|
32d6e806
|
2021-09-24T23:50:54
|
|
Vulkan: Fix rotation of interpolateAtOffset
interpolateAtOffset's offset was multiplied by flipXY*rotationMatrix,
which is a vector. This was unable to actually apply rotation to the
offset. The multiplier is changed to a matrix.
Can be tested by running interpolate_at_offset*default_framebuffer tests
in deqp gles3.1 together with pre-rotation:
./angle_deqp_gles31_tests
--deqp-case=dEQP.GLES31/functional_shaders_multisample_interpolation_interpolate_at_offset_no_qualifiers_default_framebuffer
--emulated-pre-rotation=90
Bug: angleproject:5843
Change-Id: I9727371375e999f07463b69153df3fd8b28c2a71
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3183108
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Ian Elliott <ianelliott@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
0bb3fcbb
|
2021-09-07T14:28:09
|
|
Translator: Fix precision of gl_PointSize on ES3+
In ESSL100, gl_PointSize is declared as mediump. In ESSL300+, it's
declared as highp.
Bug: angleproject:4889
Bug: angleproject:6132
Change-Id: Ie904a12fdaaca04e2507f1796d9d0fe34b154eab
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3145615
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@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>
|
|
|
9c271645
|
2021-08-27T13:07:06
|
|
Vulkan: SPIR-V Gen: Fix gl_PerVertex without clip/cull support
When the clip/cull distance extension is not supported, the
gl_ClipDistances and gl_CullDistances arrays were given a size of 0,
which made them translate as OpRuntimeArray.
This change makes sure that these arrays always have a non-zero size.
Bug: angleproject:4889
Change-Id: I8d2ffe3a7c1df9316e91eebc2aa53bdc2c87ccc1
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3123354
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: 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>
|
|
|
210773db
|
2021-08-05T10:41:59
|
|
Translator: Be more explicit about precisions
GLSL ES requires that every symbol (variable, block member, function
parameter and return value) is appropriately qualified with a precision,
either individually or through the global precision specifier.
Some tree transformations however produced symbols with EbpUndefined
precision. In text GLSL output, these would produce unqualified symbols
which was often incorrect.
In this change, the transformations are made to produce explicit / more
consistent precisions. The validation (that caught these issues) is not
included in this change as there are still a few corner cases left to
address.
Bug: angleproject:4889
Bug: angleproject:6132
Change-Id: Icca8a0a5476f8646226e7243aa8f501f44acc164
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3075127
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
061188a7
|
2021-08-04T10:07:47
|
|
Translator: General clean up
General clean up done as part of other changes, split to simplify
review.
Bug: angleproject:4889
Bug: angleproject:6132
Change-Id: Iade9954d187a759be9edd0e3754be007f4133c56
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3071598
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
a65b4056
|
2021-07-30T23:00:57
|
|
Reland "Translator: Clean up type cloning"
This is a reland of f016c4352f5203c10511df078b1ed5359afc1b35
Original change's description:
> Translator: Clean up type cloning
>
> TType has a constructor that aids cloning + helpers to convert between
> types. A number of places where a type is constructed from the
> information gathered from another type is changed to clone the type and
> then use one of these helpers.
>
> This clean up is part of an ongoing work to improve precision handling.
> This change removes many references to TType::getPrecision().
>
> Bug: angleproject:4889
> Change-Id: Ib85659ab5363b56ad298f8648fca856edc1ebf8b
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3063944
> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
> Reviewed-by: Jamie Madill <jmadill@chromium.org>
> Reviewed-by: Tim Van Patten <timvp@google.com>
Bug: angleproject:4889
Change-Id: I35772f178eb4f6cf2b64bfeb37a4a144acdb4daf
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3067802
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
c5c8a25a
|
2021-08-03T14:40:33
|
|
Revert "Translator: Clean up type cloning"
This reverts commit f016c4352f5203c10511df078b1ed5359afc1b35.
Reason for revert: Suspect for crbug.com/1236060
Original change's description:
> Translator: Clean up type cloning
>
> TType has a constructor that aids cloning + helpers to convert between
> types. A number of places where a type is constructed from the
> information gathered from another type is changed to clone the type and
> then use one of these helpers.
>
> This clean up is part of an ongoing work to improve precision handling.
> This change removes many references to TType::getPrecision().
>
> Bug: angleproject:4889
> Change-Id: Ib85659ab5363b56ad298f8648fca856edc1ebf8b
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3063944
> Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
> Reviewed-by: Jamie Madill <jmadill@chromium.org>
> Reviewed-by: Tim Van Patten <timvp@google.com>
Bug: angleproject:4889
Change-Id: I9b5ecacc410b41be382232292ceae2f7ea7906c7
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3064393
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Commit-Queue: Austin Eng <enga@chromium.org>
|
|
|
f016c435
|
2021-07-30T23:00:57
|
|
Translator: Clean up type cloning
TType has a constructor that aids cloning + helpers to convert between
types. A number of places where a type is constructed from the
information gathered from another type is changed to clone the type and
then use one of these helpers.
This clean up is part of an ongoing work to improve precision handling.
This change removes many references to TType::getPrecision().
Bug: angleproject:4889
Change-Id: Ib85659ab5363b56ad298f8648fca856edc1ebf8b
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3063944
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
ae8ef039
|
2021-07-30T00:43:14
|
|
Translator: Correctly respect precision of gl_FragColor/Data
These builtins are declared with a mediump precision. However, during
translation to Vulkan GLSL, the precision was not output and as such
glslang treated them as highp.
Discovered through comparison with SPIR-V output from ANGLE which
included more RelaxedPrecision operations.
Bug: angleproject:4889
Change-Id: I7974478e87ad20115345d679f8f035492be1f349
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3062204
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
ebe943e2
|
2021-07-29T11:19:33
|
|
Vulkan: SPIR-V Gen: Fix gl_Clip/CullDistance
These built-ins can be redeclared in the shader. The translator took
these redeclarations (and gl_LastFragData) as UserDefined symbols.
There were a number of hacks such as in name generation and
CollectVariables, to special-case these redeclarations.
This change instead makes sure that these variables continue to be
considered built-ins with the appropriate qualifiers. A number of fixes
are made accordingly:
- Hacks are removed.
- In fragment shaders, ANGLEClipDistance was initialized with
gl_ClipDistance for further use by the shader. The code generation
however mistakenly produced `gl_ClipDistance[0] =
ANGLEClipDistance[0];`, which caused compilation failures by glslang,
but passed the tests accidentally because they expected link failures
(see next item).
- CollectVariables is fixed to correctly collect gl_Clip/CullDistance
in fragment shaders; previously they were collected as output
varyings, and therefore the aforementioned link error was not produced
in the tests after the compilation error was fixed.
Additional fixes:
- The transformation of gl_Clip/CullDistance was always ever done on one
of them due to misplaced breaks in the loop that detected their
presence.
- The transformation of gl_CullDistance was skipped when it was not
redeclared.
Validation is added to make sure these built-ins always have the correct
qualifier even when redeclared. SPIR-V gen support for these variables
have been fixed as well.
Bug: angleproject:4889
Change-Id: Ic8bf2c508357035cb937ef41a28ae22ffc533ebe
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3059921
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>
|
|
|
96ccf3f4
|
2021-07-29T12:38:21
|
|
Translator: Consolidate EvqFragDepth and EvqFragDepthEXT
They were the same thing. Fixes SPIR-V generation which only handled
EvqFragDepth.
Bug: angleproject:4889
Change-Id: Id2c35f4c5a26e43ad16a420cc0306756a1add082
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3059922
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
7c6da493
|
2021-07-26T21:24:25
|
|
Translator: Make sure built-in variables are consistent
Some transformations left the tree in an inconsistent state, for example
if it used gl_SampleMask from ES3.1+OES_sample_variables and the
transformation added a usage of gl_SampleMask from ES3.2. The offending
transformations are fixed and AST validation is improved to catch such
errors.
Bug: angleproject:4889
Change-Id: I9d9ea5bb43a9408dd4c6dc3e89ec20d60dfeff73
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3054613
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
7f199d75
|
2021-07-20T00:11:23
|
|
Vulkan: SPIR-V Gen: Support geometry shaders
Geometry shader built-ins are implemented. A bug in
DeclarePerVertexBlocks is fixed w.r.t to gl_PerVertex arrays and a new
AST validation option is added to catch that sort of error.
Bug: angleproject:4889
Change-Id: I40b2929e450eac8ae840bbf1a1583e6c4c047c88
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3041623
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
0e1c90cf
|
2021-07-21T10:24:26
|
|
Translator: Validate function parameter qualifiers
This change cleans up a confusion in EvqConst and EvqConstReadOnly,
where the former was frequently used instead of the latter for function
parameters.
The change makes the following renames to make the intent of the
relevant qualifiers clearer:
EvqIn -> EvqParamIn
EvqOut -> EvqParamOut
EvqInOut -> EvqParamInOut
EvqConstReadOnly -> EvqParamConst
Bug: angleproject:4889
Change-Id: Idedd32c08a91de069b91b1657d6b783dddece04a
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3041622
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
a3df80ea
|
2021-07-19T17:15:13
|
|
Translator: Validate AST consistency w.r.t function calls
This ensures that tree transformations don't create invalid function
calls. Additionally, it's validated that no raw function calls are
issued, as that's deprecated.
Bug: angleproject:4889
Change-Id: Iabbb493dcc0a5654bbb5f930e402bf3492c89979
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3040119
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
|
|
|
05eb0a32
|
2021-07-19T16:14:34
|
|
Translator: Allow subpassLoad in AST in ESSL
subpassLoad is available to desktop GLSL, but not GLSL ES. With this
change, subpassLoad is made available internally to the AST so that
framebuffer fetch's use of this built-in can be simplified.
Additionally, this makes the tree more consistent in that there's no
function call without a corresponding function declaration.
A follow up change will add an AST validation to enforce this.
Bug: angleproject:4889
Bug: angleproject:5454
Change-Id: I3fa502dc4b29eab7f34940466e283cffd218d746
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/3040118
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
153240b2
|
2021-06-11T16:36:32
|
|
Vulkan: SPIR-V Gen: Support loops
Loops are similar to if-else in that they generate a number of blocks
where the first block specifies divergence (OpLoopMerge) and the merge
block. Differently from if-else, there is a block where the condition
is evaluated and a block which `continue;` leads to (this last block is
the only one allowed to back-jump to the beginning of the loop).
Bug: angleproject:4889
Change-Id: Ic59f4bf3e05fbf93cb5af85acd3bc4b0da8412af
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2957809
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
8b869a95
|
2021-06-13T01:09:27
|
|
Translator: Generate Ops for all built-in functions
EOpCallBuiltInFunction is removed in this change, as well as the
"op": "auto" property in builtin_function_declarations.txt. Instead,
gen_builtin_symbols.py automatically generates Ops for every built-in
function and generates the TOperator enum accordingly.
This simplifies SPIR-V code generation by allowing switches to be used
on operators instead of string comparisons.
Bug: angleproject:4589
Bug: angleproject:4889
Change-Id: Ia351524400b0e12a10a5572e27e9b88c6ec2e61c
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2958869
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
0297779c
|
2021-06-08T01:06:21
|
|
Vulkan: SPIR-V Gen: Handle invariant variables
The Invariant decoration is applied to variable declarations and block
members as necessary. When invariant is applied to a gl_PerVertex
built-in, it's applied to the appropriate member in
DeclarePerVertexBlocks.
Bug: angleproject:4889
Change-Id: Idaa8d4ab92f975381a0237d372f0a4044ff983e0
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2946116
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
af1eed2e
|
2021-06-10T01:55:59
|
|
Vulkan: Generate gl_FragColor/Data declarations in AST
gl_FragColor and gl_FragData are not available in Vulkan. Prior to this
change, their declaration as webgl_FragColor and webgl_FragData was done
in text. This change implements an AST transformation that declares a
normal fragment output variable and replaces all references to these
built-ins with those variables.
Bug: angleproject:4889
Change-Id: If224e089dec25e4aa580beb135e1be2890de7887
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2953042
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
69f2fb00
|
2021-03-08T10:49:31
|
|
Vulkan: Expose OES_geometry_shader extension
Modify the symbol table generation script to accept a list
of supported extensions. This allows for the EXT and
OES versions of the geometry shader extension to be exposed.
Test: angle_deqp_khr_gles31_tests --deqp-case=*geometry_shader*
Bug: angleproject:3571
Change-Id: Ia7127a03dbd3fce78957f0505d3ce0c9bab6cb15
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2765011
Commit-Queue: Mohan Maiya <m.maiya@samsung.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
d7aa0130
|
2021-04-26T16:56:15
|
|
Upstream Apple's direct-to-Metal backend: compile translator.
This change is meant to merge the translator changes from Apple's
direct-to-Metal backend. Taken from Kyle Piddington's CL:
https://chromium-review.googlesource.com/c/angle/angle/+/2857366/
The goal of this CL is to merge the translator code in a state that
compiles, but not to switch the Metal backend over to use this
translator backend yet.
Bug: angleproject:5505
Change-Id: I68a6354604498cd5fd1eb96c13fc56f3b38f2bd0
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2897536
Reviewed-by: Jonah Ryan-Davis <jonahr@google.com>
Commit-Queue: Jonah Ryan-Davis <jonahr@google.com>
|
|
|
c229ccfe
|
2021-05-18T15:51:12
|
|
Vulkan: SPIR-V Gen: Handle gl_PerVertex
If not declared by the shader, default gl_PerVertex is now explicitly
declared in AST prior to SPIR-V generation. The SPIR-V generation code
then doesn't need to declare them magically.
Bug: angleproject:4889
Change-Id: Icc593bc1ccc3162487bdbae7f66bc775d098778d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2905952
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
39ee3640
|
2021-05-11T22:27:08
|
|
Translator: Fix struct/uniform separation w.r.t to arrays
When a uniform of struct type was redeclared, the arrayness information
was not transferred.
Bug: chromium:1204861
Change-Id: Ic3f461a4aa349a73bc48568ee14b5418f0556238
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2889599
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
c5e344b1
|
2021-05-11T22:13:38
|
|
Translator: Fix local var inits vs struct/uniform separation
The declarations outside the global scope were being skipped in this
transformation, but if the initializer of a declaration referenced a
uniform struct that's being replaced, it should have been visited too.
Bug: chromium:1204861
Change-Id: Ie2a743126c785eb7d6e3542ba80faa4defc3509f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2889598
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
daeac238
|
2021-05-08T22:09:38
|
|
Translator: Ensure structs and blocks are uniquely defined
A new AST validation is added to ensure that the same TStructure or
TInterfaceBlock is not redundantly defined. This helps with SPIR-V
generation by allowing the id to be used as key in a hash map that looks
up the corresponding SPIR-V type id.
A bug is fixed where the Vulkan driver uniform declaration created two
identical declarations for ANGLEDepthRangeParams.
A number of other bugs are also fixed in this change, where if a
variable declaration is eliminated (for example due to constant folding,
or inactive interface variable removal) and it contained a struct
specifier, the struct declaration was also removed. OutputGLSLBase had
a hack where structs were declared on first encounter, which was
incorrect as the scope of the declaration could change. Those bugs are
fixed and this hack is removed.
Bug: angleproject:2733
Bug: angleproject:4889
Bug: angleproject:5936
Change-Id: I8e13748c0bf552ae8b052249282769a1f0775603
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2881942
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
d469dcb1
|
2021-05-06T15:13:42
|
|
Translator: remove image2DRect
There is no such thing as image2DRect in GLSL. imageRect (the right
type) is present, but the code was mistakenly using image2DRect.
Bug: angleproject:4889
Change-Id: I42bacd2ad8866c715af0d9711bebc95897607354
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2878095
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
cc56e1aa
|
2021-04-09T23:35:07
|
|
Vulkan: Remove special output of structs and keep them in AST
The code that separated out declaration of nameless structs from uniform
variable declarations is generalized to separate out declaration of all
structs from uniform variable declarations.
As a result, a pass that outputs the struct types at the top of the
shader is no longer necessary. The struct declarations are kept in the
AST to be output to GLSL as they normally would when not used in
conjunction with uniform variables.
After this change, the Vulkan and Metal translators no longer intermix
transformations and code generation; transformations are done first
entirely in AST, and code generation is done at the end.
Bug: angleproject:2461
Bug: angleproject:4889
Change-Id: Ieb4d3f7091845e50c3dc399603ab01182692521d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2818153
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
6ec55494
|
2021-04-08T22:02:50
|
|
Vulkan: Use AST transformation for gathering default uniforms
... instead of generating text. This will keep the tree in a consistent
state, as previously the references to uniform values where to
TVariable's whose declarations were removed.
Bug: angleproject:2461
Bug: angleproject:4889
Change-Id: I2ef3787dede371514a5f495f4cc8fc67ea45f451
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2815564
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
58eabfd7
|
2021-04-08T23:57:04
|
|
Translator: Validate consistent variable references
Some transformations change variable declarations. This validation
ensures that all references to said variables are appropriately replaced
as well.
Bug: angleproject:2733
Change-Id: I6c2873968eeed4cba66e70069f84eb69a1f77074
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2818140
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
776c6015
|
2021-02-26T00:00:57
|
|
Vulkan: Call glslang at compile time
With this change, the ANGLE translator immediately compiles the
generated GLSL into SPIR-V with glslang and discards the source. This
is in preparation for generating SPIR-V directly, by making the frontend
and backend already able to digest it.
This change also allows the expensive glslang calls to be parallelized,
improving the following perf test by about 20%:
LinkProgramBenchmark.Run/vulkan_compile_and_link_multi_thread
Previously, the test was run as such in the Vulkan backend:
Main Thread 1 Thread 2
Compile1 --->
Compile2 --------------------->
Translator Translator
<---
<---------------------
Link
glslang
for
shader1
glslang
for
shader2
Done
With this change, it is run as such:
Main Thread 1 Thread 2
Compile1 --->
Compile2 --------------------->
Translator Translator
glslang glslang
<---
<---------------------
Link
Done
glslang_wrapper_utils no longer interacts with glslang! A rename will
follow.
Bug: angleproject:4889
Change-Id: If4303e8ba0ba43b1a2f47f8c0a9133d0bee1a19a
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2721195
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
550f2a3e
|
2021-02-24T09:49:42
|
|
Vulkan: Shader support for EXT_shader_framebuffer_fetch_non_coherent
Translator can accept gl_LastFragData and 'inout' variable to gain
access to framebuffer attachment data. The Vulkan translator replaces
it with the SubpassInput type variable. Note that this works only for
the noncoherent version of the extension.
Bug: angleproject:5454
Test: *EXTShaderFramebufferFetchNoncoherent*.*
Change-Id: I392f84ee3ad3eb9fbd09d0b7ff83731a9a3f33f6
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2598060
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Mohan Maiya <m.maiya@samsung.com>
|
|
|
9f4b159b
|
2021-01-19T12:48:15
|
|
Translator: Avoid vector copies with multi-replacement
Turns push_backs into emplace_backs and changes the TIntermSequence
constructor argument to &&.
Bug: angleproject:5535
Change-Id: I640ce879b6ade48a28dea6385ebb7a95cb8304ff
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2636680
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
f0286e63
|
2021-01-19T12:38:46
|
|
Translator: Stop allocating TIntermSequence
The functions that take a TIntermSequence always copy out / Swap the
contents away. This change makes all TIntermSequences live on the stack
instead of being newed.
Bug: angleproject:5535
Change-Id: I942f1c5e57b00199d5308183f71bd9e18b0608bd
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2636679
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
ea71c6b6
|
2021-01-12T14:13:35
|
|
Vulkan: Emulate R32F images with R32UI
GL requires that imageAtomicExchange be supported for r32f formats.
However VK_FORMAT_FEATURE_STORAGE_*_ATOMIC_BIT is nearly unsupported
everywhere without some Vulkan extension that brings in unnecessary
support.
This GL feature is emulated by transforming the shader to use r32ui for
all images that originally specified r32f. floatToUintBits and
uintBitsToFloat is used to maintain correct usage of the image* builtin
functions.
Bug: angleproject:5535
Change-Id: Ie607089935d3283b3ffa054f4b4385b81fb8f53d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2635453
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
2b242f53
|
2021-01-17T14:24:53
|
|
Vulkan: Simplify RewriteCubeMapSamplersAs2DArray
This change is mostly done to remove an unnecessary helper in
ReplaceVariable.cpp.
Tested locally, as the bots all support ES3.1+. There are a few
failures, with or without this change.
Bug: angleproject:5556
Change-Id: I7bd2d4294c9dc164146d713db6b45cbc73e59c16
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2633437
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
186fe990
|
2021-01-15T15:49:30
|
|
Vulkan: Redo RewriteAtomicCounters
With MonomorphizeUnsupportedFunctionsInVulkanGLSL and
RewriteArrayOfArrayOfOpaqueUniforms transformations run, it is no longer
possible to encounter array of array of atomic counters, or have any
passed to functions. As a result, RewriteAtomicCounters is greatly
simplified. Additionally, it is no longer necessary to pass
binding/offset information for atomic counters around and they can use
constants. This change removes dependency on the
shaderStorageBufferArrayDynamicIndexing Vulkan feature.
Bug: angleproject:3726
Bug: angleproject:3881
Change-Id: Ia43092a668f60d009eccbbceeed5deaf105a5895
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2633687
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
e670fc72
|
2021-01-18T15:51:42
|
|
Implement shader compiler changes for Tessellation.
Numerous rule changes to support validating Tessellation Control and
Evaluation shaders. New per-patch inputs and output variable support.
Includes a new traverser step that validates barrier function calls.
Functionality changes upcoming in http://crrev.com/c/2568234
Bug: angleproject:3572
Change-Id: If8da1c21d30efa12c60ed0d6c3f8cf0b27e4c86f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2633936
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Mohan Maiya <m.maiya@samsung.com>
|
|
|
60015ff6
|
2021-01-14T02:03:07
|
|
Vulkan: Redo RewriteStructSamplers
This transformation is split into two. The first transformation solely
takes out the samplers out of structs, and potentially generates array
of array of samplers. A second transformation is added that takes any
array of array of opaque uniforms and flattens it.
A follow up change will simplify RewriteAtomicCounters which also
handles array of arrays (which is no longer possible), and removes
dependency on shaderStorageBufferArrayDynamicIndexing.
Bug: angleproject:2703
Bug: angleproject:3881
Bug: angleproject:4071
Bug: angleproject:4211
Change-Id: I352bb2bbe65ac49f4d7d753c0ba3160fa3cc925a
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2628138
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
dc99fc40
|
2021-01-13T00:21:41
|
|
Vulkan: Translator pass to monomorphize problematic functions
With array of array of samplers and images, we currently require the
shader*ArrayDynamicIndexing Vulkan features. With atomic counters, we
require the shaderStorageBufferArrayDynamicIndexing feature.
The above features are required to enable passing opaque uniforms to
functions. This change introduces a translator pass that monomorphizes
functions that receive atomic counters, or partially subscripted array
of array of samplers or images, etc by removing those arguments and
using the opaque uniform directly.
Follow up changes will include:
- Great simplification to RewriteStructSamplers, and removal of
RewriteStructSamplersOld. This will drop dependency to
shaderSampledImageArrayDynamicIndexing and
shaderStorageImageArrayDynamicIndexing.
- Great simplification to RewriteAtomicCounters. This will drop
dependency to shaderStorageBufferArrayDynamicIndexing.
- Emulation of imageAtomicExchange for r32f formats, but changing the
qualifier to r32ui.
Note that parts of RewriteStructSampler are obsolete with this change,
but will be refactored as a follow up.
Bug: angleproject:3881
Bug: angleproject:4071
Bug: angleproject:5535
Change-Id: Ifd1435b2a31ebf364815046886aeded60297da79
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2628127
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|
|
|
382bf288
|
2020-12-24T23:56:39
|
|
Organize AST transforms per backend
Most of the AST transforms are written as a workaround to an issue that
affects a single backend. This change identifies such transforms and
organizes them by backend. They are then only built if the respective
backend is.
Additionally, about half of the GL transforms are due to mac
workarounds, including the large RewriteRowMajorMatrices transform.
Mac-specific workarounds are additionally only built on said platform.
This change reduces the ANGLE binary size:
- 106KB in a Vulkan-only build on Linux
- 27KB in a GL-only build on Android (60KB on Linux)
Bug: chromium:1084580
Bug: chromium:1161513
Change-Id: I64b334332c0d4f848756c6538af0d8d96864c7e9
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/2601346
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Tim Van Patten <timvp@google.com>
|