From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 9027 invoked by alias); 20 May 2010 17:57:39 -0000 Received: (qmail 9018 invoked by uid 22791); 20 May 2010 17:57:38 -0000 X-SWARE-Spam-Status: No, hits=-2.1 required=5.0 tests=AWL,BAYES_00,T_RP_MATCHES_RCVD X-Spam-Check-By: sourceware.org Received: from smtp-outbound-2.vmware.com (HELO smtp-outbound-2.vmware.com) (65.115.85.73) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Thu, 20 May 2010 17:57:33 +0000 Received: from mailhost3.vmware.com (mailhost3.vmware.com [10.16.27.45]) by smtp-outbound-2.vmware.com (Postfix) with ESMTP id 7A4201B004; Thu, 20 May 2010 10:57:30 -0700 (PDT) Received: from msnyder-server.eng.vmware.com (promd-2s-dhcp138.eng.vmware.com [10.20.124.138]) by mailhost3.vmware.com (Postfix) with ESMTP id 68ECECD957; Thu, 20 May 2010 10:57:30 -0700 (PDT) Message-ID: <4BF5780A.2010600@vmware.com> Date: Thu, 20 May 2010 18:02:00 -0000 From: Michael Snyder User-Agent: Thunderbird 2.0.0.22 (X11/20090609) MIME-Version: 1.0 To: Pedro Alves CC: "gdb-patches@sourceware.org" Subject: Re: [RFA] (long) sepdebug.exp replace send_gdb with gdb_test References: <4BF47D1D.2030005@vmware.com> <201005201741.52364.pedro@codesourcery.com> In-Reply-To: <201005201741.52364.pedro@codesourcery.com> Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit 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: 2010-05/txt/msg00428.txt.bz2 Pedro Alves wrote: > I didn't to a thorough review, but I noticed a couple of > issues. > > On Thursday 20 May 2010 01:06:53, Michael Snyder wrote: >> -# Test deleting all breakpoints when there are none installed, >> -# GDB should not prompt for confirmation. >> -# Note that gdb-init.exp provides a "delete_breakpoints" proc >> -# for general use elsewhere. >> - >> -send_gdb "delete breakpoints\n" >> -gdb_expect { >> - -re "Delete all breakpoints.*$" { >> - send_gdb "y\n" >> - gdb_expect { >> - -re "$gdb_prompt $" { >> - fail "Delete all breakpoints when none (unexpected prompt)" >> - } >> - timeout { fail "Delete all breakpoints when none (timeout after unexpected prompt)" } >> - } >> - } >> - -re ".*$gdb_prompt $" { pass "Delete all breakpoints when none" } >> - timeout { fail "Delete all breakpoints when none (timeout)" } >> -} >> +delete_breakpoints > > delete_breakpoints doesn't do what the test was doing before. > Notice the comment.. Whether what is being tested or not has any > value in this case, is another question, but it seems > to have been just blindly copied from break.exp. You could > just delete it it seems. It was copied from break.exp, and I figured as long as it was tested there, it didn't need to be tested again here. > >> - -re "Catchpoint \[0-9\]* .fork..*$gdb_prompt $" >> - {pass $name} >> - -re "Catch of fork not yet implemented.*$gdb_prompt $" >> - {pass $name} >> - -re "$gdb_prompt $" >> - {fail $name} >> - timeout {fail "(timeout) $name"} >> +gdb_test_multiple "catch fork" $name { >> + -re "Catchpoint \[0-9\]* .fork.*" { >> + pass $name > > You should make sure to always consume the prompt with gdb_test_multiple, > otherwise, it may be left in the input stream and confuse follow up tests. Uh-oh, I didn't know that. Thanks, I'll have to redo several of these.