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

CUDA: use MMQ instead of cuBLAS by default (#8075)

Browse source
This commit is contained in:
Johannes Gäßler 2024-06-24 17:43:42 +02:00 • committed by GitHub
parent d62e4aaa02
commit a818f3028d
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
8 changed files with 124 additions and 122 deletions
Download patch file Download diff file Expand all files Collapse all files

3
Makefile
View file

@ -537,6 +537,9 @@ endif # LLAMA_CUDA_FORCE_DMMV
ifdef LLAMA_CUDA_FORCE_MMQ
MK_NVCCFLAGS += -DGGML_CUDA_FORCE_MMQ
endif # LLAMA_CUDA_FORCE_MMQ
ifdef LLAMA_CUDA_FORCE_CUBLAS
MK_NVCCFLAGS += -DGGML_CUDA_FORCE_CUBLAS
endif # LLAMA_CUDA_FORCE_CUBLAS
ifdef LLAMA_CUDA_DMMV_X
MK_NVCCFLAGS += -DGGML_CUDA_DMMV_X=$(LLAMA_CUDA_DMMV_X)
else