Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752662AbbFYVhw (ORCPT ); Thu, 25 Jun 2015 17:37:52 -0400 Received: from relay2.sgi.com ([192.48.180.65]:43599 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751746AbbFYVhp (ORCPT ); Thu, 25 Jun 2015 17:37:45 -0400 Date: Thu, 25 Jun 2015 16:37:43 -0500 From: Nathan Zimmer To: Mel Gorman Cc: Nathan Zimmer , Daniel J Blueman , Andrew Morton , Waiman Long , Dave Hansen , Scott Norton , Linux-MM , LKML , Steffen Persvold Subject: Re: [PATCH] mm: meminit: Finish initialisation of struct pages before basic setup Message-ID: <20150625213743.GB129272@asylum.americas.sgi.com> References: <20150513163157.GR2462@suse.de> <1431597783.26797.1@cpanel21.proisp.no> <20150624225028.GA97166@asylum.americas.sgi.com> <20150625204855.GC26927@suse.de> <20150625205744.GE26927@suse.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150625205744.GE26927@suse.de> User-Agent: Mutt/1.5.17 (2007-11-01) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1364 Lines: 33 On Thu, Jun 25, 2015 at 09:57:44PM +0100, Mel Gorman wrote: > On Thu, Jun 25, 2015 at 09:48:55PM +0100, Mel Gorman wrote: > > On Wed, Jun 24, 2015 at 05:50:28PM -0500, Nathan Zimmer wrote: > > > My apologies for taking so long to get back to this. > > > > > > I think I did locate two potential sources of slowdown. > > > One is the set_cpus_allowed_ptr as I have noted previously. > > > However I only notice that on the very largest boxes. > > > I did cobble together a patch that seems to help. > > > > > > > If you are using kthread_create_on_node(), is it even necessary to call > > set_cpus_allowed_ptr() at all? > > > > That aside, are you aware of any failure with this series as it currently > stands in Andrew's tree that this patch is meant to address? It seems > like a nice follow-on that would boot faster on very large machines but > if it's addressing a regression then it's very important as the series > cannot be merged with known critical failures. > Nope I haven't recorded any failures without it. I just get concerned when I see some scaling issues that something COULD go wrong. Nate -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/