maria-developers team mailing list archive
-
maria-developers team
-
Mailing list archive
-
Message #06107
Re: MariaDB mutex contention
Mutex contention from various plugins
On Tue, Aug 20, 2013 at 9:12 AM, Pavel Ivanov <pivanof@xxxxxxxxxx> wrote:
> Workaround for which problem?
>
> On Tue, Aug 20, 2013 at 8:17 AM, MARK CALLAGHAN <mdcallag@xxxxxxxxx>
> wrote:
> > Is the workaround to use static rather than dynamic plugins?
> >
> >
> > On Tue, Aug 20, 2013 at 7:31 AM, Sergei Golubchik <serg@xxxxxxxxxxx>
> wrote:
> >>
> >> Hi, Pavel!
> >>
> >> On Aug 20, Pavel Ivanov wrote:
> >> > On Tue, Aug 20, 2013 at 1:09 AM, Sergei Golubchik <serg@xxxxxxxxxxx>
> >> > wrote:
> >> > > On Aug 19, Pavel Ivanov wrote:
> >> > >> No, it's not reasonable for semi-sync to lock/unlock LOCK_plugin.
> >> > >> It's plugin infrastructure that does that.
> >> > >>
> >> > >> I've actually was terrified to learn that each call into semi-sync
> >> > >> plugin is surrounded with pthread_rwlock_rdlock/
> >> > >> pthread_rwlock_unlock (which is not cheap I believe). And also for
> >> > >> each such call it "locks"/"unlocks" the semi-sync plugin. And
> >> > >> although "locking" plugin avoids locking LOCK_plugin when plugin is
> >> > >> linked statically, "unlocking" doesn't do that.
> >> > >
> >> > Sure, but macro FOREACH_OBSERVER inside rpl_handler.cc doesn't use
> >> > this function. It uses plugin_unlock_list() which always locks
> >> > LOCK_plugin.
> >> >
> >> > BTW, MariaDB still supports compiling semi-sync plugins dynamically,
> >> > but it seems that it doesn't do anything against unloading semi-sync
> >> > plugins in the middle of transactions. Did anyone think about this?
> >>
> >> Judging from the replication plugin API - I don't think so.
> >>
> >> But if it adds that much overhead, I suppose we'll need to fix it.
> >> Then we fix the unloading too.
> >>
> >> Regards,
> >> Sergei
> >>
> >> _______________________________________________
> >> Mailing list: https://launchpad.net/~maria-developers
> >> Post to : maria-developers@xxxxxxxxxxxxxxxxxxx
> >> Unsubscribe : https://launchpad.net/~maria-developers
> >> More help : https://help.launchpad.net/ListHelp
> >
> >
> >
> >
> > --
> > Mark Callaghan
> > mdcallag@xxxxxxxxx
>
--
Mark Callaghan
mdcallag@xxxxxxxxx
Follow ups
References
-
MariaDB mutex contention (was: Re: MariaDB 10.0 benchmarks on upcoming IVB-EP processors ...)
From: Axel Schwenke, 2013-08-16
-
Re: MariaDB mutex contention (was: Re: MariaDB 10.0 benchmarks on upcoming IVB-EP processors ...)
From: Sergey Vojtovich, 2013-08-17
-
Re: MariaDB mutex contention (was: Re: MariaDB 10.0 benchmarks on upcoming IVB-EP processors ...)
From: Pavel Ivanov, 2013-08-17
-
Re: MariaDB mutex contention
From: Axel Schwenke, 2013-08-19
-
Re: MariaDB mutex contention
From: Roberto Spadim, 2013-08-19
-
Re: MariaDB mutex contention
From: Pavel Ivanov, 2013-08-19
-
Re: MariaDB mutex contention
From: MARK CALLAGHAN, 2013-08-19
-
Re: MariaDB mutex contention
From: Pavel Ivanov, 2013-08-19
-
Re: MariaDB mutex contention
From: Sergei Golubchik, 2013-08-20
-
Re: MariaDB mutex contention
From: Pavel Ivanov, 2013-08-20
-
Re: MariaDB mutex contention
From: Sergei Golubchik, 2013-08-20
-
Re: MariaDB mutex contention
From: MARK CALLAGHAN, 2013-08-20
-
Re: MariaDB mutex contention
From: Pavel Ivanov, 2013-08-20