From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 9180 invoked by alias); 7 Dec 2001 22:03:18 -0000 Mailing-List: contact gdb-patches-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sources.redhat.com Received: (qmail 9084 invoked from network); 7 Dec 2001 22:02:49 -0000 Received: from unknown (HELO nevyn.them.org) (128.2.145.6) by sources.redhat.com with SMTP; 7 Dec 2001 22:02:49 -0000 Received: from drow by nevyn.them.org with local (Exim 3.33 #1 (Debian)) id 16CT5h-0004Px-00 for ; Fri, 07 Dec 2001 17:03:33 -0500 Date: Fri, 07 Dec 2001 14:03:00 -0000 From: Daniel Jacobowitz To: gdb-patches@sources.redhat.com Subject: Re: [rfa/c++testsuite] The remaining v3 testsuite fixes (or at least most of them) Message-ID: <20011207170333.B16322@nevyn.them.org> Mail-Followup-To: gdb-patches@sources.redhat.com References: <20011205182024.A25198@nevyn.them.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.3.23i X-SW-Source: 2001-12/txt/msg00216.txt.bz2 Committed. On Fri, Dec 07, 2001 at 04:49:08PM -0500, Jim Blandy wrote: > > Please commit this, and the static member test patch above. Thanks! > > Daniel Jacobowitz writes: > > [This patch includes my previous patch today, for the static member test, > > but not its changelog; I will commit them separately, but editing it out of > > this diff would have been tricky.] > > > > These are mostly "more of the same" demangler changes: > > - const & -> const ?& > > - \\(void\\) -> \\((void|)\\) > > > > There are also changes to accept classes without the explicit vtbl member, > > or the implicit int parameter that v2 seemed to use to construct classes > > with virtual bases. I also allow the function name in "Breakpoint at " > > messages to contain the base demangled name, including argument list; makes > > sense for overloaded functions. > > > > I currently except class data printouts with explicit "_vptr.vC = $hex" > > style entries. I can suppress those if people think that better, but I'd > > rather get this in first; so I labelled the passes with FIXMEs. Our > > handling of these is a little inconsistent. > > > > These handle all the tests that should pass in my current half-submitted > > codebase, and only changes the name of one FAIL and one FAIL to PASS in the > > v2 test results: > > > > 111c111 > > < FAIL: gdb.c++/classes.exp: continue to enums2 > > --- > > > FAIL: gdb.c++/classes.exp: continue to enums2(\(\)|) > > 1342c1342 > > < FAIL: gdb.c++/namespace.exp: info func xyzq > > --- > > > PASS: gdb.c++/namespace.exp: info func xyzq > > > > (The fail was because the testsuite expected functions to be sorted by > > basename and not by full name including namespace). > > > > Is this OK to commit? -- Daniel Jacobowitz Carnegie Mellon University MontaVista Software Debian GNU/Linux Developer