On Tue, Aug 05, 2003 at 11:14:16PM +0200, Andi Kleen wrote:
> > Otherwise this will just keep on expanding.
>
> It does expand on i386 exactly because the watchdog is disabled by default.
>
> Looks like a mistake to me. It should be on because having usable backtraces
> on a deadlock/hang is useful enough that it outweights any other possible
> disadvantages. That's especially true for kernels out there at user's boxes,
> not just special debugging kernels run by developers.
>
> [if there should be any hardware where it doesn't work it should be blacklisted
> there]
the reason it's off is that certain IBM bioses corrupt the eax register on
NMI's when they collide with smm stuff... You'd be surprised how
tolerant x86 is against such corruptions... but not 100% :)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Aug 07 2003 - 22:00:30 EST