This is the mail archive of the
gdb-prs@sources.redhat.com
mailing list for the GDB project.
Re: gdb/338: gcore.exp started failing
- From: Daniel Jacobowitz <drow at mvista dot com>
- To: nobody at sources dot redhat dot com
- Cc: gdb-prs at sources dot redhat dot com,
- Date: 4 Feb 2002 22:18:01 -0000
- Subject: Re: gdb/338: gcore.exp started failing
- Reply-to: Daniel Jacobowitz <drow at mvista dot com>
The following reply was made to PR gdb/338; it has been noted by GNATS.
From: Daniel Jacobowitz <drow@mvista.com>
To: kettenis@chello.nl, gdb-gnats@sources.redhat.com
Cc:
Subject: Re: gdb/338: gcore.exp started failing
Date: Mon, 4 Feb 2002 17:12:42 -0500
On Mon, Feb 04, 2002 at 09:18:00PM -0000, kettenis@chello.nl wrote:
> The following reply was made to PR gdb/338; it has been noted by GNATS.
>
> From: kettenis@chello.nl
> To: geoffk@redhat.com
> Cc: gdb-gnats@sources.redhat.com
> Subject: Re: gdb/338: gcore.exp started failing
> Date: 4 Feb 2002 21:09:41 -0000
>
> Date: 4 Feb 2002 17:26:21 -0000
> From: geoffk@redhat.com
>
> I suspect this change:
>
> 2002-01-19 Mark Kettenis <kettenis@gnu.org>
>
> * i386gnu-nat.c: Include "i386-tdep.h".
> (fetch_fpregs): Simplify code dealing with uninitialized floating
> point states such that it doesn't require FP7_REGNUM.
>
> but have not tried backing it out to see if the problem goes away.
>
> That's not very likely :-) since i386gnu-nat.c isn't used on Linux.
> It's more likely that the following change is the culprit:
>
> 2002-01-14 Michael Snyder <msnyder@redhat.com>
>
> * linux-proc.c (linux_do_thread_registers): Ignore fpxregs
> until we can resolve portability issues.
> * gregset.h: Remove references to fpxregs.
> * gcore.c (gcore_command): Initialize note_sec to NULL.
>
> The SSE registers are no longer saved in the generated core, and upon
> loading that core GDB fills them with the dummy values you're seeing.
> We should probably XFAIL that test for i?86-*-linux-gnu*. Or we
> should correctly implement generic handling of extended registers sets
> in GDB.
Yes, that's the change that caused the problem. Before it was reverted
there were another set of problems on MIPS. You got the solution
perfectly right, too :)
--
Daniel Jacobowitz Carnegie Mellon University
MontaVista Software Debian GNU/Linux Developer