| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Up to now, rate groups could only contain vports from the same E-Switch.
This patch relaxes that restriction if the device supports it
(HCA_CAP.esw_cross_esw_sched == true) and the right conditions are met:
- Link Aggregation (LAG) is enabled.
- The E-Switches are from the same shared devlink device.
Signed-off-by: Cosmin Ratiu <cratiu@nvidia.com>
Reviewed-by: Carolina Jubran <cjubran@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260701073254.754518-13-tariqt@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
E-Switch QoS domains were added with the intention of eventually
implementing shared qos domains to support cross-esw scheduling in the
previous approach ([1]), but they are no longer necessary in the new
approach.
Remove QoS domains and switch to using the shd lock for protecting
against concurrent QoS modifications.
Enable the supported_cross_device_rate_nodes devlink ops attribute so
that all calls originating from devlink rate acquire the shd lock. Only
the additional entry points into QoS need to acquire the shd lock.
The wrinkle is that since shd can be NULL (e.g. on older HW without
serial number available), there needs to be a fallback locking
mechanism. The devlink instance lock cannot be used, as some code paths
into QoS (get, set & modify vport rate) happen with RTNL held, and the
existing devlink -> RTNL order prevents devlink lock usage there.
The other two options are either esw->state_lock or a new lock as
fallback when shd is NULL. This patch adds esw->state_lock, which
implies:
- 3 new lock/unlock helper pairs to acquire/release the missing lock:
- esw_qos_{,un}lock: acquire/release esw->state_lock when shd is NULL.
- esw_qos_shd_{,un}lock: when esw->state_lock is already held.
- esw_qos_devlink_{,un}lock: when shd is already held.
- esw_assert_qos_lock_held now asserts esw->state_lock is held when shd
is NULL.
Use the corresponding lock/unlock function in all places where either
shd or state_lock would need to be acquired.
Document all of this trickery next to esw_assert_qos_lock_held.
Enabling supported_cross_device_rate_nodes now is safe, because
mlx5_esw_qos_vport_update_parent rejects cross-esw parent updates.
This will change in the next patch.
[1]
https://lore.kernel.org/netdev/20250213180134.323929-1-tariqt@nvidia.com/
Signed-off-by: Cosmin Ratiu <cratiu@nvidia.com>
Reviewed-by: Carolina Jubran <cjubran@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260701073254.754518-12-tariqt@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
In commit [1] the concept of the root node in the qos hierarchy was
removed due to a bug with how tx_share worked. The side effect is that
in many places, there are now corner cases related to parent handling.
However, since that change, support for tc_bw was added and now, with
upcoming cross-esw support, the code is about to become even more
complicated, increasing the number of such corner cases.
Bring back the concept of the root node, to which all esw vports and
nodes are connected to. This benefits multiple operations which can
assume there's always a valid parent and don't have to do ternary
gymnastics to determine the correct esw to talk to.
As side effect, there's no longer a need to store the groups in the
qos domain, since normalization can simply iterate over all children of
the root node. Normalization gets simplified as a result.
There should be no functionality changes as a result of this change.
[1] commit 330f0f6713a3 ("net/mlx5: Remove default QoS group and attach
vports directly to root TSAR")
Signed-off-by: Cosmin Ratiu <cratiu@nvidia.com>
Reviewed-by: Carolina Jubran <cjubran@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260701073254.754518-11-tariqt@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Qos cleanup is a complex affair, because of the two modes of operation
(legacy and switchdev).
Leaf QoS is removed:
1. In legacy mode by esw_vport_cleanup() -> mlx5_esw_qos_vport_disable()
2. In switchdev mode by mlx5_esw_offloads_devlink_port_unregister() ->
mlx5_esw_qos_vport_update_parent(). A little later in the same flow, the
calls in 1 happen but they are noops.
Zooming out a bit, from both mlx5_eswitch_disable_locked() and
mlx5_eswitch_disable_sriov() the leaves are destroyed before the nodes,
which is the reverse of what should be.
For SFs there's no devl_rate_nodes_destroy() call to unparent the
affected leaf.
Sanitize all of this by:
1. Destroying nodes before leaves in both legacy and switchdev mode.
2. Only removing vport qos from esw_vport_cleanup(), reachable from both
legacy and switchdev and also reachable by SF removal.
3. Unexpose mlx5_esw_qos_vport_update_parent(), which becomes internal
to qos.
4. Remove the WARN in mlx5_esw_qos_vport_disable().
This also takes care of a theoretical corner case, when
mlx5_esw_qos_vport_update_parent() tried to reattach the vport to
the original parent on failure, which can fail as well, leaving the
vport in a broken state.
Signed-off-by: Cosmin Ratiu <cratiu@nvidia.com>
Reviewed-by: Carolina Jubran <cjubran@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260701073254.754518-10-tariqt@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Previously, the master device of the uplink netdev was queried for its
maximum link speed from the QoS layer, requiring the uplink_netdev mutex
and possibly the RTNL (if the call originated from the TC matchall
layer).
Acquiring these locks here is risky, as lock cycles could form. The
locking for the QoS layer is about to change, so to avoid issues,
replace the code querying the LAG's max link speed with the existing
infrastructure added in commit [1].
This simplifies this part and avoids potential lock cycles.
One caveat is that there's a new edge case, when the bond device is not
fully formed to represent the LAG device, the speed isn't calculated and
is left at 0. This now handled explicitly.
[1] commit f0b2fde98065 ("net/mlx5: Add support for querying bond
speed")
Signed-off-by: Cosmin Ratiu <cratiu@nvidia.com>
Reviewed-by: Carolina Jubran <cjubran@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260701073254.754518-9-tariqt@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
The PF SR-IOV enable path caches VF pci_dev pointers in
dpiring_to_vfpcidev_lut[] by iterating with pci_get_device(). Those
entries do not own a reference, because the iterator drops the previous
device reference on each step. The cached pointer is then dereferenced
later when handling OCTEON_VF_FLR_REQUEST.
Replace the cached VF mapping with runtime lookup on the mailbox DPI
ring: derive the VF index from q_no, resolve the VF via exported PCI
IOV helpers, validate it with the PF pointer and VF ID, then issue
pcie_flr() and drop the reference with pci_dev_put(). Remove the
unused VF lookup table initialization and cleanup.
Fixes: ca6139ffc67ee ("liquidio CN23XX: sysfs VF config support")
Fixes: 8c978d059224 ("liquidio CN23XX: Mailbox support")
Signed-off-by: Yuho Choi <dbgh9129@gmail.com>
Link: https://patch.msgid.link/20260701040847.1897845-1-dbgh9129@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
The rnpgbe mailbox exchanges data through 32-bit MMIO registers in
little-endian wire format. The original code had two problems:
1. FW structs (with __le16/__le32 fields) were cast to (u32 *)
before reaching the mailbox transport, hiding the endian
annotations from sparse.
2. No cpu_to_le32()/le32_to_cpu() conversion was done between
CPU-endian MMIO values and the little-endian payload, causing
data corruption on big-endian systems.
Fix by adding the missing byte-order conversions in the transport
layer and introducing union wrappers (mbx_fw_cmd_req_u,
mbx_fw_cmd_reply_u) that overlay each FW struct with a __le32
dwords[] array. Callers fill named fields using cpu_to_le16/32(),
then pass dwords[] to the transport, which now takes explicit
__le32 * instead of u32 *. This eliminates all pointer casts on
the mailbox data path and lets sparse verify the conversions.
Fixes: 4543534c3ef5 ("net: rnpgbe: Add basic mbx ops support")
Signed-off-by: Dong Yibo <dong100@mucse.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20260701032208.1843156-2-dong100@mucse.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
brcmf_notify_auth_frame_rx() takes the frame length from the firmware
event and copies the frame body with the management header offset
subtracted:
u32 mgmt_frame_len = e->datalen - sizeof(struct brcmf_rx_mgmt_data);
...
memcpy(&mgmt_frame->u, frame,
mgmt_frame_len - offsetof(struct ieee80211_mgmt, u));
The only length check is e->datalen >= sizeof(*rxframe), so mgmt_frame_len
can be anything from 0 up. offsetof(struct ieee80211_mgmt, u) is 24. When
mgmt_frame_len is below that, the subtraction wraps as an unsigned value to
a huge length. The memcpy then runs far past the kzalloc'd buffer. A
malicious or malfunctioning AP can make the frame short during the
external SAE auth exchange, so this is a remotely triggered heap overflow.
Reject frames shorter than the management header offset before the copy.
Fixes: 66f909308a7c ("wifi: brcmfmac: cyw: support external SAE authentication in station mode")
Link: https://lore.kernel.org/r/178214417708.2368577.16740907093694208834@maoyixie.com
Cc: stable@vger.kernel.org
Co-developed-by: Kaixuan Li <kaixuan.li@ntu.edu.sg>
Signed-off-by: Kaixuan Li <kaixuan.li@ntu.edu.sg>
Signed-off-by: Maoyi Xie <maoyixie.tju@gmail.com>
Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Link: https://patch.msgid.link/20260627131313.3878893-1-maoyixie.tju@gmail.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
mac80211_hwsim_stop() drops any frames left in data->pending. The loop
currently checks skb_queue_empty() and then dequeues separately.
That split is racy with TX status handling, which can remove a pending
frame under the queue lock. If the last entry is removed after the empty
check, skb_dequeue() returns NULL and the stop path passes that NULL skb
to ieee80211_free_txskb().
Use skb_dequeue() as the loop condition instead. The dequeue result is the
object that stop owns and frees, and a concurrent status completion that
empties the queue simply makes the loop terminate.
Fixes: bd18de517923 ("mac80211_hwsim: drop pending frames on stop")
Assisted-by: Codex:gpt-5.5
Signed-off-by: Cen Zhang <zzzccc427@gmail.com>
Link: https://patch.msgid.link/20260706161822.921039-1-zzzccc427@gmail.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
mac80211_hwsim_free() removes each radio from hwsim_radios before calling
mac80211_hwsim_del_radio(), but leaves the matching hwsim_radios_rht entry
in place until the whole table is destroyed.
Other radio removal paths remove both the list entry and data->rht while
holding hwsim_radio_lock, before dropping the lock and deleting the radio.
Do the same here so the all-radio cleanup path follows the same object
visibility ordering.
This helper is used while all radios are being torn down, either after
callback users have already been unregistered or while module init is
unwinding, so no hwsim_radios_generation update is needed.
Assisted-by: Codex:gpt-5.5
Signed-off-by: Cen Zhang <zzzccc427@gmail.com>
Link: https://patch.msgid.link/20260706123756.343818-1-zzzccc427@gmail.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
brcmf_sdio_probe() stores the newly allocated bus in sdiodev->bus before
allocating the ordered workqueue. If that allocation fails, the function
jumps to fail and calls brcmf_sdio_remove().
brcmf_sdio_remove() unconditionally cancels bus->datawork. Initialize the
work item before the first failure path that can reach brcmf_sdio_remove(),
so the cleanup path always observes a valid work object.
This issue was found by our static analysis tool and then confirmed by
manual review of the probe error path and the remove-time work drain. The
problem pattern is an early setup failure that reaches a cleanup helper
which cancels an embedded work item before its initializer has run.
A QEMU PoC forced alloc_ordered_workqueue() to fail at the same point in
brcmf_sdio_probe(), before INIT_WORK(&bus->datawork) is reached. The
resulting fail path calls brcmf_sdio_remove(), and DEBUG_OBJECTS reports
the invalid work drain with brcmf_sdio_probe() and brcmf_sdio_remove() in
the stack.
Fixes: 9982464379e8 ("brcmfmac: make sdio suspend wait for threads to freeze")
Signed-off-by: Runyu Xiao <runyu.xiao@seu.edu.cn>
Acked-by: Arend van Spriel <arend.vanspriel@broadcom.com>
Link: https://patch.msgid.link/20260619064401.1048976-1-runyu.xiao@seu.edu.cn
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
rsi_prepare_beacon() copies the mac80211 beacon frame after
FRAME_DESC_SZ into a management skb whose usable tailroom may be smaller
than MAX_MGMT_PKT_SIZE after alignment.
Validate the beacon length against the actual tailroom before the copy
and skb_put(). Leave ownership of the management skb with the caller on
error, matching the existing rsi_send_beacon() cleanup path.
Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
Link: https://patch.msgid.link/20260705084824.68105-1-pengpeng@iscas.ac.cn
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
libipw_rx() reads skb->data[hdrlen + 3] to extract the WEP key index in
both the software-decrypt key selection path and the hardware-decrypted
IV/ICV strip path. In both places the existing guard only checks
skb->len >= hdrlen + 3, which proves bytes up to hdrlen + 2 but not the
byte at hdrlen + 3.
Require hdrlen + 4 bytes before reading that item in both paths. This is
a local source-boundary check only; it does not change the key index
semantics.
Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
Link: https://patch.msgid.link/20260705083519.23567-1-pengpeng@iscas.ac.cn
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
rsi_send_bgscan_probe_req() allocates room for struct
rsi_bgscan_probe plus MAX_BGSCAN_PROBE_REQ_LEN bytes, but copies the
entire mac80211-generated probe request skb after the fixed header.
The probe request length depends on scan IEs and is not checked
against the fixed firmware buffer.
Reject generated probe requests that do not fit the firmware command
buffer before copying them into the skb.
Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
Link: https://patch.msgid.link/20260704011231.45593-1-pengpeng@iscas.ac.cn
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
In monitor mode, lbs_hard_start_xmit() casts skb->data to a
radiotap TX header, skips that header, and then copies the 802.11
destination address from offset 4 in the remaining frame. The
generic length check only rejects zero-length and oversized skbs, so
a short monitor frame can be read past the end of the skb data.
Require enough bytes for the radiotap TX header and the destination
address field before using the monitor-mode header layout.
Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
Link: https://patch.msgid.link/20260704011140.37639-1-pengpeng@iscas.ac.cn
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
I have an hostapd setup with a
01:00.0 Network controller: Ralink corp. RT2790 Wireless 802.11n 1T/2R PCIe
The setup work fine on 6.18.26-gentoo
It breaks on 6.18.33-gentoo (and still broken on 6.18.37)
I found an hint in dmesg:
On 6.18.26-gentoo I see:
May 31 15:48:45 trash01 kernel: ieee80211 phy0: rt2x00_set_rf: Info - RF chipset 0003 detected
On 6.18.33-gentoo I see:
May 31 15:22:57 trash01 kernel: ieee80211 phy0: rt2x00_set_rf: Info - RF chipset 0006 detected
The RF chipset seems badly detected.
The problem was the EEPROM which was badly initialized.
Probably the origin was in some PCI change but unfortunately I couldn't play
to bisect/reboot often the board with this card to do it.
Signed-off-by: Corentin Labbe <clabbe@baylibre.com>
Acked-by: Stanislaw Gruszka <stf_xl@wp.pl>
Link: https://patch.msgid.link/20260703134932.3786771-1-clabbe@baylibre.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
rsi_hal_load_key() copies tx_mic_key and rx_mic_key from data[16] and
data[24] whenever key data is present. Those offsets are only part of
the 32-byte TKIP key layout. Shorter keys used by other ciphers, such as
CCMP, do not provide those bytes, so the unconditional copies can read
past the supplied key buffer.
Only copy the MIC keys for TKIP, and reject malformed TKIP keys that are
shorter than the expected 32-byte layout.
Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
Link: https://patch.msgid.link/20260701053414.34015-1-pengpeng@iscas.ac.cn
[drop useless length check]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
p54_rx_eeprom_readback() copies the requested EEPROM slice out of a
device-supplied readback frame without checking that the skb actually holds
that many bytes. Commit da1b9a55ff11 ("wifi: p54: prevent buffer-overflow in
p54_rx_eeprom_readback()") closed the destination overflow by copying a
fixed priv->eeprom_slice_size (and rejecting a mismatched advertised len),
but the source side is still unbounded: nothing verifies the frame is long
enough to supply that many bytes.
A malicious USB device can send a short frame whose advertised len matches
priv->eeprom_slice_size while the payload is truncated. The equality check
passes and memcpy() reads past the end of the skb, leaking adjacent heap:
BUG: KASAN: slab-out-of-bounds in p54_rx (drivers/net/wireless/intersil/p54/txrx.c:507)
Read of size 1016 at addr ffff88800f077114 by task swapper/0/0
Call Trace:
<IRQ>
...
__asan_memcpy (mm/kasan/shadow.c:105)
p54_rx (drivers/net/wireless/intersil/p54/txrx.c:507)
p54u_rx_cb (drivers/net/wireless/intersil/p54/p54usb.c:163)
__usb_hcd_giveback_urb (drivers/usb/core/hcd.c:1657)
dummy_timer (drivers/usb/gadget/udc/dummy_hcd.c:2005)
...
</IRQ>
The buggy address belongs to the object at ffff88800f0770c0
which belongs to the cache skbuff_small_head of size 704
The buggy address is located 84 bytes inside of
allocated 704-byte region [ffff88800f0770c0, ffff88800f077380)
Check that the slice fits in the skb before copying.
Fixes: 7cb770729ba8 ("p54: move eeprom code into common library")
Reported-by: Weiming Shi <bestswngs@gmail.com>
Assisted-by: Claude:claude-opus-4-8
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Acked-by: Christian Lamparter <chunkeey@gmail.com>
Link: https://patch.msgid.link/20260628000510.4152481-1-xmei5@asu.edu
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
Injected HT and VHT rates store an MCS value in rates[0].idx rather
than an index into the legacy bitrate table. hwsim nevertheless passes
these rates to ieee80211_get_tx_rate() while generating monitor frames
and timestamps.
A crafted injected frame can therefore read beyond the bitrate table.
If the resulting bitrate is zero, mac80211_hwsim_write_tsf() also
divides by zero, as observed by syzbot.
Use ieee80211_get_tx_rate() only for legacy rates. The existing fallback
continues to supply a conservative bitrate where hwsim does not yet
calculate MCS rates.
Reported-by: syzbot+21629c14aa749636db9d@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=21629c14aa749636db9d
Signed-off-by: Yousef Alhouseen <alhouseenyousef@gmail.com>
Link: https://patch.msgid.link/20260628002537.23550-1-alhouseenyousef@gmail.com
[drop wrong Fixes tag]
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
helper_firmware_cb() neglects to free the single-stage firmware image
after a successful async load, leading to a memory leak in the USB
firmware-download path.
Fix this memory leak by calling release_firmware() immediately after
lbs_fw_loaded() returns.
The bug was first flagged by an experimental analysis tool we are
developing for kernel memory-management bugs while analyzing
v6.13-rc1. The tool is still under development and is not yet publicly
available. Manual inspection confirms that the bug is still present in
the current wireless tree.
An x86_64 allyesconfig build showed no new warnings. As we do not have
compatible Libertas USB hardware for exercising this firmware-download
path, no runtime testing was able to be performed.
Fixes: 1dfba3060fe7 ("libertas: move firmware lifetime handling to firmware.c")
Signed-off-by: Dawei Feng <dawei.feng@seu.edu.cn>
Link: https://patch.msgid.link/20260624085343.575508-1-dawei.feng@seu.edu.cn
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
lbtf_free_adapter() calls timer_delete(&priv->command_timer), which does
not wait for a running command_timer_fn() callback. lbtf_free_adapter()
runs on the teardown path right before ieee80211_free_hw() frees priv,
both in lbtf_remove_card() and in the probe error path. command_timer is
armed by mod_timer() in lbtf_cmd() whenever a firmware command is sent.
command_timer_fn() dereferences priv. If a command times out as the
device is removed, command_timer_fn() runs concurrently with teardown and
dereferences priv after it has been freed.
This is the same use-after-free that commit 03cc8f90d053 ("wifi: libertas:
fix use-after-free in lbs_free_adapter()") fixed in the sibling libertas
driver. The libertas_tf variant has the identical pattern and was left
unchanged. Use timer_delete_sync() so any in-flight callback completes
before priv is freed.
Fixes: 06b16ae53192 ("libertas_tf: main.c, data paths and mac80211 handlers")
Cc: stable@vger.kernel.org
Signed-off-by: Maoyi Xie <maoyixie.tju@gmail.com>
Link: https://patch.msgid.link/178211481807.2212567.8773346114561900100@maoyixie.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
hwsim_virtio_rx_work() passes the virtqueue used-ring length reported by
the device straight to skb_put() on a fixed-size receive skb. A backend
reporting a length larger than the skb tailroom drives skb_put() past the
buffer end and hits skb_over_panic() -- a host-triggerable guest panic
(denial of service).
Clamp the length to the skb's available room before skb_put(). A
conforming device never reports more than the posted buffer size, so valid
frames are unaffected; a truncated over-report then fails the
length/header checks in hwsim_virtio_handle_cmd() and is dropped, so
truncating rather than dropping here cannot be turned into a parsing
problem.
Fixes: 5d44fe7c9808 ("mac80211_hwsim: add frame transmission support over virtio")
Signed-off-by: Bryam Vargas <hexlabsecurity@proton.me>
Link: https://patch.msgid.link/20260620-b4-disp-474bee37-v1-1-1a4d37f3e2d4@proton.me
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
The memory allocated in the ipw2100_alloc_device() function is not freed
in some of the error paths in ipw2100_pci_init_one(). Fix that by
converting the direct return into a goto to the error path return.
The error path when pci_enable_device() fails cannot jump to fail, since
at this point priv is not set, so perform error handling inline.
Fixes: 2c86c275015c ("Add ipw2100 wireless driver.")
Signed-off-by: Abdun Nihaal <nihaal@cse.iitm.ac.in>
Link: https://patch.msgid.link/20260620065242.93798-1-nihaal@cse.iitm.ac.in
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
amt_get_size() incorrectly used sizeof(struct iphdr) for the sizes of
IFLA_AMT_DISCOVERY_IP, IFLA_AMT_REMOTE_IP, and IFLA_AMT_LOCAL_IP.
These attributes contain IPv4 addresses (__be32), not full IP headers.
Replace sizeof(struct iphdr) with sizeof(__be32) to avoid over-allocating
netlink message space.
Fixes: b9022b53adad ("amt: add control plane of amt interface")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Link: https://patch.msgid.link/20260701122329.3562825-1-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
The p54 wireless wiki links (wireless.wiki.kernel.org) return 404; the
content moved to the Sphinx documentation site. Point them at the
current wireless.docs.kernel.org pages.
Signed-off-by: Anas Khan <anxkhn28@gmail.com>
Acked-by: Christian Lamparter <chunkeey@gmail.com>
Link: https://patch.msgid.link/20260702102325.63955-1-anxkhn28@gmail.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
wlcore_alloc_hw() uses __get_free_pages() to allocate TX aggregation
and firmware log buffers used for software data staging.
These buffer can be allocated with kmalloc() as there's nothing special
about them to go directly to the page allocator.
kmalloc() provides a better API that does not require ugly casts and
kfree() does not need to know the size of the freed object.
Performance difference between kmalloc() and __get_free_pages() is not
measurable as both allocators take an object/page from a per-CPU list for
fast path allocations.
For the slow path the performance is anyway determined by the amount of
reclaim involved rather than by what allocator is used.
Replace use of __get_free_pages() with kzalloc() and free_pages() with
kfree().
Link: https://lore.kernel.org/all/635405e4-9423-4a25-a6e7-e03c8ea0bcbe@redhat.com
Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Link: https://patch.msgid.link/20260701-b4-drivers-wireless-v1-4-60264cdf2efe@kernel.org
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
mwifiex debugfs functions allocate buffers for formatting debug output
text using get_zeroed_page().
These buffers can be allocated with kmalloc() as there's nothing special
about them to go directly to the page allocator.
kmalloc() provides a better API that does not require ugly casts and
kfree() does not need to know the size of the freed object.
Performance difference between kmalloc() and __get_free_pages() is not
measurable as both allocators take an object/page from a per-CPU list for
fast path allocations.
For the slow path the performance is anyway determined by the amount of
reclaim involved rather than by what allocator is used.
Replace use of get_zeroed_page() with kzalloc() and free_page() with
kfree().
Link: https://lore.kernel.org/all/635405e4-9423-4a25-a6e7-e03c8ea0bcbe@redhat.com
Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Reviewed-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Link: https://patch.msgid.link/20260701-b4-drivers-wireless-v1-3-60264cdf2efe@kernel.org
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
libertas debugfs functions allocate buffers for formatting debug
output text using get_zeroed_page().
These buffers can be allocated with kmalloc() as there's nothing special
about them to go directly to the page allocator.
kmalloc() provides a better API that does not require ugly casts and
kfree() does not need to know the size of the freed object.
Performance difference between kmalloc() and __get_free_pages() is not
measurable as both allocators take an object/page from a per-CPU list for
fast path allocations.
For the slow path the performance is anyway determined by the amount of
reclaim involved rather than by what allocator is used.
Replace use of get_zeroed_page() with kzalloc() and free_page() with
kfree().
Link: https://lore.kernel.org/all/635405e4-9423-4a25-a6e7-e03c8ea0bcbe@redhat.com
Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Link: https://patch.msgid.link/20260701-b4-drivers-wireless-v1-2-60264cdf2efe@kernel.org
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
b43* debugfs functions allocate 16 KiB buffers for formatting debug output
text using __get_free_pages().
kzalloc() provides a better API that does not require ugly casts and
kfree() does not need to know the size of the freed object and for 16 Kib
allocation kzalloc() will anyway delegate it to buddy.
Replace use of __get_free_pages() with kzalloc().
Link: https://lore.kernel.org/all/635405e4-9423-4a25-a6e7-e03c8ea0bcbe@redhat.com
Signed-off-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Link: https://patch.msgid.link/20260701-b4-drivers-wireless-v1-1-60264cdf2efe@kernel.org
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
There are only two drivers left setting it, but they're
both also setting WIPHY_FLAG_SUPPORTS_MLO for the relevant
devices, so we can now remove WIPHY_FLAG_DISABLE_WEXT.
Link: https://patch.msgid.link/20260619142107.150f1bbe3b83.I9ff3d419bad54313c76fa4c3485148c122e67fb3@changeid
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
When ingress deaggregation is disabled, rmnet_map_ingress_handler() passes
the skb straight to __rmnet_map_ingress_handler(), skipping the length
validation that rmnet_map_deaggregate() performs on the aggregated path.
The parser then dereferences the MAP header and csum header/trailer based on
the on-wire pkt_len without checking skb->len, so a short frame is read out
of bounds:
BUG: KASAN: slab-out-of-bounds in rmnet_map_checksum_downlink_packet
Read of size 1 at addr ffff88801118ed00 by task exploit/147
Call Trace:
...
rmnet_map_checksum_downlink_packet (drivers/net/ethernet/qualcomm/rmnet/rmnet_map_data.c:413)
__rmnet_map_ingress_handler (drivers/net/ethernet/qualcomm/rmnet/rmnet_handlers.c:96)
rmnet_rx_handler (drivers/net/ethernet/qualcomm/rmnet/rmnet_handlers.c:129)
__netif_receive_skb_core.constprop.0 (net/core/dev.c:6089)
netif_receive_skb (net/core/dev.c:6460)
tun_get_user (drivers/net/tun.c:1955)
tun_chr_write_iter (drivers/net/tun.c:2001)
vfs_write (fs/read_write.c:688)
ksys_write (fs/read_write.c:740)
do_syscall_64 (arch/x86/entry/syscall_64.c:94)
...
Factor that validation out of rmnet_map_deaggregate() into
rmnet_map_validate_packet_len() and run it on the no-aggregation path too.
The MAP header is bounds-checked first, since this path can receive a frame
shorter than the header.
Fixes: ceed73a2cf4a ("drivers: net: ethernet: qualcomm: rmnet: Initial implementation")
Reported-by: Weiming Shi <bestswngs@gmail.com>
Suggested-by: Subash Abhinov Kasiviswanathan <subash.a.kasiviswanathan@oss.qualcomm.com>
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Reviewed-by: Subash Abhinov Kasiviswanathan <subash.a.kasiviswanathan@oss.qualcomm.com>
Link: https://patch.msgid.link/20260630174110.2003121-1-xmei5@asu.edu
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Add NL80211_EXT_FEATURE_PROBE_AP to allow drivers to advertise
support for probing the associated AP from STA/P2P-client mode.
Extend nl80211_probe_peer() to accept STA/P2P-client interfaces
when the driver advertises NL80211_EXT_FEATURE_PROBE_AP; in that
case the MAC attribute must be omitted (the peer is implied by
the association).
Update cfg80211_probe_status() to accept an optional peer address
and a link_id parameter (-1 for non-MLO), and include
NL80211_ATTR_MLO_LINK_ID in the event when link_id >= 0.
Update all callers.
Signed-off-by: Priyansha Tiwari <priyansha.tiwari@oss.qualcomm.com>
Link: https://patch.msgid.link/20260611062225.2144241-3-pritiwa@qti.qualcomm.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
Rename NL80211_CMD_PROBE_CLIENT to NL80211_CMD_PROBE_PEER in the UAPI
enum and retain NL80211_CMD_PROBE_CLIENT as a compatibility alias.
Rename the .probe_client cfg80211_ops callback to .probe_peer and
update all in-tree users (wil6210, mwifiex) and mac80211 so the
tree continues to build after this change.
Signed-off-by: Priyansha Tiwari <priyansha.tiwari@oss.qualcomm.com>
Link: https://patch.msgid.link/20260611062225.2144241-2-pritiwa@qti.qualcomm.com
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
qede_rx_build_skb() and qede_tpa_rx_build_skb() do not check for a
NULL return from qede_build_skb(). When it returns NULL under memory
pressure, the functions still consume a BD from the ring before
returning NULL. The callers then recycle additional BDs, resulting in
one extra BD being consumed (off-by-one). This desynchronizes the BD
ring, which can corrupt DMA page reference counts and lead to SLUB
freelist corruption.
Commit 4e910dbe3650 ("qede: confirm skb is allocated before using")
added a NULL check inside qede_build_skb() to prevent a NULL pointer
dereference, but did not address the missing NULL checks in the
callers, making this off-by-one reachable.
Fix this by adding NULL checks for the return value of
qede_build_skb() in both qede_rx_build_skb() and
qede_tpa_rx_build_skb(), returning NULL immediately before any BD ring
manipulation.
Fixes: 8a8633978b84 ("qede: Add build_skb() support.")
Signed-off-by: Shigeru Yoshida <syoshida@redhat.com>
Reviewed-by: Jamie Bainbridge <jamie.bainbridge@gmail.com>
Link: https://patch.msgid.link/20260630164623.3152625-1-syoshida@redhat.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Update amt_fill_info() to run under RCU read lock instead of RTNL.
The AMT device configuration fields (mode, relay_port, gw_port, local_ip,
discovery_ip, max_tunnels) and stream_dev pointer are initialized during
device creation (amt_newlink) and are immutable. Accessing them locklessly
is safe. The stream_dev net_device structure is protected from being freed
by RCU.
The only field that can change concurrently is amt->remote_ip, which is
updated in the packet receive path (amt_advertisement_handler) and
workqueue (amt_req_work). Add READ_ONCE()/WRITE_ONCE() annotations
around amt->remote_ip to prevent data races.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Link: https://patch.msgid.link/20260701125016.3650708-1-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
RX prevention disabled
Add ethtool support to query and configure the PFC (Priority Flow Control)
storm prevention timeout. When TX continuously sends PFC frames, the peer
end is suppressed from sending packets. If this persists, a PFC frame storm
may occur.
This feature allows configuring a timeout to prevent such storms.
Signed-off-by: Hao Chen <chenhao418@huawei.com>
Signed-off-by: Jijie Shao <shaojijie@huawei.com>
Link: https://patch.msgid.link/20260630134043.1532431-1-shaojijie@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
rt2x00lib_probe_dev() uses the full rt2x00lib_remove_dev() teardown for
all probe failures. However, drv_data allocation and workqueue allocation
can fail before intf_work, autowakeup_work and sleep_work have been
initialized.
Do not enter the full remove path until the probe has reached the point
where those work items are set up. Return directly for drv_data allocation
failure, and use a small early cleanup path for workqueue allocation
failure.
This issue was found by our static analysis tool and then confirmed by
manual review of rt2x00lib_probe_dev() and rt2x00lib_remove_dev(). The
early probe exits should not call a common teardown path that assumes the
later work setup has already completed.
A QEMU PoC forced alloc_ordered_workqueue() to fail before the work
initializers are reached. The resulting fail path entered
rt2x00lib_remove_dev(), and DEBUG_OBJECTS reported invalid work drains with
rt2x00lib_probe_dev() and rt2x00lib_remove_dev() in the stack.
Fixes: 1ebbc48520a0 ("rt2x00: Introduce concept of driver data in struct rt2x00_dev.")
Fixes: 0439f5367c8d ("rt2x00: Move TX/RX work into dedicated workqueue")
Cc: stable@vger.kernel.org
Signed-off-by: Runyu Xiao <runyu.xiao@seu.edu.cn>
Link: https://patch.msgid.link/20260619073104.1809161-1-runyu.xiao@seu.edu.cn
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
In order for the firmware to sleep, the driver has to confirm a
previously received sleep request. The normal sequence of evets goes
like this:
EVENT_SLEEP -> adapter->ps_state = PS_STATE_PRE_SLEEP -> sleep-confirm
-> SLEEP -> EVENT_AWAKE -> AWAKE.
Before sending the sleep-confirm command, the driver must make sure
there are no commands either running or waiting to be completed.
mwifiex_ret_802_11_associate() unconditionally sets
ps_state = PS_STATE_AWAKE when it processes the association command
response, outside of the normal powersave management flow. If
EVENT_SLEEP arrives while the association command is in flight,
ps_state is PRE_SLEEP when the association command response is parsed,
and the forced AWAKE overwrites it. The deferred sleep-confirm is
never sent.
A subsequent scan_start command is correctly acknowledged, but the
firmware doesn't generate scan_result events. The scan request never
finishes, and additional requests from userspace fail with -EBUSY.
After testing on both IW412 and W8997, I could only trigger the bug on
the IW412 and observed the firmwares behave differently. On the IW412
the firmware still sends EVENT_SLEEP while the authentication /
association process is ongoing. A W8997 under the same
conditions seems to suppress power-save for the duration of the
association, so PRE_SLEEP never coincided with the association response
even after extended periods of testing using the loops
described below (>12hours).
On the IW412, the delay between commands that triggers an EVENT_SLEEP
was empirically determined to be ~20ms. This delay can naturally occur
when the driver is outputting debugging information
(debug_mask = 0x00000037), in which situation the busy scans issue is
repeatable while running "test 1)" as described below. If the delay
between commands is less than ~20ms, the firmware stays awake and
the issue was not reproducible running the same test.
The host_mlme=false path also behaves differently. In this case, the
entire authentication / association transaction is executed by one
command (HostCmd_CMD_802_11_ASSOCIATE), and the firmware doesn't emit
EVENT_SLEEP while the command is running.
Remove the assignment so the ps_state is only manipulated in the paths
that are related to powersave event handling and on the main workqueue
for correct sleep confirmation.
The following loop tests were performed (with debugging output enabled):
1) force roaming between two AP's, one 5GHz and one 2.4GHz, same
SSID. Use wpa_cli to trigger the roaming behavior, sleep 2s
between iterations.
2) force a disconnection to AP 1 and a connection to AP 2, test
scan. Use wpa_cli to trigger the connection changes, sleep 2s
between iterations.
Each test ran in each device for at least 3 hours.
Fixes: 5e6e3a92b9a4 ("wireless: mwifiex: initial commit for Marvell mwifiex driver")
Cc: stable@vger.kernel.org
Assisted-by: Claude:claude-opus-4-7
Signed-off-by: Rafael Beims <rafael.beims@toradex.com>
Reviewed-by: Jeff Chen <jeff.chen_1@nxp.com>
Link: https://patch.msgid.link/20260612122547.1586872-2-rafael@beims.me
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
When host MLME is enabled, mwifiex_cfg80211_authenticate() transmits the
authentication frame on a remain-on-channel (ROC) reservation so that the
frame is sent on the target BSS's channel. The ROC is only configured
when priv->auth_flag is zero.
priv->auth_flag is set to HOST_MLME_AUTH_PENDING when the auth frame is
queued and advances to HOST_MLME_AUTH_DONE once authentication
completes. It is only cleared back to zero on a disconnect, deauth or
timeout path; nothing clears it when an association succeeds. It therefore
stays at HOST_MLME_AUTH_DONE for the whole connected session.
When the station later roams to a BSS on a different channel, the next
authentication finds auth_flag != 0, skips the ROC setup, and the auth
frame is transmitted on the currently-associated channel instead of the
target's channel. Authentication times out on the new AP and the device
stays connected to the original AP.
Gate the ROC setup on HOST_MLME_AUTH_PENDING instead of on auth_flag
being completely clear. This re-arms the remain-on-channel for every new
authentication attempt, while still suppressing a redundant ROC during
the multi-frame SAE exchange, where auth_flag stays PENDING between the
commit and confirm frames.
This change was tested in 3 different devices:
Verdin AM62 (IW412 SD-UART) - (16.92.21.p142)
Verdin iMX8MM (W8997 SD-SD) - (16.68.1.p197)
Verdin iMX8MP (W8997 SD-UART) - (16.92.21.p137)
There following loop tests were performed:
1) force roaming between two AP's, one 5GHz and one 2.4GHz, same
SSID. Use wpa_cli to trigger the roaming behavior, sleep 2s
between iterations.
2) force a disconnection to AP 1 and a connection to AP 2, test
scan. Use wpa_cli to trigger the connection changes, sleep 2s
between iterations.
Each test ran in each device for at least 3 hours.
Fixes: 36995892c271 ("wifi: mwifiex: add host mlme for client mode")
Cc: stable@vger.kernel.org
Assisted-by: Claude:claude-opus-4-7
Signed-off-by: Rafael Beims <rafael.beims@toradex.com>
Reviewed-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Link: https://patch.msgid.link/20260610150021.1018611-1-rafael@beims.me
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
|
|
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.
Fixes: 71c9de995260 ("net: microchip: sparx5: Add VCAP locking to protect rules")
Signed-off-by: Jens Emil Schulz Østergaard <jensemil.schulzostergaard@microchip.com>
Link: https://patch.msgid.link/20260630-microchip_fix_vcap_locking-v1-1-f60a4596734d@microchip.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
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.
Fixes: fa691d0c9c08 ("net/mlx5e: Allocate per-channel stats dynamically at first usage")
Signed-off-by: Feng Liu <feliu@nvidia.com>
Reviewed-by: Eran Ben Elisha <eranbe@nvidia.com>
Reviewed-by: Cosmin Ratiu <cratiu@nvidia.com>
Reviewed-by: Nimrod Oren <noren@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260630115151.729219-4-tariqt@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
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.
Fixes: cef35af34d6d ("net/mlx5e: Add mlx5e HV VHCA stats agent")
Signed-off-by: Feng Liu <feliu@nvidia.com>
Reviewed-by: Eran Ben Elisha <eranbe@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260630115151.729219-3-tariqt@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
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.
Fixes: fa691d0c9c08 ("net/mlx5e: Allocate per-channel stats dynamically at first usage")
Signed-off-by: Feng Liu <feliu@nvidia.com>
Reviewed-by: Eran Ben Elisha <eranbe@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260630115151.729219-2-tariqt@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Add READ_ONCE()/WRITE_ONCE() annotations on fields that can be
changed concurrently in macsec_changelink() and macsec_update_offload():
- secy->key_len
- secy->xpn
- tx_sc->encoding_sa
- tx_sc->encrypt
- secy->protect_frames
- tx_sc->send_sci
- tx_sc->end_station
- tx_sc->scb
- secy->replay_protect
- secy->validate_frames
- secy->replay_window
- macsec->offload
This allows macsec_fill_info() to run locklessly without RTNL.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Sabrina Dubroca <sd@queasysnail.net>
Cc: Andrew Lunn <andrew+netdev@lunn.ch>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Reviewed-by: Sabrina Dubroca <sd@queasysnail.net>
Link: https://patch.msgid.link/20260701094341.3218199-1-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
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
Fixes: 971b28accc09 ("net/mlx5: LAG, replace mlx5_get_dev_index with LAG sequence number")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260630112917.698313-4-tariqt@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
mlx5_mpesw_work() returned without calling complete() when
mlx5_lag_get_devcom_comp() returned NULL. A caller that queued the
work and waited on mpesww->comp would block indefinitely.
Funnel the early-return path through a new "complete" label so the
waiter is always woken.
Fixes: b430c1b4f63b ("net/mlx5: Replace global mlx5_intf_lock with HCA devcom component lock")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260630112917.698313-3-tariqt@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
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.
Fixes: ddbb5ddc43ad ("net/mlx5: LAG, Refactor lag logic")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260630112917.698313-2-tariqt@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Add READ_ONCE()/WRITE_ONCE() annotations on vlan->mode, vlan->flags,
vlan->bc_queue_len_req and port->bc_cutoff.
Fill IFLA_MACVLAN_MACADDR_DATA nested attribute and compute
on the fly the precise number of elements we put in it,
to fill an accurate IFLA_MACVLAN_MACADDR_COUNT attribute
as some user space applications could depend on its value
and the attributes order.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Link: https://patch.msgid.link/20260701082214.2974946-3-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
Both fields can be changed in macvlan_changelink() while being read
locklessly.
Add READ_ONCE()/WRITE_ONCE() annotations.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Link: https://patch.msgid.link/20260701082214.2974946-2-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/ukleinek/linux
Pull mod_devicetable.h header split from Uwe Kleine-König:
"Split <linux/mod_devicetable.h> in per subsystem headers
<linux/mod_devicetable.h> is included transitively in nearly every
driver in an x86_64 allmodconfig build of v7.1:
$ find drivers -name \*.o -not -name \*.mod.o | wc -l
21330
$ find drivers -name \*.o.cmd -not -name \*.mod.o.cmd | xargs grep -l mod_devicetable.h | wc -l
17038
The result of this mixture of different and unrelated subsystem
details is that even when touching an obscure device id struct most of
the kernel needs to be recompiled. Given that each driver typically
only needs one or two of these structures, splitting into per
subsystem headers and only including what is really needed reduces the
amount of needed recompilation.
This split is implemented in the first commit and then after some
preparatory work in the following commits, the last two replace
includes of <linux/mod_devicetable.h> by the actually needed more
specific headers.
There are still a few instances left, but the ones with high impact
(that is in headers that are used a lot) and the easy ones (.c files)
are handled. These remaining includes will be addressed during the
next merge window"
* tag 'device-id-rework' of git://git.kernel.org/pub/scm/linux/kernel/git/ukleinek/linux:
Replace <linux/mod_devicetable.h> by more specific <linux/device-id/*.h> (c files)
Replace <linux/mod_devicetable.h> by more specific <linux/device-id/*.h> (headers)
parisc: #include <linux/compiler.h> for unlikely() in <asm/ptrace.h>
media: em28xx: Add include for struct usb_device_id
LoongArch: KVM: Add include defining struct cpu_feature
ALSA: hda/core: Add include defining struct hda_device_id
usb: dwc2: Add include defining struct pci_device_id
platform/x86: int3472: Add include defining struct dmi_system_id
platform/x86: x86-android-tablets: Add include defining struct dmi_system_id
i2c: Let i2c-core.h include <linux/i2c.h>
of: Explicitly include <linux/types.h> and <linux/err.h>
platform/x86: msi-ec: Ensure dmi_system_id is defined
usb: serial: Include <linux/usb.h> in <linux/usb/serial.h>
driver core: platform: Include header for struct platform_device_id
driver: core: Include headers for acpi_device_id and of_device_id for struct device_driver
media: ti: vpe: #include <linux/platform_device.h> explicitly
mod_devicetable.h: Split into per subsystem headers
|
