develooper Front page | perl.perl5.porters | Postings from July 2013

Re: [perl.git] branch blead, updated. v5.19.0-497-g1ebabb4

Thread Previous | Thread Next
George Greer
July 15, 2013 13:19
Re: [perl.git] branch blead, updated. v5.19.0-497-g1ebabb4
Message ID:
On Sun, 14 Jul 2013, Dave Mitchell wrote:
> From this I conclude that it's not an issue with Benchmark, but an issue
> with the test file expecting that the OS will return a reasonably
> consistent CPU-seconds burn for a given task.
> Having said that, I've been completely unable to reproduce this at home,
> even on a server I ramped up to a load average of 300+. I tried having the
> high load only for the first half of the calibration loop, only on the
> second, or on both. Over many runs, I never saw more than a few %
> deviation between the two loops-per-sec values.
> But given that the smokes have failed on the platforms run by at least 3
> different volunteers, over multiple platforms (linux, darwin, win32, aix
> etc) I now propose to remove the infamous test 15 from Benchmark.t
> completely.
> After that I'll look at failures 128,129 and see if the same thing applies
> to them.

At least on my machine (Core i7-920 2.67GHz, 4core/8thread, Nehalem), the 
benchmark probably fails when the CPU adjusts its frequency.  Normally it 
runs at 2675MHz but sometimes the Test::Smoke CPU identification lists it 
as 1600MHz, which could be due to either heat throttling or sleep states.

If you have a laptop you might be able to reproduce the effect by pulling 
out the power cable and making it run on battery for the last half of the 
test, which should throttle the CPU down lower.

Not that the test could likely compensate for what the CPU/OS is up to 

George Greer

Thread Previous | Thread Next Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at | Group listing | About