[WIP] WebGPU EP [skip ci] #21904
[WIP] WebGPU EP [skip ci] #21904
Commits on Aug 29, 2024
Commits on Aug 30, 2024
-
-
make webgpu build on linux and known to most tools (#21937)
Co-authored-by: Yulong Wang <[email protected]>
-
-
-
-
-
Commits on Sep 1, 2024
Commits on Sep 3, 2024
-
-
-
-
-
-
### Description <!-- Describe your changes. --> ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. -->
-
-
compare the content of WEBGPU_BUFFER, not the address (#21967)
On linux (not sure about windows) WEBGPU_BUFFER is defined in multiple object files and comparing the address is not sufficient - use strcmp. onnxruntime uses strcmp for the most but there are some other places that compare against address which might make trouble if passed acrross object file boundary.
Commits on Sep 4, 2024
Commits on Sep 6, 2024
Commits on Sep 10, 2024
-
### Description <!-- Describe your changes. --> ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. --> --------- Co-authored-by: Yulong Wang <[email protected]>
-
-
-
-
Commits on Sep 11, 2024
-
-
-
[webgpu-native] Add transpose op (#21986)
Co-authored-by: Yulong Wang <[email protected]>
-
Commits on Sep 12, 2024
-
-
Revert "placeholder for setting proc table"
This reverts commit dce0f18.
-
Commits on Sep 14, 2024
-
[WebGPU EP] Support Shape operator (#22095)
### Description Shape operator ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. -->
Commits on Sep 17, 2024
-
[webgpu EP] Binary operators (#22112)
based on: - #22058 --------- Co-authored-by: Qin Jiajia <[email protected]>
-
-
-
Commits on Sep 20, 2024
-
[webgpu-native] Add where op (#22014)
Co-authored-by: Yulong Wang <[email protected]>
-
Commits on Sep 24, 2024
Commits on Sep 25, 2024
-
-
-
-
-
-
-
-
update build configurations to webgpu EP (#22047)
### Description --------- Co-authored-by: Scott McKay <[email protected]>
Commits on Sep 26, 2024
Commits on Sep 27, 2024
-
[webgpu native] Add RotaryEmbedding op (#22194)
### Description <!-- Describe your changes. --> ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. -->
-
[webgpu native] Add transpose shared (#22098)
### Description <!-- Describe your changes. --> ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. -->
-
[webgpu-native] Add gather (#22183)
### Description <!-- Describe your changes. --> ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. --> --------- Co-authored-by: Yulong Wang <[email protected]>
-
[Native-WebGPU] Add Concat (#22225)
### Description Add Concat operator support ### Motivation and Context Required for WebGPU EP --------- Co-authored-by: Yulong Wang <[email protected]>
-
[webgpu-native] Add MatmulNBits (#22150)
### Description <!-- Describe your changes. --> ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. --> --------- Co-authored-by: Yulong Wang <[email protected]>
Commits on Sep 30, 2024
-
[WebGPU-Native] Tile Operator (#22239)
Adds WebGPU implementation for Tile operator.
-
use Abseil OStringStream in WebGPU EP string concat (#22241)
### Description use Abseil OStringStream in WebGPU EP string concat. Allows multiple calls to shader_helper.MainFunctionBody()
-
webgpu: support MultiHeadAttention operator (#22144)
### Description <!-- Describe your changes. --> ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. --> --------- Co-authored-by: Yulong Wang <[email protected]>
Commits on Oct 1, 2024
-
[webgpu-native] support for webgpu layernorms (#22249)
adds webgpu support for LayerNormalization, SimplifiedLayerNormalization, SkipLayerNormalization, SkipSimplifiedLayerNormalization --------- Co-authored-by: Yulong Wang <[email protected]>
-
-
-
-
[webgpu-native] Fix a few build errors on Linux (#22286)
1. On Linux path strings are not std::wstring. They should be std::string. 2. "auto i =0" means i will be int. But sometimes we want to it to be size_t.
-
fix issues for e2e phi3 (#22287)
- support for input_skip_bias_sum in SkipLayerNorm - use GetElementAt in concat --------- Co-authored-by: Yulong Wang <[email protected]>
Commits on Oct 4, 2024
Commits on Oct 5, 2024
Commits on Oct 7, 2024
Commits on Oct 8, 2024
Commits on Oct 9, 2024
Commits on Oct 11, 2024
-
[webgpu-native] support webgpu profiling (#22255)
### Description <!-- Describe your changes. --> ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. --> --------- Co-authored-by: Yulong Wang <[email protected]>
-
-
[WebGPU EP] Remove unused variable. (#22412)
### Description <!-- Describe your changes. --> Remove unused variable. ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. --> Fix Android build error: ``` onnxruntime/onnxruntime/core/providers/webgpu/webgpu_context.h:122:12: error: private field 'query_time_base_' is not used [-Werror,-Wunused-private-field] 122 | uint64_t query_time_base_; | ^ 1 error generated. ```
Commits on Oct 15, 2024
Commits on Oct 16, 2024
-
allow to set a list for specifying force CPU nodes (#22431)
### Description This change allows to set EP option "forceCpuNodeNames" in which includes a list of node names that will always be running on CPU. Usage example: ```diff opt.executionProviders[0] = { name: "webgpu", validationMode: 'wgpuOnly', storageBufferCacheMode: 'bucket', + // force node "/model/embed_tokens/Gather" to be put on CPU + forceCpuNodeNames: "/model/embed_tokens/Gather" }; ```
Commits on Oct 17, 2024
Commits on Oct 21, 2024
-
[webgpu-native] opt matmulnbits (#22472)
### Description <!-- Describe your changes. --> ### Motivation and Context <!-- - Why is this change required? What problem does it solve? - If it fixes an open issue, please link to the issue here. -->
Commits on Oct 22, 2024
-
[webgpu-native] Fix CI errors on MacOS (#22535)
The MacOS device on CI is an Apple Paravirtual device, which may not support TimestampQuery. In this PR, If TimestampQuery is not supported, a warning is printed to not block profiling cpu data.
-
Commits on Oct 24, 2024
-
allow build flag --use_external_dawn (#22552)
### Description introduce this build flag --use_external_dawn to allow specifying Dawn as an external dependency. When it is specified, `dawn::native` is excluded from build of onnxruntime, and passing in dawn proc table is required. I didn't find a good way to run the test. Currently I need to build with the following flags: ```sh C:\code\onnxruntime>build --use_webgpu --use_external_dawn --skip_tests --target onnxruntime_webgpu_external_dawn_test ``` And run the test using: ```sh # run test that passing dawn proc table. expect session to initialize successfully C:\code\onnxruntime\build\Windows\Debug\Debug>onnxruntime_webgpu_external_dawn_test.exe # run test that does not pass proc table. expect exception to throw. C:\code\onnxruntime\build\Windows\Debug\Debug>onnxruntime_webgpu_external_dawn_test.exe --no_proc_table ```
-