From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 32656 invoked by alias); 26 Mar 2008 15:12:29 -0000 Received: (qmail 32646 invoked by uid 22791); 26 Mar 2008 15:12:29 -0000 X-Spam-Check-By: sourceware.org Received: from mail.codesourcery.com (HELO mail.codesourcery.com) (65.74.133.4) by sourceware.org (qpsmtpd/0.31) with ESMTP; Wed, 26 Mar 2008 15:12:04 +0000 Received: (qmail 3702 invoked from network); 26 Mar 2008 15:12:01 -0000 Received: from unknown (HELO orlando.local) (pedro@127.0.0.2) by mail.codesourcery.com with ESMTPA; 26 Mar 2008 15:12:01 -0000 From: Pedro Alves To: gdb-patches@sourceware.org Subject: Re: Cleanup MI testsuite Date: Wed, 26 Mar 2008 15:12:00 -0000 User-Agent: KMail/1.9.6 (enterprise 0.20070907.709405) References: <200803261629.53454.vladimir@codesourcery.com> In-Reply-To: <200803261629.53454.vladimir@codesourcery.com> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200803261511.57740.pedro@codesourcery.com> X-IsSubscribed: yes 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: 2008-03/txt/msg00400.txt.bz2 A Wednesday 26 March 2008 13:29:52, Vladimir Prus wrote: > When I've checked in the patch to implement per-thread varobjs, I've > implicitly assumed that MI output does not change in single-threaded > programs, so no testsuite update is necessary. However, that's not > necessary true, and with Pedro's recent patch, is not true on x86. My patch affected every linux native target, independently of arch. Were you testing on Cygwin instead for example, when you get to main, there's already 2 threads listed in info threads. The main thread and another one internal to cygwin1.dll. > > Updating the testsuite should be trivial, except that MI testsuite > is poorly designed: > > - Just about every test uses regular expression that the output > much match > - The expression is too rigid, so a new field in the output > breaks it. > > So, I've checked in the below patch that makes just about every > test for -var-create and -var-list-children use helper functions. > Along the way, it allows the thread-id field in the output of > those commands. > > Testing on x86 with and without printing of thread id for varobjs > shows no issues. > Very Nice! -- Pedro Alves