Re: I/O scheduler deadlocks on Xen virtual block devices

[Doug Neal]

On Fri, Sep 3, 2010 at 4:17 PM, Jeremy Fitzhardinge <jeremy@xxxxxxxx> wrote:
>  On 09/03/2010 05:22 AM, Pasi Kärkkäinen wrote:
>> On Wed, Sep 01, 2010 at 03:17:25PM +0100, Doug Neal wrote:
>>> Hello all,
>>>
>>> I believe I have found a bug in the I/O scheduler which results in
>>> access to a block device being blocked indefinitely.
>>>
>>> The setup:
>>>  * Xen dom0 version 3.4.2.
>>>  * domU: Ubuntu 10.04, x86_64, with kernel 2.6.32.15+drm33.5.
>>>  * Paravirtual disks and network interfaces.
>>>  * Root filesystem on /dev/xvdb1
>>>  * A scratch filesystem for the purposes of my tests on /dev/xvda
>>> mounted on /mnt
>>>  * Both filesystems are ext3, formatted and mounted with defaults
>>>  * XVBDs are backed by LVM on top of an iSCSI SAN in the dom0.
>>>
>> Hmm.. recently there was some patches to the Xen block device frontend driver
>> (sent to lkml). Maybe you should try if those make a difference?
>>
>> Jeremy: What do you think?
>
> I think this is more likely to be related to the lost interrupt bug
> which I sent fixes up for last week ("xen: use percpu interrupts for
> IPIs and VIRQs" and "xen: handle events as edge-triggered").
>
> Doug, as a workaround, try disabling irqbalanced.
>
> Thanks,
>    J

Jeremy,

That seems to be doing the trick - thanks! Will see about trying the
patches as well.

Doug
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/