| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| In the Linux kernel, the following vulnerability has been resolved:
io_uring/io-wq: don't hide a link from cancellations
A task that exits cancels its tracked requests and waits until they are
gone. io_uring_cancel_generic() looks everywhere a request can be, and
goes to sleep if it found nothing to cancel. It is woken up each time
one of the requests of the task is completed by task work or freed.
When io-wq is done with a request that has a link, there is a time in which
the next request of the link can't be found:
io_worker_handle_work()
io_assign_current_work(worker, NULL)
io_wq_free_work(work)
io_req_find_next() // takes the link off the request
io_free_req() // task work for the task
io_assign_current_work(worker, linked)
io_run_task_work()
cond_resched()
worker->cur_work = linked
It is not pending, it is not the current work of a worker, and it is not
yet on the list that it is about to be parked on. If the pass that the
exiting task makes because of io_free_req() falls into that time, it finds
nothing and the task goes to sleep. The worker then runs the link, which
IO_WQ_BIT_EXIT doesn't prevent for bounded work. If that parks the request
nothing is completed or freed, and nobody wakes up the task. If it is the
process that exits it can't be killed any more either.
NOP, IOSQE_ASYNC | IOSQE_IO_LINK
FUTEX_WAIT, FUTEX2_PRIVATE
exit of the thread, or of the process, up to 63us later
With four of these per round 5 to 8 of 10,000 rounds hang (4 CPUs, KVM):
io_uring_cancel_generic
do_exit
Commit f8b632e89a10 ("io_uring: tighten task exit cancellations") is about
the same gap, and makes the task look again for whatever happens before
io_free_req().
Free the request once the link is the current work of the worker, or
pending. Whoever is woken up by that can find the link, until it is parked
where cancellations look as well. To do so split io_wq_free_work() in two. |
| In the Linux kernel, the following vulnerability has been resolved:
io_uring: don't flood a cancelling task with task_work
io_ring_exit_work() asks each task which has used the ring to remove its
tctx node, by task_work, and repeats that until the list is empty.
io_tctx_exit_cb() doesn't touch the node while the task is in
io_uring_cancel_generic(), which removes all the nodes at its end. So the
work item queues the callback again as soon as it has completed, for
nothing.
That is more than a waste of time. task_work_run() only returns when it
finds the list empty. If the rings are released from inside the
cancellation, because requests of other rings held the last references to
their files, up to 64 work items keep sending callbacks to the one task
which is running them. It never finds the list empty, never gets back to
the cancellation, the nodes stay, and so it goes on: a task with 60 pairs
of rings which poll each other ran 280000 callbacks in 5 seconds, 20
seconds after exit_group(). It ends when all the senders happen to be late
at the same time. Until then the task keeps a CPU and the work items busy.
An exiting task can't be killed meanwhile. execve() cancels the requests in
the same way, so a live task can get stuck there as well, for minutes. If
it takes more than five, the work items run into the
WARN_ON_ONCE(time_after(jiffies, timeout)) of io_ring_exit_work(). No
privilege is needed for any of this.
If the callback has met a task in cancellation, wait a little before it is
sent again. Nobody has to wake the work item up for that: either the node
is gone by then, or it is the turn of the next round. |
| In the Linux kernel, the following vulnerability has been resolved:
io_uring/cancel: rearm local work wake up before sleep
io_uring_try_cancel_requests() sets ->cq_wait_nr to 1 so that the next
io_req_local_work_add() wakes the cancelling task. The add that takes the
count to zero is the only one that wakes, later ones see a count <= 0 and
return. io_uring_cancel_generic() however only becomes
TASK_INTERRUPTIBLE much later, so the one wake up can be spent on a
running task. That is fine as long as the entry of that add is still
queued, as the io_local_work_pending() check then skips the sleep. It is
not if the add was delayed after queueing its entry:
CPU0 (submitter in exec or exit) CPU1 (io_req_local_work_add)
mpscq_push(req A)
io_uring_try_cancel_requests()
io_run_local_work() // runs req A
io_uring_try_cancel_requests()
atomic_set(&ctx->cq_wait_nr, 1)
atomic_read(&ctx->cq_wait_nr) == 1
count taken, wake_up_state()
// task is running, no-op
prepare_to_wait(TASK_INTERRUPTIBLE)
io_local_work_pending() == false
schedule()
mpscq_push(req B)
cq_wait_nr <= 0, return
Only the sleeping task can run req B, so it is stuck in execve() until a
signal happens to wake it. An exiting task doesn't even get that.
Arm the count again once the task state is set and before the work list
is checked, for the rings whose local work this task is the one to run. |
| In the Linux kernel, the following vulnerability has been resolved:
io_uring/waitid: own the request before clearing ->head
io_waitid_remove_wq() only takes the wait queue lock if ->head is set.
io_waitid_wait() clears ->head first thing, before it tries to become the
owner of the request by bumping ->refs. If somebody else owns the request
at that point, nothing keeps it from completing the request while the
wake up handler is still going to touch it:
CPU0 (io_waitid_wait) CPU1 (io_waitid, holds a ref)
list_del_init(&wait->entry)
smp_store_release(&iw->head, NULL)
__do_wait() finds another child
io_waitid_remove_wq()
// ->head == NULL, no locking
io_waitid_finish(), IOU_COMPLETE
// request freed and reused
atomic_fetch_inc(&iw->refs)
The same goes for __io_waitid_cancel() if it claims the request in that
window. ->refs is part of the per opcode data of the request, so this
increments whatever the next user of the request keeps there. If that
happens to be zero, the handler concludes that it owns the request, and
queues io_waitid_cb() for it. With the window held open by a delay, and
the request reused for a receive that is waiting for data:
BUG: KASAN: slab-use-after-free in __wake_up_common+0x1ca/0x1f0
__wake_up_sync_key+0x36/0x50
sock_def_readable+0xf1/0x420
unix_stream_sendmsg+0x60b/0x1110
Allocated by task 87:
io_arm_apoll+0x1ed/0x940
Freed by task 87:
__io_submit_flush_completions+0x118d/0x1a10
tctx_task_work_run+0x327/0x950
Only clear ->head once the handler owns the request. If it doesn't, the
owner finds ->head set and has to wait for the lock, which is held by the
caller of the handler. Deleting the already deleted entry again is
harmless. |
| In the Linux kernel, the following vulnerability has been resolved:
io_uring/cmd: don't leave list poison in ->hash_node
io_uring_cmd_del_cancelable() uses hlist_del() to take a command off
->cancelable_uring_cmd, which leaves LIST_POISON1/2 behind in
->hash_node. Requests are recycled through the request cache without
being reinitialised, and not every user of ->hash_node sets it up before
it is used:
If a vectored futex wait is woken while futex_wait_multiple_setup() is
still queueing the futexes, io_futexv_wait() doesn't add the request to
->futex_list, and leaves the completion to the task_work queued by the
wake up. io_futexv_complete() ends in __io_futex_complete(), whose
hlist_del_init() is fine for a request that is unhashed, which is how all
other users of the node leave it. If the request was a cancelable
uring_cmd in its previous life, e.g. a ublk FETCH_REQ, it writes through
LIST_POISON2:
Oops: general protection fault, probably for non-canonical address
KASAN: maybe wild-memory-access in range [0xdead000000000120-...]
RIP: 0010:io_futexv_complete+0x219/0x420
Call Trace:
tctx_task_work_run+0x804/0x950
tctx_task_work+0x71/0xa0
task_work_run+0x118/0x200
Leave the node unhashed. |
| In the Linux kernel, the following vulnerability has been resolved:
io_uring/wait: don't spin on a stuck overflow list
As long as IO_CHECK_CQ_OVERFLOW_BIT is set, io_cqring_wait_schedule()
returns right away, before it gets to look at signals or arms the timeout,
for io_cqring_wait() to flush the overflow list. If that doesn't empty the
list, and the CQ ring doesn't hold the number of events the caller wants,
the loop starts over, with the same result. The task spins in the kernel,
and can't be killed.
The assumption is that this can't happen: if CQEs can't be flushed, the
ring is full, and min_events is capped to the size of the ring. But:
- With IORING_SETUP_CQE_MIXED a 32b CQE needs two contiguous entries. It
is stuck on the list if a single one is free, and then the ring holds
less than cq_entries events. A single threaded application that asks
for as many events as the ring has entries never gets out of
io_uring_enter().
- Whether the ring is full is up to the CQ head, which the application
can write to. While one thread waits for events on an empty ring,
another one can move the head back so that the ring looks full and
have a CQE overflow. The waiter can't be stopped any more, not even by
SIGKILL.
Only the application can make room in the CQ ring. Return to it if the
list could not be flushed. |
| In the Linux kernel, the following vulnerability has been resolved:
io_uring/io-wq: don't lose the signal that ends work
When a thread exits, only its tracked requests are cancelled. Then its
io-wq is torn down. io_wq_exit_workers() signals every worker, to get it
out of an interruptible sleep, and waits for all of them to exit. Before
they do, they go through the work that is left. Unbounded work is
cancelled. Bounded work is run, as it is expected to end by itself.
Not all of it does. An open has no file yet, so it counts as bounded, and
the open of a FIFO waits for the other end. That is what the signal is for.
But the workers are signalled once, and a worker that hasn't started the
work by then loses the signal:
io_wq_exit_workers() io_worker_handle_work()
io_wq_worker_wake()
set TIF_NOTIFY_SIGNAL
io_assign_current_work()
io_run_task_work()
clear TIF_NOTIFY_SIGNAL
io_wq_submit_work()
fifo_open()
wait_for_partner()
wait_for_completion_timeout()
The thread never gets out of do_exit():
IORING_OP_OPENAT of a FIFO without a writer, IOSQE_ASYNC
exit of the thread that has submitted it
hangs in more than 9 of 10 attempts (4 CPUs, KVM), until the process is
killed.
Set TIF_NOTIFY_SIGNAL for bounded work that is started with IO_WQ_BIT_EXIT
set. The bit is set before the workers are signalled, and it is tested
after the flag has been cleared, with clear_notify_signal() providing the
barrier. So either the flag stays, or the bit is seen.
Work that is started while the io-wq exits is treated like work that is
being run at that time: it isn't interrupted unless it waits
interruptibly. |
| In the Linux kernel, the following vulnerability has been resolved:
accel/rocket: search every core slot when looking up a scheduler
sched_to_core() walks rdev->cores[] up to rdev->num_cores, and
rocket_remove() decrements num_cores for every core it removes. Unbind a
core that is not the last one and the cores behind it fall outside the
search, so sched_to_core() returns NULL for a core that is still bound and
still running jobs. Neither caller checks the result:
rocket_job_run(): rocket_fence_create(core), core->dev
rocket_job_timedout(): dev_err(core->dev, "NPU job timed out")
Unbinding the middle core of the three on an RK3588 while three clients are
submitting to all of them faults twice, once from the surviving core's
job queue and once from its reset work:
KASAN: null-ptr-deref in range [0x0000000000000220-0x0000000000000227]
Workqueue: fdad0000.npu drm_sched_run_job_work [gpu_sched]
pc : rocket_job_run+0x234/0x838 [rocket]
Call trace:
rocket_job_run+0x234/0x838 [rocket]
drm_sched_run_job_work+0x2cc/0xad8 [gpu_sched]
process_one_work+0x640/0x14f0
KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
Workqueue: rocket-reset-2 drm_sched_job_timedout [gpu_sched]
pc : rocket_job_timedout+0xf0/0x1e0 [rocket]
Call trace:
rocket_job_timedout+0xf0/0x1e0 [rocket]
drm_sched_job_timedout+0x188/0x6a0 [gpu_sched]
Both are the third core: the workqueue names are its device and its
core->index, and it was left at slot 2 while num_cores had dropped to 2.
Search all the slots that were allocated, the way find_core_for_dev() now
does. A core that is still bound is then found, and the two callers get
the pointer they already assume they have.
This does not make unbinding one core out of several safe. An open client
keeps an entity pointing at the scheduler of the core that went away:
drm_sched reports it as not ready for every job that lands on it, and the
client waits in dma_fence_default_wait for a fence that will never signal.
Stopping the NULL dereference is what belongs in a fix; the rest wants
more thought. |
| In the Linux kernel, the following vulnerability has been resolved:
thunderbolt: Make the DP tunnel activation callback mandatory
tb_tunnel_alloc_dp() takes an optional callback which is run from
dprx_work once the DPRX capabilities read has completed. Without that
callback tb_dp_dprx_start() reads the capabilities synchronously and
never queues the work. It however always takes a tunnel reference which
is only dropped by dprx_work itself or by tb_dp_dprx_stop() when
cancel_delayed_work() actually canceled that work. That reference is
thus leaked for every tunnel without a callback.
The only tunnels without one are those from tb_tunnel_discover_dp(),
which are activated again when restoring from hibernation.
Pass the callback to tb_tunnel_discover_dp() as well and drop the
synchronous path such that the DPRX capabilities are always read from
dprx_work. Hibernation restore then also no longer blocks for up to 12
seconds while waiting for that read to complete.
Also fix up the KUnit tests. |
| In the Linux kernel, the following vulnerability has been resolved:
hwmon: (asus_rog_ryujin) Validate HID report lengths
rog_ryujin_raw_event() parses response headers and payload fields without
first checking that they are present in the received report. A short report
can therefore make the driver consume uninitialized bytes from the HID
transport buffer and expose them as sensor values through sysfs.
Validate the response header and the fields used by each response type
before parsing them. |
| In the Linux kernel, the following vulnerability has been resolved:
mm/migrate_device: clear stale mapping after freeing swapcache
__migrate_device_pages() reads the folio mapping before calling
folio_free_swap(). When folio_free_swap() succeeds, the folio is removed
from the swap cache, but the saved mapping still points to swap_space.
Passing the stale mapping to folio_migrate_mapping() makes it use the
mapped-folio path for a folio that is no longer in swapcache. It can then
operate on swap_space.i_pages with invalid reference accounting,
eventually triggering a folio reference count BUG.
After a successful split, nr still contains the number of pages in the
original large folio, although each resulting page is now a separate
order-0 folio. Reset nr to 1 so each split folio is processed separately,
including its own swapcache removal and mapping lookup.
Refresh the saved mapping after folio_free_swap() so the current folio
state is used during migration. |
| In the Linux kernel, the following vulnerability has been resolved:
netfilter: nf_conntrack_expect: use conntrack GC to reap expectations
This patch replaces the timer API by GC worker approach for
expectations, as it already happened in many other subsystems.
Use the existing conntrack GC worker to iterate over the local list of
expectations in the master conntrack to reap expired expectations.
Check IPS_HELPER_BIT to run GC for expectations, set it on for nft_ct
expectation which nevers sets it. Hold the expectation spinlock while
iterating over the master conntrack expectation list to synchronize with
nf_ct_remove_expectations(). This also performs runtime packet path
garbage collection through the expectation insertion and lookup
functions while walking over one of the chains of the global expectation
hashtables. Unconfirmed conntrack entries are skipped since ct->ext can
be reallocated and dying are skipped since those will be gone soon.
Set on IPS_HELPER_BIT if the helper ct extension is added, then the new
GC worker does not need to bump the ct refcount to check if the ct->ext
helper is available.
This removes the extra bump on the refcount for expectation timers, this
allows to remove several nf_ct_expect_put() calls after the unlink,
after this update only refcount remains at 1 while on the expectation
hashes.
This patch implicitly addresses a race with the existing timer API
allowing an expectation to access a stale exp->master pointer which has
been already released when expectation removal loses races with an
expiring timer, ie. timer_del() reporting false.
Add a new NF_CT_EXPECT_DEAD flag to reap this expectation via GC. This
is needed by nf_conntrack_unexpect_related() which is called in error
paths to invalidate newly created expectations that has been added into
the hashes. These expectactions cannot be inmediately released as GC or
nf_ct_remove_expectations() could race to make it. On expectation
insert, the runtime GC reaps stale expectations before checking the
expectation limit set by policy.
Set current timestamp in nf_ct_expect_alloc(), then add the expectation
policy timeout (or custom timeout specified added on top of this) to
specify the expectation lifetime. |
| In the Linux kernel, the following vulnerability has been resolved:
ASoC: nau8821: Cancel delayed work on component remove
Attempting to unload the driver while a jack detection work is pending
would likely crash the kernel when it is eventually scheduled for
execution:
[ 1984.896308] BUG: unable to handle page fault for address: ffffffffc10c2a20
[...]
[ 1984.896388] Hardware name: Valve Jupiter/Jupiter, BIOS F7A0131 01/30/2024
[ 1984.896396] Workqueue: events nau8821_jdet_work [snd_soc_nau8821]
[ 1984.896414] RIP: 0010:__mutex_lock+0x9f/0x11d0
[...]
[ 1984.896504] Call Trace:
[ 1984.896511] <TASK>
[ 1984.896524] ? snd_soc_dapm_disable_pin+0x26/0x60 [snd_soc_core]
[ 1984.896572] ? snd_soc_dapm_disable_pin+0x26/0x60 [snd_soc_core]
[ 1984.896596] snd_soc_dapm_disable_pin+0x26/0x60 [snd_soc_core]
[ 1984.896622] nau8821_jdet_work+0xeb/0x1e0 [snd_soc_nau8821]
[ 1984.896636] process_one_work+0x211/0x590
[ 1984.896649] ? srso_return_thunk+0x5/0x5f
[ 1984.896670] worker_thread+0x1cd/0x3a0
Cancel unscheduled jdet_work or wait for its execution to finish before
the component driver gets removed. |
| In the Linux kernel, the following vulnerability has been resolved:
gfs2: fix memory leaks in gfs2_fill_super error path
Fix two memory leaks in the gfs2_fill_super() error handling path when
transitioning a filesystem to read-write mode fails.
First leak: kthread objects (thread_struct, task_struct, etc.)
When gfs2_freeze_lock_shared() fails after init_threads() succeeds, the
created kernel threads (logd and quotad) are never destroyed. This
occurs because the fail_per_node label doesn't call
gfs2_destroy_threads().
Second leak: quota bitmap buffer (8192 bytes)
When gfs2_make_fs_rw() fails after gfs2_quota_init() succeeds but
before other operations complete, the allocated quota bitmap is never
freed.
The fix moves thread cleanup to the fail_per_node label to handle all
error paths uniformly. gfs2_destroy_threads() is safe to call
unconditionally as it checks for NULL pointers. Quota cleanup is added
in gfs2_make_fs_rw() to properly handle the withdrawal case where
quota initialization succeeds but the filesystem is then withdrawn.
Thread leak backtrace (gfs2_freeze_lock_shared failure):
unreferenced object 0xffff88801d7bca80 (size 4480):
copy_process+0x3a1/0x4670 kernel/fork.c:2422
kernel_clone+0xf3/0x6e0 kernel/fork.c:2779
kthread_create_on_node+0x100/0x150 kernel/kthread.c:478
init_threads+0xab/0x350 fs/gfs2/ops_fstype.c:611
gfs2_fill_super+0xe5c/0x1240 fs/gfs2/ops_fstype.c:1265
Quota leak backtrace (gfs2_make_fs_rw failure):
unreferenced object 0xffff88812de7c000 (size 8192):
gfs2_quota_init+0xe5/0x820 fs/gfs2/quota.c:1409
gfs2_make_fs_rw+0x7a/0xe0 fs/gfs2/super.c:149
gfs2_fill_super+0xfbb/0x1240 fs/gfs2/ops_fstype.c:1275 |
| In the Linux kernel, the following vulnerability has been resolved:
md/raid5: fix IO hang with degraded array with llbitmap
When llbitmap bit state is still unwritten, any new write should force
rcw, as bitmap_ops->blocks_synced() is checked in handle_stripe_dirtying().
However, later the same check is missing in need_this_block(), causing
stripe to deadloop during handling because handle_stripe() will decide
to go to handle_stripe_fill(), meanwhile need_this_block() always return
0 and nothing is handled. |
| In the Linux kernel, the following vulnerability has been resolved:
hwrng: core - use RCU and work_struct to fix race condition
Currently, hwrng_fill is not cleared until the hwrng_fillfn() thread
exits. Since hwrng_unregister() reads hwrng_fill outside the rng_mutex
lock, a concurrent hwrng_unregister() may call kthread_stop() again on
the same task.
Additionally, if hwrng_unregister() is called immediately after
hwrng_register(), the stopped thread may have never been executed. Thus,
hwrng_fill remains dirty even after hwrng_unregister() returns. In this
case, subsequent calls to hwrng_register() will fail to start new
threads, and hwrng_unregister() will call kthread_stop() on the same
freed task. In both cases, a use-after-free occurs:
refcount_t: addition on 0; use-after-free.
WARNING: ... at lib/refcount.c:25 refcount_warn_saturate+0xec/0x1c0
Call Trace:
kthread_stop+0x181/0x360
hwrng_unregister+0x288/0x380
virtrng_remove+0xe3/0x200
This patch fixes the race by protecting the global hwrng_fill pointer
inside the rng_mutex lock, so that hwrng_fillfn() thread is stopped only
once, and calls to kthread_run() and kthread_stop() are serialized
with the lock held.
To avoid deadlock in hwrng_fillfn() while being stopped with the lock
held, we convert current_rng to RCU, so that get_current_rng() can read
current_rng without holding the lock. To remove the lock from put_rng(),
we also delay the actual cleanup into a work_struct.
Since get_current_rng() no longer returns ERR_PTR values, the IS_ERR()
checks are removed from its callers.
With hwrng_fill protected by the rng_mutex lock, hwrng_fillfn() can no
longer clear hwrng_fill itself. Therefore, if hwrng_fillfn() returns
directly after current_rng is dropped, kthread_stop() would be called on
a freed task_struct later. To fix this, hwrng_fillfn() calls schedule()
now to keep the task alive until being stopped. The kthread_stop() call
is also moved from hwrng_unregister() to drop_current_rng(), ensuring
kthread_stop() is called on all possible paths where current_rng becomes
NULL, so that the thread would not wait forever. |
| In the Linux kernel, the following vulnerability has been resolved:
ext4: fix memory leak in ext4_ext_shift_extents()
In ext4_ext_shift_extents(), if the extent is NULL in the while loop, the
function returns immediately without releasing the path obtained via
ext4_find_extent(), leading to a memory leak.
Fix this by jumping to the out label to ensure the path is properly
released. |
| In the Linux kernel, the following vulnerability has been resolved:
drm/amdgpu: Fix memory leak in amdgpu_acpi_enumerate_xcc()
In amdgpu_acpi_enumerate_xcc(), if amdgpu_acpi_dev_init() returns -ENOMEM,
the function returns directly without releasing the allocated xcc_info,
resulting in a memory leak.
Fix this by ensuring that xcc_info is properly freed in the error paths.
Compile tested only. Issue found using a prototype static analysis tool
and code review. |
| In the Linux kernel, the following vulnerability has been resolved:
tpm: tpm_i2c_infineon: Fix locality leak on get_burstcount() failure
get_burstcount() can return -EBUSY on timeout. When this happens, the
function returns directly without releasing the locality that was
acquired at the beginning of tpm_tis_i2c_send().
Use goto out_err to ensure proper cleanup when get_burstcount() fails. |
| In the Linux kernel, the following vulnerability has been resolved:
net: stmmac: fix oops when split header is enabled
For GMAC4, when split header is enabled, in some rare cases, the
hardware does not fill buf2 of the first descriptor with payload.
Thus we cannot assume buf2 is always fully filled if it is not
the last descriptor. Otherwise, the length of buf2 of the second
descriptor will be calculated wrong and cause an oops:
Unable to handle kernel paging request at virtual address ffff00019246bfc0
...
x2 : 0000000000000040 x1 : ffff00019246bfc0 x0 : ffff00009246c000
Call trace:
dcache_inval_poc+0x28/0x58 (P)
dma_direct_sync_single_for_cpu+0x38/0x6c
__dma_sync_single_for_cpu+0x34/0x6c
stmmac_napi_poll_rx+0x8f0/0xb60
__napi_poll.constprop.0+0x30/0x144
net_rx_action+0x160/0x274
handle_softirqs+0x1b8/0x1fc
...
To fix this, the PL bit-field in RDES3 register is used for all
descriptors, whether it is the last descriptor or not. |
