This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: PATCH: propagate gdb host keyboard interrupts through gdbserver
- To: gdb-patches at sourceware dot cygnus dot com
- Subject: Re: PATCH: propagate gdb host keyboard interrupts through gdbserver
- From: Daniel Jacobowitz <dmj+ at andrew dot cmu dot edu>
- Date: Wed, 11 Jul 2001 11:27:40 -0700
- References: <5mzoabbbzd.fsf@jtc.redback.com>
On Wed, Jul 11, 2001 at 10:40:38AM -0700, J.T. Conklin wrote:
> I just committed the enclosed patch. It's a patch Greg McGary
> submitted all too long ago.
>
> I took the liberty of adding a comment explaining why code to ignore
> spurious interrupts was added. I also tweaked the change to input_-
> interrupt() to more closely resemble the original code.
>
> --jtc
>
> 2001-07-11 Greg McGary <greg@mcgary.org>
>
> * gdbserver/remote-utils.c (remote_open): Set gdbserver as "owner"
> of SIGIO.
> (input_interrupt): Don't block on read, in case we got redundant
> SIGIO. Don't gripe about redundant SIGIO.
> * gdbserver/low-hppabsd.c (mywait): Use waitpid(). Enable SIGIO
> handler while waiting.
> * gdbserver/low-linux.c (mywait): Likewise.
> * gdbserver/low-nbsd.c (mywait): Likewise.
> * gdbserver/low-sparc.c (mywait): Likewise.
What's the motivation for switching from wait to waitpid? When
gdbserver gains native thread support, we're going to be waiting for
more than one process at a time.
--
Daniel Jacobowitz Carnegie Mellon University
MontaVista Software Debian GNU/Linux Developer