It's really not a good use of time to check for overlaps between the glsl1 tests and the shader runner tests, because that would take a lot of searching. So, the basic question is: do we want to convert these tests to shader runner (and not care about overlap), keep glsl1.c, or just throw glsl1.c away?
I'm honestly not sure how to answer this question. These shader tests are pretty basic, but there are a lot of them, and they might hit on places that piglit doesn't already cover. On Wed, Oct 15, 2014 at 9:39 AM, Matt Turner <[email protected]> wrote: > On Wed, Oct 15, 2014 at 8:13 AM, Brian Paul <[email protected]> wrote: > > I'm not sure that porting this test to Piglit is worthwhile. I'd bet > that > > most of the little subtests here are already sufficiently covered by > > existing piglit shader tests. It would probably be simpler to make > shader > > test scripts for the few cases here that might not be covered already. > > I've already deleted probably a hundred tests from this that are clear > duplicated of existing shader_tests (though there might be some more > as well). > > I completely agree that we'd really like to replace these tests with > individual shader_tests where possible. >
_______________________________________________ Piglit mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/piglit
