summaryrefslogtreecommitdiffstats
path: root/fs/xfs/xfs_mount.c
diff options
context:
space:
mode:
authorDave Chinner <dchinner@redhat.com>2010-09-24 19:59:04 +1000
committerAlex Elder <aelder@sgi.com>2010-10-18 15:07:56 -0500
commit74f75a0cb7033918eb0fa4a50df25091ac75c16e (patch)
tree3885c0b357c760152d14df03ef88839fdbf5f964 /fs/xfs/xfs_mount.c
parent69b491c214d7fd4d4df972ae5377be99ca3753db (diff)
xfs: convert buffer cache hash to rbtree
The buffer cache hash is showing typical hash scalability problems. In large scale testing the number of cached items growing far larger than the hash can efficiently handle. Hence we need to move to a self-scaling cache indexing mechanism. I have selected rbtrees for indexing becuse they can have O(log n) search scalability, and insert and remove cost is not excessive, even on large trees. Hence we should be able to cache large numbers of buffers without incurring the excessive cache miss search penalties that the hash is imposing on us. To ensure we still have parallel access to the cache, we need multiple trees. Rather than hashing the buffers by disk address to select a tree, it seems more sensible to separate trees by typical access patterns. Most operations use buffers from within a single AG at a time, so rather than searching lots of different lists, separate the buffer indexes out into per-AG rbtrees. This means that searches during metadata operation have a much higher chance of hitting cache resident nodes, and that updates of the tree are less likely to disturb trees being accessed on other CPUs doing independent operations. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Alex Elder <aelder@sgi.com>
Diffstat (limited to 'fs/xfs/xfs_mount.c')
-rw-r--r--fs/xfs/xfs_mount.c2
1 files changed, 2 insertions, 0 deletions
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 59859c343e0..cfa2fb4e7f9 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -479,6 +479,8 @@ xfs_initialize_perag(
rwlock_init(&pag->pag_ici_lock);
mutex_init(&pag->pag_ici_reclaim_lock);
INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC);
+ spin_lock_init(&pag->pag_buf_lock);
+ pag->pag_buf_tree = RB_ROOT;
if (radix_tree_preload(GFP_NOFS))
goto out_unwind;