leezu commented on a change in pull request #20:
URL: https://github.com/apache/incubator-mxnet-ci/pull/20#discussion_r417731484



##########
File path: tools/jenkins-slave-creation-unix/conf-ubuntu-gpu/install.sh
##########
@@ -59,8 +59,13 @@ echo "Installed htop, java, git and python"
 #Install nvidia drivers
 #Chose the latest nvidia driver supported on Tesla driver for Ubuntu18.04
 #Refer : https://www.nvidia.com/Download/driverResults.aspx/158191/en-us
-sudo apt-get -y install nvidia-driver-435
-sudo apt-get -y install nvidia-utils-435
+wget 
https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/cuda-ubuntu1804.pin
+sudo mv cuda-ubuntu1804.pin /etc/apt/preferences.d/cuda-repository-pin-600
+wget 
http://developer.download.nvidia.com/compute/cuda/10.2/Prod/local_installers/cuda-repo-ubuntu1804-10-2-local-10.2.89-440.33.01_1.0-1_amd64.deb

Review comment:
       Well, you need to download essentially all cuda related packages once to 
create the local repo. Whereas otherwise you only download the required 
packages.
   This is only run once during AMI creation, so it may be better not to 
hardcode `10.2.89-440.33.01_1.0-1_amd64` version but rely on the latest version 
of cuda 10.2 available at time of ami creation?




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]


Reply via email to