Re: [PATCH v10 02/19] KVM: s390: pv: handle secure storage violations for protected guests

[Claudio Imbrenda]

On Thu, 5 May 2022 19:10:39 +0200
Thomas Huth <thuth@xxxxxxxxxx> wrote:

> On 14/04/2022 10.02, Claudio Imbrenda wrote:
> > With upcoming patches, protected guests will be able to trigger secure
> > storage violations in normal operation.
> > 
> > A secure storage violation is triggered when a protected guest tries to
> > access secure memory that has been mapped erroneously, or that belongs
> > to a different protected guest or to the ultravisor.
> > 
> > With upcoming patches, protected guests will be able to trigger secure
> > storage violations in normal operation.  
> 
> You've already used this sentence as 1st sentence of the patch description. 
> Looks weird to read it again. Maybe scratch the 1st sentence?

oops!

> 
> > This happens for example if a
> > protected guest is rebooted with lazy destroy enabled and the new guest
> > is also protected.
> > 
> > When the new protected guest touches pages that have not yet been
> > destroyed, and thus are accounted to the previous protected guest, a
> > secure storage violation is raised.
> > 
> > This patch adds handling of secure storage violations for protected
> > guests.
> > 
> > This exception is handled by first trying to destroy the page, because
> > it is expected to belong to a defunct protected guest where a destroy
> > should be possible. If that fails, a normal export of the page is
> > attempted.
>  >
> > Therefore, pages that trigger the exception will be made non-secure
> > before attempting to use them again for a different secure guest.  
> 
> I'm an complete ignorant here, but isn't this somewhat dangerous? Could it 
> happen that a VM could destroy/export the pages of another secure guest that 
> way?

this is a good question, perhaps I should add a comment explaining that
the destroy page UVC will only work on protected VMs with no CPUs.

Exporting instead is not an issue, if/when the page is needed, it will
get imported again. Unless some things went really wrong, but that can
only happen in case of a bug in the hypervisor.

> 
>   Thomas
>