Re: [Xen-devel] GPF Heisenbug with rumprun-xen

[Ian Jackson <Ian.Jackson@xxxxxxxxxxxxx>]

Antti Kantee writes ("Re: GPF Heisenbug with rumprun-xen"):
> On 05/02/15 15:51, Ian Jackson wrote:
> >   (Subject: Re: [Xen-devel] [rumpuserxen test] 33416: regressions - FAIL)
> 
> Aha!  I told you I don't believe in cosmic rays ;)

:-).

> The diffs for rumpuserxen and netbsdsrc between those revisions are 
> luckily small.  I couldn't spot anything in there which would 
> immediately look suspicious.  The most suspicious change is calling 
> sched_yield() as part of the bootstrap process, but that's not very 
> dramatic as far as suspicious goes.

Yes - but it could expose an existing bug.

>  TLS support was added, but I'm not sure how that would affect
> threads which do not use TLS.  That said, TLS did work right off the
> bat, so it is a bit suspicious ...

Indeed.

> Is it possible that some change in xen is tickling the bug?  That would 
> explain why attempts to reproduce the bug in other setups have failed. 
> Is it easy to fire off runs with arbitrary revisions of each repo?

It is possible that it's due to a change in Xen.  We can fire off runs
with different versions, but given the low failure probability I have
been working on adding a "do the xenstorels test many times" step to
the test run, first.

If that does what I hope, I'll be able to point osstest's automatic
bisector at it.

Ian.

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