Akarshan Biswas
SYCL: Refactor and enable FP16 in binary broadcast OPs (llama/12975)
1377b05
raw
history blame
952 Bytes
#ifndef GGML_SYCL_BINBCAST_HPP
#define GGML_SYCL_BINBCAST_HPP
#include "common.hpp"
static __dpct_inline__ float op_repeat(const float a, const float b) {
return b;
GGML_UNUSED(a);
}
static __dpct_inline__ float op_add(const float a, const float b) {
return a + b;
}
static __dpct_inline__ float op_sub(const float a, const float b) {
return a - b;
}
static __dpct_inline__ float op_mul(const float a, const float b) {
return a * b;
}
static __dpct_inline__ float op_div(const float a, const float b) {
return a / b;
}
void ggml_sycl_add(ggml_backend_sycl_context & ctx, ggml_tensor * dst);
void ggml_sycl_sub(ggml_backend_sycl_context & ctx, ggml_tensor * dst);
void ggml_sycl_mul(ggml_backend_sycl_context & ctx, ggml_tensor * dst);
void ggml_sycl_div(ggml_backend_sycl_context & ctx, ggml_tensor * dst);
void ggml_sycl_repeat(ggml_backend_sycl_context & ctx, ggml_tensor * dst);
#endif //GGML_SYCL_BINBCAST_HPP