From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 81653 invoked by alias); 29 Jul 2015 16:25:10 -0000 Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-patches-owner@gcc.gnu.org Received: (qmail 81642 invoked by uid 89); 29 Jul 2015 16:25:10 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-1.6 required=5.0 tests=AWL,BAYES_00,FREEMAIL_FROM,RCVD_IN_DNSWL_LOW,RP_MATCHES_RCVD,SPF_PASS autolearn=ham version=3.3.2 X-HELO: nm3-vm2.bullet.mail.ne1.yahoo.com Received: from nm3-vm2.bullet.mail.ne1.yahoo.com (HELO nm3-vm2.bullet.mail.ne1.yahoo.com) (98.138.91.19) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES128-SHA encrypted) ESMTPS; Wed, 29 Jul 2015 16:25:09 +0000 Received: from [98.138.100.117] by nm3.bullet.mail.ne1.yahoo.com with NNFMP; 29 Jul 2015 16:25:07 -0000 Received: from [98.138.226.132] by tm108.bullet.mail.ne1.yahoo.com with NNFMP; 29 Jul 2015 16:25:07 -0000 Received: from [127.0.0.1] by smtp219.mail.ne1.yahoo.com with NNFMP; 29 Jul 2015 16:25:07 -0000 X-Yahoo-SMTP: RhyaqECswBCSKHdmagqyBBwGHjobejNv Message-ID: <55B8FE61.7060008@yahoo.com> Date: Wed, 29 Jul 2015 17:19:00 -0000 From: Abe User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:31.0) Gecko/20100101 Thunderbird/31.4.0 MIME-Version: 1.0 To: Richard Biener CC: Alan Lawrence , "gcc-patches@gcc.gnu.org" , Sebastian Pop Subject: Re: Another benefit of the new if converter: better performance for half hammocks when running the generated code on a modern high-speed CPU with write-back caching, relative to the code produced by the old if converter given the same source code References: <55A961C1.1070206@yahoo.com> <55B7C27B.9000406@yahoo.com> In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit X-SW-Source: 2015-07/txt/msg02485.txt.bz2 > Well. We don't generally introduce regressions with changes. Understood. Regressions are bad, of course. TTBOMK the regressions in question are temporary. Once they are gone, I think we can then look at whether or not we still need to keep the old if converter in trunk. Ideally, it eventually becomes redundant and unneeded. > (well, the patch still needs review - > I hope to get to that this week). After I`ve done the SPEC-based analysis, my next planned steps on this work are to disable the code that [in my WIP] currently causes conversion to be enabled by default when autovectorization is enabled, then to re-integrate the old converter and implement the switches that will give GCC users access to the modes I described in a recent email from me. You might prefer to delay your code review until I have that all done and a new version of the patch submitted. Regards, Abe