I am experiencing this issue on a new xeon scalable machine. I can confirm that it is present with xenial linux-image-generic, and linux-image-generic-hwe, kernels 4.4.0-116, and 4.13.0-38, and I can reproduce it at will.
It no longer seems to occur on kernel 4.15.0-13. In all cases, clocksource was already set to TSC automatically on install. The soft lockup and hard lockup were triggered each time by a waf build of the ns3 project. In each case, a failure occurred while running the commands as a regular user, but the system was stable if run as root, or with sudo. ** Attachment added: "lspci-vnvn.log" https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1596866/+attachment/5100766/+files/lspci-vnvn.log -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1596866 Title: NMI watchdog: Watchdog detected hard LOCKUP on cpu 0 - Xenial - Python To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1596866/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
