summaryrefslogtreecommitdiff
path: root/fs/ocfs2/journal.c
diff options
context:
space:
mode:
authorMark Fasheh <mark.fasheh@oracle.com>2006-09-22 17:28:19 -0700
committerMark Fasheh <mark.fasheh@oracle.com>2006-09-24 13:50:46 -0700
commit24c19ef40474c3930597f31ae233dc06319bd881 (patch)
treee05b1cf72435d25bf47e67b206aa376bbea33b7d /fs/ocfs2/journal.c
parentf9e2d82e6395cfa0802446b54b63cc412089d82c (diff)
ocfs2: Remove i_generation from inode lock names
OCFS2 puts inode meta data in the "lock value block" provided by the DLM. Typically, i_generation is encoded in the lock name so that a deleted inode on and a new one in the same block don't share the same lvb. Unfortunately, that scheme means that the read in ocfs2_read_locked_inode() is potentially thrown away as soon as the meta data lock is taken - we cannot encode the lock name without first knowing i_generation, which requires a disk read. This patch encodes i_generation in the inode meta data lvb, and removes the value from the inode meta data lock name. This way, the read can be covered by a lock, and at the same time we can distinguish between an up to date and a stale LVB. This will help cold-cache stat(2) performance in particular. Since this patch changes the protocol version, we take the opportunity to do a minor re-organization of two of the LVB fields. Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
Diffstat (limited to 'fs/ocfs2/journal.c')
-rw-r--r--fs/ocfs2/journal.c3
1 files changed, 2 insertions, 1 deletions
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index f92bf1dd379a..fd9734def551 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1493,7 +1493,8 @@ static int ocfs2_queue_orphans(struct ocfs2_super *osb,
if (de->name_len == 2 && !strncmp("..", de->name, 2))
continue;
- iter = ocfs2_iget(osb, le64_to_cpu(de->inode));
+ iter = ocfs2_iget(osb, le64_to_cpu(de->inode),
+ OCFS2_FI_FLAG_NOLOCK);
if (IS_ERR(iter))
continue;