Messages in this thread | | | Date | Mon, 20 Nov 2017 10:42:37 +0100 | From | Michal Hocko <> | Subject | Re: [PATCH 1/2] mm,vmscan: Kill global shrinker lock. |
| |
On Mon 20-11-17 01:33:09, Christoph Hellwig wrote: > On Mon, Nov 20, 2017 at 10:25:26AM +0100, Michal Hocko wrote: > > On Fri 17-11-17 09:35:21, Christoph Hellwig wrote: > > > On Tue, Nov 14, 2017 at 06:37:42AM +0900, Tetsuo Handa wrote: > > > > Since do_shrink_slab() can reschedule, we cannot protect shrinker_list > > > > using one RCU section. But using atomic_inc()/atomic_dec() for each > > > > do_shrink_slab() call will not impact so much. > > > > > > But you could use SRCU.. > > > > Davidlohr has tried that already http://lkml.kernel.org/r/1434398602.1903.15.camel@stgolabs.net > > and failed. Doing SRCU inside core kernel is not seen with a great > > support... > > I can't actually find any objection in that thread. What's the problem > Davidlohr ran into?
The patch has been dropped because allnoconfig failed to compile back then http://lkml.kernel.org/r/CAP=VYLr0rPWi1aeuk4w1On9CYRNmnEWwJgGtaX=wEvGaBURtrg@mail.gmail.com I have problem to find the follow up discussion though. The main argument was that SRC is not generally available and so the core kernel should rely on it. -- Michal Hocko SUSE Labs
| |