From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753212AbaEUT5b (ORCPT ); Wed, 21 May 2014 15:57:31 -0400 Received: from mail-la0-f50.google.com ([209.85.215.50]:34359 "EHLO mail-la0-f50.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751137AbaEUT5a (ORCPT ); Wed, 21 May 2014 15:57:30 -0400 Date: Wed, 21 May 2014 23:57:23 +0400 From: Cyrill Gorcunov To: Andrew Morton Cc: "Kirill A. Shutemov" , linux-kernel@vger.kernel.org, linux-mm@kvack.org, Andrea Arcangeli , Pavel Emelyanov , Dave Hansen Subject: Re: [PATCH] mm: /prom/pid/clear_refs: avoid split_huge_page() Message-ID: <20140521195723.GD12819@moon> References: <1400699062-20123-1-git-send-email-kirill.shutemov@linux.intel.com> <20140521123446.ae45fa676cae27fffbd96cfd@linux-foundation.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140521123446.ae45fa676cae27fffbd96cfd@linux-foundation.org> 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 Wed, May 21, 2014 at 12:34:46PM -0700, Andrew Morton wrote: > On Wed, 21 May 2014 22:04:22 +0300 "Kirill A. Shutemov" wrote: > > > Currently we split all THP pages on any clear_refs request. It's not > > necessary. We can handle this on PMD level. > > > > One side effect is that soft dirty will potentially see more dirty > > memory, since we will mark whole THP page dirty at once. > > This clashes pretty badly with > http://ozlabs.org/~akpm/mmots/broken-out/clear_refs-redefine-callback-functions-for-page-table-walker.patch > > > Sanity checked with CRIU test suite. More testing is required. > > Will you be doing that testing or was this a request for Cyrill & co to > help? We've talking to Kirill how to test is and end up that criu is the best candidate (though I think I'll write selftest for vanilla too, hopefully tomorrow).