Re: [RFC PATCH 1/1] xen: Use a global mapping for runstate

[Roger Pau Monné <roger@xxxxxxx>]

On Fri, May 29, 2020 at 02:37:24PM +0100, Julien Grall wrote:
> Hi,
> 
> On 29/05/2020 14:26, Roger Pau Monné wrote:
> > TBH I would just remove the error message on Arm from the current
> > hypercall, I don't think it's useful.
> The message is part of the helpers get_page_from_gva() which is also used by
> copy_{to, from}_guest. While it may not be useful in the context of the
> runstate, it was introduced because there was some other weird bug happening
> before KPTI even existed (see [1]). I haven't yet managed to find the bottom
> line of the issue.
> 
> So I would still very much like to keep the message in place. Although we
> could reduce the number of cases where this is hapenning based on the fault.

Ack, I someone hove wrongly assumed the message was explicitly about
the runstate area, not a generic one in the helpers.

Roger.