Re: 99% System load

[Andrew Morton]

Florian Schanda <ma1flfs@xxxxxxxxxx> wrote:
>
> Finally it happened again. I attached the output of that dmesg.
> 
> > It would help if you could make a note of the PIDs of the hung processes,
> > so they can be correlated with the sysrq output.
> 
> The PIDs hanging are 3303, 3305, 3306, 3307.

It _looks_ like all four CPUs are madly taking timer interrupts.  Which
tends to point at some APIC problem, failing to clear the interrupt source.
But if that happened one wouldn't expect userspace to remain in a runnable
state, and you say that you can still run commands.

> An advance warning: I had the nvidia driver (yes, I know, closed source makes 
> debugging impossible) loaded, and before the trace begins, there are some 
> messages of that module:

It is probably unrelated, but I'd suggest that you not use the nvidia
driver for a while, verify that the lockup happens without it loaded.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/