mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2024-12-25 02:44:36 +00:00
[SYCL] Initial cmake support of SYCL for AMD GPUs (#9658)
sycl: initial cmake support of SYCL for AMD GPUs
This commit is contained in:
Alberto Cabrera Pérez
GitHub
parent
00b7317e63
commit
f536f4c439
@ -26,7 +26,7 @@
|
||||
|
||||
### Llama.cpp + SYCL
|
||||
|
||||
The llama.cpp SYCL backend is designed to support **Intel GPU** firstly. Based on the cross-platform feature of SYCL, it could support other vendor GPUs: Nvidia GPU (*AMD GPU coming*).
|
||||
The llama.cpp SYCL backend is designed to support **Intel GPU** firstly. Based on the cross-platform feature of SYCL, it also supports other vendor GPUs: Nvidia and AMD.
|
||||
|
||||
## Recommended Release
|
||||
|
||||
@ -111,10 +111,18 @@ SYCL backend supports Intel GPU Family:
|
||||
|
||||
**Verified devices**
|
||||
|
||||
| Nvidia GPU | Status | Verified Model |
|
||||
|--------------------------|---------|----------------|
|
||||
| Ampere Series | Support | A100, A4000 |
|
||||
| Ampere Series *(Mobile)* | Support | RTX 40 Series |
|
||||
| Nvidia GPU | Status | Verified Model |
|
||||
|--------------------------|-----------|----------------|
|
||||
| Ampere Series | Supported | A100, A4000 |
|
||||
| Ampere Series *(Mobile)* | Supported | RTX 40 Series |
|
||||
|
||||
| AMD GPU | Status | Verified Model |
|
||||
|--------------------------|--------------|----------------|
|
||||
| Radeon Pro | Experimental | W6800 |
|
||||
| Radeon RX | Experimental | 6700 XT |
|
||||
|
||||
Note: AMD GPU support is highly experimental and is incompatible with F16.
|
||||
Additionally, it only supports GPUs with a sub_group_size (warp size) of 32.
|
||||
|
||||
## Docker
|
||||
The docker build option is currently limited to *intel GPU* targets.
|
||||
@ -186,6 +194,10 @@ Platform #0: Intel(R) OpenCL HD Graphics
|
||||
|
||||
In order to target Nvidia GPUs through SYCL, please make sure the CUDA/CUBLAS native requirements *-found [here](README.md#cuda)-* are installed.
|
||||
|
||||
- **AMD GPU**
|
||||
|
||||
To target AMD GPUs with SYCL, the ROCm stack must be installed first.
|
||||
|
||||
2. **Install Intel® oneAPI Base toolkit**
|
||||
|
||||
- **For Intel GPU**
|
||||
@ -212,6 +224,19 @@ cmake -B buildWithCublas -DCMAKE_CXX_COMPILER=icpx -DCMAKE_C_COMPILER=icx -DENAB
|
||||
cmake --build buildWithCublas --config Release
|
||||
```
|
||||
|
||||
- **Adding support to AMD GPUs**
|
||||
|
||||
**oneAPI Plugin**: In order to enable SYCL support on AMD GPUs, please install the [Codeplay oneAPI Plugin for AMD GPUs](https://developer.codeplay.com/products/oneapi/amd/download). As with Nvidia GPUs, the user should also make sure the plugin version matches the installed base toolkit.
|
||||
|
||||
**oneMKL for rocBlas**: The current oneMKL releases *(shipped with the oneAPI base-toolkit)* doesn't contain the rocBLAS backend. A build from source of the upstream [oneMKL](https://github.com/oneapi-src/oneMKL) with the *rocBLAS* backend enabled is thus required to run it on AMD GPUs.
|
||||
|
||||
```sh
|
||||
git clone https://github.com/oneapi-src/oneMKL
|
||||
cd oneMKL
|
||||
# Find your HIPTARGET with rocminfo, under the key 'Name:'
|
||||
cmake -B buildWithrocBLAS -DCMAKE_CXX_COMPILER=icpx -DCMAKE_C_COMPILER=icx -DENABLE_MKLGPU_BACKEND=OFF -DENABLE_MKLCPU_BACKEND=OFF -DENABLE_ROCBLAS_BACKEND=ON -DHIPTARGETS=${HIPTARGET} -DTARGET_DOMAINS=blas
|
||||
cmake --build buildWithrocBLAS --config Release
|
||||
```
|
||||
|
||||
3. **Verify installation and environment**
|
||||
|
||||
@ -223,22 +248,32 @@ sycl-ls
|
||||
|
||||
- **Intel GPU**
|
||||
|
||||
When targeting an intel GPU, the user should expect one or more level-zero devices among the available SYCL devices. Please make sure that at least one GPU is present, for instance [`ext_oneapi_level_zero:gpu:0`] in the sample output below:
|
||||
When targeting an intel GPU, the user should expect one or more level-zero devices among the available SYCL devices. Please make sure that at least one GPU is present, for instance [`level_zero:gpu`] in the sample output below:
|
||||
|
||||
```
|
||||
[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]
|
||||
[opencl:acc][opencl: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][opencl: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][opencl:2] Intel(R) OpenCL Graphics, Intel(R) Arc(TM) A770 Graphics OpenCL 3.0 NEO [23.30.26918.50]
|
||||
[level_zero:gpu][level_zero:0] Intel(R) Level-Zero, Intel(R) Arc(TM) A770 Graphics 1.3 [1.3.26918]
|
||||
```
|
||||
|
||||
- **Nvidia GPU**
|
||||
|
||||
Similarly, user targeting Nvidia GPUs should expect at least one SYCL-CUDA device [`ext_oneapi_cuda:gpu`] as bellow:
|
||||
Similarly, user targeting Nvidia GPUs should expect at least one SYCL-CUDA device [`cuda:gpu`] as below:
|
||||
|
||||
```
|
||||
[opencl:acc:0] Intel(R) FPGA Emulation Platform for OpenCL(TM), Intel(R) FPGA Emulation Device OpenCL 1.2 [2023.16.12.0.12_195853.xmain-hotfix]
|
||||
[opencl:cpu:1] Intel(R) OpenCL, Intel(R) Xeon(R) Gold 6326 CPU @ 2.90GHz OpenCL 3.0 (Build 0) [2023.16.12.0.12_195853.xmain-hotfix]
|
||||
[ext_oneapi_cuda:gpu:0] NVIDIA CUDA BACKEND, NVIDIA A100-PCIE-40GB 8.0 [CUDA 12.2]
|
||||
[opencl:acc][opencl:0] Intel(R) FPGA Emulation Platform for OpenCL(TM), Intel(R) FPGA Emulation Device OpenCL 1.2 [2023.16.12.0.12_195853.xmain-hotfix]
|
||||
[opencl:cpu][opencl:1] Intel(R) OpenCL, Intel(R) Xeon(R) Gold 6326 CPU @ 2.90GHz OpenCL 3.0 (Build 0) [2023.16.12.0.12_195853.xmain-hotfix]
|
||||
[cuda:gpu][cuda:0] NVIDIA CUDA BACKEND, NVIDIA A100-PCIE-40GB 8.0 [CUDA 12.5]
|
||||
```
|
||||
|
||||
- **AMD GPU**
|
||||
|
||||
For AMD GPUs we should expect at least one SYCL-HIP device [`hip:gpu`]:
|
||||
|
||||
```
|
||||
[opencl:cpu][opencl:0] Intel(R) OpenCL, 12th Gen Intel(R) Core(TM) i9-12900K OpenCL 3.0 (Build 0) [2024.18.6.0.02_160000]
|
||||
[hip:gpu][hip:0] AMD HIP BACKEND, AMD Radeon PRO W6800 gfx1030 [HIP 60140.9]
|
||||
```
|
||||
|
||||
### II. Build llama.cpp
|
||||
@ -266,6 +301,7 @@ cmake --build build --config Release -j -v
|
||||
```
|
||||
|
||||
#### Nvidia GPU
|
||||
|
||||
```sh
|
||||
# Export relevant ENV variables
|
||||
export LD_LIBRARY_PATH=/path/to/oneMKL/buildWithCublas/lib:$LD_LIBRARY_PATH
|
||||
@ -283,7 +319,25 @@ cmake -B build -DGGML_SYCL=ON -DGGML_SYCL_TARGET=NVIDIA -DCMAKE_C_COMPILER=icx -
|
||||
|
||||
# build all binary
|
||||
cmake --build build --config Release -j -v
|
||||
```
|
||||
|
||||
#### AMD GPU
|
||||
|
||||
```sh
|
||||
# Export relevant ENV variables
|
||||
export LD_LIBRARY_PATH=/path/to/oneMKL/buildWithrocBLAS/lib:$LD_LIBRARY_PATH
|
||||
export LIBRARY_PATH=/path/to/oneMKL/buildWithrocBLAS/lib:$LIBRARY_PATH
|
||||
export CPLUS_INCLUDE_DIR=/path/to/oneMKL/buildWithrocBLAS/include:$CPLUS_INCLUDE_DIR
|
||||
|
||||
# Build LLAMA with rocBLAS acceleration through SYCL
|
||||
|
||||
## AMD
|
||||
# Use FP32, FP16 is not supported
|
||||
# Find your GGML_SYCL_HIP_TARGET with rocminfo, under the key 'Name:'
|
||||
cmake -B build -DGGML_SYCL=ON -DGGML_SYCL_TARGET=AMD -DGGML_SYCL_HIP_TARGET=${GGML_SYCL_HIP_TARGET} -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx
|
||||
|
||||
# build all binary
|
||||
cmake --build build --config Release -j -v
|
||||
```
|
||||
|
||||
### III. Run the inference
|
||||
@ -586,11 +640,11 @@ use 1 SYCL GPUs: [0] with Max compute units:512
|
||||
|
||||
#### Build
|
||||
|
||||
| Name | Value | Function |
|
||||
|--------------------|-----------------------------------|---------------------------------------------|
|
||||
| GGML_SYCL | ON (mandatory) | Enable build with SYCL code path.<br>FP32 path - recommended for better perforemance than FP16 on quantized model|
|
||||
| GGML_SYCL_TARGET | INTEL *(default)* \| NVIDIA | Set the SYCL target device type. |
|
||||
| GGML_SYCL_F16 | OFF *(default)* \|ON *(optional)* | Enable FP16 build with SYCL code path. |
|
||||
| Name | Value | Function |
|
||||
|--------------------|---------------------------------------|---------------------------------------------|
|
||||
| GGML_SYCL | ON (mandatory) | Enable build with SYCL code path.<br>FP32 path - recommended for better perforemance than FP16 on quantized model|
|
||||
| GGML_SYCL_TARGET | INTEL *(default)* \| NVIDIA \| AMD | Set the SYCL target device type. |
|
||||
| GGML_SYCL_F16 | OFF *(default)* \|ON *(optional)* | Enable FP16 build with SYCL code path. |
|
||||
| CMAKE_C_COMPILER | `icx` *(Linux)*, `icx/cl` *(Windows)* | Set `icx` compiler for SYCL code path. |
|
||||
| CMAKE_CXX_COMPILER | `icpx` *(Linux)*, `icx` *(Windows)* | Set `icpx/icx` compiler for SYCL code path. |
|
||||
|
||||
|
||||
@ -511,8 +511,8 @@ if (GGML_HIPBLAS)
|
||||
endif()
|
||||
|
||||
if (GGML_SYCL)
|
||||
if (NOT GGML_SYCL_TARGET MATCHES "^(INTEL|NVIDIA)$")
|
||||
message(FATAL_ERROR "Invalid backend chosen, supported options are INTEL or NVIDIA")
|
||||
if (NOT GGML_SYCL_TARGET MATCHES "^(INTEL|NVIDIA|AMD)$")
|
||||
message(FATAL_ERROR "Invalid backend chosen, supported options are INTEL, NVIDIA, or AMD")
|
||||
endif()
|
||||
|
||||
check_cxx_compiler_flag("-fsycl" SUPPORTS_SYCL)
|
||||
@ -532,6 +532,9 @@ if (GGML_SYCL)
|
||||
list(APPEND GGML_CDEF_PUBLIC GGML_USE_SYCL)
|
||||
|
||||
if (GGML_SYCL_F16)
|
||||
if (GGML_SYCL_TARGET STREQUAL "AMD")
|
||||
message(WARNING "AMD target does not entirely support FP16 in the SYCL backend.")
|
||||
endif()
|
||||
add_compile_definitions(GGML_SYCL_F16)
|
||||
endif()
|
||||
|
||||
@ -543,6 +546,12 @@ if (GGML_SYCL)
|
||||
|
||||
if (GGML_SYCL_TARGET STREQUAL "NVIDIA")
|
||||
add_compile_definitions(GGML_SYCL_WARP_SIZE=32)
|
||||
elseif (GGML_SYCL_TARGET STREQUAL "AMD")
|
||||
# INFO: Allowed Sub_group_sizes are not consistent through all
|
||||
# hip targets. For example, 64 is used for certain models, but the backend
|
||||
# does not support it.
|
||||
# Target archs tested working: gfx1030, gfx1031, (Only tested sub_group_size = 32)
|
||||
add_compile_definitions(GGML_SYCL_WARP_SIZE=32)
|
||||
else()
|
||||
add_compile_definitions(GGML_SYCL_WARP_SIZE=16)
|
||||
endif()
|
||||
@ -576,6 +585,12 @@ if (GGML_SYCL)
|
||||
elseif (GGML_SYCL_TARGET STREQUAL "NVIDIA")
|
||||
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsycl-targets=nvptx64-nvidia-cuda")
|
||||
list(APPEND GGML_EXTRA_LIBS_PRIVATE sycl pthread m dl onemkl)
|
||||
elseif (GGML_SYCL_TARGET STREQUAL "AMD")
|
||||
if (GGML_SYCL_HIP_TARGET STREQUAL "")
|
||||
message(ERROR "Can't enable SYCL hip backend, GGML_SYCL_HIP_TARGET has not been set.")
|
||||
endif()
|
||||
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsycl-targets=amdgcn-amd-amdhsa -Xsycl-target-backend --offload-arch=${GGML_SYCL_HIP_TARGET}")
|
||||
list(APPEND GGML_EXTRA_LIBS_PRIVATE sycl pthread m dl onemkl)
|
||||
endif()
|
||||
endif()
|
||||
endif()
|
||||
|
||||
Loading…
Reference in New Issue
Block a user