_bit(num, iommu->domain_ids);
+			set_bit(iommu->seq_id, &domain->iommu_bmp);
 			iommu->domains[num] = domain;
 			id = num;
 		}
@@ -1675,6 +1719,7 @@ static struct dmar_domain *get_domain_for_dev(struct pci_dev *pdev, int gaw)
 	unsigned long flags;
 	int bus = 0, devfn = 0;
 	int segment;
+	int ret;
 
 	domain = find_domain(pdev);
 	if (domain)
@@ -1707,6 +1752,10 @@ static struct dmar_domain *get_domain_for_dev(struct pci_dev *pdev, int gaw)
 		}
 	}
 
+	domain = alloc_domain();
+	if (!domain)
+		goto error;
+
 	/* Allocate new domain for the device */
 	drhd = dmar_find_matched_drhd_unit(pdev);
 	if (!drhd) {
@@ -1716,9 +1765,11 @@ static struct dmar_domain *get_domain_for_dev(struct pci_dev *pdev, int gaw)
 	}
 	iommu = drhd->iommu;
 
-	domain = iommu_alloc_domain(iommu);
-	if (!domain)
+	ret = iommu_attach_domain(domain, iommu);
+	if (ret) {
+		domain_exit(domain);
 		goto error;
+	}
 
 	if (domain_init(domain, gaw)) {
 		domain_exit(domain);
@@ -1792,6 +1843,8 @@ error:
 	return find_domain(pdev);
 }
 
+static int iommu_identity_mapping;
+
 static int iommu_prepare_identity_map(struct pci_dev *pdev,
 				      unsigned long long start,
 				      unsigned long long end)
@@ -1804,8 +1857,11 @@ static int iommu_prepare_identity_map(struct pci_dev *pdev,
 	printk(KERN_INFO
 		"IOMMU: Setting identity map for device %s [0x%Lx - 0x%Lx]\n",
 		pci_name(pdev), start, end);
-	/* page table init */
-	domain = get_domain_for_dev(pdev, DEFAULT_DOMAIN_ADDRESS_WIDTH);
+	if (iommu_identity_mapping)
+		domain = si_domain;
+	else
+		/* page table init */
+		domain = get_domain_for_dev(pdev, DEFAULT_DOMAIN_ADDRESS_WIDTH);
 	if (!domain)
 		return -ENOMEM;
 
@@ -1952,7 +2008,110 @@ static int __init init_context_pass_through(void)
 	return 0;
 }
 
-static int __init init_dmars(void)
+static int md_domain_init(struct dmar_domain *domain, int guest_width);
+static int si_domain_init(void)
+{
+	struct dmar_drhd_unit *drhd;
+	struct intel_iommu *iommu;
+	int ret = 0;
+
+	si_domain = alloc_domain();
+	if (!si_domain)
+		return -EFAULT;
+
+
+	for_each_active_iommu(iommu, drhd) {
+		ret = iommu_attach_domain(si_domain, iommu);
+		if (ret) {
+			domain_exit(si_domain);
+			return -EFAULT;
+		}
+	}
+
+	if (md_domain_init(si_domain, DEFAULT_DOMAIN_ADDRESS_WIDTH)) {
+		domain_exit(si_domain);
+		return -EFAULT;
+	}
+
+	si_domain->flags = DOMAIN_FLAG_STATIC_IDENTITY;
+
+	return 0;
+}
+
+static void domain_remove_one_dev_info(struct dmar_domain *domain,
+					  struct pci_dev *pdev);
+static int identity_mapping(struct pci_dev *pdev)
+{
+	struct device_domain_info *info;
+
+	if (likely(!iommu_identity_mapping))
+		return 0;
+
+
+	list_for_each_entry(info, &si_domain->devices, link)
+		if (info->dev == pdev)
+			return 1;
+	return 0;
+}
+
+static int domain_add_dev_info(struct dmar_domain *domain,
+				  struct pci_dev *pdev)
+{
+	struct device_domain_info *info;
+	unsigned long flags;
+
+	info = alloc_devinfo_mem();
+	if (!info)
+		return -ENOMEM;
+
+	info->segment = pci_domain_nr(pdev->bus);
+	info->bus = pdev->bus->number;
+	info->devfn = pdev->devfn;
+	info->dev = pdev;
+	info->domain = domain;
+
+	spin_lock_irqsave(&device_domain_lock, flags);
+	list_add(&info->link, &domain->devices);
+	list_add(&info->global, &device_domain_list);
+	pdev->dev.archdata.iommu = info;
+	spin_unlock_irqrestore(&device_domain_lock, flags);
+
+	return 0;
+}
+
+static int iommu_prepare_static_identity_mapping(void)
+{
+	int i;
+	struct pci_dev *pdev = NULL;
+	int ret;
+
+	ret = si_domain_init();
+	if (ret)
+		return -EFAULT;
+
+	printk(KERN_INFO "IOMMU: Setting identity map:\n");
+	for_each_pci_dev(pdev) {
+		for (i = 0; i < e820.nr_map; i++) {
+			struct e820entry *ei = &e820.map[i];
+
+			if (ei->type == E820_RAM) {
+				ret = iommu_prepare_identity_map(pdev,
+					ei->addr, ei->addr + ei->size);
+				if (ret)  {
+					printk(KERN_INFO "1:1 mapping to one domain failed.\n");
+					return -EFAULT;
+				}
+			}
+		}
+		ret = domain_add_dev_info(si_domain, pdev);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+int __init init_dmars(void)
 {
 	struct dmar_drhd_unit *drhd;
 	struct dmar_rmrr_unit *rmrr;
@@ -1961,6 +2120,13 @@ static int __init init_dmars(void)
 	int i, ret;
 	int pass_through = 1;
 
+	/*
+	 * In case pass through can not be enabled, iommu tries to use identity
+	 * mapping.
+	 */
+	if (iommu_pass_through)
+		iommu_identity_mapping = 1;
+
 	/*
 	 * for each drhd
 	 *    allocate root
@@ -2090,9 +2256,12 @@ static int __init init_dmars(void)
 
 	/*
 	 * If pass through is not set or not enabled, setup context entries for
-	 * identity mappings for rmrr, gfx, and isa.
+	 * identity mappings for rmrr, gfx, and isa and may fall back to static
+	 * identity mapping if iommu_identity_mapping is set.
 	 */
 	if (!iommu_pass_through) {
+		if (iommu_identity_mapping)
+			iommu_prepare_static_identity_mapping();
 		/*
 		 * For each rmrr
 		 *   for each dev attached to rmrr
@@ -2107,6 +2276,7 @@ static int __init init_dmars(void)
 		 *    endfor
 		 * endfor
 		 */
+		printk(KERN_INFO "IOMMU: Setting RMRR:\n");
 		for_each_rmrr_units(rmrr) {
 			for (i = 0; i < rmrr->devices_cnt; i++) {
 				pdev = rmrr->devices[i];
@@ -2248,6 +2418,52 @@ get_valid_domain_for_dev(struct pci_dev *pdev)
 	return domain;
 }
 
+static int iommu_dummy(struct pci_dev *pdev)
+{
+	return pdev->dev.archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO;
+}
+
+/* Check if the pdev needs to go through non-identity map and unmap process.*/
+static int iommu_no_mapping(struct pci_dev *pdev)
+{
+	int found;
+
+	if (!iommu_identity_mapping)
+		return iommu_dummy(pdev);
+
+	found = identity_mapping(pdev);
+	if (found) {
+		if (pdev->dma_mask > DMA_BIT_MASK(32))
+			return 1;
+		else {
+			/*
+			 * 32 bit DMA is removed from si_domain and fall back
+			 * to non-identity mapping.
+			 */
+			domain_remove_one_dev_info(si_domain, pdev);
+			printk(KERN_INFO "32bit %s uses non-identity mapping\n",
+			       pci_name(pdev));
+			return 0;
+		}
+	} else {
+		/*
+		 * In case of a detached 64 bit DMA device from vm, the device
+		 * is put into si_domain for identity mapping.
+		 */
+		if (pdev->dma_mask > DMA_BIT_MASK(32)) {
+			int ret;
+			ret = domain_add_dev_info(si_domain, pdev);
+			if (!ret) {
+				printk(KERN_INFO "64bit %s uses identity mapping\n",
+				       pci_name(pdev));
+				return 1;
+			}
+		}
+	}
+
+	return iommu_dummy(pdev);
+}
+
 static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr,
 				     size_t size, int dir, u64 dma_mask)
 {
@@ -2260,7 +2476,8 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr,
 	struct intel_iommu *iommu;
 
 	BUG_ON(dir == DMA_NONE);
-	if (pdev->dev.archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO)
+
+	if (iommu_no_mapping(pdev))
 		return paddr;
 
 	domain = get_valid_domain_for_dev(pdev);
@@ -2401,8 +2618,9 @@ static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr,
 	struct iova *iova;
 	struct intel_iommu *iommu;
 
-	if (pdev->dev.archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO)
+	if (iommu_no_mapping(pdev))
 		return;
+
 	domain = find_domain(pdev);
 	BUG_ON(!domain);
 
@@ -2492,7 +2710,7 @@ static void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist,
 	struct scatterlist *sg;
 	struct intel_iommu *iommu;
 
-	if (pdev->dev.archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO)
+	if (iommu_no_mapping(pdev))
 		return;
 
 	domain = find_domain(pdev);
@@ -2553,7 +2771,7 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne
 	struct intel_iommu *iommu;
 
 	BUG_ON(dir == DMA_NONE);
-	if (pdev->dev.archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO)
+	if (iommu_no_mapping(pdev))
 		return intel_nontranslate_map_sg(hwdev, sglist, nelems, dir);
 
 	domain = get_valid_domain_for_dev(pdev);
@@ -2951,31 +3169,6 @@ int __init intel_iommu_init(void)
 	return 0;
 }
 
-static int vm_domain_add_dev_info(struct dmar_domain *domain,
-				  struct pci_dev *pdev)
-{
-	struct device_domain_info *info;
-	unsigned long flags;
-
-	info = alloc_devinfo_mem();
-	if (!info)
-		return -ENOMEM;
-
-	info->segment = pci_domain_nr(pdev->bus);
-	info->bus = pdev->bus->number;
-	info->devfn = pdev->devfn;
-	info->dev = pdev;
-	info->domain = domain;
-
-	spin_lock_irqsave(&device_domain_lock, flags);
-	list_add(&info->link, &domain->devices);
-	list_add(&info->global, &device_domain_list);
-	pdev->dev.archdata.iommu = info;
-	spin_unlock_irqrestore(&device_domain_lock, flags);
-
-	return 0;
-}
-
 static void iommu_detach_dependent_devices(struct intel_iommu *iommu,
 					   struct pci_dev *pdev)
 {
@@ -3003,7 +3196,7 @@ static void iommu_detach_dependent_devices(struct intel_iommu *iommu,
 	}
 }
 
-static void vm_domain_remove_one_dev_info(struct dmar_domain *domain,
+static void domain_remove_one_dev_info(struct dmar_domain *domain,
 					  struct pci_dev *pdev)
 {
 	struct device_domain_info *info;
@@ -3136,7 +3329,7 @@ static struct dmar_domain *iommu_alloc_vm_domain(void)
 	return domain;
 }
 
-static int vm_domain_init(struct dmar_domain *domain, int guest_width)
+static int md_domain_init(struct dmar_domain *domain, int guest_width)
 {
 	int adjust_width;
 
@@ -3227,7 +3420,7 @@ static int intel_iommu_domain_init(struct iommu_domain *domain)
 			"intel_iommu_domain_init: dmar_domain == NULL\n");
 		return -ENOMEM;
 	}
-	if (vm_domain_init(dmar_domain, DEFAULT_DOMAIN_ADDRESS_WIDTH)) {
+	if (md_domain_init(dmar_domain, DEFAULT_DOMAIN_ADDRESS_WIDTH)) {
 		printk(KERN_ERR
 			"intel_iommu_domain_init() failed\n");
 		vm_domain_exit(dmar_domain);
@@ -3262,8 +3455,9 @@ static int intel_iommu_attach_device(struct iommu_domain *domain,
 
 		old_domain = find_domain(pdev);
 		if (old_domain) {
-			if (dmar_domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE)
-				vm_domain_remove_one_dev_info(old_domain, pdev);
+			if (dmar_domain->flags & DOMAIN_FLAG_VIRTUAL_MACHINE ||
+			    dmar_domain->flags & DOMAIN_FLAG_STATIC_IDENTITY)
+				domain_remove_one_dev_info(old_domain, pdev);
 			else
 				domain_remove_dev_info(old_domain);
 		}
@@ -3285,7 +3479,7 @@ static int intel_iommu_attach_device(struct iommu_domain *domain,
 		return -EFAULT;
 	}
 
-	ret = vm_domain_add_dev_info(dmar_domain, pdev);
+	ret = domain_add_dev_info(dmar_domain, pdev);
 	if (ret)
 		return ret;
 
@@ -3299,7 +3493,7 @@ static void intel_iommu_detach_device(struct iommu_domain *domain,
 	struct dmar_domain *dmar_domain = domain->priv;
 	struct pci_dev *pdev = to_pci_dev(dev);
 
-	vm_domain_remove_one_dev_info(dmar_domain, pdev);
+	domain_remove_one_dev_info(dmar_domain, pdev);
 }
 
 static int intel_iommu_map_range(struct iommu_domain *domain,
-- 
cgit v1.2.3-59-g8ed1b


From c4658b4e777bebf69884f4884a9bfb2f84dd71d9 Mon Sep 17 00:00:00 2001
From: Weidong Han <weidong.han@intel.com>
Date: Sat, 23 May 2009 00:41:14 +0800
Subject: Intel-IOMMU, intr-remap: set the whole 128bits of irte when
 modify/free it

Interrupt remapping table entry is 128bits. Currently, it only sets low
64bits of irte in modify_irte and free_irte. This ignores high 64bits
setting of irte, that means source-id setting will be ignored. This patch
sets the whole 128bits of irte when modify/free it. Following source-id
checking patch depends on this.

Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intr_remapping.c | 40 ++++++++++++++++++++++++++--------------
 1 file changed, 26 insertions(+), 14 deletions(-)

diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index 1e83c8c5f985..44025a0c2bb6 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -314,7 +314,8 @@ int modify_irte(int irq, struct irte *irte_modified)
 	index = irq_iommu->irte_index + irq_iommu->sub_handle;
 	irte = &iommu->ir_table->base[index];
 
-	set_64bit((unsigned long *)irte, irte_modified->low);
+	set_64bit((unsigned long *)&irte->low, irte_modified->low);
+	set_64bit((unsigned long *)&irte->high, irte_modified->high);
 	__iommu_flush_cache(iommu, irte, sizeof(*irte));
 
 	rc = qi_flush_iec(iommu, index, 0);
@@ -369,12 +370,32 @@ struct intel_iommu *map_dev_to_ir(struct pci_dev *dev)
 	return drhd->iommu;
 }
 
+static int clear_entries(struct irq_2_iommu *irq_iommu)
+{
+	struct irte *start, *entry, *end;
+	struct intel_iommu *iommu;
+	int index;
+
+	if (irq_iommu->sub_handle)
+		return 0;
+
+	iommu = irq_iommu->iommu;
+	index = irq_iommu->irte_index + irq_iommu->sub_handle;
+
+	start = iommu->ir_table->base + index;
+	end = start + (1 << irq_iommu->irte_mask);
+
+	for (entry = start; entry < end; entry++) {
+		set_64bit((unsigned long *)&entry->low, 0);
+		set_64bit((unsigned long *)&entry->high, 0);
+	}
+
+	return qi_flush_iec(iommu, index, irq_iommu->irte_mask);
+}
+
 int free_irte(int irq)
 {
 	int rc = 0;
-	int index, i;
-	struct irte *irte;
-	struct intel_iommu *iommu;
 	struct irq_2_iommu *irq_iommu;
 	unsigned long flags;
 
@@ -385,16 +406,7 @@ int free_irte(int irq)
 		return -1;
 	}
 
-	iommu = irq_iommu->iommu;
-
-	index = irq_iommu->irte_index + irq_iommu->sub_handle;
-	irte = &iommu->ir_table->base[index];
-
-	if (!irq_iommu->sub_handle) {
-		for (i = 0; i < (1 << irq_iommu->irte_mask); i++)
-			set_64bit((unsigned long *)(irte + i), 0);
-		rc = qi_flush_iec(iommu, index, irq_iommu->irte_mask);
-	}
+	rc = clear_entries(irq_iommu);
 
 	irq_iommu->iommu = NULL;
 	irq_iommu->irte_index = 0;
-- 
cgit v1.2.3-59-g8ed1b


From f007e99c8e2e322b8331aba72414715119a2920d Mon Sep 17 00:00:00 2001
From: Weidong Han <weidong.han@intel.com>
Date: Sat, 23 May 2009 00:41:15 +0800
Subject: Intel-IOMMU, intr-remap: source-id checking

To support domain-isolation usages, the platform hardware must be
capable of uniquely identifying the requestor (source-id) for each
interrupt message. Without source-id checking for interrupt remapping
, a rouge guest/VM with assigned devices can launch interrupt attacks
to bring down anothe guest/VM or the VMM itself.

This patch adds source-id checking for interrupt remapping, and then
really isolates interrupts for guests/VMs with assigned devices.

Because PCI subsystem is not initialized yet when set up IOAPIC
entries, use read_pci_config_byte to access PCI config space directly.

Signed-off-by: Weidong Han <weidong.han@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 arch/x86/kernel/apic/io_apic.c |   6 +++
 drivers/pci/intr_remapping.c   | 120 +++++++++++++++++++++++++++++++++++++++--
 drivers/pci/intr_remapping.h   |   2 +
 include/linux/dmar.h           |  11 ++++
 4 files changed, 136 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index b7a79207295e..4d0216fcb36c 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -1414,6 +1414,9 @@ int setup_ioapic_entry(int apic_id, int irq,
 		irte.vector = vector;
 		irte.dest_id = IRTE_DEST(destination);
 
+		/* Set source-id of interrupt request */
+		set_ioapic_sid(&irte, apic_id);
+
 		modify_irte(irq, &irte);
 
 		ir_entry->index2 = (index >> 15) & 0x1;
@@ -3290,6 +3293,9 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 		irte.vector = cfg->vector;
 		irte.dest_id = IRTE_DEST(dest);
 
+		/* Set source-id of interrupt request */
+		set_msi_sid(&irte, pdev);
+
 		modify_irte(irq, &irte);
 
 		msg->address_hi = MSI_ADDR_BASE_HI;
diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index 44025a0c2bb6..4f5b8712931f 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -10,6 +10,8 @@
 #include <linux/intel-iommu.h>
 #include "intr_remapping.h"
 #include <acpi/acpi.h>
+#include <asm/pci-direct.h>
+#include "pci.h"
 
 static struct ioapic_scope ir_ioapic[MAX_IO_APICS];
 static int ir_ioapic_num;
@@ -418,6 +420,91 @@ int free_irte(int irq)
 	return rc;
 }
 
+/*
+ * source validation type
+ */
+#define SVT_NO_VERIFY		0x0  /* no verification is required */
+#define SVT_VERIFY_SID_SQ	0x1  /* verify using SID and SQ fiels */
+#define SVT_VERIFY_BUS		0x2  /* verify bus of request-id */
+
+/*
+ * source-id qualifier
+ */
+#define SQ_ALL_16	0x0  /* verify all 16 bits of request-id */
+#define SQ_13_IGNORE_1	0x1  /* verify most significant 13 bits, ignore
+			      * the third least significant bit
+			      */
+#define SQ_13_IGNORE_2	0x2  /* verify most significant 13 bits, ignore
+			      * the second and third least significant bits
+			      */
+#define SQ_13_IGNORE_3	0x3  /* verify most significant 13 bits, ignore
+			      * the least three significant bits
+			      */
+
+/*
+ * set SVT, SQ and SID fields of irte to verify
+ * source ids of interrupt requests
+ */
+static void set_irte_sid(struct irte *irte, unsigned int svt,
+			 unsigned int sq, unsigned int sid)
+{
+	irte->svt = svt;
+	irte->sq = sq;
+	irte->sid = sid;
+}
+
+int set_ioapic_sid(struct irte *irte, int apic)
+{
+	int i;
+	u16 sid = 0;
+
+	if (!irte)
+		return -1;
+
+	for (i = 0; i < MAX_IO_APICS; i++) {
+		if (ir_ioapic[i].id == apic) {
+			sid = (ir_ioapic[i].bus << 8) | ir_ioapic[i].devfn;
+			break;
+		}
+	}
+
+	if (sid == 0) {
+		pr_warning("Failed to set source-id of IOAPIC (%d)\n", apic);
+		return -1;
+	}
+
+	set_irte_sid(irte, 1, 0, sid);
+
+	return 0;
+}
+
+int set_msi_sid(struct irte *irte, struct pci_dev *dev)
+{
+	struct pci_dev *bridge;
+
+	if (!irte || !dev)
+		return -1;
+
+	/* PCIe device or Root Complex integrated PCI device */
+	if (dev->is_pcie || !dev->bus->parent) {
+		set_irte_sid(irte, SVT_VERIFY_SID_SQ, SQ_ALL_16,
+			     (dev->bus->number << 8) | dev->devfn);
+		return 0;
+	}
+
+	bridge = pci_find_upstream_pcie_bridge(dev);
+	if (bridge) {
+		if (bridge->is_pcie) /* this is a PCIE-to-PCI/PCIX bridge */
+			set_irte_sid(irte, SVT_VERIFY_BUS, SQ_ALL_16,
+				(bridge->bus->number << 8) | dev->bus->number);
+		else /* this is a legacy PCI bridge */
+			set_irte_sid(irte, SVT_VERIFY_SID_SQ, SQ_ALL_16,
+				(bridge->bus->number << 8) | bridge->devfn);
+	}
+
+	return 0;
+}
+
 static void iommu_set_intr_remapping(struct intel_iommu *iommu, int mode)
 {
 	u64 addr;
@@ -624,6 +711,35 @@ error:
 	return -1;
 }
 
+static void ir_parse_one_ioapic_scope(struct acpi_dmar_device_scope *scope,
+				      struct intel_iommu *iommu)
+{
+	struct acpi_dmar_pci_path *path;
+	u8 bus;
+	int count;
+
+	bus = scope->bus;
+	path = (struct acpi_dmar_pci_path *)(scope + 1);
+	count = (scope->length - sizeof(struct acpi_dmar_device_scope))
+		/ sizeof(struct acpi_dmar_pci_path);
+
+	while (--count > 0) {
+		/*
+		 * Access PCI directly due to the PCI
+		 * subsystem isn't initialized yet.
+		 */
+		bus = read_pci_config_byte(bus, path->dev, path->fn,
+					   PCI_SECONDARY_BUS);
+		path++;
+	}
+
+	ir_ioapic[ir_ioapic_num].bus   = bus;
+	ir_ioapic[ir_ioapic_num].devfn = PCI_DEVFN(path->dev, path->fn);
+	ir_ioapic[ir_ioapic_num].iommu = iommu;
+	ir_ioapic[ir_ioapic_num].id    = scope->enumeration_id;
+	ir_ioapic_num++;
+}
+
 static int ir_parse_ioapic_scope(struct acpi_dmar_header *header,
 				 struct intel_iommu *iommu)
 {
@@ -648,9 +764,7 @@ static int ir_parse_ioapic_scope(struct acpi_dmar_header *header,
 			       " 0x%Lx\n", scope->enumeration_id,
 			       drhd->address);
 
-			ir_ioapic[ir_ioapic_num].iommu = iommu;
-			ir_ioapic[ir_ioapic_num].id = scope->enumeration_id;
-			ir_ioapic_num++;
+			ir_parse_one_ioapic_scope(scope, iommu);
 		}
 		start += scope->length;
 	}
diff --git a/drivers/pci/intr_remapping.h b/drivers/pci/intr_remapping.h
index ca48f0df8ac9..63a263c18415 100644
--- a/drivers/pci/intr_remapping.h
+++ b/drivers/pci/intr_remapping.h
@@ -3,6 +3,8 @@
 struct ioapic_scope {
 	struct intel_iommu *iommu;
 	unsigned int id;
+	unsigned int bus;	/* PCI bus number */
+	unsigned int devfn;	/* PCI devfn number */
 };
 
 #define IR_X2APIC_MODE(mode) (mode ? (1 << 11) : 0)
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index 1731fb5fd775..4a2b162c256a 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -126,6 +126,8 @@ extern int free_irte(int irq);
 extern int irq_remapped(int irq);
 extern struct intel_iommu *map_dev_to_ir(struct pci_dev *dev);
 extern struct intel_iommu *map_ioapic_to_ir(int apic);
+extern int set_ioapic_sid(struct irte *irte, int apic);
+extern int set_msi_sid(struct irte *irte, struct pci_dev *dev);
 #else
 static inline int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
 {
@@ -156,6 +158,15 @@ static inline struct intel_iommu *map_ioapic_to_ir(int apic)
 {
 	return NULL;
 }
+static inline int set_ioapic_sid(struct irte *irte, int apic)
+{
+	return 0;
+}
+static inline int set_msi_sid(struct irte *irte, struct pci_dev *dev)
+{
+	return 0;
+}
+
 #define irq_remapped(irq)		(0)
 #define enable_intr_remapping(mode)	(-1)
 #define disable_intr_remapping()	(0)
-- 
cgit v1.2.3-59-g8ed1b


From 76609a6928bff29ca05a94420ae3e088fbb9c2f9 Mon Sep 17 00:00:00 2001
From: Nelson Castillo <arhuaco@freaks-unidos.net>
Date: Tue, 23 Jun 2009 13:54:32 -0500
Subject: [ARM] GTA02: build fixes (s3c2410_nand_set usage)

This patch fixes two errors we get when building GTA02 kernel.

~ use_bbt is incorrect, we need flash_bbt.

~ We do not need .force_soft_ecc because we can unset
  CONFIG_MTD_NAND_S3C2410_HWECC.

Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>
[ben-linux@fluff.org: updated patch description]
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 arch/arm/mach-s3c2442/mach-gta02.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/arch/arm/mach-s3c2442/mach-gta02.c b/arch/arm/mach-s3c2442/mach-gta02.c
index e23b581aa0e1..0fb385bd9cd9 100644
--- a/arch/arm/mach-s3c2442/mach-gta02.c
+++ b/arch/arm/mach-s3c2442/mach-gta02.c
@@ -433,8 +433,7 @@ static struct s3c2410_nand_set gta02_nand_sets[] = {
 		 */
 		.name		= "neo1973-nand",
 		.nr_chips	= 1,
-		.use_bbt	= 1,
-		.force_soft_ecc	= 1,
+		.flash_bbt	= 1,
 	},
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 1ab52cf910bbbee92861227e6ed77c56b1dc233c Mon Sep 17 00:00:00 2001
From: Baruch Siach <baruch@tkos.co.il>
Date: Mon, 22 Jun 2009 16:36:29 +0300
Subject: i2c: driver for the Synopsys DesignWare I2C controller

The i2c Linux driver for the DesignWare i2c block of Synopsys, which is meant
for AMBA Peripheral Bus. This i2c block is used on SoC chips like the ARM9
based PVG610.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
---
 drivers/i2c/busses/Kconfig          |   9 +
 drivers/i2c/busses/Makefile         |   1 +
 drivers/i2c/busses/i2c-designware.c | 624 ++++++++++++++++++++++++++++++++++++
 3 files changed, 634 insertions(+)
 create mode 100644 drivers/i2c/busses/i2c-designware.c

diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index 3c259ee7ddda..aa87b6a3bbef 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -326,6 +326,15 @@ config I2C_DAVINCI
 	  devices such as DaVinci NIC.
 	  For details please see http://www.ti.com/davinci
 
+config I2C_DESIGNWARE
+	tristate "Synopsys DesignWare"
+	help
+	  If you say yes to this option, support will be included for the
+	  Synopsys DesignWare I2C adapter. Only master mode is supported.
+
+	  This driver can also be built as a module.  If so, the module
+	  will be called i2c-designware.
+
 config I2C_GPIO
 	tristate "GPIO-based bitbanging I2C"
 	depends on GENERIC_GPIO
diff --git a/drivers/i2c/busses/Makefile b/drivers/i2c/busses/Makefile
index edeabf003106..e654263bfc01 100644
--- a/drivers/i2c/busses/Makefile
+++ b/drivers/i2c/busses/Makefile
@@ -30,6 +30,7 @@ obj-$(CONFIG_I2C_AU1550)	+= i2c-au1550.o
 obj-$(CONFIG_I2C_BLACKFIN_TWI)	+= i2c-bfin-twi.o
 obj-$(CONFIG_I2C_CPM)		+= i2c-cpm.o
 obj-$(CONFIG_I2C_DAVINCI)	+= i2c-davinci.o
+obj-$(CONFIG_I2C_DESIGNWARE)	+= i2c-designware.o
 obj-$(CONFIG_I2C_GPIO)		+= i2c-gpio.o
 obj-$(CONFIG_I2C_HIGHLANDER)	+= i2c-highlander.o
 obj-$(CONFIG_I2C_IBM_IIC)	+= i2c-ibm_iic.o
diff --git a/drivers/i2c/busses/i2c-designware.c b/drivers/i2c/busses/i2c-designware.c
new file mode 100644
index 000000000000..b444762e9b9f
--- /dev/null
+++ b/drivers/i2c/busses/i2c-designware.c
@@ -0,0 +1,624 @@
+/*
+ * Synopsys Designware I2C adapter driver (master only).
+ *
+ * Based on the TI DAVINCI I2C adapter driver.
+ *
+ * Copyright (C) 2006 Texas Instruments.
+ * Copyright (C) 2007 MontaVista Software Inc.
+ * Copyright (C) 2009 Provigent Ltd.
+ *
+ * ----------------------------------------------------------------------------
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ * ----------------------------------------------------------------------------
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/i2c.h>
+#include <linux/clk.h>
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/err.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/io.h>
+
+/*
+ * Registers offset
+ */
+#define DW_IC_CON		0x0
+#define DW_IC_TAR		0x4
+#define DW_IC_DATA_CMD		0x10
+#define DW_IC_SS_SCL_HCNT	0x14
+#define DW_IC_SS_SCL_LCNT	0x18
+#define DW_IC_FS_SCL_HCNT	0x1c
+#define DW_IC_FS_SCL_LCNT	0x20
+#define DW_IC_INTR_STAT		0x2c
+#define DW_IC_INTR_MASK		0x30
+#define DW_IC_CLR_INTR		0x40
+#define DW_IC_ENABLE		0x6c
+#define DW_IC_STATUS		0x70
+#define DW_IC_TXFLR		0x74
+#define DW_IC_RXFLR		0x78
+#define DW_IC_COMP_PARAM_1	0xf4
+#define DW_IC_TX_ABRT_SOURCE	0x80
+
+#define DW_IC_CON_MASTER		0x1
+#define DW_IC_CON_SPEED_STD		0x2
+#define DW_IC_CON_SPEED_FAST		0x4
+#define DW_IC_CON_10BITADDR_MASTER	0x10
+#define DW_IC_CON_RESTART_EN		0x20
+#define DW_IC_CON_SLAVE_DISABLE		0x40
+
+#define DW_IC_INTR_TX_EMPTY	0x10
+#define DW_IC_INTR_TX_ABRT	0x40
+#define DW_IC_INTR_STOP_DET	0x200
+
+#define DW_IC_STATUS_ACTIVITY	0x1
+
+#define DW_IC_ERR_TX_ABRT	0x1
+
+/*
+ * status codes
+ */
+#define STATUS_IDLE			0x0
+#define STATUS_WRITE_IN_PROGRESS	0x1
+#define STATUS_READ_IN_PROGRESS		0x2
+
+#define TIMEOUT			20 /* ms */
+
+/*
+ * hardware abort codes from the DW_IC_TX_ABRT_SOURCE register
+ *
+ * only expected abort codes are listed here
+ * refer to the datasheet for the full list
+ */
+#define ABRT_7B_ADDR_NOACK	0
+#define ABRT_10ADDR1_NOACK	1
+#define ABRT_10ADDR2_NOACK	2
+#define ABRT_TXDATA_NOACK	3
+#define ABRT_GCALL_NOACK	4
+#define ABRT_GCALL_READ		5
+#define ABRT_SBYTE_ACKDET	7
+#define ABRT_SBYTE_NORSTRT	9
+#define ABRT_10B_RD_NORSTRT	10
+#define ARB_MASTER_DIS		11
+#define ARB_LOST		12
+
+static char *abort_sources[] = {
+	[ABRT_7B_ADDR_NOACK]	=
+		"slave address not acknowledged (7bit mode)",
+	[ABRT_10ADDR1_NOACK]	=
+		"first address byte not acknowledged (10bit mode)",
+	[ABRT_10ADDR2_NOACK]	=
+		"second address byte not acknowledged (10bit mode)",
+	[ABRT_TXDATA_NOACK]		=
+		"data not acknowledged",
+	[ABRT_GCALL_NOACK]		=
+		"no acknowledgement for a general call",
+	[ABRT_GCALL_READ]		=
+		"read after general call",
+	[ABRT_SBYTE_ACKDET]		=
+		"start byte acknowledged",
+	[ABRT_SBYTE_NORSTRT]	=
+		"trying to send start byte when restart is disabled",
+	[ABRT_10B_RD_NORSTRT]	=
+		"trying to read when restart is disabled (10bit mode)",
+	[ARB_MASTER_DIS]		=
+		"trying to use disabled adapter",
+	[ARB_LOST]			=
+		"lost arbitration",
+};
+
+/**
+ * struct dw_i2c_dev - private i2c-designware data
+ * @dev: driver model device node
+ * @base: IO registers pointer
+ * @cmd_complete: tx completion indicator
+ * @pump_msg: continue in progress transfers
+ * @lock: protect this struct and IO registers
+ * @clk: input reference clock
+ * @cmd_err: run time hadware error code
+ * @msgs: points to an array of messages currently being transfered
+ * @msgs_num: the number of elements in msgs
+ * @msg_write_idx: the element index of the current tx message in the msgs
+ *	array
+ * @tx_buf_len: the length of the current tx buffer
+ * @tx_buf: the current tx buffer
+ * @msg_read_idx: the element index of the current rx message in the msgs
+ *	array
+ * @rx_buf_len: the length of the current rx buffer
+ * @rx_buf: the current rx buffer
+ * @msg_err: error status of the current transfer
+ * @status: i2c master status, one of STATUS_*
+ * @abort_source: copy of the TX_ABRT_SOURCE register
+ * @irq: interrupt number for the i2c master
+ * @adapter: i2c subsystem adapter node
+ * @tx_fifo_depth: depth of the hardware tx fifo
+ * @rx_fifo_depth: depth of the hardware rx fifo
+ */
+struct dw_i2c_dev {
+	struct device		*dev;
+	void __iomem		*base;
+	struct completion	cmd_complete;
+	struct tasklet_struct	pump_msg;
+	struct mutex		lock;
+	struct clk		*clk;
+	int			cmd_err;
+	struct i2c_msg		*msgs;
+	int			msgs_num;
+	int			msg_write_idx;
+	u16			tx_buf_len;
+	u8			*tx_buf;
+	int			msg_read_idx;
+	u16			rx_buf_len;
+	u8			*rx_buf;
+	int			msg_err;
+	unsigned int		status;
+	u16			abort_source;
+	int			irq;
+	struct i2c_adapter	adapter;
+	unsigned int		tx_fifo_depth;
+	unsigned int		rx_fifo_depth;
+};
+
+/**
+ * i2c_dw_init() - initialize the designware i2c master hardware
+ * @dev: device private data
+ *
+ * This functions configures and enables the I2C master.
+ * This function is called during I2C init function, and in case of timeout at
+ * run time.
+ */
+static void i2c_dw_init(struct dw_i2c_dev *dev)
+{
+	u32 input_clock_khz = clk_get_rate(dev->clk) / 1000;
+	u16 ic_con;
+
+	/* Disable the adapter */
+	writeb(0, dev->base + DW_IC_ENABLE);
+
+	/* set standard and fast speed deviders for high/low periods */
+	writew((input_clock_khz * 40 / 10000)+1, /* std speed high, 4us */
+			dev->base + DW_IC_SS_SCL_HCNT);
+	writew((input_clock_khz * 47 / 10000)+1, /* std speed low, 4.7us */
+			dev->base + DW_IC_SS_SCL_LCNT);
+	writew((input_clock_khz *  6 / 10000)+1, /* fast speed high, 0.6us */
+			dev->base + DW_IC_FS_SCL_HCNT);
+	writew((input_clock_khz * 13 / 10000)+1, /* fast speed low, 1.3us */
+			dev->base + DW_IC_FS_SCL_LCNT);
+
+	/* configure the i2c master */
+	ic_con = DW_IC_CON_MASTER | DW_IC_CON_SLAVE_DISABLE |
+		DW_IC_CON_RESTART_EN | DW_IC_CON_SPEED_FAST;
+	writew(ic_con, dev->base + DW_IC_CON);
+}
+
+/*
+ * Waiting for bus not busy
+ */
+static int i2c_dw_wait_bus_not_busy(struct dw_i2c_dev *dev)
+{
+	int timeout = TIMEOUT;
+
+	while (readb(dev->base + DW_IC_STATUS) & DW_IC_STATUS_ACTIVITY) {
+		if (timeout <= 0) {
+			dev_warn(dev->dev, "timeout waiting for bus ready\n");
+			return -ETIMEDOUT;
+		}
+		timeout--;
+		mdelay(1);
+	}
+
+	return 0;
+}
+
+/*
+ * Initiate low level master read/write transaction.
+ * This function is called from i2c_dw_xfer when starting a transfer.
+ * This function is also called from dw_i2c_pump_msg to continue a transfer
+ * that is longer than the size of the TX FIFO.
+ */
+static void
+i2c_dw_xfer_msg(struct i2c_adapter *adap)
+{
+	struct dw_i2c_dev *dev = i2c_get_adapdata(adap);
+	struct i2c_msg *msgs = dev->msgs;
+	int num = dev->msgs_num;
+	u16 ic_con, intr_mask;
+	int tx_limit = dev->tx_fifo_depth - readb(dev->base + DW_IC_TXFLR);
+	int rx_limit = dev->rx_fifo_depth - readb(dev->base + DW_IC_RXFLR);
+	u16 addr = msgs[dev->msg_write_idx].addr;
+	u16 buf_len = dev->tx_buf_len;
+
+	if (!(dev->status & STATUS_WRITE_IN_PROGRESS)) {
+		/* Disable the adapter */
+		writeb(0, dev->base + DW_IC_ENABLE);
+
+		/* set the slave (target) address */
+		writew(msgs[dev->msg_write_idx].addr, dev->base + DW_IC_TAR);
+
+		/* if the slave address is ten bit address, enable 10BITADDR */
+		ic_con = readw(dev->base + DW_IC_CON);
+		if (msgs[dev->msg_write_idx].flags & I2C_M_TEN)
+			ic_con |= DW_IC_CON_10BITADDR_MASTER;
+		else
+			ic_con &= ~DW_IC_CON_10BITADDR_MASTER;
+		writew(ic_con, dev->base + DW_IC_CON);
+
+		/* Enable the adapter */
+		writeb(1, dev->base + DW_IC_ENABLE);
+	}
+
+	for (; dev->msg_write_idx < num; dev->msg_write_idx++) {
+		/* if target address has changed, we need to
+		 * reprogram the target address in the i2c
+		 * adapter when we are done with this transfer
+		 */
+		if (msgs[dev->msg_write_idx].addr != addr)
+			return;
+
+		if (msgs[dev->msg_write_idx].len == 0) {
+			dev_err(dev->dev,
+				"%s: invalid message length\n", __func__);
+			dev->msg_err = -EINVAL;
+			return;
+		}
+
+		if (!(dev->status & STATUS_WRITE_IN_PROGRESS)) {
+			/* new i2c_msg */
+			dev->tx_buf = msgs[dev->msg_write_idx].buf;
+			buf_len = msgs[dev->msg_write_idx].len;
+		}
+
+		while (buf_len > 0 && tx_limit > 0 && rx_limit > 0) {
+			if (msgs[dev->msg_write_idx].flags & I2C_M_RD) {
+				writew(0x100, dev->base + DW_IC_DATA_CMD);
+				rx_limit--;
+			} else
+				writew(*(dev->tx_buf++),
+						dev->base + DW_IC_DATA_CMD);
+			tx_limit--; buf_len--;
+		}
+	}
+
+	intr_mask = DW_IC_INTR_STOP_DET | DW_IC_INTR_TX_ABRT;
+	if (buf_len > 0) { /* more bytes to be written */
+		intr_mask |= DW_IC_INTR_TX_EMPTY;
+		dev->status |= STATUS_WRITE_IN_PROGRESS;
+	} else
+		dev->status &= ~STATUS_WRITE_IN_PROGRESS;
+	writew(intr_mask, dev->base + DW_IC_INTR_MASK);
+
+	dev->tx_buf_len = buf_len;
+}
+
+static void
+i2c_dw_read(struct i2c_adapter *adap)
+{
+	struct dw_i2c_dev *dev = i2c_get_adapdata(adap);
+	struct i2c_msg *msgs = dev->msgs;
+	int num = dev->msgs_num;
+	u16 addr = msgs[dev->msg_read_idx].addr;
+	int rx_valid = readw(dev->base + DW_IC_RXFLR);
+
+	for (; dev->msg_read_idx < num; dev->msg_read_idx++) {
+		u16 len;
+		u8 *buf;
+
+		if (!(msgs[dev->msg_read_idx].flags & I2C_M_RD))
+			continue;
+
+		/* different i2c client, reprogram the i2c adapter */
+		if (msgs[dev->msg_read_idx].addr != addr)
+			return;
+
+		if (!(dev->status & STATUS_READ_IN_PROGRESS)) {
+			len = msgs[dev->msg_read_idx].len;
+			buf = msgs[dev->msg_read_idx].buf;
+		} else {
+			len = dev->rx_buf_len;
+			buf = dev->rx_buf;
+		}
+
+		for (; len > 0 && rx_valid > 0; len--, rx_valid--)
+			*buf++ = readb(dev->base + DW_IC_DATA_CMD);
+
+		if (len > 0) {
+			dev->status |= STATUS_READ_IN_PROGRESS;
+			dev->rx_buf_len = len;
+			dev->rx_buf = buf;
+			return;
+		} else
+			dev->status &= ~STATUS_READ_IN_PROGRESS;
+	}
+}
+
+/*
+ * Prepare controller for a transaction and call i2c_dw_xfer_msg
+ */
+static int
+i2c_dw_xfer(struct i2c_adapter *adap, struct i2c_msg msgs[], int num)
+{
+	struct dw_i2c_dev *dev = i2c_get_adapdata(adap);
+	int ret;
+
+	dev_dbg(dev->dev, "%s: msgs: %d\n", __func__, num);
+
+	mutex_lock(&dev->lock);
+
+	INIT_COMPLETION(dev->cmd_complete);
+	dev->msgs = msgs;
+	dev->msgs_num = num;
+	dev->cmd_err = 0;
+	dev->msg_write_idx = 0;
+	dev->msg_read_idx = 0;
+	dev->msg_err = 0;
+	dev->status = STATUS_IDLE;
+
+	ret = i2c_dw_wait_bus_not_busy(dev);
+	if (ret < 0)
+		goto done;
+
+	/* start the transfers */
+	i2c_dw_xfer_msg(adap);
+
+	/* wait for tx to complete */
+	ret = wait_for_completion_interruptible_timeout(&dev->cmd_complete, HZ);
+	if (ret == 0) {
+		dev_err(dev->dev, "controller timed out\n");
+		i2c_dw_init(dev);
+		ret = -ETIMEDOUT;
+		goto done;
+	} else if (ret < 0)
+		goto done;
+
+	if (dev->msg_err) {
+		ret = dev->msg_err;
+		goto done;
+	}
+
+	/* no error */
+	if (likely(!dev->cmd_err)) {
+		/* read rx fifo, and disable the adapter */
+		do {
+			i2c_dw_read(adap);
+		} while (dev->status & STATUS_READ_IN_PROGRESS);
+		writeb(0, dev->base + DW_IC_ENABLE);
+		ret = num;
+		goto done;
+	}
+
+	/* We have an error */
+	if (dev->cmd_err == DW_IC_ERR_TX_ABRT) {
+		unsigned long abort_source = dev->abort_source;
+		int i;
+
+		for_each_bit(i, &abort_source, ARRAY_SIZE(abort_sources)) {
+		    dev_err(dev->dev, "%s: %s\n", __func__, abort_sources[i]);
+		}
+	}
+	ret = -EIO;
+
+done:
+	mutex_unlock(&dev->lock);
+
+	return ret;
+}
+
+static u32 i2c_dw_func(struct i2c_adapter *adap)
+{
+	return I2C_FUNC_I2C | I2C_FUNC_10BIT_ADDR;
+}
+
+static void dw_i2c_pump_msg(unsigned long data)
+{
+	struct dw_i2c_dev *dev = (struct dw_i2c_dev *) data;
+	u16 intr_mask;
+
+	i2c_dw_read(&dev->adapter);
+	i2c_dw_xfer_msg(&dev->adapter);
+
+	intr_mask = DW_IC_INTR_STOP_DET | DW_IC_INTR_TX_ABRT;
+	if (dev->status & STATUS_WRITE_IN_PROGRESS)
+		intr_mask |= DW_IC_INTR_TX_EMPTY;
+	writew(intr_mask, dev->base + DW_IC_INTR_MASK);
+}
+
+/*
+ * Interrupt service routine. This gets called whenever an I2C interrupt
+ * occurs.
+ */
+static irqreturn_t i2c_dw_isr(int this_irq, void *dev_id)
+{
+	struct dw_i2c_dev *dev = dev_id;
+	u16 stat;
+
+	stat = readw(dev->base + DW_IC_INTR_STAT);
+	dev_dbg(dev->dev, "%s: stat=0x%x\n", __func__, stat);
+	if (stat & DW_IC_INTR_TX_ABRT) {
+		dev->abort_source = readw(dev->base + DW_IC_TX_ABRT_SOURCE);
+		dev->cmd_err |= DW_IC_ERR_TX_ABRT;
+		dev->status = STATUS_IDLE;
+	} else if (stat & DW_IC_INTR_TX_EMPTY)
+		tasklet_schedule(&dev->pump_msg);
+
+	readb(dev->base + DW_IC_CLR_INTR);	/* clear interrupts */
+	writew(0, dev->base + DW_IC_INTR_MASK);	/* disable interrupts */
+	if (stat & (DW_IC_INTR_TX_ABRT | DW_IC_INTR_STOP_DET))
+		complete(&dev->cmd_complete);
+
+	return IRQ_HANDLED;
+}
+
+static struct i2c_algorithm i2c_dw_algo = {
+	.master_xfer	= i2c_dw_xfer,
+	.functionality	= i2c_dw_func,
+};
+
+static int __devinit dw_i2c_probe(struct platform_device *pdev)
+{
+	struct dw_i2c_dev *dev;
+	struct i2c_adapter *adap;
+	struct resource *mem, *irq, *ioarea;
+	int r;
+
+	/* NOTE: driver uses the static register mapping */
+	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!mem) {
+		dev_err(&pdev->dev, "no mem resource?\n");
+		return -EINVAL;
+	}
+
+	irq = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (!irq) {
+		dev_err(&pdev->dev, "no irq resource?\n");
+		return -EINVAL;
+	}
+
+	ioarea = request_mem_region(mem->start, resource_size(mem),
+			pdev->name);
+	if (!ioarea) {
+		dev_err(&pdev->dev, "I2C region already claimed\n");
+		return -EBUSY;
+	}
+
+	dev = kzalloc(sizeof(struct dw_i2c_dev), GFP_KERNEL);
+	if (!dev) {
+		r = -ENOMEM;
+		goto err_release_region;
+	}
+
+	init_completion(&dev->cmd_complete);
+	tasklet_init(&dev->pump_msg, dw_i2c_pump_msg, (unsigned long) dev);
+	mutex_init(&dev->lock);
+	dev->dev = get_device(&pdev->dev);
+	dev->irq = irq->start;
+	platform_set_drvdata(pdev, dev);
+
+	dev->clk = clk_get(&pdev->dev, NULL);
+	if (IS_ERR(dev->clk)) {
+		r = -ENODEV;
+		goto err_free_mem;
+	}
+	clk_enable(dev->clk);
+
+	dev->base = ioremap(mem->start, resource_size(mem));
+	if (dev->base == NULL) {
+		dev_err(&pdev->dev, "failure mapping io resources\n");
+		r = -EBUSY;
+		goto err_unuse_clocks;
+	}
+	{
+		u32 param1 = readl(dev->base + DW_IC_COMP_PARAM_1);
+
+		dev->tx_fifo_depth = ((param1 >> 16) & 0xff) + 1;
+		dev->rx_fifo_depth = ((param1 >> 8)  & 0xff) + 1;
+	}
+	i2c_dw_init(dev);
+
+	writew(0, dev->base + DW_IC_INTR_MASK); /* disable IRQ */
+	r = request_irq(dev->irq, i2c_dw_isr, 0, pdev->name, dev);
+	if (r) {
+		dev_err(&pdev->dev, "failure requesting irq %i\n", dev->irq);
+		goto err_iounmap;
+	}
+
+	adap = &dev->adapter;
+	i2c_set_adapdata(adap, dev);
+	adap->owner = THIS_MODULE;
+	adap->class = I2C_CLASS_HWMON;
+	strlcpy(adap->name, "Synopsys DesignWare I2C adapter",
+			sizeof(adap->name));
+	adap->algo = &i2c_dw_algo;
+	adap->dev.parent = &pdev->dev;
+
+	adap->nr = pdev->id;
+	r = i2c_add_numbered_adapter(adap);
+	if (r) {
+		dev_err(&pdev->dev, "failure adding adapter\n");
+		goto err_free_irq;
+	}
+
+	return 0;
+
+err_free_irq:
+	free_irq(dev->irq, dev);
+err_iounmap:
+	iounmap(dev->base);
+err_unuse_clocks:
+	clk_disable(dev->clk);
+	clk_put(dev->clk);
+	dev->clk = NULL;
+err_free_mem:
+	platform_set_drvdata(pdev, NULL);
+	put_device(&pdev->dev);
+	kfree(dev);
+err_release_region:
+	release_mem_region(mem->start, resource_size(mem));
+
+	return r;
+}
+
+static int __devexit dw_i2c_remove(struct platform_device *pdev)
+{
+	struct dw_i2c_dev *dev = platform_get_drvdata(pdev);
+	struct resource *mem;
+
+	platform_set_drvdata(pdev, NULL);
+	i2c_del_adapter(&dev->adapter);
+	put_device(&pdev->dev);
+
+	clk_disable(dev->clk);
+	clk_put(dev->clk);
+	dev->clk = NULL;
+
+	writeb(0, dev->base + DW_IC_ENABLE);
+	free_irq(dev->irq, dev);
+	kfree(dev);
+
+	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	release_mem_region(mem->start, resource_size(mem));
+	return 0;
+}
+
+/* work with hotplug and coldplug */
+MODULE_ALIAS("platform:i2c_designware");
+
+static struct platform_driver dw_i2c_driver = {
+	.remove		= __devexit_p(dw_i2c_remove),
+	.driver		= {
+		.name	= "i2c_designware",
+		.owner	= THIS_MODULE,
+	},
+};
+
+static int __init dw_i2c_init_driver(void)
+{
+	return platform_driver_probe(&dw_i2c_driver, dw_i2c_probe);
+}
+module_init(dw_i2c_init_driver);
+
+static void __exit dw_i2c_exit_driver(void)
+{
+	platform_driver_unregister(&dw_i2c_driver);
+}
+module_exit(dw_i2c_exit_driver);
+
+MODULE_AUTHOR("Baruch Siach <baruch@tkos.co.il>");
+MODULE_DESCRIPTION("Synopsys DesignWare I2C bus adapter");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-59-g8ed1b


From d55d87fdff8252d0e2f7c28c2d443aee17e9d70f Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 22 Jun 2009 02:25:25 +0000
Subject: net: Move rx skb_orphan call to where needed

In order to get the tun driver to account packets, we need to be
able to receive packets with destructors set.  To be on the safe
side, I added an skb_orphan call for all protocols by default since
some of them (IP in particular) cannot handle receiving packets
destructors properly.

Now it seems that at least one protocol (CAN) expects to be able
to pass skb->sk through the rx path without getting clobbered.

So this patch attempts to fix this properly by moving the skb_orphan
call to where it's actually needed.  In particular, I've added it
to skb_set_owner_[rw] which is what most users of skb->destructor
call.

This is actually an improvement for tun too since it means that
we only give back the amount charged to the socket when the skb
is passed to another socket that will also be charged accordingly.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Tested-by: Oliver Hartkopp <olver@hartkopp.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/sctp/sctp.h      | 1 +
 include/net/sock.h           | 2 ++
 net/ax25/ax25_in.c           | 3 +--
 net/core/dev.c               | 2 --
 net/irda/af_irda.c           | 3 ---
 net/irda/ircomm/ircomm_lmp.c | 1 +
 6 files changed, 5 insertions(+), 7 deletions(-)

diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h
index 9f80a7668289..d16a304cbed4 100644
--- a/include/net/sctp/sctp.h
+++ b/include/net/sctp/sctp.h
@@ -448,6 +448,7 @@ static inline void sctp_skb_set_owner_r(struct sk_buff *skb, struct sock *sk)
 {
 	struct sctp_ulpevent *event = sctp_skb2event(skb);
 
+	skb_orphan(skb);
 	skb->sk = sk;
 	skb->destructor = sctp_sock_rfree;
 	atomic_add(event->rmem_len, &sk->sk_rmem_alloc);
diff --git a/include/net/sock.h b/include/net/sock.h
index 570c7a12b54e..7f5c41cc45a9 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1250,6 +1250,7 @@ static inline int sk_has_allocations(const struct sock *sk)
 
 static inline void skb_set_owner_w(struct sk_buff *skb, struct sock *sk)
 {
+	skb_orphan(skb);
 	skb->sk = sk;
 	skb->destructor = sock_wfree;
 	/*
@@ -1262,6 +1263,7 @@ static inline void skb_set_owner_w(struct sk_buff *skb, struct sock *sk)
 
 static inline void skb_set_owner_r(struct sk_buff *skb, struct sock *sk)
 {
+	skb_orphan(skb);
 	skb->sk = sk;
 	skb->destructor = sock_rfree;
 	atomic_add(skb->truesize, &sk->sk_rmem_alloc);
diff --git a/net/ax25/ax25_in.c b/net/ax25/ax25_in.c
index 5f1d2107a1dd..de56d3983de0 100644
--- a/net/ax25/ax25_in.c
+++ b/net/ax25/ax25_in.c
@@ -437,8 +437,7 @@ free:
 int ax25_kiss_rcv(struct sk_buff *skb, struct net_device *dev,
 		  struct packet_type *ptype, struct net_device *orig_dev)
 {
-	skb->sk = NULL;		/* Initially we don't know who it's for */
-	skb->destructor = NULL;	/* Who initializes this, dammit?! */
+	skb_orphan(skb);
 
 	if (!net_eq(dev_net(dev), &init_net)) {
 		kfree_skb(skb);
diff --git a/net/core/dev.c b/net/core/dev.c
index baf2dc13a34a..60b572812278 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2310,8 +2310,6 @@ ncls:
 	if (!skb)
 		goto out;
 
-	skb_orphan(skb);
-
 	type = skb->protocol;
 	list_for_each_entry_rcu(ptype,
 			&ptype_base[ntohs(type) & PTYPE_HASH_MASK], list) {
diff --git a/net/irda/af_irda.c b/net/irda/af_irda.c
index 5922febe25c4..cb762c8723ea 100644
--- a/net/irda/af_irda.c
+++ b/net/irda/af_irda.c
@@ -913,9 +913,6 @@ static int irda_accept(struct socket *sock, struct socket *newsock, int flags)
 	/* Clean up the original one to keep it in listen state */
 	irttp_listen(self->tsap);
 
-	/* Wow ! What is that ? Jean II */
-	skb->sk = NULL;
-	skb->destructor = NULL;
 	kfree_skb(skb);
 	sk->sk_ack_backlog--;
 
diff --git a/net/irda/ircomm/ircomm_lmp.c b/net/irda/ircomm/ircomm_lmp.c
index 67c99d20857f..7ba96618660e 100644
--- a/net/irda/ircomm/ircomm_lmp.c
+++ b/net/irda/ircomm/ircomm_lmp.c
@@ -196,6 +196,7 @@ static int ircomm_lmp_data_request(struct ircomm_cb *self,
 	/* Don't forget to refcount it - see ircomm_tty_do_softint() */
 	skb_get(skb);
 
+	skb_orphan(skb);
 	skb->destructor = ircomm_lmp_flow_control;
 
 	if ((self->pkt_count++ > 7) && (self->flow_status == FLOW_START)) {
-- 
cgit v1.2.3-59-g8ed1b


From b6280b47a7a42970d098a3059f4ebe7e55e90d8d Mon Sep 17 00:00:00 2001
From: Neil Horman <nhorman@tuxdriver.com>
Date: Mon, 22 Jun 2009 10:18:53 +0000
Subject: ipv4 routing: Ensure that route cache entries are usable and
 reclaimable with caching is off

When route caching is disabled (rt_caching returns false), We still use route
cache entries that are created and passed into rt_intern_hash once.  These
routes need to be made usable for the one call path that holds a reference to
them, and they need to be reclaimed when they're finished with their use.  To be
made usable, they need to be associated with a neighbor table entry (which they
currently are not), otherwise iproute_finish2 just discards the packet, since we
don't know which L2 peer to send the packet to.  To do this binding, we need to
follow the path a bit higher up in rt_intern_hash, which calls
arp_bind_neighbour, but not assign the route entry to the hash table.
Currently, if caching is off, we simply assign the route to the rp pointer and
are reutrn success.  This patch associates us with a neighbor entry first.

Secondly, we need to make sure that any single use routes like this are known to
the garbage collector when caching is off.  If caching is off, and we try to
hash in a route, it will leak when its refcount reaches zero.  To avoid this,
this patch calls rt_free on the route cache entry passed into rt_intern_hash.
This places us on the gc list for the route cache garbage collector, so that
when its refcount reaches zero, it will be reclaimed (Thanks to Alexey for this
suggestion).

I've tested this on a local system here, and with these patches in place, I'm
able to maintain routed connectivity to remote systems, even if I set
/proc/sys/net/ipv4/rt_cache_rebuild_count to -1, which forces rt_caching to
return false.

Signed-off-by: Neil Horman <nhorman@redhat.com>
Reported-by: Jarek Poplawski <jarkao2@gmail.com>
Reported-by: Maxime Bizon <mbizon@freebox.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/route.c | 26 +++++++++++++++++++++++---
 1 file changed, 23 insertions(+), 3 deletions(-)

diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 65b3a8b11a6c..278f46f5011b 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1093,8 +1093,27 @@ restart:
 		 * If we drop it here, the callers have no way to resolve routes
 		 * when we're not caching.  Instead, just point *rp at rt, so
 		 * the caller gets a single use out of the route
+		 * Note that we do rt_free on this new route entry, so that
+		 * once its refcount hits zero, we are still able to reap it
+		 * (Thanks Alexey)
+		 * Note also the rt_free uses call_rcu.  We don't actually
+		 * need rcu protection here, this is just our path to get
+		 * on the route gc list.
 		 */
-		goto report_and_exit;
+
+		if (rt->rt_type == RTN_UNICAST || rt->fl.iif == 0) {
+			int err = arp_bind_neighbour(&rt->u.dst);
+			if (err) {
+				if (net_ratelimit())
+					printk(KERN_WARNING
+					    "Neighbour table failure & not caching routes.\n");
+				rt_drop(rt);
+				return err;
+			}
+		}
+
+		rt_free(rt);
+		goto skip_hashing;
 	}
 
 	rthp = &rt_hash_table[hash].chain;
@@ -1211,7 +1230,8 @@ restart:
 #if RT_CACHE_DEBUG >= 2
 	if (rt->u.dst.rt_next) {
 		struct rtable *trt;
-		printk(KERN_DEBUG "rt_cache @%02x: %pI4", hash, &rt->rt_dst);
+		printk(KERN_DEBUG "rt_cache @%02x: %pI4",
+		       hash, &rt->rt_dst);
 		for (trt = rt->u.dst.rt_next; trt; trt = trt->u.dst.rt_next)
 			printk(" . %pI4", &trt->rt_dst);
 		printk("\n");
@@ -1226,7 +1246,7 @@ restart:
 
 	spin_unlock_bh(rt_hash_lock_addr(hash));
 
-report_and_exit:
+skip_hashing:
 	if (rp)
 		*rp = rt;
 	else
-- 
cgit v1.2.3-59-g8ed1b


From e5a673742e34eca8ecb13c3e54ceee2c268351a0 Mon Sep 17 00:00:00 2001
From: Ron Mercer <ron.mercer@qlogic.com>
Date: Tue, 23 Jun 2009 09:00:01 +0000
Subject: qla3xxx: Give the PHY time to come out of reset.

Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/qla3xxx.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/qla3xxx.c b/drivers/net/qla3xxx.c
index bbc6d4d3cc94..68be714d356f 100644
--- a/drivers/net/qla3xxx.c
+++ b/drivers/net/qla3xxx.c
@@ -3150,7 +3150,8 @@ static int ql_adapter_initialize(struct ql3_adapter *qdev)
 	ql_write_common_reg(qdev, &port_regs->CommonRegs.serialPortInterfaceReg,
 			    (ISP_SERIAL_PORT_IF_WE |
 			     (ISP_SERIAL_PORT_IF_WE << 16)));
-
+	/* Give the PHY time to come out of reset. */
+	mdelay(100);
 	qdev->port_link_state = LS_DOWN;
 	netif_carrier_off(qdev->ndev);
 
-- 
cgit v1.2.3-59-g8ed1b


From 0f77ca928b5d1ea17afc7a95682b6534611a719c Mon Sep 17 00:00:00 2001
From: Ron Mercer <ron.mercer@qlogic.com>
Date: Tue, 23 Jun 2009 09:00:02 +0000
Subject: qla3xxx: Don't sleep while holding lock.

Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/qla3xxx.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/qla3xxx.c b/drivers/net/qla3xxx.c
index 68be714d356f..3e4b67aaa6ea 100644
--- a/drivers/net/qla3xxx.c
+++ b/drivers/net/qla3xxx.c
@@ -3142,6 +3142,7 @@ static int ql_adapter_initialize(struct ql3_adapter *qdev)
 						(void __iomem *)port_regs;
 	u32 delay = 10;
 	int status = 0;
+	unsigned long hw_flags = 0;
 
 	if(ql_mii_setup(qdev))
 		return -1;
@@ -3351,7 +3352,9 @@ static int ql_adapter_initialize(struct ql3_adapter *qdev)
 		value = ql_read_page0_reg(qdev, &port_regs->portStatus);
 		if (value & PORT_STATUS_IC)
 			break;
+		spin_unlock_irqrestore(&qdev->hw_lock, hw_flags);
 		msleep(500);
+		spin_lock_irqsave(&qdev->hw_lock, hw_flags);
 	} while (--delay);
 
 	if (delay == 0) {
-- 
cgit v1.2.3-59-g8ed1b


From 056c308d3e4859334b519033d62ef050f0e0e261 Mon Sep 17 00:00:00 2001
From: Zhang Rui <rui.zhang@intel.com>
Date: Mon, 22 Jun 2009 11:31:14 +0800
Subject: Show the physical device node of backlight class device.

Create symbol link from backlight class device to ACPI video device.

More and more laptops are shipped with multiple ACPI
video devices, while we export only one of them to userspace.

With this patch applied, we can know which ACPI video device
is used by "cat /sys/class/backlight/acpi_video0/device/path".

Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/video.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c
index 1bdfb37377e3..9de143af3625 100644
--- a/drivers/acpi/video.c
+++ b/drivers/acpi/video.c
@@ -976,6 +976,11 @@ static void acpi_video_device_find_cap(struct acpi_video_device *device)
 		device->backlight->props.max_brightness = device->brightness->count-3;
 		kfree(name);
 
+		result = sysfs_create_link(&device->backlight->dev.kobj,
+					   &device->dev->dev.kobj, "device");
+		if (result)
+			printk(KERN_ERR PREFIX "Create sysfs link\n");
+
 		device->cdev = thermal_cooling_device_register("LCD",
 					device->dev, &video_cooling_ops);
 		if (IS_ERR(device->cdev))
@@ -1990,6 +1995,7 @@ static int acpi_video_bus_put_one_device(struct acpi_video_device *device)
 	status = acpi_remove_notify_handler(device->dev->handle,
 					    ACPI_DEVICE_NOTIFY,
 					    acpi_video_device_notify);
+	sysfs_remove_link(&device->backlight->dev.kobj, "device");
 	backlight_device_unregister(device->backlight);
 	if (device->cdev) {
 		sysfs_remove_link(&device->dev->dev.kobj,
-- 
cgit v1.2.3-59-g8ed1b


From c02256be79a1a3557332ac51e653d574a2a7d2b5 Mon Sep 17 00:00:00 2001
From: Zhang Rui <rui.zhang@intel.com>
Date: Tue, 23 Jun 2009 10:20:29 +0800
Subject: ACPI: fix a deadlock in hotplug case

we used to run the hotplug code in keventd_wq.
But when hot removing the ACPI battery device,
power_supply_unregister invokes flush_scheduled_work.
This causes a deadlock. i.e
1. When dock is unplugged, all the hotplug code is run on kevent_wq.
2. the hotplug code removes all the child devices of dock device.
3. removing the child device may invoke flush_scheduled_work
4. flush_scheduled_work waits until all the work on kevent_wq to be
   finished, while this will never be true because the hotplug code
   is running on keventd_wq...

Introduce a new workqueue for hotplug in this patch.
http://bugzilla.kernel.org/show_bug.cgi?id=13533

Tested-by: Paul Martin <pm@debian.org>
Tested-by: Vojtech Gondzala <vojtech.gondzala@gmail.com>
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Reviewed-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/osl.c | 25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)

diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
index d916bea729f1..71670719d61a 100644
--- a/drivers/acpi/osl.c
+++ b/drivers/acpi/osl.c
@@ -79,6 +79,7 @@ static acpi_osd_handler acpi_irq_handler;
 static void *acpi_irq_context;
 static struct workqueue_struct *kacpid_wq;
 static struct workqueue_struct *kacpi_notify_wq;
+static struct workqueue_struct *kacpi_hotplug_wq;
 
 struct acpi_res_list {
 	resource_size_t start;
@@ -192,8 +193,10 @@ acpi_status acpi_os_initialize1(void)
 {
 	kacpid_wq = create_singlethread_workqueue("kacpid");
 	kacpi_notify_wq = create_singlethread_workqueue("kacpi_notify");
+	kacpi_hotplug_wq = create_singlethread_workqueue("kacpi_hotplug");
 	BUG_ON(!kacpid_wq);
 	BUG_ON(!kacpi_notify_wq);
+	BUG_ON(!kacpi_hotplug_wq);
 	return AE_OK;
 }
 
@@ -206,6 +209,7 @@ acpi_status acpi_os_terminate(void)
 
 	destroy_workqueue(kacpid_wq);
 	destroy_workqueue(kacpi_notify_wq);
+	destroy_workqueue(kacpi_hotplug_wq);
 
 	return AE_OK;
 }
@@ -716,6 +720,7 @@ static acpi_status __acpi_os_execute(acpi_execute_type type,
 	acpi_status status = AE_OK;
 	struct acpi_os_dpc *dpc;
 	struct workqueue_struct *queue;
+	work_func_t func;
 	int ret;
 	ACPI_DEBUG_PRINT((ACPI_DB_EXEC,
 			  "Scheduling function [%p(%p)] for deferred execution.\n",
@@ -740,15 +745,17 @@ static acpi_status __acpi_os_execute(acpi_execute_type type,
 	dpc->function = function;
 	dpc->context = context;
 
-	if (!hp) {
-		INIT_WORK(&dpc->work, acpi_os_execute_deferred);
-		queue = (type == OSL_NOTIFY_HANDLER) ?
-			kacpi_notify_wq : kacpid_wq;
-		ret = queue_work(queue, &dpc->work);
-	} else {
-		INIT_WORK(&dpc->work, acpi_os_execute_hp_deferred);
-		ret = schedule_work(&dpc->work);
-	}
+	/*
+	 * We can't run hotplug code in keventd_wq/kacpid_wq/kacpid_notify_wq
+	 * because the hotplug code may call driver .remove() functions,
+	 * which invoke flush_scheduled_work/acpi_os_wait_events_complete
+	 * to flush these workqueues.
+	 */
+	queue = hp ? kacpi_hotplug_wq :
+		(type == OSL_NOTIFY_HANDLER ? kacpi_notify_wq : kacpid_wq);
+	func = hp ? acpi_os_execute_hp_deferred : acpi_os_execute_deferred;
+	INIT_WORK(&dpc->work, func);
+	ret = queue_work(queue, &dpc->work);
 
 	if (!ret) {
 		printk(KERN_ERR PREFIX
-- 
cgit v1.2.3-59-g8ed1b


From 35a7c64fbc77bab4ca8ae477e8ab278ccd679ce2 Mon Sep 17 00:00:00 2001
From: Zhang Rui <rui.zhang@intel.com>
Date: Mon, 22 Jun 2009 11:31:17 +0800
Subject: ACPI: DMI to disable Vista compatibility on some Sony laptops

Linux claims Vista compatibility to the BIOS for a number of
reasons, but this brings hard lockup on some Sony laptops.

Disable Vista compatibility via DMI for these laptops unless
we can figure out what Vista is doing for this platform.
http://bugzilla.kernel.org/show_bug.cgi?id=12904

Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/blacklist.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/acpi/blacklist.c b/drivers/acpi/blacklist.c
index 09c69806c1fc..f6baa77deefb 100644
--- a/drivers/acpi/blacklist.c
+++ b/drivers/acpi/blacklist.c
@@ -192,6 +192,22 @@ static struct dmi_system_id acpi_osi_dmi_table[] __initdata = {
 		     DMI_MATCH(DMI_PRODUCT_NAME, "ESPRIMO Mobile V5505"),
 		},
 	},
+	{
+	.callback = dmi_disable_osi_vista,
+	.ident = "Sony VGN-NS10J_S",
+	.matches = {
+		     DMI_MATCH(DMI_SYS_VENDOR, "Sony Corporation"),
+		     DMI_MATCH(DMI_PRODUCT_NAME, "VGN-NS10J_S"),
+		},
+	},
+	{
+	.callback = dmi_disable_osi_vista,
+	.ident = "Sony VGN-SR290J",
+	.matches = {
+		     DMI_MATCH(DMI_SYS_VENDOR, "Sony Corporation"),
+		     DMI_MATCH(DMI_PRODUCT_NAME, "Sony VGN-SR290J"),
+		},
+	},
 
 	/*
 	 * BIOS invocation of _OSI(Linux) is almost always a BIOS bug.
-- 
cgit v1.2.3-59-g8ed1b


From 86e437f077c68112edcb6854ec036ed7e3f9a7f3 Mon Sep 17 00:00:00 2001
From: Zhao Yakui <yakui.zhao@intel.com>
Date: Tue, 16 Jun 2009 11:23:13 +0800
Subject: ACPI: Add the reference count to avoid unloading ACPI video bus twice

Sometimes both acpi video and i915 driver are compiled as modules.
And there exists the strict dependency between the two drivers.
The acpi video bus will be unloaded in course of unloading the i915 driver.
If we unload the acpi video driver, then the kernel oops will be triggered.

Add the reference count to avoid unloading the ACPI video bus twice.
The reference count should be checked before unregistering the acpi video bus.
If the reference count is already zero, it won't unregister it again.
And after the acpi video bus is already unregistered, the reference count
will be set to zero.

http://bugzilla.kernel.org/show_bug.cgi?id=13396

Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Acked-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/video.c                 | 41 ++++++++++++++++++++++++++++++------
 drivers/gpu/drm/i915/i915_opregion.c |  2 +-
 include/acpi/video.h                 |  4 ++--
 3 files changed, 38 insertions(+), 9 deletions(-)

diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c
index 1bdfb37377e3..a63566ba230b 100644
--- a/drivers/acpi/video.c
+++ b/drivers/acpi/video.c
@@ -76,6 +76,7 @@ MODULE_LICENSE("GPL");
 static int brightness_switch_enabled = 1;
 module_param(brightness_switch_enabled, bool, 0644);
 
+static int register_count = 0;
 static int acpi_video_bus_add(struct acpi_device *device);
 static int acpi_video_bus_remove(struct acpi_device *device, int type);
 static int acpi_video_resume(struct acpi_device *device);
@@ -2318,6 +2319,13 @@ static int __init intel_opregion_present(void)
 int acpi_video_register(void)
 {
 	int result = 0;
+	if (register_count) {
+		/*
+		 * if the function of acpi_video_register is already called,
+		 * don't register the acpi_vide_bus again and return no error.
+		 */
+		return 0;
+	}
 
 	acpi_video_dir = proc_mkdir(ACPI_VIDEO_CLASS, acpi_root_dir);
 	if (!acpi_video_dir)
@@ -2329,10 +2337,35 @@ int acpi_video_register(void)
 		return -ENODEV;
 	}
 
+	/*
+	 * When the acpi_video_bus is loaded successfully, increase
+	 * the counter reference.
+	 */
+	register_count = 1;
+
 	return 0;
 }
 EXPORT_SYMBOL(acpi_video_register);
 
+void acpi_video_unregister(void)
+{
+	if (!register_count) {
+		/*
+		 * If the acpi video bus is already unloaded, don't
+		 * unload it again and return directly.
+		 */
+		return;
+	}
+	acpi_bus_unregister_driver(&acpi_video_bus);
+
+	remove_proc_entry(ACPI_VIDEO_CLASS, acpi_root_dir);
+
+	register_count = 0;
+
+	return;
+}
+EXPORT_SYMBOL(acpi_video_unregister);
+
 /*
  * This is kind of nasty. Hardware using Intel chipsets may require
  * the video opregion code to be run first in order to initialise
@@ -2350,16 +2383,12 @@ static int __init acpi_video_init(void)
 	return acpi_video_register();
 }
 
-void acpi_video_exit(void)
+static void __exit acpi_video_exit(void)
 {
-
-	acpi_bus_unregister_driver(&acpi_video_bus);
-
-	remove_proc_entry(ACPI_VIDEO_CLASS, acpi_root_dir);
+	acpi_video_unregister();
 
 	return;
 }
-EXPORT_SYMBOL(acpi_video_exit);
 
 module_init(acpi_video_init);
 module_exit(acpi_video_exit);
diff --git a/drivers/gpu/drm/i915/i915_opregion.c b/drivers/gpu/drm/i915/i915_opregion.c
index dc425e74a268..e4b4e8898e39 100644
--- a/drivers/gpu/drm/i915/i915_opregion.c
+++ b/drivers/gpu/drm/i915/i915_opregion.c
@@ -419,7 +419,7 @@ void intel_opregion_free(struct drm_device *dev, int suspend)
 		return;
 
 	if (!suspend)
-		acpi_video_exit();
+		acpi_video_unregister();
 
 	opregion->acpi->drdy = 0;
 
diff --git a/include/acpi/video.h b/include/acpi/video.h
index af6fe95fd3d0..cf7be3dd157b 100644
--- a/include/acpi/video.h
+++ b/include/acpi/video.h
@@ -3,10 +3,10 @@
 
 #if (defined CONFIG_ACPI_VIDEO || defined CONFIG_ACPI_VIDEO_MODULE)
 extern int acpi_video_register(void);
-extern int acpi_video_exit(void);
+extern void acpi_video_unregister(void);
 #else
 static inline int acpi_video_register(void) { return 0; }
-static inline void acpi_video_exit(void) { return; }
+static inline void acpi_video_unregister(void) { return; }
 #endif
 
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From c8d72a5e76988140bfdfd8722f2228d94e7fa10f Mon Sep 17 00:00:00 2001
From: Zhang Rui <rui.zhang@intel.com>
Date: Mon, 22 Jun 2009 11:31:16 +0800
Subject: ACPI: run ACPI device hot removal in kacpi_hotplug_wq

Now that new interface is available,
convert to using it rather than creating a new kernel thread.

Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/scan.c | 20 ++++++++------------
 1 file changed, 8 insertions(+), 12 deletions(-)

diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index 8ff510b91d88..9c6e42e7cd13 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -95,7 +95,7 @@ acpi_device_modalias_show(struct device *dev, struct device_attribute *attr, cha
 }
 static DEVICE_ATTR(modalias, 0444, acpi_device_modalias_show, NULL);
 
-static int acpi_bus_hot_remove_device(void *context)
+static void acpi_bus_hot_remove_device(void *context)
 {
 	struct acpi_device *device;
 	acpi_handle handle = context;
@@ -104,10 +104,10 @@ static int acpi_bus_hot_remove_device(void *context)
 	acpi_status status = AE_OK;
 
 	if (acpi_bus_get_device(handle, &device))
-		return 0;
+		return;
 
 	if (!device)
-		return 0;
+		return;
 
 	ACPI_DEBUG_PRINT((ACPI_DB_INFO,
 		"Hot-removing device %s...\n", dev_name(&device->dev)));
@@ -115,7 +115,7 @@ static int acpi_bus_hot_remove_device(void *context)
 	if (acpi_bus_trim(device, 1)) {
 		printk(KERN_ERR PREFIX
 				"Removing device failed\n");
-		return -1;
+		return;
 	}
 
 	/* power off device */
@@ -142,9 +142,10 @@ static int acpi_bus_hot_remove_device(void *context)
 	 */
 	status = acpi_evaluate_object(handle, "_EJ0", &arg_list, NULL);
 	if (ACPI_FAILURE(status))
-		return -ENODEV;
+		printk(KERN_WARNING PREFIX
+				"Eject device failed\n");
 
-	return 0;
+	return;
 }
 
 static ssize_t
@@ -155,7 +156,6 @@ acpi_eject_store(struct device *d, struct device_attribute *attr,
 	acpi_status status;
 	acpi_object_type type = 0;
 	struct acpi_device *acpi_device = to_acpi_device(d);
-	struct task_struct *task;
 
 	if ((!count) || (buf[0] != '1')) {
 		return -EINVAL;
@@ -172,11 +172,7 @@ acpi_eject_store(struct device *d, struct device_attribute *attr,
 		goto err;
 	}
 
-	/* remove the device in another thread to fix the deadlock issue */
-	task = kthread_run(acpi_bus_hot_remove_device,
-				acpi_device->handle, "acpi_hot_remove_device");
-	if (IS_ERR(task))
-		ret = PTR_ERR(task);
+	acpi_os_hotplug_execute(acpi_bus_hot_remove_device, acpi_device->handle);
 err:
 	return ret;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 152a4e630f7ffdd7ff64427c4ba488dc0bce76af Mon Sep 17 00:00:00 2001
From: Zhang Rui <rui.zhang@intel.com>
Date: Mon, 22 Jun 2009 11:31:18 +0800
Subject: ACPI: video: DMI workaround broken Acer 7720 BIOS enabling display
 brightness

http://bugzilla.kernel.org/show_bug.cgi?id=13121

Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/video.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c
index 1bdfb37377e3..1eff1625672a 100644
--- a/drivers/acpi/video.c
+++ b/drivers/acpi/video.c
@@ -586,6 +586,14 @@ static struct dmi_system_id video_dmi_table[] __initdata = {
 		DMI_MATCH(DMI_PRODUCT_NAME, "Aspire 5315"),
 		},
 	},
+	{
+	 .callback = video_set_bqc_offset,
+	 .ident = "Acer Aspire 7720",
+	 .matches = {
+		DMI_MATCH(DMI_BOARD_VENDOR, "Acer"),
+		DMI_MATCH(DMI_PRODUCT_NAME, "Aspire 7720"),
+		},
+	},
 	{}
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 35aa901c0b66cb3c2eeee23f13624014825a44a8 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Thu, 11 Jun 2009 14:31:33 -0400
Subject: Audit: fix audit watch use after free

When an audit watch is added to a parent the temporary watch inside the
original krule from userspace is freed.  Yet the original watch is used after
the real watch was created in audit_add_rules()

Signed-off-by: Eric Paris <eparis@redhat.com>
---
 kernel/auditfilter.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index 713098ee5a02..19c0a0a2cede 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -1320,6 +1320,8 @@ static inline int audit_add_rule(struct audit_entry *entry)
 			mutex_unlock(&audit_filter_mutex);
 			goto error;
 		}
+		/* entry->rule.watch may have changed during audit_add_watch() */
+		watch = entry->rule.watch;
 		h = audit_hash_ino((u32)watch->ino);
 		list = &audit_inode_hash[h];
 	}
-- 
cgit v1.2.3-59-g8ed1b


From b87ce6e4187c24b06483c8266822ce5e6b7fa7f3 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Thu, 11 Jun 2009 14:31:34 -0400
Subject: Audit: better estimation of execve record length

The audit execve record splitting code estimates the length of the message
generated.  But it forgot to include the "" that wrap each string in its
estimation.  This means that execve messages with lots of tiny (1-2 byte)
arguments could still cause records greater than 8k to be emitted.  Simply
fix the estimate.

Signed-off-by: Eric Paris <eparis@redhat.com>
---
 kernel/auditsc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 7d6ac7c1f414..b14d234b85f3 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -1024,8 +1024,8 @@ static int audit_log_single_execve_arg(struct audit_context *context,
 {
 	char arg_num_len_buf[12];
 	const char __user *tmp_p = p;
-	/* how many digits are in arg_num? 3 is the length of " a=" */
-	size_t arg_num_len = snprintf(arg_num_len_buf, 12, "%d", arg_num) + 3;
+	/* how many digits are in arg_num? 5 is the length of ' a=""' */
+	size_t arg_num_len = snprintf(arg_num_len_buf, 12, "%d", arg_num) + 5;
 	size_t len, len_left, to_send;
 	size_t max_execve_audit_len = MAX_EXECVE_AUDIT_LEN;
 	unsigned int i, has_cntl = 0, too_long = 0;
-- 
cgit v1.2.3-59-g8ed1b


From e85188f424c8eec7f311deed9a70bec57aeed741 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Thu, 11 Jun 2009 14:31:34 -0400
Subject: Audit: dereferencing krule as if it were an audit_watch

audit_update_watch() runs all of the rules for a given watch and duplicates
them, attaches a new watch to them, and then when it finishes that process
and has called free on all of the old rules (ok maybe still inside the rcu
grace period) it proceeds to use the last element from list_for_each_entry_safe()
as if it were a krule rather than being the audit_watch which was anchoring
the list to output a message about audit rules changing.

This patch unfies the audit message from two different places into a helper
function and calls it from the correct location in audit_update_rules().  We
will now get an audit message about the config changing for each rule (with
each rules filterkey) rather than the previous garbage.

Signed-off-by: Eric Paris <eparis@redhat.com>
---
 kernel/auditfilter.c | 58 ++++++++++++++++++++++------------------------------
 1 file changed, 24 insertions(+), 34 deletions(-)

diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index 19c0a0a2cede..e7466dd145c9 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -977,6 +977,27 @@ static struct audit_entry *audit_dupe_rule(struct audit_krule *old,
 	return entry;
 }
 
+static void audit_watch_log_rule_change(struct audit_krule *r, struct audit_watch *w, char *op)
+{
+	if (audit_enabled) {
+		struct audit_buffer *ab;
+		ab = audit_log_start(NULL, GFP_NOFS, AUDIT_CONFIG_CHANGE);
+		audit_log_format(ab, "auid=%u ses=%u op=",
+				 audit_get_loginuid(current),
+				 audit_get_sessionid(current));
+		audit_log_string(ab, op);
+		audit_log_format(ab, " path=");
+		audit_log_untrustedstring(ab, w->path);
+		if (r->filterkey) {
+			audit_log_format(ab, " key=");
+			audit_log_untrustedstring(ab, r->filterkey);
+		} else
+			audit_log_format(ab, " key=(null)");
+		audit_log_format(ab, " list=%d res=1", r->listnr);
+		audit_log_end(ab);
+	}
+}
+
 /* Update inode info in audit rules based on filesystem event. */
 static void audit_update_watch(struct audit_parent *parent,
 			       const char *dname, dev_t dev,
@@ -1023,24 +1044,11 @@ static void audit_update_watch(struct audit_parent *parent,
 					     &nentry->rule.list);
 			}
 
+			audit_watch_log_rule_change(r, owatch, "updated rules");
+
 			call_rcu(&oentry->rcu, audit_free_rule_rcu);
 		}
 
-		if (audit_enabled) {
-			struct audit_buffer *ab;
-			ab = audit_log_start(NULL, GFP_NOFS,
-				AUDIT_CONFIG_CHANGE);
-			audit_log_format(ab, "auid=%u ses=%u",
-				audit_get_loginuid(current),
-				audit_get_sessionid(current));
-			audit_log_format(ab,
-				" op=updated rules specifying path=");
-			audit_log_untrustedstring(ab, owatch->path);
-			audit_log_format(ab, " with dev=%u ino=%lu\n",
-				 dev, ino);
-			audit_log_format(ab, " list=%d res=1", r->listnr);
-			audit_log_end(ab);
-		}
 		audit_remove_watch(owatch);
 		goto add_watch_to_parent; /* event applies to a single watch */
 	}
@@ -1065,25 +1073,7 @@ static void audit_remove_parent_watches(struct audit_parent *parent)
 	list_for_each_entry_safe(w, nextw, &parent->watches, wlist) {
 		list_for_each_entry_safe(r, nextr, &w->rules, rlist) {
 			e = container_of(r, struct audit_entry, rule);
-			if (audit_enabled) {
-				struct audit_buffer *ab;
-				ab = audit_log_start(NULL, GFP_NOFS,
-					AUDIT_CONFIG_CHANGE);
-				audit_log_format(ab, "auid=%u ses=%u",
-					audit_get_loginuid(current),
-					audit_get_sessionid(current));
-				audit_log_format(ab, " op=remove rule path=");
-				audit_log_untrustedstring(ab, w->path);
-				if (r->filterkey) {
-					audit_log_format(ab, " key=");
-					audit_log_untrustedstring(ab,
-							r->filterkey);
-				} else
-					audit_log_format(ab, " key=(null)");
-				audit_log_format(ab, " list=%d res=1",
-					r->listnr);
-				audit_log_end(ab);
-			}
+			audit_watch_log_rule_change(r, w, "remove rule");
 			list_del(&r->rlist);
 			list_del(&r->list);
 			list_del_rcu(&e->list);
-- 
cgit v1.2.3-59-g8ed1b


From 038cbcf65fd6a30c79e3917690b8c46321a27915 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Thu, 11 Jun 2009 14:31:35 -0400
Subject: Audit: unify the printk of an skb when auditd not around

Remove code duplication of skb printk when auditd is not around in userspace
to deal with this message.

Signed-off-by: Eric Paris <eparis@redhat.com>
---
 kernel/audit.c | 39 ++++++++++++++++++++++-----------------
 1 file changed, 22 insertions(+), 17 deletions(-)

diff --git a/kernel/audit.c b/kernel/audit.c
index 9442c3533ba9..f7ab4a479cdd 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -375,6 +375,25 @@ static void audit_hold_skb(struct sk_buff *skb)
 		kfree_skb(skb);
 }
 
+/*
+ * For one reason or another this nlh isn't getting delivered to the userspace
+ * audit daemon, just send it to printk.
+ */
+static void audit_printk_skb(struct sk_buff *skb)
+{
+	struct nlmsghdr *nlh = nlmsg_hdr(skb);
+	char *data = NLMSG_DATA(nlh);
+
+	if (nlh->nlmsg_type != AUDIT_EOE) {
+		if (printk_ratelimit())
+			printk(KERN_NOTICE "type=%d %s\n", nlh->nlmsg_type, data);
+		else
+			audit_log_lost("printk limit exceeded\n");
+	}
+
+	audit_hold_skb(skb);
+}
+
 static void kauditd_send_skb(struct sk_buff *skb)
 {
 	int err;
@@ -427,14 +446,8 @@ static int kauditd_thread(void *dummy)
 		if (skb) {
 			if (audit_pid)
 				kauditd_send_skb(skb);
-			else {
-				if (printk_ratelimit())
-					printk(KERN_NOTICE "%s\n", skb->data + NLMSG_SPACE(0));
-				else
-					audit_log_lost("printk limit exceeded\n");
-
-				audit_hold_skb(skb);
-			}
+			else
+				audit_printk_skb(skb);
 		} else {
 			DECLARE_WAITQUEUE(wait, current);
 			set_current_state(TASK_INTERRUPTIBLE);
@@ -1475,15 +1488,7 @@ void audit_log_end(struct audit_buffer *ab)
 			skb_queue_tail(&audit_skb_queue, ab->skb);
 			wake_up_interruptible(&kauditd_wait);
 		} else {
-			if (nlh->nlmsg_type != AUDIT_EOE) {
-				if (printk_ratelimit()) {
-					printk(KERN_NOTICE "type=%d %s\n",
-						nlh->nlmsg_type,
-						ab->skb->data + NLMSG_SPACE(0));
-				} else
-					audit_log_lost("printk limit exceeded\n");
-			}
-			audit_hold_skb(ab->skb);
+			audit_printk_skb(ab->skb);
 		}
 		ab->skb = NULL;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From ee080e6ce93d5993390bccf68c1df5efd9351276 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Thu, 11 Jun 2009 14:31:35 -0400
Subject: Audit: cleanup netlink mesg handling

The audit handling of netlink messages is all over the place.  Clean things
up, use predetermined macros, generally make it more readable.

Signed-off-by: Eric Paris <eparis@redhat.com>
---
 kernel/audit.c | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/kernel/audit.c b/kernel/audit.c
index f7ab4a479cdd..01082a1d2bc5 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -528,22 +528,20 @@ struct sk_buff *audit_make_reply(int pid, int seq, int type, int done,
 {
 	struct sk_buff	*skb;
 	struct nlmsghdr	*nlh;
-	int		len = NLMSG_SPACE(size);
 	void		*data;
 	int		flags = multi ? NLM_F_MULTI : 0;
 	int		t     = done  ? NLMSG_DONE  : type;
 
-	skb = alloc_skb(len, GFP_KERNEL);
+	skb = nlmsg_new(size, GFP_KERNEL);
 	if (!skb)
 		return NULL;
 
-	nlh		 = NLMSG_PUT(skb, pid, seq, t, size);
-	nlh->nlmsg_flags = flags;
-	data		 = NLMSG_DATA(nlh);
+	nlh	= NLMSG_NEW(skb, pid, seq, t, size, flags);
+	data	= NLMSG_DATA(nlh);
 	memcpy(data, payload, size);
 	return skb;
 
-nlmsg_failure:			/* Used by NLMSG_PUT */
+nlmsg_failure:			/* Used by NLMSG_NEW */
 	if (skb)
 		kfree_skb(skb);
 	return NULL;
@@ -1083,18 +1081,20 @@ static struct audit_buffer * audit_buffer_alloc(struct audit_context *ctx,
 			goto err;
 	}
 
-	ab->skb = alloc_skb(AUDIT_BUFSIZ, gfp_mask);
-	if (!ab->skb)
-		goto err;
-
 	ab->ctx = ctx;
 	ab->gfp_mask = gfp_mask;
-	nlh = (struct nlmsghdr *)skb_put(ab->skb, NLMSG_SPACE(0));
-	nlh->nlmsg_type = type;
-	nlh->nlmsg_flags = 0;
-	nlh->nlmsg_pid = 0;
-	nlh->nlmsg_seq = 0;
+
+	ab->skb = nlmsg_new(AUDIT_BUFSIZ, gfp_mask);
+	if (!ab->skb)
+		goto nlmsg_failure;
+
+	nlh = NLMSG_NEW(ab->skb, 0, 0, type, 0, 0);
+
 	return ab;
+
+nlmsg_failure:                  /* Used by NLMSG_NEW */
+	kfree_skb(ab->skb);
+	ab->skb = NULL;
 err:
 	audit_buffer_free(ab);
 	return NULL;
-- 
cgit v1.2.3-59-g8ed1b


From ea7ae60bfe39aeedfb29571c47280bf0067ee5f3 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Thu, 11 Jun 2009 14:31:35 -0400
Subject: Audit: clean up audit_receive_skb

audit_receive_skb is hard to clearly parse what it is doing to the netlink
message.  Clean the function up so it is easy and clear to see what is going
on.

Signed-off-by: Eric Paris <eparis@redhat.com>
---
 kernel/audit.c | 35 ++++++++++++++++++-----------------
 1 file changed, 18 insertions(+), 17 deletions(-)

diff --git a/kernel/audit.c b/kernel/audit.c
index 01082a1d2bc5..ce77e81a0e71 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -937,28 +937,29 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 }
 
 /*
- * Get message from skb (based on rtnetlink_rcv_skb).  Each message is
- * processed by audit_receive_msg.  Malformed skbs with wrong length are
- * discarded silently.
+ * Get message from skb.  Each message is processed by audit_receive_msg.
+ * Malformed skbs with wrong length are discarded silently.
  */
 static void audit_receive_skb(struct sk_buff *skb)
 {
-	int		err;
-	struct nlmsghdr	*nlh;
-	u32		rlen;
+	struct nlmsghdr *nlh;
+	/*
+	 * len MUST be signed for NLMSG_NEXT to be able to dec it below 0
+	 * if the nlmsg_len was not aligned
+	 */
+	int len;
+	int err;
 
-	while (skb->len >= NLMSG_SPACE(0)) {
-		nlh = nlmsg_hdr(skb);
-		if (nlh->nlmsg_len < sizeof(*nlh) || skb->len < nlh->nlmsg_len)
-			return;
-		rlen = NLMSG_ALIGN(nlh->nlmsg_len);
-		if (rlen > skb->len)
-			rlen = skb->len;
-		if ((err = audit_receive_msg(skb, nlh))) {
+	nlh = nlmsg_hdr(skb);
+	len = skb->len;
+
+	while (NLMSG_OK(nlh, len)) {
+		err = audit_receive_msg(skb, nlh);
+		/* if err or if this message says it wants a response */
+		if (err || (nlh->nlmsg_flags & NLM_F_ACK))
 			netlink_ack(skb, nlh, err);
-		} else if (nlh->nlmsg_flags & NLM_F_ACK)
-			netlink_ack(skb, nlh, 0);
-		skb_pull(skb, rlen);
+
+		nlh = NLMSG_NEXT(nlh, len);
 	}
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From cfcad62c74abfef83762dc05a556d21bdf3980a2 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Thu, 11 Jun 2009 14:31:36 -0400
Subject: audit: seperate audit inode watches into a subfile

In preparation for converting audit to use fsnotify instead of inotify we
seperate the inode watching code into it's own file.  This is similar to
how the audit tree watching code is already seperated into audit_tree.c

Signed-off-by: Eric Paris <eparis@redhat.com>
---
 kernel/Makefile      |   2 +-
 kernel/audit.c       |  16 --
 kernel/audit.h       |  39 ++--
 kernel/audit_watch.c | 534 +++++++++++++++++++++++++++++++++++++++++++++++++++
 kernel/auditfilter.c | 481 ++--------------------------------------------
 kernel/auditsc.c     |   6 +-
 6 files changed, 572 insertions(+), 506 deletions(-)
 create mode 100644 kernel/audit_watch.c

diff --git a/kernel/Makefile b/kernel/Makefile
index 0a32cb21ec97..da750010a6fc 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -70,7 +70,7 @@ obj-$(CONFIG_RESOURCE_COUNTERS) += res_counter.o
 obj-$(CONFIG_STOP_MACHINE) += stop_machine.o
 obj-$(CONFIG_KPROBES_SANITY_TEST) += test_kprobes.o
 obj-$(CONFIG_AUDIT) += audit.o auditfilter.o
-obj-$(CONFIG_AUDITSYSCALL) += auditsc.o
+obj-$(CONFIG_AUDITSYSCALL) += auditsc.o audit_watch.o
 obj-$(CONFIG_GCOV_KERNEL) += gcov/
 obj-$(CONFIG_AUDIT_TREE) += audit_tree.o
 obj-$(CONFIG_KPROBES) += kprobes.o
diff --git a/kernel/audit.c b/kernel/audit.c
index ce77e81a0e71..e07ad2340dbe 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -115,9 +115,6 @@ static atomic_t    audit_lost = ATOMIC_INIT(0);
 /* The netlink socket. */
 static struct sock *audit_sock;
 
-/* Inotify handle. */
-struct inotify_handle *audit_ih;
-
 /* Hash for inode-based rules */
 struct list_head audit_inode_hash[AUDIT_INODE_BUCKETS];
 
@@ -971,13 +968,6 @@ static void audit_receive(struct sk_buff  *skb)
 	mutex_unlock(&audit_cmd_mutex);
 }
 
-#ifdef CONFIG_AUDITSYSCALL
-static const struct inotify_operations audit_inotify_ops = {
-	.handle_event	= audit_handle_ievent,
-	.destroy_watch	= audit_free_parent,
-};
-#endif
-
 /* Initialize audit support at boot time. */
 static int __init audit_init(void)
 {
@@ -1003,12 +993,6 @@ static int __init audit_init(void)
 
 	audit_log(NULL, GFP_KERNEL, AUDIT_KERNEL, "initialized");
 
-#ifdef CONFIG_AUDITSYSCALL
-	audit_ih = inotify_init(&audit_inotify_ops);
-	if (IS_ERR(audit_ih))
-		audit_panic("cannot initialize inotify handle");
-#endif
-
 	for (i = 0; i < AUDIT_INODE_BUCKETS; i++)
 		INIT_LIST_HEAD(&audit_inode_hash[i]);
 
diff --git a/kernel/audit.h b/kernel/audit.h
index 16f18cac661b..704d5b01d9fd 100644
--- a/kernel/audit.h
+++ b/kernel/audit.h
@@ -53,18 +53,7 @@ enum audit_state {
 };
 
 /* Rule lists */
-struct audit_parent;
-
-struct audit_watch {
-	atomic_t		count;	/* reference count */
-	char			*path;	/* insertion path */
-	dev_t			dev;	/* associated superblock device */
-	unsigned long		ino;	/* associated inode number */
-	struct audit_parent	*parent; /* associated parent */
-	struct list_head	wlist;	/* entry in parent->watches list */
-	struct list_head	rules;	/* associated rules */
-};
-
+struct audit_watch;
 struct audit_tree;
 struct audit_chunk;
 
@@ -108,19 +97,31 @@ struct audit_netlink_list {
 
 int audit_send_list(void *);
 
-struct inotify_watch;
-/* Inotify handle */
-extern struct inotify_handle *audit_ih;
-
-extern void audit_free_parent(struct inotify_watch *);
-extern void audit_handle_ievent(struct inotify_watch *, u32, u32, u32,
-				const char *, struct inode *);
 extern int selinux_audit_rule_update(void);
 
 extern struct mutex audit_filter_mutex;
 extern void audit_free_rule_rcu(struct rcu_head *);
 extern struct list_head audit_filter_list[];
 
+/* audit watch functions */
+extern unsigned long audit_watch_inode(struct audit_watch *watch);
+extern dev_t audit_watch_dev(struct audit_watch *watch);
+extern void audit_put_watch(struct audit_watch *watch);
+extern void audit_get_watch(struct audit_watch *watch);
+extern int audit_to_watch(struct audit_krule *krule, char *path, int len, u32 op);
+extern int audit_get_nd(char *path, struct nameidata **ndp, struct nameidata **ndw);
+extern void audit_put_nd(struct nameidata *ndp, struct nameidata *ndw);
+extern int audit_add_watch(struct audit_krule *krule, struct nameidata *ndp,
+			   struct nameidata *ndw);
+extern void audit_remove_watch(struct audit_watch *watch);
+extern void audit_remove_watch_rule(struct audit_krule *krule, struct list_head *list);
+extern void audit_inotify_unregister(struct list_head *in_list);
+extern char *audit_watch_path(struct audit_watch *watch);
+extern struct list_head *audit_watch_rules(struct audit_watch *watch);
+
+extern struct audit_entry *audit_dupe_rule(struct audit_krule *old,
+					   struct audit_watch *watch);
+
 #ifdef CONFIG_AUDIT_TREE
 extern struct audit_chunk *audit_tree_lookup(const struct inode *);
 extern void audit_put_chunk(struct audit_chunk *);
diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c
new file mode 100644
index 000000000000..da8be6d39c1a
--- /dev/null
+++ b/kernel/audit_watch.c
@@ -0,0 +1,534 @@
+/* audit_watch.c -- watching inodes
+ *
+ * Copyright 2003-2009 Red Hat, Inc.
+ * Copyright 2005 Hewlett-Packard Development Company, L.P.
+ * Copyright 2005 IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+
+#include <linux/kernel.h>
+#include <linux/audit.h>
+#include <linux/kthread.h>
+#include <linux/mutex.h>
+#include <linux/fs.h>
+#include <linux/namei.h>
+#include <linux/netlink.h>
+#include <linux/sched.h>
+#include <linux/inotify.h>
+#include <linux/security.h>
+#include "audit.h"
+
+/*
+ * Reference counting:
+ *
+ * audit_parent: lifetime is from audit_init_parent() to receipt of an IN_IGNORED
+ * 	event.  Each audit_watch holds a reference to its associated parent.
+ *
+ * audit_watch: if added to lists, lifetime is from audit_init_watch() to
+ * 	audit_remove_watch().  Additionally, an audit_watch may exist
+ * 	temporarily to assist in searching existing filter data.  Each
+ * 	audit_krule holds a reference to its associated watch.
+ */
+
+struct audit_watch {
+	atomic_t		count;	/* reference count */
+	char			*path;	/* insertion path */
+	dev_t			dev;	/* associated superblock device */
+	unsigned long		ino;	/* associated inode number */
+	struct audit_parent	*parent; /* associated parent */
+	struct list_head	wlist;	/* entry in parent->watches list */
+	struct list_head	rules;	/* associated rules */
+};
+
+struct audit_parent {
+	struct list_head	ilist;	/* entry in inotify registration list */
+	struct list_head	watches; /* associated watches */
+	struct inotify_watch	wdata;	/* inotify watch data */
+	unsigned		flags;	/* status flags */
+};
+
+/* Inotify handle. */
+struct inotify_handle *audit_ih;
+
+/*
+ * audit_parent status flags:
+ *
+ * AUDIT_PARENT_INVALID - set anytime rules/watches are auto-removed due to
+ * a filesystem event to ensure we're adding audit watches to a valid parent.
+ * Technically not needed for IN_DELETE_SELF or IN_UNMOUNT events, as we cannot
+ * receive them while we have nameidata, but must be used for IN_MOVE_SELF which
+ * we can receive while holding nameidata.
+ */
+#define AUDIT_PARENT_INVALID	0x001
+
+/* Inotify events we care about. */
+#define AUDIT_IN_WATCH IN_MOVE|IN_CREATE|IN_DELETE|IN_DELETE_SELF|IN_MOVE_SELF
+
+static void audit_free_parent(struct inotify_watch *i_watch)
+{
+	struct audit_parent *parent;
+
+	parent = container_of(i_watch, struct audit_parent, wdata);
+	WARN_ON(!list_empty(&parent->watches));
+	kfree(parent);
+}
+
+void audit_get_watch(struct audit_watch *watch)
+{
+	atomic_inc(&watch->count);
+}
+
+void audit_put_watch(struct audit_watch *watch)
+{
+	if (atomic_dec_and_test(&watch->count)) {
+		WARN_ON(watch->parent);
+		WARN_ON(!list_empty(&watch->rules));
+		kfree(watch->path);
+		kfree(watch);
+	}
+}
+
+void audit_remove_watch(struct audit_watch *watch)
+{
+	list_del(&watch->wlist);
+	put_inotify_watch(&watch->parent->wdata);
+	watch->parent = NULL;
+	audit_put_watch(watch); /* match initial get */
+}
+
+char *audit_watch_path(struct audit_watch *watch)
+{
+	return watch->path;
+}
+
+struct list_head *audit_watch_rules(struct audit_watch *watch)
+{
+	return &watch->rules;
+}
+
+unsigned long audit_watch_inode(struct audit_watch *watch)
+{
+	return watch->ino;
+}
+
+dev_t audit_watch_dev(struct audit_watch *watch)
+{
+	return watch->dev;
+}
+
+/* Initialize a parent watch entry. */
+static struct audit_parent *audit_init_parent(struct nameidata *ndp)
+{
+	struct audit_parent *parent;
+	s32 wd;
+
+	parent = kzalloc(sizeof(*parent), GFP_KERNEL);
+	if (unlikely(!parent))
+		return ERR_PTR(-ENOMEM);
+
+	INIT_LIST_HEAD(&parent->watches);
+	parent->flags = 0;
+
+	inotify_init_watch(&parent->wdata);
+	/* grab a ref so inotify watch hangs around until we take audit_filter_mutex */
+	get_inotify_watch(&parent->wdata);
+	wd = inotify_add_watch(audit_ih, &parent->wdata,
+			       ndp->path.dentry->d_inode, AUDIT_IN_WATCH);
+	if (wd < 0) {
+		audit_free_parent(&parent->wdata);
+		return ERR_PTR(wd);
+	}
+
+	return parent;
+}
+
+/* Initialize a watch entry. */
+static struct audit_watch *audit_init_watch(char *path)
+{
+	struct audit_watch *watch;
+
+	watch = kzalloc(sizeof(*watch), GFP_KERNEL);
+	if (unlikely(!watch))
+		return ERR_PTR(-ENOMEM);
+
+	INIT_LIST_HEAD(&watch->rules);
+	atomic_set(&watch->count, 1);
+	watch->path = path;
+	watch->dev = (dev_t)-1;
+	watch->ino = (unsigned long)-1;
+
+	return watch;
+}
+
+/* Translate a watch string to kernel respresentation. */
+int audit_to_watch(struct audit_krule *krule, char *path, int len, u32 op)
+{
+	struct audit_watch *watch;
+
+	if (!audit_ih)
+		return -EOPNOTSUPP;
+
+	if (path[0] != '/' || path[len-1] == '/' ||
+	    krule->listnr != AUDIT_FILTER_EXIT ||
+	    op != Audit_equal ||
+	    krule->inode_f || krule->watch || krule->tree)
+		return -EINVAL;
+
+	watch = audit_init_watch(path);
+	if (IS_ERR(watch))
+		return PTR_ERR(watch);
+
+	audit_get_watch(watch);
+	krule->watch = watch;
+
+	return 0;
+}
+
+/* Duplicate the given audit watch.  The new watch's rules list is initialized
+ * to an empty list and wlist is undefined. */
+static struct audit_watch *audit_dupe_watch(struct audit_watch *old)
+{
+	char *path;
+	struct audit_watch *new;
+
+	path = kstrdup(old->path, GFP_KERNEL);
+	if (unlikely(!path))
+		return ERR_PTR(-ENOMEM);
+
+	new = audit_init_watch(path);
+	if (IS_ERR(new)) {
+		kfree(path);
+		goto out;
+	}
+
+	new->dev = old->dev;
+	new->ino = old->ino;
+	get_inotify_watch(&old->parent->wdata);
+	new->parent = old->parent;
+
+out:
+	return new;
+}
+
+static void audit_watch_log_rule_change(struct audit_krule *r, struct audit_watch *w, char *op)
+{
+	if (audit_enabled) {
+		struct audit_buffer *ab;
+		ab = audit_log_start(NULL, GFP_NOFS, AUDIT_CONFIG_CHANGE);
+		audit_log_format(ab, "auid=%u ses=%u op=",
+				 audit_get_loginuid(current),
+				 audit_get_sessionid(current));
+		audit_log_string(ab, op);
+		audit_log_format(ab, " path=");
+		audit_log_untrustedstring(ab, w->path);
+		if (r->filterkey) {
+			audit_log_format(ab, " key=");
+			audit_log_untrustedstring(ab, r->filterkey);
+		} else
+			audit_log_format(ab, " key=(null)");
+		audit_log_format(ab, " list=%d res=1", r->listnr);
+		audit_log_end(ab);
+	}
+}
+
+/* Update inode info in audit rules based on filesystem event. */
+static void audit_update_watch(struct audit_parent *parent,
+			       const char *dname, dev_t dev,
+			       unsigned long ino, unsigned invalidating)
+{
+	struct audit_watch *owatch, *nwatch, *nextw;
+	struct audit_krule *r, *nextr;
+	struct audit_entry *oentry, *nentry;
+
+	mutex_lock(&audit_filter_mutex);
+	list_for_each_entry_safe(owatch, nextw, &parent->watches, wlist) {
+		if (audit_compare_dname_path(dname, owatch->path, NULL))
+			continue;
+
+		/* If the update involves invalidating rules, do the inode-based
+		 * filtering now, so we don't omit records. */
+		if (invalidating && current->audit_context)
+			audit_filter_inodes(current, current->audit_context);
+
+		nwatch = audit_dupe_watch(owatch);
+		if (IS_ERR(nwatch)) {
+			mutex_unlock(&audit_filter_mutex);
+			audit_panic("error updating watch, skipping");
+			return;
+		}
+		nwatch->dev = dev;
+		nwatch->ino = ino;
+
+		list_for_each_entry_safe(r, nextr, &owatch->rules, rlist) {
+
+			oentry = container_of(r, struct audit_entry, rule);
+			list_del(&oentry->rule.rlist);
+			list_del_rcu(&oentry->list);
+
+			nentry = audit_dupe_rule(&oentry->rule, nwatch);
+			if (IS_ERR(nentry)) {
+				list_del(&oentry->rule.list);
+				audit_panic("error updating watch, removing");
+			} else {
+				int h = audit_hash_ino((u32)ino);
+				list_add(&nentry->rule.rlist, &nwatch->rules);
+				list_add_rcu(&nentry->list, &audit_inode_hash[h]);
+				list_replace(&oentry->rule.list,
+					     &nentry->rule.list);
+			}
+
+			audit_watch_log_rule_change(r, owatch, "updated rules");
+
+			call_rcu(&oentry->rcu, audit_free_rule_rcu);
+		}
+
+		audit_remove_watch(owatch);
+		goto add_watch_to_parent; /* event applies to a single watch */
+	}
+	mutex_unlock(&audit_filter_mutex);
+	return;
+
+add_watch_to_parent:
+	list_add(&nwatch->wlist, &parent->watches);
+	mutex_unlock(&audit_filter_mutex);
+	return;
+}
+
+/* Remove all watches & rules associated with a parent that is going away. */
+static void audit_remove_parent_watches(struct audit_parent *parent)
+{
+	struct audit_watch *w, *nextw;
+	struct audit_krule *r, *nextr;
+	struct audit_entry *e;
+
+	mutex_lock(&audit_filter_mutex);
+	parent->flags |= AUDIT_PARENT_INVALID;
+	list_for_each_entry_safe(w, nextw, &parent->watches, wlist) {
+		list_for_each_entry_safe(r, nextr, &w->rules, rlist) {
+			e = container_of(r, struct audit_entry, rule);
+			audit_watch_log_rule_change(r, w, "remove rule");
+			list_del(&r->rlist);
+			list_del(&r->list);
+			list_del_rcu(&e->list);
+			call_rcu(&e->rcu, audit_free_rule_rcu);
+		}
+		audit_remove_watch(w);
+	}
+	mutex_unlock(&audit_filter_mutex);
+}
+
+/* Unregister inotify watches for parents on in_list.
+ * Generates an IN_IGNORED event. */
+void audit_inotify_unregister(struct list_head *in_list)
+{
+	struct audit_parent *p, *n;
+
+	list_for_each_entry_safe(p, n, in_list, ilist) {
+		list_del(&p->ilist);
+		inotify_rm_watch(audit_ih, &p->wdata);
+		/* the unpin matching the pin in audit_do_del_rule() */
+		unpin_inotify_watch(&p->wdata);
+	}
+}
+
+/* Get path information necessary for adding watches. */
+int audit_get_nd(char *path, struct nameidata **ndp, struct nameidata **ndw)
+{
+	struct nameidata *ndparent, *ndwatch;
+	int err;
+
+	ndparent = kmalloc(sizeof(*ndparent), GFP_KERNEL);
+	if (unlikely(!ndparent))
+		return -ENOMEM;
+
+	ndwatch = kmalloc(sizeof(*ndwatch), GFP_KERNEL);
+	if (unlikely(!ndwatch)) {
+		kfree(ndparent);
+		return -ENOMEM;
+	}
+
+	err = path_lookup(path, LOOKUP_PARENT, ndparent);
+	if (err) {
+		kfree(ndparent);
+		kfree(ndwatch);
+		return err;
+	}
+
+	err = path_lookup(path, 0, ndwatch);
+	if (err) {
+		kfree(ndwatch);
+		ndwatch = NULL;
+	}
+
+	*ndp = ndparent;
+	*ndw = ndwatch;
+
+	return 0;
+}
+
+/* Release resources used for watch path information. */
+void audit_put_nd(struct nameidata *ndp, struct nameidata *ndw)
+{
+	if (ndp) {
+		path_put(&ndp->path);
+		kfree(ndp);
+	}
+	if (ndw) {
+		path_put(&ndw->path);
+		kfree(ndw);
+	}
+}
+
+/* Associate the given rule with an existing parent inotify_watch.
+ * Caller must hold audit_filter_mutex. */
+static void audit_add_to_parent(struct audit_krule *krule,
+				struct audit_parent *parent)
+{
+	struct audit_watch *w, *watch = krule->watch;
+	int watch_found = 0;
+
+	list_for_each_entry(w, &parent->watches, wlist) {
+		if (strcmp(watch->path, w->path))
+			continue;
+
+		watch_found = 1;
+
+		/* put krule's and initial refs to temporary watch */
+		audit_put_watch(watch);
+		audit_put_watch(watch);
+
+		audit_get_watch(w);
+		krule->watch = watch = w;
+		break;
+	}
+
+	if (!watch_found) {
+		get_inotify_watch(&parent->wdata);
+		watch->parent = parent;
+
+		list_add(&watch->wlist, &parent->watches);
+	}
+	list_add(&krule->rlist, &watch->rules);
+}
+
+/* Find a matching watch entry, or add this one.
+ * Caller must hold audit_filter_mutex. */
+int audit_add_watch(struct audit_krule *krule, struct nameidata *ndp,
+		    struct nameidata *ndw)
+{
+	struct audit_watch *watch = krule->watch;
+	struct inotify_watch *i_watch;
+	struct audit_parent *parent;
+	int ret = 0;
+
+	/* update watch filter fields */
+	if (ndw) {
+		watch->dev = ndw->path.dentry->d_inode->i_sb->s_dev;
+		watch->ino = ndw->path.dentry->d_inode->i_ino;
+	}
+
+	/* The audit_filter_mutex must not be held during inotify calls because
+	 * we hold it during inotify event callback processing.  If an existing
+	 * inotify watch is found, inotify_find_watch() grabs a reference before
+	 * returning.
+	 */
+	mutex_unlock(&audit_filter_mutex);
+
+	if (inotify_find_watch(audit_ih, ndp->path.dentry->d_inode,
+			       &i_watch) < 0) {
+		parent = audit_init_parent(ndp);
+		if (IS_ERR(parent)) {
+			/* caller expects mutex locked */
+			mutex_lock(&audit_filter_mutex);
+			return PTR_ERR(parent);
+		}
+	} else
+		parent = container_of(i_watch, struct audit_parent, wdata);
+
+	mutex_lock(&audit_filter_mutex);
+
+	/* parent was moved before we took audit_filter_mutex */
+	if (parent->flags & AUDIT_PARENT_INVALID)
+		ret = -ENOENT;
+	else
+		audit_add_to_parent(krule, parent);
+
+	/* match get in audit_init_parent or inotify_find_watch */
+	put_inotify_watch(&parent->wdata);
+	return ret;
+}
+
+void audit_remove_watch_rule(struct audit_krule *krule, struct list_head *list)
+{
+	struct audit_watch *watch = krule->watch;
+	struct audit_parent *parent = watch->parent;
+
+	list_del(&krule->rlist);
+
+	if (list_empty(&watch->rules)) {
+		audit_remove_watch(watch);
+
+		if (list_empty(&parent->watches)) {
+			/* Put parent on the inotify un-registration
+			 * list.  Grab a reference before releasing
+			 * audit_filter_mutex, to be released in
+			 * audit_inotify_unregister().
+			 * If filesystem is going away, just leave
+			 * the sucker alone, eviction will take
+			 * care of it. */
+			if (pin_inotify_watch(&parent->wdata))
+				list_add(&parent->ilist, list);
+		}
+	}
+}
+
+/* Update watch data in audit rules based on inotify events. */
+static void audit_handle_ievent(struct inotify_watch *i_watch, u32 wd, u32 mask,
+			 u32 cookie, const char *dname, struct inode *inode)
+{
+	struct audit_parent *parent;
+
+	parent = container_of(i_watch, struct audit_parent, wdata);
+
+	if (mask & (IN_CREATE|IN_MOVED_TO) && inode)
+		audit_update_watch(parent, dname, inode->i_sb->s_dev,
+				   inode->i_ino, 0);
+	else if (mask & (IN_DELETE|IN_MOVED_FROM))
+		audit_update_watch(parent, dname, (dev_t)-1, (unsigned long)-1, 1);
+	/* inotify automatically removes the watch and sends IN_IGNORED */
+	else if (mask & (IN_DELETE_SELF|IN_UNMOUNT))
+		audit_remove_parent_watches(parent);
+	/* inotify does not remove the watch, so remove it manually */
+	else if(mask & IN_MOVE_SELF) {
+		audit_remove_parent_watches(parent);
+		inotify_remove_watch_locked(audit_ih, i_watch);
+	} else if (mask & IN_IGNORED)
+		put_inotify_watch(i_watch);
+}
+
+static const struct inotify_operations audit_inotify_ops = {
+	.handle_event   = audit_handle_ievent,
+	.destroy_watch  = audit_free_parent,
+};
+
+static int __init audit_watch_init(void)
+{
+	audit_ih = inotify_init(&audit_inotify_ops);
+	if (IS_ERR(audit_ih))
+		audit_panic("cannot initialize inotify handle");
+	return 0;
+}
+subsys_initcall(audit_watch_init);
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index e7466dd145c9..9d4c93437de6 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -27,7 +27,6 @@
 #include <linux/namei.h>
 #include <linux/netlink.h>
 #include <linux/sched.h>
-#include <linux/inotify.h>
 #include <linux/security.h>
 #include "audit.h"
 
@@ -44,36 +43,6 @@
  * 		be written directly provided audit_filter_mutex is held.
  */
 
-/*
- * Reference counting:
- *
- * audit_parent: lifetime is from audit_init_parent() to receipt of an IN_IGNORED
- * 	event.  Each audit_watch holds a reference to its associated parent.
- *
- * audit_watch: if added to lists, lifetime is from audit_init_watch() to
- * 	audit_remove_watch().  Additionally, an audit_watch may exist
- * 	temporarily to assist in searching existing filter data.  Each
- * 	audit_krule holds a reference to its associated watch.
- */
-
-struct audit_parent {
-	struct list_head	ilist;	/* entry in inotify registration list */
-	struct list_head	watches; /* associated watches */
-	struct inotify_watch	wdata;	/* inotify watch data */
-	unsigned		flags;	/* status flags */
-};
-
-/*
- * audit_parent status flags:
- *
- * AUDIT_PARENT_INVALID - set anytime rules/watches are auto-removed due to
- * a filesystem event to ensure we're adding audit watches to a valid parent.
- * Technically not needed for IN_DELETE_SELF or IN_UNMOUNT events, as we cannot
- * receive them while we have nameidata, but must be used for IN_MOVE_SELF which
- * we can receive while holding nameidata.
- */
-#define AUDIT_PARENT_INVALID	0x001
-
 /* Audit filter lists, defined in <linux/audit.h> */
 struct list_head audit_filter_list[AUDIT_NR_FILTERS] = {
 	LIST_HEAD_INIT(audit_filter_list[0]),
@@ -97,41 +66,6 @@ static struct list_head audit_rules_list[AUDIT_NR_FILTERS] = {
 
 DEFINE_MUTEX(audit_filter_mutex);
 
-/* Inotify events we care about. */
-#define AUDIT_IN_WATCH IN_MOVE|IN_CREATE|IN_DELETE|IN_DELETE_SELF|IN_MOVE_SELF
-
-void audit_free_parent(struct inotify_watch *i_watch)
-{
-	struct audit_parent *parent;
-
-	parent = container_of(i_watch, struct audit_parent, wdata);
-	WARN_ON(!list_empty(&parent->watches));
-	kfree(parent);
-}
-
-static inline void audit_get_watch(struct audit_watch *watch)
-{
-	atomic_inc(&watch->count);
-}
-
-static void audit_put_watch(struct audit_watch *watch)
-{
-	if (atomic_dec_and_test(&watch->count)) {
-		WARN_ON(watch->parent);
-		WARN_ON(!list_empty(&watch->rules));
-		kfree(watch->path);
-		kfree(watch);
-	}
-}
-
-static void audit_remove_watch(struct audit_watch *watch)
-{
-	list_del(&watch->wlist);
-	put_inotify_watch(&watch->parent->wdata);
-	watch->parent = NULL;
-	audit_put_watch(watch); /* match initial get */
-}
-
 static inline void audit_free_rule(struct audit_entry *e)
 {
 	int i;
@@ -156,50 +90,6 @@ void audit_free_rule_rcu(struct rcu_head *head)
 	audit_free_rule(e);
 }
 
-/* Initialize a parent watch entry. */
-static struct audit_parent *audit_init_parent(struct nameidata *ndp)
-{
-	struct audit_parent *parent;
-	s32 wd;
-
-	parent = kzalloc(sizeof(*parent), GFP_KERNEL);
-	if (unlikely(!parent))
-		return ERR_PTR(-ENOMEM);
-
-	INIT_LIST_HEAD(&parent->watches);
-	parent->flags = 0;
-
-	inotify_init_watch(&parent->wdata);
-	/* grab a ref so inotify watch hangs around until we take audit_filter_mutex */
-	get_inotify_watch(&parent->wdata);
-	wd = inotify_add_watch(audit_ih, &parent->wdata,
-			       ndp->path.dentry->d_inode, AUDIT_IN_WATCH);
-	if (wd < 0) {
-		audit_free_parent(&parent->wdata);
-		return ERR_PTR(wd);
-	}
-
-	return parent;
-}
-
-/* Initialize a watch entry. */
-static struct audit_watch *audit_init_watch(char *path)
-{
-	struct audit_watch *watch;
-
-	watch = kzalloc(sizeof(*watch), GFP_KERNEL);
-	if (unlikely(!watch))
-		return ERR_PTR(-ENOMEM);
-
-	INIT_LIST_HEAD(&watch->rules);
-	atomic_set(&watch->count, 1);
-	watch->path = path;
-	watch->dev = (dev_t)-1;
-	watch->ino = (unsigned long)-1;
-
-	return watch;
-}
-
 /* Initialize an audit filterlist entry. */
 static inline struct audit_entry *audit_init_entry(u32 field_count)
 {
@@ -260,31 +150,6 @@ static inline int audit_to_inode(struct audit_krule *krule,
 	return 0;
 }
 
-/* Translate a watch string to kernel respresentation. */
-static int audit_to_watch(struct audit_krule *krule, char *path, int len,
-			  u32 op)
-{
-	struct audit_watch *watch;
-
-	if (!audit_ih)
-		return -EOPNOTSUPP;
-
-	if (path[0] != '/' || path[len-1] == '/' ||
-	    krule->listnr != AUDIT_FILTER_EXIT ||
-	    op != Audit_equal ||
-	    krule->inode_f || krule->watch || krule->tree)
-		return -EINVAL;
-
-	watch = audit_init_watch(path);
-	if (IS_ERR(watch))
-		return PTR_ERR(watch);
-
-	audit_get_watch(watch);
-	krule->watch = watch;
-
-	return 0;
-}
-
 static __u32 *classes[AUDIT_SYSCALL_CLASSES];
 
 int __init audit_register_class(int class, unsigned *list)
@@ -766,7 +631,8 @@ static struct audit_rule_data *audit_krule_to_data(struct audit_krule *krule)
 			break;
 		case AUDIT_WATCH:
 			data->buflen += data->values[i] =
-				audit_pack_string(&bufp, krule->watch->path);
+				audit_pack_string(&bufp,
+						  audit_watch_path(krule->watch));
 			break;
 		case AUDIT_DIR:
 			data->buflen += data->values[i] =
@@ -818,7 +684,8 @@ static int audit_compare_rule(struct audit_krule *a, struct audit_krule *b)
 				return 1;
 			break;
 		case AUDIT_WATCH:
-			if (strcmp(a->watch->path, b->watch->path))
+			if (strcmp(audit_watch_path(a->watch),
+				   audit_watch_path(b->watch)))
 				return 1;
 			break;
 		case AUDIT_DIR:
@@ -844,32 +711,6 @@ static int audit_compare_rule(struct audit_krule *a, struct audit_krule *b)
 	return 0;
 }
 
-/* Duplicate the given audit watch.  The new watch's rules list is initialized
- * to an empty list and wlist is undefined. */
-static struct audit_watch *audit_dupe_watch(struct audit_watch *old)
-{
-	char *path;
-	struct audit_watch *new;
-
-	path = kstrdup(old->path, GFP_KERNEL);
-	if (unlikely(!path))
-		return ERR_PTR(-ENOMEM);
-
-	new = audit_init_watch(path);
-	if (IS_ERR(new)) {
-		kfree(path);
-		goto out;
-	}
-
-	new->dev = old->dev;
-	new->ino = old->ino;
-	get_inotify_watch(&old->parent->wdata);
-	new->parent = old->parent;
-
-out:
-	return new;
-}
-
 /* Duplicate LSM field information.  The lsm_rule is opaque, so must be
  * re-initialized. */
 static inline int audit_dupe_lsm_field(struct audit_field *df,
@@ -904,8 +745,8 @@ static inline int audit_dupe_lsm_field(struct audit_field *df,
  * rule with the new rule in the filterlist, then free the old rule.
  * The rlist element is undefined; list manipulations are handled apart from
  * the initial copy. */
-static struct audit_entry *audit_dupe_rule(struct audit_krule *old,
-					   struct audit_watch *watch)
+struct audit_entry *audit_dupe_rule(struct audit_krule *old,
+				    struct audit_watch *watch)
 {
 	u32 fcount = old->field_count;
 	struct audit_entry *entry;
@@ -977,127 +818,6 @@ static struct audit_entry *audit_dupe_rule(struct audit_krule *old,
 	return entry;
 }
 
-static void audit_watch_log_rule_change(struct audit_krule *r, struct audit_watch *w, char *op)
-{
-	if (audit_enabled) {
-		struct audit_buffer *ab;
-		ab = audit_log_start(NULL, GFP_NOFS, AUDIT_CONFIG_CHANGE);
-		audit_log_format(ab, "auid=%u ses=%u op=",
-				 audit_get_loginuid(current),
-				 audit_get_sessionid(current));
-		audit_log_string(ab, op);
-		audit_log_format(ab, " path=");
-		audit_log_untrustedstring(ab, w->path);
-		if (r->filterkey) {
-			audit_log_format(ab, " key=");
-			audit_log_untrustedstring(ab, r->filterkey);
-		} else
-			audit_log_format(ab, " key=(null)");
-		audit_log_format(ab, " list=%d res=1", r->listnr);
-		audit_log_end(ab);
-	}
-}
-
-/* Update inode info in audit rules based on filesystem event. */
-static void audit_update_watch(struct audit_parent *parent,
-			       const char *dname, dev_t dev,
-			       unsigned long ino, unsigned invalidating)
-{
-	struct audit_watch *owatch, *nwatch, *nextw;
-	struct audit_krule *r, *nextr;
-	struct audit_entry *oentry, *nentry;
-
-	mutex_lock(&audit_filter_mutex);
-	list_for_each_entry_safe(owatch, nextw, &parent->watches, wlist) {
-		if (audit_compare_dname_path(dname, owatch->path, NULL))
-			continue;
-
-		/* If the update involves invalidating rules, do the inode-based
-		 * filtering now, so we don't omit records. */
-		if (invalidating && current->audit_context)
-			audit_filter_inodes(current, current->audit_context);
-
-		nwatch = audit_dupe_watch(owatch);
-		if (IS_ERR(nwatch)) {
-			mutex_unlock(&audit_filter_mutex);
-			audit_panic("error updating watch, skipping");
-			return;
-		}
-		nwatch->dev = dev;
-		nwatch->ino = ino;
-
-		list_for_each_entry_safe(r, nextr, &owatch->rules, rlist) {
-
-			oentry = container_of(r, struct audit_entry, rule);
-			list_del(&oentry->rule.rlist);
-			list_del_rcu(&oentry->list);
-
-			nentry = audit_dupe_rule(&oentry->rule, nwatch);
-			if (IS_ERR(nentry)) {
-				list_del(&oentry->rule.list);
-				audit_panic("error updating watch, removing");
-			} else {
-				int h = audit_hash_ino((u32)ino);
-				list_add(&nentry->rule.rlist, &nwatch->rules);
-				list_add_rcu(&nentry->list, &audit_inode_hash[h]);
-				list_replace(&oentry->rule.list,
-					     &nentry->rule.list);
-			}
-
-			audit_watch_log_rule_change(r, owatch, "updated rules");
-
-			call_rcu(&oentry->rcu, audit_free_rule_rcu);
-		}
-
-		audit_remove_watch(owatch);
-		goto add_watch_to_parent; /* event applies to a single watch */
-	}
-	mutex_unlock(&audit_filter_mutex);
-	return;
-
-add_watch_to_parent:
-	list_add(&nwatch->wlist, &parent->watches);
-	mutex_unlock(&audit_filter_mutex);
-	return;
-}
-
-/* Remove all watches & rules associated with a parent that is going away. */
-static void audit_remove_parent_watches(struct audit_parent *parent)
-{
-	struct audit_watch *w, *nextw;
-	struct audit_krule *r, *nextr;
-	struct audit_entry *e;
-
-	mutex_lock(&audit_filter_mutex);
-	parent->flags |= AUDIT_PARENT_INVALID;
-	list_for_each_entry_safe(w, nextw, &parent->watches, wlist) {
-		list_for_each_entry_safe(r, nextr, &w->rules, rlist) {
-			e = container_of(r, struct audit_entry, rule);
-			audit_watch_log_rule_change(r, w, "remove rule");
-			list_del(&r->rlist);
-			list_del(&r->list);
-			list_del_rcu(&e->list);
-			call_rcu(&e->rcu, audit_free_rule_rcu);
-		}
-		audit_remove_watch(w);
-	}
-	mutex_unlock(&audit_filter_mutex);
-}
-
-/* Unregister inotify watches for parents on in_list.
- * Generates an IN_IGNORED event. */
-static void audit_inotify_unregister(struct list_head *in_list)
-{
-	struct audit_parent *p, *n;
-
-	list_for_each_entry_safe(p, n, in_list, ilist) {
-		list_del(&p->ilist);
-		inotify_rm_watch(audit_ih, &p->wdata);
-		/* the unpin matching the pin in audit_do_del_rule() */
-		unpin_inotify_watch(&p->wdata);
-	}
-}
-
 /* Find an existing audit rule.
  * Caller must hold audit_filter_mutex to prevent stale rule data. */
 static struct audit_entry *audit_find_rule(struct audit_entry *entry,
@@ -1135,134 +855,6 @@ out:
 	return found;
 }
 
-/* Get path information necessary for adding watches. */
-static int audit_get_nd(char *path, struct nameidata **ndp,
-			struct nameidata **ndw)
-{
-	struct nameidata *ndparent, *ndwatch;
-	int err;
-
-	ndparent = kmalloc(sizeof(*ndparent), GFP_KERNEL);
-	if (unlikely(!ndparent))
-		return -ENOMEM;
-
-	ndwatch = kmalloc(sizeof(*ndwatch), GFP_KERNEL);
-	if (unlikely(!ndwatch)) {
-		kfree(ndparent);
-		return -ENOMEM;
-	}
-
-	err = path_lookup(path, LOOKUP_PARENT, ndparent);
-	if (err) {
-		kfree(ndparent);
-		kfree(ndwatch);
-		return err;
-	}
-
-	err = path_lookup(path, 0, ndwatch);
-	if (err) {
-		kfree(ndwatch);
-		ndwatch = NULL;
-	}
-
-	*ndp = ndparent;
-	*ndw = ndwatch;
-
-	return 0;
-}
-
-/* Release resources used for watch path information. */
-static void audit_put_nd(struct nameidata *ndp, struct nameidata *ndw)
-{
-	if (ndp) {
-		path_put(&ndp->path);
-		kfree(ndp);
-	}
-	if (ndw) {
-		path_put(&ndw->path);
-		kfree(ndw);
-	}
-}
-
-/* Associate the given rule with an existing parent inotify_watch.
- * Caller must hold audit_filter_mutex. */
-static void audit_add_to_parent(struct audit_krule *krule,
-				struct audit_parent *parent)
-{
-	struct audit_watch *w, *watch = krule->watch;
-	int watch_found = 0;
-
-	list_for_each_entry(w, &parent->watches, wlist) {
-		if (strcmp(watch->path, w->path))
-			continue;
-
-		watch_found = 1;
-
-		/* put krule's and initial refs to temporary watch */
-		audit_put_watch(watch);
-		audit_put_watch(watch);
-
-		audit_get_watch(w);
-		krule->watch = watch = w;
-		break;
-	}
-
-	if (!watch_found) {
-		get_inotify_watch(&parent->wdata);
-		watch->parent = parent;
-
-		list_add(&watch->wlist, &parent->watches);
-	}
-	list_add(&krule->rlist, &watch->rules);
-}
-
-/* Find a matching watch entry, or add this one.
- * Caller must hold audit_filter_mutex. */
-static int audit_add_watch(struct audit_krule *krule, struct nameidata *ndp,
-			   struct nameidata *ndw)
-{
-	struct audit_watch *watch = krule->watch;
-	struct inotify_watch *i_watch;
-	struct audit_parent *parent;
-	int ret = 0;
-
-	/* update watch filter fields */
-	if (ndw) {
-		watch->dev = ndw->path.dentry->d_inode->i_sb->s_dev;
-		watch->ino = ndw->path.dentry->d_inode->i_ino;
-	}
-
-	/* The audit_filter_mutex must not be held during inotify calls because
-	 * we hold it during inotify event callback processing.  If an existing
-	 * inotify watch is found, inotify_find_watch() grabs a reference before
-	 * returning.
-	 */
-	mutex_unlock(&audit_filter_mutex);
-
-	if (inotify_find_watch(audit_ih, ndp->path.dentry->d_inode,
-			       &i_watch) < 0) {
-		parent = audit_init_parent(ndp);
-		if (IS_ERR(parent)) {
-			/* caller expects mutex locked */
-			mutex_lock(&audit_filter_mutex);
-			return PTR_ERR(parent);
-		}
-	} else
-		parent = container_of(i_watch, struct audit_parent, wdata);
-
-	mutex_lock(&audit_filter_mutex);
-
-	/* parent was moved before we took audit_filter_mutex */
-	if (parent->flags & AUDIT_PARENT_INVALID)
-		ret = -ENOENT;
-	else
-		audit_add_to_parent(krule, parent);
-
-	/* match get in audit_init_parent or inotify_find_watch */
-	put_inotify_watch(&parent->wdata);
-	return ret;
-}
-
 static u64 prio_low = ~0ULL/2;
 static u64 prio_high = ~0ULL/2 - 1;
 
@@ -1297,7 +889,7 @@ static inline int audit_add_rule(struct audit_entry *entry)
 
 	/* Avoid calling path_lookup under audit_filter_mutex. */
 	if (watch) {
-		err = audit_get_nd(watch->path, &ndp, &ndw);
+		err = audit_get_nd(audit_watch_path(watch), &ndp, &ndw);
 		if (err)
 			goto error;
 	}
@@ -1312,7 +904,7 @@ static inline int audit_add_rule(struct audit_entry *entry)
 		}
 		/* entry->rule.watch may have changed during audit_add_watch() */
 		watch = entry->rule.watch;
-		h = audit_hash_ino((u32)watch->ino);
+		h = audit_hash_ino((u32)audit_watch_inode(watch));
 		list = &audit_inode_hash[h];
 	}
 	if (tree) {
@@ -1364,7 +956,7 @@ error:
 static inline int audit_del_rule(struct audit_entry *entry)
 {
 	struct audit_entry  *e;
-	struct audit_watch *watch, *tmp_watch = entry->rule.watch;
+	struct audit_watch *watch = entry->rule.watch;
 	struct audit_tree *tree = entry->rule.tree;
 	struct list_head *list;
 	LIST_HEAD(inotify_list);
@@ -1386,29 +978,8 @@ static inline int audit_del_rule(struct audit_entry *entry)
 		goto out;
 	}
 
-	watch = e->rule.watch;
-	if (watch) {
-		struct audit_parent *parent = watch->parent;
-
-		list_del(&e->rule.rlist);
-
-		if (list_empty(&watch->rules)) {
-			audit_remove_watch(watch);
-
-			if (list_empty(&parent->watches)) {
-				/* Put parent on the inotify un-registration
-				 * list.  Grab a reference before releasing
-				 * audit_filter_mutex, to be released in
-				 * audit_inotify_unregister().
-				 * If filesystem is going away, just leave
-				 * the sucker alone, eviction will take
-				 * care of it.
-				 */
-				if (pin_inotify_watch(&parent->wdata))
-					list_add(&parent->ilist, &inotify_list);
-			}
-		}
-	}
+	if (e->rule.watch)
+		audit_remove_watch_rule(&e->rule, &inotify_list);
 
 	if (e->rule.tree)
 		audit_remove_tree_rule(&e->rule);
@@ -1430,8 +1001,8 @@ static inline int audit_del_rule(struct audit_entry *entry)
 		audit_inotify_unregister(&inotify_list);
 
 out:
-	if (tmp_watch)
-		audit_put_watch(tmp_watch); /* match initial get */
+	if (watch)
+		audit_put_watch(watch); /* match initial get */
 	if (tree)
 		audit_put_tree(tree);	/* that's the temporary one */
 
@@ -1785,7 +1356,7 @@ static int update_lsm_rule(struct audit_krule *r)
 		list_del(&r->list);
 	} else {
 		if (watch) {
-			list_add(&nentry->rule.rlist, &watch->rules);
+			list_add(&nentry->rule.rlist, audit_watch_rules(watch));
 			list_del(&r->rlist);
 		} else if (tree)
 			list_replace_init(&r->rlist, &nentry->rule.rlist);
@@ -1821,27 +1392,3 @@ int audit_update_lsm_rules(void)
 
 	return err;
 }
-
-/* Update watch data in audit rules based on inotify events. */
-void audit_handle_ievent(struct inotify_watch *i_watch, u32 wd, u32 mask,
-			 u32 cookie, const char *dname, struct inode *inode)
-{
-	struct audit_parent *parent;
-
-	parent = container_of(i_watch, struct audit_parent, wdata);
-
-	if (mask & (IN_CREATE|IN_MOVED_TO) && inode)
-		audit_update_watch(parent, dname, inode->i_sb->s_dev,
-				   inode->i_ino, 0);
-	else if (mask & (IN_DELETE|IN_MOVED_FROM))
-		audit_update_watch(parent, dname, (dev_t)-1, (unsigned long)-1, 1);
-	/* inotify automatically removes the watch and sends IN_IGNORED */
-	else if (mask & (IN_DELETE_SELF|IN_UNMOUNT))
-		audit_remove_parent_watches(parent);
-	/* inotify does not remove the watch, so remove it manually */
-	else if(mask & IN_MOVE_SELF) {
-		audit_remove_parent_watches(parent);
-		inotify_remove_watch_locked(audit_ih, i_watch);
-	} else if (mask & IN_IGNORED)
-		put_inotify_watch(i_watch);
-}
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index b14d234b85f3..0b862cac6ca2 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -548,9 +548,9 @@ static int audit_filter_rules(struct task_struct *tsk,
 			}
 			break;
 		case AUDIT_WATCH:
-			if (name && rule->watch->ino != (unsigned long)-1)
-				result = (name->dev == rule->watch->dev &&
-					  name->ino == rule->watch->ino);
+			if (name && audit_watch_inode(rule->watch) != (unsigned long)-1)
+				result = (name->dev == audit_watch_dev(rule->watch) &&
+					  name->ino == audit_watch_inode(rule->watch));
 			break;
 		case AUDIT_DIR:
 			if (ctx)
-- 
cgit v1.2.3-59-g8ed1b


From 35fe4d0b1b12286a81938e9c5fdfaf639ac0ce5b Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Thu, 11 Jun 2009 14:31:36 -0400
Subject: Audit: move audit_get_nd completely into audit_watch

audit_get_nd() is only used  by audit_watch and could be more cleanly
implemented by having the audit watch functions call it when needed rather
than making the generic audit rule parsing code deal with those objects.

Signed-off-by: Eric Paris <eparis@redhat.com>
---
 kernel/audit.h       |  5 +----
 kernel/audit_watch.c | 27 ++++++++++++++++++++-------
 kernel/auditfilter.c | 15 ++-------------
 3 files changed, 23 insertions(+), 24 deletions(-)

diff --git a/kernel/audit.h b/kernel/audit.h
index 704d5b01d9fd..bb1c0d69db08 100644
--- a/kernel/audit.h
+++ b/kernel/audit.h
@@ -109,10 +109,7 @@ extern dev_t audit_watch_dev(struct audit_watch *watch);
 extern void audit_put_watch(struct audit_watch *watch);
 extern void audit_get_watch(struct audit_watch *watch);
 extern int audit_to_watch(struct audit_krule *krule, char *path, int len, u32 op);
-extern int audit_get_nd(char *path, struct nameidata **ndp, struct nameidata **ndw);
-extern void audit_put_nd(struct nameidata *ndp, struct nameidata *ndw);
-extern int audit_add_watch(struct audit_krule *krule, struct nameidata *ndp,
-			   struct nameidata *ndw);
+extern int audit_add_watch(struct audit_krule *krule);
 extern void audit_remove_watch(struct audit_watch *watch);
 extern void audit_remove_watch_rule(struct audit_krule *krule, struct list_head *list);
 extern void audit_inotify_unregister(struct list_head *in_list);
diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c
index da8be6d39c1a..b49ab019fdff 100644
--- a/kernel/audit_watch.c
+++ b/kernel/audit_watch.c
@@ -345,7 +345,7 @@ void audit_inotify_unregister(struct list_head *in_list)
 }
 
 /* Get path information necessary for adding watches. */
-int audit_get_nd(char *path, struct nameidata **ndp, struct nameidata **ndw)
+static int audit_get_nd(char *path, struct nameidata **ndp, struct nameidata **ndw)
 {
 	struct nameidata *ndparent, *ndwatch;
 	int err;
@@ -380,7 +380,7 @@ int audit_get_nd(char *path, struct nameidata **ndp, struct nameidata **ndw)
 }
 
 /* Release resources used for watch path information. */
-void audit_put_nd(struct nameidata *ndp, struct nameidata *ndw)
+static void audit_put_nd(struct nameidata *ndp, struct nameidata *ndw)
 {
 	if (ndp) {
 		path_put(&ndp->path);
@@ -426,14 +426,24 @@ static void audit_add_to_parent(struct audit_krule *krule,
 
 /* Find a matching watch entry, or add this one.
  * Caller must hold audit_filter_mutex. */
-int audit_add_watch(struct audit_krule *krule, struct nameidata *ndp,
-		    struct nameidata *ndw)
+int audit_add_watch(struct audit_krule *krule)
 {
 	struct audit_watch *watch = krule->watch;
 	struct inotify_watch *i_watch;
 	struct audit_parent *parent;
+	struct nameidata *ndp = NULL, *ndw = NULL;
 	int ret = 0;
 
+	mutex_unlock(&audit_filter_mutex);
+
+	/* Avoid calling path_lookup under audit_filter_mutex. */
+	ret = audit_get_nd(watch->path, &ndp, &ndw);
+	if (ret) {
+		/* caller expects mutex locked */
+		mutex_lock(&audit_filter_mutex);
+		goto error;
+	}
+
 	/* update watch filter fields */
 	if (ndw) {
 		watch->dev = ndw->path.dentry->d_inode->i_sb->s_dev;
@@ -445,15 +455,14 @@ int audit_add_watch(struct audit_krule *krule, struct nameidata *ndp,
 	 * inotify watch is found, inotify_find_watch() grabs a reference before
 	 * returning.
 	 */
-	mutex_unlock(&audit_filter_mutex);
-
 	if (inotify_find_watch(audit_ih, ndp->path.dentry->d_inode,
 			       &i_watch) < 0) {
 		parent = audit_init_parent(ndp);
 		if (IS_ERR(parent)) {
 			/* caller expects mutex locked */
 			mutex_lock(&audit_filter_mutex);
-			return PTR_ERR(parent);
+			ret = PTR_ERR(parent);
+			goto error;
 		}
 	} else
 		parent = container_of(i_watch, struct audit_parent, wdata);
@@ -468,7 +477,11 @@ int audit_add_watch(struct audit_krule *krule, struct nameidata *ndp,
 
 	/* match get in audit_init_parent or inotify_find_watch */
 	put_inotify_watch(&parent->wdata);
+
+error:
+	audit_put_nd(ndp, ndw);		/* NULL args OK */
 	return ret;
+
 }
 
 void audit_remove_watch_rule(struct audit_krule *krule, struct list_head *list)
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index 9d4c93437de6..21b623595aad 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -864,7 +864,6 @@ static inline int audit_add_rule(struct audit_entry *entry)
 	struct audit_entry *e;
 	struct audit_watch *watch = entry->rule.watch;
 	struct audit_tree *tree = entry->rule.tree;
-	struct nameidata *ndp = NULL, *ndw = NULL;
 	struct list_head *list;
 	int h, err;
 #ifdef CONFIG_AUDITSYSCALL
@@ -878,8 +877,8 @@ static inline int audit_add_rule(struct audit_entry *entry)
 
 	mutex_lock(&audit_filter_mutex);
 	e = audit_find_rule(entry, &list);
-	mutex_unlock(&audit_filter_mutex);
 	if (e) {
+		mutex_unlock(&audit_filter_mutex);
 		err = -EEXIST;
 		/* normally audit_add_tree_rule() will free it on failure */
 		if (tree)
@@ -887,17 +886,9 @@ static inline int audit_add_rule(struct audit_entry *entry)
 		goto error;
 	}
 
-	/* Avoid calling path_lookup under audit_filter_mutex. */
-	if (watch) {
-		err = audit_get_nd(audit_watch_path(watch), &ndp, &ndw);
-		if (err)
-			goto error;
-	}
-
-	mutex_lock(&audit_filter_mutex);
 	if (watch) {
 		/* audit_filter_mutex is dropped and re-taken during this call */
-		err = audit_add_watch(&entry->rule, ndp, ndw);
+		err = audit_add_watch(&entry->rule);
 		if (err) {
 			mutex_unlock(&audit_filter_mutex);
 			goto error;
@@ -942,11 +933,9 @@ static inline int audit_add_rule(struct audit_entry *entry)
 #endif
 	mutex_unlock(&audit_filter_mutex);
 
-	audit_put_nd(ndp, ndw);		/* NULL args OK */
  	return 0;
 
 error:
-	audit_put_nd(ndp, ndw);		/* NULL args OK */
 	if (watch)
 		audit_put_watch(watch); /* tmp watch, matches initial get */
 	return err;
-- 
cgit v1.2.3-59-g8ed1b


From 9d9609851003ebed15957f0f2ce18492739ee124 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Thu, 11 Jun 2009 14:31:37 -0400
Subject: Audit: clean up all op= output to include string quoting

A number of places in the audit system we send an op= followed by a string
that includes spaces.  Somehow this works but it's just wrong.  This patch
moves all of those that I could find to be quoted.

Example:

Change From: type=CONFIG_CHANGE msg=audit(1244666690.117:31): auid=0 ses=1
subj=unconfined_u:unconfined_r:auditctl_t:s0-s0:c0.c1023 op=remove rule
key="number2" list=4 res=0

Change To: type=CONFIG_CHANGE msg=audit(1244666690.117:31): auid=0 ses=1
subj=unconfined_u:unconfined_r:auditctl_t:s0-s0:c0.c1023 op="remove rule"
key="number2" list=4 res=0

Signed-off-by: Eric Paris <eparis@redhat.com>
---
 include/linux/audit.h |  3 +++
 kernel/audit.c        |  9 +++++++++
 kernel/audit_tree.c   | 10 ++++------
 kernel/audit_watch.c  |  6 +-----
 kernel/auditfilter.c  | 12 +++++-------
 kernel/auditsc.c      |  8 ++------
 6 files changed, 24 insertions(+), 24 deletions(-)

diff --git a/include/linux/audit.h b/include/linux/audit.h
index 4fa2810b675e..3c7a358241a7 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -599,6 +599,8 @@ extern void		    audit_log_untrustedstring(struct audit_buffer *ab,
 extern void		    audit_log_d_path(struct audit_buffer *ab,
 					     const char *prefix,
 					     struct path *path);
+extern void		    audit_log_key(struct audit_buffer *ab,
+					  char *key);
 extern void		    audit_log_lost(const char *message);
 extern int		    audit_update_lsm_rules(void);
 
@@ -621,6 +623,7 @@ extern int audit_enabled;
 #define audit_log_n_untrustedstring(a,n,s) do { ; } while (0)
 #define audit_log_untrustedstring(a,s) do { ; } while (0)
 #define audit_log_d_path(b, p, d) do { ; } while (0)
+#define audit_log_key(b, k) do { ; } while (0)
 #define audit_enabled 0
 #endif
 #endif
diff --git a/kernel/audit.c b/kernel/audit.c
index e07ad2340dbe..6194c50e2039 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -1450,6 +1450,15 @@ void audit_log_d_path(struct audit_buffer *ab, const char *prefix,
 	kfree(pathname);
 }
 
+void audit_log_key(struct audit_buffer *ab, char *key)
+{
+	audit_log_format(ab, " key=");
+	if (key)
+		audit_log_untrustedstring(ab, key);
+	else
+		audit_log_format(ab, "(null)");
+}
+
 /**
  * audit_log_end - end one audit record
  * @ab: the audit_buffer
diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c
index 1f6396d76687..3ff0731284a1 100644
--- a/kernel/audit_tree.c
+++ b/kernel/audit_tree.c
@@ -441,13 +441,11 @@ static void kill_rules(struct audit_tree *tree)
 		if (rule->tree) {
 			/* not a half-baked one */
 			ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE);
-			audit_log_format(ab, "op=remove rule dir=");
+			audit_log_format(ab, "op=");
+			audit_log_string(ab, "remove rule");
+			audit_log_format(ab, " dir=");
 			audit_log_untrustedstring(ab, rule->tree->pathname);
-			if (rule->filterkey) {
-				audit_log_format(ab, " key=");
-				audit_log_untrustedstring(ab, rule->filterkey);
-			} else
-				audit_log_format(ab, " key=(null)");
+			audit_log_key(ab, rule->filterkey);
 			audit_log_format(ab, " list=%d res=1", rule->listnr);
 			audit_log_end(ab);
 			rule->tree = NULL;
diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c
index b49ab019fdff..0e96dbc60ea9 100644
--- a/kernel/audit_watch.c
+++ b/kernel/audit_watch.c
@@ -234,11 +234,7 @@ static void audit_watch_log_rule_change(struct audit_krule *r, struct audit_watc
 		audit_log_string(ab, op);
 		audit_log_format(ab, " path=");
 		audit_log_untrustedstring(ab, w->path);
-		if (r->filterkey) {
-			audit_log_format(ab, " key=");
-			audit_log_untrustedstring(ab, r->filterkey);
-		} else
-			audit_log_format(ab, " key=(null)");
+		audit_log_key(ab, r->filterkey);
 		audit_log_format(ab, " list=%d res=1", r->listnr);
 		audit_log_end(ab);
 	}
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index 21b623595aad..a70604047f3c 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -1079,11 +1079,9 @@ static void audit_log_rule_change(uid_t loginuid, u32 sessionid, u32 sid,
 			security_release_secctx(ctx, len);
 		}
 	}
-	audit_log_format(ab, " op=%s rule key=", action);
-	if (rule->filterkey)
-		audit_log_untrustedstring(ab, rule->filterkey);
-	else
-		audit_log_format(ab, "(null)");
+	audit_log_format(ab, " op=");
+	audit_log_string(ab, action);
+	audit_log_key(ab, rule->filterkey);
 	audit_log_format(ab, " list=%d res=%d", rule->listnr, res);
 	audit_log_end(ab);
 }
@@ -1147,7 +1145,7 @@ int audit_receive_filter(int type, int pid, int uid, int seq, void *data,
 			return PTR_ERR(entry);
 
 		err = audit_add_rule(entry);
-		audit_log_rule_change(loginuid, sessionid, sid, "add",
+		audit_log_rule_change(loginuid, sessionid, sid, "add rule",
 				      &entry->rule, !err);
 
 		if (err)
@@ -1163,7 +1161,7 @@ int audit_receive_filter(int type, int pid, int uid, int seq, void *data,
 			return PTR_ERR(entry);
 
 		err = audit_del_rule(entry);
-		audit_log_rule_change(loginuid, sessionid, sid, "remove",
+		audit_log_rule_change(loginuid, sessionid, sid, "remove rule",
 				      &entry->rule, !err);
 
 		audit_free_rule(entry);
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 0b862cac6ca2..2de95d1582bc 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -1137,7 +1137,7 @@ static int audit_log_single_execve_arg(struct audit_context *context,
 		if (has_cntl)
 			audit_log_n_hex(*ab, buf, to_send);
 		else
-			audit_log_format(*ab, "\"%s\"", buf);
+			audit_log_string(*ab, buf);
 
 		p += to_send;
 		len_left -= to_send;
@@ -1372,11 +1372,7 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 
 
 	audit_log_task_info(ab, tsk);
-	if (context->filterkey) {
-		audit_log_format(ab, " key=");
-		audit_log_untrustedstring(ab, context->filterkey);
-	} else
-		audit_log_format(ab, " key=(null)");
+	audit_log_key(ab, context->filterkey);
 	audit_log_end(ab);
 
 	for (aux = context->aux; aux; aux = aux->next) {
-- 
cgit v1.2.3-59-g8ed1b


From 916d75761c971b6e630a26bd4ba472e90ac9a4b9 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 24 Jun 2009 00:02:38 -0400
Subject: Fix rule eviction order for AUDIT_DIR

If syscall removes the root of subtree being watched, we
definitely do not want the rules refering that subtree
to be destroyed without the syscall in question having
a chance to match them.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 kernel/audit.c      | 17 +---------------
 kernel/audit.h      |  7 +++++--
 kernel/audit_tree.c | 56 ++++++++++++++++++++++++++++++++++++++++++++++++-----
 kernel/auditsc.c    | 15 ++++++++++++++
 4 files changed, 72 insertions(+), 23 deletions(-)

diff --git a/kernel/audit.c b/kernel/audit.c
index 6194c50e2039..defc2e6f1e3b 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -133,7 +133,7 @@ static DECLARE_WAIT_QUEUE_HEAD(kauditd_wait);
 static DECLARE_WAIT_QUEUE_HEAD(audit_backlog_wait);
 
 /* Serialize requests from userspace. */
-static DEFINE_MUTEX(audit_cmd_mutex);
+DEFINE_MUTEX(audit_cmd_mutex);
 
 /* AUDIT_BUFSIZ is the size of the temporary buffer used for formatting
  * audit records.  Since printk uses a 1024 byte buffer, this buffer
@@ -505,21 +505,6 @@ int audit_send_list(void *_dest)
 	return 0;
 }
 
-#ifdef CONFIG_AUDIT_TREE
-static int prune_tree_thread(void *unused)
-{
-	mutex_lock(&audit_cmd_mutex);
-	audit_prune_trees();
-	mutex_unlock(&audit_cmd_mutex);
-	return 0;
-}
-
-void audit_schedule_prune(void)
-{
-	kthread_run(prune_tree_thread, NULL, "audit_prune_tree");
-}
-#endif
-
 struct sk_buff *audit_make_reply(int pid, int seq, int type, int done,
 				 int multi, void *payload, int size)
 {
diff --git a/kernel/audit.h b/kernel/audit.h
index bb1c0d69db08..208687be4f30 100644
--- a/kernel/audit.h
+++ b/kernel/audit.h
@@ -128,10 +128,9 @@ extern int audit_add_tree_rule(struct audit_krule *);
 extern int audit_remove_tree_rule(struct audit_krule *);
 extern void audit_trim_trees(void);
 extern int audit_tag_tree(char *old, char *new);
-extern void audit_schedule_prune(void);
-extern void audit_prune_trees(void);
 extern const char *audit_tree_path(struct audit_tree *);
 extern void audit_put_tree(struct audit_tree *);
+extern void audit_kill_trees(struct list_head *);
 #else
 #define audit_remove_tree_rule(rule) BUG()
 #define audit_add_tree_rule(rule) -EINVAL
@@ -140,6 +139,7 @@ extern void audit_put_tree(struct audit_tree *);
 #define audit_put_tree(tree) (void)0
 #define audit_tag_tree(old, new) -EINVAL
 #define audit_tree_path(rule) ""	/* never called */
+#define audit_kill_trees(list) BUG()
 #endif
 
 extern char *audit_unpack_string(void **, size_t *, size_t);
@@ -158,7 +158,10 @@ static inline int audit_signal_info(int sig, struct task_struct *t)
 	return 0;
 }
 extern void audit_filter_inodes(struct task_struct *, struct audit_context *);
+extern struct list_head *audit_killed_trees(void);
 #else
 #define audit_signal_info(s,t) AUDIT_DISABLED
 #define audit_filter_inodes(t,c) AUDIT_DISABLED
 #endif
+
+extern struct mutex audit_cmd_mutex;
diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c
index 3ff0731284a1..2451dc6f3282 100644
--- a/kernel/audit_tree.c
+++ b/kernel/audit_tree.c
@@ -2,6 +2,7 @@
 #include <linux/inotify.h>
 #include <linux/namei.h>
 #include <linux/mount.h>
+#include <linux/kthread.h>
 
 struct audit_tree;
 struct audit_chunk;
@@ -517,6 +518,8 @@ static void trim_marked(struct audit_tree *tree)
 	}
 }
 
+static void audit_schedule_prune(void);
+
 /* called with audit_filter_mutex */
 int audit_remove_tree_rule(struct audit_krule *rule)
 {
@@ -822,10 +825,11 @@ int audit_tag_tree(char *old, char *new)
 
 /*
  * That gets run when evict_chunk() ends up needing to kill audit_tree.
- * Runs from a separate thread, with audit_cmd_mutex held.
+ * Runs from a separate thread.
  */
-void audit_prune_trees(void)
+static int prune_tree_thread(void *unused)
 {
+	mutex_lock(&audit_cmd_mutex);
 	mutex_lock(&audit_filter_mutex);
 
 	while (!list_empty(&prune_list)) {
@@ -842,6 +846,40 @@ void audit_prune_trees(void)
 	}
 
 	mutex_unlock(&audit_filter_mutex);
+	mutex_unlock(&audit_cmd_mutex);
+	return 0;
+}
+
+static void audit_schedule_prune(void)
+{
+	kthread_run(prune_tree_thread, NULL, "audit_prune_tree");
+}
+
+/*
+ * ... and that one is done if evict_chunk() decides to delay until the end
+ * of syscall.  Runs synchronously.
+ */
+void audit_kill_trees(struct list_head *list)
+{
+	mutex_lock(&audit_cmd_mutex);
+	mutex_lock(&audit_filter_mutex);
+
+	while (!list_empty(list)) {
+		struct audit_tree *victim;
+
+		victim = list_entry(list->next, struct audit_tree, list);
+		kill_rules(victim);
+		list_del_init(&victim->list);
+
+		mutex_unlock(&audit_filter_mutex);
+
+		prune_one(victim);
+
+		mutex_lock(&audit_filter_mutex);
+	}
+
+	mutex_unlock(&audit_filter_mutex);
+	mutex_unlock(&audit_cmd_mutex);
 }
 
 /*
@@ -852,6 +890,8 @@ void audit_prune_trees(void)
 static void evict_chunk(struct audit_chunk *chunk)
 {
 	struct audit_tree *owner;
+	struct list_head *postponed = audit_killed_trees();
+	int need_prune = 0;
 	int n;
 
 	if (chunk->dead)
@@ -867,15 +907,21 @@ static void evict_chunk(struct audit_chunk *chunk)
 		owner->root = NULL;
 		list_del_init(&owner->same_root);
 		spin_unlock(&hash_lock);
-		kill_rules(owner);
-		list_move(&owner->list, &prune_list);
-		audit_schedule_prune();
+		if (!postponed) {
+			kill_rules(owner);
+			list_move(&owner->list, &prune_list);
+			need_prune = 1;
+		} else {
+			list_move(&owner->list, postponed);
+		}
 		spin_lock(&hash_lock);
 	}
 	list_del_rcu(&chunk->hash);
 	for (n = 0; n < chunk->count; n++)
 		list_del_init(&chunk->owners[n].list);
 	spin_unlock(&hash_lock);
+	if (need_prune)
+		audit_schedule_prune();
 	mutex_unlock(&audit_filter_mutex);
 }
 
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 2de95d1582bc..68d3c6a0ecd6 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -199,6 +199,7 @@ struct audit_context {
 
 	struct audit_tree_refs *trees, *first_trees;
 	int tree_count;
+	struct list_head killed_trees;
 
 	int type;
 	union {
@@ -853,6 +854,7 @@ static inline struct audit_context *audit_alloc_context(enum audit_state state)
 	if (!(context = kmalloc(sizeof(*context), GFP_KERNEL)))
 		return NULL;
 	audit_zero_context(context, state);
+	INIT_LIST_HEAD(&context->killed_trees);
 	return context;
 }
 
@@ -1545,6 +1547,8 @@ void audit_free(struct task_struct *tsk)
 	/* that can happen only if we are called from do_exit() */
 	if (context->in_syscall && context->current_state == AUDIT_RECORD_CONTEXT)
 		audit_log_exit(context, tsk);
+	if (!list_empty(&context->killed_trees))
+		audit_kill_trees(&context->killed_trees);
 
 	audit_free_context(context);
 }
@@ -1688,6 +1692,9 @@ void audit_syscall_exit(int valid, long return_code)
 	context->in_syscall = 0;
 	context->prio = context->state == AUDIT_RECORD_CONTEXT ? ~0ULL : 0;
 
+	if (!list_empty(&context->killed_trees))
+		audit_kill_trees(&context->killed_trees);
+
 	if (context->previous) {
 		struct audit_context *new_context = context->previous;
 		context->previous  = NULL;
@@ -2521,3 +2528,11 @@ void audit_core_dumps(long signr)
 	audit_log_format(ab, " sig=%ld", signr);
 	audit_log_end(ab);
 }
+
+struct list_head *audit_killed_trees(void)
+{
+	struct audit_context *ctx = current->audit_context;
+	if (likely(!ctx || !ctx->in_syscall))
+		return NULL;
+	return &ctx->killed_trees;
+}
-- 
cgit v1.2.3-59-g8ed1b


From e86435eb91b2bff114c5a02e46e16ce21b647ebe Mon Sep 17 00:00:00 2001
From: Peter Feuerer <peter@piie.net>
Date: Sun, 21 Jun 2009 18:53:03 +0200
Subject: acerhdf: Acer Aspire One fan control

Acerhdf is a driver for Acer Aspire One netbooks. It allows
to access the temperature sensor and to control the fan.

Signed-off-by: Peter Feuerer <peter@piie.net>
Signed-off-by: Andreas Mohr <andi@lisas.de>
Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 MAINTAINERS                    |   7 +
 drivers/platform/x86/Kconfig   |  17 ++
 drivers/platform/x86/Makefile  |   1 +
 drivers/platform/x86/acerhdf.c | 602 +++++++++++++++++++++++++++++++++++++++++
 4 files changed, 627 insertions(+)
 create mode 100644 drivers/platform/x86/acerhdf.c

diff --git a/MAINTAINERS b/MAINTAINERS
index cf4abddfc8a4..e9457feb7eba 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -222,6 +222,13 @@ L:	linux-acenic@sunsite.dk
 S:	Maintained
 F:	drivers/net/acenic*
 
+ACER ASPIRE ONE TEMPERATURE AND FAN DRIVER
+P: Peter Feuerer
+M: peter@piie.net
+W: http://piie.net/?section=acerhdf
+S: Maintained
+F: drivers/platform/x86/acerhdf.c
+
 ACER WMI LAPTOP EXTRAS
 P:	Carlos Corbacho
 M:	carlos@strangeworlds.co.uk
diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig
index 284ebaca6e45..7f79737b3947 100644
--- a/drivers/platform/x86/Kconfig
+++ b/drivers/platform/x86/Kconfig
@@ -34,6 +34,23 @@ config ACER_WMI
 	  If you have an ACPI-WMI compatible Acer/ Wistron laptop, say Y or M
 	  here.
 
+config ACERHDF
+	tristate "Acer Aspire One temperature and fan driver"
+	depends on THERMAL && THERMAL_HWMON && ACPI
+	---help---
+	  This is a driver for Acer Aspire One netbooks. It allows to access
+	  the temperature sensor and to control the fan.
+
+	  After loading this driver the BIOS is still in control of the fan.
+	  To let the kernel handle the fan, do:
+	  echo -n enabled > /sys/class/thermal/thermal_zone0/mode
+
+	  For more information about this driver see
+	  <http://piie.net/files/acerhdf_README.txt>
+
+	  If you have an Acer Aspire One netbook, say Y or M
+	  here.
+
 config ASUS_LAPTOP
 	tristate "Asus Laptop Extras (EXPERIMENTAL)"
 	depends on ACPI
diff --git a/drivers/platform/x86/Makefile b/drivers/platform/x86/Makefile
index e40c7bd1b87e..641b8bfa5538 100644
--- a/drivers/platform/x86/Makefile
+++ b/drivers/platform/x86/Makefile
@@ -9,6 +9,7 @@ obj-$(CONFIG_COMPAL_LAPTOP)	+= compal-laptop.o
 obj-$(CONFIG_DELL_LAPTOP)	+= dell-laptop.o
 obj-$(CONFIG_DELL_WMI)		+= dell-wmi.o
 obj-$(CONFIG_ACER_WMI)		+= acer-wmi.o
+obj-$(CONFIG_ACERHDF)		+= acerhdf.o
 obj-$(CONFIG_HP_WMI)		+= hp-wmi.o
 obj-$(CONFIG_TC1100_WMI)	+= tc1100-wmi.o
 obj-$(CONFIG_SONY_LAPTOP)	+= sony-laptop.o
diff --git a/drivers/platform/x86/acerhdf.c b/drivers/platform/x86/acerhdf.c
new file mode 100644
index 000000000000..bdfee177eefb
--- /dev/null
+++ b/drivers/platform/x86/acerhdf.c
@@ -0,0 +1,602 @@
+/*
+ * acerhdf - A driver which monitors the temperature
+ *           of the aspire one netbook, turns on/off the fan
+ *           as soon as the upper/lower threshold is reached.
+ *
+ * (C) 2009 - Peter Feuerer     peter (a) piie.net
+ *                              http://piie.net
+ *     2009 Borislav Petkov <petkovbb@gmail.com>
+ *
+ * Inspired by and many thanks to:
+ *  o acerfand   - Rachel Greenham
+ *  o acer_ec.pl - Michael Kurz     michi.kurz (at) googlemail.com
+ *               - Petr Tomasek     tomasek (#) etf,cuni,cz
+ *               - Carlos Corbacho  cathectic (at) gmail.com
+ *  o lkml       - Matthew Garrett
+ *               - Borislav Petkov
+ *               - Andreas Mohr
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+
+#define pr_fmt(fmt) "acerhdf: " fmt
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/dmi.h>
+#include <acpi/acpi_drivers.h>
+#include <linux/sched.h>
+#include <linux/thermal.h>
+#include <linux/platform_device.h>
+
+/*
+ * The driver is started with "kernel mode off" by default. That means, the BIOS
+ * is still in control of the fan. In this mode the driver allows to read the
+ * temperature of the cpu and a userspace tool may take over control of the fan.
+ * If the driver is switched to "kernel mode" (e.g. via module parameter) the
+ * driver is in full control of the fan. If you want the module to be started in
+ * kernel mode by default, define the following:
+ */
+#undef START_IN_KERNEL_MODE
+
+#define DRV_VER "0.5.13"
+
+/*
+ * According to the Atom N270 datasheet,
+ * (http://download.intel.com/design/processor/datashts/320032.pdf) the
+ * CPU's optimal operating limits denoted in junction temperature as
+ * measured by the on-die thermal monitor are within 0 <= Tj <= 90. So,
+ * assume 89°C is critical temperature.
+ */
+#define ACERHDF_TEMP_CRIT 89
+#define ACERHDF_FAN_OFF 0
+#define ACERHDF_FAN_AUTO 1
+
+/*
+ * No matter what value the user puts into the fanon variable, turn on the fan
+ * at 80 degree Celsius to prevent hardware damage
+ */
+#define ACERHDF_MAX_FANON 80
+
+/*
+ * Maximum interval between two temperature checks is 15 seconds, as the die
+ * can get hot really fast under heavy load (plus we shouldn't forget about
+ * possible impact of _external_ aggressive sources such as heaters, sun etc.)
+ */
+#define ACERHDF_MAX_INTERVAL 15
+
+#ifdef START_IN_KERNEL_MODE
+static int kernelmode = 1;
+#else
+static int kernelmode;
+#endif
+
+static unsigned int interval = 10;
+static unsigned int fanon = 63;
+static unsigned int fanoff = 58;
+static unsigned int verbose;
+static unsigned int fanstate = ACERHDF_FAN_AUTO;
+static char force_bios[16];
+static unsigned int prev_interval;
+struct thermal_zone_device *thz_dev;
+struct thermal_cooling_device *cl_dev;
+struct platform_device *acerhdf_dev;
+
+module_param(kernelmode, uint, 0);
+MODULE_PARM_DESC(kernelmode, "Kernel mode fan control on / off");
+module_param(interval, uint, 0600);
+MODULE_PARM_DESC(interval, "Polling interval of temperature check");
+module_param(fanon, uint, 0600);
+MODULE_PARM_DESC(fanon, "Turn the fan on above this temperature");
+module_param(fanoff, uint, 0600);
+MODULE_PARM_DESC(fanoff, "Turn the fan off below this temperature");
+module_param(verbose, uint, 0600);
+MODULE_PARM_DESC(verbose, "Enable verbose dmesg output");
+module_param_string(force_bios, force_bios, 16, 0);
+MODULE_PARM_DESC(force_bios, "Force BIOS version and omit BIOS check");
+
+/* BIOS settings */
+struct bios_settings_t {
+	const char *vendor;
+	const char *version;
+	unsigned char fanreg;
+	unsigned char tempreg;
+	unsigned char fancmd[2]; /* fan off and auto commands */
+};
+
+/* Register addresses and values for different BIOS versions */
+static const struct bios_settings_t bios_tbl[] = {
+	{"Acer", "v0.3109", 0x55, 0x58, {0x1f, 0x00} },
+	{"Acer", "v0.3114", 0x55, 0x58, {0x1f, 0x00} },
+	{"Acer", "v0.3301", 0x55, 0x58, {0xaf, 0x00} },
+	{"Acer", "v0.3304", 0x55, 0x58, {0xaf, 0x00} },
+	{"Acer", "v0.3305", 0x55, 0x58, {0xaf, 0x00} },
+	{"Acer", "v0.3308", 0x55, 0x58, {0x21, 0x00} },
+	{"Acer", "v0.3309", 0x55, 0x58, {0x21, 0x00} },
+	{"Acer", "v0.3310", 0x55, 0x58, {0x21, 0x00} },
+	{"Gateway", "v0.3103", 0x55, 0x58, {0x21, 0x00} },
+	{"Packard Bell", "v0.3105", 0x55, 0x58, {0x21, 0x00} },
+	{"", "", 0, 0, {0, 0} }
+};
+
+static const struct bios_settings_t *bios_cfg __read_mostly;
+
+
+static int acerhdf_get_temp(int *temp)
+{
+	u8 read_temp;
+
+	if (ec_read(bios_cfg->tempreg, &read_temp))
+		return -EINVAL;
+
+	*temp = read_temp;
+
+	return 0;
+}
+
+static int acerhdf_get_fanstate(int *state)
+{
+	u8 fan;
+	bool tmp;
+
+	if (ec_read(bios_cfg->fanreg, &fan))
+		return -EINVAL;
+
+	tmp = (fan == bios_cfg->fancmd[ACERHDF_FAN_OFF]);
+	*state = tmp ? ACERHDF_FAN_OFF : ACERHDF_FAN_AUTO;
+
+	return 0;
+}
+
+static void acerhdf_change_fanstate(int state)
+{
+	unsigned char cmd;
+
+	if (verbose)
+		pr_notice("fan %s\n", (state == ACERHDF_FAN_OFF) ?
+				"OFF" : "ON");
+
+	if ((state != ACERHDF_FAN_OFF) && (state != ACERHDF_FAN_AUTO)) {
+		pr_err("invalid fan state %d requested, setting to auto!\n",
+			state);
+		state = ACERHDF_FAN_AUTO;
+	}
+
+	cmd = bios_cfg->fancmd[state];
+	fanstate = state;
+
+	ec_write(bios_cfg->fanreg, cmd);
+}
+
+static void acerhdf_check_param(struct thermal_zone_device *thermal)
+{
+	if (fanon > ACERHDF_MAX_FANON) {
+		pr_err("fanon temperature too high, set to %d\n",
+				ACERHDF_MAX_FANON);
+		fanon = ACERHDF_MAX_FANON;
+	}
+
+	if (kernelmode && prev_interval != interval) {
+		if (interval > ACERHDF_MAX_INTERVAL) {
+			pr_err("interval too high, set to %d\n",
+				ACERHDF_MAX_INTERVAL);
+			interval = ACERHDF_MAX_INTERVAL;
+		}
+		if (verbose)
+			pr_notice("interval changed to: %d\n",
+					interval);
+		thermal->polling_delay = interval*1000;
+		prev_interval = interval;
+	}
+}
+
+/*
+ * This is the thermal zone callback which does the delayed polling of the fan
+ * state. We do check /sysfs-originating settings here in acerhdf_check_param()
+ * as late as the polling interval is since we can't do that in the respective
+ * accessors of the module parameters.
+ */
+static int acerhdf_get_ec_temp(struct thermal_zone_device *thermal,
+			       unsigned long *t)
+{
+	int temp, err = 0;
+
+	acerhdf_check_param(thermal);
+
+	err = acerhdf_get_temp(&temp);
+	if (err)
+		return err;
+
+	if (verbose)
+		pr_notice("temp %d\n", temp);
+
+	*t = temp;
+	return 0;
+}
+
+static int acerhdf_bind(struct thermal_zone_device *thermal,
+			struct thermal_cooling_device *cdev)
+{
+	/* if the cooling device is the one from acerhdf bind it */
+	if (cdev != cl_dev)
+		return 0;
+
+	if (thermal_zone_bind_cooling_device(thermal, 0, cdev)) {
+		pr_err("error binding cooling dev\n");
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int acerhdf_unbind(struct thermal_zone_device *thermal,
+			  struct thermal_cooling_device *cdev)
+{
+	if (cdev != cl_dev)
+		return 0;
+
+	if (thermal_zone_unbind_cooling_device(thermal, 0, cdev)) {
+		pr_err("error unbinding cooling dev\n");
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static inline void acerhdf_revert_to_bios_mode(void)
+{
+	acerhdf_change_fanstate(ACERHDF_FAN_AUTO);
+	kernelmode = 0;
+	if (thz_dev)
+		thz_dev->polling_delay = 0;
+	pr_notice("kernel mode fan control OFF\n");
+}
+static inline void acerhdf_enable_kernelmode(void)
+{
+	kernelmode = 1;
+
+	thz_dev->polling_delay = interval*1000;
+	thermal_zone_device_update(thz_dev);
+	pr_notice("kernel mode fan control ON\n");
+}
+
+static int acerhdf_get_mode(struct thermal_zone_device *thermal,
+			    enum thermal_device_mode *mode)
+{
+	if (verbose)
+		pr_notice("kernel mode fan control %d\n", kernelmode);
+
+	*mode = (kernelmode) ? THERMAL_DEVICE_ENABLED
+			     : THERMAL_DEVICE_DISABLED;
+
+	return 0;
+}
+
+/*
+ * set operation mode;
+ * enabled: the thermal layer of the kernel takes care about
+ *          the temperature and the fan.
+ * disabled: the BIOS takes control of the fan.
+ */
+static int acerhdf_set_mode(struct thermal_zone_device *thermal,
+			    enum thermal_device_mode mode)
+{
+	if (mode == THERMAL_DEVICE_DISABLED && kernelmode)
+		acerhdf_revert_to_bios_mode();
+	else if (mode == THERMAL_DEVICE_ENABLED && !kernelmode)
+		acerhdf_enable_kernelmode();
+
+	return 0;
+}
+
+static int acerhdf_get_trip_type(struct thermal_zone_device *thermal, int trip,
+				 enum thermal_trip_type *type)
+{
+	if (trip == 0)
+		*type = THERMAL_TRIP_ACTIVE;
+
+	return 0;
+}
+
+static int acerhdf_get_trip_temp(struct thermal_zone_device *thermal, int trip,
+				 unsigned long *temp)
+{
+	if (trip == 0)
+		*temp = fanon;
+
+	return 0;
+}
+
+static int acerhdf_get_crit_temp(struct thermal_zone_device *thermal,
+				 unsigned long *temperature)
+{
+	*temperature = ACERHDF_TEMP_CRIT;
+	return 0;
+}
+
+/* bind callback functions to thermalzone */
+struct thermal_zone_device_ops acerhdf_dev_ops = {
+	.bind = acerhdf_bind,
+	.unbind = acerhdf_unbind,
+	.get_temp = acerhdf_get_ec_temp,
+	.get_mode = acerhdf_get_mode,
+	.set_mode = acerhdf_set_mode,
+	.get_trip_type = acerhdf_get_trip_type,
+	.get_trip_temp = acerhdf_get_trip_temp,
+	.get_crit_temp = acerhdf_get_crit_temp,
+};
+
+
+/*
+ * cooling device callback functions
+ * get maximal fan cooling state
+ */
+static int acerhdf_get_max_state(struct thermal_cooling_device *cdev,
+				 unsigned long *state)
+{
+	*state = 1;
+
+	return 0;
+}
+
+static int acerhdf_get_cur_state(struct thermal_cooling_device *cdev,
+				 unsigned long *state)
+{
+	int err = 0, tmp;
+
+	err = acerhdf_get_fanstate(&tmp);
+	if (err)
+		return err;
+
+	*state = (tmp == ACERHDF_FAN_AUTO) ? 1 : 0;
+	return 0;
+}
+
+/* change current fan state - is overwritten when running in kernel mode */
+static int acerhdf_set_cur_state(struct thermal_cooling_device *cdev,
+				 unsigned long state)
+{
+	int cur_temp, cur_state, err = 0;
+
+	if (!kernelmode)
+		return 0;
+
+	err = acerhdf_get_temp(&cur_temp);
+	if (err) {
+		pr_err("error reading temperature, hand off control to BIOS\n");
+		goto err_out;
+	}
+
+	err = acerhdf_get_fanstate(&cur_state);
+	if (err) {
+		pr_err("error reading fan state, hand off control to BIOS\n");
+		goto err_out;
+	}
+
+	if (state == 0) {
+		/* turn fan off only if below fanoff temperature */
+		if ((cur_state == ACERHDF_FAN_AUTO) &&
+		    (cur_temp < fanoff))
+			acerhdf_change_fanstate(ACERHDF_FAN_OFF);
+	} else {
+		if (cur_state == ACERHDF_FAN_OFF)
+			acerhdf_change_fanstate(ACERHDF_FAN_AUTO);
+	}
+	return 0;
+
+err_out:
+	acerhdf_revert_to_bios_mode();
+	return -EINVAL;
+}
+
+/* bind fan callbacks to fan device */
+struct thermal_cooling_device_ops acerhdf_cooling_ops = {
+	.get_max_state = acerhdf_get_max_state,
+	.get_cur_state = acerhdf_get_cur_state,
+	.set_cur_state = acerhdf_set_cur_state,
+};
+
+/* suspend / resume functionality */
+static int acerhdf_suspend(struct platform_device *dev, pm_message_t state)
+{
+	if (kernelmode)
+		acerhdf_change_fanstate(ACERHDF_FAN_AUTO);
+
+	if (verbose)
+		pr_notice("going suspend\n");
+
+	return 0;
+}
+
+static int acerhdf_resume(struct platform_device *device)
+{
+	if (verbose)
+		pr_notice("resuming\n");
+
+	return 0;
+}
+
+static int __devinit acerhdf_probe(struct platform_device *device)
+{
+	return 0;
+}
+
+static int acerhdf_remove(struct platform_device *device)
+{
+	return 0;
+}
+
+struct platform_driver acerhdf_drv = {
+	.driver = {
+		.name = "acerhdf",
+		.owner = THIS_MODULE,
+	},
+	.probe = acerhdf_probe,
+	.remove = acerhdf_remove,
+	.suspend = acerhdf_suspend,
+	.resume = acerhdf_resume,
+};
+
+
+/* check hardware */
+static int acerhdf_check_hardware(void)
+{
+	char const *vendor, *version, *product;
+	int i;
+
+	/* get BIOS data */
+	vendor  = dmi_get_system_info(DMI_SYS_VENDOR);
+	version = dmi_get_system_info(DMI_BIOS_VERSION);
+	product = dmi_get_system_info(DMI_PRODUCT_NAME);
+
+	pr_info("Acer Aspire One Fan driver, v.%s\n", DRV_VER);
+
+	if (!force_bios[0]) {
+		if (strncmp(product, "AO", 2)) {
+			pr_err("no Aspire One hardware found\n");
+			return -EINVAL;
+		}
+	} else {
+		pr_info("forcing BIOS version: %s\n", version);
+		version = force_bios;
+		kernelmode = 0;
+	}
+
+	if (verbose)
+		pr_info("BIOS info: %s %s, product: %s\n",
+			vendor, version, product);
+
+	/* search BIOS version and vendor in BIOS settings table */
+	for (i = 0; bios_tbl[i].version[0]; i++) {
+		if (!strcmp(bios_tbl[i].vendor, vendor) &&
+		    !strcmp(bios_tbl[i].version, version)) {
+			bios_cfg = &bios_tbl[i];
+			break;
+		}
+	}
+
+	if (!bios_cfg) {
+		pr_err("unknown (unsupported) BIOS version %s/%s, "
+			"please report, aborting!\n", vendor, version);
+		return -EINVAL;
+	}
+
+	/*
+	 * if started with kernel mode off, prevent the kernel from switching
+	 * off the fan
+	 */
+	if (!kernelmode) {
+		pr_notice("Fan control off, to enable do:\n");
+		pr_notice("echo -n \"enabled\" > "
+			"/sys/class/thermal/thermal_zone0/mode\n");
+	}
+
+	return 0;
+}
+
+static int acerhdf_register_platform(void)
+{
+	int err = 0;
+
+	err = platform_driver_register(&acerhdf_drv);
+	if (err)
+		return err;
+
+	acerhdf_dev = platform_device_alloc("acerhdf", -1);
+	platform_device_add(acerhdf_dev);
+
+	return 0;
+}
+
+static void acerhdf_unregister_platform(void)
+{
+	if (!acerhdf_dev)
+		return;
+
+	platform_device_del(acerhdf_dev);
+	platform_driver_unregister(&acerhdf_drv);
+}
+
+static int acerhdf_register_thermal(void)
+{
+	cl_dev = thermal_cooling_device_register("acerhdf-fan", NULL,
+						 &acerhdf_cooling_ops);
+
+	if (IS_ERR(cl_dev))
+		return -EINVAL;
+
+	thz_dev = thermal_zone_device_register("acerhdf", 1, NULL,
+					      &acerhdf_dev_ops, 0, 0, 0,
+					      (kernelmode) ? interval*1000 : 0);
+	if (IS_ERR(thz_dev))
+		return -EINVAL;
+
+	return 0;
+}
+
+static void acerhdf_unregister_thermal(void)
+{
+	if (cl_dev) {
+		thermal_cooling_device_unregister(cl_dev);
+		cl_dev = NULL;
+	}
+
+	if (thz_dev) {
+		thermal_zone_device_unregister(thz_dev);
+		thz_dev = NULL;
+	}
+}
+
+static int __init acerhdf_init(void)
+{
+	int err = 0;
+
+	err = acerhdf_check_hardware();
+	if (err)
+		goto out_err;
+
+	err = acerhdf_register_platform();
+	if (err)
+		goto err_unreg;
+
+	err = acerhdf_register_thermal();
+	if (err)
+		goto err_unreg;
+
+	return 0;
+
+err_unreg:
+	acerhdf_unregister_thermal();
+	acerhdf_unregister_platform();
+
+out_err:
+	return -ENODEV;
+}
+
+static void __exit acerhdf_exit(void)
+{
+	acerhdf_change_fanstate(ACERHDF_FAN_AUTO);
+	acerhdf_unregister_thermal();
+	acerhdf_unregister_platform();
+}
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Peter Feuerer");
+MODULE_DESCRIPTION("Aspire One temperature and fan driver");
+MODULE_ALIAS("dmi:*:*Acer*:*:");
+MODULE_ALIAS("dmi:*:*Gateway*:*:");
+MODULE_ALIAS("dmi:*:*Packard Bell*:*:");
+
+module_init(acerhdf_init);
+module_exit(acerhdf_exit);
-- 
cgit v1.2.3-59-g8ed1b


From 7a04b8491a077471a34938b8ca060c37220953be Mon Sep 17 00:00:00 2001
From: Zhao Yakui <yakui.zhao@intel.com>
Date: Wed, 24 Jun 2009 11:46:44 +0800
Subject: ACPI: Rename ACPI processor device bus ID

Some BIOS re-use the same processor bus id
in different scope:

	\_SB.SCK0.CPU0
	\_SB.SCK1.CPU0

But the (deprecated) /proc/acpi/ interface
assumes the bus-id's are unique, resulting in an OOPS
when the processor driver is loaded:

WARNING: at fs/proc/generic.c:590 proc_register+0x148/0x180()
Hardware name: Sunrise Ridge
proc_dir_entry 'processor/CPU0' already registered
Call Trace:
 [<ffffffff8023f7ef>] warn_slowpath+0xb1/0xe5
 [<ffffffff8036243b>] ? ida_get_new_above+0x190/0x1b1
 [<ffffffff803625a8>] ? idr_pre_get+0x5f/0x75
 [<ffffffff8030b2f6>] proc_register+0x148/0x180
 [<ffffffff8030b4ff>] proc_mkdir_mode+0x3d/0x52
 [<ffffffff8030b525>] proc_mkdir+0x11/0x13
 [<ffffffffa0014b89>] acpi_processor_start+0x755/0x9bc [processor]

Rename the processor device bus id. And the new bus id will be
generated as the following format:
	CPU+ CPU ID

For example: If the cpu ID is 5, then the bus ID will be "CPU5".
	If the CPU ID is 10, then the bus ID will be "CPUA".

Yes, this will change the directory names seen
in /proc/acpi/processor/* on some systems.
Before this patch, those directory names where
totally arbitrary strings based on the interal AML device strings.

http://bugzilla.kernel.org/show_bug.cgi?id=13612

Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/processor_core.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c
index 23f0fb84f1c1..105562e375ac 100644
--- a/drivers/acpi/processor_core.c
+++ b/drivers/acpi/processor_core.c
@@ -649,7 +649,16 @@ static int acpi_processor_get_info(struct acpi_device *device)
 			return -ENODEV;
 		}
 	}
-
+	/*
+	 * On some boxes several processors use the same processor bus id.
+	 * But they are located in different scope. For example:
+	 * \_SB.SCK0.CPU0
+	 * \_SB.SCK1.CPU0
+	 * Rename the processor device bus id. And the new bus id will be
+	 * generated as the following format:
+	 * CPU+CPU ID.
+	 */
+	sprintf(acpi_device_bid(device), "CPU%X", pr->id);
 	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Processor [%d:%d]\n", pr->id,
 			  pr->acpi_id));
 
-- 
cgit v1.2.3-59-g8ed1b


From cede2cb6ee9b0ddaa3dbc9939418ff177a831600 Mon Sep 17 00:00:00 2001
From: Pekka Enberg <penberg@cs.helsinki.fi>
Date: Tue, 16 Jun 2009 19:28:45 +0000
Subject: eeepc-laptop: enable camera by default

If we leave the camera disabled by default, userspace programs (e.g.
Skype, Cheese) leave the user out in the cold saying that the machine
"has no camera." Therefore, it's better to enable camera by default and
let people who really don't want it just disable the thing.

To reduce power usage you should enable USB autosuspend:
echo -n auto > /sys/bus/usb/drivers/uvcvideo/*:*/../power/level

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/eeepc-laptop.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index 46b5aa5e85f0..884d76b9e8ba 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -321,6 +321,15 @@ static const struct rfkill_ops eeepc_rfkill_ops = {
 	.set_block = eeepc_rfkill_set,
 };
 
+static void __init eeepc_enable_camera(void)
+{
+	/*
+	 * If the following call to set_acpi() fails, it's because there's no
+	 * camera so we can ignore the error.
+	 */
+	set_acpi(CM_ASL_CAMERA, 1);
+}
+
 /*
  * Sys helpers
  */
@@ -983,6 +992,9 @@ static int __init eeepc_laptop_init(void)
 	result = eeepc_hwmon_init(dev);
 	if (result)
 		goto fail_hwmon;
+
+	eeepc_enable_camera();
+
 	/* Register platform stuff */
 	result = platform_driver_register(&platform_driver);
 	if (result)
-- 
cgit v1.2.3-59-g8ed1b


From 116bf2e010a0600371aede450351973821dfd9e2 Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Tue, 16 Jun 2009 19:28:46 +0000
Subject: asus-laptop: platform dev as parent for led and backlight

Makes asus-laptop platform device the parent device of
backlight and led classes.

With this patch, leds and backlight are also available in
/sys/devices/platform/asus-laptop/ like thinkpad_acpi.

Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/asus-laptop.c | 40 ++++++++++++++++++--------------------
 1 file changed, 19 insertions(+), 21 deletions(-)

diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c
index eaffe732653a..faa87d3e3983 100644
--- a/drivers/platform/x86/asus-laptop.c
+++ b/drivers/platform/x86/asus-laptop.c
@@ -1321,7 +1321,6 @@ out:
 
 static int __init asus_laptop_init(void)
 {
-	struct device *dev;
 	int result;
 
 	if (acpi_disabled)
@@ -1343,24 +1342,10 @@ static int __init asus_laptop_init(void)
 		return -ENODEV;
 	}
 
-	dev = acpi_get_physical_device(hotk->device->handle);
-
-	if (!acpi_video_backlight_support()) {
-		result = asus_backlight_init(dev);
-		if (result)
-			goto fail_backlight;
-	} else
-		printk(ASUS_INFO "Brightness ignored, must be controlled by "
-		       "ACPI video driver\n");
-
 	result = asus_input_init();
 	if (result)
 		goto fail_input;
 
-	result = asus_led_init(dev);
-	if (result)
-		goto fail_led;
-
 	/* Register platform stuff */
 	result = platform_driver_register(&asuspf_driver);
 	if (result)
@@ -1381,8 +1366,27 @@ static int __init asus_laptop_init(void)
 	if (result)
 		goto fail_sysfs;
 
+	result = asus_led_init(&asuspf_device->dev);
+	if (result)
+		goto fail_led;
+
+	if (!acpi_video_backlight_support()) {
+		result = asus_backlight_init(&asuspf_device->dev);
+		if (result)
+			goto fail_backlight;
+	} else
+		printk(ASUS_INFO "Brightness ignored, must be controlled by "
+		       "ACPI video driver\n");
+
 	return 0;
 
+fail_backlight:
+       asus_led_exit();
+
+fail_led:
+       sysfs_remove_group(&asuspf_device->dev.kobj,
+			  &asuspf_attribute_group);
+
 fail_sysfs:
 	platform_device_del(asuspf_device);
 
@@ -1393,15 +1397,9 @@ fail_platform_device1:
 	platform_driver_unregister(&asuspf_driver);
 
 fail_platform_driver:
-	asus_led_exit();
-
-fail_led:
 	asus_input_exit();
 
 fail_input:
-	asus_backlight_exit();
-
-fail_backlight:
 
 	return result;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 76593d6fb0a51cb0d666f37d91a990e36c068365 Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Tue, 16 Jun 2009 19:28:47 +0000
Subject: acpi4asus: update MAINTAINER and KConfig links

The bug tracker have moved from sourceforge to
http://dev.iksaif.net . The homepage of the project
is now http://acpi4asus.sf.net with links to the new
bug tracker. No change for the mailing list.

Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 MAINTAINERS                  |  8 +++-----
 drivers/platform/x86/Kconfig | 11 +++++++----
 2 files changed, 10 insertions(+), 9 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 7d9d3745aab9..05a4f306f1a0 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -920,8 +920,7 @@ M:	corentincj@iksaif.net
 P:	Karol Kozimor
 M:	sziwan@users.sourceforge.net
 L:	acpi4asus-user@lists.sourceforge.net
-W:	http://sourceforge.net/projects/acpi4asus
-W:	http://xf.iksaif.net/acpi4asus
+W:	http://acpi4asus.sf.net
 S:	Maintained
 F:	arch/x86/kernel/acpi/boot.c
 F:	drivers/platform/x86/asus_acpi.c
@@ -937,8 +936,7 @@ ASUS LAPTOP EXTRAS DRIVER
 P:	Corentin Chary
 M:	corentincj@iksaif.net
 L:	acpi4asus-user@lists.sourceforge.net
-W:	http://sourceforge.net/projects/acpi4asus
-W:	http://xf.iksaif.net/acpi4asus
+W:	http://acpi4asus.sf.net
 S:	Maintained
 F:	drivers/platform/x86/asus-laptop.c
 
@@ -2117,7 +2115,7 @@ EEEPC LAPTOP EXTRAS DRIVER
 P:	Corentin Chary
 M:	corentincj@iksaif.net
 L:	acpi4asus-user@lists.sourceforge.net
-W:	http://sourceforge.net/projects/acpi4asus
+W:	http://acpi4asus.sf.net
 S:	Maintained
 F:	drivers/platform/x86/eeepc-laptop.c
 
diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig
index 5613483db141..06806559c8c3 100644
--- a/drivers/platform/x86/Kconfig
+++ b/drivers/platform/x86/Kconfig
@@ -62,12 +62,12 @@ config ASUS_LAPTOP
 	---help---
 	  This is the new Linux driver for Asus laptops. It may also support some
 	  MEDION, JVC or VICTOR laptops. It makes all the extra buttons generate
-	  standard ACPI events that go through /proc/acpi/events. It also adds
+	  standard ACPI events and input events. It also adds
 	  support for video output switching, LCD backlight control, Bluetooth and
 	  Wlan control, and most importantly, allows you to blink those fancy LEDs.
 
 	  For more information and a userspace daemon for handling the extra
-	  buttons see <http://acpi4asus.sf.net/>.
+	  buttons see <http://acpi4asus.sf.net>.
 
 	  If you have an ACPI-compatible ASUS laptop, say Y or M here.
 
@@ -359,7 +359,10 @@ config EEEPC_LAPTOP
 	select HWMON
 	---help---
 	  This driver supports the Fn-Fx keys on Eee PC laptops.
-	  It also adds the ability to switch camera/wlan on/off.
+
+	  It  also gives access to some extra laptop functionalities like
+	  Bluetooth, backlight and allows powering on/off some other
+	  devices.
 
 	  If you have an Eee PC laptop, say Y or M here.
 
@@ -407,7 +410,7 @@ config ACPI_ASUS
 	  parameters.
 
 	  More information and a userspace daemon for handling the extra buttons
-	  at <http://sourceforge.net/projects/acpi4asus/>.
+	  at <http://acpi4asus.sf.net>.
 
 	  If you have an ACPI-compatible ASUS laptop, say Y or M here. This
 	  driver is still under development, so if your laptop is unsupported or
-- 
cgit v1.2.3-59-g8ed1b


From 6122af3743a48dddae19810626dd7c9c8e6c1df8 Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Tue, 16 Jun 2009 19:28:48 +0000
Subject: asus_acpi: Deprecate in favor of asus-laptop

asus-laptop have been merged in the kernel two years ago,
it is now stable and used by most distribution instead of
the old asus_acpi driver.

Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig
index 06806559c8c3..13226e4ad234 100644
--- a/drivers/platform/x86/Kconfig
+++ b/drivers/platform/x86/Kconfig
@@ -52,7 +52,7 @@ config ACERHDF
 	  here.
 
 config ASUS_LAPTOP
-	tristate "Asus Laptop Extras (EXPERIMENTAL)"
+	tristate "Asus Laptop Extras"
 	depends on ACPI
 	depends on EXPERIMENTAL && !ACPI_ASUS
 	select LEDS_CLASS
@@ -389,7 +389,7 @@ config ACPI_WMI
 	  any ACPI-WMI devices.
 
 config ACPI_ASUS
-	tristate "ASUS/Medion Laptop Extras"
+	tristate "ASUS/Medion Laptop Extras (DEPRECATED)"
 	depends on ACPI
 	select BACKLIGHT_CLASS_DEVICE
 	---help---
-- 
cgit v1.2.3-59-g8ed1b


From b7b700d4a473d56103e87e341ad555e8a7cce06d Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Tue, 16 Jun 2009 19:28:52 +0000
Subject: eeepc-laptop: sync eeepc-laptop with asus_acpi

In the default Eee PC distribution, there is a modified
asus_acpi driver. eeepc-laptop is a cleaned version of this
driver. Sync ASL enum and getter/setters with asus_acpi.

Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/eeepc-laptop.c | 19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index 884d76b9e8ba..73f3cb0fd76c 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -62,7 +62,10 @@ enum {
 	DISABLE_ASL_GPS = 0x0020,
 	DISABLE_ASL_DISPLAYSWITCH = 0x0040,
 	DISABLE_ASL_MODEM = 0x0080,
-	DISABLE_ASL_CARDREADER = 0x0100
+	DISABLE_ASL_CARDREADER = 0x0100,
+	DISABLE_ASL_3G = 0x0200,
+	DISABLE_ASL_WIMAX = 0x0400,
+	DISABLE_ASL_HWCF = 0x0800
 };
 
 enum {
@@ -87,7 +90,13 @@ enum {
 	CM_ASL_USBPORT3,
 	CM_ASL_MODEM,
 	CM_ASL_CARDREADER,
-	CM_ASL_LID
+	CM_ASL_3G,
+	CM_ASL_WIMAX,
+	CM_ASL_HWCF,
+	CM_ASL_LID,
+	CM_ASL_TYPE,
+	CM_ASL_PANELPOWER,	/*P901*/
+	CM_ASL_TPD
 };
 
 static const char *cm_getv[] = {
@@ -96,7 +105,8 @@ static const char *cm_getv[] = {
 	NULL, "PBLG", NULL, NULL,
 	"CFVG", NULL, NULL, NULL,
 	"USBG", NULL, NULL, "MODG",
-	"CRDG", "LIDG"
+	"CRDG", "M3GG", "WIMG", "HWCF",
+	"LIDG",	"TYPE", "PBPG",	"TPDG"
 };
 
 static const char *cm_setv[] = {
@@ -105,7 +115,8 @@ static const char *cm_setv[] = {
 	"SDSP", "PBLS", "HDPS", NULL,
 	"CFVS", NULL, NULL, NULL,
 	"USBG", NULL, NULL, "MODS",
-	"CRDS", NULL
+	"CRDS", "M3GS", "WIMS", NULL,
+	NULL, NULL, "PBPS", "TPDS"
 };
 
 #define EEEPC_EC	"\\_SB.PCI0.SBRG.EC0."
-- 
cgit v1.2.3-59-g8ed1b


From b31d0fde89c905673ceed0404d5ae24f2261d7c7 Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Tue, 16 Jun 2009 19:28:56 +0000
Subject: eeepc-laptop: cpufv updates

Limit cpufv input to acceptables values.
Add an available_cpufv file to show available
presets.
Change cpufv ouput format from %d to %#x, it won't
break compatibility with existing userspace tools, but
it provide a more human readable output.

Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/eeepc-laptop.c | 77 ++++++++++++++++++++++++++++++++++++-
 1 file changed, 76 insertions(+), 1 deletion(-)

diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index 73f3cb0fd76c..4207b26ff990 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -392,13 +392,88 @@ static ssize_t show_sys_acpi(int cm, char *buf)
 EEEPC_CREATE_DEVICE_ATTR(camera, CM_ASL_CAMERA);
 EEEPC_CREATE_DEVICE_ATTR(cardr, CM_ASL_CARDREADER);
 EEEPC_CREATE_DEVICE_ATTR(disp, CM_ASL_DISPLAYSWITCH);
-EEEPC_CREATE_DEVICE_ATTR(cpufv, CM_ASL_CPUFV);
+
+struct eeepc_cpufv {
+	int num;
+	int cur;
+};
+
+static int get_cpufv(struct eeepc_cpufv *c)
+{
+	c->cur = get_acpi(CM_ASL_CPUFV);
+	c->num = (c->cur >> 8) & 0xff;
+	c->cur &= 0xff;
+	if (c->cur < 0 || c->num <= 0 || c->num > 12)
+		return -ENODEV;
+	return 0;
+}
+
+static ssize_t show_available_cpufv(struct device *dev,
+				    struct device_attribute *attr,
+				    char *buf)
+{
+	struct eeepc_cpufv c;
+	int i;
+	ssize_t len = 0;
+
+	if (get_cpufv(&c))
+		return -ENODEV;
+	for (i = 0; i < c.num; i++)
+		len += sprintf(buf + len, "%d ", i);
+	len += sprintf(buf + len, "\n");
+	return len;
+}
+
+static ssize_t show_cpufv(struct device *dev,
+			  struct device_attribute *attr,
+			  char *buf)
+{
+	struct eeepc_cpufv c;
+
+	if (get_cpufv(&c))
+		return -ENODEV;
+	return sprintf(buf, "%#x\n", (c.num << 8) | c.cur);
+}
+
+static ssize_t store_cpufv(struct device *dev,
+			   struct device_attribute *attr,
+			   const char *buf, size_t count)
+{
+	struct eeepc_cpufv c;
+	int rv, value;
+
+	if (get_cpufv(&c))
+		return -ENODEV;
+	rv = parse_arg(buf, count, &value);
+	if (rv < 0)
+		return rv;
+	if (!rv || value < 0 || value >= c.num)
+		return -EINVAL;
+	set_acpi(CM_ASL_CPUFV, value);
+	return rv;
+}
+
+static struct device_attribute dev_attr_cpufv = {
+	.attr = {
+		.name = "cpufv",
+		.mode = 0644 },
+	.show   = show_cpufv,
+	.store  = store_cpufv
+};
+
+static struct device_attribute dev_attr_available_cpufv = {
+	.attr = {
+		.name = "available_cpufv",
+		.mode = 0444 },
+	.show   = show_available_cpufv
+};
 
 static struct attribute *platform_attributes[] = {
 	&dev_attr_camera.attr,
 	&dev_attr_cardr.attr,
 	&dev_attr_disp.attr,
 	&dev_attr_cpufv.attr,
+	&dev_attr_available_cpufv.attr,
 	NULL
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 2fcc23da5522b89677fb0af6043a88e88fdd09a2 Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Fri, 19 Jun 2009 14:52:03 +0200
Subject: asus-laptop: use pr_fmt and pr_<level>

Convert the unusual printk(ASUS_<level> uses to
the more standard pr_fmt and pr_<level>(.

Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/asus-laptop.c | 50 ++++++++++++++++++++------------------
 1 file changed, 26 insertions(+), 24 deletions(-)

diff --git a/drivers/platform/x86/asus-laptop.c b/drivers/platform/x86/asus-laptop.c
index faa87d3e3983..db657bbeec90 100644
--- a/drivers/platform/x86/asus-laptop.c
+++ b/drivers/platform/x86/asus-laptop.c
@@ -33,6 +33,8 @@
  *  Sam Lin        - GPS support
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -53,9 +55,10 @@
 #define ASUS_HOTK_NAME          "Asus Laptop Support"
 #define ASUS_HOTK_CLASS         "hotkey"
 #define ASUS_HOTK_DEVICE_NAME   "Hotkey"
-#define ASUS_HOTK_FILE          "asus-laptop"
+#define ASUS_HOTK_FILE          KBUILD_MODNAME
 #define ASUS_HOTK_PREFIX        "\\_SB.ATKD."
 
+
 /*
  * Some events we use, same for all Asus
  */
@@ -327,7 +330,7 @@ static int read_wireless_status(int mask)
 
 	rv = acpi_evaluate_integer(wireless_status_handle, NULL, NULL, &status);
 	if (ACPI_FAILURE(rv))
-		printk(ASUS_WARNING "Error reading Wireless status\n");
+		pr_warning("Error reading Wireless status\n");
 	else
 		return (status & mask) ? 1 : 0;
 
@@ -341,7 +344,7 @@ static int read_gps_status(void)
 
 	rv = acpi_evaluate_integer(gps_status_handle, NULL, NULL, &status);
 	if (ACPI_FAILURE(rv))
-		printk(ASUS_WARNING "Error reading GPS status\n");
+		pr_warning("Error reading GPS status\n");
 	else
 		return status ? 1 : 0;
 
@@ -381,7 +384,7 @@ static void write_status(acpi_handle handle, int out, int mask)
 	}
 
 	if (write_acpi_int(handle, NULL, out, NULL))
-		printk(ASUS_WARNING " write failed %x\n", mask);
+		pr_warning(" write failed %x\n", mask);
 }
 
 /* /sys/class/led handlers */
@@ -424,7 +427,7 @@ static int set_lcd_state(int value)
 					      NULL, NULL, NULL);
 
 		if (ACPI_FAILURE(status))
-			printk(ASUS_WARNING "Error switching LCD\n");
+			pr_warning("Error switching LCD\n");
 	}
 
 	write_status(NULL, lcd, LCD_ON);
@@ -448,7 +451,7 @@ static int read_brightness(struct backlight_device *bd)
 
 	rv = acpi_evaluate_integer(brightness_get_handle, NULL, NULL, &value);
 	if (ACPI_FAILURE(rv))
-		printk(ASUS_WARNING "Error reading brightness\n");
+		pr_warning("Error reading brightness\n");
 
 	return value;
 }
@@ -461,7 +464,7 @@ static int set_brightness(struct backlight_device *bd, int value)
 	/* 0 <= value <= 15 */
 
 	if (write_acpi_int(brightness_set_handle, NULL, value, NULL)) {
-		printk(ASUS_WARNING "Error changing brightness\n");
+		pr_warning("Error changing brightness\n");
 		ret = -EIO;
 	}
 
@@ -591,7 +594,7 @@ static ssize_t store_ledd(struct device *dev, struct device_attribute *attr,
 	rv = parse_arg(buf, count, &value);
 	if (rv > 0) {
 		if (write_acpi_int(ledd_set_handle, NULL, value, NULL))
-			printk(ASUS_WARNING "LED display write failed\n");
+			pr_warning("LED display write failed\n");
 		else
 			hotk->ledd_status = (u32) value;
 	}
@@ -636,7 +639,7 @@ static void set_display(int value)
 {
 	/* no sanity check needed for now */
 	if (write_acpi_int(display_set_handle, NULL, value, NULL))
-		printk(ASUS_WARNING "Error setting display\n");
+		pr_warning("Error setting display\n");
 	return;
 }
 
@@ -651,7 +654,7 @@ static int read_display(void)
 		rv = acpi_evaluate_integer(display_get_handle, NULL,
 					   NULL, &value);
 		if (ACPI_FAILURE(rv))
-			printk(ASUS_WARNING "Error reading display status\n");
+			pr_warning("Error reading display status\n");
 	}
 
 	value &= 0x0F;		/* needed for some models, shouldn't hurt others */
@@ -693,7 +696,7 @@ static ssize_t store_disp(struct device *dev, struct device_attribute *attr,
 static void set_light_sens_switch(int value)
 {
 	if (write_acpi_int(ls_switch_handle, NULL, value, NULL))
-		printk(ASUS_WARNING "Error setting light sensor switch\n");
+		pr_warning("Error setting light sensor switch\n");
 	hotk->light_switch = value;
 }
 
@@ -718,7 +721,7 @@ static ssize_t store_lssw(struct device *dev, struct device_attribute *attr,
 static void set_light_sens_level(int value)
 {
 	if (write_acpi_int(ls_level_handle, NULL, value, NULL))
-		printk(ASUS_WARNING "Error setting light sensor level\n");
+		pr_warning("Error setting light sensor level\n");
 	hotk->light_level = value;
 }
 
@@ -979,11 +982,11 @@ static int asus_hotk_get_info(void)
 	 */
 	status = acpi_get_table(ACPI_SIG_DSDT, 1, &asus_info);
 	if (ACPI_FAILURE(status))
-		printk(ASUS_WARNING "Couldn't get the DSDT table header\n");
+		pr_warning("Couldn't get the DSDT table header\n");
 
 	/* We have to write 0 on init this far for all ASUS models */
 	if (write_acpi_int(hotk->handle, "INIT", 0, &buffer)) {
-		printk(ASUS_ERR "Hotkey initialization failed\n");
+		pr_err("Hotkey initialization failed\n");
 		return -ENODEV;
 	}
 
@@ -991,9 +994,9 @@ static int asus_hotk_get_info(void)
 	status =
 	    acpi_evaluate_integer(hotk->handle, "BSTS", NULL, &bsts_result);
 	if (ACPI_FAILURE(status))
-		printk(ASUS_WARNING "Error calling BSTS\n");
+		pr_warning("Error calling BSTS\n");
 	else if (bsts_result)
-		printk(ASUS_NOTICE "BSTS called, 0x%02x returned\n",
+		pr_notice("BSTS called, 0x%02x returned\n",
 		       (uint) bsts_result);
 
 	/* This too ... */
@@ -1024,7 +1027,7 @@ static int asus_hotk_get_info(void)
 		return -ENOMEM;
 
 	if (*string)
-		printk(ASUS_NOTICE "  %s model detected\n", string);
+		pr_notice("  %s model detected\n", string);
 
 	ASUS_HANDLE_INIT(mled_set);
 	ASUS_HANDLE_INIT(tled_set);
@@ -1081,7 +1084,7 @@ static int asus_input_init(void)
 
 	hotk->inputdev = input_allocate_device();
 	if (!hotk->inputdev) {
-		printk(ASUS_INFO "Unable to allocate input device\n");
+		pr_info("Unable to allocate input device\n");
 		return 0;
 	}
 	hotk->inputdev->name = "Asus Laptop extra buttons";
@@ -1100,7 +1103,7 @@ static int asus_input_init(void)
 	}
 	result = input_register_device(hotk->inputdev);
 	if (result) {
-		printk(ASUS_INFO "Unable to register input device\n");
+		pr_info("Unable to register input device\n");
 		input_free_device(hotk->inputdev);
 	}
 	return result;
@@ -1117,7 +1120,7 @@ static int asus_hotk_check(void)
 	if (hotk->device->status.present) {
 		result = asus_hotk_get_info();
 	} else {
-		printk(ASUS_ERR "Hotkey device not present, aborting\n");
+		pr_err("Hotkey device not present, aborting\n");
 		return -EINVAL;
 	}
 
@@ -1133,7 +1136,7 @@ static int asus_hotk_add(struct acpi_device *device)
 	if (!device)
 		return -EINVAL;
 
-	printk(ASUS_NOTICE "Asus Laptop Support version %s\n",
+	pr_notice("Asus Laptop Support version %s\n",
 	       ASUS_LAPTOP_VERSION);
 
 	hotk = kzalloc(sizeof(struct asus_hotk), GFP_KERNEL);
@@ -1247,8 +1250,7 @@ static int asus_backlight_init(struct device *dev)
 		bd = backlight_device_register(ASUS_HOTK_FILE, dev,
 					       NULL, &asusbl_ops);
 		if (IS_ERR(bd)) {
-			printk(ASUS_ERR
-			       "Could not register asus backlight device\n");
+			pr_err("Could not register asus backlight device\n");
 			asus_backlight_device = NULL;
 			return PTR_ERR(bd);
 		}
@@ -1375,7 +1377,7 @@ static int __init asus_laptop_init(void)
 		if (result)
 			goto fail_backlight;
 	} else
-		printk(ASUS_INFO "Brightness ignored, must be controlled by "
+		pr_info("Brightness ignored, must be controlled by "
 		       "ACPI video driver\n");
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 21ab01e2fcbfcc0d1faba2b7336b3c0f7f3c1ac8 Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Fri, 19 Jun 2009 14:52:11 +0200
Subject: asus-laptop: remove EXPERIMENTAL dependency

Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig
index 13226e4ad234..7232fe7104aa 100644
--- a/drivers/platform/x86/Kconfig
+++ b/drivers/platform/x86/Kconfig
@@ -54,7 +54,7 @@ config ACERHDF
 config ASUS_LAPTOP
 	tristate "Asus Laptop Extras"
 	depends on ACPI
-	depends on EXPERIMENTAL && !ACPI_ASUS
+	depends on !ACPI_ASUS
 	select LEDS_CLASS
 	select NEW_LEDS
 	select BACKLIGHT_CLASS_DEVICE
-- 
cgit v1.2.3-59-g8ed1b


From f92e93eb5f4d56d73215f089580d53597bacd468 Mon Sep 17 00:00:00 2001
From: Jerome Glisse <jglisse@redhat.com>
Date: Mon, 22 Jun 2009 18:15:58 +0200
Subject: drm/radeon: fix radeon kms framebuffer device

smem.start is a physical address which kernel can remap to access
video memory of the fb buffer. We now pin the fb buffer into vram
by doing so we are loosing vram but fbdev need to be reworked to
allow change in framebuffer address.

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@linux.ie>
---
 drivers/gpu/drm/radeon/radeon_device.c |  4 ----
 drivers/gpu/drm/radeon/radeon_fb.c     | 27 +++++++++++++++++++--------
 drivers/gpu/drm/radeon/radeon_object.c | 30 ------------------------------
 3 files changed, 19 insertions(+), 42 deletions(-)

diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index f30aa7274a54..3f48a57531b5 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -496,7 +496,6 @@ int radeon_device_init(struct radeon_device *rdev,
 	radeon_errata(rdev);
 	/* Initialize scratch registers */
 	radeon_scratch_init(rdev);
-
 	/* TODO: disable VGA need to use VGA request */
 	/* BIOS*/
 	if (!radeon_get_bios(rdev)) {
@@ -604,9 +603,6 @@ int radeon_device_init(struct radeon_device *rdev,
 	if (r) {
 		return r;
 	}
-	if (rdev->fbdev_rfb && rdev->fbdev_rfb->obj) {
-		rdev->fbdev_robj = rdev->fbdev_rfb->obj->driver_private;
-	}
 	if (!ret) {
 		DRM_INFO("radeon: kernel modesetting successfully initialized.\n");
 	}
diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c
index fa86d398945e..09987089193e 100644
--- a/drivers/gpu/drm/radeon/radeon_fb.c
+++ b/drivers/gpu/drm/radeon/radeon_fb.c
@@ -478,14 +478,16 @@ int radeonfb_create(struct radeon_device *rdev,
 {
 	struct fb_info *info;
 	struct radeon_fb_device *rfbdev;
-	struct drm_framebuffer *fb;
+	struct drm_framebuffer *fb = NULL;
 	struct radeon_framebuffer *rfb;
 	struct drm_mode_fb_cmd mode_cmd;
 	struct drm_gem_object *gobj = NULL;
 	struct radeon_object *robj = NULL;
 	struct device *device = &rdev->pdev->dev;
 	int size, aligned_size, ret;
+	u64 fb_gpuaddr;
 	void *fbptr = NULL;
+	unsigned long tmp;
 
 	mode_cmd.width = surface_width;
 	mode_cmd.height = surface_height;
@@ -498,11 +500,12 @@ int radeonfb_create(struct radeon_device *rdev,
 	aligned_size = ALIGN(size, PAGE_SIZE);
 
 	ret = radeon_gem_object_create(rdev, aligned_size, 0,
-				       RADEON_GEM_DOMAIN_VRAM,
-				       false, ttm_bo_type_kernel,
-				       false, &gobj);
+			RADEON_GEM_DOMAIN_VRAM,
+			false, ttm_bo_type_kernel,
+			false, &gobj);
 	if (ret) {
-		printk(KERN_ERR "failed to allocate framebuffer\n");
+		printk(KERN_ERR "failed to allocate framebuffer (%d %d)\n",
+		       surface_width, surface_height);
 		ret = -ENOMEM;
 		goto out;
 	}
@@ -515,12 +518,19 @@ int radeonfb_create(struct radeon_device *rdev,
 		ret = -ENOMEM;
 		goto out_unref;
 	}
+	ret = radeon_object_pin(robj, RADEON_GEM_DOMAIN_VRAM, &fb_gpuaddr);
+	if (ret) {
+		printk(KERN_ERR "failed to pin framebuffer\n");
+		ret = -ENOMEM;
+		goto out_unref;
+	}
 
 	list_add(&fb->filp_head, &rdev->ddev->mode_config.fb_kernel_list);
 
 	rfb = to_radeon_framebuffer(fb);
 	*rfb_p = rfb;
 	rdev->fbdev_rfb = rfb;
+	rdev->fbdev_robj = robj;
 
 	info = framebuffer_alloc(sizeof(struct radeon_fb_device), device);
 	if (info == NULL) {
@@ -546,8 +556,8 @@ int radeonfb_create(struct radeon_device *rdev,
 	info->flags = FBINFO_DEFAULT;
 	info->fbops = &radeonfb_ops;
 	info->fix.line_length = fb->pitch;
-	info->screen_base = fbptr;
-	info->fix.smem_start = (unsigned long)fbptr;
+	tmp = fb_gpuaddr - rdev->mc.vram_location;
+	info->fix.smem_start = rdev->mc.aper_base + tmp;
 	info->fix.smem_len = size;
 	info->screen_base = fbptr;
 	info->screen_size = size;
@@ -644,7 +654,7 @@ out_unref:
 	if (robj) {
 		radeon_object_kunmap(robj);
 	}
-	if (ret) {
+	if (fb && ret) {
 		list_del(&fb->filp_head);
 		drm_gem_object_unreference(gobj);
 		drm_framebuffer_cleanup(fb);
@@ -813,6 +823,7 @@ int radeonfb_remove(struct drm_device *dev, struct drm_framebuffer *fb)
 		robj = rfb->obj->driver_private;
 		unregister_framebuffer(info);
 		radeon_object_kunmap(robj);
+		radeon_object_unpin(robj);
 		framebuffer_release(info);
 	}
 
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index 983e8df5e000..bac0d06c52ac 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -223,7 +223,6 @@ int radeon_object_pin(struct radeon_object *robj, uint32_t domain,
 {
 	uint32_t flags;
 	uint32_t tmp;
-	void *fbptr;
 	int r;
 
 	flags = radeon_object_flags_from_domain(domain);
@@ -242,10 +241,6 @@ int radeon_object_pin(struct radeon_object *robj, uint32_t domain,
 		DRM_ERROR("radeon: failed to reserve object for pinning it.\n");
 		return r;
 	}
-	if (robj->rdev->fbdev_robj == robj) {
-		mutex_lock(&robj->rdev->fbdev_info->lock);
-		radeon_object_kunmap(robj);
-	}
 	tmp = robj->tobj.mem.placement;
 	ttm_flag_masked(&tmp, flags, TTM_PL_MASK_MEM);
 	robj->tobj.proposed_placement = tmp | TTM_PL_FLAG_NO_EVICT | TTM_PL_MASK_CACHING;
@@ -261,23 +256,12 @@ int radeon_object_pin(struct radeon_object *robj, uint32_t domain,
 		DRM_ERROR("radeon: failed to pin object.\n");
 	}
 	radeon_object_unreserve(robj);
-	if (robj->rdev->fbdev_robj == robj) {
-		if (!r) {
-			r = radeon_object_kmap(robj, &fbptr);
-		}
-		if (!r) {
-			robj->rdev->fbdev_info->screen_base = fbptr;
-			robj->rdev->fbdev_info->fix.smem_start = (unsigned long)fbptr;
-		}
-		mutex_unlock(&robj->rdev->fbdev_info->lock);
-	}
 	return r;
 }
 
 void radeon_object_unpin(struct radeon_object *robj)
 {
 	uint32_t flags;
-	void *fbptr;
 	int r;
 
 	spin_lock(&robj->tobj.lock);
@@ -297,10 +281,6 @@ void radeon_object_unpin(struct radeon_object *robj)
 		DRM_ERROR("radeon: failed to reserve object for unpinning it.\n");
 		return;
 	}
-	if (robj->rdev->fbdev_robj == robj) {
-		mutex_lock(&robj->rdev->fbdev_info->lock);
-		radeon_object_kunmap(robj);
-	}
 	flags = robj->tobj.mem.placement;
 	robj->tobj.proposed_placement = flags & ~TTM_PL_FLAG_NO_EVICT;
 	r = ttm_buffer_object_validate(&robj->tobj,
@@ -310,16 +290,6 @@ void radeon_object_unpin(struct radeon_object *robj)
 		DRM_ERROR("radeon: failed to unpin buffer.\n");
 	}
 	radeon_object_unreserve(robj);
-	if (robj->rdev->fbdev_robj == robj) {
-		if (!r) {
-			r = radeon_object_kmap(robj, &fbptr);
-		}
-		if (!r) {
-			robj->rdev->fbdev_info->screen_base = fbptr;
-			robj->rdev->fbdev_info->fix.smem_start = (unsigned long)fbptr;
-		}
-		mutex_unlock(&robj->rdev->fbdev_info->lock);
-	}
 }
 
 int radeon_object_wait(struct radeon_object *robj)
-- 
cgit v1.2.3-59-g8ed1b


From 696d4df1dbfe0b054e94c1990b49c1727ffc1ff0 Mon Sep 17 00:00:00 2001
From: Michel Dänzer <daenzer@vmware.com>
Date: Tue, 23 Jun 2009 16:12:53 +0200
Subject: drm/radeon: Don't initialize acceleration related fields of struct
 fb_info.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Might lure userspace into trying silly things otherwise.

Signed-off-by: Michel Dänzer <daenzer@vmware.com>
Signed-off-by: Dave Airlie <airlied@linux.ie>
---
 drivers/gpu/drm/radeon/radeon_fb.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c
index 09987089193e..9e8f191eb64a 100644
--- a/drivers/gpu/drm/radeon/radeon_fb.c
+++ b/drivers/gpu/drm/radeon/radeon_fb.c
@@ -551,7 +551,7 @@ int radeonfb_create(struct radeon_device *rdev,
 	info->fix.xpanstep = 1; /* doing it in hw */
 	info->fix.ypanstep = 1; /* doing it in hw */
 	info->fix.ywrapstep = 0;
-	info->fix.accel = FB_ACCEL_I830;
+	info->fix.accel = FB_ACCEL_NONE;
 	info->fix.type_aux = 0;
 	info->flags = FBINFO_DEFAULT;
 	info->fbops = &radeonfb_ops;
@@ -572,8 +572,8 @@ int radeonfb_create(struct radeon_device *rdev,
 	info->var.width = -1;
 	info->var.xres = fb_width;
 	info->var.yres = fb_height;
-	info->fix.mmio_start = pci_resource_start(rdev->pdev, 2);
-	info->fix.mmio_len = pci_resource_len(rdev->pdev, 2);
+	info->fix.mmio_start = 0;
+	info->fix.mmio_len = 0;
 	info->pixmap.size = 64*1024;
 	info->pixmap.buf_align = 8;
 	info->pixmap.access_align = 32;
-- 
cgit v1.2.3-59-g8ed1b


From b1e3a6d1c4d0ac75ad8289bcfd69efcc9b1bc6e5 Mon Sep 17 00:00:00 2001
From: Michel Dänzer <daenzer@vmware.com>
Date: Tue, 23 Jun 2009 16:12:54 +0200
Subject: drm/radeon: Clear surface registers at initialization time.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some PowerMac firmwares set up a tiling surface at the beginning of VRAM
which messes us up otherwise.

Signed-off-by: Michel Dänzer <daenzer@vmware.com>
Signed-off-by: Dave Airlie <airlied@linux.ie>
---
 drivers/gpu/drm/radeon/radeon_device.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 3f48a57531b5..f97563db4e59 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -34,6 +34,23 @@
 #include "radeon_asic.h"
 #include "atom.h"
 
+/*
+ * Clear GPU surface registers.
+ */
+static void radeon_surface_init(struct radeon_device *rdev)
+{
+	/* FIXME: check this out */
+	if (rdev->family < CHIP_R600) {
+		int i;
+
+		for (i = 0; i < 8; i++) {
+			WREG32(RADEON_SURFACE0_INFO +
+			       i * (RADEON_SURFACE1_INFO - RADEON_SURFACE0_INFO),
+			       0);
+		}
+	}
+}
+
 /*
  * GPU scratch registers helpers function.
  */
@@ -496,6 +513,9 @@ int radeon_device_init(struct radeon_device *rdev,
 	radeon_errata(rdev);
 	/* Initialize scratch registers */
 	radeon_scratch_init(rdev);
+	/* Initialize surface registers */
+	radeon_surface_init(rdev);
+
 	/* TODO: disable VGA need to use VGA request */
 	/* BIOS*/
 	if (!radeon_get_bios(rdev)) {
-- 
cgit v1.2.3-59-g8ed1b


From e14cbee401cd00779a5267128371506b22c77bc9 Mon Sep 17 00:00:00 2001
From: Michel Dänzer <michel@daenzer.net>
Date: Tue, 23 Jun 2009 12:36:32 +0200
Subject: drm: Fix shifts which were miscalculated when converting from
 bitfields.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Looks like I managed to mess up most shifts when converting from bitfields. :(

The patch below works on my Thinkpad T500 (as well as on my PowerBook,
where the previous change worked as well, maybe out of luck...). I'd
appreciate more testing and eyes looking over it though.

Signed-off-by: Michel Dänzer <daenzer@vmware.com>
Tested-by: Michael Pyne <mpyne@kde.org>
Signed-off-by: Dave Airlie <airlied@linux.ie>
---
 drivers/gpu/drm/drm_edid.c | 12 ++++++------
 include/drm/drm_edid.h     | 38 +++++++++++++++++++-------------------
 2 files changed, 25 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 7d0835226f6e..80cc6d06d61b 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -294,10 +294,10 @@ static struct drm_display_mode *drm_mode_detailed(struct drm_device *dev,
 	unsigned vactive = (pt->vactive_vblank_hi & 0xf0) << 4 | pt->vactive_lo;
 	unsigned hblank = (pt->hactive_hblank_hi & 0xf) << 8 | pt->hblank_lo;
 	unsigned vblank = (pt->vactive_vblank_hi & 0xf) << 8 | pt->vblank_lo;
-	unsigned hsync_offset = (pt->hsync_vsync_offset_pulse_width_hi & 0x3) << 8 | pt->hsync_offset_lo;
-	unsigned hsync_pulse_width = (pt->hsync_vsync_offset_pulse_width_hi & 0xc) << 6 | pt->hsync_pulse_width_lo;
-	unsigned vsync_offset = (pt->hsync_vsync_offset_pulse_width_hi & 0x30) | (pt->vsync_offset_pulse_width_lo & 0xf);
-	unsigned vsync_pulse_width = (pt->hsync_vsync_offset_pulse_width_hi & 0xc0) >> 2 | pt->vsync_offset_pulse_width_lo >> 4;
+	unsigned hsync_offset = (pt->hsync_vsync_offset_pulse_width_hi & 0xc0) << 2 | pt->hsync_offset_lo;
+	unsigned hsync_pulse_width = (pt->hsync_vsync_offset_pulse_width_hi & 0x30) << 4 | pt->hsync_pulse_width_lo;
+	unsigned vsync_offset = (pt->hsync_vsync_offset_pulse_width_hi & 0xc) >> 2 | pt->vsync_offset_pulse_width_lo >> 4;
+	unsigned vsync_pulse_width = (pt->hsync_vsync_offset_pulse_width_hi & 0x3) << 4 | (pt->vsync_offset_pulse_width_lo & 0xf);
 
 	/* ignore tiny modes */
 	if (hactive < 64 || vactive < 64)
@@ -347,8 +347,8 @@ static struct drm_display_mode *drm_mode_detailed(struct drm_device *dev,
 	mode->flags |= (pt->misc & DRM_EDID_PT_VSYNC_POSITIVE) ?
 		DRM_MODE_FLAG_PVSYNC : DRM_MODE_FLAG_NVSYNC;
 
-	mode->width_mm = pt->width_mm_lo | (pt->width_height_mm_hi & 0xf) << 8;
-	mode->height_mm = pt->height_mm_lo | (pt->width_height_mm_hi & 0xf0) << 4;
+	mode->width_mm = pt->width_mm_lo | (pt->width_height_mm_hi & 0xf0) << 4;
+	mode->height_mm = pt->height_mm_lo | (pt->width_height_mm_hi & 0xf) << 8;
 
 	if (quirks & EDID_QUIRK_DETAILED_IN_CM) {
 		mode->width_mm *= 10;
diff --git a/include/drm/drm_edid.h b/include/drm/drm_edid.h
index c263e4d71754..7d6c9a2dfcbb 100644
--- a/include/drm/drm_edid.h
+++ b/include/drm/drm_edid.h
@@ -35,11 +35,11 @@ struct est_timings {
 } __attribute__((packed));
 
 /* 00=16:10, 01=4:3, 10=5:4, 11=16:9 */
-#define EDID_TIMING_ASPECT_SHIFT 0
+#define EDID_TIMING_ASPECT_SHIFT 6
 #define EDID_TIMING_ASPECT_MASK  (0x3 << EDID_TIMING_ASPECT_SHIFT)
 
 /* need to add 60 */
-#define EDID_TIMING_VFREQ_SHIFT  2
+#define EDID_TIMING_VFREQ_SHIFT  0
 #define EDID_TIMING_VFREQ_MASK   (0x3f << EDID_TIMING_VFREQ_SHIFT)
 
 struct std_timing {
@@ -47,11 +47,11 @@ struct std_timing {
 	u8 vfreq_aspect;
 } __attribute__((packed));
 
-#define DRM_EDID_PT_HSYNC_POSITIVE (1 << 6)
-#define DRM_EDID_PT_VSYNC_POSITIVE (1 << 5)
+#define DRM_EDID_PT_HSYNC_POSITIVE (1 << 1)
+#define DRM_EDID_PT_VSYNC_POSITIVE (1 << 2)
 #define DRM_EDID_PT_SEPARATE_SYNC  (3 << 3)
-#define DRM_EDID_PT_STEREO         (1 << 2)
-#define DRM_EDID_PT_INTERLACED     (1 << 1)
+#define DRM_EDID_PT_STEREO         (1 << 5)
+#define DRM_EDID_PT_INTERLACED     (1 << 7)
 
 /* If detailed data is pixel timing */
 struct detailed_pixel_timing {
@@ -93,7 +93,7 @@ struct detailed_data_monitor_range {
 } __attribute__((packed));
 
 struct detailed_data_wpindex {
-	u8 white_xy_lo; /* Upper 2 bits each */
+	u8 white_yx_lo; /* Lower 2 bits each */
 	u8 white_x_hi;
 	u8 white_y_hi;
 	u8 gamma; /* need to divide by 100 then add 1 */
@@ -135,21 +135,21 @@ struct detailed_timing {
 	} data;
 } __attribute__((packed));
 
-#define DRM_EDID_INPUT_SERRATION_VSYNC (1 << 7)
-#define DRM_EDID_INPUT_SYNC_ON_GREEN   (1 << 5)
-#define DRM_EDID_INPUT_COMPOSITE_SYNC  (1 << 4)
+#define DRM_EDID_INPUT_SERRATION_VSYNC (1 << 0)
+#define DRM_EDID_INPUT_SYNC_ON_GREEN   (1 << 1)
+#define DRM_EDID_INPUT_COMPOSITE_SYNC  (1 << 2)
 #define DRM_EDID_INPUT_SEPARATE_SYNCS  (1 << 3)
-#define DRM_EDID_INPUT_BLANK_TO_BLACK  (1 << 2)
-#define DRM_EDID_INPUT_VIDEO_LEVEL     (3 << 1)
-#define DRM_EDID_INPUT_DIGITAL         (1 << 0) /* bits above must be zero if set */
+#define DRM_EDID_INPUT_BLANK_TO_BLACK  (1 << 4)
+#define DRM_EDID_INPUT_VIDEO_LEVEL     (3 << 5)
+#define DRM_EDID_INPUT_DIGITAL         (1 << 7) /* bits below must be zero if set */
 
-#define DRM_EDID_FEATURE_DEFAULT_GTF      (1 << 7)
-#define DRM_EDID_FEATURE_PREFERRED_TIMING (1 << 6)
-#define DRM_EDID_FEATURE_STANDARD_COLOR   (1 << 5)
+#define DRM_EDID_FEATURE_DEFAULT_GTF      (1 << 0)
+#define DRM_EDID_FEATURE_PREFERRED_TIMING (1 << 1)
+#define DRM_EDID_FEATURE_STANDARD_COLOR   (1 << 2)
 #define DRM_EDID_FEATURE_DISPLAY_TYPE     (3 << 3) /* 00=mono, 01=rgb, 10=non-rgb, 11=unknown */
-#define DRM_EDID_FEATURE_PM_ACTIVE_OFF    (1 << 2)
-#define DRM_EDID_FEATURE_PM_SUSPEND       (1 << 1)
-#define DRM_EDID_FEATURE_PM_STANDBY       (1 << 0)
+#define DRM_EDID_FEATURE_PM_ACTIVE_OFF    (1 << 5)
+#define DRM_EDID_FEATURE_PM_SUSPEND       (1 << 6)
+#define DRM_EDID_FEATURE_PM_STANDBY       (1 << 7)
 
 struct edid {
 	u8 header[8];
-- 
cgit v1.2.3-59-g8ed1b


From 176f613e60b63f2d77e6c69f036cfc754f3aaac6 Mon Sep 17 00:00:00 2001
From: Jerome Glisse <jglisse@redhat.com>
Date: Mon, 22 Jun 2009 18:16:13 +0200
Subject: drm/radeon: fix driver initialization order so radeon kms can be
 builtin

TTM need to be initialized before radeon if KMS is enabled otherwise
the kernel will crash hard.

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
Signed-off-by: Dave Airlie <airlied@linux.ie>
---
 drivers/gpu/drm/Makefile            | 2 +-
 drivers/gpu/drm/radeon/radeon_drv.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 4e89ab08b7b8..fe23f29f7cba 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -16,6 +16,7 @@ drm-y       :=	drm_auth.o drm_bufs.o drm_cache.o \
 drm-$(CONFIG_COMPAT) += drm_ioc32.o
 
 obj-$(CONFIG_DRM)	+= drm.o
+obj-$(CONFIG_DRM_TTM)	+= ttm/
 obj-$(CONFIG_DRM_TDFX)	+= tdfx/
 obj-$(CONFIG_DRM_R128)	+= r128/
 obj-$(CONFIG_DRM_RADEON)+= radeon/
@@ -26,4 +27,3 @@ obj-$(CONFIG_DRM_I915)  += i915/
 obj-$(CONFIG_DRM_SIS)   += sis/
 obj-$(CONFIG_DRM_SAVAGE)+= savage/
 obj-$(CONFIG_DRM_VIA)	+=via/
-obj-$(CONFIG_DRM_TTM)	+= ttm/
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index 09c9fb9f6210..84ba69f48784 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -345,7 +345,7 @@ static void __exit radeon_exit(void)
 	drm_exit(driver);
 }
 
-late_initcall(radeon_init);
+module_init(radeon_init);
 module_exit(radeon_exit);
 
 MODULE_AUTHOR(DRIVER_AUTHOR);
-- 
cgit v1.2.3-59-g8ed1b


From 8b169b5f1f46da8ece1ce7304cda7155fffe3892 Mon Sep 17 00:00:00 2001
From: Huang Weiyi <weiyi.huang@gmail.com>
Date: Wed, 24 Jun 2009 16:31:50 +1000
Subject: drm: remove unused #include <linux/version.h>'s

Remove unused #include <linux/version.h>('s) in
 drivers/gpu/drm/ttm/ttm_bo_util.c
 drivers/gpu/drm/ttm/ttm_bo_vm.c
 drivers/gpu/drm/ttm/ttm_tt.c

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
---
 drivers/gpu/drm/ttm/ttm_bo_util.c | 1 -
 drivers/gpu/drm/ttm/ttm_bo_vm.c   | 1 -
 drivers/gpu/drm/ttm/ttm_tt.c      | 1 -
 3 files changed, 3 deletions(-)

diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
index 517c84559633..bdec583901eb 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_util.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
@@ -34,7 +34,6 @@
 #include <linux/highmem.h>
 #include <linux/wait.h>
 #include <linux/vmalloc.h>
-#include <linux/version.h>
 #include <linux/module.h>
 
 void ttm_bo_free_old_node(struct ttm_buffer_object *bo)
diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c b/drivers/gpu/drm/ttm/ttm_bo_vm.c
index 27b146c54fbc..40b75032ea47 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
@@ -32,7 +32,6 @@
 #include <ttm/ttm_bo_driver.h>
 #include <ttm/ttm_placement.h>
 #include <linux/mm.h>
-#include <linux/version.h>
 #include <linux/rbtree.h>
 #include <linux/module.h>
 #include <linux/uaccess.h>
diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c
index 0331fa74cd3f..75dc8bd24592 100644
--- a/drivers/gpu/drm/ttm/ttm_tt.c
+++ b/drivers/gpu/drm/ttm/ttm_tt.c
@@ -28,7 +28,6 @@
  * Authors: Thomas Hellstrom <thellstrom-at-vmware-dot-com>
  */
 
-#include <linux/version.h>
 #include <linux/vmalloc.h>
 #include <linux/sched.h>
 #include <linux/highmem.h>
-- 
cgit v1.2.3-59-g8ed1b


From 7959ea254ed18faee41160b1c50b3c9664735967 Mon Sep 17 00:00:00 2001
From: Ooiwa Naohiro <nooiwa@miraclelinux.com>
Date: Wed, 24 Jun 2009 00:19:06 -0700
Subject: bnx2: Fix the behavior of ethtool when ONBOOT=no

I found a little bug.

When configure in ifcfg-eth* is ONBOOT=no,
the behavior of ethtool command is wrong.

    # grep ONBOOT /etc/sysconfig/network-scripts/ifcfg-eth2
    ONBOOT=no
    # ethtool eth2 | tail -n1
            Link detected: yes

I think "Link detected" should be "no".

Signed-off-by: Ooiwa Naohiro <nooiwa@miraclelinux.com>
Acked-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/bnx2.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c
index 38f1c3375d7f..b70cc99962fc 100644
--- a/drivers/net/bnx2.c
+++ b/drivers/net/bnx2.c
@@ -6825,6 +6825,14 @@ bnx2_nway_reset(struct net_device *dev)
 	return 0;
 }
 
+static u32
+bnx2_get_link(struct net_device *dev)
+{
+	struct bnx2 *bp = netdev_priv(dev);
+
+	return bp->link_up;
+}
+
 static int
 bnx2_get_eeprom_len(struct net_device *dev)
 {
@@ -7392,7 +7400,7 @@ static const struct ethtool_ops bnx2_ethtool_ops = {
 	.get_wol		= bnx2_get_wol,
 	.set_wol		= bnx2_set_wol,
 	.nway_reset		= bnx2_nway_reset,
-	.get_link		= ethtool_op_get_link,
+	.get_link		= bnx2_get_link,
 	.get_eeprom_len		= bnx2_get_eeprom_len,
 	.get_eeprom		= bnx2_get_eeprom,
 	.set_eeprom		= bnx2_set_eeprom,
-- 
cgit v1.2.3-59-g8ed1b


From ffc36c7610731115c77700dcc53901920361c235 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 23 Jun 2009 03:43:00 -0700
Subject: ide: fix handling of unexpected IRQs vs request_irq()

Add ide_host_enable_irqs() helper and use it in ide_host_register()
before registering ports.  Then remove no longer needed IRQ unmasking
from in init_irq().

This should fix the problem with "screaming" shared IRQ on the first
port (after request_irq() call while we have the unexpected IRQ pending
on the second port) which was uncovered by my rework of the serialized
interfaces support.

Reported-by: Frans Pop <elendil@planet.nl>
Tested-by: Frans Pop <elendil@planet.nl>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/ide-probe.c | 23 ++++++++++++++++++++---
 1 file changed, 20 insertions(+), 3 deletions(-)

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 51af4eea0d36..1bb106f6221a 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -818,6 +818,24 @@ static int ide_port_setup_devices(ide_hwif_t *hwif)
 	return j;
 }
 
+static void ide_host_enable_irqs(struct ide_host *host)
+{
+	ide_hwif_t *hwif;
+	int i;
+
+	ide_host_for_each_port(i, hwif, host) {
+		if (hwif == NULL)
+			continue;
+
+		/* clear any pending IRQs */
+		hwif->tp_ops->read_status(hwif);
+
+		/* unmask IRQs */
+		if (hwif->io_ports.ctl_addr)
+			hwif->tp_ops->write_devctl(hwif, ATA_DEVCTL_OBS);
+	}
+}
+
 /*
  * This routine sets up the IRQ for an IDE interface.
  */
@@ -831,9 +849,6 @@ static int init_irq (ide_hwif_t *hwif)
 	if (irq_handler == NULL)
 		irq_handler = ide_intr;
 
-	if (io_ports->ctl_addr)
-		hwif->tp_ops->write_devctl(hwif, ATA_DEVCTL_OBS);
-
 	if (request_irq(hwif->irq, irq_handler, sa, hwif->name, hwif))
 		goto out_up;
 
@@ -1404,6 +1419,8 @@ int ide_host_register(struct ide_host *host, const struct ide_port_info *d,
 			ide_port_tune_devices(hwif);
 	}
 
+	ide_host_enable_irqs(host);
+
 	ide_host_for_each_port(i, hwif, host) {
 		if (hwif == NULL)
 			continue;
-- 
cgit v1.2.3-59-g8ed1b


From af054ed0018f0a69f8ea6f7546cbf34385edf13b Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 23 Jun 2009 16:01:06 -0700
Subject: ide-cd: Don't warn on bogus block size unless it actually matters.

Frans Pop reported that his CDROM drive reports a blocksize of 2352,
and this causes new warnings due to commit
e8e7b9eb11c34ee18bde8b7011af41938d1ad667 ("ide-cd: fix oops when using
growisofs").

What we're trying to do is make sure that "blocklen >> SECTOR_BITS"
is something the block layer won't choke on.

And for Frans' case "2352 >> SECTOR_BITS" is equal to
"2048 >> SECTOR_BITS", and thats "4".

So warning in this case gives no real benefit.

Reported-by: Frans Pop <elendil@planet.nl>
Tested-by: Frans Pop <elendil@planet.nl>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/ide-cd.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 4a19686fcfe9..a9a1bfb14e7c 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -876,9 +876,12 @@ static int cdrom_read_capacity(ide_drive_t *drive, unsigned long *capacity,
 		return stat;
 
 	/*
-	 * Sanity check the given block size
+	 * Sanity check the given block size, in so far as making
+	 * sure the sectors_per_frame we give to the caller won't
+	 * end up being bogus.
 	 */
 	blocklen = be32_to_cpu(capbuf.blocklen);
+	blocklen = (blocklen >> SECTOR_BITS) << SECTOR_BITS;
 	switch (blocklen) {
 	case 512:
 	case 1024:
-- 
cgit v1.2.3-59-g8ed1b


From d9ae62433e46909fc9e7d97ce74202c2851667b8 Mon Sep 17 00:00:00 2001
From: Frans Pop <elendil@planet.nl>
Date: Tue, 23 Jun 2009 16:02:58 -0700
Subject: ide-cd: Improve "weird block size" error message

Currently the error gets repeated too frequently, for example each
time HAL polls the device when a disc is present. Avoid that by using
printk_once instead of printk.

Also join the error and corrective action messages into a single line.

Signed-off-by: Frans Pop <elendil@planet.nl>
Acked-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/ide-cd.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index a9a1bfb14e7c..f0ede5953af8 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -889,10 +889,9 @@ static int cdrom_read_capacity(ide_drive_t *drive, unsigned long *capacity,
 	case 4096:
 		break;
 	default:
-		printk(KERN_ERR PFX "%s: weird block size %u\n",
+		printk_once(KERN_ERR PFX "%s: weird block size %u; "
+				"setting default block size to 2048\n",
 				drive->name, blocklen);
-		printk(KERN_ERR PFX "%s: default to 2kb block size\n",
-				drive->name);
 		blocklen = 2048;
 		break;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 346c17a6cf60375323adfaa4b8a9d841049f890e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Mon, 22 Jun 2009 07:38:26 +0000
Subject: ide: relax DMA info validity checking

There are some broken devices that report multiple DMA xfer modes
enabled at once (ATA spec doesn't allow it) but otherwise work fine
with DMA so just delete ide_id_dma_bug().

[ As discovered by detective work by Frans and Bart, due to how
  handling of the ID block was handled before commit c419993
  ("ide-iops: only clear DMA words on setting DMA mode") this
  check was always seeing zeros in the fields or other similar
  garbage.  Therefore this check wasn't actually checking anything.
  Now that the tests actually check the real bits, all we see are
  devices that trigger the check yet work perfectly fine, therefore
  killing this useless check is the best thing to do. -DaveM ]

Reported-by: Frans Pop <elendil@planet.nl>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/ide-dma.c  | 21 ---------------------
 drivers/ide/ide-iops.c |  3 ---
 include/linux/ide.h    |  2 --
 3 files changed, 26 deletions(-)

diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index 219e6fb78dc6..ee58c88dee5a 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -361,9 +361,6 @@ static int ide_tune_dma(ide_drive_t *drive)
 	if (__ide_dma_bad_drive(drive))
 		return 0;
 
-	if (ide_id_dma_bug(drive))
-		return 0;
-
 	if (hwif->host_flags & IDE_HFLAG_TRUST_BIOS_FOR_DMA)
 		return config_drive_for_dma(drive);
 
@@ -394,24 +391,6 @@ static int ide_dma_check(ide_drive_t *drive)
 	return -1;
 }
 
-int ide_id_dma_bug(ide_drive_t *drive)
-{
-	u16 *id = drive->id;
-
-	if (id[ATA_ID_FIELD_VALID] & 4) {
-		if ((id[ATA_ID_UDMA_MODES] >> 8) &&
-		    (id[ATA_ID_MWDMA_MODES] >> 8))
-			goto err_out;
-	} else if ((id[ATA_ID_MWDMA_MODES] >> 8) &&
-		   (id[ATA_ID_SWDMA_MODES] >> 8))
-		goto err_out;
-
-	return 0;
-err_out:
-	printk(KERN_ERR "%s: bad DMA info in identify block\n", drive->name);
-	return 1;
-}
-
 int ide_set_dma(ide_drive_t *drive)
 {
 	int rc;
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index fa047150a1c6..917186ec4966 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -329,9 +329,6 @@ int ide_driveid_update(ide_drive_t *drive)
 
 	kfree(id);
 
-	if ((drive->dev_flags & IDE_DFLAG_USING_DMA) && ide_id_dma_bug(drive))
-		ide_dma_off(drive);
-
 	return 1;
 out_err:
 	if (rc == 2)
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 95c6e00a72e8..cf1f3888067c 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1361,7 +1361,6 @@ int ide_in_drive_list(u16 *, const struct drive_list_entry *);
 #ifdef CONFIG_BLK_DEV_IDEDMA
 int ide_dma_good_drive(ide_drive_t *);
 int __ide_dma_bad_drive(ide_drive_t *);
-int ide_id_dma_bug(ide_drive_t *);
 
 u8 ide_find_dma_mode(ide_drive_t *, u8);
 
@@ -1402,7 +1401,6 @@ void ide_dma_lost_irq(ide_drive_t *);
 ide_startstop_t ide_dma_timeout_retry(ide_drive_t *, int);
 
 #else
-static inline int ide_id_dma_bug(ide_drive_t *drive) { return 0; }
 static inline u8 ide_find_dma_mode(ide_drive_t *drive, u8 speed) { return 0; }
 static inline u8 ide_max_dma_mode(ide_drive_t *drive) { return 0; }
 static inline void ide_dma_off_quietly(ide_drive_t *drive) { ; }
-- 
cgit v1.2.3-59-g8ed1b


From ba9413bd284e79ea43b0ae406a7a29526aaf82b3 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 23 Jun 2009 16:11:10 -0700
Subject: ide: add QUANTUM FIREBALLct20 30 with firmware APL.090 to ivb_list[]

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/ide-iops.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 917186ec4966..2892b242bbe1 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -210,6 +210,7 @@ EXPORT_SYMBOL_GPL(ide_in_drive_list);
  */
 static const struct drive_list_entry ivb_list[] = {
 	{ "QUANTUM FIREBALLlct10 05"	, "A03.0900"	},
+	{ "QUANTUM FIREBALLlct20 30"	, "APL.0900"	},
 	{ "TSSTcorp CDDVDW SH-S202J"	, "SB00"	},
 	{ "TSSTcorp CDDVDW SH-S202J"	, "SB01"	},
 	{ "TSSTcorp CDDVDW SH-S202N"	, "SB00"	},
-- 
cgit v1.2.3-59-g8ed1b


From a1317f714af7aed60ddc182d0122477cbe36ee9b Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 23 Jun 2009 23:52:17 -0700
Subject: ide: improve handling of Power Management requests

Make hwif->rq point to PM request during PM sequence and do not allow
any other types of requests to slip in (the old comment was never correct
as there should be no such requests generated during PM sequence).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/ide-io.c | 54 +++++++++++++++++++++-------------------------------
 1 file changed, 22 insertions(+), 32 deletions(-)

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 1059f809b809..93b7886a2d6e 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -476,10 +476,14 @@ void do_ide_request(struct request_queue *q)
 
 	if (!ide_lock_port(hwif)) {
 		ide_hwif_t *prev_port;
-
-		WARN_ON_ONCE(hwif->rq);
 repeat:
 		prev_port = hwif->host->cur_port;
+
+		if (drive->dev_flags & IDE_DFLAG_BLOCKED)
+			rq = hwif->rq;
+		else
+			WARN_ON_ONCE(hwif->rq);
+
 		if (drive->dev_flags & IDE_DFLAG_SLEEPING &&
 		    time_after(drive->sleep, jiffies)) {
 			ide_unlock_port(hwif);
@@ -506,43 +510,29 @@ repeat:
 		hwif->cur_dev = drive;
 		drive->dev_flags &= ~(IDE_DFLAG_SLEEPING | IDE_DFLAG_PARKED);
 
-		spin_unlock_irq(&hwif->lock);
-		spin_lock_irq(q->queue_lock);
-		/*
-		 * we know that the queue isn't empty, but this can happen
-		 * if the q->prep_rq_fn() decides to kill a request
-		 */
-		if (!rq)
+		if (rq == NULL) {
+			spin_unlock_irq(&hwif->lock);
+			spin_lock_irq(q->queue_lock);
+			/*
+			 * we know that the queue isn't empty, but this can
+			 * happen if ->prep_rq_fn() decides to kill a request
+			 */
 			rq = blk_fetch_request(drive->queue);
+			spin_unlock_irq(q->queue_lock);
+			spin_lock_irq(&hwif->lock);
 
-		spin_unlock_irq(q->queue_lock);
-		spin_lock_irq(&hwif->lock);
-
-		if (!rq) {
-			ide_unlock_port(hwif);
-			goto out;
+			if (rq == NULL) {
+				ide_unlock_port(hwif);
+				goto out;
+			}
 		}
 
 		/*
 		 * Sanity: don't accept a request that isn't a PM request
-		 * if we are currently power managed. This is very important as
-		 * blk_stop_queue() doesn't prevent the blk_fetch_request()
-		 * above to return us whatever is in the queue. Since we call
-		 * ide_do_request() ourselves, we end up taking requests while
-		 * the queue is blocked...
-		 * 
-		 * We let requests forced at head of queue with ide-preempt
-		 * though. I hope that doesn't happen too much, hopefully not
-		 * unless the subdriver triggers such a thing in its own PM
-		 * state machine.
+		 * if we are currently power managed.
 		 */
-		if ((drive->dev_flags & IDE_DFLAG_BLOCKED) &&
-		    blk_pm_request(rq) == 0 &&
-		    (rq->cmd_flags & REQ_PREEMPT) == 0) {
-			/* there should be no pending command at this point */
-			ide_unlock_port(hwif);
-			goto plug_device;
-		}
+		BUG_ON((drive->dev_flags & IDE_DFLAG_BLOCKED) &&
+		       blk_pm_request(rq) == 0);
 
 		hwif->rq = rq;
 
-- 
cgit v1.2.3-59-g8ed1b


From f7679dabfaf69840b000d238a020cee7157aca17 Mon Sep 17 00:00:00 2001
From: Roel Kluin <roel.kluin@gmail.com>
Date: Mon, 22 Jun 2009 18:42:33 +0200
Subject: perf_counter tools: Fix strbuf_fread() error path handling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

size_t res cannot be less than 0 - fread returns 0 on error.

[ Updated by: René Scharfe <rene.scharfe@lsrfire.ath.cx> ]

Reported-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Junio C Hamano <gitster@pobox.com>
LKML-Reference: <4A3FB479.2090902@lsrfire.ath.cx>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/util/strbuf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/util/strbuf.c b/tools/perf/util/strbuf.c
index eaba09306802..464e7ca898cf 100644
--- a/tools/perf/util/strbuf.c
+++ b/tools/perf/util/strbuf.c
@@ -259,7 +259,7 @@ size_t strbuf_fread(struct strbuf *sb, size_t size, FILE *f)
 	res = fread(sb->buf + sb->len, 1, size, f);
 	if (res > 0)
 		strbuf_setlen(sb, sb->len + res);
-	else if (res < 0 && oldalloc == 0)
+	else if (oldalloc == 0)
 		strbuf_release(sb);
 	return res;
 }
-- 
cgit v1.2.3-59-g8ed1b


From c14dab5c0782ef632742963a66276a195418a63c Mon Sep 17 00:00:00 2001
From: Yong Wang <yong.y.wang@linux.intel.com>
Date: Wed, 24 Jun 2009 10:13:24 +0800
Subject: perf_counter, x86: Set global control MSR correctly

Previous code made an assumption that the power on value of global
control MSR has enabled all fixed and general purpose counters properly.

However, this is not the case for certain Intel processors, such as
Atom - and it might also be firmware dependent.

Each enable bit in IA32_PERF_GLOBAL_CTRL is AND'ed with the
enable bits for all privilege levels in the respective IA32_PERFEVTSELx
or IA32_PERF_FIXED_CTR_CTRL MSRs to start/stop the counting of
respective counters. Counting is enabled if the AND'ed results is true;
counting is disabled when the result is false.

The end result is that all fixed counters are always disabled on Atom
processors because the assumption is just invalid.

Fix this by not initializing the ctrl-mask out of the global MSR,
but setting it to perf_counter_mask.

Reported-by: Stephane Eranian <eranian@googlemail.com>
Signed-off-by: Yong Wang <yong.y.wang@intel.com>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <20090624021324.GA2788@ywang-moblin2.bj.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/perf_counter.c | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c
index 22eb3a1d4f9c..a310d19faca3 100644
--- a/arch/x86/kernel/cpu/perf_counter.c
+++ b/arch/x86/kernel/cpu/perf_counter.c
@@ -969,13 +969,6 @@ fixed_mode_idx(struct perf_counter *counter, struct hw_perf_counter *hwc)
 	if (!x86_pmu.num_counters_fixed)
 		return -1;
 
-	/*
-	 * Quirk, IA32_FIXED_CTRs do not work on current Atom processors:
-	 */
-	if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL &&
-					boot_cpu_data.x86_model == 28)
-		return -1;
-
 	event = hwc->config & ARCH_PERFMON_EVENT_MASK;
 
 	if (unlikely(event == x86_pmu.event_map(PERF_COUNT_HW_INSTRUCTIONS)))
@@ -1428,8 +1421,6 @@ static int intel_pmu_init(void)
 	 */
 	x86_pmu.num_counters_fixed	= max((int)edx.split.num_counters_fixed, 3);
 
-	rdmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl);
-
 	/*
 	 * Install the hw-cache-events table:
 	 */
@@ -1514,6 +1505,7 @@ void __init init_hw_perf_counters(void)
 
 	perf_counter_mask |=
 		((1LL << x86_pmu.num_counters_fixed)-1) << X86_PMC_IDX_FIXED;
+	x86_pmu.intel_ctrl = perf_counter_mask;
 
 	perf_counters_lapic_init();
 	register_die_notifier(&perf_counter_nmi_notifier);
-- 
cgit v1.2.3-59-g8ed1b


From e1c7e2a6e67fe9db19dd15e71614526a31b5fdb1 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Wed, 24 Jun 2009 09:52:29 +0800
Subject: tracing/events: Don't increment @pos in s_start()

While testing syscall tracepoints posted by Jason, I found 3 entries
were missing when reading available_events. The output size of
available_events is < 4 pages, which means we lost 1 entry per page.

The cause is, it's wrong to increment @pos in s_start().

Actually there's another bug here -- reading avaiable_events/set_events
can race with module unload:

  # cat available_events               |
      s_start()                        |
      s_stop()                         |
                                       | # rmmod foo.ko
      s_start()                        |
        call = list_entry(m->private)  |

@call might be freed and accessing it will lead to crash.

Reviewed-by: Liming Wang <liming.wang@windriver.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A4186DD.6090405@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/trace_events.c | 28 ++++++++++++++++++++++------
 1 file changed, 22 insertions(+), 6 deletions(-)

diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index aa08be69a1b6..53c8fd376a88 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -300,10 +300,18 @@ t_next(struct seq_file *m, void *v, loff_t *pos)
 
 static void *t_start(struct seq_file *m, loff_t *pos)
 {
+	struct ftrace_event_call *call = NULL;
+	loff_t l;
+
 	mutex_lock(&event_mutex);
-	if (*pos == 0)
-		m->private = ftrace_events.next;
-	return t_next(m, NULL, pos);
+
+	m->private = ftrace_events.next;
+	for (l = 0; l <= *pos; ) {
+		call = t_next(m, NULL, &l);
+		if (!call)
+			break;
+	}
+	return call;
 }
 
 static void *
@@ -332,10 +340,18 @@ s_next(struct seq_file *m, void *v, loff_t *pos)
 
 static void *s_start(struct seq_file *m, loff_t *pos)
 {
+	struct ftrace_event_call *call = NULL;
+	loff_t l;
+
 	mutex_lock(&event_mutex);
-	if (*pos == 0)
-		m->private = ftrace_events.next;
-	return s_next(m, NULL, pos);
+
+	m->private = ftrace_events.next;
+	for (l = 0; l <= *pos; ) {
+		call = s_next(m, NULL, &l);
+		if (!call)
+			break;
+	}
+	return call;
 }
 
 static int t_show(struct seq_file *m, void *v)
-- 
cgit v1.2.3-59-g8ed1b


From c8961ec6da22ea010bf4470a8e0fb3fdad0f11c4 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Wed, 24 Jun 2009 09:52:58 +0800
Subject: tracing_bprintk: Don't increment @pos in t_start()

It's wrong to increment @pos in t_start(), otherwise we'll lose
some entries when reading printk_formats, if the output is larger
than PAGE_SIZE.

Reported-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Reviewed-by: Liming Wang <liming.wang@windriver.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A4186FA.1020106@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/trace_printk.c | 26 ++++++--------------------
 1 file changed, 6 insertions(+), 20 deletions(-)

diff --git a/kernel/trace/trace_printk.c b/kernel/trace/trace_printk.c
index 9bece9687b62..7b6278110827 100644
--- a/kernel/trace/trace_printk.c
+++ b/kernel/trace/trace_printk.c
@@ -155,25 +155,19 @@ int __ftrace_vprintk(unsigned long ip, const char *fmt, va_list ap)
 EXPORT_SYMBOL_GPL(__ftrace_vprintk);
 
 static void *
-t_next(struct seq_file *m, void *v, loff_t *pos)
+t_start(struct seq_file *m, loff_t *pos)
 {
-	const char **fmt = m->private;
-	const char **next = fmt;
-
-	(*pos)++;
+	const char **fmt = __start___trace_bprintk_fmt + *pos;
 
 	if ((unsigned long)fmt >= (unsigned long)__stop___trace_bprintk_fmt)
 		return NULL;
-
-	next = fmt;
-	m->private = ++next;
-
 	return fmt;
 }
 
-static void *t_start(struct seq_file *m, loff_t *pos)
+static void *t_next(struct seq_file *m, void * v, loff_t *pos)
 {
-	return t_next(m, NULL, pos);
+	(*pos)++;
+	return t_start(m, pos);
 }
 
 static int t_show(struct seq_file *m, void *v)
@@ -224,15 +218,7 @@ static const struct seq_operations show_format_seq_ops = {
 static int
 ftrace_formats_open(struct inode *inode, struct file *file)
 {
-	int ret;
-
-	ret = seq_open(file, &show_format_seq_ops);
-	if (!ret) {
-		struct seq_file *m = file->private_data;
-
-		m->private = __start___trace_bprintk_fmt;
-	}
-	return ret;
+	return seq_open(file, &show_format_seq_ops);
 }
 
 static const struct file_operations ftrace_formats_fops = {
-- 
cgit v1.2.3-59-g8ed1b


From 2961bf345fd1b736c3db46cad0f69855f67fbe9c Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Wed, 24 Jun 2009 09:53:26 +0800
Subject: trace_stat: Don't increment @pos in seq start()

It's wrong to increment @pos in stat_seq_start(). It causes some
stat entries lost when reading stat file, if the output of the file
is larger than PAGE_SIZE.

Reviewed-by: Liming Wang <liming.wang@windriver.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A418716.90209@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/trace_stat.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/kernel/trace/trace_stat.c b/kernel/trace/trace_stat.c
index c00643733f4c..e66f5e493342 100644
--- a/kernel/trace/trace_stat.c
+++ b/kernel/trace/trace_stat.c
@@ -199,17 +199,13 @@ static void *stat_seq_start(struct seq_file *s, loff_t *pos)
 	mutex_lock(&session->stat_mutex);
 
 	/* If we are in the beginning of the file, print the headers */
-	if (!*pos && session->ts->stat_headers) {
-		(*pos)++;
+	if (!*pos && session->ts->stat_headers)
 		return SEQ_START_TOKEN;
-	}
 
 	node = rb_first(&session->stat_root);
 	for (i = 0; node && i < *pos; i++)
 		node = rb_next(node);
 
-	(*pos)++;
-
 	return node;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From f129e965bef40c6153e4fe505f1e408286213424 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Wed, 24 Jun 2009 09:53:44 +0800
Subject: tracing: Reset iterator in t_start()

The iterator is m->private, but it's not reset to trace_types in
t_start(). If the output is larger than PAGE_SIZE and t_start()
is called the 2nd time, things will go wrong.

Reviewed-by: Liming Wang <liming.wang@windriver.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A418728.5020506@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/trace.c | 18 ++++--------------
 1 file changed, 4 insertions(+), 14 deletions(-)

diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 076fa6f0ee48..3bb31006b5cc 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -2053,25 +2053,23 @@ static int tracing_open(struct inode *inode, struct file *file)
 static void *
 t_next(struct seq_file *m, void *v, loff_t *pos)
 {
-	struct tracer *t = m->private;
+	struct tracer *t = v;
 
 	(*pos)++;
 
 	if (t)
 		t = t->next;
 
-	m->private = t;
-
 	return t;
 }
 
 static void *t_start(struct seq_file *m, loff_t *pos)
 {
-	struct tracer *t = m->private;
+	struct tracer *t;
 	loff_t l = 0;
 
 	mutex_lock(&trace_types_lock);
-	for (; t && l < *pos; t = t_next(m, t, &l))
+	for (t = trace_types; t && l < *pos; t = t_next(m, t, &l))
 		;
 
 	return t;
@@ -2107,18 +2105,10 @@ static struct seq_operations show_traces_seq_ops = {
 
 static int show_traces_open(struct inode *inode, struct file *file)
 {
-	int ret;
-
 	if (tracing_disabled)
 		return -ENODEV;
 
-	ret = seq_open(file, &show_traces_seq_ops);
-	if (!ret) {
-		struct seq_file *m = file->private_data;
-		m->private = trace_types;
-	}
-
-	return ret;
+	return seq_open(file, &show_traces_seq_ops);
 }
 
 static ssize_t
-- 
cgit v1.2.3-59-g8ed1b


From 85951842a1020669f0a9eb0f0d1853b41341f097 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Wed, 24 Jun 2009 09:54:00 +0800
Subject: ftrace: Don't increment @pos in g_start()

It's wrong to increment @pos in g_start(). It causes some entries
lost when reading set_graph_function, if the output of the file
is larger than PAGE_SIZE.

Reviewed-by: Liming Wang <liming.wang@windriver.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A418738.7090401@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/ftrace.c | 21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 3718d55fb4c3..cde74b9973b7 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -2500,32 +2500,31 @@ int ftrace_graph_count;
 unsigned long ftrace_graph_funcs[FTRACE_GRAPH_MAX_FUNCS] __read_mostly;
 
 static void *
-g_next(struct seq_file *m, void *v, loff_t *pos)
+__g_next(struct seq_file *m, loff_t *pos)
 {
 	unsigned long *array = m->private;
-	int index = *pos;
-
-	(*pos)++;
 
-	if (index >= ftrace_graph_count)
+	if (*pos >= ftrace_graph_count)
 		return NULL;
+	return &array[*pos];
+}
 
-	return &array[index];
+static void *
+g_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	(*pos)++;
+	return __g_next(m, pos);
 }
 
 static void *g_start(struct seq_file *m, loff_t *pos)
 {
-	void *p = NULL;
-
 	mutex_lock(&graph_lock);
 
 	/* Nothing, tell g_show to print all functions are enabled */
 	if (!ftrace_graph_count && !*pos)
 		return (void *)1;
 
-	p = g_next(m, p, pos);
-
-	return p;
+	return __g_next(m, pos);
 }
 
 static void g_stop(struct seq_file *m, void *p)
-- 
cgit v1.2.3-59-g8ed1b


From 694ce0a544fba37a60025a6803ee6265be8a2a22 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Wed, 24 Jun 2009 09:54:19 +0800
Subject: ftrace: Don't manipulate @pos in t_start()

It's rather confusing that in t_start(), in some cases @pos is
incremented, and in some cases it's decremented and then incremented.

This patch rewrites t_start() in a much more general way.

Thus we fix a bug that if ftrace_filtered == 1, functions have tracer
hooks won't be printed, because the branch is always unreachable:

static void *t_start(...)
{
	...
	if (!p)
		return t_hash_start(m, pos);
	return p;
}

Before:
  # echo 'sys_open' > /mnt/tracing/set_ftrace_filter
  # echo 'sys_write:traceon:4' >> /mnt/tracing/set_ftrace_filter
  sys_open

After:
  # echo 'sys_open' > /mnt/tracing/set_ftrace_filter
  # echo 'sys_write:traceon:4' >> /mnt/tracing/set_ftrace_filter
  sys_open
  sys_write:traceon:count=4

Reviewed-by: Liming Wang <liming.wang@windriver.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A41874B.4090507@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/ftrace.c | 19 ++++++++-----------
 1 file changed, 8 insertions(+), 11 deletions(-)

diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index cde74b9973b7..dc810208edde 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -1467,8 +1467,6 @@ t_next(struct seq_file *m, void *v, loff_t *pos)
 			iter->pg = iter->pg->next;
 			iter->idx = 0;
 			goto retry;
-		} else {
-			iter->idx = -1;
 		}
 	} else {
 		rec = &iter->pg->records[iter->idx++];
@@ -1497,6 +1495,7 @@ static void *t_start(struct seq_file *m, loff_t *pos)
 {
 	struct ftrace_iterator *iter = m->private;
 	void *p = NULL;
+	loff_t l;
 
 	mutex_lock(&ftrace_lock);
 	/*
@@ -1508,23 +1507,21 @@ static void *t_start(struct seq_file *m, loff_t *pos)
 		if (*pos > 0)
 			return t_hash_start(m, pos);
 		iter->flags |= FTRACE_ITER_PRINTALL;
-		(*pos)++;
 		return iter;
 	}
 
 	if (iter->flags & FTRACE_ITER_HASH)
 		return t_hash_start(m, pos);
 
-	if (*pos > 0) {
-		if (iter->idx < 0)
-			return p;
-		(*pos)--;
-		iter->idx--;
+	iter->pg = ftrace_pages_start;
+	iter->idx = 0;
+	for (l = 0; l <= *pos; ) {
+		p = t_next(m, p, &l);
+		if (!p)
+			break;
 	}
 
-	p = t_next(m, p, pos);
-
-	if (!p)
+	if (!p && iter->flags & FTRACE_ITER_FILTER)
 		return t_hash_start(m, pos);
 
 	return p;
-- 
cgit v1.2.3-59-g8ed1b


From d82d62444f87e5993af2fa82ed636b2206e052ea Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Wed, 24 Jun 2009 09:54:54 +0800
Subject: ftrace: Fix t_hash_start()

When the output of set_ftrace_filter is larger than PAGE_SIZE,
t_hash_start() will be called the 2nd time, and then we start
from the head of a hlist, which is wrong and causes some entries
to be outputed twice.

The worse is, if the hlist is large enough, reading set_ftrace_filter
won't stop but in a dead loop.

Reviewed-by: Liming Wang <liming.wang@windriver.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A41876E.2060407@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/ftrace.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index dc810208edde..71a52c172140 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -1417,10 +1417,20 @@ static void *t_hash_start(struct seq_file *m, loff_t *pos)
 {
 	struct ftrace_iterator *iter = m->private;
 	void *p = NULL;
+	loff_t l;
+
+	if (!(iter->flags & FTRACE_ITER_HASH))
+		*pos = 0;
 
 	iter->flags |= FTRACE_ITER_HASH;
 
-	return t_hash_next(m, p, pos);
+	iter->hidx = 0;
+	for (l = 0; l <= *pos; ) {
+		p = t_hash_next(m, p, &l);
+		if (!p)
+			break;
+	}
+	return p;
 }
 
 static int t_hash_show(struct seq_file *m, void *v)
-- 
cgit v1.2.3-59-g8ed1b


From 507e123151149e578c9aae33eb876c49824da5f8 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 23 Jun 2009 17:38:15 +0200
Subject: timer stats: Optimize by adding quick check to avoid function calls

When the kernel is configured with CONFIG_TIMER_STATS but timer
stats are runtime disabled we still get calls to
__timer_stats_timer_set_start_info which initializes some
fields in the corresponding struct timer_list.

So add some quick checks in the the timer stats setup functions
to avoid function calls to __timer_stats_timer_set_start_info
when timer stats are disabled.

In an artificial workload that does nothing but playing ping
pong with a single tcp packet via loopback this decreases cpu
consumption by 1 - 1.5%.

This is part of a modified function trace output on SLES11:

 perl-2497  [00] 28630647177732388 [+  125]: sk_reset_timer <-tcp_v4_rcv
 perl-2497  [00] 28630647177732513 [+  125]: mod_timer <-sk_reset_timer
 perl-2497  [00] 28630647177732638 [+  125]: __timer_stats_timer_set_start_info <-mod_timer
 perl-2497  [00] 28630647177732763 [+  125]: __mod_timer <-mod_timer
 perl-2497  [00] 28630647177732888 [+  125]: __timer_stats_timer_set_start_info <-__mod_timer
 perl-2497  [00] 28630647177733013 [+   93]: lock_timer_base <-__mod_timer

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Mustafa Mesanovic <mustafa.mesanovic@de.ibm.com>
Cc: Arjan van de Ven <arjan@infradead.org>
LKML-Reference: <20090623153811.GA4641@osiris.boeblingen.de.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/hrtimer.h   |  5 +++++
 include/linux/timer.h     |  4 ++++
 kernel/time/timer_stats.c | 16 ++++++++--------
 kernel/timer.c            |  2 ++
 4 files changed, 19 insertions(+), 8 deletions(-)

diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index 7400900de94a..54648e625efd 100644
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -21,6 +21,7 @@
 #include <linux/list.h>
 #include <linux/wait.h>
 #include <linux/percpu.h>
+#include <linux/timer.h>
 
 
 struct hrtimer_clock_base;
@@ -447,6 +448,8 @@ extern void timer_stats_update_stats(void *timer, pid_t pid, void *startf,
 
 static inline void timer_stats_account_hrtimer(struct hrtimer *timer)
 {
+	if (likely(!timer->start_pid))
+		return;
 	timer_stats_update_stats(timer, timer->start_pid, timer->start_site,
 				 timer->function, timer->start_comm, 0);
 }
@@ -456,6 +459,8 @@ extern void __timer_stats_hrtimer_set_start_info(struct hrtimer *timer,
 
 static inline void timer_stats_hrtimer_set_start_info(struct hrtimer *timer)
 {
+	if (likely(!timer_stats_active))
+		return;
 	__timer_stats_hrtimer_set_start_info(timer, __builtin_return_address(0));
 }
 
diff --git a/include/linux/timer.h b/include/linux/timer.h
index ccf882eed8f8..be62ec2ebea5 100644
--- a/include/linux/timer.h
+++ b/include/linux/timer.h
@@ -190,6 +190,8 @@ extern unsigned long get_next_timer_interrupt(unsigned long now);
  */
 #ifdef CONFIG_TIMER_STATS
 
+extern int timer_stats_active;
+
 #define TIMER_STATS_FLAG_DEFERRABLE	0x1
 
 extern void init_timer_stats(void);
@@ -203,6 +205,8 @@ extern void __timer_stats_timer_set_start_info(struct timer_list *timer,
 
 static inline void timer_stats_timer_set_start_info(struct timer_list *timer)
 {
+	if (likely(!timer_stats_active))
+		return;
 	__timer_stats_timer_set_start_info(timer, __builtin_return_address(0));
 }
 
diff --git a/kernel/time/timer_stats.c b/kernel/time/timer_stats.c
index c994530d166d..4cde8b9c716f 100644
--- a/kernel/time/timer_stats.c
+++ b/kernel/time/timer_stats.c
@@ -96,7 +96,7 @@ static DEFINE_MUTEX(show_mutex);
 /*
  * Collection status, active/inactive:
  */
-static int __read_mostly active;
+int __read_mostly timer_stats_active;
 
 /*
  * Beginning/end timestamps of measurement:
@@ -242,7 +242,7 @@ void timer_stats_update_stats(void *timer, pid_t pid, void *startf,
 	struct entry *entry, input;
 	unsigned long flags;
 
-	if (likely(!active))
+	if (likely(!timer_stats_active))
 		return;
 
 	lock = &per_cpu(lookup_lock, raw_smp_processor_id());
@@ -254,7 +254,7 @@ void timer_stats_update_stats(void *timer, pid_t pid, void *startf,
 	input.timer_flag = timer_flag;
 
 	spin_lock_irqsave(lock, flags);
-	if (!active)
+	if (!timer_stats_active)
 		goto out_unlock;
 
 	entry = tstat_lookup(&input, comm);
@@ -290,7 +290,7 @@ static int tstats_show(struct seq_file *m, void *v)
 	/*
 	 * If still active then calculate up to now:
 	 */
-	if (active)
+	if (timer_stats_active)
 		time_stop = ktime_get();
 
 	time = ktime_sub(time_stop, time_start);
@@ -368,18 +368,18 @@ static ssize_t tstats_write(struct file *file, const char __user *buf,
 	mutex_lock(&show_mutex);
 	switch (ctl[0]) {
 	case '0':
-		if (active) {
-			active = 0;
+		if (timer_stats_active) {
+			timer_stats_active = 0;
 			time_stop = ktime_get();
 			sync_access();
 		}
 		break;
 	case '1':
-		if (!active) {
+		if (!timer_stats_active) {
 			reset_entries();
 			time_start = ktime_get();
 			smp_mb();
-			active = 1;
+			timer_stats_active = 1;
 		}
 		break;
 	default:
diff --git a/kernel/timer.c b/kernel/timer.c
index 54d3912f8cad..0b36b9e5cc8b 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -380,6 +380,8 @@ static void timer_stats_account_timer(struct timer_list *timer)
 {
 	unsigned int flag = 0;
 
+	if (likely(!timer->start_site))
+		return;
 	if (unlikely(tbase_get_deferrable(timer->base)))
 		flag |= TIMER_STATS_FLAG_DEFERRABLE;
 
-- 
cgit v1.2.3-59-g8ed1b


From d7e2f36d9a92284754ed5254562766cb3d61c7ca Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Wed, 24 Jun 2009 02:36:17 -0700
Subject: ide cs5520: Initialize second port's interrupt number.

In 86ccf37c6acd74cf7e4b7751ee045de19943c5a0 the driver was modified
to deal with the removal of the pciirq argument to ide_pci_setup_ports().

But in the conversion only the first port's IRQ gets setup.

Inspired by a patch by Bartlomiej Zolnierkiewicz., and with help from
Alan Cox.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/cs5520.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/ide/cs5520.c b/drivers/ide/cs5520.c
index bd066bb9d611..09f98ed0731f 100644
--- a/drivers/ide/cs5520.c
+++ b/drivers/ide/cs5520.c
@@ -135,6 +135,7 @@ static int __devinit cs5520_init_one(struct pci_dev *dev, const struct pci_devic
 
 	ide_pci_setup_ports(dev, d, &hw[0], &hws[0]);
 	hw[0].irq = 14;
+	hw[1].irq = 15;
 
 	return ide_host_add(d, hws, 2, NULL);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 9d612beff5089b89a295a2331883a8ce3fff08c1 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Wed, 24 Jun 2009 17:33:15 +0800
Subject: tracing: Fix trace_buf_size boot option

We should be able to specify [KMG] when setting trace_buf_size
boot option, as documented in kernel-parameters.txt

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A41F2DB.4020102@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 Documentation/kernel-parameters.txt | 3 ++-
 kernel/trace/trace.c                | 5 ++---
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 92e1ab8178a8..d3f41db3ed49 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -2475,7 +2475,8 @@ and is between 256 and 4096 characters. It is defined in the file
 
 	tp720=		[HW,PS2]
 
-	trace_buf_size=nn[KMG] [ftrace] will set tracing buffer size.
+	trace_buf_size=nn[KMG]
+			[FTRACE] will set tracing buffer size.
 
 	trix=		[HW,OSS] MediaTrix AudioTrix Pro
 			Format:
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 3bb31006b5cc..3aa0a0dfdfa8 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -284,13 +284,12 @@ void trace_wake_up(void)
 static int __init set_buf_size(char *str)
 {
 	unsigned long buf_size;
-	int ret;
 
 	if (!str)
 		return 0;
-	ret = strict_strtoul(str, 0, &buf_size);
+	buf_size = memparse(str, &str);
 	/* nr_entries can not be zero */
-	if (ret < 0 || buf_size == 0)
+	if (buf_size == 0)
 		return 0;
 	trace_buf_size = buf_size;
 	return 1;
-- 
cgit v1.2.3-59-g8ed1b


From 3391faa4f18e4e33666d3d24e90e3086fcf9b922 Mon Sep 17 00:00:00 2001
From: Roel Kluin <roel.kluin@gmail.com>
Date: Mon, 22 Jun 2009 23:12:29 +0200
Subject: udf: remove redundant tests on unsigned

first_block and goal are unsigned. When negative they are wrapped and caught by
the other test.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
---
 fs/udf/balloc.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/fs/udf/balloc.c b/fs/udf/balloc.c
index e48e9a3af763..1e068535b58b 100644
--- a/fs/udf/balloc.c
+++ b/fs/udf/balloc.c
@@ -238,7 +238,7 @@ static int udf_bitmap_prealloc_blocks(struct super_block *sb,
 
 	mutex_lock(&sbi->s_alloc_mutex);
 	part_len = sbi->s_partmaps[partition].s_partition_len;
-	if (first_block < 0 || first_block >= part_len)
+	if (first_block >= part_len)
 		goto out;
 
 	if (first_block + block_count > part_len)
@@ -297,7 +297,7 @@ static int udf_bitmap_new_block(struct super_block *sb,
 	mutex_lock(&sbi->s_alloc_mutex);
 
 repeat:
-	if (goal < 0 || goal >= sbi->s_partmaps[partition].s_partition_len)
+	if (goal >= sbi->s_partmaps[partition].s_partition_len)
 		goal = 0;
 
 	nr_groups = bitmap->s_nr_groups;
@@ -666,8 +666,7 @@ static int udf_table_prealloc_blocks(struct super_block *sb,
 	int8_t etype = -1;
 	struct udf_inode_info *iinfo;
 
-	if (first_block < 0 ||
-		first_block >= sbi->s_partmaps[partition].s_partition_len)
+	if (first_block >= sbi->s_partmaps[partition].s_partition_len)
 		return 0;
 
 	iinfo = UDF_I(table);
@@ -743,7 +742,7 @@ static int udf_table_new_block(struct super_block *sb,
 		return newblock;
 
 	mutex_lock(&sbi->s_alloc_mutex);
-	if (goal < 0 || goal >= sbi->s_partmaps[partition].s_partition_len)
+	if (goal >= sbi->s_partmaps[partition].s_partition_len)
 		goal = 0;
 
 	/* We search for the closest matching block to goal. If we find
-- 
cgit v1.2.3-59-g8ed1b


From 16d11a829ed197b719723f81d82e7f1a42f5c681 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Wed, 24 Jun 2009 14:07:53 +0200
Subject: ALSA: hda - Simplify AD1986A mixer definitions

Split mixer element arrays of AD1986A models to several pieces so that
each model can share the same mixer arrays.
This removes lots of duplicated data.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/hda/patch_analog.c | 77 ++++++++++++--------------------------------
 1 file changed, 21 insertions(+), 56 deletions(-)

diff --git a/sound/pci/hda/patch_analog.c b/sound/pci/hda/patch_analog.c
index 84cc49ca9148..592423c878f2 100644
--- a/sound/pci/hda/patch_analog.c
+++ b/sound/pci/hda/patch_analog.c
@@ -669,39 +669,13 @@ static struct hda_input_mux ad1986a_automic_capture_source = {
 	},
 };
 
-static struct snd_kcontrol_new ad1986a_laptop_eapd_mixers[] = {
+static struct snd_kcontrol_new ad1986a_laptop_master_mixers[] = {
 	HDA_BIND_VOL("Master Playback Volume", &ad1986a_laptop_master_vol),
 	HDA_BIND_SW("Master Playback Switch", &ad1986a_laptop_master_sw),
-	HDA_CODEC_VOLUME("PCM Playback Volume", 0x03, 0x0, HDA_OUTPUT),
-	HDA_CODEC_MUTE("PCM Playback Switch", 0x03, 0x0, HDA_OUTPUT),
-	HDA_CODEC_VOLUME("Internal Mic Playback Volume", 0x17, 0, HDA_OUTPUT),
-	HDA_CODEC_MUTE("Internal Mic Playback Switch", 0x17, 0, HDA_OUTPUT),
-	HDA_CODEC_VOLUME("Mic Playback Volume", 0x13, 0x0, HDA_OUTPUT),
-	HDA_CODEC_MUTE("Mic Playback Switch", 0x13, 0x0, HDA_OUTPUT),
-	HDA_CODEC_VOLUME("Mic Boost", 0x0f, 0x0, HDA_OUTPUT),
-	HDA_CODEC_VOLUME("Capture Volume", 0x12, 0x0, HDA_OUTPUT),
-	HDA_CODEC_MUTE("Capture Switch", 0x12, 0x0, HDA_OUTPUT),
-	{
-		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
-		.name = "Capture Source",
-		.info = ad198x_mux_enum_info,
-		.get = ad198x_mux_enum_get,
-		.put = ad198x_mux_enum_put,
-	},
-	{
-		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
-		.name = "External Amplifier",
-		.info = ad198x_eapd_info,
-		.get = ad198x_eapd_get,
-		.put = ad198x_eapd_put,
-		.private_value = 0x1b | (1 << 8), /* port-D, inversed */
-	},
 	{ } /* end */
 };
 
-static struct snd_kcontrol_new ad1986a_samsung_mixers[] = {
-	HDA_BIND_VOL("Master Playback Volume", &ad1986a_laptop_master_vol),
-	HDA_BIND_SW("Master Playback Switch", &ad1986a_laptop_master_sw),
+static struct snd_kcontrol_new ad1986a_laptop_eapd_mixers[] = {
 	HDA_CODEC_VOLUME("PCM Playback Volume", 0x03, 0x0, HDA_OUTPUT),
 	HDA_CODEC_MUTE("PCM Playback Switch", 0x03, 0x0, HDA_OUTPUT),
 	HDA_CODEC_VOLUME("Mic Playback Volume", 0x13, 0x0, HDA_OUTPUT),
@@ -727,6 +701,12 @@ static struct snd_kcontrol_new ad1986a_samsung_mixers[] = {
 	{ } /* end */
 };
 
+static struct snd_kcontrol_new ad1986a_laptop_intmic_mixers[] = {
+	HDA_CODEC_VOLUME("Internal Mic Playback Volume", 0x17, 0, HDA_OUTPUT),
+	HDA_CODEC_MUTE("Internal Mic Playback Switch", 0x17, 0, HDA_OUTPUT),
+	{ } /* end */
+};
+
 /* re-connect the mic boost input according to the jack sensing */
 static void ad1986a_automic(struct hda_codec *codec)
 {
@@ -816,7 +796,7 @@ static int ad1986a_hp_master_sw_put(struct snd_kcontrol *kcontrol,
 	return change;
 }
 
-static struct snd_kcontrol_new ad1986a_laptop_automute_mixers[] = {
+static struct snd_kcontrol_new ad1986a_automute_master_mixers[] = {
 	HDA_BIND_VOL("Master Playback Volume", &ad1986a_laptop_master_vol),
 	{
 		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
@@ -826,33 +806,10 @@ static struct snd_kcontrol_new ad1986a_laptop_automute_mixers[] = {
 		.put = ad1986a_hp_master_sw_put,
 		.private_value = HDA_COMPOSE_AMP_VAL(0x1a, 3, 0, HDA_OUTPUT),
 	},
-	HDA_CODEC_VOLUME("PCM Playback Volume", 0x03, 0x0, HDA_OUTPUT),
-	HDA_CODEC_MUTE("PCM Playback Switch", 0x03, 0x0, HDA_OUTPUT),
-	HDA_CODEC_VOLUME("Internal Mic Playback Volume", 0x17, 0x0, HDA_OUTPUT),
-	HDA_CODEC_MUTE("Internal Mic Playback Switch", 0x17, 0x0, HDA_OUTPUT),
-	HDA_CODEC_VOLUME("Mic Playback Volume", 0x13, 0x0, HDA_OUTPUT),
-	HDA_CODEC_MUTE("Mic Playback Switch", 0x13, 0x0, HDA_OUTPUT),
-	HDA_CODEC_VOLUME("Mic Boost", 0x0f, 0x0, HDA_OUTPUT),
-	HDA_CODEC_VOLUME("Capture Volume", 0x12, 0x0, HDA_OUTPUT),
-	HDA_CODEC_MUTE("Capture Switch", 0x12, 0x0, HDA_OUTPUT),
-	{
-		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
-		.name = "Capture Source",
-		.info = ad198x_mux_enum_info,
-		.get = ad198x_mux_enum_get,
-		.put = ad198x_mux_enum_put,
-	},
-	{
-		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
-		.name = "External Amplifier",
-		.info = ad198x_eapd_info,
-		.get = ad198x_eapd_get,
-		.put = ad198x_eapd_put,
-		.private_value = 0x1b | (1 << 8), /* port-D, inversed */
-	},
 	{ } /* end */
 };
 
+
 /*
  * initialization verbs
  */
@@ -1111,7 +1068,10 @@ static int patch_ad1986a(struct hda_codec *codec)
 		spec->multiout.dac_nids = ad1986a_laptop_dac_nids;
 		break;
 	case AD1986A_LAPTOP_EAPD:
-		spec->mixers[0] = ad1986a_laptop_eapd_mixers;
+		spec->num_mixers = 3;
+		spec->mixers[0] = ad1986a_laptop_master_mixers;
+		spec->mixers[1] = ad1986a_laptop_eapd_mixers;
+		spec->mixers[2] = ad1986a_laptop_intmic_mixers;
 		spec->num_init_verbs = 2;
 		spec->init_verbs[1] = ad1986a_eapd_init_verbs;
 		spec->multiout.max_channels = 2;
@@ -1122,7 +1082,9 @@ static int patch_ad1986a(struct hda_codec *codec)
 		spec->input_mux = &ad1986a_laptop_eapd_capture_source;
 		break;
 	case AD1986A_SAMSUNG:
-		spec->mixers[0] = ad1986a_samsung_mixers;
+		spec->num_mixers = 2;
+		spec->mixers[0] = ad1986a_laptop_master_mixers;
+		spec->mixers[1] = ad1986a_laptop_eapd_mixers;
 		spec->num_init_verbs = 3;
 		spec->init_verbs[1] = ad1986a_eapd_init_verbs;
 		spec->init_verbs[2] = ad1986a_automic_verbs;
@@ -1136,7 +1098,10 @@ static int patch_ad1986a(struct hda_codec *codec)
 		codec->patch_ops.init = ad1986a_automic_init;
 		break;
 	case AD1986A_LAPTOP_AUTOMUTE:
-		spec->mixers[0] = ad1986a_laptop_automute_mixers;
+		spec->num_mixers = 3;
+		spec->mixers[0] = ad1986a_automute_master_mixers;
+		spec->mixers[1] = ad1986a_laptop_eapd_mixers;
+		spec->mixers[2] = ad1986a_laptop_intmic_mixers;
 		spec->num_init_verbs = 3;
 		spec->init_verbs[1] = ad1986a_eapd_init_verbs;
 		spec->init_verbs[2] = ad1986a_hp_init_verbs;
-- 
cgit v1.2.3-59-g8ed1b


From 6f4b67b8ff707147e14ee71045ab25aa286520f2 Mon Sep 17 00:00:00 2001
From: Shin-ichiro KAWASAKI <kawasaki@juno.dti.ne.jp>
Date: Sun, 21 Jun 2009 10:56:22 +0000
Subject: clocksource: sh_tmu: Make undefined TCOR behaviour less undefined.

Avoid undocumented vague TMU behavior when zero value is set to TCOR.

This primarily fixes up issues encountered under qemu with a zero-length
period, while the hardware itself is fairly ambivalent one way or the
other.

Signed-off-by: Shin-ichiro KAWASAKI <kawasaki@juno.dti.ne.jp>
Acked-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 drivers/clocksource/sh_tmu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/clocksource/sh_tmu.c b/drivers/clocksource/sh_tmu.c
index 9ffb05f4095d..93c2322feab7 100644
--- a/drivers/clocksource/sh_tmu.c
+++ b/drivers/clocksource/sh_tmu.c
@@ -161,7 +161,7 @@ static void sh_tmu_set_next(struct sh_tmu_priv *p, unsigned long delta,
 	if (periodic)
 		sh_tmu_write(p, TCOR, delta);
 	else
-		sh_tmu_write(p, TCOR, 0);
+		sh_tmu_write(p, TCOR, 0xffffffff);
 
 	sh_tmu_write(p, TCNT, delta);
 
-- 
cgit v1.2.3-59-g8ed1b


From 03c405ad314d3c4e049b8d04500e54e833d16747 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Wed, 24 Jun 2009 14:10:15 +0200
Subject: ALSA: hda - Generalize the pin-detect quirk for Lenovo N100

Add a new flag to ad_spec struct so that the same hack can be used for
any other models (if any).  This also allows other models to reuse the
auto-mute functions.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/hda/patch_analog.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/sound/pci/hda/patch_analog.c b/sound/pci/hda/patch_analog.c
index 592423c878f2..8c2b23f54f95 100644
--- a/sound/pci/hda/patch_analog.c
+++ b/sound/pci/hda/patch_analog.c
@@ -72,6 +72,7 @@ struct ad198x_spec {
 	hda_nid_t private_dac_nids[AUTO_CFG_MAX_OUTS];
 
 	unsigned int jack_present :1;
+	unsigned int inv_jack_detect:1;
 
 #ifdef CONFIG_SND_HDA_POWER_SAVE
 	struct hda_loopback_check loopback;
@@ -756,8 +757,9 @@ static void ad1986a_hp_automute(struct hda_codec *codec)
 	unsigned int present;
 
 	present = snd_hda_codec_read(codec, 0x1a, 0, AC_VERB_GET_PIN_SENSE, 0);
-	/* Lenovo N100 seems to report the reversed bit for HP jack-sensing */
-	spec->jack_present = !(present & 0x80000000);
+	spec->jack_present = !!(present & 0x80000000);
+	if (spec->inv_jack_detect)
+		spec->jack_present = !spec->jack_present;
 	ad1986a_update_hp(codec);
 }
 
@@ -1113,6 +1115,10 @@ static int patch_ad1986a(struct hda_codec *codec)
 		spec->input_mux = &ad1986a_laptop_eapd_capture_source;
 		codec->patch_ops.unsol_event = ad1986a_hp_unsol_event;
 		codec->patch_ops.init = ad1986a_hp_init;
+		/* Lenovo N100 seems to report the reversed bit
+		 * for HP jack-sensing
+		 */
+		spec->inv_jack_detect = 1;
 		break;
 	case AD1986A_ULTRA:
 		spec->mixers[0] = ad1986a_laptop_eapd_mixers;
-- 
cgit v1.2.3-59-g8ed1b


From c912e7a58054304575fe88574c776be7e684098e Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Wed, 24 Jun 2009 14:14:34 +0200
Subject: ALSA: hda - Fix support for Samsung P50 with AD1986A codec

Samsung P50 requires the HP auto-muting unlike other Samsung models.
Added a new model=samsung-p50 to support this.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 Documentation/sound/alsa/HD-Audio-Models.txt |  1 +
 sound/pci/hda/patch_analog.c                 | 41 ++++++++++++++++++++++++++++
 2 files changed, 42 insertions(+)

diff --git a/Documentation/sound/alsa/HD-Audio-Models.txt b/Documentation/sound/alsa/HD-Audio-Models.txt
index 0d8d23581c44..939a3dd58148 100644
--- a/Documentation/sound/alsa/HD-Audio-Models.txt
+++ b/Documentation/sound/alsa/HD-Audio-Models.txt
@@ -240,6 +240,7 @@ AD1986A
   laptop-automute 2-channel with EAPD and HP-automute (Lenovo N100)
   ultra		2-channel with EAPD (Samsung Ultra tablet PC)
   samsung	2-channel with EAPD (Samsung R65)
+  samsung-p50	2-channel with HP-automute (Samsung P50)
 
 AD1988/AD1988B/AD1989A/AD1989B
 ==============================
diff --git a/sound/pci/hda/patch_analog.c b/sound/pci/hda/patch_analog.c
index 8c2b23f54f95..1988582d1ab8 100644
--- a/sound/pci/hda/patch_analog.c
+++ b/sound/pci/hda/patch_analog.c
@@ -940,6 +940,27 @@ static struct hda_verb ad1986a_hp_init_verbs[] = {
 	{}
 };
 
+static void ad1986a_samsung_p50_unsol_event(struct hda_codec *codec,
+					    unsigned int res)
+{
+	switch (res >> 26) {
+	case AD1986A_HP_EVENT:
+		ad1986a_hp_automute(codec);
+		break;
+	case AD1986A_MIC_EVENT:
+		ad1986a_automic(codec);
+		break;
+	}
+}
+
+static int ad1986a_samsung_p50_init(struct hda_codec *codec)
+{
+	ad198x_init(codec);
+	ad1986a_hp_automute(codec);
+	ad1986a_automic(codec);
+	return 0;
+}
+
 
 /* models */
 enum {
@@ -950,6 +971,7 @@ enum {
 	AD1986A_LAPTOP_AUTOMUTE,
 	AD1986A_ULTRA,
 	AD1986A_SAMSUNG,
+	AD1986A_SAMSUNG_P50,
 	AD1986A_MODELS
 };
 
@@ -961,6 +983,7 @@ static const char *ad1986a_models[AD1986A_MODELS] = {
 	[AD1986A_LAPTOP_AUTOMUTE] = "laptop-automute",
 	[AD1986A_ULTRA]		= "ultra",
 	[AD1986A_SAMSUNG]	= "samsung",
+	[AD1986A_SAMSUNG_P50]	= "samsung-p50",
 };
 
 static struct snd_pci_quirk ad1986a_cfg_tbl[] = {
@@ -983,6 +1006,7 @@ static struct snd_pci_quirk ad1986a_cfg_tbl[] = {
 	SND_PCI_QUIRK(0x1179, 0xff40, "Toshiba", AD1986A_LAPTOP_EAPD),
 	SND_PCI_QUIRK(0x144d, 0xb03c, "Samsung R55", AD1986A_3STACK),
 	SND_PCI_QUIRK(0x144d, 0xc01e, "FSC V2060", AD1986A_LAPTOP),
+	SND_PCI_QUIRK(0x144d, 0xc024, "Samsung P50", AD1986A_SAMSUNG_P50),
 	SND_PCI_QUIRK(0x144d, 0xc027, "Samsung Q1", AD1986A_ULTRA),
 	SND_PCI_QUIRK_MASK(0x144d, 0xff00, 0xc000, "Samsung", AD1986A_SAMSUNG),
 	SND_PCI_QUIRK(0x144d, 0xc504, "Samsung Q35", AD1986A_3STACK),
@@ -1099,6 +1123,23 @@ static int patch_ad1986a(struct hda_codec *codec)
 		codec->patch_ops.unsol_event = ad1986a_automic_unsol_event;
 		codec->patch_ops.init = ad1986a_automic_init;
 		break;
+	case AD1986A_SAMSUNG_P50:
+		spec->num_mixers = 2;
+		spec->mixers[0] = ad1986a_automute_master_mixers;
+		spec->mixers[1] = ad1986a_laptop_eapd_mixers;
+		spec->num_init_verbs = 4;
+		spec->init_verbs[1] = ad1986a_eapd_init_verbs;
+		spec->init_verbs[2] = ad1986a_automic_verbs;
+		spec->init_verbs[3] = ad1986a_hp_init_verbs;
+		spec->multiout.max_channels = 2;
+		spec->multiout.num_dacs = 1;
+		spec->multiout.dac_nids = ad1986a_laptop_dac_nids;
+		if (!is_jack_available(codec, 0x25))
+			spec->multiout.dig_out_nid = 0;
+		spec->input_mux = &ad1986a_automic_capture_source;
+		codec->patch_ops.unsol_event = ad1986a_samsung_p50_unsol_event;
+		codec->patch_ops.init = ad1986a_samsung_p50_init;
+		break;
 	case AD1986A_LAPTOP_AUTOMUTE:
 		spec->num_mixers = 3;
 		spec->mixers[0] = ad1986a_automute_master_mixers;
-- 
cgit v1.2.3-59-g8ed1b


From b5450d9c84bdd38b261922057cd167da51dfae93 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Fri, 19 Jun 2009 10:30:07 +0200
Subject: reiserfs: remove stray unlock_super in reiserfs_resize

Reiserfs doesn't use lock_super anywhere internally, and ->remount_fs
which calls reiserfs_resize does have it currently but also expects it
to be held on return, so there's no business for the unlock_super here.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked by Edward Shishkin <edward.shishkin@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/reiserfs/resize.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/fs/reiserfs/resize.c b/fs/reiserfs/resize.c
index 238e9d9b31e0..18b315d3d104 100644
--- a/fs/reiserfs/resize.c
+++ b/fs/reiserfs/resize.c
@@ -82,7 +82,6 @@ int reiserfs_resize(struct super_block *s, unsigned long block_count_new)
 		if (reiserfs_allocate_list_bitmaps(s, jbitmap, bmap_nr_new) < 0) {
 			printk
 			    ("reiserfs_resize: unable to allocate memory for journal bitmaps\n");
-			unlock_super(s);
 			return -ENOMEM;
 		}
 		/* the new journal bitmaps are zero filled, now we copy in the bitmap
-- 
cgit v1.2.3-59-g8ed1b


From 654f562c526cf9dfb8d453f687341fe0777ee454 Mon Sep 17 00:00:00 2001
From: "J. R. Okajima" <hooanon05@yahoo.co.jp>
Date: Thu, 18 Jun 2009 23:30:15 +0900
Subject: vfs: fix nd->root leak in do_filp_open()

commit 2a737871108de9ba8930f7650d549f1383767f8b "Cache root in nameidata"
introduced a new member nd->root, but forgot to put it in do_filp_open().

Signed-off-by: J. R. Okajima <hooanon05@yahoo.co.jp>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/fs/namei.c b/fs/namei.c
index 527119afb6a5..5b961eb71cbf 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1698,8 +1698,11 @@ struct file *do_filp_open(int dfd, const char *pathname,
 	if (error)
 		return ERR_PTR(error);
 	error = path_walk(pathname, &nd);
-	if (error)
+	if (error) {
+		if (nd.root.mnt)
+			path_put(&nd.root);
 		return ERR_PTR(error);
+	}
 	if (unlikely(!audit_dummy_context()))
 		audit_inode(pathname, nd.path.dentry);
 
@@ -1759,6 +1762,8 @@ do_last:
 		}
 		filp = nameidata_to_filp(&nd, open_flag);
 		mnt_drop_write(nd.path.mnt);
+		if (nd.root.mnt)
+			path_put(&nd.root);
 		return filp;
 	}
 
@@ -1819,6 +1824,8 @@ ok:
 	 */
 	if (will_write)
 		mnt_drop_write(nd.path.mnt);
+	if (nd.root.mnt)
+		path_put(&nd.root);
 	return filp;
 
 exit_mutex_unlock:
@@ -1859,6 +1866,8 @@ do_link:
 		 * with "intent.open".
 		 */
 		release_open_intent(&nd);
+		if (nd.root.mnt)
+			path_put(&nd.root);
 		return ERR_PTR(error);
 	}
 	nd.flags &= ~LOOKUP_PARENT;
-- 
cgit v1.2.3-59-g8ed1b


From 3b22edc5730b87d360ee7dd7143397ba09b73a47 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Tue, 23 Jun 2009 17:29:49 -0400
Subject: VFS: Switch init_mount_tree() to use the new create_mnt_ns() helper

Eliminates some duplicated code...

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namespace.c | 11 ++---------
 1 file changed, 2 insertions(+), 9 deletions(-)

diff --git a/fs/namespace.c b/fs/namespace.c
index a7bea8c8bd46..4a86b8595164 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -2222,16 +2222,9 @@ static void __init init_mount_tree(void)
 	mnt = do_kern_mount("rootfs", 0, "rootfs", NULL);
 	if (IS_ERR(mnt))
 		panic("Can't create rootfs");
-	ns = kmalloc(sizeof(*ns), GFP_KERNEL);
-	if (!ns)
+	ns = create_mnt_ns(mnt);
+	if (IS_ERR(ns))
 		panic("Can't allocate initial namespace");
-	atomic_set(&ns->count, 1);
-	INIT_LIST_HEAD(&ns->list);
-	init_waitqueue_head(&ns->poll);
-	ns->event = 0;
-	list_add(&mnt->mnt_list, &ns->list);
-	ns->root = mnt;
-	mnt->mnt_ns = ns;
 
 	init_task.nsproxy->mnt_ns = ns;
 	get_mnt_ns(ns);
-- 
cgit v1.2.3-59-g8ed1b


From f6cc746bbb3b8a8ceb8514a7906ba582607a8cf7 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Tue, 16 Jun 2009 21:15:04 -0700
Subject: devpts: remove module-related code

These days, the devpts filesystem is closely integrated with the pty
memory management, and cannot be built as a module, even less removed
from the kernel.  Accordingly, remove all module-related stuff from
this filesystem.

[ v2: only remove code that's actually dead ]

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/devpts/inode.c | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index 9b1d285f9fe6..75efb028974b 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -423,7 +423,6 @@ static void devpts_kill_sb(struct super_block *sb)
 }
 
 static struct file_system_type devpts_fs_type = {
-	.owner		= THIS_MODULE,
 	.name		= "devpts",
 	.get_sb		= devpts_get_sb,
 	.kill_sb	= devpts_kill_sb,
@@ -564,13 +563,4 @@ static int __init init_devpts_fs(void)
 	}
 	return err;
 }
-
-static void __exit exit_devpts_fs(void)
-{
-	unregister_filesystem(&devpts_fs_type);
-	mntput(devpts_mnt);
-}
-
 module_init(init_devpts_fs)
-module_exit(exit_devpts_fs)
-MODULE_LICENSE("GPL");
-- 
cgit v1.2.3-59-g8ed1b


From 7e325d3a6b117c7288bfc0755410e9d9d2b71326 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Fri, 19 Jun 2009 20:22:37 +0200
Subject: update Documentation/filesystems/Locking

The rules for locking in many superblock operations has changed
significantly, so update the documentation for it.  Also correct some
older updates and ommissions.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 Documentation/filesystems/Locking | 43 ++++++++++++++++++++-------------------
 1 file changed, 22 insertions(+), 21 deletions(-)

diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index 229d7b7c50a3..18b9d0ca0630 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -109,27 +109,28 @@ prototypes:
 
 locking rules:
 	All may block.
-			BKL	s_lock	s_umount
-alloc_inode:		no	no	no
-destroy_inode:		no
-dirty_inode:		no				(must not sleep)
-write_inode:		no
-drop_inode:		no				!!!inode_lock!!!
-delete_inode:		no
-put_super:		yes	yes	no
-write_super:		no	yes	read
-sync_fs:		no	no	read
-freeze_fs:		?
-unfreeze_fs:		?
-statfs:			no	no	no
-remount_fs:		yes	yes	maybe		(see below)
-clear_inode:		no
-umount_begin:		yes	no	no
-show_options:		no				(vfsmount->sem)
-quota_read:		no	no	no		(see below)
-quota_write:		no	no	no		(see below)
-
-->remount_fs() will have the s_umount lock if it's already mounted.
+	None have BKL
+			s_umount
+alloc_inode:
+destroy_inode:
+dirty_inode:				(must not sleep)
+write_inode:
+drop_inode:				!!!inode_lock!!!
+delete_inode:
+put_super:		write
+write_super:		read
+sync_fs:		read
+freeze_fs:		read
+unfreeze_fs:		read
+statfs:			no
+remount_fs:		maybe		(see below)
+clear_inode:
+umount_begin:		no
+show_options:		no		(namespace_sem)
+quota_read:		no		(see below)
+quota_write:		no		(see below)
+
+->remount_fs() will have the s_umount exclusive lock if it's already mounted.
 When called from get_sb_single, it does NOT have the s_umount lock.
 ->quota_read() and ->quota_write() functions are both guaranteed to
 be the only ones operating on the quota file by the quota code (via
-- 
cgit v1.2.3-59-g8ed1b


From c63e09ecccb50f930e899d7005edc5411ee86d4f Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 24 Jun 2009 02:05:18 -0400
Subject: Make allocation of anon devices cheaper

Standard trick - add a new variable (start) such that
for each n < start n is known to be busy.  Allocation can
skip checking everything in [0..start) and if it returns
n, we can set start to n + 1.  Freeing below start sets
start to what we'd just freed.

Of course, it still sucks if we do something like
	free 0
	allocate
	allocate
in a loop - still O(n^2) time.  However, on saner loads it
improves the things a lot and the entire thing is not worth
the trouble of switching to something with better worst-case
behaviour.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/super.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/fs/super.c b/fs/super.c
index d40d53a22fb5..808ffd59e01b 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -608,6 +608,7 @@ void emergency_remount(void)
 
 static DEFINE_IDA(unnamed_dev_ida);
 static DEFINE_SPINLOCK(unnamed_dev_lock);/* protects the above */
+static int unnamed_dev_start = 0; /* don't bother trying below it */
 
 int set_anon_super(struct super_block *s, void *data)
 {
@@ -618,7 +619,8 @@ int set_anon_super(struct super_block *s, void *data)
 	if (ida_pre_get(&unnamed_dev_ida, GFP_ATOMIC) == 0)
 		return -ENOMEM;
 	spin_lock(&unnamed_dev_lock);
-	error = ida_get_new(&unnamed_dev_ida, &dev);
+	error = ida_get_new_above(&unnamed_dev_ida, unnamed_dev_start, &dev);
+	unnamed_dev_start = dev + 1;
 	spin_unlock(&unnamed_dev_lock);
 	if (error == -EAGAIN)
 		/* We raced and lost with another CPU. */
@@ -629,6 +631,7 @@ int set_anon_super(struct super_block *s, void *data)
 	if ((dev & MAX_ID_MASK) == (1 << MINORBITS)) {
 		spin_lock(&unnamed_dev_lock);
 		ida_remove(&unnamed_dev_ida, dev);
+		unnamed_dev_start = dev;
 		spin_unlock(&unnamed_dev_lock);
 		return -EMFILE;
 	}
@@ -645,6 +648,8 @@ void kill_anon_super(struct super_block *sb)
 	generic_shutdown_super(sb);
 	spin_lock(&unnamed_dev_lock);
 	ida_remove(&unnamed_dev_ida, slot);
+	if (slot < unnamed_dev_start)
+		unnamed_dev_start = slot;
 	spin_unlock(&unnamed_dev_lock);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From f21f62208a6f60e2e05440b2e438d9541822dc4d Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 24 Jun 2009 03:12:00 -0400
Subject: ... and the same for vfsmount id/mount group id

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namespace.c | 26 ++++++++++++++++++++++----
 fs/super.c     |  6 ++++--
 2 files changed, 26 insertions(+), 6 deletions(-)

diff --git a/fs/namespace.c b/fs/namespace.c
index 4a86b8595164..3dc283fd4716 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -42,6 +42,8 @@ __cacheline_aligned_in_smp DEFINE_SPINLOCK(vfsmount_lock);
 static int event;
 static DEFINE_IDA(mnt_id_ida);
 static DEFINE_IDA(mnt_group_ida);
+static int mnt_id_start = 0;
+static int mnt_group_start = 1;
 
 static struct list_head *mount_hashtable __read_mostly;
 static struct kmem_cache *mnt_cache __read_mostly;
@@ -69,7 +71,9 @@ static int mnt_alloc_id(struct vfsmount *mnt)
 retry:
 	ida_pre_get(&mnt_id_ida, GFP_KERNEL);
 	spin_lock(&vfsmount_lock);
-	res = ida_get_new(&mnt_id_ida, &mnt->mnt_id);
+	res = ida_get_new_above(&mnt_id_ida, mnt_id_start, &mnt->mnt_id);
+	if (!res)
+		mnt_id_start = mnt->mnt_id + 1;
 	spin_unlock(&vfsmount_lock);
 	if (res == -EAGAIN)
 		goto retry;
@@ -79,8 +83,11 @@ retry:
 
 static void mnt_free_id(struct vfsmount *mnt)
 {
+	int id = mnt->mnt_id;
 	spin_lock(&vfsmount_lock);
-	ida_remove(&mnt_id_ida, mnt->mnt_id);
+	ida_remove(&mnt_id_ida, id);
+	if (mnt_id_start > id)
+		mnt_id_start = id;
 	spin_unlock(&vfsmount_lock);
 }
 
@@ -91,10 +98,18 @@ static void mnt_free_id(struct vfsmount *mnt)
  */
 static int mnt_alloc_group_id(struct vfsmount *mnt)
 {
+	int res;
+
 	if (!ida_pre_get(&mnt_group_ida, GFP_KERNEL))
 		return -ENOMEM;
 
-	return ida_get_new_above(&mnt_group_ida, 1, &mnt->mnt_group_id);
+	res = ida_get_new_above(&mnt_group_ida,
+				mnt_group_start,
+				&mnt->mnt_group_id);
+	if (!res)
+		mnt_group_start = mnt->mnt_group_id + 1;
+
+	return res;
 }
 
 /*
@@ -102,7 +117,10 @@ static int mnt_alloc_group_id(struct vfsmount *mnt)
  */
 void mnt_release_group_id(struct vfsmount *mnt)
 {
-	ida_remove(&mnt_group_ida, mnt->mnt_group_id);
+	int id = mnt->mnt_group_id;
+	ida_remove(&mnt_group_ida, id);
+	if (mnt_group_start > id)
+		mnt_group_start = id;
 	mnt->mnt_group_id = 0;
 }
 
diff --git a/fs/super.c b/fs/super.c
index 808ffd59e01b..2761d3e22ed9 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -620,7 +620,8 @@ int set_anon_super(struct super_block *s, void *data)
 		return -ENOMEM;
 	spin_lock(&unnamed_dev_lock);
 	error = ida_get_new_above(&unnamed_dev_ida, unnamed_dev_start, &dev);
-	unnamed_dev_start = dev + 1;
+	if (!error)
+		unnamed_dev_start = dev + 1;
 	spin_unlock(&unnamed_dev_lock);
 	if (error == -EAGAIN)
 		/* We raced and lost with another CPU. */
@@ -631,7 +632,8 @@ int set_anon_super(struct super_block *s, void *data)
 	if ((dev & MAX_ID_MASK) == (1 << MINORBITS)) {
 		spin_lock(&unnamed_dev_lock);
 		ida_remove(&unnamed_dev_ida, dev);
-		unnamed_dev_start = dev;
+		if (unnamed_dev_start > dev)
+			unnamed_dev_start = dev;
 		spin_unlock(&unnamed_dev_lock);
 		return -EMFILE;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 01c031945f2755c7afaaf456088543312f2b72ea Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Mon, 8 Jun 2009 13:35:40 +0200
Subject: cleanup __writeback_single_inode

There is no reason to for the split between __writeback_single_inode and
__sync_single_inode, the former just does a couple of checks before
tail-calling the latter.  So merge the two, and while we're at it split
out the I_SYNC waiting case for data integrity writers, as it's
logically separate function.  Finally rename __writeback_single_inode to
writeback_single_inode.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/fs-writeback.c | 100 +++++++++++++++++++++++++++---------------------------
 1 file changed, 50 insertions(+), 50 deletions(-)

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index caf049146ca2..c54226be5294 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -278,7 +278,26 @@ int sb_has_dirty_inodes(struct super_block *sb)
 EXPORT_SYMBOL(sb_has_dirty_inodes);
 
 /*
- * Write a single inode's dirty pages and inode data out to disk.
+ * Wait for writeback on an inode to complete.
+ */
+static void inode_wait_for_writeback(struct inode *inode)
+{
+	DEFINE_WAIT_BIT(wq, &inode->i_state, __I_SYNC);
+	wait_queue_head_t *wqh;
+
+	wqh = bit_waitqueue(&inode->i_state, __I_SYNC);
+	do {
+		spin_unlock(&inode_lock);
+		__wait_on_bit(wqh, &wq, inode_wait, TASK_UNINTERRUPTIBLE);
+		spin_lock(&inode_lock);
+	} while (inode->i_state & I_SYNC);
+}
+
+/*
+ * Write out an inode's dirty pages.  Called under inode_lock.  Either the
+ * caller has ref on the inode (either via __iget or via syscall against an fd)
+ * or the inode has I_WILL_FREE set (via generic_forget_inode)
+ *
  * If `wait' is set, wait on the writeout.
  *
  * The whole writeout design is quite complex and fragile.  We want to avoid
@@ -288,13 +307,38 @@ EXPORT_SYMBOL(sb_has_dirty_inodes);
  * Called under inode_lock.
  */
 static int
-__sync_single_inode(struct inode *inode, struct writeback_control *wbc)
+writeback_single_inode(struct inode *inode, struct writeback_control *wbc)
 {
-	unsigned dirty;
 	struct address_space *mapping = inode->i_mapping;
 	int wait = wbc->sync_mode == WB_SYNC_ALL;
+	unsigned dirty;
 	int ret;
 
+	if (!atomic_read(&inode->i_count))
+		WARN_ON(!(inode->i_state & (I_WILL_FREE|I_FREEING)));
+	else
+		WARN_ON(inode->i_state & I_WILL_FREE);
+
+	if (inode->i_state & I_SYNC) {
+		/*
+		 * If this inode is locked for writeback and we are not doing
+		 * writeback-for-data-integrity, move it to s_more_io so that
+		 * writeback can proceed with the other inodes on s_io.
+		 *
+		 * We'll have another go at writing back this inode when we
+		 * completed a full scan of s_io.
+		 */
+		if (!wait) {
+			requeue_io(inode);
+			return 0;
+		}
+
+		/*
+		 * It's a data-integrity sync.  We must wait.
+		 */
+		inode_wait_for_writeback(inode);
+	}
+
 	BUG_ON(inode->i_state & I_SYNC);
 
 	/* Set I_SYNC, reset I_DIRTY */
@@ -389,50 +433,6 @@ __sync_single_inode(struct inode *inode, struct writeback_control *wbc)
 	return ret;
 }
 
-/*
- * Write out an inode's dirty pages.  Called under inode_lock.  Either the
- * caller has ref on the inode (either via __iget or via syscall against an fd)
- * or the inode has I_WILL_FREE set (via generic_forget_inode)
- */
-static int
-__writeback_single_inode(struct inode *inode, struct writeback_control *wbc)
-{
-	wait_queue_head_t *wqh;
-
-	if (!atomic_read(&inode->i_count))
-		WARN_ON(!(inode->i_state & (I_WILL_FREE|I_FREEING)));
-	else
-		WARN_ON(inode->i_state & I_WILL_FREE);
-
-	if ((wbc->sync_mode != WB_SYNC_ALL) && (inode->i_state & I_SYNC)) {
-		/*
-		 * We're skipping this inode because it's locked, and we're not
-		 * doing writeback-for-data-integrity.  Move it to s_more_io so
-		 * that writeback can proceed with the other inodes on s_io.
-		 * We'll have another go at writing back this inode when we
-		 * completed a full scan of s_io.
-		 */
-		requeue_io(inode);
-		return 0;
-	}
-
-	/*
-	 * It's a data-integrity sync.  We must wait.
-	 */
-	if (inode->i_state & I_SYNC) {
-		DEFINE_WAIT_BIT(wq, &inode->i_state, __I_SYNC);
-
-		wqh = bit_waitqueue(&inode->i_state, __I_SYNC);
-		do {
-			spin_unlock(&inode_lock);
-			__wait_on_bit(wqh, &wq, inode_wait,
-							TASK_UNINTERRUPTIBLE);
-			spin_lock(&inode_lock);
-		} while (inode->i_state & I_SYNC);
-	}
-	return __sync_single_inode(inode, wbc);
-}
-
 /*
  * Write out a superblock's list of dirty inodes.  A wait will be performed
  * upon no inodes, all inodes or the final one, depending upon sync_mode.
@@ -526,7 +526,7 @@ void generic_sync_sb_inodes(struct super_block *sb,
 		BUG_ON(inode->i_state & (I_FREEING | I_CLEAR));
 		__iget(inode);
 		pages_skipped = wbc->pages_skipped;
-		__writeback_single_inode(inode, wbc);
+		writeback_single_inode(inode, wbc);
 		if (current_is_pdflush())
 			writeback_release(bdi);
 		if (wbc->pages_skipped != pages_skipped) {
@@ -708,7 +708,7 @@ int write_inode_now(struct inode *inode, int sync)
 
 	might_sleep();
 	spin_lock(&inode_lock);
-	ret = __writeback_single_inode(inode, &wbc);
+	ret = writeback_single_inode(inode, &wbc);
 	spin_unlock(&inode_lock);
 	if (sync)
 		inode_sync_wait(inode);
@@ -732,7 +732,7 @@ int sync_inode(struct inode *inode, struct writeback_control *wbc)
 	int ret;
 
 	spin_lock(&inode_lock);
-	ret = __writeback_single_inode(inode, wbc);
+	ret = writeback_single_inode(inode, wbc);
 	spin_unlock(&inode_lock);
 	return ret;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 3e63cbb1efca7dd3137de1bb475e2e068e38ef23 Mon Sep 17 00:00:00 2001
From: Ankit Jain <me@ankitjain.org>
Date: Fri, 19 Jun 2009 14:28:07 -0400
Subject: fs: Add new pre-allocation ioctls to vfs for compatibility with
 legacy xfs ioctls

This patch adds ioctls to vfs for compatibility with legacy XFS
pre-allocation ioctls (XFS_IOC_*RESVP*). The implementation
effectively invokes sys_fallocate for the new ioctls.
Also handles the compat_ioctl case.
Note: These legacy ioctls are also implemented by OCFS2.

[AV: folded fixes from hch]

Signed-off-by: Ankit Jain <me@ankitjain.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/compat_ioctl.c      | 48 +++++++++++++++++++++++++++++++++++++++++
 fs/ioctl.c             | 35 ++++++++++++++++++++++++++++++
 fs/open.c              | 58 +++++++++++++++++++++++++-------------------------
 include/linux/falloc.h | 21 ++++++++++++++++++
 include/linux/fs.h     |  6 ++++++
 5 files changed, 139 insertions(+), 29 deletions(-)

diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index c135202c38b3..626c7483b4de 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -31,6 +31,7 @@
 #include <linux/skbuff.h>
 #include <linux/netlink.h>
 #include <linux/vt.h>
+#include <linux/falloc.h>
 #include <linux/fs.h>
 #include <linux/file.h>
 #include <linux/ppp_defs.h>
@@ -1779,6 +1780,41 @@ lp_timeout_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
 	return sys_ioctl(fd, cmd, (unsigned long)tn);
 }
 
+/* on ia32 l_start is on a 32-bit boundary */
+#if defined(CONFIG_IA64) || defined(CONFIG_X86_64)
+struct space_resv_32 {
+	__s16		l_type;
+	__s16		l_whence;
+	__s64		l_start	__attribute__((packed));
+			/* len == 0 means until end of file */
+	__s64		l_len __attribute__((packed));
+	__s32		l_sysid;
+	__u32		l_pid;
+	__s32		l_pad[4];	/* reserve area */
+};
+
+#define FS_IOC_RESVSP_32		_IOW ('X', 40, struct space_resv_32)
+#define FS_IOC_RESVSP64_32	_IOW ('X', 42, struct space_resv_32)
+
+/* just account for different alignment */
+static int compat_ioctl_preallocate(struct file *file, unsigned long arg)
+{
+	struct space_resv_32	__user *p32 = (void __user *)arg;
+	struct space_resv	__user *p = compat_alloc_user_space(sizeof(*p));
+
+	if (copy_in_user(&p->l_type,	&p32->l_type,	sizeof(s16)) ||
+	    copy_in_user(&p->l_whence,	&p32->l_whence, sizeof(s16)) ||
+	    copy_in_user(&p->l_start,	&p32->l_start,	sizeof(s64)) ||
+	    copy_in_user(&p->l_len,	&p32->l_len,	sizeof(s64)) ||
+	    copy_in_user(&p->l_sysid,	&p32->l_sysid,	sizeof(s32)) ||
+	    copy_in_user(&p->l_pid,	&p32->l_pid,	sizeof(u32)) ||
+	    copy_in_user(&p->l_pad,	&p32->l_pad,	4*sizeof(u32)))
+		return -EFAULT;
+
+	return ioctl_preallocate(file, p);
+}
+#endif
+
 
 typedef int (*ioctl_trans_handler_t)(unsigned int, unsigned int,
 					unsigned long, struct file *);
@@ -2756,6 +2792,18 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd,
 	case FIOQSIZE:
 		break;
 
+#if defined(CONFIG_IA64) || defined(CONFIG_X86_64)
+	case FS_IOC_RESVSP_32:
+	case FS_IOC_RESVSP64_32:
+		error = compat_ioctl_preallocate(filp, arg);
+		goto out_fput;
+#else
+	case FS_IOC_RESVSP:
+	case FS_IOC_RESVSP64:
+		error = ioctl_preallocate(filp, (void __user *)arg);
+		goto out_fput;
+#endif
+
 	case FIBMAP:
 	case FIGETBSZ:
 	case FIONREAD:
diff --git a/fs/ioctl.c b/fs/ioctl.c
index 001f8d3118f2..5612880fcbe7 100644
--- a/fs/ioctl.c
+++ b/fs/ioctl.c
@@ -15,6 +15,7 @@
 #include <linux/uaccess.h>
 #include <linux/writeback.h>
 #include <linux/buffer_head.h>
+#include <linux/falloc.h>
 
 #include <asm/ioctls.h>
 
@@ -403,6 +404,37 @@ EXPORT_SYMBOL(generic_block_fiemap);
 
 #endif  /*  CONFIG_BLOCK  */
 
+/*
+ * This provides compatibility with legacy XFS pre-allocation ioctls
+ * which predate the fallocate syscall.
+ *
+ * Only the l_start, l_len and l_whence fields of the 'struct space_resv'
+ * are used here, rest are ignored.
+ */
+int ioctl_preallocate(struct file *filp, void __user *argp)
+{
+	struct inode *inode = filp->f_path.dentry->d_inode;
+	struct space_resv sr;
+
+	if (copy_from_user(&sr, argp, sizeof(sr)))
+		return -EFAULT;
+
+	switch (sr.l_whence) {
+	case SEEK_SET:
+		break;
+	case SEEK_CUR:
+		sr.l_start += filp->f_pos;
+		break;
+	case SEEK_END:
+		sr.l_start += i_size_read(inode);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return do_fallocate(filp, FALLOC_FL_KEEP_SIZE, sr.l_start, sr.l_len);
+}
+
 static int file_ioctl(struct file *filp, unsigned int cmd,
 		unsigned long arg)
 {
@@ -414,6 +446,9 @@ static int file_ioctl(struct file *filp, unsigned int cmd,
 		return ioctl_fibmap(filp, p);
 	case FIONREAD:
 		return put_user(i_size_read(inode) - filp->f_pos, p);
+	case FS_IOC_RESVSP:
+	case FS_IOC_RESVSP64:
+		return ioctl_preallocate(filp, p);
 	}
 
 	return vfs_ioctl(filp, cmd, arg);
diff --git a/fs/open.c b/fs/open.c
index 7200e23d9258..dd98e8076024 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -378,63 +378,63 @@ SYSCALL_ALIAS(sys_ftruncate64, SyS_ftruncate64);
 #endif
 #endif /* BITS_PER_LONG == 32 */
 
-SYSCALL_DEFINE(fallocate)(int fd, int mode, loff_t offset, loff_t len)
+
+int do_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
 {
-	struct file *file;
-	struct inode *inode;
-	long ret = -EINVAL;
+	struct inode *inode = file->f_path.dentry->d_inode;
+	long ret;
 
 	if (offset < 0 || len <= 0)
-		goto out;
+		return -EINVAL;
 
 	/* Return error if mode is not supported */
-	ret = -EOPNOTSUPP;
 	if (mode && !(mode & FALLOC_FL_KEEP_SIZE))
-		goto out;
+		return -EOPNOTSUPP;
 
-	ret = -EBADF;
-	file = fget(fd);
-	if (!file)
-		goto out;
 	if (!(file->f_mode & FMODE_WRITE))
-		goto out_fput;
+		return -EBADF;
 	/*
 	 * Revalidate the write permissions, in case security policy has
 	 * changed since the files were opened.
 	 */
 	ret = security_file_permission(file, MAY_WRITE);
 	if (ret)
-		goto out_fput;
+		return ret;
 
-	inode = file->f_path.dentry->d_inode;
-
-	ret = -ESPIPE;
 	if (S_ISFIFO(inode->i_mode))
-		goto out_fput;
+		return -ESPIPE;
 
-	ret = -ENODEV;
 	/*
 	 * Let individual file system decide if it supports preallocation
 	 * for directories or not.
 	 */
 	if (!S_ISREG(inode->i_mode) && !S_ISDIR(inode->i_mode))
-		goto out_fput;
+		return -ENODEV;
 
-	ret = -EFBIG;
 	/* Check for wrap through zero too */
 	if (((offset + len) > inode->i_sb->s_maxbytes) || ((offset + len) < 0))
-		goto out_fput;
+		return -EFBIG;
 
-	if (inode->i_op->fallocate)
-		ret = inode->i_op->fallocate(inode, mode, offset, len);
-	else
-		ret = -EOPNOTSUPP;
+	if (!inode->i_op->fallocate)
+		return -EOPNOTSUPP;
 
-out_fput:
-	fput(file);
-out:
-	return ret;
+	return inode->i_op->fallocate(inode, mode, offset, len);
 }
+
+SYSCALL_DEFINE(fallocate)(int fd, int mode, loff_t offset, loff_t len)
+{
+	struct file *file;
+	int error = -EBADF;
+
+	file = fget(fd);
+	if (file) {
+		error = do_fallocate(file, mode, offset, len);
+		fput(file);
+	}
+
+	return error;
+}
+
 #ifdef CONFIG_HAVE_SYSCALL_WRAPPERS
 asmlinkage long SyS_fallocate(long fd, long mode, loff_t offset, loff_t len)
 {
diff --git a/include/linux/falloc.h b/include/linux/falloc.h
index 8e912ab6a072..3c155107d61f 100644
--- a/include/linux/falloc.h
+++ b/include/linux/falloc.h
@@ -3,4 +3,25 @@
 
 #define FALLOC_FL_KEEP_SIZE	0x01 /* default is extend size */
 
+#ifdef __KERNEL__
+
+/*
+ * Space reservation ioctls and argument structure
+ * are designed to be compatible with the legacy XFS ioctls.
+ */
+struct space_resv {
+	__s16		l_type;
+	__s16		l_whence;
+	__s64		l_start;
+	__s64		l_len;		/* len == 0 means until end of file */
+	__s32		l_sysid;
+	__u32		l_pid;
+	__s32		l_pad[4];	/* reserved area */
+};
+
+#define FS_IOC_RESVSP		_IOW('X', 40, struct space_resv)
+#define FS_IOC_RESVSP64		_IOW('X', 42, struct space_resv)
+
+#endif /* __KERNEL__ */
+
 #endif /* _FALLOC_H_ */
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 1ff5e4e01952..79e302ddde0f 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1906,6 +1906,8 @@ static inline int break_lease(struct inode *inode, unsigned int mode)
 
 extern int do_truncate(struct dentry *, loff_t start, unsigned int time_attrs,
 		       struct file *filp);
+extern int do_fallocate(struct file *file, int mode, loff_t offset,
+			loff_t len);
 extern long do_sys_open(int dfd, const char __user *filename, int flags,
 			int mode);
 extern struct file *filp_open(const char *, int, int);
@@ -1914,6 +1916,10 @@ extern struct file * dentry_open(struct dentry *, struct vfsmount *, int,
 extern int filp_close(struct file *, fl_owner_t id);
 extern char * getname(const char __user *);
 
+/* fs/ioctl.c */
+
+extern int ioctl_preallocate(struct file *filp, void __user *argp);
+
 /* fs/dcache.c */
 extern void __init vfs_caches_init_early(void);
 extern void __init vfs_caches_init(unsigned long);
-- 
cgit v1.2.3-59-g8ed1b


From f19d4a8fa6f9b6ccf54df0971c97ffcaa390b7b0 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 8 Jun 2009 19:50:45 -0400
Subject: add caching of ACLs in struct inode

No helpers, no conversions yet.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/inode.c         | 10 ++++++++++
 include/linux/fs.h |  7 +++++++
 2 files changed, 17 insertions(+)

diff --git a/fs/inode.c b/fs/inode.c
index f643be565df8..e193cd592fa8 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -25,6 +25,7 @@
 #include <linux/fsnotify.h>
 #include <linux/mount.h>
 #include <linux/async.h>
+#include <linux/posix_acl.h>
 
 /*
  * This is needed for the following functions:
@@ -189,6 +190,9 @@ struct inode *inode_init_always(struct super_block *sb, struct inode *inode)
 	}
 	inode->i_private = NULL;
 	inode->i_mapping = mapping;
+#ifdef CONFIG_FS_POSIX_ACL
+	inode->i_acl = inode->i_default_acl = ACL_NOT_CACHED;
+#endif
 
 #ifdef CONFIG_FSNOTIFY
 	inode->i_fsnotify_mask = 0;
@@ -227,6 +231,12 @@ void destroy_inode(struct inode *inode)
 	ima_inode_free(inode);
 	security_inode_free(inode);
 	fsnotify_inode_delete(inode);
+#ifdef CONFIG_FS_POSIX_ACL
+	if (inode->i_acl && inode->i_acl != ACL_NOT_CACHED)
+		posix_acl_release(inode->i_acl);
+	if (inode->i_default_acl && inode->i_default_acl != ACL_NOT_CACHED)
+		posix_acl_release(inode->i_default_acl);
+#endif
 	if (inode->i_sb->s_op->destroy_inode)
 		inode->i_sb->s_op->destroy_inode(inode);
 	else
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 79e302ddde0f..0872372184fe 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -710,6 +710,9 @@ static inline int mapping_writably_mapped(struct address_space *mapping)
 #define i_size_ordered_init(inode) do { } while (0)
 #endif
 
+struct posix_acl;
+#define ACL_NOT_CACHED ((void *)(-1))
+
 struct inode {
 	struct hlist_node	i_hash;
 	struct list_head	i_list;
@@ -772,6 +775,10 @@ struct inode {
 	atomic_t		i_writecount;
 #ifdef CONFIG_SECURITY
 	void			*i_security;
+#endif
+#ifdef CONFIG_FS_POSIX_ACL
+	struct posix_acl	*i_acl;
+	struct posix_acl	*i_default_acl;
 #endif
 	void			*i_private; /* fs or device private pointer */
 };
-- 
cgit v1.2.3-59-g8ed1b


From 5e78b435683daaaacadad1b2aeefb8904cf6acfb Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 8 Jun 2009 19:52:55 -0400
Subject: switch ext2 to inode->i_acl

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/ext2/acl.c   | 24 +++++++++++-------------
 fs/ext2/acl.h   |  4 ----
 fs/ext2/ext2.h  |  4 ----
 fs/ext2/inode.c |  4 ----
 fs/ext2/super.c | 16 ----------------
 5 files changed, 11 insertions(+), 41 deletions(-)

diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c
index d46e38cb85c5..d2ffddc12117 100644
--- a/fs/ext2/acl.c
+++ b/fs/ext2/acl.c
@@ -128,10 +128,10 @@ fail:
 static inline struct posix_acl *
 ext2_iget_acl(struct inode *inode, struct posix_acl **i_acl)
 {
-	struct posix_acl *acl = EXT2_ACL_NOT_CACHED;
+	struct posix_acl *acl = ACL_NOT_CACHED;
 
 	spin_lock(&inode->i_lock);
-	if (*i_acl != EXT2_ACL_NOT_CACHED)
+	if (*i_acl != ACL_NOT_CACHED)
 		acl = posix_acl_dup(*i_acl);
 	spin_unlock(&inode->i_lock);
 
@@ -143,7 +143,7 @@ ext2_iset_acl(struct inode *inode, struct posix_acl **i_acl,
 		   struct posix_acl *acl)
 {
 	spin_lock(&inode->i_lock);
-	if (*i_acl != EXT2_ACL_NOT_CACHED)
+	if (*i_acl != ACL_NOT_CACHED)
 		posix_acl_release(*i_acl);
 	*i_acl = posix_acl_dup(acl);
 	spin_unlock(&inode->i_lock);
@@ -155,7 +155,6 @@ ext2_iset_acl(struct inode *inode, struct posix_acl **i_acl,
 static struct posix_acl *
 ext2_get_acl(struct inode *inode, int type)
 {
-	struct ext2_inode_info *ei = EXT2_I(inode);
 	int name_index;
 	char *value = NULL;
 	struct posix_acl *acl;
@@ -166,15 +165,15 @@ ext2_get_acl(struct inode *inode, int type)
 
 	switch(type) {
 		case ACL_TYPE_ACCESS:
-			acl = ext2_iget_acl(inode, &ei->i_acl);
-			if (acl != EXT2_ACL_NOT_CACHED)
+			acl = ext2_iget_acl(inode, &inode->i_acl);
+			if (acl != ACL_NOT_CACHED)
 				return acl;
 			name_index = EXT2_XATTR_INDEX_POSIX_ACL_ACCESS;
 			break;
 
 		case ACL_TYPE_DEFAULT:
-			acl = ext2_iget_acl(inode, &ei->i_default_acl);
-			if (acl != EXT2_ACL_NOT_CACHED)
+			acl = ext2_iget_acl(inode, &inode->i_default_acl);
+			if (acl != ACL_NOT_CACHED)
 				return acl;
 			name_index = EXT2_XATTR_INDEX_POSIX_ACL_DEFAULT;
 			break;
@@ -200,11 +199,11 @@ ext2_get_acl(struct inode *inode, int type)
 	if (!IS_ERR(acl)) {
 		switch(type) {
 			case ACL_TYPE_ACCESS:
-				ext2_iset_acl(inode, &ei->i_acl, acl);
+				ext2_iset_acl(inode, &inode->i_acl, acl);
 				break;
 
 			case ACL_TYPE_DEFAULT:
-				ext2_iset_acl(inode, &ei->i_default_acl, acl);
+				ext2_iset_acl(inode, &inode->i_default_acl, acl);
 				break;
 		}
 	}
@@ -217,7 +216,6 @@ ext2_get_acl(struct inode *inode, int type)
 static int
 ext2_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 {
-	struct ext2_inode_info *ei = EXT2_I(inode);
 	int name_index;
 	void *value = NULL;
 	size_t size = 0;
@@ -266,11 +264,11 @@ ext2_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 	if (!error) {
 		switch(type) {
 			case ACL_TYPE_ACCESS:
-				ext2_iset_acl(inode, &ei->i_acl, acl);
+				ext2_iset_acl(inode, &inode->i_acl, acl);
 				break;
 
 			case ACL_TYPE_DEFAULT:
-				ext2_iset_acl(inode, &ei->i_default_acl, acl);
+				ext2_iset_acl(inode, &inode->i_default_acl, acl);
 				break;
 		}
 	}
diff --git a/fs/ext2/acl.h b/fs/ext2/acl.h
index b42cf578554b..ecefe478898f 100644
--- a/fs/ext2/acl.h
+++ b/fs/ext2/acl.h
@@ -53,10 +53,6 @@ static inline int ext2_acl_count(size_t size)
 
 #ifdef CONFIG_EXT2_FS_POSIX_ACL
 
-/* Value for inode->u.ext2_i.i_acl and inode->u.ext2_i.i_default_acl
-   if the ACL has not been cached */
-#define EXT2_ACL_NOT_CACHED ((void *)-1)
-
 /* acl.c */
 extern int ext2_permission (struct inode *, int);
 extern int ext2_acl_chmod (struct inode *);
diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h
index d988a718aedb..9a8a8e27a063 100644
--- a/fs/ext2/ext2.h
+++ b/fs/ext2/ext2.h
@@ -46,10 +46,6 @@ struct ext2_inode_info {
 	 * EAs.
 	 */
 	struct rw_semaphore xattr_sem;
-#endif
-#ifdef CONFIG_EXT2_FS_POSIX_ACL
-	struct posix_acl	*i_acl;
-	struct posix_acl	*i_default_acl;
 #endif
 	rwlock_t i_meta_lock;
 
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index 29ed682061f6..e27130341d4f 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -1224,10 +1224,6 @@ struct inode *ext2_iget (struct super_block *sb, unsigned long ino)
 		return inode;
 
 	ei = EXT2_I(inode);
-#ifdef CONFIG_EXT2_FS_POSIX_ACL
-	ei->i_acl = EXT2_ACL_NOT_CACHED;
-	ei->i_default_acl = EXT2_ACL_NOT_CACHED;
-#endif
 	ei->i_block_alloc_info = NULL;
 
 	raw_inode = ext2_get_inode(inode->i_sb, ino, &bh);
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 458999638c3d..1a9ffee47d56 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -152,10 +152,6 @@ static struct inode *ext2_alloc_inode(struct super_block *sb)
 	ei = (struct ext2_inode_info *)kmem_cache_alloc(ext2_inode_cachep, GFP_KERNEL);
 	if (!ei)
 		return NULL;
-#ifdef CONFIG_EXT2_FS_POSIX_ACL
-	ei->i_acl = EXT2_ACL_NOT_CACHED;
-	ei->i_default_acl = EXT2_ACL_NOT_CACHED;
-#endif
 	ei->i_block_alloc_info = NULL;
 	ei->vfs_inode.i_version = 1;
 	return &ei->vfs_inode;
@@ -198,18 +194,6 @@ static void destroy_inodecache(void)
 static void ext2_clear_inode(struct inode *inode)
 {
 	struct ext2_block_alloc_info *rsv = EXT2_I(inode)->i_block_alloc_info;
-#ifdef CONFIG_EXT2_FS_POSIX_ACL
-	struct ext2_inode_info *ei = EXT2_I(inode);
-
-	if (ei->i_acl && ei->i_acl != EXT2_ACL_NOT_CACHED) {
-		posix_acl_release(ei->i_acl);
-		ei->i_acl = EXT2_ACL_NOT_CACHED;
-	}
-	if (ei->i_default_acl && ei->i_default_acl != EXT2_ACL_NOT_CACHED) {
-		posix_acl_release(ei->i_default_acl);
-		ei->i_default_acl = EXT2_ACL_NOT_CACHED;
-	}
-#endif
 	ext2_discard_reservation(inode);
 	EXT2_I(inode)->i_block_alloc_info = NULL;
 	if (unlikely(rsv))
-- 
cgit v1.2.3-59-g8ed1b


From 6582a0e6f6bc7bf64817b9e1a424782855292ab0 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 8 Jun 2009 19:53:58 -0400
Subject: switch ext3 to inode->i_acl

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/ext3/acl.c             | 22 ++++++++++------------
 fs/ext3/acl.h             |  4 ----
 fs/ext3/inode.c           |  4 ----
 fs/ext3/super.c           | 16 ----------------
 include/linux/ext3_fs_i.h |  4 ----
 5 files changed, 10 insertions(+), 40 deletions(-)

diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c
index e0c745451715..a9707689d9e1 100644
--- a/fs/ext3/acl.c
+++ b/fs/ext3/acl.c
@@ -134,7 +134,7 @@ ext3_iget_acl(struct inode *inode, struct posix_acl **i_acl)
 	if (acl) {
 		spin_lock(&inode->i_lock);
 		acl = *i_acl;
-		if (acl != EXT3_ACL_NOT_CACHED)
+		if (acl != ACL_NOT_CACHED)
 			acl = posix_acl_dup(acl);
 		spin_unlock(&inode->i_lock);
 	}
@@ -147,7 +147,7 @@ ext3_iset_acl(struct inode *inode, struct posix_acl **i_acl,
                   struct posix_acl *acl)
 {
 	spin_lock(&inode->i_lock);
-	if (*i_acl != EXT3_ACL_NOT_CACHED)
+	if (*i_acl != ACL_NOT_CACHED)
 		posix_acl_release(*i_acl);
 	*i_acl = posix_acl_dup(acl);
 	spin_unlock(&inode->i_lock);
@@ -161,7 +161,6 @@ ext3_iset_acl(struct inode *inode, struct posix_acl **i_acl,
 static struct posix_acl *
 ext3_get_acl(struct inode *inode, int type)
 {
-	struct ext3_inode_info *ei = EXT3_I(inode);
 	int name_index;
 	char *value = NULL;
 	struct posix_acl *acl;
@@ -172,15 +171,15 @@ ext3_get_acl(struct inode *inode, int type)
 
 	switch(type) {
 		case ACL_TYPE_ACCESS:
-			acl = ext3_iget_acl(inode, &ei->i_acl);
-			if (acl != EXT3_ACL_NOT_CACHED)
+			acl = ext3_iget_acl(inode, &inode->i_acl);
+			if (acl != ACL_NOT_CACHED)
 				return acl;
 			name_index = EXT3_XATTR_INDEX_POSIX_ACL_ACCESS;
 			break;
 
 		case ACL_TYPE_DEFAULT:
-			acl = ext3_iget_acl(inode, &ei->i_default_acl);
-			if (acl != EXT3_ACL_NOT_CACHED)
+			acl = ext3_iget_acl(inode, &inode->i_default_acl);
+			if (acl != ACL_NOT_CACHED)
 				return acl;
 			name_index = EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT;
 			break;
@@ -206,11 +205,11 @@ ext3_get_acl(struct inode *inode, int type)
 	if (!IS_ERR(acl)) {
 		switch(type) {
 			case ACL_TYPE_ACCESS:
-				ext3_iset_acl(inode, &ei->i_acl, acl);
+				ext3_iset_acl(inode, &inode->i_acl, acl);
 				break;
 
 			case ACL_TYPE_DEFAULT:
-				ext3_iset_acl(inode, &ei->i_default_acl, acl);
+				ext3_iset_acl(inode, &inode->i_default_acl, acl);
 				break;
 		}
 	}
@@ -226,7 +225,6 @@ static int
 ext3_set_acl(handle_t *handle, struct inode *inode, int type,
 	     struct posix_acl *acl)
 {
-	struct ext3_inode_info *ei = EXT3_I(inode);
 	int name_index;
 	void *value = NULL;
 	size_t size = 0;
@@ -274,11 +272,11 @@ ext3_set_acl(handle_t *handle, struct inode *inode, int type,
 	if (!error) {
 		switch(type) {
 			case ACL_TYPE_ACCESS:
-				ext3_iset_acl(inode, &ei->i_acl, acl);
+				ext3_iset_acl(inode, &inode->i_acl, acl);
 				break;
 
 			case ACL_TYPE_DEFAULT:
-				ext3_iset_acl(inode, &ei->i_default_acl, acl);
+				ext3_iset_acl(inode, &inode->i_default_acl, acl);
 				break;
 		}
 	}
diff --git a/fs/ext3/acl.h b/fs/ext3/acl.h
index 42da16b8cac0..07d15a3a5969 100644
--- a/fs/ext3/acl.h
+++ b/fs/ext3/acl.h
@@ -53,10 +53,6 @@ static inline int ext3_acl_count(size_t size)
 
 #ifdef CONFIG_EXT3_FS_POSIX_ACL
 
-/* Value for inode->u.ext3_i.i_acl and inode->u.ext3_i.i_default_acl
-   if the ACL has not been cached */
-#define EXT3_ACL_NOT_CACHED ((void *)-1)
-
 /* acl.c */
 extern int ext3_permission (struct inode *, int);
 extern int ext3_acl_chmod (struct inode *);
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 05dea8132fc0..5f51fed5c750 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -2752,10 +2752,6 @@ struct inode *ext3_iget(struct super_block *sb, unsigned long ino)
 		return inode;
 
 	ei = EXT3_I(inode);
-#ifdef CONFIG_EXT3_FS_POSIX_ACL
-	ei->i_acl = EXT3_ACL_NOT_CACHED;
-	ei->i_default_acl = EXT3_ACL_NOT_CACHED;
-#endif
 	ei->i_block_alloc_info = NULL;
 
 	ret = __ext3_get_inode_loc(inode, &iloc, 0);
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 601e881e6105..524b349c6299 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -464,10 +464,6 @@ static struct inode *ext3_alloc_inode(struct super_block *sb)
 	ei = kmem_cache_alloc(ext3_inode_cachep, GFP_NOFS);
 	if (!ei)
 		return NULL;
-#ifdef CONFIG_EXT3_FS_POSIX_ACL
-	ei->i_acl = EXT3_ACL_NOT_CACHED;
-	ei->i_default_acl = EXT3_ACL_NOT_CACHED;
-#endif
 	ei->i_block_alloc_info = NULL;
 	ei->vfs_inode.i_version = 1;
 	return &ei->vfs_inode;
@@ -518,18 +514,6 @@ static void destroy_inodecache(void)
 static void ext3_clear_inode(struct inode *inode)
 {
 	struct ext3_block_alloc_info *rsv = EXT3_I(inode)->i_block_alloc_info;
-#ifdef CONFIG_EXT3_FS_POSIX_ACL
-	if (EXT3_I(inode)->i_acl &&
-			EXT3_I(inode)->i_acl != EXT3_ACL_NOT_CACHED) {
-		posix_acl_release(EXT3_I(inode)->i_acl);
-		EXT3_I(inode)->i_acl = EXT3_ACL_NOT_CACHED;
-	}
-	if (EXT3_I(inode)->i_default_acl &&
-			EXT3_I(inode)->i_default_acl != EXT3_ACL_NOT_CACHED) {
-		posix_acl_release(EXT3_I(inode)->i_default_acl);
-		EXT3_I(inode)->i_default_acl = EXT3_ACL_NOT_CACHED;
-	}
-#endif
 	ext3_discard_reservation(inode);
 	EXT3_I(inode)->i_block_alloc_info = NULL;
 	if (unlikely(rsv))
diff --git a/include/linux/ext3_fs_i.h b/include/linux/ext3_fs_i.h
index 7894dd0f3b77..ca1bfe90004f 100644
--- a/include/linux/ext3_fs_i.h
+++ b/include/linux/ext3_fs_i.h
@@ -103,10 +103,6 @@ struct ext3_inode_info {
 	 */
 	struct rw_semaphore xattr_sem;
 #endif
-#ifdef CONFIG_EXT3_FS_POSIX_ACL
-	struct posix_acl	*i_acl;
-	struct posix_acl	*i_default_acl;
-#endif
 
 	struct list_head i_orphan;	/* unlinked but open inodes */
 
-- 
cgit v1.2.3-59-g8ed1b


From d4bfe2f76d785cc77611a4bda8cedaff358d8c7d Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 8 Jun 2009 19:54:26 -0400
Subject: switch ext4 to inode->i_acl

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/ext4/acl.c   | 22 ++++++++++------------
 fs/ext4/acl.h   |  4 ----
 fs/ext4/ext4.h  |  4 ----
 fs/ext4/inode.c |  4 ----
 fs/ext4/super.c | 16 ----------------
 5 files changed, 10 insertions(+), 40 deletions(-)

diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c
index 605aeed96d68..0084e3a19d86 100644
--- a/fs/ext4/acl.c
+++ b/fs/ext4/acl.c
@@ -134,7 +134,7 @@ ext4_iget_acl(struct inode *inode, struct posix_acl **i_acl)
 	if (acl) {
 		spin_lock(&inode->i_lock);
 		acl = *i_acl;
-		if (acl != EXT4_ACL_NOT_CACHED)
+		if (acl != ACL_NOT_CACHED)
 			acl = posix_acl_dup(acl);
 		spin_unlock(&inode->i_lock);
 	}
@@ -147,7 +147,7 @@ ext4_iset_acl(struct inode *inode, struct posix_acl **i_acl,
 		struct posix_acl *acl)
 {
 	spin_lock(&inode->i_lock);
-	if (*i_acl != EXT4_ACL_NOT_CACHED)
+	if (*i_acl != ACL_NOT_CACHED)
 		posix_acl_release(*i_acl);
 	*i_acl = posix_acl_dup(acl);
 	spin_unlock(&inode->i_lock);
@@ -161,7 +161,6 @@ ext4_iset_acl(struct inode *inode, struct posix_acl **i_acl,
 static struct posix_acl *
 ext4_get_acl(struct inode *inode, int type)
 {
-	struct ext4_inode_info *ei = EXT4_I(inode);
 	int name_index;
 	char *value = NULL;
 	struct posix_acl *acl;
@@ -172,15 +171,15 @@ ext4_get_acl(struct inode *inode, int type)
 
 	switch (type) {
 	case ACL_TYPE_ACCESS:
-		acl = ext4_iget_acl(inode, &ei->i_acl);
-		if (acl != EXT4_ACL_NOT_CACHED)
+		acl = ext4_iget_acl(inode, &inode->i_acl);
+		if (acl != ACL_NOT_CACHED)
 			return acl;
 		name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS;
 		break;
 
 	case ACL_TYPE_DEFAULT:
-		acl = ext4_iget_acl(inode, &ei->i_default_acl);
-		if (acl != EXT4_ACL_NOT_CACHED)
+		acl = ext4_iget_acl(inode, &inode->i_default_acl);
+		if (acl != ACL_NOT_CACHED)
 			return acl;
 		name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT;
 		break;
@@ -206,11 +205,11 @@ ext4_get_acl(struct inode *inode, int type)
 	if (!IS_ERR(acl)) {
 		switch (type) {
 		case ACL_TYPE_ACCESS:
-			ext4_iset_acl(inode, &ei->i_acl, acl);
+			ext4_iset_acl(inode, &inode->i_acl, acl);
 			break;
 
 		case ACL_TYPE_DEFAULT:
-			ext4_iset_acl(inode, &ei->i_default_acl, acl);
+			ext4_iset_acl(inode, &inode->i_default_acl, acl);
 			break;
 		}
 	}
@@ -226,7 +225,6 @@ static int
 ext4_set_acl(handle_t *handle, struct inode *inode, int type,
 	     struct posix_acl *acl)
 {
-	struct ext4_inode_info *ei = EXT4_I(inode);
 	int name_index;
 	void *value = NULL;
 	size_t size = 0;
@@ -274,11 +272,11 @@ ext4_set_acl(handle_t *handle, struct inode *inode, int type,
 	if (!error) {
 		switch (type) {
 		case ACL_TYPE_ACCESS:
-			ext4_iset_acl(inode, &ei->i_acl, acl);
+			ext4_iset_acl(inode, &inode->i_acl, acl);
 			break;
 
 		case ACL_TYPE_DEFAULT:
-			ext4_iset_acl(inode, &ei->i_default_acl, acl);
+			ext4_iset_acl(inode, &inode->i_default_acl, acl);
 			break;
 		}
 	}
diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h
index cb45257a246e..949789d2bba6 100644
--- a/fs/ext4/acl.h
+++ b/fs/ext4/acl.h
@@ -53,10 +53,6 @@ static inline int ext4_acl_count(size_t size)
 
 #ifdef CONFIG_EXT4_FS_POSIX_ACL
 
-/* Value for inode->u.ext4_i.i_acl and inode->u.ext4_i.i_default_acl
-   if the ACL has not been cached */
-#define EXT4_ACL_NOT_CACHED ((void *)-1)
-
 /* acl.c */
 extern int ext4_permission(struct inode *, int);
 extern int ext4_acl_chmod(struct inode *);
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 17b9998680e3..0ddf7e55abe1 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -595,10 +595,6 @@ struct ext4_inode_info {
 	 */
 	struct rw_semaphore xattr_sem;
 #endif
-#ifdef CONFIG_EXT4_FS_POSIX_ACL
-	struct posix_acl	*i_acl;
-	struct posix_acl	*i_default_acl;
-#endif
 
 	struct list_head i_orphan;	/* unlinked but open inodes */
 
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 7c17ae275af4..60a26f3a6f8b 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -4453,10 +4453,6 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 		return inode;
 
 	ei = EXT4_I(inode);
-#ifdef CONFIG_EXT4_FS_POSIX_ACL
-	ei->i_acl = EXT4_ACL_NOT_CACHED;
-	ei->i_default_acl = EXT4_ACL_NOT_CACHED;
-#endif
 
 	ret = __ext4_get_inode_loc(inode, &iloc, 0);
 	if (ret < 0)
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 8bb9e2d3e4b8..8f4f079e6b9a 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -666,10 +666,6 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
 	if (!ei)
 		return NULL;
 
-#ifdef CONFIG_EXT4_FS_POSIX_ACL
-	ei->i_acl = EXT4_ACL_NOT_CACHED;
-	ei->i_default_acl = EXT4_ACL_NOT_CACHED;
-#endif
 	ei->vfs_inode.i_version = 1;
 	ei->vfs_inode.i_data.writeback_index = 0;
 	memset(&ei->i_cached_extent, 0, sizeof(struct ext4_ext_cache));
@@ -735,18 +731,6 @@ static void destroy_inodecache(void)
 
 static void ext4_clear_inode(struct inode *inode)
 {
-#ifdef CONFIG_EXT4_FS_POSIX_ACL
-	if (EXT4_I(inode)->i_acl &&
-			EXT4_I(inode)->i_acl != EXT4_ACL_NOT_CACHED) {
-		posix_acl_release(EXT4_I(inode)->i_acl);
-		EXT4_I(inode)->i_acl = EXT4_ACL_NOT_CACHED;
-	}
-	if (EXT4_I(inode)->i_default_acl &&
-			EXT4_I(inode)->i_default_acl != EXT4_ACL_NOT_CACHED) {
-		posix_acl_release(EXT4_I(inode)->i_default_acl);
-		EXT4_I(inode)->i_default_acl = EXT4_ACL_NOT_CACHED;
-	}
-#endif
 	ext4_discard_preallocations(inode);
 	if (EXT4_JOURNAL(inode))
 		jbd2_journal_release_jbd_inode(EXT4_SB(inode->i_sb)->s_journal,
-- 
cgit v1.2.3-59-g8ed1b


From 05fc0790b6c9c611129f2f712d00b6a8a364e8d2 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 8 Jun 2009 19:54:52 -0400
Subject: switch jfs to inode->i_acl

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/jfs/acl.c        | 22 +++++++++-------------
 fs/jfs/jfs_incore.h |  6 ------
 fs/jfs/super.c      | 16 ----------------
 fs/jfs/xattr.c      | 12 ++++++------
 4 files changed, 15 insertions(+), 41 deletions(-)

diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c
index 06ca1b8d2054..5fcfc9857c11 100644
--- a/fs/jfs/acl.c
+++ b/fs/jfs/acl.c
@@ -31,7 +31,6 @@ static struct posix_acl *jfs_get_acl(struct inode *inode, int type)
 {
 	struct posix_acl *acl;
 	char *ea_name;
-	struct jfs_inode_info *ji = JFS_IP(inode);
 	struct posix_acl **p_acl;
 	int size;
 	char *value = NULL;
@@ -39,17 +38,17 @@ static struct posix_acl *jfs_get_acl(struct inode *inode, int type)
 	switch(type) {
 		case ACL_TYPE_ACCESS:
 			ea_name = POSIX_ACL_XATTR_ACCESS;
-			p_acl = &ji->i_acl;
+			p_acl = &inode->i_acl;
 			break;
 		case ACL_TYPE_DEFAULT:
 			ea_name = POSIX_ACL_XATTR_DEFAULT;
-			p_acl = &ji->i_default_acl;
+			p_acl = &inode->i_default_acl;
 			break;
 		default:
 			return ERR_PTR(-EINVAL);
 	}
 
-	if (*p_acl != JFS_ACL_NOT_CACHED)
+	if (*p_acl != ACL_NOT_CACHED)
 		return posix_acl_dup(*p_acl);
 
 	size = __jfs_getxattr(inode, ea_name, NULL, 0);
@@ -80,7 +79,6 @@ static int jfs_set_acl(tid_t tid, struct inode *inode, int type,
 		       struct posix_acl *acl)
 {
 	char *ea_name;
-	struct jfs_inode_info *ji = JFS_IP(inode);
 	struct posix_acl **p_acl;
 	int rc;
 	int size = 0;
@@ -92,11 +90,11 @@ static int jfs_set_acl(tid_t tid, struct inode *inode, int type,
 	switch(type) {
 		case ACL_TYPE_ACCESS:
 			ea_name = POSIX_ACL_XATTR_ACCESS;
-			p_acl = &ji->i_acl;
+			p_acl = &inode->i_acl;
 			break;
 		case ACL_TYPE_DEFAULT:
 			ea_name = POSIX_ACL_XATTR_DEFAULT;
-			p_acl = &ji->i_default_acl;
+			p_acl = &inode->i_default_acl;
 			if (!S_ISDIR(inode->i_mode))
 				return acl ? -EACCES : 0;
 			break;
@@ -117,7 +115,7 @@ out:
 	kfree(value);
 
 	if (!rc) {
-		if (*p_acl && (*p_acl != JFS_ACL_NOT_CACHED))
+		if (*p_acl && (*p_acl != ACL_NOT_CACHED))
 			posix_acl_release(*p_acl);
 		*p_acl = posix_acl_dup(acl);
 	}
@@ -126,17 +124,15 @@ out:
 
 static int jfs_check_acl(struct inode *inode, int mask)
 {
-	struct jfs_inode_info *ji = JFS_IP(inode);
-
-	if (ji->i_acl == JFS_ACL_NOT_CACHED) {
+	if (inode->i_acl == ACL_NOT_CACHED) {
 		struct posix_acl *acl = jfs_get_acl(inode, ACL_TYPE_ACCESS);
 		if (IS_ERR(acl))
 			return PTR_ERR(acl);
 		posix_acl_release(acl);
 	}
 
-	if (ji->i_acl)
-		return posix_acl_permission(inode, ji->i_acl, mask);
+	if (inode->i_acl)
+		return posix_acl_permission(inode, inode->i_acl, mask);
 	return -EAGAIN;
 }
 
diff --git a/fs/jfs/jfs_incore.h b/fs/jfs/jfs_incore.h
index 439901d205fe..1439f119ec83 100644
--- a/fs/jfs/jfs_incore.h
+++ b/fs/jfs/jfs_incore.h
@@ -74,10 +74,6 @@ struct jfs_inode_info {
 	/* xattr_sem allows us to access the xattrs without taking i_mutex */
 	struct rw_semaphore xattr_sem;
 	lid_t	xtlid;		/* lid of xtree lock on directory */
-#ifdef CONFIG_JFS_POSIX_ACL
-	struct posix_acl *i_acl;
-	struct posix_acl *i_default_acl;
-#endif
 	union {
 		struct {
 			xtpage_t _xtroot;	/* 288: xtree root */
@@ -107,8 +103,6 @@ struct jfs_inode_info {
 #define i_inline u.link._inline
 #define i_inline_ea u.link._inline_ea
 
-#define JFS_ACL_NOT_CACHED ((void *)-1)
-
 #define IREAD_LOCK(ip, subclass) \
 	down_read_nested(&JFS_IP(ip)->rdwrlock, subclass)
 #define IREAD_UNLOCK(ip)	up_read(&JFS_IP(ip)->rdwrlock)
diff --git a/fs/jfs/super.c b/fs/jfs/super.c
index 09b1b6ee2186..37e6dcda8fc8 100644
--- a/fs/jfs/super.c
+++ b/fs/jfs/super.c
@@ -128,18 +128,6 @@ static void jfs_destroy_inode(struct inode *inode)
 		ji->active_ag = -1;
 	}
 	spin_unlock_irq(&ji->ag_lock);
-
-#ifdef CONFIG_JFS_POSIX_ACL
-	if (ji->i_acl != JFS_ACL_NOT_CACHED) {
-		posix_acl_release(ji->i_acl);
-		ji->i_acl = JFS_ACL_NOT_CACHED;
-	}
-	if (ji->i_default_acl != JFS_ACL_NOT_CACHED) {
-		posix_acl_release(ji->i_default_acl);
-		ji->i_default_acl = JFS_ACL_NOT_CACHED;
-	}
-#endif
-
 	kmem_cache_free(jfs_inode_cachep, ji);
 }
 
@@ -798,10 +786,6 @@ static void init_once(void *foo)
 	init_rwsem(&jfs_ip->xattr_sem);
 	spin_lock_init(&jfs_ip->ag_lock);
 	jfs_ip->active_ag = -1;
-#ifdef CONFIG_JFS_POSIX_ACL
-	jfs_ip->i_acl = JFS_ACL_NOT_CACHED;
-	jfs_ip->i_default_acl = JFS_ACL_NOT_CACHED;
-#endif
 	inode_init_once(&jfs_ip->vfs_inode);
 }
 
diff --git a/fs/jfs/xattr.c b/fs/jfs/xattr.c
index 61dfa8173ebc..f6e90e343593 100644
--- a/fs/jfs/xattr.c
+++ b/fs/jfs/xattr.c
@@ -727,10 +727,10 @@ static int can_set_system_xattr(struct inode *inode, const char *name,
 		/*
 		 * We're changing the ACL.  Get rid of the cached one
 		 */
-		acl =JFS_IP(inode)->i_acl;
-		if (acl != JFS_ACL_NOT_CACHED)
+		acl =inode->i_acl;
+		if (acl != ACL_NOT_CACHED)
 			posix_acl_release(acl);
-		JFS_IP(inode)->i_acl = JFS_ACL_NOT_CACHED;
+		inode->i_acl = ACL_NOT_CACHED;
 
 		return 0;
 	} else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0) {
@@ -746,10 +746,10 @@ static int can_set_system_xattr(struct inode *inode, const char *name,
 		/*
 		 * We're changing the default ACL.  Get rid of the cached one
 		 */
-		acl =JFS_IP(inode)->i_default_acl;
-		if (acl && (acl != JFS_ACL_NOT_CACHED))
+		acl = inode->i_default_acl;
+		if (acl && (acl != ACL_NOT_CACHED))
 			posix_acl_release(acl);
-		JFS_IP(inode)->i_default_acl = JFS_ACL_NOT_CACHED;
+		inode->i_default_acl = ACL_NOT_CACHED;
 
 		return 0;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 290c263bf83cd78e53b1aa3b42165f588163f2be Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 8 Jun 2009 19:55:12 -0400
Subject: switch jffs2 to inode->i_acl

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/jffs2/acl.c        | 54 ++++++++++++++++++---------------------------------
 fs/jffs2/acl.h        |  4 ----
 fs/jffs2/jffs2_fs_i.h |  4 ----
 fs/jffs2/os-linux.h   |  4 ----
 fs/jffs2/readinode.c  |  1 -
 5 files changed, 19 insertions(+), 48 deletions(-)

diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c
index 043740dde20c..ac16589ebbd1 100644
--- a/fs/jffs2/acl.c
+++ b/fs/jffs2/acl.c
@@ -158,10 +158,10 @@ static void *jffs2_acl_to_medium(const struct posix_acl *acl, size_t *size)
 
 static struct posix_acl *jffs2_iget_acl(struct inode *inode, struct posix_acl **i_acl)
 {
-	struct posix_acl *acl = JFFS2_ACL_NOT_CACHED;
+	struct posix_acl *acl = ACL_NOT_CACHED;
 
 	spin_lock(&inode->i_lock);
-	if (*i_acl != JFFS2_ACL_NOT_CACHED)
+	if (*i_acl != ACL_NOT_CACHED)
 		acl = posix_acl_dup(*i_acl);
 	spin_unlock(&inode->i_lock);
 	return acl;
@@ -170,7 +170,7 @@ static struct posix_acl *jffs2_iget_acl(struct inode *inode, struct posix_acl **
 static void jffs2_iset_acl(struct inode *inode, struct posix_acl **i_acl, struct posix_acl *acl)
 {
 	spin_lock(&inode->i_lock);
-	if (*i_acl != JFFS2_ACL_NOT_CACHED)
+	if (*i_acl != ACL_NOT_CACHED)
 		posix_acl_release(*i_acl);
 	*i_acl = posix_acl_dup(acl);
 	spin_unlock(&inode->i_lock);
@@ -178,21 +178,20 @@ static void jffs2_iset_acl(struct inode *inode, struct posix_acl **i_acl, struct
 
 static struct posix_acl *jffs2_get_acl(struct inode *inode, int type)
 {
-	struct jffs2_inode_info *f = JFFS2_INODE_INFO(inode);
 	struct posix_acl *acl;
 	char *value = NULL;
 	int rc, xprefix;
 
 	switch (type) {
 	case ACL_TYPE_ACCESS:
-		acl = jffs2_iget_acl(inode, &f->i_acl_access);
-		if (acl != JFFS2_ACL_NOT_CACHED)
+		acl = jffs2_iget_acl(inode, &inode->i_acl);
+		if (acl != ACL_NOT_CACHED)
 			return acl;
 		xprefix = JFFS2_XPREFIX_ACL_ACCESS;
 		break;
 	case ACL_TYPE_DEFAULT:
-		acl = jffs2_iget_acl(inode, &f->i_acl_default);
-		if (acl != JFFS2_ACL_NOT_CACHED)
+		acl = jffs2_iget_acl(inode, &inode->i_default_acl);
+		if (acl != ACL_NOT_CACHED)
 			return acl;
 		xprefix = JFFS2_XPREFIX_ACL_DEFAULT;
 		break;
@@ -218,10 +217,10 @@ static struct posix_acl *jffs2_get_acl(struct inode *inode, int type)
 	if (!IS_ERR(acl)) {
 		switch (type) {
 		case ACL_TYPE_ACCESS:
-			jffs2_iset_acl(inode, &f->i_acl_access, acl);
+			jffs2_iset_acl(inode, &inode->i_acl, acl);
 			break;
 		case ACL_TYPE_DEFAULT:
-			jffs2_iset_acl(inode, &f->i_acl_default, acl);
+			jffs2_iset_acl(inode, &inode->i_default_acl, acl);
 			break;
 		}
 	}
@@ -249,7 +248,6 @@ static int __jffs2_set_acl(struct inode *inode, int xprefix, struct posix_acl *a
 
 static int jffs2_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 {
-	struct jffs2_inode_info *f = JFFS2_INODE_INFO(inode);
 	int rc, xprefix;
 
 	if (S_ISLNK(inode->i_mode))
@@ -288,10 +286,10 @@ static int jffs2_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 	if (!rc) {
 		switch(type) {
 		case ACL_TYPE_ACCESS:
-			jffs2_iset_acl(inode, &f->i_acl_access, acl);
+			jffs2_iset_acl(inode, &inode->i_acl, acl);
 			break;
 		case ACL_TYPE_DEFAULT:
-			jffs2_iset_acl(inode, &f->i_acl_default, acl);
+			jffs2_iset_acl(inode, &inode->i_default_acl, acl);
 			break;
 		}
 	}
@@ -321,12 +319,11 @@ int jffs2_permission(struct inode *inode, int mask)
 
 int jffs2_init_acl_pre(struct inode *dir_i, struct inode *inode, int *i_mode)
 {
-	struct jffs2_inode_info *f = JFFS2_INODE_INFO(inode);
 	struct posix_acl *acl, *clone;
 	int rc;
 
-	f->i_acl_default = NULL;
-	f->i_acl_access = NULL;
+	inode->i_default_acl = NULL;
+	inode->i_acl = NULL;
 
 	if (S_ISLNK(*i_mode))
 		return 0;	/* Symlink always has no-ACL */
@@ -339,7 +336,7 @@ int jffs2_init_acl_pre(struct inode *dir_i, struct inode *inode, int *i_mode)
 		*i_mode &= ~current_umask();
 	} else {
 		if (S_ISDIR(*i_mode))
-			jffs2_iset_acl(inode, &f->i_acl_default, acl);
+			jffs2_iset_acl(inode, &inode->i_default_acl, acl);
 
 		clone = posix_acl_clone(acl, GFP_KERNEL);
 		if (!clone)
@@ -350,7 +347,7 @@ int jffs2_init_acl_pre(struct inode *dir_i, struct inode *inode, int *i_mode)
 			return rc;
 		}
 		if (rc > 0)
-			jffs2_iset_acl(inode, &f->i_acl_access, clone);
+			jffs2_iset_acl(inode, &inode->i_acl, clone);
 
 		posix_acl_release(clone);
 	}
@@ -359,17 +356,16 @@ int jffs2_init_acl_pre(struct inode *dir_i, struct inode *inode, int *i_mode)
 
 int jffs2_init_acl_post(struct inode *inode)
 {
-	struct jffs2_inode_info *f = JFFS2_INODE_INFO(inode);
 	int rc;
 
-	if (f->i_acl_default) {
-		rc = __jffs2_set_acl(inode, JFFS2_XPREFIX_ACL_DEFAULT, f->i_acl_default);
+	if (inode->i_default_acl) {
+		rc = __jffs2_set_acl(inode, JFFS2_XPREFIX_ACL_DEFAULT, inode->i_default_acl);
 		if (rc)
 			return rc;
 	}
 
-	if (f->i_acl_access) {
-		rc = __jffs2_set_acl(inode, JFFS2_XPREFIX_ACL_ACCESS, f->i_acl_access);
+	if (inode->i_acl) {
+		rc = __jffs2_set_acl(inode, JFFS2_XPREFIX_ACL_ACCESS, inode->i_acl);
 		if (rc)
 			return rc;
 	}
@@ -377,18 +373,6 @@ int jffs2_init_acl_post(struct inode *inode)
 	return 0;
 }
 
-void jffs2_clear_acl(struct jffs2_inode_info *f)
-{
-	if (f->i_acl_access && f->i_acl_access != JFFS2_ACL_NOT_CACHED) {
-		posix_acl_release(f->i_acl_access);
-		f->i_acl_access = JFFS2_ACL_NOT_CACHED;
-	}
-	if (f->i_acl_default && f->i_acl_default != JFFS2_ACL_NOT_CACHED) {
-		posix_acl_release(f->i_acl_default);
-		f->i_acl_default = JFFS2_ACL_NOT_CACHED;
-	}
-}
-
 int jffs2_acl_chmod(struct inode *inode)
 {
 	struct posix_acl *acl, *clone;
diff --git a/fs/jffs2/acl.h b/fs/jffs2/acl.h
index 8ca058aed384..fc929f2a14f6 100644
--- a/fs/jffs2/acl.h
+++ b/fs/jffs2/acl.h
@@ -26,13 +26,10 @@ struct jffs2_acl_header {
 
 #ifdef CONFIG_JFFS2_FS_POSIX_ACL
 
-#define JFFS2_ACL_NOT_CACHED ((void *)-1)
-
 extern int jffs2_permission(struct inode *, int);
 extern int jffs2_acl_chmod(struct inode *);
 extern int jffs2_init_acl_pre(struct inode *, struct inode *, int *);
 extern int jffs2_init_acl_post(struct inode *);
-extern void jffs2_clear_acl(struct jffs2_inode_info *);
 
 extern struct xattr_handler jffs2_acl_access_xattr_handler;
 extern struct xattr_handler jffs2_acl_default_xattr_handler;
@@ -43,6 +40,5 @@ extern struct xattr_handler jffs2_acl_default_xattr_handler;
 #define jffs2_acl_chmod(inode)			(0)
 #define jffs2_init_acl_pre(dir_i,inode,mode)	(0)
 #define jffs2_init_acl_post(inode)		(0)
-#define jffs2_clear_acl(f)
 
 #endif	/* CONFIG_JFFS2_FS_POSIX_ACL */
diff --git a/fs/jffs2/jffs2_fs_i.h b/fs/jffs2/jffs2_fs_i.h
index 4c41db91eaa4..c6923da98263 100644
--- a/fs/jffs2/jffs2_fs_i.h
+++ b/fs/jffs2/jffs2_fs_i.h
@@ -50,10 +50,6 @@ struct jffs2_inode_info {
 	uint16_t flags;
 	uint8_t usercompr;
 	struct inode vfs_inode;
-#ifdef CONFIG_JFFS2_FS_POSIX_ACL
-	struct posix_acl *i_acl_access;
-	struct posix_acl *i_acl_default;
-#endif
 };
 
 #endif /* _JFFS2_FS_I */
diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h
index 2228380c47b9..a7f03b7ebcb3 100644
--- a/fs/jffs2/os-linux.h
+++ b/fs/jffs2/os-linux.h
@@ -56,10 +56,6 @@ static inline void jffs2_init_inode_info(struct jffs2_inode_info *f)
 	f->target = NULL;
 	f->flags = 0;
 	f->usercompr = 0;
-#ifdef CONFIG_JFFS2_FS_POSIX_ACL
-	f->i_acl_access = JFFS2_ACL_NOT_CACHED;
-	f->i_acl_default = JFFS2_ACL_NOT_CACHED;
-#endif
 }
 
 
diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c
index 1fc1e92356ee..1a80301004b8 100644
--- a/fs/jffs2/readinode.c
+++ b/fs/jffs2/readinode.c
@@ -1424,7 +1424,6 @@ void jffs2_do_clear_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f)
 	struct jffs2_full_dirent *fd, *fds;
 	int deleted;
 
-	jffs2_clear_acl(f);
 	jffs2_xattr_delete_inode(c, f->inocache);
 	mutex_lock(&f->sem);
 	deleted = f->inocache && !f->inocache->pino_nlink;
-- 
cgit v1.2.3-59-g8ed1b


From 5affd88a104af43f0063a12ad1ee4c7a587945dc Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 8 Jun 2009 19:55:32 -0400
Subject: switch btrfs to inode->i_acl

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/btrfs/acl.c         | 14 +++++++-------
 fs/btrfs/btrfs_inode.h |  4 ----
 fs/btrfs/ctree.h       |  2 --
 fs/btrfs/inode.c       | 16 ++--------------
 4 files changed, 9 insertions(+), 27 deletions(-)

diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index 603972576f0f..6db8a42a3e5e 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -34,7 +34,7 @@ static void btrfs_update_cached_acl(struct inode *inode,
 				    struct posix_acl *acl)
 {
 	spin_lock(&inode->i_lock);
-	if (*p_acl && *p_acl != BTRFS_ACL_NOT_CACHED)
+	if (*p_acl && *p_acl != ACL_NOT_CACHED)
 		posix_acl_release(*p_acl);
 	*p_acl = posix_acl_dup(acl);
 	spin_unlock(&inode->i_lock);
@@ -50,11 +50,11 @@ static struct posix_acl *btrfs_get_acl(struct inode *inode, int type)
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		name = POSIX_ACL_XATTR_ACCESS;
-		p_acl = &BTRFS_I(inode)->i_acl;
+		p_acl = &inode->i_acl;
 		break;
 	case ACL_TYPE_DEFAULT:
 		name = POSIX_ACL_XATTR_DEFAULT;
-		p_acl = &BTRFS_I(inode)->i_default_acl;
+		p_acl = &inode->i_default_acl;
 		break;
 	default:
 		return ERR_PTR(-EINVAL);
@@ -67,11 +67,11 @@ static struct posix_acl *btrfs_get_acl(struct inode *inode, int type)
 
 	spin_lock(&inode->i_lock);
 	acl = *p_acl;
-	if (acl != BTRFS_ACL_NOT_CACHED)
+	if (acl != ACL_NOT_CACHED)
 		acl = posix_acl_dup(acl);
 	spin_unlock(&inode->i_lock);
 
-	if (acl != BTRFS_ACL_NOT_CACHED)
+	if (acl != ACL_NOT_CACHED)
 		return acl;
 
 	size = __btrfs_getxattr(inode, name, "", 0);
@@ -141,13 +141,13 @@ static int btrfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 		ret = 0;
 		inode->i_mode = mode;
 		name = POSIX_ACL_XATTR_ACCESS;
-		p_acl = &BTRFS_I(inode)->i_acl;
+		p_acl = &inode->i_acl;
 		break;
 	case ACL_TYPE_DEFAULT:
 		if (!S_ISDIR(inode->i_mode))
 			return acl ? -EINVAL : 0;
 		name = POSIX_ACL_XATTR_DEFAULT;
-		p_acl = &BTRFS_I(inode)->i_default_acl;
+		p_acl = &inode->i_default_acl;
 		break;
 	default:
 		return -EINVAL;
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index acb4f3517582..ea1ea0af8c0e 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -53,10 +53,6 @@ struct btrfs_inode {
 	/* used to order data wrt metadata */
 	struct btrfs_ordered_inode_tree ordered_tree;
 
-	/* standard acl pointers */
-	struct posix_acl *i_acl;
-	struct posix_acl *i_default_acl;
-
 	/* for keeping track of orphaned inodes */
 	struct list_head i_orphan;
 
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 03441a99ea38..2779c2f5360a 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -41,8 +41,6 @@ struct btrfs_ordered_sum;
 
 #define BTRFS_MAGIC "_BHRfS_M"
 
-#define BTRFS_ACL_NOT_CACHED    ((void *)-1)
-
 #define BTRFS_MAX_LEVEL 8
 
 #define BTRFS_COMPAT_EXTENT_TREE_V0
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 8612b3a09811..78ad38ddd01f 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -2123,8 +2123,8 @@ static void btrfs_read_locked_inode(struct inode *inode)
 	 */
 	maybe_acls = acls_after_inode_item(leaf, path->slots[0], inode->i_ino);
 	if (!maybe_acls) {
-		BTRFS_I(inode)->i_acl = NULL;
-		BTRFS_I(inode)->i_default_acl = NULL;
+		inode->i_acl = NULL;
+		inode->i_default_acl = NULL;
 	}
 
 	BTRFS_I(inode)->block_group = btrfs_find_block_group(root, 0,
@@ -3141,9 +3141,6 @@ static noinline void init_btrfs_i(struct inode *inode)
 {
 	struct btrfs_inode *bi = BTRFS_I(inode);
 
-	bi->i_acl = BTRFS_ACL_NOT_CACHED;
-	bi->i_default_acl = BTRFS_ACL_NOT_CACHED;
-
 	bi->generation = 0;
 	bi->sequence = 0;
 	bi->last_trans = 0;
@@ -4640,8 +4637,6 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
 	ei->last_trans = 0;
 	ei->logged_trans = 0;
 	btrfs_ordered_inode_tree_init(&ei->ordered_tree);
-	ei->i_acl = BTRFS_ACL_NOT_CACHED;
-	ei->i_default_acl = BTRFS_ACL_NOT_CACHED;
 	INIT_LIST_HEAD(&ei->i_orphan);
 	INIT_LIST_HEAD(&ei->ordered_operations);
 	return &ei->vfs_inode;
@@ -4655,13 +4650,6 @@ void btrfs_destroy_inode(struct inode *inode)
 	WARN_ON(!list_empty(&inode->i_dentry));
 	WARN_ON(inode->i_data.nrpages);
 
-	if (BTRFS_I(inode)->i_acl &&
-	    BTRFS_I(inode)->i_acl != BTRFS_ACL_NOT_CACHED)
-		posix_acl_release(BTRFS_I(inode)->i_acl);
-	if (BTRFS_I(inode)->i_default_acl &&
-	    BTRFS_I(inode)->i_default_acl != BTRFS_ACL_NOT_CACHED)
-		posix_acl_release(BTRFS_I(inode)->i_default_acl);
-
 	/*
 	 * Make sure we're properly removed from the ordered operation
 	 * lists.
-- 
cgit v1.2.3-59-g8ed1b


From d441b1c293149212045de00f346c8ea6cd41cce4 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 8 Jun 2009 19:56:34 -0400
Subject: switch nilfs2 to inode->i_acl

Actually, get rid of private analog, since nothing in there is
using ACLs at all so far.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/nilfs2/inode.c |  8 --------
 fs/nilfs2/nilfs.h |  4 ----
 fs/nilfs2/super.c | 10 ----------
 3 files changed, 22 deletions(-)

diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index 2696d6b513b7..fe9d8f2a13f8 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -309,10 +309,6 @@ struct inode *nilfs_new_inode(struct inode *dir, int mode)
 	/* ii->i_file_acl = 0; */
 	/* ii->i_dir_acl = 0; */
 	ii->i_dir_start_lookup = 0;
-#ifdef CONFIG_NILFS_FS_POSIX_ACL
-	ii->i_acl = NULL;
-	ii->i_default_acl = NULL;
-#endif
 	ii->i_cno = 0;
 	nilfs_set_inode_flags(inode);
 	spin_lock(&sbi->s_next_gen_lock);
@@ -434,10 +430,6 @@ static int __nilfs_read_inode(struct super_block *sb, unsigned long ino,
 
 	raw_inode = nilfs_ifile_map_inode(sbi->s_ifile, ino, bh);
 
-#ifdef CONFIG_NILFS_FS_POSIX_ACL
-	ii->i_acl = NILFS_ACL_NOT_CACHED;
-	ii->i_default_acl = NILFS_ACL_NOT_CACHED;
-#endif
 	if (nilfs_read_inode_common(inode, raw_inode))
 		goto failed_unmap;
 
diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h
index edf6a59d9f2a..724c63766e82 100644
--- a/fs/nilfs2/nilfs.h
+++ b/fs/nilfs2/nilfs.h
@@ -57,10 +57,6 @@ struct nilfs_inode_info {
 	 * EAs.
 	 */
 	struct rw_semaphore xattr_sem;
-#endif
-#ifdef CONFIG_NILFS_POSIX_ACL
-	struct posix_acl *i_acl;
-	struct posix_acl *i_default_acl;
 #endif
 	struct buffer_head *i_bh;	/* i_bh contains a new or dirty
 					   disk inode */
diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c
index ab785f85aa50..8e2ec43b18f4 100644
--- a/fs/nilfs2/super.c
+++ b/fs/nilfs2/super.c
@@ -189,16 +189,6 @@ static void nilfs_clear_inode(struct inode *inode)
 {
 	struct nilfs_inode_info *ii = NILFS_I(inode);
 
-#ifdef CONFIG_NILFS_POSIX_ACL
-	if (ii->i_acl && ii->i_acl != NILFS_ACL_NOT_CACHED) {
-		posix_acl_release(ii->i_acl);
-		ii->i_acl = NILFS_ACL_NOT_CACHED;
-	}
-	if (ii->i_default_acl && ii->i_default_acl != NILFS_ACL_NOT_CACHED) {
-		posix_acl_release(ii->i_default_acl);
-		ii->i_default_acl = NILFS_ACL_NOT_CACHED;
-	}
-#endif
 	/*
 	 * Free resources allocated in nilfs_read_inode(), here.
 	 */
-- 
cgit v1.2.3-59-g8ed1b


From e68888bcb60ccba4dc21df9f2d8cd7325b64dce7 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 8 Jun 2009 20:46:54 -0400
Subject: reiserfs: minimal fix for ACL caching

reiserfs uses NULL as "unknown" and ERR_PTR(-ENODATA) as "no ACL";
several codepaths store the former instead of the latter.

All those codepaths go through iset_acl() and all cases when it's
called with NULL acl are for the second variety, so the minimal
fix is to teach iset_acl() to deal with that.

Proper fix is to switch to more usual conventions and avoid back
and forth between internally used ERR_PTR(-ENODATA) and NULL
expected by the rest of the kernel.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/reiserfs/xattr_acl.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c
index c303c426fe2b..a1a7e3530e17 100644
--- a/fs/reiserfs/xattr_acl.c
+++ b/fs/reiserfs/xattr_acl.c
@@ -194,7 +194,7 @@ static inline void iset_acl(struct inode *inode, struct posix_acl **i_acl,
 	spin_lock(&inode->i_lock);
 	if (*i_acl != ERR_PTR(-ENODATA))
 		posix_acl_release(*i_acl);
-	*i_acl = posix_acl_dup(acl);
+	*i_acl = acl ? posix_acl_dup(acl) : ERR_PTR(-ENODATA);
 	spin_unlock(&inode->i_lock);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 7a77b15d9294749809de918e24bebc39e0fbc9ab Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 8 Jun 2009 21:01:13 -0400
Subject: switch reiserfs to usual conventions for caching ACLs

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/reiserfs/super.c          | 12 ++++++------
 fs/reiserfs/xattr_acl.c      | 21 ++++++++-------------
 include/linux/reiserfs_acl.h |  4 ++--
 3 files changed, 16 insertions(+), 21 deletions(-)

diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 2969773cfc22..b194451fc04b 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -530,8 +530,8 @@ static void init_once(void *foo)
 	INIT_LIST_HEAD(&ei->i_prealloc_list);
 	inode_init_once(&ei->vfs_inode);
 #ifdef CONFIG_REISERFS_FS_POSIX_ACL
-	ei->i_acl_access = NULL;
-	ei->i_acl_default = NULL;
+	ei->i_acl_access = ACL_NOT_CACHED;
+	ei->i_acl_default = ACL_NOT_CACHED;
 #endif
 }
 
@@ -586,14 +586,14 @@ static void reiserfs_clear_inode(struct inode *inode)
 	struct posix_acl *acl;
 
 	acl = REISERFS_I(inode)->i_acl_access;
-	if (acl && !IS_ERR(acl))
+	if (acl && acl != ACL_NOT_CACHED)
 		posix_acl_release(acl);
-	REISERFS_I(inode)->i_acl_access = NULL;
+	REISERFS_I(inode)->i_acl_access = ACL_NOT_CACHED;
 
 	acl = REISERFS_I(inode)->i_acl_default;
-	if (acl && !IS_ERR(acl))
+	if (acl && acl != ACL_NOT_CACHED)
 		posix_acl_release(acl);
-	REISERFS_I(inode)->i_acl_default = NULL;
+	REISERFS_I(inode)->i_acl_default = ACL_NOT_CACHED;
 }
 #else
 #define reiserfs_clear_inode NULL
diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c
index a1a7e3530e17..7b3aeb9327d3 100644
--- a/fs/reiserfs/xattr_acl.c
+++ b/fs/reiserfs/xattr_acl.c
@@ -192,19 +192,19 @@ static inline void iset_acl(struct inode *inode, struct posix_acl **i_acl,
 			    struct posix_acl *acl)
 {
 	spin_lock(&inode->i_lock);
-	if (*i_acl != ERR_PTR(-ENODATA))
+	if (*i_acl != ACL_NOT_CACHED)
 		posix_acl_release(*i_acl);
-	*i_acl = acl ? posix_acl_dup(acl) : ERR_PTR(-ENODATA);
+	*i_acl = posix_acl_dup(acl);
 	spin_unlock(&inode->i_lock);
 }
 
 static inline struct posix_acl *iget_acl(struct inode *inode,
 					 struct posix_acl **i_acl)
 {
-	struct posix_acl *acl = ERR_PTR(-ENODATA);
+	struct posix_acl *acl = ACL_NOT_CACHED;
 
 	spin_lock(&inode->i_lock);
-	if (*i_acl != ERR_PTR(-ENODATA))
+	if (*i_acl != ACL_NOT_CACHED)
 		acl = posix_acl_dup(*i_acl);
 	spin_unlock(&inode->i_lock);
 
@@ -239,15 +239,13 @@ struct posix_acl *reiserfs_get_acl(struct inode *inode, int type)
 	}
 
 	acl = iget_acl(inode, p_acl);
-	if (acl && !IS_ERR(acl))
+	if (acl != ACL_NOT_CACHED)
 		return acl;
-	else if (PTR_ERR(acl) == -ENODATA)
-		return NULL;
 
 	size = reiserfs_xattr_get(inode, name, NULL, 0);
 	if (size < 0) {
 		if (size == -ENODATA || size == -ENOSYS) {
-			*p_acl = ERR_PTR(-ENODATA);
+			*p_acl = NULL;
 			return NULL;
 		}
 		return ERR_PTR(size);
@@ -262,7 +260,7 @@ struct posix_acl *reiserfs_get_acl(struct inode *inode, int type)
 		/* This shouldn't actually happen as it should have
 		   been caught above.. but just in case */
 		acl = NULL;
-		*p_acl = ERR_PTR(-ENODATA);
+		*p_acl = acl;
 	} else if (retval < 0) {
 		acl = ERR_PTR(retval);
 	} else {
@@ -379,11 +377,8 @@ reiserfs_inherit_default_acl(struct reiserfs_transaction_handle *th,
 	}
 
 	acl = reiserfs_get_acl(dir, ACL_TYPE_DEFAULT);
-	if (IS_ERR(acl)) {
-		if (PTR_ERR(acl) == -ENODATA)
-			goto apply_umask;
+	if (IS_ERR(acl))
 		return PTR_ERR(acl);
-	}
 
 	if (acl) {
 		struct posix_acl *acl_copy;
diff --git a/include/linux/reiserfs_acl.h b/include/linux/reiserfs_acl.h
index 8cc65757e47a..8f4d8d718b10 100644
--- a/include/linux/reiserfs_acl.h
+++ b/include/linux/reiserfs_acl.h
@@ -58,12 +58,12 @@ extern struct xattr_handler reiserfs_posix_acl_access_handler;
 
 static inline void reiserfs_init_acl_access(struct inode *inode)
 {
-	REISERFS_I(inode)->i_acl_access = NULL;
+	REISERFS_I(inode)->i_acl_access = ACL_NOT_CACHED;
 }
 
 static inline void reiserfs_init_acl_default(struct inode *inode)
 {
-	REISERFS_I(inode)->i_acl_default = NULL;
+	REISERFS_I(inode)->i_acl_default = ACL_NOT_CACHED;
 }
 #else
 
-- 
cgit v1.2.3-59-g8ed1b


From 281eede0328c84a8f20e0e85b807d5b51c3de4f2 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 8 Jun 2009 21:07:04 -0400
Subject: switch reiserfs to inode->i_acl

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/reiserfs/inode.c           |  4 ----
 fs/reiserfs/super.c           | 24 ------------------------
 fs/reiserfs/xattr_acl.c       | 10 ++++------
 include/linux/reiserfs_acl.h  | 17 -----------------
 include/linux/reiserfs_fs_i.h |  4 ----
 5 files changed, 4 insertions(+), 55 deletions(-)

diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 6fd0f47e45db..a14d6cd9eeda 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -1131,8 +1131,6 @@ static void init_inode(struct inode *inode, struct treepath *path)
 	REISERFS_I(inode)->i_trans_id = 0;
 	REISERFS_I(inode)->i_jl = NULL;
 	mutex_init(&(REISERFS_I(inode)->i_mmap));
-	reiserfs_init_acl_access(inode);
-	reiserfs_init_acl_default(inode);
 	reiserfs_init_xattr_rwsem(inode);
 
 	if (stat_data_v1(ih)) {
@@ -1834,8 +1832,6 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 	    REISERFS_I(dir)->i_attrs & REISERFS_INHERIT_MASK;
 	sd_attrs_to_i_attrs(REISERFS_I(inode)->i_attrs, inode);
 	mutex_init(&(REISERFS_I(inode)->i_mmap));
-	reiserfs_init_acl_access(inode);
-	reiserfs_init_acl_default(inode);
 	reiserfs_init_xattr_rwsem(inode);
 
 	/* key to search for correct place for new stat data */
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index b194451fc04b..d3aeb061612b 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -529,10 +529,6 @@ static void init_once(void *foo)
 
 	INIT_LIST_HEAD(&ei->i_prealloc_list);
 	inode_init_once(&ei->vfs_inode);
-#ifdef CONFIG_REISERFS_FS_POSIX_ACL
-	ei->i_acl_access = ACL_NOT_CACHED;
-	ei->i_acl_default = ACL_NOT_CACHED;
-#endif
 }
 
 static int init_inodecache(void)
@@ -580,25 +576,6 @@ static void reiserfs_dirty_inode(struct inode *inode)
 	reiserfs_write_unlock(inode->i_sb);
 }
 
-#ifdef CONFIG_REISERFS_FS_POSIX_ACL
-static void reiserfs_clear_inode(struct inode *inode)
-{
-	struct posix_acl *acl;
-
-	acl = REISERFS_I(inode)->i_acl_access;
-	if (acl && acl != ACL_NOT_CACHED)
-		posix_acl_release(acl);
-	REISERFS_I(inode)->i_acl_access = ACL_NOT_CACHED;
-
-	acl = REISERFS_I(inode)->i_acl_default;
-	if (acl && acl != ACL_NOT_CACHED)
-		posix_acl_release(acl);
-	REISERFS_I(inode)->i_acl_default = ACL_NOT_CACHED;
-}
-#else
-#define reiserfs_clear_inode NULL
-#endif
-
 #ifdef CONFIG_QUOTA
 static ssize_t reiserfs_quota_write(struct super_block *, int, const char *,
 				    size_t, loff_t);
@@ -612,7 +589,6 @@ static const struct super_operations reiserfs_sops = {
 	.write_inode = reiserfs_write_inode,
 	.dirty_inode = reiserfs_dirty_inode,
 	.delete_inode = reiserfs_delete_inode,
-	.clear_inode = reiserfs_clear_inode,
 	.put_super = reiserfs_put_super,
 	.write_super = reiserfs_write_super,
 	.sync_fs = reiserfs_sync_fs,
diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c
index 7b3aeb9327d3..b6e473faa8b8 100644
--- a/fs/reiserfs/xattr_acl.c
+++ b/fs/reiserfs/xattr_acl.c
@@ -223,16 +223,15 @@ struct posix_acl *reiserfs_get_acl(struct inode *inode, int type)
 	struct posix_acl *acl, **p_acl;
 	int size;
 	int retval;
-	struct reiserfs_inode_info *reiserfs_i = REISERFS_I(inode);
 
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		name = POSIX_ACL_XATTR_ACCESS;
-		p_acl = &reiserfs_i->i_acl_access;
+		p_acl = &inode->i_acl;
 		break;
 	case ACL_TYPE_DEFAULT:
 		name = POSIX_ACL_XATTR_DEFAULT;
-		p_acl = &reiserfs_i->i_acl_default;
+		p_acl = &inode->i_default_acl;
 		break;
 	default:
 		return ERR_PTR(-EINVAL);
@@ -288,7 +287,6 @@ reiserfs_set_acl(struct reiserfs_transaction_handle *th, struct inode *inode,
 	struct posix_acl **p_acl;
 	size_t size = 0;
 	int error;
-	struct reiserfs_inode_info *reiserfs_i = REISERFS_I(inode);
 
 	if (S_ISLNK(inode->i_mode))
 		return -EOPNOTSUPP;
@@ -296,7 +294,7 @@ reiserfs_set_acl(struct reiserfs_transaction_handle *th, struct inode *inode,
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		name = POSIX_ACL_XATTR_ACCESS;
-		p_acl = &reiserfs_i->i_acl_access;
+		p_acl = &inode->i_acl;
 		if (acl) {
 			mode_t mode = inode->i_mode;
 			error = posix_acl_equiv_mode(acl, &mode);
@@ -311,7 +309,7 @@ reiserfs_set_acl(struct reiserfs_transaction_handle *th, struct inode *inode,
 		break;
 	case ACL_TYPE_DEFAULT:
 		name = POSIX_ACL_XATTR_DEFAULT;
-		p_acl = &reiserfs_i->i_acl_default;
+		p_acl = &inode->i_default_acl;
 		if (!S_ISDIR(inode->i_mode))
 			return acl ? -EACCES : 0;
 		break;
diff --git a/include/linux/reiserfs_acl.h b/include/linux/reiserfs_acl.h
index 8f4d8d718b10..b4448853900e 100644
--- a/include/linux/reiserfs_acl.h
+++ b/include/linux/reiserfs_acl.h
@@ -56,15 +56,6 @@ int reiserfs_cache_default_acl(struct inode *dir);
 extern struct xattr_handler reiserfs_posix_acl_default_handler;
 extern struct xattr_handler reiserfs_posix_acl_access_handler;
 
-static inline void reiserfs_init_acl_access(struct inode *inode)
-{
-	REISERFS_I(inode)->i_acl_access = ACL_NOT_CACHED;
-}
-
-static inline void reiserfs_init_acl_default(struct inode *inode)
-{
-	REISERFS_I(inode)->i_acl_default = ACL_NOT_CACHED;
-}
 #else
 
 #define reiserfs_cache_default_acl(inode) 0
@@ -86,12 +77,4 @@ reiserfs_inherit_default_acl(struct reiserfs_transaction_handle *th,
 {
 	return 0;
 }
-
-static inline void reiserfs_init_acl_access(struct inode *inode)
-{
-}
-
-static inline void reiserfs_init_acl_default(struct inode *inode)
-{
-}
 #endif
diff --git a/include/linux/reiserfs_fs_i.h b/include/linux/reiserfs_fs_i.h
index 76360b36ac33..89f4d3abbf5a 100644
--- a/include/linux/reiserfs_fs_i.h
+++ b/include/linux/reiserfs_fs_i.h
@@ -54,10 +54,6 @@ struct reiserfs_inode_info {
 	unsigned int i_trans_id;
 	struct reiserfs_journal_list *i_jl;
 	struct mutex i_mmap;
-#ifdef CONFIG_REISERFS_FS_POSIX_ACL
-	struct posix_acl *i_acl_access;
-	struct posix_acl *i_acl_default;
-#endif
 #ifdef CONFIG_REISERFS_FS_XATTR
 	struct rw_semaphore i_xattr_sem;
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 06b16e9f68edaa1e71aee943d3c030bcf7380af1 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 8 Jun 2009 19:56:00 -0400
Subject: switch shmem to inode->i_acl

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/shmem_fs.h |  8 --------
 mm/shmem.c               |  9 ++++-----
 mm/shmem_acl.c           | 29 ++++++-----------------------
 3 files changed, 10 insertions(+), 36 deletions(-)

diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h
index fd83f2584b15..abff6c9b413c 100644
--- a/include/linux/shmem_fs.h
+++ b/include/linux/shmem_fs.h
@@ -19,10 +19,6 @@ struct shmem_inode_info {
 	swp_entry_t		i_direct[SHMEM_NR_DIRECT]; /* first blocks */
 	struct list_head	swaplist;	/* chain of maybes on swap */
 	struct inode		vfs_inode;
-#ifdef CONFIG_TMPFS_POSIX_ACL
-	struct posix_acl	*i_acl;
-	struct posix_acl	*i_default_acl;
-#endif
 };
 
 struct shmem_sb_info {
@@ -45,7 +41,6 @@ static inline struct shmem_inode_info *SHMEM_I(struct inode *inode)
 #ifdef CONFIG_TMPFS_POSIX_ACL
 int shmem_permission(struct inode *, int);
 int shmem_acl_init(struct inode *, struct inode *);
-void shmem_acl_destroy_inode(struct inode *);
 
 extern struct xattr_handler shmem_xattr_acl_access_handler;
 extern struct xattr_handler shmem_xattr_acl_default_handler;
@@ -57,9 +52,6 @@ static inline int shmem_acl_init(struct inode *inode, struct inode *dir)
 {
 	return 0;
 }
-static inline void shmem_acl_destroy_inode(struct inode *inode)
-{
-}
 #endif  /* CONFIG_TMPFS_POSIX_ACL */
 
 #endif
diff --git a/mm/shmem.c b/mm/shmem.c
index e89d7ec18eda..5f2019fc7895 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -2379,6 +2379,10 @@ static struct inode *shmem_alloc_inode(struct super_block *sb)
 	p = (struct shmem_inode_info *)kmem_cache_alloc(shmem_inode_cachep, GFP_KERNEL);
 	if (!p)
 		return NULL;
+#ifdef CONFIG_TMPFS_POSIX_ACL
+	p->vfs_inode.i_acl = NULL;
+	p->vfs_inode.i_default_acl = NULL;
+#endif
 	return &p->vfs_inode;
 }
 
@@ -2388,7 +2392,6 @@ static void shmem_destroy_inode(struct inode *inode)
 		/* only struct inode is valid if it's an inline symlink */
 		mpol_free_shared_policy(&SHMEM_I(inode)->policy);
 	}
-	shmem_acl_destroy_inode(inode);
 	kmem_cache_free(shmem_inode_cachep, SHMEM_I(inode));
 }
 
@@ -2397,10 +2400,6 @@ static void init_once(void *foo)
 	struct shmem_inode_info *p = (struct shmem_inode_info *) foo;
 
 	inode_init_once(&p->vfs_inode);
-#ifdef CONFIG_TMPFS_POSIX_ACL
-	p->i_acl = NULL;
-	p->i_default_acl = NULL;
-#endif
 }
 
 static int init_inodecache(void)
diff --git a/mm/shmem_acl.c b/mm/shmem_acl.c
index 8e5aadd7dcd6..606a8e757a42 100644
--- a/mm/shmem_acl.c
+++ b/mm/shmem_acl.c
@@ -22,11 +22,11 @@ shmem_get_acl(struct inode *inode, int type)
 	spin_lock(&inode->i_lock);
 	switch(type) {
 		case ACL_TYPE_ACCESS:
-			acl = posix_acl_dup(SHMEM_I(inode)->i_acl);
+			acl = posix_acl_dup(inode->i_acl);
 			break;
 
 		case ACL_TYPE_DEFAULT:
-			acl = posix_acl_dup(SHMEM_I(inode)->i_default_acl);
+			acl = posix_acl_dup(inode->i_default_acl);
 			break;
 	}
 	spin_unlock(&inode->i_lock);
@@ -45,13 +45,13 @@ shmem_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 	spin_lock(&inode->i_lock);
 	switch(type) {
 		case ACL_TYPE_ACCESS:
-			free = SHMEM_I(inode)->i_acl;
-			SHMEM_I(inode)->i_acl = posix_acl_dup(acl);
+			free = inode->i_acl;
+			inode->i_acl = posix_acl_dup(acl);
 			break;
 
 		case ACL_TYPE_DEFAULT:
-			free = SHMEM_I(inode)->i_default_acl;
-			SHMEM_I(inode)->i_default_acl = posix_acl_dup(acl);
+			free = inode->i_default_acl;
+			inode->i_default_acl = posix_acl_dup(acl);
 			break;
 	}
 	spin_unlock(&inode->i_lock);
@@ -154,23 +154,6 @@ shmem_acl_init(struct inode *inode, struct inode *dir)
 	return generic_acl_init(inode, dir, &shmem_acl_ops);
 }
 
-/**
- * shmem_acl_destroy_inode  -  destroy acls hanging off the in-memory inode
- *
- * This is done before destroying the actual inode.
- */
-
-void
-shmem_acl_destroy_inode(struct inode *inode)
-{
-	if (SHMEM_I(inode)->i_acl)
-		posix_acl_release(SHMEM_I(inode)->i_acl);
-	SHMEM_I(inode)->i_acl = NULL;
-	if (SHMEM_I(inode)->i_default_acl)
-		posix_acl_release(SHMEM_I(inode)->i_default_acl);
-	SHMEM_I(inode)->i_default_acl = NULL;
-}
-
 /**
  * shmem_check_acl  -  check_acl() callback for generic_permission()
  */
-- 
cgit v1.2.3-59-g8ed1b


From 073aaa1b142461d91f83da66db1184d7c1b1edea Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 9 Jun 2009 12:11:54 -0400
Subject: helpers for acl caching + switch to those

helpers: get_cached_acl(inode, type), set_cached_acl(inode, type, acl),
forget_cached_acl(inode, type).

ubifs/xattr.c needed includes reordered, the rest is a plain switchover.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/btrfs/acl.c            | 44 +++++--------------------
 fs/ext2/acl.c             | 79 ++++++++++----------------------------------
 fs/ext3/acl.c             | 83 +++++++++++------------------------------------
 fs/ext4/acl.c             | 65 +++++--------------------------------
 fs/jffs2/acl.c            | 60 +++++++---------------------------
 fs/jfs/acl.c              | 32 ++++++++----------
 fs/jfs/xattr.c            | 10 ++----
 fs/reiserfs/xattr_acl.c   | 49 ++++++----------------------
 fs/ubifs/xattr.c          |  2 +-
 include/linux/posix_acl.h | 64 ++++++++++++++++++++++++++++++++++++
 10 files changed, 155 insertions(+), 333 deletions(-)

diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index 6db8a42a3e5e..f128427b995b 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -29,51 +29,28 @@
 
 #ifdef CONFIG_FS_POSIX_ACL
 
-static void btrfs_update_cached_acl(struct inode *inode,
-				    struct posix_acl **p_acl,
-				    struct posix_acl *acl)
-{
-	spin_lock(&inode->i_lock);
-	if (*p_acl && *p_acl != ACL_NOT_CACHED)
-		posix_acl_release(*p_acl);
-	*p_acl = posix_acl_dup(acl);
-	spin_unlock(&inode->i_lock);
-}
-
 static struct posix_acl *btrfs_get_acl(struct inode *inode, int type)
 {
 	int size;
 	const char *name;
 	char *value = NULL;
-	struct posix_acl *acl = NULL, **p_acl;
+	struct posix_acl *acl;
+
+	acl = get_cached_acl(inode, type);
+	if (acl != ACL_NOT_CACHED)
+		return acl;
 
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		name = POSIX_ACL_XATTR_ACCESS;
-		p_acl = &inode->i_acl;
 		break;
 	case ACL_TYPE_DEFAULT:
 		name = POSIX_ACL_XATTR_DEFAULT;
-		p_acl = &inode->i_default_acl;
 		break;
 	default:
-		return ERR_PTR(-EINVAL);
+		BUG();
 	}
 
-	/* Handle the cached NULL acl case without locking */
-	acl = ACCESS_ONCE(*p_acl);
-	if (!acl)
-		return acl;
-
-	spin_lock(&inode->i_lock);
-	acl = *p_acl;
-	if (acl != ACL_NOT_CACHED)
-		acl = posix_acl_dup(acl);
-	spin_unlock(&inode->i_lock);
-
-	if (acl != ACL_NOT_CACHED)
-		return acl;
-
 	size = __btrfs_getxattr(inode, name, "", 0);
 	if (size > 0) {
 		value = kzalloc(size, GFP_NOFS);
@@ -82,13 +59,13 @@ static struct posix_acl *btrfs_get_acl(struct inode *inode, int type)
 		size = __btrfs_getxattr(inode, name, value, size);
 		if (size > 0) {
 			acl = posix_acl_from_xattr(value, size);
-			btrfs_update_cached_acl(inode, p_acl, acl);
+			set_cached_acl(inode, type, acl);
 		}
 		kfree(value);
 	} else if (size == -ENOENT || size == -ENODATA || size == 0) {
 		/* FIXME, who returns -ENOENT?  I think nobody */
 		acl = NULL;
-		btrfs_update_cached_acl(inode, p_acl, acl);
+		set_cached_acl(inode, type, acl);
 	} else {
 		acl = ERR_PTR(-EIO);
 	}
@@ -121,7 +98,6 @@ static int btrfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 {
 	int ret, size = 0;
 	const char *name;
-	struct posix_acl **p_acl;
 	char *value = NULL;
 	mode_t mode;
 
@@ -141,13 +117,11 @@ static int btrfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 		ret = 0;
 		inode->i_mode = mode;
 		name = POSIX_ACL_XATTR_ACCESS;
-		p_acl = &inode->i_acl;
 		break;
 	case ACL_TYPE_DEFAULT:
 		if (!S_ISDIR(inode->i_mode))
 			return acl ? -EINVAL : 0;
 		name = POSIX_ACL_XATTR_DEFAULT;
-		p_acl = &inode->i_default_acl;
 		break;
 	default:
 		return -EINVAL;
@@ -172,7 +146,7 @@ out:
 	kfree(value);
 
 	if (!ret)
-		btrfs_update_cached_acl(inode, p_acl, acl);
+		set_cached_acl(inode, type, acl);
 
 	return ret;
 }
diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c
index d2ffddc12117..d636e1297cad 100644
--- a/fs/ext2/acl.c
+++ b/fs/ext2/acl.c
@@ -125,30 +125,6 @@ fail:
 	return ERR_PTR(-EINVAL);
 }
 
-static inline struct posix_acl *
-ext2_iget_acl(struct inode *inode, struct posix_acl **i_acl)
-{
-	struct posix_acl *acl = ACL_NOT_CACHED;
-
-	spin_lock(&inode->i_lock);
-	if (*i_acl != ACL_NOT_CACHED)
-		acl = posix_acl_dup(*i_acl);
-	spin_unlock(&inode->i_lock);
-
-	return acl;
-}
-
-static inline void
-ext2_iset_acl(struct inode *inode, struct posix_acl **i_acl,
-		   struct posix_acl *acl)
-{
-	spin_lock(&inode->i_lock);
-	if (*i_acl != ACL_NOT_CACHED)
-		posix_acl_release(*i_acl);
-	*i_acl = posix_acl_dup(acl);
-	spin_unlock(&inode->i_lock);
-}
-
 /*
  * inode->i_mutex: don't care
  */
@@ -163,23 +139,19 @@ ext2_get_acl(struct inode *inode, int type)
 	if (!test_opt(inode->i_sb, POSIX_ACL))
 		return NULL;
 
-	switch(type) {
-		case ACL_TYPE_ACCESS:
-			acl = ext2_iget_acl(inode, &inode->i_acl);
-			if (acl != ACL_NOT_CACHED)
-				return acl;
-			name_index = EXT2_XATTR_INDEX_POSIX_ACL_ACCESS;
-			break;
-
-		case ACL_TYPE_DEFAULT:
-			acl = ext2_iget_acl(inode, &inode->i_default_acl);
-			if (acl != ACL_NOT_CACHED)
-				return acl;
-			name_index = EXT2_XATTR_INDEX_POSIX_ACL_DEFAULT;
-			break;
-
-		default:
-			return ERR_PTR(-EINVAL);
+	acl = get_cached_acl(inode, type);
+	if (acl != ACL_NOT_CACHED)
+		return acl;
+
+	switch (type) {
+	case ACL_TYPE_ACCESS:
+		name_index = EXT2_XATTR_INDEX_POSIX_ACL_ACCESS;
+		break;
+	case ACL_TYPE_DEFAULT:
+		name_index = EXT2_XATTR_INDEX_POSIX_ACL_DEFAULT;
+		break;
+	default:
+		BUG();
 	}
 	retval = ext2_xattr_get(inode, name_index, "", NULL, 0);
 	if (retval > 0) {
@@ -196,17 +168,9 @@ ext2_get_acl(struct inode *inode, int type)
 		acl = ERR_PTR(retval);
 	kfree(value);
 
-	if (!IS_ERR(acl)) {
-		switch(type) {
-			case ACL_TYPE_ACCESS:
-				ext2_iset_acl(inode, &inode->i_acl, acl);
-				break;
+	if (!IS_ERR(acl))
+		set_cached_acl(inode, type, acl);
 
-			case ACL_TYPE_DEFAULT:
-				ext2_iset_acl(inode, &inode->i_default_acl, acl);
-				break;
-		}
-	}
 	return acl;
 }
 
@@ -261,17 +225,8 @@ ext2_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 	error = ext2_xattr_set(inode, name_index, "", value, size, 0);
 
 	kfree(value);
-	if (!error) {
-		switch(type) {
-			case ACL_TYPE_ACCESS:
-				ext2_iset_acl(inode, &inode->i_acl, acl);
-				break;
-
-			case ACL_TYPE_DEFAULT:
-				ext2_iset_acl(inode, &inode->i_default_acl, acl);
-				break;
-		}
-	}
+	if (!error)
+		set_cached_acl(inode, type, acl);
 	return error;
 }
 
diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c
index a9707689d9e1..e167bae37ef0 100644
--- a/fs/ext3/acl.c
+++ b/fs/ext3/acl.c
@@ -126,33 +126,6 @@ fail:
 	return ERR_PTR(-EINVAL);
 }
 
-static inline struct posix_acl *
-ext3_iget_acl(struct inode *inode, struct posix_acl **i_acl)
-{
-	struct posix_acl *acl = ACCESS_ONCE(*i_acl);
-
-	if (acl) {
-		spin_lock(&inode->i_lock);
-		acl = *i_acl;
-		if (acl != ACL_NOT_CACHED)
-			acl = posix_acl_dup(acl);
-		spin_unlock(&inode->i_lock);
-	}
-
-	return acl;
-}
-
-static inline void
-ext3_iset_acl(struct inode *inode, struct posix_acl **i_acl,
-                  struct posix_acl *acl)
-{
-	spin_lock(&inode->i_lock);
-	if (*i_acl != ACL_NOT_CACHED)
-		posix_acl_release(*i_acl);
-	*i_acl = posix_acl_dup(acl);
-	spin_unlock(&inode->i_lock);
-}
-
 /*
  * Inode operation get_posix_acl().
  *
@@ -169,24 +142,21 @@ ext3_get_acl(struct inode *inode, int type)
 	if (!test_opt(inode->i_sb, POSIX_ACL))
 		return NULL;
 
-	switch(type) {
-		case ACL_TYPE_ACCESS:
-			acl = ext3_iget_acl(inode, &inode->i_acl);
-			if (acl != ACL_NOT_CACHED)
-				return acl;
-			name_index = EXT3_XATTR_INDEX_POSIX_ACL_ACCESS;
-			break;
-
-		case ACL_TYPE_DEFAULT:
-			acl = ext3_iget_acl(inode, &inode->i_default_acl);
-			if (acl != ACL_NOT_CACHED)
-				return acl;
-			name_index = EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT;
-			break;
-
-		default:
-			return ERR_PTR(-EINVAL);
+	acl = get_cached_acl(inode, type);
+	if (acl != ACL_NOT_CACHED)
+		return acl;
+
+	switch (type) {
+	case ACL_TYPE_ACCESS:
+		name_index = EXT3_XATTR_INDEX_POSIX_ACL_ACCESS;
+		break;
+	case ACL_TYPE_DEFAULT:
+		name_index = EXT3_XATTR_INDEX_POSIX_ACL_DEFAULT;
+		break;
+	default:
+		BUG();
 	}
+
 	retval = ext3_xattr_get(inode, name_index, "", NULL, 0);
 	if (retval > 0) {
 		value = kmalloc(retval, GFP_NOFS);
@@ -202,17 +172,9 @@ ext3_get_acl(struct inode *inode, int type)
 		acl = ERR_PTR(retval);
 	kfree(value);
 
-	if (!IS_ERR(acl)) {
-		switch(type) {
-			case ACL_TYPE_ACCESS:
-				ext3_iset_acl(inode, &inode->i_acl, acl);
-				break;
+	if (!IS_ERR(acl))
+		set_cached_acl(inode, type, acl);
 
-			case ACL_TYPE_DEFAULT:
-				ext3_iset_acl(inode, &inode->i_default_acl, acl);
-				break;
-		}
-	}
 	return acl;
 }
 
@@ -269,17 +231,10 @@ ext3_set_acl(handle_t *handle, struct inode *inode, int type,
 				      value, size, 0);
 
 	kfree(value);
-	if (!error) {
-		switch(type) {
-			case ACL_TYPE_ACCESS:
-				ext3_iset_acl(inode, &inode->i_acl, acl);
-				break;
 
-			case ACL_TYPE_DEFAULT:
-				ext3_iset_acl(inode, &inode->i_default_acl, acl);
-				break;
-		}
-	}
+	if (!error)
+		set_cached_acl(inode, type, acl);
+
 	return error;
 }
 
diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c
index 0084e3a19d86..f6d8967149ca 100644
--- a/fs/ext4/acl.c
+++ b/fs/ext4/acl.c
@@ -126,33 +126,6 @@ fail:
 	return ERR_PTR(-EINVAL);
 }
 
-static inline struct posix_acl *
-ext4_iget_acl(struct inode *inode, struct posix_acl **i_acl)
-{
-	struct posix_acl *acl = ACCESS_ONCE(*i_acl);
-
-	if (acl) {
-		spin_lock(&inode->i_lock);
-		acl = *i_acl;
-		if (acl != ACL_NOT_CACHED)
-			acl = posix_acl_dup(acl);
-		spin_unlock(&inode->i_lock);
-	}
-
-	return acl;
-}
-
-static inline void
-ext4_iset_acl(struct inode *inode, struct posix_acl **i_acl,
-		struct posix_acl *acl)
-{
-	spin_lock(&inode->i_lock);
-	if (*i_acl != ACL_NOT_CACHED)
-		posix_acl_release(*i_acl);
-	*i_acl = posix_acl_dup(acl);
-	spin_unlock(&inode->i_lock);
-}
-
 /*
  * Inode operation get_posix_acl().
  *
@@ -169,23 +142,19 @@ ext4_get_acl(struct inode *inode, int type)
 	if (!test_opt(inode->i_sb, POSIX_ACL))
 		return NULL;
 
+	acl = get_cached_acl(inode, type);
+	if (acl != ACL_NOT_CACHED)
+		return acl;
+
 	switch (type) {
 	case ACL_TYPE_ACCESS:
-		acl = ext4_iget_acl(inode, &inode->i_acl);
-		if (acl != ACL_NOT_CACHED)
-			return acl;
 		name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS;
 		break;
-
 	case ACL_TYPE_DEFAULT:
-		acl = ext4_iget_acl(inode, &inode->i_default_acl);
-		if (acl != ACL_NOT_CACHED)
-			return acl;
 		name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT;
 		break;
-
 	default:
-		return ERR_PTR(-EINVAL);
+		BUG();
 	}
 	retval = ext4_xattr_get(inode, name_index, "", NULL, 0);
 	if (retval > 0) {
@@ -202,17 +171,9 @@ ext4_get_acl(struct inode *inode, int type)
 		acl = ERR_PTR(retval);
 	kfree(value);
 
-	if (!IS_ERR(acl)) {
-		switch (type) {
-		case ACL_TYPE_ACCESS:
-			ext4_iset_acl(inode, &inode->i_acl, acl);
-			break;
+	if (!IS_ERR(acl))
+		set_cached_acl(inode, type, acl);
 
-		case ACL_TYPE_DEFAULT:
-			ext4_iset_acl(inode, &inode->i_default_acl, acl);
-			break;
-		}
-	}
 	return acl;
 }
 
@@ -269,17 +230,9 @@ ext4_set_acl(handle_t *handle, struct inode *inode, int type,
 				      value, size, 0);
 
 	kfree(value);
-	if (!error) {
-		switch (type) {
-		case ACL_TYPE_ACCESS:
-			ext4_iset_acl(inode, &inode->i_acl, acl);
-			break;
+	if (!error)
+		set_cached_acl(inode, type, acl);
 
-		case ACL_TYPE_DEFAULT:
-			ext4_iset_acl(inode, &inode->i_default_acl, acl);
-			break;
-		}
-	}
 	return error;
 }
 
diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c
index ac16589ebbd1..edd2ad6416d8 100644
--- a/fs/jffs2/acl.c
+++ b/fs/jffs2/acl.c
@@ -156,47 +156,25 @@ static void *jffs2_acl_to_medium(const struct posix_acl *acl, size_t *size)
 	return ERR_PTR(-EINVAL);
 }
 
-static struct posix_acl *jffs2_iget_acl(struct inode *inode, struct posix_acl **i_acl)
-{
-	struct posix_acl *acl = ACL_NOT_CACHED;
-
-	spin_lock(&inode->i_lock);
-	if (*i_acl != ACL_NOT_CACHED)
-		acl = posix_acl_dup(*i_acl);
-	spin_unlock(&inode->i_lock);
-	return acl;
-}
-
-static void jffs2_iset_acl(struct inode *inode, struct posix_acl **i_acl, struct posix_acl *acl)
-{
-	spin_lock(&inode->i_lock);
-	if (*i_acl != ACL_NOT_CACHED)
-		posix_acl_release(*i_acl);
-	*i_acl = posix_acl_dup(acl);
-	spin_unlock(&inode->i_lock);
-}
-
 static struct posix_acl *jffs2_get_acl(struct inode *inode, int type)
 {
 	struct posix_acl *acl;
 	char *value = NULL;
 	int rc, xprefix;
 
+	acl = get_cached_acl(inode, type);
+	if (acl != ACL_NOT_CACHED)
+		return acl;
+
 	switch (type) {
 	case ACL_TYPE_ACCESS:
-		acl = jffs2_iget_acl(inode, &inode->i_acl);
-		if (acl != ACL_NOT_CACHED)
-			return acl;
 		xprefix = JFFS2_XPREFIX_ACL_ACCESS;
 		break;
 	case ACL_TYPE_DEFAULT:
-		acl = jffs2_iget_acl(inode, &inode->i_default_acl);
-		if (acl != ACL_NOT_CACHED)
-			return acl;
 		xprefix = JFFS2_XPREFIX_ACL_DEFAULT;
 		break;
 	default:
-		return ERR_PTR(-EINVAL);
+		BUG();
 	}
 	rc = do_jffs2_getxattr(inode, xprefix, "", NULL, 0);
 	if (rc > 0) {
@@ -214,16 +192,8 @@ static struct posix_acl *jffs2_get_acl(struct inode *inode, int type)
 	}
 	if (value)
 		kfree(value);
-	if (!IS_ERR(acl)) {
-		switch (type) {
-		case ACL_TYPE_ACCESS:
-			jffs2_iset_acl(inode, &inode->i_acl, acl);
-			break;
-		case ACL_TYPE_DEFAULT:
-			jffs2_iset_acl(inode, &inode->i_default_acl, acl);
-			break;
-		}
-	}
+	if (!IS_ERR(acl))
+		set_cached_acl(inode, type, acl);
 	return acl;
 }
 
@@ -283,16 +253,8 @@ static int jffs2_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 		return -EINVAL;
 	}
 	rc = __jffs2_set_acl(inode, xprefix, acl);
-	if (!rc) {
-		switch(type) {
-		case ACL_TYPE_ACCESS:
-			jffs2_iset_acl(inode, &inode->i_acl, acl);
-			break;
-		case ACL_TYPE_DEFAULT:
-			jffs2_iset_acl(inode, &inode->i_default_acl, acl);
-			break;
-		}
-	}
+	if (!rc)
+		set_cached_acl(inode, type, acl);
 	return rc;
 }
 
@@ -336,7 +298,7 @@ int jffs2_init_acl_pre(struct inode *dir_i, struct inode *inode, int *i_mode)
 		*i_mode &= ~current_umask();
 	} else {
 		if (S_ISDIR(*i_mode))
-			jffs2_iset_acl(inode, &inode->i_default_acl, acl);
+			set_cached_acl(inode, ACL_TYPE_DEFAULT, acl);
 
 		clone = posix_acl_clone(acl, GFP_KERNEL);
 		if (!clone)
@@ -347,7 +309,7 @@ int jffs2_init_acl_pre(struct inode *dir_i, struct inode *inode, int *i_mode)
 			return rc;
 		}
 		if (rc > 0)
-			jffs2_iset_acl(inode, &inode->i_acl, clone);
+			set_cached_acl(inode, ACL_TYPE_ACCESS, clone);
 
 		posix_acl_release(clone);
 	}
diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c
index 5fcfc9857c11..f272bf032e1e 100644
--- a/fs/jfs/acl.c
+++ b/fs/jfs/acl.c
@@ -31,26 +31,24 @@ static struct posix_acl *jfs_get_acl(struct inode *inode, int type)
 {
 	struct posix_acl *acl;
 	char *ea_name;
-	struct posix_acl **p_acl;
 	int size;
 	char *value = NULL;
 
+	acl = get_cached_acl(inode, type);
+	if (acl != ACL_NOT_CACHED)
+		return acl;
+
 	switch(type) {
 		case ACL_TYPE_ACCESS:
 			ea_name = POSIX_ACL_XATTR_ACCESS;
-			p_acl = &inode->i_acl;
 			break;
 		case ACL_TYPE_DEFAULT:
 			ea_name = POSIX_ACL_XATTR_DEFAULT;
-			p_acl = &inode->i_default_acl;
 			break;
 		default:
 			return ERR_PTR(-EINVAL);
 	}
 
-	if (*p_acl != ACL_NOT_CACHED)
-		return posix_acl_dup(*p_acl);
-
 	size = __jfs_getxattr(inode, ea_name, NULL, 0);
 
 	if (size > 0) {
@@ -61,17 +59,18 @@ static struct posix_acl *jfs_get_acl(struct inode *inode, int type)
 	}
 
 	if (size < 0) {
-		if (size == -ENODATA) {
-			*p_acl = NULL;
+		if (size == -ENODATA)
 			acl = NULL;
-		} else
+		else
 			acl = ERR_PTR(size);
 	} else {
 		acl = posix_acl_from_xattr(value, size);
-		if (!IS_ERR(acl))
-			*p_acl = posix_acl_dup(acl);
 	}
 	kfree(value);
+	if (!IS_ERR(acl)) {
+		set_cached_acl(inode, type, acl);
+		posix_acl_release(acl);
+	}
 	return acl;
 }
 
@@ -79,7 +78,6 @@ static int jfs_set_acl(tid_t tid, struct inode *inode, int type,
 		       struct posix_acl *acl)
 {
 	char *ea_name;
-	struct posix_acl **p_acl;
 	int rc;
 	int size = 0;
 	char *value = NULL;
@@ -90,11 +88,9 @@ static int jfs_set_acl(tid_t tid, struct inode *inode, int type,
 	switch(type) {
 		case ACL_TYPE_ACCESS:
 			ea_name = POSIX_ACL_XATTR_ACCESS;
-			p_acl = &inode->i_acl;
 			break;
 		case ACL_TYPE_DEFAULT:
 			ea_name = POSIX_ACL_XATTR_DEFAULT;
-			p_acl = &inode->i_default_acl;
 			if (!S_ISDIR(inode->i_mode))
 				return acl ? -EACCES : 0;
 			break;
@@ -114,11 +110,9 @@ static int jfs_set_acl(tid_t tid, struct inode *inode, int type,
 out:
 	kfree(value);
 
-	if (!rc) {
-		if (*p_acl && (*p_acl != ACL_NOT_CACHED))
-			posix_acl_release(*p_acl);
-		*p_acl = posix_acl_dup(acl);
-	}
+	if (!rc)
+		set_cached_acl(inode, type, acl);
+
 	return rc;
 }
 
diff --git a/fs/jfs/xattr.c b/fs/jfs/xattr.c
index f6e90e343593..fad364548bc9 100644
--- a/fs/jfs/xattr.c
+++ b/fs/jfs/xattr.c
@@ -727,10 +727,7 @@ static int can_set_system_xattr(struct inode *inode, const char *name,
 		/*
 		 * We're changing the ACL.  Get rid of the cached one
 		 */
-		acl =inode->i_acl;
-		if (acl != ACL_NOT_CACHED)
-			posix_acl_release(acl);
-		inode->i_acl = ACL_NOT_CACHED;
+		forget_cached_acl(inode, ACL_TYPE_ACCESS);
 
 		return 0;
 	} else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0) {
@@ -746,10 +743,7 @@ static int can_set_system_xattr(struct inode *inode, const char *name,
 		/*
 		 * We're changing the default ACL.  Get rid of the cached one
 		 */
-		acl = inode->i_default_acl;
-		if (acl && (acl != ACL_NOT_CACHED))
-			posix_acl_release(acl);
-		inode->i_default_acl = ACL_NOT_CACHED;
+		forget_cached_acl(inode, ACL_TYPE_DEFAULT);
 
 		return 0;
 	}
diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c
index b6e473faa8b8..35d6e672a279 100644
--- a/fs/reiserfs/xattr_acl.c
+++ b/fs/reiserfs/xattr_acl.c
@@ -188,29 +188,6 @@ static void *posix_acl_to_disk(const struct posix_acl *acl, size_t * size)
 	return ERR_PTR(-EINVAL);
 }
 
-static inline void iset_acl(struct inode *inode, struct posix_acl **i_acl,
-			    struct posix_acl *acl)
-{
-	spin_lock(&inode->i_lock);
-	if (*i_acl != ACL_NOT_CACHED)
-		posix_acl_release(*i_acl);
-	*i_acl = posix_acl_dup(acl);
-	spin_unlock(&inode->i_lock);
-}
-
-static inline struct posix_acl *iget_acl(struct inode *inode,
-					 struct posix_acl **i_acl)
-{
-	struct posix_acl *acl = ACL_NOT_CACHED;
-
-	spin_lock(&inode->i_lock);
-	if (*i_acl != ACL_NOT_CACHED)
-		acl = posix_acl_dup(*i_acl);
-	spin_unlock(&inode->i_lock);
-
-	return acl;
-}
-
 /*
  * Inode operation get_posix_acl().
  *
@@ -220,31 +197,29 @@ static inline struct posix_acl *iget_acl(struct inode *inode,
 struct posix_acl *reiserfs_get_acl(struct inode *inode, int type)
 {
 	char *name, *value;
-	struct posix_acl *acl, **p_acl;
+	struct posix_acl *acl;
 	int size;
 	int retval;
 
+	acl = get_cached_acl(inode, type);
+	if (acl != ACL_NOT_CACHED)
+		return acl;
+
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		name = POSIX_ACL_XATTR_ACCESS;
-		p_acl = &inode->i_acl;
 		break;
 	case ACL_TYPE_DEFAULT:
 		name = POSIX_ACL_XATTR_DEFAULT;
-		p_acl = &inode->i_default_acl;
 		break;
 	default:
-		return ERR_PTR(-EINVAL);
+		BUG();
 	}
 
-	acl = iget_acl(inode, p_acl);
-	if (acl != ACL_NOT_CACHED)
-		return acl;
-
 	size = reiserfs_xattr_get(inode, name, NULL, 0);
 	if (size < 0) {
 		if (size == -ENODATA || size == -ENOSYS) {
-			*p_acl = NULL;
+			set_cached_acl(inode, type, NULL);
 			return NULL;
 		}
 		return ERR_PTR(size);
@@ -259,14 +234,13 @@ struct posix_acl *reiserfs_get_acl(struct inode *inode, int type)
 		/* This shouldn't actually happen as it should have
 		   been caught above.. but just in case */
 		acl = NULL;
-		*p_acl = acl;
 	} else if (retval < 0) {
 		acl = ERR_PTR(retval);
 	} else {
 		acl = posix_acl_from_disk(value, retval);
-		if (!IS_ERR(acl))
-			iset_acl(inode, p_acl, acl);
 	}
+	if (!IS_ERR(acl))
+		set_cached_acl(inode, type, acl);
 
 	kfree(value);
 	return acl;
@@ -284,7 +258,6 @@ reiserfs_set_acl(struct reiserfs_transaction_handle *th, struct inode *inode,
 {
 	char *name;
 	void *value = NULL;
-	struct posix_acl **p_acl;
 	size_t size = 0;
 	int error;
 
@@ -294,7 +267,6 @@ reiserfs_set_acl(struct reiserfs_transaction_handle *th, struct inode *inode,
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		name = POSIX_ACL_XATTR_ACCESS;
-		p_acl = &inode->i_acl;
 		if (acl) {
 			mode_t mode = inode->i_mode;
 			error = posix_acl_equiv_mode(acl, &mode);
@@ -309,7 +281,6 @@ reiserfs_set_acl(struct reiserfs_transaction_handle *th, struct inode *inode,
 		break;
 	case ACL_TYPE_DEFAULT:
 		name = POSIX_ACL_XATTR_DEFAULT;
-		p_acl = &inode->i_default_acl;
 		if (!S_ISDIR(inode->i_mode))
 			return acl ? -EACCES : 0;
 		break;
@@ -342,7 +313,7 @@ reiserfs_set_acl(struct reiserfs_transaction_handle *th, struct inode *inode,
 	kfree(value);
 
 	if (!error)
-		iset_acl(inode, p_acl, acl);
+		set_cached_acl(inode, type, acl);
 
 	return error;
 }
diff --git a/fs/ubifs/xattr.c b/fs/ubifs/xattr.c
index cfd31e229c89..adafcf556531 100644
--- a/fs/ubifs/xattr.c
+++ b/fs/ubifs/xattr.c
@@ -55,9 +55,9 @@
  * ACL support is not implemented.
  */
 
+#include "ubifs.h"
 #include <linux/xattr.h>
 #include <linux/posix_acl_xattr.h>
-#include "ubifs.h"
 
 /*
  * Limit the number of extended attributes per inode so that the total size
diff --git a/include/linux/posix_acl.h b/include/linux/posix_acl.h
index 4bc241290c24..0cdba01b7756 100644
--- a/include/linux/posix_acl.h
+++ b/include/linux/posix_acl.h
@@ -83,4 +83,68 @@ extern int posix_acl_chmod_masq(struct posix_acl *, mode_t);
 extern struct posix_acl *get_posix_acl(struct inode *, int);
 extern int set_posix_acl(struct inode *, int, struct posix_acl *);
 
+static inline struct posix_acl *get_cached_acl(struct inode *inode, int type)
+{
+	struct posix_acl **p, *acl;
+	switch (type) {
+	case ACL_TYPE_ACCESS:
+		p = &inode->i_acl;
+		break;
+	case ACL_TYPE_DEFAULT:
+		p = &inode->i_default_acl;
+		break;
+	default:
+		return ERR_PTR(-EINVAL);
+	}
+	acl = ACCESS_ONCE(*p);
+	if (acl) {
+		spin_lock(&inode->i_lock);
+		acl = *p;
+		if (acl != ACL_NOT_CACHED)
+			acl = posix_acl_dup(acl);
+		spin_unlock(&inode->i_lock);
+	}
+	return acl;
+}
+
+static inline void set_cached_acl(struct inode *inode,
+				  int type,
+				  struct posix_acl *acl)
+{
+	struct posix_acl *old = NULL;
+	spin_lock(&inode->i_lock);
+	switch (type) {
+	case ACL_TYPE_ACCESS:
+		old = inode->i_acl;
+		inode->i_acl = posix_acl_dup(acl);
+		break;
+	case ACL_TYPE_DEFAULT:
+		old = inode->i_default_acl;
+		inode->i_default_acl = posix_acl_dup(acl);
+		break;
+	}
+	spin_unlock(&inode->i_lock);
+	if (old != ACL_NOT_CACHED)
+		posix_acl_release(old);
+}
+
+static inline void forget_cached_acl(struct inode *inode, int type)
+{
+	struct posix_acl *old = NULL;
+	spin_lock(&inode->i_lock);
+	switch (type) {
+	case ACL_TYPE_ACCESS:
+		old = inode->i_acl;
+		inode->i_acl = ACL_NOT_CACHED;
+		break;
+	case ACL_TYPE_DEFAULT:
+		old = inode->i_default_acl;
+		inode->i_default_acl = ACL_NOT_CACHED;
+		break;
+	}
+	spin_unlock(&inode->i_lock);
+	if (old != ACL_NOT_CACHED)
+		posix_acl_release(old);
+}
+
 #endif  /* __LINUX_POSIX_ACL_H */
-- 
cgit v1.2.3-59-g8ed1b


From 1cbd20d820c36f52543e3e4cd0067ebf52aa388f Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 9 Jun 2009 13:29:39 -0400
Subject: switch xfs to generic acl caching helpers

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/xfs/linux-2.6/xfs_acl.c | 73 ++++++----------------------------------------
 fs/xfs/xfs_acl.h           |  4 ---
 fs/xfs/xfs_iget.c          |  2 --
 fs/xfs/xfs_inode.h         |  5 ----
 4 files changed, 9 insertions(+), 75 deletions(-)

diff --git a/fs/xfs/linux-2.6/xfs_acl.c b/fs/xfs/linux-2.6/xfs_acl.c
index 1e9d1246eebc..b23a54506446 100644
--- a/fs/xfs/linux-2.6/xfs_acl.c
+++ b/fs/xfs/linux-2.6/xfs_acl.c
@@ -25,14 +25,10 @@
 #include <linux/posix_acl_xattr.h>
 
 
-#define XFS_ACL_NOT_CACHED	((void *)-1)
-
 /*
  * Locking scheme:
  *  - all ACL updates are protected by inode->i_mutex, which is taken before
  *    calling into this file.
- *  - access and updates to the ip->i_acl and ip->i_default_acl pointers are
- *    protected by inode->i_lock.
  */
 
 STATIC struct posix_acl *
@@ -102,59 +98,35 @@ xfs_acl_to_disk(struct xfs_acl *aclp, const struct posix_acl *acl)
 	}
 }
 
-/*
- * Update the cached ACL pointer in the inode.
- *
- * Because we don't hold any locks while reading/writing the attribute
- * from/to disk another thread could have raced and updated the cached
- * ACL value before us. In that case we release the previous cached value
- * and update it with our new value.
- */
-STATIC void
-xfs_update_cached_acl(struct inode *inode, struct posix_acl **p_acl,
-		struct posix_acl *acl)
-{
-	spin_lock(&inode->i_lock);
-	if (*p_acl && *p_acl != XFS_ACL_NOT_CACHED)
-		posix_acl_release(*p_acl);
-	*p_acl = posix_acl_dup(acl);
-	spin_unlock(&inode->i_lock);
-}
-
 struct posix_acl *
 xfs_get_acl(struct inode *inode, int type)
 {
 	struct xfs_inode *ip = XFS_I(inode);
-	struct posix_acl *acl = NULL, **p_acl;
+	struct posix_acl *acl;
 	struct xfs_acl *xfs_acl;
 	int len = sizeof(struct xfs_acl);
 	char *ea_name;
 	int error;
 
+	acl = get_cached_acl(inode, type);
+	if (acl != ACL_NOT_CACHED)
+		return acl;
+
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		ea_name = SGI_ACL_FILE;
-		p_acl = &ip->i_acl;
 		break;
 	case ACL_TYPE_DEFAULT:
 		ea_name = SGI_ACL_DEFAULT;
-		p_acl = &ip->i_default_acl;
 		break;
 	default:
-		return ERR_PTR(-EINVAL);
+		BUG();
 	}
 
-	spin_lock(&inode->i_lock);
-	if (*p_acl != XFS_ACL_NOT_CACHED)
-		acl = posix_acl_dup(*p_acl);
-	spin_unlock(&inode->i_lock);
-
 	/*
 	 * If we have a cached ACLs value just return it, not need to
 	 * go out to the disk.
 	 */
-	if (acl)
-		return acl;
 
 	xfs_acl = kzalloc(sizeof(struct xfs_acl), GFP_KERNEL);
 	if (!xfs_acl)
@@ -165,7 +137,7 @@ xfs_get_acl(struct inode *inode, int type)
 		/*
 		 * If the attribute doesn't exist make sure we have a negative
 		 * cache entry, for any other error assume it is transient and
-		 * leave the cache entry as XFS_ACL_NOT_CACHED.
+		 * leave the cache entry as ACL_NOT_CACHED.
 		 */
 		if (error == -ENOATTR) {
 			acl = NULL;
@@ -179,7 +151,7 @@ xfs_get_acl(struct inode *inode, int type)
 		goto out;
 
  out_update_cache:
-	xfs_update_cached_acl(inode, p_acl, acl);
+	set_cached_acl(inode, type, acl);
  out:
 	kfree(xfs_acl);
 	return acl;
@@ -189,7 +161,6 @@ STATIC int
 xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 {
 	struct xfs_inode *ip = XFS_I(inode);
-	struct posix_acl **p_acl;
 	char *ea_name;
 	int error;
 
@@ -199,13 +170,11 @@ xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		ea_name = SGI_ACL_FILE;
-		p_acl = &ip->i_acl;
 		break;
 	case ACL_TYPE_DEFAULT:
 		if (!S_ISDIR(inode->i_mode))
 			return acl ? -EACCES : 0;
 		ea_name = SGI_ACL_DEFAULT;
-		p_acl = &ip->i_default_acl;
 		break;
 	default:
 		return -EINVAL;
@@ -242,7 +211,7 @@ xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
 	}
 
 	if (!error)
-		xfs_update_cached_acl(inode, p_acl, acl);
+		set_cached_acl(inode, type, acl);
 	return error;
 }
 
@@ -384,30 +353,6 @@ xfs_acl_chmod(struct inode *inode)
 	return error;
 }
 
-void
-xfs_inode_init_acls(struct xfs_inode *ip)
-{
-	/*
-	 * No need for locking, inode is not live yet.
-	 */
-	ip->i_acl = XFS_ACL_NOT_CACHED;
-	ip->i_default_acl = XFS_ACL_NOT_CACHED;
-}
-
-void
-xfs_inode_clear_acls(struct xfs_inode *ip)
-{
-	/*
-	 * No need for locking here, the inode is not live anymore
-	 * and just about to be freed.
-	 */
-	if (ip->i_acl != XFS_ACL_NOT_CACHED)
-		posix_acl_release(ip->i_acl);
-	if (ip->i_default_acl != XFS_ACL_NOT_CACHED)
-		posix_acl_release(ip->i_default_acl);
-}
-
-
 /*
  * System xattr handlers.
  *
diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h
index 63dc1f2efad5..947b150df8ed 100644
--- a/fs/xfs/xfs_acl.h
+++ b/fs/xfs/xfs_acl.h
@@ -46,8 +46,6 @@ extern int xfs_check_acl(struct inode *inode, int mask);
 extern struct posix_acl *xfs_get_acl(struct inode *inode, int type);
 extern int xfs_inherit_acl(struct inode *inode, struct posix_acl *default_acl);
 extern int xfs_acl_chmod(struct inode *inode);
-extern void xfs_inode_init_acls(struct xfs_inode *ip);
-extern void xfs_inode_clear_acls(struct xfs_inode *ip);
 extern int posix_acl_access_exists(struct inode *inode);
 extern int posix_acl_default_exists(struct inode *inode);
 
@@ -57,8 +55,6 @@ extern struct xattr_handler xfs_xattr_system_handler;
 # define xfs_get_acl(inode, type)			NULL
 # define xfs_inherit_acl(inode, default_acl)		0
 # define xfs_acl_chmod(inode)				0
-# define xfs_inode_init_acls(ip)
-# define xfs_inode_clear_acls(ip)
 # define posix_acl_access_exists(inode)			0
 # define posix_acl_default_exists(inode)		0
 #endif /* CONFIG_XFS_POSIX_ACL */
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index 76c540f719e4..5fcec6f020a7 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -83,7 +83,6 @@ xfs_inode_alloc(
 	memset(&ip->i_d, 0, sizeof(xfs_icdinode_t));
 	ip->i_size = 0;
 	ip->i_new_size = 0;
-	xfs_inode_init_acls(ip);
 
 	/*
 	 * Initialize inode's trace buffers.
@@ -560,7 +559,6 @@ xfs_ireclaim(
 	ASSERT(atomic_read(&ip->i_pincount) == 0);
 	ASSERT(!spin_is_locked(&ip->i_flags_lock));
 	ASSERT(completion_done(&ip->i_flush));
-	xfs_inode_clear_acls(ip);
 	kmem_zone_free(xfs_inode_zone, ip);
 }
 
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 77016702938b..1804f866a71d 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -273,11 +273,6 @@ typedef struct xfs_inode {
 	/* VFS inode */
 	struct inode		i_vnode;	/* embedded VFS inode */
 
-#ifdef CONFIG_XFS_POSIX_ACL
-	struct posix_acl	*i_acl;
-	struct posix_acl	*i_default_acl;
-#endif
-
 	/* Trace buffers per inode. */
 #ifdef XFS_INODE_TRACE
 	struct ktrace		*i_trace;	/* general inode trace */
-- 
cgit v1.2.3-59-g8ed1b


From 3d63259583278262d9153316094e315f73ebfcb5 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Wed, 24 Jun 2009 18:19:34 +0530
Subject: perf stat: Remove dead code

Remove dead code and do some code alignment.

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <1245847774.2681.2.camel@ht.satnam>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-stat.c | 44 +++++++++++++-------------------------------
 1 file changed, 13 insertions(+), 31 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 5e04fcc8d077..8420ec589506 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -59,42 +59,27 @@ static struct perf_counter_attr default_attrs[MAX_COUNTERS] = {
 
 };
 
+#define MAX_RUN			100
+
 static int			system_wide			=  0;
-static int			inherit				=  1;
 static int			verbose				=  0;
-
-static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
-
-static int			target_pid			= -1;
 static int			nr_cpus				=  0;
-static unsigned int		page_size;
+static int			run_idx				=  0;
 
+static int			run_count			=  1;
+static int			inherit				=  1;
 static int			scale				=  1;
+static int			target_pid			= -1;
 
-static const unsigned int default_count[] = {
-	1000000,
-	1000000,
-	  10000,
-	  10000,
-	1000000,
-	  10000,
-};
-
-#define MAX_RUN 100
-
-static int			run_count		=  1;
-static int			run_idx			=  0;
-
-static u64			event_res[MAX_RUN][MAX_COUNTERS][3];
-static u64			event_scaled[MAX_RUN][MAX_COUNTERS];
-
-//static u64			event_hist[MAX_RUN][MAX_COUNTERS][3];
-
+static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
 
 static u64			runtime_nsecs[MAX_RUN];
 static u64			walltime_nsecs[MAX_RUN];
 static u64			runtime_cycles[MAX_RUN];
 
+static u64			event_res[MAX_RUN][MAX_COUNTERS][3];
+static u64			event_scaled[MAX_RUN][MAX_COUNTERS];
+
 static u64			event_res_avg[MAX_COUNTERS][3];
 static u64			event_res_noise[MAX_COUNTERS][3];
 
@@ -109,7 +94,6 @@ static u64			walltime_nsecs_noise;
 static u64			runtime_cycles_avg;
 static u64			runtime_cycles_noise;
 
-
 #define ERR_PERF_OPEN \
 "Error: counter %d, sys_perf_counter_open() syscall returned with %d (%s)\n"
 
@@ -470,9 +454,9 @@ static const struct option options[] = {
 	OPT_INTEGER('p', "pid", &target_pid,
 		    "stat events on existing pid"),
 	OPT_BOOLEAN('a', "all-cpus", &system_wide,
-			    "system-wide collection from all CPUs"),
+		    "system-wide collection from all CPUs"),
 	OPT_BOOLEAN('S', "scale", &scale,
-			    "scale/normalize counters"),
+		    "scale/normalize counters"),
 	OPT_BOOLEAN('v', "verbose", &verbose,
 		    "be more verbose (show counter open errors, etc)"),
 	OPT_INTEGER('r', "repeat", &run_count,
@@ -484,8 +468,6 @@ int cmd_stat(int argc, const char **argv, const char *prefix)
 {
 	int status;
 
-	page_size = sysconf(_SC_PAGE_SIZE);
-
 	memcpy(attrs, default_attrs, sizeof(attrs));
 
 	argc = parse_options(argc, argv, options, stat_usage, 0);
@@ -515,7 +497,7 @@ int cmd_stat(int argc, const char **argv, const char *prefix)
 	status = 0;
 	for (run_idx = 0; run_idx < run_count; run_idx++) {
 		if (run_count != 1 && verbose)
-			fprintf(stderr, "[ perf stat: executing run #%d ... ]\n", run_idx+1);
+			fprintf(stderr, "[ perf stat: executing run #%d ... ]\n", run_idx + 1);
 		status = run_perf_stat(argc, argv);
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From d94d4adb7dd05b4e25f3c317a1b932ec74272a12 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 24 Jun 2009 22:35:30 +0900
Subject: sh: make set_perf_counter_pending() static inline.

Fixes up a recently introduced build error.

Reported-by: Kyle McMartin <kyle@mcmartin.ca>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/include/asm/perf_counter.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/sh/include/asm/perf_counter.h b/arch/sh/include/asm/perf_counter.h
index a8153c2aa6fa..61c2b40c802c 100644
--- a/arch/sh/include/asm/perf_counter.h
+++ b/arch/sh/include/asm/perf_counter.h
@@ -2,6 +2,6 @@
 #define __ASM_SH_PERF_COUNTER_H
 
 /* SH only supports software counters through this interface. */
-#define set_perf_counter_pending()	do { } while (0)
+static inline void set_perf_counter_pending(void) {}
 
 #endif /* __ASM_SH_PERF_COUNTER_H */
-- 
cgit v1.2.3-59-g8ed1b


From 17659c60629618c0aa67eb3cb6a77d2c52486d2e Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Wed, 24 Jun 2009 15:35:15 +0100
Subject: mtd: maps: Remove BUS_ID_SIZE from integrator_flash

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Tested-by: Catalin Marinas <catalin.marinas@arm.com>
---
 drivers/mtd/maps/integrator-flash.c | 22 ++++++++++++++--------
 1 file changed, 14 insertions(+), 8 deletions(-)

diff --git a/drivers/mtd/maps/integrator-flash.c b/drivers/mtd/maps/integrator-flash.c
index b08a798ee254..2aac41bde8b3 100644
--- a/drivers/mtd/maps/integrator-flash.c
+++ b/drivers/mtd/maps/integrator-flash.c
@@ -42,10 +42,8 @@
 #include <mach/hardware.h>
 #include <asm/system.h>
 
-#define SUBDEV_NAME_SIZE	(BUS_ID_SIZE + 2)
-
 struct armflash_subdev_info {
-	char			name[SUBDEV_NAME_SIZE];
+	char			*name;
 	struct mtd_info		*mtd;
 	struct map_info		map;
 	struct flash_platform_data *plat;
@@ -134,6 +132,8 @@ static void armflash_subdev_remove(struct armflash_subdev_info *subdev)
 		map_destroy(subdev->mtd);
 	if (subdev->map.virt)
 		iounmap(subdev->map.virt);
+	kfree(subdev->name);
+	subdev->name = NULL;
 	release_mem_region(subdev->map.phys, subdev->map.size);
 }
 
@@ -177,16 +177,22 @@ static int armflash_probe(struct platform_device *dev)
 
 		if (nr == 1)
 			/* No MTD concatenation, just use the default name */
-			snprintf(subdev->name, SUBDEV_NAME_SIZE, "%s",
-				 dev_name(&dev->dev));
+			subdev->name = kstrdup(dev_name(&dev->dev), GFP_KERNEL);
 		else
-			snprintf(subdev->name, SUBDEV_NAME_SIZE, "%s-%d",
-				 dev_name(&dev->dev), i);
+			subdev->name = kasprintf(GFP_KERNEL, "%s-%d",
+						 dev_name(&dev->dev), i);
+		if (!subdev->name) {
+			err = -ENOMEM;
+			break;
+		}
 		subdev->plat = plat;
 
 		err = armflash_subdev_probe(subdev, res);
-		if (err)
+		if (err) {
+			kfree(subdev->name);
+			subdev->name = NULL;
 			break;
+		}
 	}
 	info->nr_subdev = i;
 
-- 
cgit v1.2.3-59-g8ed1b


From 9c26f52b900f7207135bafc8789e1a4f5d43e096 Mon Sep 17 00:00:00 2001
From: Cliff Wickman <cpw@sgi.com>
Date: Wed, 24 Jun 2009 09:41:59 -0500
Subject: x86: Fix uv bau sending buffer initialization

The initialization of the UV Broadcast Assist Unit's sending
buffers was making an invalid assumption about the
initialization of an MMR that defines its address.

The BIOS will not be providing that MMR.  So
uv_activation_descriptor_init() should unconditionally set it.

Tested on UV simulator.

Signed-off-by: Cliff Wickman <cpw@sgi.com>
Cc: <stable@kernel.org> # for v2.6.30.x
LKML-Reference: <E1MJTfj-0005i1-W8@eag09.americas.sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/tlb_uv.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c
index 124d40c575df..8ccabb8a2f6a 100644
--- a/arch/x86/kernel/tlb_uv.c
+++ b/arch/x86/kernel/tlb_uv.c
@@ -711,7 +711,6 @@ uv_activation_descriptor_init(int node, int pnode)
 	unsigned long pa;
 	unsigned long m;
 	unsigned long n;
-	unsigned long mmr_image;
 	struct bau_desc *adp;
 	struct bau_desc *ad2;
 
@@ -727,12 +726,8 @@ uv_activation_descriptor_init(int node, int pnode)
 	n = pa >> uv_nshift;
 	m = pa & uv_mmask;
 
-	mmr_image = uv_read_global_mmr64(pnode, UVH_LB_BAU_SB_DESCRIPTOR_BASE);
-	if (mmr_image) {
-		uv_write_global_mmr64(pnode, (unsigned long)
-				      UVH_LB_BAU_SB_DESCRIPTOR_BASE,
-				      (n << UV_DESC_BASE_PNODE_SHIFT | m));
-	}
+	uv_write_global_mmr64(pnode, UVH_LB_BAU_SB_DESCRIPTOR_BASE,
+			      (n << UV_DESC_BASE_PNODE_SHIFT | m));
 
 	/*
 	 * initializing all 8 (UV_ITEMS_PER_DESCRIPTOR) descriptors for each
-- 
cgit v1.2.3-59-g8ed1b


From a10b32db34898d0db58a58ef76a70c374931bbff Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Wed, 24 Jun 2009 18:34:34 +0100
Subject: kgdb: kgdboc console poll hooks for serial_txx9 uart

Implement the serial polling hooks for the serial_txx9 uart for use with
kgdboc.

This patch once got SOB from Jason on Jul 2008 and (perhaps) merged into
kgdb-next branch, but lost somewhere then.  I resend it now with Jason's
Acked-by.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Acked-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/serial_txx9.c | 113 +++++++++++++++++++++++++++++++++++--------
 1 file changed, 92 insertions(+), 21 deletions(-)

diff --git a/drivers/serial/serial_txx9.c b/drivers/serial/serial_txx9.c
index 7313c2edcb83..54dd16d66a4b 100644
--- a/drivers/serial/serial_txx9.c
+++ b/drivers/serial/serial_txx9.c
@@ -461,6 +461,94 @@ static void serial_txx9_break_ctl(struct uart_port *port, int break_state)
 	spin_unlock_irqrestore(&up->port.lock, flags);
 }
 
+#if defined(CONFIG_SERIAL_TXX9_CONSOLE) || (CONFIG_CONSOLE_POLL)
+/*
+ *	Wait for transmitter & holding register to empty
+ */
+static void wait_for_xmitr(struct uart_txx9_port *up)
+{
+	unsigned int tmout = 10000;
+
+	/* Wait up to 10ms for the character(s) to be sent. */
+	while (--tmout &&
+	       !(sio_in(up, TXX9_SICISR) & TXX9_SICISR_TXALS))
+		udelay(1);
+
+	/* Wait up to 1s for flow control if necessary */
+	if (up->port.flags & UPF_CONS_FLOW) {
+		tmout = 1000000;
+		while (--tmout &&
+		       (sio_in(up, TXX9_SICISR) & TXX9_SICISR_CTSS))
+			udelay(1);
+	}
+}
+#endif
+
+#ifdef CONFIG_CONSOLE_POLL
+/*
+ * Console polling routines for writing and reading from the uart while
+ * in an interrupt or debug context.
+ */
+
+static int serial_txx9_get_poll_char(struct uart_port *port)
+{
+	unsigned int ier;
+	unsigned char c;
+	struct uart_txx9_port *up = (struct uart_txx9_port *)port;
+
+	/*
+	 *	First save the IER then disable the interrupts
+	 */
+	ier = sio_in(up, TXX9_SIDICR);
+	sio_out(up, TXX9_SIDICR, 0);
+
+	while (sio_in(up, TXX9_SIDISR) & TXX9_SIDISR_UVALID)
+		;
+
+	c = sio_in(up, TXX9_SIRFIFO);
+
+	/*
+	 *	Finally, clear RX interrupt status
+	 *	and restore the IER
+	 */
+	sio_mask(up, TXX9_SIDISR, TXX9_SIDISR_RDIS);
+	sio_out(up, TXX9_SIDICR, ier);
+	return c;
+}
+
+
+static void serial_txx9_put_poll_char(struct uart_port *port, unsigned char c)
+{
+	unsigned int ier;
+	struct uart_txx9_port *up = (struct uart_txx9_port *)port;
+
+	/*
+	 *	First save the IER then disable the interrupts
+	 */
+	ier = sio_in(up, TXX9_SIDICR);
+	sio_out(up, TXX9_SIDICR, 0);
+
+	wait_for_xmitr(up);
+	/*
+	 *	Send the character out.
+	 *	If a LF, also do CR...
+	 */
+	sio_out(up, TXX9_SITFIFO, c);
+	if (c == 10) {
+		wait_for_xmitr(up);
+		sio_out(up, TXX9_SITFIFO, 13);
+	}
+
+	/*
+	 *	Finally, wait for transmitter to become empty
+	 *	and restore the IER
+	 */
+	wait_for_xmitr(up);
+	sio_out(up, TXX9_SIDICR, ier);
+}
+
+#endif /* CONFIG_CONSOLE_POLL */
+
 static int serial_txx9_startup(struct uart_port *port)
 {
 	struct uart_txx9_port *up = (struct uart_txx9_port *)port;
@@ -781,6 +869,10 @@ static struct uart_ops serial_txx9_pops = {
 	.release_port	= serial_txx9_release_port,
 	.request_port	= serial_txx9_request_port,
 	.config_port	= serial_txx9_config_port,
+#ifdef CONFIG_CONSOLE_POLL
+	.poll_get_char	= serial_txx9_get_poll_char,
+	.poll_put_char	= serial_txx9_put_poll_char,
+#endif
 };
 
 static struct uart_txx9_port serial_txx9_ports[UART_NR];
@@ -803,27 +895,6 @@ static void __init serial_txx9_register_ports(struct uart_driver *drv,
 
 #ifdef CONFIG_SERIAL_TXX9_CONSOLE
 
-/*
- *	Wait for transmitter & holding register to empty
- */
-static inline void wait_for_xmitr(struct uart_txx9_port *up)
-{
-	unsigned int tmout = 10000;
-
-	/* Wait up to 10ms for the character(s) to be sent. */
-	while (--tmout &&
-	       !(sio_in(up, TXX9_SICISR) & TXX9_SICISR_TXALS))
-		udelay(1);
-
-	/* Wait up to 1s for flow control if necessary */
-	if (up->port.flags & UPF_CONS_FLOW) {
-		tmout = 1000000;
-		while (--tmout &&
-		       (sio_in(up, TXX9_SICISR) & TXX9_SICISR_CTSS))
-			udelay(1);
-	}
-}
-
 static void serial_txx9_console_putchar(struct uart_port *port, int ch)
 {
 	struct uart_txx9_port *up = (struct uart_txx9_port *)port;
-- 
cgit v1.2.3-59-g8ed1b


From 2a13373cf84477460365c32842cda9a6374b845d Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Wed, 24 Jun 2009 18:34:43 +0100
Subject: jsm: clean up "serial: jsm: correctly support 4 8 port boards"

Remove unneeded casts.

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/jsm/jsm_tty.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/serial/jsm/jsm_tty.c b/drivers/serial/jsm/jsm_tty.c
index 107ce2e187b8..00f4577d2f7f 100644
--- a/drivers/serial/jsm/jsm_tty.c
+++ b/drivers/serial/jsm/jsm_tty.c
@@ -467,7 +467,7 @@ int __devinit jsm_uart_port_init(struct jsm_board *brd)
 			printk(KERN_INFO "jsm: linemap is full, added device failed\n");
 			continue;
 		} else
-			set_bit((int)line, linemap);
+			set_bit(line, linemap);
 		brd->channels[i]->uart_port.line = line;
 		if (uart_add_one_port (&jsm_uart_driver, &brd->channels[i]->uart_port))
 			printk(KERN_INFO "jsm: add device failed\n");
@@ -503,7 +503,7 @@ int jsm_remove_uart_port(struct jsm_board *brd)
 
 		ch = brd->channels[i];
 
-		clear_bit((int)(ch->uart_port.line), linemap);
+		clear_bit(ch->uart_port.line, linemap);
 		uart_remove_one_port(&jsm_uart_driver, &brd->channels[i]->uart_port);
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From ce89294c056805019d8369b3b74bb52ef51b4708 Mon Sep 17 00:00:00 2001
From: Paul Fulghum <paulkf@microgate.com>
Date: Wed, 24 Jun 2009 18:34:51 +0100
Subject: synclink_gt: fix transmit race and timeout

Fix race condition when adding transmit data to active DMA buffer ring
that can cause transmit stall.

Update transmit timeout when adding data to active DMA buffer ring.
Base transmit timeout on amount of buffered data instead of using fixed
value.

Signed-off-by: Paul Fulghum <paulkf@microgate.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/synclink_gt.c | 72 +++++++++++++++++++---------------------------
 1 file changed, 30 insertions(+), 42 deletions(-)

diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index 1386625fc4ca..a2e67e6df3a1 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -467,7 +467,6 @@ static unsigned int free_tbuf_count(struct slgt_info *info);
 static unsigned int tbuf_bytes(struct slgt_info *info);
 static void reset_tbufs(struct slgt_info *info);
 static void tdma_reset(struct slgt_info *info);
-static void tdma_start(struct slgt_info *info);
 static void tx_load(struct slgt_info *info, const char *buf, unsigned int count);
 
 static void get_signals(struct slgt_info *info);
@@ -795,6 +794,18 @@ static void set_termios(struct tty_struct *tty, struct ktermios *old_termios)
 	}
 }
 
+static void update_tx_timer(struct slgt_info *info)
+{
+	/*
+	 * use worst case speed of 1200bps to calculate transmit timeout
+	 * based on data in buffers (tbuf_bytes) and FIFO (128 bytes)
+	 */
+	if (info->params.mode == MGSL_MODE_HDLC) {
+		int timeout  = (tbuf_bytes(info) * 7) + 1000;
+		mod_timer(&info->tx_timer, jiffies + msecs_to_jiffies(timeout));
+	}
+}
+
 static int write(struct tty_struct *tty,
 		 const unsigned char *buf, int count)
 {
@@ -838,8 +849,18 @@ start:
 		spin_lock_irqsave(&info->lock,flags);
 		if (!info->tx_active)
 		 	tx_start(info);
-		else
-			tdma_start(info);
+		else if (!(rd_reg32(info, TDCSR) & BIT0)) {
+			/* transmit still active but transmit DMA stopped */
+			unsigned int i = info->tbuf_current;
+			if (!i)
+				i = info->tbuf_count;
+			i--;
+			/* if DMA buf unsent must try later after tx idle */
+			if (desc_count(info->tbufs[i]))
+				ret = 0;
+		}
+		if (ret > 0)
+			update_tx_timer(info);
 		spin_unlock_irqrestore(&info->lock,flags);
  	}
 
@@ -1502,10 +1523,9 @@ static int hdlcdev_xmit(struct sk_buff *skb, struct net_device *dev)
 	/* save start time for transmit timeout detection */
 	dev->trans_start = jiffies;
 
-	/* start hardware transmitter if necessary */
 	spin_lock_irqsave(&info->lock,flags);
-	if (!info->tx_active)
-	 	tx_start(info);
+	tx_start(info);
+	update_tx_timer(info);
 	spin_unlock_irqrestore(&info->lock,flags);
 
 	return 0;
@@ -3946,50 +3966,19 @@ static void tx_start(struct slgt_info *info)
 			slgt_irq_on(info, IRQ_TXUNDER + IRQ_TXIDLE);
 			/* clear tx idle and underrun status bits */
 			wr_reg16(info, SSR, (unsigned short)(IRQ_TXIDLE + IRQ_TXUNDER));
-			if (info->params.mode == MGSL_MODE_HDLC)
-				mod_timer(&info->tx_timer, jiffies +
-						msecs_to_jiffies(5000));
 		} else {
 			slgt_irq_off(info, IRQ_TXDATA);
 			slgt_irq_on(info, IRQ_TXIDLE);
 			/* clear tx idle status bit */
 			wr_reg16(info, SSR, IRQ_TXIDLE);
 		}
-		tdma_start(info);
+		/* set 1st descriptor address and start DMA */
+		wr_reg32(info, TDDAR, info->tbufs[info->tbuf_start].pdesc);
+		wr_reg32(info, TDCSR, BIT2 + BIT0);
 		info->tx_active = true;
 	}
 }
 
-/*
- * start transmit DMA if inactive and there are unsent buffers
- */
-static void tdma_start(struct slgt_info *info)
-{
-	unsigned int i;
-
-	if (rd_reg32(info, TDCSR) & BIT0)
-		return;
-
-	/* transmit DMA inactive, check for unsent buffers */
-	i = info->tbuf_start;
-	while (!desc_count(info->tbufs[i])) {
-		if (++i == info->tbuf_count)
-			i = 0;
-		if (i == info->tbuf_current)
-			return;
-	}
-	info->tbuf_start = i;
-
-	/* there are unsent buffers, start transmit DMA */
-
-	/* reset needed if previous error condition */
-	tdma_reset(info);
-
-	/* set 1st descriptor address */
-	wr_reg32(info, TDDAR, info->tbufs[info->tbuf_start].pdesc);
-	wr_reg32(info, TDCSR, BIT2 + BIT0); /* IRQ + DMA enable */
-}
-
 static void tx_stop(struct slgt_info *info)
 {
 	unsigned short val;
@@ -5004,8 +4993,7 @@ static void tx_timeout(unsigned long context)
 		info->icount.txtimeout++;
 	}
 	spin_lock_irqsave(&info->lock,flags);
-	info->tx_active = false;
-	info->tx_count = 0;
+	tx_stop(info);
 	spin_unlock_irqrestore(&info->lock,flags);
 
 #if SYNCLINK_GENERIC_HDLC
-- 
cgit v1.2.3-59-g8ed1b


From 24ed3abaa13a9499d7454a1ed9830bb53b689b94 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Wed, 24 Jun 2009 18:34:58 +0100
Subject: pci: use pci_ioremap_bar() in drivers/serial

Use the newly introduced pci_ioremap_bar() function in drivers/serial.
pci_ioremap_bar() just takes a pci device and a bar number, with the goal
of making it really hard to get wrong, while also having a central place
to stick sanity checks.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/8250_pci.c | 6 ++----
 drivers/serial/icom.c     | 3 +--
 2 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c
index e371a9c15341..a07015d646dd 100644
--- a/drivers/serial/8250_pci.c
+++ b/drivers/serial/8250_pci.c
@@ -398,8 +398,7 @@ static int sbs_init(struct pci_dev *dev)
 {
 	u8 __iomem *p;
 
-	p = ioremap_nocache(pci_resource_start(dev, 0),
-						pci_resource_len(dev, 0));
+	p = pci_ioremap_bar(dev, 0);
 
 	if (p == NULL)
 		return -ENOMEM;
@@ -423,8 +422,7 @@ static void __devexit sbs_exit(struct pci_dev *dev)
 {
 	u8 __iomem *p;
 
-	p = ioremap_nocache(pci_resource_start(dev, 0),
-					pci_resource_len(dev, 0));
+	p = pci_ioremap_bar(dev, 0);
 	/* FIXME: What if resource_len < OCT_REG_CR_OFF */
 	if (p != NULL)
 		writeb(0, p + OCT_REG_CR_OFF);
diff --git a/drivers/serial/icom.c b/drivers/serial/icom.c
index 9f2891c2c4a2..cd1b6a45bb82 100644
--- a/drivers/serial/icom.c
+++ b/drivers/serial/icom.c
@@ -1548,8 +1548,7 @@ static int __devinit icom_probe(struct pci_dev *dev,
 		goto probe_exit1;
 	}
 
-	 icom_adapter->base_addr = ioremap(icom_adapter->base_addr_pci,
-						pci_resource_len(dev, 0));
+	 icom_adapter->base_addr = pci_ioremap_bar(dev, 0);
 
 	if (!icom_adapter->base_addr)
 		goto probe_exit1;
-- 
cgit v1.2.3-59-g8ed1b


From 6af9a43d58f2ec455b752fb9534cf05c7e855dbe Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Wed, 24 Jun 2009 18:35:05 +0100
Subject: tty: fix tty_port_block_til_ready waiting

Since commit 3e3b5c087799e536871c8261b05bc28e4783c8da ("tty: use
prepare/finish_wait"), tty_port_block_til_ready() is using
prepare_to_wait()/finish_wait().  Those functions require that the
wait_queue_t be initialised with .func=autoremove_wake_function, via
DEFINE_WAIT().

But the conversion from DECLARE_WAITQUEUE() to DEFINE_WAIT() was not made,
so this code will oops in finish_wait().

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/tty_port.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/char/tty_port.c b/drivers/char/tty_port.c
index 62dadfc95e34..4e862a75f7ff 100644
--- a/drivers/char/tty_port.c
+++ b/drivers/char/tty_port.c
@@ -193,7 +193,7 @@ int tty_port_block_til_ready(struct tty_port *port,
 {
 	int do_clocal = 0, retval;
 	unsigned long flags;
-	DECLARE_WAITQUEUE(wait, current);
+	DEFINE_WAIT(wait);
 	int cd;
 
 	/* block if port is in the process of being closed */
-- 
cgit v1.2.3-59-g8ed1b


From 4d8d4d251df8eaaa3dae71c8cfa7fbf4510d967d Mon Sep 17 00:00:00 2001
From: Chuck Ebbert <cebbert@redhat.com>
Date: Wed, 24 Jun 2009 18:35:13 +0100
Subject: Remove low_latency flag setting from nozomi and mxser drivers

The kernel oopses if this flag is set.

[and neither driver should set it as they call tty_flip_buffer_push from IRQ
 paths so have always been buggy]

Signed-off-by: Chuck Ebbert <cebbert@redhat.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/mxser.c  | 2 --
 drivers/char/nozomi.c | 2 --
 2 files changed, 4 deletions(-)

diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index 9533f43a30bb..52d953eb30c3 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -1048,8 +1048,6 @@ static int mxser_open(struct tty_struct *tty, struct file *filp)
 	if (retval)
 		return retval;
 
-	/* unmark here for very high baud rate (ex. 921600 bps) used */
-	tty->low_latency = 1;
 	return 0;
 }
 
diff --git a/drivers/char/nozomi.c b/drivers/char/nozomi.c
index d6102b644b55..574f1c79b6e6 100644
--- a/drivers/char/nozomi.c
+++ b/drivers/char/nozomi.c
@@ -1591,8 +1591,6 @@ static int ntty_open(struct tty_struct *tty, struct file *file)
 
 	/* Enable interrupt downlink for channel */
 	if (port->port.count == 1) {
-		/* FIXME: is this needed now ? */
-		tty->low_latency = 1;
 		tty->driver_data = port;
 		tty_port_tty_set(&port->port, tty);
 		DBG1("open: %d", port->token_dl);
-- 
cgit v1.2.3-59-g8ed1b


From 9937ac0cc087b03d6d73f46a5d6b38c43626e60e Mon Sep 17 00:00:00 2001
From: Jesper Nilsson <jesper.nilsson@axis.com>
Date: Wed, 24 Jun 2009 09:33:19 +0200
Subject: MAINTAINERS: Change mailing list info for CRIS

Posting to the dev-etrax mailing list is only allowed for subscribers,
and the list is more geared toward user applications than kernel
developers.

Change to newly created mailing list for CRIS.

Signed-off-by: Jesper Nilsson <jesper.nilsson@axis.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 MAINTAINERS | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 05a4f306f1a0..fa2a16def17a 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1641,7 +1641,7 @@ P:	Mikael Starvik
 M:	starvik@axis.com
 P:	Jesper Nilsson
 M:	jesper.nilsson@axis.com
-L:	dev-etrax@axis.com
+L:	linux-cris-kernel@axis.com
 W:	http://developer.axis.com
 S:	Maintained
 F:	arch/cris/
-- 
cgit v1.2.3-59-g8ed1b


From 7433ab770327b471399f7b5baacad78e160b5393 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 25 Jun 2009 02:30:10 +0900
Subject: sh: Hook up page fault events for software perf counters.

This adds page fault instrumentation for the software performance
counters. Follows the x86 and powerpc changes.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/fault_32.c | 61 +++++++++++++++++++++++++++------------------------
 1 file changed, 32 insertions(+), 29 deletions(-)

diff --git a/arch/sh/mm/fault_32.c b/arch/sh/mm/fault_32.c
index cc8ddbdf3d7a..71925946f1e1 100644
--- a/arch/sh/mm/fault_32.c
+++ b/arch/sh/mm/fault_32.c
@@ -15,12 +15,28 @@
 #include <linux/mm.h>
 #include <linux/hardirq.h>
 #include <linux/kprobes.h>
-#include <linux/marker.h>
+#include <linux/perf_counter.h>
 #include <asm/io_trapped.h>
 #include <asm/system.h>
 #include <asm/mmu_context.h>
 #include <asm/tlbflush.h>
 
+static inline int notify_page_fault(struct pt_regs *regs, int trap)
+{
+	int ret = 0;
+
+#ifdef CONFIG_KPROBES
+	if (!user_mode(regs)) {
+		preempt_disable();
+		if (kprobe_running() && kprobe_fault_handler(regs, trap))
+			ret = 1;
+		preempt_enable();
+	}
+#endif
+
+	return ret;
+}
+
 /*
  * This routine handles page faults.  It determines the address,
  * and the problem, and then passes it off to one of the appropriate
@@ -87,13 +103,16 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
 		return;
 	}
 
+	mm = tsk->mm;
+
+	if (unlikely(notify_page_fault(regs, lookup_exception_vector())))
+		return;
+
 	/* Only enable interrupts if they were on before the fault */
-	if ((regs->sr & SR_IMASK) != SR_IMASK) {
-		trace_hardirqs_on();
+	if ((regs->sr & SR_IMASK) != SR_IMASK)
 		local_irq_enable();
-	}
 
-	mm = tsk->mm;
+	perf_swcounter_event(PERF_COUNT_SW_PAGE_FAULTS, 1, 0, regs, address);
 
 	/*
 	 * If we're in an interrupt or have no user
@@ -141,10 +160,15 @@ survive:
 			goto do_sigbus;
 		BUG();
 	}
-	if (fault & VM_FAULT_MAJOR)
+	if (fault & VM_FAULT_MAJOR) {
 		tsk->maj_flt++;
-	else
+		perf_swcounter_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1, 0,
+				     regs, address);
+	} else {
 		tsk->min_flt++;
+		perf_swcounter_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1, 0,
+				     regs, address);
+	}
 
 	up_read(&mm->mmap_sem);
 	return;
@@ -245,22 +269,6 @@ do_sigbus:
 		goto no_context;
 }
 
-static inline int notify_page_fault(struct pt_regs *regs, int trap)
-{
-	int ret = 0;
-
-#ifdef CONFIG_KPROBES
-	if (!user_mode(regs)) {
-		preempt_disable();
-		if (kprobe_running() && kprobe_fault_handler(regs, trap))
-			ret = 1;
-		preempt_enable();
-	}
-#endif
-
-	return ret;
-}
-
 /*
  * Called with interrupts disabled.
  */
@@ -273,12 +281,7 @@ asmlinkage int __kprobes __do_page_fault(struct pt_regs *regs,
 	pmd_t *pmd;
 	pte_t *pte;
 	pte_t entry;
-	int ret = 0;
-
-	if (notify_page_fault(regs, lookup_exception_vector()))
-		goto out;
-
-	ret = 1;
+	int ret = 1;
 
 	/*
 	 * We don't take page faults for P1, P2, and parts of P4, these
-- 
cgit v1.2.3-59-g8ed1b


From 0ca5921e791fb2011d4d6de787f6485b3900703d Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 19 Jun 2009 13:51:28 +0100
Subject: MIPS: bug.h Build fix - include <linux/compiler.h>.

In the past this file somehow used to be dragged in.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/include/asm/bug.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/mips/include/asm/bug.h b/arch/mips/include/asm/bug.h
index 08ea46863fe5..6cf29c26e873 100644
--- a/arch/mips/include/asm/bug.h
+++ b/arch/mips/include/asm/bug.h
@@ -1,6 +1,7 @@
 #ifndef __ASM_BUG_H
 #define __ASM_BUG_H
 
+#include <linux/compiler.h>
 #include <asm/sgidefs.h>
 
 #ifdef CONFIG_BUG
-- 
cgit v1.2.3-59-g8ed1b


From 631330f5847b3f8a7ea67d689e9f7c56833ccaa6 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 19 Jun 2009 14:05:26 +0100
Subject: MIPS: Build fix - include <linux/smp.h> into all smp_processor_id()
 users.

Some of the were relying into smp.h being dragged in by another header
which of course is fragile.  <asm/cpu-info.h> uses smp_processor_id()
only in macros and including smp.h there leads to an include loop, so
don't change cpu-info.h.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/cavium-octeon/octeon-irq.c | 2 +-
 arch/mips/cavium-octeon/setup.c      | 1 +
 arch/mips/include/asm/bugs.h         | 1 +
 arch/mips/include/asm/irq.h          | 1 +
 arch/mips/include/asm/mmu_context.h  | 1 +
 arch/mips/include/asm/smp.h          | 1 +
 arch/mips/include/asm/sn/addrs.h     | 1 +
 arch/mips/jazz/irq.c                 | 1 +
 arch/mips/kernel/cevt-bcm1480.c      | 1 +
 arch/mips/kernel/cevt-r4k.c          | 1 +
 arch/mips/kernel/cevt-sb1250.c       | 1 +
 arch/mips/kernel/cevt-smtc.c         | 1 +
 arch/mips/kernel/cpu-probe.c         | 1 +
 arch/mips/kernel/i8253.c             | 1 +
 arch/mips/kernel/irq-gic.c           | 1 +
 arch/mips/kernel/kgdb.c              | 1 +
 arch/mips/kernel/smp-cmp.c           | 1 +
 arch/mips/kernel/smp.c               | 1 +
 arch/mips/kernel/smtc.c              | 1 +
 arch/mips/mm/c-octeon.c              | 1 +
 arch/mips/mm/c-r3k.c                 | 1 +
 arch/mips/mm/c-r4k.c                 | 1 +
 arch/mips/mm/c-tx39.c                | 1 +
 arch/mips/mm/highmem.c               | 1 +
 arch/mips/mm/init.c                  | 1 +
 arch/mips/mm/page.c                  | 1 +
 arch/mips/mm/tlb-r3k.c               | 1 +
 arch/mips/mm/tlb-r4k.c               | 1 +
 arch/mips/mm/tlb-r8k.c               | 1 +
 arch/mips/mm/tlbex.c                 | 1 +
 arch/mips/mti-malta/malta-int.c      | 1 +
 arch/mips/pci/pci-ip27.c             | 1 +
 arch/mips/pmc-sierra/yosemite/smp.c  | 1 +
 arch/mips/sgi-ip27/ip27-init.c       | 1 +
 arch/mips/sgi-ip27/ip27-irq.c        | 1 +
 arch/mips/sgi-ip27/ip27-timer.c      | 1 +
 arch/mips/sgi-ip27/ip27-xtalk.c      | 1 +
 arch/mips/sibyte/bcm1480/irq.c       | 1 +
 arch/mips/sni/time.c                 | 1 +
 39 files changed, 39 insertions(+), 1 deletion(-)

diff --git a/arch/mips/cavium-octeon/octeon-irq.c b/arch/mips/cavium-octeon/octeon-irq.c
index 8dfa009e0070..3090ee37e109 100644
--- a/arch/mips/cavium-octeon/octeon-irq.c
+++ b/arch/mips/cavium-octeon/octeon-irq.c
@@ -7,7 +7,7 @@
  */
 #include <linux/irq.h>
 #include <linux/interrupt.h>
-#include <linux/hardirq.h>
+#include <linux/smp.h>
 
 #include <asm/octeon/octeon.h>
 #include <asm/octeon/cvmx-pexp-defs.h>
diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c
index 5f4e49ba4713..da559249cc2f 100644
--- a/arch/mips/cavium-octeon/setup.c
+++ b/arch/mips/cavium-octeon/setup.c
@@ -13,6 +13,7 @@
 #include <linux/io.h>
 #include <linux/irq.h>
 #include <linux/serial.h>
+#include <linux/smp.h>
 #include <linux/types.h>
 #include <linux/string.h>	/* for memset */
 #include <linux/tty.h>
diff --git a/arch/mips/include/asm/bugs.h b/arch/mips/include/asm/bugs.h
index 9dc10df32078..b160a706795d 100644
--- a/arch/mips/include/asm/bugs.h
+++ b/arch/mips/include/asm/bugs.h
@@ -11,6 +11,7 @@
 
 #include <linux/bug.h>
 #include <linux/delay.h>
+#include <linux/smp.h>
 
 #include <asm/cpu.h>
 #include <asm/cpu-info.h>
diff --git a/arch/mips/include/asm/irq.h b/arch/mips/include/asm/irq.h
index 4f1eed107b08..09b08d05ff72 100644
--- a/arch/mips/include/asm/irq.h
+++ b/arch/mips/include/asm/irq.h
@@ -10,6 +10,7 @@
 #define _ASM_IRQ_H
 
 #include <linux/linkage.h>
+#include <linux/smp.h>
 
 #include <asm/mipsmtregs.h>
 
diff --git a/arch/mips/include/asm/mmu_context.h b/arch/mips/include/asm/mmu_context.h
index d7f3eb03ad12..d3bea88d8744 100644
--- a/arch/mips/include/asm/mmu_context.h
+++ b/arch/mips/include/asm/mmu_context.h
@@ -13,6 +13,7 @@
 
 #include <linux/errno.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/slab.h>
 #include <asm/cacheflush.h>
 #include <asm/tlbflush.h>
diff --git a/arch/mips/include/asm/smp.h b/arch/mips/include/asm/smp.h
index 40e5ef1d4d26..2f83fa8631db 100644
--- a/arch/mips/include/asm/smp.h
+++ b/arch/mips/include/asm/smp.h
@@ -13,6 +13,7 @@
 
 #include <linux/bitops.h>
 #include <linux/linkage.h>
+#include <linux/smp.h>
 #include <linux/threads.h>
 #include <linux/cpumask.h>
 
diff --git a/arch/mips/include/asm/sn/addrs.h b/arch/mips/include/asm/sn/addrs.h
index 3a56d90abfa6..2367b56dcdef 100644
--- a/arch/mips/include/asm/sn/addrs.h
+++ b/arch/mips/include/asm/sn/addrs.h
@@ -11,6 +11,7 @@
 
 
 #ifndef __ASSEMBLY__
+#include <linux/smp.h>
 #include <linux/types.h>
 #endif /* !__ASSEMBLY__ */
 
diff --git a/arch/mips/jazz/irq.c b/arch/mips/jazz/irq.c
index d9b6a5b5399d..7fd170d007e7 100644
--- a/arch/mips/jazz/irq.c
+++ b/arch/mips/jazz/irq.c
@@ -10,6 +10,7 @@
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/kernel.h>
+#include <linux/smp.h>
 #include <linux/spinlock.h>
 
 #include <asm/irq_cpu.h>
diff --git a/arch/mips/kernel/cevt-bcm1480.c b/arch/mips/kernel/cevt-bcm1480.c
index a5182a207696..e02f79b1eb51 100644
--- a/arch/mips/kernel/cevt-bcm1480.c
+++ b/arch/mips/kernel/cevt-bcm1480.c
@@ -18,6 +18,7 @@
 #include <linux/clockchips.h>
 #include <linux/interrupt.h>
 #include <linux/percpu.h>
+#include <linux/smp.h>
 
 #include <asm/addrspace.h>
 #include <asm/io.h>
diff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-r4k.c
index 0015e442572b..2652362ce047 100644
--- a/arch/mips/kernel/cevt-r4k.c
+++ b/arch/mips/kernel/cevt-r4k.c
@@ -9,6 +9,7 @@
 #include <linux/clockchips.h>
 #include <linux/interrupt.h>
 #include <linux/percpu.h>
+#include <linux/smp.h>
 
 #include <asm/smtc_ipi.h>
 #include <asm/time.h>
diff --git a/arch/mips/kernel/cevt-sb1250.c b/arch/mips/kernel/cevt-sb1250.c
index 340f53e5c6b1..ac5903d1b20e 100644
--- a/arch/mips/kernel/cevt-sb1250.c
+++ b/arch/mips/kernel/cevt-sb1250.c
@@ -18,6 +18,7 @@
 #include <linux/clockchips.h>
 #include <linux/interrupt.h>
 #include <linux/percpu.h>
+#include <linux/smp.h>
 
 #include <asm/addrspace.h>
 #include <asm/io.h>
diff --git a/arch/mips/kernel/cevt-smtc.c b/arch/mips/kernel/cevt-smtc.c
index df6f5bc60572..98bd7de75778 100644
--- a/arch/mips/kernel/cevt-smtc.c
+++ b/arch/mips/kernel/cevt-smtc.c
@@ -10,6 +10,7 @@
 #include <linux/clockchips.h>
 #include <linux/interrupt.h>
 #include <linux/percpu.h>
+#include <linux/smp.h>
 
 #include <asm/smtc_ipi.h>
 #include <asm/time.h>
diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c
index b13b8eb30596..1abe9905c9c1 100644
--- a/arch/mips/kernel/cpu-probe.c
+++ b/arch/mips/kernel/cpu-probe.c
@@ -14,6 +14,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/ptrace.h>
+#include <linux/smp.h>
 #include <linux/stddef.h>
 
 #include <asm/bugs.h>
diff --git a/arch/mips/kernel/i8253.c b/arch/mips/kernel/i8253.c
index ed20e7fe65e3..f7d8d5d0ddbf 100644
--- a/arch/mips/kernel/i8253.c
+++ b/arch/mips/kernel/i8253.c
@@ -7,6 +7,7 @@
 #include <linux/interrupt.h>
 #include <linux/jiffies.h>
 #include <linux/module.h>
+#include <linux/smp.h>
 #include <linux/spinlock.h>
 
 #include <asm/delay.h>
diff --git a/arch/mips/kernel/irq-gic.c b/arch/mips/kernel/irq-gic.c
index 3f43c2e3aa5a..39000f103f2c 100644
--- a/arch/mips/kernel/irq-gic.c
+++ b/arch/mips/kernel/irq-gic.c
@@ -2,6 +2,7 @@
 
 #include <linux/bitmap.h>
 #include <linux/init.h>
+#include <linux/smp.h>
 
 #include <asm/io.h>
 #include <asm/gic.h>
diff --git a/arch/mips/kernel/kgdb.c b/arch/mips/kernel/kgdb.c
index 6e152c80cd4a..50c9bb880667 100644
--- a/arch/mips/kernel/kgdb.c
+++ b/arch/mips/kernel/kgdb.c
@@ -26,6 +26,7 @@
 #include <linux/kgdb.h>
 #include <linux/kdebug.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <asm/inst.h>
 #include <asm/fpu.h>
 #include <asm/cacheflush.h>
diff --git a/arch/mips/kernel/smp-cmp.c b/arch/mips/kernel/smp-cmp.c
index f27beca4b26d..653be061b9ec 100644
--- a/arch/mips/kernel/smp-cmp.c
+++ b/arch/mips/kernel/smp-cmp.c
@@ -20,6 +20,7 @@
 
 #include <linux/kernel.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/cpumask.h>
 #include <linux/interrupt.h>
 #include <linux/compiler.h>
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index c937506a03aa..58f4679bbd43 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -22,6 +22,7 @@
 #include <linux/delay.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
+#include <linux/smp.h>
 #include <linux/spinlock.h>
 #include <linux/threads.h>
 #include <linux/module.h>
diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c
index 37d51cd124e9..8a0626cbb108 100644
--- a/arch/mips/kernel/smtc.c
+++ b/arch/mips/kernel/smtc.c
@@ -20,6 +20,7 @@
 #include <linux/clockchips.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/cpumask.h>
 #include <linux/interrupt.h>
 #include <linux/kernel_stat.h>
diff --git a/arch/mips/mm/c-octeon.c b/arch/mips/mm/c-octeon.c
index 44d01a0a8490..b165cdcb2818 100644
--- a/arch/mips/mm/c-octeon.c
+++ b/arch/mips/mm/c-octeon.c
@@ -8,6 +8,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/mm.h>
 #include <linux/bitops.h>
 #include <linux/cpu.h>
diff --git a/arch/mips/mm/c-r3k.c b/arch/mips/mm/c-r3k.c
index 5500c20c79ae..54e5f7b9f440 100644
--- a/arch/mips/mm/c-r3k.c
+++ b/arch/mips/mm/c-r3k.c
@@ -12,6 +12,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/mm.h>
 
 #include <asm/page.h>
diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c
index 71fe4cb778cd..6721ee2b1e8b 100644
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -13,6 +13,7 @@
 #include <linux/kernel.h>
 #include <linux/linkage.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/mm.h>
 #include <linux/module.h>
 #include <linux/bitops.h>
diff --git a/arch/mips/mm/c-tx39.c b/arch/mips/mm/c-tx39.c
index f7c8f9ce39c1..6515b4418714 100644
--- a/arch/mips/mm/c-tx39.c
+++ b/arch/mips/mm/c-tx39.c
@@ -11,6 +11,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/mm.h>
 
 #include <asm/cacheops.h>
diff --git a/arch/mips/mm/highmem.c b/arch/mips/mm/highmem.c
index 2b1309b2580a..e274fda329f4 100644
--- a/arch/mips/mm/highmem.c
+++ b/arch/mips/mm/highmem.c
@@ -1,5 +1,6 @@
 #include <linux/module.h>
 #include <linux/highmem.h>
+#include <linux/smp.h>
 #include <asm/fixmap.h>
 #include <asm/tlbflush.h>
 
diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c
index c5511294a9ee..0e820508ff23 100644
--- a/arch/mips/mm/init.c
+++ b/arch/mips/mm/init.c
@@ -13,6 +13,7 @@
 #include <linux/module.h>
 #include <linux/signal.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/kernel.h>
 #include <linux/errno.h>
 #include <linux/string.h>
diff --git a/arch/mips/mm/page.c b/arch/mips/mm/page.c
index 48060c635acd..f5c73754d664 100644
--- a/arch/mips/mm/page.c
+++ b/arch/mips/mm/page.c
@@ -10,6 +10,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/mm.h>
 #include <linux/module.h>
 #include <linux/proc_fs.h>
diff --git a/arch/mips/mm/tlb-r3k.c b/arch/mips/mm/tlb-r3k.c
index 1c0048a6f5cf..0f5ab236ab69 100644
--- a/arch/mips/mm/tlb-r3k.c
+++ b/arch/mips/mm/tlb-r3k.c
@@ -13,6 +13,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/mm.h>
 
 #include <asm/page.h>
diff --git a/arch/mips/mm/tlb-r4k.c b/arch/mips/mm/tlb-r4k.c
index f60fe513eb60..cee502caf398 100644
--- a/arch/mips/mm/tlb-r4k.c
+++ b/arch/mips/mm/tlb-r4k.c
@@ -10,6 +10,7 @@
  */
 #include <linux/init.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/mm.h>
 #include <linux/hugetlb.h>
 
diff --git a/arch/mips/mm/tlb-r8k.c b/arch/mips/mm/tlb-r8k.c
index 4ec95cc2df2f..2b82f23df1a1 100644
--- a/arch/mips/mm/tlb-r8k.c
+++ b/arch/mips/mm/tlb-r8k.c
@@ -10,6 +10,7 @@
  */
 #include <linux/init.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/mm.h>
 
 #include <asm/cpu.h>
diff --git a/arch/mips/mm/tlbex.c b/arch/mips/mm/tlbex.c
index 8f606ead826e..9a17bf8395df 100644
--- a/arch/mips/mm/tlbex.c
+++ b/arch/mips/mm/tlbex.c
@@ -23,6 +23,7 @@
 #include <linux/bug.h>
 #include <linux/kernel.h>
 #include <linux/types.h>
+#include <linux/smp.h>
 #include <linux/string.h>
 #include <linux/init.h>
 
diff --git a/arch/mips/mti-malta/malta-int.c b/arch/mips/mti-malta/malta-int.c
index ea176113fea9..b4eaf137e4a7 100644
--- a/arch/mips/mti-malta/malta-int.c
+++ b/arch/mips/mti-malta/malta-int.c
@@ -24,6 +24,7 @@
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/slab.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
diff --git a/arch/mips/pci/pci-ip27.c b/arch/mips/pci/pci-ip27.c
index dda6f2058665..a0e726eb039a 100644
--- a/arch/mips/pci/pci-ip27.c
+++ b/arch/mips/pci/pci-ip27.c
@@ -10,6 +10,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/pci.h>
+#include <linux/smp.h>
 #include <asm/sn/arch.h>
 #include <asm/pci/bridge.h>
 #include <asm/paccess.h>
diff --git a/arch/mips/pmc-sierra/yosemite/smp.c b/arch/mips/pmc-sierra/yosemite/smp.c
index f78c29b68d77..8ace27716232 100644
--- a/arch/mips/pmc-sierra/yosemite/smp.c
+++ b/arch/mips/pmc-sierra/yosemite/smp.c
@@ -1,5 +1,6 @@
 #include <linux/linkage.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 
 #include <asm/pmon.h>
 #include <asm/titan_dep.h>
diff --git a/arch/mips/sgi-ip27/ip27-init.c b/arch/mips/sgi-ip27/ip27-init.c
index 4a500e8cd3cc..51d3a4f2d7e1 100644
--- a/arch/mips/sgi-ip27/ip27-init.c
+++ b/arch/mips/sgi-ip27/ip27-init.c
@@ -9,6 +9,7 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/sched.h>
+#include <linux/smp.h>
 #include <linux/mm.h>
 #include <linux/module.h>
 #include <linux/cpumask.h>
diff --git a/arch/mips/sgi-ip27/ip27-irq.c b/arch/mips/sgi-ip27/ip27-irq.c
index 1bb692a3b319..c1c8e40d65d6 100644
--- a/arch/mips/sgi-ip27/ip27-irq.c
+++ b/arch/mips/sgi-ip27/ip27-irq.c
@@ -18,6 +18,7 @@
 #include <linux/ioport.h>
 #include <linux/timex.h>
 #include <linux/slab.h>
+#include <linux/smp.h>
 #include <linux/random.h>
 #include <linux/kernel.h>
 #include <linux/kernel_stat.h>
diff --git a/arch/mips/sgi-ip27/ip27-timer.c b/arch/mips/sgi-ip27/ip27-timer.c
index f10a7cd64f7e..6d0e59ffba2e 100644
--- a/arch/mips/sgi-ip27/ip27-timer.c
+++ b/arch/mips/sgi-ip27/ip27-timer.c
@@ -10,6 +10,7 @@
 #include <linux/interrupt.h>
 #include <linux/kernel_stat.h>
 #include <linux/param.h>
+#include <linux/smp.h>
 #include <linux/time.h>
 #include <linux/timex.h>
 #include <linux/mm.h>
diff --git a/arch/mips/sgi-ip27/ip27-xtalk.c b/arch/mips/sgi-ip27/ip27-xtalk.c
index 6ae64e8dfc40..5e871e75a8d9 100644
--- a/arch/mips/sgi-ip27/ip27-xtalk.c
+++ b/arch/mips/sgi-ip27/ip27-xtalk.c
@@ -9,6 +9,7 @@
 
 #include <linux/init.h>
 #include <linux/kernel.h>
+#include <linux/smp.h>
 #include <asm/sn/types.h>
 #include <asm/sn/klconfig.h>
 #include <asm/sn/hub.h>
diff --git a/arch/mips/sibyte/bcm1480/irq.c b/arch/mips/sibyte/bcm1480/irq.c
index 690de06bde90..ba59839a021e 100644
--- a/arch/mips/sibyte/bcm1480/irq.c
+++ b/arch/mips/sibyte/bcm1480/irq.c
@@ -19,6 +19,7 @@
 #include <linux/init.h>
 #include <linux/linkage.h>
 #include <linux/interrupt.h>
+#include <linux/smp.h>
 #include <linux/spinlock.h>
 #include <linux/mm.h>
 #include <linux/slab.h>
diff --git a/arch/mips/sni/time.c b/arch/mips/sni/time.c
index 69f5f88711cc..0d9ec1a5c24a 100644
--- a/arch/mips/sni/time.c
+++ b/arch/mips/sni/time.c
@@ -1,5 +1,6 @@
 #include <linux/types.h>
 #include <linux/interrupt.h>
+#include <linux/smp.h>
 #include <linux/time.h>
 #include <linux/clockchips.h>
 
-- 
cgit v1.2.3-59-g8ed1b


From 44eeab67416711db9b84610ef18c99a60415dff8 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 19 Jun 2009 15:01:44 +0100
Subject: MIPS: Hibernation: Remove SMP TLB and cacheflushing code.

We can't perform any flushes on SMP from swsusp_arch_resume because
interrupts are disabled.  A cross-CPU flush is unnecessary anyway
because all but the local CPU have already been disabled.  A local
flush is not needed either because we didn't change any mappings.  So
just delete the code.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/power/hibernate.S | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/arch/mips/power/hibernate.S b/arch/mips/power/hibernate.S
index 486bd3fd01a1..4b8174b382d7 100644
--- a/arch/mips/power/hibernate.S
+++ b/arch/mips/power/hibernate.S
@@ -43,15 +43,6 @@ LEAF(swsusp_arch_resume)
 	bne t1, t3, 1b
 	PTR_L t0, PBE_NEXT(t0)
 	bnez t0, 0b
-	/* flush caches to make sure context is in memory */
-	PTR_L t0, __flush_cache_all
-	jalr t0
-	/* flush tlb entries */
-#ifdef CONFIG_SMP
-	jal	flush_tlb_all
-#else
-	jal	local_flush_tlb_all
-#endif
 	PTR_LA t0, saved_regs
 	PTR_L ra, PT_R31(t0)
 	PTR_L sp, PT_R29(t0)
-- 
cgit v1.2.3-59-g8ed1b


From ab7f6f3010a6c5ae147541168705a446cee511e7 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Mon, 22 Jun 2009 15:48:27 +0100
Subject: MIPS: MIPSsim: Fix build error if MSC01E_INT_BASE is undefined.

This fixes kernel.org bugzilla 13595, see
http://bugzilla.kernel.org/show_bug.cgi?id=13595

Reported-by: dvice_null@yahoo.com

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mipssim/sim_time.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/mips/mipssim/sim_time.c b/arch/mips/mipssim/sim_time.c
index 881ecbc1fa23..0cea932f1241 100644
--- a/arch/mips/mipssim/sim_time.c
+++ b/arch/mips/mipssim/sim_time.c
@@ -91,6 +91,7 @@ unsigned __cpuinit get_c0_compare_int(void)
 		mips_cpu_timer_irq = MSC01E_INT_BASE + MSC01E_INT_CPUCTR;
 	} else {
 #endif
+	       {
 		if (cpu_has_vint)
 			set_vi_handler(cp0_compare_irq, mips_timer_dispatch);
 		mips_cpu_timer_irq = MIPS_CPU_IRQ_BASE + cp0_compare_irq;
-- 
cgit v1.2.3-59-g8ed1b


From 2e25406fb878e2313a9d8e302ed7ff3c2831198f Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Mon, 22 Jun 2009 16:17:52 +0100
Subject: MIPS: Sibyte: Fix build error if CONFIG_SERIAL_SB1250_DUART is
 undefined.

This fixes kernel.org bugzilla 13596, see
http://bugzilla.kernel.org/show_bug.cgi?id=13596

Reported-by: dvice_null@yahoo.com

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/sibyte/common/cfe_console.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/arch/mips/sibyte/common/cfe_console.c b/arch/mips/sibyte/common/cfe_console.c
index 81e3d54376e9..1ad2da103fe9 100644
--- a/arch/mips/sibyte/common/cfe_console.c
+++ b/arch/mips/sibyte/common/cfe_console.c
@@ -51,12 +51,13 @@ static int cfe_console_setup(struct console *cons, char *str)
 			setleds("u0cn");
 		} else if (!strcmp(consdev, "uart1")) {
 			setleds("u1cn");
+		} else
 #endif
 #ifdef CONFIG_VGA_CONSOLE
-		} else if (!strcmp(consdev, "pcconsole0")) {
-			setleds("pccn");
-#endif
+		       if (!strcmp(consdev, "pcconsole0")) {
+				setleds("pccn");
 		} else
+#endif
 			return -ENODEV;
 	}
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 4ac4aa5cc3b00cc558575065ae71043e92d1a69a Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Wed, 17 Jun 2009 13:08:31 -0700
Subject: DMA: txx9dmac: use dma_unmap_single if
 DMA_COMPL_{SRC,DEST}_UNMAP_SINGLE set

This patch does not change actual behaviour since dma_unmap_page is just
an alias of dma_unmap_single on MIPS.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Cc: Ralf Baechle <ralf@linux-mips.org>
Acked-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 drivers/dma/txx9dmac.c | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/drivers/dma/txx9dmac.c b/drivers/dma/txx9dmac.c
index 9aa9ea9822c8..88dab52926f4 100644
--- a/drivers/dma/txx9dmac.c
+++ b/drivers/dma/txx9dmac.c
@@ -432,23 +432,27 @@ txx9dmac_descriptor_complete(struct txx9dmac_chan *dc,
 	list_splice_init(&txd->tx_list, &dc->free_list);
 	list_move(&desc->desc_node, &dc->free_list);
 
-	/*
-	 * We use dma_unmap_page() regardless of how the buffers were
-	 * mapped before they were submitted...
-	 */
 	if (!ds) {
 		dma_addr_t dmaaddr;
 		if (!(txd->flags & DMA_COMPL_SKIP_DEST_UNMAP)) {
 			dmaaddr = is_dmac64(dc) ?
 				desc->hwdesc.DAR : desc->hwdesc32.DAR;
-			dma_unmap_page(chan2parent(&dc->chan), dmaaddr,
-				       desc->len, DMA_FROM_DEVICE);
+			if (txd->flags & DMA_COMPL_DEST_UNMAP_SINGLE)
+				dma_unmap_single(chan2parent(&dc->chan),
+					dmaaddr, desc->len, DMA_FROM_DEVICE);
+			else
+				dma_unmap_page(chan2parent(&dc->chan),
+					dmaaddr, desc->len, DMA_FROM_DEVICE);
 		}
 		if (!(txd->flags & DMA_COMPL_SKIP_SRC_UNMAP)) {
 			dmaaddr = is_dmac64(dc) ?
 				desc->hwdesc.SAR : desc->hwdesc32.SAR;
-			dma_unmap_page(chan2parent(&dc->chan), dmaaddr,
-				       desc->len, DMA_TO_DEVICE);
+			if (txd->flags & DMA_COMPL_SRC_UNMAP_SINGLE)
+				dma_unmap_single(chan2parent(&dc->chan),
+					dmaaddr, desc->len, DMA_TO_DEVICE);
+			else
+				dma_unmap_page(chan2parent(&dc->chan),
+					dmaaddr, desc->len, DMA_TO_DEVICE);
 		}
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 1b2bc75c1bde6581d2694cb3ed7fb06b69685008 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 23 Jun 2009 10:00:31 +0100
Subject: MIPS: Add arch generic CPU hotplug

Each platform has to add support for CPU hotplugging itself by providing
suitable definitions for the cpu_disable and cpu_die of the smp_ops
methods and setting SYS_SUPPORTS_HOTPLUG_CPU.  A platform should only set
SYS_SUPPORTS_HOTPLUG_CPU once all it's smp_ops definitions have the
necessary changes.  This patch contains the changes to the dummy smp_ops
definition for uni-processor systems.

Parts of the code contributed by Cavium Inc.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig               | 11 ++++++++++-
 arch/mips/include/asm/smp-ops.h |  4 ++++
 arch/mips/include/asm/smp.h     | 19 +++++++++++++++++++
 arch/mips/kernel/process.c      | 13 ++++++++++++-
 arch/mips/kernel/smp-up.c       | 16 ++++++++++++++++
 arch/mips/kernel/smp.c          | 17 +++++++++++++----
 arch/mips/kernel/topology.c     |  5 ++++-
 7 files changed, 78 insertions(+), 7 deletions(-)

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index b29f0280d712..eb7e8d795c6a 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -784,8 +784,17 @@ config SYS_HAS_EARLY_PRINTK
 	bool
 
 config HOTPLUG_CPU
+	bool "Support for hot-pluggable CPUs"
+	depends on SMP && HOTPLUG && SYS_SUPPORTS_HOTPLUG_CPU
+	help
+	  Say Y here to allow turning CPUs off and on. CPUs can be
+	  controlled through /sys/devices/system/cpu.
+	  (Note: power management support will enable this option
+	    automatically on SMP systems. )
+	  Say N if you want to disable CPU hotplug.
+
+config SYS_SUPPORTS_HOTPLUG_CPU
 	bool
-	default n
 
 config I8259
 	bool
diff --git a/arch/mips/include/asm/smp-ops.h b/arch/mips/include/asm/smp-ops.h
index 64ffc0290b84..fd545547b8aa 100644
--- a/arch/mips/include/asm/smp-ops.h
+++ b/arch/mips/include/asm/smp-ops.h
@@ -26,6 +26,10 @@ struct plat_smp_ops {
 	void (*boot_secondary)(int cpu, struct task_struct *idle);
 	void (*smp_setup)(void);
 	void (*prepare_cpus)(unsigned int max_cpus);
+#ifdef CONFIG_HOTPLUG_CPU
+	int (*cpu_disable)(void);
+	void (*cpu_die)(unsigned int cpu);
+#endif
 };
 
 extern void register_smp_ops(struct plat_smp_ops *ops);
diff --git a/arch/mips/include/asm/smp.h b/arch/mips/include/asm/smp.h
index 2f83fa8631db..01f813dc3888 100644
--- a/arch/mips/include/asm/smp.h
+++ b/arch/mips/include/asm/smp.h
@@ -41,6 +41,7 @@ extern int __cpu_logical_map[NR_CPUS];
 /* Octeon - Tell another core to flush its icache */
 #define SMP_ICACHE_FLUSH	0x4
 
+extern cpumask_t cpu_callin_map;
 
 extern void asmlinkage smp_bootstrap(void);
 
@@ -56,6 +57,24 @@ static inline void smp_send_reschedule(int cpu)
 	mp_ops->send_ipi_single(cpu, SMP_RESCHEDULE_YOURSELF);
 }
 
+#ifdef CONFIG_HOTPLUG_CPU
+static inline int __cpu_disable(void)
+{
+	extern struct plat_smp_ops *mp_ops;     /* private */
+
+	return mp_ops->cpu_disable();
+}
+
+static inline void __cpu_die(unsigned int cpu)
+{
+	extern struct plat_smp_ops *mp_ops;     /* private */
+
+	mp_ops->cpu_die(cpu);
+}
+
+extern void play_dead(void);
+#endif
+
 extern asmlinkage void smp_call_function_interrupt(void);
 
 extern void arch_send_call_function_single_ipi(int cpu);
diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index 1eaaa450e20c..c09d681b7181 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -50,10 +50,15 @@
  */
 void __noreturn cpu_idle(void)
 {
+	int cpu;
+
+	/* CPU is going idle. */
+	cpu = smp_processor_id();
+
 	/* endless idle loop with no priority at all */
 	while (1) {
 		tick_nohz_stop_sched_tick(1);
-		while (!need_resched()) {
+		while (!need_resched() && cpu_online(cpu)) {
 #ifdef CONFIG_MIPS_MT_SMTC
 			extern void smtc_idle_loop_hook(void);
 
@@ -62,6 +67,12 @@ void __noreturn cpu_idle(void)
 			if (cpu_wait)
 				(*cpu_wait)();
 		}
+#ifdef CONFIG_HOTPLUG_CPU
+		if (!cpu_online(cpu) && !cpu_isset(cpu, cpu_callin_map) &&
+		    (system_state == SYSTEM_RUNNING ||
+		     system_state == SYSTEM_BOOTING))
+			play_dead();
+#endif
 		tick_nohz_restart_sched_tick();
 		preempt_enable_no_resched();
 		schedule();
diff --git a/arch/mips/kernel/smp-up.c b/arch/mips/kernel/smp-up.c
index 878e3733bbb2..2508d55d68fd 100644
--- a/arch/mips/kernel/smp-up.c
+++ b/arch/mips/kernel/smp-up.c
@@ -55,6 +55,18 @@ static void __init up_prepare_cpus(unsigned int max_cpus)
 {
 }
 
+#ifdef CONFIG_HOTPLUG_CPU
+static int up_cpu_disable(void)
+{
+	return -ENOSYS;
+}
+
+static void up_cpu_die(unsigned int cpu)
+{
+	BUG();
+}
+#endif
+
 struct plat_smp_ops up_smp_ops = {
 	.send_ipi_single	= up_send_ipi_single,
 	.send_ipi_mask		= up_send_ipi_mask,
@@ -64,4 +76,8 @@ struct plat_smp_ops up_smp_ops = {
 	.boot_secondary		= up_boot_secondary,
 	.smp_setup		= up_smp_setup,
 	.prepare_cpus		= up_prepare_cpus,
+#ifdef CONFIG_HOTPLUG_CPU
+	.cpu_disable		= up_cpu_disable,
+	.cpu_die		= up_cpu_die,
+#endif
 };
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index 58f4679bbd43..bc7d9b05e2f4 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -45,7 +45,7 @@
 #include <asm/mipsmtregs.h>
 #endif /* CONFIG_MIPS_MT_SMTC */
 
-static volatile cpumask_t cpu_callin_map;	/* Bitmask of started secondaries */
+volatile cpumask_t cpu_callin_map;	/* Bitmask of started secondaries */
 int __cpu_number_map[NR_CPUS];		/* Map physical to logical */
 int __cpu_logical_map[NR_CPUS];		/* Map logical to physical */
 
@@ -201,6 +201,8 @@ void __devinit smp_prepare_boot_cpu(void)
  * and keep control until "cpu_online(cpu)" is set.  Note: cpu is
  * physical, not logical.
  */
+static struct task_struct *cpu_idle_thread[NR_CPUS];
+
 int __cpuinit __cpu_up(unsigned int cpu)
 {
 	struct task_struct *idle;
@@ -210,9 +212,16 @@ int __cpuinit __cpu_up(unsigned int cpu)
 	 * The following code is purely to make sure
 	 * Linux can schedule processes on this slave.
 	 */
-	idle = fork_idle(cpu);
-	if (IS_ERR(idle))
-		panic(KERN_ERR "Fork failed for CPU %d", cpu);
+	if (!cpu_idle_thread[cpu]) {
+		idle = fork_idle(cpu);
+		cpu_idle_thread[cpu] = idle;
+
+		if (IS_ERR(idle))
+			panic(KERN_ERR "Fork failed for CPU %d", cpu);
+	} else {
+		idle = cpu_idle_thread[cpu];
+		init_idle(idle, cpu);
+	}
 
 	mp_ops->boot_secondary(cpu, idle);
 
diff --git a/arch/mips/kernel/topology.c b/arch/mips/kernel/topology.c
index 660e44ed44d7..cf3eb61fad12 100644
--- a/arch/mips/kernel/topology.c
+++ b/arch/mips/kernel/topology.c
@@ -17,7 +17,10 @@ static int __init topology_init(void)
 #endif /* CONFIG_NUMA */
 
 	for_each_present_cpu(i) {
-		ret = register_cpu(&per_cpu(cpu_devices, i), i);
+		struct cpu *c = &per_cpu(cpu_devices, i);
+
+		c->hotpluggable = 1;
+		ret = register_cpu(c, i);
 		if (ret)
 			printk(KERN_WARNING "topology_init: register_cpu %d "
 			       "failed (%d)\n", i, ret);
-- 
cgit v1.2.3-59-g8ed1b


From 9801b321ecdb6708365b6825bf728c8e433fca00 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 23 Jun 2009 10:20:56 +0100
Subject: MIPS: SMP: Allow suspend and hibernation if CPU hotplug is available

The SMP implementation of suspend and hibernate depends on CPU hotplugging.
In the past we didn't have CPU hotplug so suspend and hibernation were not
possible on SMP systems.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index eb7e8d795c6a..3cce4ed5473d 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -2145,11 +2145,11 @@ menu "Power management options"
 
 config ARCH_HIBERNATION_POSSIBLE
 	def_bool y
-	depends on !SMP
+	depends on SYS_SUPPORTS_HOTPLUG_CPU
 
 config ARCH_SUSPEND_POSSIBLE
 	def_bool y
-	depends on !SMP
+	depends on SYS_SUPPORTS_HOTPLUG_CPU
 
 source "kernel/power/Kconfig"
 
-- 
cgit v1.2.3-59-g8ed1b


From 773cb77d0e32f0a3c36edf5aaeb9642c18038cd2 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 23 Jun 2009 10:36:38 +0100
Subject: MIPS: Cavium: Add CPU hotplugging code.

Thanks to Cavium Inc. for the code contribution and help.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig                     |   2 +
 arch/mips/cavium-octeon/octeon-irq.c  |  59 +++++++++
 arch/mips/cavium-octeon/octeon_boot.h |  70 ++++++++++
 arch/mips/cavium-octeon/smp.c         | 234 +++++++++++++++++++++++++++++++++-
 arch/mips/include/asm/smp.h           |   2 +-
 5 files changed, 365 insertions(+), 2 deletions(-)
 create mode 100644 arch/mips/cavium-octeon/octeon_boot.h

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 3cce4ed5473d..8c4be1f301cf 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -601,6 +601,7 @@ config CAVIUM_OCTEON_SIMULATOR
 	select SYS_SUPPORTS_64BIT_KERNEL
 	select SYS_SUPPORTS_BIG_ENDIAN
 	select SYS_SUPPORTS_HIGHMEM
+	select SYS_SUPPORTS_HOTPLUG_CPU
 	select SYS_HAS_CPU_CAVIUM_OCTEON
 	help
 	  The Octeon simulator is software performance model of the Cavium
@@ -615,6 +616,7 @@ config CAVIUM_OCTEON_REFERENCE_BOARD
 	select SYS_SUPPORTS_64BIT_KERNEL
 	select SYS_SUPPORTS_BIG_ENDIAN
 	select SYS_SUPPORTS_HIGHMEM
+	select SYS_SUPPORTS_HOTPLUG_CPU
 	select SYS_HAS_EARLY_PRINTK
 	select SYS_HAS_CPU_CAVIUM_OCTEON
 	select SWAP_IO_SPACE
diff --git a/arch/mips/cavium-octeon/octeon-irq.c b/arch/mips/cavium-octeon/octeon-irq.c
index 3090ee37e109..384f1842bfb1 100644
--- a/arch/mips/cavium-octeon/octeon-irq.c
+++ b/arch/mips/cavium-octeon/octeon-irq.c
@@ -501,3 +501,62 @@ asmlinkage void plat_irq_dispatch(void)
 		}
 	}
 }
+
+#ifdef CONFIG_HOTPLUG_CPU
+static int is_irq_enabled_on_cpu(unsigned int irq, unsigned int cpu)
+{
+       unsigned int isset;
+#ifdef CONFIG_SMP
+       int coreid = cpu_logical_map(cpu);
+#else
+	int coreid = cvmx_get_core_num();
+#endif
+	int bit = (irq < OCTEON_IRQ_WDOG0) ?
+		irq - OCTEON_IRQ_WORKQ0 : irq - OCTEON_IRQ_WDOG0;
+       if (irq < 64) {
+		isset = (cvmx_read_csr(CVMX_CIU_INTX_EN0(coreid * 2)) &
+			(1ull << bit)) >> bit;
+       } else {
+	       isset = (cvmx_read_csr(CVMX_CIU_INTX_EN1(coreid * 2 + 1)) &
+			(1ull << bit)) >> bit;
+       }
+       return isset;
+}
+
+void fixup_irqs(void)
+{
+       int irq;
+
+	for (irq = OCTEON_IRQ_SW0; irq <= OCTEON_IRQ_TIMER; irq++)
+		octeon_irq_core_disable_local(irq);
+
+	for (irq = OCTEON_IRQ_WORKQ0; irq <= OCTEON_IRQ_GPIO15; irq++) {
+		if (is_irq_enabled_on_cpu(irq, smp_processor_id())) {
+			/* ciu irq migrates to next cpu */
+			octeon_irq_chip_ciu0.disable(irq);
+			octeon_irq_ciu0_set_affinity(irq, &cpu_online_map);
+		}
+	}
+
+#if 0
+	for (irq = OCTEON_IRQ_MBOX0; irq <= OCTEON_IRQ_MBOX1; irq++)
+		octeon_irq_mailbox_mask(irq);
+#endif
+	for (irq = OCTEON_IRQ_UART0; irq <= OCTEON_IRQ_BOOTDMA; irq++) {
+		if (is_irq_enabled_on_cpu(irq, smp_processor_id())) {
+			/* ciu irq migrates to next cpu */
+			octeon_irq_chip_ciu0.disable(irq);
+			octeon_irq_ciu0_set_affinity(irq, &cpu_online_map);
+		}
+	}
+
+	for (irq = OCTEON_IRQ_UART2; irq <= OCTEON_IRQ_RESERVED135; irq++) {
+		if (is_irq_enabled_on_cpu(irq, smp_processor_id())) {
+			/* ciu irq migrates to next cpu */
+			octeon_irq_chip_ciu1.disable(irq);
+			octeon_irq_ciu1_set_affinity(irq, &cpu_online_map);
+		}
+	}
+}
+
+#endif /* CONFIG_HOTPLUG_CPU */
diff --git a/arch/mips/cavium-octeon/octeon_boot.h b/arch/mips/cavium-octeon/octeon_boot.h
new file mode 100644
index 000000000000..0f7f84accf9a
--- /dev/null
+++ b/arch/mips/cavium-octeon/octeon_boot.h
@@ -0,0 +1,70 @@
+/*
+ * (C) Copyright 2004, 2005 Cavium Networks
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of
+ * the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston,
+ * MA 02111-1307 USA
+ */
+
+#ifndef __OCTEON_BOOT_H__
+#define __OCTEON_BOOT_H__
+
+#include <linux/types.h>
+
+struct boot_init_vector {
+	uint32_t stack_addr;
+	uint32_t code_addr;
+	uint32_t app_start_func_addr;
+	uint32_t k0_val;
+	uint32_t flags;
+	uint32_t boot_info_addr;
+	uint32_t pad;
+	uint32_t pad2;
+};
+
+/* similar to bootloader's linux_app_boot_info but without global data */
+struct linux_app_boot_info {
+	uint32_t labi_signature;
+	uint32_t start_core0_addr;
+	uint32_t avail_coremask;
+	uint32_t pci_console_active;
+	uint32_t icache_prefetch_disable;
+	uint32_t InitTLBStart_addr;
+	uint32_t start_app_addr;
+	uint32_t cur_exception_base;
+	uint32_t no_mark_private_data;
+	uint32_t compact_flash_common_base_addr;
+	uint32_t compact_flash_attribute_base_addr;
+	uint32_t led_display_base_addr;
+};
+
+/* If not to copy a lot of bootloader's structures
+   here is only offset of requested member */
+#define AVAIL_COREMASK_OFFSET_IN_LINUX_APP_BOOT_BLOCK    0x765c
+
+/* hardcoded in bootloader */
+#define  LABI_ADDR_IN_BOOTLOADER                         0x700
+
+#define LINUX_APP_BOOT_BLOCK_NAME "linux-app-boot"
+
+#define LABI_SIGNATURE 0xAABBCCDD
+
+/*  from uboot-headers/octeon_mem_map.h */
+#define EXCEPTION_BASE_INCR     (4 * 1024)
+			       /* Increment size for exception base addresses (4k minimum) */
+#define EXCEPTION_BASE_BASE     0
+#define BOOTLOADER_PRIV_DATA_BASE       (EXCEPTION_BASE_BASE + 0x800)
+#define BOOTLOADER_BOOT_VECTOR          (BOOTLOADER_PRIV_DATA_BASE)
+
+#endif /* __OCTEON_BOOT_H__ */
diff --git a/arch/mips/cavium-octeon/smp.c b/arch/mips/cavium-octeon/smp.c
index 24e0ad63980a..0b891a9c6253 100644
--- a/arch/mips/cavium-octeon/smp.c
+++ b/arch/mips/cavium-octeon/smp.c
@@ -5,6 +5,7 @@
  *
  * Copyright (C) 2004-2008 Cavium Networks
  */
+#include <linux/cpu.h>
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <linux/smp.h>
@@ -19,10 +20,16 @@
 
 #include <asm/octeon/octeon.h>
 
+#include "octeon_boot.h"
+
 volatile unsigned long octeon_processor_boot = 0xff;
 volatile unsigned long octeon_processor_sp;
 volatile unsigned long octeon_processor_gp;
 
+#ifdef CONFIG_HOTPLUG_CPU
+static unsigned int InitTLBStart_addr;
+#endif
+
 static irqreturn_t mailbox_interrupt(int irq, void *dev_id)
 {
 	const int coreid = cvmx_get_core_num();
@@ -67,8 +74,28 @@ static inline void octeon_send_ipi_mask(cpumask_t mask, unsigned int action)
 }
 
 /**
- * Detect available CPUs, populate phys_cpu_present_map
+ * Detect available CPUs, populate cpu_possible_map
  */
+static void octeon_smp_hotplug_setup(void)
+{
+#ifdef CONFIG_HOTPLUG_CPU
+	uint32_t labi_signature;
+
+	labi_signature =
+		cvmx_read64_uint32(CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+					LABI_ADDR_IN_BOOTLOADER +
+					offsetof(struct linux_app_boot_info,
+						    labi_signature)));
+	if (labi_signature != LABI_SIGNATURE)
+		pr_err("The bootloader version on this board is incorrect\n");
+	InitTLBStart_addr =
+		cvmx_read64_uint32(CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+				   LABI_ADDR_IN_BOOTLOADER +
+					   offsetof(struct linux_app_boot_info,
+						    InitTLBStart_addr)));
+#endif
+}
+
 static void octeon_smp_setup(void)
 {
 	const int coreid = cvmx_get_core_num();
@@ -91,6 +118,9 @@ static void octeon_smp_setup(void)
 			cpus++;
 		}
 	}
+	cpu_present_map = cpu_possible_map;
+
+	octeon_smp_hotplug_setup();
 }
 
 /**
@@ -128,6 +158,17 @@ static void octeon_init_secondary(void)
 	const int coreid = cvmx_get_core_num();
 	union cvmx_ciu_intx_sum0 interrupt_enable;
 
+#ifdef CONFIG_HOTPLUG_CPU
+	unsigned int cur_exception_base;
+
+	cur_exception_base = cvmx_read64_uint32(
+		CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+			     LABI_ADDR_IN_BOOTLOADER +
+			     offsetof(struct linux_app_boot_info,
+				      cur_exception_base)));
+	/* cur_exception_base is incremented in bootloader after setting */
+	write_c0_ebase((unsigned int)(cur_exception_base - EXCEPTION_BASE_INCR));
+#endif
 	octeon_check_cpu_bist();
 	octeon_init_cvmcount();
 	/*
@@ -199,6 +240,193 @@ static void octeon_cpus_done(void)
 #endif
 }
 
+#ifdef CONFIG_HOTPLUG_CPU
+
+/* State of each CPU. */
+DEFINE_PER_CPU(int, cpu_state);
+
+extern void fixup_irqs(void);
+
+static DEFINE_SPINLOCK(smp_reserve_lock);
+
+static int octeon_cpu_disable(void)
+{
+	unsigned int cpu = smp_processor_id();
+
+	if (cpu == 0)
+		return -EBUSY;
+
+	spin_lock(&smp_reserve_lock);
+
+	cpu_clear(cpu, cpu_online_map);
+	cpu_clear(cpu, cpu_callin_map);
+	local_irq_disable();
+	fixup_irqs();
+	local_irq_enable();
+
+	flush_cache_all();
+	local_flush_tlb_all();
+
+	spin_unlock(&smp_reserve_lock);
+
+	return 0;
+}
+
+static void octeon_cpu_die(unsigned int cpu)
+{
+	int coreid = cpu_logical_map(cpu);
+	uint32_t avail_coremask;
+	struct cvmx_bootmem_named_block_desc *block_desc;
+
+#ifdef CONFIG_CAVIUM_OCTEON_WATCHDOG
+	/* Disable the watchdog */
+	cvmx_ciu_wdogx_t ciu_wdog;
+	ciu_wdog.u64 = cvmx_read_csr(CVMX_CIU_WDOGX(cpu));
+	ciu_wdog.s.mode = 0;
+	cvmx_write_csr(CVMX_CIU_WDOGX(cpu), ciu_wdog.u64);
+#endif
+
+	while (per_cpu(cpu_state, cpu) != CPU_DEAD)
+		cpu_relax();
+
+	/*
+	 * This is a bit complicated strategics of getting/settig available
+	 * cores mask, copied from bootloader
+	 */
+	/* LINUX_APP_BOOT_BLOCK is initialized in bootoct binary */
+	block_desc = cvmx_bootmem_find_named_block(LINUX_APP_BOOT_BLOCK_NAME);
+
+	if (!block_desc) {
+		avail_coremask =
+			cvmx_read64_uint32(CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+						   LABI_ADDR_IN_BOOTLOADER +
+						   offsetof
+						   (struct linux_app_boot_info,
+						    avail_coremask)));
+	} else {		       /* alternative, already initialized */
+	       avail_coremask =
+		   cvmx_read64_uint32(CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+						   block_desc->base_addr +
+						  AVAIL_COREMASK_OFFSET_IN_LINUX_APP_BOOT_BLOCK));
+	}
+
+	avail_coremask |= 1 << coreid;
+
+	/* Setting avail_coremask for bootoct binary */
+	if (!block_desc) {
+		cvmx_write64_uint32(CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+						LABI_ADDR_IN_BOOTLOADER +
+						offsetof(struct linux_app_boot_info,
+							 avail_coremask)),
+				   avail_coremask);
+	} else {
+		cvmx_write64_uint32(CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+						block_desc->base_addr +
+						AVAIL_COREMASK_OFFSET_IN_LINUX_APP_BOOT_BLOCK),
+				   avail_coremask);
+	}
+
+	pr_info("Reset core %d. Available Coremask = %x \n", coreid,
+		avail_coremask);
+	cvmx_write_csr(CVMX_CIU_PP_RST, 1 << coreid);
+	cvmx_write_csr(CVMX_CIU_PP_RST, 0);
+}
+
+void play_dead(void)
+{
+	int coreid = cvmx_get_core_num();
+
+	idle_task_exit();
+	octeon_processor_boot = 0xff;
+	per_cpu(cpu_state, coreid) = CPU_DEAD;
+
+	while (1)	/* core will be reset here */
+		;
+}
+
+extern void kernel_entry(unsigned long arg1, ...);
+
+static void start_after_reset(void)
+{
+	kernel_entry(0, 0, 0);  /* set a2 = 0 for secondary core */
+}
+
+int octeon_update_boot_vector(unsigned int cpu)
+{
+
+	int coreid = cpu_logical_map(cpu);
+	unsigned int avail_coremask;
+	struct cvmx_bootmem_named_block_desc *block_desc;
+	struct boot_init_vector *boot_vect =
+		(struct boot_init_vector *) cvmx_phys_to_ptr(0x0 +
+						  BOOTLOADER_BOOT_VECTOR);
+
+	block_desc = cvmx_bootmem_find_named_block(LINUX_APP_BOOT_BLOCK_NAME);
+
+	if (!block_desc) {
+		avail_coremask =
+			cvmx_read64_uint32(CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+					   LABI_ADDR_IN_BOOTLOADER +
+						offsetof(struct linux_app_boot_info,
+						avail_coremask)));
+	} else {		       /* alternative, already initialized */
+	       avail_coremask =
+		   cvmx_read64_uint32(CVMX_ADD_SEG(CVMX_MIPS_SPACE_XKPHYS,
+						   block_desc->base_addr +
+						   AVAIL_COREMASK_OFFSET_IN_LINUX_APP_BOOT_BLOCK));
+	}
+
+	if (!(avail_coremask & (1 << coreid))) {
+		/* core not available, assume, that catched by simple-executive */
+		cvmx_write_csr(CVMX_CIU_PP_RST, 1 << coreid);
+		cvmx_write_csr(CVMX_CIU_PP_RST, 0);
+	}
+
+	boot_vect[coreid].app_start_func_addr =
+		(uint32_t) (unsigned long) start_after_reset;
+	boot_vect[coreid].code_addr = InitTLBStart_addr;
+
+	CVMX_SYNC;
+
+	cvmx_write_csr(CVMX_CIU_NMI, (1 << coreid) & avail_coremask);
+
+	return 0;
+}
+
+static int __cpuinit octeon_cpu_callback(struct notifier_block *nfb,
+	unsigned long action, void *hcpu)
+{
+	unsigned int cpu = (unsigned long)hcpu;
+
+	switch (action) {
+	case CPU_UP_PREPARE:
+		octeon_update_boot_vector(cpu);
+		break;
+	case CPU_ONLINE:
+		pr_info("Cpu %d online\n", cpu);
+		break;
+	case CPU_DEAD:
+		break;
+	}
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block __cpuinitdata octeon_cpu_notifier = {
+	.notifier_call = octeon_cpu_callback,
+};
+
+static int __cpuinit register_cavium_notifier(void)
+{
+	register_hotcpu_notifier(&octeon_cpu_notifier);
+
+	return 0;
+}
+
+late_initcall(register_cavium_notifier);
+
+#endif  /* CONFIG_HOTPLUG_CPU */
+
 struct plat_smp_ops octeon_smp_ops = {
 	.send_ipi_single	= octeon_send_ipi_single,
 	.send_ipi_mask		= octeon_send_ipi_mask,
@@ -208,4 +436,8 @@ struct plat_smp_ops octeon_smp_ops = {
 	.boot_secondary		= octeon_boot_secondary,
 	.smp_setup		= octeon_smp_setup,
 	.prepare_cpus		= octeon_prepare_cpus,
+#ifdef CONFIG_HOTPLUG_CPU
+	.cpu_disable		= octeon_cpu_disable,
+	.cpu_die		= octeon_cpu_die,
+#endif
 };
diff --git a/arch/mips/include/asm/smp.h b/arch/mips/include/asm/smp.h
index 01f813dc3888..aaa2d4ab26dc 100644
--- a/arch/mips/include/asm/smp.h
+++ b/arch/mips/include/asm/smp.h
@@ -41,7 +41,7 @@ extern int __cpu_logical_map[NR_CPUS];
 /* Octeon - Tell another core to flush its icache */
 #define SMP_ICACHE_FLUSH	0x4
 
-extern cpumask_t cpu_callin_map;
+extern volatile cpumask_t cpu_callin_map;
 
 extern void asmlinkage smp_bootstrap(void);
 
-- 
cgit v1.2.3-59-g8ed1b


From f696a10838ffab85e5bc07e7cff0d0e1870a30d7 Mon Sep 17 00:00:00 2001
From: David Daney <ddaney@caviumnetworks.com>
Date: Tue, 23 Jun 2009 11:34:08 -0700
Subject: Staging: octeon-ethernet: Convert to use net_device_ops.

Convert the driver to use net_device_ops as it is now mandatory.

Also compensate for the removal of struct sk_buff's dst field.

The changes are mostly mechanical, the content of ethernet-common.c
was moved to ethernet.c and ethernet-common.{c,h} are removed.

Signed-off-by: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 drivers/staging/octeon/Makefile          |   1 -
 drivers/staging/octeon/ethernet-common.c | 328 --------------------------
 drivers/staging/octeon/ethernet-common.h |  29 ---
 drivers/staging/octeon/ethernet-rgmii.c  |   9 +-
 drivers/staging/octeon/ethernet-sgmii.c  |   9 +-
 drivers/staging/octeon/ethernet-spi.c    |   1 -
 drivers/staging/octeon/ethernet-tx.c     |   6 +-
 drivers/staging/octeon/ethernet-xaui.c   |   9 +-
 drivers/staging/octeon/ethernet.c        | 383 +++++++++++++++++++++++++++++--
 drivers/staging/octeon/octeon-ethernet.h |  11 +
 10 files changed, 390 insertions(+), 396 deletions(-)
 delete mode 100644 drivers/staging/octeon/ethernet-common.c
 delete mode 100644 drivers/staging/octeon/ethernet-common.h

diff --git a/drivers/staging/octeon/Makefile b/drivers/staging/octeon/Makefile
index 3c839e37d37f..c0a583cc2227 100644
--- a/drivers/staging/octeon/Makefile
+++ b/drivers/staging/octeon/Makefile
@@ -12,7 +12,6 @@
 obj-${CONFIG_OCTEON_ETHERNET} :=  octeon-ethernet.o
 
 octeon-ethernet-objs := ethernet.o
-octeon-ethernet-objs += ethernet-common.o
 octeon-ethernet-objs += ethernet-mdio.o
 octeon-ethernet-objs += ethernet-mem.o
 octeon-ethernet-objs += ethernet-proc.o
diff --git a/drivers/staging/octeon/ethernet-common.c b/drivers/staging/octeon/ethernet-common.c
deleted file mode 100644
index 3e6f5b8cc63d..000000000000
--- a/drivers/staging/octeon/ethernet-common.c
+++ /dev/null
@@ -1,328 +0,0 @@
-/**********************************************************************
- * Author: Cavium Networks
- *
- * Contact: support@caviumnetworks.com
- * This file is part of the OCTEON SDK
- *
- * Copyright (c) 2003-2007 Cavium Networks
- *
- * This file is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License, Version 2, as
- * published by the Free Software Foundation.
- *
- * This file is distributed in the hope that it will be useful, but
- * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
- * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
- * NONINFRINGEMENT.  See the GNU General Public License for more
- * details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this file; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- * or visit http://www.gnu.org/licenses/.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium Networks for more information
-**********************************************************************/
-#include <linux/kernel.h>
-#include <linux/mii.h>
-#include <net/dst.h>
-
-#include <asm/atomic.h>
-#include <asm/octeon/octeon.h>
-
-#include "ethernet-defines.h"
-#include "ethernet-tx.h"
-#include "ethernet-mdio.h"
-#include "ethernet-util.h"
-#include "octeon-ethernet.h"
-#include "ethernet-common.h"
-
-#include "cvmx-pip.h"
-#include "cvmx-pko.h"
-#include "cvmx-fau.h"
-#include "cvmx-helper.h"
-
-#include "cvmx-gmxx-defs.h"
-
-/**
- * Get the low level ethernet statistics
- *
- * @dev:    Device to get the statistics from
- * Returns Pointer to the statistics
- */
-static struct net_device_stats *cvm_oct_common_get_stats(struct net_device *dev)
-{
-	cvmx_pip_port_status_t rx_status;
-	cvmx_pko_port_status_t tx_status;
-	struct octeon_ethernet *priv = netdev_priv(dev);
-
-	if (priv->port < CVMX_PIP_NUM_INPUT_PORTS) {
-		if (octeon_is_simulation()) {
-			/* The simulator doesn't support statistics */
-			memset(&rx_status, 0, sizeof(rx_status));
-			memset(&tx_status, 0, sizeof(tx_status));
-		} else {
-			cvmx_pip_get_port_status(priv->port, 1, &rx_status);
-			cvmx_pko_get_port_status(priv->port, 1, &tx_status);
-		}
-
-		priv->stats.rx_packets += rx_status.inb_packets;
-		priv->stats.tx_packets += tx_status.packets;
-		priv->stats.rx_bytes += rx_status.inb_octets;
-		priv->stats.tx_bytes += tx_status.octets;
-		priv->stats.multicast += rx_status.multicast_packets;
-		priv->stats.rx_crc_errors += rx_status.inb_errors;
-		priv->stats.rx_frame_errors += rx_status.fcs_align_err_packets;
-
-		/*
-		 * The drop counter must be incremented atomically
-		 * since the RX tasklet also increments it.
-		 */
-#ifdef CONFIG_64BIT
-		atomic64_add(rx_status.dropped_packets,
-			     (atomic64_t *)&priv->stats.rx_dropped);
-#else
-		atomic_add(rx_status.dropped_packets,
-			     (atomic_t *)&priv->stats.rx_dropped);
-#endif
-	}
-
-	return &priv->stats;
-}
-
-/**
- * Set the multicast list. Currently unimplemented.
- *
- * @dev:    Device to work on
- */
-static void cvm_oct_common_set_multicast_list(struct net_device *dev)
-{
-	union cvmx_gmxx_prtx_cfg gmx_cfg;
-	struct octeon_ethernet *priv = netdev_priv(dev);
-	int interface = INTERFACE(priv->port);
-	int index = INDEX(priv->port);
-
-	if ((interface < 2)
-	    && (cvmx_helper_interface_get_mode(interface) !=
-		CVMX_HELPER_INTERFACE_MODE_SPI)) {
-		union cvmx_gmxx_rxx_adr_ctl control;
-		control.u64 = 0;
-		control.s.bcst = 1;	/* Allow broadcast MAC addresses */
-
-		if (dev->mc_list || (dev->flags & IFF_ALLMULTI) ||
-		    (dev->flags & IFF_PROMISC))
-			/* Force accept multicast packets */
-			control.s.mcst = 2;
-		else
-			/* Force reject multicat packets */
-			control.s.mcst = 1;
-
-		if (dev->flags & IFF_PROMISC)
-			/*
-			 * Reject matches if promisc. Since CAM is
-			 * shut off, should accept everything.
-			 */
-			control.s.cam_mode = 0;
-		else
-			/* Filter packets based on the CAM */
-			control.s.cam_mode = 1;
-
-		gmx_cfg.u64 =
-		    cvmx_read_csr(CVMX_GMXX_PRTX_CFG(index, interface));
-		cvmx_write_csr(CVMX_GMXX_PRTX_CFG(index, interface),
-			       gmx_cfg.u64 & ~1ull);
-
-		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CTL(index, interface),
-			       control.u64);
-		if (dev->flags & IFF_PROMISC)
-			cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM_EN
-				       (index, interface), 0);
-		else
-			cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM_EN
-				       (index, interface), 1);
-
-		cvmx_write_csr(CVMX_GMXX_PRTX_CFG(index, interface),
-			       gmx_cfg.u64);
-	}
-}
-
-/**
- * Set the hardware MAC address for a device
- *
- * @dev:    Device to change the MAC address for
- * @addr:   Address structure to change it too. MAC address is addr + 2.
- * Returns Zero on success
- */
-static int cvm_oct_common_set_mac_address(struct net_device *dev, void *addr)
-{
-	struct octeon_ethernet *priv = netdev_priv(dev);
-	union cvmx_gmxx_prtx_cfg gmx_cfg;
-	int interface = INTERFACE(priv->port);
-	int index = INDEX(priv->port);
-
-	memcpy(dev->dev_addr, addr + 2, 6);
-
-	if ((interface < 2)
-	    && (cvmx_helper_interface_get_mode(interface) !=
-		CVMX_HELPER_INTERFACE_MODE_SPI)) {
-		int i;
-		uint8_t *ptr = addr;
-		uint64_t mac = 0;
-		for (i = 0; i < 6; i++)
-			mac = (mac << 8) | (uint64_t) (ptr[i + 2]);
-
-		gmx_cfg.u64 =
-		    cvmx_read_csr(CVMX_GMXX_PRTX_CFG(index, interface));
-		cvmx_write_csr(CVMX_GMXX_PRTX_CFG(index, interface),
-			       gmx_cfg.u64 & ~1ull);
-
-		cvmx_write_csr(CVMX_GMXX_SMACX(index, interface), mac);
-		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM0(index, interface),
-			       ptr[2]);
-		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM1(index, interface),
-			       ptr[3]);
-		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM2(index, interface),
-			       ptr[4]);
-		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM3(index, interface),
-			       ptr[5]);
-		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM4(index, interface),
-			       ptr[6]);
-		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM5(index, interface),
-			       ptr[7]);
-		cvm_oct_common_set_multicast_list(dev);
-		cvmx_write_csr(CVMX_GMXX_PRTX_CFG(index, interface),
-			       gmx_cfg.u64);
-	}
-	return 0;
-}
-
-/**
- * Change the link MTU. Unimplemented
- *
- * @dev:     Device to change
- * @new_mtu: The new MTU
- *
- * Returns Zero on success
- */
-static int cvm_oct_common_change_mtu(struct net_device *dev, int new_mtu)
-{
-	struct octeon_ethernet *priv = netdev_priv(dev);
-	int interface = INTERFACE(priv->port);
-	int index = INDEX(priv->port);
-#if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
-	int vlan_bytes = 4;
-#else
-	int vlan_bytes = 0;
-#endif
-
-	/*
-	 * Limit the MTU to make sure the ethernet packets are between
-	 * 64 bytes and 65535 bytes.
-	 */
-	if ((new_mtu + 14 + 4 + vlan_bytes < 64)
-	    || (new_mtu + 14 + 4 + vlan_bytes > 65392)) {
-		pr_err("MTU must be between %d and %d.\n",
-		       64 - 14 - 4 - vlan_bytes, 65392 - 14 - 4 - vlan_bytes);
-		return -EINVAL;
-	}
-	dev->mtu = new_mtu;
-
-	if ((interface < 2)
-	    && (cvmx_helper_interface_get_mode(interface) !=
-		CVMX_HELPER_INTERFACE_MODE_SPI)) {
-		/* Add ethernet header and FCS, and VLAN if configured. */
-		int max_packet = new_mtu + 14 + 4 + vlan_bytes;
-
-		if (OCTEON_IS_MODEL(OCTEON_CN3XXX)
-		    || OCTEON_IS_MODEL(OCTEON_CN58XX)) {
-			/* Signal errors on packets larger than the MTU */
-			cvmx_write_csr(CVMX_GMXX_RXX_FRM_MAX(index, interface),
-				       max_packet);
-		} else {
-			/*
-			 * Set the hardware to truncate packets larger
-			 * than the MTU and smaller the 64 bytes.
-			 */
-			union cvmx_pip_frm_len_chkx frm_len_chk;
-			frm_len_chk.u64 = 0;
-			frm_len_chk.s.minlen = 64;
-			frm_len_chk.s.maxlen = max_packet;
-			cvmx_write_csr(CVMX_PIP_FRM_LEN_CHKX(interface),
-				       frm_len_chk.u64);
-		}
-		/*
-		 * Set the hardware to truncate packets larger than
-		 * the MTU. The jabber register must be set to a
-		 * multiple of 8 bytes, so round up.
-		 */
-		cvmx_write_csr(CVMX_GMXX_RXX_JABBER(index, interface),
-			       (max_packet + 7) & ~7u);
-	}
-	return 0;
-}
-
-/**
- * Per network device initialization
- *
- * @dev:    Device to initialize
- * Returns Zero on success
- */
-int cvm_oct_common_init(struct net_device *dev)
-{
-	static int count;
-	char mac[8] = { 0x00, 0x00,
-		octeon_bootinfo->mac_addr_base[0],
-		octeon_bootinfo->mac_addr_base[1],
-		octeon_bootinfo->mac_addr_base[2],
-		octeon_bootinfo->mac_addr_base[3],
-		octeon_bootinfo->mac_addr_base[4],
-		octeon_bootinfo->mac_addr_base[5] + count
-	};
-	struct octeon_ethernet *priv = netdev_priv(dev);
-
-	/*
-	 * Force the interface to use the POW send if always_use_pow
-	 * was specified or it is in the pow send list.
-	 */
-	if ((pow_send_group != -1)
-	    && (always_use_pow || strstr(pow_send_list, dev->name)))
-		priv->queue = -1;
-
-	if (priv->queue != -1) {
-		dev->hard_start_xmit = cvm_oct_xmit;
-		if (USE_HW_TCPUDP_CHECKSUM)
-			dev->features |= NETIF_F_IP_CSUM;
-	} else
-		dev->hard_start_xmit = cvm_oct_xmit_pow;
-	count++;
-
-	dev->get_stats = cvm_oct_common_get_stats;
-	dev->set_mac_address = cvm_oct_common_set_mac_address;
-	dev->set_multicast_list = cvm_oct_common_set_multicast_list;
-	dev->change_mtu = cvm_oct_common_change_mtu;
-	dev->do_ioctl = cvm_oct_ioctl;
-	/* We do our own locking, Linux doesn't need to */
-	dev->features |= NETIF_F_LLTX;
-	SET_ETHTOOL_OPS(dev, &cvm_oct_ethtool_ops);
-#ifdef CONFIG_NET_POLL_CONTROLLER
-	dev->poll_controller = cvm_oct_poll_controller;
-#endif
-
-	cvm_oct_mdio_setup_device(dev);
-	dev->set_mac_address(dev, mac);
-	dev->change_mtu(dev, dev->mtu);
-
-	/*
-	 * Zero out stats for port so we won't mistakenly show
-	 * counters from the bootloader.
-	 */
-	memset(dev->get_stats(dev), 0, sizeof(struct net_device_stats));
-
-	return 0;
-}
-
-void cvm_oct_common_uninit(struct net_device *dev)
-{
-	/* Currently nothing to do */
-}
diff --git a/drivers/staging/octeon/ethernet-common.h b/drivers/staging/octeon/ethernet-common.h
deleted file mode 100644
index 2bd9cd76a398..000000000000
--- a/drivers/staging/octeon/ethernet-common.h
+++ /dev/null
@@ -1,29 +0,0 @@
-/*********************************************************************
- * Author: Cavium Networks
- *
- * Contact: support@caviumnetworks.com
- * This file is part of the OCTEON SDK
- *
- * Copyright (c) 2003-2007 Cavium Networks
- *
- * This file is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License, Version 2, as
- * published by the Free Software Foundation.
- *
- * This file is distributed in the hope that it will be useful, but
- * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
- * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
- * NONINFRINGEMENT.  See the GNU General Public License for more
- * details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this file; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- * or visit http://www.gnu.org/licenses/.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium Networks for more information
-*********************************************************************/
-
-int cvm_oct_common_init(struct net_device *dev);
-void cvm_oct_common_uninit(struct net_device *dev);
diff --git a/drivers/staging/octeon/ethernet-rgmii.c b/drivers/staging/octeon/ethernet-rgmii.c
index 8579f1670d1e..8704133fe127 100644
--- a/drivers/staging/octeon/ethernet-rgmii.c
+++ b/drivers/staging/octeon/ethernet-rgmii.c
@@ -33,7 +33,6 @@
 
 #include "ethernet-defines.h"
 #include "octeon-ethernet.h"
-#include "ethernet-common.h"
 #include "ethernet-util.h"
 
 #include "cvmx-helper.h"
@@ -265,7 +264,7 @@ static irqreturn_t cvm_oct_rgmii_rml_interrupt(int cpl, void *dev_id)
 	return return_status;
 }
 
-static int cvm_oct_rgmii_open(struct net_device *dev)
+int cvm_oct_rgmii_open(struct net_device *dev)
 {
 	union cvmx_gmxx_prtx_cfg gmx_cfg;
 	struct octeon_ethernet *priv = netdev_priv(dev);
@@ -286,7 +285,7 @@ static int cvm_oct_rgmii_open(struct net_device *dev)
 	return 0;
 }
 
-static int cvm_oct_rgmii_stop(struct net_device *dev)
+int cvm_oct_rgmii_stop(struct net_device *dev)
 {
 	union cvmx_gmxx_prtx_cfg gmx_cfg;
 	struct octeon_ethernet *priv = netdev_priv(dev);
@@ -305,9 +304,7 @@ int cvm_oct_rgmii_init(struct net_device *dev)
 	int r;
 
 	cvm_oct_common_init(dev);
-	dev->open = cvm_oct_rgmii_open;
-	dev->stop = cvm_oct_rgmii_stop;
-	dev->stop(dev);
+	dev->netdev_ops->ndo_stop(dev);
 
 	/*
 	 * Due to GMX errata in CN3XXX series chips, it is necessary
diff --git a/drivers/staging/octeon/ethernet-sgmii.c b/drivers/staging/octeon/ethernet-sgmii.c
index 58fa39c1d675..2b54996bd85d 100644
--- a/drivers/staging/octeon/ethernet-sgmii.c
+++ b/drivers/staging/octeon/ethernet-sgmii.c
@@ -34,13 +34,12 @@
 #include "ethernet-defines.h"
 #include "octeon-ethernet.h"
 #include "ethernet-util.h"
-#include "ethernet-common.h"
 
 #include "cvmx-helper.h"
 
 #include "cvmx-gmxx-defs.h"
 
-static int cvm_oct_sgmii_open(struct net_device *dev)
+int cvm_oct_sgmii_open(struct net_device *dev)
 {
 	union cvmx_gmxx_prtx_cfg gmx_cfg;
 	struct octeon_ethernet *priv = netdev_priv(dev);
@@ -61,7 +60,7 @@ static int cvm_oct_sgmii_open(struct net_device *dev)
 	return 0;
 }
 
-static int cvm_oct_sgmii_stop(struct net_device *dev)
+int cvm_oct_sgmii_stop(struct net_device *dev)
 {
 	union cvmx_gmxx_prtx_cfg gmx_cfg;
 	struct octeon_ethernet *priv = netdev_priv(dev);
@@ -113,9 +112,7 @@ int cvm_oct_sgmii_init(struct net_device *dev)
 {
 	struct octeon_ethernet *priv = netdev_priv(dev);
 	cvm_oct_common_init(dev);
-	dev->open = cvm_oct_sgmii_open;
-	dev->stop = cvm_oct_sgmii_stop;
-	dev->stop(dev);
+	dev->netdev_ops->ndo_stop(dev);
 	if (!octeon_is_simulation())
 		priv->poll = cvm_oct_sgmii_poll;
 
diff --git a/drivers/staging/octeon/ethernet-spi.c b/drivers/staging/octeon/ethernet-spi.c
index e0971bbe4ddc..66190b0cb68f 100644
--- a/drivers/staging/octeon/ethernet-spi.c
+++ b/drivers/staging/octeon/ethernet-spi.c
@@ -33,7 +33,6 @@
 
 #include "ethernet-defines.h"
 #include "octeon-ethernet.h"
-#include "ethernet-common.h"
 #include "ethernet-util.h"
 
 #include "cvmx-spi.h"
diff --git a/drivers/staging/octeon/ethernet-tx.c b/drivers/staging/octeon/ethernet-tx.c
index 77b7122c8fdb..bfd3dd2fcef8 100644
--- a/drivers/staging/octeon/ethernet-tx.c
+++ b/drivers/staging/octeon/ethernet-tx.c
@@ -253,10 +253,10 @@ int cvm_oct_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	/*
 	 * The skbuff will be reused without ever being freed. We must
-	 * cleanup a bunch of Linux stuff.
+	 * cleanup a bunch of core things.
 	 */
-	dst_release(skb->dst);
-	skb->dst = NULL;
+	dst_release(skb_dst(skb));
+	skb_dst_set(skb, NULL);
 #ifdef CONFIG_XFRM
 	secpath_put(skb->sp);
 	skb->sp = NULL;
diff --git a/drivers/staging/octeon/ethernet-xaui.c b/drivers/staging/octeon/ethernet-xaui.c
index f08eb32e04fc..0c2e7cc40f35 100644
--- a/drivers/staging/octeon/ethernet-xaui.c
+++ b/drivers/staging/octeon/ethernet-xaui.c
@@ -33,14 +33,13 @@
 
 #include "ethernet-defines.h"
 #include "octeon-ethernet.h"
-#include "ethernet-common.h"
 #include "ethernet-util.h"
 
 #include "cvmx-helper.h"
 
 #include "cvmx-gmxx-defs.h"
 
-static int cvm_oct_xaui_open(struct net_device *dev)
+int cvm_oct_xaui_open(struct net_device *dev)
 {
 	union cvmx_gmxx_prtx_cfg gmx_cfg;
 	struct octeon_ethernet *priv = netdev_priv(dev);
@@ -60,7 +59,7 @@ static int cvm_oct_xaui_open(struct net_device *dev)
 	return 0;
 }
 
-static int cvm_oct_xaui_stop(struct net_device *dev)
+int cvm_oct_xaui_stop(struct net_device *dev)
 {
 	union cvmx_gmxx_prtx_cfg gmx_cfg;
 	struct octeon_ethernet *priv = netdev_priv(dev);
@@ -112,9 +111,7 @@ int cvm_oct_xaui_init(struct net_device *dev)
 {
 	struct octeon_ethernet *priv = netdev_priv(dev);
 	cvm_oct_common_init(dev);
-	dev->open = cvm_oct_xaui_open;
-	dev->stop = cvm_oct_xaui_stop;
-	dev->stop(dev);
+	dev->netdev_ops->ndo_stop(dev);
 	if (!octeon_is_simulation())
 		priv->poll = cvm_oct_xaui_poll;
 
diff --git a/drivers/staging/octeon/ethernet.c b/drivers/staging/octeon/ethernet.c
index e8ef9e0b791f..2d9356dfbca6 100644
--- a/drivers/staging/octeon/ethernet.c
+++ b/drivers/staging/octeon/ethernet.c
@@ -40,9 +40,9 @@
 #include "ethernet-mem.h"
 #include "ethernet-rx.h"
 #include "ethernet-tx.h"
+#include "ethernet-mdio.h"
 #include "ethernet-util.h"
 #include "ethernet-proc.h"
-#include "ethernet-common.h"
 #include "octeon-ethernet.h"
 
 #include "cvmx-pip.h"
@@ -51,6 +51,7 @@
 #include "cvmx-ipd.h"
 #include "cvmx-helper.h"
 
+#include "cvmx-gmxx-defs.h"
 #include "cvmx-smix-defs.h"
 
 #if defined(CONFIG_CAVIUM_OCTEON_NUM_PACKET_BUFFERS) \
@@ -164,7 +165,7 @@ static void cvm_do_timer(unsigned long arg)
 						    lock);
 				}
 			}
-			cvm_oct_device[port]->get_stats(cvm_oct_device[port]);
+			cvm_oct_device[port]->netdev_ops->ndo_get_stats(cvm_oct_device[port]);
 		}
 		port++;
 		/* Poll the next port in a 50th of a second.
@@ -245,6 +246,362 @@ int cvm_oct_free_work(void *work_queue_entry)
 }
 EXPORT_SYMBOL(cvm_oct_free_work);
 
+/**
+ * Get the low level ethernet statistics
+ *
+ * @dev:    Device to get the statistics from
+ * Returns Pointer to the statistics
+ */
+static struct net_device_stats *cvm_oct_common_get_stats(struct net_device *dev)
+{
+	cvmx_pip_port_status_t rx_status;
+	cvmx_pko_port_status_t tx_status;
+	struct octeon_ethernet *priv = netdev_priv(dev);
+
+	if (priv->port < CVMX_PIP_NUM_INPUT_PORTS) {
+		if (octeon_is_simulation()) {
+			/* The simulator doesn't support statistics */
+			memset(&rx_status, 0, sizeof(rx_status));
+			memset(&tx_status, 0, sizeof(tx_status));
+		} else {
+			cvmx_pip_get_port_status(priv->port, 1, &rx_status);
+			cvmx_pko_get_port_status(priv->port, 1, &tx_status);
+		}
+
+		priv->stats.rx_packets += rx_status.inb_packets;
+		priv->stats.tx_packets += tx_status.packets;
+		priv->stats.rx_bytes += rx_status.inb_octets;
+		priv->stats.tx_bytes += tx_status.octets;
+		priv->stats.multicast += rx_status.multicast_packets;
+		priv->stats.rx_crc_errors += rx_status.inb_errors;
+		priv->stats.rx_frame_errors += rx_status.fcs_align_err_packets;
+
+		/*
+		 * The drop counter must be incremented atomically
+		 * since the RX tasklet also increments it.
+		 */
+#ifdef CONFIG_64BIT
+		atomic64_add(rx_status.dropped_packets,
+			     (atomic64_t *)&priv->stats.rx_dropped);
+#else
+		atomic_add(rx_status.dropped_packets,
+			     (atomic_t *)&priv->stats.rx_dropped);
+#endif
+	}
+
+	return &priv->stats;
+}
+
+/**
+ * Change the link MTU. Unimplemented
+ *
+ * @dev:     Device to change
+ * @new_mtu: The new MTU
+ *
+ * Returns Zero on success
+ */
+static int cvm_oct_common_change_mtu(struct net_device *dev, int new_mtu)
+{
+	struct octeon_ethernet *priv = netdev_priv(dev);
+	int interface = INTERFACE(priv->port);
+	int index = INDEX(priv->port);
+#if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
+	int vlan_bytes = 4;
+#else
+	int vlan_bytes = 0;
+#endif
+
+	/*
+	 * Limit the MTU to make sure the ethernet packets are between
+	 * 64 bytes and 65535 bytes.
+	 */
+	if ((new_mtu + 14 + 4 + vlan_bytes < 64)
+	    || (new_mtu + 14 + 4 + vlan_bytes > 65392)) {
+		pr_err("MTU must be between %d and %d.\n",
+		       64 - 14 - 4 - vlan_bytes, 65392 - 14 - 4 - vlan_bytes);
+		return -EINVAL;
+	}
+	dev->mtu = new_mtu;
+
+	if ((interface < 2)
+	    && (cvmx_helper_interface_get_mode(interface) !=
+		CVMX_HELPER_INTERFACE_MODE_SPI)) {
+		/* Add ethernet header and FCS, and VLAN if configured. */
+		int max_packet = new_mtu + 14 + 4 + vlan_bytes;
+
+		if (OCTEON_IS_MODEL(OCTEON_CN3XXX)
+		    || OCTEON_IS_MODEL(OCTEON_CN58XX)) {
+			/* Signal errors on packets larger than the MTU */
+			cvmx_write_csr(CVMX_GMXX_RXX_FRM_MAX(index, interface),
+				       max_packet);
+		} else {
+			/*
+			 * Set the hardware to truncate packets larger
+			 * than the MTU and smaller the 64 bytes.
+			 */
+			union cvmx_pip_frm_len_chkx frm_len_chk;
+			frm_len_chk.u64 = 0;
+			frm_len_chk.s.minlen = 64;
+			frm_len_chk.s.maxlen = max_packet;
+			cvmx_write_csr(CVMX_PIP_FRM_LEN_CHKX(interface),
+				       frm_len_chk.u64);
+		}
+		/*
+		 * Set the hardware to truncate packets larger than
+		 * the MTU. The jabber register must be set to a
+		 * multiple of 8 bytes, so round up.
+		 */
+		cvmx_write_csr(CVMX_GMXX_RXX_JABBER(index, interface),
+			       (max_packet + 7) & ~7u);
+	}
+	return 0;
+}
+
+/**
+ * Set the multicast list. Currently unimplemented.
+ *
+ * @dev:    Device to work on
+ */
+static void cvm_oct_common_set_multicast_list(struct net_device *dev)
+{
+	union cvmx_gmxx_prtx_cfg gmx_cfg;
+	struct octeon_ethernet *priv = netdev_priv(dev);
+	int interface = INTERFACE(priv->port);
+	int index = INDEX(priv->port);
+
+	if ((interface < 2)
+	    && (cvmx_helper_interface_get_mode(interface) !=
+		CVMX_HELPER_INTERFACE_MODE_SPI)) {
+		union cvmx_gmxx_rxx_adr_ctl control;
+		control.u64 = 0;
+		control.s.bcst = 1;	/* Allow broadcast MAC addresses */
+
+		if (dev->mc_list || (dev->flags & IFF_ALLMULTI) ||
+		    (dev->flags & IFF_PROMISC))
+			/* Force accept multicast packets */
+			control.s.mcst = 2;
+		else
+			/* Force reject multicat packets */
+			control.s.mcst = 1;
+
+		if (dev->flags & IFF_PROMISC)
+			/*
+			 * Reject matches if promisc. Since CAM is
+			 * shut off, should accept everything.
+			 */
+			control.s.cam_mode = 0;
+		else
+			/* Filter packets based on the CAM */
+			control.s.cam_mode = 1;
+
+		gmx_cfg.u64 =
+		    cvmx_read_csr(CVMX_GMXX_PRTX_CFG(index, interface));
+		cvmx_write_csr(CVMX_GMXX_PRTX_CFG(index, interface),
+			       gmx_cfg.u64 & ~1ull);
+
+		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CTL(index, interface),
+			       control.u64);
+		if (dev->flags & IFF_PROMISC)
+			cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM_EN
+				       (index, interface), 0);
+		else
+			cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM_EN
+				       (index, interface), 1);
+
+		cvmx_write_csr(CVMX_GMXX_PRTX_CFG(index, interface),
+			       gmx_cfg.u64);
+	}
+}
+
+/**
+ * Set the hardware MAC address for a device
+ *
+ * @dev:    Device to change the MAC address for
+ * @addr:   Address structure to change it too. MAC address is addr + 2.
+ * Returns Zero on success
+ */
+static int cvm_oct_common_set_mac_address(struct net_device *dev, void *addr)
+{
+	struct octeon_ethernet *priv = netdev_priv(dev);
+	union cvmx_gmxx_prtx_cfg gmx_cfg;
+	int interface = INTERFACE(priv->port);
+	int index = INDEX(priv->port);
+
+	memcpy(dev->dev_addr, addr + 2, 6);
+
+	if ((interface < 2)
+	    && (cvmx_helper_interface_get_mode(interface) !=
+		CVMX_HELPER_INTERFACE_MODE_SPI)) {
+		int i;
+		uint8_t *ptr = addr;
+		uint64_t mac = 0;
+		for (i = 0; i < 6; i++)
+			mac = (mac << 8) | (uint64_t) (ptr[i + 2]);
+
+		gmx_cfg.u64 =
+		    cvmx_read_csr(CVMX_GMXX_PRTX_CFG(index, interface));
+		cvmx_write_csr(CVMX_GMXX_PRTX_CFG(index, interface),
+			       gmx_cfg.u64 & ~1ull);
+
+		cvmx_write_csr(CVMX_GMXX_SMACX(index, interface), mac);
+		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM0(index, interface),
+			       ptr[2]);
+		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM1(index, interface),
+			       ptr[3]);
+		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM2(index, interface),
+			       ptr[4]);
+		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM3(index, interface),
+			       ptr[5]);
+		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM4(index, interface),
+			       ptr[6]);
+		cvmx_write_csr(CVMX_GMXX_RXX_ADR_CAM5(index, interface),
+			       ptr[7]);
+		cvm_oct_common_set_multicast_list(dev);
+		cvmx_write_csr(CVMX_GMXX_PRTX_CFG(index, interface),
+			       gmx_cfg.u64);
+	}
+	return 0;
+}
+
+/**
+ * Per network device initialization
+ *
+ * @dev:    Device to initialize
+ * Returns Zero on success
+ */
+int cvm_oct_common_init(struct net_device *dev)
+{
+	static int count;
+	char mac[8] = { 0x00, 0x00,
+		octeon_bootinfo->mac_addr_base[0],
+		octeon_bootinfo->mac_addr_base[1],
+		octeon_bootinfo->mac_addr_base[2],
+		octeon_bootinfo->mac_addr_base[3],
+		octeon_bootinfo->mac_addr_base[4],
+		octeon_bootinfo->mac_addr_base[5] + count
+	};
+	struct octeon_ethernet *priv = netdev_priv(dev);
+
+	/*
+	 * Force the interface to use the POW send if always_use_pow
+	 * was specified or it is in the pow send list.
+	 */
+	if ((pow_send_group != -1)
+	    && (always_use_pow || strstr(pow_send_list, dev->name)))
+		priv->queue = -1;
+
+	if (priv->queue != -1 && USE_HW_TCPUDP_CHECKSUM)
+		dev->features |= NETIF_F_IP_CSUM;
+
+	count++;
+
+	/* We do our own locking, Linux doesn't need to */
+	dev->features |= NETIF_F_LLTX;
+	SET_ETHTOOL_OPS(dev, &cvm_oct_ethtool_ops);
+
+	cvm_oct_mdio_setup_device(dev);
+	dev->netdev_ops->ndo_set_mac_address(dev, mac);
+	dev->netdev_ops->ndo_change_mtu(dev, dev->mtu);
+
+	/*
+	 * Zero out stats for port so we won't mistakenly show
+	 * counters from the bootloader.
+	 */
+	memset(dev->netdev_ops->ndo_get_stats(dev), 0,
+	       sizeof(struct net_device_stats));
+
+	return 0;
+}
+
+void cvm_oct_common_uninit(struct net_device *dev)
+{
+	/* Currently nothing to do */
+}
+
+static const struct net_device_ops cvm_oct_npi_netdev_ops = {
+	.ndo_init		= cvm_oct_common_init,
+	.ndo_uninit		= cvm_oct_common_uninit,
+	.ndo_start_xmit		= cvm_oct_xmit,
+	.ndo_set_multicast_list	= cvm_oct_common_set_multicast_list,
+	.ndo_set_mac_address	= cvm_oct_common_set_mac_address,
+	.ndo_do_ioctl		= cvm_oct_ioctl,
+	.ndo_change_mtu		= cvm_oct_common_change_mtu,
+	.ndo_get_stats		= cvm_oct_common_get_stats,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller	= cvm_oct_poll_controller,
+#endif
+};
+static const struct net_device_ops cvm_oct_xaui_netdev_ops = {
+	.ndo_init		= cvm_oct_xaui_init,
+	.ndo_uninit		= cvm_oct_xaui_uninit,
+	.ndo_open		= cvm_oct_xaui_open,
+	.ndo_stop		= cvm_oct_xaui_stop,
+	.ndo_start_xmit		= cvm_oct_xmit,
+	.ndo_set_multicast_list	= cvm_oct_common_set_multicast_list,
+	.ndo_set_mac_address	= cvm_oct_common_set_mac_address,
+	.ndo_do_ioctl		= cvm_oct_ioctl,
+	.ndo_change_mtu		= cvm_oct_common_change_mtu,
+	.ndo_get_stats		= cvm_oct_common_get_stats,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller	= cvm_oct_poll_controller,
+#endif
+};
+static const struct net_device_ops cvm_oct_sgmii_netdev_ops = {
+	.ndo_init		= cvm_oct_sgmii_init,
+	.ndo_uninit		= cvm_oct_sgmii_uninit,
+	.ndo_open		= cvm_oct_sgmii_open,
+	.ndo_stop		= cvm_oct_sgmii_stop,
+	.ndo_start_xmit		= cvm_oct_xmit,
+	.ndo_set_multicast_list	= cvm_oct_common_set_multicast_list,
+	.ndo_set_mac_address	= cvm_oct_common_set_mac_address,
+	.ndo_do_ioctl		= cvm_oct_ioctl,
+	.ndo_change_mtu		= cvm_oct_common_change_mtu,
+	.ndo_get_stats		= cvm_oct_common_get_stats,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller	= cvm_oct_poll_controller,
+#endif
+};
+static const struct net_device_ops cvm_oct_spi_netdev_ops = {
+	.ndo_init		= cvm_oct_spi_init,
+	.ndo_uninit		= cvm_oct_spi_uninit,
+	.ndo_start_xmit		= cvm_oct_xmit,
+	.ndo_set_multicast_list	= cvm_oct_common_set_multicast_list,
+	.ndo_set_mac_address	= cvm_oct_common_set_mac_address,
+	.ndo_do_ioctl		= cvm_oct_ioctl,
+	.ndo_change_mtu		= cvm_oct_common_change_mtu,
+	.ndo_get_stats		= cvm_oct_common_get_stats,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller	= cvm_oct_poll_controller,
+#endif
+};
+static const struct net_device_ops cvm_oct_rgmii_netdev_ops = {
+	.ndo_init		= cvm_oct_rgmii_init,
+	.ndo_uninit		= cvm_oct_rgmii_uninit,
+	.ndo_open		= cvm_oct_rgmii_open,
+	.ndo_stop		= cvm_oct_rgmii_stop,
+	.ndo_start_xmit		= cvm_oct_xmit,
+	.ndo_set_multicast_list	= cvm_oct_common_set_multicast_list,
+	.ndo_set_mac_address	= cvm_oct_common_set_mac_address,
+	.ndo_do_ioctl		= cvm_oct_ioctl,
+	.ndo_change_mtu		= cvm_oct_common_change_mtu,
+	.ndo_get_stats		= cvm_oct_common_get_stats,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller	= cvm_oct_poll_controller,
+#endif
+};
+static const struct net_device_ops cvm_oct_pow_netdev_ops = {
+	.ndo_init		= cvm_oct_common_init,
+	.ndo_start_xmit		= cvm_oct_xmit_pow,
+	.ndo_set_multicast_list	= cvm_oct_common_set_multicast_list,
+	.ndo_set_mac_address	= cvm_oct_common_set_mac_address,
+	.ndo_do_ioctl		= cvm_oct_ioctl,
+	.ndo_change_mtu		= cvm_oct_common_change_mtu,
+	.ndo_get_stats		= cvm_oct_common_get_stats,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+	.ndo_poll_controller	= cvm_oct_poll_controller,
+#endif
+};
+
 /**
  * Module/ driver initialization. Creates the linux network
  * devices.
@@ -303,7 +660,7 @@ static int __init cvm_oct_init_module(void)
 			struct octeon_ethernet *priv = netdev_priv(dev);
 			memset(priv, 0, sizeof(struct octeon_ethernet));
 
-			dev->init = cvm_oct_common_init;
+			dev->netdev_ops = &cvm_oct_pow_netdev_ops;
 			priv->imode = CVMX_HELPER_INTERFACE_MODE_DISABLED;
 			priv->port = CVMX_PIP_NUM_INPUT_PORTS;
 			priv->queue = -1;
@@ -372,44 +729,38 @@ static int __init cvm_oct_init_module(void)
 				break;
 
 			case CVMX_HELPER_INTERFACE_MODE_NPI:
-				dev->init = cvm_oct_common_init;
-				dev->uninit = cvm_oct_common_uninit;
+				dev->netdev_ops = &cvm_oct_npi_netdev_ops;
 				strcpy(dev->name, "npi%d");
 				break;
 
 			case CVMX_HELPER_INTERFACE_MODE_XAUI:
-				dev->init = cvm_oct_xaui_init;
-				dev->uninit = cvm_oct_xaui_uninit;
+				dev->netdev_ops = &cvm_oct_xaui_netdev_ops;
 				strcpy(dev->name, "xaui%d");
 				break;
 
 			case CVMX_HELPER_INTERFACE_MODE_LOOP:
-				dev->init = cvm_oct_common_init;
-				dev->uninit = cvm_oct_common_uninit;
+				dev->netdev_ops = &cvm_oct_npi_netdev_ops;
 				strcpy(dev->name, "loop%d");
 				break;
 
 			case CVMX_HELPER_INTERFACE_MODE_SGMII:
-				dev->init = cvm_oct_sgmii_init;
-				dev->uninit = cvm_oct_sgmii_uninit;
+				dev->netdev_ops = &cvm_oct_sgmii_netdev_ops;
 				strcpy(dev->name, "eth%d");
 				break;
 
 			case CVMX_HELPER_INTERFACE_MODE_SPI:
-				dev->init = cvm_oct_spi_init;
-				dev->uninit = cvm_oct_spi_uninit;
+				dev->netdev_ops = &cvm_oct_spi_netdev_ops;
 				strcpy(dev->name, "spi%d");
 				break;
 
 			case CVMX_HELPER_INTERFACE_MODE_RGMII:
 			case CVMX_HELPER_INTERFACE_MODE_GMII:
-				dev->init = cvm_oct_rgmii_init;
-				dev->uninit = cvm_oct_rgmii_uninit;
+				dev->netdev_ops = &cvm_oct_rgmii_netdev_ops;
 				strcpy(dev->name, "eth%d");
 				break;
 			}
 
-			if (!dev->init) {
+			if (!dev->netdev_ops) {
 				kfree(dev);
 			} else if (register_netdev(dev) < 0) {
 				pr_err("Failed to register ethernet device "
diff --git a/drivers/staging/octeon/octeon-ethernet.h b/drivers/staging/octeon/octeon-ethernet.h
index b3199076ef5e..3aef9878fc0a 100644
--- a/drivers/staging/octeon/octeon-ethernet.h
+++ b/drivers/staging/octeon/octeon-ethernet.h
@@ -111,12 +111,23 @@ static inline int cvm_oct_transmit(struct net_device *dev,
 
 extern int cvm_oct_rgmii_init(struct net_device *dev);
 extern void cvm_oct_rgmii_uninit(struct net_device *dev);
+extern int cvm_oct_rgmii_open(struct net_device *dev);
+extern int cvm_oct_rgmii_stop(struct net_device *dev);
+
 extern int cvm_oct_sgmii_init(struct net_device *dev);
 extern void cvm_oct_sgmii_uninit(struct net_device *dev);
+extern int cvm_oct_sgmii_open(struct net_device *dev);
+extern int cvm_oct_sgmii_stop(struct net_device *dev);
+
 extern int cvm_oct_spi_init(struct net_device *dev);
 extern void cvm_oct_spi_uninit(struct net_device *dev);
 extern int cvm_oct_xaui_init(struct net_device *dev);
 extern void cvm_oct_xaui_uninit(struct net_device *dev);
+extern int cvm_oct_xaui_open(struct net_device *dev);
+extern int cvm_oct_xaui_stop(struct net_device *dev);
+
+extern int cvm_oct_common_init(struct net_device *dev);
+extern void cvm_oct_common_uninit(struct net_device *dev);
 
 extern int always_use_pow;
 extern int pow_send_group;
-- 
cgit v1.2.3-59-g8ed1b


From a620c1632629b42369e78448acc7b384fe1faf48 Mon Sep 17 00:00:00 2001
From: David Daney <ddaney@caviumnetworks.com>
Date: Tue, 23 Jun 2009 16:20:56 -0700
Subject: Staging: octeon-ethernet: Fix race freeing transmit buffers.

The existing code had the following race:

Thread-1                       Thread-2

inc/read in_use
                               inc/read in_use
inc tx_free_list[qos].len
                               inc tx_free_list[qos].len

The actual in_use value was incremented twice, but thread-1 is going
to free memory based on its stale value, and will free one too many
times.  The result is that memory is freed back to the kernel while
its packet is still in the transmit buffer.  If the memory is
overwritten before it is transmitted, the hardware will put a valid
checksum on it and send it out (just like it does with good packets).
If by chance the TCP flags are clobbered but not the addresses or
ports, the result can be a broken TCP stream.

The fix is to track the number of freed packets in a single location
(a Fetch-and-Add Unit register).  That way it can never get out of sync
with itself.

We try to free up to MAX_SKB_TO_FREE (currently 10) buffers at a time.
If fewer are available we adjust the free count with the difference.
The action of claiming buffers to free is atomic so two threads cannot
claim the same buffers.

Signed-off-by: David Daney <ddaney@caviumnetworks.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 drivers/staging/octeon/ethernet-defines.h |  2 +
 drivers/staging/octeon/ethernet-tx.c      | 56 +++++++++++--------
 drivers/staging/octeon/ethernet-tx.h      | 25 +++++++++
 drivers/staging/octeon/ethernet.c         | 89 ++++++++++++++++---------------
 4 files changed, 106 insertions(+), 66 deletions(-)

diff --git a/drivers/staging/octeon/ethernet-defines.h b/drivers/staging/octeon/ethernet-defines.h
index 8f7374e7664c..f13131b03c33 100644
--- a/drivers/staging/octeon/ethernet-defines.h
+++ b/drivers/staging/octeon/ethernet-defines.h
@@ -117,6 +117,8 @@
 
 /* Maximum number of packets to process per interrupt. */
 #define MAX_RX_PACKETS 120
+/* Maximum number of SKBs to try to free per xmit packet. */
+#define MAX_SKB_TO_FREE 10
 #define MAX_OUT_QUEUE_DEPTH 1000
 
 #ifndef CONFIG_SMP
diff --git a/drivers/staging/octeon/ethernet-tx.c b/drivers/staging/octeon/ethernet-tx.c
index bfd3dd2fcef8..81a851390f1b 100644
--- a/drivers/staging/octeon/ethernet-tx.c
+++ b/drivers/staging/octeon/ethernet-tx.c
@@ -47,6 +47,7 @@
 
 #include "ethernet-defines.h"
 #include "octeon-ethernet.h"
+#include "ethernet-tx.h"
 #include "ethernet-util.h"
 
 #include "cvmx-wqe.h"
@@ -82,8 +83,10 @@ int cvm_oct_xmit(struct sk_buff *skb, struct net_device *dev)
 	uint64_t old_scratch2;
 	int dropped;
 	int qos;
+	int queue_it_up;
 	struct octeon_ethernet *priv = netdev_priv(dev);
-	int32_t in_use;
+	int32_t skb_to_free;
+	int32_t undo;
 	int32_t buffers_to_free;
 #if REUSE_SKBUFFS_WITHOUT_FREE
 	unsigned char *fpa_head;
@@ -120,15 +123,15 @@ int cvm_oct_xmit(struct sk_buff *skb, struct net_device *dev)
 		old_scratch2 = cvmx_scratch_read64(CVMX_SCR_SCRATCH + 8);
 
 		/*
-		 * Assume we're going to be able t osend this
-		 * packet. Fetch and increment the number of pending
-		 * packets for output.
+		 * Fetch and increment the number of packets to be
+		 * freed.
 		 */
 		cvmx_fau_async_fetch_and_add32(CVMX_SCR_SCRATCH + 8,
 					       FAU_NUM_PACKET_BUFFERS_TO_FREE,
 					       0);
 		cvmx_fau_async_fetch_and_add32(CVMX_SCR_SCRATCH,
-					       priv->fau + qos * 4, 1);
+					       priv->fau + qos * 4,
+					       MAX_SKB_TO_FREE);
 	}
 
 	/*
@@ -286,15 +289,29 @@ dont_put_skbuff_in_hw:
 	if (USE_ASYNC_IOBDMA) {
 		/* Get the number of skbuffs in use by the hardware */
 		CVMX_SYNCIOBDMA;
-		in_use = cvmx_scratch_read64(CVMX_SCR_SCRATCH);
+		skb_to_free = cvmx_scratch_read64(CVMX_SCR_SCRATCH);
 		buffers_to_free = cvmx_scratch_read64(CVMX_SCR_SCRATCH + 8);
 	} else {
 		/* Get the number of skbuffs in use by the hardware */
-		in_use = cvmx_fau_fetch_and_add32(priv->fau + qos * 4, 1);
+		skb_to_free = cvmx_fau_fetch_and_add32(priv->fau + qos * 4,
+						       MAX_SKB_TO_FREE);
 		buffers_to_free =
 		    cvmx_fau_fetch_and_add32(FAU_NUM_PACKET_BUFFERS_TO_FREE, 0);
 	}
 
+	/*
+	 * We try to claim MAX_SKB_TO_FREE buffers.  If there were not
+	 * that many available, we have to un-claim (undo) any that
+	 * were in excess.  If skb_to_free is positive we will free
+	 * that many buffers.
+	 */
+	undo = skb_to_free > 0 ?
+		MAX_SKB_TO_FREE : skb_to_free + MAX_SKB_TO_FREE;
+	if (undo > 0)
+		cvmx_fau_atomic_add32(priv->fau+qos*4, -undo);
+	skb_to_free = -skb_to_free > MAX_SKB_TO_FREE ?
+		MAX_SKB_TO_FREE : -skb_to_free;
+
 	/*
 	 * If we're sending faster than the receive can free them then
 	 * don't do the HW free.
@@ -330,38 +347,31 @@ dont_put_skbuff_in_hw:
 		cvmx_scratch_write64(CVMX_SCR_SCRATCH + 8, old_scratch2);
 	}
 
+	queue_it_up = 0;
 	if (unlikely(dropped)) {
 		dev_kfree_skb_any(skb);
-		cvmx_fau_atomic_add32(priv->fau + qos * 4, -1);
 		priv->stats.tx_dropped++;
 	} else {
 		if (USE_SKBUFFS_IN_HW) {
 			/* Put this packet on the queue to be freed later */
 			if (pko_command.s.dontfree)
-				skb_queue_tail(&priv->tx_free_list[qos], skb);
-			else {
+				queue_it_up = 1;
+			else
 				cvmx_fau_atomic_add32
 				    (FAU_NUM_PACKET_BUFFERS_TO_FREE, -1);
-				cvmx_fau_atomic_add32(priv->fau + qos * 4, -1);
-			}
 		} else {
 			/* Put this packet on the queue to be freed later */
-			skb_queue_tail(&priv->tx_free_list[qos], skb);
+			queue_it_up = 1;
 		}
 	}
 
-	/* Free skbuffs not in use by the hardware, possibly two at a time */
-	if (skb_queue_len(&priv->tx_free_list[qos]) > in_use) {
+	if (queue_it_up) {
 		spin_lock(&priv->tx_free_list[qos].lock);
-		/*
-		 * Check again now that we have the lock. It might
-		 * have changed.
-		 */
-		if (skb_queue_len(&priv->tx_free_list[qos]) > in_use)
-			dev_kfree_skb(__skb_dequeue(&priv->tx_free_list[qos]));
-		if (skb_queue_len(&priv->tx_free_list[qos]) > in_use)
-			dev_kfree_skb(__skb_dequeue(&priv->tx_free_list[qos]));
+		__skb_queue_tail(&priv->tx_free_list[qos], skb);
+		cvm_oct_free_tx_skbs(priv, skb_to_free, qos, 0);
 		spin_unlock(&priv->tx_free_list[qos].lock);
+	} else {
+		cvm_oct_free_tx_skbs(priv, skb_to_free, qos, 1);
 	}
 
 	return 0;
diff --git a/drivers/staging/octeon/ethernet-tx.h b/drivers/staging/octeon/ethernet-tx.h
index 5106236fe981..c0bebf750bc0 100644
--- a/drivers/staging/octeon/ethernet-tx.h
+++ b/drivers/staging/octeon/ethernet-tx.h
@@ -30,3 +30,28 @@ int cvm_oct_xmit_pow(struct sk_buff *skb, struct net_device *dev);
 int cvm_oct_transmit_qos(struct net_device *dev, void *work_queue_entry,
 			 int do_free, int qos);
 void cvm_oct_tx_shutdown(struct net_device *dev);
+
+/**
+ * Free dead transmit skbs.
+ *
+ * @priv:		The driver data
+ * @skb_to_free:	The number of SKBs to free (free none if negative).
+ * @qos:		The queue to free from.
+ * @take_lock:		If true, acquire the skb list lock.
+ */
+static inline void cvm_oct_free_tx_skbs(struct octeon_ethernet *priv,
+					int skb_to_free,
+					int qos, int take_lock)
+{
+	/* Free skbuffs not in use by the hardware.  */
+	if (skb_to_free > 0) {
+		if (take_lock)
+			spin_lock(&priv->tx_free_list[qos].lock);
+		while (skb_to_free > 0) {
+			dev_kfree_skb(__skb_dequeue(&priv->tx_free_list[qos]));
+			skb_to_free--;
+		}
+		if (take_lock)
+			spin_unlock(&priv->tx_free_list[qos].lock);
+	}
+}
diff --git a/drivers/staging/octeon/ethernet.c b/drivers/staging/octeon/ethernet.c
index 2d9356dfbca6..b8479517dce2 100644
--- a/drivers/staging/octeon/ethernet.c
+++ b/drivers/staging/octeon/ethernet.c
@@ -37,13 +37,14 @@
 #include <asm/octeon/octeon.h>
 
 #include "ethernet-defines.h"
+#include "octeon-ethernet.h"
 #include "ethernet-mem.h"
 #include "ethernet-rx.h"
 #include "ethernet-tx.h"
 #include "ethernet-mdio.h"
 #include "ethernet-util.h"
 #include "ethernet-proc.h"
-#include "octeon-ethernet.h"
+
 
 #include "cvmx-pip.h"
 #include "cvmx-pko.h"
@@ -130,53 +131,55 @@ extern struct semaphore mdio_sem;
  */
 static void cvm_do_timer(unsigned long arg)
 {
+	int32_t skb_to_free, undo;
+	int queues_per_port;
+	int qos;
+	struct octeon_ethernet *priv;
 	static int port;
-	if (port < CVMX_PIP_NUM_INPUT_PORTS) {
-		if (cvm_oct_device[port]) {
-			int queues_per_port;
-			int qos;
-			struct octeon_ethernet *priv =
-				netdev_priv(cvm_oct_device[port]);
-			if (priv->poll) {
-				/* skip polling if we don't get the lock */
-				if (!down_trylock(&mdio_sem)) {
-					priv->poll(cvm_oct_device[port]);
-					up(&mdio_sem);
-				}
-			}
 
-			queues_per_port = cvmx_pko_get_num_queues(port);
-			/* Drain any pending packets in the free list */
-			for (qos = 0; qos < queues_per_port; qos++) {
-				if (skb_queue_len(&priv->tx_free_list[qos])) {
-					spin_lock(&priv->tx_free_list[qos].
-						  lock);
-					while (skb_queue_len
-					       (&priv->tx_free_list[qos]) >
-					       cvmx_fau_fetch_and_add32(priv->
-									fau +
-									qos * 4,
-									0))
-						dev_kfree_skb(__skb_dequeue
-							      (&priv->
-							       tx_free_list
-							       [qos]));
-					spin_unlock(&priv->tx_free_list[qos].
-						    lock);
-				}
-			}
-			cvm_oct_device[port]->netdev_ops->ndo_get_stats(cvm_oct_device[port]);
-		}
-		port++;
-		/* Poll the next port in a 50th of a second.
-		   This spreads the polling of ports out a little bit */
-		mod_timer(&cvm_oct_poll_timer, jiffies + HZ / 50);
-	} else {
+	if (port >= CVMX_PIP_NUM_INPUT_PORTS) {
+		/*
+		 * All ports have been polled. Start the next
+		 * iteration through the ports in one second.
+		 */
 		port = 0;
-		/* All ports have been polled. Start the next iteration through
-		   the ports in one second */
 		mod_timer(&cvm_oct_poll_timer, jiffies + HZ);
+		return;
+	}
+	if (!cvm_oct_device[port])
+		goto out;
+
+	priv = netdev_priv(cvm_oct_device[port]);
+	if (priv->poll) {
+		/* skip polling if we don't get the lock */
+		if (!down_trylock(&mdio_sem)) {
+			priv->poll(cvm_oct_device[port]);
+			up(&mdio_sem);
+		}
+	}
+
+	queues_per_port = cvmx_pko_get_num_queues(port);
+	/* Drain any pending packets in the free list */
+	for (qos = 0; qos < queues_per_port; qos++) {
+		if (skb_queue_len(&priv->tx_free_list[qos]) == 0)
+			continue;
+		skb_to_free = cvmx_fau_fetch_and_add32(priv->fau + qos * 4,
+						       MAX_SKB_TO_FREE);
+		undo = skb_to_free > 0 ?
+			MAX_SKB_TO_FREE : skb_to_free + MAX_SKB_TO_FREE;
+		if (undo > 0)
+			cvmx_fau_atomic_add32(priv->fau+qos*4, -undo);
+		skb_to_free = -skb_to_free > MAX_SKB_TO_FREE ?
+			MAX_SKB_TO_FREE : -skb_to_free;
+		cvm_oct_free_tx_skbs(priv, skb_to_free, qos, 1);
 	}
+	cvm_oct_device[port]->netdev_ops->ndo_get_stats(cvm_oct_device[port]);
+
+out:
+	port++;
+	/* Poll the next port in a 50th of a second.
+	   This spreads the polling of ports out a little bit */
+	mod_timer(&cvm_oct_poll_timer, jiffies + HZ / 50);
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 163b2f0ba93e9298b3d5fff2337d860c3872ec60 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 25 Jun 2009 02:49:03 +0900
Subject: sh64: Hook up page fault events for software perf counters.

sh64 can use these as well, so tie them up there as well.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/tlbflush_64.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/arch/sh/mm/tlbflush_64.c b/arch/sh/mm/tlbflush_64.c
index fcbb6e135cef..3ce40ea34824 100644
--- a/arch/sh/mm/tlbflush_64.c
+++ b/arch/sh/mm/tlbflush_64.c
@@ -3,7 +3,7 @@
  *
  * Copyright (C) 2000, 2001  Paolo Alberelli
  * Copyright (C) 2003  Richard Curnow (/proc/tlb, bug fixes)
- * Copyright (C) 2003  Paul Mundt
+ * Copyright (C) 2003 - 2009 Paul Mundt
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
@@ -20,6 +20,7 @@
 #include <linux/mman.h>
 #include <linux/mm.h>
 #include <linux/smp.h>
+#include <linux/perf_counter.h>
 #include <linux/interrupt.h>
 #include <asm/system.h>
 #include <asm/io.h>
@@ -115,6 +116,8 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
 	/* Not an IO address, so reenable interrupts */
 	local_irq_enable();
 
+	perf_swcounter_event(PERF_COUNT_SW_PAGE_FAULTS, 1, 0, regs, address);
+
 	/*
 	 * If we're in an interrupt or have no user
 	 * context, we must not take the fault..
@@ -195,10 +198,16 @@ survive:
 			goto do_sigbus;
 		BUG();
 	}
-	if (fault & VM_FAULT_MAJOR)
+
+	if (fault & VM_FAULT_MAJOR) {
 		tsk->maj_flt++;
-	else
+		perf_swcounter_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1, 0,
+				     regs, address);
+	} else {
 		tsk->min_flt++;
+		perf_swcounter_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1, 0,
+				     regs, address);
+	}
 
 	/* If we get here, the page fault has been handled.  Do the TLB refill
 	   now from the newly-setup PTE, to avoid having to fault again right
-- 
cgit v1.2.3-59-g8ed1b


From 1b173f77dd0d5fd4f0ff18034aaa79e30da068b9 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 24 Jun 2009 19:54:29 +0200
Subject: perf_counter tools: Add CREDITS file for Git contributors

Much of perf's libraries comes from the Git project. I noticed
that the files (in tools/perf/util/*.[ch] and elsewhere) are
quite spartan wrt. credits, so lets add a CREDITS file that
includes an (incomplete!) list of main contributors.

Thanks guys, these libraries are really useful. Special thanks
go to Johannes Schindelin and Junio C Hamano for coming up with
this list.

List-Composed-By: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Cc: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/CREDITS | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)
 create mode 100644 tools/perf/CREDITS

diff --git a/tools/perf/CREDITS b/tools/perf/CREDITS
new file mode 100644
index 000000000000..c2ddcb3acbd0
--- /dev/null
+++ b/tools/perf/CREDITS
@@ -0,0 +1,30 @@
+Most of the infrastructure that 'perf' uses here has been reused
+from the Git project, as of version:
+
+    66996ec: Sync with 1.6.2.4
+
+Here is an (incomplete!) list of main contributors to those files
+in util/* and elsewhere:
+
+ Alex Riesen
+ Christian Couder
+ Dmitry Potapov
+ Jeff King
+ Johannes Schindelin
+ Johannes Sixt
+ Junio C Hamano
+ Linus Torvalds
+ Matthias Kestenholz
+ Michal Ostrowski
+ Miklos Vajna
+ Petr Baudis
+ Pierre Habouzit
+ René Scharfe
+ Samuel Tardieu
+ Shawn O. Pearce
+ Steffen Prohaska
+ Steve Haslam
+
+Thanks guys!
+
+The full history of the files can be found in the upstream Git commits.
-- 
cgit v1.2.3-59-g8ed1b


From 4923abf9f1a4c1864af438a57c1f3686548230e9 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Wed, 24 Jun 2009 12:16:49 -0700
Subject: Don't warn about order-1 allocations with __GFP_NOFAIL

Traditionally, we never failed small orders (even regardless of any
__GFP_NOFAIL flags), and slab will allocate order-1 allocations even for
small allocations that could fit in a single page (in order to avoid
excessive fragmentation).

Maybe we should remove this warning entirely, but before making that
judgement, at least limit it to bigger allocations.

Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/page_alloc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index aecc9cdfdfce..5d714f8fb303 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1153,10 +1153,10 @@ again:
 			 * properly detect and handle allocation failures.
 			 *
 			 * We most definitely don't want callers attempting to
-			 * allocate greater than single-page units with
+			 * allocate greater than order-1 page units with
 			 * __GFP_NOFAIL.
 			 */
-			WARN_ON_ONCE(order > 0);
+			WARN_ON_ONCE(order > 1);
 		}
 		spin_lock_irqsave(&zone->lock, flags);
 		page = __rmqueue(zone, order, migratetype);
-- 
cgit v1.2.3-59-g8ed1b


From ba52270d18fb17ce2cf176b35419dab1e43fe4a3 Mon Sep 17 00:00:00 2001
From: Pekka Enberg <penberg@cs.helsinki.fi>
Date: Wed, 24 Jun 2009 21:59:51 +0300
Subject: SLUB: Don't pass __GFP_FAIL for the initial allocation

SLUB uses higher order allocations by default but falls back to small
orders under memory pressure. Make sure the GFP mask used in the initial
allocation doesn't include __GFP_NOFAIL.

Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/slub.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/mm/slub.c b/mm/slub.c
index ce62b770e2fc..819f056b39c6 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1085,11 +1085,17 @@ static struct page *allocate_slab(struct kmem_cache *s, gfp_t flags, int node)
 {
 	struct page *page;
 	struct kmem_cache_order_objects oo = s->oo;
+	gfp_t alloc_gfp;
 
 	flags |= s->allocflags;
 
-	page = alloc_slab_page(flags | __GFP_NOWARN | __GFP_NORETRY, node,
-									oo);
+	/*
+	 * Let the initial higher-order allocation fail under memory pressure
+	 * so we fall-back to the minimum order allocation.
+	 */
+	alloc_gfp = (flags | __GFP_NOWARN | __GFP_NORETRY) & ~__GFP_NOFAIL;
+
+	page = alloc_slab_page(alloc_gfp, node, oo);
 	if (unlikely(!page)) {
 		oo = s->min;
 		/*
-- 
cgit v1.2.3-59-g8ed1b


From d0725992c8a6fb63a16bc9e8b2a50094cc4db3cd Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 11 Jun 2009 23:15:43 +0200
Subject: futex: Fix the write access fault problem for real

commit 64d1304a64 (futex: setup writeable mapping for futex ops which
modify user space data) did address only half of the problem of write
access faults.

The patch was made on two wrong assumptions:

1) access_ok(VERIFY_WRITE,...) would actually check write access.

   On x86 it does _NOT_. It's a pure address range check.

2) a RW mapped region can not go away under us.

   That's wrong as well. Nobody can prevent another thread to call
   mprotect(PROT_READ) on that region where the futex resides. If that
   call hits between the get_user_pages_fast() verification and the
   actual write access in the atomic region we are toast again.

The solution is to not rely on access_ok and get_user() for any write
access related fault on private and shared futexes. Instead we need to
fault it in with verification of write access.

There is no generic non destructive write mechanism which would fault
the user page in trough a #PF, but as we already know that we will
fault we can as well call get_user_pages() directly and avoid the #PF
overhead.

If get_user_pages() returns -EFAULT we know that we can not fix it
anymore and need to bail out to user space.

Remove a bunch of confusing comments on this issue as well.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@kernel.org
---
 kernel/futex.c | 45 ++++++++++++++++++++++++---------------------
 1 file changed, 24 insertions(+), 21 deletions(-)

diff --git a/kernel/futex.c b/kernel/futex.c
index 80b5ce716596..1c337112335c 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -284,6 +284,25 @@ void put_futex_key(int fshared, union futex_key *key)
 	drop_futex_key_refs(key);
 }
 
+/*
+ * fault_in_user_writeable - fault in user address and verify RW access
+ * @uaddr:	pointer to faulting user space address
+ *
+ * Slow path to fixup the fault we just took in the atomic write
+ * access to @uaddr.
+ *
+ * We have no generic implementation of a non destructive write to the
+ * user address. We know that we faulted in the atomic pagefault
+ * disabled section so we can as well avoid the #PF overhead by
+ * calling get_user_pages() right away.
+ */
+static int fault_in_user_writeable(u32 __user *uaddr)
+{
+	int ret = get_user_pages(current, current->mm, (unsigned long)uaddr,
+				 sizeof(*uaddr), 1, 0, NULL, NULL);
+	return ret < 0 ? ret : 0;
+}
+
 /**
  * futex_top_waiter() - Return the highest priority waiter on a futex
  * @hb:     the hash bucket the futex_q's reside in
@@ -896,7 +915,6 @@ retry:
 retry_private:
 	op_ret = futex_atomic_op_inuser(op, uaddr2);
 	if (unlikely(op_ret < 0)) {
-		u32 dummy;
 
 		double_unlock_hb(hb1, hb2);
 
@@ -914,7 +932,7 @@ retry_private:
 			goto out_put_keys;
 		}
 
-		ret = get_user(dummy, uaddr2);
+		ret = fault_in_user_writeable(uaddr2);
 		if (ret)
 			goto out_put_keys;
 
@@ -1204,7 +1222,7 @@ retry_private:
 			double_unlock_hb(hb1, hb2);
 			put_futex_key(fshared, &key2);
 			put_futex_key(fshared, &key1);
-			ret = get_user(curval2, uaddr2);
+			ret = fault_in_user_writeable(uaddr2);
 			if (!ret)
 				goto retry;
 			goto out;
@@ -1482,7 +1500,7 @@ retry:
 handle_fault:
 	spin_unlock(q->lock_ptr);
 
-	ret = get_user(uval, uaddr);
+	ret = fault_in_user_writeable(uaddr);
 
 	spin_lock(q->lock_ptr);
 
@@ -1807,7 +1825,6 @@ static int futex_lock_pi(u32 __user *uaddr, int fshared,
 {
 	struct hrtimer_sleeper timeout, *to = NULL;
 	struct futex_hash_bucket *hb;
-	u32 uval;
 	struct futex_q q;
 	int res, ret;
 
@@ -1909,16 +1926,9 @@ out:
 	return ret != -EINTR ? ret : -ERESTARTNOINTR;
 
 uaddr_faulted:
-	/*
-	 * We have to r/w  *(int __user *)uaddr, and we have to modify it
-	 * atomically.  Therefore, if we continue to fault after get_user()
-	 * below, we need to handle the fault ourselves, while still holding
-	 * the mmap_sem.  This can occur if the uaddr is under contention as
-	 * we have to drop the mmap_sem in order to call get_user().
-	 */
 	queue_unlock(&q, hb);
 
-	ret = get_user(uval, uaddr);
+	ret = fault_in_user_writeable(uaddr);
 	if (ret)
 		goto out_put_key;
 
@@ -2013,17 +2023,10 @@ out:
 	return ret;
 
 pi_faulted:
-	/*
-	 * We have to r/w  *(int __user *)uaddr, and we have to modify it
-	 * atomically.  Therefore, if we continue to fault after get_user()
-	 * below, we need to handle the fault ourselves, while still holding
-	 * the mmap_sem.  This can occur if the uaddr is under contention as
-	 * we have to drop the mmap_sem in order to call get_user().
-	 */
 	spin_unlock(&hb->lock);
 	put_futex_key(fshared, &key);
 
-	ret = get_user(uval, uaddr);
+	ret = fault_in_user_writeable(uaddr);
 	if (!ret)
 		goto retry;
 
-- 
cgit v1.2.3-59-g8ed1b


From 641cf4a668e9e69d2bc061e953422ff72a91f86e Mon Sep 17 00:00:00 2001
From: Markus Trippelsdorf <markus@trippelsdorf.de>
Date: Wed, 24 Jun 2009 22:28:52 +0200
Subject: inline functions left without protection of ifdef (acl)

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/posix_acl.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/linux/posix_acl.h b/include/linux/posix_acl.h
index 0cdba01b7756..c513466c7dc7 100644
--- a/include/linux/posix_acl.h
+++ b/include/linux/posix_acl.h
@@ -83,6 +83,7 @@ extern int posix_acl_chmod_masq(struct posix_acl *, mode_t);
 extern struct posix_acl *get_posix_acl(struct inode *, int);
 extern int set_posix_acl(struct inode *, int, struct posix_acl *);
 
+#ifdef CONFIG_FS_POSIX_ACL
 static inline struct posix_acl *get_cached_acl(struct inode *inode, int type)
 {
 	struct posix_acl **p, *acl;
@@ -146,5 +147,5 @@ static inline void forget_cached_acl(struct inode *inode, int type)
 	if (old != ACL_NOT_CACHED)
 		posix_acl_release(old);
 }
-
+#endif
 #endif  /* __LINUX_POSIX_ACL_H */
-- 
cgit v1.2.3-59-g8ed1b


From 3a6a6c16be78472a52f6dd7d88913373b42ad0f7 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Wed, 24 Jun 2009 16:09:01 -0400
Subject: audit: inode watches depend on CONFIG_AUDIT not CONFIG_AUDIT_SYSCALL

Even though one cannot make use of the audit watch code without
CONFIG_AUDIT_SYSCALL the spaghetti nature of the audit code means that
the audit rule filtering requires that it at least be compiled.

Thus build the audit_watch code when we build auditfilter like it was
before cfcad62c74abfef83762dc05a556d21bdf3980a2

Clearly this is a point of potential future cleanup..

Reported-by: Frans Pop <elendil@planet.nl>
Signed-off-by: Eric Paris <eparis@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 kernel/Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/Makefile b/kernel/Makefile
index da750010a6fc..780c8dcf4516 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -69,8 +69,8 @@ obj-$(CONFIG_IKCONFIG) += configs.o
 obj-$(CONFIG_RESOURCE_COUNTERS) += res_counter.o
 obj-$(CONFIG_STOP_MACHINE) += stop_machine.o
 obj-$(CONFIG_KPROBES_SANITY_TEST) += test_kprobes.o
-obj-$(CONFIG_AUDIT) += audit.o auditfilter.o
-obj-$(CONFIG_AUDITSYSCALL) += auditsc.o audit_watch.o
+obj-$(CONFIG_AUDIT) += audit.o auditfilter.o audit_watch.o
+obj-$(CONFIG_AUDITSYSCALL) += auditsc.o
 obj-$(CONFIG_GCOV_KERNEL) += gcov/
 obj-$(CONFIG_AUDIT_TREE) += audit_tree.o
 obj-$(CONFIG_KPROBES) += kprobes.o
-- 
cgit v1.2.3-59-g8ed1b


From 72c04902d1e27c8a324014cff1d4475c11b1cecd Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 24 Jun 2009 16:58:48 -0400
Subject: Get "no acls for this inode" right, fix shmem breakage

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/btrfs/inode.c          | 6 ++----
 fs/jffs2/acl.c            | 3 +--
 include/linux/posix_acl.h | 9 +++++++++
 mm/shmem.c                | 5 +----
 4 files changed, 13 insertions(+), 10 deletions(-)

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 78ad38ddd01f..dbe1aabf96cd 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -2122,10 +2122,8 @@ static void btrfs_read_locked_inode(struct inode *inode)
 	 * any xattrs or acls
 	 */
 	maybe_acls = acls_after_inode_item(leaf, path->slots[0], inode->i_ino);
-	if (!maybe_acls) {
-		inode->i_acl = NULL;
-		inode->i_default_acl = NULL;
-	}
+	if (!maybe_acls)
+		cache_no_acl(inode);
 
 	BTRFS_I(inode)->block_group = btrfs_find_block_group(root, 0,
 						alloc_group_block, 0);
diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c
index edd2ad6416d8..8fcb6239218e 100644
--- a/fs/jffs2/acl.c
+++ b/fs/jffs2/acl.c
@@ -284,8 +284,7 @@ int jffs2_init_acl_pre(struct inode *dir_i, struct inode *inode, int *i_mode)
 	struct posix_acl *acl, *clone;
 	int rc;
 
-	inode->i_default_acl = NULL;
-	inode->i_acl = NULL;
+	cache_no_acl(inode);
 
 	if (S_ISLNK(*i_mode))
 		return 0;	/* Symlink always has no-ACL */
diff --git a/include/linux/posix_acl.h b/include/linux/posix_acl.h
index c513466c7dc7..065a3652a3ea 100644
--- a/include/linux/posix_acl.h
+++ b/include/linux/posix_acl.h
@@ -148,4 +148,13 @@ static inline void forget_cached_acl(struct inode *inode, int type)
 		posix_acl_release(old);
 }
 #endif
+
+static inline void cache_no_acl(struct inode *inode)
+{
+#ifdef CONFIG_FS_POSIX_ACL
+	inode->i_acl = NULL;
+	inode->i_default_acl = NULL;
+#endif
+}
+
 #endif  /* __LINUX_POSIX_ACL_H */
diff --git a/mm/shmem.c b/mm/shmem.c
index 5f2019fc7895..d713239ce2ce 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1558,6 +1558,7 @@ static struct inode *shmem_get_inode(struct super_block *sb, int mode,
 		spin_lock_init(&info->lock);
 		info->flags = flags & VM_NORESERVE;
 		INIT_LIST_HEAD(&info->swaplist);
+		cache_no_acl(inode);
 
 		switch (mode & S_IFMT) {
 		default:
@@ -2379,10 +2380,6 @@ static struct inode *shmem_alloc_inode(struct super_block *sb)
 	p = (struct shmem_inode_info *)kmem_cache_alloc(shmem_inode_cachep, GFP_KERNEL);
 	if (!p)
 		return NULL;
-#ifdef CONFIG_TMPFS_POSIX_ACL
-	p->vfs_inode.i_acl = NULL;
-	p->vfs_inode.i_default_acl = NULL;
-#endif
 	return &p->vfs_inode;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From d5bb68adda7cc179e8efadeaa3a283cb470f13a6 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 24 Jun 2009 17:02:42 -0400
Subject: another race fix in jfs_check_acl()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/jfs/acl.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c
index f272bf032e1e..91fa3ad6e8c2 100644
--- a/fs/jfs/acl.c
+++ b/fs/jfs/acl.c
@@ -118,15 +118,16 @@ out:
 
 static int jfs_check_acl(struct inode *inode, int mask)
 {
-	if (inode->i_acl == ACL_NOT_CACHED) {
-		struct posix_acl *acl = jfs_get_acl(inode, ACL_TYPE_ACCESS);
-		if (IS_ERR(acl))
-			return PTR_ERR(acl);
+	struct posix_acl *acl = jfs_get_acl(inode, ACL_TYPE_ACCESS);
+
+	if (IS_ERR(acl))
+		return PTR_ERR(acl);
+	if (acl) {
+		int error = posix_acl_permission(inode, acl, mask);
 		posix_acl_release(acl);
+		return error;
 	}
 
-	if (inode->i_acl)
-		return posix_acl_permission(inode, inode->i_acl, mask);
 	return -EAGAIN;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 236e946b53ffd5e2f5d7e6abebbe72a9f0826d15 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Wed, 24 Jun 2009 16:23:03 -0700
Subject: Revert "PCI: use ACPI _CRS data by default"

This reverts commit 9e9f46c44e487af0a82eb61b624553e2f7118f5b.

Quoting from the commit message:

 "At this point, it seems to solve more problems than it causes, so let's
  try using it by default.  It's an easy revert if it ends up causing
  trouble."

And guess what? The _CRS code causes trouble.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/kernel-parameters.txt | 2 +-
 arch/x86/include/asm/pci_x86.h      | 2 +-
 arch/x86/pci/acpi.c                 | 2 +-
 arch/x86/pci/amd_bus.c              | 2 +-
 arch/x86/pci/common.c               | 4 ++--
 5 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 040fee607282..d08759aa0903 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1855,7 +1855,7 @@ and is between 256 and 4096 characters. It is defined in the file
 				IRQ routing is enabled.
 		noacpi		[X86] Do not use ACPI for IRQ routing
 				or for PCI scanning.
-		nocrs		[X86] Don't use _CRS for PCI resource
+		use_crs		[X86] Use _CRS for PCI resource
 				allocation.
 		routeirq	Do IRQ routing for all PCI devices.
 				This is normally done in pci_enable_device(),
diff --git a/arch/x86/include/asm/pci_x86.h b/arch/x86/include/asm/pci_x86.h
index d419f5c02669..b399988eee3a 100644
--- a/arch/x86/include/asm/pci_x86.h
+++ b/arch/x86/include/asm/pci_x86.h
@@ -25,7 +25,7 @@
 #define PCI_BIOS_IRQ_SCAN	0x2000
 #define PCI_ASSIGN_ALL_BUSSES	0x4000
 #define PCI_CAN_SKIP_ISA_ALIGN	0x8000
-#define PCI_NO_ROOT_CRS		0x10000
+#define PCI_USE__CRS		0x10000
 #define PCI_CHECK_ENABLE_AMD_MMCONF	0x20000
 #define PCI_HAS_IO_ECS		0x40000
 #define PCI_NOASSIGN_ROMS	0x80000
diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c
index 16c3fda85bba..b26626dc517c 100644
--- a/arch/x86/pci/acpi.c
+++ b/arch/x86/pci/acpi.c
@@ -238,7 +238,7 @@ struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_device *device, int do
 #endif
 	}
 
-	if (bus && !(pci_probe & PCI_NO_ROOT_CRS))
+	if (bus && (pci_probe & PCI_USE__CRS))
 		get_current_resources(device, busnum, domain, bus);
 	return bus;
 }
diff --git a/arch/x86/pci/amd_bus.c b/arch/x86/pci/amd_bus.c
index 2255f880678b..f893d6a6e803 100644
--- a/arch/x86/pci/amd_bus.c
+++ b/arch/x86/pci/amd_bus.c
@@ -101,7 +101,7 @@ void x86_pci_root_bus_res_quirks(struct pci_bus *b)
 	struct pci_root_info *info;
 
 	/* don't go for it if _CRS is used */
-	if (!(pci_probe & PCI_NO_ROOT_CRS))
+	if (pci_probe & PCI_USE__CRS)
 		return;
 
 	/* if only one root bus, don't need to anything */
diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c
index 4740119e4bb7..2202b6257b82 100644
--- a/arch/x86/pci/common.c
+++ b/arch/x86/pci/common.c
@@ -515,8 +515,8 @@ char * __devinit  pcibios_setup(char *str)
 	} else if (!strcmp(str, "assign-busses")) {
 		pci_probe |= PCI_ASSIGN_ALL_BUSSES;
 		return NULL;
-	} else if (!strcmp(str, "nocrs")) {
-		pci_probe |= PCI_NO_ROOT_CRS;
+	} else if (!strcmp(str, "use_crs")) {
+		pci_probe |= PCI_USE__CRS;
 		return NULL;
 	} else if (!strcmp(str, "earlydump")) {
 		pci_early_dump_regs = 1;
-- 
cgit v1.2.3-59-g8ed1b


From 28d0325ce6e0a52f53d8af687e6427fee59004d3 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Wed, 24 Jun 2009 16:25:37 -0700
Subject: Linux 2.6.31-rc1

---
 Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/Makefile b/Makefile
index 46e1c9d03d51..d1216fea0c92 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 2
 PATCHLEVEL = 6
-SUBLEVEL = 30
-EXTRAVERSION =
+SUBLEVEL = 31
+EXTRAVERSION = -rc1
 NAME = Man-Eating Seals of Antiquity
 
 # *DOCUMENTATION*
-- 
cgit v1.2.3-59-g8ed1b


From 268875b9d1dd1bf0b523c59e736da9bc20c8ce1f Mon Sep 17 00:00:00 2001
From: Steve French <sfrench@us.ibm.com>
Date: Thu, 25 Jun 2009 00:29:21 +0000
Subject: [CIFS] Do not send tree disconnect if session is already disconnected

Noticed this when tree connect timed out (due to Samba server crash) -
we try to send a tree disconnect for a tid that does not exist
since we don't have a valid tree id yet. This checks that the
session is valid before sending the tree disconnect to handle
this case.

Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/CHANGES   | 2 +-
 fs/cifs/cifssmb.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES
index b48689839428..3a9b7a58a51d 100644
--- a/fs/cifs/CHANGES
+++ b/fs/cifs/CHANGES
@@ -5,7 +5,7 @@ client generated ones by default (mount option "serverino" turned
 on by default if server supports it).  Add forceuid and forcegid
 mount options (so that when negotiating unix extensions specifying
 which uid mounted does not immediately force the server's reported
-uids to be overridden).
+uids to be overridden).  Add support for scope moutn parm.
 
 Version 1.58
 ------------
diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c
index b84c61d5bca4..58d65a9a79c3 100644
--- a/fs/cifs/cifssmb.c
+++ b/fs/cifs/cifssmb.c
@@ -729,7 +729,7 @@ CIFSSMBTDis(const int xid, struct cifsTconInfo *tcon)
 	 * the tcon is no longer on the list, so no need to take lock before
 	 * checking this.
 	 */
-	if (tcon->need_reconnect)
+	if ((tcon->need_reconnect) || (tcon->ses->need_reconnect))
 		return 0;
 
 	rc = small_smb_init(SMB_COM_TREE_DISCONNECT, 0, tcon,
-- 
cgit v1.2.3-59-g8ed1b


From 245acb87729bc76ba65c7476665c01837e0cdccb Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 24 Jun 2009 03:55:41 -0700
Subject: ipsec: Fix name of CAST algorithm

Our CAST algorithm is called cast5, not cast128.  Clearly nobody
has ever used it :)

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/xfrm/xfrm_algo.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/xfrm/xfrm_algo.c b/net/xfrm/xfrm_algo.c
index d31ccb487730..faf54c6bf96b 100644
--- a/net/xfrm/xfrm_algo.c
+++ b/net/xfrm/xfrm_algo.c
@@ -292,8 +292,8 @@ static struct xfrm_algo_desc ealg_list[] = {
 	}
 },
 {
-	.name = "cbc(cast128)",
-	.compat = "cast128",
+	.name = "cbc(cast5)",
+	.compat = "cast5",
 
 	.uinfo = {
 		.encr = {
-- 
cgit v1.2.3-59-g8ed1b


From 6a9b6546164fb380a019f92ca4d76443202fdc4f Mon Sep 17 00:00:00 2001
From: Florian Fainelli <florian@openwrt.org>
Date: Wed, 24 Jun 2009 16:32:33 -0700
Subject: cpmac: fix compilation failure introduced with netdev_ops conversion

This patch fixes and obvious typo in the netdev_ops initialization:
ndo_so_ioctl should be ndo_do_ioctl.

Signed-off-by: Florian Fainelli <florian@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/cpmac.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/cpmac.c b/drivers/net/cpmac.c
index 58afafbd3b9c..fd5e32cbcb87 100644
--- a/drivers/net/cpmac.c
+++ b/drivers/net/cpmac.c
@@ -1097,7 +1097,7 @@ static const struct net_device_ops cpmac_netdev_ops = {
 	.ndo_start_xmit		= cpmac_start_xmit,
 	.ndo_tx_timeout		= cpmac_tx_timeout,
 	.ndo_set_multicast_list	= cpmac_set_multicast_list,
-	.ndo_so_ioctl		= cpmac_ioctl,
+	.ndo_do_ioctl		= cpmac_ioctl,
 	.ndo_set_config		= cpmac_config,
 	.ndo_change_mtu		= eth_change_mtu,
 	.ndo_validate_addr	= eth_validate_addr,
-- 
cgit v1.2.3-59-g8ed1b


From 681bf72e4893a187cf6b6b62c08fc193f81c8c2f Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Thu, 11 Jun 2009 10:27:31 -0400
Subject: cifs: have cifs parse scope_id out of IPv6 addresses and use it

This patch has CIFS look for a '%' in an IPv6 address. If one is
present then it will try to treat that value as a numeric interface
index suitable for stuffing into the sin6_scope_id field.

This should allow people to mount servers on IPv6 link-local addresses.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Acked-by: David Holder <david@erion.co.uk>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/connect.c     |  6 ++++--
 fs/cifs/dns_resolve.c |  4 ++--
 fs/cifs/netmisc.c     | 32 +++++++++++++++++++++++++++-----
 3 files changed, 33 insertions(+), 9 deletions(-)

diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index c368ad658236..3fb799ff55c9 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -1386,8 +1386,10 @@ cifs_find_tcp_session(struct sockaddr_storage *addr)
 		     server->addr.sockAddr.sin_addr.s_addr))
 			continue;
 		else if (addr->ss_family == AF_INET6 &&
-			 !ipv6_addr_equal(&server->addr.sockAddr6.sin6_addr,
-					  &addr6->sin6_addr))
+			 (!ipv6_addr_equal(&server->addr.sockAddr6.sin6_addr,
+					   &addr6->sin6_addr) ||
+			  server->addr.sockAddr6.sin6_scope_id !=
+					   addr6->sin6_scope_id))
 			continue;
 
 		++server->srv_count;
diff --git a/fs/cifs/dns_resolve.c b/fs/cifs/dns_resolve.c
index 91b5500755bf..87948147d7ec 100644
--- a/fs/cifs/dns_resolve.c
+++ b/fs/cifs/dns_resolve.c
@@ -35,7 +35,7 @@
  * 		0 - name is not IP
  */
 static int
-is_ip(const char *name)
+is_ip(char *name)
 {
 	struct sockaddr_storage ss;
 
@@ -57,7 +57,7 @@ dns_resolver_instantiate(struct key *key, const void *data,
 	ip[datalen] = '\0';
 
 	/* make sure this looks like an address */
-	if (!is_ip((const char *) ip)) {
+	if (!is_ip(ip)) {
 		kfree(ip);
 		return -EINVAL;
 	}
diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c
index f9a54da97d34..bd6d6895730d 100644
--- a/fs/cifs/netmisc.c
+++ b/fs/cifs/netmisc.c
@@ -158,25 +158,47 @@ cifs_inet_pton(const int address_family, const char *cp, void *dst)
 /*
  * Try to convert a string to an IPv4 address and then attempt to convert
  * it to an IPv6 address if that fails. Set the family field if either
- * succeeds.
+ * succeeds. If it's an IPv6 address and it has a '%' sign in it, try to
+ * treat the part following it as a numeric sin6_scope_id.
  *
  * Returns 0 on failure.
  */
 int
 cifs_convert_address(char *src, void *dst)
 {
+	int rc;
+	char *pct, *endp;
 	struct sockaddr_in *s4 = (struct sockaddr_in *) dst;
 	struct sockaddr_in6 *s6 = (struct sockaddr_in6 *) dst;
 
+	/* IPv4 address */
 	if (cifs_inet_pton(AF_INET, src, &s4->sin_addr.s_addr)) {
 		s4->sin_family = AF_INET;
 		return 1;
-	} else if (cifs_inet_pton(AF_INET6, src, &s6->sin6_addr.s6_addr)) {
-		s6->sin6_family = AF_INET6;
-		return 1;
 	}
 
-	return 0;
+	/* temporarily terminate string */
+	pct = strchr(src, '%');
+	if (pct)
+		*pct = '\0';
+
+	rc = cifs_inet_pton(AF_INET6, src, &s6->sin6_addr.s6_addr);
+
+	/* repair temp termination (if any) and make pct point to scopeid */
+	if (pct)
+		*pct++ = '%';
+
+	if (!rc)
+		return rc;
+
+	s6->sin6_family = AF_INET6;
+	if (pct) {
+		s6->sin6_scope_id = (u32) simple_strtoul(pct, &endp, 0);
+		if (!*pct || *endp)
+			return 0;
+	}
+
+	return rc;
 }
 
 /*****************************************************************************
-- 
cgit v1.2.3-59-g8ed1b


From b48a485884b5afb3e33b1871bcbd246b67491923 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Thu, 25 Jun 2009 00:56:54 -0400
Subject: cifs: fix problems with earlier patches

cifs: fix problems with earlier patches

cifs_show_address hasn't been introduced yet, and fix a typo that was
silently fixed by a later patch in the series.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/cifsfs.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index ddef913ff3e6..b5e9f398c2e5 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -374,8 +374,6 @@ cifs_show_options(struct seq_file *s, struct vfsmount *m)
 	if (tcon->ses->domainName)
 		seq_printf(s, ",domain=%s", tcon->ses->domainName);
 
-	cifs_show_address(s, tcon->ses->server);
-
 	seq_printf(s, ",uid=%d", cifs_sb->mnt_uid);
 	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID)
 		seq_printf(s, ",forceuid");
-- 
cgit v1.2.3-59-g8ed1b


From 6459340cfcc6f6d165b27c3dd955aeb55a1b73d3 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Thu, 25 Jun 2009 00:56:55 -0400
Subject: cifs: remove rw/ro options

cifs: remove rw/ro options

These options are handled at the VFS layer. They only ever set the
option in the smb_vol struct. Nothing was ever done with them afterward
anyway.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/connect.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 3fb799ff55c9..a581cfa2ba82 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -70,7 +70,6 @@ struct smb_vol {
 	mode_t file_mode;
 	mode_t dir_mode;
 	unsigned secFlg;
-	bool rw:1;
 	bool retry:1;
 	bool intr:1;
 	bool setuids:1;
@@ -832,7 +831,6 @@ cifs_parse_mount_options(char *options, const char *devname,
 	vol->dir_mode = vol->file_mode = S_IRUGO | S_IXUGO | S_IWUSR;
 
 	/* vol->retry default is 0 (i.e. "soft" limited retry not hard retry) */
-	vol->rw = true;
 	/* default is always to request posix paths. */
 	vol->posix_paths = 1;
 	/* default to using server inode numbers where available */
@@ -1198,8 +1196,6 @@ cifs_parse_mount_options(char *options, const char *devname,
 			/* ignore */
 		} else if (strnicmp(data, "guest", 5) == 0) {
 			/* ignore */
-		} else if (strnicmp(data, "rw", 2) == 0) {
-			vol->rw = true;
 		} else if (strnicmp(data, "noblocksend", 11) == 0) {
 			vol->noblocksnd = 1;
 		} else if (strnicmp(data, "noautotune", 10) == 0) {
@@ -1218,8 +1214,6 @@ cifs_parse_mount_options(char *options, const char *devname,
 			    parse these options again and set anything and it
 			    is ok to just ignore them */
 			continue;
-		} else if (strnicmp(data, "ro", 2) == 0) {
-			vol->rw = false;
 		} else if (strnicmp(data, "hard", 4) == 0) {
 			vol->retry = 1;
 		} else if (strnicmp(data, "soft", 4) == 0) {
-- 
cgit v1.2.3-59-g8ed1b


From 6debdbc0ba6253ac519cd5a3d22e30f1f9f1dd12 Mon Sep 17 00:00:00 2001
From: Simo Leone <simo@archlinux.org>
Date: Thu, 25 Jun 2009 02:44:43 +0000
Subject: [CIFS] Copy struct *after* setting the port, instead of before.

Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Simo Leone <simo@archlinux.org>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/connect.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index a581cfa2ba82..12c2cf693555 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -1496,14 +1496,14 @@ cifs_get_tcp_session(struct smb_vol *volume_info)
 		cFYI(1, ("attempting ipv6 connect"));
 		/* BB should we allow ipv6 on port 139? */
 		/* other OS never observed in Wild doing 139 with v6 */
+		sin_server6->sin6_port = htons(volume_info->port);
 		memcpy(&tcp_ses->addr.sockAddr6, sin_server6,
 			sizeof(struct sockaddr_in6));
-		sin_server6->sin6_port = htons(volume_info->port);
 		rc = ipv6_connect(tcp_ses);
 	} else {
+		sin_server->sin_port = htons(volume_info->port);
 		memcpy(&tcp_ses->addr.sockAddr, sin_server,
 			sizeof(struct sockaddr_in));
-		sin_server->sin_port = htons(volume_info->port);
 		rc = ipv4_connect(tcp_ses);
 	}
 	if (rc < 0) {
-- 
cgit v1.2.3-59-g8ed1b


From f46c7234e472ceee39afea4fb5a4365843e1850a Mon Sep 17 00:00:00 2001
From: Steve French <sfrench@us.ibm.com>
Date: Thu, 25 Jun 2009 03:04:20 +0000
Subject: [CIFS] cleanup asn handling for ntlmssp

Also removes obsolete distinction between rawntlmssp and ntlmssp (in asn/SPNEGO)
since as jra noted we can always send raw ntlmssp in session setup now.

remove check for experimental runtime flag (/proc/fs/cifs/Experimental) in
ntlmssp path.

Reviewed-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/asn1.c     | 55 ++++++++++++++++++++++++++++++++++++++++++++++++------
 fs/cifs/cifsglob.h |  2 +-
 fs/cifs/cifssmb.c  |  2 +-
 fs/cifs/sess.c     |  2 +-
 4 files changed, 52 insertions(+), 9 deletions(-)

diff --git a/fs/cifs/asn1.c b/fs/cifs/asn1.c
index 1b09f1670061..20692fbfdb24 100644
--- a/fs/cifs/asn1.c
+++ b/fs/cifs/asn1.c
@@ -49,6 +49,7 @@
 #define ASN1_OJI	6	/* Object Identifier  */
 #define ASN1_OJD	7	/* Object Description */
 #define ASN1_EXT	8	/* External */
+#define ASN1_ENUM	10	/* Enumerated */
 #define ASN1_SEQ	16	/* Sequence */
 #define ASN1_SET	17	/* Set */
 #define ASN1_NUMSTR	18	/* Numerical String */
@@ -78,10 +79,12 @@
 #define SPNEGO_OID_LEN 7
 #define NTLMSSP_OID_LEN  10
 #define KRB5_OID_LEN  7
+#define KRB5U2U_OID_LEN  8
 #define MSKRB5_OID_LEN  7
 static unsigned long SPNEGO_OID[7] = { 1, 3, 6, 1, 5, 5, 2 };
 static unsigned long NTLMSSP_OID[10] = { 1, 3, 6, 1, 4, 1, 311, 2, 2, 10 };
 static unsigned long KRB5_OID[7] = { 1, 2, 840, 113554, 1, 2, 2 };
+static unsigned long KRB5U2U_OID[8] = { 1, 2, 840, 113554, 1, 2, 2, 3 };
 static unsigned long MSKRB5_OID[7] = { 1, 2, 840, 48018, 1, 2, 2 };
 
 /*
@@ -122,6 +125,28 @@ asn1_octet_decode(struct asn1_ctx *ctx, unsigned char *ch)
 	return 1;
 }
 
+#if 0 /* will be needed later by spnego decoding/encoding of ntlmssp */
+static unsigned char
+asn1_enum_decode(struct asn1_ctx *ctx, __le32 *val)
+{
+	unsigned char ch;
+
+	if (ctx->pointer >= ctx->end) {
+		ctx->error = ASN1_ERR_DEC_EMPTY;
+		return 0;
+	}
+
+	ch = *(ctx->pointer)++; /* ch has 0xa, ptr points to lenght octet */
+	if ((ch) == ASN1_ENUM)  /* if ch value is ENUM, 0xa */
+		*val = *(++(ctx->pointer)); /* value has enum value */
+	else
+		return 0;
+
+	ctx->pointer++;
+	return 1;
+}
+#endif
+
 static unsigned char
 asn1_tag_decode(struct asn1_ctx *ctx, unsigned int *tag)
 {
@@ -476,10 +501,9 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 	unsigned int cls, con, tag, oidlen, rc;
 	bool use_ntlmssp = false;
 	bool use_kerberos = false;
+	bool use_kerberosu2u = false;
 	bool use_mskerberos = false;
 
-	*secType = NTLM; /* BB eventually make Kerberos or NLTMSSP the default*/
-
 	/* cifs_dump_mem(" Received SecBlob ", security_blob, length); */
 
 	asn1_open(&ctx, security_blob, length);
@@ -515,6 +539,7 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 		return 0;
 	}
 
+	/* SPNEGO */
 	if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) {
 		cFYI(1, ("Error decoding negTokenInit"));
 		return 0;
@@ -526,6 +551,7 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 		return 0;
 	}
 
+	/* negTokenInit */
 	if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) {
 		cFYI(1, ("Error decoding negTokenInit"));
 		return 0;
@@ -537,6 +563,7 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 		return 0;
 	}
 
+	/* sequence */
 	if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) {
 		cFYI(1, ("Error decoding 2nd part of negTokenInit"));
 		return 0;
@@ -548,6 +575,7 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 		return 0;
 	}
 
+	/* sequence of */
 	if (asn1_header_decode
 	    (&ctx, &sequence_end, &cls, &con, &tag) == 0) {
 		cFYI(1, ("Error decoding 2nd part of negTokenInit"));
@@ -560,6 +588,7 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 		return 0;
 	}
 
+	/* list of security mechanisms */
 	while (!asn1_eoc_decode(&ctx, sequence_end)) {
 		rc = asn1_header_decode(&ctx, &end, &cls, &con, &tag);
 		if (!rc) {
@@ -576,11 +605,15 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 
 				if (compare_oid(oid, oidlen, MSKRB5_OID,
 						MSKRB5_OID_LEN) &&
-						!use_kerberos)
+						!use_mskerberos)
 					use_mskerberos = true;
+				else if (compare_oid(oid, oidlen, KRB5U2U_OID,
+						     KRB5U2U_OID_LEN) &&
+						     !use_kerberosu2u)
+					use_kerberosu2u = true;
 				else if (compare_oid(oid, oidlen, KRB5_OID,
 						     KRB5_OID_LEN) &&
-						     !use_mskerberos)
+						     !use_kerberos)
 					use_kerberos = true;
 				else if (compare_oid(oid, oidlen, NTLMSSP_OID,
 						     NTLMSSP_OID_LEN))
@@ -593,7 +626,12 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 		}
 	}
 
+	/* mechlistMIC */
 	if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) {
+		/* Check if we have reached the end of the blob, but with
+		   no mechListMic (e.g. NTLMSSP instead of KRB5) */
+		if (ctx.error == ASN1_ERR_DEC_EMPTY)
+			goto decode_negtoken_exit;
 		cFYI(1, ("Error decoding last part negTokenInit exit3"));
 		return 0;
 	} else if ((cls != ASN1_CTX) || (con != ASN1_CON)) {
@@ -602,6 +640,8 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 			 cls, con, tag, end, *end));
 		return 0;
 	}
+
+	/* sequence */
 	if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) {
 		cFYI(1, ("Error decoding last part negTokenInit exit5"));
 		return 0;
@@ -611,6 +651,7 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 			cls, con, tag, end, *end));
 	}
 
+	/* sequence of */
 	if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) {
 		cFYI(1, ("Error decoding last part negTokenInit exit 7"));
 		return 0;
@@ -619,6 +660,8 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 			 cls, con, tag, end, *end));
 		return 0;
 	}
+
+	/* general string */
 	if (asn1_header_decode(&ctx, &end, &cls, &con, &tag) == 0) {
 		cFYI(1, ("Error decoding last part negTokenInit exit9"));
 		return 0;
@@ -630,13 +673,13 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 	}
 	cFYI(1, ("Need to call asn1_octets_decode() function for %s",
 		 ctx.pointer));	/* is this UTF-8 or ASCII? */
-
+decode_negtoken_exit:
 	if (use_kerberos)
 		*secType = Kerberos;
 	else if (use_mskerberos)
 		*secType = MSKerberos;
 	else if (use_ntlmssp)
-		*secType = NTLMSSP;
+		*secType = RawNTLMSSP;
 
 	return 1;
 }
diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h
index a61ab772c6f6..e1225e6ded2f 100644
--- a/fs/cifs/cifsglob.h
+++ b/fs/cifs/cifsglob.h
@@ -83,7 +83,7 @@ enum securityEnum {
 	NTLM,			/* Legacy NTLM012 auth with NTLM hash */
 	NTLMv2,			/* Legacy NTLM auth with NTLMv2 hash */
 	RawNTLMSSP,		/* NTLMSSP without SPNEGO, NTLMv2 hash */
-	NTLMSSP,		/* NTLMSSP via SPNEGO, NTLMv2 hash */
+/*	NTLMSSP, */ /* can use rawNTLMSSP instead of NTLMSSP via SPNEGO */
 	Kerberos,		/* Kerberos via SPNEGO */
 	MSKerberos,		/* MS Kerberos via SPNEGO */
 };
diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c
index 58d65a9a79c3..61007c627497 100644
--- a/fs/cifs/cifssmb.c
+++ b/fs/cifs/cifssmb.c
@@ -594,7 +594,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses)
 	else if (secFlags & CIFSSEC_MAY_KRB5)
 		server->secType = Kerberos;
 	else if (secFlags & CIFSSEC_MAY_NTLMSSP)
-		server->secType = NTLMSSP;
+		server->secType = RawNTLMSSP;
 	else if (secFlags & CIFSSEC_MAY_LANMAN)
 		server->secType = LANMAN;
 /* #ifdef CONFIG_CIFS_EXPERIMENTAL
diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c
index 897a052270f9..7085a6275c4c 100644
--- a/fs/cifs/sess.c
+++ b/fs/cifs/sess.c
@@ -802,7 +802,7 @@ ssetup_ntlmssp_authenticate:
 #endif /* CONFIG_CIFS_UPCALL */
 	} else {
 #ifdef CONFIG_CIFS_EXPERIMENTAL
-		if ((experimEnabled > 1) && (type == RawNTLMSSP)) {
+		if (type == RawNTLMSSP) {
 			if ((pSMB->req.hdr.Flags2 & SMBFLG2_UNICODE) == 0) {
 				cERROR(1, ("NTLMSSP requires Unicode support"));
 				rc = -ENOSYS;
-- 
cgit v1.2.3-59-g8ed1b


From febe8345353e8873e43f2c2c9792d062c770b22b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 25 Jun 2009 14:41:57 +0900
Subject: perf_counter tools: add cpu_relax()/rmb() definitions for sh.

Simple cpu_relax()/rmb() stubs that perf needs, which were inadvertently
omitted from the sh HAVE_PERF_COUNTERS patch.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 tools/perf/perf.h | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index ceb68aa51f7f..f735b6924a2f 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -19,6 +19,16 @@
 #define cpu_relax()	asm volatile("" ::: "memory");
 #endif
 
+#ifdef __sh__
+#include "../../arch/sh/include/asm/unistd.h"
+#if defined(__SH4A__) || defined(__SH5__)
+# define rmb()		asm volatile("synco" ::: "memory")
+#else
+# define rmb()		asm volatile("" ::: "memory")
+#endif
+#define cpu_relax()	asm volatile("" ::: "memory")
+#endif
+
 #include <time.h>
 #include <unistd.h>
 #include <sys/types.h>
-- 
cgit v1.2.3-59-g8ed1b


From 261c2407401ca26fa17f05667ea68f51e12c5303 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Thu, 25 Jun 2009 08:13:40 +0200
Subject: ALSA: hda - Add pin-sense trigger when needed for Realtek codecs

Realtek codecs require the pin-sense trigger call before actually
reading the pin-sense.  Without this, the pin-detection might not be
done accurately.

This patch adds the pin-capability check and issues the trigger call
if required.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/hda/patch_realtek.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 334533197425..98ac24adf39c 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -945,12 +945,13 @@ static void alc_fix_pll_init(struct hda_codec *codec, hda_nid_t nid,
 static void alc_automute_pin(struct hda_codec *codec)
 {
 	struct alc_spec *spec = codec->spec;
-	unsigned int present;
+	unsigned int present, pincap;
 	unsigned int nid = spec->autocfg.hp_pins[0];
 	int i;
 
-	/* need to execute and sync at first */
-	snd_hda_codec_read(codec, nid, 0, AC_VERB_SET_PIN_SENSE, 0);
+	pincap = snd_hda_query_pin_caps(codec, nid);
+	if (pincap & AC_PINCAP_TRIG_REQ) /* need trigger? */
+		snd_hda_codec_read(codec, nid, 0, AC_VERB_SET_PIN_SENSE, 0);
 	present = snd_hda_codec_read(codec, nid, 0,
 				     AC_VERB_GET_PIN_SENSE, 0);
 	spec->jack_present = (present & AC_PINSENSE_PRESENCE) != 0;
@@ -1392,7 +1393,7 @@ static struct hda_verb alc888_fujitsu_xa3530_verbs[] = {
 static void alc_automute_amp(struct hda_codec *codec)
 {
 	struct alc_spec *spec = codec->spec;
-	unsigned int val, mute;
+	unsigned int val, mute, pincap;
 	hda_nid_t nid;
 	int i;
 
@@ -1401,6 +1402,10 @@ static void alc_automute_amp(struct hda_codec *codec)
 		nid = spec->autocfg.hp_pins[i];
 		if (!nid)
 			break;
+		pincap = snd_hda_query_pin_caps(codec, nid);
+		if (pincap & AC_PINCAP_TRIG_REQ) /* need trigger? */
+			snd_hda_codec_read(codec, nid, 0,
+					   AC_VERB_SET_PIN_SENSE, 0);
 		val = snd_hda_codec_read(codec, nid, 0,
 					 AC_VERB_GET_PIN_SENSE, 0);
 		if (val & AC_PINSENSE_PRESENCE) {
-- 
cgit v1.2.3-59-g8ed1b


From 320d592001acbfd76bf856b5370319f144285489 Mon Sep 17 00:00:00 2001
From: Emilio López <buhitoescolar@gmail.com>
Date: Thu, 25 Jun 2009 08:18:44 +0200
Subject: ALSA: hda - Fix acer-aspire-6530g model quirk
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the following bugs of acer-aspire-6530g model with ALC888:
- HP jack to mute all speaker outputs including LFE
- Make digital built-in mic working

Signed-off-by: Emilio López <buhitoescolar@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/hda/patch_realtek.c | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 98ac24adf39c..7ebe5216b4b5 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -1476,6 +1476,10 @@ static struct hda_verb alc888_acer_aspire_4930g_verbs[] = {
 static struct hda_verb alc888_acer_aspire_6530g_verbs[] = {
 /* Bias voltage on for external mic port */
 	{0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN | PIN_VREF80},
+/* Front Mic: set to PIN_IN (empty by default) */
+	{0x12, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN},
+/* Unselect Front Mic by default in input mixer 3 */
+	{0x22, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0xb)},
 /* Enable unsolicited event for HP jack */
 	{0x15, AC_VERB_SET_UNSOLICITED_ENABLE, ALC880_HP_EVENT | AC_USRSP_EN},
 /* Enable speaker output */
@@ -1644,6 +1648,17 @@ static void alc888_acer_aspire_4930g_init_hook(struct hda_codec *codec)
 	alc_automute_amp(codec);
 }
 
+static void alc888_acer_aspire_6530g_init_hook(struct hda_codec *codec)
+{
+	struct alc_spec *spec = codec->spec;
+
+	spec->autocfg.hp_pins[0] = 0x15;
+	spec->autocfg.speaker_pins[0] = 0x14;
+	spec->autocfg.speaker_pins[1] = 0x16;
+	spec->autocfg.speaker_pins[2] = 0x17;
+	alc_automute_amp(codec);
+}
+
 static void alc889_acer_aspire_8930g_init_hook(struct hda_codec *codec)
 {
 	struct alc_spec *spec = codec->spec;
@@ -9322,7 +9337,7 @@ static struct alc_config_preset alc883_presets[] = {
 			ARRAY_SIZE(alc888_2_capture_sources),
 		.input_mux = alc888_acer_aspire_6530_sources,
 		.unsol_event = alc_automute_amp_unsol_event,
-		.init_hook = alc888_acer_aspire_4930g_init_hook,
+		.init_hook = alc888_acer_aspire_6530g_init_hook,
 	},
 	[ALC888_ACER_ASPIRE_8930G] = {
 		.mixers = { alc888_base_mixer,
-- 
cgit v1.2.3-59-g8ed1b


From dde6535686aa4e78e8b85850d1f3fccd8a581622 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Thu, 25 Jun 2009 08:25:35 +0200
Subject: ALSA: hda - Use model=acer-aspire-6530g for Acer Aspire 6930G

For Acer Aspire 6930G (1025:015e), acre-aspire-6530g model matches
obviously better.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/hda/patch_realtek.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 7ebe5216b4b5..2ed514030e75 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -9084,7 +9084,7 @@ static struct snd_pci_quirk alc883_cfg_tbl[] = {
 	SND_PCI_QUIRK(0x1025, 0x0157, "Acer X3200", ALC883_AUTO),
 	SND_PCI_QUIRK(0x1025, 0x0158, "Acer AX1700-U3700A", ALC883_AUTO),
 	SND_PCI_QUIRK(0x1025, 0x015e, "Acer Aspire 6930G",
-		ALC888_ACER_ASPIRE_4930G),
+		ALC888_ACER_ASPIRE_6530G),
 	SND_PCI_QUIRK(0x1025, 0x0166, "Acer Aspire 6530G",
 		ALC888_ACER_ASPIRE_6530G),
 	/* default Acer -- disabled as it causes more problems.
-- 
cgit v1.2.3-59-g8ed1b


From 28d27aae9432c300857722a917be4065c6d7abff Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Wed, 24 Jun 2009 22:13:35 -0700
Subject: sound: Use PCI_VDEVICE

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/oss/kahlua.c          |  2 +-
 sound/pci/atiixp.c          |  8 ++++----
 sound/pci/atiixp_modem.c    |  4 ++--
 sound/pci/au88x0/au8810.c   |  3 +--
 sound/pci/au88x0/au8820.c   |  3 +--
 sound/pci/au88x0/au8830.c   |  3 +--
 sound/pci/cmipci.c          | 10 +++++-----
 sound/pci/cs4281.c          |  2 +-
 sound/pci/cs46xx/cs46xx.c   |  6 +++---
 sound/pci/ens1370.c         |  6 +++---
 sound/pci/es1938.c          |  2 +-
 sound/pci/ice1712/ice1712.c |  2 +-
 sound/pci/ice1712/ice1724.c |  2 +-
 sound/pci/intel8x0.c        | 46 ++++++++++++++++++++++-----------------------
 sound/pci/intel8x0m.c       | 34 ++++++++++++++++-----------------
 sound/pci/mixart/mixart.c   |  2 +-
 sound/pci/nm256/nm256.c     |  6 +++---
 sound/pci/rme32.c           |  9 +++------
 sound/pci/rme96.c           | 12 ++++--------
 sound/pci/sonicvibes.c      |  2 +-
 sound/pci/via82xx.c         |  4 ++--
 sound/pci/via82xx_modem.c   |  2 +-
 sound/pci/ymfpci/ymfpci.c   | 12 ++++++------
 23 files changed, 86 insertions(+), 96 deletions(-)

diff --git a/sound/oss/kahlua.c b/sound/oss/kahlua.c
index c180598f1710..89466b056be7 100644
--- a/sound/oss/kahlua.c
+++ b/sound/oss/kahlua.c
@@ -199,7 +199,7 @@ MODULE_LICENSE("GPL");
  */
 
 static struct pci_device_id id_tbl[] = {
-	{ PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5530_AUDIO, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
+	{ PCI_VDEVICE(CYRIX, PCI_DEVICE_ID_CYRIX_5530_AUDIO), 0 },
 	{ }
 };
 
diff --git a/sound/pci/atiixp.c b/sound/pci/atiixp.c
index 71515ddb4593..d6752dff2a44 100644
--- a/sound/pci/atiixp.c
+++ b/sound/pci/atiixp.c
@@ -287,10 +287,10 @@ struct atiixp {
 /*
  */
 static struct pci_device_id snd_atiixp_ids[] = {
-	{ 0x1002, 0x4341, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* SB200 */
-	{ 0x1002, 0x4361, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* SB300 */
-	{ 0x1002, 0x4370, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* SB400 */
-	{ 0x1002, 0x4382, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* SB600 */
+	{ PCI_VDEVICE(ATI, 0x4341), 0 }, /* SB200 */
+	{ PCI_VDEVICE(ATI, 0x4361), 0 }, /* SB300 */
+	{ PCI_VDEVICE(ATI, 0x4370), 0 }, /* SB400 */
+	{ PCI_VDEVICE(ATI, 0x4382), 0 }, /* SB600 */
 	{ 0, }
 };
 
diff --git a/sound/pci/atiixp_modem.c b/sound/pci/atiixp_modem.c
index c3136cccc559..e7e147bf8eb2 100644
--- a/sound/pci/atiixp_modem.c
+++ b/sound/pci/atiixp_modem.c
@@ -262,8 +262,8 @@ struct atiixp_modem {
 /*
  */
 static struct pci_device_id snd_atiixp_ids[] = {
-	{ 0x1002, 0x434d, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* SB200 */
-	{ 0x1002, 0x4378, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* SB400 */
+	{ PCI_VDEVICE(ATI, 0x434d), 0 }, /* SB200 */
+	{ PCI_VDEVICE(ATI, 0x4378), 0 }, /* SB400 */
 	{ 0, }
 };
 
diff --git a/sound/pci/au88x0/au8810.c b/sound/pci/au88x0/au8810.c
index fce22c7af0ea..c0e8c6b295cb 100644
--- a/sound/pci/au88x0/au8810.c
+++ b/sound/pci/au88x0/au8810.c
@@ -1,8 +1,7 @@
 #include "au8810.h"
 #include "au88x0.h"
 static struct pci_device_id snd_vortex_ids[] = {
-	{PCI_VENDOR_ID_AUREAL, PCI_DEVICE_ID_AUREAL_ADVANTAGE,
-	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1,},
+	{PCI_VDEVICE(AUREAL, PCI_DEVICE_ID_AUREAL_ADVANTAGE), 1,},
 	{0,}
 };
 
diff --git a/sound/pci/au88x0/au8820.c b/sound/pci/au88x0/au8820.c
index d1fbcce07257..a6527330df58 100644
--- a/sound/pci/au88x0/au8820.c
+++ b/sound/pci/au88x0/au8820.c
@@ -1,8 +1,7 @@
 #include "au8820.h"
 #include "au88x0.h"
 static struct pci_device_id snd_vortex_ids[] = {
-	{PCI_VENDOR_ID_AUREAL, PCI_DEVICE_ID_AUREAL_VORTEX_1,
-	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0,},
+	{PCI_VDEVICE(AUREAL, PCI_DEVICE_ID_AUREAL_VORTEX_1), 0,},
 	{0,}
 };
 
diff --git a/sound/pci/au88x0/au8830.c b/sound/pci/au88x0/au8830.c
index d4f2717c14fb..6c702ad4352a 100644
--- a/sound/pci/au88x0/au8830.c
+++ b/sound/pci/au88x0/au8830.c
@@ -1,8 +1,7 @@
 #include "au8830.h"
 #include "au88x0.h"
 static struct pci_device_id snd_vortex_ids[] = {
-	{PCI_VENDOR_ID_AUREAL, PCI_DEVICE_ID_AUREAL_VORTEX_2,
-	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0,},
+	{PCI_VDEVICE(AUREAL, PCI_DEVICE_ID_AUREAL_VORTEX_2), 0,},
 	{0,}
 };
 
diff --git a/sound/pci/cmipci.c b/sound/pci/cmipci.c
index 449fe02f666e..ddcd4a9fd7e6 100644
--- a/sound/pci/cmipci.c
+++ b/sound/pci/cmipci.c
@@ -2797,11 +2797,11 @@ static inline void snd_cmipci_proc_init(struct cmipci *cm) {}
 
 
 static struct pci_device_id snd_cmipci_ids[] = {
-	{PCI_VENDOR_ID_CMEDIA, PCI_DEVICE_ID_CMEDIA_CM8338A, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
-	{PCI_VENDOR_ID_CMEDIA, PCI_DEVICE_ID_CMEDIA_CM8338B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
-	{PCI_VENDOR_ID_CMEDIA, PCI_DEVICE_ID_CMEDIA_CM8738, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
-	{PCI_VENDOR_ID_CMEDIA, PCI_DEVICE_ID_CMEDIA_CM8738B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
-	{PCI_VENDOR_ID_AL, PCI_DEVICE_ID_CMEDIA_CM8738, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+	{PCI_VDEVICE(CMEDIA, PCI_DEVICE_ID_CMEDIA_CM8338A), 0},
+	{PCI_VDEVICE(CMEDIA, PCI_DEVICE_ID_CMEDIA_CM8338B), 0},
+	{PCI_VDEVICE(CMEDIA, PCI_DEVICE_ID_CMEDIA_CM8738), 0},
+	{PCI_VDEVICE(CMEDIA, PCI_DEVICE_ID_CMEDIA_CM8738B), 0},
+	{PCI_VDEVICE(AL, PCI_DEVICE_ID_CMEDIA_CM8738), 0},
 	{0,},
 };
 
diff --git a/sound/pci/cs4281.c b/sound/pci/cs4281.c
index f6286f84a221..e2e0359bb056 100644
--- a/sound/pci/cs4281.c
+++ b/sound/pci/cs4281.c
@@ -495,7 +495,7 @@ struct cs4281 {
 static irqreturn_t snd_cs4281_interrupt(int irq, void *dev_id);
 
 static struct pci_device_id snd_cs4281_ids[] = {
-	{ 0x1013, 0x6005, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },	/* CS4281 */
+	{ PCI_VDEVICE(CIRRUS, 0x6005), 0, },	/* CS4281 */
 	{ 0, }
 };
 
diff --git a/sound/pci/cs46xx/cs46xx.c b/sound/pci/cs46xx/cs46xx.c
index c9b3e3d48cbc..033aec430117 100644
--- a/sound/pci/cs46xx/cs46xx.c
+++ b/sound/pci/cs46xx/cs46xx.c
@@ -65,9 +65,9 @@ module_param_array(mmap_valid, bool, NULL, 0444);
 MODULE_PARM_DESC(mmap_valid, "Support OSS mmap.");
 
 static struct pci_device_id snd_cs46xx_ids[] = {
-        { 0x1013, 0x6001, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },   /* CS4280 */
-        { 0x1013, 0x6003, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },   /* CS4612 */
-        { 0x1013, 0x6004, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },   /* CS4615 */
+	{ PCI_VDEVICE(CIRRUS, 0x6001), 0, },   /* CS4280 */
+	{ PCI_VDEVICE(CIRRUS, 0x6003), 0, },   /* CS4612 */
+	{ PCI_VDEVICE(CIRRUS, 0x6004), 0, },   /* CS4615 */
 	{ 0, }
 };
 
diff --git a/sound/pci/ens1370.c b/sound/pci/ens1370.c
index 18f4d1e98c46..d589bbc516e2 100644
--- a/sound/pci/ens1370.c
+++ b/sound/pci/ens1370.c
@@ -445,11 +445,11 @@ static irqreturn_t snd_audiopci_interrupt(int irq, void *dev_id);
 
 static struct pci_device_id snd_audiopci_ids[] = {
 #ifdef CHIP1370
-	{ 0x1274, 0x5000, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },	/* ES1370 */
+	{ PCI_VDEVICE(ENSONIQ, 0x5000), 0, },	/* ES1370 */
 #endif
 #ifdef CHIP1371
-	{ 0x1274, 0x1371, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },	/* ES1371 */
-	{ 0x1274, 0x5880, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },	/* ES1373 - CT5880 */
+	{ PCI_VDEVICE(ENSONIQ, 0x1371), 0, },	/* ES1371 */
+	{ PCI_VDEVICE(ENSONIQ, 0x5880), 0, },	/* ES1373 - CT5880 */
 	{ 0x1102, 0x8938, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },	/* Ectiva EV1938 */
 #endif
 	{ 0, }
diff --git a/sound/pci/es1938.c b/sound/pci/es1938.c
index fbd2ac09aa34..820318ee62c1 100644
--- a/sound/pci/es1938.c
+++ b/sound/pci/es1938.c
@@ -244,7 +244,7 @@ struct es1938 {
 static irqreturn_t snd_es1938_interrupt(int irq, void *dev_id);
 
 static struct pci_device_id snd_es1938_ids[] = {
-        { 0x125d, 0x1969, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },   /* Solo-1 */
+	{ PCI_VDEVICE(ESS, 0x1969), 0, },   /* Solo-1 */
 	{ 0, }
 };
 
diff --git a/sound/pci/ice1712/ice1712.c b/sound/pci/ice1712/ice1712.c
index 0d0cdbdb4486..cecf1ffeeaaa 100644
--- a/sound/pci/ice1712/ice1712.c
+++ b/sound/pci/ice1712/ice1712.c
@@ -107,7 +107,7 @@ MODULE_PARM_DESC(dxr_enable, "Enable DXR support for Terratec DMX6FIRE.");
 
 
 static const struct pci_device_id snd_ice1712_ids[] = {
-	{ PCI_VENDOR_ID_ICE, PCI_DEVICE_ID_ICE_1712, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },   /* ICE1712 */
+	{ PCI_VDEVICE(ICE, PCI_DEVICE_ID_ICE_1712), 0 },   /* ICE1712 */
 	{ 0, }
 };
 
diff --git a/sound/pci/ice1712/ice1724.c b/sound/pci/ice1712/ice1724.c
index 36ade77cf371..cc84a831eb21 100644
--- a/sound/pci/ice1712/ice1724.c
+++ b/sound/pci/ice1712/ice1724.c
@@ -93,7 +93,7 @@ MODULE_PARM_DESC(model, "Use the given board model.");
 
 /* Both VT1720 and VT1724 have the same PCI IDs */
 static const struct pci_device_id snd_vt1724_ids[] = {
-	{ PCI_VENDOR_ID_ICE, PCI_DEVICE_ID_VT1724, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
+	{ PCI_VDEVICE(ICE, PCI_DEVICE_ID_VT1724), 0 },
 	{ 0, }
 };
 
diff --git a/sound/pci/intel8x0.c b/sound/pci/intel8x0.c
index 8aa5687f392a..171ada535209 100644
--- a/sound/pci/intel8x0.c
+++ b/sound/pci/intel8x0.c
@@ -421,29 +421,29 @@ struct intel8x0 {
 };
 
 static struct pci_device_id snd_intel8x0_ids[] = {
-	{ 0x8086, 0x2415, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* 82801AA */
-	{ 0x8086, 0x2425, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* 82901AB */
-	{ 0x8086, 0x2445, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* 82801BA */
-	{ 0x8086, 0x2485, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* ICH3 */
-	{ 0x8086, 0x24c5, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL_ICH4 }, /* ICH4 */
-	{ 0x8086, 0x24d5, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL_ICH4 }, /* ICH5 */
-	{ 0x8086, 0x25a6, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL_ICH4 }, /* ESB */
-	{ 0x8086, 0x266e, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL_ICH4 }, /* ICH6 */
-	{ 0x8086, 0x27de, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL_ICH4 }, /* ICH7 */
-	{ 0x8086, 0x2698, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL_ICH4 }, /* ESB2 */
-	{ 0x8086, 0x7195, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* 440MX */
-	{ 0x1039, 0x7012, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_SIS },	/* SI7012 */
-	{ 0x10de, 0x01b1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_NFORCE },	/* NFORCE */
-	{ 0x10de, 0x003a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_NFORCE },	/* MCP04 */
-	{ 0x10de, 0x006a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_NFORCE },	/* NFORCE2 */
-	{ 0x10de, 0x0059, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_NFORCE },	/* CK804 */
-	{ 0x10de, 0x008a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_NFORCE },	/* CK8 */
-	{ 0x10de, 0x00da, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_NFORCE },	/* NFORCE3 */
-	{ 0x10de, 0x00ea, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_NFORCE },	/* CK8S */
-	{ 0x10de, 0x026b, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_NFORCE },	/* MCP51 */
-	{ 0x1022, 0x746d, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* AMD8111 */
-	{ 0x1022, 0x7445, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* AMD768 */
-	{ 0x10b9, 0x5455, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_ALI },   /* Ali5455 */
+	{ PCI_VDEVICE(INTEL, 0x2415), DEVICE_INTEL },	/* 82801AA */
+	{ PCI_VDEVICE(INTEL, 0x2425), DEVICE_INTEL },	/* 82901AB */
+	{ PCI_VDEVICE(INTEL, 0x2445), DEVICE_INTEL },	/* 82801BA */
+	{ PCI_VDEVICE(INTEL, 0x2485), DEVICE_INTEL },	/* ICH3 */
+	{ PCI_VDEVICE(INTEL, 0x24c5), DEVICE_INTEL_ICH4 }, /* ICH4 */
+	{ PCI_VDEVICE(INTEL, 0x24d5), DEVICE_INTEL_ICH4 }, /* ICH5 */
+	{ PCI_VDEVICE(INTEL, 0x25a6), DEVICE_INTEL_ICH4 }, /* ESB */
+	{ PCI_VDEVICE(INTEL, 0x266e), DEVICE_INTEL_ICH4 }, /* ICH6 */
+	{ PCI_VDEVICE(INTEL, 0x27de), DEVICE_INTEL_ICH4 }, /* ICH7 */
+	{ PCI_VDEVICE(INTEL, 0x2698), DEVICE_INTEL_ICH4 }, /* ESB2 */
+	{ PCI_VDEVICE(INTEL, 0x7195), DEVICE_INTEL },	/* 440MX */
+	{ PCI_VDEVICE(SI, 0x7012), DEVICE_SIS },	/* SI7012 */
+	{ PCI_VDEVICE(NVIDIA, 0x01b1), DEVICE_NFORCE },	/* NFORCE */
+	{ PCI_VDEVICE(NVIDIA, 0x003a), DEVICE_NFORCE },	/* MCP04 */
+	{ PCI_VDEVICE(NVIDIA, 0x006a), DEVICE_NFORCE },	/* NFORCE2 */
+	{ PCI_VDEVICE(NVIDIA, 0x0059), DEVICE_NFORCE },	/* CK804 */
+	{ PCI_VDEVICE(NVIDIA, 0x008a), DEVICE_NFORCE },	/* CK8 */
+	{ PCI_VDEVICE(NVIDIA, 0x00da), DEVICE_NFORCE },	/* NFORCE3 */
+	{ PCI_VDEVICE(NVIDIA, 0x00ea), DEVICE_NFORCE },	/* CK8S */
+	{ PCI_VDEVICE(NVIDIA, 0x026b), DEVICE_NFORCE },	/* MCP51 */
+	{ PCI_VDEVICE(AMD, 0x746d), DEVICE_INTEL },	/* AMD8111 */
+	{ PCI_VDEVICE(AMD, 0x7445), DEVICE_INTEL },	/* AMD768 */
+	{ PCI_VDEVICE(AL, 0x5455), DEVICE_ALI },   /* Ali5455 */
 	{ 0, }
 };
 
diff --git a/sound/pci/intel8x0m.c b/sound/pci/intel8x0m.c
index 6ec0fc50d6be..9e7d12e7673f 100644
--- a/sound/pci/intel8x0m.c
+++ b/sound/pci/intel8x0m.c
@@ -220,24 +220,24 @@ struct intel8x0m {
 };
 
 static struct pci_device_id snd_intel8x0m_ids[] = {
-	{ 0x8086, 0x2416, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* 82801AA */
-	{ 0x8086, 0x2426, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* 82901AB */
-	{ 0x8086, 0x2446, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* 82801BA */
-	{ 0x8086, 0x2486, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* ICH3 */
-	{ 0x8086, 0x24c6, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL }, /* ICH4 */
-	{ 0x8086, 0x24d6, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL }, /* ICH5 */
-	{ 0x8086, 0x266d, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* ICH6 */
-	{ 0x8086, 0x27dd, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* ICH7 */
-	{ 0x8086, 0x7196, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* 440MX */
-	{ 0x1022, 0x7446, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* AMD768 */
-	{ 0x1039, 0x7013, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_SIS },	/* SI7013 */
-	{ 0x10de, 0x01c1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_NFORCE }, /* NFORCE */
-	{ 0x10de, 0x0069, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_NFORCE }, /* NFORCE2 */
-	{ 0x10de, 0x0089, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_NFORCE }, /* NFORCE2s */
-	{ 0x10de, 0x00d9, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_NFORCE }, /* NFORCE3 */
+	{ PCI_VDEVICE(INTEL, 0x2416), DEVICE_INTEL },	/* 82801AA */
+	{ PCI_VDEVICE(INTEL, 0x2426), DEVICE_INTEL },	/* 82901AB */
+	{ PCI_VDEVICE(INTEL, 0x2446), DEVICE_INTEL },	/* 82801BA */
+	{ PCI_VDEVICE(INTEL, 0x2486), DEVICE_INTEL },	/* ICH3 */
+	{ PCI_VDEVICE(INTEL, 0x24c6), DEVICE_INTEL }, /* ICH4 */
+	{ PCI_VDEVICE(INTEL, 0x24d6), DEVICE_INTEL }, /* ICH5 */
+	{ PCI_VDEVICE(INTEL, 0x266d), DEVICE_INTEL },	/* ICH6 */
+	{ PCI_VDEVICE(INTEL, 0x27dd), DEVICE_INTEL },	/* ICH7 */
+	{ PCI_VDEVICE(INTEL, 0x7196), DEVICE_INTEL },	/* 440MX */
+	{ PCI_VDEVICE(AMD, 0x7446), DEVICE_INTEL },	/* AMD768 */
+	{ PCI_VDEVICE(SI, 0x7013), DEVICE_SIS },	/* SI7013 */
+	{ PCI_VDEVICE(NVIDIA, 0x01c1), DEVICE_NFORCE }, /* NFORCE */
+	{ PCI_VDEVICE(NVIDIA, 0x0069), DEVICE_NFORCE }, /* NFORCE2 */
+	{ PCI_VDEVICE(NVIDIA, 0x0089), DEVICE_NFORCE }, /* NFORCE2s */
+	{ PCI_VDEVICE(NVIDIA, 0x00d9), DEVICE_NFORCE }, /* NFORCE3 */
 #if 0
-	{ 0x1022, 0x746d, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_INTEL },	/* AMD8111 */
-	{ 0x10b9, 0x5455, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_ALI },   /* Ali5455 */
+	{ PCI_VDEVICE(AMD, 0x746d), DEVICE_INTEL },	/* AMD8111 */
+	{ PCI_VDEVICE(AL, 0x5455), DEVICE_ALI },   /* Ali5455 */
 #endif
 	{ 0, }
 };
diff --git a/sound/pci/mixart/mixart.c b/sound/pci/mixart/mixart.c
index 82bc5b9e7629..a83d1968a845 100644
--- a/sound/pci/mixart/mixart.c
+++ b/sound/pci/mixart/mixart.c
@@ -61,7 +61,7 @@ MODULE_PARM_DESC(enable, "Enable Digigram " CARD_NAME " soundcard.");
  */
 
 static struct pci_device_id snd_mixart_ids[] = {
-	{ 0x1057, 0x0003, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, /* MC8240 */
+	{ PCI_VDEVICE(MOTOROLA, 0x0003), 0, }, /* MC8240 */
 	{ 0, }
 };
 
diff --git a/sound/pci/nm256/nm256.c b/sound/pci/nm256/nm256.c
index 522a040855d4..97a0731331a1 100644
--- a/sound/pci/nm256/nm256.c
+++ b/sound/pci/nm256/nm256.c
@@ -263,9 +263,9 @@ struct nm256 {
  * PCI ids
  */
 static struct pci_device_id snd_nm256_ids[] = {
-	{PCI_VENDOR_ID_NEOMAGIC, PCI_DEVICE_ID_NEOMAGIC_NM256AV_AUDIO, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
-	{PCI_VENDOR_ID_NEOMAGIC, PCI_DEVICE_ID_NEOMAGIC_NM256ZX_AUDIO, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
-	{PCI_VENDOR_ID_NEOMAGIC, PCI_DEVICE_ID_NEOMAGIC_NM256XL_PLUS_AUDIO, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+	{PCI_VDEVICE(NEOMAGIC, PCI_DEVICE_ID_NEOMAGIC_NM256AV_AUDIO), 0},
+	{PCI_VDEVICE(NEOMAGIC, PCI_DEVICE_ID_NEOMAGIC_NM256ZX_AUDIO), 0},
+	{PCI_VDEVICE(NEOMAGIC, PCI_DEVICE_ID_NEOMAGIC_NM256XL_PLUS_AUDIO), 0},
 	{0,},
 };
 
diff --git a/sound/pci/rme32.c b/sound/pci/rme32.c
index d7b966e7c4cf..f977dba7cbd0 100644
--- a/sound/pci/rme32.c
+++ b/sound/pci/rme32.c
@@ -227,12 +227,9 @@ struct rme32 {
 };
 
 static struct pci_device_id snd_rme32_ids[] = {
-	{PCI_VENDOR_ID_XILINX_RME, PCI_DEVICE_ID_RME_DIGI32,
-	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0,},
-	{PCI_VENDOR_ID_XILINX_RME, PCI_DEVICE_ID_RME_DIGI32_8,
-	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0,},
-	{PCI_VENDOR_ID_XILINX_RME, PCI_DEVICE_ID_RME_DIGI32_PRO,
-	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0,},
+	{PCI_VDEVICE(XILINX_RME, PCI_DEVICE_ID_RME_DIGI32), 0,},
+	{PCI_VDEVICE(XILINX_RME, PCI_DEVICE_ID_RME_DIGI32_8), 0,},
+	{PCI_VDEVICE(XILINX_RME, PCI_DEVICE_ID_RME_DIGI32_PRO), 0,},
 	{0,}
 };
 
diff --git a/sound/pci/rme96.c b/sound/pci/rme96.c
index 55fb1c131f58..2ba5c0fd55db 100644
--- a/sound/pci/rme96.c
+++ b/sound/pci/rme96.c
@@ -232,14 +232,10 @@ struct rme96 {
 };
 
 static struct pci_device_id snd_rme96_ids[] = {
-	{ PCI_VENDOR_ID_XILINX, PCI_DEVICE_ID_RME_DIGI96,
-	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },
-	{ PCI_VENDOR_ID_XILINX, PCI_DEVICE_ID_RME_DIGI96_8,
-	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },
-	{ PCI_VENDOR_ID_XILINX, PCI_DEVICE_ID_RME_DIGI96_8_PRO,
-	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },
-	{ PCI_VENDOR_ID_XILINX, PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST,
-	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, 
+	{ PCI_VDEVICE(XILINX, PCI_DEVICE_ID_RME_DIGI96), 0, },
+	{ PCI_VDEVICE(XILINX, PCI_DEVICE_ID_RME_DIGI96_8), 0, },
+	{ PCI_VDEVICE(XILINX, PCI_DEVICE_ID_RME_DIGI96_8_PRO), 0, },
+	{ PCI_VDEVICE(XILINX, PCI_DEVICE_ID_RME_DIGI96_8_PAD_OR_PST), 0, },
 	{ 0, }
 };
 
diff --git a/sound/pci/sonicvibes.c b/sound/pci/sonicvibes.c
index 7dc60ad4772e..1f6406c4534d 100644
--- a/sound/pci/sonicvibes.c
+++ b/sound/pci/sonicvibes.c
@@ -243,7 +243,7 @@ struct sonicvibes {
 };
 
 static struct pci_device_id snd_sonic_ids[] = {
-	{ 0x5333, 0xca00, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },
+	{ PCI_VDEVICE(S3, 0xca00), 0, },
         { 0, }
 };
 
diff --git a/sound/pci/via82xx.c b/sound/pci/via82xx.c
index 949fcaf6b70e..acfa4760da49 100644
--- a/sound/pci/via82xx.c
+++ b/sound/pci/via82xx.c
@@ -402,9 +402,9 @@ struct via82xx {
 
 static struct pci_device_id snd_via82xx_ids[] = {
 	/* 0x1106, 0x3058 */
-	{ PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_5, PCI_ANY_ID, PCI_ANY_ID, 0, 0, TYPE_CARD_VIA686, },	/* 686A */
+	{ PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_82C686_5), TYPE_CARD_VIA686, },	/* 686A */
 	/* 0x1106, 0x3059 */
-	{ PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8233_5, PCI_ANY_ID, PCI_ANY_ID, 0, 0, TYPE_CARD_VIA8233, },	/* VT8233 */
+	{ PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_8233_5), TYPE_CARD_VIA8233, },	/* VT8233 */
 	{ 0, }
 };
 
diff --git a/sound/pci/via82xx_modem.c b/sound/pci/via82xx_modem.c
index 0d54e3503c1e..47eb61561dfc 100644
--- a/sound/pci/via82xx_modem.c
+++ b/sound/pci/via82xx_modem.c
@@ -261,7 +261,7 @@ struct via82xx_modem {
 };
 
 static struct pci_device_id snd_via82xx_modem_ids[] = {
-	{ 0x1106, 0x3068, PCI_ANY_ID, PCI_ANY_ID, 0, 0, TYPE_CARD_VIA82XX_MODEM, },
+	{ PCI_VDEVICE(VIA, 0x3068), TYPE_CARD_VIA82XX_MODEM, },
 	{ 0, }
 };
 
diff --git a/sound/pci/ymfpci/ymfpci.c b/sound/pci/ymfpci/ymfpci.c
index 4af66661f9b0..e6b18b90d451 100644
--- a/sound/pci/ymfpci/ymfpci.c
+++ b/sound/pci/ymfpci/ymfpci.c
@@ -67,12 +67,12 @@ module_param_array(rear_switch, bool, NULL, 0444);
 MODULE_PARM_DESC(rear_switch, "Enable shared rear/line-in switch");
 
 static struct pci_device_id snd_ymfpci_ids[] = {
-        { 0x1073, 0x0004, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },   /* YMF724 */
-        { 0x1073, 0x000d, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },   /* YMF724F */
-        { 0x1073, 0x000a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },   /* YMF740 */
-        { 0x1073, 0x000c, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },   /* YMF740C */
-        { 0x1073, 0x0010, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },   /* YMF744 */
-        { 0x1073, 0x0012, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },   /* YMF754 */
+	{ PCI_VDEVICE(YAMAHA, 0x0004), 0, },   /* YMF724 */
+	{ PCI_VDEVICE(YAMAHA, 0x000d), 0, },   /* YMF724F */
+	{ PCI_VDEVICE(YAMAHA, 0x000a), 0, },   /* YMF740 */
+	{ PCI_VDEVICE(YAMAHA, 0x000c), 0, },   /* YMF740C */
+	{ PCI_VDEVICE(YAMAHA, 0x0010), 0, },   /* YMF744 */
+	{ PCI_VDEVICE(YAMAHA, 0x0012), 0, },   /* YMF754 */
 	{ 0, }
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 0d7392e54435476243ce08ba57745ab52d639cbb Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Wed, 24 Jun 2009 23:18:02 -0700
Subject: sound: Use PCI_VDEVICE for CREATIVE and ECTIVA

Here's a patch on top of the others to use CREATIVE and ECTIVA

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/ca0106/ca0106_main.c | 2 +-
 sound/pci/emu10k1/emu10k1.c    | 6 +++---
 sound/pci/emu10k1/emu10k1x.c   | 2 +-
 sound/pci/ens1370.c            | 2 +-
 4 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/sound/pci/ca0106/ca0106_main.c b/sound/pci/ca0106/ca0106_main.c
index 57b992a5c057..f24bf1ecb36d 100644
--- a/sound/pci/ca0106/ca0106_main.c
+++ b/sound/pci/ca0106/ca0106_main.c
@@ -1876,7 +1876,7 @@ static int snd_ca0106_resume(struct pci_dev *pci)
 
 // PCI IDs
 static struct pci_device_id snd_ca0106_ids[] = {
-	{ 0x1102, 0x0007, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },	/* Audigy LS or Live 24bit */
+	{ PCI_VDEVICE(CREATIVE, 0x0007), 0 },	/* Audigy LS or Live 24bit */
 	{ 0, }
 };
 MODULE_DEVICE_TABLE(pci, snd_ca0106_ids);
diff --git a/sound/pci/emu10k1/emu10k1.c b/sound/pci/emu10k1/emu10k1.c
index c7f3b994101c..168af67d938e 100644
--- a/sound/pci/emu10k1/emu10k1.c
+++ b/sound/pci/emu10k1/emu10k1.c
@@ -77,9 +77,9 @@ MODULE_PARM_DESC(subsystem, "Force card subsystem model.");
  * Class 0401: 1102:0008 (rev 00) Subsystem: 1102:1001 -> Audigy2 Value  Model:SB0400
  */
 static struct pci_device_id snd_emu10k1_ids[] = {
-	{ 0x1102, 0x0002, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },	/* EMU10K1 */
-	{ 0x1102, 0x0004, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1 },	/* Audigy */
-	{ 0x1102, 0x0008, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1 },	/* Audigy 2 Value SB0400 */
+	{ PCI_VDEVICE(CREATIVE, 0x0002), 0 },	/* EMU10K1 */
+	{ PCI_VDEVICE(CREATIVE, 0x0004), 1 },	/* Audigy */
+	{ PCI_VDEVICE(CREATIVE, 0x0008), 1 },	/* Audigy 2 Value SB0400 */
 	{ 0, }
 };
 
diff --git a/sound/pci/emu10k1/emu10k1x.c b/sound/pci/emu10k1/emu10k1x.c
index 4d3ad793e98f..36e08bd2b3cc 100644
--- a/sound/pci/emu10k1/emu10k1x.c
+++ b/sound/pci/emu10k1/emu10k1x.c
@@ -1607,7 +1607,7 @@ static void __devexit snd_emu10k1x_remove(struct pci_dev *pci)
 
 // PCI IDs
 static struct pci_device_id snd_emu10k1x_ids[] = {
-	{ 0x1102, 0x0006, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },	/* Dell OEM version (EMU10K1) */
+	{ PCI_VDEVICE(CREATIVE, 0x0006), 0 },	/* Dell OEM version (EMU10K1) */
 	{ 0, }
 };
 MODULE_DEVICE_TABLE(pci, snd_emu10k1x_ids);
diff --git a/sound/pci/ens1370.c b/sound/pci/ens1370.c
index d589bbc516e2..2b82c5c723e1 100644
--- a/sound/pci/ens1370.c
+++ b/sound/pci/ens1370.c
@@ -450,7 +450,7 @@ static struct pci_device_id snd_audiopci_ids[] = {
 #ifdef CHIP1371
 	{ PCI_VDEVICE(ENSONIQ, 0x1371), 0, },	/* ES1371 */
 	{ PCI_VDEVICE(ENSONIQ, 0x5880), 0, },	/* ES1373 - CT5880 */
-	{ 0x1102, 0x8938, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, },	/* Ectiva EV1938 */
+	{ PCI_VDEVICE(ECTIVA, 0x8938), 0, },	/* Ectiva EV1938 */
 #endif
 	{ 0, }
 };
-- 
cgit v1.2.3-59-g8ed1b


From 342ba1039ad7cf464c7927ddf1ddc10d48a3716b Mon Sep 17 00:00:00 2001
From: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Date: Wed, 24 Jun 2009 18:39:09 -0300
Subject: mtd: cmdlineparts: Use 64-bit format when printing a debug message.

Commit 69423d99fc182a81f3c5db3eb5c140acc6fc64be ("[MTD] update internal
API to support 64-bit device size") has changed some structure values
to 64-bit and has not updated this debug message, since it's not built
by default.

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/cmdlinepart.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/cmdlinepart.c b/drivers/mtd/cmdlinepart.c
index 5011fa73f918..1479da6d3aa6 100644
--- a/drivers/mtd/cmdlinepart.c
+++ b/drivers/mtd/cmdlinepart.c
@@ -194,7 +194,7 @@ static struct mtd_partition * newpart(char *s,
 	parts[this_part].name = extra_mem;
 	extra_mem += name_len + 1;
 
-	dbg(("partition %d: name <%s>, offset %x, size %x, mask flags %x\n",
+	dbg(("partition %d: name <%s>, offset %llx, size %llx, mask flags %x\n",
 	     this_part,
 	     parts[this_part].name,
 	     parts[this_part].offset,
-- 
cgit v1.2.3-59-g8ed1b


From ae27a7ab2c74f9c075e03730c5f493163d048c62 Mon Sep 17 00:00:00 2001
From: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Date: Wed, 24 Jun 2009 18:40:46 -0300
Subject: mtd: atmel_nand: Fix typo s/parititions/partitions/

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/atmel_nand.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/nand/atmel_nand.c b/drivers/mtd/nand/atmel_nand.c
index 2802992b39da..20c828ba9405 100644
--- a/drivers/mtd/nand/atmel_nand.c
+++ b/drivers/mtd/nand/atmel_nand.c
@@ -534,7 +534,7 @@ static int __init atmel_nand_probe(struct platform_device *pdev)
 							 &num_partitions);
 
 	if ((!partitions) || (num_partitions == 0)) {
-		printk(KERN_ERR "atmel_nand: No parititions defined, or unsupported device.\n");
+		printk(KERN_ERR "atmel_nand: No partitions defined, or unsupported device.\n");
 		res = ENXIO;
 		goto err_no_partitions;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 7e895cfaad51c862932ea7db0c428761076412e5 Mon Sep 17 00:00:00 2001
From: Tim Blechmann <tim@klingt.org>
Date: Thu, 25 Jun 2009 09:41:46 +0200
Subject: ALSA: lx6464es - configure ethersound io channels

as long as the io channel number is not set by the driver, the card
is not visible from the ethersound network

Signed-off-by: Tim Blechmann <tim@klingt.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/lx6464es/lx6464es.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/sound/pci/lx6464es/lx6464es.c b/sound/pci/lx6464es/lx6464es.c
index 18da2ef04d09..11b8c6514b3d 100644
--- a/sound/pci/lx6464es/lx6464es.c
+++ b/sound/pci/lx6464es/lx6464es.c
@@ -654,13 +654,12 @@ static int __devinit lx_init_ethersound_config(struct lx6464es *chip)
 	int i;
 	u32 orig_conf_es = lx_dsp_reg_read(chip, eReg_CONFES);
 
-	u32 default_conf_es = (64 << IOCR_OUTPUTS_OFFSET) |
+	/* configure 64 io channels */
+	u32 conf_es = (orig_conf_es & CONFES_READ_PART_MASK) |
 		(64 << IOCR_INPUTS_OFFSET) |
+		(64 << IOCR_OUTPUTS_OFFSET) |
 		(FREQ_RATIO_SINGLE_MODE << FREQ_RATIO_OFFSET);
 
-	u32 conf_es = (orig_conf_es & CONFES_READ_PART_MASK)
-		| (default_conf_es & CONFES_WRITE_PART_MASK);
-
 	snd_printdd("->lx_init_ethersound\n");
 
 	chip->freq_ratio = FREQ_RATIO_SINGLE_MODE;
-- 
cgit v1.2.3-59-g8ed1b


From 00e54d087afb3867b0b461aef6c1ff433d0df564 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Thu, 25 Jun 2009 14:05:27 +0800
Subject: ftrace: Remove duplicate newline

Before:
  # echo 'sys_open:traceon:' > set_ftrace_filter
  # echo 'sys_close:traceoff:5' > set_ftrace_filter
  # cat set_ftrace_filter
  #### all functions enabled ####
  sys_open:traceon:unlimited

  sys_close:traceoff:count=0

After:
  # cat set_ftrace_filter
  #### all functions enabled ####
  sys_open:traceon:unlimited
  sys_close:traceoff:count=0

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A4313A7.7030105@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/trace_functions.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c
index 90f134764837..7402144bff21 100644
--- a/kernel/trace/trace_functions.c
+++ b/kernel/trace/trace_functions.c
@@ -302,8 +302,7 @@ ftrace_trace_onoff_print(struct seq_file *m, unsigned long ip,
 	if (count == -1)
 		seq_printf(m, ":unlimited\n");
 	else
-		seq_printf(m, ":count=%ld", count);
-	seq_putc(m, '\n');
+		seq_printf(m, ":count=%ld\n", count);
 
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 1155de47cd66d0c496d5a6fb2223e980ef1285b2 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 25 Jun 2009 14:30:12 +0900
Subject: ring-buffer: Make it generally available

In hunting down the cause for the hwlat_detector ring buffer spew in
my failed -next builds it became obvious that folks are now treating
ring_buffer as something that is generic independent of tracing and thus,
suitable for public driver consumption.

Given that there are only a few minor areas in ring_buffer that have any
reliance on CONFIG_TRACING or CONFIG_FUNCTION_TRACER, provide stubs for
those and make it generally available.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Cc: Jon Masters <jcm@jonmasters.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
LKML-Reference: <20090625053012.GB19944@linux-sh.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/Makefile            |  1 +
 kernel/trace/ring_buffer.c | 11 +++++++++++
 kernel/trace/trace.h       |  7 +++++++
 3 files changed, 19 insertions(+)

diff --git a/kernel/Makefile b/kernel/Makefile
index 0a32cb21ec97..0630e293cd49 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -96,6 +96,7 @@ obj-$(CONFIG_HAVE_GENERIC_DMA_COHERENT) += dma-coherent.o
 obj-$(CONFIG_FUNCTION_TRACER) += trace/
 obj-$(CONFIG_TRACING) += trace/
 obj-$(CONFIG_X86_DS) += trace/
+obj-$(CONFIG_RING_BUFFER) += trace/
 obj-$(CONFIG_SMP) += sched_cpupri.o
 obj-$(CONFIG_SLOW_WORK) += slow-work.o
 obj-$(CONFIG_PERF_COUNTERS) += perf_counter.o
diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 04dac2638258..bf27bb7a63e2 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -1563,6 +1563,8 @@ rb_reserve_next_event(struct ring_buffer_per_cpu *cpu_buffer,
 	return NULL;
 }
 
+#ifdef CONFIG_TRACING
+
 #define TRACE_RECURSIVE_DEPTH 16
 
 static int trace_recursive_lock(void)
@@ -1593,6 +1595,13 @@ static void trace_recursive_unlock(void)
 	current->trace_recursion--;
 }
 
+#else
+
+#define trace_recursive_lock()		(0)
+#define trace_recursive_unlock()	do { } while (0)
+
+#endif
+
 static DEFINE_PER_CPU(int, rb_need_resched);
 
 /**
@@ -3104,6 +3113,7 @@ int ring_buffer_read_page(struct ring_buffer *buffer,
 }
 EXPORT_SYMBOL_GPL(ring_buffer_read_page);
 
+#ifdef CONFIG_TRACING
 static ssize_t
 rb_simple_read(struct file *filp, char __user *ubuf,
 	       size_t cnt, loff_t *ppos)
@@ -3171,6 +3181,7 @@ static __init int rb_init_debugfs(void)
 }
 
 fs_initcall(rb_init_debugfs);
+#endif
 
 #ifdef CONFIG_HOTPLUG_CPU
 static int rb_cpu_notify(struct notifier_block *self,
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 6e735d4771f8..3548ae5cc780 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -597,6 +597,7 @@ print_graph_function(struct trace_iterator *iter)
 
 extern struct pid *ftrace_pid_trace;
 
+#ifdef CONFIG_FUNCTION_TRACER
 static inline int ftrace_trace_task(struct task_struct *task)
 {
 	if (!ftrace_pid_trace)
@@ -604,6 +605,12 @@ static inline int ftrace_trace_task(struct task_struct *task)
 
 	return test_tsk_trace_trace(task);
 }
+#else
+static inline int ftrace_trace_task(struct task_struct *task)
+{
+	return 1;
+}
+#endif
 
 /*
  * trace_iterator_flags is an enumeration that defines bit
-- 
cgit v1.2.3-59-g8ed1b


From a9d9058abab4ac17b79d500506e6c74bd16cecdc Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Thu, 25 Jun 2009 10:16:11 +0100
Subject: kmemleak: Allow the early log buffer to be configurable.

(feature suggested by Sergey Senozhatsky)

Kmemleak needs to track all the memory allocations but some of these
happen before kmemleak is initialised. These are stored in an internal
buffer which may be exceeded in some kernel configurations. This patch
adds a configuration option with a default value of 400 and also removes
the stack dump when the early log buffer is exceeded.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Sergey Senozhatsky <sergey.senozhatsky@mail.by>
---
 Documentation/kmemleak.txt |  4 ++++
 lib/Kconfig.debug          | 12 ++++++++++++
 mm/kmemleak.c              |  5 +++--
 3 files changed, 19 insertions(+), 2 deletions(-)

diff --git a/Documentation/kmemleak.txt b/Documentation/kmemleak.txt
index 0112da3b9ab8..f655308064d7 100644
--- a/Documentation/kmemleak.txt
+++ b/Documentation/kmemleak.txt
@@ -41,6 +41,10 @@ Memory scanning parameters can be modified at run-time by writing to the
 Kmemleak can also be disabled at boot-time by passing "kmemleak=off" on
 the kernel command line.
 
+Memory may be allocated or freed before kmemleak is initialised and
+these actions are stored in an early log buffer. The size of this buffer
+is configured via the CONFIG_DEBUG_KMEMLEAK_EARLY_LOG_SIZE option.
+
 Basic Algorithm
 ---------------
 
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 4c32b1a1a06e..12327b2bb785 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -359,6 +359,18 @@ config DEBUG_KMEMLEAK
 	  In order to access the kmemleak file, debugfs needs to be
 	  mounted (usually at /sys/kernel/debug).
 
+config DEBUG_KMEMLEAK_EARLY_LOG_SIZE
+	int "Maximum kmemleak early log entries"
+	depends on DEBUG_KMEMLEAK
+	range 200 2000
+	default 400
+	help
+	  Kmemleak must track all the memory allocations to avoid
+	  reporting false positives. Since memory may be allocated or
+	  freed before kmemleak is initialised, an early log buffer is
+	  used to store these actions. If kmemleak reports "early log
+	  buffer exceeded", please increase this value.
+
 config DEBUG_KMEMLEAK_TEST
 	tristate "Simple test for the kernel memory leak detector"
 	depends on DEBUG_KMEMLEAK
diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index c96f2c8700aa..17096d1b59b2 100644
--- a/mm/kmemleak.c
+++ b/mm/kmemleak.c
@@ -235,7 +235,7 @@ struct early_log {
 };
 
 /* early logging buffer and current position */
-static struct early_log early_log[200];
+static struct early_log early_log[CONFIG_DEBUG_KMEMLEAK_EARLY_LOG_SIZE];
 static int crt_early_log;
 
 static void kmemleak_disable(void);
@@ -696,7 +696,8 @@ static void log_early(int op_type, const void *ptr, size_t size,
 	struct early_log *log;
 
 	if (crt_early_log >= ARRAY_SIZE(early_log)) {
-		kmemleak_stop("Early log buffer exceeded\n");
+		pr_warning("Early log buffer exceeded\n");
+		kmemleak_disable();
 		return;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 76c64c5e4c47b6d28deb3cae8dfa07a93c2229dc Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@cmpxchg.org>
Date: Wed, 24 Jun 2009 21:08:36 +0200
Subject: perf record: Fix filemap pathname parsing in /proc/pid/maps

Looking backward for the first space from the end of a line in
/proc/pid/maps does not find the start of the pathname of the mapped
file if it contains a space.

Since the only slashes we have in this file occur in the (absolute!)
pathname column of file mappings, looking for the first slash in a
line is a safe method to find the name.

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Stefani Seibold <stefani@seibold.net>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20090624190835.GA25548@cmpxchg.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-record.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index d7ebbd757543..9b899ba1b410 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -306,12 +306,11 @@ static void pid_synthesize_mmap_samples(pid_t pid)
 			continue;
 		pbf += n + 3;
 		if (*pbf == 'x') { /* vm_exec */
-			char *execname = strrchr(bf, ' ');
+			char *execname = strchr(bf, '/');
 
-			if (execname == NULL || execname[1] != '/')
+			if (execname == NULL)
 				continue;
 
-			execname += 1;
 			size = strlen(execname);
 			execname[size - 1] = '\0'; /* Remove \n */
 			memcpy(mmap_ev.filename, execname, size);
-- 
cgit v1.2.3-59-g8ed1b


From 11687a1099583273a8a98ec42af62b5bb5a69e45 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 25 Jun 2009 02:45:42 -0700
Subject: Revert "veth: prevent oops caused by netdev destructor"

This reverts commit ae0e8e82205c903978a79ebf5e31c670b61fa5b4.

This change had two problems:

1) Since it frees the stats in the drivers' close method, we
   can OOPS in the transmit routine.

2) stats are no longer remembered across ifdown/ifup which
   disagrees with how every other device operates.

Thanks to analysis and test patch from Serge E. Hallyn
and initial OOPS report by Sachin Sant.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/veth.c | 41 +++++++++++++++++++++++++----------------
 1 file changed, 25 insertions(+), 16 deletions(-)

diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 87197dd9c788..1097c72e44d5 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -208,11 +208,14 @@ rx_drop:
 
 static struct net_device_stats *veth_get_stats(struct net_device *dev)
 {
-	struct veth_priv *priv = netdev_priv(dev);
-	struct net_device_stats *dev_stats = &dev->stats;
-	unsigned int cpu;
+	struct veth_priv *priv;
+	struct net_device_stats *dev_stats;
+	int cpu;
 	struct veth_net_stats *stats;
 
+	priv = netdev_priv(dev);
+	dev_stats = &dev->stats;
+
 	dev_stats->rx_packets = 0;
 	dev_stats->tx_packets = 0;
 	dev_stats->rx_bytes = 0;
@@ -220,17 +223,16 @@ static struct net_device_stats *veth_get_stats(struct net_device *dev)
 	dev_stats->tx_dropped = 0;
 	dev_stats->rx_dropped = 0;
 
-	if (priv->stats)
-		for_each_online_cpu(cpu) {
-			stats = per_cpu_ptr(priv->stats, cpu);
+	for_each_online_cpu(cpu) {
+		stats = per_cpu_ptr(priv->stats, cpu);
 
-			dev_stats->rx_packets += stats->rx_packets;
-			dev_stats->tx_packets += stats->tx_packets;
-			dev_stats->rx_bytes += stats->rx_bytes;
-			dev_stats->tx_bytes += stats->tx_bytes;
-			dev_stats->tx_dropped += stats->tx_dropped;
-			dev_stats->rx_dropped += stats->rx_dropped;
-		}
+		dev_stats->rx_packets += stats->rx_packets;
+		dev_stats->tx_packets += stats->tx_packets;
+		dev_stats->rx_bytes += stats->rx_bytes;
+		dev_stats->tx_bytes += stats->tx_bytes;
+		dev_stats->tx_dropped += stats->tx_dropped;
+		dev_stats->rx_dropped += stats->rx_dropped;
+	}
 
 	return dev_stats;
 }
@@ -257,8 +259,6 @@ static int veth_close(struct net_device *dev)
 	netif_carrier_off(dev);
 	netif_carrier_off(priv->peer);
 
-	free_percpu(priv->stats);
-	priv->stats = NULL;
 	return 0;
 }
 
@@ -289,6 +289,15 @@ static int veth_dev_init(struct net_device *dev)
 	return 0;
 }
 
+static void veth_dev_free(struct net_device *dev)
+{
+	struct veth_priv *priv;
+
+	priv = netdev_priv(dev);
+	free_percpu(priv->stats);
+	free_netdev(dev);
+}
+
 static const struct net_device_ops veth_netdev_ops = {
 	.ndo_init            = veth_dev_init,
 	.ndo_open            = veth_open,
@@ -306,7 +315,7 @@ static void veth_setup(struct net_device *dev)
 	dev->netdev_ops = &veth_netdev_ops;
 	dev->ethtool_ops = &veth_ethtool_ops;
 	dev->features |= NETIF_F_LLTX;
-	dev->destructor = free_netdev;
+	dev->destructor = veth_dev_free;
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From c7a1a4c80f873d5d6ecd173035bb80eba489f380 Mon Sep 17 00:00:00 2001
From: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Date: Wed, 24 Jun 2009 01:07:44 +0000
Subject: Phonet: publicize the Netlink notification function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/phonet/pn_dev.h | 1 +
 net/phonet/pn_netlink.c     | 4 ++--
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/include/net/phonet/pn_dev.h b/include/net/phonet/pn_dev.h
index 5054dc5ea2c2..29d126736611 100644
--- a/include/net/phonet/pn_dev.h
+++ b/include/net/phonet/pn_dev.h
@@ -45,6 +45,7 @@ int phonet_address_add(struct net_device *dev, u8 addr);
 int phonet_address_del(struct net_device *dev, u8 addr);
 u8 phonet_address_get(struct net_device *dev, u8 addr);
 int phonet_address_lookup(struct net *net, u8 addr);
+void phonet_address_notify(int event, struct net_device *dev, u8 addr);
 
 #define PN_NO_ADDR	0xff
 
diff --git a/net/phonet/pn_netlink.c b/net/phonet/pn_netlink.c
index cec4e5951681..f8b4cee434c2 100644
--- a/net/phonet/pn_netlink.c
+++ b/net/phonet/pn_netlink.c
@@ -32,7 +32,7 @@
 static int fill_addr(struct sk_buff *skb, struct net_device *dev, u8 addr,
 		     u32 pid, u32 seq, int event);
 
-static void rtmsg_notify(int event, struct net_device *dev, u8 addr)
+void phonet_address_notify(int event, struct net_device *dev, u8 addr)
 {
 	struct sk_buff *skb;
 	int err = -ENOBUFS;
@@ -94,7 +94,7 @@ static int addr_doit(struct sk_buff *skb, struct nlmsghdr *nlh, void *attr)
 	else
 		err = phonet_address_del(dev, pnaddr);
 	if (!err)
-		rtmsg_notify(nlh->nlmsg_type, dev, pnaddr);
+		phonet_address_notify(nlh->nlmsg_type, dev, pnaddr);
 	return err;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 2be6fa4c7e5731375cc5e70843a3444293c27514 Mon Sep 17 00:00:00 2001
From: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Date: Wed, 24 Jun 2009 01:07:45 +0000
Subject: Phonet: generate Netlink RTM_DELADDR when destroying a device
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Netlink address deletion events were not sent when a network device
vanished neither when Phonet was unloaded.

Signed-off-by: Rémi Denis-Courmont <remi.denis-courmont@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/phonet/pn_dev.c | 52 +++++++++++++++++++++++++++++++---------------------
 1 file changed, 31 insertions(+), 21 deletions(-)

diff --git a/net/phonet/pn_dev.c b/net/phonet/pn_dev.c
index 80a322d77909..b0d6ddd82a9d 100644
--- a/net/phonet/pn_dev.c
+++ b/net/phonet/pn_dev.c
@@ -69,10 +69,27 @@ static struct phonet_device *__phonet_get(struct net_device *dev)
 	return NULL;
 }
 
-static void __phonet_device_free(struct phonet_device *pnd)
+static void phonet_device_destroy(struct net_device *dev)
 {
-	list_del(&pnd->list);
-	kfree(pnd);
+	struct phonet_device_list *pndevs = phonet_device_list(dev_net(dev));
+	struct phonet_device *pnd;
+
+	ASSERT_RTNL();
+
+	spin_lock_bh(&pndevs->lock);
+	pnd = __phonet_get(dev);
+	if (pnd)
+		list_del(&pnd->list);
+	spin_unlock_bh(&pndevs->lock);
+
+	if (pnd) {
+		u8 addr;
+
+		for (addr = find_first_bit(pnd->addrs, 64); addr < 64;
+			addr = find_next_bit(pnd->addrs, 64, 1+addr))
+			phonet_address_notify(RTM_DELADDR, dev, addr);
+		kfree(pnd);
+	}
 }
 
 struct net_device *phonet_device_get(struct net *net)
@@ -126,8 +143,10 @@ int phonet_address_del(struct net_device *dev, u8 addr)
 	pnd = __phonet_get(dev);
 	if (!pnd || !test_and_clear_bit(addr >> 2, pnd->addrs))
 		err = -EADDRNOTAVAIL;
-	else if (bitmap_empty(pnd->addrs, 64))
-		__phonet_device_free(pnd);
+	else if (bitmap_empty(pnd->addrs, 64)) {
+		list_del(&pnd->list);
+		kfree(pnd);
+	}
 	spin_unlock_bh(&pndevs->lock);
 	return err;
 }
@@ -181,18 +200,8 @@ static int phonet_device_notify(struct notifier_block *me, unsigned long what,
 {
 	struct net_device *dev = arg;
 
-	if (what == NETDEV_UNREGISTER) {
-		struct phonet_device_list *pndevs;
-		struct phonet_device *pnd;
-
-		/* Destroy phonet-specific device data */
-		pndevs = phonet_device_list(dev_net(dev));
-		spin_lock_bh(&pndevs->lock);
-		pnd = __phonet_get(dev);
-		if (pnd)
-			__phonet_device_free(pnd);
-		spin_unlock_bh(&pndevs->lock);
-	}
+	if (what == NETDEV_UNREGISTER)
+		phonet_device_destroy(dev);
 	return 0;
 
 }
@@ -218,11 +227,12 @@ static int phonet_init_net(struct net *net)
 static void phonet_exit_net(struct net *net)
 {
 	struct phonet_net *pnn = net_generic(net, phonet_net_id);
-	struct phonet_device *pnd, *n;
-
-	list_for_each_entry_safe(pnd, n, &pnn->pndevs.list, list)
-		__phonet_device_free(pnd);
+	struct net_device *dev;
 
+	rtnl_lock();
+	for_each_netdev(net, dev)
+		phonet_device_destroy(dev);
+	rtnl_unlock();
 	kfree(pnn);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From d8146bb23ea045fb75c56b4e3b53f0964eed4076 Mon Sep 17 00:00:00 2001
From: Brandon Philips <brandon@ifup.org>
Date: Wed, 24 Jun 2009 14:09:14 +0000
Subject: atl1*: add device_set_wakeup_enable to atl1*_set_wol

Tell PCI core that atl1* device can wakeup the system when WOL is
enabled by calling device_set_wakeup_enable.

Joerg noted that his atl1e device WOL fine after enabling it with
ethtool and changing /sys/class/net/eth0/device/power/wakeup to enabled
Tested on atl1e: https://bugzilla.novell.com/show_bug.cgi?id=493214

Tested by: Joerg Reuter <jreuter@novell.com>
Signed-off-by: Brandon Philips <bphilips@suse.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/atl1c/atl1c_ethtool.c | 2 ++
 drivers/net/atl1e/atl1e_ethtool.c | 2 ++
 2 files changed, 4 insertions(+)

diff --git a/drivers/net/atl1c/atl1c_ethtool.c b/drivers/net/atl1c/atl1c_ethtool.c
index e4afbd628c23..607007d75b6f 100644
--- a/drivers/net/atl1c/atl1c_ethtool.c
+++ b/drivers/net/atl1c/atl1c_ethtool.c
@@ -281,6 +281,8 @@ static int atl1c_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 	if (wol->wolopts & WAKE_PHY)
 		adapter->wol |= AT_WUFC_LNKC;
 
+	device_set_wakeup_enable(&adapter->pdev->dev, adapter->wol);
+
 	return 0;
 }
 
diff --git a/drivers/net/atl1e/atl1e_ethtool.c b/drivers/net/atl1e/atl1e_ethtool.c
index 619c6583e1aa..4003955d7a96 100644
--- a/drivers/net/atl1e/atl1e_ethtool.c
+++ b/drivers/net/atl1e/atl1e_ethtool.c
@@ -365,6 +365,8 @@ static int atl1e_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol)
 	if (wol->wolopts & WAKE_PHY)
 		adapter->wol |= AT_WUFC_LNKC;
 
+	device_set_wakeup_enable(&adapter->pdev->dev, adapter->wol);
+
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 06813f6c743420c16f9248ab59bd2e68a2de57ba Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Thu, 25 Jun 2009 17:16:07 +0530
Subject: perf_counter tools: Check for valid cache operations

Made new table for cache operartion stat 'hw_cache_stat' as:

 L1I : Read and prefetch only
 ITLB and BPU : Read-only

introduce is_cache_op_valid() for cache operation validity

And checks for valid cache operations.

Reported-by : Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <1245930367.5308.33.camel@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/util/parse-events.c | 33 +++++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 06af2fadcd87..7939a21130d2 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -90,6 +90,34 @@ static char *hw_cache_result[][MAX_ALIASES] = {
 	{ "Miss"							},
 };
 
+#define C(x)		PERF_COUNT_HW_CACHE_##x
+#define CACHE_READ	(1 << C(OP_READ))
+#define CACHE_WRITE	(1 << C(OP_WRITE))
+#define CACHE_PREFETCH	(1 << C(OP_PREFETCH))
+#define COP(x)		(1 << x)
+
+/*
+ * cache operartion stat
+ * L1I : Read and prefetch only
+ * ITLB and BPU : Read-only
+ */
+static unsigned long hw_cache_stat[C(MAX)] = {
+ [C(L1D)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(L1I)]	= (CACHE_READ | CACHE_PREFETCH),
+ [C(LL)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(DTLB)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(ITLB)]	= (CACHE_READ),
+ [C(BPU)]	= (CACHE_READ),
+};
+
+static int is_cache_op_valid(u8 cache_type, u8 cache_op)
+{
+	if (hw_cache_stat[cache_type] & COP(cache_op))
+		return 1;	/* valid */
+	else
+		return 0;	/* invalid */
+}
+
 char *event_name(int counter)
 {
 	u64 config = attrs[counter].config;
@@ -123,6 +151,8 @@ char *event_name(int counter)
 		if (cache_result > PERF_COUNT_HW_CACHE_RESULT_MAX)
 			return "unknown-ext-hardware-cache-result";
 
+		if (!is_cache_op_valid(cache_type, cache_op))
+			return "invalid-cache";
 		sprintf(name, "%s-Cache-%s-%ses",
 			hw_cache[cache_type][0],
 			hw_cache_op[cache_op][0],
@@ -179,6 +209,9 @@ parse_generic_hw_symbols(const char *str, struct perf_counter_attr *attr)
 	if (cache_op == -1)
 		cache_op = PERF_COUNT_HW_CACHE_OP_READ;
 
+	if (!is_cache_op_valid(cache_type, cache_op))
+		return -EINVAL;
+
 	cache_result = parse_aliases(str, hw_cache_result,
 					PERF_COUNT_HW_CACHE_RESULT_MAX);
 	/*
-- 
cgit v1.2.3-59-g8ed1b


From 1fbcf37128cc19bd67d9a736fb634dc444e907d7 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Thu, 25 Jun 2009 21:17:19 +0900
Subject: sh: Kill off unused DEBUG_BOOTMEM symbol.

This was killed off in generic code some time ago, kill off the left over
symbol.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig.debug | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/arch/sh/Kconfig.debug b/arch/sh/Kconfig.debug
index 8ece0b5bd028..39224b57c6ef 100644
--- a/arch/sh/Kconfig.debug
+++ b/arch/sh/Kconfig.debug
@@ -61,10 +61,6 @@ config EARLY_PRINTK
 	  select both the EARLY_SCIF_CONSOLE and SH_STANDARD_BIOS, using
 	  the kernel command line option to toggle back and forth.
 
-config DEBUG_BOOTMEM
-	depends on DEBUG_KERNEL
-	bool "Debug BOOTMEM initialization"
-
 config DEBUG_STACKOVERFLOW
 	bool "Check for stack overflows"
 	depends on DEBUG_KERNEL && SUPERH32
-- 
cgit v1.2.3-59-g8ed1b


From aa715284b4d28cabde6c25c568d769a6be712bc8 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 25 Jun 2009 14:27:58 +0200
Subject: futex: request only one page from get_user_pages()

Yanmin noticed that fault_in_user_writeable() requests 4 pages instead
of one.

That's the result of blindly trusting Linus' proposal :) I even looked
up the prototype to verify the correctness: the argument in question
is confusingly enough named "len" while in reality it means number of
pages.

Pointed-out-by: Yanmin Zhang <yanmin_zhang@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 kernel/futex.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/futex.c b/kernel/futex.c
index 1c337112335c..794c862125fe 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -299,7 +299,7 @@ void put_futex_key(int fshared, union futex_key *key)
 static int fault_in_user_writeable(u32 __user *uaddr)
 {
 	int ret = get_user_pages(current, current->mm, (unsigned long)uaddr,
-				 sizeof(*uaddr), 1, 0, NULL, NULL);
+				 1, 1, 0, NULL, NULL);
 	return ret < 0 ? ret : 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From dd5e8e6b1d4c218d2bafe002231ec460459ab5c4 Mon Sep 17 00:00:00 2001
From: Peter Ujfalusi <peter.ujfalusi@nokia.com>
Date: Thu, 25 Jun 2009 13:26:09 +0300
Subject: MAINTAINERS: Add entry for twl4030 series soc codec driver

New MAINTAINERS entry for twl4030 series soc codec driver
with Peter Ujfalusi as maintainer.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@nokia.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
---
 MAINTAINERS | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 1ae5a81f4436..a20ac54f0250 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5523,6 +5523,13 @@ M:	dsaxena@plexity.net
 S:	Maintained
 F:	drivers/char/hw_random/omap-rng.c
 
+TI TWL4030 SERIES SOC CODEC DRIVER
+P:	Peter Ujfalusi
+M:	peter.ujfalusi@nokia.com
+L:	alsa-devel@alsa-project.org (moderated for non-subscribers)
+S:	Maintained
+F:	sound/soc/codecs/twl4030*
+
 TIPC NETWORK LAYER
 P:	Per Liden
 M:	per.liden@ericsson.com
-- 
cgit v1.2.3-59-g8ed1b


From 7aa5514e7170c6179272bc638a980adc1738fd29 Mon Sep 17 00:00:00 2001
From: Aaro Koskinen <aaro.koskinen@nokia.com>
Date: Mon, 22 Jun 2009 09:23:36 +0100
Subject: [ARM] 5560/1: Avoid buffer overrun in case of an invalid IRQ

handle_bad_irq() expects the IRQ number to be valid (used for statistics),
so it cannot be called with an illegal vector. The problem was reported
by a static analysis tool.

The change makes bad_irq_desc redundant, so delete it.

Signed-off-by: Aaro Koskinen <aaro.koskinen@nokia.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/irq.c | 24 ++++++------------------
 1 file changed, 6 insertions(+), 18 deletions(-)

diff --git a/arch/arm/kernel/irq.c b/arch/arm/kernel/irq.c
index 096f600dc8d8..b7c3490eaa24 100644
--- a/arch/arm/kernel/irq.c
+++ b/arch/arm/kernel/irq.c
@@ -98,17 +98,6 @@ unlock:
 	return 0;
 }
 
-/* Handle bad interrupts */
-static struct irq_desc bad_irq_desc = {
-	.handle_irq = handle_bad_irq,
-	.lock = __SPIN_LOCK_UNLOCKED(bad_irq_desc.lock),
-};
-
-#ifdef CONFIG_CPUMASK_OFFSTACK
-/* We are not allocating bad_irq_desc.affinity or .pending_mask */
-#error "ARM architecture does not support CONFIG_CPUMASK_OFFSTACK."
-#endif
-
 /*
  * do_IRQ handles all hardware IRQ's.  Decoded IRQs should not
  * come via this function.  Instead, they should provide their
@@ -124,10 +113,13 @@ asmlinkage void __exception asm_do_IRQ(unsigned int irq, struct pt_regs *regs)
 	 * Some hardware gives randomly wrong interrupts.  Rather
 	 * than crashing, do something sensible.
 	 */
-	if (irq >= NR_IRQS)
-		handle_bad_irq(irq, &bad_irq_desc);
-	else
+	if (unlikely(irq >= NR_IRQS)) {
+		if (printk_ratelimit())
+			printk(KERN_WARNING "Bad IRQ%u\n", irq);
+		ack_bad_irq(irq);
+	} else {
 		generic_handle_irq(irq);
+	}
 
 	/* AT91 specific workaround */
 	irq_finish(irq);
@@ -165,10 +157,6 @@ void __init init_IRQ(void)
 	for (irq = 0; irq < NR_IRQS; irq++)
 		irq_desc[irq].status |= IRQ_NOREQUEST | IRQ_NOPROBE;
 
-#ifdef CONFIG_SMP
-	cpumask_setall(bad_irq_desc.affinity);
-	bad_irq_desc.node = smp_processor_id();
-#endif
 	init_arch_irq();
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From f6430a938dc6d77e33722aaf6a58382b3423935d Mon Sep 17 00:00:00 2001
From: Linus Walleij <linus.walleij@stericsson.com>
Date: Wed, 24 Jun 2009 23:38:56 +0100
Subject: [ARM] 5565/2: Use PAGE_SIZE and RO_DATA() in link script

Update the link script for ARM to use PAGE_SIZE instead of hard-
coded 4096. Also the old RODATA macro is deprecated
for the RO_DATA(PAGE_SIZE) macro. As a consequence the PAGE_SIZE
was changed from (1UL << PAGE_SHIFT) to (_AC(1,UL) << PAGE_SHIFT)
because the linker does not understand the "UL" suffix to numeric
constants.

Signed-off-by: Linus Walleij <linus.walleij@stericsson.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/include/asm/page.h   |  2 +-
 arch/arm/kernel/vmlinux.lds.S | 15 ++++++++-------
 2 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/arch/arm/include/asm/page.h b/arch/arm/include/asm/page.h
index be962c1349c4..9c746af1bf6e 100644
--- a/arch/arm/include/asm/page.h
+++ b/arch/arm/include/asm/page.h
@@ -12,7 +12,7 @@
 
 /* PAGE_SHIFT determines the page size */
 #define PAGE_SHIFT		12
-#define PAGE_SIZE		(1UL << PAGE_SHIFT)
+#define PAGE_SIZE		(_AC(1,UL) << PAGE_SHIFT)
 #define PAGE_MASK		(~(PAGE_SIZE-1))
 
 #ifndef __ASSEMBLY__
diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S
index 4340bf3d2c84..69371028a202 100644
--- a/arch/arm/kernel/vmlinux.lds.S
+++ b/arch/arm/kernel/vmlinux.lds.S
@@ -6,6 +6,7 @@
 #include <asm-generic/vmlinux.lds.h>
 #include <asm/thread_info.h>
 #include <asm/memory.h>
+#include <asm/page.h>
 	
 OUTPUT_ARCH(arm)
 ENTRY(stext)
@@ -63,7 +64,7 @@ SECTIONS
 			usr/built-in.o(.init.ramfs)
 		__initramfs_end = .;
 #endif
-		. = ALIGN(4096);
+		. = ALIGN(PAGE_SIZE);
 		__per_cpu_load = .;
 		__per_cpu_start = .;
 			*(.data.percpu.page_aligned)
@@ -73,7 +74,7 @@ SECTIONS
 #ifndef CONFIG_XIP_KERNEL
 		__init_begin = _stext;
 		INIT_DATA
-		. = ALIGN(4096);
+		. = ALIGN(PAGE_SIZE);
 		__init_end = .;
 #endif
 	}
@@ -118,7 +119,7 @@ SECTIONS
 		*(.got)			/* Global offset table		*/
 	}
 
-	RODATA
+	RO_DATA(PAGE_SIZE)
 
 	_etext = .;			/* End of text and rodata section */
 
@@ -158,17 +159,17 @@ SECTIONS
 		*(.data.init_task)
 
 #ifdef CONFIG_XIP_KERNEL
-		. = ALIGN(4096);
+		. = ALIGN(PAGE_SIZE);
 		__init_begin = .;
 		INIT_DATA
-		. = ALIGN(4096);
+		. = ALIGN(PAGE_SIZE);
 		__init_end = .;
 #endif
 
-		. = ALIGN(4096);
+		. = ALIGN(PAGE_SIZE);
 		__nosave_begin = .;
 		*(.data.nosave)
-		. = ALIGN(4096);
+		. = ALIGN(PAGE_SIZE);
 		__nosave_end = .;
 
 		/*
-- 
cgit v1.2.3-59-g8ed1b


From 14744d7da2e6ab5c6d8e82c84dc280e3c0dd8552 Mon Sep 17 00:00:00 2001
From: Clemens Ladisch <clemens@ladisch.de>
Date: Thu, 25 Jun 2009 14:28:49 +0200
Subject: sound: oxygen: make mic volume control mono

The microphone input and its volume register have only one channel, so
we have to make the corresponding mixer control a mono control.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/oxygen/oxygen_mixer.c | 28 ++++++++++++++++++----------
 1 file changed, 18 insertions(+), 10 deletions(-)

diff --git a/sound/pci/oxygen/oxygen_mixer.c b/sound/pci/oxygen/oxygen_mixer.c
index 304da169bfdc..5401c547c4e3 100644
--- a/sound/pci/oxygen/oxygen_mixer.c
+++ b/sound/pci/oxygen/oxygen_mixer.c
@@ -575,8 +575,10 @@ static int ac97_switch_put(struct snd_kcontrol *ctl,
 static int ac97_volume_info(struct snd_kcontrol *ctl,
 			    struct snd_ctl_elem_info *info)
 {
+	int stereo = (ctl->private_value >> 16) & 1;
+
 	info->type = SNDRV_CTL_ELEM_TYPE_INTEGER;
-	info->count = 2;
+	info->count = stereo ? 2 : 1;
 	info->value.integer.min = 0;
 	info->value.integer.max = 0x1f;
 	return 0;
@@ -587,6 +589,7 @@ static int ac97_volume_get(struct snd_kcontrol *ctl,
 {
 	struct oxygen *chip = ctl->private_data;
 	unsigned int codec = (ctl->private_value >> 24) & 1;
+	int stereo = (ctl->private_value >> 16) & 1;
 	unsigned int index = ctl->private_value & 0xff;
 	u16 reg;
 
@@ -594,7 +597,8 @@ static int ac97_volume_get(struct snd_kcontrol *ctl,
 	reg = oxygen_read_ac97(chip, codec, index);
 	mutex_unlock(&chip->mutex);
 	value->value.integer.value[0] = 31 - (reg & 0x1f);
-	value->value.integer.value[1] = 31 - ((reg >> 8) & 0x1f);
+	if (stereo)
+		value->value.integer.value[1] = 31 - ((reg >> 8) & 0x1f);
 	return 0;
 }
 
@@ -603,6 +607,7 @@ static int ac97_volume_put(struct snd_kcontrol *ctl,
 {
 	struct oxygen *chip = ctl->private_data;
 	unsigned int codec = (ctl->private_value >> 24) & 1;
+	int stereo = (ctl->private_value >> 16) & 1;
 	unsigned int index = ctl->private_value & 0xff;
 	u16 oldreg, newreg;
 	int change;
@@ -612,8 +617,11 @@ static int ac97_volume_put(struct snd_kcontrol *ctl,
 	newreg = oldreg;
 	newreg = (newreg & ~0x1f) |
 		(31 - (value->value.integer.value[0] & 0x1f));
-	newreg = (newreg & ~0x1f00) |
-		((31 - (value->value.integer.value[0] & 0x1f)) << 8);
+	if (stereo)
+		newreg = (newreg & ~0x1f00) |
+			((31 - (value->value.integer.value[1] & 0x1f)) << 8);
+	else
+		newreg = (newreg & ~0x1f00) | ((newreg & 0x1f) << 8);
 	change = newreg != oldreg;
 	if (change)
 		oxygen_write_ac97(chip, codec, index, newreg);
@@ -673,7 +681,7 @@ static int ac97_fp_rec_volume_put(struct snd_kcontrol *ctl,
 		.private_value = ((codec) << 24) | ((invert) << 16) | \
 				 ((bitnr) << 8) | (index), \
 	}
-#define AC97_VOLUME(xname, codec, index) { \
+#define AC97_VOLUME(xname, codec, index, stereo) { \
 		.iface = SNDRV_CTL_ELEM_IFACE_MIXER, \
 		.name = xname, \
 		.access = SNDRV_CTL_ELEM_ACCESS_READWRITE | \
@@ -682,7 +690,7 @@ static int ac97_fp_rec_volume_put(struct snd_kcontrol *ctl,
 		.get = ac97_volume_get, \
 		.put = ac97_volume_put, \
 		.tlv = { .p = ac97_db_scale, }, \
-		.private_value = ((codec) << 24) | (index), \
+		.private_value = ((codec) << 24) | ((stereo) << 16) | (index), \
 	}
 
 static DECLARE_TLV_DB_SCALE(monitor_db_scale, -1000, 1000, 0);
@@ -882,18 +890,18 @@ static const struct {
 };
 
 static const struct snd_kcontrol_new ac97_controls[] = {
-	AC97_VOLUME("Mic Capture Volume", 0, AC97_MIC),
+	AC97_VOLUME("Mic Capture Volume", 0, AC97_MIC, 0),
 	AC97_SWITCH("Mic Capture Switch", 0, AC97_MIC, 15, 1),
 	AC97_SWITCH("Mic Boost (+20dB)", 0, AC97_MIC, 6, 0),
 	AC97_SWITCH("Line Capture Switch", 0, AC97_LINE, 15, 1),
-	AC97_VOLUME("CD Capture Volume", 0, AC97_CD),
+	AC97_VOLUME("CD Capture Volume", 0, AC97_CD, 1),
 	AC97_SWITCH("CD Capture Switch", 0, AC97_CD, 15, 1),
-	AC97_VOLUME("Aux Capture Volume", 0, AC97_AUX),
+	AC97_VOLUME("Aux Capture Volume", 0, AC97_AUX, 1),
 	AC97_SWITCH("Aux Capture Switch", 0, AC97_AUX, 15, 1),
 };
 
 static const struct snd_kcontrol_new ac97_fp_controls[] = {
-	AC97_VOLUME("Front Panel Playback Volume", 1, AC97_HEADPHONE),
+	AC97_VOLUME("Front Panel Playback Volume", 1, AC97_HEADPHONE, 1),
 	AC97_SWITCH("Front Panel Playback Switch", 1, AC97_HEADPHONE, 15, 1),
 	{
 		.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
-- 
cgit v1.2.3-59-g8ed1b


From 308ff823ebd749a94d3b6ac26b95bc0eb114c39e Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <hawk@comx.dk>
Date: Thu, 25 Jun 2009 16:32:52 +0200
Subject: nf_conntrack: Use rcu_barrier()

RCU barriers, rcu_barrier(), is inserted two places.

 In nf_conntrack_expect.c nf_conntrack_expect_fini() before the
 kmem_cache_destroy().  Firstly to make sure the callback to the
 nf_ct_expect_free_rcu() code is still around.  Secondly because I'm
 unsure about the consequence of having in flight
 nf_ct_expect_free_rcu/kmem_cache_free() calls while doing a
 kmem_cache_destroy() slab destroy.

 And in nf_conntrack_extend.c nf_ct_extend_unregister(), inorder to
 wait for completion of callbacks to __nf_ct_ext_free_rcu(), which is
 invoked by __nf_ct_ext_add().  It might be more efficient to call
 rcu_barrier() in nf_conntrack_core.c nf_conntrack_cleanup_net(), but
 thats make it more difficult to read the code (as the callback code
 in located in nf_conntrack_extend.c).

Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 net/netfilter/nf_conntrack_expect.c | 4 +++-
 net/netfilter/nf_conntrack_extend.c | 2 +-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c
index afde8f991646..2032dfe25ca8 100644
--- a/net/netfilter/nf_conntrack_expect.c
+++ b/net/netfilter/nf_conntrack_expect.c
@@ -617,8 +617,10 @@ err1:
 void nf_conntrack_expect_fini(struct net *net)
 {
 	exp_proc_remove(net);
-	if (net_eq(net, &init_net))
+	if (net_eq(net, &init_net)) {
+		rcu_barrier(); /* Wait for call_rcu() before destroy */
 		kmem_cache_destroy(nf_ct_expect_cachep);
+	}
 	nf_ct_free_hashtable(net->ct.expect_hash, net->ct.expect_vmalloc,
 			     nf_ct_expect_hsize);
 }
diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c
index 4b2c769d555f..fef95be334bd 100644
--- a/net/netfilter/nf_conntrack_extend.c
+++ b/net/netfilter/nf_conntrack_extend.c
@@ -186,6 +186,6 @@ void nf_ct_extend_unregister(struct nf_ct_ext_type *type)
 	rcu_assign_pointer(nf_ct_ext_types[type->id], NULL);
 	update_alloc_size(type);
 	mutex_unlock(&nf_ct_ext_type_mutex);
-	synchronize_rcu();
+	rcu_barrier(); /* Wait for completion of call_rcu()'s */
 }
 EXPORT_SYMBOL_GPL(nf_ct_extend_unregister);
-- 
cgit v1.2.3-59-g8ed1b


From e5c59547791f171b280bc4c4b2c3ff171824c1a3 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Thu, 25 Jun 2009 18:25:22 +0530
Subject: perf_counter tools: Shorten names for events

Added new alias for events.

On AMD box:

 $ ./perf stat -e l1d -e l1d-misses -e l1d-write -e l1d-prefetch -e l1d-prefetch-miss -e l1i -e l1i-misses -e l1i-prefetch -e l2 -e l2-misses -e l2-write -e dtlb -e dtlb-misses -e itlb -e itlb-misses -e bpu -e bpu-misses -- ls -lR /usr/include/ > /dev/null

Before :

 Performance counter stats for 'ls -lR /usr/include/':

      248064467  L1-data-Cache-Load-Referencees  (scaled from 23.27%)
        1001433  L1-data-Cache-Load-Misses  (scaled from 23.34%)
         153691  L1-data-Cache-Store-Referencees  (scaled from 23.34%)
         423248  L1-data-Cache-Prefetch-Referencees  (scaled from 23.33%)
         302138  L1-data-Cache-Prefetch-Misses  (scaled from 23.25%)
      251217546  L1-instruction-Cache-Load-Referencees  (scaled from 23.25%)
        5757005  L1-instruction-Cache-Load-Misses  (scaled from 23.23%)
          93435  L1-instruction-Cache-Prefetch-Referencees  (scaled from 23.24%)
        6496073  L2-Cache-Load-Referencees  (scaled from 23.32%)
         609485  L2-Cache-Load-Misses  (scaled from 23.45%)
        6876991  L2-Cache-Store-Referencees  (scaled from 23.71%)
      248922840  Data-TLB-Cache-Load-Referencees  (scaled from 23.94%)
        5828386  Data-TLB-Cache-Load-Misses  (scaled from 24.17%)
      257613506  Instruction-TLB-Cache-Load-Referencees  (scaled from 24.20%)
           6833  Instruction-TLB-Cache-Load-Misses  (scaled from 23.88%)
      109043606  Branch-Cache-Load-Referencees  (scaled from 23.64%)
        5552296  Branch-Cache-Load-Misses  (scaled from 23.42%)

    0.413702461  seconds time elapsed.

After :

 Peformance counter stats for 'ls -lR /usr/include/':

      266590464  L1-d$-loads           (scaled from 23.03%)
        1222273  L1-d$-load-misses     (scaled from 23.58%)
         146204  L1-d$-stores          (scaled from 23.83%)
         406344  L1-d$-prefetches      (scaled from 24.09%)
         283748  L1-d$-prefetch-misses (scaled from 24.10%)
      249650965  L1-i$-loads           (scaled from 23.80%)
        3353961  L1-i$-load-misses     (scaled from 23.82%)
         104599  L1-i$-prefetches      (scaled from 23.68%)
        4836405  LLC-loads             (scaled from 23.67%)
         498214  LLC-load-misses       (scaled from 23.66%)
        4953994  LLC-stores            (scaled from 23.64%)
      243354097  dTLB-loads            (scaled from 23.77%)
        6468584  dTLB-load-misses      (scaled from 23.74%)
      249719549  iTLB-loads            (scaled from 23.25%)
           5060  iTLB-load-misses      (scaled from 23.00%)
      112343016  branch-loads          (scaled from 22.76%)
        5528876  branch-load-misses    (scaled from 22.54%)

    0.427154051  seconds time elapsed.

Reported-by : Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <1245934522.5308.39.camel@hpdv5.satnam>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/util/parse-events.c | 45 ++++++++++++++++++++++++++----------------
 1 file changed, 28 insertions(+), 17 deletions(-)

diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 7939a21130d2..430f06083201 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -71,23 +71,23 @@ static char *sw_event_names[] = {
 #define MAX_ALIASES 8
 
 static char *hw_cache[][MAX_ALIASES] = {
-	{ "L1-data",		"l1-d",		"l1d"			},
-	{ "L1-instruction",	"l1-i",		"l1i"			},
-	{ "L2",			"l2"					},
-	{ "Data-TLB",		"dtlb",		"d-tlb"			},
-	{ "Instruction-TLB",	"itlb",		"i-tlb"			},
-	{ "Branch",		"bpu" ,		"btb",		"bpc"	},
+ { "L1-d$",	"l1-d",		"L1-data",				},
+ { "L1-i$",	"l1-i",		"L1-instruction",			},
+ { "LLC",	"L2"							},
+ { "dTLB",	"d-tlb",	"Data-TLB",				},
+ { "iTLB",	"i-tlb",	"Instruction-TLB",			},
+ { "branch",	"branches",	"bpu",		"btb",		"bpc",	},
 };
 
 static char *hw_cache_op[][MAX_ALIASES] = {
-	{ "Load",		"read"					},
-	{ "Store",		"write"					},
-	{ "Prefetch",		"speculative-read", "speculative-load"	},
+ { "load",	"loads",	"read",					},
+ { "store",	"stores",	"write",				},
+ { "prefetch",	"prefetches",	"speculative-read", "speculative-load",	},
 };
 
 static char *hw_cache_result[][MAX_ALIASES] = {
-	{ "Reference",		"ops",		"access"		},
-	{ "Miss"							},
+ { "refs",	"Reference",	"ops",		"access",		},
+ { "misses",	"miss",							},
 };
 
 #define C(x)		PERF_COUNT_HW_CACHE_##x
@@ -118,6 +118,22 @@ static int is_cache_op_valid(u8 cache_type, u8 cache_op)
 		return 0;	/* invalid */
 }
 
+static char *event_cache_name(u8 cache_type, u8 cache_op, u8 cache_result)
+{
+	static char name[50];
+
+	if (cache_result) {
+		sprintf(name, "%s-%s-%s", hw_cache[cache_type][0],
+			hw_cache_op[cache_op][0],
+			hw_cache_result[cache_result][0]);
+	} else {
+		sprintf(name, "%s-%s", hw_cache[cache_type][0],
+			hw_cache_op[cache_op][1]);
+	}
+
+	return name;
+}
+
 char *event_name(int counter)
 {
 	u64 config = attrs[counter].config;
@@ -137,7 +153,6 @@ char *event_name(int counter)
 
 	case PERF_TYPE_HW_CACHE: {
 		u8 cache_type, cache_op, cache_result;
-		static char name[100];
 
 		cache_type   = (config >>  0) & 0xff;
 		if (cache_type > PERF_COUNT_HW_CACHE_MAX)
@@ -153,12 +168,8 @@ char *event_name(int counter)
 
 		if (!is_cache_op_valid(cache_type, cache_op))
 			return "invalid-cache";
-		sprintf(name, "%s-Cache-%s-%ses",
-			hw_cache[cache_type][0],
-			hw_cache_op[cache_op][0],
-			hw_cache_result[cache_result][0]);
 
-		return name;
+		return event_cache_name(cache_type, cache_op, cache_result);
 	}
 
 	case PERF_TYPE_SOFTWARE:
-- 
cgit v1.2.3-59-g8ed1b


From e08afeb7e69f45e4ab9fbb8530fe433484b96606 Mon Sep 17 00:00:00 2001
From: Brian King <brking@linux.vnet.ibm.com>
Date: Tue, 23 Jun 2009 17:14:01 -0500
Subject: [SCSI] ibmvscsi: Fix module load hang

Fixes a regression seen in the ibmvscsi driver when using the VSCSI
server in SLES 9 and SLES 10. The VSCSI server in these releases
has a bug in it in which it does not send responses to unknown MADs.
Check the OS Type field in the adapter info response and do not send
these unsupported commands when talking to an older server.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/ibmvscsi/ibmvscsi.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/ibmvscsi/ibmvscsi.c b/drivers/scsi/ibmvscsi/ibmvscsi.c
index 869a11bdccbd..9928704e235f 100644
--- a/drivers/scsi/ibmvscsi/ibmvscsi.c
+++ b/drivers/scsi/ibmvscsi/ibmvscsi.c
@@ -1095,9 +1095,14 @@ static void adapter_info_rsp(struct srp_event_struct *evt_struct)
 				MAX_INDIRECT_BUFS);
 			hostdata->host->sg_tablesize = MAX_INDIRECT_BUFS;
 		}
+
+		if (hostdata->madapter_info.os_type == 3) {
+			enable_fast_fail(hostdata);
+			return;
+		}
 	}
 
-	enable_fast_fail(hostdata);
+	send_srp_login(hostdata);
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 87a2d34b0372dcf6bc4caf4d97a7889f5e62a1af Mon Sep 17 00:00:00 2001
From: Roel Kluin <roel.kluin@gmail.com>
Date: Tue, 23 Jun 2009 01:06:40 +0200
Subject: [SCSI] fnic: remove redundant BUG_ONs and fix checks on unsigned

The shost sg tablesize is set to FNIC_MAX_SG_DESC_CNT and fnic uses
scsi_dma_map, so both BUG_ONs can be removed.

scsi_dma_map may return -ENOMEM, sg_count should be int to catch that.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/fnic/fnic_scsi.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/scsi/fnic/fnic_scsi.c b/drivers/scsi/fnic/fnic_scsi.c
index eabf36502856..bfc996971b81 100644
--- a/drivers/scsi/fnic/fnic_scsi.c
+++ b/drivers/scsi/fnic/fnic_scsi.c
@@ -245,7 +245,7 @@ static inline int fnic_queue_wq_copy_desc(struct fnic *fnic,
 					  struct vnic_wq_copy *wq,
 					  struct fnic_io_req *io_req,
 					  struct scsi_cmnd *sc,
-					  u32 sg_count)
+					  int sg_count)
 {
 	struct scatterlist *sg;
 	struct fc_rport *rport = starget_to_rport(scsi_target(sc->device));
@@ -260,9 +260,6 @@ static inline int fnic_queue_wq_copy_desc(struct fnic *fnic,
 	char msg[2];
 
 	if (sg_count) {
-		BUG_ON(sg_count < 0);
-		BUG_ON(sg_count > FNIC_MAX_SG_DESC_CNT);
-
 		/* For each SGE, create a device desc entry */
 		desc = io_req->sgl_list;
 		for_each_sg(scsi_sglist(sc), sg, sg_count, i) {
@@ -344,7 +341,7 @@ int fnic_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *))
 	struct fnic *fnic;
 	struct vnic_wq_copy *wq;
 	int ret;
-	u32 sg_count;
+	int sg_count;
 	unsigned long flags;
 	unsigned long ptr;
 
-- 
cgit v1.2.3-59-g8ed1b


From e3f47cc74bddea8121560026185ede4770170043 Mon Sep 17 00:00:00 2001
From: Abhijeet Joglekar <abjoglek@cisco.com>
Date: Wed, 24 Jun 2009 07:42:25 -0700
Subject: [SCSI] fnic: use DMA_BIT_MASK(nn) instead of deprecated
 DMA_nnBIT_MASK

Robert Love reported warning while building fnic_main.c:
drivers/scsi/fnic/fnic_main.c:478: warning: `DMA_nnBIT_MASK' is deprecated.

Replaced use of DMA_nnBIT_MASK by DMA_BIT_MASK(nn)

Signed-off-by: Abhijeet Joglekar <abjoglek@cisco.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/fnic/fnic_main.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/fnic/fnic_main.c b/drivers/scsi/fnic/fnic_main.c
index a84072865fc2..2c266c01dc5a 100644
--- a/drivers/scsi/fnic/fnic_main.c
+++ b/drivers/scsi/fnic/fnic_main.c
@@ -473,16 +473,16 @@ static int __devinit fnic_probe(struct pci_dev *pdev,
 	 * limitation for the device.  Try 40-bit first, and
 	 * fail to 32-bit.
 	 */
-	err = pci_set_dma_mask(pdev, DMA_40BIT_MASK);
+	err = pci_set_dma_mask(pdev, DMA_BIT_MASK(40));
 	if (err) {
-		err = pci_set_dma_mask(pdev, DMA_32BIT_MASK);
+		err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
 		if (err) {
 			shost_printk(KERN_ERR, fnic->lport->host,
 				     "No usable DMA configuration "
 				     "aborting\n");
 			goto err_out_release_regions;
 		}
-		err = pci_set_consistent_dma_mask(pdev, DMA_32BIT_MASK);
+		err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32));
 		if (err) {
 			shost_printk(KERN_ERR, fnic->lport->host,
 				     "Unable to obtain 32-bit DMA "
@@ -490,7 +490,7 @@ static int __devinit fnic_probe(struct pci_dev *pdev,
 			goto err_out_release_regions;
 		}
 	} else {
-		err = pci_set_consistent_dma_mask(pdev, DMA_40BIT_MASK);
+		err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(40));
 		if (err) {
 			shost_printk(KERN_ERR, fnic->lport->host,
 				     "Unable to obtain 40-bit DMA "
-- 
cgit v1.2.3-59-g8ed1b


From d3a263a8168f78874254ea9da9595cfb0f3e96d7 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@HansenPartnership.com>
Date: Wed, 24 Jun 2009 19:55:22 +0000
Subject: [SCSI] zalon: fix oops on attach failure

I recently discovered on my zalon that if the attachment fails because
of a bus misconfiguration (I scrapped my HVD array, so the card is now
unterminated) then the system oopses.  The reason is that if
ncr_attach() returns NULL (signalling failure) that NULL is passed by
the goto failed straight into ncr_detach() which oopses.

The fix is just to return -ENODEV in this case.

Cc: Stable Tree <stable@kernel.org>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/zalon.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/zalon.c b/drivers/scsi/zalon.c
index 97f3158fa7b5..27e84e4b1fa9 100644
--- a/drivers/scsi/zalon.c
+++ b/drivers/scsi/zalon.c
@@ -134,7 +134,7 @@ zalon_probe(struct parisc_device *dev)
 
 	host = ncr_attach(&zalon7xx_template, unit, &device);
 	if (!host)
-		goto fail;
+		return -ENODEV;
 
 	if (request_irq(dev->irq, ncr53c8xx_intr, IRQF_SHARED, "zalon", host)) {
 	  dev_printk(KERN_ERR, &dev->dev, "irq problem with %d, detaching\n ",
-- 
cgit v1.2.3-59-g8ed1b


From 47749b14e55cd167632f9a27a4fc439e591e5268 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 25 Jun 2009 08:27:14 +0200
Subject: i2c: fix build bug in i2c-designware.c

This build error triggers on x86:

 drivers/built-in.o: In function `i2c_dw_init':
 i2c-designware.c:(.text+0x4e37ca): undefined reference to `clk_get_rate'
 drivers/built-in.o: In function `dw_i2c_probe':
 i2c-designware.c:(.devinit.text+0x51f5e): undefined reference to `clk_get'
 i2c-designware.c:(.devinit.text+0x51f76): undefined reference to `clk_enable'
 i2c-designware.c:(.devinit.text+0x520ff): undefined reference to `clk_disable'
 i2c-designware.c:(.devinit.text+0x52108): undefined reference to `clk_put'

Because this new driver uses the clk_*() facilities which is an
ARM-only thing currently.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/i2c/busses/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index aa87b6a3bbef..8206442fbabd 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -328,6 +328,7 @@ config I2C_DAVINCI
 
 config I2C_DESIGNWARE
 	tristate "Synopsys DesignWare"
+	depends on HAVE_CLK
 	help
 	  If you say yes to this option, support will be included for the
 	  Synopsys DesignWare I2C adapter. Only master mode is supported.
-- 
cgit v1.2.3-59-g8ed1b


From ab420e6d9c2511b862d753b70efb4e979faa0714 Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Thu, 25 Jun 2009 00:17:15 -0700
Subject: UML: Fix some apparent bitrot

UML: Fix some apparent bitrot

- migration of net_device methods into net_device_ops
- dma_sync_single() changes

Signed-off-by: Paul Menage <menage@google.com>
Acked-by: Amerigo Wang <xiyou.wangcong@gmail.com>
--

This version is split from my earlier patch, including just the
portions that ar required for Linus' tree.

Fixes the following compile errors:

include/linux/dma-mapping.h:113: error: redefinition of 'dma_sync_single'
arch/um/include/asm/dma-mapping.h:84: error: previous definition of 'dma_sync_single' was here
include/linux/dma-mapping.h: In function 'dma_sync_single':
include/linux/dma-mapping.h:117: error: implicit declaration of function 'dma_sync_single_for_cpu'
include/linux/dma-mapping.h: At top level:
include/linux/dma-mapping.h:120: error: redefinition of 'dma_sync_sg'
arch/um/include/asm/dma-mapping.h:91: error: previous definition of 'dma_sync_sg' was here
include/linux/dma-mapping.h: In function 'dma_sync_sg':
include/linux/dma-mapping.h:124: error: implicit declaration of function 'dma_sync_sg_for_cpu'

arch/um/drivers/slirp_kern.c: In function 'slirp_init':
arch/um/drivers/slirp_kern.c:35: error: 'struct net_device' has no member named 'init'
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/um/drivers/slip_kern.c       | 1 -
 arch/um/drivers/slirp_kern.c      | 1 -
 arch/um/include/asm/dma-mapping.h | 4 ++--
 3 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/arch/um/drivers/slip_kern.c b/arch/um/drivers/slip_kern.c
index 5ec17563142e..dd2aadc14af0 100644
--- a/arch/um/drivers/slip_kern.c
+++ b/arch/um/drivers/slip_kern.c
@@ -30,7 +30,6 @@ static void slip_init(struct net_device *dev, void *data)
 
 	slip_proto_init(&spri->slip);
 
-	dev->init = NULL;
 	dev->hard_header_len = 0;
 	dev->header_ops = NULL;
 	dev->addr_len = 0;
diff --git a/arch/um/drivers/slirp_kern.c b/arch/um/drivers/slirp_kern.c
index f15a6e7654f3..e376284f0fb7 100644
--- a/arch/um/drivers/slirp_kern.c
+++ b/arch/um/drivers/slirp_kern.c
@@ -32,7 +32,6 @@ void slirp_init(struct net_device *dev, void *data)
 
 	slip_proto_init(&spri->slip);
 
-	dev->init = NULL;
 	dev->hard_header_len = 0;
 	dev->header_ops = NULL;
 	dev->addr_len = 0;
diff --git a/arch/um/include/asm/dma-mapping.h b/arch/um/include/asm/dma-mapping.h
index 90fc708b320e..378de4bbf49f 100644
--- a/arch/um/include/asm/dma-mapping.h
+++ b/arch/um/include/asm/dma-mapping.h
@@ -79,14 +79,14 @@ dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
 }
 
 static inline void
-dma_sync_single(struct device *dev, dma_addr_t dma_handle, size_t size,
+dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size,
 		enum dma_data_direction direction)
 {
 	BUG();
 }
 
 static inline void
-dma_sync_sg(struct device *dev, struct scatterlist *sg, int nelems,
+dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
 	    enum dma_data_direction direction)
 {
 	BUG();
-- 
cgit v1.2.3-59-g8ed1b


From 9d73777e500929b71dcfed16eec05f6760e345a6 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <peterz@infradead.org>
Date: Thu, 25 Jun 2009 11:58:55 +0200
Subject: clarify get_user_pages() prototype

Currently the 4th parameter of get_user_pages() is called len, but its
in pages, not bytes. Rename the thing to nr_pages to avoid future
confusion.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/mm.h |  2 +-
 mm/memory.c        | 26 ++++++++++++--------------
 mm/nommu.c         | 12 +++++-------
 3 files changed, 18 insertions(+), 22 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index d006e93d5c93..ba3a7cb1eaa0 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -826,7 +826,7 @@ extern int make_pages_present(unsigned long addr, unsigned long end);
 extern int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write);
 
 int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
-			unsigned long start, int len, int write, int force,
+			unsigned long start, int nr_pages, int write, int force,
 			struct page **pages, struct vm_area_struct **vmas);
 int get_user_pages_fast(unsigned long start, int nr_pages, int write,
 			struct page **pages);
diff --git a/mm/memory.c b/mm/memory.c
index f46ac18ba231..65216194eb8d 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1207,8 +1207,8 @@ static inline int use_zero_page(struct vm_area_struct *vma)
 
 
 int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
-		     unsigned long start, int len, int flags,
-		struct page **pages, struct vm_area_struct **vmas)
+		     unsigned long start, int nr_pages, int flags,
+		     struct page **pages, struct vm_area_struct **vmas)
 {
 	int i;
 	unsigned int vm_flags = 0;
@@ -1217,7 +1217,7 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 	int ignore = !!(flags & GUP_FLAGS_IGNORE_VMA_PERMISSIONS);
 	int ignore_sigkill = !!(flags & GUP_FLAGS_IGNORE_SIGKILL);
 
-	if (len <= 0)
+	if (nr_pages <= 0)
 		return 0;
 	/* 
 	 * Require read or write permissions.
@@ -1269,7 +1269,7 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 				vmas[i] = gate_vma;
 			i++;
 			start += PAGE_SIZE;
-			len--;
+			nr_pages--;
 			continue;
 		}
 
@@ -1280,7 +1280,7 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 
 		if (is_vm_hugetlb_page(vma)) {
 			i = follow_hugetlb_page(mm, vma, pages, vmas,
-						&start, &len, i, write);
+						&start, &nr_pages, i, write);
 			continue;
 		}
 
@@ -1357,9 +1357,9 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 				vmas[i] = vma;
 			i++;
 			start += PAGE_SIZE;
-			len--;
-		} while (len && start < vma->vm_end);
-	} while (len);
+			nr_pages--;
+		} while (nr_pages && start < vma->vm_end);
+	} while (nr_pages);
 	return i;
 }
 
@@ -1368,7 +1368,7 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
  * @tsk:	task_struct of target task
  * @mm:		mm_struct of target mm
  * @start:	starting user address
- * @len:	number of pages from start to pin
+ * @nr_pages:	number of pages from start to pin
  * @write:	whether pages will be written to by the caller
  * @force:	whether to force write access even if user mapping is
  *		readonly. This will result in the page being COWed even
@@ -1380,7 +1380,7 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
  *		Or NULL if the caller does not require them.
  *
  * Returns number of pages pinned. This may be fewer than the number
- * requested. If len is 0 or negative, returns 0. If no pages
+ * requested. If nr_pages is 0 or negative, returns 0. If no pages
  * were pinned, returns -errno. Each page returned must be released
  * with a put_page() call when it is finished with. vmas will only
  * remain valid while mmap_sem is held.
@@ -1414,7 +1414,7 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
  * See also get_user_pages_fast, for performance critical applications.
  */
 int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
-		unsigned long start, int len, int write, int force,
+		unsigned long start, int nr_pages, int write, int force,
 		struct page **pages, struct vm_area_struct **vmas)
 {
 	int flags = 0;
@@ -1424,9 +1424,7 @@ int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 	if (force)
 		flags |= GUP_FLAGS_FORCE;
 
-	return __get_user_pages(tsk, mm,
-				start, len, flags,
-				pages, vmas);
+	return __get_user_pages(tsk, mm, start, nr_pages, flags, pages, vmas);
 }
 
 EXPORT_SYMBOL(get_user_pages);
diff --git a/mm/nommu.c b/mm/nommu.c
index 2fd2ad5da98e..bf0cc762a7d2 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -173,8 +173,8 @@ unsigned int kobjsize(const void *objp)
 }
 
 int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
-		     unsigned long start, int len, int flags,
-		struct page **pages, struct vm_area_struct **vmas)
+		     unsigned long start, int nr_pages, int flags,
+		     struct page **pages, struct vm_area_struct **vmas)
 {
 	struct vm_area_struct *vma;
 	unsigned long vm_flags;
@@ -189,7 +189,7 @@ int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 	vm_flags  = write ? (VM_WRITE | VM_MAYWRITE) : (VM_READ | VM_MAYREAD);
 	vm_flags &= force ? (VM_MAYREAD | VM_MAYWRITE) : (VM_READ | VM_WRITE);
 
-	for (i = 0; i < len; i++) {
+	for (i = 0; i < nr_pages; i++) {
 		vma = find_vma(mm, start);
 		if (!vma)
 			goto finish_or_fault;
@@ -224,7 +224,7 @@ finish_or_fault:
  * - don't permit access to VMAs that don't support it, such as I/O mappings
  */
 int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
-	unsigned long start, int len, int write, int force,
+	unsigned long start, int nr_pages, int write, int force,
 	struct page **pages, struct vm_area_struct **vmas)
 {
 	int flags = 0;
@@ -234,9 +234,7 @@ int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 	if (force)
 		flags |= GUP_FLAGS_FORCE;
 
-	return __get_user_pages(tsk, mm,
-				start, len, flags,
-				pages, vmas);
+	return __get_user_pages(tsk, mm, start, nr_pages, flags, pages, vmas);
 }
 EXPORT_SYMBOL(get_user_pages);
 
-- 
cgit v1.2.3-59-g8ed1b


From 42dd2aa6496a2e87e496aac5494d2e1d6096c85b Mon Sep 17 00:00:00 2001
From: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Date: Thu, 25 Jun 2009 14:41:24 +0100
Subject: acm: Return ENODEV instead of EINVAL when trying to open ACM device.

This is required, otherwise a user will get a EINVAL while opening a
non-existing device, instead of ENODEV.

This is what I get with this patch applied now instead of an "Invalid
argument".

  cascardo@vespa:~$ cat /dev/ttyACM0
  cat: /dev/ttyACM0: No such device
  cascardo@vespa:~$

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/usb/class/cdc-acm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c
index 38bfdb0f6660..02eb60bb6795 100644
--- a/drivers/usb/class/cdc-acm.c
+++ b/drivers/usb/class/cdc-acm.c
@@ -550,7 +550,7 @@ static void acm_waker(struct work_struct *waker)
 static int acm_tty_open(struct tty_struct *tty, struct file *filp)
 {
 	struct acm *acm;
-	int rv = -EINVAL;
+	int rv = -ENODEV;
 	int i;
 	dbg("Entering acm_tty_open.");
 
-- 
cgit v1.2.3-59-g8ed1b


From 922b13565b6a826a925f9f91f053dc9cb0d6210e Mon Sep 17 00:00:00 2001
From: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Date: Thu, 25 Jun 2009 14:41:30 +0100
Subject: acm: Fix oops when closing ACM tty device right after open has
 failed.

This commit 10077d4a6674f535abdbe25cdecb1202af7948f1 has stopped
checking if there was a valid acm device associated to the tty, which is
not true right after open fails and tty subsystem tries to close the
device.

As an example, open fails with a non-existing device, when probe has
never been called, because the device has never been plugged. This is
common in systems with static modules and no udev.

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/usb/class/cdc-acm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c
index 02eb60bb6795..3f1045993474 100644
--- a/drivers/usb/class/cdc-acm.c
+++ b/drivers/usb/class/cdc-acm.c
@@ -677,7 +677,7 @@ static void acm_tty_close(struct tty_struct *tty, struct file *filp)
 
 	/* Perform the closing process and see if we need to do the hardware
 	   shutdown */
-	if (tty_port_close_start(&acm->port, tty, filp) == 0)
+	if (!acm || tty_port_close_start(&acm->port, tty, filp) == 0)
 		return;
 	acm_port_down(acm, 0);
 	tty_port_close_end(&acm->port, tty);
-- 
cgit v1.2.3-59-g8ed1b


From f4fa446883959c1c5f314a043e750dbfe3728c55 Mon Sep 17 00:00:00 2001
From: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Date: Thu, 25 Jun 2009 14:41:37 +0100
Subject: usb_serial: Fix oops when unexisting usb serial device is opened.

This commit 335f8514f200e63d689113d29cb7253a5c282967 has stopped
properly checking if there is any usb serial associated with the tty in
the close function. It happens the close function is called by releasing
the terminal right after opening the device fails.

As an example, open fails with a non-existing device, when probe has
never been called, because the device has never been plugged. This is
common in systems with static modules and no udev.

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/usb/serial/usb-serial.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c
index d595aa5586a7..a84216464ca0 100644
--- a/drivers/usb/serial/usb-serial.c
+++ b/drivers/usb/serial/usb-serial.c
@@ -333,6 +333,9 @@ static void serial_close(struct tty_struct *tty, struct file *filp)
 {
 	struct usb_serial_port *port = tty->driver_data;
 
+	if (!port)
+		return;
+
 	dbg("%s - port %d", __func__, port->number);
 
 
-- 
cgit v1.2.3-59-g8ed1b


From 0f3bc09ee1b7fcadd5bfdc5ed2e1643f658fe23d Mon Sep 17 00:00:00 2001
From: Suresh Jayaraman <sjayaraman@suse.de>
Date: Thu, 25 Jun 2009 18:12:34 +0530
Subject: cifs: Fix incorrect return code being printed in cFYI messages

FreeXid() along with freeing Xid does add a cifsFYI debug message that
prints rc (return code) as well. In some code paths where we set/return
error code after calling FreeXid(), incorrect error code is being
printed when cifsFYI is enabled.

This could be misleading in few cases. For eg.
In cifs_open() if cifs_fill_filedata() returns a valid pointer to
cifsFileInfo, FreeXid() prints rc=-13 whereas 0 is actually being
returned. Fix this by setting rc before calling FreeXid().

Basically convert

FreeXid(xid);			rc = -ERR;
return -ERR;		=>	FreeXid(xid);
				return rc;

[Note that Christoph would like to replace the GetXid/FreeXid
calls, which are primarily used for debugging.  This seems
like a good longer term goal, but although there is an
alternative tracing facility, there are no examples yet
available that I know of that we can use (yet) to
convert this cifs function entry/exit logging, and for
creating an identifier that we can use to correlate
all dmesg log entries for a particular vfs operation
(ie identify all log entries for a particular vfs
request to cifs: e.g. a particular close or read or write
or byte range lock call ... and just using the thread id
is harder).  Eventually when a replacement
for this is available (e.g. when NFS switches over and various
samples to look at in other file systems) we can remove the
GetXid/FreeXid macro but in the meantime multiple people
use this run time configurable logging all the time
for debugging, and Suresh's patch fixes a problem
which made it harder to notice some low
memory problems in the log so it is worthwhile
to fix this problem until a better logging
approach is able to be used]

Acked-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Suresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/dir.c   |  6 ++++--
 fs/cifs/file.c  | 24 ++++++++++++++++--------
 fs/cifs/inode.c | 15 ++++++++++-----
 fs/cifs/link.c  |  3 ++-
 fs/cifs/xattr.c | 12 ++++++++----
 5 files changed, 40 insertions(+), 20 deletions(-)

diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
index 3758965d73d5..7dc6b74f9def 100644
--- a/fs/cifs/dir.c
+++ b/fs/cifs/dir.c
@@ -307,8 +307,9 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode,
 
 	full_path = build_path_from_dentry(direntry);
 	if (full_path == NULL) {
+		rc = -ENOMEM;
 		FreeXid(xid);
-		return -ENOMEM;
+		return rc;
 	}
 
 	if (oplockEnabled)
@@ -540,8 +541,9 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode,
 			buf = kmalloc(sizeof(FILE_ALL_INFO), GFP_KERNEL);
 			if (buf == NULL) {
 				kfree(full_path);
+				rc = -ENOMEM;
 				FreeXid(xid);
-				return -ENOMEM;
+				return rc;
 			}
 
 			rc = CIFSSMBOpen(xid, pTcon, full_path,
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 06866841b97f..ebdbe62a829c 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -300,14 +300,16 @@ int cifs_open(struct inode *inode, struct file *file)
 	pCifsInode = CIFS_I(file->f_path.dentry->d_inode);
 	pCifsFile = cifs_fill_filedata(file);
 	if (pCifsFile) {
+		rc = 0;
 		FreeXid(xid);
-		return 0;
+		return rc;
 	}
 
 	full_path = build_path_from_dentry(file->f_path.dentry);
 	if (full_path == NULL) {
+		rc = -ENOMEM;
 		FreeXid(xid);
-		return -ENOMEM;
+		return rc;
 	}
 
 	cFYI(1, ("inode = 0x%p file flags are 0x%x for %s",
@@ -494,8 +496,9 @@ static int cifs_reopen_file(struct file *file, bool can_flush)
 	mutex_unlock(&pCifsFile->fh_mutex);
 	if (!pCifsFile->invalidHandle) {
 		mutex_lock(&pCifsFile->fh_mutex);
+		rc = 0;
 		FreeXid(xid);
-		return 0;
+		return rc;
 	}
 
 	if (file->f_path.dentry == NULL) {
@@ -845,8 +848,9 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock)
 	tcon = cifs_sb->tcon;
 
 	if (file->private_data == NULL) {
+		rc = -EBADF;
 		FreeXid(xid);
-		return -EBADF;
+		return rc;
 	}
 	netfid = ((struct cifsFileInfo *)file->private_data)->netfid;
 
@@ -1805,8 +1809,9 @@ ssize_t cifs_user_read(struct file *file, char __user *read_data,
 	pTcon = cifs_sb->tcon;
 
 	if (file->private_data == NULL) {
+		rc = -EBADF;
 		FreeXid(xid);
-		return -EBADF;
+		return rc;
 	}
 	open_file = (struct cifsFileInfo *)file->private_data;
 
@@ -1885,8 +1890,9 @@ static ssize_t cifs_read(struct file *file, char *read_data, size_t read_size,
 	pTcon = cifs_sb->tcon;
 
 	if (file->private_data == NULL) {
+		rc = -EBADF;
 		FreeXid(xid);
-		return -EBADF;
+		return rc;
 	}
 	open_file = (struct cifsFileInfo *)file->private_data;
 
@@ -2019,8 +2025,9 @@ static int cifs_readpages(struct file *file, struct address_space *mapping,
 
 	xid = GetXid();
 	if (file->private_data == NULL) {
+		rc = -EBADF;
 		FreeXid(xid);
-		return -EBADF;
+		return rc;
 	}
 	open_file = (struct cifsFileInfo *)file->private_data;
 	cifs_sb = CIFS_SB(file->f_path.dentry->d_sb);
@@ -2185,8 +2192,9 @@ static int cifs_readpage(struct file *file, struct page *page)
 	xid = GetXid();
 
 	if (file->private_data == NULL) {
+		rc = -EBADF;
 		FreeXid(xid);
-		return -EBADF;
+		return rc;
 	}
 
 	cFYI(1, ("readpage %p at offset %d 0x%x\n",
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index fad882b075ba..155c9e785d0c 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -988,8 +988,9 @@ int cifs_unlink(struct inode *dir, struct dentry *dentry)
 	 * sb->s_vfs_rename_mutex here */
 	full_path = build_path_from_dentry(dentry);
 	if (full_path == NULL) {
+		rc = -ENOMEM;
 		FreeXid(xid);
-		return -ENOMEM;
+		return rc;
 	}
 
 	if ((tcon->ses->capabilities & CAP_UNIX) &&
@@ -1118,8 +1119,9 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode)
 
 	full_path = build_path_from_dentry(direntry);
 	if (full_path == NULL) {
+		rc = -ENOMEM;
 		FreeXid(xid);
-		return -ENOMEM;
+		return rc;
 	}
 
 	if ((pTcon->ses->capabilities & CAP_UNIX) &&
@@ -1303,8 +1305,9 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry)
 
 	full_path = build_path_from_dentry(direntry);
 	if (full_path == NULL) {
+		rc = -ENOMEM;
 		FreeXid(xid);
-		return -ENOMEM;
+		return rc;
 	}
 
 	rc = CIFSSMBRmDir(xid, pTcon, full_path, cifs_sb->local_nls,
@@ -1508,8 +1511,9 @@ int cifs_revalidate(struct dentry *direntry)
 	   since that would deadlock */
 	full_path = build_path_from_dentry(direntry);
 	if (full_path == NULL) {
+		rc = -ENOMEM;
 		FreeXid(xid);
-		return -ENOMEM;
+		return rc;
 	}
 	cFYI(1, ("Revalidate: %s inode 0x%p count %d dentry: 0x%p d_time %ld "
 		 "jiffies %ld", full_path, direntry->d_inode,
@@ -1911,8 +1915,9 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs)
 
 	full_path = build_path_from_dentry(direntry);
 	if (full_path == NULL) {
+		rc = -ENOMEM;
 		FreeXid(xid);
-		return -ENOMEM;
+		return rc;
 	}
 
 	/*
diff --git a/fs/cifs/link.c b/fs/cifs/link.c
index cd83c53fcbb5..fc1e0487eaee 100644
--- a/fs/cifs/link.c
+++ b/fs/cifs/link.c
@@ -172,8 +172,9 @@ cifs_symlink(struct inode *inode, struct dentry *direntry, const char *symname)
 	full_path = build_path_from_dentry(direntry);
 
 	if (full_path == NULL) {
+		rc = -ENOMEM;
 		FreeXid(xid);
-		return -ENOMEM;
+		return rc;
 	}
 
 	cFYI(1, ("Full path: %s", full_path));
diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c
index e9527eedc639..a75afa3dd9e1 100644
--- a/fs/cifs/xattr.c
+++ b/fs/cifs/xattr.c
@@ -64,8 +64,9 @@ int cifs_removexattr(struct dentry *direntry, const char *ea_name)
 
 	full_path = build_path_from_dentry(direntry);
 	if (full_path == NULL) {
+		rc = -ENOMEM;
 		FreeXid(xid);
-		return -ENOMEM;
+		return rc;
 	}
 	if (ea_name == NULL) {
 		cFYI(1, ("Null xattr names not supported"));
@@ -118,8 +119,9 @@ int cifs_setxattr(struct dentry *direntry, const char *ea_name,
 
 	full_path = build_path_from_dentry(direntry);
 	if (full_path == NULL) {
+		rc = -ENOMEM;
 		FreeXid(xid);
-		return -ENOMEM;
+		return rc;
 	}
 	/* return dos attributes as pseudo xattr */
 	/* return alt name if available as pseudo attr */
@@ -225,8 +227,9 @@ ssize_t cifs_getxattr(struct dentry *direntry, const char *ea_name,
 
 	full_path = build_path_from_dentry(direntry);
 	if (full_path == NULL) {
+		rc = -ENOMEM;
 		FreeXid(xid);
-		return -ENOMEM;
+		return rc;
 	}
 	/* return dos attributes as pseudo xattr */
 	/* return alt name if available as pseudo attr */
@@ -351,8 +354,9 @@ ssize_t cifs_listxattr(struct dentry *direntry, char *data, size_t buf_size)
 
 	full_path = build_path_from_dentry(direntry);
 	if (full_path == NULL) {
+		rc = -ENOMEM;
 		FreeXid(xid);
-		return -ENOMEM;
+		return rc;
 	}
 	/* return dos attributes as pseudo xattr */
 	/* return alt name if available as pseudo attr */
-- 
cgit v1.2.3-59-g8ed1b


From dfc2f91ac29f5ef50e74bf15a1a6b6aa6b952e62 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Fri, 26 Jun 2009 04:31:57 +0900
Subject: nommu: provide follow_pfn().

With the introduction of follow_pfn() as an exported symbol, modules have
begun making use of it. Unfortunately this was not reflected on nommu at
the time, so the in-tree users have subsequently all blown up with link
errors there.

This provides a simple follow_pfn() that just returns addr >> PAGE_SHIFT,
which will do the right thing on nommu. There is no need to do range
checking within the vma, as the find_vma() case will already take care of
this.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 mm/nommu.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/mm/nommu.c b/mm/nommu.c
index 2fd2ad5da98e..598bc871487a 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -240,6 +240,27 @@ int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 }
 EXPORT_SYMBOL(get_user_pages);
 
+/**
+ * follow_pfn - look up PFN at a user virtual address
+ * @vma: memory mapping
+ * @address: user virtual address
+ * @pfn: location to store found PFN
+ *
+ * Only IO mappings and raw PFN mappings are allowed.
+ *
+ * Returns zero and the pfn at @pfn on success, -ve otherwise.
+ */
+int follow_pfn(struct vm_area_struct *vma, unsigned long address,
+	unsigned long *pfn)
+{
+	if (!(vma->vm_flags & (VM_IO | VM_PFNMAP)))
+		return -EINVAL;
+
+	*pfn = address >> PAGE_SHIFT;
+	return 0;
+}
+EXPORT_SYMBOL(follow_pfn);
+
 DEFINE_RWLOCK(vmlist_lock);
 struct vm_struct *vmlist;
 
-- 
cgit v1.2.3-59-g8ed1b


From 7c6a1c65bbd3be688e581511f45818663efc1877 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 25 Jun 2009 17:05:54 +0200
Subject: perf_counter tools: Rework the file format

Create a structured file format that includes the full
perf_counter_attr and all its relevant counter IDs so that
the reporting program has full information.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/Makefile         |   3 +-
 tools/perf/builtin-record.c | 100 +++++++++++-------
 tools/perf/builtin-report.c |  37 +++++--
 tools/perf/perf.h           |   8 +-
 tools/perf/types.h          |  17 ----
 tools/perf/util/header.c    | 242 ++++++++++++++++++++++++++++++++++++++++++++
 tools/perf/util/header.h    |  37 +++++++
 tools/perf/util/string.h    |   2 +-
 tools/perf/util/symbol.h    |   2 +-
 tools/perf/util/types.h     |  17 ++++
 10 files changed, 394 insertions(+), 71 deletions(-)
 delete mode 100644 tools/perf/types.h
 create mode 100644 tools/perf/util/header.c
 create mode 100644 tools/perf/util/header.h
 create mode 100644 tools/perf/util/types.h

diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index 36d7eef49913..d3887ed51a64 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -290,7 +290,7 @@ LIB_FILE=libperf.a
 
 LIB_H += ../../include/linux/perf_counter.h
 LIB_H += perf.h
-LIB_H += types.h
+LIB_H += util/types.h
 LIB_H += util/list.h
 LIB_H += util/rbtree.h
 LIB_H += util/levenshtein.h
@@ -328,6 +328,7 @@ LIB_OBJS += util/sigchain.o
 LIB_OBJS += util/symbol.o
 LIB_OBJS += util/color.o
 LIB_OBJS += util/pager.o
+LIB_OBJS += util/header.o
 
 BUILTIN_OBJS += builtin-annotate.o
 BUILTIN_OBJS += builtin-help.o
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 9b899ba1b410..f4f0240d2302 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -14,6 +14,8 @@
 #include "util/parse-events.h"
 #include "util/string.h"
 
+#include "util/header.h"
+
 #include <unistd.h>
 #include <sched.h>
 
@@ -52,7 +54,8 @@ static int			nr_poll;
 static int			nr_cpu;
 
 static int			file_new = 1;
-static struct perf_file_header	file_header;
+
+struct perf_header		*header;
 
 struct mmap_event {
 	struct perf_event_header	header;
@@ -328,7 +331,7 @@ static void pid_synthesize_mmap_samples(pid_t pid)
 	fclose(fp);
 }
 
-static void synthesize_samples(void)
+static void synthesize_all(void)
 {
 	DIR *proc;
 	struct dirent dirent, *next;
@@ -352,10 +355,35 @@ static void synthesize_samples(void)
 
 static int group_fd;
 
+static struct perf_header_attr *get_header_attr(struct perf_counter_attr *a, int nr)
+{
+	struct perf_header_attr *h_attr;
+
+	if (nr < header->attrs) {
+		h_attr = header->attr[nr];
+	} else {
+		h_attr = perf_header_attr__new(a);
+		perf_header__add_attr(header, h_attr);
+	}
+
+	return h_attr;
+}
+
 static void create_counter(int counter, int cpu, pid_t pid)
 {
 	struct perf_counter_attr *attr = attrs + counter;
-	int track = 1;
+	struct perf_header_attr *h_attr;
+	int track = !counter; /* only the first counter needs these */
+	struct {
+		u64 count;
+		u64 time_enabled;
+		u64 time_running;
+		u64 id;
+	} read_data;
+
+	attr->read_format	= PERF_FORMAT_TOTAL_TIME_ENABLED |
+				  PERF_FORMAT_TOTAL_TIME_RUNNING |
+				  PERF_FORMAT_ID;
 
 	attr->sample_type	= PERF_SAMPLE_IP | PERF_SAMPLE_TID;
 
@@ -368,22 +396,11 @@ static void create_counter(int counter, int cpu, pid_t pid)
 	if (call_graph)
 		attr->sample_type	|= PERF_SAMPLE_CALLCHAIN;
 
-	if (file_new) {
-		file_header.sample_type = attr->sample_type;
-	} else {
-		if (file_header.sample_type != attr->sample_type) {
-			fprintf(stderr, "incompatible append\n");
-			exit(-1);
-		}
-	}
-
 	attr->mmap		= track;
 	attr->comm		= track;
 	attr->inherit		= (cpu < 0) && inherit;
 	attr->disabled		= 1;
 
-	track = 0; /* only the first counter needs these */
-
 try_again:
 	fd[nr_cpu][counter] = sys_perf_counter_open(attr, pid, cpu, group_fd, 0);
 
@@ -414,6 +431,19 @@ try_again:
 		exit(-1);
 	}
 
+	h_attr = get_header_attr(attr, counter);
+
+	if (!file_new) {
+		if (memcmp(&h_attr->attr, attr, sizeof(*attr))) {
+			fprintf(stderr, "incompatible append\n");
+			exit(-1);
+		}
+	}
+
+	read(fd[nr_cpu][counter], &read_data, sizeof(read_data));
+
+	perf_header_attr__add_id(h_attr, read_data.id);
+
 	assert(fd[nr_cpu][counter] >= 0);
 	fcntl(fd[nr_cpu][counter], F_SETFL, O_NONBLOCK);
 
@@ -444,11 +474,6 @@ static void open_counters(int cpu, pid_t pid)
 {
 	int counter;
 
-	if (pid > 0) {
-		pid_synthesize_comm_event(pid, 0);
-		pid_synthesize_mmap_samples(pid);
-	}
-
 	group_fd = -1;
 	for (counter = 0; counter < nr_counters; counter++)
 		create_counter(counter, cpu, pid);
@@ -458,17 +483,16 @@ static void open_counters(int cpu, pid_t pid)
 
 static void atexit_header(void)
 {
-	file_header.data_size += bytes_written;
+	header->data_size += bytes_written;
 
-	if (pwrite(output, &file_header, sizeof(file_header), 0) == -1)
-		perror("failed to write on file headers");
+	perf_header__write(header, output);
 }
 
 static int __cmd_record(int argc, const char **argv)
 {
 	int i, counter;
 	struct stat st;
-	pid_t pid;
+	pid_t pid = 0;
 	int flags;
 	int ret;
 
@@ -499,22 +523,31 @@ static int __cmd_record(int argc, const char **argv)
 		exit(-1);
 	}
 
-	if (!file_new) {
-		if (read(output, &file_header, sizeof(file_header)) == -1) {
-			perror("failed to read file headers");
-			exit(-1);
-		}
-
-		lseek(output, file_header.data_size, SEEK_CUR);
-	}
+	if (!file_new)
+		header = perf_header__read(output);
+	else
+		header = perf_header__new();
 
 	atexit(atexit_header);
 
 	if (!system_wide) {
-		open_counters(-1, target_pid != -1 ? target_pid : getpid());
+		pid = target_pid;
+		if (pid == -1)
+			pid = getpid();
+
+		open_counters(-1, pid);
 	} else for (i = 0; i < nr_cpus; i++)
 		open_counters(i, target_pid);
 
+	if (file_new)
+		perf_header__write(header, output);
+
+	if (!system_wide) {
+		pid_synthesize_comm_event(pid, 0);
+		pid_synthesize_mmap_samples(pid);
+	} else
+		synthesize_all();
+
 	if (target_pid == -1 && argc) {
 		pid = fork();
 		if (pid < 0)
@@ -538,9 +571,6 @@ static int __cmd_record(int argc, const char **argv)
 		}
 	}
 
-	if (system_wide)
-		synthesize_samples();
-
 	while (!done) {
 		int hits = samples;
 
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index b4e76f75ba87..e575f3039766 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -17,6 +17,7 @@
 #include "util/string.h"
 
 #include "perf.h"
+#include "util/header.h"
 
 #include "util/parse-options.h"
 #include "util/parse-events.h"
@@ -1385,13 +1386,27 @@ process_event(event_t *event, unsigned long offset, unsigned long head)
 	return 0;
 }
 
-static struct perf_file_header		file_header;
+static struct perf_header	*header;
+
+static int perf_header__has_sample(u64 sample_mask)
+{
+	int i;
+
+	for (i = 0; i < header->attrs; i++) {
+		struct perf_header_attr *attr = header->attr[i];
+
+		if (!(attr->attr.sample_type & sample_mask))
+			return 0;
+	}
+
+	return 1;
+}
 
 static int __cmd_report(void)
 {
 	int ret, rc = EXIT_FAILURE;
 	unsigned long offset = 0;
-	unsigned long head = sizeof(file_header);
+	unsigned long head, shift;
 	struct stat stat;
 	event_t *event;
 	uint32_t size;
@@ -1419,13 +1434,11 @@ static int __cmd_report(void)
 		exit(0);
 	}
 
-	if (read(input, &file_header, sizeof(file_header)) == -1) {
-		perror("failed to read file headers");
-		exit(-1);
-	}
+	header = perf_header__read(input);
+	head = header->data_offset;
 
 	if (sort__has_parent &&
-	    !(file_header.sample_type & PERF_SAMPLE_CALLCHAIN)) {
+	    !perf_header__has_sample(PERF_SAMPLE_CALLCHAIN)) {
 		fprintf(stderr, "selected --sort parent, but no callchain data\n");
 		exit(-1);
 	}
@@ -1445,6 +1458,11 @@ static int __cmd_report(void)
 		cwd = NULL;
 		cwdlen = 0;
 	}
+
+	shift = page_size * (head / page_size);
+	offset += shift;
+	head -= shift;
+
 remap:
 	buf = (char *)mmap(NULL, page_size * mmap_window, PROT_READ,
 			   MAP_SHARED, input, offset);
@@ -1461,9 +1479,10 @@ more:
 		size = 8;
 
 	if (head + event->header.size >= page_size * mmap_window) {
-		unsigned long shift = page_size * (head / page_size);
 		int ret;
 
+		shift = page_size * (head / page_size);
+
 		ret = munmap(buf, page_size * mmap_window);
 		assert(ret == 0);
 
@@ -1501,7 +1520,7 @@ more:
 
 	head += size;
 
-	if (offset + head >= sizeof(file_header) + file_header.data_size)
+	if (offset + head >= header->data_offset + header->data_size)
 		goto done;
 
 	if (offset + head < stat.st_size)
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index bccb529dac08..16c84fd73c86 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -19,7 +19,7 @@
 #include <sys/syscall.h>
 
 #include "../../include/linux/perf_counter.h"
-#include "types.h"
+#include "util/types.h"
 
 /*
  * prctl(PR_TASK_PERF_COUNTERS_DISABLE) will (cheaply) disable all
@@ -66,10 +66,4 @@ sys_perf_counter_open(struct perf_counter_attr *attr,
 #define MAX_COUNTERS			256
 #define MAX_NR_CPUS			256
 
-struct perf_file_header {
-	u64	version;
-	u64	sample_type;
-	u64	data_size;
-};
-
 #endif
diff --git a/tools/perf/types.h b/tools/perf/types.h
deleted file mode 100644
index 5e75f9005940..000000000000
--- a/tools/perf/types.h
+++ /dev/null
@@ -1,17 +0,0 @@
-#ifndef _PERF_TYPES_H
-#define _PERF_TYPES_H
-
-/*
- * We define u64 as unsigned long long for every architecture
- * so that we can print it with %Lx without getting warnings.
- */
-typedef unsigned long long u64;
-typedef signed long long   s64;
-typedef unsigned int	   u32;
-typedef signed int	   s32;
-typedef unsigned short	   u16;
-typedef signed short	   s16;
-typedef unsigned char	   u8;
-typedef signed char	   s8;
-
-#endif /* _PERF_TYPES_H */
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
new file mode 100644
index 000000000000..450384b3bbe5
--- /dev/null
+++ b/tools/perf/util/header.c
@@ -0,0 +1,242 @@
+#include <sys/types.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+#include "util.h"
+#include "header.h"
+
+/*
+ *
+ */
+
+struct perf_header_attr *perf_header_attr__new(struct perf_counter_attr *attr)
+{
+	struct perf_header_attr *self = malloc(sizeof(*self));
+
+	if (!self)
+		die("nomem");
+
+	self->attr = *attr;
+	self->ids = 0;
+	self->size = 1;
+	self->id = malloc(sizeof(u64));
+
+	if (!self->id)
+		die("nomem");
+
+	return self;
+}
+
+void perf_header_attr__add_id(struct perf_header_attr *self, u64 id)
+{
+	int pos = self->ids;
+
+	self->ids++;
+	if (self->ids > self->size) {
+		self->size *= 2;
+		self->id = realloc(self->id, self->size * sizeof(u64));
+		if (!self->id)
+			die("nomem");
+	}
+	self->id[pos] = id;
+}
+
+/*
+ *
+ */
+
+struct perf_header *perf_header__new(void)
+{
+	struct perf_header *self = malloc(sizeof(*self));
+
+	if (!self)
+		die("nomem");
+
+	self->frozen = 0;
+
+	self->attrs = 0;
+	self->size = 1;
+	self->attr = malloc(sizeof(void *));
+
+	if (!self->attr)
+		die("nomem");
+
+	self->data_offset = 0;
+	self->data_size = 0;
+
+	return self;
+}
+
+void perf_header__add_attr(struct perf_header *self,
+			   struct perf_header_attr *attr)
+{
+	int pos = self->attrs;
+
+	if (self->frozen)
+		die("frozen");
+
+	self->attrs++;
+	if (self->attrs > self->size) {
+		self->size *= 2;
+		self->attr = realloc(self->attr, self->size * sizeof(void *));
+		if (!self->attr)
+			die("nomem");
+	}
+	self->attr[pos] = attr;
+}
+
+static const char *__perf_magic = "PERFFILE";
+
+#define PERF_MAGIC	(*(u64 *)__perf_magic)
+
+struct perf_file_section {
+	u64 offset;
+	u64 size;
+};
+
+struct perf_file_attr {
+	struct perf_counter_attr	attr;
+	struct perf_file_section	ids;
+};
+
+struct perf_file_header {
+	u64				magic;
+	u64				size;
+	u64				attr_size;
+	struct perf_file_section	attrs;
+	struct perf_file_section	data;
+};
+
+static void do_write(int fd, void *buf, size_t size)
+{
+	while (size) {
+		int ret = write(fd, buf, size);
+
+		if (ret < 0)
+			die("failed to write");
+
+		size -= ret;
+		buf += ret;
+	}
+}
+
+void perf_header__write(struct perf_header *self, int fd)
+{
+	struct perf_file_header f_header;
+	struct perf_file_attr   f_attr;
+	struct perf_header_attr	*attr;
+	int i;
+
+	lseek(fd, sizeof(f_header), SEEK_SET);
+
+
+	for (i = 0; i < self->attrs; i++) {
+		attr = self->attr[i];
+
+		attr->id_offset = lseek(fd, 0, SEEK_CUR);
+		do_write(fd, attr->id, attr->ids * sizeof(u64));
+	}
+
+
+	self->attr_offset = lseek(fd, 0, SEEK_CUR);
+
+	for (i = 0; i < self->attrs; i++) {
+		attr = self->attr[i];
+
+		f_attr = (struct perf_file_attr){
+			.attr = attr->attr,
+			.ids  = {
+				.offset = attr->id_offset,
+				.size   = attr->ids * sizeof(u64),
+			}
+		};
+		do_write(fd, &f_attr, sizeof(f_attr));
+	}
+
+
+	self->data_offset = lseek(fd, 0, SEEK_CUR);
+
+	f_header = (struct perf_file_header){
+		.magic	   = PERF_MAGIC,
+		.size	   = sizeof(f_header),
+		.attr_size = sizeof(f_attr),
+		.attrs = {
+			.offset = self->attr_offset,
+			.size   = self->attrs * sizeof(f_attr),
+		},
+		.data = {
+			.offset = self->data_offset,
+			.size	= self->data_size,
+		},
+	};
+
+	lseek(fd, 0, SEEK_SET);
+	do_write(fd, &f_header, sizeof(f_header));
+	lseek(fd, self->data_offset + self->data_size, SEEK_SET);
+
+	self->frozen = 1;
+}
+
+static void do_read(int fd, void *buf, size_t size)
+{
+	while (size) {
+		int ret = read(fd, buf, size);
+
+		if (ret < 0)
+			die("failed to read");
+
+		size -= ret;
+		buf += ret;
+	}
+}
+
+struct perf_header *perf_header__read(int fd)
+{
+	struct perf_header	*self = perf_header__new();
+	struct perf_file_header f_header;
+	struct perf_file_attr	f_attr;
+	u64			f_id;
+
+	int nr_attrs, nr_ids, i, j;
+
+	lseek(fd, 0, SEEK_SET);
+	do_read(fd, &f_header, sizeof(f_header));
+
+	if (f_header.magic	!= PERF_MAGIC		||
+	    f_header.size	!= sizeof(f_header)	||
+	    f_header.attr_size	!= sizeof(f_attr))
+		die("incompatible file format");
+
+	nr_attrs = f_header.attrs.size / sizeof(f_attr);
+	lseek(fd, f_header.attrs.offset, SEEK_SET);
+
+	for (i = 0; i < nr_attrs; i++) {
+		struct perf_header_attr *attr;
+		off_t tmp = lseek(fd, 0, SEEK_CUR);
+
+		do_read(fd, &f_attr, sizeof(f_attr));
+
+		attr = perf_header_attr__new(&f_attr.attr);
+
+		nr_ids = f_attr.ids.size / sizeof(u64);
+		lseek(fd, f_attr.ids.offset, SEEK_SET);
+
+		for (j = 0; j < nr_ids; j++) {
+			do_read(fd, &f_id, sizeof(f_id));
+
+			perf_header_attr__add_id(attr, f_id);
+		}
+		perf_header__add_attr(self, attr);
+		lseek(fd, tmp, SEEK_SET);
+	}
+
+	self->data_offset = f_header.data.offset;
+	self->data_size   = f_header.data.size;
+
+	lseek(fd, self->data_offset + self->data_size, SEEK_SET);
+
+	self->frozen = 1;
+
+	return self;
+}
diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h
new file mode 100644
index 000000000000..b5ef53ad4c7a
--- /dev/null
+++ b/tools/perf/util/header.h
@@ -0,0 +1,37 @@
+#ifndef _PERF_HEADER_H
+#define _PERF_HEADER_H
+
+#include "../../../include/linux/perf_counter.h"
+#include <sys/types.h>
+#include "types.h"
+
+struct perf_header_attr {
+	struct perf_counter_attr attr;
+	int ids, size;
+	u64 *id;
+	off_t id_offset;
+};
+
+struct perf_header {
+	int frozen;
+	int attrs, size;
+	struct perf_header_attr **attr;
+	off_t attr_offset;
+	u64 data_offset;
+	u64 data_size;
+};
+
+struct perf_header *perf_header__read(int fd);
+void perf_header__write(struct perf_header *self, int fd);
+
+void perf_header__add_attr(struct perf_header *self,
+			   struct perf_header_attr *attr);
+
+struct perf_header_attr *
+perf_header_attr__new(struct perf_counter_attr *attr);
+void perf_header_attr__add_id(struct perf_header_attr *self, u64 id);
+
+
+struct perf_header *perf_header__new(void);
+
+#endif /* _PERF_HEADER_H */
diff --git a/tools/perf/util/string.h b/tools/perf/util/string.h
index 37b03255b425..3dca2f654cd0 100644
--- a/tools/perf/util/string.h
+++ b/tools/perf/util/string.h
@@ -1,7 +1,7 @@
 #ifndef _PERF_STRING_H_
 #define _PERF_STRING_H_
 
-#include "../types.h"
+#include "types.h"
 
 int hex2u64(const char *ptr, u64 *val);
 
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index ea332e56e458..940b432db16e 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -2,7 +2,7 @@
 #define _PERF_SYMBOL_ 1
 
 #include <linux/types.h>
-#include "../types.h"
+#include "types.h"
 #include "list.h"
 #include "rbtree.h"
 
diff --git a/tools/perf/util/types.h b/tools/perf/util/types.h
new file mode 100644
index 000000000000..5e75f9005940
--- /dev/null
+++ b/tools/perf/util/types.h
@@ -0,0 +1,17 @@
+#ifndef _PERF_TYPES_H
+#define _PERF_TYPES_H
+
+/*
+ * We define u64 as unsigned long long for every architecture
+ * so that we can print it with %Lx without getting warnings.
+ */
+typedef unsigned long long u64;
+typedef signed long long   s64;
+typedef unsigned int	   u32;
+typedef signed int	   s32;
+typedef unsigned short	   u16;
+typedef signed short	   s16;
+typedef unsigned char	   u8;
+typedef signed char	   s8;
+
+#endif /* _PERF_TYPES_H */
-- 
cgit v1.2.3-59-g8ed1b


From 41f95331b972a039f519ae0c70f051b7121f7346 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Tue, 23 Jun 2009 17:55:18 +0200
Subject: perf_counter: Split the mmap control page in two parts

Since there are two distinct sections to the control page,
move them apart so that possible extentions don't overlap.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_counter.h | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/include/linux/perf_counter.h b/include/linux/perf_counter.h
index e7213e46cf9c..489d5cbfbcca 100644
--- a/include/linux/perf_counter.h
+++ b/include/linux/perf_counter.h
@@ -233,6 +233,12 @@ struct perf_counter_mmap_page {
 	__u32	index;			/* hardware counter identifier */
 	__s64	offset;			/* add to hardware counter value */
 
+		/*
+		 * Hole for extension of the self monitor capabilities
+		 */
+
+	__u64	__reserved[125];	/* align to 1k */
+
 	/*
 	 * Control data for the mmap() data buffer.
 	 *
-- 
cgit v1.2.3-59-g8ed1b


From 7f8b4e4e0988dadfd22330fd147ad2453e19f510 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 22 Jun 2009 14:34:35 +0200
Subject: perf_counter: Add scale information to the mmap control page

Add the needed time scale to the self-profile mmap information.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_counter.h | 4 +++-
 kernel/perf_counter.c        | 6 ++++++
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/include/linux/perf_counter.h b/include/linux/perf_counter.h
index 489d5cbfbcca..bcbf1c43ed42 100644
--- a/include/linux/perf_counter.h
+++ b/include/linux/perf_counter.h
@@ -232,12 +232,14 @@ struct perf_counter_mmap_page {
 	__u32	lock;			/* seqlock for synchronization */
 	__u32	index;			/* hardware counter identifier */
 	__s64	offset;			/* add to hardware counter value */
+	__u64	time_enabled;		/* time counter active */
+	__u64	time_running;		/* time counter on cpu */
 
 		/*
 		 * Hole for extension of the self monitor capabilities
 		 */
 
-	__u64	__reserved[125];	/* align to 1k */
+	__u64	__reserved[123];	/* align to 1k */
 
 	/*
 	 * Control data for the mmap() data buffer.
diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index c2b19c111718..23614adab475 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -1782,6 +1782,12 @@ void perf_counter_update_userpage(struct perf_counter *counter)
 	if (counter->state == PERF_COUNTER_STATE_ACTIVE)
 		userpg->offset -= atomic64_read(&counter->hw.prev_count);
 
+	userpg->time_enabled = counter->total_time_enabled +
+			atomic64_read(&counter->child_total_time_enabled);
+
+	userpg->time_running = counter->total_time_running +
+			atomic64_read(&counter->child_total_time_running);
+
 	barrier();
 	++userpg->lock;
 	preempt_enable();
-- 
cgit v1.2.3-59-g8ed1b


From 194002b274e9169a04beb1b23dcc132159bb566c Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 22 Jun 2009 16:35:24 +0200
Subject: perf_counter, x86: Add mmap counter read support

Update the mmap control page with the needed information to
use the userspace RDPMC instruction for self monitoring.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/powerpc/include/asm/perf_counter.h |  2 ++
 arch/x86/include/asm/perf_counter.h     |  3 +++
 arch/x86/kernel/cpu/perf_counter.c      |  6 ++++++
 kernel/perf_counter.c                   | 10 +++++++++-
 4 files changed, 20 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/include/asm/perf_counter.h b/arch/powerpc/include/asm/perf_counter.h
index 8ccd4e155768..0ea0639fcf75 100644
--- a/arch/powerpc/include/asm/perf_counter.h
+++ b/arch/powerpc/include/asm/perf_counter.h
@@ -61,6 +61,8 @@ struct pt_regs;
 extern unsigned long perf_misc_flags(struct pt_regs *regs);
 extern unsigned long perf_instruction_pointer(struct pt_regs *regs);
 
+#define PERF_COUNTER_INDEX_OFFSET	1
+
 /*
  * Only override the default definitions in include/linux/perf_counter.h
  * if we have hardware PMU support.
diff --git a/arch/x86/include/asm/perf_counter.h b/arch/x86/include/asm/perf_counter.h
index 5fb33e160ea0..fa64e401589d 100644
--- a/arch/x86/include/asm/perf_counter.h
+++ b/arch/x86/include/asm/perf_counter.h
@@ -87,6 +87,9 @@ union cpuid10_edx {
 #ifdef CONFIG_PERF_COUNTERS
 extern void init_hw_perf_counters(void);
 extern void perf_counters_lapic_init(void);
+
+#define PERF_COUNTER_INDEX_OFFSET			0
+
 #else
 static inline void init_hw_perf_counters(void)		{ }
 static inline void perf_counters_lapic_init(void)	{ }
diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c
index a310d19faca3..b83474b6021a 100644
--- a/arch/x86/kernel/cpu/perf_counter.c
+++ b/arch/x86/kernel/cpu/perf_counter.c
@@ -912,6 +912,8 @@ x86_perf_counter_set_period(struct perf_counter *counter,
 	err = checking_wrmsrl(hwc->counter_base + idx,
 			     (u64)(-left) & x86_pmu.counter_mask);
 
+	perf_counter_update_userpage(counter);
+
 	return ret;
 }
 
@@ -1034,6 +1036,8 @@ try_generic:
 	x86_perf_counter_set_period(counter, hwc, idx);
 	x86_pmu.enable(hwc, idx);
 
+	perf_counter_update_userpage(counter);
+
 	return 0;
 }
 
@@ -1126,6 +1130,8 @@ static void x86_pmu_disable(struct perf_counter *counter)
 	x86_perf_counter_update(counter, hwc, idx);
 	cpuc->counters[idx] = NULL;
 	clear_bit(idx, cpuc->used_mask);
+
+	perf_counter_update_userpage(counter);
 }
 
 /*
diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index 23614adab475..02994a719e27 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -1753,6 +1753,14 @@ int perf_counter_task_disable(void)
 	return 0;
 }
 
+static int perf_counter_index(struct perf_counter *counter)
+{
+	if (counter->state != PERF_COUNTER_STATE_ACTIVE)
+		return 0;
+
+	return counter->hw.idx + 1 - PERF_COUNTER_INDEX_OFFSET;
+}
+
 /*
  * Callers need to ensure there can be no nesting of this function, otherwise
  * the seqlock logic goes bad. We can not serialize this because the arch
@@ -1777,7 +1785,7 @@ void perf_counter_update_userpage(struct perf_counter *counter)
 	preempt_disable();
 	++userpg->lock;
 	barrier();
-	userpg->index = counter->hw.idx;
+	userpg->index = perf_counter_index(counter);
 	userpg->offset = atomic64_read(&counter->count);
 	if (counter->state == PERF_COUNTER_STATE_ACTIVE)
 		userpg->offset -= atomic64_read(&counter->hw.prev_count);
-- 
cgit v1.2.3-59-g8ed1b


From 38b200d67636a30cb8dc1508137908e7a649b5c9 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Tue, 23 Jun 2009 20:13:11 +0200
Subject: perf_counter: Add PERF_EVENT_READ

Provide a read() like event which can be used to log the
counter value at specific sites such as child->parent
folding on exit.

In order to be useful, we log the counter parent ID, not the
actual counter ID, since userspace can only relate parent
IDs to perf_counter_attr constructs.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_counter.h | 12 ++++++++
 kernel/perf_counter.c        | 72 +++++++++++++++++++++++++++++++++++++++++---
 2 files changed, 80 insertions(+), 4 deletions(-)

diff --git a/include/linux/perf_counter.h b/include/linux/perf_counter.h
index bcbf1c43ed42..6a384f04755a 100644
--- a/include/linux/perf_counter.h
+++ b/include/linux/perf_counter.h
@@ -334,6 +334,18 @@ enum perf_event_type {
 	 */
 	PERF_EVENT_FORK			= 7,
 
+	/*
+	 * struct {
+	 * 	struct perf_event_header	header;
+	 * 	u32				pid, tid;
+	 * 	u64				value;
+	 * 	{ u64		time_enabled; 	} && PERF_FORMAT_ENABLED
+	 * 	{ u64		time_running; 	} && PERF_FORMAT_RUNNING
+	 * 	{ u64		parent_id;	} && PERF_FORMAT_ID
+	 * };
+	 */
+	PERF_EVENT_READ			= 8,
+
 	/*
 	 * When header.misc & PERF_EVENT_MISC_OVERFLOW the event_type field
 	 * will be PERF_SAMPLE_*
diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index 02994a719e27..a72c20e91953 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -2623,6 +2623,66 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
 	perf_output_end(&handle);
 }
 
+/*
+ * read event
+ */
+
+struct perf_read_event {
+	struct perf_event_header	header;
+
+	u32				pid;
+	u32				tid;
+	u64				value;
+	u64				format[3];
+};
+
+static void
+perf_counter_read_event(struct perf_counter *counter,
+			struct task_struct *task)
+{
+	struct perf_output_handle handle;
+	struct perf_read_event event = {
+		.header = {
+			.type = PERF_EVENT_READ,
+			.misc = 0,
+			.size = sizeof(event) - sizeof(event.format),
+		},
+		.pid = perf_counter_pid(counter, task),
+		.tid = perf_counter_tid(counter, task),
+		.value = atomic64_read(&counter->count),
+	};
+	int ret, i = 0;
+
+	if (counter->attr.read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) {
+		event.header.size += sizeof(u64);
+		event.format[i++] = counter->total_time_enabled;
+	}
+
+	if (counter->attr.read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) {
+		event.header.size += sizeof(u64);
+		event.format[i++] = counter->total_time_running;
+	}
+
+	if (counter->attr.read_format & PERF_FORMAT_ID) {
+		u64 id;
+
+		event.header.size += sizeof(u64);
+		if (counter->parent)
+			id = counter->parent->id;
+		else
+			id = counter->id;
+
+		event.format[i++] = id;
+	}
+
+	ret = perf_output_begin(&handle, counter, event.header.size, 0, 0);
+	if (ret)
+		return;
+
+	perf_output_copy(&handle, &event, event.header.size);
+	perf_output_end(&handle);
+}
+
 /*
  * fork tracking
  */
@@ -3985,10 +4045,13 @@ static int inherit_group(struct perf_counter *parent_counter,
 }
 
 static void sync_child_counter(struct perf_counter *child_counter,
-			       struct perf_counter *parent_counter)
+			       struct task_struct *child)
 {
+	struct perf_counter *parent_counter = child_counter->parent;
 	u64 child_val;
 
+	perf_counter_read_event(child_counter, child);
+
 	child_val = atomic64_read(&child_counter->count);
 
 	/*
@@ -4017,7 +4080,8 @@ static void sync_child_counter(struct perf_counter *child_counter,
 
 static void
 __perf_counter_exit_task(struct perf_counter *child_counter,
-			 struct perf_counter_context *child_ctx)
+			 struct perf_counter_context *child_ctx,
+			 struct task_struct *child)
 {
 	struct perf_counter *parent_counter;
 
@@ -4031,7 +4095,7 @@ __perf_counter_exit_task(struct perf_counter *child_counter,
 	 * counters need to be zapped - but otherwise linger.
 	 */
 	if (parent_counter) {
-		sync_child_counter(child_counter, parent_counter);
+		sync_child_counter(child_counter, child);
 		free_counter(child_counter);
 	}
 }
@@ -4093,7 +4157,7 @@ void perf_counter_exit_task(struct task_struct *child)
 again:
 	list_for_each_entry_safe(child_counter, tmp, &child_ctx->counter_list,
 				 list_entry)
-		__perf_counter_exit_task(child_counter, child_ctx);
+		__perf_counter_exit_task(child_counter, child_ctx, child);
 
 	/*
 	 * If the last counter was a group counter, it will have appended all
-- 
cgit v1.2.3-59-g8ed1b


From bfbd3381e63aa2a14c6706afb50ce4630aa0d9a2 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Wed, 24 Jun 2009 21:11:59 +0200
Subject: perf_counter: Implement more accurate per task statistics

With the introduction of PERF_EVENT_READ we have the
possibility to provide accurate counter values for
individual tasks in a task hierarchy.

However, due to the lazy context switching used for similar
counter contexts our current per task counts are way off.

In order to maintain some of the lazy switch benefits we
don't disable it out-right, but simply iterate the active
counters and flip the values between the contexts.

This only reads the counters but does not need to reprogram
the full PMU.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_counter.h |  4 ++-
 kernel/perf_counter.c        | 83 ++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 83 insertions(+), 4 deletions(-)

diff --git a/include/linux/perf_counter.h b/include/linux/perf_counter.h
index 6a384f04755a..de70a10b5ec8 100644
--- a/include/linux/perf_counter.h
+++ b/include/linux/perf_counter.h
@@ -178,8 +178,9 @@ struct perf_counter_attr {
 				mmap           :  1, /* include mmap data     */
 				comm	       :  1, /* include comm data     */
 				freq           :  1, /* use freq, not period  */
+				inherit_stat   :  1, /* per task counts       */
 
-				__reserved_1   : 53;
+				__reserved_1   : 52;
 
 	__u32			wakeup_events;	/* wakeup every n events */
 	__u32			__reserved_2;
@@ -602,6 +603,7 @@ struct perf_counter_context {
 	int				nr_counters;
 	int				nr_active;
 	int				is_active;
+	int				nr_stat;
 	atomic_t			refcount;
 	struct task_struct		*task;
 
diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index a72c20e91953..385ca51c6e60 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -236,6 +236,8 @@ list_add_counter(struct perf_counter *counter, struct perf_counter_context *ctx)
 
 	list_add_rcu(&counter->event_entry, &ctx->event_list);
 	ctx->nr_counters++;
+	if (counter->attr.inherit_stat)
+		ctx->nr_stat++;
 }
 
 /*
@@ -250,6 +252,8 @@ list_del_counter(struct perf_counter *counter, struct perf_counter_context *ctx)
 	if (list_empty(&counter->list_entry))
 		return;
 	ctx->nr_counters--;
+	if (counter->attr.inherit_stat)
+		ctx->nr_stat--;
 
 	list_del_init(&counter->list_entry);
 	list_del_rcu(&counter->event_entry);
@@ -1006,6 +1010,76 @@ static int context_equiv(struct perf_counter_context *ctx1,
 		&& !ctx1->pin_count && !ctx2->pin_count;
 }
 
+static void __perf_counter_read(void *counter);
+
+static void __perf_counter_sync_stat(struct perf_counter *counter,
+				     struct perf_counter *next_counter)
+{
+	u64 value;
+
+	if (!counter->attr.inherit_stat)
+		return;
+
+	/*
+	 * Update the counter value, we cannot use perf_counter_read()
+	 * because we're in the middle of a context switch and have IRQs
+	 * disabled, which upsets smp_call_function_single(), however
+	 * we know the counter must be on the current CPU, therefore we
+	 * don't need to use it.
+	 */
+	switch (counter->state) {
+	case PERF_COUNTER_STATE_ACTIVE:
+		__perf_counter_read(counter);
+		break;
+
+	case PERF_COUNTER_STATE_INACTIVE:
+		update_counter_times(counter);
+		break;
+
+	default:
+		break;
+	}
+
+	/*
+	 * In order to keep per-task stats reliable we need to flip the counter
+	 * values when we flip the contexts.
+	 */
+	value = atomic64_read(&next_counter->count);
+	value = atomic64_xchg(&counter->count, value);
+	atomic64_set(&next_counter->count, value);
+
+	/*
+	 * XXX also sync time_enabled and time_running ?
+	 */
+}
+
+#define list_next_entry(pos, member) \
+	list_entry(pos->member.next, typeof(*pos), member)
+
+static void perf_counter_sync_stat(struct perf_counter_context *ctx,
+				   struct perf_counter_context *next_ctx)
+{
+	struct perf_counter *counter, *next_counter;
+
+	if (!ctx->nr_stat)
+		return;
+
+	counter = list_first_entry(&ctx->event_list,
+				   struct perf_counter, event_entry);
+
+	next_counter = list_first_entry(&next_ctx->event_list,
+					struct perf_counter, event_entry);
+
+	while (&counter->event_entry != &ctx->event_list &&
+	       &next_counter->event_entry != &next_ctx->event_list) {
+
+		__perf_counter_sync_stat(counter, next_counter);
+
+		counter = list_next_entry(counter, event_entry);
+		next_counter = list_next_entry(counter, event_entry);
+	}
+}
+
 /*
  * Called from scheduler to remove the counters of the current task,
  * with interrupts disabled.
@@ -1061,6 +1135,8 @@ void perf_counter_task_sched_out(struct task_struct *task,
 			ctx->task = next;
 			next_ctx->task = task;
 			do_switch = 0;
+
+			perf_counter_sync_stat(ctx, next_ctx);
 		}
 		spin_unlock(&next_ctx->lock);
 		spin_unlock(&ctx->lock);
@@ -1350,7 +1426,7 @@ void perf_counter_task_tick(struct task_struct *curr, int cpu)
 /*
  * Cross CPU call to read the hardware counter
  */
-static void __read(void *info)
+static void __perf_counter_read(void *info)
 {
 	struct perf_counter *counter = info;
 	struct perf_counter_context *ctx = counter->ctx;
@@ -1372,7 +1448,7 @@ static u64 perf_counter_read(struct perf_counter *counter)
 	 */
 	if (counter->state == PERF_COUNTER_STATE_ACTIVE) {
 		smp_call_function_single(counter->oncpu,
-					 __read, counter, 1);
+					 __perf_counter_read, counter, 1);
 	} else if (counter->state == PERF_COUNTER_STATE_INACTIVE) {
 		update_counter_times(counter);
 	}
@@ -4050,7 +4126,8 @@ static void sync_child_counter(struct perf_counter *child_counter,
 	struct perf_counter *parent_counter = child_counter->parent;
 	u64 child_val;
 
-	perf_counter_read_event(child_counter, child);
+	if (child_counter->attr.inherit_stat)
+		perf_counter_read_event(child_counter, child);
 
 	child_val = atomic64_read(&child_counter->count);
 
-- 
cgit v1.2.3-59-g8ed1b


From e6e18ec79b023d5fe84226cef533cf0e3770ce93 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 25 Jun 2009 11:27:12 +0200
Subject: perf_counter: Rework the sample ABI

The PERF_EVENT_READ implementation made me realize we don't
actually need the sample_type int the output sample, since
we already have that in the perf_counter_attr information.

Therefore, remove the PERF_EVENT_MISC_OVERFLOW bit and the
event->type overloading, and imply put counter overflow
samples in a PERF_EVENT_SAMPLE type.

This also fixes the issue that event->type was only 32-bit
and sample_type had 64 usable bits.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_counter.h  | 10 +++++-----
 kernel/perf_counter.c         | 36 +++++++++++++++---------------------
 tools/perf/builtin-annotate.c |  8 ++++----
 tools/perf/builtin-report.c   | 32 +++++++++++++++++++-------------
 tools/perf/builtin-top.c      | 11 ++++++-----
 5 files changed, 49 insertions(+), 48 deletions(-)

diff --git a/include/linux/perf_counter.h b/include/linux/perf_counter.h
index de70a10b5ec8..3078e23c91eb 100644
--- a/include/linux/perf_counter.h
+++ b/include/linux/perf_counter.h
@@ -262,7 +262,6 @@ struct perf_counter_mmap_page {
 #define PERF_EVENT_MISC_KERNEL			(1 << 0)
 #define PERF_EVENT_MISC_USER			(2 << 0)
 #define PERF_EVENT_MISC_HYPERVISOR		(3 << 0)
-#define PERF_EVENT_MISC_OVERFLOW		(1 << 2)
 
 struct perf_event_header {
 	__u32	type;
@@ -348,9 +347,6 @@ enum perf_event_type {
 	PERF_EVENT_READ			= 8,
 
 	/*
-	 * When header.misc & PERF_EVENT_MISC_OVERFLOW the event_type field
-	 * will be PERF_SAMPLE_*
-	 *
 	 * struct {
 	 *	struct perf_event_header	header;
 	 *
@@ -358,8 +354,9 @@ enum perf_event_type {
 	 *	{ u32			pid, tid; } && PERF_SAMPLE_TID
 	 *	{ u64			time;     } && PERF_SAMPLE_TIME
 	 *	{ u64			addr;     } && PERF_SAMPLE_ADDR
-	 *	{ u64			config;   } && PERF_SAMPLE_CONFIG
+	 *	{ u64			id;	  } && PERF_SAMPLE_ID
 	 *	{ u32			cpu, res; } && PERF_SAMPLE_CPU
+	 * 	{ u64			period;   } && PERF_SAMPLE_PERIOD
 	 *
 	 *	{ u64			nr;
 	 *	  { u64 id, val; }	cnt[nr];  } && PERF_SAMPLE_GROUP
@@ -368,6 +365,9 @@ enum perf_event_type {
 	 *	  u64			ips[nr];  } && PERF_SAMPLE_CALLCHAIN
 	 * };
 	 */
+	PERF_EVENT_SAMPLE		= 9,
+
+	PERF_EVENT_MAX,			/* non-ABI */
 };
 
 enum perf_callchain_context {
diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index 385ca51c6e60..f2f232696587 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -2575,15 +2575,14 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
 		u32 cpu, reserved;
 	} cpu_entry;
 
-	header.type = 0;
+	header.type = PERF_EVENT_SAMPLE;
 	header.size = sizeof(header);
 
-	header.misc = PERF_EVENT_MISC_OVERFLOW;
+	header.misc = 0;
 	header.misc |= perf_misc_flags(data->regs);
 
 	if (sample_type & PERF_SAMPLE_IP) {
 		ip = perf_instruction_pointer(data->regs);
-		header.type |= PERF_SAMPLE_IP;
 		header.size += sizeof(ip);
 	}
 
@@ -2592,7 +2591,6 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
 		tid_entry.pid = perf_counter_pid(counter, current);
 		tid_entry.tid = perf_counter_tid(counter, current);
 
-		header.type |= PERF_SAMPLE_TID;
 		header.size += sizeof(tid_entry);
 	}
 
@@ -2602,34 +2600,25 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
 		 */
 		time = sched_clock();
 
-		header.type |= PERF_SAMPLE_TIME;
 		header.size += sizeof(u64);
 	}
 
-	if (sample_type & PERF_SAMPLE_ADDR) {
-		header.type |= PERF_SAMPLE_ADDR;
+	if (sample_type & PERF_SAMPLE_ADDR)
 		header.size += sizeof(u64);
-	}
 
-	if (sample_type & PERF_SAMPLE_ID) {
-		header.type |= PERF_SAMPLE_ID;
+	if (sample_type & PERF_SAMPLE_ID)
 		header.size += sizeof(u64);
-	}
 
 	if (sample_type & PERF_SAMPLE_CPU) {
-		header.type |= PERF_SAMPLE_CPU;
 		header.size += sizeof(cpu_entry);
 
 		cpu_entry.cpu = raw_smp_processor_id();
 	}
 
-	if (sample_type & PERF_SAMPLE_PERIOD) {
-		header.type |= PERF_SAMPLE_PERIOD;
+	if (sample_type & PERF_SAMPLE_PERIOD)
 		header.size += sizeof(u64);
-	}
 
 	if (sample_type & PERF_SAMPLE_GROUP) {
-		header.type |= PERF_SAMPLE_GROUP;
 		header.size += sizeof(u64) +
 			counter->nr_siblings * sizeof(group_entry);
 	}
@@ -2639,10 +2628,9 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
 
 		if (callchain) {
 			callchain_size = (1 + callchain->nr) * sizeof(u64);
-
-			header.type |= PERF_SAMPLE_CALLCHAIN;
 			header.size += callchain_size;
-		}
+		} else
+			header.size += sizeof(u64);
 	}
 
 	ret = perf_output_begin(&handle, counter, header.size, nmi, 1);
@@ -2693,8 +2681,14 @@ static void perf_counter_output(struct perf_counter *counter, int nmi,
 		}
 	}
 
-	if (callchain)
-		perf_output_copy(&handle, callchain, callchain_size);
+	if (sample_type & PERF_SAMPLE_CALLCHAIN) {
+		if (callchain)
+			perf_output_copy(&handle, callchain, callchain_size);
+		else {
+			u64 nr = 0;
+			perf_output_put(&handle, nr);
+		}
+	}
 
 	perf_output_end(&handle);
 }
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index 7e58e3ad1508..722c0f54e549 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -855,7 +855,7 @@ static unsigned long total = 0,
 		     total_unknown = 0;
 
 static int
-process_overflow_event(event_t *event, unsigned long offset, unsigned long head)
+process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 {
 	char level;
 	int show = 0;
@@ -1013,10 +1013,10 @@ process_period_event(event_t *event, unsigned long offset, unsigned long head)
 static int
 process_event(event_t *event, unsigned long offset, unsigned long head)
 {
-	if (event->header.misc & PERF_EVENT_MISC_OVERFLOW)
-		return process_overflow_event(event, offset, head);
-
 	switch (event->header.type) {
+	case PERF_EVENT_SAMPLE:
+		return process_sample_event(event, offset, head);
+
 	case PERF_EVENT_MMAP:
 		return process_mmap_event(event, offset, head);
 
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index e575f3039766..ec5361c67bf5 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -53,6 +53,8 @@ static regex_t		parent_regex;
 
 static int		exclude_other = 1;
 
+static u64		sample_type;
+
 struct ip_event {
 	struct perf_event_header header;
 	u64 ip;
@@ -1135,7 +1137,7 @@ static int validate_chain(struct ip_callchain *chain, event_t *event)
 }
 
 static int
-process_overflow_event(event_t *event, unsigned long offset, unsigned long head)
+process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 {
 	char level;
 	int show = 0;
@@ -1147,12 +1149,12 @@ process_overflow_event(event_t *event, unsigned long offset, unsigned long head)
 	void *more_data = event->ip.__more_data;
 	struct ip_callchain *chain = NULL;
 
-	if (event->header.type & PERF_SAMPLE_PERIOD) {
+	if (sample_type & PERF_SAMPLE_PERIOD) {
 		period = *(u64 *)more_data;
 		more_data += sizeof(u64);
 	}
 
-	dprintf("%p [%p]: PERF_EVENT (IP, %d): %d: %p period: %Ld\n",
+	dprintf("%p [%p]: PERF_EVENT_SAMPLE (IP, %d): %d: %p period: %Ld\n",
 		(void *)(offset + head),
 		(void *)(long)(event->header.size),
 		event->header.misc,
@@ -1160,7 +1162,7 @@ process_overflow_event(event_t *event, unsigned long offset, unsigned long head)
 		(void *)(long)ip,
 		(long long)period);
 
-	if (event->header.type & PERF_SAMPLE_CALLCHAIN) {
+	if (sample_type & PERF_SAMPLE_CALLCHAIN) {
 		int i;
 
 		chain = (void *)more_data;
@@ -1352,10 +1354,10 @@ process_event(event_t *event, unsigned long offset, unsigned long head)
 {
 	trace_event(event);
 
-	if (event->header.misc & PERF_EVENT_MISC_OVERFLOW)
-		return process_overflow_event(event, offset, head);
-
 	switch (event->header.type) {
+	case PERF_EVENT_SAMPLE:
+		return process_sample_event(event, offset, head);
+
 	case PERF_EVENT_MMAP:
 		return process_mmap_event(event, offset, head);
 
@@ -1388,18 +1390,21 @@ process_event(event_t *event, unsigned long offset, unsigned long head)
 
 static struct perf_header	*header;
 
-static int perf_header__has_sample(u64 sample_mask)
+static u64 perf_header__sample_type(void)
 {
+	u64 sample_type = 0;
 	int i;
 
 	for (i = 0; i < header->attrs; i++) {
 		struct perf_header_attr *attr = header->attr[i];
 
-		if (!(attr->attr.sample_type & sample_mask))
-			return 0;
+		if (!sample_type)
+			sample_type = attr->attr.sample_type;
+		else if (sample_type != attr->attr.sample_type)
+			die("non matching sample_type");
 	}
 
-	return 1;
+	return sample_type;
 }
 
 static int __cmd_report(void)
@@ -1437,8 +1442,9 @@ static int __cmd_report(void)
 	header = perf_header__read(input);
 	head = header->data_offset;
 
-	if (sort__has_parent &&
-	    !perf_header__has_sample(PERF_SAMPLE_CALLCHAIN)) {
+	sample_type = perf_header__sample_type();
+
+	if (sort__has_parent && !(sample_type & PERF_SAMPLE_CALLCHAIN)) {
 		fprintf(stderr, "selected --sort parent, but no callchain data\n");
 		exit(-1);
 	}
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 5352b5e352ed..cf0d21f1ae10 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -392,11 +392,11 @@ static void record_ip(u64 ip, int counter)
 	samples--;
 }
 
-static void process_event(u64 ip, int counter)
+static void process_event(u64 ip, int counter, int user)
 {
 	samples++;
 
-	if (ip < min_ip || ip > max_ip) {
+	if (user) {
 		userspace_samples++;
 		return;
 	}
@@ -509,9 +509,10 @@ static void mmap_read_counter(struct mmap_data *md)
 
 		old += size;
 
-		if (event->header.misc & PERF_EVENT_MISC_OVERFLOW) {
-			if (event->header.type & PERF_SAMPLE_IP)
-				process_event(event->ip.ip, md->counter);
+		if (event->header.type == PERF_EVENT_SAMPLE) {
+			int user =
+	(event->header.misc & PERF_EVENT_MISC_CPUMODE_MASK) == PERF_EVENT_MISC_USER;
+			process_event(event->ip.ip, md->counter, user);
 		}
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 649c48a9e7fafcc72bfcc99471d9dea98d789d59 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Wed, 24 Jun 2009 21:12:48 +0200
Subject: perf-report: Add modes for inherited stats and no-samples

Now that we can collect per task statistics, add modes that
make use of that facility.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-record.c | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index f4f0240d2302..798a56d890e5 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -41,6 +41,8 @@ static int			force				= 0;
 static int			append_file			= 0;
 static int			call_graph			= 0;
 static int			verbose				= 0;
+static int			inherit_stat			= 0;
+static int			no_samples			= 0;
 
 static long			samples;
 static struct timeval		last_read;
@@ -393,6 +395,12 @@ static void create_counter(int counter, int cpu, pid_t pid)
 		attr->sample_freq	= freq;
 	}
 
+	if (no_samples)
+		attr->sample_freq = 0;
+
+	if (inherit_stat)
+		attr->inherit_stat = 1;
+
 	if (call_graph)
 		attr->sample_type	|= PERF_SAMPLE_CALLCHAIN;
 
@@ -571,7 +579,7 @@ static int __cmd_record(int argc, const char **argv)
 		}
 	}
 
-	while (!done) {
+	for (;;) {
 		int hits = samples;
 
 		for (i = 0; i < nr_cpu; i++) {
@@ -579,8 +587,11 @@ static int __cmd_record(int argc, const char **argv)
 				mmap_read(&mmap_array[i][counter]);
 		}
 
-		if (hits == samples)
+		if (hits == samples) {
+			if (done)
+				break;
 			ret = poll(event_array, nr_poll, 100);
+		}
 	}
 
 	/*
@@ -629,6 +640,10 @@ static const struct option options[] = {
 		    "do call-graph (stack chain/backtrace) recording"),
 	OPT_BOOLEAN('v', "verbose", &verbose,
 		    "be more verbose (show counter open errors, etc)"),
+	OPT_BOOLEAN('s', "stat", &inherit_stat,
+		    "per thread counts"),
+	OPT_BOOLEAN('n', "no-samples", &no_samples,
+		    "don't sample"),
 	OPT_END()
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From e9ea2fde7a07ae60a119171a2946ed2ae778271e Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Wed, 24 Jun 2009 22:46:04 +0200
Subject: perf-report: Add bare minimum PERF_EVENT_READ parsing

Provide the basic infrastructure to provide per task stats.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-report.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index ec5361c67bf5..681c2233f882 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -100,6 +100,13 @@ struct lost_event {
 	u64 lost;
 };
 
+struct read_event {
+	struct perf_event_header header;
+	u32 pid,tid;
+	u64 value;
+	u64 format[3];
+};
+
 typedef union event_union {
 	struct perf_event_header	header;
 	struct ip_event			ip;
@@ -108,6 +115,7 @@ typedef union event_union {
 	struct fork_event		fork;
 	struct period_event		period;
 	struct lost_event		lost;
+	struct read_event		read;
 } event_t;
 
 static LIST_HEAD(dsos);
@@ -1349,6 +1357,19 @@ static void trace_event(event_t *event)
 	dprintf(".\n");
 }
 
+static int
+process_read_event(event_t *event, unsigned long offset, unsigned long head)
+{
+	dprintf("%p [%p]: PERF_EVENT_READ: %d %d %Lu\n",
+			(void *)(offset + head),
+			(void *)(long)(event->header.size),
+			event->read.pid,
+			event->read.tid,
+			event->read.value);
+
+	return 0;
+}
+
 static int
 process_event(event_t *event, unsigned long offset, unsigned long head)
 {
@@ -1373,6 +1394,9 @@ process_event(event_t *event, unsigned long offset, unsigned long head)
 	case PERF_EVENT_LOST:
 		return process_lost_event(event, offset, head);
 
+	case PERF_EVENT_READ:
+		return process_read_event(event, offset, head);
+
 	/*
 	 * We dont process them right now but they are fine:
 	 */
-- 
cgit v1.2.3-59-g8ed1b


From 4418351f06d9ce73acc846158c20186965f920f3 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Thu, 25 Jun 2009 21:27:42 +0530
Subject: perf_counter tools: Add alias for 'l1d' and 'l1i'

Add 'l1d' and 'l1i' aliases again as shortcuts - just dont make them
the primary display alias.

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1245945462.9157.11.camel@hpdv5.satnam>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/util/parse-events.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 430f06083201..4d042f104cdc 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -71,8 +71,8 @@ static char *sw_event_names[] = {
 #define MAX_ALIASES 8
 
 static char *hw_cache[][MAX_ALIASES] = {
- { "L1-d$",	"l1-d",		"L1-data",				},
- { "L1-i$",	"l1-i",		"L1-instruction",			},
+ { "L1-d$",	"l1-d",		"l1d",		"L1-data",		},
+ { "L1-i$",	"l1-i",		"l1i",		"L1-instruction",	},
  { "LLC",	"L2"							},
  { "dTLB",	"d-tlb",	"Data-TLB",				},
  { "iTLB",	"i-tlb",	"Instruction-TLB",			},
-- 
cgit v1.2.3-59-g8ed1b


From 5211a242d0cbdded372aee59da18f80552b0a80a Mon Sep 17 00:00:00 2001
From: Kurt Garloff <garloff@suse.de>
Date: Wed, 24 Jun 2009 14:32:11 -0700
Subject: x86: Add sysctl to allow panic on IOCK NMI error

This patch introduces a new sysctl:

    /proc/sys/kernel/panic_on_io_nmi

which defaults to 0 (off).

When enabled, the kernel panics when the kernel receives an NMI
caused by an IO error.

The IO error triggered NMI indicates a serious system
condition, which could result in IO data corruption. Rather
than contiuing, panicing and dumping might be a better choice,
so one can figure out what's causing the IO error.

This could be especially important to companies running IO
intensive applications where corruption must be avoided, e.g. a
bank's databases.

[ SuSE has been shipping it for a while, it was done at the
  request of a large database vendor, for their users. ]

Signed-off-by: Kurt Garloff <garloff@suse.de>
Signed-off-by: Roberto Angelino <robertangelino@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
LKML-Reference: <20090624213211.GA11291@kroah.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/dumpstack.c | 1 +
 arch/x86/kernel/traps.c     | 3 +++
 include/linux/kernel.h      | 1 +
 kernel/sysctl.c             | 8 ++++++++
 4 files changed, 13 insertions(+)

diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c
index 95ea5fa7d444..c8405718a4c3 100644
--- a/arch/x86/kernel/dumpstack.c
+++ b/arch/x86/kernel/dumpstack.c
@@ -22,6 +22,7 @@
 #include "dumpstack.h"
 
 int panic_on_unrecovered_nmi;
+int panic_on_io_nmi;
 unsigned int code_bytes = 64;
 int kstack_depth_to_print = 3 * STACKSLOTS_PER_LINE;
 static int die_counter;
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index a0f48f5671c0..5204332f475d 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -346,6 +346,9 @@ io_check_error(unsigned char reason, struct pt_regs *regs)
 	printk(KERN_EMERG "NMI: IOCK error (debug interrupt?)\n");
 	show_registers(regs);
 
+	if (panic_on_io_nmi)
+		panic("NMI IOCK error: Not continuing");
+
 	/* Re-enable the IOCK line, wait for a few seconds */
 	reason = (reason & 0xf) | 8;
 	outb(reason, 0x61);
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index fac104e7186a..d6320a3e8def 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -303,6 +303,7 @@ extern int oops_in_progress;		/* If set, an oops, panic(), BUG() or die() is in
 extern int panic_timeout;
 extern int panic_on_oops;
 extern int panic_on_unrecovered_nmi;
+extern int panic_on_io_nmi;
 extern const char *print_tainted(void);
 extern void add_taint(unsigned flag);
 extern int test_taint(unsigned flag);
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 62e4ff9968b5..fba42eda8de2 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -743,6 +743,14 @@ static struct ctl_table kern_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "panic_on_io_nmi",
+		.data		= &panic_on_io_nmi,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
 	{
 		.ctl_name	= KERN_BOOTLOADER_TYPE,
 		.procname	= "bootloader_type",
-- 
cgit v1.2.3-59-g8ed1b


From 3928ddbe994cce1da1b6365b0db04d5765f254f4 Mon Sep 17 00:00:00 2001
From: Frederic Weisbecker <fweisbec@gmail.com>
Date: Thu, 25 Jun 2009 22:21:27 +0200
Subject: perf record: Fix unhandled io return value
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Building latest perfcounter fails on the following error:

 builtin-record.c: In function ‘create_counter’:
 builtin-record.c:451: erreur: ignoring return value of ‘read’, declared with attribute warn_unused_result
 make: *** [builtin-record.o] Erreur 1

Just check if we successfully read the perf file descriptor.

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <1245961287-5327-1-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-record.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 798a56d890e5..d18546f37d7c 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -448,7 +448,10 @@ try_again:
 		}
 	}
 
-	read(fd[nr_cpu][counter], &read_data, sizeof(read_data));
+	if (read(fd[nr_cpu][counter], &read_data, sizeof(read_data)) == -1) {
+		perror("Unable to read perf file descriptor\n");
+		exit(-1);
+	}
 
 	perf_header_attr__add_id(h_attr, read_data.id);
 
-- 
cgit v1.2.3-59-g8ed1b


From 5be6066a7f8d917db347d94f1b359b9b70dcb572 Mon Sep 17 00:00:00 2001
From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Date: Wed, 24 Jun 2009 09:21:10 +0900
Subject: x86, mce: percpu mcheck_timer should be pinned

If CONFIG_NO_HZ + CONFIG_SMP, timer added via add_timer() might
be migrated on other cpu.  Use add_timer_on() instead.

Avoids the following failure:

Maciej Rutecki wrote:
> > After normal boot I try:
> >
> > echo 1 > /sys/devices/system/machinecheck/machinecheck0/check_interval
> >
> > I found this in dmesg:
> >
> > [  141.704025] ------------[ cut here ]------------
> > [  141.704039] WARNING: at arch/x86/kernel/cpu/mcheck/mce.c:1102
> > mcheck_timer+0xf5/0x100()

Reported-by: Maciej Rutecki <maciej.rutecki@gmail.com>
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Tested-by: Maciej Rutecki <maciej.rutecki@gmail.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/kernel/cpu/mcheck/mce.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 284d1de968bc..af425b83202b 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -1117,7 +1117,7 @@ static void mcheck_timer(unsigned long data)
 		*n = min(*n*2, (int)round_jiffies_relative(check_interval*HZ));
 
 	t->expires = jiffies + *n;
-	add_timer(t);
+	add_timer_on(t, smp_processor_id());
 }
 
 static void mce_do_trigger(struct work_struct *work)
@@ -1321,7 +1321,7 @@ static void mce_init_timer(void)
 		return;
 	setup_timer(t, mcheck_timer, smp_processor_id());
 	t->expires = round_jiffies(jiffies + *n);
-	add_timer(t);
+	add_timer_on(t, smp_processor_id());
 }
 
 /*
-- 
cgit v1.2.3-59-g8ed1b


From 22f4319d6bc0155e6c0ae560729baa6c09dc09e7 Mon Sep 17 00:00:00 2001
From: "Robert P. J. Day" <rpjday@crashcourse.ca>
Date: Thu, 25 Jun 2009 16:20:48 -0400
Subject: x86, setup: Fix typo "CONFIG_x86_64" in <asm/boot.h>

CONFIG_X86_64 was misspelled (wrong case), which caused the x86-64
kernel to advertise itself as more relocatable than it really is.
This could in theory cause boot failures once bootloaders start
support the new relocation fields.

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/boot.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/include/asm/boot.h b/arch/x86/include/asm/boot.h
index 418e632d4a80..6f02b9a53848 100644
--- a/arch/x86/include/asm/boot.h
+++ b/arch/x86/include/asm/boot.h
@@ -16,7 +16,7 @@
 				& ~(CONFIG_PHYSICAL_ALIGN - 1))
 
 /* Minimum kernel alignment, as a power of two */
-#ifdef CONFIG_x86_64
+#ifdef CONFIG_X86_64
 #define MIN_KERNEL_ALIGN_LG2	PMD_SHIFT
 #else
 #define MIN_KERNEL_ALIGN_LG2	(PAGE_SHIFT+1)
-- 
cgit v1.2.3-59-g8ed1b


From 658dbfeb5e7ab35d440c665d643a6285e43fddcd Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Thu, 25 Jun 2009 15:16:06 -0700
Subject: x86, setup: correct include file in <asm/boot.h>

<asm/boot.h> needs <asm/pgtable_types.h>, not <asm/page_types.h> in
order to resolve PMD_SHIFT.  Also, correct a +1 which really should be
+ THREAD_ORDER.

This is a build error which was masked by a typoed #ifdef.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/include/asm/boot.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86/include/asm/boot.h b/arch/x86/include/asm/boot.h
index 6f02b9a53848..7a1065958ba9 100644
--- a/arch/x86/include/asm/boot.h
+++ b/arch/x86/include/asm/boot.h
@@ -8,7 +8,7 @@
 
 #ifdef __KERNEL__
 
-#include <asm/page_types.h>
+#include <asm/pgtable_types.h>
 
 /* Physical address where kernel should be loaded. */
 #define LOAD_PHYSICAL_ADDR ((CONFIG_PHYSICAL_START \
@@ -19,7 +19,7 @@
 #ifdef CONFIG_X86_64
 #define MIN_KERNEL_ALIGN_LG2	PMD_SHIFT
 #else
-#define MIN_KERNEL_ALIGN_LG2	(PAGE_SHIFT+1)
+#define MIN_KERNEL_ALIGN_LG2	(PAGE_SHIFT + THREAD_ORDER)
 #endif
 #define MIN_KERNEL_ALIGN	(_AC(1, UL) << MIN_KERNEL_ALIGN_LG2)
 
-- 
cgit v1.2.3-59-g8ed1b


From e888d7facd1f1460a638151036d15b6cfb3ccc74 Mon Sep 17 00:00:00 2001
From: "Pallipadi, Venkatesh" <venkatesh.pallipadi@intel.com>
Date: Thu, 25 Jun 2009 16:44:31 -0700
Subject: x86, delay: tsc based udelay should have rdtsc_barrier

delay_tsc needs rdtsc_barrier to provide proper delay.

Output from a test driver using hpet to cross check delay
provided by udelay().

Before:
[   86.794363] Expected delay 5us actual 4679ns
[   87.154362] Expected delay 5us actual 698ns
[   87.514162] Expected delay 5us actual 4539ns
[   88.653716] Expected delay 5us actual 4539ns
[   94.664106] Expected delay 10us actual 9638ns
[   95.049351] Expected delay 10us actual 10126ns
[   95.416110] Expected delay 10us actual 9568ns
[   95.799216] Expected delay 10us actual 9638ns
[  103.624104] Expected delay 10us actual 9707ns
[  104.020619] Expected delay 10us actual 768ns
[  104.419951] Expected delay 10us actual 9707ns

After:
[   50.983320] Expected delay 5us actual 5587ns
[   51.261807] Expected delay 5us actual 5587ns
[   51.565715] Expected delay 5us actual 5657ns
[   51.861171] Expected delay 5us actual 5587ns
[   52.164704] Expected delay 5us actual 5726ns
[   52.487457] Expected delay 5us actual 5657ns
[   52.789338] Expected delay 5us actual 5726ns
[   57.119680] Expected delay 10us actual 10755ns
[   57.893997] Expected delay 10us actual 10615ns
[   58.261287] Expected delay 10us actual 10755ns
[   58.620505] Expected delay 10us actual 10825ns
[   58.941035] Expected delay 10us actual 10755ns
[   59.320903] Expected delay 10us actual 10615ns
[   61.306311] Expected delay 10us actual 10755ns
[   61.520542] Expected delay 10us actual 10615ns

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
---
 arch/x86/lib/delay.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/x86/lib/delay.c b/arch/x86/lib/delay.c
index f4568605d7d5..ff485d361182 100644
--- a/arch/x86/lib/delay.c
+++ b/arch/x86/lib/delay.c
@@ -55,8 +55,10 @@ static void delay_tsc(unsigned long loops)
 
 	preempt_disable();
 	cpu = smp_processor_id();
+	rdtsc_barrier();
 	rdtscl(bclock);
 	for (;;) {
+		rdtsc_barrier();
 		rdtscl(now);
 		if ((now - bclock) >= loops)
 			break;
@@ -78,6 +80,7 @@ static void delay_tsc(unsigned long loops)
 		if (unlikely(cpu != smp_processor_id())) {
 			loops -= (now - bclock);
 			cpu = smp_processor_id();
+			rdtsc_barrier();
 			rdtscl(bclock);
 		}
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 14a2ff6ed28931f796d2c2c8a440227a5d90f441 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Thu, 25 Jun 2009 19:00:47 -0700
Subject: sparc64: Don't use alloc_bootmem() in init_IRQ() code paths.

The page allocator and SLAB are available at this point now,
and if we still try to use bootmem allocations here the kernel
spits out warnings.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc/kernel/irq_64.c | 45 +++++++++++++++++++--------------------------
 1 file changed, 19 insertions(+), 26 deletions(-)

diff --git a/arch/sparc/kernel/irq_64.c b/arch/sparc/kernel/irq_64.c
index bd075054942b..f0ee79055409 100644
--- a/arch/sparc/kernel/irq_64.c
+++ b/arch/sparc/kernel/irq_64.c
@@ -20,7 +20,6 @@
 #include <linux/delay.h>
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
-#include <linux/bootmem.h>
 #include <linux/irq.h>
 
 #include <asm/ptrace.h>
@@ -914,25 +913,19 @@ void __cpuinit notrace sun4v_register_mondo_queues(int this_cpu)
 			   tb->nonresum_qmask);
 }
 
-static void __init alloc_one_mondo(unsigned long *pa_ptr, unsigned long qmask)
-{
-	unsigned long size = PAGE_ALIGN(qmask + 1);
-	void *p = __alloc_bootmem(size, size, 0);
-	if (!p) {
-		prom_printf("SUN4V: Error, cannot allocate mondo queue.\n");
-		prom_halt();
-	}
-
-	*pa_ptr = __pa(p);
-}
-
-static void __init alloc_one_kbuf(unsigned long *pa_ptr, unsigned long qmask)
+/* Each queue region must be a power of 2 multiple of 64 bytes in
+ * size.  The base real address must be aligned to the size of the
+ * region.  Thus, an 8KB queue must be 8KB aligned, for example.
+ */
+static void __init alloc_one_queue(unsigned long *pa_ptr, unsigned long qmask)
 {
 	unsigned long size = PAGE_ALIGN(qmask + 1);
-	void *p = __alloc_bootmem(size, size, 0);
+	unsigned long order = get_order(size);
+	unsigned long p;
 
+	p = __get_free_pages(GFP_KERNEL, order);
 	if (!p) {
-		prom_printf("SUN4V: Error, cannot allocate kbuf page.\n");
+		prom_printf("SUN4V: Error, cannot allocate queue.\n");
 		prom_halt();
 	}
 
@@ -942,11 +935,11 @@ static void __init alloc_one_kbuf(unsigned long *pa_ptr, unsigned long qmask)
 static void __init init_cpu_send_mondo_info(struct trap_per_cpu *tb)
 {
 #ifdef CONFIG_SMP
-	void *page;
+	unsigned long page;
 
 	BUILD_BUG_ON((NR_CPUS * sizeof(u16)) > (PAGE_SIZE - 64));
 
-	page = alloc_bootmem_pages(PAGE_SIZE);
+	page = get_zeroed_page(GFP_KERNEL);
 	if (!page) {
 		prom_printf("SUN4V: Error, cannot allocate cpu mondo page.\n");
 		prom_halt();
@@ -965,13 +958,13 @@ static void __init sun4v_init_mondo_queues(void)
 	for_each_possible_cpu(cpu) {
 		struct trap_per_cpu *tb = &trap_block[cpu];
 
-		alloc_one_mondo(&tb->cpu_mondo_pa, tb->cpu_mondo_qmask);
-		alloc_one_mondo(&tb->dev_mondo_pa, tb->dev_mondo_qmask);
-		alloc_one_mondo(&tb->resum_mondo_pa, tb->resum_qmask);
-		alloc_one_kbuf(&tb->resum_kernel_buf_pa, tb->resum_qmask);
-		alloc_one_mondo(&tb->nonresum_mondo_pa, tb->nonresum_qmask);
-		alloc_one_kbuf(&tb->nonresum_kernel_buf_pa,
-			       tb->nonresum_qmask);
+		alloc_one_queue(&tb->cpu_mondo_pa, tb->cpu_mondo_qmask);
+		alloc_one_queue(&tb->dev_mondo_pa, tb->dev_mondo_qmask);
+		alloc_one_queue(&tb->resum_mondo_pa, tb->resum_qmask);
+		alloc_one_queue(&tb->resum_kernel_buf_pa, tb->resum_qmask);
+		alloc_one_queue(&tb->nonresum_mondo_pa, tb->nonresum_qmask);
+		alloc_one_queue(&tb->nonresum_kernel_buf_pa,
+				tb->nonresum_qmask);
 	}
 }
 
@@ -999,7 +992,7 @@ void __init init_IRQ(void)
 	kill_prom_timer();
 
 	size = sizeof(struct ino_bucket) * NUM_IVECS;
-	ivector_table = alloc_bootmem(size);
+	ivector_table = kzalloc(size, GFP_KERNEL);
 	if (!ivector_table) {
 		prom_printf("Fatal error, cannot allocate ivector_table\n");
 		prom_halt();
-- 
cgit v1.2.3-59-g8ed1b


From 413ee282a510afb2f18975a189501f39d279a906 Mon Sep 17 00:00:00 2001
From: Julian Calaby <julian.calaby@gmail.com>
Date: Sun, 21 Jun 2009 16:44:13 +0000
Subject: sparc64: Fix build warnings in piggyback_64.c

This patch fixes the following build warnings:
arch/sparc/boot/piggyback_64.c: In function 'main':
arch/sparc/boot/piggyback_64.c:44: warning: 'end' may be used uninitialized in this function
arch/sparc/boot/piggyback_64.c:44: warning: 'start' may be used uninitialized in this function

Signed-off-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc/boot/piggyback_64.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/sparc/boot/piggyback_64.c b/arch/sparc/boot/piggyback_64.c
index de364bfed0bb..c63fd1b6bdd4 100644
--- a/arch/sparc/boot/piggyback_64.c
+++ b/arch/sparc/boot/piggyback_64.c
@@ -46,6 +46,7 @@ int main(int argc,char **argv)
 	struct stat s;
 	int image, tail;
 	
+	start = end = 0;
 	if (stat (argv[3], &s) < 0) die (argv[3]);
 	map = fopen (argv[2], "r");
 	if (!map) die(argv[2]);
-- 
cgit v1.2.3-59-g8ed1b


From 22b096a8907e5184f25fafd1b73f0b3633d52495 Mon Sep 17 00:00:00 2001
From: Robert Reif <reif@earthlink.net>
Date: Sun, 21 Jun 2009 16:45:44 +0000
Subject: sparc32: Fix obvious build issues for tftpboot.img build.

Signed-off-by: Robert Reif <reif@earthlink.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc/boot/Makefile       | 2 +-
 arch/sparc/boot/piggyback_32.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/sparc/boot/Makefile b/arch/sparc/boot/Makefile
index 96041a8d39e8..5e2caeb31109 100644
--- a/arch/sparc/boot/Makefile
+++ b/arch/sparc/boot/Makefile
@@ -58,7 +58,7 @@ $(obj)/image: $(obj)/btfix.o FORCE
 $(obj)/zImage: $(obj)/image
 	$(call if_changed,strip)
 
-$(obj)/tftpboot.img: $(obj)/piggyback $(obj)/System.map $(obj)/image FORCE
+$(obj)/tftpboot.img: $(obj)/piggyback_32 $(obj)/System.map $(obj)/image FORCE
 	$(call if_changed,elftoaout)
 	$(call if_changed,piggy)
 
diff --git a/arch/sparc/boot/piggyback_32.c b/arch/sparc/boot/piggyback_32.c
index c9f500c1a8b2..3f0f93354543 100644
--- a/arch/sparc/boot/piggyback_32.c
+++ b/arch/sparc/boot/piggyback_32.c
@@ -70,7 +70,7 @@ void die(char *str)
 int main(int argc,char **argv)
 {
 	static char aout_magic[] = { 0x01, 0x03, 0x01, 0x07 };
-	unsigned char buffer[1024], *q, *r;
+	char buffer[1024], *q, *r;
 	unsigned int i, j, k, start, end, offset;
 	FILE *map;
 	struct stat s;
-- 
cgit v1.2.3-59-g8ed1b


From 52da82cfb569b44e26e15395a6727277758580fe Mon Sep 17 00:00:00 2001
From: Sam Ravnborg <sam@ravnborg.org>
Date: Sun, 21 Jun 2009 16:46:10 +0000
Subject: sparc: fix tftpboot.img build

Kjetil Oftedal mentioned that piggyback_32 was failing
when building a sparc image.

I tracked this down to the fact that the kernel no longer
provided an absolute symbol named "end".

Commit 86ed40bd6fe511d26bb8f3fa65a84cb65c235366 ("sparc: unify sections.h")
renamed end to _end but failed to update piggyback_32.

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Cc: Kjetil Oftedal <oftedal@gmail.com>
Cc: Robert Reif <reif@earthlink.net>
Signed-off-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc/boot/piggyback_32.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/sparc/boot/piggyback_32.c b/arch/sparc/boot/piggyback_32.c
index 3f0f93354543..e8dc9adfcd61 100644
--- a/arch/sparc/boot/piggyback_32.c
+++ b/arch/sparc/boot/piggyback_32.c
@@ -84,7 +84,7 @@ int main(int argc,char **argv)
 	while (fgets (buffer, 1024, map)) {
 		if (!strcmp (buffer + 8, " T start\n") || !strcmp (buffer + 16, " T start\n"))
 			start = strtoul (buffer, NULL, 16);
-		else if (!strcmp (buffer + 8, " A end\n") || !strcmp (buffer + 16, " A end\n"))
+		else if (!strcmp (buffer + 8, " A _end\n") || !strcmp (buffer + 16, " A _end\n"))
 			end = strtoul (buffer, NULL, 16);
 	}
 	fclose (map);
-- 
cgit v1.2.3-59-g8ed1b


From 3e05c5e2ce40066582dc34aa8335baa328815a09 Mon Sep 17 00:00:00 2001
From: Julian Calaby <julian.calaby@gmail.com>
Date: Sun, 21 Jun 2009 16:45:01 +0000
Subject: sparc32: Fix tftpboot.img Makefile

Signed-off-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc/boot/Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/sparc/boot/Makefile b/arch/sparc/boot/Makefile
index 5e2caeb31109..500a2f87f28f 100644
--- a/arch/sparc/boot/Makefile
+++ b/arch/sparc/boot/Makefile
@@ -15,7 +15,7 @@ quiet_cmd_elftoaout	= ELFTOAOUT $@
 
 ifeq ($(CONFIG_SPARC32),y)
 quiet_cmd_piggy		= PIGGY   $@
-      cmd_piggy		= $(obj)/piggyback_32 $@ $(obj)/System.map $(ROOT_IMG)
+      cmd_piggy		= $(obj)/piggyback_32 $@ System.map $(ROOT_IMG)
 quiet_cmd_btfix		= BTFIX   $@
       cmd_btfix		= $(OBJDUMP) -x vmlinux | $(obj)/btfixupprep > $@
 quiet_cmd_sysmap        = SYSMAP  $(obj)/System.map
@@ -58,7 +58,7 @@ $(obj)/image: $(obj)/btfix.o FORCE
 $(obj)/zImage: $(obj)/image
 	$(call if_changed,strip)
 
-$(obj)/tftpboot.img: $(obj)/piggyback_32 $(obj)/System.map $(obj)/image FORCE
+$(obj)/tftpboot.img: $(obj)/piggyback_32 System.map $(ROOT_IMG) FORCE
 	$(call if_changed,elftoaout)
 	$(call if_changed,piggy)
 
-- 
cgit v1.2.3-59-g8ed1b


From 8944146daa2c38dd85bc489d1b84fb9abc108837 Mon Sep 17 00:00:00 2001
From: Julian Calaby <julian.calaby@gmail.com>
Date: Tue, 23 Jun 2009 01:45:46 +0000
Subject: sparc32: Fix makefile not generating required files

The tftpboot build was failing with missing file errors.

It turns out that $(obj)/image wasn't being generated which was causing the a.out conversion to be skipped and hence piggyback to be called with nonexistent files.

Signed-off-by: Julian Calaby <julian.calaby@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc/boot/Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/sparc/boot/Makefile b/arch/sparc/boot/Makefile
index 500a2f87f28f..1ff0fd924756 100644
--- a/arch/sparc/boot/Makefile
+++ b/arch/sparc/boot/Makefile
@@ -58,7 +58,7 @@ $(obj)/image: $(obj)/btfix.o FORCE
 $(obj)/zImage: $(obj)/image
 	$(call if_changed,strip)
 
-$(obj)/tftpboot.img: $(obj)/piggyback_32 System.map $(ROOT_IMG) FORCE
+$(obj)/tftpboot.img: $(obj)/image $(obj)/piggyback_32 System.map $(ROOT_IMG) FORCE
 	$(call if_changed,elftoaout)
 	$(call if_changed,piggy)
 
@@ -79,7 +79,7 @@ $(obj)/image: vmlinux FORCE
 	$(call if_changed,strip)
 	@echo '  kernel: $@ is ready'
 
-$(obj)/tftpboot.img: vmlinux $(obj)/piggyback_64 System.map $(ROOT_IMG) FORCE
+$(obj)/tftpboot.img: $(obj)/image $(obj)/piggyback_64 System.map $(ROOT_IMG) FORCE
 	$(call if_changed,elftoaout)
 	$(call if_changed,piggy)
 	@echo '  kernel: $@ is ready'
-- 
cgit v1.2.3-59-g8ed1b


From 1ac530b3553e0b4dc1e18a32bed57cfa84cd57cb Mon Sep 17 00:00:00 2001
From: Wei Yongjun <yjwei@cn.fujitsu.com>
Date: Wed, 24 Jun 2009 22:29:31 +0000
Subject: tcp: missing check ACK flag of received segment in FIN-WAIT-2 state

RFC0793 defined that in FIN-WAIT-2 state if the ACK bit is off drop
the segment and return[Page 72]. But this check is missing in function
tcp_timewait_state_process(). This cause the segment with FIN flag but
no ACK has two diffent action:

Case 1:
    Node A                      Node B
              <-------------    FIN,ACK
                                (enter FIN-WAIT-1)
    ACK       ------------->
                                (enter FIN-WAIT-2)
    FIN       ------------->    discard
                                (move sk to tw list)

Case 2:
    Node A                      Node B
              <-------------    FIN,ACK
                                (enter FIN-WAIT-1)
    ACK       ------------->
                                (enter FIN-WAIT-2)
                                (move sk to tw list)
    FIN       ------------->

              <-------------    ACK

This patch fixed the problem.

Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_minisocks.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 43bbba7926ee..f8d67ccc64f3 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -128,7 +128,8 @@ tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb,
 			goto kill_with_rst;
 
 		/* Dup ACK? */
-		if (!after(TCP_SKB_CB(skb)->end_seq, tcptw->tw_rcv_nxt) ||
+		if (!th->ack ||
+		    !after(TCP_SKB_CB(skb)->end_seq, tcptw->tw_rcv_nxt) ||
 		    TCP_SKB_CB(skb)->end_seq == TCP_SKB_CB(skb)->seq) {
 			inet_twsk_put(tw);
 			return TCP_TW_SUCCESS;
-- 
cgit v1.2.3-59-g8ed1b


From a1faa69810b2af562b70b2a71c116c7d03575dd3 Mon Sep 17 00:00:00 2001
From: Jens Rosenboom <me@jayr.de>
Date: Thu, 25 Jun 2009 04:55:50 +0000
Subject: ipv6: avoid wraparound for expired preferred lifetime

Avoid showing wrong high values when the preferred lifetime of an address
is expired.

Signed-off-by: Jens Rosenboom <me@jayr.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/addrconf.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 8c1e86afbbf5..3883b4036a74 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -3362,7 +3362,10 @@ static int inet6_fill_ifaddr(struct sk_buff *skb, struct inet6_ifaddr *ifa,
 		valid = ifa->valid_lft;
 		if (preferred != INFINITY_LIFE_TIME) {
 			long tval = (jiffies - ifa->tstamp)/HZ;
-			preferred -= tval;
+			if (preferred > tval)
+				preferred -= tval;
+			else
+				preferred = 0;
 			if (valid != INFINITY_LIFE_TIME)
 				valid -= tval;
 		}
-- 
cgit v1.2.3-59-g8ed1b


From e2a61fa31382b1ac2b22f76f9c439892e5dc4b86 Mon Sep 17 00:00:00 2001
From: Ionut Nicu <ionut.nicu@freescale.com>
Date: Wed, 24 Jun 2009 22:23:39 +0000
Subject: fsl_pq_mdio: Fix fsl_pq_mdio to work with modules

This patch fixes the case when ucc_geth or gianfar are compiled
as modules. Without this patch the call to phy_connect() fails.

Signed-off-by: Ionut Nicu <ionut.nicu@freescale.com>
Acked-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/fsl_pq_mdio.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/net/fsl_pq_mdio.c b/drivers/net/fsl_pq_mdio.c
index 3af581303ca2..d167090248e2 100644
--- a/drivers/net/fsl_pq_mdio.c
+++ b/drivers/net/fsl_pq_mdio.c
@@ -188,7 +188,7 @@ static int fsl_pq_mdio_find_free(struct mii_bus *new_bus)
 }
 
 
-#ifdef CONFIG_GIANFAR
+#if defined(CONFIG_GIANFAR) || defined(CONFIG_GIANFAR_MODULE)
 static u32 __iomem *get_gfar_tbipa(struct fsl_pq_mdio __iomem *regs)
 {
 	struct gfar __iomem *enet_regs;
@@ -206,7 +206,7 @@ static u32 __iomem *get_gfar_tbipa(struct fsl_pq_mdio __iomem *regs)
 #endif
 
 
-#ifdef CONFIG_UCC_GETH
+#if defined(CONFIG_UCC_GETH) || defined(CONFIG_UCC_GETH_MODULE)
 static int get_ucc_id_for_range(u64 start, u64 end, u32 *ucc_id)
 {
 	struct device_node *np = NULL;
@@ -291,7 +291,7 @@ static int fsl_pq_mdio_probe(struct of_device *ofdev,
 	if (of_device_is_compatible(np, "fsl,gianfar-mdio") ||
 			of_device_is_compatible(np, "fsl,gianfar-tbi") ||
 			of_device_is_compatible(np, "gianfar")) {
-#ifdef CONFIG_GIANFAR
+#if defined(CONFIG_GIANFAR) || defined(CONFIG_GIANFAR_MODULE)
 		tbipa = get_gfar_tbipa(regs);
 #else
 		err = -ENODEV;
@@ -299,7 +299,7 @@ static int fsl_pq_mdio_probe(struct of_device *ofdev,
 #endif
 	} else if (of_device_is_compatible(np, "fsl,ucc-mdio") ||
 			of_device_is_compatible(np, "ucc_geth_phy")) {
-#ifdef CONFIG_UCC_GETH
+#if defined(CONFIG_UCC_GETH) || defined(CONFIG_UCC_GETH_MODULE)
 		u32 id;
 		static u32 mii_mng_master;
 
-- 
cgit v1.2.3-59-g8ed1b


From 37c8ae3acf39108b3b7866897f1e3e9f277efc50 Mon Sep 17 00:00:00 2001
From: roel kluin <roel.kluin@gmail.com>
Date: Mon, 22 Jun 2009 07:38:00 +0000
Subject: sh_eth: remove redundant test on unsigned

Unsigned boguscnt cannot be less than 0.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/sh_eth.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/drivers/net/sh_eth.c b/drivers/net/sh_eth.c
index 341882f959f3..a2d82ddb3b4d 100644
--- a/drivers/net/sh_eth.c
+++ b/drivers/net/sh_eth.c
@@ -865,8 +865,7 @@ static irqreturn_t sh_eth_interrupt(int irq, void *netdev)
 	struct sh_eth_private *mdp = netdev_priv(ndev);
 	struct sh_eth_cpu_data *cd = mdp->cd;
 	irqreturn_t ret = IRQ_NONE;
-	u32 ioaddr, boguscnt = RX_RING_SIZE;
-	u32 intr_status = 0;
+	u32 ioaddr, intr_status = 0;
 
 	ioaddr = ndev->base_addr;
 	spin_lock(&mdp->lock);
@@ -901,12 +900,6 @@ static irqreturn_t sh_eth_interrupt(int irq, void *netdev)
 	if (intr_status & cd->eesr_err_check)
 		sh_eth_error(ndev, intr_status);
 
-	if (--boguscnt < 0) {
-		printk(KERN_WARNING
-		       "%s: Too much work at interrupt, status=0x%4.4x.\n",
-		       ndev->name, intr_status);
-	}
-
 other_irq:
 	spin_unlock(&mdp->lock);
 
-- 
cgit v1.2.3-59-g8ed1b


From 30767636e5896c650f33db5f7f0a9b0e82f3e8c4 Mon Sep 17 00:00:00 2001
From: Nicolas Reinecke <nr@das-labor.org>
Date: Thu, 25 Jun 2009 02:55:31 +0000
Subject: mdio add missing GPL flag

Add missing GPL flag and description.

mdio: module license 'unspecified' taints kernel.
Disabling lock debugging due to kernel taint

Signed-off-by: Nicolas Reinecke <nr <at> das-labor.org>
Acked-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/mdio.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/net/mdio.c b/drivers/net/mdio.c
index dc45e9856c35..6851bdb2ce29 100644
--- a/drivers/net/mdio.c
+++ b/drivers/net/mdio.c
@@ -14,6 +14,10 @@
 #include <linux/mdio.h>
 #include <linux/module.h>
 
+MODULE_DESCRIPTION("Generic support for MDIO-compatible transceivers");
+MODULE_AUTHOR("Copyright 2006-2009 Solarflare Communications Inc.");
+MODULE_LICENSE("GPL");
+
 /**
  * mdio45_probe - probe for an MDIO (clause 45) device
  * @mdio: MDIO interface
-- 
cgit v1.2.3-59-g8ed1b


From ad8034f19792736db5c259103c2eaaf72887bbb4 Mon Sep 17 00:00:00 2001
From: Steve French <sfrench@us.ibm.com>
Date: Fri, 26 Jun 2009 03:25:49 +0000
Subject: [CIFS] remove bkl usage from umount begin

The lock_kernel call moved into the fs for umount_begin
is not needed.  This adds a check to make sure we don't
call umount_begin twice on the same fs.

umount_begin for cifs is probably not needed and
may eventually be able to be removed, but in
the meantime this smaller patch is safe and
gets rid of the bkl from this path which provides
some benefit.

Acked-by: Jeff Layton <redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/cifsfs.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index b5e9f398c2e5..9f669f982c4d 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -537,9 +537,14 @@ static void cifs_umount_begin(struct super_block *sb)
 	if (tcon == NULL)
 		return;
 
-	lock_kernel();
 	read_lock(&cifs_tcp_ses_lock);
-	if (tcon->tc_count == 1)
+	if ((tcon->tc_count > 1) || (tcon->tidStatus == CifsExiting)) {
+		/* we have other mounts to same share or we have
+		   already tried to force umount this and woken up
+		   all waiting network requests, nothing to do */
+		read_unlock(&cifs_tcp_ses_lock);
+		return;
+	} else if (tcon->tc_count == 1)
 		tcon->tidStatus = CifsExiting;
 	read_unlock(&cifs_tcp_ses_lock);
 
@@ -554,9 +559,7 @@ static void cifs_umount_begin(struct super_block *sb)
 		wake_up_all(&tcon->ses->server->response_q);
 		msleep(1);
 	}
-/* BB FIXME - finish add checks for tidStatus BB */
 
-	unlock_kernel();
 	return;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 71a394faaad07090af5de5c075ec2f5bca0fbb35 Mon Sep 17 00:00:00 2001
From: Steve French <sfrench@us.ibm.com>
Date: Fri, 26 Jun 2009 04:07:18 +0000
Subject: [CIFS] remove unknown mount option warning message

Jeff's previous patch which removed the unneeded rw/ro
parsing can cause a minor warning in dmesg (about the
unknown rw or ro mount option) at mount time. This
patch makes cifs ignore them in kernel to remove the warning
(they are already handled in the mount helper and VFS).

Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/connect.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 12c2cf693555..e16d7592116a 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -1196,6 +1196,10 @@ cifs_parse_mount_options(char *options, const char *devname,
 			/* ignore */
 		} else if (strnicmp(data, "guest", 5) == 0) {
 			/* ignore */
+		} else if (strnicmp(data, "rw", 2) == 0) {
+			/* ignore */
+		} else if (strnicmp(data, "ro", 2) == 0) {
+			/* ignore */
 		} else if (strnicmp(data, "noblocksend", 11) == 0) {
 			vol->noblocksnd = 1;
 		} else if (strnicmp(data, "noautotune", 10) == 0) {
-- 
cgit v1.2.3-59-g8ed1b


From 2b121bc262fa03c94e653b2d44356c2f86c1bcdc Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Thu, 25 Jun 2009 13:25:36 +0200
Subject: eeepc-laptop: Register as a pci-hotplug device

The eee contains a logically (but not physically) hotpluggable PCIe slot.
Currently this is handled by adding or removing the PCI device in response
to rfkill events, but if a user has forced pciehp to bind to it (with the
force=1 argument) then both drivers will try to handle the event and
hilarity (in the form of oopses) will ensue. This can be avoided by having
eee-laptop register the slot as a hotplug slot. Only one of pciehp and
eee-laptop will successfully register this, avoiding the problem.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Tested-by: Darren Salt <linux@youmustbejoking.demon.co.uk>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/Kconfig        |  2 +
 drivers/platform/x86/eeepc-laptop.c | 87 +++++++++++++++++++++++++++++++++++++
 2 files changed, 89 insertions(+)

diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig
index 7232fe7104aa..fee6a4022bc1 100644
--- a/drivers/platform/x86/Kconfig
+++ b/drivers/platform/x86/Kconfig
@@ -357,6 +357,8 @@ config EEEPC_LAPTOP
 	depends on RFKILL || RFKILL = n
 	select BACKLIGHT_CLASS_DEVICE
 	select HWMON
+	select HOTPLUG
+	select HOTPLUG_PCI if PCI
 	---help---
 	  This driver supports the Fn-Fx keys on Eee PC laptops.
 
diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index 4207b26ff990..c0b203ca29fb 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -31,6 +31,7 @@
 #include <linux/input.h>
 #include <linux/rfkill.h>
 #include <linux/pci.h>
+#include <linux/pci_hotplug.h>
 
 #define EEEPC_LAPTOP_VERSION	"0.1"
 
@@ -143,6 +144,7 @@ struct eeepc_hotk {
 	u16 *keycode_map;
 	struct rfkill *eeepc_wlan_rfkill;
 	struct rfkill *eeepc_bluetooth_rfkill;
+	struct hotplug_slot *hotplug_slot;
 };
 
 /* The actual device the driver binds to */
@@ -213,6 +215,15 @@ static struct acpi_driver eeepc_hotk_driver = {
 	},
 };
 
+/* PCI hotplug ops */
+static int eeepc_get_adapter_status(struct hotplug_slot *slot, u8 *value);
+
+static struct hotplug_slot_ops eeepc_hotplug_slot_ops = {
+	.owner = THIS_MODULE,
+	.get_adapter_status = eeepc_get_adapter_status,
+	.get_power_status = eeepc_get_adapter_status,
+};
+
 /* The backlight device /sys/class/backlight */
 static struct backlight_device *eeepc_backlight_device;
 
@@ -612,6 +623,19 @@ static int notify_brn(void)
 	return -1;
 }
 
+static int eeepc_get_adapter_status(struct hotplug_slot *hotplug_slot,
+				    u8 *value)
+{
+	int val = get_acpi(CM_ASL_WLAN);
+
+	if (val == 1 || val == 0)
+		*value = val;
+	else
+		return -EINVAL;
+
+	return 0;
+}
+
 static void eeepc_rfkill_hotplug(void)
 {
 	struct pci_dev *dev;
@@ -744,6 +768,54 @@ static void eeepc_unregister_rfkill_notifier(char *node)
 	}
 }
 
+static void eeepc_cleanup_pci_hotplug(struct hotplug_slot *hotplug_slot)
+{
+	kfree(hotplug_slot->info);
+	kfree(hotplug_slot);
+}
+
+static int eeepc_setup_pci_hotplug(void)
+{
+	int ret = -ENOMEM;
+	struct pci_bus *bus = pci_find_bus(0, 1);
+
+	if (!bus) {
+		printk(EEEPC_ERR "Unable to find wifi PCI bus\n");
+		return -ENODEV;
+	}
+
+	ehotk->hotplug_slot = kzalloc(sizeof(struct hotplug_slot), GFP_KERNEL);
+	if (!ehotk->hotplug_slot)
+		goto error_slot;
+
+	ehotk->hotplug_slot->info = kzalloc(sizeof(struct hotplug_slot_info),
+					    GFP_KERNEL);
+	if (!ehotk->hotplug_slot->info)
+		goto error_info;
+
+	ehotk->hotplug_slot->private = ehotk;
+	ehotk->hotplug_slot->release = &eeepc_cleanup_pci_hotplug;
+	ehotk->hotplug_slot->ops = &eeepc_hotplug_slot_ops;
+	eeepc_get_adapter_status(ehotk->hotplug_slot,
+				 &ehotk->hotplug_slot->info->adapter_status);
+
+	ret = pci_hp_register(ehotk->hotplug_slot, bus, 0, "eeepc-wifi");
+	if (ret) {
+		printk(EEEPC_ERR "Unable to register hotplug slot - %d\n", ret);
+		goto error_register;
+	}
+
+	return 0;
+
+error_register:
+	kfree(ehotk->hotplug_slot->info);
+error_info:
+	kfree(ehotk->hotplug_slot);
+	ehotk->hotplug_slot = NULL;
+error_slot:
+	return ret;
+}
+
 static int eeepc_hotk_add(struct acpi_device *device)
 {
 	int result;
@@ -802,8 +874,21 @@ static int eeepc_hotk_add(struct acpi_device *device)
 			goto bluetooth_fail;
 	}
 
+	result = eeepc_setup_pci_hotplug();
+	/*
+	 * If we get -EBUSY then something else is handling the PCI hotplug -
+	 * don't fail in this case
+	 */
+	if (result == -EBUSY)
+		return 0;
+	else if (result)
+		goto pci_fail;
+
 	return 0;
 
+ pci_fail:
+	if (ehotk->eeepc_bluetooth_rfkill)
+		rfkill_unregister(ehotk->eeepc_bluetooth_rfkill);
  bluetooth_fail:
 	rfkill_destroy(ehotk->eeepc_bluetooth_rfkill);
 	rfkill_unregister(ehotk->eeepc_wlan_rfkill);
@@ -825,6 +910,8 @@ static int eeepc_hotk_remove(struct acpi_device *device, int type)
 
 	eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P6");
 	eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P7");
+	if (ehotk->hotplug_slot)
+		pci_hp_deregister(ehotk->hotplug_slot);
 
 	kfree(ehotk);
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From 19b532892834b7f1c04b2940ac73177dc566fed5 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Thu, 25 Jun 2009 13:25:37 +0200
Subject: eeepc-laptop.c: use pr_fmt and pr_<level>

Convert the unusual printk(EEEPC_<level> uses to
the more standard pr_fmt and pr_<level>(.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/eeepc-laptop.c | 55 +++++++++++++++----------------------
 1 file changed, 22 insertions(+), 33 deletions(-)

diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index c0b203ca29fb..d14f7149cb13 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -16,6 +16,8 @@
  *  GNU General Public License for more details.
  */
 
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -41,11 +43,6 @@
 #define EEEPC_HOTK_DEVICE_NAME	"Hotkey"
 #define EEEPC_HOTK_HID		"ASUS010"
 
-#define EEEPC_LOG	EEEPC_HOTK_FILE ": "
-#define EEEPC_ERR	KERN_ERR	EEEPC_LOG
-#define EEEPC_WARNING	KERN_WARNING	EEEPC_LOG
-#define EEEPC_NOTICE	KERN_NOTICE	EEEPC_LOG
-#define EEEPC_INFO	KERN_INFO	EEEPC_LOG
 
 /*
  * Definitions for Asus EeePC
@@ -285,7 +282,7 @@ static int set_acpi(int cm, int value)
 		if (method == NULL)
 			return -ENODEV;
 		if (write_acpi_int(ehotk->handle, method, value, NULL))
-			printk(EEEPC_WARNING "Error writing %s\n", method);
+			pr_warning("Error writing %s\n", method);
 	}
 	return 0;
 }
@@ -298,7 +295,7 @@ static int get_acpi(int cm)
 		if (method == NULL)
 			return -ENODEV;
 		if (read_acpi_int(ehotk->handle, method, &value))
-			printk(EEEPC_WARNING "Error reading %s\n", method);
+			pr_warning("Error reading %s\n", method);
 	}
 	return value;
 }
@@ -562,26 +559,23 @@ static int eeepc_hotk_check(void)
 	if (ehotk->device->status.present) {
 		if (write_acpi_int(ehotk->handle, "INIT", ehotk->init_flag,
 				    &buffer)) {
-			printk(EEEPC_ERR "Hotkey initialization failed\n");
+			pr_err("Hotkey initialization failed\n");
 			return -ENODEV;
 		} else {
-			printk(EEEPC_NOTICE "Hotkey init flags 0x%x\n",
-			       ehotk->init_flag);
+			pr_notice("Hotkey init flags 0x%x\n", ehotk->init_flag);
 		}
 		/* get control methods supported */
 		if (read_acpi_int(ehotk->handle, "CMSG"
 				   , &ehotk->cm_supported)) {
-			printk(EEEPC_ERR
-			       "Get control methods supported failed\n");
+			pr_err("Get control methods supported failed\n");
 			return -ENODEV;
 		} else {
-			printk(EEEPC_INFO
-			       "Get control methods supported: 0x%x\n",
-			       ehotk->cm_supported);
+			pr_info("Get control methods supported: 0x%x\n",
+				ehotk->cm_supported);
 		}
 		ehotk->inputdev = input_allocate_device();
 		if (!ehotk->inputdev) {
-			printk(EEEPC_INFO "Unable to allocate input device\n");
+			pr_info("Unable to allocate input device\n");
 			return 0;
 		}
 		ehotk->inputdev->name = "Asus EeePC extra buttons";
@@ -600,12 +594,12 @@ static int eeepc_hotk_check(void)
 		}
 		result = input_register_device(ehotk->inputdev);
 		if (result) {
-			printk(EEEPC_INFO "Unable to register input device\n");
+			pr_info("Unable to register input device\n");
 			input_free_device(ehotk->inputdev);
 			return 0;
 		}
 	} else {
-		printk(EEEPC_ERR "Hotkey device not present, aborting\n");
+		pr_err("Hotkey device not present, aborting\n");
 		return -EINVAL;
 	}
 	return 0;
@@ -643,7 +637,7 @@ static void eeepc_rfkill_hotplug(void)
 	bool blocked;
 
 	if (!bus) {
-		printk(EEEPC_WARNING "Unable to find PCI bus 1?\n");
+		pr_warning("Unable to find PCI bus 1?\n");
 		return;
 	}
 
@@ -659,7 +653,7 @@ static void eeepc_rfkill_hotplug(void)
 		if (dev) {
 			pci_bus_assign_resources(bus);
 			if (pci_bus_add_device(dev))
-				printk(EEEPC_ERR "Unable to hotplug wifi\n");
+				pr_err("Unable to hotplug wifi\n");
 		}
 	} else {
 		dev = pci_get_slot(bus, 0);
@@ -742,8 +736,7 @@ static int eeepc_register_rfkill_notifier(char *node)
 						     eeepc_rfkill_notify,
 						     NULL);
 		if (ACPI_FAILURE(status))
-			printk(EEEPC_WARNING
-			       "Failed to register notify on %s\n", node);
+			pr_warning("Failed to register notify on %s\n", node);
 	} else
 		return -ENODEV;
 
@@ -762,8 +755,7 @@ static void eeepc_unregister_rfkill_notifier(char *node)
 						     ACPI_SYSTEM_NOTIFY,
 						     eeepc_rfkill_notify);
 		if (ACPI_FAILURE(status))
-			printk(EEEPC_ERR
-			       "Error removing rfkill notify handler %s\n",
+			pr_err("Error removing rfkill notify handler %s\n",
 				node);
 	}
 }
@@ -780,7 +772,7 @@ static int eeepc_setup_pci_hotplug(void)
 	struct pci_bus *bus = pci_find_bus(0, 1);
 
 	if (!bus) {
-		printk(EEEPC_ERR "Unable to find wifi PCI bus\n");
+		pr_err("Unable to find wifi PCI bus\n");
 		return -ENODEV;
 	}
 
@@ -801,7 +793,7 @@ static int eeepc_setup_pci_hotplug(void)
 
 	ret = pci_hp_register(ehotk->hotplug_slot, bus, 0, "eeepc-wifi");
 	if (ret) {
-		printk(EEEPC_ERR "Unable to register hotplug slot - %d\n", ret);
+		pr_err("Unable to register hotplug slot - %d\n", ret);
 		goto error_register;
 	}
 
@@ -822,7 +814,7 @@ static int eeepc_hotk_add(struct acpi_device *device)
 
 	if (!device)
 		 return -EINVAL;
-	printk(EEEPC_NOTICE EEEPC_HOTK_NAME "\n");
+	pr_notice(EEEPC_HOTK_NAME "\n");
 	ehotk = kzalloc(sizeof(struct eeepc_hotk), GFP_KERNEL);
 	if (!ehotk)
 		return -ENOMEM;
@@ -1105,8 +1097,7 @@ static int eeepc_backlight_init(struct device *dev)
 	bd = backlight_device_register(EEEPC_HOTK_FILE, dev,
 				       NULL, &eeepcbl_ops);
 	if (IS_ERR(bd)) {
-		printk(EEEPC_ERR
-		       "Could not register eeepc backlight device\n");
+		pr_err("Could not register eeepc backlight device\n");
 		eeepc_backlight_device = NULL;
 		return PTR_ERR(bd);
 	}
@@ -1125,8 +1116,7 @@ static int eeepc_hwmon_init(struct device *dev)
 
 	hwmon = hwmon_device_register(dev);
 	if (IS_ERR(hwmon)) {
-		printk(EEEPC_ERR
-		       "Could not register eeepc hwmon device\n");
+		pr_err("Could not register eeepc hwmon device\n");
 		eeepc_hwmon_device = NULL;
 		return PTR_ERR(hwmon);
 	}
@@ -1159,8 +1149,7 @@ static int __init eeepc_laptop_init(void)
 		if (result)
 			goto fail_backlight;
 	} else
-		printk(EEEPC_INFO "Backlight controlled by ACPI video "
-		       "driver\n");
+		pr_info("Backlight controlled by ACPI video driver\n");
 
 	result = eeepc_hwmon_init(dev);
 	if (result)
-- 
cgit v1.2.3-59-g8ed1b


From 7de39389d8f61aa517ce2a8b4d925acc62696ae5 Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Thu, 25 Jun 2009 13:25:38 +0200
Subject: eeepc-laptop: rfkill refactoring

Refactor rfkill code, because we'll add another
rfkill for wwan3g later.

Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/eeepc-laptop.c | 160 +++++++++++++++++++-----------------
 1 file changed, 84 insertions(+), 76 deletions(-)

diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index d14f7149cb13..e46981a5f20b 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -139,8 +139,8 @@ struct eeepc_hotk {
 	u16 event_count[128];		/* count for each event */
 	struct input_dev *inputdev;
 	u16 *keycode_map;
-	struct rfkill *eeepc_wlan_rfkill;
-	struct rfkill *eeepc_bluetooth_rfkill;
+	struct rfkill *wlan_rfkill;
+	struct rfkill *bluetooth_rfkill;
 	struct hotplug_slot *hotplug_slot;
 };
 
@@ -663,7 +663,7 @@ static void eeepc_rfkill_hotplug(void)
 		}
 	}
 
-	rfkill_set_sw_state(ehotk->eeepc_wlan_rfkill, blocked);
+	rfkill_set_sw_state(ehotk->wlan_rfkill, blocked);
 }
 
 static void eeepc_rfkill_notify(acpi_handle handle, u32 event, void *data)
@@ -828,66 +828,8 @@ static int eeepc_hotk_add(struct acpi_device *device)
 	if (result)
 		goto ehotk_fail;
 
-	eeepc_register_rfkill_notifier("\\_SB.PCI0.P0P6");
-	eeepc_register_rfkill_notifier("\\_SB.PCI0.P0P7");
-
-	if (get_acpi(CM_ASL_WLAN) != -1) {
-		ehotk->eeepc_wlan_rfkill = rfkill_alloc("eeepc-wlan",
-							&device->dev,
-							RFKILL_TYPE_WLAN,
-							&eeepc_rfkill_ops,
-							(void *)CM_ASL_WLAN);
-
-		if (!ehotk->eeepc_wlan_rfkill)
-			goto wlan_fail;
-
-		rfkill_init_sw_state(ehotk->eeepc_wlan_rfkill,
-				     get_acpi(CM_ASL_WLAN) != 1);
-		result = rfkill_register(ehotk->eeepc_wlan_rfkill);
-		if (result)
-			goto wlan_fail;
-	}
-
-	if (get_acpi(CM_ASL_BLUETOOTH) != -1) {
-		ehotk->eeepc_bluetooth_rfkill =
-			rfkill_alloc("eeepc-bluetooth",
-				     &device->dev,
-				     RFKILL_TYPE_BLUETOOTH,
-				     &eeepc_rfkill_ops,
-				     (void *)CM_ASL_BLUETOOTH);
-
-		if (!ehotk->eeepc_bluetooth_rfkill)
-			goto bluetooth_fail;
-
-		rfkill_init_sw_state(ehotk->eeepc_bluetooth_rfkill,
-				     get_acpi(CM_ASL_BLUETOOTH) != 1);
-		result = rfkill_register(ehotk->eeepc_bluetooth_rfkill);
-		if (result)
-			goto bluetooth_fail;
-	}
-
-	result = eeepc_setup_pci_hotplug();
-	/*
-	 * If we get -EBUSY then something else is handling the PCI hotplug -
-	 * don't fail in this case
-	 */
-	if (result == -EBUSY)
-		return 0;
-	else if (result)
-		goto pci_fail;
-
 	return 0;
 
- pci_fail:
-	if (ehotk->eeepc_bluetooth_rfkill)
-		rfkill_unregister(ehotk->eeepc_bluetooth_rfkill);
- bluetooth_fail:
-	rfkill_destroy(ehotk->eeepc_bluetooth_rfkill);
-	rfkill_unregister(ehotk->eeepc_wlan_rfkill);
- wlan_fail:
-	rfkill_destroy(ehotk->eeepc_wlan_rfkill);
-	eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P6");
-	eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P7");
  ehotk_fail:
 	kfree(ehotk);
 	ehotk = NULL;
@@ -900,18 +842,13 @@ static int eeepc_hotk_remove(struct acpi_device *device, int type)
 	if (!device || !acpi_driver_data(device))
 		 return -EINVAL;
 
-	eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P6");
-	eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P7");
-	if (ehotk->hotplug_slot)
-		pci_hp_deregister(ehotk->hotplug_slot);
-
 	kfree(ehotk);
 	return 0;
 }
 
 static int eeepc_hotk_resume(struct acpi_device *device)
 {
-	if (ehotk->eeepc_wlan_rfkill) {
+	if (ehotk->wlan_rfkill) {
 		bool wlan;
 
 		/* Workaround - it seems that _PTS disables the wireless
@@ -923,14 +860,13 @@ static int eeepc_hotk_resume(struct acpi_device *device)
 		wlan = get_acpi(CM_ASL_WLAN);
 		set_acpi(CM_ASL_WLAN, wlan);
 
-		rfkill_set_sw_state(ehotk->eeepc_wlan_rfkill,
-				    wlan != 1);
+		rfkill_set_sw_state(ehotk->wlan_rfkill, wlan != 1);
 
 		eeepc_rfkill_hotplug();
 	}
 
-	if (ehotk->eeepc_bluetooth_rfkill)
-		rfkill_set_sw_state(ehotk->eeepc_bluetooth_rfkill,
+	if (ehotk->bluetooth_rfkill)
+		rfkill_set_sw_state(ehotk->bluetooth_rfkill,
 				    get_acpi(CM_ASL_BLUETOOTH) != 1);
 
 	return 0;
@@ -1052,10 +988,14 @@ static void eeepc_backlight_exit(void)
 
 static void eeepc_rfkill_exit(void)
 {
-	if (ehotk->eeepc_wlan_rfkill)
-		rfkill_unregister(ehotk->eeepc_wlan_rfkill);
-	if (ehotk->eeepc_bluetooth_rfkill)
-		rfkill_unregister(ehotk->eeepc_bluetooth_rfkill);
+	eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P6");
+	eeepc_unregister_rfkill_notifier("\\_SB.PCI0.P0P7");
+	if (ehotk->wlan_rfkill)
+		rfkill_unregister(ehotk->wlan_rfkill);
+	if (ehotk->bluetooth_rfkill)
+		rfkill_unregister(ehotk->bluetooth_rfkill);
+	if (ehotk->hotplug_slot)
+		pci_hp_deregister(ehotk->hotplug_slot);
 }
 
 static void eeepc_input_exit(void)
@@ -1090,6 +1030,67 @@ static void __exit eeepc_laptop_exit(void)
 	platform_driver_unregister(&platform_driver);
 }
 
+static int eeepc_new_rfkill(struct rfkill **rfkill,
+			    const char *name, struct device *dev,
+			    enum rfkill_type type, int cm)
+{
+	int result;
+
+	if (get_acpi(cm) == -1)
+		return -ENODEV;
+
+	*rfkill = rfkill_alloc(name, dev, type,
+			       &eeepc_rfkill_ops, (void *)(unsigned long)cm);
+
+	if (!*rfkill)
+		return -EINVAL;
+
+	rfkill_init_sw_state(*rfkill, get_acpi(cm) != 1);
+	result = rfkill_register(*rfkill);
+	if (result) {
+		rfkill_destroy(*rfkill);
+		*rfkill = NULL;
+		return result;
+	}
+	return 0;
+}
+
+
+static int eeepc_rfkill_init(struct device *dev)
+{
+	int result = 0;
+
+	eeepc_register_rfkill_notifier("\\_SB.PCI0.P0P6");
+	eeepc_register_rfkill_notifier("\\_SB.PCI0.P0P7");
+
+	result = eeepc_new_rfkill(&ehotk->wlan_rfkill,
+				  "eeepc-wlan", dev,
+				  RFKILL_TYPE_WLAN, CM_ASL_WLAN);
+
+	if (result && result != -ENODEV)
+		goto exit;
+
+	result = eeepc_new_rfkill(&ehotk->bluetooth_rfkill,
+				  "eeepc-bluetooth", dev,
+				  RFKILL_TYPE_BLUETOOTH, CM_ASL_BLUETOOTH);
+
+	if (result && result != -ENODEV)
+		goto exit;
+
+	result = eeepc_setup_pci_hotplug();
+	/*
+	 * If we get -EBUSY then something else is handling the PCI hotplug -
+	 * don't fail in this case
+	 */
+	if (result == -EBUSY)
+		result = 0;
+
+exit:
+	if (result && result != -ENODEV)
+		eeepc_rfkill_exit();
+	return result;
+}
+
 static int eeepc_backlight_init(struct device *dev)
 {
 	struct backlight_device *bd;
@@ -1173,7 +1174,15 @@ static int __init eeepc_laptop_init(void)
 				    &platform_attribute_group);
 	if (result)
 		goto fail_sysfs;
+
+	result = eeepc_rfkill_init(dev);
+	if (result)
+		goto fail_rfkill;
+
 	return 0;
+fail_rfkill:
+	sysfs_remove_group(&platform_device->dev.kobj,
+			   &platform_attribute_group);
 fail_sysfs:
 	platform_device_del(platform_device);
 fail_platform_device2:
@@ -1186,7 +1195,6 @@ fail_hwmon:
 	eeepc_backlight_exit();
 fail_backlight:
 	eeepc_input_exit();
-	eeepc_rfkill_exit();
 	return result;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 1ddec2f9435e77b4d3f50eced68c4c942f2bcd4b Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Thu, 25 Jun 2009 13:25:39 +0200
Subject: eeepc-laptop: right parent device

Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/eeepc-laptop.c | 34 ++++++++++++++++++----------------
 1 file changed, 18 insertions(+), 16 deletions(-)

diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index e46981a5f20b..5b102c2f66a0 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -1143,18 +1143,6 @@ static int __init eeepc_laptop_init(void)
 		acpi_bus_unregister_driver(&eeepc_hotk_driver);
 		return -ENODEV;
 	}
-	dev = acpi_get_physical_device(ehotk->device->handle);
-
-	if (!acpi_video_backlight_support()) {
-		result = eeepc_backlight_init(dev);
-		if (result)
-			goto fail_backlight;
-	} else
-		pr_info("Backlight controlled by ACPI video driver\n");
-
-	result = eeepc_hwmon_init(dev);
-	if (result)
-		goto fail_hwmon;
 
 	eeepc_enable_camera();
 
@@ -1175,12 +1163,30 @@ static int __init eeepc_laptop_init(void)
 	if (result)
 		goto fail_sysfs;
 
+	dev = &platform_device->dev;
+
+	if (!acpi_video_backlight_support()) {
+		result = eeepc_backlight_init(dev);
+		if (result)
+			goto fail_backlight;
+	} else
+		pr_info("Backlight controlled by ACPI video "
+			"driver\n");
+
+	result = eeepc_hwmon_init(dev);
+	if (result)
+		goto fail_hwmon;
+
 	result = eeepc_rfkill_init(dev);
 	if (result)
 		goto fail_rfkill;
 
 	return 0;
 fail_rfkill:
+	eeepc_hwmon_exit();
+fail_hwmon:
+	eeepc_backlight_exit();
+fail_backlight:
 	sysfs_remove_group(&platform_device->dev.kobj,
 			   &platform_attribute_group);
 fail_sysfs:
@@ -1190,10 +1196,6 @@ fail_platform_device2:
 fail_platform_device1:
 	platform_driver_unregister(&platform_driver);
 fail_platform_driver:
-	eeepc_hwmon_exit();
-fail_hwmon:
-	eeepc_backlight_exit();
-fail_backlight:
 	eeepc_input_exit();
 	return result;
 }
-- 
cgit v1.2.3-59-g8ed1b


From f36509e7248631671d02f48d1a88f56cdeb54ed8 Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Thu, 25 Jun 2009 13:25:40 +0200
Subject: eeepc-laptop: makes get_acpi() returns -ENODEV

If there is there is no getter defined, get_acpi()
will return -ENODEV.

Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/eeepc-laptop.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index 5b102c2f66a0..19cc9ae7db5a 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -289,7 +289,7 @@ static int set_acpi(int cm, int value)
 
 static int get_acpi(int cm)
 {
-	int value = -1;
+	int value = -ENODEV;
 	if ((ehotk->cm_supported & (0x1 << cm))) {
 		const char *method = cm_getv[cm];
 		if (method == NULL)
@@ -367,13 +367,19 @@ static ssize_t store_sys_acpi(int cm, const char *buf, size_t count)
 
 	rv = parse_arg(buf, count, &value);
 	if (rv > 0)
-		set_acpi(cm, value);
+		value = set_acpi(cm, value);
+	if (value < 0)
+		return value;
 	return rv;
 }
 
 static ssize_t show_sys_acpi(int cm, char *buf)
 {
-	return sprintf(buf, "%d\n", get_acpi(cm));
+	int value = get_acpi(cm);
+
+	if (value < 0)
+		return value;
+	return sprintf(buf, "%d\n", value);
 }
 
 #define EEEPC_CREATE_DEVICE_ATTR(_name, _cm)				\
@@ -1036,8 +1042,9 @@ static int eeepc_new_rfkill(struct rfkill **rfkill,
 {
 	int result;
 
-	if (get_acpi(cm) == -1)
-		return -ENODEV;
+	result = get_acpi(cm);
+	if (result < 0)
+		return result;
 
 	*rfkill = rfkill_alloc(name, dev, type,
 			       &eeepc_rfkill_ops, (void *)(unsigned long)cm);
-- 
cgit v1.2.3-59-g8ed1b


From dbfa3ba90dfe353a56e107cff5bce9fb7976f06f Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Thu, 25 Jun 2009 13:25:41 +0200
Subject: eeepc-laptop: get the right value for CMSG

CMSG is an ACPI method used to find features available on
an Eee PC. But some features are never repported, even if present.

If the getter of a feature is present, this patch will set
the corresponding bit in cmsg.

Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/eeepc-laptop.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index 19cc9ae7db5a..f5d8473ea66f 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -553,6 +553,28 @@ static int eeepc_setkeycode(struct input_dev *dev, int scancode, int keycode)
 	return -EINVAL;
 }
 
+static void cmsg_quirk(int cm, const char *name)
+{
+	int dummy;
+
+	/* Some BIOSes do not report cm although it is avaliable.
+	   Check if cm_getv[cm] works and, if yes, assume cm should be set. */
+	if (!(ehotk->cm_supported & (1 << cm))
+	    && !read_acpi_int(ehotk->handle, cm_getv[cm], &dummy)) {
+		pr_info("%s (%x) not reported by BIOS,"
+			" enabling anyway\n", name, 1 << cm);
+		ehotk->cm_supported |= 1 << cm;
+	}
+}
+
+static void cmsg_quirks(void)
+{
+	cmsg_quirk(CM_ASL_LID, "LID");
+	cmsg_quirk(CM_ASL_TYPE, "TYPE");
+	cmsg_quirk(CM_ASL_PANELPOWER, "PANELPOWER");
+	cmsg_quirk(CM_ASL_TPD, "TPD");
+}
+
 static int eeepc_hotk_check(void)
 {
 	const struct key_entry *key;
@@ -576,6 +598,7 @@ static int eeepc_hotk_check(void)
 			pr_err("Get control methods supported failed\n");
 			return -ENODEV;
 		} else {
+			cmsg_quirks();
 			pr_info("Get control methods supported: 0x%x\n",
 				ehotk->cm_supported);
 		}
-- 
cgit v1.2.3-59-g8ed1b


From 3cd530b5aaffd27b231f9717730f2f6684c00bda Mon Sep 17 00:00:00 2001
From: Corentin Chary <corentincj@iksaif.net>
Date: Thu, 25 Jun 2009 13:25:42 +0200
Subject: eeepc-laptop: add rfkill support for the 3G modem in Eee PC 901 Go

Signed-off-by: Janne Grunau <j@jannau.net>
Signed-off-by: Corentin Chary <corentincj@iksaif.net>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/platform/x86/eeepc-laptop.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/platform/x86/eeepc-laptop.c b/drivers/platform/x86/eeepc-laptop.c
index f5d8473ea66f..ec560f16d720 100644
--- a/drivers/platform/x86/eeepc-laptop.c
+++ b/drivers/platform/x86/eeepc-laptop.c
@@ -141,6 +141,7 @@ struct eeepc_hotk {
 	u16 *keycode_map;
 	struct rfkill *wlan_rfkill;
 	struct rfkill *bluetooth_rfkill;
+	struct rfkill *wwan3g_rfkill;
 	struct hotplug_slot *hotplug_slot;
 };
 
@@ -1023,6 +1024,8 @@ static void eeepc_rfkill_exit(void)
 		rfkill_unregister(ehotk->wlan_rfkill);
 	if (ehotk->bluetooth_rfkill)
 		rfkill_unregister(ehotk->bluetooth_rfkill);
+	if (ehotk->wwan3g_rfkill)
+		rfkill_unregister(ehotk->wwan3g_rfkill);
 	if (ehotk->hotplug_slot)
 		pci_hp_deregister(ehotk->hotplug_slot);
 }
@@ -1107,6 +1110,13 @@ static int eeepc_rfkill_init(struct device *dev)
 	if (result && result != -ENODEV)
 		goto exit;
 
+	result = eeepc_new_rfkill(&ehotk->wwan3g_rfkill,
+				  "eeepc-wwan3g", dev,
+				  RFKILL_TYPE_WWAN, CM_ASL_3G);
+
+	if (result && result != -ENODEV)
+		goto exit;
+
 	result = eeepc_setup_pci_hotplug();
 	/*
 	 * If we get -EBUSY then something else is handling the PCI hotplug -
-- 
cgit v1.2.3-59-g8ed1b


From 412af97838828bc6d035a1902c8974f944663da6 Mon Sep 17 00:00:00 2001
From: Troy Moure <twmoure@szypr.net>
Date: Thu, 25 Jun 2009 17:05:35 -0600
Subject: ACPI: video: prevent NULL deref in acpi_get_pci_dev()

ref: http://thread.gmane.org/gmane.linux.kernel/857228/focus=857468

When the ACPI video driver initializes, it does a namespace walk
looking for for supported devices. When we find an appropriate
handle, we walk up the ACPI tree looking for a PCI root bus, and
then walk back down the PCI bus, assuming that every device
inbetween is a P2P bridge.

This assumption is not correct, and is reported broken on at
least:

	Dell Latitude E6400
	ThinkPad X61
	Dell XPS M1330

Add a NULL deref check to prevent boot panics.

Reported-by: Alessandro Suardi <alessandro.suardi@gmail.com>
Signed-off-by: Troy Moure <twmoure@szypr.net>
Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
 drivers/acpi/pci_root.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c
index 8a5bf3b356fa..55b5b90c2a44 100644
--- a/drivers/acpi/pci_root.c
+++ b/drivers/acpi/pci_root.c
@@ -395,7 +395,7 @@ struct pci_dev *acpi_get_pci_dev(acpi_handle handle)
 		fn  = adr & 0xffff;
 
 		pdev = pci_get_slot(pbus, PCI_DEVFN(dev, fn));
-		if (hnd == handle)
+		if (!pdev || hnd == handle)
 			break;
 
 		pbus = pdev->subordinate;
-- 
cgit v1.2.3-59-g8ed1b


From 6f0b1c6094b3e8eeeb13f8f16c1b2ef452a6f519 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Mon, 22 Jun 2009 23:13:48 +0000
Subject: powerpc: Swiotlb breaks pseries

Turning on SWIOTLB selects or enables PPC_NEED_DMA_SYNC_OPS, which means
we get the non empty versions of dma_sync_* in asm/dma-mapping.h

On my pseries machine the dma_ops have no such routines and we die with
a null pointer - this patch gets it booting, is there a more elegant way
to do it?

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/dma-mapping.h | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)

diff --git a/arch/powerpc/include/asm/dma-mapping.h b/arch/powerpc/include/asm/dma-mapping.h
index 3d9e887c3c0c..b44aaabdd1a6 100644
--- a/arch/powerpc/include/asm/dma-mapping.h
+++ b/arch/powerpc/include/asm/dma-mapping.h
@@ -309,7 +309,9 @@ static inline void dma_sync_single_for_cpu(struct device *dev,
 	struct dma_mapping_ops *dma_ops = get_dma_ops(dev);
 
 	BUG_ON(!dma_ops);
-	dma_ops->sync_single_range_for_cpu(dev, dma_handle, 0,
+
+	if (dma_ops->sync_single_range_for_cpu)
+		dma_ops->sync_single_range_for_cpu(dev, dma_handle, 0,
 					   size, direction);
 }
 
@@ -320,7 +322,9 @@ static inline void dma_sync_single_for_device(struct device *dev,
 	struct dma_mapping_ops *dma_ops = get_dma_ops(dev);
 
 	BUG_ON(!dma_ops);
-	dma_ops->sync_single_range_for_device(dev, dma_handle,
+
+	if (dma_ops->sync_single_range_for_device)
+		dma_ops->sync_single_range_for_device(dev, dma_handle,
 					      0, size, direction);
 }
 
@@ -331,7 +335,9 @@ static inline void dma_sync_sg_for_cpu(struct device *dev,
 	struct dma_mapping_ops *dma_ops = get_dma_ops(dev);
 
 	BUG_ON(!dma_ops);
-	dma_ops->sync_sg_for_cpu(dev, sgl, nents, direction);
+
+	if (dma_ops->sync_sg_for_cpu)
+		dma_ops->sync_sg_for_cpu(dev, sgl, nents, direction);
 }
 
 static inline void dma_sync_sg_for_device(struct device *dev,
@@ -341,7 +347,9 @@ static inline void dma_sync_sg_for_device(struct device *dev,
 	struct dma_mapping_ops *dma_ops = get_dma_ops(dev);
 
 	BUG_ON(!dma_ops);
-	dma_ops->sync_sg_for_device(dev, sgl, nents, direction);
+
+	if (dma_ops->sync_sg_for_device)
+		dma_ops->sync_sg_for_device(dev, sgl, nents, direction);
 }
 
 static inline void dma_sync_single_range_for_cpu(struct device *dev,
@@ -351,7 +359,9 @@ static inline void dma_sync_single_range_for_cpu(struct device *dev,
 	struct dma_mapping_ops *dma_ops = get_dma_ops(dev);
 
 	BUG_ON(!dma_ops);
-	dma_ops->sync_single_range_for_cpu(dev, dma_handle,
+
+	if (dma_ops->sync_single_range_for_cpu)
+		dma_ops->sync_single_range_for_cpu(dev, dma_handle,
 					   offset, size, direction);
 }
 
@@ -362,7 +372,9 @@ static inline void dma_sync_single_range_for_device(struct device *dev,
 	struct dma_mapping_ops *dma_ops = get_dma_ops(dev);
 
 	BUG_ON(!dma_ops);
-	dma_ops->sync_single_range_for_device(dev, dma_handle, offset,
+
+	if (dma_ops->sync_single_range_for_device)
+		dma_ops->sync_single_range_for_device(dev, dma_handle, offset,
 					      size, direction);
 }
 #else /* CONFIG_PPC_NEED_DMA_SYNC_OPS */
-- 
cgit v1.2.3-59-g8ed1b


From 5a2642f620eb6e40792822fa0eafe23046fbb55e Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Mon, 22 Jun 2009 16:47:59 +0000
Subject: powerpc/mpic: Fix mapping of "DCR" based MPIC variants

Commit 31207dab7d2e63795eb15823947bd2f7025b08e2
"Fix incorrect allocation of interrupt rev-map"
introduced a regression crashing on boot on machines using
a "DCR" based MPIC, such as the Cell blades.

The reason is that the irq host data structure is initialized
much later as a result of that patch, causing our calls to
mpic_map() do be done before we have a host setup.

Unfortunately, this breaks _mpic_map_dcr() which uses the
mpic->irqhost to get to the device node.

This fixes it by, instead, passing the device node explicitely
to mpic_map().

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Akira Tsukamoto <akirat@rd.scei.sony.co.jp>
---
 arch/powerpc/sysdev/mpic.c | 29 ++++++++++++++++-------------
 1 file changed, 16 insertions(+), 13 deletions(-)

diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c
index 9c3af5045495..32a2e950f563 100644
--- a/arch/powerpc/sysdev/mpic.c
+++ b/arch/powerpc/sysdev/mpic.c
@@ -279,28 +279,29 @@ static void _mpic_map_mmio(struct mpic *mpic, phys_addr_t phys_addr,
 }
 
 #ifdef CONFIG_PPC_DCR
-static void _mpic_map_dcr(struct mpic *mpic, struct mpic_reg_bank *rb,
+static void _mpic_map_dcr(struct mpic *mpic, struct device_node *node,
+			  struct mpic_reg_bank *rb,
 			  unsigned int offset, unsigned int size)
 {
 	const u32 *dbasep;
 
-	dbasep = of_get_property(mpic->irqhost->of_node, "dcr-reg", NULL);
+	dbasep = of_get_property(node, "dcr-reg", NULL);
 
-	rb->dhost = dcr_map(mpic->irqhost->of_node, *dbasep + offset, size);
+	rb->dhost = dcr_map(node, *dbasep + offset, size);
 	BUG_ON(!DCR_MAP_OK(rb->dhost));
 }
 
-static inline void mpic_map(struct mpic *mpic, phys_addr_t phys_addr,
-			    struct mpic_reg_bank *rb, unsigned int offset,
-			    unsigned int size)
+static inline void mpic_map(struct mpic *mpic, struct device_node *node,
+			    phys_addr_t phys_addr, struct mpic_reg_bank *rb,
+			    unsigned int offset, unsigned int size)
 {
 	if (mpic->flags & MPIC_USES_DCR)
-		_mpic_map_dcr(mpic, rb, offset, size);
+		_mpic_map_dcr(mpic, node, rb, offset, size);
 	else
 		_mpic_map_mmio(mpic, phys_addr, rb, offset, size);
 }
 #else /* CONFIG_PPC_DCR */
-#define mpic_map(m,p,b,o,s)	_mpic_map_mmio(m,p,b,o,s)
+#define mpic_map(m,n,p,b,o,s)	_mpic_map_mmio(m,p,b,o,s)
 #endif /* !CONFIG_PPC_DCR */
 
 
@@ -1152,8 +1153,8 @@ struct mpic * __init mpic_alloc(struct device_node *node,
 	}
 
 	/* Map the global registers */
-	mpic_map(mpic, paddr, &mpic->gregs, MPIC_INFO(GREG_BASE), 0x1000);
-	mpic_map(mpic, paddr, &mpic->tmregs, MPIC_INFO(TIMER_BASE), 0x1000);
+	mpic_map(mpic, node, paddr, &mpic->gregs, MPIC_INFO(GREG_BASE), 0x1000);
+	mpic_map(mpic, node, paddr, &mpic->tmregs, MPIC_INFO(TIMER_BASE), 0x1000);
 
 	/* Reset */
 	if (flags & MPIC_WANTS_RESET) {
@@ -1194,7 +1195,7 @@ struct mpic * __init mpic_alloc(struct device_node *node,
 
 	/* Map the per-CPU registers */
 	for (i = 0; i < mpic->num_cpus; i++) {
-		mpic_map(mpic, paddr, &mpic->cpuregs[i],
+		mpic_map(mpic, node, paddr, &mpic->cpuregs[i],
 			 MPIC_INFO(CPU_BASE) + i * MPIC_INFO(CPU_STRIDE),
 			 0x1000);
 	}
@@ -1202,7 +1203,7 @@ struct mpic * __init mpic_alloc(struct device_node *node,
 	/* Initialize main ISU if none provided */
 	if (mpic->isu_size == 0) {
 		mpic->isu_size = mpic->num_sources;
-		mpic_map(mpic, paddr, &mpic->isus[0],
+		mpic_map(mpic, node, paddr, &mpic->isus[0],
 			 MPIC_INFO(IRQ_BASE), MPIC_INFO(IRQ_STRIDE) * mpic->isu_size);
 	}
 	mpic->isu_shift = 1 + __ilog2(mpic->isu_size - 1);
@@ -1256,8 +1257,10 @@ void __init mpic_assign_isu(struct mpic *mpic, unsigned int isu_num,
 
 	BUG_ON(isu_num >= MPIC_MAX_ISU);
 
-	mpic_map(mpic, paddr, &mpic->isus[isu_num], 0,
+	mpic_map(mpic, mpic->irqhost->of_node,
+		 paddr, &mpic->isus[isu_num], 0,
 		 MPIC_INFO(IRQ_STRIDE) * mpic->isu_size);
+
 	if ((isu_first + mpic->isu_size) > mpic->num_sources)
 		mpic->num_sources = isu_first + mpic->isu_size;
 }
-- 
cgit v1.2.3-59-g8ed1b


From b810c6ec5c659c80b3641580b112db877a0f1f45 Mon Sep 17 00:00:00 2001
From: Jon Smirl <jonsmirl@gmail.com>
Date: Sun, 21 Jun 2009 15:28:00 +0000
Subject: powerpc: Have git ignore generated files from dtc compile

Have git ignore generated files from dtc compile

Signed-off-by: Jon Smirl <jonsmirl@gmail.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Sean MacLennan <smaclennan@pikatech.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/boot/.gitignore | 10 ++++++++++
 scripts/dtc/.gitignore       |  5 +++++
 2 files changed, 15 insertions(+)
 create mode 100644 scripts/dtc/.gitignore

diff --git a/arch/powerpc/boot/.gitignore b/arch/powerpc/boot/.gitignore
index 2f50acd11a60..3d80c3e9cf60 100644
--- a/arch/powerpc/boot/.gitignore
+++ b/arch/powerpc/boot/.gitignore
@@ -36,3 +36,13 @@ zImage.pseries
 zconf.h
 zlib.h
 zutil.h
+fdt.c
+fdt.h
+fdt_ro.c
+fdt_rw.c
+fdt_strerror.c
+fdt_sw.c
+fdt_wip.c
+libfdt.h
+libfdt_internal.h
+
diff --git a/scripts/dtc/.gitignore b/scripts/dtc/.gitignore
new file mode 100644
index 000000000000..095acb49a374
--- /dev/null
+++ b/scripts/dtc/.gitignore
@@ -0,0 +1,5 @@
+dtc
+dtc-lexer.lex.c
+dtc-parser.tab.c
+dtc-parser.tab.h
+
-- 
cgit v1.2.3-59-g8ed1b


From 3984114f056203d833251af85501721f7b00fd18 Mon Sep 17 00:00:00 2001
From: Sean MacLennan <smaclennan@pikatech.com>
Date: Fri, 19 Jun 2009 19:43:59 +0000
Subject: powerpc/warp: Platform fix for i2c change

A change to the i2c subsystem breaks the warp platform code. The patch
is cleaner anyway, the old way was a bit crufty.

For those with keen eyes, the gratuitous change in the string from
PIKA to Warp is just so the logs look a bit nicer. The following two
lines tend to be printed one after another.

  Warp POST OK
  Warp DTM thread running.

Yeah, this will be the third patch to warp.c submitted in this
release....

Cheers,
   Sean

The i2c_client struct changed, breaking the code that looked for the ad7414
chip. Use the new of_find_i2c_device_by_node function added in 2.6.29.

Signed-off-by: Sean MacLennan <smaclennan@pikatech.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/platforms/44x/warp.c | 44 +++++++++++++--------------------------
 1 file changed, 14 insertions(+), 30 deletions(-)

diff --git a/arch/powerpc/platforms/44x/warp.c b/arch/powerpc/platforms/44x/warp.c
index 42e09a9f77e2..0362c88f47d7 100644
--- a/arch/powerpc/platforms/44x/warp.c
+++ b/arch/powerpc/platforms/44x/warp.c
@@ -16,6 +16,7 @@
 #include <linux/interrupt.h>
 #include <linux/delay.h>
 #include <linux/of_gpio.h>
+#include <linux/of_i2c.h>
 
 #include <asm/machdep.h>
 #include <asm/prom.h>
@@ -65,7 +66,6 @@ define_machine(warp) {
 
 static u32 post_info;
 
-/* I am not sure this is the best place for this... */
 static int __init warp_post_info(void)
 {
 	struct device_node *np;
@@ -194,9 +194,9 @@ static int pika_setup_leds(void)
 	return 0;
 }
 
-static void pika_setup_critical_temp(struct i2c_client *client)
+static void pika_setup_critical_temp(struct device_node *np,
+				     struct i2c_client *client)
 {
-	struct device_node *np;
 	int irq, rc;
 
 	/* Do this before enabling critical temp interrupt since we
@@ -208,14 +208,7 @@ static void pika_setup_critical_temp(struct i2c_client *client)
 	i2c_smbus_write_byte_data(client, 2, 65); /* Thigh */
 	i2c_smbus_write_byte_data(client, 3,  0); /* Tlow */
 
-	np = of_find_compatible_node(NULL, NULL, "adi,ad7414");
-	if (np == NULL) {
-		printk(KERN_ERR __FILE__ ": Unable to find ad7414\n");
-		return;
-	}
-
 	irq = irq_of_parse_and_map(np, 0);
-	of_node_put(np);
 	if (irq  == NO_IRQ) {
 		printk(KERN_ERR __FILE__ ": Unable to get ad7414 irq\n");
 		return;
@@ -244,32 +237,24 @@ static inline void pika_dtm_check_fan(void __iomem *fpga)
 
 static int pika_dtm_thread(void __iomem *fpga)
 {
-	struct i2c_adapter *adap;
+	struct device_node *np;
 	struct i2c_client *client;
 
-	/* We loop in case either driver was compiled as a module and
-	 * has not been insmoded yet.
-	 */
-	while (!(adap = i2c_get_adapter(0))) {
-		set_current_state(TASK_INTERRUPTIBLE);
-		schedule_timeout(HZ);
-	}
-
-	while (1) {
-		list_for_each_entry(client, &adap->clients, list)
-			if (client->addr == 0x4a)
-				goto found_it;
+	np = of_find_compatible_node(NULL, NULL, "adi,ad7414");
+	if (np == NULL)
+		return -ENOENT;
 
-		set_current_state(TASK_INTERRUPTIBLE);
-		schedule_timeout(HZ);
+	client = of_find_i2c_device_by_node(np);
+	if (client == NULL) {
+		of_node_put(np);
+		return -ENOENT;
 	}
 
-found_it:
-	pika_setup_critical_temp(client);
+	pika_setup_critical_temp(np, client);
 
-	i2c_put_adapter(adap);
+	of_node_put(np);
 
-	printk(KERN_INFO "PIKA DTM thread running.\n");
+	printk(KERN_INFO "Warp DTM thread running.\n");
 
 	while (!kthread_should_stop()) {
 		int val;
@@ -291,7 +276,6 @@ found_it:
 	return 0;
 }
 
-
 static int __init pika_dtm_start(void)
 {
 	struct task_struct *dtm_thread;
-- 
cgit v1.2.3-59-g8ed1b


From 6bb2ae535f2eee0334802724a542701bd969d055 Mon Sep 17 00:00:00 2001
From: Gerhard Pircher <gerhard_pircher@gmx.net>
Date: Fri, 19 Jun 2009 11:42:36 +0000
Subject: powerpc/amigaone: Limit ISA I/O range to 4k in the device tree

The kernel reserves the I/O address space from 0x0 to 0xfff for legacy
ISA devices. Change the ranges property for the PCI2ISA bridge to match
the kernels behavior, even if the ranges property isn't used for now.

Signed-off-by: Gerhard Pircher <gerhard_pircher@gmx.net>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/boot/dts/amigaone.dts | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/boot/dts/amigaone.dts b/arch/powerpc/boot/dts/amigaone.dts
index 26549fca2ed4..49ac36b16dd7 100644
--- a/arch/powerpc/boot/dts/amigaone.dts
+++ b/arch/powerpc/boot/dts/amigaone.dts
@@ -70,8 +70,8 @@
 			devsel-speed = <0x00000001>;
 			min-grant = <0>;
 			max-latency = <0>;
-			/* First 64k for I/O at 0x0 on PCI mapped to 0x0 on ISA. */
-			ranges = <0x00000001 0 0x01000000 0 0x00000000 0x00010000>;
+			/* First 4k for I/O at 0x0 on PCI mapped to 0x0 on ISA. */
+			ranges = <0x00000001 0 0x01000000 0 0x00000000 0x00001000>;
 			interrupt-parent = <&i8259>;
 			#interrupt-cells = <2>;
 			#address-cells = <2>;
-- 
cgit v1.2.3-59-g8ed1b


From 7ccbe504b5ee766d33211a507189a06f3079b29b Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Thu, 18 Jun 2009 23:30:07 +0000
Subject: powerpc/pmac: Fix issues with PowerMac "PowerSurge" SMP

The old PowerSurge SMP (ie, dual or quad 604 machines) code has
numerous issues in modern world.

One is cpu_possible_map is set too late (the device-tree is bogus)
so we fail to allocate the interrupt stacks and crash. Another
problem is the fact the timebase is frozen by the bringup of the
second CPU so the delays in the generic code will hang, we need
to move some of the calling procedure to inside the powermac code.

This makes it boot again for me

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/smp.c               |   3 +-
 arch/powerpc/platforms/powermac/setup.c |  41 ++------
 arch/powerpc/platforms/powermac/smp.c   | 166 ++++++++++++++++++++------------
 3 files changed, 112 insertions(+), 98 deletions(-)

diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 65484b2200b3..0b47de07302d 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -68,7 +68,8 @@ EXPORT_PER_CPU_SYMBOL(cpu_core_map);
 /* SMP operations for this machine */
 struct smp_ops_t *smp_ops;
 
-static volatile unsigned int cpu_callin_map[NR_CPUS];
+/* Can't be static due to PowerMac hackery */
+volatile unsigned int cpu_callin_map[NR_CPUS];
 
 int smt_enabled_at_boot = 1;
 
diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c
index 86f69a4eb49b..c20522656367 100644
--- a/arch/powerpc/platforms/powermac/setup.c
+++ b/arch/powerpc/platforms/powermac/setup.c
@@ -103,11 +103,6 @@ unsigned long smu_cmdbuf_abs;
 EXPORT_SYMBOL(smu_cmdbuf_abs);
 #endif
 
-#ifdef CONFIG_SMP
-extern struct smp_ops_t psurge_smp_ops;
-extern struct smp_ops_t core99_smp_ops;
-#endif /* CONFIG_SMP */
-
 static void pmac_show_cpuinfo(struct seq_file *m)
 {
 	struct device_node *np;
@@ -341,34 +336,6 @@ static void __init pmac_setup_arch(void)
 		ROOT_DEV = DEFAULT_ROOT_DEVICE;
 #endif
 
-#ifdef CONFIG_SMP
-	/* Check for Core99 */
-	ic = of_find_node_by_name(NULL, "uni-n");
-	if (!ic)
-		ic = of_find_node_by_name(NULL, "u3");
-	if (!ic)
-		ic = of_find_node_by_name(NULL, "u4");
-	if (ic) {
-		of_node_put(ic);
-		smp_ops = &core99_smp_ops;
-	}
-#ifdef CONFIG_PPC32
-	else {
-		/*
-		 * We have to set bits in cpu_possible_map here since the
-		 * secondary CPU(s) aren't in the device tree, and
-		 * setup_per_cpu_areas only allocates per-cpu data for
-		 * CPUs in the cpu_possible_map.
-		 */
-		int cpu;
-
-		for (cpu = 1; cpu < 4 && cpu < NR_CPUS; ++cpu)
-			cpu_set(cpu, cpu_possible_map);
-		smp_ops = &psurge_smp_ops;
-	}
-#endif
-#endif /* CONFIG_SMP */
-
 #ifdef CONFIG_ADB
 	if (strstr(cmd_line, "adb_sync")) {
 		extern int __adb_probe_sync;
@@ -512,6 +479,14 @@ static void __init pmac_init_early(void)
 #ifdef CONFIG_PPC64
 	iommu_init_early_dart();
 #endif
+
+	/* SMP Init has to be done early as we need to patch up
+	 * cpu_possible_map before interrupt stacks are allocated
+	 * or kaboom...
+	 */
+#ifdef CONFIG_SMP
+	pmac_setup_smp();
+#endif
 }
 
 static int __init pmac_declare_of_platform_devices(void)
diff --git a/arch/powerpc/platforms/powermac/smp.c b/arch/powerpc/platforms/powermac/smp.c
index cf1dbe758890..6d4da7b46b41 100644
--- a/arch/powerpc/platforms/powermac/smp.c
+++ b/arch/powerpc/platforms/powermac/smp.c
@@ -64,10 +64,11 @@
 extern void __secondary_start_pmac_0(void);
 extern int pmac_pfunc_base_install(void);
 
-#ifdef CONFIG_PPC32
+static void (*pmac_tb_freeze)(int freeze);
+static u64 timebase;
+static int tb_req;
 
-/* Sync flag for HW tb sync */
-static volatile int sec_tb_reset = 0;
+#ifdef CONFIG_PPC32
 
 /*
  * Powersurge (old powermac SMP) support.
@@ -294,6 +295,9 @@ static int __init smp_psurge_probe(void)
 		psurge_quad_init();
 		/* All released cards using this HW design have 4 CPUs */
 		ncpus = 4;
+		/* No sure how timebase sync works on those, let's use SW */
+		smp_ops->give_timebase = smp_generic_give_timebase;
+		smp_ops->take_timebase = smp_generic_take_timebase;
 	} else {
 		iounmap(quad_base);
 		if ((in_8(hhead_base + HHEAD_CONFIG) & 0x02) == 0) {
@@ -308,18 +312,15 @@ static int __init smp_psurge_probe(void)
 	psurge_start = ioremap(PSURGE_START, 4);
 	psurge_pri_intr = ioremap(PSURGE_PRI_INTR, 4);
 
-	/*
-	 * This is necessary because OF doesn't know about the
+	/* This is necessary because OF doesn't know about the
 	 * secondary cpu(s), and thus there aren't nodes in the
 	 * device tree for them, and smp_setup_cpu_maps hasn't
-	 * set their bits in cpu_possible_map and cpu_present_map.
+	 * set their bits in cpu_present_map.
 	 */
 	if (ncpus > NR_CPUS)
 		ncpus = NR_CPUS;
-	for (i = 1; i < ncpus ; ++i) {
+	for (i = 1; i < ncpus ; ++i)
 		cpu_set(i, cpu_present_map);
-		set_hard_smp_processor_id(i, i);
-	}
 
 	if (ppc_md.progress) ppc_md.progress("smp_psurge_probe - done", 0x352);
 
@@ -329,8 +330,14 @@ static int __init smp_psurge_probe(void)
 static void __init smp_psurge_kick_cpu(int nr)
 {
 	unsigned long start = __pa(__secondary_start_pmac_0) + nr * 8;
-	unsigned long a;
-	int i;
+	unsigned long a, flags;
+	int i, j;
+
+	/* Defining this here is evil ... but I prefer hiding that
+	 * crap to avoid giving people ideas that they can do the
+	 * same.
+	 */
+	extern volatile unsigned int cpu_callin_map[NR_CPUS];
 
 	/* may need to flush here if secondary bats aren't setup */
 	for (a = KERNELBASE; a < KERNELBASE + 0x800000; a += 32)
@@ -339,47 +346,52 @@ static void __init smp_psurge_kick_cpu(int nr)
 
 	if (ppc_md.progress) ppc_md.progress("smp_psurge_kick_cpu", 0x353);
 
+	/* This is going to freeze the timeebase, we disable interrupts */
+	local_irq_save(flags);
+
 	out_be32(psurge_start, start);
 	mb();
 
 	psurge_set_ipi(nr);
+
 	/*
 	 * We can't use udelay here because the timebase is now frozen.
 	 */
 	for (i = 0; i < 2000; ++i)
-		barrier();
+		asm volatile("nop" : : : "memory");
 	psurge_clr_ipi(nr);
 
-	if (ppc_md.progress) ppc_md.progress("smp_psurge_kick_cpu - done", 0x354);
-}
-
-/*
- * With the dual-cpu powersurge board, the decrementers and timebases
- * of both cpus are frozen after the secondary cpu is started up,
- * until we give the secondary cpu another interrupt.  This routine
- * uses this to get the timebases synchronized.
- *  -- paulus.
- */
-static void __init psurge_dual_sync_tb(int cpu_nr)
-{
-	int t;
-
-	set_dec(tb_ticks_per_jiffy);
-	/* XXX fixme */
-	set_tb(0, 0);
-
-	if (cpu_nr > 0) {
+	/*
+	 * Also, because the timebase is frozen, we must not return to the
+	 * caller which will try to do udelay's etc... Instead, we wait -here-
+	 * for the CPU to callin.
+	 */
+	for (i = 0; i < 100000 && !cpu_callin_map[nr]; ++i) {
+		for (j = 1; j < 10000; j++)
+			asm volatile("nop" : : : "memory");
+		asm volatile("sync" : : : "memory");
+	}
+	if (!cpu_callin_map[nr])
+		goto stuck;
+
+	/* And we do the TB sync here too for standard dual CPU cards */
+	if (psurge_type == PSURGE_DUAL) {
+		while(!tb_req)
+			barrier();
+		tb_req = 0;
+		mb();
+		timebase = get_tb();
+		mb();
+		while (timebase)
+			barrier();
 		mb();
-		sec_tb_reset = 1;
-		return;
 	}
+ stuck:
+	/* now interrupt the secondary, restarting both TBs */
+	if (psurge_type == PSURGE_DUAL)
+		psurge_set_ipi(1);
 
-	/* wait for the secondary to have reset its TB before proceeding */
-	for (t = 10000000; t > 0 && !sec_tb_reset; --t)
-		;
-
-	/* now interrupt the secondary, starting both TBs */
-	psurge_set_ipi(1);
+	if (ppc_md.progress) ppc_md.progress("smp_psurge_kick_cpu - done", 0x354);
 }
 
 static struct irqaction psurge_irqaction = {
@@ -390,36 +402,35 @@ static struct irqaction psurge_irqaction = {
 
 static void __init smp_psurge_setup_cpu(int cpu_nr)
 {
+	if (cpu_nr != 0)
+		return;
 
-	if (cpu_nr == 0) {
-		/* If we failed to start the second CPU, we should still
-		 * send it an IPI to start the timebase & DEC or we might
-		 * have them stuck.
-		 */
-		if (num_online_cpus() < 2) {
-			if (psurge_type == PSURGE_DUAL)
-				psurge_set_ipi(1);
-			return;
-		}
-		/* reset the entry point so if we get another intr we won't
-		 * try to startup again */
-		out_be32(psurge_start, 0x100);
-		if (setup_irq(30, &psurge_irqaction))
-			printk(KERN_ERR "Couldn't get primary IPI interrupt");
-	}
-
-	if (psurge_type == PSURGE_DUAL)
-		psurge_dual_sync_tb(cpu_nr);
+	/* reset the entry point so if we get another intr we won't
+	 * try to startup again */
+	out_be32(psurge_start, 0x100);
+	if (setup_irq(30, &psurge_irqaction))
+		printk(KERN_ERR "Couldn't get primary IPI interrupt");
 }
 
 void __init smp_psurge_take_timebase(void)
 {
-	/* Dummy implementation */
+	if (psurge_type != PSURGE_DUAL)
+		return;
+
+	tb_req = 1;
+	mb();
+	while (!timebase)
+		barrier();
+	mb();
+	set_tb(timebase >> 32, timebase & 0xffffffff);
+	timebase = 0;
+	mb();
+	set_dec(tb_ticks_per_jiffy/2);
 }
 
 void __init smp_psurge_give_timebase(void)
 {
-	/* Dummy implementation */
+	/* Nothing to do here */
 }
 
 /* PowerSurge-style Macs */
@@ -437,9 +448,6 @@ struct smp_ops_t psurge_smp_ops = {
  * Core 99 and later support
  */
 
-static void (*pmac_tb_freeze)(int freeze);
-static u64 timebase;
-static int tb_req;
 
 static void smp_core99_give_timebase(void)
 {
@@ -478,7 +486,6 @@ static void __devinit smp_core99_take_timebase(void)
 	set_tb(timebase >> 32, timebase & 0xffffffff);
 	timebase = 0;
 	mb();
-	set_dec(tb_ticks_per_jiffy/2);
 
 	local_irq_restore(flags);
 }
@@ -920,3 +927,34 @@ struct smp_ops_t core99_smp_ops = {
 # endif
 #endif
 };
+
+void __init pmac_setup_smp(void)
+{
+	struct device_node *np;
+
+	/* Check for Core99 */
+	np = of_find_node_by_name(NULL, "uni-n");
+	if (!np)
+		np = of_find_node_by_name(NULL, "u3");
+	if (!np)
+		np = of_find_node_by_name(NULL, "u4");
+	if (np) {
+		of_node_put(np);
+		smp_ops = &core99_smp_ops;
+	}
+#ifdef CONFIG_PPC32
+	else {
+		/* We have to set bits in cpu_possible_map here since the
+		 * secondary CPU(s) aren't in the device tree. Various
+		 * things won't be initialized for CPUs not in the possible
+		 * map, so we really need to fix it up here.
+		 */
+		int cpu;
+
+		for (cpu = 1; cpu < 4 && cpu < NR_CPUS; ++cpu)
+			cpu_set(cpu, cpu_possible_map);
+		smp_ops = &psurge_smp_ops;
+	}
+#endif /* CONFIG_PPC32 */
+}
+
-- 
cgit v1.2.3-59-g8ed1b


From a2367194183d6ab6b05e5d7d9b40db6ba48afc06 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@kernel.crashing.org>
Date: Thu, 18 Jun 2009 22:29:55 +0000
Subject: powerpc: Fix output from show_regs

For some reason we've had an explicit KERN_INFO for GPR dumps.  With
recent changes we get output like:

<6>GPR00: 00000000 ef855eb0 ef858000 00000001 000000d0 f1000000 ffbc8000 ffffffff

The KERN_INFO is causing the <6>.  Don't see any reason to keep it
around.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/process.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index 3e7135bbe40f..892a9f2e6d76 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -528,7 +528,7 @@ void show_regs(struct pt_regs * regs)
 
 	for (i = 0;  i < 32;  i++) {
 		if ((i % REGS_PER_LINE) == 0)
-			printk("\n" KERN_INFO "GPR%02d: ", i);
+			printk("\nGPR%02d: ", i);
 		printk(REG " ", regs->gpr[i]);
 		if (i == LAST_VOLATILE && !FULL_REGS(regs))
 			break;
-- 
cgit v1.2.3-59-g8ed1b


From 85355bb272db31a3f2dd99d547eef794805e1319 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@kernel.crashing.org>
Date: Thu, 18 Jun 2009 22:01:20 +0000
Subject: powerpc: Fix mpic alloc warning

Since we can use kmalloc earlier we are getting the following since the
mpic_alloc() code calls alloc_bootmem().  Move to using kzalloc() to
remove the warning.

------------[ cut here ]------------
Badness at c0583248 [verbose debug info unavailable]
NIP: c0583248 LR: c0583210 CTR: 00000004
REGS: c0741de0 TRAP: 0700   Not tainted  (2.6.30-06736-g12a31df)
MSR: 00021000 <ME,CE>  CR: 22024024  XER: 00000000
TASK = c070d3b8[0] 'swapper' THREAD: c0740000 CPU: 0
<6>GPR00: 00000001 c0741e90 c070d3b8 00000001 00000210 00000020 3fffffff 00000000
<6>GPR08: 00000000 c0c85700 c04f8c40 0000002d 22044022 1004a388 7ffd9400 00000000
<6>GPR16: 00000000 7ffcd100 7ffcd100 7ffcd100 c04f8c40 00000000 c059f62c c075a0c0
<6>GPR24: c059f648 00000000 0000000f 00000210 00000020 00000000 3fffffff 00000210
NIP [c0583248] alloc_arch_preferred_bootmem+0x50/0x80
LR [c0583210] alloc_arch_preferred_bootmem+0x18/0x80
Call Trace:
[c0741e90] [c07343b0] devtree_lock+0x0/0x24 (unreliable)
[c0741ea0] [c0583b14] ___alloc_bootmem_nopanic+0x54/0x108
[c0741ee0] [c0583e18] ___alloc_bootmem+0x18/0x50
[c0741ef0] [c057b9cc] mpic_alloc+0x48/0x710
[c0741f40] [c057ecf4] mpc85xx_ds_pic_init+0x190/0x1b8
[c0741f90] [c057633c] init_IRQ+0x24/0x34
[c0741fa0] [c05738b8] start_kernel+0x260/0x3dc
[c0741ff0] [c00003c8] skpinv+0x2e0/0x31c
Instruction dump:
409e001c 7c030378 80010014 83e1000c 38210010 7c0803a6 4e800020 3d20c0c8
39295700 80090004 7c000034 5400d97e <0f000000> 2f800000 409e001c 38800000

BenH: Changed to use GFP_KERNEL, the allocator will do the right thing

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/sysdev/mpic.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c
index 32a2e950f563..d46de1f0f3ee 100644
--- a/arch/powerpc/sysdev/mpic.c
+++ b/arch/powerpc/sysdev/mpic.c
@@ -1053,11 +1053,10 @@ struct mpic * __init mpic_alloc(struct device_node *node,
 	int		intvec_top;
 	u64		paddr = phys_addr;
 
-	mpic = alloc_bootmem(sizeof(struct mpic));
+	mpic = kzalloc(sizeof(struct mpic), GFP_KERNEL);
 	if (mpic == NULL)
 		return NULL;
-	
-	memset(mpic, 0, sizeof(struct mpic));
+
 	mpic->name = name;
 
 	mpic->hc_irq = mpic_irq_chip;
-- 
cgit v1.2.3-59-g8ed1b


From 850f6ac316cf84bba63fdb775c897834eccbfaa3 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Thu, 18 Jun 2009 19:25:00 +0000
Subject: powerpc/mm: Make k(un)map_atomic out of line

Those functions are way too big to be inline, besides, kmap_atomic()
wants to call debug_kmap_atomic() which isn't exported for modules
and causes module link failures.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/highmem.h | 57 ++--------------------------
 arch/powerpc/mm/Makefile           |  1 +
 arch/powerpc/mm/highmem.c          | 77 ++++++++++++++++++++++++++++++++++++++
 3 files changed, 82 insertions(+), 53 deletions(-)
 create mode 100644 arch/powerpc/mm/highmem.c

diff --git a/arch/powerpc/include/asm/highmem.h b/arch/powerpc/include/asm/highmem.h
index 684a73f4324f..a74c4ee6c020 100644
--- a/arch/powerpc/include/asm/highmem.h
+++ b/arch/powerpc/include/asm/highmem.h
@@ -22,9 +22,7 @@
 
 #ifdef __KERNEL__
 
-#include <linux/init.h>
 #include <linux/interrupt.h>
-#include <linux/highmem.h>
 #include <asm/kmap_types.h>
 #include <asm/tlbflush.h>
 #include <asm/page.h>
@@ -62,6 +60,9 @@ extern pte_t *pkmap_page_table;
 
 extern void *kmap_high(struct page *page);
 extern void kunmap_high(struct page *page);
+extern void *kmap_atomic_prot(struct page *page, enum km_type type,
+			      pgprot_t prot);
+extern void kunmap_atomic(void *kvaddr, enum km_type type);
 
 static inline void *kmap(struct page *page)
 {
@@ -79,62 +80,11 @@ static inline void kunmap(struct page *page)
 	kunmap_high(page);
 }
 
-/*
- * The use of kmap_atomic/kunmap_atomic is discouraged - kmap/kunmap
- * gives a more generic (and caching) interface. But kmap_atomic can
- * be used in IRQ contexts, so in some (very limited) cases we need
- * it.
- */
-static inline void *kmap_atomic_prot(struct page *page, enum km_type type, pgprot_t prot)
-{
-	unsigned int idx;
-	unsigned long vaddr;
-
-	/* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */
-	pagefault_disable();
-	if (!PageHighMem(page))
-		return page_address(page);
-
-	debug_kmap_atomic(type);
-	idx = type + KM_TYPE_NR*smp_processor_id();
-	vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
-#ifdef CONFIG_DEBUG_HIGHMEM
-	BUG_ON(!pte_none(*(kmap_pte-idx)));
-#endif
-	__set_pte_at(&init_mm, vaddr, kmap_pte-idx, mk_pte(page, prot), 1);
-	local_flush_tlb_page(NULL, vaddr);
-
-	return (void*) vaddr;
-}
-
 static inline void *kmap_atomic(struct page *page, enum km_type type)
 {
 	return kmap_atomic_prot(page, type, kmap_prot);
 }
 
-static inline void kunmap_atomic(void *kvaddr, enum km_type type)
-{
-#ifdef CONFIG_DEBUG_HIGHMEM
-	unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
-	enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id();
-
-	if (vaddr < __fix_to_virt(FIX_KMAP_END)) {
-		pagefault_enable();
-		return;
-	}
-
-	BUG_ON(vaddr != __fix_to_virt(FIX_KMAP_BEGIN + idx));
-
-	/*
-	 * force other mappings to Oops if they'll try to access
-	 * this pte without first remap it
-	 */
-	pte_clear(&init_mm, vaddr, kmap_pte-idx);
-	local_flush_tlb_page(NULL, vaddr);
-#endif
-	pagefault_enable();
-}
-
 static inline struct page *kmap_atomic_to_page(void *ptr)
 {
 	unsigned long idx, vaddr = (unsigned long) ptr;
@@ -148,6 +98,7 @@ static inline struct page *kmap_atomic_to_page(void *ptr)
 	return pte_page(*pte);
 }
 
+
 #define flush_cache_kmaps()	flush_cache_all()
 
 #endif /* __KERNEL__ */
diff --git a/arch/powerpc/mm/Makefile b/arch/powerpc/mm/Makefile
index 2d2192e48de7..3e68363405b7 100644
--- a/arch/powerpc/mm/Makefile
+++ b/arch/powerpc/mm/Makefile
@@ -30,3 +30,4 @@ obj-$(CONFIG_PPC_MM_SLICES)	+= slice.o
 obj-$(CONFIG_HUGETLB_PAGE)	+= hugetlbpage.o
 obj-$(CONFIG_PPC_SUBPAGE_PROT)	+= subpage-prot.o
 obj-$(CONFIG_NOT_COHERENT_CACHE) += dma-noncoherent.o
+obj-$(CONFIG_HIGHMEM)		+= highmem.o
diff --git a/arch/powerpc/mm/highmem.c b/arch/powerpc/mm/highmem.c
new file mode 100644
index 000000000000..c2186c74c85a
--- /dev/null
+++ b/arch/powerpc/mm/highmem.c
@@ -0,0 +1,77 @@
+/*
+ * highmem.c: virtual kernel memory mappings for high memory
+ *
+ * PowerPC version, stolen from the i386 version.
+ *
+ * Used in CONFIG_HIGHMEM systems for memory pages which
+ * are not addressable by direct kernel virtual addresses.
+ *
+ * Copyright (C) 1999 Gerhard Wichert, Siemens AG
+ *		      Gerhard.Wichert@pdb.siemens.de
+ *
+ *
+ * Redesigned the x86 32-bit VM architecture to deal with
+ * up to 16 Terrabyte physical memory. With current x86 CPUs
+ * we now support up to 64 Gigabytes physical RAM.
+ *
+ * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
+ *
+ * Reworked for PowerPC by various contributors. Moved from
+ * highmem.h by Benjamin Herrenschmidt (c) 2009 IBM Corp.
+ */
+
+#include <linux/highmem.h>
+#include <linux/module.h>
+
+/*
+ * The use of kmap_atomic/kunmap_atomic is discouraged - kmap/kunmap
+ * gives a more generic (and caching) interface. But kmap_atomic can
+ * be used in IRQ contexts, so in some (very limited) cases we need
+ * it.
+ */
+void *kmap_atomic_prot(struct page *page, enum km_type type, pgprot_t prot)
+{
+	unsigned int idx;
+	unsigned long vaddr;
+
+	/* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */
+	pagefault_disable();
+	if (!PageHighMem(page))
+		return page_address(page);
+
+	debug_kmap_atomic(type);
+	idx = type + KM_TYPE_NR*smp_processor_id();
+	vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
+#ifdef CONFIG_DEBUG_HIGHMEM
+	BUG_ON(!pte_none(*(kmap_pte-idx)));
+#endif
+	__set_pte_at(&init_mm, vaddr, kmap_pte-idx, mk_pte(page, prot), 1);
+	local_flush_tlb_page(NULL, vaddr);
+
+	return (void*) vaddr;
+}
+EXPORT_SYMBOL(kmap_atomic_prot);
+
+void kunmap_atomic(void *kvaddr, enum km_type type)
+{
+#ifdef CONFIG_DEBUG_HIGHMEM
+	unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
+	enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id();
+
+	if (vaddr < __fix_to_virt(FIX_KMAP_END)) {
+		pagefault_enable();
+		return;
+	}
+
+	BUG_ON(vaddr != __fix_to_virt(FIX_KMAP_BEGIN + idx));
+
+	/*
+	 * force other mappings to Oops if they'll try to access
+	 * this pte without first remap it
+	 */
+	pte_clear(&init_mm, vaddr, kmap_pte-idx);
+	local_flush_tlb_page(NULL, vaddr);
+#endif
+	pagefault_enable();
+}
+EXPORT_SYMBOL(kunmap_atomic);
-- 
cgit v1.2.3-59-g8ed1b


From 3514141aedc16c7344117d5bd79ec1310edf8fb3 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Thu, 18 Jun 2009 19:20:51 +0000
Subject: powerpc/pmac: Fix DMA ops for MacIO devices

The macio_dev's created to map devices inside the MacIO ASICs
don't have proper dma_ops. This causes crashes on some machines
since the SCSI code calls dma_map_* on our behalf using the
device we hang from.

This fixes it by copying the parent PCI device dma_ops into
the macio_dev when creating it.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 drivers/macintosh/macio_asic.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/macintosh/macio_asic.c b/drivers/macintosh/macio_asic.c
index 6e149f4a1fff..a0f68386c12f 100644
--- a/drivers/macintosh/macio_asic.c
+++ b/drivers/macintosh/macio_asic.c
@@ -378,6 +378,17 @@ static struct macio_dev * macio_add_one_device(struct macio_chip *chip,
 	dev->ofdev.dev.bus = &macio_bus_type;
 	dev->ofdev.dev.release = macio_release_dev;
 
+#ifdef CONFIG_PCI
+	/* Set the DMA ops to the ones from the PCI device, this could be
+	 * fishy if we didn't know that on PowerMac it's always direct ops
+	 * or iommu ops that will work fine
+	 */
+	dev->ofdev.dev.archdata.dma_ops =
+		chip->lbus.pdev->dev.archdata.dma_ops;
+	dev->ofdev.dev.archdata.dma_data =
+		chip->lbus.pdev->dev.archdata.dma_data;
+#endif /* CONFIG_PCI */
+
 #ifdef DEBUG
 	printk("preparing mdev @%p, ofdev @%p, dev @%p, kobj @%p\n",
 	       dev, &dev->ofdev, &dev->ofdev.dev, &dev->ofdev.dev.kobj);
-- 
cgit v1.2.3-59-g8ed1b


From 4a5cbf17c49a6024a6d7baf03efdffb8ed252bb1 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Thu, 18 Jun 2009 19:17:39 +0000
Subject: powerpc: Map more memory early on 601 processors

The 32-bit kernel relies on some memory being mapped covering
the kernel text,data and bss at least, early during boot before
the full MMU setup is done. On 32-bit "classic" processors, this
is done using BAT registers.

On 601, the size of BATs is limited to 8M and we use 2 of them
for that initial mapping. This can become quite tight when enabling
features like lockdep, so let's use a 3rd one to bump that mapping
from 16M to 24M. We keep the 4th BAT free as it can be useful for
debugging early boot code to map things like serial ports.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/head_32.S | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/arch/powerpc/kernel/head_32.S b/arch/powerpc/kernel/head_32.S
index 48469463f89e..fc2132942754 100644
--- a/arch/powerpc/kernel/head_32.S
+++ b/arch/powerpc/kernel/head_32.S
@@ -1124,9 +1124,8 @@ mmu_off:
 	RFI
 
 /*
- * Use the first pair of BAT registers to map the 1st 16MB
- * of RAM to PAGE_OFFSET.  From this point on we can't safely
- * call OF any more.
+ * On 601, we use 3 BATs to map up to 24M of RAM at _PAGE_OFFSET
+ * (we keep one for debugging) and on others, we use one 256M BAT.
  */
 initial_bats:
 	lis	r11,PAGE_OFFSET@h
@@ -1136,12 +1135,16 @@ initial_bats:
 	bne	4f
 	ori	r11,r11,4		/* set up BAT registers for 601 */
 	li	r8,0x7f			/* valid, block length = 8MB */
-	oris	r9,r11,0x800000@h	/* set up BAT reg for 2nd 8M */
-	oris	r10,r8,0x800000@h	/* set up BAT reg for 2nd 8M */
 	mtspr	SPRN_IBAT0U,r11		/* N.B. 601 has valid bit in */
 	mtspr	SPRN_IBAT0L,r8		/* lower BAT register */
-	mtspr	SPRN_IBAT1U,r9
-	mtspr	SPRN_IBAT1L,r10
+	addis	r11,r11,0x800000@h
+	addis	r8,r8,0x800000@h
+	mtspr	SPRN_IBAT1U,r11
+	mtspr	SPRN_IBAT1L,r8
+	addis	r11,r11,0x800000@h
+	addis	r8,r8,0x800000@h
+	mtspr	SPRN_IBAT2U,r11
+	mtspr	SPRN_IBAT2L,r8
 	isync
 	blr
 
-- 
cgit v1.2.3-59-g8ed1b


From e4031d52c57b17c76bbdb15fcf1a32a9f87d9756 Mon Sep 17 00:00:00 2001
From: Sonny Rao <sonnyrao@us.ibm.com>
Date: Thu, 18 Jun 2009 15:14:36 +0000
Subject: powerpc/BSR: add 4096 byte BSR size

Add a 4096 byte BSR size which will be used on new machines.  Also, remove
the warning when we run into an unknown size, as this can spam the kernel
log excessively.

Signed-off-by: Sonny Rao <sonnyrao@us.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 drivers/char/bsr.c | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/drivers/char/bsr.c b/drivers/char/bsr.c
index 140ea10ecb88..7d9fd8a8dfd0 100644
--- a/drivers/char/bsr.c
+++ b/drivers/char/bsr.c
@@ -75,12 +75,13 @@ static struct class *bsr_class;
 static int bsr_major;
 
 enum {
-	BSR_8   = 0,
-	BSR_16  = 1,
-	BSR_64  = 2,
-	BSR_128 = 3,
-	BSR_UNKNOWN = 4,
-	BSR_MAX = 5,
+	BSR_8    = 0,
+	BSR_16   = 1,
+	BSR_64   = 2,
+	BSR_128  = 3,
+	BSR_4096 = 4,
+	BSR_UNKNOWN = 5,
+	BSR_MAX  = 6,
 };
 
 static unsigned bsr_types[BSR_MAX];
@@ -218,9 +219,11 @@ static int bsr_add_node(struct device_node *bn)
 		case 128:
 			cur->bsr_type = BSR_128;
 			break;
+		case 4096:
+			cur->bsr_type = BSR_4096;
+			break;
 		default:
 			cur->bsr_type = BSR_UNKNOWN;
-			printk(KERN_INFO "unknown BSR size %d\n",cur->bsr_bytes);
 		}
 
 		cur->bsr_num = bsr_types[cur->bsr_type];
-- 
cgit v1.2.3-59-g8ed1b


From 04a85d1234d7e1682a612565e663e6b760918643 Mon Sep 17 00:00:00 2001
From: Sonny Rao <sonnyrao@us.ibm.com>
Date: Thu, 18 Jun 2009 15:13:04 +0000
Subject: powerpc/BSR: Fix BSR to allow mmap of small BSR on 64k kernel

On Mon, Nov 17, 2008 at 01:26:13AM -0600, Sonny Rao wrote:
> On Fri, Nov 07, 2008 at 04:28:29PM +1100, Paul Mackerras wrote:
> > Sonny Rao writes:
> >
> > > Fix the BSR driver to allow small BSR devices, which are limited to a
> > > single 4k space, on a 64k page kernel.  Previously the driver would
> > > reject the mmap since the size was smaller than PAGESIZE (or because
> > > the size was greater than the size of the device).  Now, we check for
> > > this case use remap_4k_pfn(). Also, take out code to set vm_flags,
> > > as the remap_pfn functions will do this for us.
> >
> > Thanks.
> >
> > Do we know that the BSR size will always be 4k if it's not a multiple
> > of 64k?  Is it possible that we could get 8k, 16k or 32k or BSRs?
> > If it is possible, what does the user need to be able to do?  Do they
> > just want to map 4k, or might then want to map the whole thing?
>
>
> Hi Paul, I took a look at changing the driver to reject a request for
> mapping more than a single 4k page, however the only indication we get
> of the requested size in the mmap function is the vma size, and this
> is always one page at minimum.  So, it's not possible to determine if
> the user wants one 4k page or more.  As I noted in my first response,
> there is only one case where this is even possible and I don't think
> it is a significant concern.
>
> I did notice that I left out the check to see if the user is trying to
> map more than the device length, so I fixed that.  Here's the revised
> patch.

Alright, I've reworked this now so that if we get one of these cases
where there's a bsr that's > 4k and < 64k on a 64k kernel we'll only
advertise that it is a 4k BSR to userspace.  I think this is the best
solution since user programs are only supposed to look at sysfs to
determine how much can be mapped, and libbsr does this as well.

Please consider for 2.6.31 as a fix, thanks.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 drivers/char/bsr.c | 25 +++++++++++++++++++------
 1 file changed, 19 insertions(+), 6 deletions(-)

diff --git a/drivers/char/bsr.c b/drivers/char/bsr.c
index 7d9fd8a8dfd0..c02db01f736e 100644
--- a/drivers/char/bsr.c
+++ b/drivers/char/bsr.c
@@ -27,6 +27,7 @@
 #include <linux/cdev.h>
 #include <linux/list.h>
 #include <linux/mm.h>
+#include <asm/pgtable.h>
 #include <asm/io.h>
 
 /*
@@ -118,15 +119,22 @@ static int bsr_mmap(struct file *filp, struct vm_area_struct *vma)
 {
 	unsigned long size   = vma->vm_end - vma->vm_start;
 	struct bsr_dev *dev = filp->private_data;
+	int ret;
 
-	if (size > dev->bsr_len || (size & (PAGE_SIZE-1)))
-		return -EINVAL;
-
-	vma->vm_flags |= (VM_IO | VM_DONTEXPAND);
 	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
 
-	if (io_remap_pfn_range(vma, vma->vm_start, dev->bsr_addr >> PAGE_SHIFT,
-			       size, vma->vm_page_prot))
+	/* check for the case of a small BSR device and map one 4k page for it*/
+	if (dev->bsr_len < PAGE_SIZE && size == PAGE_SIZE)
+		ret = remap_4k_pfn(vma, vma->vm_start, dev->bsr_addr >> 12,
+				   vma->vm_page_prot);
+	else if (size <= dev->bsr_len)
+		ret = io_remap_pfn_range(vma, vma->vm_start,
+					 dev->bsr_addr >> PAGE_SHIFT,
+					 size, vma->vm_page_prot);
+	else
+		return -EINVAL;
+
+	if (ret)
 		return -EAGAIN;
 
 	return 0;
@@ -206,6 +214,11 @@ static int bsr_add_node(struct device_node *bn)
 		cur->bsr_stride = bsr_stride[i];
 		cur->bsr_dev    = MKDEV(bsr_major, i + total_bsr_devs);
 
+		/* if we have a bsr_len of > 4k and less then PAGE_SIZE (64k pages) */
+		/* we can only map 4k of it, so only advertise the 4k in sysfs */
+		if (cur->bsr_len > 4096 && cur->bsr_len < PAGE_SIZE)
+			cur->bsr_len = 4096;
+
 		switch(cur->bsr_bytes) {
 		case 8:
 			cur->bsr_type = BSR_8;
-- 
cgit v1.2.3-59-g8ed1b


From 5d38902c483881645ba16058cffaa478b81e5cfa Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 17 Jun 2009 17:43:59 +0000
Subject: powerpc: Add irqtrace support for 32-bit powerpc

Based on initial work from: Dale Farnsworth <dale@farnsworth.org>

Add the low level irq tracing hooks for 32-bit powerpc needed
to enable full lockdep functionality.

The approach taken to deal with the code in entry_32.S is that
we don't trace all the transitions of MSR:EE when we just turn
it off to peek at TI_FLAGS without races. Only when we are
calling into C code or returning from exceptions with a state
that have changed from what lockdep thinks.

There's a little bugger though: If we take an exception that
keeps interrupts enabled (such as an alignment exception) while
interrupts are enabled, we will call trace_hardirqs_on() on the
way back spurriously. Not a big deal, but to get rid of it would
require remembering in pt_regs that the exception was one of the
type that kept interrupts enabled which we don't know at this
stage. (Well, we could test all cases for regs->trap but that
sucks too much).

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Tested-by: Kumar Gala <galak@kernel.crashing.org>
---
 arch/powerpc/Kconfig              |   1 -
 arch/powerpc/include/asm/hw_irq.h |  20 +++---
 arch/powerpc/kernel/entry_32.S    | 127 ++++++++++++++++++++++++++++++++++++--
 arch/powerpc/kernel/setup_32.c    |   2 +
 4 files changed, 135 insertions(+), 15 deletions(-)

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index bf6cedfa05db..d00131ca0835 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -62,7 +62,6 @@ config HAVE_LATENCYTOP_SUPPORT
 
 config TRACE_IRQFLAGS_SUPPORT
 	bool
-	depends on PPC64
 	default y
 
 config LOCKDEP_SUPPORT
diff --git a/arch/powerpc/include/asm/hw_irq.h b/arch/powerpc/include/asm/hw_irq.h
index 867ab8ed69b3..8b505eaaa38a 100644
--- a/arch/powerpc/include/asm/hw_irq.h
+++ b/arch/powerpc/include/asm/hw_irq.h
@@ -68,13 +68,13 @@ static inline int irqs_disabled_flags(unsigned long flags)
 
 #if defined(CONFIG_BOOKE)
 #define SET_MSR_EE(x)	mtmsr(x)
-#define local_irq_restore(flags)	__asm__ __volatile__("wrtee %0" : : "r" (flags) : "memory")
+#define raw_local_irq_restore(flags)	__asm__ __volatile__("wrtee %0" : : "r" (flags) : "memory")
 #else
 #define SET_MSR_EE(x)	mtmsr(x)
-#define local_irq_restore(flags)	mtmsr(flags)
+#define raw_local_irq_restore(flags)	mtmsr(flags)
 #endif
 
-static inline void local_irq_disable(void)
+static inline void raw_local_irq_disable(void)
 {
 #ifdef CONFIG_BOOKE
 	__asm__ __volatile__("wrteei 0": : :"memory");
@@ -86,7 +86,7 @@ static inline void local_irq_disable(void)
 #endif
 }
 
-static inline void local_irq_enable(void)
+static inline void raw_local_irq_enable(void)
 {
 #ifdef CONFIG_BOOKE
 	__asm__ __volatile__("wrteei 1": : :"memory");
@@ -98,7 +98,7 @@ static inline void local_irq_enable(void)
 #endif
 }
 
-static inline void local_irq_save_ptr(unsigned long *flags)
+static inline void raw_local_irq_save_ptr(unsigned long *flags)
 {
 	unsigned long msr;
 	msr = mfmsr();
@@ -110,12 +110,12 @@ static inline void local_irq_save_ptr(unsigned long *flags)
 #endif
 }
 
-#define local_save_flags(flags)	((flags) = mfmsr())
-#define local_irq_save(flags)	local_irq_save_ptr(&flags)
-#define irqs_disabled()		((mfmsr() & MSR_EE) == 0)
+#define raw_local_save_flags(flags)	((flags) = mfmsr())
+#define raw_local_irq_save(flags)	raw_local_irq_save_ptr(&flags)
+#define raw_irqs_disabled()		((mfmsr() & MSR_EE) == 0)
+#define raw_irqs_disabled_flags(flags)	(((flags) & MSR_EE) == 0)
 
-#define hard_irq_enable()	local_irq_enable()
-#define hard_irq_disable()	local_irq_disable()
+#define hard_irq_disable()		raw_local_irq_disable()
 
 static inline int irqs_disabled_flags(unsigned long flags)
 {
diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S
index 4dd38f129153..3cadba60a4b6 100644
--- a/arch/powerpc/kernel/entry_32.S
+++ b/arch/powerpc/kernel/entry_32.S
@@ -191,11 +191,49 @@ transfer_to_handler_cont:
 	mflr	r9
 	lwz	r11,0(r9)		/* virtual address of handler */
 	lwz	r9,4(r9)		/* where to go when done */
+#ifdef CONFIG_TRACE_IRQFLAGS
+	lis	r12,reenable_mmu@h
+	ori	r12,r12,reenable_mmu@l
+	mtspr	SPRN_SRR0,r12
+	mtspr	SPRN_SRR1,r10
+	SYNC
+	RFI
+reenable_mmu:				/* re-enable mmu so we can */
+	mfmsr	r10
+	lwz	r12,_MSR(r1)
+	xor	r10,r10,r12
+	andi.	r10,r10,MSR_EE		/* Did EE change? */
+	beq	1f
+
+	/* Save handler and return address into the 2 unused words
+	 * of the STACK_FRAME_OVERHEAD (sneak sneak sneak). Everything
+	 * else can be recovered from the pt_regs except r3 which for
+	 * normal interrupts has been set to pt_regs and for syscalls
+	 * is an argument, so we temporarily use ORIG_GPR3 to save it
+	 */
+	stw	r9,8(r1)
+	stw	r11,12(r1)
+	stw	r3,ORIG_GPR3(r1)
+	bl	trace_hardirqs_off
+	lwz	r0,GPR0(r1)
+	lwz	r3,ORIG_GPR3(r1)
+	lwz	r4,GPR4(r1)
+	lwz	r5,GPR5(r1)
+	lwz	r6,GPR6(r1)
+	lwz	r7,GPR7(r1)
+	lwz	r8,GPR8(r1)
+	lwz	r9,8(r1)
+	lwz	r11,12(r1)
+1:	mtctr	r11
+	mtlr	r9
+	bctr				/* jump to handler */
+#else /* CONFIG_TRACE_IRQFLAGS */
 	mtspr	SPRN_SRR0,r11
 	mtspr	SPRN_SRR1,r10
 	mtlr	r9
 	SYNC
 	RFI				/* jump to handler, enable MMU */
+#endif /* CONFIG_TRACE_IRQFLAGS */
 
 #if defined (CONFIG_6xx) || defined(CONFIG_E500)
 4:	rlwinm	r12,r12,0,~_TLF_NAPPING
@@ -251,6 +289,31 @@ _GLOBAL(DoSyscall)
 #ifdef SHOW_SYSCALLS
 	bl	do_show_syscall
 #endif /* SHOW_SYSCALLS */
+#ifdef CONFIG_TRACE_IRQFLAGS
+	/* Return from syscalls can (and generally will) hard enable
+	 * interrupts. You aren't supposed to call a syscall with
+	 * interrupts disabled in the first place. However, to ensure
+	 * that we get it right vs. lockdep if it happens, we force
+	 * that hard enable here with appropriate tracing if we see
+	 * that we have been called with interrupts off
+	 */
+	mfmsr	r11
+	andi.	r12,r11,MSR_EE
+	bne+	1f
+	/* We came in with interrupts disabled, we enable them now */
+	bl	trace_hardirqs_on
+	mfmsr	r11
+	lwz	r0,GPR0(r1)
+	lwz	r3,GPR3(r1)
+	lwz	r4,GPR4(r1)
+	ori	r11,r11,MSR_EE
+	lwz	r5,GPR5(r1)
+	lwz	r6,GPR6(r1)
+	lwz	r7,GPR7(r1)
+	lwz	r8,GPR8(r1)
+	mtmsr	r11
+1:
+#endif /* CONFIG_TRACE_IRQFLAGS */
 	rlwinm	r10,r1,0,0,(31-THREAD_SHIFT)	/* current_thread_info() */
 	lwz	r11,TI_FLAGS(r10)
 	andi.	r11,r11,_TIF_SYSCALL_T_OR_A
@@ -275,6 +338,7 @@ ret_from_syscall:
 	rlwinm	r12,r1,0,0,(31-THREAD_SHIFT)	/* current_thread_info() */
 	/* disable interrupts so current_thread_info()->flags can't change */
 	LOAD_MSR_KERNEL(r10,MSR_KERNEL)	/* doesn't include MSR_EE */
+	/* Note: We don't bother telling lockdep about it */
 	SYNC
 	MTMSRD(r10)
 	lwz	r9,TI_FLAGS(r12)
@@ -288,6 +352,19 @@ ret_from_syscall:
 	oris	r11,r11,0x1000	/* Set SO bit in CR */
 	stw	r11,_CCR(r1)
 syscall_exit_cont:
+	lwz	r8,_MSR(r1)
+#ifdef CONFIG_TRACE_IRQFLAGS
+	/* If we are going to return from the syscall with interrupts
+	 * off, we trace that here. It shouldn't happen though but we
+	 * want to catch the bugger if it does right ?
+	 */
+	andi.	r10,r8,MSR_EE
+	bne+	1f
+	stw	r3,GPR3(r1)
+	bl      trace_hardirqs_off
+	lwz	r3,GPR3(r1)
+1:
+#endif /* CONFIG_TRACE_IRQFLAGS */
 #if defined(CONFIG_4xx) || defined(CONFIG_BOOKE)
 	/* If the process has its own DBCR0 value, load it up.  The internal
 	   debug mode bit tells us that dbcr0 should be loaded. */
@@ -311,7 +388,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_NEED_PAIRED_STWCX)
 	mtlr	r4
 	mtcr	r5
 	lwz	r7,_NIP(r1)
-	lwz	r8,_MSR(r1)
 	FIX_SRR1(r8, r0)
 	lwz	r2,GPR2(r1)
 	lwz	r1,GPR1(r1)
@@ -394,7 +470,9 @@ syscall_exit_work:
 	andi.	r0,r9,(_TIF_SYSCALL_T_OR_A|_TIF_SINGLESTEP)
 	beq	ret_from_except
 
-	/* Re-enable interrupts */
+	/* Re-enable interrupts. There is no need to trace that with
+	 * lockdep as we are supposed to have IRQs on at this point
+	 */
 	ori	r10,r10,MSR_EE
 	SYNC
 	MTMSRD(r10)
@@ -705,6 +783,7 @@ ret_from_except:
 	/* Hard-disable interrupts so that current_thread_info()->flags
 	 * can't change between when we test it and when we return
 	 * from the interrupt. */
+	/* Note: We don't bother telling lockdep about it */
 	LOAD_MSR_KERNEL(r10,MSR_KERNEL)
 	SYNC			/* Some chip revs have problems here... */
 	MTMSRD(r10)		/* disable interrupts */
@@ -744,11 +823,24 @@ resume_kernel:
 	beq+	restore
 	andi.	r0,r3,MSR_EE	/* interrupts off? */
 	beq	restore		/* don't schedule if so */
+#ifdef CONFIG_TRACE_IRQFLAGS
+	/* Lockdep thinks irqs are enabled, we need to call
+	 * preempt_schedule_irq with IRQs off, so we inform lockdep
+	 * now that we -did- turn them off already
+	 */
+	bl	trace_hardirqs_off
+#endif
 1:	bl	preempt_schedule_irq
 	rlwinm	r9,r1,0,0,(31-THREAD_SHIFT)
 	lwz	r3,TI_FLAGS(r9)
 	andi.	r0,r3,_TIF_NEED_RESCHED
 	bne-	1b
+#ifdef CONFIG_TRACE_IRQFLAGS
+	/* And now, to properly rebalance the above, we tell lockdep they
+	 * are being turned back on, which will happen when we return
+	 */
+	bl	trace_hardirqs_on
+#endif
 #else
 resume_kernel:
 #endif /* CONFIG_PREEMPT */
@@ -765,6 +857,28 @@ restore:
 	stw	r6,icache_44x_need_flush@l(r4)
 1:
 #endif  /* CONFIG_44x */
+
+	lwz	r9,_MSR(r1)
+#ifdef CONFIG_TRACE_IRQFLAGS
+	/* Lockdep doesn't know about the fact that IRQs are temporarily turned
+	 * off in this assembly code while peeking at TI_FLAGS() and such. However
+	 * we need to inform it if the exception turned interrupts off, and we
+	 * are about to trun them back on.
+	 *
+	 * The problem here sadly is that we don't know whether the exceptions was
+	 * one that turned interrupts off or not. So we always tell lockdep about
+	 * turning them on here when we go back to wherever we came from with EE
+	 * on, even if that may meen some redudant calls being tracked. Maybe later
+	 * we could encode what the exception did somewhere or test the exception
+	 * type in the pt_regs but that sounds overkill
+	 */
+	andi.	r10,r9,MSR_EE
+	beq	1f
+	bl	trace_hardirqs_on
+	lwz	r9,_MSR(r1)
+1:
+#endif /* CONFIG_TRACE_IRQFLAGS */
+
 	lwz	r0,GPR0(r1)
 	lwz	r2,GPR2(r1)
 	REST_4GPRS(3, r1)
@@ -782,7 +896,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_NEED_PAIRED_STWCX)
 	stwcx.	r0,0,r1			/* to clear the reservation */
 
 #if !(defined(CONFIG_4xx) || defined(CONFIG_BOOKE))
-	lwz	r9,_MSR(r1)
 	andi.	r10,r9,MSR_RI		/* check if this exception occurred */
 	beql	nonrecoverable		/* at a bad place (MSR:RI = 0) */
 
@@ -805,7 +918,6 @@ END_FTR_SECTION_IFSET(CPU_FTR_NEED_PAIRED_STWCX)
 	MTMSRD(r10)		/* clear the RI bit */
 	.globl exc_exit_restart
 exc_exit_restart:
-	lwz	r9,_MSR(r1)
 	lwz	r12,_NIP(r1)
 	FIX_SRR1(r9,r10)
 	mtspr	SPRN_SRR0,r12
@@ -1035,11 +1147,18 @@ do_work:			/* r10 contains MSR_KERNEL here */
 	beq	do_user_signal
 
 do_resched:			/* r10 contains MSR_KERNEL here */
+	/* Note: We don't need to inform lockdep that we are enabling
+	 * interrupts here. As far as it knows, they are already enabled
+	 */
 	ori	r10,r10,MSR_EE
 	SYNC
 	MTMSRD(r10)		/* hard-enable interrupts */
 	bl	schedule
 recheck:
+	/* Note: And we don't tell it we are disabling them again
+	 * neither. Those disable/enable cycles used to peek at
+	 * TI_FLAGS aren't advertised.
+	 */
 	LOAD_MSR_KERNEL(r10,MSR_KERNEL)
 	SYNC
 	MTMSRD(r10)		/* disable interrupts */
diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c
index 1d154248cf40..e1e3059cf34b 100644
--- a/arch/powerpc/kernel/setup_32.c
+++ b/arch/powerpc/kernel/setup_32.c
@@ -119,6 +119,8 @@ notrace unsigned long __init early_init(unsigned long dt_ptr)
  */
 notrace void __init machine_init(unsigned long dt_ptr)
 {
+	lockdep_init();
+
 	/* Enable early debugging if any specified (see udbg.h) */
 	udbg_early_init();
 
-- 
cgit v1.2.3-59-g8ed1b


From f97bb36f705da0a86b3ea77bfeee3415fee0b025 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 16 Jun 2009 16:42:49 +0000
Subject: powerpc/rtas: Turn rtas lock into a raw spinlock

RTAS currently uses a normal spinlock. However it can be called from
contexts where this is not necessarily a good idea. For example, it
can be called while syncing timebases, with the core timebase being
frozen. Unfortunately, that will deadlock in case of lock contention
when spinlock debugging is enabled as the spin lock debugging code
will try to use __delay() which ... relies on the timebase being
enabled.

Also RTAS can be used in some low level IRQ handling code path so it
may as well be a raw spinlock for -rt sake.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/rtas.h |  2 +-
 arch/powerpc/kernel/rtas.c      | 38 +++++++++++++++++++++++++++++---------
 2 files changed, 30 insertions(+), 10 deletions(-)

diff --git a/arch/powerpc/include/asm/rtas.h b/arch/powerpc/include/asm/rtas.h
index 01c12339b304..0af42d20b692 100644
--- a/arch/powerpc/include/asm/rtas.h
+++ b/arch/powerpc/include/asm/rtas.h
@@ -58,7 +58,7 @@ struct rtas_t {
 	unsigned long entry;		/* physical address pointer */
 	unsigned long base;		/* physical address pointer */
 	unsigned long size;
-	spinlock_t lock;
+	raw_spinlock_t lock;
 	struct rtas_args args;
 	struct device_node *dev;	/* virtual address pointer */
 };
diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c
index ee4c7609b649..d9a9974c6938 100644
--- a/arch/powerpc/kernel/rtas.c
+++ b/arch/powerpc/kernel/rtas.c
@@ -40,7 +40,7 @@
 #include <asm/atomic.h>
 
 struct rtas_t rtas = {
-	.lock = SPIN_LOCK_UNLOCKED
+	.lock = __RAW_SPIN_LOCK_UNLOCKED
 };
 EXPORT_SYMBOL(rtas);
 
@@ -67,6 +67,28 @@ unsigned long rtas_rmo_buf;
 void (*rtas_flash_term_hook)(int);
 EXPORT_SYMBOL(rtas_flash_term_hook);
 
+/* RTAS use home made raw locking instead of spin_lock_irqsave
+ * because those can be called from within really nasty contexts
+ * such as having the timebase stopped which would lockup with
+ * normal locks and spinlock debugging enabled
+ */
+static unsigned long lock_rtas(void)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	preempt_disable();
+	__raw_spin_lock_flags(&rtas.lock, flags);
+	return flags;
+}
+
+static void unlock_rtas(unsigned long flags)
+{
+	__raw_spin_unlock(&rtas.lock);
+	local_irq_restore(flags);
+	preempt_enable();
+}
+
 /*
  * call_rtas_display_status and call_rtas_display_status_delay
  * are designed only for very early low-level debugging, which
@@ -79,7 +101,7 @@ static void call_rtas_display_status(char c)
 
 	if (!rtas.base)
 		return;
-	spin_lock_irqsave(&rtas.lock, s);
+	s = lock_rtas();
 
 	args->token = 10;
 	args->nargs = 1;
@@ -89,7 +111,7 @@ static void call_rtas_display_status(char c)
 
 	enter_rtas(__pa(args));
 
-	spin_unlock_irqrestore(&rtas.lock, s);
+	unlock_rtas(s);
 }
 
 static void call_rtas_display_status_delay(char c)
@@ -411,8 +433,7 @@ int rtas_call(int token, int nargs, int nret, int *outputs, ...)
 	if (!rtas.entry || token == RTAS_UNKNOWN_SERVICE)
 		return -1;
 
-	/* Gotta do something different here, use global lock for now... */
-	spin_lock_irqsave(&rtas.lock, s);
+	s = lock_rtas();
 	rtas_args = &rtas.args;
 
 	rtas_args->token = token;
@@ -439,8 +460,7 @@ int rtas_call(int token, int nargs, int nret, int *outputs, ...)
 			outputs[i] = rtas_args->rets[i+1];
 	ret = (nret > 0)? rtas_args->rets[0]: 0;
 
-	/* Gotta do something different here, use global lock for now... */
-	spin_unlock_irqrestore(&rtas.lock, s);
+	unlock_rtas(s);
 
 	if (buff_copy) {
 		log_error(buff_copy, ERR_TYPE_RTAS_LOG, 0);
@@ -837,7 +857,7 @@ asmlinkage int ppc_rtas(struct rtas_args __user *uargs)
 
 	buff_copy = get_errorlog_buffer();
 
-	spin_lock_irqsave(&rtas.lock, flags);
+	flags = lock_rtas();
 
 	rtas.args = args;
 	enter_rtas(__pa(&rtas.args));
@@ -848,7 +868,7 @@ asmlinkage int ppc_rtas(struct rtas_args __user *uargs)
 	if (args.rets[0] == -1)
 		errbuf = __fetch_rtas_last_error(buff_copy);
 
-	spin_unlock_irqrestore(&rtas.lock, flags);
+	unlock_rtas(flags);
 
 	if (buff_copy) {
 		if (errbuf)
-- 
cgit v1.2.3-59-g8ed1b


From c4007a2fbf5f82b7e694c22b5929c87e38415a56 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 16 Jun 2009 16:42:50 +0000
Subject: powerpc: Use one common impl. of RTAS timebase sync and use raw
 spinlock

Several platforms use their own copy of what is essentially the same code,
using RTAS to synchronize the timebases when bringing up new CPUs. This
moves it all into a single common implementation and additionally
turns the spinlock into a raw spinlock since the former can rely on
the timebase not being frozen when spinlock debugging is enabled, and finally
masks interrupts while the timebase is disabled.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/rtas.h      |  3 +++
 arch/powerpc/kernel/rtas.c           | 31 +++++++++++++++++++++++++++++++
 arch/powerpc/platforms/cell/smp.c    | 30 ++----------------------------
 arch/powerpc/platforms/chrp/smp.c    | 33 ++-------------------------------
 arch/powerpc/platforms/pseries/smp.c | 30 ++----------------------------
 5 files changed, 40 insertions(+), 87 deletions(-)

diff --git a/arch/powerpc/include/asm/rtas.h b/arch/powerpc/include/asm/rtas.h
index 0af42d20b692..168fce726201 100644
--- a/arch/powerpc/include/asm/rtas.h
+++ b/arch/powerpc/include/asm/rtas.h
@@ -245,5 +245,8 @@ static inline u32 rtas_config_addr(int busno, int devfn, int reg)
 			(devfn << 8) | (reg & 0xff);
 }
 
+extern void __cpuinit rtas_give_timebase(void);
+extern void __cpuinit rtas_take_timebase(void);
+
 #endif /* __KERNEL__ */
 #endif /* _POWERPC_RTAS_H */
diff --git a/arch/powerpc/kernel/rtas.c b/arch/powerpc/kernel/rtas.c
index d9a9974c6938..c434823b8c83 100644
--- a/arch/powerpc/kernel/rtas.c
+++ b/arch/powerpc/kernel/rtas.c
@@ -38,6 +38,7 @@
 #include <asm/syscalls.h>
 #include <asm/smp.h>
 #include <asm/atomic.h>
+#include <asm/time.h>
 
 struct rtas_t rtas = {
 	.lock = __RAW_SPIN_LOCK_UNLOCKED
@@ -971,3 +972,33 @@ int __init early_init_dt_scan_rtas(unsigned long node,
 	/* break now */
 	return 1;
 }
+
+static raw_spinlock_t timebase_lock;
+static u64 timebase = 0;
+
+void __cpuinit rtas_give_timebase(void)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	hard_irq_disable();
+	__raw_spin_lock(&timebase_lock);
+	rtas_call(rtas_token("freeze-time-base"), 0, 1, NULL);
+	timebase = get_tb();
+	__raw_spin_unlock(&timebase_lock);
+
+	while (timebase)
+		barrier();
+	rtas_call(rtas_token("thaw-time-base"), 0, 1, NULL);
+	local_irq_restore(flags);
+}
+
+void __cpuinit rtas_take_timebase(void)
+{
+	while (!timebase)
+		barrier();
+	__raw_spin_lock(&timebase_lock);
+	set_tb(timebase >> 32, timebase & 0xffffffff);
+	timebase = 0;
+	__raw_spin_unlock(&timebase_lock);
+}
diff --git a/arch/powerpc/platforms/cell/smp.c b/arch/powerpc/platforms/cell/smp.c
index 9046803c8276..bc97fada48c6 100644
--- a/arch/powerpc/platforms/cell/smp.c
+++ b/arch/powerpc/platforms/cell/smp.c
@@ -36,7 +36,6 @@
 #include <asm/prom.h>
 #include <asm/smp.h>
 #include <asm/paca.h>
-#include <asm/time.h>
 #include <asm/machdep.h>
 #include <asm/cputable.h>
 #include <asm/firmware.h>
@@ -140,31 +139,6 @@ static void __devinit smp_cell_setup_cpu(int cpu)
 	mtspr(SPRN_DABRX, DABRX_KERNEL | DABRX_USER);
 }
 
-static DEFINE_SPINLOCK(timebase_lock);
-static unsigned long timebase = 0;
-
-static void __devinit cell_give_timebase(void)
-{
-	spin_lock(&timebase_lock);
-	rtas_call(rtas_token("freeze-time-base"), 0, 1, NULL);
-	timebase = get_tb();
-	spin_unlock(&timebase_lock);
-
-	while (timebase)
-		barrier();
-	rtas_call(rtas_token("thaw-time-base"), 0, 1, NULL);
-}
-
-static void __devinit cell_take_timebase(void)
-{
-	while (!timebase)
-		barrier();
-	spin_lock(&timebase_lock);
-	set_tb(timebase >> 32, timebase & 0xffffffff);
-	timebase = 0;
-	spin_unlock(&timebase_lock);
-}
-
 static void __devinit smp_cell_kick_cpu(int nr)
 {
 	BUG_ON(nr < 0 || nr >= NR_CPUS);
@@ -224,8 +198,8 @@ void __init smp_init_cell(void)
 
 	/* Non-lpar has additional take/give timebase */
 	if (rtas_token("freeze-time-base") != RTAS_UNKNOWN_SERVICE) {
-		smp_ops->give_timebase = cell_give_timebase;
-		smp_ops->take_timebase = cell_take_timebase;
+		smp_ops->give_timebase = rtas_give_timebase;
+		smp_ops->take_timebase = rtas_take_timebase;
 	}
 
 	DBG(" <- smp_init_cell()\n");
diff --git a/arch/powerpc/platforms/chrp/smp.c b/arch/powerpc/platforms/chrp/smp.c
index 10a4a4d063b6..02cafecc90e3 100644
--- a/arch/powerpc/platforms/chrp/smp.c
+++ b/arch/powerpc/platforms/chrp/smp.c
@@ -26,7 +26,6 @@
 #include <asm/io.h>
 #include <asm/prom.h>
 #include <asm/smp.h>
-#include <asm/time.h>
 #include <asm/machdep.h>
 #include <asm/mpic.h>
 #include <asm/rtas.h>
@@ -42,40 +41,12 @@ static void __devinit smp_chrp_setup_cpu(int cpu_nr)
 	mpic_setup_this_cpu();
 }
 
-static DEFINE_SPINLOCK(timebase_lock);
-static unsigned int timebase_upper = 0, timebase_lower = 0;
-
-void __devinit smp_chrp_give_timebase(void)
-{
-	spin_lock(&timebase_lock);
-	rtas_call(rtas_token("freeze-time-base"), 0, 1, NULL);
-	timebase_upper = get_tbu();
-	timebase_lower = get_tbl();
-	spin_unlock(&timebase_lock);
-
-	while (timebase_upper || timebase_lower)
-		barrier();
-	rtas_call(rtas_token("thaw-time-base"), 0, 1, NULL);
-}
-
-void __devinit smp_chrp_take_timebase(void)
-{
-	while (!(timebase_upper || timebase_lower))
-		barrier();
-	spin_lock(&timebase_lock);
-	set_tb(timebase_upper, timebase_lower);
-	timebase_upper = 0;
-	timebase_lower = 0;
-	spin_unlock(&timebase_lock);
-	printk("CPU %i taken timebase\n", smp_processor_id());
-}
-
 /* CHRP with openpic */
 struct smp_ops_t chrp_smp_ops = {
 	.message_pass = smp_mpic_message_pass,
 	.probe = smp_mpic_probe,
 	.kick_cpu = smp_chrp_kick_cpu,
 	.setup_cpu = smp_chrp_setup_cpu,
-	.give_timebase = smp_chrp_give_timebase,
-	.take_timebase = smp_chrp_take_timebase,
+	.give_timebase = rtas_give_timebase,
+	.take_timebase = rtas_take_timebase,
 };
diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c
index 1a231c389ba0..1f8f6cfb94f7 100644
--- a/arch/powerpc/platforms/pseries/smp.c
+++ b/arch/powerpc/platforms/pseries/smp.c
@@ -35,7 +35,6 @@
 #include <asm/prom.h>
 #include <asm/smp.h>
 #include <asm/paca.h>
-#include <asm/time.h>
 #include <asm/machdep.h>
 #include <asm/cputable.h>
 #include <asm/firmware.h>
@@ -118,31 +117,6 @@ static void __devinit smp_xics_setup_cpu(int cpu)
 }
 #endif /* CONFIG_XICS */
 
-static DEFINE_SPINLOCK(timebase_lock);
-static unsigned long timebase = 0;
-
-static void __devinit pSeries_give_timebase(void)
-{
-	spin_lock(&timebase_lock);
-	rtas_call(rtas_token("freeze-time-base"), 0, 1, NULL);
-	timebase = get_tb();
-	spin_unlock(&timebase_lock);
-
-	while (timebase)
-		barrier();
-	rtas_call(rtas_token("thaw-time-base"), 0, 1, NULL);
-}
-
-static void __devinit pSeries_take_timebase(void)
-{
-	while (!timebase)
-		barrier();
-	spin_lock(&timebase_lock);
-	set_tb(timebase >> 32, timebase & 0xffffffff);
-	timebase = 0;
-	spin_unlock(&timebase_lock);
-}
-
 static void __devinit smp_pSeries_kick_cpu(int nr)
 {
 	BUG_ON(nr < 0 || nr >= NR_CPUS);
@@ -209,8 +183,8 @@ static void __init smp_init_pseries(void)
 
 	/* Non-lpar has additional take/give timebase */
 	if (rtas_token("freeze-time-base") != RTAS_UNKNOWN_SERVICE) {
-		smp_ops->give_timebase = pSeries_give_timebase;
-		smp_ops->take_timebase = pSeries_take_timebase;
+		smp_ops->give_timebase = rtas_give_timebase;
+		smp_ops->take_timebase = rtas_take_timebase;
 	}
 
 	pr_debug(" <- smp_init_pSeries()\n");
-- 
cgit v1.2.3-59-g8ed1b


From 6893ce6c1cdcf489b7ca8e6b6596208aa971a083 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 16 Jun 2009 16:42:51 +0000
Subject: powerpc/pasemi: Use raw spinlock in SMP TB sync

spin_lock() can hang if called while the timebase is frozen,
so use a raw lock instead, also disable interrupts while
at it.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/platforms/pasemi/setup.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/arch/powerpc/platforms/pasemi/setup.c b/arch/powerpc/platforms/pasemi/setup.c
index 153051eb6d93..a4619347aa7e 100644
--- a/arch/powerpc/platforms/pasemi/setup.c
+++ b/arch/powerpc/platforms/pasemi/setup.c
@@ -71,20 +71,25 @@ static void pas_restart(char *cmd)
 }
 
 #ifdef CONFIG_SMP
-static DEFINE_SPINLOCK(timebase_lock);
+static raw_spinlock_t timebase_lock;
 static unsigned long timebase;
 
 static void __devinit pas_give_timebase(void)
 {
-	spin_lock(&timebase_lock);
+	unsigned long flags;
+
+	local_irq_save(flags);
+	hard_irq_disable();
+	__raw_spin_lock(&timebase_lock);
 	mtspr(SPRN_TBCTL, TBCTL_FREEZE);
 	isync();
 	timebase = get_tb();
-	spin_unlock(&timebase_lock);
+	__raw_spin_unlock(&timebase_lock);
 
 	while (timebase)
 		barrier();
 	mtspr(SPRN_TBCTL, TBCTL_RESTART);
+	local_irq_restore(flags);
 }
 
 static void __devinit pas_take_timebase(void)
@@ -92,10 +97,10 @@ static void __devinit pas_take_timebase(void)
 	while (!timebase)
 		smp_rmb();
 
-	spin_lock(&timebase_lock);
+	__raw_spin_lock(&timebase_lock);
 	set_tb(timebase >> 32, timebase & 0xffffffff);
 	timebase = 0;
-	spin_unlock(&timebase_lock);
+	__raw_spin_unlock(&timebase_lock);
 }
 
 struct smp_ops_t pas_smp_ops = {
-- 
cgit v1.2.3-59-g8ed1b


From 03c01aa740d4137ea2884328f2c29956c6084834 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 16 Jun 2009 15:55:18 +0000
Subject: powerpc/of: Fix usage of dev_set_name() in of_device_alloc()

dev_set_name() takes a format string, so use it properly and avoid
a warning with recent gcc's

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/of_device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/powerpc/kernel/of_device.c b/arch/powerpc/kernel/of_device.c
index fa983a59c4ce..a359cb08e900 100644
--- a/arch/powerpc/kernel/of_device.c
+++ b/arch/powerpc/kernel/of_device.c
@@ -76,7 +76,7 @@ struct of_device *of_device_alloc(struct device_node *np,
 	dev->dev.archdata.of_node = np;
 
 	if (bus_id)
-		dev_set_name(&dev->dev, bus_id);
+		dev_set_name(&dev->dev, "%s", bus_id);
 	else
 		of_device_make_bus_id(dev);
 
-- 
cgit v1.2.3-59-g8ed1b


From f694cda89250f38782a5fd0b2d32fe33a39dcf37 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Tue, 16 Jun 2009 15:55:19 +0000
Subject: powerpc/440: Fix warning early debug code

The function udbg_44x_as1_flush() has the wrong prototype causing
a warning when enabling 440 early debug.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/udbg_16550.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/powerpc/kernel/udbg_16550.c b/arch/powerpc/kernel/udbg_16550.c
index 0362a891e54e..acb74a17bbbf 100644
--- a/arch/powerpc/kernel/udbg_16550.c
+++ b/arch/powerpc/kernel/udbg_16550.c
@@ -219,7 +219,7 @@ void udbg_init_pas_realmode(void)
 #ifdef CONFIG_PPC_EARLY_DEBUG_44x
 #include <platforms/44x/44x.h>
 
-static int udbg_44x_as1_flush(void)
+static void udbg_44x_as1_flush(void)
 {
 	if (udbg_comport) {
 		while ((as1_readb(&udbg_comport->lsr) & LSR_THRE) == 0)
-- 
cgit v1.2.3-59-g8ed1b


From 6c16a74d423f584ed80815ee7b944f5b578dd37a Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Mon, 15 Jun 2009 16:53:43 +0000
Subject: powerpc/mm: Fix potential access to freed pages when using hugetlbfs

When using 64k page sizes, our PTE pages are split in two halves,
the second half containing the "extension" used to keep track of
individual 4k pages when not using HW 64k pages.

However, our page tables used for hugetlb have a slightly different
format and don't carry that "second half".

Our code that batched PTEs to be invalidated unconditionally reads
the "second half" (to put it into the batch), which means that when
called to invalidate hugetlb PTEs, it will access unrelated memory.

It breaks when CONFIG_DEBUG_PAGEALLOC is enabled.

This fixes it by only accessing the second half when the _PAGE_COMBO
bit is set in the first half, which indicates that we are dealing with
a "combo" page which represents 16x4k subpages. Anything else shouldn't
have this bit set and thus not require loading from the second half.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/include/asm/pte-hash64-64k.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/include/asm/pte-hash64-64k.h b/arch/powerpc/include/asm/pte-hash64-64k.h
index e05d26fa372f..82b72207c51c 100644
--- a/arch/powerpc/include/asm/pte-hash64-64k.h
+++ b/arch/powerpc/include/asm/pte-hash64-64k.h
@@ -47,7 +47,8 @@
  * generic accessors and iterators here
  */
 #define __real_pte(e,p) 	((real_pte_t) { \
-	(e), pte_val(*((p) + PTRS_PER_PTE)) })
+			(e), ((e) & _PAGE_COMBO) ? \
+				(pte_val(*((p) + PTRS_PER_PTE))) : 0 })
 #define __rpte_to_hidx(r,index)	((pte_val((r).pte) & _PAGE_COMBO) ? \
         (((r).hidx >> ((index)<<2)) & 0xf) : ((pte_val((r).pte) >> 12) & 0xf))
 #define __rpte_to_pte(r)	((r).pte)
-- 
cgit v1.2.3-59-g8ed1b


From 5ba762c9bb3ce2cc11e9e111cb3c476e84b91668 Mon Sep 17 00:00:00 2001
From: Adrian Reber <adrian@lisas.de>
Date: Thu, 26 Mar 2009 02:05:42 +0000
Subject: powerpc/rtas: Fix watchdog driver temperature read functionality

Using the RTAS watchdog driver to read out the temperature crashes
on a PXCAB:

Unable to handle kernel paging request for data at address 0xfe347b50
Faulting instruction address: 0xc00000000001af64
Oops: Kernel access of bad area, sig: 11 [#1]

The wrong usage of "(void *)__pa(&temperature)" in rtas_call() is
removed by using the function rtas_get_sensor() which does the
right thing.

Signed-off-by: Adrian Reber <adrian@lisas.de>
Acked-by: Utz Bacher <utz.bacher@de.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 drivers/watchdog/wdrtas.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/watchdog/wdrtas.c b/drivers/watchdog/wdrtas.c
index a4fe7a38d9b0..3bde56bce63a 100644
--- a/drivers/watchdog/wdrtas.c
+++ b/drivers/watchdog/wdrtas.c
@@ -218,16 +218,14 @@ static void wdrtas_timer_keepalive(void)
  */
 static int wdrtas_get_temperature(void)
 {
-	long result;
+	int result;
 	int temperature = 0;
 
-	result = rtas_call(wdrtas_token_get_sensor_state, 2, 2,
-			   (void *)__pa(&temperature),
-			   WDRTAS_THERMAL_SENSOR, 0);
+	result = rtas_get_sensor(WDRTAS_THERMAL_SENSOR, 0, &temperature);
 
 	if (result < 0)
 		printk(KERN_WARNING "wdrtas: reading the thermal sensor "
-		       "faild: %li\n", result);
+		       "failed: %i\n", result);
 	else
 		temperature = ((temperature * 9) / 5) + 32; /* fahrenheit */
 
-- 
cgit v1.2.3-59-g8ed1b


From 789547508f22e482825f52f813b59680408ec2c7 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 23 Jun 2009 11:26:06 +0000
Subject: ide: fix ide_kill_rq() for special ide-{floppy,tape} driver requests

Such requests should be failed with -EIO (like all other requests
in this function) instead of being completed successfully.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/ide-io.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 93b7886a2d6e..95db5f03f6a2 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -152,7 +152,7 @@ void ide_kill_rq(ide_drive_t *drive, struct request *rq)
 
 	if ((media == ide_floppy || media == ide_tape) && drv_req) {
 		rq->errors = 0;
-		ide_complete_rq(drive, 0, blk_rq_bytes(rq));
+		ide_complete_rq(drive, -EIO, blk_rq_bytes(rq));
 	} else {
 		if (media == ide_tape)
 			rq->errors = IDE_DRV_ERROR_GENERAL;
-- 
cgit v1.2.3-59-g8ed1b


From 5e955245d6cf49c5ed26c7add7392ff5a6762bf4 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Tue, 23 Jun 2009 11:27:27 +0000
Subject: ide: always kill the whole request on error

* Use blk_rq_bytes() instead of obsolete ide_rq_bytes() in ide_kill_rq()
  and ide_floppy_do_request() for failed requests.
  [ bugfix part ]

* Use blk_rq_bytes() instead of obsolete ide_rq_bytes() in ide_do_devset()
  and ide_complete_drive_reset().  Then remove ide_rq_bytes().
  [ cleanup part ]

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/ide-devsets.c |  2 +-
 drivers/ide/ide-eh.c      |  2 +-
 drivers/ide/ide-floppy.c  |  2 +-
 drivers/ide/ide-io.c      | 14 ++------------
 include/linux/ide.h       |  1 -
 5 files changed, 5 insertions(+), 16 deletions(-)

diff --git a/drivers/ide/ide-devsets.c b/drivers/ide/ide-devsets.c
index 5bf958e5b1d5..1099bf7cf968 100644
--- a/drivers/ide/ide-devsets.c
+++ b/drivers/ide/ide-devsets.c
@@ -183,6 +183,6 @@ ide_startstop_t ide_do_devset(ide_drive_t *drive, struct request *rq)
 	err = setfunc(drive, *(int *)&rq->cmd[1]);
 	if (err)
 		rq->errors = err;
-	ide_complete_rq(drive, err, ide_rq_bytes(rq));
+	ide_complete_rq(drive, err, blk_rq_bytes(rq));
 	return ide_stopped;
 }
diff --git a/drivers/ide/ide-eh.c b/drivers/ide/ide-eh.c
index 2b9141979613..e9abf2c3c335 100644
--- a/drivers/ide/ide-eh.c
+++ b/drivers/ide/ide-eh.c
@@ -149,7 +149,7 @@ static inline void ide_complete_drive_reset(ide_drive_t *drive, int err)
 	if (rq && blk_special_request(rq) && rq->cmd[0] == REQ_DRIVE_RESET) {
 		if (err <= 0 && rq->errors == 0)
 			rq->errors = -EIO;
-		ide_complete_rq(drive, err ? err : 0, ide_rq_bytes(rq));
+		ide_complete_rq(drive, err ? err : 0, blk_rq_bytes(rq));
 	}
 }
 
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 8b3f204f7d73..fefbdfc8db06 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -293,7 +293,7 @@ out_end:
 	drive->failed_pc = NULL;
 	if (blk_fs_request(rq) == 0 && rq->errors == 0)
 		rq->errors = -EIO;
-	ide_complete_rq(drive, -EIO, ide_rq_bytes(rq));
+	ide_complete_rq(drive, -EIO, blk_rq_bytes(rq));
 	return ide_stopped;
 }
 
diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
index 95db5f03f6a2..d5f3c77beadd 100644
--- a/drivers/ide/ide-io.c
+++ b/drivers/ide/ide-io.c
@@ -112,16 +112,6 @@ void ide_complete_cmd(ide_drive_t *drive, struct ide_cmd *cmd, u8 stat, u8 err)
 	}
 }
 
-/* obsolete, blk_rq_bytes() should be used instead */
-unsigned int ide_rq_bytes(struct request *rq)
-{
-	if (blk_pc_request(rq))
-		return blk_rq_bytes(rq);
-	else
-		return blk_rq_cur_sectors(rq) << 9;
-}
-EXPORT_SYMBOL_GPL(ide_rq_bytes);
-
 int ide_complete_rq(ide_drive_t *drive, int error, unsigned int nr_bytes)
 {
 	ide_hwif_t *hwif = drive->hwif;
@@ -152,14 +142,14 @@ void ide_kill_rq(ide_drive_t *drive, struct request *rq)
 
 	if ((media == ide_floppy || media == ide_tape) && drv_req) {
 		rq->errors = 0;
-		ide_complete_rq(drive, -EIO, blk_rq_bytes(rq));
 	} else {
 		if (media == ide_tape)
 			rq->errors = IDE_DRV_ERROR_GENERAL;
 		else if (blk_fs_request(rq) == 0 && rq->errors == 0)
 			rq->errors = -EIO;
-		ide_complete_rq(drive, -EIO, ide_rq_bytes(rq));
 	}
+
+	ide_complete_rq(drive, -EIO, blk_rq_bytes(rq));
 }
 
 static void ide_tf_set_specify_cmd(ide_drive_t *drive, struct ide_taskfile *tf)
diff --git a/include/linux/ide.h b/include/linux/ide.h
index cf1f3888067c..c6af7c44d46c 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1062,7 +1062,6 @@ int generic_ide_ioctl(ide_drive_t *, struct block_device *, unsigned, unsigned l
 extern int ide_vlb_clk;
 extern int ide_pci_clk;
 
-unsigned int ide_rq_bytes(struct request *);
 int ide_end_rq(ide_drive_t *, struct request *, int, unsigned int);
 void ide_kill_rq(ide_drive_t *, struct request *);
 
-- 
cgit v1.2.3-59-g8ed1b


From a80cad950f2a562e60db1869dd29bc007c5a4b66 Mon Sep 17 00:00:00 2001
From: Kuninori Morimoto <morimoto.kuninori@renesas.com>
Date: Fri, 26 Jun 2009 07:05:39 +0000
Subject: sh: ms7724se: Add sh_eth support

Signed-off-by: Kuninori Morimoto <morimoto.kuninori@renesas.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/mach-se/7724/setup.c   | 106 +++++++++++++++++++++++++++++++++-
 arch/sh/include/mach-se/mach/se7724.h |   5 ++
 2 files changed, 109 insertions(+), 2 deletions(-)

diff --git a/arch/sh/boards/mach-se/7724/setup.c b/arch/sh/boards/mach-se/7724/setup.c
index 9cd04bd558b8..21d18005fb4a 100644
--- a/arch/sh/boards/mach-se/7724/setup.c
+++ b/arch/sh/boards/mach-se/7724/setup.c
@@ -23,6 +23,8 @@
 #include <media/sh_mobile_ceu.h>
 #include <asm/io.h>
 #include <asm/heartbeat.h>
+#include <asm/sh_eth.h>
+#include <asm/clock.h>
 #include <asm/sh_keysc.h>
 #include <cpu/sh7724.h>
 #include <mach-se/mach/se7724.h>
@@ -272,6 +274,34 @@ static struct platform_device keysc_device = {
 	},
 };
 
+/* SH Eth */
+static struct resource sh_eth_resources[] = {
+	[0] = {
+		.start = SH_ETH_ADDR,
+		.end   = SH_ETH_ADDR + 0x1FC,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = 91,
+		.flags = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHLEVEL,
+	},
+};
+
+struct sh_eth_plat_data sh_eth_plat = {
+	.phy = 0x1f, /* SMSC LAN8187 */
+	.edmac_endian = EDMAC_LITTLE_ENDIAN,
+};
+
+static struct platform_device sh_eth_device = {
+	.name = "sh-eth",
+	.id	= 0,
+	.dev = {
+		.platform_data = &sh_eth_plat,
+	},
+	.num_resources = ARRAY_SIZE(sh_eth_resources),
+	.resource = sh_eth_resources,
+};
+
 static struct platform_device *ms7724se_devices[] __initdata = {
 	&heartbeat_device,
 	&smc91x_eth_device,
@@ -280,8 +310,57 @@ static struct platform_device *ms7724se_devices[] __initdata = {
 	&ceu0_device,
 	&ceu1_device,
 	&keysc_device,
+	&sh_eth_device,
 };
 
+#define EEPROM_OP   0xBA206000
+#define EEPROM_ADR  0xBA206004
+#define EEPROM_DATA 0xBA20600C
+#define EEPROM_STAT 0xBA206010
+#define EEPROM_STRT 0xBA206014
+static int __init sh_eth_is_eeprom_ready(void)
+{
+	int t = 10000;
+
+	while (t--) {
+		if (!ctrl_inw(EEPROM_STAT))
+			return 1;
+		cpu_relax();
+	}
+
+	printk(KERN_ERR "ms7724se can not access to eeprom\n");
+	return 0;
+}
+
+static void __init sh_eth_init(void)
+{
+	int i;
+	u16 mac[3];
+
+	/* check EEPROM status */
+	if (!sh_eth_is_eeprom_ready())
+		return;
+
+	/* read MAC addr from EEPROM */
+	for (i = 0 ; i < 3 ; i++) {
+		ctrl_outw(0x0, EEPROM_OP); /* read */
+		ctrl_outw(i*2, EEPROM_ADR);
+		ctrl_outw(0x1, EEPROM_STRT);
+		if (!sh_eth_is_eeprom_ready())
+			return;
+
+		mac[i] = ctrl_inw(EEPROM_DATA);
+		mac[i] = ((mac[i] & 0xFF) << 8) | (mac[i] >> 8); /* swap */
+	}
+
+	/* reset sh-eth */
+	ctrl_outl(0x1, SH_ETH_ADDR + 0x0);
+
+	/* set MAC addr */
+	ctrl_outl(((mac[0] << 16) | (mac[1])), SH_ETH_MAHR);
+	ctrl_outl((mac[2]), SH_ETH_MALR);
+}
+
 #define SW4140    0xBA201000
 #define FPGA_OUT  0xBA200400
 #define PORT_HIZA 0xA4050158
@@ -302,7 +381,8 @@ static int __init devices_setup(void)
 	ctrl_outw(ctrl_inw(FPGA_OUT) &
 		  ~((1 << 1)  | /* LAN */
 		    (1 << 6)  | /* VIDEO DAC */
-		    (1 << 12)), /* USB0 */
+		    (1 << 12) | /* USB0 */
+		    (1 << 14)), /* RMII */
 		  FPGA_OUT);
 
 	/* enable IRQ 0,1,2 */
@@ -404,6 +484,28 @@ static int __init devices_setup(void)
 	gpio_request(GPIO_FN_KEYOUT1,     NULL);
 	gpio_request(GPIO_FN_KEYOUT0,     NULL);
 
+	/*
+	 * enable SH-Eth
+	 *
+	 * please remove J33 pin from your board !!
+	 *
+	 * ms7724 board should not use GPIO_FN_LNKSTA pin
+	 * So, This time PTX5 is set to input pin
+	 */
+	gpio_request(GPIO_FN_RMII_RXD0,    NULL);
+	gpio_request(GPIO_FN_RMII_RXD1,    NULL);
+	gpio_request(GPIO_FN_RMII_TXD0,    NULL);
+	gpio_request(GPIO_FN_RMII_TXD1,    NULL);
+	gpio_request(GPIO_FN_RMII_REF_CLK, NULL);
+	gpio_request(GPIO_FN_RMII_TX_EN,   NULL);
+	gpio_request(GPIO_FN_RMII_RX_ER,   NULL);
+	gpio_request(GPIO_FN_RMII_CRS_DV,  NULL);
+	gpio_request(GPIO_FN_MDIO,         NULL);
+	gpio_request(GPIO_FN_MDC,          NULL);
+	gpio_request(GPIO_PTX5, NULL);
+	gpio_direction_input(GPIO_PTX5);
+	sh_eth_init();
+
 	if (sw & SW41_B) {
 		/* SVGA */
 		lcdc_info.ch[0].lcd_cfg.xres         = 800;
@@ -437,7 +539,7 @@ static int __init devices_setup(void)
 	}
 
 	return platform_add_devices(ms7724se_devices,
-				ARRAY_SIZE(ms7724se_devices));
+				    ARRAY_SIZE(ms7724se_devices));
 }
 device_initcall(devices_setup);
 
diff --git a/arch/sh/include/mach-se/mach/se7724.h b/arch/sh/include/mach-se/mach/se7724.h
index 74164b60d0db..29514a39d0f5 100644
--- a/arch/sh/include/mach-se/mach/se7724.h
+++ b/arch/sh/include/mach-se/mach/se7724.h
@@ -20,6 +20,11 @@
  */
 #include <asm/addrspace.h>
 
+/* SH Eth */
+#define SH_ETH_ADDR	(0xA4600000)
+#define SH_ETH_MAHR	(SH_ETH_ADDR + 0x1C0)
+#define SH_ETH_MALR	(SH_ETH_ADDR + 0x1C8)
+
 #define PA_LED		(0xba203000)	/* 8bit LED */
 #define IRQ_MODE	(0xba200010)
 #define IRQ0_SR		(0xba200014)
-- 
cgit v1.2.3-59-g8ed1b


From 0296e4254f3318e0dcad9706fa1daf8e5addc1e9 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 26 Jun 2009 11:15:37 +0800
Subject: ftrace: Fix the output of profile

The first entry of the ftrace profile was always skipped when
reading trace_stat/functionX.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A443D59.4080307@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/ftrace.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 71a52c172140..f3716bf04df6 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -291,7 +291,9 @@ function_stat_next(void *v, int idx)
 	pg = (struct ftrace_profile_page *)((unsigned long)rec & PAGE_MASK);
 
  again:
-	rec++;
+	if (idx != 0)
+		rec++;
+
 	if ((void *)rec >= (void *)&pg->records[pg->index]) {
 		pg = pg->next;
 		if (!pg)
-- 
cgit v1.2.3-59-g8ed1b


From 7e25a2422987a37729706b18583d177966919d2a Mon Sep 17 00:00:00 2001
From: Chris Wright <chrisw@redhat.com>
Date: Thu, 25 Jun 2009 18:52:05 -0700
Subject: intel-iommu: fix Identity Mapping to be arch independent

Drop the e820 scanning and use existing function for finding valid
RAM regions to add to 1:1 mapping.

Signed-off-by: Chris Wright <chrisw@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 19 +++++--------------
 1 file changed, 5 insertions(+), 14 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index e53eacd75c8d..420afa887283 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -39,7 +39,6 @@
 #include <linux/sysdev.h>
 #include <asm/cacheflush.h>
 #include <asm/iommu.h>
-#include <asm/e820.h>
 #include "pci.h"
 
 #define ROOT_SIZE		VTD_PAGE_SIZE
@@ -1908,7 +1907,6 @@ static inline int iommu_prepare_rmrr_dev(struct dmar_rmrr_unit *rmrr,
 		rmrr->end_address + 1);
 }
 
-#ifdef CONFIG_DMAR_GFX_WA
 struct iommu_prepare_data {
 	struct pci_dev *pdev;
 	int ret;
@@ -1943,6 +1941,7 @@ static int __init iommu_prepare_with_active_regions(struct pci_dev *pdev)
 	return data.ret;
 }
 
+#ifdef CONFIG_DMAR_GFX_WA
 static void __init iommu_prepare_gfx_mapping(void)
 {
 	struct pci_dev *pdev = NULL;
@@ -2081,7 +2080,6 @@ static int domain_add_dev_info(struct dmar_domain *domain,
 
 static int iommu_prepare_static_identity_mapping(void)
 {
-	int i;
 	struct pci_dev *pdev = NULL;
 	int ret;
 
@@ -2091,17 +2089,10 @@ static int iommu_prepare_static_identity_mapping(void)
 
 	printk(KERN_INFO "IOMMU: Setting identity map:\n");
 	for_each_pci_dev(pdev) {
-		for (i = 0; i < e820.nr_map; i++) {
-			struct e820entry *ei = &e820.map[i];
-
-			if (ei->type == E820_RAM) {
-				ret = iommu_prepare_identity_map(pdev,
-					ei->addr, ei->addr + ei->size);
-				if (ret)  {
-					printk(KERN_INFO "1:1 mapping to one domain failed.\n");
-					return -EFAULT;
-				}
-			}
+		ret = iommu_prepare_with_active_regions(pdev);
+		if (ret) {
+			printk(KERN_INFO "1:1 mapping to one domain failed.\n");
+			return -EFAULT;
 		}
 		ret = domain_add_dev_info(si_domain, pdev);
 		if (ret)
-- 
cgit v1.2.3-59-g8ed1b


From 584fcff428bde3b9985ba21498764e9dba2fd3ce Mon Sep 17 00:00:00 2001
From: Borislav Petkov <borislav.petkov@amd.com>
Date: Wed, 10 Jun 2009 18:29:54 +0200
Subject: amd64_edac: check only ECC bit in amd64_determine_edac_cap

Checking whether the machine is using ECC enabled DRAM is done through
testing the DimmEccEn bit in the DRAM Cfg Low register (F2x[1,0]90). Do
that instead of testing all bits from the DimmEccEn upwards.

Also, remove mci->edac_cap assignment and use value returned from
amd64_determine_edac_cap().

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
---
 drivers/edac/amd64_edac.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c
index c36bf40568cf..3b76605330e9 100644
--- a/drivers/edac/amd64_edac.c
+++ b/drivers/edac/amd64_edac.c
@@ -754,13 +754,13 @@ static void amd64_cpu_display_info(struct amd64_pvt *pvt)
 static enum edac_type amd64_determine_edac_cap(struct amd64_pvt *pvt)
 {
 	int bit;
-	enum dev_type edac_cap = EDAC_NONE;
+	enum dev_type edac_cap = EDAC_FLAG_NONE;
 
 	bit = (boot_cpu_data.x86 > 0xf || pvt->ext_model >= OPTERON_CPU_REV_F)
 		? 19
 		: 17;
 
-	if (pvt->dclr0 >> BIT(bit))
+	if (pvt->dclr0 & BIT(bit))
 		edac_cap = EDAC_FLAG_SECDED;
 
 	return edac_cap;
@@ -3006,7 +3006,6 @@ static void amd64_setup_mci_misc_attributes(struct mem_ctl_info *mci)
 
 	mci->mtype_cap		= MEM_FLAG_DDR2 | MEM_FLAG_RDDR2;
 	mci->edac_ctl_cap	= EDAC_FLAG_NONE;
-	mci->edac_cap		= EDAC_FLAG_NONE;
 
 	if (pvt->nbcap & K8_NBCAP_SECDED)
 		mci->edac_ctl_cap |= EDAC_FLAG_SECDED;
-- 
cgit v1.2.3-59-g8ed1b


From 30c875cbc1836a03a1acc6c998fa8a04f29f8f73 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <borislav.petkov@amd.com>
Date: Mon, 22 Jun 2009 19:42:24 +0200
Subject: amd64_edac: fix ecc_enable_override handling

amd64_check_ecc_enabled() returns non-zero status when ECC
checking/correcting is disabled and this fails further loading of the
driver even when 'ecc_enable_override' boot param is used.

Fix that by clearing return status in that case.

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
---
 drivers/edac/amd64_edac.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c
index 3b76605330e9..8497963a61f6 100644
--- a/drivers/edac/amd64_edac.c
+++ b/drivers/edac/amd64_edac.c
@@ -2966,7 +2966,12 @@ static int amd64_check_ecc_enabled(struct amd64_pvt *pvt)
 				"    Use of the override can cause "
 				"unknown side effects.\n");
 			ret = -ENODEV;
-		}
+		} else
+			/*
+			 * enable further driver loading if ECC enable is
+			 * overridden.
+			 */
+			ret = 0;
 	} else {
 		amd64_printk(KERN_INFO,
 			"ECC is enabled by BIOS, Proceeding "
-- 
cgit v1.2.3-59-g8ed1b


From 37da045067b4e923190662e21029005ea53bfaa1 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <borislav.petkov@amd.com>
Date: Wed, 10 Jun 2009 17:36:57 +0200
Subject: amd64_edac: misc small cleanups

- cleanup debug calls
- shorten function names
- cleanup error exit paths

Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
---
 drivers/edac/amd64_edac.c | 13 +++++++------
 drivers/edac/amd64_edac.h |  2 +-
 2 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/drivers/edac/amd64_edac.c b/drivers/edac/amd64_edac.c
index 8497963a61f6..858fe6037223 100644
--- a/drivers/edac/amd64_edac.c
+++ b/drivers/edac/amd64_edac.c
@@ -1269,7 +1269,7 @@ static int f10_early_channel_count(struct amd64_pvt *pvt)
 	if (channels == 0)
 		channels = 1;
 
-	debugf0("DIMM count= %d\n", channels);
+	debugf0("MCT channel count: %d\n", channels);
 
 	return channels;
 
@@ -3056,7 +3056,7 @@ static int amd64_probe_one_instance(struct pci_dev *dram_f2_ctl,
 	if (!pvt)
 		goto err_exit;
 
-	pvt->mc_node_id = get_mc_node_id_from_pdev(dram_f2_ctl);
+	pvt->mc_node_id = get_node_id(dram_f2_ctl);
 
 	pvt->dram_f2_ctl	= dram_f2_ctl;
 	pvt->ext_model		= boot_cpu_data.x86_model >> 4;
@@ -3183,8 +3183,7 @@ static int __devinit amd64_init_one_instance(struct pci_dev *pdev,
 {
 	int ret = 0;
 
-	debugf0("(MC node=%d,mc_type='%s')\n",
-		get_mc_node_id_from_pdev(pdev),
+	debugf0("(MC node=%d,mc_type='%s')\n", get_node_id(pdev),
 		get_amd_family_name(mc_type->driver_data));
 
 	ret = pci_enable_device(pdev);
@@ -3323,15 +3322,17 @@ static int __init amd64_edac_init(void)
 
 		err = amd64_init_2nd_stage(pvt_lookup[nb]);
 		if (err)
-			goto err_exit;
+			goto err_2nd_stage;
 	}
 
 	amd64_setup_pci_device();
 
 	return 0;
 
+err_2nd_stage:
+	debugf0("2nd stage failed\n");
+
 err_exit:
-	debugf0("'finish_setup' stage failed\n");
 	pci_unregister_driver(&amd64_pci_driver);
 
 	return err;
diff --git a/drivers/edac/amd64_edac.h b/drivers/edac/amd64_edac.h
index a159957e167b..ba73015af8e4 100644
--- a/drivers/edac/amd64_edac.h
+++ b/drivers/edac/amd64_edac.h
@@ -444,7 +444,7 @@ enum {
 #define K8_MSR_MC4ADDR			0x0412
 
 /* AMD sets the first MC device at device ID 0x18. */
-static inline int get_mc_node_id_from_pdev(struct pci_dev *pdev)
+static inline int get_node_id(struct pci_dev *pdev)
 {
 	return PCI_SLOT(pdev->devfn) - 0x18;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 8cb76d99d715741637b6d0884f389e17e9cb05d2 Mon Sep 17 00:00:00 2001
From: Frederic Weisbecker <fweisbec@gmail.com>
Date: Fri, 26 Jun 2009 16:28:00 +0200
Subject: perf_counter tools: Prepare a small callchain framework

We plan to display the callchains depending on some user-configurable
parameters.

To gather the callchains stats from the recorded stream in a fast way,
this patch introduces an ad hoc radix tree adapted for callchains and also
a rbtree to sort these callchains once we have gathered every events
from the stream.

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1246026481-8314-2-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/Makefile         |   1 +
 tools/perf/builtin-report.c |   5 --
 tools/perf/perf.h           |   5 ++
 tools/perf/util/callchain.c | 174 ++++++++++++++++++++++++++++++++++++++++++++
 tools/perf/util/callchain.h |  33 +++++++++
 5 files changed, 213 insertions(+), 5 deletions(-)
 create mode 100644 tools/perf/util/callchain.c
 create mode 100644 tools/perf/util/callchain.h

diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index d3887ed51a64..1c1296d8a64b 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -329,6 +329,7 @@ LIB_OBJS += util/symbol.o
 LIB_OBJS += util/color.o
 LIB_OBJS += util/pager.o
 LIB_OBJS += util/header.o
+LIB_OBJS += util/callchain.o
 
 BUILTIN_OBJS += builtin-annotate.o
 BUILTIN_OBJS += builtin-help.o
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 681c2233f882..28d1cb2127e9 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -62,11 +62,6 @@ struct ip_event {
 	unsigned char __more_data[];
 };
 
-struct ip_callchain {
-	u64 nr;
-	u64 ips[0];
-};
-
 struct mmap_event {
 	struct perf_event_header header;
 	u32 pid, tid;
diff --git a/tools/perf/perf.h b/tools/perf/perf.h
index 16c84fd73c86..a49842b69a59 100644
--- a/tools/perf/perf.h
+++ b/tools/perf/perf.h
@@ -66,4 +66,9 @@ sys_perf_counter_open(struct perf_counter_attr *attr,
 #define MAX_COUNTERS			256
 #define MAX_NR_CPUS			256
 
+struct ip_callchain {
+	u64 nr;
+	u64 ips[0];
+};
+
 #endif
diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
new file mode 100644
index 000000000000..ad3c28578961
--- /dev/null
+++ b/tools/perf/util/callchain.c
@@ -0,0 +1,174 @@
+/*
+ * Copyright (C) 2009, Frederic Weisbecker <fweisbec@gmail.com>
+ *
+ * Handle the callchains from the stream in an ad-hoc radix tree and then
+ * sort them in an rbtree.
+ *
+ */
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <errno.h>
+
+#include "callchain.h"
+
+
+static void rb_insert_callchain(struct rb_root *root, struct callchain_node *chain)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct callchain_node *rnode;
+
+	while (*p) {
+		parent = *p;
+		rnode = rb_entry(parent, struct callchain_node, rb_node);
+
+		if (rnode->hit < chain->hit)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&chain->rb_node, parent, p);
+	rb_insert_color(&chain->rb_node, root);
+}
+
+/*
+ * Once we get every callchains from the stream, we can now
+ * sort them by hit
+ */
+void sort_chain_to_rbtree(struct rb_root *rb_root, struct callchain_node *node)
+{
+	struct callchain_node *child;
+
+	list_for_each_entry(child, &node->children, brothers)
+		sort_chain_to_rbtree(rb_root, child);
+
+	if (node->hit)
+		rb_insert_callchain(rb_root, node);
+}
+
+static struct callchain_node *create_child(struct callchain_node *parent)
+{
+	struct callchain_node *new;
+
+	new = malloc(sizeof(*new));
+	if (!new) {
+		perror("not enough memory to create child for code path tree");
+		return NULL;
+	}
+	new->parent = parent;
+	INIT_LIST_HEAD(&new->children);
+	INIT_LIST_HEAD(&new->val);
+	list_add_tail(&new->brothers, &parent->children);
+
+	return new;
+}
+
+static void
+fill_node(struct callchain_node *node, struct ip_callchain *chain, int start)
+{
+	int i;
+
+	for (i = start; i < chain->nr; i++) {
+		struct callchain_list *call;
+
+		call = malloc(sizeof(*chain));
+		if (!call) {
+			perror("not enough memory for the code path tree");
+			return;
+		}
+		call->ip = chain->ips[i];
+		list_add_tail(&call->list, &node->val);
+	}
+	node->val_nr = i - start;
+}
+
+static void add_child(struct callchain_node *parent, struct ip_callchain *chain)
+{
+	struct callchain_node *new;
+
+	new = create_child(parent);
+	fill_node(new, chain, parent->val_nr);
+
+	new->hit = 1;
+}
+
+static void
+split_add_child(struct callchain_node *parent, struct ip_callchain *chain,
+		struct callchain_list *to_split, int idx)
+{
+	struct callchain_node *new;
+
+	/* split */
+	new = create_child(parent);
+	list_move_tail(&to_split->list, &new->val);
+	new->hit = parent->hit;
+	parent->hit = 0;
+	parent->val_nr = idx;
+
+	/* create the new one */
+	add_child(parent, chain);
+}
+
+static int
+__append_chain(struct callchain_node *root, struct ip_callchain *chain,
+		int start);
+
+static int
+__append_chain_children(struct callchain_node *root, struct ip_callchain *chain)
+{
+	struct callchain_node *rnode;
+
+	/* lookup in childrens */
+	list_for_each_entry(rnode, &root->children, brothers) {
+		int ret = __append_chain(rnode, chain, root->val_nr);
+		if (!ret)
+			return 0;
+	}
+	return -1;
+}
+
+static int
+__append_chain(struct callchain_node *root, struct ip_callchain *chain,
+		int start)
+{
+	struct callchain_list *cnode;
+	int i = start;
+	bool found = false;
+
+	/* lookup in the current node */
+	list_for_each_entry(cnode, &root->val, list) {
+		if (cnode->ip != chain->ips[i++])
+			break;
+		if (!found)
+			found = true;
+		if (i == chain->nr)
+			break;
+	}
+
+	/* matches not, relay on the parent */
+	if (!found)
+		return -1;
+
+	/* we match only a part of the node. Split it and add the new chain */
+	if (i < root->val_nr) {
+		split_add_child(root, chain, cnode, i);
+		return 0;
+	}
+
+	/* we match 100% of the path, increment the hit */
+	if (i == root->val_nr) {
+		root->hit++;
+		return 0;
+	}
+
+	return __append_chain_children(root, chain);
+}
+
+void append_chain(struct callchain_node *root, struct ip_callchain *chain)
+{
+	if (__append_chain_children(root, chain) == -1)
+		add_child(root, chain);
+}
diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h
new file mode 100644
index 000000000000..fa1cd2f71fd3
--- /dev/null
+++ b/tools/perf/util/callchain.h
@@ -0,0 +1,33 @@
+#ifndef __PERF_CALLCHAIN_H
+#define __PERF_CALLCHAIN_H
+
+#include "../perf.h"
+#include "list.h"
+#include "rbtree.h"
+
+
+struct callchain_node {
+	struct callchain_node	*parent;
+	struct list_head	brothers;
+	struct list_head 	children;
+	struct list_head 	val;
+	struct rb_node		rb_node;
+	int			val_nr;
+	int			hit;
+};
+
+struct callchain_list {
+	unsigned long		ip;
+	struct list_head	list;
+};
+
+static inline void callchain_init(struct callchain_node *node)
+{
+	INIT_LIST_HEAD(&node->brothers);
+	INIT_LIST_HEAD(&node->children);
+	INIT_LIST_HEAD(&node->val);
+}
+
+void append_chain(struct callchain_node *root, struct ip_callchain *chain);
+void sort_chain_to_rbtree(struct rb_root *rb_root, struct callchain_node *node);
+#endif
-- 
cgit v1.2.3-59-g8ed1b


From f55c555226b1010b249730ec6b232e5470286950 Mon Sep 17 00:00:00 2001
From: Frederic Weisbecker <fweisbec@gmail.com>
Date: Fri, 26 Jun 2009 16:28:01 +0200
Subject: perf report: Print sorted callchains per histogram entries

Use the newly created callchains radix tree to gather the chains stats
from the recorded events and then print the callchains for all of them,
sorted by hits, using the "-c" parameter with perf report.

Example:

 66.15%  [k] atm_clip_exit
            63.08%
                0xffffffffffffff80
                0xffffffff810196a8
                0xffffffff810c14c8
                0xffffffff8101a79c
                0xffffffff810194f3
                0xffffffff8106ab7f
                0xffffffff8106abe5
                0xffffffff8106acde
                0xffffffff8100d94b
                0xffffffff8153e7ea
                [...]

             1.54%
                0xffffffffffffff80
                0xffffffff810196a8
                0xffffffff810c14c8
                0xffffffff8101a79c
		[...]

Symbols are not yet resolved.

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1246026481-8314-3-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-report.c | 82 +++++++++++++++++++++++++++++++++++++++------
 1 file changed, 71 insertions(+), 11 deletions(-)

diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 28d1cb2127e9..ed391db9e0f8 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -15,6 +15,7 @@
 #include "util/rbtree.h"
 #include "util/symbol.h"
 #include "util/string.h"
+#include "util/callchain.h"
 
 #include "perf.h"
 #include "util/header.h"
@@ -52,6 +53,7 @@ static char		*parent_pattern = default_parent_pattern;
 static regex_t		parent_regex;
 
 static int		exclude_other = 1;
+static int		callchain;
 
 static u64		sample_type;
 
@@ -488,17 +490,19 @@ static size_t threads__fprintf(FILE *fp)
 static struct rb_root hist;
 
 struct hist_entry {
-	struct rb_node	 rb_node;
-
-	struct thread	 *thread;
-	struct map	 *map;
-	struct dso	 *dso;
-	struct symbol	 *sym;
-	struct symbol	 *parent;
-	u64		 ip;
-	char		 level;
-
-	u64		 count;
+	struct rb_node		rb_node;
+
+	struct thread		*thread;
+	struct map		*map;
+	struct dso		*dso;
+	struct symbol		*sym;
+	struct symbol		*parent;
+	u64			ip;
+	char			level;
+	struct callchain_node	callchain;
+	struct rb_root		sorted_chain;
+
+	u64			count;
 };
 
 /*
@@ -768,6 +772,48 @@ hist_entry__collapse(struct hist_entry *left, struct hist_entry *right)
 	return cmp;
 }
 
+static size_t
+callchain__fprintf(FILE *fp, struct callchain_node *self, u64 total_samples)
+{
+	struct callchain_list *chain;
+	size_t ret = 0;
+
+	if (!self)
+		return 0;
+
+	ret += callchain__fprintf(fp, self->parent, total_samples);
+
+
+	list_for_each_entry(chain, &self->val, list)
+		ret += fprintf(fp, "                %p\n", (void *)chain->ip);
+
+	return ret;
+}
+
+static size_t
+hist_entry_callchain__fprintf(FILE *fp, struct hist_entry *self,
+			      u64 total_samples)
+{
+	struct rb_node *rb_node;
+	struct callchain_node *chain;
+	size_t ret = 0;
+
+	rb_node = rb_first(&self->sorted_chain);
+	while (rb_node) {
+		double percent;
+
+		chain = rb_entry(rb_node, struct callchain_node, rb_node);
+		percent = chain->hit * 100.0 / total_samples;
+		ret += fprintf(fp, "           %6.2f%%\n", percent);
+		ret += callchain__fprintf(fp, chain, total_samples);
+		ret += fprintf(fp, "\n");
+		rb_node = rb_next(rb_node);
+	}
+
+	return ret;
+}
+
+
 static size_t
 hist_entry__fprintf(FILE *fp, struct hist_entry *self, u64 total_samples)
 {
@@ -808,6 +854,9 @@ hist_entry__fprintf(FILE *fp, struct hist_entry *self, u64 total_samples)
 
 	ret += fprintf(fp, "\n");
 
+	if (callchain)
+		hist_entry_callchain__fprintf(fp, self, total_samples);
+
 	return ret;
 }
 
@@ -892,6 +941,7 @@ hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
 		.level	= level,
 		.count	= count,
 		.parent = NULL,
+		.sorted_chain = RB_ROOT
 	};
 	int cmp;
 
@@ -934,6 +984,8 @@ hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
 
 		if (!cmp) {
 			he->count += count;
+			if (callchain)
+				append_chain(&he->callchain, chain);
 			return 0;
 		}
 
@@ -947,6 +999,10 @@ hist_entry__add(struct thread *thread, struct map *map, struct dso *dso,
 	if (!he)
 		return -ENOMEM;
 	*he = entry;
+	if (callchain) {
+		callchain_init(&he->callchain);
+		append_chain(&he->callchain, chain);
+	}
 	rb_link_node(&he->rb_node, parent, p);
 	rb_insert_color(&he->rb_node, &hist);
 
@@ -1023,6 +1079,9 @@ static void output__insert_entry(struct hist_entry *he)
 	struct rb_node *parent = NULL;
 	struct hist_entry *iter;
 
+	if (callchain)
+		sort_chain_to_rbtree(&he->sorted_chain, &he->callchain);
+
 	while (*p != NULL) {
 		parent = *p;
 		iter = rb_entry(parent, struct hist_entry, rb_node);
@@ -1599,6 +1658,7 @@ static const struct option options[] = {
 		   "regex filter to identify parent, see: '--sort parent'"),
 	OPT_BOOLEAN('x', "exclude-other", &exclude_other,
 		    "Only display entries with parent-match"),
+	OPT_BOOLEAN('c', "callchain", &callchain, "Display callchains"),
 	OPT_END()
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From 39562e783928e3ea9ee2cbce99a756ab48d3c06a Mon Sep 17 00:00:00 2001
From: Christof Schmitt <christof.schmitt@de.ibm.com>
Date: Fri, 26 Jun 2009 16:30:43 +0200
Subject: [SCSI] FC transport: Locking fix for common-code FC pass-through
 patch

Fix this:
------------[ cut here ]------------
Badness at block/blk-core.c:244
CPU: 0 Tainted: G        W  2.6.31-rc1-00004-gd3a263a #3
Process zfcp_wq (pid: 901, task: 000000002fb7a038, ksp: 000000002f02bc78)
Krnl PSW : 0704300180000000 00000000002141ba (blk_remove_plug+0xb2/0xb8)
           R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:0 CC:3 PM:0 EA:3
Krnl GPRS: 0000000000000001 0000000000000001 0000000022811440 0000000022811798
           000000000027ff4e 0000000000000000 0000000000000000 000000002f00f000
           070000000006a0f4 000000002af70000 000000002af2a800 00000000228d1c00
           0000000022811440 000000000050c708 000000002f02bca8 000000002f02bc80
Krnl Code: 00000000002141b0: b9140022		lgfr	%r2,%r2
           00000000002141b4: 07fe		bcr	15,%r14
           00000000002141b6: a7f40001		brc	15,2141b8
          >00000000002141ba: a7f4ffbe		brc	15,214136
           00000000002141be: 0707		bcr	0,%r7
           00000000002141c0: ebaff0680024	stmg %r10,%r15,104(%r15)
           00000000002141c6: c0d00017c2a9	larl	%r13,50c718
           00000000002141cc: a7f13fc0		tmll	%r15,16320
Call Trace:
([<000000000050e7d8>] C.272.16122+0x88/0x110)
 [<00000000002141ec>] __blk_run_queue+0x2c/0x154
 [<000000000028013a>] fc_remote_port_add+0x85e/0x95c
 [<000000000037596e>] zfcp_scsi_rport_work+0xe6/0x148
 [<000000000006908c>] worker_thread+0x25c/0x318
 [<000000000006f10c>] kthread+0x94/0x9c
 [<000000000001c2b2>] kernel_thread_starter+0x6/0xc
 [<000000000001c2ac>] kernel_thread_starter+0x0/0xc
INFO: lockdep is turned off.
Last Breaking-Event-Address:
 [<00000000002141b6>] blk_remove_plug+0xae/0xb8

The FC pass-through support triggers the WARN_ON(!irqs_disabled()) in
blk_plug_device. Since blk_plug_device requires being called with
disabled interrupts, use spin_lock_irqsave in fc_bsg_goose_queue to
disable the interrupts before calling into the block layer.

Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
Acked-by: James Smart <james.smart@emulex.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/scsi_transport_fc.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c
index 2eee9e6e4fe8..292c02f810d0 100644
--- a/drivers/scsi/scsi_transport_fc.c
+++ b/drivers/scsi/scsi_transport_fc.c
@@ -3670,13 +3670,14 @@ static void
 fc_bsg_goose_queue(struct fc_rport *rport)
 {
 	int flagset;
+	unsigned long flags;
 
 	if (!rport->rqst_q)
 		return;
 
 	get_device(&rport->dev);
 
-	spin_lock(rport->rqst_q->queue_lock);
+	spin_lock_irqsave(rport->rqst_q->queue_lock, flags);
 	flagset = test_bit(QUEUE_FLAG_REENTER, &rport->rqst_q->queue_flags) &&
 		  !test_bit(QUEUE_FLAG_REENTER, &rport->rqst_q->queue_flags);
 	if (flagset)
@@ -3684,7 +3685,7 @@ fc_bsg_goose_queue(struct fc_rport *rport)
 	__blk_run_queue(rport->rqst_q);
 	if (flagset)
 		queue_flag_clear(QUEUE_FLAG_REENTER, rport->rqst_q);
-	spin_unlock(rport->rqst_q->queue_lock);
+	spin_unlock_irqrestore(rport->rqst_q->queue_lock, flags);
 
 	put_device(&rport->dev);
 }
-- 
cgit v1.2.3-59-g8ed1b


From 19d2e755436054dfc2be640bffc32e427c37ac3d Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri, 26 Jun 2009 13:10:23 +0200
Subject: perf_counter: Complete counter swap

Complete the counter swap by indeed switching the times too and
updating the userpage after modifying the counter values.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1246014623.31755.195.camel@twins>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/perf_counter.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index f2f232696587..66ab1e9d1294 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -1048,9 +1048,14 @@ static void __perf_counter_sync_stat(struct perf_counter *counter,
 	value = atomic64_xchg(&counter->count, value);
 	atomic64_set(&next_counter->count, value);
 
+	swap(counter->total_time_enabled, next_counter->total_time_enabled);
+	swap(counter->total_time_running, next_counter->total_time_running);
+
 	/*
-	 * XXX also sync time_enabled and time_running ?
+	 * Since we swizzled the values, update the user visible data too.
 	 */
+	perf_counter_update_userpage(counter);
+	perf_counter_update_userpage(next_counter);
 }
 
 #define list_next_entry(pos, member) \
-- 
cgit v1.2.3-59-g8ed1b


From b9389796fa4c87fbdff33816e317cdae5f36dd0b Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Fri, 26 Jun 2009 09:28:42 -0700
Subject: sky2: Fix checksum endianness

sky2 driver on PowerPC targets floods kernel log with following errors:

  eth1: hw csum failure.
  Call Trace:
  [ef84b8a0] [c00075e4] show_stack+0x50/0x160 (unreliable)
  [ef84b8d0] [c02fa178] netdev_rx_csum_fault+0x3c/0x5c
  [ef84b8f0] [c02f6920] __skb_checksum_complete_head+0x7c/0x84
  [ef84b900] [c02f693c] __skb_checksum_complete+0x14/0x24
  [ef84b910] [c0337e08] tcp_v4_rcv+0x4c8/0x6f8
  [ef84b940] [c031a9c8] ip_local_deliver+0x98/0x210
  [ef84b960] [c031a788] ip_rcv+0x38c/0x534
  [ef84b990] [c0300338] netif_receive_skb+0x260/0x36c
  [ef84b9c0] [c025de00] sky2_poll+0x5dc/0xcf8
  [ef84ba20] [c02fb7fc] net_rx_action+0xc0/0x144

The NIC is Yukon-2 EC chip revision 1.

Converting checksum field from le16 to CPU byte order fixes the issue.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/sky2.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index 7681d28c53d7..daf961ab68bc 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -2495,7 +2495,7 @@ static int sky2_status_intr(struct sky2_hw *hw, int to_do, u16 idx)
 			if (likely(status >> 16 == (status & 0xffff))) {
 				skb = sky2->rx_ring[sky2->rx_next].skb;
 				skb->ip_summed = CHECKSUM_COMPLETE;
-				skb->csum = status & 0xffff;
+				skb->csum = le16_to_cpu(status);
 			} else {
 				printk(KERN_NOTICE PFX "%s: hardware receive "
 				       "checksum problem (status = %#x)\n",
-- 
cgit v1.2.3-59-g8ed1b


From e0a2a1601bec01243bcad44414d06f59dae2eedb Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 26 Jun 2009 17:38:25 +0100
Subject: kmemleak: Enable task stacks scanning by default

This is to reduce the number of false positives reported.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 Documentation/kmemleak.txt | 8 ++++----
 mm/kmemleak.c              | 2 +-
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/Documentation/kmemleak.txt b/Documentation/kmemleak.txt
index f655308064d7..9426e94f291a 100644
--- a/Documentation/kmemleak.txt
+++ b/Documentation/kmemleak.txt
@@ -31,12 +31,12 @@ Memory scanning parameters can be modified at run-time by writing to the
 /sys/kernel/debug/kmemleak file. The following parameters are supported:
 
   off		- disable kmemleak (irreversible)
-  stack=on	- enable the task stacks scanning
+  stack=on	- enable the task stacks scanning (default)
   stack=off	- disable the tasks stacks scanning
-  scan=on	- start the automatic memory scanning thread
+  scan=on	- start the automatic memory scanning thread (default)
   scan=off	- stop the automatic memory scanning thread
-  scan=<secs>	- set the automatic memory scanning period in seconds (0
-		  to disable it)
+  scan=<secs>	- set the automatic memory scanning period in seconds
+		  (default 600, 0 to stop the automatic scanning)
 
 Kmemleak can also be disabled at boot-time by passing "kmemleak=off" on
 the kernel command line.
diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index 17096d1b59b2..a38418a95d33 100644
--- a/mm/kmemleak.c
+++ b/mm/kmemleak.c
@@ -194,7 +194,7 @@ static unsigned long jiffies_min_age;
 /* delay between automatic memory scannings */
 static signed long jiffies_scan_wait;
 /* enables or disables the task stacks scanning */
-static int kmemleak_stack_scan;
+static int kmemleak_stack_scan = 1;
 /* mutex protecting the memory scanning */
 static DEFINE_MUTEX(scan_mutex);
 /* mutex protecting the access to the /sys/kernel/debug/kmemleak file */
-- 
cgit v1.2.3-59-g8ed1b


From bab4a34afc301fdb81b6ea0e3098d96fc356e03a Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 26 Jun 2009 17:38:26 +0100
Subject: kmemleak: Simplify the reports logged by the scanning thread

Because of false positives, the memory scanning thread may print too
much information. This patch changes the scanning thread to only print
the number of newly suspected leaks. Further information can be read
from the /sys/kernel/debug/kmemleak file.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 Documentation/kmemleak.txt |  6 ++---
 mm/kmemleak.c              | 61 ++++++++++++----------------------------------
 2 files changed, 19 insertions(+), 48 deletions(-)

diff --git a/Documentation/kmemleak.txt b/Documentation/kmemleak.txt
index 9426e94f291a..c06f7ba64993 100644
--- a/Documentation/kmemleak.txt
+++ b/Documentation/kmemleak.txt
@@ -16,9 +16,9 @@ Usage
 -----
 
 CONFIG_DEBUG_KMEMLEAK in "Kernel hacking" has to be enabled. A kernel
-thread scans the memory every 10 minutes (by default) and prints any new
-unreferenced objects found. To trigger an intermediate scan and display
-all the possible memory leaks:
+thread scans the memory every 10 minutes (by default) and prints the
+number of new unreferenced objects found. To trigger an intermediate
+scan and display the details of all the possible memory leaks:
 
   # mount -t debugfs nodev /sys/kernel/debug/
   # cat /sys/kernel/debug/kmemleak
diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index a38418a95d33..4130a4889fa9 100644
--- a/mm/kmemleak.c
+++ b/mm/kmemleak.c
@@ -278,15 +278,6 @@ static int color_gray(const struct kmemleak_object *object)
 	return object->min_count != -1 && object->count >= object->min_count;
 }
 
-/*
- * Objects are considered referenced if their color is gray and they have not
- * been deleted.
- */
-static int referenced_object(struct kmemleak_object *object)
-{
-	return (object->flags & OBJECT_ALLOCATED) && color_gray(object);
-}
-
 /*
  * Objects are considered unreferenced only if their color is white, they have
  * not be deleted and have a minimum age to avoid false positives caused by
@@ -299,38 +290,23 @@ static int unreferenced_object(struct kmemleak_object *object)
 }
 
 /*
- * Printing of the (un)referenced objects information, either to the seq file
- * or to the kernel log. The print_referenced/print_unreferenced functions
- * must be called with the object->lock held.
+ * Printing of the unreferenced objects information to the seq file. The
+ * print_unreferenced function must be called with the object->lock held.
  */
-#define print_helper(seq, x...)	do {	\
-	struct seq_file *s = (seq);	\
-	if (s)				\
-		seq_printf(s, x);	\
-	else				\
-		pr_info(x);		\
-} while (0)
-
-static void print_referenced(struct kmemleak_object *object)
-{
-	pr_info("referenced object 0x%08lx (size %zu)\n",
-		object->pointer, object->size);
-}
-
 static void print_unreferenced(struct seq_file *seq,
 			       struct kmemleak_object *object)
 {
 	int i;
 
-	print_helper(seq, "unreferenced object 0x%08lx (size %zu):\n",
-		     object->pointer, object->size);
-	print_helper(seq, "  comm \"%s\", pid %d, jiffies %lu\n",
-		     object->comm, object->pid, object->jiffies);
-	print_helper(seq, "  backtrace:\n");
+	seq_printf(seq, "unreferenced object 0x%08lx (size %zu):\n",
+		   object->pointer, object->size);
+	seq_printf(seq, "  comm \"%s\", pid %d, jiffies %lu\n",
+		   object->comm, object->pid, object->jiffies);
+	seq_printf(seq, "  backtrace:\n");
 
 	for (i = 0; i < object->trace_len; i++) {
 		void *ptr = (void *)object->trace[i];
-		print_helper(seq, "    [<%p>] %pS\n", ptr, ptr);
+		seq_printf(seq, "    [<%p>] %pS\n", ptr, ptr);
 	}
 }
 
@@ -571,8 +547,6 @@ static void delete_object(unsigned long ptr)
 	 * cannot be freed when it is being scanned.
 	 */
 	spin_lock_irqsave(&object->lock, flags);
-	if (object->flags & OBJECT_REPORTED)
-		print_referenced(object);
 	object->flags &= ~OBJECT_ALLOCATED;
 	spin_unlock_irqrestore(&object->lock, flags);
 	put_object(object);
@@ -1073,33 +1047,30 @@ static int kmemleak_scan_thread(void *arg)
 	while (!kthread_should_stop()) {
 		struct kmemleak_object *object;
 		signed long timeout = jiffies_scan_wait;
+		int new_leaks = 0;
 
 		mutex_lock(&scan_mutex);
 
 		kmemleak_scan();
-		reported_leaks = 0;
 
 		rcu_read_lock();
 		list_for_each_entry_rcu(object, &object_list, object_list) {
 			unsigned long flags;
 
-			if (reported_leaks >= REPORTS_NR)
-				break;
 			spin_lock_irqsave(&object->lock, flags);
-			if (!(object->flags & OBJECT_REPORTED) &&
-			    unreferenced_object(object)) {
-				print_unreferenced(NULL, object);
+			if (unreferenced_object(object) &&
+			    !(object->flags & OBJECT_REPORTED)) {
 				object->flags |= OBJECT_REPORTED;
-				reported_leaks++;
-			} else if ((object->flags & OBJECT_REPORTED) &&
-				   referenced_object(object)) {
-				print_referenced(object);
-				object->flags &= ~OBJECT_REPORTED;
+				new_leaks++;
 			}
 			spin_unlock_irqrestore(&object->lock, flags);
 		}
 		rcu_read_unlock();
 
+		if (new_leaks)
+			pr_info("%d new suspected memory leaks (see "
+				"/sys/kernel/debug/kmemleak)\n", new_leaks);
+
 		mutex_unlock(&scan_mutex);
 		/* wait before the next scan */
 		while (timeout && !kthread_should_stop())
-- 
cgit v1.2.3-59-g8ed1b


From 4698c1f2bbe44ce852ef1a6716973c1f5401a4c4 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 26 Jun 2009 17:38:27 +0100
Subject: kmemleak: Do not trigger a scan when reading the debug/kmemleak file

Since there is a kernel thread for automatically scanning the memory, it
makes sense for the debug/kmemleak file to only show its findings. This
patch also adds support for "echo scan > debug/kmemleak" to trigger an
intermediate memory scan and eliminates the kmemleak_mutex (scan_mutex
covers all the cases now).

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 Documentation/kmemleak.txt |  9 +++--
 mm/kmemleak.c              | 90 +++++++++++++++++++++-------------------------
 2 files changed, 47 insertions(+), 52 deletions(-)

diff --git a/Documentation/kmemleak.txt b/Documentation/kmemleak.txt
index c06f7ba64993..89068030b01b 100644
--- a/Documentation/kmemleak.txt
+++ b/Documentation/kmemleak.txt
@@ -17,12 +17,16 @@ Usage
 
 CONFIG_DEBUG_KMEMLEAK in "Kernel hacking" has to be enabled. A kernel
 thread scans the memory every 10 minutes (by default) and prints the
-number of new unreferenced objects found. To trigger an intermediate
-scan and display the details of all the possible memory leaks:
+number of new unreferenced objects found. To display the details of all
+the possible memory leaks:
 
   # mount -t debugfs nodev /sys/kernel/debug/
   # cat /sys/kernel/debug/kmemleak
 
+To trigger an intermediate memory scan:
+
+  # echo scan > /sys/kernel/debug/kmemleak
+
 Note that the orphan objects are listed in the order they were allocated
 and one object at the beginning of the list may cause other subsequent
 objects to be reported as orphan.
@@ -37,6 +41,7 @@ Memory scanning parameters can be modified at run-time by writing to the
   scan=off	- stop the automatic memory scanning thread
   scan=<secs>	- set the automatic memory scanning period in seconds
 		  (default 600, 0 to stop the automatic scanning)
+  scan		- trigger a memory scan
 
 Kmemleak can also be disabled at boot-time by passing "kmemleak=off" on
 the kernel command line.
diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index 4130a4889fa9..e96e0ec6a56e 100644
--- a/mm/kmemleak.c
+++ b/mm/kmemleak.c
@@ -48,10 +48,10 @@
  *   scanned. This list is only modified during a scanning episode when the
  *   scan_mutex is held. At the end of a scan, the gray_list is always empty.
  *   Note that the kmemleak_object.use_count is incremented when an object is
- *   added to the gray_list and therefore cannot be freed
- * - kmemleak_mutex (mutex): prevents multiple users of the "kmemleak" debugfs
- *   file together with modifications to the memory scanning parameters
- *   including the scan_thread pointer
+ *   added to the gray_list and therefore cannot be freed. This mutex also
+ *   prevents multiple users of the "kmemleak" debugfs file together with
+ *   modifications to the memory scanning parameters including the scan_thread
+ *   pointer
  *
  * The kmemleak_object structures have a use_count incremented or decremented
  * using the get_object()/put_object() functions. When the use_count becomes
@@ -195,10 +195,8 @@ static unsigned long jiffies_min_age;
 static signed long jiffies_scan_wait;
 /* enables or disables the task stacks scanning */
 static int kmemleak_stack_scan = 1;
-/* mutex protecting the memory scanning */
+/* protects the memory scanning, parameters and debug/kmemleak file access */
 static DEFINE_MUTEX(scan_mutex);
-/* mutex protecting the access to the /sys/kernel/debug/kmemleak file */
-static DEFINE_MUTEX(kmemleak_mutex);
 
 /* number of leaks reported (for limitation purposes) */
 static int reported_leaks;
@@ -927,6 +925,7 @@ static void kmemleak_scan(void)
 	struct kmemleak_object *object, *tmp;
 	struct task_struct *task;
 	int i;
+	int new_leaks = 0;
 
 	/* prepare the kmemleak_object's */
 	rcu_read_lock();
@@ -1024,6 +1023,26 @@ static void kmemleak_scan(void)
 		object = tmp;
 	}
 	WARN_ON(!list_empty(&gray_list));
+
+	/*
+	 * Scanning result reporting.
+	 */
+	rcu_read_lock();
+	list_for_each_entry_rcu(object, &object_list, object_list) {
+		spin_lock_irqsave(&object->lock, flags);
+		if (unreferenced_object(object) &&
+		    !(object->flags & OBJECT_REPORTED)) {
+			object->flags |= OBJECT_REPORTED;
+			new_leaks++;
+		}
+		spin_unlock_irqrestore(&object->lock, flags);
+	}
+	rcu_read_unlock();
+
+	if (new_leaks)
+		pr_info("%d new suspected memory leaks (see "
+			"/sys/kernel/debug/kmemleak)\n", new_leaks);
+
 }
 
 /*
@@ -1045,33 +1064,12 @@ static int kmemleak_scan_thread(void *arg)
 	}
 
 	while (!kthread_should_stop()) {
-		struct kmemleak_object *object;
 		signed long timeout = jiffies_scan_wait;
-		int new_leaks = 0;
 
 		mutex_lock(&scan_mutex);
-
 		kmemleak_scan();
-
-		rcu_read_lock();
-		list_for_each_entry_rcu(object, &object_list, object_list) {
-			unsigned long flags;
-
-			spin_lock_irqsave(&object->lock, flags);
-			if (unreferenced_object(object) &&
-			    !(object->flags & OBJECT_REPORTED)) {
-				object->flags |= OBJECT_REPORTED;
-				new_leaks++;
-			}
-			spin_unlock_irqrestore(&object->lock, flags);
-		}
-		rcu_read_unlock();
-
-		if (new_leaks)
-			pr_info("%d new suspected memory leaks (see "
-				"/sys/kernel/debug/kmemleak)\n", new_leaks);
-
 		mutex_unlock(&scan_mutex);
+
 		/* wait before the next scan */
 		while (timeout && !kthread_should_stop())
 			timeout = schedule_timeout_interruptible(timeout);
@@ -1084,7 +1082,7 @@ static int kmemleak_scan_thread(void *arg)
 
 /*
  * Start the automatic memory scanning thread. This function must be called
- * with the kmemleak_mutex held.
+ * with the scan_mutex held.
  */
 void start_scan_thread(void)
 {
@@ -1099,7 +1097,7 @@ void start_scan_thread(void)
 
 /*
  * Stop the automatic memory scanning thread. This function must be called
- * with the kmemleak_mutex held.
+ * with the scan_mutex held.
  */
 void stop_scan_thread(void)
 {
@@ -1119,10 +1117,8 @@ static void *kmemleak_seq_start(struct seq_file *seq, loff_t *pos)
 	struct kmemleak_object *object;
 	loff_t n = *pos;
 
-	if (!n) {
-		kmemleak_scan();
+	if (!n)
 		reported_leaks = 0;
-	}
 	if (reported_leaks >= REPORTS_NR)
 		return NULL;
 
@@ -1206,13 +1202,10 @@ static int kmemleak_open(struct inode *inode, struct file *file)
 	if (!atomic_read(&kmemleak_enabled))
 		return -EBUSY;
 
-	ret = mutex_lock_interruptible(&kmemleak_mutex);
+	ret = mutex_lock_interruptible(&scan_mutex);
 	if (ret < 0)
 		goto out;
 	if (file->f_mode & FMODE_READ) {
-		ret = mutex_lock_interruptible(&scan_mutex);
-		if (ret < 0)
-			goto kmemleak_unlock;
 		ret = seq_open(file, &kmemleak_seq_ops);
 		if (ret < 0)
 			goto scan_unlock;
@@ -1221,8 +1214,6 @@ static int kmemleak_open(struct inode *inode, struct file *file)
 
 scan_unlock:
 	mutex_unlock(&scan_mutex);
-kmemleak_unlock:
-	mutex_unlock(&kmemleak_mutex);
 out:
 	return ret;
 }
@@ -1231,11 +1222,9 @@ static int kmemleak_release(struct inode *inode, struct file *file)
 {
 	int ret = 0;
 
-	if (file->f_mode & FMODE_READ) {
+	if (file->f_mode & FMODE_READ)
 		seq_release(inode, file);
-		mutex_unlock(&scan_mutex);
-	}
-	mutex_unlock(&kmemleak_mutex);
+	mutex_unlock(&scan_mutex);
 
 	return ret;
 }
@@ -1250,6 +1239,7 @@ static int kmemleak_release(struct inode *inode, struct file *file)
  *   scan=off	- stop the automatic memory scanning thread
  *   scan=...	- set the automatic memory scanning period in seconds (0 to
  *		  disable it)
+ *   scan	- trigger a memory scan
  */
 static ssize_t kmemleak_write(struct file *file, const char __user *user_buf,
 			      size_t size, loff_t *ppos)
@@ -1287,7 +1277,9 @@ static ssize_t kmemleak_write(struct file *file, const char __user *user_buf,
 			jiffies_scan_wait = msecs_to_jiffies(secs * 1000);
 			start_scan_thread();
 		}
-	} else
+	} else if (strncmp(buf, "scan", 4) == 0)
+		kmemleak_scan();
+	else
 		return -EINVAL;
 
 	/* ignore the rest of the buffer, only one command at a time */
@@ -1312,11 +1304,9 @@ static int kmemleak_cleanup_thread(void *arg)
 {
 	struct kmemleak_object *object;
 
-	mutex_lock(&kmemleak_mutex);
+	mutex_lock(&scan_mutex);
 	stop_scan_thread();
-	mutex_unlock(&kmemleak_mutex);
 
-	mutex_lock(&scan_mutex);
 	rcu_read_lock();
 	list_for_each_entry_rcu(object, &object_list, object_list)
 		delete_object(object->pointer);
@@ -1458,9 +1448,9 @@ static int __init kmemleak_late_init(void)
 				     &kmemleak_fops);
 	if (!dentry)
 		pr_warning("Failed to create the debugfs kmemleak file\n");
-	mutex_lock(&kmemleak_mutex);
+	mutex_lock(&scan_mutex);
 	start_scan_thread();
-	mutex_unlock(&kmemleak_mutex);
+	mutex_unlock(&scan_mutex);
 
 	pr_info("Kernel memory leak detector initialized\n");
 
-- 
cgit v1.2.3-59-g8ed1b


From acf4968ec9dea49387ca8b3d36dfaa0850bdb2d5 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Fri, 26 Jun 2009 17:38:29 +0100
Subject: kmemleak: Slightly change the policy on newly allocated objects

Newly allocated objects are more likely to be reported as false
positives. Kmemleak ignores the reporting of objects younger than 5
seconds. However, this age was calculated after the memory scanning
completed which usually takes longer than 5 seconds. This patch
make the minimum object age calculation in relation to the start of the
memory scanning.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 mm/kmemleak.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index e96e0ec6a56e..c37e8e50e4de 100644
--- a/mm/kmemleak.c
+++ b/mm/kmemleak.c
@@ -190,7 +190,9 @@ static unsigned long max_addr;
 static unsigned long next_scan_yield;
 static struct task_struct *scan_thread;
 static unsigned long jiffies_scan_yield;
+/* used to avoid reporting of recently allocated objects */
 static unsigned long jiffies_min_age;
+static unsigned long jiffies_last_scan;
 /* delay between automatic memory scannings */
 static signed long jiffies_scan_wait;
 /* enables or disables the task stacks scanning */
@@ -284,7 +286,8 @@ static int color_gray(const struct kmemleak_object *object)
 static int unreferenced_object(struct kmemleak_object *object)
 {
 	return (object->flags & OBJECT_ALLOCATED) && color_white(object) &&
-		time_is_before_eq_jiffies(object->jiffies + jiffies_min_age);
+		time_before_eq(object->jiffies + jiffies_min_age,
+			       jiffies_last_scan);
 }
 
 /*
@@ -927,6 +930,8 @@ static void kmemleak_scan(void)
 	int i;
 	int new_leaks = 0;
 
+	jiffies_last_scan = jiffies;
+
 	/* prepare the kmemleak_object's */
 	rcu_read_lock();
 	list_for_each_entry_rcu(object, &object_list, object_list) {
-- 
cgit v1.2.3-59-g8ed1b


From 89bb871e96cdc3d78b7f69f0bacc94b21bbaccfd Mon Sep 17 00:00:00 2001
From: "Steven A. Falco" <sfalco@harris.com>
Date: Fri, 26 Jun 2009 12:42:47 -0400
Subject: mtd: m25p80 timeout too short for worst-case m25p16 devices

The m25p16 data sheet from numonyx lists the worst-case bulk erase time
(tBE) as 40 seconds.

Signed-off-by: Steven A. Falco <sfalco@harris.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/devices/m25p80.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/devices/m25p80.c b/drivers/mtd/devices/m25p80.c
index 59c46126a5ce..ae5fe91867e1 100644
--- a/drivers/mtd/devices/m25p80.c
+++ b/drivers/mtd/devices/m25p80.c
@@ -54,7 +54,7 @@
 #define	SR_SRWD			0x80	/* SR write protect */
 
 /* Define max times to check status register before we give up. */
-#define	MAX_READY_WAIT_JIFFIES	(10 * HZ)	/* eg. M25P128 specs 6s max sector erase */
+#define	MAX_READY_WAIT_JIFFIES	(40 * HZ)	/* M25P16 specs 40s max chip erase */
 #define	CMD_SIZE		4
 
 #ifdef CONFIG_M25PXX_USE_FAST_READ
-- 
cgit v1.2.3-59-g8ed1b


From 9c72ebef5aabf3532469d602a9d87beceea268b1 Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@gmail.com>
Date: Fri, 26 Jun 2009 11:22:37 -0700
Subject: ide-cd: handle fragmented packet commands gracefully

There are some devices in the wild that clear the DRQ bit during the
last word of a packet command and therefore could use a "second chance"
for that last word of data to be xferred instead of simply failing the
request. Do that by attempting to suck in those last bytes in PIO mode.

In addition, the ATA_ERR bit has to be cleared for we cannot be sure the
data is valid otherwise.

See http://bugzilla.kernel.org/show_bug.cgi?id=13399 for details.

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Acked-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/ide-cd.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index f0ede5953af8..6a9a769bffc1 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -592,9 +592,19 @@ static ide_startstop_t cdrom_newpc_intr(ide_drive_t *drive)
 			}
 		} else if (!blk_pc_request(rq)) {
 			ide_cd_request_sense_fixup(drive, cmd);
-			/* complain if we still have data left to transfer */
+
 			uptodate = cmd->nleft ? 0 : 1;
-			if (uptodate == 0)
+
+			/*
+			 * suck out the remaining bytes from the drive in an
+			 * attempt to complete the data xfer. (see BZ#13399)
+			 */
+			if (!(stat & ATA_ERR) && !uptodate && thislen) {
+				ide_pio_bytes(drive, cmd, write, thislen);
+				uptodate = cmd->nleft ? 0 : 1;
+			}
+
+			if (!uptodate)
 				rq->cmd_flags |= REQ_FAILED;
 		}
 		goto out_end;
-- 
cgit v1.2.3-59-g8ed1b


From 10e85448019097e4fcfa535f612f51d0d31a34f4 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <hawk@comx.dk>
Date: Fri, 26 Jun 2009 10:46:08 +0000
Subject: decnet: Use rcu_barrier() on module unload.

The decnet module unloading as been disabled with a '#if 0' statement,
because it have had issues.

We add a rcu_barrier() anyhow for correctness.

The maintainer (Chrissie Caulfield) will look into the unload issue
when time permits.

Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Chrissie Caulfield <christine.caulfield@googlemail.com>
Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/decnet/af_decnet.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c
index d351b8db0df5..77d40289653c 100644
--- a/net/decnet/af_decnet.c
+++ b/net/decnet/af_decnet.c
@@ -2413,6 +2413,8 @@ static void __exit decnet_exit(void)
 	proc_net_remove(&init_net, "decnet");
 
 	proto_unregister(&dn_proto);
+
+	rcu_barrier_bh(); /* Wait for completion of call_rcu_bh()'s */
 }
 module_exit(decnet_exit);
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From 1f2ccd00f224a4e2d6d26f590f3e6851f3deef99 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <hawk@comx.dk>
Date: Fri, 26 Jun 2009 10:46:03 +0000
Subject: ipv6: Use rcu_barrier() on module unload.

The ipv6 module uses rcu_call() thus it should use rcu_barrier() on
module unload.

Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/af_inet6.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 85b3d0036afd..caa0278d30a9 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -1284,6 +1284,8 @@ static void __exit inet6_exit(void)
 	proto_unregister(&udplitev6_prot);
 	proto_unregister(&udpv6_prot);
 	proto_unregister(&tcpv6_prot);
+
+	rcu_barrier(); /* Wait for completion of call_rcu()'s */
 }
 module_exit(inet6_exit);
 
-- 
cgit v1.2.3-59-g8ed1b


From 473c22d759e73cbbe604f41105b497817cc2ee8e Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <hawk@comx.dk>
Date: Fri, 26 Jun 2009 10:45:48 +0000
Subject: bridge: Use rcu_barrier() instead of syncronize_net() on unload.

When unloading modules that uses call_rcu() callbacks, then we must
use rcu_barrier().  This module uses syncronize_net() which is not
enough to be sure that all callback has been completed.

Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/bridge/br.c b/net/bridge/br.c
index 9aac5213105a..e1241c76239a 100644
--- a/net/bridge/br.c
+++ b/net/bridge/br.c
@@ -93,7 +93,7 @@ static void __exit br_deinit(void)
 
 	unregister_pernet_subsys(&br_net_ops);
 
-	synchronize_net();
+	rcu_barrier(); /* Wait for completion of call_rcu()'s */
 
 	br_netfilter_fini();
 #if defined(CONFIG_ATM_LANE) || defined(CONFIG_ATM_LANE_MODULE)
-- 
cgit v1.2.3-59-g8ed1b


From 75de874f5c35f679c6370fccc2bf4930e638ef3b Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <hawk@comx.dk>
Date: Fri, 26 Jun 2009 10:45:58 +0000
Subject: sunrpc: Use rcu_barrier() on unload.

The sunrpc module uses rcu_call() thus it should use rcu_barrier() on
module unload.

Have not verified that the possibility for new call_rcu() callbacks
has been disabled.  As a hint for checking, the functions calling
call_rcu() (unx_destroy_cred and generic_destroy_cred) are
registered as crdestroy function pointer in struct rpc_credops.

Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sunrpc/sunrpc_syms.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/sunrpc/sunrpc_syms.c b/net/sunrpc/sunrpc_syms.c
index 843629f55763..adaa81982f74 100644
--- a/net/sunrpc/sunrpc_syms.c
+++ b/net/sunrpc/sunrpc_syms.c
@@ -66,6 +66,7 @@ cleanup_sunrpc(void)
 #ifdef CONFIG_PROC_FS
 	rpc_proc_exit();
 #endif
+	rcu_barrier(); /* Wait for completion of call_rcu()'s */
 }
 MODULE_LICENSE("GPL");
 module_init(init_sunrpc);
-- 
cgit v1.2.3-59-g8ed1b


From 4a27096bbe2cad4c6e78802a0d9dfe0e598a1129 Mon Sep 17 00:00:00 2001
From: Jesper Dangaard Brouer <hawk@comx.dk>
Date: Fri, 26 Jun 2009 10:45:53 +0000
Subject: mac80211: Use rcu_barrier() on unload.

The mac80211 module uses rcu_call() thus it should use rcu_barrier()
on module unload.

The rcu_barrier() is placed in mech.c ieee80211_stop_mesh() which is
invoked from ieee80211_stop() in case vif.type == NL80211_IFTYPE_MESH_POINT.

Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Jesper Dangaard Brouer <hawk@comx.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/mac80211/mesh.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c
index fc712e60705d..11cf45bce38a 100644
--- a/net/mac80211/mesh.c
+++ b/net/mac80211/mesh.c
@@ -494,7 +494,7 @@ void ieee80211_stop_mesh(struct ieee80211_sub_if_data *sdata)
 	 * should it be using the interface and enqueuing
 	 * frames at this very time on another CPU.
 	 */
-	synchronize_rcu();
+	rcu_barrier(); /* Wait for RX path and call_rcu()'s */
 	skb_queue_purge(&sdata->u.mesh.skb_queue);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 71f9dacd2e4d233029e9e956ca3f79531f411827 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 26 Jun 2009 19:22:37 -0700
Subject: inet: Call skb_orphan before tproxy activates

As transparent proxying looks up the socket early and assigns
it to the skb for later processing, we must drop any existing
socket ownership prior to that in order to distinguish between
the case where tproxy is active and where it is not.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/ip_input.c  | 3 +++
 net/ipv6/ip6_input.c | 3 +++
 2 files changed, 6 insertions(+)

diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index 490ce20faf38..db46b4b5b2b9 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -440,6 +440,9 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt,
 	/* Remove any debris in the socket control block */
 	memset(IPCB(skb), 0, sizeof(struct inet_skb_parm));
 
+	/* Must drop socket now because of tproxy. */
+	skb_orphan(skb);
+
 	return NF_HOOK(PF_INET, NF_INET_PRE_ROUTING, skb, dev, NULL,
 		       ip_rcv_finish);
 
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index c3a07d75b5f5..6d6a4277c677 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -139,6 +139,9 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt
 
 	rcu_read_unlock();
 
+	/* Must drop socket now because of tproxy. */
+	skb_orphan(skb);
+
 	return NF_HOOK(PF_INET6, NF_INET_PRE_ROUTING, skb, dev, NULL,
 		       ip6_rcv_finish);
 err:
-- 
cgit v1.2.3-59-g8ed1b


From ff780cd8f2fa928b193554f593b36d1243554212 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 26 Jun 2009 19:27:04 -0700
Subject: gro: Flush GRO packets in napi_disable_pending path

When NAPI is disabled while we're in net_rx_action, we end up
calling __napi_complete without flushing GRO packets.  This is
a bug as it would cause the GRO packets to linger, of course it
also literally BUGs to catch error like this :)

This patch changes it to napi_complete, with the obligatory IRQ
reenabling.  This should be safe because we've only just disabled
IRQs and it does not materially affect the test conditions in
between.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/dev.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 60b572812278..70c27e0c7c32 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2823,9 +2823,11 @@ static void net_rx_action(struct softirq_action *h)
 		 * move the instance around on the list at-will.
 		 */
 		if (unlikely(work == weight)) {
-			if (unlikely(napi_disable_pending(n)))
-				__napi_complete(n);
-			else
+			if (unlikely(napi_disable_pending(n))) {
+				local_irq_enable();
+				napi_complete(n);
+				local_irq_disable();
+			} else
 				list_move_tail(&n->poll_list, list);
 		}
 
-- 
cgit v1.2.3-59-g8ed1b


From fde953c1c67986e1c381fa50d8207b1578b5cefa Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sat, 27 Jun 2009 06:06:39 +0200
Subject: perf_counter tools: Remove dead code

Vince Weaver reported that there's a handful of #ifdef __MINGW32__
sections in the code.

Remove them as they are in essence dead code - as unlike upstream
Git, the perf tool is unlikely to be ported to Windows.

Reported-by: Vince Weaver <vince@deater.net>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/util/help.c        | 15 -------
 tools/perf/util/pager.c       |  5 +--
 tools/perf/util/run-command.c | 95 +------------------------------------------
 tools/perf/util/run-command.h |  5 ---
 tools/perf/util/util.h        | 15 -------
 5 files changed, 3 insertions(+), 132 deletions(-)

diff --git a/tools/perf/util/help.c b/tools/perf/util/help.c
index 6653f7dd1d78..17a00e0df2c4 100644
--- a/tools/perf/util/help.c
+++ b/tools/perf/util/help.c
@@ -126,21 +126,6 @@ static int is_executable(const char *name)
 	    !S_ISREG(st.st_mode))
 		return 0;
 
-#ifdef __MINGW32__
-	/* cannot trust the executable bit, peek into the file instead */
-	char buf[3] = { 0 };
-	int n;
-	int fd = open(name, O_RDONLY);
-	st.st_mode &= ~S_IXUSR;
-	if (fd >= 0) {
-		n = read(fd, buf, 2);
-		if (n == 2)
-			/* DOS executables start with "MZ" */
-			if (!strcmp(buf, "#!") || !strcmp(buf, "MZ"))
-				st.st_mode |= S_IXUSR;
-		close(fd);
-	}
-#endif
 	return st.st_mode & S_IXUSR;
 }
 
diff --git a/tools/perf/util/pager.c b/tools/perf/util/pager.c
index a28bccae5458..1915de20dcac 100644
--- a/tools/perf/util/pager.c
+++ b/tools/perf/util/pager.c
@@ -9,7 +9,6 @@
 
 static int spawned_pager;
 
-#ifndef __MINGW32__
 static void pager_preexec(void)
 {
 	/*
@@ -24,7 +23,6 @@ static void pager_preexec(void)
 
 	setenv("LESS", "FRSX", 0);
 }
-#endif
 
 static const char *pager_argv[] = { "sh", "-c", NULL, NULL };
 static struct child_process pager_process;
@@ -70,9 +68,8 @@ void setup_pager(void)
 	pager_argv[2] = pager;
 	pager_process.argv = pager_argv;
 	pager_process.in = -1;
-#ifndef __MINGW32__
 	pager_process.preexec_cb = pager_preexec;
-#endif
+
 	if (start_command(&pager_process))
 		return;
 
diff --git a/tools/perf/util/run-command.c b/tools/perf/util/run-command.c
index b2f5e854f40a..a3935343091a 100644
--- a/tools/perf/util/run-command.c
+++ b/tools/perf/util/run-command.c
@@ -65,7 +65,6 @@ int start_command(struct child_process *cmd)
 		cmd->err = fderr[0];
 	}
 
-#ifndef __MINGW32__
 	fflush(NULL);
 	cmd->pid = fork();
 	if (!cmd->pid) {
@@ -118,71 +117,6 @@ int start_command(struct child_process *cmd)
 		}
 		exit(127);
 	}
-#else
-	int s0 = -1, s1 = -1, s2 = -1;	/* backups of stdin, stdout, stderr */
-	const char **sargv = cmd->argv;
-	char **env = environ;
-
-	if (cmd->no_stdin) {
-		s0 = dup(0);
-		dup_devnull(0);
-	} else if (need_in) {
-		s0 = dup(0);
-		dup2(fdin[0], 0);
-	} else if (cmd->in) {
-		s0 = dup(0);
-		dup2(cmd->in, 0);
-	}
-
-	if (cmd->no_stderr) {
-		s2 = dup(2);
-		dup_devnull(2);
-	} else if (need_err) {
-		s2 = dup(2);
-		dup2(fderr[1], 2);
-	}
-
-	if (cmd->no_stdout) {
-		s1 = dup(1);
-		dup_devnull(1);
-	} else if (cmd->stdout_to_stderr) {
-		s1 = dup(1);
-		dup2(2, 1);
-	} else if (need_out) {
-		s1 = dup(1);
-		dup2(fdout[1], 1);
-	} else if (cmd->out > 1) {
-		s1 = dup(1);
-		dup2(cmd->out, 1);
-	}
-
-	if (cmd->dir)
-		die("chdir in start_command() not implemented");
-	if (cmd->env) {
-		env = copy_environ();
-		for (; *cmd->env; cmd->env++)
-			env = env_setenv(env, *cmd->env);
-	}
-
-	if (cmd->perf_cmd) {
-		cmd->argv = prepare_perf_cmd(cmd->argv);
-	}
-
-	cmd->pid = mingw_spawnvpe(cmd->argv[0], cmd->argv, env);
-
-	if (cmd->env)
-		free_environ(env);
-	if (cmd->perf_cmd)
-		free(cmd->argv);
-
-	cmd->argv = sargv;
-	if (s0 >= 0)
-		dup2(s0, 0), close(s0);
-	if (s1 >= 0)
-		dup2(s1, 1), close(s1);
-	if (s2 >= 0)
-		dup2(s2, 2), close(s2);
-#endif
 
 	if (cmd->pid < 0) {
 		int err = errno;
@@ -288,14 +222,6 @@ int run_command_v_opt_cd_env(const char **argv, int opt, const char *dir, const
 	return run_command(&cmd);
 }
 
-#ifdef __MINGW32__
-static __stdcall unsigned run_thread(void *data)
-{
-	struct async *async = data;
-	return async->proc(async->fd_for_proc, async->data);
-}
-#endif
-
 int start_async(struct async *async)
 {
 	int pipe_out[2];
@@ -304,7 +230,6 @@ int start_async(struct async *async)
 		return error("cannot create pipe: %s", strerror(errno));
 	async->out = pipe_out[0];
 
-#ifndef __MINGW32__
 	/* Flush stdio before fork() to avoid cloning buffers */
 	fflush(NULL);
 
@@ -319,33 +244,17 @@ int start_async(struct async *async)
 		exit(!!async->proc(pipe_out[1], async->data));
 	}
 	close(pipe_out[1]);
-#else
-	async->fd_for_proc = pipe_out[1];
-	async->tid = (HANDLE) _beginthreadex(NULL, 0, run_thread, async, 0, NULL);
-	if (!async->tid) {
-		error("cannot create thread: %s", strerror(errno));
-		close_pair(pipe_out);
-		return -1;
-	}
-#endif
+
 	return 0;
 }
 
 int finish_async(struct async *async)
 {
-#ifndef __MINGW32__
 	int ret = 0;
 
 	if (wait_or_whine(async->pid))
 		ret = error("waitpid (async) failed");
-#else
-	DWORD ret = 0;
-	if (WaitForSingleObject(async->tid, INFINITE) != WAIT_OBJECT_0)
-		ret = error("waiting for thread failed: %lu", GetLastError());
-	else if (!GetExitCodeThread(async->tid, &ret))
-		ret = error("cannot get thread exit code: %lu", GetLastError());
-	CloseHandle(async->tid);
-#endif
+
 	return ret;
 }
 
diff --git a/tools/perf/util/run-command.h b/tools/perf/util/run-command.h
index 328289f23669..cc1837deba88 100644
--- a/tools/perf/util/run-command.h
+++ b/tools/perf/util/run-command.h
@@ -79,12 +79,7 @@ struct async {
 	int (*proc)(int fd, void *data);
 	void *data;
 	int out;	/* caller reads from here and closes it */
-#ifndef __MINGW32__
 	pid_t pid;
-#else
-	HANDLE tid;
-	int fd_for_proc;
-#endif
 };
 
 int start_async(struct async *async);
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index b8cfed776d81..b4be6071c105 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -67,7 +67,6 @@
 #include <assert.h>
 #include <regex.h>
 #include <utime.h>
-#ifndef __MINGW32__
 #include <sys/wait.h>
 #include <sys/poll.h>
 #include <sys/socket.h>
@@ -81,20 +80,6 @@
 #include <netdb.h>
 #include <pwd.h>
 #include <inttypes.h>
-#if defined(__CYGWIN__)
-#undef _XOPEN_SOURCE
-#include <grp.h>
-#define _XOPEN_SOURCE 600
-#include "compat/cygwin.h"
-#else
-#undef _ALL_SOURCE /* AIX 5.3L defines a struct list with _ALL_SOURCE. */
-#include <grp.h>
-#define _ALL_SOURCE 1
-#endif
-#else 	/* __MINGW32__ */
-/* pull in Windows compatibility stuff */
-#include "compat/mingw.h"
-#endif	/* __MINGW32__ */
 
 #ifndef NO_ICONV
 #include <iconv.h>
-- 
cgit v1.2.3-59-g8ed1b


From 0cfb7a13b8e4e0afd4b856156ab16a182de7505b Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sat, 27 Jun 2009 06:10:30 +0200
Subject: perf stat: Add -n/--null option to run without counters

Allow a no-counters run. This can be useful to measure just
elapsed wall-clock time - or to assess the raw overhead of perf
stat itself, without running any counters.

Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-stat.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 8420ec589506..cdcd058fac08 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -70,6 +70,7 @@ static int			run_count			=  1;
 static int			inherit				=  1;
 static int			scale				=  1;
 static int			target_pid			= -1;
+static int			null_run			=  0;
 
 static int			fd[MAX_NR_CPUS][MAX_COUNTERS];
 
@@ -461,6 +462,8 @@ static const struct option options[] = {
 		    "be more verbose (show counter open errors, etc)"),
 	OPT_INTEGER('r', "repeat", &run_count,
 		    "repeat command and print average + stddev (max: 100)"),
+	OPT_BOOLEAN('n', "null", &null_run,
+		    "null run - dont start any counters"),
 	OPT_END()
 };
 
@@ -476,7 +479,7 @@ int cmd_stat(int argc, const char **argv, const char *prefix)
 	if (run_count <= 0 || run_count > MAX_RUN)
 		usage_with_options(stat_usage, options);
 
-	if (!nr_counters)
+	if (!null_run && !nr_counters)
 		nr_counters = 8;
 
 	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
-- 
cgit v1.2.3-59-g8ed1b


From 566747e6298289c5cb02d4939cb3abf1c4fe7e5a Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sat, 27 Jun 2009 06:24:32 +0200
Subject: perf stat: Fix multi-run stats

In multi-run (-r/--repeat) printouts, print out the noise of
the wall-clock average as well.

Also, fix a bug in printing out scaled counters: if it was not
scaled then we should not update the average with -1.

Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-stat.c | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index cdcd058fac08..52c176cc683e 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -353,8 +353,11 @@ static void calc_avg(void)
 				event_res_avg[j]+1, event_res[i][j]+1);
 			update_avg("counter/2", j,
 				event_res_avg[j]+2, event_res[i][j]+2);
-			update_avg("scaled", j,
-				event_scaled_avg + j, event_scaled[i]+j);
+			if (event_scaled[i][j] != -1)
+				update_avg("scaled", j,
+					event_scaled_avg + j, event_scaled[i]+j);
+			else
+				event_scaled_avg[j] = -1;
 		}
 	}
 	runtime_nsecs_avg /= run_count;
@@ -420,9 +423,13 @@ static void print_stat(int argc, const char **argv)
 
 
 	fprintf(stderr, "\n");
-	fprintf(stderr, " %14.9f  seconds time elapsed.\n",
+	fprintf(stderr, " %14.9f  seconds time elapsed",
 			(double)walltime_nsecs_avg/1e9);
-	fprintf(stderr, "\n");
+	if (run_count > 1) {
+		fprintf(stderr, "   ( +- %7.3f%% )",
+			100.0*(double)walltime_nsecs_noise/(double)walltime_nsecs_avg);
+	}
+	fprintf(stderr, "\n\n");
 }
 
 static volatile int signr = -1;
-- 
cgit v1.2.3-59-g8ed1b


From 70ec3bb8ea3f8c55b255f41d122c7d4d8c0d00b4 Mon Sep 17 00:00:00 2001
From: Julia Lawall <julia@diku.dk>
Date: Sat, 27 Jun 2009 09:55:32 +0200
Subject: mtd: Use BLOCK_NIL consistently in NFTL/INFTL

Use BLOCK_NIL consistently rather than sometimes 0xffff and sometimes
BLOCK_NIL.

The semantic patch that finds this issue is below
(http://www.emn.fr/x-info/coccinelle/).  On the other hand, the changes
were made by hand, in part because drivers/mtd/inftlcore.c contains dead
code that causes spatch to ignore a relevant function.  Specifically, the
function INFTL_findwriteunit contains a do-while loop, but always takes a
return that leaves the loop on the first iteration.

// <smpl>
@r exists@
identifier f,C;
@@

f(...) { ... return C; }

@s@
identifier r.C;
expression E;
@@

@@
identifier r.f,r.C,I;
expression s.E;
@@

f(...) {
 <...
(
  I
|
- E
+ C
)
 ...>
}

// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/inftlcore.c | 11 ++++++-----
 drivers/mtd/nftlcore.c  | 16 ++++++++--------
 2 files changed, 14 insertions(+), 13 deletions(-)

diff --git a/drivers/mtd/inftlcore.c b/drivers/mtd/inftlcore.c
index 73f05227dc8c..d8cf29c01cc4 100644
--- a/drivers/mtd/inftlcore.c
+++ b/drivers/mtd/inftlcore.c
@@ -226,7 +226,7 @@ static u16 INFTL_findfreeblock(struct INFTLrecord *inftl, int desperate)
 	if (!desperate && inftl->numfreeEUNs < 2) {
 		DEBUG(MTD_DEBUG_LEVEL1, "INFTL: there are too few free "
 			"EUNs (%d)\n", inftl->numfreeEUNs);
-		return 0xffff;
+		return BLOCK_NIL;
 	}
 
 	/* Scan for a free block */
@@ -281,7 +281,8 @@ static u16 INFTL_foldchain(struct INFTLrecord *inftl, unsigned thisVUC, unsigned
 	silly = MAX_LOOPS;
 	while (thisEUN < inftl->nb_blocks) {
 		for (block = 0; block < inftl->EraseSize/SECTORSIZE; block ++) {
-			if ((BlockMap[block] != 0xffff) || BlockDeleted[block])
+			if ((BlockMap[block] != BLOCK_NIL) ||
+			    BlockDeleted[block])
 				continue;
 
 			if (inftl_read_oob(mtd, (thisEUN * inftl->EraseSize)
@@ -525,7 +526,7 @@ static inline u16 INFTL_findwriteunit(struct INFTLrecord *inftl, unsigned block)
 			if (!silly--) {
 				printk(KERN_WARNING "INFTL: infinite loop in "
 					"Virtual Unit Chain 0x%x\n", thisVUC);
-				return 0xffff;
+				return BLOCK_NIL;
 			}
 
 			/* Skip to next block in chain */
@@ -549,7 +550,7 @@ hitused:
 			 * waiting to be picked up. We're going to have to fold
 			 * a chain to make room.
 			 */
-			thisEUN = INFTL_makefreeblock(inftl, 0xffff);
+			thisEUN = INFTL_makefreeblock(inftl, BLOCK_NIL);
 
 			/*
 			 * Hopefully we free something, lets try again.
@@ -631,7 +632,7 @@ hitused:
 
 	printk(KERN_WARNING "INFTL: error folding to make room for Virtual "
 		"Unit Chain 0x%x\n", thisVUC);
-	return 0xffff;
+	return BLOCK_NIL;
 }
 
 /*
diff --git a/drivers/mtd/nftlcore.c b/drivers/mtd/nftlcore.c
index e3f8495a94c2..fb86cacd5bdb 100644
--- a/drivers/mtd/nftlcore.c
+++ b/drivers/mtd/nftlcore.c
@@ -208,7 +208,7 @@ static u16 NFTL_findfreeblock(struct NFTLrecord *nftl, int desperate )
 	/* Normally, we force a fold to happen before we run out of free blocks completely */
 	if (!desperate && nftl->numfreeEUNs < 2) {
 		DEBUG(MTD_DEBUG_LEVEL1, "NFTL_findfreeblock: there are too few free EUNs\n");
-		return 0xffff;
+		return BLOCK_NIL;
 	}
 
 	/* Scan for a free block */
@@ -230,11 +230,11 @@ static u16 NFTL_findfreeblock(struct NFTLrecord *nftl, int desperate )
 			printk("Argh! No free blocks found! LastFreeEUN = %d, "
 			       "FirstEUN = %d\n", nftl->LastFreeEUN,
 			       le16_to_cpu(nftl->MediaHdr.FirstPhysicalEUN));
-			return 0xffff;
+			return BLOCK_NIL;
 		}
 	} while (pot != nftl->LastFreeEUN);
 
-	return 0xffff;
+	return BLOCK_NIL;
 }
 
 static u16 NFTL_foldchain (struct NFTLrecord *nftl, unsigned thisVUC, unsigned pendingblock )
@@ -431,7 +431,7 @@ static u16 NFTL_foldchain (struct NFTLrecord *nftl, unsigned thisVUC, unsigned p
 
 	/* add the header so that it is now a valid chain */
 	oob.u.a.VirtUnitNum = oob.u.a.SpareVirtUnitNum = cpu_to_le16(thisVUC);
-	oob.u.a.ReplUnitNum = oob.u.a.SpareReplUnitNum = 0xffff;
+	oob.u.a.ReplUnitNum = oob.u.a.SpareReplUnitNum = BLOCK_NIL;
 
 	nftl_write_oob(mtd, (nftl->EraseSize * targetEUN) + 8,
 		       8, &retlen, (char *)&oob.u);
@@ -515,7 +515,7 @@ static u16 NFTL_makefreeblock( struct NFTLrecord *nftl , unsigned pendingblock)
 	if (ChainLength < 2) {
 		printk(KERN_WARNING "No Virtual Unit Chains available for folding. "
 		       "Failing request\n");
-		return 0xffff;
+		return BLOCK_NIL;
 	}
 
 	return NFTL_foldchain (nftl, LongestChain, pendingblock);
@@ -578,7 +578,7 @@ static inline u16 NFTL_findwriteunit(struct NFTLrecord *nftl, unsigned block)
 				printk(KERN_WARNING
 				       "Infinite loop in Virtual Unit Chain 0x%x\n",
 				       thisVUC);
-				return 0xffff;
+				return BLOCK_NIL;
 			}
 
 			/* Skip to next block in chain */
@@ -601,7 +601,7 @@ static inline u16 NFTL_findwriteunit(struct NFTLrecord *nftl, unsigned block)
 			//u16 startEUN = nftl->EUNtable[thisVUC];
 
 			//printk("Write to VirtualUnitChain %d, calling makefreeblock()\n", thisVUC);
-			writeEUN = NFTL_makefreeblock(nftl, 0xffff);
+			writeEUN = NFTL_makefreeblock(nftl, BLOCK_NIL);
 
 			if (writeEUN == BLOCK_NIL) {
 				/* OK, we accept that the above comment is
@@ -673,7 +673,7 @@ static inline u16 NFTL_findwriteunit(struct NFTLrecord *nftl, unsigned block)
 
 	printk(KERN_WARNING "Error folding to make room for Virtual Unit Chain 0x%x\n",
 	       thisVUC);
-	return 0xffff;
+	return BLOCK_NIL;
 }
 
 static int nftl_writeblock(struct mtd_blktrans_dev *mbd, unsigned long block,
-- 
cgit v1.2.3-59-g8ed1b


From 9a24ee03aebc55cec00cc388b6727bff24ed433f Mon Sep 17 00:00:00 2001
From: Nicolas Ferre <nicolas.ferre@atmel.com>
Date: Wed, 24 Jun 2009 17:13:48 +0100
Subject: [ARM] 5563/1: at91: at91sam9rlek lcd interface correction

Here is a little update to the at91sam9rlek lcd interface.
This will correct the power pin of the LCD.
It will also add precision to the struct atmel_lcdfb_info
scructure: backlight enabling  and wiring mode correction:
RGB wiring on the -EK board.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-at91/board-sam9rlek.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-at91/board-sam9rlek.c b/arch/arm/mach-at91/board-sam9rlek.c
index 35e12a49d1a6..f6b5672cabd6 100644
--- a/arch/arm/mach-at91/board-sam9rlek.c
+++ b/arch/arm/mach-at91/board-sam9rlek.c
@@ -186,19 +186,21 @@ static struct fb_monspecs at91fb_default_monspecs = {
 static void at91_lcdc_power_control(int on)
 {
 	if (on)
-		at91_set_gpio_value(AT91_PIN_PA30, 0);	/* power up */
+		at91_set_gpio_value(AT91_PIN_PC1, 0);	/* power up */
 	else
-		at91_set_gpio_value(AT91_PIN_PA30, 1);	/* power down */
+		at91_set_gpio_value(AT91_PIN_PC1, 1);	/* power down */
 }
 
 /* Driver datas */
 static struct atmel_lcdfb_info __initdata ek_lcdc_data = {
+	.lcdcon_is_backlight            = true,
 	.default_bpp			= 16,
 	.default_dmacon			= ATMEL_LCDC_DMAEN,
 	.default_lcdcon2		= AT91SAM9RL_DEFAULT_LCDCON2,
 	.default_monspecs		= &at91fb_default_monspecs,
 	.atmel_lcdfb_power_control	= at91_lcdc_power_control,
 	.guard_time			= 1,
+	.lcd_wiring_mode		= ATMEL_LCDC_WIRING_RGB,
 };
 
 #else
-- 
cgit v1.2.3-59-g8ed1b


From 6e750a8fc009fd0ae98704525d1d8e80d60e8cc9 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 27 Jun 2009 03:02:07 +0530
Subject: perf stat: Improve output

Increase size for event name to handle bigger names like
'L1-d$-prefetch-misses'

Changed scaled counters from percentage to a multiplicative
factor because the latter is more expressive.

Also aligned the scaling factor, otherwise sometimes it looks
like:

            384  iTLB-load-misses           (4.74x scaled)
         452029  branch-loads               (8.00x scaled)
           5892  branch-load-misses         (20.39x scaled)
         972315  iTLB-loads                 (3.24x scaled)

Before:
         150708  L1-d$-stores          (scaled from 23.57%)
         428804  L1-d$-prefetches      (scaled from 23.47%)
         314446  L1-d$-prefetch-misses  (scaled from 23.42%)
      252626137  L1-i$-loads           (scaled from 23.24%)
        5297550  dTLB-load-misses      (scaled from 23.96%)
      106992392  branch-loads          (scaled from 23.67%)
        5239561  branch-load-misses    (scaled from 23.43%)

After:
        1731713  L1-d$-loads               (  14.25x scaled)
          44241  L1-d$-prefetches          (   3.88x scaled)
          21076  L1-d$-prefetch-misses     (   3.40x scaled)
        5789421  L1-i$-loads               (   3.78x scaled)
          29645  dTLB-load-misses          (   2.95x scaled)
         461474  branch-loads              (   6.52x scaled)
           7493  branch-load-misses        (  26.57x scaled)

Reported-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <1246051927.2988.10.camel@hpdv5.satnam>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-stat.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 52c176cc683e..3840a70f05b7 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -32,6 +32,7 @@
  *   Wu Fengguang <fengguang.wu@intel.com>
  *   Mike Galbraith <efault@gmx.de>
  *   Paul Mackerras <paulus@samba.org>
+ *   Jaswinder Singh Rajput <jaswinder@kernel.org>
  *
  * Released under the GPL v2. (and only v2, not any later version)
  */
@@ -251,7 +252,7 @@ static void nsec_printout(int counter, u64 *count, u64 *noise)
 {
 	double msecs = (double)count[0] / 1000000;
 
-	fprintf(stderr, " %14.6f  %-20s", msecs, event_name(counter));
+	fprintf(stderr, " %14.6f  %-24s", msecs, event_name(counter));
 
 	if (attrs[counter].type == PERF_TYPE_SOFTWARE &&
 		attrs[counter].config == PERF_COUNT_SW_TASK_CLOCK) {
@@ -265,7 +266,7 @@ static void nsec_printout(int counter, u64 *count, u64 *noise)
 
 static void abs_printout(int counter, u64 *count, u64 *noise)
 {
-	fprintf(stderr, " %14Ld  %-20s", count[0], event_name(counter));
+	fprintf(stderr, " %14Ld  %-24s", count[0], event_name(counter));
 
 	if (runtime_cycles_avg &&
 		attrs[counter].type == PERF_TYPE_HARDWARE &&
@@ -295,7 +296,7 @@ static void print_counter(int counter)
 	scaled = event_scaled_avg[counter];
 
 	if (scaled == -1) {
-		fprintf(stderr, " %14s  %-20s\n",
+		fprintf(stderr, " %14s  %-24s\n",
 			"<not counted>", event_name(counter));
 		return;
 	}
@@ -306,8 +307,7 @@ static void print_counter(int counter)
 		abs_printout(counter, count, noise);
 
 	if (scaled)
-		fprintf(stderr, "  (scaled from %.2f%%)",
-			(double) count[2] / count[1] * 100);
+		fprintf(stderr, "  (%7.2fx scaled)", (double)count[1]/count[2]);
 
 	fprintf(stderr, "\n");
 }
@@ -421,7 +421,6 @@ static void print_stat(int argc, const char **argv)
 	for (counter = 0; counter < nr_counters; counter++)
 		print_counter(counter);
 
-
 	fprintf(stderr, "\n");
 	fprintf(stderr, " %14.9f  seconds time elapsed",
 			(double)walltime_nsecs_avg/1e9);
-- 
cgit v1.2.3-59-g8ed1b


From a222ad1a4b2e3ca177a538482c99c519c1ce94d1 Mon Sep 17 00:00:00 2001
From: Karen Xie <kxie@chelsio.com>
Date: Fri, 26 Jun 2009 15:17:29 -0700
Subject: [SCSI] cxgb3i: fix connection error when vlan is enabled

There is a bug when VLAN is configured on the cxgb3 interface, the iscsi
conn. would be denied with message "cxgb3i: NOT going through cxgbi device."

This patch adds code to get the real egress net_device when vlan is configured.

Signed-off-by: Karen Xie <kxie@chelsio.com>
Reviewed-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
---
 drivers/scsi/cxgb3i/cxgb3i_iscsi.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/scsi/cxgb3i/cxgb3i_iscsi.c b/drivers/scsi/cxgb3i/cxgb3i_iscsi.c
index 74369a3f963b..c399f485aa7d 100644
--- a/drivers/scsi/cxgb3i/cxgb3i_iscsi.c
+++ b/drivers/scsi/cxgb3i/cxgb3i_iscsi.c
@@ -13,6 +13,7 @@
 
 #include <linux/inet.h>
 #include <linux/crypto.h>
+#include <linux/if_vlan.h>
 #include <net/dst.h>
 #include <net/tcp.h>
 #include <scsi/scsi_cmnd.h>
@@ -184,6 +185,9 @@ static struct cxgb3i_hba *cxgb3i_hba_find_by_netdev(struct net_device *ndev)
 	struct cxgb3i_adapter *snic;
 	int i;
 
+	if (ndev->priv_flags & IFF_802_1Q_VLAN)
+		ndev = vlan_dev_real_dev(ndev);
+
 	read_lock(&cxgb3i_snic_rwlock);
 	list_for_each_entry(snic, &cxgb3i_snic_list, list_head) {
 		for (i = 0; i < snic->hba_cnt; i++) {
-- 
cgit v1.2.3-59-g8ed1b


From f0a71eb820596bd8f6abf64beb4cb181edaa2341 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Sat, 27 Jun 2009 07:04:55 -0400
Subject: cifs: fix fh_mutex locking in cifs_reopen_file

Fixes a regression caused by commit a6ce4932fbdbcd8f8e8c6df76812014351c32892

When this lock was converted to a mutex, the locks were turned into
unlocks and vice-versa.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Acked-by: Shirish Pargaonkar <shirishp@us.ibm.com>
Cc: Stable Tree <stable@kernel.org>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/file.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index ebdbe62a829c..97ce4bf89d15 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -493,9 +493,9 @@ static int cifs_reopen_file(struct file *file, bool can_flush)
 		return -EBADF;
 
 	xid = GetXid();
-	mutex_unlock(&pCifsFile->fh_mutex);
+	mutex_lock(&pCifsFile->fh_mutex);
 	if (!pCifsFile->invalidHandle) {
-		mutex_lock(&pCifsFile->fh_mutex);
+		mutex_unlock(&pCifsFile->fh_mutex);
 		rc = 0;
 		FreeXid(xid);
 		return rc;
@@ -527,7 +527,7 @@ static int cifs_reopen_file(struct file *file, bool can_flush)
 	if (full_path == NULL) {
 		rc = -ENOMEM;
 reopen_error_exit:
-		mutex_lock(&pCifsFile->fh_mutex);
+		mutex_unlock(&pCifsFile->fh_mutex);
 		FreeXid(xid);
 		return rc;
 	}
@@ -569,14 +569,14 @@ reopen_error_exit:
 			 cifs_sb->local_nls, cifs_sb->mnt_cifs_flags &
 				CIFS_MOUNT_MAP_SPECIAL_CHR);
 	if (rc) {
-		mutex_lock(&pCifsFile->fh_mutex);
+		mutex_unlock(&pCifsFile->fh_mutex);
 		cFYI(1, ("cifs_open returned 0x%x", rc));
 		cFYI(1, ("oplock: %d", oplock));
 	} else {
 reopen_success:
 		pCifsFile->netfid = netfid;
 		pCifsFile->invalidHandle = false;
-		mutex_lock(&pCifsFile->fh_mutex);
+		mutex_unlock(&pCifsFile->fh_mutex);
 		pCifsInode = CIFS_I(inode);
 		if (pCifsInode) {
 			if (can_flush) {
-- 
cgit v1.2.3-59-g8ed1b


From ff8a4bae459a9b6455504127fcb78fdbc8e50e4c Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Sat, 27 Jun 2009 12:22:27 -0700
Subject: Revert "x86: cap iomem_resource to addressable physical memory"

This reverts commit 95ee14e4379c5e19c0897c872350570402014742.
Mikael Petterson <mikepe@it.uu.se> reported that at least one of his
systems will not boot as a result.  We have ruled out the detection
algorithm malfunctioning, so it is not a matter of producing the
incorrect bitmasks; rather, something in the application of them
fails.

Revert the commit until we can root cause and correct this problem.

-stable team: this means the underlying commit should be rejected.

Reported-and-isolated-by: Mikael Petterson <mikpe@it.uu.se>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
LKML-Reference: <200906261559.n5QFxJH8027336@pilspetsen.it.uu.se>
Cc: stable@kernel.org
Cc: Grant Grundler <grundler@parisc-linux.org>
---
 arch/x86/kernel/cpu/common.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 6b26d4deada0..f1961c07af9a 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -848,9 +848,6 @@ static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 #if defined(CONFIG_NUMA) && defined(CONFIG_X86_64)
 	numa_add_cpu(smp_processor_id());
 #endif
-
-	/* Cap the iomem address space to what is addressable on all CPUs */
-	iomem_resource.end &= (1ULL << c->x86_phys_bits) - 1;
 }
 
 #ifdef CONFIG_X86_64
-- 
cgit v1.2.3-59-g8ed1b


From c276aca46d26aa2347320096f8ecdf5016795c14 Mon Sep 17 00:00:00 2001
From: vimal singh <vimalsingh@ti.com>
Date: Sat, 27 Jun 2009 11:07:06 +0530
Subject: mtd: nand: fix build failure and incorrect return from omap_wait()

We need to include jiffies.h manually in some cases, and the status
returned from omap_wait() was broken in two separate ways.

Also add cond_resched() to the loop.

Signed-off-by: Vimal Singh <vimalsingh@ti.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/omap2.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/omap2.c b/drivers/mtd/nand/omap2.c
index 0cd76f89f4b0..ebd07e95b814 100644
--- a/drivers/mtd/nand/omap2.c
+++ b/drivers/mtd/nand/omap2.c
@@ -11,6 +11,8 @@
 #include <linux/platform_device.h>
 #include <linux/dma-mapping.h>
 #include <linux/delay.h>
+#include <linux/jiffies.h>
+#include <linux/sched.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/nand.h>
 #include <linux/mtd/partitions.h>
@@ -541,7 +543,7 @@ static int omap_wait(struct mtd_info *mtd, struct nand_chip *chip)
 	struct omap_nand_info *info = container_of(mtd, struct omap_nand_info,
 							mtd);
 	unsigned long timeo = jiffies;
-	int status, state = this->state;
+	int status = NAND_STATUS_FAIL, state = this->state;
 
 	if (state == FL_ERASING)
 		timeo += (HZ * 400) / 1000;
@@ -556,8 +558,9 @@ static int omap_wait(struct mtd_info *mtd, struct nand_chip *chip)
 
 	while (time_before(jiffies, timeo)) {
 		status = __raw_readb(this->IO_ADDR_R);
-		if (!(status & 0x40))
+		if (status & NAND_STATUS_READY)
 			break;
+		cond_resched();
 	}
 	return status;
 }
-- 
cgit v1.2.3-59-g8ed1b


From e244584fe3a5c20deddeca246548ac86dbc6e1d1 Mon Sep 17 00:00:00 2001
From: Izik Eidus <ieidus@redhat.com>
Date: Wed, 10 Jun 2009 19:23:24 +0300
Subject: KVM: Fix dirty bit tracking for slots with large pages

When slot is already allocated and being asked to be tracked we need
to break the large pages.

This code flush the mmu when someone ask a slot to start dirty bit
tracking.

Cc: stable@kernel.org
Signed-off-by: Izik Eidus <ieidus@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 virt/kvm/kvm_main.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 764554350ed8..013a5b3e9f75 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -1194,6 +1194,8 @@ int __kvm_set_memory_region(struct kvm *kvm,
 		if (!new.dirty_bitmap)
 			goto out_free;
 		memset(new.dirty_bitmap, 0, dirty_bytes);
+		if (old.npages)
+			kvm_arch_flush_shadow(kvm);
 	}
 #endif /* not defined CONFIG_S390 */
 
-- 
cgit v1.2.3-59-g8ed1b


From 29a4b9333bf9ffef12b3dd7cbf2e3dbe01152968 Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@redhat.com>
Date: Tue, 19 May 2009 13:29:27 +0300
Subject: KVM: MMU: Allow 4K ptes with bit 7 (PAT) set

Bit 7 is perfectly legal in the 4K page leve; it is used for the PAT.

Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/x86/kvm/mmu.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 5c3d6e81a7dc..7030b5f911bf 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -2157,7 +2157,7 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, int level)
 		else
 			/* 32 bits PSE 4MB page */
 			context->rsvd_bits_mask[1][1] = rsvd_bits(13, 21);
-		context->rsvd_bits_mask[1][0] = ~0ull;
+		context->rsvd_bits_mask[1][0] = context->rsvd_bits_mask[1][0];
 		break;
 	case PT32E_ROOT_LEVEL:
 		context->rsvd_bits_mask[0][2] =
@@ -2170,7 +2170,7 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, int level)
 		context->rsvd_bits_mask[1][1] = exb_bit_rsvd |
 			rsvd_bits(maxphyaddr, 62) |
 			rsvd_bits(13, 20);		/* large page */
-		context->rsvd_bits_mask[1][0] = ~0ull;
+		context->rsvd_bits_mask[1][0] = context->rsvd_bits_mask[1][0];
 		break;
 	case PT64_ROOT_LEVEL:
 		context->rsvd_bits_mask[0][3] = exb_bit_rsvd |
@@ -2186,7 +2186,7 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, int level)
 		context->rsvd_bits_mask[1][1] = exb_bit_rsvd |
 			rsvd_bits(maxphyaddr, 51) |
 			rsvd_bits(13, 20);		/* large page */
-		context->rsvd_bits_mask[1][0] = ~0ull;
+		context->rsvd_bits_mask[1][0] = context->rsvd_bits_mask[1][0];
 		break;
 	}
 }
-- 
cgit v1.2.3-59-g8ed1b


From 84261923d3dddb766736023bead6fa07b7e218d5 Mon Sep 17 00:00:00 2001
From: Marcelo Tosatti <mtosatti@redhat.com>
Date: Wed, 17 Jun 2009 10:53:47 -0300
Subject: KVM: protect concurrent make_all_cpus_request

make_all_cpus_request contains a race condition which can
trigger false request completed status, as follows:

CPU0                                              CPU1

if (test_and_set_bit(req,&vcpu->requests))
   ....                                        	   if (test_and_set_bit(req,&vcpu->requests))
   ..                                                  return
proceed to smp_call_function_many(wait=1)

Use a spinlock to serialize concurrent CPUs.

Cc: stable@kernel.org
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 include/linux/kvm_host.h | 1 +
 virt/kvm/kvm_main.c      | 3 +++
 2 files changed, 4 insertions(+)

diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index aacc5449f586..16713dc672e4 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -125,6 +125,7 @@ struct kvm_kernel_irq_routing_entry {
 struct kvm {
 	struct mutex lock; /* protects the vcpus array and APIC accesses */
 	spinlock_t mmu_lock;
+	spinlock_t requests_lock;
 	struct rw_semaphore slots_lock;
 	struct mm_struct *mm; /* userspace tied to this vm */
 	int nmemslots;
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 013a5b3e9f75..2884baf1d5f9 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -746,6 +746,7 @@ static bool make_all_cpus_request(struct kvm *kvm, unsigned int req)
 		cpumask_clear(cpus);
 
 	me = get_cpu();
+	spin_lock(&kvm->requests_lock);
 	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
 		vcpu = kvm->vcpus[i];
 		if (!vcpu)
@@ -762,6 +763,7 @@ static bool make_all_cpus_request(struct kvm *kvm, unsigned int req)
 		smp_call_function_many(cpus, ack_flush, NULL, 1);
 	else
 		called = false;
+	spin_unlock(&kvm->requests_lock);
 	put_cpu();
 	free_cpumask_var(cpus);
 	return called;
@@ -982,6 +984,7 @@ static struct kvm *kvm_create_vm(void)
 	kvm->mm = current->mm;
 	atomic_inc(&kvm->mm->mm_count);
 	spin_lock_init(&kvm->mmu_lock);
+	spin_lock_init(&kvm->requests_lock);
 	kvm_io_bus_init(&kvm->pio_bus);
 	mutex_init(&kvm->lock);
 	kvm_io_bus_init(&kvm->mmio_bus);
-- 
cgit v1.2.3-59-g8ed1b


From ffdfa071bd6fa8f5e5964569bef41e067540d424 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <jes@sgi.com>
Date: Wed, 17 Jun 2009 11:08:08 +0200
Subject: KVM: ia64: fix ia64 build due to missing kallsyms_lookup() and double
 export

Fix problem with double export of certain symbols from vsprintf.c
which we do not wish to export from the kvm-intel.ko module.

In addition, we do not have access to kallsyms_lookup() from the
module, so make sure to #undef CONFIG_KALLSYMS

Signed-off-by: Jes Sorensen <jes@sgi.com>
Acked-by: Xiantao Zhang <xiantao.zhang@intel.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/ia64/kvm/kvm_lib.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/arch/ia64/kvm/kvm_lib.c b/arch/ia64/kvm/kvm_lib.c
index a85cb611ecd7..f1268b8e6f9e 100644
--- a/arch/ia64/kvm/kvm_lib.c
+++ b/arch/ia64/kvm/kvm_lib.c
@@ -11,5 +11,11 @@
  *
  */
 #undef CONFIG_MODULES
+#include <linux/module.h>
+#undef CONFIG_KALLSYMS
+#undef EXPORT_SYMBOL
+#undef EXPORT_SYMBOL_GPL
+#define EXPORT_SYMBOL(sym)
+#define EXPORT_SYMBOL_GPL(sym)
 #include "../../../lib/vsprintf.c"
 #include "../../../lib/ctype.c"
-- 
cgit v1.2.3-59-g8ed1b


From a3f9d3981cd82d65232b733eb792382237d686bd Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Thu, 18 Jun 2009 16:53:25 +0530
Subject: KVM: kvm/x86_emulate.c toggle_interruptibility() should be static

toggle_interruptibility() is used only by same file, it should be static.

Fixed following sparse warning :

  arch/x86/kvm/x86_emulate.c:1364:6: warning: symbol 'toggle_interruptibility' was not declared. Should it be static?

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/x86/kvm/x86_emulate.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kvm/x86_emulate.c b/arch/x86/kvm/x86_emulate.c
index c1b6c232e02b..616de4628d60 100644
--- a/arch/x86/kvm/x86_emulate.c
+++ b/arch/x86/kvm/x86_emulate.c
@@ -1361,7 +1361,7 @@ static inline int writeback(struct x86_emulate_ctxt *ctxt,
 	return 0;
 }
 
-void toggle_interruptibility(struct x86_emulate_ctxt *ctxt, u32 mask)
+static void toggle_interruptibility(struct x86_emulate_ctxt *ctxt, u32 mask)
 {
 	u32 int_shadow = kvm_x86_ops->get_interrupt_shadow(ctxt->vcpu, mask);
 	/*
-- 
cgit v1.2.3-59-g8ed1b


From ef50f7ac7e234f9696555e41eab3de69c3d86166 Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <borntraeger@de.ibm.com>
Date: Tue, 23 Jun 2009 17:24:07 +0200
Subject: KVM: s390: Allow stfle instruction in the guest

2.6.31-rc introduced an architecture level set checker based on facility
bits. e.g. if the kernel is compiled to run only on z9, several facility
bits are checked very early and the kernel refuses to boot if a z9 specific
facility is missing.
Until now kvm on s390 did not implement the store facility extended (STFLE)
instruction. A 2.6.31-rc kernel that was compiled for z9 or higher did not
boot in kvm. This patch implements stfle.

This patch should go in before 2.6.31.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/s390/include/asm/kvm_host.h |  4 +++-
 arch/s390/kvm/kvm-s390.c         | 23 ++++++++++++++++++++++-
 arch/s390/kvm/priv.c             |  2 +-
 3 files changed, 26 insertions(+), 3 deletions(-)

diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index a27d0d5a6f86..1cd02f6073a0 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -99,7 +99,9 @@ struct kvm_s390_sie_block {
 	__u8	reservedd0[48];		/* 0x00d0 */
 	__u64	gcr[16];		/* 0x0100 */
 	__u64	gbea;			/* 0x0180 */
-	__u8	reserved188[120];	/* 0x0188 */
+	__u8	reserved188[24];	/* 0x0188 */
+	__u32	fac;			/* 0x01a0 */
+	__u8	reserved1a4[92];	/* 0x01a4 */
 } __attribute__((packed));
 
 struct kvm_vcpu_stat {
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index c18b21d6991c..90d9d1ba258b 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -25,6 +25,7 @@
 #include <asm/lowcore.h>
 #include <asm/pgtable.h>
 #include <asm/nmi.h>
+#include <asm/system.h>
 #include "kvm-s390.h"
 #include "gaccess.h"
 
@@ -69,6 +70,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ NULL }
 };
 
+static unsigned long long *facilities;
 
 /* Section: not file related */
 void kvm_arch_hardware_enable(void *garbage)
@@ -288,6 +290,7 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 	vcpu->arch.sie_block->gmsor = vcpu->kvm->arch.guest_origin;
 	vcpu->arch.sie_block->ecb   = 2;
 	vcpu->arch.sie_block->eca   = 0xC1002001U;
+	vcpu->arch.sie_block->fac   = (int) (long) facilities;
 	hrtimer_init(&vcpu->arch.ckc_timer, CLOCK_REALTIME, HRTIMER_MODE_ABS);
 	tasklet_init(&vcpu->arch.tasklet, kvm_s390_tasklet,
 		     (unsigned long) vcpu);
@@ -739,11 +742,29 @@ gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn)
 
 static int __init kvm_s390_init(void)
 {
-	return kvm_init(NULL, sizeof(struct kvm_vcpu), THIS_MODULE);
+	int ret;
+	ret = kvm_init(NULL, sizeof(struct kvm_vcpu), THIS_MODULE);
+	if (ret)
+		return ret;
+
+	/*
+	 * guests can ask for up to 255+1 double words, we need a full page
+	 * to hold the maximum amount of facilites. On the other hand, we
+	 * only set facilities that are known to work in KVM.
+	 */
+	facilities = (unsigned long long *) get_zeroed_page(GFP_DMA);
+	if (!facilities) {
+		kvm_exit();
+		return -ENOMEM;
+	}
+	stfle(facilities, 1);
+	facilities[0] &= 0xff00fff3f0700000ULL;
+	return 0;
 }
 
 static void __exit kvm_s390_exit(void)
 {
+	free_page((unsigned long) facilities);
 	kvm_exit();
 }
 
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index 93ecd06e1a74..d426aac8095d 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -158,7 +158,7 @@ static int handle_stfl(struct kvm_vcpu *vcpu)
 
 	vcpu->stat.instruction_stfl++;
 	/* only pass the facility bits, which we can handle */
-	facility_list &= 0xfe00fff3;
+	facility_list &= 0xff00fff3;
 
 	rc = copy_to_guest(vcpu, offsetof(struct _lowcore, stfl_fac_list),
 			   &facility_list, sizeof(facility_list));
-- 
cgit v1.2.3-59-g8ed1b


From e3c7cb6ad7191e92ba89d00a7ae5f5dd1ca0c214 Mon Sep 17 00:00:00 2001
From: Avi Kivity <avi@redhat.com>
Date: Tue, 16 Jun 2009 14:19:52 +0300
Subject: KVM: VMX: Handle vmx instruction vmexits

IF a guest tries to use vmx instructions, inject a #UD to let it know the
instruction is not implemented, rather than crashing.

This prevents guest userspace from crashing the guest kernel.

Cc: stable@kernel.org
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/x86/kvm/vmx.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index e770bf349ec4..356a0ce85c68 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -3012,6 +3012,12 @@ static int handle_vmcall(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	return 1;
 }
 
+static int handle_vmx_insn(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
+{
+	kvm_queue_exception(vcpu, UD_VECTOR);
+	return 1;
+}
+
 static int handle_invlpg(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 {
 	unsigned long exit_qualification = vmcs_readl(EXIT_QUALIFICATION);
@@ -3198,6 +3204,15 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu,
 	[EXIT_REASON_HLT]                     = handle_halt,
 	[EXIT_REASON_INVLPG]		      = handle_invlpg,
 	[EXIT_REASON_VMCALL]                  = handle_vmcall,
+	[EXIT_REASON_VMCLEAR]	              = handle_vmx_insn,
+	[EXIT_REASON_VMLAUNCH]                = handle_vmx_insn,
+	[EXIT_REASON_VMPTRLD]                 = handle_vmx_insn,
+	[EXIT_REASON_VMPTRST]                 = handle_vmx_insn,
+	[EXIT_REASON_VMREAD]                  = handle_vmx_insn,
+	[EXIT_REASON_VMRESUME]                = handle_vmx_insn,
+	[EXIT_REASON_VMWRITE]                 = handle_vmx_insn,
+	[EXIT_REASON_VMOFF]                   = handle_vmx_insn,
+	[EXIT_REASON_VMON]                    = handle_vmx_insn,
 	[EXIT_REASON_TPR_BELOW_THRESHOLD]     = handle_tpr_below_threshold,
 	[EXIT_REASON_APIC_ACCESS]             = handle_apic_access,
 	[EXIT_REASON_WBINVD]                  = handle_wbinvd,
-- 
cgit v1.2.3-59-g8ed1b


From 9e6996240afcbe61682eab8eeaeb65c34333164d Mon Sep 17 00:00:00 2001
From: Amit Shah <amit.shah@redhat.com>
Date: Mon, 15 Jun 2009 13:25:34 +0530
Subject: KVM: Ignore reads to K7 EVNTSEL MSRs

In commit 7fe29e0faacb650d31b9e9f538203a157bec821d we ignored the
reads to the P6 EVNTSEL MSRs. That fixed crashes on Intel machines.

Ignore the reads to K7 EVNTSEL MSRs as well to fix this on AMD
hosts.

This fixes Kaspersky antivirus crashing Windows guests on AMD hosts.

Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/x86/kvm/x86.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 249540f98513..fe5474aec41a 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -898,6 +898,7 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
 	case MSR_VM_HSAVE_PA:
 	case MSR_P6_EVNTSEL0:
 	case MSR_P6_EVNTSEL1:
+	case MSR_K7_EVNTSEL0:
 		data = 0;
 		break;
 	case MSR_MTRRcap:
-- 
cgit v1.2.3-59-g8ed1b


From bde892232532ed522bb56b04576d07f91e59b3c7 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Wed, 20 May 2009 09:59:35 +0530
Subject: KVM: shut up uninit compiler warning in paging_tmpl.h
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Dixes compilation warning:
  CC      arch/x86/kernel/io_delay.o
 arch/x86/kvm/paging_tmpl.h: In function ‘paging64_fetch’:
 arch/x86/kvm/paging_tmpl.h:279: warning: ‘sptep’ may be used uninitialized in this function
 arch/x86/kvm/paging_tmpl.h: In function ‘paging32_fetch’:
 arch/x86/kvm/paging_tmpl.h:279: warning: ‘sptep’ may be used uninitialized in this function

warning is bogus (always have a least one level), but need to shut the compiler
up.

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/x86/kvm/paging_tmpl.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 258e4591e1ca..67785f635399 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -281,7 +281,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr,
 {
 	unsigned access = gw->pt_access;
 	struct kvm_mmu_page *shadow_page;
-	u64 spte, *sptep;
+	u64 spte, *sptep = NULL;
 	int direct;
 	gfn_t table_gfn;
 	int r;
-- 
cgit v1.2.3-59-g8ed1b


From c3043569dc8fbe9228b76174f15d1a7152c48a20 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
Date: Sat, 27 Jun 2009 23:49:09 +0530
Subject: perf stat: Micro-optimize the code: memcpy is only required if no
 event is selected and !null_run

Set attrs and nr_counters if no event is selected and !null_run.

Setting of attrs should depend on number of counters,
so we need to memcpy only for sizeof(default_attrs)

Also set nr_counters as ARRAY_SIZE(default_attrs) in place of
hardcoded value.

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <1246126749.32198.16.camel@hpdv5.satnam>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-stat.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 3840a70f05b7..3e5ea4e2e5fd 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -46,7 +46,7 @@
 #include <sys/prctl.h>
 #include <math.h>
 
-static struct perf_counter_attr default_attrs[MAX_COUNTERS] = {
+static struct perf_counter_attr default_attrs[] = {
 
   { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK	},
   { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CONTEXT_SWITCHES},
@@ -477,16 +477,17 @@ int cmd_stat(int argc, const char **argv, const char *prefix)
 {
 	int status;
 
-	memcpy(attrs, default_attrs, sizeof(attrs));
-
 	argc = parse_options(argc, argv, options, stat_usage, 0);
 	if (!argc)
 		usage_with_options(stat_usage, options);
 	if (run_count <= 0 || run_count > MAX_RUN)
 		usage_with_options(stat_usage, options);
 
-	if (!null_run && !nr_counters)
-		nr_counters = 8;
+	/* Set attrs and nr_counters if no event is selected and !null_run */
+	if (!null_run && !nr_counters) {
+		memcpy(attrs, default_attrs, sizeof(default_attrs));
+		nr_counters = ARRAY_SIZE(default_attrs);
+	}
 
 	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
 	assert(nr_cpus <= MAX_NR_CPUS);
-- 
cgit v1.2.3-59-g8ed1b


From 46690f3718d95e9bb712b6f2b5c869f8494521de Mon Sep 17 00:00:00 2001
From: Mimi Zohar <zohar@linux.vnet.ibm.com>
Date: Fri, 26 Jun 2009 11:24:05 -0400
Subject: integrity: ima mq_open imbalance msg fix

This patch fixes an imbalance message as reported by Sanchin Sant.
As we don't need to measure the message queue, just increment the
counters.

Reported-by: Sanchin Sant <sanchinp@in.ibm.com>
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 ipc/mqueue.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index e35ba2c3a8d7..c5e68adc6732 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -32,6 +32,7 @@
 #include <linux/nsproxy.h>
 #include <linux/pid.h>
 #include <linux/ipc_namespace.h>
+#include <linux/ima.h>
 
 #include <net/sock.h>
 #include "util.h"
@@ -733,6 +734,7 @@ SYSCALL_DEFINE4(mq_open, const char __user *, u_name, int, oflag, mode_t, mode,
 		error = PTR_ERR(filp);
 		goto out_putfd;
 	}
+	ima_counts_get(filp);
 
 	fd_install(fd, filp);
 	goto out_upsem;
-- 
cgit v1.2.3-59-g8ed1b


From 79b854c549c62c54fa27f87e04465c01db889f8d Mon Sep 17 00:00:00 2001
From: Mimi Zohar <zohar@linux.vnet.ibm.com>
Date: Fri, 26 Jun 2009 11:25:00 -0400
Subject: integrity: ima audit hash_exists fix

Audit the file name, not the template name.

Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 security/integrity/ima/ima_queue.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/security/integrity/ima/ima_queue.c b/security/integrity/ima/ima_queue.c
index 7ec94314ac0c..a0880e9c8e05 100644
--- a/security/integrity/ima/ima_queue.c
+++ b/security/integrity/ima/ima_queue.c
@@ -134,7 +134,8 @@ int ima_add_template_entry(struct ima_template_entry *entry, int violation,
 	}
 out:
 	mutex_unlock(&ima_extend_list_mutex);
-	integrity_audit_msg(AUDIT_INTEGRITY_PCR, inode, entry->template_name,
+	integrity_audit_msg(AUDIT_INTEGRITY_PCR, inode,
+			    entry->template.file_name,
 			    op, audit_cause, result, audit_info);
 	return result;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 94e5d714f604d4cb4cb13163f01ede278e69258b Mon Sep 17 00:00:00 2001
From: Mimi Zohar <zohar@linux.vnet.ibm.com>
Date: Fri, 26 Jun 2009 14:05:27 -0400
Subject: integrity: add ima_counts_put (updated)

This patch fixes an imbalance message as reported by J.R. Okajima.
The IMA file counters are incremented in ima_path_check. If the
actual open fails, such as ETXTBSY, decrement the counters to
prevent unnecessary imbalance messages.

Reported-by: J.R. Okajima <hooanon05@yahoo.co.jp>
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: James Morris <jmorris@namei.org>
---
 fs/namei.c                        |  7 +++++++
 include/linux/ima.h               |  6 ++++++
 security/integrity/ima/ima_main.c | 29 ++++++++++++++++++++++++++++-
 3 files changed, 41 insertions(+), 1 deletion(-)

diff --git a/fs/namei.c b/fs/namei.c
index 5b961eb71cbf..f3c5b278895a 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1761,6 +1761,10 @@ do_last:
 			goto exit;
 		}
 		filp = nameidata_to_filp(&nd, open_flag);
+		if (IS_ERR(filp))
+			ima_counts_put(&nd.path,
+				       acc_mode & (MAY_READ | MAY_WRITE |
+						   MAY_EXEC));
 		mnt_drop_write(nd.path.mnt);
 		if (nd.root.mnt)
 			path_put(&nd.root);
@@ -1817,6 +1821,9 @@ ok:
 		goto exit;
 	}
 	filp = nameidata_to_filp(&nd, open_flag);
+	if (IS_ERR(filp))
+		ima_counts_put(&nd.path,
+			       acc_mode & (MAY_READ | MAY_WRITE | MAY_EXEC));
 	/*
 	 * It is now safe to drop the mnt write
 	 * because the filp has had a write taken
diff --git a/include/linux/ima.h b/include/linux/ima.h
index b1b827d091a9..0e3f2a4c25f6 100644
--- a/include/linux/ima.h
+++ b/include/linux/ima.h
@@ -24,6 +24,7 @@ extern int ima_path_check(struct path *path, int mask, int update_counts);
 extern void ima_file_free(struct file *file);
 extern int ima_file_mmap(struct file *file, unsigned long prot);
 extern void ima_counts_get(struct file *file);
+extern void ima_counts_put(struct path *path, int mask);
 
 #else
 static inline int ima_bprm_check(struct linux_binprm *bprm)
@@ -60,5 +61,10 @@ static inline void ima_counts_get(struct file *file)
 {
 	return;
 }
+
+static inline void ima_counts_put(struct path *path, int mask)
+{
+	return;
+}
 #endif /* CONFIG_IMA_H */
 #endif /* _LINUX_IMA_H */
diff --git a/security/integrity/ima/ima_main.c b/security/integrity/ima/ima_main.c
index 6f611874d10e..101c512564ec 100644
--- a/security/integrity/ima/ima_main.c
+++ b/security/integrity/ima/ima_main.c
@@ -238,7 +238,34 @@ out:
 }
 
 /*
- * ima_opens_get - increment file counts
+ * ima_counts_put - decrement file counts
+ *
+ * File counts are incremented in ima_path_check. On file open
+ * error, such as ETXTBSY, decrement the counts to prevent
+ * unnecessary imbalance messages.
+ */
+void ima_counts_put(struct path *path, int mask)
+{
+	struct inode *inode = path->dentry->d_inode;
+	struct ima_iint_cache *iint;
+
+	if (!ima_initialized || !S_ISREG(inode->i_mode))
+		return;
+	iint = ima_iint_find_insert_get(inode);
+	if (!iint)
+		return;
+
+	mutex_lock(&iint->mutex);
+	iint->opencount--;
+	if ((mask & MAY_WRITE) || (mask == 0))
+		iint->writecount--;
+	else if (mask & (MAY_READ | MAY_EXEC))
+		iint->readcount--;
+	mutex_unlock(&iint->mutex);
+}
+
+/*
+ * ima_counts_get - increment file counts
  *
  * - for IPC shm and shmat file.
  * - for nfsd exported files.
-- 
cgit v1.2.3-59-g8ed1b


From bd46cb6cf11867130a41ea9546dd65688b71f3c2 Mon Sep 17 00:00:00 2001
From: Ajit Khaparde <ajitk@serverengines.com>
Date: Fri, 26 Jun 2009 02:51:07 +0000
Subject: be2net: Fix to avoid a crash seen on PPC with LRO and Jumbo frames.

While testing the driver on PPC, we ran into a crash with LRO, Jumbo frames.
With CONFIG_PPC_64K_PAGES configured (a default in PPC), MAX_SKB_FRAGS drops to 3 and we were crossing the array limits on skb_shinfo(skb)->frags[].
Now we coalesce the frags from the same physical page into one slot in
skb_shinfo(skb)->frags[] and go to the next index when the frag is from

different physical page.

This patch is against the net-2.6 tree.

Signed-off-by: Ajit Khaparde <ajitk@serverengines.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/benet/be.h         |  2 +-
 drivers/net/benet/be_ethtool.c |  4 ++--
 drivers/net/benet/be_main.c    | 45 ++++++++++++++++++++++++++++++------------
 3 files changed, 35 insertions(+), 16 deletions(-)

diff --git a/drivers/net/benet/be.h b/drivers/net/benet/be.h
index f703758f0a6e..5b4bf3d2cdc2 100644
--- a/drivers/net/benet/be.h
+++ b/drivers/net/benet/be.h
@@ -73,7 +73,7 @@ static inline char *nic_name(struct pci_dev *pdev)
 #define RX_FRAGS_REFILL_WM	(RX_Q_LEN - MAX_RX_POST)
 
 #define BE_MAX_LRO_DESCRIPTORS  16
-#define BE_MAX_FRAGS_PER_FRAME  16
+#define BE_MAX_FRAGS_PER_FRAME  (min((u32) 16, (u32) MAX_SKB_FRAGS))
 
 struct be_dma_mem {
 	void *va;
diff --git a/drivers/net/benet/be_ethtool.c b/drivers/net/benet/be_ethtool.c
index 9592f22e4c8c..cccc5419ad72 100644
--- a/drivers/net/benet/be_ethtool.c
+++ b/drivers/net/benet/be_ethtool.c
@@ -162,8 +162,8 @@ be_set_coalesce(struct net_device *netdev, struct ethtool_coalesce *coalesce)
 		return -EINVAL;
 
 	adapter->max_rx_coal = coalesce->rx_max_coalesced_frames;
-	if (adapter->max_rx_coal > MAX_SKB_FRAGS)
-		adapter->max_rx_coal = MAX_SKB_FRAGS - 1;
+	if (adapter->max_rx_coal > BE_MAX_FRAGS_PER_FRAME)
+		adapter->max_rx_coal = BE_MAX_FRAGS_PER_FRAME;
 
 	/* if AIC is being turned on now, start with an EQD of 0 */
 	if (rx_eq->enable_aic == 0 &&
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index 66c10c87f517..308eb09ca56b 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -666,7 +666,7 @@ static void skb_fill_rx_data(struct be_adapter *adapter,
 {
 	struct be_queue_info *rxq = &adapter->rx_obj.q;
 	struct be_rx_page_info *page_info;
-	u16 rxq_idx, i, num_rcvd;
+	u16 rxq_idx, i, num_rcvd, j;
 	u32 pktsize, hdr_len, curr_frag_len;
 	u8 *start;
 
@@ -709,22 +709,33 @@ static void skb_fill_rx_data(struct be_adapter *adapter,
 
 	/* More frags present for this completion */
 	pktsize -= curr_frag_len; /* account for above copied frag */
-	for (i = 1; i < num_rcvd; i++) {
+	for (i = 1, j = 0; i < num_rcvd; i++) {
 		index_inc(&rxq_idx, rxq->len);
 		page_info = get_rx_page_info(adapter, rxq_idx);
 
 		curr_frag_len = min(pktsize, rx_frag_size);
 
-		skb_shinfo(skb)->frags[i].page = page_info->page;
-		skb_shinfo(skb)->frags[i].page_offset = page_info->page_offset;
-		skb_shinfo(skb)->frags[i].size = curr_frag_len;
+		/* Coalesce all frags from the same physical page in one slot */
+		if (page_info->page_offset == 0) {
+			/* Fresh page */
+			j++;
+			skb_shinfo(skb)->frags[j].page = page_info->page;
+			skb_shinfo(skb)->frags[j].page_offset =
+							page_info->page_offset;
+			skb_shinfo(skb)->frags[j].size = 0;
+			skb_shinfo(skb)->nr_frags++;
+		} else {
+			put_page(page_info->page);
+		}
+
+		skb_shinfo(skb)->frags[j].size += curr_frag_len;
 		skb->len += curr_frag_len;
 		skb->data_len += curr_frag_len;
-		skb_shinfo(skb)->nr_frags++;
 		pktsize -= curr_frag_len;
 
 		memset(page_info, 0, sizeof(*page_info));
 	}
+	BUG_ON(j > MAX_SKB_FRAGS);
 
 done:
 	be_rx_stats_update(adapter, pktsize, num_rcvd);
@@ -786,7 +797,7 @@ static void be_rx_compl_process_lro(struct be_adapter *adapter,
 	struct skb_frag_struct rx_frags[BE_MAX_FRAGS_PER_FRAME];
 	struct be_queue_info *rxq = &adapter->rx_obj.q;
 	u32 num_rcvd, pkt_size, remaining, vlanf, curr_frag_len;
-	u16 i, rxq_idx = 0, vid;
+	u16 i, rxq_idx = 0, vid, j;
 
 	num_rcvd = AMAP_GET_BITS(struct amap_eth_rx_compl, numfrags, rxcp);
 	pkt_size = AMAP_GET_BITS(struct amap_eth_rx_compl, pktsize, rxcp);
@@ -794,20 +805,28 @@ static void be_rx_compl_process_lro(struct be_adapter *adapter,
 	rxq_idx = AMAP_GET_BITS(struct amap_eth_rx_compl, fragndx, rxcp);
 
 	remaining = pkt_size;
-	for (i = 0; i < num_rcvd; i++) {
+	for (i = 0, j = -1; i < num_rcvd; i++) {
 		page_info = get_rx_page_info(adapter, rxq_idx);
 
 		curr_frag_len = min(remaining, rx_frag_size);
 
-		rx_frags[i].page = page_info->page;
-		rx_frags[i].page_offset = page_info->page_offset;
-		rx_frags[i].size = curr_frag_len;
-		remaining -= curr_frag_len;
+		/* Coalesce all frags from the same physical page in one slot */
+		if (i == 0 || page_info->page_offset == 0) {
+			/* First frag or Fresh page */
+			j++;
+			rx_frags[j].page = page_info->page;
+			rx_frags[j].page_offset = page_info->page_offset;
+			rx_frags[j].size = 0;
+		} else {
+			put_page(page_info->page);
+		}
+		rx_frags[j].size += curr_frag_len;
 
+		remaining -= curr_frag_len;
 		index_inc(&rxq_idx, rxq->len);
-
 		memset(page_info, 0, sizeof(*page_info));
 	}
+	BUG_ON(j > MAX_SKB_FRAGS);
 
 	if (likely(!vlanf)) {
 		lro_receive_frags(&adapter->rx_obj.lro_mgr, rx_frags, pkt_size,
-- 
cgit v1.2.3-59-g8ed1b


From 684a88429c5ab04d8b1894de9a1ef62de6f601b7 Mon Sep 17 00:00:00 2001
From: Tony Vroon <tony@linx.net>
Date: Fri, 26 Jun 2009 09:27:50 +0100
Subject: ALSA: hda - Line In for Acer Inspire 6530G model

The Line In connector is set up as PIN_IN by default, using
VREF_HIZ. It is connected to both ADCs, so add it to both
input selectors.
Also add the ability to use the input mix (on a SoundBlaster
one would call this "What You Hear").

Signed-off-by: Tony Vroon <tony@linx.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/hda/patch_realtek.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 2ed514030e75..08846d222cbe 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -1569,18 +1569,22 @@ static struct hda_input_mux alc888_2_capture_sources[2] = {
 static struct hda_input_mux alc888_acer_aspire_6530_sources[2] = {
 	/* Interal mic only available on one ADC */
 	{
-		.num_items = 3,
+		.num_items = 5,
 		.items = {
 			{ "Ext Mic", 0x0 },
+			{ "Line In", 0x2 },
 			{ "CD", 0x4 },
+			{ "Input Mix", 0xa },
 			{ "Int Mic", 0xb },
 		},
 	},
 	{
-		.num_items = 2,
+		.num_items = 4,
 		.items = {
 			{ "Ext Mic", 0x0 },
+			{ "Line In", 0x2 },
 			{ "CD", 0x4 },
+			{ "Input Mix", 0xa },
 		},
 	}
 };
@@ -8209,6 +8213,8 @@ static struct snd_kcontrol_new alc888_acer_aspire_6530_mixer[] = {
 	HDA_BIND_MUTE("Front Playback Switch", 0x0c, 2, HDA_INPUT),
 	HDA_CODEC_VOLUME("LFE Playback Volume", 0x0f, 0x0, HDA_OUTPUT),
 	HDA_BIND_MUTE("LFE Playback Switch", 0x0f, 2, HDA_INPUT),
+	HDA_CODEC_VOLUME("Line Playback Volume", 0x0b, 0x02, HDA_INPUT),
+	HDA_CODEC_MUTE("Line Playback Switch", 0x0b, 0x02, HDA_INPUT),
 	HDA_CODEC_VOLUME("CD Playback Volume", 0x0b, 0x04, HDA_INPUT),
 	HDA_CODEC_MUTE("CD Playback Switch", 0x0b, 0x04, HDA_INPUT),
 	HDA_CODEC_VOLUME("Mic Playback Volume", 0x0b, 0x0, HDA_INPUT),
-- 
cgit v1.2.3-59-g8ed1b


From c2a30d711852e4f39c8a79135b3caa701f7a8e02 Mon Sep 17 00:00:00 2001
From: Ondrej Zary <linux@rainbow-software.org>
Date: Sat, 27 Jun 2009 16:17:08 +0200
Subject: ALSA: cmi8330: fix MPU-401 PnP init copy&paste bug

Fix copy&paste bug in PnP MPU-401 initialization.

Signed-off-by: Ondrej Zary <linux@rainbow-software.org>
Cc: <stable@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/isa/cmi8330.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/isa/cmi8330.c b/sound/isa/cmi8330.c
index de83608719ea..3ee0269e5bd0 100644
--- a/sound/isa/cmi8330.c
+++ b/sound/isa/cmi8330.c
@@ -338,7 +338,7 @@ static int __devinit snd_cmi8330_pnp(int dev, struct snd_cmi8330 *acard,
 		return -EBUSY;
 
 	acard->mpu = pnp_request_card_device(card, id->devs[2].id, NULL);
-	if (acard->play == NULL)
+	if (acard->mpu == NULL)
 		return -EBUSY;
 
 	pdev = acard->cap;
-- 
cgit v1.2.3-59-g8ed1b


From 9a660a6e1a0ebef72cf792583777234483a40022 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 28 Jun 2009 16:26:13 +0000
Subject: arch/sh/boards/mach-se/7206/io.c: Remove unnecessary semicolons

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/mach-se/7206/io.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/sh/boards/mach-se/7206/io.c b/arch/sh/boards/mach-se/7206/io.c
index 9c3a33210d61..180455642a43 100644
--- a/arch/sh/boards/mach-se/7206/io.c
+++ b/arch/sh/boards/mach-se/7206/io.c
@@ -50,7 +50,7 @@ unsigned char se7206_inb_p(unsigned long port)
 
 unsigned short se7206_inw(unsigned long port)
 {
-	return *port2adr(port);;
+	return *port2adr(port);
 }
 
 void se7206_outb(unsigned char value, unsigned long port)
-- 
cgit v1.2.3-59-g8ed1b


From ec87805c63a96e8863385d79c864801a572a4ab8 Mon Sep 17 00:00:00 2001
From: Kuninori Morimoto <morimoto.kuninori@renesas.com>
Date: Mon, 29 Jun 2009 07:40:00 +0000
Subject: sh: ms7724se: Enable sh_eth in defconfig.

Signed-off-by: Kuninori Morimoto <morimoto.kuninori@renesas.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/configs/se7724_defconfig | 25 +++++++++++++++++++++----
 1 file changed, 21 insertions(+), 4 deletions(-)

diff --git a/arch/sh/configs/se7724_defconfig b/arch/sh/configs/se7724_defconfig
index 3840270283e4..3ee783a0a075 100644
--- a/arch/sh/configs/se7724_defconfig
+++ b/arch/sh/configs/se7724_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
 # Linux kernel version: 2.6.30
-# Thu Jun 18 16:09:05 2009
+# Mon Jun 29 16:28:43 2009
 #
 CONFIG_SUPERH=y
 CONFIG_SUPERH32=y
@@ -14,6 +14,7 @@ CONFIG_GENERIC_HWEIGHT=y
 CONFIG_GENERIC_HARDIRQS=y
 CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ=y
 CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_IRQ_PER_CPU=y
 CONFIG_GENERIC_GPIO=y
 CONFIG_GENERIC_TIME=y
 CONFIG_GENERIC_CLOCKEVENTS=y
@@ -28,7 +29,9 @@ CONFIG_HAVE_LATENCYTOP_SUPPORT=y
 # CONFIG_ARCH_HAS_ILOG2_U64 is not set
 CONFIG_ARCH_NO_VIRT_TO_BUS=y
 CONFIG_ARCH_HAS_DEFAULT_IDLE=y
+CONFIG_ARCH_HAS_CPU_IDLE_WAIT=y
 CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+CONFIG_CONSTRUCTORS=y
 
 #
 # General setup
@@ -88,10 +91,12 @@ CONFIG_TIMERFD=y
 CONFIG_EVENTFD=y
 CONFIG_SHMEM=y
 CONFIG_AIO=y
+CONFIG_HAVE_PERF_COUNTERS=y
 
 #
 # Performance Counters
 #
+# CONFIG_PERF_COUNTERS is not set
 CONFIG_VM_EVENT_COUNTERS=y
 # CONFIG_STRIP_ASM_SYMS is not set
 CONFIG_COMPAT_BRK=y
@@ -107,6 +112,10 @@ CONFIG_HAVE_KRETPROBES=y
 CONFIG_HAVE_ARCH_TRACEHOOK=y
 CONFIG_HAVE_CLK=y
 CONFIG_HAVE_DMA_API_DEBUG=y
+
+#
+# GCOV-based kernel profiling
+#
 # CONFIG_SLOW_WORK is not set
 CONFIG_HAVE_GENERIC_DMA_COHERENT=y
 CONFIG_SLABINFO=y
@@ -119,7 +128,7 @@ CONFIG_MODULE_UNLOAD=y
 # CONFIG_MODVERSIONS is not set
 # CONFIG_MODULE_SRCVERSION_ALL is not set
 CONFIG_BLOCK=y
-# CONFIG_LBD is not set
+CONFIG_LBDAF=y
 # CONFIG_BLK_DEV_BSG is not set
 # CONFIG_BLK_DEV_INTEGRITY is not set
 
@@ -584,7 +593,6 @@ CONFIG_SCSI_WAIT_SCAN=m
 # CONFIG_SCSI_SRP_ATTRS is not set
 CONFIG_SCSI_LOWLEVEL=y
 # CONFIG_ISCSI_TCP is not set
-# CONFIG_SCSI_BNX2_ISCSI is not set
 # CONFIG_LIBFC is not set
 # CONFIG_LIBFCOE is not set
 # CONFIG_SCSI_DEBUG is not set
@@ -624,7 +632,7 @@ CONFIG_NET_ETHERNET=y
 CONFIG_MII=y
 # CONFIG_AX88796 is not set
 # CONFIG_STNIC is not set
-# CONFIG_SH_ETH is not set
+CONFIG_SH_ETH=y
 CONFIG_SMC91X=y
 # CONFIG_ENC28J60 is not set
 # CONFIG_ETHOC is not set
@@ -801,6 +809,11 @@ CONFIG_SPI_BITBANG=y
 #
 # CONFIG_SPI_SPIDEV is not set
 # CONFIG_SPI_TLE62X0 is not set
+
+#
+# PPS support
+#
+# CONFIG_PPS is not set
 CONFIG_ARCH_REQUIRE_GPIOLIB=y
 CONFIG_GPIOLIB=y
 # CONFIG_GPIO_SYSFS is not set
@@ -851,6 +864,8 @@ CONFIG_SSB_POSSIBLE=y
 # CONFIG_MFD_WM8400 is not set
 # CONFIG_MFD_WM8350_I2C is not set
 # CONFIG_MFD_PCF50633 is not set
+# CONFIG_AB3100_CORE is not set
+# CONFIG_EZX_PCAP is not set
 # CONFIG_REGULATOR is not set
 CONFIG_MEDIA_SUPPORT=y
 
@@ -1196,6 +1211,7 @@ CONFIG_RTC_DRV_PCF8563=y
 # CONFIG_RTC_DRV_S35390A is not set
 # CONFIG_RTC_DRV_FM3130 is not set
 # CONFIG_RTC_DRV_RX8581 is not set
+# CONFIG_RTC_DRV_RX8025 is not set
 
 #
 # SPI RTC drivers
@@ -1260,6 +1276,7 @@ CONFIG_FS_MBCACHE=y
 # CONFIG_JFS_FS is not set
 CONFIG_FS_POSIX_ACL=y
 # CONFIG_XFS_FS is not set
+# CONFIG_GFS2_FS is not set
 # CONFIG_OCFS2_FS is not set
 # CONFIG_BTRFS_FS is not set
 CONFIG_FILE_LOCKING=y
-- 
cgit v1.2.3-59-g8ed1b


From 4078c444cf667f018c3fc7ebf141131a2b7c9480 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Mon, 29 Jun 2009 00:41:11 -0700
Subject: perf_counter, x86: Update x86_pmu after WARN()

The print out should read the value before changing the value.

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <4A487017.4090007@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/cpu/perf_counter.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c
index b83474b6021a..d4cf4ce19aac 100644
--- a/arch/x86/kernel/cpu/perf_counter.c
+++ b/arch/x86/kernel/cpu/perf_counter.c
@@ -1496,17 +1496,17 @@ void __init init_hw_perf_counters(void)
 	pr_cont("%s PMU driver.\n", x86_pmu.name);
 
 	if (x86_pmu.num_counters > X86_PMC_MAX_GENERIC) {
-		x86_pmu.num_counters = X86_PMC_MAX_GENERIC;
 		WARN(1, KERN_ERR "hw perf counters %d > max(%d), clipping!",
 		     x86_pmu.num_counters, X86_PMC_MAX_GENERIC);
+		x86_pmu.num_counters = X86_PMC_MAX_GENERIC;
 	}
 	perf_counter_mask = (1 << x86_pmu.num_counters) - 1;
 	perf_max_counters = x86_pmu.num_counters;
 
 	if (x86_pmu.num_counters_fixed > X86_PMC_MAX_FIXED) {
-		x86_pmu.num_counters_fixed = X86_PMC_MAX_FIXED;
 		WARN(1, KERN_ERR "hw perf counters fixed %d > max(%d), clipping!",
 		     x86_pmu.num_counters_fixed, X86_PMC_MAX_FIXED);
+		x86_pmu.num_counters_fixed = X86_PMC_MAX_FIXED;
 	}
 
 	perf_counter_mask |=
-- 
cgit v1.2.3-59-g8ed1b


From 1d955ebd4db7961dc4e772a23288b3d5c6f191be Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Mon, 29 Jun 2009 11:33:53 +0200
Subject: ALSA: hda - Add missing initializations for ALC268 and ALC269

During the changes to clean up / fix the realtek codec initialization
routines in commit 4a79ba34cada6a5a4ee86ed53aa8a73ba1e6fc51,
I forgot to add the check for ALC268 and ALC269.
This resulted in the missing EAPD and COEF setup for these codecs.

This patch adds the missing checks for these codecs.

Reference: bko#13633
	http://bugzilla.kernel.org/show_bug.cgi?id=13633

Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/pci/hda/patch_realtek.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 08846d222cbe..3a8e58c483df 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -12463,6 +12463,8 @@ static int alc268_parse_auto_config(struct hda_codec *codec)
 	if (err < 0)
 		return err;
 
+	alc_ssid_check(codec, 0x15, 0x1b, 0x14);
+
 	return 1;
 }
 
@@ -13371,6 +13373,8 @@ static int alc269_parse_auto_config(struct hda_codec *codec)
 	if (!spec->cap_mixer && !spec->no_analog)
 		set_capture_mixer(spec);
 
+	alc_ssid_check(codec, 0x15, 0x1b, 0x14);
+
 	return 1;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 59c5fe6d848cae8fd51498d17532f5aad3866f98 Mon Sep 17 00:00:00 2001
From: Nicolas Ferre <nicolas.ferre@atmel.com>
Date: Mon, 29 Jun 2009 12:11:50 +0100
Subject: [ARM] 5562/2: at91: add gpio button support for at91sam9g20ek

This adds input keyboard gpio support on at91sam9g20ek board.
It adds button 3 and 4.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-at91/board-sam9g20ek.c | 54 ++++++++++++++++++++++++++++++++++++
 1 file changed, 54 insertions(+)

diff --git a/arch/arm/mach-at91/board-sam9g20ek.c b/arch/arm/mach-at91/board-sam9g20ek.c
index cc270beadd5d..a55398ed1211 100644
--- a/arch/arm/mach-at91/board-sam9g20ek.c
+++ b/arch/arm/mach-at91/board-sam9g20ek.c
@@ -24,6 +24,8 @@
 #include <linux/platform_device.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/at73c213.h>
+#include <linux/gpio_keys.h>
+#include <linux/input.h>
 #include <linux/clk.h>
 
 #include <mach/hardware.h>
@@ -218,6 +220,56 @@ static struct gpio_led ek_leds[] = {
 	}
 };
 
+
+/*
+ * GPIO Buttons
+ */
+#if defined(CONFIG_KEYBOARD_GPIO) || defined(CONFIG_KEYBOARD_GPIO_MODULE)
+static struct gpio_keys_button ek_buttons[] = {
+	{
+		.gpio		= AT91_PIN_PA30,
+		.code		= BTN_3,
+		.desc		= "Button 3",
+		.active_low	= 1,
+		.wakeup		= 1,
+	},
+	{
+		.gpio		= AT91_PIN_PA31,
+		.code		= BTN_4,
+		.desc		= "Button 4",
+		.active_low	= 1,
+		.wakeup		= 1,
+	}
+};
+
+static struct gpio_keys_platform_data ek_button_data = {
+	.buttons	= ek_buttons,
+	.nbuttons	= ARRAY_SIZE(ek_buttons),
+};
+
+static struct platform_device ek_button_device = {
+	.name		= "gpio-keys",
+	.id		= -1,
+	.num_resources	= 0,
+	.dev		= {
+		.platform_data	= &ek_button_data,
+	}
+};
+
+static void __init ek_add_device_buttons(void)
+{
+	at91_set_gpio_input(AT91_PIN_PA30, 1);	/* btn3 */
+	at91_set_deglitch(AT91_PIN_PA30, 1);
+	at91_set_gpio_input(AT91_PIN_PA31, 1);	/* btn4 */
+	at91_set_deglitch(AT91_PIN_PA31, 1);
+
+	platform_device_register(&ek_button_device);
+}
+#else
+static void __init ek_add_device_buttons(void) {}
+#endif
+
+
 static struct i2c_board_info __initdata ek_i2c_devices[] = {
 	{
 		I2C_BOARD_INFO("24c512", 0x50),
@@ -245,6 +297,8 @@ static void __init ek_board_init(void)
 	at91_add_device_i2c(ek_i2c_devices, ARRAY_SIZE(ek_i2c_devices));
 	/* LEDs */
 	at91_gpio_leds(ek_leds, ARRAY_SIZE(ek_leds));
+	/* Push Buttons */
+	ek_add_device_buttons();
 	/* PCK0 provides MCLK to the WM8731 */
 	at91_set_B_periph(AT91_PIN_PC1, 0);
 	/* SSC (for WM8731) */
-- 
cgit v1.2.3-59-g8ed1b


From bf92df30df909710c498d05620e2df1be1ef779b Mon Sep 17 00:00:00 2001
From: Yu Zhao <yu.zhao@intel.com>
Date: Mon, 29 Jun 2009 11:31:45 +0800
Subject: intel-iommu: Only avoid flushing device IOTLB for domain ID 0 in
 caching mode

In caching mode, domain ID 0 is reserved for non-present to present
mapping flush. Device IOTLB doesn't need to be flushed in this case.

Previously we were avoiding the flush for domain zero, even if the IOMMU
wasn't in caching mode and domain zero wasn't special.

Signed-off-by: Yu Zhao <yu.zhao@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 420afa887283..3cad7006ed8e 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1054,7 +1054,12 @@ static void iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did,
 	else
 		iommu->flush.flush_iotlb(iommu, did, addr, mask,
 						DMA_TLB_PSI_FLUSH);
-	if (did)
+
+	/*
+	 * In caching mode, domain ID 0 is reserved for non-present to present
+	 * mapping flush. Device IOTLB doesn't need to be flushed in this case.
+	 */
+	if (!cap_caching_mode(iommu->cap) || did)
 		iommu_flush_dev_iotlb(iommu->domains[did], addr, mask);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From b213203e475212a69ad6fedfb73464087e317148 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Fri, 26 Jun 2009 18:50:28 +0100
Subject: intel-iommu: Create new iommu_domain_identity_map() function

We'll want to do this to a _domain_ (the si_domain) rather than a PCI device.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 62 +++++++++++++++++++++++++++--------------------
 1 file changed, 36 insertions(+), 26 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 3cad7006ed8e..3a4f347e2f88 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1849,25 +1849,12 @@ error:
 
 static int iommu_identity_mapping;
 
-static int iommu_prepare_identity_map(struct pci_dev *pdev,
-				      unsigned long long start,
-				      unsigned long long end)
+static int iommu_domain_identity_map(struct dmar_domain *domain,
+				     unsigned long long start,
+				     unsigned long long end)
 {
-	struct dmar_domain *domain;
 	unsigned long size;
 	unsigned long long base;
-	int ret;
-
-	printk(KERN_INFO
-		"IOMMU: Setting identity map for device %s [0x%Lx - 0x%Lx]\n",
-		pci_name(pdev), start, end);
-	if (iommu_identity_mapping)
-		domain = si_domain;
-	else
-		/* page table init */
-		domain = get_domain_for_dev(pdev, DEFAULT_DOMAIN_ADDRESS_WIDTH);
-	if (!domain)
-		return -ENOMEM;
 
 	/* The address might not be aligned */
 	base = start & PAGE_MASK;
@@ -1876,31 +1863,54 @@ static int iommu_prepare_identity_map(struct pci_dev *pdev,
 	if (!reserve_iova(&domain->iovad, IOVA_PFN(base),
 			IOVA_PFN(base + size) - 1)) {
 		printk(KERN_ERR "IOMMU: reserve iova failed\n");
-		ret = -ENOMEM;
-		goto error;
+		return -ENOMEM;
 	}
 
-	pr_debug("Mapping reserved region %lx@%llx for %s\n",
-		size, base, pci_name(pdev));
+	pr_debug("Mapping reserved region %lx@%llx for domain %d\n",
+		 size, base, domain->id);
 	/*
 	 * RMRR range might have overlap with physical memory range,
 	 * clear it first
 	 */
 	dma_pte_clear_range(domain, base, base + size);
 
-	ret = domain_page_mapping(domain, base, base, size,
-		DMA_PTE_READ|DMA_PTE_WRITE);
+	return domain_page_mapping(domain, base, base, size,
+				   DMA_PTE_READ|DMA_PTE_WRITE);
+}
+
+static int iommu_prepare_identity_map(struct pci_dev *pdev,
+				      unsigned long long start,
+				      unsigned long long end)
+{
+	struct dmar_domain *domain;
+	int ret;
+
+	printk(KERN_INFO
+	       "IOMMU: Setting identity map for device %s [0x%Lx - 0x%Lx]\n",
+	       pci_name(pdev), start, end);
+
+	if (iommu_identity_mapping)
+		domain = si_domain;
+	else
+		/* page table init */
+		domain = get_domain_for_dev(pdev, DEFAULT_DOMAIN_ADDRESS_WIDTH);
+	if (!domain)
+		return -ENOMEM;
+
+	ret = iommu_domain_identity_map(domain, start, end);
 	if (ret)
 		goto error;
 
 	/* context entry init */
 	ret = domain_context_mapping(domain, pdev, CONTEXT_TT_MULTI_LEVEL);
-	if (!ret)
-		return 0;
-error:
+	if (ret)
+		goto error;
+
+	return 0;
+
+ error:
 	domain_exit(domain);
 	return ret;
-
 }
 
 static inline int iommu_prepare_rmrr_dev(struct dmar_rmrr_unit *rmrr,
-- 
cgit v1.2.3-59-g8ed1b


From c7ab48d2acaf959e4d59c3f55d12fdb7ca9afd7c Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Fri, 26 Jun 2009 19:10:36 +0100
Subject: intel-iommu: Clean up identity mapping code, remove
 CONFIG_DMAR_GFX_WA

There's no need for the GFX workaround now we have 'iommu=pt' for the
cases where people really care about performance. There's no need to
have a special case for just one type of device.

This also speeds up the iommu=pt path and reduces memory usage by
setting up the si_domain _once_ and then using it for all devices,
rather than giving each device its own private page tables.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 arch/x86/Kconfig          |  15 +------
 drivers/pci/intel-iommu.c | 107 ++++++++++++++--------------------------------
 2 files changed, 34 insertions(+), 88 deletions(-)

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index d1430ef6b4f9..c07f72205909 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1913,25 +1913,14 @@ config DMAR_DEFAULT_ON
 	  recommended you say N here while the DMAR code remains
 	  experimental.
 
-config DMAR_GFX_WA
-	def_bool y
-	prompt "Support for Graphics workaround"
-	depends on DMAR
-	---help---
-	  Current Graphics drivers tend to use physical address
-	  for DMA and avoid using DMA APIs. Setting this config
-	  option permits the IOMMU driver to set a unity map for
-	  all the OS-visible memory. Hence the driver can continue
-	  to use physical addresses for DMA.
-
 config DMAR_FLOPPY_WA
 	def_bool y
 	depends on DMAR
 	---help---
-	  Floppy disk drivers are know to bypass DMA API calls
+	  Floppy disk drivers are known to bypass DMA API calls
 	  thereby failing to work when IOMMU is enabled. This
 	  workaround will setup a 1:1 mapping for the first
-	  16M to make floppy (an ISA device) work.
+	  16MiB to make floppy (an ISA device) work.
 
 config INTR_REMAP
 	bool "Support for Interrupt Remapping (EXPERIMENTAL)"
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 3a4f347e2f88..fc121967cb5b 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1889,11 +1889,7 @@ static int iommu_prepare_identity_map(struct pci_dev *pdev,
 	       "IOMMU: Setting identity map for device %s [0x%Lx - 0x%Lx]\n",
 	       pci_name(pdev), start, end);
 
-	if (iommu_identity_mapping)
-		domain = si_domain;
-	else
-		/* page table init */
-		domain = get_domain_for_dev(pdev, DEFAULT_DOMAIN_ADDRESS_WIDTH);
+	domain = get_domain_for_dev(pdev, DEFAULT_DOMAIN_ADDRESS_WIDTH);
 	if (!domain)
 		return -ENOMEM;
 
@@ -1922,64 +1918,6 @@ static inline int iommu_prepare_rmrr_dev(struct dmar_rmrr_unit *rmrr,
 		rmrr->end_address + 1);
 }
 
-struct iommu_prepare_data {
-	struct pci_dev *pdev;
-	int ret;
-};
-
-static int __init iommu_prepare_work_fn(unsigned long start_pfn,
-					 unsigned long end_pfn, void *datax)
-{
-	struct iommu_prepare_data *data;
-
-	data = (struct iommu_prepare_data *)datax;
-
-	data->ret = iommu_prepare_identity_map(data->pdev,
-				start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT);
-	return data->ret;
-
-}
-
-static int __init iommu_prepare_with_active_regions(struct pci_dev *pdev)
-{
-	int nid;
-	struct iommu_prepare_data data;
-
-	data.pdev = pdev;
-	data.ret = 0;
-
-	for_each_online_node(nid) {
-		work_with_active_regions(nid, iommu_prepare_work_fn, &data);
-		if (data.ret)
-			return data.ret;
-	}
-	return data.ret;
-}
-
-#ifdef CONFIG_DMAR_GFX_WA
-static void __init iommu_prepare_gfx_mapping(void)
-{
-	struct pci_dev *pdev = NULL;
-	int ret;
-
-	for_each_pci_dev(pdev) {
-		if (pdev->dev.archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO ||
-				!IS_GFX_DEVICE(pdev))
-			continue;
-		printk(KERN_INFO "IOMMU: gfx device %s 1-1 mapping\n",
-			pci_name(pdev));
-		ret = iommu_prepare_with_active_regions(pdev);
-		if (ret)
-			printk(KERN_ERR "IOMMU: mapping reserved region failed\n");
-	}
-}
-#else /* !CONFIG_DMAR_GFX_WA */
-static inline void iommu_prepare_gfx_mapping(void)
-{
-	return;
-}
-#endif
-
 #ifdef CONFIG_DMAR_FLOPPY_WA
 static inline void iommu_prepare_isa(void)
 {
@@ -1990,12 +1928,12 @@ static inline void iommu_prepare_isa(void)
 	if (!pdev)
 		return;
 
-	printk(KERN_INFO "IOMMU: Prepare 0-16M unity mapping for LPC\n");
+	printk(KERN_INFO "IOMMU: Prepare 0-16MiB unity mapping for LPC\n");
 	ret = iommu_prepare_identity_map(pdev, 0, 16*1024*1024);
 
 	if (ret)
-		printk(KERN_ERR "IOMMU: Failed to create 0-64M identity map, "
-			"floppy might not work\n");
+		printk(KERN_ERR "IOMMU: Failed to create 0-16MiB identity map; "
+		       "floppy might not work\n");
 
 }
 #else
@@ -2023,16 +1961,30 @@ static int __init init_context_pass_through(void)
 }
 
 static int md_domain_init(struct dmar_domain *domain, int guest_width);
+
+static int __init si_domain_work_fn(unsigned long start_pfn,
+				    unsigned long end_pfn, void *datax)
+{
+	int *ret = datax;
+
+	*ret = iommu_domain_identity_map(si_domain,
+					 (uint64_t)start_pfn << PAGE_SHIFT,
+					 (uint64_t)end_pfn << PAGE_SHIFT);
+	return *ret;
+
+}
+
 static int si_domain_init(void)
 {
 	struct dmar_drhd_unit *drhd;
 	struct intel_iommu *iommu;
-	int ret = 0;
+	int nid, ret = 0;
 
 	si_domain = alloc_domain();
 	if (!si_domain)
 		return -EFAULT;
 
+	pr_debug("Identity mapping domain is domain %d\n", si_domain->id);
 
 	for_each_active_iommu(iommu, drhd) {
 		ret = iommu_attach_domain(si_domain, iommu);
@@ -2049,6 +2001,12 @@ static int si_domain_init(void)
 
 	si_domain->flags = DOMAIN_FLAG_STATIC_IDENTITY;
 
+	for_each_online_node(nid) {
+		work_with_active_regions(nid, si_domain_work_fn, &ret);
+		if (ret)
+			return ret;
+	}
+
 	return 0;
 }
 
@@ -2102,13 +2060,14 @@ static int iommu_prepare_static_identity_mapping(void)
 	if (ret)
 		return -EFAULT;
 
-	printk(KERN_INFO "IOMMU: Setting identity map:\n");
 	for_each_pci_dev(pdev) {
-		ret = iommu_prepare_with_active_regions(pdev);
-		if (ret) {
-			printk(KERN_INFO "1:1 mapping to one domain failed.\n");
-			return -EFAULT;
-		}
+		printk(KERN_INFO "IOMMU: identity mapping for device %s\n",
+		       pci_name(pdev));
+
+		ret = domain_context_mapping(si_domain, pdev,
+					     CONTEXT_TT_MULTI_LEVEL);
+		if (ret)
+			return ret;
 		ret = domain_add_dev_info(si_domain, pdev);
 		if (ret)
 			return ret;
@@ -2299,8 +2258,6 @@ int __init init_dmars(void)
 			}
 		}
 
-		iommu_prepare_gfx_mapping();
-
 		iommu_prepare_isa();
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From dd4e831960e4f0214480fa96a53ca9bb7dd04927 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sat, 27 Jun 2009 16:21:20 +0100
Subject: intel-iommu: Change dma_set_pte_addr() to dma_set_pte_pfn()

Add some helpers for converting between VT-d and normal system pfns,
since system pages can be larger than VT-d pages.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 28 ++++++++++++++++++++++++----
 1 file changed, 24 insertions(+), 4 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index fc121967cb5b..852f40a913d4 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -65,6 +65,26 @@
 #define PHYSICAL_PAGE_MASK PAGE_MASK
 #endif
 
+/* VT-d pages must always be _smaller_ than MM pages. Otherwise things
+   are never going to work. */
+static inline unsigned long dma_to_mm_pfn(unsigned long dma_pfn)
+{
+	return dma_pfn >> (PAGE_SHIFT - VTD_PAGE_SHIFT);
+}
+
+static inline unsigned long mm_to_dma_pfn(unsigned long mm_pfn)
+{
+	return mm_pfn << (PAGE_SHIFT - VTD_PAGE_SHIFT);
+}
+static inline unsigned long page_to_dma_pfn(struct page *pg)
+{
+	return mm_to_dma_pfn(page_to_pfn(pg));
+}
+static inline unsigned long virt_to_dma_pfn(void *p)
+{
+	return page_to_dma_pfn(virt_to_page(p));
+}
+
 /* global iommu list, set NULL for ignored DMAR units */
 static struct intel_iommu **g_iommus;
 
@@ -207,9 +227,9 @@ static inline u64 dma_pte_addr(struct dma_pte *pte)
 	return (pte->val & VTD_PAGE_MASK);
 }
 
-static inline void dma_set_pte_addr(struct dma_pte *pte, u64 addr)
+static inline void dma_set_pte_pfn(struct dma_pte *pte, unsigned long pfn)
 {
-	pte->val |= (addr & VTD_PAGE_MASK);
+	pte->val |= (uint64_t)pfn << VTD_PAGE_SHIFT;
 }
 
 static inline bool dma_pte_present(struct dma_pte *pte)
@@ -702,7 +722,7 @@ static struct dma_pte * addr_to_dma_pte(struct dmar_domain *domain, u64 addr)
 				return NULL;
 			}
 			domain_flush_cache(domain, tmp_page, PAGE_SIZE);
-			dma_set_pte_addr(pte, virt_to_phys(tmp_page));
+			dma_set_pte_pfn(pte, virt_to_dma_pfn(tmp_page));
 			/*
 			 * high level table always sets r/w, last level page
 			 * table control read/write
@@ -1648,7 +1668,7 @@ domain_page_mapping(struct dmar_domain *domain, dma_addr_t iova,
 		 * touches the iova range
 		 */
 		BUG_ON(dma_pte_addr(pte));
-		dma_set_pte_addr(pte, start_pfn << VTD_PAGE_SHIFT);
+		dma_set_pte_pfn(pte, start_pfn);
 		dma_set_pte_prot(pte, prot);
 		if (prot & DMA_PTE_SNP)
 			dma_set_pte_snp(pte);
-- 
cgit v1.2.3-59-g8ed1b


From 77dfa56c94d2855a25ff552b74980a5538e129f8 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sat, 27 Jun 2009 16:40:08 +0100
Subject: intel-iommu: Change address_level_offset() to pfn_level_offset()

We're shifting the inputs for now, but that'll change...

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 852f40a913d4..529c1c13048f 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -671,9 +671,9 @@ static inline unsigned int level_to_offset_bits(int level)
 	return (12 + (level - 1) * LEVEL_STRIDE);
 }
 
-static inline int address_level_offset(u64 addr, int level)
+static inline int pfn_level_offset(unsigned long pfn, int level)
 {
-	return ((addr >> level_to_offset_bits(level)) & LEVEL_MASK);
+	return (pfn >> (level_to_offset_bits(level) - 12)) & LEVEL_MASK;
 }
 
 static inline u64 level_mask(int level)
@@ -708,7 +708,7 @@ static struct dma_pte * addr_to_dma_pte(struct dmar_domain *domain, u64 addr)
 	while (level > 0) {
 		void *tmp_page;
 
-		offset = address_level_offset(addr, level);
+		offset = pfn_level_offset(addr >> VTD_PAGE_SHIFT, level);
 		pte = &parent[offset];
 		if (level == 1)
 			break;
@@ -749,7 +749,7 @@ static struct dma_pte *dma_addr_level_pte(struct dmar_domain *domain, u64 addr,
 
 	parent = domain->pgd;
 	while (level <= total) {
-		offset = address_level_offset(addr, total);
+		offset = pfn_level_offset(addr >> VTD_PAGE_SHIFT, total);
 		pte = &parent[offset];
 		if (level == total)
 			return pte;
-- 
cgit v1.2.3-59-g8ed1b


From 90dcfb5eb2fd427b16135a14f176a6902750b6b4 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sat, 27 Jun 2009 17:14:59 +0100
Subject: intel-iommu: Change dma_addr_level_pte() to dma_pfn_level_pte()

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 529c1c13048f..edd39d348a98 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -740,8 +740,9 @@ static struct dma_pte * addr_to_dma_pte(struct dmar_domain *domain, u64 addr)
 }
 
 /* return address's pte at specific level */
-static struct dma_pte *dma_addr_level_pte(struct dmar_domain *domain, u64 addr,
-		int level)
+static struct dma_pte *dma_pfn_level_pte(struct dmar_domain *domain,
+					 unsigned long pfn,
+					 int level)
 {
 	struct dma_pte *parent, *pte = NULL;
 	int total = agaw_to_level(domain->agaw);
@@ -749,7 +750,7 @@ static struct dma_pte *dma_addr_level_pte(struct dmar_domain *domain, u64 addr,
 
 	parent = domain->pgd;
 	while (level <= total) {
-		offset = pfn_level_offset(addr >> VTD_PAGE_SHIFT, total);
+		offset = pfn_level_offset(pfn, total);
 		pte = &parent[offset];
 		if (level == total)
 			return pte;
@@ -768,7 +769,7 @@ static void dma_pte_clear_one(struct dmar_domain *domain, u64 addr)
 	struct dma_pte *pte = NULL;
 
 	/* get last level pte */
-	pte = dma_addr_level_pte(domain, addr, 1);
+	pte = dma_pfn_level_pte(domain, addr >> VTD_PAGE_SHIFT, 1);
 
 	if (pte) {
 		dma_clear_pte(pte);
@@ -817,7 +818,8 @@ static void dma_pte_free_pagetable(struct dmar_domain *domain,
 			return;
 
 		while (tmp < end) {
-			pte = dma_addr_level_pte(domain, tmp, level);
+			pte = dma_pfn_level_pte(domain, tmp >> VTD_PAGE_SHIFT,
+						level);
 			if (pte) {
 				free_pgtable_page(
 					phys_to_virt(dma_pte_addr(pte)));
-- 
cgit v1.2.3-59-g8ed1b


From a75f7cf94f01717c5103138319b96752ee2a2be9 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sat, 27 Jun 2009 17:44:39 +0100
Subject: intel-iommu: Make dma_pte_clear_one() take pfn not address

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index edd39d348a98..40eae2097aca 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -764,12 +764,12 @@ static struct dma_pte *dma_pfn_level_pte(struct dmar_domain *domain,
 }
 
 /* clear one page's page table */
-static void dma_pte_clear_one(struct dmar_domain *domain, u64 addr)
+static void dma_pte_clear_one(struct dmar_domain *domain, unsigned long pfn)
 {
 	struct dma_pte *pte = NULL;
 
 	/* get last level pte */
-	pte = dma_pfn_level_pte(domain, addr >> VTD_PAGE_SHIFT, 1);
+	pte = dma_pfn_level_pte(domain, pfn, 1);
 
 	if (pte) {
 		dma_clear_pte(pte);
@@ -792,7 +792,7 @@ static void dma_pte_clear_range(struct dmar_domain *domain, u64 start, u64 end)
 
 	/* we don't need lock here, nobody else touches the iova range */
 	while (npages--) {
-		dma_pte_clear_one(domain, start);
+		dma_pte_clear_one(domain, start >> VTD_PAGE_SHIFT);
 		start += VTD_PAGE_SIZE;
 	}
 }
-- 
cgit v1.2.3-59-g8ed1b


From 66eae8469e4e4ba6f4ca7ef82103c78f6d645583 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sat, 27 Jun 2009 19:00:32 +0100
Subject: intel-iommu: Don't just mask out too-big physical addresses; BUG()
 instead

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 40eae2097aca..ad367f53a2bb 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -700,8 +700,7 @@ static struct dma_pte * addr_to_dma_pte(struct dmar_domain *domain, u64 addr)
 	unsigned long flags;
 
 	BUG_ON(!domain->pgd);
-
-	addr &= (((u64)1) << addr_width) - 1;
+	BUG_ON(addr >> addr_width);
 	parent = domain->pgd;
 
 	spin_lock_irqsave(&domain->mapping_lock, flags);
@@ -783,8 +782,9 @@ static void dma_pte_clear_range(struct dmar_domain *domain, u64 start, u64 end)
 	int addr_width = agaw_to_width(domain->agaw);
 	int npages;
 
-	start &= (((u64)1) << addr_width) - 1;
-	end &= (((u64)1) << addr_width) - 1;
+	BUG_ON(start >> addr_width);
+	BUG_ON((end-1) >> addr_width);
+
 	/* in case it's partial page */
 	start &= PAGE_MASK;
 	end = PAGE_ALIGN(end);
@@ -807,8 +807,8 @@ static void dma_pte_free_pagetable(struct dmar_domain *domain,
 	int level;
 	u64 tmp;
 
-	start &= (((u64)1) << addr_width) - 1;
-	end &= (((u64)1) << addr_width) - 1;
+	BUG_ON(start >> addr_width);
+	BUG_ON(end >> addr_width);
 
 	/* we don't need lock here, nobody else touches the iova range */
 	level = 2;
@@ -1654,7 +1654,7 @@ domain_page_mapping(struct dmar_domain *domain, dma_addr_t iova,
 	int index;
 	int addr_width = agaw_to_width(domain->agaw);
 
-	hpa &= (((u64)1) << addr_width) - 1;
+	BUG_ON(hpa >> addr_width);
 
 	if ((prot & (DMA_PTE_READ|DMA_PTE_WRITE)) == 0)
 		return -EINVAL;
-- 
cgit v1.2.3-59-g8ed1b


From a3a9f79e361e864f0e9d75ebe2a0cb43d17c4272 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 29 Jun 2009 14:07:56 +0200
Subject: netfilter: tcp conntrack: fix unacknowledged data detection with NAT

When NAT helpers change the TCP packet size, the highest seen sequence
number needs to be corrected. This is currently only done upwards, when
the packet size is reduced the sequence number is unchanged. This causes
TCP conntrack to falsely detect unacknowledged data and decrease the
timeout.

Fix by updating the highest seen sequence number in both directions after
packet mangling.

Tested-by: Krzysztof Piotr Oledzki <ole@ans.pl>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/net/netfilter/nf_conntrack.h   |  4 ++--
 net/ipv4/netfilter/nf_nat_helper.c     | 17 +++++++++++------
 net/netfilter/nf_conntrack_proto_tcp.c |  6 +++---
 3 files changed, 16 insertions(+), 11 deletions(-)

diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h
index a632689b61b4..cbdd6284996d 100644
--- a/include/net/netfilter/nf_conntrack.h
+++ b/include/net/netfilter/nf_conntrack.h
@@ -258,8 +258,8 @@ static inline bool nf_ct_kill(struct nf_conn *ct)
 /* Update TCP window tracking data when NAT mangles the packet */
 extern void nf_conntrack_tcp_update(const struct sk_buff *skb,
 				    unsigned int dataoff,
-				    struct nf_conn *ct,
-				    int dir);
+				    struct nf_conn *ct, int dir,
+				    s16 offset);
 
 /* Fake conntrack entry for untracked connections */
 extern struct nf_conn nf_conntrack_untracked;
diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c
index 155c008626c8..09172a65d9b6 100644
--- a/net/ipv4/netfilter/nf_nat_helper.c
+++ b/net/ipv4/netfilter/nf_nat_helper.c
@@ -191,7 +191,8 @@ nf_nat_mangle_tcp_packet(struct sk_buff *skb,
 				    ct, ctinfo);
 		/* Tell TCP window tracking about seq change */
 		nf_conntrack_tcp_update(skb, ip_hdrlen(skb),
-					ct, CTINFO2DIR(ctinfo));
+					ct, CTINFO2DIR(ctinfo),
+					(int)rep_len - (int)match_len);
 
 		nf_conntrack_event_cache(IPCT_NATSEQADJ, ct);
 	}
@@ -377,6 +378,7 @@ nf_nat_seq_adjust(struct sk_buff *skb,
 	struct tcphdr *tcph;
 	int dir;
 	__be32 newseq, newack;
+	s16 seqoff, ackoff;
 	struct nf_conn_nat *nat = nfct_nat(ct);
 	struct nf_nat_seq *this_way, *other_way;
 
@@ -390,15 +392,18 @@ nf_nat_seq_adjust(struct sk_buff *skb,
 
 	tcph = (void *)skb->data + ip_hdrlen(skb);
 	if (after(ntohl(tcph->seq), this_way->correction_pos))
-		newseq = htonl(ntohl(tcph->seq) + this_way->offset_after);
+		seqoff = this_way->offset_after;
 	else
-		newseq = htonl(ntohl(tcph->seq) + this_way->offset_before);
+		seqoff = this_way->offset_before;
 
 	if (after(ntohl(tcph->ack_seq) - other_way->offset_before,
 		  other_way->correction_pos))
-		newack = htonl(ntohl(tcph->ack_seq) - other_way->offset_after);
+		ackoff = other_way->offset_after;
 	else
-		newack = htonl(ntohl(tcph->ack_seq) - other_way->offset_before);
+		ackoff = other_way->offset_before;
+
+	newseq = htonl(ntohl(tcph->seq) + seqoff);
+	newack = htonl(ntohl(tcph->ack_seq) - ackoff);
 
 	inet_proto_csum_replace4(&tcph->check, skb, tcph->seq, newseq, 0);
 	inet_proto_csum_replace4(&tcph->check, skb, tcph->ack_seq, newack, 0);
@@ -413,7 +418,7 @@ nf_nat_seq_adjust(struct sk_buff *skb,
 	if (!nf_nat_sack_adjust(skb, tcph, ct, ctinfo))
 		return 0;
 
-	nf_conntrack_tcp_update(skb, ip_hdrlen(skb), ct, dir);
+	nf_conntrack_tcp_update(skb, ip_hdrlen(skb), ct, dir, seqoff);
 
 	return 1;
 }
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index 33fc0a443f3d..97a82ba75376 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -720,8 +720,8 @@ static bool tcp_in_window(const struct nf_conn *ct,
 /* Caller must linearize skb at tcp header. */
 void nf_conntrack_tcp_update(const struct sk_buff *skb,
 			     unsigned int dataoff,
-			     struct nf_conn *ct,
-			     int dir)
+			     struct nf_conn *ct, int dir,
+			     s16 offset)
 {
 	const struct tcphdr *tcph = (const void *)skb->data + dataoff;
 	const struct ip_ct_tcp_state *sender = &ct->proto.tcp.seen[dir];
@@ -734,7 +734,7 @@ void nf_conntrack_tcp_update(const struct sk_buff *skb,
 	/*
 	 * We have to worry for the ack in the reply packet only...
 	 */
-	if (after(end, ct->proto.tcp.seen[dir].td_end))
+	if (ct->proto.tcp.seen[dir].td_end + offset == end)
 		ct->proto.tcp.seen[dir].td_end = end;
 	ct->proto.tcp.last_end = end;
 	spin_unlock_bh(&ct->lock);
-- 
cgit v1.2.3-59-g8ed1b


From 04b18e65dd5a3e544f07f4bcfa8fb52704a1833b Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sat, 27 Jun 2009 19:15:01 +0100
Subject: intel-iommu: Make dma_pte_clear_range() use pfns

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 22 +++++++++-------------
 1 file changed, 9 insertions(+), 13 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index ad367f53a2bb..d4217f737159 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -779,21 +779,17 @@ static void dma_pte_clear_one(struct dmar_domain *domain, unsigned long pfn)
 /* clear last level pte, a tlb flush should be followed */
 static void dma_pte_clear_range(struct dmar_domain *domain, u64 start, u64 end)
 {
-	int addr_width = agaw_to_width(domain->agaw);
-	int npages;
-
-	BUG_ON(start >> addr_width);
-	BUG_ON((end-1) >> addr_width);
+	unsigned long start_pfn = IOVA_PFN(start);
+	unsigned long end_pfn = IOVA_PFN(end-1);
+	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
 
-	/* in case it's partial page */
-	start &= PAGE_MASK;
-	end = PAGE_ALIGN(end);
-	npages = (end - start) / VTD_PAGE_SIZE;
+	BUG_ON(addr_width < BITS_PER_LONG && start_pfn >> addr_width);
+	BUG_ON(addr_width < BITS_PER_LONG && end_pfn >> addr_width);
 
-	/* we don't need lock here, nobody else touches the iova range */
-	while (npages--) {
-		dma_pte_clear_one(domain, start >> VTD_PAGE_SHIFT);
-		start += VTD_PAGE_SIZE;
+	/* we don't need lock here; nobody else touches the iova range */
+	while (start_pfn <= end_pfn) {
+		dma_pte_clear_one(domain, start_pfn);
+		start_pfn++;
 	}
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 595badf5d65d50300319e6178e6df005ea501f70 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sat, 27 Jun 2009 22:09:11 +0100
Subject: intel-iommu: Make dma_pte_clear_range() take pfns as argument

Noting that this is now an _inclusive_ range.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 32 +++++++++++++++++++-------------
 1 file changed, 19 insertions(+), 13 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index d4217f737159..ff8b7ce4a013 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -56,6 +56,7 @@
 #define MAX_AGAW_WIDTH 64
 
 #define DOMAIN_MAX_ADDR(gaw) ((((u64)1) << gaw) - 1)
+#define DOMAIN_MAX_PFN(gaw)  ((((u64)1) << (gaw-VTD_PAGE_SHIFT)) - 1)
 
 #define IOVA_PFN(addr)		((addr) >> PAGE_SHIFT)
 #define DMA_32BIT_PFN		IOVA_PFN(DMA_BIT_MASK(32))
@@ -777,17 +778,17 @@ static void dma_pte_clear_one(struct dmar_domain *domain, unsigned long pfn)
 }
 
 /* clear last level pte, a tlb flush should be followed */
-static void dma_pte_clear_range(struct dmar_domain *domain, u64 start, u64 end)
+static void dma_pte_clear_range(struct dmar_domain *domain,
+				unsigned long start_pfn,
+				unsigned long last_pfn)
 {
-	unsigned long start_pfn = IOVA_PFN(start);
-	unsigned long end_pfn = IOVA_PFN(end-1);
 	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
 
 	BUG_ON(addr_width < BITS_PER_LONG && start_pfn >> addr_width);
-	BUG_ON(addr_width < BITS_PER_LONG && end_pfn >> addr_width);
+	BUG_ON(addr_width < BITS_PER_LONG && last_pfn >> addr_width);
 
 	/* we don't need lock here; nobody else touches the iova range */
-	while (start_pfn <= end_pfn) {
+	while (start_pfn <= last_pfn) {
 		dma_pte_clear_one(domain, start_pfn);
 		start_pfn++;
 	}
@@ -1424,7 +1425,7 @@ static void domain_exit(struct dmar_domain *domain)
 	end = end & (~PAGE_MASK);
 
 	/* clear ptes */
-	dma_pte_clear_range(domain, 0, end);
+	dma_pte_clear_range(domain, 0, DOMAIN_MAX_PFN(domain->gaw));
 
 	/* free page tables */
 	dma_pte_free_pagetable(domain, 0, end);
@@ -1890,7 +1891,8 @@ static int iommu_domain_identity_map(struct dmar_domain *domain,
 	 * RMRR range might have overlap with physical memory range,
 	 * clear it first
 	 */
-	dma_pte_clear_range(domain, base, base + size);
+	dma_pte_clear_range(domain, base >> VTD_PAGE_SHIFT,
+			    (base + size - 1) >> VTD_PAGE_SHIFT);
 
 	return domain_page_mapping(domain, base, base, size,
 				   DMA_PTE_READ|DMA_PTE_WRITE);
@@ -2618,7 +2620,8 @@ static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr,
 		pci_name(pdev), size, (unsigned long long)start_addr);
 
 	/*  clear the whole page */
-	dma_pte_clear_range(domain, start_addr, start_addr + size);
+	dma_pte_clear_range(domain, start_addr >> VTD_PAGE_SHIFT,
+			    (start_addr + size - 1) >> VTD_PAGE_SHIFT);
 	/* free page tables */
 	dma_pte_free_pagetable(domain, start_addr, start_addr + size);
 	if (intel_iommu_strict) {
@@ -2710,7 +2713,8 @@ static void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist,
 	start_addr = iova->pfn_lo << PAGE_SHIFT;
 
 	/*  clear the whole page */
-	dma_pte_clear_range(domain, start_addr, start_addr + size);
+	dma_pte_clear_range(domain, start_addr >> VTD_PAGE_SHIFT,
+			    (start_addr + size - 1) >> VTD_PAGE_SHIFT);
 	/* free page tables */
 	dma_pte_free_pagetable(domain, start_addr, start_addr + size);
 
@@ -2792,8 +2796,9 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne
 					  size, prot);
 		if (ret) {
 			/*  clear the page */
-			dma_pte_clear_range(domain, start_addr,
-				  start_addr + offset);
+			dma_pte_clear_range(domain,
+					    start_addr >> VTD_PAGE_SHIFT,
+					    (start_addr + offset - 1) >> VTD_PAGE_SHIFT);
 			/* free page tables */
 			dma_pte_free_pagetable(domain, start_addr,
 				  start_addr + offset);
@@ -3382,7 +3387,7 @@ static void vm_domain_exit(struct dmar_domain *domain)
 	end = end & (~VTD_PAGE_MASK);
 
 	/* clear ptes */
-	dma_pte_clear_range(domain, 0, end);
+	dma_pte_clear_range(domain, 0, DOMAIN_MAX_PFN(domain->gaw));
 
 	/* free page tables */
 	dma_pte_free_pagetable(domain, 0, end);
@@ -3526,7 +3531,8 @@ static void intel_iommu_unmap_range(struct iommu_domain *domain,
 	/* The address might not be aligned */
 	base = iova & VTD_PAGE_MASK;
 	size = VTD_PAGE_ALIGN(size);
-	dma_pte_clear_range(dmar_domain, base, base + size);
+	dma_pte_clear_range(dmar_domain, base >> VTD_PAGE_SHIFT,
+			    (base + size - 1) >> VTD_PAGE_SHIFT);
 
 	if (dmar_domain->max_addr == base + size)
 		dmar_domain->max_addr = base;
-- 
cgit v1.2.3-59-g8ed1b


From 8a3af79361e85db6fec4173ef1916322471c19e3 Mon Sep 17 00:00:00 2001
From: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Date: Mon, 29 Jun 2009 14:28:27 +0200
Subject: netfilter: headers_check fix: linux/netfilter/xt_osf.h

fix the following 'make headers_check' warnings:

  usr/include/linux/netfilter/xt_osf.h:40: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter/xt_osf.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/linux/netfilter/xt_osf.h b/include/linux/netfilter/xt_osf.h
index fd2272e0959a..18afa495f973 100644
--- a/include/linux/netfilter/xt_osf.h
+++ b/include/linux/netfilter/xt_osf.h
@@ -20,6 +20,8 @@
 #ifndef _XT_OSF_H
 #define _XT_OSF_H
 
+#include <linux/types.h>
+
 #define MAXGENRELEN		32
 
 #define XT_OSF_GENRE		(1<<0)
-- 
cgit v1.2.3-59-g8ed1b


From 6660c63a79a639b86e3a709e25a8c4fc3ab24770 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sat, 27 Jun 2009 22:41:00 +0100
Subject: intel-iommu: Make dma_pte_free_pagetable() use pfns

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 40 ++++++++++++++++++++++------------------
 1 file changed, 22 insertions(+), 18 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index ff8b7ce4a013..1526864a9d6f 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -669,27 +669,27 @@ static inline int width_to_agaw(int width)
 
 static inline unsigned int level_to_offset_bits(int level)
 {
-	return (12 + (level - 1) * LEVEL_STRIDE);
+	return (level - 1) * LEVEL_STRIDE;
 }
 
 static inline int pfn_level_offset(unsigned long pfn, int level)
 {
-	return (pfn >> (level_to_offset_bits(level) - 12)) & LEVEL_MASK;
+	return (pfn >> level_to_offset_bits(level)) & LEVEL_MASK;
 }
 
-static inline u64 level_mask(int level)
+static inline unsigned long level_mask(int level)
 {
-	return ((u64)-1 << level_to_offset_bits(level));
+	return -1UL << level_to_offset_bits(level);
 }
 
-static inline u64 level_size(int level)
+static inline unsigned long level_size(int level)
 {
-	return ((u64)1 << level_to_offset_bits(level));
+	return 1UL << level_to_offset_bits(level);
 }
 
-static inline u64 align_to_level(u64 addr, int level)
+static inline unsigned long align_to_level(unsigned long pfn, int level)
 {
-	return ((addr + level_size(level) - 1) & level_mask(level));
+	return (pfn + level_size(level) - 1) & level_mask(level);
 }
 
 static struct dma_pte * addr_to_dma_pte(struct dmar_domain *domain, u64 addr)
@@ -798,25 +798,29 @@ static void dma_pte_clear_range(struct dmar_domain *domain,
 static void dma_pte_free_pagetable(struct dmar_domain *domain,
 	u64 start, u64 end)
 {
-	int addr_width = agaw_to_width(domain->agaw);
+	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
+	unsigned long start_pfn = start >> VTD_PAGE_SHIFT;
+	unsigned long last_pfn = (end-1) >> VTD_PAGE_SHIFT;
 	struct dma_pte *pte;
 	int total = agaw_to_level(domain->agaw);
 	int level;
-	u64 tmp;
+	unsigned long tmp;
 
-	BUG_ON(start >> addr_width);
-	BUG_ON(end >> addr_width);
+	BUG_ON(addr_width < BITS_PER_LONG && start_pfn >> addr_width);
+	BUG_ON(addr_width < BITS_PER_LONG && last_pfn >> addr_width);
 
 	/* we don't need lock here, nobody else touches the iova range */
 	level = 2;
 	while (level <= total) {
-		tmp = align_to_level(start, level);
-		if (tmp >= end || (tmp + level_size(level) > end))
+		tmp = align_to_level(start_pfn, level);
+
+		/* Only clear this pte/pmd if we're asked to clear its
+		   _whole_ range */
+		if (tmp + level_size(level) - 1 > last_pfn)
 			return;
 
-		while (tmp < end) {
-			pte = dma_pfn_level_pte(domain, tmp >> VTD_PAGE_SHIFT,
-						level);
+		while (tmp <= last_pfn) {
+			pte = dma_pfn_level_pte(domain, tmp, level);
 			if (pte) {
 				free_pgtable_page(
 					phys_to_virt(dma_pte_addr(pte)));
@@ -828,7 +832,7 @@ static void dma_pte_free_pagetable(struct dmar_domain *domain,
 		level++;
 	}
 	/* free pgd */
-	if (start == 0 && end >= ((((u64)1) << addr_width) - 1)) {
+	if (start == 0 && last_pfn == DOMAIN_MAX_PFN(domain->gaw)) {
 		free_pgtable_page(domain->pgd);
 		domain->pgd = NULL;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From d794dc9b302c2781c571c10dedb8094e223d31b8 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 00:27:49 +0100
Subject: intel-iommu: Make dma_pte_free_pagetable() take pfns as argument

With some cleanup of intel_unmap_page(), intel_unmap_sg() and
vm_domain_exit() to no longer play with 64-bit addresses.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 68 +++++++++++++++++++----------------------------
 1 file changed, 28 insertions(+), 40 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 1526864a9d6f..fc593adb049a 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -796,11 +796,10 @@ static void dma_pte_clear_range(struct dmar_domain *domain,
 
 /* free page table pages. last level pte should already be cleared */
 static void dma_pte_free_pagetable(struct dmar_domain *domain,
-	u64 start, u64 end)
+				   unsigned long start_pfn,
+				   unsigned long last_pfn)
 {
 	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
-	unsigned long start_pfn = start >> VTD_PAGE_SHIFT;
-	unsigned long last_pfn = (end-1) >> VTD_PAGE_SHIFT;
 	struct dma_pte *pte;
 	int total = agaw_to_level(domain->agaw);
 	int level;
@@ -832,7 +831,7 @@ static void dma_pte_free_pagetable(struct dmar_domain *domain,
 		level++;
 	}
 	/* free pgd */
-	if (start == 0 && last_pfn == DOMAIN_MAX_PFN(domain->gaw)) {
+	if (start_pfn == 0 && last_pfn == DOMAIN_MAX_PFN(domain->gaw)) {
 		free_pgtable_page(domain->pgd);
 		domain->pgd = NULL;
 	}
@@ -1416,7 +1415,6 @@ static void domain_exit(struct dmar_domain *domain)
 {
 	struct dmar_drhd_unit *drhd;
 	struct intel_iommu *iommu;
-	u64 end;
 
 	/* Domain 0 is reserved, so dont process it */
 	if (!domain)
@@ -1425,14 +1423,12 @@ static void domain_exit(struct dmar_domain *domain)
 	domain_remove_dev_info(domain);
 	/* destroy iovas */
 	put_iova_domain(&domain->iovad);
-	end = DOMAIN_MAX_ADDR(domain->gaw);
-	end = end & (~PAGE_MASK);
 
 	/* clear ptes */
 	dma_pte_clear_range(domain, 0, DOMAIN_MAX_PFN(domain->gaw));
 
 	/* free page tables */
-	dma_pte_free_pagetable(domain, 0, end);
+	dma_pte_free_pagetable(domain, 0, DOMAIN_MAX_PFN(domain->gaw));
 
 	for_each_active_iommu(iommu, drhd)
 		if (test_bit(iommu->seq_id, &domain->iommu_bmp))
@@ -2601,7 +2597,7 @@ static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr,
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 	struct dmar_domain *domain;
-	unsigned long start_addr;
+	unsigned long start_pfn, last_pfn;
 	struct iova *iova;
 	struct intel_iommu *iommu;
 
@@ -2617,20 +2613,22 @@ static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr,
 	if (!iova)
 		return;
 
-	start_addr = iova->pfn_lo << PAGE_SHIFT;
-	size = aligned_size((u64)dev_addr, size);
+	start_pfn = mm_to_dma_pfn(iova->pfn_lo);
+	last_pfn = mm_to_dma_pfn(iova->pfn_hi + 1) - 1;
 
-	pr_debug("Device %s unmapping: %zx@%llx\n",
-		pci_name(pdev), size, (unsigned long long)start_addr);
+	pr_debug("Device %s unmapping: pfn %lx-%lx\n",
+		 pci_name(pdev), start_pfn, last_pfn);
 
 	/*  clear the whole page */
-	dma_pte_clear_range(domain, start_addr >> VTD_PAGE_SHIFT,
-			    (start_addr + size - 1) >> VTD_PAGE_SHIFT);
+	dma_pte_clear_range(domain, start_pfn, last_pfn);
+
 	/* free page tables */
-	dma_pte_free_pagetable(domain, start_addr, start_addr + size);
+	dma_pte_free_pagetable(domain, start_pfn, last_pfn);
+
 	if (intel_iommu_strict) {
-		iommu_flush_iotlb_psi(iommu, domain->id, start_addr,
-				      size >> VTD_PAGE_SHIFT);
+		iommu_flush_iotlb_psi(iommu, domain->id,
+				      start_pfn << VTD_PAGE_SHIFT,
+				      last_pfn - start_pfn + 1);
 		/* free iova */
 		__free_iova(&domain->iovad, iova);
 	} else {
@@ -2688,14 +2686,10 @@ static void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist,
 			   int nelems, enum dma_data_direction dir,
 			   struct dma_attrs *attrs)
 {
-	int i;
 	struct pci_dev *pdev = to_pci_dev(hwdev);
 	struct dmar_domain *domain;
-	unsigned long start_addr;
+	unsigned long start_pfn, last_pfn;
 	struct iova *iova;
-	size_t size = 0;
-	phys_addr_t addr;
-	struct scatterlist *sg;
 	struct intel_iommu *iommu;
 
 	if (iommu_no_mapping(pdev))
@@ -2709,21 +2703,19 @@ static void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist,
 	iova = find_iova(&domain->iovad, IOVA_PFN(sglist[0].dma_address));
 	if (!iova)
 		return;
-	for_each_sg(sglist, sg, nelems, i) {
-		addr = page_to_phys(sg_page(sg)) + sg->offset;
-		size += aligned_size((u64)addr, sg->length);
-	}
 
-	start_addr = iova->pfn_lo << PAGE_SHIFT;
+	start_pfn = mm_to_dma_pfn(iova->pfn_lo);
+	last_pfn = mm_to_dma_pfn(iova->pfn_hi + 1) - 1;
 
 	/*  clear the whole page */
-	dma_pte_clear_range(domain, start_addr >> VTD_PAGE_SHIFT,
-			    (start_addr + size - 1) >> VTD_PAGE_SHIFT);
+	dma_pte_clear_range(domain, start_pfn, last_pfn);
+
 	/* free page tables */
-	dma_pte_free_pagetable(domain, start_addr, start_addr + size);
+	dma_pte_free_pagetable(domain, start_pfn, last_pfn);
 
-	iommu_flush_iotlb_psi(iommu, domain->id, start_addr,
-			      size >> VTD_PAGE_SHIFT);
+	iommu_flush_iotlb_psi(iommu, domain->id,
+			      start_pfn << VTD_PAGE_SHIFT,
+			      (last_pfn - start_pfn + 1));
 
 	/* free iova */
 	__free_iova(&domain->iovad, iova);
@@ -2804,8 +2796,8 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne
 					    start_addr >> VTD_PAGE_SHIFT,
 					    (start_addr + offset - 1) >> VTD_PAGE_SHIFT);
 			/* free page tables */
-			dma_pte_free_pagetable(domain, start_addr,
-				  start_addr + offset);
+			dma_pte_free_pagetable(domain, start_addr >> VTD_PAGE_SHIFT,
+					       (start_addr + offset - 1) >> VTD_PAGE_SHIFT);
 			/* free iova */
 			__free_iova(&domain->iovad, iova);
 			return 0;
@@ -3378,8 +3370,6 @@ static void iommu_free_vm_domain(struct dmar_domain *domain)
 
 static void vm_domain_exit(struct dmar_domain *domain)
 {
-	u64 end;
-
 	/* Domain 0 is reserved, so dont process it */
 	if (!domain)
 		return;
@@ -3387,14 +3377,12 @@ static void vm_domain_exit(struct dmar_domain *domain)
 	vm_domain_remove_all_dev_info(domain);
 	/* destroy iovas */
 	put_iova_domain(&domain->iovad);
-	end = DOMAIN_MAX_ADDR(domain->gaw);
-	end = end & (~VTD_PAGE_MASK);
 
 	/* clear ptes */
 	dma_pte_clear_range(domain, 0, DOMAIN_MAX_PFN(domain->gaw));
 
 	/* free page tables */
-	dma_pte_free_pagetable(domain, 0, end);
+	dma_pte_free_pagetable(domain, 0, DOMAIN_MAX_PFN(domain->gaw));
 
 	iommu_free_vm_domain(domain);
 	free_domain_mem(domain);
-- 
cgit v1.2.3-59-g8ed1b


From 163cc52ccd2cc5c5ae4e1c886f6fde8547feed2a Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 00:51:17 +0100
Subject: intel-iommu: Clean up intel_iommu_unmap_range()

Use unaligned address for domain->max_addr. That algorithm isn't ideal
anyway -- we should probably just look at the last iova in the tree.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index fc593adb049a..21dc77311863 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -3491,7 +3491,7 @@ static int intel_iommu_map_range(struct iommu_domain *domain,
 	if ((iommu_prot & IOMMU_CACHE) && dmar_domain->iommu_snooping)
 		prot |= DMA_PTE_SNP;
 
-	max_addr = (iova & VTD_PAGE_MASK) + VTD_PAGE_ALIGN(size);
+	max_addr = iova + size;
 	if (dmar_domain->max_addr < max_addr) {
 		int min_agaw;
 		u64 end;
@@ -3518,16 +3518,12 @@ static void intel_iommu_unmap_range(struct iommu_domain *domain,
 				    unsigned long iova, size_t size)
 {
 	struct dmar_domain *dmar_domain = domain->priv;
-	dma_addr_t base;
 
-	/* The address might not be aligned */
-	base = iova & VTD_PAGE_MASK;
-	size = VTD_PAGE_ALIGN(size);
-	dma_pte_clear_range(dmar_domain, base >> VTD_PAGE_SHIFT,
-			    (base + size - 1) >> VTD_PAGE_SHIFT);
+	dma_pte_clear_range(dmar_domain, iova >> VTD_PAGE_SHIFT,
+			    (iova + size - 1) >> VTD_PAGE_SHIFT);
 
-	if (dmar_domain->max_addr == base + size)
-		dmar_domain->max_addr = base;
+	if (dmar_domain->max_addr == iova + size)
+		dmar_domain->max_addr = iova;
 }
 
 static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,
-- 
cgit v1.2.3-59-g8ed1b


From d6d3f08b0fd998b647a05540cedd11a067b72867 Mon Sep 17 00:00:00 2001
From: Jan Engelhardt <jengelh@medozas.de>
Date: Mon, 29 Jun 2009 14:31:46 +0200
Subject: netfilter: xtables: conntrack match revision 2

As reported by Philip, the UNTRACKED state bit does not fit within
the 8-bit state_mask member. Enlarge state_mask and give status_mask
a few more bits too.

Reported-by: Philip Craig <philipc@snapgear.com>
References: http://markmail.org/thread/b7eg6aovfh4agyz7
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
---
 include/linux/netfilter/xt_conntrack.h | 13 +++++++
 net/netfilter/xt_conntrack.c           | 66 ++++++++++++++++++++++++++++++----
 2 files changed, 73 insertions(+), 6 deletions(-)

diff --git a/include/linux/netfilter/xt_conntrack.h b/include/linux/netfilter/xt_conntrack.h
index 3430c7751948..7ae05338e94c 100644
--- a/include/linux/netfilter/xt_conntrack.h
+++ b/include/linux/netfilter/xt_conntrack.h
@@ -81,4 +81,17 @@ struct xt_conntrack_mtinfo1 {
 	__u8 state_mask, status_mask;
 };
 
+struct xt_conntrack_mtinfo2 {
+	union nf_inet_addr origsrc_addr, origsrc_mask;
+	union nf_inet_addr origdst_addr, origdst_mask;
+	union nf_inet_addr replsrc_addr, replsrc_mask;
+	union nf_inet_addr repldst_addr, repldst_mask;
+	__u32 expires_min, expires_max;
+	__u16 l4proto;
+	__be16 origsrc_port, origdst_port;
+	__be16 replsrc_port, repldst_port;
+	__u16 match_flags, invert_flags;
+	__u16 state_mask, status_mask;
+};
+
 #endif /*_XT_CONNTRACK_H*/
diff --git a/net/netfilter/xt_conntrack.c b/net/netfilter/xt_conntrack.c
index 0b7139f3dd78..fc581800698e 100644
--- a/net/netfilter/xt_conntrack.c
+++ b/net/netfilter/xt_conntrack.c
@@ -129,7 +129,7 @@ conntrack_addrcmp(const union nf_inet_addr *kaddr,
 
 static inline bool
 conntrack_mt_origsrc(const struct nf_conn *ct,
-                     const struct xt_conntrack_mtinfo1 *info,
+                     const struct xt_conntrack_mtinfo2 *info,
 		     u_int8_t family)
 {
 	return conntrack_addrcmp(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3,
@@ -138,7 +138,7 @@ conntrack_mt_origsrc(const struct nf_conn *ct,
 
 static inline bool
 conntrack_mt_origdst(const struct nf_conn *ct,
-                     const struct xt_conntrack_mtinfo1 *info,
+                     const struct xt_conntrack_mtinfo2 *info,
 		     u_int8_t family)
 {
 	return conntrack_addrcmp(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.u3,
@@ -147,7 +147,7 @@ conntrack_mt_origdst(const struct nf_conn *ct,
 
 static inline bool
 conntrack_mt_replsrc(const struct nf_conn *ct,
-                     const struct xt_conntrack_mtinfo1 *info,
+                     const struct xt_conntrack_mtinfo2 *info,
 		     u_int8_t family)
 {
 	return conntrack_addrcmp(&ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3,
@@ -156,7 +156,7 @@ conntrack_mt_replsrc(const struct nf_conn *ct,
 
 static inline bool
 conntrack_mt_repldst(const struct nf_conn *ct,
-                     const struct xt_conntrack_mtinfo1 *info,
+                     const struct xt_conntrack_mtinfo2 *info,
 		     u_int8_t family)
 {
 	return conntrack_addrcmp(&ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3,
@@ -164,7 +164,7 @@ conntrack_mt_repldst(const struct nf_conn *ct,
 }
 
 static inline bool
-ct_proto_port_check(const struct xt_conntrack_mtinfo1 *info,
+ct_proto_port_check(const struct xt_conntrack_mtinfo2 *info,
                     const struct nf_conn *ct)
 {
 	const struct nf_conntrack_tuple *tuple;
@@ -204,7 +204,7 @@ ct_proto_port_check(const struct xt_conntrack_mtinfo1 *info,
 static bool
 conntrack_mt(const struct sk_buff *skb, const struct xt_match_param *par)
 {
-	const struct xt_conntrack_mtinfo1 *info = par->matchinfo;
+	const struct xt_conntrack_mtinfo2 *info = par->matchinfo;
 	enum ip_conntrack_info ctinfo;
 	const struct nf_conn *ct;
 	unsigned int statebit;
@@ -278,6 +278,16 @@ conntrack_mt(const struct sk_buff *skb, const struct xt_match_param *par)
 	return true;
 }
 
+static bool
+conntrack_mt_v1(const struct sk_buff *skb, const struct xt_match_param *par)
+{
+	const struct xt_conntrack_mtinfo2 *const *info = par->matchinfo;
+	struct xt_match_param newpar = *par;
+
+	newpar.matchinfo = *info;
+	return conntrack_mt(skb, &newpar);
+}
+
 static bool conntrack_mt_check(const struct xt_mtchk_param *par)
 {
 	if (nf_ct_l3proto_try_module_get(par->family) < 0) {
@@ -288,11 +298,45 @@ static bool conntrack_mt_check(const struct xt_mtchk_param *par)
 	return true;
 }
 
+static bool conntrack_mt_check_v1(const struct xt_mtchk_param *par)
+{
+	struct xt_conntrack_mtinfo1 *info = par->matchinfo;
+	struct xt_conntrack_mtinfo2 *up;
+	int ret = conntrack_mt_check(par);
+
+	if (ret < 0)
+		return ret;
+
+	up = kmalloc(sizeof(*up), GFP_KERNEL);
+	if (up == NULL) {
+		nf_ct_l3proto_module_put(par->family);
+		return -ENOMEM;
+	}
+
+	/*
+	 * The strategy here is to minimize the overhead of v1 matching,
+	 * by prebuilding a v2 struct and putting the pointer into the
+	 * v1 dataspace.
+	 */
+	memcpy(up, info, offsetof(typeof(*info), state_mask));
+	up->state_mask  = info->state_mask;
+	up->status_mask = info->status_mask;
+	*(void **)info  = up;
+	return true;
+}
+
 static void conntrack_mt_destroy(const struct xt_mtdtor_param *par)
 {
 	nf_ct_l3proto_module_put(par->family);
 }
 
+static void conntrack_mt_destroy_v1(const struct xt_mtdtor_param *par)
+{
+	struct xt_conntrack_mtinfo2 **info = par->matchinfo;
+	kfree(*info);
+	conntrack_mt_destroy(par);
+}
+
 #ifdef CONFIG_COMPAT
 struct compat_xt_conntrack_info
 {
@@ -363,6 +407,16 @@ static struct xt_match conntrack_mt_reg[] __read_mostly = {
 		.revision   = 1,
 		.family     = NFPROTO_UNSPEC,
 		.matchsize  = sizeof(struct xt_conntrack_mtinfo1),
+		.match      = conntrack_mt_v1,
+		.checkentry = conntrack_mt_check_v1,
+		.destroy    = conntrack_mt_destroy_v1,
+		.me         = THIS_MODULE,
+	},
+	{
+		.name       = "conntrack",
+		.revision   = 2,
+		.family     = NFPROTO_UNSPEC,
+		.matchsize  = sizeof(struct xt_conntrack_mtinfo2),
 		.match      = conntrack_mt,
 		.checkentry = conntrack_mt_check,
 		.destroy    = conntrack_mt_destroy,
-- 
cgit v1.2.3-59-g8ed1b


From b026fd28ea23af24a3eea6e5be3f3d0193a8e87d Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 10:37:25 +0100
Subject: intel-iommu: Change addr_to_dma_pte() to pfn_to_dma_pte()

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 21dc77311863..dfbabd151a9c 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -692,23 +692,24 @@ static inline unsigned long align_to_level(unsigned long pfn, int level)
 	return (pfn + level_size(level) - 1) & level_mask(level);
 }
 
-static struct dma_pte * addr_to_dma_pte(struct dmar_domain *domain, u64 addr)
+static struct dma_pte *pfn_to_dma_pte(struct dmar_domain *domain,
+				      unsigned long pfn)
 {
-	int addr_width = agaw_to_width(domain->agaw);
+	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
 	struct dma_pte *parent, *pte = NULL;
 	int level = agaw_to_level(domain->agaw);
 	int offset;
 	unsigned long flags;
 
 	BUG_ON(!domain->pgd);
-	BUG_ON(addr >> addr_width);
+	BUG_ON(addr_width < BITS_PER_LONG && pfn >> addr_width);
 	parent = domain->pgd;
 
 	spin_lock_irqsave(&domain->mapping_lock, flags);
 	while (level > 0) {
 		void *tmp_page;
 
-		offset = pfn_level_offset(addr >> VTD_PAGE_SHIFT, level);
+		offset = pfn_level_offset(pfn, level);
 		pte = &parent[offset];
 		if (level == 1)
 			break;
@@ -1660,7 +1661,7 @@ domain_page_mapping(struct dmar_domain *domain, dma_addr_t iova,
 	end_pfn = (VTD_PAGE_ALIGN(((u64)hpa) + size)) >> VTD_PAGE_SHIFT;
 	index = 0;
 	while (start_pfn < end_pfn) {
-		pte = addr_to_dma_pte(domain, iova + VTD_PAGE_SIZE * index);
+		pte = pfn_to_dma_pte(domain, (iova >> VTD_PAGE_SHIFT) + index);
 		if (!pte)
 			return -ENOMEM;
 		/* We don't need lock here, nobody else
@@ -3533,7 +3534,7 @@ static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain,
 	struct dma_pte *pte;
 	u64 phys = 0;
 
-	pte = addr_to_dma_pte(dmar_domain, iova);
+	pte = pfn_to_dma_pte(dmar_domain, iova >> VTD_PAGE_SHIFT);
 	if (pte)
 		phys = dma_pte_addr(pte);
 
-- 
cgit v1.2.3-59-g8ed1b


From 1c5a46ed49e37f56f8aa9000bb1c2ac59670c372 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 10:53:37 +0100
Subject: intel-iommu: Clean up address handling in domain_page_mapping()

No more masking and alignment; just use pfns.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index dfbabd151a9c..f08d7865fe00 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1647,20 +1647,18 @@ static int
 domain_page_mapping(struct dmar_domain *domain, dma_addr_t iova,
 			u64 hpa, size_t size, int prot)
 {
-	u64 start_pfn, end_pfn;
+	unsigned long start_pfn = hpa >> VTD_PAGE_SHIFT;
+	unsigned long last_pfn = (hpa + size - 1) >> VTD_PAGE_SHIFT;
 	struct dma_pte *pte;
-	int index;
-	int addr_width = agaw_to_width(domain->agaw);
+	int index = 0;
+	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
 
-	BUG_ON(hpa >> addr_width);
+	BUG_ON(addr_width < BITS_PER_LONG && last_pfn >> addr_width);
 
 	if ((prot & (DMA_PTE_READ|DMA_PTE_WRITE)) == 0)
 		return -EINVAL;
-	iova &= PAGE_MASK;
-	start_pfn = ((u64)hpa) >> VTD_PAGE_SHIFT;
-	end_pfn = (VTD_PAGE_ALIGN(((u64)hpa) + size)) >> VTD_PAGE_SHIFT;
-	index = 0;
-	while (start_pfn < end_pfn) {
+
+	while (start_pfn <= last_pfn) {
 		pte = pfn_to_dma_pte(domain, (iova >> VTD_PAGE_SHIFT) + index);
 		if (!pte)
 			return -ENOMEM;
-- 
cgit v1.2.3-59-g8ed1b


From 61df744314079e8cb8cdec75f517cf0e704e41ef Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 11:55:58 +0100
Subject: intel-iommu: Introduce domain_pfn_mapping()

... and use it in the trivial cases; the other callers want individual
(and bisectable) attention, since I screwed them up the first time...

Make the BUG_ON() happen on too-large virtual address rather than
physical address, too. That's the one we care about.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 38 ++++++++++++++++++++++++--------------
 1 file changed, 24 insertions(+), 14 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index f08d7865fe00..7540ef91d5f7 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1643,40 +1643,48 @@ static int domain_context_mapped(struct pci_dev *pdev)
 					     tmp->devfn);
 }
 
-static int
-domain_page_mapping(struct dmar_domain *domain, dma_addr_t iova,
-			u64 hpa, size_t size, int prot)
+static int domain_pfn_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
+			      unsigned long phys_pfn, unsigned long nr_pages,
+			      int prot)
 {
-	unsigned long start_pfn = hpa >> VTD_PAGE_SHIFT;
-	unsigned long last_pfn = (hpa + size - 1) >> VTD_PAGE_SHIFT;
 	struct dma_pte *pte;
-	int index = 0;
 	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
 
-	BUG_ON(addr_width < BITS_PER_LONG && last_pfn >> addr_width);
+	BUG_ON(addr_width < BITS_PER_LONG && (iov_pfn + nr_pages - 1) >> addr_width);
 
 	if ((prot & (DMA_PTE_READ|DMA_PTE_WRITE)) == 0)
 		return -EINVAL;
 
-	while (start_pfn <= last_pfn) {
-		pte = pfn_to_dma_pte(domain, (iova >> VTD_PAGE_SHIFT) + index);
+	while (nr_pages--) {
+		pte = pfn_to_dma_pte(domain, iov_pfn);
 		if (!pte)
 			return -ENOMEM;
 		/* We don't need lock here, nobody else
 		 * touches the iova range
 		 */
 		BUG_ON(dma_pte_addr(pte));
-		dma_set_pte_pfn(pte, start_pfn);
+		dma_set_pte_pfn(pte, phys_pfn);
 		dma_set_pte_prot(pte, prot);
 		if (prot & DMA_PTE_SNP)
 			dma_set_pte_snp(pte);
 		domain_flush_cache(domain, pte, sizeof(*pte));
-		start_pfn++;
-		index++;
+		iov_pfn++;
+		phys_pfn++;
 	}
 	return 0;
 }
 
+static int domain_page_mapping(struct dmar_domain *domain, dma_addr_t iova,
+			       u64 hpa, size_t size, int prot)
+{
+	unsigned long first_pfn = hpa >> VTD_PAGE_SHIFT;
+	unsigned long last_pfn = (hpa + size - 1) >> VTD_PAGE_SHIFT;
+
+	return domain_pfn_mapping(domain, iova >> VTD_PAGE_SHIFT, first_pfn,
+				  last_pfn - first_pfn + 1, prot);
+
+}
+
 static void iommu_detach_dev(struct intel_iommu *iommu, u8 bus, u8 devfn)
 {
 	if (!iommu)
@@ -1893,8 +1901,10 @@ static int iommu_domain_identity_map(struct dmar_domain *domain,
 	dma_pte_clear_range(domain, base >> VTD_PAGE_SHIFT,
 			    (base + size - 1) >> VTD_PAGE_SHIFT);
 
-	return domain_page_mapping(domain, base, base, size,
-				   DMA_PTE_READ|DMA_PTE_WRITE);
+	return domain_pfn_mapping(domain, base >> VTD_PAGE_SHIFT,
+				  base >> VTD_PAGE_SHIFT,
+				  size >> VTD_PAGE_SHIFT,
+				  DMA_PTE_READ|DMA_PTE_WRITE);
 }
 
 static int iommu_prepare_identity_map(struct pci_dev *pdev,
-- 
cgit v1.2.3-59-g8ed1b


From 0ab36de274ab094c3992b50c9c48c5c89072ec94 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 14:01:43 +0100
Subject: intel-iommu: Use domain_pfn_mapping() in __intel_map_single()

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 18 ++++++++----------
 1 file changed, 8 insertions(+), 10 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 7540ef91d5f7..dccd0a7b7a5f 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -2477,14 +2477,12 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr,
 		return 0;
 
 	iommu = domain_get_iommu(domain);
-	size = aligned_size((u64)paddr, size);
+	size = aligned_size(paddr, size) >> VTD_PAGE_SHIFT;
 
-	iova = __intel_alloc_iova(hwdev, domain, size, pdev->dma_mask);
+	iova = __intel_alloc_iova(hwdev, domain, size << VTD_PAGE_SHIFT, pdev->dma_mask);
 	if (!iova)
 		goto error;
 
-	start_paddr = (phys_addr_t)iova->pfn_lo << PAGE_SHIFT;
-
 	/*
 	 * Check if DMAR supports zero-length reads on write only
 	 * mappings..
@@ -2500,20 +2498,20 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr,
 	 * might have two guest_addr mapping to the same host paddr, but this
 	 * is not a big problem
 	 */
-	ret = domain_page_mapping(domain, start_paddr,
-				  ((u64)paddr) & PHYSICAL_PAGE_MASK,
-				  size, prot);
+	ret = domain_pfn_mapping(domain, mm_to_dma_pfn(iova->pfn_lo),
+				 paddr >> VTD_PAGE_SHIFT, size, prot);
 	if (ret)
 		goto error;
 
+	start_paddr = (phys_addr_t)iova->pfn_lo << PAGE_SHIFT;
+
 	/* it's a non-present to present mapping. Only flush if caching mode */
 	if (cap_caching_mode(iommu->cap))
-		iommu_flush_iotlb_psi(iommu, 0, start_paddr,
-				      size >> VTD_PAGE_SHIFT);
+		iommu_flush_iotlb_psi(iommu, 0, start_paddr, size);
 	else
 		iommu_flush_write_buffer(iommu);
 
-	return start_paddr + ((u64)paddr & (~PAGE_MASK));
+	return start_paddr + (paddr & (~PAGE_MASK));
 
 error:
 	if (iova)
-- 
cgit v1.2.3-59-g8ed1b


From ad05122162b67f64d5a1c6d35e001f7a88619b88 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 14:22:28 +0100
Subject: intel-iommu: Use domain_pfn_mapping() in intel_iommu_map_range()

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index dccd0a7b7a5f..a490b39ca3d5 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -3516,8 +3516,11 @@ static int intel_iommu_map_range(struct iommu_domain *domain,
 		}
 		dmar_domain->max_addr = max_addr;
 	}
-
-	ret = domain_page_mapping(dmar_domain, iova, hpa, size, prot);
+	/* Round up size to next multiple of PAGE_SIZE, if it and
+	   the low bits of hpa would take us onto the next page */
+	size = aligned_size(hpa, size) >> VTD_PAGE_SHIFT;
+	ret = domain_pfn_mapping(dmar_domain, iova >> VTD_PAGE_SHIFT,
+				 hpa >> VTD_PAGE_SHIFT, size, prot);
 	return ret;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From b536d24d212c994a7d98469ea3a8891573d45fd4 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 14:49:31 +0100
Subject: intel-iommu: Clean up intel_map_sg(), remove domain_page_mapping()

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 54 +++++++++++++++++------------------------------
 1 file changed, 19 insertions(+), 35 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index a490b39ca3d5..bc49b121c667 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1674,17 +1674,6 @@ static int domain_pfn_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
 	return 0;
 }
 
-static int domain_page_mapping(struct dmar_domain *domain, dma_addr_t iova,
-			       u64 hpa, size_t size, int prot)
-{
-	unsigned long first_pfn = hpa >> VTD_PAGE_SHIFT;
-	unsigned long last_pfn = (hpa + size - 1) >> VTD_PAGE_SHIFT;
-
-	return domain_pfn_mapping(domain, iova >> VTD_PAGE_SHIFT, first_pfn,
-				  last_pfn - first_pfn + 1, prot);
-
-}
-
 static void iommu_detach_dev(struct intel_iommu *iommu, u8 bus, u8 devfn)
 {
 	if (!iommu)
@@ -2745,17 +2734,16 @@ static int intel_nontranslate_map_sg(struct device *hddev,
 static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int nelems,
 			enum dma_data_direction dir, struct dma_attrs *attrs)
 {
-	phys_addr_t addr;
 	int i;
 	struct pci_dev *pdev = to_pci_dev(hwdev);
 	struct dmar_domain *domain;
 	size_t size = 0;
 	int prot = 0;
-	size_t offset = 0;
+	size_t offset_pfn = 0;
 	struct iova *iova = NULL;
 	int ret;
 	struct scatterlist *sg;
-	unsigned long start_addr;
+	unsigned long start_vpfn;
 	struct intel_iommu *iommu;
 
 	BUG_ON(dir == DMA_NONE);
@@ -2768,10 +2756,8 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne
 
 	iommu = domain_get_iommu(domain);
 
-	for_each_sg(sglist, sg, nelems, i) {
-		addr = page_to_phys(sg_page(sg)) + sg->offset;
-		size += aligned_size((u64)addr, sg->length);
-	}
+	for_each_sg(sglist, sg, nelems, i)
+		size += aligned_size(sg->offset, sg->length);
 
 	iova = __intel_alloc_iova(hwdev, domain, size, pdev->dma_mask);
 	if (!iova) {
@@ -2789,36 +2775,34 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne
 	if (dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL)
 		prot |= DMA_PTE_WRITE;
 
-	start_addr = iova->pfn_lo << PAGE_SHIFT;
-	offset = 0;
+	start_vpfn = mm_to_dma_pfn(iova->pfn_lo);
+	offset_pfn = 0;
 	for_each_sg(sglist, sg, nelems, i) {
-		addr = page_to_phys(sg_page(sg)) + sg->offset;
-		size = aligned_size((u64)addr, sg->length);
-		ret = domain_page_mapping(domain, start_addr + offset,
-					  ((u64)addr) & PHYSICAL_PAGE_MASK,
-					  size, prot);
+		int nr_pages = aligned_size(sg->offset, sg->length) >> VTD_PAGE_SHIFT;
+		ret = domain_pfn_mapping(domain, start_vpfn + offset_pfn,
+					 page_to_dma_pfn(sg_page(sg)),
+					 nr_pages, prot);
 		if (ret) {
 			/*  clear the page */
-			dma_pte_clear_range(domain,
-					    start_addr >> VTD_PAGE_SHIFT,
-					    (start_addr + offset - 1) >> VTD_PAGE_SHIFT);
+			dma_pte_clear_range(domain, start_vpfn,
+					    start_vpfn + offset_pfn);
 			/* free page tables */
-			dma_pte_free_pagetable(domain, start_addr >> VTD_PAGE_SHIFT,
-					       (start_addr + offset - 1) >> VTD_PAGE_SHIFT);
+			dma_pte_free_pagetable(domain, start_vpfn,
+					       start_vpfn + offset_pfn);
 			/* free iova */
 			__free_iova(&domain->iovad, iova);
 			return 0;
 		}
-		sg->dma_address = start_addr + offset +
-				((u64)addr & (~PAGE_MASK));
+		sg->dma_address = ((dma_addr_t)(start_vpfn + offset_pfn)
+				   << VTD_PAGE_SHIFT) + sg->offset;
 		sg->dma_length = sg->length;
-		offset += size;
+		offset_pfn += nr_pages;
 	}
 
 	/* it's a non-present to present mapping. Only flush if caching mode */
 	if (cap_caching_mode(iommu->cap))
-		iommu_flush_iotlb_psi(iommu, 0, start_addr,
-				      offset >> VTD_PAGE_SHIFT);
+		iommu_flush_iotlb_psi(iommu, 0, start_vpfn << VTD_PAGE_SHIFT,
+				      offset_pfn);
 	else
 		iommu_flush_write_buffer(iommu);
 
-- 
cgit v1.2.3-59-g8ed1b


From 88cb6a7424d9465faf6caaaadff5af0766c93991 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 15:03:06 +0100
Subject: intel-iommu: Change aligned_size() to aligned_nrpages()

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 21 ++++++++++++---------
 1 file changed, 12 insertions(+), 9 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index bc49b121c667..22add36fd731 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -2320,11 +2320,13 @@ error:
 	return ret;
 }
 
-static inline u64 aligned_size(u64 host_addr, size_t size)
+static inline unsigned long aligned_nrpages(unsigned long host_addr,
+					    size_t size)
 {
-	u64 addr;
-	addr = (host_addr & (~PAGE_MASK)) + size;
-	return PAGE_ALIGN(addr);
+	host_addr &= ~PAGE_MASK;
+	host_addr += size + PAGE_SIZE - 1;
+
+	return host_addr >> VTD_PAGE_SHIFT;
 }
 
 struct iova *
@@ -2466,7 +2468,7 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr,
 		return 0;
 
 	iommu = domain_get_iommu(domain);
-	size = aligned_size(paddr, size) >> VTD_PAGE_SHIFT;
+	size = aligned_nrpages(paddr, size);
 
 	iova = __intel_alloc_iova(hwdev, domain, size << VTD_PAGE_SHIFT, pdev->dma_mask);
 	if (!iova)
@@ -2757,9 +2759,10 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne
 	iommu = domain_get_iommu(domain);
 
 	for_each_sg(sglist, sg, nelems, i)
-		size += aligned_size(sg->offset, sg->length);
+		size += aligned_nrpages(sg->offset, sg->length);
 
-	iova = __intel_alloc_iova(hwdev, domain, size, pdev->dma_mask);
+	iova = __intel_alloc_iova(hwdev, domain, size << VTD_PAGE_SHIFT,
+				  pdev->dma_mask);
 	if (!iova) {
 		sglist->dma_length = 0;
 		return 0;
@@ -2778,7 +2781,7 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne
 	start_vpfn = mm_to_dma_pfn(iova->pfn_lo);
 	offset_pfn = 0;
 	for_each_sg(sglist, sg, nelems, i) {
-		int nr_pages = aligned_size(sg->offset, sg->length) >> VTD_PAGE_SHIFT;
+		int nr_pages = aligned_nrpages(sg->offset, sg->length);
 		ret = domain_pfn_mapping(domain, start_vpfn + offset_pfn,
 					 page_to_dma_pfn(sg_page(sg)),
 					 nr_pages, prot);
@@ -3502,7 +3505,7 @@ static int intel_iommu_map_range(struct iommu_domain *domain,
 	}
 	/* Round up size to next multiple of PAGE_SIZE, if it and
 	   the low bits of hpa would take us onto the next page */
-	size = aligned_size(hpa, size) >> VTD_PAGE_SHIFT;
+	size = aligned_nrpages(hpa, size);
 	ret = domain_pfn_mapping(dmar_domain, iova >> VTD_PAGE_SHIFT,
 				 hpa >> VTD_PAGE_SHIFT, size, prot);
 	return ret;
-- 
cgit v1.2.3-59-g8ed1b


From 03d6a2461ab1704c171ce21081c5022378ef7a91 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 15:33:46 +0100
Subject: intel-iommu: Make iommu_flush_iotlb_psi() take pfn as argument

Most of its callers are having to shift for themselves anyway, so we might
as well do it in iommu_flush_iotlb_psi().

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 21 +++++++++------------
 1 file changed, 9 insertions(+), 12 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 22add36fd731..6afe44cb6815 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1058,11 +1058,11 @@ static void iommu_flush_dev_iotlb(struct dmar_domain *domain,
 }
 
 static void iommu_flush_iotlb_psi(struct intel_iommu *iommu, u16 did,
-				  u64 addr, unsigned int pages)
+				  unsigned long pfn, unsigned int pages)
 {
 	unsigned int mask = ilog2(__roundup_pow_of_two(pages));
+	uint64_t addr = (uint64_t)pfn << VTD_PAGE_SHIFT;
 
-	BUG_ON(addr & (~VTD_PAGE_MASK));
 	BUG_ON(pages == 0);
 
 	/*
@@ -2494,15 +2494,15 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr,
 	if (ret)
 		goto error;
 
-	start_paddr = (phys_addr_t)iova->pfn_lo << PAGE_SHIFT;
-
 	/* it's a non-present to present mapping. Only flush if caching mode */
 	if (cap_caching_mode(iommu->cap))
-		iommu_flush_iotlb_psi(iommu, 0, start_paddr, size);
+		iommu_flush_iotlb_psi(iommu, 0, mm_to_dma_pfn(iova->pfn_lo), size);
 	else
 		iommu_flush_write_buffer(iommu);
 
-	return start_paddr + (paddr & (~PAGE_MASK));
+	start_paddr = (phys_addr_t)iova->pfn_lo << PAGE_SHIFT;
+	start_paddr += paddr & ~PAGE_MASK;
+	return start_paddr;
 
 error:
 	if (iova)
@@ -2624,8 +2624,7 @@ static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr,
 	dma_pte_free_pagetable(domain, start_pfn, last_pfn);
 
 	if (intel_iommu_strict) {
-		iommu_flush_iotlb_psi(iommu, domain->id,
-				      start_pfn << VTD_PAGE_SHIFT,
+		iommu_flush_iotlb_psi(iommu, domain->id, start_pfn,
 				      last_pfn - start_pfn + 1);
 		/* free iova */
 		__free_iova(&domain->iovad, iova);
@@ -2711,8 +2710,7 @@ static void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist,
 	/* free page tables */
 	dma_pte_free_pagetable(domain, start_pfn, last_pfn);
 
-	iommu_flush_iotlb_psi(iommu, domain->id,
-			      start_pfn << VTD_PAGE_SHIFT,
+	iommu_flush_iotlb_psi(iommu, domain->id, start_pfn,
 			      (last_pfn - start_pfn + 1));
 
 	/* free iova */
@@ -2804,8 +2802,7 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne
 
 	/* it's a non-present to present mapping. Only flush if caching mode */
 	if (cap_caching_mode(iommu->cap))
-		iommu_flush_iotlb_psi(iommu, 0, start_vpfn << VTD_PAGE_SHIFT,
-				      offset_pfn);
+		iommu_flush_iotlb_psi(iommu, 0, start_vpfn, offset_pfn);
 	else
 		iommu_flush_write_buffer(iommu);
 
-- 
cgit v1.2.3-59-g8ed1b


From 1a4a45516d7a57de0691352d899d7008f2e090d1 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 16:00:42 +0100
Subject: intel-iommu: Remove last use of PHYSICAL_PAGE_MASK, for reserving PCI
 BARs

This is fairly broken anyway -- it doesn't take hotplug into account.
We should probably be checking page_is_ram() instead.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 13 +++----------
 1 file changed, 3 insertions(+), 10 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 6afe44cb6815..a55f5fb06b14 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -62,9 +62,6 @@
 #define DMA_32BIT_PFN		IOVA_PFN(DMA_BIT_MASK(32))
 #define DMA_64BIT_PFN		IOVA_PFN(DMA_BIT_MASK(64))
 
-#ifndef PHYSICAL_PAGE_MASK
-#define PHYSICAL_PAGE_MASK PAGE_MASK
-#endif
 
 /* VT-d pages must always be _smaller_ than MM pages. Otherwise things
    are never going to work. */
@@ -1307,7 +1304,6 @@ static void dmar_init_reserved_ranges(void)
 	struct pci_dev *pdev = NULL;
 	struct iova *iova;
 	int i;
-	u64 addr, size;
 
 	init_iova_domain(&reserved_iova_list, DMA_32BIT_PFN);
 
@@ -1330,12 +1326,9 @@ static void dmar_init_reserved_ranges(void)
 			r = &pdev->resource[i];
 			if (!r->flags || !(r->flags & IORESOURCE_MEM))
 				continue;
-			addr = r->start;
-			addr &= PHYSICAL_PAGE_MASK;
-			size = r->end - addr;
-			size = PAGE_ALIGN(size);
-			iova = reserve_iova(&reserved_iova_list, IOVA_PFN(addr),
-				IOVA_PFN(size + addr) - 1);
+			iova = reserve_iova(&reserved_iova_list,
+					    IOVA_PFN(r->start),
+					    IOVA_PFN(r->end));
 			if (!iova)
 				printk(KERN_ERR "Reserve iova failed\n");
 		}
-- 
cgit v1.2.3-59-g8ed1b


From c5395d5c4a82159889cb650de93b591ea51d8c56 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 16:35:56 +0100
Subject: intel-iommu: Clean up iommu_domain_identity_map()

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 24 +++++++++---------------
 1 file changed, 9 insertions(+), 15 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index a55f5fb06b14..c5caf7d63a0f 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1861,31 +1861,25 @@ static int iommu_domain_identity_map(struct dmar_domain *domain,
 				     unsigned long long start,
 				     unsigned long long end)
 {
-	unsigned long size;
-	unsigned long long base;
+	unsigned long first_vpfn = start >> VTD_PAGE_SHIFT;
+	unsigned long last_vpfn = end >> VTD_PAGE_SHIFT;
 
-	/* The address might not be aligned */
-	base = start & PAGE_MASK;
-	size = end - base;
-	size = PAGE_ALIGN(size);
-	if (!reserve_iova(&domain->iovad, IOVA_PFN(base),
-			IOVA_PFN(base + size) - 1)) {
+	if (!reserve_iova(&domain->iovad, dma_to_mm_pfn(first_vpfn),
+			  dma_to_mm_pfn(last_vpfn))) {
 		printk(KERN_ERR "IOMMU: reserve iova failed\n");
 		return -ENOMEM;
 	}
 
-	pr_debug("Mapping reserved region %lx@%llx for domain %d\n",
-		 size, base, domain->id);
+	pr_debug("Mapping reserved region %llx-%llx for domain %d\n",
+		 start, end, domain->id);
 	/*
 	 * RMRR range might have overlap with physical memory range,
 	 * clear it first
 	 */
-	dma_pte_clear_range(domain, base >> VTD_PAGE_SHIFT,
-			    (base + size - 1) >> VTD_PAGE_SHIFT);
+	dma_pte_clear_range(domain, first_vpfn, last_vpfn);
 
-	return domain_pfn_mapping(domain, base >> VTD_PAGE_SHIFT,
-				  base >> VTD_PAGE_SHIFT,
-				  size >> VTD_PAGE_SHIFT,
+	return domain_pfn_mapping(domain, first_vpfn, first_vpfn,
+				  last_vpfn - first_vpfn + 1,
 				  DMA_PTE_READ|DMA_PTE_WRITE);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 310a5ab93cb4ce29367238f682affd9ac352f4d0 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 18:52:20 +0100
Subject: intel-iommu: Performance improvement for dma_pte_clear_range()

It's a bit silly to repeatedly call domain_flush_cache() for each PTE
individually, as we clear it. Instead, batch them up and flush a whole
range at a time. We might as well refrain from recalculating the PTE
address from scratch each time round the loop too.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 31 +++++++++++++++----------------
 1 file changed, 15 insertions(+), 16 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index c5caf7d63a0f..ba7e37f7111a 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -761,34 +761,33 @@ static struct dma_pte *dma_pfn_level_pte(struct dmar_domain *domain,
 	return NULL;
 }
 
-/* clear one page's page table */
-static void dma_pte_clear_one(struct dmar_domain *domain, unsigned long pfn)
-{
-	struct dma_pte *pte = NULL;
-
-	/* get last level pte */
-	pte = dma_pfn_level_pte(domain, pfn, 1);
-
-	if (pte) {
-		dma_clear_pte(pte);
-		domain_flush_cache(domain, pte, sizeof(*pte));
-	}
-}
-
 /* clear last level pte, a tlb flush should be followed */
 static void dma_pte_clear_range(struct dmar_domain *domain,
 				unsigned long start_pfn,
 				unsigned long last_pfn)
 {
 	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
+	struct dma_pte *first_pte, *pte;
 
 	BUG_ON(addr_width < BITS_PER_LONG && start_pfn >> addr_width);
 	BUG_ON(addr_width < BITS_PER_LONG && last_pfn >> addr_width);
 
 	/* we don't need lock here; nobody else touches the iova range */
 	while (start_pfn <= last_pfn) {
-		dma_pte_clear_one(domain, start_pfn);
-		start_pfn++;
+		first_pte = pte = dma_pfn_level_pte(domain, start_pfn, 1);
+		if (!pte) {
+			start_pfn = align_to_level(start_pfn + 1, 2);
+			continue;
+		}
+		while (start_pfn <= last_pfn &&
+		       (unsigned long)pte >> VTD_PAGE_SHIFT ==
+		       (unsigned long)first_pte >> VTD_PAGE_SHIFT) {
+			dma_clear_pte(pte);
+			start_pfn++;
+			pte++;
+		}
+		domain_flush_cache(domain, first_pte,
+				   (void *)pte - (void *)first_pte);
 	}
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 6f6a00e40aa3fdd3b29c30e3ef1fc9690506bc03 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 20:38:49 +0100
Subject: intel-iommu: Performance improvement for domain_pfn_mapping()

As with dma_pte_clear_range(), don't keep flushing a single PTE at a
time. And also micro-optimise the setting of PTE values rather than
using the helper functions to do all the masking.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 26 +++++++++++++++++---------
 1 file changed, 17 insertions(+), 9 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index ba7e37f7111a..f8074236bcce 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1639,7 +1639,7 @@ static int domain_pfn_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
 			      unsigned long phys_pfn, unsigned long nr_pages,
 			      int prot)
 {
-	struct dma_pte *pte;
+	struct dma_pte *first_pte = NULL, *pte = NULL;
 	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
 
 	BUG_ON(addr_width < BITS_PER_LONG && (iov_pfn + nr_pages - 1) >> addr_width);
@@ -1647,19 +1647,27 @@ static int domain_pfn_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
 	if ((prot & (DMA_PTE_READ|DMA_PTE_WRITE)) == 0)
 		return -EINVAL;
 
+	prot &= DMA_PTE_READ | DMA_PTE_WRITE | DMA_PTE_SNP;
+
 	while (nr_pages--) {
-		pte = pfn_to_dma_pte(domain, iov_pfn);
-		if (!pte)
-			return -ENOMEM;
+		if (!pte) {
+			first_pte = pte = pfn_to_dma_pte(domain, iov_pfn);
+			if (!pte)
+				return -ENOMEM;
+		}
 		/* We don't need lock here, nobody else
 		 * touches the iova range
 		 */
 		BUG_ON(dma_pte_addr(pte));
-		dma_set_pte_pfn(pte, phys_pfn);
-		dma_set_pte_prot(pte, prot);
-		if (prot & DMA_PTE_SNP)
-			dma_set_pte_snp(pte);
-		domain_flush_cache(domain, pte, sizeof(*pte));
+		pte->val = (phys_pfn << VTD_PAGE_SHIFT) | prot;
+		pte++;
+		if (!nr_pages ||
+		    (unsigned long)pte >> VTD_PAGE_SHIFT !=
+		    (unsigned long)first_pte >> VTD_PAGE_SHIFT) {
+			domain_flush_cache(domain, first_pte,
+					   (void *)pte - (void *)first_pte);
+			pte = NULL;
+		}
 		iov_pfn++;
 		phys_pfn++;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 875764de6f0ddb23d270c29357d5a339232a0488 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Sun, 28 Jun 2009 21:20:51 +0100
Subject: intel-iommu: Simplify __intel_alloc_iova()

There's no need for the separate iommu_alloc_iova() function, and
certainly not for it to be global. Remove the underscores while we're at
it.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 49 +++++++++++++++++------------------------------
 1 file changed, 18 insertions(+), 31 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index f8074236bcce..11a23201445a 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -2323,43 +2323,31 @@ static inline unsigned long aligned_nrpages(unsigned long host_addr,
 	return host_addr >> VTD_PAGE_SHIFT;
 }
 
-struct iova *
-iommu_alloc_iova(struct dmar_domain *domain, size_t size, u64 end)
-{
-	struct iova *piova;
-
-	/* Make sure it's in range */
-	end = min_t(u64, DOMAIN_MAX_ADDR(domain->gaw), end);
-	if (!size || (IOVA_START_ADDR + size > end))
-		return NULL;
-
-	piova = alloc_iova(&domain->iovad,
-			size >> PAGE_SHIFT, IOVA_PFN(end), 1);
-	return piova;
-}
-
-static struct iova *
-__intel_alloc_iova(struct device *dev, struct dmar_domain *domain,
-		   size_t size, u64 dma_mask)
+static struct iova *intel_alloc_iova(struct device *dev,
+				     struct dmar_domain *domain,
+				     unsigned long nrpages, uint64_t dma_mask)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
 	struct iova *iova = NULL;
 
-	if (dma_mask <= DMA_BIT_MASK(32) || dmar_forcedac)
-		iova = iommu_alloc_iova(domain, size, dma_mask);
-	else {
+	/* Restrict dma_mask to the width that the iommu can handle */
+	dma_mask = min_t(uint64_t, DOMAIN_MAX_ADDR(domain->gaw), dma_mask);
+
+	if (!dmar_forcedac && dma_mask > DMA_BIT_MASK(32)) {
 		/*
 		 * First try to allocate an io virtual address in
 		 * DMA_BIT_MASK(32) and if that fails then try allocating
 		 * from higher range
 		 */
-		iova = iommu_alloc_iova(domain, size, DMA_BIT_MASK(32));
-		if (!iova)
-			iova = iommu_alloc_iova(domain, size, dma_mask);
-	}
-
-	if (!iova) {
-		printk(KERN_ERR"Allocating iova for %s failed", pci_name(pdev));
+		iova = alloc_iova(&domain->iovad, nrpages,
+				  IOVA_PFN(DMA_BIT_MASK(32)), 1);
+		if (iova)
+			return iova;
+	}
+	iova = alloc_iova(&domain->iovad, nrpages, IOVA_PFN(dma_mask), 1);
+	if (unlikely(!iova)) {
+		printk(KERN_ERR "Allocating %ld-page iova for %s failed",
+		       nrpages, pci_name(pdev));
 		return NULL;
 	}
 
@@ -2464,7 +2452,7 @@ static dma_addr_t __intel_map_single(struct device *hwdev, phys_addr_t paddr,
 	iommu = domain_get_iommu(domain);
 	size = aligned_nrpages(paddr, size);
 
-	iova = __intel_alloc_iova(hwdev, domain, size << VTD_PAGE_SHIFT, pdev->dma_mask);
+	iova = intel_alloc_iova(hwdev, domain, size, pdev->dma_mask);
 	if (!iova)
 		goto error;
 
@@ -2753,8 +2741,7 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne
 	for_each_sg(sglist, sg, nelems, i)
 		size += aligned_nrpages(sg->offset, sg->length);
 
-	iova = __intel_alloc_iova(hwdev, domain, size << VTD_PAGE_SHIFT,
-				  pdev->dma_mask);
+	iova = intel_alloc_iova(hwdev, domain, size, pdev->dma_mask);
 	if (!iova) {
 		sglist->dma_length = 0;
 		return 0;
-- 
cgit v1.2.3-59-g8ed1b


From e6ce3066010a21bde961d8f8cefe0b69cae78a0f Mon Sep 17 00:00:00 2001
From: Jeremy Kerr <jk@ozlabs.org>
Date: Mon, 29 Jun 2009 14:31:58 +0800
Subject: fs: allow d_instantiate to be called with negative parent dentry

The new fsnotify infrastructure (starting at 90586523) causes an oops in
spufs, where we populate a directory with files before instantiating the
directory itself. The new changes seem to have introduced an assumption
that a dentry's parent will be positive when instantiating.

This change makes it once again possible to d_instantiate a dentry
with a negative parent, and brings __fsnotify_d_instantiate() into
line with inotify_d_instantiate(), which already has this NULL check.

Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
Signed-off-by: Eric Paris <eparis@redhat.com>
---
 include/linux/fsnotify_backend.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
index 44848aa830dc..6c3de999fb34 100644
--- a/include/linux/fsnotify_backend.h
+++ b/include/linux/fsnotify_backend.h
@@ -280,7 +280,7 @@ static inline void __fsnotify_update_dcache_flags(struct dentry *dentry)
 	assert_spin_locked(&dentry->d_lock);
 
 	parent = dentry->d_parent;
-	if (fsnotify_inode_watches_children(parent->d_inode))
+	if (parent->d_inode && fsnotify_inode_watches_children(parent->d_inode))
 		dentry->d_flags |= DCACHE_FSNOTIFY_PARENT_WATCHED;
 	else
 		dentry->d_flags &= ~DCACHE_FSNOTIFY_PARENT_WATCHED;
-- 
cgit v1.2.3-59-g8ed1b


From e0af6062aa4f89081afb8a1a4269605775d354de Mon Sep 17 00:00:00 2001
From: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Date: Thu, 18 Jun 2009 13:05:49 +0400
Subject: MAINTAINERS: ieee802154 lists are moderated for non-subscribers.

Note that our mailing list is moderated for non-subscribers.

Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
---
 MAINTAINERS | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index fa2a16def17a..28c150e916a2 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2886,7 +2886,7 @@ P:	Dmitry Eremin-Solenikov
 M:	dbaryshkov@gmail.com
 P:	Sergey Lapin
 M:	slapin@ossfans.org
-L:	linux-zigbee-devel@lists.sourceforge.net
+L:	linux-zigbee-devel@lists.sourceforge.net (moderated for non-subscribers)
 W:	http://apps.sourceforge.net/trac/linux-zigbee
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/lowpan/lowpan.git
 S:	Maintained
-- 
cgit v1.2.3-59-g8ed1b


From 932c1329acebc03ef5efa3647c9c3a967b59d0c4 Mon Sep 17 00:00:00 2001
From: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Date: Fri, 19 Jun 2009 17:00:08 +0400
Subject: nl802154: fix Oops in ieee802154_nl_get_dev

ieee802154_nl_get_dev() lacks check for the existance of the device
that was returned by dev_get_XXX, thus resulting in Oops for non-existing
devices. Fix it.

Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
---
 net/ieee802154/netlink.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/ieee802154/netlink.c b/net/ieee802154/netlink.c
index 105ad10876af..332b947ae812 100644
--- a/net/ieee802154/netlink.c
+++ b/net/ieee802154/netlink.c
@@ -276,6 +276,9 @@ static struct net_device *ieee802154_nl_get_dev(struct genl_info *info)
 	else
 		return NULL;
 
+	if (!dev)
+		return NULL;
+
 	if (dev->type != ARPHRD_IEEE802154) {
 		dev_put(dev);
 		return NULL;
-- 
cgit v1.2.3-59-g8ed1b


From dfd06fe8246c0425f8d6850b8e2c872b0d691ec3 Mon Sep 17 00:00:00 2001
From: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Date: Fri, 19 Jun 2009 17:02:09 +0400
Subject: nl802154: add module license and description

Signed-off-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
---
 net/ieee802154/netlink.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/ieee802154/netlink.c b/net/ieee802154/netlink.c
index 332b947ae812..27eda9fdf3c2 100644
--- a/net/ieee802154/netlink.c
+++ b/net/ieee802154/netlink.c
@@ -524,3 +524,6 @@ static void __exit ieee802154_nl_exit(void)
 }
 module_exit(ieee802154_nl_exit);
 
+MODULE_LICENSE("GPL v2");
+MODULE_DESCRIPTION("ieee 802.15.4 configuration interface");
+
-- 
cgit v1.2.3-59-g8ed1b


From 9ea5ca75a2aebb7172094a7d77acf6ff7600cc56 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Mon, 29 Jun 2009 18:03:34 +0200
Subject: sound: OSS: mpu401, fix deadlock

mpu401_chk_version is called with a spin lock already held. Don't take it
again.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
---
 sound/oss/mpu401.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/sound/oss/mpu401.c b/sound/oss/mpu401.c
index 6c0a770ed054..1b2316f35b1f 100644
--- a/sound/oss/mpu401.c
+++ b/sound/oss/mpu401.c
@@ -926,31 +926,21 @@ static struct midi_operations mpu401_midi_operations[MAX_MIDI_DEV];
 static void mpu401_chk_version(int n, struct mpu_config *devc)
 {
 	int tmp;
-	unsigned long flags;
 
 	devc->version = devc->revision = 0;
 
-	spin_lock_irqsave(&devc->lock,flags);
-	if ((tmp = mpu_cmd(n, 0xAC, 0)) < 0)
-	{
-		spin_unlock_irqrestore(&devc->lock,flags);
+	tmp = mpu_cmd(n, 0xAC, 0);
+	if (tmp < 0)
 		return;
-	}
 	if ((tmp & 0xf0) > 0x20)	/* Why it's larger than 2.x ??? */
-	{
-		spin_unlock_irqrestore(&devc->lock,flags);
 		return;
-	}
 	devc->version = tmp;
 
-	if ((tmp = mpu_cmd(n, 0xAD, 0)) < 0)
-	{
+	if ((tmp = mpu_cmd(n, 0xAD, 0)) < 0) {
 		devc->version = 0;
-		spin_unlock_irqrestore(&devc->lock,flags);
 		return;
 	}
 	devc->revision = tmp;
-	spin_unlock_irqrestore(&devc->lock,flags);
 }
 
 int attach_mpu401(struct address_info *hw_config, struct module *owner)
-- 
cgit v1.2.3-59-g8ed1b


From 17bb9e0d906b625f86e9d31740bb1c35bc0f63d7 Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Mon, 29 Jun 2009 17:13:56 +0100
Subject: kmemleak: Do not report new leaked objects if the scanning was
 stopped

If the scanning was stopped with a signal, it is possible that some
objects are left with a white colour (potential leaks) and reported. Add
a check to avoid reporting such objects.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 mm/kmemleak.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index c37e8e50e4de..e094c4dbdf55 100644
--- a/mm/kmemleak.c
+++ b/mm/kmemleak.c
@@ -1029,6 +1029,12 @@ static void kmemleak_scan(void)
 	}
 	WARN_ON(!list_empty(&gray_list));
 
+	/*
+	 * If scanning was stopped do not report any new unreferenced objects.
+	 */
+	if (scan_should_stop())
+		return;
+
 	/*
 	 * Scanning result reporting.
 	 */
@@ -1184,11 +1190,10 @@ static int kmemleak_seq_show(struct seq_file *seq, void *v)
 	unsigned long flags;
 
 	spin_lock_irqsave(&object->lock, flags);
-	if (!unreferenced_object(object))
-		goto out;
-	print_unreferenced(seq, object);
-	reported_leaks++;
-out:
+	if ((object->flags & OBJECT_REPORTED) && unreferenced_object(object)) {
+		print_unreferenced(seq, object);
+		reported_leaks++;
+	}
 	spin_unlock_irqrestore(&object->lock, flags);
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From b6e687221eb840bacd4d4a991e5f8e7ed3ae910a Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Mon, 29 Jun 2009 17:13:57 +0100
Subject: kmemleak: Do not warn if an unknown object is freed

vmap'ed memory blocks are not tracked by kmemleak (yet) but they may be
released with vfree() which is tracked. The corresponding kmemleak
warning is only enabled in debug mode. Future patch will add support for
ioremap and vmap.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 mm/kmemleak.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index e094c4dbdf55..eeece2deace2 100644
--- a/mm/kmemleak.c
+++ b/mm/kmemleak.c
@@ -531,8 +531,10 @@ static void delete_object(unsigned long ptr)
 	write_lock_irqsave(&kmemleak_lock, flags);
 	object = lookup_object(ptr, 0);
 	if (!object) {
+#ifdef DEBUG
 		kmemleak_warn("Freeing unknown object at 0x%08lx\n",
 			      ptr);
+#endif
 		write_unlock_irqrestore(&kmemleak_lock, flags);
 		return;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 12de38b186c2af97bf0b4a1f907f766df46b1def Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Mon, 29 Jun 2009 17:13:55 +0100
Subject: kmemleak: Inform kmemleak about pid_hash

Kmemleak does not track alloc_bootmem calls but the pid_hash allocated
in pidhash_init() would need to be scanned as it contains pointers to
struct pid objects.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
---
 kernel/pid.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/kernel/pid.c b/kernel/pid.c
index 31310b5d3f50..5fa1db48d8b7 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -36,6 +36,7 @@
 #include <linux/pid_namespace.h>
 #include <linux/init_task.h>
 #include <linux/syscalls.h>
+#include <linux/kmemleak.h>
 
 #define pid_hashfn(nr, ns)	\
 	hash_long((unsigned long)nr + (unsigned long)ns, pidhash_shift)
@@ -512,6 +513,12 @@ void __init pidhash_init(void)
 	pid_hash = alloc_bootmem(pidhash_size *	sizeof(*(pid_hash)));
 	if (!pid_hash)
 		panic("Could not alloc pidhash!\n");
+	/*
+	 * pid_hash contains references to allocated struct pid objects and it
+	 * must be scanned by kmemleak to avoid false positives.
+	 */
+	kmemleak_alloc(pid_hash, pidhash_size *	sizeof(*(pid_hash)), 0,
+		       GFP_KERNEL);
 	for (i = 0; i < pidhash_size; i++)
 		INIT_HLIST_HEAD(&pid_hash[i]);
 }
-- 
cgit v1.2.3-59-g8ed1b


From aef29bc2603014cb28dfe39bab8d888546fe18e7 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@linux.intel.com>
Date: Mon, 29 Jun 2009 15:21:47 +0100
Subject: tty: Fix the leak in tty_ldisc_release

Currently we reinit the ldisc on final tty close which is what the old code
did to ensure that if the device retained its termios settings then it had the
right ldisc. tty_ldisc_reinit does that but also leaves us with the reset
ldisc reference which is then leaked.

At this point we know the port will be recycled so we can kill the ldisc
off completely rather than try and add another ldisc free up when the kref
count hits zero.

At this point it is safe to keep the ldisc closed as tty_ldisc waiting
methods are only used from the user side, and as the final close we are
the last such reference. Interrupt/driver side methods will always use the
non wait version and get back a NULL.

Found with kmemleak and investigated/identified by Catalin Marinas.

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/tty_ldisc.c | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/drivers/char/tty_ldisc.c b/drivers/char/tty_ldisc.c
index a19e935847b0..913aa8d3f1c5 100644
--- a/drivers/char/tty_ldisc.c
+++ b/drivers/char/tty_ldisc.c
@@ -867,15 +867,22 @@ void tty_ldisc_release(struct tty_struct *tty, struct tty_struct *o_tty)
 	tty_ldisc_wait_idle(tty);
 
 	/*
-	 * Shutdown the current line discipline, and reset it to N_TTY.
-	 *
-	 * FIXME: this MUST get fixed for the new reflocking
+	 * Now kill off the ldisc
 	 */
+	tty_ldisc_close(tty, tty->ldisc);
+	tty_ldisc_put(tty->ldisc);
+	/* Force an oops if we mess this up */
+	tty->ldisc = NULL;
+
+	/* Ensure the next open requests the N_TTY ldisc */
+	tty_set_termios_ldisc(tty, N_TTY);
 
-	tty_ldisc_reinit(tty);
 	/* This will need doing differently if we need to lock */
 	if (o_tty)
 		tty_ldisc_release(o_tty, NULL);
+
+	/* And the memory resources remaining (buffers, termios) will be
+	   disposed of when the kref hits zero */
 }
 
 /**
-- 
cgit v1.2.3-59-g8ed1b


From 44b3615b8cb3b016a49eb7ef4236e77a77793cec Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 29 Jun 2009 10:07:54 +0200
Subject: eeepc-laptop: Fix build failure with HOTPLUG_PCI && !SYSFS

FYI, there's a post-rc1 build regression with certain configs:

 drivers/built-in.o: In function `pci_hp_deregister':
 (.text+0xb166): undefined reference to `pci_hp_remove_module_link'
 drivers/built-in.o: In function `pci_hp_deregister':
 (.text+0xb19f): undefined reference to `pci_destroy_slot'
 drivers/built-in.o: In function `__pci_hp_register':
 (.text+0xb583): undefined reference to `pci_create_slot'
 drivers/built-in.o: In function `__pci_hp_register':
 (.text+0xb5b1): undefined reference to `pci_hp_create_module_link'
 make: *** [.tmp_vmlinux1] Error 1

Caused by:

| 2b121bc262fa03c94e653b2d44356c2f86c1bcdc is first bad commit
| commit 2b121bc262fa03c94e653b2d44356c2f86c1bcdc
| Date:   Thu Jun 25 13:25:36 2009 +0200
|
|     eeepc-laptop: Register as a pci-hotplug device

which changed the driver to use the PCI hotplug infrastructure, but
didn't do a good job on the Kconfig rules.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Acked-by: Len Brown <len.brown@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/platform/x86/Kconfig | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig
index fee6a4022bc1..46dad12f952f 100644
--- a/drivers/platform/x86/Kconfig
+++ b/drivers/platform/x86/Kconfig
@@ -355,10 +355,9 @@ config EEEPC_LAPTOP
 	depends on INPUT
 	depends on EXPERIMENTAL
 	depends on RFKILL || RFKILL = n
+	depends on HOTPLUG_PCI
 	select BACKLIGHT_CLASS_DEVICE
 	select HWMON
-	select HOTPLUG
-	select HOTPLUG_PCI if PCI
 	---help---
 	  This driver supports the Fn-Fx keys on Eee PC laptops.
 
-- 
cgit v1.2.3-59-g8ed1b


From 210ad39fb7ef0bc0494483f517f42524f16bb2a7 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 29 Jun 2009 21:50:54 +0200
Subject: perf stat: Use percentages for scaling output

Peter expressed a strong preference for percentage based
display of scaled values - so revert to that from the
recently introduced multiplication-factor unit.

Reported-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Jaswinder Singh Rajput <jaswinder@kernel.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-stat.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 3e5ea4e2e5fd..c5a290727a92 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -307,7 +307,8 @@ static void print_counter(int counter)
 		abs_printout(counter, count, noise);
 
 	if (scaled)
-		fprintf(stderr, "  (%7.2fx scaled)", (double)count[1]/count[2]);
+		fprintf(stderr, "  (scaled from %.2f%%)",
+			(double) count[2] / count[1] * 100);
 
 	fprintf(stderr, "\n");
 }
-- 
cgit v1.2.3-59-g8ed1b


From 051ae7f7344f453616b6b10332d4d8e1d40ed823 Mon Sep 17 00:00:00 2001
From: Paul Mackerras <paulus@samba.org>
Date: Mon, 29 Jun 2009 21:13:21 +1000
Subject: perf_counter tools: Reduce perf stat measurement overhead/skew

Vince Weaver reported a 'perf stat' measurement overhead in the
count of retired instructions, which can amount to a +6000
instructions inflated count in the reported count.

At present, perf stat creates its counters on the perf process.  Thus
the counters count the fork and various other activity in both the
parent and child, such as the resolver overhead for resolving PLT
entries for any libc functions that haven't been called before, such
as execvp.

This reduces the overhead by creating the counters on the child process
after the fork, using a couple of pipes to synchronize so that the
child process waits until the parent has created the counters before
doing the exec.  To eliminate the PLT resolution overhead on calling
execvp, this does a dummy execvp first which will always fail.

With this, the overhead of executing a program goes down from over
4800 instructions to about 90 instructions on powerpc (32-bit).
This was measured with a statically-linked program written in
assembler which only does the 3 instructions needed to call _exit(0).

Before:

$ perf stat -e 0:1:u ./three

 Performance counter stats for './three':

           4858  instructions

    0.001274523  seconds time elapsed

After:

$ perf stat -e 0:1:u ./three

 Performance counter stats for './three':

             92  instructions

    0.000468153  seconds time elapsed

Reported-by: Vince Weaver <vince@deater.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <19016.41425.814043.870352@cargo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/builtin-stat.c | 64 ++++++++++++++++++++++++++++++++++++-----------
 1 file changed, 50 insertions(+), 14 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index c5a290727a92..201ef2367dcb 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -99,7 +99,7 @@ static u64			runtime_cycles_noise;
 #define ERR_PERF_OPEN \
 "Error: counter %d, sys_perf_counter_open() syscall returned with %d (%s)\n"
 
-static void create_perf_stat_counter(int counter)
+static void create_perf_stat_counter(int counter, int pid)
 {
 	struct perf_counter_attr *attr = attrs + counter;
 
@@ -119,7 +119,7 @@ static void create_perf_stat_counter(int counter)
 		attr->inherit	= inherit;
 		attr->disabled	= 1;
 
-		fd[0][counter] = sys_perf_counter_open(attr, 0, -1, -1, 0);
+		fd[0][counter] = sys_perf_counter_open(attr, pid, -1, -1, 0);
 		if (fd[0][counter] < 0 && verbose)
 			fprintf(stderr, ERR_PERF_OPEN, counter,
 				fd[0][counter], strerror(errno));
@@ -205,12 +205,58 @@ static int run_perf_stat(int argc, const char **argv)
 	int status = 0;
 	int counter;
 	int pid;
+	int child_ready_pipe[2], go_pipe[2];
+	char buf;
 
 	if (!system_wide)
 		nr_cpus = 1;
 
+	if (pipe(child_ready_pipe) < 0 || pipe(go_pipe) < 0) {
+		perror("failed to create pipes");
+		exit(1);
+	}
+
+	if ((pid = fork()) < 0)
+		perror("failed to fork");
+
+	if (!pid) {
+		close(child_ready_pipe[0]);
+		close(go_pipe[1]);
+		fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC);
+
+		/*
+		 * Do a dummy execvp to get the PLT entry resolved,
+		 * so we avoid the resolver overhead on the real
+		 * execvp call.
+		 */
+		execvp("", (char **)argv);
+
+		/*
+		 * Tell the parent we're ready to go
+		 */
+		close(child_ready_pipe[1]);
+
+		/*
+		 * Wait until the parent tells us to go.
+		 */
+		read(go_pipe[0], &buf, 1);
+
+		execvp(argv[0], (char **)argv);
+
+		perror(argv[0]);
+		exit(-1);
+	}
+
+	/*
+	 * Wait for the child to be ready to exec.
+	 */
+	close(child_ready_pipe[1]);
+	close(go_pipe[0]);
+	read(child_ready_pipe[0], &buf, 1);
+	close(child_ready_pipe[0]);
+
 	for (counter = 0; counter < nr_counters; counter++)
-		create_perf_stat_counter(counter);
+		create_perf_stat_counter(counter, pid);
 
 	/*
 	 * Enable counters and exec the command:
@@ -218,19 +264,9 @@ static int run_perf_stat(int argc, const char **argv)
 	t0 = rdclock();
 	prctl(PR_TASK_PERF_COUNTERS_ENABLE);
 
-	if ((pid = fork()) < 0)
-		perror("failed to fork");
-
-	if (!pid) {
-		if (execvp(argv[0], (char **)argv)) {
-			perror(argv[0]);
-			exit(-1);
-		}
-	}
-
+	close(go_pipe[1]);
 	wait(&status);
 
-	prctl(PR_TASK_PERF_COUNTERS_DISABLE);
 	t1 = rdclock();
 
 	walltime_nsecs[run_idx] = t1 - t0;
-- 
cgit v1.2.3-59-g8ed1b


From 2bf427b25b79eb7cea27963a66c3d4684cae0e0c Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Mon, 29 Jun 2009 19:20:42 -0700
Subject: ide: fix resume for CONFIG_BLK_DEV_IDEACPI=y

commit 2f0d0fd2a605666d38e290c5c0d2907484352dc4 ("ide-acpi: cleanup
do_drive_get_GTF()") didn't account for the lack of hwif->acpidata
check in generic_ide_suspend() [ indirect user of do_drive_get_GTF()
through ide_acpi_exec_tfs() ] resulting in broken resume when ACPI
support is enabled but ACPI data is unavailable.

Fix it by adding ide_port_acpi() helper for checking if port needs
ACPI handling and cleaning generic_ide_{suspend,resume}() to use it
instead of hiding hwif->acpidata and ide_noacpi checks in IDE ACPI
helpers (this should help in preventing similar bugs in the future).

While at it:
- kill superfluous debugging printks in ide_acpi_{get,push}_timing()

Reported-and-tested-by: Etienne Basset <etienne.basset@numericable.fr>
Also-reported-and-tested-by: Jeff Chua <jeff.chua.linux@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/ide-acpi.c | 37 +++++++------------------------------
 drivers/ide/ide-pm.c   | 30 ++++++++++++++++++------------
 include/linux/ide.h    |  2 ++
 3 files changed, 27 insertions(+), 42 deletions(-)

diff --git a/drivers/ide/ide-acpi.c b/drivers/ide/ide-acpi.c
index 77f79d26b264..c509c9916464 100644
--- a/drivers/ide/ide-acpi.c
+++ b/drivers/ide/ide-acpi.c
@@ -92,6 +92,11 @@ int ide_acpi_init(void)
 	return 0;
 }
 
+bool ide_port_acpi(ide_hwif_t *hwif)
+{
+	return ide_noacpi == 0 && hwif->acpidata;
+}
+
 /**
  * ide_get_dev_handle - finds acpi_handle and PCI device.function
  * @dev: device to locate
@@ -352,9 +357,6 @@ int ide_acpi_exec_tfs(ide_drive_t *drive)
 	unsigned long	gtf_address;
 	unsigned long	obj_loc;
 
-	if (ide_noacpi)
-		return 0;
-
 	DEBPRINT("call get_GTF, drive=%s port=%d\n", drive->name, drive->dn);
 
 	ret = do_drive_get_GTF(drive, &gtf_length, &gtf_address, &obj_loc);
@@ -389,16 +391,6 @@ void ide_acpi_get_timing(ide_hwif_t *hwif)
 	struct acpi_buffer	output;
 	union acpi_object 	*out_obj;
 
-	if (ide_noacpi)
-		return;
-
-	DEBPRINT("ENTER:\n");
-
-	if (!hwif->acpidata) {
-		DEBPRINT("no ACPI data for %s\n", hwif->name);
-		return;
-	}
-
 	/* Setting up output buffer for _GTM */
 	output.length = ACPI_ALLOCATE_BUFFER;
 	output.pointer = NULL;	/* ACPI-CA sets this; save/free it later */
@@ -479,16 +471,6 @@ void ide_acpi_push_timing(ide_hwif_t *hwif)
 	struct ide_acpi_drive_link	*master = &hwif->acpidata->master;
 	struct ide_acpi_drive_link	*slave = &hwif->acpidata->slave;
 
-	if (ide_noacpi)
-		return;
-
-	DEBPRINT("ENTER:\n");
-
-	if (!hwif->acpidata) {
-		DEBPRINT("no ACPI data for %s\n", hwif->name);
-		return;
-	}
-
 	/* Give the GTM buffer + drive Identify data to the channel via the
 	 * _STM method: */
 	/* setup input parameters buffer for _STM */
@@ -527,16 +509,11 @@ void ide_acpi_set_state(ide_hwif_t *hwif, int on)
 	ide_drive_t *drive;
 	int i;
 
-	if (ide_noacpi || ide_noacpi_psx)
+	if (ide_noacpi_psx)
 		return;
 
 	DEBPRINT("ENTER:\n");
 
-	if (!hwif->acpidata) {
-		DEBPRINT("no ACPI data for %s\n", hwif->name);
-		return;
-	}
-
 	/* channel first and then drives for power on and verse versa for power off */
 	if (on)
 		acpi_bus_set_power(hwif->acpidata->obj_handle, ACPI_STATE_D0);
@@ -616,7 +593,7 @@ void ide_acpi_port_init_devices(ide_hwif_t *hwif)
 				 drive->name, err);
 	}
 
-	if (!ide_acpionboot) {
+	if (ide_noacpi || ide_acpionboot == 0) {
 		DEBPRINT("ACPI methods disabled on boot\n");
 		return;
 	}
diff --git a/drivers/ide/ide-pm.c b/drivers/ide/ide-pm.c
index c14ca144cffe..ad7be2669dcb 100644
--- a/drivers/ide/ide-pm.c
+++ b/drivers/ide/ide-pm.c
@@ -10,9 +10,11 @@ int generic_ide_suspend(struct device *dev, pm_message_t mesg)
 	struct request_pm_state rqpm;
 	int ret;
 
-	/* call ACPI _GTM only once */
-	if ((drive->dn & 1) == 0 || pair == NULL)
-		ide_acpi_get_timing(hwif);
+	if (ide_port_acpi(hwif)) {
+		/* call ACPI _GTM only once */
+		if ((drive->dn & 1) == 0 || pair == NULL)
+			ide_acpi_get_timing(hwif);
+	}
 
 	memset(&rqpm, 0, sizeof(rqpm));
 	rq = blk_get_request(drive->queue, READ, __GFP_WAIT);
@@ -26,9 +28,11 @@ int generic_ide_suspend(struct device *dev, pm_message_t mesg)
 	ret = blk_execute_rq(drive->queue, NULL, rq, 0);
 	blk_put_request(rq);
 
-	/* call ACPI _PS3 only after both devices are suspended */
-	if (ret == 0 && ((drive->dn & 1) || pair == NULL))
-		ide_acpi_set_state(hwif, 0);
+	if (ret == 0 && ide_port_acpi(hwif)) {
+		/* call ACPI _PS3 only after both devices are suspended */
+		if ((drive->dn & 1) || pair == NULL)
+			ide_acpi_set_state(hwif, 0);
+	}
 
 	return ret;
 }
@@ -42,13 +46,15 @@ int generic_ide_resume(struct device *dev)
 	struct request_pm_state rqpm;
 	int err;
 
-	/* call ACPI _PS0 / _STM only once */
-	if ((drive->dn & 1) == 0 || pair == NULL) {
-		ide_acpi_set_state(hwif, 1);
-		ide_acpi_push_timing(hwif);
-	}
+	if (ide_port_acpi(hwif)) {
+		/* call ACPI _PS0 / _STM only once */
+		if ((drive->dn & 1) == 0 || pair == NULL) {
+			ide_acpi_set_state(hwif, 1);
+			ide_acpi_push_timing(hwif);
+		}
 
-	ide_acpi_exec_tfs(drive);
+		ide_acpi_exec_tfs(drive);
+	}
 
 	memset(&rqpm, 0, sizeof(rqpm));
 	rq = blk_get_request(drive->queue, READ, __GFP_WAIT);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index c6af7c44d46c..edc93a6d931d 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1419,6 +1419,7 @@ static inline void ide_dma_unmap_sg(ide_drive_t *drive,
 
 #ifdef CONFIG_BLK_DEV_IDEACPI
 int ide_acpi_init(void);
+bool ide_port_acpi(ide_hwif_t *hwif);
 extern int ide_acpi_exec_tfs(ide_drive_t *drive);
 extern void ide_acpi_get_timing(ide_hwif_t *hwif);
 extern void ide_acpi_push_timing(ide_hwif_t *hwif);
@@ -1427,6 +1428,7 @@ void ide_acpi_port_init_devices(ide_hwif_t *);
 extern void ide_acpi_set_state(ide_hwif_t *hwif, int on);
 #else
 static inline int ide_acpi_init(void) { return 0; }
+static inline bool ide_port_acpi(ide_hwif_t *hwif) { return 0; }
 static inline int ide_acpi_exec_tfs(ide_drive_t *drive) { return 0; }
 static inline void ide_acpi_get_timing(ide_hwif_t *hwif) { ; }
 static inline void ide_acpi_push_timing(ide_hwif_t *hwif) { ; }
-- 
cgit v1.2.3-59-g8ed1b


From e18ed145c7f556f1de8350c32739bf35b26df705 Mon Sep 17 00:00:00 2001
From: Christian Engelmayer <christian.engelmayer@frequentis.com>
Date: Mon, 29 Jun 2009 19:31:41 -0700
Subject: ide: memory overrun in ide_get_identity_ioctl() on big endian
 machines using ioctl HDIO_OBSOLETE_IDENTITY

This patch fixes a memory overrun in function ide_get_identity_ioctl() which
chooses the size of a memory buffer depending on the ioctl command that led
to the function call, however, passes that buffer to a function which needs the
buffer size to be always chosen unconditionally.

Due to conditional compilation the memory overrun can only happen on big endian
machines. The error can be triggered using ioctl HDIO_OBSOLETE_IDENTITY. Usage
of ioctl HDIO_GET_IDENTITY is safe.

Signed-off-by: Christian Engelmayer <christian.engelmayer@frequentis.com>
Acked-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/ide/ide-ioctls.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/ide/ide-ioctls.c b/drivers/ide/ide-ioctls.c
index 82f252c3ee6e..e246d3d3fbcc 100644
--- a/drivers/ide/ide-ioctls.c
+++ b/drivers/ide/ide-ioctls.c
@@ -64,7 +64,8 @@ static int ide_get_identity_ioctl(ide_drive_t *drive, unsigned int cmd,
 		goto out;
 	}
 
-	id = kmalloc(size, GFP_KERNEL);
+	/* ata_id_to_hd_driveid() relies on 'id' to be fully allocated. */
+	id = kmalloc(ATA_ID_WORDS * 2, GFP_KERNEL);
 	if (id == NULL) {
 		rc = -ENOMEM;
 		goto out;
-- 
cgit v1.2.3-59-g8ed1b


From 8e5b9dda99cc86bdbd822935fcc37c5808e271b3 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 28 Jun 2009 18:03:30 +0000
Subject: tcp: Stop non-TSO packets morphing into TSO

If a socket starts out on a non-TSO route, and then switches to
a TSO route, then the tail on the tx queue can morph into a TSO
packet, causing mischief because the rest of the stack does not
expect a partially linear TSO packet.

This patch fixes this by ensuring that skb->ip_summed is set to
CHECKSUM_PARTIAL before declaring a packet as TSO.

Reported-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_output.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 416fc4c2e7eb..5bdf08d312d9 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -725,7 +725,8 @@ static void tcp_queue_skb(struct sock *sk, struct sk_buff *skb)
 static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb,
 				 unsigned int mss_now)
 {
-	if (skb->len <= mss_now || !sk_can_gso(sk)) {
+	if (skb->len <= mss_now || !sk_can_gso(sk) ||
+	    skb->ip_summed == CHECKSUM_NONE) {
 		/* Avoid the costly divide in the normal
 		 * non-TSO case.
 		 */
-- 
cgit v1.2.3-59-g8ed1b


From 6828b92bd21acd65113dfe0541f19f5df0d9668f Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Sun, 28 Jun 2009 18:06:41 +0000
Subject: tcp: Do not tack on TSO data to non-TSO packet

If a socket starts out on a non-TSO route, and then switches to
a TSO route, then we will tack on data to the tail of the tx queue
even if it started out life as non-TSO.  This is suboptimal because
all of it will then be copied and checksummed unnecessarily.

This patch fixes this by ensuring that skb->ip_summed is set to
CHECKSUM_PARTIAL before appending extra data beyond the MSS.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 17b89c523f9d..7870a535dac6 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -903,13 +903,17 @@ int tcp_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
 		iov++;
 
 		while (seglen > 0) {
-			int copy;
+			int copy = 0;
+			int max = size_goal;
 
 			skb = tcp_write_queue_tail(sk);
+			if (tcp_send_head(sk)) {
+				if (skb->ip_summed == CHECKSUM_NONE)
+					max = mss_now;
+				copy = max - skb->len;
+			}
 
-			if (!tcp_send_head(sk) ||
-			    (copy = size_goal - skb->len) <= 0) {
-
+			if (copy <= 0) {
 new_segment:
 				/* Allocate new segment. If the interface is SG,
 				 * allocate skb fitting to single page.
@@ -930,6 +934,7 @@ new_segment:
 
 				skb_entail(sk, skb);
 				copy = size_goal;
+				max = size_goal;
 			}
 
 			/* Try to append data to the end of skb. */
@@ -1028,7 +1033,7 @@ new_segment:
 			if ((seglen -= copy) == 0 && iovlen == 0)
 				goto out;
 
-			if (skb->len < size_goal || (flags & MSG_OOB))
+			if (skb->len < max || (flags & MSG_OOB))
 				continue;
 
 			if (forced_push(tp)) {
-- 
cgit v1.2.3-59-g8ed1b


From 1802571b9865c0fc1d8d0fa39cf73275f3a75af3 Mon Sep 17 00:00:00 2001
From: Wei Yongjun <yjwei@cn.fujitsu.com>
Date: Sun, 28 Jun 2009 18:42:53 +0000
Subject: xfrm: use xfrm_addr_cmp() instead of compare addresses directly

Clean up to use xfrm_addr_cmp() instead of compare addresses directly.

Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/xfrm/xfrm_state.c | 57 ++++++++-------------------------------------------
 1 file changed, 8 insertions(+), 49 deletions(-)

diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 5f1f86565f16..f2f7c638083e 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -668,22 +668,10 @@ static struct xfrm_state *__xfrm_state_lookup(struct net *net, xfrm_address_t *d
 	hlist_for_each_entry(x, entry, net->xfrm.state_byspi+h, byspi) {
 		if (x->props.family != family ||
 		    x->id.spi       != spi ||
-		    x->id.proto     != proto)
+		    x->id.proto     != proto ||
+		    xfrm_addr_cmp(&x->id.daddr, daddr, family))
 			continue;
 
-		switch (family) {
-		case AF_INET:
-			if (x->id.daddr.a4 != daddr->a4)
-				continue;
-			break;
-		case AF_INET6:
-			if (!ipv6_addr_equal((struct in6_addr *)daddr,
-					     (struct in6_addr *)
-					     x->id.daddr.a6))
-				continue;
-			break;
-		}
-
 		xfrm_state_hold(x);
 		return x;
 	}
@@ -699,26 +687,11 @@ static struct xfrm_state *__xfrm_state_lookup_byaddr(struct net *net, xfrm_addre
 
 	hlist_for_each_entry(x, entry, net->xfrm.state_bysrc+h, bysrc) {
 		if (x->props.family != family ||
-		    x->id.proto     != proto)
+		    x->id.proto     != proto ||
+		    xfrm_addr_cmp(&x->id.daddr, daddr, family) ||
+		    xfrm_addr_cmp(&x->props.saddr, saddr, family))
 			continue;
 
-		switch (family) {
-		case AF_INET:
-			if (x->id.daddr.a4 != daddr->a4 ||
-			    x->props.saddr.a4 != saddr->a4)
-				continue;
-			break;
-		case AF_INET6:
-			if (!ipv6_addr_equal((struct in6_addr *)daddr,
-					     (struct in6_addr *)
-					     x->id.daddr.a6) ||
-			    !ipv6_addr_equal((struct in6_addr *)saddr,
-					     (struct in6_addr *)
-					     x->props.saddr.a6))
-				continue;
-			break;
-		}
-
 		xfrm_state_hold(x);
 		return x;
 	}
@@ -1001,25 +974,11 @@ static struct xfrm_state *__find_acq_core(struct net *net, unsigned short family
 		    x->props.family != family ||
 		    x->km.state     != XFRM_STATE_ACQ ||
 		    x->id.spi       != 0 ||
-		    x->id.proto	    != proto)
+		    x->id.proto	    != proto ||
+		    xfrm_addr_cmp(&x->id.daddr, daddr, family) ||
+		    xfrm_addr_cmp(&x->props.saddr, saddr, family))
 			continue;
 
-		switch (family) {
-		case AF_INET:
-			if (x->id.daddr.a4    != daddr->a4 ||
-			    x->props.saddr.a4 != saddr->a4)
-				continue;
-			break;
-		case AF_INET6:
-			if (!ipv6_addr_equal((struct in6_addr *)x->id.daddr.a6,
-					     (struct in6_addr *)daddr) ||
-			    !ipv6_addr_equal((struct in6_addr *)
-					     x->props.saddr.a6,
-					     (struct in6_addr *)saddr))
-				continue;
-			break;
-		}
-
 		xfrm_state_hold(x);
 		return x;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From d51e9b0d94336db56a13fdc65bb30751e3ea33b7 Mon Sep 17 00:00:00 2001
From: Graf Yang <graf.yang@analog.com>
Date: Mon, 29 Jun 2009 09:34:20 +0000
Subject: net/irda: convert bfin_sir to net_device_ops

Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/irda/bfin_sir.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/net/irda/bfin_sir.c b/drivers/net/irda/bfin_sir.c
index f3eed6a8fba5..911c082cee5a 100644
--- a/drivers/net/irda/bfin_sir.c
+++ b/drivers/net/irda/bfin_sir.c
@@ -677,6 +677,14 @@ static int bfin_sir_init_iobuf(iobuff_t *io, int size)
 	return 0;
 }
 
+static const struct net_device_ops bfin_sir_ndo = {
+	.ndo_open		= bfin_sir_open,
+	.ndo_stop		= bfin_sir_stop,
+	.ndo_start_xmit		= bfin_sir_hard_xmit,
+	.ndo_do_ioctl		= bfin_sir_ioctl,
+	.ndo_get_stats		= bfin_sir_stats,
+};
+
 static int __devinit bfin_sir_probe(struct platform_device *pdev)
 {
 	struct net_device *dev;
@@ -718,12 +726,8 @@ static int __devinit bfin_sir_probe(struct platform_device *pdev)
 	if (err)
 		goto err_mem_3;
 
-	dev->hard_start_xmit = bfin_sir_hard_xmit;
-	dev->open            = bfin_sir_open;
-	dev->stop            = bfin_sir_stop;
-	dev->do_ioctl        = bfin_sir_ioctl;
-	dev->get_stats       = bfin_sir_stats;
-	dev->irq             = sir_port->irq;
+	dev->netdev_ops = &bfin_sir_ndo;
+	dev->irq = sir_port->irq;
 
 	irda_init_max_qos_capabilies(&self->qos);
 
-- 
cgit v1.2.3-59-g8ed1b


From ff0ac74afb5b9916641723a78796d4ee7937c2ea Mon Sep 17 00:00:00 2001
From: Wei Yongjun <yjwei@cn.fujitsu.com>
Date: Sun, 28 Jun 2009 22:49:37 +0000
Subject: sctp: xmit sctp packet always return no route error

Commit 'net: skb->dst accessors'(adf30907d63893e4208dfe3f5c88ae12bc2f25d5)
broken the sctp protocol stack, the sctp packet can never be sent out after
Eric Dumazet's patch, which have typo in the sctp code.

Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Vlad Yasevich <vladisalv.yasevich@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sctp/output.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/sctp/output.c b/net/sctp/output.c
index b76411444515..b94c21190566 100644
--- a/net/sctp/output.c
+++ b/net/sctp/output.c
@@ -407,7 +407,7 @@ int sctp_packet_transmit(struct sctp_packet *packet)
 	}
 	dst = dst_clone(tp->dst);
 	skb_dst_set(nskb, dst);
-	if (dst)
+	if (!dst)
 		goto no_route;
 
 	/* Build the SCTP header.  */
-- 
cgit v1.2.3-59-g8ed1b


From e1605495c716ef4eebdb7606bcd1b593f28e2837 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Mon, 29 Jun 2009 11:17:38 +0100
Subject: intel-iommu: Introduce domain_sg_mapping() to speed up intel_map_sg()

Instead of calling domain_pfn_mapping() repeatedly with single or
small numbers of pages, just pass the sglist in. It can optimise the
number of cache flushes like domain_pfn_mapping() does, and gives a huge
speedup for large scatterlists.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 83 +++++++++++++++++++++++++++++++++++------------
 1 file changed, 62 insertions(+), 21 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 11a23201445a..28bd5f2d78fc 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1635,6 +1635,56 @@ static int domain_context_mapped(struct pci_dev *pdev)
 					     tmp->devfn);
 }
 
+static int domain_sg_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
+			     struct scatterlist *sg, unsigned long nr_pages,
+			     int prot)
+{
+	struct dma_pte *first_pte = NULL, *pte = NULL;
+	uint64_t pteval;
+	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
+	unsigned long sg_res = 0;
+
+	BUG_ON(addr_width < BITS_PER_LONG && (iov_pfn + nr_pages - 1) >> addr_width);
+
+	if ((prot & (DMA_PTE_READ|DMA_PTE_WRITE)) == 0)
+		return -EINVAL;
+
+	prot &= DMA_PTE_READ | DMA_PTE_WRITE | DMA_PTE_SNP;
+
+	while (nr_pages--) {
+		if (!sg_res) {
+			sg_res = (sg->offset + sg->length + VTD_PAGE_SIZE - 1) >> VTD_PAGE_SHIFT;
+			sg->dma_address = ((dma_addr_t)iov_pfn << VTD_PAGE_SHIFT) + sg->offset;
+			sg->dma_length = sg->length;
+			pteval = page_to_phys(sg_page(sg)) | prot;
+		}
+		if (!pte) {
+			first_pte = pte = pfn_to_dma_pte(domain, iov_pfn);
+			if (!pte)
+				return -ENOMEM;
+		}
+		/* We don't need lock here, nobody else
+		 * touches the iova range
+		 */
+		BUG_ON(dma_pte_addr(pte));
+		pte->val = pteval;
+		pte++;
+		if (!nr_pages ||
+		    (unsigned long)pte >> VTD_PAGE_SHIFT !=
+		    (unsigned long)first_pte >> VTD_PAGE_SHIFT) {
+			domain_flush_cache(domain, first_pte,
+					   (void *)pte - (void *)first_pte);
+			pte = NULL;
+		}
+		iov_pfn++;
+		pteval += VTD_PAGE_SIZE;
+		sg_res--;
+		if (!sg_res)
+			sg = sg_next(sg);
+	}
+	return 0;
+}
+
 static int domain_pfn_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
 			      unsigned long phys_pfn, unsigned long nr_pages,
 			      int prot)
@@ -2758,27 +2808,18 @@ static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int ne
 		prot |= DMA_PTE_WRITE;
 
 	start_vpfn = mm_to_dma_pfn(iova->pfn_lo);
-	offset_pfn = 0;
-	for_each_sg(sglist, sg, nelems, i) {
-		int nr_pages = aligned_nrpages(sg->offset, sg->length);
-		ret = domain_pfn_mapping(domain, start_vpfn + offset_pfn,
-					 page_to_dma_pfn(sg_page(sg)),
-					 nr_pages, prot);
-		if (ret) {
-			/*  clear the page */
-			dma_pte_clear_range(domain, start_vpfn,
-					    start_vpfn + offset_pfn);
-			/* free page tables */
-			dma_pte_free_pagetable(domain, start_vpfn,
-					       start_vpfn + offset_pfn);
-			/* free iova */
-			__free_iova(&domain->iovad, iova);
-			return 0;
-		}
-		sg->dma_address = ((dma_addr_t)(start_vpfn + offset_pfn)
-				   << VTD_PAGE_SHIFT) + sg->offset;
-		sg->dma_length = sg->length;
-		offset_pfn += nr_pages;
+
+	ret = domain_sg_mapping(domain, start_vpfn, sglist, mm_to_dma_pfn(size), prot);
+	if (unlikely(ret)) {
+		/*  clear the page */
+		dma_pte_clear_range(domain, start_vpfn,
+				    start_vpfn + size - 1);
+		/* free page tables */
+		dma_pte_free_pagetable(domain, start_vpfn,
+				       start_vpfn + size - 1);
+		/* free iova */
+		__free_iova(&domain->iovad, iova);
+		return 0;
 	}
 
 	/* it's a non-present to present mapping. Only flush if caching mode */
-- 
cgit v1.2.3-59-g8ed1b


From 9051aa0268dc1c3e42cd79a802b0af1f2bfcadae Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Mon, 29 Jun 2009 12:30:54 +0100
Subject: intel-iommu: Combine domain_pfn_mapping() and domain_sg_mapping()

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 62 +++++++++++++++++------------------------------
 1 file changed, 22 insertions(+), 40 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 28bd5f2d78fc..14308533b1cb 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1635,14 +1635,14 @@ static int domain_context_mapped(struct pci_dev *pdev)
 					     tmp->devfn);
 }
 
-static int domain_sg_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
-			     struct scatterlist *sg, unsigned long nr_pages,
-			     int prot)
+static int __domain_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
+			    struct scatterlist *sg, unsigned long phys_pfn,
+			    unsigned long nr_pages, int prot)
 {
 	struct dma_pte *first_pte = NULL, *pte = NULL;
-	uint64_t pteval;
+	phys_addr_t uninitialized_var(pteval);
 	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
-	unsigned long sg_res = 0;
+	unsigned long sg_res;
 
 	BUG_ON(addr_width < BITS_PER_LONG && (iov_pfn + nr_pages - 1) >> addr_width);
 
@@ -1651,6 +1651,13 @@ static int domain_sg_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
 
 	prot &= DMA_PTE_READ | DMA_PTE_WRITE | DMA_PTE_SNP;
 
+	if (sg)
+		sg_res = 0;
+	else {
+		sg_res = nr_pages + 1;
+		pteval = ((phys_addr_t)phys_pfn << VTD_PAGE_SHIFT) | prot;
+	}
+
 	while (nr_pages--) {
 		if (!sg_res) {
 			sg_res = (sg->offset + sg->length + VTD_PAGE_SIZE - 1) >> VTD_PAGE_SHIFT;
@@ -1685,43 +1692,18 @@ static int domain_sg_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
 	return 0;
 }
 
-static int domain_pfn_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
-			      unsigned long phys_pfn, unsigned long nr_pages,
-			      int prot)
+static inline int domain_sg_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
+				    struct scatterlist *sg, unsigned long nr_pages,
+				    int prot)
 {
-	struct dma_pte *first_pte = NULL, *pte = NULL;
-	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
-
-	BUG_ON(addr_width < BITS_PER_LONG && (iov_pfn + nr_pages - 1) >> addr_width);
-
-	if ((prot & (DMA_PTE_READ|DMA_PTE_WRITE)) == 0)
-		return -EINVAL;
-
-	prot &= DMA_PTE_READ | DMA_PTE_WRITE | DMA_PTE_SNP;
+	return __domain_mapping(domain, iov_pfn, sg, 0, nr_pages, prot);
+}
 
-	while (nr_pages--) {
-		if (!pte) {
-			first_pte = pte = pfn_to_dma_pte(domain, iov_pfn);
-			if (!pte)
-				return -ENOMEM;
-		}
-		/* We don't need lock here, nobody else
-		 * touches the iova range
-		 */
-		BUG_ON(dma_pte_addr(pte));
-		pte->val = (phys_pfn << VTD_PAGE_SHIFT) | prot;
-		pte++;
-		if (!nr_pages ||
-		    (unsigned long)pte >> VTD_PAGE_SHIFT !=
-		    (unsigned long)first_pte >> VTD_PAGE_SHIFT) {
-			domain_flush_cache(domain, first_pte,
-					   (void *)pte - (void *)first_pte);
-			pte = NULL;
-		}
-		iov_pfn++;
-		phys_pfn++;
-	}
-	return 0;
+static inline int domain_pfn_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
+				     unsigned long phys_pfn, unsigned long nr_pages,
+				     int prot)
+{
+	return __domain_mapping(domain, iov_pfn, NULL, phys_pfn, nr_pages, prot);
 }
 
 static void iommu_detach_dev(struct intel_iommu *iommu, u8 bus, u8 devfn)
-- 
cgit v1.2.3-59-g8ed1b


From 1bf20f0dc5629032ddd07617139d9fbca66c1642 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Mon, 29 Jun 2009 22:06:43 +0100
Subject: intel-iommu: dump mappings but don't die on pte already set

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 14308533b1cb..40ce5a03f18f 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1673,7 +1673,16 @@ static int __domain_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
 		/* We don't need lock here, nobody else
 		 * touches the iova range
 		 */
-		BUG_ON(dma_pte_addr(pte));
+		if (unlikely(dma_pte_addr(pte))) {
+			static int dumps = 5;
+			printk(KERN_CRIT "ERROR: DMA PTE for vPFN 0x%lx already set (to %llx)\n",
+			       iov_pfn, pte->val);
+			if (dumps) {
+				dumps--;
+				debug_dma_dump_mappings(NULL);
+			}
+			WARN_ON(1);
+		}
 		pte->val = pteval;
 		pte++;
 		if (!nr_pages ||
-- 
cgit v1.2.3-59-g8ed1b


From 3d7b0e4154b4963d6bd39991ec8eaa09caeb3994 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Tue, 30 Jun 2009 03:38:09 +0100
Subject: intel-iommu: Don't free too much in dma_pte_free_pagetable()

The loop condition was wrong -- we should free a PMD only if its
_entire_ range is within the range we're intending to clear. The
early-termination condition was right, but not the loop.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 40ce5a03f18f..35bdd2a06caa 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -815,7 +815,7 @@ static void dma_pte_free_pagetable(struct dmar_domain *domain,
 		if (tmp + level_size(level) - 1 > last_pfn)
 			return;
 
-		while (tmp <= last_pfn) {
+		while (tmp + level_size(level) - 1 <= last_pfn) {
 			pte = dma_pfn_level_pte(domain, tmp, level);
 			if (pte) {
 				free_pgtable_page(
-- 
cgit v1.2.3-59-g8ed1b


From f3a0a52fff4dbfdea2dccc908d00c038481d888e Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Tue, 30 Jun 2009 03:40:07 +0100
Subject: intel-iommu: Performance improvement for dma_pte_free_pagetable()

As with other functions, batch the CPU data cache flushes and don't keep
recalculating PTE addresses.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 27 +++++++++++++++++----------
 1 file changed, 17 insertions(+), 10 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 35bdd2a06caa..ec7e032d5ab5 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -797,7 +797,7 @@ static void dma_pte_free_pagetable(struct dmar_domain *domain,
 				   unsigned long last_pfn)
 {
 	int addr_width = agaw_to_width(domain->agaw) - VTD_PAGE_SHIFT;
-	struct dma_pte *pte;
+	struct dma_pte *first_pte, *pte;
 	int total = agaw_to_level(domain->agaw);
 	int level;
 	unsigned long tmp;
@@ -805,25 +805,32 @@ static void dma_pte_free_pagetable(struct dmar_domain *domain,
 	BUG_ON(addr_width < BITS_PER_LONG && start_pfn >> addr_width);
 	BUG_ON(addr_width < BITS_PER_LONG && last_pfn >> addr_width);
 
-	/* we don't need lock here, nobody else touches the iova range */
+	/* We don't need lock here; nobody else touches the iova range */
 	level = 2;
 	while (level <= total) {
 		tmp = align_to_level(start_pfn, level);
 
-		/* Only clear this pte/pmd if we're asked to clear its
-		   _whole_ range */
+		/* If we can't even clear one PTE at this level, we're done */
 		if (tmp + level_size(level) - 1 > last_pfn)
 			return;
 
 		while (tmp + level_size(level) - 1 <= last_pfn) {
-			pte = dma_pfn_level_pte(domain, tmp, level);
-			if (pte) {
-				free_pgtable_page(
-					phys_to_virt(dma_pte_addr(pte)));
+			first_pte = pte = dma_pfn_level_pte(domain, tmp, level);
+			if (!pte) {
+				tmp = align_to_level(tmp + 1, level + 1);
+				continue;
+			}
+			while (tmp + level_size(level) - 1 <= last_pfn &&
+			       (unsigned long)pte >> VTD_PAGE_SHIFT ==
+			       (unsigned long)first_pte >> VTD_PAGE_SHIFT) {
+				free_pgtable_page(phys_to_virt(dma_pte_addr(pte)));
 				dma_clear_pte(pte);
-				domain_flush_cache(domain, pte, sizeof(*pte));
+				pte++;
+				tmp += level_size(level);
 			}
-			tmp += level_size(level);
+			domain_flush_cache(domain, first_pte,
+					   (void *)pte - (void *)first_pte);
+			
 		}
 		level++;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From 57e7986ed142417498155ebcd5eaf617ac37136d Mon Sep 17 00:00:00 2001
From: Paul Mackerras <paulus@samba.org>
Date: Tue, 30 Jun 2009 16:07:19 +1000
Subject: perf_counter: Provide a way to enable counters on exec

This provides a way to mark a counter to be enabled on the next
exec. This is useful for measuring the total activity of a
program without including overhead from the process that
launches it.

This also changes the perf stat command to use this new
facility.

Signed-off-by: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <19017.43927.838745.689203@cargo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/perf_counter.h |  3 ++-
 kernel/perf_counter.c        | 50 ++++++++++++++++++++++++++++++++++++++++++++
 tools/perf/builtin-stat.c    |  6 +++---
 3 files changed, 55 insertions(+), 4 deletions(-)

diff --git a/include/linux/perf_counter.h b/include/linux/perf_counter.h
index 3078e23c91eb..5e970c7d3fd5 100644
--- a/include/linux/perf_counter.h
+++ b/include/linux/perf_counter.h
@@ -179,8 +179,9 @@ struct perf_counter_attr {
 				comm	       :  1, /* include comm data     */
 				freq           :  1, /* use freq, not period  */
 				inherit_stat   :  1, /* per task counts       */
+				enable_on_exec :  1, /* next exec enables     */
 
-				__reserved_1   : 52;
+				__reserved_1   : 51;
 
 	__u32			wakeup_events;	/* wakeup every n events */
 	__u32			__reserved_2;
diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index 66ab1e9d1294..d55a50da2347 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -1428,6 +1428,53 @@ void perf_counter_task_tick(struct task_struct *curr, int cpu)
 		perf_counter_task_sched_in(curr, cpu);
 }
 
+/*
+ * Enable all of a task's counters that have been marked enable-on-exec.
+ * This expects task == current.
+ */
+static void perf_counter_enable_on_exec(struct task_struct *task)
+{
+	struct perf_counter_context *ctx;
+	struct perf_counter *counter;
+	unsigned long flags;
+	int enabled = 0;
+
+	local_irq_save(flags);
+	ctx = task->perf_counter_ctxp;
+	if (!ctx || !ctx->nr_counters)
+		goto out;
+
+	__perf_counter_task_sched_out(ctx);
+
+	spin_lock(&ctx->lock);
+
+	list_for_each_entry(counter, &ctx->counter_list, list_entry) {
+		if (!counter->attr.enable_on_exec)
+			continue;
+		counter->attr.enable_on_exec = 0;
+		if (counter->state >= PERF_COUNTER_STATE_INACTIVE)
+			continue;
+		counter->state = PERF_COUNTER_STATE_INACTIVE;
+		counter->tstamp_enabled =
+			ctx->time - counter->total_time_enabled;
+		enabled = 1;
+	}
+
+	/*
+	 * Unclone this context if we enabled any counter.
+	 */
+	if (enabled && ctx->parent_ctx) {
+		put_ctx(ctx->parent_ctx);
+		ctx->parent_ctx = NULL;
+	}
+
+	spin_unlock(&ctx->lock);
+
+	perf_counter_task_sched_in(task, smp_processor_id());
+ out:
+	local_irq_restore(flags);
+}
+
 /*
  * Cross CPU call to read the hardware counter
  */
@@ -2949,6 +2996,9 @@ void perf_counter_comm(struct task_struct *task)
 {
 	struct perf_comm_event comm_event;
 
+	if (task->perf_counter_ctxp)
+		perf_counter_enable_on_exec(task);
+
 	if (!atomic_read(&nr_comm_counters))
 		return;
 
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 201ef2367dcb..2e03524a1de0 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -116,8 +116,9 @@ static void create_perf_stat_counter(int counter, int pid)
 					fd[cpu][counter], strerror(errno));
 		}
 	} else {
-		attr->inherit	= inherit;
-		attr->disabled	= 1;
+		attr->inherit	     = inherit;
+		attr->disabled	     = 1;
+		attr->enable_on_exec = 1;
 
 		fd[0][counter] = sys_perf_counter_open(attr, pid, -1, -1, 0);
 		if (fd[0][counter] < 0 && verbose)
@@ -262,7 +263,6 @@ static int run_perf_stat(int argc, const char **argv)
 	 * Enable counters and exec the command:
 	 */
 	t0 = rdclock();
-	prctl(PR_TASK_PERF_COUNTERS_ENABLE);
 
 	close(go_pipe[1]);
 	wait(&status);
-- 
cgit v1.2.3-59-g8ed1b


From 874d2f61d31e596c36af7732dc1b3aa2dc233824 Mon Sep 17 00:00:00 2001
From: Milan Broz <mbroz@redhat.com>
Date: Tue, 30 Jun 2009 15:18:14 +0100
Subject: dm exception store: really fix type lookup

Fix exception store name handling.

We need to reference exception store by zero terminated string.

Fixes regression introduced in commit f6bd4eb73cdf2a5bf954e497972842f39cabb7e3

Cc: Yi Yang <yi.y.yang@intel.com>
Cc: Jonathan Brassow <jbrassow@redhat.com>
Cc: stable@kernel.org
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Milan Broz <mbroz@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-exception-store.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/md/dm-exception-store.c b/drivers/md/dm-exception-store.c
index c3ae51584b12..3710ff88fc10 100644
--- a/drivers/md/dm-exception-store.c
+++ b/drivers/md/dm-exception-store.c
@@ -195,7 +195,7 @@ int dm_exception_store_create(struct dm_target *ti, int argc, char **argv,
 			      struct dm_exception_store **store)
 {
 	int r = 0;
-	struct dm_exception_store_type *type;
+	struct dm_exception_store_type *type = NULL;
 	struct dm_exception_store *tmp_store;
 	char persistent;
 
@@ -211,12 +211,15 @@ int dm_exception_store_create(struct dm_target *ti, int argc, char **argv,
 	}
 
 	persistent = toupper(*argv[1]);
-	if (persistent != 'P' && persistent != 'N') {
+	if (persistent == 'P')
+		type = get_type("P");
+	else if (persistent == 'N')
+		type = get_type("N");
+	else {
 		ti->error = "Persistent flag is not P or N";
 		return -EINVAL;
 	}
 
-	type = get_type(&persistent);
 	if (!type) {
 		ti->error = "Exception store type not recognised";
 		r = -EINVAL;
-- 
cgit v1.2.3-59-g8ed1b


From ea9df47cc92573b159ef3b4fda516c32cba9c4fd Mon Sep 17 00:00:00 2001
From: Mike Snitzer <snitzer@redhat.com>
Date: Tue, 30 Jun 2009 15:18:17 +0100
Subject: dm table: fix blk_stack_limits arg to use bytes not sectors

The offset passed to blk_stack_limits() must be in bytes not sectors.
Fixes false warnings like the following:
device-mapper: table: 254:1: target device sda6 is misaligned

Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Reported-by: Frans Pop <elendil@planet.nl>
Tested-by: Frans Pop <elendil@planet.nl>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
---
 drivers/md/dm-table.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 4899ebe767c8..2cba557d9e61 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -495,7 +495,7 @@ int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
 		return 0;
 	}
 
-	if (blk_stack_limits(limits, &q->limits, start) < 0)
+	if (blk_stack_limits(limits, &q->limits, start << 9) < 0)
 		DMWARN("%s: target device %s is misaligned",
 		       dm_device_name(ti->table->md), bdevname(bdev, b));
 
-- 
cgit v1.2.3-59-g8ed1b


From f5812a7a336fb952d819e4427b9a2dce02368e82 Mon Sep 17 00:00:00 2001
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Date: Tue, 30 Jun 2009 11:43:17 -0300
Subject: perf_counter tools: Adjust only prelinked symbol's addresses

I.e. we can't handle these two kinds of files in the same way:

1) prelinked system library:

[acme@doppio pahole]$ readelf -s /usr/lib64/libdw-0.141.so | egrep 'FUNC.+GLOBAL.+dwfl_report_elf'
   278: 00000030450105a0   261 FUNC    GLOBAL DEFAULT   12 dwfl_report_elf@@ELFUTILS_0.122

2) not prelinked library with debug information from a -debuginfo package:

[acme@doppio pahole]$ readelf -s /usr/lib/debug/usr/lib64/libdw-0.141.so.debug | egrep 'FUNC.+GLOBAL.+dwfl_report_elf'
   629: 00000000000105a0   261 FUNC    GLOBAL DEFAULT   12 dwfl_report_elf
[acme@doppio pahole]$

Now the numbers I got for a pahole perf run are in line with
the numbers I get from oprofile.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20090630144317.GB12663@ghostprotocols.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/util/symbol.c | 16 +++++++++++-----
 tools/perf/util/symbol.h |  3 ++-
 2 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index 9c659ef6aec2..78c2efde01b7 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -520,7 +520,9 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
 	nr_syms = shdr.sh_size / shdr.sh_entsize;
 
 	memset(&sym, 0, sizeof(sym));
-
+	self->prelinked = elf_section_by_name(elf, &ehdr, &shdr,
+					      ".gnu.prelink_undo",
+					      NULL) != NULL;
 	elf_symtab__for_each_symbol(syms, nr_syms, index, sym) {
 		struct symbol *f;
 		u64 obj_start;
@@ -535,11 +537,13 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
 		gelf_getshdr(sec, &shdr);
 		obj_start = sym.st_value;
 
-		if (verbose >= 2)
-			printf("adjusting symbol: st_value: %Lx sh_addr: %Lx sh_offset: %Lx\n",
-				(u64)sym.st_value, (u64)shdr.sh_addr, (u64)shdr.sh_offset);
+		if (self->prelinked) {
+			if (verbose >= 2)
+				printf("adjusting symbol: st_value: %Lx sh_addr: %Lx sh_offset: %Lx\n",
+					(u64)sym.st_value, (u64)shdr.sh_addr, (u64)shdr.sh_offset);
 
-		sym.st_value -= shdr.sh_addr - shdr.sh_offset;
+			sym.st_value -= shdr.sh_addr - shdr.sh_offset;
+		}
 
 		f = symbol__new(sym.st_value, sym.st_size,
 				elf_sym__name(&sym, symstrs),
@@ -573,6 +577,8 @@ int dso__load(struct dso *self, symbol_filter_t filter, int verbose)
 	if (!name)
 		return -1;
 
+	self->prelinked = 0;
+
 	if (strncmp(self->name, "/tmp/perf-", 10) == 0)
 		return dso__load_perf_map(self, filter, verbose);
 
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index 940b432db16e..2c48ace8203b 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -20,8 +20,9 @@ struct symbol {
 struct dso {
 	struct list_head node;
 	struct rb_root	 syms;
-	unsigned int	 sym_priv_size;
 	struct symbol    *(*find_symbol)(struct dso *, u64 ip);
+	unsigned int	 sym_priv_size;
+	unsigned char	 prelinked;
 	char		 name[0];
 };
 
-- 
cgit v1.2.3-59-g8ed1b


From b4c458b3a23d76936e76678f2074b1528f129f7a Mon Sep 17 00:00:00 2001
From: Csaba Henk <csaba@gluster.com>
Date: Mon, 29 Jun 2009 03:26:53 +0200
Subject: fuse: fix return value of fuse_dev_write()

On 64 bit systems -- where sizeof(ssize_t) > sizeof(int) -- the following test
exposes a bug due to a non-careful return of an int or unsigned value:

implement a FUSE filesystem which sends an unsolicited notification to
the kernel with invalid opcode. The respective write to /dev/fuse
will return (1 << 32) - EINVAL with errno == 0 instead of -1 with
errno == EINVAL.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
CC: stable@kernel.org
---
 fs/fuse/dev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index 8fed2ed12f38..8a11a8c67c42 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -910,7 +910,7 @@ static ssize_t fuse_dev_write(struct kiocb *iocb, const struct iovec *iov,
 			       unsigned long nr_segs, loff_t pos)
 {
 	int err;
-	unsigned nbytes = iov_length(iov, nr_segs);
+	size_t nbytes = iov_length(iov, nr_segs);
 	struct fuse_req *req;
 	struct fuse_out_header oh;
 	struct fuse_copy_state cs;
-- 
cgit v1.2.3-59-g8ed1b


From 201fa69a2849536ef2912e8e971ec0b01c04eff4 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Tue, 30 Jun 2009 20:06:24 +0200
Subject: fuse: fix bad return value in fuse_file_poll()

Fix fuse_file_poll() which returned a -errno value instead of a poll
mask.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
CC: stable@kernel.org
---
 fs/fuse/file.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index fce6ce694fde..cbc464043b6f 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -1922,7 +1922,7 @@ unsigned fuse_file_poll(struct file *file, poll_table *wait)
 
 	req = fuse_get_req(fc);
 	if (IS_ERR(req))
-		return PTR_ERR(req);
+		return POLLERR;
 
 	req->in.h.opcode = FUSE_POLL;
 	req->in.h.nodeid = ff->nodeid;
-- 
cgit v1.2.3-59-g8ed1b


From e0a43ddcc08c34dbd666d93600fd23914505f4aa Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Tue, 30 Jun 2009 20:12:23 +0200
Subject: fuse: allow umask processing in userspace
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch lets filesystems handle masking the file mode on creation.
This is needed if filesystem is using ACLs.

 - The CREATE, MKDIR and MKNOD requests are extended with a "umask"
   parameter.

 - A new FUSE_DONT_MASK flag is added to the INIT request/reply.  With
   this the filesystem may request that the create mode is not masked.

CC: Jean-Pierre André <jean-pierre.andre@wanadoo.fr>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
---
 fs/fuse/dir.c        | 20 +++++++++++++++++---
 fs/fuse/fuse_i.h     |  3 +++
 fs/fuse/inode.c      |  9 ++++++++-
 include/linux/fuse.h | 20 ++++++++++++++++++--
 4 files changed, 46 insertions(+), 6 deletions(-)

diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index b3089a083d30..6b700734e519 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -375,7 +375,7 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry, int mode,
 	struct fuse_conn *fc = get_fuse_conn(dir);
 	struct fuse_req *req;
 	struct fuse_req *forget_req;
-	struct fuse_open_in inarg;
+	struct fuse_create_in inarg;
 	struct fuse_open_out outopen;
 	struct fuse_entry_out outentry;
 	struct fuse_file *ff;
@@ -399,15 +399,20 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry, int mode,
 	if (!ff)
 		goto out_put_request;
 
+	if (!fc->dont_mask)
+		mode &= ~current_umask();
+
 	flags &= ~O_NOCTTY;
 	memset(&inarg, 0, sizeof(inarg));
 	memset(&outentry, 0, sizeof(outentry));
 	inarg.flags = flags;
 	inarg.mode = mode;
+	inarg.umask = current_umask();
 	req->in.h.opcode = FUSE_CREATE;
 	req->in.h.nodeid = get_node_id(dir);
 	req->in.numargs = 2;
-	req->in.args[0].size = sizeof(inarg);
+	req->in.args[0].size = fc->minor < 12 ? sizeof(struct fuse_open_in) :
+						sizeof(inarg);
 	req->in.args[0].value = &inarg;
 	req->in.args[1].size = entry->d_name.len + 1;
 	req->in.args[1].value = entry->d_name.name;
@@ -546,12 +551,17 @@ static int fuse_mknod(struct inode *dir, struct dentry *entry, int mode,
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
+	if (!fc->dont_mask)
+		mode &= ~current_umask();
+
 	memset(&inarg, 0, sizeof(inarg));
 	inarg.mode = mode;
 	inarg.rdev = new_encode_dev(rdev);
+	inarg.umask = current_umask();
 	req->in.h.opcode = FUSE_MKNOD;
 	req->in.numargs = 2;
-	req->in.args[0].size = sizeof(inarg);
+	req->in.args[0].size = fc->minor < 12 ? FUSE_COMPAT_MKNOD_IN_SIZE :
+						sizeof(inarg);
 	req->in.args[0].value = &inarg;
 	req->in.args[1].size = entry->d_name.len + 1;
 	req->in.args[1].value = entry->d_name.name;
@@ -578,8 +588,12 @@ static int fuse_mkdir(struct inode *dir, struct dentry *entry, int mode)
 	if (IS_ERR(req))
 		return PTR_ERR(req);
 
+	if (!fc->dont_mask)
+		mode &= ~current_umask();
+
 	memset(&inarg, 0, sizeof(inarg));
 	inarg.mode = mode;
+	inarg.umask = current_umask();
 	req->in.h.opcode = FUSE_MKDIR;
 	req->in.numargs = 2;
 	req->in.args[0].size = sizeof(inarg);
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index aaf2f9ff970e..ede4f77b2d6c 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -446,6 +446,9 @@ struct fuse_conn {
 	/** Do multi-page cached writes */
 	unsigned big_writes:1;
 
+	/** Don't apply umask to creation modes */
+	unsigned dont_mask:1;
+
 	/** The number of requests waiting for completion */
 	atomic_t num_waiting;
 
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index d8673ccf90b7..6cc501bd0187 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -725,6 +725,8 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_req *req)
 			}
 			if (arg->flags & FUSE_BIG_WRITES)
 				fc->big_writes = 1;
+			if (arg->flags & FUSE_DONT_MASK)
+				fc->dont_mask = 1;
 		} else {
 			ra_pages = fc->max_read / PAGE_CACHE_SIZE;
 			fc->no_lock = 1;
@@ -748,7 +750,7 @@ static void fuse_send_init(struct fuse_conn *fc, struct fuse_req *req)
 	arg->minor = FUSE_KERNEL_MINOR_VERSION;
 	arg->max_readahead = fc->bdi.ra_pages * PAGE_CACHE_SIZE;
 	arg->flags |= FUSE_ASYNC_READ | FUSE_POSIX_LOCKS | FUSE_ATOMIC_O_TRUNC |
-		FUSE_EXPORT_SUPPORT | FUSE_BIG_WRITES;
+		FUSE_EXPORT_SUPPORT | FUSE_BIG_WRITES | FUSE_DONT_MASK;
 	req->in.h.opcode = FUSE_INIT;
 	req->in.numargs = 1;
 	req->in.args[0].size = sizeof(*arg);
@@ -864,6 +866,11 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent)
 	if (err)
 		goto err_put_conn;
 
+	/* Handle umasking inside the fuse code */
+	if (sb->s_flags & MS_POSIXACL)
+		fc->dont_mask = 1;
+	sb->s_flags |= MS_POSIXACL;
+
 	fc->release = fuse_free_conn;
 	fc->flags = d.flags;
 	fc->user_id = d.user_id;
diff --git a/include/linux/fuse.h b/include/linux/fuse.h
index d41ed593f79f..e2b816a62488 100644
--- a/include/linux/fuse.h
+++ b/include/linux/fuse.h
@@ -25,6 +25,9 @@
  *  - add IOCTL message
  *  - add unsolicited notification support
  *  - add POLL message and NOTIFY_POLL notification
+ *
+ * 7.12
+ *  - add umask flag to input argument of open, mknod and mkdir
  */
 
 #ifndef _LINUX_FUSE_H
@@ -36,7 +39,7 @@
 #define FUSE_KERNEL_VERSION 7
 
 /** Minor version number of this interface */
-#define FUSE_KERNEL_MINOR_VERSION 11
+#define FUSE_KERNEL_MINOR_VERSION 12
 
 /** The node ID of the root inode */
 #define FUSE_ROOT_ID 1
@@ -112,6 +115,7 @@ struct fuse_file_lock {
  * INIT request/reply flags
  *
  * FUSE_EXPORT_SUPPORT: filesystem handles lookups of "." and ".."
+ * FUSE_DONT_MASK: don't apply umask to file mode on create operations
  */
 #define FUSE_ASYNC_READ		(1 << 0)
 #define FUSE_POSIX_LOCKS	(1 << 1)
@@ -119,6 +123,7 @@ struct fuse_file_lock {
 #define FUSE_ATOMIC_O_TRUNC	(1 << 3)
 #define FUSE_EXPORT_SUPPORT	(1 << 4)
 #define FUSE_BIG_WRITES		(1 << 5)
+#define FUSE_DONT_MASK		(1 << 6)
 
 /**
  * CUSE INIT request/reply flags
@@ -262,14 +267,18 @@ struct fuse_attr_out {
 	struct fuse_attr attr;
 };
 
+#define FUSE_COMPAT_MKNOD_IN_SIZE 8
+
 struct fuse_mknod_in {
 	__u32	mode;
 	__u32	rdev;
+	__u32	umask;
+	__u32	padding;
 };
 
 struct fuse_mkdir_in {
 	__u32	mode;
-	__u32	padding;
+	__u32	umask;
 };
 
 struct fuse_rename_in {
@@ -300,8 +309,15 @@ struct fuse_setattr_in {
 };
 
 struct fuse_open_in {
+	__u32	flags;
+	__u32	unused;
+};
+
+struct fuse_create_in {
 	__u32	flags;
 	__u32	mode;
+	__u32	umask;
+	__u32	padding;
 };
 
 struct fuse_open_out {
-- 
cgit v1.2.3-59-g8ed1b


From 3b463ae0c6264f70e5d4c0a9c46af20fed43c96e Mon Sep 17 00:00:00 2001
From: John Muir <muirj@nortel.com>
Date: Sun, 31 May 2009 11:13:57 -0400
Subject: fuse: invalidation reverse calls

Add notification messages that allow the filesystem to invalidate VFS
caches.

Two notifications are added:

 1) inode invalidation

   - invalidate cached attributes
   - invalidate a range of pages in the page cache (this is optional)

 2) dentry invalidation

   - try to invalidate a subtree in the dentry cache

Care must be taken while accessing the 'struct super_block' for the
mount, as it can go away while an invalidation is in progress.  To
prevent this, introduce a rw-semaphore, that is taken for read during
the invalidation and taken for write in the ->kill_sb callback.

Cc: Csaba Henk <csaba@gluster.com>
Cc: Anand Avati <avati@zresearch.com>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
---
 fs/fuse/dev.c        | 81 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 fs/fuse/dir.c        | 37 ++++++++++++++++++++++++
 fs/fuse/fuse_i.h     | 24 ++++++++++++++++
 fs/fuse/inode.c      | 59 ++++++++++++++++++++++++++++++++++++--
 include/linux/fuse.h | 16 +++++++++++
 5 files changed, 214 insertions(+), 3 deletions(-)

diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index 8a11a8c67c42..f58ecbc416c8 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -849,6 +849,81 @@ err:
 	return err;
 }
 
+static int fuse_notify_inval_inode(struct fuse_conn *fc, unsigned int size,
+				   struct fuse_copy_state *cs)
+{
+	struct fuse_notify_inval_inode_out outarg;
+	int err = -EINVAL;
+
+	if (size != sizeof(outarg))
+		goto err;
+
+	err = fuse_copy_one(cs, &outarg, sizeof(outarg));
+	if (err)
+		goto err;
+	fuse_copy_finish(cs);
+
+	down_read(&fc->killsb);
+	err = -ENOENT;
+	if (!fc->sb)
+		goto err_unlock;
+
+	err = fuse_reverse_inval_inode(fc->sb, outarg.ino,
+				       outarg.off, outarg.len);
+
+err_unlock:
+	up_read(&fc->killsb);
+	return err;
+
+err:
+	fuse_copy_finish(cs);
+	return err;
+}
+
+static int fuse_notify_inval_entry(struct fuse_conn *fc, unsigned int size,
+				   struct fuse_copy_state *cs)
+{
+	struct fuse_notify_inval_entry_out outarg;
+	int err = -EINVAL;
+	char buf[FUSE_NAME_MAX+1];
+	struct qstr name;
+
+	if (size < sizeof(outarg))
+		goto err;
+
+	err = fuse_copy_one(cs, &outarg, sizeof(outarg));
+	if (err)
+		goto err;
+
+	err = -ENAMETOOLONG;
+	if (outarg.namelen > FUSE_NAME_MAX)
+		goto err;
+
+	name.name = buf;
+	name.len = outarg.namelen;
+	err = fuse_copy_one(cs, buf, outarg.namelen + 1);
+	if (err)
+		goto err;
+	fuse_copy_finish(cs);
+	buf[outarg.namelen] = 0;
+	name.hash = full_name_hash(name.name, name.len);
+
+	down_read(&fc->killsb);
+	err = -ENOENT;
+	if (!fc->sb)
+		goto err_unlock;
+
+	err = fuse_reverse_inval_entry(fc->sb, outarg.parent, &name);
+
+err_unlock:
+	up_read(&fc->killsb);
+	return err;
+
+err:
+	fuse_copy_finish(cs);
+	return err;
+}
+
 static int fuse_notify(struct fuse_conn *fc, enum fuse_notify_code code,
 		       unsigned int size, struct fuse_copy_state *cs)
 {
@@ -856,6 +931,12 @@ static int fuse_notify(struct fuse_conn *fc, enum fuse_notify_code code,
 	case FUSE_NOTIFY_POLL:
 		return fuse_notify_poll(fc, size, cs);
 
+	case FUSE_NOTIFY_INVAL_INODE:
+		return fuse_notify_inval_inode(fc, size, cs);
+
+	case FUSE_NOTIFY_INVAL_ENTRY:
+		return fuse_notify_inval_entry(fc, size, cs);
+
 	default:
 		fuse_copy_finish(cs);
 		return -EINVAL;
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 6b700734e519..e703654e7f40 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -859,6 +859,43 @@ int fuse_update_attributes(struct inode *inode, struct kstat *stat,
 	return err;
 }
 
+int fuse_reverse_inval_entry(struct super_block *sb, u64 parent_nodeid,
+			     struct qstr *name)
+{
+	int err = -ENOTDIR;
+	struct inode *parent;
+	struct dentry *dir;
+	struct dentry *entry;
+
+	parent = ilookup5(sb, parent_nodeid, fuse_inode_eq, &parent_nodeid);
+	if (!parent)
+		return -ENOENT;
+
+	mutex_lock(&parent->i_mutex);
+	if (!S_ISDIR(parent->i_mode))
+		goto unlock;
+
+	err = -ENOENT;
+	dir = d_find_alias(parent);
+	if (!dir)
+		goto unlock;
+
+	entry = d_lookup(dir, name);
+	dput(dir);
+	if (!entry)
+		goto unlock;
+
+	fuse_invalidate_attr(parent);
+	fuse_invalidate_entry(entry);
+	dput(entry);
+	err = 0;
+
+ unlock:
+	mutex_unlock(&parent->i_mutex);
+	iput(parent);
+	return err;
+}
+
 /*
  * Calling into a user-controlled filesystem gives the filesystem
  * daemon ptrace-like capabilities over the requester process.  This
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index ede4f77b2d6c..52b641fc0faf 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -484,6 +484,12 @@ struct fuse_conn {
 
 	/** Called on final put */
 	void (*release)(struct fuse_conn *);
+
+	/** Super block for this connection. */
+	struct super_block *sb;
+
+	/** Read/write semaphore to hold when accessing sb. */
+	struct rw_semaphore killsb;
 };
 
 static inline struct fuse_conn *get_fuse_conn_super(struct super_block *sb)
@@ -511,6 +517,11 @@ extern const struct file_operations fuse_dev_operations;
 
 extern const struct dentry_operations fuse_dentry_operations;
 
+/**
+ * Inode to nodeid comparison.
+ */
+int fuse_inode_eq(struct inode *inode, void *_nodeidp);
+
 /**
  * Get a filled in inode
  */
@@ -711,6 +722,19 @@ void fuse_release_nowrite(struct inode *inode);
 
 u64 fuse_get_attr_version(struct fuse_conn *fc);
 
+/**
+ * File-system tells the kernel to invalidate cache for the given node id.
+ */
+int fuse_reverse_inval_inode(struct super_block *sb, u64 nodeid,
+			     loff_t offset, loff_t len);
+
+/**
+ * File-system tells the kernel to invalidate parent attributes and
+ * the dentry matching parent/name.
+ */
+int fuse_reverse_inval_entry(struct super_block *sb, u64 parent_nodeid,
+			     struct qstr *name);
+
 int fuse_do_open(struct fuse_conn *fc, u64 nodeid, struct file *file,
 		 bool isdir);
 ssize_t fuse_direct_io(struct file *file, const char __user *buf,
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 6cc501bd0187..f91ccc4a189d 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -206,7 +206,7 @@ static void fuse_init_inode(struct inode *inode, struct fuse_attr *attr)
 		BUG();
 }
 
-static int fuse_inode_eq(struct inode *inode, void *_nodeidp)
+int fuse_inode_eq(struct inode *inode, void *_nodeidp)
 {
 	u64 nodeid = *(u64 *) _nodeidp;
 	if (get_node_id(inode) == nodeid)
@@ -257,6 +257,31 @@ struct inode *fuse_iget(struct super_block *sb, u64 nodeid,
 	return inode;
 }
 
+int fuse_reverse_inval_inode(struct super_block *sb, u64 nodeid,
+			     loff_t offset, loff_t len)
+{
+	struct inode *inode;
+	pgoff_t pg_start;
+	pgoff_t pg_end;
+
+	inode = ilookup5(sb, nodeid, fuse_inode_eq, &nodeid);
+	if (!inode)
+		return -ENOENT;
+
+	fuse_invalidate_attr(inode);
+	if (offset >= 0) {
+		pg_start = offset >> PAGE_CACHE_SHIFT;
+		if (len <= 0)
+			pg_end = -1;
+		else
+			pg_end = (offset + len - 1) >> PAGE_CACHE_SHIFT;
+		invalidate_inode_pages2_range(inode->i_mapping,
+					      pg_start, pg_end);
+	}
+	iput(inode);
+	return 0;
+}
+
 static void fuse_umount_begin(struct super_block *sb)
 {
 	fuse_abort_conn(get_fuse_conn_super(sb));
@@ -480,6 +505,7 @@ void fuse_conn_init(struct fuse_conn *fc)
 	memset(fc, 0, sizeof(*fc));
 	spin_lock_init(&fc->lock);
 	mutex_init(&fc->inst_mutex);
+	init_rwsem(&fc->killsb);
 	atomic_set(&fc->count, 1);
 	init_waitqueue_head(&fc->waitq);
 	init_waitqueue_head(&fc->blocked_waitq);
@@ -862,6 +888,7 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent)
 	fuse_conn_init(fc);
 
 	fc->dev = sb->s_dev;
+	fc->sb = sb;
 	err = fuse_bdi_init(fc, sb);
 	if (err)
 		goto err_put_conn;
@@ -948,12 +975,25 @@ static int fuse_get_sb(struct file_system_type *fs_type,
 	return get_sb_nodev(fs_type, flags, raw_data, fuse_fill_super, mnt);
 }
 
+static void fuse_kill_sb_anon(struct super_block *sb)
+{
+	struct fuse_conn *fc = get_fuse_conn_super(sb);
+
+	if (fc) {
+		down_write(&fc->killsb);
+		fc->sb = NULL;
+		up_write(&fc->killsb);
+	}
+
+	kill_anon_super(sb);
+}
+
 static struct file_system_type fuse_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "fuse",
 	.fs_flags	= FS_HAS_SUBTYPE,
 	.get_sb		= fuse_get_sb,
-	.kill_sb	= kill_anon_super,
+	.kill_sb	= fuse_kill_sb_anon,
 };
 
 #ifdef CONFIG_BLOCK
@@ -965,11 +1005,24 @@ static int fuse_get_sb_blk(struct file_system_type *fs_type,
 			   mnt);
 }
 
+static void fuse_kill_sb_blk(struct super_block *sb)
+{
+	struct fuse_conn *fc = get_fuse_conn_super(sb);
+
+	if (fc) {
+		down_write(&fc->killsb);
+		fc->sb = NULL;
+		up_write(&fc->killsb);
+	}
+
+	kill_block_super(sb);
+}
+
 static struct file_system_type fuseblk_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "fuseblk",
 	.get_sb		= fuse_get_sb_blk,
-	.kill_sb	= kill_block_super,
+	.kill_sb	= fuse_kill_sb_blk,
 	.fs_flags	= FS_REQUIRES_DEV | FS_HAS_SUBTYPE,
 };
 
diff --git a/include/linux/fuse.h b/include/linux/fuse.h
index e2b816a62488..cf593bf9fd32 100644
--- a/include/linux/fuse.h
+++ b/include/linux/fuse.h
@@ -28,6 +28,8 @@
  *
  * 7.12
  *  - add umask flag to input argument of open, mknod and mkdir
+ *  - add notification messages for invalidation of inodes and
+ *    directory entries
  */
 
 #ifndef _LINUX_FUSE_H
@@ -229,6 +231,8 @@ enum fuse_opcode {
 
 enum fuse_notify_code {
 	FUSE_NOTIFY_POLL   = 1,
+	FUSE_NOTIFY_INVAL_INODE = 2,
+	FUSE_NOTIFY_INVAL_ENTRY = 3,
 	FUSE_NOTIFY_CODE_MAX,
 };
 
@@ -524,4 +528,16 @@ struct fuse_dirent {
 #define FUSE_DIRENT_SIZE(d) \
 	FUSE_DIRENT_ALIGN(FUSE_NAME_OFFSET + (d)->namelen)
 
+struct fuse_notify_inval_inode_out {
+	__u64	ino;
+	__s64	off;
+	__s64	len;
+};
+
+struct fuse_notify_inval_entry_out {
+	__u64	parent;
+	__u32	namelen;
+	__u32	padding;
+};
+
 #endif /* _LINUX_FUSE_H */
-- 
cgit v1.2.3-59-g8ed1b


From 01e532981460594fffbf9b992ecfc96a78369924 Mon Sep 17 00:00:00 2001
From: Naohiro Ooiwa <nooiwa@miraclelinux.com>
Date: Tue, 30 Jun 2009 12:44:19 -0700
Subject: bnx2x: Fix the behavior of ethtool when ONBOOT=no

This is the same fix as commit
7959ea254ed18faee41160b1c50b3c9664735967 ("bnx2: Fix the behavior of
ethtool when ONBOOT=no"), but for bnx2x:

--------------------
    When configure in ifcfg-eth* is ONBOOT=no,
    the behavior of ethtool command is wrong.

        # grep ONBOOT /etc/sysconfig/network-scripts/ifcfg-eth2
        ONBOOT=no
        # ethtool eth2 | tail -n1
                Link detected: yes

    I think "Link detected" should be "no".
--------------------

Signed-off-by: Naohiro Ooiwa <nooiwa@miraclelinux.com>
Acked-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/bnx2x_main.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/net/bnx2x_main.c b/drivers/net/bnx2x_main.c
index fbf1352e9c1c..951714a7f90a 100644
--- a/drivers/net/bnx2x_main.c
+++ b/drivers/net/bnx2x_main.c
@@ -8637,6 +8637,14 @@ static int bnx2x_nway_reset(struct net_device *dev)
 	return 0;
 }
 
+static u32
+bnx2x_get_link(struct net_device *dev)
+{
+	struct bnx2x *bp = netdev_priv(dev);
+
+	return bp->link_vars.link_up;
+}
+
 static int bnx2x_get_eeprom_len(struct net_device *dev)
 {
 	struct bnx2x *bp = netdev_priv(dev);
@@ -10034,7 +10042,7 @@ static struct ethtool_ops bnx2x_ethtool_ops = {
 	.get_msglevel		= bnx2x_get_msglevel,
 	.set_msglevel		= bnx2x_set_msglevel,
 	.nway_reset		= bnx2x_nway_reset,
-	.get_link		= ethtool_op_get_link,
+	.get_link		= bnx2x_get_link,
 	.get_eeprom_len		= bnx2x_get_eeprom_len,
 	.get_eeprom		= bnx2x_get_eeprom,
 	.set_eeprom		= bnx2x_set_eeprom,
-- 
cgit v1.2.3-59-g8ed1b


From 008440e3ad4b72f5048d1b1f6f5ed894fdc5ad08 Mon Sep 17 00:00:00 2001
From: Jarek Poplawski <jarkao2@gmail.com>
Date: Tue, 30 Jun 2009 12:47:19 -0700
Subject: ipv4: Fix fib_trie rebalancing, part 3

Alas current delaying of freeing old tnodes by RCU in trie_rebalance
is still not enough because we can free a top tnode before updating a
t->trie pointer.

Reported-by: Pawel Staszewski <pstaszewski@itcare.pl>
Tested-by: Pawel Staszewski <pstaszewski@itcare.pl>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/fib_trie.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 012cf5a68581..00a54b246dfe 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1021,6 +1021,9 @@ static void trie_rebalance(struct trie *t, struct tnode *tn)
 				      (struct node *)tn, wasfull);
 
 		tp = node_parent((struct node *) tn);
+		if (!tp)
+			rcu_assign_pointer(t->trie, (struct node *)tn);
+
 		tnode_free_flush();
 		if (!tp)
 			break;
-- 
cgit v1.2.3-59-g8ed1b


From f7c2df9b55212d5ec94169a4de11e44c683e0af4 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Tue, 30 Jun 2009 21:10:13 +0100
Subject: AFS: Fix lock imbalance

Don't unlock on vfs_rejected_lock path in afs_do_setlk, since the lock
is unlocked after abort_attempt label.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/afs/flock.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/fs/afs/flock.c b/fs/afs/flock.c
index 210acafe4a9b..3ff8bdd18fb3 100644
--- a/fs/afs/flock.c
+++ b/fs/afs/flock.c
@@ -432,7 +432,6 @@ vfs_rejected_lock:
 	list_del_init(&fl->fl_u.afs.link);
 	if (list_empty(&vnode->granted_locks))
 		afs_defer_unlock(vnode, key);
-	spin_unlock(&vnode->lock);
 	goto abort_attempt;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From 2be8412c6cef97b01dfaae71c04bf585d3d93a3b Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@linux.intel.com>
Date: Tue, 30 Jun 2009 14:02:00 -0700
Subject: [IA64] sprintf should not be used with same source & destination
 address

This happens to work at the moment but isn't a good idea so fix it the
simple way.

Resolves-bug: http://bugzilla.kernel.org/show_bug.cgi?id=13576

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/io_common.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/ia64/sn/kernel/io_common.c b/arch/ia64/sn/kernel/io_common.c
index 76645cf6ac5d..25831c47c579 100644
--- a/arch/ia64/sn/kernel/io_common.c
+++ b/arch/ia64/sn/kernel/io_common.c
@@ -435,7 +435,8 @@ void sn_generate_path(struct pci_bus *pci_bus, char *address)
 	bricktype = MODULE_GET_BTYPE(moduleid);
 	if ((bricktype == L1_BRICKTYPE_191010) ||
 	    (bricktype == L1_BRICKTYPE_1932))
-			sprintf(address, "%s^%d", address, geo_slot(geoid));
+			sprintf(address + strlen(address), "^%d",
+						geo_slot(geoid));
 }
 
 void __devinit
-- 
cgit v1.2.3-59-g8ed1b


From 58782b34e9ffcc04619634efe9658263344ed188 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Sun, 28 Jun 2009 09:26:07 -0700
Subject: [IA64] Remove unnecessary semicolons

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/esi.c  | 2 +-
 arch/ia64/kvm/process.c | 6 +++---
 arch/ia64/kvm/vcpu.c    | 2 +-
 arch/ia64/kvm/vtlb.c    | 4 ++--
 4 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/arch/ia64/kernel/esi.c b/arch/ia64/kernel/esi.c
index ebf4e988e78c..d5764a3d74af 100644
--- a/arch/ia64/kernel/esi.c
+++ b/arch/ia64/kernel/esi.c
@@ -65,7 +65,7 @@ static int __init esi_init (void)
 	}
 
 	if (!esi)
-		return -ENODEV;;
+		return -ENODEV;
 
 	systab = __va(esi);
 
diff --git a/arch/ia64/kvm/process.c b/arch/ia64/kvm/process.c
index a8f84da04b49..bb862fb224f2 100644
--- a/arch/ia64/kvm/process.c
+++ b/arch/ia64/kvm/process.c
@@ -130,7 +130,7 @@ static void collect_interruption(struct kvm_vcpu *vcpu)
 	if (vdcr & IA64_DCR_PP) {
 		vpsr |= IA64_PSR_PP;
 	} else {
-		vpsr &= ~IA64_PSR_PP;;
+		vpsr &= ~IA64_PSR_PP;
 	}
 
 	vcpu_set_psr(vcpu, vpsr);
@@ -594,11 +594,11 @@ static void set_pal_call_data(struct kvm_vcpu *vcpu)
 		p->u.pal_data.gr30 = vcpu_get_gr(vcpu, 30);
 		break;
 	case PAL_BRAND_INFO:
-		p->u.pal_data.gr29 = gr29;;
+		p->u.pal_data.gr29 = gr29;
 		p->u.pal_data.gr30 = kvm_trans_pal_call_args(vcpu, gr30);
 		break;
 	default:
-		p->u.pal_data.gr29 = gr29;;
+		p->u.pal_data.gr29 = gr29;
 		p->u.pal_data.gr30 = vcpu_get_gr(vcpu, 30);
 	}
 	p->u.pal_data.gr28 = gr28;
diff --git a/arch/ia64/kvm/vcpu.c b/arch/ia64/kvm/vcpu.c
index a2c6c15e4761..46b02cbcc874 100644
--- a/arch/ia64/kvm/vcpu.c
+++ b/arch/ia64/kvm/vcpu.c
@@ -406,7 +406,7 @@ void getreg(unsigned long regnum, unsigned long *val,
 	 * Now look at registers in [0-31] range and init correct UNAT
 	 */
 	addr = (unsigned long)regs;
-	unat = &regs->eml_unat;;
+	unat = &regs->eml_unat;
 
 	addr += gr_info[regnum];
 
diff --git a/arch/ia64/kvm/vtlb.c b/arch/ia64/kvm/vtlb.c
index 4290a429bf7c..20b3852f7a6e 100644
--- a/arch/ia64/kvm/vtlb.c
+++ b/arch/ia64/kvm/vtlb.c
@@ -135,7 +135,7 @@ struct thash_data *__vtr_lookup(struct kvm_vcpu *vcpu, u64 va, int type)
 	u64 rid;
 
 	rid = vcpu_get_rr(vcpu, va);
-	rid = rid & RR_RID_MASK;;
+	rid = rid & RR_RID_MASK;
 	if (type == D_TLB) {
 		if (vcpu_quick_region_check(vcpu->arch.dtr_regions, va)) {
 			for (trp = (struct thash_data *)&vcpu->arch.dtrs, i = 0;
@@ -518,7 +518,7 @@ struct thash_data *vtlb_lookup(struct kvm_vcpu *v, u64 va, int is_data)
 
 	struct thash_cb *hcb = &v->arch.vtlb;
 
-	cch = __vtr_lookup(v, va, is_data);;
+	cch = __vtr_lookup(v, va, is_data);
 	if (cch)
 		return cch;
 
-- 
cgit v1.2.3-59-g8ed1b


From fa276f36f3d8743295e067fb483b42dca8bd1ece Mon Sep 17 00:00:00 2001
From: Jan Beulich <JBeulich@novell.com>
Date: Tue, 30 Jun 2009 12:01:57 +0100
Subject: [IA64] address compiler warnings perfmon.c/salinfo.c

perfmon.c has a dubious cast directly from "int" to "void *". Add
an intermediate cast to "long" to keep gcc happy.

salinfo.c uses "down_trylock()" in a highly creative way (explained
in the comments in the file) ... but it does kick out this warning:

 arch/ia64/kernel/salinfo.c:195: warning: ignoring return value of 'down_trylock'

which people occasionally try to "fix" in ways that do not work. Use some
casts to keep gcc quiet.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/perfmon.c | 2 +-
 arch/ia64/kernel/salinfo.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index abce2468a40b..f1782705b1f7 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -5603,7 +5603,7 @@ pfm_interrupt_handler(int irq, void *arg)
  * /proc/perfmon interface, for debug only
  */
 
-#define PFM_PROC_SHOW_HEADER	((void *)nr_cpu_ids+1)
+#define PFM_PROC_SHOW_HEADER	((void *)(long)nr_cpu_ids+1)
 
 static void *
 pfm_proc_start(struct seq_file *m, loff_t *pos)
diff --git a/arch/ia64/kernel/salinfo.c b/arch/ia64/kernel/salinfo.c
index 7053c55b7649..e6676fca4828 100644
--- a/arch/ia64/kernel/salinfo.c
+++ b/arch/ia64/kernel/salinfo.c
@@ -192,7 +192,7 @@ struct salinfo_platform_oemdata_parms {
 static void
 salinfo_work_to_do(struct salinfo_data *data)
 {
-	down_trylock(&data->mutex);
+	(void)(down_trylock(&data->mutex) ?: 0);
 	up(&data->mutex);
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From fadfd2b6ba8838a6cc458dbae214993a177a3ee9 Mon Sep 17 00:00:00 2001
From: Tony Luck <tony.luck@intel.com>
Date: Tue, 30 Jun 2009 14:28:54 -0700
Subject: Add Fenghua Yu as temporary co-maintainer for ia64

I'm taking my sabbatical from Intel for July/August 2009.
Fenghua Yu will handle ia64 architecture while I'm gone.

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 MAINTAINERS | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 05a9a563042f..f3bbc759360c 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2851,7 +2851,9 @@ S:	Maintained
 
 IA64 (Itanium) PLATFORM
 P:	Tony Luck
+P:	Fenghua Yu
 M:	tony.luck@intel.com
+M:	fenghua.yu@intel.com
 L:	linux-ia64@vger.kernel.org
 W:	http://www.ia64-linux.org/
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6.git
-- 
cgit v1.2.3-59-g8ed1b


From 25903407da21552419e0955705d6d8c8e601cb2e Mon Sep 17 00:00:00 2001
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Date: Tue, 30 Jun 2009 19:01:20 -0300
Subject: perf report: Add --dsos parameter

So that we can filter by dso. Symbols in other dsos won't be
accounted for.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1246399282-20934-2-git-send-email-acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/Documentation/perf-report.txt |   4 +
 tools/perf/Makefile                      |   2 +
 tools/perf/builtin-report.c              |  16 +++
 tools/perf/util/strlist.c                | 184 +++++++++++++++++++++++++++++++
 tools/perf/util/strlist.h                |  32 ++++++
 5 files changed, 238 insertions(+)
 create mode 100644 tools/perf/util/strlist.c
 create mode 100644 tools/perf/util/strlist.h

diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
index 40c1db83a16d..13d85ca8c914 100644
--- a/tools/perf/Documentation/perf-report.txt
+++ b/tools/perf/Documentation/perf-report.txt
@@ -20,6 +20,10 @@ OPTIONS
 -i::
 --input=::
         Input file name. (default: perf.data)
+-d::
+--dsos=::
+	Only consider symbols in these dsos. CSV that understands
+	file://filename entries.
 
 SEE ALSO
 --------
diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index 1c1296d8a64b..9c6d0ae3708e 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -301,6 +301,7 @@ LIB_H += util/util.h
 LIB_H += util/help.h
 LIB_H += util/strbuf.h
 LIB_H += util/string.h
+LIB_H += util/strlist.h
 LIB_H += util/run-command.h
 LIB_H += util/sigchain.h
 LIB_H += util/symbol.h
@@ -322,6 +323,7 @@ LIB_OBJS += util/run-command.o
 LIB_OBJS += util/quote.o
 LIB_OBJS += util/strbuf.o
 LIB_OBJS += util/string.o
+LIB_OBJS += util/strlist.o
 LIB_OBJS += util/usage.o
 LIB_OBJS += util/wrapper.o
 LIB_OBJS += util/sigchain.o
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index ed391db9e0f8..7c6b6e776718 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -16,6 +16,7 @@
 #include "util/symbol.h"
 #include "util/string.h"
 #include "util/callchain.h"
+#include "util/strlist.h"
 
 #include "perf.h"
 #include "util/header.h"
@@ -32,6 +33,8 @@ static char		*vmlinux = NULL;
 
 static char		default_sort_order[] = "comm,dso";
 static char		*sort_order = default_sort_order;
+static char		*dso_list_str;
+static struct strlist	*dso_list;
 
 static int		input;
 static int		show_mask = SHOW_KERNEL | SHOW_USER | SHOW_HV;
@@ -1272,6 +1275,9 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 	if (show & show_mask) {
 		struct symbol *sym = resolve_symbol(thread, &map, &dso, &ip);
 
+		if (dso_list && dso && dso->name && !strlist__has_entry(dso_list, dso->name))
+			return 0;
+
 		if (hist_entry__add(thread, map, dso, sym, ip, chain, level, period)) {
 			eprintf("problem incrementing symbol count, skipping event\n");
 			return -1;
@@ -1659,6 +1665,8 @@ static const struct option options[] = {
 	OPT_BOOLEAN('x', "exclude-other", &exclude_other,
 		    "Only display entries with parent-match"),
 	OPT_BOOLEAN('c', "callchain", &callchain, "Display callchains"),
+	OPT_STRING('d', "dsos", &dso_list_str, "dso[,dso...]",
+		   "only consider symbols in these dsos"),
 	OPT_END()
 };
 
@@ -1698,6 +1706,14 @@ int cmd_report(int argc, const char **argv, const char *prefix)
 	if (argc)
 		usage_with_options(report_usage, options);
 
+	if (dso_list_str) {
+		dso_list = strlist__new(true, dso_list_str);
+		if (!dso_list) {
+			fprintf(stderr, "problems parsing dso list\n");
+			exit(129);
+		}
+	}
+
 	setup_pager();
 
 	return __cmd_report();
diff --git a/tools/perf/util/strlist.c b/tools/perf/util/strlist.c
new file mode 100644
index 000000000000..025a78edfffe
--- /dev/null
+++ b/tools/perf/util/strlist.c
@@ -0,0 +1,184 @@
+/*
+ * (c) 2009 Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Licensed under the GPLv2.
+ */
+
+#include "strlist.h"
+#include <errno.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+static struct str_node *str_node__new(const char *s, bool dupstr)
+{
+	struct str_node *self = malloc(sizeof(*self));
+
+	if (self != NULL) {
+		if (dupstr) {
+			s = strdup(s);
+			if (s == NULL)
+				goto out_delete;
+		}
+		self->s = s;
+	}
+
+	return self;
+
+out_delete:
+	free(self);
+	return NULL;
+}
+
+static void str_node__delete(struct str_node *self, bool dupstr)
+{
+	if (dupstr)
+		free((void *)self->s);
+	free(self);
+}
+
+int strlist__add(struct strlist *self, const char *new_entry)
+{
+	struct rb_node **p = &self->entries.rb_node;
+	struct rb_node *parent = NULL;
+	struct str_node *sn;
+
+	while (*p != NULL) {
+		int rc;
+
+		parent = *p;
+		sn = rb_entry(parent, struct str_node, rb_node);
+		rc = strcmp(sn->s, new_entry);
+
+		if (rc > 0)
+			p = &(*p)->rb_left;
+		else if (rc < 0)
+			p = &(*p)->rb_right;
+		else
+			return -EEXIST;
+	}
+
+	sn = str_node__new(new_entry, self->dupstr);
+	if (sn == NULL)
+		return -ENOMEM;
+
+	rb_link_node(&sn->rb_node, parent, p);
+	rb_insert_color(&sn->rb_node, &self->entries);
+
+	return 0;
+}
+
+int strlist__load(struct strlist *self, const char *filename)
+{
+	char entry[1024];
+	int err;
+	FILE *fp = fopen(filename, "r");
+
+	if (fp == NULL)
+		return errno;
+
+	while (fgets(entry, sizeof(entry), fp) != NULL) {
+		const size_t len = strlen(entry);
+
+		if (len == 0)
+			continue;
+		entry[len - 1] = '\0';
+
+		err = strlist__add(self, entry);
+		if (err != 0)
+			goto out;
+	}
+
+	err = 0;
+out:
+	fclose(fp);
+	return err;
+}
+
+void strlist__remove(struct strlist *self, struct str_node *sn)
+{
+	rb_erase(&sn->rb_node, &self->entries);
+	str_node__delete(sn, self->dupstr);
+}
+
+bool strlist__has_entry(struct strlist *self, const char *entry)
+{
+	struct rb_node **p = &self->entries.rb_node;
+	struct rb_node *parent = NULL;
+
+	while (*p != NULL) {
+		struct str_node *sn;
+		int rc;
+
+		parent = *p;
+		sn = rb_entry(parent, struct str_node, rb_node);
+		rc = strcmp(sn->s, entry);
+
+		if (rc > 0)
+			p = &(*p)->rb_left;
+		else if (rc < 0)
+			p = &(*p)->rb_right;
+		else
+			return true;
+	}
+
+	return false;
+}
+
+static int strlist__parse_list_entry(struct strlist *self, const char *s)
+{
+	if (strncmp(s, "file://", 7) == 0)
+		return strlist__load(self, s + 7);
+
+	return strlist__add(self, s);
+}
+
+int strlist__parse_list(struct strlist *self, const char *s)
+{
+	char *sep;
+	int err;
+
+	while ((sep = strchr(s, ',')) != NULL) {
+		*sep = '\0';
+		err = strlist__parse_list_entry(self, s);
+		*sep = ',';
+		if (err != 0)
+			return err;
+		s = sep + 1;
+	}
+
+	return *s ? strlist__parse_list_entry(self, s) : 0;
+}
+
+struct strlist *strlist__new(bool dupstr, const char *slist)
+{
+	struct strlist *self = malloc(sizeof(*self));
+
+	if (self != NULL) {
+		self->entries = RB_ROOT;
+		self->dupstr = dupstr;
+		if (slist && strlist__parse_list(self, slist) != 0)
+			goto out_error;
+	}
+
+	return self;
+out_error:
+	free(self);
+	return NULL;
+}
+
+void strlist__delete(struct strlist *self)
+{
+	if (self != NULL) {
+		struct str_node *pos;
+		struct rb_node *next = rb_first(&self->entries);
+
+		while (next) {
+			pos = rb_entry(next, struct str_node, rb_node);
+			next = rb_next(&pos->rb_node);
+			strlist__remove(self, pos);
+		}
+		self->entries = RB_ROOT;
+		free(self);
+	}
+}
diff --git a/tools/perf/util/strlist.h b/tools/perf/util/strlist.h
new file mode 100644
index 000000000000..2fb117fb4b67
--- /dev/null
+++ b/tools/perf/util/strlist.h
@@ -0,0 +1,32 @@
+#ifndef STRLIST_H_
+#define STRLIST_H_
+
+#include "rbtree.h"
+#include <stdbool.h>
+
+struct str_node {
+	struct rb_node rb_node;
+	const char     *s;
+};
+
+struct strlist {
+	struct rb_root entries;
+	bool dupstr;
+};
+
+struct strlist *strlist__new(bool dupstr, const char *slist);
+void strlist__delete(struct strlist *self);
+
+void strlist__remove(struct strlist *self, struct str_node *sn);
+int strlist__load(struct strlist *self, const char *filename);
+int strlist__add(struct strlist *self, const char *str);
+
+bool strlist__has_entry(struct strlist *self, const char *entry);
+
+static inline bool strlist__empty(const struct strlist *self)
+{
+	return rb_first(&self->entries) == NULL;
+}
+
+int strlist__parse_list(struct strlist *self, const char *s);
+#endif /* STRLIST_H_ */
-- 
cgit v1.2.3-59-g8ed1b


From cc8b88b15ab8e5ae162a46c4b6b286b555190dd1 Mon Sep 17 00:00:00 2001
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Date: Tue, 30 Jun 2009 19:01:21 -0300
Subject: perf report: Add --comms parameter

So that we can filter by comm. Symbols in other comms won't be
accounted for.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1246399282-20934-3-git-send-email-acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/Documentation/perf-report.txt |  4 ++++
 tools/perf/builtin-report.c              | 33 ++++++++++++++++++++++----------
 2 files changed, 27 insertions(+), 10 deletions(-)

diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
index 13d85ca8c914..4c44ef1747b9 100644
--- a/tools/perf/Documentation/perf-report.txt
+++ b/tools/perf/Documentation/perf-report.txt
@@ -24,6 +24,10 @@ OPTIONS
 --dsos=::
 	Only consider symbols in these dsos. CSV that understands
 	file://filename entries.
+-C::
+--comms=::
+	Only consider symbols in these comms. CSV that understands
+	file://filename entries.
 
 SEE ALSO
 --------
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 7c6b6e776718..8143477b7ef7 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -33,8 +33,8 @@ static char		*vmlinux = NULL;
 
 static char		default_sort_order[] = "comm,dso";
 static char		*sort_order = default_sort_order;
-static char		*dso_list_str;
-static struct strlist	*dso_list;
+static char		*dso_list_str, *comm_list_str;
+static struct strlist	*dso_list, *comm_list;
 
 static int		input;
 static int		show_mask = SHOW_KERNEL | SHOW_USER | SHOW_HV;
@@ -240,7 +240,7 @@ static u64 vdso__map_ip(struct map *map, u64 ip)
 
 static inline int is_anon_memory(const char *filename)
 {
-     return strcmp(filename, "//anon") == 0;
+	return strcmp(filename, "//anon") == 0;
 }
 
 static struct map *map__new(struct mmap_event *event)
@@ -1253,6 +1253,9 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 		return -1;
 	}
 
+	if (comm_list && !strlist__has_entry(comm_list, thread->comm))
+		return 0;
+
 	if (event->header.misc & PERF_EVENT_MISC_KERNEL) {
 		show = SHOW_KERNEL;
 		level = 'k';
@@ -1667,6 +1670,8 @@ static const struct option options[] = {
 	OPT_BOOLEAN('c', "callchain", &callchain, "Display callchains"),
 	OPT_STRING('d', "dsos", &dso_list_str, "dso[,dso...]",
 		   "only consider symbols in these dsos"),
+	OPT_STRING('C', "comms", &comm_list_str, "comm[,comm...]",
+		   "only consider symbols in these comms"),
 	OPT_END()
 };
 
@@ -1685,6 +1690,19 @@ static void setup_sorting(void)
 	free(str);
 }
 
+static void setup_list(struct strlist **list, const char *list_str,
+		       const char *list_name)
+{
+	if (list_str) {
+		*list = strlist__new(true, list_str);
+		if (!*list) {
+			fprintf(stderr, "problems parsing %s list\n",
+				list_name);
+			exit(129);
+		}
+	}
+}
+
 int cmd_report(int argc, const char **argv, const char *prefix)
 {
 	symbol__init();
@@ -1706,13 +1724,8 @@ int cmd_report(int argc, const char **argv, const char *prefix)
 	if (argc)
 		usage_with_options(report_usage, options);
 
-	if (dso_list_str) {
-		dso_list = strlist__new(true, dso_list_str);
-		if (!dso_list) {
-			fprintf(stderr, "problems parsing dso list\n");
-			exit(129);
-		}
-	}
+	setup_list(&dso_list, dso_list_str, "dso");
+	setup_list(&comm_list, comm_list_str, "comm");
 
 	setup_pager();
 
-- 
cgit v1.2.3-59-g8ed1b


From 7bec7a9134c25cecb0d7029199b59f7b1bef35b8 Mon Sep 17 00:00:00 2001
From: Arnaldo Carvalho de Melo <acme@redhat.com>
Date: Tue, 30 Jun 2009 19:01:22 -0300
Subject: perf report: Add --symbols parameter

So that we can filter by symbol name.

The 'pfunct' utility in the 'dwarves' package can be used to
create a file with the functions one wants.

Example:

[acme@doppio pahole]$ pfunct /usr/lib/debug/usr/lib64/libdw-0.141.so.debug | grep dwarf > /tmp/dwarf.symbols
[acme@doppio pahole]$ wc -l /tmp/dwarf.symbols
93 /tmp/dwarf.symbols
[acme@doppio pahole]$ head -3 /tmp/dwarf.symbols
dwfl_addrdwarf
dwfl_module_getdwarf
dwfl_getdwarf
[acme@doppio pahole]$ perf report --sort comm,dso,symbol --comms pahole --dsos /usr/lib64/libdw-0.141.so --symbols file:///tmp/dwarf.symbols

    33.99%            pahole  /usr/lib64/libdw-0.141.so  [.] dwarf_tag
    29.07%            pahole  /usr/lib64/libdw-0.141.so  [.] dwarf_decl_file
    27.71%            pahole  /usr/lib64/libdw-0.141.so  [.] dwarf_getsrclines
     4.54%            pahole  /usr/lib64/libdw-0.141.so  0x00000000007400
     3.93%            pahole  /usr/lib64/libdw-0.141.so  [.] dwarf_decl_line
     0.46%            pahole  /usr/lib64/libdw-0.141.so  [.] dwarf_getlocation
     0.18%            pahole  /usr/lib64/libdw-0.141.so  [.] __libdwarf_next_prime
     0.13%            pahole  /usr/lib64/libdw-0.141.so  [.] dwarf_diecu

[acme@doppio pahole]$

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1246399282-20934-4-git-send-email-acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 tools/perf/Documentation/perf-report.txt |  4 ++++
 tools/perf/builtin-report.c              | 10 ++++++++--
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt
index 4c44ef1747b9..8aa3f8c88707 100644
--- a/tools/perf/Documentation/perf-report.txt
+++ b/tools/perf/Documentation/perf-report.txt
@@ -28,6 +28,10 @@ OPTIONS
 --comms=::
 	Only consider symbols in these comms. CSV that understands
 	file://filename entries.
+-S::
+--symbols=::
+	Only consider these symbols. CSV that understands
+	file://filename entries.
 
 SEE ALSO
 --------
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 8143477b7ef7..135b7837e6bf 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -33,8 +33,8 @@ static char		*vmlinux = NULL;
 
 static char		default_sort_order[] = "comm,dso";
 static char		*sort_order = default_sort_order;
-static char		*dso_list_str, *comm_list_str;
-static struct strlist	*dso_list, *comm_list;
+static char		*dso_list_str, *comm_list_str, *sym_list_str;
+static struct strlist	*dso_list, *comm_list, *sym_list;
 
 static int		input;
 static int		show_mask = SHOW_KERNEL | SHOW_USER | SHOW_HV;
@@ -1281,6 +1281,9 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head)
 		if (dso_list && dso && dso->name && !strlist__has_entry(dso_list, dso->name))
 			return 0;
 
+		if (sym_list && sym && !strlist__has_entry(sym_list, sym->name))
+			return 0;
+
 		if (hist_entry__add(thread, map, dso, sym, ip, chain, level, period)) {
 			eprintf("problem incrementing symbol count, skipping event\n");
 			return -1;
@@ -1672,6 +1675,8 @@ static const struct option options[] = {
 		   "only consider symbols in these dsos"),
 	OPT_STRING('C', "comms", &comm_list_str, "comm[,comm...]",
 		   "only consider symbols in these comms"),
+	OPT_STRING('S', "symbols", &sym_list_str, "symbol[,symbol...]",
+		   "only consider these symbols"),
 	OPT_END()
 };
 
@@ -1726,6 +1731,7 @@ int cmd_report(int argc, const char **argv, const char *prefix)
 
 	setup_list(&dso_list, dso_list_str, "dso");
 	setup_list(&comm_list, comm_list_str, "comm");
+	setup_list(&sym_list, sym_list_str, "symbol");
 
 	setup_pager();
 
-- 
cgit v1.2.3-59-g8ed1b


From 8f6c2e4b325a8e9f8f47febb2fd0ed4fae7d45a9 Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Wed, 1 Jul 2009 11:13:45 +1000
Subject: md: Use new topology calls to indicate alignment and I/O sizes

Switch MD over to the new disk_stack_limits() function which checks for
aligment and adjusts preferred I/O sizes when stacking.

Also indicate preferred I/O sizes where applicable.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/linear.c    |  4 ++--
 drivers/md/multipath.c |  7 ++++---
 drivers/md/raid0.c     |  9 +++++++--
 drivers/md/raid1.c     |  9 ++++-----
 drivers/md/raid10.c    | 19 +++++++++++++------
 drivers/md/raid5.c     | 10 +++++++++-
 6 files changed, 39 insertions(+), 19 deletions(-)

diff --git a/drivers/md/linear.c b/drivers/md/linear.c
index 15c8b7b25a9b..5810fa906af0 100644
--- a/drivers/md/linear.c
+++ b/drivers/md/linear.c
@@ -166,8 +166,8 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
 			rdev->sectors = sectors * mddev->chunk_sectors;
 		}
 
-		blk_queue_stack_limits(mddev->queue,
-				       rdev->bdev->bd_disk->queue);
+		disk_stack_limits(mddev->gendisk, rdev->bdev,
+				  rdev->data_offset << 9);
 		/* as we don't honour merge_bvec_fn, we must never risk
 		 * violating it, so limit ->max_sector to one PAGE, as
 		 * a one page request is never in violation.
diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c
index cbe368fa6598..237fe3fd235c 100644
--- a/drivers/md/multipath.c
+++ b/drivers/md/multipath.c
@@ -294,7 +294,8 @@ static int multipath_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
 	for (path = first; path <= last; path++)
 		if ((p=conf->multipaths+path)->rdev == NULL) {
 			q = rdev->bdev->bd_disk->queue;
-			blk_queue_stack_limits(mddev->queue, q);
+			disk_stack_limits(mddev->gendisk, rdev->bdev,
+					  rdev->data_offset << 9);
 
 		/* as we don't honour merge_bvec_fn, we must never risk
 		 * violating it, so limit ->max_sector to one PAGE, as
@@ -463,9 +464,9 @@ static int multipath_run (mddev_t *mddev)
 
 		disk = conf->multipaths + disk_idx;
 		disk->rdev = rdev;
+		disk_stack_limits(mddev->gendisk, rdev->bdev,
+				  rdev->data_offset << 9);
 
-		blk_queue_stack_limits(mddev->queue,
-				       rdev->bdev->bd_disk->queue);
 		/* as we don't honour merge_bvec_fn, we must never risk
 		 * violating it, not that we ever expect a device with
 		 * a merge_bvec_fn to be involved in multipath */
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index ab4a489d8695..335f490dcad6 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -170,8 +170,8 @@ static int create_strip_zones(mddev_t *mddev)
 		}
 		dev[j] = rdev1;
 
-		blk_queue_stack_limits(mddev->queue,
-				       rdev1->bdev->bd_disk->queue);
+		disk_stack_limits(mddev->gendisk, rdev1->bdev,
+				  rdev1->data_offset << 9);
 		/* as we don't honour merge_bvec_fn, we must never risk
 		 * violating it, so limit ->max_sector to one PAGE, as
 		 * a one page request is never in violation.
@@ -250,6 +250,11 @@ static int create_strip_zones(mddev_t *mddev)
 		       mddev->chunk_sectors << 9);
 		goto abort;
 	}
+
+	blk_queue_io_min(mddev->queue, mddev->chunk_sectors << 9);
+	blk_queue_io_opt(mddev->queue,
+			 (mddev->chunk_sectors << 9) * mddev->raid_disks);
+
 	printk(KERN_INFO "raid0: done.\n");
 	mddev->private = conf;
 	return 0;
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 89939a7aef57..0569efba0c02 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -1123,8 +1123,8 @@ static int raid1_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
 	for (mirror = first; mirror <= last; mirror++)
 		if ( !(p=conf->mirrors+mirror)->rdev) {
 
-			blk_queue_stack_limits(mddev->queue,
-					       rdev->bdev->bd_disk->queue);
+			disk_stack_limits(mddev->gendisk, rdev->bdev,
+					  rdev->data_offset << 9);
 			/* as we don't honour merge_bvec_fn, we must never risk
 			 * violating it, so limit ->max_sector to one PAGE, as
 			 * a one page request is never in violation.
@@ -1988,9 +1988,8 @@ static int run(mddev_t *mddev)
 		disk = conf->mirrors + disk_idx;
 
 		disk->rdev = rdev;
-
-		blk_queue_stack_limits(mddev->queue,
-				       rdev->bdev->bd_disk->queue);
+		disk_stack_limits(mddev->gendisk, rdev->bdev,
+				  rdev->data_offset << 9);
 		/* as we don't honour merge_bvec_fn, we must never risk
 		 * violating it, so limit ->max_sector to one PAGE, as
 		 * a one page request is never in violation.
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index ae12ceafe10c..7298a5e5a183 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -1151,8 +1151,8 @@ static int raid10_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
 	for ( ; mirror <= last ; mirror++)
 		if ( !(p=conf->mirrors+mirror)->rdev) {
 
-			blk_queue_stack_limits(mddev->queue,
-					       rdev->bdev->bd_disk->queue);
+			disk_stack_limits(mddev->gendisk, rdev->bdev,
+					  rdev->data_offset << 9);
 			/* as we don't honour merge_bvec_fn, we must never risk
 			 * violating it, so limit ->max_sector to one PAGE, as
 			 * a one page request is never in violation.
@@ -2044,7 +2044,7 @@ raid10_size(mddev_t *mddev, sector_t sectors, int raid_disks)
 static int run(mddev_t *mddev)
 {
 	conf_t *conf;
-	int i, disk_idx;
+	int i, disk_idx, chunk_size;
 	mirror_info_t *disk;
 	mdk_rdev_t *rdev;
 	int nc, fc, fo;
@@ -2130,6 +2130,14 @@ static int run(mddev_t *mddev)
 	spin_lock_init(&conf->device_lock);
 	mddev->queue->queue_lock = &conf->device_lock;
 
+	chunk_size = mddev->chunk_sectors << 9;
+	blk_queue_io_min(mddev->queue, chunk_size);
+	if (conf->raid_disks % conf->near_copies)
+		blk_queue_io_opt(mddev->queue, chunk_size * conf->raid_disks);
+	else
+		blk_queue_io_opt(mddev->queue, chunk_size *
+				 (conf->raid_disks / conf->near_copies));
+
 	list_for_each_entry(rdev, &mddev->disks, same_set) {
 		disk_idx = rdev->raid_disk;
 		if (disk_idx >= mddev->raid_disks
@@ -2138,9 +2146,8 @@ static int run(mddev_t *mddev)
 		disk = conf->mirrors + disk_idx;
 
 		disk->rdev = rdev;
-
-		blk_queue_stack_limits(mddev->queue,
-				       rdev->bdev->bd_disk->queue);
+		disk_stack_limits(mddev->gendisk, rdev->bdev,
+				  rdev->data_offset << 9);
 		/* as we don't honour merge_bvec_fn, we must never risk
 		 * violating it, so limit ->max_sector to one PAGE, as
 		 * a one page request is never in violation.
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index f9f991e6e138..92ef9b6abfc7 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -4452,7 +4452,7 @@ static raid5_conf_t *setup_conf(mddev_t *mddev)
 static int run(mddev_t *mddev)
 {
 	raid5_conf_t *conf;
-	int working_disks = 0;
+	int working_disks = 0, chunk_size;
 	mdk_rdev_t *rdev;
 
 	if (mddev->recovery_cp != MaxSector)
@@ -4607,6 +4607,14 @@ static int run(mddev_t *mddev)
 	md_set_array_sectors(mddev, raid5_size(mddev, 0, 0));
 
 	blk_queue_merge_bvec(mddev->queue, raid5_mergeable_bvec);
+	chunk_size = mddev->chunk_sectors << 9;
+	blk_queue_io_min(mddev->queue, chunk_size);
+	blk_queue_io_opt(mddev->queue, chunk_size *
+			 (conf->raid_disks - conf->max_degraded));
+
+	list_for_each_entry(rdev, &mddev->disks, same_set)
+		disk_stack_limits(mddev->gendisk, rdev->bdev,
+				  rdev->data_offset << 9);
 
 	return 0;
 abort:
-- 
cgit v1.2.3-59-g8ed1b


From b8d966efd9a46a9a35beac50cbff6e30565125ef Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Wed, 1 Jul 2009 11:14:04 +1000
Subject: md: avoid dereferencing NULL pointer when accessing suspend_* sysfs
 attributes.

If we try to modify one of the md/ sysfs files
  suspend_lo or suspend_hi
when the array is not active, we dereference a NULL.
Protect against that.

Cc: stable@kernel.org
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/md.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 09be637d52cb..2166af8a7654 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -3573,7 +3573,8 @@ suspend_lo_store(mddev_t *mddev, const char *buf, size_t len)
 	char *e;
 	unsigned long long new = simple_strtoull(buf, &e, 10);
 
-	if (mddev->pers->quiesce == NULL)
+	if (mddev->pers == NULL || 
+	    mddev->pers->quiesce == NULL)
 		return -EINVAL;
 	if (buf == e || (*e && *e != '\n'))
 		return -EINVAL;
@@ -3601,7 +3602,8 @@ suspend_hi_store(mddev_t *mddev, const char *buf, size_t len)
 	char *e;
 	unsigned long long new = simple_strtoull(buf, &e, 10);
 
-	if (mddev->pers->quiesce == NULL)
+	if (mddev->pers == NULL ||
+	    mddev->pers->quiesce == NULL)
 		return -EINVAL;
 	if (buf == e || (*e && *e != '\n'))
 		return -EINVAL;
-- 
cgit v1.2.3-59-g8ed1b


From 133890103b9de08904f909995973e4b5c08a780e Mon Sep 17 00:00:00 2001
From: Davide Libenzi <davidel@xmailserver.org>
Date: Tue, 30 Jun 2009 11:41:11 -0700
Subject: eventfd: revised interface and cleanups

Change the eventfd interface to de-couple the eventfd memory context, from
the file pointer instance.

Without such change, there is no clean way to racely free handle the
POLLHUP event sent when the last instance of the file* goes away.  Also,
now the internal eventfd APIs are using the eventfd context instead of the
file*.

This patch is required by KVM's IRQfd code, which is still under
development.

Signed-off-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Gregory Haskins <ghaskins@novell.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Avi Kivity <avi@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/lguest/lg.h          |   2 +-
 drivers/lguest/lguest_user.c |   4 +-
 fs/aio.c                     |  24 +++------
 fs/eventfd.c                 | 122 ++++++++++++++++++++++++++++++++++++++-----
 include/linux/aio.h          |   4 +-
 include/linux/eventfd.h      |  35 ++++++++++---
 6 files changed, 149 insertions(+), 42 deletions(-)

diff --git a/drivers/lguest/lg.h b/drivers/lguest/lg.h
index d4e8979735cb..9c3138265f8e 100644
--- a/drivers/lguest/lg.h
+++ b/drivers/lguest/lg.h
@@ -82,7 +82,7 @@ struct lg_cpu {
 
 struct lg_eventfd {
 	unsigned long addr;
-	struct file *event;
+	struct eventfd_ctx *event;
 };
 
 struct lg_eventfd_map {
diff --git a/drivers/lguest/lguest_user.c b/drivers/lguest/lguest_user.c
index 32e297121058..9f9a2953b383 100644
--- a/drivers/lguest/lguest_user.c
+++ b/drivers/lguest/lguest_user.c
@@ -50,7 +50,7 @@ static int add_eventfd(struct lguest *lg, unsigned long addr, int fd)
 
 	/* Now append new entry. */
 	new->map[new->num].addr = addr;
-	new->map[new->num].event = eventfd_fget(fd);
+	new->map[new->num].event = eventfd_ctx_fdget(fd);
 	if (IS_ERR(new->map[new->num].event)) {
 		kfree(new);
 		return PTR_ERR(new->map[new->num].event);
@@ -357,7 +357,7 @@ static int close(struct inode *inode, struct file *file)
 
 	/* Release any eventfds they registered. */
 	for (i = 0; i < lg->eventfds->num; i++)
-		fput(lg->eventfds->map[i].event);
+		eventfd_ctx_put(lg->eventfds->map[i].event);
 	kfree(lg->eventfds);
 
 	/* If lg->dead doesn't contain an error code it will be NULL or a
diff --git a/fs/aio.c b/fs/aio.c
index 76da12537956..d065b2c3273e 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -485,6 +485,8 @@ static inline void really_put_req(struct kioctx *ctx, struct kiocb *req)
 {
 	assert_spin_locked(&ctx->ctx_lock);
 
+	if (req->ki_eventfd != NULL)
+		eventfd_ctx_put(req->ki_eventfd);
 	if (req->ki_dtor)
 		req->ki_dtor(req);
 	if (req->ki_iovec != &req->ki_inline_vec)
@@ -509,8 +511,6 @@ static void aio_fput_routine(struct work_struct *data)
 		/* Complete the fput(s) */
 		if (req->ki_filp != NULL)
 			__fput(req->ki_filp);
-		if (req->ki_eventfd != NULL)
-			__fput(req->ki_eventfd);
 
 		/* Link the iocb into the context's free list */
 		spin_lock_irq(&ctx->ctx_lock);
@@ -528,8 +528,6 @@ static void aio_fput_routine(struct work_struct *data)
  */
 static int __aio_put_req(struct kioctx *ctx, struct kiocb *req)
 {
-	int schedule_putreq = 0;
-
 	dprintk(KERN_DEBUG "aio_put(%p): f_count=%ld\n",
 		req, atomic_long_read(&req->ki_filp->f_count));
 
@@ -549,24 +547,16 @@ static int __aio_put_req(struct kioctx *ctx, struct kiocb *req)
 	 * we would not be holding the last reference to the file*, so
 	 * this function will be executed w/out any aio kthread wakeup.
 	 */
-	if (unlikely(atomic_long_dec_and_test(&req->ki_filp->f_count)))
-		schedule_putreq++;
-	else
-		req->ki_filp = NULL;
-	if (req->ki_eventfd != NULL) {
-		if (unlikely(atomic_long_dec_and_test(&req->ki_eventfd->f_count)))
-			schedule_putreq++;
-		else
-			req->ki_eventfd = NULL;
-	}
-	if (unlikely(schedule_putreq)) {
+	if (unlikely(atomic_long_dec_and_test(&req->ki_filp->f_count))) {
 		get_ioctx(ctx);
 		spin_lock(&fput_lock);
 		list_add(&req->ki_list, &fput_head);
 		spin_unlock(&fput_lock);
 		queue_work(aio_wq, &fput_work);
-	} else
+	} else {
+		req->ki_filp = NULL;
 		really_put_req(ctx, req);
+	}
 	return 1;
 }
 
@@ -1622,7 +1612,7 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
 		 * an eventfd() fd, and will be signaled for each completed
 		 * event using the eventfd_signal() function.
 		 */
-		req->ki_eventfd = eventfd_fget((int) iocb->aio_resfd);
+		req->ki_eventfd = eventfd_ctx_fdget((int) iocb->aio_resfd);
 		if (IS_ERR(req->ki_eventfd)) {
 			ret = PTR_ERR(req->ki_eventfd);
 			req->ki_eventfd = NULL;
diff --git a/fs/eventfd.c b/fs/eventfd.c
index 3f0e1974abdc..31d12de83a2a 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -14,35 +14,44 @@
 #include <linux/list.h>
 #include <linux/spinlock.h>
 #include <linux/anon_inodes.h>
-#include <linux/eventfd.h>
 #include <linux/syscalls.h>
 #include <linux/module.h>
+#include <linux/kref.h>
+#include <linux/eventfd.h>
 
 struct eventfd_ctx {
+	struct kref kref;
 	wait_queue_head_t wqh;
 	/*
 	 * Every time that a write(2) is performed on an eventfd, the
 	 * value of the __u64 being written is added to "count" and a
 	 * wakeup is performed on "wqh". A read(2) will return the "count"
 	 * value to userspace, and will reset "count" to zero. The kernel
-	 * size eventfd_signal() also, adds to the "count" counter and
+	 * side eventfd_signal() also, adds to the "count" counter and
 	 * issue a wakeup.
 	 */
 	__u64 count;
 	unsigned int flags;
 };
 
-/*
- * Adds "n" to the eventfd counter "count". Returns "n" in case of
- * success, or a value lower then "n" in case of coutner overflow.
- * This function is supposed to be called by the kernel in paths
- * that do not allow sleeping. In this function we allow the counter
- * to reach the ULLONG_MAX value, and we signal this as overflow
- * condition by returining a POLLERR to poll(2).
+/**
+ * eventfd_signal - Adds @n to the eventfd counter.
+ * @ctx: [in] Pointer to the eventfd context.
+ * @n: [in] Value of the counter to be added to the eventfd internal counter.
+ *          The value cannot be negative.
+ *
+ * This function is supposed to be called by the kernel in paths that do not
+ * allow sleeping. In this function we allow the counter to reach the ULLONG_MAX
+ * value, and we signal this as overflow condition by returining a POLLERR
+ * to poll(2).
+ *
+ * Returns @n in case of success, a non-negative number lower than @n in case
+ * of overflow, or the following error codes:
+ *
+ * -EINVAL    : The value of @n is negative.
  */
-int eventfd_signal(struct file *file, int n)
+int eventfd_signal(struct eventfd_ctx *ctx, int n)
 {
-	struct eventfd_ctx *ctx = file->private_data;
 	unsigned long flags;
 
 	if (n < 0)
@@ -59,9 +68,45 @@ int eventfd_signal(struct file *file, int n)
 }
 EXPORT_SYMBOL_GPL(eventfd_signal);
 
+static void eventfd_free(struct kref *kref)
+{
+	struct eventfd_ctx *ctx = container_of(kref, struct eventfd_ctx, kref);
+
+	kfree(ctx);
+}
+
+/**
+ * eventfd_ctx_get - Acquires a reference to the internal eventfd context.
+ * @ctx: [in] Pointer to the eventfd context.
+ *
+ * Returns: In case of success, returns a pointer to the eventfd context.
+ */
+struct eventfd_ctx *eventfd_ctx_get(struct eventfd_ctx *ctx)
+{
+	kref_get(&ctx->kref);
+	return ctx;
+}
+EXPORT_SYMBOL_GPL(eventfd_ctx_get);
+
+/**
+ * eventfd_ctx_put - Releases a reference to the internal eventfd context.
+ * @ctx: [in] Pointer to eventfd context.
+ *
+ * The eventfd context reference must have been previously acquired either
+ * with eventfd_ctx_get() or eventfd_ctx_fdget()).
+ */
+void eventfd_ctx_put(struct eventfd_ctx *ctx)
+{
+	kref_put(&ctx->kref, eventfd_free);
+}
+EXPORT_SYMBOL_GPL(eventfd_ctx_put);
+
 static int eventfd_release(struct inode *inode, struct file *file)
 {
-	kfree(file->private_data);
+	struct eventfd_ctx *ctx = file->private_data;
+
+	wake_up_poll(&ctx->wqh, POLLHUP);
+	eventfd_ctx_put(ctx);
 	return 0;
 }
 
@@ -185,6 +230,16 @@ static const struct file_operations eventfd_fops = {
 	.write		= eventfd_write,
 };
 
+/**
+ * eventfd_fget - Acquire a reference of an eventfd file descriptor.
+ * @fd: [in] Eventfd file descriptor.
+ *
+ * Returns a pointer to the eventfd file structure in case of success, or the
+ * following error pointer:
+ *
+ * -EBADF    : Invalid @fd file descriptor.
+ * -EINVAL   : The @fd file descriptor is not an eventfd file.
+ */
 struct file *eventfd_fget(int fd)
 {
 	struct file *file;
@@ -201,6 +256,48 @@ struct file *eventfd_fget(int fd)
 }
 EXPORT_SYMBOL_GPL(eventfd_fget);
 
+/**
+ * eventfd_ctx_fdget - Acquires a reference to the internal eventfd context.
+ * @fd: [in] Eventfd file descriptor.
+ *
+ * Returns a pointer to the internal eventfd context, otherwise the error
+ * pointers returned by the following functions:
+ *
+ * eventfd_fget
+ */
+struct eventfd_ctx *eventfd_ctx_fdget(int fd)
+{
+	struct file *file;
+	struct eventfd_ctx *ctx;
+
+	file = eventfd_fget(fd);
+	if (IS_ERR(file))
+		return (struct eventfd_ctx *) file;
+	ctx = eventfd_ctx_get(file->private_data);
+	fput(file);
+
+	return ctx;
+}
+EXPORT_SYMBOL_GPL(eventfd_ctx_fdget);
+
+/**
+ * eventfd_ctx_fileget - Acquires a reference to the internal eventfd context.
+ * @file: [in] Eventfd file pointer.
+ *
+ * Returns a pointer to the internal eventfd context, otherwise the error
+ * pointer:
+ *
+ * -EINVAL   : The @fd file descriptor is not an eventfd file.
+ */
+struct eventfd_ctx *eventfd_ctx_fileget(struct file *file)
+{
+	if (file->f_op != &eventfd_fops)
+		return ERR_PTR(-EINVAL);
+
+	return eventfd_ctx_get(file->private_data);
+}
+EXPORT_SYMBOL_GPL(eventfd_ctx_fileget);
+
 SYSCALL_DEFINE2(eventfd2, unsigned int, count, int, flags)
 {
 	int fd;
@@ -217,6 +314,7 @@ SYSCALL_DEFINE2(eventfd2, unsigned int, count, int, flags)
 	if (!ctx)
 		return -ENOMEM;
 
+	kref_init(&ctx->kref);
 	init_waitqueue_head(&ctx->wqh);
 	ctx->count = count;
 	ctx->flags = flags;
diff --git a/include/linux/aio.h b/include/linux/aio.h
index b16a957030f8..47f7d932a01d 100644
--- a/include/linux/aio.h
+++ b/include/linux/aio.h
@@ -121,9 +121,9 @@ struct kiocb {
 
 	/*
 	 * If the aio_resfd field of the userspace iocb is not zero,
-	 * this is the underlying file* to deliver event to.
+	 * this is the underlying eventfd context to deliver events to.
 	 */
-	struct file		*ki_eventfd;
+	struct eventfd_ctx	*ki_eventfd;
 };
 
 #define is_sync_kiocb(iocb)	((iocb)->ki_key == KIOCB_SYNC_KEY)
diff --git a/include/linux/eventfd.h b/include/linux/eventfd.h
index f45a8ae5f828..3b85ba6479f4 100644
--- a/include/linux/eventfd.h
+++ b/include/linux/eventfd.h
@@ -8,10 +8,8 @@
 #ifndef _LINUX_EVENTFD_H
 #define _LINUX_EVENTFD_H
 
-#ifdef CONFIG_EVENTFD
-
-/* For O_CLOEXEC and O_NONBLOCK */
 #include <linux/fcntl.h>
+#include <linux/file.h>
 
 /*
  * CAREFUL: Check include/asm-generic/fcntl.h when defining
@@ -27,16 +25,37 @@
 #define EFD_SHARED_FCNTL_FLAGS (O_CLOEXEC | O_NONBLOCK)
 #define EFD_FLAGS_SET (EFD_SHARED_FCNTL_FLAGS | EFD_SEMAPHORE)
 
+#ifdef CONFIG_EVENTFD
+
+struct eventfd_ctx *eventfd_ctx_get(struct eventfd_ctx *ctx);
+void eventfd_ctx_put(struct eventfd_ctx *ctx);
 struct file *eventfd_fget(int fd);
-int eventfd_signal(struct file *file, int n);
+struct eventfd_ctx *eventfd_ctx_fdget(int fd);
+struct eventfd_ctx *eventfd_ctx_fileget(struct file *file);
+int eventfd_signal(struct eventfd_ctx *ctx, int n);
 
 #else /* CONFIG_EVENTFD */
 
-#define eventfd_fget(fd) ERR_PTR(-ENOSYS)
-static inline int eventfd_signal(struct file *file, int n)
-{ return 0; }
+/*
+ * Ugly ugly ugly error layer to support modules that uses eventfd but
+ * pretend to work in !CONFIG_EVENTFD configurations. Namely, AIO.
+ */
+static inline struct eventfd_ctx *eventfd_ctx_fdget(int fd)
+{
+	return ERR_PTR(-ENOSYS);
+}
+
+static inline int eventfd_signal(struct eventfd_ctx *ctx, int n)
+{
+	return -ENOSYS;
+}
+
+static inline void eventfd_ctx_put(struct eventfd_ctx *ctx)
+{
+
+}
 
-#endif /* CONFIG_EVENTFD */
+#endif
 
 #endif /* _LINUX_EVENTFD_H */
 
-- 
cgit v1.2.3-59-g8ed1b


From 2a2325e6e8a3782795fb520220c36fd805775972 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 30 Jun 2009 11:41:13 -0700
Subject: gcov: fix __ctors_start alignment

The ctors section for each object file is eight byte aligned (on 64 bit).
However the __ctors_start symbol starts at an arbitrary address dependent
on the size of the previous sections.

Therefore the linker may add some zeroes after __ctors_start to make sure
the ctors contents are properly aligned.  However the extra zeroes at the
beginning aren't expected by the code.  When walking the functions
pointers contained in there and extra zeroes are added this may result in
random jumps.  So make sure that the __ctors_start symbol is always
aligned as well.

Fixes this crash on an allyesconfig on s390:

[    0.582482] Kernel BUG at 0000000000000012 [verbose debug info unavailable]
[    0.582489] illegal operation: 0001 [#1] SMP DEBUG_PAGEALLOC
[    0.582496] Modules linked in:
[    0.582501] CPU: 0 Tainted: G        W  2.6.31-rc1-dirty #273
[    0.582506] Process swapper (pid: 1, task: 000000003f218000, ksp: 000000003f2238e8)
[    0.582510] Krnl PSW : 0704200180000000 0000000000000012 (0x12)
[    0.582518]            R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:0 CC:2 PM:0 EA:3
[    0.582524] Krnl GPRS: 0000000000036727 0000000000000010 0000000000000001 0000000000000001
[    0.582529]            00000000001dfefa 0000000000000000 0000000000000000 0000000000000040
[    0.582534]            0000000001fff0f0 0000000001790628 0000000002296048 0000000002296048
[    0.582540]            00000000020c438e 0000000001786000 0000000002014a66 000000003f223e60
[    0.582553] Krnl Code:>0000000000000012: 0000                unknown
[    0.582559]            0000000000000014: 0000                unknown
[    0.582564]            0000000000000016: 0000                unknown
[    0.582570]            0000000000000018: 0000                unknown
[    0.582575]            000000000000001a: 0000                unknown
[    0.582580]            000000000000001c: 0000                unknown
[    0.582585]            000000000000001e: 0000                unknown
[    0.582591]            0000000000000020: 0000                unknown
[    0.582596] Call Trace:
[    0.582599] ([<0000000002014a46>] kernel_init+0x622/0x7a0)
[    0.582607]  [<0000000000113e22>] kernel_thread_starter+0x6/0xc
[    0.582615]  [<0000000000113e1c>] kernel_thread_starter+0x0/0xc
[    0.582621] INFO: lockdep is turned off.
[    0.582624] Last Breaking-Event-Address:
[    0.582627]  [<0000000002014a64>] kernel_init+0x640/0x7a0

Cc: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-generic/vmlinux.lds.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 92b73b6140ff..dccdbed05848 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -441,7 +441,8 @@
 	}
 
 #ifdef CONFIG_CONSTRUCTORS
-#define KERNEL_CTORS()	VMLINUX_SYMBOL(__ctors_start) = .; \
+#define KERNEL_CTORS()	. = ALIGN(8);			   \
+			VMLINUX_SYMBOL(__ctors_start) = .; \
 			*(.ctors)			   \
 			VMLINUX_SYMBOL(__ctors_end) = .;
 #else
-- 
cgit v1.2.3-59-g8ed1b


From c15e504bd008aedfcd2219051055b66bffdb6148 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Tue, 30 Jun 2009 11:41:13 -0700
Subject: MAINTAINERS: update EDAC-I82975X

As per Ranganathan's request.

Signed-off-by: Joe Perches <joe@perches.com>
Cc: Ranganathan Desikan <ravi@jetztechnologies.com>
Cc: Arvind R. <arvind@jetztechnologies.com>
Cc: Doug Thompson <norsk5@yahoo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 MAINTAINERS | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 05a9a563042f..92fe0796cd1b 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2130,9 +2130,9 @@ F:	drivers/edac/i5400_edac.c
 
 EDAC-I82975X
 P:	Ranganathan Desikan
-M:	rdesikan@jetzbroadband.com
+M:	ravi@jetztechnologies.com
 P:	Arvind R.
-M:	arvind@acarlab.com
+M:	arvind@jetztechnologies.com
 L:	bluesmoke-devel@lists.sourceforge.net (moderated for non-subscribers)
 W:	bluesmoke.sourceforge.net
 S:	Maintained
-- 
cgit v1.2.3-59-g8ed1b


From 15e3252464432a29c5461325cb5243471bd2a219 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 30 Jun 2009 11:41:15 -0700
Subject: fbdev: work around old compiler bug

When building with a 4.1.x compiler on powerpc64 (at least) we get this
error:

drivers/video/logo/logo_linux_mono.c:81: error: logo_linux_mono causes a section type conflict

This was introduced by commit ae52bb2384f721562f15f719de1acb8e934733cb
("fbdev: move logo externs to header file").  This is a partial revert of
that commit sufficient to not hit the compiler bug.

Also convert _clut arrays from __initconst to __initdata.

Sam said:

  Al analysed this some time ago.  When we say something is const then
  _sometimes_ gcc annotate the section as const(?) - sometimes not.  So if
  we have two variables/functions annotated __*const and gcc decides to
  annotate the section const only in one case we get a section type
  conflict.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Cc: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/pnmtologo.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/scripts/pnmtologo.c b/scripts/pnmtologo.c
index 64f5ddb09ea6..5c113123ed9f 100644
--- a/scripts/pnmtologo.c
+++ b/scripts/pnmtologo.c
@@ -237,7 +237,7 @@ static void write_header(void)
     fprintf(out, " *  Linux logo %s\n", logoname);
     fputs(" */\n\n", out);
     fputs("#include <linux/linux_logo.h>\n\n", out);
-    fprintf(out, "static const unsigned char %s_data[] __initconst = {\n",
+    fprintf(out, "static unsigned char %s_data[] __initdata = {\n",
 	    logoname);
 }
 
@@ -374,7 +374,7 @@ static void write_logo_clut224(void)
     fputs("\n};\n\n", out);
 
     /* write logo clut */
-    fprintf(out, "static const unsigned char %s_clut[] __initconst = {\n",
+    fprintf(out, "static unsigned char %s_clut[] __initdata = {\n",
 	    logoname);
     write_hex_cnt = 0;
     for (i = 0; i < logo_clutsize; i++) {
-- 
cgit v1.2.3-59-g8ed1b


From b01e8dc34379f4ba2f454390e340a025edbaaa7e Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Tue, 30 Jun 2009 11:41:18 -0700
Subject: alpha: fix percpu build breakage

alpha percpu access requires custom SHIFT_PERCPU_PTR() definition for
modules to work around addressing range limitation.  This is done via
generating inline assembly using C preprocessing which forces the
assembler to generate external reference.  This happens behind the
compiler's back and makes the compiler think that static percpu variables
in modules are unused.

This used to be worked around by using __unused attribute for percpu
variables which prevent the compiler from omitting the variable; however,
recent declare/definition attribute unification change broke this as
__used can't be used for declaration.  Also, in the process,
PER_CPU_ATTRIBUTES definition in alpha percpu.h got broken.

This patch adds PER_CPU_DEF_ATTRIBUTES which is only used for definitions
and make alpha use it to add __used for percpu variables in modules.  This
also fixes the PER_CPU_ATTRIBUTES double definition bug.

Signed-off-by: Tejun Heo <tj@kernel.org>
Tested-by: maximilian attems <max@stro.at>
Acked-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Richard Henderson <rth@twiddle.net>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/alpha/include/asm/percpu.h | 6 +++---
 include/asm-generic/percpu.h    | 4 ++++
 include/linux/percpu-defs.h     | 3 ++-
 3 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/arch/alpha/include/asm/percpu.h b/arch/alpha/include/asm/percpu.h
index 06c5c7a4afd3..b663f1f10b6a 100644
--- a/arch/alpha/include/asm/percpu.h
+++ b/arch/alpha/include/asm/percpu.h
@@ -30,7 +30,7 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
 
 #ifndef MODULE
 #define SHIFT_PERCPU_PTR(var, offset) RELOC_HIDE(&per_cpu_var(var), (offset))
-#define PER_CPU_ATTRIBUTES
+#define PER_CPU_DEF_ATTRIBUTES
 #else
 /*
  * To calculate addresses of locally defined variables, GCC uses 32-bit
@@ -49,7 +49,7 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
 		: "=&r"(__ptr), "=&r"(tmp_gp));		\
 	(typeof(&per_cpu_var(var)))(__ptr + (offset)); })
 
-#define PER_CPU_ATTRIBUTES	__used
+#define PER_CPU_DEF_ATTRIBUTES	__used
 
 #endif /* MODULE */
 
@@ -71,7 +71,7 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
 #define __get_cpu_var(var)		per_cpu_var(var)
 #define __raw_get_cpu_var(var)		per_cpu_var(var)
 
-#define PER_CPU_ATTRIBUTES
+#define PER_CPU_DEF_ATTRIBUTES
 
 #endif /* SMP */
 
diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index d7d50d7ee51e..aa00800adacc 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -97,4 +97,8 @@ extern void setup_per_cpu_areas(void);
 #define PER_CPU_ATTRIBUTES
 #endif
 
+#ifndef PER_CPU_DEF_ATTRIBUTES
+#define PER_CPU_DEF_ATTRIBUTES
+#endif
+
 #endif /* _ASM_GENERIC_PERCPU_H_ */
diff --git a/include/linux/percpu-defs.h b/include/linux/percpu-defs.h
index 8f921d74f49f..68438e18fff4 100644
--- a/include/linux/percpu-defs.h
+++ b/include/linux/percpu-defs.h
@@ -24,7 +24,8 @@
 
 #define DEFINE_PER_CPU_SECTION(type, name, section)			\
 	__attribute__((__section__(PER_CPU_BASE_SECTION section)))	\
-	PER_CPU_ATTRIBUTES __typeof__(type) per_cpu__##name
+	PER_CPU_ATTRIBUTES PER_CPU_DEF_ATTRIBUTES			\
+	__typeof__(type) per_cpu__##name
 
 /*
  * Variant on the per-CPU variable declaration/definition theme used for
-- 
cgit v1.2.3-59-g8ed1b


From 972c71a3183ab41c0b1a9e50842be7e3e980954f Mon Sep 17 00:00:00 2001
From: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Date: Tue, 30 Jun 2009 11:41:20 -0700
Subject: gcov: fix documentation

Commonly available versions of cp and tar don't work well with special
files created using seq_file.  Mention this problem in the gcov
documentation and update the helper script example to work around these
problems.

Signed-off-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/gcov.txt | 25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)

diff --git a/Documentation/gcov.txt b/Documentation/gcov.txt
index e716aadb3a33..40ec63352760 100644
--- a/Documentation/gcov.txt
+++ b/Documentation/gcov.txt
@@ -188,13 +188,18 @@ Solution: Exclude affected source files from profiling by specifying
           GCOV_PROFILE := n or GCOV_PROFILE_basename.o := n in the
           corresponding Makefile.
 
+Problem:  Files copied from sysfs appear empty or incomplete.
+Cause:    Due to the way seq_file works, some tools such as cp or tar
+          may not correctly copy files from sysfs.
+Solution: Use 'cat' to read .gcda files and 'cp -d' to copy links.
+          Alternatively use the mechanism shown in Appendix B.
+
 
 Appendix A: gather_on_build.sh
 ==============================
 
 Sample script to gather coverage meta files on the build machine
 (see 6a):
-
 #!/bin/bash
 
 KSRC=$1
@@ -226,7 +231,7 @@ Appendix B: gather_on_test.sh
 Sample script to gather coverage data files on the test machine
 (see 6b):
 
-#!/bin/bash
+#!/bin/bash -e
 
 DEST=$1
 GCDA=/sys/kernel/debug/gcov
@@ -236,11 +241,13 @@ if [ -z "$DEST" ] ; then
   exit 1
 fi
 
-find $GCDA -name '*.gcno' -o -name '*.gcda' | tar cfz $DEST -T -
+TEMPDIR=$(mktemp -d)
+echo Collecting data..
+find $GCDA -type d -exec mkdir -p $TEMPDIR/\{\} \;
+find $GCDA -name '*.gcda' -exec sh -c 'cat < $0 > '$TEMPDIR'/$0' {} \;
+find $GCDA -name '*.gcno' -exec sh -c 'cp -d $0 '$TEMPDIR'/$0' {} \;
+tar czf $DEST -C $TEMPDIR sys
+rm -rf $TEMPDIR
 
-if [ $? -eq 0 ] ; then
-  echo "$DEST successfully created, copy to build system and unpack with:"
-  echo "  tar xfz $DEST"
-else
-  echo "Could not create file $DEST"
-fi
+echo "$DEST successfully created, copy to build system and unpack with:"
+echo "  tar xfz $DEST"
-- 
cgit v1.2.3-59-g8ed1b


From c4285b47b0514e2103584ee829246f813e7ae323 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Tue, 30 Jun 2009 11:41:21 -0700
Subject: parport/serial: add support for NetMos 9901 Multi-IO card

Add support for the PCI-Express NetMos 9901 Multi-IO card.

0001:06:00.0 Serial controller [0700]: NetMos Technology Device [9710:9901] (prog-if 02 [16550])
        Subsystem: Device [a000:1000]
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0, Cache Line Size: 64 bytes
        Interrupt: pin A routed to IRQ 65
        Region 0: I/O ports at 0030 [size=8]
        Region 1: Memory at 80105000 (32-bit, non-prefetchable) [size=4K]
        Region 4: Memory at 80104000 (32-bit, non-prefetchable) [size=4K]
        Capabilities: <access denied>
        Kernel driver in use: serial
        Kernel modules: 8250_pci

0001:06:00.1 Serial controller [0700]: NetMos Technology Device [9710:9901] (prog-if 02 [16550])
        Subsystem: Device [a000:1000]
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0, Cache Line Size: 64 bytes
        Interrupt: pin B routed to IRQ 65
        Region 0: I/O ports at 0020 [size=8]
        Region 1: Memory at 80103000 (32-bit, non-prefetchable) [size=4K]
        Region 4: Memory at 80102000 (32-bit, non-prefetchable) [size=4K]
        Capabilities: <access denied>
        Kernel driver in use: serial
        Kernel modules: 8250_pci

0001:06:00.2 Parallel controller [0701]: NetMos Technology Device [9710:9901] (prog-if 03 [IEEE1284])
        Subsystem: Device [a000:2000]
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0, Cache Line Size: 64 bytes
        Interrupt: pin C routed to IRQ 65
        Region 0: I/O ports at 0010 [size=8]
        Region 1: I/O ports at <unassigned>
        Region 2: Memory at 80101000 (32-bit, non-prefetchable) [size=4K]
        Region 4: Memory at 80100000 (32-bit, non-prefetchable) [size=4K]
        Capabilities: <access denied>
        Kernel driver in use: parport_pc
        Kernel modules: parport_pc

[   16.760181] PCI parallel port detected: 416c:0100, I/O at 0x812010(0x0), IRQ 65
[   16.760225] parport0: PC-style at 0x812010, irq 65 [PCSPP,TRISTATE,EPP]
[   16.851842] serial 0001:06:00.0: enabling device (0004 -> 0007)
[   16.883776] 0001:06:00.0: ttyS0 at I/O 0x812030 (irq = 65) is a ST16650V2
[   16.893832] serial 0001:06:00.1: enabling device (0004 -> 0007)
[   16.926537] 0001:06:00.1: ttyS1 at I/O 0x812020 (irq = 65) is a ST16650V2

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/parport/parport_pc.c | 5 ++++-
 drivers/serial/8250_pci.c    | 6 ++++++
 include/linux/pci_ids.h      | 1 +
 3 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/parport/parport_pc.c b/drivers/parport/parport_pc.c
index 1032d5fdbd42..2597145a066e 100644
--- a/drivers/parport/parport_pc.c
+++ b/drivers/parport/parport_pc.c
@@ -2907,6 +2907,7 @@ enum parport_pc_pci_cards {
 	netmos_9755,
 	netmos_9805,
 	netmos_9815,
+	netmos_9901,
 	quatech_sppxp100,
 };
 
@@ -2987,7 +2988,7 @@ static struct parport_pc_pci {
 	/* netmos_9755 */               { 2, { { 0, 1 }, { 2, 3 },} },
 	/* netmos_9805 */               { 1, { { 0, -1 }, } },
 	/* netmos_9815 */               { 2, { { 0, -1 }, { 2, -1 }, } },
-
+	/* netmos_9901 */               { 1, { { 0, -1 }, } },
 	/* quatech_sppxp100 */		{ 1, { { 0, 1 }, } },
 };
 
@@ -3089,6 +3090,8 @@ static const struct pci_device_id parport_pc_pci_tbl[] = {
 	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, netmos_9805 },
 	{ PCI_VENDOR_ID_NETMOS, PCI_DEVICE_ID_NETMOS_9815,
 	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, netmos_9815 },
+	{ PCI_VENDOR_ID_NETMOS, PCI_DEVICE_ID_NETMOS_9901,
+	  0xA000, 0x2000, 0, 0, netmos_9901 },
 	/* Quatech SPPXP-100 Parallel port PCI ExpressCard */
 	{ PCI_VENDOR_ID_QUATECH, PCI_DEVICE_ID_QUATECH_SPPXP_100,
 	  PCI_ANY_ID, PCI_ANY_ID, 0, 0, quatech_sppxp100 },
diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c
index a07015d646dd..6160e03f410c 100644
--- a/drivers/serial/8250_pci.c
+++ b/drivers/serial/8250_pci.c
@@ -759,6 +759,8 @@ static int pci_netmos_init(struct pci_dev *dev)
 	/* subdevice 0x00PS means <P> parallel, <S> serial */
 	unsigned int num_serial = dev->subsystem_device & 0xf;
 
+	if (dev->device == PCI_DEVICE_ID_NETMOS_9901)
+		return 0;
 	if (dev->subsystem_vendor == PCI_VENDOR_ID_IBM &&
 			dev->subsystem_device == 0x0299)
 		return 0;
@@ -3557,6 +3559,10 @@ static struct pci_device_id serial_pci_tbl[] = {
 		PCI_VENDOR_ID_IBM, 0x0299,
 		0, 0, pbn_b0_bt_2_115200 },
 
+	{	PCI_VENDOR_ID_NETMOS, PCI_DEVICE_ID_NETMOS_9901,
+		0xA000, 0x1000,
+		0, 0, pbn_b0_1_115200 },
+
 	/*
 	 * These entries match devices with class COMMUNICATION_SERIAL,
 	 * COMMUNICATION_MODEM or COMMUNICATION_MULTISERIAL
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index a3b000365795..73b46b6b904f 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2645,6 +2645,7 @@
 #define PCI_DEVICE_ID_NETMOS_9835	0x9835
 #define PCI_DEVICE_ID_NETMOS_9845	0x9845
 #define PCI_DEVICE_ID_NETMOS_9855	0x9855
+#define PCI_DEVICE_ID_NETMOS_9901	0x9901
 
 #define PCI_VENDOR_ID_3COM_2		0xa727
 
-- 
cgit v1.2.3-59-g8ed1b


From b1cfebc9231a69d46d66982a2c856ba41ef6d6b9 Mon Sep 17 00:00:00 2001
From: Yang Shi <yang.shi@windriver.com>
Date: Tue, 30 Jun 2009 11:41:22 -0700
Subject: edac: add DDR3 memory type for MPC85xx EDAC

Since some new MPC85xx SOCs support DDR3 memory now, so add DDR3 memory
type for MPC85xx EDAC.

Signed-off-by: Yang Shi <yang.shi@windriver.com>
Cc: Doug Thompson <norsk5@yahoo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/edac_core.h     | 4 ++++
 drivers/edac/edac_mc_sysfs.c | 4 +++-
 drivers/edac/mpc85xx_edac.c  | 6 ++++++
 drivers/edac/mpc85xx_edac.h  | 1 +
 4 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/drivers/edac/edac_core.h b/drivers/edac/edac_core.h
index 3493c6bdb820..871c13b4c148 100644
--- a/drivers/edac/edac_core.h
+++ b/drivers/edac/edac_core.h
@@ -150,6 +150,8 @@ enum mem_type {
 	MEM_FB_DDR2,		/* fully buffered DDR2 */
 	MEM_RDDR2,		/* Registered DDR2 RAM */
 	MEM_XDR,		/* Rambus XDR */
+	MEM_DDR3,		/* DDR3 RAM */
+	MEM_RDDR3,		/* Registered DDR3 RAM */
 };
 
 #define MEM_FLAG_EMPTY		BIT(MEM_EMPTY)
@@ -167,6 +169,8 @@ enum mem_type {
 #define MEM_FLAG_FB_DDR2        BIT(MEM_FB_DDR2)
 #define MEM_FLAG_RDDR2          BIT(MEM_RDDR2)
 #define MEM_FLAG_XDR            BIT(MEM_XDR)
+#define MEM_FLAG_DDR3		 BIT(MEM_DDR3)
+#define MEM_FLAG_RDDR3		 BIT(MEM_RDDR3)
 
 /* chipset Error Detection and Correction capabilities and mode */
 enum edac_type {
diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c
index ad218fe4942d..e1d4ce083481 100644
--- a/drivers/edac/edac_mc_sysfs.c
+++ b/drivers/edac/edac_mc_sysfs.c
@@ -94,7 +94,9 @@ static const char *mem_types[] = {
 	[MEM_DDR2] = "Unbuffered-DDR2",
 	[MEM_FB_DDR2] = "FullyBuffered-DDR2",
 	[MEM_RDDR2] = "Registered-DDR2",
-	[MEM_XDR] = "XDR"
+	[MEM_XDR] = "XDR",
+	[MEM_DDR3] = "Unbuffered-DDR3",
+	[MEM_RDDR3] = "Registered-DDR3"
 };
 
 static const char *dev_types[] = {
diff --git a/drivers/edac/mpc85xx_edac.c b/drivers/edac/mpc85xx_edac.c
index 7c8c2d72916f..3f2ccfc6407c 100644
--- a/drivers/edac/mpc85xx_edac.c
+++ b/drivers/edac/mpc85xx_edac.c
@@ -757,6 +757,9 @@ static void __devinit mpc85xx_init_csrows(struct mem_ctl_info *mci)
 		case DSC_SDTYPE_DDR2:
 			mtype = MEM_RDDR2;
 			break;
+		case DSC_SDTYPE_DDR3:
+			mtype = MEM_RDDR3;
+			break;
 		default:
 			mtype = MEM_UNKNOWN;
 			break;
@@ -769,6 +772,9 @@ static void __devinit mpc85xx_init_csrows(struct mem_ctl_info *mci)
 		case DSC_SDTYPE_DDR2:
 			mtype = MEM_DDR2;
 			break;
+		case DSC_SDTYPE_DDR3:
+			mtype = MEM_DDR3;
+			break;
 		default:
 			mtype = MEM_UNKNOWN;
 			break;
diff --git a/drivers/edac/mpc85xx_edac.h b/drivers/edac/mpc85xx_edac.h
index 135b3539a030..52432ee7c4b9 100644
--- a/drivers/edac/mpc85xx_edac.h
+++ b/drivers/edac/mpc85xx_edac.h
@@ -53,6 +53,7 @@
 
 #define DSC_SDTYPE_DDR		0x02000000
 #define DSC_SDTYPE_DDR2		0x03000000
+#define DSC_SDTYPE_DDR3		0x07000000
 #define DSC_X32_EN	0x00000020
 
 /* Err_Int_En */
-- 
cgit v1.2.3-59-g8ed1b


From 341c87bf346f57748230628c5ad6ee69219250e8 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Tue, 30 Jun 2009 11:41:23 -0700
Subject: elf: limit max map count to safe value

With ELF, at generating coredump, some more headers other than used
vmas are added.

When max_map_count == 65536, a core generated by following kinds of
code can be unreadable because the number of ELF's program header is
written in 16bit in Ehdr (please see elf.h) and the number overflows.

==
	... = mmap(); (munmap, mprotect, etc...)
	if (failed)
		abort();
==

This can happen in mmap/munmap/mprotect/etc...which calls split_vma().

I think 65536 is not safe as _default_ and reduce it to 65530 is good
for avoiding unexpected corrupted core.

Anyway, max_map_count can be enlarged by sysctl if a user is brave..

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: Jakub Jelinek <jakub@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/binfmt_elf.c       |  5 ++++-
 include/linux/sched.h | 16 ++++++++++++++--
 2 files changed, 18 insertions(+), 3 deletions(-)

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 9fa212b014a5..f1867900e459 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -1929,7 +1929,10 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, un
 	elf = kmalloc(sizeof(*elf), GFP_KERNEL);
 	if (!elf)
 		goto out;
-	
+	/*
+	 * The number of segs are recored into ELF header as 16bit value.
+	 * Please check DEFAULT_MAX_MAP_COUNT definition when you modify here.
+	 */
 	segs = current->mm->map_count;
 #ifdef ELF_CORE_EXTRA_PHDRS
 	segs += ELF_CORE_EXTRA_PHDRS;
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4d0754269884..0085d758d645 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -349,8 +349,20 @@ extern int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner);
 struct nsproxy;
 struct user_namespace;
 
-/* Maximum number of active map areas.. This is a random (large) number */
-#define DEFAULT_MAX_MAP_COUNT	65536
+/*
+ * Default maximum number of active map areas, this limits the number of vmas
+ * per mm struct. Users can overwrite this number by sysctl but there is a
+ * problem.
+ *
+ * When a program's coredump is generated as ELF format, a section is created
+ * per a vma. In ELF, the number of sections is represented in unsigned short.
+ * This means the number of sections should be smaller than 65535 at coredump.
+ * Because the kernel adds some informative sections to a image of program at
+ * generating coredump, we need some margin. The number of extra sections is
+ * 1-3 now and depends on arch. We use "5" as safe margin, here.
+ */
+#define MAPCOUNT_ELF_CORE_MARGIN	(5)
+#define DEFAULT_MAX_MAP_COUNT	(USHORT_MAX - MAPCOUNT_ELF_CORE_MARGIN)
 
 extern int sysctl_max_map_count;
 
-- 
cgit v1.2.3-59-g8ed1b


From 4d6c13f87db12ae1ce35ea6a15688ac72419b133 Mon Sep 17 00:00:00 2001
From: Bryan Donlan <bdonlan@gmail.com>
Date: Tue, 30 Jun 2009 11:41:24 -0700
Subject: ext2: return -EIO not -ESTALE on directory traversal through deleted
 inode

ext2_iget() returns -ESTALE if invoked on a deleted inode, in order to
report errors to NFS properly.  However, in ext[234]_lookup(), this
-ESTALE can be propagated to userspace if the filesystem is corrupted such
that a directory entry references a deleted inode.  This leads to a
misleading error message - "Stale NFS file handle" - and confusion on the
part of the admin.

The bug can be easily reproduced by creating a new filesystem, making a
link to an unused inode using debugfs, then mounting and attempting to ls
-l said link.

This patch thus changes ext2_lookup to return -EIO if it receives -ESTALE
from ext2_iget(), as ext2 does for other filesystem metadata corruption;
and also invokes the appropriate ext*_error functions when this case is
detected.

Signed-off-by: Bryan Donlan <bdonlan@gmail.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext2/namei.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c
index 6524ecaebb7a..e1dedb0f7873 100644
--- a/fs/ext2/namei.c
+++ b/fs/ext2/namei.c
@@ -66,8 +66,16 @@ static struct dentry *ext2_lookup(struct inode * dir, struct dentry *dentry, str
 	inode = NULL;
 	if (ino) {
 		inode = ext2_iget(dir->i_sb, ino);
-		if (IS_ERR(inode))
-			return ERR_CAST(inode);
+		if (unlikely(IS_ERR(inode))) {
+			if (PTR_ERR(inode) == -ESTALE) {
+				ext2_error(dir->i_sb, __func__,
+						"deleted inode referenced: %lu",
+						ino);
+				return ERR_PTR(-EIO);
+			} else {
+				return ERR_CAST(inode);
+			}
+		}
 	}
 	return d_splice_alias(inode, dentry);
 }
-- 
cgit v1.2.3-59-g8ed1b


From c49568235dd7b4a2ffad63aa950562f4ffb9455f Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 30 Jun 2009 11:41:25 -0700
Subject: dmapools: protect page_list walk in show_pools()

show_pools() walks the page_list of a pool w/o protection against the list
modifications in alloc/free.  Take pool->lock to avoid stomping into
nirvana.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/dmapool.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/mm/dmapool.c b/mm/dmapool.c
index b1f0885dda22..3df063706f53 100644
--- a/mm/dmapool.c
+++ b/mm/dmapool.c
@@ -86,10 +86,12 @@ show_pools(struct device *dev, struct device_attribute *attr, char *buf)
 		unsigned pages = 0;
 		unsigned blocks = 0;
 
+		spin_lock_irq(&pool->lock);
 		list_for_each_entry(page, &pool->page_list, page_list) {
 			pages++;
 			blocks += page->in_use;
 		}
+		spin_unlock_irq(&pool->lock);
 
 		/* per-pool info, no real statistics yet */
 		temp = scnprintf(next, size, "%-16s %4u %4Zu %4Zu %2u\n",
-- 
cgit v1.2.3-59-g8ed1b


From b55f627feeb9d48fdbde3835e18afbc76712e49b Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Tue, 30 Jun 2009 11:41:26 -0700
Subject: spi: new spi->mode bits

Add two new spi_device.mode bits to accomodate more protocol options, and
pass them through to usermode drivers:

 * SPI_NO_CS ... a second 3-wire variant, where the chipselect
   line is removed instead of a data line; transfers are still
   full duplex.

   This obviously has STRONG protocol implications since the
   chipselect transitions can't be used to synchronize state
   transitions with the SPI master.

 * SPI_READY ... defines open drain signal that's pulled low
   to pause the clock.  This defines a 5-wire variant (normal
   4-wire SPI plus READY) and two 4-wire variants (READY plus
   each of the 3-wire flavors).

   Such hardware flow control can be a big win.  There are ADC
   converters and flash chips that expose READY signals, but not
   many host controllers support it today.

The spi_bitbang code should be changed to use SPI_NO_CS instead of its
current nonportable hack.  That's a mode most hardware can easily support
(unlike SPI_READY).

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: "Paulraj, Sandeep" <s-paulraj@ti.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/spi/spidev_test.c | 10 +++++++++-
 drivers/spi/spidev.c            | 17 +++++++++++------
 include/linux/spi/spi.h         |  2 ++
 include/linux/spi/spidev.h      |  2 ++
 4 files changed, 24 insertions(+), 7 deletions(-)

diff --git a/Documentation/spi/spidev_test.c b/Documentation/spi/spidev_test.c
index cf0e3ce0d526..c1a5aad3c75a 100644
--- a/Documentation/spi/spidev_test.c
+++ b/Documentation/spi/spidev_test.c
@@ -99,11 +99,13 @@ void parse_opts(int argc, char *argv[])
 			{ "lsb",     0, 0, 'L' },
 			{ "cs-high", 0, 0, 'C' },
 			{ "3wire",   0, 0, '3' },
+			{ "no-cs",   0, 0, 'N' },
+			{ "ready",   0, 0, 'R' },
 			{ NULL, 0, 0, 0 },
 		};
 		int c;
 
-		c = getopt_long(argc, argv, "D:s:d:b:lHOLC3", lopts, NULL);
+		c = getopt_long(argc, argv, "D:s:d:b:lHOLC3NR", lopts, NULL);
 
 		if (c == -1)
 			break;
@@ -139,6 +141,12 @@ void parse_opts(int argc, char *argv[])
 		case '3':
 			mode |= SPI_3WIRE;
 			break;
+		case 'N':
+			mode |= SPI_NO_CS;
+			break;
+		case 'R':
+			mode |= SPI_READY;
+			break;
 		default:
 			print_usage(argv[0]);
 			break;
diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c
index 5d869c4d3eb2..606e7a40a8da 100644
--- a/drivers/spi/spidev.c
+++ b/drivers/spi/spidev.c
@@ -58,15 +58,20 @@ static unsigned long	minors[N_SPI_MINORS / BITS_PER_LONG];
 
 
 /* Bit masks for spi_device.mode management.  Note that incorrect
- * settings for CS_HIGH and 3WIRE can cause *lots* of trouble for other
- * devices on a shared bus:  CS_HIGH, because this device will be
- * active when it shouldn't be;  3WIRE, because when active it won't
- * behave as it should.
+ * settings for some settings can cause *lots* of trouble for other
+ * devices on a shared bus:
  *
- * REVISIT should changing those two modes be privileged?
+ *  - CS_HIGH ... this device will be active when it shouldn't be
+ *  - 3WIRE ... when active, it won't behave as it should
+ *  - NO_CS ... there will be no explicit message boundaries; this
+ *	is completely incompatible with the shared bus model
+ *  - READY ... transfers may proceed when they shouldn't.
+ *
+ * REVISIT should changing those flags be privileged?
  */
 #define SPI_MODE_MASK		(SPI_CPHA | SPI_CPOL | SPI_CS_HIGH \
-				| SPI_LSB_FIRST | SPI_3WIRE | SPI_LOOP)
+				| SPI_LSB_FIRST | SPI_3WIRE | SPI_LOOP \
+				| SPI_NO_CS | SPI_READY)
 
 struct spidev_data {
 	dev_t			devt;
diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h
index 9c4cd27f4685..743c933ac4e7 100644
--- a/include/linux/spi/spi.h
+++ b/include/linux/spi/spi.h
@@ -80,6 +80,8 @@ struct spi_device {
 #define	SPI_LSB_FIRST	0x08			/* per-word bits-on-wire */
 #define	SPI_3WIRE	0x10			/* SI/SO signals shared */
 #define	SPI_LOOP	0x20			/* loopback mode */
+#define	SPI_NO_CS	0x40			/* 1 dev/bus, no chipselect */
+#define	SPI_READY	0x80			/* slave pulls low to pause */
 	u8			bits_per_word;
 	int			irq;
 	void			*controller_state;
diff --git a/include/linux/spi/spidev.h b/include/linux/spi/spidev.h
index 95251ccd5a07..bf0570a84f7a 100644
--- a/include/linux/spi/spidev.h
+++ b/include/linux/spi/spidev.h
@@ -40,6 +40,8 @@
 #define SPI_LSB_FIRST		0x08
 #define SPI_3WIRE		0x10
 #define SPI_LOOP		0x20
+#define SPI_NO_CS		0x40
+#define SPI_READY		0x80
 
 /*---------------------------------------------------------------------------*/
 
-- 
cgit v1.2.3-59-g8ed1b


From 70d6027ff2bc8bab180273b77e7ab3e8a62cca51 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Tue, 30 Jun 2009 11:41:27 -0700
Subject: spi: add spi_master flag word

Add a new spi_master.flags word listing constraints relevant to that
controller.  Define the first constraint bit: a half duplex restriction.
Include that constraint in the OMAP1 MicroWire controller driver.

Have the mmc_spi host be the first customer of this flag.  Its coding
relies heavily on full duplex transfers, so it must fail when the
underlying controller driver won't perform them.

(The spi_write_then_read routine could use it too: use the
temporarily-withdrawn full-duplex speedup unless this flag is set, in
which case the existing code applies.  Similarly, any spi_master
implementing only SPI_3WIRE should set the flag.)

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mmc/host/mmc_spi.c | 6 ++++++
 drivers/spi/omap_uwire.c   | 2 ++
 include/linux/spi/spi.h    | 4 ++++
 3 files changed, 12 insertions(+)

diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
index 240608cc7ae9..a461017ce5ce 100644
--- a/drivers/mmc/host/mmc_spi.c
+++ b/drivers/mmc/host/mmc_spi.c
@@ -1313,6 +1313,12 @@ static int mmc_spi_probe(struct spi_device *spi)
 	struct mmc_spi_host	*host;
 	int			status;
 
+	/* We rely on full duplex transfers, mostly to reduce
+	 * per-transfer overheads (by making fewer transfers).
+	 */
+	if (spi->master->flags & SPI_MASTER_HALF_DUPLEX)
+		return -EINVAL;
+
 	/* MMC and SD specs only seem to care that sampling is on the
 	 * rising edge ... meaning SPI modes 0 or 3.  So either SPI mode
 	 * should be legit.  We'll use mode 0 since the steady state is 0,
diff --git a/drivers/spi/omap_uwire.c b/drivers/spi/omap_uwire.c
index aa90ddb37066..8980a5640bd9 100644
--- a/drivers/spi/omap_uwire.c
+++ b/drivers/spi/omap_uwire.c
@@ -514,6 +514,8 @@ static int __init uwire_probe(struct platform_device *pdev)
 	/* the spi->mode bits understood by this driver: */
 	master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH;
 
+	master->flags = SPI_MASTER_HALF_DUPLEX;
+
 	master->bus_num = 2;	/* "official" */
 	master->num_chipselect = 4;
 	master->setup = uwire_setup;
diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h
index 743c933ac4e7..c47c4b4da97e 100644
--- a/include/linux/spi/spi.h
+++ b/include/linux/spi/spi.h
@@ -250,6 +250,10 @@ struct spi_master {
 	/* spi_device.mode flags understood by this controller driver */
 	u16			mode_bits;
 
+	/* other constraints relevant to this driver */
+	u16			flags;
+#define SPI_MASTER_HALF_DUPLEX	BIT(0)		/* can't do full duplex */
+
 	/* Setup mode and clock, etc (spi driver may call many times).
 	 *
 	 * IMPORTANT:  this may be called when transfers to another
-- 
cgit v1.2.3-59-g8ed1b


From 537a1bf059fa312355696fa6db80726e655e7f17 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Tue, 30 Jun 2009 11:41:29 -0700
Subject: fbdev: add mutex for fb_mmap locking

Add a mutex to avoid a circular locking problem between the mm layer
semaphore and fbdev ioctl mutex through the fb_mmap() call.

Also, add mutex to all places where smem_start and smem_len fields change
so the mutex inside the fb_mmap() is actually used.  Changing of these
fields before calling the framebuffer_register() are not mutexed.

This is 2.6.31 material.  It removes one lockdep (fb_mmap() and
register_framebuffer()) but there is still another one (fb_release() and
register_framebuffer()).  It also cleans up handling of the smem_start and
smem_len fields used by mutexed section of the fb_mmap().

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/atafb.c                 |  7 ++++++-
 drivers/video/atmel_lcdfb.c           |  2 ++
 drivers/video/fbmem.c                 | 13 +++++--------
 drivers/video/fsl-diu-fb.c            | 14 +++++++++-----
 drivers/video/i810/i810_main.c        |  2 ++
 drivers/video/matrox/matroxfb_base.c  |  3 +++
 drivers/video/matrox/matroxfb_crtc2.c |  5 ++++-
 drivers/video/mx3fb.c                 | 17 +++++++++++------
 drivers/video/omap/omapfb_main.c      |  4 ++++
 drivers/video/platinumfb.c            |  2 ++
 drivers/video/pxafb.c                 |  2 ++
 drivers/video/sh7760fb.c              | 19 ++++++-------------
 drivers/video/sis/sis_main.c          |  2 ++
 drivers/video/sm501fb.c               | 21 +++++++++++++--------
 drivers/video/w100fb.c                |  2 ++
 include/linux/fb.h                    |  1 +
 16 files changed, 74 insertions(+), 42 deletions(-)

diff --git a/drivers/video/atafb.c b/drivers/video/atafb.c
index 018850c116c6..497ff8af03ed 100644
--- a/drivers/video/atafb.c
+++ b/drivers/video/atafb.c
@@ -2414,7 +2414,10 @@ static int atafb_get_fix(struct fb_fix_screeninfo *fix, struct fb_info *info)
 	if (err)
 		return err;
 	memset(fix, 0, sizeof(struct fb_fix_screeninfo));
-	return fbhw->encode_fix(fix, &par);
+	mutex_lock(&info->mm_lock);
+	err = fbhw->encode_fix(fix, &par);
+	mutex_unlock(&info->mm_lock);
+	return err;
 }
 
 static int atafb_get_var(struct fb_var_screeninfo *var, struct fb_info *info)
@@ -2743,7 +2746,9 @@ static int atafb_set_par(struct fb_info *info)
 
 	/* Decode wanted screen parameters */
 	fbhw->decode_var(&info->var, par);
+	mutex_lock(&info->mm_lock);
 	fbhw->encode_fix(&info->fix, par);
+	mutex_unlock(&info->mm_lock);
 
 	/* Set new videomode */
 	ata_set_par(par);
diff --git a/drivers/video/atmel_lcdfb.c b/drivers/video/atmel_lcdfb.c
index 5afd64482f55..cb88394ba995 100644
--- a/drivers/video/atmel_lcdfb.c
+++ b/drivers/video/atmel_lcdfb.c
@@ -270,7 +270,9 @@ static int atmel_lcdfb_alloc_video_memory(struct atmel_lcdfb_info *sinfo)
 
 	smem_len = (var->xres_virtual * var->yres_virtual
 		    * ((var->bits_per_pixel + 7) / 8));
+	mutex_lock(&info->mm_lock);
 	info->fix.smem_len = max(smem_len, sinfo->smem_len);
+	mutex_unlock(&info->mm_lock);
 
 	info->screen_base = dma_alloc_writecombine(info->device, info->fix.smem_len,
 					(dma_addr_t *)&info->fix.smem_start, GFP_KERNEL);
diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c
index f8a09bf8d0cd..53ea05645ff8 100644
--- a/drivers/video/fbmem.c
+++ b/drivers/video/fbmem.c
@@ -1310,8 +1310,6 @@ static long fb_compat_ioctl(struct file *file, unsigned int cmd,
 
 static int
 fb_mmap(struct file *file, struct vm_area_struct * vma)
-__acquires(&info->lock)
-__releases(&info->lock)
 {
 	int fbidx = iminor(file->f_path.dentry->d_inode);
 	struct fb_info *info = registered_fb[fbidx];
@@ -1325,16 +1323,14 @@ __releases(&info->lock)
 	off = vma->vm_pgoff << PAGE_SHIFT;
 	if (!fb)
 		return -ENODEV;
+	mutex_lock(&info->mm_lock);
 	if (fb->fb_mmap) {
 		int res;
-		mutex_lock(&info->lock);
 		res = fb->fb_mmap(info, vma);
-		mutex_unlock(&info->lock);
+		mutex_unlock(&info->mm_lock);
 		return res;
 	}
 
-	mutex_lock(&info->lock);
-
 	/* frame buffer memory */
 	start = info->fix.smem_start;
 	len = PAGE_ALIGN((start & ~PAGE_MASK) + info->fix.smem_len);
@@ -1342,13 +1338,13 @@ __releases(&info->lock)
 		/* memory mapped io */
 		off -= len;
 		if (info->var.accel_flags) {
-			mutex_unlock(&info->lock);
+			mutex_unlock(&info->mm_lock);
 			return -EINVAL;
 		}
 		start = info->fix.mmio_start;
 		len = PAGE_ALIGN((start & ~PAGE_MASK) + info->fix.mmio_len);
 	}
-	mutex_unlock(&info->lock);
+	mutex_unlock(&info->mm_lock);
 	start &= PAGE_MASK;
 	if ((vma->vm_end - vma->vm_start + off) > len)
 		return -EINVAL;
@@ -1518,6 +1514,7 @@ register_framebuffer(struct fb_info *fb_info)
 			break;
 	fb_info->node = i;
 	mutex_init(&fb_info->lock);
+	mutex_init(&fb_info->mm_lock);
 
 	fb_info->dev = device_create(fb_class, fb_info->device,
 				     MKDEV(FB_MAJOR, i), NULL, "fb%d", i);
diff --git a/drivers/video/fsl-diu-fb.c b/drivers/video/fsl-diu-fb.c
index f153c581cbd7..0bf2190928d0 100644
--- a/drivers/video/fsl-diu-fb.c
+++ b/drivers/video/fsl-diu-fb.c
@@ -750,24 +750,26 @@ static void update_lcdc(struct fb_info *info)
 static int map_video_memory(struct fb_info *info)
 {
 	phys_addr_t phys;
+	u32 smem_len = info->fix.line_length * info->var.yres_virtual;
 
 	pr_debug("info->var.xres_virtual = %d\n", info->var.xres_virtual);
 	pr_debug("info->var.yres_virtual = %d\n", info->var.yres_virtual);
 	pr_debug("info->fix.line_length  = %d\n", info->fix.line_length);
+	pr_debug("MAP_VIDEO_MEMORY: smem_len = %u\n", smem_len);
 
-	info->fix.smem_len = info->fix.line_length * info->var.yres_virtual;
-	pr_debug("MAP_VIDEO_MEMORY: smem_len = %d\n", info->fix.smem_len);
-	info->screen_base = fsl_diu_alloc(info->fix.smem_len, &phys);
+	info->screen_base = fsl_diu_alloc(smem_len, &phys);
 	if (info->screen_base == NULL) {
 		printk(KERN_ERR "Unable to allocate fb memory\n");
 		return -ENOMEM;
 	}
+	mutex_lock(&info->mm_lock);
 	info->fix.smem_start = (unsigned long) phys;
+	info->fix.smem_len = smem_len;
+	mutex_unlock(&info->mm_lock);
 	info->screen_size = info->fix.smem_len;
 
 	pr_debug("Allocated fb @ paddr=0x%08lx, size=%d.\n",
-				info->fix.smem_start,
-		info->fix.smem_len);
+		 info->fix.smem_start, info->fix.smem_len);
 	pr_debug("screen base %p\n", info->screen_base);
 
 	return 0;
@@ -776,9 +778,11 @@ static int map_video_memory(struct fb_info *info)
 static void unmap_video_memory(struct fb_info *info)
 {
 	fsl_diu_free(info->screen_base, info->fix.smem_len);
+	mutex_lock(&info->mm_lock);
 	info->screen_base = NULL;
 	info->fix.smem_start = 0;
 	info->fix.smem_len = 0;
+	mutex_unlock(&info->mm_lock);
 }
 
 /*
diff --git a/drivers/video/i810/i810_main.c b/drivers/video/i810/i810_main.c
index 2e940199fc89..71960672d721 100644
--- a/drivers/video/i810/i810_main.c
+++ b/drivers/video/i810/i810_main.c
@@ -1090,8 +1090,10 @@ static int encode_fix(struct fb_fix_screeninfo *fix, struct fb_info *info)
     	memset(fix, 0, sizeof(struct fb_fix_screeninfo));
 
     	strcpy(fix->id, "I810");
+	mutex_lock(&info->mm_lock);
     	fix->smem_start = par->fb.physical;
     	fix->smem_len = par->fb.size;
+	mutex_unlock(&info->mm_lock);
     	fix->type = FB_TYPE_PACKED_PIXELS;
     	fix->type_aux = 0;
 	fix->xpanstep = 8;
diff --git a/drivers/video/matrox/matroxfb_base.c b/drivers/video/matrox/matroxfb_base.c
index 8e7a275df50c..59c3a2e14913 100644
--- a/drivers/video/matrox/matroxfb_base.c
+++ b/drivers/video/matrox/matroxfb_base.c
@@ -724,8 +724,10 @@ static void matroxfb_update_fix(WPMINFO2)
 	struct fb_fix_screeninfo *fix = &ACCESS_FBINFO(fbcon).fix;
 	DBG(__func__)
 
+	mutex_lock(&ACCESS_FBINFO(fbcon).mm_lock);
 	fix->smem_start = ACCESS_FBINFO(video.base) + ACCESS_FBINFO(curr.ydstorg.bytes);
 	fix->smem_len = ACCESS_FBINFO(video.len_usable) - ACCESS_FBINFO(curr.ydstorg.bytes);
+	mutex_unlock(&ACCESS_FBINFO(fbcon).mm_lock);
 }
 
 static int matroxfb_check_var(struct fb_var_screeninfo *var, struct fb_info *info)
@@ -2081,6 +2083,7 @@ static int matroxfb_probe(struct pci_dev* pdev, const struct pci_device_id* dumm
 	spin_lock_init(&ACCESS_FBINFO(lock.accel));
 	init_rwsem(&ACCESS_FBINFO(crtc2.lock));
 	init_rwsem(&ACCESS_FBINFO(altout.lock));
+	mutex_init(&ACCESS_FBINFO(fbcon).mm_lock);
 	ACCESS_FBINFO(irq_flags) = 0;
 	init_waitqueue_head(&ACCESS_FBINFO(crtc1.vsync.wait));
 	init_waitqueue_head(&ACCESS_FBINFO(crtc2.vsync.wait));
diff --git a/drivers/video/matrox/matroxfb_crtc2.c b/drivers/video/matrox/matroxfb_crtc2.c
index 7ac4c5f6145d..909e10a11898 100644
--- a/drivers/video/matrox/matroxfb_crtc2.c
+++ b/drivers/video/matrox/matroxfb_crtc2.c
@@ -289,13 +289,16 @@ static int matroxfb_dh_release(struct fb_info* info, int user) {
 #undef m2info
 }
 
-static void matroxfb_dh_init_fix(struct matroxfb_dh_fb_info *m2info) {
+static void matroxfb_dh_init_fix(struct matroxfb_dh_fb_info *m2info)
+{
 	struct fb_fix_screeninfo *fix = &m2info->fbcon.fix;
 
 	strcpy(fix->id, "MATROX DH");
 
+	mutex_lock(&m2info->fbcon.mm_lock);
 	fix->smem_start = m2info->video.base;
 	fix->smem_len = m2info->video.len_usable;
+	mutex_unlock(&m2info->fbcon.mm_lock);
 	fix->ypanstep = 1;
 	fix->ywrapstep = 0;
 	fix->xpanstep = 8;	/* TBD */
diff --git a/drivers/video/mx3fb.c b/drivers/video/mx3fb.c
index b7af5256e887..567fb944bd2a 100644
--- a/drivers/video/mx3fb.c
+++ b/drivers/video/mx3fb.c
@@ -669,7 +669,7 @@ static uint32_t bpp_to_pixfmt(int bpp)
 }
 
 static int mx3fb_blank(int blank, struct fb_info *fbi);
-static int mx3fb_map_video_memory(struct fb_info *fbi);
+static int mx3fb_map_video_memory(struct fb_info *fbi, unsigned int mem_len);
 static int mx3fb_unmap_video_memory(struct fb_info *fbi);
 
 /**
@@ -742,8 +742,7 @@ static int mx3fb_set_par(struct fb_info *fbi)
 		if (fbi->fix.smem_start)
 			mx3fb_unmap_video_memory(fbi);
 
-		fbi->fix.smem_len = mem_len;
-		if (mx3fb_map_video_memory(fbi) < 0) {
+		if (mx3fb_map_video_memory(fbi, mem_len) < 0) {
 			mutex_unlock(&mx3_fbi->mutex);
 			return -ENOMEM;
 		}
@@ -1198,6 +1197,7 @@ static int mx3fb_resume(struct platform_device *pdev)
 /**
  * mx3fb_map_video_memory() - allocates the DRAM memory for the frame buffer.
  * @fbi:	framebuffer information pointer
+ * @mem_len:	length of mapped memory
  * @return:	Error code indicating success or failure
  *
  * This buffer is remapped into a non-cached, non-buffered, memory region to
@@ -1205,23 +1205,26 @@ static int mx3fb_resume(struct platform_device *pdev)
  * area is remapped, all virtual memory access to the video memory should occur
  * at the new region.
  */
-static int mx3fb_map_video_memory(struct fb_info *fbi)
+static int mx3fb_map_video_memory(struct fb_info *fbi, unsigned int mem_len)
 {
 	int retval = 0;
 	dma_addr_t addr;
 
 	fbi->screen_base = dma_alloc_writecombine(fbi->device,
-						  fbi->fix.smem_len,
+						  mem_len,
 						  &addr, GFP_DMA);
 
 	if (!fbi->screen_base) {
 		dev_err(fbi->device, "Cannot allocate %u bytes framebuffer memory\n",
-			fbi->fix.smem_len);
+			mem_len);
 		retval = -EBUSY;
 		goto err0;
 	}
 
+	mutex_lock(&fbi->mm_lock);
 	fbi->fix.smem_start = addr;
+	fbi->fix.smem_len = mem_len;
+	mutex_unlock(&fbi->mm_lock);
 
 	dev_dbg(fbi->device, "allocated fb @ p=0x%08x, v=0x%p, size=%d.\n",
 		(uint32_t) fbi->fix.smem_start, fbi->screen_base, fbi->fix.smem_len);
@@ -1251,8 +1254,10 @@ static int mx3fb_unmap_video_memory(struct fb_info *fbi)
 			      fbi->screen_base, fbi->fix.smem_start);
 
 	fbi->screen_base = 0;
+	mutex_lock(&fbi->mm_lock);
 	fbi->fix.smem_start = 0;
 	fbi->fix.smem_len = 0;
+	mutex_unlock(&fbi->mm_lock);
 	return 0;
 }
 
diff --git a/drivers/video/omap/omapfb_main.c b/drivers/video/omap/omapfb_main.c
index 060d72fe57cb..4ea99bfc37b4 100644
--- a/drivers/video/omap/omapfb_main.c
+++ b/drivers/video/omap/omapfb_main.c
@@ -393,8 +393,10 @@ static void set_fb_fix(struct fb_info *fbi)
 
 	rg = &plane->fbdev->mem_desc.region[plane->idx];
 	fbi->screen_base	= rg->vaddr;
+	mutex_lock(&fbi->mm_lock);
 	fix->smem_start		= rg->paddr;
 	fix->smem_len		= rg->size;
+	mutex_unlock(&fbi->mm_lock);
 
 	fix->type = FB_TYPE_PACKED_PIXELS;
 	bpp = var->bits_per_pixel;
@@ -886,8 +888,10 @@ static int omapfb_setup_mem(struct fb_info *fbi, struct omapfb_mem_info *mi)
 				 * plane memory is dealloce'd, the other
 				 * screen parameters in var / fix are invalid.
 				 */
+				mutex_lock(&fbi->mm_lock);
 				fbi->fix.smem_start = 0;
 				fbi->fix.smem_len = 0;
+				mutex_unlock(&fbi->mm_lock);
 			}
 		}
 	}
diff --git a/drivers/video/platinumfb.c b/drivers/video/platinumfb.c
index 03b3670130a0..bacfabd9ce16 100644
--- a/drivers/video/platinumfb.c
+++ b/drivers/video/platinumfb.c
@@ -141,7 +141,9 @@ static int platinumfb_set_par (struct fb_info *info)
   		offset = 0x10;
 
 	info->screen_base = pinfo->frame_buffer + init->fb_offset + offset;
+	mutex_lock(&info->mm_lock);
 	info->fix.smem_start = (pinfo->frame_buffer_phys) + init->fb_offset + offset;
+	mutex_unlock(&info->mm_lock);
 	info->fix.visual = (pinfo->cmode == CMODE_8) ?
 		FB_VISUAL_PSEUDOCOLOR : FB_VISUAL_DIRECTCOLOR;
  	info->fix.line_length = vmode_attrs[pinfo->vmode-1].hres * (1<<pinfo->cmode)
diff --git a/drivers/video/pxafb.c b/drivers/video/pxafb.c
index 0889d50c3288..6506117c134b 100644
--- a/drivers/video/pxafb.c
+++ b/drivers/video/pxafb.c
@@ -815,8 +815,10 @@ static int overlayfb_map_video_memory(struct pxafb_layer *ofb)
 	ofb->video_mem_phys = virt_to_phys(ofb->video_mem);
 	ofb->video_mem_size = size;
 
+	mutex_lock(&ofb->fb.mm_lock);
 	ofb->fb.fix.smem_start	= ofb->video_mem_phys;
 	ofb->fb.fix.smem_len	= ofb->fb.fix.line_length * var->yres_virtual;
+	mutex_unlock(&ofb->fb.mm_lock);
 	ofb->fb.screen_base	= ofb->video_mem;
 	return 0;
 }
diff --git a/drivers/video/sh7760fb.c b/drivers/video/sh7760fb.c
index 653bdfee3057..9f6d6e61f0cc 100644
--- a/drivers/video/sh7760fb.c
+++ b/drivers/video/sh7760fb.c
@@ -120,18 +120,6 @@ static int sh7760_setcolreg (u_int regno,
 	return 0;
 }
 
-static void encode_fix(struct fb_fix_screeninfo *fix, struct fb_info *info,
-		       unsigned long stride)
-{
-	memset(fix, 0, sizeof(struct fb_fix_screeninfo));
-	strcpy(fix->id, "sh7760-lcdc");
-
-	fix->smem_start = (unsigned long)info->screen_base;
-	fix->smem_len = info->screen_size;
-
-	fix->line_length = stride;
-}
-
 static int sh7760fb_get_color_info(struct device *dev,
 				   u16 lddfr, int *bpp, int *gray)
 {
@@ -334,7 +322,8 @@ static int sh7760fb_set_par(struct fb_info *info)
 
 	iowrite32(ldsarl, par->base + LDSARL);	/* mem for lower half of DSTN */
 
-	encode_fix(&info->fix, info, stride);
+	info->fix.line_length = stride;
+
 	sh7760fb_check_var(&info->var, info);
 
 	sh7760fb_blank(FB_BLANK_UNBLANK, info);	/* panel on! */
@@ -435,6 +424,8 @@ static int sh7760fb_alloc_mem(struct fb_info *info)
 
 	info->screen_base = fbmem;
 	info->screen_size = vram;
+	info->fix.smem_start = (unsigned long)info->screen_base;
+	info->fix.smem_len = info->screen_size;
 
 	return 0;
 }
@@ -520,6 +511,8 @@ static int __devinit sh7760fb_probe(struct platform_device *pdev)
 	info->var.transp.length = 0;
 	info->var.transp.msb_right = 0;
 
+	strcpy(info->fix.id, "sh7760-lcdc");
+
 	/* set the DON2 bit now, before cmap allocation, as it will randomize
 	 * palette memory.
 	 */
diff --git a/drivers/video/sis/sis_main.c b/drivers/video/sis/sis_main.c
index 7072d19080d5..fd33455389b8 100644
--- a/drivers/video/sis/sis_main.c
+++ b/drivers/video/sis/sis_main.c
@@ -1847,8 +1847,10 @@ sisfb_get_fix(struct fb_fix_screeninfo *fix, int con, struct fb_info *info)
 
 	strcpy(fix->id, ivideo->myid);
 
+	mutex_lock(&info->mm_lock);
 	fix->smem_start  = ivideo->video_base + ivideo->video_offset;
 	fix->smem_len    = ivideo->sisfb_mem;
+	mutex_unlock(&info->mm_lock);
 	fix->type        = FB_TYPE_PACKED_PIXELS;
 	fix->type_aux    = 0;
 	fix->visual      = (ivideo->video_bpp == 8) ? FB_VISUAL_PSEUDOCOLOR : FB_VISUAL_TRUECOLOR;
diff --git a/drivers/video/sm501fb.c b/drivers/video/sm501fb.c
index eb5d73a06702..98f24f0ec00d 100644
--- a/drivers/video/sm501fb.c
+++ b/drivers/video/sm501fb.c
@@ -145,7 +145,7 @@ static inline void sm501fb_sync_regs(struct sm501fb_info *info)
 #define SM501_MEMF_ACCEL		(8)
 
 static int sm501_alloc_mem(struct sm501fb_info *inf, struct sm501_mem *mem,
-			   unsigned int why, size_t size)
+			   unsigned int why, size_t size, u32 smem_len)
 {
 	struct sm501fb_par *par;
 	struct fb_info *fbi;
@@ -172,7 +172,7 @@ static int sm501_alloc_mem(struct sm501fb_info *inf, struct sm501_mem *mem,
 		if (ptr > 0)
 			ptr &= ~(PAGE_SIZE - 1);
 
-		if (fbi && ptr < fbi->fix.smem_len)
+		if (fbi && ptr < smem_len)
 			return -ENOMEM;
 
 		break;
@@ -197,7 +197,7 @@ static int sm501_alloc_mem(struct sm501fb_info *inf, struct sm501_mem *mem,
 
 	case SM501_MEMF_ACCEL:
 		fbi = inf->fb[HEAD_CRT];
-		ptr = fbi ? fbi->fix.smem_len : 0;
+		ptr = fbi ? smem_len : 0;
 
 		fbi = inf->fb[HEAD_PANEL];
 		if (fbi) {
@@ -413,6 +413,7 @@ static int sm501fb_set_par_common(struct fb_info *info,
 	unsigned int mem_type;
 	unsigned int clock_type;
 	unsigned int head_addr;
+	unsigned int smem_len;
 
 	dev_dbg(fbi->dev, "%s: %dx%d, bpp = %d, virtual %dx%d\n",
 		__func__, var->xres, var->yres, var->bits_per_pixel,
@@ -453,18 +454,20 @@ static int sm501fb_set_par_common(struct fb_info *info,
 
 	/* allocate fb memory within 501 */
 	info->fix.line_length = (var->xres_virtual * var->bits_per_pixel)/8;
-	info->fix.smem_len    = info->fix.line_length * var->yres_virtual;
+	smem_len = info->fix.line_length * var->yres_virtual;
 
 	dev_dbg(fbi->dev, "%s: line length = %u\n", __func__,
 		info->fix.line_length);
 
-	if (sm501_alloc_mem(fbi, &par->screen, mem_type,
-			    info->fix.smem_len)) {
+	if (sm501_alloc_mem(fbi, &par->screen, mem_type, smem_len, smem_len)) {
 		dev_err(fbi->dev, "no memory available\n");
 		return -ENOMEM;
 	}
 
+	mutex_lock(&info->mm_lock);
 	info->fix.smem_start = fbi->fbmem_res->start + par->screen.sm_addr;
+	info->fix.smem_len   = smem_len;
+	mutex_unlock(&info->mm_lock);
 
 	info->screen_base = fbi->fbmem + par->screen.sm_addr;
 	info->screen_size = info->fix.smem_len;
@@ -637,7 +640,8 @@ static int sm501fb_set_par_crt(struct fb_info *info)
 	if ((control & SM501_DC_CRT_CONTROL_SEL) == 0) {
 		/* the head is displaying panel data... */
 
-		sm501_alloc_mem(fbi, &par->screen, SM501_MEMF_CRT, 0);
+		sm501_alloc_mem(fbi, &par->screen, SM501_MEMF_CRT, 0,
+				info->fix.smem_len);
 		goto out_update;
 	}
 
@@ -1289,7 +1293,8 @@ static int sm501_init_cursor(struct fb_info *fbi, unsigned int reg_base)
 
 	par->cursor_regs = info->regs + reg_base;
 
-	ret = sm501_alloc_mem(info, &par->cursor, SM501_MEMF_CURSOR, 1024);
+	ret = sm501_alloc_mem(info, &par->cursor, SM501_MEMF_CURSOR, 1024,
+			      fbi->fix.smem_len);
 	if (ret < 0)
 		return ret;
 
diff --git a/drivers/video/w100fb.c b/drivers/video/w100fb.c
index d0674f1e3f10..8a141c2c637b 100644
--- a/drivers/video/w100fb.c
+++ b/drivers/video/w100fb.c
@@ -523,6 +523,7 @@ static int w100fb_set_par(struct fb_info *info)
 		info->fix.ywrapstep = 0;
 		info->fix.line_length = par->xres * BITS_PER_PIXEL / 8;
 
+		mutex_lock(&info->mm_lock);
 		if ((par->xres*par->yres*BITS_PER_PIXEL/8) > (MEM_INT_SIZE+1)) {
 			par->extmem_active = 1;
 			info->fix.smem_len = par->mach->mem->size+1;
@@ -530,6 +531,7 @@ static int w100fb_set_par(struct fb_info *info)
 			par->extmem_active = 0;
 			info->fix.smem_len = MEM_INT_SIZE+1;
 		}
+		mutex_unlock(&info->mm_lock);
 
 		w100fb_activate_var(par);
 	}
diff --git a/include/linux/fb.h b/include/linux/fb.h
index dd68358996b7..f847df9e99b6 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -819,6 +819,7 @@ struct fb_info {
 	int node;
 	int flags;
 	struct mutex lock;		/* Lock for open/release/ioctl funcs */
+	struct mutex mm_lock;		/* Lock for fb_mmap and smem_* fields */
 	struct fb_var_screeninfo var;	/* Current var */
 	struct fb_fix_screeninfo fix;	/* Current fix */
 	struct fb_monspecs monspecs;	/* Current Monitor specs */
-- 
cgit v1.2.3-59-g8ed1b


From 529ba0d9669386157457a1cb96294d2fe79b3f88 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Tue, 30 Jun 2009 11:41:30 -0700
Subject: spi: bitbang bugfix in message setup

Bugfix to spi_bitbang infrastructure: make sure to always set transfer
parameters on the first pass through the message's per-transfer loop.
This can matter with drivers that replace the per-word or per-buffer
transfer primitives, on busses with multiple SPI devices.

Previously, this could have started messages using the settings left after
previous messages.  The problem was observed when a high speed chip
(m25p80 type flash) was running very slowly because a low speed device
(avr8 microcontroller) had previously used the bus.  Similar faults could
have driven the low speed device too fast, or used an unexpected word
size.

Acked-by: Steven A. Falco <sfalco@harris.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/spi/spi_bitbang.c | 24 ++++++++++++++----------
 1 file changed, 14 insertions(+), 10 deletions(-)

diff --git a/drivers/spi/spi_bitbang.c b/drivers/spi/spi_bitbang.c
index 2a5abc08e857..f1db395dd889 100644
--- a/drivers/spi/spi_bitbang.c
+++ b/drivers/spi/spi_bitbang.c
@@ -258,6 +258,11 @@ static void bitbang_work(struct work_struct *work)
 	struct spi_bitbang	*bitbang =
 		container_of(work, struct spi_bitbang, work);
 	unsigned long		flags;
+	int			do_setup = -1;
+	int			(*setup_transfer)(struct spi_device *,
+					struct spi_transfer *);
+
+	setup_transfer = bitbang->setup_transfer;
 
 	spin_lock_irqsave(&bitbang->lock, flags);
 	bitbang->busy = 1;
@@ -269,8 +274,6 @@ static void bitbang_work(struct work_struct *work)
 		unsigned		tmp;
 		unsigned		cs_change;
 		int			status;
-		int			(*setup_transfer)(struct spi_device *,
-						struct spi_transfer *);
 
 		m = container_of(bitbang->queue.next, struct spi_message,
 				queue);
@@ -287,19 +290,19 @@ static void bitbang_work(struct work_struct *work)
 		tmp = 0;
 		cs_change = 1;
 		status = 0;
-		setup_transfer = NULL;
 
 		list_for_each_entry (t, &m->transfers, transfer_list) {
 
-			/* override or restore speed and wordsize */
-			if (t->speed_hz || t->bits_per_word) {
-				setup_transfer = bitbang->setup_transfer;
+			/* override speed or wordsize? */
+			if (t->speed_hz || t->bits_per_word)
+				do_setup = 1;
+
+			/* init (-1) or override (1) transfer params */
+			if (do_setup != 0) {
 				if (!setup_transfer) {
 					status = -ENOPROTOOPT;
 					break;
 				}
-			}
-			if (setup_transfer) {
 				status = setup_transfer(spi, t);
 				if (status < 0)
 					break;
@@ -363,9 +366,10 @@ static void bitbang_work(struct work_struct *work)
 		m->status = status;
 		m->complete(m->context);
 
-		/* restore speed and wordsize */
-		if (setup_transfer)
+		/* restore speed and wordsize if it was overridden */
+		if (do_setup == 1)
 			setup_transfer(spi, NULL);
+		do_setup = 0;
 
 		/* normally deactivate chipselect ... unless no error and
 		 * cs_change has hinted that the next message will probably
-- 
cgit v1.2.3-59-g8ed1b


From 8bc1ad7dd301b7ca7454013519fa92e8c53655ff Mon Sep 17 00:00:00 2001
From: Zhang Rui <rui.zhang@intel.com>
Date: Tue, 30 Jun 2009 11:41:31 -0700
Subject: kernel/resource.c: fix sign extension in reserve_setup()

When the 32-bit signed quantities get assigned to the u64 resource_size_t,
they are incorrectly sign-extended.

Addresses http://bugzilla.kernel.org/show_bug.cgi?id=13253
Addresses http://bugzilla.kernel.org/show_bug.cgi?id=9905

Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Reported-by: Leann Ogasawara <leann@ubuntu.com>
Cc: Pierre Ossman <drzeus@drzeus.cx>
Reported-by: <pablomme@googlemail.com>
Tested-by: <pablomme@googlemail.com>
Cc: <stable@kernel.org>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/resource.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/resource.c b/kernel/resource.c
index ac5f3a36923f..78b087221c15 100644
--- a/kernel/resource.c
+++ b/kernel/resource.c
@@ -787,7 +787,7 @@ static int __init reserve_setup(char *str)
 	static struct resource reserve[MAXRESERVE];
 
 	for (;;) {
-		int io_start, io_num;
+		unsigned int io_start, io_num;
 		int x = reserved;
 
 		if (get_option (&str, &io_start) != 2)
-- 
cgit v1.2.3-59-g8ed1b


From b4f90189dc2c7a7e6926ea480ae3404ec3de4581 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Tue, 30 Jun 2009 11:41:32 -0700
Subject: MAINTAINERS: STARFIRE/DURALAN update

Ion's cs.columbia.edu email address no longer works.

Signed-off-by: Joe Perches <joe@perches.com>
Acked-by: Ion Badulescu <ionut@badula.org>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 MAINTAINERS | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 92fe0796cd1b..6a89e312a7b4 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5576,8 +5576,8 @@ F:	drivers/staging/
 
 STARFIRE/DURALAN NETWORK DRIVER
 P:	Ion Badulescu
-M:	ionut@cs.columbia.edu
-S:	Maintained
+M:	ionut@badula.org
+S:	Odd Fixes
 F:	drivers/net/starfire*
 
 STARMODE RADIO IP (STRIP) PROTOCOL DRIVER
-- 
cgit v1.2.3-59-g8ed1b


From df279ca8966c3de83105428e3391ab17690802a9 Mon Sep 17 00:00:00 2001
From: Renaud Lottiaux <renaud.lottiaux@kerlabs.com>
Date: Tue, 30 Jun 2009 11:41:34 -0700
Subject: bsdacct: fix access to invalid filp in acct_on()

The file opened in acct_on and freshly stored in the ns->bacct struct can
be closed in acct_file_reopen by a concurrent call after we release
acct_lock and before we call mntput(file->f_path.mnt).

Record file->f_path.mnt in a local variable and use this variable only.

Signed-off-by: Renaud Lottiaux <renaud.lottiaux@kerlabs.com>
Signed-off-by: Louis Rilling <louis.rilling@kerlabs.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/acct.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/kernel/acct.c b/kernel/acct.c
index 7afa31564162..9f3391090b3e 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -215,6 +215,7 @@ static void acct_file_reopen(struct bsd_acct_struct *acct, struct file *file,
 static int acct_on(char *name)
 {
 	struct file *file;
+	struct vfsmount *mnt;
 	int error;
 	struct pid_namespace *ns;
 	struct bsd_acct_struct *acct = NULL;
@@ -256,11 +257,12 @@ static int acct_on(char *name)
 		acct = NULL;
 	}
 
-	mnt_pin(file->f_path.mnt);
+	mnt = file->f_path.mnt;
+	mnt_pin(mnt);
 	acct_file_reopen(ns->bacct, file, ns);
 	spin_unlock(&acct_lock);
 
-	mntput(file->f_path.mnt); /* it's pinned, now give up active reference */
+	mntput(mnt); /* it's pinned, now give up active reference */
 	kfree(acct);
 
 	return 0;
-- 
cgit v1.2.3-59-g8ed1b


From d7831a0bdf06b9f722b947bb0c205ff7d77cebd8 Mon Sep 17 00:00:00 2001
From: Richard Kennedy <richard@rsk.demon.co.uk>
Date: Tue, 30 Jun 2009 11:41:35 -0700
Subject: mm: prevent balance_dirty_pages() from doing too much work

balance_dirty_pages can overreact and move all of the dirty pages to
writeback unnecessarily.

balance_dirty_pages makes its decision to throttle based on the number of
dirty plus writeback pages that are over the calculated limit,so it will
continue to move pages even when there are plenty of pages in writeback
and less than the threshold still dirty.

This allows it to overshoot its limits and move all the dirty pages to
writeback while waiting for the drives to catch up and empty the writeback
list.

A simple fio test easily demonstrates this problem.

fio --name=f1 --directory=/disk1 --size=2G -rw=write --name=f2 --directory=/disk2 --size=1G --rw=write --startdelay=10

This is the simplest fix I could find, but I'm not entirely sure that it
alone will be enough for all cases.  But it certainly is an improvement on
my desktop machine writing to 2 disks.

Do we need something more for machines with large arrays where
bdi_threshold * number_of_drives is greater than the dirty_ratio ?

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/page-writeback.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 7b0dcea4935b..7687879253b9 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -541,8 +541,11 @@ static void balance_dirty_pages(struct address_space *mapping)
 		 * filesystems (i.e. NFS) in which data may have been
 		 * written to the server's write cache, but has not yet
 		 * been flushed to permanent storage.
+		 * Only move pages to writeback if this bdi is over its
+		 * threshold otherwise wait until the disk writes catch
+		 * up.
 		 */
-		if (bdi_nr_reclaimable) {
+		if (bdi_nr_reclaimable > bdi_thresh) {
 			writeback_inodes(&wbc);
 			pages_written += write_chunk - wbc.nr_to_write;
 			get_dirty_limits(&background_thresh, &dirty_thresh,
-- 
cgit v1.2.3-59-g8ed1b


From b37f2d4de6dfce4bfd6df311af80e4d61458ee1e Mon Sep 17 00:00:00 2001
From: Nikanth Karthikesan <knikanth@suse.de>
Date: Tue, 30 Jun 2009 11:41:36 -0700
Subject: cpusets: document adding/removing cpus to cpuset elaborately

By writing a tasks's pid to the file, a process adds that task to that
cgroup/cpuset.  But to add a cpu/mem to a cpuset, the new list of cpus
should be written to the cpuset.mems file which would replace the old list
of cpus.  Make this clearer in the documentation.

Signed-off-by: Nikanth Karthikesan <knikanth@suse.de>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/cgroups/cpusets.txt | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/Documentation/cgroups/cpusets.txt b/Documentation/cgroups/cpusets.txt
index f9ca389dddf4..1d7e9784439a 100644
--- a/Documentation/cgroups/cpusets.txt
+++ b/Documentation/cgroups/cpusets.txt
@@ -777,6 +777,18 @@ in cpuset directories:
 # /bin/echo 1-4 > cpus		-> set cpus list to cpus 1,2,3,4
 # /bin/echo 1,2,3,4 > cpus	-> set cpus list to cpus 1,2,3,4
 
+To add a CPU to a cpuset, write the new list of CPUs including the
+CPU to be added. To add 6 to the above cpuset:
+
+# /bin/echo 1-4,6 > cpus	-> set cpus list to cpus 1,2,3,4,6
+
+Similarly to remove a CPU from a cpuset, write the new list of CPUs
+without the CPU to be removed.
+
+To remove all the CPUs:
+
+# /bin/echo "" > cpus		-> clear cpus list
+
 2.3 Setting flags
 -----------------
 
-- 
cgit v1.2.3-59-g8ed1b


From 66918dcdf91ad101194c749c18099e836ba3de2b Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Tue, 30 Jun 2009 11:41:37 -0700
Subject: x86: only clear node_states for 64bit

Nathan reported that

| commit 73d60b7f747176dbdff826c4127d22e1fd3f9f74
| Author: Yinghai Lu <yinghai@kernel.org>
| Date:   Tue Jun 16 15:33:00 2009 -0700
|
|    page-allocator: clear N_HIGH_MEMORY map before we set it again
|
|    SRAT tables may contains nodes of very small size.  The arch code may
|    decide to not activate such a node.  However, currently the early boot
|    code sets N_HIGH_MEMORY for such nodes.  These nodes therefore seem to be
|    active although these nodes have no present pages.
|
|    For 64bit N_HIGH_MEMORY == N_NORMAL_MEMORY, so that works for 64 bit too

unintentionally and incorrectly clears the cpuset.mems cgroup attribute on
an i386 kvm guest, meaning that cpuset.mems can not be used.

Fix this by only clearing node_states[N_NORMAL_MEMORY] for 64bit only.
and need to do save/restore for that in find_zone_movable_pfn

Reported-by: Nathan Lynch <ntl@pobox.com>
Tested-by: Nathan Lynch <ntl@pobox.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@elte.hu>,
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/mm/init_64.c |  2 ++
 mm/page_alloc.c       | 13 +++++++------
 2 files changed, 9 insertions(+), 6 deletions(-)

diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index c4378f4fd4a5..b177652251a4 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -598,6 +598,8 @@ void __init paging_init(void)
 
 	sparse_memory_present_with_active_regions(MAX_NUMNODES);
 	sparse_init();
+	/* clear the default setting with node 0 */
+	nodes_clear(node_states[N_NORMAL_MEMORY]);
 	free_area_init_nodes(max_zone_pfns);
 }
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 5d714f8fb303..e0f2cdf9d8b1 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -4032,6 +4032,8 @@ static void __init find_zone_movable_pfns_for_nodes(unsigned long *movable_pfn)
 	int i, nid;
 	unsigned long usable_startpfn;
 	unsigned long kernelcore_node, kernelcore_remaining;
+	/* save the state before borrow the nodemask */
+	nodemask_t saved_node_state = node_states[N_HIGH_MEMORY];
 	unsigned long totalpages = early_calculate_totalpages();
 	int usable_nodes = nodes_weight(node_states[N_HIGH_MEMORY]);
 
@@ -4059,7 +4061,7 @@ static void __init find_zone_movable_pfns_for_nodes(unsigned long *movable_pfn)
 
 	/* If kernelcore was not specified, there is no ZONE_MOVABLE */
 	if (!required_kernelcore)
-		return;
+		goto out;
 
 	/* usable_startpfn is the lowest possible pfn ZONE_MOVABLE can be at */
 	find_usable_zone_for_movable();
@@ -4158,6 +4160,10 @@ restart:
 	for (nid = 0; nid < MAX_NUMNODES; nid++)
 		zone_movable_pfn[nid] =
 			roundup(zone_movable_pfn[nid], MAX_ORDER_NR_PAGES);
+
+out:
+	/* restore the node_state */
+	node_states[N_HIGH_MEMORY] = saved_node_state;
 }
 
 /* Any regular memory on that node ? */
@@ -4242,11 +4248,6 @@ void __init free_area_init_nodes(unsigned long *max_zone_pfn)
 						early_node_map[i].start_pfn,
 						early_node_map[i].end_pfn);
 
-	/*
-	 * find_zone_movable_pfns_for_nodes/early_calculate_totalpages init
-	 * that node_mask, clear it at first
-	 */
-	nodes_clear(node_states[N_HIGH_MEMORY]);
 	/* Initialise every node */
 	mminit_verify_pageflags_layout();
 	setup_nr_node_ids();
-- 
cgit v1.2.3-59-g8ed1b


From 79d7f4ee23d41571d9e4663521b5e6604c55729a Mon Sep 17 00:00:00 2001
From: Baruch Siach <baruch@tkos.co.il>
Date: Tue, 30 Jun 2009 11:41:38 -0700
Subject: gpio: pl061: fix probe error handling code

Note that IRQ has not been initialized when kmalloc() fails.

Also, use DECLARE_BITMAP() to make the code clearer.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/gpio/pl061.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpio/pl061.c b/drivers/gpio/pl061.c
index aa8e7cb020d9..80e483986699 100644
--- a/drivers/gpio/pl061.c
+++ b/drivers/gpio/pl061.c
@@ -221,7 +221,7 @@ static int __init pl061_probe(struct amba_device *dev, struct amba_id *id)
 	struct pl061_gpio *chip;
 	struct list_head *chip_list;
 	int ret, irq, i;
-	static unsigned long init_irq[BITS_TO_LONGS(NR_IRQS)];
+	static DECLARE_BITMAP(init_irq, NR_IRQS);
 
 	pdata = dev->dev.platform_data;
 	if (pdata == NULL)
@@ -280,6 +280,7 @@ static int __init pl061_probe(struct amba_device *dev, struct amba_id *id)
 	if (!test_and_set_bit(irq, init_irq)) { /* list initialized? */
 		chip_list = kmalloc(sizeof(*chip_list), GFP_KERNEL);
 		if (chip_list == NULL) {
+			clear_bit(irq, init_irq);
 			ret = -ENOMEM;
 			goto iounmap;
 		}
-- 
cgit v1.2.3-59-g8ed1b


From 50efacf6711e6c75595afd9b92aa15c1e4f7c79d Mon Sep 17 00:00:00 2001
From: Baruch Siach <baruch@tkos.co.il>
Date: Tue, 30 Jun 2009 11:41:39 -0700
Subject: gpio: pl061: fix IRQ handling for GPIOs >= PL061_GPIO_NR

IRQ handling is wrong for any GPIO >= PL061_GPIO_NR.

Fix this by implementing and using a proper .to_irq method.

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/gpio/pl061.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/drivers/gpio/pl061.c b/drivers/gpio/pl061.c
index 80e483986699..4ee4c8367a3f 100644
--- a/drivers/gpio/pl061.c
+++ b/drivers/gpio/pl061.c
@@ -109,6 +109,16 @@ static void pl061_set_value(struct gpio_chip *gc, unsigned offset, int value)
 	writeb(!!value << offset, chip->base + (1 << (offset + 2)));
 }
 
+static int pl061_to_irq(struct gpio_chip *gc, unsigned offset)
+{
+	struct pl061_gpio *chip = container_of(gc, struct pl061_gpio, gc);
+
+	if (chip->irq_base == (unsigned) -1)
+		return -EINVAL;
+
+	return chip->irq_base + offset;
+}
+
 /*
  * PL061 GPIO IRQ
  */
@@ -200,7 +210,7 @@ static void pl061_irq_handler(unsigned irq, struct irq_desc *desc)
 	desc->chip->ack(irq);
 	list_for_each(ptr, chip_list) {
 		unsigned long pending;
-		int gpio;
+		int offset;
 
 		chip = list_entry(ptr, struct pl061_gpio, list);
 		pending = readb(chip->base + GPIOMIS);
@@ -209,8 +219,8 @@ static void pl061_irq_handler(unsigned irq, struct irq_desc *desc)
 		if (pending == 0)
 			continue;
 
-		for_each_bit(gpio, &pending, PL061_GPIO_NR)
-			generic_handle_irq(gpio_to_irq(gpio));
+		for_each_bit(offset, &pending, PL061_GPIO_NR)
+			generic_handle_irq(pl061_to_irq(&chip->gc, offset));
 	}
 	desc->chip->unmask(irq);
 }
@@ -251,6 +261,7 @@ static int __init pl061_probe(struct amba_device *dev, struct amba_id *id)
 	chip->gc.direction_output = pl061_direction_output;
 	chip->gc.get = pl061_get_value;
 	chip->gc.set = pl061_set_value;
+	chip->gc.to_irq = pl061_to_irq;
 	chip->gc.base = pdata->gpio_base;
 	chip->gc.ngpio = PL061_GPIO_NR;
 	chip->gc.label = dev_name(&dev->dev);
-- 
cgit v1.2.3-59-g8ed1b


From eafad22a05fdaca60f06433ffe8810aaa920d539 Mon Sep 17 00:00:00 2001
From: Ville Syrjala <syrjala@sci.fi>
Date: Tue, 30 Jun 2009 11:41:40 -0700
Subject: atyfb: fix HP OmniBook 500 reboot hang

Apparently HP OmniBook 500's BIOS doesn't like the way atyfb reprograms
the hardware. The BIOS will simply hang after a reboot. Fix the problem
by restoring the hardware to it's original state on reboot.

Signed-off-by: Ville Syrjala <syrjala@sci.fi>
Cc: Mikulas Patocka <mpatocka@redhat.com>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/aty/atyfb.h      |  2 +
 drivers/video/aty/atyfb_base.c | 89 +++++++++++++++++++++++++++++++++++++-----
 2 files changed, 82 insertions(+), 9 deletions(-)

diff --git a/drivers/video/aty/atyfb.h b/drivers/video/aty/atyfb.h
index 7691e73823d3..0369653b5d88 100644
--- a/drivers/video/aty/atyfb.h
+++ b/drivers/video/aty/atyfb.h
@@ -187,6 +187,8 @@ struct atyfb_par {
 	int mtrr_reg;
 #endif
 	u32 mem_cntl;
+	struct crtc saved_crtc;
+	union aty_pll saved_pll;
 };
 
     /*
diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c
index 1207c208a30b..06782906daf5 100644
--- a/drivers/video/aty/atyfb_base.c
+++ b/drivers/video/aty/atyfb_base.c
@@ -66,6 +66,8 @@
 #include <linux/spinlock.h>
 #include <linux/wait.h>
 #include <linux/backlight.h>
+#include <linux/reboot.h>
+#include <linux/dmi.h>
 
 #include <asm/io.h>
 #include <linux/uaccess.h>
@@ -249,8 +251,6 @@ static int aty_init(struct fb_info *info);
 static int store_video_par(char *videopar, unsigned char m64_num);
 #endif
 
-static struct crtc saved_crtc;
-static union aty_pll saved_pll;
 static void aty_get_crtc(const struct atyfb_par *par, struct crtc *crtc);
 
 static void aty_set_crtc(const struct atyfb_par *par, const struct crtc *crtc);
@@ -261,6 +261,8 @@ static void set_off_pitch(struct atyfb_par *par, const struct fb_info *info);
 static int read_aty_sense(const struct atyfb_par *par);
 #endif
 
+static DEFINE_MUTEX(reboot_lock);
+static struct fb_info *reboot_info;
 
     /*
      *  Interface used by the world
@@ -2390,9 +2392,9 @@ static int __devinit aty_init(struct fb_info *info)
 #endif /* CONFIG_FB_ATY_CT */
 
 	/* save previous video mode */
-	aty_get_crtc(par, &saved_crtc);
+	aty_get_crtc(par, &par->saved_crtc);
 	if(par->pll_ops->get_pll)
-		par->pll_ops->get_pll(info, &saved_pll);
+		par->pll_ops->get_pll(info, &par->saved_pll);
 
 	par->mem_cntl = aty_ld_le32(MEM_CNTL, par);
 	gtb_memsize = M64_HAS(GTB_DSP);
@@ -2667,8 +2669,8 @@ static int __devinit aty_init(struct fb_info *info)
 
 aty_init_exit:
 	/* restore video mode */
-	aty_set_crtc(par, &saved_crtc);
-	par->pll_ops->set_pll(info, &saved_pll);
+	aty_set_crtc(par, &par->saved_crtc);
+	par->pll_ops->set_pll(info, &par->saved_pll);
 
 #ifdef CONFIG_MTRR
 	if (par->mtrr_reg >= 0) {
@@ -3502,6 +3504,11 @@ static int __devinit atyfb_pci_probe(struct pci_dev *pdev, const struct pci_devi
 	par->mmap_map[1].prot_flag = _PAGE_E;
 #endif /* __sparc__ */
 
+	mutex_lock(&reboot_lock);
+	if (!reboot_info)
+		reboot_info = info;
+	mutex_unlock(&reboot_lock);
+
 	return 0;
 
 err_release_io:
@@ -3614,8 +3621,8 @@ static void __devexit atyfb_remove(struct fb_info *info)
 	struct atyfb_par *par = (struct atyfb_par *) info->par;
 
 	/* restore video mode */
-	aty_set_crtc(par, &saved_crtc);
-	par->pll_ops->set_pll(info, &saved_pll);
+	aty_set_crtc(par, &par->saved_crtc);
+	par->pll_ops->set_pll(info, &par->saved_pll);
 
 	unregister_framebuffer(info);
 
@@ -3661,6 +3668,11 @@ static void __devexit atyfb_pci_remove(struct pci_dev *pdev)
 {
 	struct fb_info *info = pci_get_drvdata(pdev);
 
+	mutex_lock(&reboot_lock);
+	if (reboot_info == info)
+		reboot_info = NULL;
+	mutex_unlock(&reboot_lock);
+
 	atyfb_remove(info);
 }
 
@@ -3808,6 +3820,56 @@ static int __init atyfb_setup(char *options)
 }
 #endif  /*  MODULE  */
 
+static int atyfb_reboot_notify(struct notifier_block *nb,
+			       unsigned long code, void *unused)
+{
+	struct atyfb_par *par;
+
+	if (code != SYS_RESTART)
+		return NOTIFY_DONE;
+
+	mutex_lock(&reboot_lock);
+
+	if (!reboot_info)
+		goto out;
+
+	if (!lock_fb_info(reboot_info))
+		goto out;
+
+	par = reboot_info->par;
+
+	/*
+	 * HP OmniBook 500's BIOS doesn't like the state of the
+	 * hardware after atyfb has been used. Restore the hardware
+	 * to the original state to allow successful reboots.
+	 */
+	aty_set_crtc(par, &par->saved_crtc);
+	par->pll_ops->set_pll(reboot_info, &par->saved_pll);
+
+	unlock_fb_info(reboot_info);
+ out:
+	mutex_unlock(&reboot_lock);
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block atyfb_reboot_notifier = {
+	.notifier_call = atyfb_reboot_notify,
+};
+
+static const struct dmi_system_id atyfb_reboot_ids[] = {
+	{
+		.ident = "HP OmniBook 500",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "HP OmniBook PC"),
+			DMI_MATCH(DMI_PRODUCT_VERSION, "HP OmniBook 500 FA"),
+		},
+	},
+
+	{ }
+};
+
 static int __init atyfb_init(void)
 {
     int err1 = 1, err2 = 1;
@@ -3826,11 +3888,20 @@ static int __init atyfb_init(void)
     err2 = atyfb_atari_probe();
 #endif
 
-    return (err1 && err2) ? -ENODEV : 0;
+    if (err1 && err2)
+	return -ENODEV;
+
+    if (dmi_check_system(atyfb_reboot_ids))
+	register_reboot_notifier(&atyfb_reboot_notifier);
+
+    return 0;
 }
 
 static void __exit atyfb_exit(void)
 {
+	if (dmi_check_system(atyfb_reboot_ids))
+		unregister_reboot_notifier(&atyfb_reboot_notifier);
+
 #ifdef CONFIG_PCI
 	pci_unregister_driver(&atyfb_driver);
 #endif
-- 
cgit v1.2.3-59-g8ed1b


From ee905d0c58a440a5bd10c845e8305f6f7f706be2 Mon Sep 17 00:00:00 2001
From: Ville Syrjala <syrjala@sci.fi>
Date: Tue, 30 Jun 2009 11:41:42 -0700
Subject: atyfb: fix alignment for block writes

Block writes require 64 byte alignment.  Since block writes could be used
with SGRAM or WRAM also refine the memory type detection to check for
either type before deciding to use the 64 byte alignment.

Signed-off-by: Ville Syrjala <syrjala@sci.fi>
Tested-by: Mikulas Patocka <mpatocka@redhat.com>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/aty/atyfb.h        |  1 +
 drivers/video/aty/atyfb_base.c   | 52 ++++++++++++++++++++++++++++++----------
 drivers/video/aty/mach64_accel.c |  7 ++++--
 3 files changed, 46 insertions(+), 14 deletions(-)

diff --git a/drivers/video/aty/atyfb.h b/drivers/video/aty/atyfb.h
index 0369653b5d88..1f39a62f899b 100644
--- a/drivers/video/aty/atyfb.h
+++ b/drivers/video/aty/atyfb.h
@@ -219,6 +219,7 @@ struct atyfb_par {
 #define M64F_XL_DLL		0x00080000
 #define M64F_MFB_FORCE_4	0x00100000
 #define M64F_HW_TRIPLE		0x00200000
+#define M64F_XL_MEM		0x00400000
     /*
      *  Register access
      */
diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c
index 06782906daf5..63d3739d43a8 100644
--- a/drivers/video/aty/atyfb_base.c
+++ b/drivers/video/aty/atyfb_base.c
@@ -363,8 +363,8 @@ static unsigned long phys_guiregbase[FB_MAX] __devinitdata = { 0, };
 #define ATI_CHIP_264GTPRO  (ATI_MODERN_SET | M64F_SDRAM_MAGIC_PLL | M64F_HW_TRIPLE | M64F_FIFO_32 | M64F_RESET_3D)
 #define ATI_CHIP_264LTPRO  (ATI_MODERN_SET | M64F_HW_TRIPLE | M64F_FIFO_32 | M64F_RESET_3D)
 
-#define ATI_CHIP_264XL     (ATI_MODERN_SET | M64F_HW_TRIPLE | M64F_FIFO_32 | M64F_RESET_3D | M64F_XL_DLL | M64F_MFB_FORCE_4)
-#define ATI_CHIP_MOBILITY  (ATI_MODERN_SET | M64F_HW_TRIPLE | M64F_FIFO_32 | M64F_RESET_3D | M64F_XL_DLL | M64F_MFB_FORCE_4 | M64F_MOBIL_BUS)
+#define ATI_CHIP_264XL     (ATI_MODERN_SET | M64F_HW_TRIPLE | M64F_FIFO_32 | M64F_RESET_3D | M64F_XL_DLL | M64F_MFB_FORCE_4 | M64F_XL_MEM)
+#define ATI_CHIP_MOBILITY  (ATI_MODERN_SET | M64F_HW_TRIPLE | M64F_FIFO_32 | M64F_RESET_3D | M64F_XL_DLL | M64F_MFB_FORCE_4 | M64F_XL_MEM | M64F_MOBIL_BUS)
 
 static struct {
 	u16 pci_id;
@@ -541,6 +541,7 @@ static char ram_edo[] __devinitdata = "EDO";
 static char ram_sdram[] __devinitdata = "SDRAM (1:1)";
 static char ram_sgram[] __devinitdata = "SGRAM (1:1)";
 static char ram_sdram32[] __devinitdata = "SDRAM (2:1) (32-bit)";
+static char ram_wram[] __devinitdata = "WRAM";
 static char ram_off[] __devinitdata = "OFF";
 #endif /* CONFIG_FB_ATY_CT */
 
@@ -554,6 +555,10 @@ static char *aty_gx_ram[8] __devinitdata = {
 
 #ifdef CONFIG_FB_ATY_CT
 static char *aty_ct_ram[8] __devinitdata = {
+	ram_off, ram_dram, ram_edo, ram_edo,
+	ram_sdram, ram_sgram, ram_wram, ram_resv
+};
+static char *aty_xl_ram[8] __devinitdata = {
 	ram_off, ram_dram, ram_edo, ram_edo,
 	ram_sdram, ram_sgram, ram_sdram32, ram_resv
 };
@@ -762,6 +767,17 @@ static void aty_set_crtc(const struct atyfb_par *par, const struct crtc *crtc)
 #endif /* CONFIG_FB_ATY_GENERIC_LCD */
 }
 
+static u32 calc_line_length(struct atyfb_par *par, u32 vxres, u32 bpp)
+{
+	u32 line_length = vxres * bpp / 8;
+
+	if (par->ram_type == SGRAM ||
+	    (!M64_HAS(XL_MEM) && par->ram_type == WRAM))
+		line_length = (line_length + 63) & ~63;
+
+	return line_length;
+}
+
 static int aty_var_to_crtc(const struct fb_info *info,
 	const struct fb_var_screeninfo *var, struct crtc *crtc)
 {
@@ -771,13 +787,14 @@ static int aty_var_to_crtc(const struct fb_info *info,
 	u32 h_total, h_disp, h_sync_strt, h_sync_end, h_sync_dly, h_sync_wid, h_sync_pol;
 	u32 v_total, v_disp, v_sync_strt, v_sync_end, v_sync_wid, v_sync_pol, c_sync;
 	u32 pix_width, dp_pix_width, dp_chain_mask;
+	u32 line_length;
 
 	/* input */
-	xres = var->xres;
+	xres = (var->xres + 7) & ~7;
 	yres = var->yres;
-	vxres = var->xres_virtual;
+	vxres = (var->xres_virtual + 7) & ~7;
 	vyres = var->yres_virtual;
-	xoffset = var->xoffset;
+	xoffset = (var->xoffset + 7) & ~7;
 	yoffset = var->yoffset;
 	bpp = var->bits_per_pixel;
 	if (bpp == 16)
@@ -829,7 +846,9 @@ static int aty_var_to_crtc(const struct fb_info *info,
 	} else
 		FAIL("invalid bpp");
 
-	if (vxres * vyres * bpp / 8 > info->fix.smem_len)
+	line_length = calc_line_length(par, vxres, bpp);
+
+	if (vyres * line_length > info->fix.smem_len)
 		FAIL("not enough video RAM");
 
 	h_sync_pol = sync & FB_SYNC_HOR_HIGH_ACT ? 0 : 1;
@@ -971,7 +990,9 @@ static int aty_var_to_crtc(const struct fb_info *info,
 	crtc->xoffset = xoffset;
 	crtc->yoffset = yoffset;
 	crtc->bpp = bpp;
-	crtc->off_pitch = ((yoffset*vxres+xoffset)*bpp/64) | (vxres<<19);
+	crtc->off_pitch =
+		((yoffset * line_length + xoffset * bpp / 8) / 8) |
+		((line_length / bpp) << 22);
 	crtc->vline_crnt_vline = 0;
 
 	crtc->h_tot_disp = h_total | (h_disp<<16);
@@ -1396,7 +1417,9 @@ static int atyfb_set_par(struct fb_info *info)
 	}
 	aty_st_8(DAC_MASK, 0xff, par);
 
-	info->fix.line_length = var->xres_virtual * var->bits_per_pixel/8;
+	info->fix.line_length = calc_line_length(par, var->xres_virtual,
+						 var->bits_per_pixel);
+
 	info->fix.visual = var->bits_per_pixel <= 8 ?
 		FB_VISUAL_PSEUDOCOLOR : FB_VISUAL_DIRECTCOLOR;
 
@@ -1507,10 +1530,12 @@ static void set_off_pitch(struct atyfb_par *par, const struct fb_info *info)
 {
 	u32 xoffset = info->var.xoffset;
 	u32 yoffset = info->var.yoffset;
-	u32 vxres = par->crtc.vxres;
+	u32 line_length = info->fix.line_length;
 	u32 bpp = info->var.bits_per_pixel;
 
-	par->crtc.off_pitch = ((yoffset * vxres + xoffset) * bpp / 64) | (vxres << 19);
+	par->crtc.off_pitch =
+		((yoffset * line_length + xoffset * bpp / 8) / 8) |
+		((line_length / bpp) << 22);
 }
 
 
@@ -2203,7 +2228,7 @@ static void __devinit aty_calc_mem_refresh(struct atyfb_par *par, int xclk)
 	const int *refresh_tbl;
 	int i, size;
 
-	if (IS_XL(par->pci_id) || IS_MOBILITY(par->pci_id)) {
+	if (M64_HAS(XL_MEM)) {
 		refresh_tbl = ragexl_tbl;
 		size = ARRAY_SIZE(ragexl_tbl);
 	} else {
@@ -2337,7 +2362,10 @@ static int __devinit aty_init(struct fb_info *info)
 		par->pll_ops = &aty_pll_ct;
 		par->bus_type = PCI;
 		par->ram_type = (aty_ld_le32(CNFG_STAT0, par) & 0x07);
-		ramname = aty_ct_ram[par->ram_type];
+		if (M64_HAS(XL_MEM))
+			ramname = aty_xl_ram[par->ram_type];
+		else
+			ramname = aty_ct_ram[par->ram_type];
 		/* for many chips, the mclk is 67 MHz for SDRAM, 63 MHz otherwise */
 		if (par->pll_limits.mclk == 67 && par->ram_type < SDRAM)
 			par->pll_limits.mclk = 63;
diff --git a/drivers/video/aty/mach64_accel.c b/drivers/video/aty/mach64_accel.c
index 0cc9724e61a2..51fcc0a2c94a 100644
--- a/drivers/video/aty/mach64_accel.c
+++ b/drivers/video/aty/mach64_accel.c
@@ -63,14 +63,17 @@ static void reset_GTC_3D_engine(const struct atyfb_par *par)
 void aty_init_engine(struct atyfb_par *par, struct fb_info *info)
 {
 	u32 pitch_value;
+	u32 vxres;
 
 	/* determine modal information from global mode structure */
-	pitch_value = info->var.xres_virtual;
+	pitch_value = info->fix.line_length / (info->var.bits_per_pixel / 8);
+	vxres = info->var.xres_virtual;
 
 	if (info->var.bits_per_pixel == 24) {
 		/* In 24 bpp, the engine is in 8 bpp - this requires that all */
 		/* horizontal coordinates and widths must be adjusted */
 		pitch_value *= 3;
+		vxres *= 3;
 	}
 
 	/* On GTC (RagePro), we need to reset the 3D engine before */
@@ -133,7 +136,7 @@ void aty_init_engine(struct atyfb_par *par, struct fb_info *info)
 	aty_st_le32(SC_LEFT, 0, par);
 	aty_st_le32(SC_TOP, 0, par);
 	aty_st_le32(SC_BOTTOM, par->crtc.vyres - 1, par);
-	aty_st_le32(SC_RIGHT, pitch_value - 1, par);
+	aty_st_le32(SC_RIGHT, vxres - 1, par);
 
 	/* set background color to minimum value (usually BLACK) */
 	aty_st_le32(DP_BKGD_CLR, 0, par);
-- 
cgit v1.2.3-59-g8ed1b


From 9980060bad5607ca6db7fb8683de671b522e56a4 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Tue, 30 Jun 2009 11:41:43 -0700
Subject: bfin: delay IRQ registration until driver is ready

Make sure we do not actually request the RTC IRQ until the device driver
is fully ready to handle and process any interrupt.  This way a spurious
interrupt won't crash the system (which may happen if the bootloader was
poking the RTC right before booting Linux).

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/rtc-bfin.c | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/drivers/rtc/rtc-bfin.c b/drivers/rtc/rtc-bfin.c
index aafd3e6ebb0d..a118eb0f1e67 100644
--- a/drivers/rtc/rtc-bfin.c
+++ b/drivers/rtc/rtc-bfin.c
@@ -1,8 +1,8 @@
 /*
  * Blackfin On-Chip Real Time Clock Driver
- *  Supports BF52[257]/BF53[123]/BF53[467]/BF54[24789]
+ *  Supports BF51x/BF52x/BF53[123]/BF53[467]/BF54x
  *
- * Copyright 2004-2008 Analog Devices Inc.
+ * Copyright 2004-2009 Analog Devices Inc.
  *
  * Enter bugs at http://blackfin.uclinux.org/
  *
@@ -363,7 +363,7 @@ static int __devinit bfin_rtc_probe(struct platform_device *pdev)
 	struct bfin_rtc *rtc;
 	struct device *dev = &pdev->dev;
 	int ret = 0;
-	unsigned long timeout;
+	unsigned long timeout = jiffies + HZ;
 
 	dev_dbg_stamp(dev);
 
@@ -374,32 +374,32 @@ static int __devinit bfin_rtc_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, rtc);
 	device_init_wakeup(dev, 1);
 
+	/* Register our RTC with the RTC framework */
+	rtc->rtc_dev = rtc_device_register(pdev->name, dev, &bfin_rtc_ops,
+						THIS_MODULE);
+	if (unlikely(IS_ERR(rtc->rtc_dev))) {
+		ret = PTR_ERR(rtc->rtc_dev);
+		goto err;
+	}
+
 	/* Grab the IRQ and init the hardware */
 	ret = request_irq(IRQ_RTC, bfin_rtc_interrupt, IRQF_SHARED, pdev->name, dev);
 	if (unlikely(ret))
-		goto err;
+		goto err_reg;
 	/* sometimes the bootloader touched things, but the write complete was not
 	 * enabled, so let's just do a quick timeout here since the IRQ will not fire ...
 	 */
-	timeout = jiffies + HZ;
 	while (bfin_read_RTC_ISTAT() & RTC_ISTAT_WRITE_PENDING)
 		if (time_after(jiffies, timeout))
 			break;
 	bfin_rtc_reset(dev, RTC_ISTAT_WRITE_COMPLETE);
 	bfin_write_RTC_SWCNT(0);
 
-	/* Register our RTC with the RTC framework */
-	rtc->rtc_dev = rtc_device_register(pdev->name, dev, &bfin_rtc_ops, THIS_MODULE);
-	if (unlikely(IS_ERR(rtc->rtc_dev))) {
-		ret = PTR_ERR(rtc->rtc_dev);
-		goto err_irq;
-	}
-
 	return 0;
 
- err_irq:
-	free_irq(IRQ_RTC, dev);
- err:
+err_reg:
+	rtc_device_unregister(rtc->rtc_dev);
+err:
 	kfree(rtc);
 	return ret;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 8516a500029890a72622d245f8ed32c4e30969b7 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Tue, 30 Jun 2009 11:41:44 -0700
Subject: floppy: fix lock imbalance

A crappy macro prevents us unlocking on a fail path.

Expand the macro and unlock appropriatelly.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/block/floppy.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index 862b40c90181..91b753013780 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -3327,7 +3327,10 @@ static inline int set_geometry(unsigned int cmd, struct floppy_struct *g,
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
 		mutex_lock(&open_lock);
-		LOCK_FDC(drive, 1);
+		if (lock_fdc(drive, 1)) {
+			mutex_unlock(&open_lock);
+			return -EINTR;
+		}
 		floppy_type[type] = *g;
 		floppy_type[type].name = "user format";
 		for (cnt = type << 2; cnt < (type << 2) + 4; cnt++)
-- 
cgit v1.2.3-59-g8ed1b


From 752fa51e4c5182c3c257f1cede90577a7e213c58 Mon Sep 17 00:00:00 2001
From: Wolfgang Illmeyer <wolfgang@illmeyer.com>
Date: Tue, 30 Jun 2009 11:41:44 -0700
Subject: hostfs: set maximum filesize in superblock for proper LFS support

Maximum file size for hostfs mounts defaults to 2GB, so bigger files cannot be
read/written through hostfs. This patch initializes the maximum file size to
MAX_LFS_SIZE.

Addresses http://bugzilla.kernel.org/show_bug.cgi?id=13531

Signed-off-by: Wolfgang Illmeyer <wolfgang@illmeyer.com>
Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/hostfs/hostfs_kern.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index fe02ad4740e7..032604e5ef2c 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -972,6 +972,7 @@ static int hostfs_fill_sb_common(struct super_block *sb, void *d, int silent)
 	sb->s_blocksize_bits = 10;
 	sb->s_magic = HOSTFS_SUPER_MAGIC;
 	sb->s_op = &hostfs_sbops;
+	sb->s_maxbytes = MAX_LFS_FILESIZE;
 
 	/* NULL is printed as <NULL> by sprintf: avoid that. */
 	if (req_root == NULL)
-- 
cgit v1.2.3-59-g8ed1b


From aee3ff1b413cff44e7d91dd1901cacd8988ce9cf Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 30 Jun 2009 22:24:54 +0100
Subject: FRV: Wire up new syscalls

Wire up new syscalls rt_tgsigqueueinfo and perf_counter_open.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/frv/include/asm/unistd.h | 4 +++-
 arch/frv/kernel/entry.S       | 2 ++
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/arch/frv/include/asm/unistd.h b/arch/frv/include/asm/unistd.h
index 96d78d5d2c41..4a8fb427ce0a 100644
--- a/arch/frv/include/asm/unistd.h
+++ b/arch/frv/include/asm/unistd.h
@@ -341,10 +341,12 @@
 #define __NR_inotify_init1	332
 #define __NR_preadv		333
 #define __NR_pwritev		334
+#define __NR_rt_tgsigqueueinfo	335
+#define __NR_perf_counter_open	336
 
 #ifdef __KERNEL__
 
-#define NR_syscalls 335
+#define NR_syscalls 337
 
 #define __ARCH_WANT_IPC_PARSE_VERSION
 /* #define __ARCH_WANT_OLD_READDIR */
diff --git a/arch/frv/kernel/entry.S b/arch/frv/kernel/entry.S
index 356e0e327a89..fde1e446b440 100644
--- a/arch/frv/kernel/entry.S
+++ b/arch/frv/kernel/entry.S
@@ -1524,5 +1524,7 @@ sys_call_table:
 	.long sys_inotify_init1
 	.long sys_preadv
 	.long sys_pwritev
+	.long sys_rt_tgsigqueueinfo	/* 335 */
+	.long sys_perf_counter_open
 
 syscall_table_size = (. - sys_call_table)
-- 
cgit v1.2.3-59-g8ed1b


From 6086071005674eb982d898c75269c931240154cf Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 30 Jun 2009 22:33:15 +0100
Subject: MN10300: Wire up new syscalls

Wire up new syscalls rt_tgsigqueueinfo and perf_counter_open.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/mn10300/include/asm/unistd.h | 4 +++-
 arch/mn10300/kernel/entry.S       | 2 ++
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/arch/mn10300/include/asm/unistd.h b/arch/mn10300/include/asm/unistd.h
index fef5b434dadc..fad68616af32 100644
--- a/arch/mn10300/include/asm/unistd.h
+++ b/arch/mn10300/include/asm/unistd.h
@@ -346,10 +346,12 @@
 #define __NR_inotify_init1	333
 #define __NR_preadv		334
 #define __NR_pwritev		335
+#define __NR_rt_tgsigqueueinfo	336
+#define __NR_perf_counter_open	337
 
 #ifdef __KERNEL__
 
-#define NR_syscalls 326
+#define NR_syscalls 338
 
 /*
  * specify the deprecated syscalls we want to support on this arch
diff --git a/arch/mn10300/kernel/entry.S b/arch/mn10300/kernel/entry.S
index 7408a27199f3..e0d2563af4f2 100644
--- a/arch/mn10300/kernel/entry.S
+++ b/arch/mn10300/kernel/entry.S
@@ -722,6 +722,8 @@ ENTRY(sys_call_table)
 	.long sys_inotify_init1
 	.long sys_preadv
 	.long sys_pwritev		/* 335 */
+	.long sys_rt_tgsigqueueinfo
+	.long sys_perf_counter_open
 
 
 nr_syscalls=(.-sys_call_table)/4
-- 
cgit v1.2.3-59-g8ed1b


From 1ec22eb2b4a2e1a763106bce36b11c02eaa84e61 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Wed, 1 Jul 2009 12:27:21 +1000
Subject: md: fix error path when duplicate name is found on md device
 creation.

When an md device is created by name (rather than number) we need to
check that the name is not already in use.  If this check finds a
duplicate, we return an error without dropping the lock or freeing
the newly create mddev.
This patch fixes that.

Cc: stable@kernel.org
Found-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/md.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 2166af8a7654..58bee2366ea8 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -3862,6 +3862,8 @@ static int md_alloc(dev_t dev, char *name)
 			if (mddev2->gendisk &&
 			    strcmp(mddev2->gendisk->disk_name, name) == 0) {
 				spin_unlock(&all_mddevs_lock);
+				mutex_unlock(&disks_mutex);
+				mddev_put(mddev);
 				return -EEXIST;
 			}
 		spin_unlock(&all_mddevs_lock);
-- 
cgit v1.2.3-59-g8ed1b


From 0909dc448c98ed5021c87ffdfc09fb473aa464ab Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Wed, 1 Jul 2009 12:27:21 +1000
Subject: md: tidy up error paths in md_alloc

As the recent bug in md_alloc showed, having a single exit path for
unlocking and putting is a good idea.  So restructure md_alloc to have
a single mutex_unlock and mddev_put, and use gotos where necessary.

Found-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/md.c | 38 ++++++++++++++++++--------------------
 1 file changed, 18 insertions(+), 20 deletions(-)

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 58bee2366ea8..65fe35b5e34a 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -3846,11 +3846,9 @@ static int md_alloc(dev_t dev, char *name)
 	flush_scheduled_work();
 
 	mutex_lock(&disks_mutex);
-	if (mddev->gendisk) {
-		mutex_unlock(&disks_mutex);
-		mddev_put(mddev);
-		return -EEXIST;
-	}
+	error = -EEXIST;
+	if (mddev->gendisk)
+		goto abort;
 
 	if (name) {
 		/* Need to ensure that 'name' is not a duplicate.
@@ -3862,19 +3860,15 @@ static int md_alloc(dev_t dev, char *name)
 			if (mddev2->gendisk &&
 			    strcmp(mddev2->gendisk->disk_name, name) == 0) {
 				spin_unlock(&all_mddevs_lock);
-				mutex_unlock(&disks_mutex);
-				mddev_put(mddev);
-				return -EEXIST;
+				goto abort;
 			}
 		spin_unlock(&all_mddevs_lock);
 	}
 
+	error = -ENOMEM;
 	mddev->queue = blk_alloc_queue(GFP_KERNEL);
-	if (!mddev->queue) {
-		mutex_unlock(&disks_mutex);
-		mddev_put(mddev);
-		return -ENOMEM;
-	}
+	if (!mddev->queue)
+		goto abort;
 	mddev->queue->queuedata = mddev;
 
 	/* Can be unlocked because the queue is new: no concurrency */
@@ -3884,11 +3878,9 @@ static int md_alloc(dev_t dev, char *name)
 
 	disk = alloc_disk(1 << shift);
 	if (!disk) {
-		mutex_unlock(&disks_mutex);
 		blk_cleanup_queue(mddev->queue);
 		mddev->queue = NULL;
-		mddev_put(mddev);
-		return -ENOMEM;
+		goto abort;
 	}
 	disk->major = MAJOR(mddev->unit);
 	disk->first_minor = unit << shift;
@@ -3910,16 +3902,22 @@ static int md_alloc(dev_t dev, char *name)
 	mddev->gendisk = disk;
 	error = kobject_init_and_add(&mddev->kobj, &md_ktype,
 				     &disk_to_dev(disk)->kobj, "%s", "md");
-	mutex_unlock(&disks_mutex);
-	if (error)
+	if (error) {
+		/* This isn't possible, but as kobject_init_and_add is marked
+		 * __must_check, we must do something with the result
+		 */
 		printk(KERN_WARNING "md: cannot register %s/md - name in use\n",
 		       disk->disk_name);
-	else {
+		error = 0;
+	}
+ abort:
+	mutex_unlock(&disks_mutex);
+	if (!error) {
 		kobject_uevent(&mddev->kobj, KOBJ_ADD);
 		mddev->sysfs_state = sysfs_get_dirent(mddev->kobj.sd, "array_state");
 	}
 	mddev_put(mddev);
-	return 0;
+	return error;
 }
 
 static struct kobject *md_probe(dev_t dev, int *part, void *data)
-- 
cgit v1.2.3-59-g8ed1b


From eaea43abf30c8ccb447c190e7c94b46b5f75eae6 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 29 Jun 2009 16:49:40 +0000
Subject: cdc_eem: Use netdev stats structure

Now that netdev has its own stats structure we should use that
instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/usb/cdc_eem.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/usb/cdc_eem.c b/drivers/net/usb/cdc_eem.c
index 80e01778dd3b..cd35d50e46d4 100644
--- a/drivers/net/usb/cdc_eem.c
+++ b/drivers/net/usb/cdc_eem.c
@@ -319,7 +319,7 @@ static int eem_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 				return crc == crc2;
 
 			if (unlikely(crc != crc2)) {
-				dev->stats.rx_errors++;
+				dev->net->stats.rx_errors++;
 				dev_kfree_skb_any(skb2);
 			} else
 				usbnet_skb_return(dev, skb2);
-- 
cgit v1.2.3-59-g8ed1b


From 9612101cb33862cc160069cc8423926d61db51f8 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 29 Jun 2009 16:50:51 +0000
Subject: dm9601: Use netdev stats structure

Now that netdev has its own stats structure we should use that
instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/usb/dm9601.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/usb/dm9601.c b/drivers/net/usb/dm9601.c
index 7ae82446b93a..1d3730d6690f 100644
--- a/drivers/net/usb/dm9601.c
+++ b/drivers/net/usb/dm9601.c
@@ -513,11 +513,11 @@ static int dm9601_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 	len = (skb->data[1] | (skb->data[2] << 8)) - 4;
 
 	if (unlikely(status & 0xbf)) {
-		if (status & 0x01) dev->stats.rx_fifo_errors++;
-		if (status & 0x02) dev->stats.rx_crc_errors++;
-		if (status & 0x04) dev->stats.rx_frame_errors++;
-		if (status & 0x20) dev->stats.rx_missed_errors++;
-		if (status & 0x90) dev->stats.rx_length_errors++;
+		if (status & 0x01) dev->net->stats.rx_fifo_errors++;
+		if (status & 0x02) dev->net->stats.rx_crc_errors++;
+		if (status & 0x04) dev->net->stats.rx_frame_errors++;
+		if (status & 0x20) dev->net->stats.rx_missed_errors++;
+		if (status & 0x90) dev->net->stats.rx_length_errors++;
 		return 0;
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From a22d2b36a2c4ca58c5914072a88704377bbd34f8 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 29 Jun 2009 16:51:40 +0000
Subject: net1080: Use netdev stats structure

Now that netdev has its own stats structure we should use that
instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/usb/net1080.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/usb/net1080.c b/drivers/net/usb/net1080.c
index 034e8a73ca6b..aeb1ab03a9ee 100644
--- a/drivers/net/usb/net1080.c
+++ b/drivers/net/usb/net1080.c
@@ -433,7 +433,7 @@ static int net1080_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 		dbg("rx framesize %d range %d..%d mtu %d", skb->len,
 			net->hard_header_len, dev->hard_mtu, net->mtu);
 #endif
-		dev->stats.rx_frame_errors++;
+		dev->net->stats.rx_frame_errors++;
 		nc_ensure_sync(dev);
 		return 0;
 	}
@@ -442,12 +442,12 @@ static int net1080_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 	hdr_len = le16_to_cpup(&header->hdr_len);
 	packet_len = le16_to_cpup(&header->packet_len);
 	if (FRAMED_SIZE(packet_len) > NC_MAX_PACKET) {
-		dev->stats.rx_frame_errors++;
+		dev->net->stats.rx_frame_errors++;
 		dbg("packet too big, %d", packet_len);
 		nc_ensure_sync(dev);
 		return 0;
 	} else if (hdr_len < MIN_HEADER) {
-		dev->stats.rx_frame_errors++;
+		dev->net->stats.rx_frame_errors++;
 		dbg("header too short, %d", hdr_len);
 		nc_ensure_sync(dev);
 		return 0;
@@ -465,21 +465,21 @@ static int net1080_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 
 	if ((packet_len & 0x01) == 0) {
 		if (skb->data [packet_len] != PAD_BYTE) {
-			dev->stats.rx_frame_errors++;
+			dev->net->stats.rx_frame_errors++;
 			dbg("bad pad");
 			return 0;
 		}
 		skb_trim(skb, skb->len - 1);
 	}
 	if (skb->len != packet_len) {
-		dev->stats.rx_frame_errors++;
+		dev->net->stats.rx_frame_errors++;
 		dbg("bad packet len %d (expected %d)",
 			skb->len, packet_len);
 		nc_ensure_sync(dev);
 		return 0;
 	}
 	if (header->packet_id != get_unaligned(&trailer->packet_id)) {
-		dev->stats.rx_fifo_errors++;
+		dev->net->stats.rx_fifo_errors++;
 		dbg("(2+ dropped) rx packet_id mismatch 0x%x 0x%x",
 			le16_to_cpu(header->packet_id),
 			le16_to_cpu(trailer->packet_id));
-- 
cgit v1.2.3-59-g8ed1b


From 58e2e7d5913e7a2a6d87ef30d3b52e66b88e6e1d Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 29 Jun 2009 16:52:26 +0000
Subject: rndis_host: Use netdev stats structure

Now that netdev has its own stats structure we should use that
instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/usb/rndis_host.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/usb/rndis_host.c b/drivers/net/usb/rndis_host.c
index 1bf243ef950e..2232232b7989 100644
--- a/drivers/net/usb/rndis_host.c
+++ b/drivers/net/usb/rndis_host.c
@@ -487,7 +487,7 @@ int rndis_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 		if (unlikely(hdr->msg_type != RNDIS_MSG_PACKET
 				|| skb->len < msg_len
 				|| (data_offset + data_len + 8) > msg_len)) {
-			dev->stats.rx_frame_errors++;
+			dev->net->stats.rx_frame_errors++;
 			devdbg(dev, "bad rndis message %d/%d/%d/%d, len %d",
 				le32_to_cpu(hdr->msg_type),
 				msg_len, data_offset, data_len, skb->len);
-- 
cgit v1.2.3-59-g8ed1b


From 80667ac13a6cf2c3a3ff275a2a72809671299acb Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 29 Jun 2009 16:53:00 +0000
Subject: smsc95xx: Use netdev stats structure

Now that netdev has its own stats structure we should use that
instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/usb/smsc95xx.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/usb/smsc95xx.c b/drivers/net/usb/smsc95xx.c
index 89a91f8c22de..fe045896406b 100644
--- a/drivers/net/usb/smsc95xx.c
+++ b/drivers/net/usb/smsc95xx.c
@@ -1108,18 +1108,18 @@ static int smsc95xx_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 		if (unlikely(header & RX_STS_ES_)) {
 			if (netif_msg_rx_err(dev))
 				devdbg(dev, "Error header=0x%08x", header);
-			dev->stats.rx_errors++;
-			dev->stats.rx_dropped++;
+			dev->net->stats.rx_errors++;
+			dev->net->stats.rx_dropped++;
 
 			if (header & RX_STS_CRC_) {
-				dev->stats.rx_crc_errors++;
+				dev->net->stats.rx_crc_errors++;
 			} else {
 				if (header & (RX_STS_TL_ | RX_STS_RF_))
-					dev->stats.rx_frame_errors++;
+					dev->net->stats.rx_frame_errors++;
 
 				if ((header & RX_STS_LE_) &&
 					(!(header & RX_STS_FT_)))
-					dev->stats.rx_length_errors++;
+					dev->net->stats.rx_length_errors++;
 			}
 		} else {
 			/* ETH_FRAME_LEN + 4(CRC) + 2(COE) + 4(Vlan) */
-- 
cgit v1.2.3-59-g8ed1b


From 7963837f933df8a8ada56fa8f8205ebab40f84d0 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 29 Jun 2009 16:53:28 +0000
Subject: usbnet: Use netdev stats structure

Now that netdev has its own stats structure we should use that
instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/usb/usbnet.c | 30 +++++++++++++++---------------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index 22c0585a0319..edfd9e10ceba 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -234,8 +234,8 @@ void usbnet_skb_return (struct usbnet *dev, struct sk_buff *skb)
 	int	status;
 
 	skb->protocol = eth_type_trans (skb, dev->net);
-	dev->stats.rx_packets++;
-	dev->stats.rx_bytes += skb->len;
+	dev->net->stats.rx_packets++;
+	dev->net->stats.rx_bytes += skb->len;
 
 	if (netif_msg_rx_status (dev))
 		devdbg (dev, "< rx, len %zu, type 0x%x",
@@ -397,7 +397,7 @@ static inline void rx_process (struct usbnet *dev, struct sk_buff *skb)
 		if (netif_msg_rx_err (dev))
 			devdbg (dev, "drop");
 error:
-		dev->stats.rx_errors++;
+		dev->net->stats.rx_errors++;
 		skb_queue_tail (&dev->done, skb);
 	}
 }
@@ -420,8 +420,8 @@ static void rx_complete (struct urb *urb)
 	case 0:
 		if (skb->len < dev->net->hard_header_len) {
 			entry->state = rx_cleanup;
-			dev->stats.rx_errors++;
-			dev->stats.rx_length_errors++;
+			dev->net->stats.rx_errors++;
+			dev->net->stats.rx_length_errors++;
 			if (netif_msg_rx_err (dev))
 				devdbg (dev, "rx length %d", skb->len);
 		}
@@ -433,7 +433,7 @@ static void rx_complete (struct urb *urb)
 	 * storm, recovering as needed.
 	 */
 	case -EPIPE:
-		dev->stats.rx_errors++;
+		dev->net->stats.rx_errors++;
 		usbnet_defer_kevent (dev, EVENT_RX_HALT);
 		// FALLTHROUGH
 
@@ -451,7 +451,7 @@ static void rx_complete (struct urb *urb)
 	case -EPROTO:
 	case -ETIME:
 	case -EILSEQ:
-		dev->stats.rx_errors++;
+		dev->net->stats.rx_errors++;
 		if (!timer_pending (&dev->delay)) {
 			mod_timer (&dev->delay, jiffies + THROTTLE_JIFFIES);
 			if (netif_msg_link (dev))
@@ -465,12 +465,12 @@ block:
 
 	/* data overrun ... flush fifo? */
 	case -EOVERFLOW:
-		dev->stats.rx_over_errors++;
+		dev->net->stats.rx_over_errors++;
 		// FALLTHROUGH
 
 	default:
 		entry->state = rx_cleanup;
-		dev->stats.rx_errors++;
+		dev->net->stats.rx_errors++;
 		if (netif_msg_rx_err (dev))
 			devdbg (dev, "rx status %d", urb_status);
 		break;
@@ -583,8 +583,8 @@ int usbnet_stop (struct net_device *net)
 
 	if (netif_msg_ifdown (dev))
 		devinfo (dev, "stop stats: rx/tx %ld/%ld, errs %ld/%ld",
-			dev->stats.rx_packets, dev->stats.tx_packets,
-			dev->stats.rx_errors, dev->stats.tx_errors
+			net->stats.rx_packets, net->stats.tx_packets,
+			net->stats.rx_errors, net->stats.tx_errors
 			);
 
 	// ensure there are no more active urbs
@@ -891,10 +891,10 @@ static void tx_complete (struct urb *urb)
 	struct usbnet		*dev = entry->dev;
 
 	if (urb->status == 0) {
-		dev->stats.tx_packets++;
-		dev->stats.tx_bytes += entry->length;
+		dev->net->stats.tx_packets++;
+		dev->net->stats.tx_bytes += entry->length;
 	} else {
-		dev->stats.tx_errors++;
+		dev->net->stats.tx_errors++;
 
 		switch (urb->status) {
 		case -EPIPE:
@@ -1020,7 +1020,7 @@ int usbnet_start_xmit (struct sk_buff *skb, struct net_device *net)
 			devdbg (dev, "drop, code %d", retval);
 drop:
 		retval = NET_XMIT_SUCCESS;
-		dev->stats.tx_dropped++;
+		dev->net->stats.tx_dropped++;
 		if (skb)
 			dev_kfree_skb_any (skb);
 		usb_free_urb (urb);
-- 
cgit v1.2.3-59-g8ed1b


From d9d62f3f2c6fa609883714f6fd6cd710a83d307f Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 29 Jun 2009 16:54:12 +0000
Subject: usbnet: Remove private stats structure

Now that nothing uses the private stats structure we can remove it.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/usb/usbnet.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/include/linux/usb/usbnet.h b/include/linux/usb/usbnet.h
index 5d44059f6d63..310e18a880ff 100644
--- a/include/linux/usb/usbnet.h
+++ b/include/linux/usb/usbnet.h
@@ -42,7 +42,6 @@ struct usbnet {
 
 	/* protocol/interface state */
 	struct net_device	*net;
-	struct net_device_stats	stats;
 	int			msg_enable;
 	unsigned long		data [5];
 	u32			xid;
-- 
cgit v1.2.3-59-g8ed1b


From 88d2b81f4ee8f9ea3798dbe6105beb5609844317 Mon Sep 17 00:00:00 2001
From: Don Skidmore <donald.c.skidmore@intel.com>
Date: Tue, 30 Jun 2009 11:43:55 +0000
Subject: ixgbe: Fix SFP log messages

We had a wide range of log messages for the same sort of SFP
failure.  This patch makes them all more similar and less
confusing along with converting them to dev_err.

Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ixgbe/ixgbe_main.c | 25 +++++++++++++++++--------
 1 file changed, 17 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index e756e220db32..30d8c0e41a9d 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -2701,7 +2701,10 @@ static int ixgbe_up_complete(struct ixgbe_adapter *adapter)
 	 */
 	err = hw->phy.ops.identify(hw);
 	if (err == IXGBE_ERR_SFP_NOT_SUPPORTED) {
-		DPRINTK(PROBE, ERR, "PHY not supported on this NIC %d\n", err);
+		dev_err(&adapter->pdev->dev, "failed to initialize because "
+			"an unsupported SFP+ module type was detected.\n"
+			"Reload the driver after installing a supported "
+			"module.\n");
 		ixgbe_down(adapter);
 		return err;
 	}
@@ -3720,10 +3723,11 @@ static void ixgbe_sfp_task(struct work_struct *work)
 			goto reschedule;
 		ret = hw->phy.ops.reset(hw);
 		if (ret == IXGBE_ERR_SFP_NOT_SUPPORTED) {
-			DPRINTK(PROBE, ERR, "failed to initialize because an "
-			        "unsupported SFP+ module type was detected.\n"
-			        "Reload the driver after installing a "
-			        "supported module.\n");
+			dev_err(&adapter->pdev->dev, "failed to initialize "
+				"because an unsupported SFP+ module type "
+				"was detected.\n"
+				"Reload the driver after installing a "
+				"supported module.\n");
 			unregister_netdev(adapter->netdev);
 		} else {
 			DPRINTK(PROBE, INFO, "detected SFP+: %d\n",
@@ -4526,7 +4530,10 @@ static void ixgbe_sfp_config_module_task(struct work_struct *work)
 	adapter->flags |= IXGBE_FLAG_IN_SFP_MOD_TASK;
 	err = hw->phy.ops.identify_sfp(hw);
 	if (err == IXGBE_ERR_SFP_NOT_SUPPORTED) {
-		DPRINTK(PROBE, ERR, "PHY not supported on this NIC %d\n", err);
+		dev_err(&adapter->pdev->dev, "failed to initialize because "
+			"an unsupported SFP+ module type was detected.\n"
+			"Reload the driver after installing a supported "
+			"module.\n");
 		ixgbe_down(adapter);
 		return;
 	}
@@ -5513,8 +5520,10 @@ static int __devinit ixgbe_probe(struct pci_dev *pdev,
 			  round_jiffies(jiffies + (2 * HZ)));
 		err = 0;
 	} else if (err == IXGBE_ERR_SFP_NOT_SUPPORTED) {
-		dev_err(&adapter->pdev->dev, "failed to load because an "
-		        "unsupported SFP+ module type was detected.\n");
+		dev_err(&adapter->pdev->dev, "failed to initialize because "
+			"an unsupported SFP+ module type was detected.\n"
+			"Reload the driver after installing a supported "
+			"module.\n");
 		goto err_sw_init;
 	} else if (err) {
 		dev_err(&adapter->pdev->dev, "HW Init failed: %d\n", err);
-- 
cgit v1.2.3-59-g8ed1b


From a380137900fca5c79e6daa9500bdb6ea5649188e Mon Sep 17 00:00:00 2001
From: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Date: Tue, 30 Jun 2009 11:44:16 +0000
Subject: ixgbe: Fix device capabilities of 82599 single speed fiber NICs.

82599 single speed fiber modules only support 10G/Full. Return
proper device capabilities while querrying the adapter and error
while changing device advertisement/speed/duplex capabilities.

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ixgbe/ixgbe_ethtool.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe_ethtool.c b/drivers/net/ixgbe/ixgbe_ethtool.c
index 86f4f3e36f27..0f7b6a3a2e68 100644
--- a/drivers/net/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ixgbe/ixgbe_ethtool.c
@@ -139,7 +139,7 @@ static int ixgbe_get_settings(struct net_device *netdev,
 	ecmd->autoneg = AUTONEG_ENABLE;
 	ecmd->transceiver = XCVR_EXTERNAL;
 	if ((hw->phy.media_type == ixgbe_media_type_copper) ||
-	    (hw->mac.type == ixgbe_mac_82599EB)) {
+	    (hw->phy.multispeed_fiber)) {
 		ecmd->supported |= (SUPPORTED_1000baseT_Full |
 		                    SUPPORTED_Autoneg);
 
@@ -217,7 +217,7 @@ static int ixgbe_set_settings(struct net_device *netdev,
 	s32 err = 0;
 
 	if ((hw->phy.media_type == ixgbe_media_type_copper) ||
-	    (hw->mac.type == ixgbe_mac_82599EB)) {
+	    (hw->phy.multispeed_fiber)) {
 		/* 10000/copper and 1000/copper must autoneg
 		 * this function does not support any duplex forcing, but can
 		 * limit the advertising of the adapter to only 10000 or 1000 */
@@ -245,6 +245,7 @@ static int ixgbe_set_settings(struct net_device *netdev,
 	} else {
 		/* in this case we currently only support 10Gb/FULL */
 		if ((ecmd->autoneg == AUTONEG_ENABLE) ||
+		    (ecmd->advertising != ADVERTISED_10000baseT_Full) ||
 		    (ecmd->speed + ecmd->duplex != SPEED_10000 + DUPLEX_FULL))
 			return -EINVAL;
 	}
-- 
cgit v1.2.3-59-g8ed1b


From a1f25324b93ecdab1cbb27d3e9c4cafecb06ceda Mon Sep 17 00:00:00 2001
From: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Date: Tue, 30 Jun 2009 11:44:36 +0000
Subject: ixgbe: Fix link capabilities during adapter resets

Adapter link advertisement capabilities were not persistent during
adapter resets. While configuring multispeed fiber link check for
phy autoneg_advertised settings before overwriting with default
link capabilities

Signed-off-by: Mallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ixgbe/ixgbe_main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index 30d8c0e41a9d..fce2ef49b3a7 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -4506,7 +4506,8 @@ static void ixgbe_multispeed_fiber_task(struct work_struct *work)
 	u32 autoneg;
 
 	adapter->flags |= IXGBE_FLAG_IN_SFP_LINK_TASK;
-	if (hw->mac.ops.get_link_capabilities)
+	autoneg = hw->phy.autoneg_advertised;
+	if ((!autoneg) && (hw->mac.ops.get_link_capabilities))
 		hw->mac.ops.get_link_capabilities(hw, &autoneg,
 		                                  &hw->mac.autoneg);
 	if (hw->mac.ops.setup_link_speed)
-- 
cgit v1.2.3-59-g8ed1b


From 4f57ca6e17edfc56ddde5c87eb893e47e0d2d343 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Tue, 30 Jun 2009 11:44:56 +0000
Subject: ixgbe: fix unmap length bug

This patch addresses three WARN_ON statements from DMA-API debug code

ixgbe is mapping more than it unmaps, reduce the length of the map call and
remove the "used once" local variable.

found by Joerg Roedel <joerg.roedel@amd.com> in 2.6.30, so is a candidate
for -stable.

in addition, fix missing ->dma = 0 after unmap to prevent double free with
pci_unmap_single

and lastly, don't unmap (half) pages that aren't mapped.

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
CC: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ixgbe/ixgbe_main.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index fce2ef49b3a7..5588ef493a3d 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -563,7 +563,6 @@ static void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter,
 	union ixgbe_adv_rx_desc *rx_desc;
 	struct ixgbe_rx_buffer *bi;
 	unsigned int i;
-	unsigned int bufsz = rx_ring->rx_buf_len + NET_IP_ALIGN;
 
 	i = rx_ring->next_to_use;
 	bi = &rx_ring->rx_buffer_info[i];
@@ -593,7 +592,9 @@ static void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter,
 
 		if (!bi->skb) {
 			struct sk_buff *skb;
-			skb = netdev_alloc_skb(adapter->netdev, bufsz);
+			skb = netdev_alloc_skb(adapter->netdev,
+			                       (rx_ring->rx_buf_len +
+			                        NET_IP_ALIGN));
 
 			if (!skb) {
 				adapter->alloc_rx_buff_failed++;
@@ -608,7 +609,8 @@ static void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter,
 			skb_reserve(skb, NET_IP_ALIGN);
 
 			bi->skb = skb;
-			bi->dma = pci_map_single(pdev, skb->data, bufsz,
+			bi->dma = pci_map_single(pdev, skb->data,
+			                         rx_ring->rx_buf_len,
 			                         PCI_DMA_FROMDEVICE);
 		}
 		/* Refresh the desc even if buffer_addrs didn't change because
@@ -732,6 +734,7 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 			pci_unmap_single(pdev, rx_buffer_info->dma,
 			                 rx_ring->rx_buf_len,
 			                 PCI_DMA_FROMDEVICE);
+			rx_buffer_info->dma = 0;
 			skb_put(skb, len);
 		}
 
@@ -2815,9 +2818,11 @@ static void ixgbe_clean_rx_ring(struct ixgbe_adapter *adapter,
 		}
 		if (!rx_buffer_info->page)
 			continue;
-		pci_unmap_page(pdev, rx_buffer_info->page_dma, PAGE_SIZE / 2,
-		               PCI_DMA_FROMDEVICE);
-		rx_buffer_info->page_dma = 0;
+		if (rx_buffer_info->page_dma) {
+			pci_unmap_page(pdev, rx_buffer_info->page_dma,
+			               PAGE_SIZE / 2, PCI_DMA_FROMDEVICE);
+			rx_buffer_info->page_dma = 0;
+		}
 		put_page(rx_buffer_info->page);
 		rx_buffer_info->page = NULL;
 		rx_buffer_info->page_offset = 0;
-- 
cgit v1.2.3-59-g8ed1b


From 91615f765a2935b6cbae424b9eee1585ed681ae6 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Tue, 30 Jun 2009 12:45:15 +0000
Subject: igb: fix unmap length bug

driver was mixing NET_IP_ALIGN count bytes in map/unmap calls
unevenly. Only map the bytes that the hardware might dma into

also fix unmap related bug where ->dma was not being cleared
after unmap

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/igb/igb_main.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index ea17319624aa..468356d124ea 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -4549,11 +4549,12 @@ static bool igb_clean_rx_irq_adv(struct igb_ring *rx_ring,
 		cleaned = true;
 		cleaned_count++;
 
+		/* this is the fast path for the non-packet split case */
 		if (!adapter->rx_ps_hdr_size) {
 			pci_unmap_single(pdev, buffer_info->dma,
-					 adapter->rx_buffer_len +
-					   NET_IP_ALIGN,
+					 adapter->rx_buffer_len,
 					 PCI_DMA_FROMDEVICE);
+			buffer_info->dma = 0;
 			skb_put(skb, length);
 			goto send_up;
 		}
@@ -4570,8 +4571,9 @@ static bool igb_clean_rx_irq_adv(struct igb_ring *rx_ring,
 
 		if (!skb_shinfo(skb)->nr_frags) {
 			pci_unmap_single(pdev, buffer_info->dma,
-					 adapter->rx_ps_hdr_size + NET_IP_ALIGN,
+					 adapter->rx_ps_hdr_size,
 					 PCI_DMA_FROMDEVICE);
+			buffer_info->dma = 0;
 			skb_put(skb, hlen);
 		}
 
@@ -4713,7 +4715,6 @@ static void igb_alloc_rx_buffers_adv(struct igb_ring *rx_ring,
 		bufsz = adapter->rx_ps_hdr_size;
 	else
 		bufsz = adapter->rx_buffer_len;
-	bufsz += NET_IP_ALIGN;
 
 	while (cleaned_count--) {
 		rx_desc = E1000_RX_DESC_ADV(*rx_ring, i);
@@ -4737,7 +4738,7 @@ static void igb_alloc_rx_buffers_adv(struct igb_ring *rx_ring,
 		}
 
 		if (!buffer_info->skb) {
-			skb = netdev_alloc_skb(netdev, bufsz);
+			skb = netdev_alloc_skb(netdev, bufsz + NET_IP_ALIGN);
 			if (!skb) {
 				adapter->alloc_rx_buff_failed++;
 				goto no_buffers;
-- 
cgit v1.2.3-59-g8ed1b


From 679be3ba0c493eb66d22c206273729ce50925e85 Mon Sep 17 00:00:00 2001
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
Date: Tue, 30 Jun 2009 12:45:34 +0000
Subject: e1000: fix unmap bug

as reported by kerneloops.org

[  121.781161] ------------[ cut here ]------------
[  121.781171] WARNING: at lib/dma-debug.c:793 check_unmap+0x14e/0x577()
[  121.781173] Hardware name: S5520HC
[  121.781177] e1000 0000:0a:00.0: DMA-API: device driver tries to free DMA
memory it has not allocated [device address=0x00000001d688b0fa] [size=1522
bytes]
[  121.781180] Modules linked in: e1000 mdio  dca [last unloaded: ixgbe]
[  121.781187] Pid: 4793, comm: bash Tainted: P 2.6.30-master-06161113 #3
[  121.781190] Call Trace:
[  121.781195]  [<ffffffff8123056f>] ? check_unmap+0x14e/0x577
[  121.781201]  [<ffffffff81057a19>] warn_slowpath_common+0x77/0x8f
[  121.781205]  [<ffffffff81057ae1>] warn_slowpath_fmt+0x9f/0xa1
[  121.781212]  [<ffffffff81477ce2>] ? _spin_lock_irqsave+0x3f/0x49
[  121.781216]  [<ffffffff8122fa97>] ? get_hash_bucket+0x28/0x33
[  121.781220]  [<ffffffff8123056f>] check_unmap+0x14e/0x577
[  121.781225]  [<ffffffff810e4f48>] ? check_bytes_and_report+0x38/0xcb
[  121.781230]  [<ffffffff81230bbf>] debug_dma_unmap_page+0x80/0x92
[  121.781234]  [<ffffffff8122e549>] ? unmap_single+0x1a/0x4e
[  121.781239]  [<ffffffff813901e1>] ? __kfree_skb+0x74/0x78
[  121.781250]  [<ffffffffa00662ef>] pci_unmap_single+0x64/0x6d [e1000]
[  121.781259]  [<ffffffffa0066344>] e1000_clean_rx_ring+0x4c/0xbf [e1000]
[  121.781268]  [<ffffffffa00663df>] e1000_clean_all_rx_rings+0x28/0x36 [e1000]
[  121.781277]  [<ffffffffa0067464>] e1000_down+0x138/0x141 [e1000]
[  121.781286]  [<ffffffffa00681c2>] __e1000_shutdown+0x6b/0x198 [e1000]
[  121.781296]  [<ffffffffa0068405>] e1000_suspend+0x17/0x50 [e1000]
[  121.781301]  [<ffffffff81237665>] pci_legacy_suspend+0x3b/0xbe
[  121.781305]  [<ffffffff81237bc6>] pci_pm_suspend+0x3e/0xf1
[  121.781310]  [<ffffffff812eaf1c>] pm_op+0x57/0xde
[  121.781314]  [<ffffffff812eb444>] dpm_suspend_start+0x31e/0x470
[  121.781319]  [<ffffffff810877da>] suspend_devices_and_enter+0x3e/0x1a2
[  121.781323]  [<ffffffff81087a0f>] enter_state+0xd1/0x127
[  121.781327]  [<ffffffff8108717a>] state_store+0xa7/0xc9
[  121.781332]  [<ffffffff81221843>] kobj_attr_store+0x17/0x19
[  121.781336]  [<ffffffff8113c01e>] sysfs_write_file+0xe5/0x121
[  121.781341]  [<ffffffff810ed165>] vfs_write+0xab/0x105
[  121.781344]  [<ffffffff810ed279>] sys_write+0x47/0x6d
[  121.781349]  [<ffffffff81027aab>] system_call_fastpath+0x16/0x1b
[  121.781352] ---[ end trace 97bacaaac2ed7786 ]---

Fix is to correctly zero out internal ->dma value when unmapping
and make sure never to unmap unless there specifically was a mapping done.

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/e1000/e1000_main.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c
index 5e3356f8eb5a..972e06d984c8 100644
--- a/drivers/net/e1000/e1000_main.c
+++ b/drivers/net/e1000/e1000_main.c
@@ -2185,12 +2185,16 @@ static void e1000_clean_rx_ring(struct e1000_adapter *adapter,
 	/* Free all the Rx ring sk_buffs */
 	for (i = 0; i < rx_ring->count; i++) {
 		buffer_info = &rx_ring->buffer_info[i];
-		if (buffer_info->skb) {
+		if (buffer_info->dma) {
 			pci_unmap_single(pdev,
 					 buffer_info->dma,
 					 buffer_info->length,
 					 PCI_DMA_FROMDEVICE);
+		}
 
+		buffer_info->dma = 0;
+
+		if (buffer_info->skb) {
 			dev_kfree_skb(buffer_info->skb);
 			buffer_info->skb = NULL;
 		}
@@ -4033,6 +4037,7 @@ static bool e1000_clean_rx_irq(struct e1000_adapter *adapter,
 		                 buffer_info->dma,
 		                 buffer_info->length,
 		                 PCI_DMA_FROMDEVICE);
+		buffer_info->dma = 0;
 
 		length = le16_to_cpu(rx_desc->length);
 		/* !EOP means multiple descriptors were used to store a single
@@ -4222,6 +4227,7 @@ map_skb:
 			pci_unmap_single(pdev, buffer_info->dma,
 					 adapter->rx_buffer_len,
 					 PCI_DMA_FROMDEVICE);
+			buffer_info->dma = 0;
 
 			break; /* while !buffer_info->skb */
 		}
-- 
cgit v1.2.3-59-g8ed1b


From eab633021c26025b34f36f79f0311d3d99f40ceb Mon Sep 17 00:00:00 2001
From: Andre Detsch <adetsch@br.ibm.com>
Date: Tue, 30 Jun 2009 12:46:13 +0000
Subject: e1000: return PCI_ERS_RESULT_DISCONNECT on permanent error

PCI drivers that implement the io_error_detected callback
should return PCI_ERS_RESULT_DISCONNECT if the state
passed in is pci_channel_io_perm_failure.  This state is
not checked in many of the network drivers.

The patch fixes the omission in the e1000 driver.

Based on Mike Mason's similar patch for e1000e.

Signed-off-by: Andre Detsch <adetsch@br.ibm.com>
CC: Mike Mason <mmlnx@us.ibm.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/e1000/e1000_main.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c
index 972e06d984c8..5b8cbdb4b520 100644
--- a/drivers/net/e1000/e1000_main.c
+++ b/drivers/net/e1000/e1000_main.c
@@ -4823,6 +4823,9 @@ static pci_ers_result_t e1000_io_error_detected(struct pci_dev *pdev,
 
 	netif_device_detach(netdev);
 
+	if (state == pci_channel_io_perm_failure)
+		return PCI_ERS_RESULT_DISCONNECT;
+
 	if (netif_running(netdev))
 		e1000_down(adapter);
 	pci_disable_device(pdev);
-- 
cgit v1.2.3-59-g8ed1b


From c93b5a76d58656158d195a7df507ebc660010969 Mon Sep 17 00:00:00 2001
From: Mike Mason <mmlnx@us.ibm.com>
Date: Tue, 30 Jun 2009 12:45:53 +0000
Subject: e1000e: io_error_detected callback should return
 PCI_ERS_RESULT_DISCONNECT

on permanent failure

PCI drivers that implement the io_error_detected callback
should return PCI_ERS_RESULT_DISCONNECT if the state
passed in is pci_channel_io_perm_failure.  This state is not
checked in many of the network drivers.

This patch fixes the omission in the e1000e driver.

Signed-off-by: Mike Mason <mmlnx@us.ibm.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/e1000e/netdev.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c
index 679885a122b4..63415bb6f48f 100644
--- a/drivers/net/e1000e/netdev.c
+++ b/drivers/net/e1000e/netdev.c
@@ -4785,6 +4785,9 @@ static pci_ers_result_t e1000_io_error_detected(struct pci_dev *pdev,
 
 	netif_device_detach(netdev);
 
+	if (state == pci_channel_io_perm_failure)
+		return PCI_ERS_RESULT_DISCONNECT;
+
 	if (netif_running(netdev))
 		e1000e_down(adapter);
 	pci_disable_device(pdev);
-- 
cgit v1.2.3-59-g8ed1b


From 59ed6eecff4aa00c5c5d18ffd180acac108d596e Mon Sep 17 00:00:00 2001
From: Alexander Duyck <alexander.h.duyck@intel.com>
Date: Tue, 30 Jun 2009 12:46:34 +0000
Subject: igb: return PCI_ERS_RESULT_DISCONNECT on permanent error

PCI drivers that implement the io_error_detected callback should return
PCI_ERS_RESULT_DISCONNECT if the state passed in is
pci_channel_io_perm_failure.  This patch fixes the issue for igb.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/igb/igb_main.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/igb/igb_main.c b/drivers/net/igb/igb_main.c
index 468356d124ea..be480292aba1 100644
--- a/drivers/net/igb/igb_main.c
+++ b/drivers/net/igb/igb_main.c
@@ -5339,6 +5339,9 @@ static pci_ers_result_t igb_io_error_detected(struct pci_dev *pdev,
 
 	netif_device_detach(netdev);
 
+	if (state == pci_channel_io_perm_failure)
+		return PCI_ERS_RESULT_DISCONNECT;
+
 	if (netif_running(netdev))
 		igb_down(adapter);
 	pci_disable_device(pdev);
-- 
cgit v1.2.3-59-g8ed1b


From f8a68e752bc4e39644843403168137663c984524 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Tue, 30 Jun 2009 16:27:17 +0000
Subject: Revert "ipv4: arp announce, arp_proxy and windows ip conflict
 verification"

This reverts commit 73ce7b01b4496a5fbf9caf63033c874be692333f.

After discovering that we don't listen to gratuitious arps in 2.6.30
I tracked the failure down to this commit.

The patch makes absolutely no sense.  RFC2131 RFC3927 and RFC5227.
are all in agreement that an arp request with sip == 0 should be used
for the probe (to prevent learning) and an arp request with sip == tip
should be used for the gratitous announcement that people can learn
from.

It appears the author of the broken patch got those two cases confused
and modified the code to drop all gratuitous arp traffic.  Ouch!

Cc: stable@kernel.org
Signed-off-by: Eric W. Biederman <ebiederm@aristanetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/arp.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c
index 8a3881e28aca..c29d75d8f1b1 100644
--- a/net/ipv4/arp.c
+++ b/net/ipv4/arp.c
@@ -801,11 +801,8 @@ static int arp_process(struct sk_buff *skb)
  *  cache.
  */
 
-	/*
-	 *  Special case: IPv4 duplicate address detection packet (RFC2131)
-	 *  and Gratuitous ARP/ARP Announce. (RFC3927, Section 2.4)
-	 */
-	if (sip == 0 || tip == sip) {
+	/* Special case: IPv4 duplicate address detection packet (RFC2131) */
+	if (sip == 0) {
 		if (arp->ar_op == htons(ARPOP_REQUEST) &&
 		    inet_addr_type(net, tip) == RTN_LOCAL &&
 		    !arp_ignore(in_dev, sip, tip))
-- 
cgit v1.2.3-59-g8ed1b


From a5c308d4d1659b1f4833b863394e3e24cdbdfc6e Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Wed, 1 Jul 2009 13:15:35 +1000
Subject: md/raid5: suspend shouldn't affect read requests.

md allows write to regions on an array to be suspended temporarily.
This allows user-space to participate is aspects of reshape.
In particular, data can be copied with not risk of a race.
We should not be blocking read requests though, so don't.

Cc: stable@kernel.org
Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/raid5.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 92ef9b6abfc7..1f444ae07f89 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -3702,7 +3702,8 @@ static int make_request(struct request_queue *q, struct bio * bi)
 			/* FIXME what if we get a false positive because these
 			 * are being updated.
 			 */
-			if (logical_sector >= mddev->suspend_lo &&
+			if (bio_data_dir(bi) == WRITE &&
+			    logical_sector >= mddev->suspend_lo &&
 			    logical_sector < mddev->suspend_hi) {
 				release_stripe(sh);
 				schedule();
-- 
cgit v1.2.3-59-g8ed1b


From e62e58a5ffdc98ac28d8dbd070c857620d541f99 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Wed, 1 Jul 2009 13:15:35 +1000
Subject: md: use interruptible wait when duration is controlled by userspace.

User space can set various limits on an md array so that resync waits
when it gets to a certain point, or so that I/O is blocked for a short
while.
When md is waiting against one of these limit, it should use an
interruptible wait so as not to add to the load average, and so are
not to trigger a warning if the wait goes on for too long.

Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/md.c    | 14 ++++++++++----
 drivers/md/raid5.c | 15 +++++++++++----
 2 files changed, 21 insertions(+), 8 deletions(-)

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 65fe35b5e34a..0f4a70c43ffc 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -6336,10 +6336,16 @@ void md_do_sync(mddev_t *mddev)
 			sysfs_notify(&mddev->kobj, NULL, "sync_completed");
 		}
 
-		if (j >= mddev->resync_max)
-			wait_event(mddev->recovery_wait,
-				   mddev->resync_max > j
-				   || kthread_should_stop());
+		while (j >= mddev->resync_max && !kthread_should_stop()) {
+			/* As this condition is controlled by user-space,
+			 * we can block indefinitely, so use '_interruptible'
+			 * to avoid triggering warnings.
+			 */
+			flush_signals(current); /* just in case */
+			wait_event_interruptible(mddev->recovery_wait,
+						 mddev->resync_max > j
+						 || kthread_should_stop());
+		}
 
 		if (kthread_should_stop())
 			goto interrupted;
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 1f444ae07f89..37835538b58e 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -3699,14 +3699,21 @@ static int make_request(struct request_queue *q, struct bio * bi)
 					goto retry;
 				}
 			}
-			/* FIXME what if we get a false positive because these
-			 * are being updated.
-			 */
+
 			if (bio_data_dir(bi) == WRITE &&
 			    logical_sector >= mddev->suspend_lo &&
 			    logical_sector < mddev->suspend_hi) {
 				release_stripe(sh);
-				schedule();
+				/* As the suspend_* range is controlled by
+				 * userspace, we want an interruptible
+				 * wait.
+				 */
+				flush_signals(current);
+				prepare_to_wait(&conf->wait_for_overlap,
+						&w, TASK_INTERRUPTIBLE);
+				if (logical_sector >= mddev->suspend_lo &&
+				    logical_sector < mddev->suspend_hi)
+					schedule();
 				goto retry;
 			}
 
-- 
cgit v1.2.3-59-g8ed1b


From 84f7597c116f811cfcb762e645fa258b7e585155 Mon Sep 17 00:00:00 2001
From: Magnus Damm <damm@igel.co.jp>
Date: Wed, 1 Jul 2009 04:55:35 +0000
Subject: sh: fix se7724 ceu names

Use "ceu0" and "ceu1" as CEU names instead of "ceu".
This fixes "memchunk" kernel command line selection
on the solution engine 7724 board.

With this patch applied use "memchunk.ceu0=1m" or
"memchunk.ceu1=1m" on kernel command line to override
physically memory size to one meg for CEU0 or CEU1.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/mach-se/7724/setup.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/sh/boards/mach-se/7724/setup.c b/arch/sh/boards/mach-se/7724/setup.c
index 21d18005fb4a..c050a8d76dfd 100644
--- a/arch/sh/boards/mach-se/7724/setup.c
+++ b/arch/sh/boards/mach-se/7724/setup.c
@@ -454,7 +454,7 @@ static int __init devices_setup(void)
 	gpio_request(GPIO_FN_VIO0_CLK, NULL);
 	gpio_request(GPIO_FN_VIO0_FLD, NULL);
 	gpio_request(GPIO_FN_VIO0_HD,  NULL);
-	platform_resource_setup_memory(&ceu0_device, "ceu", 4 << 20);
+	platform_resource_setup_memory(&ceu0_device, "ceu0", 4 << 20);
 
 	/* enable CEU1 */
 	gpio_request(GPIO_FN_VIO1_D7,  NULL);
@@ -469,7 +469,7 @@ static int __init devices_setup(void)
 	gpio_request(GPIO_FN_VIO1_HD,  NULL);
 	gpio_request(GPIO_FN_VIO1_VD,  NULL);
 	gpio_request(GPIO_FN_VIO1_CLK, NULL);
-	platform_resource_setup_memory(&ceu1_device, "ceu", 4 << 20);
+	platform_resource_setup_memory(&ceu1_device, "ceu1", 4 << 20);
 
 	/* KEYSC */
 	gpio_request(GPIO_FN_KEYOUT5_IN5, NULL);
-- 
cgit v1.2.3-59-g8ed1b


From 0802d9e55cc6be6bd1b13f9dc9ef5aa6d24bca77 Mon Sep 17 00:00:00 2001
From: Magnus Damm <damm@igel.co.jp>
Date: Wed, 1 Jul 2009 05:16:31 +0000
Subject: sh: re-add LCDC fbdev support to the Migo-R defconfig

Re-add LCDC fbdev support to the Migo-R defconfig.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/configs/migor_defconfig | 53 +++++++++++++++++++++++++++++++++++++++--
 1 file changed, 51 insertions(+), 2 deletions(-)

diff --git a/arch/sh/configs/migor_defconfig b/arch/sh/configs/migor_defconfig
index da627d22c009..b18cfd39cac6 100644
--- a/arch/sh/configs/migor_defconfig
+++ b/arch/sh/configs/migor_defconfig
@@ -309,7 +309,7 @@ CONFIG_ZERO_PAGE_OFFSET=0x00001000
 CONFIG_BOOT_LINK_OFFSET=0x00800000
 CONFIG_ENTRY_OFFSET=0x00001000
 CONFIG_CMDLINE_BOOL=y
-CONFIG_CMDLINE="console=ttySC0,115200 earlyprintk=serial ip=on root=/dev/nfs ip=dhcp"
+CONFIG_CMDLINE="console=tty0 console=ttySC0,115200 earlyprintk=serial ip=on root=/dev/nfs ip=dhcp"
 
 #
 # Bus options
@@ -858,7 +858,35 @@ CONFIG_VIDEO_SH_MOBILE_CEU=y
 #
 # CONFIG_VGASTATE is not set
 # CONFIG_VIDEO_OUTPUT_CONTROL is not set
-# CONFIG_FB is not set
+CONFIG_FB=y
+# CONFIG_FIRMWARE_EDID is not set
+# CONFIG_FB_DDC is not set
+# CONFIG_FB_BOOT_VESA_SUPPORT is not set
+# CONFIG_FB_CFB_FILLRECT is not set
+# CONFIG_FB_CFB_COPYAREA is not set
+# CONFIG_FB_CFB_IMAGEBLIT is not set
+# CONFIG_FB_CFB_REV_PIXELS_IN_BYTE is not set
+CONFIG_FB_SYS_FILLRECT=y
+CONFIG_FB_SYS_COPYAREA=y
+CONFIG_FB_SYS_IMAGEBLIT=y
+# CONFIG_FB_FOREIGN_ENDIAN is not set
+CONFIG_FB_SYS_FOPS=y
+CONFIG_FB_DEFERRED_IO=y
+# CONFIG_FB_SVGALIB is not set
+# CONFIG_FB_MACMODES is not set
+# CONFIG_FB_BACKLIGHT is not set
+# CONFIG_FB_MODE_HELPERS is not set
+# CONFIG_FB_TILEBLITTING is not set
+
+#
+# Frame buffer hardware drivers
+#
+# CONFIG_FB_S1D13XXX is not set
+CONFIG_FB_SH_MOBILE_LCDC=y
+# CONFIG_FB_VIRTUAL is not set
+# CONFIG_FB_METRONOME is not set
+# CONFIG_FB_MB862XX is not set
+# CONFIG_FB_BROADSHEET is not set
 # CONFIG_BACKLIGHT_LCD_SUPPORT is not set
 
 #
@@ -870,6 +898,27 @@ CONFIG_VIDEO_SH_MOBILE_CEU=y
 # Console display driver support
 #
 CONFIG_DUMMY_CONSOLE=y
+CONFIG_FRAMEBUFFER_CONSOLE=y
+CONFIG_FRAMEBUFFER_CONSOLE_DETECT_PRIMARY=y
+# CONFIG_FRAMEBUFFER_CONSOLE_ROTATION is not set
+CONFIG_FONTS=y
+# CONFIG_FONT_8x8 is not set
+# CONFIG_FONT_8x16 is not set
+# CONFIG_FONT_6x11 is not set
+# CONFIG_FONT_7x14 is not set
+# CONFIG_FONT_PEARL_8x8 is not set
+# CONFIG_FONT_ACORN_8x8 is not set
+CONFIG_FONT_MINI_4x6=y
+# CONFIG_FONT_SUN8x16 is not set
+# CONFIG_FONT_SUN12x22 is not set
+# CONFIG_FONT_10x18 is not set
+CONFIG_LOGO=y
+# CONFIG_LOGO_LINUX_MONO is not set
+# CONFIG_LOGO_LINUX_VGA16 is not set
+# CONFIG_LOGO_LINUX_CLUT224 is not set
+# CONFIG_LOGO_SUPERH_MONO is not set
+CONFIG_LOGO_SUPERH_VGA16=y
+# CONFIG_LOGO_SUPERH_CLUT224 is not set
 # CONFIG_SOUND is not set
 CONFIG_HID_SUPPORT=y
 CONFIG_HID=y
-- 
cgit v1.2.3-59-g8ed1b


From 61fd21670d048017c81e62f60894ef1b04b481db Mon Sep 17 00:00:00 2001
From: Andre Noll <maan@systemlinux.org>
Date: Thu, 25 Jun 2009 13:03:04 +0200
Subject: Trivial typo fixes in Documentation/block/data-integrity.txt.

Signed-off-by: Andre Noll <maan@systemlinux.org>
Acked-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 Documentation/block/data-integrity.txt | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/Documentation/block/data-integrity.txt b/Documentation/block/data-integrity.txt
index e8ca040ba2cf..2d735b0ae383 100644
--- a/Documentation/block/data-integrity.txt
+++ b/Documentation/block/data-integrity.txt
@@ -50,7 +50,7 @@ encouraged them to allow separation of the data and integrity metadata
 scatter-gather lists.
 
 The controller will interleave the buffers on write and split them on
-read.  This means that the Linux can DMA the data buffers to and from
+read.  This means that Linux can DMA the data buffers to and from
 host memory without changes to the page cache.
 
 Also, the 16-bit CRC checksum mandated by both the SCSI and SATA specs
@@ -66,7 +66,7 @@ software RAID5).
 
 The IP checksum is weaker than the CRC in terms of detecting bit
 errors.  However, the strength is really in the separation of the data
-buffers and the integrity metadata.  These two distinct buffers much
+buffers and the integrity metadata.  These two distinct buffers must
 match up for an I/O to complete.
 
 The separation of the data and integrity metadata buffers as well as
-- 
cgit v1.2.3-59-g8ed1b


From d5036d770f871bd34c9cfd955e6dee692e1e8e81 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Fri, 26 Jun 2009 10:44:34 +0200
Subject: cfq-iosched: move cfqq initialization out of cfq_find_alloc_queue()

We're going to be needing that init code outside of that function
to get rid of the __GFP_NOFAIL in cfqq allocation.

Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/cfq-iosched.c | 37 +++++++++++++++++++++----------------
 1 file changed, 21 insertions(+), 16 deletions(-)

diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 833ec18eaa63..c760ae7019dd 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -1641,6 +1641,26 @@ static void cfq_ioc_set_ioprio(struct io_context *ioc)
 	ioc->ioprio_changed = 0;
 }
 
+static void cfq_init_cfqq(struct cfq_data *cfqd, struct cfq_queue *cfqq,
+			  pid_t pid, int is_sync)
+{
+	RB_CLEAR_NODE(&cfqq->rb_node);
+	RB_CLEAR_NODE(&cfqq->p_node);
+	INIT_LIST_HEAD(&cfqq->fifo);
+
+	atomic_set(&cfqq->ref, 0);
+	cfqq->cfqd = cfqd;
+
+	cfq_mark_cfqq_prio_changed(cfqq);
+
+	if (is_sync) {
+		if (!cfq_class_idle(cfqq))
+			cfq_mark_cfqq_idle_window(cfqq);
+		cfq_mark_cfqq_sync(cfqq);
+	}
+	cfqq->pid = pid;
+}
+
 static struct cfq_queue *
 cfq_find_alloc_queue(struct cfq_data *cfqd, int is_sync,
 		     struct io_context *ioc, gfp_t gfp_mask)
@@ -1678,23 +1698,8 @@ retry:
 				goto out;
 		}
 
-		RB_CLEAR_NODE(&cfqq->rb_node);
-		RB_CLEAR_NODE(&cfqq->p_node);
-		INIT_LIST_HEAD(&cfqq->fifo);
-
-		atomic_set(&cfqq->ref, 0);
-		cfqq->cfqd = cfqd;
-
-		cfq_mark_cfqq_prio_changed(cfqq);
-
+		cfq_init_cfqq(cfqd, cfqq, current->pid, is_sync);
 		cfq_init_prio_data(cfqq, ioc);
-
-		if (is_sync) {
-			if (!cfq_class_idle(cfqq))
-				cfq_mark_cfqq_idle_window(cfqq);
-			cfq_mark_cfqq_sync(cfqq);
-		}
-		cfqq->pid = current->pid;
 		cfq_log_cfqq(cfqd, cfqq, "alloced");
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 6118b70b3a0b4c583439bb77600194c82f220ce3 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Tue, 30 Jun 2009 09:34:12 +0200
Subject: cfq-iosched: get rid of the need for __GFP_NOFAIL in
 cfq_find_alloc_queue()

Setup an emergency fallback cfqq that we allocate at IO scheduler init
time. If the slab allocation fails in cfq_find_alloc_queue(), we'll just
punt IO to that cfqq instead. This ensures that cfq_find_alloc_queue()
never fails without having to ensure free memory.

On cfqq lookup, always try to allocate a new cfqq if the given cfq io
context has the oom_cfqq assigned. This ensures that we only temporarily
punt to this shared queue.

Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/cfq-iosched.c | 137 ++++++++++++++++++++++++++++------------------------
 1 file changed, 73 insertions(+), 64 deletions(-)

diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index c760ae7019dd..1d9160ffa26d 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -70,6 +70,51 @@ struct cfq_rb_root {
 };
 #define CFQ_RB_ROOT	(struct cfq_rb_root) { RB_ROOT, NULL, }
 
+/*
+ * Per process-grouping structure
+ */
+struct cfq_queue {
+	/* reference count */
+	atomic_t ref;
+	/* various state flags, see below */
+	unsigned int flags;
+	/* parent cfq_data */
+	struct cfq_data *cfqd;
+	/* service_tree member */
+	struct rb_node rb_node;
+	/* service_tree key */
+	unsigned long rb_key;
+	/* prio tree member */
+	struct rb_node p_node;
+	/* prio tree root we belong to, if any */
+	struct rb_root *p_root;
+	/* sorted list of pending requests */
+	struct rb_root sort_list;
+	/* if fifo isn't expired, next request to serve */
+	struct request *next_rq;
+	/* requests queued in sort_list */
+	int queued[2];
+	/* currently allocated requests */
+	int allocated[2];
+	/* fifo list of requests in sort_list */
+	struct list_head fifo;
+
+	unsigned long slice_end;
+	long slice_resid;
+	unsigned int slice_dispatch;
+
+	/* pending metadata requests */
+	int meta_pending;
+	/* number of requests that are on the dispatch list or inside driver */
+	int dispatched;
+
+	/* io prio of this group */
+	unsigned short ioprio, org_ioprio;
+	unsigned short ioprio_class, org_ioprio_class;
+
+	pid_t pid;
+};
+
 /*
  * Per block device queue structure
  */
@@ -135,51 +180,11 @@ struct cfq_data {
 	unsigned int cfq_slice_idle;
 
 	struct list_head cic_list;
-};
 
-/*
- * Per process-grouping structure
- */
-struct cfq_queue {
-	/* reference count */
-	atomic_t ref;
-	/* various state flags, see below */
-	unsigned int flags;
-	/* parent cfq_data */
-	struct cfq_data *cfqd;
-	/* service_tree member */
-	struct rb_node rb_node;
-	/* service_tree key */
-	unsigned long rb_key;
-	/* prio tree member */
-	struct rb_node p_node;
-	/* prio tree root we belong to, if any */
-	struct rb_root *p_root;
-	/* sorted list of pending requests */
-	struct rb_root sort_list;
-	/* if fifo isn't expired, next request to serve */
-	struct request *next_rq;
-	/* requests queued in sort_list */
-	int queued[2];
-	/* currently allocated requests */
-	int allocated[2];
-	/* fifo list of requests in sort_list */
-	struct list_head fifo;
-
-	unsigned long slice_end;
-	long slice_resid;
-	unsigned int slice_dispatch;
-
-	/* pending metadata requests */
-	int meta_pending;
-	/* number of requests that are on the dispatch list or inside driver */
-	int dispatched;
-
-	/* io prio of this group */
-	unsigned short ioprio, org_ioprio;
-	unsigned short ioprio_class, org_ioprio_class;
-
-	pid_t pid;
+	/*
+	 * Fallback dummy cfqq for extreme OOM conditions
+	 */
+	struct cfq_queue oom_cfqq;
 };
 
 enum cfqq_state_flags {
@@ -1673,41 +1678,40 @@ retry:
 	/* cic always exists here */
 	cfqq = cic_to_cfqq(cic, is_sync);
 
-	if (!cfqq) {
+	/*
+	 * Always try a new alloc if we fell back to the OOM cfqq
+	 * originally, since it should just be a temporary situation.
+	 */
+	if (!cfqq || cfqq == &cfqd->oom_cfqq) {
+		cfqq = NULL;
 		if (new_cfqq) {
 			cfqq = new_cfqq;
 			new_cfqq = NULL;
 		} else if (gfp_mask & __GFP_WAIT) {
-			/*
-			 * Inform the allocator of the fact that we will
-			 * just repeat this allocation if it fails, to allow
-			 * the allocator to do whatever it needs to attempt to
-			 * free memory.
-			 */
 			spin_unlock_irq(cfqd->queue->queue_lock);
 			new_cfqq = kmem_cache_alloc_node(cfq_pool,
-					gfp_mask | __GFP_NOFAIL | __GFP_ZERO,
+					gfp_mask | __GFP_ZERO,
 					cfqd->queue->node);
 			spin_lock_irq(cfqd->queue->queue_lock);
-			goto retry;
+			if (new_cfqq)
+				goto retry;
 		} else {
 			cfqq = kmem_cache_alloc_node(cfq_pool,
 					gfp_mask | __GFP_ZERO,
 					cfqd->queue->node);
-			if (!cfqq)
-				goto out;
 		}
 
-		cfq_init_cfqq(cfqd, cfqq, current->pid, is_sync);
-		cfq_init_prio_data(cfqq, ioc);
-		cfq_log_cfqq(cfqd, cfqq, "alloced");
+		if (cfqq) {
+			cfq_init_cfqq(cfqd, cfqq, current->pid, is_sync);
+			cfq_init_prio_data(cfqq, ioc);
+			cfq_log_cfqq(cfqd, cfqq, "alloced");
+		} else
+			cfqq = &cfqd->oom_cfqq;
 	}
 
 	if (new_cfqq)
 		kmem_cache_free(cfq_pool, new_cfqq);
 
-out:
-	WARN_ON((gfp_mask & __GFP_WAIT) && !cfqq);
 	return cfqq;
 }
 
@@ -1740,11 +1744,8 @@ cfq_get_queue(struct cfq_data *cfqd, int is_sync, struct io_context *ioc,
 		cfqq = *async_cfqq;
 	}
 
-	if (!cfqq) {
+	if (!cfqq)
 		cfqq = cfq_find_alloc_queue(cfqd, is_sync, ioc, gfp_mask);
-		if (!cfqq)
-			return NULL;
-	}
 
 	/*
 	 * pin the queue now that it's allocated, scheduler exit will prune it
@@ -2470,6 +2471,14 @@ static void *cfq_init_queue(struct request_queue *q)
 	for (i = 0; i < CFQ_PRIO_LISTS; i++)
 		cfqd->prio_trees[i] = RB_ROOT;
 
+	/*
+	 * Our fallback cfqq if cfq_find_alloc_queue() runs into OOM issues.
+	 * Grab a permanent reference to it, so that the normal code flow
+	 * will not attempt to free it.
+	 */
+	cfq_init_cfqq(cfqd, &cfqd->oom_cfqq, 1, 0);
+	atomic_inc(&cfqd->oom_cfqq.ref);
+
 	INIT_LIST_HEAD(&cfqd->cic_list);
 
 	cfqd->queue = q;
-- 
cgit v1.2.3-59-g8ed1b


From 7878cba9f0037f5599004b03a1260b32d9050360 Mon Sep 17 00:00:00 2001
From: "Martin K. Petersen" <martin.petersen@oracle.com>
Date: Fri, 26 Jun 2009 15:37:49 +0200
Subject: block: Create bip slabs with embedded integrity vectors

This patch restores stacking ability to the block layer integrity
infrastructure by creating a set of dedicated bip slabs.  Each bip slab
has an embedded bio_vec array at the end.  This cuts down on memory
allocations and also simplifies the code compared to the original bvec
version.  Only the largest bip slab is backed by a mempool.  The pool is
contained in the bio_set so stacking drivers can ensure forward
progress.

Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Jens Axboe <axboe@carl.(none)>
---
 block/blk-core.c    |   2 +-
 drivers/md/dm.c     |   4 +-
 fs/bio-integrity.c  | 170 ++++++++++++++++++++++++++++++++++++++--------------
 fs/bio.c            |  11 +++-
 include/linux/bio.h |  22 +++++--
 5 files changed, 152 insertions(+), 57 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index b06cf5c2a829..345d99da8d41 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -2365,7 +2365,7 @@ int blk_rq_prep_clone(struct request *rq, struct request *rq_src,
 		__bio_clone(bio, bio_src);
 
 		if (bio_integrity(bio_src) &&
-		    bio_integrity_clone(bio, bio_src, gfp_mask))
+		    bio_integrity_clone(bio, bio_src, gfp_mask, bs))
 			goto free_and_out;
 
 		if (bio_ctr && bio_ctr(bio, bio_src, data))
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 3c6d4ee8921d..9acd54a5cffb 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1017,7 +1017,7 @@ static struct bio *split_bvec(struct bio *bio, sector_t sector,
 	clone->bi_flags |= 1 << BIO_CLONED;
 
 	if (bio_integrity(bio)) {
-		bio_integrity_clone(clone, bio, GFP_NOIO);
+		bio_integrity_clone(clone, bio, GFP_NOIO, bs);
 		bio_integrity_trim(clone,
 				   bio_sector_offset(bio, idx, offset), len);
 	}
@@ -1045,7 +1045,7 @@ static struct bio *clone_bio(struct bio *bio, sector_t sector,
 	clone->bi_flags &= ~(1 << BIO_SEG_VALID);
 
 	if (bio_integrity(bio)) {
-		bio_integrity_clone(clone, bio, GFP_NOIO);
+		bio_integrity_clone(clone, bio, GFP_NOIO, bs);
 
 		if (idx != bio->bi_idx || clone->bi_size < bio->bi_size)
 			bio_integrity_trim(clone,
diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c
index 31c46a241bac..49a34e7f7306 100644
--- a/fs/bio-integrity.c
+++ b/fs/bio-integrity.c
@@ -1,7 +1,7 @@
 /*
  * bio-integrity.c - bio data integrity extensions
  *
- * Copyright (C) 2007, 2008 Oracle Corporation
+ * Copyright (C) 2007, 2008, 2009 Oracle Corporation
  * Written by: Martin K. Petersen <martin.petersen@oracle.com>
  *
  * This program is free software; you can redistribute it and/or
@@ -25,63 +25,121 @@
 #include <linux/bio.h>
 #include <linux/workqueue.h>
 
-static struct kmem_cache *bio_integrity_slab __read_mostly;
-static mempool_t *bio_integrity_pool;
-static struct bio_set *integrity_bio_set;
+struct integrity_slab {
+	struct kmem_cache *slab;
+	unsigned short nr_vecs;
+	char name[8];
+};
+
+#define IS(x) { .nr_vecs = x, .name = "bip-"__stringify(x) }
+struct integrity_slab bip_slab[BIOVEC_NR_POOLS] __read_mostly = {
+	IS(1), IS(4), IS(16), IS(64), IS(128), IS(BIO_MAX_PAGES),
+};
+#undef IS
+
 static struct workqueue_struct *kintegrityd_wq;
 
+static inline unsigned int vecs_to_idx(unsigned int nr)
+{
+	switch (nr) {
+	case 1:
+		return 0;
+	case 2 ... 4:
+		return 1;
+	case 5 ... 16:
+		return 2;
+	case 17 ... 64:
+		return 3;
+	case 65 ... 128:
+		return 4;
+	case 129 ... BIO_MAX_PAGES:
+		return 5;
+	default:
+		BUG();
+	}
+}
+
+static inline int use_bip_pool(unsigned int idx)
+{
+	if (idx == BIOVEC_NR_POOLS)
+		return 1;
+
+	return 0;
+}
+
 /**
- * bio_integrity_alloc - Allocate integrity payload and attach it to bio
+ * bio_integrity_alloc_bioset - Allocate integrity payload and attach it to bio
  * @bio:	bio to attach integrity metadata to
  * @gfp_mask:	Memory allocation mask
  * @nr_vecs:	Number of integrity metadata scatter-gather elements
+ * @bs:		bio_set to allocate from
  *
  * Description: This function prepares a bio for attaching integrity
  * metadata.  nr_vecs specifies the maximum number of pages containing
  * integrity metadata that can be attached.
  */
-struct bio_integrity_payload *bio_integrity_alloc(struct bio *bio,
-						  gfp_t gfp_mask,
-						  unsigned int nr_vecs)
+struct bio_integrity_payload *bio_integrity_alloc_bioset(struct bio *bio,
+							 gfp_t gfp_mask,
+							 unsigned int nr_vecs,
+							 struct bio_set *bs)
 {
 	struct bio_integrity_payload *bip;
-	struct bio_vec *iv;
-	unsigned long idx;
+	unsigned int idx = vecs_to_idx(nr_vecs);
 
 	BUG_ON(bio == NULL);
+	bip = NULL;
 
-	bip = mempool_alloc(bio_integrity_pool, gfp_mask);
-	if (unlikely(bip == NULL)) {
-		printk(KERN_ERR "%s: could not alloc bip\n", __func__);
-		return NULL;
-	}
+	/* Lower order allocations come straight from slab */
+	if (!use_bip_pool(idx))
+		bip = kmem_cache_alloc(bip_slab[idx].slab, gfp_mask);
 
-	memset(bip, 0, sizeof(*bip));
+	/* Use mempool if lower order alloc failed or max vecs were requested */
+	if (bip == NULL) {
+		bip = mempool_alloc(bs->bio_integrity_pool, gfp_mask);
 
-	iv = bvec_alloc_bs(gfp_mask, nr_vecs, &idx, integrity_bio_set);
-	if (unlikely(iv == NULL)) {
-		printk(KERN_ERR "%s: could not alloc bip_vec\n", __func__);
-		mempool_free(bip, bio_integrity_pool);
-		return NULL;
+		if (unlikely(bip == NULL)) {
+			printk(KERN_ERR "%s: could not alloc bip\n", __func__);
+			return NULL;
+		}
 	}
 
-	bip->bip_pool = idx;
-	bip->bip_vec = iv;
+	memset(bip, 0, sizeof(*bip));
+
+	bip->bip_slab = idx;
 	bip->bip_bio = bio;
 	bio->bi_integrity = bip;
 
 	return bip;
 }
+EXPORT_SYMBOL(bio_integrity_alloc_bioset);
+
+/**
+ * bio_integrity_alloc - Allocate integrity payload and attach it to bio
+ * @bio:	bio to attach integrity metadata to
+ * @gfp_mask:	Memory allocation mask
+ * @nr_vecs:	Number of integrity metadata scatter-gather elements
+ *
+ * Description: This function prepares a bio for attaching integrity
+ * metadata.  nr_vecs specifies the maximum number of pages containing
+ * integrity metadata that can be attached.
+ */
+struct bio_integrity_payload *bio_integrity_alloc(struct bio *bio,
+						  gfp_t gfp_mask,
+						  unsigned int nr_vecs)
+{
+	return bio_integrity_alloc_bioset(bio, gfp_mask, nr_vecs, fs_bio_set);
+}
 EXPORT_SYMBOL(bio_integrity_alloc);
 
 /**
  * bio_integrity_free - Free bio integrity payload
  * @bio:	bio containing bip to be freed
+ * @bs:		bio_set this bio was allocated from
  *
  * Description: Used to free the integrity portion of a bio. Usually
  * called from bio_free().
  */
-void bio_integrity_free(struct bio *bio)
+void bio_integrity_free(struct bio *bio, struct bio_set *bs)
 {
 	struct bio_integrity_payload *bip = bio->bi_integrity;
 
@@ -92,8 +150,10 @@ void bio_integrity_free(struct bio *bio)
 	    && bip->bip_buf != NULL)
 		kfree(bip->bip_buf);
 
-	bvec_free_bs(integrity_bio_set, bip->bip_vec, bip->bip_pool);
-	mempool_free(bip, bio_integrity_pool);
+	if (use_bip_pool(bip->bip_slab))
+		mempool_free(bip, bs->bio_integrity_pool);
+	else
+		kmem_cache_free(bip_slab[bip->bip_slab].slab, bip);
 
 	bio->bi_integrity = NULL;
 }
@@ -114,7 +174,7 @@ int bio_integrity_add_page(struct bio *bio, struct page *page,
 	struct bio_integrity_payload *bip = bio->bi_integrity;
 	struct bio_vec *iv;
 
-	if (bip->bip_vcnt >= bvec_nr_vecs(bip->bip_pool)) {
+	if (bip->bip_vcnt >= bvec_nr_vecs(bip->bip_slab)) {
 		printk(KERN_ERR "%s: bip_vec full\n", __func__);
 		return 0;
 	}
@@ -647,8 +707,8 @@ void bio_integrity_split(struct bio *bio, struct bio_pair *bp, int sectors)
 	bp->iv1 = bip->bip_vec[0];
 	bp->iv2 = bip->bip_vec[0];
 
-	bp->bip1.bip_vec = &bp->iv1;
-	bp->bip2.bip_vec = &bp->iv2;
+	bp->bip1.bip_vec[0] = bp->iv1;
+	bp->bip2.bip_vec[0] = bp->iv2;
 
 	bp->iv1.bv_len = sectors * bi->tuple_size;
 	bp->iv2.bv_offset += sectors * bi->tuple_size;
@@ -667,17 +727,19 @@ EXPORT_SYMBOL(bio_integrity_split);
  * @bio:	New bio
  * @bio_src:	Original bio
  * @gfp_mask:	Memory allocation mask
+ * @bs:		bio_set to allocate bip from
  *
  * Description:	Called to allocate a bip when cloning a bio
  */
-int bio_integrity_clone(struct bio *bio, struct bio *bio_src, gfp_t gfp_mask)
+int bio_integrity_clone(struct bio *bio, struct bio *bio_src,
+			gfp_t gfp_mask, struct bio_set *bs)
 {
 	struct bio_integrity_payload *bip_src = bio_src->bi_integrity;
 	struct bio_integrity_payload *bip;
 
 	BUG_ON(bip_src == NULL);
 
-	bip = bio_integrity_alloc(bio, gfp_mask, bip_src->bip_vcnt);
+	bip = bio_integrity_alloc_bioset(bio, gfp_mask, bip_src->bip_vcnt, bs);
 
 	if (bip == NULL)
 		return -EIO;
@@ -693,25 +755,43 @@ int bio_integrity_clone(struct bio *bio, struct bio *bio_src, gfp_t gfp_mask)
 }
 EXPORT_SYMBOL(bio_integrity_clone);
 
-static int __init bio_integrity_init(void)
+int bioset_integrity_create(struct bio_set *bs, int pool_size)
 {
-	kintegrityd_wq = create_workqueue("kintegrityd");
+	unsigned int max_slab = vecs_to_idx(BIO_MAX_PAGES);
+
+	bs->bio_integrity_pool =
+		mempool_create_slab_pool(pool_size, bip_slab[max_slab].slab);
 
+	if (!bs->bio_integrity_pool)
+		return -1;
+
+	return 0;
+}
+EXPORT_SYMBOL(bioset_integrity_create);
+
+void bioset_integrity_free(struct bio_set *bs)
+{
+	if (bs->bio_integrity_pool)
+		mempool_destroy(bs->bio_integrity_pool);
+}
+EXPORT_SYMBOL(bioset_integrity_free);
+
+void __init bio_integrity_init(void)
+{
+	unsigned int i;
+
+	kintegrityd_wq = create_workqueue("kintegrityd");
 	if (!kintegrityd_wq)
 		panic("Failed to create kintegrityd\n");
 
-	bio_integrity_slab = KMEM_CACHE(bio_integrity_payload,
-					SLAB_HWCACHE_ALIGN|SLAB_PANIC);
+	for (i = 0 ; i < BIOVEC_NR_POOLS ; i++) {
+		unsigned int size;
 
-	bio_integrity_pool = mempool_create_slab_pool(BIO_POOL_SIZE,
-						      bio_integrity_slab);
-	if (!bio_integrity_pool)
-		panic("bio_integrity: can't allocate bip pool\n");
+		size = sizeof(struct bio_integrity_payload)
+			+ bip_slab[i].nr_vecs * sizeof(struct bio_vec);
 
-	integrity_bio_set = bioset_create(BIO_POOL_SIZE, 0);
-	if (!integrity_bio_set)
-		panic("bio_integrity: can't allocate bio_set\n");
-
-	return 0;
+		bip_slab[i].slab =
+			kmem_cache_create(bip_slab[i].name, size, 0,
+					  SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL);
+	}
 }
-subsys_initcall(bio_integrity_init);
diff --git a/fs/bio.c b/fs/bio.c
index 24c914043532..1486b19fc431 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -238,7 +238,7 @@ void bio_free(struct bio *bio, struct bio_set *bs)
 		bvec_free_bs(bs, bio->bi_io_vec, BIO_POOL_IDX(bio));
 
 	if (bio_integrity(bio))
-		bio_integrity_free(bio);
+		bio_integrity_free(bio, bs);
 
 	/*
 	 * If we have front padding, adjust the bio pointer before freeing
@@ -341,7 +341,7 @@ struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs)
 static void bio_kmalloc_destructor(struct bio *bio)
 {
 	if (bio_integrity(bio))
-		bio_integrity_free(bio);
+		bio_integrity_free(bio, fs_bio_set);
 	kfree(bio);
 }
 
@@ -472,7 +472,7 @@ struct bio *bio_clone(struct bio *bio, gfp_t gfp_mask)
 	if (bio_integrity(bio)) {
 		int ret;
 
-		ret = bio_integrity_clone(b, bio, gfp_mask);
+		ret = bio_integrity_clone(b, bio, gfp_mask, fs_bio_set);
 
 		if (ret < 0) {
 			bio_put(b);
@@ -1539,6 +1539,7 @@ void bioset_free(struct bio_set *bs)
 	if (bs->bio_pool)
 		mempool_destroy(bs->bio_pool);
 
+	bioset_integrity_free(bs);
 	biovec_free_pools(bs);
 	bio_put_slab(bs);
 
@@ -1579,6 +1580,9 @@ struct bio_set *bioset_create(unsigned int pool_size, unsigned int front_pad)
 	if (!bs->bio_pool)
 		goto bad;
 
+	if (bioset_integrity_create(bs, pool_size))
+		goto bad;
+
 	if (!biovec_create_pools(bs, pool_size))
 		return bs;
 
@@ -1616,6 +1620,7 @@ static int __init init_bio(void)
 	if (!bio_slabs)
 		panic("bio: can't allocate bios\n");
 
+	bio_integrity_init();
 	biovec_init_slabs();
 
 	fs_bio_set = bioset_create(BIO_POOL_SIZE, 0);
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 2a04eb54c0dd..2892b710771c 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -319,7 +319,6 @@ static inline int bio_has_allocated_vec(struct bio *bio)
  */
 struct bio_integrity_payload {
 	struct bio		*bip_bio;	/* parent bio */
-	struct bio_vec		*bip_vec;	/* integrity data vector */
 
 	sector_t		bip_sector;	/* virtual start sector */
 
@@ -328,11 +327,12 @@ struct bio_integrity_payload {
 
 	unsigned int		bip_size;
 
-	unsigned short		bip_pool;	/* pool the ivec came from */
+	unsigned short		bip_slab;	/* slab the bip came from */
 	unsigned short		bip_vcnt;	/* # of integrity bio_vecs */
 	unsigned short		bip_idx;	/* current bip_vec index */
 
 	struct work_struct	bip_work;	/* I/O completion */
+	struct bio_vec		bip_vec[0];	/* embedded bvec array */
 };
 #endif /* CONFIG_BLK_DEV_INTEGRITY */
 
@@ -430,6 +430,9 @@ struct bio_set {
 	unsigned int front_pad;
 
 	mempool_t *bio_pool;
+#if defined(CONFIG_BLK_DEV_INTEGRITY)
+	mempool_t *bio_integrity_pool;
+#endif
 	mempool_t *bvec_pool;
 };
 
@@ -634,8 +637,9 @@ static inline struct bio *bio_list_get(struct bio_list *bl)
 
 #define bio_integrity(bio) (bio->bi_integrity != NULL)
 
+extern struct bio_integrity_payload *bio_integrity_alloc_bioset(struct bio *, gfp_t, unsigned int, struct bio_set *);
 extern struct bio_integrity_payload *bio_integrity_alloc(struct bio *, gfp_t, unsigned int);
-extern void bio_integrity_free(struct bio *);
+extern void bio_integrity_free(struct bio *, struct bio_set *);
 extern int bio_integrity_add_page(struct bio *, struct page *, unsigned int, unsigned int);
 extern int bio_integrity_enabled(struct bio *bio);
 extern int bio_integrity_set_tag(struct bio *, void *, unsigned int);
@@ -645,21 +649,27 @@ extern void bio_integrity_endio(struct bio *, int);
 extern void bio_integrity_advance(struct bio *, unsigned int);
 extern void bio_integrity_trim(struct bio *, unsigned int, unsigned int);
 extern void bio_integrity_split(struct bio *, struct bio_pair *, int);
-extern int bio_integrity_clone(struct bio *, struct bio *, gfp_t);
+extern int bio_integrity_clone(struct bio *, struct bio *, gfp_t, struct bio_set *);
+extern int bioset_integrity_create(struct bio_set *, int);
+extern void bioset_integrity_free(struct bio_set *);
+extern void bio_integrity_init(void);
 
 #else /* CONFIG_BLK_DEV_INTEGRITY */
 
 #define bio_integrity(a)		(0)
+#define bioset_integrity_create(a, b)	(0)
 #define bio_integrity_prep(a)		(0)
 #define bio_integrity_enabled(a)	(0)
-#define bio_integrity_clone(a, b, c)	(0)
-#define bio_integrity_free(a)		do { } while (0)
+#define bio_integrity_clone(a, b, c, d)	(0)
+#define bioset_integrity_free(a)	do { } while (0)
+#define bio_integrity_free(a, b)	do { } while (0)
 #define bio_integrity_endio(a, b)	do { } while (0)
 #define bio_integrity_advance(a, b)	do { } while (0)
 #define bio_integrity_trim(a, b, c)	do { } while (0)
 #define bio_integrity_split(a, b, c)	do { } while (0)
 #define bio_integrity_set_tag(a, b, c)	do { } while (0)
 #define bio_integrity_get_tag(a, b, c)	do { } while (0)
+#define bio_integrity_init(a)		do { } while (0)
 
 #endif /* CONFIG_BLK_DEV_INTEGRITY */
 
-- 
cgit v1.2.3-59-g8ed1b


From 018e0446890661504783f92388ecce7138c1566d Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Fri, 26 Jun 2009 16:27:10 +0200
Subject: block: get rid of queue-private command filter

The initial patches to support this through sysfs export were broken
and have been if 0'ed out in any release. So lets just kill the code
and reclaim some space in struct request_queue, if anyone would later
like to fixup the sysfs bits, the git history can easily restore
the removed bits.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/Makefile         |   2 +-
 block/blk-core.c       |   2 -
 block/bsg.c            |   2 +-
 block/cmd-filter.c     | 233 -------------------------------------------------
 block/scsi_ioctl.c     |  43 +++++++--
 drivers/scsi/sg.c      |   4 +-
 include/linux/blkdev.h |  15 +---
 7 files changed, 42 insertions(+), 259 deletions(-)
 delete mode 100644 block/cmd-filter.c

diff --git a/block/Makefile b/block/Makefile
index e9fa4dd690f2..6c54ed0ff755 100644
--- a/block/Makefile
+++ b/block/Makefile
@@ -5,7 +5,7 @@
 obj-$(CONFIG_BLOCK) := elevator.o blk-core.o blk-tag.o blk-sysfs.o \
 			blk-barrier.o blk-settings.o blk-ioc.o blk-map.o \
 			blk-exec.o blk-merge.o blk-softirq.o blk-timeout.o \
-			ioctl.o genhd.o scsi_ioctl.o cmd-filter.o
+			ioctl.o genhd.o scsi_ioctl.o
 
 obj-$(CONFIG_BLK_DEV_BSG)	+= bsg.o
 obj-$(CONFIG_IOSCHED_NOOP)	+= noop-iosched.o
diff --git a/block/blk-core.c b/block/blk-core.c
index 345d99da8d41..02b87134a167 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -595,8 +595,6 @@ blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id)
 
 	q->sg_reserved_size = INT_MAX;
 
-	blk_set_cmd_filter_defaults(&q->cmd_filter);
-
 	/*
 	 * all done
 	 */
diff --git a/block/bsg.c b/block/bsg.c
index e7d475254248..5f184bb3ff9e 100644
--- a/block/bsg.c
+++ b/block/bsg.c
@@ -186,7 +186,7 @@ static int blk_fill_sgv4_hdr_rq(struct request_queue *q, struct request *rq,
 		return -EFAULT;
 
 	if (hdr->subprotocol == BSG_SUB_PROTOCOL_SCSI_CMD) {
-		if (blk_verify_command(&q->cmd_filter, rq->cmd, has_write_perm))
+		if (blk_verify_command(rq->cmd, has_write_perm))
 			return -EPERM;
 	} else if (!capable(CAP_SYS_RAWIO))
 		return -EPERM;
diff --git a/block/cmd-filter.c b/block/cmd-filter.c
deleted file mode 100644
index 572bbc2f900d..000000000000
--- a/block/cmd-filter.c
+++ /dev/null
@@ -1,233 +0,0 @@
-/*
- * Copyright 2004 Peter M. Jones <pjones@redhat.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- *
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public Licens
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-
- *
- */
-
-#include <linux/list.h>
-#include <linux/genhd.h>
-#include <linux/spinlock.h>
-#include <linux/capability.h>
-#include <linux/bitops.h>
-#include <linux/blkdev.h>
-
-#include <scsi/scsi.h>
-#include <linux/cdrom.h>
-
-int blk_verify_command(struct blk_cmd_filter *filter,
-		       unsigned char *cmd, fmode_t has_write_perm)
-{
-	/* root can do any command. */
-	if (capable(CAP_SYS_RAWIO))
-		return 0;
-
-	/* if there's no filter set, assume we're filtering everything out */
-	if (!filter)
-		return -EPERM;
-
-	/* Anybody who can open the device can do a read-safe command */
-	if (test_bit(cmd[0], filter->read_ok))
-		return 0;
-
-	/* Write-safe commands require a writable open */
-	if (test_bit(cmd[0], filter->write_ok) && has_write_perm)
-		return 0;
-
-	return -EPERM;
-}
-EXPORT_SYMBOL(blk_verify_command);
-
-#if 0
-/* and now, the sysfs stuff */
-static ssize_t rcf_cmds_show(struct blk_cmd_filter *filter, char *page,
-			     int rw)
-{
-	char *npage = page;
-	unsigned long *okbits;
-	int i;
-
-	if (rw == READ)
-		okbits = filter->read_ok;
-	else
-		okbits = filter->write_ok;
-
-	for (i = 0; i < BLK_SCSI_MAX_CMDS; i++) {
-		if (test_bit(i, okbits)) {
-			npage += sprintf(npage, "0x%02x", i);
-			if (i < BLK_SCSI_MAX_CMDS - 1)
-				sprintf(npage++, " ");
-		}
-	}
-
-	if (npage != page)
-		npage += sprintf(npage, "\n");
-
-	return npage - page;
-}
-
-static ssize_t rcf_readcmds_show(struct blk_cmd_filter *filter, char *page)
-{
-	return rcf_cmds_show(filter, page, READ);
-}
-
-static ssize_t rcf_writecmds_show(struct blk_cmd_filter *filter,
-				 char *page)
-{
-	return rcf_cmds_show(filter, page, WRITE);
-}
-
-static ssize_t rcf_cmds_store(struct blk_cmd_filter *filter,
-			      const char *page, size_t count, int rw)
-{
-	unsigned long okbits[BLK_SCSI_CMD_PER_LONG], *target_okbits;
-	int cmd, set;
-	char *p, *status;
-
-	if (rw == READ) {
-		memcpy(&okbits, filter->read_ok, sizeof(okbits));
-		target_okbits = filter->read_ok;
-	} else {
-		memcpy(&okbits, filter->write_ok, sizeof(okbits));
-		target_okbits = filter->write_ok;
-	}
-
-	while ((p = strsep((char **)&page, " ")) != NULL) {
-		set = 1;
-
-		if (p[0] == '+') {
-			p++;
-		} else if (p[0] == '-') {
-			set = 0;
-			p++;
-		}
-
-		cmd = simple_strtol(p, &status, 16);
-
-		/* either of these cases means invalid input, so do nothing. */
-		if ((status == p) || cmd >= BLK_SCSI_MAX_CMDS)
-			return -EINVAL;
-
-		if (set)
-			__set_bit(cmd, okbits);
-		else
-			__clear_bit(cmd, okbits);
-	}
-
-	memcpy(target_okbits, okbits, sizeof(okbits));
-	return count;
-}
-
-static ssize_t rcf_readcmds_store(struct blk_cmd_filter *filter,
-				  const char *page, size_t count)
-{
-	return rcf_cmds_store(filter, page, count, READ);
-}
-
-static ssize_t rcf_writecmds_store(struct blk_cmd_filter *filter,
-				   const char *page, size_t count)
-{
-	return rcf_cmds_store(filter, page, count, WRITE);
-}
-
-struct rcf_sysfs_entry {
-	struct attribute attr;
-	ssize_t (*show)(struct blk_cmd_filter *, char *);
-	ssize_t (*store)(struct blk_cmd_filter *, const char *, size_t);
-};
-
-static struct rcf_sysfs_entry rcf_readcmds_entry = {
-	.attr = { .name = "read_table", .mode = S_IRUGO | S_IWUSR },
-	.show = rcf_readcmds_show,
-	.store = rcf_readcmds_store,
-};
-
-static struct rcf_sysfs_entry rcf_writecmds_entry = {
-	.attr = {.name = "write_table", .mode = S_IRUGO | S_IWUSR },
-	.show = rcf_writecmds_show,
-	.store = rcf_writecmds_store,
-};
-
-static struct attribute *default_attrs[] = {
-	&rcf_readcmds_entry.attr,
-	&rcf_writecmds_entry.attr,
-	NULL,
-};
-
-#define to_rcf(atr) container_of((atr), struct rcf_sysfs_entry, attr)
-
-static ssize_t
-rcf_attr_show(struct kobject *kobj, struct attribute *attr, char *page)
-{
-	struct rcf_sysfs_entry *entry = to_rcf(attr);
-	struct blk_cmd_filter *filter;
-
-	filter = container_of(kobj, struct blk_cmd_filter, kobj);
-	if (entry->show)
-		return entry->show(filter, page);
-
-	return 0;
-}
-
-static ssize_t
-rcf_attr_store(struct kobject *kobj, struct attribute *attr,
-			const char *page, size_t length)
-{
-	struct rcf_sysfs_entry *entry = to_rcf(attr);
-	struct blk_cmd_filter *filter;
-
-	if (!capable(CAP_SYS_RAWIO))
-		return -EPERM;
-
-	if (!entry->store)
-		return -EINVAL;
-
-	filter = container_of(kobj, struct blk_cmd_filter, kobj);
-	return entry->store(filter, page, length);
-}
-
-static struct sysfs_ops rcf_sysfs_ops = {
-	.show = rcf_attr_show,
-	.store = rcf_attr_store,
-};
-
-static struct kobj_type rcf_ktype = {
-	.sysfs_ops = &rcf_sysfs_ops,
-	.default_attrs = default_attrs,
-};
-
-int blk_register_filter(struct gendisk *disk)
-{
-	int ret;
-	struct blk_cmd_filter *filter = &disk->queue->cmd_filter;
-
-	ret = kobject_init_and_add(&filter->kobj, &rcf_ktype,
-				   &disk_to_dev(disk)->kobj,
-				   "%s", "cmd_filter");
-	if (ret < 0)
-		return ret;
-
-	return 0;
-}
-EXPORT_SYMBOL(blk_register_filter);
-
-void blk_unregister_filter(struct gendisk *disk)
-{
-	struct blk_cmd_filter *filter = &disk->queue->cmd_filter;
-
-	kobject_put(&filter->kobj);
-}
-EXPORT_SYMBOL(blk_unregister_filter);
-#endif
diff --git a/block/scsi_ioctl.c b/block/scsi_ioctl.c
index 5f8e798ede4e..f0e0ce0a607d 100644
--- a/block/scsi_ioctl.c
+++ b/block/scsi_ioctl.c
@@ -32,6 +32,11 @@
 #include <scsi/scsi_ioctl.h>
 #include <scsi/scsi_cmnd.h>
 
+struct blk_cmd_filter {
+	unsigned long read_ok[BLK_SCSI_CMD_PER_LONG];
+	unsigned long write_ok[BLK_SCSI_CMD_PER_LONG];
+} blk_default_cmd_filter;
+
 /* Command group 3 is reserved and should never be used.  */
 const unsigned char scsi_command_size_tbl[8] =
 {
@@ -105,7 +110,7 @@ static int sg_emulated_host(struct request_queue *q, int __user *p)
 	return put_user(1, p);
 }
 
-void blk_set_cmd_filter_defaults(struct blk_cmd_filter *filter)
+static void blk_set_cmd_filter_defaults(struct blk_cmd_filter *filter)
 {
 	/* Basic read-only commands */
 	__set_bit(TEST_UNIT_READY, filter->read_ok);
@@ -187,14 +192,37 @@ void blk_set_cmd_filter_defaults(struct blk_cmd_filter *filter)
 	__set_bit(GPCMD_SET_STREAMING, filter->write_ok);
 	__set_bit(GPCMD_SET_READ_AHEAD, filter->write_ok);
 }
-EXPORT_SYMBOL_GPL(blk_set_cmd_filter_defaults);
+
+int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm)
+{
+	struct blk_cmd_filter *filter = &blk_default_cmd_filter;
+
+	/* root can do any command. */
+	if (capable(CAP_SYS_RAWIO))
+		return 0;
+
+	/* if there's no filter set, assume we're filtering everything out */
+	if (!filter)
+		return -EPERM;
+
+	/* Anybody who can open the device can do a read-safe command */
+	if (test_bit(cmd[0], filter->read_ok))
+		return 0;
+
+	/* Write-safe commands require a writable open */
+	if (test_bit(cmd[0], filter->write_ok) && has_write_perm)
+		return 0;
+
+	return -EPERM;
+}
+EXPORT_SYMBOL(blk_verify_command);
 
 static int blk_fill_sghdr_rq(struct request_queue *q, struct request *rq,
 			     struct sg_io_hdr *hdr, fmode_t mode)
 {
 	if (copy_from_user(rq->cmd, hdr->cmdp, hdr->cmd_len))
 		return -EFAULT;
-	if (blk_verify_command(&q->cmd_filter, rq->cmd, mode & FMODE_WRITE))
+	if (blk_verify_command(rq->cmd, mode & FMODE_WRITE))
 		return -EPERM;
 
 	/*
@@ -427,7 +455,7 @@ int sg_scsi_ioctl(struct request_queue *q, struct gendisk *disk, fmode_t mode,
 	if (in_len && copy_from_user(buffer, sic->data + cmdlen, in_len))
 		goto error;
 
-	err = blk_verify_command(&q->cmd_filter, rq->cmd, mode & FMODE_WRITE);
+	err = blk_verify_command(rq->cmd, mode & FMODE_WRITE);
 	if (err)
 		goto error;
 
@@ -645,5 +673,10 @@ int scsi_cmd_ioctl(struct request_queue *q, struct gendisk *bd_disk, fmode_t mod
 	blk_put_queue(q);
 	return err;
 }
-
 EXPORT_SYMBOL(scsi_cmd_ioctl);
+
+int __init blk_scsi_ioctl_init(void)
+{
+	blk_set_cmd_filter_defaults(&blk_default_cmd_filter);
+	return 0;
+}
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index 8201387b4daa..ef142fd47a83 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -210,13 +210,11 @@ static void sg_put_dev(Sg_device *sdp);
 static int sg_allow_access(struct file *filp, unsigned char *cmd)
 {
 	struct sg_fd *sfp = (struct sg_fd *)filp->private_data;
-	struct request_queue *q = sfp->parentdp->device->request_queue;
 
 	if (sfp->parentdp->device->type == TYPE_SCANNER)
 		return 0;
 
-	return blk_verify_command(&q->cmd_filter,
-				  cmd, filp->f_mode & FMODE_WRITE);
+	return blk_verify_command(cmd, filp->f_mode & FMODE_WRITE);
 }
 
 static int
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 8963d9149b5f..49ae07951d55 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -301,12 +301,6 @@ struct blk_queue_tag {
 #define BLK_SCSI_MAX_CMDS	(256)
 #define BLK_SCSI_CMD_PER_LONG	(BLK_SCSI_MAX_CMDS / (sizeof(long) * 8))
 
-struct blk_cmd_filter {
-	unsigned long read_ok[BLK_SCSI_CMD_PER_LONG];
-	unsigned long write_ok[BLK_SCSI_CMD_PER_LONG];
-	struct kobject kobj;
-};
-
 struct queue_limits {
 	unsigned long		bounce_pfn;
 	unsigned long		seg_boundary_mask;
@@ -445,7 +439,6 @@ struct request_queue
 #if defined(CONFIG_BLK_DEV_BSG)
 	struct bsg_class_device bsg_dev;
 #endif
-	struct blk_cmd_filter cmd_filter;
 };
 
 #define QUEUE_FLAG_CLUSTER	0	/* cluster several segments into 1 */
@@ -998,13 +991,7 @@ static inline int sb_issue_discard(struct super_block *sb,
 	return blkdev_issue_discard(sb->s_bdev, block, nr_blocks, GFP_KERNEL);
 }
 
-/*
-* command filter functions
-*/
-extern int blk_verify_command(struct blk_cmd_filter *filter,
-			      unsigned char *cmd, fmode_t has_write_perm);
-extern void blk_unregister_filter(struct gendisk *disk);
-extern void blk_set_cmd_filter_defaults(struct blk_cmd_filter *filter);
+extern int blk_verify_command(unsigned char *cmd, fmode_t has_write_perm);
 
 #define MAX_PHYS_SEGMENTS 128
 #define MAX_HW_SEGMENTS 128
-- 
cgit v1.2.3-59-g8ed1b


From db64f680ba4b5c56c4be59f0698000df89ff0281 Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Tue, 30 Jun 2009 09:35:44 +0200
Subject: blocK: Restore barrier support for md and probably other virtual
 devices.

The next_ordered flag is only meaningful for devices that use __make_request.
So move the test against next_ordered out of generic code and in to
__make_request

Since this test was added, barriers have not worked on md or any
devices that don't use __make_request and so don't bother to set
next_ordered.  (dm explicitly sets something other than
QUEUE_ORDERED_NONE since
  commit 99360b4c18f7675b50d283301d46d755affe75fd
but notes in the comments that it is otherwise meaningless).

Cc: Ken Milmore <ken.milmore@googlemail.com>
Cc: stable@kernel.org
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/blk-core.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index 02b87134a167..4b45435c6eaf 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1170,6 +1170,11 @@ static int __make_request(struct request_queue *q, struct bio *bio)
 	const int unplug = bio_unplug(bio);
 	int rw_flags;
 
+	if (bio_barrier(bio) && bio_has_data(bio) &&
+	    (q->next_ordered == QUEUE_ORDERED_NONE)) {
+		bio_endio(bio, -EOPNOTSUPP);
+		return 0;
+	}
 	/*
 	 * low level driver can indicate that it wants pages above a
 	 * certain limit bounced to low memory (ie for highmem, or even
@@ -1470,11 +1475,6 @@ static inline void __generic_make_request(struct bio *bio)
 			err = -EOPNOTSUPP;
 			goto end_io;
 		}
-		if (bio_barrier(bio) && bio_has_data(bio) &&
-		    (q->next_ordered == QUEUE_ORDERED_NONE)) {
-			err = -EOPNOTSUPP;
-			goto end_io;
-		}
 
 		ret = q->make_request_fn(q, bio);
 	} while (ret);
-- 
cgit v1.2.3-59-g8ed1b


From b706f64281b24d8b1fdc8ae883700131d365c412 Mon Sep 17 00:00:00 2001
From: Shan Wei <shanwei@cn.fujitsu.com>
Date: Wed, 1 Jul 2009 12:41:14 +0200
Subject: cfq-iosched: remove redundant check for NULL cfqq in
 cfq_set_request()

With the changes for falling back to an oom_cfqq, we never fail
to find/allocate a queue in cfq_get_queue(). So remove the check.

Signed-off-by: Shan Wei <shanwei@cn.fujitsu.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/cfq-iosched.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 1d9160ffa26d..87276eb83f7f 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -2313,10 +2313,6 @@ cfq_set_request(struct request_queue *q, struct request *rq, gfp_t gfp_mask)
 	cfqq = cic_to_cfqq(cic, is_sync);
 	if (!cfqq) {
 		cfqq = cfq_get_queue(cfqd, is_sync, cic->ioc, gfp_mask);
-
-		if (!cfqq)
-			goto queue_fail;
-
 		cic_set_cfqq(cic, cfqq, is_sync);
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From 57d81f6f393b245894ca0cd828f80ce7e3294f39 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 1 Jul 2009 09:43:53 +0200
Subject: kmemleak: Fix scheduling-while-atomic bug

One of the kmemleak changes caused the following
scheduling-while-holding-the-tasklist-lock regression on x86:

BUG: sleeping function called from invalid context at mm/kmemleak.c:795
in_atomic(): 1, irqs_disabled(): 0, pid: 1737, name: kmemleak
2 locks held by kmemleak/1737:
 #0:  (scan_mutex){......}, at: [<c10c4376>] kmemleak_scan_thread+0x45/0x86
 #1:  (tasklist_lock){......}, at: [<c10c3bb4>] kmemleak_scan+0x1a9/0x39c
Pid: 1737, comm: kmemleak Not tainted 2.6.31-rc1-tip #59266
Call Trace:
 [<c105ac0f>] ? __debug_show_held_locks+0x1e/0x20
 [<c102e490>] __might_sleep+0x10a/0x111
 [<c10c38d5>] scan_yield+0x17/0x3b
 [<c10c3970>] scan_block+0x39/0xd4
 [<c10c3bc6>] kmemleak_scan+0x1bb/0x39c
 [<c10c4331>] ? kmemleak_scan_thread+0x0/0x86
 [<c10c437b>] kmemleak_scan_thread+0x4a/0x86
 [<c104d73e>] kthread+0x6e/0x73
 [<c104d6d0>] ? kthread+0x0/0x73
 [<c100959f>] kernel_thread_helper+0x7/0x10
kmemleak: 834 new suspected memory leaks (see /sys/kernel/debug/kmemleak)

The bit causing it is highly dubious:

static void scan_yield(void)
{
        might_sleep();

        if (time_is_before_eq_jiffies(next_scan_yield)) {
                schedule();
                next_scan_yield = jiffies + jiffies_scan_yield;
        }
}

It called deep inside the codepath and in a conditional way,
and that is what crapped up when one of the new scan_block()
uses grew a tasklist_lock dependency.

This minimal patch removes that yielding stuff and adds the
proper cond_resched().

The background scanning thread could probably also be reniced
to +10.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/kmemleak.c | 31 +------------------------------
 1 file changed, 1 insertion(+), 30 deletions(-)

diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index eeece2deace2..e766e1da09d2 100644
--- a/mm/kmemleak.c
+++ b/mm/kmemleak.c
@@ -105,7 +105,6 @@
 #define MAX_TRACE		16	/* stack trace length */
 #define REPORTS_NR		50	/* maximum number of reported leaks */
 #define MSECS_MIN_AGE		5000	/* minimum object age for reporting */
-#define MSECS_SCAN_YIELD	10	/* CPU yielding period */
 #define SECS_FIRST_SCAN		60	/* delay before the first scan */
 #define SECS_SCAN_WAIT		600	/* subsequent auto scanning delay */
 
@@ -186,10 +185,7 @@ static atomic_t kmemleak_error = ATOMIC_INIT(0);
 static unsigned long min_addr = ULONG_MAX;
 static unsigned long max_addr;
 
-/* used for yielding the CPU to other tasks during scanning */
-static unsigned long next_scan_yield;
 static struct task_struct *scan_thread;
-static unsigned long jiffies_scan_yield;
 /* used to avoid reporting of recently allocated objects */
 static unsigned long jiffies_min_age;
 static unsigned long jiffies_last_scan;
@@ -785,21 +781,6 @@ void kmemleak_no_scan(const void *ptr)
 }
 EXPORT_SYMBOL(kmemleak_no_scan);
 
-/*
- * Yield the CPU so that other tasks get a chance to run.  The yielding is
- * rate-limited to avoid excessive number of calls to the schedule() function
- * during memory scanning.
- */
-static void scan_yield(void)
-{
-	might_sleep();
-
-	if (time_is_before_eq_jiffies(next_scan_yield)) {
-		schedule();
-		next_scan_yield = jiffies + jiffies_scan_yield;
-	}
-}
-
 /*
  * Memory scanning is a long process and it needs to be interruptable. This
  * function checks whether such interrupt condition occured.
@@ -840,15 +821,6 @@ static void scan_block(void *_start, void *_end,
 		if (scan_should_stop())
 			break;
 
-		/*
-		 * When scanning a memory block with a corresponding
-		 * kmemleak_object, the CPU yielding is handled in the calling
-		 * code since it holds the object->lock to avoid the block
-		 * freeing.
-		 */
-		if (!scanned)
-			scan_yield();
-
 		object = find_and_get_object(pointer, 1);
 		if (!object)
 			continue;
@@ -1014,7 +986,7 @@ static void kmemleak_scan(void)
 	 */
 	object = list_entry(gray_list.next, typeof(*object), gray_list);
 	while (&object->gray_list != &gray_list) {
-		scan_yield();
+		cond_resched();
 
 		/* may add new objects to the list */
 		if (!scan_should_stop())
@@ -1385,7 +1357,6 @@ void __init kmemleak_init(void)
 	int i;
 	unsigned long flags;
 
-	jiffies_scan_yield = msecs_to_jiffies(MSECS_SCAN_YIELD);
 	jiffies_min_age = msecs_to_jiffies(MSECS_MIN_AGE);
 	jiffies_scan_wait = msecs_to_jiffies(SECS_SCAN_WAIT * 1000);
 
-- 
cgit v1.2.3-59-g8ed1b


From 3238c0c4d68d9a9022b411a11a4b933fbdb53a14 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Wed, 1 Jul 2009 18:56:16 +0100
Subject: intel-iommu: Make iommu=pt work on i386 too

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 arch/x86/kernel/pci-dma.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index 47630479b067..1a041bcf506b 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -211,11 +211,11 @@ static __init int iommu_setup(char *p)
 #ifdef CONFIG_SWIOTLB
 		if (!strncmp(p, "soft", 4))
 			swiotlb = 1;
+#endif
 		if (!strncmp(p, "pt", 2)) {
 			iommu_pass_through = 1;
 			return 1;
 		}
-#endif
 
 		gart_parse_options(p);
 
-- 
cgit v1.2.3-59-g8ed1b


From e2dbe12557d85d81f4527879499f55681c3cca4f Mon Sep 17 00:00:00 2001
From: Amerigo Wang <amwang@redhat.com>
Date: Wed, 1 Jul 2009 01:06:26 -0400
Subject: elf: fix one check-after-use

Check before use it.

Signed-off-by: WANG Cong <amwang@redhat.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: David Howells <dhowells@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/binfmt_elf.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index f1867900e459..b7c1603cd4bd 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -1522,11 +1522,11 @@ static int fill_note_info(struct elfhdr *elf, int phdrs,
 	info->thread = NULL;
 
 	psinfo = kmalloc(sizeof(*psinfo), GFP_KERNEL);
-	fill_note(&info->psinfo, "CORE", NT_PRPSINFO, sizeof(*psinfo), psinfo);
-
 	if (psinfo == NULL)
 		return 0;
 
+	fill_note(&info->psinfo, "CORE", NT_PRPSINFO, sizeof(*psinfo), psinfo);
+
 	/*
 	 * Figure out how many notes we're going to need for each thread.
 	 */
-- 
cgit v1.2.3-59-g8ed1b


From 788d84bba47ea3eb377f7a3ae4fd1ee84b84877b Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Wed, 1 Jul 2009 18:34:52 +0100
Subject: Fix pci_unmap_addr() et al on i386.

We can run a 32-bit kernel on boxes with an IOMMU, so we need
pci_unmap_addr() etc. to work -- without it, drivers will leak mappings.

To be honest, this whole thing looks like it's more pain than it's
worth; I'm half inclined to remove the no-op #else case altogether.

But this is the minimal fix, which just does the right thing if
CONFIG_DMAR is set.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Cc: stable@kernel.org  [ for 2.6.30 ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/include/asm/pci.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/include/asm/pci.h b/arch/x86/include/asm/pci.h
index 927958d13c19..1ff685ca221c 100644
--- a/arch/x86/include/asm/pci.h
+++ b/arch/x86/include/asm/pci.h
@@ -91,7 +91,7 @@ extern void pci_iommu_alloc(void);
 
 #define PCI_DMA_BUS_IS_PHYS (dma_ops->is_phys)
 
-#if defined(CONFIG_X86_64) || defined(CONFIG_DMA_API_DEBUG)
+#if defined(CONFIG_X86_64) || defined(CONFIG_DMAR) || defined(CONFIG_DMA_API_DEBUG)
 
 #define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)       \
 	        dma_addr_t ADDR_NAME;
-- 
cgit v1.2.3-59-g8ed1b


From a15a519ed6e5e644f5a33c213c00b0c1d3cfe683 Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Wed, 1 Jul 2009 18:49:06 +0100
Subject: Fix iommu address space allocation

This fixes kernel.org bug #13584. The IOVA code attempted to optimise
the insertion of new ranges into the rbtree, with the unfortunate result
that some ranges just didn't get inserted into the tree at all. Then
those ranges would be handed out more than once, and things kind of go
downhill from there.

Introduced after 2.6.25 by ddf02886cbe665d67ca750750196ea5bf524b10b
("PCI: iova RB tree setup tweak").

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Cc: mark gross <mgross@linux.intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/pci/iova.c | 26 ++++++++++++++++++++++----
 1 file changed, 22 insertions(+), 4 deletions(-)

diff --git a/drivers/pci/iova.c b/drivers/pci/iova.c
index 2287116e9822..46dd440e2315 100644
--- a/drivers/pci/iova.c
+++ b/drivers/pci/iova.c
@@ -1,9 +1,19 @@
 /*
- * Copyright (c) 2006, Intel Corporation.
+ * Copyright © 2006-2009, Intel Corporation.
  *
- * This file is released under the GPLv2.
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
+ * Place - Suite 330, Boston, MA 02111-1307 USA.
  *
- * Copyright (C) 2006-2008 Intel Corporation
  * Author: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
  */
 
@@ -123,7 +133,15 @@ move_left:
 	/* Insert the new_iova into domain rbtree by holding writer lock */
 	/* Add new node and rebalance tree. */
 	{
-		struct rb_node **entry = &((prev)), *parent = NULL;
+		struct rb_node **entry, *parent = NULL;
+
+		/* If we have 'prev', it's a valid place to start the
+		   insertion. Otherwise, start from the root. */
+		if (prev)
+			entry = &prev;
+		else
+			entry = &iovad->rbroot.rb_node;
+
 		/* Figure out where to put new node */
 		while (*entry) {
 			struct iova *this = container_of(*entry,
-- 
cgit v1.2.3-59-g8ed1b


From c85994e4771025ef2a66533eb1a4c6c2217b9cda Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Wed, 1 Jul 2009 19:21:24 +0100
Subject: intel-iommu: Ensure that PTE writes are 64-bit atomic, even on i386

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 37 +++++++++++++++++++++++--------------
 1 file changed, 23 insertions(+), 14 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index ec7e032d5ab5..eea1006c860a 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -222,7 +222,12 @@ static inline void dma_set_pte_prot(struct dma_pte *pte, unsigned long prot)
 
 static inline u64 dma_pte_addr(struct dma_pte *pte)
 {
-	return (pte->val & VTD_PAGE_MASK);
+#ifdef CONFIG_64BIT
+	return pte->val & VTD_PAGE_MASK;
+#else
+	/* Must have a full atomic 64-bit read */
+	return  __cmpxchg64(pte, 0ULL, 0ULL) & VTD_PAGE_MASK;
+#endif
 }
 
 static inline void dma_set_pte_pfn(struct dma_pte *pte, unsigned long pfn)
@@ -712,6 +717,8 @@ static struct dma_pte *pfn_to_dma_pte(struct dmar_domain *domain,
 			break;
 
 		if (!dma_pte_present(pte)) {
+			uint64_t pteval;
+
 			tmp_page = alloc_pgtable_page();
 
 			if (!tmp_page) {
@@ -719,15 +726,15 @@ static struct dma_pte *pfn_to_dma_pte(struct dmar_domain *domain,
 					flags);
 				return NULL;
 			}
-			domain_flush_cache(domain, tmp_page, PAGE_SIZE);
-			dma_set_pte_pfn(pte, virt_to_dma_pfn(tmp_page));
-			/*
-			 * high level table always sets r/w, last level page
-			 * table control read/write
-			 */
-			dma_set_pte_readable(pte);
-			dma_set_pte_writable(pte);
-			domain_flush_cache(domain, pte, sizeof(*pte));
+			domain_flush_cache(domain, tmp_page, VTD_PAGE_SIZE);
+			pteval = (virt_to_dma_pfn(tmp_page) << VTD_PAGE_SHIFT) | DMA_PTE_READ | DMA_PTE_WRITE;
+			if (cmpxchg64(&pte->val, 0ULL, pteval)) {
+				/* Someone else set it while we were thinking; use theirs. */
+				free_pgtable_page(tmp_page);
+			} else {
+				dma_pte_addr(pte);
+				domain_flush_cache(domain, pte, sizeof(*pte));
+			}
 		}
 		parent = phys_to_virt(dma_pte_addr(pte));
 		level--;
@@ -1666,6 +1673,8 @@ static int __domain_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
 	}
 
 	while (nr_pages--) {
+		uint64_t tmp;
+
 		if (!sg_res) {
 			sg_res = (sg->offset + sg->length + VTD_PAGE_SIZE - 1) >> VTD_PAGE_SHIFT;
 			sg->dma_address = ((dma_addr_t)iov_pfn << VTD_PAGE_SHIFT) + sg->offset;
@@ -1680,17 +1689,17 @@ static int __domain_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
 		/* We don't need lock here, nobody else
 		 * touches the iova range
 		 */
-		if (unlikely(dma_pte_addr(pte))) {
+		tmp = cmpxchg64(&pte->val, 0ULL, pteval);
+		if (tmp) {
 			static int dumps = 5;
-			printk(KERN_CRIT "ERROR: DMA PTE for vPFN 0x%lx already set (to %llx)\n",
-			       iov_pfn, pte->val);
+			printk(KERN_CRIT "ERROR: DMA PTE for vPFN 0x%lx already set (to %llx not %llx)\n",
+			       iov_pfn, tmp, (unsigned long long)pteval);
 			if (dumps) {
 				dumps--;
 				debug_dma_dump_mappings(NULL);
 			}
 			WARN_ON(1);
 		}
-		pte->val = pteval;
 		pte++;
 		if (!nr_pages ||
 		    (unsigned long)pte >> VTD_PAGE_SHIFT !=
-- 
cgit v1.2.3-59-g8ed1b


From d960eea974f5e500c0dcb95a934239cc1f481cfd Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Mon, 29 Jun 2009 14:54:11 -0700
Subject: kernel-doc: move ignoring kmemcheck

Somehow I managed to generate a diff that put these 2 lines
into the wrong function:  should have been in dump_struct()
instead of in dump_enum().

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/kernel-doc | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/scripts/kernel-doc b/scripts/kernel-doc
index ed591e9b7d1d..b52d340d759d 100755
--- a/scripts/kernel-doc
+++ b/scripts/kernel-doc
@@ -1426,6 +1426,8 @@ sub dump_struct($$) {
 	# strip comments:
 	$members =~ s/\/\*.*?\*\///gos;
 	$nested =~ s/\/\*.*?\*\///gos;
+	# strip kmemcheck_bitfield_{begin,end}.*;
+	$members =~ s/kmemcheck_bitfield_.*?;//gos;
 
 	create_parameterlist($members, ';', $file);
 	check_sections($file, $declaration_name, "struct", $sectcheck, $struct_actual, $nested);
@@ -1468,8 +1470,6 @@ sub dump_enum($$) {
 	    }
 
 	}
-	# strip kmemcheck_bitfield_{begin,end}.*;
-	$members =~ s/kmemcheck_bitfield_.*?;//gos;
 
 	output_declaration($declaration_name,
 			   'enum',
-- 
cgit v1.2.3-59-g8ed1b


From 34e19ada994fb9cb3d11873f2b734602e2135f3a Mon Sep 17 00:00:00 2001
From: Matt Fleming <matt@console-pimps.org>
Date: Wed, 1 Jul 2009 18:14:18 +0000
Subject: sh: Fix compiler error and include the definition of IS_ERR_VALUE

When arch/sh/include/asm/syscall_32.h is included from a file that
doesn't also include linux/err.h the following error is produced,

In file included from /home/matt/src/kernels/sh-2.6/arch/sh/include/asm/syscall.h:5,
                 from kernel/trace/trace_syscalls.c:3:
/home/matt/src/kernels/sh-2.6/arch/sh/include/asm/syscall_32.h: In function 'syscall_get_error':
/home/matt/src/kernels/sh-2.6/arch/sh/include/asm/syscall_32.h:28: error: implicit declaration of function 'IS_ERR_VALUE'
make[2]: *** [kernel/trace/trace_syscalls.o] Error 1
make[1]: *** [kernel/trace] Error 2
make: *** [kernel] Error 2

Signed-off-by: Matt Fleming <matt@console-pimps.org>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/include/asm/syscall_32.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/sh/include/asm/syscall_32.h b/arch/sh/include/asm/syscall_32.h
index 5bc34681d994..6f83f2cc45c1 100644
--- a/arch/sh/include/asm/syscall_32.h
+++ b/arch/sh/include/asm/syscall_32.h
@@ -3,6 +3,7 @@
 
 #include <linux/kernel.h>
 #include <linux/sched.h>
+#include <linux/err.h>
 #include <asm/ptrace.h>
 
 /* The system call number is given by the user in R3 */
-- 
cgit v1.2.3-59-g8ed1b


From 1c6a307a54668eda556f499c94e75086aaf8f80f Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 1 Jul 2009 06:50:31 +0000
Subject: sh: LCDC dcache flush for deferred io

Since writenotify on uncached vmas is unsupported in 2.6.31,
live with cached framebuffer memory in the deferred io
case for now and flush the dcache before forcing refresh.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Acked-by: Magnus damm <damm@igel.co.jp>
---
 drivers/video/sh_mobile_lcdcfb.c | 53 ++++++++++++++++++++++++++++++++++------
 1 file changed, 45 insertions(+), 8 deletions(-)

diff --git a/drivers/video/sh_mobile_lcdcfb.c b/drivers/video/sh_mobile_lcdcfb.c
index f10d2fbeda06..da983b720f08 100644
--- a/drivers/video/sh_mobile_lcdcfb.c
+++ b/drivers/video/sh_mobile_lcdcfb.c
@@ -17,6 +17,7 @@
 #include <linux/platform_device.h>
 #include <linux/dma-mapping.h>
 #include <linux/interrupt.h>
+#include <linux/vmalloc.h>
 #include <video/sh_mobile_lcdc.h>
 #include <asm/atomic.h>
 
@@ -33,6 +34,7 @@ struct sh_mobile_lcdc_chan {
 	struct fb_info info;
 	dma_addr_t dma_handle;
 	struct fb_deferred_io defio;
+	struct scatterlist *sglist;
 	unsigned long frame_end;
 	wait_queue_head_t frame_end_wait;
 };
@@ -206,16 +208,38 @@ static void sh_mobile_lcdc_clk_on(struct sh_mobile_lcdc_priv *priv) {}
 static void sh_mobile_lcdc_clk_off(struct sh_mobile_lcdc_priv *priv) {}
 #endif
 
+static int sh_mobile_lcdc_sginit(struct fb_info *info,
+				  struct list_head *pagelist)
+{
+	struct sh_mobile_lcdc_chan *ch = info->par;
+	unsigned int nr_pages_max = info->fix.smem_len >> PAGE_SHIFT;
+	struct page *page;
+	int nr_pages = 0;
+
+	sg_init_table(ch->sglist, nr_pages_max);
+
+	list_for_each_entry(page, pagelist, lru)
+		sg_set_page(&ch->sglist[nr_pages++], page, PAGE_SIZE, 0);
+
+	return nr_pages;
+}
+
 static void sh_mobile_lcdc_deferred_io(struct fb_info *info,
 				       struct list_head *pagelist)
 {
 	struct sh_mobile_lcdc_chan *ch = info->par;
+	unsigned int nr_pages;
 
 	/* enable clocks before accessing hardware */
 	sh_mobile_lcdc_clk_on(ch->lcdc);
 
+	nr_pages = sh_mobile_lcdc_sginit(info, pagelist);
+	dma_map_sg(info->dev, ch->sglist, nr_pages, DMA_TO_DEVICE);
+
 	/* trigger panel update */
 	lcdc_write_chan(ch, LDSM2R, 1);
+
+	dma_unmap_sg(info->dev, ch->sglist, nr_pages, DMA_TO_DEVICE);
 }
 
 static void sh_mobile_lcdc_deferred_io_touch(struct fb_info *info)
@@ -846,21 +870,31 @@ static int __init sh_mobile_lcdc_probe(struct platform_device *pdev)
 	}
 
 	for (i = 0; i < j; i++) {
-		error = register_framebuffer(&priv->ch[i].info);
+		struct sh_mobile_lcdc_chan *ch = priv->ch + i;
+
+		info = &ch->info;
+
+		if (info->fbdefio) {
+			priv->ch->sglist = vmalloc(sizeof(struct scatterlist) *
+					info->fix.smem_len >> PAGE_SHIFT);
+			if (!priv->ch->sglist) {
+				dev_err(&pdev->dev, "cannot allocate sglist\n");
+				goto err1;
+			}
+		}
+
+		error = register_framebuffer(info);
 		if (error < 0)
 			goto err1;
-	}
 
-	for (i = 0; i < j; i++) {
-		info = &priv->ch[i].info;
 		dev_info(info->dev,
 			 "registered %s/%s as %dx%d %dbpp.\n",
 			 pdev->name,
-			 (priv->ch[i].cfg.chan == LCDC_CHAN_MAINLCD) ?
+			 (ch->cfg.chan == LCDC_CHAN_MAINLCD) ?
 			 "mainlcd" : "sublcd",
-			 (int) priv->ch[i].cfg.lcd_cfg.xres,
-			 (int) priv->ch[i].cfg.lcd_cfg.yres,
-			 priv->ch[i].cfg.bpp);
+			 (int) ch->cfg.lcd_cfg.xres,
+			 (int) ch->cfg.lcd_cfg.yres,
+			 ch->cfg.bpp);
 
 		/* deferred io mode: disable clock to save power */
 		if (info->fbdefio)
@@ -892,6 +926,9 @@ static int sh_mobile_lcdc_remove(struct platform_device *pdev)
 		if (!info->device)
 			continue;
 
+		if (priv->ch[i].sglist)
+			vfree(priv->ch[i].sglist);
+
 		dma_free_coherent(&pdev->dev, info->fix.smem_len,
 				  info->screen_base, priv->ch[i].dma_handle);
 		fb_dealloc_cmap(&info->cmap);
-- 
cgit v1.2.3-59-g8ed1b


From 206a73c102fc480ba072a9388bc2142c303113aa Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Wed, 1 Jul 2009 19:30:28 +0100
Subject: intel-iommu: Kill superfluous mapping_lock

Since we're using cmpxchg64() anyway (because that's the only way to do
an atomic 64-bit store on i386), we might as well ditch the extra
locking and just use cmpxchg64() to ensure that we don't add the page
twice.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 12 ++----------
 1 file changed, 2 insertions(+), 10 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index eea1006c860a..02223e2e27d4 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -267,7 +267,6 @@ struct dmar_domain {
 	struct iova_domain iovad;	/* iova's that belong to this domain */
 
 	struct dma_pte	*pgd;		/* virtual address */
-	spinlock_t	mapping_lock;	/* page table lock */
 	int		gaw;		/* max guest address width */
 
 	/* adjusted guest address width, 0 is level 2 30-bit */
@@ -701,13 +700,11 @@ static struct dma_pte *pfn_to_dma_pte(struct dmar_domain *domain,
 	struct dma_pte *parent, *pte = NULL;
 	int level = agaw_to_level(domain->agaw);
 	int offset;
-	unsigned long flags;
 
 	BUG_ON(!domain->pgd);
 	BUG_ON(addr_width < BITS_PER_LONG && pfn >> addr_width);
 	parent = domain->pgd;
 
-	spin_lock_irqsave(&domain->mapping_lock, flags);
 	while (level > 0) {
 		void *tmp_page;
 
@@ -721,11 +718,9 @@ static struct dma_pte *pfn_to_dma_pte(struct dmar_domain *domain,
 
 			tmp_page = alloc_pgtable_page();
 
-			if (!tmp_page) {
-				spin_unlock_irqrestore(&domain->mapping_lock,
-					flags);
+			if (!tmp_page)
 				return NULL;
-			}
+
 			domain_flush_cache(domain, tmp_page, VTD_PAGE_SIZE);
 			pteval = (virt_to_dma_pfn(tmp_page) << VTD_PAGE_SHIFT) | DMA_PTE_READ | DMA_PTE_WRITE;
 			if (cmpxchg64(&pte->val, 0ULL, pteval)) {
@@ -740,7 +735,6 @@ static struct dma_pte *pfn_to_dma_pte(struct dmar_domain *domain,
 		level--;
 	}
 
-	spin_unlock_irqrestore(&domain->mapping_lock, flags);
 	return pte;
 }
 
@@ -1375,7 +1369,6 @@ static int domain_init(struct dmar_domain *domain, int guest_width)
 	unsigned long sagaw;
 
 	init_iova_domain(&domain->iovad, DMA_32BIT_PFN);
-	spin_lock_init(&domain->mapping_lock);
 	spin_lock_init(&domain->iommu_lock);
 
 	domain_reserve_special_ranges(domain);
@@ -3336,7 +3329,6 @@ static int md_domain_init(struct dmar_domain *domain, int guest_width)
 	int adjust_width;
 
 	init_iova_domain(&domain->iovad, DMA_32BIT_PFN);
-	spin_lock_init(&domain->mapping_lock);
 	spin_lock_init(&domain->iommu_lock);
 
 	domain_reserve_special_ranges(domain);
-- 
cgit v1.2.3-59-g8ed1b


From 85b98276f2ffa66b25ae6328b00bfadfd74b74e7 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Wed, 1 Jul 2009 19:27:53 +0100
Subject: intel-iommu: Warn about unmatched unmap requests

This would have found the bug in i386 pci_unmap_addr() a long time ago.
We shouldn't just silently return without doing anything.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 02223e2e27d4..2bbc3fc88dc9 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -2640,7 +2640,8 @@ static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr,
 	iommu = domain_get_iommu(domain);
 
 	iova = find_iova(&domain->iovad, IOVA_PFN(dev_addr));
-	if (!iova)
+	if (WARN_ONCE(!iova, "Driver unmaps unmatched page at PFN %llx\n",
+		      (unsigned long long)dev_addr))
 		return;
 
 	start_pfn = mm_to_dma_pfn(iova->pfn_lo);
@@ -2730,7 +2731,8 @@ static void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist,
 	iommu = domain_get_iommu(domain);
 
 	iova = find_iova(&domain->iovad, IOVA_PFN(sglist[0].dma_address));
-	if (!iova)
+	if (WARN_ONCE(!iova, "Driver unmaps unmatched sglist at PFN %llx\n",
+		      (unsigned long long)sglist[0].dma_address))
 		return;
 
 	start_pfn = mm_to_dma_pfn(iova->pfn_lo);
-- 
cgit v1.2.3-59-g8ed1b


From 7766a3fb905f0b078b05f5d6a6be8df4c64b9f51 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Wed, 1 Jul 2009 20:27:03 +0100
Subject: intel-iommu: Use cmpxchg64_local() for setting PTEs

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 2bbc3fc88dc9..2c1b2babfdc5 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1682,7 +1682,7 @@ static int __domain_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
 		/* We don't need lock here, nobody else
 		 * touches the iova range
 		 */
-		tmp = cmpxchg64(&pte->val, 0ULL, pteval);
+		tmp = cmpxchg64_local(&pte->val, 0ULL, pteval);
 		if (tmp) {
 			static int dumps = 5;
 			printk(KERN_CRIT "ERROR: DMA PTE for vPFN 0x%lx already set (to %llx not %llx)\n",
-- 
cgit v1.2.3-59-g8ed1b


From 00460f41fffc0435dbb6ab4b058a190163d57ce6 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Thu, 2 Jul 2009 00:46:16 +0100
Subject: FRV: Implement atomic64_t

Implement atomic64_t and its ops for FRV.  Tested with the following patch:

	diff --git a/arch/frv/kernel/setup.c b/arch/frv/kernel/setup.c
	index 55e4fab..086d50d 100644
	--- a/arch/frv/kernel/setup.c
	+++ b/arch/frv/kernel/setup.c
	@@ -746,6 +746,52 @@ static void __init parse_cmdline_early(char *cmdline)

	 } /* end parse_cmdline_early() */

	+static atomic64_t xxx;
	+
	+static void test_atomic64(void)
	+{
	+	atomic64_set(&xxx, 0x12300000023LL);
	+
	+	mb();
	+	BUG_ON(atomic64_read(&xxx) != 0x12300000023LL);
	+	mb();
	+	if (atomic64_inc_return(&xxx) != 0x12300000024LL)
	+		BUG();
	+	mb();
	+	BUG_ON(atomic64_read(&xxx) != 0x12300000024LL);
	+	mb();
	+	if (atomic64_sub_return(0x36900000050LL, &xxx) != -0x2460000002cLL)
	+		BUG();
	+	mb();
	+	BUG_ON(atomic64_read(&xxx) != -0x2460000002cLL);
	+	mb();
	+	if (atomic64_dec_return(&xxx) != -0x2460000002dLL)
	+		BUG();
	+	mb();
	+	BUG_ON(atomic64_read(&xxx) != -0x2460000002dLL);
	+	mb();
	+	if (atomic64_add_return(0x36800000001LL, &xxx) != 0x121ffffffd4LL)
	+		BUG();
	+	mb();
	+	BUG_ON(atomic64_read(&xxx) != 0x121ffffffd4LL);
	+	mb();
	+	if (atomic64_cmpxchg(&xxx, 0x123456789abcdefLL, 0x121ffffffd4LL) != 0x121ffffffd4LL)
	+		BUG();
	+	mb();
	+	BUG_ON(atomic64_read(&xxx) != 0x121ffffffd4LL);
	+	mb();
	+	if (atomic64_cmpxchg(&xxx, 0x121ffffffd4LL, 0x123456789abcdefLL) != 0x121ffffffd4LL)
	+		BUG();
	+	mb();
	+	BUG_ON(atomic64_read(&xxx) != 0x123456789abcdefLL);
	+	mb();
	+	if (atomic64_xchg(&xxx, 0xabcdef123456789LL) != 0x123456789abcdefLL)
	+		BUG();
	+	mb();
	+	BUG_ON(atomic64_read(&xxx) != 0xabcdef123456789LL);
	+	mb();
	+}
	+
	 /*****************************************************************************/
	 /*
	  *
	@@ -845,6 +891,8 @@ void __init setup_arch(char **cmdline_p)
	 //	asm volatile("movgs %0,timerd" :: "r"(10000000));
	 //	__set_HSR(0, __get_HSR(0) | HSR0_ETMD);

	+	test_atomic64();
	+
	 } /* end setup_arch() */

	 #if 0

Note that this doesn't cover all the trivial wrappers, but does cover all the
substantial implementations.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/frv/include/asm/atomic.h |  68 +++++++++++++++++-
 arch/frv/include/asm/system.h |   2 +
 arch/frv/kernel/frv_ksyms.c   |   4 ++
 arch/frv/lib/Makefile         |   2 +-
 arch/frv/lib/atomic-ops.S     |   3 +-
 arch/frv/lib/atomic64-ops.S   | 162 ++++++++++++++++++++++++++++++++++++++++++
 6 files changed, 236 insertions(+), 5 deletions(-)
 create mode 100644 arch/frv/lib/atomic64-ops.S

diff --git a/arch/frv/include/asm/atomic.h b/arch/frv/include/asm/atomic.h
index 0409d981fd39..00a57af79afc 100644
--- a/arch/frv/include/asm/atomic.h
+++ b/arch/frv/include/asm/atomic.h
@@ -121,10 +121,72 @@ static inline void atomic_dec(atomic_t *v)
 #define atomic_dec_and_test(v)		(atomic_sub_return(1, (v)) == 0)
 #define atomic_inc_and_test(v)		(atomic_add_return(1, (v)) == 0)
 
+/*
+ * 64-bit atomic ops
+ */
+typedef struct {
+	volatile long long counter;
+} atomic64_t;
+
+#define ATOMIC64_INIT(i)	{ (i) }
+
+static inline long long atomic64_read(atomic64_t *v)
+{
+	long long counter;
+
+	asm("ldd%I1 %M1,%0"
+	    : "=e"(counter)
+	    : "m"(v->counter));
+	return counter;
+}
+
+static inline void atomic64_set(atomic64_t *v, long long i)
+{
+	asm volatile("std%I0 %1,%M0"
+		     : "=m"(v->counter)
+		     : "e"(i));
+}
+
+extern long long atomic64_inc_return(atomic64_t *v);
+extern long long atomic64_dec_return(atomic64_t *v);
+extern long long atomic64_add_return(long long i, atomic64_t *v);
+extern long long atomic64_sub_return(long long i, atomic64_t *v);
+
+static inline long long atomic64_add_negative(long long i, atomic64_t *v)
+{
+	return atomic64_add_return(i, v) < 0;
+}
+
+static inline void atomic64_add(long long i, atomic64_t *v)
+{
+	atomic64_add_return(i, v);
+}
+
+static inline void atomic64_sub(long long i, atomic64_t *v)
+{
+	atomic64_sub_return(i, v);
+}
+
+static inline void atomic64_inc(atomic64_t *v)
+{
+	atomic64_inc_return(v);
+}
+
+static inline void atomic64_dec(atomic64_t *v)
+{
+	atomic64_dec_return(v);
+}
+
+#define atomic64_sub_and_test(i,v)	(atomic64_sub_return((i), (v)) == 0)
+#define atomic64_dec_and_test(v)	(atomic64_dec_return((v)) == 0)
+#define atomic64_inc_and_test(v)	(atomic64_inc_return((v)) == 0)
+
 /*****************************************************************************/
 /*
  * exchange value with memory
  */
+extern uint64_t __xchg_64(uint64_t i, volatile void *v);
+
 #ifndef CONFIG_FRV_OUTOFLINE_ATOMIC_OPS
 
 #define xchg(ptr, x)								\
@@ -174,8 +236,10 @@ extern uint32_t __xchg_32(uint32_t i, volatile void *v);
 
 #define tas(ptr) (xchg((ptr), 1))
 
-#define atomic_cmpxchg(v, old, new) (cmpxchg(&((v)->counter), old, new))
-#define atomic_xchg(v, new) (xchg(&((v)->counter), new))
+#define atomic_cmpxchg(v, old, new)	(cmpxchg(&(v)->counter, old, new))
+#define atomic_xchg(v, new)		(xchg(&(v)->counter, new))
+#define atomic64_cmpxchg(v, old, new)	(__cmpxchg_64(old, new, &(v)->counter))
+#define atomic64_xchg(v, new)		(__xchg_64(new, &(v)->counter))
 
 static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
 {
diff --git a/arch/frv/include/asm/system.h b/arch/frv/include/asm/system.h
index 7742ec000cc4..efd22d9077ac 100644
--- a/arch/frv/include/asm/system.h
+++ b/arch/frv/include/asm/system.h
@@ -208,6 +208,8 @@ extern void free_initmem(void);
  * - if (*ptr == test) then orig = *ptr; *ptr = test;
  * - if (*ptr != test) then orig = *ptr;
  */
+extern uint64_t __cmpxchg_64(uint64_t test, uint64_t new, volatile uint64_t *v);
+
 #ifndef CONFIG_FRV_OUTOFLINE_ATOMIC_OPS
 
 #define cmpxchg(ptr, test, new)							\
diff --git a/arch/frv/kernel/frv_ksyms.c b/arch/frv/kernel/frv_ksyms.c
index 0316b3c50eff..a89803b58b9a 100644
--- a/arch/frv/kernel/frv_ksyms.c
+++ b/arch/frv/kernel/frv_ksyms.c
@@ -67,6 +67,10 @@ EXPORT_SYMBOL(atomic_sub_return);
 EXPORT_SYMBOL(__xchg_32);
 EXPORT_SYMBOL(__cmpxchg_32);
 #endif
+EXPORT_SYMBOL(atomic64_add_return);
+EXPORT_SYMBOL(atomic64_sub_return);
+EXPORT_SYMBOL(__xchg_64);
+EXPORT_SYMBOL(__cmpxchg_64);
 
 EXPORT_SYMBOL(__debug_bug_printk);
 EXPORT_SYMBOL(__delay_loops_MHz);
diff --git a/arch/frv/lib/Makefile b/arch/frv/lib/Makefile
index 08be305c9f44..4ff2fb1e6b16 100644
--- a/arch/frv/lib/Makefile
+++ b/arch/frv/lib/Makefile
@@ -4,5 +4,5 @@
 
 lib-y := \
 	__ashldi3.o __lshrdi3.o __muldi3.o __ashrdi3.o __negdi2.o __ucmpdi2.o \
-	checksum.o memcpy.o memset.o atomic-ops.o \
+	checksum.o memcpy.o memset.o atomic-ops.o atomic64-ops.o \
 	outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o
diff --git a/arch/frv/lib/atomic-ops.S b/arch/frv/lib/atomic-ops.S
index ee0ac905fb08..5e9e6ab5dd0e 100644
--- a/arch/frv/lib/atomic-ops.S
+++ b/arch/frv/lib/atomic-ops.S
@@ -163,11 +163,10 @@ __cmpxchg_32:
 	ld.p		@(gr11,gr0),gr8
 	orcr		cc7,cc7,cc3
 	subcc		gr8,gr9,gr7,icc0
-	bne		icc0,#0,1f
+	bnelr		icc0,#0
 	cst.p		gr10,@(gr11,gr0)	,cc3,#1
 	corcc		gr29,gr29,gr0		,cc3,#1
 	beq		icc3,#0,0b
-1:
 	bralr
 
 	.size		__cmpxchg_32, .-__cmpxchg_32
diff --git a/arch/frv/lib/atomic64-ops.S b/arch/frv/lib/atomic64-ops.S
new file mode 100644
index 000000000000..b6194eeac127
--- /dev/null
+++ b/arch/frv/lib/atomic64-ops.S
@@ -0,0 +1,162 @@
+/* kernel atomic64 operations
+ *
+ * For an explanation of how atomic ops work in this arch, see:
+ *   Documentation/frv/atomic-ops.txt
+ *
+ * Copyright (C) 2009 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <asm/spr-regs.h>
+
+	.text
+	.balign 4
+
+
+###############################################################################
+#
+# long long atomic64_inc_return(atomic64_t *v)
+#
+###############################################################################
+	.globl		atomic64_inc_return
+        .type		atomic64_inc_return,@function
+atomic64_inc_return:
+	or.p		gr8,gr8,gr10
+0:
+	orcc		gr0,gr0,gr0,icc3		/* set ICC3.Z */
+	ckeq		icc3,cc7
+	ldd.p		@(gr10,gr0),gr8			/* LDD.P/ORCR must be atomic */
+	orcr		cc7,cc7,cc3			/* set CC3 to true */
+	addicc		gr9,#1,gr9,icc0
+	addxi		gr8,#0,gr8,icc0
+	cstd.p		gr8,@(gr10,gr0)		,cc3,#1
+	corcc		gr29,gr29,gr0		,cc3,#1	/* clear ICC3.Z if store happens */
+	beq		icc3,#0,0b
+	bralr
+
+	.size		atomic64_inc_return, .-atomic64_inc_return
+
+###############################################################################
+#
+# long long atomic64_dec_return(atomic64_t *v)
+#
+###############################################################################
+	.globl		atomic64_dec_return
+        .type		atomic64_dec_return,@function
+atomic64_dec_return:
+	or.p		gr8,gr8,gr10
+0:
+	orcc		gr0,gr0,gr0,icc3		/* set ICC3.Z */
+	ckeq		icc3,cc7
+	ldd.p		@(gr10,gr0),gr8			/* LDD.P/ORCR must be atomic */
+	orcr		cc7,cc7,cc3			/* set CC3 to true */
+	subicc		gr9,#1,gr9,icc0
+	subxi		gr8,#0,gr8,icc0
+	cstd.p		gr8,@(gr10,gr0)		,cc3,#1
+	corcc		gr29,gr29,gr0		,cc3,#1	/* clear ICC3.Z if store happens */
+	beq		icc3,#0,0b
+	bralr
+
+	.size		atomic64_dec_return, .-atomic64_dec_return
+
+###############################################################################
+#
+# long long atomic64_add_return(long long i, atomic64_t *v)
+#
+###############################################################################
+	.globl		atomic64_add_return
+        .type		atomic64_add_return,@function
+atomic64_add_return:
+	or.p		gr8,gr8,gr4
+	or		gr9,gr9,gr5
+0:
+	orcc		gr0,gr0,gr0,icc3		/* set ICC3.Z */
+	ckeq		icc3,cc7
+	ldd.p		@(gr10,gr0),gr8			/* LDD.P/ORCR must be atomic */
+	orcr		cc7,cc7,cc3			/* set CC3 to true */
+	addcc		gr9,gr5,gr9,icc0
+	addx		gr8,gr4,gr8,icc0
+	cstd.p		gr8,@(gr10,gr0)		,cc3,#1
+	corcc		gr29,gr29,gr0		,cc3,#1	/* clear ICC3.Z if store happens */
+	beq		icc3,#0,0b
+	bralr
+
+	.size		atomic64_add_return, .-atomic64_add_return
+
+###############################################################################
+#
+# long long atomic64_sub_return(long long i, atomic64_t *v)
+#
+###############################################################################
+	.globl		atomic64_sub_return
+        .type		atomic64_sub_return,@function
+atomic64_sub_return:
+	or.p		gr8,gr8,gr4
+	or		gr9,gr9,gr5
+0:
+	orcc		gr0,gr0,gr0,icc3		/* set ICC3.Z */
+	ckeq		icc3,cc7
+	ldd.p		@(gr10,gr0),gr8			/* LDD.P/ORCR must be atomic */
+	orcr		cc7,cc7,cc3			/* set CC3 to true */
+	subcc		gr9,gr5,gr9,icc0
+	subx		gr8,gr4,gr8,icc0
+	cstd.p		gr8,@(gr10,gr0)		,cc3,#1
+	corcc		gr29,gr29,gr0		,cc3,#1	/* clear ICC3.Z if store happens */
+	beq		icc3,#0,0b
+	bralr
+
+	.size		atomic64_sub_return, .-atomic64_sub_return
+
+###############################################################################
+#
+# uint64_t __xchg_64(uint64_t i, uint64_t *v)
+#
+###############################################################################
+	.globl		__xchg_64
+        .type		__xchg_64,@function
+__xchg_64:
+	or.p		gr8,gr8,gr4
+	or		gr9,gr9,gr5
+0:
+	orcc		gr0,gr0,gr0,icc3		/* set ICC3.Z */
+	ckeq		icc3,cc7
+	ldd.p		@(gr10,gr0),gr8			/* LDD.P/ORCR must be atomic */
+	orcr		cc7,cc7,cc3			/* set CC3 to true */
+	cstd.p		gr4,@(gr10,gr0)		,cc3,#1
+	corcc		gr29,gr29,gr0		,cc3,#1	/* clear ICC3.Z if store happens */
+	beq		icc3,#0,0b
+	bralr
+
+	.size		__xchg_64, .-__xchg_64
+
+###############################################################################
+#
+# uint64_t __cmpxchg_64(uint64_t test, uint64_t new, uint64_t *v)
+#
+###############################################################################
+	.globl		__cmpxchg_64
+        .type		__cmpxchg_64,@function
+__cmpxchg_64:
+	or.p		gr8,gr8,gr4
+	or		gr9,gr9,gr5
+0:
+	orcc		gr0,gr0,gr0,icc3		/* set ICC3.Z */
+	ckeq		icc3,cc7
+	ldd.p		@(gr12,gr0),gr8			/* LDD.P/ORCR must be atomic */
+	orcr		cc7,cc7,cc3
+	subcc		gr8,gr4,gr0,icc0
+	subcc.p		gr9,gr5,gr0,icc1
+	bnelr		icc0,#0
+	bnelr		icc1,#0
+	cstd.p		gr10,@(gr12,gr0)	,cc3,#1
+	corcc		gr29,gr29,gr0		,cc3,#1	/* clear ICC3.Z if store happens */
+	beq		icc3,#0,0b
+	bralr
+
+	.size		__cmpxchg_64, .-__cmpxchg_64
+
-- 
cgit v1.2.3-59-g8ed1b


From 42ca4fb69126dd9d0e25112edca1cacf846aa5c3 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Thu, 2 Jul 2009 00:46:21 +0100
Subject: FRV: Add basic performance counter support

Add basic performance counter support to the FRV arch.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/frv/Kconfig                    |  1 +
 arch/frv/include/asm/perf_counter.h | 17 +++++++++++++++++
 arch/frv/lib/Makefile               |  2 +-
 arch/frv/lib/perf_counter.c         | 19 +++++++++++++++++++
 4 files changed, 38 insertions(+), 1 deletion(-)
 create mode 100644 arch/frv/include/asm/perf_counter.h
 create mode 100644 arch/frv/lib/perf_counter.c

diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig
index 8a5bd7a9c6f5..b86e19c9b5b0 100644
--- a/arch/frv/Kconfig
+++ b/arch/frv/Kconfig
@@ -7,6 +7,7 @@ config FRV
 	default y
 	select HAVE_IDE
 	select HAVE_ARCH_TRACEHOOK
+	select HAVE_PERF_COUNTERS
 
 config ZONE_DMA
 	bool
diff --git a/arch/frv/include/asm/perf_counter.h b/arch/frv/include/asm/perf_counter.h
new file mode 100644
index 000000000000..ccf726e61b2e
--- /dev/null
+++ b/arch/frv/include/asm/perf_counter.h
@@ -0,0 +1,17 @@
+/* FRV performance counter support
+ *
+ * Copyright (C) 2009 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public Licence
+ * as published by the Free Software Foundation; either version
+ * 2 of the Licence, or (at your option) any later version.
+ */
+
+#ifndef _ASM_PERF_COUNTER_H
+#define _ASM_PERF_COUNTER_H
+
+#define PERF_COUNTER_INDEX_OFFSET	0
+
+#endif /* _ASM_PERF_COUNTER_H */
diff --git a/arch/frv/lib/Makefile b/arch/frv/lib/Makefile
index 4ff2fb1e6b16..0a377210c89b 100644
--- a/arch/frv/lib/Makefile
+++ b/arch/frv/lib/Makefile
@@ -5,4 +5,4 @@
 lib-y := \
 	__ashldi3.o __lshrdi3.o __muldi3.o __ashrdi3.o __negdi2.o __ucmpdi2.o \
 	checksum.o memcpy.o memset.o atomic-ops.o atomic64-ops.o \
-	outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o
+	outsl_ns.o outsl_sw.o insl_ns.o insl_sw.o cache.o perf_counter.o
diff --git a/arch/frv/lib/perf_counter.c b/arch/frv/lib/perf_counter.c
new file mode 100644
index 000000000000..2000feecd571
--- /dev/null
+++ b/arch/frv/lib/perf_counter.c
@@ -0,0 +1,19 @@
+/* Performance counter handling
+ *
+ * Copyright (C) 2009 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public Licence
+ * as published by the Free Software Foundation; either version
+ * 2 of the Licence, or (at your option) any later version.
+ */
+
+#include <linux/perf_counter.h>
+
+/*
+ * mark the performance counter as pending
+ */
+void set_perf_counter_pending(void)
+{
+}
-- 
cgit v1.2.3-59-g8ed1b


From 75e6bf9638992dfc0fec9c3ca10444c8e0d6a638 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Thu, 2 Jul 2009 11:21:16 +0100
Subject: intel-iommu: Introduce first_pte_in_page() to simplify PTE-setting
 loops

On Wed, 2009-07-01 at 16:59 -0700, Linus Torvalds wrote:
> I also _really_ hate how you do
>
>         (unsigned long)pte >> VTD_PAGE_SHIFT ==
>         (unsigned long)first_pte >> VTD_PAGE_SHIFT

Kill this, in favour of just looking to see if the incremented pte
pointer has 'wrapped' onto the next page. Which means we have to check
it _after_ incrementing it, not before.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 2c1b2babfdc5..dcf0295a9b60 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -240,6 +240,11 @@ static inline bool dma_pte_present(struct dma_pte *pte)
 	return (pte->val & 3) != 0;
 }
 
+static inline int first_pte_in_page(struct dma_pte *pte)
+{
+	return !((unsigned long)pte & ~VTD_PAGE_MASK);
+}
+
 /*
  * This domain is a statically identity mapping domain.
  *	1. This domain creats a static 1:1 mapping to all usable memory.
@@ -780,13 +785,12 @@ static void dma_pte_clear_range(struct dmar_domain *domain,
 			start_pfn = align_to_level(start_pfn + 1, 2);
 			continue;
 		}
-		while (start_pfn <= last_pfn &&
-		       (unsigned long)pte >> VTD_PAGE_SHIFT ==
-		       (unsigned long)first_pte >> VTD_PAGE_SHIFT) {
+		do { 
 			dma_clear_pte(pte);
 			start_pfn++;
 			pte++;
-		}
+		} while (start_pfn <= last_pfn && !first_pte_in_page(pte));
+
 		domain_flush_cache(domain, first_pte,
 				   (void *)pte - (void *)first_pte);
 	}
@@ -821,14 +825,14 @@ static void dma_pte_free_pagetable(struct dmar_domain *domain,
 				tmp = align_to_level(tmp + 1, level + 1);
 				continue;
 			}
-			while (tmp + level_size(level) - 1 <= last_pfn &&
-			       (unsigned long)pte >> VTD_PAGE_SHIFT ==
-			       (unsigned long)first_pte >> VTD_PAGE_SHIFT) {
+			do {
 				free_pgtable_page(phys_to_virt(dma_pte_addr(pte)));
 				dma_clear_pte(pte);
 				pte++;
 				tmp += level_size(level);
-			}
+			} while (!first_pte_in_page(pte) &&
+				 tmp + level_size(level) - 1 <= last_pfn);
+
 			domain_flush_cache(domain, first_pte,
 					   (void *)pte - (void *)first_pte);
 			
@@ -1694,9 +1698,7 @@ static int __domain_mapping(struct dmar_domain *domain, unsigned long iov_pfn,
 			WARN_ON(1);
 		}
 		pte++;
-		if (!nr_pages ||
-		    (unsigned long)pte >> VTD_PAGE_SHIFT !=
-		    (unsigned long)first_pte >> VTD_PAGE_SHIFT) {
+		if (!nr_pages || first_pte_in_page(pte)) {
 			domain_flush_cache(domain, first_pte,
 					   (void *)pte - (void *)first_pte);
 			pte = NULL;
-- 
cgit v1.2.3-59-g8ed1b


From 6a43e574c5af7d9bd084992b1c9c3cdbc3b6c0e9 Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Thu, 2 Jul 2009 12:02:34 +0100
Subject: intel-iommu: Don't keep freeing page zero in dma_pte_free_pagetable()

Check dma_pte_present() and only free the page if there _is_ one.
Kind of surprising that there was no warning about this.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/pci/intel-iommu.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index dcf0295a9b60..53075424a434 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -826,8 +826,10 @@ static void dma_pte_free_pagetable(struct dmar_domain *domain,
 				continue;
 			}
 			do {
-				free_pgtable_page(phys_to_virt(dma_pte_addr(pte)));
-				dma_clear_pte(pte);
+				if (dma_pte_present(pte)) {
+					free_pgtable_page(phys_to_virt(dma_pte_addr(pte)));
+					dma_clear_pte(pte);
+				}
 				pte++;
 				tmp += level_size(level);
 			} while (!first_pte_in_page(pte) &&
-- 
cgit v1.2.3-59-g8ed1b


From bdae997f44535ac4ebe1e055ffe59eeee946f453 Mon Sep 17 00:00:00 2001
From: Keith Packard <keithp@keithp.com>
Date: Wed, 1 Jul 2009 21:56:38 -0700
Subject: fs/notify/inotify: decrement user inotify count on close

The per-user inotify_devs value is incremented each time a new file is
allocated, but never decremented. This led to inotify_init failing after a
limited number of calls.

Signed-off-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Eric Paris <eparis@redhat.com>
---
 fs/notify/inotify/inotify_user.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index ff231ad23895..ff27a2965844 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -296,12 +296,15 @@ static int inotify_fasync(int fd, struct file *file, int on)
 static int inotify_release(struct inode *ignored, struct file *file)
 {
 	struct fsnotify_group *group = file->private_data;
+	struct user_struct *user = group->inotify_data.user;
 
 	fsnotify_clear_marks_by_group(group);
 
 	/* free this group, matching get was inotify_init->fsnotify_obtain_group */
 	fsnotify_put_group(group);
 
+	atomic_dec(&user->inotify_devs);
+
 	return 0;
 }
 
-- 
cgit v1.2.3-59-g8ed1b


From f597bb19ccd034cbcf05e1194238e2c8d9505a8a Mon Sep 17 00:00:00 2001
From: Chris Mason <chris.mason@oracle.com>
Date: Sat, 27 Jun 2009 21:06:22 -0400
Subject: Btrfs: don't log the inode in file_write while growing the file

---
 fs/btrfs/file.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 126477eaecf5..7c3cd248d8d6 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -151,7 +151,10 @@ static noinline int dirty_and_release_pages(struct btrfs_trans_handle *trans,
 	}
 	if (end_pos > isize) {
 		i_size_write(inode, end_pos);
-		btrfs_update_inode(trans, root, inode);
+		/* we've only changed i_size in ram, and we haven't updated
+		 * the disk i_size.  There is no need to log the inode
+		 * at this time.
+		 */
 	}
 	err = btrfs_end_transaction(trans, root);
 out_unlock:
-- 
cgit v1.2.3-59-g8ed1b


From c8a894d77de4a1e0a544577fd4eabc9aacd453a8 Mon Sep 17 00:00:00 2001
From: Chris Mason <chris.mason@oracle.com>
Date: Sat, 27 Jun 2009 21:07:03 -0400
Subject: Btrfs: fix the file clone ioctl for preallocated extents

---
 fs/btrfs/ioctl.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index eff18f5b5362..9f4db848db10 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1028,7 +1028,8 @@ static long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 						struct btrfs_file_extent_item);
 			comp = btrfs_file_extent_compression(leaf, extent);
 			type = btrfs_file_extent_type(leaf, extent);
-			if (type == BTRFS_FILE_EXTENT_REG) {
+			if (type == BTRFS_FILE_EXTENT_REG ||
+			    type == BTRFS_FILE_EXTENT_PREALLOC) {
 				disko = btrfs_file_extent_disk_bytenr(leaf,
 								      extent);
 				diskl = btrfs_file_extent_disk_num_bytes(leaf,
@@ -1051,7 +1052,8 @@ static long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 			new_key.objectid = inode->i_ino;
 			new_key.offset = key.offset + destoff - off;
 
-			if (type == BTRFS_FILE_EXTENT_REG) {
+			if (type == BTRFS_FILE_EXTENT_REG ||
+			    type == BTRFS_FILE_EXTENT_PREALLOC) {
 				ret = btrfs_insert_empty_item(trans, root, path,
 							      &new_key, size);
 				if (ret)
-- 
cgit v1.2.3-59-g8ed1b


From a970b0a16cc416a509d5ae8b1d70978664e6f4fe Mon Sep 17 00:00:00 2001
From: Josef Bacik <jbacik@redhat.com>
Date: Sat, 27 Jun 2009 21:07:34 -0400
Subject: Btrfs: account for space we may use in fallocate

Using Eric Sandeen's xfstest for fallocate, you can easily trigger a ENOSPC
panic on btrfs.  This is because we do not account for data we may use when
doing the fallocate.  This patch fixes the problem by properly reserving space,
and then just freeing it when we are done.  The reservation stuff was made with
delalloc in mind, so its a little crude for this case, but it keeps the box
from panicing.

Signed-off-by: Josef Bacik <jbacik@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 5b68330f8585..1eacc78f6614 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5103,6 +5103,7 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 	u64 mask = BTRFS_I(inode)->root->sectorsize - 1;
 	struct extent_map *em;
 	struct btrfs_trans_handle *trans;
+	struct btrfs_root *root;
 	int ret;
 
 	alloc_start = offset & ~mask;
@@ -5121,6 +5122,13 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 			goto out;
 	}
 
+	root = BTRFS_I(inode)->root;
+
+	ret = btrfs_check_data_free_space(root, inode,
+					  alloc_end - alloc_start);
+	if (ret)
+		goto out;
+
 	locked_end = alloc_end - 1;
 	while (1) {
 		struct btrfs_ordered_extent *ordered;
@@ -5128,7 +5136,7 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 		trans = btrfs_start_transaction(BTRFS_I(inode)->root, 1);
 		if (!trans) {
 			ret = -EIO;
-			goto out;
+			goto out_free;
 		}
 
 		/* the extent lock is ordered inside the running
@@ -5189,6 +5197,8 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 		      GFP_NOFS);
 
 	btrfs_end_transaction(trans, BTRFS_I(inode)->root);
+out_free:
+	btrfs_free_reserved_data_space(root, inode, alloc_end - alloc_start);
 out:
 	mutex_unlock(&inode->i_mutex);
 	return ret;
-- 
cgit v1.2.3-59-g8ed1b


From 2c47e605a91dde6b0514f689645e7ab336c8592a Mon Sep 17 00:00:00 2001
From: Yan Zheng <zheng.yan@oracle.com>
Date: Sat, 27 Jun 2009 21:07:35 -0400
Subject: Btrfs: update backrefs while dropping snapshot

The new backref format has restriction on type of backref item.  If a tree
block isn't referenced by its owner tree, full backrefs must be used for the
pointers in it. When a tree block loses its owner tree's reference, backrefs
for the pointers in it should be updated to full backrefs. Current
btrfs_drop_snapshot misses the code that updates backrefs, so it's unsafe for
general use.

This patch adds backrefs update code to btrfs_drop_snapshot.  It isn't a
problem in the restricted form btrfs_drop_snapshot is used today, but for
general snapshot deletion this update is required.

Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/ctree.h       |   3 +-
 fs/btrfs/extent-tree.c | 564 ++++++++++++++++++++++++++++++++++---------------
 fs/btrfs/relocation.c  |   5 +-
 fs/btrfs/transaction.c |   4 +-
 4 files changed, 395 insertions(+), 181 deletions(-)

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 03441a99ea38..a404ecc53eb1 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -2076,8 +2076,7 @@ static inline int btrfs_insert_empty_item(struct btrfs_trans_handle *trans,
 int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path);
 int btrfs_prev_leaf(struct btrfs_root *root, struct btrfs_path *path);
 int btrfs_leaf_free_space(struct btrfs_root *root, struct extent_buffer *leaf);
-int btrfs_drop_snapshot(struct btrfs_trans_handle *trans, struct btrfs_root
-			*root);
+int btrfs_drop_snapshot(struct btrfs_root *root, int update_ref);
 int btrfs_drop_subtree(struct btrfs_trans_handle *trans,
 			struct btrfs_root *root,
 			struct extent_buffer *node,
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index edc7d208c5ce..cd64cfc14f7f 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -990,15 +990,13 @@ static inline int extent_ref_type(u64 parent, u64 owner)
 	return type;
 }
 
-static int find_next_key(struct btrfs_path *path, struct btrfs_key *key)
+static int find_next_key(struct btrfs_path *path, int level,
+			 struct btrfs_key *key)
 
 {
-	int level;
-	BUG_ON(!path->keep_locks);
-	for (level = 0; level < BTRFS_MAX_LEVEL; level++) {
+	for (; level < BTRFS_MAX_LEVEL; level++) {
 		if (!path->nodes[level])
 			break;
-		btrfs_assert_tree_locked(path->nodes[level]);
 		if (path->slots[level] + 1 >=
 		    btrfs_header_nritems(path->nodes[level]))
 			continue;
@@ -1158,7 +1156,8 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans,
 		 * For simplicity, we just do not add new inline back
 		 * ref if there is any kind of item for this block
 		 */
-		if (find_next_key(path, &key) == 0 && key.objectid == bytenr &&
+		if (find_next_key(path, 0, &key) == 0 &&
+		    key.objectid == bytenr &&
 		    key.type < BTRFS_BLOCK_GROUP_ITEM_KEY) {
 			err = -EAGAIN;
 			goto out;
@@ -4128,6 +4127,7 @@ struct extent_buffer *btrfs_alloc_free_block(struct btrfs_trans_handle *trans,
 	return buf;
 }
 
+#if 0
 int btrfs_drop_leaf_ref(struct btrfs_trans_handle *trans,
 			struct btrfs_root *root, struct extent_buffer *leaf)
 {
@@ -4171,8 +4171,6 @@ int btrfs_drop_leaf_ref(struct btrfs_trans_handle *trans,
 	return 0;
 }
 
-#if 0
-
 static noinline int cache_drop_leaf_ref(struct btrfs_trans_handle *trans,
 					struct btrfs_root *root,
 					struct btrfs_leaf_ref *ref)
@@ -4553,262 +4551,471 @@ out:
 }
 #endif
 
+struct walk_control {
+	u64 refs[BTRFS_MAX_LEVEL];
+	u64 flags[BTRFS_MAX_LEVEL];
+	struct btrfs_key update_progress;
+	int stage;
+	int level;
+	int shared_level;
+	int update_ref;
+	int keep_locks;
+};
+
+#define DROP_REFERENCE	1
+#define UPDATE_BACKREF	2
+
 /*
- * helper function for drop_subtree, this function is similar to
- * walk_down_tree. The main difference is that it checks reference
- * counts while tree blocks are locked.
+ * hepler to process tree block while walking down the tree.
+ *
+ * when wc->stage == DROP_REFERENCE, this function checks
+ * reference count of the block. if the block is shared and
+ * we need update back refs for the subtree rooted at the
+ * block, this function changes wc->stage to UPDATE_BACKREF
+ *
+ * when wc->stage == UPDATE_BACKREF, this function updates
+ * back refs for pointers in the block.
+ *
+ * NOTE: return value 1 means we should stop walking down.
  */
-static noinline int walk_down_tree(struct btrfs_trans_handle *trans,
+static noinline int walk_down_proc(struct btrfs_trans_handle *trans,
 				   struct btrfs_root *root,
-				   struct btrfs_path *path, int *level)
+				   struct btrfs_path *path,
+				   struct walk_control *wc)
 {
-	struct extent_buffer *next;
-	struct extent_buffer *cur;
-	struct extent_buffer *parent;
-	u64 bytenr;
-	u64 ptr_gen;
-	u64 refs;
-	u64 flags;
-	u32 blocksize;
+	int level = wc->level;
+	struct extent_buffer *eb = path->nodes[level];
+	struct btrfs_key key;
+	u64 flag = BTRFS_BLOCK_FLAG_FULL_BACKREF;
 	int ret;
 
-	cur = path->nodes[*level];
-	ret = btrfs_lookup_extent_info(trans, root, cur->start, cur->len,
-				       &refs, &flags);
-	BUG_ON(ret);
-	if (refs > 1)
-		goto out;
+	if (wc->stage == UPDATE_BACKREF &&
+	    btrfs_header_owner(eb) != root->root_key.objectid)
+		return 1;
 
-	BUG_ON(!(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF));
+	/*
+	 * when reference count of tree block is 1, it won't increase
+	 * again. once full backref flag is set, we never clear it.
+	 */
+	if ((wc->stage == DROP_REFERENCE && wc->refs[level] != 1) ||
+	    (wc->stage == UPDATE_BACKREF && !(wc->flags[level] & flag))) {
+		BUG_ON(!path->locks[level]);
+		ret = btrfs_lookup_extent_info(trans, root,
+					       eb->start, eb->len,
+					       &wc->refs[level],
+					       &wc->flags[level]);
+		BUG_ON(ret);
+		BUG_ON(wc->refs[level] == 0);
+	}
 
-	while (*level >= 0) {
-		cur = path->nodes[*level];
-		if (*level == 0) {
-			ret = btrfs_drop_leaf_ref(trans, root, cur);
-			BUG_ON(ret);
-			clean_tree_block(trans, root, cur);
-			break;
-		}
-		if (path->slots[*level] >= btrfs_header_nritems(cur)) {
-			clean_tree_block(trans, root, cur);
-			break;
+	if (wc->stage == DROP_REFERENCE &&
+	    wc->update_ref && wc->refs[level] > 1) {
+		BUG_ON(eb == root->node);
+		BUG_ON(path->slots[level] > 0);
+		if (level == 0)
+			btrfs_item_key_to_cpu(eb, &key, path->slots[level]);
+		else
+			btrfs_node_key_to_cpu(eb, &key, path->slots[level]);
+		if (btrfs_header_owner(eb) == root->root_key.objectid &&
+		    btrfs_comp_cpu_keys(&key, &wc->update_progress) >= 0) {
+			wc->stage = UPDATE_BACKREF;
+			wc->shared_level = level;
 		}
+	}
 
-		bytenr = btrfs_node_blockptr(cur, path->slots[*level]);
-		blocksize = btrfs_level_size(root, *level - 1);
-		ptr_gen = btrfs_node_ptr_generation(cur, path->slots[*level]);
+	if (wc->stage == DROP_REFERENCE) {
+		if (wc->refs[level] > 1)
+			return 1;
 
-		next = read_tree_block(root, bytenr, blocksize, ptr_gen);
-		btrfs_tree_lock(next);
-		btrfs_set_lock_blocking(next);
+		if (path->locks[level] && !wc->keep_locks) {
+			btrfs_tree_unlock(eb);
+			path->locks[level] = 0;
+		}
+		return 0;
+	}
 
-		ret = btrfs_lookup_extent_info(trans, root, bytenr, blocksize,
-					       &refs, &flags);
+	/* wc->stage == UPDATE_BACKREF */
+	if (!(wc->flags[level] & flag)) {
+		BUG_ON(!path->locks[level]);
+		ret = btrfs_inc_ref(trans, root, eb, 1);
 		BUG_ON(ret);
-		if (refs > 1) {
-			parent = path->nodes[*level];
-			ret = btrfs_free_extent(trans, root, bytenr,
-						blocksize, parent->start,
-						btrfs_header_owner(parent),
-						*level - 1, 0);
+		ret = btrfs_dec_ref(trans, root, eb, 0);
+		BUG_ON(ret);
+		ret = btrfs_set_disk_extent_flags(trans, root, eb->start,
+						  eb->len, flag, 0);
+		BUG_ON(ret);
+		wc->flags[level] |= flag;
+	}
+
+	/*
+	 * the block is shared by multiple trees, so it's not good to
+	 * keep the tree lock
+	 */
+	if (path->locks[level] && level > 0) {
+		btrfs_tree_unlock(eb);
+		path->locks[level] = 0;
+	}
+	return 0;
+}
+
+/*
+ * hepler to process tree block while walking up the tree.
+ *
+ * when wc->stage == DROP_REFERENCE, this function drops
+ * reference count on the block.
+ *
+ * when wc->stage == UPDATE_BACKREF, this function changes
+ * wc->stage back to DROP_REFERENCE if we changed wc->stage
+ * to UPDATE_BACKREF previously while processing the block.
+ *
+ * NOTE: return value 1 means we should stop walking up.
+ */
+static noinline int walk_up_proc(struct btrfs_trans_handle *trans,
+				 struct btrfs_root *root,
+				 struct btrfs_path *path,
+				 struct walk_control *wc)
+{
+	int ret = 0;
+	int level = wc->level;
+	struct extent_buffer *eb = path->nodes[level];
+	u64 parent = 0;
+
+	if (wc->stage == UPDATE_BACKREF) {
+		BUG_ON(wc->shared_level < level);
+		if (level < wc->shared_level)
+			goto out;
+
+		BUG_ON(wc->refs[level] <= 1);
+		ret = find_next_key(path, level + 1, &wc->update_progress);
+		if (ret > 0)
+			wc->update_ref = 0;
+
+		wc->stage = DROP_REFERENCE;
+		wc->shared_level = -1;
+		path->slots[level] = 0;
+
+		/*
+		 * check reference count again if the block isn't locked.
+		 * we should start walking down the tree again if reference
+		 * count is one.
+		 */
+		if (!path->locks[level]) {
+			BUG_ON(level == 0);
+			btrfs_tree_lock(eb);
+			btrfs_set_lock_blocking(eb);
+			path->locks[level] = 1;
+
+			ret = btrfs_lookup_extent_info(trans, root,
+						       eb->start, eb->len,
+						       &wc->refs[level],
+						       &wc->flags[level]);
 			BUG_ON(ret);
-			path->slots[*level]++;
-			btrfs_tree_unlock(next);
-			free_extent_buffer(next);
-			continue;
+			BUG_ON(wc->refs[level] == 0);
+			if (wc->refs[level] == 1) {
+				btrfs_tree_unlock(eb);
+				path->locks[level] = 0;
+				return 1;
+			}
+		} else {
+			BUG_ON(level != 0);
 		}
+	}
 
-		BUG_ON(!(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF));
+	/* wc->stage == DROP_REFERENCE */
+	BUG_ON(wc->refs[level] > 1 && !path->locks[level]);
 
-		*level = btrfs_header_level(next);
-		path->nodes[*level] = next;
-		path->slots[*level] = 0;
-		path->locks[*level] = 1;
-		cond_resched();
+	if (wc->refs[level] == 1) {
+		if (level == 0) {
+			if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF)
+				ret = btrfs_dec_ref(trans, root, eb, 1);
+			else
+				ret = btrfs_dec_ref(trans, root, eb, 0);
+			BUG_ON(ret);
+		}
+		/* make block locked assertion in clean_tree_block happy */
+		if (!path->locks[level] &&
+		    btrfs_header_generation(eb) == trans->transid) {
+			btrfs_tree_lock(eb);
+			btrfs_set_lock_blocking(eb);
+			path->locks[level] = 1;
+		}
+		clean_tree_block(trans, root, eb);
+	}
+
+	if (eb == root->node) {
+		if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF)
+			parent = eb->start;
+		else
+			BUG_ON(root->root_key.objectid !=
+			       btrfs_header_owner(eb));
+	} else {
+		if (wc->flags[level + 1] & BTRFS_BLOCK_FLAG_FULL_BACKREF)
+			parent = path->nodes[level + 1]->start;
+		else
+			BUG_ON(root->root_key.objectid !=
+			       btrfs_header_owner(path->nodes[level + 1]));
 	}
-out:
-	if (path->nodes[*level] == root->node)
-		parent = path->nodes[*level];
-	else
-		parent = path->nodes[*level + 1];
-	bytenr = path->nodes[*level]->start;
-	blocksize = path->nodes[*level]->len;
 
-	ret = btrfs_free_extent(trans, root, bytenr, blocksize, parent->start,
-				btrfs_header_owner(parent), *level, 0);
+	ret = btrfs_free_extent(trans, root, eb->start, eb->len, parent,
+				root->root_key.objectid, level, 0);
 	BUG_ON(ret);
+out:
+	wc->refs[level] = 0;
+	wc->flags[level] = 0;
+	return ret;
+}
+
+static noinline int walk_down_tree(struct btrfs_trans_handle *trans,
+				   struct btrfs_root *root,
+				   struct btrfs_path *path,
+				   struct walk_control *wc)
+{
+	struct extent_buffer *next;
+	struct extent_buffer *cur;
+	u64 bytenr;
+	u64 ptr_gen;
+	u32 blocksize;
+	int level = wc->level;
+	int ret;
+
+	while (level >= 0) {
+		cur = path->nodes[level];
+		BUG_ON(path->slots[level] >= btrfs_header_nritems(cur));
 
-	if (path->locks[*level]) {
-		btrfs_tree_unlock(path->nodes[*level]);
-		path->locks[*level] = 0;
+		ret = walk_down_proc(trans, root, path, wc);
+		if (ret > 0)
+			break;
+
+		if (level == 0)
+			break;
+
+		bytenr = btrfs_node_blockptr(cur, path->slots[level]);
+		blocksize = btrfs_level_size(root, level - 1);
+		ptr_gen = btrfs_node_ptr_generation(cur, path->slots[level]);
+
+		next = read_tree_block(root, bytenr, blocksize, ptr_gen);
+		btrfs_tree_lock(next);
+		btrfs_set_lock_blocking(next);
+
+		level--;
+		BUG_ON(level != btrfs_header_level(next));
+		path->nodes[level] = next;
+		path->slots[level] = 0;
+		path->locks[level] = 1;
+		wc->level = level;
 	}
-	free_extent_buffer(path->nodes[*level]);
-	path->nodes[*level] = NULL;
-	*level += 1;
-	cond_resched();
 	return 0;
 }
 
-/*
- * helper for dropping snapshots.  This walks back up the tree in the path
- * to find the first node higher up where we haven't yet gone through
- * all the slots
- */
 static noinline int walk_up_tree(struct btrfs_trans_handle *trans,
 				 struct btrfs_root *root,
 				 struct btrfs_path *path,
-				 int *level, int max_level)
+				 struct walk_control *wc, int max_level)
 {
-	struct btrfs_root_item *root_item = &root->root_item;
-	int i;
-	int slot;
+	int level = wc->level;
 	int ret;
 
-	for (i = *level; i < max_level && path->nodes[i]; i++) {
-		slot = path->slots[i];
-		if (slot + 1 < btrfs_header_nritems(path->nodes[i])) {
-			/*
-			 * there is more work to do in this level.
-			 * Update the drop_progress marker to reflect
-			 * the work we've done so far, and then bump
-			 * the slot number
-			 */
-			path->slots[i]++;
-			WARN_ON(*level == 0);
-			if (max_level == BTRFS_MAX_LEVEL) {
-				btrfs_node_key(path->nodes[i],
-					       &root_item->drop_progress,
-					       path->slots[i]);
-				root_item->drop_level = i;
-			}
-			*level = i;
+	path->slots[level] = btrfs_header_nritems(path->nodes[level]);
+	while (level < max_level && path->nodes[level]) {
+		wc->level = level;
+		if (path->slots[level] + 1 <
+		    btrfs_header_nritems(path->nodes[level])) {
+			path->slots[level]++;
 			return 0;
 		} else {
-			struct extent_buffer *parent;
-
-			/*
-			 * this whole node is done, free our reference
-			 * on it and go up one level
-			 */
-			if (path->nodes[*level] == root->node)
-				parent = path->nodes[*level];
-			else
-				parent = path->nodes[*level + 1];
+			ret = walk_up_proc(trans, root, path, wc);
+			if (ret > 0)
+				return 0;
 
-			clean_tree_block(trans, root, path->nodes[i]);
-			ret = btrfs_free_extent(trans, root,
-						path->nodes[i]->start,
-						path->nodes[i]->len,
-						parent->start,
-						btrfs_header_owner(parent),
-						*level, 0);
-			BUG_ON(ret);
-			if (path->locks[*level]) {
-				btrfs_tree_unlock(path->nodes[i]);
-				path->locks[i] = 0;
+			if (path->locks[level]) {
+				btrfs_tree_unlock(path->nodes[level]);
+				path->locks[level] = 0;
 			}
-			free_extent_buffer(path->nodes[i]);
-			path->nodes[i] = NULL;
-			*level = i + 1;
+			free_extent_buffer(path->nodes[level]);
+			path->nodes[level] = NULL;
+			level++;
 		}
 	}
 	return 1;
 }
 
 /*
- * drop the reference count on the tree rooted at 'snap'.  This traverses
- * the tree freeing any blocks that have a ref count of zero after being
- * decremented.
+ * drop a subvolume tree.
+ *
+ * this function traverses the tree freeing any blocks that only
+ * referenced by the tree.
+ *
+ * when a shared tree block is found. this function decreases its
+ * reference count by one. if update_ref is true, this function
+ * also make sure backrefs for the shared block and all lower level
+ * blocks are properly updated.
  */
-int btrfs_drop_snapshot(struct btrfs_trans_handle *trans, struct btrfs_root
-			*root)
+int btrfs_drop_snapshot(struct btrfs_root *root, int update_ref)
 {
-	int ret = 0;
-	int wret;
-	int level;
 	struct btrfs_path *path;
-	int update_count;
+	struct btrfs_trans_handle *trans;
+	struct btrfs_root *tree_root = root->fs_info->tree_root;
 	struct btrfs_root_item *root_item = &root->root_item;
+	struct walk_control *wc;
+	struct btrfs_key key;
+	int err = 0;
+	int ret;
+	int level;
 
 	path = btrfs_alloc_path();
 	BUG_ON(!path);
 
-	level = btrfs_header_level(root->node);
+	wc = kzalloc(sizeof(*wc), GFP_NOFS);
+	BUG_ON(!wc);
+
+	trans = btrfs_start_transaction(tree_root, 1);
+
 	if (btrfs_disk_key_objectid(&root_item->drop_progress) == 0) {
+		level = btrfs_header_level(root->node);
 		path->nodes[level] = btrfs_lock_root_node(root);
 		btrfs_set_lock_blocking(path->nodes[level]);
 		path->slots[level] = 0;
 		path->locks[level] = 1;
+		memset(&wc->update_progress, 0,
+		       sizeof(wc->update_progress));
 	} else {
-		struct btrfs_key key;
-		struct btrfs_disk_key found_key;
-		struct extent_buffer *node;
-
 		btrfs_disk_key_to_cpu(&key, &root_item->drop_progress);
+		memcpy(&wc->update_progress, &key,
+		       sizeof(wc->update_progress));
+
 		level = root_item->drop_level;
+		BUG_ON(level == 0);
 		path->lowest_level = level;
-		wret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
-		if (wret < 0) {
-			ret = wret;
+		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
+		path->lowest_level = 0;
+		if (ret < 0) {
+			err = ret;
 			goto out;
 		}
-		node = path->nodes[level];
-		btrfs_node_key(node, &found_key, path->slots[level]);
-		WARN_ON(memcmp(&found_key, &root_item->drop_progress,
-			       sizeof(found_key)));
+		btrfs_node_key_to_cpu(path->nodes[level], &key,
+				      path->slots[level]);
+		WARN_ON(memcmp(&key, &wc->update_progress, sizeof(key)));
+
 		/*
 		 * unlock our path, this is safe because only this
 		 * function is allowed to delete this snapshot
 		 */
 		btrfs_unlock_up_safe(path, 0);
+
+		level = btrfs_header_level(root->node);
+		while (1) {
+			btrfs_tree_lock(path->nodes[level]);
+			btrfs_set_lock_blocking(path->nodes[level]);
+
+			ret = btrfs_lookup_extent_info(trans, root,
+						path->nodes[level]->start,
+						path->nodes[level]->len,
+						&wc->refs[level],
+						&wc->flags[level]);
+			BUG_ON(ret);
+			BUG_ON(wc->refs[level] == 0);
+
+			if (level == root_item->drop_level)
+				break;
+
+			btrfs_tree_unlock(path->nodes[level]);
+			WARN_ON(wc->refs[level] != 1);
+			level--;
+		}
 	}
+
+	wc->level = level;
+	wc->shared_level = -1;
+	wc->stage = DROP_REFERENCE;
+	wc->update_ref = update_ref;
+	wc->keep_locks = 0;
+
 	while (1) {
-		unsigned long update;
-		wret = walk_down_tree(trans, root, path, &level);
-		if (wret > 0)
+		ret = walk_down_tree(trans, root, path, wc);
+		if (ret < 0) {
+			err = ret;
 			break;
-		if (wret < 0)
-			ret = wret;
+		}
 
-		wret = walk_up_tree(trans, root, path, &level,
-				    BTRFS_MAX_LEVEL);
-		if (wret > 0)
+		ret = walk_up_tree(trans, root, path, wc, BTRFS_MAX_LEVEL);
+		if (ret < 0) {
+			err = ret;
 			break;
-		if (wret < 0)
-			ret = wret;
-		if (trans->transaction->in_commit ||
-		    trans->transaction->delayed_refs.flushing) {
-			ret = -EAGAIN;
+		}
+
+		if (ret > 0) {
+			BUG_ON(wc->stage != DROP_REFERENCE);
 			break;
 		}
-		for (update_count = 0; update_count < 16; update_count++) {
+
+		if (wc->stage == DROP_REFERENCE) {
+			level = wc->level;
+			btrfs_node_key(path->nodes[level],
+				       &root_item->drop_progress,
+				       path->slots[level]);
+			root_item->drop_level = level;
+		}
+
+		BUG_ON(wc->level == 0);
+		if (trans->transaction->in_commit ||
+		    trans->transaction->delayed_refs.flushing) {
+			ret = btrfs_update_root(trans, tree_root,
+						&root->root_key,
+						root_item);
+			BUG_ON(ret);
+
+			btrfs_end_transaction(trans, tree_root);
+			trans = btrfs_start_transaction(tree_root, 1);
+		} else {
+			unsigned long update;
 			update = trans->delayed_ref_updates;
 			trans->delayed_ref_updates = 0;
 			if (update)
-				btrfs_run_delayed_refs(trans, root, update);
-			else
-				break;
+				btrfs_run_delayed_refs(trans, tree_root,
+						       update);
 		}
 	}
+	btrfs_release_path(root, path);
+	BUG_ON(err);
+
+	ret = btrfs_del_root(trans, tree_root, &root->root_key);
+	BUG_ON(ret);
+
+	free_extent_buffer(root->node);
+	free_extent_buffer(root->commit_root);
+	kfree(root);
 out:
+	btrfs_end_transaction(trans, tree_root);
+	kfree(wc);
 	btrfs_free_path(path);
-	return ret;
+	return err;
 }
 
+/*
+ * drop subtree rooted at tree block 'node'.
+ *
+ * NOTE: this function will unlock and release tree block 'node'
+ */
 int btrfs_drop_subtree(struct btrfs_trans_handle *trans,
 			struct btrfs_root *root,
 			struct extent_buffer *node,
 			struct extent_buffer *parent)
 {
 	struct btrfs_path *path;
+	struct walk_control *wc;
 	int level;
 	int parent_level;
 	int ret = 0;
 	int wret;
 
+	BUG_ON(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID);
+
 	path = btrfs_alloc_path();
 	BUG_ON(!path);
 
+	wc = kzalloc(sizeof(*wc), GFP_NOFS);
+	BUG_ON(!wc);
+
 	btrfs_assert_tree_locked(parent);
 	parent_level = btrfs_header_level(parent);
 	extent_buffer_get(parent);
@@ -4817,24 +5024,33 @@ int btrfs_drop_subtree(struct btrfs_trans_handle *trans,
 
 	btrfs_assert_tree_locked(node);
 	level = btrfs_header_level(node);
-	extent_buffer_get(node);
 	path->nodes[level] = node;
 	path->slots[level] = 0;
+	path->locks[level] = 1;
+
+	wc->refs[parent_level] = 1;
+	wc->flags[parent_level] = BTRFS_BLOCK_FLAG_FULL_BACKREF;
+	wc->level = level;
+	wc->shared_level = -1;
+	wc->stage = DROP_REFERENCE;
+	wc->update_ref = 0;
+	wc->keep_locks = 1;
 
 	while (1) {
-		wret = walk_down_tree(trans, root, path, &level);
-		if (wret < 0)
+		wret = walk_down_tree(trans, root, path, wc);
+		if (wret < 0) {
 			ret = wret;
-		if (wret != 0)
 			break;
+		}
 
-		wret = walk_up_tree(trans, root, path, &level, parent_level);
+		wret = walk_up_tree(trans, root, path, wc, parent_level);
 		if (wret < 0)
 			ret = wret;
 		if (wret != 0)
 			break;
 	}
 
+	kfree(wc);
 	btrfs_free_path(path);
 	return ret;
 }
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index b23dc209ae10..008397934778 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -1788,7 +1788,7 @@ static void merge_func(struct btrfs_work *work)
 		btrfs_end_transaction(trans, root);
 	}
 
-	btrfs_drop_dead_root(reloc_root);
+	btrfs_drop_snapshot(reloc_root, 0);
 
 	if (atomic_dec_and_test(async->num_pending))
 		complete(async->done);
@@ -2075,9 +2075,6 @@ static int do_relocation(struct btrfs_trans_handle *trans,
 
 			ret = btrfs_drop_subtree(trans, root, eb, upper->eb);
 			BUG_ON(ret);
-
-			btrfs_tree_unlock(eb);
-			free_extent_buffer(eb);
 		}
 		if (!lowest) {
 			btrfs_tree_unlock(upper->eb);
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 4e83457ea253..2dbf1c1f56ee 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -593,6 +593,7 @@ int btrfs_defrag_root(struct btrfs_root *root, int cacheonly)
 	return 0;
 }
 
+#if 0
 /*
  * when dropping snapshots, we generate a ton of delayed refs, and it makes
  * sense not to join the transaction while it is trying to flush the current
@@ -681,6 +682,7 @@ int btrfs_drop_dead_root(struct btrfs_root *root)
 	btrfs_btree_balance_dirty(tree_root, nr);
 	return ret;
 }
+#endif
 
 /*
  * new snapshots need to be created at a very specific time in the
@@ -1081,7 +1083,7 @@ int btrfs_clean_old_snapshots(struct btrfs_root *root)
 	while (!list_empty(&list)) {
 		root = list_entry(list.next, struct btrfs_root, root_list);
 		list_del_init(&root->root_list);
-		btrfs_drop_dead_root(root);
+		btrfs_drop_snapshot(root, 0);
 	}
 	return 0;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 9427216476d4fa75103f39d4b228c47d56ba20da Mon Sep 17 00:00:00 2001
From: Chris Mason <chris.mason@oracle.com>
Date: Thu, 2 Jul 2009 12:26:06 -0400
Subject: Btrfs: honor nodatacow/sum mount options for new files

The btrfs attr patches unconditionally inherited the inode flags field
without honoring nodatacow and nodatasum.  This fix makes sure
we properly record the nodatacow/sum mount options in new inodes.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/inode.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 1eacc78f6614..a48c084f6d3a 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -3587,12 +3587,6 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 		owner = 1;
 	BTRFS_I(inode)->block_group =
 			btrfs_find_block_group(root, 0, alloc_hint, owner);
-	if ((mode & S_IFREG)) {
-		if (btrfs_test_opt(root, NODATASUM))
-			BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM;
-		if (btrfs_test_opt(root, NODATACOW))
-			BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW;
-	}
 
 	key[0].objectid = objectid;
 	btrfs_set_key_type(&key[0], BTRFS_INODE_ITEM_KEY);
@@ -3647,6 +3641,13 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 
 	btrfs_inherit_iflags(inode, dir);
 
+	if ((mode & S_IFREG)) {
+		if (btrfs_test_opt(root, NODATASUM))
+			BTRFS_I(inode)->flags |= BTRFS_INODE_NODATASUM;
+		if (btrfs_test_opt(root, NODATACOW))
+			BTRFS_I(inode)->flags |= BTRFS_INODE_NODATACOW;
+	}
+
 	insert_inode_hash(inode);
 	inode_tree_add(inode);
 	return inode;
-- 
cgit v1.2.3-59-g8ed1b


From 9b627e9bf49ebfeb060dfae0435bdba06cf27cb8 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Thu, 2 Jul 2009 13:50:58 -0400
Subject: Btrfs: fix use after free in btrfs_start_workers fail path

worker memory is already freed on one fail path in btrfs_start_workers,
but is still dereferenced. Switch the dereference and kfree.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/async-thread.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 7f88628a1a72..6e4f6c50a120 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -299,8 +299,8 @@ int btrfs_start_workers(struct btrfs_workers *workers, int num_workers)
 					   "btrfs-%s-%d", workers->name,
 					   workers->num_workers + i);
 		if (IS_ERR(worker->task)) {
-			kfree(worker);
 			ret = PTR_ERR(worker->task);
+			kfree(worker);
 			goto fail;
 		}
 
-- 
cgit v1.2.3-59-g8ed1b


From 68f5a38c3ea4ae9cc7a40f86ff6d6d031583d93a Mon Sep 17 00:00:00 2001
From: Hu Tao <cnhutao@gmail.com>
Date: Thu, 2 Jul 2009 13:55:45 -0400
Subject: Btrfs: fix error message formatting

Make an error msg look nicer by inserting a space between number and word.

Signed-off-by: Hu Tao <hu.taoo@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
---
 fs/btrfs/extent-tree.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index cd64cfc14f7f..a5aca3997d42 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -2696,7 +2696,7 @@ again:
 
 		printk(KERN_ERR "no space left, need %llu, %llu delalloc bytes"
 		       ", %llu bytes_used, %llu bytes_reserved, "
-		       "%llu bytes_pinned, %llu bytes_readonly, %llu may use"
+		       "%llu bytes_pinned, %llu bytes_readonly, %llu may use "
 		       "%llu total\n", (unsigned long long)bytes,
 		       (unsigned long long)data_sinfo->bytes_delalloc,
 		       (unsigned long long)data_sinfo->bytes_used,
-- 
cgit v1.2.3-59-g8ed1b


From 43644679a1e80f53e6e0155ab75b1093ba3c0365 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Thu, 2 Jul 2009 12:05:10 -0700
Subject: x86: fix power-of-2 round_up/round_down macros

These macros had two bugs:
 - the type of the mask was not correctly expanded to the full size of
   the argument being expanded, resulting in possible loss of high bits
   when mixing types.
 - the alignment argument was evaluated twice, despite the macro looking
   like a fancy function (but it really does need to be a macro, since
   it works on arbitrary integer types)

Noticed by Peter Anvin, and with a fix that is a modification of his
suggestion (bug noticed by Yinghai Lu).

Cc: Peter Anvin <hpa@zytor.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/include/asm/proto.h | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/arch/x86/include/asm/proto.h b/arch/x86/include/asm/proto.h
index 49fb3ecf3bb3..621f56d73121 100644
--- a/arch/x86/include/asm/proto.h
+++ b/arch/x86/include/asm/proto.h
@@ -22,7 +22,14 @@ extern int reboot_force;
 
 long do_arch_prctl(struct task_struct *task, int code, unsigned long addr);
 
-#define round_up(x, y) (((x) + (y) - 1) & ~((y) - 1))
-#define round_down(x, y) ((x) & ~((y) - 1))
+/*
+ * This looks more complex than it should be. But we need to
+ * get the type for the ~ right in round_down (it needs to be
+ * as wide as the result!), and we want to evaluate the macro
+ * arguments just once each.
+ */
+#define __round_mask(x,y) ((__typeof__(x))((y)-1))
+#define round_up(x,y) ((((x)-1) | __round_mask(x,y))+1)
+#define round_down(x,y) ((x) & ~__round_mask(x,y))
 
 #endif /* _ASM_X86_PROTO_H */
-- 
cgit v1.2.3-59-g8ed1b


From 7c5371c403abb29f01bc6cff6c5096abdf2dc524 Mon Sep 17 00:00:00 2001
From: Yinghai Lu <yinghai@kernel.org>
Date: Wed, 1 Jul 2009 12:32:18 -0700
Subject: x86: add boundary check for 32bit res before expand e820 resource to
 alignment

fix hang with HIGHMEM_64G and 32bit resource.  According to hpa and
Linus, use (resource_size_t)-1 to fend off big ranges.

Analyzed by hpa

Reported-and-tested-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/kernel/e820.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 7271fa33d791..c4ca89d9aaf4 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -1383,6 +1383,8 @@ static unsigned long ram_alignment(resource_size_t pos)
 	return 32*1024*1024;
 }
 
+#define MAX_RESOURCE_SIZE ((resource_size_t)-1)
+
 void __init e820_reserve_resources_late(void)
 {
 	int i;
@@ -1400,17 +1402,19 @@ void __init e820_reserve_resources_late(void)
 	 * avoid stolen RAM:
 	 */
 	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *entry = &e820_saved.map[i];
-		resource_size_t start, end;
+		struct e820entry *entry = &e820.map[i];
+		u64 start, end;
 
 		if (entry->type != E820_RAM)
 			continue;
 		start = entry->addr + entry->size;
-		end = round_up(start, ram_alignment(start));
-		if (start == end)
+		end = round_up(start, ram_alignment(start)) - 1;
+		if (end > MAX_RESOURCE_SIZE)
+			end = MAX_RESOURCE_SIZE;
+		if (start >= end)
 			continue;
-		reserve_region_with_split(&iomem_resource, start,
-						  end - 1, "RAM buffer");
+		reserve_region_with_split(&iomem_resource, start, end,
+					  "RAM buffer");
 	}
 }
 
-- 
cgit v1.2.3-59-g8ed1b