Messages in this thread | | | Date | Wed, 17 May 2006 11:13:28 +1000 | From | Peter Chubb <> | Subject | Re: [RFC] [Patch 7/8] statistics infrastructure - exploitation prerequisite |
| |
>>>>> "Martin" == Martin Peschke <mp3@de.ibm.com> writes:
Martin> I would be happy to exploit an API that may result from that Martin> discussion. I would plead for exporting such an API to Martin> modules. I don't see how to implement statistics for Martin> latencies, otherwise.
Martin> Any other hints on how to replace my sched_clock() calls are Martin> welcome. (I want to measure elapsed times in units that are Martin> understandable to users without hardware manuals and Martin> calculator, such as milliseconds.)
You may wish to look at the microstate accounting patches at http://www.gelato.unsw.edu.au/cgi-bin/viewcvs.cgi/cvs/kernel/microstate/ I made the clock source configurable.
It's acually rather difficult to find a reliable cross-platform monotonic clock with good resolution. And different statistics may want different clocks; for my purposes a cycle counter is best (because I'm running on a machine that varies the clock speed for power management reasons), other people may want nanoseconds.
-- Dr Peter Chubb http://www.gelato.unsw.edu.au peterc AT gelato.unsw.edu.au http://www.ertos.nicta.com.au ERTOS within National ICT Australia - 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/
| |