Re: [Xen-devel] [xen-unstable test] 94672: regressions - trouble: blocked/broken/fail/pass

[Wei Liu <wei.liu2@xxxxxxxxxx>]

On Sun, May 22, 2016 at 04:03:47PM +0100, Andrew Cooper wrote:
> On 22/05/16 15:58, Wei Liu wrote:
> > On Sun, May 22, 2016 at 09:14:36AM +0000, osstest service owner wrote:
> >> flight 94672 xen-unstable real [real]
> >> http://logs.test-lab.xenproject.org/osstest/logs/94672/
> >>
> >> Regressions :-(
> >>
> >> Tests which did not succeed and are blocking,
> >> including tests which could not be run:
> >>  test-amd64-amd64-xl-qemuu-ovmf-amd64 17 guest-start/debianhvm.repeat fail 
> >> REGR. vs. 94580
> > This has been consistently failing in the last 5 flights.
> >
> > The most likely culprit seems to be
> >
> >  x86emul: suppress writeback upon unsuccessful MMX/SSE/AVX insn emulation
> 
> The bisector (message 
> <E1b4M5H-0007yh-V5@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> ) fingers your bump
> of the ovmf version, which precedes this MMX change.
> 
> However, it is curious as your bump of the ovmf has already passed the
> push gate.
> 
> Either the bisector is right, in which case why did your ovmf change get
> into master in the first place, or the bisector is wrong, in which case
> why didn't it identify this change as the culprit.  I am confused.
> 

Indeed. I missed the bisection email.  I only checked commits from
master to staging.

The test starts failing on merlot1 since OVMF commit gets updated. That
host is a bit special because IIRC that's the only host that would cause
the guest to have more than 4G ram. I will have a closer look tomorrow.

Wei.

> ~Andrew

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