diff options
| author | Ingo Molnar <mingo@elte.hu> | 2009-04-08 17:02:50 +0200 |
|---|---|---|
| committer | Ingo Molnar <mingo@elte.hu> | 2009-04-08 17:02:57 +0200 |
| commit | ff96e612cba32510e263e17b213235fe5746397e (patch) | |
| tree | a8df57d76b10e0901a4fb76cd2987eb9826a560a /fs/btrfs/locking.c | |
| parent | cd84a42f315e50edd454c27a3da3951ccd3d735a (diff) | |
| parent | 577c9c456f0e1371cbade38eaf91ae8e8a308555 (diff) | |
Merge commit 'v2.6.30-rc1' into core/urgent
Merge reason: need latest upstream to queue up dependent fix
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Diffstat (limited to 'fs/btrfs/locking.c')
| -rw-r--r-- | fs/btrfs/locking.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c index a5310c0f41e2..1c36e5cd8f55 100644 --- a/fs/btrfs/locking.c +++ b/fs/btrfs/locking.c @@ -60,8 +60,8 @@ void btrfs_clear_lock_blocking(struct extent_buffer *eb) /* * unfortunately, many of the places that currently set a lock to blocking - * don't end up blocking for every long, and often they don't block - * at all. For a dbench 50 run, if we don't spin one the blocking bit + * don't end up blocking for very long, and often they don't block + * at all. For a dbench 50 run, if we don't spin on the blocking bit * at all, the context switch rate can jump up to 400,000/sec or more. * * So, we're still stuck with this crummy spin on the blocking bit, |