Note: This is a public test instance of Red Hat Bugzilla. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback at bugzilla.redhat.com.
Bug 1227877 - SMT guests enter an infinite loop at boot time on POWER7
Summary: SMT guests enter an infinite loop at boot time on POWER7
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Fedora
Classification: Fedora
Component: kernel
Version: 21
Hardware: ppc64
OS: All
unspecified
urgent
Target Milestone: ---
Assignee: Kernel Maintainer List
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2015-06-03 17:10 UTC by IBM Bug Proxy
Modified: 2015-06-21 00:31 UTC (History)
7 users (show)

Fixed In Version: kernel-4.0.5-300.fc22
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2015-06-20 23:59:16 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
IBM Linux Technology Center 125850 0 None None None Never

Description IBM Bug Proxy 2015-06-03 17:10:17 UTC

Comment 1 IBM Bug Proxy 2015-06-03 17:10:19 UTC
== Comment: #0 - Greg Kurz <KURZGREG.com> - 2015-06-03 12:01:08 ==
If I try to start a SMT capable guest (threads > 1 in the CPU topology) on a POWER7 fedora hypervisor, the guest kernel ends up looping and won't boot further than the "Enabling Asymmetric SMT scheduling" message.

It is a bug in the powernv code that is related to the following commits:

7cba160ad "powernv/cpuidle: Redesign idle states management"
77b54e9f2 "powernv/powerpc: Add winkle support for offline cpus"

I could verify that changes from 7cba160ad and 77b54e9f2  are present in the following kernels:

 fedora 20: 3.19.5-100.fc20.ppc64p7
 fedora 21: 3.19.5-200.fc21.ppc64p7 and 4.0.4-202.fc21.ppc64p7
 fedora 22: 4.0.4-303.fc22.ppc64p7

It was fixed recently upstream:

0aab37470 "powerpc/powernv: Restore non-volatile CRs after nap"

The above commit applies flawlessly to all fedora kernels and fixes the issue.

Comment 2 Josh Boyer 2015-06-04 11:34:10 UTC
Thanks for the pointer to the patch.  I've applied it to the F20-F22 branches.  Rawhide already contains the commit.

Comment 3 Fedora Update System 2015-06-09 12:57:54 UTC
kernel-4.0.5-300.fc22 has been submitted as an update for Fedora 22.
https://admin.fedoraproject.org/updates/kernel-4.0.5-300.fc22

Comment 4 Fedora Update System 2015-06-09 12:59:44 UTC
kernel-4.0.5-200.fc21 has been submitted as an update for Fedora 21.
https://admin.fedoraproject.org/updates/kernel-4.0.5-200.fc21

Comment 5 IBM Bug Proxy 2015-06-10 08:30:34 UTC
------- Comment From KURZGREG.com 2015-06-10 08:21 EDT-------
(In reply to comment #5)
> kernel-4.0.5-300.fc22 has been submitted as an update for Fedora 22.
> https://admin.fedoraproject.org/updates/kernel-4.0.5-300.fc22
>
> kernel-4.0.5-200.fc21 has been submitted as an update for Fedora 21.
> https://admin.fedoraproject.org/updates/kernel-4.0.5-200.fc21

When will the ppc64p7 packages be available ?

Comment 6 IBM Bug Proxy 2015-06-10 09:31:30 UTC
------- Comment From KURZGREG.com 2015-06-10 09:27 EDT-------
(In reply to comment #6)
> (In reply to comment #5)
> > kernel-4.0.5-300.fc22 has been submitted as an update for Fedora 22.
> > https://admin.fedoraproject.org/updates/kernel-4.0.5-300.fc22
> >
> > kernel-4.0.5-200.fc21 has been submitted as an update for Fedora 21.
> > https://admin.fedoraproject.org/updates/kernel-4.0.5-200.fc21
>
> When will the ppc64p7 packages be available ?

FWIW I have rebuilt kernel-4.0.5-200.fc21 for ppc64p7 from the source RPM and tested: the issue is fixed.

Comment 7 Josh Boyer 2015-06-10 13:54:30 UTC
The ppc64p7 packages are built by the powerpc secondary arch team.  I have no idea when they'll ship updates that include this.  From a Fedora kernel standpoint, there's nothing left for us to do.

Comment 8 IBM Bug Proxy 2015-06-10 14:50:33 UTC
------- Comment From KURZGREG.com 2015-06-10 14:45 EDT-------
(In reply to comment #8)
> The ppc64p7 packages are built by the powerpc secondary arch team.  I have
> no idea when they'll ship updates that include this.  From a Fedora kernel
> standpoint, there's nothing left for us to do.

I guess we can close the bug then.

Comment 9 IBM Bug Proxy 2015-06-10 15:00:32 UTC
------- Comment From lagarcia.com 2015-06-10 14:53 EDT-------
Gustavo,

Would you be able to tell us when this src.rpm will be built into a ppc64p7 binary RPM so that we can go ahead and close this bug?

------- Comment From gusld.com 2015-06-10 14:59 EDT-------
(In reply to comment #10)
> Gustavo,
>
> Would you be able to tell us when this src.rpm will be built into a ppc64p7
> binary RPM so that we can go ahead and close this bug?

Releasing updates is a manual process that we do regularly. We try to do that every few weeks... I will check with the release engineer when he is planning to release the next batch of updates.

[]'s
Gustavo

Comment 10 Fedora Update System 2015-06-10 19:09:33 UTC
Package kernel-4.0.5-200.fc21:
* should fix your issue,
* was pushed to the Fedora 21 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing kernel-4.0.5-200.fc21'
as soon as you are able to, then reboot.
Please go to the following url:
https://admin.fedoraproject.org/updates/FEDORA-2015-9704/kernel-4.0.5-200.fc21
then log in and leave karma (feedback).

Comment 11 Fedora Update System 2015-06-20 23:59:16 UTC
kernel-4.0.5-200.fc21 has been pushed to the Fedora 21 stable repository.  If problems still persist, please make note of it in this bug report.

Comment 12 Fedora Update System 2015-06-21 00:31:47 UTC
kernel-4.0.5-300.fc22 has been pushed to the Fedora 22 stable repository.  If problems still persist, please make note of it in this bug report.


Note You need to log in before you can comment on or make changes to this bug.