ExternalVendorCode/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2024-12-18 20:27:57 +00:00
Code Issues Actions 12 Packages Projects Releases Wiki Activity

feat(build): only build llama.cpp relevant targets (#2659)

Browse Source 
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
This commit is contained in:
Ettore Di Giacinto 2024-06-26 14:58:38 +02:00 committed by GitHub
parent 59af0e77af
commit b783c811db
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 8 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
Makefile
View File

@ -54,7 +54,7 @@ override LD_FLAGS += -X "github.com/go-skynet/LocalAI/internal.Commit=$(shell gi
OPTIONAL_TARGETS?= OPTIONAL_TARGETS?=
OS := $(shell uname -s) export OS := $(shell uname -s)
ARCH := $(shell uname -m) ARCH := $(shell uname -m)
GREEN := $(shell tput -Txterm setaf 2) GREEN := $(shell tput -Txterm setaf 2)
YELLOW := $(shell tput -Txterm setaf 3) YELLOW := $(shell tput -Txterm setaf 3)
@ -764,7 +764,7 @@ backend-assets/grpc/llama-cpp-grpc: backend-assets/grpc
cp -rf backend/cpp/llama backend/cpp/llama-grpc cp -rf backend/cpp/llama backend/cpp/llama-grpc
$(MAKE) -C backend/cpp/llama-grpc purge $(MAKE) -C backend/cpp/llama-grpc purge
$(info ${GREEN}I llama-cpp build info:grpc${RESET}) $(info ${GREEN}I llama-cpp build info:grpc${RESET})
CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_RPC=ON -DLLAMA_AVX=off -DLLAMA_AVX2=off -DLLAMA_AVX512=off -DLLAMA_FMA=off -DLLAMA_F16C=off" $(MAKE) VARIANT="llama-grpc" build-llama-cpp-grpc-server CMAKE_ARGS="$(CMAKE_ARGS) -DLLAMA_RPC=ON -DLLAMA_AVX=off -DLLAMA_AVX2=off -DLLAMA_AVX512=off -DLLAMA_FMA=off -DLLAMA_F16C=off" TARGET="--target grpc-server --target rpc-server" $(MAKE) VARIANT="llama-grpc" build-llama-cpp-grpc-server
cp -rfv backend/cpp/llama-grpc/grpc-server backend-assets/grpc/llama-cpp-grpc cp -rfv backend/cpp/llama-grpc/grpc-server backend-assets/grpc/llama-cpp-grpc
backend-assets/util/llama-cpp-rpc-server: backend-assets/grpc/llama-cpp-grpc backend-assets/util/llama-cpp-rpc-server: backend-assets/grpc/llama-cpp-grpc

9
backend/cpp/llama/Makefile
View File

@ -4,6 +4,7 @@ LLAMA_VERSION?=
CMAKE_ARGS?= CMAKE_ARGS?=
BUILD_TYPE?= BUILD_TYPE?=
ONEAPI_VARS?=/opt/intel/oneapi/setvars.sh ONEAPI_VARS?=/opt/intel/oneapi/setvars.sh
TARGET?=--target grpc-server
# If build type is cublas, then we set -DLLAMA_CUBLAS=ON to CMAKE_ARGS automatically # If build type is cublas, then we set -DLLAMA_CUBLAS=ON to CMAKE_ARGS automatically
ifeq ($(BUILD_TYPE),cublas) ifeq ($(BUILD_TYPE),cublas)
@ -20,9 +21,11 @@ else ifeq ($(BUILD_TYPE),hipblas)
CMAKE_ARGS+=-DLLAMA_HIPBLAS=ON CMAKE_ARGS+=-DLLAMA_HIPBLAS=ON
# If it's OSX, DO NOT embed the metal library - -DLLAMA_METAL_EMBED_LIBRARY=ON requires further investigation # If it's OSX, DO NOT embed the metal library - -DLLAMA_METAL_EMBED_LIBRARY=ON requires further investigation
# But if it's OSX without metal, disable it here # But if it's OSX without metal, disable it here
else ifeq ($(OS),darwin) else ifeq ($(OS),Darwin)
ifneq ($(BUILD_TYPE),metal) ifneq ($(BUILD_TYPE),metal)
CMAKE_ARGS+=-DLLAMA_METAL=OFF CMAKE_ARGS+=-DLLAMA_METAL=OFF
else
TARGET+=--target ggml-metal
endif endif
endif endif
@ -62,8 +65,8 @@ grpc-server: llama.cpp llama.cpp/examples/grpc-server
@echo "Building grpc-server with $(BUILD_TYPE) build type and $(CMAKE_ARGS)" @echo "Building grpc-server with $(BUILD_TYPE) build type and $(CMAKE_ARGS)"
ifneq (,$(findstring sycl,$(BUILD_TYPE))) ifneq (,$(findstring sycl,$(BUILD_TYPE)))
bash -c "source $(ONEAPI_VARS); \ bash -c "source $(ONEAPI_VARS); \
cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && $(MAKE)" cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && cmake --build . --config Release $(TARGET)"
else else
cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && $(MAKE) cd llama.cpp && mkdir -p build && cd build && cmake .. $(CMAKE_ARGS) && cmake --build . --config Release $(TARGET)
endif endif
cp llama.cpp/build/bin/grpc-server . cp llama.cpp/build/bin/grpc-server .