happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Fixed format consistency of repack gemm and gemv declarations

This commit is contained in:
Alberto Cabrera 2025-12-15 15:58:03 +00:00
parent 360cbd94f2
commit 275237c771
1 changed files with 6 additions and 48 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

54
ggml/src/ggml-cpu/repack.cpp
View File

@ -2008,25 +2008,11 @@ template <> void gemv<block_iq4_nl, 8, 8, GGML_TYPE_Q8_0>(int n, float * s, size
ggml_gemv_iq4_nl_8x8_q8_0(n, s, bs, vx, vy, nr, nc);
}
template <>
void gemv<block_q8_0, 4, 4, GGML_TYPE_Q8_0>(int n,
float * s,
size_t bs,
const void * vx,
const void * vy,
int nr,
int nc) {
template <> void gemv<block_q8_0, 4, 4, GGML_TYPE_Q8_0>(int n, float * s, size_t bs, const void * vx, const void * vy, int nr, int nc) {
ggml_gemv_q8_0_4x4_q8_0(n, s, bs, vx, vy, nr, nc);
}
template <>
void gemv<block_q8_0, 8, 4, GGML_TYPE_Q8_0>(int n,
float * s,
size_t bs,
const void * vx,
const void * vy,
int nr,
int nc) {
template <> void gemv<block_q8_0, 8, 4, GGML_TYPE_Q8_0>(int n, float * s, size_t bs, const void * vx, const void * vy, int nr, int nc) {
ggml_gemv_q8_0_4x8_q8_0(n, s, bs, vx, vy, nr, nc);
}
@ -2050,25 +2036,11 @@ template <> void gemm<block_q4_0, 8, 8, GGML_TYPE_Q8_0>(int n, float * s, size_t
ggml_gemm_q4_0_8x8_q8_0(n, s, bs, vx, vy, nr, nc);
}
template <>
void gemm<block_q4_K, 8, 8, GGML_TYPE_Q8_K>(int n,
float * s,
size_t bs,
const void * vx,
const void * vy,
int nr,
int nc) {
template <> void gemm<block_q4_K, 8, 8, GGML_TYPE_Q8_K>(int n, float * s, size_t bs, const void * vx, const void * vy, int nr, int nc) {
ggml_gemm_q4_K_8x8_q8_K(n, s, bs, vx, vy, nr, nc);
}
template <>
void gemm<block_q2_K, 8, 8, GGML_TYPE_Q8_K>(int n,
float * s,
size_t bs,
const void * vx,
const void * vy,
int nr,
int nc) {
template <> void gemm<block_q2_K, 8, 8, GGML_TYPE_Q8_K>(int n, float * s, size_t bs, const void * vx, const void * vy, int nr, int nc) {
ggml_gemm_q2_K_8x8_q8_K(n, s, bs, vx, vy, nr, nc);
}
@ -2080,25 +2052,11 @@ template <> void gemm<block_iq4_nl, 8, 8, GGML_TYPE_Q8_0>(int n, float * s, size
ggml_gemm_iq4_nl_8x8_q8_0(n, s, bs, vx, vy, nr, nc);
}
template <>
void gemm<block_q8_0, 4, 4, GGML_TYPE_Q8_0>(int n,
float * s,
size_t bs,
const void * vx,
const void * vy,
int nr,
int nc) {
template <> void gemm<block_q8_0, 4, 4, GGML_TYPE_Q8_0>(int n, float * s, size_t bs, const void * vx, const void * vy, int nr, int nc) {
ggml_gemm_q8_0_4x4_q8_0(n, s, bs, vx, vy, nr, nc);
}
template <>
void gemm<block_q8_0, 8, 4, GGML_TYPE_Q8_0>(int n,
float * s,
size_t bs,
const void * vx,
const void * vy,
int nr,
int nc) {
template <> void gemm<block_q8_0, 8, 4, GGML_TYPE_Q8_0>(int n, float * s, size_t bs, const void * vx, const void * vy, int nr, int nc) {
ggml_gemm_q8_0_4x8_q8_0(n, s, bs, vx, vy, nr, nc);
}