IOMMU page mapping issues on x86 T[his CNA information record relates to multiple CVEs; the text explains which aspects/vulnerabilities correspond to which CVE.] Both AMD and Intel allow ACPI tables to specify regions of memory which should be left untranslated, which typically means these addresses should pass the translation phase unaltered. While these are typically device specific ACPI properties, they can also be specified to apply to a range of devices, or even all devices. On all systems with such regions Xen failed to prevent guests from undoing/replacing such mappings (CVE-2021-28694). On AMD systems, where a discontinuous range is specified by firmware, the supposedly-excluded middle range will also be identity-mapped (CVE-2021-28695). Further, on AMD systems, upon de-assigment of a physical device from a guest, the identity mappings would be left in place, allowing a guest continued access to ranges of memory which it shouldn't have access to anymore (CVE-2021-28696).

grant table v2 status pages may remain accessible after de-allocation Guest get permitted access to certain Xen-owned pages of memory. The majority of such pages remain allocated / associated with a guest for its entire lifetime. Grant table v2 status pages, however, get de-allocated when a guest switched (back) from v2 to v1. The freeing of such pages requires that the hypervisor know where in the guest these pages were mapped. The hypervisor tracks only one use within guest space, but racing requests from the guest to insert mappings of these pages may result in any of them to become mapped in multiple locations. Upon switching back from v2 to v1, the guest would then retain access to a page that was freed and perhaps re-used for other purposes.

PCI devices with RMRRs not deassigned correctly 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.

An issue was discovered in the Linux kernel through 5.11.3. Certain iSCSI data structures do not have appropriate length constraints or checks, and can exceed the PAGE_SIZE value. An unprivileged user can send a Netlink message that is associated with iSCSI, and has a length up to the maximum length of a Netlink message.

An issue was discovered in the Linux kernel 3.11 through 5.10.16, as used by Xen. To service requests to the PV backend, the driver maps grant references provided by the frontend. In this process, errors may be encountered. In one case, an error encountered earlier might be discarded by later processing, resulting in the caller assuming successful mapping, and hence subsequent operations trying to access space that wasn't mapped. In another case, internal state would be insufficiently updated, preventing safe recovery from the error. This affects drivers/block/xen-blkback/blkback.c.

An issue was discovered in the Linux kernel 4.18 through 5.10.16, as used by Xen. The backend allocation (aka be-alloc) mode of the drm_xen_front drivers was not meant to be a supported configuration, but this wasn't stated accordingly in its support status entry.

avahi-daemon-check-dns.sh in the Debian avahi package through 0.8-4 is executed as root via /etc/network/if-up.d/avahi-daemon, and allows a local attacker to cause a denial of service or create arbitrary empty files via a symlink attack on files under /run/avahi-daemon. NOTE: this only affects the packaging for Debian GNU/Linux (used indirectly by SUSE), not the upstream Avahi product.

Use After Free vulnerability in nfc sockets in the Linux Kernel before 5.12.4 allows local attackers to elevate their privileges. In typical configurations, the issue can only be triggered by a privileged local user with the CAP_NET_RAW capability.

The btrfs_xattr_set_acl function in fs/btrfs/acl.c in btrfs in the Linux kernel 2.6.34 and earlier does not check file ownership before setting an ACL, which allows local users to bypass file permissions by setting arbitrary ACLs, as demonstrated using setfacl.

The do_gfs2_set_flags function in fs/gfs2/file.c in the Linux kernel before 2.6.34-git10 does not verify the ownership of a file, which allows local users to bypass intended access restrictions via a SETFLAGS ioctl request.

drbd8 allows local users to bypass intended restrictions for certain actions via netlink packets, similar to CVE-2009-3725.

rssh version 2.3.4 contains a CWE-77: Improper Neutralization of Special Elements used in a Command ('Command Injection') vulnerability in allowscp permission that can result in Local command execution. This attack appear to be exploitable via An authorized SSH user with the allowscp permission.

Heap-based buffer overflow in the pcnet_receive function in hw/net/pcnet.c in QEMU allows guest OS administrators to cause a denial of service (instance crash) or possibly execute arbitrary code via a series of packets in loopback mode.

Insufficient validation of environment variables in the telnet client supplied in Junos OS can lead to stack-based buffer overflows, which can be exploited to bypass veriexec restrictions on Junos OS. A stack-based overflow is present in the handling of environment variables when connecting via the telnet client to remote telnet servers. This issue only affects the telnet client — accessible from the CLI or shell — in Junos OS. Inbound telnet services are not affected by this issue. This issue affects: Juniper Networks Junos OS: 12.3 versions prior to 12.3R12-S13; 12.3X48 versions prior to 12.3X48-D80; 14.1X53 versions prior to 14.1X53-D130, 14.1X53-D49; 15.1 versions prior to 15.1F6-S12, 15.1R7-S4; 15.1X49 versions prior to 15.1X49-D170; 15.1X53 versions prior to 15.1X53-D237, 15.1X53-D496, 15.1X53-D591, 15.1X53-D69; 16.1 versions prior to 16.1R3-S11, 16.1R7-S4; 16.2 versions prior to 16.2R2-S9; 17.1 versions prior to 17.1R3; 17.2 versions prior to 17.2R1-S8, 17.2R2-S7, 17.2R3-S1; 17.3 versions prior to 17.3R3-S4; 17.4 versions prior to 17.4R1-S6, 17.4R2-S3, 17.4R3; 18.1 versions prior to 18.1R2-S4, 18.1R3-S3; 18.2 versions prior to 18.2R1-S5, 18.2R2-S2, 18.2R3; 18.2X75 versions prior to 18.2X75-D40; 18.3 versions prior to 18.3R1-S3, 18.3R2; 18.4 versions prior to 18.4R1-S2, 18.4R2.

The load_multiboot function in hw/i386/multiboot.c in Quick Emulator (aka QEMU) allows local guest OS users to execute arbitrary code on the QEMU host via a mh_load_end_addr value greater than mh_bss_end_addr, which triggers an out-of-bounds read or write memory access.

The Linux kernel 4.15 has a Buffer Overflow via an SNDRV_SEQ_IOCTL_SET_CLIENT_POOL ioctl write operation to /dev/snd/seq by a local user.

The futex_requeue function in kernel/futex.c in the Linux kernel before 4.14.15 might allow attackers to cause a denial of service (integer overflow) or possibly have unspecified other impact by triggering a negative wake or requeue value.

util/virlog.c in libvirt does not properly determine the hostname on LXC container startup, which allows local guest OS users to bypass an intended container protection mechanism and execute arbitrary commands via a crafted NSS module.

In the function wmi_set_ie(), the length validation code does not handle unsigned integer overflow properly. As a result, a large value of the 'ie_len' argument can cause a buffer overflow in all Android releases from CAF (Android for MSM, Firefox OS for MSM, QRD Android) using the Linux Kernel.

Go before 1.8.7, Go 1.9.x before 1.9.4, and Go 1.10 pre-releases before Go 1.10rc2 allow "go get" remote command execution during source code build, by leveraging the gcc or clang plugin feature, because -fplugin= and -plugin= arguments were not blocked.

NetApp OnCommand Unified Manager for Linux versions 7.2 though 7.3 ship with the Java Debug Wire Protocol (JDWP) enabled which allows unauthorized local attackers to execute arbitrary code.

The svpn component of the F5 BIG-IP APM client prior to version 7.1.7 for Linux and Mac OS X runs as a privileged process and can allow an unprivileged user to assume super-user privileges on the local client host. A malicious local unprivileged user may gain knowledge of sensitive information, manipulate certain data, or disrupt service.

In the Linux kernel through 4.14.13, drivers/block/loop.c mishandles lo_release serialization, which allows attackers to cause a denial of service (__lock_acquire use-after-free) or possibly have unspecified other impact.

In macOS High Sierra before 10.13.5, an issue existed in CUPS. This issue was addressed with improved access restrictions.

Use-after-free vulnerability in the path_openat function in fs/namei.c in the Linux kernel 3.x and 4.x before 4.0.4 allows local users to cause a denial of service or possibly have unspecified other impact via O_TMPFILE filesystem operations that leverage a duplicate cleanup operation.

Integer overflow in the sg_start_req function in drivers/scsi/sg.c in the Linux kernel 2.6.x through 4.x before 4.1 allows local users to cause a denial of service or possibly have unspecified other impact via a large iov_count value in a write request.

CA Common Services, as used in CA Client Automation r12.5 SP01, r12.8, and r12.9; CA Network and Systems Management r11.0, r11.1, and r11.2; CA NSM Job Management Option r11.0, r11.1, and r11.2; CA Universal Job Management Agent; CA Virtual Assurance for Infrastructure Managers (aka SystemEDGE) 12.6, 12.7, 12.8, and 12.9; and CA Workload Automation AE r11, r11.3, r11.3.5, and r11.3.6 on UNIX, allows local users to gain privileges via an unspecified environment variable.

CA Common Services, as used in CA Client Automation r12.5 SP01, r12.8, and r12.9; CA Network and Systems Management r11.0, r11.1, and r11.2; CA NSM Job Management Option r11.0, r11.1, and r11.2; CA Universal Job Management Agent; CA Virtual Assurance for Infrastructure Managers (aka SystemEDGE) 12.6, 12.7, 12.8, and 12.9; and CA Workload Automation AE r11, r11.3, r11.3.5, and r11.3.6 on UNIX, does not properly validate an unspecified variable, which allows local users to gain privileges via unknown vectors.

net/llc/sysctl_net_llc.c in the Linux kernel before 3.19 uses an incorrect data type in a sysctl table, which allows local users to obtain potentially sensitive information from kernel memory or possibly have unspecified other impact by accessing a sysctl entry.

Heap-based buffer overflow in openfs.c in the libext2fs library in e2fsprogs before 1.42.12 allows local users to execute arbitrary code via crafted block group descriptor data in a filesystem image.

The Linux kernel through 3.17.4 does not properly restrict dropping of supplemental group memberships in certain namespace scenarios, which allows local users to bypass intended file permissions by leveraging a POSIX ACL containing an entry for the group category that is more restrictive than the entry for the other category, aka a "negative groups" issue, related to kernel/groups.c, kernel/uid16.c, and kernel/user_namespace.c.

The kvm_iommu_map_pages function in virt/kvm/iommu.c in the Linux kernel through 3.17.2 miscalculates the number of pages during the handling of a mapping failure, which allows guest OS users to cause a denial of service (host OS page unpinning) or possibly have unspecified other impact by leveraging guest OS privileges. NOTE: this vulnerability exists because of an incorrect fix for CVE-2014-3601.

An issue was discovered in the Linux kernel before 4.18.7. In block/blk-core.c, there is an __blk_drain_queue() use-after-free because a certain error case is mishandled.

An issue was discovered in fs/xfs/xfs_super.c in the Linux kernel before 4.18. A use after free exists, related to xfs_fs_fill_super failure.

The BPF subsystem in the Linux kernel before 4.17 mishandles situations with a long jump over an instruction sequence where inner instructions require substantial expansions into multiple BPF instructions, leading to an overflow. This affects kernel/bpf/core.c and net/core/filter.c.

A crafted NTFS image can cause a heap-based buffer overflow in ntfs_check_log_client_array in NTFS-3G through 2021.8.22.

An issue was discovered in Xen through 4.14.x. There are missing memory barriers when accessing/allocating an event channel. Event channels control structures can be accessed lockless as long as the port is considered to be valid. Such a sequence is missing an appropriate memory barrier (e.g., smp_*mb()) to prevent both the compiler and CPU from re-ordering access. A malicious guest may be able to cause a hypervisor crash resulting in a Denial of Service (DoS). Information leak and privilege escalation cannot be excluded. Systems running all versions of Xen are affected. Whether a system is vulnerable will depend on the CPU and compiler used to build Xen. For all systems, the presence and the scope of the vulnerability depend on the precise re-ordering performed by the compiler used to build Xen. We have not been able to survey compilers; consequently we cannot say which compiler(s) might produce vulnerable code (with which code generation options). GCC documentation clearly suggests that re-ordering is possible. Arm systems will also be vulnerable if the CPU is able to re-order memory access. Please consult your CPU vendor. x86 systems are only vulnerable if a compiler performs re-ordering.

IBM Informix Dynamic Server 14.10 is vulnerable to a stack based buffer overflow, caused by improper bounds checking. A local privileged user could overflow a buffer and execute arbitrary code on the system or cause a denial of service condition. IBM X-Force ID: 198366.