@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / ff39141a49
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 16, 02:00 UTC

[HPU][misc] add comments for explanation (#12034)

Signed-off-by: youkaichao <youkaichao@gmail.com>

youkaichao committed Jan 14, 2025 at 19:24 UTC ff39141a49928540c3975cc8e2a6e9bfa51f0bef
1 file changed +5
.buildkite/run-hpu-test.sh
+5
@@ -8,6 +8,11 @@ set -ex
8 docker build -t hpu-test-env -f Dockerfile.hpu .
9
10 # Setup cleanup
11 +# certain versions of HPU software stack have a bug that can
12 +# override the exit code of the script, so we need to use
13 +# separate remove_docker_container and remove_docker_container_and_exit
14 +# functions, while other platforms only need one remove_docker_container
15 +# function.
16 EXITCODE=1
17 remove_docker_container() { docker rm -f hpu-test || true; }
18 remove_docker_container_and_exit() { remove_docker_container; exit $EXITCODE; }