[torch.compile] support all attention backends (#10558)

Signed-off-by: youkaichao <youkaichao@gmail.com>
This commit is contained in:
youkaichao
2024-11-22 14:04:42 -08:00
committed by GitHub
parent db100c5cde
commit eebad39f26
77 changed files with 876 additions and 648 deletions
+4
View File
@@ -57,6 +57,10 @@ class DeviceCapability(NamedTuple):
class Platform:
_enum: PlatformEnum
device_type: str
# available dispatch keys:
# check https://github.com/pytorch/pytorch/blob/313dac6c1ca0fa0cde32477509cce32089f8532a/torchgen/model.py#L134 # noqa
# use "CPU" as a fallback for platforms not registered in PyTorch
dispatch_key: str = "CPU"
def is_cuda(self) -> bool:
return self._enum == PlatformEnum.CUDA