From: "Jan Beulich" Subject: [PATCH][XSA-59] VT-d: suppress UR signaling for further desktop chipsets Date: Tue, 16 Sep 2014 08:18:36 +0100 Message-ID: <5418006C02000078000353DA@mail.emea.novell.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="=__Part93A6DD5C.1__=" Return-path: Received: from mail6.bemta14.messagelabs.com ([193.109.254.103]) by lists.xen.org with esmtp (Exim 4.72) (envelope-from ) id 1XTn27-0006lU-Uq for xen-devel@lists.xenproject.org; Tue, 16 Sep 2014 07:18:40 +0000 List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: xen-devel Cc: Yang Z Zhang , Kevin Tian List-Id: xen-devel@lists.xenproject.org This is a MIME message. If you are reading this text, you may want to consider changing to a mail reader or gateway that understands how to properly handle MIME multipart messages. --=__Part93A6DD5C.1__= Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Content-Disposition: inline This extends commit d6cb14b34f ("VT-d: suppress UR signaling for desktop chipsets") as per the finally obtained list of affected chipsets from Intel. Also pad the IDs we had listed there before to full 4 hex digits. This is CVE-2013-3495 / XSA-59. Signed-off-by: Jan Beulich --- I already raised the obvious questions connected to the appearance of Broadwell here: Are its server chipsets affected then too, i.e. does the server side set of IDs need extension as well? Is the issue going to be fixed in hardware at some point, or will we need to indefinitely extend the set of IDs here? --- a/xen/drivers/passthrough/vtd/quirks.c +++ b/xen/drivers/passthrough/vtd/quirks.c @@ -466,10 +466,12 @@ void pci_vtd_quirk(const struct pci_dev=20 action, seg, bus, dev, func); break; =20 - case 0x100: case 0x104: case 0x108: /* Sandybridge */ - case 0x150: case 0x154: case 0x158: /* Ivybridge */ - case 0xa04: /* Haswell ULT */ - case 0xc00: case 0xc04: case 0xc08: /* Haswell */ + case 0x0040: case 0x0044: case 0x0048: /* Nehalem/Westmere */ + case 0x0100: case 0x0104: case 0x0108: /* Sandybridge */ + case 0x0150: case 0x0154: case 0x0158: /* Ivybridge */ + case 0x0a04: /* Haswell ULT */ + case 0x0c00: case 0x0c04: case 0x0c08: /* Haswell */ + case 0x1600: case 0x1604: case 0x1608: /* Broadwell */ bar =3D pci_conf_read32(seg, bus, dev, func, 0x6c); bar =3D (bar << 32) | pci_conf_read32(seg, bus, dev, func, 0x68); pa =3D bar & 0x7ffffff000UL; /* bits 12...38 */ --=__Part93A6DD5C.1__= Content-Type: text/plain; name="VT-d-mask-UR-more-desktop-chipsets.patch" Content-Transfer-Encoding: quoted-printable Content-Disposition: attachment; filename="VT-d-mask-UR-more-desktop-chipsets.patch" VT-d: suppress UR signaling for further desktop chipsets=0A=0AThis extends = commit d6cb14b34f ("VT-d: suppress UR signaling for=0Adesktop chipsets") = as per the finally obtained list of affected=0Achipsets from Intel.=0A=0AAl= so pad the IDs we had listed there before to full 4 hex digits.=0A=0AThis = is CVE-2013-3495 / XSA-59.=0A=0ASigned-off-by: Jan Beulich =0A---=0AI already raised the obvious questions connected to the = appearance of=0ABroadwell here: Are its server chipsets affected then too, = i.e. does=0Athe server side set of IDs need extension as well? Is the = issue going=0Ato be fixed in hardware at some point, or will we need to = indefinitely=0Aextend the set of IDs here?=0A=0A--- a/xen/drivers/passthrou= gh/vtd/quirks.c=0A+++ b/xen/drivers/passthrough/vtd/quirks.c=0A@@ -466,10 = +466,12 @@ void pci_vtd_quirk(const struct pci_dev =0A = action, seg, bus, dev, func);=0A break;=0A =0A- case 0x100: = case 0x104: case 0x108: /* Sandybridge */=0A- case 0x150: case 0x154: = case 0x158: /* Ivybridge */=0A- case 0xa04: /* Haswell ULT */=0A- = case 0xc00: case 0xc04: case 0xc08: /* Haswell */=0A+ case 0x0040: case = 0x0044: case 0x0048: /* Nehalem/Westmere */=0A+ case 0x0100: case = 0x0104: case 0x0108: /* Sandybridge */=0A+ case 0x0150: case 0x0154: = case 0x0158: /* Ivybridge */=0A+ case 0x0a04: /* Haswell ULT */=0A+ = case 0x0c00: case 0x0c04: case 0x0c08: /* Haswell */=0A+ case 0x1600: = case 0x1604: case 0x1608: /* Broadwell */=0A bar =3D pci_conf_read3= 2(seg, bus, dev, func, 0x6c);=0A bar =3D (bar << 32) | pci_conf_rea= d32(seg, bus, dev, func, 0x68);=0A pa =3D bar & 0x7ffffff000UL; /* = bits 12...38 */=0A --=__Part93A6DD5C.1__= Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel --=__Part93A6DD5C.1__=--