From: Vlastimil Babka Subject: Re: [PATCH 6/6] mm: page_alloc: Reduce cost of the fair zone allocation policy Date: Mon, 11 Aug 2014 14:34:05 +0200 Message-ID: <53E8B83D.1070004@suse.cz> References: <1404893588-21371-1-git-send-email-mgorman@suse.de> <1404893588-21371-7-git-send-email-mgorman@suse.de> <53E4EC53.1050904@suse.cz> <20140811121241.GD7970@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Cc: Andrew Morton , Linux Kernel , Linux-MM , Linux-FSDevel , Johannes Weiner To: Mel Gorman Return-path: In-Reply-To: <20140811121241.GD7970@suse.de> Sender: owner-linux-mm@kvack.org List-Id: linux-fsdevel.vger.kernel.org On 08/11/2014 02:12 PM, Mel Gorman wrote: > On Fri, Aug 08, 2014 at 05:27:15PM +0200, Vlastimil Babka wrote: >> On 07/09/2014 10:13 AM, Mel Gorman wrote: >>> --- a/mm/page_alloc.c >>> +++ b/mm/page_alloc.c >>> @@ -1604,6 +1604,9 @@ again: >>> } >>> >>> __mod_zone_page_state(zone, NR_ALLOC_BATCH, -(1 << order)); >> >> This can underflow zero, right? >> > > Yes, because of per-cpu accounting drift. I meant mainly because of order > 0. >>> + if (zone_page_state(zone, NR_ALLOC_BATCH) == 0 && >> >> AFAICS, zone_page_state will correct negative values to zero only for >> CONFIG_SMP. Won't this check be broken on !CONFIG_SMP? >> > > On !CONFIG_SMP how can there be per-cpu accounting drift that would make > that counter negative? Well original code used "if (zone_page_state(zone, NR_ALLOC_BATCH) <= 0)" elsewhere, that you are replacing with zone_is_fair_depleted check. I assumed it's because it can get negative due to order > 0. I might have not looked thoroughly enough but it seems to me there's nothing that would prevent it, such as skipping a zone because its remaining batch is lower than 1 << order. So I think the check should be "<= 0" to be safe. Vlastimil -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org