On 2018-09-06 8:11 p.m., Marek Olšák wrote: > On Thu, Sep 6, 2018 at 6:25 AM, Michel Dänzer <mic...@daenzer.net> wrote: >> On 2018-09-05 11:49 p.m., Marek Olšák wrote: >>> On Tue, Sep 4, 2018 at 2:35 PM, Marek Olšák <mar...@gmail.com> wrote: >>>> On Mon, Sep 3, 2018 at 6:08 AM, Michel Dänzer <mic...@daenzer.net> wrote: >>>>> On 2018-09-01 8:54 a.m., Marek Olšák wrote: >>>>>> From: Marek Olšák <marek.ol...@amd.com> >>>>>> >>>>>> +41% performance in debug builds >>>>>> (testing piglit/drawoverhead + u_threaded_context) >>>>> >>>>> Nice, but please include ministat output. >>>> >>>> I don't use ministat. >>> >>> What I meant is that ministat isn't useful with drawoverhead, because >>> drawoverhead repeats testing if results are volatile until it finds a >>> good average. >> >> Instead of that, it would be better to run the test multiple times, and >> let ministat take care of the statistical details. > > Oh come on. The patch is too trivial to warrant so much testing.
My feedback to you (in general, not just for this series) is not to put meaningless numbers (which most likely suggest more accuracy than you were able to measure) in commit logs. Either produce more meaningful numbers (use "git log --grep='(n='" for examples), or don't mention any numbers. > Besides, the improvement is consistently around ~41%. More meaningful numbers should look very good then. :) -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
