mirror of
https://git.datalinker.icu/vllm-project/vllm.git
synced 2025-12-26 22:49:44 +08:00
9239bf718e
Signed-off-by: ElizaWszola <eliza@neuralmagic.com> Signed-off-by: ElizaWszola <ewszola@redhat.com> Co-authored-by: Lucas Wilkinson <wilkinson.lucas@gmail.com>
80 lines
3.4 KiB
Plaintext
80 lines
3.4 KiB
Plaintext
#pragma once
|
|
|
|
#include <cuda.h>
|
|
#include <torch/all.h>
|
|
#include <c10/cuda/CUDAStream.h>
|
|
|
|
#include "core/scalar_type.hpp"
|
|
#include "cutlass/bfloat16.h"
|
|
#include "cutlass/float8.h"
|
|
|
|
template <typename ElementAB, typename ElementC, typename ElementAccumulator>
|
|
__global__ void get_group_gemm_starts(
|
|
int32_t* expert_offsets, ElementAB** a_offsets, ElementAB** b_offsets,
|
|
ElementC** out_offsets, ElementAccumulator** a_scales_offsets,
|
|
ElementAccumulator** b_scales_offsets, ElementAB* a_base_as_int,
|
|
ElementAB* b_base_as_int, ElementC* out_base_as_int,
|
|
ElementAccumulator* a_scales_base_as_int,
|
|
ElementAccumulator* b_scales_base_as_int, int64_t n, int64_t k,
|
|
bool per_act_token, bool per_out_ch) {
|
|
int expert_id = threadIdx.x;
|
|
|
|
int64_t expert_offset = expert_offsets[expert_id];
|
|
|
|
a_offsets[expert_id] = a_base_as_int + expert_offset * k;
|
|
b_offsets[expert_id] = b_base_as_int + expert_id * k * n;
|
|
out_offsets[expert_id] = out_base_as_int + expert_offset * n;
|
|
a_scales_offsets[expert_id] =
|
|
a_scales_base_as_int + (per_act_token ? expert_offset : 0);
|
|
b_scales_offsets[expert_id] =
|
|
b_scales_base_as_int + (per_out_ch ? n * expert_id : expert_id);
|
|
}
|
|
|
|
#define __CALL_GET_STARTS_KERNEL(TENSOR_C_TYPE, C_TYPE) \
|
|
else if (out_tensors.dtype() == TENSOR_C_TYPE) { \
|
|
get_group_gemm_starts<cutlass::float_e4m3_t, C_TYPE, float> \
|
|
<<<1, num_experts, 0, stream>>>( \
|
|
static_cast<int32_t*>(expert_offsets.data_ptr()), \
|
|
static_cast<cutlass::float_e4m3_t**>(a_ptrs.data_ptr()), \
|
|
static_cast<cutlass::float_e4m3_t**>(b_ptrs.data_ptr()), \
|
|
static_cast<C_TYPE**>(out_ptrs.data_ptr()), \
|
|
static_cast<float**>(a_scales_ptrs.data_ptr()), \
|
|
static_cast<float**>(b_scales_ptrs.data_ptr()), \
|
|
static_cast<cutlass::float_e4m3_t*>(a_tensors.data_ptr()), \
|
|
static_cast<cutlass::float_e4m3_t*>(b_tensors.data_ptr()), \
|
|
static_cast<C_TYPE*>(out_tensors.data_ptr()), \
|
|
static_cast<float*>(a_scales.data_ptr()), \
|
|
static_cast<float*>(b_scales.data_ptr()), out_tensors.size(1), \
|
|
a_tensors.size(1), per_act_token, per_out_ch); \
|
|
}
|
|
|
|
namespace {
|
|
|
|
void run_get_group_gemm_starts(
|
|
torch::Tensor const& expert_offsets, torch::Tensor& a_ptrs,
|
|
torch::Tensor& b_ptrs, torch::Tensor& out_ptrs,
|
|
torch::Tensor& a_scales_ptrs, torch::Tensor& b_scales_ptrs,
|
|
torch::Tensor const& a_tensors, torch::Tensor const& b_tensors,
|
|
torch::Tensor& out_tensors, torch::Tensor const& a_scales,
|
|
torch::Tensor const& b_scales) {
|
|
TORCH_CHECK(a_tensors.dtype() == torch::kFloat8_e4m3fn);
|
|
TORCH_CHECK(b_tensors.dtype() == torch::kFloat8_e4m3fn);
|
|
TORCH_CHECK(a_scales.dtype() == torch::kFloat32);
|
|
TORCH_CHECK(b_scales.dtype() == torch::kFloat32);
|
|
|
|
int num_experts = static_cast<int>(expert_offsets.size(0));
|
|
bool per_act_token = a_scales.numel() != 1;
|
|
bool per_out_ch = b_scales.numel() != num_experts;
|
|
|
|
auto stream = at::cuda::getCurrentCUDAStream(a_tensors.device().index());
|
|
|
|
if (false) {
|
|
}
|
|
__CALL_GET_STARTS_KERNEL(torch::kBFloat16, cutlass::bfloat16_t)
|
|
__CALL_GET_STARTS_KERNEL(torch::kFloat16, half)
|
|
else {
|
|
TORCH_CHECK(false, "Invalid output type (must be float16 or bfloat16)");
|
|
}
|
|
}
|
|
|
|
} // namespace |