commit cedff600f1 upstream.
FSCTL_DUPLICATE_EXTENTS_TO_FILE passes the source file directly to
vfs_clone_file_range() or vfs_copy_file_range() without checking the SMB
access mask granted to the source handle. A handle opened with attribute
access can consequently be used to copy file contents into an
attacker-readable destination.
Require FILE_READ_DATA on the source handle before either VFS operation,
matching other ksmbd data-copy paths.
Cc: stable@vger.kernel.org
Reported-by: Musaab Khan <musaab.khan@protonmail.com>
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit d20d1c8ba5 upstream.
When a blocking byte-range lock request is deferred in the
FILE_LOCK_DEFERRED path, ksmbd registers the asynchronous work into
the connection's async_requests list via setup_async_work(). The cancel
callback smb2_remove_blocked_lock() holds a reference to the flock.
If the lock waiter is subsequently woken up but the work state is no
longer KSMBD_WORK_ACTIVE (e.g., due to a concurrent cancellation), the
cleanup path calls locks_free_lock(flock) without dequeuing the work from
the async_requests list. Concurrently, smb2_cancel() walks the list
under conn->request_lock and invokes the cancel callback, which then
dereferences the already freed 'flock'. This leads to a slab-use-after-free
inside __wake_up_common.
Fix this by restructuring the cleanup logic after the worker returns
from ksmbd_vfs_posix_lock_wait(). Move list_del(&smb_lock->llist) and
release_async_work(work) to the top of the cleanup block. This guarantees
that the async work is completely dequeued and serialized under
conn->request_lock before locks_free_lock(flock) is called, rendering
the flock unreachable for any concurrent smb2_cancel().
Cc: stable@vger.kernel.org
Signed-off-by: Davide Ornaghi <d.ornaghi97@gmail.com>
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit be6d26bf27 upstream.
smb2_query_dir() stores a pointer to its stack-allocated private data in
the ksmbd_file readdir_data. Concurrent QUERY_DIRECTORY requests using the
same file handle can overwrite this pointer while an iterate_dir() callback
is still using it, resulting in a stack use-after-free.
Add a per-file mutex and hold it while accessing the shared directory
enumeration state. The lock covers scan restart, dot entry state,
readdir_data setup and iteration, and response construction. This prevents
another request from replacing readdir_data.private before the current
request has finished using it and also serializes the shared file position.
Cc: stable@vger.kernel.org
Reported-by: zdi-disclosures@trendmicro.com # ZDI-CAN-30527
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 3320ba0681 upstream.
FSCTL_SET_ZERO_DATA in smb2_ioctl() destroys file data via
ksmbd_vfs_zero_data() -> vfs_fallocate(PUNCH_HOLE/ZERO_RANGE) after
checking only the share-level KSMBD_TREE_CONN_FLAG_WRITABLE, with no
per-handle access check. A handle opened with only FILE_WRITE_ATTRIBUTES
still yields an FMODE_WRITE filp (FILE_WRITE_ATTRIBUTES is part of
FILE_WRITE_DESIRE_ACCESS_LE, so smb2_create_open_flags() opens it
O_WRONLY), so the vfs_fallocate FMODE_WRITE check does not stop it; only
the missing fp->daccess gate would. Reproduced on mainline 7.1-rc7 with
KASAN by an authenticated SMB client: a FILE_WRITE_ATTRIBUTES-only handle
zeroed 4096 bytes of file data it had no FILE_WRITE_DATA right to
(6/6; a FILE_READ_DATA-only handle was correctly denied).
This is the unfixed sibling of commit cc57232cae ("ksmbd: fix FSCTL
permission bypass by adding a permission check for FSCTL_SET_SPARSE").
Because SET_ZERO_DATA writes data (not an attribute), require
FILE_WRITE_DATA.
Cc: stable@vger.kernel.org
Signed-off-by: Gil Portnoy <dddhkts1@gmail.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 388e4139db upstream.
The FSCTL_DUPLICATE_EXTENTS_TO_FILE arm of smb2_ioctl() overwrites the
destination file's data via vfs_clone_file_range() with neither the
share-level KSMBD_TREE_CONN_FLAG_WRITABLE check nor a per-handle
fp->daccess check that the other write-bearing arms carry. A client can
overwrite destination data on a read-only share, or from a handle opened
with only FILE_WRITE_ATTRIBUTES (which still yields an FMODE_WRITE filp).
FILE_WRITE_ATTRIBUTES-only destination handle overwrote the file's data via
the clone. Add both checks, matching the FSCTL_SET_SPARSE permission fix;
require FILE_WRITE_DATA since this writes data.
Cc: stable@vger.kernel.org
Signed-off-by: Gil Portnoy <dddhkts1@gmail.com>
Acked-by: Namjae Jeon <linkinjeon@kernel.org>
Signed-off-by: Steve French <stfrench@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 61f28012e5 upstream.
The "*num_sgs" variable is a u32 so "ERR_PTR(*num_sgs)" doesn't work.
We would have to do something similar to the previous line where it's
cast to int and then long. However, it's simpler to store the return in
an int ret variable.
This bug would eventually result in a crash when dereference the invalid
error pointer.
Fixes: d08089f649 ("cifs: Change the I/O paths to use an iterator rather than a page list")
Cc: stable@kernel.org
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Steve French <stfrench@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit ec457f9afe upstream.
cifs_swn_notify() looks up a witness registration by id under
cifs_swnreg_idr_mutex, drops the mutex, and then uses the registration's
cached tcon pointer. That pointer is not a lifetime reference, and it is
not a stable representative once cifs_get_swn_reg() lets multiple tcons
for the same net/share name share one registration id.
A same-share second mount can keep the cifs_swn_reg alive after the first
tcon unregisters and is freed. The registration then still points at the
freed first tcon, so taking tc_lock or incrementing tc_count through
swnreg->tcon only moves the use-after-free earlier. Taking tc_lock while
holding cifs_swnreg_idr_mutex also violates the documented CIFS lock
order.
Fix this by making the registration store only the stable witness
identity: id, net name, share name, and notify flags. When a notify
arrives, copy that identity under cifs_swnreg_idr_mutex, drop the mutex,
then find and pin a live witness tcon that currently matches the net/share
pair under the normal cifs_tcp_ses_lock -> tc_lock order. The notification
path uses that pinned tcon directly and drops the reference when done.
Registration and unregister messages now use the live tcon passed by the
caller instead of a cached tcon in the registration. The final unregister
send is folded into cifs_swn_unregister() while the registration is still
protected by cifs_swnreg_idr_mutex. This removes the previous
find/drop/reacquire raw-pointer window. The release path only removes the
idr entry and frees the stable identity strings.
This preserves the intended one-registration/many-tcon behavior: a
registration id represents a net/share pair, and notify handling acts on a
live representative selected at use time. It also preserves CLIENT_MOVE
ordering for the representative tcon because the old-IP unregister is sent
before cifs_swn_register() sends the new-IP register.
Fixes: fed979a7e0 ("cifs: Set witness notification handler for messages from userspace daemon")
Cc: stable@vger.kernel.org
Signed-off-by: Michael Bommarito <michael.bommarito@gmail.com>
Assisted-by: Claude:claude-opus-4-7
Signed-off-by: Steve French <stfrench@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 98495b5a4d upstream.
When the SMB sink is used as a perf AUX sink, smb_update_buffer() calls
smb_sync_perf_buffer() to copy hardware trace data into the perf AUX ring
buffer pages. It derives pg_idx = head >> PAGE_SHIFT from @head, which is
handle->head, and indexes dst_pages[pg_idx]. The pg_idx %= nr_pages
normalization is only applied after the first loop iteration.
This leaves the initial page index underived from the buffer size, which
can result in an out-of-bounds write past dst_pages[] when head exceeds
the AUX buffer size.
Normalize head modulo the AUX buffer size before deriving the page index
and offset, mirroring tmc_etr_sync_perf_buffer().
Fixes: 06f5c2926a ("drivers/coresight: Add UltraSoc System Memory Buffer driver")
Reported-by: Yuhao Jiang <danisjiang@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Junrui Luo <moonafterrain@outlook.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/SYBPR01MB788156B3380A36835DB22290AF102@SYBPR01MB7881.ausprd01.prod.outlook.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ Upstream commit d7ea8495fd ]
When a compressed or sparse attribute has its clusters frame-aligned,
vcn is rounded down to the frame start using cmask, which can result
in vcn != vcn0. In this case, vcn and vcn0 may reside in different
attribute segments.
The code already handles the case where vcn is in a different segment
by loading its runs before allocation. However, it fails to load runs
for vcn0 when vcn0 resides in a different segment than vcn. This causes
run_lookup_entry() to return SPARSE_LCN for vcn0 since its segment was
never loaded into the in-memory run list, triggering the WARN_ON(1).
Fix this by adding a missing check for vcn0 after the existing vcn
segment check. If vcn0 falls outside the current segment range
[svcn, evcn1), find and load the attribute segment containing vcn0
before performing the run lookup.
The following scenario triggers the bug:
attr_data_get_block_locked()
vcn = vcn0 & cmask <- vcn != vcn0 after frame alignment
load runs for vcn segment <- vcn0 segment not loaded!
attr_allocate_clusters() <- allocation succeeds
run_lookup_entry(vcn0) <- vcn0 not in run -> SPARSE_LCN
WARN_ON(1) <- bug fires here!
Reported-by: syzbot+c1e9aedbd913fadad617@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=c1e9aedbd913fadad617
Fixes: c380b52f6c ("fs/ntfs3: Change new sparse cluster processing")
Signed-off-by: Deepanshu Kartikey <Kartikey406@gmail.com>
Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 356fa24816 ]
Handle ntfs_read_folio() early when the folio offset is beyond i_valid
by zero-filling the folio and marking it uptodate. This avoids needless
I/O and locking, improves read performance.
Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
Stable-dep-of: d7ea8495fd ("fs/ntfs3: fix missing run load for vcn0 in attr_data_get_block_locked()")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit dcd9d6a471 ]
Some xfstests expect fsync() on a file or directory to also persist
directory metadata up the parent chain. Using generic_file_fsync() is not
sufficient for ntfs, because parent directories are not explicitly
written out.
Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
Stable-dep-of: d7ea8495fd ("fs/ntfs3: fix missing run load for vcn0 in attr_data_get_block_locked()")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 4248f563f0 ]
The old "readpage" naming is still used in ni_readpage_cmpr(), even though
the vfs has transitioned to the folio-based read_folio() API.
This patch performs a straightforward renaming of the helper:
ni_readpage_cmpr() -> ni_read_folio_cmpr().
Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
Stable-dep-of: d7ea8495fd ("fs/ntfs3: fix missing run load for vcn0 in attr_data_get_block_locked()")
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit c3b1edea37 upstream.
The Intel VT-d PASID table entry is 512 bits (64 bytes). When replacing
an active PASID entry (e.g., during domain replacement), the current
implementation calculates a new entry on the stack and copies it to the
table using a single structure assignment.
struct pasid_entry *pte, new_pte;
pte = intel_pasid_get_entry(dev, pasid);
pasid_pte_config_first_level(iommu, &new_pte, ...);
*pte = new_pte;
Because the hardware may fetch the 512-bit PASID entry in multiple
128-bit chunks, updating the entire entry while it is active (Present
bit set) risks a "torn" read. In this scenario, the IOMMU hardware
could observe an inconsistent state — partially new data and partially
old data — leading to unpredictable behavior or spurious faults.
Fix this by removing the unsafe "replace" helpers and following the
"clear-then-update" flow, which ensures the Present bit is cleared and
the required invalidation handshake is completed before the new
configuration is applied.
Fixes: 7543ee63e8 ("iommu/vt-d: Add pasid replace helpers")
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Samiullah Khawaja <skhawaja@google.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Link: https://lore.kernel.org/r/20260120061816.2132558-4-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
[vineeth: adjusted for 6.18.y: the removed intel_pasid_replace_second_level()
in 6.18 still computed pgd/pgd_val locally and used the older
pasid_pte_config_second_level() signature; since the function is deleted
entirely, the resolution only affects removed lines. The prerequisite
"Clear Present bit" patches are already in 6.18.y.
Built and tested on 6.18.38: PASID replace, iommufd RID replace and DMA
sanity on QEMU VT-d in scalable mode (FS and SS), no regressions.]
Signed-off-by: Vineeth Pillai (Google) <vineeth@bitbyteword.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 687617555c upstream.
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: 7c9cbd0b5e ("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>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 2641a9e0a1 upstream.
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:
WARNING: possible circular locking dependency detected
process_pending_rx+0x21/0x2a [vuln_msv]
l2cap_conn_del.constprop.0+0x3f/0x4e [vuln_msv]
*** DEADLOCK ***
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: 7ab56c3a6e ("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>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit d5541eb148 upstream.
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: 7a17308c17 ("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>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 1b0d946d6f upstream.
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.
Fixes: c1bb9336ae ("Bluetooth: hci_uart: fix UAFs and race conditions in close and init paths")
Link: https://lore.kernel.org/linux-bluetooth/07e0a28650773abec711ee492fdb1bf5d21a6c98.camel@iki.fi/
Cc: stable@vger.kernel.org
Signed-off-by: Pauli Virtanen <pav@iki.fi>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 12917f591c upstream.
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: a13f316e90 ("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>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 4bd0b27405 upstream.
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: ab1513597c ("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>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit badff6c3be upstream.
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.
nxpdev->fw_dnld_v3_offset = offset - nxpdev->fw_v3_offset_correction;
serdev_device_write_buf(nxpdev->serdev, nxpdev->fw->data +
nxpdev->fw_dnld_v3_offset, len);
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: 689ca16e52 ("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>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a257407e2b upstream.
Every once in a while we see a hung btmtksdio_flush() task:
INFO: task kworker/u17:0:189 blocked for more than 122 seconds.
__cancel_work_timer+0x3f4/0x460
cancel_work_sync+0x1c/0x2c
btmtksdio_flush+0x2c/0x40
hci_dev_open_sync+0x10c4/0x2190
[..]
It all boils down to incorrect time_is_before_jiffies() usage in
btmtksdio_txrx_work(). The btmtksdio_txrx_work() loop is expected
to be terminated if running for longer than 5*HZ. However the
timeout check is twisted: time_is_before_jiffies(old_jiffies + 5*HZ)
evaluates to true when old_jiffies + 5*HZ is in the past i.e. when a
timeout has occurred. Using OR with time_is_before_jiffies(txrx_timeout)
means that:
- before the 5-second timeout: the condition is `int_status || false`,
so it loops as long as there are pending interrupts.
- after the 5-second timeout: the condition becomes `int_status || true`,
which is always true.
When the loop becomes infinite btmtksdio_txrx_work() loop never
terminates and never releases the SDIO host.
Fix loop termination condition to actually enforce a 5*HZ timeout.
Fixes: 26270bc189 ("Bluetooth: btmtksdio: move interrupt service to work")
Cc: stable@vger.kernel.org
Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Reviewed-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 53b3e60edb upstream.
OpenWrt has recently migrated many platforms to kernel 6.18. On the
MediaTek platform, which supports hardware network offloading, WiFi
connections accelerated via the WED path were observed to drop after
roughly 300 seconds.
After several debugging sessions, assisted by the Claude LLM, the
problem was narrowed down as follows:
nf_flow_table_extend_ct_timeout() extends ct->timeout for offloaded
flows using:
cmpxchg(&ct->timeout, expires, new_timeout);
'expires' comes from nf_ct_expires(ct) and is a relative value, while
ct->timeout holds an absolute timestamp. The two are never equal, so
the cmpxchg always fails and the timeout is never extended.
This goes unnoticed for most flows, but a long-lived hardware (WED)
offloaded flow on MediaTek MT7986 eventually has ct->timeout decay to
zero, the conntrack entry is reaped and the connection breaks.
Open-code the relative value from a single READ_ONCE(ct->timeout)
snapshot and compare against that same absolute snapshot in the
cmpxchg, so the timeout extension actually takes effect while the
datapath remains authoritative if it updates ct->timeout concurrently.
Fixes: 03428ca5ce ("netfilter: conntrack: rework offload nf_conn timeout extension logic")
Cc: stable@vger.kernel.org
Suggested-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Adrian Bente <adibente@gmail.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a622d2e960 upstream.
update_counters() and compat_update_counters() forward a user-supplied
32-byte table name to find_table_lock() without NUL-terminating it. On a
lookup miss, find_inlist_lock() calls try_then_request_module(..., "%s%s",
"ebtable_", name), and vsnprintf() reads past the name field and the
stack object until it hits a zero byte.
BUG: KASAN: stack-out-of-bounds in string (lib/vsprintf.c:648 lib/vsprintf.c:730)
Read of size 1 at addr ffff8880119dfb20 by task exploit/147
Call Trace:
...
string (lib/vsprintf.c:648 lib/vsprintf.c:730)
vsnprintf (lib/vsprintf.c:2945)
__request_module (kernel/module/kmod.c:150)
do_update_counters.isra.0 (net/bridge/netfilter/ebtables.c:371 net/bridge/netfilter/ebtables.c:380)
update_counters (net/bridge/netfilter/ebtables.c:1440)
do_ebt_set_ctl (net/bridge/netfilter/ebtables.c:2573)
nf_setsockopt (net/netfilter/nf_sockopt.c:101)
ip_setsockopt (net/ipv4/ip_sockglue.c:1424)
raw_setsockopt (net/ipv4/raw.c:847)
__sys_setsockopt (net/socket.c:2393)
...
compat_do_replace() shares the same unterminated name via
compat_copy_ebt_replace_from_user(); terminate it there too so all
find_table_lock() callers behave alike. The other callers already
terminate the name after the copy.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Fixes: 81e675c227 ("netfilter: ebtables: add CONFIG_COMPAT support")
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>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 084d23f818 upstream.
We need to explicitly check the length, else we may pass non-null
terminated string to request_module().
Cc: stable@vger.kernel.org
Fixes: bcf4934288 ("netfilter: ebtables: Fix extension lookup with identical name")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit cbfe53599e upstream.
sashiko reports:
looking at ebtables table
translation, could a sparse cpu_possible_mask lead to an uninitialized pointer
free?
If cpu_possible_mask is sparse (for example, CPU 0 and CPU 2 are possible,
but CPU 1 is not), the allocation loop skips CPU 1. If vmalloc_node() fails at
CPU 2, the cleanup loop will blindly decrement and call vfree() on
newinfo->chainstack[1].
Not a real-world bug, such allocation isn't expected to fail
in the first place.
Cc: stable@vger.kernel.org
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit da5b58478a upstream.
sashiko reports:
When an skb with unreadable fragments (such as from devmem TCP, where
skb_frags_readable(skb) returns false) is processed by the u32 module,
skb_copy_bits() will safely return a negative error code [..]
xt_u32: bail out with hotdrop in this case.
gather_frags: return -1, just as if we had no fragment header.
nfnetlink_queue: restrict to the linear part.
nfnetlink_log: restrict to the linear part.
v2:
- skb_zerocopy helpers don't copy readable flag, i.e. nfnetlink_queue
is broken too
xt_u32 shouldn't return true if hotdrop was set.
Fixes: 65249feb6b ("net: add support for skbs with unreadable frags")
Cc: stable@vger.kernel.org
Acked-by: Mina Almasry <almasrymina@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a7f57320bb upstream.
Commit 02a3231b6d ("netfilter: nf_conntrack_expect: store netns and zone in expectation")
introduced exp->net so RCU-only expectation paths no longer need to
dereference exp->master for netns lookups.
Commit 3db5647984 ("netfilter: nf_conntrack_expect: skip expectations in other netns via proc")
updated the proc path accordingly, but ctnetlink_exp_dump_table() still
compares against nf_ct_net(exp->master).
Use nf_ct_exp_net(exp) here as well so the netlink dump path matches
the rest of the March 2026 expectation netns/RCU cleanup.
Fixes: 02a3231b6d ("netfilter: nf_conntrack_expect: store netns and zone in expectation")
Cc: stable@vger.kernel.org
Signed-off-by: Pratham Gupta <pratham36gupta@gmail.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 66366d291f upstream.
We hit a real softlockup in an internal stress test environment. The
workload was LTP memory/swap stress on a large arm64 machine, with 320
CPUs, about 1TB memory and an 8.6GB swap device. The system was under
heavy load and the swap device had a large number of full clusters. The
softlockup was triggered during a stress test after about 3 days.
So, add periodic cond_resched() calls during large full_clusters
reclaim operations to prevent softlockup issues.
Detailed call trace as follow:
PID: 3817773 TASK: ffff0883bb28b780 CPU: 48 COMMAND: "kworker/48:7"
#0 [ffff800080183d10] __crash_kexec at ffffa4c1361e5de4
#1 [ffff800080183d90] panic at ffffa4c1360d5e9c
#2 [ffff800080183e20] watchdog_timer_fn at ffffa4c136231fa8
...
#16 [ffff8000c4ad3cb0] swap_cache_del_folio at ffffa4c1363e1614
#17 [ffff8000c4ad3ce0] __try_to_reclaim_swap at ffffa4c1363e4bfc
#18 [ffff8000c4ad3d40] swap_reclaim_full_clusters at ffffa4c1363e5474
#19 [ffff8000c4ad3da0] swap_reclaim_work at ffffa4c1363e550c
#20 [ffff8000c4ad3dc0] process_one_work at ffffa4c136102edc
#21 [ffff8000c4ad3e10] worker_thread at ffffa4c136103398
#22 [ffff8000c4ad3e70] kthread at ffffa4c13610d95c
Link: https://lore.kernel.org/20260506130919.2298807-1-kerayhuang@tencent.com
Fixes: 5168a68eb7 ("mm, swap: avoid over reclaim of full clusters")
Signed-off-by: Zijiang Huang <kerayhuang@tencent.com>
Reviewed-by: Kairui Song <kasong@tencent.com>
Reviewed-by: Hao Peng <flyingpeng@tencent.com>
Reviewed-by: albinwyang <albinwyang@tencent.com>
Reviewed-by: Baoquan He <baoquan.he@linux.dev>
Acked-by: Chris Li <chrisl@kernel.org>
Cc: Barry Song <baohua@kernel.org>
Cc: Kairui Song <kasong@tencent.com>
Cc: Kemeng Shi <shikemeng@huaweicloud.com>
Cc: Nhat Pham <nphamcs@gmail.com>
Cc: Youngjun Park <youngjun.park@lge.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 63b02a9409 upstream.
lookup_swap_cgroup_id() passes swap_cgroup_ctrl[type].map to
__swap_cgroup_id_lookup() without checking that the type was ever
registered via swap_cgroup_swapon(). On a swapless host every ctrl->map
is NULL, so __swap_cgroup_id_lookup() dereferences NULL + a scaled
swp_offset().
Since commit bea67dcc5e ("mm: attempt to batch free swap entries for
zap_pte_range()"), zap_pte_range() -> swap_pte_batch() calls
lookup_swap_cgroup_id() on any non-present, non-none PTE that decodes as a
real swap entry, without first validating it against swap_info[]. A
single PTE corrupted into a type-0 swap entry takes the host down at
process exit.
We hit this in production on a swapless 6.12.58 host: ~1s of
"get_swap_device: Bad swap file entry 3f800204222bb" (do_swap_page() being
correctly defensive about the same entry) followed by
BUG: unable to handle page fault for address: 000003f800204220
RIP: 0010:lookup_swap_cgroup_id+0x2b/0x60
Call Trace:
swap_pte_batch+0xbf/0x230
zap_pte_range+0x4c8/0x780
unmap_page_range+0x190/0x3e0
exit_mmap+0xd9/0x3c0
do_exit+0x20c/0x4b0
syzbot has reported the identical stack.
The source of the PTE corruption is a separate bug; this change makes the
teardown path as robust as the fault path already is. Every other caller
of lookup_swap_cgroup_id() is downstream of a get_swap_device() that has
already validated the entry, so the new branch is cold.
Link: https://lore.kernel.org/20260504-swap-cgroup-fix-7-0-v1-1-f53ff41ee553@linux.dev
Fixes: bea67dcc5e ("mm: attempt to batch free swap entries for zap_pte_range()")
Signed-off-by: Jose Fernandez (Anthropic) <jose.fernandez@linux.dev>
Reported-by: syzbot+e12bd9ca48157add237a@syzkaller.appspotmail.com
Link: https://lore.kernel.org/r/69859728.050a0220.3b3015.0033.GAE@google.com
Assisted-by: Claude:unspecified
Cc: Barry Song <baohua@kernel.org>
Cc: David Hildenbrand <david@kernel.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Kairui Song <ryncsn@gmail.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Muchun Song <muchun.song@linux.dev>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Shakeel Butt <shakeel.butt@linux.dev>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit e30453c61e upstream.
shrinker_debugfs_add() creates both "count" and "scan" debugfs files
unconditionally.
That assumes every shrinker implements both count_objects() and
scan_objects(), which is not guaranteed. For example, the xen-backend
shrinker sets count_objects() but leaves scan_objects() NULL, so writing
to its scan file calls through a NULL function pointer and panics the
kernel:
BUG: kernel NULL pointer dereference, address: 0000000000000000
RIP: 0010:0x0
Code: Unable to access opcode bytes at 0xffffffffffffffd6.
Call Trace:
<TASK>
shrinker_debugfs_scan_write+0x12e/0x270
full_proxy_write+0x5f/0x90
vfs_write+0xde/0x420
? filp_flush+0x75/0x90
? filp_close+0x1d/0x30
? do_dup2+0xb8/0x120
ksys_write+0x68/0xf0
? filp_flush+0x75/0x90
do_syscall_64+0xb3/0x5b0
entry_SYSCALL_64_after_hwframe+0x76/0x7e
The count path has the same issue in principle if a shrinker omits
count_objects().
To fix it, only create "count" and "scan" debugfs files when the
corresponding callbacks are present.
Link: https://lore.kernel.org/20260617090052.27325-1-qi.zheng@linux.dev
Fixes: bbf535fd6f ("mm: shrinkers: add scan interface for shrinker debugfs")
Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com>
Reviewed-by: Muchun Song <muchun.song@linux.dev>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 65476d31d8 upstream.
expand_shrinker_info() iterates all visible memcgs under shrinker_mutex,
including memcgs that have not finished ->css_online() yet.
Once pn->shrinker_info has been published, teardown must stay serialized
with expand_shrinker_info() until that memcg is either fully online or no
longer visible to iteration. Today alloc_shrinker_info() breaks that rule
by dropping shrinker_mutex before freeing a partially initialized
shrinker_info array, which may cause the following race:
CPU0 CPU1
==== ====
css_create
--> list_add_tail_rcu(&css->sibling, &parent_css->children);
online_css
--> mem_cgroup_css_online
--> alloc_shrinker_info
--> alloc node0 info
rcu_assign_pointer(C->node0->shrinker_info, old0)
alloc node1 info -> FAIL -> goto err
mutex_unlock(shrinker_mutex)
shrinker_alloc()
--> shrinker_memcg_alloc
--> mutex_lock(shrinker_mutex)
expand_shrinker_info
--> mem_cgroup_iter see the memcg
expand_one_shrinker_info
--> old0 = C->node0->shrinker_info
memcpy(new->unit, old0->unit, ...);
free_shrinker_info
--> kvfree(old0);
/* double free !! */
kvfree_rcu(old0, rcu);
The same problem exists later in mem_cgroup_css_online(). If
alloc_shrinker_info() succeeds but a subsequent objcg allocation fails,
the free_objcg -> free_shrinker_info() unwind path tears down the already
published pn->shrinker_info arrays without shrinker_mutex. The
expand_one_shrinker_info() can race with that teardown in the same way,
leading to use-after-free or double-free of the old shrinker_info.
Fix this by serializing shrinker_info teardown with shrinker_mutex, and by
keeping alloc_shrinker_info() error cleanup inside the locked section.
Link: https://lore.kernel.org/20260617085658.27096-1-qi.zheng@linux.dev
Fixes: 307bececcd ("mm: shrinker: add a secondary array for shrinker_info::{map, nr_deferred}")
Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com>
Acked-by: Muchun Song <muchun.song@linux.dev>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Qi Zheng <zhengqi.arch@bytedance.com>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit b902890c62 upstream.
Reading the debugfs "count" file of a memcg-aware shrinker can sleep
inside an RCU read-side critical section:
BUG: sleeping function called from invalid context at kernel/cgroup/rstat.c:421
RCU nest depth: 1, expected: 0
css_rstat_flush
mem_cgroup_flush_stats
zswap_shrinker_count
shrinker_debugfs_count_show
shrinker_debugfs_count_show() invokes the ->count_objects() callback under
rcu_read_lock(). The zswap callback flushes memcg stats via
css_rstat_flush(), which may sleep, so it must not run under RCU.
The RCU lock is not needed here. mem_cgroup_iter() takes RCU internally
and returns a memcg holding a css reference (dropped on the next iteration
or by mem_cgroup_iter_break()), so the memcg stays alive without it. The
shrinker is kept alive by the open debugfs file: shrinker_free() removes
the debugfs entries via debugfs_remove_recursive(), which waits for
in-flight readers to drain, before call_rcu(..., shrinker_free_rcu_cb).
The sibling "scan" handler already invokes the sleeping ->scan_objects()
callback with no RCU section.
Drop the rcu_read_lock()/rcu_read_unlock().
Link: https://lore.kernel.org/20260610232048.62930-1-shakeel.butt@linux.dev
Fixes: 5035ebc644 ("mm: shrinkers: introduce debugfs interface for memory shrinkers")
Signed-off-by: Shakeel Butt <shakeel.butt@linux.dev>
Reported-by: Zenghui Yu <zenghui.yu@linux.dev>
Closes: https://lore.kernel.org/all/c052a064-cddb-494f-a0d8-f8a10b4b1c4d@linux.dev/
Suggested-by: Nhat Pham <nphamcs@gmail.com>
Reviewed-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Qi Zheng <qi.zheng@linux.dev>
Tested-by: Zenghui Yu (Huawei) <zenghui.yu@linux.dev>
Reviewed-by: Nhat Pham <nphamcs@gmail.com>
Acked-by: Muchun Song <muchun.song@linux.dev>
Reviewed-by: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Dave Chinner <david@fromorbit.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 8b2c1d41bc upstream.
If ec_device_probe() fails, cros_ec_class_release releases memory for the
cros_ec_dev structure. However, because the drvdata was already set,
sub-drivers like cros_ec_typec can still retrieve the stale pointer via the
platform device. This leads to a use-after-free when cros_ec_typec attempts
to access &typec->ec->ec->dev on a device that has already been released.
Move dev_set_drvdata() to ensure that the pointer is only made available
once all initialization steps have succeeded.
sysfs: cannot create duplicate filename '/class/chromeos/cros_ec'
Call trace:
sysfs_do_create_link_sd+0x94/0xdc
sysfs_create_link+0x30/0x44
device_add_class_symlinks+0x90/0x13c
device_add+0xf0/0x50c
ec_device_probe+0x150/0x4f0
platform_probe+0xa0/0xe0
...
BUG: KASAN: invalid-access in __memcpy+0x44/0x230
Write at addr f5ffff809e2d33ac by task kworker/u32:5/125
Pointer tag: [f5], memory tag: [fe]
Tainted : [W]=WARN, [O]=OOT_MODULE
Hardware name: Google Navi unprovisioned 0x7FFFFFFF/sku0 board/sku3
Workqueue: events_unbound deferred_probe_work_func
Call trace:
__memcpy+0x44/0x230
cros_ec_check_features+0x60/0xcc [cros_ec_proto]
cros_typec_probe+0xe8/0x6e0 [cros_ec_typec]
platform_probe+0xa0/0xe0
Cc: stable@vger.kernel.org
Fixes: 1c1d152cc5 ("platform/chrome: cros_ec_dev - utilize new cdev_device_add helper function")
Co-developed-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Signed-off-by: Andrei Kuchynski <akuchynski@chromium.org>
Reviewed-by: Benson Leung <bleung@chromium.org>
Link: https://patch.msgid.link/20260427131721.1165078-1-akuchynski@chromium.org
Signed-off-by: Lee Jones <lee@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit efb8763d7b upstream.
Reject invalid `net.ipv4.tcp_reordering` values before they reach TCP
socket state. The sysctl is stored as an `int` but copied into the
`u32` `tp->reordering` field for new sockets, so negative writes wrap
to large values.
With `tcp_mtu_probing=2`, the wrapped value can overflow the
`tcp_mtu_probe()` size calculation and drive the MTU probing path into
an out-of-bounds read. Route `tcp_reordering` writes through
`proc_dointvec_minmax()` and require it to be at least 1. Also require
`tcp_max_reordering` to be at least 1 so the configured maximum cannot
become negative either.
When registering the table for a non-init network namespace, relocate
`extra2` pointers that refer into `init_net.ipv4` so the
`tcp_reordering` upper bound follows that namespace's
`tcp_max_reordering`.
Harden `tcp_mtu_probe()` itself by computing `size_needed` as `u64`.
This keeps the send queue and window checks from being bypassed through
signed integer overflow.
Fixes: 91cc17c0e5 ("[TCP]: MTUprobe: receiver window & data available checks fixed")
Cc: stable@vger.kernel.org
Reported-by: Yuan Tan <yuantan098@gmail.com>
Reported-by: Zhengchuan Liang <zcliangcn@gmail.com>
Reported-by: Xin Liu <bird@lzu.edu.cn>
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Wyatt Feng <bronzed_45_vested@icloud.com>
Signed-off-by: Ren Wei <n05ec@lzu.edu.cn>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/1a5b7e1ef4d70fbad8c8ee0b82d8405f3c964a3d.1781395200.git.bronzed_45_vested@icloud.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 7993211bde upstream.
When a device is destroyed under RTNL, ip_mc_destroy_dev() iterates through
the multicast list and calls ip_ma_put() on each membership, scheduling
them for RCU reclamation. However, they are not unlinked from the device's
multicast hash table (mc_hash).
Since the device remains published in dev->ip_ptr until after
ip_mc_destroy_dev() completes, concurrent RCU readers traversing mc_hash
can still locate and access the multicast group after its refcount is
decremented. If the RCU callback runs and frees the group while a reader is
accessing it, a use-after-free occurs.
Fix this by unlinking the multicast group from mc_hash using
ip_mc_hash_remove() before scheduling it for reclamation.
BUG: KASAN: slab-use-after-free in ip_check_mc_rcu+0x149/0x3f0
Read of size 4 at addr ffff888009bf1408 by task mausezahn/2276
Call Trace:
<IRQ>
dump_stack_lvl+0x67/0x90
print_report+0x175/0x7c0
kasan_report+0x147/0x180
ip_check_mc_rcu+0x149/0x3f0
udp_v4_early_demux+0x36d/0x12d0
ip_rcv_finish_core+0xb8b/0x1390
ip_rcv_finish+0x54/0x120
NF_HOOK+0x213/0x2b0
__netif_receive_skb+0x126/0x340
process_backlog+0x4f2/0xf00
__napi_poll+0x92/0x2c0
net_rx_action+0x583/0xc60
handle_softirqs+0x236/0x7f0
do_softirq+0x57/0x80
</IRQ>
Allocated by task 2239:
kasan_save_track+0x3e/0x80
__kasan_kmalloc+0x72/0x90
____ip_mc_inc_group+0x31a/0xa40
__ip_mc_join_group+0x334/0x3f0
do_ip_setsockopt+0x16fa/0x2010
ip_setsockopt+0x3f/0x90
do_sock_setsockopt+0x1ad/0x300
Freed by task 0:
kasan_save_track+0x3e/0x80
kasan_save_free_info+0x40/0x50
__kasan_slab_free+0x3a/0x60
__rcu_free_sheaf_prepare+0xd4/0x220
rcu_free_sheaf+0x36/0x190
rcu_core+0x8d9/0x12f0
handle_softirqs+0x236/0x7f0
Fixes: e989707135 ("igmp: hash a hash table to speedup ip_check_mc_rcu()")
Cc: stable@vger.kernel.org
Signed-off-by: Yuyang Huang <yuyanghuang@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20260701235014.73505-1-yuyanghuang@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 45f1458a85 upstream.
There is a use-after-free error on netpoll, which is clearly detected by
KASAN.
BUG: KASAN: slab-use-after-free in _raw_spin_lock_irqsave+0x3b/0x80
Read of size 1 at addr ... by task kworker/9:1
Workqueue: events queue_process
Call Trace:
skb_dequeue+0x1e/0xb0
queue_process+0x2c/0x600
process_scheduled_works+0x4b6/0x850
worker_thread+0x414/0x5a0
Allocated by task 242:
__netpoll_setup+0x201/0x4a0
netpoll_setup+0x249/0x550
enabled_store+0x32f/0x380
Freed by task 0:
kfree+0x1b7/0x540
rcu_core+0x3f8/0x7a0
The problem happens when there is a pending TX worker running in
parallel with the cleanup path.
This is what happens on netpoll shutdown path:
1) __netpoll_cleanup() is called
2) set dev->npinfo to NULL
3) call_rcu() with rcu_cleanup_netpoll_info()
3.1) rcu_cleanup_netpoll_info() tries to cancel all workers with
cancel_delayed_work(), but doesn't wait for the worker to finish
4) and kfree(npinfo);
Because 3.1) doesn't really cancel the work, as the comment says "we
can't call cancel_delayed_work_sync here, as we are in softirq", the TX
worker can run after 4).
Tl;DR: queue_process() is not an RCU reader, it reaches npinfo through
the work item via container_of().
Use disable_delayed_work_sync() to ensure the worker is completely
stopped and prevent any future re-arming attempts. Once npinfo is set
to NULL, senders will bail out and not queue new work. The disable flag
ensures any in-flight re-arming attempts also fail silently.
In the future, we can do the cleanup inline here without needing the
npinfo->rcu rcu_head, but that is net-next material.
Cc: stable@vger.kernel.org
Fixes: 38e6bc185d ("netpoll: make __netpoll_cleanup non-block")
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://patch.msgid.link/20260625-netpoll_rcu_fix-v2-1-0748ffac1e98@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c554246ff4 upstream.
A partial read will store the completed byte count in io->bytes_done.
The regular completion path applies io_fixup_rw_res() so that, when the
following operation reaches EOF, the number of bytes already read is
returned.
The iopoll completion path does not apply this fixup to the return value
and can return zero instead.
Use the fixup result when updating the CQE, and the raw result for the
reissue check.
Cc: stable@vger.kernel.org
Fixes: 4d9cb92ca4 ("io_uring/rw: fix short rw error handling")
Signed-off-by: Michael Wigham <michael@wigham.net>
Link: https://patch.msgid.link/20260613225240.34032-1-michael@wigham.net
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 29bef9934b upstream.
commit 10dc959398 ("io_uring/io-wq: check IO_WQ_BIT_EXIT inside work
run loop") fixed the obvious case where io_worker_handle_work() took one
exit-bit snapshot before draining pending work, but the fix stops one
level too early.
io_worker_handle_work() now re-checks IO_WQ_BIT_EXIT in its outer work
run loop, yet it still snapshots that bit once before processing a whole
dependent linked-work chain. If io_wq_exit_start() sets IO_WQ_BIT_EXIT
after the first linked item has started, the remaining linked items can
still reuse stale do_kill = false, skip IO_WQ_WORK_CANCEL, and continue
running after exit has begun.
Move the check further inside, so it covers linked items too. Note: this
is a syzbot special as it loves setting up tons of slow linked work on
weird devices like msr that take forever to read, and immediately close
the ring. Exit then takes a long time.
Fixes: 10dc959398 ("io_uring/io-wq: check IO_WQ_BIT_EXIT inside work run loop")
Cc: stable@vger.kernel.org
Signed-off-by: Runyu Xiao <runyu.xiao@seu.edu.cn>
Link: https://patch.msgid.link/20260527172203.2043962-1-runyu.xiao@seu.edu.cn
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 2564ca2e31 upstream.
NOP file-acquisition support choses between a fixed (registered) file and
a normal fget()'d file based on its own IORING_NOP_FIXED_FILE flag in
sqe->nop_flags. However, a request's REQ_F_FIXED_FILE is set
independently from the generic IOSQE_FIXED_FILE sqe flag during request
init, before the issue handler runs.
If a NOP is submitted with IOSQE_FIXED_FILE set (so REQ_F_FIXED_FILE is
set) but without IORING_NOP_FIXED_FILE, io_nop() takes the normal path
and grabs a real reference via io_file_get_normal(). On completion,
io_put_file() only drops the reference when REQ_F_FIXED_FILE is clear,
so the fget()'d file is never released and leaks:
BUG: memory leak
unreferenced object 0xffff88800f42c240 (size 176):
kmem_cache_alloc_noprof+0x358/0x440
alloc_empty_file+0x57/0x180
path_openat+0x44/0x1e50
do_file_open+0x121/0x200
do_sys_openat2+0xa7/0x150
__x64_sys_openat+0x82/0xf0
Decide between fixed and normal file acquisition from REQ_F_FIXED_FILE,
the same way io_assign_file() does for every other opcode, and fold
IORING_NOP_FIXED_FILE into REQ_F_FIXED_FILE at prep time.
Cc: stable@vger.kernel.org
Fixes: a85f31052b ("io_uring/nop: add support for testing registered files and buffers")
Reported-by: syzbot+2cd473471e77bda12b0e@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?id=879092631b98f73a28ea405adacfa5bb34a14a25
Signed-off-by: Vasileios Almpanis <vasilisalmpanis@gmail.com>
Link: https://patch.msgid.link/20260615144619.482749-1-vasilisalmpanis@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 286533cb14 upstream.
sch_irq_unmask() enables the GPIO IRQ and then updates the controller
state through sch_irq_mask_unmask(), which takes sch->lock with
spin_lock_irqsave(). The callback can be reached from irq_startup()
while setting up a requested IRQ. That path is not sleepable, but on
PREEMPT_RT a regular spinlock_t becomes a sleeping lock.
This issue was found by our static analysis tool and then manually
reviewed against the current tree.
The grounded PoC kept the request_threaded_irq() -> __setup_irq() ->
irq_startup() -> sch_irq_unmask() -> sch_irq_mask_unmask() carrier and
used the original spin_lock_irqsave(&sch->lock) edge. Lockdep reported:
BUG: sleeping function called from invalid context
hardirqs last disabled at ... __setup_irq.constprop.0 ... [vuln_msv]
sch_rt_spin_lock_irqsave+0x1c/0x30 [vuln_msv]
sch_irq_mask_unmask.constprop.0+0x31/0x70 [vuln_msv]
__setup_irq.constprop.0+0xd/0x30 [vuln_msv]
Convert the SCH controller lock to raw_spinlock_t. The same lock is
also used by the GPIO direction and value callbacks, but those critical
sections only update MMIO-backed GPIO registers and do not contain
sleepable operations. Keeping this register lock non-sleeping is
therefore appropriate for the irqchip callbacks and does not change the
GPIO-side locking contract.
Fixes: 7a81638485 ("gpio: sch: Add edge event support")
Cc: stable@vger.kernel.org
Signed-off-by: Runyu Xiao <runyu.xiao@seu.edu.cn>
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Link: https://patch.msgid.link/20260617154035.1199948-2-runyu.xiao@seu.edu.cn
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 90f0109019 upstream.
sprd_eic_irq_unmask() enables the GPIO IRQ and then updates controller
state through sprd_eic_update(), which takes sprd_eic->lock with
spin_lock_irqsave(). The callback can be reached from irq_startup()
while setting up a requested IRQ. That path is not sleepable, but on
PREEMPT_RT a regular spinlock_t becomes a sleeping lock.
This issue was found by our static analysis tool and then manually
reviewed against the current tree.
The grounded PoC kept the request_threaded_irq() -> __setup_irq() ->
irq_startup() -> sprd_eic_irq_unmask() -> sprd_eic_update() carrier and
used the original spin_lock_irqsave(&sprd_eic->lock) edge. Lockdep
reported:
BUG: sleeping function called from invalid context
hardirqs last disabled at ... __setup_irq.constprop.0 ... [vuln_msv]
sprd_rt_spin_lock_irqsave+0x1c/0x30 [vuln_msv]
sprd_eic_update.constprop.0+0x48/0x90 [vuln_msv]
sprd_eic_irq_unmask.constprop.0+0x35/0x50 [vuln_msv]
__setup_irq.constprop.0+0xd/0x30 [vuln_msv]
Convert the Spreadtrum EIC controller lock to raw_spinlock_t. The
locked section only serializes MMIO register updates and does not contain
sleepable operations, so keeping it non-sleeping is appropriate for the
irqchip callbacks.
Fixes: 25518e024e ("gpio: Add Spreadtrum EIC driver support")
Cc: stable@vger.kernel.org
Signed-off-by: Runyu Xiao <runyu.xiao@seu.edu.cn>
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://patch.msgid.link/20260617154035.1199948-3-runyu.xiao@seu.edu.cn
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 4dcddc1c79 upstream.
ntb_epf_vec_isr() calls pci_irq_vector() in hardirq context to derive
the vector number. pci_irq_vector() calls msi_get_virq() that takes a
mutex and can therefore trigger "scheduling while atomic" splats:
BUG: scheduling while atomic: kworker/u33:0/55/0x00010001
...
Call trace:
...
schedule+0x38/0x110
schedule_preempt_disabled+0x28/0x50
__mutex_lock.constprop.0+0x848/0x908
__mutex_lock_slowpath+0x18/0x30
mutex_lock+0x4c/0x60
msi_domain_get_virq+0xe8/0x138
pci_irq_vector+0x2c/0x60
ntb_epf_vec_isr+0x28/0x120 [ntb_hw_epf]
__handle_irq_event_percpu+0x70/0x3a8
handle_irq_event+0x48/0x100
handle_edge_irq+0x100/0x1c8
...
Cache the Linux IRQ number for vector 0 when vectors are allocated and
use it as a base in the ISR. Running the ISR in a threaded IRQ handler
would also avoid the problem, but that would be unnecessary here.
Fixes: 812ce2f8d1 ("NTB: Add support for EPF PCI Non-Transparent Bridge")
Signed-off-by: Koichiro Den <den@valinux.co.jp>
Signed-off-by: Manivannan Sadhasivam <mani@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Cc: stable@vger.kernel.org # v5.12+
Link: https://patch.msgid.link/20260304083028.1391068-3-den@valinux.co.jp
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 57382ec6ac upstream.
In the analysis pass of $LogFile journal replay, log_replay() copies
LCNs from each action log record into an existing Dirty Page Table
(DPT) entry without bounding the destination index. A crafted NTFS
image with DPT entry lcns_follow=1 and an action log record with
lcns_follow=2 produces a kernel slab out-of-bounds write at mount
time:
BUG: KASAN: slab-out-of-bounds in log_replay+0x654c/0xdb60
Write of size 8 at addr ffff8880095e1040 by task mount
Two attacker-controlled fields can drive j+i past the allocated
page_lcns[] array:
1. dp->lcns_follow (capacity) can be smaller than lrh->lcns_follow.
2. lrh->target_vcn may be smaller than dp->vcn, making the u64
subtraction wrap to a huge size_t.
Validate target VCN delta and per-record LCN count against the
DPT entry capacity, bail via the existing out: cleanup label with
-EINVAL.
This mirrors the bounds-check pattern added in commit b2bc7c44ed
("fs/ntfs3: Fix slab-out-of-bounds read in DeleteIndexEntryRoot")
and commit 0ca0485e4b ("fs/ntfs3: validate rec->used in
journal-replay file record check").
Fixes: b46acd6a6a ("fs/ntfs3: Add NTFS journal")
Reported-by: Yunpeng Tian <shionthanatos@gmail.com>
Reported-by: Mingda Zhang <npczmd@qq.com>
Reported-by: Gongming Wang <gmwgg05@gmail.com>
Reported-by: Peiyuan Xu <paulbucket12@gmail.com>
Reported-by: Qinrun Dai <jupmouse@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Yunpeng Tian <shionthanatos@gmail.com>
Signed-off-by: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit b81dde13cc upstream.
syzbot reported a puzzling splat:
WARNING: kernel/time/hrtimer.c:443 at stub_timer+0xa/0x20
stub_timer() is installed as timer callback function in
hrtimer_fixup_assert_init(), which is invoked when
debug_object_assert_init() can't find a shadow object. In that case debug
objects emits a warning about it before invoking the fixup.
Though the provided console log lacks this warning and instead has the
following a few seconds before the splat:
ODEBUG: Out of memory. ODEBUG disabled
So the object was looked up in debug_object_assert_init() and the lookup
failed due a concurrent out of memory situation which disabled debug
objects and freed the shadow objects:
debug_object_assert_init()
if (!debug_objects_enabled)
return; obj = alloc();
if (!obj) {
// Out of memory
debug_objects_enabled = false;
free_objects();
obj = lookup_or_alloc();
// The lookup failed because the other side
// removed the objects, so this returns
// an error code as the object in question
// is not statically initialized
if (!IS_ERR_OR_NULL(obj))
return;
if (!obj) {
debug_oom();
return;
}
print(...)
if (!debug_objects_enabled)
return;
fixup(...)
The debug object splat is skipped because debug_objects_enabled is false,
but the fixup callback is invoked unconditionally, which makes the timer
disfunctional.
This is only a problem in debug_object_assert_init() and
debug_object_activate() as both have to handle statically initialized
objects and therefore must handle the error pointer return case
gracefully. All other places only handle the found/not found case and the
NULL pointer return is a signal for OOM. Otherwise they get a valid shadow
object.
Plug the hole by checking whether debug objects are still enabled before
invoking the print and fixup function in those two places.
Fixes: b84d435cc2 ("debugobjects: Extend to assert that an object is initialized")
Reported-by: syzbot+5e8dda76ca21dae314b6@syzkaller.appspotmail.com
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/874iiwlzlb.ffs@fw13
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit fa09f08ede upstream.
The etb10 miscdevice uses drvdata->reading as a shared exclusivity gate
for userspace buffer access. etb_open() claims that gate with
local_cmpxchg(), and etb_release() clears it with local_set().
That gate is shared per-device state rather than CPU-local state. A
running system can reach it whenever /dev/<etb> is opened, closed, and
reopened by different tasks while the device remains registered, so the
same drvdata->reading variable may be claimed on one CPU and later
cleared on another.
This code used to use atomic_t for the same gate, but commit
27b10da8ff ("coresight: etb10: moving to local atomic operations")
changed it to local_t even though the access pattern remained cross-task
and cross-CPU. Restore atomic_t together with atomic_cmpxchg() and
atomic_set() so the exclusivity gate again uses a primitive intended
for shared state.
The issue was found on Linux v6.18.21 by our static analysis tool while
scanning surviving local_t-on-shared-state sites, and then manually
reviewed against the live etb10 file-op path.
It was runtime-validated with a reproducible QEMU no-device KCSAN PoC
that kept the same report-local contract:
1. use one shared struct etb_drvdata carrier and its
drvdata->reading gate;
2. call etb_open() and etb_release() sequentially on that gate to
confirm the original claim/clear path;
3. bind the open side to CPU0 and the release side to CPU1 for the
same gate to show cross-CPU ownership;
4. run bound workers that repeatedly race etb_open() and
etb_release() on the same gate until KCSAN reports a target hit.
The harness recorded:
L1 passed open=1 release=1
reading_after_open=1 reading_after_release=0
L2 passed open_cpu=0 release_cpu=1
cross_cpu_release=1 reading_after=0 open_ret=0
Representative KCSAN excerpt from the no-device validation run:
BUG: KCSAN: data-race in etb_open.constprop.0.isra.0 [vuln_msv]
write to 0xffffffffc0003810 of 4 bytes by task 216 on cpu 1:
etb_open.constprop.0.isra.0+0x38/0x80 [vuln_msv]
l3_worker_thread_fn+0x4f/0xf0 [vuln_msv]
kthread+0x17e/0x1c0
ret_from_fork+0x22/0x30
read to 0xffffffffc0003810 of 4 bytes by task 215 on cpu 0:
etb_open.constprop.0.isra.0+0x18/0x80 [vuln_msv]
l3_worker_thread_fn+0x4f/0xf0 [vuln_msv]
kthread+0x17e/0x1c0
ret_from_fork+0x22/0x30
value changed: 0x00000000 -> 0x00000001
Reported by Kernel Concurrency Sanitizer on:
CPU: 0 PID: 215 Comm: etb10_l3_a Tainted: G O 6.1.66 #2
This no-device harness is not a real ETB10 hardware end-to-end run, but
it preserves the same shared drvdata->reading gate and the same
etb_open()/etb_release() claim/clear contract. No real ETB10 hardware
was available for runtime testing.
Build-tested with:
make olddefconfig
make -j"$(nproc)" drivers/hwtracing/coresight/coresight-etb10.o
Fixes: 27b10da8ff ("coresight: etb10: moving to local atomic operations")
Cc: stable@vger.kernel.org
Signed-off-by: Runyu Xiao <runyu.xiao@seu.edu.cn>
Reviewed-by: James Clark <james.clark@linaro.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20260528165201.319452-1-runyu.xiao@seu.edu.cn
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit b66774b48d upstream.
l2cap_chan_timeout() runs asynchronously and accesses chan->conn. If
the connection is torn down while the timer is running or pending,
chan->conn can be freed, leading to a use-after-free when the timer
worker attempts to lock conn->lock:
| BUG: KASAN: slab-use-after-free in instrument_atomic_read_write include/linux/instrumented.h:112 [inline]
| BUG: KASAN: slab-use-after-free in atomic_long_try_cmpxchg_acquire include/linux/atomic/atomic-instrumented.h:4456 [inline]
| BUG: KASAN: slab-use-after-free in __mutex_trylock_fast kernel/locking/mutex.c:161 [inline]
| BUG: KASAN: slab-use-after-free in mutex_lock+0x4f/0xa0 kernel/locking/mutex.c:318
| Write of size 8 at addr ffff8881298d9550 by task kworker/2:1/83
|
| CPU: 2 UID: 0 PID: 83 Comm: kworker/2:1 Not tainted 7.1.0-rc6-next-20260601-dirty #6 PREEMPT(full)
| Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.17.0-debian-1.17.0-1 04/01/2014
| Workqueue: events l2cap_chan_timeout
| Call Trace:
| <TASK>
| instrument_atomic_read_write include/linux/instrumented.h:112 [inline]
| atomic_long_try_cmpxchg_acquire include/linux/atomic/atomic-instrumented.h:4456 [inline]
| __mutex_trylock_fast kernel/locking/mutex.c:161 [inline]
| mutex_lock+0x4f/0xa0 kernel/locking/mutex.c:318
| l2cap_chan_timeout+0x5d/0x1b0 net/bluetooth/l2cap_core.c:422
| process_one_work kernel/workqueue.c:3326 [inline]
| process_scheduled_works+0x7c8/0xfb0 kernel/workqueue.c:3409
| worker_thread+0x8a9/0xcf0 kernel/workqueue.c:3490
| kthread+0x346/0x430 kernel/kthread.c:436
| ret_from_fork+0x1a3/0x470 arch/x86/kernel/process.c:158
| ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
| </TASK>
|
| Allocated by task 320:
| l2cap_conn_add+0xa7/0x820 net/bluetooth/l2cap_core.c:7075
| l2cap_connect_cfm+0xdb/0xd70 net/bluetooth/l2cap_core.c:7452
| hci_connect_cfm include/net/bluetooth/hci_core.h:2139 [inline]
| hci_remote_features_evt+0x52f/0x9f0 net/bluetooth/hci_event.c:3760
| hci_event_func net/bluetooth/hci_event.c:7796 [inline]
| hci_event_packet+0x561/0xa70 net/bluetooth/hci_event.c:7847
| hci_rx_work+0x370/0x890 net/bluetooth/hci_core.c:4040
| process_one_work kernel/workqueue.c:3326 [inline]
| process_scheduled_works+0x7c8/0xfb0 kernel/workqueue.c:3409
| worker_thread+0x8a9/0xcf0 kernel/workqueue.c:3490
| kthread+0x346/0x430 kernel/kthread.c:436
| ret_from_fork+0x1a3/0x470 arch/x86/kernel/process.c:158
| ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
|
| Freed by task 322:
| hci_disconn_cfm include/net/bluetooth/hci_core.h:2154 [inline]
| hci_conn_hash_flush+0x101/0x1f0 net/bluetooth/hci_conn.c:2736
| hci_dev_close_sync+0x889/0xde0 net/bluetooth/hci_sync.c:5405
| hci_dev_do_close net/bluetooth/hci_core.c:502 [inline]
| hci_unregister_dev+0x1f7/0x370 net/bluetooth/hci_core.c:2679
| vhci_release+0x12a/0x180 drivers/bluetooth/hci_vhci.c:690
| __fput+0x369/0x890 fs/file_table.c:510
| task_work_run+0x160/0x1d0 kernel/task_work.c:233
| get_signal+0xf5b/0x1120 kernel/signal.c:2810
| arch_do_signal_or_restart+0x4d/0x600 arch/x86/kernel/signal.c:337
| __exit_to_user_mode_loop kernel/entry/common.c:64 [inline]
| exit_to_user_mode_loop+0x85/0x510 kernel/entry/common.c:98
| do_syscall_64+0x263/0x3d0 arch/x86/entry/syscall_64.c:100
| entry_SYSCALL_64_after_hwframe+0x77/0x7f
|
| The buggy address belongs to the object at ffff8881298d9400
| which belongs to the cache kmalloc-512 of size 512
| The buggy address is located 336 bytes inside of
| freed 512-byte region [ffff8881298d9400, ffff8881298d9600)
Fix it by having chan->conn hold a reference to l2cap_conn (via
l2cap_conn_get) when the channel is added to the connection, and
releasing it in the channel destructor. This ensures the l2cap_conn
remains alive as long as the channel exists.
A new FLAG_DEL channel flag is introduced to indicate that the channel
has been deleted from its connection. l2cap_chan_del() atomically sets
this flag using test_and_set_bit() instead of setting chan->conn to
NULL. All asynchronous workers (l2cap_chan_timeout, l2cap_ack_timeout,
l2cap_monitor_timeout, l2cap_retrans_timeout) and l2cap_chan_send()
check FLAG_DEL to determine whether the channel has been torn down,
rather than testing chan->conn for NULL.
Fixes: 8c8e620467 ("Bluetooth: L2CAP: use chan timer to close channels in cleanup_listen()")
Cc: <stable@vger.kernel.org>
Cc: Siwei Zhang <oss@fourdim.xyz>
Cc: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Assisted-by: Gemini:gemini-3.1-pro-preview
Reported-by: https://sashiko.dev/#/patchset/20260521021249.3258069-1-oss%40fourdim.xyz
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c9a71daaec upstream.
Multiple readers access audit_queue.qlen via skb_queue_len() without
holding the queue lock or using READ_ONCE(), while kauditd writes to
this field via the skb_dequeue() → __skb_unlink() path with WRITE_ONCE()
protected by a spinlock. This constitutes data races.
All affected skb_queue_len(&audit_queue) call sites:
- kauditd_thread() wait_event_freezable() condition
- audit_receive_msg() AUDIT_GET handler (s.backlog assignment)
- audit_receive() backlog check
- audit_log_start() backlog check and pr_warn()
KCSAN reports the following conflicting access pattern (one example):
==================================================================
BUG: KCSAN: data-race in audit_log_start / skb_dequeue
write (marked) to 0xffffffff8512ee20 of 4 bytes by task 661 on cpu 57:
skb_dequeue+0x70/0xf0
kauditd_send_queue+0x71/0x220
kauditd_thread+0x1cb/0x430
kthread+0x1c2/0x210
ret_from_fork+0x162/0x1a0
ret_from_fork_asm+0x1a/0x30
read to 0xffffffff8512ee20 of 4 bytes by task 36586 on cpu 1:
audit_log_start+0x2a0/0x6b0
audit_core_dumps+0x64/0xa0
do_coredump+0x14b/0x1260
get_signal+0xeb2/0xf70
arch_do_signal_or_restart+0x41/0x170
exit_to_user_mode_loop+0xa2/0x1c0
do_syscall_64+0x1a3/0x1c0
entry_SYSCALL_64_after_hwframe+0x76/0xe0
value changed: 0x00000001 -> 0x00000000
==================================================================
Resolve the race by switching to lockless helper skb_queue_len_lockless(),
which internally uses READ_ONCE() and properly pairs with the WRITE_ONCE()
write accesses already present on the writer side.
Cc: stable@vger.kernel.org
Fixes: 3197542482 ("audit: rework audit_log_start()")
Signed-off-by: Chi Wang <wangchi@kylinos.cn>
Reviewed-by: Ricardo Robaina <rrobaina@redhat.com>
[PM: line length tweak]
Signed-off-by: Paul Moore <paul@paul-moore.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit d129c3177d upstream.
pfkey_msg2xfrm_state() handles the IPComp (SADB_X_SATYPE_IPCOMP) case by
allocating x->calg and copying only the algorithm name:
x->calg = kmalloc_obj(*x->calg);
if (!x->calg) {
err = -ENOMEM;
goto out;
}
strcpy(x->calg->alg_name, a->name);
x->props.calgo = sa->sadb_sa_encrypt;
Unlike the authentication (x->aalg) and encryption (x->ealg) branches of
the same function, the compression branch never initializes
calg->alg_key_len. IPComp carries no key and the allocation only
reserves sizeof(struct xfrm_algo) (i.e. no room for a key), so the field
is left containing uninitialized slab data.
calg->alg_key_len is later used as a length by xfrm_algo_clone() when an
IPComp state is cloned during XFRM_MSG_MIGRATE:
xfrm_state_migrate()
xfrm_state_clone_and_setup()
x->calg = xfrm_algo_clone(orig->calg);
kmemdup(orig, xfrm_alg_len(orig));
where xfrm_alg_len() returns sizeof(*alg) + (alg_key_len + 7) / 8. With
a non-zero garbage alg_key_len, kmemdup() reads past the end of the
68-byte calg object. Adding an IPComp SA via PF_KEY and then migrating
it triggers (net-next, KASAN, init_on_alloc=0):
BUG: KASAN: slab-out-of-bounds in kmemdup_noprof+0x44/0x60
Read of size 4164 at addr ff11000025a74980 by task diag2/9287
CPU: 3 UID: 0 PID: 9287 Comm: diag2 7.1.0-rc6-g903db046d557 #1
Call Trace:
<TASK>
dump_stack_lvl+0x10e/0x1f0
print_report+0xf7/0x600
kasan_report+0xe4/0x120
kasan_check_range+0x105/0x1b0
__asan_memcpy+0x23/0x60
kmemdup_noprof+0x44/0x60
xfrm_state_migrate+0x70a/0x1da0
xfrm_migrate+0x753/0x18a0
xfrm_do_migrate+0xb47/0xf10
xfrm_user_rcv_msg+0x411/0xb50
netlink_rcv_skb+0x158/0x420
xfrm_netlink_rcv+0x71/0x90
netlink_unicast+0x584/0x850
netlink_sendmsg+0x8b0/0xdc0
____sys_sendmsg+0x9f7/0xb90
___sys_sendmsg+0x134/0x1d0
__sys_sendmsg+0x16d/0x220
do_syscall_64+0x116/0x7d0
entry_SYSCALL_64_after_hwframe+0x77/0x7f
</TASK>
Allocated by task 9287:
kasan_save_stack+0x33/0x60
kasan_save_track+0x14/0x30
__kasan_kmalloc+0xaa/0xb0
pfkey_add+0x2652/0x2ea0
pfkey_process+0x6d0/0x830
pfkey_sendmsg+0x42c/0x850
__sys_sendto+0x461/0x4b0
__x64_sys_sendto+0xe0/0x1c0
do_syscall_64+0x116/0x7d0
entry_SYSCALL_64_after_hwframe+0x77/0x7f
The buggy address belongs to the object at ff11000025a74980
which belongs to the cache kmalloc-96 of size 96
The buggy address is located 0 bytes inside of
allocated 68-byte region [ff11000025a74980, ff11000025a749c4)
Depending on the uninitialized value the same field can instead request
an oversized kmemdup() allocation and make the migration clone fail.
The XFRM netlink path is not affected: verify_one_alg() rejects an
XFRMA_ALG_COMP attribute shorter than xfrm_alg_len(), so a calg added via
XFRM_MSG_NEWSA is always self-consistent.
Initialize calg->alg_key_len to 0, matching the aalg/ealg branches.
Fixes: 80c9abaabf ("[XFRM]: Extension for dynamic update of endpoint address(es)")
Cc: stable@vger.kernel.org
Signed-off-by: Zijing Yin <yzjaurora@gmail.com>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Namjae Jeon
Davide Ornaghi
Gil Portnoy
Dan Carpenter
Michael Bommarito
Junrui Luo
Deepanshu Kartikey
Konstantin Komarov
Lu Baolu
Muhammad Bilal
Runyu Xiao
Pauli Virtanen
Siwei Zhang
Yousef Alhouseen
Maoyi Xie
Sergey Senozhatsky
Adrian Bente
Xiang Mei
Florian Westphal
Pratham Gupta
Zijiang Huang
Jose Fernandez (Anthropic)
Qi Zheng
Shakeel Butt
Andrei Kuchynski
Wyatt Feng
Yuyang Huang
Breno Leitao
Michael Wigham
Vasileios Almpanis
Koichiro Den
Yunpeng Tian
Thomas Gleixner
Samuel Page
Marco Elver
Chi Wang
Zijing Yin