From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:35056) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aaFyD-0004T9-DW for qemu-devel@nongnu.org; Mon, 29 Feb 2016 00:02:10 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1aaFyC-0000va-AI for qemu-devel@nongnu.org; Mon, 29 Feb 2016 00:02:09 -0500 Date: Mon, 29 Feb 2016 14:06:44 +1100 From: David Gibson Message-ID: <20160229030644.GC5427@voom.redhat.com> References: <1456486323-8047-1-git-send-email-david@gibson.dropbear.id.au> <1456486323-8047-11-git-send-email-david@gibson.dropbear.id.au> <56D3A204.2080806@ozlabs.ru> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="w7PDEPdKQumQfZlR" Content-Disposition: inline In-Reply-To: <56D3A204.2080806@ozlabs.ru> Subject: Re: [Qemu-devel] [PATCH 10/12] spapr_pci: (Mostly) remove spapr-pci-vfio-host-bridge List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alexey Kardashevskiy Cc: qemu-devel@nongnu.org, mdroth@linux.vnet.ibm.com, gwshan@au1.ibm.com, agraf@suse.de, alex.williamson@redhat.com, qemu-ppc@nongnu.org --w7PDEPdKQumQfZlR Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Feb 29, 2016 at 12:42:28PM +1100, Alexey Kardashevskiy wrote: > On 02/26/2016 10:32 PM, David Gibson wrote: > >Now that the regular spapr-pci-host-bridge can handle EEH, there are only > >two things that spapr-pci-vfio-host-bridge does differently: > > 1. automatically sizes its DMA window to match the host IOMMU > > 2. checks if the attached VFIO container is backed by the > > VFIO_SPAPR_TCE_IOMMU type on the host > > > >(1) is not particularly useful, since the default window used by the > >regular host bridge will work with the host IOMMU configuration on all > >current systems anyway. > > > >Plus, automatically changing guest visible configuration (such as the DMA > >window) based on host settings is generally a bad idea. >=20 > btw why exactly is it a bad idea? So, as noted, in this case it's probably ok. But in general changing configuration based on host state will break migration, because the source and destination VMs, with apparently identical setup won't actually be the same. It also makes it more difficult for management layers to know exactly what they're constructing, which is potentially an issue even for this non-migratable device. >=20 >=20 > Anyway, > Reviewed-by: Alexey Kardashevskiy >=20 >=20 --=20 David Gibson | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson --w7PDEPdKQumQfZlR Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJW07XEAAoJEGw4ysog2bOSY1YP/imSbTgKL0NUCpCcKyHj8wi6 JvSdzwqJ7EwEMdEJmJ2FIYkfkrXX/58pl1hOCz9aIMX0936XdHyafEwYPsAtjvYc EPl3w9hP7au32nKT6Nebm/JIJfeMd4vSolcv9qQDhNdP3+MB1y3MT63BEpaq83PN InKji6fj5bllnWiMxTsklr4vfp+n1Ny7zmrUG4Ks/k0PolhxfF/tZS6B+vUQS1pg HNs0BmbPeiLc0U7xENcld+m7jfvi6QUY9GcWTRctK+dhmd2dnFjON52qGYple2AF 401xk2VkmHwfQfN5zcPxc9BkcIchDhUlQ1+gedSuUZC39lVL9h0ZHHGCGerVmjyw HNkAj1pKpWpuksGObQ9Es62anW1gZei14lqi7L8WGIiZaDn+vlXP7zHOBubKD86G PQZN8Duhjty8vALxq5ZJR4M8WY9Wc0axtdcvJZRe1gdCOHgRZxp+epFJbNXUYe7h dhTJ9KDqHzS9hEyjNed9ordamrf5uNJ3OBy3BlaQmB8RH543k/EfIukZ+6QYfdQN uwnRMGTw8zKtf7up+gsIUEDDWIAgaCHuFpaihjkM91m3UZ3nfxAZdQFPaV/Kwrcs Xo8MXwa4cSxn291ERKeAPW459u2BDosB+nh5qy3BQk9y9HGXzGnjGCQs9/AKBWd0 rUsJj+9d8gfcWM0M3evP =ugb1 -----END PGP SIGNATURE----- --w7PDEPdKQumQfZlR--