Описание
ELSA-2026-45192: kernel security, bug fix, and enhancement update (IMPORTANT)
[5.14.0-687.30.1]
- Disable UKI signing [Orabug: 36571828]
- Update Oracle Linux certificates (Kevin Lyons)
- Disable signing for aarch64 (Ilya Okomin)
- Oracle Linux RHCK Module Signing Key was added to the kernel trusted keys list (olkmod_signing_key.pem) [Orabug: 29539237]
- Update x509.genkey [Orabug: 24817676]
- Conflict with shim-ia32 and shim-x64 <= 15.3-1.0.5]
- Remove upstream reference during boot (Kevin Lyons) [Orabug: 34729535]
- Add Oracle Linux IMA certificates
- Add new Oracle Linux Driver Signing (key 1) certificate [Orabug: 37985764]
[5.14.0-687.30.1]
- KVM: x86/mmu: Ensure hugepage is in by slot before checking max mapping level (Aidan Wallace) [RHEL-213468] {CVE-2026-63807}
- KVM: nVMX: Hide shadow VMCS right after VMCLEAR (Aidan Wallace) [RHEL-213468]
- KVM: x86: Check for invalid/obsolete root after making MMU pages available (Aidan Wallace) [RHEL-213468]
- KVM: nVMX: Put vmcs12 pages if nested VM-Enter fails due to invalid guest state (Aidan Wallace) [RHEL-213468]
- KVM: nVMX: Add helper to put (unmap) vmcs12 pages (Paolo Bonzini) [RHEL-211277]
- KVM: x86: Don't (re)check L1 intercepts when completing userspace I/O (Aidan Wallace) [RHEL-211277] {CVE-2025-40026}
- scsi: core: Run queues for all non-SDEV_DEL devices from scsi_run_host_queues (David Jeffery) [RHEL-187412]
- drm/xe/dma-buf: fix UAF with retry loop (Anusha Srivatsa) [RHEL-192228] {CVE-2026-52950}
- drm/xe/dma-buf: handle empty bo and UAF races (Anusha Srivatsa) [RHEL-192228] {CVE-2026-52950}
- drm/xe: Fix bo leak in xe_dma_buf_init_obj() on allocation failure (Anusha Srivatsa) [RHEL-192228] {CVE-2026-52950}
- drm/xe: Fix dma-buf attachment leak in xe_gem_prime_import() (Anusha Srivatsa) [RHEL-192228] {CVE-2026-52950}
- ipv6: fix possible UAF in icmpv6_rcv() (CKI Backport Bot) [RHEL-192220] {CVE-2026-53006}
- drm/xe: Fix error cleanup in xe_exec_queue_create_ioctl() (CKI Backport Bot) [RHEL-188647] {CVE-2026-52976}
- dm log: fix out-of-bounds write due to region_count overflow (CKI Backport Bot) [RHEL-188541] {CVE-2026-53059}
Обновленные пакеты
Oracle Linux 9
Oracle Linux aarch64
kernel-cross-headers
5.14.0-687.30.1.el9_8
kernel-tools-libs-devel
5.14.0-687.30.1.el9_8
libperf
5.14.0-687.30.1.el9_8
kernel-headers
5.14.0-687.30.1.el9_8
perf
5.14.0-687.30.1.el9_8
python3-perf
5.14.0-687.30.1.el9_8
rtla
5.14.0-687.30.1.el9_8
rv
5.14.0-687.30.1.el9_8
kernel-tools
5.14.0-687.30.1.el9_8
kernel-tools-libs
5.14.0-687.30.1.el9_8
Oracle Linux x86_64
kernel-debug
5.14.0-687.30.1.el9_8
kernel-debug-core
5.14.0-687.30.1.el9_8
kernel-debug-modules
5.14.0-687.30.1.el9_8
kernel-debug-modules-core
5.14.0-687.30.1.el9_8
kernel-debug-modules-extra
5.14.0-687.30.1.el9_8
kernel-modules
5.14.0-687.30.1.el9_8
kernel-tools
5.14.0-687.30.1.el9_8
kernel-uki-virt
5.14.0-687.30.1.el9_8
kernel-debug-devel
5.14.0-687.30.1.el9_8
kernel-debug-devel-matched
5.14.0-687.30.1.el9_8
kernel-devel
5.14.0-687.30.1.el9_8
kernel-devel-matched
5.14.0-687.30.1.el9_8
kernel-doc
5.14.0-687.30.1.el9_8
kernel-headers
5.14.0-687.30.1.el9_8
perf
5.14.0-687.30.1.el9_8
python3-perf
5.14.0-687.30.1.el9_8
rtla
5.14.0-687.30.1.el9_8
rv
5.14.0-687.30.1.el9_8
kernel-cross-headers
5.14.0-687.30.1.el9_8
kernel-tools-libs-devel
5.14.0-687.30.1.el9_8
libperf
5.14.0-687.30.1.el9_8
kernel
5.14.0-687.30.1.el9_8
kernel-abi-stablelists
5.14.0-687.30.1.el9_8
kernel-core
5.14.0-687.30.1.el9_8
kernel-debug-uki-virt
5.14.0-687.30.1.el9_8
kernel-modules-core
5.14.0-687.30.1.el9_8
kernel-modules-extra
5.14.0-687.30.1.el9_8
kernel-tools-libs
5.14.0-687.30.1.el9_8
kernel-uki-virt-addons
5.14.0-687.30.1.el9_8
Ссылки на источники
Связанные уязвимости
In the Linux kernel, the following vulnerability has been resolved: KVM: x86: Don't (re)check L1 intercepts when completing userspace I/O When completing emulation of instruction that generated a userspace exit for I/O, don't recheck L1 intercepts as KVM has already finished that phase of instruction execution, i.e. has already committed to allowing L2 to perform I/O. If L1 (or host userspace) modifies the I/O permission bitmaps during the exit to userspace, KVM will treat the access as being intercepted despite already having emulated the I/O access. Pivot on EMULTYPE_NO_DECODE to detect that KVM is completing emulation. Of the three users of EMULTYPE_NO_DECODE, only complete_emulated_io() (the intended "recipient") can reach the code in question. gp_interception()'s use is mutually exclusive with is_guest_mode(), and complete_emulated_insn_gp() unconditionally pairs EMULTYPE_NO_DECODE with EMULTYPE_SKIP. The bad behavior was detected by a syzkaller program that toggles port I/O...
In the Linux kernel, the following vulnerability has been resolved: KVM: x86: Don't (re)check L1 intercepts when completing userspace I/O When completing emulation of instruction that generated a userspace exit for I/O, don't recheck L1 intercepts as KVM has already finished that phase of instruction execution, i.e. has already committed to allowing L2 to perform I/O. If L1 (or host userspace) modifies the I/O permission bitmaps during the exit to userspace, KVM will treat the access as being intercepted despite already having emulated the I/O access. Pivot on EMULTYPE_NO_DECODE to detect that KVM is completing emulation. Of the three users of EMULTYPE_NO_DECODE, only complete_emulated_io() (the intended "recipient") can reach the code in question. gp_interception()'s use is mutually exclusive with is_guest_mode(), and complete_emulated_insn_gp() unconditionally pairs EMULTYPE_NO_DECODE with EMULTYPE_SKIP. The bad behavior was detected by a syzkaller program that toggles port I/O...