------- Comment From boris.m...@de.ibm.com 2025-03-25 06:12 EDT------- Solution (provided by <schne...@linux.ibm.com>):
The first part of the problem is solved by not ignoring PFs in standby. The second issue is solved by explicitly checking if the parent PF of a given VF is configured and enabled. Since a PFn can only have child VFs while enabled this is always the case if this configuration has the parent PF. Reproduction: The first part of the issue can be reproduced by putting a PF in standby, then reboot, configure it and try creating child VFs which then appear in their own PCI domains. For the second part two LPARs with both PFs+VFs from a single card sharing candidate lists are required. Then a VF with its parent PF in the other LPAR will incorrectly be placed in the shared PCI domain of its "uncle PF". Upstream-ID: dc287e4c9149ab54a5003b4d4da007818b5fda3d 05793884a1f30509e477de9da233ab73584b1c8c 2844ddbd540fc84d7571cca65d6c43088e4d6952 Preventive: yes -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/2104120 Title: [Ubuntu 25.04] s390/pci: Fix PCHID per port multi-function detection To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu-z-systems/+bug/2104120/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
