From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755936AbaFQPpi (ORCPT ); Tue, 17 Jun 2014 11:45:38 -0400 Received: from zene.cmpxchg.org ([85.214.230.12]:37062 "EHLO zene.cmpxchg.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755401AbaFQPpg (ORCPT ); Tue, 17 Jun 2014 11:45:36 -0400 Date: Tue, 17 Jun 2014 11:45:27 -0400 From: Johannes Weiner To: Michal Hocko Cc: Andrew Morton , Hugh Dickins , Tejun Heo , Vladimir Davydov , cgroups@vger.kernel.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: Re: [patch 04/12] mm: memcontrol: retry reclaim for oom-disabled and __GFP_NOFAIL charges Message-ID: <20140617154527.GC7331@cmpxchg.org> References: <1402948472-8175-1-git-send-email-hannes@cmpxchg.org> <1402948472-8175-5-git-send-email-hannes@cmpxchg.org> <20140617135344.GC19886@dhcp22.suse.cz> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140617135344.GC19886@dhcp22.suse.cz> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Jun 17, 2014 at 03:53:44PM +0200, Michal Hocko wrote: > On Mon 16-06-14 15:54:24, Johannes Weiner wrote: > > There is no reason why oom-disabled and __GFP_NOFAIL charges should > > try to reclaim only once when every other charge tries several times > > before giving up. Make them all retry the same number of times. > > OK, this makes sense for oom-disabled and __GFP_NOFAIL but does it make > sense to do additional reclaim for tasks with fatal_signal_pending? > > It is little bit unexpected, because we bypass if the condition happens > before the reclaim but then we ignore it. "mm: memcontrol: rearrange charging fast path", moves the pending signal check inside the retry block, right before reclaim.