From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 14420 invoked by alias); 28 Apr 2002 16:39:33 -0000 Mailing-List: contact gdb-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-owner@sources.redhat.com Received: (qmail 14410 invoked from network); 28 Apr 2002 16:39:31 -0000 Received: from unknown (HELO cygnus.com) (205.180.83.203) by sources.redhat.com with SMTP; 28 Apr 2002 16:39:31 -0000 Received: from redhat.com (remus.sfbay.redhat.com [172.16.27.252]) by runyon.cygnus.com (8.8.7-cygnus/8.8.7) with ESMTP id JAA19522; Sun, 28 Apr 2002 09:39:14 -0700 (PDT) Message-ID: <3CCC24DB.A37605D5@redhat.com> Date: Sun, 28 Apr 2002 09:39:00 -0000 From: Fernando Nasser Organization: Red Hat Canada X-Accept-Language: en MIME-Version: 1.0 To: Stan Shebs CC: Elena Zannoni , gdb@sources.redhat.com, Andrew Cagney Subject: Re: Adding target specific tests References: <15561.39436.53833.917524@localhost.redhat.com> <3CC9B863.6AABD20C@apple.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-SW-Source: 2002-04/txt/msg00476.txt.bz2 Stan Shebs wrote: > > Elena Zannoni wrote: > > > > I am about to add 2 new testfiles to the gdb testsuite. Both are > > specific to Altivec. One is a generic test for the ABI (function calls > > and return values), the other deals with the AltiVec registers. > > > > Where should I put them? > > I would suggest a gdb.arch or gdb.target for random target > specific tests, and put the altivec tests directly in it, > unless you anticipate scores or hundreds of altivec test > files a la gdb.hp. > That is not a bad idea. I liked the gdb.altivec one because we could configure or not that directory, but we can perhaps do the same with the files in a gdb.target (or gdb.arch) directory. I would like to configure them out to avoid lots of UNSUPPORTED tests. I would add a guard (testing for the right target) anyway. P.S.: Question: how would the GDB testsuite handle multiarched targets? We will need to "set arch" to one architecture, run the tests, set arch to another, re-run the tests (all?). I guess our tests for target will have to stop being on a configuration triplet and start using a gdb command to set the arch and confirm that it is available). All the xfail/kfail mechanism would have to change as well... -- Fernando Nasser Red Hat Canada Ltd. E-Mail: fnasser@redhat.com 2323 Yonge Street, Suite #300 Toronto, Ontario M4P 2C9