Re: [Xen-devel] [PATCH v2 07/10] xen/arm: Release maintenance interrupt when CPU is hot-unplugged

[Julien Grall <julien.grall@xxxxxxx>]

On 09/05/18 11:10, Mirela Simonovic wrote:
> On Fri, Apr 27, 2018 at 5:12 PM, Julien Grall <julien.grall@xxxxxxx> wrote:
> > On 27/04/18 15:38, Mirela Simonovic wrote:
> > > On Fri, Apr 27, 2018 at 4:15 PM, Tim Deegan <tim@xxxxxxx> wrote:
> > > > At 10:28 +0100 on 27 Apr (1524824906), Julien Grall wrote:
> > > > > On 26/04/18 15:23, Tim Deegan wrote:
> > > > > > At 11:08 +0100 on 26 Apr (1524740921), Julien Grall wrote:
> > > > > > > > > > > On 20/04/18 13:25, Mirela Simonovic wrote:
> > While I guess this code makes no harm, it does not do what is expected (i.e
> > draining the interrupt). I can't see any reason to keep wrong code, we
> > should really aim to have code that match the architecture. And better to
> > fix it when we discover the problem rather than waiting until we
> > rediscovered it later.
> >
> > So at least a patch to update the code/comment should be done.
> I don't feel comfortable removing these 3 lines because I have no way
> to test and guarantee that the change will not introduce any issues.
The work you are doing (suspend/resume, hotplug) is not easy to test and 
very subtle to get it right. Testing can only uncover obvious bug on 
your platform. IMHO, we can only rely on the specifications (ARM ARM, 
PSCI...) and extensive review of the series (and the code around).
> However, if despite all you really want me to remove these lines
> within this series I don't have a problem doing that in a separate
> patch. Please just confirm the plan.
I am quite confident that this code should not be there or at least not 
in its current form.
Cheers,

--
Julien Grall

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel