ceph: hold on to exclusive caps on complete directories

If a directory is complete, we want to keep the exclusive
cap. So that MDS does not end up revoking the shared cap
on every create/unlink operation.

Signed-off-by: Yan, Zheng <zyan@redhat.com>
This commit is contained in:
Yan, Zheng 2015-03-26 19:06:00 +08:00 committed by Ilya Dryomov
parent 9571eb4f96
commit 32ec439775

View file

@ -1561,7 +1561,19 @@ void ceph_check_caps(struct ceph_inode_info *ci, int flags,
if (!mdsc->stopping && inode->i_nlink > 0) {
if (want) {
retain |= CEPH_CAP_ANY; /* be greedy */
} else if (S_ISDIR(inode->i_mode) &&
(issued & CEPH_CAP_FILE_SHARED) &&
__ceph_dir_is_complete(ci)) {
/*
* If a directory is complete, we want to keep
* the exclusive cap. So that MDS does not end up
* revoking the shared cap on every create/unlink
* operation.
*/
want = CEPH_CAP_ANY_SHARED | CEPH_CAP_FILE_EXCL;
retain |= want;
} else {
retain |= CEPH_CAP_ANY_SHARED;
/*
* keep RD only if we didn't have the file open RW,