Re: Using sched_clock for mmio-trace

From: Frank Ch. Eigler
Date: Fri Feb 16 2007 - 16:27:05 EST


Hi -

On Fri, Feb 16, 2007 at 09:03:23PM +0100, Andi Kleen wrote:
> > We in systemtap land have the same problem, and so far made do with
> > slightly postprocessed per-cpu TSC values.
>
> 90+% likely you're not solving your problem correctly this way.

Yes, it was done as a last resort.

We need facility that we can call from even more demanding contexts
than interrupt handlers, considering that kprobes can be placed nearly
anywhere. This is one of the reasons why we don't just use good old
do_gettimeofday(), since it takes locks and can lead to lock recursion
if parts of itself are probed.

- FChE
-
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/