Messages in this thread | | | Date | Wed, 6 Jul 2011 11:35:01 +0200 | From | Ingo Molnar <> | Subject | Re: [PATCH 1/4] perf, x86: Add Intel Nehalem/Westmere uncore pmu |
| |
* Andi Kleen <andi@firstfloor.org> wrote:
> > > > >> But it's not strictly required I would say, > >> 44(?) bits are probably enough for near all use cases. > > > > 44bits is in the hours range for pure cycle counts, which is > > so-so. I bet you're going to be very annoyed when you find your > > counters are wrecked after your 5 hour test run finishes. > > For these kinds of long running measurements you usually care about > ratios, not absolutes. Some infrequent wrapping should not > substantially change. those.
That's just empty blathering - those who *do* take a look at long-running absolute values obviously care.
There's absolutely no reason to be sloppy here - this needs to be implemented correctly.
Thanks,
Ingo
| |