Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758128Ab1DNCLI (ORCPT ); Wed, 13 Apr 2011 22:11:08 -0400 Received: from mail-wy0-f174.google.com ([74.125.82.174]:39175 "EHLO mail-wy0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752580Ab1DNCLF (ORCPT ); Wed, 13 Apr 2011 22:11:05 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=subject:from:to:cc:in-reply-to:references:content-type:date :message-id:mime-version:x-mailer:content-transfer-encoding; b=LFleSoMsX03nfQMh5p1iPvGhaZfIkI6hhz/Ur1GOiJzupn5jT/mkkVQPSyfzQNKtaD BDq5ZKesYdB2O3JwQc2QCFNPKP3f1tIrN51pbXg1UwTspMvb8Bvq4mnMyqg9c/UxrPfk cYB1DpYmkQN4KjQ070v7koILL4D2NicIb2Olw= Subject: Re: Regression from 2.6.36 From: Eric Dumazet To: Andrew Morton Cc: Changli Gao , =?ISO-8859-1?Q?Am=E9rico?= Wang , Jiri Slaby , azurIt , linux-kernel@vger.kernel.org, linux-mm@kvack.org, linux-fsdevel@vger.kernel.org, Jiri Slaby , Mel Gorman In-Reply-To: <20110413141600.28793661.akpm@linux-foundation.org> References: <20110315132527.130FB80018F1@mail1005.cent> <20110317001519.GB18911@kroah.com> <20110407120112.E08DCA03@pobox.sk> <4D9D8FAA.9080405@suse.cz> <1302177428.3357.25.camel@edumazet-laptop> <1302178426.3357.34.camel@edumazet-laptop> <1302190586.3357.45.camel@edumazet-laptop> <20110412154906.70829d60.akpm@linux-foundation.org> <20110412183132.a854bffc.akpm@linux-foundation.org> <1302662256.2811.27.camel@edumazet-laptop> <20110413141600.28793661.akpm@linux-foundation.org> Content-Type: text/plain; charset="UTF-8" Date: Thu, 14 Apr 2011 04:10:58 +0200 Message-ID: <1302747058.3549.7.camel@edumazet-laptop> Mime-Version: 1.0 X-Mailer: Evolution 2.30.3 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2660 Lines: 86 Le mercredi 13 avril 2011 à 14:16 -0700, Andrew Morton a écrit : > So am I correct in believing that this regression is due to the > high-order allocations putting excess stress onto page reclaim? > I suppose so. > If so, then how large _are_ these allocations? This perhaps can be > determined from /proc/slabinfo. They must be pretty huge, because slub > likes to do excessively-large allocations and the system handles that > reasonably well. > > I suppose that a suitable fix would be > > > From: Andrew Morton > > Azurit reports large increases in system time after 2.6.36 when running > Apache. It was bisected down to a892e2d7dcdfa6c76e6 ("vfs: use kmalloc() > to allocate fdmem if possible"). > > That patch caused the vfs to use kmalloc() for very large allocations and > this is causing excessive work (and presumably excessive reclaim) within > the page allocator. > > Fix it by falling back to vmalloc() earlier - when the allocation attempt > would have been considered "costly" by reclaim. > > Reported-by: azurIt > Cc: Changli Gao > Cc: Americo Wang > Cc: Jiri Slaby > Cc: Eric Dumazet > Cc: Mel Gorman > Signed-off-by: Andrew Morton > --- > > fs/file.c | 17 ++++++++++------- > 1 file changed, 10 insertions(+), 7 deletions(-) > > diff -puN fs/file.c~a fs/file.c > --- a/fs/file.c~a > +++ a/fs/file.c > @@ -39,14 +39,17 @@ int sysctl_nr_open_max = 1024 * 1024; /* > */ > static DEFINE_PER_CPU(struct fdtable_defer, fdtable_defer_list); > > -static inline void *alloc_fdmem(unsigned int size) > +static void *alloc_fdmem(unsigned int size) > { > - void *data; > - > - data = kmalloc(size, GFP_KERNEL|__GFP_NOWARN); > - if (data != NULL) > - return data; > - > + /* > + * Very large allocations can stress page reclaim, so fall back to > + * vmalloc() if the allocation size will be considered "large" by the VM. > + */ > + if (size <= (PAGE_SIZE << PAGE_ALLOC_COSTLY_ORDER) { > + void *data = kmalloc(size, GFP_KERNEL|__GFP_NOWARN); > + if (data != NULL) > + return data; > + } > return vmalloc(size); > } > > _ > Acked-by: Eric Dumazet #define PAGE_ALLOC_COSTLY_ORDER 3 On x86_64, this means we try kmalloc() up to 4096 files in fdtable. Thanks ! -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/