From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 124560 invoked by alias); 21 Sep 2018 17:27:28 -0000 Mailing-List: contact libc-alpha-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: libc-alpha-owner@sourceware.org Received: (qmail 124546 invoked by uid 89); 21 Sep 2018 17:27:28 -0000 Authentication-Results: sourceware.org; auth=none X-Spam-SWARE-Status: No, score=-0.6 required=5.0 tests=BAYES_00,KHOP_DYNAMIC,RCVD_IN_DNSWL_LOW,SPF_PASS autolearn=no version=3.3.2 spammy= X-HELO: mx0a-001b2d01.pphosted.com From: Tulio Magno Quites Machado Filho To: Florian Weimer , Joseph Myers Cc: libc-alpha@sourceware.org Cc: Subject: Re: [PATCH] Fix ldbl-128ibm fma (Inf, Inf, finite) (bug 23272) In-Reply-To: <8736u3f1yq.fsf@oldenburg.str.redhat.com> References: <20180613220136.11438-1-tuliom@linux.ibm.com> <87r2hnhxhg.fsf@oldenburg.str.redhat.com> <8736u3f1yq.fsf@oldenburg.str.redhat.com> User-Agent: Notmuch/0.27 (http://notmuchmail.org) Emacs/25.3.1 (x86_64-redhat-linux-gnu) Date: Fri, 21 Sep 2018 17:27:00 -0000 MIME-Version: 1.0 Content-Type: text/plain x-cbid: 18092117-0016-0000-0000-000009321B5E X-IBM-SpamModules-Scores: X-IBM-SpamModules-Versions: BY=3.00009747; HX=3.00000242; KW=3.00000007; PH=3.00000004; SC=3.00000266; SDB=6.01091622; UDB=6.00564047; IPR=6.00871647; MB=3.00023437; MTD=3.00000008; XFM=3.00000015; UTC=2018-09-21 17:27:24 X-IBM-AV-DETECTION: SAVI=unused REMOTE=unused XFE=unused x-cbparentid: 18092117-0017-0000-0000-0000406F4CFC Message-Id: <875zyyn4lj.fsf@linux.ibm.com> X-SW-Source: 2018-09/txt/msg00409.txt.bz2 Florian Weimer writes: > * Joseph Myers: >> That answers the first question about the symptoms (both spurious >> overflow exceptions and bad results, apparently coming from __gcc_qadd >> mishandling certain cases involving infinities), but doesn't answer >> the second question about what happens when the infinite x * y comes >> from (Inf * finite) (I'd guess the same problem, in which case the >> proposed patch needs fixing further, but it looks like tests for that >> case also need adding to libm-test-fma.inc). > > Thanks. I have to admit this is a bit over my head. > > Tulio, can you take care of this, or should someone else pick this up? Yes, it's in my task list. I still have to confirm Joseph's points and will get back to you. I still need more time, though. -- Tulio Magno