Export XPU oneDNN header to the public (#139177)

# Motivation
Export oneDNN header to the public, for example, the third-party extension now could use `GpuStreamManager` to manage `dnnl::stream` to submit oneDNN kernel.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/139177
Approved by: https://github.com/gujinghui, https://github.com/EikanWang, https://github.com/malfet
This commit is contained in:
Yu, Guangye
2024-10-30 18:33:51 +00:00
committed by PyTorch MergeBot
parent 04382efe5e
commit 46bca8a4b6
2 changed files with 4 additions and 1 deletions

View File

@ -1280,6 +1280,8 @@ def main():
"include/ATen/native/hip/*.h",
"include/ATen/native/hip/*.cuh",
"include/ATen/native/mps/*.h",
"include/ATen/native/mkldnn/xpu/*.h",
"include/ATen/native/mkldnn/xpu/detail/*.h",
"include/ATen/native/nested/*.h",
"include/ATen/native/quantized/*.h",
"include/ATen/native/quantized/cpu/*.h",