From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([209.51.188.92]:48426) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gyCXp-0007Wj-Us for qemu-devel@nongnu.org; Mon, 25 Feb 2019 04:27:30 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gyCXo-0006mF-0Y for qemu-devel@nongnu.org; Mon, 25 Feb 2019 04:27:29 -0500 Received: from 4.mo179.mail-out.ovh.net ([46.105.36.149]:44793) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1gyCXj-0006Tq-TB for qemu-devel@nongnu.org; Mon, 25 Feb 2019 04:27:25 -0500 Received: from player737.ha.ovh.net (unknown [10.109.143.146]) by mo179.mail-out.ovh.net (Postfix) with ESMTP id 0270811C5EC for ; Mon, 25 Feb 2019 10:26:58 +0100 (CET) Date: Mon, 25 Feb 2019 10:26:51 +0100 From: Greg Kurz Message-ID: <20190225102651.762e2520@bahia.lan> In-Reply-To: <20190224233711.GD7668@umbus.fritz.box> References: <20190214043916.22128-1-david@gibson.dropbear.id.au> <20190214043916.22128-3-david@gibson.dropbear.id.au> <20190222100822.73515a4d@bahia.lan> <20190224233711.GD7668@umbus.fritz.box> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha256; boundary="Sig_/8=QYiygxEyObaqsx9qOLpor"; protocol="application/pgp-signature" Subject: Re: [Qemu-devel] [Qemu-ppc] [PATCH 2/5] virtio-balloon: Corrections to address verification List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: David Gibson Cc: mst@redhat.com, qemu-devel@nongnu.org, David Hildenbrand , qemu-ppc@nongnu.org --Sig_/8=QYiygxEyObaqsx9qOLpor Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable On Mon, 25 Feb 2019 10:37:11 +1100 David Gibson wrote: > On Fri, Feb 22, 2019 at 10:08:22AM +0100, Greg Kurz wrote: > > On Thu, 14 Feb 2019 15:39:13 +1100 > > David Gibson wrote: > > =20 > > > The virtio-balloon device's verification of the address given to it b= y the > > > guest has a number of faults: > > > * The addresses here are guest physical addresses, which should be > > > 'hwaddr' rather than 'ram_addr_t' (the distinction is admittedly > > > pretty subtle and confusing) > > > * We don't check for section.mr being NULL, which is the main way= that > > > memory_region_find() reports basic failures. We really need to= check > > > that before looking at any other section fields, because > > > memory_region_find() doesn't initialize them on the failure path > > > * We're passing a length of '1' to memory_region_find(), but real= ly the > > > guest is requesting that we put the entire page into the balloo= n, > > > so it makes more sense to call it with BALLOON_PAGE_SIZE > > >=20 > > > Signed-off-by: David Gibson > > > Reviewed-by: David Hildenbrand > > > Reviewed-by: Michael S. Tsirkin > > > --- > > > hw/virtio/virtio-balloon.c | 17 ++++++++++------- > > > 1 file changed, 10 insertions(+), 7 deletions(-) > > >=20 > > > diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c > > > index 43af521884..eb357824d8 100644 > > > --- a/hw/virtio/virtio-balloon.c > > > +++ b/hw/virtio/virtio-balloon.c > > > @@ -221,17 +221,20 @@ static void virtio_balloon_handle_output(VirtIO= Device *vdev, VirtQueue *vq) > > > } > > > =20 > > > while (iov_to_buf(elem->out_sg, elem->out_num, offset, &pfn,= 4) =3D=3D 4) { > > > - ram_addr_t pa; > > > - ram_addr_t addr; > > > + hwaddr pa; > > > + hwaddr addr; > > > int p =3D virtio_ldl_p(vdev, &pfn); > > > =20 > > > - pa =3D (ram_addr_t) p << VIRTIO_BALLOON_PFN_SHIFT; > > > + pa =3D (hwaddr) p << VIRTIO_BALLOON_PFN_SHIFT; > > > offset +=3D 4; > > > =20 > > > - /* FIXME: remove get_system_memory(), but how? */ > > > - section =3D memory_region_find(get_system_memory(), pa, = 1); > > > - if (!int128_nz(section.size) || > > > - !memory_region_is_ram(section.mr) || > > > + section =3D memory_region_find(get_system_memory(), pa, > > > + BALLOON_PAGE_SIZE); > > > + if (!section.mr) { > > > + trace_virtio_balloon_bad_addr(pa); > > > + continue; > > > + } =20 > >=20 > > memory_region_unref(section.mr) with section.mr =3D=3D NULL is safe and > > resolves to a nop. Not sure you need a separate if to handle this > > case. =20 >=20 > memory_region_is_ram() and friends are not, however - they will > dereference their argument unconditionally. >=20 Indeed but the two ifs can be merged anyway: if (!section.mr || !memory_region_is_ram(section.mr) || memory_region_is_rom(section.mr) || memory_region_is_romd(section.mr)) { trace_virtio_balloon_bad_addr(pa); memory_region_unref(section.mr); continue; } > >=20 > > Apart from that, > >=20 > > Reviewed-by: Greg Kurz > > =20 > > > + if (!memory_region_is_ram(section.mr) || > > > memory_region_is_rom(section.mr) || > > > memory_region_is_romd(section.mr)) { > > > trace_virtio_balloon_bad_addr(pa); =20 > > =20 >=20 --Sig_/8=QYiygxEyObaqsx9qOLpor Content-Type: application/pgp-signature Content-Description: OpenPGP digital signature -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEtIKLr5QxQM7yo0kQcdTV5YIvc9YFAlxztNsACgkQcdTV5YIv c9bi0Q/+JObWXZqwTsp9L8YEl7a/3CoNRs25T3qQZZasrk0AFmwQdX+u8Pj+qME5 VxiJH4OK8kI6zGnOrHStkBo5Y6UFEy3ayXWiSpMcSuLyOqrfr2crFAZak5aa/Ewv fSmP6115SgWzf72Y2m1LNGGVCYG9MfXYwN1E3dHWtHjD4NDq44KeU2hC+knm0hwb wzPuMvQWkMELLLwva/7T7RKGgGmU0YM1Mtyz02fDPZbyRZM60PKhdaAFfAeRyriC 4oeaWOjYkvOPOJGi8o9ia6VGHdMNMwAurJVkH1uATUCWxj8UH2YTkqYksc4QkI3O 6X5Y2tmxDHQxXyXmJU2VE4ikv6IhDFuxlq3zhLSyILdLoUSEV8VF+HjRaYGWsf2I y1DM442qGaqnCuq80oF/W1dlbFAMh7r6aeGGxgrBRjkTBoh6EMXJNm8xUinFoalz 7wARvKTM4MFnLt0Unw3EH9XagnTNHY9h+YKcIfpvkbYZMi40VeXAm/VJi4+ijrc/ ydOaV1n6eYE/eZd5scru15sHUVdiqx89T7qujgkE1GzYNl1EIM0aEOXEDuHO3HHF xXcOhI04q7ML2LYtFm02AsTXqkaWhHD28TTp81leRCJR+ex3tSfzhChoZJGBk0OL sflIfMvD6mVIKsua6wqCNrrwPCHsopGHGRECSmLXQb8N6EIBY9A= =cEb+ -----END PGP SIGNATURE----- --Sig_/8=QYiygxEyObaqsx9qOLpor--