From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 939 invoked by alias); 18 Jul 2009 12:55:57 -0000 Received: (qmail 926 invoked by uid 22791); 18 Jul 2009 12:55:56 -0000 X-SWARE-Spam-Status: No, hits=-1.9 required=5.0 tests=AWL,BAYES_00,SPF_SOFTFAIL X-Spam-Check-By: sourceware.org Received: from qmta05.emeryville.ca.mail.comcast.net (HELO QMTA05.emeryville.ca.mail.comcast.net) (76.96.30.48) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Sat, 18 Jul 2009 12:55:47 +0000 Received: from OMTA19.emeryville.ca.mail.comcast.net ([76.96.30.76]) by QMTA05.emeryville.ca.mail.comcast.net with comcast id HQpe1c0021eYJf8A5Qvn7H; Sat, 18 Jul 2009 12:55:47 +0000 Received: from [192.168.0.101] ([24.10.247.15]) by OMTA19.emeryville.ca.mail.comcast.net with comcast id HQvl1c0080Lg2Gw01QvmEu; Sat, 18 Jul 2009 12:55:46 +0000 Message-ID: <4A61C658.7040707@byu.net> Date: Sat, 18 Jul 2009 12:55:00 -0000 From: Eric Blake User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.22) Gecko/20090605 Thunderbird/2.0.0.22 Mnenhy/0.7.6.666 MIME-Version: 1.0 To: cygwin-talk@cygwin.com Subject: Re: fun? with libsigsegv References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-IsSubscribed: yes Mailing-List: contact cygwin-talk-help@cygwin.com; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Post: List-Help: , Sender: cygwin-talk-owner@cygwin.com Reply-To: The Vulgar and Unprofessional Cygwin-Talk List Mail-Followup-To: cygwin-talk@cygwin.com X-SW-Source: 2009-q3/txt/msg00021.txt.bz2 -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 According to Eric Blake on 7/17/2009 4:50 PM: >> Without libsigsegv, things operate as before. But with libsigsegv, I now > see... > > Aargh. I spoke too soon. For my previous mail, I fat-fingered my command > line I hate it when a typo changes the course of my entire weekend. I thought for sure that I was home free, until I realized that I wasn't testing the snapshot like I thought I was. Almost as bad as the feeling you get when you see a hippo charging you. With m4, it's interesting how many pieces have to interact, and yet how concise a reproducible test case is: echo | m4 >&- It requires cygwin 1.7 (1.5 passes, as does Linux), closed stdout (who does that, normally?), and libsigsegv (no crash if the library is omitted from m4) - in other words, not your most common combinations. At any rate, it looks like the crash can be independently avoided from any one of three locations (gnulib's use of fflush[1], newlib's fault in fflush[2], and libsigsegv's overreacting to faults that cygwin wants to claim[3]). But although any one of the patches fixes the simple m4 test case, I'm hoping that all three get accepted in their respective upstream repositories. [1] http://thread.gmane.org/gmane.comp.lib.gnulib.bugs/18046/focus=18052 [2] http://sourceware.org/ml/newlib/2009/msg00770.html [3] URL to libsigsegv patch not yet known - wanna help me write it? ;) - -- Don't work too hard, make some time for fun as well! Eric Blake ebb9@byu.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (Cygwin) Comment: Public key at home.comcast.net/~ericblake/eblake.gpg Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkphxlgACgkQ84KuGfSFAYCDpQCgkMnCxOiWq3MInHukLETKHv3N DtMAoM6pqNOg0jF3M/mF1EB4rzOtMyZY =bRRu -----END PGP SIGNATURE-----