diff options
author | Neo Zhang Jianyu <jianyu.zhang@intel.com> | 2024-04-14 10:42:29 +0800 |
---|---|---|
committer | GitHub <noreply@github.com> | 2024-04-14 10:42:29 +0800 |
commit | de17e3f7455dc7fd298cc61d86798533b9ca7a29 (patch) | |
tree | 65595beccf54a5db5a31c3dff5fd0b519c75b275 /examples | |
parent | b5e7285baffb0da8a6619567b52d8e67de41291d (diff) |
fix memcpy() crash, add missed cmd in guide, fix softmax (#6622)
* disable mmap to fix memcpy crash, add missed cmd in guide, fix softmax
* refactor to disable mmap for SYCL backend
* fix compile error in other os
* refactor the solution, use host buf to fix it, instead of disable mmap
* keep to support mmap()
* use host buff to reduce malloc times
* revert to malloc/free solution, for threaad safe
Diffstat (limited to 'examples')
-rwxr-xr-x | examples/sycl/build.sh | 2 | ||||
-rwxr-xr-x | examples/sycl/run-llama2.sh | 1 |
2 files changed, 2 insertions, 1 deletions
diff --git a/examples/sycl/build.sh b/examples/sycl/build.sh index f20391d7..db46d57c 100755 --- a/examples/sycl/build.sh +++ b/examples/sycl/build.sh @@ -20,4 +20,4 @@ cmake .. -DLLAMA_SYCL=ON -DCMAKE_C_COMPILER=icx -DCMAKE_CXX_COMPILER=icpx #cmake --build . --config Release --target llama-bench #build all binary -cmake --build . --config Release -v +cmake --build . --config Release -j -v diff --git a/examples/sycl/run-llama2.sh b/examples/sycl/run-llama2.sh index c979a52f..7b39a18c 100755 --- a/examples/sycl/run-llama2.sh +++ b/examples/sycl/run-llama2.sh @@ -12,6 +12,7 @@ if [ $# -gt 0 ]; then GGML_SYCL_SINGLE_GPU=1 else GGML_SYCL_DEVICE=0 + GGML_SYCL_SINGLE_GPU=0 fi #export GGML_SYCL_DEBUG=1 |