Files
pytorch/torch/nativert/executor/OpKernelKind.h
dolpm 4f72d932fe re-land triton runtime implementation" (#162217)
Summary: original pr - https://github.com/pytorch/pytorch/pull/161798

Test Plan:
ci

Rollback Plan:

Differential Revision: D81724234

Pull Request resolved: https://github.com/pytorch/pytorch/pull/162217
Approved by: https://github.com/SherlockNoMad
2025-09-06 00:52:29 +00:00

18 lines
318 B
C++

#pragma once
#include <cstdint>
namespace torch::nativert {
enum class OpKernelKind : uint8_t {
kPrimKernel,
kStaticDispatchKernel,
kInterpreterFallbackKernel,
// static dispatch kernels that don't reuse
// out TensorImpl
kNativeStaticDispatchKernel,
kTritonKernel,
};
} // namespace torch::nativert