[RFA] Fix handling of "set sysroot foo"

Doug Evans dje@google.com
Tue Apr 23 07:14:00 GMT 2013


On Fri, Apr 19, 2013 at 12:44 PM, Tom Tromey <tromey@redhat.com> wrote:
>>>>>> "Doug" == Doug Evans <dje@google.com> writes:
>
> Doug> 2013-04-12  Doug Evans  <dje@google.com>
> Doug>   * solist.h (struct target_so_ops): New member clear_solist.
> Doug>   * solib-svr4.c (svr4_clear_solist): New function.
> Doug>   (_initialize_svr4_solib): Set svr4_so_ops.clear_solist.
> Doug>   * solib.c (clear_solist): Renamed from free_so_symbols.
> Doug>   All callers updated.  Call target clear_solist if it exists.
>
> This looks reasonable to me.
>
> How hard would it be to write a regression test for it?

Well, I'm sufficiently traumatized by it if that's what you're asking.  1/2 :-)

http://sourceware.org/ml/gdb-patches/2013-04/msg00676.html



More information about the Gdb-patches mailing list