Everyone keeps around a bit of their own... http://people.freedesktop.org/~imirkin/ has some links to some results, including cross-hardware combinations, and the source files are at https://people.freedesktop.org/~imirkin/piglit/ . Getting thousands of failures is definitely unexpected. (My results are a bit old, admittedly.) You can see Martin's runs at http://piglit.mupuf.org/results/ on a variety of (nvidia-based) hardware. [I highly recommend only ever going to problems.html - all.html takes forever to load.)
I'm not aware of a centralized repository for this. Unfortunately there tends to be a lot of factors, including piglit/mesa versions, various support libraries (e.g. llvm for radeonsi, etc), kernel versions, etc. This can make direct or automated comparisons tricky. -ilia On Sat, Sep 3, 2016 at 8:23 PM, Dan Kegel <[email protected]> wrote: > Hi all, > I'm using piglit to make sure my continuous build rigs have good > opengl hardware/software. > The number of test failures in piglit is rather high (thousands), and > I suspect that's > not unexpected. > > Is there an archive like > https://people.freedesktop.org/~nh/piglit/results/all/index.html > anywhere, but much larger, of piglit results vs. hardware/software > combinations? > _______________________________________________ > Piglit mailing list > [email protected] > https://lists.freedesktop.org/mailman/listinfo/piglit _______________________________________________ Piglit mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/piglit
