Skip to content

[CUDA] Fix MultiHeadAttention thread safe and bias support #7321

[CUDA] Fix MultiHeadAttention thread safe and bias support

[CUDA] Fix MultiHeadAttention thread safe and bias support #7321

Triggered via pull request July 30, 2024 18:13
Status Cancelled
Total duration 1h 6m 46s
Artifacts

sca.yml

on: pull_request
Onnxruntime-SCA-training-CUDA
48m 43s
Onnxruntime-SCA-training-CUDA
Onnxruntime-SCA-win32-WINML-x64
47m 29s
Onnxruntime-SCA-win32-WINML-x64
Onnxruntime-SCA-win32-WINML-x86
43m 31s
Onnxruntime-SCA-win32-WINML-x86
Fit to window
Zoom out
Zoom in

Annotations

6 errors
Onnxruntime-SCA-win32-WINML-x86
Canceling since a higher priority waiting request for 'Windows_SCA-tlwu/fix_dmmha_input_check' exists
Onnxruntime-SCA-win32-WINML-x86
The operation was canceled.
Onnxruntime-SCA-win32-WINML-x64
Canceling since a higher priority waiting request for 'Windows_SCA-tlwu/fix_dmmha_input_check' exists
Onnxruntime-SCA-win32-WINML-x64
The operation was canceled.
Onnxruntime-SCA-training-CUDA
Canceling since a higher priority waiting request for 'Windows_SCA-tlwu/fix_dmmha_input_check' exists
Onnxruntime-SCA-training-CUDA
The operation was canceled.