From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 22495 invoked by alias); 7 Feb 2014 15:45:45 -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 22477 invoked by uid 89); 7 Feb 2014 15:45:44 -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,SPF_PASS autolearn=ham version=3.3.2 X-HELO: mail-lb0-f169.google.com Received: from mail-lb0-f169.google.com (HELO mail-lb0-f169.google.com) (209.85.217.169) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES128-SHA encrypted) ESMTPS; Fri, 07 Feb 2014 15:45:43 +0000 Received: by mail-lb0-f169.google.com with SMTP id q8so2831733lbi.0 for ; Fri, 07 Feb 2014 07:45:39 -0800 (PST) MIME-Version: 1.0 X-Received: by 10.152.1.3 with SMTP id 3mr1267137lai.58.1391787939847; Fri, 07 Feb 2014 07:45:39 -0800 (PST) Received: by 10.112.22.231 with HTTP; Fri, 7 Feb 2014 07:45:39 -0800 (PST) In-Reply-To: <8761orknyz.fsf@igel.home> References: <1391594162-22269-1-git-send-email-will.newton@linaro.org> <20140206221127.6E28974443@topped-with-meat.com> <8761orknyz.fsf@igel.home> Date: Fri, 07 Feb 2014 15:45:00 -0000 Message-ID: Subject: Re: [PATCH v2] ARM: Add SystemTap probes to longjmp and setjmp. From: "Jonathan S. Shapiro" To: Andreas Schwab Cc: Will Newton , Roland McGrath , "libc-ports@sourceware.org" , Patch Tracking Content-Type: text/plain; charset=ISO-8859-1 X-IsSubscribed: yes X-SW-Source: 2014-02/txt/msg00014.txt.bz2 On Fri, Feb 7, 2014 at 6:16 AM, Andreas Schwab wrote: > According to ports/sysdeps/arm/bits/setjmp.h the layout of jmp_buf is > part of the ABI. Yes. The layout of jmp_buf is part of the ABI. And not just the libc ABI, but for some platforms it's part of the platform standard (e.g. SVID specifies it). In consequence, changing the structure layout isn't an option in either the current libc release or any foreseeable future libc release. The patch point needs to be reworked so as to operate given the current structure layout. On the bright side, reworking the code to fit the current layout means that (a) this can be integrated sooner, and (b) we don't have to accept a small regression in GDB functionality in order to implement encrypted pointers. Jonathan