From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-oa0-f41.google.com (mail-oa0-f41.google.com [209.85.219.41]) by kanga.kvack.org (Postfix) with ESMTP id D88836B0036 for ; Wed, 7 May 2014 13:34:38 -0400 (EDT) Received: by mail-oa0-f41.google.com with SMTP id m1so1670737oag.0 for ; Wed, 07 May 2014 10:34:38 -0700 (PDT) Received: from g4t3426.houston.hp.com (g4t3426.houston.hp.com. [15.201.208.54]) by mx.google.com with ESMTPS id ta10si11301285obc.150.2014.05.07.10.34.37 for (version=TLSv1 cipher=RC4-SHA bits=128/128); Wed, 07 May 2014 10:34:38 -0700 (PDT) Message-ID: <1399484069.4567.5.camel@buesod1.americas.hpqcorp.net> Subject: Re: [RFC] Heterogeneous memory management (mirror process address space on a device mmu). From: Davidlohr Bueso Date: Wed, 07 May 2014 10:34:29 -0700 In-Reply-To: <20140507130032.GM30445@twins.programming.kicks-ass.net> References: <1399038730-25641-1-git-send-email-j.glisse@gmail.com> <20140506102925.GD11096@twins.programming.kicks-ass.net> <1399429987.2581.25.camel@buesod1.americas.hpqcorp.net> <20140507130032.GM30445@twins.programming.kicks-ass.net> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: Peter Zijlstra Cc: j.glisse@gmail.com, linux-mm@kvack.org, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, Mel Gorman , "H. Peter Anvin" , Andrew Morton , Linda Wang , Kevin E Martin , Jerome Glisse , Andrea Arcangeli , Johannes Weiner , Larry Woodman , Rik van Riel , Dave Airlie , Jeff Law , Brendan Conoboy , Joe Donohue , Duncan Poole , Sherry Cheung , Subhash Gutti , John Hubbard , Mark Hairgrove , Lucien Dunning , Cameron Buschardt , Arvind Gopalakrishnan , Haggai Eran , Or Gerlitz , Sagi Grimberg , Shachar Raindel , Liran Liss , Roland Dreier , "Sander, Ben" , "Stoner, Greg" , "Bridgman, John" , "Mantor, Michael" , "Blinzer, Paul" , "Morichetti, Laurent" , "Deucher, Alexander" , "Gabbay, Oded" , Linus Torvalds On Wed, 2014-05-07 at 15:00 +0200, Peter Zijlstra wrote: > On Tue, May 06, 2014 at 07:33:07PM -0700, Davidlohr Bueso wrote: > > > So I've been running benchmarks (mostly aim7, which nicely exercises our > > locks) comparing my recent v4 for rwsem optimistic spinning against > > previous implementation ideas for the anon-vma lock, mostly: > > > - rwlock_t > > - qrwlock_t > > Which reminds me; can you provide the numbers for rwlock_t vs qrwlock_t > in a numeric form so I can include them in the qrwlock_t changelog. Ah, right. I was lazy and just showed you the graphs. > That way I can queue those patches for inclusion, I think we want a fair > rwlock_t if we can show (and you graphs do iirc) that it doesn't cost us > performance. I agree, fairness is much welcome here. And I agree that despite my good numbers, and that we should keep the anon vma lock as a rwsem, its still worth merging the qrwlock stuff. I'll cookup my regular numeric table today. Thanks, Davidlohr -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org