char-fork/llama.cpp

git clone https://git.t4t.associates/char-fork/llama.cpp

shalinib-ibmQ4/Q8 Tiled Gemm Optimization. (#16999)3a0d10533

char/aliceai-t5-moe
brancheschar/aliceai-t5-moeall refs →
524 B25 linesraw
1#pragma once
2#include <stdint.h>
3#include <stdbool.h>
4
5#if defined(__VXE__) || defined(__VXE2__)
6#include <vecintrin.h>
7#endif
8
9#ifdef _MSC_VER
10#define NOINLINE __declspec(noinline)
11#else
12#define NOINLINE __attribute__((__noinline__))
13#endif
14
15#ifdef __cplusplus
16extern "C" {
17#endif
18
19bool llamafile_sgemm(const struct ggml_compute_params * params, int64_t, int64_t, int64_t,
20                     const void *, int64_t, const void *, int64_t, void *, int64_t,
21                     int, int, int);
22
23#ifdef __cplusplus
24}
25#endif