Kiran K [Tue, 30 Jun 2026 16:59:19 +0000 (22:29 +0530)]
Bluetooth: btintel_pcie: Refactor FLR to use device_reprobe()
The FLR branch in btintel_pcie_reset_work() open-coded the entire
re-init sequence: btintel_pcie_release_hdev() (hci_unregister_dev +
hci_free_dev), pci_try_reset_function(), enable_interrupts /
config_msix / enable_bt / reset_ia / start_rx, then
btintel_pcie_setup_hdev() (hci_alloc_dev_priv + hci_register_dev).
Every probe() init step had to be kept in sync with this second
copy in the reset path, and any failure mid-sequence left state to
unwind by hand.
The PLDR path already delegates teardown and re-init to the PCI
core via device_reprobe(): .remove() destroys data through devres
and unregisters hdev, then .probe() rebuilds everything from
scratch. Apply the same model to FLR.
Introduce btintel_pcie_perform_flr() mirroring perform_pldr(). It
runs pci_try_reset_function() (required to avoid the device_lock
ABBA against btintel_pcie_remove(), which calls
disable_work_sync(&reset_work) while holding device_lock) followed
by device_reprobe(). On success, data is destroyed and a fresh
probe re-INIT_WORKs coredump_work with disable count 0, so
enable_work() must not be called; on failure, data is still alive
and the caller balances the earlier disable_work_sync(). The
contract is documented on the helper and reiterated at the
reset_work() call site.
reset_work() shrinks to interrupt/worker drain, dispatch on
reset_type, and the single asymmetry between the two paths. The
out_enable label, the manual unregister/register pair, and the
forward declaration of btintel_pcie_setup_hdev() are dropped.
No intended functional change; FLR and PLDR now share one
teardown contract.
Fixes: 256ab9520d15 ("Bluetooth: btintel_pcie: Support Function level reset") Assisted-by: GitHub-Copilot:claude-4.7-opus Signed-off-by: Kiran K <kiran.k@intel.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Siwei Zhang [Mon, 29 Jun 2026 13:49:58 +0000 (09:49 -0400)]
Bluetooth: L2CAP: Fix use-after-free in l2cap_sock_new_connection_cb()
l2cap_sock_new_connection_cb() returned l2cap_pi(sk)->chan after
release_sock(parent). Once the parent lock is dropped the newly
enqueued child socket sk is reachable via the accept queue, so another
task can accept and free it before the callback dereferences sk,
resulting in a use-after-free.
Rework the ->new_connection() op so the core, rather than the callback,
owns the child channel's lifetime. The op now receives a pre-allocated
new_chan and returns an errno instead of allocating and returning a
channel. l2cap_new_connection() allocates the child channel and links
it into the conn list via __l2cap_chan_add() before invoking the
callback, so the conn-list reference keeps the channel alive once
release_sock(parent) exposes the socket to other tasks.
Channel configuration that was duplicated in l2cap_sock_init() and the
various new_connection callbacks is consolidated into
l2cap_chan_set_defaults(), which now inherits from the parent channel
when one is supplied.
Fixes: 8ffb929098a5 ("Bluetooth: Remove parent socket usage from l2cap_core.c") Cc: stable@kernel.org Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Siwei Zhang <oss@fourdim.xyz> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Yousef Alhouseen [Sun, 28 Jun 2026 00:23:05 +0000 (02:23 +0200)]
Bluetooth: fix UAF in bt_accept_dequeue()
bt_accept_get() takes a temporary reference before dropping the accept
queue lock. bt_accept_dequeue() currently drops that reference before
bt_accept_unlink(), leaving only the queue reference.
bt_accept_unlink() drops the queue reference. The subsequent
sock_hold() therefore accesses freed memory if it was the final
reference, as observed by KASAN during listening L2CAP socket cleanup.
Retain the temporary queue-walk reference through unlink and hand it to
the caller on success. Drop it explicitly on the closed and
not-yet-connected paths.
Fixes: ab1513597c6c ("Bluetooth: fix UAF in l2cap_sock_cleanup_listen() vs l2cap_conn_del()") Reported-by: syzbot+674ff7e4d7fdfd572afc@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=674ff7e4d7fdfd572afc Cc: stable@vger.kernel.org Signed-off-by: Yousef Alhouseen <alhouseenyousef@gmail.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Yousef Alhouseen [Sun, 28 Jun 2026 00:50:58 +0000 (02:50 +0200)]
Bluetooth: bnep: pin L2CAP connection during netdev registration
bnep_add_connection() reads the L2CAP connection without holding the
channel lock, then passes its HCI device to register_netdev(). Controller
teardown can clear and release that connection concurrently, leaving the
network device registration path to dereference a freed parent device.
Take a reference to the L2CAP connection while holding the channel lock.
Retain it until register_netdev() has taken the parent device reference.
Fixes: 65f53e9802db ("Bluetooth: Access BNEP session addresses through L2CAP channel") Reported-by: syzbot+fed5dce4553262f3b35c@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=fed5dce4553262f3b35c Cc: stable@vger.kernel.org Signed-off-by: Yousef Alhouseen <alhouseenyousef@gmail.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Sungwoo Kim [Wed, 24 Jun 2026 21:33:04 +0000 (17:33 -0400)]
Bluetooth: sco: Fix a race condition in sco_sock_timeout()
sco_sock_timeout() runs asynchronously and lock_sock(sk). If the socket
is closing while the timer is running, it holds the same lock
(lock_sock(sk)) twice, leading to a deadlock.
CPU 0 CPU 1
==================== ======================
sco_sock_close()
sco_sock_timeout()
lock_sock(sk) // <-- LOCK
__sco_sock_close()
sco_chan_del()
sco_conn_put()
sco_conn_free()
disable_delayed_work_sync()
lock(sk) // <-- SAME LOCK
Fix this by moving disable_delayed_work_sync() outside of lock_sock(sk),
ensuring that no lock_sock(sk) is held before sco_sock_timeout().
Lockdep splat:
WARNING: possible circular locking dependency detected
6.13.0-rc4 #7 Not tainted
Fixes: e6720779ae61 ("Bluetooth: SCO: Use kref to track lifetime of sco_conn") Acked-by: Dave Tian <daveti@purdue.edu> Signed-off-by: Sungwoo Kim <iam@sung-woo.kim> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
hci_add_adv_monitor() publishes a new adv_monitor in
hdev->adv_monitors_idr before the powered MSFT setup step. The MSFT
offload add path can then fail either locally before the controller add
command completes, or in the MSFT add callback. In the current queued
management add flow, hci_cmd_sync_work() still invokes
mgmt_add_adv_patterns_monitor_complete() with the original pending command
after msft_add_monitor_pattern() returns.
The buggy scenario involves two paths, with each column showing the order
within that path:
Local MSFT setup failures have the other half of the same ownership bug:
they return an error after the IDR insertion, but no later code removes the
failed monitor from the IDR.
Keep ownership with the pending management command until its completion.
For normal management adds, the MSFT add callback now records successful
controller state and returns errors to its caller. The management
completion frees the monitor on non-success after copying the response
handle, while resume/reregister callback-error cleanup remains in the
MSFT callback. The success path keeps the existing bookkeeping.
Validation reproduced this kernel report:
BUG: KASAN: slab-use-after-free in mgmt_add_adv_patterns_monitor_complete+0xfb/0x260 [bluetooth]
Allocated by task 471 on cpu 3 at 285.205389s:
kasan_save_stack+0x33/0x60
kasan_save_track+0x17/0x60
__kasan_kmalloc+0xaa/0xb0
add_adv_patterns_monitor_rssi+0xd5/0x230 [bluetooth]
hci_sock_sendmsg+0x96b/0xf80 [bluetooth]
__sys_sendto+0x2bc/0x2d0
__x64_sys_sendto+0x76/0x90
do_syscall_64+0x115/0x6a0
entry_SYSCALL_64_after_hwframe+0x77/0x7f
Freed by task 454 on cpu 2 at 285.217112s:
kasan_save_stack+0x33/0x60
kasan_save_track+0x17/0x60
kasan_save_free_info+0x3b/0x60
__kasan_slab_free+0x5f/0x80
kfree+0x313/0x590
msft_add_monitor_sync+0x54a/0x570 [bluetooth]
hci_add_adv_monitor+0x133/0x180 [bluetooth]
hci_cmd_sync_work+0x187/0x210 [bluetooth]
process_one_work+0x4fd/0xbc0
worker_thread+0x2d8/0x570
kthread+0x1ad/0x1f0
ret_from_fork+0x3c9/0x540
ret_from_fork_asm+0x1a/0x30
Fixes: a2a4dedf88ab ("Bluetooth: advmon offload MSFT add monitor") Assisted-by: Codex:gpt-5.5 Signed-off-by: Cen Zhang <zzzccc427@gmail.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Cen Zhang [Tue, 23 Jun 2026 16:12:59 +0000 (00:12 +0800)]
Bluetooth: 6lowpan: hold L2CAP conn across debugfs control
get_l2cap_conn() looks up an LE hci_conn under hdev protection, but
then drops that protection before reading hcon->l2cap_data and before
lowpan_control_write() later dereferences conn->hcon. A disconnect or
device close can tear down the same L2CAP connection in that window.
The buggy scenario involves two paths, with each column showing the order
within that path:
6LoWPAN control write: HCI disconnect/device close:
1. get_l2cap_conn() finds hcon 1. hci_disconn_cfm() dispatches
and hcon->l2cap_data. the L2CAP disconnect callback.
2. get_l2cap_conn() drops hdev 2. l2cap_conn_del() clears
protection and returns conn. hcon->l2cap_data and drops the
L2CAP connection reference.
3. lowpan_control_write() reads 3. hci_conn_del() removes and drops
conn->hcon. the HCI connection.
Take a reference to the L2CAP connection with
l2cap_conn_hold_unless_zero() while hdev is still locked, and drop that
reference after the debugfs command's last use of conn. This mirrors the
existing L2CAP ACL receive-side handoff and keeps the connection
dereferenceable after leaving hdev protection. Export the existing helper
so the bluetooth_6lowpan module can use the same lifetime primitive.
Validation reproduced this kernel report:
BUG: KASAN: slab-use-after-free in lowpan_control_write+0x374/0x520
The buggy address belongs to the object at ffff888111b9d000 which belongs
to the cache kmalloc-1k of size 1024
The buggy address is located 0 bytes inside of freed 1024-byte region
[ffff888111b9d000, ffff888111b9d400)
Read of size 8
Call trace:
dump_stack_lvl+0x66/0xa0
print_report+0xce/0x5f0
lowpan_control_write+0x374/0x520 (net/bluetooth/6lowpan.c:1131)
srso_alias_return_thunk+0x5/0xfbef5
__virt_addr_valid+0x19f/0x330
kasan_report+0xe0/0x110
__debugfs_file_get+0xf7/0x400
full_proxy_write+0x9e/0xd0
vfs_write+0x1b0/0x810
ksys_write+0xd2/0x170
dnotify_flush+0x32/0x220
do_syscall_64+0x115/0x6a0 (arch/x86/entry/syscall_64.c:87)
entry_SYSCALL_64_after_hwframe+0x77/0x7f
Allocated by task stack:
kasan_save_stack+0x33/0x60
kasan_save_track+0x17/0x60
__kasan_kmalloc+0xaa/0xb0
l2cap_conn_add+0x45/0x520
l2cap_chan_connect+0xac6/0xd90
l2cap_sock_connect+0x216/0x350
__sys_connect+0x101/0x130
__x64_sys_connect+0x40/0x50
do_syscall_64+0x115/0x6a0 (arch/x86/entry/syscall_64.c:87)
entry_SYSCALL_64_after_hwframe+0x77/0x7f
Freed by task stack:
kasan_save_stack+0x33/0x60
kasan_save_track+0x17/0x60
kasan_save_free_info+0x3b/0x60
__kasan_slab_free+0x5f/0x80
kfree+0x313/0x590
hci_conn_hash_flush+0xc0/0x140
hci_dev_close_sync+0x41a/0xb00
hci_dev_close+0x12f/0x160
hci_sock_ioctl+0x157/0x570
sock_do_ioctl+0xf7/0x210
sock_ioctl+0x32f/0x490
__x64_sys_ioctl+0xc7/0x110
do_syscall_64+0x115/0x6a0 (arch/x86/entry/syscall_64.c:87)
entry_SYSCALL_64_after_hwframe+0x77/0x7f
kasan_record_aux_stack+0xa7/0xc0
insert_work+0x32/0x100
__queue_work+0x262/0xa60
queue_work_on+0xad/0xb0
l2cap_connect_cfm+0x4ef/0x670
hci_le_remote_feat_complete_evt+0x247/0x430
hci_event_packet+0x360/0x6f0
hci_rx_work+0x2ae/0x7a0
process_one_work+0x4fd/0xbc0
worker_thread+0x2d8/0x570
kthread+0x1ad/0x1f0
ret_from_fork+0x3c9/0x540
ret_from_fork_asm+0x1a/0x30
Fixes: 6b8d4a6a0314 ("Bluetooth: 6LoWPAN: Use connected oriented channel instead of fixed one") Assisted-by: Codex:gpt-5.5 Signed-off-by: Cen Zhang <zzzccc427@gmail.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Cen Zhang [Tue, 23 Jun 2026 16:12:29 +0000 (00:12 +0800)]
Bluetooth: 6lowpan: avoid untracked enable work
lowpan_enable_set() allocates a temporary work item and schedules
do_enable_set() on system_wq, then returns to debugfs. The debugfs active
operation has ended at that point, but the worker still executes module
text and manipulates enable_6lowpan and listen_chan.
bt_6lowpan_exit() removes the debugfs files and immediately closes and
puts listen_chan. It has no pointer to the queued work item, so it cannot
cancel or flush it before tearing down the state that the worker uses.
The buggy scenario involves two paths, with each column showing the order
within that path:
debugfs enable write module exit
1. lowpan_enable_set() allocates 1. bt_6lowpan_exit() removes
set_enable work the debugfs file
2. schedule_work() queues 2. bt_6lowpan_exit() closes
do_enable_set() and puts listen_chan
3. the write operation returns 3. module teardown can continue
4. do_enable_set() later runs
against stale state
Run the enable state transition synchronously in lowpan_enable_set()
instead. The simple debugfs setter can sleep, and this file already handles
the 6LoWPAN control write synchronously under the same set_lock. Once the
setter returns, debugfs removal covers the whole operation and exit can no
longer race with an untracked work item.
Validation reproduced this kernel report:
BUG: KASAN: slab-use-after-free in do_enable_set+0x113/0x2e0
Workqueue: events do_enable_set [bluetooth_6lowpan]
The buggy address belongs to the object at ffff888109cb8000
Fixes: 90305829635d ("Bluetooth: 6lowpan: Converting rwlocks to use RCU") Assisted-by: Codex:gpt-5.5 Signed-off-by: Cen Zhang <zzzccc427@gmail.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Siwei Zhang [Mon, 15 Jun 2026 15:33:05 +0000 (11:33 -0400)]
Bluetooth: hci_conn: Fix null ptr deref in hci_abort_conn()
hci_abort_conn() read hci_skb_event(hdev->sent_cmd) when a connection
was pending, but hdev->sent_cmd can be NULL while req_status is still
HCI_REQ_PEND, leading to a NULL pointer dereference and a general
protection fault from the hci_rx_work() receive path.
Instead of inspecting hdev->sent_cmd, track the in-flight create
connection command with a new per-connection HCI_CONN_CREATE flag and
route all cancellation through hci_cancel_connect_sync(), which
dispatches to a dedicated per-type cancel function. The create command
is in exactly one of two states: still queued, or in flight. The cancel
function holds cmd_sync_work_lock across the whole decision: the worker
takes this lock to dequeue every entry, so while it is held a queued
command cannot start running and an in-flight command cannot complete
and let the next command become pending. This keeps the flag test and
hci_cmd_sync_cancel() atomic with respect to the worker, so a queued
command is simply dequeued, and an in-flight command owned by this
connection is cancelled without the risk of cancelling an unrelated
command that became pending in the meantime. CIS uses the same flag
mechanism via HCI_CONN_CREATE_CIS but cannot be dequeued per-connection.
hci_acl_create_conn_sync() and hci_le_create_conn_sync() clear
HCI_CONN_CREATE after the create command completes, but the command
status handler can free conn via hci_conn_del() (for example when the
controller rejects the connection) while the worker is still blocked on
the connection complete event. Hold a reference on conn across the
create command so the flag can be cleared without a use-after-free.
Fixes: a13f316e90fd ("Bluetooth: hci_conn: Consolidate code for aborting connections") Cc: stable@vger.kernel.org Suggested-by: XIAO WU <xiaowu.417@qq.com> Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Siwei Zhang <oss@fourdim.xyz> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Maoyi Xie [Wed, 17 Jun 2026 08:36:52 +0000 (16:36 +0800)]
Bluetooth: btnxpuart: Fix out-of-bounds firmware read in nxp_recv_fw_req_v3()
During the v3 firmware download the controller sends a v3_data_req with a
32 bit offset and a 16 bit len. nxp_recv_fw_req_v3() checks only the lower
bound of the offset and then sends firmware from that offset.
Nothing checks that fw_dnld_v3_offset + len stays within nxpdev->fw->size,
so a controller that asks for an offset or length past the firmware image
makes the driver read past the end of nxpdev->fw->data and send that
memory back over UART.
nxp_recv_fw_req_v1() already bounds the same write. Add the equivalent
check to the v3 path, reject the request when it falls outside the firmware
image, and zero len on the error path so the fw_v3_prev_sent bookkeeping at
free_skb stays consistent.
Fixes: 689ca16e5232 ("Bluetooth: NXP: Add protocol support for NXP Bluetooth chipsets") Suggested-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com> Reviewed-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com> Cc: stable@vger.kernel.org Signed-off-by: Maoyi Xie <maoyixie.tju@gmail.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Muhammad Bilal [Sat, 20 Jun 2026 19:56:35 +0000 (00:56 +0500)]
Bluetooth: L2CAP: validate option length before reading conf opt value
l2cap_get_conf_opt() derives the option length from the
attacker-controlled opt->len field and immediately dereferences
opt->val (as u8, get_unaligned_le16() or get_unaligned_le32(), or a
raw pointer for the default case) before any caller has confirmed
that opt->len bytes are present in the buffer. The callers
(l2cap_parse_conf_req(), l2cap_parse_conf_rsp() and
l2cap_conf_rfc_get()) only detect a malformed option afterwards, once
the running length has gone negative, by which point the
out-of-bounds read has already executed.
An existing post-hoc length check keeps the garbage value from being
consumed, so this is not a data leak in the current control flow. It
is still a validate-after-use ordering bug: up to 4 bytes are read
past the end of the buffer before it is known to contain them, and it
is fragile to future changes in the callers.
Fix it at the source. Pass the end of the buffer into
l2cap_get_conf_opt() and refuse to touch opt->val unless the full
option (header + value) fits. Each caller computes an end pointer
once before the loop and checks the return value directly instead of
inferring the error from a negative length.
Fixes: 7c9cbd0b5e38 ("Bluetooth: Verify that l2cap_get_conf_opt provides large enough buffer") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal <meatuni001@gmail.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Runyu Xiao [Wed, 17 Jun 2026 15:36:13 +0000 (23:36 +0800)]
Bluetooth: L2CAP: cancel pending_rx_work before taking conn->lock
l2cap_conn_del() takes conn->lock and then calls cancel_work_sync() for
pending_rx_work. process_pending_rx() takes the same mutex, so teardown
can deadlock against the worker it is flushing.
This issue was found by our static analysis tool and then manually
reviewed against the current tree.
The grounded PoC kept the l2cap_conn_ready() -> queue_work(...,
&conn->pending_rx_work) submit path, the l2cap_conn_del() ->
cancel_work_sync(&conn->pending_rx_work) teardown path, and the
process_pending_rx() -> mutex_lock(&conn->lock) worker edge. Lockdep
reported:
Cancel pending_rx_work before taking conn->lock, matching the existing
lock-before-drain ordering used for the two delayed works in the same
teardown path. The pending_rx queue is still purged after the work has
been cancelled and conn->lock has been acquired.
Fixes: 7ab56c3a6ecc ("Bluetooth: Fix deadlock in l2cap_conn_del()") Cc: stable@vger.kernel.org Signed-off-by: Runyu Xiao <runyu.xiao@seu.edu.cn> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Muhammad Bilal [Sun, 21 Jun 2026 16:23:05 +0000 (21:23 +0500)]
Bluetooth: ISO: avoid NULL deref of conn in iso_conn_big_sync()
iso_conn_big_sync() drops the socket lock to call hci_get_route() and
then re-acquires it, but dereferences iso_pi(sk)->conn->hcon afterwards
without re-checking that conn is still valid.
While the lock is dropped, the connection can be torn down under the
same socket lock: iso_disconn_cfm() -> iso_conn_del() -> iso_chan_del()
sets iso_pi(sk)->conn to NULL (and the broadcast teardown path can also
clear conn->hcon on its own). When iso_conn_big_sync() re-acquires the
lock and reads conn->hcon, conn may be NULL, causing a NULL pointer
dereference (hcon is the first member of struct iso_conn).
This path is reached from iso_sock_recvmsg() for a PA-sync broadcast
sink socket (BT_SK_DEFER_SETUP | BT_SK_PA_SYNC), so the dropped-lock
window can race with connection teardown driven by controller events.
Re-validate iso_pi(sk)->conn and its hcon after re-acquiring the socket
lock and bail out if the connection went away, as already done in the
sibling iso_sock_rebind_bc().
Fixes: 7a17308c17880d ("Bluetooth: iso: Fix circular lock in iso_conn_big_sync") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal <meatuni001@gmail.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Samuel Page [Mon, 15 Jun 2026 15:09:22 +0000 (16:09 +0100)]
Bluetooth: MGMT: Fix UAF of hci_conn_params in add_device_complete
add_device_complete() runs from the hci_cmd_sync_work kworker, which
holds only hci_req_sync_lock and *not* hci_dev_lock. It calls
hci_conn_params_lookup() and then dereferences the returned object
(params->flags) without taking hci_dev_lock:
hci_conn_params_lookup() walks hdev->le_conn_params and is documented to
require hdev->lock. A concurrent MGMT_OP_REMOVE_DEVICE
(remove_device()), which does run under hci_dev_lock, can call
hci_conn_params_free() to list_del() and kfree() the very object the
lookup returned, so the subsequent params->flags read touches freed
memory [0].
Hold hci_dev_lock() across the hci_conn_params_lookup() and the read of
params->flags (and the matching event emission) so the lookup result
cannot be freed by a concurrent remove_device() before it is used,
honouring the locking contract of hci_conn_params_lookup().
[0]: (trailing page/memory-state dump trimmed)
BUG: KASAN: slab-use-after-free in add_device_complete+0x358/0x3d8 net/bluetooth/mgmt.c:7671
Read of size 1 at addr ffff000017ab26c1 by task kworker/u9:8/388
Bluetooth: 6lowpan: Fix using chan->conn as indication to no remote netdev
b66774b48dd9 ("Bluetooth: L2CAP: Fix UAF in channel timeout by holding
conn ref") don't reset the chan->conn to NULL anymore making the bt#
netdev not be remove once the last l2cap_chan_del is removed.
Instead of restoring the original behavior this remove the logic of
keeping the interface after the last channel is removed because it
never worked as intended and the l2cap_chan_del always detach its
l2cap_conn which results in always removing the channel anyway.
Fixes: b66774b48dd9 ("Bluetooth: L2CAP: Fix UAF in channel timeout by holding conn ref") Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Pauli Virtanen [Sat, 13 Jun 2026 18:43:37 +0000 (21:43 +0300)]
Bluetooth: hci_uart: clear HCI_UART_SENDING when write_work is canceled
HCI_UART_SENDING bit in tx_state means write_work is pending and blocks
queueing it again. Currently this bit is not cleared when canceling the
work in hci_uart_close(), which blocks future writes when device is
reopened later if write_work was pending.
Fix by clearing HCI_UART_SENDING when canceling the work.
Also make clearing of tx_skb safe by using disable_work_sync +
enable_work instead of just cancel_work_sync. hci_uart_flush() purges
the proto tx queue so we can cancel the pending write_work there,
instead of doing it just in hci_uart_close(). Re-enable and possibly
requeue the work after queue flush.
Eric Dumazet [Wed, 1 Jul 2026 12:23:29 +0000 (12:23 +0000)]
amt: fix size calculation in amt_get_size()
amt_get_size() incorrectly used sizeof(struct iphdr) for the sizes of
IFLA_AMT_DISCOVERY_IP, IFLA_AMT_REMOTE_IP, and IFLA_AMT_LOCAL_IP.
These attributes contain IPv4 addresses (__be32), not full IP headers.
Replace sizeof(struct iphdr) with sizeof(__be32) to avoid over-allocating
netlink message space.
Fixes: b9022b53adad ("amt: add control plane of amt interface") Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com> Link: https://patch.msgid.link/20260701122329.3562825-1-edumazet@google.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Xiang Mei [Tue, 30 Jun 2026 18:32:27 +0000 (11:32 -0700)]
net/smc: fix UAF in smc_cdc_rx_handler() by pinning the socket
smc_cdc_rx_handler() looks up the connection by token under the link
group's conns_lock, drops the lock, and then dereferences conn and the
smc_sock derived from it, ending in sock_hold(&smc->sk) inside
smc_cdc_msg_recv(). No reference is held across the lock release.
The only reference pinning the socket while the connection is
discoverable in the link group is taken in smc_lgr_register_conn()
(sock_hold) and dropped in __smc_lgr_unregister_conn() (sock_put), both
under conns_lock. Once the handler drops conns_lock, a concurrent
close() -> smc_release() -> smc_conn_free() -> smc_lgr_unregister_conn()
can drop that reference and free the smc_sock, so the handler's later
sock_hold() runs on freed memory:
WARNING: lib/refcount.c:25 at refcount_warn_saturate
Workqueue: rxe_wq do_work
refcount_warn_saturate (lib/refcount.c:25)
smc_cdc_msg_recv (net/smc/smc_cdc.c:430)
smc_cdc_rx_handler (net/smc/smc_cdc.c:502)
smc_wr_rx_tasklet_fn (net/smc/smc_wr.c:445)
tasklet_action_common (kernel/softirq.c:938)
handle_softirqs (kernel/softirq.c:622)
Kernel panic - not syncing: panic_on_warn set
Only SMC-R is affected. The SMC-D receive tasklet is stopped by
tasklet_kill(&conn->rx_tsklet) in smc_conn_free() before the connection
is unregistered, so it cannot run concurrently with the free.
Take the socket reference while still holding conns_lock, so the
registration reference can no longer be the last one, and drop it once
the handler is done.
Fixes: d7b0e37c1ac1 ("net/smc: restructure CDC message reception") Reported-by: Weiming Shi <bestswngs@gmail.com> Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Xiang Mei <xmei5@asu.edu> Link: https://patch.msgid.link/20260630183227.2044998-1-xmei5@asu.edu Signed-off-by: Paolo Abeni <pabeni@redhat.com>
net/sched: act_pedit: fix TOCTOU heap OOB write in tc offload
There is a TOCTOU race condition in flower lockless approach between sizing
a flow_rule buffer and filling it.
zdi-disclosures@trendmicro.com reports:
The cls_flower classifier operates with TCF_PROTO_OPS_DOIT_UNLOCKED
(fl_change runs without RTNL), while RTM_NEWACTION holds RTNL, so the
independent locking domains make the race reachable in practice. KASAN
confirms:
BUG: KASAN: slab-out-of-bounds in tcf_pedit_offload_act_setup+0x81b/0x930
Write of size 4 at addr ffff888001f27520 by task poc-toctou/312
The buggy address is located 0 bytes to the right of
allocated 288-byte region [ffff888001f27400, ffff888001f27520)
(cache kmalloc-512)
Note: The result is a heap OOB write attacker-controlled content into the
adjacent slab object (requires CAP_NET_ADMIN).
The fix introduces reading tcfp_nkeys under act->tcfa_lock in all places
using a new tcf_pedit_nkeys_locked() which replaces the old tcf_pedit_nkeys().
Additionally we close the remaining TOCTOU window between the sizing read and
the fill reads by more careful accounting.
Rather than silently truncating the key count, which leads to incorrect
action semantics offloaded to hardware and secondary OOB writes if
the remaining capacity is zero or consumed by prior actions, we enforce
remaining capacity checks and return -ENOSPC if the required space exceeds
the remaining capacity.
Fixes: 71d0ed7079df ("net/act_pedit: Support using offset relative to the conventional network headers") Reported-by: zdi-disclosures@trendmicro.com Tested-by: Victor Nogueira <victor@mojatatu.com> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Link: https://patch.msgid.link/20260701161912.125355-1-jhs@mojatatu.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Xiang Mei [Tue, 30 Jun 2026 17:41:09 +0000 (10:41 -0700)]
net: qualcomm: rmnet: validate MAP frame length before ingress parsing
When ingress deaggregation is disabled, rmnet_map_ingress_handler() passes
the skb straight to __rmnet_map_ingress_handler(), skipping the length
validation that rmnet_map_deaggregate() performs on the aggregated path.
The parser then dereferences the MAP header and csum header/trailer based on
the on-wire pkt_len without checking skb->len, so a short frame is read out
of bounds:
BUG: KASAN: slab-out-of-bounds in rmnet_map_checksum_downlink_packet
Read of size 1 at addr ffff88801118ed00 by task exploit/147
Call Trace:
...
rmnet_map_checksum_downlink_packet (drivers/net/ethernet/qualcomm/rmnet/rmnet_map_data.c:413)
__rmnet_map_ingress_handler (drivers/net/ethernet/qualcomm/rmnet/rmnet_handlers.c:96)
rmnet_rx_handler (drivers/net/ethernet/qualcomm/rmnet/rmnet_handlers.c:129)
__netif_receive_skb_core.constprop.0 (net/core/dev.c:6089)
netif_receive_skb (net/core/dev.c:6460)
tun_get_user (drivers/net/tun.c:1955)
tun_chr_write_iter (drivers/net/tun.c:2001)
vfs_write (fs/read_write.c:688)
ksys_write (fs/read_write.c:740)
do_syscall_64 (arch/x86/entry/syscall_64.c:94)
...
Factor that validation out of rmnet_map_deaggregate() into
rmnet_map_validate_packet_len() and run it on the no-aggregation path too.
The MAP header is bounds-checked first, since this path can receive a frame
shorter than the header.
Paolo Abeni [Mon, 6 Jul 2026 10:39:42 +0000 (12:39 +0200)]
Merge tag 'nf-26-07-03' of https://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf
Florian Westphal says:
====================
netfilter: updates for net
The following patchset contains Netfilter fixes for *net*, all
for ancient problems. Patch 7 raised drive-by sashiko findings,
but those are not related to the change itself.
1) Rebuild the nf_nat_sip data pointer to prevent stale access after SKB
reallocation. Restrict UDP mangling to UDP streams to avoid TCP packet
corruption.
2) Prevent undefined behavior in xt_u32 caused by invalid shift counts.
From Wyatt Feng.
3) Use u64 variables to prevent incorrect comparisons on links exceeding
34 Gbps in xt_rateest. From Feng Wu.
4) Cap the number of expectations per master during nfnetlink_cthelper
updates. From Pablo Neira Ayuso.
5) Mark malformed IPv6 extension headers for hotdrop in ip6tables.
From Zhixing Chen.
6) Skip the end element of an open interval during the get command when its
closest match is the interval's start element. Also from Pablo Neira Ayuso.
7) Fix PMTU calculation for GUE/GRE tunnels in IPVS during ICMP fragmentation
error handling. Include additional tunnel header length when computing the
new MTU. From Yizhou Zhao.
8) Reset full ip_vs_seq structures in ip_vs_conn_new. Also from Yizhou Zhao.
9) Reject invalid shift parameters in xt_connmark. Also from Wyatt Feng.
netfilter pull request nf-26-07-03
* tag 'nf-26-07-03' of https://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
netfilter: xt_connmark: reject invalid shift parameters
ipvs: reset full ip_vs_seq structs in ip_vs_conn_new
ipvs: fix PMTU for GUE/GRE tunnel ICMP errors
netfilter: nft_set_rbtree: get command skips end element with open interval
netfilter: ip6tables: mark malformed IPv6 extension headers for hotdrop
netfilter: nfnetlink_cthelper: cap to maximum number of expectation per master on updates
netfilter: xt_rateest: fix u64 truncation in xt_rateest_mt()
netfilter: xt_u32: reject invalid shift counts
netfilter: nf_nat_sip: reload possible stale data pointer
====================
Nirmoy Das [Tue, 30 Jun 2026 16:51:57 +0000 (09:51 -0700)]
selftests: net: make busywait timeout clock portable
loopy_wait() expects millisecond timestamps. However, Ubuntu Resolute
can use uutils date, where `date -u +%s%3N` returns seconds plus full
nanoseconds instead of a 3-digit millisecond field. This makes
busywait expire too early and can make vlan_bridge_binding.sh read a
stale operstate.
Shigeru Yoshida [Tue, 30 Jun 2026 16:46:20 +0000 (01:46 +0900)]
qede: fix off-by-one in BD ring consumption on build_skb failure
qede_rx_build_skb() and qede_tpa_rx_build_skb() do not check for a
NULL return from qede_build_skb(). When it returns NULL under memory
pressure, the functions still consume a BD from the ring before
returning NULL. The callers then recycle additional BDs, resulting in
one extra BD being consumed (off-by-one). This desynchronizes the BD
ring, which can corrupt DMA page reference counts and lead to SLUB
freelist corruption.
Commit 4e910dbe3650 ("qede: confirm skb is allocated before using")
added a NULL check inside qede_build_skb() to prevent a NULL pointer
dereference, but did not address the missing NULL checks in the
callers, making this off-by-one reachable.
Fix this by adding NULL checks for the return value of
qede_build_skb() in both qede_rx_build_skb() and
qede_tpa_rx_build_skb(), returning NULL immediately before any BD ring
manipulation.
Fixes: 8a8633978b84 ("qede: Add build_skb() support.") Signed-off-by: Shigeru Yoshida <syoshida@redhat.com> Reviewed-by: Jamie Bainbridge <jamie.bainbridge@gmail.com> Link: https://patch.msgid.link/20260630164623.3152625-1-syoshida@redhat.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
net: microchip: vcap: fix races on the shared Super VCAP block
The VCAP instances on a chip are not independent, yet they are locked
independently. On sparx5 and lan969x the IS0 and IS2 instances are
backed by the same Super VCAP hardware block and share its cache and
command registers: every access drives the shared VCAP_SUPER_CTRL
register and moves data through the shared cache registers.
Accessing one instance therefore races with accessing another. The
per-instance admin->lock cannot prevent this, as each instance takes a
different lock.
The locking issue is mostly disguised by the fact that the core usage of
the vcap api runs under rtnl. However, the full rule dump in debugfs
decodes rules straight from hardware (a READ command followed by a cache
read) and runs outside rtnl, so it races a concurrent tc-flower rule
write to another Super VCAP instance.
Besides corrupting the dump, the read repopulates the shared cache
between the writers cache fill and its write command, so the writer
commits the wrong data and corrupts the hardware entry.
Introduce vcap_lock() and vcap_unlock() helpers and route every rule
lock site in the VCAP API and its debugfs code through them. Replace the
per-instance admin->lock with a single mutex in struct vcap_control that
serializes access to all instances. The helpers reach it through a new
admin->vctrl back-pointer, and the clients initialise and destroy the
control lock instead of a per-instance one.
No path holds more than one instance lock, so collapsing them onto a
single mutex cannot self-deadlock.
Shuangpeng Bai [Tue, 30 Jun 2026 19:48:56 +0000 (15:48 -0400)]
llc: fix SAP refcount leak in llc_ui_autobind()
llc_ui_autobind() opens a SAP after choosing a dynamic LSAP.
llc_sap_open() returns a reference owned by the caller, and
llc_sap_add_socket() takes a second reference for the socket's
membership in the SAP hash tables.
llc_ui_bind() drops the caller's reference after adding the socket,
but llc_ui_autobind() keeps it. When the socket is closed,
llc_sap_remove_socket() releases only the socket reference, leaving
the SAP on llc_sap_list with sk_count == 0.
This is user-visible because repeated autobind and close cycles can consume
all dynamic SAP values and make later autobinds fail with -EUSERS.
Drop the caller's reference after a successful autobind, matching
llc_ui_bind()'s ownership model.
====================
net/mlx5e: Fix crashes in dynamic per-channel stats and HV VHCA agent
Since per-channel stats were converted to be allocated and published
lazily at first channel open in commit fa691d0c9c08 ("net/mlx5e:
Allocate per-channel stats dynamically at first usage"),
priv->channel_stats[] and priv->stats_nch are filled in
incrementally during interface bring-up. This opened a window in
which the various stats readers - most of them reachable from
userspace via netlink/netdev stats queries - can race with
mlx5e_open_channel() on another CPU and observe partially
initialized state. The HV VHCA stats agent, which is created
before the channels are opened, hits related problems of its own.
Feng Liu [Tue, 30 Jun 2026 11:51:51 +0000 (14:51 +0300)]
net/mlx5e: Fix publication race for priv->channel_stats[]
mlx5e_channel_stats_alloc() publishes a new entry to
priv->channel_stats[] and then increments priv->stats_nch as a
publication token, but neither store carries any memory barrier:
priv->channel_stats[ix] = kvzalloc_node(...);
if (!priv->channel_stats[ix])
return -ENOMEM;
priv->stats_nch++;
Concurrent readers compute the loop bound from priv->stats_nch and
then dereference priv->channel_stats[i] using plain accesses, e.g.
for (i = 0; i < priv->stats_nch; i++) {
struct mlx5e_channel_stats *cs = priv->channel_stats[i];
... cs->rq.packets ...
}
On weakly-ordered architectures (ARM, PowerPC, RISC-V) the writes to
channel_stats[ix] and stats_nch may become visible to other CPUs out
of program order. A reader can observe stats_nch == N while still
seeing channel_stats[N-1] == NULL, leading to a NULL pointer
dereference in the channel_stats loop.
This has been observed in production on BlueField-3 DPUs (arm64),
where ovs-vswitchd queries netdev statistics over netlink during NIC
bringup, racing mlx5e_open_channel() -> mlx5e_channel_stats_alloc()
on another CPU:
Add mlx5e_stats_nch_write() and mlx5e_stats_nch_read() helpers in en.h
that wrap the smp_store_release()/smp_load_acquire() pair on stats_nch.
The release/acquire pair establishes the contract:
stats_nch == N => channel_stats[0..N-1] are visible and non-NULL.
Publish the stats_nch increment via mlx5e_stats_nch_write() in the
writer (mlx5e_channel_stats_alloc()), and read stats_nch via
mlx5e_stats_nch_read() in all readers: mlx5e RX/TX queue stats,
mlx5e_get_base_stats(), ethtool channels stats, IPoIB stats, the
sw_stats fold and the HV VHCA stats agent.
Fixes: fa691d0c9c08 ("net/mlx5e: Allocate per-channel stats dynamically at first usage") Signed-off-by: Feng Liu <feliu@nvidia.com> Reviewed-by: Eran Ben Elisha <eranbe@nvidia.com> Reviewed-by: Cosmin Ratiu <cratiu@nvidia.com> Reviewed-by: Nimrod Oren <noren@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Link: https://patch.msgid.link/20260630115151.729219-4-tariqt@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
mlx5e_hv_vhca_stats_create() registers the stats agent through
mlx5_hv_vhca_agent_create(). The helper publishes the agent in
hv_vhca->agents[type] under agents_lock and immediately schedules an
asynchronous control invalidation on the HV VHCA workqueue before
returning to mlx5e.
The asynchronous invalidation invokes the control agent's invalidate
callback, which reads the hypervisor control block and forwards the
command to mlx5e_hv_vhca_stats_control(). That callback may either:
- call cancel_delayed_work_sync(&priv->stats_agent.work), or
- call queue_delayed_work(priv->wq, &sagent->work, sagent->delay).
However, the delayed_work and priv->stats_agent.agent are only
initialized after mlx5_hv_vhca_agent_create() returns to mlx5e:
agent = mlx5_hv_vhca_agent_create(...); /* publish + invalidate */
...
priv->stats_agent.agent = agent; /* too late */
INIT_DELAYED_WORK(&priv->stats_agent.work, ...); /* too late */
If the asynchronous control path runs before the two assignments
above, it can:
- Operate on an uninitialized delayed_work whose timer.function is
NULL. queue_delayed_work() calls add_timer() unconditionally, so
when the timer expires the timer softirq invokes a NULL function
pointer.
- Re-initialize the timer later through INIT_DELAYED_WORK() while
the timer is already enqueued in the timer wheel, corrupting the
hlist (entry.pprev cleared while the previous bucket node still
points at this entry).
- When the worker eventually runs, mlx5e_hv_vhca_stats_work() reads
sagent->agent (NULL) and dereferences it inside
mlx5_hv_vhca_agent_write().
Fix this by:
- Initializing priv->stats_agent.work before invoking
mlx5_hv_vhca_agent_create(), so the work is always in a valid
state when the control callback observes it.
- Adding a struct mlx5_hv_vhca_agent **ctx_update out-parameter
to mlx5_hv_vhca_agent_create(). The helper writes the agent
pointer to *ctx_update before publishing into hv_vhca->agents[]
and triggering the agents_update flow, so any callback
subsequently invoked from that flow already sees a valid
priv->stats_agent.agent. This avoids having the control
callback participate in agent initialization.
While at it, access priv->stats_agent.agent with
READ_ONCE()/WRITE_ONCE() for the cross-CPU access with the worker, and
clear priv->stats_agent.buf on the agent_create() failure path.
Fixes: cef35af34d6d ("net/mlx5e: Add mlx5e HV VHCA stats agent") Signed-off-by: Feng Liu <feliu@nvidia.com> Reviewed-by: Eran Ben Elisha <eranbe@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Link: https://patch.msgid.link/20260630115151.729219-3-tariqt@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
mlx5e_hv_vhca_stats_create() is called from mlx5e_nic_enable(),
before mlx5e_open(). At that point priv->stats_nch is still zero,
because it is only ever incremented in mlx5e_channel_stats_alloc(),
which is reached only from mlx5e_open_channel().
mlx5e_hv_vhca_stats_buf_size() therefore returns 0, and
kvzalloc(0, GFP_KERNEL) returns ZERO_SIZE_PTR ((void *)16) rather
than NULL. The "if (!buf)" guard does not catch this, and
mlx5e_hv_vhca_stats_create() completes "successfully" with
priv->stats_agent.buf set to ZERO_SIZE_PTR.
Once channels are opened (priv->stats_nch > 0) and the hypervisor
enables stats reporting, mlx5e_hv_vhca_stats_work() recomputes
buf_len using the new non-zero stats_nch and calls
memset(buf, 0, buf_len) on ZERO_SIZE_PTR, faulting at address 0x10.
Allocate the buffer based on priv->max_nch, which is set in
mlx5e_priv_init() and is the upper bound on stats_nch:
- Add a separate helper mlx5e_hv_vhca_stats_buf_max_size() that
returns sizeof(per_ring_stats) * max(max_nch, stats_nch), and
use it for the kvzalloc() in mlx5e_hv_vhca_stats_create().
- Keep mlx5e_hv_vhca_stats_buf_size() (which returns based on
stats_nch) for the worker's active payload size, so the wire
format (block->rings = stats_nch) and the amount of data filled
by mlx5e_hv_vhca_fill_stats() are unchanged.
The max(max_nch, stats_nch) guard handles the rare case where
mlx5e_attach_netdev() recomputes max_nch downward across a
detach/resume cycle while priv->stats_nch persists (mlx5e_detach_netdev
does not call mlx5e_priv_cleanup, so stats_nch is only reset when
the netdev is destroyed). Without the guard, the worker could compute
buf_len from stats_nch and overrun the smaller buffer allocated based
on the reduced max_nch.
Allocating a non-zero buffer also makes the kvzalloc() failure path in
mlx5e_hv_vhca_stats_create() reachable for the first time: it returns
early without (re)creating the agent. Clear
priv->stats_agent.{agent,buf} in mlx5e_hv_vhca_stats_destroy() after
freeing them, so that if a later create() bails out on this path, a
subsequent teardown does not double-free the stale agent/buffer left
from a previous enable/disable cycle.
This mirrors the existing mlx5e pattern of preallocating arrays of
size max_nch (e.g. priv->channel_stats) and lazily populating
entries up to stats_nch on demand.
Fixes: fa691d0c9c08 ("net/mlx5e: Allocate per-channel stats dynamically at first usage") Signed-off-by: Feng Liu <feliu@nvidia.com> Reviewed-by: Eran Ben Elisha <eranbe@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Link: https://patch.msgid.link/20260630115151.729219-2-tariqt@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Paolo Abeni [Fri, 3 Jul 2026 16:39:03 +0000 (18:39 +0200)]
Merge branch 'net-mlx5-lag-bug-fixes'
Tariq Toukan says:
====================
net/mlx5: LAG bug fixes
Three bug fixes by Shay in the mlx5 LAG subsystem.
Patch 1 fixes an off-by-one in the error rollback path of
mlx5_lag_create_single_fdb_filter(): the loop started from the
failed index i, potentially operating on uninitialized state or
double-tearing-down an entry that had already self-rolled-back.
The rollback should start from i - 1.
Patch 2 fixes a hang in mlx5_mpesw_work(): when
mlx5_lag_get_devcom_comp() returns NULL the function returned
early without calling complete(), blocking any caller waiting on
mpesww->comp indefinitely.
Patch 3 fixes a kernel crash during teardown when
mlx5_lag_get_dev_seq() returns an error because no device is
marked as master or the peer is no longer in the LAG. The peer
flow cleanup is now skipped instead of proceeding with a bad
pointer.
This series by Shay fixes three bugs in the mlx5 LAG subsystem.
Shay Drory [Tue, 30 Jun 2026 11:29:17 +0000 (14:29 +0300)]
net/mlx5e: TC, skip peer flow cleanup when LAG seq is unavailable
mlx5_lag_get_dev_seq() will return error when the peer isn't in the LAG
or when no device is marked as master. Result bad memory access and kernel
crash[1].
Hence, skip the peer when lookup fails.
Note: In case there are peer flows, they are cleaned before LAG cleared
the master mark.
Shay Drory [Tue, 30 Jun 2026 11:29:16 +0000 (14:29 +0300)]
net/mlx5: LAG, MPESW, Fix missing complete() on devcom error
mlx5_mpesw_work() returned without calling complete() when
mlx5_lag_get_devcom_comp() returned NULL. A caller that queued the
work and waited on mpesww->comp would block indefinitely.
Funnel the early-return path through a new "complete" label so the
waiter is always woken.
Fixes: b430c1b4f63b ("net/mlx5: Replace global mlx5_intf_lock with HCA devcom component lock") Signed-off-by: Shay Drory <shayd@nvidia.com> Reviewed-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Link: https://patch.msgid.link/20260630112917.698313-3-tariqt@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Shay Drory [Tue, 30 Jun 2026 11:29:15 +0000 (14:29 +0300)]
net/mlx5: LAG, Fix off-by-one in single-FDB error rollback
On failure at index i, the reverse cleanup loop in
mlx5_lag_create_single_fdb() starts from i, so the failed index
itself is rolled back. That can operate on uninitialized state or
double-tear-down a rule the add_one path already self-rolled-back.
Start the rollback from i - 1 so only successfully-installed entries
are undone.
Fixes: ddbb5ddc43ad ("net/mlx5: LAG, Refactor lag logic") Signed-off-by: Shay Drory <shayd@nvidia.com> Reviewed-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Link: https://patch.msgid.link/20260630112917.698313-2-tariqt@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
mac802154: remove interfaces with RCU list deletion
Queue wake, stop, and disable paths walk local->interfaces under RCU.
The bulk hardware teardown path removes entries with list_del(), so an
asynchronous transmit completion can follow a poisoned list node in
ieee802154_wake_queue().
Use list_del_rcu() as in the single-interface removal path. The following
unregister_netdevice() waits for in-flight RCU readers before freeing the
netdevice, so no separate grace-period wait is needed.
Jamal Hadi Salim [Tue, 30 Jun 2026 15:09:22 +0000 (11:09 -0400)]
net/sched: sch_teql: move rcu_read_lock()/spin_lock() from _bh variants
This is a followup based on sashiko comments [1] on commit e5b811fe7931
("net/sched: sch_teql: Introduce slaves_lock to avoid race condition and UAF")
Use plain rcu_read_lock()/spin_lock() in teql_master_xmit() instead of the
_bh variants, since ndo_start_xmit is already invoked with BH disabled
by the core stack and the _bh primitives can warn in_hardirq() when xmit
is reached through netpoll or a softirq xmit path with hard IRQs disabled.
Moves rcu_read_lock() after restart: label + adds rcu_read_unlock() before
goto restart (fixes the unbounded RCU hold across retries)
Revision 2 of the CONNMARK target accepts user-controlled shift
parameters and applies them to 32-bit mark values in
connmark_tg_shift().
A shift_bits value of 32 or more triggers an undefined-shift bug when
the rule is evaluated. Invalid shift_dir values are also accepted and
silently fall back to the left-shift path.
Reject invalid revision-2 shift parameters in connmark_tg_check() so
malformed rules fail at installation time, before they can reach the
packet path.
Fixes: 472a73e00757 ("netfilter: xt_conntrack: Support bit-shifting for CONNMARK & MARK targets.") Reported-by: Yuan Tan <yuantan098@gmail.com> Reported-by: Zhengchuan Liang <zcliangcn@gmail.com> Reported-by: Xin Liu <dstsmallbird@foxmail.com> Assisted-by: Codex:GPT-5.4 Signed-off-by: Wyatt Feng <bronzed_45_vested@icloud.com> Reviewed-by: Ren Wei <enjou1224z@gmail.com> Reviewed-by: Phil Sutter <phil@nwl.cc> Signed-off-by: Florian Westphal <fw@strlen.de>
ipvs: reset full ip_vs_seq structs in ip_vs_conn_new
Commit 9a05475cebdd ("ipvs: avoid kmem_cache_zalloc in
ip_vs_conn_new") changed ip_vs_conn_new() to allocate an ip_vs_conn
object with kmem_cache_alloc(). The function then initializes many
fields explicitly, but only resets in_seq.delta and out_seq.delta in the
two struct ip_vs_seq members.
That leaves init_seq and previous_delta uninitialized. This is normally
harmless while the corresponding IP_VS_CONN_F_IN_SEQ or
IP_VS_CONN_F_OUT_SEQ flag is clear. For connections learned from a sync
message, however, ip_vs_proc_conn() preserves those flags from
IP_VS_CONN_F_BACKUP_MASK and passes opt=NULL when the message omits
IPVS_OPT_SEQ_DATA. In that case the new connection can be hashed with
SEQ flags set but with the rest of in_seq/out_seq still containing stale
slab data.
When a packet for such a connection is later handled by an IPVS
application helper, vs_fix_seq() and vs_fix_ack_seq() use
previous_delta and init_seq to rewrite TCP sequence numbers. A malformed
sync message can therefore make forwarded packets carry stale slab bytes
in their TCP seq/ack numbers, and can also corrupt the forwarded TCP
flow.
Reset both struct ip_vs_seq members completely before publishing the
connection. This matches the existing "reset struct ip_vs_seq" comment
and keeps the sequence-adjustment gates inactive unless valid sequence
data is installed later.
Fixes: 9a05475cebdd ("ipvs: avoid kmem_cache_zalloc in ip_vs_conn_new") Cc: stable@vger.kernel.org Reported-by: Yizhou Zhao <zhaoyz24@mails.tsinghua.edu.cn> Reported-by: Yuxiang Yang <yangyx22@mails.tsinghua.edu.cn> Reported-by: Ao Wang <wangao@seu.edu.cn> Reported-by: Xuewei Feng <fengxw06@126.com> Reported-by: Qi Li <qli01@tsinghua.edu.cn> Reported-by: Ke Xu <xuke@tsinghua.edu.cn> Assisted-by: Claude-Code:GLM-5.2 Signed-off-by: Yizhou Zhao <zhaoyz24@mails.tsinghua.edu.cn> Signed-off-by: Florian Westphal <fw@strlen.de>
When an ICMP Fragmentation Needed error is received for a tunneled IPVS
connection, ip_vs_in_icmp() recomputes the MTU that the original packet
can use by subtracting the tunnel overhead from the reported next-hop
MTU.
The current code always subtracts sizeof(struct iphdr), which is only
the IPIP overhead. For GUE and GRE tunnels, ipvs_udp_decap() and
ipvs_gre_decap() already compute the additional tunnel header length,
but that value is scoped to the decapsulation block and is lost before
the ICMP_FRAG_NEEDED handling. As a result, the ICMP error sent back to
the client advertises an MTU that is too large, so PMTUD can fail to
converge for GUE/GRE-tunneled real servers.
With a reported next-hop MTU of 1400, a GUE tunnel currently returns
1380 to the client. The correct value is 1368:
Hoist the tunnel header length into the main ip_vs_in_icmp() scope and
subtract sizeof(struct iphdr) + ulen in the Fragmentation Needed path.
The IPIP path keeps ulen as 0, so its existing 1400 - 20 = 1380 result
is unchanged.
Fixes: 508f744c0de3 ("ipvs: strip udp tunnel headers from icmp errors") Cc: stable@vger.kernel.org Reported-by: Yizhou Zhao <zhaoyz24@mails.tsinghua.edu.cn> Reported-by: Yuxiang Yang <yangyx22@mails.tsinghua.edu.cn> Reported-by: Ao Wang <wangao@seu.edu.cn> Reported-by: Xuewei Feng <fengxw06@126.com> Reported-by: Qi Li <qli01@tsinghua.edu.cn> Reported-by: Ke Xu <xuke@tsinghua.edu.cn> Assisted-by: Claude-Code:GLM-5.2 Signed-off-by: Yizhou Zhao <zhaoyz24@mails.tsinghua.edu.cn> Acked-by: Julian Anastasov <ja@ssi.bg> Signed-off-by: Florian Westphal <fw@strlen.de>
netfilter: nft_set_rbtree: get command skips end element with open interval
The get command on intervals provide partial matches such as subranges
for usability reasons. However, an open interval has no closing end
element. If the closing element matches within the range of the open
internal, ie. its closest match is the start element of the open range,
then, return 0 but offer no matching element to userspace through
netlink as a special case. Userspace provides at least a matching start
element in this case and the closing end element matching the open
interal is ignored.
Another possibility is to report the matching start element of the open
interval for this end interval. However, this results in duplicated
matching being listed in userspace because userspace does not expect a
start element as response to a end element.
Fixes: 2aa34191f06f ("netfilter: nft_set_rbtree: use binary search array in get command") Reported-by: Melbin K Mathew <mlbnkm1@gmail.com> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Florian Westphal <fw@strlen.de>
netfilter: ip6tables: mark malformed IPv6 extension headers for hotdrop
The ah, hbh and rt matches check that the fixed extension header is
present, then use the header length field to derive the advertised
extension header length for matching.
For the ah match, add the missing advertised-length check. For hbh
and rt, update the existing advertised-length checks. In all three
cases, set hotdrop to true before returning false when the advertised
extension header length exceeds the available skb data.
Returning false treats the packet as a rule mismatch. Set hotdrop to
true and drop malformed packets so they cannot bypass rules intended
to drop packets with these IPv6 extension headers.
netfilter: nfnetlink_cthelper: cap to maximum number of expectation per master on updates
Really cap it to NF_CT_EXPECT_MAX_CNT (255) on updates.
The commit ("netfilter: nfnetlink_cthelper: cap to maximum number of
expectation per master") only covers creation of helpers, not updates.
Fixes: 397c8300972f ("netfilter: nf_conntrack_helper: cap maximum number of expectation at helper registration") Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Florian Westphal <fw@strlen.de>
Feng Wu [Thu, 25 Jun 2026 08:44:25 +0000 (08:44 +0000)]
netfilter: xt_rateest: fix u64 truncation in xt_rateest_mt()
On links faster than ~34 Gbps, where byte rate may exceed 2^32-1
(~ 4.3 GBps), the comparison result becomes incorrect because the
truncated value no longer reflects the actual estimator rate.
Wyatt Feng [Sun, 28 Jun 2026 08:05:54 +0000 (16:05 +0800)]
netfilter: xt_u32: reject invalid shift counts
u32_match_it() executes rule-supplied shift operands on a 32-bit
value. A malformed u32 rule can provide a shift count of 32 or more,
triggering an undefined shift out-of-bounds during packet evaluation.
Validate XT_U32_LEFTSH and XT_U32_RIGHTSH operands in
u32_mt_checkentry() and reject malformed rules before they reach the
packet path.
If a cloned or fragmented SKB is reallocated by skb_ensure_writable(), the
old data buffer is freed. However, nf_nat_sip() fails to update *dptr to
point to the new buffer.
It also appears to use nf_nat_mangle_udp_packet() on what could be a TCP
packet, which would overwrite the sequence number with a checksum update.
------------------------------------------------------------------------
nf_conntrack_sip linerizes skbs, hence no fragmented skb can be seen.
But clones are possible, so rebuild dptr.
Disable nf_nat_mangle_udp_packet() branch for TCP streams.
It doesn't look like this can ever happen, else we should have received
bug reports about this, so just check the conntrack is UDP and drop
otherwise.
The calling conntrack_sip set ->forced_dport for SIP_HDR_VIA_UDP messages,
so I don't think this is ever expected to be true for a TCP stream.
Fixes: 7266507d8999 ("netfilter: nf_ct_sip: support Cisco 7941/7945 IP phones") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-sonnet-4-6 Signed-off-by: Florian Westphal <fw@strlen.de>
GUE private flags can indicate that remote checksum offload metadata is
present. The private flags field itself is accounted for by
guehdr_flags_len(), but guehdr_priv_flags_len() currently returns 0 even
when GUE_PFLAG_REMCSUM is set.
This lets a packet with only the private flags field pass
validate_gue_flags(), after which gue_remcsum() and gue_gro_remcsum()
read the missing REMCSUM start/offset fields from the following bytes.
Account for GUE_PLEN_REMCSUM when GUE_PFLAG_REMCSUM is present so that
malformed packets are rejected during option validation.
Fixes: c1aa8347e73e ("gue: Protocol constants for remote checksum offload") Signed-off-by: Qihang <q.h.hack.winter@gmail.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Dawei Feng [Tue, 30 Jun 2026 07:16:25 +0000 (15:16 +0800)]
octeontx2-pf: fix SQB pointer leak on init failure
otx2_init_hw_resources() initializes SQ aura and pool resources before
several later setup steps. On failure, err_free_sq_ptrs only frees SQB
pages, leaving the per-SQ sqb_ptrs arrays behind.
Use otx2_free_sq_res() for the SQ unwind path and let it free sqb_ptrs
even when sq->sqe has not been allocated yet.
The bug was first flagged by an experimental analysis tool we are
developing for kernel memory-management bugs while analyzing
v6.13-rc1. The tool is still under development and is not yet publicly
available. Manual inspection confirms that the bug is still
present in v7.1.1.
An x86_64 allyesconfig build showed no new warnings. As we do not have an
OcteonTX2 PF device and the corresponding AF mailbox setup to test with,
no runtime testing was able to be performed.
Xiang Mei [Tue, 30 Jun 2026 04:51:21 +0000 (21:51 -0700)]
net: usb: net1080: validate packet_len before pad-byte access in rx_fixup
For an even packet_len, net1080_rx_fixup() reads the pad byte at
skb->data[packet_len] before the skb->len != packet_len check further
down, and packet_len is only bounded against NC_MAX_PACKET. A malicious
NetChip 1080 device can send a short frame advertising a large even
packet_len (e.g. 0x4000), so the pad-byte read lands past the end of the
skb:
BUG: KASAN: slab-out-of-bounds in net1080_rx_fixup
Read of size 1 at addr ffff8880106c83c6 by task ksoftirqd/0/14
...
net1080_rx_fixup (drivers/net/usb/net1080.c:384)
usbnet_bh (drivers/net/usb/usbnet.c:1589)
process_one_work (kernel/workqueue.c:3322)
bh_worker (kernel/workqueue.c:3708)
tasklet_action (kernel/softirq.c:965)
handle_softirqs (kernel/softirq.c:622)
...
Reject the frame when packet_len >= skb->len before reading.
Fixes: 904813cd8a0b ("[PATCH] USB: usbnet (4/9) module for net1080 cables") Reported-by: Weiming Shi <bestswngs@gmail.com> Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Xiang Mei <xmei5@asu.edu> Link: https://patch.msgid.link/20260630045121.1565324-1-xmei5@asu.edu Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Merge tag 'net-7.2-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Pull networking fixes from Paolo Abeni:
"Including fixes from netfilter and batman-adv.
Current release - new code bugs:
- netfilter: cthelper: cap to maximum number of expectation per master
Previous releases - regressions:
- netpoll: fix a use-after-free on shutdown path
- tcp: restore RCU grace period in tcp_ao_destroy_sock
- ipv6: fix NULL deref in fib6_walk_continiue() on multi-batch dump
- batman-adv: dat: ensure accessible eth_hdr proto field
- eth:
- virtio_net: disable cb when NAPI is busy-polled
- lan743x: Initialize eth_syslock spinlock before use
Previous releases - always broken:
- netfilter:
- nft_set_pipapo: don't leak bad clone into future transaction
- sched:
- sch_teql: Introduce slaves_lock to avoid race condition and UAF
- replace direct dequeue call with peek and qdisc_dequeue_peeked
- sctp: add INIT verification after cookie unpacking
- tipc: fix out-of-bounds read in broadcast Gap ACK blocks
- seg6: validate SRH length before reading fixed fields
- eth:
- mlx5e: fix use-after-free of metadata_dst on RX SC delete
- enetc: check the number of BDs needed for xdp_frame
- fbnic: don't cache shinfo across skb realloc"
* tag 'net-7.2-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (58 commits)
net/mlx5: HWS, fix matcher leak on resize target setup failure
net/sched: hhf: clear heavy-hitter state on reset
net/sched: dualpi2: clear stale classification on filter miss
net/sched: act_bpf: use rcu_dereference_bh() to read the filter
selftests: drv-net: tso: don't touch dangerous feature bits
cxgb4: Fix decode strings dump for T6 adapters
virtio_net: disable cb when NAPI is busy-polled
sctp: fix addr_wq_timer race in sctp_free_addr_wq()
selftests: net: bump default cmd() timeout to 20 seconds
bridge: stp: Fix a potential use-after-free when deleting a bridge
net/sched: sch_teql: Introduce slaves_lock to avoid race condition and UAF
net: gianfar: dispose irq mappings on probe failure and device removal
net: lan743x: Initialize eth_syslock spinlock before use
net: libwx: fix VMDQ mask for 1-queue mode
net: airoha: fix max receive size configuration
fsl/fman: Free init resources on KeyGen failure in fman_init()
netfilter: nftables: restrict checkum update offset
netfilter: nftables: restrict linklayer and network header writes
netfilter: nfnetlink_queue: restrict writes to network header
netfilter: nft_fib: reject fib expression on the netdev egress hook
...
Paolo Abeni [Thu, 2 Jul 2026 08:34:05 +0000 (10:34 +0200)]
Merge tag 'batadv-net-pullrequest-20260630' of https://git.open-mesh.org/batadv
Simon Wunderlich says:
====================
Here are some batman-adv bugfix, all by Sven Eckelmann:
- fix pointers after potential skb reallocs (5 patches)
- dat: ensure accessible eth_hdr proto field
* tag 'batadv-net-pullrequest-20260630' of https://git.open-mesh.org/batadv:
batman-adv: dat: ensure accessible eth_hdr proto field
batman-adv: bla: reacquire gw address after skb realloc
batman-adv: dat: acquire ARP hw source only after skb realloc
batman-adv: gw: acquire ethernet header only after skb realloc
batman-adv: access unicast_ttvn skb->data only after skb realloc
batman-adv: retrieve ethhdr after potential skb realloc on RX
====================
Dawei Feng [Mon, 29 Jun 2026 06:40:49 +0000 (14:40 +0800)]
net/mlx5: HWS, fix matcher leak on resize target setup failure
hws_bwc_matcher_move() allocates a replacement matcher before setting it
as the resize target. If mlx5hws_matcher_resize_set_target() fails, the
replacement matcher is not attached anywhere and is leaked.
Fix the leak by destroying the replacement matcher before returning from
the resize-target failure path.
The bug was first flagged by an experimental analysis tool we are
developing for kernel memory-management bugs while analyzing
v6.13-rc1. The tool is still under development and is not yet publicly
available. Manual inspection confirms that the bug is still
present in v7.1.1.
An x86_64 allyesconfig build showed no new warnings. As we do not have a
mlx5 HWS-capable device to test with, no runtime testing was able to be
performed.
Merge tag 'bootconfig-fixes-v7.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull bootconfig fix from Masami Hiramatsu:
- bootconfig: Fix NULL-pointer arithmetic
Fix undefined pointer arithmetic in xbc_snprint_cmdline() when
probing the buffer length with NULL and size 0. Track the written
length as a size_t instead to prevent build-time UBSan/FORTIFY_SOURCE
failures.
* tag 'bootconfig-fixes-v7.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
bootconfig: fix NULL-pointer arithmetic in xbc_snprint_cmdline()
Paolo Abeni [Wed, 1 Jul 2026 16:56:27 +0000 (18:56 +0200)]
Merge tag 'nf-26-06-30' of https://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf
Florian Westphal says:
====================
netfilter: updates for net
The following patchset contains Netfilter fixes for *net*.
Due to bug volume the plan is to make a second *net* pull request
this Friday.
1) Zero nf_conntrack_expect at allocation to prevent uninitialized data
leaks to userspace. Add missing exp->dir initialization.
2) Prevent out-of-bounds writes in nft_set_pipapo caused by inconsistent
clones during allocation failures. Fail operations if the clone enters an
error state. This was a day-0 bug.
3) Fix use-after-free race between ipset dump and array resizing. Protect
array pointer access with rcu_read_lock(). From Xiang Mei. Bug existed
since v4.20.
4) Validate skb_dst() exists before access in nf_conntrack_sip.
This Prevent crash when called from tc ingress or openvswitch.
From Pablo Neira Ayuso. Bug added in 4.3 when ovs gained support
for conntrack helpers.
5) Cap the maximum number of expectations to NF_CT_EXPECT_MAX_CNT during
userspace helper policy updates. Also from Pablo.
6) Prevent NULL pointer dereference in nft_fib on netdev egress hooks. Add
nft_fib_netdev_validate() to restrict fib expressions to appropriate
netdev hooks. Restrict nft_fib_validate() to IPv4, IPv6, and INET
protocols. From Theodor Arsenij Larionov-Trichkine.
Bug was exposed in v5.16 when egress hooks got added.
7) Restrict nfnetlink_queue writes to network headers. Validate IP/IPv6
header length and disable extension headers or IP option modifications.
Disable bridge modification for now, its unlikely anyone is using this.
8) Restrict arbitrary writes to link-layer and network headers in nftables.
Prevent link-layer modifications from spilling into network headers.
Prevent writes to IP version and length fields.
9) Restrict L3 checksum update offset to IPv4. Else csum offset can be
used to munge arbitrary header offsets, rendering the previous change moot.
These three patches are follow-ups to a 7.1 change that disabled
header rewrite ability in unprivileged network namespaces.
unprivileged netns support is not yet enabled again here.
netfilter pull request nf-26-06-30
* tag 'nf-26-06-30' of https://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf:
netfilter: nftables: restrict checkum update offset
netfilter: nftables: restrict linklayer and network header writes
netfilter: nfnetlink_queue: restrict writes to network header
netfilter: nft_fib: reject fib expression on the netdev egress hook
netfilter: nfnetlink_cthelper: cap to maximum number of expectation per master
netfilter: nf_conntrack_sip: validate skb_dst() before accessing it
netfilter: ipset: fix race between dump and ip_set_list resize
netfilter: nft_set_pipapo: don't leak bad clone into future transaction
netfilter: nf_conntrack_expect: zero at allocation time
====================
Samuel Moelius [Mon, 29 Jun 2026 16:44:59 +0000 (16:44 +0000)]
net/sched: hhf: clear heavy-hitter state on reset
HHF reset does not clear the classifier state used to identify heavy
hitters. Packets after reset can therefore be scheduled using flow
history from before the reset.
The reset operation should return the qdisc to an empty state.
Clear the heavy-hitter classifier tables when HHF is reset.
Fixes: 10239edf86f1 ("net-qdisc-hhf: Heavy-Hitter Filter (HHF) qdisc") Assisted-by: Codex:gpt-5.5-cyber-preview Signed-off-by: Samuel Moelius <sam.moelius@trailofbits.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Samuel Moelius [Sun, 28 Jun 2026 13:48:47 +0000 (13:48 +0000)]
net/sched: dualpi2: clear stale classification on filter miss
DualPI2 leaves previous classification state attached to an skb when
filter classification returns no match. The enqueue path can then act
on stale state from an earlier classification attempt.
A filter miss should fall back to the default class without reusing old
per-packet classification data.
Initialize the classification result to CLASSIC before running the
classifier. Explicit L4S, priority, and successful filter
classification can still override that default.
Fixes: 8f9516daedd6 ("sched: Add enqueue/dequeue of dualpi2 qdisc") Assisted-by: Codex:gpt-5.5-cyber-preview Signed-off-by: Samuel Moelius <sam.moelius@trailofbits.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Merge tag 'probes-fixes-v7.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull probes fixes from Masami Hiramatsu:
"fprobe fixes and spelling typos:
- Fix NULL pointer dereference in fprobe_fgraph_entry(). Prevent
general protection faults by checking shadow-stack reservation
bounds. Skip mid-flight registered fprobes that were not counted
during sizing.
eprobe: fix string pointer extraction
- Correct the casting of string pointers read from the ringbuffer to
prevent truncation of base event pointer variables when
dereferencing FILTER_PTR_STRING fields.
tracing/probes: clean up argument parsing and BTF helper logic
- Make the $ prefix mandatory for comm access: Require the $ prefix
for special fetcharg variables like $comm and $COMM, preventing
naming conflicts with regular BTF-based event fields.
- Fix double addition of offset for @+FOFFSET: Clear the temporary
offset variable after setting the FETCH_OP_FOFFS instruction to
avoid applying the offset multiple times.
- Remove WARN_ON_ONCE from parse_btf_arg: Prevent triggering a kernel
warning via user-space input when creating a kprobe event on a raw
address.
- Fix typo in a log message: Correct a spelling error ("$-valiable")
in trace probe log messages.
samples/trace_events: improve error checking
- Validate the thread pointer returned from kthread_run() in the
trace events sample code to properly handle thread creation
failures"
* tag 'probes-fixes-v7.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
tracing/probes: Make the $ prefix mandatory for comm access
tracing/fprobe: Fix NULL pointer dereference in fprobe_fgraph_entry()
tracing/probes: Fix double addition of offset for @+FOFFSET
tracing: eprobe: read the complete FILTER_PTR_STRING pointer
tracing/events: Fix to check the simple_tsk_fn creation
tracing/probes: Remove WARN_ON_ONCE from parse_btf_arg
tracing: probes: fix typo in a log message
query_nic_features() detects which offloads depend on tx-gso-partial
by enabling everything, turning tx-gso-partial off, and seeing which
active features drop out. Enabling all hw features is dangerous:
we may end up enabling rx-fcs and loopback for example. For the
ice driver we end up getting into problems with feature dependencies
so the cleanup isn't successful either, and the test exits with
rx-fcs and loopback enabled.
Scope the feature probing just to segmentation bits.
Fixes: 266b835e5e84 ("selftests: drv-net: tso: enable test cases based on hw_features") Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com> Reviewed-by: Daniel Zahka <daniel.zahka@gmail.com> Link: https://patch.msgid.link/20260629233923.2151144-1-kuba@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Gleb Markov [Mon, 29 Jun 2026 13:08:54 +0000 (16:08 +0300)]
cxgb4: Fix decode strings dump for T6 adapters
Depending on the value of chip_version, the correct decode set is selected.
However, the subsequent matching with the t4 encoding type in the if-else
block results in a reassignment, which leads to the loss of support for
t6_decode as well as reinitializing of values t4_decode and t5_decode.
The component history shows that the if-else block previously used for
this purpose, as well as the execution order, was not affected by the
change.
Furthermore, it is suggested by the execution order that the scenario with
overwriting and loss of support will be implemented.
Delete the if-else block.
Fixes: 6df397539cb0 ("cxgb4: Update correct encoding of SGE Ingress DMA States for T6 adapter") Signed-off-by: Gleb Markov <markov.gi@npc-ksb.ru> Reviewed-by: Potnuri Bharat Teja <bharat@chelsio.com> Link: https://patch.msgid.link/20260629130856.1168-1-markov.gi@npc-ksb.ru Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Longjun Tang [Mon, 29 Jun 2026 02:42:30 +0000 (10:42 +0800)]
virtio_net: disable cb when NAPI is busy-polled
When busy-poll is active, napi_schedule_prep() returns false in
virtqueue_napi_schedule(), so virtqueue_disable_cb() is skipped.
The device may keep firing irqs until reaches virtqueue_napi_complete().
Under load (received == budget), it will lead to a large number
of spurious interrupts.
Fix it by disabling the callback at the virtnet_poll() entry.
This keeps the callback off while we poll and it is re-enabled by
virtqueue_napi_complete() when going idle.
Fixes: ceef438d613f ("virtio_net: remove custom busy_poll") Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Longjun Tang <tanglongjun@kylinos.cn> Link: https://patch.msgid.link/20260629024230.37325-1-lange_tang@163.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Xin Long [Mon, 29 Jun 2026 18:31:14 +0000 (14:31 -0400)]
sctp: fix addr_wq_timer race in sctp_free_addr_wq()
sctp_free_addr_wq() previously removed addr_wq_timer using timer_delete()
while holding addr_wq_lock. However, timer_delete() does not guarantee that
a currently running timer handler has completed.
This allows a race with sctp_addr_wq_timeout_handler(), where the handler
may still run after addr_waitq has been freed, acquire addr_wq_lock, and
access freed memory, leading to a use-after-free.
Fix this by calling timer_shutdown_sync() before taking addr_wq_lock. This
guarantees that any in-flight timer handler has finished and prevents the
timer from being re-armed during teardown, making subsequent cleanup safe.
Jakub Kicinski [Mon, 29 Jun 2026 23:33:48 +0000 (16:33 -0700)]
selftests: net: bump default cmd() timeout to 20 seconds
We always used 5 sec as the default command timeout. But soon after
it was introduced, David effectively made us ignore the timeout
(it was passed to process.communicate() as the wrong argument).
Gal recently fixed that, but turns out the 5 sec is not enough
for a lot of tests and setups. The fix caused regressions.
In particular running reconfig commands (e.g. XDP attach) on mlx5
with 32 rings and 9k MTU, on a heavily-debug-enabled kernel takes
more than 5 sec. The XDP installation command will time out after
5 sec but since the sleeps in the kernel are non interruptible
the command finishes anyway, leaving the XDP program attached,
but with non-zero exit code. defer()ed cleanups are not installed,
breaking the environment for subsequent tests.
Since "install XDP" is a pretty normal command a "point fix"
does not seem appropriate. 32 rings is a fairly reasonable
config, too, so we should just increase the timeout to 20 sec.
There's no real reason behind the value of 20.
Fixes: 1cf270424218 ("net: selftest: add test for netdev netlink queue-get API") Fixes: f0bd19316663 ("selftests: net: fix timeout passed as positional argument to communicate()") Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com> Acked-by: Breno Leitao <leitao@debian.org> Reviewed-by: Nimrod Oren <noren@nvidia.com> Link: https://patch.msgid.link/20260629233348.2145841-1-kuba@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Breno Leitao [Fri, 26 Jun 2026 12:50:10 +0000 (05:50 -0700)]
bootconfig: fix NULL-pointer arithmetic in xbc_snprint_cmdline()
xbc_snprint_cmdline() is meant to be called twice: first with
buf=NULL, size=0 to probe the rendered length, then with a real
buffer to fill it (the standard snprintf() two-pass pattern). The
probe call makes the function compute "buf + size" (NULL + 0) and,
on every iteration, advance "buf += ret" from that NULL base and
pass the result back into snprintf().
Pointer arithmetic on a NULL pointer is undefined behavior. It is
harmless in the in-kernel callers today, but the follow-up patches
run this same code in the userspace tools/bootconfig parser at kernel
build time, where host UBSan / FORTIFY_SOURCE abort the build.
Track a running written length (size_t) instead of mutating @buf, and
only form "buf + len" when @buf is non-NULL. snprintf(NULL, 0, ...)
is itself well defined and returns the would-be length, so the
two-pass "probe then fill" usage returns identical byte counts.
tracing/probes: Make the $ prefix mandatory for comm access
Since $comm or $COMM are not event field but special fetcharg
variables to access current->comm, It should not be accessed
without '$' prefix even with typecast.
Sechang Lim [Wed, 24 Jun 2026 23:34:48 +0000 (08:34 +0900)]
tracing/fprobe: Fix NULL pointer dereference in fprobe_fgraph_entry()
fprobe_fgraph_entry() sizes a shadow-stack reservation in one walk of
the per-ip fprobe list and fills it in a second walk, both under
rcu_read_lock() only. A fprobe registered on an already-live ip can
become visible between the two walks, so the fill walk processes an
exit_handler the sizing walk did not count and used runs past
reserved_words. If the sizing walk counted nothing, fgraph_data is NULL
and the first write_fprobe_header() faults:
Oops: general protection fault, probably for non-canonical address ...
KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
RIP: 0010:fprobe_fgraph_entry+0xa38/0xf10 kernel/trace/fprobe.c:167
Call Trace:
<TASK>
function_graph_enter_regs+0x44c/0xa10 kernel/trace/fgraph.c:677
ftrace_graph_func+0xc5/0x140 arch/x86/kernel/ftrace.c:671
__kernel_text_address+0x9/0x40 kernel/extable.c:78
arch_stack_walk+0x117/0x170 arch/x86/kernel/stacktrace.c:26
kmem_cache_free+0x188/0x580 mm/slub.c:6378
tcp_data_queue+0x18d/0x6550 net/ipv4/tcp_input.c:5590
[...]
</TASK>
The list cannot be frozen across the two walks, so skip a node that does
not fit the reservation and count it as missed.
tracing/probes: Fix double addition of offset for @+FOFFSET
Since commit 533059281ee5 ("tracing: probeevent: Introduce new argument
fetching code") wrongly use @offset local variable during the parsing,
the offset value is added twice when dereferencing.
Reset the @offset after setting it in FETCH_OP_FOFFS.
Ido Schimmel [Mon, 29 Jun 2026 07:21:17 +0000 (10:21 +0300)]
bridge: stp: Fix a potential use-after-free when deleting a bridge
The three STP timers are not supposed to be armed while the bridge is
administratively down. They are synchronously deactivated when the
bridge is put administratively down and the various call sites check for
'IFF_UP' before arming them.
This check is missing from br_topology_change_detection() and it is
possible to engineer a situation in which the topology change timer is
armed while the bridge is administratively down, resulting in a
use-after-free [1] when the bridge is deleted.
Fix by adding the missing check and for good measures synchronously
shutdown the three timers when the bridge is deleted.
[1]
ODEBUG: free active (active state 0) object: ffff88811662b9b0 object type: timer_list hint: br_topology_change_timer_expired (net/bridge/br_stp_timer.c:120)
WARNING: lib/debugobjects.c:629 at debug_print_object+0x1bc/0x450, CPU#9: ip/359
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: Noam Rathaus <noamr@ssd-disclosure.com> Reported-by: Neil Young <contact@ssd-disclosure.com> Acked-by: Nikolay Aleksandrov <nikolay@nvidia.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: Breno Leitao <leitao@debian.org> Link: https://patch.msgid.link/20260629072117.497959-1-idosch@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Jamal Hadi Salim [Sun, 28 Jun 2026 11:12:29 +0000 (07:12 -0400)]
net/sched: sch_teql: Introduce slaves_lock to avoid race condition and UAF
The teql master->slaves singly linked list is not protected against
multiple writes. It can be mod'ed concurently from teql_master_xmit(),
teql_dequeue(), teql_init() and teql_destroy() without holding any list
lock or RCU protection.
zdi-disclosures@trendmicro.com has demonstrated that the qdisc is freed
after an RCU grace period, but teql_master_xmit() running on another
CPU can still hold a stale pointer into the list, resulting in a
slab-use-after-free:
BUG: KASAN: slab-use-after-free in teql_master_xmit+0xf0f/0x16b0
Read of size 8 at addr ffff888013fb0440 by task poc/332
Freed 512-byte region [ffff888013fb0400, ffff888013fb0600) (kmalloc-512)
The fix?
Add a per-master slaves_lock spinlock that serializes all mutations of
master->slaves and the NEXT_SLAVE() links in teql_destroy() and
teql_qdisc_init(). teql_master_xmit() also takes the same slaves_lock
around those updates.
Annotate master->slaves and the per-slave ->next pointer with __rcu and
use the appropriate RCU accessors everywhere they are touched:
rcu_assign_pointer() on the writer side (under slaves_lock),
rcu_dereference_protected() for the writer-side loads (also under
slaves_lock), rcu_dereference_bh() for the loads in teql_master_xmit() and
rtnl_dereference() for the loads in teql_master_open()/teql_master_mtu(),
which run under RTNL.
Pair this with rcu_read_lock_bh()/rcu_read_unlock_bh() around the list
traversal in teql_master_xmit(), so that readers either observe a fully
linked list or are deferred until the in-flight mutation completes. The two
early-return paths in teql_master_xmit() are updated to release the RCU-bh
read-side critical section before returning, since leaving it held would
disable BH on that CPU for good.
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: zdi-disclosures@trendmicro.com Tested-by: Victor Nogueira <victor@mojatatu.com> Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com> Link: https://patch.msgid.link/20260628111229.669751-1-jhs@mojatatu.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Rosen Penev [Fri, 26 Jun 2026 22:52:28 +0000 (15:52 -0700)]
net: gianfar: dispose irq mappings on probe failure and device removal
irq_of_parse_and_map() creates irqdomain mappings that should be
balanced with irq_dispose_mapping(). The driver never called
irq_dispose_mapping(), leaking mappings on probe failure and
device removal.
Fix by adding irq_dispose_mapping() in free_gfar_dev() and
expanding its loop from priv->num_grps to MAXGROUPS so the
error path also catches partially-initialized groups. All
irqinfo pointers are pre-initialized to NULL in gfar_of_init(),
making the NULL-guarded walk in free_gfar_dev() safe for every
scenario.
gfar_parse_group() itself is left as a simple parse function
with no resource management; cleanup is centralized in the
caller's error path.
Assisted-by: opencode:big-pickle Fixes: b31a1d8b4151 ("gianfar: Convert gianfar to an of_platform_driver") Signed-off-by: Rosen Penev <rosenp@gmail.com> Link: https://patch.msgid.link/20260626225228.427392-1-rosenp@gmail.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Andrea Righi [Fri, 26 Jun 2026 16:32:18 +0000 (18:32 +0200)]
net: lan743x: Initialize eth_syslock spinlock before use
lan743x_hardware_init() calls pci11x1x_strap_get_status() during the
PCI11x1x probe sequence. That helper acquires the Ethernet subsystem
hardware lock via lan743x_hs_syslock_acquire(), which relies on
adapter->eth_syslock_spinlock to serialize access.
The spinlock is currently initialized only after the strap status is
read. With CONFIG_DEBUG_SPINLOCK enabled, taking the zeroed initialized
spinlock can trip the spinlock debug check.
Fix by initializing adapter->eth_syslock_spinlock before reading the
strap status so the probe path never attempts to lock an uninitialized
spinlock.
Fixes: 46b777ad9a8c ("net: lan743x: Add support to SGMII 1G and 2.5G") Cc: stable@vger.kernel.org # v6.0+ Signed-off-by: Andrea Righi <arighi@nvidia.com> Reviewed-by: David Thompson <davthompson@nvidia.com> Reviewed-by: Thangaraj Samynathan<Thangaraj.s@microchip.com> Link: https://patch.msgid.link/20260626163218.3591486-1-arighi@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Jiawen Wu [Fri, 26 Jun 2026 09:25:30 +0000 (17:25 +0800)]
net: libwx: fix VMDQ mask for 1-queue mode
In wx_set_vmdq_queues(), the VMDQ mask was not set for the devices not
supporting WX_FLAG_MULTI_64_FUNC, i.e., NGBE devices. A mask of 0 causes
__ALIGN_MASK(1, ~vmdq->mask) to return 0, which incorrectly sets
q_per_pool to 0 in wx_write_qde().
Fix the VMDQ 1-queue mask to 0x7F then ensures that __ALIGN_MASK(1,
~0x7F) correctly evaluates to 1.
Lorenzo Bianconi [Thu, 25 Jun 2026 06:49:23 +0000 (08:49 +0200)]
net: airoha: fix max receive size configuration
Set the GDM maximum receive size to AIROHA_MAX_RX_SIZE unconditionally
during hardware initialization instead of updating it according to the
configured MTU. This avoids dropping incoming frames that exceed the
current MTU but could still be processed by the networking stack, which
is able to fragment the reply on the TX side (e.g. ICMP echo requests).
Move the per-port MTU configuration to the PPE egress path where it
belongs, and set the tx frame size running airoha_ppe_set_xmit_frame_size()
to dynamically track the maximum MTU across running interfaces sharing
the same PPE instance.
Fix the PPE MTU register addressing to pack two port entries per
register word and add WAN_MTU0 configuration for non-LAN GDM devices.
Fixes: 54d989d58d2a ("net: airoha: Move min/max packet len configuration in airoha_dev_open()") Tested-by: Madhur Agrawal <madhur.agrawal@airoha.com> Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> Link: https://patch.msgid.link/20260625-airoha-fix-rx-max-len-v1-1-45b9b827358d@kernel.org Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Haoxiang Li [Thu, 25 Jun 2026 00:48:34 +0000 (08:48 +0800)]
fsl/fman: Free init resources on KeyGen failure in fman_init()
fman_muram_alloc() allocates initialization resources before
initializing the KeyGen block. If keygen_init() fails, the
function returns -EINVAL directly and leaves those resources
allocated. Free the initialization resources before returning
from the KeyGen failure path.
After previous patch, writes to network header are restricted.
However, there is another way to manipulate the l3 header: The
checksum update function.
Restrict this for network header writes, only the ipv4 header is
allowed. This needs run-time checks because BRIDGE, INET, NETDEV
families can carry l3 headers other than IP.
checksum updates to the udp/tcp (l4) headers are not restricted.
netfilter: nftables: restrict linklayer and network header writes
Don't permit arbitrary writes to linklayer and network header data.
Several spots in network stack trust header validation performed in
ipv4/ipv6 before PRE_ROUTING hook.
For linklayer, allow writes for netdev ingress. For other hooks, only
allow link layer writes that do not spill into network header.
For network header, check the offset/length combinations:
- changing dscp requires store at offset 0 for checsum fixups, so
make sure ip version + length field isn't altered.
- ip6 dscp starts directly after the version field, so make sure it
remains 6.
Several of these checks could already be done at rule insertion time.
Risk is that this might cause ruleset load failures for existing
rulesets. With this change such writes are silently skipped and packet
passes unchanged.
Transport and inner header bases are not checked / restricted.
netfilter: nfnetlink_queue: restrict writes to network header
nfnetlink_queue doesn't allow selective replacements of some part of the
payload, only complete replacement.
If the new data is shorter, skb is trimmed, otherwise expanded.
Add minimal validation of the new ip/ipv6 header. Check total len
matches skb length. Disallow ip option modifications.
IPv6 extension headers are also disabled.
IP options and exthdrs could be allowed later after validation pass or
ip option recompile.
Transport header is not checked.
Bridge modifications are rejected. Given userspace doesn't even receive
L2 headers, use is limited and I don't think there are any users of
bridge nfnetlink_queue, let alone users that modifiy payload.
netfilter: nft_fib: reject fib expression on the netdev egress hook
A fib expression in a netdev egress base chain dereferences nft_in(pkt),
NULL on the transmit path, causing a NULL pointer dereference at eval.
nft_fib_validate() masks the hook with NF_INET_* values, but netdev hook
numbers are a separate enum that aliases them (NF_NETDEV_EGRESS ==
NF_INET_LOCAL_IN), so an egress chain passes validation and then faults.
Add nft_fib_netdev_validate() that limits each result/flag to the netdev
hook where the device it reads exists: the input-device cases (OIF,
OIFNAME, ADDRTYPE with F_IIF) to ingress, the output-device case (ADDRTYPE
with F_OIF) to egress, ADDRTYPE with no device flag to both. Also restrict
nft_fib_validate() to NFPROTO_IPV4/IPV6/INET so its NF_INET_* masks are
not applied to another family's hooks.
netfilter: nfnetlink_cthelper: cap to maximum number of expectation per master
If userspace helper policy updates sets maximum number of expectation to
zero, cap it to NF_CT_EXPECT_MAX_CNT (255) on updates too.
Fixes: 397c8300972f ("netfilter: nf_conntrack_helper: cap maximum number of expectation at helper registration") Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: Florian Westphal <fw@strlen.de>
netfilter: nf_conntrack_sip: validate skb_dst() before accessing it
tc ingress and openvswitch do not guarantee routing information to be
available. These subsystems use the conntrack helper infrastructure, and
the SIP helper relies on the skb_dst() to be present if
sip_external_media is set to 1 (which is disabled by default as a module
parameter).
This effectively disables the sip_external_media toggle for these
subsystems without resulting in a crash.
Xiang Mei [Thu, 25 Jun 2026 01:00:06 +0000 (18:00 -0700)]
netfilter: ipset: fix race between dump and ip_set_list resize
The release path of ip_set_dump_do() and ip_set_dump_done() read
inst->ip_set_list via ip_set_ref_netlink(), a plain rcu_dereference_raw()
of the array pointer. These run from netlink_recvmsg() without the nfnl
mutex and without an RCU read-side critical section.
A concurrent ip_set_create() can grow the array: it publishes the new
array, calls synchronize_net() and then kvfree()s the old one. Since the
dump paths read the array outside any RCU reader, synchronize_net() does
not wait for them and the old array can be freed while they still index
into it, causing a use-after-free.
The dumped set itself stays pinned via set->ref_netlink, so only the
array load needs protecting. Take rcu_read_lock() around it, matching
ip_set_get_byname() and __ip_set_put_byindex().
BUG: KASAN: slab-use-after-free in ip_set_dump_do (net/netfilter/ipset/ip_set_core.c:1697)
Read of size 8 at addr ffff88800b5c4018 by task exploit/150
Call Trace:
...
kasan_report (mm/kasan/report.c:595)
ip_set_dump_do (net/netfilter/ipset/ip_set_core.c:1697)
netlink_dump (net/netlink/af_netlink.c:2325)
netlink_recvmsg (net/netlink/af_netlink.c:1976)
sock_recvmsg (net/socket.c:1159)
__sys_recvfrom (net/socket.c:2315)
...
Oops: general protection fault, probably for non-canonical address ... KASAN NOPTI
KASAN: maybe wild-memory-access in range [0x02d6...d0-0x02d6...d7]
RIP: 0010:ip_set_dump_do (net/netfilter/ipset/ip_set_core.c:1698)
Kernel panic - not syncing: Fatal exception
Fixes: 8a02bdd50b2e ("netfilter: ipset: Fix calling ip_set() macro at dumping") Cc: stable@vger.kernel.org Reported-by: Weiming Shi <bestswngs@gmail.com> Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Xiang Mei <xmei5@asu.edu> Acked-by: Jozsef Kadlecsik <kadlec@netfilter.org> Signed-off-by: Florian Westphal <fw@strlen.de>
Florian Westphal [Tue, 16 Jun 2026 11:26:26 +0000 (13:26 +0200)]
netfilter: nft_set_pipapo: don't leak bad clone into future transaction
On memory allocation failure the cloned nft_pipapo_match can enter a bad
state:
- some fields can have their lookup tables resized while others did
not
- bits might have been toggled
- scratch map can be undersized which also means m->bsize_max can be
lower than what is required
This means that the next insertion in the same batch can trigger
out-of-bounds writes.
Furthermore, a failure in the first can result in the bad clone to
leak into the next transaction because the abort callback is never
executed in this case (the upper layer saw an error and no attempt to
allocate a transactional request was made).
Record a state for the nft_pipapo_match structure:
- NEW (pristine clone)
- MOD (modified clone with good state)
- ERR (potentially bogus content)
Then make it so that deletes and insertions fail when the clone
entered ERR state.
In case the very first insert attempt results in an error, free the
clone right away.
Fixes: 3c4287f62044 ("nf_tables: Add set type for arbitrary concatenation of ranges") Cc: stable@vger.kernel.org Reported-and-tested-by: Seesee <cjc000013@gmail.com> Reviewed-by: Stefano Brivio <sbrivio@redhat.com> Signed-off-by: Florian Westphal <fw@strlen.de>
Florian Westphal [Wed, 24 Jun 2026 21:06:43 +0000 (23:06 +0200)]
netfilter: nf_conntrack_expect: zero at allocation time
There are occasional LLM hints wrt. leaking uninitialized data to
userspace via ctnetlink. Just zero at allocation time,
expectations are not frequently used these days.
Intentionally keeps _init as-is because we could theoretically
support re-init, so add the missing exp->dir there.
v4 tried to fix the RollBall regression from 8fe125892f40 by deferring the
bridge probe to PHY discovery time. Maxime Chevallier and Aleksander
Bajkowski both tested that on genuine RollBall hardware and confirmed it
does not restore PHY detection (the module still is not ready when the
probe runs), and the Sashiko static review flagged the same path.
So this version drops the deferred-probe patch and instead reverts 8fe125892f40, restoring the pre-regression behaviour for genuine RollBall
modules. A proper fix for slow-initializing modules needs per-module init
timing (a longer module_t_wait / a per-module quirk) and genuine RollBall
hardware to validate; that is better owned as a follow-up by someone with
such a module.
====================
Petr Wozniak [Sat, 27 Jun 2026 17:32:42 +0000 (19:32 +0200)]
Revert "net: phy: sfp: probe for RollBall I2C-to-MDIO bridge in mdio-i2c"
This reverts commit 8fe125892f40 ("net: phy: sfp: probe for RollBall
I2C-to-MDIO bridge in mdio-i2c").
That commit added a RollBall bridge probe at MDIO bus creation time, in
i2c_mii_init_rollball(), to avoid a multi-minute PHY probe retry loop on
modules without a bridge (e.g. RTL8261BE). The probe runs in SFP_S_INIT,
before genuine RollBall modules have finished their firmware/bridge
initialization, so the bridge does not yet answer CMD_READ/CMD_DONE. The
probe times out, mdio_protocol is set to MDIO_I2C_NONE, and PHY detection
is then skipped for genuine RollBall modules that worked before the commit.
This was confirmed on hardware by Maxime Chevallier and Aleksander
Bajkowski: their RollBall modules no longer detect a PHY, and work again
on v7.0 (before the bridge probing was introduced). The Sashiko static
review flagged the same path.
Deferring the probe to PHY discovery time does not fix it either: at that
point a slow module may still be initializing, so the probe still returns
-ENODEV. A proper fix needs per-module init timing (a longer module_t_wait
or a per-module quirk, per SFF-8472 the host must also wait at least 300 ms
after insertion), which requires genuine RollBall hardware to develop and
validate. Revert to restore the previous, working behaviour in the meantime.
The RTL8261BE retry-loop latency that the reverted commit addressed is
handled in our downstream tree, so reverting upstream is safe on our side.
Petr Wozniak [Sat, 27 Jun 2026 17:32:41 +0000 (19:32 +0200)]
net: phy: sfp: free mii_bus in sfp_i2c_mdiobus_destroy
sfp_i2c_mdiobus_create() allocates the I2C MDIO bus with mdio_i2c_alloc(),
a plain (non-devm) allocation, and registers it. sfp_i2c_mdiobus_destroy()
only unregisters the bus and clears sfp->i2c_mii without calling
mdiobus_free(). As the only reference to the bus is then cleared, the
struct mii_bus is leaked.
This is hit whenever a copper/RollBall SFP module that instantiated an MDIO
bus is removed: sfp_sm_main() takes the global teardown path and calls
sfp_i2c_mdiobus_destroy(). sfp_cleanup(), on driver unbind, frees
sfp->i2c_mii directly, which is why the leak only triggered on module
hot-removal and not on unbind.
Free the bus in sfp_i2c_mdiobus_destroy() to match the allocation done in
sfp_i2c_mdiobus_create().
Fixes: e85b1347ace6 ("net: sfp: create/destroy I2C mdiobus before PHY probe/after PHY release") Signed-off-by: Petr Wozniak <petr.wozniak@gmail.com> Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com> Reviewed-by: Larysa Zaremba <larysa.zaremba@intel.com> Link: https://patch.msgid.link/312bde8176fc429aa89524e3be250137f034ba84.1782581445.git.petr.wozniak@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Xiang Mei [Sat, 27 Jun 2026 20:53:53 +0000 (13:53 -0700)]
usbnet: gl620a: fix out-of-bounds read in genelink_rx_fixup()
genelink_rx_fixup() splits an aggregated RX frame into its individual
packets, using a per-packet length taken from device-supplied data. That
length is only bounded by GL_MAX_PACKET_LEN (1514); it is never compared
against how many bytes were actually received.
A malicious GeneLink (GL620A) device can therefore send a short URB whose
header claims packet_count > 1 and a first packet of up to 1514 bytes.
skb_put_data(gl_skb, packet->packet_data, size);
then copies past the end of the receive buffer and hands the adjacent slab
contents up the network stack, an out-of-bounds read that leaks kernel heap.
No privilege is required: the path runs in the usbnet RX softirq as soon as
the interface is up.
BUG: KASAN: slab-out-of-bounds in genelink_rx_fixup (drivers/net/usb/gl620a.c:112)
Read of size 1514 at addr ffff888011309708 by task ksoftirqd/0/14
Call Trace:
...
__asan_memcpy (mm/kasan/shadow.c:105)
genelink_rx_fixup (include/linux/skbuff.h:2814 drivers/net/usb/gl620a.c:112)
usbnet_bh (drivers/net/usb/usbnet.c:572 drivers/net/usb/usbnet.c:1589)
process_one_work (kernel/workqueue.c:3322)
bh_worker (kernel/workqueue.c:3405)
tasklet_action (kernel/softirq.c:965)
handle_softirqs (kernel/softirq.c:622)
run_ksoftirqd (kernel/softirq.c:1076)
...
skb_pull() already verifies that the requested length fits the buffer and
returns NULL otherwise. Move it ahead of the copy and check its result, so
a packet that overruns the received data is rejected before it is read.
Well-formed frames, whose packets are fully present, are unaffected.
Fixes: 47ee3051c856 ("[PATCH] USB: usbnet (5/9) module for genesys gl620a cables") Reported-by: Weiming Shi <bestswngs@gmail.com> Signed-off-by: Xiang Mei <xmei5@asu.edu> Link: https://patch.msgid.link/20260627205353.4000788-1-xmei5@asu.edu Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Doruk Tan Ozturk [Sat, 27 Jun 2026 22:30:59 +0000 (00:30 +0200)]
net/mlx5e: macsec: fix use-after-free of metadata_dst on RX SC delete
When an offloaded MACsec RX SC is deleted, macsec_del_rxsc_ctx() freed
the per-SC metadata_dst with metadata_dst_free(), which kfree()s the
object unconditionally and ignores the dst reference count. The RX
datapath in mlx5e_macsec_offload_handle_rx_skb() looks up the SC under
rcu_read_lock() via xa_load(), takes a reference with dst_hold() and
attaches the dst to the skb with skb_dst_set(). A reader that already
obtained the rx_sc pointer can race with the delete path and operate on
freed memory.
Fix the owner side by dropping the reference with dst_release() instead
of freeing unconditionally, and convert the RX datapath to
dst_hold_safe() so a reader racing the SC delete cannot attach a dst
whose last reference was just dropped; only attach it when a reference
was actually taken.
mlx5e_macsec_add_rxsc() also published sc_xarray_element via xa_alloc()
before rx_sc->md_dst was allocated and initialised, so a datapath reader
that looked the SC up by fs_id could observe rx_sc with md_dst still
NULL or, on weakly-ordered architectures, a non-NULL md_dst pointer
whose contents were not yet visible. NULL-check the xa_load() result and
md_dst on the datapath, and reorder add_rxsc() so the xa_alloc() publish
happens only after md_dst is fully initialised; the xarray RCU publish
then pairs with the rcu_read_lock()/xa_load() in the datapath.
Note: macsec_del_rxsc_ctx() also kfree()s rx_sc->sc_xarray_element
without an RCU grace period while the same datapath reads it under
rcu_read_lock(); that is a separate pre-existing issue left to a
follow-up patch.
Found by 0sec automated security-research tooling (https://0sec.ai).
Fixes: b7c9400cbc48 ("net/mlx5e: Implement MACsec Rx data path using MACsec skb_metadata_dst") Cc: stable@vger.kernel.org Signed-off-by: Doruk Tan Ozturk <doruk@0sec.ai> Reviewed-by: Tariq Toukan <tariqt@nvidia.com> Link: https://patch.msgid.link/20260627223059.29917-1-doruk@0sec.ai Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Yousef Alhouseen [Sun, 28 Jun 2026 00:28:04 +0000 (02:28 +0200)]
netdevsim: remove ethtool debugfs files before freeing netdev
The ethtool debugfs files point directly into struct netdevsim, which is
allocated as net_device private data. Their containing port directory is
removed only after nsim_destroy() calls free_netdev().
An open simple-attribute file can consequently dereference the freed
private data before the directory is removed. KASAN observed this in
debugfs_u32_get() during network namespace teardown.
Track and remove the ethtool subtree before free_netdev() on both the
normal and registration-failure paths. debugfs removal drains active
file users before returning.
Fixes: ff1f7c17fb20 ("netdevsim: add pause frame stats") Reported-by: syzbot+6c25f4750230faf70be9@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=6c25f4750230faf70be9 Cc: <stable+noautosel@kernel.org> # netdevsim is a test harness, it's never loaded on production systems Signed-off-by: Yousef Alhouseen <alhouseenyousef@gmail.com> Link: https://patch.msgid.link/20260628002804.24214-1-alhouseenyousef@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Pengfei Zhang [Thu, 25 Jun 2026 07:05:17 +0000 (15:05 +0800)]
ipv6: fib6: fix NULL deref in fib6_walk_continue() on multi-batch dump
inet6_dump_fib() saves its progress in cb->args[1] as a positional
index within the current hash chain. Between batches, a concurrent
fib6_new_table() can insert a new table at the chain head, shifting
all existing entries. The saved index then lands on a different
table, causing fib6_dump_table() to set w->root to the wrong table
while w->node still points into the previous one.
fib6_walk_continue() dereferences w->node->parent (NULL) and panics:
Fix by storing tb->tb6_id in cb->args[1] instead of a positional
index. On resume, skip entries until the id matches; a concurrent
head-insert can never match the saved id, so the walker always
resumes on the correct table.
Fixes: 1b43af5480c3 ("[IPV6]: Increase number of possible routing tables to 2^32") Signed-off-by: Pengfei Zhang <zhangfeionline@gmail.com> Reviewed-by: Ido Schimmel <idosch@nvidia.com> Link: https://patch.msgid.link/20260625070517.965597-1-zhangfeionline@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Dmitry Safonov [Thu, 25 Jun 2026 18:21:41 +0000 (19:21 +0100)]
tcp: Decrement tcp_md5_needed static branch
In case of early freeing an unwanted TCP-MD5 key on TCP-AO connect(),
md5sig_info is freed right away (and set to NULL). Later, at
the moment of socket destruction, the static branch counter
is not getting decremented.
Add a missing decrement for TCP-MD5 static branch.
Reported-by: Qihang <q.h.hack.winter@gmail.com> Fixes: 0aadc73995d0 ("net/tcp: Prevent TCP-MD5 with TCP-AO being set") Cc: stable@vger.kernel.org Signed-off-by: Dmitry Safonov <0x7f454c46@gmail.com> Link: https://patch.msgid.link/20260625-tcp-md5-connect-v3-3-1fd313d6c1e0@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
tcp: defer md5sig_info kfree past RCU grace period in tcp_connect
The md5+ao reconciliation in tcp_connect() (net/ipv4/tcp_output.c)
has two symmetric branches:
if (needs_md5) {
tcp_ao_destroy_sock(sk, false);
} else if (needs_ao) {
tcp_clear_md5_list(sk);
kfree(rcu_replace_pointer(tp->md5sig_info, NULL, ...));
}
Both branches free a per-socket auth-info object while the socket is
in TCP_SYN_SENT and is already on the inet ehash (inserted by
inet_hash_connect() in tcp_v4_connect()). Both branches are reachable
by softirq RX-path readers that load the corresponding info pointer
via implicit RCU before bh_lock_sock_nested() is taken.
The needs_md5 branch is fixed in the prior patch by re-introducing
the call_rcu() free in tcp_ao_destroy_sock(): the equivalent per-key
loop runs inside tcp_ao_info_free_rcu(), the RCU callback, so by the
time it frees each tcp_ao_key all softirq readers that captured the
container have already completed rcu_read_unlock().
The needs_ao branch is not symmetric in the same way. The container
free can be deferred via kfree_rcu(md5sig, rcu) -- struct
tcp_md5sig_info already has the required rcu member
(include/net/tcp.h:1999-2002), and the rest of the tree already does
this in the tcp_md5sig_info_add() rollback paths
(net/ipv4/tcp_ipv4.c:1410, 1436). But the per-key teardown is done
by tcp_clear_md5_list() in process context BEFORE the container's
RCU grace period: it walks &md5sig->head and frees each
tcp_md5sig_key with bare hlist_del + kfree. A concurrent softirq
reader in __tcp_md5_do_lookup() / __tcp_md5_do_lookup_exact()
(tcp_ipv4.c:1253, 1298) walks the same list via
hlist_for_each_entry_rcu() and races with that bare kfree on the
keys themselves -- a per-key slab use-after-free of the same class
as the TCP-AO bug, on the same race window.
Fix this in two halves:
1. Convert the bare kfree() in tcp_connect() to kfree_rcu() so the
md5sig_info container joins the rest of the md5sig lifecycle.
The local-variable lift is mechanical and required because
kfree_rcu() is a macro that expects an lvalue.
2. Make tcp_clear_md5_list() RCU-safe by replacing hlist_del +
kfree(key) with hlist_del_rcu + kfree_rcu(key, rcu). struct
tcp_md5sig_key already carries the rcu member
(include/net/tcp.h:1995) and tcp_md5_do_del()
(net/ipv4/tcp_ipv4.c:1456) already uses kfree_rcu, so this
restores the lifecycle invariant the rest of the file follows
rather than introducing a one-off.
The other caller of tcp_clear_md5_list() is tcp_md5_destruct_sock()
(net/ipv4/tcp.c:412), which runs from the sock destructor when the
socket is already unhashed and unreachable; the extra grace period
there is unnecessary but harmless. Making the helper unconditionally
RCU-safe is the cleaner contract.
The needs_ao branch is not reachable by the userns reproducer used
to demonstrate the AO-side splat (the repro installs both keys but
ends up in the needs_md5 branch because the connect peer matches
the MD5 key, not the AO key); however the symmetric race exists
and a maintainer touching this code should not have to think about
which branch escapes RCU and which one does not.
Fixes: 51e547e8c89c ("tcp: Free TCP-AO/TCP-MD5 info/keys without RCU") Cc: stable@vger.kernel.org # v6.18+ Suggested-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Michael Bommarito <michael.bommarito@gmail.com> Reviewed-by: Dmitry Safonov <dima@arista.com> Reviewed-by: Eric Dumazet <edumazet@google.com>
[also credits to Qihang, who found that this races with tcp-diag] Reported-by: Qihang <q.h.hack.winter@gmail.com> Signed-off-by: Dmitry Safonov <0x7f454c46@gmail.com> Link: https://patch.msgid.link/20260625-tcp-md5-connect-v3-2-1fd313d6c1e0@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
tcp: restore RCU grace period in tcp_ao_destroy_sock
Commit 51e547e8c89c ("tcp: Free TCP-AO/TCP-MD5 info/keys without RCU")
removed the call_rcu() callback from tcp_ao_destroy_sock(), arguing that
"the destruction of info/keys is delayed until the socket destructor"
and therefore "no one can discover it anymore".
That argument does not hold for the call site in tcp_connect()
(net/ipv4/tcp_output.c:4327-4332). At that point the socket is in
TCP_SYN_SENT, has already been inserted into the inet ehash by
inet_hash_connect() in tcp_v4_connect(), and is therefore very much
discoverable: any softirq running tcp_v4_rcv() on another CPU can take
the socket out of the ehash, walk into tcp_inbound_hash(), and load
tp->ao_info via implicit RCU before bh_lock_sock_nested() is taken on
the destroying CPU.
The reader path then enters __tcp_ao_do_lookup() (net/ipv4/tcp_ao.c:208)
which re-loads tp->ao_info via rcu_dereference_check(); the re-load can
still observe the (about-to-be-freed) pointer because there is no
synchronize_rcu() between rcu_assign_pointer(tp->ao_info, NULL) and
tcp_ao_info_free() in tcp_ao_destroy_sock(). The captured pointer is
then walked at line 223:
The writer's synchronous kfree() is free to complete between the line
218 re-fetch and the line 223 hlist iteration. The slab is reused
(or simply LIST_POISON1-stamped if not yet reused) and the iteration
walks attacker-controlled or poison memory in softirq context.
Reproducer (no debug shim, stock x86_64 v7.1-rc2 SMP+KASAN, QEMU+KVM):
an unprivileged uid=1000 process inside CLONE_NEWUSER|CLONE_NEWNET
installs TCP_MD5SIG + TCP_AO_ADD_KEY on a TCP socket, sprays forged
TCP-AO segments toward its eventual 4-tuple via raw sockets, then
calls connect(). The md5-wins reconciliation in tcp_connect() fires
tcp_ao_destroy_sock(); the softirq backlog reader on the loopback
NAPI path crashes on the freed ao->head.first walk:
Oops: general protection fault, probably for non-canonical
address 0xfbd59c000000002f
KASAN: maybe wild-memory-access in range
[0xdead000000000178-0xdead00000000017f]
CPU: 0 UID: 1000 PID: 100 Comm: repro_userns
RIP: 0010:__tcp_ao_do_lookup+0x107/0x1c0
Call Trace: <IRQ>
__tcp_ao_do_lookup+0x107/0x1c0
tcp_ao_inbound_lookup.constprop.0+0x12a/0x200
tcp_inbound_ao_hash+0x5ea/0x1520
tcp_inbound_hash+0x7ce/0x1240
tcp_v4_rcv+0x1e7a/0x3e10
...
Restore the RCU grace period: re-add struct rcu_head to tcp_ao_info
and replace the synchronous tcp_ao_info_free() with a call_rcu()
callback. Readers that captured tp->ao_info before rcu_assign_pointer
NULLed it now see the object remain valid until rcu_read_unlock().
With the patch applied the reproducer runs cleanly for 2000 iterations
on the same kernel build.
Fixes: 51e547e8c89c ("tcp: Free TCP-AO/TCP-MD5 info/keys without RCU") Cc: stable@vger.kernel.org # v6.18+ Reviewed-by: Dmitry Safonov <dima@arista.com> Signed-off-by: Michael Bommarito <michael.bommarito@gmail.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Dmitry Safonov <0x7f454c46@gmail.com> Link: https://patch.msgid.link/20260625-tcp-md5-connect-v3-1-1fd313d6c1e0@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Maoyi Xie [Thu, 25 Jun 2026 06:17:28 +0000 (14:17 +0800)]
net: wwan: iosm: bound device offsets in the MUX downlink decoder
mux_dl_adb_decode() walks a chain of aggregated datagram tables using
offsets and lengths taken from the modem. first_table_index,
next_table_index, table_length, datagram_index and datagram_length are
all device supplied le values. Only first_table_index was checked, and
only for being non zero. The decoder then formed adth = block +
adth_index and read the table header and the datagram entries with no
bound against the received skb. A modem that reports an index or a
length past the downlink buffer makes the decoder read out of bounds.
The buffer is IPC_MEM_MAX_DL_MUX_LITE_BUF_SIZE and skb->len is at most
that, so skb->len is the real limit, but none of these in band offsets
were checked against it.
The table chain is also followed with no forward progress check. The loop
takes the next table from adth->next_table_index and stops only when that
reaches zero. A modem can stage two tables that point at each other, so
the loop never ends. It runs in softirq and clones the skb on every pass.
Validate every device offset and length against skb->len before use.
The block header must fit. Each table header, on entry and after every
next_table_index, must lie inside the skb. The datagram table must fit.
Each datagram index and length must stay inside the skb. The header
padding must not exceed the datagram length so the receive length does
not wrap. Require each next_table_index to move forward so the chain
cannot cycle.
This was reproduced under KASAN as a slab out of bounds read on a normal
downlink receive once the iosm net device is up.
projects / linux / log