|
d896fab8
|
2023-11-09T15:03:05
|
|
Vulkan: Fix texture self-copy
A new layout is introduced to support self-copy.
Bug: angleproject:7289
Change-Id: Ib914c433d55b9a79cfeb7a91f8a2b8680824d473
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5018797
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
|
|
|
4c500944
|
2023-11-08T16:16:23
|
|
Translator: Mark most builtins as not having side effect
This helps the translator eliminate stray uses of these builtins.
Bug: angleproject:6061
Change-Id: Ic0e72bddad5730fb1f906a4c8e53191ab3318d29
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5012988
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
2155534b
|
2023-11-06T11:49:20
|
|
Don't set dirty bits for attribs that are out of range.
PrivateState::setAllDirtyBits sets all bits in mDirtyCurrentValues.
When the context has fewer max attibutes than MAX_VERTEX_ATTRIBS, this
can cause out-of-bounds access to
PrivateState::mVertexAttribCurrentValues if the dirty bits are iterated
over without range validation.
Bug: chromium:1496378
Change-Id: I65481c432263a6e353a9361bba741b97dc5e20b1
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5008034
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
1bb2c5ec
|
2023-11-07T15:44:41
|
|
support reclient in capture_replay_tests
This also uses autoninja to start and stop reclient automatically.
This will be used with https://crrev.com/c/4797373.
Bug: angleproject:8309
Change-Id: I2e17c87c8ff0e253f16201297ef979337bf30292
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4796170
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Auto-Submit: Takuto Ikuta <tikuta@chromium.org>
|
|
9a7e8b7b
|
2023-11-07T17:02:02
|
|
Perf tests: log hwmon temps on linux
Syslog on a failed bot indicates possible overheating.
Log all hwmon temps which include CPU temps but also some other system
temps (disk probably not included). We might need to self-throttle if
crashes are correlated with higher temps.
Bug: b/296921272
Change-Id: I25f8dd899dd25880fc1567ecc18e102759fb868f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5008089
Auto-Submit: Roman Lavrov <romanl@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
383df961
|
2023-11-06T22:00:52
|
|
Metal: only convert vertex if offset is not multiple of 4.
Previously we always convert vertex attribute if its binding offset
is not multiple of the attribute's size. This requirement seems to be
unnecessary.
This CL removes that requirement so the only requirement left for offset
is that it must be multiple of 4.
Bug: chromium:1496807
Change-Id: I35c421951c7817b77bd0c006ed4b72cd04b5a8d7
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5006359
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Quyen Le <lehoangquyen@chromium.org>
|
|
32f209b8
|
2023-11-03T09:57:00
|
|
Vulkan: Fix __samplerExternal2DY2YEXTtnd combined with swizzle
Pixel camera app is hitting assertion due to swizzle applied to swizzle.
This CL adds ReswizzleYUVOpsTraverser::visitSwizzle() method and detects
there is a swizzle on YUV sampler and apply YUV swizzle first and then
original swizzle and then fold two swizzle into one.
This CL added a test to reproduce the bug.
This CL also makes ShCompileOptions in ShaderExtensionTest class a class
member instead of local variable so that subclass can modify for testing
or debugging purpose.
Bug: b/309480316
Change-Id: I72353fbad8dcacd77ca17a9f44e84485f4656a34
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5001614
Reviewed-by: Chris Forbes <chrisforbes@google.com>
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
34e5620b
|
2023-11-01T17:52:34
|
|
Metal: Reuse UBO conversion buffer only if size matches
BufferMtl::getUniformConversionBuffer will recycle an existing buffer if
the request has the same buffer index and offset that is a multiple of
existing block size. This disregards the GL std size of the request
leading to incorrect allocation of blocks in the conversion buffer. The
fix is to create a new conversion buffer when stdSize !=
buffer.uniformBufferBlockSize.
Bug: angleproject:8393
Change-Id: Ife87309e66f65baa7f636b734a5586d0ae3b9768
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4995450
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Quyen Le <lehoangquyen@chromium.org>
|
|
|
ae3af8ee
|
2023-11-03T16:20:12
|
|
Manual roll VK-GL-CTS from 7d738783bf28 to b127977ddad4 (29 revisions)
https://chromium.googlesource.com/external/github.com/KhronosGroup/VK-GL-CTS.git/+log/7d738783bf28..b127977ddad4
2023-11-03 gleese@broadcom.com Simplifications for float_controls tests
2023-11-03 javed@igalia.com Add tests for verifying VK_FORMAT_UNDEFINED
as input format.
2023-11-03 rgarcia@igalia.com Remove ESO tests which do not actually use
ESO
2023-11-03 rgarcia@igalia.com Add missing mandatory features #ifdef
2023-11-03 tzlatinski@nvidia.com Fix the verification of multi-layered
decode images
2023-11-03 gleese@broadcom.com Remove unused code from integer function
tests
2023-11-03 lorenzo@khronosgroup.org Merge vk-gl-cts/vulkan-cts-1.3.7
into vk-gl-cts/main
2023-10-30 lorenzo@khronosgroup.org Update external dependencies to the
latest version
2023-10-27 lorenzo@khronosgroup.org Merge remote-tracking branch
'vk-gl-cts/dev/VK_EXT_nested_command_buffer'
2023-10-27 lorenzo@khronosgroup.org Merge vk-gl-cts/opengl-es-cts-3.2.9
into vk-gl-cts/main
2023-10-27 rgarcia@igalia.com Fix warning when building deqp-vksc
2023-10-26 gleese@broadcom.com Fix offsets in float controls tests
2023-10-26 lorenzo@khronosgroup.org Merge vk-gl-cts/vulkan-cts-1.3.7
into vk-gl-cts/main
2023-10-20 Tyler.Schneider@amd.com Ray Query multiple queries tests
cases: dEQP-VK.ray_query.multiple_ray_queries.*
2023-10-20 rgarcia@igalia.com Test multiple geometry query stats
simultaneously
2023-10-20 lorenzo@khronosgroup.org Merge vk-gl-cts/vulkan-cts-1.3.7
into vk-gl-cts/main
2023-10-20 lorenzo@khronosgroup.org Merge vk-gl-cts/vulkan-cts-1.3.7
into vk-gl-cts/main
2023-10-19 mateusz.bahyrycz@mobica.com Tests for depth stencil buffer
resolve
2023-10-19 gleese@broadcom.com Add coverage of compute workgroup sizing
2023-10-19 gleese@broadcom.com Simplifying changes to spirv_assembly
test structure
2023-10-19 rgarcia@igalia.com Remove redundant ESO and feedback loop
layout tests
2023-10-19 mateusz.bahyrycz@mobica.com Tests for sparse resources on
transfer queue
2023-10-19 amber@igalia.com Add tests for inherithed framebuffers with
VK_EXT_depth_bias_control.
2023-10-19 cturner@igalia.com Remove the custom device code in the
copies and blitting tests
2023-10-19 piotr.byszewski@mobica.com Improve compile times of some
generated glsl tests
2023-10-19 gleese@broadcom.com Simplify BorderSwizzle TestNode
2023-10-18 ziga@lunarg.com Fix duplicated files in cmakelists.txt
2023-10-18 cturner@igalia.com Fix an unused variable warning in release
mode
2023-10-16 lorenzo@khronosgroup.org Merge "Merge branch
'dev/VK_ANDROID_external_format_resolve'" into main
If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/vk-gl-cts-angle-autoroll
Please CC angle-team@google.com,syoussefi@google.com on the revert to
ensure that a human
is aware of the problem.
To file a bug in ANGLE:
https://bugs.chromium.org/p/angleproject/issues/entry
To report a problem with the AutoRoller itself, please file a bug:
https://issues.skia.org/issues/new?component=1389291&template=1850622
Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+doc/main/autoroll/README.md
Bug: angleproject:8402
Change-Id: I99ac9c068080abf59f31e0f391da85cc0eb130b2
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5002011
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
f441e3ee
|
2023-11-01T15:22:55
|
|
Metal: Fix missing mix(float,float,bool) overload
ANGLE has an ANGLE_mix_bool overload when the last parameter is a bool
but it is not needed when the other parameters are scalar.
Bug: chromium:1493243
Change-Id: I6ea904ed63055b49cf3eaded475eafa607ca5976
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4995278
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
59f7467a
|
2023-11-01T14:53:42
|
|
Metal: Add missing ScalarMatrix ops
Translations for the following built-ins were missing:
scalar - matrix
scalar + matrix
scalar / matrix
Bug: angleproject:7525
Change-Id: I5da85781829c195906cc689ffee55cb681f0337c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4995668
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
67222ef4
|
2023-10-20T22:56:09
|
|
Reland "Vulkan: merge client buffer data."
This is a reland of commit 563569acfcaf56ea87916d2ab5d50f09c8e0094e
Original change's description:
> Vulkan: merge client buffer data.
>
> some old fashion game still use lots of interleaved
> client buffer. instead of copy each attrib alone.
> this cl try to merge all the attrib ranges. reduce
> allocated memory and do whole range memcpy.
>
> Bug: b/306763053
> Change-Id: I493d7f0e1ef593fb7059c36ae0ed2149c4595e42
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4960642
> Commit-Queue: Hailin Zhang <hailinzhang@google.com>
> Reviewed-by: Charlie Lao <cclao@google.com>
Bug: b/306763053
Change-Id: If079ab055b7b7a2d14235bee3311fd628f657f35
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4997325
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Hailin Zhang <hailinzhang@google.com>
|
|
221117ab
|
2023-11-02T08:48:51
|
|
Tests: Add The Sims Mobile trace
Test: angle_trace_tests --gtest_filter=TraceTest.the_sims_mobile
Bug: b/308927628
Change-Id: I5de44b7eedb904c318168fbc7e308af218dc5ab1
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5002145
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Commit-Queue: Cody Northrop <cnorthrop@google.com>
|
|
|
4cc9e0af
|
2023-10-31T11:19:51
|
|
Capture Tests: Fix unintentional gen-on-bind usage
A handful of tests are binding handles without glGen'ing them first.
This causes mismatched handles in capture_replay_test.py now that replay
inserts additional gen calls for resources that aren't explicitly gen'd.
For handles that are intentionally gen-on-bind, clean up the GL state so
it doesn't miscompare.
Test: capture_replay_tests.py --gtest_filter=*/ES3_Vulkan_SwiftShader
Bug: b/303100333
Change-Id: I5ce8060ff62264b3d648722b3c4542a136d95db5
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4994414
Reviewed-by: Roman Lavrov <romanl@google.com>
Commit-Queue: Mike Schuchardt <mikes@lunarg.com>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
|
|
918028a2
|
2023-10-31T17:53:07
|
|
Update bot detection in capture_replay_tests
Since GOMA is going away, check username instead.
Bug: angleproject:6085, angleproject:8309
Change-Id: Ie4fa4ec7fad8b52450c1ef8be9e5099f30c8fc7c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4994667
Reviewed-by: Roman Lavrov <romanl@google.com>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
|
|
|
9a5d75de
|
2023-10-30T11:59:19
|
|
Vulkan: Fix incompatible redefinition of cube faces
The TextureVk::mRedefinedLevels bitmask tracked which levels are
incompatibly redefined, greatly reducing the complexity of dealing with
GL's mutable textures.
It did not however take into account the fact that GL allows each
cubemap face to be separately redefined (unlike 2D arrays, where all
layers are defined together). This change turns the bitmask into an
array of bitmasks. Previously, a single bit represented whether the
level is incompatibly redefined. Now, elements of the array track the
same information for each cube face. For non-cube-map textures, only
element 0 is used.
Bug: chromium:1494664
Change-Id: I69568d3da2391796bf5f01505861fee42c6c8924
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4986289
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
c0c20673
|
2023-10-31T09:03:16
|
|
Remove expections for some AHB tests which are now fixed
Bug: b/223456677
Change-Id: Ibdc5b529e249c39c241a8e8e2b78ac10ff3d5ad2
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4988292
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Charlie Lao <cclao@google.com>
|
|
b6607672
|
2023-10-30T00:00:00
|
|
Test sampling from a texture with renderable levels below base
Bug: angleproject:8378
Change-Id: Id847849e1920c0d1706a4b3441c907ea2451cb6e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4994791
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
36f41997
|
2023-10-30T15:14:09
|
|
D3D11: CopySubTexture: Handle sRGB source with D3D11
Use typeless formats for sRGB textures that can be reinterpreted as
linear. Add a parameter to SRV creation to request samplers that
do not do sRGB for copying sRGB textures.
Bug: angleproject:7907
Change-Id: I3d94dfa5b25a1c2c531cd9bef4247097aa4bfb61
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4986290
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
|
|
|
118a9b14
|
2023-10-30T00:00:00
|
|
Test sampling from a texture with extra renderable levels
Bug: angleproject:8378
Change-Id: I4790d59ce193a2aa4ba9e468debf0933a6e3fe47
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4987639
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
c0f94afb
|
2023-10-30T11:07:02
|
|
Vulkan: Limit ImageTestES3.RGBXAHBImportMultipleLayers to 1 layer
ImageTestES3.RGBXAHBImportMultipleLayers is testing 3 layer AHB with CPU
write. It assumes alignment between each layer is 4096 which appears not
always true in ARM driver. I am also not seeing any API that would tell
us the underlying layer alignment. So for now restrict this test to
single layer.
Bug: b/223456677
Change-Id: I447990d08d31c2a20e8765686834f43b81560840
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4990044
Reviewed-by: Chris Forbes <chrisforbes@google.com>
Commit-Queue: Charlie Lao <cclao@google.com>
|
|
|
5fa65ec8
|
2023-10-27T16:12:46
|
|
ImageTest: check support before test cubemap
ImageTestES31.SourceAHBCubeArrayTargetCubeArray crahes because AHB
creation failed. We should check kAHBUsageGPUCubeMap support and bail
out if not supported.
Bug: b/223456677
Change-Id: I0d66830ac892e673bc39cca8caf5e25822957775
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4985412
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Chris Forbes <chrisforbes@google.com>
|
|
e519da9a
|
2023-10-29T14:33:06
|
|
Tests: Add Streets of Rage 4 trace
Test: angle_trace_tests --gtest_filter="*streets_of_rage_4*"
Bug: b/308334509
Change-Id: I3cc2d41c83bfcc6a32e6d1aec5e306cefb15afdc
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4988132
Commit-Queue: Cody Northrop <cnorthrop@google.com>
Reviewed-by: Mike Schuchardt <mikes@lunarg.com>
|
|
|
bcd331f1
|
2023-10-27T15:21:35
|
|
ImageTest: Allow writeAHBData to fail
ImageTest::SourceNativeClientBufferTargetRenderbuffer test is using EGL
to create a AHB. eglCreateNativeClientBufferANDROID does not allow us to
specify CPU access. So driver may allocate AHB in a way that prohibits
CPU access. That will cause AHB lock to fail and writeAHBData crash.
This is within AHB specification. This CL allows this to fail so that we
wont crash.
Bug: b/223456677
Change-Id: I549b940d1283165cec1dc48ea5b43ba917081252
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4985627
Reviewed-by: Chris Forbes <chrisforbes@google.com>
Commit-Queue: Charlie Lao <cclao@google.com>
|
|
|
53c5b145
|
2023-10-10T09:42:08
|
|
Capture/Replay: Restore texture bindings on reset
Store active texture unit and all bound textures at capture start. Track
dirty state for each unit+target pair during capture. Restore bound
textures for each dirty binding unit when resetting context state. Also
restore active texture unit if necessary.
Test: Capture The Sims Mobile
Bug: angleproject:4599
Change-Id: Iae3e03e6c5b0a35d5bb95bf07397d7894355502b
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4927869
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Commit-Queue: Mike Schuchardt <mikes@lunarg.com>
|
|
|
3b555e8d
|
2023-10-26T00:00:00
|
|
GL: Add emulateClipOrigin workaround
Older Adreno drivers apply GL_CLIP_ORIGIN_EXT
to the framebuffer instead of the clip space,
thus causing various side-effects.
Used a uniform to emulate the correct behavior.
Drive-by:
* Fixed a typo in ClipControlTest.OriginFrontFacing
Fixed: angleproject:8392
Change-Id: I2f7145977ab0e11dc88e8dbfec2cd32c4c31f830
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4987326
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
95fe7e47
|
2023-10-27T14:07:52
|
|
[Android] Update NDK version check for hardware buffer.
The function used for checking if hardware buffer is supported requires
NDK 29, but is currently checking if 26+ is in use.
This was not an issue before, but when updating the minimum SDK (and
NDK) version in Chromium, this led to an inconsistency.
For now, increasing the version requirement to 29+, and therefore also
removing the unnecessary branch for lock planes support.
Bug: chromium:1497004
Change-Id: I3d6e60e51db0831ee81d817aec90760d7680cc31
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4985624
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Kenneth Russell <kbr@chromium.org>
|
|
|
a7882c41
|
2023-10-27T09:32:07
|
|
Skip InstancingTestES3.LargeDivisor on Linux ASAN
Tests a large divisor and hits an emulation path on SwiftShader, causing
flaky timeouts on asan bots.
Bug: angleproject:7900
Change-Id: Ia82941e67867816c8330472f4f2ea8d897fa9636
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4982910
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Auto-Submit: Roman Lavrov <romanl@google.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Commit-Queue: Roman Lavrov <romanl@google.com>
Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
|
|
fdf4c6a6
|
2023-10-24T12:09:51
|
|
Add test for attrib rebinding from ARRAY_BUFFER to client.
Failure was found in Dragonmania game, where on switching from
ARRAY_BUFFER to client attrib pointer without updating the stride,
ANGLE assumed no change in the pipeline.
Bug: b/306472834
Change-Id: I30f7a478ed62f4356317156b50558166cb7c4c01
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4977168
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Charlie Lao <cclao@google.com>
|
|
|
729e19ac
|
2023-10-24T15:31:17
|
|
Remove essl3 extension requirement for ES3 texture parameters
The extensions are not clear about support for the swizzle parameters
with external textures. Other parameters are mentioned with specific
validation (which is handled in ValidateTexParameterBase). Simply
let the other validation handle these parameters and allow the
swizzle to be forwarded.
Bug: chromium:1493448
Change-Id: Ic34d4c124c5493fef796322b5237541e557fd7ab
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4973157
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
|
|
|
13776ac3
|
2023-10-24T11:10:25
|
|
Compile with optimize_max instead of default_optimization
Android builds default to aggressive optimization for size (-Oz) which
results in subpar performance in some spots.
AFAICT the effect of this change on cflags is:
-Oz -enable-ml-inliner=release
to
-O2
Note that -enable-ml-inliner=release has a more significant impact on
the binary size than -Oz/-O2:
With it: -Oz 5.7M, -O2 6.15M
Without it: -Oz: 6.3M, -O2 7.1M
So this change goes 5.7M -> 7.1M (+1.4M). That's in standalone ANGLE
builds. Unsure if Chrome builds might be different.
Skipping a few dEQP tests where this switch seems to uncover UB.
Bug: b/306391262
Bug: b/307584642
Change-Id: I44d0f3bcf25de0c36a4228895cad19ea38502cbf
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4968419
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Commit-Queue: Roman Lavrov <romanl@google.com>
|
|
|
562d4be3
|
2023-10-20T14:33:05
|
|
Clear pending program linking in Context::onDestroy
When destroying Context, ANGLE resets any internal state before
releasing allocated objects, such as Programs and Shaders. When
destroying a program, any pending program linking is resolved via
Program::resolveLink. This results in trying to access the Context state
that’s just been reset.
To work around this, we ensure there are no pending links before
resetting the Context state.
Bug: angleproject:8380
Change-Id: I599c7e5a82b0e6bf14df1263006e10d8a9cb0b38
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4957715
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
|
|
c22f46b5
|
2023-10-23T11:52:20
|
|
Update rgb-565-no-depth-no-stencil test suite names
Replace the "-" with "_" in the test suite names
to be consistent with other test suite name formats.
Bug: b/297901032
Change-Id: Idbfc1bc7aa7cef4382e3cdd3e223db28f414f15e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4968792
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
|
|
|
65f6c2ea
|
2023-10-22T18:07:55
|
|
Metal: always implicitly resolve MSAA render buffers on tiled GPUs.
On tile based GPUs, implicitly resolving MSAA render buffers to
single-sampled is preferred. Because it would save bandwidth by avoiding
storing the MSAA textures to memory. Furthermore resolving as
StoreAction is almost a free operation on these GPUs.
Traditional desktop GPUs almost always store MSAA textures to memory
anyway, so this feature would have no benefit besides adding additional
resolve step as well as memory overhead of the hidden single-sampled
textures.
Bug: chromium:1486094
Change-Id: I5eb3b1314560024dd5c0834b0c0b43a6b4d3d51b
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4962114
Commit-Queue: Quyen Le <lehoangquyen@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
4b59e4c6
|
2023-10-17T17:47:20
|
|
Add a test to check glReadPixels format and type
In GLES 3.2 spec
https://registry.khronos.org/OpenGL/specs/es/3.2/es_spec_3.2.pdf,
section 16.1.2, it says the following glReadPixels type
and format should work:
The second is an implementation-chosen format from
those defined in table 8.2... The values of format and
type for this format may be determined by calling
GetIntegerv with the symbolic constants
IMPLEMENTATION_COLOR_READ_FORMAT and
IMPLEMENTATION_COLOR_READ_TYPE, respectively.
The implementation-chosen format may vary depending
on the format of the selected read buffer of the
currently bound read framebuffer.
Add a test to check that when reading from the default
framebuffer, using the format and type retrieved from
IMPLEMENTATION_COLOR_READ_FORMAT and
IMPLEMENTATION_COLOR_READ_TYPE works.
Bug: b/306256081
Change-Id: I1e94087aac8c803bcc5db3515ed56291349daf28
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4950844
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
|
|
ba65feb4
|
2023-10-18T17:33:38
|
|
Vulkan: Limit mutable texture flush to one update
In case there are many updates for a mutable texture, flushing
it preemptively can reduce performance, especially if it is done
repeatedly.
* Added getLevelUpdateCount() to ImageHelper.
* Previous mutable textures will now be flushed only if they have
exactly one update per mip level/cubemap face (if defined).
* This means that mutable textures with no data will also not be
flushed.
* Added unit tests for single-level texture flushing and situations
with no updates or more than one update.
Bug: b/285613719
Change-Id: I1592ecf502051a55ebfbb7fcd22577c9ce87bf43
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4953847
Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
49fd8eda
|
2023-10-20T14:23:38
|
|
android_helper: improve _RemovePrefix logging
Random stuff gets in there sometimes, such as:
"cmd: Failure calling service package: Broken pipe (32)"
RemovePrefix just raises that it doesn't start with "package:" so the
issue doesn't show up in logs.
Bug: b/296921272
Change-Id: Ifae4e3c0c7717514401bf1109febf8e70fff6527
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4959473
Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
Reviewed-by: Roman Lavrov <romanl@google.com>
Auto-Submit: Roman Lavrov <romanl@google.com>
Commit-Queue: Roman Lavrov <romanl@google.com>
|
|
|
cfc40d2b
|
2023-07-19T17:47:13
|
|
Vulkan: Adjust clear color precision for GL_RGB5_A1 FBO
When clear color has more precision than the framebuffer
format can hold, dithering is automatically applied on
some hardware.
This issue causes below dEQP tests to fail when the FBO
color attachment format is RGB5_A1:
KHR-GLES31.core.draw_buffers_indexed.color_masks
KHR-GLES32.core.draw_buffers_indexed.color_masks
Adjust the clear color precision for RGB5_A1 format to
workaround the issue.
We can remove this workaround once the vulkan driver
fixes the auto-dithering problem.
Bug: b/292282210
Change-Id: Ic3ffebd2d20c8782612619a60d1ec2cc6d613c22
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4937472
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Roman Lavrov <romanl@google.com>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
|
|
|
36439d9c
|
2023-10-12T00:00:00
|
|
Metal: Fix PBO uploads of PVRTC1 textures
Do not override row pitch values for PVRTC1
textures during internal backend operations.
Instead, set bytesPerRow to 0 before calling
replaceRegion just to pass Metal Validation.
Enforce CPU path for PBO uploads until
GPU block linearization is implemented.
Bug: angleproject:8376
Change-Id: I80b9ec69364d55cd9f51005f4d1a180e05e71c2c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4949350
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
3b28bca0
|
2023-10-11T09:49:54
|
|
Allow using RGBA IOSurfaces with GL_RGBA format for EAGL backend
As https://crrev.com/c/4877334 switched RGBA IOSurfaces to GL_RGBA
to work with ANGLE/Metal, some tests(e.g. web tests) on iOS failed
to create an IOSurface because ANGLE'S EGL backend still still
binds IOSurfaces as GL_BGRA_EXT even though we expect the IOSurface
to really be RGBA. To fix it, this CL allows using RGBA IOSurfaces
with GL_RGBA format by adding a GL_RGBA entry to kIOSurfaceFormats.
See also the comment[1] for better explanation.
[1] https://crrev.com/c/4918590/comments/1e44e971_50c3d780
Bug: chromium:1489586
Change-Id: Ifa215318401edb892fd5527c35c7d30871e82a83
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4930309
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Sunny Sachanandani <sunnyps@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
1ea49a22
|
2023-10-13T11:28:41
|
|
Move uniform dirty bits to ProgramExecutable
Rather than try to funnel them through Program and ProgramPipeline to
the executable in the backend, just move them to ProgramExecutable in
the front end.
This fixes Dota Underlords at the same time due to not needing to set
the Program dirty to propagate bits.
Test: Dota Underlords
Test: ProgramPipelineTest31.ProgramPipelineBindBufferRange
Bug: b/299532942
Change-Id: Ic73c45608e22f89ca400ebf684f8cd287ed2f43a
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4922969
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Cody Northrop <cnorthrop@google.com>
|
|
3892ac14
|
2023-10-12T13:35:33
|
|
Do not flush normal float constants to zero.
It's ok to flush denormalized constants to zero.
It's not ok to flush perfectly valid normal float constants >= FLT_MIN
to zero.
Two problems:
1) Values when parsed as doubles with a value less than FLT_MIN are
being flushed to zero. This is incorrect when the comparison is done
in double, since some values below FLT_MIN in double are equal to
FLT_MIN when cast to float. The fix is to perform the comparison in
float.
2) Values with a decimal exponent less than FLT_MIN_10_EXP are being
flushed to zero. FLT_MIN_10_EXP is -37 but FLT_MIN is 1.1754943E-38.
10^-37 may be the "minimum negative integer such that 10 raised to
that power is a normalized float", but being constrained to powers of
ten it's above FLT_MIN (which is 2^-126). Since this comparison is
done before #1 above, it's only present (AFAIK) to ensure that the
exponent will not make the pow() function overflow. Comparing against
-38 (FLT_MIN_10_EXP - 1) instead will do the trick.
Bug: angleproject:8373, dawn:2077
Change-Id: I1ddf410c2caa9f0d1ba3529ace693dcd326a2cb3
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4936714
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Stephen White <senorblanco@chromium.org>
|
|
|
c54c2f81
|
2023-10-12T00:00:00
|
|
Sync border color on format updates
When a texture is redefined, its border color should
be resynced because of format-specific adjustments.
Used memcmp for comparing ColorGeneric structs.
Bug: angleproject:7969
Change-Id: If0e3e0bc7761c26dab6d00ab24462c789f5b90bb
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4935866
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
|
1988108e
|
2023-10-12T00:00:00
|
|
Tighten draw buffers validation for dual-source blending
Use the front-end DRAW_BUFFER* state instead of the
internal mask, which is unset for missing attachments.
Bug: angleproject:1085
Bug: angleproject:7177
Change-Id: I5311d1f26043f15035e51957920bf9cd553c3007
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4935858
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
b870956e
|
2023-10-05T00:00:00
|
|
GL: Limit depth/stencil texture mode updates
The DEPTH_STENCIL_TEXTURE_MODE state must affect only
DEPTH_STENCIL textures (OpenGL ES 3.2, Section 8.16).
Some drivers do not follow this rule and exhibit various
side effects when this mode is set to STENCIL_INDEX for
textures of other formats.
Bug: angleproject:2373
Change-Id: I20a8168d858cf070762ff0b3b7111cdf30e8166d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4936374
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
|
43acadab
|
2023-10-06T15:05:36
|
|
Add missing configurations for khr tests
KHR-GLES* tests are ran with configurations defined in
glcKhronosMustpassEs.hpp:
https://github.com/KhronosGroup/VK-GL-CTS/blob/main/external/openglcts/modules/runner/glcKhronosMustpassEs.hpp,
when launching the conformance test from Khronos
Conformance Package. This change adds the missing
configurations in test_suites.pyl, for vulkan backends
and swiftshader backends.
Bug: b/297901032
Change-Id: I87e77ca8756b883ef08551919f67904ecf21aa97
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4918582
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
|
|
|
a0967770
|
2023-10-06T18:01:45
|
|
Add egl/wrapper headers to deqp opengl sources
gn gen --check catches these and "include not allowed"
Bug: b/303881910
Change-Id: I26728dc2618436daec232152c56d59a7233ca91e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4916365
Auto-Submit: Roman Lavrov <romanl@google.com>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
|
|
|
0afc3ec9
|
2023-10-05T00:00:00
|
|
Metal: Fix robust initialization of PVRTC1 textures
Replace Region Validation: rowBytes must be 0 for:
* MTLPixelFormatPVRTC_RGB_2BPP
* MTLPixelFormatPVRTC_RGB_2BPP_sRGB
* MTLPixelFormatPVRTC_RGB_4BPP
* MTLPixelFormatPVRTC_RGB_4BPP_sRGB
* MTLPixelFormatPVRTC_RGBA_2BPP
* MTLPixelFormatPVRTC_RGBA_2BPP_sRGB
* MTLPixelFormatPVRTC_RGBA_4BPP
* MTLPixelFormatPVRTC_RGBA_4BPP_sRGB
Fixed: angleproject:8370
Change-Id: I153bea5bb487f79b0966be0b8670a2ee4e3aeb39
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4915704
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
|
7e8907db
|
2023-10-06T15:41:05
|
|
Add egl/wrapper to deqp opengl includes
Needed to roll VK-GL-CTS upstream change:
https://issuetracker.google.com/303881910
Bug: b/303881910
Change-Id: Idffa299cc6c1d277702167bf50e1d944febb9e7e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4915605
Commit-Queue: Roman Lavrov <romanl@google.com>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Auto-Submit: Roman Lavrov <romanl@google.com>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
|
|
|
0015d1e2
|
2023-10-05T00:00:00
|
|
Allow ETC1_RGB8 for 2D array textures
When GL_EXT_compressed_ETC1_RGB8_sub_texture is enabled,
ETC1_RGB8 format must be accepted for 2D array textures.
Bug: angleproject:5731
Change-Id: Ic5e647ffd4e7d2a171e25ea8cf3f41514cf37072
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4915918
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
65e8ae30
|
2023-10-06T09:42:45
|
|
Vulkan: ANGLE_device_vulkan is incompatible w/ asyncCommandQueue
The application may use EGL_ANGLE_device_vulkan to retrieve ANGLE's
Vulkan queue for direct usage. There is no mechanism for the
application to synchronize with the internal thread that uses the same
queue when the asyncCommandQueue feature is enabled.
This was causing flakiness where VVL reported this error. This change
makes sure an ASSERT catches this incompatible usage.
Bug: angleproject:7204
Bug: b/299693666
Change-Id: Idf349f99d80b7bbddeb9a0b2d72e3dd7c707fa5c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4915594
Reviewed-by: Roman Lavrov <romanl@google.com>
Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
d6bf8e0d
|
2023-10-02T18:10:36
|
|
Improve capture_replay_tests (win-trace) logging
* Get rid of the CRASH expectation as it is always confusing. Replace
items in expectations with SKIP_FOR_CAPTURE. This will no longer run
the corresponding tests (which was making logs even more confusing)
* Use CaptureFailed and ReplayFailed instead of Crashed status (the
actual meaning is a non-zero return code from the test harness)
* Introduce batch names by indexing them as they are created:
batch_000, batch_001, ...
This enables easy searching in logs.
* When logging a batch (result) include its name and why it is logged
* Failed batches are printed with full output instead of "abbreviated"
output. One issue with b/303065176 was that an important part
of the output (errors) was truncated by this "abbreviation"
* Remove "regression log" and "Summary" as these are often just
confusing: regression log just duplicates information printed earlier
and summary may not be clear w.r.t test expectations. The part that
prints what was unexpected is still there. It also now includes
batch names so it's much more obvious what comes from where. Example:
Unexpected 'CaptureFailed' (8):
!= Pass: batch_000 FramebufferTest_ES3.RenderAndInvalidateImmutableTextureWithBellowBaseLevelLOD/ES3_Vulkan_SwiftShader
!= Pass: batch_000 FramebufferTest_ES3.RenderAndInvalidateImmutableTextureWithBellowBaseLevelLOD/ES3_Vulkan_SwiftShader_AsyncCommandQueue
!= Pass: batch_000 FramebufferTest_ES3.RenderAndInvalidateImmutableTextureWithBellowBaseLevelLOD/ES3_Vulkan_SwiftShader_EnableParallelCompileAndLink
!= Pass: batch_000 FramebufferTest_ES3.RenderAndInvalidateImmutableTextureWithSubImageWithBeyondMaxLevel/ES3_Vulkan_SwiftShader
!= Pass: batch_000 FramebufferTest_ES3.RenderAndInvalidateImmutableTextureWithSubImageWithBeyondMaxLevel/ES3_Vulkan_SwiftShader_AsyncCommandQueue
!= Pass: batch_000 FramebufferTest_ES3.RenderAndInvalidateImmutableTextureWithSubImageWithBeyondMaxLevel/ES3_Vulkan_SwiftShader_EnableParallelCompileAndLink
!= Pass: batch_000 FramebufferTest_ES3.RenderImmutableTextureWithSubImageWithBeyondMaxLevel/ES3_Vulkan_SwiftShader
!= Pass: batch_000 FramebufferTest_ES3.RenderImmutableTextureWithSubImageWithBeyondMaxLevel/ES3_Vulkan_SwiftShader_AsyncCommandQueue
Unexpected 'FailedToTrace' (1):
!= Pass: batch_001 FramebufferTest_ES3.RenderSampleDepthTextureWithExcludedLevel/ES3_Vulkan_SwiftShader
Bug: angleproject:7519
Bug: b/303065176
Change-Id: Ie0c4734c7c4ac965aa6fd38b0c1783e0b92af7a5
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4908349
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Roman Lavrov <romanl@google.com>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
|
|
|
8c341cfd
|
2023-10-04T12:49:59
|
|
Vulkan: Fix blits involving 3D images
The layer vs depth value involved with 3D images when calling
vkCmdBlitImage is fixed in this change.
However, that brought to light that the combination of
VUID-vkCmdBlitImage-srcImage-00240 and
VUID-vkCmdBlitImage-dstImage-00252 make it impossible to blit between 3D
and 2D array images, which is likely a spec oversight.
This change makes 3D<->2DArray blits fall back to draw-based blit. This
in turn exposed the fact that 3D images as src were not handled in
BlitResolve.frag. A new Blit3DSrc.frag shader is added which shares
code with BlitResolve.frag to implement this. This is a separate shader
to avoid creating unnecessary and invalid combinations of shaders.
VK_EXT_image_2d_view_of_3d could have been used to avoid this new
shader, but that is not ubiquitous.
Bug: angleproject:7291
Bug: dawn:1962
Change-Id: I6a96162f95829304b4731d43208d9d054f538105
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4911800
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
|
|
|
856a6ec0
|
2023-10-03T17:17:35
|
|
Enable rgb-565-no-depth-no-stencil tests on Android Bots
Enable gles3 and gles31 rgb-565-no-depth-no-stencil tests
on Android Bots.
Khronos conformance package ran the gles3 and gles31 tests with
rgb-565-no-depth-no-stencil configurations on Android devices:
https://github.com/KhronosGroup/VK-GL-CTS/blob/main/external/openglcts/modules/runner/glcAospMustpassEs.hpp#L41
Enable the rgb-565-no-depth-no-stencil configuration
tests on Android bots to cover them.
Non-Android platforms do not support rgb-565-no-depth-no-stencil
configuration. We should not enable it on non-Android platforms.
This CL also disables this configuration from
building on non-android platforms in src/tests/BUILD.gn.
Bug: b/297901032
Change-Id: I23d83213818884ea679dd76bf4b001c3b636dffc
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4908171
Reviewed-by: Roman Lavrov <romanl@google.com>
|
|
|
ffb32380
|
2023-10-02T16:01:37
|
|
Vulkan: Fix device creation on multi-queue devices
When multiple eligible queue families are exposed by the device, ANGLE
defers devices creation until a surface is made current, at which point
it selects a queue family that supports presentation to that surface.
This path was largely untested and was broken:
- Some post-device-creation initialization was not deferred until the
device was actually created
- Some cap calculation depended on the chosen queue family index
- Query of device capabilities was done too late, such that ANGLE could
not correctly determine the level of support for ES versions.
Bug: angleproject:8300
Change-Id: I2a35396d1fd08ed26d217dff07d10e9a9c1ac55f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4907895
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Roman Lavrov <romanl@google.com>
|
|
|
1eccf863
|
2023-10-03T16:18:07
|
|
Handle non-debuggable com.android.angle.test on device
Noticed in capture_tests on CI, and angle_perftests also runs into this.
When a non-debuggable test apk is installed on the device the hash check
fails as run-as does not work for non-debuggable apps.
It's just an optimization to avoid re-installing the apk though, so just
warn about the error and re-install the apk.
Bug: b/303276423
Change-Id: Iecc68de285454043fc7f090163af373a22e11e41
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4910152
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Commit-Queue: Roman Lavrov <romanl@google.com>
|
|
|
7f5143c2
|
2023-10-02T15:38:15
|
|
Vulkan: Notify VAO when VBO's mBufferWithUserSize changed.
When buffer robust access is enabled, and bufferData is called with
different size and we end up reusing the underline storage, we will have
to recreate VkBuffer with user's size, and driver is relying on
VkBuffer's size to implement robust access. The bug here is that we
notify VAO when storage changes. But when storage is reused and we have
dedicated VkBufer with user size and that VkBuffer changed, we were not
notifying the VAO. This CL adds that notification so that VAO gets
notified and dirty bits processed and its cache of VkBuffer gets updated
Bug: chromium:1488055
Bug: b/303138134
Change-Id: Ie693c92c2edde9a22a41a25f5bde493397550d95
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4906568
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
8b9500ae
|
2023-09-28T16:15:21
|
|
Enable multisample deqp tests on bots
Enable gles3 and gles31 deqp multisample tests on bots.
Khronos conformance package ran the tests in gles3 and
gles31 mustpass list with multisample config rgba8888d24s8ms4:
https://github.com/KhronosGroup/VK-GL-CTS/blob/main/external/openglcts/modules/runner/glcAospMustpassEs.hpp#L24
Enable the multisample config tests on bots cover the tests.
This CL only enables the multisample tests on Vulkan
backends (including swiftshader) on Android, Linux, Windows,
as we are prioritizing vulkan backend development.
Bug: b/297901032
Change-Id: Iaa830bd466e3564bb1e83c98cc360bd5a85abf86
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4903430
Reviewed-by: Roman Lavrov <romanl@google.com>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
|
|
|
e9719f5c
|
2023-10-02T12:15:35
|
|
Capture/Replay: Disable VK_EXT_host_image_copy during capture
Bug: b/303065176
Change-Id: I87f9dd11cd87115f097bb35d44f024db19274d51
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4903077
Reviewed-by: Roman Lavrov <romanl@google.com>
Commit-Queue: Roman Lavrov <romanl@google.com>
Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
4ace4da1
|
2023-09-28T00:00:00
|
|
Metal: Add 2D array and 3D texture type mappings
Do not try to copy texture levels with mismatching
layer counts to avoid triggering Metal validation.
Fixed: angleproject:8364
Change-Id: I4bc72be753a62dd78dc131562c62e14eb540ab2a
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4901993
Reviewed-by: Quyen Le <lehoangquyen@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
|
|
|
961ed7a2
|
2023-10-02T15:07:44
|
|
RepeatedEglInitAndTerminate: 100 -> 50 repetitions
One of the variants
ES3_1_Vulkan_SwiftShader_EnableParallelCompileAndLink
is taking >60s which seems a bit too much time for this kind of test
Bug: b/303089709
Change-Id: I75584bf376cfac8e4ddd61b21ccd75b9ac6d2967
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4908348
Commit-Queue: Roman Lavrov <romanl@google.com>
Auto-Submit: Roman Lavrov <romanl@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
32ef34df
|
2023-10-02T10:30:48
|
|
Remove the dEQP-EGL.functional.color_clears*/render* test suppression
The dEQP-EGL.functional.color_clears* and
dEQP-EGL.functional.render* test failures on Pixel6 have been
fixed in https://gerrit.khronos.org/c/vk-gl-cts/+/12272.
Remove the test suppression from the expectation files.
Bug: b/295394803
Bug: angleproject:8214
Change-Id: I2d99f9ed282eba43dc9c18e4427f5abd5014224e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4908191
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
|
|
|
1cab871c
|
2023-09-27T22:11:25
|
|
GL: Fix ScalarizeVecAndMatConstructorArgs vs precision
If the constructor being expanded has no precision (such as if it's
entirely made up of constants and bools), the workaround didn't produce
a precision for the helper function, leading to GLES compilation errors.
Bug: chromium:1420130
Bug: angleproject:8361
Change-Id: I0de312dfd1bfbf45317db4fb88d685a843052817
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4898697
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
49c4d87a
|
2023-08-21T00:00:00
|
|
Apply fragment color broadcasting to secondary outputs
When GL_EXT_draw_buffers is enabled,
gl_FragColor is replaced with gl_FragData.
Apply the same transformation to gl_SecondaryFragColorEXT
if it is statically used to avoid compiler failures.
Bug: angleproject:1085
Change-Id: If154cd20a8e14a92da520847a9c4508971a45fef
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4900641
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
|
7f5b5f74
|
2023-09-19T07:35:16
|
|
Tests: Add Batman: The Telltale Series trace
Test: angle_trace_tests --gtest_filter="*batman_telltale*"
Bug: b/301024256
Change-Id: Ice5aa3ed093901c04486fca64121592eb8805f93
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4883428
Auto-Submit: Cody Northrop <cnorthrop@google.com>
Reviewed-by: Roman Lavrov <romanl@google.com>
Commit-Queue: Roman Lavrov <romanl@google.com>
|
|
6b0ddce0
|
2023-09-22T13:13:38
|
|
Backends need to handle invalid render area during a clear
1. Frontend no longer noops empty scissors during clear
2. FramebufferVk is updated to handle invalid render area by restaging
deferred clears in clearImpl(...) if render area is invalid
Bug: angleproject:8348
Tests: EGLSurfaceTest*WindowThenScissoredClear*
Change-Id: Iec51914a083a59bad7f939798c932dffada56a6c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4867641
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: mohan maiya <m.maiya@samsung.com>
|
|
|
9fc3baf5
|
2023-07-26T10:29:13
|
|
Add the missing GraphicsPipelineDesc legacy dither bit update
When the app calls glEnable(GL_DITHER) or glDisable(GL_DITHER),
we need to update the legacy dither bit of the renderpass that
belongs to ContextVk::mGraphicsPipelineDesc. If not, there is a
change that the graphics pipeline will be created with a
renderpass that has outdated legacy dither bit. This results
the dither being applied to the render results incorrectly:
e.g. the app calls glDisable(GL_DITHER),
but the render results have dithering applied.
Bug: b/286921997
Bug: b/292282210
Bug: b/293349058
Bug: b/284462263
Change-Id: Ie24b95898526c9021be6e3cb7620e4050f9faaaf
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4722446
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Roman Lavrov <romanl@google.com>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
|
|
|
e1d2e88a
|
2023-09-20T11:53:15
|
|
Check pending garbage after some buffer releases
* Embedded BufferHelper::releaseBufferAndDescriptorSetCache() inside
a new ContextVk method: releaseBufferAllocation()
* After releasing the buffer, there is a check for excess pending
garbage. If the tracked pending garbage size is larger than the
threshold, the context will be flushed.
* Unskipped the test "BufferDataInLoopManyTimes", which was failing
on Android devices.
Bug: b/280304441
Change-Id: Ib34319f3291dd2200fc1a92e30645f9d1da8e2b9
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4879086
Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
a1e3abc0
|
2023-09-22T10:14:43
|
|
Add ES2 support for copying GL_SRGB8_ALPHA8 to GL_RGBA8
Bug: angleproject:7907
Change-Id: I02ae9785ea1f942e85983c18155cfd75ec39b5c2
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4887271
Commit-Queue: ccameron chromium <ccameron@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
7b1b8a01
|
2023-09-21T20:15:29
|
|
Metal: Allow using GL_RGB with either RGBA/BGRA IOSurface.
Previously we only allowed GL_RGB internal format to be used with BGRA
IOSurface and GL_RGBX8_ANGLE to be used with RGBA IOSurface
respectively.
However, there are currently many places in Chrome prefer GL_RGB to be
able to be used with both RGBA and BGRA IOSurface.
This CL allows such combinations. Instead of deducing angle::FormatID of
the IOSurface pbuffer based on the input GL internal format & type, we
will take into account the pixel format of the IOSurface as well.
For example, when we call eglCreatePbufferFromClientBuffer with GL_RGB
internal format attribute:
- if IOSurface's pixel format is 'RGBA' ->
deduced angle::FormatID is R8G8B8A8_UNORM.
- if IOSurface's pixel format is 'BGRA' ->
deduced angle::FormatID is B8G8R8A8_UNORM.
This CL also removes GL_RGBX8_ANGLE support from Metal backend. Because
there are many places in Chrome that use this format enum for both RGBA
& BGRA IOSurface when the extension is available. It's redundant to
support that since GL_RGB already covers most of the required cases.
Bug: angleproject:8350
Change-Id: I5a121a97e031a42d0779721d4348f373dfaee9a0
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4881742
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Quyen Le <lehoangquyen@chromium.org>
Reviewed-by: Sunny Sachanandani <sunnyps@chromium.org>
|
|
|
120e90eb
|
2023-09-08T11:23:11
|
|
Add test for copying GL_SRGB8_ALPHA8 to GL_RGBA8
The pixel values should be the same (no conversion should happen).
Include sub-tests with and without alpha premultiplication.
Bug: angleproject:7907
Change-Id: I2e3d657d5b8513cd06df357e84027572ebce03ff
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4852258
Commit-Queue: ccameron chromium <ccameron@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
f5842350
|
2023-08-21T00:00:00
|
|
Vulkan: Flatten secondary fragment output arrays
Added a SPIR-V transformation that makes them private
and copies their values into regular fragment outputs.
Fixed: angleproject:8336
Change-Id: Ide62dd183426e0de70fd4ac6d2b8c6a4b4b33994
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4875327
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
|
bfae8072
|
2023-09-20T06:48:06
|
|
Metal: Support RGBA IOSurface pbuffer.
Currently only BGRA IOSurface is supported by ANGLE.
This CL adds RGBA IOSurface support to Metal backend.
- Added GL_RGBX8_ANGLE & GL_UNSIGNED_BYTE attributes combo to specify
that the IOSurface is RGBX. Only supported on Metal backend for now.
- Added GL_RGBA & GL_UNSIGNED_BYTE combo to specify that the IOSurface
is RGBA. Only supported on Metal backend for now.
Bug: angleproject:8350
Change-Id: I0e05762870c9c034bca78e8989aedf346406df57
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4874484
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Quyen Le <lehoangquyen@chromium.org>
|
|
|
ae9c7857
|
2023-09-19T13:40:52
|
|
Suppress BufferDataInLoopManyTimes for Linux/Intel
* Skipped BufferDataInLoopManyTimes for Linux/Intel due to the long
execution time, suspected to be causing errors on the bots.
Bug: angleproject:8351
Change-Id: Ib7b640800edf5b30e251b47cfabfd34cc1ecea28
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4877693
Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
|
|
ecf11eca
|
2023-09-19T15:18:04
|
|
Add ContextMutex unit tests
Test: angle_unittests --gtest_filter=ContextMutex*
Bug: angleproject:8226
Change-Id: I55b69ce8bb89d572a2bf5d28f428635505771fa1
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4874388
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Igor Nazarov <i.nazarov@samsung.com>
|
|
|
eb0d5997
|
2023-09-15T16:41:13
|
|
Move set/get uniform machinery to ProgramExecutable
This is done because some uniforms are internally added by the compiler
(draw ID, base vertex, and base instance) and are automatically set **on
the installed executable**.
This change fixes scenarios where a draw is done after a program has
failed a relink, and therefore is unable to correctly set the uniforms
(as it does not have access to the executable that is installed).
It also fixes draws that use those uniforms in a PPO.
Bug: angleproject:8297
Change-Id: Id74b4984b88aa09b5b81be1c91412d6c91711136
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4864693
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
a1f52f1b
|
2023-09-07T14:44:24
|
|
Vulkan: Flush pending image garbage more often
* Added a counter to the context object to keep track of the size of
the pending image garbage: mEstimatedPendingImageGarbageSize.
* Modified hasExcessPendingGarbage() to use the sum of the size of
the image and and suballocation garbage.
* RendererVk::calculatePendingGarbageSizeLimit() provides the limit.
* Currently the limit is based on the available heap sizes. It will
use a fraction of the largest memory heap size.
* The portion is currently kGarbageSizeLimitCoefficient = 0.2f.
* Unskipped the test "TextureDataInLoopManyTimes", which was failing
on Android devices.
Bug: b/280304441
Change-Id: Ibcced1d118ea8a1f347028b62d29cfbd9e38e8c0
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4851252
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
|
|
|
1b450b92
|
2023-09-15T11:07:25
|
|
Vulkan: Fix buffer storage reuse bug when robustAccess is enabled
There is an optimization in vulkan backend that when the bufferData is
called and current storage size is big enough for new bufferData call,
we just reuse the storage. Mean while, when hasRobustAccess() is true,
we must use the VkBuffer with the exact user size that glBufferData call
provides so that driver can set proper access boundary. In order to
satisfy both requirement, if robust resource access is enabled, we
create a separate VkBuffer with the exact user provided size but bind to
the same memory. There is a bug here that if robustAccess is true, this
buffer of user provided size is not been recreated when storage is
reused but with different user size (both has same allocation size).
This causes we keep using the smaller VkBuffer and subsequently causes
missing triangles. This CL clears mBufferWithUserSize when size changes
and storage is reused.
The other bug here is that previously we are checking
isRobustResourceInitEnabled, which is incorrect. We should check
hasRobustAccess. This appears works for chrome possibly due to both are
enabled. This CL switches it to check hasRobustAccess.
This CL also renames mBufferForVertexArray to mBufferWithUserSize to
reflect what its true meaning.
Bug: chromium:1476475
Change-Id: I843cc3a705f8a582a97bc0307f03aa1eb9fad3ff
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4864003
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
503c0db1
|
2023-09-12T18:01:23
|
|
Add the missing multisample and rgb565-no-depth-no-stencil tests.
We need different --deqp-gl-config-name values
for different test applications, to create RenderTarget
with different properties needed for different test
applications, such as multisample, rgb formats.
This CL adds an array of
vector<const char*> gTestSuiteConfigParameters
to store the --deqp-gl-config-name
values for different test applications. Each element
in this array can hold multiple const char* values.
In the upcoming CLs, we can add more customized config
values for each test applications, such as
--deqp-screen-rotation for angle_deqp_gles*rotate*tests.
Rename the gEGLConfigName to gEGLConfigNameFromCmdLine.
The default value for gEGLConfigNameFromCmdLine is empty
string, and its value can be overwritten by specifying
--deqp-gl-config-name when running the tests from command
line. If we pass --deqp-gl-config-name from command line,
then use the value passed from the command line to replace
the values in gTestSuiteConfigParameters.
Bug: b/297901032
Change-Id: Ibf8762c55d7d0cdf7bcf33f384568587023b369c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4851048
Reviewed-by: Roman Lavrov <romanl@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
|
|
|
f1345d1c
|
2023-08-02T14:23:59
|
|
Call ReleaseThread() entry point from ThreadCleanupCallback
Change relevant only for the Android platform.
Fixes problems:
1. Current code calls `thread->setCurrent(nullptr)` in
`Display::threadCleanup()` method, while not executing other code
that will actually unmake Context from current. This will affect
following EGL calls from the application's side in its own thread
cleanup callback. For example, calls to `eglMakeCurrent()` (to unmake
from current) or `eglReleaseThread()` will be no-op. This may lead to
memory leak if context was previously marked for destruction.
2. The `Display::threadCleanup()` did not also clears
the `CurrentValidContext`. This may cause crashes or other UB if an
application will call some GLES API in its thread termination
callback (for example `glFinish()`), if ANGLE already destroyed the
context in its ThreadCleanupCallback (use after delete).
3. Context will remain current to a thread that terminated. Current
implementation of GLES drivers on Mali/Adreno GPUs automatically
unmade context from current, so the application may make this context
current to some other thread. ANGLE should mimic this behavior to
avoid possible bugs.
All of the above problems may be fixed by simply calling from the
ThreadCleanupCallback either:
- MakeCurrent(t, d, EGL_NO_SURFACE, EGL_NO_SURFACE, EGL_NO_CONTEXT)
- ReleaseThread(t)
This CL choose to call `ReleaseThread()` because this is the API that
the application may call in its own thread cleanup callback. And if
application's callback will be called later than ANGLE's callback, then
application's `eglReleaseThread()` call we be a no-op.
Even if the application does not have its own thread cleanup callback
(like most of the applications), there is no harm to call
`ReleaseThread()` anyway. The only difference with `MakeCurrent()` is
that it will also cleanup invalid objects and some other memory
depending on the backend.
This CL naturally replaces existing logic with Display termination when
there is no active threads thanks to the previous CL:
"Perform Display terminate(InternalCleanup) from makeCurrent()".
Bug: angleproject:6723
Bug: angleproject:8283
Test: angle_end2end_tests --gtest_filter=EGLContextSharingTest.ThreadCleanupCallback*
Test: angle_end2end_tests --gtest_filter=EGLContextSharingTest.UnmakeFromCurrentOnThreadExit*
Change-Id: I8c5b31f34f91c4bfdac165ac30c213ff5fef992e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4742383
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Igor Nazarov <i.nazarov@samsung.com>
|
|
|
c38f3374
|
2023-09-15T19:22:44
|
|
Fix "win-asan-test" random failures and timeouts
While exact reason for the failures is unknown, it seems like disabling
Vulkan Loader DLL unloading fixes the problem. More details in the issue
8347.
This CL also sets the variable on the Linux platform, because it does
not hurt and will probably help with similar issues.
Bug: angleproject:8347
Change-Id: I398d55a7c13f1ce6790988983fdf8fee7cc5776c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4868220
Commit-Queue: Igor Nazarov <i.nazarov@samsung.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
22b6ca1c
|
2023-08-21T00:00:00
|
|
Remove obsolete EXT_blend_func_extended expectations
Bug: angleproject:5981
Bug: angleproject:6585
Bug: angleproject:7177
Change-Id: I25da401f03ff0bfa7a77c3ce7a3fecd0a119d6a4
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4870916
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
9d99f85f
|
2023-09-14T15:01:14
|
|
Add multithreaded texture upload test
Bug: angleproject:8341
Change-Id: I8561a64fb28681509a1d65c490888dfa1a423835
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4864467
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
|
|
|
b2e6a196
|
2023-09-11T15:27:20
|
|
Vulkan: Use VK_EXT_host_image_copy for texture uploads
Of all the scenarios where host image copy may be useful, this is likely
the most common case. There are numerous conditions for when the copy
may be done on the host:
- The image format must support it,
- It must be unused by the GPU,
- It must not have any pending updates (this can potentially be
mitigated if needed), and
- It must be in a host-copyable layout.
However, many texture uploads are done:
- To compressed formats, where support is highly likely,
- On init, where:
- the image is never previously used,
- the image has no previous uploads
- the image is in the UNDEFINED layout
which satisfies the conditions above.
As a result of this change, when the upload is done on the host,
creation of a temp buffer is avoided which greatly reduces memory
pressure (specially during app loading which is when most texture data
is uploaded) and may even improve performance (due to avoiding a double
copy).
Testing the first 3 frames of the following traces with a SwiftShader
implementation shows the amount of buffer allocated for staged uploads
changed as such:
- Black Desert: 185MB -> 65MB
- Genshin Impact: 125MB -> 12MB
- Asphalt 9: 138MB -> 0MB
Bug: angleproject:8341
Change-Id: Id71dcc4a7a0f8b67960d2d283fe9d19ce7429a03
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4856676
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
ffd6ec26
|
2023-09-07T15:39:19
|
|
Reland "Make egl surface uncurrent when being destroyed"
This relands commit 497440cdcb7d2ee59bca612dd07fc13cf09a6a57.
This is to workaround errors when app does below behaviors:
1) while there is a context still bound to the current
rendering thread and the surface, call eglDestroySurface()
2) create a new surface eglCreateWindowSurface()
3) call eglMakeCurrent() with the surface created in step 2)
4) does work on the new surface
The old surface won't be destroyed in step 1) because
it was still bound by the context of the current rendering
thread. When creating new surface on step 2), some hardware
will return error code EGL_BAD_ALLOC, because the old egl
surface is still associated with the native window.
To workaround, when destroying surface, if the surface
is still bound by the context of the current rendering
thread, release the context and surface by passing
EGL_NO_CONTEXT and EGL_NO_SURFACE to eglMakeCurrent().
The workaround is controlled by a frontend feature
uncurrentEglSurfaceUponSurfaceDestroy. This feature
is only enabled on vulkan backends.
Bug: b/292285899
Change-Id: Id4c47f1b20e0f90d1013a68893fd70e917c030e2
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4867066
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
|
|
|
b4852ef9
|
2023-02-08T14:18:06
|
|
Vulkan: Drop support for Vulkan 1.0
Bug: angleproject:7959
Change-Id: Ib673679ea1a503af22b37092dbff1ee1fd34fba6
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4233092
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Ian Elliott <ianelliott@google.com>
|
|
|
8a7ad933
|
2023-09-13T15:11:46
|
|
Revert "Make egl surface uncurrent when being destroyed"
This reverts commit 497440cdcb7d2ee59bca612dd07fc13cf09a6a57.
Reason for revert: this caused chromium webview tests failures: https://chromium-review.googlesource.com/c/chromium/src/+/4860891.
Original change's description:
> Make egl surface uncurrent when being destroyed
>
> This is to workaround errors when app does below behaviors:
>
> 1) while there is a context still bound to the current
> rendering thread and the surface, call eglDestroySurface()
> 2) create a new surface eglCreateWindowSurface()
> 3) call eglMakeCurrent() with the surface created in step 2)
> 4) does work on the new surface
>
> The old surface won't be destroyed in step 1) because
> it was still bound by the context of the current rendering
> thread. When creating new surface on step 2), some hardware
> will return error code EGL_BAD_ALLOC, because the old egl
> surface is still associated with the native window.
>
> To workaround, when destroying surface, if the surface
> is still bound by the context of the current rendering
> thread, release the context and surface by passing
> EGL_NO_CONTEXT and EGL_NO_SURFACE to eglMakeCurrent().
>
> The workaround is controlled by a frontend feature
> uncurrentEglSurfaceUponSurfaceDestroy. This feature
> is only enabled on vulkan and gl backends.
>
> Bug: b/292285899
> Change-Id: I872d2e116ba6860f58d1176f011a5ef7c5a5af4e
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4851255
> Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
> Reviewed-by: Charlie Lao <cclao@google.com>
> Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
> Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Bug: b/292285899
Change-Id: I760054d856294e6691e79e165fd73ce9e560621f
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4862958
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Auto-Submit: Yuxin Hu <yuxinhu@google.com>
Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
|
|
|
e7a77053
|
2023-09-12T13:40:47
|
|
Update VulkanMemoryTest after adding context flush
After adding context flushing to the memory allocation fallbacks
in a previous CL (https://crrev.com/c/4787949), some tests in the
VulkanMemoryTest suite no longer test the success path of the fallback
to wait for submitted commands to finish.
* Updated AllocateVMAImageAfterFreeing2DGarbageWhenDeviceOOM
* The test now uses an additional allocation and draw after the
first sync, followed by a second sync. During the allocation
of the last texture, it would wait to finish the submitted
command and free the garbage without flushing the whole context.
Bug: b/280304441
Change-Id: Idc583e120b8d9e1cc550f6cba4cb0b2b197e4944
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4860367
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
|
|
|
497440cd
|
2023-09-07T15:39:19
|
|
Make egl surface uncurrent when being destroyed
This is to workaround errors when app does below behaviors:
1) while there is a context still bound to the current
rendering thread and the surface, call eglDestroySurface()
2) create a new surface eglCreateWindowSurface()
3) call eglMakeCurrent() with the surface created in step 2)
4) does work on the new surface
The old surface won't be destroyed in step 1) because
it was still bound by the context of the current rendering
thread. When creating new surface on step 2), some hardware
will return error code EGL_BAD_ALLOC, because the old egl
surface is still associated with the native window.
To workaround, when destroying surface, if the surface
is still bound by the context of the current rendering
thread, release the context and surface by passing
EGL_NO_CONTEXT and EGL_NO_SURFACE to eglMakeCurrent().
The workaround is controlled by a frontend feature
uncurrentEglSurfaceUponSurfaceDestroy. This feature
is only enabled on vulkan and gl backends.
Bug: b/292285899
Change-Id: I872d2e116ba6860f58d1176f011a5ef7c5a5af4e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4851255
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
|
|
|
e7418836
|
2023-08-16T14:25:52
|
|
Vulkan: Add context flushing as OOM fallback
* As a new fallback for out-of-memory errors, if an allocation results
in device OOM, the context is flushed and the allocation is retried.
* Functions related to buffer/image allocations now return a VkResult
value instead of angle::Result, which will be bubbled up to a higher
level for safer handling.
* The OOM is no longer handled at the level where the allocation
happens, but is moved up to the context.
* Added two functions to ContextVk for allocating memory for images and
buffer suballocations, which also include the fallback options.
* initBufferAllocation(): Uses BufferHelper::initSuballocation()
* initImageAllocation(): Uses ImageHelper::initMemory()
* Moved initNonZeroMemory() out of the following functions:
* BufferHelper::initSuballocation()
* Moved to ContextVk::initBufferAllocation().
* ImageHelper::initMemory()
* Moved to ContextVk::initImageAllocation().
* Also moved to new function:
ImageHelper::initMemoryAndNonZeroFillIfNeeded().
This function replaced the rest of initMemory() usages outside
initImageAllocation().
* New macros for memory allocation
* VK_RESULT_TRY()
* If the output of the command inside it is not VK_SUCCESS, it will
return with the error result from the command.
* VK_RESULT_CHECK()
* If the output of the command inside it is not VK_SUCCESS, it will
return with the input error.
* Added a test in which allocation would fail due to too much pending
garbage without the fix on some platforms. The test ends once there
has been a submission.
* New suite: UniformBufferMemoryTest
* Added a similar test for flushing texture-related pending garbage.
* New suite: Texture2DMemoryTestES3
Bug: b/280304441
Change-Id: I60248ce39eae80b5a8ffe4723d8a1c5641087f23
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4787949
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
2fe7bbae
|
2023-09-07T13:21:29
|
|
Tests: Add Oxenfree trace
Test: angle_trace_tests --gtest_filter=TraceTest.oxenfree
Bug: b/298613720
Change-Id: Ie577dff75faf58ee97c0898a28766e9ed7c87459
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4851250
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Commit-Queue: Cody Northrop <cnorthrop@google.com>
|
|
|
ce263437
|
2023-09-11T12:25:28
|
|
Vulkan: Perform CPU wait in clientWait outside the global lock
Leverage UnlockedTailCall and move the CPU side wait
during a clientWait outside of the global mutex lock.
Bug: angleproject:8340
Tests: FenceSyncTest.BasicOperations*
Tests: EGLSyncTest.EglClientWaitSync*
Change-Id: I8c05e62e74cc64d38bf8797d28faaf49135e71fc
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4851649
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: mohan maiya <m.maiya@samsung.com>
Reviewed-by: Charlie Lao <cclao@google.com>
|
|
|
ee7f6ed5
|
2023-09-07T16:50:54
|
|
Refactor dEQP test classes
Rename dEQP to dEQPTest.
Move the static functions and variables that
are shared between all individual dEQP tests
in the same test suite into the dEQPTestSuiteStat
class.
This new dEQPTestSuiteStat class is responsible
for updating the test counts, succeeded tests,
and failed tests of a test suite. Add a static member
of type dEQPTestSuiteStat to the dEQPTest class,
so that all of the dEQPTest instances shared the
same instance of dEQPTestSuiteStat, and as each dEQPTest
TestBody() executes, it can update the data in the same
dEQPTestSuiteStat.
Each deqp test application will create multiple
dEQPTest instances based tests listed in one of the
mustpass list in gCaseListFiles array. And each
deqp test application will only create one test suite.
This is done through testing::RegisterTest(). The
testSuiteName passed to testing::RegisterTest()
for each dEQP test application remains
the same for all of the dEQP tests created.
This change also replaces the below legacy function
names with the new function names:
SetUpTestCase() --> SetUpTestSuite()
TearDownTestCase() --> TearDownTestSuite()
Bug: b/297901032
Change-Id: I9439d7ac314d70a9a5d9ad72c0a3f58cd11e559b
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4851047
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Reviewed-by: Roman Lavrov <romanl@google.com>
|
|
|
eb5acc3e
|
2023-09-08T15:34:44
|
|
Tests: Add ONE PIECE TREASURE CRUISE trace
Test: angle_trace_tests --gtest_filter="*one_piece_treasure_cruise*"
Bug: b/299681746
Change-Id: I3b5a9de18613df37181feb3c207105dba411d2a7
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4854449
Commit-Queue: Cody Northrop <cnorthrop@google.com>
Reviewed-by: Roman Lavrov <romanl@google.com>
|
|
|
1249fc1f
|
2023-09-07T14:39:26
|
|
Tests: Support capture_tests on Android and enable on CI
Enables angle_with_capture_by_default = true on Android bots
similarly to Linux and Windows.
Enables the capture test on Pixel 4 and 6 bots (it takes only a few
seconds to run)
Enables support for capture in android_helper in general.
For example, this now works the same way it would on say Linux, but
runs the capture on the device an pulls files to OUT_DIR.
% ANGLE_CAPTURE_LABEL=gacha_club ANGLE_CAPTURE_OUT_DIR=/tmp/capt2 ANGLE_CAPTURE_FRAME_START=16 ANGLE_CAPTURE_FRAME_END=16 out/AndroidPerformance/angle_trace_tests --gtest_filter=TraceTest.gacha_club --use-angle=vulkan --retrace-mode --max-steps-performed=20
For end2end tests it's a bit trickier as we're still using the Chromium
wrappers for e.g. out/AndroidPerformance/angle_end2end_tests - which is
why capture_tests.py is currently using angle_android_test_runner.py
Bug: b/296397568
Change-Id: I2606e74a17c90e381bb49ec8d555a6d24851b4f7
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4850889
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Commit-Queue: Roman Lavrov <romanl@google.com>
|
|
|
a412b149
|
2023-09-07T12:08:40
|
|
Suppress out of bounds on no robust contexts tests
dEQP-EGL.functional.robustness.reset_context.shaders.out_of_bounds_non_robust.reset_status.writes.uniform_block.vertex
failed occasionally due to VVL error.
The test first results in vulkan device lost,
which is an acceptable behavior. When we clean up the
vulkan resources, the VVL is complaning the resources
have not finished execution. However, VVL should check
if the vulkan device is already lost, as there is no way to
finish resource execution when device is lost.
Will remove the test from the skip list when the VVL
issue filed
https://github.com/KhronosGroup/Vulkan-ValidationLayers/issues/6310
is resolved.
Bug: b/289544394
Change-Id: Ideb676e6629cc17a4f8101ccc083a331a1d3c864
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4851049
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
|
|
|
cdbc45a9
|
2023-09-07T15:45:11
|
|
Tests: capture_tests: remove ANGLE_FEATURE_OVERRIDES_ENABLED
Unnecessary for this test, should only matter for tests involving
serialization etc.
Bug: b/296397568
Change-Id: I1b3e12b11358778cf23dd6d7a5aab69be41c1ab4
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4850548
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Commit-Queue: Roman Lavrov <romanl@google.com>
|
|
|
9c0748d3
|
2023-08-16T09:15:22
|
|
Capture/Replay: Trim inactive Texture resources
As is done with Shader and Program objects, trim inactive Texture
objects from default trace execution playback and update
CapturedTest with new use cases.
Bug: b/297031925
Change-Id: I734d8f1fa66a24be7790391deaeee24cf570d12a
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4802281
Reviewed-by: Roman Lavrov <romanl@google.com>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Commit-Queue: Mark Łobodziński <mark@lunarg.com>
|
|
|
06411d16
|
2023-09-06T13:23:34
|
|
GL: Use the executable instead of program
In a few places, the program was still being directly referenced instead
of the executable (in particular when dealing with multiview).
Bug: angleproject:8297
Change-Id: I15d0865bf58376a9f85efeec739dd93b49ceaea7
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4846475
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
7b0bb0f6
|
2023-09-01T13:52:28
|
|
Properly "install" program executables
According to GL:
- The program has an executable
- The executable is overwritten during link.
- After a failed link, queries of the executable may return
half-linked information
- On glUseProgram, the executable is installed in the context
- On glUseProgramStages, the executable is installed in the program
pipeline
- After a successful link, the executable is updated wherever the
previous executable of the program was installed.
This change implements exactly the above:
- The program's and the program pipeline's executables are now
shared_ptr. References to an executable in the context and PPO are
also through a shared_ptr. Installing an executable thus translates
to sharing the executable.
- The context and PPOs are made to not reference the program directly,
but work solely through the executable. As a result, the program is
free to create a new executable for link.
With this change, the link job will be free to modify the executable as
necessary because that will not be accessed until the link is done.
Note that previous changes made the backend executable accessed through
the frontend one, and moved all link results to the frontend and backend
executables as appropriate.
Bug: angleproject:6358
Bug: angleproject:8297
Change-Id: Ie636b23ff7420ad284d18b525ec4f5fb559dd9d1
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4823089
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
|