From 2dc75d3c3b49c64fd26b4832a7efb75546cb3fc5 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Thu, 11 Sep 2008 14:20:23 +0200
Subject: block: disable sysfs parts of the disk command filter

We still have life time issues with the sysfs command filter kobject,
so disable it for 2.6.27 release. We can revisit this and make it work
properly for 2.6.28, for 2.6.27 release it's too risky.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 include/linux/blkdev.h | 2 --
 1 file changed, 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 44710d7e7bff..53ea933cf60b 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -843,8 +843,6 @@ extern int blkdev_issue_flush(struct block_device *, sector_t *);
 */
 extern int blk_verify_command(struct blk_cmd_filter *filter,
 			      unsigned char *cmd, int has_write_perm);
-extern int blk_register_filter(struct gendisk *disk);
-extern void blk_unregister_filter(struct gendisk *disk);
 extern void blk_set_cmd_filter_defaults(struct blk_cmd_filter *filter);
 
 #define MAX_PHYS_SEGMENTS 128
-- 
cgit v1.3-7-g2ca7


From 97b697a11b07e2ebfa69c488132596cc5eb24119 Mon Sep 17 00:00:00 2001
From: Taisuke Yamada <tai@rakugaki.org>
Date: Sat, 13 Sep 2008 16:46:15 -0400
Subject: [libata] LBA28/LBA48 off-by-one bug in ata.h

I recently bought 3 HGST P7K500-series 500GB SATA drives and
had trouble accessing the block right on the LBA28-LBA48 border.
Here's how it fails (same for all 3 drives):

  # dd if=/dev/sdc bs=512 count=1 skip=268435455 > /dev/null
  dd: reading `/dev/sdc': Input/output error
  0+0 records in
  0+0 records out
  0 bytes (0 B) copied, 0.288033 seconds, 0.0 kB/s
  # dmesg
  ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
  ata1.00: BMDMA stat 0x25
  ata1.00: cmd c8/00:08:f8:ff:ff/00:00:00:00:00/ef tag 0 dma 4096 in
  res 51/04:08:f8:ff:ff/00:00:00:00:00/ef Emask 0x1 (device error)
  ata1.00: status: { DRDY ERR }
  ata1.00: error: { ABRT }
  ata1.00: configured for UDMA/33
  ata1: EH complete
  ...

After some investigations, it turned out this seems to be caused
by misinterpretation of the ATA specification on LBA28 access.
Following part is the code in question:

  === include/linux/ata.h ===
  static inline int lba_28_ok(u64 block, u32 n_block)
  {
    /* check the ending block number */
    return ((block + n_block - 1) < ((u64)1 << 28)) && (n_block <= 256);
  }

HGST drive (sometimes) fails with LBA28 access of {block = 0xfffffff,
n_block = 1}, and this behavior seems to be comformant. Other drives,
including other HGST drives are not that strict, through.

>From the ATA specification:
(http://www.t13.org/Documents/UploadedDocuments/project/d1410r3b-ATA-ATAPI-6.pdf)

  8.15.29  Word (61:60): Total number of user addressable sectors
  This field contains a value that is one greater than the total number
  of user addressable sectors (see 6.2). The maximum value that shall
  be placed in this field is 0FFFFFFFh.

So the driver shouldn't use the value of 0xfffffff for LBA28 request
as this exceeds maximum user addressable sector. The logical maximum
value for LBA28 is 0xffffffe.

The obvious fix is to cut "- 1" part, and the patch attached just do
that. I've been using the patched kernel for about a month now, and
the same fix is also floating on the net for some time. So I believe
this fix works reliably.

Just FYI, many Windows/Intel platform users also seems to be struck
by this, and HGST has issued a note pointing to Intel ICH8/9 driver.

  "28-bit LBA command is being used to access LBAs 29-bits in length"
http://www.hitachigst.com/hddt/knowtree.nsf/cffe836ed7c12018862565b000530c74/b531b8bce8745fb78825740f00580e23

Also, *BSDs seems to have similar fix included sometime around ~2004,
through I have not checked out exact portion of the code.

Signed-off-by: Taisuke Yamada <tai@rakugaki.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/ata.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/ata.h b/include/linux/ata.h
index 1ce19c1ef0e9..8a12d718c169 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -745,7 +745,7 @@ static inline int ata_ok(u8 status)
 static inline int lba_28_ok(u64 block, u32 n_block)
 {
 	/* check the ending block number */
-	return ((block + n_block - 1) < ((u64)1 << 28)) && (n_block <= 256);
+	return ((block + n_block) < ((u64)1 << 28)) && (n_block <= 256);
 }
 
 static inline int lba_48_ok(u64 block, u32 n_block)
-- 
cgit v1.3-7-g2ca7


From dea420ce0e2973e8ef1fd11fde6804c8d03a82ad Mon Sep 17 00:00:00 2001
From: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
Date: Sat, 13 Sep 2008 02:33:07 -0700
Subject: include/linux/ioport.h: add missing macro argument for devm_release_*
 family

akpm: these have no callers at this time, but they shall soon, so let's
get them right.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
Cc: Tony Lindgren <tony@atomide.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/ioport.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/ioport.h b/include/linux/ioport.h
index 8d3b7a9afd17..350033e8f4e1 100644
--- a/include/linux/ioport.h
+++ b/include/linux/ioport.h
@@ -159,9 +159,9 @@ extern struct resource * __devm_request_region(struct device *dev,
 				struct resource *parent, resource_size_t start,
 				resource_size_t n, const char *name);
 
-#define devm_release_region(start,n) \
+#define devm_release_region(dev, start, n) \
 	__devm_release_region(dev, &ioport_resource, (start), (n))
-#define devm_release_mem_region(start,n) \
+#define devm_release_mem_region(dev, start, n) \
 	__devm_release_region(dev, &iomem_resource, (start), (n))
 
 extern void __devm_release_region(struct device *dev, struct resource *parent,
-- 
cgit v1.3-7-g2ca7


From 5bead2a0680687b9576d57c177988e8aa082b922 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Sat, 13 Sep 2008 02:33:19 -0700
Subject: mm: mark the correct zone as full when scanning zonelists

The iterator for_each_zone_zonelist() uses a struct zoneref *z cursor when
scanning zonelists to keep track of where in the zonelist it is.  The
zoneref that is returned corresponds to the the next zone that is to be
scanned, not the current one.  It was intended to be treated as an opaque
list.

When the page allocator is scanning a zonelist, it marks elements in the
zonelist corresponding to zones that are temporarily full.  As the
zonelist is being updated, it uses the cursor here;

  if (NUMA_BUILD)
        zlc_mark_zone_full(zonelist, z);

This is intended to prevent rescanning in the near future but the zoneref
cursor does not correspond to the zone that has been found to be full.
This is an easy misunderstanding to make so this patch corrects the
problem by changing zoneref cursor to be the current zone being scanned
instead of the next one.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: <stable@kernel.org>		[2.6.26.x]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/mmzone.h | 12 ++++++------
 mm/mmzone.c            |  2 +-
 2 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 443bc7cd8c62..428328a05fa1 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -751,8 +751,9 @@ static inline int zonelist_node_idx(struct zoneref *zoneref)
  *
  * This function returns the next zone at or below a given zone index that is
  * within the allowed nodemask using a cursor as the starting point for the
- * search. The zoneref returned is a cursor that is used as the next starting
- * point for future calls to next_zones_zonelist().
+ * search. The zoneref returned is a cursor that represents the current zone
+ * being examined. It should be advanced by one before calling
+ * next_zones_zonelist again.
  */
 struct zoneref *next_zones_zonelist(struct zoneref *z,
 					enum zone_type highest_zoneidx,
@@ -768,9 +769,8 @@ struct zoneref *next_zones_zonelist(struct zoneref *z,
  *
  * This function returns the first zone at or below a given zone index that is
  * within the allowed nodemask. The zoneref returned is a cursor that can be
- * used to iterate the zonelist with next_zones_zonelist. The cursor should
- * not be used by the caller as it does not match the value of the zone
- * returned.
+ * used to iterate the zonelist with next_zones_zonelist by advancing it by
+ * one before calling.
  */
 static inline struct zoneref *first_zones_zonelist(struct zonelist *zonelist,
 					enum zone_type highest_zoneidx,
@@ -795,7 +795,7 @@ static inline struct zoneref *first_zones_zonelist(struct zonelist *zonelist,
 #define for_each_zone_zonelist_nodemask(zone, z, zlist, highidx, nodemask) \
 	for (z = first_zones_zonelist(zlist, highidx, nodemask, &zone);	\
 		zone;							\
-		z = next_zones_zonelist(z, highidx, nodemask, &zone))	\
+		z = next_zones_zonelist(++z, highidx, nodemask, &zone))	\
 
 /**
  * for_each_zone_zonelist - helper macro to iterate over valid zones in a zonelist at or below a given zone index
diff --git a/mm/mmzone.c b/mm/mmzone.c
index 486ed595ee6f..16ce8b955dcf 100644
--- a/mm/mmzone.c
+++ b/mm/mmzone.c
@@ -69,6 +69,6 @@ struct zoneref *next_zones_zonelist(struct zoneref *z,
 				(z->zone && !zref_in_nodemask(z, nodes)))
 			z++;
 
-	*zone = zonelist_zone(z++);
+	*zone = zonelist_zone(z);
 	return z;
 }
-- 
cgit v1.3-7-g2ca7


From 8e82f8c34b1759ae0d80fe96101746ec51fb1ba4 Mon Sep 17 00:00:00 2001
From: Alex Dubov <oakad@yahoo.com>
Date: Sat, 13 Sep 2008 02:33:26 -0700
Subject: memstick: fix MSProHG 8-bit interface mode support

- 8-bit interface mode never worked properly.  The only adapter I have
  which supports the 8b mode (the Jmicron) had some problems with its
  clock wiring and they discovered it only now.  We also discovered that
  ProHG media is more sensitive to the ordering of initialization
  commands.

- Make the driver fall back to highest supported mode instead of always
  falling back to serial.  The driver will attempt the switch to 8b mode
  for any new MSPro card, but not all of them support it.  Previously,
  these new cards ended up in serial mode, which is not the best idea
  (they work fine with 4b, after all).

- Edit some macros for better conformance to Sony documentation

Signed-off-by: Alex Dubov <oakad@yahoo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/memstick/core/memstick.c    | 10 ++--
 drivers/memstick/core/mspro_block.c | 33 ++++++++++---
 drivers/memstick/host/jmb38x_ms.c   | 39 ++++++++++-----
 include/linux/memstick.h            | 97 +++++++++++++++++++------------------
 4 files changed, 106 insertions(+), 73 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/memstick/core/memstick.c b/drivers/memstick/core/memstick.c
index a38005008a20..cea46906408e 100644
--- a/drivers/memstick/core/memstick.c
+++ b/drivers/memstick/core/memstick.c
@@ -185,7 +185,7 @@ static void memstick_free(struct device *dev)
 }
 
 static struct class memstick_host_class = {
-	.name       = "memstick_host",
+	.name        = "memstick_host",
 	.dev_release = memstick_free
 };
 
@@ -264,7 +264,7 @@ EXPORT_SYMBOL(memstick_new_req);
  * @sg - TPC argument
  */
 void memstick_init_req_sg(struct memstick_request *mrq, unsigned char tpc,
-			  struct scatterlist *sg)
+			  const struct scatterlist *sg)
 {
 	mrq->tpc = tpc;
 	if (tpc & 8)
@@ -294,7 +294,7 @@ EXPORT_SYMBOL(memstick_init_req_sg);
  * user supplied buffer.
  */
 void memstick_init_req(struct memstick_request *mrq, unsigned char tpc,
-		       void *buf, size_t length)
+		       const void *buf, size_t length)
 {
 	mrq->tpc = tpc;
 	if (tpc & 8)
@@ -439,7 +439,7 @@ static void memstick_check(struct work_struct *work)
 	if (!host->card) {
 		if (memstick_power_on(host))
 			goto out_power_off;
-	} else
+	} else if (host->card->stop)
 		host->card->stop(host->card);
 
 	card = memstick_alloc_card(host);
@@ -458,7 +458,7 @@ static void memstick_check(struct work_struct *work)
 			    || !(host->card->check(host->card))) {
 				device_unregister(&host->card->dev);
 				host->card = NULL;
-			} else
+			} else if (host->card->start)
 				host->card->start(host->card);
 		}
 
diff --git a/drivers/memstick/core/mspro_block.c b/drivers/memstick/core/mspro_block.c
index 44b1817f2f2f..d2d2318dafa4 100644
--- a/drivers/memstick/core/mspro_block.c
+++ b/drivers/memstick/core/mspro_block.c
@@ -30,6 +30,8 @@ module_param(major, int, 0644);
 #define MSPRO_BLOCK_SIGNATURE        0xa5c3
 #define MSPRO_BLOCK_MAX_ATTRIBUTES   41
 
+#define MSPRO_BLOCK_PART_SHIFT 3
+
 enum {
 	MSPRO_BLOCK_ID_SYSINFO         = 0x10,
 	MSPRO_BLOCK_ID_MODELNAME       = 0x15,
@@ -195,7 +197,7 @@ static int mspro_block_bd_open(struct inode *inode, struct file *filp)
 static int mspro_block_disk_release(struct gendisk *disk)
 {
 	struct mspro_block_data *msb = disk->private_data;
-	int disk_id = disk->first_minor >> MEMSTICK_PART_SHIFT;
+	int disk_id = disk->first_minor >> MSPRO_BLOCK_PART_SHIFT;
 
 	mutex_lock(&mspro_block_disk_lock);
 
@@ -877,6 +879,7 @@ static int mspro_block_switch_interface(struct memstick_dev *card)
 	struct mspro_block_data *msb = memstick_get_drvdata(card);
 	int rc = 0;
 
+try_again:
 	if (msb->caps & MEMSTICK_CAP_PAR4)
 		rc = mspro_block_set_interface(card, MEMSTICK_SYS_PAR4);
 	else
@@ -930,6 +933,18 @@ static int mspro_block_switch_interface(struct memstick_dev *card)
 		rc = memstick_set_rw_addr(card);
 		if (!rc)
 			rc = mspro_block_set_interface(card, msb->system);
+
+		if (!rc) {
+			msleep(150);
+			rc = mspro_block_wait_for_ced(card);
+			if (rc)
+				return rc;
+
+			if (msb->caps & MEMSTICK_CAP_PAR8) {
+				msb->caps &= ~MEMSTICK_CAP_PAR8;
+				goto try_again;
+			}
+		}
 	}
 	return rc;
 }
@@ -1117,14 +1132,16 @@ static int mspro_block_init_card(struct memstick_dev *card)
 		return -EIO;
 
 	msb->caps = host->caps;
-	rc = mspro_block_switch_interface(card);
+
+	msleep(150);
+	rc = mspro_block_wait_for_ced(card);
 	if (rc)
 		return rc;
 
-	msleep(200);
-	rc = mspro_block_wait_for_ced(card);
+	rc = mspro_block_switch_interface(card);
 	if (rc)
 		return rc;
+
 	dev_dbg(&card->dev, "card activated\n");
 	if (msb->system != MEMSTICK_SYS_SERIAL)
 		msb->caps |= MEMSTICK_CAP_AUTO_GET_INT;
@@ -1192,12 +1209,12 @@ static int mspro_block_init_disk(struct memstick_dev *card)
 	if (rc)
 		return rc;
 
-	if ((disk_id << MEMSTICK_PART_SHIFT) > 255) {
+	if ((disk_id << MSPRO_BLOCK_PART_SHIFT) > 255) {
 		rc = -ENOSPC;
 		goto out_release_id;
 	}
 
-	msb->disk = alloc_disk(1 << MEMSTICK_PART_SHIFT);
+	msb->disk = alloc_disk(1 << MSPRO_BLOCK_PART_SHIFT);
 	if (!msb->disk) {
 		rc = -ENOMEM;
 		goto out_release_id;
@@ -1220,7 +1237,7 @@ static int mspro_block_init_disk(struct memstick_dev *card)
 				   MSPRO_BLOCK_MAX_PAGES * msb->page_size);
 
 	msb->disk->major = major;
-	msb->disk->first_minor = disk_id << MEMSTICK_PART_SHIFT;
+	msb->disk->first_minor = disk_id << MSPRO_BLOCK_PART_SHIFT;
 	msb->disk->fops = &ms_block_bdops;
 	msb->usage_count = 1;
 	msb->disk->private_data = msb;
@@ -1416,7 +1433,7 @@ out_unlock:
 
 static struct memstick_device_id mspro_block_id_tbl[] = {
 	{MEMSTICK_MATCH_ALL, MEMSTICK_TYPE_PRO, MEMSTICK_CATEGORY_STORAGE_DUO,
-	 MEMSTICK_CLASS_GENERIC_DUO},
+	 MEMSTICK_CLASS_DUO},
 	{}
 };
 
diff --git a/drivers/memstick/host/jmb38x_ms.c b/drivers/memstick/host/jmb38x_ms.c
index 3485c63d20b0..2fb95a5b72eb 100644
--- a/drivers/memstick/host/jmb38x_ms.c
+++ b/drivers/memstick/host/jmb38x_ms.c
@@ -81,6 +81,8 @@ struct jmb38x_ms {
 #define TPC_CODE_SZ_MASK       0x00000700
 #define TPC_DATA_SZ_MASK       0x00000007
 
+#define HOST_CONTROL_TDELAY_EN 0x00040000
+#define HOST_CONTROL_HW_OC_P   0x00010000
 #define HOST_CONTROL_RESET_REQ 0x00008000
 #define HOST_CONTROL_REI       0x00004000
 #define HOST_CONTROL_LED       0x00000400
@@ -88,6 +90,7 @@ struct jmb38x_ms {
 #define HOST_CONTROL_RESET     0x00000100
 #define HOST_CONTROL_POWER_EN  0x00000080
 #define HOST_CONTROL_CLOCK_EN  0x00000040
+#define HOST_CONTROL_REO       0x00000008
 #define HOST_CONTROL_IF_SHIFT  4
 
 #define HOST_CONTROL_IF_SERIAL 0x0
@@ -133,11 +136,15 @@ struct jmb38x_ms {
 #define PAD_PU_PD_ON_MS_SOCK1 0x0f0f0000
 
 #define CLOCK_CONTROL_40MHZ   0x00000001
-#define CLOCK_CONTROL_50MHZ   0x00000002
+#define CLOCK_CONTROL_50MHZ   0x0000000a
 #define CLOCK_CONTROL_60MHZ   0x00000008
 #define CLOCK_CONTROL_62_5MHZ 0x0000000c
 #define CLOCK_CONTROL_OFF     0x00000000
 
+#define PCI_CTL_CLOCK_DLY_ADDR   0x000000b0
+#define PCI_CTL_CLOCK_DLY_MASK_A 0x00000f00
+#define PCI_CTL_CLOCK_DLY_MASK_B 0x0000f000
+
 enum {
 	CMD_READY    = 0x01,
 	FIFO_READY   = 0x02,
@@ -367,8 +374,7 @@ static int jmb38x_ms_issue_cmd(struct memstick_host *msh)
 		return host->req->error;
 	}
 
-	dev_dbg(&msh->dev, "control %08x\n",
-		readl(host->addr + HOST_CONTROL));
+	dev_dbg(&msh->dev, "control %08x\n", readl(host->addr + HOST_CONTROL));
 	dev_dbg(&msh->dev, "status %08x\n", readl(host->addr + INT_STATUS));
 	dev_dbg(&msh->dev, "hstatus %08x\n", readl(host->addr + STATUS));
 
@@ -637,7 +643,7 @@ static int jmb38x_ms_reset(struct jmb38x_ms_host *host)
 		ndelay(20);
 	}
 	dev_dbg(&host->chip->pdev->dev, "reset_req timeout\n");
-	return -EIO;
+	/* return -EIO; */
 
 reset_next:
 	writel(HOST_CONTROL_RESET | HOST_CONTROL_CLOCK_EN
@@ -680,7 +686,9 @@ static int jmb38x_ms_set_param(struct memstick_host *msh,
 
 			host_ctl = 7;
 			host_ctl |= HOST_CONTROL_POWER_EN
-				 | HOST_CONTROL_CLOCK_EN;
+				    | HOST_CONTROL_CLOCK_EN
+				    | HOST_CONTROL_HW_OC_P
+				    | HOST_CONTROL_TDELAY_EN;
 			writel(host_ctl, host->addr + HOST_CONTROL);
 
 			writel(host->id ? PAD_PU_PD_ON_MS_SOCK1
@@ -704,33 +712,40 @@ static int jmb38x_ms_set_param(struct memstick_host *msh,
 		break;
 	case MEMSTICK_INTERFACE:
 		host_ctl &= ~(3 << HOST_CONTROL_IF_SHIFT);
+		pci_read_config_dword(host->chip->pdev,
+				      PCI_CTL_CLOCK_DLY_ADDR,
+				      &clock_delay);
+		clock_delay &= host->id ? ~PCI_CTL_CLOCK_DLY_MASK_B
+					: ~PCI_CTL_CLOCK_DLY_MASK_A;
 
 		if (value == MEMSTICK_SERIAL) {
 			host_ctl &= ~HOST_CONTROL_FAST_CLK;
+			host_ctl &= ~HOST_CONTROL_REO;
 			host_ctl |= HOST_CONTROL_IF_SERIAL
 				    << HOST_CONTROL_IF_SHIFT;
 			host_ctl |= HOST_CONTROL_REI;
 			clock_ctl = CLOCK_CONTROL_40MHZ;
-			clock_delay = 0;
 		} else if (value == MEMSTICK_PAR4) {
-			host_ctl |= HOST_CONTROL_FAST_CLK;
+			host_ctl |= HOST_CONTROL_FAST_CLK | HOST_CONTROL_REO;
 			host_ctl |= HOST_CONTROL_IF_PAR4
 				    << HOST_CONTROL_IF_SHIFT;
 			host_ctl &= ~HOST_CONTROL_REI;
 			clock_ctl = CLOCK_CONTROL_40MHZ;
-			clock_delay = 4;
+			clock_delay |= host->id ? (4 << 12) : (4 << 8);
 		} else if (value == MEMSTICK_PAR8) {
 			host_ctl |= HOST_CONTROL_FAST_CLK;
 			host_ctl |= HOST_CONTROL_IF_PAR8
 				    << HOST_CONTROL_IF_SHIFT;
-			host_ctl &= ~HOST_CONTROL_REI;
-			clock_ctl = CLOCK_CONTROL_60MHZ;
-			clock_delay = 0;
+			host_ctl &= ~(HOST_CONTROL_REI | HOST_CONTROL_REO);
+			clock_ctl = CLOCK_CONTROL_50MHZ;
 		} else
 			return -EINVAL;
+
 		writel(host_ctl, host->addr + HOST_CONTROL);
 		writel(clock_ctl, host->addr + CLOCK_CONTROL);
-		writel(clock_delay, host->addr + CLOCK_DELAY);
+		pci_write_config_dword(host->chip->pdev,
+				       PCI_CTL_CLOCK_DLY_ADDR,
+				       clock_delay);
 		break;
 	};
 	return 0;
diff --git a/include/linux/memstick.h b/include/linux/memstick.h
index a9f998a3f48b..d0c37e682234 100644
--- a/include/linux/memstick.h
+++ b/include/linux/memstick.h
@@ -21,30 +21,30 @@
 struct ms_status_register {
 	unsigned char reserved;
 	unsigned char interrupt;
-#define MEMSTICK_INT_CMDNAK             0x0001
-#define MEMSTICK_INT_IOREQ              0x0008
-#define MEMSTICK_INT_IOBREQ             0x0010
-#define MEMSTICK_INT_BREQ               0x0020
-#define MEMSTICK_INT_ERR                0x0040
-#define MEMSTICK_INT_CED                0x0080
+#define MEMSTICK_INT_CMDNAK 0x01
+#define MEMSTICK_INT_IOREQ  0x08
+#define MEMSTICK_INT_IOBREQ 0x10
+#define MEMSTICK_INT_BREQ   0x20
+#define MEMSTICK_INT_ERR    0x40
+#define MEMSTICK_INT_CED    0x80
 
 	unsigned char status0;
-#define MEMSTICK_STATUS0_WP             0x0001
-#define MEMSTICK_STATUS0_SL             0x0002
-#define MEMSTICK_STATUS0_BF             0x0010
-#define MEMSTICK_STATUS0_BE             0x0020
-#define MEMSTICK_STATUS0_FB0            0x0040
-#define MEMSTICK_STATUS0_MB             0x0080
+#define MEMSTICK_STATUS0_WP  0x01
+#define MEMSTICK_STATUS0_SL  0x02
+#define MEMSTICK_STATUS0_BF  0x10
+#define MEMSTICK_STATUS0_BE  0x20
+#define MEMSTICK_STATUS0_FB0 0x40
+#define MEMSTICK_STATUS0_MB  0x80
 
 	unsigned char status1;
-#define MEMSTICK_STATUS1_UCFG           0x0001
-#define MEMSTICK_STATUS1_FGER           0x0002
-#define MEMSTICK_STATUS1_UCEX           0x0004
-#define MEMSTICK_STATUS1_EXER           0x0008
-#define MEMSTICK_STATUS1_UCDT           0x0010
-#define MEMSTICK_STATUS1_DTER           0x0020
-#define MEMSTICK_STATUS1_FBI            0x0040
-#define MEMSTICK_STATUS1_MB             0x0080
+#define MEMSTICK_STATUS1_UCFG 0x01
+#define MEMSTICK_STATUS1_FGER 0x02
+#define MEMSTICK_STATUS1_UCEX 0x04
+#define MEMSTICK_STATUS1_EXER 0x08
+#define MEMSTICK_STATUS1_UCDT 0x10
+#define MEMSTICK_STATUS1_DTER 0x20
+#define MEMSTICK_STATUS1_FB1  0x40
+#define MEMSTICK_STATUS1_MB   0x80
 } __attribute__((packed));
 
 struct ms_id_register {
@@ -56,32 +56,32 @@ struct ms_id_register {
 
 struct ms_param_register {
 	unsigned char system;
-#define MEMSTICK_SYS_ATEN 0xc0
-#define MEMSTICK_SYS_BAMD 0x80
 #define MEMSTICK_SYS_PAM  0x08
+#define MEMSTICK_SYS_BAMD 0x80
 
 	unsigned char block_address_msb;
 	unsigned short block_address;
 	unsigned char cp;
-#define MEMSTICK_CP_BLOCK               0x0000
-#define MEMSTICK_CP_PAGE                0x0020
-#define MEMSTICK_CP_EXTRA               0x0040
-#define MEMSTICK_CP_OVERWRITE           0x0080
+#define MEMSTICK_CP_BLOCK     0x00
+#define MEMSTICK_CP_PAGE      0x20
+#define MEMSTICK_CP_EXTRA     0x40
+#define MEMSTICK_CP_OVERWRITE 0x80
 
 	unsigned char page_address;
 } __attribute__((packed));
 
 struct ms_extra_data_register {
 	unsigned char  overwrite_flag;
-#define MEMSTICK_OVERWRITE_UPDATA       0x0010
-#define MEMSTICK_OVERWRITE_PAGE         0x0060
-#define MEMSTICK_OVERWRITE_BLOCK        0x0080
+#define MEMSTICK_OVERWRITE_UDST  0x10
+#define MEMSTICK_OVERWRITE_PGST1 0x20
+#define MEMSTICK_OVERWRITE_PGST0 0x40
+#define MEMSTICK_OVERWRITE_BKST  0x80
 
 	unsigned char  management_flag;
-#define MEMSTICK_MANAGEMENT_SYSTEM      0x0004
-#define MEMSTICK_MANAGEMENT_TRANS_TABLE 0x0008
-#define MEMSTICK_MANAGEMENT_COPY        0x0010
-#define MEMSTICK_MANAGEMENT_ACCESS      0x0020
+#define MEMSTICK_MANAGEMENT_SYSFLG 0x04
+#define MEMSTICK_MANAGEMENT_ATFLG  0x08
+#define MEMSTICK_MANAGEMENT_SCMS1  0x10
+#define MEMSTICK_MANAGEMENT_SCMS0  0x20
 
 	unsigned short logical_address;
 } __attribute__((packed));
@@ -96,9 +96,9 @@ struct ms_register {
 
 struct mspro_param_register {
 	unsigned char  system;
-#define MEMSTICK_SYS_SERIAL 0x80
 #define MEMSTICK_SYS_PAR4   0x00
 #define MEMSTICK_SYS_PAR8   0x40
+#define MEMSTICK_SYS_SERIAL 0x80
 
 	unsigned short data_count;
 	unsigned int   data_address;
@@ -147,7 +147,7 @@ struct ms_register_addr {
 	unsigned char w_length;
 } __attribute__((packed));
 
-enum {
+enum memstick_tpc {
 	MS_TPC_READ_MG_STATUS   = 0x01,
 	MS_TPC_READ_LONG_DATA   = 0x02,
 	MS_TPC_READ_SHORT_DATA  = 0x03,
@@ -167,7 +167,7 @@ enum {
 	MS_TPC_SET_CMD          = 0x0e
 };
 
-enum {
+enum memstick_command {
 	MS_CMD_BLOCK_END       = 0x33,
 	MS_CMD_RESET           = 0x3c,
 	MS_CMD_BLOCK_WRITE     = 0x55,
@@ -201,8 +201,6 @@ enum {
 
 /*** Driver structures and functions ***/
 
-#define MEMSTICK_PART_SHIFT 3
-
 enum memstick_param { MEMSTICK_POWER = 1, MEMSTICK_INTERFACE };
 
 #define MEMSTICK_POWER_OFF 0
@@ -215,24 +213,27 @@ enum memstick_param { MEMSTICK_POWER = 1, MEMSTICK_INTERFACE };
 struct memstick_host;
 struct memstick_driver;
 
+struct memstick_device_id {
+	unsigned char match_flags;
 #define MEMSTICK_MATCH_ALL            0x01
 
+	unsigned char type;
 #define MEMSTICK_TYPE_LEGACY          0xff
 #define MEMSTICK_TYPE_DUO             0x00
 #define MEMSTICK_TYPE_PRO             0x01
 
+	unsigned char category;
 #define MEMSTICK_CATEGORY_STORAGE     0xff
 #define MEMSTICK_CATEGORY_STORAGE_DUO 0x00
+#define MEMSTICK_CATEGORY_IO          0x01
+#define MEMSTICK_CATEGORY_IO_PRO      0x10
 
-#define MEMSTICK_CLASS_GENERIC        0xff
-#define MEMSTICK_CLASS_GENERIC_DUO    0x00
-
-
-struct memstick_device_id {
-	unsigned char match_flags;
-	unsigned char type;
-	unsigned char category;
 	unsigned char class;
+#define MEMSTICK_CLASS_FLASH          0xff
+#define MEMSTICK_CLASS_DUO            0x00
+#define MEMSTICK_CLASS_ROM            0x01
+#define MEMSTICK_CLASS_RO             0x02
+#define MEMSTICK_CLASS_WP             0x03
 };
 
 struct memstick_request {
@@ -319,9 +320,9 @@ void memstick_suspend_host(struct memstick_host *host);
 void memstick_resume_host(struct memstick_host *host);
 
 void memstick_init_req_sg(struct memstick_request *mrq, unsigned char tpc,
-			  struct scatterlist *sg);
+			  const struct scatterlist *sg);
 void memstick_init_req(struct memstick_request *mrq, unsigned char tpc,
-		       void *buf, size_t length);
+		       const void *buf, size_t length);
 int memstick_next_req(struct memstick_host *host,
 		      struct memstick_request **mrq);
 void memstick_new_req(struct memstick_host *host);
-- 
cgit v1.3-7-g2ca7


From 29bdc88384c2b24e37e5760df0dc898546083d6b Mon Sep 17 00:00:00 2001
From: Vladimir Sokolovsky <vlad@mellanox.co.il>
Date: Mon, 15 Sep 2008 14:25:23 -0700
Subject: IB/mlx4: Fix up fast register page list format

Byte swap the addresses in the page list for fast register work requests
to big endian to match what the HCA expectx.  Also, the addresses must
have the "present" bit set so that the HCA knows it can access them.
Otherwise the HCA will fault the first time it accesses the memory
region.

Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mlx4/qp.c | 6 ++++++
 drivers/net/mlx4/mr.c           | 2 --
 include/linux/mlx4/device.h     | 4 ++++
 3 files changed, 10 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index f29dbb767e87..9559248f265b 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -1342,6 +1342,12 @@ static __be32 convert_access(int acc)
 static void set_fmr_seg(struct mlx4_wqe_fmr_seg *fseg, struct ib_send_wr *wr)
 {
 	struct mlx4_ib_fast_reg_page_list *mfrpl = to_mfrpl(wr->wr.fast_reg.page_list);
+	int i;
+
+	for (i = 0; i < wr->wr.fast_reg.page_list_len; ++i)
+		wr->wr.fast_reg.page_list->page_list[i] =
+			cpu_to_be64(wr->wr.fast_reg.page_list->page_list[i] |
+				    MLX4_MTT_FLAG_PRESENT);
 
 	fseg->flags		= convert_access(wr->wr.fast_reg.access_flags);
 	fseg->mem_key		= cpu_to_be32(wr->wr.fast_reg.rkey);
diff --git a/drivers/net/mlx4/mr.c b/drivers/net/mlx4/mr.c
index 644adf074a35..d1dd5b48dbd1 100644
--- a/drivers/net/mlx4/mr.c
+++ b/drivers/net/mlx4/mr.c
@@ -71,8 +71,6 @@ struct mlx4_mpt_entry {
 #define MLX4_MPT_PD_FLAG_RAE	    (1 << 28)
 #define MLX4_MPT_PD_FLAG_EN_INV	    (3 << 24)
 
-#define MLX4_MTT_FLAG_PRESENT		1
-
 #define MLX4_MPT_STATUS_SW		0xF0
 #define MLX4_MPT_STATUS_HW		0x00
 
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 655ea0d1ee14..b2f944468313 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -141,6 +141,10 @@ enum {
 	MLX4_STAT_RATE_OFFSET	= 5
 };
 
+enum {
+	MLX4_MTT_FLAG_PRESENT		= 1
+};
+
 static inline u64 mlx4_fw_ver(u64 major, u64 minor, u64 subminor)
 {
 	return (major << 32) | (minor << 16) | subminor;
-- 
cgit v1.3-7-g2ca7


From b08508c40adf3fd1330aabc4f37d3254179776c4 Mon Sep 17 00:00:00 2001
From: Greg KH <gregkh@suse.de>
Date: Tue, 26 Aug 2008 08:20:34 -0700
Subject: PCI: fix compiler warnings in pci_get_subsys()

pci_get_subsys() changed in 2.6.26 so that the from pointer is modified
when the call is being invoked, so fix up the 'const' marking of it that
the compiler is complaining about.

Reported-by: Rufus & Azrael <rufus-azrael@numericable.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/search.c | 6 +++---
 include/linux/pci.h  | 8 ++++----
 2 files changed, 7 insertions(+), 7 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/pci/search.c b/drivers/pci/search.c
index 3b3b5f178797..4edfc4731bd4 100644
--- a/drivers/pci/search.c
+++ b/drivers/pci/search.c
@@ -162,7 +162,7 @@ EXPORT_SYMBOL(pci_find_slot);
  * time.
  */
 struct pci_dev *pci_find_device(unsigned int vendor, unsigned int device,
-				const struct pci_dev *from)
+				struct pci_dev *from)
 {
 	struct pci_dev *pdev;
 
@@ -263,7 +263,7 @@ static int match_pci_dev_by_id(struct device *dev, void *data)
  * this file.
  */
 static struct pci_dev *pci_get_dev_by_id(const struct pci_device_id *id,
-					 const struct pci_dev *from)
+					 struct pci_dev *from)
 {
 	struct device *dev;
 	struct device *dev_start = NULL;
@@ -303,7 +303,7 @@ static struct pci_dev *pci_get_dev_by_id(const struct pci_device_id *id,
  */
 struct pci_dev *pci_get_subsys(unsigned int vendor, unsigned int device,
 			       unsigned int ss_vendor, unsigned int ss_device,
-			       const struct pci_dev *from)
+			       struct pci_dev *from)
 {
 	struct pci_dev *pdev;
 	struct pci_device_id *id;
diff --git a/include/linux/pci.h b/include/linux/pci.h
index c0e14008a3c2..98dc6243a706 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -534,7 +534,7 @@ extern void pci_sort_breadthfirst(void);
 #ifdef CONFIG_PCI_LEGACY
 struct pci_dev __deprecated *pci_find_device(unsigned int vendor,
 					     unsigned int device,
-					     const struct pci_dev *from);
+					     struct pci_dev *from);
 struct pci_dev __deprecated *pci_find_slot(unsigned int bus,
 					   unsigned int devfn);
 #endif /* CONFIG_PCI_LEGACY */
@@ -550,7 +550,7 @@ struct pci_dev *pci_get_device(unsigned int vendor, unsigned int device,
 				struct pci_dev *from);
 struct pci_dev *pci_get_subsys(unsigned int vendor, unsigned int device,
 				unsigned int ss_vendor, unsigned int ss_device,
-				const struct pci_dev *from);
+				struct pci_dev *from);
 struct pci_dev *pci_get_slot(struct pci_bus *bus, unsigned int devfn);
 struct pci_dev *pci_get_bus_and_slot(unsigned int bus, unsigned int devfn);
 struct pci_dev *pci_get_class(unsigned int class, struct pci_dev *from);
@@ -816,7 +816,7 @@ _PCI_NOP_ALL(write,)
 
 static inline struct pci_dev *pci_find_device(unsigned int vendor,
 					      unsigned int device,
-					      const struct pci_dev *from)
+					      struct pci_dev *from)
 {
 	return NULL;
 }
@@ -838,7 +838,7 @@ static inline struct pci_dev *pci_get_subsys(unsigned int vendor,
 					     unsigned int device,
 					     unsigned int ss_vendor,
 					     unsigned int ss_device,
-					     const struct pci_dev *from)
+					     struct pci_dev *from)
 {
 	return NULL;
 }
-- 
cgit v1.3-7-g2ca7


From ef3d7714f6b75b51825ad0384b5ce48358427e50 Mon Sep 17 00:00:00 2001
From: David Miller <davem@davemloft.net>
Date: Tue, 16 Sep 2008 15:00:11 -0700
Subject: Fix PNP build failure, bugzilla #11276

This fill fix the following regression list entry:

Bug-Entry	: http://bugzilla.kernel.org/show_bug.cgi?id=11276
Subject		: build error: CONFIG_OPTIMIZE_INLINING=y causes gcc 4.2 to do stupid things
Submitter	: Randy Dunlap <randy.dunlap@oracle.com>
Date		: 2008-08-06 17:18 (38 days old)
References	: http://marc.info/?l=linux-kernel&m=121804329014332&w=4
		  http://lkml.org/lkml/2008/7/22/353
Handled-By	: Bjorn Helgaas <bjorn.helgaas@hp.com>
Patch		: http://lkml.org/lkml/2008/7/22/364

with what I believe is a better fix than the one referenced
in the regression entry above.

These PNP header interfaces try to work in such a way that
you can reference some of them even if PNP is not enabled,
and the compiler was expected to optimize everything away.

Which is mostly fine, except that there was one interface
for which there was not provided an inline "NOP" implementation.

Once we add that, all of these compile failures cannot handle
any more.

pnp: Provide NOP inline implementation of pnp_get_resource() when !PNP

Fixes kernel bugzilla #11276.

Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/pnp.h | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/pnp.h b/include/linux/pnp.h
index 1ce54b63085d..be764e514e35 100644
--- a/include/linux/pnp.h
+++ b/include/linux/pnp.h
@@ -21,7 +21,14 @@ struct pnp_dev;
 /*
  * Resource Management
  */
+#ifdef CONFIG_PNP
 struct resource *pnp_get_resource(struct pnp_dev *, unsigned int, unsigned int);
+#else
+static inline struct resource *pnp_get_resource(struct pnp_dev *dev, unsigned int type, unsigned int num)
+{
+	return NULL;
+}
+#endif
 
 static inline int pnp_resource_valid(struct resource *res)
 {
-- 
cgit v1.3-7-g2ca7


From c32a162fd420fe8dfb049db941b2438061047fcc Mon Sep 17 00:00:00 2001
From: "Kirill A. Shutemov" <kirill@shutemov.name>
Date: Mon, 22 Sep 2008 13:57:43 -0700
Subject: smb.h: do not include linux/time.h in userspace

linux/time.h conflicts with time.h from glibc

It breaks building smbmount from samba.  It's regression introduced by
commit 76308da (" smb.h: uses struct timespec but didn't include
linux/time.h").

Signed-off-by: Kirill A. Shutemov <kirill@shutemov.name>
Cc: <stable@kernel.org>             [2.6.26.x]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/smb.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/smb.h b/include/linux/smb.h
index caa43b2370cb..82fefddc5987 100644
--- a/include/linux/smb.h
+++ b/include/linux/smb.h
@@ -11,7 +11,9 @@
 
 #include <linux/types.h>
 #include <linux/magic.h>
+#ifdef __KERNEL__
 #include <linux/time.h>
+#endif
 
 enum smb_protocol { 
 	SMB_PROTOCOL_NONE, 
-- 
cgit v1.3-7-g2ca7


From b4f151ff899362fec952c45d166252c9912c041f Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 24 Sep 2008 17:48:26 +0100
Subject: MN10300: Move asm-arm/cnt32_to_63.h to include/linux/

Move asm-arm/cnt32_to_63.h to include/linux/ so that MN10300 can make
use of it too.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/arm/mach-pxa/time.c       |  2 +-
 arch/arm/mach-sa1100/generic.c |  2 +-
 arch/arm/mach-versatile/core.c |  2 +-
 include/linux/cnt32_to_63.h    | 80 ++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 83 insertions(+), 3 deletions(-)
 create mode 100644 include/linux/cnt32_to_63.h

(limited to 'include/linux')

diff --git a/arch/arm/mach-pxa/time.c b/arch/arm/mach-pxa/time.c
index 67e18509d7bf..b0d6b32654cf 100644
--- a/arch/arm/mach-pxa/time.c
+++ b/arch/arm/mach-pxa/time.c
@@ -17,9 +17,9 @@
 #include <linux/interrupt.h>
 #include <linux/clockchips.h>
 #include <linux/sched.h>
+#include <linux/cnt32_to_63.h>
 
 #include <asm/div64.h>
-#include <asm/cnt32_to_63.h>
 #include <asm/mach/irq.h>
 #include <asm/mach/time.h>
 #include <mach/pxa-regs.h>
diff --git a/arch/arm/mach-sa1100/generic.c b/arch/arm/mach-sa1100/generic.c
index 1362994c78aa..b422526f6d8b 100644
--- a/arch/arm/mach-sa1100/generic.c
+++ b/arch/arm/mach-sa1100/generic.c
@@ -18,9 +18,9 @@
 #include <linux/ioport.h>
 #include <linux/sched.h>	/* just for sched_clock() - funny that */
 #include <linux/platform_device.h>
+#include <linux/cnt32_to_63.h>
 
 #include <asm/div64.h>
-#include <asm/cnt32_to_63.h>
 #include <mach/hardware.h>
 #include <asm/system.h>
 #include <asm/pgtable.h>
diff --git a/arch/arm/mach-versatile/core.c b/arch/arm/mach-versatile/core.c
index d75e795c893e..b638f10411e8 100644
--- a/arch/arm/mach-versatile/core.c
+++ b/arch/arm/mach-versatile/core.c
@@ -28,8 +28,8 @@
 #include <linux/amba/clcd.h>
 #include <linux/clocksource.h>
 #include <linux/clockchips.h>
+#include <linux/cnt32_to_63.h>
 
-#include <asm/cnt32_to_63.h>
 #include <asm/system.h>
 #include <mach/hardware.h>
 #include <asm/io.h>
diff --git a/include/linux/cnt32_to_63.h b/include/linux/cnt32_to_63.h
new file mode 100644
index 000000000000..8c0f9505b48c
--- /dev/null
+++ b/include/linux/cnt32_to_63.h
@@ -0,0 +1,80 @@
+/*
+ *  Extend a 32-bit counter to 63 bits
+ *
+ *  Author:	Nicolas Pitre
+ *  Created:	December 3, 2006
+ *  Copyright:	MontaVista Software, Inc.
+ *
+ * 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_CNT32_TO_63_H__
+#define __LINUX_CNT32_TO_63_H__
+
+#include <linux/compiler.h>
+#include <linux/types.h>
+#include <asm/byteorder.h>
+
+/* this is used only to give gcc a clue about good code generation */
+union cnt32_to_63 {
+	struct {
+#if defined(__LITTLE_ENDIAN)
+		u32 lo, hi;
+#elif defined(__BIG_ENDIAN)
+		u32 hi, lo;
+#endif
+	};
+	u64 val;
+};
+
+
+/**
+ * cnt32_to_63 - Expand a 32-bit counter to a 63-bit counter
+ * @cnt_lo: The low part of the counter
+ *
+ * Many hardware clock counters are only 32 bits wide and therefore have
+ * a relatively short period making wrap-arounds rather frequent.  This
+ * is a problem when implementing sched_clock() for example, where a 64-bit
+ * non-wrapping monotonic value is expected to be returned.
+ *
+ * To overcome that limitation, let's extend a 32-bit counter to 63 bits
+ * in a completely lock free fashion. Bits 0 to 31 of the clock are provided
+ * by the hardware while bits 32 to 62 are stored in memory.  The top bit in
+ * memory is used to synchronize with the hardware clock half-period.  When
+ * the top bit of both counters (hardware and in memory) differ then the
+ * memory is updated with a new value, incrementing it when the hardware
+ * counter wraps around.
+ *
+ * Because a word store in memory is atomic then the incremented value will
+ * always be in synch with the top bit indicating to any potential concurrent
+ * reader if the value in memory is up to date or not with regards to the
+ * needed increment.  And any race in updating the value in memory is harmless
+ * as the same value would simply be stored more than once.
+ *
+ * The only restriction for the algorithm to work properly is that this
+ * code must be executed at least once per each half period of the 32-bit
+ * counter to properly update the state bit in memory. This is usually not a
+ * problem in practice, but if it is then a kernel timer could be scheduled
+ * to manage for this code to be executed often enough.
+ *
+ * Note that the top bit (bit 63) in the returned value should be considered
+ * as garbage.  It is not cleared here because callers are likely to use a
+ * multiplier on the returned value which can get rid of the top bit
+ * implicitly by making the multiplier even, therefore saving on a runtime
+ * clear-bit instruction. Otherwise caller must remember to clear the top
+ * bit explicitly.
+ */
+#define cnt32_to_63(cnt_lo) \
+({ \
+	static volatile u32 __m_cnt_hi; \
+	union cnt32_to_63 __x; \
+	__x.hi = __m_cnt_hi; \
+	__x.lo = (cnt_lo); \
+	if (unlikely((s32)(__x.hi ^ __x.lo) < 0)) \
+		__m_cnt_hi = __x.hi = (__x.hi ^ 0x80000000) + (__x.hi >> 31); \
+	__x.val; \
+})
+
+#endif
-- 
cgit v1.3-7-g2ca7


From 82ef04fb4c82542b3eda81cca461f0594ce9cd0b Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 31 Jul 2008 17:02:40 +0900
Subject: libata: make SCR access ops per-link

Logically, SCR access ops should take @link; however, there was no
compelling reason to convert all SCR access ops when adding @link
abstraction as there's one-to-one mapping between a port and a non-PMP
link.  However, that assumption won't hold anymore with the scheduled
addition of slave link.

Make SCR access ops per-link.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/ahci.c            | 24 ++++++++++++------------
 drivers/ata/ata_piix.c        | 15 +++++++++++----
 drivers/ata/libata-core.c     | 13 ++++---------
 drivers/ata/sata_fsl.c        | 26 +++++++++++++-------------
 drivers/ata/sata_inic162x.c   |  8 ++++----
 drivers/ata/sata_mv.c         | 28 ++++++++++++++--------------
 drivers/ata/sata_nv.c         | 16 ++++++++--------
 drivers/ata/sata_promise.c    | 16 +++++++++-------
 drivers/ata/sata_qstor.c      | 12 ++++++------
 drivers/ata/sata_sil.c        | 16 ++++++++--------
 drivers/ata/sata_sil24.c      | 12 ++++++------
 drivers/ata/sata_sis.c        | 28 ++++++++++++++++------------
 drivers/ata/sata_svw.c        | 10 ++++++----
 drivers/ata/sata_uli.c        | 24 ++++++++++++------------
 drivers/ata/sata_via.c        | 24 ++++++++++++------------
 drivers/ata/sata_vsc.c        | 10 ++++++----
 drivers/scsi/libsas/sas_ata.c |  8 ++++----
 include/linux/libata.h        |  4 ++--
 18 files changed, 153 insertions(+), 141 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 2e1a7cb2ed5f..6acea41eb7ca 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -267,8 +267,8 @@ struct ahci_port_priv {
 					 	 * per PM slot */
 };
 
-static int ahci_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val);
-static int ahci_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val);
+static int ahci_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val);
+static int ahci_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val);
 static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
 static unsigned int ahci_qc_issue(struct ata_queued_cmd *qc);
 static bool ahci_qc_fill_rtf(struct ata_queued_cmd *qc);
@@ -820,10 +820,10 @@ static unsigned ahci_scr_offset(struct ata_port *ap, unsigned int sc_reg)
 	return 0;
 }
 
-static int ahci_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
+static int ahci_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val)
 {
-	void __iomem *port_mmio = ahci_port_base(ap);
-	int offset = ahci_scr_offset(ap, sc_reg);
+	void __iomem *port_mmio = ahci_port_base(link->ap);
+	int offset = ahci_scr_offset(link->ap, sc_reg);
 
 	if (offset) {
 		*val = readl(port_mmio + offset);
@@ -832,10 +832,10 @@ static int ahci_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
 	return -EINVAL;
 }
 
-static int ahci_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
+static int ahci_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val)
 {
-	void __iomem *port_mmio = ahci_port_base(ap);
-	int offset = ahci_scr_offset(ap, sc_reg);
+	void __iomem *port_mmio = ahci_port_base(link->ap);
+	int offset = ahci_scr_offset(link->ap, sc_reg);
 
 	if (offset) {
 		writel(val, port_mmio + offset);
@@ -973,7 +973,7 @@ static void ahci_disable_alpm(struct ata_port *ap)
 	writel(PORT_IRQ_PHYRDY, port_mmio + PORT_IRQ_STAT);
 
 	/* go ahead and clean out PhyRdy Change from Serror too */
-	ahci_scr_write(ap, SCR_ERROR, ((1 << 16) | (1 << 18)));
+	ahci_scr_write(&ap->link, SCR_ERROR, ((1 << 16) | (1 << 18)));
 
 	/*
  	 * Clear flag to indicate that we should ignore all PhyRdy
@@ -1937,8 +1937,8 @@ static void ahci_error_intr(struct ata_port *ap, u32 irq_stat)
 	ata_ehi_push_desc(host_ehi, "irq_stat 0x%08x", irq_stat);
 
 	/* AHCI needs SError cleared; otherwise, it might lock up */
-	ahci_scr_read(ap, SCR_ERROR, &serror);
-	ahci_scr_write(ap, SCR_ERROR, serror);
+	ahci_scr_read(&ap->link, SCR_ERROR, &serror);
+	ahci_scr_write(&ap->link, SCR_ERROR, serror);
 	host_ehi->serror |= serror;
 
 	/* some controllers set IRQ_IF_ERR on device errors, ignore it */
@@ -2027,7 +2027,7 @@ static void ahci_port_intr(struct ata_port *ap)
 	if ((hpriv->flags & AHCI_HFLAG_NO_HOTPLUG) &&
 		(status & PORT_IRQ_PHYRDY)) {
 		status &= ~PORT_IRQ_PHYRDY;
-		ahci_scr_write(ap, SCR_ERROR, ((1 << 16) | (1 << 18)));
+		ahci_scr_write(&ap->link, SCR_ERROR, ((1 << 16) | (1 << 18)));
 	}
 
 	if (unlikely(status & PORT_IRQ_ERROR)) {
diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index e6b4606e36b6..81387ff48937 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -165,8 +165,10 @@ static void piix_set_dmamode(struct ata_port *ap, struct ata_device *adev);
 static void ich_set_dmamode(struct ata_port *ap, struct ata_device *adev);
 static int ich_pata_cable_detect(struct ata_port *ap);
 static u8 piix_vmw_bmdma_status(struct ata_port *ap);
-static int piix_sidpr_scr_read(struct ata_port *ap, unsigned int reg, u32 *val);
-static int piix_sidpr_scr_write(struct ata_port *ap, unsigned int reg, u32 val);
+static int piix_sidpr_scr_read(struct ata_link *link,
+			       unsigned int reg, u32 *val);
+static int piix_sidpr_scr_write(struct ata_link *link,
+				unsigned int reg, u32 val);
 #ifdef CONFIG_PM
 static int piix_pci_device_suspend(struct pci_dev *pdev, pm_message_t mesg);
 static int piix_pci_device_resume(struct pci_dev *pdev);
@@ -971,8 +973,10 @@ static u32 piix_merge_scr(u32 val0, u32 val1, const int * const *merge_tbl)
 	return val;
 }
 
-static int piix_sidpr_scr_read(struct ata_port *ap, unsigned int reg, u32 *val)
+static int piix_sidpr_scr_read(struct ata_link *link,
+			       unsigned int reg, u32 *val)
 {
+	struct ata_port *ap = link->ap;
 	const int * const sstatus_merge_tbl[] = {
 		/* DET */ (const int []){ 1, 3, 0, 4, 3, -1 },
 		/* SPD */ (const int []){ 2, 1, 0, -1 },
@@ -1013,8 +1017,11 @@ static int piix_sidpr_scr_read(struct ata_port *ap, unsigned int reg, u32 *val)
 	return 0;
 }
 
-static int piix_sidpr_scr_write(struct ata_port *ap, unsigned int reg, u32 val)
+static int piix_sidpr_scr_write(struct ata_link *link,
+				unsigned int reg, u32 val)
 {
+	struct ata_port *ap = link->ap;
+
 	if (reg >= ARRAY_SIZE(piix_sidx_map))
 		return -EINVAL;
 
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 79e3a8e7a84a..825461a33abe 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4868,10 +4868,8 @@ int sata_scr_valid(struct ata_link *link)
 int sata_scr_read(struct ata_link *link, int reg, u32 *val)
 {
 	if (ata_is_host_link(link)) {
-		struct ata_port *ap = link->ap;
-
 		if (sata_scr_valid(link))
-			return ap->ops->scr_read(ap, reg, val);
+			return link->ap->ops->scr_read(link, reg, val);
 		return -EOPNOTSUPP;
 	}
 
@@ -4897,10 +4895,8 @@ int sata_scr_read(struct ata_link *link, int reg, u32 *val)
 int sata_scr_write(struct ata_link *link, int reg, u32 val)
 {
 	if (ata_is_host_link(link)) {
-		struct ata_port *ap = link->ap;
-
 		if (sata_scr_valid(link))
-			return ap->ops->scr_write(ap, reg, val);
+			return link->ap->ops->scr_write(link, reg, val);
 		return -EOPNOTSUPP;
 	}
 
@@ -4925,13 +4921,12 @@ int sata_scr_write(struct ata_link *link, int reg, u32 val)
 int sata_scr_write_flush(struct ata_link *link, int reg, u32 val)
 {
 	if (ata_is_host_link(link)) {
-		struct ata_port *ap = link->ap;
 		int rc;
 
 		if (sata_scr_valid(link)) {
-			rc = ap->ops->scr_write(ap, reg, val);
+			rc = link->ap->ops->scr_write(link, reg, val);
 			if (rc == 0)
-				rc = ap->ops->scr_read(ap, reg, &val);
+				rc = link->ap->ops->scr_read(link, reg, &val);
 			return rc;
 		}
 		return -EOPNOTSUPP;
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c
index 3924e7209a44..1a56db92ff7a 100644
--- a/drivers/ata/sata_fsl.c
+++ b/drivers/ata/sata_fsl.c
@@ -469,10 +469,10 @@ static bool sata_fsl_qc_fill_rtf(struct ata_queued_cmd *qc)
 	return true;
 }
 
-static int sata_fsl_scr_write(struct ata_port *ap, unsigned int sc_reg_in,
-			       u32 val)
+static int sata_fsl_scr_write(struct ata_link *link,
+			      unsigned int sc_reg_in, u32 val)
 {
-	struct sata_fsl_host_priv *host_priv = ap->host->private_data;
+	struct sata_fsl_host_priv *host_priv = link->ap->host->private_data;
 	void __iomem *ssr_base = host_priv->ssr_base;
 	unsigned int sc_reg;
 
@@ -493,10 +493,10 @@ static int sata_fsl_scr_write(struct ata_port *ap, unsigned int sc_reg_in,
 	return 0;
 }
 
-static int sata_fsl_scr_read(struct ata_port *ap, unsigned int sc_reg_in,
-			u32 *val)
+static int sata_fsl_scr_read(struct ata_link *link,
+			     unsigned int sc_reg_in, u32 *val)
 {
-	struct sata_fsl_host_priv *host_priv = ap->host->private_data;
+	struct sata_fsl_host_priv *host_priv = link->ap->host->private_data;
 	void __iomem *ssr_base = host_priv->ssr_base;
 	unsigned int sc_reg;
 
@@ -645,12 +645,12 @@ static int sata_fsl_port_start(struct ata_port *ap)
 	 * Workaround for 8315DS board 3gbps link-up issue,
 	 * currently limit SATA port to GEN1 speed
 	 */
-	sata_fsl_scr_read(ap, SCR_CONTROL, &temp);
+	sata_fsl_scr_read(&ap->link, SCR_CONTROL, &temp);
 	temp &= ~(0xF << 4);
 	temp |= (0x1 << 4);
-	sata_fsl_scr_write(ap, SCR_CONTROL, temp);
+	sata_fsl_scr_write(&ap->link, SCR_CONTROL, temp);
 
-	sata_fsl_scr_read(ap, SCR_CONTROL, &temp);
+	sata_fsl_scr_read(&ap->link, SCR_CONTROL, &temp);
 	dev_printk(KERN_WARNING, dev, "scr_control, speed limited to %x\n",
 			temp);
 #endif
@@ -868,7 +868,7 @@ issue_srst:
 			ioread32(CQ + hcr_base),
 			ioread32(CA + hcr_base), ioread32(CC + hcr_base));
 
-		sata_fsl_scr_read(ap, SCR_ERROR, &Serror);
+		sata_fsl_scr_read(&ap->link, SCR_ERROR, &Serror);
 
 		DPRINTK("HStatus = 0x%x\n", ioread32(hcr_base + HSTATUS));
 		DPRINTK("HControl = 0x%x\n", ioread32(hcr_base + HCONTROL));
@@ -972,9 +972,9 @@ static void sata_fsl_error_intr(struct ata_port *ap)
 	 * Handle & Clear SError
 	 */
 
-	sata_fsl_scr_read(ap, SCR_ERROR, &SError);
+	sata_fsl_scr_read(&ap->link, SCR_ERROR, &SError);
 	if (unlikely(SError & 0xFFFF0000)) {
-		sata_fsl_scr_write(ap, SCR_ERROR, SError);
+		sata_fsl_scr_write(&ap->link, SCR_ERROR, SError);
 	}
 
 	DPRINTK("error_intr,hStat=0x%x,CE=0x%x,DE =0x%x,SErr=0x%x\n",
@@ -1091,7 +1091,7 @@ static void sata_fsl_host_intr(struct ata_port *ap)
 
 	hstatus = ioread32(hcr_base + HSTATUS);
 
-	sata_fsl_scr_read(ap, SCR_ERROR, &SError);
+	sata_fsl_scr_read(&ap->link, SCR_ERROR, &SError);
 
 	if (unlikely(SError & 0xFFFF0000)) {
 		DPRINTK("serror @host_intr : 0x%x\n", SError);
diff --git a/drivers/ata/sata_inic162x.c b/drivers/ata/sata_inic162x.c
index 5032c32fa505..fbbd87c96f10 100644
--- a/drivers/ata/sata_inic162x.c
+++ b/drivers/ata/sata_inic162x.c
@@ -269,9 +269,9 @@ static void inic_reset_port(void __iomem *port_base)
 	writeb(0xff, port_base + PORT_IRQ_STAT);
 }
 
-static int inic_scr_read(struct ata_port *ap, unsigned sc_reg, u32 *val)
+static int inic_scr_read(struct ata_link *link, unsigned sc_reg, u32 *val)
 {
-	void __iomem *scr_addr = inic_port_base(ap) + PORT_SCR;
+	void __iomem *scr_addr = inic_port_base(link->ap) + PORT_SCR;
 	void __iomem *addr;
 
 	if (unlikely(sc_reg >= ARRAY_SIZE(scr_map)))
@@ -286,9 +286,9 @@ static int inic_scr_read(struct ata_port *ap, unsigned sc_reg, u32 *val)
 	return 0;
 }
 
-static int inic_scr_write(struct ata_port *ap, unsigned sc_reg, u32 val)
+static int inic_scr_write(struct ata_link *link, unsigned sc_reg, u32 val)
 {
-	void __iomem *scr_addr = inic_port_base(ap) + PORT_SCR;
+	void __iomem *scr_addr = inic_port_base(link->ap) + PORT_SCR;
 
 	if (unlikely(sc_reg >= ARRAY_SIZE(scr_map)))
 		return -EINVAL;
diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
index c815f8ecf6e6..2b24ae58b52e 100644
--- a/drivers/ata/sata_mv.c
+++ b/drivers/ata/sata_mv.c
@@ -493,10 +493,10 @@ struct mv_hw_ops {
 	void (*reset_bus)(struct ata_host *host, void __iomem *mmio);
 };
 
-static int mv_scr_read(struct ata_port *ap, unsigned int sc_reg_in, u32 *val);
-static int mv_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val);
-static int mv5_scr_read(struct ata_port *ap, unsigned int sc_reg_in, u32 *val);
-static int mv5_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val);
+static int mv_scr_read(struct ata_link *link, unsigned int sc_reg_in, u32 *val);
+static int mv_scr_write(struct ata_link *link, unsigned int sc_reg_in, u32 val);
+static int mv5_scr_read(struct ata_link *link, unsigned int sc_reg_in, u32 *val);
+static int mv5_scr_write(struct ata_link *link, unsigned int sc_reg_in, u32 val);
 static int mv_port_start(struct ata_port *ap);
 static void mv_port_stop(struct ata_port *ap);
 static int mv_qc_defer(struct ata_queued_cmd *qc);
@@ -1070,23 +1070,23 @@ static unsigned int mv_scr_offset(unsigned int sc_reg_in)
 	return ofs;
 }
 
-static int mv_scr_read(struct ata_port *ap, unsigned int sc_reg_in, u32 *val)
+static int mv_scr_read(struct ata_link *link, unsigned int sc_reg_in, u32 *val)
 {
 	unsigned int ofs = mv_scr_offset(sc_reg_in);
 
 	if (ofs != 0xffffffffU) {
-		*val = readl(mv_ap_base(ap) + ofs);
+		*val = readl(mv_ap_base(link->ap) + ofs);
 		return 0;
 	} else
 		return -EINVAL;
 }
 
-static int mv_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val)
+static int mv_scr_write(struct ata_link *link, unsigned int sc_reg_in, u32 val)
 {
 	unsigned int ofs = mv_scr_offset(sc_reg_in);
 
 	if (ofs != 0xffffffffU) {
-		writelfl(val, mv_ap_base(ap) + ofs);
+		writelfl(val, mv_ap_base(link->ap) + ofs);
 		return 0;
 	} else
 		return -EINVAL;
@@ -2251,11 +2251,11 @@ static unsigned int mv5_scr_offset(unsigned int sc_reg_in)
 	return ofs;
 }
 
-static int mv5_scr_read(struct ata_port *ap, unsigned int sc_reg_in, u32 *val)
+static int mv5_scr_read(struct ata_link *link, unsigned int sc_reg_in, u32 *val)
 {
-	struct mv_host_priv *hpriv = ap->host->private_data;
+	struct mv_host_priv *hpriv = link->ap->host->private_data;
 	void __iomem *mmio = hpriv->base;
-	void __iomem *addr = mv5_phy_base(mmio, ap->port_no);
+	void __iomem *addr = mv5_phy_base(mmio, link->ap->port_no);
 	unsigned int ofs = mv5_scr_offset(sc_reg_in);
 
 	if (ofs != 0xffffffffU) {
@@ -2265,11 +2265,11 @@ static int mv5_scr_read(struct ata_port *ap, unsigned int sc_reg_in, u32 *val)
 		return -EINVAL;
 }
 
-static int mv5_scr_write(struct ata_port *ap, unsigned int sc_reg_in, u32 val)
+static int mv5_scr_write(struct ata_link *link, unsigned int sc_reg_in, u32 val)
 {
-	struct mv_host_priv *hpriv = ap->host->private_data;
+	struct mv_host_priv *hpriv = link->ap->host->private_data;
 	void __iomem *mmio = hpriv->base;
-	void __iomem *addr = mv5_phy_base(mmio, ap->port_no);
+	void __iomem *addr = mv5_phy_base(mmio, link->ap->port_no);
 	unsigned int ofs = mv5_scr_offset(sc_reg_in);
 
 	if (ofs != 0xffffffffU) {
diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index 1e1f3f3757ae..88fd4aeacde0 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -302,8 +302,8 @@ static void nv_ck804_host_stop(struct ata_host *host);
 static irqreturn_t nv_generic_interrupt(int irq, void *dev_instance);
 static irqreturn_t nv_nf2_interrupt(int irq, void *dev_instance);
 static irqreturn_t nv_ck804_interrupt(int irq, void *dev_instance);
-static int nv_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val);
-static int nv_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val);
+static int nv_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val);
+static int nv_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val);
 
 static void nv_nf2_freeze(struct ata_port *ap);
 static void nv_nf2_thaw(struct ata_port *ap);
@@ -1492,21 +1492,21 @@ static irqreturn_t nv_ck804_interrupt(int irq, void *dev_instance)
 	return ret;
 }
 
-static int nv_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
+static int nv_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
 
-	*val = ioread32(ap->ioaddr.scr_addr + (sc_reg * 4));
+	*val = ioread32(link->ap->ioaddr.scr_addr + (sc_reg * 4));
 	return 0;
 }
 
-static int nv_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
+static int nv_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
 
-	iowrite32(val, ap->ioaddr.scr_addr + (sc_reg * 4));
+	iowrite32(val, link->ap->ioaddr.scr_addr + (sc_reg * 4));
 	return 0;
 }
 
@@ -2184,9 +2184,9 @@ static void nv_swncq_host_interrupt(struct ata_port *ap, u16 fis)
 	if (!pp->qc_active)
 		return;
 
-	if (ap->ops->scr_read(ap, SCR_ERROR, &serror))
+	if (ap->ops->scr_read(&ap->link, SCR_ERROR, &serror))
 		return;
-	ap->ops->scr_write(ap, SCR_ERROR, serror);
+	ap->ops->scr_write(&ap->link, SCR_ERROR, serror);
 
 	if (ata_stat & ATA_ERR) {
 		ata_ehi_clear_desc(ehi);
diff --git a/drivers/ata/sata_promise.c b/drivers/ata/sata_promise.c
index 030665ba76b7..750d8cdc00cd 100644
--- a/drivers/ata/sata_promise.c
+++ b/drivers/ata/sata_promise.c
@@ -137,8 +137,8 @@ struct pdc_port_priv {
 	dma_addr_t		pkt_dma;
 };
 
-static int pdc_sata_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val);
-static int pdc_sata_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val);
+static int pdc_sata_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val);
+static int pdc_sata_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val);
 static int pdc_ata_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
 static int pdc_common_port_start(struct ata_port *ap);
 static int pdc_sata_port_start(struct ata_port *ap);
@@ -386,19 +386,21 @@ static int pdc_sata_cable_detect(struct ata_port *ap)
 	return ATA_CBL_SATA;
 }
 
-static int pdc_sata_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
+static int pdc_sata_scr_read(struct ata_link *link,
+			     unsigned int sc_reg, u32 *val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
-	*val = readl(ap->ioaddr.scr_addr + (sc_reg * 4));
+	*val = readl(link->ap->ioaddr.scr_addr + (sc_reg * 4));
 	return 0;
 }
 
-static int pdc_sata_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
+static int pdc_sata_scr_write(struct ata_link *link,
+			      unsigned int sc_reg, u32 val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
-	writel(val, ap->ioaddr.scr_addr + (sc_reg * 4));
+	writel(val, link->ap->ioaddr.scr_addr + (sc_reg * 4));
 	return 0;
 }
 
@@ -731,7 +733,7 @@ static void pdc_error_intr(struct ata_port *ap, struct ata_queued_cmd *qc,
 	if (sata_scr_valid(&ap->link)) {
 		u32 serror;
 
-		pdc_sata_scr_read(ap, SCR_ERROR, &serror);
+		pdc_sata_scr_read(&ap->link, SCR_ERROR, &serror);
 		ehi->serror |= serror;
 	}
 
diff --git a/drivers/ata/sata_qstor.c b/drivers/ata/sata_qstor.c
index 1600107047cf..a000c86ac859 100644
--- a/drivers/ata/sata_qstor.c
+++ b/drivers/ata/sata_qstor.c
@@ -111,8 +111,8 @@ struct qs_port_priv {
 	qs_state_t		state;
 };
 
-static int qs_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val);
-static int qs_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val);
+static int qs_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val);
+static int qs_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val);
 static int qs_ata_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
 static int qs_port_start(struct ata_port *ap);
 static void qs_host_stop(struct ata_host *host);
@@ -242,11 +242,11 @@ static int qs_prereset(struct ata_link *link, unsigned long deadline)
 	return ata_sff_prereset(link, deadline);
 }
 
-static int qs_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
+static int qs_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
-	*val = readl(ap->ioaddr.scr_addr + (sc_reg * 8));
+	*val = readl(link->ap->ioaddr.scr_addr + (sc_reg * 8));
 	return 0;
 }
 
@@ -256,11 +256,11 @@ static void qs_error_handler(struct ata_port *ap)
 	ata_std_error_handler(ap);
 }
 
-static int qs_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
+static int qs_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
-	writel(val, ap->ioaddr.scr_addr + (sc_reg * 8));
+	writel(val, link->ap->ioaddr.scr_addr + (sc_reg * 8));
 	return 0;
 }
 
diff --git a/drivers/ata/sata_sil.c b/drivers/ata/sata_sil.c
index 88bf4212590f..031d7b7dee34 100644
--- a/drivers/ata/sata_sil.c
+++ b/drivers/ata/sata_sil.c
@@ -115,8 +115,8 @@ static int sil_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
 static int sil_pci_device_resume(struct pci_dev *pdev);
 #endif
 static void sil_dev_config(struct ata_device *dev);
-static int sil_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val);
-static int sil_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val);
+static int sil_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val);
+static int sil_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val);
 static int sil_set_mode(struct ata_link *link, struct ata_device **r_failed);
 static void sil_freeze(struct ata_port *ap);
 static void sil_thaw(struct ata_port *ap);
@@ -317,9 +317,9 @@ static inline void __iomem *sil_scr_addr(struct ata_port *ap,
 	return NULL;
 }
 
-static int sil_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
+static int sil_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val)
 {
-	void __iomem *mmio = sil_scr_addr(ap, sc_reg);
+	void __iomem *mmio = sil_scr_addr(link->ap, sc_reg);
 
 	if (mmio) {
 		*val = readl(mmio);
@@ -328,9 +328,9 @@ static int sil_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
 	return -EINVAL;
 }
 
-static int sil_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
+static int sil_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val)
 {
-	void __iomem *mmio = sil_scr_addr(ap, sc_reg);
+	void __iomem *mmio = sil_scr_addr(link->ap, sc_reg);
 
 	if (mmio) {
 		writel(val, mmio);
@@ -352,8 +352,8 @@ static void sil_host_intr(struct ata_port *ap, u32 bmdma2)
 		 * controllers continue to assert IRQ as long as
 		 * SError bits are pending.  Clear SError immediately.
 		 */
-		sil_scr_read(ap, SCR_ERROR, &serror);
-		sil_scr_write(ap, SCR_ERROR, serror);
+		sil_scr_read(&ap->link, SCR_ERROR, &serror);
+		sil_scr_write(&ap->link, SCR_ERROR, serror);
 
 		/* Sometimes spurious interrupts occur, double check
 		 * it's PHYRDY CHG.
diff --git a/drivers/ata/sata_sil24.c b/drivers/ata/sata_sil24.c
index 84ffcc26a74b..4621807a1a6a 100644
--- a/drivers/ata/sata_sil24.c
+++ b/drivers/ata/sata_sil24.c
@@ -340,8 +340,8 @@ struct sil24_port_priv {
 };
 
 static void sil24_dev_config(struct ata_device *dev);
-static int sil24_scr_read(struct ata_port *ap, unsigned sc_reg, u32 *val);
-static int sil24_scr_write(struct ata_port *ap, unsigned sc_reg, u32 val);
+static int sil24_scr_read(struct ata_link *link, unsigned sc_reg, u32 *val);
+static int sil24_scr_write(struct ata_link *link, unsigned sc_reg, u32 val);
 static int sil24_qc_defer(struct ata_queued_cmd *qc);
 static void sil24_qc_prep(struct ata_queued_cmd *qc);
 static unsigned int sil24_qc_issue(struct ata_queued_cmd *qc);
@@ -504,9 +504,9 @@ static int sil24_scr_map[] = {
 	[SCR_ACTIVE]	= 3,
 };
 
-static int sil24_scr_read(struct ata_port *ap, unsigned sc_reg, u32 *val)
+static int sil24_scr_read(struct ata_link *link, unsigned sc_reg, u32 *val)
 {
-	void __iomem *scr_addr = sil24_port_base(ap) + PORT_SCONTROL;
+	void __iomem *scr_addr = sil24_port_base(link->ap) + PORT_SCONTROL;
 
 	if (sc_reg < ARRAY_SIZE(sil24_scr_map)) {
 		void __iomem *addr;
@@ -517,9 +517,9 @@ static int sil24_scr_read(struct ata_port *ap, unsigned sc_reg, u32 *val)
 	return -EINVAL;
 }
 
-static int sil24_scr_write(struct ata_port *ap, unsigned sc_reg, u32 val)
+static int sil24_scr_write(struct ata_link *link, unsigned sc_reg, u32 val)
 {
-	void __iomem *scr_addr = sil24_port_base(ap) + PORT_SCONTROL;
+	void __iomem *scr_addr = sil24_port_base(link->ap) + PORT_SCONTROL;
 
 	if (sc_reg < ARRAY_SIZE(sil24_scr_map)) {
 		void __iomem *addr;
diff --git a/drivers/ata/sata_sis.c b/drivers/ata/sata_sis.c
index 1010b3069bd5..9c43b4e7c4a6 100644
--- a/drivers/ata/sata_sis.c
+++ b/drivers/ata/sata_sis.c
@@ -64,8 +64,8 @@ enum {
 };
 
 static int sis_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
-static int sis_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val);
-static int sis_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val);
+static int sis_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val);
+static int sis_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val);
 
 static const struct pci_device_id sis_pci_tbl[] = {
 	{ PCI_VDEVICE(SI, 0x0180), sis_180 },	/* SiS 964/180 */
@@ -134,10 +134,11 @@ static unsigned int get_scr_cfg_addr(struct ata_port *ap, unsigned int sc_reg)
 	return addr;
 }
 
-static u32 sis_scr_cfg_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
+static u32 sis_scr_cfg_read(struct ata_link *link,
+			    unsigned int sc_reg, u32 *val)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
-	unsigned int cfg_addr = get_scr_cfg_addr(ap, sc_reg);
+	struct pci_dev *pdev = to_pci_dev(link->ap->host->dev);
+	unsigned int cfg_addr = get_scr_cfg_addr(link->ap, sc_reg);
 	u32 val2 = 0;
 	u8 pmr;
 
@@ -158,10 +159,11 @@ static u32 sis_scr_cfg_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
 	return 0;
 }
 
-static int sis_scr_cfg_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
+static int sis_scr_cfg_write(struct ata_link *link,
+			     unsigned int sc_reg, u32 val)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
-	unsigned int cfg_addr = get_scr_cfg_addr(ap, sc_reg);
+	struct pci_dev *pdev = to_pci_dev(link->ap->host->dev);
+	unsigned int cfg_addr = get_scr_cfg_addr(link->ap, sc_reg);
 	u8 pmr;
 
 	if (sc_reg == SCR_ERROR) /* doesn't exist in PCI cfg space */
@@ -178,8 +180,9 @@ static int sis_scr_cfg_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
 	return 0;
 }
 
-static int sis_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
+static int sis_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val)
 {
+	struct ata_port *ap = link->ap;
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	u8 pmr;
 
@@ -187,7 +190,7 @@ static int sis_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
 		return -EINVAL;
 
 	if (ap->flags & SIS_FLAG_CFGSCR)
-		return sis_scr_cfg_read(ap, sc_reg, val);
+		return sis_scr_cfg_read(link, sc_reg, val);
 
 	pci_read_config_byte(pdev, SIS_PMR, &pmr);
 
@@ -202,8 +205,9 @@ static int sis_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
 	return 0;
 }
 
-static int sis_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
+static int sis_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val)
 {
+	struct ata_port *ap = link->ap;
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	u8 pmr;
 
@@ -213,7 +217,7 @@ static int sis_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
 	pci_read_config_byte(pdev, SIS_PMR, &pmr);
 
 	if (ap->flags & SIS_FLAG_CFGSCR)
-		return sis_scr_cfg_write(ap, sc_reg, val);
+		return sis_scr_cfg_write(link, sc_reg, val);
 	else {
 		iowrite32(val, ap->ioaddr.scr_addr + (sc_reg * 4));
 		if ((pdev->device == 0x0182) || (pdev->device == 0x0183) ||
diff --git a/drivers/ata/sata_svw.c b/drivers/ata/sata_svw.c
index fb13b82aacba..609d147813ae 100644
--- a/drivers/ata/sata_svw.c
+++ b/drivers/ata/sata_svw.c
@@ -123,20 +123,22 @@ static int k2_sata_check_atapi_dma(struct ata_queued_cmd *qc)
 	}
 }
 
-static int k2_sata_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
+static int k2_sata_scr_read(struct ata_link *link,
+			    unsigned int sc_reg, u32 *val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
-	*val = readl(ap->ioaddr.scr_addr + (sc_reg * 4));
+	*val = readl(link->ap->ioaddr.scr_addr + (sc_reg * 4));
 	return 0;
 }
 
 
-static int k2_sata_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
+static int k2_sata_scr_write(struct ata_link *link,
+			     unsigned int sc_reg, u32 val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
-	writel(val, ap->ioaddr.scr_addr + (sc_reg * 4));
+	writel(val, link->ap->ioaddr.scr_addr + (sc_reg * 4));
 	return 0;
 }
 
diff --git a/drivers/ata/sata_uli.c b/drivers/ata/sata_uli.c
index db529b849948..019575bb3e08 100644
--- a/drivers/ata/sata_uli.c
+++ b/drivers/ata/sata_uli.c
@@ -57,8 +57,8 @@ struct uli_priv {
 };
 
 static int uli_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
-static int uli_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val);
-static int uli_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val);
+static int uli_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val);
+static int uli_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val);
 
 static const struct pci_device_id uli_pci_tbl[] = {
 	{ PCI_VDEVICE(AL, 0x5289), uli_5289 },
@@ -107,39 +107,39 @@ static unsigned int get_scr_cfg_addr(struct ata_port *ap, unsigned int sc_reg)
 	return hpriv->scr_cfg_addr[ap->port_no] + (4 * sc_reg);
 }
 
-static u32 uli_scr_cfg_read(struct ata_port *ap, unsigned int sc_reg)
+static u32 uli_scr_cfg_read(struct ata_link *link, unsigned int sc_reg)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
-	unsigned int cfg_addr = get_scr_cfg_addr(ap, sc_reg);
+	struct pci_dev *pdev = to_pci_dev(link->ap->host->dev);
+	unsigned int cfg_addr = get_scr_cfg_addr(link->ap, sc_reg);
 	u32 val;
 
 	pci_read_config_dword(pdev, cfg_addr, &val);
 	return val;
 }
 
-static void uli_scr_cfg_write(struct ata_port *ap, unsigned int scr, u32 val)
+static void uli_scr_cfg_write(struct ata_link *link, unsigned int scr, u32 val)
 {
-	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
-	unsigned int cfg_addr = get_scr_cfg_addr(ap, scr);
+	struct pci_dev *pdev = to_pci_dev(link->ap->host->dev);
+	unsigned int cfg_addr = get_scr_cfg_addr(link->ap, scr);
 
 	pci_write_config_dword(pdev, cfg_addr, val);
 }
 
-static int uli_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
+static int uli_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
 
-	*val = uli_scr_cfg_read(ap, sc_reg);
+	*val = uli_scr_cfg_read(link, sc_reg);
 	return 0;
 }
 
-static int uli_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
+static int uli_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val)
 {
 	if (sc_reg > SCR_CONTROL) //SCR_CONTROL=2, SCR_ERROR=1, SCR_STATUS=0
 		return -EINVAL;
 
-	uli_scr_cfg_write(ap, sc_reg, val);
+	uli_scr_cfg_write(link, sc_reg, val);
 	return 0;
 }
 
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index 96deeb354e16..1cfa74535d91 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -68,8 +68,8 @@ enum {
 };
 
 static int svia_init_one(struct pci_dev *pdev, const struct pci_device_id *ent);
-static int svia_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val);
-static int svia_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val);
+static int svia_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val);
+static int svia_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val);
 static void svia_noop_freeze(struct ata_port *ap);
 static int vt6420_prereset(struct ata_link *link, unsigned long deadline);
 static int vt6421_pata_cable_detect(struct ata_port *ap);
@@ -152,19 +152,19 @@ MODULE_LICENSE("GPL");
 MODULE_DEVICE_TABLE(pci, svia_pci_tbl);
 MODULE_VERSION(DRV_VERSION);
 
-static int svia_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
+static int svia_scr_read(struct ata_link *link, unsigned int sc_reg, u32 *val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
-	*val = ioread32(ap->ioaddr.scr_addr + (4 * sc_reg));
+	*val = ioread32(link->ap->ioaddr.scr_addr + (4 * sc_reg));
 	return 0;
 }
 
-static int svia_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
+static int svia_scr_write(struct ata_link *link, unsigned int sc_reg, u32 val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
-	iowrite32(val, ap->ioaddr.scr_addr + (4 * sc_reg));
+	iowrite32(val, link->ap->ioaddr.scr_addr + (4 * sc_reg));
 	return 0;
 }
 
@@ -210,20 +210,20 @@ static int vt6420_prereset(struct ata_link *link, unsigned long deadline)
 		goto skip_scr;
 
 	/* Resume phy.  This is the old SATA resume sequence */
-	svia_scr_write(ap, SCR_CONTROL, 0x300);
-	svia_scr_read(ap, SCR_CONTROL, &scontrol); /* flush */
+	svia_scr_write(link, SCR_CONTROL, 0x300);
+	svia_scr_read(link, SCR_CONTROL, &scontrol); /* flush */
 
 	/* wait for phy to become ready, if necessary */
 	do {
 		msleep(200);
-		svia_scr_read(ap, SCR_STATUS, &sstatus);
+		svia_scr_read(link, SCR_STATUS, &sstatus);
 		if ((sstatus & 0xf) != 1)
 			break;
 	} while (time_before(jiffies, timeout));
 
 	/* open code sata_print_link_status() */
-	svia_scr_read(ap, SCR_STATUS, &sstatus);
-	svia_scr_read(ap, SCR_CONTROL, &scontrol);
+	svia_scr_read(link, SCR_STATUS, &sstatus);
+	svia_scr_read(link, SCR_CONTROL, &scontrol);
 
 	online = (sstatus & 0xf) == 0x3;
 
@@ -232,7 +232,7 @@ static int vt6420_prereset(struct ata_link *link, unsigned long deadline)
 			online ? "up" : "down", sstatus, scontrol);
 
 	/* SStatus is read one more time */
-	svia_scr_read(ap, SCR_STATUS, &sstatus);
+	svia_scr_read(link, SCR_STATUS, &sstatus);
 
 	if (!online) {
 		/* tell EH to bail */
diff --git a/drivers/ata/sata_vsc.c b/drivers/ata/sata_vsc.c
index f3d635c0a2e9..c57cdff9e6bd 100644
--- a/drivers/ata/sata_vsc.c
+++ b/drivers/ata/sata_vsc.c
@@ -98,20 +98,22 @@ enum {
 			      VSC_SATA_INT_PHY_CHANGE),
 };
 
-static int vsc_sata_scr_read(struct ata_port *ap, unsigned int sc_reg, u32 *val)
+static int vsc_sata_scr_read(struct ata_link *link,
+			     unsigned int sc_reg, u32 *val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
-	*val = readl(ap->ioaddr.scr_addr + (sc_reg * 4));
+	*val = readl(link->ap->ioaddr.scr_addr + (sc_reg * 4));
 	return 0;
 }
 
 
-static int vsc_sata_scr_write(struct ata_port *ap, unsigned int sc_reg, u32 val)
+static int vsc_sata_scr_write(struct ata_link *link,
+			      unsigned int sc_reg, u32 val)
 {
 	if (sc_reg > SCR_CONTROL)
 		return -EINVAL;
-	writel(val, ap->ioaddr.scr_addr + (sc_reg * 4));
+	writel(val, link->ap->ioaddr.scr_addr + (sc_reg * 4));
 	return 0;
 }
 
diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c
index 48ee8c7f5bdd..e1872989710a 100644
--- a/drivers/scsi/libsas/sas_ata.c
+++ b/drivers/scsi/libsas/sas_ata.c
@@ -294,10 +294,10 @@ static void sas_ata_post_internal(struct ata_queued_cmd *qc)
 	}
 }
 
-static int sas_ata_scr_write(struct ata_port *ap, unsigned int sc_reg_in,
+static int sas_ata_scr_write(struct ata_link *link, unsigned int sc_reg_in,
 			      u32 val)
 {
-	struct domain_device *dev = ap->private_data;
+	struct domain_device *dev = link->ap->private_data;
 
 	SAS_DPRINTK("STUB %s\n", __func__);
 	switch (sc_reg_in) {
@@ -319,10 +319,10 @@ static int sas_ata_scr_write(struct ata_port *ap, unsigned int sc_reg_in,
 	return 0;
 }
 
-static int sas_ata_scr_read(struct ata_port *ap, unsigned int sc_reg_in,
+static int sas_ata_scr_read(struct ata_link *link, unsigned int sc_reg_in,
 			    u32 *val)
 {
-	struct domain_device *dev = ap->private_data;
+	struct domain_device *dev = link->ap->private_data;
 
 	SAS_DPRINTK("STUB %s\n", __func__);
 	switch (sc_reg_in) {
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 225bfc5bd9ec..ffd622fa319c 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -772,8 +772,8 @@ struct ata_port_operations {
 	/*
 	 * Optional features
 	 */
-	int  (*scr_read)(struct ata_port *ap, unsigned int sc_reg, u32 *val);
-	int  (*scr_write)(struct ata_port *ap, unsigned int sc_reg, u32 val);
+	int  (*scr_read)(struct ata_link *link, unsigned int sc_reg, u32 *val);
+	int  (*scr_write)(struct ata_link *link, unsigned int sc_reg, u32 val);
 	void (*pmp_attach)(struct ata_port *ap);
 	void (*pmp_detach)(struct ata_port *ap);
 	int  (*enable_pm)(struct ata_port *ap, enum link_pm policy);
-- 
cgit v1.3-7-g2ca7


From aadffb682cc5572f48cc24883681db65530bd284 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 31 Jul 2008 17:02:41 +0900
Subject: libata: reimplement link iterator

Implement __ata_port_next_link() and reimplement
__ata_port_for_each_link() and ata_port_for_each_link() using it.
This removes relatively large inlined code and makes iteration easier
to extend.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c | 30 ++++++++++++++++++++++++++++++
 include/linux/libata.h    | 33 ++++++++-------------------------
 2 files changed, 38 insertions(+), 25 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 825461a33abe..d156616f45f5 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -163,6 +163,35 @@ MODULE_LICENSE("GPL");
 MODULE_VERSION(DRV_VERSION);
 
 
+/*
+ * Iterator helpers.  Don't use directly.
+ *
+ * LOCKING:
+ * Host lock or EH context.
+ */
+struct ata_link *__ata_port_next_link(struct ata_port *ap,
+				      struct ata_link *link, bool dev_only)
+{
+	/* NULL link indicates start of iteration */
+	if (!link) {
+		if (dev_only && sata_pmp_attached(ap))
+			return ap->pmp_link;
+		return &ap->link;
+	}
+
+	/* we just iterated over the host link, what's next? */
+	if (ata_is_host_link(link)) {
+		if (!sata_pmp_attached(ap))
+			return NULL;
+		return ap->pmp_link;
+	}
+
+	/* iterate to the next PMP link */
+	if (++link < ap->pmp_link + ap->nr_pmp_links)
+		return link;
+	return NULL;
+}
+
 /**
  *	ata_force_cbl - force cable type according to libata.force
  *	@ap: ATA port of interest
@@ -6255,6 +6284,7 @@ EXPORT_SYMBOL_GPL(ata_base_port_ops);
 EXPORT_SYMBOL_GPL(sata_port_ops);
 EXPORT_SYMBOL_GPL(ata_dummy_port_ops);
 EXPORT_SYMBOL_GPL(ata_dummy_port_info);
+EXPORT_SYMBOL_GPL(__ata_port_next_link);
 EXPORT_SYMBOL_GPL(ata_std_bios_param);
 EXPORT_SYMBOL_GPL(ata_host_init);
 EXPORT_SYMBOL_GPL(ata_host_alloc);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index ffd622fa319c..3eaca347ce29 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1265,34 +1265,17 @@ static inline int ata_link_active(struct ata_link *link)
 	return ata_tag_valid(link->active_tag) || link->sactive;
 }
 
-static inline struct ata_link *ata_port_first_link(struct ata_port *ap)
-{
-	if (sata_pmp_attached(ap))
-		return ap->pmp_link;
-	return &ap->link;
-}
-
-static inline struct ata_link *ata_port_next_link(struct ata_link *link)
-{
-	struct ata_port *ap = link->ap;
-
-	if (ata_is_host_link(link)) {
-		if (!sata_pmp_attached(ap))
-			return NULL;
-		return ap->pmp_link;
-	}
-
-	if (++link < ap->nr_pmp_links + ap->pmp_link)
-		return link;
-	return NULL;
-}
+extern struct ata_link *__ata_port_next_link(struct ata_port *ap,
+					     struct ata_link *link,
+					     bool dev_only);
 
-#define __ata_port_for_each_link(lk, ap) \
-	for ((lk) = &(ap)->link; (lk); (lk) = ata_port_next_link(lk))
+#define __ata_port_for_each_link(link, ap) \
+	for ((link) = __ata_port_next_link((ap), NULL, false); (link); \
+	     (link) = __ata_port_next_link((ap), (link), false))
 
 #define ata_port_for_each_link(link, ap) \
-	for ((link) = ata_port_first_link(ap); (link); \
-	     (link) = ata_port_next_link(link))
+	for ((link) = __ata_port_next_link((ap), NULL, true); (link); \
+	     (link) = __ata_port_next_link((ap), (link), true))
 
 #define ata_link_for_each_dev(dev, link) \
 	for ((dev) = (link)->device; \
-- 
cgit v1.3-7-g2ca7


From b5b3fa386b8f96c7fa92e507e5deddc2637924b4 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 31 Jul 2008 17:02:42 +0900
Subject: libata: misc updates to prepare for slave link

* Add ATA_EH_ALL_ACTIONS.

* Make sata_link_{on|off}_line() return bool instead of int.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c | 16 ++++++++--------
 include/linux/libata.h    |  6 ++++--
 2 files changed, 12 insertions(+), 10 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index d156616f45f5..71024e94c576 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4976,16 +4976,16 @@ int sata_scr_write_flush(struct ata_link *link, int reg, u32 val)
  *	None.
  *
  *	RETURNS:
- *	1 if the port online status is available and online.
+ *	True if the port online status is available and online.
  */
-int ata_link_online(struct ata_link *link)
+bool ata_link_online(struct ata_link *link)
 {
 	u32 sstatus;
 
 	if (sata_scr_read(link, SCR_STATUS, &sstatus) == 0 &&
 	    (sstatus & 0xf) == 0x3)
-		return 1;
-	return 0;
+		return true;
+	return false;
 }
 
 /**
@@ -5000,16 +5000,16 @@ int ata_link_online(struct ata_link *link)
  *	None.
  *
  *	RETURNS:
- *	1 if the port offline status is available and offline.
+ *	True if the port offline status is available and offline.
  */
-int ata_link_offline(struct ata_link *link)
+bool ata_link_offline(struct ata_link *link)
 {
 	u32 sstatus;
 
 	if (sata_scr_read(link, SCR_STATUS, &sstatus) == 0 &&
 	    (sstatus & 0xf) != 0x3)
-		return 1;
-	return 0;
+		return true;
+	return false;
 }
 
 #ifdef CONFIG_PM
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 3eaca347ce29..0c7e6f3c28eb 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -321,6 +321,8 @@ enum {
 	ATA_EH_LPM		= (1 << 4),  /* link power management action */
 
 	ATA_EH_PERDEV_MASK	= ATA_EH_REVALIDATE,
+	ATA_EH_ALL_ACTIONS	= ATA_EH_REVALIDATE | ATA_EH_RESET |
+				  ATA_EH_ENABLE_LINK | ATA_EH_LPM,
 
 	/* ata_eh_info->flags */
 	ATA_EHI_HOTPLUGGED	= (1 << 0),  /* could have been hotplugged */
@@ -920,8 +922,8 @@ extern int sata_scr_valid(struct ata_link *link);
 extern int sata_scr_read(struct ata_link *link, int reg, u32 *val);
 extern int sata_scr_write(struct ata_link *link, int reg, u32 val);
 extern int sata_scr_write_flush(struct ata_link *link, int reg, u32 val);
-extern int ata_link_online(struct ata_link *link);
-extern int ata_link_offline(struct ata_link *link);
+extern bool ata_link_online(struct ata_link *link);
+extern bool ata_link_offline(struct ata_link *link);
 #ifdef CONFIG_PM
 extern int ata_host_suspend(struct ata_host *host, pm_message_t mesg);
 extern void ata_host_resume(struct ata_host *host);
-- 
cgit v1.3-7-g2ca7


From b1c72916abbdd0a55015c87358536ca0ebaf6735 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Thu, 31 Jul 2008 17:02:43 +0900
Subject: libata: implement slave_link

Explanation taken from the comment of ata_slave_link_init().

 In libata, a port contains links and a link contains devices.  There
 is single host link but if a PMP is attached to it, there can be
 multiple fan-out links.  On SATA, there's usually a single device
 connected to a link but PATA and SATA controllers emulating TF based
 interface can have two - master and slave.

 However, there are a few controllers which don't fit into this
 abstraction too well - SATA controllers which emulate TF interface
 with both master and slave devices but also have separate SCR
 register sets for each device.  These controllers need separate links
 for physical link handling (e.g. onlineness, link speed) but should
 be treated like a traditional M/S controller for everything else
 (e.g. command issue, softreset).

 slave_link is libata's way of handling this class of controllers
 without impacting core layer too much.  For anything other than
 physical link handling, the default host link is used for both master
 and slave.  For physical link handling, separate @ap->slave_link is
 used.  All dirty details are implemented inside libata core layer.
 From LLD's POV, the only difference is that prereset, hardreset and
 postreset are called once more for the slave link, so the reset
 sequence looks like the following.

 prereset(M) -> prereset(S) -> hardreset(M) -> hardreset(S) ->
 softreset(M) -> postreset(M) -> postreset(S)

 Note that softreset is called only for the master.  Softreset resets
 both M/S by definition, so SRST on master should handle both (the
 standard method will work just fine).

As slave_link excludes PMP support and only code paths which deal with
the attributes of physical link are affected, all the changes are
localized to libata.h, libata-core.c and libata-eh.c.

 * ata_is_host_link() updated so that slave_link is considered as host
   link too.

 * iterator extended to iterate over the slave_link when using the
   underbarred version.

 * force param handling updated such that devno 16 is mapped to the
   slave link/device.

 * ata_link_on/offline() updated to return the combined result from
   master and slave link.  ata_phys_link_on/offline() are the direct
   versions.

 * EH autopsy and report are performed separately for master slave
   links.  Reset is udpated to implement the above described reset
   sequence.

Except for reset update, most changes are minor, many of them just
modifying dev->link to ata_dev_phys_link(dev) or using phys online
test instead.

After this update, LLDs can take full advantage of per-dev SCR
registers by simply turning on slave link.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/libata-core.c | 202 ++++++++++++++++++++++++++++++++++++++++------
 drivers/ata/libata-eh.c   | 142 ++++++++++++++++++++++++++------
 drivers/ata/libata.h      |   3 +
 include/linux/libata.h    |   8 +-
 4 files changed, 303 insertions(+), 52 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 71024e94c576..6eed58e35e12 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -179,19 +179,51 @@ struct ata_link *__ata_port_next_link(struct ata_port *ap,
 		return &ap->link;
 	}
 
-	/* we just iterated over the host link, what's next? */
-	if (ata_is_host_link(link)) {
-		if (!sata_pmp_attached(ap))
+	/* we just iterated over the host master link, what's next? */
+	if (link == &ap->link) {
+		if (!sata_pmp_attached(ap)) {
+			if (unlikely(ap->slave_link) && !dev_only)
+				return ap->slave_link;
 			return NULL;
+		}
 		return ap->pmp_link;
 	}
 
+	/* slave_link excludes PMP */
+	if (unlikely(link == ap->slave_link))
+		return NULL;
+
 	/* iterate to the next PMP link */
 	if (++link < ap->pmp_link + ap->nr_pmp_links)
 		return link;
 	return NULL;
 }
 
+/**
+ *	ata_dev_phys_link - find physical link for a device
+ *	@dev: ATA device to look up physical link for
+ *
+ *	Look up physical link which @dev is attached to.  Note that
+ *	this is different from @dev->link only when @dev is on slave
+ *	link.  For all other cases, it's the same as @dev->link.
+ *
+ *	LOCKING:
+ *	Don't care.
+ *
+ *	RETURNS:
+ *	Pointer to the found physical link.
+ */
+struct ata_link *ata_dev_phys_link(struct ata_device *dev)
+{
+	struct ata_port *ap = dev->link->ap;
+
+	if (!ap->slave_link)
+		return dev->link;
+	if (!dev->devno)
+		return &ap->link;
+	return ap->slave_link;
+}
+
 /**
  *	ata_force_cbl - force cable type according to libata.force
  *	@ap: ATA port of interest
@@ -235,7 +267,8 @@ void ata_force_cbl(struct ata_port *ap)
  *	the host link and all fan-out ports connected via PMP.  If the
  *	device part is specified as 0 (e.g. 1.00:), it specifies the
  *	first fan-out link not the host link.  Device number 15 always
- *	points to the host link whether PMP is attached or not.
+ *	points to the host link whether PMP is attached or not.  If the
+ *	controller has slave link, device number 16 points to it.
  *
  *	LOCKING:
  *	EH context.
@@ -243,12 +276,11 @@ void ata_force_cbl(struct ata_port *ap)
 static void ata_force_link_limits(struct ata_link *link)
 {
 	bool did_spd = false;
-	int linkno, i;
+	int linkno = link->pmp;
+	int i;
 
 	if (ata_is_host_link(link))
-		linkno = 15;
-	else
-		linkno = link->pmp;
+		linkno += 15;
 
 	for (i = ata_force_tbl_size - 1; i >= 0; i--) {
 		const struct ata_force_ent *fe = &ata_force_tbl[i];
@@ -295,9 +327,9 @@ static void ata_force_xfermask(struct ata_device *dev)
 	int alt_devno = devno;
 	int i;
 
-	/* allow n.15 for the first device attached to host port */
-	if (ata_is_host_link(dev->link) && devno == 0)
-		alt_devno = 15;
+	/* allow n.15/16 for devices attached to host port */
+	if (ata_is_host_link(dev->link))
+		alt_devno += 15;
 
 	for (i = ata_force_tbl_size - 1; i >= 0; i--) {
 		const struct ata_force_ent *fe = &ata_force_tbl[i];
@@ -349,9 +381,9 @@ static void ata_force_horkage(struct ata_device *dev)
 	int alt_devno = devno;
 	int i;
 
-	/* allow n.15 for the first device attached to host port */
-	if (ata_is_host_link(dev->link) && devno == 0)
-		alt_devno = 15;
+	/* allow n.15/16 for devices attached to host port */
+	if (ata_is_host_link(dev->link))
+		alt_devno += 15;
 
 	for (i = 0; i < ata_force_tbl_size; i++) {
 		const struct ata_force_ent *fe = &ata_force_tbl[i];
@@ -2710,7 +2742,7 @@ static void sata_print_link_status(struct ata_link *link)
 		return;
 	sata_scr_read(link, SCR_CONTROL, &scontrol);
 
-	if (ata_link_online(link)) {
+	if (ata_phys_link_online(link)) {
 		tmp = (sstatus >> 4) & 0xf;
 		ata_link_printk(link, KERN_INFO,
 				"SATA link up %s (SStatus %X SControl %X)\n",
@@ -3401,6 +3433,12 @@ int ata_wait_ready(struct ata_link *link, unsigned long deadline,
 	unsigned long nodev_deadline = ata_deadline(start, ATA_TMOUT_FF_WAIT);
 	int warned = 0;
 
+	/* Slave readiness can't be tested separately from master.  On
+	 * M/S emulation configuration, this function should be called
+	 * only on the master and it will handle both master and slave.
+	 */
+	WARN_ON(link == link->ap->slave_link);
+
 	if (time_after(nodev_deadline, deadline))
 		nodev_deadline = deadline;
 
@@ -3622,7 +3660,7 @@ int ata_std_prereset(struct ata_link *link, unsigned long deadline)
 	}
 
 	/* no point in trying softreset on offline link */
-	if (ata_link_offline(link))
+	if (ata_phys_link_offline(link))
 		ehc->i.action &= ~ATA_EH_SOFTRESET;
 
 	return 0;
@@ -3700,7 +3738,7 @@ int sata_link_hardreset(struct ata_link *link, const unsigned long *timing,
 	if (rc)
 		goto out;
 	/* if link is offline nothing more to do */
-	if (ata_link_offline(link))
+	if (ata_phys_link_offline(link))
 		goto out;
 
 	/* Link is online.  From this point, -ENODEV too is an error. */
@@ -4965,7 +5003,7 @@ int sata_scr_write_flush(struct ata_link *link, int reg, u32 val)
 }
 
 /**
- *	ata_link_online - test whether the given link is online
+ *	ata_phys_link_online - test whether the given link is online
  *	@link: ATA link to test
  *
  *	Test whether @link is online.  Note that this function returns
@@ -4978,7 +5016,7 @@ int sata_scr_write_flush(struct ata_link *link, int reg, u32 val)
  *	RETURNS:
  *	True if the port online status is available and online.
  */
-bool ata_link_online(struct ata_link *link)
+bool ata_phys_link_online(struct ata_link *link)
 {
 	u32 sstatus;
 
@@ -4989,7 +5027,7 @@ bool ata_link_online(struct ata_link *link)
 }
 
 /**
- *	ata_link_offline - test whether the given link is offline
+ *	ata_phys_link_offline - test whether the given link is offline
  *	@link: ATA link to test
  *
  *	Test whether @link is offline.  Note that this function
@@ -5002,7 +5040,7 @@ bool ata_link_online(struct ata_link *link)
  *	RETURNS:
  *	True if the port offline status is available and offline.
  */
-bool ata_link_offline(struct ata_link *link)
+bool ata_phys_link_offline(struct ata_link *link)
 {
 	u32 sstatus;
 
@@ -5012,6 +5050,58 @@ bool ata_link_offline(struct ata_link *link)
 	return false;
 }
 
+/**
+ *	ata_link_online - test whether the given link is online
+ *	@link: ATA link to test
+ *
+ *	Test whether @link is online.  This is identical to
+ *	ata_phys_link_online() when there's no slave link.  When
+ *	there's a slave link, this function should only be called on
+ *	the master link and will return true if any of M/S links is
+ *	online.
+ *
+ *	LOCKING:
+ *	None.
+ *
+ *	RETURNS:
+ *	True if the port online status is available and online.
+ */
+bool ata_link_online(struct ata_link *link)
+{
+	struct ata_link *slave = link->ap->slave_link;
+
+	WARN_ON(link == slave);	/* shouldn't be called on slave link */
+
+	return ata_phys_link_online(link) ||
+		(slave && ata_phys_link_online(slave));
+}
+
+/**
+ *	ata_link_offline - test whether the given link is offline
+ *	@link: ATA link to test
+ *
+ *	Test whether @link is offline.  This is identical to
+ *	ata_phys_link_offline() when there's no slave link.  When
+ *	there's a slave link, this function should only be called on
+ *	the master link and will return true if both M/S links are
+ *	offline.
+ *
+ *	LOCKING:
+ *	None.
+ *
+ *	RETURNS:
+ *	True if the port offline status is available and offline.
+ */
+bool ata_link_offline(struct ata_link *link)
+{
+	struct ata_link *slave = link->ap->slave_link;
+
+	WARN_ON(link == slave);	/* shouldn't be called on slave link */
+
+	return ata_phys_link_offline(link) &&
+		(!slave || ata_phys_link_offline(slave));
+}
+
 #ifdef CONFIG_PM
 static int ata_host_request_pm(struct ata_host *host, pm_message_t mesg,
 			       unsigned int action, unsigned int ehi_flags,
@@ -5151,11 +5241,11 @@ int ata_port_start(struct ata_port *ap)
  */
 void ata_dev_init(struct ata_device *dev)
 {
-	struct ata_link *link = dev->link;
+	struct ata_link *link = ata_dev_phys_link(dev);
 	struct ata_port *ap = link->ap;
 	unsigned long flags;
 
-	/* SATA spd limit is bound to the first device */
+	/* SATA spd limit is bound to the attached device, reset together */
 	link->sata_spd_limit = link->hw_sata_spd_limit;
 	link->sata_spd = 0;
 
@@ -5318,6 +5408,7 @@ static void ata_host_release(struct device *gendev, void *res)
 			scsi_host_put(ap->scsi_host);
 
 		kfree(ap->pmp_link);
+		kfree(ap->slave_link);
 		kfree(ap);
 		host->ports[i] = NULL;
 	}
@@ -5438,6 +5529,68 @@ struct ata_host *ata_host_alloc_pinfo(struct device *dev,
 	return host;
 }
 
+/**
+ *	ata_slave_link_init - initialize slave link
+ *	@ap: port to initialize slave link for
+ *
+ *	Create and initialize slave link for @ap.  This enables slave
+ *	link handling on the port.
+ *
+ *	In libata, a port contains links and a link contains devices.
+ *	There is single host link but if a PMP is attached to it,
+ *	there can be multiple fan-out links.  On SATA, there's usually
+ *	a single device connected to a link but PATA and SATA
+ *	controllers emulating TF based interface can have two - master
+ *	and slave.
+ *
+ *	However, there are a few controllers which don't fit into this
+ *	abstraction too well - SATA controllers which emulate TF
+ *	interface with both master and slave devices but also have
+ *	separate SCR register sets for each device.  These controllers
+ *	need separate links for physical link handling
+ *	(e.g. onlineness, link speed) but should be treated like a
+ *	traditional M/S controller for everything else (e.g. command
+ *	issue, softreset).
+ *
+ *	slave_link is libata's way of handling this class of
+ *	controllers without impacting core layer too much.  For
+ *	anything other than physical link handling, the default host
+ *	link is used for both master and slave.  For physical link
+ *	handling, separate @ap->slave_link is used.  All dirty details
+ *	are implemented inside libata core layer.  From LLD's POV, the
+ *	only difference is that prereset, hardreset and postreset are
+ *	called once more for the slave link, so the reset sequence
+ *	looks like the following.
+ *
+ *	prereset(M) -> prereset(S) -> hardreset(M) -> hardreset(S) ->
+ *	softreset(M) -> postreset(M) -> postreset(S)
+ *
+ *	Note that softreset is called only for the master.  Softreset
+ *	resets both M/S by definition, so SRST on master should handle
+ *	both (the standard method will work just fine).
+ *
+ *	LOCKING:
+ *	Should be called before host is registered.
+ *
+ *	RETURNS:
+ *	0 on success, -errno on failure.
+ */
+int ata_slave_link_init(struct ata_port *ap)
+{
+	struct ata_link *link;
+
+	WARN_ON(ap->slave_link);
+	WARN_ON(ap->flags & ATA_FLAG_PMP);
+
+	link = kzalloc(sizeof(*link), GFP_KERNEL);
+	if (!link)
+		return -ENOMEM;
+
+	ata_link_init(ap, link, 1);
+	ap->slave_link = link;
+	return 0;
+}
+
 static void ata_host_stop(struct device *gendev, void *res)
 {
 	struct ata_host *host = dev_get_drvdata(gendev);
@@ -5664,6 +5817,8 @@ int ata_host_register(struct ata_host *host, struct scsi_host_template *sht)
 
 		/* init sata_spd_limit to the current value */
 		sata_link_init_spd(&ap->link);
+		if (ap->slave_link)
+			sata_link_init_spd(ap->slave_link);
 
 		/* print per-port info to dmesg */
 		xfer_mask = ata_pack_xfermask(ap->pio_mask, ap->mwdma_mask,
@@ -6289,6 +6444,7 @@ EXPORT_SYMBOL_GPL(ata_std_bios_param);
 EXPORT_SYMBOL_GPL(ata_host_init);
 EXPORT_SYMBOL_GPL(ata_host_alloc);
 EXPORT_SYMBOL_GPL(ata_host_alloc_pinfo);
+EXPORT_SYMBOL_GPL(ata_slave_link_init);
 EXPORT_SYMBOL_GPL(ata_host_start);
 EXPORT_SYMBOL_GPL(ata_host_register);
 EXPORT_SYMBOL_GPL(ata_host_activate);
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index c1db2f234d2e..99037a4860d9 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -1756,7 +1756,7 @@ static unsigned int ata_eh_speed_down_verdict(struct ata_device *dev)
 static unsigned int ata_eh_speed_down(struct ata_device *dev,
 				unsigned int eflags, unsigned int err_mask)
 {
-	struct ata_link *link = dev->link;
+	struct ata_link *link = ata_dev_phys_link(dev);
 	int xfer_ok = 0;
 	unsigned int verdict;
 	unsigned int action = 0;
@@ -1880,7 +1880,8 @@ static void ata_eh_link_autopsy(struct ata_link *link)
 	for (tag = 0; tag < ATA_MAX_QUEUE; tag++) {
 		struct ata_queued_cmd *qc = __ata_qc_from_tag(ap, tag);
 
-		if (!(qc->flags & ATA_QCFLAG_FAILED) || qc->dev->link != link)
+		if (!(qc->flags & ATA_QCFLAG_FAILED) ||
+		    ata_dev_phys_link(qc->dev) != link)
 			continue;
 
 		/* inherit upper level err_mask */
@@ -1967,6 +1968,23 @@ void ata_eh_autopsy(struct ata_port *ap)
 	ata_port_for_each_link(link, ap)
 		ata_eh_link_autopsy(link);
 
+	/* Handle the frigging slave link.  Autopsy is done similarly
+	 * but actions and flags are transferred over to the master
+	 * link and handled from there.
+	 */
+	if (ap->slave_link) {
+		struct ata_eh_context *mehc = &ap->link.eh_context;
+		struct ata_eh_context *sehc = &ap->slave_link->eh_context;
+
+		ata_eh_link_autopsy(ap->slave_link);
+
+		ata_eh_about_to_do(ap->slave_link, NULL, ATA_EH_ALL_ACTIONS);
+		mehc->i.action		|= sehc->i.action;
+		mehc->i.dev_action[1]	|= sehc->i.dev_action[1];
+		mehc->i.flags		|= sehc->i.flags;
+		ata_eh_done(ap->slave_link, NULL, ATA_EH_ALL_ACTIONS);
+	}
+
 	/* Autopsy of fanout ports can affect host link autopsy.
 	 * Perform host link autopsy last.
 	 */
@@ -2001,7 +2019,8 @@ static void ata_eh_link_report(struct ata_link *link)
 	for (tag = 0; tag < ATA_MAX_QUEUE; tag++) {
 		struct ata_queued_cmd *qc = __ata_qc_from_tag(ap, tag);
 
-		if (!(qc->flags & ATA_QCFLAG_FAILED) || qc->dev->link != link ||
+		if (!(qc->flags & ATA_QCFLAG_FAILED) ||
+		    ata_dev_phys_link(qc->dev) != link ||
 		    ((qc->flags & ATA_QCFLAG_QUIET) &&
 		     qc->err_mask == AC_ERR_DEV))
 			continue;
@@ -2068,7 +2087,7 @@ static void ata_eh_link_report(struct ata_link *link)
 		char cdb_buf[70] = "";
 
 		if (!(qc->flags & ATA_QCFLAG_FAILED) ||
-		    qc->dev->link != link || !qc->err_mask)
+		    ata_dev_phys_link(qc->dev) != link || !qc->err_mask)
 			continue;
 
 		if (qc->dma_dir != DMA_NONE) {
@@ -2160,12 +2179,14 @@ void ata_eh_report(struct ata_port *ap)
 }
 
 static int ata_do_reset(struct ata_link *link, ata_reset_fn_t reset,
-			unsigned int *classes, unsigned long deadline)
+			unsigned int *classes, unsigned long deadline,
+			bool clear_classes)
 {
 	struct ata_device *dev;
 
-	ata_link_for_each_dev(dev, link)
-		classes[dev->devno] = ATA_DEV_UNKNOWN;
+	if (clear_classes)
+		ata_link_for_each_dev(dev, link)
+			classes[dev->devno] = ATA_DEV_UNKNOWN;
 
 	return reset(link, classes, deadline);
 }
@@ -2187,17 +2208,20 @@ int ata_eh_reset(struct ata_link *link, int classify,
 		 ata_reset_fn_t hardreset, ata_postreset_fn_t postreset)
 {
 	struct ata_port *ap = link->ap;
+	struct ata_link *slave = ap->slave_link;
 	struct ata_eh_context *ehc = &link->eh_context;
+	struct ata_eh_context *sehc = &slave->eh_context;
 	unsigned int *classes = ehc->classes;
 	unsigned int lflags = link->flags;
 	int verbose = !(ehc->i.flags & ATA_EHI_QUIET);
 	int max_tries = 0, try = 0;
+	struct ata_link *failed_link;
 	struct ata_device *dev;
 	unsigned long deadline, now;
 	ata_reset_fn_t reset;
 	unsigned long flags;
 	u32 sstatus;
-	int nr_known, rc;
+	int nr_unknown, rc;
 
 	/*
 	 * Prepare to reset
@@ -2252,8 +2276,30 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	}
 
 	if (prereset) {
-		rc = prereset(link,
-			      ata_deadline(jiffies, ATA_EH_PRERESET_TIMEOUT));
+		unsigned long deadline = ata_deadline(jiffies,
+						      ATA_EH_PRERESET_TIMEOUT);
+
+		if (slave) {
+			sehc->i.action &= ~ATA_EH_RESET;
+			sehc->i.action |= ehc->i.action;
+		}
+
+		rc = prereset(link, deadline);
+
+		/* If present, do prereset on slave link too.  Reset
+		 * is skipped iff both master and slave links report
+		 * -ENOENT or clear ATA_EH_RESET.
+		 */
+		if (slave && (rc == 0 || rc == -ENOENT)) {
+			int tmp;
+
+			tmp = prereset(slave, deadline);
+			if (tmp != -ENOENT)
+				rc = tmp;
+
+			ehc->i.action |= sehc->i.action;
+		}
+
 		if (rc) {
 			if (rc == -ENOENT) {
 				ata_link_printk(link, KERN_DEBUG,
@@ -2302,25 +2348,51 @@ int ata_eh_reset(struct ata_link *link, int classify,
 		else
 			ehc->i.flags |= ATA_EHI_DID_SOFTRESET;
 
-		rc = ata_do_reset(link, reset, classes, deadline);
-		if (rc && rc != -EAGAIN)
+		rc = ata_do_reset(link, reset, classes, deadline, true);
+		if (rc && rc != -EAGAIN) {
+			failed_link = link;
 			goto fail;
+		}
+
+		/* hardreset slave link if existent */
+		if (slave && reset == hardreset) {
+			int tmp;
+
+			if (verbose)
+				ata_link_printk(slave, KERN_INFO,
+						"hard resetting link\n");
 
+			ata_eh_about_to_do(slave, NULL, ATA_EH_RESET);
+			tmp = ata_do_reset(slave, reset, classes, deadline,
+					   false);
+			switch (tmp) {
+			case -EAGAIN:
+				rc = -EAGAIN;
+			case 0:
+				break;
+			default:
+				failed_link = slave;
+				rc = tmp;
+				goto fail;
+			}
+		}
+
+		/* perform follow-up SRST if necessary */
 		if (reset == hardreset &&
 		    ata_eh_followup_srst_needed(link, rc, classes)) {
-			/* okay, let's do follow-up softreset */
 			reset = softreset;
 
 			if (!reset) {
 				ata_link_printk(link, KERN_ERR,
 						"follow-up softreset required "
 						"but no softreset avaliable\n");
+				failed_link = link;
 				rc = -EINVAL;
 				goto fail;
 			}
 
 			ata_eh_about_to_do(link, NULL, ATA_EH_RESET);
-			rc = ata_do_reset(link, reset, classes, deadline);
+			rc = ata_do_reset(link, reset, classes, deadline, true);
 		}
 	} else {
 		if (verbose)
@@ -2341,7 +2413,7 @@ int ata_eh_reset(struct ata_link *link, int classify,
 		dev->pio_mode = XFER_PIO_0;
 		dev->flags &= ~ATA_DFLAG_SLEEPING;
 
-		if (ata_link_offline(link))
+		if (ata_phys_link_offline(ata_dev_phys_link(dev)))
 			continue;
 
 		/* apply class override */
@@ -2354,6 +2426,8 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	/* record current link speed */
 	if (sata_scr_read(link, SCR_STATUS, &sstatus) == 0)
 		link->sata_spd = (sstatus >> 4) & 0xf;
+	if (slave && sata_scr_read(slave, SCR_STATUS, &sstatus) == 0)
+		slave->sata_spd = (sstatus >> 4) & 0xf;
 
 	/* thaw the port */
 	if (ata_is_host_link(link))
@@ -2366,12 +2440,17 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	 * reset and here.  This race is mediated by cross checking
 	 * link onlineness and classification result later.
 	 */
-	if (postreset)
+	if (postreset) {
 		postreset(link, classes);
+		if (slave)
+			postreset(slave, classes);
+	}
 
 	/* clear cached SError */
 	spin_lock_irqsave(link->ap->lock, flags);
 	link->eh_info.serror = 0;
+	if (slave)
+		slave->eh_info.serror = 0;
 	spin_unlock_irqrestore(link->ap->lock, flags);
 
 	/* Make sure onlineness and classification result correspond.
@@ -2381,19 +2460,21 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	 * link onlineness and classification result, those conditions
 	 * can be reliably detected and retried.
 	 */
-	nr_known = 0;
+	nr_unknown = 0;
 	ata_link_for_each_dev(dev, link) {
 		/* convert all ATA_DEV_UNKNOWN to ATA_DEV_NONE */
-		if (classes[dev->devno] == ATA_DEV_UNKNOWN)
+		if (classes[dev->devno] == ATA_DEV_UNKNOWN) {
 			classes[dev->devno] = ATA_DEV_NONE;
-		else
-			nr_known++;
+			if (ata_phys_link_online(ata_dev_phys_link(dev)))
+				nr_unknown++;
+		}
 	}
 
-	if (classify && !nr_known && ata_link_online(link)) {
+	if (classify && nr_unknown) {
 		if (try < max_tries) {
 			ata_link_printk(link, KERN_WARNING, "link online but "
 				       "device misclassified, retrying\n");
+			failed_link = link;
 			rc = -EAGAIN;
 			goto fail;
 		}
@@ -2404,6 +2485,8 @@ int ata_eh_reset(struct ata_link *link, int classify,
 
 	/* reset successful, schedule revalidation */
 	ata_eh_done(link, NULL, ATA_EH_RESET);
+	if (slave)
+		ata_eh_done(slave, NULL, ATA_EH_RESET);
 	ehc->last_reset = jiffies;
 	ehc->i.action |= ATA_EH_REVALIDATE;
 
@@ -2411,6 +2494,8 @@ int ata_eh_reset(struct ata_link *link, int classify,
  out:
 	/* clear hotplug flag */
 	ehc->i.flags &= ~ATA_EHI_HOTPLUGGED;
+	if (slave)
+		sehc->i.flags &= ~ATA_EHI_HOTPLUGGED;
 
 	spin_lock_irqsave(ap->lock, flags);
 	ap->pflags &= ~ATA_PFLAG_RESETTING;
@@ -2431,7 +2516,7 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	if (time_before(now, deadline)) {
 		unsigned long delta = deadline - now;
 
-		ata_link_printk(link, KERN_WARNING,
+		ata_link_printk(failed_link, KERN_WARNING,
 			"reset failed (errno=%d), retrying in %u secs\n",
 			rc, DIV_ROUND_UP(jiffies_to_msecs(delta), 1000));
 
@@ -2439,8 +2524,13 @@ int ata_eh_reset(struct ata_link *link, int classify,
 			delta = schedule_timeout_uninterruptible(delta);
 	}
 
-	if (rc == -EPIPE || try == max_tries - 1)
+	if (try == max_tries - 1) {
 		sata_down_spd_limit(link);
+		if (slave)
+			sata_down_spd_limit(slave);
+	} else if (rc == -EPIPE)
+		sata_down_spd_limit(failed_link);
+
 	if (hardreset)
 		reset = hardreset;
 	goto retry;
@@ -2472,7 +2562,7 @@ static int ata_eh_revalidate_and_attach(struct ata_link *link,
 		if ((action & ATA_EH_REVALIDATE) && ata_dev_enabled(dev)) {
 			WARN_ON(dev->class == ATA_DEV_PMP);
 
-			if (ata_link_offline(link)) {
+			if (ata_phys_link_offline(ata_dev_phys_link(dev))) {
 				rc = -EIO;
 				goto err;
 			}
@@ -2697,7 +2787,7 @@ static int ata_eh_handle_dev_fail(struct ata_device *dev, int err)
 			/* This is the last chance, better to slow
 			 * down than lose it.
 			 */
-			sata_down_spd_limit(dev->link);
+			sata_down_spd_limit(ata_dev_phys_link(dev));
 			ata_down_xfermask_limit(dev, ATA_DNXFER_PIO);
 		}
 	}
@@ -2707,7 +2797,7 @@ static int ata_eh_handle_dev_fail(struct ata_device *dev, int err)
 		ata_dev_disable(dev);
 
 		/* detach if offline */
-		if (ata_link_offline(dev->link))
+		if (ata_phys_link_offline(ata_dev_phys_link(dev)))
 			ata_eh_detach_dev(dev);
 
 		/* schedule probe if necessary */
diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h
index ade5c75b6144..e96de96e3020 100644
--- a/drivers/ata/libata.h
+++ b/drivers/ata/libata.h
@@ -70,6 +70,7 @@ extern int atapi_passthru16;
 extern int libata_fua;
 extern int libata_noacpi;
 extern int libata_allow_tpm;
+extern struct ata_link *ata_dev_phys_link(struct ata_device *dev);
 extern void ata_force_cbl(struct ata_port *ap);
 extern u64 ata_tf_to_lba(const struct ata_taskfile *tf);
 extern u64 ata_tf_to_lba48(const struct ata_taskfile *tf);
@@ -107,6 +108,8 @@ extern void ata_qc_issue(struct ata_queued_cmd *qc);
 extern void __ata_qc_complete(struct ata_queued_cmd *qc);
 extern int atapi_check_dma(struct ata_queued_cmd *qc);
 extern void swap_buf_le16(u16 *buf, unsigned int buf_words);
+extern bool ata_phys_link_online(struct ata_link *link);
+extern bool ata_phys_link_offline(struct ata_link *link);
 extern void ata_dev_init(struct ata_device *dev);
 extern void ata_link_init(struct ata_port *ap, struct ata_link *link, int pmp);
 extern int sata_link_init_spd(struct ata_link *link);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 0c7e6f3c28eb..244ff601559a 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -690,7 +690,8 @@ struct ata_port {
 	unsigned int		qc_active;
 	int			nr_active_links; /* #links with active qcs */
 
-	struct ata_link		link;	/* host default link */
+	struct ata_link		link;		/* host default link */
+	struct ata_link		*slave_link;	/* see ata_slave_link_init() */
 
 	int			nr_pmp_links;	/* nr of available PMP links */
 	struct ata_link		*pmp_link;	/* array of PMP links */
@@ -897,6 +898,7 @@ extern void ata_port_disable(struct ata_port *);
 extern struct ata_host *ata_host_alloc(struct device *dev, int max_ports);
 extern struct ata_host *ata_host_alloc_pinfo(struct device *dev,
 			const struct ata_port_info * const * ppi, int n_ports);
+extern int ata_slave_link_init(struct ata_port *ap);
 extern int ata_host_start(struct ata_host *host);
 extern int ata_host_register(struct ata_host *host,
 			     struct scsi_host_template *sht);
@@ -1136,7 +1138,7 @@ static inline bool sata_pmp_attached(struct ata_port *ap)
 
 static inline int ata_is_host_link(const struct ata_link *link)
 {
-	return link == &link->ap->link;
+	return link == &link->ap->link || link == link->ap->slave_link;
 }
 #else /* CONFIG_SATA_PMP */
 static inline bool sata_pmp_supported(struct ata_port *ap)
@@ -1169,7 +1171,7 @@ static inline int sata_srst_pmp(struct ata_link *link)
 	printk("%sata%u: "fmt, lv, (ap)->print_id , ##args)
 
 #define ata_link_printk(link, lv, fmt, args...) do { \
-	if (sata_pmp_attached((link)->ap)) \
+	if (sata_pmp_attached((link)->ap) || (link)->ap->slave_link)	\
 		printk("%sata%u.%02u: "fmt, lv, (link)->ap->print_id,	\
 		       (link)->pmp , ##args); \
 	else \
-- 
cgit v1.3-7-g2ca7


From ea6ce53cd5d005455ec0a3cc1d45d3af0cb90919 Mon Sep 17 00:00:00 2001
From: Elias Oltmanns <eo@nebensachen.de>
Date: Fri, 19 Sep 2008 23:46:01 +0200
Subject: [libata] Introduce ata_id_has_unload()

Add a function to check an ATA device's id for head unload support as
specified in ATA-7.

Signed-off-by: Elias Oltmanns <eo@nebensachen.de>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/ata.h | 9 +++++++++
 1 file changed, 9 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/ata.h b/include/linux/ata.h
index 8a12d718c169..a26ebd25bac1 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -667,6 +667,15 @@ static inline int ata_id_has_dword_io(const u16 *id)
 	return 0;
 }
 
+static inline int ata_id_has_unload(const u16 *id)
+{
+	if (ata_id_major_version(id) >= 7 &&
+	    (id[ATA_ID_CFSSE] & 0xC000) == 0x4000 &&
+	    id[ATA_ID_CFSSE] & (1 << 13))
+		return 1;
+	return 0;
+}
+
 static inline int ata_id_current_chs_valid(const u16 *id)
 {
 	/* For ATA-1 devices, if the INITIALIZE DEVICE PARAMETERS command
-- 
cgit v1.3-7-g2ca7


From 45fabbb77bd95adff7a80bde1c7a0ace1075fde6 Mon Sep 17 00:00:00 2001
From: Elias Oltmanns <eo@nebensachen.de>
Date: Sun, 21 Sep 2008 11:54:08 +0200
Subject: libata: Implement disk shock protection support

On user request (through sysfs), the IDLE IMMEDIATE command with UNLOAD
FEATURE as specified in ATA-7 is issued to the device and processing of
the request queue is stopped thereafter until the specified timeout
expires or user space asks to resume normal operation. This is supposed
to prevent the heads of a hard drive from accidentally crashing onto the
platter when a heavy shock is anticipated (like a falling laptop
expected to hit the floor). In fact, the whole port stops processing
commands until the timeout has expired in order to avoid any resets due
to failed commands on another device.

Signed-off-by: Elias Oltmanns <eo@nebensachen.de>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 drivers/ata/ahci.c        |   1 +
 drivers/ata/libata-core.c |   1 +
 drivers/ata/libata-eh.c   | 126 +++++++++++++++++++++++++++++++++++++++++++++-
 drivers/ata/libata-scsi.c | 108 +++++++++++++++++++++++++++++++++++++++
 include/linux/libata.h    |  13 ++++-
 5 files changed, 246 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 6acea41eb7ca..aeadd00411a1 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -316,6 +316,7 @@ static struct device_attribute *ahci_shost_attrs[] = {
 
 static struct device_attribute *ahci_sdev_attrs[] = {
 	&dev_attr_sw_activity,
+	&dev_attr_unload_heads,
 	NULL
 };
 
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 6eed58e35e12..1ee9499bd343 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -5378,6 +5378,7 @@ struct ata_port *ata_port_alloc(struct ata_host *host)
 	INIT_WORK(&ap->scsi_rescan_task, ata_scsi_dev_rescan);
 	INIT_LIST_HEAD(&ap->eh_done_q);
 	init_waitqueue_head(&ap->eh_wait_q);
+	init_completion(&ap->park_req_pending);
 	init_timer_deferrable(&ap->fastdrain_timer);
 	ap->fastdrain_timer.function = ata_eh_fastdrain_timerfn;
 	ap->fastdrain_timer.data = (unsigned long)ap;
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 99037a4860d9..33ac5ea4f531 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -2536,6 +2536,80 @@ int ata_eh_reset(struct ata_link *link, int classify,
 	goto retry;
 }
 
+static inline void ata_eh_pull_park_action(struct ata_port *ap)
+{
+	struct ata_link *link;
+	struct ata_device *dev;
+	unsigned long flags;
+
+	/*
+	 * This function can be thought of as an extended version of
+	 * ata_eh_about_to_do() specially crafted to accommodate the
+	 * requirements of ATA_EH_PARK handling. Since the EH thread
+	 * does not leave the do {} while () loop in ata_eh_recover as
+	 * long as the timeout for a park request to *one* device on
+	 * the port has not expired, and since we still want to pick
+	 * up park requests to other devices on the same port or
+	 * timeout updates for the same device, we have to pull
+	 * ATA_EH_PARK actions from eh_info into eh_context.i
+	 * ourselves at the beginning of each pass over the loop.
+	 *
+	 * Additionally, all write accesses to &ap->park_req_pending
+	 * through INIT_COMPLETION() (see below) or complete_all()
+	 * (see ata_scsi_park_store()) are protected by the host lock.
+	 * As a result we have that park_req_pending.done is zero on
+	 * exit from this function, i.e. when ATA_EH_PARK actions for
+	 * *all* devices on port ap have been pulled into the
+	 * respective eh_context structs. If, and only if,
+	 * park_req_pending.done is non-zero by the time we reach
+	 * wait_for_completion_timeout(), another ATA_EH_PARK action
+	 * has been scheduled for at least one of the devices on port
+	 * ap and we have to cycle over the do {} while () loop in
+	 * ata_eh_recover() again.
+	 */
+
+	spin_lock_irqsave(ap->lock, flags);
+	INIT_COMPLETION(ap->park_req_pending);
+	ata_port_for_each_link(link, ap) {
+		ata_link_for_each_dev(dev, link) {
+			struct ata_eh_info *ehi = &link->eh_info;
+
+			link->eh_context.i.dev_action[dev->devno] |=
+				ehi->dev_action[dev->devno] & ATA_EH_PARK;
+			ata_eh_clear_action(link, dev, ehi, ATA_EH_PARK);
+		}
+	}
+	spin_unlock_irqrestore(ap->lock, flags);
+}
+
+static void ata_eh_park_issue_cmd(struct ata_device *dev, int park)
+{
+	struct ata_eh_context *ehc = &dev->link->eh_context;
+	struct ata_taskfile tf;
+	unsigned int err_mask;
+
+	ata_tf_init(dev, &tf);
+	if (park) {
+		ehc->unloaded_mask |= 1 << dev->devno;
+		tf.command = ATA_CMD_IDLEIMMEDIATE;
+		tf.feature = 0x44;
+		tf.lbal = 0x4c;
+		tf.lbam = 0x4e;
+		tf.lbah = 0x55;
+	} else {
+		ehc->unloaded_mask &= ~(1 << dev->devno);
+		tf.command = ATA_CMD_CHK_POWER;
+	}
+
+	tf.flags |= ATA_TFLAG_DEVICE | ATA_TFLAG_ISADDR;
+	tf.protocol |= ATA_PROT_NODATA;
+	err_mask = ata_exec_internal(dev, &tf, NULL, DMA_NONE, NULL, 0, 0);
+	if (park && (err_mask || tf.lbal != 0xc4)) {
+		ata_dev_printk(dev, KERN_ERR, "head unload failed!\n");
+		ehc->unloaded_mask &= ~(1 << dev->devno);
+	}
+}
+
 static int ata_eh_revalidate_and_attach(struct ata_link *link,
 					struct ata_device **r_failed_dev)
 {
@@ -2845,7 +2919,7 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 	struct ata_device *dev;
 	int nr_failed_devs;
 	int rc;
-	unsigned long flags;
+	unsigned long flags, deadline;
 
 	DPRINTK("ENTER\n");
 
@@ -2919,6 +2993,56 @@ int ata_eh_recover(struct ata_port *ap, ata_prereset_fn_t prereset,
 		}
 	}
 
+	do {
+		unsigned long now;
+
+		/*
+		 * clears ATA_EH_PARK in eh_info and resets
+		 * ap->park_req_pending
+		 */
+		ata_eh_pull_park_action(ap);
+
+		deadline = jiffies;
+		ata_port_for_each_link(link, ap) {
+			ata_link_for_each_dev(dev, link) {
+				struct ata_eh_context *ehc = &link->eh_context;
+				unsigned long tmp;
+
+				if (dev->class != ATA_DEV_ATA)
+					continue;
+				if (!(ehc->i.dev_action[dev->devno] &
+				      ATA_EH_PARK))
+					continue;
+				tmp = dev->unpark_deadline;
+				if (time_before(deadline, tmp))
+					deadline = tmp;
+				else if (time_before_eq(tmp, jiffies))
+					continue;
+				if (ehc->unloaded_mask & (1 << dev->devno))
+					continue;
+
+				ata_eh_park_issue_cmd(dev, 1);
+			}
+		}
+
+		now = jiffies;
+		if (time_before_eq(deadline, now))
+			break;
+
+		deadline = wait_for_completion_timeout(&ap->park_req_pending,
+						       deadline - now);
+	} while (deadline);
+	ata_port_for_each_link(link, ap) {
+		ata_link_for_each_dev(dev, link) {
+			if (!(link->eh_context.unloaded_mask &
+			      (1 << dev->devno)))
+				continue;
+
+			ata_eh_park_issue_cmd(dev, 0);
+			ata_eh_done(link, dev, ATA_EH_PARK);
+		}
+	}
+
 	/* the rest */
 	ata_port_for_each_link(link, ap) {
 		struct ata_eh_context *ehc = &link->eh_context;
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index b9d3ba423cb2..fccd5e496c62 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -183,6 +183,105 @@ DEVICE_ATTR(link_power_management_policy, S_IRUGO | S_IWUSR,
 		ata_scsi_lpm_show, ata_scsi_lpm_put);
 EXPORT_SYMBOL_GPL(dev_attr_link_power_management_policy);
 
+static ssize_t ata_scsi_park_show(struct device *device,
+				  struct device_attribute *attr, char *buf)
+{
+	struct scsi_device *sdev = to_scsi_device(device);
+	struct ata_port *ap;
+	struct ata_link *link;
+	struct ata_device *dev;
+	unsigned long flags;
+	unsigned int uninitialized_var(msecs);
+	int rc = 0;
+
+	ap = ata_shost_to_port(sdev->host);
+
+	spin_lock_irqsave(ap->lock, flags);
+	dev = ata_scsi_find_dev(ap, sdev);
+	if (!dev) {
+		rc = -ENODEV;
+		goto unlock;
+	}
+	if (dev->flags & ATA_DFLAG_NO_UNLOAD) {
+		rc = -EOPNOTSUPP;
+		goto unlock;
+	}
+
+	link = dev->link;
+	if (ap->pflags & ATA_PFLAG_EH_IN_PROGRESS &&
+	    link->eh_context.unloaded_mask & (1 << dev->devno) &&
+	    time_after(dev->unpark_deadline, jiffies))
+		msecs = jiffies_to_msecs(dev->unpark_deadline - jiffies);
+	else
+		msecs = 0;
+
+unlock:
+	spin_unlock_irq(ap->lock);
+
+	return rc ? rc : snprintf(buf, 20, "%u\n", msecs);
+}
+
+static ssize_t ata_scsi_park_store(struct device *device,
+				   struct device_attribute *attr,
+				   const char *buf, size_t len)
+{
+	struct scsi_device *sdev = to_scsi_device(device);
+	struct ata_port *ap;
+	struct ata_device *dev;
+	long int input;
+	unsigned long flags;
+	int rc;
+
+	rc = strict_strtol(buf, 10, &input);
+	if (rc || input < -2)
+		return -EINVAL;
+	if (input > ATA_TMOUT_MAX_PARK) {
+		rc = -EOVERFLOW;
+		input = ATA_TMOUT_MAX_PARK;
+	}
+
+	ap = ata_shost_to_port(sdev->host);
+
+	spin_lock_irqsave(ap->lock, flags);
+	dev = ata_scsi_find_dev(ap, sdev);
+	if (unlikely(!dev)) {
+		rc = -ENODEV;
+		goto unlock;
+	}
+	if (dev->class != ATA_DEV_ATA) {
+		rc = -EOPNOTSUPP;
+		goto unlock;
+	}
+
+	if (input >= 0) {
+		if (dev->flags & ATA_DFLAG_NO_UNLOAD) {
+			rc = -EOPNOTSUPP;
+			goto unlock;
+		}
+
+		dev->unpark_deadline = ata_deadline(jiffies, input);
+		dev->link->eh_info.dev_action[dev->devno] |= ATA_EH_PARK;
+		ata_port_schedule_eh(ap);
+		complete(&ap->park_req_pending);
+	} else {
+		switch (input) {
+		case -1:
+			dev->flags &= ~ATA_DFLAG_NO_UNLOAD;
+			break;
+		case -2:
+			dev->flags |= ATA_DFLAG_NO_UNLOAD;
+			break;
+		}
+	}
+unlock:
+	spin_unlock_irqrestore(ap->lock, flags);
+
+	return rc ? rc : len;
+}
+DEVICE_ATTR(unload_heads, S_IRUGO | S_IWUSR,
+	    ata_scsi_park_show, ata_scsi_park_store);
+EXPORT_SYMBOL_GPL(dev_attr_unload_heads);
+
 static void ata_scsi_set_sense(struct scsi_cmnd *cmd, u8 sk, u8 asc, u8 ascq)
 {
 	cmd->result = (DRIVER_SENSE << 24) | SAM_STAT_CHECK_CONDITION;
@@ -269,6 +368,12 @@ DEVICE_ATTR(sw_activity, S_IWUGO | S_IRUGO, ata_scsi_activity_show,
 			ata_scsi_activity_store);
 EXPORT_SYMBOL_GPL(dev_attr_sw_activity);
 
+struct device_attribute *ata_common_sdev_attrs[] = {
+	&dev_attr_unload_heads,
+	NULL
+};
+EXPORT_SYMBOL_GPL(ata_common_sdev_attrs);
+
 static void ata_scsi_invalid_field(struct scsi_cmnd *cmd,
 				   void (*done)(struct scsi_cmnd *))
 {
@@ -954,6 +1059,9 @@ static int atapi_drain_needed(struct request *rq)
 static int ata_scsi_dev_config(struct scsi_device *sdev,
 			       struct ata_device *dev)
 {
+	if (!ata_id_has_unload(dev->id))
+		dev->flags |= ATA_DFLAG_NO_UNLOAD;
+
 	/* configure max sectors */
 	blk_queue_max_sectors(sdev->request_queue, dev->max_sectors);
 
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 244ff601559a..1f44cfb847e1 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -146,6 +146,7 @@ enum {
 	ATA_DFLAG_SPUNDOWN	= (1 << 14), /* XXX: for spindown_compat */
 	ATA_DFLAG_SLEEPING	= (1 << 15), /* device is sleeping */
 	ATA_DFLAG_DUBIOUS_XFER	= (1 << 16), /* data transfer not verified */
+	ATA_DFLAG_NO_UNLOAD	= (1 << 17), /* device doesn't support unload */
 	ATA_DFLAG_INIT_MASK	= (1 << 24) - 1,
 
 	ATA_DFLAG_DETACH	= (1 << 24),
@@ -244,6 +245,7 @@ enum {
 	ATA_TMOUT_BOOT		= 30000,	/* heuristic */
 	ATA_TMOUT_BOOT_QUICK	=  7000,	/* heuristic */
 	ATA_TMOUT_INTERNAL_QUICK = 5000,
+	ATA_TMOUT_MAX_PARK	= 30000,
 
 	/* FIXME: GoVault needs 2s but we can't afford that without
 	 * parallel probing.  800ms is enough for iVDR disk
@@ -319,8 +321,9 @@ enum {
 	ATA_EH_RESET		= ATA_EH_SOFTRESET | ATA_EH_HARDRESET,
 	ATA_EH_ENABLE_LINK	= (1 << 3),
 	ATA_EH_LPM		= (1 << 4),  /* link power management action */
+	ATA_EH_PARK		= (1 << 5), /* unload heads and stop I/O */
 
-	ATA_EH_PERDEV_MASK	= ATA_EH_REVALIDATE,
+	ATA_EH_PERDEV_MASK	= ATA_EH_REVALIDATE | ATA_EH_PARK,
 	ATA_EH_ALL_ACTIONS	= ATA_EH_REVALIDATE | ATA_EH_RESET |
 				  ATA_EH_ENABLE_LINK | ATA_EH_LPM,
 
@@ -454,6 +457,7 @@ enum link_pm {
 	MEDIUM_POWER,
 };
 extern struct device_attribute dev_attr_link_power_management_policy;
+extern struct device_attribute dev_attr_unload_heads;
 extern struct device_attribute dev_attr_em_message_type;
 extern struct device_attribute dev_attr_em_message;
 extern struct device_attribute dev_attr_sw_activity;
@@ -566,6 +570,7 @@ struct ata_device {
 	/* n_sector is used as CLEAR_OFFSET, read comment above CLEAR_OFFSET */
 	u64			n_sectors;	/* size of device, if ATA */
 	unsigned int		class;		/* ATA_DEV_xxx */
+	unsigned long		unpark_deadline;
 
 	u8			pio_mode;
 	u8			dma_mode;
@@ -623,6 +628,7 @@ struct ata_eh_context {
 					       [ATA_EH_CMD_TIMEOUT_TABLE_SIZE];
 	unsigned int		classes[ATA_MAX_DEVICES];
 	unsigned int		did_probe_mask;
+	unsigned int		unloaded_mask;
 	unsigned int		saved_ncq_enabled;
 	u8			saved_xfer_mode[ATA_MAX_DEVICES];
 	/* timestamp for the last reset attempt or success */
@@ -712,6 +718,7 @@ struct ata_port {
 	struct list_head	eh_done_q;
 	wait_queue_head_t	eh_wait_q;
 	int			eh_tries;
+	struct completion	park_req_pending;
 
 	pm_message_t		pm_mesg;
 	int			*pm_result;
@@ -1102,6 +1109,7 @@ extern void ata_std_error_handler(struct ata_port *ap);
  */
 extern const struct ata_port_operations ata_base_port_ops;
 extern const struct ata_port_operations sata_port_ops;
+extern struct device_attribute *ata_common_sdev_attrs[];
 
 #define ATA_BASE_SHT(drv_name)					\
 	.module			= THIS_MODULE,			\
@@ -1116,7 +1124,8 @@ extern const struct ata_port_operations sata_port_ops;
 	.proc_name		= drv_name,			\
 	.slave_configure	= ata_scsi_slave_config,	\
 	.slave_destroy		= ata_scsi_slave_destroy,	\
-	.bios_param		= ata_std_bios_param
+	.bios_param		= ata_std_bios_param,		\
+	.sdev_attrs		= ata_common_sdev_attrs
 
 #define ATA_NCQ_SHT(drv_name)					\
 	ATA_BASE_SHT(drv_name),					\
-- 
cgit v1.3-7-g2ca7


From 6866e7bc83f13a1bc6de59099930e9db1ab0042f Mon Sep 17 00:00:00 2001
From: Richard Kennedy <richard@rsk.demon.co.uk>
Date: Mon, 22 Sep 2008 14:47:13 -0700
Subject: libata: reorder ata_device to remove 8 bytes of padding on 64 bits

reduce size by 8 bytes from 1160 to 1152 allowing it to fit in 1 fewer
cachelines.

Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
---
 include/linux/libata.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/include/linux/libata.h b/include/linux/libata.h
index 1f44cfb847e1..947cf84e555d 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -560,8 +560,8 @@ struct ata_ering {
 struct ata_device {
 	struct ata_link		*link;
 	unsigned int		devno;		/* 0 or 1 */
-	unsigned long		flags;		/* ATA_DFLAG_xxx */
 	unsigned int		horkage;	/* List of broken features */
+	unsigned long		flags;		/* ATA_DFLAG_xxx */
 	struct scsi_device	*sdev;		/* attached SCSI device */
 #ifdef CONFIG_ATA_ACPI
 	acpi_handle		acpi_handle;
-- 
cgit v1.3-7-g2ca7


From b00c1a99e7758f794923c61e5cd55268d61c9469 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 29 Sep 2008 15:44:46 +0200
Subject: hrtimer: mark migration state

Impact: during migration active hrtimers can be seen as inactive

The migration code removes the hrtimers from the queues of the dead
CPU and sets the state temporary to INACTIVE. The enqueue code sets it
to ACTIVE/PENDING again.

Prevent that the wrong state can be seen by using a separate migration
state bit.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/hrtimer.h |  4 +++-
 kernel/hrtimer.c        | 12 ++++++++++--
 2 files changed, 13 insertions(+), 3 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index 6d93dce61cbb..bdd88df1b4e5 100644
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -67,9 +67,10 @@ enum hrtimer_cb_mode {
  * 0x02		callback function running
  * 0x04		callback pending (high resolution mode)
  *
- * Special case:
+ * Special cases:
  * 0x03		callback function running and enqueued
  *		(was requeued on another CPU)
+ * 0x09		timer was migrated on CPU hotunplug
  * The "callback function running and enqueued" status is only possible on
  * SMP. It happens for example when a posix timer expired and the callback
  * queued a signal. Between dropping the lock which protects the posix timer
@@ -87,6 +88,7 @@ enum hrtimer_cb_mode {
 #define HRTIMER_STATE_ENQUEUED	0x01
 #define HRTIMER_STATE_CALLBACK	0x02
 #define HRTIMER_STATE_PENDING	0x04
+#define HRTIMER_STATE_MIGRATE	0x08
 
 /**
  * struct hrtimer - the basic hrtimer structure
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index ac2f6d6d4868..ace723dd1e52 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -1602,7 +1602,13 @@ static int migrate_hrtimer_list(struct hrtimer_clock_base *old_base,
 		timer = rb_entry(node, struct hrtimer, node);
 		BUG_ON(hrtimer_callback_running(timer));
 		debug_hrtimer_deactivate(timer);
-		__remove_hrtimer(timer, old_base, HRTIMER_STATE_INACTIVE, 0);
+
+		/*
+		 * Mark it as STATE_MIGRATE not INACTIVE otherwise the
+		 * timer could be seen as !active and just vanish away
+		 * under us on another CPU
+		 */
+		__remove_hrtimer(timer, old_base, HRTIMER_STATE_MIGRATE, 0);
 		timer->base = new_base;
 		/*
 		 * Enqueue the timer. Allow reprogramming of the event device
@@ -1620,13 +1626,15 @@ static int migrate_hrtimer_list(struct hrtimer_clock_base *old_base,
 		 * state, we need to do that otherwise we end up with
 		 * a stale timer.
 		 */
-		if (timer->state == HRTIMER_STATE_INACTIVE) {
+		if (timer->state == HRTIMER_STATE_MIGRATE) {
 			timer->state = HRTIMER_STATE_PENDING;
 			list_add_tail(&timer->cb_entry,
 				      &new_base->cpu_base->cb_pending);
 			raise = 1;
 		}
 #endif
+		/* Clear the migration state bit */
+		timer->state &= ~HRTIMER_STATE_MIGRATE;
 	}
 	return raise;
 }
-- 
cgit v1.3-7-g2ca7


From ccc7dadf736639da86f3e0c86832c11a66fc8221 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 29 Sep 2008 15:47:42 +0200
Subject: hrtimer: prevent migration of per CPU hrtimers

Impact: per CPU hrtimers can be migrated from a dead CPU

The hrtimer code has no knowledge about per CPU timers, but we need to
prevent the migration of such timers and warn when such a timer is
active at migration time.

Explicitely mark the timers as per CPU and use a more understandable
mode descriptor for the interrupts safe unlocked callback mode, which
is used by hrtimer_sleeper and the scheduler code.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/hrtimer.h      | 14 +++++++++++---
 kernel/hrtimer.c             | 37 +++++++++++++++++++++++++------------
 kernel/sched.c               |  4 ++--
 kernel/time/tick-sched.c     |  2 +-
 kernel/trace/trace_sysprof.c |  2 +-
 5 files changed, 40 insertions(+), 19 deletions(-)

(limited to 'include/linux')

diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index bdd88df1b4e5..2f245fe63bda 100644
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -47,14 +47,22 @@ enum hrtimer_restart {
  *	HRTIMER_CB_IRQSAFE:		Callback may run in hardirq context
  *	HRTIMER_CB_IRQSAFE_NO_RESTART:	Callback may run in hardirq context and
  *					does not restart the timer
- *	HRTIMER_CB_IRQSAFE_NO_SOFTIRQ:	Callback must run in hardirq context
- *					Special mode for tick emultation
+ *	HRTIMER_CB_IRQSAFE_PERCPU:	Callback must run in hardirq context
+ *					Special mode for tick emulation and
+ *					scheduler timer. Such timers are per
+ *					cpu and not allowed to be migrated on
+ *					cpu unplug.
+ *	HRTIMER_CB_IRQSAFE_UNLOCKED:	Callback should run in hardirq context
+ *					with timer->base lock unlocked
+ *					used for timers which call wakeup to
+ *					avoid lock order problems with rq->lock
  */
 enum hrtimer_cb_mode {
 	HRTIMER_CB_SOFTIRQ,
 	HRTIMER_CB_IRQSAFE,
 	HRTIMER_CB_IRQSAFE_NO_RESTART,
-	HRTIMER_CB_IRQSAFE_NO_SOFTIRQ,
+	HRTIMER_CB_IRQSAFE_PERCPU,
+	HRTIMER_CB_IRQSAFE_UNLOCKED,
 };
 
 /*
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index ace723dd1e52..cdec83e722fa 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -672,13 +672,14 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
 			 */
 			BUG_ON(timer->function(timer) != HRTIMER_NORESTART);
 			return 1;
-		case HRTIMER_CB_IRQSAFE_NO_SOFTIRQ:
+		case HRTIMER_CB_IRQSAFE_PERCPU:
+		case HRTIMER_CB_IRQSAFE_UNLOCKED:
 			/*
 			 * This is solely for the sched tick emulation with
 			 * dynamic tick support to ensure that we do not
 			 * restart the tick right on the edge and end up with
 			 * the tick timer in the softirq ! The calling site
-			 * takes care of this.
+			 * takes care of this. Also used for hrtimer sleeper !
 			 */
 			debug_hrtimer_deactivate(timer);
 			return 1;
@@ -1245,7 +1246,8 @@ static void __run_hrtimer(struct hrtimer *timer)
 	timer_stats_account_hrtimer(timer);
 
 	fn = timer->function;
-	if (timer->cb_mode == HRTIMER_CB_IRQSAFE_NO_SOFTIRQ) {
+	if (timer->cb_mode == HRTIMER_CB_IRQSAFE_PERCPU ||
+	    timer->cb_mode == HRTIMER_CB_IRQSAFE_UNLOCKED) {
 		/*
 		 * Used for scheduler timers, avoid lock inversion with
 		 * rq->lock and tasklist_lock.
@@ -1452,7 +1454,7 @@ void hrtimer_init_sleeper(struct hrtimer_sleeper *sl, struct task_struct *task)
 	sl->timer.function = hrtimer_wakeup;
 	sl->task = task;
 #ifdef CONFIG_HIGH_RES_TIMERS
-	sl->timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ;
+	sl->timer.cb_mode = HRTIMER_CB_IRQSAFE_UNLOCKED;
 #endif
 }
 
@@ -1592,7 +1594,7 @@ static void __cpuinit init_hrtimers_cpu(int cpu)
 #ifdef CONFIG_HOTPLUG_CPU
 
 static int migrate_hrtimer_list(struct hrtimer_clock_base *old_base,
-				struct hrtimer_clock_base *new_base)
+				struct hrtimer_clock_base *new_base, int dcpu)
 {
 	struct hrtimer *timer;
 	struct rb_node *node;
@@ -1603,6 +1605,18 @@ static int migrate_hrtimer_list(struct hrtimer_clock_base *old_base,
 		BUG_ON(hrtimer_callback_running(timer));
 		debug_hrtimer_deactivate(timer);
 
+		/*
+		 * Should not happen. Per CPU timers should be
+		 * canceled _before_ the migration code is called
+		 */
+		if (timer->cb_mode == HRTIMER_CB_IRQSAFE_PERCPU) {
+			__remove_hrtimer(timer, old_base,
+					 HRTIMER_STATE_INACTIVE, 0);
+			WARN(1, "hrtimer (%p %p)active but cpu %d dead\n",
+			     timer, timer->function, dcpu);
+			continue;
+		}
+
 		/*
 		 * Mark it as STATE_MIGRATE not INACTIVE otherwise the
 		 * timer could be seen as !active and just vanish away
@@ -1619,12 +1633,11 @@ static int migrate_hrtimer_list(struct hrtimer_clock_base *old_base,
 		/*
 		 * Happens with high res enabled when the timer was
 		 * already expired and the callback mode is
-		 * HRTIMER_CB_IRQSAFE_NO_SOFTIRQ
-		 * (hrtimer_sleeper). The enqueue code does not move
-		 * them to the soft irq pending list for
-		 * performance/latency reasons, but in the migration
-		 * state, we need to do that otherwise we end up with
-		 * a stale timer.
+		 * HRTIMER_CB_IRQSAFE_UNLOCKED (hrtimer_sleeper). The
+		 * enqueue code does not move them to the soft irq
+		 * pending list for performance/latency reasons, but
+		 * in the migration state, we need to do that
+		 * otherwise we end up with a stale timer.
 		 */
 		if (timer->state == HRTIMER_STATE_MIGRATE) {
 			timer->state = HRTIMER_STATE_PENDING;
@@ -1682,7 +1695,7 @@ static void migrate_hrtimers(int cpu)
 
 	for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) {
 		if (migrate_hrtimer_list(&old_base->clock_base[i],
-					 &new_base->clock_base[i]))
+					 &new_base->clock_base[i], cpu))
 			raise = 1;
 	}
 
diff --git a/kernel/sched.c b/kernel/sched.c
index 13dd2db9fb2d..ad1962dc0aa2 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -201,7 +201,7 @@ void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime)
 	hrtimer_init(&rt_b->rt_period_timer,
 			CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	rt_b->rt_period_timer.function = sched_rt_period_timer;
-	rt_b->rt_period_timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ;
+	rt_b->rt_period_timer.cb_mode = HRTIMER_CB_IRQSAFE_UNLOCKED;
 }
 
 static void start_rt_bandwidth(struct rt_bandwidth *rt_b)
@@ -1119,7 +1119,7 @@ static void init_rq_hrtick(struct rq *rq)
 
 	hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	rq->hrtick_timer.function = hrtick;
-	rq->hrtick_timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ;
+	rq->hrtick_timer.cb_mode = HRTIMER_CB_IRQSAFE_PERCPU;
 }
 #else
 static inline void hrtick_clear(struct rq *rq)
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 39019b3f7621..cb02324bdb88 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -625,7 +625,7 @@ void tick_setup_sched_timer(void)
 	 */
 	hrtimer_init(&ts->sched_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
 	ts->sched_timer.function = tick_sched_timer;
-	ts->sched_timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ;
+	ts->sched_timer.cb_mode = HRTIMER_CB_IRQSAFE_PERCPU;
 
 	/* Get the next period (per cpu) */
 	ts->sched_timer.expires = tick_init_jiffy_update();
diff --git a/kernel/trace/trace_sysprof.c b/kernel/trace/trace_sysprof.c
index bb948e52ce20..db58fb66a135 100644
--- a/kernel/trace/trace_sysprof.c
+++ b/kernel/trace/trace_sysprof.c
@@ -202,7 +202,7 @@ static void start_stack_timer(int cpu)
 
 	hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
 	hrtimer->function = stack_trace_timer_fn;
-	hrtimer->cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ;
+	hrtimer->cb_mode = HRTIMER_CB_IRQSAFE_PERCPU;
 
 	hrtimer_start(hrtimer, ns_to_ktime(sample_period), HRTIMER_MODE_REL);
 }
-- 
cgit v1.3-7-g2ca7


From 4b19de6d1cb07c8bcb6778e771f9cfd5bcfdfd3e Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Thu, 2 Oct 2008 14:50:16 -0700
Subject: mm: tiny-shmem nommu fix

The previous patch db203d53d474aa068984e409d807628f5841da1b ("mm:
tiny-shmem fix lock ordering: mmap_sem vs i_mutex") to fix the lock
ordering in tiny-shmem breaks shared anonymous and IPC memory on NOMMU
architectures because it was using the expanding truncate to signal ramfs
to allocate a physically contiguous RAM backing the inode (otherwise it is
unusable for "memory mapping" it to userspace).

However do_truncate is what caused the lock ordering error, due to it
taking i_mutex.  In this case, we can actually just call ramfs directly to
allocate memory for the mapping, rather than go via truncate.

Acked-by: David Howells <dhowells@redhat.com>
Acked-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ramfs/file-nommu.c | 2 +-
 include/linux/ramfs.h | 1 +
 mm/tiny-shmem.c       | 6 ++++++
 3 files changed, 8 insertions(+), 1 deletion(-)

(limited to 'include/linux')

diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c
index 52312ec93ff4..5145cb9125af 100644
--- a/fs/ramfs/file-nommu.c
+++ b/fs/ramfs/file-nommu.c
@@ -58,7 +58,7 @@ const struct inode_operations ramfs_file_inode_operations = {
  * size 0 on the assumption that it's going to be used for an mmap of shared
  * memory
  */
-static int ramfs_nommu_expand_for_mapping(struct inode *inode, size_t newsize)
+int ramfs_nommu_expand_for_mapping(struct inode *inode, size_t newsize)
 {
 	struct pagevec lru_pvec;
 	unsigned long npages, xpages, loop, limit;
diff --git a/include/linux/ramfs.h b/include/linux/ramfs.h
index b160fb18e8d6..37aaf2b39863 100644
--- a/include/linux/ramfs.h
+++ b/include/linux/ramfs.h
@@ -6,6 +6,7 @@ extern int ramfs_get_sb(struct file_system_type *fs_type,
 	 int flags, const char *dev_name, void *data, struct vfsmount *mnt);
 
 #ifndef CONFIG_MMU
+extern int ramfs_nommu_expand_for_mapping(struct inode *inode, size_t newsize);
 extern unsigned long ramfs_nommu_get_unmapped_area(struct file *file,
 						   unsigned long addr,
 						   unsigned long len,
diff --git a/mm/tiny-shmem.c b/mm/tiny-shmem.c
index d17cb6f6ab10..8d7a27a6335c 100644
--- a/mm/tiny-shmem.c
+++ b/mm/tiny-shmem.c
@@ -80,6 +80,12 @@ struct file *shmem_file_setup(char *name, loff_t size, unsigned long flags)
 	inode->i_nlink = 0;	/* It is unlinked */
 	init_file(file, shm_mnt, dentry, FMODE_WRITE | FMODE_READ,
 			&ramfs_file_operations);
+
+#ifndef CONFIG_MMU
+	error = ramfs_nommu_expand_for_mapping(inode, size);
+	if (error)
+		goto close_file;
+#endif
 	return file;
 
 close_file:
-- 
cgit v1.3-7-g2ca7


From 897312bd240357c88ce906633703c324c6f0a5cd Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Fri, 3 Oct 2008 15:23:41 -0700
Subject: include/linux/stacktrace.h: declare struct task_struct

include/linux/stacktrace.h:13: warning:
 'struct task_struct' declared inside parameter list

(This might be a hard error on sparc64, which uses this header and has
-Werror)

Reported-by: "Randy.Dunlap" <rdunlap@xenotime.net>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/stacktrace.h | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'include/linux')

diff --git a/include/linux/stacktrace.h b/include/linux/stacktrace.h
index 5da9794b2d78..b106fd8e0d5c 100644
--- a/include/linux/stacktrace.h
+++ b/include/linux/stacktrace.h
@@ -1,6 +1,8 @@
 #ifndef __LINUX_STACKTRACE_H
 #define __LINUX_STACKTRACE_H
 
+struct task_struct;
+
 #ifdef CONFIG_STACKTRACE
 struct stack_trace {
 	unsigned int nr_entries, max_entries;
-- 
cgit v1.3-7-g2ca7


From f20f258603ebc5da91e76884cf0c0d7ac9804b1c Mon Sep 17 00:00:00 2001
From: Borislav Petkov <petkovbb@gmail.com>
Date: Sun, 5 Oct 2008 18:23:27 +0200
Subject: ide-cd: temporary tray close fix

This one fixes http://bugzilla.kernel.org/show_bug.cgi?id=11602.

A more generic fix for drives which cannot autoclose tray will follow.

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
[bart: add an extra parentheses for consistency with the rest of kernel code]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-cd.c | 5 ++++-
 include/linux/ide.h  | 4 +++-
 2 files changed, 7 insertions(+), 2 deletions(-)

(limited to 'include/linux')

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 49a8c589e346..e1cd7d520fc0 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -1661,7 +1661,9 @@ static int ide_cdrom_probe_capabilities(ide_drive_t *drive)
 		cdi->mask &= ~CDC_PLAY_AUDIO;
 
 	mechtype = buf[8 + 6] >> 5;
-	if (mechtype == mechtype_caddy || mechtype == mechtype_popup)
+	if (mechtype == mechtype_caddy ||
+	    mechtype == mechtype_popup ||
+	    (drive->atapi_flags & IDE_AFLAG_NO_AUTOCLOSE))
 		cdi->mask |= CDC_CLOSE_TRAY;
 
 	if (cdi->sanyo_slot > 0) {
@@ -1859,6 +1861,7 @@ static const struct cd_list_entry ide_cd_quirks_list[] = {
 	{ "MATSHITADVD-ROM SR-8176", NULL,   IDE_AFLAG_PLAY_AUDIO_OK	     },
 	{ "MATSHITADVD-ROM SR-8174", NULL,   IDE_AFLAG_PLAY_AUDIO_OK	     },
 	{ "Optiarc DVD RW AD-5200A", NULL,   IDE_AFLAG_PLAY_AUDIO_OK	     },
+	{ "Optiarc DVD RW AD-7543A", NULL,   IDE_AFLAG_NO_AUTOCLOSE	     },
 	{ NULL, NULL, 0 }
 };
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 1524829f73f2..6514db8fd2e4 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -366,7 +366,9 @@ enum {
 	/* Currently on a filemark */
 	IDE_AFLAG_FILEMARK		= (1 << 25),
 	/* 0 = no tape is loaded, so we don't rewind after ejecting */
-	IDE_AFLAG_MEDIUM_PRESENT	= (1 << 26)
+	IDE_AFLAG_MEDIUM_PRESENT	= (1 << 26),
+
+	IDE_AFLAG_NO_AUTOCLOSE		= (1 << 27),
 };
 
 struct ide_drive_s {
-- 
cgit v1.3-7-g2ca7