From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 19532 invoked by alias); 17 May 2005 03:44:15 -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 19511 invoked from network); 17 May 2005 03:44:10 -0000 Received: from unknown (HELO nevyn.them.org) (66.93.172.17) by sourceware.org with SMTP; 17 May 2005 03:44:10 -0000 Received: from drow by nevyn.them.org with local (Exim 4.50) id 1DXt06-0005ZB-Co; Mon, 16 May 2005 23:44:10 -0400 Date: Tue, 17 May 2005 05:46:00 -0000 From: Daniel Jacobowitz To: Joel Brobecker Cc: gdb-patches@sources.redhat.com Subject: Re: [RFA/alpha] Add handling of FP control insn in software-single step Message-ID: <20050517034410.GA21351@nevyn.them.org> Mail-Followup-To: Joel Brobecker , gdb-patches@sources.redhat.com References: <20050517023400.GW1462@adacore.com> <20050517023835.GA19735@nevyn.them.org> <20050517025048.GL12565@adacore.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20050517025048.GL12565@adacore.com> User-Agent: Mutt/1.5.8i X-SW-Source: 2005-05/txt/msg00410.txt.bz2 On Tue, May 17, 2005 at 12:50:48PM +1000, Joel Brobecker wrote: > > On Tue, May 17, 2005 at 12:34:00PM +1000, Joel Brobecker wrote: > > > Tested on alpha-tru64 5.1a with no regression. It fixes a problem with > > > some code but I can't contribute it, as it has been given to us by a > > > customer. > > > > Can you write a test for this in gdb.arch, using an asm to provide the > > fbeq instruction that was causing a problem? > > Let me see what I can do. I tend to prefer tests written in C or Ada, > as it can be used on all platforms rather than just on alpha. But > perhaps the usage in this case is to prefer asm? The feature you're implementing is specific to Alpha; think of it in the nature of unit testing. That's not to say a test for "single step over a floating point compare-and-branch" would be a bad thing to have in the rest of the testsuite. But I'm sure you can imagine how hard it is to predict what a compiler will generate. -- Daniel Jacobowitz CodeSourcery, LLC