|
25390156
|
2025-08-21T00:13:19
|
|
Suppress unsafe buffers on a file-by-file basis in src/ [1 of N]
In this CL, we suppress many files but stop short of actually
enabling the warning by not removing the line from the
unsafe_buffers_paths.txt file. That will happen in a follow-on
CL, along with resolving any stragglers missed here.
This is mostly a manual change so as to familiarize myself with
the kinds of issues faced by the Angle codebase when applying buffer
safety warnings.
-- Re-generate affected hashes.
-- Clang-format applied to all changed files.
-- Add a few missing .reserve() calls to vectors as noticed.
-- Fix some mismatches between file names and header comments.
-- Be more consistent with header comment format (blank lines and
trailing //-only lines when a filename comment adjoins license
boilerplate).
Bug: b/436880895
Change-Id: I3bde5cc2059acbe8345057289214f1a26f1c34aa
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6869022
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
beae1b4a
|
2025-06-19T00:00:00
|
|
Metal: Remove obsolete workarounds
Removed Intel-specific Metal workarounds
that are not used on macOS 12 and later.
Removed a redundant version check.
Bug: angleproject:427600175
Change-Id: I34c7e53108f7e030512c9436ab2b9ae38ad17946
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6672312
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
4d478713
|
2025-04-03T08:10:40
|
|
Revert "Metal: Make StateCache descriptors hash consistent"
This reverts commit 4334125be506e0c2c5dde4cfc29b31a7ff2bdec6.
Reason for revert: Pipeline creation failures.
Bug: chromium:407310859
Original change's description:
> Metal: Make StateCache descriptors hash consistent
>
> Various descriptors would compare equality and do hashing based on the
> underlying storage. The storage contains alignment padding and
> bitfields. The contents of these are not defined, even when the
> constructors tried to memset and memcpy.
>
> The various hash and compare functions also seemed to be missing some
> elements, like rasterSampleCount.
>
> Missing == on fields and hashing unused memory may lead to inconsistency
> where a==b is true but hash(a) != hash(b).
>
> Fix by:
> Remove the memset/memcpy and write out the operator== and
> hash() consistently.
>
> Initialize the members to their default values.
>
> Use uint32_t : 1 instead of bool if the struct tries to optimize for
> size, since mixing types stops the packing.
>
> Use uint32_t to get the alignment to uint32_t, which is mostly what
> is expected.
>
> Use uint32 someVar : N where N is payload bits + padding bits for
> the last member. This way the assignment clears the padding bits.
> The goal os to get the operator== to compare the padding bits, which
> lets the compiler elide the memberwise compares and use word-wise
> compares where it makes sense.
>
> This is an attempt to fix rare crashes related to map inserts
> and lookups. std::unordered_map will crash if operator== is not
> consistent with hash.
>
> Bug: angleproject:403372465
> Change-Id: I0a3a2ff327ac9f65e7d41bc9585cda54ce93ab86
> Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6355828
> Commit-Queue: Kimmo Kinnunen <kkinnunen@apple.com>
> Auto-Submit: Kimmo Kinnunen <kkinnunen@apple.com>
> Reviewed-by: Alexey Knyazev <lexa.knyazev@gmail.com>
> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Bug: angleproject:403372465
Change-Id: Id111c0728b603776fd8085fd395819bb5aa14722
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6428466
Commit-Queue: Quyen Le <lehoangquyen@chromium.org>
Reviewed-by: Quyen Le <lehoangquyen@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
4334125b
|
2025-03-14T16:12:32
|
|
Metal: Make StateCache descriptors hash consistent
Various descriptors would compare equality and do hashing based on the
underlying storage. The storage contains alignment padding and
bitfields. The contents of these are not defined, even when the
constructors tried to memset and memcpy.
The various hash and compare functions also seemed to be missing some
elements, like rasterSampleCount.
Missing == on fields and hashing unused memory may lead to inconsistency
where a==b is true but hash(a) != hash(b).
Fix by:
Remove the memset/memcpy and write out the operator== and
hash() consistently.
Initialize the members to their default values.
Use uint32_t : 1 instead of bool if the struct tries to optimize for
size, since mixing types stops the packing.
Use uint32_t to get the alignment to uint32_t, which is mostly what
is expected.
Use uint32 someVar : N where N is payload bits + padding bits for
the last member. This way the assignment clears the padding bits.
The goal os to get the operator== to compare the padding bits, which
lets the compiler elide the memberwise compares and use word-wise
compares where it makes sense.
This is an attempt to fix rare crashes related to map inserts
and lookups. std::unordered_map will crash if operator== is not
consistent with hash.
Bug: angleproject:403372465
Change-Id: I0a3a2ff327ac9f65e7d41bc9585cda54ce93ab86
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6355828
Commit-Queue: Kimmo Kinnunen <kkinnunen@apple.com>
Auto-Submit: Kimmo Kinnunen <kkinnunen@apple.com>
Reviewed-by: Alexey Knyazev <lexa.knyazev@gmail.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
d4083c79
|
2025-01-31T15:06:36
|
|
Metal: Make ObjCPtr available to all modules
Move ObjCPtr implementation to src/common/ObjCPtr.h,
angle::ObjCPtr so that it's available also outside libANGLE.
Bug: angleproject:393263506
Change-Id: I8ecd5632c7ae33ef4b409fc820fad229e91bc0ab
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6219318
Auto-Submit: Kimmo Kinnunen <kkinnunen@apple.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Kimmo Kinnunen <kkinnunen@apple.com>
|
|
|
4e5f2ca3
|
2025-01-31T14:30:05
|
|
Metal: Rewrite AutoObjCPtr as ObjCPtr
Rewrite without WrappedObject base class, the base class only
convolutes things and does not help. This way the class can be
moved upwards in the software layers in subsequent commits.
Fix the class so that it can be used with ObjCPtr<T> and ObjCPtr<id<P>>
signatures, to resemble other smart pointers.
Remove the Auto prefix, ObjCPtr should be relatable to other smart
pointers such as unique_ptr.
Bug: angleproject:393263506
Change-Id: Ia5a5728514b7f9182c2ec8380c04716e8b9d2351
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6219317
Auto-Submit: Kimmo Kinnunen <kkinnunen@apple.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Kimmo Kinnunen <kkinnunen@apple.com>
|
|
|
bee34ea7
|
2025-01-30T18:39:00
|
|
Metal: Remove redundant AutoObjCObj template alias
Some members were held with AutoObjCObj<T> and some with
AutoObjCPtr<T *>. Use the latter.
Bug: angleproject:393263506
Change-Id: I9d8d56e9ffe3281d5e9087de7959249cbb4b9bd4
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6216872
Auto-Submit: Kimmo Kinnunen <kkinnunen@apple.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
5325904b
|
2025-01-29T14:43:35
|
|
Metal: Simplify error checking
Use normal ANGLE_CHECK, ANGLE_CHECK_GL_ALLOC, ANGLE_CHECK_GL_MATH
when appropriate.
Use ANGLE_MTL_CHECK for checking Metal NSError return value.
Remove mtl::ErrorHandler::handleError variant that responds to NSError.
Uses "Internal error." that occur due to implementation bugs. Binary
size is already expended with __FILE__ et al to disambiguate the
location. The descriptions were not actionable for the API client and
sometimes wrong cut-and-paste.
Bug: angleproject:392938089
Change-Id: If9525d3d1610d5bc757855053600d78d2a1526c3
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/6211841
Auto-Submit: Kimmo Kinnunen <kkinnunen@apple.com>
Commit-Queue: Kimmo Kinnunen <kkinnunen@apple.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
d550d96f
|
2024-09-30T00:00:00
|
|
Metal: Remove allowSamplerCompareLod feature
* Apple2 GPUs do not support setting texture sampler compare state
via Metal API. As a result, all sample_compare functions including
those without LOD options do not work on Apple2 GPUs. This feature
was attempting to incorrectly emulate a subset of sample_compare
functions with a hard-coded compare state.
There are no plans to support shadow samplers on Apple2 GPUs, as it
would require non-trivial emulation, so this feature is not needed.
* Metal on macOS 10.15 only supports constant zero as
the level parameter for sample_compare functions.
Ignoring the requested level and using zero on old OS versions is a
better fallback than dropping the parameter altogether because many
applications set the level to zero anyway.
Bug: angleproject:365066518
Bug: angleproject:368059227
Change-Id: I2cdc0bda5842842d62ad5663427da7f1f37bddda
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5907619
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
|
a36c459e
|
2024-08-20T00:00:00
|
|
Metal: Remove redundant symbol redeclarations
MTLBlitOptionRowLinearPVRTC
- macOS 11.0 (min compile 12.0)
- iOS/tvOS 9.0 (min compile 17.0)
MTLBarrierScope
- macOS 10.14 (min compile 12.0)
- iOS/tvOS 12.0 (min compile 17.0)
MTLBarrierScopeRenderTargets
- Moved compile guards to call sites
MTLRenderStages, MTLRenderStageVertex, MTLRenderStageFragment
- macOS 10.13 (min compile 12.0)
- iOS/tvOS 10.0 (min compile 17.0)
Bug: angleproject:360147119
Change-Id: I7dac46accde63be486a7de918284bbd6e8367a4e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5806896
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Kimmo Kinnunen <kkinnunen@apple.com>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
|
5cf94ffa
|
2024-08-20T10:41:40
|
|
Metal: Use pipeline rasterSampleCount
Use MTLRenderPipelineDescriptor rasterSampleCount instead of
sampleCount. This avoids the use of the deprecation macro.
Availability: iOS 8.0 (min: 15.0), mac Catalyst 13.1 (min: 16.5),
macOS 10.11 (min: 10.15), tvOS 11.0 (min: 15.0), visionOS 1.0+
Bug: angleproject:360147118
Change-Id: I7009a7d7b39ecb763e23f4b97441fb45a64a891d
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5797134
Auto-Submit: Kimmo Kinnunen <kkinnunen@apple.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Alexey Knyazev <lexa.knyazev@gmail.com>
Commit-Queue: Kimmo Kinnunen <kkinnunen@apple.com>
|
|
62d9f847
|
2024-07-17T13:27:07
|
|
Speed up UniformBuffer support on Metal
If the sizes are different but all fields match
then only padding remains. So, as long as the
buffer can safely handle the offset and size of
usage, allow drawing without converting the
uniform buffer.
Fixed: angleproject:348270693
Change-Id: I91196ca9839cf064b573d912f125ff4879a85b02
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5719193
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Quyen Le <lehoangquyen@chromium.org>
Auto-Submit: Gregg Tavares <gman@chromium.org>
Reviewed-by: Quyen Le <lehoangquyen@chromium.org>
|
|
|
8ed4d427
|
2024-07-11T00:00:00
|
|
Metal: Remove link tasks dependency on context
Bug: angleproject:351165323
Change-Id: If9e5d141cb15a6fc6df8a5fd9a2cfe70efe19dfc
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5701902
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
|
|
|
c5ab1ceb
|
2024-07-11T00:00:00
|
|
Metal: Remove library creation dependency on ContextDevice
Bug: angleproject:351165323
Change-Id: Idc560a75e569fe917685bb247625d7df8472af46
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5698885
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
|
b7d4c5d1
|
2024-07-04T00:00:00
|
|
Metal: Disable fast math if isinf or isnan are used
Fixed: angleproject:42266714
Fixed: angleproject:42267063
Change-Id: Ibf6508ffc7d458a6b91b9218b2b488b6aa5b1001
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5678416
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
|
18bb6e54
|
2024-06-06T00:00:00
|
|
Metal: Support non-layered 2D array shader images
Fixed: angleproject:8125
Change-Id: Id148c09d948487bb677c4a3af746e6ea151d84a7
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5604225
Reviewed-by: Quyen Le <lehoangquyen@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
4109a90e
|
2024-04-16T17:31:11
|
|
LinkedUniform: avoid frequent GLenum -> index conversion
Certain functions such as getElementComponents() are frequently called
in driver_overhead benchmark, causing repeated GLenum -> index
conversion of the uniform type which shows up in profiling
(driver_overhead_2 trace)
Change LinkedUniform.pod.type to LinkedUniform.pod.typeIndex storing the
UniformTypeInfo index with conversion helpers.
Bug: b/335295728
Change-Id: Iae5cd58f4e2703589d23b8e52991fc4b97c5fb08
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5458741
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Roman Lavrov <romanl@google.com>
Reviewed-by: Charlie Lao <cclao@google.com>
|
|
0c4d6446
|
2024-01-24T10:38:45
|
|
Rework uniform block <-> uniform buffer mapping
In GLES, the shader declares which buffer binding a block (uniform,
storage or atomic counter) is bound to. For example:
layout(binding = 1) uniform ubo0 { ... };
layout(binding = 2) uniform ubo1 { ... };
layout(binding = 1) uniform ubo2 { ... };
In the above, ubo0 and ubo2 use data from the buffer bound to index 2
(through glBindBufferRange), while ubo1 uses data from the buffer bound
to index 1. For uniform blocks in particular, omitting the binding
is allowed, in which case it is implicitly bound to buffer 0.
GLES allows uniform blocks (and only uniform blocks) to remap their
bindings through calls to glUniformBlockBinding. This means that the
mapping of uniform blocks in the program (ubo0, ubo1, ubo2) to the
buffer bindings is not constant. For storage blocks and atomic counter
buffers, this binding _is_ constant and is determined at link time.
At link time, the mapping of blocks to buffers is determined based on
values specified in the shaders. This info is stored was stored in
gl::InterfaceBlock::binding (for UBOs and SSBOs), and
gl::AtomicCounterBuffer::binding. For clarity, this change renames
these members to ...::inShaderBinding.
When glUniformBlockBinding is called, the mapping is updated. Prior to
this change, gl::InterfaceBlock::binding was directly updated, trumping
the mapping determined at link time. A bug here was that after a call
to glProgramBinary, GL expects the mappings to reset to their original
link-time values, but instead ANGLE restored the mappings to what was
configured at the time the binary was retrieved.
This change tracks the uniform block -> buffer binding mapping
separately from the link results so that the original values can be
restored during glProgramBinary. In the process, the support data
structures for tracking this mapping are moved to ProgramExecutable and
the algorithms are simplified. Program Pipeline Objects maintain this
mapping identically to Programs and no longer require a special and more
costly path when a buffer state changes.
This change prepares for but does not yet fix the more fundamental bug
that the dirty bits are tracked in the program executable instead of the
context state, which makes changes not propagate to all contexts
correctly.
Bug: angleproject:8493
Change-Id: Ib0999f49be24db06ebe9a4917d06b90af899611e
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5235883
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
6d9f0aee
|
2023-11-30T00:00:00
|
|
Metal: Ensure helper invocations when derivatives are used
Set sample coverage mask to ensure that
fragment shader helper invocations are
performed when derivatives are used.
Fixed the following tests on Apple silicon:
dEQP-GLES2.functional.texture.mipmap
.2d.projected.nearest_linear_repeat
.2d.projected.linear_linear_repeat
.cube.projected.linear_nearest
.cube.projected.linear_linear
dEQP-GLES3.functional.texture.mipmap
.2d.projected.nearest_nearest_mirror
.2d.projected.linear_nearest_clamp
.2d.projected.linear_nearest_repeat
.2d.projected.linear_nearest_mirror
.2d.projected.nearest_linear_repeat
.2d.projected.linear_linear_clamp
.2d.projected.linear_linear_repeat
.3d.projected.nearest_nearest_mirror
.3d.projected.linear_nearest_clamp
.3d.projected.linear_nearest_repeat
.3d.projected.nearest_linear_clamp
.3d.projected.nearest_linear_mirror
.3d.projected.linear_linear_clamp
.3d.projected.linear_linear_repeat
Fixed: angleproject:8443
Fixed: angleproject:7023
Change-Id: I13fc52532b94de098dcf3e9f5f02d48a6c8913c7
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/5089131
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Alexey Knyazev <lexa.knyazev@gmail.com>
|
|
34e5620b
|
2023-11-01T17:52:34
|
|
Metal: Reuse UBO conversion buffer only if size matches
BufferMtl::getUniformConversionBuffer will recycle an existing buffer if
the request has the same buffer index and offset that is a multiple of
existing block size. This disregards the GL std size of the request
leading to incorrect allocation of blocks in the conversion buffer. The
fix is to create a new conversion buffer when stdSize !=
buffer.uniformBufferBlockSize.
Bug: angleproject:8393
Change-Id: Ife87309e66f65baa7f636b734a5586d0ae3b9768
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4995450
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Quyen Le <lehoangquyen@chromium.org>
|
|
|
e076d6cf
|
2023-10-17T11:39:10
|
|
Move invalid uniform protection to the frontend.
The frontend potentialy indexes into mUniformLocations with invalid
uniform locations while validation is disabled too. Move the validation
from the Metal backend to the frontend.
Bug: chromium:1484878
Change-Id: I92bc43aa28cfa26d601bb28f318860375f618608
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4947652
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
d55893cd
|
2023-09-27T16:40:58
|
|
Issue 1484512: Out of bounds access to Metal uniform blocks
Fixed conversion offset for Uniform buffers: Dynamic buffers
were being allocated and copied from the 'updateBufferSubData'
offset, than from the previously allocated offset. This meant that when
creating a new dynamic buffer to hold the metal uniforms,
ANGLE didn't re-convert blocks from the initial source offset,
nor did we allocate enough space to read those blocks.
Important use cases to test include any UBO-heavy sites.
Bug: chromium:1484512
Change-Id: Ia9f505e3df871a401d40682d737019bd5d3f443f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4899369
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Kyle Piddington <kpiddington@apple.com>
|
|
|
2c0f9d20
|
2023-09-26T11:50:54
|
|
Metal: Protect against invalid locations in setUniformImpl
Chrome disables validation and a common bug is that the wrong program
is used during uniform setting in Skia. Make this bug less catastrophic
while the root cause is investigated.
Bug: chromium:1484878
Change-Id: I521e99cdf6cebc56775e484fa8d61ad18f506479
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4886384
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
8fcd4a50
|
2023-09-19T14:08:14
|
|
Cleanup POD struct usage to make them more consistent
In recent months, I have made many CLs to wrap various data structures
into a trivially copy-able struct. Some of them uses different
terminology. This CL replaces all of these to use a consistent name
"pod" in the struct, and "mPod" in a class.
This CL also turns ActiveVariable methods into macros to remove the code
duplication.
This CL also moves ProgramInput/ProgramOutput struct implementations
from Program.cpp to ProgramExecutable.cpp
Bug: b/275102061
Change-Id: Ia2a4210a9ea633f3d323bebe674ee74f8b90b363
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4877335
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Roman Lavrov <romanl@google.com>
|
|
|
9ca025d2
|
2023-09-18T15:50:48
|
|
Flatten BufferVariable/ShaderVariableBuffer/InterfaceBlock struct
InterfaceBlock inherits from ShaderVariableBuffer, ShaderVariableBuffer
is not a trivially copyable struct, this made InterfaceBlock not
trivially copyable. InterfaceBlock is being used by some app traces for
uniform blocks. BufferVariable inherits from sh::ShaderVariable which is
very complicated and not trivially copyable. This CL flattens all of
these three structs to simple structs without inheritance, and wraps all
trivially copyable data into one POD struct, thus load/save are cheaper.
Bug: b/275102061
Change-Id: I96f89176ce3d3131cb1d3ea3280c3c36c257560f
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4874610
Commit-Queue: Charlie Lao <cclao@google.com>
Reviewed-by: Roman Lavrov <romanl@google.com>
Reviewed-by: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
eb0d5997
|
2023-09-15T16:41:13
|
|
Move set/get uniform machinery to ProgramExecutable
This is done because some uniforms are internally added by the compiler
(draw ID, base vertex, and base instance) and are automatically set **on
the installed executable**.
This change fixes scenarios where a draw is done after a program has
failed a relink, and therefore is unable to correctly set the uniforms
(as it does not have access to the executable that is installed).
It also fixes draws that use those uniforms in a PPO.
Bug: angleproject:8297
Change-Id: Id74b4984b88aa09b5b81be1c91412d6c91711136
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4864693
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
68bfa1ed
|
2023-08-22T22:02:15
|
|
Support for link to be entirely parallelized
The link job is split as such:
- Front-end link
- Back-end link
- Independent back-end link subtasks (typically native driver compile
jobs)
- Post-link finalization
Each step depends on the previous. These steps are executed as such:
1. Program::link calls into ProgramImpl::link
- ProgramImpl::link runs whatever needs the Context, such as releasing
resources
- ProgramImpl::link returns a LinkTask
2. Program::link implements a closure that calls the front-end link and
passes the results to the backend's LinkTask.
3. The LinkTask potentially returns a set of LinkSubTasks to be
scheduled by the worker pool
4. Once the link is resolved, the post-link finalization is run
In the above, steps 1 and 4 are done under the share group lock. Steps
2 and 3 can be done in threads or without holding the share group lock
if the backend supports it. Step 2 is not yet made independent of the
Context on some backends, and a frontend feature is used to make that
step either run on the main thread or as a worker thread.
Bug: angleproject:8297
Change-Id: I12f1e6bbaf365543dfcac969e166e0b5aa622104
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4808191
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Shahbaz Youssefi <syoussefi@google.com>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
99d39241
|
2023-09-07T16:48:22
|
|
Metal: Increase the size of the default uniform block
Use a buffer pool per shader type and size it to be a nice multiple of
the block that will be allocated for each uniform update.
Set the buffer pool size to unbounded. When the pool runs out of
buffers it currently does a full CPU/GPU sync which is catastrophic
for performance when uniforms are updated every draw call.
Bug: angleproject:8301
Change-Id: I7f826feb4e766fd1cf1b2db01954af7089e3fe90
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4850950
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
d664543f
|
2023-09-01T23:11:59
|
|
Metal: Move setupDraw and associated code to executable
Last bits of state left in ProgramMtl that are now moved to
ProgramExecutableMtl, and now ContextMtl does not reference the program
at all.
This fix was necessary for a follow up change that allows the program to
be modified while the executable is installed, and the metal backend was
crashing after a failed relink due its direct access to the program.
Bug: angleproject:8297
Change-Id: Iadf623bf6baf612767ff372adee2e7f4eeedb593
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4833624
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
|
|
|
6ff209bc
|
2023-08-30T00:02:30
|
|
Clean up InfoLog usage during link
The info log doesn't really belong to ProgramExecutable; it belongs to
ProgramState. However, it is placed there for convenience since many
functions access it.
This change cleans up usage of InfoLog so the one in ProgramExecutable
is consistently used, but also that is turned into a reference to
ProgramState's InfoLog.
This is necessary for a follow up change that restores the previous
executable on link failure (and would thus otherwise lose the info log
of the failing link).
Bug: angleproject:8297
Change-Id: I088408e3fce9ebb35b1ec4ad3dc599bdb90bf5c5
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4825624
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Yuxin Hu <yuxinhu@google.com>
|
|
|
ae67a24b
|
2023-08-25T14:51:50
|
|
Metal: Move program state to ProgramExecutableMtl
Bug: angleproject:8297
Change-Id: I1d13f7aee1ff5b0ce799b486d8a57c83c4481983
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4812047
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
bb135f0e
|
2023-08-24T15:29:11
|
|
Make ProgramExecutableImpl managed by ProgramExecutable
This change allows both parts of the program executable to be safely
backed up and swapped on link.
Bug: angleproject:8297
Change-Id: I17e4b6c05e4e481a66a227d6047dbf943d2c2603
Reviewed-on: https://chromium-review.googlesource.com/c/angle/angle/+/4812138
Commit-Queue: Shahbaz Youssefi <syoussefi@chromium.org>
Reviewed-by: Charlie Lao <cclao@google.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|