Re: [PATCH][4.17?] x86: also zap secondary time area handles during soft reset

["Roger Pau Monné" <roger.pau@citrix.com>]

On Tue, Oct 25, 2022 at 05:58:10PM +0200, Jan Beulich wrote:
> On 25.10.2022 17:23, Roger Pau Monné wrote:
> > On Thu, Oct 13, 2022 at 08:48:21AM +0200, Jan Beulich wrote:
> I wasn't sure about moving arch_domain_soft_reset() as a whole, but
> yes, if that wouldn't cause other interaction issues this might be
> an option.
> 
> > In any case it's unlikely for a domain that was attempting a soft
> > reset to survive the hypervisor rejecting the operation, so it doesn't
> > matter much whether the domain is crashed by Xen or left as-is I would
> > think.
> 
> I'm sorry, I don't think I understand what you're saying here. For
> PV I'd very much expect the guest to survive; it may of course then
> be crashed or destroyed by a further tool stack operation.

I was expecting a domain that goes to the length of preparing for a
soft reset operation to either perform such soft reset, or die as a
result (and perform a non-soft reset), as recovering into the previous
state won't be feasible.  But maybe I'm wrong.

Thanks, Roger.