From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751836AbZGPEDR (ORCPT ); Thu, 16 Jul 2009 00:03:17 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751725AbZGPEDQ (ORCPT ); Thu, 16 Jul 2009 00:03:16 -0400 Received: from smtp1.linux-foundation.org ([140.211.169.13]:38195 "EHLO smtp1.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751675AbZGPEDO (ORCPT ); Thu, 16 Jul 2009 00:03:14 -0400 Date: Wed, 15 Jul 2009 21:02:53 -0700 From: Andrew Morton To: Rik van Riel Cc: KOSAKI Motohiro , LKML , linux-mm , Wu Fengguang Subject: Re: [PATCH -mm] throttle direct reclaim when too many pages are isolated already (v3) Message-Id: <20090715210253.bc137b2d.akpm@linux-foundation.org> In-Reply-To: <20090715235318.6d2f5247@bree.surriel.com> References: <20090715223854.7548740a@bree.surriel.com> <20090715194820.237a4d77.akpm@linux-foundation.org> <4A5E9A33.3030704@redhat.com> <20090715202114.789d36f7.akpm@linux-foundation.org> <4A5E9E4E.5000308@redhat.com> <20090715203854.336de2d5.akpm@linux-foundation.org> <20090715235318.6d2f5247@bree.surriel.com> X-Mailer: Sylpheed 2.4.8 (GTK+ 2.12.5; x86_64-redhat-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 15 Jul 2009 23:53:18 -0400 Rik van Riel wrote: > @@ -1049,6 +1074,14 @@ static unsigned long shrink_inactive_lis > struct zone_reclaim_stat *reclaim_stat = get_reclaim_stat(zone, sc); > int lumpy_reclaim = 0; > > + while (unlikely(too_many_isolated(zone, file, sc))) { > + congestion_wait(WRITE, HZ/10); > + > + /* We are about to die and free our memory. Return now. */ > + if (fatal_signal_pending(current)) > + return SWAP_CLUSTER_MAX; > + } mutter. While I agree that handling fatal signals on the direct reclaim path is probably a good thing, this seems like a fairly random place at which to start the enhancement. If we were to step back and approach this in a broader fashion, perhaps we would find some commonality with the existing TIF_MEMDIE handling, dunno. And I question the testedness of v3 :)