From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 73788 invoked by alias); 31 May 2019 15:48:23 -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 73692 invoked by uid 89); 31 May 2019 15:48:14 -0000 Authentication-Results: sourceware.org; auth=none X-Spam-SWARE-Status: No, score=-1.9 required=5.0 tests=BAYES_00,SPF_PASS autolearn=ham version=3.3.1 spammy= X-HELO: mail3-relais-sop.national.inria.fr Received: from mail3-relais-sop.national.inria.fr (HELO mail3-relais-sop.national.inria.fr) (192.134.164.104) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with ESMTP; Fri, 31 May 2019 15:48:04 +0000 Received: from aaubervilliers-653-1-65-238.w86-198.abo.wanadoo.fr (HELO stedding) ([86.198.56.238]) by mail3-relais-sop.national.inria.fr with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 31 May 2019 17:48:02 +0200 Date: Fri, 31 May 2019 15:51:00 -0000 From: Marc Glisse To: Aldy Hernandez cc: Jeff Law , Richard Biener , Martin Sebor , GCC Patches Subject: Re: Simplify more EXACT_DIV_EXPR comparisons In-Reply-To: Message-ID: References: <2d73a72a-fac5-0e70-77b0-6aeaf24c8f49@redhat.com> User-Agent: Alpine 2.21 (DEB 202 2017-01-01) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII; format=flowed X-SW-Source: 2019-05/txt/msg02140.txt.bz2 On Fri, 31 May 2019, Aldy Hernandez wrote: > I've never been too happy with the too large due to cast warnings. For that > matter, it seems like a lot of the unbounded alloca warning variants were > artifacts of the way we couldn't get precise ranges after vrp asserts had > disappeared and we were trying to guess at what the actual range in the > original code was. It's fragile at best. Yes, very fragile. > I haven't been paying too much attention to walloca because the ranger gets > considerably better context ranges in the ranger walloca version, and we > are getting correct warnings for a variety of things we couldn't before. So > I was hoping to ignore this until we all agreed on what range, vrp etc will > look like going forward. Seems sensible. > That being said, I could take a closer look at this xfail on Monday if > y'all would like. But I don't currently have strong opinions either way. I > guess it'll all change in the next few months. As long as you are ok with one Walloca testcase being xfailed until the VRP work lands, I don't think there is a need to spend time on it now. -- Marc Glisse