Skip to content

torch-2.9: support Windows with XPU variant #9

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
dvrogozh wants to merge 1 commit into
base: master
Choose a base branch
from
Open

torch-2.9: support Windows with XPU variant #9

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
36 changes: 18 additions & 18 deletions configs/torch-2.9/torch_variant_config.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -121,27 +121,27 @@ deps_add_list = [
"nvidia-nvtx==13.0.39; platform_system == 'Linux' and variant_label == 'cuda13.0'",

# XPU
"dpcpp-cpp-rt==2025.2.1; platform_system == 'Linux' and 'intel' in variant_namespaces",
"dpcpp-cpp-rt==2025.2.1; 'intel' in variant_namespaces",
"impi-rt==2021.16.1; platform_system == 'Linux' and platform_machine == 'x86_64' and 'intel' in variant_namespaces",
"intel-cmplr-lib-rt==2025.2.1; platform_system == 'Linux' and 'intel' in variant_namespaces",
"intel-cmplr-lib-ur==2025.2.1; platform_system == 'Linux' and 'intel' in variant_namespaces",
"intel-cmplr-lic-rt==2025.2.1; platform_system == 'Linux' and 'intel' in variant_namespaces",
"intel-opencl-rt==2025.2.1; platform_system == 'Linux' and 'intel' in variant_namespaces",
"intel-openmp==2025.2.1; platform_system == 'Linux' and 'intel' in variant_namespaces",
"intel-pti==0.13.1; platform_system == 'Linux' and 'intel' in variant_namespaces",
"intel-sycl-rt==2025.2.1; platform_system == 'Linux' and 'intel' in variant_namespaces",
"mkl==2025.2.0; platform_system == 'Linux' and 'intel' in variant_namespaces",
"intel-cmplr-lib-rt==2025.2.1; 'intel' in variant_namespaces",
Copy link
Author

@dvrogozh dvrogozh Oct 14, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

For reviewers:

Limiting the following 3 package to platform_system == 'Linux' and platform_machine == 'x86_64' is not a mistake as these packages are needed for XCCL distributed backend and it's not available for Linux:

  • impi-rt
  • oneccl-devel
  • oneccl

"intel-cmplr-lib-ur==2025.2.1; 'intel' in variant_namespaces",
"intel-cmplr-lic-rt==2025.2.1; 'intel' in variant_namespaces",
"intel-opencl-rt==2025.2.1; 'intel' in variant_namespaces",
"intel-openmp==2025.2.1; 'intel' in variant_namespaces",
"intel-pti==0.13.1; 'intel' in variant_namespaces",
"intel-sycl-rt==2025.2.1; 'intel' in variant_namespaces",
"mkl==2025.2.0; 'intel' in variant_namespaces",
"oneccl-devel==2021.16.1; platform_system == 'Linux' and platform_machine == 'x86_64' and 'intel' in variant_namespaces",
"oneccl==2021.16.1; platform_system == 'Linux' and platform_machine == 'x86_64' and 'intel' in variant_namespaces",
"onemkl-sycl-blas==2025.2.0; platform_system == 'Linux' and 'intel' in variant_namespaces",
"onemkl-sycl-dft==2025.2.0; platform_system == 'Linux' and 'intel' in variant_namespaces",
"onemkl-sycl-lapack==2025.2.0; platform_system == 'Linux' and 'intel' in variant_namespaces",
"onemkl-sycl-rng==2025.2.0; platform_system == 'Linux' and 'intel' in variant_namespaces",
"onemkl-sycl-sparse==2025.2.0; platform_system == 'Linux' and 'intel' in variant_namespaces",
"pytorch-triton-xpu==3.5.0; platform_system == 'Linux' and 'intel' in variant_namespaces",
"tbb==2022.2.0; platform_system == 'Linux' and 'intel' in variant_namespaces",
"tcmlib==1.4.0; platform_system == 'Linux' and 'intel' in variant_namespaces",
"umf==0.11.0; platform_system == 'Linux' and 'intel' in variant_namespaces",
"onemkl-sycl-blas==2025.2.0; 'intel' in variant_namespaces",
"onemkl-sycl-dft==2025.2.0; 'intel' in variant_namespaces",
"onemkl-sycl-lapack==2025.2.0; 'intel' in variant_namespaces",
"onemkl-sycl-rng==2025.2.0; 'intel' in variant_namespaces",
"onemkl-sycl-sparse==2025.2.0; 'intel' in variant_namespaces",
"pytorch-triton-xpu==3.5.0; 'intel' in variant_namespaces",
"tbb==2022.2.0; 'intel' in variant_namespaces",
"tcmlib==1.4.0; 'intel' in variant_namespaces",
"umf==0.11.0; 'intel' in variant_namespaces",

# Common to all ROCm Variants
'pytorch-triton-rocm==3.5.0; platform_system == "Linux" and platform_machine == "x86_64" and "amd" in variant_namespaces',
Expand Down