aboutsummaryrefslogtreecommitdiff
path: root/Makefile
diff options
context:
space:
mode:
Diffstat (limited to 'Makefile')
-rw-r--r--Makefile8
1 files changed, 7 insertions, 1 deletions
diff --git a/Makefile b/Makefile
index 7141566..6068cbe 100644
--- a/Makefile
+++ b/Makefile
@@ -163,7 +163,12 @@ ifdef LLAMA_CUBLAS
LDFLAGS += -lcublas -lculibos -lcudart -lcublasLt -lpthread -ldl -lrt -L/usr/local/cuda/lib64 -L/opt/cuda/lib64 -L$(CUDA_PATH)/targets/x86_64-linux/lib
OBJS += ggml-cuda.o
NVCC = nvcc
- NVCCFLAGS = --forward-unknown-to-host-compiler -arch=native
+ NVCCFLAGS = --forward-unknown-to-host-compiler
+ifdef CUDA_DOCKER_ARCH
+ NVCCFLAGS += -Wno-deprecated-gpu-targets -arch=$(CUDA_DOCKER_ARCH)
+else
+ NVCCFLAGS += -arch=native
+endif # CUDA_DOCKER_ARCH
ifdef LLAMA_CUDA_FORCE_DMMV
NVCCFLAGS += -DGGML_CUDA_FORCE_DMMV
endif # LLAMA_CUDA_FORCE_DMMV
@@ -187,6 +192,7 @@ ifdef LLAMA_CUDA_KQUANTS_ITER
else
NVCCFLAGS += -DK_QUANTS_PER_ITERATION=2
endif
+
ggml-cuda.o: ggml-cuda.cu ggml-cuda.h
$(NVCC) $(NVCCFLAGS) $(CXXFLAGS) -Wno-pedantic -c $< -o $@
endif # LLAMA_CUBLAS