Re: [Xen-devel] [xen-unstable test] 146578: regressions - FAIL

[Jan Beulich <jbeulich@xxxxxxxx>]

On 30.01.2020 10:18, Roger Pau Monné wrote:
> On Thu, Jan 30, 2020 at 10:12:39AM +0100, Jan Beulich wrote:
>> On 30.01.2020 04:56, osstest service owner wrote:
>>> flight 146578 xen-unstable real [real]
>>> http://logs.test-lab.xenproject.org/osstest/logs/146578/
>>>
>>> Regressions :-(
>>>
>>> Tests which did not succeed and are blocking,
>>> including tests which could not be run:
>>>  test-amd64-amd64-xl-qemuu-win7-amd64  7 xen-boot         fail REGR. vs. 
>>> 146563
>>
>> Afaict it is not "xen-boot" which failed here, but the going
>> down of the prior native kernel which was used to do the
>> install:
>>
>> Jan 29 23:19:34.370663 [  182.472073] e1000e: eth0 NIC Link is Down
>> Jan 29 23:19:36.470587 [  182.476728] kvm: exiting hardware virtualization
>> Jan 29 23:19:36.470660 [  182.481574] sd 3:0:0:0: [sdb] Synchronizing SCSI 
>> cache
>> Jan 29 23:19:36.482617 [  182.486874] sd 2:0:0:0: [sda] Synchronizing SCSI 
>> cache
>> Jan 29 23:19:36.482683 [  182.528263] e1000e: EEE TX LPI TIMER: 00000000
>> Jan 29 23:19:36.530516 [  182.537036] e1000e: EEE TX LPI TIMER: 00000000
>> Jan 29 23:19:36.530582 [  182.544504] usb 6-2: USB disconnect, device number 
>> 2
>> Jan 29 23:19:36.542579 [  182.624125] reboot: Restarting system
>> Jan 29 23:19:36.614554 [  182.625527] AMD-Vi: Event logged [
>> Jan 29 23:19:36.626671 [  182.625527] IO_PAGE_FAULT device=00:13.1 
>> domain=0x0006 address=0x0000000000000080 flags=0x0020]
>>
>> These last two lines then look to repeat indefinitely. Is
>> there a new kernel version in use which has regressed on this
>> particular hardware?
> 
> osstest was recently upgraded to use Linux 4.19, but that seems to be
> a sporadic failure, other tests from the xen-unstable branch seem to
> work fine:
> 
> http://logs.test-lab.xenproject.org/osstest/results/host/pinot0.html
> 
> Maybe we could disable the iommu on the Linux command line to
> workarouund this?

Well, I don't know in how far there is control over it. It's after
all not something actually mattering for the intended testing. But
yes, if there is such control, then this might be a (n ugly) option.

Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel