From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on archive.lwn.net X-Spam-Level: X-Spam-Status: No, score=-6.0 required=5.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,RCVD_IN_DNSWL_HI autolearn=unavailable autolearn_force=no version=3.4.2 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by archive.lwn.net (Postfix) with ESMTP id 98EEB7D8A3 for ; Mon, 7 Jan 2019 16:22:38 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1730090AbfAGQW3 (ORCPT ); Mon, 7 Jan 2019 11:22:29 -0500 Received: from mx1.redhat.com ([209.132.183.28]:38564 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728683AbfAGQW3 (ORCPT ); Mon, 7 Jan 2019 11:22:29 -0500 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id B0387804ED; Mon, 7 Jan 2019 16:22:28 +0000 (UTC) Received: from redhat.com (ovpn-122-143.rdu2.redhat.com [10.10.122.143]) by smtp.corp.redhat.com (Postfix) with SMTP id 9A08E65F76; Mon, 7 Jan 2019 16:22:23 +0000 (UTC) Date: Mon, 7 Jan 2019 11:22:23 -0500 From: "Michael S. Tsirkin" To: Peter Zijlstra Cc: Jason Wang , linux-kernel@vger.kernel.org, Alan Stern , Andrea Parri , Will Deacon , Boqun Feng , Nicholas Piggin , David Howells , Jade Alglave , Luc Maranget , "Paul E. McKenney" , Akira Yokosawa , Daniel Lustig , linux-arch@vger.kernel.org, netdev@vger.kernel.org, virtualization@lists.linux-foundation.org, Jonathan Corbet , Richard Henderson , Ivan Kokshaysky , Matt Turner , Arnd Bergmann , Luc Van Oostenryck , linux-doc@vger.kernel.org, linux-alpha@vger.kernel.org, linux-sparse@vger.kernel.org Subject: Re: [PATCH RFC 3/4] barriers: convert a control to a data dependency Message-ID: <20190107112007-mutt-send-email-mst@kernel.org> References: <20190102205715.14054-1-mst@redhat.com> <20190102205715.14054-4-mst@redhat.com> <86023cbe-d1ae-a0d6-7b75-26556f1a0c1f@redhat.com> <20190106231756-mutt-send-email-mst@kernel.org> <20190107094610.GA2861@worktop.programming.kicks-ass.net> <20190107082223-mutt-send-email-mst@kernel.org> <20190107155423.GM2218@hirez.programming.kicks-ass.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20190107155423.GM2218@hirez.programming.kicks-ass.net> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.27]); Mon, 07 Jan 2019 16:22:29 +0000 (UTC) Sender: linux-doc-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-doc@vger.kernel.org On Mon, Jan 07, 2019 at 04:54:23PM +0100, Peter Zijlstra wrote: > On Mon, Jan 07, 2019 at 08:36:36AM -0500, Michael S. Tsirkin wrote: > > On Mon, Jan 07, 2019 at 10:46:10AM +0100, Peter Zijlstra wrote: > > > > How about naming the thing: dependent_ptr() ? That is without any (r)mb > > > implications at all. The address dependency is strictly weaker than an > > > rmb in that it will only order the two loads in qestion and not, like > > > rmb, any prior to any later load. > > > > So I'm fine with this as it's enough for virtio, but I would like to point out two things: > > > > 1. E.g. on x86 both SMP and DMA variants can be NOPs but > > the madatory one can't, so assuming we do not want > > it to be stronger than rmp then either we want > > smp_dependent_ptr(), dma_dependent_ptr(), dependent_ptr() > > or we just will specify that dependent_ptr() works for > > both DMA and SMP. > > The latter; the construct simply generates dependent loads. It is up to > the CPU as to what all that works for. But not on intel right? On intel loads are ordered so it can be a nop. > > 2. Down the road, someone might want to order a store after a load. > > Address dependency does that for us too. Assuming we make > > dependent_ptr a NOP on x86, we will want an mb variant > > which isn't a NOP on x86. Will we want to rename > > dependent_ptr to dependent_ptr_rmb at that point? > > Not sure; what is the actual overhead of the construct on x86 vs the > NOP? I'll have to check. There's a pipeline stall almost for sure - that's why we put it there after all :). -- MST