Merge vk-gl-cts/vulkan-cts-1.1.6 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I3ed8d5f4de16750c664e4c8908b8b971f48f7593

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.1.6

Change-Id: I004160d9169fcb42a928c10d76b5355dd671e5a8

Fix slow run time of OpPhi wide test

dEQP-VK.spirv_assembly.instruction.compute.opphi.wide was using 1024
arguments for OpPhi SpirV instruction. This was causing very slow
run/compile time on some devices and therefore failing the test due to
timeout.

This CL halves the number of arguments given to the OpPhi instruction to
avoid timeouts.

Affects:

dEQP-VK.spirv_assembly.instruction.compute.opphi.wide

VK-GL-CTS Issue: 2681

Components: Vulkan
Change-Id: Ice41e983b8da866916e07c1b52d7ccfd65001077

Fixes descriptor indexing tests cache issue

Tests were not invalidating the results
buffer data before verifying.

Components: Vulkan
VK-GL-CTS Issue: 2682

Affects:
dEQP-VK.descriptor_indexing.*

Change-Id: Ia2afa6697d12fd3f4ebc8ef29ae01d166f290a7f

Loosen basic subgroup elect test vertex checks

This commit takes into account any shader in the vertex pipeline may run
multiple times per vertex when checking results in the basic subgroup
elect test.

In addition, it contains a few indentation fixes as well as fixes for
several validation errors in the whole test group.

Affected tests:
dEQP-VK.subgroups.basic.*

Components: Vulkan
VK-GL-CTS issue: 2162

Change-Id: Ib0aaf929bca5ce4939a3e2fd98a69a0812f1aa0b

Add tests for VK_EXT_subgroup_size_control

For compute and graphics pipelines:

* Check the required subgroup size set by the tests is honored by the
  driver (including for full subgroups and allow varying cases for
  computer shaders).
* Check that enabling allow varying subgroup size reports a subgroup
  size that falls within the limits of the driver.
* For compute pipeline, check that require full subgroups works as
  expected.

On top of that, added requiredSubgroupSize case for subgroup's compute
shader tests that check all possible required subgroup sizes announced
by the driver.

Also there is a test that checks subgroup size control limits sanity.

Affected tests:

  dEQP-VK.subgroups.*

Components: Vulkan
VK-GL-CTS issue: 1864

Change-Id: Ia48deb9318420889754bd97545045f85b3c6fed7

Fixes sparse resource SNORM tests

Tests verifies value out of range

Components: Vulkan
VK-GL-CTS Issue: 2598

Affects:
dEQP-VK.sparse_resources.image_sparse_binding.*snorm*
dEQP-VK.sparse_resources.mipmap_sparse_residency.*snorm*

Change-Id: Id11addbc60c44ca7f20f86e11dbc7c833bbe7da7

Merge vk-gl-cts/vulkan-cts-1.2.0 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: Ie6ac4a6779ee2dd624590df63f152f661c089223

Merge vk-gl-cts/vulkan-cts-1.1.6 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I8f4da52990ac1b7ae96414375e9bea652ddbc7d8

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.1.6

Change-Id: Ie294402abe02966f55089430d9557b5f0bc80c5f

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.2.0

Change-Id: Icda35a4b6ca926511422ec96c4675b9012973ef1

Permit not flushing operands for FP16 DenormFlushToZero add, dot, vmulm

Where `d` is a denormal value such that `2 * d` is not denormal:

- The F16 OpFAdd tests assumed (d + d) is zero when DenormFlushToZero is
  used. This is not true if the operands (d) are not flushed to zero by
  the implementation, which is allowed by the Vulkan specification.
- The F16 OpDot test assumed dot(vec2(d, d), vec2(1, 1)) is zero. Since
  that is equivalent to (d + d), it has the same problem as OpFAdd.
- The F16 OpVectorTimesMatrix test made the same assumption as the OpDot
  test but for both columns of a 2-by-2 matrix, with the same problem.

These tests are changed to accept `2 * d` as a valid result alongside
zero, such that they will pass both on implementations which flush
operands to zero and on implementations which only flush results to
zero.

Modified tests:

    dEQP-VK.spirv_assembly.instruction.*.float_controls.fp16.input_args.denorm_*flush_to_zero*
    dEQP-VK.spirv_assembly.instruction.compute.float_controls.independence_settings.denorm_ind_*_fp16_flush_*

VK-GL-CTS Issue: 2644

Components: Vulkan

Change-Id: I0ef4a57915e3bdab100f55f52d21913f87af4927

Amend Vulkan Conformance Submission Package Requirements

The current practice is that all bugfixes should be merged
into the CTS repo before being used in a submission.
This change promotes this practices to a requirement.

Compontents: Vulkan

Change-Id: I61eef5522aecf97a3d42c49777681eb671affe87

Splits long shader in memory model tests

Components: Vulkan
VK-GL-CTS Issue: 2625

Affects:
dEQP-VK.memory_model.*

Change-Id: I8478e268fa5988b015acd2dd70cd370bfc4c082c

Merge vk-gl-cts/vulkan-cts-1.2.0 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: Ia15f227ae78f00e0d142f0b2ce3d966d10a09cab

Merge vk-gl-cts/vulkan-cts-1.1.6 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I4c86e4fe92acca4c4260ab43a492ace9f527f986

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.2.0

Change-Id: Id37d7b8198e8d5867e4b802e638eb1e0010c2351

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.1.6

Change-Id: I840b7caa71134a32ba49b484b716b97c31ace9ee

Merge vk-gl-cts/vulkan-cts-1.1.4 into vk-gl-cts/vulkan-cts-1.1.5

Change-Id: I04485faa2b85d5af689da6143b130e3c92bd9253

Merge vk-gl-cts/vulkan-cts-1.1.3 into vk-gl-cts/vulkan-cts-1.1.4

Change-Id: Ic81486b5b158f0edf964d29a6e084a9c6e2dee00

Fixes multisample resolve sRGB cases

CTS expects downsampling in linear space,
where spec says implementation dependent

Components: Vulkan
VK-GL-CTS Issue: 2584

Affects:
dEQP-VK.renderpass.suballocation.multisample_resolve.*_srgb*

Change-Id: I8512ef058b03fe55c5429d3add5d2f4dc890c663

Fix validation errors in subgroups tests

Some subgroups tests were attempting to create a descriptor pool of size
0, which is invalid. Instead, skip the whole construction of descriptor
sets since they aren't needed.

Components: Vulkan
Affects: dEQP-VK.subgroups.*.graphics.*

Change-Id: I5abedcc2360a0336c521aea855a9725b127c9a0f
(cherry picked from commit a887be07dd7b63ce89794af9f70f893dc2de5955)

Fixes multisample resolve sRGB cases

CTS expects downsampling in linear space,
where spec says implementation dependent

Components: Vulkan
VK-GL-CTS Issue: 2584

Affects:
dEQP-VK.renderpass.suballocation.multisample_resolve.*_srgb*

Change-Id: I9ab4ebf54e38c5c2cd1cc626dcf77f904d0e1e4d

Coverage not correct for pixels on shared edges

Added coverage check for the second "friend" triangle.

Components: Vulkan
VK-GL-CTS Issue: 2580

Affects:
dEQP-VK.rasterization.*

Change-Id: If30814ed81fb9b91ace4655c0ba50dfbcff7b37b

Allow Vulkan CTS 1.2.1.2

Components: Vulkan

Change-Id: Ibc9f72ef89e85469c2e54a74b5cf80367ad5deed

Fixes missing timeline semaphore checks

Components: Vulkan
VK-GL-CTS Issue: 2572

Affects:
dEQP-VK.synchronization.op.multi_queue.*
dEQP-VK.synchronization.smoke.timeline_semaphores

Change-Id: I9829ae9f09ca3a45c6ce7a6aad5ad21c6caad3cd

Fixes sparse snorm tests

Test using values outside of defined
snorm range.

Components: Vulkan
VK-GL-CTS Issue: 2573

Affects:
dEQP-VK.sparse_resources.mipmap_sparse_residency.*snorm*

Change-Id: Id0027c252ea348dd4756809b280d92072219975c

Fixes timeline sempahore queue flags

Components: Vulkan
VK-GL-CTS Issue: 2571

Affects:
dEQP-VK.synchronization.signal_order.timeline_semaphore.*

Change-Id: Ibba5d7ac7279cb87a53433ef4c04b57050629efb

Merge vk-gl-cts/vulkan-cts-1.2.0 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I96350d6395f5f21a2d33e74ecd153eac1b420bc9

Merge vk-gl-cts/vulkan-cts-1.1.6 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: Ia8467b7749d07e2f73b8aaef4c9a51a8b7789495

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.1.6

Change-Id: I19b5ee4378122e348e1d1fa427e0322c0900a94b

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.2.0

Change-Id: Ib36fe69650d2c93592c3d1ea48d5871326871bae

Merge vk-gl-cts/vulkan-cts-1.1.4 into vk-gl-cts/vulkan-cts-1.1.5

Change-Id: Iea4251d6635a9fa68f0426714641acd50f87e166

Revert back to libpng 1.6.27

Components: Framework

Change-Id: I6576aeff8d0a1c9375f829518cb4035c53af5e82
(cherry picked from commit f3b873529b8f1b62e8a79055045c1a368ce21ea4)

Merge vk-gl-cts/vulkan-cts-1.2.0 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I7819afa5a0f79fa9859828376c59273bee3ab120

Merge vk-gl-cts/vulkan-cts-1.1.6 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I79af85ae37015656853e9f12f9b1beefe0a1df8f

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.1.6

Change-Id: I39a6bf72421c7adc0e28469622fe6880d8b54c48

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.2.0

Change-Id: Ib748d52d82158a63ba3b79a528bcb54276e1df57

Merge vk-gl-cts/vulkan-cts-1.1.4 into vk-gl-cts/vulkan-cts-1.1.5

Change-Id: I32ee9cf4f1a17df56e76e36d57490aa66a20e972

Update libpng

Components: Framework

Change-Id: I5f218837c8a58e35b2d37ca170a93eb01d6b0057

Merge vk-gl-cts/vulkan-cts-1.2.0 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: Ie4f286836d822239983ba886eee14cd991676afe

Merge vk-gl-cts/vulkan-cts-1.1.6 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I6bb8696c337a0cef0501cf6879a43bc4f95bdfcd

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.1.6

Change-Id: I4801810e09dab1c07ec862b7a3add86291408a22

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.2.0

Change-Id: I0c4266d49b22f1d2f4f02d5c3a02c12665be38ec

Merge vk-gl-cts/vulkan-cts-1.1.4 into vk-gl-cts/vulkan-cts-1.1.5

Change-Id: I08d446d669112864315659de6c9443e176e9f575

Merge vk-gl-cts/vulkan-cts-1.1.3 into vk-gl-cts/vulkan-cts-1.1.4

Change-Id: I8dd42b54d5d9b559f86affd2518cce0ea13f4b33

Fix renderdoc library name on android

Components: Framework, Vulkan
VK-GL-CTS Issue: 2562

Change-Id: I05657bd83f7bc97107792dd4d742c346883f589b

Fixes descriptor indexing tests

Tests were exceeding the max number of input
attachments

Components: Vulkan
VK-GL-CTS Issue: 2561

Affects:
dEQP-VK.descriptor_indexing.*

Change-Id: If135ecee8aca6480d28f2e66ca574a5a9fbbc201

Fix incorrect buffer type in some protected memory tests

Certain Vulkan tests, such as
dEQP-VK.protected_memory.buffer.fill.<type>_buffer.*, create a BufferView with
a format that is always set to VK_FORMAT_R32G32B32A32_UINT, independently of
the actual test type (float_buffer, integer_buffer, unsigned_buffer). Reading
floats from an integer BufferView may return incorrect values on some
implementations.

This patch propagates a VkFormat parameter so that each test can create a
BufferView with the appropriate format.

Components: Vulkan

Affects:
 - dEQP-VK.protected_memory.buffer.copy.*
 - dEQP-VK.protected_memory.buffer.fill.*
 - dEQP-VK.protected_memory.buffer.update.*

VK-GL-CTS Issue: 2551

Change-Id: I1e79ae2761a028025148c3a63775a0d464278285
(cherry picked from commit 8d42d5d9c0e48ea76fad2cb4e232334b4a849d0f)

Fix support check for sparse input attachment test

Components: Vulkan
VK-GL-CTS Issue: 2528

Affects:
dEQP-VK.renderpass.suballocation.attachment_sparse_filling.*

Change-Id: I4c143273e9f337f7d5dcfc0230c800955304633b

Fixes capability checks in subgroup tests

Components: Vulkan
VK-GL-CTS Issue: 2517

Affects:
dEQP-VK.subgroups.vote.*
dEQP-VK.subgroups.vote.frag_helper.*

Change-Id: I49832e7fe34eddbc655ca4d5d7e155d311685c8a

Splits protected memory test shaders

Components: Vulkan
VK-GL-CTS Issue: 2516

Affects:
dEQP-VK.protected_memory.stack.stacksize_*

Change-Id: I5ac52229cd59e9b4b92914a60b0ffa37cdff70c7

Merge vk-gl-cts/vulkan-cts-1.2.0 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: Ife743420baa1f4998a8d0f1b9ee32dd93a5a74f4

Merge vk-gl-cts/vulkan-cts-1.1.6 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I7bac3dcb8580348fe7aa1238145c55640a3f75da

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.2.0

Change-Id: Ic03b0ca5deecc32b574df399b12314d16150574f

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.1.6

Change-Id: I9257feb431f2edf54f39f0912df10c115115f75a

Merge vk-gl-cts/vulkan-cts-1.1.4 into vk-gl-cts/vulkan-cts-1.1.5

Change-Id: I9a0205ade4419a8e0260ac487fa3e443e93609cc

Stop 10bit formats requiring GL_SRGB encoding

The EGL_KHR_gl_colorspace spec does not require formats that
are not SRGB8 to respect the given EGL_GL_COLORSPACE_KHR.

Components: EGL

VK-GL-CTS issue: 2288

Affects:

dEQP-EGL.functional.wide_color.window_1010102_colorspace_srgb
dEQP-EGL.functional.wide_color.pbuffer_1010102_colorspace_srgb

Change-Id: I1d9c502ef2f7068cf13d13564abdfe2d3c75b93b

Add calls to touchWatchdog()

...to prevent a timeout when creating eight pipelines with large compute
shaders.

Components: Vulkan

Affects:

dEQP-VK.subgroups.*

Change-Id: Icab408e382a5c4d582786128bb88188c657dd9da

Merge vk-gl-cts/vulkan-cts-1.2.0 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: Ie62afeb1af0f212d51ae8ad9999078068587e19c

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.2.0

Change-Id: Iab64c9ee8090eadf98c8383996d53ff8c9f488f1

Merge vk-gl-cts/vulkan-cts-1.1.6 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I14c19183e4f63e28bcb091aaef1073009355909d

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.1.6

Change-Id: Iaf4e658989b8e022896e313159dc14fe59e25b3d

Merge vk-gl-cts/vulkan-cts-1.1.4 into vk-gl-cts/vulkan-cts-1.1.5

Change-Id: Ifb9683901fedc005a2b4e0d4173b402747dad99c

Merge vk-gl-cts/vulkan-cts-1.1.3 into vk-gl-cts/vulkan-cts-1.1.4

Change-Id: Ie4f2ff790ce6428d29182a61c44ef3706291d9df

Enable unused AllocationCallbackRecorder

computeDeviceMemorySystemMemFootprint used in Vulkan memory mapping and
allocation tests creates an AllocationCallbackRecorder but didn't pass
the callbacks to allocateMemory, passing zero records to
validateAllocationCallbacks.

Affects:
dEQP-VK.memory.mapping.suballocation.random.*
dEQP-VK.memory.allocation.random.*

Components: Vulkan

VK-GL-CTS issue: 2434

Change-Id: I208ea3e5a449c9a69b2981067ad38ea3e54699c2

Fix multiview dependency access mask for depth stencil attachments

Vulkan spec states that in case if depth stencil attachment storeOp
and/or stencilStoreOp is set to VK_ATTACHMENT_STORE_OP_STORE in its
description, corresponding access type should be write permissive.

This change sets subpass depencency source access mask to
VK_ACCESS_DEPTH_STENCIL_ATTACHMENT_WRITE_BIT in case if DS
attachment is present.

Components: Vulkan

Affects: dEQP-VK.multiview.*

VK-GL-CTS Issue: 2428

Change-Id: Ie34a6d93a8bdfad163150483b2fb8feb780ce0db

Always dynamically lookup egl functions

Android's latest NDK includes support for EGL 1.5 functions.
However, only for API level 29. But due to unified headers
folks building for older API levels will also see the new
functions but won't be able to link. Making the egl functions
dynamic avoids build time issues.

Change-Id: I718e28c7422faf3c7b964ca70b86a1711a08ae54
Affects: EGL
Bug: 1873
(cherry picked from commit 9af8b5c1ad98342a634b79a1f9c1213632248c64)

Add missing writes to PointSize in subgroups tests

When shaderTessellationAndGeometryPointSize is enabled the PointSize
builtin is undefined unless it is written to in the last vertex
processing phase so propagate the write through all vertex stages.

Affects:
dEQP-VK.subgroups.*

Components: Vulkan

VK-GL-CTS issue: 2404

Change-Id: Id8ffe8401c0a69d1e13847d168bafcec7e3c7a52
(cherry picked from commit 75f9abc189d4bef82d9496f55284c382b3b7a43f)

Add missing barriers from device  to host read access

Add barriers from various device write accesses to
host read access just before validating device written-data
on host.

Additionally correct one host memory op before validating data:
invalidate should be used instead of flush.

Components: Vulkan

VK-GL-CTS Issue: 2429

Change-Id: I09c845e7f306eadb012bf33cf49e1a9cfa8e2e0c
Affects: dEQP-VK.conditional_rendering.*

Fixes advanced blending tests support check

Tests were only running with
advancedBlendAllOperations supported

Components: Vulkan
VK-GL-CTS Issue: 2406

Affects:
dEQP-VK.pipeline.blend_operation_advanced.*

Change-Id: I387f1fd724bb14f6a34e784e663680abc178d5bb

Merge vk-gl-cts/vulkan-cts-1.1.6 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I90e38666d58a9b1e0af157dc5dbaa0516c681b56

Merge vk-gl-cts/vulkan-cts-1.2.0 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I04bbf72e6cc5268c82d352aa2ba638d66445457c

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.1.6

Change-Id: Ie554eb9583f468a3bd85375fefd8549da51b9389

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.2.0

Change-Id: Iddeb0a37d7c73f0539c4723040b59d22418f8f05

Fixes timeline semaphore test queues

Components: Vulkan
VK-GL-CTS Issue: 2416

Affects:
dEQP-VK.synchronization.timeline_semaphore

Change-Id: I3c101c869267c7bb747c53193195108df5996e70

Merge vk-gl-cts/vulkan-cts-1.1.4 into vk-gl-cts/vulkan-cts-1.1.5

Change-Id: I997b74efec17bfa3b4f70e7c3cd99e34ca726d30

Merge vk-gl-cts/vulkan-cts-1.1.3 into vk-gl-cts/vulkan-cts-1.1.4

Change-Id: I19c871d2823652543622ebe6d11cf1da6b72fb76

Fix image layouts in graphics spirv_assembly tests

This fixes regression from 789c733eef56b.

After above change, the command buffer recording and submission
is looped. However, image layouts are not properly handled at
the beginning and end of each loop
- transitioning image from UNDEFINED layout at the beginning
   of each loop, corrupting the data from previous iterations.
- transitioning the image to TRANSFER_SRC layout at the end of
   each loop, making it incompatible with render pass in next
   iteration.

This change moves the barriers only to the first and last
iteration of the loop.

Components: Vulkan

VK-GL-CTS Issue: 2405

Affects: dEQP-VK.spirv_assembly.*

Change-Id: I7f5d8493f11b05eea8569090e6da2ff768579a09
(cherry picked from commit 17e9601b282c94f4cc5a9614a300f536684dc70f)

Fix maxMipMap calculation for images

This change uses the floor of the log2 instead of the ceiling as
described in vulkan specification for normal images.

Vulkan spec section: 11.3.3. Image Miplevel Sizing

Components: Vulkan

Affects:
dEQP-VK.api.info.image_format_properties.*

VK-GL-CTS Issue: 2407

Change-Id: Ic5c7afe5890ebf1b5cc00abd1534a4a4d0c9c3c9
(cherry picked from commit 3d81feb1d686a02a92ad0d1e8bd519869c2d700a)

Fix image layouts in graphics spirv_assembly tests

This fixes regression from 789c733eef56b.

After above change, the command buffer recording and submission
is looped. However, image layouts are not properly handled at
the beginning and end of each loop
- transitioning image from UNDEFINED layout at the beginning
   of each loop, corrupting the data from previous iterations.
- transitioning the image to TRANSFER_SRC layout at the end of
   each loop, making it incompatible with render pass in next
   iteration.

This change moves the barriers only to the first and last
iteration of the loop.

Components: Vulkan

VK-GL-CTS Issue: 2405

Affects: dEQP-VK.spirv_assembly.*

Change-Id: I7f5d8493f11b05eea8569090e6da2ff768579a09

Use float16 precision for testing mediump

The precision requirements for mediump are not in the spec, they're just
for the conformance test, so it makes sense to use the simplest thing.
Allow the same error margins for mediump's 16-bit as for true 16-bit
floats. This slightly increases the precision requirements on mediump
exp but loosens them for atan.

Note, however, that the requirements for fp16 are still stricter than
for mediump because of the FloatFormat involved. fp16 requires support
for a wider exponent range than mediump.

Components: Vulkan
Affects: dEQP-VK.glsl.builtin.precision.exp*.mediump.*
         dEQP-VK.glsl.builtin.precision.atan*.mediump.*

Change-Id: I600a53984ba6a4f624bbb1525636f392d74ae96f
(cherry picked from commit b7dea11e6a395552a1423009c33d5518d9a0b2ec)

Clear unused image aspects in buffer to depth/stencil copy tests

Some of the buffer to depth/stencil tests used a combined
depth/stencil format but copied only one of those aspects.
The result checking still always compared both aspects. Now
the aspects not being copied are cleared for both result
and reference images to ignore these values.

VK-GL-CTS Issue: 2394

Affects:

dEQP-VK.api.copy_and_blit.core.buffer_to_depthstencil.*

Components: Vulkan
Change-Id: Ie5eba010d4f5f6d484d4635e275492da197eef79

Split spirv_assembly tests workload

The dEQP-VK.spirv_assembly.* tests should allow the render workload to
be split into multiple submissions if required to prevent GPU timeouts
encountered in some tests.

An additional flag is added to allow specific test groups to request
that the test workload is split.

Affects: dEQP-VK.spirv_assembly.*

VK-GL-CTS Issue: 2294

Components: Vulkan

Change-Id: I9295f675c06b3b966d89afae98334ebb5c9530c8
(cherry picked from commit 789c733eef56bd443692d751431177eb4fb2519f)

Add missing barrier before host access in draw tests

There is a missing barrier between in utilities for reading images,
between device write access and host read access.

Components: Vulkan

VK-GL-CTS Issue: 2401

Affects: dEQP-VK.draw.*

Change-Id: I7a8ec444b790c4afbf80f798222912398dfb818e

Merge vk-gl-cts/vulkan-cts-1.2.0 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I4c275dc2b38e9e722afd8a79227535bd59d3cb5e

Allow not flusing denorms in operands

This fixes tests for denorm value flush to zero in Ceil(), Log(),
 Log2(), Sqrt(), Sqrt2() operations. The test was assuming, that
operands of denormal value are flushed before the operation, however
the specification does not require that:

"If the entry point is declared with the DenormFlushToZero execution mode then for the
affected instuctions the denormalized result must be flushed to zero and the denormalized
operands **may** be flushed to zero."

VK-GL-CTS Issue: 2242

Affects: dEQP-VK.spirv_assembly.instruction.compute.float_controls.*.input_args.*_denorm_flush_to_zero

Components: Vulkan

Change-Id: Iedeb0d18fde3749a83be3262dba5efabf365aea2
(cherry picked from commit 0f323f593e9989190f0b8a5768aa25b594d7c1be)

Merge vk-gl-cts/vulkan-cts-1.2.0 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I8faea9ae44404bb48cd0cacc20f10cf8441f3b59

Merge vk-gl-cts/vulkan-cts-1.1.6 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I888b86ca0ad11da7329e4aace426274c44439d9e

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.2.0

Change-Id: Iee8c3acc0f091ebb4b8a32c300ebddbcaa4fc73f

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.1.6

Change-Id: I1aaa8a159b94d1396c0d53377d1adc98ce684e2b

Merge vk-gl-cts/vulkan-cts-1.1.4 into vk-gl-cts/vulkan-cts-1.1.5

Change-Id: Ia8efca269d913a5aa28a847596348fd6ed1e1571

Merge vk-gl-cts/vulkan-cts-1.1.3 into vk-gl-cts/vulkan-cts-1.1.4

Change-Id: I5f4f177d073c81c46c3dc2996f5909353e5cd4ff

Merge vk-gl-cts/vulkan-cts-1.1.5 into vk-gl-cts/vulkan-cts-1.1.6

Change-Id: I824fd73bbca4f8632d0fa7f982a8aa9e8c9294a5

Merge vk-gl-cts/vulkan-cts-1.1.4 into vk-gl-cts/vulkan-cts-1.1.5

Change-Id: I2d3f55a79614417544130cd255cd32da9e5aa423

Stop 10bit formats requiring GL_SRGB encoding

The EGL_KHR_gl_colorspace spec does not require formats that
are not SRGB8 to respect the given EGL_GL_COLORSPACE_KHR.

Components: EGL

VK-GL-CTS issue: 2288

Affects:

dEQP-EGL.functional.wide_color.window_1010102_colorspace_srgb
dEQP-EGL.functional.wide_color.pbuffer_1010102_colorspace_srgb

Change-Id: I41ab6c6a0467a29cf7e5e764bf2b2cfb78c51db3

Merge vk-gl-cts/vulkan-cts-1.2.0 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: I6cf7fcf5c0fab0a28a09b7fd25fcaa89b02e7973

Do not use 2D array d/s attachment created from 3D texture

Spec says, on VkFramebufferCreateInfo
"""
If flags does not include VK_FRAMEBUFFER_CREATE_IMAGELESS_BIT, each
element of pAttachments that is a 2D or 2D array image view taken from a 3D
image must not be a depth/stencil format
"""

This change changes image type to be 2D array, instead of 3D

Components: Vulkan

VK-GL-CTS Issue: 2379

Affects: dEQP-VK.geometry.layered.*.readback

Change-Id: Icd957c6badb17a0fcef82015c1ab2fba0896923b

Fixes clipping tests supported check

Adjacent line modes need to have check
that geometry shaders are supported

Component: Vulkan
VK-GL-CTS Issue: 2383

Affects:
dEQP-VK.clipping.clip_volume.depth_clamp.*

Change-Id: Ifcdc533ab0e9d54e834a7ac8fb609dec7c6ac248

Merge vk-gl-cts/vulkan-cts-1.2.0 into vk-gl-cts/vulkan-cts-1.2.1

Change-Id: If852bb48bdfbf7cc5168d68c6f296fc98bdb7f3c