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

ggml : update get_rows f16 and q

Browse source
This commit is contained in:
slaren 2023-12-09 22:38:22 +01:00
parent ac3f7d8e23
commit 2e4db48291
2 changed files with 26 additions and 18 deletions
Download patch file Download diff file Expand all files Collapse all files

5
Makefile
View file

@ -396,6 +396,11 @@ ifdef LLAMA_CUBLAS
MK_LDFLAGS += -lcublas -lculibos -lcudart -lcublasLt -lpthread -ldl -lrt -L/usr/local/cuda/lib64 -L/opt/cuda/lib64 -L$(CUDA_PATH)/targets/x86_64-linux/lib
OBJS += ggml-cuda.o
NVCCFLAGS = --forward-unknown-to-host-compiler -use_fast_math
ifdef LLAMA_DEBUG
NVCCFLAGS += -lineinfo
endif
ifdef LLAMA_CUDA_NVCC
NVCC = $(LLAMA_CUDA_NVCC)
else