Описание
ELSA-2026-39494: kernel security, bug fix, and enhancement update (IMPORTANT)
[6.12.0-211.34.1]
- Add new Oracle Linux Driver Signing (key 1) certificate [Orabug: 37985782]
- 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
- Update module name for cryptographic module [Orabug: 37400433]
- Clean git history at setup stage
[6.12.0-211.34.1]
- crypto: ccp - copy IV using skcipher ivsize (CKI Backport Bot) [RHEL-188463] {CVE-2026-53016}
- xfs: resample the data fork mapping after cycling ILOCK (Carlos Maiolino) [RHEL-193945]
- xfrm: esp: restore combined single-frag length gate (CKI Backport Bot) [RHEL-178326]
[6.12.0-211.33.1]
- futex/requeue: Revert 'Prevent NULL pointer dereference in remove_waiter() on self-deadlock'' (CKI Backport Bot) [RHEL-193250] {CVE-2026-53166}
- futex/requeue: Prevent NULL pointer dereference in remove_waiter() on self-deadlock (CKI Backport Bot) [RHEL-193250] {CVE-2026-43499}
- locking/rtmutex: Skip remove_waiter() when waiter is not enqueued (CKI Backport Bot) [RHEL-193153] {CVE-2026-43499}
- rtmutex: Use waiter::task instead of current in remove_waiter() (CKI Backport Bot) [RHEL-193153] {CVE-2026-43499}
- md/bitmap: fix GPF in write_page caused by resize race (Nigel Croxon) [RHEL-174092] {CVE-2026-43163}
[6.12.0-211.32.1]
- KVM: x86: Fix shadow paging use-after-free due to unexpected role (Paolo Bonzini) [RHEL-192407] {CVE-2026-53359}
- KVM: x86: Fix shadow paging use-after-free due to unexpected GFN (CKI Backport Bot) [RHEL-186698] {CVE-2026-46113}
- net/sched: ets: Always remove class from active list before deleting in ets_qdisc_change (CKI Backport Bot) [RHEL-183007] {CVE-2025-71066}
- sctp: revalidate list cursor after sctp_sendmsg_to_asoc() in SCTP_SENDALL (CKI Backport Bot) [RHEL-179854] {CVE-2026-46227}
[6.12.0-211.31.1]
- eventpoll: refresh epi_fget() / ep_remove_file() comments (Ian Kent) [RHEL-180776]
- eventpoll: Fix integer overflow in ep_loop_check_proc() (Ian Kent) [RHEL-180776]
- eventpoll: drop vestigial epi->dying flag (Ian Kent) [RHEL-180776]
- eventpoll: Fix semi-unbounded recursion (Ian Kent) [RHEL-180776] {CVE-2025-38614}
- eventpoll: drop dead bool return from ep_remove_epi() (Ian Kent) [RHEL-180776]
- eventpoll: refresh eventpoll_release() fast-path comment (Ian Kent) [RHEL-180776]
- eventpoll: move f_lock acquisition into ep_remove_file() (Ian Kent) [RHEL-180776]
- eventpoll: fix ep_remove struct eventpoll / struct file UAF (Ian Kent) [RHEL-180776] {CVE-2026-46242}
- eventpoll: drop vestigial __ prefix from ep_remove_{file,epi}() (Ian Kent) [RHEL-180776]
- eventpoll: move epi_fget() up (Ian Kent) [RHEL-180776]
- eventpoll: rename ep_remove_safe() back to ep_remove() (Ian Kent) [RHEL-180776]
- eventpoll: kill __ep_remove() (Ian Kent) [RHEL-180776]
- eventpoll: split __ep_remove() (Ian Kent) [RHEL-180776]
- eventpoll: use hlist_is_singular_node() in __ep_remove() (Ian Kent) [RHEL-180776]
- eventpoll: defer struct eventpoll free to RCU grace period (Ian Kent) [RHEL-173832] {CVE-2026-43074}
- epoll: annotate racy check (Ian Kent) [RHEL-180776]
- smb/client: fix out-of-bounds read in smb2_compound_op() (Paulo Alcantara) [RHEL-180048] {CVE-2026-46155}
- smb: client: fix off-by-8 bounds check in check_wsl_eas() (Paulo Alcantara) [RHEL-180048]
- tg3: Fix race for querying speed/duplex (CKI Backport Bot) [RHEL-182770]
- procfs: fix missing RCU protection when reading real_parent in do_task_stat() (CKI Backport Bot) [RHEL-181902] {CVE-2026-46259}
- net/ipv6: ioam6: prevent schema length wraparound in trace fill (Antoine Tenart) [RHEL-174786] {CVE-2026-43341}
- net/mana: Null service_wq on setup error to prevent double destroy (CKI Backport Bot) [RHEL-180277] {CVE-2026-43276}
- net: mana: Fix double destroy_workqueue on service rescan PCI path (CKI Backport Bot) [RHEL-180277] {CVE-2026-43276}
- net: page_pool: avoid false positive warning if NAPI was never added (Ivan Vecera) [RHEL-162140]
[6.12.0-211.30.1]
- ipv6: account for fraggap on the paged allocation path (Sabrina Dubroca) [RHEL-191349]
- ipv4: account for fraggap on the paged allocation path (Sabrina Dubroca) [RHEL-191349]
- fs/smb/client: fix out-of-bounds read in cifs_sanitize_prepath (CKI Backport Bot) [RHEL-189491] {CVE-2026-43112}
- Enable workaround for ARM64 ERRATUM 4118414 (Mark Salter) [RHEL-183592] {CVE-2025-10263}
- arm64: errata: Mitigate TLBI errata on Microsoft Azure Cobalt 100 CPU (Mark Salter) [RHEL-183592] {CVE-2025-10263}
- arm64: errata: Mitigate TLBI errata on NVIDIA Olympus CPU (Mark Salter) [RHEL-183592] {CVE-2025-10263}
- arm64: errata: Mitigate TLBI errata on various Arm CPUs (Mark Salter) [RHEL-183592] {CVE-2025-10263}
- arm64: cputype: Add C1-Ultra definitions (Mark Salter) [RHEL-183592] {CVE-2025-10263}
- arm64: cputype: Add C1-Premium definitions (Mark Salter) [RHEL-183592] {CVE-2025-10263}
- arm64: cputype: Add C1-Pro definitions (Mark Salter) [RHEL-183592] {CVE-2025-10263}
- crypto: krb5 - filter out async aead implementations at alloc (CKI Backport Bot) [RHEL-182537]
- crypto: krb5enc - fix async decrypt skipping hash verification (CKI Backport Bot) [RHEL-182537]
- crypto: krb5enc - fix sleepable flag handling in encrypt dispatch (CKI Backport Bot) [RHEL-182537]
- crypto: testmgr - allow authenc(hmac(sha{256,384}),cts(cbc(aes))) in FIPS mode (CKI Backport Bot) [RHEL-182537]
- ASoC: SDCA: fix NULL pointer dereference in sdca_dev_unregister_functions (CKI Backport Bot) [RHEL-185110]
- KVM: arm64: vgic-its: Drop the translation cache reference only for the erased entry (CKI Backport Bot) [RHEL-183965] {CVE-2026-46316}
- netfilter: nft_inner: Fix IPv6 inner_thoff desync (CKI Backport Bot) [RHEL-181936] {CVE-2026-46244}
- drm/gem: Fix inconsistent plane dimension calculation in drm_gem_fb_init_with_funcs() (CKI Backport Bot) [RHEL-179912] {CVE-2026-46209}
- rxrpc: Fix rxrpc_input_call_event() to only unshare DATA packets (CKI Backport Bot) [RHEL-179566] {CVE-2026-45998}
- rxrpc: Fix potential UAF after skb_unshare() failure (CKI Backport Bot) [RHEL-179566] {CVE-2026-45998}
[6.12.0-211.29.1]
- RDMA/mlx5: Fix error path fall-through in mlx5_ib_dev_res_srq_init() (CKI Backport Bot) [RHEL-180001] {CVE-2026-46176}
- crypto: tegra - Return ENOMEM when input buffer allocation fails for ccm (Vladislav Dronov) [RHEL-182759]
- crypto: tegra - Fix dma_free_coherent size error (Vladislav Dronov) [RHEL-182759]
- crypto: tegra - Don't touch bo refcount in host1x bo pin/unpin (Vladislav Dronov) [RHEL-182759]
- crypto: tegra - Add missing CRYPTO_ALG_ASYNC (Vladislav Dronov) [RHEL-182759] {CVE-2026-31739}
- crypto: tegra - Use int type to store negative error codes (Vladislav Dronov) [RHEL-182759]
- crypto: tegra - Remove the use of dev_err_probe() (Vladislav Dronov) [RHEL-182759]
- crypto: tegra - use API helpers to setup fallback request (Vladislav Dronov) [RHEL-182759]
- crypto: tegra - remove unneeded crypto_engine_stop() call (Vladislav Dronov) [RHEL-182759]
- scsi: qla2xxx: Add support to report MPI FW state (Ewan D. Milne) [RHEL-181887]
- RDMA/vmw_pvrdma: Fix double free on pvrdma_alloc_ucontext() error path (CKI Backport Bot) [RHEL-179965] {CVE-2026-46189}
- ALSA: aloop: Fix peer runtime UAF during format-change stop (CKI Backport Bot) [RHEL-179317] {CVE-2026-46090}
- ALSA: usb-audio: Add sanity check for OOB writes at silencing (CKI Backport Bot) [RHEL-173937] {CVE-2026-43279}
[6.12.0-211.28.1]
- selinux: RHEL-only hotfix for execmem regression (Ondrej Mosnacek) [RHEL-185117]
- tcp: fix potential race in tcp_v6_syn_recv_sock() (Paolo Abeni) [RHEL-174242] {CVE-2026-43198}
- scsi: qla2xxx: Completely fix fcport double free (Ewan D. Milne) [RHEL-179755] {CVE-2026-43414}
- RDMA/mana: Remove user triggerable WARN_ON() in mana_ib_create_qp_rss() (CKI Backport Bot) [RHEL-180155] {CVE-2026-46117}
- RDMA/iwcm: Fix workqueue list corruption by removing work_list (CKI Backport Bot) [RHEL-179665] {CVE-2026-45898}
- rxrpc: fix RESPONSE authenticator parser OOB read (CKI Backport Bot) [RHEL-171456] {CVE-2026-31636}
[6.12.0-211.27.1]
- nvmet-tcp: fix race between ICReq handling and queue teardown (CKI Backport Bot) [RHEL-180106] {CVE-2026-46135}
- RDMA/mana: Validate rx_hash_key_len (CKI Backport Bot) [RHEL-180090] {CVE-2026-46145}
- ipv6: icmp: clear skb2->cb[] in ip6_err_gen_icmpv6_unreach() (Guillaume Nault) [RHEL-172675] {CVE-2026-43038}
[6.12.0-211.26.1]
- net/sched: fix pedit partial COW leading to page cache corruption (Ivan Vecera) [RHEL-177380] {CVE-2026-46331}
- exit: prevent preemption of oopsing TASK_DEAD task (CKI Backport Bot) [RHEL-180009] {CVE-2026-46173}
- crypto: caam - fix overflow on long hmac keys (CKI Backport Bot) [RHEL-179769] {CVE-2026-43330}
- net: mana: fix use-after-free in add_adev() error path (CKI Backport Bot) [RHEL-172772] {CVE-2026-43056}
Обновленные пакеты
Oracle Linux 10
Oracle Linux aarch64
kernel-headers
6.12.0-211.34.1.el10_2
perf
6.12.0-211.34.1.el10_2
python3-perf
6.12.0-211.34.1.el10_2
rtla
6.12.0-211.34.1.el10_2
rv
6.12.0-211.34.1.el10_2
kernel-tools
6.12.0-211.34.1.el10_2
kernel-tools-libs
6.12.0-211.34.1.el10_2
kernel-cross-headers
6.12.0-211.34.1.el10_2
kernel-tools-libs-devel
6.12.0-211.34.1.el10_2
libperf
6.12.0-211.34.1.el10_2
Oracle Linux x86_64
kernel
6.12.0-211.34.1.el10_2
kernel-abi-stablelists
6.12.0-211.34.1.el10_2
kernel-core
6.12.0-211.34.1.el10_2
kernel-debug
6.12.0-211.34.1.el10_2
kernel-debug-core
6.12.0-211.34.1.el10_2
kernel-debug-modules-core
6.12.0-211.34.1.el10_2
kernel-debug-uki-virt
6.12.0-211.34.1.el10_2
kernel-modules
6.12.0-211.34.1.el10_2
kernel-modules-core
6.12.0-211.34.1.el10_2
kernel-modules-extra
6.12.0-211.34.1.el10_2
kernel-modules-extra-matched
6.12.0-211.34.1.el10_2
kernel-tools
6.12.0-211.34.1.el10_2
kernel-tools-libs
6.12.0-211.34.1.el10_2
kernel-uki-virt
6.12.0-211.34.1.el10_2
kernel-uki-virt-addons
6.12.0-211.34.1.el10_2
kernel-debug-devel
6.12.0-211.34.1.el10_2
kernel-debug-devel-matched
6.12.0-211.34.1.el10_2
kernel-devel
6.12.0-211.34.1.el10_2
kernel-devel-matched
6.12.0-211.34.1.el10_2
kernel-doc
6.12.0-211.34.1.el10_2
kernel-headers
6.12.0-211.34.1.el10_2
perf
6.12.0-211.34.1.el10_2
python3-perf
6.12.0-211.34.1.el10_2
rtla
6.12.0-211.34.1.el10_2
rv
6.12.0-211.34.1.el10_2
kernel-debug-modules
6.12.0-211.34.1.el10_2
kernel-debug-modules-extra
6.12.0-211.34.1.el10_2
kernel-cross-headers
6.12.0-211.34.1.el10_2
kernel-tools-libs-devel
6.12.0-211.34.1.el10_2
libperf
6.12.0-211.34.1.el10_2
Связанные CVE
Связанные уязвимости
In the Linux kernel, the following vulnerability has been resolved: xfs: resample the data fork mapping after cycling ILOCK xfs_reflink_fill_{cow_hole,delalloc} are both presented with an inode, a data fork mapping, and a cow fork mapping. Unfortunately, these two helpers cycle the ILOCK to grab a transaction, which means that the mappings are stale as soon as we reacquire the ILOCK. Currently we refresh the cow fork mapping by re-calling xfs_find_trim_cow_extent, but we don't refresh the data fork mapping beforehand, which means that the xfs_bmap_trim_cow in that function queries the refcount btree about the wrong physical blocks and returns an inaccurate value in *shared. If *shared is now false, the directio write proceeds with a stale data fork mapping. Fix this by querying the data fork mapping if the sequence counter changes across the ILOCK cycle.
In the Linux kernel, the following vulnerability has been resolved: xfs: resample the data fork mapping after cycling ILOCK xfs_reflink_fill_{cow_hole,delalloc} are both presented with an inode, a data fork mapping, and a cow fork mapping. Unfortunately, these two helpers cycle the ILOCK to grab a transaction, which means that the mappings are stale as soon as we reacquire the ILOCK. Currently we refresh the cow fork mapping by re-calling xfs_find_trim_cow_extent, but we don't refresh the data fork mapping beforehand, which means that the xfs_bmap_trim_cow in that function queries the refcount btree about the wrong physical blocks and returns an inaccurate value in *shared. If *shared is now false, the directio write proceeds with a stale data fork mapping. Fix this by querying the data fork mapping if the sequence counter changes across the ILOCK cycle.