NVIDIA/cuda-quantum

Execution hangs on OpenSUSE when installed via installer

Opened this issue · 1 comments

Required prerequisites

  • Consult the security policy. If reporting a security vulnerability, do not report the bug using this form. Use the process described in the policy to report the issue.
  • Make sure you've read the documentation. Your issue may be addressed there.
  • Search the issue tracker to verify that this hasn't already been reported. +1 or comment there if it has.
  • If possible, make a PR with a failing test to give us a starting point to work on!

Describe the bug

Upon validating the libc++ support, I realized that the execution on GPU-accelerated backends on OpenSUSE simply seems to hang. It succeeds at some point, but clearly only after something timed out.

Steps to reproduce the bug

The issue is reproduceable following the validation our pipelines do (build docker/releases/installer.Dockerfile, then docker/test/installer/linux.Dockerfile with the arguments defined by .github/workflows/config/validation_config.json), then try running on the nvidia backend.

Expected behavior

The execution should start immediately and be as fast as other operating systems.

Is this a regression? If it is, put the last known working version (or commit) here.

Not a regression

Environment

  • CUDA Quantum version: 0.5.0 - 0.7.1
  • Python version: N/A
  • C++ compiler: CUDA-Q built with GCC 11
  • Operating system: OpenSUSE 15

Suggestions

The issue is fixed by the changes made in #1603.

Is there a link to a log file available?