Re: Console client, signal handeling patch

[Marcus Brinkmann]

On Thu, May 01, 2003 at 01:47:24AM +0200, marco@linuxrulez.nl wrote:
> This patch makes the console client handle signals like SIGTERM and
> SIGSEGV. The signal handler calls console_exit to restore the previous
> state, otherwise some hardware state won't be restored when it crashes or
> is killed.

Ah, nice.
 
> I also added a switch, --debug, to disable this handeling for signals that
> make the console client coredump because it makes debugging with gdb
> difficult. By default it is on (signal handling for those signals).

The question is, do we really want to catch those signals at all?  I think
that we only want to catch those signals we really expect, like SIGTERM. 
Other signals point out a bug, and we should just fix the bug instead.
Crashing is not really acceptable.  (Compare this with sync'ing of
filesystems).

> 2003-05-01  Marco Gerards  <metgerards@student.han.nl>
>         * console.c (options): Added option to disable catching of signals
>         that can cause a coredump...
>         (parse_opt): ...and parse it here.
>         (sighandler): New function.
>         (handle_signal): New macro.
>         (main): Handle signals.

Nitpick: Empty line between date/name and entries.  Space before and after 
"...".

Thanks,
Marcus

-- 
`Rhubarb is no Egyptian god.' GNU      http://www.gnu.org    marcus@gnu.org
Marcus Brinkmann              The Hurd http://www.gnu.org/software/hurd/
Marcus.Brinkmann@ruhr-uni-bochum.de
http://www.marcus-brinkmann.de/