1) Append "Shadow" to samplers representing depth textures.
2) Sampling a depth texture returns f32, not vec4<f32>
3) Sampling a depth texture requires a Dref parameter, so we must
generate one if none is provided.
4) GLSL requires Dref to be appended to the texture coordinates vector,
*unless* it's a samplerCubeArrayShadow, since this would require vec5.
In that case, it's passed as a separate parameter.
5) GLSL's textureGather() with a depth sampler always requires a refZ
parameter, so provide zero to emulate WGSL's compare-less textureGather().
6) texelFetch() does not support depth textures, so this will have to be
validated out.
7) textureOffset() does not support sampler2DArrayShadow in GLES, so this will
have to be validated out.
Bug: tint:1298
Change-Id: Idaebe89cac6c1ec97c50a361b1d3aa3b84fb6c12
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78760
Reviewed-by: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Stephen White <senorblanco@chromium.org>
Split IsAnyOf() into log(n) stages, where each stage performs a hashcode
check.
Previously there was a single hash test across the bitwise-or of all the
types being considered. If this passed, then each type would be tested
with Is<T>() individually. With this change, the list of types will be
recursively split into two, which each block hash-code checked. This is
repeated until we reach fewer than 4 types to check, where the test
decays to using Is<T>() for each type.
Also renamed `combined_hashcode` to `full_hashcode`, and used the term
CombinedHash for new helpers that bitwise-or the hashes from a number
of types.
Bug: tint:1383
Change-Id: Id056b9f7a9792430bd75ce554cb5fe73221ca4c7
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78580
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Kokoro: Ben Clayton <bclayton@google.com>
When converting a for-loop to a loop, we were not cloning the for-loop's
body, but rather the statements within it. This worked fine, except if
we also hoisted a variable to a let within that body, which requires the
body to be cloned for the 'insert before' to work. This change clones
the for-loop body, which fixes the problem, but introduces a block in
the destination AST, which is ugly, but not incorrect.
Bug: tint:1300
Change-Id: I478244d87f8cf58837102004242ba1c835e21710
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78821
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Ben Clayton <bclayton@google.com>
Commit-Queue: Antonio Maiorano <amaiorano@google.com>
This has been deprecated since M97.
Fixed: tint:1214
Change-Id: I970898f2ae8baf1916e2f8d43230d8b724641730
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78785
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: James Price <jrprice@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
It's too easy to copy diagnostics around and lose track of Source::File
ownership. Ideally we'd place the shared_ptr on the Source, but Sources
are copied _very_ frequently, and we'd lose a huge amount of
performance. Typically, Source::Files are owned externally. The only
time we really need to hold a shared_ptr to these is when a Source::File
is generated by an ICE, as the File points to the C++ source file that
raised the error.
Bug: chromium:1292829
Bug: tint:1383
Change-Id: I2706de8775bc3366115865b5a94785c0b2fefaae
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78782
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
The documentation for setThreadgroupMemoryLength states:
length - The size of the threadgroup memory, in bytes. Must be a multiple of 16 bytes.
This is the likely cause of a number of CTS flakes for 'webgpu:shader,execution,zero_init:compute,zero_init:storageClass="workgroup";*'
Bug: dawn:1277
Change-Id: Ib8c271fccacac6e68cdf0ddb5c6b5a41756173b5
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/78801
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Ben Clayton <bclayton@chromium.org>
As part of the tint -> dawn merge.
Bug: tint:1275
Change-Id: I0ebb8afe9971c21ca90f90f246edcfde01d5f4bc
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/78800
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
- WriteBuffer to 1D textures
- copyBufferToTexture to 1D textures
- copyTextureToBuffer from 1D textures
- copyTextureToTexture with 1D textures
Test coverage will be provided by the CTS. At the time this
commit is up for review, the validation tests have been change
to test 1D texture copies comprehensively, except for copies
between 1D textures and textures of other dimensions.
Bug: dawn:814
Change-Id: I91a278d9c113a06a108fecb124aa21fea5e691b3
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/64544
Reviewed-by: Loko Kung <lokokung@google.com>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Don't terminate on first error. Sleep a bit and try again.
Post a message to a gerrit change if it cannot be built. The fact the PS has a message from perfmon will prevent it from retrying the same change.
Remove trailing newlines from log.Printf() messages, they're automatically added.
Bug: tint:1383
Change-Id: I78a627c53c492e7da33a74470d5a064e90a7a753
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78783
Auto-Submit: Ben Clayton <bclayton@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Previously the conversion lost precision because of the repeated
float additions and rounding of float to uint. It is reworked to operate
using only integers, giving a better precision (3e-5 instead of 2e-3)
for all periods between 1ns and 2^16ns (which all GPUs seem to fall in
to).
The QueryHelper test is reworked to test multiple periods to provide
better certainty that the maths in the conversion is correct.
Bug: dawn:1250
Change-Id: I43703bb3a40c4e362d78126e3bf0d830690bc1d7
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/77780
Reviewed-by: Loko Kung <lokokung@google.com>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
The two paths were duplicated with extremely similar code.
Unifying them will help implement the buffer<->1D texture
copies by having a single place to change.
Bug: dawn:814
Change-Id: Id574bf62fc85e5e72ab54ba5066e52ff37000e87
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/78723
Reviewed-by: Loko Kung <lokokung@google.com>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
The SPIR-V cleanup transforms were only being run if the
DecomposeStridedMatrix transform needed to be run - despite
the RemoveUnreachableStatements transform needing to be run, always.
Change-Id: I5dc28a79200ab67c96fb793980412a5632e26026
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78780
Reviewed-by: James Price <jrprice@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
The current directory and namespace collides quite spectacularly with
the google benchmark include directory and namespace.
This becomes very hard to work around when there's a .cc file in src/
that wants to include "benchmark/benchmark.h", as MSVC appears to
resolve this to the relative path, while GCC and Clang resolve to the
compiler specified include directory.
Bug: tint:1383
Change-Id: Icc8891718d1d8a1b55c2ac4b2bb1487e8d09e629
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78740
Kokoro: Kokoro <noreply+kokoro@google.com>
Auto-Submit: Ben Clayton <bclayton@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Commit-Queue: Antonio Maiorano <amaiorano@google.com>
SPIRV-Tools was updated to fix a typo (preceeded->preceded) and the
string matching of these tests started failing.
Bug: tint:1406
Change-Id: If7affbf2e34e4f8d3e929c38ecdec7e3a624ee19
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78722
Reviewed-by: Ben Clayton <bclayton@google.com>
Auto-Submit: Corentin Wallez <cwallez@chromium.org>
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Without this, users of ExternalTextureBase will try to instantiate the
default destructor, which will cause a call to ~Ref<BufferBase> which
will fail when Buffer.h isn't included.
Fixes the roll of Dawn in google3.
Bug: None
Change-Id: Ib88d4338033468779f416973e5808910eb4ef110
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/78721
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Auto-Submit: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Ben Clayton <bclayton@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
VK_PIPELINE_STAGE_2_NONE_KHR
was renamed to
VK_PIPELINE_STAGE_2_NONE
in https://github.com/KhronosGroup/Vulkan-ValidationLayers/pull/3705
Removing the comma in the end of string will make it catch both cases.
Bug: chromium:1292106
Change-Id: Idca97cc6b2d53476b7d108595e329e7c287d086f
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/78640
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Auto-Submit: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Move builtin_to_string() and builtin_type() to
the CanonicalizeEntryPointIO transform. Use the former to
rename entry point IO variables to the gl_ names, and the latter
to cast values to the correct type.
Change-Id: Iddfad574ddd660ff1bfd89a399a001b967b6b67e
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78380
Reviewed-by: James Price <jrprice@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Stephen White <senorblanco@chromium.org>
Readonly depth/stencil attachment has already been supported on
D3D12, Metal and Vulkan. Its implement on GL and GLES will be
at post V1 via WebGPU-compat.
Bug: dawn:485
Change-Id: I1c13232f405b10b36f66c1c452449a864ca8fedc
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/78441
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Yunchao He <yunchao.he@intel.com>
The diff format changed.
Change-Id: I13930e0aeb450384d95d71460f9623adb5c1b192
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78581
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Kokoro: Ben Clayton <bclayton@chromium.org>
Commit-Queue: Ben Clayton <bclayton@chromium.org>
The transform was not correctly inserting the intrinsic call after array element types.
Fixed: chromium:1290107
Change-Id: I7199d1846cb98305d789cf0bc362eb5872d9b917
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78542
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: James Price <jrprice@google.com>
When TINT_BUILD_WGSL_WRITER is enabled.
This allows printing of the AST for debugging purposes.
Change-Id: I92b5911c16cb1e5fd22e81def00de33e9257f575
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78541
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: James Price <jrprice@google.com>
In recent optimizations, this got broken: when enabled the input program was printed over and over again, instead of the current transformed program.
Change-Id: I0c5acaf40a88ac759d0114f85febc052d577a698
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78540
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
As part of the tint -> dawn merge.
Bug: dawn:1275
Change-Id: Ice0c9d2f03f6d7e96471cf8398aecd16273c833f
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/78400
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
Initializes a uniform buffer on ExternalTexture objects that holds data
used by in shaders. Includes modifications to some Mock tests, which
should be using mock external texture objects instead of real ones.
Bug: dawn:1082
Change-Id: I34c02eadaf38aebf22630ac99098e0637ca1279c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/78240
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Commit-Queue: Brandon1 Jones <brandon1.jones@intel.com>
The `std::string file_path` carried around by every Source was very expensive for heap allocations.
Push this into the Source::File.
Bug: tint:1383
Change-Id: Id9e3bdef1cf24aea5f3c83f348c05f5cf8ef4bbb
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78321
Reviewed-by: Antonio Maiorano <amaiorano@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Kokoro: Ben Clayton <bclayton@google.com>
This doesn't exist in WGSL, and we don't have a supported path to
implement this in MSL.
This was preventing "image" from being used as an identifier.
Change-Id: I724e46c866a4dd488f45fbc1215ef23a67355a78
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/78280
Auto-Submit: James Price <jrprice@google.com>
Kokoro: Kokoro <noreply+kokoro@google.com>
Reviewed-by: David Neto <dneto@google.com>
Commit-Queue: David Neto <dneto@google.com>
Dawn Autoroller
Stephen White
Ben Clayton
Antonio Maiorano
Corentin Wallez
Yuly Novikov
Yunchao He
Brandon Jones
James Price