This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: RFC: fix latent bug in syms_from_objfile_1
- From: Jan Kratochvil <jan dot kratochvil at redhat dot com>
- To: Tom Tromey <tromey at redhat dot com>
- Cc: gdb-patches at sourceware dot org
- Date: Wed, 20 Mar 2013 19:40:34 +0100
- Subject: Re: RFC: fix latent bug in syms_from_objfile_1
- References: <87ip4m7wxy dot fsf at fleche dot redhat dot com>
On Wed, 20 Mar 2013 19:22:49 +0100, Tom Tromey wrote:
> --- symfile.c 14 Mar 2013 20:26:19 -0000 1.367
> +++ symfile.c 20 Mar 2013 18:19:42 -0000
> @@ -997,7 +997,7 @@
>
> We no longer warn if the lowest section is not a text segment (as
> happens for the PA64 port. */
> - if (addrs && addrs->other[0].name)
> + if (addrs && addrs->num_sections > 0 && addrs->other[0].name)
> addr_info_make_relative (addrs, objfile->obfd);
>
> /* Initialize symbol reading routines for this objfile, allow complaints to
I find it just fixes a regression introduced by:
commit 2679ab4be931bedfc7987d215bc336b912364906
Author: Elena Zannoni <ezannoni@kwikemart.cygnus.com>
Date: Fri Jun 27 13:11:17 2003 +0000
And it also matches how addr_info_make_relative and other code handles ADDRS.
The problem is that alloc_section_addr_info sets NUM_SECTIONS while they are
not yet filled in which leads to those ugly "&& ->other[x].name" checks...
Thanks,
Jan