KonstantinSeurer
/
mesa
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
mesa/src/freedreno/ci/deqp-freedreno-a530-fails.txt

545 lines
47 KiB
Plaintext
Raw Normal View History

ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES2.functional.clipping.line.wide_line_clip_viewport_center,Fail
dEQP-GLES2.functional.clipping.line.wide_line_clip_viewport_corner,Fail
dEQP-GLES2.functional.clipping.point.wide_point_clip,Fail
dEQP-GLES2.functional.clipping.point.wide_point_clip_viewport_center,Fail
dEQP-GLES2.functional.clipping.point.wide_point_clip_viewport_corner,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES31.functional.image_load_store.early_fragment_tests.early_fragment_tests_stencil,Crash
dEQP-GLES31.functional.image_load_store.early_fragment_tests.early_fragment_tests_stencil_fbo,Crash
dEQP-GLES31.functional.image_load_store.early_fragment_tests.early_fragment_tests_stencil_fbo_with_no_stencil,Crash
dEQP-GLES31.functional.image_load_store.early_fragment_tests.no_early_fragment_tests_stencil,Crash
dEQP-GLES31.functional.image_load_store.early_fragment_tests.no_early_fragment_tests_stencil_fbo,Crash
dEQP-GLES31.functional.image_load_store.early_fragment_tests.no_early_fragment_tests_stencil_fbo_with_no_stencil,Crash
dEQP-GLES31.functional.layout_binding.image.image2d.fragment_binding_array,Crash
dEQP-GLES31.functional.layout_binding.image.image2d.fragment_binding_max,Crash
dEQP-GLES31.functional.layout_binding.image.image3d.fragment_binding_array,Crash
dEQP-GLES31.functional.layout_binding.image.image3d.fragment_binding_max,Crash
dEQP-GLES31.functional.texture.border_clamp.depth_compare_mode.depth24_stencil8.gather_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.depth_compare_mode.depth32f_stencil8.gather_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.depth_compare_mode.depth32f_stencil8.linear_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.depth_compare_mode.depth32f_stencil8.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.depth_compare_mode.depth_component16.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_r11_eac.gather_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_rg11_eac.gather_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_rgb8_etc2.gather_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_rgb8_etc2.nearest_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_rgba8_etc2_eac.gather_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_rgba8_etc2_eac.linear_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_rgba8_etc2_eac.nearest_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_signed_r11_eac.gather_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_signed_r11_eac.linear_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_signed_r11_eac.nearest_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_srgb8_alpha8_etc2_eac.gather_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_srgb8_alpha8_etc2_eac.nearest_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_srgb8_etc2.gather_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_srgb8_etc2.linear_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_srgb8_etc2.nearest_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_srgb8_punchthrough_alpha1_etc2.gather_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.compressed_srgb8_punchthrough_alpha1_etc2.linear_size_tile_multiple,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.depth_component32f.gather_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.depth_component32f.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.luminance_alpha.gather_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.luminance.linear_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.luminance.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.r11f_g11f_b10f.linear_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.r16f.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.r32f.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.r8.gather_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.r8_snorm.gather_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.r8_snorm.linear_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.r8_snorm.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.r8ui.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rg16f.linear_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rg16i.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rg32f.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rg32i.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rg8.gather_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rg8.linear_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rg8.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rg8_snorm.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgb10_a2.gather_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgb10_a2.linear_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgb10_a2.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgb16i.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgb32ui.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgb565.gather_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgb5_a1.linear_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgb5_a1.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgb9_e5.linear_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgba16i.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgba32i.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgba32ui.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgba4.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgba8.gather_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgba8i.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgba8_snorm.linear_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.rgba8_snorm.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.formats.stencil_index8.nearest_size_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.compressed_color.gather.s_mirrored_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.compressed_color.gather.s_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.compressed_color.nearest.s_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.float_color.nearest.s_clamp_to_edge_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.float_color.nearest.s_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.float_depth.gather.s_clamp_to_edge_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.float_depth.gather.s_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.float_depth.nearest.s_mirrored_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.int_color.nearest.s_mirrored_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.snorm_color.gather.s_mirrored_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.snorm_color.linear.s_clamp_to_edge_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.snorm_color.linear.s_mirrored_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.snorm_color.linear.s_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.snorm_color.nearest.s_mirrored_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.unorm_color.gather.s_mirrored_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.unorm_color.nearest.s_clamp_to_edge_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.unorm_depth.nearest.s_mirrored_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_2d.unorm_depth.nearest.s_repeat_t_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_3d.snorm_color.linear.s_mirrored_repeat_t_clamp_to_border_r_repeat_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_3d.snorm_color.linear.s_repeat_t_mirrored_repeat_r_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_3d.snorm_color.nearest.s_clamp_to_border_t_clamp_to_border_r_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_3d.snorm_color.nearest.s_clamp_to_border_t_clamp_to_border_r_repeat_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_3d.uint_color.nearest.s_clamp_to_border_t_clamp_to_border_r_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_3d.uint_color.nearest.s_mirrored_repeat_t_clamp_to_border_r_repeat_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_3d.uint_color.nearest.s_repeat_t_mirrored_repeat_r_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_3d.unorm_color.linear.s_clamp_to_border_t_clamp_to_border_r_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_3d.unorm_color.linear.s_clamp_to_border_t_clamp_to_border_r_repeat_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.per_axis_wrap_mode.texture_3d.unorm_color.linear.s_repeat_t_mirrored_repeat_r_clamp_to_border_pot,Fail
dEQP-GLES31.functional.texture.border_clamp.range_clamp.linear_compressed_color,Fail
dEQP-GLES31.functional.texture.border_clamp.sampler.int_color,Fail
dEQP-GLES31.functional.texture.border_clamp.sampler.uint_stencil,Fail
dEQP-GLES31.functional.texture.border_clamp.sampler.unorm_depth,Fail
dEQP-GLES31.functional.texture.border_clamp.unused_channels.r16i,Fail
dEQP-GLES31.functional.texture.border_clamp.unused_channels.r8,Fail
dEQP-GLES3.functional.clipping.line.wide_line_clip_viewport_corner,Fail
dEQP-GLES3.functional.clipping.point.wide_point_clip_viewport_center,Fail
dEQP-GLES3.functional.fbo.blit.depth_stencil.depth24_stencil8_scale,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.fbo.blit.depth_stencil.depth24_stencil8_stencil_only,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.fbo.blit.depth_stencil.depth32f_stencil8_scale,Fail
dEQP-GLES3.functional.fbo.blit.depth_stencil.depth32f_stencil8_stencil_only,Fail
dEQP-GLES3.functional.fbo.blit.rect.nearest_consistency_mag,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.fbo.blit.rect.nearest_consistency_mag_reverse_dst_y,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.fbo.blit.rect.nearest_consistency_mag_reverse_src_dst_y,Fail
dEQP-GLES3.functional.fbo.blit.rect.nearest_consistency_mag_reverse_src_y,Fail
dEQP-GLES3.functional.fbo.blit.rect.nearest_consistency_min_reverse_dst_x,Fail
dEQP-GLES3.functional.fbo.blit.rect.nearest_consistency_min_reverse_src_dst_x,Fail
dEQP-GLES3.functional.fbo.blit.rect.nearest_consistency_min_reverse_src_x,Fail
dEQP-GLES3.functional.fbo.invalidate.sub.unbind_blit_depth,Fail
dEQP-GLES3.functional.fbo.invalidate.sub.unbind_blit_msaa_depth,Fail
dEQP-GLES3.functional.fbo.invalidate.whole.unbind_blit_color,Fail
dEQP-GLES3.functional.fbo.invalidate.whole.unbind_blit_msaa_color,Fail
dEQP-GLES3.functional.fbo.msaa.2_samples.depth32f_stencil8,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.fbo.msaa.2_samples.stencil_index8,Fail
dEQP-GLES3.functional.fbo.msaa.4_samples.depth32f_stencil8,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.fbo.msaa.4_samples.stencil_index8,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.shaders.texture_functions.texturegradoffset.sampler2darray_fixed_vertex,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.shaders.texture_functions.texturegradoffset.sampler2dshadow_vertex,Fail
dEQP-GLES3.functional.shaders.texture_functions.texturegradoffset.sampler3d_fixed_vertex,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.shaders.texture_functions.texturegradoffset.sampler3d_float_vertex,Fail
dEQP-GLES3.functional.shaders.texture_functions.textureprojgradoffset.sampler2dshadow_vertex,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.shaders.texture_functions.textureprojgradoffset.sampler3d_fixed_vertex,Fail
dEQP-GLES3.functional.shaders.texture_functions.textureprojgradoffset.sampler3d_float_vertex,Fail
dEQP-GLES3.functional.shaders.texture_functions.textureprojgrad.sampler2dshadow_vertex,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.highp_float,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.highp_ivec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.highp_ivec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.highp_mat2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.highp_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.highp_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.highp_uvec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.highp_vec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.lowp_float,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.lowp_ivec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.lowp_ivec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.lowp_mat2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.lowp_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.lowp_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.lowp_uvec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.lowp_vec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.mediump_int,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.mediump_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.mediump_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.mediump_uint,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.mediump_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.mediump_vec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.lines.mediump_vec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.highp_int,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.highp_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.highp_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.highp_uint,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.highp_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.highp_vec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.highp_vec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.lowp_int,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.lowp_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.lowp_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.lowp_uint,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.lowp_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.lowp_vec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.lowp_vec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.mediump_float,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.mediump_ivec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.mediump_ivec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.mediump_mat2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.mediump_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.mediump_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.mediump_uvec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.interleaved.points.mediump_vec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.highp_int,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.highp_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.highp_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.highp_uint,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.highp_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.highp_vec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.highp_vec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.lowp_int,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.lowp_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.lowp_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.lowp_uint,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.lowp_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.lowp_vec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.lowp_vec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.mediump_float,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.mediump_ivec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.mediump_ivec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.mediump_mat2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.mediump_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.mediump_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.mediump_uvec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.lines.mediump_vec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.highp_float,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.highp_ivec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.highp_ivec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.highp_mat2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.highp_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.highp_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.highp_uvec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.highp_vec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.lowp_float,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.lowp_ivec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.lowp_ivec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.lowp_mat2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.lowp_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.lowp_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.lowp_uvec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.lowp_vec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.mediump_int,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.mediump_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.mediump_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.mediump_uint,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.mediump_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.mediump_vec2,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.points.mediump_vec4,Fail
dEQP-GLES3.functional.transform_feedback.array_element.separate.triangles.mediump_vec4,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.highp_float,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.highp_ivec2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.highp_ivec4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.highp_mat2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.highp_mat2x4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.highp_mat3,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.highp_mat4,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.highp_mat4x2,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.highp_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.highp_uvec4,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.highp_vec3,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.lowp_float,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.lowp_ivec2,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.lowp_ivec4,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.lowp_mat2,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.lowp_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.lowp_mat3,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.lowp_mat4,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.lowp_mat4x2,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.lowp_uvec2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.lowp_uvec4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.lowp_vec3,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.mediump_int,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.mediump_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.mediump_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.mediump_mat3x2,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.mediump_mat3x4,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.mediump_mat4x3,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.mediump_uint,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.mediump_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.mediump_vec2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.lines.mediump_vec4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.highp_int,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.highp_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.highp_mat2x3,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.highp_mat3x2,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.highp_mat3x4,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.highp_mat4x3,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.highp_uint,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.highp_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.highp_vec2,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.highp_vec4,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.lowp_int,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.lowp_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.lowp_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.lowp_mat3x2,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.lowp_mat3x4,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.lowp_mat4x3,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.lowp_uint,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.lowp_uvec3,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.lowp_vec2,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.lowp_vec4,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.mediump_float,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.mediump_ivec2,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.mediump_ivec4,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.mediump_mat2,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.mediump_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.mediump_mat3,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.mediump_mat4,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.mediump_mat4x2,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.mediump_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.mediump_uvec4,Fail
dEQP-GLES3.functional.transform_feedback.array.interleaved.points.mediump_vec3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.highp_int,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.highp_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.highp_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.highp_mat3x2,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.lines.highp_mat3x4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.lines.highp_mat4x3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.highp_uint,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.lines.highp_uvec3,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.lines.highp_vec2,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.highp_vec4,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.lines.lowp_int,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.lines.lowp_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.lowp_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.lowp_mat3x2,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.lowp_mat3x4,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.lowp_mat4x3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.lowp_uint,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.lowp_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.lowp_vec2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.lines.lowp_vec4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.lines.mediump_float,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.mediump_ivec2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.lines.mediump_ivec4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.lines.mediump_mat2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.lines.mediump_mat2x4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.lines.mediump_mat3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.mediump_mat4,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.mediump_mat4x2,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.mediump_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.mediump_uvec4,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.lines.mediump_vec3,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.highp_float,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.highp_ivec2,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.highp_ivec4,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.highp_mat2,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.highp_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.highp_mat3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.highp_mat4,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.highp_mat4x2,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.highp_uvec2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.highp_uvec4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.highp_vec3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.lowp_float,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.lowp_ivec2,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.lowp_ivec4,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.lowp_mat2,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.lowp_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.lowp_mat3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.lowp_mat4,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.lowp_mat4x2,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.lowp_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.lowp_uvec4,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.lowp_vec3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.mediump_int,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.mediump_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.mediump_mat2x3,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.mediump_mat3x2,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.mediump_mat3x4,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.mediump_mat4x3,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.mediump_uint,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.array.separate.points.mediump_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.mediump_vec2,Fail
dEQP-GLES3.functional.transform_feedback.array.separate.points.mediump_vec4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.highp_float,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.highp_ivec2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.highp_ivec4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.highp_mat2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.highp_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.highp_mat3,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.highp_mat4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.highp_mat4x2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.highp_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.highp_uvec4,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.highp_vec3,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.lowp_float,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.lowp_ivec2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.lowp_ivec4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.lowp_mat2,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.lowp_mat2x4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.lowp_mat3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.lowp_mat4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.lowp_mat4x2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.lowp_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.lowp_uvec4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.lowp_vec3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.mediump_int,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.mediump_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.mediump_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.mediump_mat3x2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.mediump_mat3x4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.mediump_mat4x3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.mediump_uint,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.mediump_uvec3,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.mediump_vec2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.lines.mediump_vec4,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.highp_int,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.highp_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.highp_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.highp_mat3x2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.highp_mat3x4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.highp_mat4x3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.highp_uint,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.highp_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.highp_vec2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.highp_vec4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.lowp_int,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.lowp_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.lowp_mat2x3,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.lowp_mat3x2,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.lowp_mat3x4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.lowp_mat4x3,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.lowp_uint,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.lowp_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.lowp_vec2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.lowp_vec4,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.mediump_float,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.mediump_ivec2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.mediump_ivec4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.mediump_mat2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.mediump_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.mediump_mat3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.mediump_mat4,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.mediump_mat4x2,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.mediump_uvec2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.mediump_uvec4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.interleaved.points.mediump_vec3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.highp_int,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.highp_ivec3,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.highp_mat2x3,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.highp_mat3x2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.highp_mat3x4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.highp_mat4x3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.highp_uint,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.highp_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.highp_vec2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.highp_vec4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.lowp_int,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.lowp_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.lowp_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.lowp_mat3x2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.lowp_mat3x4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.lowp_mat4x3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.lowp_uint,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.lowp_uvec3,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.lowp_vec2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.lowp_vec4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.mediump_float,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.mediump_ivec2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.mediump_ivec4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.mediump_mat2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.mediump_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.mediump_mat3,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.mediump_mat4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.mediump_mat4x2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.mediump_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.mediump_uvec4,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.lines.mediump_vec3,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.highp_float,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.highp_ivec2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.highp_ivec4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.highp_mat2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.highp_mat2x4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.highp_mat3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.highp_mat4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.highp_mat4x2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.highp_uvec2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.highp_uvec4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.highp_vec3,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.lowp_float,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.lowp_ivec2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.lowp_ivec4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.lowp_mat2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.lowp_mat2x4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.lowp_mat3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.lowp_mat4,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.lowp_mat4x2,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.lowp_uvec2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.lowp_uvec4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.lowp_vec3,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.mediump_int,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.mediump_ivec3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.mediump_mat2x3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.mediump_mat3x2,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.mediump_mat3x4,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.mediump_mat4x3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.mediump_uint,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.mediump_uvec3,Fail
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.mediump_vec2,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.basic_types.separate.points.mediump_vec4,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.interpolation.centroid.highp_vec4_lines_separate,Fail
dEQP-GLES3.functional.transform_feedback.interpolation.centroid.highp_vec4_points_separate,Fail
dEQP-GLES3.functional.transform_feedback.interpolation.centroid.lowp_vec4_lines_separate,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.interpolation.centroid.lowp_vec4_points_separate,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.interpolation.centroid.mediump_vec4_lines_separate,Fail
dEQP-GLES3.functional.transform_feedback.interpolation.centroid.mediump_vec4_points_separate,Fail
dEQP-GLES3.functional.transform_feedback.interpolation.flat.highp_vec4_lines_separate,Fail
dEQP-GLES3.functional.transform_feedback.interpolation.flat.highp_vec4_points_separate,Fail
dEQP-GLES3.functional.transform_feedback.interpolation.flat.lowp_vec4_lines_separate,Fail
dEQP-GLES3.functional.transform_feedback.interpolation.flat.lowp_vec4_points_separate,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.interpolation.flat.mediump_vec4_lines_separate,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.interpolation.flat.mediump_vec4_points_separate,Fail
dEQP-GLES3.functional.transform_feedback.interpolation.smooth.highp_vec4_lines_separate,Fail
dEQP-GLES3.functional.transform_feedback.interpolation.smooth.highp_vec4_points_separate,Fail
dEQP-GLES3.functional.transform_feedback.interpolation.smooth.lowp_vec4_lines_separate,Fail
dEQP-GLES3.functional.transform_feedback.interpolation.smooth.lowp_vec4_points_separate,Fail
dEQP-GLES3.functional.transform_feedback.interpolation.smooth.mediump_vec4_lines_separate,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.interpolation.smooth.mediump_vec4_points_separate,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.point_size.lines_separate,Fail
dEQP-GLES3.functional.transform_feedback.point_size.points_separate,Fail
ci/deqp: Switch to a new dEQP runner written in Rust. I found the C++ runner hard to develop on, and we had stability issues and outstanding feature needs that made me want something I felt good about hacking on. Thus, Rewrite It In Rust of the deqp runner. The new runner includes: - Skip lists don't reshuffle the test list. - Known-flake handling without resorting to skip lists (fixing our main CI reliability issue on a3xx right now). - Per-thread Vulkan shader caches should speed up VK CI runtime. - Tracking of crashes separate from fails (so we can see progress on that front). - Logging of deqp stderr spam (particularly assertion failures!) in the CI log. - Integrated QPA filtering so we don't have bash perf issues for it. - Logging of what caselist to go look at for a given error report (in red, so it's easier to find in your CI log). - The code is 1/3 unit tests, and easy to extend for more coverage. - Non-LAVA CI runs create a failures.csv in artifacts that you can check in as your deqp-*-fails.txt file. - Test runtime is included in results.csv so you can debug how to speed up your CI job. - Pretty summary at the end of the run of slow/flaky/failed tests. Since this is a new runner with a different RNG, the test groups are shuffled one more time. This seems to result in some panfrost T720 stability issues (See its new deqp-panfrost-t720-flakes.txt), and one new flake in freedreno a630. Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7434>
2020-10-29 17:29:28 +00:00
dEQP-GLES3.functional.transform_feedback.position.lines_separate,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.position.points_separate,Fail
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.interleaved.lines.1,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.interleaved.lines.3,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.interleaved.lines.5,Fail
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.interleaved.lines.7,Fail
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.interleaved.lines.9,Fail
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.interleaved.points.1,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.interleaved.points.3,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.interleaved.points.5,Fail
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.interleaved.points.7,Fail
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.interleaved.points.9,Fail
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.separate.lines.1,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.separate.lines.3,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.separate.lines.5,Fail
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.separate.lines.7,Fail
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.separate.lines.9,Fail
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.separate.points.1,Fail
ci/a5xx: Increase our dEQP GLES3 fraction by 4x. Now that we've got SMP, we can get a lot more of this test suite covered in our 10-minute job window. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9144>
2021-02-18 20:28:27 +00:00
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.separate.points.3,Fail
ci/a5xx: Increase the gles3/31 coverage. Now that there's more time available in our budget per board, we can run all of gles31, and half of gles3, instead of 10%. Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9314>
2021-02-26 20:26:08 +00:00
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.separate.points.5,Fail
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.separate.points.7,Fail
dEQP-GLES3.functional.transform_feedback.random_full_array_capture.separate.points.9,Fail
dEQP-GLES3.functional.transform_feedback.random.interleaved.lines.1,Fail
dEQP-GLES3.functional.transform_feedback.random.interleaved.lines.3,Fail
dEQP-GLES3.functional.transform_feedback.random.interleaved.lines.5,Fail
dEQP-GLES3.functional.transform_feedback.random.interleaved.lines.7,Fail
dEQP-GLES3.functional.transform_feedback.random.interleaved.lines.9,Fail
dEQP-GLES3.functional.transform_feedback.random.interleaved.points.1,Fail
dEQP-GLES3.functional.transform_feedback.random.interleaved.points.3,Fail
dEQP-GLES3.functional.transform_feedback.random.interleaved.points.5,Fail
dEQP-GLES3.functional.transform_feedback.random.interleaved.points.7,Fail
dEQP-GLES3.functional.transform_feedback.random.interleaved.points.9,Fail
dEQP-GLES3.functional.transform_feedback.random.separate.lines.1,Fail
dEQP-GLES3.functional.transform_feedback.random.separate.lines.3,Fail
dEQP-GLES3.functional.transform_feedback.random.separate.lines.5,Fail
dEQP-GLES3.functional.transform_feedback.random.separate.lines.7,Fail
dEQP-GLES3.functional.transform_feedback.random.separate.lines.9,Fail
dEQP-GLES3.functional.transform_feedback.random.separate.points.1,Fail
dEQP-GLES3.functional.transform_feedback.random.separate.points.3,Fail
dEQP-GLES3.functional.transform_feedback.random.separate.points.5,Fail
dEQP-GLES3.functional.transform_feedback.random.separate.points.7,Fail
dEQP-GLES3.functional.transform_feedback.random.separate.points.9,Fail