On Fri, Jun 09, 2006 at 11:15:34PM +0200, Frank Kardel wrote:
> 
> Maybe we should consider adding a fast unsynchronized time abstraction
> for raw counter samples that can be scaled later to favor fast light weight
> time stamping (syscall-, user-, system-, trap- and interrupt-time  
> measurement come immediately to mind here).

And process executions times.

I have some code 95% working that measures the time spent in each
system call number.  Since the units are irrelevant this uses the
existing timecounter32() values.
In this case it doesn't even matter if the frequency changes...

I just need to sort out the trap and interrupt entries.

	David

-- 
David Laight: david@l8s.co.uk