[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Xen Security Advisory 386 v2 (CVE-2021-28702) - PCI devices with RMRRs not deassigned correctly
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Xen Security Advisory CVE-2021-28702 / XSA-386 version 2 PCI devices with RMRRs not deassigned correctly UPDATES IN VERSION 2 ==================== Updated/corrected information about vulnerable versions. Upstream Xen 4.12 is not affected. There is no harm from applying the patch to an unaffected version. ISSUE DESCRIPTION ================= Certain PCI devices in a system might be assigned Reserved Memory Regions (specified via Reserved Memory Region Reporting, "RMRR"). These are typically used for platform tasks such as legacy USB emulation. If such a device is passed through to a guest, then on guest shutdown the device is not properly deassigned. The IOMMU configuration for these devices which are not properly deassigned ends up pointing to a freed data structure, including the IO Pagetables. Subsequent DMA or interrupts from the device will have unpredictable behaviour, ranging from IOMMU faults to memory corruption. This bug has existed since at least Xen 4.4 But it was previously masked by a tangentially-related misbehaviour; that misbehaviour was corrected in f591755823a7 IOMMU/PCI: don't let domain cleanup continue when device de-assignment failed which was backported to supported stable branches. IMPACT ====== Administrators of guests which have been assigned RMRR-using PCI devices can cause denial of service and other problems, possibly including escalation of privilege. VULNERABLE SYSTEMS ================== For stable Xen releases: 4.13.4, 4.14.3 and 4.15.1 are vulnerable. Other versions of Xen released by the Xen Project are not affected. For Xen git branches: the HEADs of 4.13 and later (including xen-unstable) were vulnerable, up until 2021-10-05 (when the patch in this advisory was committed). 4.12 and earlier are not affected. In detail: code that has the following patch applied, is vulnerable: IOMMU/PCI: don't let domain cleanup continue when device de-assignment failed That patch is currently in upstream stable branches 4.13 onwards and was included in the most recent stable point releases of each Xen version. Other downstream Xen builds may be affected if that patch was backported. Only Intel x86 systems are affected. AMD x86 systems, and Arm systems, are all unaffected. Only systems using PCI passthrough are affected. (And then, only if the assigned devices have RMRRs, but whether a device advertises RMRRs is not easy to discern.) MITIGATION ========== There is no mitigation (other than not passing through PCI devices with RMRRs to guests). RESOLUTION ========== Applying the appropriate attached patch resolves this issue. Note that patches for released versions are generally prepared to apply to the stable branches, and may not apply cleanly to the most recent release tarball. Downstreams are encouraged to update to the tip of the stable branch before applying these patches. xsa386.patch xen-unstable - Xen 4.13.x $ sha256sum xsa386* f2f83c825e249bba9454437b48bbd8307fe7a224f56484388a67af124dfd279b xsa386.patch $ NOTE CONCERNING LACK OF EMBARGO =============================== This issue was reported and debugged in public before the security nature became apparent. -----BEGIN PGP SIGNATURE----- iQFABAEBCAAqFiEEI+MiLBRfRHX6gGCng/4UyVfoK9kFAmFfBvkMHHBncEB4ZW4u b3JnAAoJEIP+FMlX6CvZY20H/jWe2XVSU6R+cOv4GbhWL5sWBv4skLZ07yq77p8i JB9nJXdVkyHJPSkENzggGGiygiHMJFSD5cLvczJp1IbAlQKQlZt/oVG9oTWHHeqO joabwgZ9UyNW8/beCigRo1PYdiWI7tMsLp3D/LAjE8+ZhBRjD0NKLyWK26Uw0R8A Su5tApmlBGx0BJzQm4BUWiyog86fPoNcBkP1hRJfj1BfXRjVYB5MsaPCtMhsqBlG CFjDJ51Wn4Esxkg22e/429MbbExIAJUZoxuOWDk/D7nQShQNBTfqci4pfcaf5E+f Mxi32bIr/XY5LLgf0Opu5Sl2JP3s7Ik3IDlSa+wYoGIZWB4= =Ti35 -----END PGP SIGNATURE----- Attachment:
xsa386.patch |
Lists.xenproject.org is hosted with RackSpace, monitoring our |