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 S1754796AbcBPKIq (ORCPT <rfc822;w@1wt.eu>);
	Tue, 16 Feb 2016 05:08:46 -0500
Received: from mga01.intel.com ([192.55.52.88]:13311 "EHLO mga01.intel.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1754538AbcBPKIn (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Tue, 16 Feb 2016 05:08:43 -0500
X-ExtLoop1: 1
X-IronPort-AV: E=Sophos;i="5.22,454,1449561600"; 
   d="scan'208";a="885890511"
Date: Tue, 16 Feb 2016 13:08:38 +0300
From: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
To: Dave Hansen <dave.hansen@intel.com>
Cc: Hugh Dickins <hughd@google.com>, Andrea Arcangeli <aarcange@redhat.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Vlastimil Babka <vbabka@suse.cz>, Christoph Lameter <cl@gentwo.org>,
        Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>,
        Jerome Marchand <jmarchan@redhat.com>, Yang Shi <yang.shi@linaro.org>,
        Sasha Levin <sasha.levin@oracle.com>, linux-kernel@vger.kernel.org,
        linux-mm@kvack.org
Subject: Re: [PATCHv2 15/28] thp: handle file COW faults
Message-ID: <20160216100838.GD46557@black.fi.intel.com>
References: <1455200516-132137-1-git-send-email-kirill.shutemov@linux.intel.com>
 <1455200516-132137-16-git-send-email-kirill.shutemov@linux.intel.com>
 <56BE2629.90001@intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <56BE2629.90001@intel.com>
User-Agent: Mutt/1.5.23.1 (2014-03-12)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Fri, Feb 12, 2016 at 10:36:25AM -0800, Dave Hansen wrote:
> On 02/11/2016 06:21 AM, Kirill A. Shutemov wrote:
> > File COW for THP is handled on pte level: just split the pmd.
> 
> More changelog.  More comments, please.

Okay, I'll add more.

> We don't want to COW THP's because we'll waste memory?  A COW that we
> could handle with 4k, we would have to handle with 2M, and that's
> inefficient and high-latency?

All of above.i

It's not clear how benefitial THP file COW mappings. And it would require
some code to make them work.

I think at some point we can consider teaching khugepaged to collapse such
pages, but allocating huge on fault is probably overkill.

> Seems like a good idea to me.  It would just be nice to ensure every
> reviewer doesn't have to think their way through it.

-- 
 Kirill A. Shutemov