Re: [Xen-devel] [xen-4.6-testing test] 107186: regressions - trouble: broken/fail/pass

[Andrew Cooper <andrew.cooper3@xxxxxxxxxx>]

On 05/04/17 10:13, Jan Beulich wrote:
>>>> On 05.04.17 at 06:27, <osstest-admin@xxxxxxxxxxxxxx> wrote:
>> flight 107186 xen-4.6-testing real [real]
>> http://logs.test-lab.xenproject.org/osstest/logs/107186/ 
>>
>> Regressions :-(
>>
>> Tests which did not succeed and are blocking,
>> including tests which could not be run:
>>  test-amd64-i386-freebsd10-amd64  3 host-install(3)     broken REGR. vs. 
>> 107151
>>  test-xtf-amd64-amd64-5   20 xtf/test-hvm32-invlpg~shadow fail REGR. vs. 
>> 107151
>>  test-xtf-amd64-amd64-5 33 xtf/test-hvm32pae-invlpg~shadow fail REGR. vs. 
>> 107151
>>  test-xtf-amd64-amd64-5   44 xtf/test-hvm64-invlpg~shadow fail REGR. vs. 
>> 107151
> Are these tests CPU vendor dependent in some way? The
> previous pass was on an Intel box, the failures here are on an
> AMD one.

This is because the XTF tests aren't locked to specific hardware.  I
pointed this problem out when the 5-way split was being considered.

These failures are wont-fix on AMD hardware on Xen 4.6, due to the
invasiveness of the backports, but having the different flights
scheduled across hardware will cause false negatives as shown here.

This is an OSS test bug, not a Xen bug.

~Andrew

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel