|
9b1ff7b1
|
2023-05-25T14:13:13
|
|
Translator: Fix SPIR-V translation without "variables" flag
This flag is now assumed to always be true for the SPIR-V (and MSL)
translators. They use the collected variables during translation.
Bug: chromium:1447314
Change-Id: I69d31ce8144aec0bd2a7263bad6e5a366e922740
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4567830
Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
3f633521
|
2023-05-18T21:27:00
|
|
Vulkan: Fix handling inactive unnamed I/O blocks in SPIR-V
When an unnamed I/O block already exists, later inactive unnamed I/O
blocks were not being added to the list of inactive varyings, causing
the SPIR-V transformer to trip up
Bug: b/283017896
Change-Id: Ida5b38e1a03da2b336d5904caa44f53dbdfb8aa3
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4544590
Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Charlie Lao <cclao@google.com>
|
|
336bfd60
|
2023-03-30T00:16:40
|
|
GL: Fix the RewriteRepeatedAssignToSwizzled workaround
This transformation splits chained assignments where the LHS is
assigning to a swizzle. However, instead of retaining the assignment
operator (like *=, += etc), it was accidentally unconditionally using
`=`.
Bug: chromium:1394345
Change-Id: Icaef159e1fb57e720e08d4b4f6603f08d9cd2bfe
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4381648
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
|
|
57dd0e5f
|
2023-03-27T13:05:36
|
|
Translator: Initialize out arguments too for WebGL
Local variables were initialized for WebGL, but not `out` arguments.
However, `out` arguments completely overwrite the passed-in variable at
call site, so they should also be zeroed out.
Bug: chromium:1425685
Change-Id: Ib2f61a962eea74c4933aafa4d5ad377189cec578
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4375137
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
1ee27fcd
|
2023-03-24T00:00:00
|
|
Fix translation of noperspective interpolation qualifier
Auxiliary storage qualifiers used with 'noperspective'
interpolation must be retained during ESSL translation
* Added new qualifiers:
* EvqNoPerspectiveCentroid
* EvqNoPerspectiveCentroidIn
* EvqNoPerspectiveCentroidOut
* EvqNoPerspectiveSample
* EvqNoPerspectiveSampleIn
* EvqNoPerspectiveSampleOut
* Adjusted GLSL and SPIR-V output
* Adjusted 'sample' rank to be the same as 'centroid'
Fixed: angleproject:4388
Change-Id: Iffc80a7a3b057beffd3bcbbc9446eeed77183c99
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4368471
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
afc1e5dc
|
2023-03-27T00:00:00
|
|
GL: Support OES_sample_variables
Bug: angleproject:8097
Change-Id: I983d6e83fe9ca942c91cbdf7274b46306e38f11f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4368476
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
f5695244
|
2023-03-21T00:00:00
|
|
Fix sample qualifier validation
* `smooth sample` is the same as `sample`
* `flat sample` is the same as `flat`
Bug: angleproject:3589
Change-Id: I8615fa73f4b0cea729b8e09d86f1b8183669de2b
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4356666
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
4cbe8548
|
2023-02-28T21:12:17
|
|
Expose shader extensions based on ESSL version
Previously, all suported extensions were always exposed in the
preprocessor. This broke some games which relied on ESSL1-only extension
macros not being defined in ESSL3 shaders. This change adds min/max
version information to list of extensions so the preprocessor can
conditionally expose extensions based on the shader language version,
both via the extension name macros and the #extension directive.
Test: angle_unittests --gtest_filter="VersionTest.*"
Test: angle_end2end_tests --gtest_filter="*ESSL*ExtensionMacros*"
Test: Run com.gameloft.android.ANMP.GloftGGHM on Pixel 6
Bug: b/268091452
Change-Id: I2332a6cb964f54c47d23e2ef6b24e99a0b5c8202
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4304907
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Mike Schuchardt <mikes@lunarg.com>
|
|
facd07ec
|
2023-02-14T14:11:00
|
|
Vulkan: Prune all statements without side effect
Bug: chromium:1417461
Bug: angleproject:6061
Change-Id: I6e2b74589021ee60c99448e43ee3eb7f06796ede
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4250278
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
9d84ad7d
|
2023-02-14T10:47:54
|
|
Translator: Fix sampler-in-struct when struct is unused
Standalone symbols as statements are also pruned. For example in the
below:
```
x;
g();
```
The line that only contains `x;` is pruned. In particular, this was
causing a bug with sampler-in-struct processing where the code ASSERTs
that if an instance of the struct is visited during traversal, that
it's already processed as part of an index chain (like s.member), but
in the above situation that is not true.
Bug: chromium:1416152
Change-Id: I49e9a5826fc4d8c1fef1fa6f1d38c65f33f8acfd
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4247584
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
|
|
70b5620e
|
2023-01-10T17:16:41
|
|
Metal: skip out variable rewrite for array element references.
ANGLE_elem_ref already returns a reference which is incompatible with
the ANGLE_out and ANGLE_inout builtins. Skip the rewrite for this
builtin, similarly to swizzle_ref for which it was already skipped.
Add a small regression test which catches this bug.
Fixed: angleproject:7939
Change-Id: Iff8ead697220466ceb063ea2f8eebdb7d9e5896d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4152866
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Auto-Submit: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
78e28cb8
|
2022-11-03T11:05:48
|
|
Vulkan: Remove glslang fallback
ANGLE's SPIR-V generator is battle tested by now.
Bug: angleproject:6210
Change-Id: I2ed88642e3ede435e3f906fbd701133be6873fb7
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4002868
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
|
|
03eccad2
|
2022-10-18T10:39:53
|
|
Tests: Add reduced crashing shader test
Reduced case of a shader that crashes an app during SPIR-V compile.
Test: GLSLTest_ES3.NestedPowFromUniform
Bug: b/217922015
Change-Id: I8d0ccd8b2c4f4e8e51ac39c9dbfaafbf48423157
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3965051
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Cody Northrop <cnorthrop@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
70c434b2
|
2022-08-29T17:09:23
|
|
Check max shader storage blocks support in test
If the shader source code uses shader storage
buffer objects, we need to check the corresponding
GL_MAX_*_SHADER_STORAGE_BLOCKS is at least of size 1.
Bug: angleproject:7583
Change-Id: I7f0e537e4dcfca8414cccb0088be8d34735633d1
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3863111
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
ac2aeccd
|
2022-08-25T17:43:43
|
|
Check GL_MAX_VERTEX_SHADER_STORAGE_BLOCKS in test
If the vertex shader uses shader storage buffer objects,
we need to check GL_MAX_VERTEX_SHADER_STORAGE_BLOCKS
is at least of size 1.
Bug: angleproject:7185
Change-Id: I98d83f9f0cc30e92fdcaee15541fdde96652b38e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3858144
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
955adb77
|
2022-08-12T10:14:48
|
|
Cache compiled shader
By storing the compiled shader in the blob cache, the time to
recompile the same shader is reduced.
Based on work by <hckim.kim@samsung.com>
Bug: angleproject:7036
Change-Id: I884ae40e715c49a9ccd12903012e8327811e3557
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3808235
Commit-Queue: Cody Northrop <cnorthrop@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
|
|
58392791
|
2022-04-07T16:14:12
|
|
Translator: Fix crashes with monomorphizing loops
Monomorphize functions crashes when trying to deep-copy a while loop,
and a continue statement. Null check these values before performing
a copy, as this is still valid AST syntax.
Bug: angleproject:7190
Change-Id: I822c0aa062ed844f86aa2b603899d73cbd48255e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3573079
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Kyle Piddington <kpiddington@apple.com>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
|
|
aef9603f
|
2022-08-02T18:30:08
|
|
Separate Struct declarations earlier
Metal translator backend asserts with compound struct definitions.
Move SeparateCompoundStructDeclarations before
SeparateCompoundExpressions. This change does revert some changes
for angleproject:6386, but the tests fixed there continue to pass.
Bug: angleproject:6489
Change-Id: I6adc606ac0b14453eb5e3e90f6501bf02c2d2768
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3806616
Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Reviewed-by: Gregg Tavares <gman@chromium.org>
Commit-Queue: Kyle Piddington <kpiddington@apple.com>
|
|
6f80f0f0
|
2022-08-06T02:29:19
|
|
Translator: Clean up the compile flag passing interface
Historically, compile flags were sent to the translator as a bitmask.
Recently, we were getting close to running out of bits. Additionally,
direct-to-metal work had started to introduce constants to be passed to
the translator, which were misplaced in ShBuiltInResources and Caps.
Recent work on Pixel Local Storage adds even more constants, aggravating
the situation.
In this change, the interface to passing compile flags is reworked. A
struct is passed (instead of a bitmask) that has one bit for each flag.
This can be indefinitely extended. Additionally, the constants needed
by metal and PLS are also placed in this struct. In turn, the backends
can set these options directly, and don't have to hack them into Caps to
further get hacked into ShBuiltInResources.
Bug: angleproject:7559
Change-Id: If93f1e1b8818ad3a0ac708ab04ab93b4b397d114
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3812562
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
Commit-Queue: Amirali Abdolrashidi <abdolrashidi@google.com>
Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
8474a95e
|
2022-08-05T22:15:04
|
|
Translator: Don't validate structs on non-root nodes
Struct validation needs to know about structs that may have been defined
outside the scope of the node, so this validation (similar to variable
and function call validation) is disabled when validation a subtree of
the AST.
Bug: angleproject:7531
Bug: b/239207278
Change-Id: Id6dcee041a40f09f0c36a9ef3bfc238433b69c58
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3812047
Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Yuxin Hu <yuxinhu@google.com>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
|
|
4e046256
|
2022-07-25T17:17:28
|
|
Translator: Fix deep copy of loops and branches
Fixed nullptr dereference when the loop or branch doesn't have an
expression.
Bug: angleproject:7518
Change-Id: Ib888634e409dcbc27ef060bb46017e8911097eef
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3785294
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
3baa867c
|
2022-07-14T23:08:42
|
|
Vulkan: Fix image arguments passed to functions in GLSL
It's impossible to provide the format qualifier to an image function
parameter in GLSL, because `layout()` is not acceptable there. SPIR-V
doesn't accept formatless image parameters without the
StorageImageReadWithoutFormat and StorageImageWriteWithoutFormat
capabilities, which are not universally available.
Instead, this change monomorphizes functions with image parameters.
Note that this was already previously done for R32F images for emulation
purposes.
Bug: angleproject:7484
Change-Id: Id287c4a93b32deb3fd8be982c32fa58fd2c848d8
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3763074
Reviewed-by: Amirali Abdolrashidi <abdolrashidi@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
eaf77ef3
|
2022-07-07T10:25:16
|
|
Metal fails to compile uniform blocks in ternary statements
When creating an intermediate type for type expansion, Metal was copying
over information about which uniform block a type appeared in. This led
to a later pass, RewriteNamelessUniformBlocks, failing.
This patch helps address a compile failure in the Godot engine.
Bug: angleproject:7487
Change-Id: If0724f93017e41260ff11e59a5c28e8578563891
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3751106
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Kyle Piddington <kpiddington@apple.com>
Reviewed-by: Gregg Tavares <gman@chromium.org>
|
|
3e5de4ac
|
2022-07-05T10:22:54
|
|
Add a test that passes image2D handles as function args
Bug: angleproject:7484
Change-Id: I3565bd42deeefa22566d4556a25ec9778c6f8550
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3747098
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
785353fd
|
2022-05-24T12:40:16
|
|
Support Desktop OpenGL context creation in end2end tests
Validation of Desktop GL versions and profile masks is unimplemented.
Bug: angleproject:7360
Change-Id: Ifae94215b6aada895c2b02318a1d05c9515e9b96
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3664916
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
|
|
89e38b57
|
2022-06-22T15:04:08
|
|
Refactor to use ANGLETest vs ANGLETestWithParam
Bug: angleproject:6747
Change-Id: I72ad52d0268eae0e1a401f12f3e94cc5efa402f2
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3719002
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Cody Northrop <cnorthrop@google.com>
|
|
79185677
|
2022-06-22T16:22:35
|
|
Translator: Fix validation of struct with same-name struct field
In the following:
```
struct S
{
S s;
};
```
The `S` in the field refers to a previous declaration of `S` in an outer
scope. The validation code mistakenly attempted to ensure it refers to
the same `S` being declared.
Bug: chromium:1337628
Change-Id: I86184be63bc59e608fe5b24369cd4c40b3bb3b31
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3716965
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Auto-Submit: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
d9a2027f
|
2022-05-19T16:52:28
|
|
Add a test for volatile aliases of the same shader image
Bug: angleproject:3032
Bug: angleproject:7343
Bug: angleproject:7398
Change-Id: I6ea6abea0b9dafd93bc297499df23be27904c996
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3656314
Reviewed-by: Stephen White <senorblanco@chromium.org>
Commit-Queue: Stephen White <senorblanco@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
4b911686
|
2022-06-06T22:25:37
|
|
Converts from sprintf() to snprintf().
sprintf() is deprecated in Xcode 14, so update to safer equivalents in
order to keep the compiler happy on iOS and macOS.
Bug: chromium:1331345
Change-Id: Id5348088bf69cbd360d9251e6323596cb710666d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3690747
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
9800a3cf
|
2022-05-19T16:52:28
|
|
Add a test for imageLoad after glClear to the texture
Bug: angleproject:7355
Change-Id: Ib1ddf685f2fa6cb24161615224971a7185234000
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3659504
Auto-Submit: Chris Dalton <chris@rive.app>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
98c2e169
|
2022-05-20T16:17:49
|
|
Vulkan: Reduce pre-rotation spec const to bool
The specialization constant now only dictates whether x and y should be
swapped. The complete 8 possible states of rotation and y-flip are
achieved by using this swap in combination with a driver uniform for x
and y flip.
Swapping is still a specialization constant to avoid degrading
performance of dFdx/dFdy which otherwise would need both to be evaluated
instead of one. On platforms which don't support pre-rotation, the
specialization constant will never change and driver uniforms entirely
govern y-flip. On platforms that do support pre-rotation, only two
variations of the pipeline are needed.
Bug: angleproject:7366
Change-Id: I73f84e89fa9349d2098fa5b21573aee57d93a30c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3663151
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Lingfeng Yang <lfy@google.com>
Reviewed-by: Charlie Lao <cclao@google.com>
|
|
b4ec8eac
|
2022-04-12T12:16:40
|
|
D3D11: implement image load/store for non-compute shaders.
Output image metadata and image2d replacement string in
other shader types.
Implement the actual HLSL transform for fragment shaders.
Re-enable problematic test.
Bug: angleproject:7121
Change-Id: I49b815695757e822e6fdfe599e07ceb5b1d6a75c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3550544
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Stephen White <senorblanco@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
c8c4109c
|
2022-05-02T11:29:24
|
|
HLSL: Disambiguate functions that have int/uint parameters.
If a bit-shift expression is passed as a function parameter in HLSL,
the compiler cannot tell if it is intended as a uint or int when
doing overload resolution. Explicitly disambiguate functions that have
int and uint parameters when generating the HLSL.
Bug: chromium:1319332
Change-Id: I11c9518e060e9940550bbb04dd7cb953d99c2bb8
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3621316
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Lingfeng Yang <lfy@google.com>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
5b4f180c
|
2022-04-29T16:56:34
|
|
Fix formatting
Unrelated changes created by running "git cl format"
Bug: None
Change-Id: I880c7d50aead70353f22502d953efa7aa60ff5d5
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3616882
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
fc828287
|
2022-04-24T00:12:11
|
|
Tests for Nvidia/Vulkan bugs with tessellation control
Bug: angleproject:7236
Change-Id: I9fec5ef0fe03e1bb5e11067d6d3cf2e7c0b560fa
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3604077
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
3b38b379
|
2022-04-20T10:44:24
|
|
Vulkan: Add feature avoid HOST_VISIBLE and DEVICE_LOCAL combination
Discrete GPUs device local memory usually is not CPU accessible. This
adds a feature flag to control that.
Fixed bug in BufferVk that when mapRangeImpl is called from angle
internal, unmapImpl was using front end mapping parameters that is
incorrect. We have to cache the mapping parameters in the backend to
hangle the mapRangeImpl/unmapImpl calls from internal.
Fixed the test bug in ComputeShaderTest.BufferImageBufferMapWrite that
we are calling glMapBufferRange with GL_MAP_READ_BIT but are actually
writing to the map pointer. This should result in undefined behavior per
spec.
Fixed the test bug in GLSLTest.* that VerifyBuffer calls
glMapBufferRange, but was giving incorrect length which result in data
only been partially copied. This bug was hidden due to previously all
buffers are CPU accessible and there is no copy needed.
Fixed the test bug in ReadPixelsPBOTest.* and ReadPixelsPBONVTest.* that
calls glMapBufferRangeEXT, but was giving incorrect length which result
in data only been partially copied. This bug was hidden due to
previously all buffers are CPU accessible and there is no copy needed.
Added new skipped syncval messages. Because this CL triggers a
copyToBuffer call for some of the buffers and that changes the syncval
message signature for the same reasons (i.e, feedback loop or synval
does not know the exact range of buffer been used for vertex buffers
etc).
Bug: angleproject:7047
Change-Id: I28c96ae0f23db8e5b51af8259e5b97e12e8b91f2
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3597711
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Charlie Lao <cclao@google.com>
|
|
039660a0
|
2022-04-14T13:32:33
|
|
Translator: Drop const from opaque parameter types
`const` is ineffective on opaque types like sampler2D. That qualifier
is now dropped as it was tripping SPIR-V gen up.
New validation is added to make sure such parameters are not
reintroduced.
Test credit of Lubosz Sarnecki <lubosz.sarnecki@collabora.com>
Test: angle_end2end_tests --gtest_filter="GLSLTest.ConstSamplerParameter*/*_Vulkan"
Bug: angleproject:7201
Bug: angleproject:7205
Change-Id: I310a47a8eb162ddafb2ab44682520f546f8a5512
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3587345
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
797e627e
|
2022-04-08T22:49:51
|
|
Autogenerate list of features as enum
The WithX() and WithNoX() helpers are removed and replaced with enable()
and disable() member functions that take the name of the feature (as a
Feature::X enum constant). This has two benefits:
- Adding tests that override a feature no longer requires additional
helper functions to be written.
- There's no mistaking the feature name.
This change doesn't yet fix the main issue in anglebug.com/6435, but
does fix the following helpers using an old feature name (so they were
ineffective):
- WithMetalForcedBufferGPUStorage
- WithNoVulkanViewportFlip
A follow up would remove the old way of overriding features in tests and
replaces them with the new way.
Bug: angleproject:6435
Change-Id: Ida02b26ec72bc40d7a8938c76a93815bb903ca05
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3580982
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
6b94a71c
|
2022-04-01T10:12:07
|
|
Vulkan: Lift SwS suppressions.
Bug: angleproject:4092
Change-Id: I0e72b95ba5ba2b60420d5b685349c405864c2e6e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3563513
Auto-Submit: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@google.com>
|
|
cd3c74af
|
2022-03-30T11:48:41
|
|
SPIR-V Gen: Fix aliasing out parameters
In ANGLE, when an unindexed lvalue was passed as an out parameter to a
function, SPIR-V was generated such that the lvalue is passed in
directly. A Skia test revealed a difference in SPIR-V and GLSL
semantics where aliasing out parameters are expected to work on local
copies until the end of the function.
Bug: b/226904235
Change-Id: I476af01eb7d065272825967111cd208faf88c275
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3561278
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
ce964d66
|
2022-03-26T18:56:50
|
|
Vulkan: Add tessellation to GetPipelineStage helper
The tessellation stages were missing from a helper, which was silently
returning an invalid value.
Add a test and an assert, which fires before the fix.
Test: GFXBench Car Chase
Test: GLSLTest_ES31.TessellationTextureBufferAccess
Bug: angleproject:7135
Bug: b/218314686
Change-Id: I2bc8d374300fc1470e52affabab7491698c99cee
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3554575
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Cody Northrop <cnorthrop@google.com>
|
|
40afbe4b
|
2022-03-20T23:39:57
|
|
Translator: Fix SimplifyLoopConditions producing dead code
This causes the tree to fail validation. Dead code could be generated
if the loop body ended in a branch. In such a case, no additional
instructions are added to the transformed loop body after the original
body.
Bug: chromium:1300782
Change-Id: I4ec46c170ab5a45fa9b9b73d4e2b62c9c4fa24a3
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3539442
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
7eb7596a
|
2022-03-10T14:32:58
|
|
Ensure non float uniform are actively used
In mesa, the non float uniforms are optimized out when they are
not active. This change ensures the unforms are active.
Reference: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6119
Bug: angleproject:7091
Change-Id: Iedff4a4b72a7204348c226778cfe940374632a4a
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3515640
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jiajia Qin <jiajia.qin@intel.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
8dc9e83e
|
2022-02-16T21:48:53
|
|
Translator: Fix RunAtTheEndOfShader w.r.t discard
RunAtTheEndOfShader wrapped main() if it ended in discard. However, it
didn't account for the fact that the discard instruction could be
wrapped in a block. This change makes RunAtTheEndOfShader more
conservative w.r.t discard and has it wrap main() on any encounter of
discard.
The change additionally adds AST validation to ensure transformations
don't generate dead code after branches.
Test credit to Cody Northrop.
Test: GLSLTest_ES3.ConstantConditionGuardingDiscard
Test: GLSLTest_ES3.NestedUnconditionalDiscards
Bug: angleproject:7033
Change-Id: Ie9d5210a5cfbb13449720a8a3f44666df9443d98
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3471014
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
eeddb050
|
2022-01-27T11:36:10
|
|
Metal: Incorrect parenthization for assignment expressions
Fix for Webgl tests conformance/ogles/GL/build/build_017_to_024.html
Create intermediate values for assign-equals operations, as they
cause issues on expansion later.
Bug: angleproject:6489
Change-Id: I7fea162bd1f59c7c765868af739d6ddfc1c1747f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3421527
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Kyle Piddington <kpiddington@apple.com>
|
|
06edae72
|
2022-01-11T11:13:20
|
|
Vulkan: Allow duplicated block name in different shader interface
In OpenGL ES 3.2 Shading Language specification(4.3.9), A block name is
allowed to have different definitions in different shader interfaces
within the same shader.
Bug: angleproject:5557
Test: KHR-GLES32.core.tessellation_shader.single.max_patch_vertices
Change-Id: I49f149cfd8f6f063fc5045aa154cc401d3e38cfb
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3429684
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Brandon Schade <b.schade@samsung.com>
|
|
72529c19
|
2022-01-31T14:37:42
|
|
Vulkan: Fix a few framebuffer fetch bugs
A few fixes to framebuffer fetch:
- The built-in symbol for coherent and non-coherent are now the same for
gl_LastFragData, so extension check can choose one or the other.
- The temporary variable used in a transformation is renamed to not be
gl_LastFragData.
- inout usage is now validated correctly for shader type and version.
Bug: angleproject:6951
Bug: angleproject:6905
Bug: angleproject:6907
Change-Id: I9af341a3e8e01ed375a8230618184494813d3ece
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3427577
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
aadc6434
|
2022-01-24T18:28:19
|
|
Vulkan: Fix to correctly redefine gl_PerVertex in GS
Angle translator structures were inconsistent of data
lookup of gl_PerVertex data when it is user defined.
Bug: angleproject:5579
Test: KHR-GLES32.core.geometry_shader.api.getProgramiv3
Test: GLSLTest_ES31.PerVertex*
Change-Id: Ied9ea25ce86ade29261f4f75ce0b48af15061760
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2962352
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Brandon Schade <b.schade@samsung.com>
|
|
b9f2d9cc
|
2021-12-16T13:08:22
|
|
Use text expectations to skip a test
This instead of exiting the test in C++.
Bug: angleproject:6826
Change-Id: I1e91f38f0f581821178b3b54ba809e59570b440a
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3345615
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Gregg Tavares <gman@chromium.org>
|
|
a0924015
|
2021-11-22T19:44:05
|
|
Metal:Fix GLSLTest_ES3.GLVertexIDIntegerTextureDrawElements Metal
The test was failing if the "first" argument to glDrawElements
was odd. When odd the offset was not a multiple of 2 bytes
and the code was going down a different path than when the
offset was event.
The fix for that just removes a check for alignment to 4 bytes.
Checking with UNSIGNED_BYTE indices though and it still
failed. That was because it was encoding the index conversion
path on one command buffer and the provoking index prep
on another and so they were happening out of order.
Putting both on the same command buffer fixes that.
Bug: angleproject:6688
Change-Id: I89b68ba965d073e0fa70b6d5b444dc743b5e34a7
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3296631
Commit-Queue: Gregg Tavares <gman@chromium.org>
Reviewed-by: Kyle Piddington <kpiddington@apple.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
edc2cfe5
|
2021-11-19T14:57:16
|
|
Fix signed integer overflow in angle_end2end_tests.
The test math was causing an signed overflow in the most extreme
case.
Test: GLSLTest_ES3.GLVertexIDIntegerTextureDrawArrays/ES3_Vulkan_SwiftShader
Bug: angleproject:5795
Change-Id: Id6b845577cb6ce585ec67251dcb6111d5773012f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3289942
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
d87d10cd
|
2021-11-12T11:02:01
|
|
Fix ASAN issue with GLVertexIDIntegerTextureDrawElements
Was
GLint clearData[] = {42};
glClearBufferiv(GL_COLOR, 0, clearData);
Now
GLint clearData[4] = {42};
glClearBufferiv(GL_COLOR, 0, clearData);
Bug: angleproject:6688
Change-Id: Ia8fa0c7d406626046a32e613fda3a165a06941b6
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3278489
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
c6d86da2
|
2021-11-11T15:30:41
|
|
Add test for gl_VertexID and DrawElements
This fails on the WebGL CTS conformance2/rendering/vertex-id.html.
Repoing here to make it easier to debug/find/fix.
Bug: angleproject:6688
Change-Id: I0a4bafc5700c77e0b6c44b067167fc4a60e63c22
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3276708
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Gregg Tavares <gman@chromium.org>
|
|
5a5996aa
|
2021-11-09T21:42:43
|
|
Vulkan: Handle missing decorations in SPIR-V precision fixer
Bug: angleproject:6674
Change-Id: Ibfc72f14b9eeb17d7c70cf6d2384d970b2e66650
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3271551
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
52d6e363
|
2021-11-03T12:46:10
|
|
Translator: Fix zero-init of nameless I/O blocks
Bug: angleproject:6609
Change-Id: I680fec85d53709f9abd3680a9dc5794ecfae6588
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3260326
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
e10768f0
|
2021-11-03T12:23:11
|
|
Translator: Fix zero-init of I/O block arrays
Bug: angleproject:6609
Change-Id: I8245e868fe6bc4798e74088916e1d8cdd0727b6b
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3260325
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
5c914b57
|
2021-10-28T13:27:04
|
|
Vulkan: SPIR-V Gen: Fix float+matrix
Found by a fuzzer, float+matrix was not handled correctly by the SPIR-V
output.
Bug: chromium:1264212
Change-Id: I09b13c3e48374621228f5fab4de68c33973ddfd4
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3251585
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
066fb91b
|
2021-10-18T15:31:27
|
|
Vulkan: SPIR-V Gen: Handle scalar(const) produced by index clamp
Normally scalar(const) is folded into a constant. The index clamp
transformation may produce such a code where the index looks dynamic at
first (for example `false ? uniform_value : constant`), but becomes
constant after folding.
This change makes SPIR-V generation robust in that case. A potential
future change could avoid the clamp entirely by making FoldExpressions
adjust the op of the EOpIndexIndirect node whose index is being replaced
with a constant with EOpIndexDirect (and apply the clamp on the
argument).
Bug: chromium:1260651
Change-Id: I552b7527d821d1cb52e0e53212cc481285674861
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3226311
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
ef237faf
|
2021-10-18T16:22:40
|
|
Translator: Validate precision for function args
In fragment shaders, float does not have a default precision. Any
declaration of this type must therefore specify the precision if a
default one is not provided.
This was not validated for function arguments.
Bug: chromium:1255089
Change-Id: I0d17e226ec88610692ec7dd18793cf4d471f12e7
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3226314
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
7f87a326
|
2021-10-18T13:06:29
|
|
Vulkan: SPIR-V Gen: Fix crash in array of struct constant
Bug: chromium:1260690
Change-Id: I51fe85a2ebc23c3fcaa3c961c4ebf84688bbed31
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3226309
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
f3d5dac3
|
2021-08-23T17:25:15
|
|
Vulkan: SPIR-V Gen: Drop dependency to glslang
The SPIR-V gen path is now made default. Compilation through glslang is
still supported for debugging, and is enabled on the GLSL* end2end tests
for smoke testing. On release builds, glslang is not supported.
To test with glslang, add the following gn arg (only necessary if dcheck
is disabled):
angle_enable_spirv_gen_through_glslang = true
Then enable the generateSPIRVThroughGlslang feature. This can be done
by setting an environment variable:
ANGLE_FEATURE_OVERRIDES_ENABLED=generateSPIRVThroughGlslang ./angle_deqp_gles2_tests
Binary size saving:
- 1.3MB on Linux (SPIR-V gen itself: 240KB)
- 730KB on Android (SPIR-V gen itself: 140KB)
Perf tests:
- LinkProgramBenchmark.Run/vulkan_compile_single_thread
* Through glslang:
truncated mean: 1287033.36
* Direct SPIR-V Gen:
truncated mean: 244495.91 (~80% reduction)
- LinkProgramBenchmark.Run/vulkan_compile_multi_thread
* Through glslang:
truncated mean: 4565894.83
* Direct SPIR-V Gen:
truncated mean: 1158164.10 (~75% reduction)
Bug: angleproject:4889
Bug: angleproject:6210
Change-Id: I486342702977c8114e90073b97183aba115a8b2d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3115140
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
44bef8ae
|
2021-10-08T15:01:37
|
|
Remove traces of Feature Level 9_3 support from tests.
Bug: angleproject:1284
Bug: angleproject:3042
Change-Id: Ic2f77d315e98c3b15c5fc8b0359168bce00867d4
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3213294
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
0f66c861
|
2021-10-07T09:32:35
|
|
Add test that covers HLSL compiler bug.
Bug: angleproject:2445
Change-Id: I820b2dba3748bd10ca516ef47f226ff537b78849
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3212328
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
212f4592
|
2021-09-22T12:02:47
|
|
Enable direct-to-metal backend by default
We are switching over to Apple's direct-to-metal backend instead
of generating SPIRV in the metal backend. This CL enables the
direct-to-metal generation by default, but the SPIRV backend is
still accessible by overriding the feature directMetalGeneration.
This CL comes with a change in test expectations to catch new
failures and clean up newly passing tests.
Bug: angleproject:6080
Change-Id: I4b10ad93c641b88857079a08fb45d3dc575d71f1
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3175664
Reviewed-by: Gregg Tavares <gman@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Jonah Ryan-Davis <jonahr@google.com>
|
|
daaaba0a
|
2021-10-04T22:28:56
|
|
Translator: Fix uniform/struct separation for anonymous arrays
When a uniform array of anonymous structs was separated into a struct
declaration and uniform array, the uniform symbol was being replaced,
but not the intermediate nodes that index it. This caused an AST
validation error.
Bug: angleproject:4740
Change-Id: Ib9d0115bd01c39d1a007f644e46b5e35c12c9304
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3204590
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
dde409be
|
2021-10-04T22:05:47
|
|
Translator: Produce error on func(void, int)
If void is specified, another parameter cannot follow. This change
ensures that this produces an error.
Bug: angleproject:6338
Change-Id: I18602ea4b52c96e498f4f2dc627d4ba2210d6fa7
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3203795
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
234ebd58
|
2021-10-05T08:34:12
|
|
Lift old AMD/Win/GL expectations.
No longer tested on the bots.
Bug: angleproject:1643
Change-Id: I5e1ca8b9c2f673ebcb44c67879829f9fd346edf0
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3205795
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
696f8533
|
2021-09-28T15:21:51
|
|
fix GLSLTest_ES3.GLVertexIDIntegerTextureDrawArrays
The test was using a gl_PointSize of 1000.0 but GL_POINTS
are only required to support a max size of 1.0
Bug: angleproject:6463
Change-Id: I8a44607085583cbc7cef06f21b118121cc1fef24
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3192177
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Gregg Tavares <gman@chromium.org>
|
|
bc8fe7dd
|
2021-09-24T22:50:32
|
|
Translator: Fix removing inactive output vs SH_INIT_OUTPUT_VARIABLES
The code that SH_INIT_OUTPUT_VARIABLES adds for inactive outputs is also
removed with this change.
Bug: chromium:1253016
Change-Id: I3caeb628487354de943f7b53f7462f58baa3eca8
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3183107
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
d71dc819
|
2021-09-22T16:01:06
|
|
Fixes for RowMajor test failures
row_major matrices get translated to column major matrices
and the expressions rewritten. To do this temporary variables
that contain the column and row index expressions are moved
to separte statements before the matrix expression itself.
When this happens it breaks side effects and short circuits
in the expression.
The solution is to call SeparateCompoundExpressions before
calling RewriteRowMajorMatrices.
Bug: angleproject:6386
Change-Id: I4d819ade959e2a875495f62733e11d86560d3fab
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3177337
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Gregg Tavares <gman@chromium.org>
|
|
e1a9af6a
|
2021-09-20T10:42:03
|
|
Translator: Fix HLSL vs for loop's init not having initializer
The handleExcessiveLoop code expected that if a for loop has an init
expression that it must necessarily have an initializer as well. That
is not true in (fuzzer-produced) code such as following:
for (int i; ...; ...)
Bug: chromium:1238182
Change-Id: Id761e937a34a6b2feac1266e22207350a3bd616c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3170114
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
46149c0e
|
2021-09-17T15:38:47
|
|
Translator: Fix pruning functions that declare a struct
When pruning unused functions, the entire declaration of the function is
removed. If the function declares a struct as part of its return value,
the declaration of that struct was lost. This change makes sure that
declaration survives the pruning. For example:
struct S { vec4 v; } unused_func() { ... }
is now replaced with:
struct S { vec4 v; };
This change also makes struct usage validation more stringent.
Bug: chromium:1248753
Change-Id: Idd9a87fd2e785135775cfea62995fd33adaf3c09
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3169691
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
085ab6e8
|
2021-09-09T10:22:04
|
|
Turn on Direct Metal tests
Bug: angleproject:5505
Change-Id: I8de95f28698785e5467dd82d04cbc7636d6df042
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3152156
Commit-Queue: Gregg Tavares <gman@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
3ae1f336
|
2021-09-03T09:34:10
|
|
WebGL: Make unsuccessful links fail subsequent draw calls.
This protects against incomplete state updates during a failed
link call that can interfere with draw calls.
Bug: angleproject:6358
Bug: chromium:1241123
Change-Id: Ie892654c3a58c69d6e35ba3c41758ab6269d8193
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3140496
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
|
|
ab187c35
|
2021-09-02T14:46:29
|
|
Handle comparing arrays of scalars/vectors/matrices
Fixes:
UniformBlockWithOneLargeArrayMemberTest.MemberArrayOperations/ES3_Metal
Bug: angleproject:5505
Change-Id: Ife35db2146ac8ce8c0880cdb98df57dc75cb35cb
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3140712
Commit-Queue: Gregg Tavares <gman@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
b6e99bb3
|
2021-09-02T13:47:34
|
|
Rename and expand shader variable init feature.
This feature now forces both output and unitialized locals to be
initialized by the shader translator. This feature is needed by the
trace validator to ensure we get deterministic behaviour in traces
that exhibit some undefined results.
Bug: angleproject:5133
Change-Id: Id1242cd077a57e891eed217f7671976ce1631a58
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3140216
Reviewed-by: Cody Northrop <cnorthrop@google.com>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
9d0e2851
|
2021-09-01T00:42:58
|
|
Fix SeparateDeclarations vs struct specifiers
Bug: chromium:1237696
Change-Id: I3b00f3797800e814ca83226a8e4f25b2a43cc641
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3133824
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
a29b07d8
|
2021-08-26T13:11:29
|
|
InitializeVariables: Init shader IO block outputs.
These variables were not handled in the pass because we had no prior
test coverage of them. Some frame capture testing uncovered this gap.
The variables must be initialized field-by-field, since there seems
to be no defined way to initialize an entire block at once.
Bug: angleproject:6326
Change-Id: Ib7aecfb76b97b4236d786b44b3dfb706c573e221
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3123228
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
800e82c6
|
2021-08-23T11:05:23
|
|
Translator: Validate precisions
When declaring a variable, a struct field, function parameter etc,
there's a precision necessarily applied to the entity being declared.
AST Validation is added to enforce this. Intermediate nodes derive
their precision from these entities automatically.
Consistency of intermediate nodes is not validated. This is because AST
transformations replace a node with a transformed one, and that may not
have the same precision. Take the following code:
mediump float x = ...;
mediump float y = ...;
... x + y ...
and assume is transformed as such:
highp float driver_uniform;
... (x * driver_uniform) + y ...
The addition was originally done in mediump, but would seemingly need to
be done in highp after transformation. There are a number of options
here:
- Make sure that when nodes are replaced, the precision is unaffected.
This can be intrusive, requiring temp variables.
- Bubble up the new precision
- Accept the discrepancy
ANGLE opts for the last option, which actually respects the original
shader's intended precision for operations, even if some transformation
needs to temporarily evaluate an expression at a higher precision.
Bug: angleproject:4889
Bug: angleproject:6132
Change-Id: Ibcde3a230de159157783b1c6d5ef1cd63ceb4d8f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3114027
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
a141cb52
|
2021-08-23T13:52:57
|
|
Test vector and matrix scalarization does not affect output
Vector and matrix scalarization was turned off because
it affects the rendering output. We need to turn it
back on in some form for Metal. This test is added to
check that rendering output is not affected.
Test adapted from WebGL CTS test
https://github.com/KhronosGroup/WebGL/pull/3214
Bug: angleproject:6306
Change-Id: Ic1db18920064726761d6168f9dd5111772028598
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3115013
Commit-Queue: Gregg Tavares <gman@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
|
|
8ffb4602
|
2021-08-18T16:55:56
|
|
Turn on the first few direct-to-Metal tests by default.
All of the subtests of these previously-asserting tests are passing
with --bot-mode:
ProgramBinaryES3Test
UniformTest
WebGLGLSLTest
so start running them on ANGLE's CQ.
Bug: angleproject:5505
Change-Id: Ibdaa814cec846379e2c073117efa816fc871be33
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3105442
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
d20afde4
|
2021-08-05T23:06:44
|
|
Vulkan: Retain precision of uniform variables
When gathering in a uniform block, the precision of default uniforms was
dropped.
Bug: angleproject:4889
Bug: angleproject:6132
Change-Id: Ie6a8e2e7cef7cd3808cee08d20f886fc4e762cd5
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3076124
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
63248e7f
|
2021-08-05T23:49:23
|
|
Translator: Fix dead-code-elimination corner case
The DCE code had a corner case bug where a switch case containing
multiple DCE'ed `break`s followed by another case would cause some of
the DCE'ed statements in the first case to be doubly pruned, failing on
an assertion.
This was due to the fact that visitBlock() was asking traversal to
continue if a new case was visited while pruning nodes, but the
traversal does not take into account that some statements need to be
skipped and so would also visit the pruned statements. If the pruned
statements contain a branch instruction, they get re-pruned.
The visitBlock() function that does the pruning is reworked so that it
more clearly traverses the statements.
Bug: chromium:1237200
Change-Id: Ib078c2ea73ade756c7d7ef5a5c489fa53c39f352
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3077659
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
09d5047b
|
2021-08-03T01:31:03
|
|
GL: Remove the rewriteVectorScalarArithmetic workaround
This workaround was implemented for a bug in Nvidia driver 387 which is
end-of-life. More recent driver branches (390 and 4XX) are fixed, and
so this workaround is no longer required. The implementation of the
workaround itself could introduce bugs.
Bug: chromium:772651
Bug: chromium:1201084
Change-Id: I3db179eb90d9124235bdad2daacc712302906d8e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3067952
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
|
|
aca9f2e2
|
2021-07-29T17:49:26
|
|
Vulkan: SPIR-V Gen: Fix constructors with uniform parameters
When the constructor contains parameters from interface blocks, they may
have SPIR-V type specializations that was not previously accounted for.
The code is made more robust by retrieving the column/component types
when generating constructors through TTypes instead of modifying
SpirvTypes directly.
Bug was caught by the WebGL CTS.
Bug: angleproject:4889
Change-Id: I6c9df17a97fac8622dea55ed67c7bae1ce16d1bd
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3061640
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
a83c9108
|
2021-07-29T17:13:41
|
|
Vulkan: SPIR-V Gen: Fix nested ternary operators
The OpPhi instruction used to implement ternaries referenced to the
blocks pregenerated for the true and false cases. However, if there
were other code nested inside that could generate blocks (other
ternary expressions or short-circuits), that would be wrong.
Instead, the last block id is taken every time the true or false case is
entirely visited (similar to how it's done in the short-circuit
implementation).
Bug was caught by the WebGL CTS.
Bug: angleproject:4899
Change-Id: I71be86989dfe9596a13940ce657a7e67849e86bf
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3061639
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
2b23aae4
|
2021-07-29T16:39:28
|
|
Vulkan: SPIR-V Gen: Support vec(..., mat)
Matrices specified in vector constructor arguments were not correctly
handled w.r.t to casting (caught by the WebGL CTS).
Bug: angleproject:4889
Change-Id: I70253f049a651fc5f439b56d1e4d8f785813b605
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3061638
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
aec5e65c
|
2021-07-28T00:36:12
|
|
Get direct-to-Metal backend to run angle_end2end_tests.
Cherry-pick nameless struct fix from Apple in
https://bugs.webkit.org/show_bug.cgi?id=227482 .
Fix SeparateCompoundStructDeclarations pass to stop generating
multiple declarations; thanks syoussefi@ for advice.
Incorporate additional passes from TranslatorVulkan
(MonomorphizeUnsupportedFunctionsInVulkanGLSL,
RewriteArrayOfArrayOfOpaqueUniforms,
SeparateStructFromUniformDeclarations) needed by RewriteStructSamplers
pass in TranslatorMetalDirect. Fixes many assertion failures in GLSL
tests. Moved these passes out of tree_ops/vulkan. Thanks again to
syoussefi@ for advice and help.
Disable a validation check related to the RewritePipelines pass. Skip
two tests that were failing for other reasons.
With these changes, angle_end2end_tests runs to completion when the
direct-to-Metal backend is turned on. There are still ~1300 failures
of the ~4000 tests which will be investigated next.
Bug: angleproject:5505
Change-Id: Ibca77822543e8e8e8d2a8c862e92cdf74bfa3545
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3058524
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Kenneth Russell <kbr@chromium.org>
|
|
8bf2e5ec
|
2021-07-23T16:54:16
|
|
Vulkan: SPIR-V Gen: Fix switch with default at the end
A minor bug was making a switch with default at the end to jump to the
merge block instead of the default block.
Bug: angleproject:4889
Change-Id: Ied434fab949b10d45a0db1242c1b8535a5f4f773
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3048321
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
d0c03ff4
|
2021-07-22T14:55:51
|
|
Vulkan: SPIR-V Gen: Fix lvalues passed to in/inout parameters
In GLSL, these values are semantically copied when passed to a function
as an in or inout parameter. For example in:
bool f(inout vec4 a, inout vec4 b)
{
a = vec4(0);
return all(equal(a, b));
}
var = vec4(1);
bool result = f(var, var);
result is expected to be false. In SPIR-V, every parameter is
semantically passed by "reference".
glslang conservatively uses temporaries to pass to functions. An
optimization in ANGLE didn't create temporaries for unindexed lvalues,
which did not take into account the above fact. This optimization is
limited to out parameters now.
Bug: angleproject:4889
Change-Id: Ie1b4b1cecba847ba63d5810d01d0856823b89ddc
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3046103
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
9d23ae62
|
2021-07-22T12:32:44
|
|
Vulkan: SPIR-V Gen: Fix conditionals with pruned blocks
Rewored the visitIfElse function to simplify the logic and correctly
handle if-else constructs where the true block is pruned.
Bug: angleproject:4889
Change-Id: Ib968a2fe65f4b6463158fd76e7d67757115ee832
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3046101
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
951b04bb
|
2021-07-21T12:20:13
|
|
Vulkan: SPIR-V Gen: Support scalar constructors
In GLSL, the scalar constructor can be given a vector and matrix, in
which case the first element is selected. This case was not previously
handled.
Bug: angleproject:4889
Change-Id: Ibf276883fc7396b750981a4e469ff9b152c6e700
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3042554
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
6e261453
|
2021-07-20T22:34:33
|
|
Vulkan: SPIR-V Gen: Fix I/O block arrays
Since the qualifier on intermediate nodes are EvqTemporary, the code
that autodeduced the block storage of I/O blocks was incorrectly
assigning them std140 when an indexed I/O block array was encountered.
This resulted in duplicate types in the SPIR-V. This is generally
benign (other than creating an unnecessary type) except for gl_PerVertex
as it adds BuiltIn decorations. gl_PerVertex may be an array in
geometry and tessellation shaders.
Bug: angleproject:4889
Change-Id: Iaa8e414ae01a4be127dc52df0e9406546b23d24c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3041621
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
a8807cbf
|
2021-07-17T00:00:27
|
|
Vulkan: SPIR-V Gen: Support interpolateAt*
The interpolateAt* builtins take their first argument as a pointer, so
only the OpAccessChain result should be given to the corresponding
SPIR-V instruction.
A test is written in this change to handle interpolateAt* when given a
swizzled lvalue, which is not representable in SPIR-V. This can be
supported by having the AST transform:
interpolateAtX(value.zy)
to:
interpolateAtX(value).zy
which can be done right when the function is parsed and the AST node for
it is created (by taking the swizzle out of the parameter and applying
it to the node). However, swizzled lvalues as parameter to
interpolateAt* is only allowed in desktop GLSL, and so the test is
disabled. From the GLSL ES3.2 spec:
> Component selection operators (e.g. .xy) may not be used when
> specifying interpolant.
Bug: angleproject:4889
Change-Id: I043969f22011e6171c7ae225ded3a5013e8cfa4e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3035588
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
32196d8f
|
2021-07-16T14:49:08
|
|
Vulkan: SPIR-V Gen: Support built-ins with out parameters
Some built-ins have out parameters; modf, frexp, uaddCarry, usubBorrow,
umulExtended and imulExtended. In SPIR-V, they are translated as such:
- modf and frexp: Returns the same result as GLSL, with a pointer to
store the output parameter. Since SPIR-V cannot create pointers to
swizzles, a temp variable is used in that case.
- uaddCarry, usubBorrow, umulExtended and imulExtended: These return a
struct in SPIR-V, which contains the two output parameters; with lsb
in field 0 and msb in field 1.
Bug: angleproject:4889
Change-Id: Ic5114f74b9c1013093007e91939f78d5218f2a07
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3036088
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
f4ce6d00
|
2021-07-15T11:07:25
|
|
Vulkan: SPIR-V Gen: Support bool in interface blocks
SPIR-V doesn't allow bool in interface blocks. Another type
specialization is added for this purpose, turning those bools to uint.
Bug: angleproject:4889
Change-Id: I803bffcf5ea58c913d4df6e7aae3386c67901b25
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3032021
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
9c6fb52c
|
2021-07-13T14:10:41
|
|
Vulkan: SPIR-V Gen: Assignment between mismatching SPIR-V types
In general, GLSL qualifiers translate to SPIR-V decorations on SPIR-V
variables. In the case of blocks (struct or interface block),
OpMemberDecorate is used, which due to its specification in SPIR-V, can
only apply decorations to direct members of a block. This makes it
impossible for example to decorate a nested member of a block through
its variable id.
As such, some decorations such as RowMajor and Invariant apply to
members of a block given its _type_ id. Unfortunately SPIR-V requires
ArrayStride to also be applied to a type directly, rather than a member
of a block. This implies that some types, such as structs used in
uniform/buffer interface block, or decorated with invariant or
row_major, as well as arrays (of any type) used inside and outside
interface blocks to produce different SPIR-V types from the same GLSL
type. The SpirvTypeSpec data previously introduced specialize these
types.
It's necessary to "cast" between these types when needed. The
translator handles casts at load/store boundary:
- Upon load, the value is cast to the type with the default
SpirvTypeSpec.
- Upon store, the value is cast from the default to the store target
SpirvTypeSpec.
- All intermediate results use the default SpirvTypeSpec.
Bug: angleproject:4889
Change-Id: I6fa28e518ec6b517ff163f44b6892859eb4b10fd
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3026145
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
07c39616
|
2021-07-08T17:01:31
|
|
Vulkan: SPIR-V Gen: Enable GLSLTest*
This change includes an assortment of small fixes to enable most of
GLSLTest* end2end tests.
- User-defined std140 storage buffers were mistakenly turned to std430.
- External and WEBGL samplers were redundantly redeclaring the sampler2D
type.
- `invariant` specified on a field of struct type didn't apply it to
said struct's members.
- Arrays of struct as a member of an interface block didn't have their
size aligned, producing incorrect stride
- Interpolation and auxiliary qualifiers are now generated
- Arrays of opaque uniforms are now correctly passed to functions when
indexed.
- atan(x, y)'s built-in Op is fixed
- early_fragment_tests is now generated
Additionally, the SPIR-V transformer didn't handle the Invariant
decoration applied to gl_PerVertex built-ins, which is possible as a
result of:
#pragma STDGL invariant(all)
This bug existed in the transformer due to two bugs:
- ANGLE didn't generate this line when generating GLSL to feed to
glslang. This is fixed by this change.
- glslang itself ignores this line:
https://github.com/KhronosGroup/glslang/issues/2689
Bug: angleproject:4889
Change-Id: I15ad9fcc178abc1eae944e7a58f4e6153fa5dfd9
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3016762
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
afdf378f
|
2021-07-01T21:55:58
|
|
Vulkan: SPIR-V Gen: Support == and != on complex types
In GLSL, == and != produce a bool. In SPIR-V, there are only
component-wise comparison operations. For ==, OpAll should be used to
reduce it to a single bool, and for !=, OpAny. For matrices, the
comparison is done column by column with a similar reduction for each
column, and one for the final result. For structs and arrays, the
comparison is similarly done field by field or element by element
respectively.
Bug: angleproject:4889
Change-Id: I8157c1931b7d1dedd74a3825967f5e212d346900
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3001905
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
3b0fcf6a
|
2021-06-30T15:14:48
|
|
Vulkan: SPIR-V Gen: Support type casts in constructors
GLSL basic, vector and matrix constructors can convert between types.
This was already done for constants used in constructors. This change
implements the cast for non-constant expressions.
Bug: angleproject:4889
Change-Id: I0a8c1a6e97ffced0d1652032a41fb87c70be16ca
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2999022
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|