Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753486Ab3F2HYr (ORCPT ); Sat, 29 Jun 2013 03:24:47 -0400 Received: from mail-wi0-f176.google.com ([209.85.212.176]:32867 "EHLO mail-wi0-f176.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751513Ab3F2HYp (ORCPT ); Sat, 29 Jun 2013 03:24:45 -0400 Date: Sat, 29 Jun 2013 09:24:41 +0200 From: Ingo Molnar To: Nathan Zimmer Cc: Daniel J Blueman , Andrew Morton , Mike Travis , "H. Peter Anvin" , holt@sgi.com, rob@landley.net, Thomas Gleixner , Ingo Molnar , yinghai@kernel.org, Greg KH , x86@kernel.org, linux-doc@vger.kernel.org, Linux Kernel , Linus Torvalds , Peter Zijlstra , Steffen Persvold Subject: Re: [RFC] Transparent on-demand memory setup initialization embedded in the (GFP) buddy allocator Message-ID: <20130629072441.GA15394@gmail.com> References: <51CBB2F7.3050604@numascale-asia.com> <51CDF417.3050406@sgi.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <51CDF417.3050406@sgi.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2491 Lines: 61 * Nathan Zimmer wrote: > On 06/26/2013 10:35 PM, Daniel J Blueman wrote: > >On Wednesday, June 26, 2013 9:30:02 PM UTC+8, Andrew Morton wrote: > >> > >> On Wed, 26 Jun 2013 11:22:48 +0200 Ingo Molnar > > wrote: > >> > >> > except that on 32 TB > >> > systems we don't spend ~2 hours initializing 8,589,934,592 > >page heads. > >> > >> That's about a million a second which is crazy slow - even my > >prehistoric desktop > >> is 100x faster than that. > >> > >> Where's all this time actually being spent? > > > > The complexity of a directory-lookup architecture to make the > > (intrinsically unscalable) cache-coherency protocol scalable gives you > > a ~1us roundtrip to remote NUMA nodes. > > > > Probably a lot of time is spent in some memsets, and RMW cycles which > > are setting page bits, which are intrinsically synchronous, so the > > initialising core can't get to 12 or so outstanding memory > > transactions. > > > > Since EFI memory ranges have a flag to state if they are zerod (which > > may be a fair assumption for memory on non-bootstrap processor NUMA > > nodes), we can probably collapse the RMWs to just writes. > > > > A normal write will require a coherency cycle, then a fetch and a > > writeback when it's evicted from the cache. For this purpose, > > non-temporal writes would eliminate the cache line fetch and give a > > massive increase in bandwidth. We wouldn't even need a store-fence as > > the initialising core is the only one online. > > Could you elaborate a bit more? or suggest a specific area to look at? > > After some experiments with trying to just set some fields in the struct > page directly I haven't been able to produce any improvements. Of > course there is lots about the area which I don't have much experience > with. Any such improvement will at most be in the 10-20% range. I'd suggest first concentrating on the 1000-fold boot time initialization speedup that the buddy allocator delayed initialization can offer, and speeding up whatever remains after that stage - in a much more development-friendly environment. (You'll be able to run 'perf record ./calloc-1TB' after bootup and get meaningful results, etc.) Thanks, Ingo -- 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/