| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| In the Linux kernel, the following vulnerability has been resolved:
btrfs: prevent copying too big compressed lzo segment
Compressed length can be corrupted to be a lot larger than memory
we have allocated for buffer.
This will cause memcpy in copy_compressed_segment to write outside
of allocated memory.
This mostly results in stuck read syscall but sometimes when using
btrfs send can get #GP
kernel: general protection fault, probably for non-canonical address 0x841551d5c1000: 0000 [#1] PREEMPT SMP NOPTI
kernel: CPU: 17 PID: 264 Comm: kworker/u256:7 Tainted: P OE 5.17.0-rc2-1 #12
kernel: Workqueue: btrfs-endio btrfs_work_helper [btrfs]
kernel: RIP: 0010:lzo_decompress_bio (./include/linux/fortify-string.h:225 fs/btrfs/lzo.c:322 fs/btrfs/lzo.c:394) btrfs
Code starting with the faulting instruction
===========================================
0:* 48 8b 06 mov (%rsi),%rax <-- trapping instruction
3: 48 8d 79 08 lea 0x8(%rcx),%rdi
7: 48 83 e7 f8 and $0xfffffffffffffff8,%rdi
b: 48 89 01 mov %rax,(%rcx)
e: 44 89 f0 mov %r14d,%eax
11: 48 8b 54 06 f8 mov -0x8(%rsi,%rax,1),%rdx
kernel: RSP: 0018:ffffb110812efd50 EFLAGS: 00010212
kernel: RAX: 0000000000001000 RBX: 000000009ca264c8 RCX: ffff98996e6d8ff8
kernel: RDX: 0000000000000064 RSI: 000841551d5c1000 RDI: ffffffff9500435d
kernel: RBP: ffff989a3be856c0 R08: 0000000000000000 R09: 0000000000000000
kernel: R10: 0000000000000000 R11: 0000000000001000 R12: ffff98996e6d8000
kernel: R13: 0000000000000008 R14: 0000000000001000 R15: 000841551d5c1000
kernel: FS: 0000000000000000(0000) GS:ffff98a09d640000(0000) knlGS:0000000000000000
kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
kernel: CR2: 00001e9f984d9ea8 CR3: 000000014971a000 CR4: 00000000003506e0
kernel: Call Trace:
kernel: <TASK>
kernel: end_compressed_bio_read (fs/btrfs/compression.c:104 fs/btrfs/compression.c:1363 fs/btrfs/compression.c:323) btrfs
kernel: end_workqueue_fn (fs/btrfs/disk-io.c:1923) btrfs
kernel: btrfs_work_helper (fs/btrfs/async-thread.c:326) btrfs
kernel: process_one_work (./arch/x86/include/asm/jump_label.h:27 ./include/linux/jump_label.h:212 ./include/trace/events/workqueue.h:108 kernel/workqueue.c:2312)
kernel: worker_thread (./include/linux/list.h:292 kernel/workqueue.c:2455)
kernel: ? process_one_work (kernel/workqueue.c:2397)
kernel: kthread (kernel/kthread.c:377)
kernel: ? kthread_complete_and_exit (kernel/kthread.c:332)
kernel: ret_from_fork (arch/x86/entry/entry_64.S:301)
kernel: </TASK> |
| In the Linux kernel, the following vulnerability has been resolved:
cifs: fix double free race when mount fails in cifs_get_root()
When cifs_get_root() fails during cifs_smb3_do_mount() we call
deactivate_locked_super() which eventually will call delayed_free() which
will free the context.
In this situation we should not proceed to enter the out: section in
cifs_smb3_do_mount() and free the same resources a second time.
[Thu Feb 10 12:59:06 2022] BUG: KASAN: use-after-free in rcu_cblist_dequeue+0x32/0x60
[Thu Feb 10 12:59:06 2022] Read of size 8 at addr ffff888364f4d110 by task swapper/1/0
[Thu Feb 10 12:59:06 2022] CPU: 1 PID: 0 Comm: swapper/1 Tainted: G OE 5.17.0-rc3+ #4
[Thu Feb 10 12:59:06 2022] Hardware name: Microsoft Corporation Virtual Machine/Virtual Machine, BIOS Hyper-V UEFI Release v4.0 12/17/2019
[Thu Feb 10 12:59:06 2022] Call Trace:
[Thu Feb 10 12:59:06 2022] <IRQ>
[Thu Feb 10 12:59:06 2022] dump_stack_lvl+0x5d/0x78
[Thu Feb 10 12:59:06 2022] print_address_description.constprop.0+0x24/0x150
[Thu Feb 10 12:59:06 2022] ? rcu_cblist_dequeue+0x32/0x60
[Thu Feb 10 12:59:06 2022] kasan_report.cold+0x7d/0x117
[Thu Feb 10 12:59:06 2022] ? rcu_cblist_dequeue+0x32/0x60
[Thu Feb 10 12:59:06 2022] __asan_load8+0x86/0xa0
[Thu Feb 10 12:59:06 2022] rcu_cblist_dequeue+0x32/0x60
[Thu Feb 10 12:59:06 2022] rcu_core+0x547/0xca0
[Thu Feb 10 12:59:06 2022] ? call_rcu+0x3c0/0x3c0
[Thu Feb 10 12:59:06 2022] ? __this_cpu_preempt_check+0x13/0x20
[Thu Feb 10 12:59:06 2022] ? lock_is_held_type+0xea/0x140
[Thu Feb 10 12:59:06 2022] rcu_core_si+0xe/0x10
[Thu Feb 10 12:59:06 2022] __do_softirq+0x1d4/0x67b
[Thu Feb 10 12:59:06 2022] __irq_exit_rcu+0x100/0x150
[Thu Feb 10 12:59:06 2022] irq_exit_rcu+0xe/0x30
[Thu Feb 10 12:59:06 2022] sysvec_hyperv_stimer0+0x9d/0xc0
...
[Thu Feb 10 12:59:07 2022] Freed by task 58179:
[Thu Feb 10 12:59:07 2022] kasan_save_stack+0x26/0x50
[Thu Feb 10 12:59:07 2022] kasan_set_track+0x25/0x30
[Thu Feb 10 12:59:07 2022] kasan_set_free_info+0x24/0x40
[Thu Feb 10 12:59:07 2022] ____kasan_slab_free+0x137/0x170
[Thu Feb 10 12:59:07 2022] __kasan_slab_free+0x12/0x20
[Thu Feb 10 12:59:07 2022] slab_free_freelist_hook+0xb3/0x1d0
[Thu Feb 10 12:59:07 2022] kfree+0xcd/0x520
[Thu Feb 10 12:59:07 2022] cifs_smb3_do_mount+0x149/0xbe0 [cifs]
[Thu Feb 10 12:59:07 2022] smb3_get_tree+0x1a0/0x2e0 [cifs]
[Thu Feb 10 12:59:07 2022] vfs_get_tree+0x52/0x140
[Thu Feb 10 12:59:07 2022] path_mount+0x635/0x10c0
[Thu Feb 10 12:59:07 2022] __x64_sys_mount+0x1bf/0x210
[Thu Feb 10 12:59:07 2022] do_syscall_64+0x5c/0xc0
[Thu Feb 10 12:59:07 2022] entry_SYSCALL_64_after_hwframe+0x44/0xae
[Thu Feb 10 12:59:07 2022] Last potentially related work creation:
[Thu Feb 10 12:59:07 2022] kasan_save_stack+0x26/0x50
[Thu Feb 10 12:59:07 2022] __kasan_record_aux_stack+0xb6/0xc0
[Thu Feb 10 12:59:07 2022] kasan_record_aux_stack_noalloc+0xb/0x10
[Thu Feb 10 12:59:07 2022] call_rcu+0x76/0x3c0
[Thu Feb 10 12:59:07 2022] cifs_umount+0xce/0xe0 [cifs]
[Thu Feb 10 12:59:07 2022] cifs_kill_sb+0xc8/0xe0 [cifs]
[Thu Feb 10 12:59:07 2022] deactivate_locked_super+0x5d/0xd0
[Thu Feb 10 12:59:07 2022] cifs_smb3_do_mount+0xab9/0xbe0 [cifs]
[Thu Feb 10 12:59:07 2022] smb3_get_tree+0x1a0/0x2e0 [cifs]
[Thu Feb 10 12:59:07 2022] vfs_get_tree+0x52/0x140
[Thu Feb 10 12:59:07 2022] path_mount+0x635/0x10c0
[Thu Feb 10 12:59:07 2022] __x64_sys_mount+0x1bf/0x210
[Thu Feb 10 12:59:07 2022] do_syscall_64+0x5c/0xc0
[Thu Feb 10 12:59:07 2022] entry_SYSCALL_64_after_hwframe+0x44/0xae |
| In the Linux kernel, the following vulnerability has been resolved:
netfilter: nf_queue: fix possible use-after-free
Eric Dumazet says:
The sock_hold() side seems suspect, because there is no guarantee
that sk_refcnt is not already 0.
On failure, we cannot queue the packet and need to indicate an
error. The packet will be dropped by the caller.
v2: split skb prefetch hunk into separate change |
| In the Linux kernel, the following vulnerability has been resolved:
drm/virtio: Fix GEM handle creation UAF
Userspace can guess the handle value and try to race GEM object creation
with handle close, resulting in a use-after-free if we dereference the
object after dropping the handle's reference. For that reason, dropping
the handle's reference must be done *after* we are done dereferencing
the object. |
| In the Linux kernel, the following vulnerability has been resolved:
drm/vmwgfx: Remove rcu locks from user resources
User resource lookups used rcu to avoid two extra atomics. Unfortunately
the rcu paths were buggy and it was easy to make the driver crash by
submitting command buffers from two different threads. Because the
lookups never show up in performance profiles replace them with a
regular spin lock which fixes the races in accesses to those shared
resources.
Fixes kernel oops'es in IGT's vmwgfx execution_buffer stress test and
seen crashes with apps using shared resources. |
| In the Linux kernel, the following vulnerability has been resolved:
misc: fastrpc: Fix use-after-free race condition for maps
It is possible that in between calling fastrpc_map_get() until
map->fl->lock is taken in fastrpc_free_map(), another thread can call
fastrpc_map_lookup() and get a reference to a map that is about to be
deleted.
Rewrite fastrpc_map_get() to only increase the reference count of a map
if it's non-zero. Propagate this to callers so they can know if a map is
about to be deleted.
Fixes this warning:
refcount_t: addition on 0; use-after-free.
WARNING: CPU: 5 PID: 10100 at lib/refcount.c:25 refcount_warn_saturate
...
Call trace:
refcount_warn_saturate
[fastrpc_map_get inlined]
[fastrpc_map_lookup inlined]
fastrpc_map_create
fastrpc_internal_invoke
fastrpc_device_ioctl
__arm64_sys_ioctl
invoke_syscall |
| In the Linux kernel, the following vulnerability has been resolved:
tty: serial: qcom-geni-serial: fix slab-out-of-bounds on RX FIFO buffer
Driver's probe allocates memory for RX FIFO (port->rx_fifo) based on
default RX FIFO depth, e.g. 16. Later during serial startup the
qcom_geni_serial_port_setup() updates the RX FIFO depth
(port->rx_fifo_depth) to match real device capabilities, e.g. to 32.
The RX UART handle code will read "port->rx_fifo_depth" number of words
into "port->rx_fifo" buffer, thus exceeding the bounds. This can be
observed in certain configurations with Qualcomm Bluetooth HCI UART
device and KASAN:
Bluetooth: hci0: QCA Product ID :0x00000010
Bluetooth: hci0: QCA SOC Version :0x400a0200
Bluetooth: hci0: QCA ROM Version :0x00000200
Bluetooth: hci0: QCA Patch Version:0x00000d2b
Bluetooth: hci0: QCA controller version 0x02000200
Bluetooth: hci0: QCA Downloading qca/htbtfw20.tlv
bluetooth hci0: Direct firmware load for qca/htbtfw20.tlv failed with error -2
Bluetooth: hci0: QCA Failed to request file: qca/htbtfw20.tlv (-2)
Bluetooth: hci0: QCA Failed to download patch (-2)
==================================================================
BUG: KASAN: slab-out-of-bounds in handle_rx_uart+0xa8/0x18c
Write of size 4 at addr ffff279347d578c0 by task swapper/0/0
CPU: 0 PID: 0 Comm: swapper/0 Not tainted 6.1.0-rt5-00350-gb2450b7e00be-dirty #26
Hardware name: Qualcomm Technologies, Inc. Robotics RB5 (DT)
Call trace:
dump_backtrace.part.0+0xe0/0xf0
show_stack+0x18/0x40
dump_stack_lvl+0x8c/0xb8
print_report+0x188/0x488
kasan_report+0xb4/0x100
__asan_store4+0x80/0xa4
handle_rx_uart+0xa8/0x18c
qcom_geni_serial_handle_rx+0x84/0x9c
qcom_geni_serial_isr+0x24c/0x760
__handle_irq_event_percpu+0x108/0x500
handle_irq_event+0x6c/0x110
handle_fasteoi_irq+0x138/0x2cc
generic_handle_domain_irq+0x48/0x64
If the RX FIFO depth changes after probe, be sure to resize the buffer. |
| In the Linux kernel, the following vulnerability has been resolved:
vdpa/mlx5: add validation for VIRTIO_NET_CTRL_MQ_VQ_PAIRS_SET command
When control vq receives a VIRTIO_NET_CTRL_MQ_VQ_PAIRS_SET command
request from the driver, presently there is no validation against the
number of queue pairs to configure, or even if multiqueue had been
negotiated or not is unverified. This may lead to kernel panic due to
uninitialized resource for the queues were there any bogus request
sent down by untrusted driver. Tie up the loose ends there. |
| In the Linux kernel, the following vulnerability has been resolved:
staging: gdm724x: fix use after free in gdm_lte_rx()
The netif_rx_ni() function frees the skb so we can't dereference it to
save the skb->len. |
| In the Linux kernel, the following vulnerability has been resolved:
net/packet: fix slab-out-of-bounds access in packet_recvmsg()
syzbot found that when an AF_PACKET socket is using PACKET_COPY_THRESH
and mmap operations, tpacket_rcv() is queueing skbs with
garbage in skb->cb[], triggering a too big copy [1]
Presumably, users of af_packet using mmap() already gets correct
metadata from the mapped buffer, we can simply make sure
to clear 12 bytes that might be copied to user space later.
BUG: KASAN: stack-out-of-bounds in memcpy include/linux/fortify-string.h:225 [inline]
BUG: KASAN: stack-out-of-bounds in packet_recvmsg+0x56c/0x1150 net/packet/af_packet.c:3489
Write of size 165 at addr ffffc9000385fb78 by task syz-executor233/3631
CPU: 0 PID: 3631 Comm: syz-executor233 Not tainted 5.17.0-rc7-syzkaller-02396-g0b3660695e80 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106
print_address_description.constprop.0.cold+0xf/0x336 mm/kasan/report.c:255
__kasan_report mm/kasan/report.c:442 [inline]
kasan_report.cold+0x83/0xdf mm/kasan/report.c:459
check_region_inline mm/kasan/generic.c:183 [inline]
kasan_check_range+0x13d/0x180 mm/kasan/generic.c:189
memcpy+0x39/0x60 mm/kasan/shadow.c:66
memcpy include/linux/fortify-string.h:225 [inline]
packet_recvmsg+0x56c/0x1150 net/packet/af_packet.c:3489
sock_recvmsg_nosec net/socket.c:948 [inline]
sock_recvmsg net/socket.c:966 [inline]
sock_recvmsg net/socket.c:962 [inline]
____sys_recvmsg+0x2c4/0x600 net/socket.c:2632
___sys_recvmsg+0x127/0x200 net/socket.c:2674
__sys_recvmsg+0xe2/0x1a0 net/socket.c:2704
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7fdfd5954c29
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 41 15 00 00 90 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 c0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffcf8e71e48 EFLAGS: 00000246 ORIG_RAX: 000000000000002f
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007fdfd5954c29
RDX: 0000000000000000 RSI: 0000000020000500 RDI: 0000000000000005
RBP: 0000000000000000 R08: 000000000000000d R09: 000000000000000d
R10: 0000000000000000 R11: 0000000000000246 R12: 00007ffcf8e71e60
R13: 00000000000f4240 R14: 000000000000c1ff R15: 00007ffcf8e71e54
</TASK>
addr ffffc9000385fb78 is located in stack of task syz-executor233/3631 at offset 32 in frame:
____sys_recvmsg+0x0/0x600 include/linux/uio.h:246
this frame has 1 object:
[32, 160) 'addr'
Memory state around the buggy address:
ffffc9000385fa80: 00 04 f3 f3 f3 f3 f3 00 00 00 00 00 00 00 00 00
ffffc9000385fb00: 00 00 00 00 00 00 00 00 00 00 00 f1 f1 f1 f1 00
>ffffc9000385fb80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 f3
^
ffffc9000385fc00: f3 f3 f3 00 00 00 00 00 00 00 00 00 00 00 00 f1
ffffc9000385fc80: f1 f1 f1 00 f2 f2 f2 00 f2 f2 f2 00 00 00 00 00
================================================================== |
| In the Linux kernel, the following vulnerability has been resolved:
can: isotp: fix potential CAN frame reception race in isotp_rcv()
When receiving a CAN frame the current code logic does not consider
concurrently receiving processes which do not show up in real world
usage.
Ziyang Xuan writes:
The following syz problem is one of the scenarios. so->rx.len is
changed by isotp_rcv_ff() during isotp_rcv_cf(), so->rx.len equals
0 before alloc_skb() and equals 4096 after alloc_skb(). That will
trigger skb_over_panic() in skb_put().
=======================================================
CPU: 1 PID: 19 Comm: ksoftirqd/1 Not tainted 5.16.0-rc8-syzkaller #0
RIP: 0010:skb_panic+0x16c/0x16e net/core/skbuff.c:113
Call Trace:
<TASK>
skb_over_panic net/core/skbuff.c:118 [inline]
skb_put.cold+0x24/0x24 net/core/skbuff.c:1990
isotp_rcv_cf net/can/isotp.c:570 [inline]
isotp_rcv+0xa38/0x1e30 net/can/isotp.c:668
deliver net/can/af_can.c:574 [inline]
can_rcv_filter+0x445/0x8d0 net/can/af_can.c:635
can_receive+0x31d/0x580 net/can/af_can.c:665
can_rcv+0x120/0x1c0 net/can/af_can.c:696
__netif_receive_skb_one_core+0x114/0x180 net/core/dev.c:5465
__netif_receive_skb+0x24/0x1b0 net/core/dev.c:5579
Therefore we make sure the state changes and data structures stay
consistent at CAN frame reception time by adding a spin_lock in
isotp_rcv(). This fixes the issue reported by syzkaller but does not
affect real world operation. |
| In the Linux kernel, the following vulnerability has been resolved:
NFSD: Fix NFSv3 SETATTR/CREATE's handling of large file sizes
iattr::ia_size is a loff_t, so these NFSv3 procedures must be
careful to deal with incoming client size values that are larger
than s64_max without corrupting the value.
Silently capping the value results in storing a different value
than the client passed in which is unexpected behavior, so remove
the min_t() check in decode_sattr3().
Note that RFC 1813 permits only the WRITE procedure to return
NFS3ERR_FBIG. We believe that NFSv3 reference implementations
also return NFS3ERR_FBIG when ia_size is too large. |
| In the Linux kernel, the following vulnerability has been resolved:
NFSD: Fix ia_size underflow
iattr::ia_size is a loff_t, which is a signed 64-bit type. NFSv3 and
NFSv4 both define file size as an unsigned 64-bit type. Thus there
is a range of valid file size values an NFS client can send that is
already larger than Linux can handle.
Currently decode_fattr4() dumps a full u64 value into ia_size. If
that value happens to be larger than S64_MAX, then ia_size
underflows. I'm about to fix up the NFSv3 behavior as well, so let's
catch the underflow in the common code path: nfsd_setattr(). |
| In the Linux kernel, the following vulnerability has been resolved:
NFSD: Fix the behavior of READ near OFFSET_MAX
Dan Aloni reports:
> Due to commit 8cfb9015280d ("NFS: Always provide aligned buffers to
> the RPC read layers") on the client, a read of 0xfff is aligned up
> to server rsize of 0x1000.
>
> As a result, in a test where the server has a file of size
> 0x7fffffffffffffff, and the client tries to read from the offset
> 0x7ffffffffffff000, the read causes loff_t overflow in the server
> and it returns an NFS code of EINVAL to the client. The client as
> a result indefinitely retries the request.
The Linux NFS client does not handle NFS?ERR_INVAL, even though all
NFS specifications permit servers to return that status code for a
READ.
Instead of NFS?ERR_INVAL, have out-of-range READ requests succeed
and return a short result. Set the EOF flag in the result to prevent
the client from retrying the READ request. This behavior appears to
be consistent with Solaris NFS servers.
Note that NFSv3 and NFSv4 use u64 offset values on the wire. These
must be converted to loff_t internally before use -- an implicit
type cast is not adequate for this purpose. Otherwise VFS checks
against sb->s_maxbytes do not work properly. |
| In the Linux kernel, the following vulnerability has been resolved:
SUNRPC: lock against ->sock changing during sysfs read
->sock can be set to NULL asynchronously unless ->recv_mutex is held.
So it is important to hold that mutex. Otherwise a sysfs read can
trigger an oops.
Commit 17f09d3f619a ("SUNRPC: Check if the xprt is connected before
handling sysfs reads") appears to attempt to fix this problem, but it
only narrows the race window. |
| In the Linux kernel, the following vulnerability has been resolved:
phy: ti: Fix missing sentinel for clk_div_table
_get_table_maxdiv() tries to access "clk_div_table" array out of bound
defined in phy-j721e-wiz.c. Add a sentinel entry to prevent
the following global-out-of-bounds error reported by enabling KASAN.
[ 9.552392] BUG: KASAN: global-out-of-bounds in _get_maxdiv+0xc0/0x148
[ 9.558948] Read of size 4 at addr ffff8000095b25a4 by task kworker/u4:1/38
[ 9.565926]
[ 9.567441] CPU: 1 PID: 38 Comm: kworker/u4:1 Not tainted 5.16.0-116492-gdaadb3bd0e8d-dirty #360
[ 9.576242] Hardware name: Texas Instruments J721e EVM (DT)
[ 9.581832] Workqueue: events_unbound deferred_probe_work_func
[ 9.587708] Call trace:
[ 9.590174] dump_backtrace+0x20c/0x218
[ 9.594038] show_stack+0x18/0x68
[ 9.597375] dump_stack_lvl+0x9c/0xd8
[ 9.601062] print_address_description.constprop.0+0x78/0x334
[ 9.606830] kasan_report+0x1f0/0x260
[ 9.610517] __asan_load4+0x9c/0xd8
[ 9.614030] _get_maxdiv+0xc0/0x148
[ 9.617540] divider_determine_rate+0x88/0x488
[ 9.622005] divider_round_rate_parent+0xc8/0x124
[ 9.626729] wiz_clk_div_round_rate+0x54/0x68
[ 9.631113] clk_core_determine_round_nolock+0x124/0x158
[ 9.636448] clk_core_round_rate_nolock+0x68/0x138
[ 9.641260] clk_core_set_rate_nolock+0x268/0x3a8
[ 9.645987] clk_set_rate+0x50/0xa8
[ 9.649499] cdns_sierra_phy_init+0x88/0x248
[ 9.653794] phy_init+0x98/0x108
[ 9.657046] cdns_pcie_enable_phy+0xa0/0x170
[ 9.661340] cdns_pcie_init_phy+0x250/0x2b0
[ 9.665546] j721e_pcie_probe+0x4b8/0x798
[ 9.669579] platform_probe+0x8c/0x108
[ 9.673350] really_probe+0x114/0x630
[ 9.677037] __driver_probe_device+0x18c/0x220
[ 9.681505] driver_probe_device+0xac/0x150
[ 9.685712] __device_attach_driver+0xec/0x170
[ 9.690178] bus_for_each_drv+0xf0/0x158
[ 9.694124] __device_attach+0x184/0x210
[ 9.698070] device_initial_probe+0x14/0x20
[ 9.702277] bus_probe_device+0xec/0x100
[ 9.706223] deferred_probe_work_func+0x124/0x180
[ 9.710951] process_one_work+0x4b0/0xbc0
[ 9.714983] worker_thread+0x74/0x5d0
[ 9.718668] kthread+0x214/0x230
[ 9.721919] ret_from_fork+0x10/0x20
[ 9.725520]
[ 9.727032] The buggy address belongs to the variable:
[ 9.732183] clk_div_table+0x24/0x440 |
| In the Linux kernel, the following vulnerability has been resolved:
mm: don't try to NUMA-migrate COW pages that have other uses
Oded Gabbay reports that enabling NUMA balancing causes corruption with
his Gaudi accelerator test load:
"All the details are in the bug, but the bottom line is that somehow,
this patch causes corruption when the numa balancing feature is
enabled AND we don't use process affinity AND we use GUP to pin pages
so our accelerator can DMA to/from system memory.
Either disabling numa balancing, using process affinity to bind to
specific numa-node or reverting this patch causes the bug to
disappear"
and Oded bisected the issue to commit 09854ba94c6a ("mm: do_wp_page()
simplification").
Now, the NUMA balancing shouldn't actually be changing the writability
of a page, and as such shouldn't matter for COW. But it appears it
does. Suspicious.
However, regardless of that, the condition for enabling NUMA faults in
change_pte_range() is nonsensical. It uses "page_mapcount(page)" to
decide if a COW page should be NUMA-protected or not, and that makes
absolutely no sense.
The number of mappings a page has is irrelevant: not only does GUP get a
reference to a page as in Oded's case, but the other mappings migth be
paged out and the only reference to them would be in the page count.
Since we should never try to NUMA-balance a page that we can't move
anyway due to other references, just fix the code to use 'page_count()'.
Oded confirms that that fixes his issue.
Now, this does imply that something in NUMA balancing ends up changing
page protections (other than the obvious one of making the page
inaccessible to get the NUMA faulting information). Otherwise the COW
simplification wouldn't matter - since doing the GUP on the page would
make sure it's writable.
The cause of that permission change would be good to figure out too,
since it clearly results in spurious COW events - but fixing the
nonsensical test that just happened to work before is obviously the
CorrectThing(tm) to do regardless. |
| In the Linux kernel, the following vulnerability has been resolved:
nvme: fix a possible use-after-free in controller reset during load
Unlike .queue_rq, in .submit_async_event drivers may not check the ctrl
readiness for AER submission. This may lead to a use-after-free
condition that was observed with nvme-tcp.
The race condition may happen in the following scenario:
1. driver executes its reset_ctrl_work
2. -> nvme_stop_ctrl - flushes ctrl async_event_work
3. ctrl sends AEN which is received by the host, which in turn
schedules AEN handling
4. teardown admin queue (which releases the queue socket)
5. AEN processed, submits another AER, calling the driver to submit
6. driver attempts to send the cmd
==> use-after-free
In order to fix that, add ctrl state check to validate the ctrl
is actually able to accept the AER submission.
This addresses the above race in controller resets because the driver
during teardown should:
1. change ctrl state to RESETTING
2. flush async_event_work (as well as other async work elements)
So after 1,2, any other AER command will find the
ctrl state to be RESETTING and bail out without submitting the AER. |
| In the Linux kernel, the following vulnerability has been resolved:
nvme-tcp: fix possible use-after-free in transport error_recovery work
While nvme_tcp_submit_async_event_work is checking the ctrl and queue
state before preparing the AER command and scheduling io_work, in order
to fully prevent a race where this check is not reliable the error
recovery work must flush async_event_work before continuing to destroy
the admin queue after setting the ctrl state to RESETTING such that
there is no race .submit_async_event and the error recovery handler
itself changing the ctrl state. |
| In the Linux kernel, the following vulnerability has been resolved:
nvme-rdma: fix possible use-after-free in transport error_recovery work
While nvme_rdma_submit_async_event_work is checking the ctrl and queue
state before preparing the AER command and scheduling io_work, in order
to fully prevent a race where this check is not reliable the error
recovery work must flush async_event_work before continuing to destroy
the admin queue after setting the ctrl state to RESETTING such that
there is no race .submit_async_event and the error recovery handler
itself changing the ctrl state. |
