Re: [PATCH v2 4/4] mm/memcg: Protect memcg_stock with a local_lock_t

From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
To: Johannes Weiner <hannes@cmpxchg.org>
Cc: cgroups@vger.kernel.org, linux-mm@kvack.org,
	"Andrew Morton" <akpm@linux-foundation.org>,
	"Michal Hocko" <mhocko@kernel.org>,
	"Michal Koutný" <mkoutny@suse.com>,
	"Peter Zijlstra" <peterz@infradead.org>,
	"Thomas Gleixner" <tglx@linutronix.de>,
	"Vladimir Davydov" <vdavydov.dev@gmail.com>,
	"Waiman Long" <longman@redhat.com>,
	"kernel test robot" <oliver.sang@intel.com>
Subject: Re: [PATCH v2 4/4] mm/memcg: Protect memcg_stock with a local_lock_t
Date: Thu, 17 Feb 2022 10:28:57 +0100	[thread overview]
Message-ID: <Yg4VWfeCLGFnjQ1W@linutronix.de> (raw)
In-Reply-To: <Yg09t/j5Z0X9L7aX@cmpxchg.org>

On 2022-02-16 13:08:55 [-0500], Johannes Weiner wrote:
> I thought you'd fold it into yours, but separate patch works too,
> thanks!
> 
> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>

Thank you.

> One important note, though:
> 
> > @@ -3151,8 +3155,17 @@ static void drain_obj_stock(struct memcg_stock_pcp *stock)
> >  		unsigned int nr_pages = stock->nr_bytes >> PAGE_SHIFT;
> >  		unsigned int nr_bytes = stock->nr_bytes & (PAGE_SIZE - 1);
> >  
> > -		if (nr_pages)
> > -			obj_cgroup_uncharge_pages(old, nr_pages);
> > +		if (nr_pages) {
> > +			struct mem_cgroup *memcg;
> > +
> > +			memcg = get_mem_cgroup_from_objcg(old);
> > +
> > +			if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))
> > +				page_counter_uncharge(&memcg->kmem, nr_pages);
> > +			__refill_stock(memcg, nr_pages);
> 
> This doesn't take "memcg: add per-memcg total kernel memory stat"
> queued in -mm into account and so will break kmem accounting.
>
> Make sure to rebase the patches to the -mm tree before sending it
> out. You can find it here: https://github.com/hnaz/linux-mm

Thank you, rebased on-top.

Sebastian