CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-18 11:46:58 +00:00
Code Issues Packages Projects Releases Wiki Activity

HIP: Cleanup hipification header (#15285)

Browse Source 
add expicit conversion operator to support older versions of rocm
Switch over to hip_bf16 from legacy hip_bfloat16
Simplify RDNA3 define
Reduce swap over of new hipblas api to rocm 6.5 as this version is used for rocm 7.0 previews

---------

Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
This commit is contained in:
uvos
2025-08-14 16:23:56 +02:00
committed by GitHub
parent b204a5a234
commit 5ba36f6103
7 changed files with 32 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
ggml/src/ggml-cuda/cpy-utils.cuh
View File

@@ -1,15 +1,7 @@
#pragma once
#include "ggml-common.h"
template<typename src_t, typename dst_t>
static __device__ __forceinline__ void convert_flt(const src_t * src, dst_t * dst) {
if constexpr (std::is_same_v<src_t, dst_t>) {
*dst = *src;
} else {
*dst = float(*src);
}
}
#include "convert.cuh"
static __device__ __forceinline__ int best_index_int8(int n, const int8_t * val, float x) {
if (x <= val[0]) return 0;
@@ -221,5 +213,5 @@ static __device__ void cpy_blck_f32_iq4_nl(const char * cxi, char * cdsti) {
template<typename src_t, typename dst_t>
static __device__ void cpy_1_flt(const char * cxi, char * cdsti) {
convert_flt((const src_t *)cxi, (dst_t *)cdsti);
*(dst_t *) cdsti = ggml_cuda_cast<dst_t>(*(const src_t *) cxi);
}