-
Notifications
You must be signed in to change notification settings - Fork 184
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
CUDA RT IB fails to build onnxruntime
pkg
#8571
Comments
A new Issue was created by @aandvalenzuela Andrea Valenzuela. @Dr15Jones, @perrotta, @dpiparo, @rappoccio, @makortel, @smuzaffar can you please review it and eventually sign/assign? Thanks. cms-bot commands are listed here |
Tensorflow fails with the same error. |
looks like this IB is using cuda 11.5 . please update it to cude 11.8.0 version (I think automatic forward port of #8545 to cudart branch did not work as cuda spec are different in master and cudart branches) |
also make sure that cuda 11.8.0 is installed in special cvmfs area ( /cvmfs/projects.cern.ch/cms-restricted/$(uname -m)/rhel8/external/cuda ) , if it is not then you can run https://cmssdt.cern.ch/jenkins/view/CVMFS-CMS/job/cvmfs-install-cuda/ job to install it |
Closing since CUDA update solved the issue. Thanks! |
Hello,
The dedicated IB for cuda runtime fails to build
onnxruntime
pkg. From the latest CUDART IB log:It seems this issue raised as of the
gcc
update to 11.4.1 (#8545).I think it is the same issue exposed in NVIDIA/nccl#650 (comment). I am moving forward to apply the proposed patch to see if it solves the issue.
FYI, @smuzaffar @fwyzard
The text was updated successfully, but these errors were encountered: