vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

try AMD fix

Browse source
This commit is contained in:
Johannes Gäßler 2024-05-04 09:50:43 +02:00
parent 86636bd115
commit 617f129e43
1 changed files with 1 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

3
ggml-cuda/common.cuh
View file

@ -317,8 +317,7 @@ static __device__ __forceinline__ int __dp4a(const int a, const int b, int c) {
} }
#endif // defined(GGML_USE_HIPBLAS) #endif // defined(GGML_USE_HIPBLAS)
#define FP16_AVAILABLE defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__) ? \ #define FP16_AVAILABLE (defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__)) || __CUDA_ARCH__ >= CC_PASCAL
defined(RDNA1) || defined(RDNA2) || defined(RDNA3) : __CUDA_ARCH__ >= CC_PASCAL
#define FP16_MMA_AVAILABLE !(defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__)) && __CUDA_ARCH__ >= CC_VOLTA #define FP16_MMA_AVAILABLE !(defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__)) && __CUDA_ARCH__ >= CC_VOLTA