No worries. I'd rather get this right the first time before we convert 100+ tests than afterwards :)
I also appreciate that you've converted some of the tests to use piglit_run_selected_subtests. I've lamented with others that it's unfortunate that we've never really used that feature, since it would be nice for debugging. I've left you a few comments on github, but we should send it to the list before we land it so that others have a chance to look at it. Dylan Quoting Fabian Bieler (2018-01-23 15:35:32) > Hello! > > So after some consideration I think I prefer your first approach > (explicitly calling the enumeration function) after all. > > First of all: Sorry for the back and forth on my part and the excess > work you put into this series because of it. Please don't hate me! 🙇 > > However, I think this approach will be more DRY and less WET. > In particular when used with the existing struct piglit_subtest and > piglit_run_selected_subtests. > > I updated the series accordingly (and fixed a type-o in a docstring in > 5/15 and the commit message of 6/15 and a small arithmetical error in > 11/15). > > The result can be found here: > https://github.com/fabianbieler/piglit/tree/enumerate_subtests > If this isn't acceptable for intermediate review, I'll post to the list, > too. > > Note that I leveraged piglit_run_selected_subtests for gl-2.1-pbo, > fbo-incomplete, degenerate-prims and linestipple and used > piglit_register_subtests in fbo-storage-formats. > > Regards (and sorry again) > Fabian >
signature.asc
Description: signature
_______________________________________________ Piglit mailing list Piglit@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/piglit
