tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-29 12:35:16 +00:00
Code Issues Packages Projects Releases Wiki Activity

ggml : riscv: add 128-bit RVV support (#12530)

Browse Source 
* ggml : add 128-bit RVV support

* ggml : revert to old RVV 256+ q2_K, q3_K, q4_K, q6_K impl

* remove trailing whitespaces

* restructure vector length selection code
This commit is contained in:
xctan
2025-03-27 14:38:34 +08:00
committed by GitHub
parent f28bc4c286
commit 24feaec057
4 changed files with 781 additions and 397 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
ggml/src/ggml-impl.h
View File

@ -381,6 +381,35 @@ GGML_API void ggml_aligned_free(void * ptr, size_t size);
return r;
}
#elif defined(__riscv) && defined(GGML_RV_ZFH)
static inline float ggml_compute_fp16_to_fp32(ggml_fp16_t h) {
float f;
__asm__(
"fmv.h.x %[f], %[h]\n\t"
"fcvt.s.h %[f], %[f]"
: [f] "=&f" (f)
: [h] "r" (h)
);
return f;
}
static inline ggml_fp16_t ggml_compute_fp32_to_fp16(float f) {
ggml_fp16_t res;
__asm__(
"fcvt.h.s %[f], %[f]\n\t"
"fmv.x.h %[h], %[f]"
: [h] "=&r" (res)
: [f] "f" (f)
);
return res;
}
#define GGML_COMPUTE_FP16_TO_FP32(x) ggml_compute_fp16_to_fp32(x)
#define GGML_COMPUTE_FP32_TO_FP16(x) ggml_compute_fp32_to_fp16(x)
#define GGML_FP16_TO_FP32(x) GGML_COMPUTE_FP16_TO_FP32(x)
#define GGML_FP32_TO_FP16(x) GGML_COMPUTE_FP32_TO_FP16(x)
#else
// FP16 <-> FP32