On 15.05.2014 21:44, Brian Paul wrote: > On 05/14/2014 09:53 PM, Michel Dänzer wrote: >> On 14.05.2014 10:06, Brian Paul wrote: >>> >>> diff --git a/tests/spec/arb_occlusion_query/occlusion_query.c >>> b/tests/spec/arb_occlusion_query/occlusion_query.c >>> index e099f06..b8ea664 100644 >>> --- a/tests/spec/arb_occlusion_query/occlusion_query.c >>> +++ b/tests/spec/arb_occlusion_query/occlusion_query.c >>> @@ -34,9 +34,6 @@ >>> PIGLIT_GL_TEST_CONFIG_BEGIN >>> >>> config.supports_gl_compat_version = 10; >>> - >>> - config.window_width = 180; >>> - config.window_height = 100; >> >> This change breaks the test for me, because the window is no longer wide >> enough to contain everything rendered by the test. >> >> >> The spec/NV_conditional_render/clear test was also broken for me by the >> corresponding change, but there it's not obvious to me whether it's a >> test or driver bug. >> >> >> How are you testing changes like these? > > Generally using nvidia's driver, plus _some_ double-checking with > llvmpipe (when it supports the extension in question). I messed up on > the occlusion_query.c change - I'll revert that. > spec/NV_conditional_render/clear passes before and after the change w/ > nvidia's driver. > > Sorry for the mistake, Michel.
No worries, thanks for fixing it Brian. -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer _______________________________________________ Piglit mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/piglit
