-
CVE-2026-93287
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
i2c: smbus: reject oversized block transfers in the common path
The SMBus block transfer length data-block[0] is validated in
i2c_smbus_xfer_emulated() but that check runs too late for tracepoints
and is skipped entirely when the adapter provides a native smbus_xfer
implementation. This allows user-controlled oversized block lengths to
reach tracepoint memcpy calls and driver callbacks unchecked.
Add an early validation in __i2c_smbus_xfer() that rejects block
transfers whose caller-supplied length is zero or exceeds
I2C_SMBUS_BLOCK_MAX before any tracepoint fires or driver callback
runs. data->block[0] is filled in by the device on SMBus block reads,
so the check is scoped to operations where the length is actually
supplied by the caller. This is consistent with the existing -EINVAL
convention in the emulated path and protects all downstream consumers
at once: the smbus_write tracepoint, all native smbus_xfer driver
implementations, and the emulated path.
Two distinct bugs are fixed by this change:
Bug 1: smbus_write tracepoint OOB (include/trace/events/smbus.h)
trace_smbus_write() fires before any validation and copies
data->block[0]+1 bytes into a 34-byte event buffer. With
block[0]=0xfe the tracepoint copies 255 bytes, overflowing by 221.
BUG: KASAN: stack-out-of-bounds in trace_event_raw_event_smbus_write+0x27c/0x530
Read of size 255 at addr ffff88800d98fcf8 by task poc_smbus/91
Call Trace:
TASK>
__asan_memcpy+0x23/0x80
trace_event_raw_event_smbus_write+0x27c/0x530
__i2c_smbus_xfer+0x43a/0xa40
i2c_smbus_xfer+0x19e/0x340
i2cdev_ioctl_smbus+0x38f/0x7f0
i2cdev_ioctl+0x35e/0x680
__x64_sys_ioctl+0x147/0x1e0
do_syscall_64+0xcf/0x15a0
entry_SYSCALL_64_after_hwframe+0x76/0x7e
Bug 2: i2c-stub I2C_SMBUS_I2C_BLOCK_DATA OOB (drivers/i2c/i2c-stub.c)
stub_xfer() implements .smbus_xfer directly and only clamps
block[0] against 256-command, not I2C_SMBUS_BLOCK_MAX. With
block[0]=0xff and command=0 the loop accesses block[1+i] for
i up to 254, far past the 34-byte union.
UBSAN: array-index-out-of-bounds in drivers/i2c/i2c-stub.c:223:44
index 34 is out of range for type '__u8 [34]'
Call Trace:
__ubsan_handle_out_of_bounds+0xd7/0x120
stub_xfer+0x1971/0x198f [i2c_stub]
__i2c_smbus_xfer+0x306/0xa40
i2c_smbus_xfer+0x19e/0x340
i2cdev_ioctl_smbus+0x38f/0x7f0
i2cdev_ioctl+0x35e/0x680
__x64_sys_ioctl+0x147/0x1e0
do_syscall_64+0xcf/0x15a0
entry_SYSCALL_64_after_hwframe+0x76/0x7e
Both traces reproduced on v7.0-rc6+i2c/for-current with KASAN+UBSAN.
-
CVE-2026-93286
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
net: appletalk: fix NULL pointer dereference in aarp_send_ddp()
aarp_send_ddp() calls atalk_find_dev_addr(dev) in the LocalTalk fast
path without checking for NULL. When the device has no AppleTalk
interface configured (dev-atalk_ptr == NULL), this leads to a NULL
pointer dereference at the at->s_net access.
KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
RIP: 0010:aarp_send_ddp (net/appletalk/aarp.c:552 (discriminator 2))
Call Trace:
TASK>
atalk_sendmsg (net/appletalk/ddp.c:1715)
__sys_sendto (net/socket.c:2265 (discriminator 1))
__x64_sys_sendto (net/socket.c:2272)
do_syscall_64 (arch/x86/entry/syscall_64.c:94)
entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121)
Add a NULL check consistent with the other callers of
atalk_find_dev_addr().
-
CVE-2026-93277
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
RDMA/bnxt_re: Validate udata before executing commands
The destroy callbacks currently zero the udata output after tearing down
driver resources. If the userspace access fails, uverbs preserves the
uobject and allows the destroy callback to run again, even though the
driver resource has already been freed.
Call ib_no_udata_io() before teardown so udata failures are detected
while the resource is still intact, then return success after teardown
completes.
As part of this change, move ib_respond_empty_udata() to the start of
the create and modify flows. While this is not strictly required for
general create flows, as the core layer unwinds uobjects on failure, it
is necessary for create AH. In _rdma_create_ah(), the HW object is
otherwise leaked.
-
CVE-2026-93276
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
phy: renesas: phy-rcar-gen3-usb2: Fix devm action registration for disabled VBUS regulator
devm_regulator_get_exclusive() initialises the regulator with
enable_count = 1, requiring the consumer to disable it before release.
The devm disable action was previously only registered when the caller
explicitly requested enable, so when the regulator was left in its initial
enabled state without an explicit enable call, the cleanup path skipped
decrementing enable_count, triggering a WARN_ON during regulator
release on device removal.
Fix this by always registering the devm disable action based on the actual
enabled state via regulator_is_enabled(), regardless of whether the
caller requested an explicit enable. This covers both the explicitly
enabled case and the initial state set by devm_regulator_get_exclusive().
-
CVE-2026-93275
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
perf/x86/intel/pt: Fix stop/start with no update
If pt_event_stop() is called without PERF_EF_UPDATE flag, then
perf_aux_output_end() is not called. A subsequent call to pt_event_start()
will call perf_aux_output_begin() again which violates the rule against
nesting and triggers a WARNING in perf_aux_output_begin().
Originally, pt_event_stop() was never called without PERF_EF_UPDATE,
because the only code paths to do so are from event overflow, and Intel PT
does not do that.
However the introduction of group throttling by commit 9734e25fbf5ae
("perf: Fix the throttle logic for a group") meant that an Intel PT event
could be throttled if it was part of a group. Throttling calls PMU
-stop() / ->start() callbacks without flags.
An example is when AUX area sampling is used. The following commands
hit the issue:
echo 10000 > /proc/sys/kernel/perf_event_max_sample_rate
perf record -F32000 --aux-sample -e '{intel_pt//u,cycles:u}' \
-- bash -c 'for i in `seq 1 100000` ; do true ; done'
Use PERF_HES_UPTODATE to track whether perf_aux_output_begin() and
perf_aux_output_end() are balanced. A cleared PERF_HES_UPTODATE bit
indicates that an AUX output context is still open.
Amend pt_event_start() / pt_event_stop() accordingly so that begin/end
stay balanced:
- In non-snapshot mode, stop() always closes the buffer (the buffer may
have run out of space, and that accounting is done by the update), so
a following start() opens a fresh one as before.
- In snapshot/overwrite mode, stop() without PERF_EF_UPDATE leaves the
buffer open so that pt_event_snapshot_aux() can still copy from it,
and start() then only re-enables tracing instead of calling
perf_aux_output_begin() again.
Note that pt_event_del() calls pt_event_stop() with PERF_EF_UPDATE flag set
(as is required by the documentation), so a final call to
perf_aux_output_end() is assured.
-
CVE-2026-93274
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
pinctrl: bcm2835: Don't remove an unregistered GPIO chip
If the devm_pinctrl_register() function fails,
bcm2835_pinctrl_probe() calls gpiochip_remove()
before gpiochip_add_data() has registered the GPIO chip.
This means that upon failure the gpio_chip.gpiodev
is NULL resulting in a null pointer dereference
inside the gpiochip_remove() function.
Remove the unnecessary function call to gpiochip_remove().
No GPIO cleanup is required because the GPIO chip
has not yet been registered. Without this change there
is potential for a kernel panic upon registration failure
-
CVE-2026-93270
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
bpf: Disallow interpreter fallback for BPF_ADDR_PERCPU insn
The BPF_MOV64_PERCPU_REG insn requires JIT to emit native code to for
'dst_reg = src_reg + percpu_base_off'.
However, the interpreter ignores the 'off' at its ALU64_MOV_X label.
The 'off' indicates the insn is BPF_MOV64_PERCPU_REG insn. Then, when
the interpreter loads memory from the register, it will hit a page
fault.
[ 2.545572] BUG: unable to handle page fault for address: ffffffffacaaf034
[ 2.546485] #PF: supervisor read access in kernel mode
[ 2.547167] #PF: error_code(0x0000) - not-present page
[ 2.547850] PGD 134e63067 P4D 134e63067 PUD 134e64063 PMD 10021c063 PTE 800ffffeca550062
[ 2.548912] Oops: Oops: 0000 [#1] SMP PTI
Set jit_required as true in order to disallow interpreter fallback in
core.c::__bpf_prog_select_runtime(), if any BPF_ADDR_PERCPU insn is
patched to the prog.
BTW, rename the helper bpf_map_supports_cpu_flags() to
bpf_map_is_percpu_map().
-
CVE-2026-93265
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
PCI/pwrctrl: tc9563: Fix parsing the integrated Ethernet MAC Endpoint node
DSP3 has an integrated Ethernet MAC Endpoint which has its own set of
config registers for configuring settings such as ASPM. The Endpoint device
has two physical functions and those two functions share the same settings.
Parse the Endpoint node under DSP3 instead of parsing both functions. The
existing parsing logic also has one OOB issue as parsing both functions
will result in accessing past the tc9563_pwrctrl-cfg array.
-
CVE-2026-93260
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
powerpc/xive: propagate IPI init errors to prevent use-after-free
When xive_init_ipis() fails (e.g. irq_domain_alloc_irqs() fails),
the error path frees the global xive_ipis array. However,
xive_smp_probe() previously ignored this failure and proceeded to
call xive_setup_cpu_ipi(), which dereferences the already-freed
xive_ipis pointer -- a use-after-free.
Now that xive_smp_probe() returns int (previous patch), propagate
the error from xive_init_ipis() and xive_setup_cpu_ipi() through
xive_smp_probe(). Check the return value in both pnv_smp_probe()
and pSeries_smp_probe() so that IPI setup is aborted cleanly on
failure, avoiding the use-after-free.
-
CVE-2026-93259
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
powerpc/irq: Fix missing r2 clobber in PCREL inline assembly
In CONFIG_PPC_KERNEL_PCREL mode, r2 is no longer reserved for the TOC
pointer and is available as a caller-saved register [0].
Both call_do_irq() and call_do_softirq() use inline assembly to call
functions with stack switching, but fail to list r2 in their clobber
lists. This causes the compiler to assume r2 is preserved across these
calls, leading to register corruption when the called functions
(__do_irq and __do_softirq) clobber r2.
As a result of this kernel crash during interrupt handling is seen and
the kernel fails to boot:
BUG: Unable to handle kernel data access on write at 0xc000000404697638
Faulting instruction address: 0xc0000000000181ec
Oops: Kernel access of bad area, sig: 11 [#1]
NIP [c0000000000181ec] __do_IRQ+0x6c/0xc0
With older GCC, the compiler would conservatively allocate
callee-saved registers (like r31) for values spanning function calls,
accidentally avoiding the bug:
__do_IRQ:
00 00 00 60 nop
a6 02 08 7c mflr r0
f8 ff e1 fb std r31,-8(r1)
f0 ff c1 fb std r30,-16(r1)
2d 03 10 06 pla r31,53297316
...
3d e8 ff 4b bl c0000000000165ac <__do_irq>
00 00 21 e8 ld r1,0(r1)
28 00 4d e9 ld r10,40(r13)
40 00 21 38 addi r1,r1,64
2a f9 aa 7f stdx r29,r10,r31
With newer GCC 14, the compiler uses r2 for such values, exposing the
missing clobber specification:
<__do_IRQ>:
00 00 00 60 nop
a6 02 08 7c mflr r0
f0 ff c1 fb std r30,-16(r1)
f8 ff e1 fb std r31,-8(r1)
29 02 10 06 pla r2,36252592 # c0000000022aadc0 <__irq_regs>
...
85 dc ff 4b bl c000000000015ee0 <__do_irq>
00 00 21 e8 ld r1,0(r1)
28 00 2d e9 ld r9,40(r13)
30 00 21 38 addi r1,r1,48
2a 11 c9 7f stdx r30,r9,r2
Fix this by adding r2 to the clobber list for both call_do_irq() and
call_do_softirq() when CONFIG_PPC_KERNEL_PCREL is enabled.
[0]: https://www.mail-archive.com/gcc-patches@gcc.gnu.org/msg313226.html
-
CVE-2026-93247
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
Bluetooth: mgmt: fix 'hdev-discovery.uuids' NULL dereference
'uuid_count' member of struct 'discovery_state' is assigned and read
without any locks, so there is a chance of situation when
uuid_count != 0, but uuids is NULL and there will be NULL pointer
dereference.
Possible race:
'hci_update_passive_scan_sync'
'hci_discovery_filter_clear'
hdev->discovery.uuid_count = 0;
----------------------preempted----------------------------->
'start_service_discovery'
// Set uuid_count to value != 0
hdev->discovery.uuid_count = uuid_count;
hdev->discovery.uuids = kmemdup(...);
<----------------------preempted----------------------------->
spin_lock(&hdev->discovery.lock);
kfree(hdev->discovery.uuids);
hdev->discovery.uuids = NULL;
spin_unlock(&hdev->discovery.lock);
Now uuids == NULL and uuid_count != 0.
So 'mgmt_device_found' -> 'is_filter_match' -> 'eir_has_uuids' receives
non consistent discovery state, where NULL dereference of uuids happens.
To fix it let's add discovery.lock around every read/write of uuid_count,
uuids pair of struct members. It is also important to assign uuid_count
value only after success kmemdup() allocation in
start_service_discovery(), otherwise uuids is NULL, because kmemdup failed,
but uuid_count is already assigned to non zero value.
The following panic happens:
[ ] ------------[ cut here ]------------
[ ] Unable to handle kernel NULL pointer dereference at virtual
address 0000000000000000
[ ] Internal error: Oops: 0000000096000006 [#1] PREEMPT SMP
[ ] CPU: 0 PID: 15056 Comm: kworker/u9:2
[ ] Workqueue: hci0 hci_rx_work
[ ] pstate: 10400009 (nzcV daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ ] pc : eir_has_uuids+0x2d8/0x590
[ ] lr : is_filter_match+0x258/0x320
...
[ ] Call trace:
[ ] eir_has_uuids+0x2d8/0x590
[ ] is_filter_match+0x258/0x320
[ ] mgmt_device_found+0x5b0/0xafc
[ ] process_adv_report.part.0+0x8c8/0xf14
[ ] hci_le_adv_report_evt+0x338/0x3f0
[ ] hci_le_meta_evt+0x1f0/0x4c8
[ ] hci_event_packet+0x440/0xc9c
[ ] hci_rx_work+0x44c/0xaf8
[ ] process_one_work+0x54c/0x103c
[ ] worker_thread+0x6c4/0x10c4
[ ] kthread+0x274/0x2ec
[ ] ret_from_fork+0x10/0x20
[ ] Code: 14000004 91004021 eb14003f 54000180 (f9400024)
[ ] ---[ end trace 0000000000000000 ]---
-
CVE-2026-93245
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
apparmor: policy_int make sure list heads are initialized before fail path
If profile create fails before policy_init is complete the list heads
are not properly initialized causing profile_free() sanity checks to
trigger the following splat.
AppArmor WARN aa_policy_destroy: (((!list_empty(&policy-profiles) && (&policy->profiles)->prev != ((void *) 0x122 + (0xdead000000000000UL))))):
WARNING: security/apparmor/lib.c:509 at aa_policy_destroy+0x164/0x1b0 security/apparmor/lib.c:509, CPU#0: syz.0.17/5541
Modules linked in:
CPU: 0 UID: 0 PID: 5541 Comm: syz.0.17 Not tainted syzkaller #0 PREEMPT(full)
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
RIP: 0010:aa_policy_destroy+0x16b/0x1b0 security/apparmor/lib.c:509
Code: 85 ed 7e 4d e8 96 bc 37 fd 5b 41 5c 41 5e 41 5f 5d e9 19 27 4e 07 cc e8 83 bc 37 fd 48 8d 3d 0c f0 d3 0b 48 c7 c6 a4 eb 38 8e 67> 48 0f b9 3a e9 04 ff ff ff e8 66 bc 37 fd 48 8d 3d ff ef d3 0b
RSP: 0018:ffffc9000345eaa0 EFLAGS: 00010293
RAX: ffffffff848f530d RBX: ffff88803f734800 RCX: ffff88801af2a580
RDX: 0000000000000000 RSI: ffffffff8e38eba4 RDI: ffffffff90634320
RBP: 0000000000000000 R08: 0000000000000cc0 R09: 00000000ffffffff
R10: dffffc0000000000 R11: fffffbfff1d95913 R12: dead000000000122
R13: ffff88803f734800 R14: ffff88803f734828 R15: dffffc0000000000
FS: 00007f5f6a1836c0(0000) GS:ffff88808c519000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000055d02407b048 CR3: 0000000012aa9000 CR4: 0000000000352ef0
Call Trace:
aa_free_profile+0x9d/0x9f0 security/apparmor/policy.c:334
aa_alloc_profile+0x1e4/0x3e0 security/apparmor/policy.c:416
unpack_profile security/apparmor/policy_unpack.c:1153 [inline]
aa_unpack+0x17db/0x7430 security/apparmor/policy_unpack.c:1748
aa_replace_profiles+0x226/0x2a20 security/apparmor/policy.c:1183
policy_update+0x234/0x4a0 security/apparmor/apparmorfs.c:505
profile_load+0x1cb/0x320 security/apparmor/apparmorfs.c:522
vfs_write+0x296/0xba0 fs/read_write.c:685
ksys_write+0x150/0x270 fs/read_write.c:739
do_syscall_x64 arch/x86/entry/syscall_64.c:61 [inline]
do_syscall_64+0x166/0x520 arch/x86/entry/syscall_64.c:84
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f5f6939e0d9
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f5f6a183028 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
RAX: ffffffffffffffda RBX: 00007f5f69625fa0 RCX: 00007f5f6939e0d9
RDX: 0000000000000041 RSI: 0000200000000400 RDI: 0000000000000003
RBP: 00007f5f6a183090 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000001
R13: 00007f5f69626038 R14: 00007f5f69625fa0 R15: 00007ffe23725c18
-
CVE-2026-93241
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
memcg: bypass the reclaim and oom killer for dying tasks once oom_reaper is done
At Meta, we are seeing instances where an OOM killed job is stuck in the
exit path for several hours. In one particular case, the job was stuck
for more than 8 hours and I had to manually remove the memory.max limits
to allow the process to exit.
The job was a single process job and had ~55 GiB memory.max and zswap
enabled. It had almost 0 anon in memory and ~111 GiB in zswap compressed
to ~51 GiB zswap pool (i.e. almost all of memory.current was zswap).
Nothing was left on the LRUs to reclaim.
On further inspection, I observed ~20k threads of that process stuck with
the following stack:
[0] mem_cgroup_out_of_memory+0x4e/0xa0
[<0>] charge_memcg+0x8bf/0x990
[<0>] mem_cgroup_swapin_charge_folio+0x4e/0x80
[<0>] __read_swap_cache_async+0x10c/0x260
[<0>] swapin_readahead+0x116/0x3f0
[<0>] do_swap_page+0x13c/0x1ce0
[<0>] handle_mm_fault+0x61d/0x11f0
[<0>] do_user_addr_fault+0x3e7/0x6d0
[<0>] exc_page_fault+0x8f/0x110
[<0>] asm_exc_page_fault+0x22/0x30
[<0>] __get_user_8+0x14/0x20
[<0>] futex_cleanup+0x27/0x1c0
[<0>] futex_exit_release+0x47/0x60
[<0>] do_exit+0x107/0x940
[<0>] do_group_exit+0x81/0xa0
[<0>] get_signal+0x2b1/0x6e0
[<0>] arch_do_signal_or_restart+0x1a/0x1c0
[<0>] exit_to_user_mode_loop+0xa8/0x1c0
[<0>] do_syscall_64+0x152/0x250
[<0>] entry_SYSCALL_64_after_hwframe+0x4b/0x53
In addition the dmesg was filled with "Out of memory and no killable
processes..." messages.
I have no idea why oom reaper was not able to reap/unmap the process. My
guess is that since oom reaper tries to acquire mmap_lock in read mode
limited number of times and then gives up, there might be a thread of that
process which had mmap_lock in write mode at that time.
My initial suspicion was the futex_cleanup and kernel page fault causing
infinite fault and charge retries but that was put to rest in previous
discussions happened on similar problem [1].
My current theory is that it is just a simple slow serialization behind
the oom_lock. Unlike page allocator, memcg charge code takes the oom_lock
without the "try". Though memcg oom code uses mutex_lock_killable(), note
that in the call stack get_signal() consumes SIGKILL (or
sigdelset(SIGKILL)) before calling do_group_exit(). So this
mutex_lock_killable() is just a mutex_lock() here. Therefore 10s of
thousands of threads are waiting on oom_lock and one by one they get
-EFAULT from get_user() in the futex cleanup code and bails out.
Discussion from [1] led to commit a75ffa26122b ("memcg, oom: do not bypass
oom killer for dying tasks") which routes dying tasks into the OOM path
precisely so the oom_reaper can reap their mm and free the memory
asynchronously. But the reaper is best-effort and one-shot: if it cannot
take mmap_lock for read (e.g. a sibling thread holds it for write) it
sets MMF_OOM_SKIP and never retries, leaving only the glacial
oom_lock-serialized synchronous drain.
Once MMF_OOM_SKIP is set there is no more asynchronous reclaim coming for
the mm, so a dying task charging against it has nothing left to wait for:
it frees its memory only once it finishes exiting. Running reclaim and
the (no-victim) OOM killer for it is then pointless, and doing it for 10s
of thousands of exiting threads is what serializes them behind oom_lock.
So before reclaim, if current is an OOM victim whose reaper is done, fail
the charge.
Reproduced with 20k threads, each parking a robust futex head on its own
zswapped page, OOM-group-killed while a sibling holds mmap_lock for write
so the reaper gives up and sets MMF_OOM_SKIP. Tested on next-20260728 and
baseline show ~90 seconds exit time while with the patch the exit time
reduced to ~3 seconds.
-
CVE-2026-93238
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
s390/vfio-ap: fix potential use of uninitialized apm_filtered bitmap
The DECLARE_BITMAP(apm_filtered, AP_DEVICES) macro allocates the bitmap
on the stack without zero-initializing it.
In vfio_ap_mdev_hot_plug_cfg(), the vfio_ap_mdev_filter_matrix() function
is only called to initialize and populate apm_filtered if either
filter_adapters or filter_domains is true. If the hot plug configuration
change only adds control domains (meaning filter_cdoms is true, but
filter_adapters and filter_domains are both false),
vfio_ap_mdev_filter_matrix() is bypassed.
Consequently, apm_filtered is passed to reset_queues_for_apids() with
uninitialized stack garbage. This can cause reset_queues_for_apids() to
interpret arbitrary stack garbage bits as valid APIDs to reset, potentially
performing unintended guest hardware queue resets.
Fix this by zero-initializing the apm_filtered bitmap at the beginning of
vfio_ap_mdev_hot_plug_cfg() using bitmap_zero().
-
CVE-2026-93233
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
drm/nouveau/dmem: fix callocated underflow on large folio split
nouveau_dmem_folio_free() drops chunk-callocated once per freed folio,
while a large (compound) device-private folio is only counted once when
it is allocated. When such a folio is split, the mm core invokes
->folio_split() (nouveau_dmem_folio_split()) once for each new
sub-folio, but the hook only fixes up the sub-folio metadata and leaves
chunk->callocated unchanged.
Each resulting sub-folio is later freed separately, so after a split
the single allocation (+1) is met by N frees (-N), leaving
chunk->callocated short by N-1. On the first split/free cycle it
underflows: WARN_ON(!chunk->callocated) fires, the unsigned counter
wraps and never returns to zero, so the chunk can no longer be
reclaimed (nouveau_dmem_fini() also warns on the leaked count).
Account for the new sub-folio in the split hook, under the same lock as
nouveau_dmem_folio_free(), so the count stays balanced.
-
CVE-2026-93225
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
phy: fsl-imx8mq-usb: fix typec switch leak on probe error path
If probe fails after imx95_usb_phy_get_tca() succeeds, the typec
switch leaks because the only cleanup path was in .remove(), which
never runs on probe failure.
Use devm_add_action_or_reset() so the switch is cleaned up on both
probe failure and driver removal. The imx95_usb_phy_put_tca() is no
longer needed, it will be removed in .remove() too.
-
CVE-2026-93224
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
svcrdma: Fix unmatched rn_unregister on failed accept
When svc_rdma_accept() takes the errout path before
rpcrdma_rn_register() has succeeded, the existing cleanup block
calls rpcrdma_rn_unregister(dev, &newxprt-sc_rn) unconditionally.
svcxprt_rdma is kzalloc'd, so on that path sc_rn.rn_index is 0 and
sc_rn.rn_done is NULL; the unregister therefore xa_erase()s another
caller's slot 0 and performs an unmatched kref_put() on the
rpcrdma_device's rd_kref.
The same errout also brackets the cleanup with svc_xprt_get()/
svc_xprt_put() around the kref_init() birth reference. The kref
goes 1 -> 2 -> 1 and never reaches 0, so the svcxprt_rdma (and the
net/ns_tracker it pinned) is leaked on every failed accept.
rpcrdma_rn_register() writes rn->rn_done last, only after xa_alloc()
and kref_get() have both succeeded, so rn_done == NULL is a natural
"never registered" sentinel. Guard rpcrdma_rn_unregister() with an
early return when rn_done is NULL, and clear rn_done before the
matching xa_erase() so a repeated unregister is also a no-op.
With that guard in place, the accept errout drops the kref_init()
birth reference via svc_xprt_put(), which dispatches svc_rdma_free().
Teardown of sc_qp, sc_sq_cq, sc_rq_cq, and sc_pd runs under existing
IS_ERR/NULL guards in svc_rdma_free(); sc_rn is covered by the new
rn_done sentinel; sc_cm_id is non-NULL on every errout path because
svc_rdma_accept() dereferences it above the first goto errout.
svc_xprt_free() drops the module reference associated with the freed
transport, and svc_handle_xprt() drops its pre-acquired reference
when ->xpo_accept() returns NULL. Take a replacement module reference
before svc_xprt_put() so the two module_put()s remain balanced.
The rn_done guard also covers svc_rdma_free()'s non-listener call
to rpcrdma_rn_unregister() for transports whose register attempt
failed or never ran.
-
CVE-2026-93221
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
nfsd: convert nfsd_net boolean flags to unsigned long flags word
nfsd_net contains several boolean fields that are accessed from
concurrent contexts without serialization. In particular,
nfsd4_end_grace() guards its drain path with a plain bool:
if (nn-grace_ended)
return;
nn->grace_ended = true;
The read and the write are independent, and nothing in struct
nfsd_net serializes them. At least two contexts can reach this
code with no lock held:
laundromat path
laundry_wq kworker
nfs4_laundromat()
nfsd4_end_grace()
RECLAIM_COMPLETE path
nfsd compound kthread
nfsd4_reclaim_complete()
inc_reclaim_complete()
nfsd4_end_grace()
Both callers can observe grace_ended == false on different CPUs,
both store true, and both proceed into nfsd4_record_grace_done(),
which invokes the active client_tracking_ops->grace_done callback.
For tracking ops that drain reclaim_str_hashtbl (legacy_tracking_ops
via nfsd4_recdir_purge_old, and the cld v1+ ops via
nfsd4_cld_grace_done), grace_done calls nfs4_release_reclaim(),
which walks every bucket of reclaim_str_hashtbl with no lock and
calls nfs4_remove_reclaim_record() (list_del + kfree) on each
entry. Two concurrent walkers corrupt the list and double-free
every nfs4_client_reclaim. A concurrent nfsd4_find_reclaim_client()
iterating the same bucket reads through freed memory.
A third call site exists in nfs4_state_start_net() on the
skip_grace startup path, but it runs under nfsd_mutex before any
client has connected and before the laundromat's first delayed
work fires, so it cannot race with the two callers above.
Replace the scattered boolean fields in nfsd_net with a single
unsigned long flags word and an enum nfsd_net_flag for the bit
positions. The grace_ended race is fixed by using
test_and_set_bit(), which is atomic on all architectures. The
remaining flags (grace_end_forced, in_grace, somebody_reclaimed,
track_reclaim_completes, nfsd_net_up, lockd_up) are converted to
use test_bit/set_bit/clear_bit for consistency. This avoids
sub-word cmpxchg issues on architectures like Hexagon that only
support word-sized atomic operations.
-
CVE-2026-93218
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
mm/huge_memory: skip device-private PMDs in madvise_free_huge_pmd
madvise_free_pte_range() checks pmd_trans_huge(*pmd) unlocked, then
madvise_free_huge_pmd() takes pmd_trans_huge_lock(). pmd_is_huge()
returns true for a device-private PMD, so orig_pmd can be device-private
and enter the !pmd_present() branch.
Skip device-private PMDs in that non-present branch and continue to out
before calling pmd_folio(). Downgrade the check to VM_WARN_ON_ONCE() so
an unexpected PMD softleaf logs a warning rather than panicking. Drop the
thp_migration_supported() guard: it expands to
IS_ENABLED(CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF), and both
pmd_is_migration_entry() and pmd_is_device_private_entry() already return
false when that config is not selected, so the guard suppresses only the
case where the warning would already be silent.
Potential trigger: an HMM-based GPU driver races with madvise(MADV_FREE):
migrate_vma_pages() flips the PMD to a device-private entry between the
caller's pmd_trans_huge() check and the callee's pmd_trans_huge_lock().
-
CVE-2026-93217
•
published on September 24, 2026
In the Linux kernel, the following vulnerability has been resolved:
mm/madvise: skip device-private PMDs in cold and pageout walks
madvise_cold_or_pageout_pte_range() takes pmd_trans_huge_lock(), whose
pmd_is_huge() check returns true for a device-private PMD. The subsequent
!pmd_present() branch has a VM_BUG_ON() asserting migration is the only
allowed non-present case; a device-private PMD trips it.
Skip device-private PMDs in that non-present branch and continue to
huge_unlock before calling pmd_folio(). Downgrade the check to
VM_WARN_ON_ONCE() so an unexpected PMD softleaf logs a warning rather than
panicking. Drop the thp_migration_supported() guard: it expands to
IS_ENABLED(CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF), and both
pmd_is_migration_entry() and pmd_is_device_private_entry() already return
false when that config is not selected, so the guard suppresses only the
case where the warning would already be silent.
Potential trigger: an HMM-based GPU driver races with
madvise(MADV_COLD)/MADV_PAGEOUT: pmd_trans_huge(*pmd) reads true, then
migrate_vma_pages() flips the PMD to a device-private entry before the PMD
lock is acquired.