| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| In the Linux kernel, the following vulnerability has been resolved:
dm thin metadata: fix metadata snapshot consistency on commit failure
__reserve_metadata_snap() and __release_metadata_snap() modify the
superblock's held_root directly in the block_manager's buffer. If the
subsequent metadata commit fails, the held_root gets flushed to disk
through the abort_transaction path, resulting in inconsistent metadata.
Reproducer 1: __reserve_metadata_snap()
1. Create a 2 MiB metadata device and make the region after the 14th
block inaccessible, to trigger metadata commit failure in the
subsequent reserve_metadata_snap operation. The 14th block will be
the shadow destination for the index block.
dmsetup create tmeta --table "0 112 linear /dev/sdc 0
112 3984 error"
2. Create a 16 MiB thin-pool
dmsetup create tdata --table "0 32768 zero"
dd if=/dev/zero of=/dev/mapper/tmeta bs=4k count=1
dmsetup create tpool --table "0 32768 thin-pool /dev/mapper/tmeta \
/dev/mapper/tdata 128 0 1 skip_block_zeroing"
3. Take a metadata snapshot to trigger metadata commit failure and
transaction abort. However, the held_root is written to disk,
breaking metadata consistency.
dmsetup message tpool 0 "reserve_metadata_snap"
thin_check v1.2.2 result:
Bad reference count for metadata block 6. Expected 2, but space map contains 1.
Bad reference count for metadata block 7. Expected 2, but space map contains 1.
Bad reference count for metadata block 13. Expected 1, but space map contains 0.
Reproducer 2: __release_metadata_snap()
1. Create a 2 MiB metadata device and make the region after the 16th
block inaccessible, to trigger metadata commit failure in the
subsequent release_metadata_snap operation. The 16th block will be
the shadow destination for the index block.
dmsetup create tmeta --table "0 128 linear /dev/sdc 0
128 3968 error"
2. Create a 16 MiB thin-pool
dmsetup create tdata --table "0 32768 zero"
dd if=/dev/zero of=/dev/mapper/tmeta bs=4k count=1
dmsetup create tpool --table "0 32768 thin-pool /dev/mapper/tmeta \
/dev/mapper/tdata 128 0 1 skip_block_zeroing"
3. Reserve then release the metadata snapshot, to trigger metadata
commit failure and transaction abort. The held_root gets removed
from the on-disk superblock, causing inconsistent metadata.
dmsetup message tpool 0 "reserve_metadata_snap"
dmsetup message tpool 0 "release_metadata_snap"
thin_check v1.2.2 result:
Bad reference count for metadata block 6. Expected 1, but space map contains 2.
Bad reference count for metadata block 7. Expected 1, but space map contains 2.
1 metadata blocks have leaked.
Fix by deferring the held_root update to commit time.
Additionally, move the existing-snapshot check in __reserve_metadata_snap
before the shadow operation to avoid unnecessary work. In
__release_metadata_snap, clear pmd->held_root before btree deletion so
partial failure leaks blocks rather than leaving a stale reference, and
unlock the snapshot block before decrementing its refcount. |
| In the Linux kernel, the following vulnerability has been resolved:
wifi: cfg80211: use wiphy work for socket owner autodisconnect
nl80211_netlink_notify() walks the cfg80211 wireless device list when a
NETLINK_GENERIC socket is released. If the socket owns a connection, the
notifier queues the embedded wdev->disconnect_wk work item.
That work is a plain work_struct today. NETDEV_GOING_DOWN cancels it, but a
NETLINK_URELEASE notifier that already observed conn_owner_nlportid can
queue it after that cancel returns. _cfg80211_unregister_wdev() then
removes the wdev from the list and waits for RCU readers, but
synchronize_net() does not drain work queued by such a reader.
Make the autodisconnect work a wiphy_work instead. The callback already
needs the wiphy mutex, and wiphy_work runs under that mutex. This lets
teardown cancel pending autodisconnect work while holding the mutex,
without a cancel_work_sync() vs. worker locking concern.
Also cancel the wiphy work after list_del_rcu() and synchronize_net(). Any
NETLINK_URELEASE notifier that had already reached the wdev list has then
either queued the work and it is removed, or can no longer find the wdev. |
| In the Linux kernel, the following vulnerability has been resolved:
drm/xe/vf: Fix VF CCS attach/detach race with in-flight BO moves
xe_bo_move() attaches VF CCS read/write batch buffers (BBs) to a BO
after it transitions NULL/SYSTEM -> TT, and detaches them after it
transitions TT -> SYSTEM. Both operations were done synchronously on
the CPU immediately after building the move's copy/clear fence,
without waiting for that fence to signal. This creates two races with
VF migration:
- Attach happens too late relative to the copy job it is meant to
protect. If the copy job is submitted before the CCS BBs are
attached, a VF migration event that pauses execution mid-copy can
observe partially copied CCS metadata without the attach state
needed to correctly save/restore it.
- Detach happens too early relative to the copy job that moves data
out of TT. The CCS BBs are torn down right after the copy fence is
obtained, while the actual blit may still be in flight. A VF
migration event that pauses execution mid-copy can then race the
save/restore path against the still-running blit, and the CCS BBs
it would need to make sense of the paused state have already been
removed.
Fix both races:
- Move the attach call to before the copy/clear job is submitted, so
the CCS BBs are already registered by the time the copy runs. On
attach failure, unwind and bail out of the move. xe_migrate_ccs_rw_copy()
now takes the destination resource explicitly, since bo->ttm.resource
is not updated to the new resource until after the move commits.
- Detach only after explicitly waiting for the copy fence to signal,
instead of tearing down the CCS BBs immediately after obtaining it.
While here, also fix xe_sriov_vf_ccs_attach_bo() to properly unwind and
propagate errors: the per-context loop previously never broke out on
error, silently discarding earlier failures. Unwind by clearing each
attached context directly via xe_migrate_ccs_rw_copy_clear() instead of
reusing xe_sriov_vf_ccs_detach_bo(), which requires both contexts to be
attached before it will clean up either one.
(cherry picked from commit d45ad0aa7a1eb5d7288b5ed948b05695611dc39e) |
| In the Linux kernel, the following vulnerability has been resolved:
iommu/amd: Wait for completion instead of returning early in iommu_completion_wait()
need_sync is a per-IOMMU flag shared by all domains and devices behind
that IOMMU. It is set whenever a command is queued with sync == true and
cleared when a completion-wait (CWAIT) command is queued. However, a
cleared need_sync only means that a covering CWAIT has been queued, not
that all previously queued commands have actually completed in hardware.
iommu_completion_wait() read need_sync locklessly and returned early
when it was false. This breaks the "block until all previously queued
commands have completed" contract in a multi-CPU scenario:
CPU2: queue inv-B => need_sync = true
CPU1: queue CWAIT(N); need_sync = false; then wait_on_sem(N)
CPU2: read need_sync == false => return 0 (no wait!)
CPU2 returns without waiting for any sequence number even though its
inv-B may not have completed yet (CWAIT(N), queued after inv-B, has not
been signaled). CPU2 then proceeds to, for example, free page-table
pages while the IOMMU can still walk stale translations, opening a
use-after-free window. This is a logical race in the meaning of the
flag, not a memory-visibility issue, so barriers alone do not help.
Fix it without losing the optimization of avoiding redundant CWAIT
commands: take iommu->lock before testing need_sync, and when it is
false do not return early but wait for the last allocated sequence
number (cmd_sem_val). Since need_sync == false implies no sync command
was queued after the last CWAIT, that CWAIT is FIFO-ordered after every
not-yet-completed command, so waiting for its sequence number guarantees
all prior commands (possibly queued by another CPU) have completed. The
common path with pending work is unchanged and no extra hardware command
is issued. |
| In the Linux kernel, the following vulnerability has been resolved:
proc: protect ptrace_may_access() with exec_update_lock (FD links)
proc_pid_get_link() and proc_pid_readlink() currently look up the task from
the pid once, then do the ptrace access check on that task, then look up
the task from the pid a second time to do the actual access.
That's racy in several ways.
To fix it, pass the task to the ->proc_get_link() handler, and instead of
proc_fd_access_allowed(), introduce a new helper call_proc_get_link() that
looks up and locks the task, does the access check, and calls
->proc_get_link(). |
| In the Linux kernel, the following vulnerability has been resolved:
KVM: x86: Ensure vendor's exit handler runs before fastpath userspace exits
Move the handling of fastpath userspace exits into vendor code to ensure
KVM runs vendor specific operations that need to run before userspace gains
control of the vCPU. E.g. for VMX (and soon to be for SVM as well), KVM
needs to flush the PML buffer prior to exiting to userspace, otherwise any
memory written by the final KVM_RUN might never be flagged as dirty.
Note, waiting to snapshot CR0 and CR3 until svm_handle_exit() is flawed in
general, as that risks consuming stale state in a fastpath handler. That
will be addressed in a future change. |
| In the Linux kernel, the following vulnerability has been resolved:
i2c: core: fix adapter deregistration race
Adapters can be looked up by their id using i2c_get_adapter() which
takes a reference to the embedded struct device.
Remove the adapter from the IDR before tearing it down during
deregistration (and on registration failure) to make sure its resources
are not accessed after having been freed (e.g. the device name). |
| In the Linux kernel, the following vulnerability has been resolved:
rbd: eliminate a race in lock_dwork draining on unmap
Given how rbd_lock_add_request() and rbd_img_exclusive_lock() are
written, lock_dwork may be (re)queued more than it's actually needed:
for example in case a new I/O request comes in while we are in the
middle of rbd_acquire_lock() on behalf of another I/O request. This is
expected and with rbd_release_lock() preemptively canceling lock_dwork
is benign under normal operation.
A more problematic example is maybe_kick_acquire():
if (have_requests || delayed_work_pending(&rbd_dev->lock_dwork)) {
dout("%s rbd_dev %p kicking lock_dwork\n", __func__, rbd_dev);
mod_delayed_work(rbd_dev->task_wq, &rbd_dev->lock_dwork, 0);
}
It's not unrealistic for lock_dwork to get canceled right after
delayed_work_pending() returns true and for mod_delayed_work() to
requeue it right there anyway. This is a classic TOCTOU race.
When it comes to unmapping the image, there is an implicit assumption
of no self-initiated exclusive lock activity past the point of return
from rbd_dev_image_unlock() which unlocks the lock if it happens to be
held. This unlock is assumed to be final and lock_dwork (as well as
all other exclusive lock tasks, really) isn't expected to get queued
again. However, lock_dwork is canceled only in cancel_tasks_sync()
(i.e. later in the unmap sequence) and on top of that the cancellation
can get in effect nullified by maybe_kick_acquire(). This may result
in rbd_acquire_lock() executing after rbd_dev_device_release() and
rbd_dev_image_release() run and free and/or reset a bunch of things.
One of the possible failure modes then is a violated
rbd_assert(rbd_image_format_valid(rbd_dev->image_format));
in rbd_dev_header_info() which is called via rbd_dev_refresh() from
rbd_post_acquire_action().
Redo exclusive lock task draining to provide saner semantics and try
to meet the assumptions around rbd_dev_image_unlock(). |
| In the Linux kernel, the following vulnerability has been resolved:
net: mana: Fix TOCTOU double-fetch of hwc_msg_id from DMA buffer
In mana_hwc_rx_event_handler(), resp->response.hwc_msg_id is read from
DMA-coherent memory and bounds-checked, then mana_hwc_handle_resp()
re-reads the same field from the same DMA buffer for test_bit() and
pointer arithmetic.
DMA-coherent memory is mapped uncacheable on x86 and is shared,
unencrypted, in Confidential VMs (SEV-SNP/TDX), so each load goes
directly to host-visible memory. A H/W can modify the value
between the check and the use, bypassing the bounds validation.
Fix this by reading hwc_msg_id exactly once using READ_ONCE() into a
stack-local variable in mana_hwc_rx_event_handler(), and passing the
validated value as a parameter to mana_hwc_handle_resp(). |
| In the Linux kernel, the following vulnerability has been resolved:
net: shaper: rework the VALID marking (again)
Recent commit changed the semantics from NOT_VALID to VALID.
I didn't realize that the flags are not stored atomically
with the entry in XArray. There's still a race of reader
observing a VALID mark for a slot, getting interrupted,
writer replacing the entry with a different one, reader
continuing, fetching the entry which is now a different
pointer than the pointer for which VALID was meant.
The biggest consequence of this is that we may see a UAF
since net_shaper_rollback() assumed that entries without
VALID can be freed without observing RCU.
Looks like the XArray marks are buying us nothing at this
point. Let's convert the code to an explicit valid field.
The smp_load_acquire() / smp_store_release() barriers are
marginally cleaner. |
| In the Linux kernel, the following vulnerability has been resolved:
sctp: fix race between sctp_wait_for_connect and peeloff
sctp_wait_for_connect() drops and re-acquires the socket lock while
waiting for the association to reach ESTABLISHED state. During this
window, another thread can peeloff the association to a new socket via
getsockopt(SCTP_SOCKOPT_PEELOFF), changing asoc->base.sk. After
re-acquiring the old socket lock, sctp_wait_for_connect() returns
success without noticing the migration — the caller then accesses
the association under the wrong lock in sctp_datamsg_from_user().
Add the same sk != asoc->base.sk check that sctp_wait_for_sndbuf()
already has, returning an error if the association was migrated while
we slept. |
| In the Linux kernel, the following vulnerability has been resolved:
KVM: SEV: Use READ_ONCE() when reading entries/indices from PSC buffer
Use READ_ONCE() when reading entries/indices from the guest-accessible
Page State Change buffer to defend against TOCTOU bugs.
Don't bother with READ_ONCE()/WRITE_ONCE() for cases where KVM is writing
(and not consuming the result!), as the guest isn't supposed to touch the
buffer while it's being processed. I.e. using READ_ONCE() is all about
protecting against misbehaving guests. |
| In the Linux kernel, the following vulnerability has been resolved:
xsk: cache csum_start/csum_offset to fix TOCTOU in xsk_skb_metadata()
The TX metadata area resides in the UMEM buffer which is memory-mapped
and concurrently writable by userspace. In xsk_skb_metadata(),
csum_start and csum_offset are read from shared memory for bounds
validation, then read again for skb assignment. A malicious userspace
application can race to overwrite these values between the two reads,
bypassing the bounds check and causing out-of-bounds memory access
during checksum computation in the transmit path.
Fix this by reading csum_start and csum_offset into local variables
once, then using the local copies for both validation and assignment.
Note that other metadata fields (flags, launch_time) and the cached
csum fields may be mutually inconsistent due to concurrent userspace
writes, but this is benign: the only security-critical invariant is
that each field's validated value is the same one used, which local
caching guarantees. |
| In the Linux kernel, the following vulnerability has been resolved:
sched/psi: fix race between file release and pressure write
A potential race condition exists between pressure write and cgroup file
release regarding the priv member of struct kernfs_open_file, which
triggers the uaf reported in [1].
Consider the following scenario involving execution on two separate CPUs:
CPU0 CPU1
==== ====
vfs_rmdir()
kernfs_iop_rmdir()
cgroup_rmdir()
cgroup_kn_lock_live()
cgroup_destroy_locked()
cgroup_addrm_files()
cgroup_rm_file()
kernfs_remove_by_name()
kernfs_remove_by_name_ns()
vfs_write() __kernfs_remove()
new_sync_write() kernfs_drain()
kernfs_fop_write_iter() kernfs_drain_open_files()
cgroup_file_write() kernfs_release_file()
pressure_write() cgroup_file_release()
ctx = of->priv;
kfree(ctx);
of->priv = NULL;
cgroup_kn_unlock()
cgroup_kn_lock_live()
cgroup_get(cgrp)
cgroup_kn_unlock()
if (ctx->psi.trigger) // here, trigger uaf for ctx, that is of->priv
The cgroup_rmdir() is protected by the cgroup_mutex, it also safeguards
the memory deallocation of of->priv performed within cgroup_file_release().
However, the operations involving of->priv executed within pressure_write()
are not entirely covered by the protection of cgroup_mutex. Consequently,
if the code in pressure_write(), specifically the section handling the
ctx variable executes after cgroup_file_release() has completed, a uaf
vulnerability involving of->priv is triggered.
Therefore, the issue can be resolved by extending the scope of the
cgroup_mutex lock within pressure_write() to encompass all code paths
involving of->priv, thereby properly synchronizing the race condition
occurring between cgroup_file_release() and pressure_write().
And, if an live kn lock can be successfully acquired while executing
the pressure write operation, it indicates that the cgroup deletion
process has not yet reached its final stage; consequently, the priv
pointer within open_file cannot be NULL. Therefore, the operation to
retrieve the ctx value must be moved to a point *after* the live kn
lock has been successfully acquired.
In another situation, specifically after entering cgroup_kn_lock_live()
but before acquiring cgroup_mutex, there exists a different class of
race condition:
CPU0: write memory.pressure CPU1: write cgroup.pressure=0
=========================== =============================
kernfs_fop_write_iter()
kernfs_get_active_of(of)
pressure_write()
cgroup_kn_lock_live(memory.pressure)
cgroup_tryget(cgrp)
kernfs_break_active_protection(kn)
... blocks on cgroup_mutex
cgroup_pressure_write()
cgroup_kn_lock_live(cgroup.pressure)
cgroup_file_show(memory.pressure, false)
kernfs_show(false)
kernfs_drain_open_files()
cgroup_file_release(of)
kfree(ctx)
of->priv = NULL
cgroup_kn_unlock()
... acquires cgroup_mutex
ctx = of->priv; // may now be NULL
if (ctx->psi.trigger) // NULL dereference
Consequently, there is a possibility that of->priv is NULL, the pressure
write needs to check for this.
Now that the scope of the cgroup_mutex has been expanded, the original
explicit cgroup_get/put operations are no longer necessary, this is
because acquiring/releasing the live kn lock inherently executes a
cgroup get/put operation.
[1]
BUG: KASAN: slab-use-after-free in pressure_write+0xa4/0x210 kernel/cgroup/cgroup.c:4011
Call Trace:
pressure_write+0xa4/0x210 kernel/cgroup/cgroup.c:4011
cgroup_file_write+0x36f/0x790 kernel/cgroup/cgroup.c:43
---truncated--- |
| In the Linux kernel, the following vulnerability has been resolved:
Revert "wireguard: device: enable threaded NAPI"
This reverts commit 933466fc50a8e4eb167acbd0d8ec96a078462e9c which is
commit db9ae3b6b43c79b1ba87eea849fd65efa05b4b2e upstream.
We have had three independent production user reports in combination
with Cilium utilizing WireGuard as encryption underneath that k8s Pod
E/W traffic to certain peer nodes fully stalled. The situation appears
as follows:
- Occurs very rarely but at random times under heavy networking load.
- Once the issue triggers the decryption side stops working completely
for that WireGuard peer, other peers keep working fine. The stall
happens also for newly initiated connections towards that particular
WireGuard peer.
- Only the decryption side is affected, never the encryption side.
- Once it triggers, it never recovers and remains in this state,
the CPU/mem on that node looks normal, no leak, busy loop or crash.
- bpftrace on the affected system shows that wg_prev_queue_enqueue
fails, thus the MAX_QUEUED_PACKETS (1024 skbs!) for the peer's
rx_queue is reached.
- Also, bpftrace shows that wg_packet_rx_poll for that peer is never
called again after reaching this state for that peer. For other
peers wg_packet_rx_poll does get called normally.
- Commit db9ae3b ("wireguard: device: enable threaded NAPI")
switched WireGuard to threaded NAPI by default. The default has
not been changed for triggering the issue, neither did CPU
hotplugging occur (i.e. 5bd8de2 ("wireguard: queueing: always
return valid online CPU in wg_cpumask_choose_online()")).
- The issue has been observed with stable kernels of v5.15 as well as
v6.1. It was reported to us that v5.10 stable is working fine, and
no report on v6.6 stable either (somewhat related discussion in [0]
though).
- In the WireGuard driver the only material difference between v5.10
stable and v5.15 stable is the switch to threaded NAPI by default.
[0] https://lore.kernel.org/netdev/CA+wXwBTT74RErDGAnj98PqS=wvdh8eM1pi4q6tTdExtjnokKqA@mail.gmail.com/
Breakdown of the problem:
1) skbs arriving for decryption are enqueued to the peer->rx_queue in
wg_packet_consume_data via wg_queue_enqueue_per_device_and_peer.
2) The latter only moves the skb into the MPSC peer queue if it does
not surpass MAX_QUEUED_PACKETS (1024) which is kept track in an
atomic counter via wg_prev_queue_enqueue.
3) In case enqueueing was successful, the skb is also queued up
in the device queue, round-robin picks a next online CPU, and
schedules the decryption worker.
4) The wg_packet_decrypt_worker, once scheduled, picks these up
from the queue, decrypts the packets and once done calls into
wg_queue_enqueue_per_peer_rx.
5) The latter updates the state to PACKET_STATE_CRYPTED on success
and calls napi_schedule on the per peer->napi instance.
6) NAPI then polls via wg_packet_rx_poll. wg_prev_queue_peek checks
on the peer->rx_queue. It will wg_prev_queue_dequeue if the
queue->peeked skb was not cached yet, or just return the latter
otherwise. (wg_prev_queue_drop_peeked later clears the cache.)
7) From an ordering perspective, the peer->rx_queue has skbs in order
while the device queue with the per-CPU worker threads from a
global ordering PoV can finish the decryption and signal the skb
PACKET_STATE_CRYPTED out of order.
8) A situation can be observed that the first packet coming in will
be stuck waiting for the decryption worker to be scheduled for
a longer time when the system is under pressure.
9) While this is the case, the other CPUs in the meantime finish
decryption and call into napi_schedule.
10) Now in wg_packet_rx_poll it picks up the first in-order skb
from the peer->rx_queue and sees that its state is still
PACKET_STATE_UNCRYPTED. The NAPI poll routine then exits e
---truncated--- |
| In the Linux kernel, the following vulnerability has been resolved:
dm: fix unlocked test for dm_suspended_md
The function dm_blk_report_zones tests if the device is suspended with
the "dm_suspended_md" call. However, this function is called without
holding any locks, so the device may be suspended just after it.
Move the call to dm_suspended_md after dm_get_live_table, so that the
device can't be suspended after the suspended state was tested. |
| In the Linux kernel, the following vulnerability has been resolved:
sctp: revalidate list cursor after sctp_sendmsg_to_asoc() in SCTP_SENDALL
The SCTP_SENDALL path in sctp_sendmsg() iterates ep->asocs with
list_for_each_entry_safe(), which caches the next entry in @tmp before
the loop body runs. The body calls sctp_sendmsg_to_asoc(), which may
drop the socket lock inside sctp_wait_for_sndbuf().
While the lock is dropped, another thread can SCTP_SOCKOPT_PEELOFF the
association cached in @tmp, migrating it to a new endpoint via
sctp_sock_migrate() (list_del_init() + list_add_tail() to
newep->asocs), and optionally close the new socket which frees the
association via kfree_rcu(). The cached @tmp can also be freed by a
network ABORT for that association, processed in softirq while the
lock is dropped.
sctp_wait_for_sndbuf() revalidates @asoc (the current entry) on re-lock
via the "sk != asoc->base.sk" and "asoc->base.dead" checks, but nothing
revalidates @tmp. After a successful return, the iterator advances to
the stale @tmp, yielding either a use-after-free (if the peeled socket
was closed) or a list-walk onto the new endpoint's list head (type
confusion of &newep->asocs as a struct sctp_association *).
Both are reachable from CapEff=0; the type-confusion path gives
controlled indirect call via the outqueue.sched->init_sid pointer.
Fix by re-deriving @tmp from @asoc after sctp_sendmsg_to_asoc()
returns. @asoc is known to still be on ep->asocs at that point: the
only callers that list_del an association from ep->asocs are
sctp_association_free() (which sets asoc->base.dead) and
sctp_assoc_migrate() (which changes asoc->base.sk), and
sctp_wait_for_sndbuf() checks both under the lock before any
successful return; a tripped check propagates as err < 0 and the loop
bails before the re-derive.
The SCTP_ABORT path in sctp_sendmsg_check_sflags() returns 0 and the
loop hits 'continue' before sctp_sendmsg_to_asoc() is ever called, so
the @tmp cached by list_for_each_entry_safe() still covers the
lock-held free that ba59fb027307 ("sctp: walk the list of asoc
safely") was added for. |
| In the Linux kernel, the following vulnerability has been resolved:
iommu/vt-d: Clear Present bit before tearing down context entry
When tearing down a context entry, the current implementation zeros the
entire 128-bit entry using multiple 64-bit writes. This creates a window
where the hardware can fetch a "torn" entry — where some fields are
already zeroed while the 'Present' bit is still set — leading to
unpredictable behavior or spurious faults.
While x86 provides strong write ordering, the compiler may reorder writes
to the two 64-bit halves of the context entry. Even without compiler
reordering, the hardware fetch is not guaranteed to be atomic with
respect to multiple CPU writes.
Align with the "Guidance to Software for Invalidations" in the VT-d spec
(Section 6.5.3.3) by implementing the recommended ownership handshake:
1. Clear only the 'Present' (P) bit of the context entry first to
signal the transition of ownership from hardware to software.
2. Use dma_wmb() to ensure the cleared bit is visible to the IOMMU.
3. Perform the required cache and context-cache invalidation to ensure
hardware no longer has cached references to the entry.
4. Fully zero out the entry only after the invalidation is complete.
Also, add a dma_wmb() to context_set_present() to ensure the entry
is fully initialized before the 'Present' bit becomes visible. |
| In the Linux kernel, the following vulnerability has been resolved:
ext4: fix e4b bitmap inconsistency reports
A bitmap inconsistency issue was observed during stress tests under
mixed huge-page workloads. Ext4 reported multiple e4b bitmap check
failures like:
ext4_mb_complex_scan_group:2508: group 350, 8179 free clusters as
per group info. But got 8192 blocks
Analysis and experimentation confirmed that the issue is caused by a
race condition between page migration and bitmap modification. Although
this timing window is extremely narrow, it is still hit in practice:
folio_lock ext4_mb_load_buddy
__migrate_folio
check ref count
folio_mc_copy __filemap_get_folio
folio_try_get(folio)
......
mb_mark_used
ext4_mb_unload_buddy
__folio_migrate_mapping
folio_ref_freeze
folio_unlock
The root cause of this issue is that the fast path of load_buddy only
increments the folio's reference count, which is insufficient to prevent
concurrent folio migration. We observed that the folio migration process
acquires the folio lock. Therefore, we can determine whether to take the
fast path in load_buddy by checking the lock status. If the folio is
locked, we opt for the slow path (which acquires the lock) to close this
concurrency window.
Additionally, this change addresses the following issues:
When the DOUBLE_CHECK macro is enabled to inspect bitmap-related
issues, the following error may be triggered:
corruption in group 324 at byte 784(6272): f in copy != ff on
disk/prealloc
Analysis reveals that this is a false positive. There is a specific race
window where the bitmap and the group descriptor become momentarily
inconsistent, leading to this error report:
ext4_mb_load_buddy ext4_mb_load_buddy
__filemap_get_folio(create|lock)
folio_lock
ext4_mb_init_cache
folio_mark_uptodate
__filemap_get_folio(no lock)
......
mb_mark_used
mb_mark_used_double
mb_cmp_bitmaps
mb_set_bits(e4b->bd_bitmap)
folio_unlock
The original logic assumed that since mb_cmp_bitmaps is called when the
bitmap is newly loaded from disk, the folio lock would be sufficient to
prevent concurrent access. However, this overlooks a specific race
condition: if another process attempts to load buddy and finds the folio
is already in an uptodate state, it will immediately begin using it without
holding folio lock. |
| In the Linux kernel, the following vulnerability has been resolved:
iommu/vt-d: Clear Present bit before tearing down PASID entry
The Intel VT-d Scalable Mode PASID table entry consists of 512 bits (64
bytes). When tearing down an entry, the current implementation zeros the
entire 64-byte structure immediately using multiple 64-bit writes.
Since the IOMMU hardware may fetch these 64 bytes using multiple
internal transactions (e.g., four 128-bit bursts), updating or zeroing
the entire entry while it is active (P=1) risks a "torn" read. If a
hardware fetch occurs simultaneously with the CPU zeroing the entry, the
hardware could observe an inconsistent state, leading to unpredictable
behavior or spurious faults.
Follow the "Guidance to Software for Invalidations" in the VT-d spec
(Section 6.5.3.3) by implementing the recommended ownership handshake:
1. Clear only the 'Present' (P) bit of the PASID entry.
2. Use a dma_wmb() to ensure the cleared bit is visible to hardware
before proceeding.
3. Execute the required invalidation sequence (PASID cache, IOTLB, and
Device-TLB flush) to ensure the hardware has released all cached
references.
4. Only after the flushes are complete, zero out the remaining fields
of the PASID entry.
Also, add a dma_wmb() in pasid_set_present() to ensure that all other
fields of the PASID entry are visible to the hardware before the Present
bit is set. |
