- 20 May, 2022 1 commit
-
-
Fix defects reported by Coverity Scan. Resource leak (RESOURCE_LEAK) leaked_storage: Variable attachments going out of scope leaks the storage it points to. Fixes: deec4392 ("ext_direct_state_access: add test for Framebuffer functions") Signed-off-by:
Vinson Lee <vlee@freedesktop.org> Reviewed-by:
Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> Part-of: <mesa/piglit!679>
-
- 19 May, 2022 1 commit
-
-
Pierre-Eric Pelloux-Prayer authored
GL_MAX_TEXTURE_BUFFER_SIZE_ARB + 1 won't necessarly fit in an int32_t so use a int64_t in the computation. Reviewed-By:
Mike Blumenkrantz <michael.blumenkrantz@gmail.com>
Part-of: <mesa/piglit!675>
-
- 18 May, 2022 2 commits
-
-
Vinson Lee authored
Fix defects reported by Coverity Scan. Resource leak (RESOURCE_LEAK) leaked_storage: Variable dst_data going out of scope leaks the storage it points to. leaked_storage: Variable src_data going out of scope leaks the storage it points to. leaked_storage: Variable exp_data going out of scope leaks the storage it points to. Fixes: c02fca9f ("Ported logicop test from Glean to Piglit.") Fixes: c773ae68 ("test/logicop: expand tests to also cover MSAA render target") Signed-off-by:
Iago Toral Quiroga <itoral@igalia.com> Part-of: <!666>
-
Jose Fonseca authored
Fixes #69 Reviewed-by:
Roland Scheidegger <sroland@vmware.com> Part-of: <!678>
-
- 13 May, 2022 1 commit
-
-
Vinson Lee authored
prop_count is of type uint32_t. Fix defect reported by Coverity Scan. Unsigned compared against 0 (NO_EFFECT) unsigned_compare: This less-than-zero comparison of an unsigned value is never true. prop_count < 0U. Fixes: a0722042 ("ext_external_objects: Selecting a queue that supports graphics") Signed-off-by:
David Heidelberg <david.heidelberg@collabora.com>
Part-of: <!663>
-
- 12 May, 2022 2 commits
-
-
Waffle 1.6.0 was released over 2 years ago. Even Debian stable has version 1.6.3. As result remove all the ifdef version checks. Signed-off-by:
Emil Velikov <emil.velikov@collabora.com> Reviewed-by:
Corentin Noël <corentin.noel@collabora.com> Part-of: <mesa/piglit!592>
-
It is an EGL surfaceless platform, so might as well guard it properly. Otherwise we would end building it on Windows where (barring ANGLE) it is not a thing. Signed-off-by:
-
- 11 May, 2022 2 commits
-
-
David Heidelberg authoredVestigial reference which accidentally got in. Fixes: mesa/piglit!529 Reviewed-by:
Tomeu Vizoso <tomeu.vizoso@collabora.com> Signed-off-by:
David Heidelberg <david.heidelberg@collabora.com>
Part-of: <mesa/piglit!672> -
Vinson Lee authored
Fix defect reported by Coverity Scan. Dereference before null check (REVERSE_INULL) check_after_deref: Null-checking attachments suggests that it may be null, but it has already been dereferenced on all paths leading to the check. Fixes: 8713fbda ("ext_external_objects: Vulkan framework additions") Signed-off-by:
Tapani Pälli <tapani.palli@intel.com>
Part-of: <mesa/piglit!660>
-
- 10 May, 2022 1 commit
-
-
The test will fail if the images aren't exactly the same, so have the image comparison match that so we can see what pixels are different. Signed-off-by:
David Heidelberg <david.heidelberg@collabora.com>
Part-of: <mesa/piglit!669>
-
- 09 May, 2022 5 commits
-
-
Mark Janes authored
Since 1fc96325, read_file is called with both a pathname and an open file handle. It is clear that it is meant to be called with a pathname. Fixes: ef87faef ("Make it work with the latest pytest version") Closes: mesa/piglit#70 Reviewed-by:
-
David Heidelberg authored
Piglit crashes, when is run without device comparing single trace. It shouldn't be needed. f.e. ``` PIGLIT_SOURCE_DIR="../" ./replayer.py compare trace /home/projects/collabora/traces-db/0ad/0ad-v2.trace 0 ``` should be valid, but does crash with: ``` ... File "/home/projects/collabora/piglit/replayer/../framework/replay/compare_replay.py", line 81, in _check_trace dir_in_results = path.join('trace', OPTIONS.device_name, trace_dir) File "/usr/lib/python3.10/posixpath.py", line 90, in join genericpath._check_arg_types('join', a, *p) File "/usr/lib/python3.10/genericpath.py", line 152, in _check_arg_types raise TypeError(f'{funcname}() argument must be str, bytes, or ' TypeError: join() argument must be str, bytes, or os.PathLike object, not 'NoneType' ``` Reviewed-by:David Heidelberg <david.heidelberg@collabora.com>
Part-of: <mesa/piglit!668> -
Fix all deprecation warnings. Remove testing for python 3.6. Signed-off-by:
Emma Anholt <emma@anholt.net> Part-of: <mesa/piglit!626>
-
They are already defined somewhere else in this case. Signed-off-by:
-
Signed-off-by:
-
- 06 May, 2022 2 commits
-
-
Vinson Lee authored
Fix defect reported by Coverity Scan. Resource leak (RESOURCE_LEAK) leaked_storage: Variable texel_fetch_macro going out of scope leaks the storage it points to. Fixes: 8e842367 ("ext_framebuffer_multisample_blit_scaled: Rewrite the fragment shader") Signed-off-by:
Anuj Phogat <anuj.phogat@gmail.com> Part-of: <mesa/piglit!657>
-
See mesa/mesa!10765 v2: Fix out-of-source builds. Remove unused variable. Define NOGDI to avoid Polygon symbol clash. Fix build when Git is not found. -- Jose Fonseca <jfonseca@vmware.com> Reviewed-by:
-
- 05 May, 2022 2 commits
-
-
Verify behaviour when file has not matching content length, when content length missing and download success with later retry. Acked-by:
Sergi Blanch Torne <sergi.blanch.torne@collabora.com>
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Signed-off-by: Guilherme Gallo <guilherme.gallo@collabora.com> Part-of: <mesa/piglit!653>
-
David Heidelberg authored
- support retries (use HTTPAdapter + Retry) - drop checking status_code which is handled by raise_for_status() - enforce content length checking - increase chunk size - misc: add spaces (linting) Acked-by:
Sergi Blanch Torne <sergi.blanch.torne@collabora.com>
Signed-off-by: David Heidelberg <david.heidelberg@collabora.com>
Part-of: <mesa/piglit!653>
-
- 04 May, 2022 4 commits
-
-
Vinson Lee authored
Fix defect reported by Coverity Scan. Resource leak (RESOURCE_LEAK) leaked_storage: Variable offset_coords going out of scope leaks the storage it points to. Fixes: 95241b0c ("textureGather: make sure to test out min/max offsets") Signed-off-by:
Ilia Mirkin <imirkin@alum.mit.edu> Reviewed-by:
Timothy Arceri <tarceri@itsqueeze.com> Part-of: <mesa/piglit!655>
-
Noticed in updated deqp-runner which reports Missing instead of just Fail (seems reasonable for a missing binary). Reviewed-by:
Marek Olšák <marek.olsak@amd.com> Part-of: <mesa/piglit!650>
-
"'GL_PROGRAM_INPUT(vs,fs) active resources' expected 2 but got 6" is not very useful without also seeing the list of resources you got. Reviewed-by:
-
From the ARB_shader_objects spec: The error INVALID_VALUE is generated by any command that takes one or more handles as input, and one or more of these handles are not an object handle generated by OpenGL. However, Mesa's implementation of glDeleteObjectARB does not emit that error at the time of writing. Let's add a test to make sure we don't regress this in the future. Reviewed-by:
-
- 03 May, 2022 1 commit
-
-
Reviewed-by:
-
- 30 Apr, 2022 4 commits
-
-
Vinson Lee authored
Fix defect reported by Coverity Scan. Resource leak (RESOURCE_LEAK) leaked_storage: Variable name going out of scope leaks the storage it points to. Fixes: 98f9fb5f ("texture integer getTexImage clamping: Remove GLenum string literals.") Signed-off-by:
-
This fixes the logic that prevents windows to take focus that was specifically added for glx tests. Reviewed-by:
-
piglit_get_glx_window calls it. Reviewed-by:
-
Reviewed-by:
-
- 27 Apr, 2022 1 commit
-
-
Vinson Lee authored
Fix defects reported by Coverity Scan. Resource leak (RESOURCE_LEAK) leaked_storage: Variable name going out of scope leaks the storage it points to. Fixes: 9be78360 ("use a glsl1.30 version of texture integer") Fixes: 08b9bf31 ("texture integer glsl130: Remove GLenum string literals.") Signed-off-by:
Dave Airlie <airlied@redhat.com> Part-of: <mesa/piglit!654>
-
- 21 Apr, 2022 1 commit
-
-
In order to check the values in gl_SampleMaskIn, we need to make sure multisampling rasterization is turned on. Signed-off-by:
Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: ed153b51 ("Add GL_INTEL_conservative_rasterization tests")
Part-of: <mesa/piglit!649>
-
- 12 Apr, 2022 2 commits
-
-
Tests that texture buffer's size cannot exceed GL_MAX_TEXTURE_BUFFER_SIZE_ARB even if underlying buffer allows it to be bigger. In other words the size of texture buffer must be clamped to the GL_MAX_TEXTURE_BUFFER_SIZE_ARB by the driver. Closes: mesa/mesa#1723 Signed-off-by:
Dmytro Nester <dmytro.nester@globallogic.com> Signed-off-by:
Sergii Melikhov <sergii.v.melikhov@globallogic.com> Signed-off-by:
Viktoriia Palianytsia <v.palianytsia@globallogic.com> Reviewed-by:
Nanley Chery <nanley.g.chery@intel.com>
Part-of: <mesa/piglit!231> -
Tests that glClearDepth store on gpu exactly the transferred values with different depth's values and with different depth formats. Signed-off-by:
Illia Iorin <illia.iorin@globallogic.com> Signed-off-by:
Illia Abernikhin <illia.abernikhin@globallogic.com> Reviewed-by:
Nanley Chery <nanley.g.chery@intel.com>
Part-of: <mesa/piglit!98>
-
- 07 Apr, 2022 1 commit
-
-
From Python docs, proc.terminate() is the right way to do it. Reviewed-by:
-
- 29 Mar, 2022 1 commit
-
-
Mike Blumenkrantz authoredthis is supposed to clear each buffer to the expected alpha fixes the test for zink Reviewed-by:
Adam Jackson <ajax@redhat.com> Part-of: <mesa/piglit!647>
-
- 17 Mar, 2022 6 commits
-
-
Reviewed-by:
Ian Romanick <ian.d.romanick@intel.com> Part-of: <mesa/piglit!609>
-
From section 5.7 "Structure and Array Operations" of the GLSL 1.30 spec: "Behavior is undefined if a shader subscripts an array with an index less than 0 or greater than or equal to the size the array was declared with." The behaviour becomes defined only in robustness extensions, however even if driver is technically allowed to crash or hang, it most likely doesn't want to. Robustness extensions encourage to us to return zero on out-of-bounds read, however it's not clearly applicable to the array of opaque types. What will happen when we call a built-in function which expects the opaque type with the instance returned from oob access? In the spirit of the spec - I think neither compilation should fail nor a hang occur. Signed-off-by:
Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Acked-by: -
From section 5.7 "Structure and Array Operations" of the GLSL 1.30 spec: "Behavior is undefined if a shader subscripts an array with an index less than 0 or greater than or equal to the size the array was declared with." The behaviour becomes defined only in robustness extensions, however even if driver is technically allowed to crash or hang, it most likely doesn't want to. Large local array may be spilled, so out-of-bounds access should be tested for them separately. Signed-off-by:
Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Acked-by: -
From section 5.7 "Structure and Array Operations" of the GLSL 1.30 spec: "Behavior is undefined if a shader subscripts an array with an index less than 0 or greater than or equal to the size the array was declared with." The behaviour becomes defined only in robustness extensions, however even if driver is technically allowed to crash or hang, it most likely doesn't want to. gl_SampleMaskIn may be handled differently in a driver than ordinary arrays. Signed-off-by:Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Acked-by: -
From section 5.7 "Structure and Array Operations" of the GLSL 1.30 spec: "Behavior is undefined if a shader subscripts an array with an index less than 0 or greater than or equal to the size the array was declared with." The behaviour becomes defined only in robustness extensions, however even if driver is technically allowed to crash or hang, it most likely doesn't want to. gl_TessLevel* may be handled differently in a driver than ordinary arrays. Signed-off-by:Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Acked-by: -
From section 5.7 "Structure and Array Operations" of the GLSL 1.30 spec: "Behavior is undefined if a shader subscripts an array with an index less than 0 or greater than or equal to the size the array was declared with." The behaviour becomes defined only in robustness extensions, however even if driver is technically allowed to crash or hang, it most likely doesn't want to. gl_ClipDistance may be handled differently in a driver than ordinary arrays. Signed-off-by:
Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Acked-by:
-
