Messages in this thread |  | | Date | Mon, 18 Nov 2013 16:39:41 +0100 | From | Ingo Molnar <> | Subject | Re: [PROBLEM] possible divide by 0 in kernel/sched/cputime.c scale_stime() |
| |
* Peter Zijlstra <peterz@infradead.org> wrote:
> I would be tempted to say you should simply mark the tsc unstable on > boot and live with that -- we fully assume the sched_clock stuff is > not going backwards in an 'observable' way.
BIOS crap and actual hardware bugs do happen - so kernel code needs to consider TSC input with a pinch of salt, assuming that it's untrusted external data.
> That said, it might be nice to not crash either..
Indeed, not crashing on weird TSC input is absolutely required!
Thanks,
Ingo
|  |