# Linux Kernel 6.18.47

- Product: Linux Kernel (https://whatsnew.fyi/product/linux-kernel)
- Vendor: Linux Kernel Organization
- Date: 2026-08-27
- Version: 6.18.47
- Original notes: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.18.47
- Permalink: https://whatsnew.fyi/product/linux-kernel/releases/6.18.47
- Labels: longterm

What's New is an index, not a publisher: every entry below links to the vendor's own release notes, which are the authoritative source. Entries are labelled where they are hand-curated sample data, pre-releases, or drawn from a secondary source such as a developer blog.

Reuse: the summaries, labels and curation here are © What's New. Quote freely with attribution and a link back; wholesale republication of the corpus is not permitted — terms: https://whatsnew.fyi/terms. The vendors' own release notes remain their publishers'.

---

- **fixed** — net: gro: properly validate BIG TCP aggregation criteria
- **fixed** — ptp: vmclock: prevent read-only mappings from becoming writable
- **fixed** — futex: Avoid private hash use-after-free on final put
- **fixed** — Bluetooth: hci_aml: validate firmware segment lengths
- **fixed** — Bluetooth: MGMT: reject HCI_CMD_SYNC params_len above 255
- **fixed** — Bluetooth: ISO: zero the sockaddr before returning it in getname
- **fixed** — Bluetooth: ISO: do not force BT_LISTEN after a failed BIG sync
- **fixed** — Bluetooth: hci_sync: Fix accept list UAF during suspend
- **fixed** — Bluetooth: hci_event: validate LE Set CIG Parameters response
- **fixed** — Bluetooth: hci_event: fix LE list UAF on reset
- **fixed** — HID: hyperv: validate initial device info bounds
- **fixed** — HID: uclogic: fix use-after-free of inrange_timer on remove
- **fixed** — HID: sensor: custom: Fix use-after-free in enable_sensor
- **fixed** — HID: core: fix number/pointer type confusion on long items
- **fixed** — HID: nintendo: stop device IO before hid_hw_stop on probe failure
- **fixed** — HID: nintendo: register input device after capabilities are set
- **fixed** — HID: nintendo: fix out-of-bounds read in joycon_ctlr_read_handler()
- **fixed** — futex: Fix might_sleep() warning in futex_pivot_pending()
- **fixed** — futex: Fix race on the initial mm->futex.phash.ref allocation
- **fixed** — futex/pi: Plug private futex exec() race

94 commits in this release:

- Linux 6.18.47
- net: gro: properly validate BIG TCP aggregation criteria
- ptp: vmclock: prevent read-only mappings from becoming writable
- futex: Avoid private hash use-after-free on final put
- Bluetooth: hci_aml: validate firmware segment lengths
- Bluetooth: MGMT: reject HCI_CMD_SYNC params_len above 255
- Bluetooth: ISO: zero the sockaddr before returning it in getname
- Bluetooth: ISO: do not force BT_LISTEN after a failed BIG sync
- Bluetooth: hci_sync: Fix accept list UAF during suspend
- Bluetooth: hci_event: validate LE Set CIG Parameters response
- Bluetooth: hci_event: fix LE list UAF on reset
- HID: hyperv: validate initial device info bounds
- HID: uclogic: fix use-after-free of inrange_timer on remove
- HID: sensor: custom: Fix use-after-free in enable_sensor
- HID: core: fix number/pointer type confusion on long items
- HID: nintendo: stop device IO before hid_hw_stop on probe failure
- HID: nintendo: register input device after capabilities are set
- HID: nintendo: fix out-of-bounds read in joycon_ctlr_read_handler()
- futex: Fix might_sleep() warning in futex_pivot_pending()
- futex: Fix race on the initial mm->futex.phash.ref allocation
- futex/pi: Plug private futex exec() race
- futex: Sanitize and document task_struct::futex::state transitions
- futex/pi: Reject cross-mm private futex owners
- Input: atkbd - skip deactivate for HONOR ZQC-P
- Input: atkbd - skip deactivate for HONOR FMB-P's internal keyboard
- xfrm: fix sk_dst_cache double-free in xfrm_user_policy()
- net/ionic: avoid OOB TX partner lookup for hwstamp RXQ
- HID: pidff: fix OOB write when hid->inputs is empty
- HID: core: fix OOB read of field->usage in hid_set_field()
- HID: magicmouse: Prevent out-of-bounds (OOB) read during DOUBLE_REPORT_ID
- HID: magicmouse: do not keep a stale msc->input if no input is claimed
- HID: magicmouse: re-enable multitouch after reset-resume
- HID: magicmouse: fix battery reporting for Bluetooth Magic Trackpad USB-C
- ASoC: codecs: lpass-tx-macro: Fix enum kcontrol accesses
- mptcp: pm: fix memory leak from alloc-during-teardown race
- mptcp: pm: uniform announced addresses helpers
- mptcp: pm: rename add_entry structure to add_addr
- mptcp: pm: use for_each_subflow helper
- nvmet: pci-epf: put CQ ref on create_cq mapping failure
- nvmet: pci-epf: fix use-after-free in nvmet_pci_epf_exec_iod_work()
- nvmet-tcp: Do not WARN on remotely-controlled oversized SGL allocations
- nvmet-tcp: bound SGL data length before allocating command buffers
- nvmet-fc: fix invalid free in LS IOD error path
- nvmet-auth: zero the AUTH_RECEIVE response buffer
- dmaengine: fsl-edma: Add error handling for devm_kasprintf
- mailbox: mchp-ipc-sbi: Add null check for devm_kasprintf()
- ipv6: fix use-after-free in ip6_finish_output2()
- ipv4: reject undersized MTUs in ip_do_fragment()
- drm/xe: Fix DPT allocation paths.
- nfc: nci: free destination parameters when closing a connection
- nfc: nci: fix uninit-value in the RF discover/activated NTF handlers
- nfc: nci: fix out-of-bounds write in nci_target_auto_activated()
- nfc: nci: add data_len bound checks to activation parameter extractors
- nfc: st21nfca: validate ATR_REQ length against the received frame
- nfc: pn533: purge fragmented skbs during cleanup
- nfc: llcp: reject PDUs shorter than the LLCP header
- nfc: llcp: fix OOB read and u8 offset wrap in TLV parsers
- nfc: llcp: bound the connect_sn TLV walk to the skb
- nfc: microread: validate target discovery payload lengths
- nfc: fdp: bound the device-reported read length and fix an skb leak
- nfc: digital: clamp SENSF_RES length to the destination buffer
- libceph: fix OOB read in decode_watchers() via missing bounds check
- xfs: validate attr entry pointer before field access
- ext4: fix incorrect function call when initializing s_resgid
- ext4: don't enable DAX on new encrypted files
- ext4: propagate errors from fast commit range replay
- ext4: avoid tail write_begin walk for uptodate folios

_[Truncated at 4000 characters — full notes: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.18.47]_