Message ID | 20210728155220.4234-2-taihsiang.ho@canonical.com |
---|---|
State | New |
Headers | show |
Series | ubuntu_performance_deep_learning: fix TTY error | expand |
Acked-by: Po-Hsu Lin <po-hsu.lin@canonical.com>
Applied and pushed.
Thanks
Sam
diff --git a/ubuntu_performance_deep_learning/ubuntu_performance_tensor_flow.sh b/ubuntu_performance_deep_learning/ubuntu_performance_tensor_flow.sh index 77efaf39..a0ba5670 100755 --- a/ubuntu_performance_deep_learning/ubuntu_performance_tensor_flow.sh +++ b/ubuntu_performance_deep_learning/ubuntu_performance_tensor_flow.sh @@ -33,7 +33,7 @@ run_test() { --shm-size=1g \ --ulimit memlock=-1 \ --ulimit stack=67108864 \ - -ti --rm nvcr.io/nvidia/tensorflow:${CONTAINER_VER}-tf1-py3 -- \ + --rm nvcr.io/nvidia/tensorflow:${CONTAINER_VER}-tf1-py3 -- \ mpiexec \ --bind-to socket \ --allow-run-as-root \
It is unnecessary to invoke nvidia-docker, which is essentially a wrapper of docker executable, with an allocated pseudo-TTY when automating this TensorFlow framework test job, otherwise such TTY error message may raise if the automation job is launched without TTY: the input device is not a TTY Signed-off-by: Taihsiang Ho (tai271828) <taihsiang.ho@canonical.com> --- .../ubuntu_performance_tensor_flow.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)