Re: [linux-3.18 test] 59041: regressions - trouble: blocked/broken/fail/pass

[Ian Campbell <ian.campbell@citrix.com>]

On Mon, 2015-07-06 at 11:22 -0400, Boris Ostrovsky wrote:
> On 07/06/2015 04:32 AM, Ian Campbell wrote:
> > On Fri, 2015-07-03 at 22:21 +0000, osstest service owner wrote:
> >> flight 59041 linux-3.18 real [real]
> >> http://logs.test-lab.xenproject.org/osstest/logs/59041/
> >>
> >> Regressions :-(
> >>
> >> Tests which did not succeed and are blocking,
> >> including tests which could not be run:
> >>   test-amd64-amd64-xl-pvh-intel 11 guest-start              fail REGR. vs. 58581
> > Not really sure who is on the hook for pvh fails these days, copying
> > some likely suspects + x86 maintainers. Looks like the guest is triple
> > faulting.
> 
> 
> Looks like something happened between 2015-06-15 and 2015-06-29 (flights 
> 58581 and 58976)

Yes, and a bit of the original report I should have quoted is:

version targeted for testing:
 linux                ea5dd38e93b3bec3427e5d3eef000bbf5d637e76
baseline version:
 linux                d048c068d00da7d4cfa5ea7651933b99026958cf

See also the complete history of this test on this branch at
http://logs.test-lab.xenproject.org/osstest/results/history/test-amd64-amd64-xl-pvh-intel/linux-3.18.html

>  and it's only 3.18 kernel, right?

As far as I know. But: older linux-X.Y doesn't support PVH, so this test
has always failed at guest-start in those flights, before we would have
gotten to this failure.

The only newer stable branch which we test is linux-4.1, flight 59054
from didn't show this issue, but the next one might pick up something
newer, or 4.1.(y+1) might still be in the pipeline upstream and  the
issue will show up later.

We also test linux-linus and linux-next, so far neither of them seems to
exhibit this problem:
http://logs.test-lab.xenproject.org/osstest/results/history/test-amd64-amd64-xl-pvh-intel/linux-linus.html
http://logs.test-lab.xenproject.org/osstest/results/history/test-amd64-amd64-xl-pvh-intel/linux-next.html

Which is interesting in itself I suppose.

Bisection was broken for linux-3.18 until this morning, Ian fixed the
config and it should pick up on this failure and start investigating
once the next flight completes (tonight some time).

Ian.