aboutsummaryrefslogtreecommitdiffstats
path: root/fs/ceph
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ceph')
-rw-r--r--fs/ceph/addr.c6
-rw-r--r--fs/ceph/caps.c17
-rw-r--r--fs/ceph/dir.c23
-rw-r--r--fs/ceph/file.c65
-rw-r--r--fs/ceph/inode.c50
-rw-r--r--fs/ceph/ioctl.h2
-rw-r--r--fs/ceph/locks.c94
-rw-r--r--fs/ceph/mds_client.c49
-rw-r--r--fs/ceph/mds_client.h33
-rw-r--r--fs/ceph/super.h4
10 files changed, 221 insertions, 122 deletions
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index e9c874abc9e1..561438b6a50c 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -204,7 +204,7 @@ static int readpage_nounlock(struct file *filp, struct page *page)
err = ceph_osdc_readpages(osdc, ceph_vino(inode), &ci->i_layout,
page->index << PAGE_CACHE_SHIFT, &len,
ci->i_truncate_seq, ci->i_truncate_size,
- &page, 1);
+ &page, 1, 0);
if (err == -ENOENT)
err = 0;
if (err < 0) {
@@ -287,7 +287,7 @@ static int ceph_readpages(struct file *file, struct address_space *mapping,
rc = ceph_osdc_readpages(osdc, ceph_vino(inode), &ci->i_layout,
offset, &len,
ci->i_truncate_seq, ci->i_truncate_size,
- pages, nr_pages);
+ pages, nr_pages, 0);
if (rc == -ENOENT)
rc = 0;
if (rc < 0)
@@ -774,7 +774,7 @@ get_more_pages:
snapc, do_sync,
ci->i_truncate_seq,
ci->i_truncate_size,
- &inode->i_mtime, true, 1);
+ &inode->i_mtime, true, 1, 0);
max_pages = req->r_num_pages;
alloc_page_vec(fsc, req);
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 98ab13e2b71d..60d27bc9eb83 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -1430,8 +1430,8 @@ static int try_nonblocking_invalidate(struct inode *inode)
invalidating_gen == ci->i_rdcache_gen) {
/* success. */
dout("try_nonblocking_invalidate %p success\n", inode);
- ci->i_rdcache_gen = 0;
- ci->i_rdcache_revoking = 0;
+ /* save any racing async invalidate some trouble */
+ ci->i_rdcache_revoking = ci->i_rdcache_gen - 1;
return 0;
}
dout("try_nonblocking_invalidate %p failed\n", inode);
@@ -2273,8 +2273,7 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
{
struct ceph_inode_info *ci = ceph_inode(inode);
int mds = session->s_mds;
- unsigned seq = le32_to_cpu(grant->seq);
- unsigned issue_seq = le32_to_cpu(grant->issue_seq);
+ int seq = le32_to_cpu(grant->seq);
int newcaps = le32_to_cpu(grant->caps);
int issued, implemented, used, wanted, dirty;
u64 size = le64_to_cpu(grant->size);
@@ -2286,8 +2285,8 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
int revoked_rdcache = 0;
int queue_invalidate = 0;
- dout("handle_cap_grant inode %p cap %p mds%d seq %u/%u %s\n",
- inode, cap, mds, seq, issue_seq, ceph_cap_string(newcaps));
+ dout("handle_cap_grant inode %p cap %p mds%d seq %d %s\n",
+ inode, cap, mds, seq, ceph_cap_string(newcaps));
dout(" size %llu max_size %llu, i_size %llu\n", size, max_size,
inode->i_size);
@@ -2383,7 +2382,6 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
}
cap->seq = seq;
- cap->issue_seq = issue_seq;
/* file layout may have changed */
ci->i_layout = grant->layout;
@@ -2691,6 +2689,11 @@ static void handle_cap_import(struct ceph_mds_client *mdsc,
NULL /* no caps context */);
try_flush_caps(inode, session, NULL);
up_read(&mdsc->snap_rwsem);
+
+ /* make sure we re-request max_size, if necessary */
+ spin_lock(&inode->i_lock);
+ ci->i_requested_max_size = 0;
+ spin_unlock(&inode->i_lock);
}
/*
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index e0a2dc6fcafc..d902948a90d8 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -40,7 +40,8 @@ int ceph_init_dentry(struct dentry *dentry)
if (dentry->d_fsdata)
return 0;
- if (ceph_snap(dentry->d_parent->d_inode) == CEPH_NOSNAP)
+ if (dentry->d_parent == NULL || /* nfs fh_to_dentry */
+ ceph_snap(dentry->d_parent->d_inode) == CEPH_NOSNAP)
dentry->d_op = &ceph_dentry_ops;
else if (ceph_snap(dentry->d_parent->d_inode) == CEPH_SNAPDIR)
dentry->d_op = &ceph_snapdir_dentry_ops;
@@ -114,8 +115,8 @@ static int __dcache_readdir(struct file *filp,
spin_lock(&dcache_lock);
/* start at beginning? */
- if (filp->f_pos == 2 || (last &&
- filp->f_pos < ceph_dentry(last)->offset)) {
+ if (filp->f_pos == 2 || last == NULL ||
+ filp->f_pos < ceph_dentry(last)->offset) {
if (list_empty(&parent->d_subdirs))
goto out_unlock;
p = parent->d_subdirs.prev;
@@ -336,7 +337,10 @@ more:
if (req->r_reply_info.dir_end) {
kfree(fi->last_name);
fi->last_name = NULL;
- fi->next_offset = 2;
+ if (ceph_frag_is_rightmost(frag))
+ fi->next_offset = 2;
+ else
+ fi->next_offset = 0;
} else {
rinfo = &req->r_reply_info;
err = note_last_dentry(fi,
@@ -355,18 +359,22 @@ more:
u64 pos = ceph_make_fpos(frag, off);
struct ceph_mds_reply_inode *in =
rinfo->dir_in[off - fi->offset].in;
+ struct ceph_vino vino;
+ ino_t ino;
+
dout("readdir off %d (%d/%d) -> %lld '%.*s' %p\n",
off, off - fi->offset, rinfo->dir_nr, pos,
rinfo->dir_dname_len[off - fi->offset],
rinfo->dir_dname[off - fi->offset], in);
BUG_ON(!in);
ftype = le32_to_cpu(in->mode) >> 12;
+ vino.ino = le64_to_cpu(in->ino);
+ vino.snap = le64_to_cpu(in->snapid);
+ ino = ceph_vino_to_ino(vino);
if (filldir(dirent,
rinfo->dir_dname[off - fi->offset],
rinfo->dir_dname_len[off - fi->offset],
- pos,
- le64_to_cpu(in->ino),
- ftype) < 0) {
+ pos, ino, ftype) < 0) {
dout("filldir stopping us...\n");
return 0;
}
@@ -414,6 +422,7 @@ static void reset_readdir(struct ceph_file_info *fi)
fi->last_readdir = NULL;
}
kfree(fi->last_name);
+ fi->last_name = NULL;
fi->next_offset = 2; /* compensate for . and .. */
if (fi->dentry) {
dput(fi->dentry);
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index e77c28cf3690..7d0e4a82d898 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -154,11 +154,13 @@ int ceph_open(struct inode *inode, struct file *file)
}
/*
- * No need to block if we have any caps. Update wanted set
+ * No need to block if we have caps on the auth MDS (for
+ * write) or any MDS (for read). Update wanted set
* asynchronously.
*/
spin_lock(&inode->i_lock);
- if (__ceph_is_any_real_caps(ci)) {
+ if (__ceph_is_any_real_caps(ci) &&
+ (((fmode & CEPH_FILE_MODE_WR) == 0) || ci->i_auth_cap)) {
int mds_wanted = __ceph_caps_mds_wanted(ci);
int issued = __ceph_caps_issued(ci, NULL);
@@ -280,11 +282,13 @@ int ceph_release(struct inode *inode, struct file *file)
static int striped_read(struct inode *inode,
u64 off, u64 len,
struct page **pages, int num_pages,
- int *checkeof)
+ int *checkeof, bool align_to_pages,
+ unsigned long buf_align)
{
struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
struct ceph_inode_info *ci = ceph_inode(inode);
u64 pos, this_len;
+ int io_align, page_align;
int page_off = off & ~PAGE_CACHE_MASK; /* first byte's offset in page */
int left, pages_left;
int read;
@@ -300,14 +304,19 @@ static int striped_read(struct inode *inode,
page_pos = pages;
pages_left = num_pages;
read = 0;
+ io_align = off & ~PAGE_MASK;
more:
+ if (align_to_pages)
+ page_align = (pos - io_align + buf_align) & ~PAGE_MASK;
+ else
+ page_align = pos & ~PAGE_MASK;
this_len = left;
ret = ceph_osdc_readpages(&fsc->client->osdc, ceph_vino(inode),
&ci->i_layout, pos, &this_len,
ci->i_truncate_seq,
ci->i_truncate_size,
- page_pos, pages_left);
+ page_pos, pages_left, page_align);
hit_stripe = this_len < left;
was_short = ret >= 0 && ret < this_len;
if (ret == -ENOENT)
@@ -368,32 +377,34 @@ static ssize_t ceph_sync_read(struct file *file, char __user *data,
struct inode *inode = file->f_dentry->d_inode;
struct page **pages;
u64 off = *poff;
- int num_pages = calc_pages_for(off, len);
- int ret;
+ int num_pages, ret;
dout("sync_read on file %p %llu~%u %s\n", file, off, len,
(file->f_flags & O_DIRECT) ? "O_DIRECT" : "");
if (file->f_flags & O_DIRECT) {
- pages = ceph_get_direct_page_vector(data, num_pages, off, len);
-
- /*
- * flush any page cache pages in this range. this
- * will make concurrent normal and O_DIRECT io slow,
- * but it will at least behave sensibly when they are
- * in sequence.
- */
+ num_pages = calc_pages_for((unsigned long)data, len);
+ pages = ceph_get_direct_page_vector(data, num_pages, true);
} else {
+ num_pages = calc_pages_for(off, len);
pages = ceph_alloc_page_vector(num_pages, GFP_NOFS);
}
if (IS_ERR(pages))
return PTR_ERR(pages);
+ /*
+ * flush any page cache pages in this range. this
+ * will make concurrent normal and sync io slow,
+ * but it will at least behave sensibly when they are
+ * in sequence.
+ */
ret = filemap_write_and_wait(inode->i_mapping);
if (ret < 0)
goto done;
- ret = striped_read(inode, off, len, pages, num_pages, checkeof);
+ ret = striped_read(inode, off, len, pages, num_pages, checkeof,
+ file->f_flags & O_DIRECT,
+ (unsigned long)data & ~PAGE_MASK);
if (ret >= 0 && (file->f_flags & O_DIRECT) == 0)
ret = ceph_copy_page_vector_to_user(pages, data, off, ret);
@@ -402,7 +413,7 @@ static ssize_t ceph_sync_read(struct file *file, char __user *data,
done:
if (file->f_flags & O_DIRECT)
- ceph_put_page_vector(pages, num_pages);
+ ceph_put_page_vector(pages, num_pages, true);
else
ceph_release_page_vector(pages, num_pages);
dout("sync_read result %d\n", ret);
@@ -448,6 +459,8 @@ static ssize_t ceph_sync_write(struct file *file, const char __user *data,
int flags;
int do_sync = 0;
int check_caps = 0;
+ int page_align, io_align;
+ unsigned long buf_align;
int ret;
struct timespec mtime = CURRENT_TIME;
@@ -462,6 +475,9 @@ static ssize_t ceph_sync_write(struct file *file, const char __user *data,
else
pos = *offset;
+ io_align = pos & ~PAGE_MASK;
+ buf_align = (unsigned long)data & ~PAGE_MASK;
+
ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + left);
if (ret < 0)
return ret;
@@ -486,20 +502,27 @@ static ssize_t ceph_sync_write(struct file *file, const char __user *data,
*/
more:
len = left;
+ if (file->f_flags & O_DIRECT) {
+ /* write from beginning of first page, regardless of
+ io alignment */
+ page_align = (pos - io_align + buf_align) & ~PAGE_MASK;
+ num_pages = calc_pages_for((unsigned long)data, len);
+ } else {
+ page_align = pos & ~PAGE_MASK;
+ num_pages = calc_pages_for(pos, len);
+ }
req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout,
ceph_vino(inode), pos, &len,
CEPH_OSD_OP_WRITE, flags,
ci->i_snap_realm->cached_context,
do_sync,
ci->i_truncate_seq, ci->i_truncate_size,
- &mtime, false, 2);
+ &mtime, false, 2, page_align);
if (!req)
return -ENOMEM;
- num_pages = calc_pages_for(pos, len);
-
if (file->f_flags & O_DIRECT) {
- pages = ceph_get_direct_page_vector(data, num_pages, pos, len);
+ pages = ceph_get_direct_page_vector(data, num_pages, false);
if (IS_ERR(pages)) {
ret = PTR_ERR(pages);
goto out;
@@ -549,7 +572,7 @@ more:
}
if (file->f_flags & O_DIRECT)
- ceph_put_page_vector(pages, num_pages);
+ ceph_put_page_vector(pages, num_pages, false);
else if (file->f_flags & O_SYNC)
ceph_release_page_vector(pages, num_pages);
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 1d6a45b5a04c..bf1286588f26 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -2,7 +2,6 @@
#include <linux/module.h>
#include <linux/fs.h>
-#include <linux/smp_lock.h>
#include <linux/slab.h>
#include <linux/string.h>
#include <linux/uaccess.h>
@@ -471,7 +470,9 @@ void ceph_fill_file_time(struct inode *inode, int issued,
if (issued & (CEPH_CAP_FILE_EXCL|
CEPH_CAP_FILE_WR|
- CEPH_CAP_FILE_BUFFER)) {
+ CEPH_CAP_FILE_BUFFER|
+ CEPH_CAP_AUTH_EXCL|
+ CEPH_CAP_XATTR_EXCL)) {
if (timespec_compare(ctime, &inode->i_ctime) > 0) {
dout("ctime %ld.%09ld -> %ld.%09ld inc w/ cap\n",
inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec,
@@ -511,7 +512,7 @@ void ceph_fill_file_time(struct inode *inode, int issued,
warn = 1;
}
} else {
- /* we have no write caps; whatever the MDS says is true */
+ /* we have no write|excl caps; whatever the MDS says is true */
if (ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) >= 0) {
inode->i_ctime = *ctime;
inode->i_mtime = *mtime;
@@ -567,12 +568,17 @@ static int fill_inode(struct inode *inode,
/*
* provided version will be odd if inode value is projected,
- * even if stable. skip the update if we have a newer info
- * (e.g., due to inode info racing form multiple MDSs), or if
- * we are getting projected (unstable) inode info.
+ * even if stable. skip the update if we have newer stable
+ * info (ours>=theirs, e.g. due to racing mds replies), unless
+ * we are getting projected (unstable) info (in which case the
+ * version is odd, and we want ours>theirs).
+ * us them
+ * 2 2 skip
+ * 3 2 skip
+ * 3 3 update
*/
if (le64_to_cpu(info->version) > 0 &&
- (ci->i_version & ~1) > le64_to_cpu(info->version))
+ (ci->i_version & ~1) >= le64_to_cpu(info->version))
goto no_change;
issued = __ceph_caps_issued(ci, &implemented);
@@ -606,7 +612,14 @@ static int fill_inode(struct inode *inode,
le32_to_cpu(info->time_warp_seq),
&ctime, &mtime, &atime);
- ci->i_max_size = le64_to_cpu(info->max_size);
+ /* only update max_size on auth cap */
+ if ((info->cap.flags & CEPH_CAP_FLAG_AUTH) &&
+ ci->i_max_size != le64_to_cpu(info->max_size)) {
+ dout("max_size %lld -> %llu\n", ci->i_max_size,
+ le64_to_cpu(info->max_size));
+ ci->i_max_size = le64_to_cpu(info->max_size);
+ }
+
ci->i_layout = info->layout;
inode->i_blkbits = fls(le32_to_cpu(info->layout.fl_stripe_unit)) - 1;
@@ -1055,7 +1068,8 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
ininfo = rinfo->targeti.in;
vino.ino = le64_to_cpu(ininfo->ino);
vino.snap = le64_to_cpu(ininfo->snapid);
- if (!dn->d_inode) {
+ in = dn->d_inode;
+ if (!in) {
in = ceph_get_inode(sb, vino);
if (IS_ERR(in)) {
pr_err("fill_trace bad get_inode "
@@ -1386,11 +1400,8 @@ static void ceph_invalidate_work(struct work_struct *work)
spin_lock(&inode->i_lock);
dout("invalidate_pages %p gen %d revoking %d\n", inode,
ci->i_rdcache_gen, ci->i_rdcache_revoking);
- if (ci->i_rdcache_gen == 0 ||
- ci->i_rdcache_revoking != ci->i_rdcache_gen) {
- BUG_ON(ci->i_rdcache_revoking > ci->i_rdcache_gen);
+ if (ci->i_rdcache_revoking != ci->i_rdcache_gen) {
/* nevermind! */
- ci->i_rdcache_revoking = 0;
spin_unlock(&inode->i_lock);
goto out;
}
@@ -1400,15 +1411,16 @@ static void ceph_invalidate_work(struct work_struct *work)
ceph_invalidate_nondirty_pages(inode->i_mapping);
spin_lock(&inode->i_lock);
- if (orig_gen == ci->i_rdcache_gen) {
+ if (orig_gen == ci->i_rdcache_gen &&
+ orig_gen == ci->i_rdcache_revoking) {
dout("invalidate_pages %p gen %d successful\n", inode,
ci->i_rdcache_gen);
- ci->i_rdcache_gen = 0;
- ci->i_rdcache_revoking = 0;
+ ci->i_rdcache_revoking--;
check = 1;
} else {
- dout("invalidate_pages %p gen %d raced, gen now %d\n",
- inode, orig_gen, ci->i_rdcache_gen);
+ dout("invalidate_pages %p gen %d raced, now %d revoking %d\n",
+ inode, orig_gen, ci->i_rdcache_gen,
+ ci->i_rdcache_revoking);
}
spin_unlock(&inode->i_lock);
@@ -1739,7 +1751,7 @@ int ceph_do_getattr(struct inode *inode, int mask)
return 0;
}
- dout("do_getattr inode %p mask %s\n", inode, ceph_cap_string(mask));
+ dout("do_getattr inode %p mask %s mode 0%o\n", inode, ceph_cap_string(mask), inode->i_mode);
if (ceph_caps_issued_mask(ceph_inode(inode), mask, 1))
return 0;
diff --git a/fs/ceph/ioctl.h b/fs/ceph/ioctl.h
index a6ce54e94eb5..52e8fd74d450 100644
--- a/fs/ceph/ioctl.h
+++ b/fs/ceph/ioctl.h
@@ -4,7 +4,7 @@
#include <linux/ioctl.h>
#include <linux/types.h>
-#define CEPH_IOCTL_MAGIC 0x98
+#define CEPH_IOCTL_MAGIC 0x97
/* just use u64 to align sanely on all archs */
struct ceph_ioctl_layout {
diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c
index 40abde93c345..476b329867d4 100644
--- a/fs/ceph/locks.c
+++ b/fs/ceph/locks.c
@@ -11,40 +11,68 @@
* Implement fcntl and flock locking functions.
*/
static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file,
- u64 pid, u64 pid_ns,
- int cmd, u64 start, u64 length, u8 wait)
+ int cmd, u8 wait, struct file_lock *fl)
{
struct inode *inode = file->f_dentry->d_inode;
struct ceph_mds_client *mdsc =
ceph_sb_to_client(inode->i_sb)->mdsc;
struct ceph_mds_request *req;
int err;
+ u64 length = 0;
req = ceph_mdsc_create_request(mdsc, operation, USE_AUTH_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
req->r_inode = igrab(inode);
+ /* mds requires start and length rather than start and end */
+ if (LLONG_MAX == fl->fl_end)
+ length = 0;
+ else
+ length = fl->fl_end - fl->fl_start + 1;
+
dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, "
"length: %llu, wait: %d, type`: %d", (int)lock_type,
- (int)operation, pid, start, length, wait, cmd);
+ (int)operation, (u64)fl->fl_pid, fl->fl_start,
+ length, wait, fl->fl_type);
+
req->r_args.filelock_change.rule = lock_type;
req->r_args.filelock_change.type = cmd;
- req->r_args.filelock_change.pid = cpu_to_le64(pid);
+ req->r_args.filelock_change.pid = cpu_to_le64((u64)fl->fl_pid);
/* This should be adjusted, but I'm not sure if
namespaces actually get id numbers*/
req->r_args.filelock_change.pid_namespace =
- cpu_to_le64((u64)pid_ns);
- req->r_args.filelock_change.start = cpu_to_le64(start);
+ cpu_to_le64((u64)(unsigned long)fl->fl_nspid);
+ req->r_args.filelock_change.start = cpu_to_le64(fl->fl_start);
req->r_args.filelock_change.length = cpu_to_le64(length);
req->r_args.filelock_change.wait = wait;
err = ceph_mdsc_do_request(mdsc, inode, req);
+
+ if ( operation == CEPH_MDS_OP_GETFILELOCK){
+ fl->fl_pid = le64_to_cpu(req->r_reply_info.filelock_reply->pid);
+ if (CEPH_LOCK_SHARED == req->r_reply_info.filelock_reply->type)
+ fl->fl_type = F_RDLCK;
+ else if (CEPH_LOCK_EXCL == req->r_reply_info.filelock_reply->type)
+ fl->fl_type = F_WRLCK;
+ else
+ fl->fl_type = F_UNLCK;
+
+ fl->fl_start = le64_to_cpu(req->r_reply_info.filelock_reply->start);
+ length = le64_to_cpu(req->r_reply_info.filelock_reply->start) +
+ le64_to_cpu(req->r_reply_info.filelock_reply->length);
+ if (length >= 1)
+ fl->fl_end = length -1;
+ else
+ fl->fl_end = 0;
+
+ }
ceph_mdsc_put_request(req);
dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, "
- "length: %llu, wait: %d, type`: %d err code %d", (int)lock_type,
- (int)operation, pid, start, length, wait, cmd, err);
+ "length: %llu, wait: %d, type`: %d, err code %d", (int)lock_type,
+ (int)operation, (u64)fl->fl_pid, fl->fl_start,
+ length, wait, fl->fl_type, err);
return err;
}
@@ -54,7 +82,6 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file,
*/
int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
{
- u64 length;
u8 lock_cmd;
int err;
u8 wait = 0;
@@ -76,29 +103,20 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
else
lock_cmd = CEPH_LOCK_UNLOCK;
- if (LLONG_MAX == fl->fl_end)
- length = 0;
- else
- length = fl->fl_end - fl->fl_start + 1;
-
- err = ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
- (u64)fl->fl_pid,
- (u64)(unsigned long)fl->fl_nspid,
- lock_cmd, fl->fl_start,
- length, wait);
+ err = ceph_lock_message(CEPH_LOCK_FCNTL, op, file, lock_cmd, wait, fl);
if (!err) {
- dout("mds locked, locking locally");
- err = posix_lock_file(file, fl, NULL);
- if (err && (CEPH_MDS_OP_SETFILELOCK == op)) {
- /* undo! This should only happen if the kernel detects
- * local deadlock. */
- ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
- (u64)fl->fl_pid,
- (u64)(unsigned long)fl->fl_nspid,
- CEPH_LOCK_UNLOCK, fl->fl_start,
- length, 0);
- dout("got %d on posix_lock_file, undid lock", err);
+ if ( op != CEPH_MDS_OP_GETFILELOCK ){
+ dout("mds locked, locking locally");
+ err = posix_lock_file(file, fl, NULL);
+ if (err && (CEPH_MDS_OP_SETFILELOCK == op)) {
+ /* undo! This should only happen if the kernel detects
+ * local deadlock. */
+ ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
+ CEPH_LOCK_UNLOCK, 0, fl);
+ dout("got %d on posix_lock_file, undid lock", err);
+ }
}
+
} else {
dout("mds returned error code %d", err);
}
@@ -107,7 +125,6 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
int ceph_flock(struct file *file, int cmd, struct file_lock *fl)
{
- u64 length;
u8 lock_cmd;
int err;
u8 wait = 1;
@@ -127,26 +144,15 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl)
lock_cmd = CEPH_LOCK_EXCL;
else
lock_cmd = CEPH_LOCK_UNLOCK;
- /* mds requires start and length rather than start and end */
- if (LLONG_MAX == fl->fl_end)
- length = 0;
- else
- length = fl->fl_end - fl->fl_start + 1;
err = ceph_lock_message(CEPH_LOCK_FLOCK, CEPH_MDS_OP_SETFILELOCK,
- file, (u64)fl->fl_pid,
- (u64)(unsigned long)fl->fl_nspid,
- lock_cmd, fl->fl_start,
- length, wait);
+ file, lock_cmd, wait, fl);
if (!err) {
err = flock_lock_file_wait(file, fl);
if (err) {
ceph_lock_message(CEPH_LOCK_FLOCK,
CEPH_MDS_OP_SETFILELOCK,
- file, (u64)fl->fl_pid,
- (u64)(unsigned long)fl->fl_nspid,
- CEPH_LOCK_UNLOCK, fl->fl_start,
- length, 0);
+ file, CEPH_LOCK_UNLOCK, 0, fl);
dout("got %d on flock_lock_file_wait, undid lock", err);
}
} else {
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 3142b15940c2..38800eaa81d0 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -6,7 +6,6 @@
#include <linux/sched.h>
#include <linux/debugfs.h>
#include <linux/seq_file.h>
-#include <linux/smp_lock.h>
#include "super.h"
#include "mds_client.h"
@@ -203,6 +202,38 @@ out_bad:
}
/*
+ * parse fcntl F_GETLK results
+ */
+static int parse_reply_info_filelock(void **p, void *end,
+ struct ceph_mds_reply_info_parsed *info)
+{
+ if (*p + sizeof(*info->filelock_reply) > end)
+ goto bad;
+
+ info->filelock_reply = *p;
+ *p += sizeof(*info->filelock_reply);
+
+ if (unlikely(*p != end))
+ goto bad;
+ return 0;
+
+bad:
+ return -EIO;
+}
+
+/*
+ * parse extra results
+ */
+static int parse_reply_info_extra(void **p, void *end,
+ struct ceph_mds_reply_info_parsed *info)
+{
+ if (info->head->op == CEPH_MDS_OP_GETFILELOCK)
+ return parse_reply_info_filelock(p, end, info);
+ else
+ return parse_reply_info_dir(p, end, info);
+}
+
+/*
* parse entire mds reply
*/
static int parse_reply_info(struct ceph_msg *msg,
@@ -224,10 +255,10 @@ static int parse_reply_info(struct ceph_msg *msg,
goto out_bad;
}
- /* dir content */
+ /* extra */
ceph_decode_32_safe(&p, end, len, bad);
if (len > 0) {
- err = parse_reply_info_dir(&p, p+len, info);
+ err = parse_reply_info_extra(&p, p+len, info);
if (err < 0)
goto out_bad;
}
@@ -529,6 +560,9 @@ static void __register_request(struct ceph_mds_client *mdsc,
ceph_mdsc_get_request(req);
__insert_request(mdsc, req);
+ req->r_uid = current_fsuid();
+ req->r_gid = current_fsgid();
+
if (dir) {
struct ceph_inode_info *ci = ceph_inode(dir);
@@ -1588,8 +1622,8 @@ static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc,
head->mdsmap_epoch = cpu_to_le32(mdsc->mdsmap->m_epoch);
head->op = cpu_to_le32(req->r_op);
- head->caller_uid = cpu_to_le32(current_fsuid());
- head->caller_gid = cpu_to_le32(current_fsgid());
+ head->caller_uid = cpu_to_le32(req->r_uid);
+ head->caller_gid = cpu_to_le32(req->r_gid);
head->args = req->r_args;
ceph_encode_filepath(&p, end, ino1, path1);
@@ -2072,7 +2106,7 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
mutex_lock(&session->s_mutex);
if (err < 0) {
- pr_err("mdsc_handle_reply got corrupt reply mds%d\n", mds);
+ pr_err("mdsc_handle_reply got corrupt reply mds%d(tid:%lld)\n", mds, tid);
ceph_msg_dump(msg);
goto out_err;
}
@@ -2092,7 +2126,8 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
mutex_lock(&req->r_fill_mutex);
err = ceph_fill_trace(mdsc->fsc->sb, req, req->r_session);
if (err == 0) {
- if (result == 0 && rinfo->dir_nr)
+ if (result == 0 && req->r_op != CEPH_MDS_OP_GETFILELOCK &&
+ rinfo->dir_nr)
ceph_readdir_prepopulate(req, req->r_session);
ceph_unreserve_caps(mdsc, &req->r_caps_reservation);
}
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index d66d63c72355..aabe563b54db 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -42,26 +42,37 @@ struct ceph_mds_reply_info_in {
};
/*
- * parsed info about an mds reply, including information about the
- * target inode and/or its parent directory and dentry, and directory
- * contents (for readdir results).
+ * parsed info about an mds reply, including information about
+ * either: 1) the target inode and/or its parent directory and dentry,
+ * and directory contents (for readdir results), or
+ * 2) the file range lock info (for fcntl F_GETLK results).
*/
struct ceph_mds_reply_info_parsed {
struct ceph_mds_reply_head *head;
+ /* trace */
struct ceph_mds_reply_info_in diri, targeti;
struct ceph_mds_reply_dirfrag *dirfrag;
char *dname;
u32 dname_len;
struct ceph_mds_reply_lease *dlease;
- struct ceph_mds_reply_dirfrag *dir_dir;
- int dir_nr;
- char **dir_dname;
- u32 *dir_dname_len;
- struct ceph_mds_reply_lease **dir_dlease;
- struct ceph_mds_reply_info_in *dir_in;
- u8 dir_complete, dir_end;
+ /* extra */
+ union {
+ /* for fcntl F_GETLK results */
+ struct ceph_filelock *filelock_reply;
+
+ /* for readdir results */
+ struct {
+ struct ceph_mds_reply_dirfrag *dir_dir;
+ int dir_nr;
+ char **dir_dname;
+ u32 *dir_dname_len;
+ struct ceph_mds_reply_lease **dir_dlease;
+ struct ceph_mds_reply_info_in *dir_in;
+ u8 dir_complete, dir_end;
+ };
+ };
/* encoded blob describing snapshot contexts for certain
operations (e.g., open) */
@@ -170,6 +181,8 @@ struct ceph_mds_request {
union ceph_mds_request_args r_args;
int r_fmode; /* file mode, if expecting cap */
+ uid_t r_uid;
+ gid_t r_gid;
/* for choosing which mds to send this request to */
int r_direct_mode;
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index 1886294e12f7..7f01728a4657 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -293,9 +293,7 @@ struct ceph_inode_info {
int i_rd_ref, i_rdcache_ref, i_wr_ref;
int i_wrbuffer_ref, i_wrbuffer_ref_head;
u32 i_shared_gen; /* increment each time we get FILE_SHARED */
- u32 i_rdcache_gen; /* we increment this each time we get
- FILE_CACHE. If it's non-zero, we
- _may_ have cached pages. */
+ u32 i_rdcache_gen; /* incremented each time we get FILE_CACHE. */
u32 i_rdcache_revoking; /* RDCACHE gen to async invalidate, if any */
struct list_head i_unsafe_writes; /* uncommitted sync writes */