From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 17840 invoked by alias); 30 Aug 2013 15:18:32 -0000 Mailing-List: contact libc-ports-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Post: List-Help: , Sender: libc-ports-owner@sourceware.org Received: (qmail 17830 invoked by uid 89); 30 Aug 2013 15:18:32 -0000 Received: from relay1.mentorg.com (HELO relay1.mentorg.com) (192.94.38.131) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with ESMTP; Fri, 30 Aug 2013 15:18:32 +0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-2.7 required=5.0 tests=AWL,BAYES_00,KHOP_THREADED,RDNS_NONE,SPF_HELO_FAIL autolearn=no version=3.3.2 X-HELO: relay1.mentorg.com Received: from svr-orw-fem-01.mgc.mentorg.com ([147.34.98.93]) by relay1.mentorg.com with esmtp id 1VFQSy-0006Bk-Im from joseph_myers@mentor.com ; Fri, 30 Aug 2013 08:18:28 -0700 Received: from SVR-IES-FEM-01.mgc.mentorg.com ([137.202.0.104]) by svr-orw-fem-01.mgc.mentorg.com over TLS secured channel with Microsoft SMTPSVC(6.0.3790.4675); Fri, 30 Aug 2013 08:18:28 -0700 Received: from digraph.polyomino.org.uk (137.202.0.76) by SVR-IES-FEM-01.mgc.mentorg.com (137.202.0.104) with Microsoft SMTP Server id 14.2.247.3; Fri, 30 Aug 2013 16:18:26 +0100 Received: from jsm28 (helo=localhost) by digraph.polyomino.org.uk with local-esmtp (Exim 4.76) (envelope-from ) id 1VFQSv-0004fO-41; Fri, 30 Aug 2013 15:18:25 +0000 Date: Fri, 30 Aug 2013 15:18:00 -0000 From: "Joseph S. Myers" To: Will Newton CC: "libc-ports@sourceware.org" , Patch Tracking Subject: Re: [PATCH] sysdeps/arm/armv7/multiarch/memcpy_impl.S: Improve performance. In-Reply-To: Message-ID: References: <520894D5.7060207@linaro.org> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" X-SW-Source: 2013-08/txt/msg00083.txt.bz2 On Fri, 30 Aug 2013, Will Newton wrote: > > There are various comments regarding alignment, whether stating "LDRD/STRD > > support unaligned word accesses" or referring to the mutual alignment that > > applies for particular code. Does this patch make any of them out of > > date? (If code can now only be reached with common 64-bit alignment, but > > in fact requires only 32-bit alignment, the comment should probably state > > both those things explicitly.) > > I've reviewed the comments and they all look ok as far as I can tell. Are you sure? For example, where it says "SRC and DST have the same mutual 32-bit alignment", is it not in fact the case after the patch that they now have the same mutual 64-bit alignment, even if this code doesn't currently rely on that? I think the comments in each place should be explicit about both things - what preconditions the code relies on, and what possibly stronger preconditions are in fact true given the other code in this file. Saying the mutual alignment is 32-bit when the reader can see from the code not far above that it's 64-bit just seems liable to confuse the reader, even if the comment is still formally true. Similarly for the requirements on unaligned word accesses - after this patch, which uses of ldrd/strd do require that? -- Joseph S. Myers joseph@codesourcery.com