summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2019-06-26arb_vertex_program: Verify that unsupported functions generate errorsHEADmasterIan Romanick3-0/+116
Once upon a time, Mesa accidentally enabled GL_ARB_vertex_program in Core Profile. We fixed that, but continued to export the functions. We eventually fixed that too. It turns out that, especially in the piglit framework, it's really hard to verify that a function is not there. The piglit framework will detect that a function is being called that is not supported by the implementation, and it will print a friendly message. We don't want that. We want to call the function specifically because we know it's not supported. As a result, glXGetProcAddress has to be used directly. Using glXGetProcAddress poses two problems. First, it can either return NULL or it can return a pointer to a dispatch trampoile function. Some closed-source libGL implementations will return NULL for functions that they and the associated drivers have never heard of. The open-source libGL implementations will never return NULL because the application might later load a new driver that supports the requested function. Try glXGetProcAddress((const GLubyte *)"glHamsandwich") on your favorite libGL. Second, the trampoline function may or may not call a valid function. Depending on the driver and the libGL the dispatch table entry used by the trampoline function may: 1. Point at garbage. 2. Point at a default function that always sets GL_INVALID_OPERATION. 3. Point at a real implementation of the function that may or may not set GL_INVALID_OPERATION. The only way to determine which you've got is by calling the trampoline function. In case #1, this will result in some signal (probably either SIGSEGV or SIGILL) that leads to program termination. By the definition of the GL spec, this is actually success, so crashing is not a good way to reflect that. To deal with this a combination of signal and sigsetjmp are used. For each function to be tested, 1. Call glXGetProcAddress on the function. If glXGetProcAddress returns NULL, the function passes. 2. Set signal handlers for every reasonable signal that calling outer space might generate. The signal handler will siglongjmp back to the caller. 3. Use sigsetjmp to set the location to which the signal handler should return. 4. Call the function. If GL_INVALID_OPERATION is set, the function passes. If the signal handler is invoked, the call to glGetError will never happen. If any other condition is set, the function fails. This particular test only calls a subset of the GL_ARB_vertex_program functions. The remaining functions require that a valid vertex program be bound. If all the functions included in this test are successful, either the functions would be too (with possible false positives) or an api-errors test should detect their failures. Either way, I didn't see much point in trying them. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Tested-by: Albert Freeman <albertwdfreeman@gmail.com> Reviewed-by: Albert Freeman <albertwdfreeman@gmail.com>
2019-06-26degenerate-prims: Add immediate-mode drawing testsIan Romanick1-19/+56
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=38109
2019-06-26degenerate-prims: Add tests that draw one primitive and have some extra dataIan Romanick1-1/+7
For GL_LINES, GL_TRIANGLES, GL_QUADS, and GL_QUAD_STRIP, add tests that are one vertex short of two primitives. This should draw something. We don't check exactly what's drawn because, well, it's hard. This is mostly try to reproduce a crash in some Mesa drivers. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=38109
2019-06-26degenerate-prims: Add the ability for tests to expect some drawingIan Romanick1-13/+29
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2019-06-26degenerate-prims: Don't log the subtest result in the subtestIan Romanick1-3/+0
This causes every subtest result to be logged twice: PIGLIT: {"subtest": {"Primitive: GL_POINTS" : "pass"}} PIGLIT: {"subtest": {"GL_POINTS" : "pass"}} PIGLIT: {"subtest": {"Primitive: GL_LINES" : "pass"}} PIGLIT: {"subtest": {"GL_LINES" : "pass"}} PIGLIT: {"subtest": {"Primitive: GL_LINE_STRIP" : "pass"}} PIGLIT: {"subtest": {"GL_LINE_STRIP" : "pass"}} PIGLIT: {"subtest": {"Primitive: GL_LINE_LOOP" : "pass"}} PIGLIT: {"subtest": {"GL_LINE_LOOP" : "pass"}} PIGLIT: {"subtest": {"Primitive: GL_TRIANGLES" : "pass"}} PIGLIT: {"subtest": {"GL_TRIANGLES" : "pass"}} PIGLIT: {"subtest": {"Primitive: GL_TRIANGLE_STRIP" : "pass"}} PIGLIT: {"subtest": {"GL_TRIANGLE_STRIP" : "pass"}} PIGLIT: {"subtest": {"Primitive: GL_TRIANGLE_FAN" : "pass"}} PIGLIT: {"subtest": {"GL_TRIANGLE_FAN" : "pass"}} PIGLIT: {"subtest": {"Primitive: GL_QUADS" : "pass"}} PIGLIT: {"subtest": {"GL_QUADS" : "pass"}} PIGLIT: {"subtest": {"Primitive: GL_QUAD_STRIP" : "pass"}} PIGLIT: {"subtest": {"GL_QUAD_STRIP" : "pass"}} PIGLIT: {"subtest": {"Primitive: GL_POLYGON" : "pass"}} PIGLIT: {"subtest": {"GL_POLYGON" : "pass"}} This will be even more annoying when a later patch adds parameters other than the primitive that control the subtest name. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
2019-06-26arb_shader_objects: Test some edge cases of glGetAttachedObjectARBIan Romanick3-0/+342
NOTE: On builds of Mesa that include commit 739ac3d3, this test fails with: Testing glGetAttachedObjectsARB... Unexpected GL error: GL_OUT_OF_MEMORY 0x505 (Error at tests/spec/arb_shader_objects/getattachedobjects.c:156) Expected GL error: GL_INVALID_VALUE 0x501 Segmentation fault (core dumped) I verified that the test passes with my distro Mesa: OpenGL version string: 3.0 Mesa 11.1.0 (git-525f3c2) Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Cc: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
2019-06-26arb_direct_state_access: Also test texture parameters around ↵Ian Romanick1-0/+198
glGenerateTextureMipmap NOTE: This causes the test to fail on Mesa drivers that use meta (e.g., i965). Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93717
2019-06-26fbo-integer: Allow test to run on either GLSL 1.30 or GL_EXT_gpu_shader4Ian Romanick1-8/+9
Cc: Vinson Lee <vlee@freedesktop.org> Cc: Brian Paul <brianp@vmware.com>
2019-06-26fbo-integer: Don't require rendering to GL_RGB_INTEGER_EXTIan Romanick1-2/+7
The RGB base formats aren't required. Don't report failure if the driver reports GL_FRAMEBUFFER_UNSUPPORTED_EXT. Cc: Vinson Lee <vlee@freedesktop.org> Cc: Brian Paul <brianp@vmware.com>
2019-06-26fbo-integer: Remove glDrawPixels testIan Romanick1-87/+0
Original GL_EXT_texture_integer allowed glDrawPixels(..., GL_RGBA_INTEGER_EXT, GL_INT, ...), but OpenGL 3.0 does not. According to comments in Mesa's src/mesa/main/drawpix.c, NVIDIA generates the required OpenGL 3.0 error even on implmentations that advertise GL_EXT_texture_integer. It doesn't seem like this should pass on any implementaiton. Cc: Vinson Lee <vlee@freedesktop.org> Cc: Brian Paul <brianp@vmware.com>
2019-06-26arb_transform_feedback2: Misc. API error checksIan Romanick3-0/+748
This covers most of the errors mentioned in the spec that aren't already covered in cannot-bind-when-active.c and gen-names-only.c. Most of the other errors should be covered by existing EXT_transform_feedback tests. Mesa currently passes all of these tests. NVIDIA (304.64 on GTX 260) fails several subtests. Pause active feedback only: Incorrectly generates error in glEndTransformFeedback. TransformFeedbackVaryings only when inactive: Doesn't generate any errors when it should. Draw only from "ended" object: Generates GL_INVALID_VALUE instead of GL_INVALID_OPERATION. AMD has not been tested. v2: Convert to use piglit subtest reporting. Fix fs_only_prog to actually only have a fragment shader attached. v3: Quite significant re-write. Per Paul's suggestion, each of the tests is no a subtest that can be individually selected from the command line. This uses infrastructure in piglit-framework-gl to handle the subtests. This also eliminates the work-around for NVIDIA's glEndTransformFeedback-while-paused bug. The code currently in all.tests for getting the list of subtests is just a place holder. We'll want to refactor this out somewhere else before pushing. v4: Fix GLSL checks. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com> Cc: Dylan Baker <dylan@pnwbakers.com>
2019-06-26Don't check for OpenGL 1.1Ian Romanick4-4/+0
Seriously. The Linux OpenGL ABI has always required at least OpenGL 1.2. The Windows OpenGL ABI has always required at least OpenGL 1.1. I don't know what the Mac OS X ABI requires, but every supported device has had 1.1 at least as far back as 10.5.8. SGI O2 and Octane support at least OpenGL 1.1. I don't think piglit needs to fail gracefully on systems from before 1996.
2019-06-26Fix many incorrect GLSL checksIan Romanick43-40/+72
piglit_require_GLSL() only checks that shader objects and the shading language exist. It does *NOT* check that any particular shader stage exists. It is perfectly valid to have an OpenGL 1.5 implementation that has only vertex shaders or only fragment shaders. piglit_require_GLSL_version(110) and piglit_require_GLSL_version(120) have the same failings as piglit_require_GLSL() and more. GLSL 1.10 is the minimum GLSL version that actually exists, so anything that has GLSL has version 1.10. GLSL 1.20 can be exposed by an implementation, so checking that version doesn't guarantee anything either. GLSL 1.30 implicitly requires OpenGL 3.0, so that is a strong check. All of the non-shader drivers in Mesa advertise shader objects and GLSL 1.20 and NEITHER vertex nor fragment stages. This fixes these tests to just skip on those drivers.
2019-06-26Remove fake GL_EXT_gpu_shader4 supportIan Romanick2-15/+2
All of the shaders in these tests explicitly say, "#version 130".
2019-06-26glsl-1.30: Replace open-coded version of piglit_require_GLSL_version()Ian Romanick1-15/+1
2019-06-26Delete spurious version checksIan Romanick46-70/+1
Remove check for: 1. GLSL version checks that are redundant with GL version checks (e.g., checking for GLSL 1.30 when the test already required OpenGL 3.0). 2. GL version checks that are redundant with the requirements section. 3. One call to piglit_require_GLSL() when no GLSL related API calls are made.
2019-06-26genmipmap-errors: Check for GL_ARB_depth_texture before creating a depth textureIan Romanick1-2/+2
Neither GL_EXT_packed_depth_stencil nor GL_ARB_depth_buffer_float implies that the implementation can create depth textures. This affects the R200 driver in Mesa, for example.
2019-06-26intel_shader_atomic_float_minmax: Explicitly request std430 layoutIan Romanick10-10/+10
These tests preinit the ssbo contents but expect the driver to be doing std430 packing by default, just specify std430 packing. Edits done by: grep -lr ' buffer ' tests/spec/intel_shader_atomic_float_minmax/execution |\ while read f; do sed --in-place -e 's/layout(binding = 0) buffer/layout(binding = 0, std430) buffer/' $f done
2019-06-25shader_runner: some extra skip reasong loggingAlejandro Piñeiro1-3/+10
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
2019-06-24glsl-es-1.00: Only unroll a do-while-false loop onceIan Romanick1-0/+47
This was the smallest test case that I could make that would reproduce the bug, generate actual instructions (instead of constant folding to nothing), and be small enough to run on any GPU supported by Mesa that has GLES 2.0 (i.e., i915). Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110953 Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira@intel.com> Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
2019-06-22tests/fbo: init texture contentPierre-Eric Pelloux-Prayer1-3/+13
The "variant 1" subtest uses glBlendFunc(GL_DST_ALPHA, ...) so we must initialize the dst texture or we might get inconsistent results. This fixes random failures on radeonsi hardware. Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
2019-06-21tests/arb_direct_state_access: Memory leakYevhenii Kolesnikov1-0/+2
Free memory allocated for random image. Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com> Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
2019-06-20util: Optionally validate SPIR-V after assemblingCaio Marcelo de Oliveira Filho1-0/+24
PIGLIT_SPIRV_VALIDATE enables the feature, PIGLIT_SPIRV_VAL_BINARY let the user specify which "spirv-val" to use. For now, validate against opengl4.5 target environment. Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
2019-06-20util: Assemble SPIR-V with the correct target environmentCaio Marcelo de Oliveira Filho1-0/+1
In the future we should consider decide this based on the shader_test requirements. For now allow us to generate SPIR-V binary in the expected version -- useful for adding validation later. Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
2019-06-18tests: Free memory returned by glXChooseFBConfigYevhenii Kolesnikov6-0/+17
Memory, allocated in glXGetFBConfigs, has to be freed manually. Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com> Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
2019-06-10test/textureGather: memory leakSergii Romantsov1-1/+3
Test uses asprintf which allocates memory, but not freed it Fixes: f62e74f61b39(ARB_texture_gather: add initial execution test) Signed-off-by: Sergii Romantsov <sergii.romantsov@globallogic.com> Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
2019-06-10tests/texelFetch: Memory leakYevhenii Kolesnikov1-0/+1
Free memory, allocated by asprintf. Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com> Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
2019-06-10arb_copy_buffer: NULL check for glMapBuffer outcomeAlejandro Piñeiro1-1/+1
test_copy checks if the outcome of glMapBuffer is the same that some expected data. But we found that for drivers failing on returnign the proper data, it is possible/likely that glMapBuffer would just return NULL. So it would be better to check for NULL to avoid a crash, as technically the driver is not crashing, but the test (even if the reason glMapBuffer still needs to be investigated). Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
2019-06-10util/piglit-util-gl: Memory leaks in a number of testsYevhenii Kolesnikov1-0/+2
Memory has been allocated but never freed, which causes memory leaks in a number of tests: arb_point_sprite-mipmap glsl-fs-bug25902 glsl-fs-sampler-numbering tex-border-1 gl-2.0-link-empty-prog arb_point_sprite-checkerboard Fixes: 3def81d05a (Unify piglit_checkerboard_texture between GL and GLES) Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com> Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
2019-06-07tests/fp-indirections2: Memory isn't freedYevhenii Kolesnikov1-0/+2
Memory isn't properly freed, which causes significant memory leaks. Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com> Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
2019-06-05tests/arb_copy_image: Memory leakYevhenii Kolesnikov2-0/+4
Free memory, allocated for fs_src and img_data. Fixes memory leaks in arb_copy_image-formats and arb_copy_image-targets tests. Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com> Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
2019-06-04arb_texture_buffer_range: Fix buffer alignment in ranges-2 testAnthony Pesch1-6/+24
The ranges-2 test was failing due to glTexBufferRange returning GL_INVALID_VALUE when the offset parameter wasn't GL_TEXTURE_BUFFER_OFFSET_ALIGNMENT byte aligned. From the OpenGL 4.6 Core spec: An INVALID_VALUE error is generated if offset is not an integer multiple of the value of TEXTURE_BUFFER_OFFSET_ALIGNMENT. Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
2019-06-04tests/vertex-program-two-side: Memory leakYevhenii Kolesnikov1-0/+2
Memory is being allocated by asprintf, but never freed. Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com> Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
2019-06-03draw-prim-rate: Fix Clang build error.Vinson Lee1-1/+1
draw-prim-rate.c:415:3: error: initializer element is not a compile-time constant ceil(sqrt(0.5 * 1000)), ^~~~~~~~~~~~~~~~~~~~~~ Fixes: 6e1bb1e769d2 ("Add a test that measures primitive rate") Signed-off-by: Vinson Lee <vlee@freedesktop.org> Reviewed-by: Marek Olšák <marek.olsak@amd.com>
2019-06-03Add a test that measures primitive rateMarek Olšák5-5/+524
The output looks like this (from the initial version, the current version is slightly different): Measuring GPrims/second, , Number of primitives Draw Call , Cull Method , 2K, 4K, 8K, 16K, 32K, 64K, 256K --------------,----------------------,------,------,------,------,------,------,------ glDrawElements, none , 2.80, 2.69, 2.91, 2.89, 2.91, 2.92, 2.42 glDrawElements, rasterizer discard , 4.80, 4.77, 4.95, 4.84, 4.91, 4.85, 4.93 glDrawElements, 100% back faces , 3.27, 3.19, 3.29, 3.21, 3.26, 3.32, 3.33 glDrawElements, 75% back faces , 3.27, 3.47, 3.29, 3.54, 3.53, 3.60, 3.49 glDrawElements, 50% back faces , 3.92, 3.83, 3.34, 3.58, 3.68, 3.63, 2.76 glDrawElements, 25% back faces , 3.66, 3.52, 3.12, 3.18, 3.00, 2.78, 3.45 glDrawElements, 100% culled by view , 4.85, 4.75, 4.94, 4.68, 4.91, 4.80, 4.94 glDrawElements, 75% culled by view , 4.82, 4.68, 4.77, 4.76, 4.80, 4.65, 3.20 glDrawElements, 50% culled by view , 4.73, 4.65, 4.46, 3.40, 4.86, 4.04, 2.99 glDrawElements, 25% culled by view , 3.67, 3.48, 3.26, 2.70, 2.76, 2.60, 2.46 glDrawElements, 100% degenerate prims, 1.67, 1.66, 1.68, 1.66, 1.68, 1.68, 1.68 glDrawElements, 75% degenerate prims, 1.65, 1.90, 1.67, 1.96, 1.86, 2.01, 1.83 glDrawElements, 50% degenerate prims, 2.43, 2.37, 1.66, 2.44, 1.90, 2.24, 1.98 glDrawElements, 25% degenerate prims, 2.49, 2.94, 1.67, 2.03, 2.76, 2.79, 2.15 glDrawElements, 98 small prims/pixel , 4.82, 4.65, 4.85, 4.80, 4.90, 4.77, 4.30 glDrawElements, 32 small prims/pixel , 4.86, 4.71, 4.80, 4.69, 4.81, 4.16, 4.95 glDrawElements, 8 small prims/pixel , 4.73, 4.67, 4.92, 4.85, 4.91, 4.86, 3.20 glDrawArrays , none , 1.67, 1.66, 1.65, 1.47, 1.52, 1.24, 1.60 etc.
2019-06-03opengl.py: fix incorrect ext_transform_instanced extension nameMarek Olšák1-1/+1
trivial
2019-06-03rendermode-feedback: fix a crashMarek Olšák1-2/+1
For some reason, the last element is uninitialized if I don't clear the structure like this.
2019-06-03test: fix resource leaking for depthstencil-render-miplevelsSergii Romantsov1-7/+45
Usage test 'depthstencil-render-miplevels 200 s=z24_s8' causes memory leaks. Fixed: memory allocation/deallocation is controlled more carefully. CC: Eric Anholt <eric@anholt.net> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108819 Issue: https://gitlab.freedesktop.org/mesa/piglit/issues/9 Fixes: 7a0e61d7792f (depthstencil-render-miplevels: Present the results in non-auto mode.) Signed-off-by: Sergii Romantsov <sergii.romantsov@globallogic.com> Reviewed-by: Marek Olšák <marek.olsak@amd.com>
2019-06-03test: fix core dump for depthstencil-render-miplevelsSergii Romantsov1-2/+2
Usage test 'depthstencil-render-miplevels 200 s=z24_s8' causes core dump on exit. Fixed: array of pointers is allocated to size max_miplevel, but used as max_miplevel + 1. CC: Eric Anholt <eric@anholt.net> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108819 Issue: https://gitlab.freedesktop.org/mesa/piglit/issues/9 Fixes: 7a0e61d7792f (depthstencil-render-miplevels: Present the results in non-auto mode.) Signed-off-by: Sergii Romantsov <sergii.romantsov@globallogic.com> Reviewed-by: Marek Olšák <marek.olsak@amd.com>
2019-06-03ext_dsa: fix invalid empty shader in testPierre-Eric Pelloux-Prayer1-1/+1
2019-06-03ext_direct_state_access: test glTextureImage* with GL_PROXY_TEXTURE_* targetsPierre-Eric Pelloux-Prayer2-0/+140
2019-06-03ext_direct_state_access: verify possible failures when using GL_COMPILEPierre-Eric Pelloux-Prayer4-27/+138
2019-06-03util/dma_buf: allow custom rendernodeSergii Romantsov1-10/+29
Piglit tests based on dma-buf may not finish properly due to wrongly selected rendernode. Patch adds possibility to override device by variable WAFFLE_GBM_DEVICE (in format /dev/dri/renderD128). -v2: added WAFFLE_GBM_DEVICE (Marek Olšák) CC: Chad Versace <chad.versace@intel.com> Fixes: 8147ec81a71d (util/dma_buf: Use rendernode if available) Fixes: https://gitlab.freedesktop.org/mesa/piglit/issues/15 Signed-off-by: Sergii Romantsov <sergii.romantsov@globallogic.com> Reviewed-by: Marek Olšák <marek.olsak@amd.com>
2019-06-03tests/overlapping-locations-input-attribs: Memory leakYevhenii Kolesnikov1-0/+2
Memory is being allocated by asprintf, but it is never freed. Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com> Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
2019-05-31tests/line-aa-width: Memory leakYevhenii Kolesnikov1-0/+2
Big chunk of memory is allocated, but not freed. Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com> Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
2019-05-28fbo-generatemipmap-formats: Modify NPOT tex_widthNanley Chery1-1/+4
The i965 driver's surface layout component (ISL) can calculate the location of miplevels incorrectly when the surface is compressed. Set the width to 257px to demonstrate this issue. Reviewed-by: Marek Olšák <marek.olsak@amd.com>
2019-05-24glx-multithread-buffer: Add a new test reproducing a radeonsi bugMarek Olšák3-0/+201
2019-05-24piglit_drm_dma_buf: fix NV12, YUV420, and YVU420 import tests for radeonsiMarek Olšák1-10/+11
- We have to use the linear layout, otherwise the offsets can't be modified. - The GBM image has only 1 plane, so the gbm functions can't be used for other planes. - Don't manipulate with the stride. It has to be the same.
2019-05-24Revert "cmake: make gbm a required build dependency"Marek Olšák2-3/+3
This reverts commit c878a27d0032c6c97f13f84f2a223fafe41e1e63. pushed by accident
2019-05-24shader_runner: don't abort on atomic counter subtest failureMarek Olšák1-2/+2
Reviewed-by: Dave Airlie <airlied@redhat.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-10-16 08:48:59 +0000