CVE-2026-74537 Bluetooth: ISO: hold sk properly in iso_conn_ready

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: ISO: hold sk properly in isoconnready sk deref in isoconnready must be done either under conn-lock, or holding a refcount, to avoid concurrent close. conn-sk is currently accessed without either: Task 1 Task 2...

EUVD-2026-59724

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hciconn: hold conn reference in abortconnsync There is theoretical UAF if the conn is freed while the hcisync task is running. Hold refcount to avoid that...

CVE-2026-74530

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: hold conn in hciconnectbigsync callback There is theoretical UAF if the conn is freed while the hcisync task is running. Hold refcount to avoid that. Handle NULL hcon, return 0 + do nothing to match the previo...

EUVD-2026-59723

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: hold conn in hciconnectbigsync callback There is theoretical UAF if the conn is freed while the hcisync task is running. Hold refcount to avoid that. Handle NULL hcon, return 0 + do nothing to match the previo...

CVE-2026-74531

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hciconn: hold conn reference in abortconnsync There is theoretical UAF if the conn is freed while the hcisync task is running. Hold refcount to avoid that...

CVE-2026-74530

CVE-2026-74530 : Linux kernel Bluetooth vulnerability in the hci_sync path (hci_connect_big_sync) may allow a use-after-free if the conn is freed while the hci_sync task runs. The root cause is a potential UAF due to a conn lifecycle race; the fix holds a refcount to prevent use-after-free and ha...

CVE-2026-74530 Bluetooth: hci_sync: hold conn in hci_connect_big_sync() callback

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: hold conn in hciconnectbigsync callback There is theoretical UAF if the conn is freed while the hcisync task is running. Hold refcount to avoid that. Handle NULL hcon, return 0 + do nothing to match the previo...

CVE-2026-74528

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: hold conn in hcipastsync callback Avoids giving freed pointers to hciconnvalid, which kmalloc may have reused. Hold refcount to avoid that...

CVE-2026-74528 Bluetooth: hci_sync: hold conn in hci_past_sync() callback

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: hold conn in hcipastsync callback Avoids giving freed pointers to hciconnvalid, which kmalloc may have reused. Hold refcount to avoid that...

CVE-2026-74529

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: hold conn in hciconnectpasync callback There is theoretical UAF if the conn is freed while the hcisync task is running. Hold refcount to avoid that...

CVE-2026-74529 Bluetooth: hci_sync: hold conn in hci_connect_pa_sync() callback

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: hold conn in hciconnectpasync callback There is theoretical UAF if the conn is freed while the hcisync task is running. Hold refcount to avoid that...

CVE-2026-74529 Bluetooth: hci_sync: hold conn in hci_connect_pa_sync() callback

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: hold conn in hciconnectpasync callback There is theoretical UAF if the conn is freed while the hcisync task is running. Hold refcount to avoid that...

CVE-2026-74529

In the Linux kernel Bluetooth subsystem, the issue is in the hci_sync path: the conn object could be freed while the hci_sync task is running in hci_connect_pa_sync(), creating a theoretical use-after-free. The documented fix is to hold a refcount on the conn to prevent deallocation during the ca...

EUVD-2026-59721

In the Linux kernel, the following vulnerability has been resolved: Bluetooth: hcisync: hold conn in hcipastsync callback Avoids giving freed pointers to hciconnvalid, which kmalloc may have reused. Hold refcount to avoid that...

CVE-2026-74528

The CVE refers to the Linux kernel Bluetooth subsystem: hci_sync in the hci_past_sync() callback, which previously could dereference freed pointers to hci_conn_valid() due to kmalloc reuse. The fix holds a reference count on the connection to ensure resources remain valid, preventing use-after-fr...

CVE-2026-74513 dibs: fix use-after-free of dmb_node in loopback attach/detach/unregister

In the Linux kernel, the following vulnerability has been resolved: dibs: fix use-after-free of dmbnode in loopback attach/detach/unregister dibsloattachdmb, dibslodetachdmb and dibslounregisterdmb look up the dmbnode under dmbhtlock, drop the lock and only then operate on the node's refcount...

CVE-2026-74513

In the Linux kernel, the following vulnerability has been resolved: dibs: fix use-after-free of dmbnode in loopback attach/detach/unregister dibsloattachdmb, dibslodetachdmb and dibslounregisterdmb look up the dmbnode under dmbhtlock, drop the lock and only then operate on the node's refcount...

EUVD-2026-59671

In the Linux kernel, the following vulnerability has been resolved: um: vector: fix use-after-free in vectormmsgrx When vectormmsgrx discards a packet whose overlay header fails verifyheader, it frees the skb and continues the loop: if headercheck estats.rxencapserrors++; continue; The normal and...

CVE-2026-74347

In the Linux kernel, the following vulnerability has been resolved: netfilter: cttimeout: detach dataplane timeout policy and repurpose refcount Add a refcount for struct nfcttimeout which is used by ct extension to set the custom ct timeout policy, this tells us that the ct timeout is being used...

CVE-2026-74289

In the Linux kernel, the following vulnerability has been resolved: ipv4: fib: Don't dump dying fibinfo in fibleafnotify. syzbot reported use-after-free in nsimfib4prepareevent. 0 The problem is that the following functions call fibinfohold / refcountinc while dumping fibinfo under RCU, which is...