Re: [Xen-devel] [PATCH for-4.12] arm/p2m: do not invalidate p2m root if it is shared with IOMMU

[Julien Grall <julien.grall@xxxxxxx>]

Hi Andrii,

On 29/01/2019 14:33, Andrii Anisov wrote:
> On 29.01.19 15:53, Julien Grall wrote:
> > Using Set/Way is fundamentally broken partly because it does not deal with 
> > system caches.
> I already know it. I've looked at KVM presentation from 2015, your last year 
> video from China, reading through ARM ARM.
>
> > It also only applies to a given cache and the behavior will not be replicated 
> > to other CPUs.
> In our case cpu cache flushing by Set/Way it is propagated to all cpus with 
> `on_each_cpu()`. (Here should be a facepalm emoji)
:/

> > Running such software is a call for nasty behavior in your guest.
> I do understand that.
> But we have to handle that. And it is a question to us how to do that.
No we don't have to. They have been lucky to see this working even on baremetal.
Set/Way operations have been dropped from Linux for a long time, so I really 
can't see why a proprietary driver is still using them.
I think the policy for Set/Way operations is correct in Xen. This works to avoid 
breaking basic case but the most complex one are going to break.
You solution is only delaying the real fix (i.e removing Set/Way operation from 
the software). So here the best solution is to go to the vendor and ask them to 
fix their software.
Cheers,

--
Julien Grall

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