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 14, 05:30 UTC
[CI][TPU] Fix docker clean up (#22271)
Signed-off-by: Siyuan Liu <lsiyuan@google.com>
Siyuan Liu committed
Aug 5, 2025 at 16:54 UTC
4b29d2784b3753fd5434cded25cbcf0bce7b7da7
5 files changed
+3
-7
.buildkite/scripts/hardware_ci/run-tpu-v1-test-part2.sh
+1
-2
@@ -4,8 +4,7 @@ set -xu
4
5
6
remove_docker_container() {
7
- docker rm -f tpu-test || true;
8
- docker rm -f vllm-tpu || true;
7
+ docker rm -f tpu-test || true;
8
}
9
10
trap remove_docker_container EXIT
.buildkite/scripts/hardware_ci/run-tpu-v1-test.sh
-1
@@ -5,7 +5,6 @@ set -xu
5
6
remove_docker_container() {
7
docker rm -f tpu-test || true;
8
- docker rm -f vllm-tpu || true;
8
}
9
10
trap remove_docker_container EXIT
.buildkite/scripts/tpu/config_v6e_1.env
+1
-1
@@ -1,6 +1,6 @@
1
# Environment config
2
TEST_NAME=llama8b
3
-CONTAINER_NAME=vllm-tpu
3
+CONTAINER_NAME=tpu-test
4
5
# vllm config
6
MODEL=meta-llama/Llama-3.1-8B-Instruct
.buildkite/scripts/tpu/docker_run_bm.sh
-2
@@ -12,8 +12,6 @@ source /etc/environment
12
source $ENV_FILE
13
14
remove_docker_container() {
15
- docker rm -f tpu-test || true;
16
- docker rm -f vllm-tpu || true;
15
docker rm -f $CONTAINER_NAME || true;
16
}
17
.buildkite/scripts/tpu/quantized_v6e_1.env
+1
-1
@@ -1,6 +1,6 @@
1
# Environment config
2
TEST_NAME=llama8bw8a8
3
-CONTAINER_NAME=vllm-tpu
3
+CONTAINER_NAME=tpu-test
4
5
# vllm config
6
MODEL=RedHatAI/Meta-Llama-3.1-8B-Instruct-quantized.w8a8