Re: [PATCH 4.14 103/115] x86/xen: dont write ptes directly in 32-bit PV guests

[Eduardo Valentin]

On Fri, Sep 14, 2018 at 08:19:27PM +0200, Greg Kroah-Hartman wrote:
> On Fri, Sep 14, 2018 at 10:12:22AM -0700, Eduardo Valentin wrote:
> > Hello,
> > 
> > On Thu, Sep 13, 2018 at 03:32:03PM +0200, Greg Kroah-Hartman wrote:
> > > 4.14-stable review patch.  If anyone has any objections, please let me know.
> > > 
> > > ------------------
> > > 
> > > From: Juergen Gross <jgross@xxxxxxxx>
> > > 
> > > commit f7c90c2aa4004808dff777ba6ae2c7294dd06851 upstream.
> > > 
> > > In some cases 32-bit PAE PV guests still write PTEs directly instead of
> > > using hypercalls. This is especially bad when clearing a PTE as this is
> > > done via 32-bit writes which will produce intermediate L1TF attackable
> > > PTEs.
> > > 
> > > Change the code to use hypercalls instead.
> > > 
> > 
> > Good that we got this in. Should we also take this one:
> > b2d7a075a1cc ("x86/pae: use 64 bit atomic xchg function in native_ptep_get_and_clear")
> > 
> > Which is under the line of the same issue?
> 
> It's also in this patch series, the very next one in fact :)

Cool! Missed that. Sorry for the noise.

Thanks for including the patches!

> 
> thanks,
> 
> greg k-h
> 

-- 
All the best,
Eduardo Valentin