From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Sat, 19 Aug 2000 20:39:31 -0500 From: Steven Bytnar To: linuxppc-dev@lists.linuxppc.org Subject: Pseudo Oops for ppc Message-ID: <20000819203931.A15224@openprojects.rarcoa.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-linuxppc-dev@lists.linuxppc.org List-Id: http://www.stg.com/udi/ppc-oops-1.patch This simple patch improves the default panic backtraceing by walking the kernel symbol table to find function names. It also attempts to kill and cleanup the task that caused the panic by calling do_exit(SIGKILL). This seems to work most of the time and leaves the machine in a mostly usable state, although you should probably reboot soon thereafter. The end result... you can panic the machine and keep working after waiting about 10 seconds. (So you can see the panic message.) You will want to tune the delay loop value. Why is there a delay loop?! I didn't feel like figuring out how to read the USB keyboard driver. This patch isn't very useful if you have kgdb or xmon enabled. Hopefully someone will find this useful. It's been a time saver for me while I've been working on developing kernel modules on a G4. Enjoy, --Steve ** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/