CVE-2023-54033

In the Linux kernel, the following vulnerability has been resolved: bpf: fix a memory leak in the LRU and LRU_PERCPU hash maps The LRU and LRU_PERCPU maps allocate a new element on update before locking the target hash table bucket. Right after that the maps try to lock the bucket. If this fails, then maps return -EBUSY to the caller without releasing the allocated element. This makes the element untracked: it doesn't belong to either of free lists, and it doesn't belong to the hash table, so can't be re-used; this eventually leads to the permanent -ENOMEM on LRU map updates, which is unexpected. Fix this by returning the element to the local free list if bucket locking fails.

In the Linux kernel, the following vulnerability has been resolved:

bpf: fix a memory leak in the LRU and LRU_PERCPU hash maps

The LRU and LRU_PERCPU maps allocate a new element on update before locking the target hash table bucket. Right after that the maps try to lock the bucket. If this fails, then maps return -EBUSY to the caller without releasing the allocated element. This makes the element untracked: it doesn't belong to either of free lists, and it doesn't belong to the hash table, so can't be re-used; this eventually leads to the permanent -ENOMEM on LRU map updates, which is unexpected. Fix this by returning the element to the local free list if bucket locking fails.

In the Linux kernel, the following vulnerability has been resolved:

bpf: fix a memory leak in the LRU and LRU_PERCPU hash maps

The LRU and LRU_PERCPU maps allocate a new element on update before locking the target hash table bucket. Right after that the maps try to lock the bucket. If this fails, then maps return -EBUSY to the caller without releasing the allocated element. This makes the element untracked: it doesn't belong to either of free lists, and it doesn't belong to the hash table, so can't be re-used; this eventually leads to the permanent -ENOMEM on LRU map updates, which is unexpected. Fix this by returning the element to the local free list if bucket locking fails.

In the Linux kernel, the following vulnerability has been resolved:

bpf: fix a memory leak in the LRU and LRU_PERCPU hash maps

The LRU and LRU_PERCPU maps allocate a new element on update before locking the target hash table bucket. Right after that the maps try to lock the bucket. If this fails, then maps return -EBUSY to the caller without releasing the allocated element. This makes the element untracked: it doesn't belong to either of free lists, and it doesn't belong to the hash table, so can't be re-used; this eventually leads to the permanent -ENOMEM on LRU map updates, which is unexpected. Fix this by returning the element to the local free list if bucket locking fails.

In the Linux kernel, the following vulnerability has been resolved:

bpf: fix a memory leak in the LRU and LRU_PERCPU hash maps

The LRU and LRU_PERCPU maps allocate a new element on update before locking the target hash table bucket. Right after that the maps try to lock the bucket. If this fails, then maps return -EBUSY to the caller without releasing the allocated element. This makes the element untracked: it doesn't belong to either of free lists, and it doesn't belong to the hash table, so can't be re-used; this eventually leads to the permanent -ENOMEM on LRU map updates, which is unexpected. Fix this by returning the element to the local free list if bucket locking fails.

• wifi: ath11k: update channel list in reg notifier instead reg worker {CVE-2025-23133}
• wifi: cfg80211: call reg_notifier for self managed wiphy from driver hint {CVE-2025-23133}
• KVM: SVM: Forcibly leave SMM mode on SHUTDOWN interception {CVE-2025-37957}
• KVM: x86: forcibly leave nested mode on vCPU reset {CVE-2022-3344}
• RDMA/umem: Fix truncation for block sizes >= 4G {CVE-2026-53133}
• crypto: arm64/poly1305 - fix a read out-of-bound {CVE-2022-50231}
• tls: handle data disappearing from under the TLS ULP {CVE-2025-38616}
• Bluetooth: btmtk: validate WMT event SKB length before struct access {CVE-2026-46140}
• Bluetooth: L2CAP: complete CVE-2026-23395 hardening (chan_lock + ident clear + memset ordering)
• wireguard: device: use exit_rtnl callback instead of manual rtnl_lock in pre_exit {CVE-2026-31579}
• net: add exit_batch_rtnl() method
• rxrpc: Fix data-race warning and potential load/store tearing {CVE-2026-23118}
• virt/coco/sev-guest: Double-buffer messages {CVE-2023-53769}
• ASoC: SOF: avoid a NULL dereference with unsupported widgets {CVE-2023-54233}
• bnxt_en: Fix memory corruption when FW resources change during ifdown {CVE-2025-39810}
• scsi: qla2xxx: Free sp in error path to fix system crash {CVE-2025-71232}
• sched/deadline: Fix missing ENQUEUE_REPLENISH during PI de-boosting {CVE-2026-23371}
• net: rfkill: prevent unlimited numbers of rfkill events from being created {CVE-2026-31670}
• intel_th: fix device leak on output open() {CVE-2026-23091}
• batman-adv: reject oversized global TT response buffers {CVE-2026-31659}
• rxrpc: Fix call removal to use RCU safe deletion {CVE-2026-31642}
• i2c: s3c24xx: check the size of the SMBUS message before using it {CVE-2026-31627}
• Bluetooth: L2CAP: Fix ERTM re-init and zero pdu_len infinite loop {CVE-2026-31498}
• mtd: rawnand: serialize lock/unlock against other NAND operations {CVE-2026-23434}
• x86/efi: defer freeing of boot services memory {CVE-2026-23352}
• KVM: Don't clobber irqfd routing type when deassigning irqfd {CVE-2026-23198}
• nvmet-tcp: add bounds checks in nvmet_tcp_build_pdu_iovec {CVE-2026-23112}
• usb: typec: ucsi: Handle incorrect num_connectors capability {CVE-2025-71108}
• KVM: x86: Fix VM hard lockup after prolonged inactivity with periodic HV timer {CVE-2025-71104}
• drm/ttm: Avoid NULL pointer deref for evicted BOs {CVE-2025-71083}
• amba: bus: fix refcount leak {CVE-2023-54230}
• exfat: use kvmalloc_array/kvfree instead of kmalloc_array/kfree {CVE-2023-54194}
• octeontx2-af: Add validation for lmac type {CVE-2023-54129}
• rcu: dump vmalloc memory info safely {CVE-2023-54113}
• wifi: rtw88: use work to update rate to avoid RCU warning {CVE-2023-54071}
• bpf: fix a memory leak in the LRU and LRU_PERCPU hash maps {CVE-2023-54033}
• ASoC: soc-compress: Reposition and add pcm_mutex {CVE-2023-53866}
• netlink: do not hard code device address lenth in fdb dumps {CVE-2023-53863}
• md/raid5-cache: fix a deadlock in r5l_exit_log() {CVE-2023-53848}
• tpm: tpm_tis: Add the missed acpi_put_table() to fix memory leak {CVE-2022-50824}
• net: stmmac: fix possible memory leak in stmmac_dvr_probe() {CVE-2022-50663}
• bpf: Fix reference state management for synchronous callbacks {CVE-2022-50650}
• blk-mq: fix tags leak when shrink nr_hw_queues {CVE-2023-54227}
• usbnet: Run unregister_netdev() before unbind() again {CVE-2022-49501}
• scsi: core: Wake up the error handler when final completions race against each other {CVE-2026-23110}
• scsi: core: Move scsi_host_busy() out of host lock for waking up EH handler
• perf/core: Prevent VMA split of buffer mappings {CVE-2025-38563}
• Bluetooth: L2CAP: Fix accepting multiple L2CAP_ECRED_CONN_REQ {CVE-2026-23395}
• Bluetooth: L2CAP: Fix responding with wrong PDU type
• netfilter: nf_tables: release flowtable after rcu grace period on error {CVE-2026-23392}

In the Linux kernel, the following vulnerability has been resolved:

bpf: fix a memory leak in the LRU and LRU_PERCPU hash maps

The LRU and LRU_PERCPU maps allocate a new element on update before locking the target hash table bucket. Right after that the maps try to lock the bucket. If this fails, then maps return -EBUSY to the caller without releasing the allocated element. This makes the element untracked: it doesn't belong to either of free lists, and it doesn't belong to the hash table, so can't be re-used; this eventually leads to the permanent -ENOMEM on LRU map updates, which is unexpected. Fix this by returning the element to the local free list if bucket locking fails.

In the Linux kernel, the following vulnerability has been resolved: bpf: fix a memory leak in the LRU and LRU_PERCPU hash maps The LRU and LRU_PERCPU maps allocate a new element on update before locking the target hash table bucket. Right after that the maps try to lock the bucket. If this fails, then maps return -EBUSY to the caller without releasing the allocated element. This makes the element untracked: it doesn't belong to either of free lists, and it doesn't belong to the hash table, so can't be re-used; this eventually leads to the permanent -ENOMEM on LRU map updates, which is unexpected. Fix this by returning the element to the local free list if bucket locking fails.