Front page | perl.perl5.porters |
Postings from January 2001
On Tue, Jan 23, 2001 at 05:08:28PM +0000, Nick Ing-Simmons wrote: > Nicholas Clark <nick@ccl4.org> writes: > > > >real 2m16.388s > >user 1m3.610s > >sys 0m11.000s > > > > > >(which is from time ./perl TEST) > > > >I was under the impression that real is wallclock, and the other two > >are user and system call CPU time, not elapsed time. > > Of course. And the wall clock time is known useless. > But if you get consistent user/sys times running the same perl > on the same suite then you are lucky. > > I seem to get at least ±5% without changing ANYTHING apart from typing > !!<cr> > at the prompt. This is the sort of variation I was seeing some of the time. However, today the times for the 2 inline cases seems reasonably repeatable. Maybe as part of a test harness suite it would be useful it had "run the test n times" and could show the average (and the variation) to give a better estimate. Nicholas ClarkThread Previous | Thread Next