Blocks for the allocation btree are allocated from and release 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@xxxxxx>
Index: xfs/fs/xfs/xfs_ag.h
===================================================================
--- xfs.orig/fs/xfs/xfs_ag.h 2011-03-31 12:27:55.475651908 +0200
+++ xfs/fs/xfs/xfs_ag.h 2011-03-31 12:32:18.650592820 +0200
@@ -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 */
};
/*
Index: xfs/fs/xfs/xfs_alloc_btree.c
===================================================================
--- xfs.orig/fs/xfs/xfs_alloc_btree.c 2011-03-31 12:22:56.000000000 +0200
+++ xfs/fs/xfs/xfs_alloc_btree.c 2011-03-31 12:32:18.650592820 +0200
@@ -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;
}
Index: xfs/fs/xfs/xfs_alloc.c
===================================================================
--- xfs.orig/fs/xfs/xfs_alloc.c 2011-03-31 12:30:54.407636028 +0200
+++ xfs/fs/xfs/xfs_alloc.c 2011-03-31 12:32:18.658593173 +0200
@@ -2458,7 +2458,7 @@ xfs_free_extent(
#endif
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;
@@ -2469,7 +2469,8 @@ 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;
@@ -2493,6 +2494,7 @@ 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);
@@ -2982,7 +2984,7 @@ xfs_alloc_busy_prepare_discard(
pag = xfs_perag_get(mp, busyp->agno);
spin_lock(&pag->pagb_lock);
- if (!busyp->length)
+ if (!busyp->length || (busyp->flags & XFS_ALLOC_BUSY_SKIP_DISCARD))
ret = false;
busyp->flags = XFS_ALLOC_BUSY_DISCARDED;
spin_unlock(&pag->pagb_lock);
Index: xfs/fs/xfs/xfs_alloc.h
===================================================================
--- xfs.orig/fs/xfs/xfs_alloc.h 2011-03-31 12:27:55.483617993 +0200
+++ xfs/fs/xfs/xfs_alloc.h 2011-03-31 12:32:19.222130801 +0200
@@ -137,7 +137,7 @@ xfs_alloc_longest_free_extent(struct xfs
#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 xfs_busy_extent *busyp);
_______________________________________________
xfs mailing list
xfs@xxxxxxxxxxx
http://oss.sgi.com/mailman/listinfo/xfs
