This CL cleans up some interfaces and methods in the MSL generator and
generator impl classes.
Bug: tint:8
Change-Id: I9aabeb1ecb69baeaa9cb6bbc15e313806923dddc
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/25163
Reviewed-by: David Neto <dneto@google.com>
This CL adds support for handling uniform data. Currently the uniform is
added to a buffer where the number is the binding value. This will need
to be updated to accept the correct mapping from the embedder.
Bug: tint:8
Change-Id: Icccccbe599a9555defa6136e384745f4093df020
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/25104
Reviewed-by: David Neto <dneto@google.com>
Old way:
- struct decorated with BufferBlock
- Uniform storage class
New way
- struct decorated with Block
- StorageBuffer storage class
Also fixes the result type for an access chain.
Bug: tint:99
Change-Id: I2324ba94bb19b369d206313de798bdfec6099fe0
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24605
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Set apiVersion to Vulkan 1.0 if the instance only supports
Vulkan 1.0. Otherwise set apiVersion to Vulkan 1.2, treat
1.2 as the highest API version dawn targets.
Bug: dawn:426
Change-Id: I322eaa0a93a518df36b86717c2ed5a98c5d056ea
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/25065
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Xinghua Cao <xinghua.cao@intel.com>
This patch adds the check and implementations of buffer lazy
initialization before CopyBufferToTexture().
The support of buffer lazy initialization before CopyTextureToBuffer()
is much more complicated than what we do for CopyBufferToTexture(), so
we decide to put it in another CL instead of writing them together with
CopyBufferToTexture().
BUG=dawn:414
TEST=dawn_end2end_tests
Change-Id: I45fdcdde2c9a0dafff23623815fc35c877990ef1
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/25140
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
Reviewed-by: Austin Eng <enga@chromium.org>
The CL that added offset/size args landed about the same time
as the CL that adding zero init mapping tests (without the args).
Tbr=cwallez@chromium.org
Bug: dawn:445, dawn:414
Change-Id: I388399d425c3a3c12be79b160686caaf3d81a89c
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/25162
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Austin Eng <enga@chromium.org>
This should not be necessary as the command pool should own the
memory if it was created without the creation flag
VK_COMMAND_POOL_CREATE_RESET_COMMAND_BUFFER_BIT. However, some drivers
leak memory if vkFreeCommandBuffers is not called.
Bug: chromium:1082181
Change-Id: Ia437fc17b2a304a248b9227b43cfff1868f9f10e
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/25062
Commit-Queue: Austin Eng <enga@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
Depth texture values outside the 0-1 range are not supported without
an extension.
Bug: dawn:413, dawn:367
Change-Id: I75e7f0de054d6f9a4bac7c6b955c65db810fe36e
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/24681
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Austin Eng <enga@chromium.org>
Projects using Dawn that are compiled with Emscripten may use some
of our common headers and utilities. ex.) creating a ShaderModule
from SPIR-V. Add handling for __EMSCRIPTEN__ to Platform.h so
compiling to WASM works.
Also, implement GetExecutablePath on Emscripten as unreachable.
This is only ever used to search native paths in the Vulkan backend.
Bug: none
Change-Id: I5d9d6e27e287cb34e7af3214add1c4bc8c6a0f70
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/25064
Reviewed-by: Kai Ninomiya <kainino@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Austin Eng <enga@chromium.org>
Added an additional alignment on the staging buffer offset in
WriteTexture on Vulkan to match
VkPhysicalDeviceLimits::optimalBufferCopyOffsetAlignment.
We now allocate some additional memory in the staging buffer for that
purpose.
Bug: dawn:483
Change-Id: Ic5b95aa57f6c241e0ab9f9853b4c3152e52e5630
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/24982
Commit-Queue: Tomek Ponitka <tommek@google.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Changed the order of validation blocks in ValidateLinearTextureData.
It doesn't match the order in spec now, but for the algorithm to
compute required bytes in copy we need the conditions which come
after the ones relating to it. Also switched the order of
ValidateLinearTextureData and ValidateTextureCopyRange for similar
reasons.
Bug: dawn:483
Change-Id: If00ae769d170ea12494258721916ec55d79e2880
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/25041
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This patch implements buffer lazy initialization beffor MapAsync() and
buffer creation with BufferDescriptor.mappedAtCreation == true.
Note that this patch doesn't initialize buffers in MapReadAsyc() and
MapWriteAsync() because they are deprecated and will be removed soon.
BUG=dawn:414
TEST=dawn_end2end_tests
Change-Id: Ifea99833897081f599c45797e0829c57de1ac926
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/24687
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
Fixing an earlier TODO about aligning bytesPerRow in WriteTextureImpl
to VkPhysicalDeviceLimits::optimalBufferCopyRowPitch.
Bug: dawn:483
Change-Id: Ided2d367177f2f886a84f232c77f1f9f0d50d05d
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/24981
Commit-Queue: Tomek Ponitka <tommek@google.com>
Reviewed-by: Austin Eng <enga@chromium.org>
This changes the format of the server->client callback for async mapping
to match MapAsync. Previously there were two callbacks, one for
MapReadAsync and one for MapWriteAsync.
Bug: dawn:445
Change-Id: I3330c07ac8bb6d1fa9019563e9c946875e852639
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/24821
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
vkCreateImage when bound to imported memory, must also have
a VkExternalImageCreateInfo struct in its pNext chain that
has VK_EXTERNAL_MEMORY_HANDLE_TYPE_TEMP_ZIRCON_VMO_BIT_FUCHSIA
as a handleType.
This was discovered when tests started failing after a roll
of the Vulkan validation layers.
Bug: none
Change-Id: Ic7fc95f963062c8740e714a490adf6a020fed220
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/25100
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
WebGPU currently allows as many as 108 view descriptors per bind
group. This is too many to have one descriptor allocator per size,
so we need to bucket them by size.
Bug: dawn:443, dawn:488
Change-Id: I4fc8cf7cd0dc8292bb6a8488fd2ceb7575e1e5f7
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/24787
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Bryan Bernhart <bryan.bernhart@intel.com>
Commit-Queue: Austin Eng <enga@chromium.org>
Emits it as a module-level variable. Deconstruct and throw away
the gl_PerVertex struct.
Not handled: unusual patterns that are technically valid but
which don't occur in practice:
- loading, storing, or producing intermediate values of the whole structure.
- multiple definitions of the per-vertex structure (e.g. if someone had
put both a vertex shader and a tessellation shader in the same
module.)
Bug: tint:3, tint:99
Change-Id: I3ad9ff6ab780a002367f01f385bfa7d6ddba6db9
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24880
Reviewed-by: dan sinclair <dsinclair@chromium.org>
This CL adds code to emit program constants in the MSL backend.
Bug: tint:8
Change-Id: I63e40983253349d2e293904fbe9b6f543b885b34
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24940
Reviewed-by: David Neto <dneto@google.com>
This changes the format of the client->server command for async mapping
to match MapAsync and updates the doer on the server to implement it via
MapAsync. Previously the command had an isWrite boolean that routed to
either MapReadAsync or MapWriteAsync.
Also updates the Wire tests using mock_webgpu.h for the new server-side
traces.
Bug: dawn:445
Change-Id: I673cbe5e08b7c71539e8678dbb07b4043a9fadb2
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/24820
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Also removes suppressions for tests that pass after
https://swiftshader-review.googlesource.com/c/SwiftShader/+/46509https://swiftshader.googlesource.com/SwiftShader/+log/ad65826a2ab1..e8dd233c7a85
$ git log ad65826a2..e8dd233c7 --date=short --no-merges --format='%ad %ae %s'
2020-07-16 cwallez Fix B<->T copies of multiple array layers and unpacked rowpitch
2020-07-15 bclayton Regres: Collate and add new documentation for Regres
2020-04-08 natsu Support Gralloc3 in SwiftShader
2020-07-14 swiftshader.regress Regres: Update test lists @ cd0af645
2020-07-13 capn Undefine Bool after the headers that define them
2020-07-11 capn Remove dead ASTC code for GLES
2020-07-09 capn Rename command classes to match API calls
2020-06-04 srisser Add decoder for BC6h
2020-06-15 sugoi Image dirtiness
2020-07-09 amaiorano Android: build against llvm-10 instead of llvm-7
Created with:
roll-dep third_party/swiftshader
Bug: None
Change-Id: Ife6dcf9f832e4aaab40de3bdb62d4b1eb258d2f8
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/25042
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
This CL fixes up the array constructors to emit `{}` instead of
`<type>[<size>]()`. The matrix example is also updated to have the
correct data format for WGSL matrices which fixes the MSL output.
Bug: tint:8
Change-Id: I3a08a8814d4b8b38a57b6324e2182a271c958ef3
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/25060
Reviewed-by: David Neto <dneto@google.com>
This CL adds a `_stride_` into the array type name to differentiate an
array of 8 elements and a runtime array of stride 8.
Bug: tint:102
Change-Id: Iaf10fe5957acde16a9ccdf2a0fd8a83e47bb57bc
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24962
Reviewed-by: David Neto <dneto@google.com>
This CL adds support for injecting padding into structs in the MSL
backend.
Bug: tint:8
Change-Id: I83631a71ce4a2f00b61974ee2c0c7ca1b97f3028
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24601
Reviewed-by: David Neto <dneto@google.com>
This CL extends module scoped variables to include support for builtins.
Bug: tint:8
Change-Id: I9e4363be32401bfdd45ad5d1727d9432aca206fe
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24786
Reviewed-by: David Neto <dneto@google.com>
This CL adds a helper to get referenced builtins and referenced
locations from the ast::Function.
Change-Id: I95cf7efd6b0fe9569c5e514d3245112e78147ffe
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24783
Reviewed-by: David Neto <dneto@google.com>
This gives Chromium the option to give Dawn the correct dxgi adapter to use
Bug: chromium:1036711
Change-Id: Ica544d2e76d1c300038fa07b5b639a35c43f60b9
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/24761
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Rafael Cintron <rafael.cintron@microsoft.com>
Commit-Queue: Natasha Lee <natlee@microsoft.com>
This CL cleans up some of the entry point variable code.
Bug: tint:8
Change-Id: Ic89ff3608d34af26eeb9fe9408cdbd5c7de22dda
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24782
Reviewed-by: David Neto <dneto@google.com>
Added implementation of writeTexture in Metal. It's using a
staging buffer instead of writing directly from the CPU to
the texture, because Dawn uses the private storage mode for
most of the Metal textures.
Bug: dawn:483
Change-Id: I6b85ee8bbe343881337bdb203a122dc1f1523177
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/24581
Commit-Queue: Tomek Ponitka <tommek@google.com>
Reviewed-by: Austin Eng <enga@chromium.org>
This CL generates entry point functions and duplicate functions as
needed to call from the entry points.
Bug: tint:8
Change-Id: I8092ce463248e7a887c26ae05b0774e8fa21ab94
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24764
Reviewed-by: David Neto <dneto@google.com>
GN recently added support for Apple frameworks to link, rather than
overloading the libs lists. This pulls .frameworks out of the libs
lists, so that GN can stop supporting .frameworks in libs in the
future.
Bug: chromium:1052560
Change-Id: I0b57a3c4c995887a57937317db30b67cf7a7f0f7
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/24281
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
This patch optimizes the implementation of buffer-to-texture and
texture-to-buffer computations on D3D12 backends by computing
TextureCopySplits once for all the 2D texture array layers in the
copy instead of computing it once per layer. You can see the
comments in the function D3D12::ComputeTextureCopySplits() for more
details.
BUG=dawn:453
TEST=dawn_end2end_tests
Change-Id: I1b66d24d2418147957fbe03e2c25144bd043a62e
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/24643
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Commit-Queue: Jiawei Shao <jiawei.shao@intel.com>
kMaxBindingsPerGroup is not a real WebGPU limit. Code in Dawn that
uses it now uses dynamically sized vectors, optimistic stack_vecs or
a different limit like kMaxDynamicBuffersPerPipelineLayout.
The CL introduces kMaxOptimalBindingsPerGroup=32 which is used to size
the static portion of stack_vecs, but the value is tentative.
Bug: dawn:443
Change-Id: I08e06bed6531bed8b4365f36cf2fc0579ac5f180
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/23502
Commit-Queue: Austin Eng <enga@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
This CL adds support for calling user defined functions from the MSL
backend. Intrinsics and imports are not handled yet.
Bug: tint:8
Change-Id: I45c3078d014ab89cc0eec76dd626759077e1a890
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24763
Reviewed-by: David Neto <dneto@google.com>
A given caller should inherit the globals referenced from a callee. This
way, a given entry point will have a list of all the variables used up
the stack which it needs to reference.
Change-Id: Ib6efcdd5c3347749ad2d54aecfa425bd966a62fd
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24762
Reviewed-by: David Neto <dneto@google.com>
This CL adds generation of the input/output structures for entry points.
Bug: tint:8
Change-Id: I93942496bcea0a2eea944e5e1cd0baf383530f5e
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24721
Reviewed-by: David Neto <dneto@google.com>
This Cl updates the type determiner to annotate each function with the
name of any entry points which call into the given function. This will
allow determining in the backends if we need to duplicate the function
due to differing entry point parameter requirements.
Bug: tint:8
Change-Id: Icd7c4ccab72dd6eabcf0abaf1159319949c4ecf5
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24760
Reviewed-by: David Neto <dneto@google.com>
MapAsync in dawn_native is fully implemented and only missing
a couple cleanups that can be done once MapRead/WriteAsync are
removed.
MapAsync in dawn_wire is left as a pure shim on top of
MapRead/WriteAsync and will be transitioned to its own commands
in follow-ups.
All MapRead/WriteAsync end2end and validation tests are duplicated
for MapAsync.
Bug: dawn:445
Change-Id: Ib1430b9257149917be19a84f13e0ddd2a8eccc32
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/24260
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Stephen White <senorblanco@chromium.org>
