diff options
author | Pranith Kumar K <pranithk@gluster.com> | 2012-09-25 10:35:14 +0530 |
---|---|---|
committer | Anand Avati <avati@redhat.com> | 2012-09-25 10:45:08 -0700 |
commit | 7add67bcca82e438b4aece5aad096a08a866e26c (patch) | |
tree | dc5327b365357a9e4e8bd722f6ffec5757043c0a /xlators | |
parent | 5ad96fb5a8c79fe50c21a726a7cbda4eaa3b96cc (diff) |
cluster/afr: Wake up post-op on non-co-operative transaction
Problem:
The problem is observed when kernel untar is done. One file untar
happens every second. The reason for this is, setattr lock is blocked
on the prev fd data-transaction full-lock (because of eager-lock).
Because of post-op-delay the post-op (xattrop + unlock) of the prev
data-transaction happens after 1 sec.
Until this the setattr is blocked resulting in performance problems
in untar.
Fix:
Whenever an loc data, meta-data transaction comes, it should wakeup
the prev-post-op on the same process' fd.
Tests:
The performance problem in untar went away. I put a breakpoint in
client_finodelk for a 2G file dd and the inodelk is hit only 4 times.
This confirms that the change does not affect post-op-delay in a
-ve way.
Change-Id: Ice3c2a1211f4dca6520a19bc4ba6cb9efb2902ad
BUG: 845754
Signed-off-by: Pranith Kumar K <pranithk@gluster.com>
Reviewed-on: http://review.gluster.org/3975
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Jeff Darcy <jdarcy@redhat.com>
Reviewed-by: Anand Avati <avati@redhat.com>
Diffstat (limited to 'xlators')
-rw-r--r-- | xlators/cluster/afr/src/afr-transaction.c | 23 |
1 files changed, 23 insertions, 0 deletions
diff --git a/xlators/cluster/afr/src/afr-transaction.c b/xlators/cluster/afr/src/afr-transaction.c index ddf0af31794..f5094e1cfd9 100644 --- a/xlators/cluster/afr/src/afr-transaction.c +++ b/xlators/cluster/afr/src/afr-transaction.c @@ -1433,11 +1433,25 @@ afr_transaction_fop_failed (call_frame_t *frame, xlator_t *this, int child_index child_index, local->transaction.type); } +gf_boolean_t +_does_transaction_conflict_with_delayed_post_op (call_frame_t *frame) +{ + afr_local_t *local = frame->local; + //check if it is going to compete with inode lock from the fd + if (local->fd) + return _gf_false; + if ((local->transaction.type == AFR_DATA_TRANSACTION) || + (local->transaction.type == AFR_METADATA_TRANSACTION)) + return _gf_true; + return _gf_false; +} + int afr_transaction (call_frame_t *frame, xlator_t *this, afr_transaction_type type) { afr_local_t * local = NULL; afr_private_t * priv = NULL; + fd_t *fd = NULL; local = frame->local; priv = this->private; @@ -1453,6 +1467,15 @@ afr_transaction (call_frame_t *frame, xlator_t *this, afr_transaction_type type) local->transaction.resume = afr_transaction_resume; local->transaction.type = type; + if (_does_transaction_conflict_with_delayed_post_op (frame) && + local->loc.inode) { + fd = fd_lookup (local->loc.inode, frame->root->pid); + if (fd) { + afr_delayed_changelog_wake_up (this, fd); + fd_unref (fd); + } + } + if (afr_lock_server_count (priv, local->transaction.type) == 0) { afr_internal_lock_finish (frame, this); } else { |