Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752637AbXIHIPv (ORCPT ); Sat, 8 Sep 2007 04:15:51 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751378AbXIHIPn (ORCPT ); Sat, 8 Sep 2007 04:15:43 -0400 Received: from smtp110.mail.mud.yahoo.com ([209.191.85.220]:24085 "HELO smtp110.mail.mud.yahoo.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1751370AbXIHIPm (ORCPT ); Sat, 8 Sep 2007 04:15:42 -0400 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com.au; h=Received:X-YMail-OSG:From:To:Subject:Date:User-Agent:Cc:References:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Content-Disposition:Message-Id; b=VQjhI9O/hhH5wSDTaicivYHVsXXfoZD1O5MxGobCdEutUtm2ATH3WE6ZZrr5jb5ydzsZZr57xecU3xSJotOBIQG2naQuFR/laPSyJkkUKRoFhA8HhaH/+NOgCRvI7txvZNRJcObExj95K0N46+e/znxvpV+oi1xLgL2CSmTFHnA= ; X-YMail-OSG: ckER2D0VM1lGfnyKWhapaJDRw1WBVHbpkaaIyTx1LUDSaqWdLGTjOrcK1dwVErKMaqa6wSbPVA-- From: Nick Piggin To: Linus Torvalds Subject: Re: Intel Memory Ordering White Paper Date: Sat, 8 Sep 2007 04:13:12 +1000 User-Agent: KMail/1.9.5 Cc: ak@suse.de, Jesse Barnes , linux-kernel@vger.kernel.org References: <200709071526.51169.jesse.barnes@intel.com> <200709090334.27677.nickpiggin@yahoo.com.au> <200709090348.28076.nickpiggin@yahoo.com.au> In-Reply-To: <200709090348.28076.nickpiggin@yahoo.com.au> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200709080413.12282.nickpiggin@yahoo.com.au> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 881 Lines: 20 On Sunday 09 September 2007 03:48, Nick Piggin wrote: > There is some suggestion in the source code that non-temporal stores > (movntq) are weakly ordered. But AFAIKS from the documents, it is ordered > when operating on wb memory. What's the situation there? Sorry, it looks from the AMD document like nontemporal stores to wb memory can go out of order. It is a bit hard to decipher what the types mean. If this is the case, we can either retain the sfence in smp_wmb(), or noop it, and put explicit sfences around any place that performs nontemporal stores... Anyway, the lfence should be able to go away without so much trouble. - 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/