haraldwolff/llama.cpp

Fork 0

Neo Zhang Jianyu 59c17f02de

add blog link (#6222 )

2024-03-22 15:19:37 +08:00

18 KiB

Raw Blame History

llama.cpp for SYCL

Background
News
OS
Intel GPU
Docker
Linux
Windows
Environment Variable
Known Issue
Q&A
Todo

Background

SYCL is a higher-level programming model to improve programming productivity on various hardware accelerators—such as CPUs, GPUs, and FPGAs. It is a single-source embedded domain-specific language based on pure C++17.

oneAPI is a specification that is open and standards-based, supporting multiple architecture types including but not limited to GPU, CPU, and FPGA. The spec has both direct programming and API-based programming paradigms.

Intel uses the SYCL as direct programming language to support CPU, GPUs and FPGAs.

To avoid to re-invent the wheel, this code refer other code paths in llama.cpp (like OpenBLAS, cuBLAS, CLBlast). We use a open-source tool SYCLomatic (Commercial release Intel® DPC++ Compatibility Tool) migrate to SYCL.

The llama.cpp for SYCL is used to support Intel GPUs.

For Intel CPU, recommend to use llama.cpp for X86 (Intel MKL building).

News

2024.3
- A blog is published: Run LLM on all Intel GPUs Using llama.cpp: intel.com or medium.com.
- New base line is ready: tag b2437.
- Support multiple cards: --split-mode: [none|layer]; not support [row], it's on developing.
- Support to assign main GPU by --main-gpu, replace $GGML_SYCL_DEVICE.
- Support detecting all GPUs with level-zero and same top Max compute units.
- Support OPs
  - hardsigmoid
  - hardswish
  - pool2d
2024.1
- Create SYCL backend for Intel GPU.
- Support Windows build

OS

OS	Status	Verified
Linux	Support	Ubuntu 22.04, Fedora Silverblue 39
Windows	Support	Windows 11

Intel GPU

Verified

Intel GPU	Status	Verified Model
Intel Data Center Max Series	Support	Max 1550
Intel Data Center Flex Series	Support	Flex 170
Intel Arc Series	Support	Arc 770, 730M
Intel built-in Arc GPU	Support	built-in Arc GPU in Meteor Lake
Intel iGPU	Support	iGPU in i5-1250P, i7-1260P, i7-1165G7

Note: If the EUs (Execution Unit) in iGPU is less than 80, the inference speed will be too slow to use.

Memory

The memory is a limitation to run LLM on GPUs.

When run llama.cpp, there is print log to show the applied memory on GPU. You could know how much memory to be used in your case. Like llm_load_tensors: buffer size = 3577.56 MiB.

For iGPU, please make sure the shared memory from host memory is enough. For llama-2-7b.Q4_0, recommend the host memory is 8GB+.

For dGPU, please make sure the device memory is enough. For llama-2-7b.Q4_0, recommend the device memory is 4GB+.

Nvidia GPU

Verified

Intel GPU	Status	Verified Model
Ampere Series	Support	A100

oneMKL for CUDA

The current oneMKL release does not contain the oneMKL cuBlas backend. As a result for Nvidia GPU's oneMKL must be built from source.

git clone https://github.com/oneapi-src/oneMKL
cd oneMKL
mkdir build
cd build
cmake -G Ninja .. -DCMAKE_CXX_COMPILER=icpx -DCMAKE_C_COMPILER=icx -DENABLE_MKLGPU_BACKEND=OFF -DENABLE_MKLCPU_BACKEND=OFF -DENABLE_CUBLAS_BACKEND=ON
ninja
// Add paths as necessary

Docker

Note:

Only docker on Linux is tested. Docker on WSL may not work.
You may need to install Intel GPU driver on the host machine (See the Linux section to know how to do that)

Build the image

You can choose between F16 and F32 build. F16 is faster for long-prompt inference.

# For F16:
#docker build -t llama-cpp-sycl --build-arg="LLAMA_SYCL_F16=ON" -f .devops/main-intel.Dockerfile .

# Or, for F32:
docker build -t llama-cpp-sycl -f .devops/main-intel.Dockerfile .

# Note: you can also use the ".devops/server-intel.Dockerfile", which compiles the "server" example

Run

# Firstly, find all the DRI cards:
ls -la /dev/dri
# Then, pick the card that you want to use.

# For example with "/dev/dri/card1"
docker run -it --rm -v "$(pwd):/app:Z" --device /dev/dri/renderD128:/dev/dri/renderD128 --device /dev/dri/card1:/dev/dri/card1 llama-cpp-sycl -m "/app/models/YOUR_MODEL_FILE" -p "Building a website can be done in 10 simple steps:" -n 400 -e -ngl 33

Linux

Setup Environment

Install Intel GPU driver.

a. Please install Intel GPU driver by official guide: Install GPU Drivers.

Note: for iGPU, please install the client GPU driver.

b. Add user to group: video, render.

sudo usermod -aG render username
sudo usermod -aG video username

Note: re-login to enable it.

c. Check

sudo apt install clinfo
sudo clinfo -l

Output (example):

Platform #0: Intel(R) OpenCL Graphics
 `-- Device #0: Intel(R) Arc(TM) A770 Graphics


Platform #0: Intel(R) OpenCL HD Graphics
 `-- Device #0: Intel(R) Iris(R) Xe Graphics [0x9a49]

Install Intel® oneAPI Base toolkit.

a. Please follow the procedure in Get the Intel® oneAPI Base Toolkit .

Recommend to install to default folder: /opt/intel/oneapi.

Following guide use the default folder as example. If you use other folder, please modify the following guide info with your folder.

b. Check

source /opt/intel/oneapi/setvars.sh

sycl-ls

There should be one or more level-zero devices. Please confirm that at least one GPU is present, like [ext_oneapi_level_zero:gpu:0].

Output (example):

[opencl:acc:0] Intel(R) FPGA Emulation Platform for OpenCL(TM), Intel(R) FPGA Emulation Device OpenCL 1.2  [2023.16.10.0.17_160000]
[opencl:cpu:1] Intel(R) OpenCL, 13th Gen Intel(R) Core(TM) i7-13700K OpenCL 3.0 (Build 0) [2023.16.10.0.17_160000]
[opencl:gpu:2] Intel(R) OpenCL Graphics, Intel(R) Arc(TM) A770 Graphics OpenCL 3.0 NEO  [23.30.26918.50]
[ext_oneapi_level_zero:gpu:0] Intel(R) Level-Zero, Intel(R) Arc(TM) A770 Graphics 1.3 [1.3.26918]

Build locally:

Note:

You can choose between F16 and F32 build. F16 is faster for long-prompt inference.
By default, it will build for all binary files. It will take more time. To reduce the time, we recommend to build for example/main only.

mkdir -p build
cd build
source /opt/intel/oneapi/setvars.sh

# For FP16:
#cmake .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx -DLLAMA_SYCL_F16=ON

# Or, for FP32:
cmake .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx

# For Nvidia GPUs
cmake .. -DLLAMA_SYCL=ON -DLLAMA_SYCL_TARGET=NVIDIA -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx

# Build example/main only
#cmake --build . --config Release --target main

# Or, build all binary
cmake --build . --config Release -v

cd ..

./examples/sycl/build.sh

Run

Put model file to folder models

You could download llama-2-7b.Q4_0.gguf as example.

Enable oneAPI running environment

source /opt/intel/oneapi/setvars.sh

List device ID

Run without parameter:

./build/bin/ls-sycl-device

# or running the "main" executable and look at the output log:

./build/bin/main

Check the ID in startup log, like:

found 6 SYCL devices:
|  |                  |                                             |Compute   |Max compute|Max work|Max sub|               |
|ID|       Device Type|                                         Name|capability|units      |group   |group  |Global mem size|
|--|------------------|---------------------------------------------|----------|-----------|--------|-------|---------------|
| 0|[level_zero:gpu:0]|               Intel(R) Arc(TM) A770 Graphics|       1.3|        512|    1024|     32|    16225243136|
| 1|[level_zero:gpu:1]|                    Intel(R) UHD Graphics 770|       1.3|         32|     512|     32|    53651849216|
| 2|    [opencl:gpu:0]|               Intel(R) Arc(TM) A770 Graphics|       3.0|        512|    1024|     32|    16225243136|
| 3|    [opencl:gpu:1]|                    Intel(R) UHD Graphics 770|       3.0|         32|     512|     32|    53651849216|
| 4|    [opencl:cpu:0]|         13th Gen Intel(R) Core(TM) i7-13700K|       3.0|         24|    8192|     64|    67064815616|
| 5|    [opencl:acc:0]|               Intel(R) FPGA Emulation Device|       1.2|         24|67108864|     64|    67064815616|

Attribute	Note
compute capability 1.3	Level-zero running time, recommended
compute capability 3.0	OpenCL running time, slower than level-zero in most cases

Device selection and execution of llama.cpp

There are two device selection modes:

Single device: Use one device assigned by user.
Multiple devices: Automatically choose the devices with the same biggest Max compute units.

Device selection	Parameter
Single device	--split-mode none --main-gpu DEVICE_ID
Multiple devices	--split-mode layer (default)

Examples:

Use device 0:

ZES_ENABLE_SYSMAN=1 ./build/bin/main -m models/llama-2-7b.Q4_0.gguf -p "Building a website can be done in 10 simple steps:" -n 400 -e -ngl 33 -sm none -mg 0

or run by script:

./examples/sycl/run_llama2.sh 0

Use multiple devices:

ZES_ENABLE_SYSMAN=1 ./build/bin/main -m models/llama-2-7b.Q4_0.gguf -p "Building a website can be done in 10 simple steps:" -n 400 -e -ngl 33 -sm layer

or run by script:

./examples/sycl/run_llama2.sh

Note:

By default, mmap is used to read model file. In some cases, it leads to the hang issue. Recommend to use parameter --no-mmap to disable mmap() to skip this issue.

Verify the device ID in output

Verify to see if the selected GPU is shown in the output, like:

detect 1 SYCL GPUs: [0] with top Max compute units:512

use 1 SYCL GPUs: [0] with Max compute units:512

Windows

Setup Environment

Install Intel GPU driver.

Please install Intel GPU driver by official guide: Install GPU Drivers.

Note: The driver is mandatory for compute function.

Install Visual Studio.

Please install Visual Studio which impact oneAPI environment enabling in Windows.

Install Intel® oneAPI Base toolkit.

a. Please follow the procedure in Get the Intel® oneAPI Base Toolkit .

Recommend to install to default folder: C:\Program Files (x86)\Intel\oneAPI.

Following guide uses the default folder as example. If you use other folder, please modify the following guide info with your folder.

b. Enable oneAPI running environment:

In Search, input 'oneAPI'.

Search & open "Intel oneAPI command prompt for Intel 64 for Visual Studio 2022"

In Run:

In CMD:

"C:\Program Files (x86)\Intel\oneAPI\setvars.bat" intel64

c. Check GPU

In oneAPI command line:

sycl-ls

There should be one or more level-zero devices. Please confirm that at least one GPU is present, like [ext_oneapi_level_zero:gpu:0].

Output (example):

[opencl:acc:0] Intel(R) FPGA Emulation Platform for OpenCL(TM), Intel(R) FPGA Emulation Device OpenCL 1.2  [2023.16.10.0.17_160000]
[opencl:cpu:1] Intel(R) OpenCL, 11th Gen Intel(R) Core(TM) i7-1185G7 @ 3.00GHz OpenCL 3.0 (Build 0) [2023.16.10.0.17_160000]
[opencl:gpu:2] Intel(R) OpenCL Graphics, Intel(R) Iris(R) Xe Graphics OpenCL 3.0 NEO  [31.0.101.5186]
[ext_oneapi_level_zero:gpu:0] Intel(R) Level-Zero, Intel(R) Iris(R) Xe Graphics 1.3 [1.3.28044]

Install cmake & make

a. Download & install cmake for Windows: https://cmake.org/download/

b. Download & install mingw-w64 make for Windows provided by w64devkit

Download the 1.19.0 version of w64devkit.
Extract w64devkit on your pc.
Add the bin folder path in the Windows system PATH environment, like C:\xxx\w64devkit\bin\.

Build locally:

In oneAPI command line window:

mkdir -p build
cd build
@call "C:\Program Files (x86)\Intel\oneAPI\setvars.bat" intel64 --force

::  for FP16
::  faster for long-prompt inference
::  cmake -G "MinGW Makefiles" ..  -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icx  -DCMAKE_BUILD_TYPE=Release -DLLAMA_SYCL_F16=ON

::  for FP32
cmake -G "MinGW Makefiles" ..  -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icx  -DCMAKE_BUILD_TYPE=Release


::  build example/main only
::  make main

::  build all binary
make -j
cd ..

.\examples\sycl\win-build-sycl.bat

Note:

By default, it will build for all binary files. It will take more time. To reduce the time, we recommend to build for example/main only.

Run

Put model file to folder models

You could download llama-2-7b.Q4_0.gguf as example.

Enable oneAPI running environment

In Search, input 'oneAPI'.

Search & open "Intel oneAPI command prompt for Intel 64 for Visual Studio 2022"

In Run:

In CMD:

"C:\Program Files (x86)\Intel\oneAPI\setvars.bat" intel64

List device ID

Run without parameter:

build\bin\ls-sycl-device.exe

or

build\bin\main.exe

Check the ID in startup log, like:

found 6 SYCL devices:
|  |                  |                                             |Compute   |Max compute|Max work|Max sub|               |
|ID|       Device Type|                                         Name|capability|units      |group   |group  |Global mem size|
|--|------------------|---------------------------------------------|----------|-----------|--------|-------|---------------|
| 0|[level_zero:gpu:0]|               Intel(R) Arc(TM) A770 Graphics|       1.3|        512|    1024|     32|    16225243136|
| 1|[level_zero:gpu:1]|                    Intel(R) UHD Graphics 770|       1.3|         32|     512|     32|    53651849216|
| 2|    [opencl:gpu:0]|               Intel(R) Arc(TM) A770 Graphics|       3.0|        512|    1024|     32|    16225243136|
| 3|    [opencl:gpu:1]|                    Intel(R) UHD Graphics 770|       3.0|         32|     512|     32|    53651849216|
| 4|    [opencl:cpu:0]|         13th Gen Intel(R) Core(TM) i7-13700K|       3.0|         24|    8192|     64|    67064815616|
| 5|    [opencl:acc:0]|               Intel(R) FPGA Emulation Device|       1.2|         24|67108864|     64|    67064815616|

Attribute	Note
compute capability 1.3	Level-zero running time, recommended
compute capability 3.0	OpenCL running time, slower than level-zero in most cases

Device selection and execution of llama.cpp

There are two device selection modes:

Single device: Use one device assigned by user.
Multiple devices: Automatically choose the devices with the same biggest Max compute units.

Device selection	Parameter
Single device	--split-mode none --main-gpu DEVICE_ID
Multiple devices	--split-mode layer (default)

Examples:

Use device 0:

build\bin\main.exe -m models\llama-2-7b.Q4_0.gguf -p "Building a website can be done in 10 simple steps:\nStep 1:" -n 400 -e -ngl 33 -s 0 -sm none -mg 0

Use multiple devices:

build\bin\main.exe -m models\llama-2-7b.Q4_0.gguf -p "Building a website can be done in 10 simple steps:\nStep 1:" -n 400 -e -ngl 33 -s 0 -sm layer

or run by script:

.\examples\sycl\win-run-llama2.bat

Note:

By default, mmap is used to read model file. In some cases, it leads to the hang issue. Recommend to use parameter --no-mmap to disable mmap() to skip this issue.

Verify the device ID in output

Verify to see if the selected GPU is shown in the output, like:

detect 1 SYCL GPUs: [0] with top Max compute units:512

use 1 SYCL GPUs: [0] with Max compute units:512

Environment Variable

Build

Name	Value	Function
LLAMA_SYCL	ON (mandatory)	Enable build with SYCL code path. For FP32/FP16, LLAMA_SYCL=ON is mandatory.
LLAMA_SYCL_F16	ON (optional)	Enable FP16 build with SYCL code path. Faster for long-prompt inference. For FP32, not set it.
CMAKE_C_COMPILER	icx	Use icx compiler for SYCL code path
CMAKE_CXX_COMPILER	icpx (Linux), icx (Windows)	use icpx/icx for SYCL code path

Running

Name	Value	Function
GGML_SYCL_DEBUG	0 (default) or 1	Enable log function by macro: GGML_SYCL_DEBUG
ZES_ENABLE_SYSMAN	0 (default) or 1	Support to get free memory of GPU by sycl::aspect::ext_intel_free_memory. Recommended to use when --split-mode = layer

Known Issue

Hang during startup

llama.cpp use mmap as default way to read model file and copy to GPU. In some system, memcpy will be abnormal and block.

Solution: add --no-mmap or --mmap 0.
Split-mode: [row] is not supported

It's on developing.

Q&A

Note: please add prefix [SYCL] in issue title, so that we will check it as soon as possible.

Error: error while loading shared libraries: libsycl.so.7: cannot open shared object file: No such file or directory.

Miss to enable oneAPI running environment.

Install oneAPI base toolkit and enable it by: source /opt/intel/oneapi/setvars.sh.
In Windows, no result, not error.

Miss to enable oneAPI running environment.
Meet compile error.

Remove folder build and try again.
I can not see [ext_oneapi_level_zero:gpu:0] afer install GPU driver in Linux.

Please run sudo sycl-ls.

If you see it in result, please add video/render group to your ID:
```
sudo usermod -aG render username
sudo usermod -aG video username
```
Then relogin.

If you do not see it, please check the installation GPU steps again.

Todo

Support row layer split for multiple card runs.

18 KiB Raw Blame History

llama.cpp for SYCL

Background

News

OS

Intel GPU

Verified

Memory

Nvidia GPU

Verified

oneMKL for CUDA

Docker

Build the image

Run

Linux

Setup Environment

Run

Windows

Setup Environment

Build locally:

Run

Environment Variable

Build

Running

Known Issue

Q&A

Todo

18 KiB

Raw Blame History