feature: add blis support

This commit is contained in:
zenix 2023-05-17 19:33:57 +09:00
parent 2d5db48371
commit 0926278434
4 changed files with 95 additions and 0 deletions

67
BLIS.md Normal file
View file

@ -0,0 +1,67 @@
BLIS Installation Manual
------------------------
BLIS is a portable software framework for high-performance BLAS-like dense linear algebra libraries. It has received awards and recognition, including the 2023 James H. Wilkinson Prize for Numerical Software and the 2020 SIAM Activity Group on Supercomputing Best Paper Prize. BLIS provides a new BLAS-like API and a compatibility layer for traditional BLAS routine calls. It offers features such as object-based API, typed API, BLAS and CBLAS compatibility layers.
Project URL: https://github.com/flame/blis
### Prepare:
Compile BLIS:
```bash
git clone https://github.com/flame/blis
cd blis
./configure --enable-cblas -t openmp,pthreads auto
# will install to /usr/local/ by default.
make -j
```
Install BLIS:
```bash
sudo make install
```
We recommend using openmp since it's easier to modify the cores been used.
### llama.cpp compilation
Makefile:
```bash
make LLAMA_BLIS=1 -j
# make LLAMA_BLIS=1 benchmark-matmult
```
CMake:
```bash
mkdir build
cd build
cmake -DLLAMA_BLIS=ON ..
make -j
```
### llama.cpp execution
According to the BLIS documentation, we could set the following
environment variables to modify the behavior of openmp:
```
export GOMP_GPU_AFFINITY="0-19"
export BLIS_NUM_THREADS=14
```
And then run the binaries as normal.
### Intel specific issue
Some might get the error message saying that `libimf.so` cannot be found.
Please follow this [stackoverflow page](https://stackoverflow.com/questions/70687930/intel-oneapi-2022-libimf-so-no-such-file-or-directory-during-openmpi-compila).
### Reference:
1. https://github.com/flame/blis#getting-started
2. https://github.com/flame/blis/blob/master/docs/Multithreading.md

View file

@ -66,6 +66,7 @@ endif()
# 3rd party libs
option(LLAMA_ACCELERATE "llama: enable Accelerate framework" ON)
option(LLAMA_OPENBLAS "llama: use OpenBLAS" OFF)
option(LLAMA_BLIS "llama: use blis" OFF)
option(LLAMA_CUBLAS "llama: use cuBLAS" OFF)
option(LLAMA_CLBLAST "llama: use CLBlast" OFF)
@ -178,6 +179,24 @@ if (LLAMA_OPENBLAS)
endif()
endif()
if (LLAMA_BLIS)
add_compile_definitions(GGML_USE_BLIS)
# we don't directly call BLIS apis, use cblas wrapper instead
add_compile_definitions(GGML_USE_OPENBLAS)
set(BLIS_INCLUDE_SEARCH_PATHS
/usr/include
/usr/include/blis
/usr/local/include
/usr/local/include/blis
$ENV{BLIS_HOME}
$ENV{BLIS_HOME}/include
)
find_path(BLIS_INC NAMES blis.h PATHS ${BLIS_INCLUDE_SEARCH_PATHS})
add_compile_definitions(BLIS_ENABLE_CBLAS)
add_link_options(-lblis)
add_compile_options(-I${BLIS_INC})
endif()
if (LLAMA_CUBLAS)
cmake_minimum_required(VERSION 3.17)

View file

@ -122,6 +122,10 @@ ifdef LLAMA_OPENBLAS
LDFLAGS += -lopenblas
endif
endif
ifdef LLAMA_BLIS
CFLAGS += -DGGML_USE_OPENBLAS -DGGML_USE_BLIS -I/usr/local/include/blis -I/usr/include/blis
LDFLAGS += -lblis -L/usr/local/lib
endif
ifdef LLAMA_CUBLAS
CFLAGS += -DGGML_USE_CUBLAS -I/usr/local/cuda/include -I/opt/cuda/include -I$(CUDA_PATH)/targets/x86_64-linux/include
CXXFLAGS += -DGGML_USE_CUBLAS -I/usr/local/cuda/include -I/opt/cuda/include -I$(CUDA_PATH)/targets/x86_64-linux/include

View file

@ -58,6 +58,7 @@ The main goal of `llama.cpp` is to run the LLaMA model using 4-bit integer quant
- Runs on the CPU
- OpenBLAS support
- cuBLAS and CLBlast support
- BLIS support (cblas wrapper)
The original implementation of `llama.cpp` was [hacked in an evening](https://github.com/ggerganov/llama.cpp/issues/33#issuecomment-1465108022).
Since then, the project has improved significantly thanks to many contributions. This project is for educational purposes and serves
@ -278,6 +279,10 @@ Building the program with BLAS support may lead to some performance improvements
cmake --build . --config Release
```
- BLIS
Check [BLIS.md](BLIS.md) for more information.
- cuBLAS
This provides BLAS acceleration using the CUDA cores of your Nvidia GPU. Make sure to have the CUDA toolkit installed. You can download it from your Linux distro's package manager or from here: [CUDA Toolkit](https://developer.nvidia.com/cuda-downloads).