Alexander Galazin [Mon, 23 Jan 2017 00:51:34 +0000 (01:51 +0100)]
Merge vulkancts/opengl-cts-es-3.2.2 into vulkancts/opengl-cts-dev
Change-Id: Idbcb8437a3598c158d3e4f7d57000b6b668dcf05
Alexander Galazin [Fri, 20 Jan 2017 09:00:27 +0000 (10:00 +0100)]
Improve submission verification scripts
Do not report warnings if git logs match exactly the release tag commit
Components: Framework
Change-Id: I9833026c5924bd22adfce8d26de2314a52fd88b9
(cherry picked from commit
244a26241b3887ccdca2eccbd45b914d3b5e6883)
Alexander Galazin [Fri, 20 Jan 2017 09:00:27 +0000 (10:00 +0100)]
Improve submission verification scripts
Do not report warnings if git logs match exactly the release tag commit
Components: Framework
Change-Id: I9833026c5924bd22adfce8d26de2314a52fd88b9
Graeme Leese [Fri, 20 Jan 2017 13:44:51 +0000 (13:44 +0000)]
Map buffers for writing in atomic counter tests
These tests attempt to sort the values in the returned buffers in place,
so the buffers must be mapped for writing. See VK-GL-CTS issue: 39.
Components: OpenGL
Affects: KHR-GLES31.core.shader_atomic_counters.*
Change-Id: Icda40fe8d27c86ecd9912767c34ec81e7fa3a35a
Alexander Galazin [Fri, 20 Jan 2017 10:21:49 +0000 (11:21 +0100)]
Merge vulkancts/opengl-cts-es-3.2.2 into vulkancts/opengl-cts-dev
Change-Id: Ia246cacb67f8e9f01f84533337efff66e40be757
Alexander Galazin [Thu, 19 Jan 2017 20:00:10 +0000 (21:00 +0100)]
Make RGB16F/RGB32F not color-renderable in ES 3.2
GL_RGB16F is color-renderable only if
GL_EXT_color_buffer_half_float is supported.
GL_RGB32F is color-renderable only if
GL_EXT_color_buffer_float is supported.
Affects: dEQP-GLES31.functional.fbo.color.texcubearray.rgb16f
VK-GL-CTS issue: 40
Components: AOSP
Change-Id: I2cf3eb7e8d5c361ac23a64128b0d27958dc48438
Pyry Haulos [Thu, 19 Jan 2017 19:43:08 +0000 (11:43 -0800)]
Merge vulkancts/vulkan-cts-1.0-dev into vulkancts/opengl-cts-dev
Change-Id: I54cbd7d0021d911bad32411f54437fb270a60b64
Pyry Haulos [Thu, 19 Jan 2017 19:41:57 +0000 (11:41 -0800)]
Merge vulkancts/vulkan-cts-1.0.2 into vulkancts/vulkan-cts-1.0-dev
Change-Id: I42976523a4e4b0c9b66fadc7b28f4acaa3f93c9e
Pyry Haulos [Thu, 19 Jan 2017 17:10:20 +0000 (09:10 -0800)]
Remove PointCoord test from mustpass
Test doesn't take into account difference between real (unrounded) point
origin and (rounded) texel center when computing PointCoord reference
values.
Vulkan CTS issue: 582
Change-Id: I0ffaf40c1d99247b68d6ee16f2f94df74c832884
Pyry Haulos [Thu, 19 Jan 2017 17:08:02 +0000 (09:08 -0800)]
Remove broken pipeline barrier tests from mustpass
These tests attempt to store a value that doesn't fit in 16-bit uint
into a RelaxedPrecision-qualified uint push constant.
Vulkan CTS issue: 583
Change-Id: I9ae6da8469b8213c2f1d0e5d90b1ada8640bef6e
Alexander Galazin [Thu, 19 Jan 2017 09:42:20 +0000 (10:42 +0100)]
Update README.md in response to feedback
- Add more details about GLCTS_GTF_TARGET
- Clarify that KC CTS is necessary only for conformance submissions
- Clarify that GTF builds and conformance runs must use matching
API versions
Components: OpenGL
Change-Id: I76a015e66918cf074823e6ab139c140369f7a073
Courtney Goeltzenleuchter [Thu, 12 Jan 2017 01:17:08 +0000 (18:17 -0700)]
Add NV extensions to known instance extension list
Change-Id: Idf87f8718fcfdd55764d81dcbf478685225b3635
Maciej Jesionowski [Wed, 7 Dec 2016 15:14:55 +0000 (16:14 +0100)]
Test OpVectorShuffle with 0xFFFFFFFF and OpUndef
Modified tests:
- dEQP-VK.spirv_assembly.instruction.compute.opspecconstantop.vector_related
- dEQP-VK.spirv_assembly.instruction.graphics.opspecconstantop.vector_related_*
Khronos issue: #498
Change-Id: I6f370c5300cdcb3f9afa3c20e22ab5f2ccfaeffa
Akos Dirner [Tue, 28 Jun 2016 13:15:08 +0000 (15:15 +0200)]
Add new texture function tests with 1d, 1d array and cube array samplers
New tests:
- dEQP-VK.glsl.texture_functions.*sampler1d*
- dEQP-VK.glsl.texture_functions.*sampler1darray*
- dEQP-VK.glsl.texture_functions.*samplercubearray*
Affected tests:
- dEQP-VK.glsl.texture_functions.*
Fixes #413
Change-Id: I9fb5096ea0e77e2973df190b0d6eff91e17dcd0e
Kantoch [Thu, 13 Oct 2016 12:29:53 +0000 (14:29 +0200)]
Multisample Shader BuiltIn Tests
Tests for gl_SampleID, gl_SamplePosition, gl_SampleMaskIn[], gl_SampleMask[]
New tests:
- dEQP-VK.pipeline.multisample_shader_builtin.*
Issue #450
Change-Id: I82875fc20d01232127f01186644a04d67ddb8833
Slawomir Cygan [Mon, 28 Nov 2016 18:28:01 +0000 (19:28 +0100)]
Add tests for resolving between images of different sizes
New tests: dEQP-VK.api.copy_and_blit.resolve_image.diff_image_size*
This change adds new tests, which try to resolve data between images of
different sizes. As some implementations may optimize resolves of whole
images, this tests tricks implementation by having resolve region equal
one of dst/src image size, while the other image is larger.
`FILL_MODE_MULTISAMPLE` logic was tweaked to work properly on non-square src images
- this is required by new tests.
Change-Id: If15f500e0e4b2b574f1441da71e5d57e690f8f43
Pyry Haulos [Wed, 23 Nov 2016 19:02:57 +0000 (11:02 -0800)]
Relax LOD computation bounds; add mipmap.2d.projected to mustpass
This change updates LOD approximation rules to match proposed spec
language in Vulkan texture filtering and mipmap tests.
dEQP-VK.texture.mipmap.2d.projected.* is added back to the mustpass.
Affects: dEQP-VK.texture.*
Change-Id: Ic786bc733913a76d27f0cba31f3a61a10f8144fc
Gabor Abraham [Fri, 18 Nov 2016 10:22:53 +0000 (11:22 +0100)]
Improve the test generation in the copies and blitting tests
The generateBuffer function didn't handle the integer formats well.
Affected tests:
- dEQP-VK.api.copy_and_blit.image_to_image.all_formats.*int*
- dEQP-VK.api.copy_and_blit.blit_image.all_formats.*int*
Change-Id: I248136606d2535c176432bbc50de45e1e4e55b80
Maciej Jesionowski [Tue, 27 Dec 2016 13:15:58 +0000 (14:15 +0100)]
Add NEAREST filtered blitting tests
This fixes a couple of issues with nearest filtered blitting tests
and adds them back to the mustpass.
- Corrected test case names (esp. for D/S formats)
- Modified common code used by linear cases too (shouldn't have
an observable effect on tests)
- Used tcuTexLookupVerifier to sample source image with an offset
that accounts for rounding errors in coordinate calculation
- Enabled nearest cases in mustpass
New tests:
- dEQP-VK.api.copy_and_blit.blit_image.all_formats.*_nearest
Modified tests:
- dEQP-VK.api.copy_and_blit.blit_image.all_formats.*
Fixes #483, #565
Change-Id: I176c2b50a656268a31f1f35fef1ff90647039727
Eduardo Lima Mitev [Mon, 12 Dec 2016 18:46:57 +0000 (19:46 +0100)]
opengl/khr_debug: Push/pop the custom render context
KHR_Debug tests create their own render context. However, the
framework's test iterator handles basic operations on the render context
between test executions (e.g, swapping buffers), but it always assumes
that the default render context is the one being used. This causes
conflicts on KHR_debug tests, where the postIterate() method throws a
ResourceError, aborting the test execution, and breaking the logging
system.
This patch changes all KHR_debug tests to first save the original render
context, then sets the created render context as that of the base
context to give the framework a good, active render context; and finally
sets back the riginal render context when the test finishes.
If the test aborts in the middle, the destructor of the test (in TestBase)
will still try to restore the original context.
To allow the above, a setRenderContext() method was added to the test
context. So we need to have a mechanism to temporarily set the render
context of the default test context during the execution of these tests.
Fixes all tests in:
* GLXX-CTS.khr_debug.*
Fixes #28
Change-Id: If2c3eea5cb0882c11d2baf8334ed9ee6208b6c65
Adam Czupryna [Fri, 16 Dec 2016 14:58:18 +0000 (15:58 +0100)]
Extended CTS_ARB_sparse_texture2 MS34
* Added ShaderExtensionTestCase,
* Added SparseTexture2LookupTestCase
Change-Id: I444205daec25d875ac0eea3662ad28482e12695e
Alexander Galazin [Wed, 18 Jan 2017 07:28:57 +0000 (08:28 +0100)]
Merge vulkancts/opengl-cts-es-3.2.2 into vulkancts/opengl-cts-dev
Change-Id: I014d453d1fbd0f028783765bdc78ff548f5d12ab
Alexander Galazin [Mon, 16 Jan 2017 08:15:59 +0000 (09:15 +0100)]
Update zlib dependency to zlib 1.2.11
Fixes Vulkan CTS issue: #579
Change-Id: Ib0cd05616cbb4759d63bcf15ba8e8f6f374e291d
(cherry picked from commit
1be44d17818bd119885ad24bd9665ab3db933834)
Pyry Haulos [Tue, 17 Jan 2017 22:25:03 +0000 (14:25 -0800)]
Merge vulkancts/vulkan-cts-1.0-dev into vulkancts/opengl-cts-dev
Change-Id: Ica9b259ebbd8281891f3fd5c8154a96a2ad4bffe
Pyry Haulos [Tue, 17 Jan 2017 22:25:01 +0000 (14:25 -0800)]
Merge vulkancts/vulkan-cts-1.0.2 into vulkancts/vulkan-cts-1.0-dev
Change-Id: I42ef5e23579c3c51667f102ac3379daf02537153
Pyry Haulos [Tue, 17 Jan 2017 22:24:59 +0000 (14:24 -0800)]
Merge vulkancts/vulkan-cts-1.0.1 into vulkancts/vulkan-cts-1.0.2
Change-Id: I0ddead63d83e79809d35df7f3108ade58a13d0e6
Pyry Haulos [Tue, 17 Jan 2017 22:24:57 +0000 (14:24 -0800)]
Merge vulkancts/vulkan-cts-1.0.0 into vulkancts/vulkan-cts-1.0.1
Change-Id: I4742b81f85d9b09efd77b954d432e28fefabf818
Alexander Galazin [Mon, 16 Jan 2017 08:15:59 +0000 (09:15 +0100)]
Update zlib dependency to zlib 1.2.11
Fixes Vulkan CTS issue: #579
Change-Id: Ib0cd05616cbb4759d63bcf15ba8e8f6f374e291d
Alexander Galazin [Fri, 13 Jan 2017 21:41:32 +0000 (22:41 +0100)]
Update GL CTS README to point to GitHub VK GL CTS wiki
Change-Id: Ia06ec8801cb840e25dbe6012de34ce38ffd79467
(cherry picked from commit
41833f93fa1a551affefa71a16fc73e7afdf2ef8)
Alexander Galazin [Fri, 13 Jan 2017 21:41:32 +0000 (22:41 +0100)]
Update GL CTS README to point to GitHub VK GL CTS wiki
Change-Id: Ia06ec8801cb840e25dbe6012de34ce38ffd79467
Pyry Haulos [Fri, 13 Jan 2017 20:20:47 +0000 (12:20 -0800)]
Merge vulkancts/vulkan-cts-1.0-dev into vulkancts/opengl-cts-dev
Change-Id: I3b3e5e6367a0b8aa998ddebd0ad238870c778e6b
Pyry Haulos [Fri, 13 Jan 2017 20:20:44 +0000 (12:20 -0800)]
Merge vulkancts/vulkan-cts-1.0.2 into vulkancts/vulkan-cts-1.0-dev
Change-Id: I23e7239a4af6fbb1925c30e8a9b73c7e942d3ffc
Maciej Jesionowski [Thu, 12 Jan 2017 12:42:45 +0000 (13:42 +0100)]
Bugfix: Allow UBO sparse allocation larger than maxUniformBufferRange
Modified tests:
- dEQP-VK.sparse_resources.buffer.ubo.*
Fixes #573
Change-Id: I33d2098d5b6b783fa8293c7cb1af7e782b51df77
Alexander Galazin [Fri, 13 Jan 2017 17:21:09 +0000 (18:21 +0100)]
Merge vulkancts/opengl-cts-es-3.2.2 into vulkancts/opengl-cts-dev
Change-Id: I35a77f7ae74c294d760df6c7a897c0cd7d5f4643
Filip Demczewski [Thu, 24 Nov 2016 16:51:50 +0000 (17:51 +0100)]
Added CTS_ARB_texture_filter_minmax tests
Change-Id: I30d0ee7ef7e6796a2de85c4affd911d552dd106e
Adam Czupryna [Wed, 30 Nov 2016 15:54:57 +0000 (16:54 +0100)]
Extended CTS_ARB_sparse_texture2 MS33
* Added SparseTexture2CommitmentTestCase
* Added UncommittedRegionsAccessTestCase
* Added extension support check
Change-Id: I6f5b80631ae1f7bd9897db7544238ae5ff19213b
Alexander Galazin [Wed, 11 Jan 2017 06:51:53 +0000 (07:51 +0100)]
Run GTF EGL tests only in 'window' configs
Fixes OSS CTS #40
Change-Id: I6967bb63fb58863bcf143a27c01e84ccf536ccf8
Filip Demczewski [Thu, 8 Dec 2016 10:08:31 +0000 (11:08 +0100)]
Added CTS_ARB_shader_viewport_layer_array specification proposal
Change-Id: Id391811a9e5d780b2afac5efeceb1c3f67f8d7cb
Alexander Galazin [Thu, 15 Dec 2016 17:54:41 +0000 (18:54 +0100)]
DrawElementsBaseVertexTestBase: call base class init() method
Fixes OpenGL OSS-CTS issue 26
Without it, test fail with:
InternalError (Value for parameter 'VERSION' not found in map at tcuStringTemplate.cpp:113)
because the shader m_specializationMap is not initialized.
Original MR https://gitlab.khronos.org/opengl/oss-cts/merge_requests/129
by Nicolai Hähnle
Change-Id: Ifff450f4b91bc21d368bd6bfbd145849ae80765a
Nicolai Hähnle [Fri, 23 Dec 2016 11:01:15 +0000 (12:01 +0100)]
gl4cComputeShaderTests: fix pipeline-compute-chain for changes in tcuMatrix
tcuMatrix now stores data in column-major order (i.e. matches the default
setting of GLSL). Previously, it stored data in row-major order, so the
test needs to be updated to reflect this change.
Change-Id: I89f13acbb1dd2858fa4768eca5efa40ee516805e
Nicolai Hähnle [Sat, 10 Dec 2016 18:46:11 +0000 (19:46 +0100)]
gl4cPipelineStatisticsQueryTests: fix expected vertex count
Restore the orginal version of
getExpectedVerticesSubmittedQueryResult. It got replaced by logic
that computes the number of primitives instead.
Change-Id: I7fbd8cfe2a2310ab8ec43c7cc8e7605d0060f65f
Pyry Haulos [Thu, 12 Jan 2017 23:27:44 +0000 (15:27 -0800)]
Fix memoryTypeBits validation in memory.requirements.buffer
Memory type bits returned for a subset of all possible usage flags must
be a superset of memory type bits returned for all usage flags case, not
the other way around.
Affects: dEQP-VK.memory.requirements.buffer.*
Vulkan CTS issue: 578
Change-Id: Iff34ffbf571fa7b0bd2d5e769e32d82a4dc9229e
Maciej Jesionowski [Thu, 12 Jan 2017 11:23:44 +0000 (12:23 +0100)]
Bugfix: provide enough VkClearValues in ImageSamplingInstance
Affected tests:
- dEQP-VK.pipeline.image.*
- dEQP-VK.pipeline.image_view.*
Fixes #575
Change-Id: Ie2fbdf9b926c677c54aca4b4cfd6b107be451c7f
Slawomir Cygan [Tue, 10 Jan 2017 15:59:29 +0000 (16:59 +0100)]
Do not require unsupported sparse images in image_format_properties test
Affects tests: dEQP-VK.api.info.image_format_properties*
This change removes the requirement for supporting sparse residency images:
- with compressed format
- with non-color format
- with format of pixel size that is not power ot two (for example: RGB).
Additionally do not require 1D sparse residency images (not in spec),
and require 2D/3D residency images basing on the device capabilities.
Spec 28.1. Sparse Resource Features:
"A sparse image created using VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT
supports all non-compressed color formats with power-of-two element
size that non-sparse usage supports.
(...)
sparseResidencyImage2D: Support for creating 2D single-sampled VkImage
objects with VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.
sparseResidencyImage3D: Support for creating 3D VkImage objects with
VK_IMAGE_CREATE_SPARSE_RESIDENCY_BIT.
"
Fixes bug: gitlab!567
Change-Id: If7d72c4392299e4c449a911140632dcc114fc908
Alexander Galazin [Thu, 12 Jan 2017 11:53:30 +0000 (12:53 +0100)]
Fix VkCmdImageBlit usage in blitting tests
Some tests call VkCmdImageBlit with a 2D destination image and a region
with dstOffset[0].z=1, dstOffset[1].z=0.
The spec says that it should be the opposite.
Affects:
dEQP-VK.api.copy_and_blit.blit_image.simple_tests.mirror_xy_nearest
dEQP-VK.api.copy_and_blit.blit_image.simple_tests.mirror_y_nearest
Components: Vulkan
Fixes Vulkancts issue: #576
Change-Id: Iaa05f3b38cd46c39d313bd044b37535c112b77dd
Maciej Jesionowski [Tue, 10 Jan 2017 12:15:45 +0000 (13:15 +0100)]
Use deAlign64 in api.buffer and sparse_resources
Change-Id: I087d99db106555a806c6864e67c9af12fb5c67b2
Maciej Jesionowski [Tue, 10 Jan 2017 12:09:14 +0000 (13:09 +0100)]
Add deInt64 alignment functions to delibs
Change-Id: Icf66e7157ae6e972e0d7ef19486fa5b776da1eb0
Alexander Galazin [Wed, 11 Jan 2017 07:49:11 +0000 (08:49 +0100)]
Revert "Generate flat triangles in polygon offset tests."
This reverts commit
b95e37d04894664aae9cb96a38063684fb4a070d.
Alexander Galazin [Wed, 11 Jan 2017 07:37:13 +0000 (08:37 +0100)]
Generate flat triangles in polygon offset tests.
The tests setup triangles with a depth slope and then call
glPolygonOffset(factor, units) with variable units and
always 0 slope scaled factor.
According to the GLES 3.2 spec 13.7.2 Depth Offset:
"void PolygonOffset( float factor, float units );
factor scales the maximum depth slope of the polygon, and units scales an
implementation-dependent constant that relates to the usable resolution of the
depth buffer."
If the polygon has a depth slope,
the depth slope factor has to be not equal to zero.
The rationale of the slope scaled factor is to ensure that the offset
(equation 13.11) is greater than max(dz/dx, dz/dy) for all resolutions,
which may not be the case for the constant offset since it doesn't take
the screen resolution into account.
Change-Id: Ic87d641f814e2f977ccccc45944bef7f4e254add
Pyry Haulos [Tue, 10 Jan 2017 20:21:17 +0000 (12:21 -0800)]
Merge gerrit/vulkan-cts-1.0-dev into gerrit/opengl-cts-dev
Change-Id: I027046f847315da035a19ae1267d18bea84e67b2
Pyry Haulos [Tue, 10 Jan 2017 20:21:14 +0000 (12:21 -0800)]
Merge gerrit/vulkan-cts-1.0.2 into gerrit/vulkan-cts-1.0-dev
Change-Id: I104cae5b8ad882dbead6e15b6cf6514b8b9f48cd
Pyry Haulos [Tue, 10 Jan 2017 20:20:15 +0000 (12:20 -0800)]
Update mustpass version to 1.0.2 in README.md
Change-Id: Ia683c10238ede6a712a46d14d936ef0032937e9e
Pyry Haulos [Tue, 10 Jan 2017 20:18:40 +0000 (12:18 -0800)]
Merge gerrit/vulkan-cts-1.0-dev into gerrit/opengl-cts-dev
Change-Id: I3a85bd404f90b45cfb1305f6125c236518ffa4f4
Pyry Haulos [Tue, 10 Jan 2017 20:18:34 +0000 (12:18 -0800)]
Merge gerrit/vulkan-cts-1.0.2 into gerrit/vulkan-cts-1.0-dev
Change-Id: I74387a3ed98feb4162b3d49a411cb220d0f050e9
Pyry Haulos [Sun, 8 Jan 2017 22:42:32 +0000 (14:42 -0800)]
Allow unknown KHX extensions
Affects:
dEQP-VK.api.info.instance.extensions
dEQP-VK.api.info.device.extensions
Change-Id: Iece3f83514f5f6e197fd0c1e1cf628f7cfe68bd2
Weiwan Liu [Fri, 23 Dec 2016 00:25:19 +0000 (16:25 -0800)]
Correct expected error to INVALID_OPERATION
Correct error code per ES 3.2 spec:
"An INVALID_OPERATION error is generated if attachment is
COLOR_ATTACHMENTm where m is greater than or equal to the
value of MAX_COLOR_ATTACHMENTS."
Change-Id: I5ee47fbb5d1f099d62cee9e20fd8ae0b55d766b3
Maciej Jesionowski [Mon, 9 Jan 2017 11:05:32 +0000 (12:05 +0100)]
Remove pooled objects from alloc_callback_fail group
The pooled objects (VkDescriptorSet and VkCommandBuffer) are properly
handled in the alloc_callback_fail_multiple group. The removed cases
were redundant, and also didn't work correctly with pooled objects.
Removed tests:
- dEQP-VK.api.object_management.alloc_callback_fail.descriptor_set
- dEQP-VK.api.object_management.alloc_callback_fail.command_buffer_primary
- dEQP-VK.api.object_management.alloc_callback_fail.command_buffer_secondary
Change-Id: I212b544096e6e1fd5b98c7985d836e41247107e9
Alexander Galazin [Mon, 9 Jan 2017 14:17:02 +0000 (15:17 +0100)]
Skip anisotropic filtering tests if the feature is not supported.
dEQP-VK.texture.filtering_anisotropy.* tests assume that an
implementation supports anisotropic filtering and should be skipped in
case VkPhysicalDeviceFeatures::samplerAnisotropy is VK_FALSE.
Fixes #570
Change-Id: I0601fe67a0e81b71670dac7ec8b83b4dc35c576c
Anuj Phogat [Fri, 30 Sep 2016 17:46:49 +0000 (10:46 -0700)]
ES{2,3}: Clamp the x, y location of the viewport's bottom-left corner
Affected tests:
dEQP-GLES2.functional.state_query.integers.viewport_getfloat
dEQP-GLES2.functional.state_query.integers.viewport_getinteger
dEQP-GLES3.functional.state_query.integers.viewport_getfloat
dEQP-GLES3.functional.state_query.integers.viewport_getinteger
dEQP-GLES3.functional.state_query.integers.viewport_getinteger64
Above CTS tests are not clamping the x, y location of the viewport's
bottom-left corner as recommended by ARB_viewport_array and
OES_viewport_array.
Tests later compare the clamped x, y values read back from the driver
with reference unclamped x, y values resulting in to false failures.
From OES_viewport_array:
"The location of the viewport's bottom-left corner, given by (x,y), are
clamped to be within the implementation-dependent viewport bounds range.
The viewport bounds range [min, max] tuple may be determined by calling
GetFloatv with the symbolic constant VIEWPORT_BOUNDS_RANGE_OES"
Change-Id: I380e0d03291d3b4b8cbc698ff80624860b0f2f05
(cherry picked from commit
555ed53736072fc9b50993cbc0550a70902a0bf9)
Maciej Jesionowski [Thu, 5 Jan 2017 10:21:40 +0000 (11:21 +0100)]
Update zlib and libpng dependencies
Fixes #568
Change-Id: I8ee890a2176bf6d42d3dd8b119b7263034127960
(cherry picked from commit
8744b7d393cbdba8a2c356e9d4e4d625b9468f9a)
Anuj Phogat [Fri, 30 Sep 2016 17:46:49 +0000 (10:46 -0700)]
ES{2,3}: Clamp the x, y location of the viewport's bottom-left corner
Affected tests:
dEQP-GLES2.functional.state_query.integers.viewport_getfloat
dEQP-GLES2.functional.state_query.integers.viewport_getinteger
dEQP-GLES3.functional.state_query.integers.viewport_getfloat
dEQP-GLES3.functional.state_query.integers.viewport_getinteger
dEQP-GLES3.functional.state_query.integers.viewport_getinteger64
Above CTS tests are not clamping the x, y location of the viewport's
bottom-left corner as recommended by ARB_viewport_array and
OES_viewport_array.
Tests later compare the clamped x, y values read back from the driver
with reference unclamped x, y values resulting in to false failures.
From OES_viewport_array:
"The location of the viewport's bottom-left corner, given by (x,y), are
clamped to be within the implementation-dependent viewport bounds range.
The viewport bounds range [min, max] tuple may be determined by calling
GetFloatv with the symbolic constant VIEWPORT_BOUNDS_RANGE_OES"
Change-Id: I380e0d03291d3b4b8cbc698ff80624860b0f2f05
Nicolai Hähnle [Thu, 22 Dec 2016 14:54:02 +0000 (15:54 +0100)]
DrawElementsBaseVertexTestBase: add missing '}'
Fixes GL45-CTS.draw_elements_base_vertex_tests.invalid_*
Change-Id: I1d1a3a1e326340631c1b4def18f7fa1fdc0c36a9
Nicolai Hähnle [Mon, 12 Dec 2016 09:10:52 +0000 (10:10 +0100)]
DrawElementsBaseVertexTestBase: call base class init() method
Without it, test fail with:
InternalError (Value for parameter 'VERSION' not found in map at tcuStringTemplate.cpp:113)
because the shader m_specializationMap is not initialized.
Change-Id: I993ae7a5f16be4f00b1c10b3013ae0333ef9777d
Pyry Haulos [Fri, 6 Jan 2017 20:45:02 +0000 (12:45 -0800)]
Merge gerrit/vulkan-cts-1.0-dev into gerrit/opengl-cts-dev
Change-Id: I326010e9a243d5c94c63d23c48c39e1dcab34320
Pyry Haulos [Fri, 6 Jan 2017 20:44:58 +0000 (12:44 -0800)]
Merge gerrit/vulkan-cts-1.0.2 into gerrit/vulkan-cts-1.0-dev
Change-Id: I5d98762111c8dba323a81a7179026d009f39934e
Pyry Haulos [Fri, 6 Jan 2017 20:44:56 +0000 (12:44 -0800)]
Merge gerrit/vulkan-cts-1.0.1 into gerrit/vulkan-cts-1.0.2
Change-Id: Ic1b09c684e6535fcaba6e4d7dd407b2fbbe40b6d
Pyry Haulos [Wed, 4 Jan 2017 22:30:07 +0000 (14:30 -0800)]
Optimize swapchain OOM tests
Swapchain OOM tests were hitting timeouts on some platforms. This commit
optimizes the tests in two ways:
* VkInstance, VkDevice, and VkSurface are now re-used across all
sub-cases.
* Loop is re-written to avoid re-creating swapchains for parameters
0..N-1 when testing parameter combination N.
Affects: dEQP-VK.wsi.*.swapchain.simulate_oom.*
Bug:
33555898
Change-Id: Ib5e1a4944e4c400a5d430c5a3d20fe0930c6eed0
Lionel Landwerlin [Fri, 23 Dec 2016 17:11:14 +0000 (17:11 +0000)]
Skip unsupported attachment formats in granularity tests
Change-Id: Iffbd79d0ddb3c212f75895c266b3e3ec7d2a01da
Lionel Landwerlin [Fri, 23 Dec 2016 17:37:56 +0000 (17:37 +0000)]
Fix framebuffer number of layers in granularity tests
Change-Id: Idf2f1ee6f6f9ab5996ad5d07a7fed84241c57e9b
Pyry Haulos [Wed, 4 Jan 2017 21:46:48 +0000 (13:46 -0800)]
Reset Android window state when it is released
This avoids leaking window state between tests in WSI tests.
Affects: dEQP-VK.wsi.android.*
Change-Id: I618485fbbc1bb83c86437bb04fbcbfa65b4c04c2
Jason Ekstrand [Tue, 6 Dec 2016 14:27:56 +0000 (15:27 +0100)]
Add OpSMod and OpSRem tests
New tests:
- dEQP-VK.spirv_assembly.instruction.compute.opsrem.positive
- dEQP-VK.spirv_assembly.instruction.compute.opsrem.all
- dEQP-VK.spirv_assembly.instruction.compute.opsmod.positive
- dEQP-VK.spirv_assembly.instruction.compute.opsmod.all
- dEQP-VK.spirv_assembly.instruction.graphics.srem.positive_*
- dEQP-VK.spirv_assembly.instruction.graphics.srem.all_*
- dEQP-VK.spirv_assembly.instruction.graphics.smod.positive_*
- dEQP-VK.spirv_assembly.instruction.graphics.smod.all_*
Fixes #520
Change-Id: I16fe1e920960b7537ef097952e48dcdcc9141246
Zoltan Janos Szabo [Thu, 17 Nov 2016 11:53:14 +0000 (12:53 +0100)]
Add getDeviceMemoryCommitment test
Fixes #553
Change-Id: Ia029f895e5f13187a2bc0fb11c583ab3119f48f7
Pyry Haulos [Thu, 5 Jan 2017 18:24:10 +0000 (10:24 -0800)]
Merge gerrit/vulkan-cts-1.0-dev into gerrit/opengl-cts-dev
Change-Id: I0e30246d8afce9f4b3704c941c57e9fe88098268
Pyry Haulos [Thu, 5 Jan 2017 18:04:19 +0000 (10:04 -0800)]
Merge aosp/master into gerrit/vulkan-cts-1.0-dev
Change-Id: I6d61156e336937d8c2f534a410b6f2a92987f037
Pyry Haulos [Thu, 5 Jan 2017 17:58:20 +0000 (09:58 -0800)]
Merge gerrit/vulkan-cts-1.0-dev into gerrit/opengl-cts-dev
Change-Id: If9a06423b0fb4be658187ee4df49817dcc9b0394
Pyry Haulos [Thu, 5 Jan 2017 17:58:17 +0000 (09:58 -0800)]
Merge gerrit/vulkan-cts-1.0.2 into gerrit/vulkan-cts-1.0-dev
Change-Id: If550d50796597ec466efb30b1cbbd9aab4b13890
Pyry Haulos [Thu, 5 Jan 2017 17:58:15 +0000 (09:58 -0800)]
Merge gerrit/vulkan-cts-1.0.1 into gerrit/vulkan-cts-1.0.2
Change-Id: I4dc1513b1528b8182cfd75f4ed9184fa8f383e47
Pyry Haulos [Thu, 5 Jan 2017 17:58:14 +0000 (09:58 -0800)]
Merge gerrit/vulkan-cts-1.0.0 into gerrit/vulkan-cts-1.0.1
Change-Id: I3ace822f33fb20acd6109886458b0d16c1624bd5
Iago Toral Quiroga [Mon, 2 Jan 2017 11:22:29 +0000 (12:22 +0100)]
Fix assert for combined depth/stencil formats
This was using tcu::isCombinedDepthStencilType(format.type) for this,
but this only checks the type and for 24bit depth formats with a
x8 d24 setup it will incorrectly identify them as combined
depth/stencil even though the underlying format is depth-only. Fix
this by testing if the format has depth and stencil components
instead.
Prevents crashes in tests with the following signature (144 tests):
dEQP-VK.texture.shadow.*x8_d24*
Fixes #563
Change-Id: Ie43e140a0461de49c02e0d97eeed38fa8ef24040
Maciej Jesionowski [Thu, 15 Dec 2016 15:45:31 +0000 (16:45 +0100)]
Fix buffer size and alignment issues in api.buffer
- Don't use maxTexelBufferElements, it doesn't apply here
- Remove unused throw/catch blocks (catch block was never reachable)
- Add a case that exercises huge buffer sizes
- Add a check for sparseAddressSpaceSize
Modified tests:
- dEQP-VK.api.buffer.create_buffer_*
Fixes #558
Change-Id: Ia6b46d8c34a4eef4a6136532d0cb626d0b64b560
Maciej Jesionowski [Thu, 5 Jan 2017 10:21:40 +0000 (11:21 +0100)]
Update zlib and libpng dependencies
Fixes #568
Change-Id: I8ee890a2176bf6d42d3dd8b119b7263034127960
Treehugger Robot [Wed, 4 Jan 2017 19:03:50 +0000 (19:03 +0000)]
Merge "Merge aosp/upstream-vulkan-cts-1.0-dev into aosp/master"
Pyry Haulos [Wed, 4 Jan 2017 18:52:48 +0000 (18:52 +0000)]
DO NOT MERGE: Remove changed texel buffer access tests from mustpass
am:
302bd7c6b3 -s ours
Change-Id: I1cefd7be538f766444576ae02a4b28b639a3e8cb
Pyry Haulos [Wed, 4 Jan 2017 18:52:44 +0000 (18:52 +0000)]
DO NOT MERGE: Remove a few changed Vulkan tests from mustpass
am:
09e8b66d03 -s ours
Change-Id: I595b0c3b589b73c806145602957875b57cffe84a
Pyry Haulos [Wed, 4 Jan 2017 17:32:13 +0000 (09:32 -0800)]
Merge aosp/upstream-vulkan-cts-1.0-dev into aosp/master
Change-Id: Id23cc2f5a2ff4238fbf56d8c9cd4330dd166971a
Mark Adams [Wed, 14 Dec 2016 22:02:52 +0000 (17:02 -0500)]
Add support for EGL_EXT_pixel_format_float
This updates the egl.xml version and identifies/skips the float configs
Bug:
32969314
Change-Id: Iadeb0d4a76a06b2be716f3e76a1cc9fb69aff16a
Pyry Haulos [Wed, 4 Jan 2017 16:20:43 +0000 (08:20 -0800)]
DO NOT MERGE: Remove changed texel buffer access tests from mustpass
NYC CTS bug-fix changes the SPIR-V for these tests causing drivers that
passed old CTS fail the new version.
Bug:
33041922
Change-Id: I7e23a700693d5e6ee7c4264e05d449f49f820516
Pyry Haulos [Wed, 4 Jan 2017 00:45:19 +0000 (16:45 -0800)]
DO NOT MERGE: Remove a few changed Vulkan tests from mustpass
Original NYC CTS was decorating struct member locations differently and
this was later fixed causing drivers that pass original NYC CTS fail.
Bug:
34060750
Bug:
33041922
Change-Id: I3fe33960a74fb8f568aa7e1327bc09de8bc3df41
Pyry Haulos [Wed, 4 Jan 2017 00:43:24 +0000 (16:43 -0800)]
Merge gerrit/vulkan-cts-1.0-dev into gerrit/opengl-cts-dev
Change-Id: Icef189b9af29e090fbb4940af970d91951c2d6e6
Pyry Haulos [Wed, 4 Jan 2017 00:43:21 +0000 (16:43 -0800)]
Merge gerrit/vulkan-cts-1.0.2 into gerrit/vulkan-cts-1.0-dev
Change-Id: I598c6d52ece7b00a8dd0af10b31ea144ebc6ba80
Pyry Haulos [Wed, 4 Jan 2017 00:42:53 +0000 (16:42 -0800)]
Revert "Add new files to Android.mk"
This reverts commit
4ed5a72717f9dc776f7f63aa21aca2c727524d0f.
Pyry Haulos [Wed, 4 Jan 2017 00:25:16 +0000 (16:25 -0800)]
Merge gerrit/vulkan-cts-1.0-dev into gerrit/opengl-cts-dev
Change-Id: Ieccd2e74ca11d5a4b7b06f78bae0094e5a870b69
Pyry Haulos [Wed, 4 Jan 2017 00:25:13 +0000 (16:25 -0800)]
Merge gerrit/vulkan-cts-1.0.2 into gerrit/vulkan-cts-1.0-dev
Change-Id: I33cca201217f9c9a258c61e0bb6178bb7694a96e
Pyry Haulos [Wed, 4 Jan 2017 00:24:51 +0000 (16:24 -0800)]
Add new files to Android.mk
Change-Id: Ib396f9504a3b5523314cd2d5ab9375a346dbb219
Pyry Haulos [Wed, 4 Jan 2017 00:20:18 +0000 (16:20 -0800)]
Merge gerrit/vulkan-cts-1.0-dev into gerrit/opengl-cts-dev
Change-Id: I6ae311b56b24f91aae76a908b8086bcc8afeaacf
Pyry Haulos [Wed, 4 Jan 2017 00:20:14 +0000 (16:20 -0800)]
Merge gerrit/vulkan-cts-1.0.2 into gerrit/vulkan-cts-1.0-dev
Change-Id: I4b9cb490746a875501c5213b0a5899a3a487be03
Pyry Haulos [Wed, 4 Jan 2017 00:07:06 +0000 (16:07 -0800)]
Merge gerrit/vulkan-cts-1.0.1 into gerrit/vulkan-cts-1.0.2
Jorg Wagner [Thu, 15 Dec 2016 21:52:06 +0000 (22:52 +0100)]
Skip integer formats that expose a max. sample count of 1
Actually fixes #538
As agreed in the ticket we need to skip integer formats if they
need to be sampled and only expose a max. sample count of 1.
Change-Id: Ia67c8d84cedcae09804e83cfefeb279b329d07b9
Kalle Raita [Fri, 16 Dec 2016 17:45:41 +0000 (09:45 -0800)]
install.py continues if adb uninstall fails
ADB uninstall fails with non-zero return code if the package to be
uninstalled does not exist. To allow the installation proceed when
there's no previous package, the non-zero return value should be ignored
for the uninstall step.
There are historical reasons for not relying on -r -d for
install (replace, allow downgrade), which I can't recall right now.
Test: Installed succesfully on device without deqp package.
Bug:
33961937
Change-Id: I401f3861f8d68f352747c041bd60cd372b48134b
Tina Zhou [Thu, 22 Dec 2016 05:20:13 +0000 (05:20 +0000)]
Merge "Blacklist match_different_[member_]struct_names SSO validation tests." into nougat-cts-dev
am:
4acf725866
Change-Id: I2cc66f7fb73fecf721da7704aa5a35384c5f5d4d