Re: [PATCH 1/2] mm,vmscan: Kill global shrinker lock.

From: Shakeel Butt <shakeelb@google.com>
To: Christoph Hellwig <hch@infradead.org>
Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>,
	Minchan Kim <minchan@kernel.org>,
	Huang Ying <ying.huang@intel.com>,
	Mel Gorman <mgorman@techsingularity.net>,
	Vladimir Davydov <vdavydov.dev@gmail.com>,
	Michal Hocko <mhocko@kernel.org>,
	Johannes Weiner <hannes@cmpxchg.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Greg Thelen <gthelen@google.com>, Linux MM <linux-mm@kvack.org>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 1/2] mm,vmscan: Kill global shrinker lock.
Date: Fri, 17 Nov 2017 09:53:22 -0800	[thread overview]
Message-ID: <CALvZod5b8EBgJ=jTWfAs97zn3D9WPDP5j-2qAR5FGEYrn0GM6Q@mail.gmail.com> (raw)
In-Reply-To: <CALvZod7Mrs8=5A2j=x96vaUcjCMSxVYi6RVLaKF23UENcAPLvw@mail.gmail.com>

On Fri, Nov 17, 2017 at 9:41 AM, Shakeel Butt <shakeelb@google.com> wrote:
> On Fri, Nov 17, 2017 at 9:35 AM, Christoph Hellwig <hch@infradead.org> 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..
>
> I looked into that but was advised to not go through that route due to
> SRCU behind the CONFIG_SRCU. However now I see the precedence of
> "#ifdef CONFIG_SRCU" in drivers/base/core.c and I think if we can take
> that route if even after Minchan's patch the issue persists.

Too many 'ifs' in the last sentence. I just wanted to say we can
consider SRCU if the issue persists even after Minchan's patch.