This is the mail archive of the
mailing list for the GCC project.
Re: [Tru64 PATCH] PR target/22223: .frame vs. large stack frames
- From: Roger Sayle <roger at eyesopen dot com>
- To: Albert Chin <gcc-patches at mlists dot thewrittenword dot com>
- Cc: gcc-patches at gcc dot gnu dot org, Rainer Orth <ro at TechFak dot Uni-Bielefeld dot DE>
- Date: Wed, 6 Sep 2006 14:41:36 -0600 (MDT)
- Subject: Re: [Tru64 PATCH] PR target/22223: .frame vs. large stack frames
On Wed, 6 Sep 2006, Albert Chin wrote:
> On Fri, Sep 01, 2006 at 07:40:01AM -0600, Roger Sayle wrote:
> > ... This restriction is less severe than it may seem, as gdb has
> > never worked well on this platform (breakpoints trigger segmentation
> > faults);
> Really? gdb-6.4 works on a really simple program:
> $ gcc-3.4.3 -g a.c
> $ gdb-6.4 a.out
Interesting, gdb 6.4 does work with your a.out created by gcc v3.4.3!
(conveniently both the versions I have installed on my Tru64 box).
However, changing the experiment just a little to compile the same
program to a.out with current mainline gcc:
alpha% GCC/clean/gcc/xgcc -BGCC/clean/gcc -g a.c
alpha% gdb a.out
GNU gdb 6.4
Copyright 2005 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for details.
This GDB was configured as "alphaev67-dec-osf5.1"...Segmentation fault
And of course, you get exactly the same core dump attempting to use
gdb on GCC's cc1. But you're right that on simple variants of your
test program breakpoints, segmentation faults, displaying arguments and
backtraces all appear to be functional. Perhaps things currently work
much better than I remember.