Misleading output for differing values of warmup JVM forks and benchmark JVM forks

[Daniel Mitterdorfer]

That was fast... Regarding 0.4: Sorry, got it wrong... . I just saw 0.4
already mentioned in commit comments and misleadingly assumed it was
already out. Thanks anyway. :)

Daniel


2014-02-19 20:45 GMT+01:00 Aleksey Shipilev <aleksey.shipilev at oracle.com>:

> On 02/19/2014 11:41 PM, Daniel Mitterdorfer wrote:
> > I've just stumpled upon a minor problem regarding the command line output
> > in jmh 0.4.
>
> 0.4 have not yet pushed, and we can probably include the fix there :)
>
> > Consider a benchmark annotated with @Fork(value = 3, warmups =
> > 5). JMH will then output:
> >
> > # Warmup Fork: 1 of 3
> > # Warmup Fork: 2 of 3
> > # Warmup Fork: 3 of 3
> > # Warmup Fork: 4 of 3
> > # Warmup Fork: 5 of 3
> >
> > which puzzled me a bit at first. :-)
> >
> > It boils down to org.openjdk.jmh.runner.Runner#runSeparate(..):
> >
> > out.println("# Warmup Fork: " + (i + 1) + " of " + forkCount);
> >
> > which should be
> >
> > out.println("# Warmup Fork: " + (i + 1) + " of " + warmupForkCount);
>
> Yup, thanks! Fixed:
>   http://hg.openjdk.java.net/code-tools/jmh/rev/a3a905ff7a94
>
> -Aleksey.
>