gered/ollama
1
0
Fork 0
Code Issues Pull requests Actions Packages Projects Releases Wiki Activity

Improve maintainability of Radeon card list

Browse source 
This moves the list of AMD GPUs to an easier to maintain list which
should make it easier to update over time.
This commit is contained in:
Daniel Hiltgen 2024-01-03 15:12:29 -08:00
parent c7ea8f237e
commit 16f4603b67
1 changed files with 23 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

24
llm/llama.cpp/gen_linux.sh
View file

@ -16,6 +16,28 @@
set -ex
set -o pipefail
# See https://llvm.org/docs/AMDGPUUsage.html#processors for reference
amdGPUs() {
GPU_LIST=(
"gfx803"
"gfx900"
"gfx906:xnack-"
"gfx908:xnack-"
"gfx90a:xnack+"
"gfx90a:xnack-"
"gfx1010"
"gfx1012"
"gfx1030"
"gfx1100"
"gfx1101"
"gfx1102"
)
(
IFS=$';'
echo "'${GPU_LIST[*]}'"
)
}
echo "Starting linux generate script"
if [ -z "${CUDACXX}" -a -x /usr/local/cuda/bin/nvcc ]; then
export CUDACXX=/usr/local/cuda/bin/nvcc
@ -72,7 +94,7 @@ fi
if [ -d "${ROCM_PATH}" ]; then
echo "ROCm libraries detected - building dynamic ROCm library"
init_vars
CMAKE_DEFS="${COMMON_CMAKE_DEFS} ${CMAKE_DEFS} -DLLAMA_HIPBLAS=on -DCMAKE_C_COMPILER=$ROCM_PATH/llvm/bin/clang -DCMAKE_CXX_COMPILER=$ROCM_PATH/llvm/bin/clang++ -DAMDGPU_TARGETS='gfx803;gfx900;gfx906:xnack-;gfx908:xnack-;gfx90a:xnack+;gfx90a:xnack-;gfx1010;gfx1012;gfx1030;gfx1100;gfx1101;gfx1102' -DGPU_TARGETS='gfx803;gfx900;gfx906:xnack-;gfx908:xnack-;gfx90a:xnack+;gfx90a:xnack-;gfx1010;gfx1012;gfx1030;gfx1100;gfx1101;gfx1102'"
CMAKE_DEFS="${COMMON_CMAKE_DEFS} ${CMAKE_DEFS} -DLLAMA_HIPBLAS=on -DCMAKE_C_COMPILER=$ROCM_PATH/llvm/bin/clang -DCMAKE_CXX_COMPILER=$ROCM_PATH/llvm/bin/clang++ -DAMDGPU_TARGETS=$(amdGPUs) -DGPU_TARGETS=$(amdGPUs)"
BUILD_DIR="gguf/build/linux/rocm"
build
install