Files
triton-tvm-ffi/examples/mm/mm.cc
2026-02-08 22:24:12 +08:00

56 lines
2.2 KiB
C++

#include <ATen/DLConvertor.h>
#include <ATen/dlpack.h>
#include <tvm/ffi/extra/cuda/cubin_launcher.h>
#include <tvm/ffi/tvm_ffi.h>
#ifndef MATMUL_KERNEL_STUB
#define MATMUL_KERNEL_STUB(grid, device, stream, args, kwargs)
#endif
#ifndef MATMUL_NAME
#define MATMUL_NAME ""
#endif
tvm::ffi::Tensor Matmul(tvm::ffi::Tensor a, tvm::ffi::Tensor b,
tvm::ffi::String activation) {
at::Tensor atorch = at::fromDLPack(a.ToDLPack()),
btorch = at::fromDLPack(b.ToDLPack());
const int32_t M = atorch.size(0), K = atorch.size(1), N = btorch.size(1);
at::Tensor ctorch = at::empty({M, N}, atorch.options());
tvm::ffi::Function grid = tvm::ffi::Function::FromTyped(
[M, N](const tvm::ffi::Map<tvm::ffi::String, tvm::ffi::Any> &meta)
-> tvm::ffi::Tuple<int32_t, int32_t, int32_t> {
const int32_t BLOCK_SIZE_M = meta["BLOCK_SIZE_M"].cast<int32_t>(),
BLOCK_SIZE_N = meta["BLOCK_SIZE_N"].cast<int32_t>();
return tvm::ffi::Tuple<int32_t, int32_t, int32_t>{
(M + BLOCK_SIZE_M - 1) / BLOCK_SIZE_M *
((N + BLOCK_SIZE_N - 1) / BLOCK_SIZE_N),
1, 1};
});
DLDevice device = a.device();
void *stream = TVMFFIEnvGetStream(device.device_type, device.device_id);
tvm::ffi::Tensor c = tvm::ffi::Tensor::FromDLPack(at::toDLPack(ctorch));
tvm::ffi::Array<tvm::ffi::Any> args = {a,
b,
c,
M,
N,
K,
atorch.stride(0),
atorch.stride(1),
btorch.stride(0),
btorch.stride(1),
ctorch.stride(0),
ctorch.stride(1)};
tvm::ffi::Map<tvm::ffi::String, tvm::ffi::Any> kwargs = {
{"ACTIVATION", activation},
};
MATMUL_KERNEL_STUB(grid, device.device_id, stream, args, kwargs);
return c;
}
TVM_FFI_STATIC_INIT_BLOCK() {
namespace refl = tvm::ffi::reflection;
refl::GlobalDef().def(MATMUL_NAME, Matmul);
}