Re: [PATCH 5/6] mm/page_alloc: Limit the number of pages on PCP lists when reclaim is active

From: Mel Gorman <mgorman@techsingularity.net>
To: Dave Hansen <dave.hansen@intel.com>
Cc: Linux-MM <linux-mm@kvack.org>,
	Dave Hansen <dave.hansen@linux.intel.com>,
	Matthew Wilcox <willy@infradead.org>,
	Vlastimil Babka <vbabka@suse.cz>,
	Michal Hocko <mhocko@kernel.org>,
	Nicholas Piggin <npiggin@gmail.com>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 5/6] mm/page_alloc: Limit the number of pages on PCP lists when reclaim is active
Date: Mon, 24 May 2021 10:22:25 +0100	[thread overview]
Message-ID: <20210524092225.GD30378@techsingularity.net> (raw)
In-Reply-To: <3f0ebd90-1aca-1dfc-3b92-bdb991d0fb29@intel.com>

On Fri, May 21, 2021 at 03:44:49PM -0700, Dave Hansen wrote:
> On 5/21/21 3:28 AM, Mel Gorman wrote:
> > +static int nr_pcp_high(struct per_cpu_pages *pcp, struct zone *zone)
> > +{
> > +	int high = READ_ONCE(pcp->high);
> > +
> > +	if (unlikely(!high))
> > +		return 0;
> > +
> > +	if (!test_bit(ZONE_RECLAIM_ACTIVE, &zone->flags))
> > +		return high;
> > +
> > +	/*
> > +	 * If reclaim is active, limit the number of pages that can be
> > +	 * stored on pcp lists
> > +	 */
> > +	return READ_ONCE(pcp->batch) << 2;
> > +}
> 
> Should there be a sanity check on this?  Let's say we had one of those
> weirdo zones with tons of CPUs and a small low_wmark_pages().  Could we
> have a case where:
> 
> 	pcp->high < pcp->batch<<2
> 
> and this effectively *raises* nr_pcp_high()?
> 
> It's not possible with the current pcp->high calculation, but does
> anything prevent it now?

I don't think it would happen as pcp->batch is reduced for small zones
but a sanity check does not hurt so I added one.

-- 
Mel Gorman
SUSE Labs