CVE-2026-68284

In the Linux kernel, the following vulnerability has been resolved: bpf, sockmap: Fix cork use-after-free in tcpbpfsendmsg tcpbpfsendmsg keeps msgtx across skstreamwaitmemory, which drops and reacquires the socket lock. Its error path tries to decide whether msgtx names the local temporary messag...

EUVD-2026-55385

In the Linux kernel, the following vulnerability has been resolved: bpf, sockmap: Fix cork use-after-free in tcpbpfsendmsg tcpbpfsendmsg keeps msgtx across skstreamwaitmemory, which drops and reacquires the socket lock. Its error path tries to decide whether msgtx names the local temporary messag...

CVE-2026-68284

CVE-2026-68284 affects the Linux kernel BPF/sockmap path. The bug is a use-after-free in tcp_bpf_sendmsg() when two threads send on the same socket: msg_tx is compared to psock->cork while cork may have been freed, leading to freeing a stale cork. A fix ensures tmp cork is checked directly, pr...

CVE-2026-68284 bpf, sockmap: Fix cork use-after-free in tcp_bpf_sendmsg()

In the Linux kernel, the following vulnerability has been resolved: bpf, sockmap: Fix cork use-after-free in tcpbpfsendmsg tcpbpfsendmsg keeps msgtx across skstreamwaitmemory, which drops and reacquires the socket lock. Its error path tries to decide whether msgtx names the local temporary messag...

CVE-2026-68284 bpf, sockmap: Fix cork use-after-free in tcp_bpf_sendmsg()

In the Linux kernel, the following vulnerability has been resolved: bpf, sockmap: Fix cork use-after-free in tcpbpfsendmsg tcpbpfsendmsg keeps msgtx across skstreamwaitmemory, which drops and reacquires the socket lock. Its error path tries to decide whether msgtx names the local temporary messag...

CVE-2026-68239 drm/ttm: Account for NULL and handle pages in ttm_pool_backup

In the Linux kernel, the following vulnerability has been resolved: drm/ttm: Account for NULL and handle pages in ttmpoolbackup Pages in ttmpoolbackup can be NULL or backup handles ttmbackuppageptrishandle, neither of which can be passed to setpagesarraywb or freed. Add a dedicated WB pass before...

CVE-2026-68239

In the Linux kernel, the following vulnerability has been resolved: drm/ttm: Account for NULL and handle pages in ttmpoolbackup Pages in ttmpoolbackup can be NULL or backup handles ttmbackuppageptrishandle, neither of which can be passed to setpagesarraywb or freed. Add a dedicated WB pass before...

EUVD-2026-55340

In the Linux kernel, the following vulnerability has been resolved: drm/ttm: Account for NULL and handle pages in ttmpoolbackup Pages in ttmpoolbackup can be NULL or backup handles ttmbackuppageptrishandle, neither of which can be passed to setpagesarraywb or freed. Add a dedicated WB pass before...

CVE-2026-68239 drm/ttm: Account for NULL and handle pages in ttm_pool_backup

In the Linux kernel, the following vulnerability has been resolved: drm/ttm: Account for NULL and handle pages in ttmpoolbackup Pages in ttmpoolbackup can be NULL or backup handles ttmbackuppageptrishandle, neither of which can be passed to setpagesarraywb or freed. Add a dedicated WB pass before...

CVE-2026-68239

CVE-2026-68239 affects the Linux kernel DRM TTM code, specifically the ttm_pool_backup path. The issue arises because ttm_pool_backup could encounter NULL or backup handle entries (ttm_backup_page_ptr_is_handle()) which were not safely guarded for set_pages_array_wb() and could be freed erroneous...

EUVD-2026-55338

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/userq: fix indefinite fence wait during GPU reset prereset only force-completes fences of MAPPED queues. A queue in any other state e.g. mid-eviction keeps its lastfence pending; after a GPU reset that fence never...

CVE-2026-68237

CVE-2026-68237 resolves an indefinite fence wait in drm/amdgpu/userq during a GPU reset. The pre_reset path only force-completes fences of MAPPED queues, leaving fences for queues in states like mid-eviction pending; after a reset, those fences may never signal, causing eviction/suspend workers a...

CVE-2026-68237 drm/amdgpu/userq: fix indefinite fence wait during GPU reset

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/userq: fix indefinite fence wait during GPU reset prereset only force-completes fences of MAPPED queues. A queue in any other state e.g. mid-eviction keeps its lastfence pending; after a GPU reset that fence never...

CVE-2026-68237

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/userq: fix indefinite fence wait during GPU reset prereset only force-completes fences of MAPPED queues. A queue in any other state e.g. mid-eviction keeps its lastfence pending; after a GPU reset that fence never...

CVE-2026-68237 drm/amdgpu/userq: fix indefinite fence wait during GPU reset

In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/userq: fix indefinite fence wait during GPU reset prereset only force-completes fences of MAPPED queues. A queue in any other state e.g. mid-eviction keeps its lastfence pending; after a GPU reset that fence never...

EUVD-2026-55290

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: Protect UUID list traversal The hcisync conversion moved class-of-device and EIR generation from an HCI request built under hdev-lock to asynchronous command sync work. The worker holds hdev-reqlock, but that...

CVE-2026-68189 Bluetooth: hci_sync: Protect UUID list traversal

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: Protect UUID list traversal The hcisync conversion moved class-of-device and EIR generation from an HCI request built under hdev-lock to asynchronous command sync work. The worker holds hdev-reqlock, but that...

CVE-2026-68189

CVE-2026-68189 affects the Linux kernel Bluetooth hci_sync component. A race between asynchronous command sync work and UUID list updates can cause a use-after-free in eir_create due to non-serialized access to hdev->uuids. The fix/mitigation is to hold hdev->lock while generating and commi...

CVE-2026-68189

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: Protect UUID list traversal The hcisync conversion moved class-of-device and EIR generation from an HCI request built under hdev-lock to asynchronous command sync work. The worker holds hdev-reqlock, but that...

CVE-2026-68189 Bluetooth: hci_sync: Protect UUID list traversal

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: Protect UUID list traversal The hcisync conversion moved class-of-device and EIR generation from an HCI request built under hdev-lock to asynchronous command sync work. The worker holds hdev-reqlock, but that...