From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1423996AbcFMNzJ (ORCPT ); Mon, 13 Jun 2016 09:55:09 -0400 Received: from mail-wm0-f65.google.com ([74.125.82.65]:34417 "EHLO mail-wm0-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1423391AbcFMNzF (ORCPT ); Mon, 13 Jun 2016 09:55:05 -0400 Date: Mon, 13 Jun 2016 15:55:02 +0200 From: Michal Hocko To: Ebru Akagunduz Cc: linux-mm@kvack.org, hughd@google.com, riel@redhat.com, akpm@linux-foundation.org, kirill.shutemov@linux.intel.com, n-horiguchi@ah.jp.nec.com, aarcange@redhat.com, iamjoonsoo.kim@lge.com, gorcunov@openvz.org, linux-kernel@vger.kernel.org, mgorman@suse.de, rientjes@google.com, vbabka@suse.cz, aneesh.kumar@linux.vnet.ibm.com, hannes@cmpxchg.org, boaz@plexistor.com Subject: Re: [PATCH 0/3] mm, thp: remove duplication and fix locking issues in swapin Message-ID: <20160613135502.GI6518@dhcp22.suse.cz> References: <1464023651-19420-1-git-send-email-ebru.akagunduz@gmail.com> <20160523172929.GA4406@debian> <20160527131247.GM27686@dhcp22.suse.cz> <20160611192106.GA6662@debian> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20160611192106.GA6662@debian> User-Agent: Mutt/1.6.0 (2016-04-01) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sat 11-06-16 22:21:06, Ebru Akagunduz wrote: > On Fri, May 27, 2016 at 03:12:47PM +0200, Michal Hocko wrote: [...] > > IMHO we should drop the current ALLOCSTALL heuristic and replace it with > > ~__GFP_DIRECT_RECLAIM. > > Actually, I don't lean towards to touch do_swap_page giving gfp parameter. > do_swap_page is also used by do_page_fault, it can cause many side effect > that I can't see. The point of the patch is to keep the current gfp mask from all places except for the optimistic swapin. This was what my example patch did AFAIR. > I've just sent a patch series for converting from optimistic to > conservative and take back allocstall. Maybe that way can be easier to > be approved and less problemitical. I will try to have a look but I will be travelling this week so cannot promise anything. -- Michal Hocko SUSE Labs