From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1763421AbYEFAoM (ORCPT ); Mon, 5 May 2008 20:44:12 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1757691AbYEFAn5 (ORCPT ); Mon, 5 May 2008 20:43:57 -0400 Received: from mx1.redhat.com ([66.187.233.31]:54693 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756997AbYEFAn5 (ORCPT ); Mon, 5 May 2008 20:43:57 -0400 Date: Mon, 5 May 2008 20:43:18 -0400 From: Rik van Riel To: "KOSAKI Motohiro" Cc: LKML , linux-mm , "Andrew Morton" Subject: Re: [-mm][PATCH 4/5] core of reclaim throttle Message-ID: <20080505204318.3f95c83c@bree.surriel.com> In-Reply-To: <2f11576a0805051523h730fce0foa51f1fdbf9c46cbe@mail.gmail.com> References: <20080504201343.8F52.KOSAKI.MOTOHIRO@jp.fujitsu.com> <20080504215819.8F5E.KOSAKI.MOTOHIRO@jp.fujitsu.com> <20080504221043.8F64.KOSAKI.MOTOHIRO@jp.fujitsu.com> <20080505175142.7de3f27b@cuia.bos.redhat.com> <2f11576a0805051523h730fce0foa51f1fdbf9c46cbe@mail.gmail.com> Organization: Red Hat, Inc. X-Mailer: Claws Mail 3.0.2 (GTK+ 2.10.4; 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 Tue, 6 May 2008 07:23:18 +0900 "KOSAKI Motohiro" wrote: > hmmm, AFAIK, > on current kernel, sometimes __GFP_IO task wait for non __GFP_IO task > by lock_page(). > Is this wrong? This is fine. The problem is adding a code path that causes non __GFP_IO tasks to wait on __GFP_IO tasks. Then you can have a deadlock. > therefore my patch care only recursive reclaim situation. > I don't object to your opinion. but I hope understand exactly your opinion. I believe not all non __GFP_IO or non __GFP_FS calls are recursive reclaim, but there are some other code paths too. For example from fs/buffer.c -- All rights reversed.