frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-21 05:34:18 +08:00
Code Issues Packages Projects Releases Wiki Activity

[HPU] Add hpu to fused kernels supported devices (#148666)

Browse Source 
This change adds "hpu" to the list of device types that support fused kernels in the optimizer, ensuring
compatibility with HPU backend.

Without this change, when `test_all_gather_extension_outer_size_stride` of `pytorch/test/distributed/_composable/fsdp/test_fully_shard_extensions.py` is run on 'hpu' backend, it fails with:

RuntimeError: fused=True requires all the params to be floating point Tensors
of supported devices: ['mps', 'cuda', 'xpu', 'cpu', 'privateuseone']
but torch.float32 and hpu

Pull Request resolved: https://github.com/pytorch/pytorch/pull/148666
Approved by: https://github.com/albanD
This commit is contained in:
Nitin Singh
2025-03-07 04:28:30 +00:00
committed by PyTorch MergeBot
parent 33f8ab2f58
commit fe4b88f6aa
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
torch/utils/_foreach_utils.py
View File

@ -11,7 +11,7 @@ def _get_foreach_kernels_supported_devices() -> list[str]:
def _get_fused_kernels_supported_devices() -> list[str]:
r"""Return the device type list that supports fused kernels in optimizer."""
return ["mps", "cuda", "xpu", "cpu", torch._C._get_privateuse1_backend_name()]
return ["mps", "cuda", "xpu", "hpu", "cpu", torch._C._get_privateuse1_backend_name()]
TensorListList: TypeAlias = list[list[Optional[Tensor]]]
Indices: TypeAlias = list[int]