From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753697AbbCWNCJ (ORCPT ); Mon, 23 Mar 2015 09:02:09 -0400 Received: from cantor2.suse.de ([195.135.220.15]:38661 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753676AbbCWNCF (ORCPT ); Mon, 23 Mar 2015 09:02:05 -0400 Date: Mon, 23 Mar 2015 14:02:01 +0100 From: Michal Hocko To: Dave Chinner Cc: Rik van Riel , Andrew Morton , Al Viro , Johannes Weiner , Mel Gorman , Neil Brown , Tetsuo Handa , Sage Weil , Mark Fasheh , linux-mm@kvack.org, LKML Subject: Re: [PATCH] mm: Use GFP_KERNEL allocation for the page cache in page_cache_read Message-ID: <20150323130201.GB29084@dhcp22.suse.cz> References: <1426687766-518-1-git-send-email-mhocko@suse.cz> <55098F3B.7070000@redhat.com> <20150318145528.GK17241@dhcp22.suse.cz> <20150319071439.GE28621@dastard> <20150319124441.GC12466@dhcp22.suse.cz> <20150320034820.GH28621@dastard> <20150320131453.GA4821@dhcp22.suse.cz> <20150320225139.GL28621@dastard> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150320225139.GL28621@dastard> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sat 21-03-15 09:51:39, Dave Chinner wrote: > On Fri, Mar 20, 2015 at 02:14:53PM +0100, Michal Hocko wrote: > > On Fri 20-03-15 14:48:20, Dave Chinner wrote: > > > On Thu, Mar 19, 2015 at 01:44:41PM +0100, Michal Hocko wrote: > > > > > allocations where the caller cannot pass a gfp_mask need to obey > > > > > the mapping_gfp_mask that is set by the mapping owner.... > > > > > > > > Hmm, I thought this is true only when the function might be called from > > > > the fs path. > > > > > > How do you know in, say, mpage_readpages, you aren't being called > > > from a fs path that holds locks? e.g. we can get there from ext4 > > > doing readdir, so it is holding an i_mutex lock at that point. > > > > > > Many other paths into mpages_readpages don't hold locks, but there > > > are some that do, and those that do need functionals like this to > > > obey the mapping_gfp_mask because it is set appropriately for the > > > allocation context of the inode that owns the mapping.... > > > > What about the following? > > --- > > From 5d905cb291138d61bbab056845d6e53bc4451ec8 Mon Sep 17 00:00:00 2001 > > From: Michal Hocko > > Date: Thu, 19 Mar 2015 14:56:56 +0100 > > Subject: [PATCH 1/2] mm: do not ignore mapping_gfp_mask in page cache > > allocation paths > > Looks reasonable, though I though there were more places that that > in the mapping paths that need to be careful... I have focused on those which involve page cache allocation because those are obvious. We might need others but I do not see them right now. I will include this patch for the next submit after I manage to wrap my head around up-coming xfs changes and come up with something for page_cache_read vs OOM killer issue. -- Michal Hocko SUSE Labs