masahi commented on pull request #10758: URL: https://github.com/apache/tvm/pull/10758#issuecomment-1086274838
No idea, this happens only when linked with `libtorch.so`, right? https://discuss.pytorch.org/t/building-pytorch-from-source-failed-on-ubuntu-16-04/43026 suggests our cuda install is broken, but since CI has passed with PT 1.11, I don't think that's the case. One thing that could be related is that our `ci-gpu` is using fairly old cuda - 11.0 according to https://github.com/apache/tvm/blob/main/docker/Dockerfile.ci_gpu#L20. Maybe this is incompatible with the latest `libtorch.so`? Even if that's the case, I don't get why we didn't get any errors on the CI test. In the worst case, we can install CPU-version of PT, since we don't need cuda capability for CI purpose. cc @driazati -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
