- Jun 11, 2020
-
-
Dave Airlie authored
-
Dave Airlie authored
The resource may have been written to as images previously. KHR-GL45.shader_image_load_store.advanced-sync-imageAccess2
-
Dave Airlie authored
Fixes: dEQP-VK.robustness.buffer_access*
-
Dave Airlie authored
TGSI expect vec4 of constants for it's current code paths, and when doing indirect accesses it does the comparison on vec4 indexes, however NIR does the indexing on packed float indexes. This also align the compute path with the other shaders, and should improve robustness (at least under Vulkan)
-
Dave Airlie authored
This fixes two bugs, one in clearing and one in sign extensions for S8 only types, and enables it for use.
-
Dave Airlie authored
-
Dave Airlie authored
Fixes: GTF-GL45.gtf30.GL3Tests.sgis_texture_lod.sgis_texture_lod_basic_lod_selection when running with filter hacks disable
-
Dave Airlie authored
-
Dave Airlie authored
The way we construt vertex/geom shaders means these can diverge, so we have to just hammer it out manually, there are likely optimisation opportuniities in here
-
Dave Airlie authored
This causes correct flushing: KHR-GL46.shader_image_load_store.non-layered_binding
-
Dave Airlie authored
-
Dave Airlie authored
-
Dave Airlie authored
This fixes tess factors with gl spirv
-
Dave Airlie authored
-
Dave Airlie authored
-
Dave Airlie authored
-
Dave Airlie authored
Do this earlier, gets rid of lots of useless operations
-
Dave Airlie authored
Images only use the format field, and images in an array seem to have the same format(??), so we just have to handle the case where no image is bound, when format is NONE.
-
Dave Airlie authored
Only good enough for GL
-
Dave Airlie authored
This fixes remaining test for ARB_gpu_shader5
-
Dave Airlie authored
-
Dave Airlie authored
-
Dave Airlie authored
This passes the dynamic offset value through the sampler code.
-
Dave Airlie authored
-
Dave Airlie authored
-
Dave Airlie authored
Only good enough for GL
-
Dave Airlie authored
-
Dave Airlie authored
We have to run invocations per primitive, not primitives per invocation TODO fix softpipe
-
Dave Airlie authored
-
Dave Airlie authored
Need to think about the sizings here a bit more
-
Dave Airlie authored
The current code wasn't tracking the primitve lenghts properly per stream and wasn't handling the end primitives properly for the non-0 stream. If the 0 stream has no primitives, the code would end instead of doing streamout.
-
Dave Airlie authored
Fixes: KHR-GL45.transform_feedback.draw_xfb_stream_test
-
Dave Airlie authored
This fixes piglit arb_viewport_array-render-viewport-2
-
Dave Airlie authored
This doesn't pass thie piglits because currently they are broken for case where GL upgrades 2 samples to 4
-
Dave Airlie authored
Some tests were using a 1x1 fb bound, with a 2x2 viewport, and all 4 pixels were getting rendered. Test if the fb bounds need planes added or not. Closes: mesa/mesa#3101
-
- Jun 10, 2020
-
-
Dave Airlie authored
This has to be unsigned, so clamping works properly for border colors. Fixes dEQP-GLES31.functional.texture.border_clamp.range_clamp.nearest_uint_stencil
-
Dave Airlie authored
I'm not sure why this code was if (0), but if (1) for it fixes dEQP-GLES31.functional.texture.border_clamp.range_clamp.nearest_float_color This test expects +inf to get mapped to 255 and -inf to 0, both values were ending up at 0.
-
Dave Airlie authored
This fixes: dEQP-GLES31.functional.draw_indirect.random.2 which ends up with 3x32-bit USCALED values going down this path some of which have the top bit set, and end up converted to signed float instead of unsigned float values.
-
Dave Airlie authored
This fixes some vulkan tests later.
-
Dave Airlie authored
Fixes crash in dEQP-GLES31.functional.synchronization.inter_invocation.image_write_read
-