linux-kernel - RE: [RFC v2 1/2] powerpc/pseries/iommu: Share the per-cpu TCE page with the hypervisor.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-Id: <20191112011530.GC5159@oc0525413822.ibm.com>
Date:   Mon, 11 Nov 2019 17:15:30 -0800
From:   Ram Pai <linuxram@...ibm.com>
To:     David Gibson <david@...son.dropbear.id.au>
Cc:     linuxppc-dev@...ts.ozlabs.org, benh@...nel.crashing.org,
        mpe@...erman.id.au, paulus@...abs.org, mdroth@...ux.vnet.ibm.com,
        hch@....de, andmike@...ibm.com, sukadev@...ux.vnet.ibm.com,
        mst@...hat.com, ram.n.pai@...il.com, aik@...abs.ru, cai@....pw,
        tglx@...utronix.de, bauerman@...ux.ibm.com,
        linux-kernel@...r.kernel.org
Subject: RE: [RFC v2 1/2] powerpc/pseries/iommu: Share the per-cpu TCE page with the
 hypervisor.

On Sun, Nov 10, 2019 at 07:40:06PM +0000, David Gibson wrote:
> On Fri, Nov 08, 2019 at 03:00:10PM -0800, Ram Pai wrote:
> > The hypervisor needs to access the contents of the page holding the TCE
> > entries while setting up the TCE entries in the IOMMU's TCE table.
> > 
> > For SecureVMs, since this page is encrypted, the hypervisor cannot
> > access valid entries. Share the page with the hypervisor. This ensures
> > that the hypervisor sees those valid entries.
> > 
> > Why is this safe?
> >    The page contains only TCE entries; not any sensitive data
> >    belonging to the Secure VM. The hypervisor has a genuine need to know
> >    the value of the TCE entries, without which it will not be able to
> >    DMA to/from the pages pointed to by the TCE entries. In a Secure
> >    VM the TCE entries point to pages that are also shared with the
> >    hypervisor; example: pages containing bounce buffers.
> 
> The bit that may not be obvious to reviewers from the above is this:
> 
> This is *not* a page of "live" TCEs which are actively used for
> translation.  Instead this is just a transient buffer with a batch of
> TCEs to set, passed to the hypervisor with the H_PUT_TCE_INDIRECT call.


That is true.  I should have stated that explicitly. Thanks. will
update the commit log.


RP