Re: [Xen-devel] [xen-4.6-testing bisection] complete test-xtf-amd64-amd64-5

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 04.04.17 at 19:23, <dario.faggioli@xxxxxxxxxx> wrote:
> On Mon, 2017-04-03 at 02:02 -0600, Jan Beulich wrote:
>> > > > On 03.04.17 at 00:44, <osstest-admin@xxxxxxxxxxxxxx> wrote:
>> >   commit 7ff6d9fc19ef88d2ca3304a312c0e8a46b61f546
>> >   Author: Dario Faggioli <dario.faggioli@xxxxxxxxxx>
>> >   Date:   Fri Mar 31 09:03:32 2017 +0200
>> >   
>> >       xen: sched: don't call hooks of the wrong scheduler via
>> > VCPU2OP
>> 
>> I must have completely screwed up on this backport, so I've simply
>> reverted it for now. I didn't look closely yet whether this is some
>> trivial oversight of mine, or whether perhaps it shouldn't have been
>> backported that far back in the first place.
>> 
> I had a look. The backport looks fine in itself to me.
> 
> If I'm not mistaken, this is not in staging-4.6:
> 
>  f3d47501db2b7bb8dfd6a3c9710b7aff4b1fc55b
>  xen: fix a (latent) cpupool-related race during domain destroy
> 
> I can't be positive this is what's missing, but it's my best guess. :-)
> 
> I also haven't tried backporting it, but I'm happy to. Let me know if I
> should.

I think we could leave 4.6 alone in this regard, unless you see a
strong need for the fix there, or unless anyone's aware of
someone having actively encountered the problem there. Or in
other words - if you feel like making sure we know of all prereqs
(and providing backports if those won't apply as is), I'd happily
apply them, but I don't think you absolutely need to.

Jan


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