| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| In the Linux kernel, the following vulnerability has been resolved:
block: Use RCU in blk_mq_[un]quiesce_tagset() instead of set->tag_list_lock
blk_mq_{add,del}_queue_tag_set() functions add and remove queues from
tagset, the functions make sure that tagset and queues are marked as
shared when two or more queues are attached to the same tagset.
Initially a tagset starts as unshared and when the number of added
queues reaches two, blk_mq_add_queue_tag_set() marks it as shared along
with all the queues attached to it. When the number of attached queues
drops to 1 blk_mq_del_queue_tag_set() need to mark both the tagset and
the remaining queues as unshared.
Both functions need to freeze current queues in tagset before setting on
unsetting BLK_MQ_F_TAG_QUEUE_SHARED flag. While doing so, both functions
hold set->tag_list_lock mutex, which makes sense as we do not want
queues to be added or deleted in the process. This used to work fine
until commit 98d81f0df70c ("nvme: use blk_mq_[un]quiesce_tagset")
made the nvme driver quiesce tagset instead of quiscing individual
queues. blk_mq_quiesce_tagset() does the job and quiesce the queues in
set->tag_list while holding set->tag_list_lock also.
This results in deadlock between two threads with these stacktraces:
__schedule+0x47c/0xbb0
? timerqueue_add+0x66/0xb0
schedule+0x1c/0xa0
schedule_preempt_disabled+0xa/0x10
__mutex_lock.constprop.0+0x271/0x600
blk_mq_quiesce_tagset+0x25/0xc0
nvme_dev_disable+0x9c/0x250
nvme_timeout+0x1fc/0x520
blk_mq_handle_expired+0x5c/0x90
bt_iter+0x7e/0x90
blk_mq_queue_tag_busy_iter+0x27e/0x550
? __blk_mq_complete_request_remote+0x10/0x10
? __blk_mq_complete_request_remote+0x10/0x10
? __call_rcu_common.constprop.0+0x1c0/0x210
blk_mq_timeout_work+0x12d/0x170
process_one_work+0x12e/0x2d0
worker_thread+0x288/0x3a0
? rescuer_thread+0x480/0x480
kthread+0xb8/0xe0
? kthread_park+0x80/0x80
ret_from_fork+0x2d/0x50
? kthread_park+0x80/0x80
ret_from_fork_asm+0x11/0x20
__schedule+0x47c/0xbb0
? xas_find+0x161/0x1a0
schedule+0x1c/0xa0
blk_mq_freeze_queue_wait+0x3d/0x70
? destroy_sched_domains_rcu+0x30/0x30
blk_mq_update_tag_set_shared+0x44/0x80
blk_mq_exit_queue+0x141/0x150
del_gendisk+0x25a/0x2d0
nvme_ns_remove+0xc9/0x170
nvme_remove_namespaces+0xc7/0x100
nvme_remove+0x62/0x150
pci_device_remove+0x23/0x60
device_release_driver_internal+0x159/0x200
unbind_store+0x99/0xa0
kernfs_fop_write_iter+0x112/0x1e0
vfs_write+0x2b1/0x3d0
ksys_write+0x4e/0xb0
do_syscall_64+0x5b/0x160
entry_SYSCALL_64_after_hwframe+0x4b/0x53
The top stacktrace is showing nvme_timeout() called to handle nvme
command timeout. timeout handler is trying to disable the controller and
as a first step, it needs to blk_mq_quiesce_tagset() to tell blk-mq not
to call queue callback handlers. The thread is stuck waiting for
set->tag_list_lock as it tries to walk the queues in set->tag_list.
The lock is held by the second thread in the bottom stack which is
waiting for one of queues to be frozen. The queue usage counter will
drop to zero after nvme_timeout() finishes, and this will not happen
because the thread will wait for this mutex forever.
Given that [un]quiescing queue is an operation that does not need to
sleep, update blk_mq_[un]quiesce_tagset() to use RCU instead of taking
set->tag_list_lock, update blk_mq_{add,del}_queue_tag_set() to use RCU
safe list operations. Also, delete INIT_LIST_HEAD(&q->tag_set_list)
in blk_mq_del_queue_tag_set() because we can not re-initialize it while
the list is being traversed under RCU. The deleted queue will not be
added/deleted to/from a tagset and it will be freed in blk_free_queue()
after the end of RCU grace period. |
| In the Linux kernel, the following vulnerability has been resolved:
iavf: Implement settime64 with -EOPNOTSUPP
ptp_clock_settime() assumes every ptp_clock has implemented settime64().
Stub it with -EOPNOTSUPP to prevent a NULL dereference.
The fix is similar to commit 329d050bbe63 ("gve: Implement settime64
with -EOPNOTSUPP"). |
| Integer overflow or wraparound in the Linux kernel-mode driver for some Intel(R) 800 Series Ethernet before version 1.17.2 may allow an authenticated user to potentially enable escalation of privilege via local access. |
| In the Linux kernel, the following vulnerability has been resolved:
arm64: mte: Do not warn if the page is already tagged in copy_highpage()
The arm64 copy_highpage() assumes that the destination page is newly
allocated and not MTE-tagged (PG_mte_tagged unset) and warns
accordingly. However, following commit 060913999d7a ("mm: migrate:
support poisoned recover from migrate folio"), folio_mc_copy() is called
before __folio_migrate_mapping(). If the latter fails (-EAGAIN), the
copy will be done again to the same destination page. Since
copy_highpage() already set the PG_mte_tagged flag, this second copy
will warn.
Replace the WARN_ON_ONCE(page already tagged) in the arm64
copy_highpage() with a comment. |
| In the Linux kernel, the following vulnerability has been resolved:
ext4: verify orphan file size is not too big
In principle orphan file can be arbitrarily large. However orphan replay
needs to traverse it all and we also pin all its buffers in memory. Thus
filesystems with absurdly large orphan files can lead to big amounts of
memory consumed. Limit orphan file size to a sane value and also use
kvmalloc() for allocating array of block descriptor structures to avoid
large order allocations for sane but large orphan files. |
| In the Linux kernel, the following vulnerability has been resolved:
dmaengine: xilinx: xdma: Fix regmap max_register
The max_register field is assigned the size of the register memory
region instead of the offset of the last register.
The result is that reading from the regmap via debugfs can cause
a segmentation fault:
tail /sys/kernel/debug/regmap/xdma.1.auto/registers
Unable to handle kernel paging request at virtual address ffff800082f70000
Mem abort info:
ESR = 0x0000000096000007
EC = 0x25: DABT (current EL), IL = 32 bits
SET = 0, FnV = 0
EA = 0, S1PTW = 0
FSC = 0x07: level 3 translation fault
[...]
Call trace:
regmap_mmio_read32le+0x10/0x30
_regmap_bus_reg_read+0x74/0xc0
_regmap_read+0x68/0x198
regmap_read+0x54/0x88
regmap_read_debugfs+0x140/0x380
regmap_map_read_file+0x30/0x48
full_proxy_read+0x68/0xc8
vfs_read+0xcc/0x310
ksys_read+0x7c/0x120
__arm64_sys_read+0x24/0x40
invoke_syscall.constprop.0+0x64/0x108
do_el0_svc+0xb0/0xd8
el0_svc+0x38/0x130
el0t_64_sync_handler+0x120/0x138
el0t_64_sync+0x194/0x198
Code: aa1e03e9 d503201f f9400000 8b214000 (b9400000)
---[ end trace 0000000000000000 ]---
note: tail[1217] exited with irqs disabled
note: tail[1217] exited with preempt_count 1
Segmentation fault |
| In the Linux kernel, the following vulnerability has been resolved:
ring-buffer: Do not warn in ring_buffer_map_get_reader() when reader catches up
The function ring_buffer_map_get_reader() is a bit more strict than the
other get reader functions, and except for certain situations the
rb_get_reader_page() should not return NULL. If it does, it triggers a
warning.
This warning was triggering but after looking at why, it was because
another acceptable situation was happening and it wasn't checked for.
If the reader catches up to the writer and there's still data to be read
on the reader page, then the rb_get_reader_page() will return NULL as
there's no new page to get.
In this situation, the reader page should not be updated and no warning
should trigger. |
| In the Linux kernel, the following vulnerability has been resolved:
drm/virtio: Check whether transferred 2D BO is shmem
Transferred 2D BO always must be a shmem BO. Add check for that to prevent
NULL dereference if userspace passes a VRAM BO. |
| In the Linux kernel, the following vulnerability has been resolved:
mm/memfd: fix information leak in hugetlb folios
When allocating hugetlb folios for memfd, three initialization steps are
missing:
1. Folios are not zeroed, leading to kernel memory disclosure to userspace
2. Folios are not marked uptodate before adding to page cache
3. hugetlb_fault_mutex is not taken before hugetlb_add_to_page_cache()
The memfd allocation path bypasses the normal page fault handler
(hugetlb_no_page) which would handle all of these initialization steps.
This is problematic especially for udmabuf use cases where folios are
pinned and directly accessed by userspace via DMA.
Fix by matching the initialization pattern used in hugetlb_no_page():
- Zero the folio using folio_zero_user() which is optimized for huge pages
- Mark it uptodate with folio_mark_uptodate()
- Take hugetlb_fault_mutex before adding to page cache to prevent races
The folio_zero_user() change also fixes a potential security issue where
uninitialized kernel memory could be disclosed to userspace through read()
or mmap() operations on the memfd. |
| In the Linux kernel, the following vulnerability has been resolved:
tty: serial: imx: disable Ageing Timer interrupt request irq
There maybe pending USR interrupt before requesting irq, however
uart_add_one_port has not executed, so there will be kernel panic:
[ 0.795668] Unable to handle kernel NULL pointer dereference at virtual addre
ss 0000000000000080
[ 0.802701] Mem abort info:
[ 0.805367] ESR = 0x0000000096000004
[ 0.808950] EC = 0x25: DABT (current EL), IL = 32 bits
[ 0.814033] SET = 0, FnV = 0
[ 0.816950] EA = 0, S1PTW = 0
[ 0.819950] FSC = 0x04: level 0 translation fault
[ 0.824617] Data abort info:
[ 0.827367] ISV = 0, ISS = 0x00000004
[ 0.831033] CM = 0, WnR = 0
[ 0.833866] [0000000000000080] user address but active_mm is swapper
[ 0.839951] Internal error: Oops: 0000000096000004 [#1] PREEMPT SMP
[ 0.845953] Modules linked in:
[ 0.848869] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 6.1.1+g56321e101aca #1
[ 0.855617] Hardware name: Freescale i.MX8MP EVK (DT)
[ 0.860452] pstate: 000000c5 (nzcv daIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[ 0.867117] pc : __imx_uart_rxint.constprop.0+0x11c/0x2c0
[ 0.872283] lr : imx_uart_int+0xf8/0x1ec
The issue only happends in the inmate linux when Jailhouse hypervisor
enabled. The test procedure is:
while true; do
jailhouse enable imx8mp.cell
jailhouse cell linux xxxx
sleep 10
jailhouse cell destroy 1
jailhouse disable
sleep 5
done
And during the upper test, press keys to the 2nd linux console.
When `jailhouse cell destroy 1`, the 2nd linux has no chance to put
the uart to a quiese state, so USR1/2 may has pending interrupts. Then
when `jailhosue cell linux xx` to start 2nd linux again, the issue
trigger.
In order to disable irqs before requesting them, both UCR1 and UCR2 irqs
should be disabled, so here fix that, disable the Ageing Timer interrupt
in UCR2 as UCR1 does. |
| In the Linux kernel, the following vulnerability has been resolved:
media: b2c2: Fix use-after-free causing by irq_check_work in flexcop_pci_remove
The original code uses cancel_delayed_work() in flexcop_pci_remove(), which
does not guarantee that the delayed work item irq_check_work has fully
completed if it was already running. This leads to use-after-free scenarios
where flexcop_pci_remove() may free the flexcop_device while irq_check_work
is still active and attempts to dereference the device.
A typical race condition is illustrated below:
CPU 0 (remove) | CPU 1 (delayed work callback)
flexcop_pci_remove() | flexcop_pci_irq_check_work()
cancel_delayed_work() |
flexcop_device_kfree(fc_pci->fc_dev) |
| fc = fc_pci->fc_dev; // UAF
This is confirmed by a KASAN report:
==================================================================
BUG: KASAN: slab-use-after-free in __run_timer_base.part.0+0x7d7/0x8c0
Write of size 8 at addr ffff8880093aa8c8 by task bash/135
...
Call Trace:
<IRQ>
dump_stack_lvl+0x55/0x70
print_report+0xcf/0x610
? __run_timer_base.part.0+0x7d7/0x8c0
kasan_report+0xb8/0xf0
? __run_timer_base.part.0+0x7d7/0x8c0
__run_timer_base.part.0+0x7d7/0x8c0
? __pfx___run_timer_base.part.0+0x10/0x10
? __pfx_read_tsc+0x10/0x10
? ktime_get+0x60/0x140
? lapic_next_event+0x11/0x20
? clockevents_program_event+0x1d4/0x2a0
run_timer_softirq+0xd1/0x190
handle_softirqs+0x16a/0x550
irq_exit_rcu+0xaf/0xe0
sysvec_apic_timer_interrupt+0x70/0x80
</IRQ>
...
Allocated by task 1:
kasan_save_stack+0x24/0x50
kasan_save_track+0x14/0x30
__kasan_kmalloc+0x7f/0x90
__kmalloc_noprof+0x1be/0x460
flexcop_device_kmalloc+0x54/0xe0
flexcop_pci_probe+0x1f/0x9d0
local_pci_probe+0xdc/0x190
pci_device_probe+0x2fe/0x470
really_probe+0x1ca/0x5c0
__driver_probe_device+0x248/0x310
driver_probe_device+0x44/0x120
__driver_attach+0xd2/0x310
bus_for_each_dev+0xed/0x170
bus_add_driver+0x208/0x500
driver_register+0x132/0x460
do_one_initcall+0x89/0x300
kernel_init_freeable+0x40d/0x720
kernel_init+0x1a/0x150
ret_from_fork+0x10c/0x1a0
ret_from_fork_asm+0x1a/0x30
Freed by task 135:
kasan_save_stack+0x24/0x50
kasan_save_track+0x14/0x30
kasan_save_free_info+0x3a/0x60
__kasan_slab_free+0x3f/0x50
kfree+0x137/0x370
flexcop_device_kfree+0x32/0x50
pci_device_remove+0xa6/0x1d0
device_release_driver_internal+0xf8/0x210
pci_stop_bus_device+0x105/0x150
pci_stop_and_remove_bus_device_locked+0x15/0x30
remove_store+0xcc/0xe0
kernfs_fop_write_iter+0x2c3/0x440
vfs_write+0x871/0xd70
ksys_write+0xee/0x1c0
do_syscall_64+0xac/0x280
entry_SYSCALL_64_after_hwframe+0x77/0x7f
...
Replace cancel_delayed_work() with cancel_delayed_work_sync() to ensure
that the delayed work item is properly canceled and any executing delayed
work has finished before the device memory is deallocated.
This bug was initially identified through static analysis. To reproduce
and test it, I simulated the B2C2 FlexCop PCI device in QEMU and introduced
artificial delays within the flexcop_pci_irq_check_work() function to
increase the likelihood of triggering the bug. |
| Uncontrolled resource consumption in the Linux kernel-mode driver for some Intel(R) 700 Series Ethernet before version 2.28.5 may allow an authenticated user to potentially enable denial of service. |
| In the Linux kernel, the following vulnerability has been resolved:
bpf: Fix issue in verifying allow_ptr_leaks
After we converted the capabilities of our networking-bpf program from
cap_sys_admin to cap_net_admin+cap_bpf, our networking-bpf program
failed to start. Because it failed the bpf verifier, and the error log
is "R3 pointer comparison prohibited".
A simple reproducer as follows,
SEC("cls-ingress")
int ingress(struct __sk_buff *skb)
{
struct iphdr *iph = (void *)(long)skb->data + sizeof(struct ethhdr);
if ((long)(iph + 1) > (long)skb->data_end)
return TC_ACT_STOLEN;
return TC_ACT_OK;
}
Per discussion with Yonghong and Alexei [1], comparison of two packet
pointers is not a pointer leak. This patch fixes it.
Our local kernel is 6.1.y and we expect this fix to be backported to
6.1.y, so stable is CCed.
[1]. https://lore.kernel.org/bpf/CAADnVQ+Nmspr7Si+pxWn8zkE7hX-7s93ugwC+94aXSy4uQ9vBg@mail.gmail.com/ |
| In the Linux kernel, the following vulnerability has been resolved:
netdevsim: fix memory leak in nsim_bus_dev_new()
If device_register() failed in nsim_bus_dev_new(), the value of reference
in nsim_bus_dev->dev is 1. obj->name in nsim_bus_dev->dev will not be
released.
unreferenced object 0xffff88810352c480 (size 16):
comm "echo", pid 5691, jiffies 4294945921 (age 133.270s)
hex dump (first 16 bytes):
6e 65 74 64 65 76 73 69 6d 31 00 00 00 00 00 00 netdevsim1......
backtrace:
[<000000005e2e5e26>] __kmalloc_node_track_caller+0x3a/0xb0
[<0000000094ca4fc8>] kvasprintf+0xc3/0x160
[<00000000aad09bcc>] kvasprintf_const+0x55/0x180
[<000000009bac868d>] kobject_set_name_vargs+0x56/0x150
[<000000007c1a5d70>] dev_set_name+0xbb/0xf0
[<00000000ad0d126b>] device_add+0x1f8/0x1cb0
[<00000000c222ae24>] new_device_store+0x3b6/0x5e0
[<0000000043593421>] bus_attr_store+0x72/0xa0
[<00000000cbb1833a>] sysfs_kf_write+0x106/0x160
[<00000000d0dedb8a>] kernfs_fop_write_iter+0x3a8/0x5a0
[<00000000770b66e2>] vfs_write+0x8f0/0xc80
[<0000000078bb39be>] ksys_write+0x106/0x210
[<00000000005e55a4>] do_syscall_64+0x35/0x80
[<00000000eaa40bbc>] entry_SYSCALL_64_after_hwframe+0x46/0xb0 |
| In the Linux kernel, the following vulnerability has been resolved:
iavf: use internal state to free traffic IRQs
If the system tries to close the netdev while iavf_reset_task() is
running, __LINK_STATE_START will be cleared and netif_running() will
return false in iavf_reinit_interrupt_scheme(). This will result in
iavf_free_traffic_irqs() not being called and a leak as follows:
[7632.489326] remove_proc_entry: removing non-empty directory 'irq/999', leaking at least 'iavf-enp24s0f0v0-TxRx-0'
[7632.490214] WARNING: CPU: 0 PID: 10 at fs/proc/generic.c:718 remove_proc_entry+0x19b/0x1b0
is shown when pci_disable_msix() is later called. Fix by using the
internal adapter state. The traffic IRQs will always exist if
state == __IAVF_RUNNING. |
| In the Linux kernel, the following vulnerability has been resolved:
backlight: led-bl: Add devlink to supplier LEDs
LED Backlight is a consumer of one or multiple LED class devices, but
devlink is currently unable to create correct supplier-producer links when
the supplier is a class device. It creates instead a link where the
supplier is the parent of the expected device.
One consequence is that removal order is not correctly enforced.
Issues happen for example with the following sections in a device tree
overlay:
// An LED driver chip
pca9632@62 {
compatible = "nxp,pca9632";
reg = <0x62>;
// ...
addon_led_pwm: led-pwm@3 {
reg = <3>;
label = "addon:led:pwm";
};
};
backlight-addon {
compatible = "led-backlight";
leds = <&addon_led_pwm>;
brightness-levels = <255>;
default-brightness-level = <255>;
};
In this example, the devlink should be created between the backlight-addon
(consumer) and the pca9632@62 (supplier). Instead it is created between the
backlight-addon (consumer) and the parent of the pca9632@62, which is
typically the I2C bus adapter.
On removal of the above overlay, the LED driver can be removed before the
backlight device, resulting in:
Unable to handle kernel NULL pointer dereference at virtual address 0000000000000010
...
Call trace:
led_put+0xe0/0x140
devm_led_release+0x6c/0x98
Another way to reproduce the bug without any device tree overlays is
unbinding the LED class device (pca9632@62) before unbinding the consumer
(backlight-addon):
echo 11-0062 >/sys/bus/i2c/drivers/leds-pca963x/unbind
echo ...backlight-dock >/sys/bus/platform/drivers/led-backlight/unbind
Fix by adding a devlink between the consuming led-backlight device and the
supplying LED device, as other drivers and subsystems do as well. |
| In the Linux kernel, the following vulnerability has been resolved:
wifi: ath11k: fix peer HE MCS assignment
In ath11k_wmi_send_peer_assoc_cmd(), peer's transmit MCS is sent to
firmware as receive MCS while peer's receive MCS sent as transmit MCS,
which goes against firmwire's definition.
While connecting to a misbehaved AP that advertises 0xffff (meaning not
supported) for 160 MHz transmit MCS map, firmware crashes due to 0xffff
is assigned to he_mcs->rx_mcs_set field.
Ext Tag: HE Capabilities
[...]
Supported HE-MCS and NSS Set
[...]
Rx and Tx MCS Maps 160 MHz
[...]
Tx HE-MCS Map 160 MHz: 0xffff
Swap the assignment to fix this issue.
As the HE rate control mask is meant to limit our own transmit MCS, it
needs to go via he_mcs->rx_mcs_set field. With the aforementioned swapping
done, change is needed as well to apply it to the peer's receive MCS.
Tested-on: WCN6855 hw2.1 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.41
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.4.1-00199-QCAHKSWPL_SILICONZ-1 |
| In the Linux kernel, the following vulnerability has been resolved:
comedi: pcl818: fix null-ptr-deref in pcl818_ai_cancel()
Syzbot identified an issue [1] in pcl818_ai_cancel(), which stems from
the fact that in case of early device detach via pcl818_detach(),
subdevice dev->read_subdev may not have initialized its pointer to
&struct comedi_async as intended. Thus, any such dereferencing of
&s->async->cmd will lead to general protection fault and kernel crash.
Mitigate this problem by removing a call to pcl818_ai_cancel() from
pcl818_detach() altogether. This way, if the subdevice setups its
support for async commands, everything async-related will be
handled via subdevice's own ->cancel() function in
comedi_device_detach_locked() even before pcl818_detach(). If no
support for asynchronous commands is provided, there is no need
to cancel anything either.
[1] Syzbot crash:
Oops: general protection fault, probably for non-canonical address 0xdffffc0000000005: 0000 [#1] SMP KASAN PTI
KASAN: null-ptr-deref in range [0x0000000000000028-0x000000000000002f]
CPU: 1 UID: 0 PID: 6050 Comm: syz.0.18 Not tainted syzkaller #0 PREEMPT(full)
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 08/18/2025
RIP: 0010:pcl818_ai_cancel+0x69/0x3f0 drivers/comedi/drivers/pcl818.c:762
...
Call Trace:
<TASK>
pcl818_detach+0x66/0xd0 drivers/comedi/drivers/pcl818.c:1115
comedi_device_detach_locked+0x178/0x750 drivers/comedi/drivers.c:207
do_devconfig_ioctl drivers/comedi/comedi_fops.c:848 [inline]
comedi_unlocked_ioctl+0xcde/0x1020 drivers/comedi/comedi_fops.c:2178
vfs_ioctl fs/ioctl.c:51 [inline]
__do_sys_ioctl fs/ioctl.c:597 [inline]
... |
| In the Linux kernel, the following vulnerability has been resolved:
usb: storage: Fix memory leak in USB bulk transport
A kernel memory leak was identified by the 'ioctl_sg01' test from Linux
Test Project (LTP). The following bytes were mainly observed: 0x53425355.
When USB storage devices incorrectly skip the data phase with status data,
the code extracts/validates the CSW from the sg buffer, but fails to clear
it afterwards. This leaves status protocol data in srb's transfer buffer,
such as the US_BULK_CS_SIGN 'USBS' signature observed here. Thus, this can
lead to USB protocols leaks to user space through SCSI generic (/dev/sg*)
interfaces, such as the one seen here when the LTP test requested 512 KiB.
Fix the leak by zeroing the CSW data in srb's transfer buffer immediately
after the validation of devices that skip data phase.
Note: Differently from CVE-2018-1000204, which fixed a big leak by zero-
ing pages at allocation time, this leak occurs after allocation, when USB
protocol data is written to already-allocated sg pages. |
| In the Linux kernel, the following vulnerability has been resolved:
KVM: SVM: Don't skip unrelated instruction if INT3/INTO is replaced
When re-injecting a soft interrupt from an INT3, INT0, or (select) INTn
instruction, discard the exception and retry the instruction if the code
stream is changed (e.g. by a different vCPU) between when the CPU
executes the instruction and when KVM decodes the instruction to get the
next RIP.
As effectively predicted by commit 6ef88d6e36c2 ("KVM: SVM: Re-inject
INT3/INTO instead of retrying the instruction"), failure to verify that
the correct INTn instruction was decoded can effectively clobber guest
state due to decoding the wrong instruction and thus specifying the
wrong next RIP.
The bug most often manifests as "Oops: int3" panics on static branch
checks in Linux guests. Enabling or disabling a static branch in Linux
uses the kernel's "text poke" code patching mechanism. To modify code
while other CPUs may be executing that code, Linux (temporarily)
replaces the first byte of the original instruction with an int3 (opcode
0xcc), then patches in the new code stream except for the first byte,
and finally replaces the int3 with the first byte of the new code
stream. If a CPU hits the int3, i.e. executes the code while it's being
modified, then the guest kernel must look up the RIP to determine how to
handle the #BP, e.g. by emulating the new instruction. If the RIP is
incorrect, then this lookup fails and the guest kernel panics.
The bug reproduces almost instantly by hacking the guest kernel to
repeatedly check a static branch[1] while running a drgn script[2] on
the host to constantly swap out the memory containing the guest's TSS.
[1]: https://gist.github.com/osandov/44d17c51c28c0ac998ea0334edf90b5a
[2]: https://gist.github.com/osandov/10e45e45afa29b11e0c7209247afc00b |