Re: [Xen-devel] BUG: soft lockup - CPU#1 stuck for 61s! [udevd:4865]

[Jeremy Fitzhardinge <jeremy@xxxxxxxx>]

 On 09/02/2010 05:20 PM, James Harper wrote:
> I've had "BUG: soft lockup - CPU#1 stuck for 61s! [udevd:4865]" happen
> twice now and the server fairly quickly (over the course of 5-10
> minutes) becomes unusable after that happens. The call trace appears to
> be:
>
> Call Trace:
>  [<ffffffff8100e8e9>] ? xen_force_evtchn_callback+0x9/0xa
>  [<ffffffff8100ef72>] ? check_events+0x12/0x20
>  [<ffffffff8100ef19>] ? xen_irq_enable_direct_end+0x0/0x7
>  [<ffffffff8102fcc1>] ? do_page_fault+0x9e/0x27b
>  [<ffffffff812b39a5>] ? page_fault+0x25/0x30
>  [<ffffffff8116f28d>] ? __put_user_4+0x1d/0x30
>  [<ffffffff81045265>] ? schedule_tail+0x92/0x96
>  [<ffffffff81011983>] ? ret_from_fork+0x13/0x80
>
> That was under 2.6.32.17-g69a73fa. I've since upgraded to
> 2.6.32.18-ge6b9b2c but the changelogs don't show anything obviously
> related to the crash (but as usual there are a lot of them so maybe I've
> missed something?)
>
> Is this a known problem under 2.6.32.17-g69a73fa that has since been
> fixed?

Hard to know; nothing springs to mind right now.  What else is going on
at the time?  What's the full softlockup message?  Is there any more
context?

There have been some fixes relating to long-standing interrupt migration
bugs, but they're after 69a73fa.

    J

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel