vllm/csrc/quantization/gptq_marlin/kernel.h

39 lines
1.9 KiB
C++

#ifndef MARLIN_NAMESPACE_NAME
#define MARLIN_NAMESPACE_NAME marlin
#endif
#include "marlin.cuh"
#include "marlin_dtypes.cuh"
#include "core/scalar_type.hpp"
#define MARLIN_KERNEL_PARAMS \
const int4 *__restrict__ A, const int4 *__restrict__ B, \
int4 *__restrict__ C, int4 *__restrict__ C_tmp, \
const int4 *__restrict__ scales_ptr, \
const uint16_t *__restrict__ scale2_ptr, \
const int4 *__restrict__ zp_ptr, const int *__restrict__ g_idx, \
int num_groups, int prob_m, int prob_n, int prob_k, int lda, int *locks, \
bool use_atomic_add, bool use_fp32_reduce, int max_shared_mem
namespace MARLIN_NAMESPACE_NAME {
template <typename scalar_t, // compute dtype, half or nv_float16
const vllm::ScalarTypeId w_type_id, // weight ScalarType id
const int threads, // number of threads in a threadblock
const int thread_m_blocks, // number of 16x16 blocks in the m
// dimension (batchsize) of the
// threadblock
const int thread_n_blocks, // same for n dimension (output)
const int thread_k_blocks, // same for k dimension (reduction)
const bool m_block_size_8, // whether m_block_size == 8
// only works when thread_m_blocks == 1
const int stages, // number of stages for the async global->shared
// fetch pipeline
const int group_blocks, // number of consecutive 16x16 blocks
// with a separate quantization scale
const bool is_zp_float // is zero point of float16 type?
>
__global__ void Marlin(MARLIN_KERNEL_PARAMS);
}