Red Hat Bugzilla – Bug 807016
RFE: Better power management support for KVM device assignment
Last modified: 2014-06-17 23:15:23 EDT
Description of problem:
The power management registers for an assigned device are read-only for the guest. This works fine for most drivers, but some (tg3) refuse to release the device because of being unable to transition the power state. Investigate either allowing power state transitions or emulating them (more likely).
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Attempt to hot-unplug and assigned BCM5755 or similar tg3 device
Guest refuses to release the device
libvirt fails to notice the hotunplug failed
This request was not resolved in time for the current release.
Red Hat invites you to ask your support representative to
propose this request, if still desired, for consideration in
the next release of Red Hat Enterprise Linux.
This request was erroneously removed from consideration in Red Hat Enterprise Linux 6.4, which is currently under development. This request will be evaluated for inclusion in Red Hat Enterprise Linux 6.4.
How can I get a copy of RHEL 7 to try now?
(In reply to comment #6)
> How can I get a copy of RHEL 7 to try now?
There's no beta available yet. The support we'll be targeting for this is the new vfio-based device assignment where power management is already accessible to userspace. The 3.9 kernel has a fix to enable this and an additional fix will be in QEMU 1.5. Please contact me if you need help getting vfio setup. There's not yet libvirt support, but it should be available in the next upstream release. Thanks.
For RHEL7, this functionality requested by this BZ is provided by VFIO,
which is available in the 3.10 kernel and qemu 1.5.
Setting this BZ to POST (on Alex's behalf).
This request was resolved in Red Hat Enterprise Linux 7.0.
Contact your manager or support representative in case you have further questions about the request.