xfs: do not discard alloc btree blocks

Blocks for the allocation btree are allocated from and released to
the AGFL, and thus frequently reused.  Even worse we do not have an
easy way to avoid using an AGFL block when it is discarded due to
the simple FILO list of free blocks, and thus can frequently stall
on blocks that are currently undergoing a discard.

Add a flag to the busy extent tracking structure to skip the discard
for allocation btree blocks.  In normal operation these blocks are
reused frequently enough that there is no need to discard them
anyway, but if they spill over to the allocation btree as part of a
balance we "leak" blocks that we would otherwise discard.  We could
fix this by adding another flag and keeping these block in the
rbtree even after they aren't busy any more so that we could discard
them when they migrate out of the AGFL.  Given that this would cause
significant overhead I don't think it's worthwile for now.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>

4 files changed, 10 insertions, 5 deletions

diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h
index 8d52ba4c87e5..6530769a999b 100644
--- a/fs/xfs/xfs_ag.h
+++ b/fs/xfs/xfs_ag.h
@@ -189,6 +189,7 @@ struct xfs_busy_extent {
         xfs_extlen_t    length;
         unsigned int    flags;
 #define XFS_ALLOC_BUSY_DISCARDED        0x01    /* undergoing a discard op. */
+#define XFS_ALLOC_BUSY_SKIP_DISCARD     0x02    /* do not discard */
 };
 
 /*
diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c
index 721db22c6ec9..95862bbff56b 100644
--- a/fs/xfs/xfs_alloc.c
+++ b/fs/xfs/xfs_alloc.c
@@ -2469,7 +2469,7 @@ xfs_free_extent(
 
         error = xfs_free_ag_extent(tp, args.agbp, args.agno, args.agbno, len, 0);
         if (!error)
-                xfs_alloc_busy_insert(tp, args.agno, args.agbno, len);
+                xfs_alloc_busy_insert(tp, args.agno, args.agbno, len, 0);
 error0:
         xfs_perag_put(args.pag);
         return error;
@@ -2480,7 +2480,8 @@ xfs_alloc_busy_insert(
         struct xfs_trans        *tp,
         xfs_agnumber_t          agno,
         xfs_agblock_t           bno,
-        xfs_extlen_t            len)
+        xfs_extlen_t            len,
+        unsigned int            flags)
 {
         struct xfs_busy_extent  *new;
         struct xfs_busy_extent  *busyp;
@@ -2504,6 +2505,7 @@ xfs_alloc_busy_insert(
         new->bno = bno;
         new->length = len;
         INIT_LIST_HEAD(&new->list);
+        new->flags = flags;
 
         /* trace before insert to be able to see failed inserts */
         trace_xfs_alloc_busy(tp->t_mountp, agno, bno, len);
@@ -3018,7 +3020,8 @@ xfs_alloc_busy_clear(
                         agno = busyp->agno;
                 }
 
-                if (do_discard && busyp->length)
+                if (do_discard && busyp->length &&
+                    !(busyp->flags & XFS_ALLOC_BUSY_SKIP_DISCARD))
                         busyp->flags = XFS_ALLOC_BUSY_DISCARDED;
                 else
                         xfs_alloc_busy_clear_one(mp, pag, busyp);
diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/xfs_alloc.h
index 06aa8217452b..2f52b924be79 100644
--- a/fs/xfs/xfs_alloc.h
+++ b/fs/xfs/xfs_alloc.h
@@ -137,7 +137,7 @@ xfs_alloc_longest_free_extent(struct xfs_mount *mp,
 #ifdef __KERNEL__
 void
 xfs_alloc_busy_insert(struct xfs_trans *tp, xfs_agnumber_t agno,
-        xfs_agblock_t bno, xfs_extlen_t len);
+        xfs_agblock_t bno, xfs_extlen_t len, unsigned int flags);
 
 void
 xfs_alloc_busy_clear(struct xfs_mount *mp, struct list_head *list,
diff --git a/fs/xfs/xfs_alloc_btree.c b/fs/xfs/xfs_alloc_btree.c
index 8b469d53599f..2b3518826a69 100644
--- a/fs/xfs/xfs_alloc_btree.c
+++ b/fs/xfs/xfs_alloc_btree.c
@@ -120,7 +120,8 @@ xfs_allocbt_free_block(
         if (error)
                 return error;
 
-        xfs_alloc_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1);
+        xfs_alloc_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1,
+                              XFS_ALLOC_BUSY_SKIP_DISCARD);
         xfs_trans_agbtree_delta(cur->bc_tp, -1);
         return 0;
 }