dearwolf/LocalAI
1
0
Fork 0
mirror of https://github.com/mudler/LocalAI.git synced 2025-05-20 02:24:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 21

deps(llama.cpp): bump to latest, update build variables (#2669)

Browse source 
* arrow_up: Update ggerganov/llama.cpp

Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* deps(llama.cpp): update build variables to follow upstream

Update build recipes with https://github.com/ggerganov/llama.cpp/pull/8006

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* Disable shared libs by default in llama.cpp

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* Disable shared libs in llama.cpp Makefile

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* Disable metal embedding for now, until it is tested

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* fix(mac): explicitly enable metal

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* debug

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* fix typo

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

---------

Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Co-authored-by: mudler <2420543+mudler@users.noreply.github.com>
This commit is contained in:
Ettore Di Giacinto 2024-06-27 23:10:04 +02:00 • committed by GitHub
parent 30b883affe
commit 7b1e792732
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
9 changed files with 39 additions and 32 deletions
Download patch file Download diff file Expand all files Collapse all files

29
backend/cpp/llama/Makefile
View file

@ -6,35 +6,42 @@ BUILD_TYPE?=
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
# Disable Shared libs as we are linking on static gRPC and we can't mix shared and static
CMAKE_ARGS+=-DBUILD_SHARED_LIBS=OFF
# If build type is cublas, then we set -DGGML_CUDA=ON to CMAKE_ARGS automatically
ifeq ($(BUILD_TYPE),cublas)
CMAKE_ARGS+=-DLLAMA_CUBLAS=ON
# If build type is openblas then we set -DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS
CMAKE_ARGS+=-DGGML_CUDA=ON
# If build type is openblas then we set -DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS
# to CMAKE_ARGS automatically
else ifeq ($(BUILD_TYPE),openblas)
CMAKE_ARGS+=-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS
# If build type is clblas (openCL) we set -DLLAMA_CLBLAST=ON -DCLBlast_DIR=/some/path
CMAKE_ARGS+=-DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS
# If build type is clblas (openCL) we set -DGGML_CLBLAST=ON -DCLBlast_DIR=/some/path
else ifeq ($(BUILD_TYPE),clblas)
CMAKE_ARGS+=-DLLAMA_CLBLAST=ON -DCLBlast_DIR=/some/path
CMAKE_ARGS+=-DGGML_CLBLAST=ON -DCLBlast_DIR=/some/path
# If it's hipblas we do have also to set CC=/opt/rocm/llvm/bin/clang CXX=/opt/rocm/llvm/bin/clang++
else ifeq ($(BUILD_TYPE),hipblas)
CMAKE_ARGS+=-DLLAMA_HIPBLAS=ON
# If it's OSX, DO NOT embed the metal library - -DLLAMA_METAL_EMBED_LIBRARY=ON requires further investigation
CMAKE_ARGS+=-DGGML_HIPBLAS=ON
# If it's OSX, DO NOT embed the metal library - -DGGML_METAL_EMBED_LIBRARY=ON requires further investigation
# But if it's OSX without metal, disable it here
else ifeq ($(OS),Darwin)
ifneq ($(BUILD_TYPE),metal)
CMAKE_ARGS+=-DLLAMA_METAL=OFF
CMAKE_ARGS+=-DGGML_METAL=OFF
else
CMAKE_ARGS+=-DGGML_METAL=ON
# Until this is tested properly, we disable embedded metal file
# as we already embed it as part of the LocalAI assets
CMAKE_ARGS+=-DGGML_METAL_EMBED_LIBRARY=OFF
TARGET+=--target ggml-metal
endif
endif
ifeq ($(BUILD_TYPE),sycl_f16)
CMAKE_ARGS+=-DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_SYCL_F16=ON
CMAKE_ARGS+=-DGGML_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DGGML_SYCL_F16=ON
endif
ifeq ($(BUILD_TYPE),sycl_f32)
CMAKE_ARGS+=-DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
CMAKE_ARGS+=-DGGML_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
endif
llama.cpp: