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

cuda : rename build flag to LLAMA_CUDA (#6299)

Browse source
This commit is contained in:
slaren 2024-03-26 01:16:01 +01:00 • committed by GitHub
parent b06c16ef9f
commit 280345968d
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
28 changed files with 129 additions and 115 deletions
Download patch file Download diff file Expand all files Collapse all files

26
Makefile
View file

@ -390,12 +390,17 @@ ifdef LLAMA_BLIS
endif # LLAMA_BLIS
ifdef LLAMA_CUBLAS
# LLAMA_CUBLAS is deprecated and will be removed in the future
LLAMA_CUDA := 1
endif
ifdef LLAMA_CUDA
ifneq ('', '$(wildcard /opt/cuda)')
CUDA_PATH ?= /opt/cuda
else
CUDA_PATH ?= /usr/local/cuda
endif
MK_CPPFLAGS += -DGGML_USE_CUBLAS -I$(CUDA_PATH)/include -I$(CUDA_PATH)/targets/$(UNAME_M)-linux/include
MK_CPPFLAGS += -DGGML_USE_CUDA -I$(CUDA_PATH)/include -I$(CUDA_PATH)/targets/$(UNAME_M)-linux/include
MK_LDFLAGS += -lcuda -lcublas -lculibos -lcudart -lcublasLt -lpthread -ldl -lrt -L$(CUDA_PATH)/lib64 -L/usr/lib64 -L$(CUDA_PATH)/targets/$(UNAME_M)-linux/lib -L/usr/lib/wsl/lib
OBJS += ggml-cuda.o
OBJS += $(patsubst %.cu,%.o,$(wildcard ggml-cuda/*.cu))
@ -462,7 +467,7 @@ endif
ifdef JETSON_EOL_MODULE_DETECT
define NVCC_COMPILE
$(NVCC) -I. -Icommon -D_XOPEN_SOURCE=600 -D_GNU_SOURCE -DNDEBUG -DGGML_USE_CUBLAS -I/usr/local/cuda/include -I/opt/cuda/include -I/usr/local/cuda/targets/aarch64-linux/include -std=c++11 -O3 $(NVCCFLAGS) $(CPPFLAGS) -Xcompiler "$(CUDA_CXXFLAGS)" -c $< -o $@
$(NVCC) -I. -Icommon -D_XOPEN_SOURCE=600 -D_GNU_SOURCE -DNDEBUG -DGGML_USE_CUDA -I/usr/local/cuda/include -I/opt/cuda/include -I/usr/local/cuda/targets/aarch64-linux/include -std=c++11 -O3 $(NVCCFLAGS) $(CPPFLAGS) -Xcompiler "$(CUDA_CXXFLAGS)" -c $< -o $@
endef # NVCC_COMPILE
else
define NVCC_COMPILE
@ -476,7 +481,7 @@ ggml-cuda/%.o: ggml-cuda/%.cu ggml-cuda/%.cuh ggml.h ggml-common.h ggml-cuda/com
ggml-cuda.o: ggml-cuda.cu ggml-cuda.h ggml.h ggml-backend.h ggml-backend-impl.h ggml-common.h $(wildcard ggml-cuda/*.cuh)
$(NVCC_COMPILE)
endif # LLAMA_CUBLAS
endif # LLAMA_CUDA
ifdef LLAMA_CLBLAST
@ -533,7 +538,7 @@ ifdef LLAMA_HIPBLAS
LLAMA_CUDA_DMMV_X ?= 32
LLAMA_CUDA_MMV_Y ?= 1
LLAMA_CUDA_KQUANTS_ITER ?= 2
MK_CPPFLAGS += -DGGML_USE_HIPBLAS -DGGML_USE_CUBLAS
MK_CPPFLAGS += -DGGML_USE_HIPBLAS -DGGML_USE_CUDA
ifdef LLAMA_HIP_UMA
MK_CPPFLAGS += -DGGML_HIP_UMA
endif # LLAMA_HIP_UMA
@ -609,7 +614,7 @@ override NVCCFLAGS := $(MK_NVCCFLAGS) $(NVCCFLAGS)
override LDFLAGS := $(MK_LDFLAGS) $(LDFLAGS)
# identify CUDA host compiler
ifdef LLAMA_CUBLAS
ifdef LLAMA_CUDA
GF_CC := $(NVCC) $(NVCCFLAGS) 2>/dev/null .c -Xcompiler
include scripts/get-flags.mk
CUDA_CXXFLAGS := $(BASE_CXXFLAGS) $(GF_CXXFLAGS) -Wno-pedantic
@ -634,7 +639,7 @@ $(info I NVCCFLAGS: $(NVCCFLAGS))
$(info I LDFLAGS: $(LDFLAGS))
$(info I CC: $(shell $(CC) --version | head -n 1))
$(info I CXX: $(shell $(CXX) --version | head -n 1))
ifdef LLAMA_CUBLAS
ifdef LLAMA_CUDA
$(info I NVCC: $(shell $(NVCC) --version | tail -n 1))
CUDA_VERSION := $(shell $(NVCC) --version | grep -oP 'release (\K[0-9]+\.[0-9])')
ifeq ($(shell awk -v "v=$(CUDA_VERSION)" 'BEGIN { print (v < 11.7) }'),1)
@ -644,9 +649,16 @@ $(error I ERROR: For CUDA versions < 11.7 a target CUDA architecture must be exp
endif # CUDA_POWER_ARCH
endif # CUDA_DOCKER_ARCH
endif # eq ($(shell echo "$(CUDA_VERSION) < 11.7" | bc),1)
endif # LLAMA_CUBLAS
endif # LLAMA_CUDA
$(info )
ifdef LLAMA_CUBLAS
$(info !!!!)
$(info LLAMA_CUBLAS is deprecated and will be removed in the future. Use LLAMA_CUDA instead.)
$(info !!!!)
$(info )
endif
#
# Build library
#