| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| In the Linux kernel, the following vulnerability has been resolved:
drm/xe: fix NPD in bo_meminfo()
When a buffer object is purged, its ttm.resource is set to NULL via the
TTM pipeline gutting flow. However, the BO remains in the client's
object list until userspace explicitly closes the GEM handle. If memory
stats are queried during this time, accessing bo->ttm.resource->mem_type
will result in a NULL pointer dereference.
Fix this by safely skipping purged BOs in bo_meminfo, as they no longer
consume any memory.
User is getting NPD on device resume, and possible theory is that in
bo_move(), if we need to evict something to SYSTEM to save the CCS state,
but the BO is marked as dontneed, this won't trigger a move but will
nuke the pages, leaving us with a NULL bo resource. And the meminfo()
doesn't look ready to handle a NULL resource.
v2 (Sashiko):
- There could potentially be other cases where we might end up with a
NULL resource, so make this a general NULL check for now.
(cherry picked from commit c9a8e7daa0afe3161111e27fd92176e608c7f186) |
| In the Linux kernel, the following vulnerability has been resolved:
drm/xe/pt: prevent invalid cursor access for purged BOs
During a page table walk for binding, xe_pt_stage_bind() explicitly
skips initializing the xe_res_cursor for purged BOs, treating them
similarly to NULL VMAs by only setting the cursor size.
However, xe_pt_hugepte_possible() and xe_pt_scan_64K() did not check
if the BO was purged before attempting to walk the cursor using
xe_res_dma() and xe_res_next(). Because the cursor was left
uninitialized for purged BOs, this falls through and triggers
warnings like:
WARNING: drivers/gpu/drm/xe/xe_res_cursor.h:274 at xe_res_next
Fix this by explicitly checking if the BO is purged in both
xe_pt_hugepte_possible() and xe_pt_scan_64K(), returning early just
as we do for NULL VMAs, avoiding the invalid cursor accesses entirely.
As a precaution, also zero-initialize the cursor in xe_pt_stage_bind()
to ensure we don't pass garbage data into the page table walkers
if we ever hit a similar edge case in the future.
(cherry picked from commit 4c7b9c6ece32440e5a435a92076d049450cd2d2e) |
| In the Linux kernel, the following vulnerability has been resolved:
uprobes/x86: Use proper mm_struct in __in_uprobe_trampoline
In the unregister path we use __in_uprobe_trampoline check with
current->mm for the VMA lookup, which is wrong, because we are
in the tracer context, not the traced process.
Add mm_struct pointer argument to __in_uprobe_trampoline and
changing related callers to pass proper mm_struct pointer. |
| In the Linux kernel, the following vulnerability has been resolved:
cifs: Fix missing credit release on failure in cifs_issue_read()
Fix missing release of credits in the failure path in cifs_issue_read()
lest retrying the subreq just overwrites the credits value. |
| In the Linux kernel, the following vulnerability has been resolved:
netfs: Fix barriering when walking subrequest list
Fix the barriering used when walking the subrequest list in retry as
there's a possibility of seeing a subreq that's just been added by the
application thread. |
| In the Linux kernel, the following vulnerability has been resolved:
ntfs: avoid stale runlist element dereference in MFT writeback
ntfs_write_mft_block() maps each $MFT record through the $MFT data
runlist. For sub-folio clusters it looks up a struct runlist_element under
ni->runlist.lock, drops the lock, and later uses rl->length and rl->vcn
when choosing folio_sz.
That pointer is only borrowed from ni->runlist.rl. Concurrent $MFT
allocation extension can merge a replacement runlist under the same lock,
and ntfs_rl_realloc() can free the old backing array. If that happens
between the lookup and the later folio_sz decision, writeback can
dereference freed runlist storage.
The buggy scenario involves two paths, with each column showing the order
within that path:
MFT writeback path: $MFT allocation extension:
1. Look up rl under 1. Extend the $MFT data allocation.
ni->runlist.lock. 2. Publish a replacement runlist.
2. Drop ni->runlist.lock. 3. Free the old runlist array.
3. Read rl->length and rl->vcn
to choose folio_sz.
Compute the remaining run length while ni->runlist.lock is still held, and
use that scalar after unlock. This preserves the existing folio sizing
decision without carrying a borrowed runlist_element across the lock
boundary.
Validation reproduced this kernel report:
BUG: KASAN: slab-use-after-free in ntfs_mft_writepages+0x1c8d/0x1fb0
Call Trace:
<TASK>
dump_stack_lvl+0x66/0xa0
print_report+0xce/0x630
? ntfs_mft_writepages+0x1c8d/0x1fb0
? srso_alias_return_thunk+0x5/0xfbef5
? __virt_addr_valid+0x20d/0x410
? ntfs_mft_writepages+0x1c8d/0x1fb0
kasan_report+0xe0/0x110
? ntfs_mft_writepages+0x1c8d/0x1fb0
ntfs_mft_writepages+0x1c8d/0x1fb0
? __pfx_ntfs_mft_writepages+0x10/0x10
? __pfx___mutex_unlock_slowpath+0x10/0x10
? srso_alias_return_thunk+0x5/0xfbef5
? iput+0x92/0xa80
do_writepages+0x219/0x530
? __pfx_do_writepages+0x10/0x10
__writeback_single_inode+0x117/0xf50
? do_raw_spin_lock+0x130/0x270
? __pfx_do_raw_spin_lock+0x10/0x10
? __pfx___writeback_single_inode+0x10/0x10
? srso_alias_return_thunk+0x5/0xfbef5
writeback_sb_inodes+0x65b/0x1810
? srso_alias_return_thunk+0x5/0xfbef5
? lock_acquire+0x2b8/0x2f0
? __pfx_writeback_sb_inodes+0x10/0x10
? lock_release+0x1e0/0x280
? _raw_spin_unlock+0x23/0x40
? move_expired_inodes+0x2b8/0x850
__writeback_inodes_wb+0xf4/0x270
? __pfx___writeback_inodes_wb+0x10/0x10
? srso_alias_return_thunk+0x5/0xfbef5
? queue_io+0x2e4/0x410
wb_writeback+0x666/0x880
? srso_alias_return_thunk+0x5/0xfbef5
? __pfx_wb_writeback+0x10/0x10
? srso_alias_return_thunk+0x5/0xfbef5
? srso_alias_return_thunk+0x5/0xfbef5
? get_nr_dirty_inodes+0x1c/0x170
wb_workfn+0x75e/0xbb0
? srso_alias_return_thunk+0x5/0xfbef5
? _raw_spin_unlock_irqrestore+0x27/0x60
? __pfx_wb_workfn+0x10/0x10
? __pfx_debug_object_deactivate+0x10/0x10
? lock_acquire+0x2b8/0x2f0
? srso_alias_return_thunk+0x5/0xfbef5
? lock_release+0x1e0/0x280
process_one_work+0x8d0/0x1870
? __pfx_process_one_work+0x10/0x10
? srso_alias_return_thunk+0x5/0xfbef5
worker_thread+0x575/0xf80
? __pfx_worker_thread+0x10/0x10
kthread+0x2e7/0x3c0
? __pfx_kthread+0x10/0x10
ret_from_fork+0x576/0x810
? __pfx_ret_from_fork+0x10/0x10
? srso_alias_return_thunk+0x5/0xfbef5
? __switch_to+0x57e/0xe10
? __switch_to_asm+0x33/0x70
? __pfx_kthread+0x10/0x10
ret_from_fork_asm+0x1a/0x30
</TASK>
Allocated by task 970:
kasan_save_stack+0x33/0x60
kasan_save_track+0x14/0x30
__kasan_kmalloc+0xaa/0xb0
__kvmalloc_node_noprof+0x353/0x920
ntfs_rl_realloc+0x3c/0x80
ntfs_runlists_merge+0x1212/0x3010
ntfs_mft_data_extend_allocation_nolock+0x3e0/0x1f40
ntfs_mft_record_alloc+0x1ab4/0x4f10
__ntfs_create+0x680/0x2e50
ntfs_create+0x1e6/0x3a0
path_openat+0x2b55/0x3c10
do_file_open+0x1f4/0x460
do_sys_openat2+0xde/0x170
__x64_sys_openat+0x122/0x1e0
do_syscall_64+0x115/0x6a0
entry_SYSCALL_64_after_hwframe+0x77/0x7f
Freed by task 1294:
kasan_save_
---truncated--- |
| In the Linux kernel, the following vulnerability has been resolved:
ntfs: avoid stale runlist element dereference in fallocate
ntfs_attr_fallocate() allocates holes and delayed allocations inside
initialized size by looking up the current runlist element under
ni->runlist.lock. The returned struct runlist_element is only a borrowed
pointer into ni->runlist.rl. A writer can replace and free that array
after the read lock is dropped, so later reads of rl->lcn, rl->length and
rl->vcn can touch freed memory.
The buggy scenario involves two paths, with each column showing the order
within that path:
ntfs_attr_fallocate():
1. Take ni->runlist.lock for read.
2. Get rl from ntfs_attr_find_vcn_nolock().
3. Drop ni->runlist.lock.
4. Read rl->lcn, rl->length and rl->vcn.
mmap page_mkwrite:
1. Enter ntfs_filemap_page_mkwrite().
2. Reach __ntfs_write_iomap_begin() and ntfs_attr_map_cluster().
3. Merge allocation state with ntfs_runlists_merge().
4. Reallocate ni->runlist.rl in ntfs_rl_realloc(), freeing the old array.
Validation reproduced this kernel report:
BUG: KASAN: slab-use-after-free in ntfs_attr_fallocate+0xbb8/0xd00
Call Trace:
<TASK>
dump_stack_lvl+0x66/0xa0
print_report+0xce/0x630
? ntfs_attr_fallocate+0xbb8/0xd00
? srso_alias_return_thunk+0x5/0xfbef5
? __virt_addr_valid+0x20d/0x410
? ntfs_attr_fallocate+0xbb8/0xd00
kasan_report+0xe0/0x110
? ntfs_attr_fallocate+0xbb8/0xd00
ntfs_attr_fallocate+0xbb8/0xd00
? lock_acquire+0x2b8/0x2f0
? __pfx_ntfs_attr_fallocate+0x10/0x10
? 0xffffffffc0000095
? down_write+0x10d/0x1e0
ntfs_fallocate+0x5c9/0x1d00
? __pfx_ntfs_fallocate+0x10/0x10
? srso_alias_return_thunk+0x5/0xfbef5
? lock_acquire+0x2b8/0x2f0
? srso_alias_return_thunk+0x5/0xfbef5
? selinux_file_permission+0x3a7/0x510
vfs_fallocate+0x29d/0xd30
__x64_sys_fallocate+0xc7/0x150
? do_syscall_64+0x81/0x6a0
do_syscall_64+0x115/0x6a0
entry_SYSCALL_64_after_hwframe+0x77/0x7f
Allocated by task 410:
kasan_save_stack+0x33/0x60
kasan_save_track+0x14/0x30
__kasan_kmalloc+0xaa/0xb0
__kvmalloc_node_noprof+0x353/0x920
ntfs_rl_realloc+0x3f/0x110
ntfs_runlists_merge+0xaa3/0x3010
ntfs_attr_map_cluster+0x4e5/0xf80
ntfs_attr_fallocate+0x53f/0xd00
ntfs_fallocate+0x5c9/0x1d00
vfs_fallocate+0x29d/0xd30
__x64_sys_fallocate+0xc7/0x150
do_syscall_64+0x115/0x6a0
entry_SYSCALL_64_after_hwframe+0x77/0x7f
Freed by task 424:
kasan_save_stack+0x33/0x60
kasan_save_track+0x14/0x30
kasan_save_free_info+0x3b/0x60
__kasan_slab_free+0x5f/0x80
kfree+0x307/0x580
ntfs_rl_realloc+0x6f/0x110
ntfs_runlists_merge+0x7b1/0x3010
ntfs_attr_map_cluster+0x4e5/0xf80
__ntfs_write_iomap_begin+0x8cd/0x2280
iomap_iter+0x6de/0x11e0
iomap_page_mkwrite+0x391/0x650
ntfs_filemap_page_mkwrite+0x1ac/0x400
do_page_mkwrite+0x15c/0x280
__handle_mm_fault+0xd6d/0x1ca0
handle_mm_fault+0x19c/0x470
do_user_addr_fault+0x23b/0x9c0
exc_page_fault+0x5c/0xc0
asm_exc_page_fault+0x26/0x30
Fix this by copying the needed runlist fields while the read lock is still
held and using only those scalar snapshots after unlocking.
After the snapshot, ntfs_attr_map_cluster() can also find that the range
is already mapped and return balloc=false. Only call ntfs_dio_zero_range()
when new clusters were allocated, matching the write iomap path and
preserving the zero-newly-allocated-holes behavior. |
| In the Linux kernel, the following vulnerability has been resolved:
HID: bpf: Fix hid_bpf_get_data() range check
hid_bpf_get_data() returns a pointer into the HID-BPF context data when
the caller-provided offset and size fit inside ctx->allocated_size.
The current check adds rdwr_buf_size and offset before comparing the
result against ctx->allocated_size. Since both values are unsigned, a
very large size can wrap the sum below ctx->allocated_size and make the
helper return a pointer even though the requested range is not contained
in the backing buffer.
Use check_add_overflow() to reject wrapped range ends before comparing
the requested range end against ctx->allocated_size. |
| In the Linux kernel, the following vulnerability has been resolved:
gue: validate REMCSUM private option length
GUE private flags can indicate that remote checksum offload metadata is
present. The private flags field itself is accounted for by
guehdr_flags_len(), but guehdr_priv_flags_len() currently returns 0 even
when GUE_PFLAG_REMCSUM is set.
This lets a packet with only the private flags field pass
validate_gue_flags(), after which gue_remcsum() and gue_gro_remcsum()
read the missing REMCSUM start/offset fields from the following bytes.
Account for GUE_PLEN_REMCSUM when GUE_PFLAG_REMCSUM is present so that
malformed packets are rejected during option validation. |
| In the Linux kernel, the following vulnerability has been resolved:
netfilter: xt_u32: reject invalid shift counts
u32_match_it() executes rule-supplied shift operands on a 32-bit
value. A malformed u32 rule can provide a shift count of 32 or more,
triggering an undefined shift out-of-bounds during packet evaluation.
Validate XT_U32_LEFTSH and XT_U32_RIGHTSH operands in
u32_mt_checkentry() and reject malformed rules before they reach the
packet path. |
| In the Linux kernel, the following vulnerability has been resolved:
netfilter: xt_rateest: fix u64 truncation in xt_rateest_mt()
On links faster than ~34 Gbps, where byte rate may exceed 2^32-1
(~ 4.3 GBps), the comparison result becomes incorrect because the
truncated value no longer reflects the actual estimator rate.
Fix by changing the local variables to u64. |
| In the Linux kernel, the following vulnerability has been resolved:
netfilter: ip6tables: mark malformed IPv6 extension headers for hotdrop
The ah, hbh and rt matches check that the fixed extension header is
present, then use the header length field to derive the advertised
extension header length for matching.
For the ah match, add the missing advertised-length check. For hbh
and rt, update the existing advertised-length checks. In all three
cases, set hotdrop to true before returning false when the advertised
extension header length exceeds the available skb data.
Returning false treats the packet as a rule mismatch. Set hotdrop to
true and drop malformed packets so they cannot bypass rules intended
to drop packets with these IPv6 extension headers. |
| In the Linux kernel, the following vulnerability has been resolved:
netfilter: xt_connmark: reject invalid shift parameters
Revision 2 of the CONNMARK target accepts user-controlled shift
parameters and applies them to 32-bit mark values in
connmark_tg_shift().
A shift_bits value of 32 or more triggers an undefined-shift bug when
the rule is evaluated. Invalid shift_dir values are also accepted and
silently fall back to the left-shift path.
Reject invalid revision-2 shift parameters in connmark_tg_check() so
malformed rules fail at installation time, before they can reach the
packet path. |
| In the Linux kernel, the following vulnerability has been resolved:
platform/x86: bitland-mifs-wmi: Fix NULL pointer dereference during suspend/resume
The driver registers two distinct WMI devices: a control device
(BITLAND_WMI_CONTROL) and an event device (BITLAND_WMI_EVENT). During
the probe phase, the event device handling path returns early before
initializing the platform profile device (data->pp_dev), leaving it
NULL.
However, the PM sleep operations are registered globally for the WMI
driver and are triggered for both devices. When entering suspend, the
event device invokes bitland_mifs_wmi_suspend(), which passes the
uninitialized data->pp_dev (NULL) into laptop_profile_get(). This leads
to a NULL pointer dereference inside dev_get_drvdata(), causing a
kernel Oops and halting the suspend sequence.
Fix this by adding a validity check for data->pp_dev in both the suspend
and resume callbacks, safely skipping profile operations for the event
device. |
| In the Linux kernel, the following vulnerability has been resolved:
net/mlx5: LAG, Fix off-by-one in single-FDB error rollback
On failure at index i, the reverse cleanup loop in
mlx5_lag_create_single_fdb() starts from i, so the failed index
itself is rolled back. That can operate on uninitialized state or
double-tear-down a rule the add_one path already self-rolled-back.
Start the rollback from i - 1 so only successfully-installed entries
are undone. |
| In the Linux kernel, the following vulnerability has been resolved:
net/mlx5e: TC, skip peer flow cleanup when LAG seq is unavailable
mlx5_lag_get_dev_seq() will return error when the peer isn't in the LAG
or when no device is marked as master. Result bad memory access and kernel
crash[1].
Hence, skip the peer when lookup fails.
Note: In case there are peer flows, they are cleaned before LAG cleared
the master mark.
[1]
RIP: 0010:mlx5e_tc_del_fdb_peers_flow+0x3d/0x350 [mlx5_core]
Call Trace:
<TASK>
mlx5e_tc_clean_fdb_peer_flows+0xc1/0x130 [mlx5_core]
mlx5_esw_offloads_unpair+0x3a/0x400 [mlx5_core]
mlx5_esw_offloads_devcom_event+0xee/0x360 [mlx5_core]
mlx5_devcom_send_event+0x7a/0x140 [mlx5_core]
mlx5_esw_offloads_devcom_cleanup+0x2f/0x90 [mlx5_core]
mlx5e_tc_esw_cleanup+0x28/0xf0 [mlx5_core]
mlx5e_rep_tc_cleanup+0x19/0x30 [mlx5_core]
mlx5e_cleanup_uplink_rep_tx+0x36/0x40 [mlx5_core]
mlx5e_cleanup_rep_tx+0x55/0x60 [mlx5_core]
mlx5e_detach_netdev+0x96/0xf0 [mlx5_core]
mlx5e_netdev_change_profile+0x5b/0x120 [mlx5_core]
mlx5e_netdev_attach_nic_profile+0x1b/0x30 [mlx5_core]
mlx5e_vport_rep_unload+0xdd/0x110 [mlx5_core]
__esw_offloads_unload_rep+0x81/0xb0 [mlx5_core]
mlx5_eswitch_unregister_vport_reps+0x1d7/0x220 [mlx5_core]
mlx5e_rep_remove+0x22/0x30 [mlx5_core]
device_release_driver_internal+0x194/0x1f0
bus_remove_device+0xe8/0x1b0
device_del+0x159/0x3c0
mlx5_rescan_drivers_locked+0xbc/0x2d0 [mlx5_core]
mlx5_unregister_device+0x54/0x80 [mlx5_core]
mlx5_uninit_one+0x73/0x130 [mlx5_core]
remove_one+0x78/0xe0 [mlx5_core]
pci_device_remove+0x39/0xa0 |
| In the Linux kernel, the following vulnerability has been resolved:
net/mlx5e: Fix HV VHCA stats zero-sized buffer allocation
mlx5e_hv_vhca_stats_create() is called from mlx5e_nic_enable(),
before mlx5e_open(). At that point priv->stats_nch is still zero,
because it is only ever incremented in mlx5e_channel_stats_alloc(),
which is reached only from mlx5e_open_channel().
mlx5e_hv_vhca_stats_buf_size() therefore returns 0, and
kvzalloc(0, GFP_KERNEL) returns ZERO_SIZE_PTR ((void *)16) rather
than NULL. The "if (!buf)" guard does not catch this, and
mlx5e_hv_vhca_stats_create() completes "successfully" with
priv->stats_agent.buf set to ZERO_SIZE_PTR.
Once channels are opened (priv->stats_nch > 0) and the hypervisor
enables stats reporting, mlx5e_hv_vhca_stats_work() recomputes
buf_len using the new non-zero stats_nch and calls
memset(buf, 0, buf_len) on ZERO_SIZE_PTR, faulting at address 0x10.
Allocate the buffer based on priv->max_nch, which is set in
mlx5e_priv_init() and is the upper bound on stats_nch:
- Add a separate helper mlx5e_hv_vhca_stats_buf_max_size() that
returns sizeof(per_ring_stats) * max(max_nch, stats_nch), and
use it for the kvzalloc() in mlx5e_hv_vhca_stats_create().
- Keep mlx5e_hv_vhca_stats_buf_size() (which returns based on
stats_nch) for the worker's active payload size, so the wire
format (block->rings = stats_nch) and the amount of data filled
by mlx5e_hv_vhca_fill_stats() are unchanged.
The max(max_nch, stats_nch) guard handles the rare case where
mlx5e_attach_netdev() recomputes max_nch downward across a
detach/resume cycle while priv->stats_nch persists (mlx5e_detach_netdev
does not call mlx5e_priv_cleanup, so stats_nch is only reset when
the netdev is destroyed). Without the guard, the worker could compute
buf_len from stats_nch and overrun the smaller buffer allocated based
on the reduced max_nch.
Allocating a non-zero buffer also makes the kvzalloc() failure path in
mlx5e_hv_vhca_stats_create() reachable for the first time: it returns
early without (re)creating the agent. Clear
priv->stats_agent.{agent,buf} in mlx5e_hv_vhca_stats_destroy() after
freeing them, so that if a later create() bails out on this path, a
subsequent teardown does not double-free the stale agent/buffer left
from a previous enable/disable cycle.
This mirrors the existing mlx5e pattern of preallocating arrays of
size max_nch (e.g. priv->channel_stats) and lazily populating
entries up to stats_nch on demand. |
| In the Linux kernel, the following vulnerability has been resolved:
net/mlx5e: Fix HV VHCA stats agent registration race
mlx5e_hv_vhca_stats_create() registers the stats agent through
mlx5_hv_vhca_agent_create(). The helper publishes the agent in
hv_vhca->agents[type] under agents_lock and immediately schedules an
asynchronous control invalidation on the HV VHCA workqueue before
returning to mlx5e.
The asynchronous invalidation invokes the control agent's invalidate
callback, which reads the hypervisor control block and forwards the
command to mlx5e_hv_vhca_stats_control(). That callback may either:
- call cancel_delayed_work_sync(&priv->stats_agent.work), or
- call queue_delayed_work(priv->wq, &sagent->work, sagent->delay).
However, the delayed_work and priv->stats_agent.agent are only
initialized after mlx5_hv_vhca_agent_create() returns to mlx5e:
agent = mlx5_hv_vhca_agent_create(...); /* publish + invalidate */
...
priv->stats_agent.agent = agent; /* too late */
INIT_DELAYED_WORK(&priv->stats_agent.work, ...); /* too late */
If the asynchronous control path runs before the two assignments
above, it can:
- Operate on an uninitialized delayed_work whose timer.function is
NULL. queue_delayed_work() calls add_timer() unconditionally, so
when the timer expires the timer softirq invokes a NULL function
pointer.
- Re-initialize the timer later through INIT_DELAYED_WORK() while
the timer is already enqueued in the timer wheel, corrupting the
hlist (entry.pprev cleared while the previous bucket node still
points at this entry).
- When the worker eventually runs, mlx5e_hv_vhca_stats_work() reads
sagent->agent (NULL) and dereferences it inside
mlx5_hv_vhca_agent_write().
Fix this by:
- Initializing priv->stats_agent.work before invoking
mlx5_hv_vhca_agent_create(), so the work is always in a valid
state when the control callback observes it.
- Adding a struct mlx5_hv_vhca_agent **ctx_update out-parameter
to mlx5_hv_vhca_agent_create(). The helper writes the agent
pointer to *ctx_update before publishing into hv_vhca->agents[]
and triggering the agents_update flow, so any callback
subsequently invoked from that flow already sees a valid
priv->stats_agent.agent. This avoids having the control
callback participate in agent initialization.
While at it, access priv->stats_agent.agent with
READ_ONCE()/WRITE_ONCE() for the cross-CPU access with the worker, and
clear priv->stats_agent.buf on the agent_create() failure path. |
| In the Linux kernel, the following vulnerability has been resolved:
net/mlx5e: Fix publication race for priv->channel_stats[]
mlx5e_channel_stats_alloc() publishes a new entry to
priv->channel_stats[] and then increments priv->stats_nch as a
publication token, but neither store carries any memory barrier:
priv->channel_stats[ix] = kvzalloc_node(...);
if (!priv->channel_stats[ix])
return -ENOMEM;
priv->stats_nch++;
Concurrent readers compute the loop bound from priv->stats_nch and
then dereference priv->channel_stats[i] using plain accesses, e.g.
for (i = 0; i < priv->stats_nch; i++) {
struct mlx5e_channel_stats *cs = priv->channel_stats[i];
... cs->rq.packets ...
}
On weakly-ordered architectures (ARM, PowerPC, RISC-V) the writes to
channel_stats[ix] and stats_nch may become visible to other CPUs out
of program order. A reader can observe stats_nch == N while still
seeing channel_stats[N-1] == NULL, leading to a NULL pointer
dereference in the channel_stats loop.
This has been observed in production on BlueField-3 DPUs (arm64),
where ovs-vswitchd queries netdev statistics over netlink during NIC
bringup, racing mlx5e_open_channel() -> mlx5e_channel_stats_alloc()
on another CPU:
Unable to handle kernel NULL pointer dereference at virtual address 0x840
Hardware name: BlueField-3 DPU
pc : mlx5e_fold_sw_stats64+0x30/0x180 [mlx5_core]
Call trace:
mlx5e_fold_sw_stats64+0x30/0x180 [mlx5_core]
dev_get_stats+0x50/0xc0
ovs_vport_get_stats+0x38/0xac [openvswitch]
ovs_vport_cmd_fill_info+0x194/0x290 [openvswitch]
ovs_vport_cmd_get+0xbc/0x10c [openvswitch]
genl_family_rcv_msg_doit+0xd0/0x160
genl_rcv_msg+0xec/0x1f0
netlink_rcv_skb+0x64/0x130
genl_rcv+0x40/0x60
netlink_unicast+0x2fc/0x370
netlink_sendmsg+0x1dc/0x454
...
__arm64_sys_sendmsg+0x2c/0x40
Add mlx5e_stats_nch_write() and mlx5e_stats_nch_read() helpers in en.h
that wrap the smp_store_release()/smp_load_acquire() pair on stats_nch.
The release/acquire pair establishes the contract:
stats_nch == N => channel_stats[0..N-1] are visible and non-NULL.
Publish the stats_nch increment via mlx5e_stats_nch_write() in the
writer (mlx5e_channel_stats_alloc()), and read stats_nch via
mlx5e_stats_nch_read() in all readers: mlx5e RX/TX queue stats,
mlx5e_get_base_stats(), ethtool channels stats, IPoIB stats, the
sw_stats fold and the HV VHCA stats agent. |
| In the Linux kernel, the following vulnerability has been resolved:
net: microchip: vcap: fix races on the shared Super VCAP block
The VCAP instances on a chip are not independent, yet they are locked
independently. On sparx5 and lan969x the IS0 and IS2 instances are
backed by the same Super VCAP hardware block and share its cache and
command registers: every access drives the shared VCAP_SUPER_CTRL
register and moves data through the shared cache registers.
Accessing one instance therefore races with accessing another. The
per-instance admin->lock cannot prevent this, as each instance takes a
different lock.
The locking issue is mostly disguised by the fact that the core usage of
the vcap api runs under rtnl. However, the full rule dump in debugfs
decodes rules straight from hardware (a READ command followed by a cache
read) and runs outside rtnl, so it races a concurrent tc-flower rule
write to another Super VCAP instance.
Besides corrupting the dump, the read repopulates the shared cache
between the writers cache fill and its write command, so the writer
commits the wrong data and corrupts the hardware entry.
Introduce vcap_lock() and vcap_unlock() helpers and route every rule
lock site in the VCAP API and its debugfs code through them. Replace the
per-instance admin->lock with a single mutex in struct vcap_control that
serializes access to all instances. The helpers reach it through a new
admin->vctrl back-pointer, and the clients initialise and destroy the
control lock instead of a per-instance one.
No path holds more than one instance lock, so collapsing them onto a
single mutex cannot self-deadlock. |
