From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S932568AbcHKP5Z (ORCPT <rfc822;w@1wt.eu>);
	Thu, 11 Aug 2016 11:57:25 -0400
Received: from verein.lst.de ([213.95.11.211]:42202 "EHLO newverein.lst.de"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S932375AbcHKP5X (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Thu, 11 Aug 2016 11:57:23 -0400
Date: Thu, 11 Aug 2016 17:57:21 +0200
From: Christoph Hellwig <hch@lst.de>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: "Huang, Ying" <ying.huang@intel.com>, Dave Chinner <david@fromorbit.com>,
        LKML <linux-kernel@vger.kernel.org>,
        Bob Peterson <rpeterso@redhat.com>,
        Wu Fengguang <fengguang.wu@intel.com>, LKP <lkp@01.org>,
        Christoph Hellwig <hch@lst.de>
Subject: Re: [LKP] [lkp] [xfs] 68a9f5e700: aim7.jobs-per-min -13.6%
	regression
Message-ID: <20160811155721.GA23015@lst.de>
References: <20160809143359.GA11220@yexl-desktop> <CA+55aFyJaT3ufm7kfU=PGi0YtHzBEYYLxcA+PUrka8uQ3=5+bg@mail.gmail.com> <20160810230840.GS16044@dastard> <CA+55aFyR3DxeQimeU+j4gEXku0WJhEuDZPr7PNSDRUQHTaTAXA@mail.gmail.com> <87eg5w18iu.fsf@yhuang-mobile.sh.intel.com> <87a8gk17x7.fsf@yhuang-mobile.sh.intel.com> <CA+55aFyQ6cjKOQZKEuQV9ch+aerW1Q6uGdv+pCS81rX0yjyR6w@mail.gmail.com> <8760r816wf.fsf@yhuang-mobile.sh.intel.com> <CA+55aFy=xeEKgfHWisfxsZNqrgMJxvKvhrWrPmdpMxu0pnOXig@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CA+55aFy=xeEKgfHWisfxsZNqrgMJxvKvhrWrPmdpMxu0pnOXig@mail.gmail.com>
User-Agent: Mutt/1.5.17 (2007-11-01)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Wed, Aug 10, 2016 at 06:00:24PM -0700, Linus Torvalds wrote:
> The biggest difference is that we have "mark_page_accessed()" show up
> after, and not before. There was also a lot of LRU noise in the
> non-profile data. I wonder if that is the reason here: the old model
> of using generic_perform_write/block_page_mkwrite didn't mark the
> pages accessed, and now with iomap_file_buffered_write() they get
> marked as active and that screws up the LRU list, and makes us not
> flush out the dirty pages well (because they are seen as active and
> not good for writeback), and then you get bad memory use.

And that's actually a "bug" in the new code - mostly because I failed
to pick up changes to the core code happening after we 'forked' it,
in this case commit 2457ae ("mm: non-atomically mark page accessed during page
cache allocation where possible").

The one liner below (not tested yet) to simply remove it should fix that
up.  I also noticed we have a spurious pagefault_disable/enable, I
need to dig into the history of that first, though.

diff --git a/fs/iomap.c b/fs/iomap.c
index 48141b8..f39c318 100644
--- a/fs/iomap.c
+++ b/fs/iomap.c
@@ -199,7 +199,6 @@ again:
 		pagefault_enable();
 
 		flush_dcache_page(page);
-		mark_page_accessed(page);
 
 		status = iomap_write_end(inode, pos, bytes, copied, page);
 		if (unlikely(status < 0))