]> git.apps.os.sepia.ceph.com Git - ceph.git/commit
projects / ceph.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 5b89c47) | patch
librbd/cache/pwl/ssd/WriteLog: decrement m_bytes_allocated when retiring
authorIlya Dryomov <idryomov@gmail.com>
Wed, 28 Apr 2021 12:27:12 +0000 (14:27 +0200)
committerIlya Dryomov <idryomov@gmail.com>
Thu, 29 Apr 2021 20:09:16 +0000 (22:09 +0200)
commite2bbf4167fc8816d329ceae5c07c9e61599d9a17
tree880a198e3b46a0ea948a582ba05ff8a537a01be7tree | snapshot
parent5b89c47ec5139a8d07be09c9f0021d90fe4a663bcommit | diff
librbd/cache/pwl/ssd/WriteLog: decrement m_bytes_allocated when retiring

Currently if ssd cache is filled to capacity, all future I/O hangs
indefinitely because even though the cache eventually becomes clean
and retires enough entries to get back under RETIRE_HIGH_WATER, this
isn't communicated to AbstractWriteLog::check_allocation().

Fixes: https://tracker.ceph.com/issues/50560
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
src/librbd/cache/pwl/ssd/WriteLog.cc diff | blob | history
Unnamed repository; edit this file 'description' to name the repository.