]> git-server-git.apps.pok.os.sepia.ceph.com Git - ceph.git/commit
projects / ceph.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: fb5b59f) | patch
mds: fix handling very fast delete ops 13899/head
authorJohn Spray <john.spray@redhat.com>
Thu, 9 Mar 2017 10:43:32 +0000 (10:43 +0000)
committerJohn Spray <john.spray@redhat.com>
Thu, 9 Mar 2017 11:28:51 +0000 (11:28 +0000)
commitc315be9894d56b4e7e3b9e4c2442af5bae7ecbbd
tree4bc0f933e39225e96f0315a9a4ccc489acf00bc0tree | snapshot
parentfb5b59fc394167cab5808ef147d291152c4ce5cdcommit | diff
mds: fix handling very fast delete ops

Deletions can complete so quickly that we're
ready to call execute_item_complete before
the end of execute_item.  We could use a finisher
here so that the completion was always called
from outside PurgeQueue::lock, but it's more efficient
to just handle the inline case directly.

Fixes: http://tracker.ceph.com/issues/19245
Signed-off-by: John Spray <john.spray@redhat.com>
src/mds/PurgeQueue.cc diff | blob | history
src/mds/PurgeQueue.h diff | blob | history
Unnamed repository; edit this file 'description' to name the repository.