This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [patchv2] GDB 7.2: new feature for "backtrace" that cuts path to file (remain filename)
- From: Jan Kratochvil <jan dot kratochvil at redhat dot com>
- To: Andrew Burgess <aburgess at broadcom dot com>
- Cc: gdb-patches at sourceware dot org
- Date: Mon, 3 Dec 2012 14:21:21 +0100
- Subject: Re: [patchv2] GDB 7.2: new feature for "backtrace" that cuts path to file (remain filename)
- References: <20121130185653.GA15394@host2.jankratochvil.net> <50BC7AB0.3070204@broadcom.com>
On Mon, 03 Dec 2012 11:10:56 +0100, Andrew Burgess wrote:
> I think we should consider switching some (or all) other places we print
> symtab->filename to use the get_filename_display_from_sal function
> instead.
That was intended, I find the patch useless/buggy without it.
"frame" worked fine but I have found for example this case does not:
(gdb) frame
#0 0x00007ffff637d7d0 in __poll_nocancel () at /usr/src/debug/glibc-2.16-75f0d304/sysdeps/unix/syscall-template.S:81
81 T_PSEUDO (SYSCALL_SYMBOL, SYSCALL_NAME, SYSCALL_NARGS)
(gdb) info line *$pc
Line 81 of "../sysdeps/unix/syscall-template.S" starts at address 0x7ffff637d7c0 <poll> and ends at 0x7ffff637d808 <poll+72>.
> Even if this work is not done in this patch (which would be
> fine with me)
Not with me.
> As such, I would rather the gdb option be "set filename-display" rather
> than "set backtrace filename-display",
I was thinking about such change before but I have left it according to the
original submitters proposal which may match best where the users would search
for it. OTOH when I check now the former thread it even did not come as the
original submitters proposal so it could be changed.
Thanks,
Jan