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

mmq implementation for non k-quants

Browse source
This commit is contained in:
JohannesGaessler 2023-07-08 19:12:39 +02:00
parent d91f3f0c55
commit ddb37bf8a0
2 changed files with 784 additions and 265 deletions
Download patch file Download diff file Expand all files Collapse all files

2
Makefile
View file

@ -194,7 +194,7 @@ ifdef LLAMA_CUBLAS
CXXFLAGS += -DGGML_USE_CUBLAS -I/usr/local/cuda/include -I/opt/cuda/include -I$(CUDA_PATH)/targets/x86_64-linux/include
LDFLAGS += -lcublas -lculibos -lcudart -lcublasLt -lpthread -ldl -lrt -L/usr/local/cuda/lib64 -L/opt/cuda/lib64 -L$(CUDA_PATH)/targets/x86_64-linux/lib
OBJS += ggml-cuda.o
NVCCFLAGS = --forward-unknown-to-host-compiler
NVCCFLAGS = --forward-unknown-to-host-compiler -use_fast_math
ifdef LLAMA_CUDA_NVCC
NVCC = $(LLAMA_CUDA_NVCC)
else

1047
ggml-cuda.cu
View file

File diff suppressed because it is too large Load diff