This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: [patch/rfc] Use frame_type for sigtramp test in infrun.c
- From: Joel Brobecker <brobecker at gnat dot com>
- To: Andrew Cagney <cagney at gnu dot org>
- Cc: Ulrich Weigand <weigand at i1 dot informatik dot uni-erlangen dot de>,gdb-patches at sources dot redhat dot com
- Date: Fri, 2 Apr 2004 15:57:22 -0800
- Subject: Re: [patch/rfc] Use frame_type for sigtramp test in infrun.c
- References: <200403292338.BAA16799@faui1d.informatik.uni-erlangen.de> <406DD226.1080104@gnu.org>
> Joel, from memory you had a change to:
>
> if (((stop_pc == ecs->stop_func_start /* Quick test */
> || in_prologue (stop_pc, ecs->stop_func_start))
> && !IN_SOLIB_RETURN_TRAMPOLINE (stop_pc, ecs->stop_func_name))
> || IN_SOLIB_CALL_TRAMPOLINE (stop_pc, ecs->stop_func_name)
> || ecs->stop_func_name == 0)
> {
> /* It's a subroutine call. */
> handle_step_into_function (ecs);
> return;
> }
>
> pending? If we do pull the sigtramp code I think it would be prudent to
> first have that committed - Joel's change greatly clarifies the logic.
Yes. Something that would isolate this code to the platforms who haven't
been framified yet (are there still any?). But the problem is that I
got some regressions mostly on HP/UX, but also one on Tru64. The
testsuite results were clean on sparc-solaris and x86-linux.
The problem was that it was difficult for me to investigate the problems
on HP/UX given the current status. So I started looking at the current
failures on HP/UX to see if I could improve them a bit.
I should probably redo some testing and see where we stand. Yeah, let
me do that.
--
Joel