On 2018/12/12 下午9:50, David Sterba wrote: > On Wed, Dec 12, 2018 at 09:20:33PM +0800, Qu Wenruo wrote: >> >> >> On 2018/12/12 下午8:37, David Sterba wrote: >>> On Wed, Dec 12, 2018 at 08:39:46AM +0800, Qu Wenruo wrote: >>>> changelog: >>>> v2: >>>> - Rebase to v4.20-rc1. >>>> - Instead commit transaction after each reloc tree merge, delay it until >>>> merge_reloc_roots() finishes. >>>> - This provides a more natural behavior, and reduce the unnecessary >>>> transaction commits. >>>> v3: >>>> - Fix backref walk deadlock by not triggering it at all. >>>> This also removes the need for @exec_post refactor and replace the >>>> patch to allow @old_root unpopulated. >>>> - Include the patch that fixes the unexpected data rsv free. >>> >>> I've tested the deadlock fix but it still locks up and there are some >>> lockdep warnings. I can't spend more time on that given the pre-merge >>> window timing so I'm going to postpone all qgruop things until the next >>> cycle. >> >> No problem. >> >> But would you please provide the reproducer? >> My test script can't reproduce lockup this time. > > git://github.com/kdave/testunion > > in the directory test-subvol-qg, it needs a 15GB tarred dump of > wikipedia, the url is in the repo. You can possibly replace it with > something else, the scripts need tmux and maybe some other utilites. Run > with startme script and path to fresh filesystem. >
Just in case, the github branch name changed. It's now "qgroup_delayed_subtree", no "_rebased" suffix. If still lock up, then it's going to be tricky. Thanks, Qu
signature.asc
Description: OpenPGP digital signature
