The condition for css_sch_device_unregister(sch) also caught my eye, calling it unconditionally is probably closer to right because it was called in the !cdev case before, and in the attached patch it would no longer be called in this case. However I think in the short term the revert is the safest option, since the code will match what we already know was working in the groovy kernel. Once a fix is committed upstream, we can trade out the revert for that patch.
-- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1925211 Title: Hot-unplug of disks leaves broken block devices around in Hirsute on s390x To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu-z-systems/+bug/1925211/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
