From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754663AbdBNQat (ORCPT ); Tue, 14 Feb 2017 11:30:49 -0500 Received: from gum.cmpxchg.org ([85.214.110.215]:35230 "EHLO gum.cmpxchg.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753734AbdBNQal (ORCPT ); Tue, 14 Feb 2017 11:30:41 -0500 Date: Tue, 14 Feb 2017 11:30:05 -0500 From: Johannes Weiner To: Aleksa Sarai Cc: Ingo Molnar , Peter Zijlstra , Andrew Morton , Michal Hocko , Oleg Nesterov , linux-kernel@vger.kernel.org, linux-mm@kvack.org, cyphar@cyphar.com Subject: Re: [PATCH] oom_reaper: switch to struct list_head for reap queue Message-ID: <20170214163005.GA2450@cmpxchg.org> References: <20170214150714.6195-1-asarai@suse.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170214150714.6195-1-asarai@suse.de> User-Agent: Mutt/1.7.2 (2016-11-26) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Feb 15, 2017 at 02:07:14AM +1100, Aleksa Sarai wrote: > Rather than implementing an open addressing linked list structure > ourselves, use the standard list_head structure to improve consistency > with the rest of the kernel and reduce confusion. > > Cc: Michal Hocko > Cc: Oleg Nesterov > Signed-off-by: Aleksa Sarai > --- > include/linux/sched.h | 6 +++++- > kernel/fork.c | 4 ++++ > mm/oom_kill.c | 24 +++++++++++++----------- > 3 files changed, 22 insertions(+), 12 deletions(-) > > diff --git a/include/linux/sched.h b/include/linux/sched.h > index e93594b88130..d8bcd0f8c5fe 100644 > --- a/include/linux/sched.h > +++ b/include/linux/sched.h > @@ -1960,7 +1960,11 @@ struct task_struct { > #endif > int pagefault_disabled; > #ifdef CONFIG_MMU > - struct task_struct *oom_reaper_list; > + /* > + * List of threads that have to be reaped by OOM (rooted at > + * &oom_reaper_list in mm/oom_kill.c). > + */ > + struct list_head oom_reaper_list; This is an extra pointer to task_struct and more lines of code to accomplish the same thing. Why would we want to do that?