nikola/llama_cpp
1
0
Fork 0
Code Issues Pull Requests Actions 18 Packages Projects Releases Wiki Activity

[SYCL] Enhance build script to use half cores to build, avoid OS hang (#21093)

Browse Source 
* use half cores to build, avoid OS hang

* reduce the output text num to short test time

* avoid to return 0
This commit is contained in:
Neo Zhang
2026-03-29 09:02:45 +08:00
committed by GitHub
parent 65097181e4
commit afe65aa282
2 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/sycl/run-llama2.sh
+2 -2
View File
@@ -23,9 +23,9 @@ if [ $# -gt 0 ]; then
GGML_SYCL_DEVICE=$1
echo "use $GGML_SYCL_DEVICE as main GPU"
#use signle GPU only
ZES_ENABLE_SYSMAN=1 ./build/bin/llama-completion -m ${MODEL_FILE} -no-cnv -p "${INPUT_PROMPT}" -n 400 -e -ngl ${NGL} -s 0 -c ${CONTEXT} -mg $GGML_SYCL_DEVICE -sm none ${LOAD_MODE}
ZES_ENABLE_SYSMAN=1 ./build/bin/llama-completion -m ${MODEL_FILE} -no-cnv -p "${INPUT_PROMPT}" -n 200 -e -ngl ${NGL} -s 0 -c ${CONTEXT} -mg $GGML_SYCL_DEVICE -sm none ${LOAD_MODE}
else
#use multiple GPUs with same max compute units
ZES_ENABLE_SYSMAN=1 ./build/bin/llama-completion -m ${MODEL_FILE} -no-cnv -p "${INPUT_PROMPT}" -n 400 -e -ngl ${NGL} -s 0 -c ${CONTEXT} ${LOAD_MODE}
ZES_ENABLE_SYSMAN=1 ./build/bin/llama-completion -m ${MODEL_FILE} -no-cnv -p "${INPUT_PROMPT}" -n 200 -e -ngl ${NGL} -s 0 -c ${CONTEXT} ${LOAD_MODE}
fi