|To:||Peter Leckie <pleckie@xxxxxxx>|
|Subject:||Re: [PATCH v2] Use atomic_t and wait_event to track dquot pincount|
|From:||Lachlan McIlroy <lachlan@xxxxxxx>|
|Date:||Fri, 26 Sep 2008 11:26:38 +1000|
|References:||<48D9C1DD.6030607@xxxxxxx> <48D9EB8F.1070104@xxxxxxx> <48D9EF6E.8010505@xxxxxxx> <20080924074604.GK5448@disturbed> <48D9F718.4010905@xxxxxxx> <20080925010318.GB27997@disturbed> <48DB4F3F.8040307@xxxxxxx> <20080926003401.GG27997@disturbed> <48DC3638.3050601@xxxxxxx>|
|User-agent:||Thunderbird 188.8.131.52 (X11/20080707)|
Peter Leckie wrote:
Dave Chinner wrote:Yeah I already had this idea I just have not posted a patch because Lachlan thoughbut it doesn't fix the underlying problem that was causing the spurious wakeups, which is the fact that xfs_qm_dqflush() is not obeying non-blocking flush directions. The patch below should fix that. Can you please test it before you add your patch?it might introduce a deadlock.
I suggested some changes a while back to make tail pushing non-blocking and Dave thought it might cause a deadlock. http://oss.sgi.com/archives/xfs/2008-07/msg00472.html I actually did hit a deadlock with this change but could not figure out why. It may have been the same issue Pete is trying to fix here.
If you think this is a good Idea I will update my patchto be non blocking.
|<Prev in Thread]||Current Thread||[Next in Thread>|
|Previous by Date:||Re: [PATCH v2] Use atomic_t and wait_event to track dquot pincount, Peter Leckie|
|Next by Date:||Re: [PATCH v2] Use atomic_t and wait_event to track dquot pincount, Lachlan McIlroy|
|Previous by Thread:||Re: [PATCH v2] Use atomic_t and wait_event to track dquot pincount, Peter Leckie|
|Next by Thread:||Re: [PATCH v2] Use atomic_t and wait_event to track dquot pincount, Dave Chinner|
|Indexes:||[Date] [Thread] [Top] [All Lists]|