Messages in this thread | | | Date | Thu, 6 May 1999 21:57:22 +0100 | From | Steve Dodd <> | Subject | Re: Oops assist... |
| |
On Thu, May 06, 1999 at 04:26:31PM +0200, BROWN Nick wrote:
> NT has several options when the machine crashes: [snip]
I think this is the way we have to go. I don't think you stand a whelk's chance in a supernova of finding a mechanism that will work for every machine even of the same architecture.
I'd suggest a submenu in the kernel config, allowing options like (off the top of my head): i) write oops data to console / serial / parallel ii) broadcast oops data to network (UDP? raw ethernet frames?) iii) write to (reserved?) hard disk area via BIOS iv) write to IDE hard disk area v) save to nvram vi) store in RAM area conserved across reboot vii) store in VGA card framebuffer conserved across reboot ...
For maximum reliability (if you can use such a word after an oops!), the systems to do the above should try to relying on as little of the rest of the (kernel) world as possible.
Maybe even a module interface so you load an oops-handler at boot time (you wouldn't want on-demand kmod loading, obviously <g>)...
S.
-- "If the bottom is falling out of your world, then have a drink here and the world will fall out of your bottom."
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |