Re: [PATCH] perf: accommodate for MacOSX

[Johannes Schindelin <Johannes.Schindelin@gmx.de>]

Hi Lars,

On Tue, 21 Jun 2016, Lars Schneider wrote:

> > On 21 Jun 2016, at 13:55, Johannes Schindelin <Johannes.Schindelin@gmx.de> wrote:
> > 
> >> ...
> >> If we don't run any perf tests by default on Travis CI then I wouldn't
> >> take the ".travis.yml" part of the patch just to keep our Travis CI
> >> setup as lean as possible.
> > 
> > Maybe commented-out, so that people like me have a chance to use Travis
> > for MacOSX perf testing?
> > 
> > Could you let me know whether a commented-out
> > 
> > 	# Uncomment this if you want to run perf tests:
> > 	# brew install gnu-time
> > 
> > would be acceptable? I will reroll the patch accordingly.
> 
> Commented-out would be fine with me!

Okay! Updated patch coming in a moment.

> >> Running perf tests on Travis CI is probably bogus anyways because we
> >> never know on what hardware our jobs run and what other jobs run in
> >> parallel on that hardware.
> > 
> > While I agree that the absolute timings cannot be trusted, I have to point
> > out that the relative timings on Linux at least are consistent with what I
> > could test locally.
>
> Given that the relative timings are consistent for you. Maybe there is
> value to run the performance tests (e.g. only on the master branch)
> in a separate Travis job. Then we could chart the timings over releases.
> I dunno.

Sure, that would be a fine addition, if there is a way, somehow, to chart
the timings (keep in mind that Travis runs for each and every iteration of
each and every PR, in addition to the branch updates).

I do have enough on my plate, though, so I will have to hope that somebody
else picks this up.

Ciao,
Dscho