From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 30652 invoked by alias); 26 Dec 2009 05:36:36 -0000 Received: (qmail 30639 invoked by uid 22791); 26 Dec 2009 05:36:35 -0000 X-SWARE-Spam-Status: No, hits=-2.5 required=5.0 tests=AWL,BAYES_00 X-Spam-Check-By: sourceware.org Received: from rock.gnat.com (HELO rock.gnat.com) (205.232.38.15) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Sat, 26 Dec 2009 05:36:30 +0000 Received: from localhost (localhost.localdomain [127.0.0.1]) by filtered-rock.gnat.com (Postfix) with ESMTP id AD13C2BABA7; Sat, 26 Dec 2009 00:36:18 -0500 (EST) Received: from rock.gnat.com ([127.0.0.1]) by localhost (rock.gnat.com [127.0.0.1]) (amavisd-new, port 10024) with LMTP id bErIjetw0tfc; Sat, 26 Dec 2009 00:36:18 -0500 (EST) Received: from joel.gnat.com (localhost.localdomain [127.0.0.1]) by rock.gnat.com (Postfix) with ESMTP id 343242BAB9E; Sat, 26 Dec 2009 00:36:18 -0500 (EST) Received: by joel.gnat.com (Postfix, from userid 1000) id D74C9F5937; Sat, 26 Dec 2009 06:35:55 +0100 (CET) Date: Sat, 26 Dec 2009 05:36:00 -0000 From: Joel Brobecker To: Jan Kratochvil Cc: gdb-patches@sourceware.org Subject: Re: [patch] Fix watchpoint-at-epilogue 7.0 regression (s390x) Message-ID: <20091226053555.GZ2788@adacore.com> References: <20091225142031.GA5442@host0.dyn.jankratochvil.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20091225142031.GA5442@host0.dyn.jankratochvil.net> User-Agent: Mutt/1.5.20 (2009-06-14) Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2009-12/txt/msg00396.txt.bz2 > gdb/ > 2009-12-25 Jan Kratochvil > > * breakpoint.c (watchpoint_check): Check the call > gdbarch_in_function_epilogue_p before calling frame_find_by_id. > Extend the comment. This part looks OK to me. On a more general side, I wondered for a while if frame_find_by_id should be modified to never raise an exception (and return NULL instead). But decided that this was probably a bad idea. Using the same kind of example as provided by Jan, if our program was to stop inside the epilogue a function called by the block where the expression is valid, we might end up deleting the watchpoint prematurely. > gdb/testsuite/ > 2009-12-25 Jan Kratochvil > > * gdb.base/watchpoint-cond-gone.exp, gdb.base/watchpoint-cond-gone.c, > gdb.base/watchpoint-cond-gone-stripped.c: New. OK, with a few changes: - The copyright notices needs to be GPL v3; - One nit, see below. > + gdb_suppress_entire_file "Testcase compile failed, so all tests in this file will automatically fail." We are now trying to get rid of gdb_suppress_entire_file. Can you just "return" instead? BTW: Thank you for using some of the convenience routines. This reminded me that I wanted to start writing a GDB testcase writing cookbook. Your testcase had a lot of stuff that I wanted to put in there, so I just finally started it! Will send an email to gdb@ when I'm done... -- Joel