1 files changed, 43 insertions, 60 deletions

diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index 44c2b0ef9a41..ea79072f5210 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -37,14 +37,14 @@
 
 #include "xfs_sb.h"
 #include "xfs_inum.h"
+#include "xfs_log.h"
 #include "xfs_ag.h"
-#include "xfs_dmapi.h"
 #include "xfs_mount.h"
 #include "xfs_trace.h"
 
 static kmem_zone_t *xfs_buf_zone;
 STATIC int xfsbufd(void *);
-STATIC int xfsbufd_wakeup(int, gfp_t);
+STATIC int xfsbufd_wakeup(struct shrinker *, int, gfp_t);
 STATIC void xfs_buf_delwri_queue(xfs_buf_t *, int);
 static struct shrinker xfs_buf_shake = {
         .shrink = xfsbufd_wakeup,
@@ -339,7 +339,7 @@ _xfs_buf_lookup_pages(
                                         __func__, gfp_mask);
 
                         XFS_STATS_INC(xb_page_retries);
-                        xfsbufd_wakeup(0, gfp_mask);
+                        xfsbufd_wakeup(NULL, 0, gfp_mask);
                         congestion_wait(BLK_RW_ASYNC, HZ/50);
                         goto retry;
                 }
@@ -578,9 +578,9 @@ _xfs_buf_read(
                         XBF_READ_AHEAD | _XBF_RUN_QUEUES);
 
         status = xfs_buf_iorequest(bp);
-        if (!status && !(flags & XBF_ASYNC))
-                status = xfs_buf_iowait(bp);
-        return status;
+        if (status || XFS_BUF_ISERROR(bp) || (flags & XBF_ASYNC))
+                return status;
+        return xfs_buf_iowait(bp);
 }
 
 xfs_buf_t *
@@ -850,6 +850,12 @@ xfs_buf_lock_value(
  *      Note that this in no way locks the underlying pages, so it is only
  *      useful for synchronizing concurrent use of buffer objects, not for
  *      synchronizing independent access to the underlying pages.
+ *
+ *      If we come across a stale, pinned, locked buffer, we know that we
+ *      are being asked to lock a buffer that has been reallocated. Because
+ *      it is pinned, we know that the log has not been pushed to disk and
+ *      hence it will still be locked. Rather than sleeping until someone
+ *      else pushes the log, push it ourselves before trying to get the lock.
  */
 void
 xfs_buf_lock(
@@ -857,6 +863,8 @@ xfs_buf_lock(
 {
         trace_xfs_buf_lock(bp, _RET_IP_);
 
+        if (atomic_read(&bp->b_pin_count) && (bp->b_flags & XBF_STALE))
+                xfs_log_force(bp->b_mount, 0);
         if (atomic_read(&bp->b_io_remaining))
                 blk_run_address_space(bp->b_target->bt_mapping);
         down(&bp->b_sema);
@@ -888,36 +896,6 @@ xfs_buf_unlock(
         trace_xfs_buf_unlock(bp, _RET_IP_);
 }
 
-
-/*
- *      Pinning Buffer Storage in Memory
- *      Ensure that no attempt to force a buffer to disk will succeed.
- */
-void
-xfs_buf_pin(
-        xfs_buf_t               *bp)
-{
-        trace_xfs_buf_pin(bp, _RET_IP_);
-        atomic_inc(&bp->b_pin_count);
-}
-
-void
-xfs_buf_unpin(
-        xfs_buf_t               *bp)
-{
-        trace_xfs_buf_unpin(bp, _RET_IP_);
-
-        if (atomic_dec_and_test(&bp->b_pin_count))
-                wake_up_all(&bp->b_waiters);
-}
-
-int
-xfs_buf_ispin(
-        xfs_buf_t               *bp)
-{
-        return atomic_read(&bp->b_pin_count);
-}
-
 STATIC void
 xfs_buf_wait_unpin(
         xfs_buf_t               *bp)
@@ -1007,25 +985,19 @@ xfs_bwrite(
         struct xfs_mount        *mp,
         struct xfs_buf          *bp)
 {
-        int                     iowait = (bp->b_flags & XBF_ASYNC) == 0;
-        int                     error = 0;
+        int                     error;
 
-        bp->b_strat = xfs_bdstrat_cb;
         bp->b_mount = mp;
         bp->b_flags |= XBF_WRITE;
-        if (!iowait)
-                bp->b_flags |= _XBF_RUN_QUEUES;
+        bp->b_flags &= ~(XBF_ASYNC | XBF_READ);
 
         xfs_buf_delwri_dequeue(bp);
-        xfs_buf_iostrategy(bp);
-
-        if (iowait) {
-                error = xfs_buf_iowait(bp);
-                if (error)
-                        xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
-                xfs_buf_relse(bp);
-        }
+        xfs_bdstrat_cb(bp);
 
+        error = xfs_buf_iowait(bp);
+        if (error)
+                xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
+        xfs_buf_relse(bp);
         return error;
 }
 
@@ -1036,7 +1008,6 @@ xfs_bdwrite(
 {
         trace_xfs_buf_bdwrite(bp, _RET_IP_);
 
-        bp->b_strat = xfs_bdstrat_cb;
         bp->b_mount = mp;
 
         bp->b_flags &= ~XBF_READ;
@@ -1071,7 +1042,6 @@ xfs_bioerror(
         XFS_BUF_UNDONE(bp);
         XFS_BUF_STALE(bp);
 
-        XFS_BUF_CLR_BDSTRAT_FUNC(bp);
         xfs_biodone(bp);
 
         return EIO;
@@ -1101,7 +1071,6 @@ xfs_bioerror_relse(
         XFS_BUF_DONE(bp);
         XFS_BUF_STALE(bp);
         XFS_BUF_CLR_IODONE_FUNC(bp);
-        XFS_BUF_CLR_BDSTRAT_FUNC(bp);
         if (!(fl & XBF_ASYNC)) {
                 /*
                  * Mark b_error and B_ERROR _both_.
@@ -1307,8 +1276,19 @@ submit_io:
                 if (size)
                         goto next_chunk;
         } else {
-                bio_put(bio);
+                /*
+                 * if we get here, no pages were added to the bio. However,
+                 * we can't just error out here - if the pages are locked then
+                 * we have to unlock them otherwise we can hang on a later
+                 * access to the page.
+                 */
                 xfs_buf_ioerror(bp, EIO);
+                if (bp->b_flags & _XBF_PAGE_LOCKED) {
+                        int i;
+                        for (i = 0; i < bp->b_page_count; i++)
+                                unlock_page(bp->b_pages[i]);
+                }
+                bio_put(bio);
         }
 }
 
@@ -1614,7 +1594,8 @@ xfs_mapping_buftarg(
 
 STATIC int
 xfs_alloc_delwrite_queue(
-        xfs_buftarg_t           *btp)
+        xfs_buftarg_t           *btp,
+        const char              *fsname)
 {
         int     error = 0;
 
@@ -1622,7 +1603,7 @@ xfs_alloc_delwrite_queue(
         INIT_LIST_HEAD(&btp->bt_delwrite_queue);
         spin_lock_init(&btp->bt_delwrite_lock);
         btp->bt_flags = 0;
-        btp->bt_task = kthread_run(xfsbufd, btp, "xfsbufd");
+        btp->bt_task = kthread_run(xfsbufd, btp, "xfsbufd/%s", fsname);
         if (IS_ERR(btp->bt_task)) {
                 error = PTR_ERR(btp->bt_task);
                 goto out_error;
@@ -1635,7 +1616,8 @@ out_error:
 xfs_buftarg_t *
 xfs_alloc_buftarg(
         struct block_device     *bdev,
-        int                     external)
+        int                     external,
+        const char              *fsname)
 {
         xfs_buftarg_t           *btp;
 
@@ -1647,7 +1629,7 @@ xfs_alloc_buftarg(
                 goto error;
         if (xfs_mapping_buftarg(btp, bdev))
                 goto error;
-        if (xfs_alloc_delwrite_queue(btp))
+        if (xfs_alloc_delwrite_queue(btp, fsname))
                 goto error;
         xfs_alloc_bufhash(btp, external);
         return btp;
@@ -1756,6 +1738,7 @@ xfs_buf_runall_queues(
 
 STATIC int
 xfsbufd_wakeup(
+        struct shrinker         *shrink,
         int                     priority,
         gfp_t                   mask)
 {
@@ -1797,7 +1780,7 @@ xfs_buf_delwri_split(
                 trace_xfs_buf_delwri_split(bp, _RET_IP_);
                 ASSERT(bp->b_flags & XBF_DELWRI);
 
-                if (!xfs_buf_ispin(bp) && !xfs_buf_cond_lock(bp)) {
+                if (!XFS_BUF_ISPINNED(bp) && !xfs_buf_cond_lock(bp)) {
                         if (!force &&
                             time_before(jiffies, bp->b_queuetime + age)) {
                                 xfs_buf_unlock(bp);
@@ -1882,7 +1865,7 @@ xfsbufd(
                         struct xfs_buf *bp;
                         bp = list_first_entry(&tmp, struct xfs_buf, b_list);
                         list_del_init(&bp->b_list);
-                        xfs_buf_iostrategy(bp);
+                        xfs_bdstrat_cb(bp);
                         count++;
                 }
                 if (count)
@@ -1929,7 +1912,7 @@ xfs_flush_buftarg(
                         bp->b_flags &= ~XBF_ASYNC;
                         list_add(&bp->b_list, &wait_list);
                 }
-                xfs_buf_iostrategy(bp);
+                xfs_bdstrat_cb(bp);
         }
 
         if (wait) {