From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 3140 invoked by alias); 1 Dec 2011 13:41:22 -0000 Received: (qmail 2995 invoked by uid 22791); 1 Dec 2011 13:41:21 -0000 X-SWARE-Spam-Status: No, hits=-1.3 required=5.0 tests=AWL,BAYES_00,SPF_FAIL X-Spam-Check-By: sourceware.org Received: from gbenson.demon.co.uk (HELO gbenson.demon.co.uk) (80.177.220.214) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Thu, 01 Dec 2011 13:41:00 +0000 Date: Thu, 01 Dec 2011 13:41:00 -0000 From: Gary Benson To: gdb-patches@sourceware.org Subject: Re: [RFA] Allow setting breakpoints on inline functions (PR 10738) Message-ID: <20111201134058.GB3708@redhat.com> Mail-Followup-To: gdb-patches@sourceware.org References: <20111129150200.GB3425@redhat.com> <20111130201848.GA12763@host2.jankratochvil.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20111130201848.GA12763@host2.jankratochvil.net> 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: 2011-12/txt/msg00002.txt.bz2 Jan Kratochvil wrote: > On Tue, 29 Nov 2011 16:02:00 +0100, Gary Benson wrote: > > I'm also interested in feedback about the tests I've written, > > since this is the first time I've worked on the testsuite. > > I would prefer definitely a copy in gdb.dwarf2/ . gdb.opt/ I find a > bad idea as it breaks too much across GCC changes. just gcc -S -dA > is probably OK. > > If there is gdb.dwarf2/ then I would even drop the gdb.opt/ one. > Not sure if it makes sense for non-DWARF targets, the functionality > gets tested by the DWARF targets anyway. Are you saying I should move the testcase from gdb.opt and into gdb.dwarf2? I can do that. Is there some difference between the way the various directories of tests are treated? And, is there something I should read to help decide where to put tests? > > --- /dev/null > > +++ b/gdb/testsuite/gdb.opt/inline-break.exp > ... > > +if { [prepare_for_testing $testfile.exp $testfile $testfile.c \ > > + {debug optimize=-O2 additional_flags=-Winline}] } { > > This -Winline is questionable. > > You make the .c file compatible even if __GNUC__ is not defined but > then you use GCC-specific -Winline option. Either just make the > testcase whole GCC-specific or make a fallback if the compilation > with -Winline fails try also non-Winline. Or maybe just drop that > -Winline, there is no -Werror anyway so it was more just for the > testcase development. As I understand it the -Winline is there to cause the test to fail if the methods don't get inlined. I may very well be wrong, I don't understand the syntax 100%, but as I read it any unexpected compiler output causes gdb_compile to assume the compilation failed. I copied this (somewhat) from the other gdb.opt/inline* testcases, so if there is a problem with it then I guess they should be fixed too. > As I suggested the gdb.dwarf2/ way this whole -Winline is offtopic > then. > > > +# > > +# func1 is a static inlined function that is called once. > > +# The result should be a single-location breakpoint. > > +# > > +gdb_test "break func1" \ > > + "Breakpoint.*at.* file .*$testfile\.c, line.*" > > As you use "" and not {} you should use \\.c and not \.c . This way > it is the same as .c . > > It is there several times. > > > +# func2 is a non-static inlined function that is called once. > > +# The result should be a breakpoint with two locations: the > > +# out-of-line function and the single inlined instance. > > +# > > +gdb_test "break func2" \ > > + "Breakpoint.*at.*func2.*(2 locations).*" > > You wanted to use \\( and \\) here. > > It is there several times. Thanks, I added the extra '\'s on my branch. Would it be better to to use {} here, or does that make other changes? Cheers, Gary -- http://gbenson.net/