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

ggml-cpu : fix RISC-V Q4_0 repack select and RVV feature reporting (#17951) 

* ggml-cpu:fix RISC-V Q4_0 repack select and RVV feature reporting

Signed-off-by: Wang Yang <yangwang@iscas.ac.cn>

* using the name VLEN instead of CNT

* Update ggml/include/ggml-cpu.h

---------

Signed-off-by: Wang Yang <yangwang@iscas.ac.cn>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
ixgbe 2025-12-12 22:26:03 +08:00 committed by GitHub
parent 17158965ac
commit 51604435e8
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
4 changed files with 33 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
ggml/include/ggml-cpu.h
View File

@ -99,6 +99,7 @@ extern "C" {
GGML_BACKEND_API int ggml_cpu_has_sme (void);
// other
GGML_BACKEND_API int ggml_cpu_has_riscv_v (void);
GGML_BACKEND_API int ggml_cpu_get_rvv_vlen (void); // risc-v vector length in bytes
GGML_BACKEND_API int ggml_cpu_has_vsx (void);
GGML_BACKEND_API int ggml_cpu_has_vxe (void);
GGML_BACKEND_API int ggml_cpu_has_wasm_simd (void);

26
ggml/src/ggml-cpu/ggml-cpu.c
View File

@ -81,6 +81,11 @@ struct ggml_arm_arch_features_type {
} ggml_arm_arch_features = { 0 };
#endif
#if defined(__riscv)
struct ggml_riscv_arch_features_type {
int rvv_vlen;
} ggml_riscv_arch_features = { 0 };
#endif
#if defined(_WIN32)
@ -703,6 +708,15 @@ static void ggml_init_arm_arch_features(void) {}
#endif
#endif // __ARM_ARCH
#if defined(__riscv) && defined(__riscv_v_intrinsic)
#include <riscv_vector.h>
static void ggml_init_riscv_arch_features(void) {
ggml_riscv_arch_features.rvv_vlen = __riscv_vlenb();
}
#else
static void ggml_init_riscv_arch_features(void) {}
#endif
struct ggml_tensor * ggml_new_i32(struct ggml_context * ctx, int32_t value) {
GGML_ASSERT(!ggml_get_no_alloc(ctx));
@ -3459,6 +3473,14 @@ int ggml_cpu_has_riscv_v(void) {
#endif
}
int ggml_cpu_get_rvv_vlen(void) {
#if defined(__riscv) && defined(__riscv_v_intrinsic)
return ggml_riscv_arch_features.rvv_vlen;
#else
return 0;
#endif
}
int ggml_cpu_has_f16c(void) {
#if defined(__F16C__)
return 1;
@ -3625,6 +3647,10 @@ void ggml_cpu_init(void) {
ggml_init_arm_arch_features();
#endif
#if defined(__riscv)
ggml_init_riscv_arch_features();
#endif
is_first_call = false;
}

4
ggml/src/ggml-cpu/ggml-cpu.cpp
View File

@ -583,6 +583,10 @@ static ggml_backend_feature * ggml_backend_cpu_get_features(ggml_backend_reg_t r
if (ggml_cpu_has_riscv_v()) {
features.push_back({ "RISCV_V", "1" });
}
if (ggml_cpu_get_rvv_vlen() > 0) {
static std::string rvv_vlen = std::to_string(ggml_cpu_get_rvv_vlen());
features.push_back({ "RVV_VLEN", rvv_vlen.c_str() });
}
if (ggml_cpu_has_vsx()) {
features.push_back({ "VSX", "1" });
}

3
ggml/src/ggml-cpu/repack.cpp
View File

@ -2169,7 +2169,8 @@ static const ggml::cpu::tensor_traits * ggml_repack_get_optimal_repack_type(cons
static const ggml::cpu::repack::tensor_traits<block_iq4_nl, 8, 8, GGML_TYPE_Q8_0> iq4_nl_8x8_q8_0;
if (cur->type == GGML_TYPE_Q4_0) {
if (ggml_cpu_has_avx2() || (ggml_cpu_has_sve() && ggml_cpu_has_matmul_int8() && ggml_cpu_get_sve_cnt() == QK8_0)) {
if (ggml_cpu_has_avx2() || (ggml_cpu_has_sve() && ggml_cpu_has_matmul_int8() && ggml_cpu_get_sve_cnt() == QK8_0)
|| (ggml_cpu_has_riscv_v() && (ggml_cpu_get_rvv_vlen() >= QK4_0))) {
if (cur->ne[1] % 8 == 0) {
return &q4_0_8x8_q8_0;
}