This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: Xtensa GDB port -- revised patch
- From: Daniel Jacobowitz <drow at false dot org>
- To: Maxim Grigoriev <maxim at tensilica dot com>
- Cc: gdb-patches at sources dot redhat dot com, Bob Wilson <bwilson at tensilica dot com>, Marc Gauthier <marc at tensilica dot com>
- Date: Fri, 10 Nov 2006 15:39:08 -0500
- Subject: Re: Xtensa GDB port -- revised patch
- References: <451B202C.3090804@hq.tensilica.com> <20060928011910.GA20142@nevyn.them.org> <45259B4B.3010101@hq.tensilica.com>
On Thu, Oct 05, 2006 at 04:54:51PM -0700, Maxim Grigoriev wrote:
> Hi Daniel and GDB maintainers,
>
> >>Can you use the previous frame's stack pointer instead?
> >>Is that going to work?
>
> Yes, it works fine.
>
> I've attached a modified patch.
Thanks. This is almost OK. There are a few small issues left:
- The port needs a maintainer, and an entry in MAINTAINERS. Who will
be the maintainer?
- I noticed that your prev_register recursively calls
frame_register_unwind. This isn't a big problem but it is
incorrect. If you defer to another frame, you should set
e.g. realnump to point the number for the next frame. See
trad_frame_get_prev_register for an example.
- There's one stray hunk:
> @@ -1330,7 +1332,6 @@
> done
>
> Makefile: Makefile.in config.status @frags@
> - # Regenerate the Makefile and the tm.h / nm.h links.
> CONFIG_FILES=Makefile \
- There's a few instances of operators without spaces around them,
like "-=4" and "(op1)==0x6c" and "8? 8". I don't think you need to
go back and fix them at this point, but please be careful in the
future.
With the first three things fixed, I think we could commit the patch.
--
Daniel Jacobowitz
CodeSourcery