Add support to request SoC stat from firmware through HTT stat
type 38. This stat gives drop count of SoC.
Note: MCC firmware version -
WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 does not
support tag HTT_STATS_SOC_TXRX_STATS_COMMON_TAG(125).
Sample output:
-------------
echo 38 > /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats_type
cat /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats
HTT_SOC_COMMON_STATS_TLV:
soc_drop_count = 0x0000000000000000
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Signed-off-by: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Signed-off-by: Roopni Devanathan <quic_rdevanat@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Acked-by: Kalle Valo <kvalo@kernel.org>
Link: https://patch.msgid.link/20241115062854.1919672-4-quic_rdevanat@quicinc.com
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Add support to request counters and Transmission Power Control
(TPC) stats through HTT stats type 37. These stats give
information about counters like received packet count, CRC pass
count, error count, transmit abort count, etc., about counter reset
like reset cause, channel frequency, number and mode, channel flags,
etc., about TPC like transmit power scale, maximum transmit power,
gain cap, EIRP, etc.
Note: MCC firmware version -
WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 does not
support HTT stats type 37, i.e., the firmware does not respond to the
command requesting stats.
Sample output:
-------------
echo 37 > /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats_type
cat /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats
HTT_PHY_STATS_TLV:
bdf_nf_chain[0] = -92
bdf_nf_chain[1] = -94
bdf_nf_chain[2] = -94
bdf_nf_chain[3] = -93
.....
HTT_PHY_COUNTERS_TLV:
rx_ofdma_timing_err_cnt = 18068
rx_cck_fail_cnt = 0
mactx_abort_cnt = 2612
macrx_abort_cnt = 0
.....
HTT_PHY_RESET_STATS_TLV:
pdev_id = 0
chan_mhz = 0
chan_band_center_freq1 = 0
chan_band_center_freq2 = 0
.....
HTT_PHY_RESET_COUNTERS_TLV:
pdev_id = 0
cf_active_low_fail_cnt = 0
cf_active_low_pass_cnt = 0
phy_off_through_vreg_cnt = 0
.....
HTT_PHY_TPC_STATS_TLV:
pdev_id = 0
tx_power_scale = 0
tx_power_scale_db = 0
min_negative_tx_power = 0
.....
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Signed-off-by: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Signed-off-by: Roopni Devanathan <quic_rdevanat@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Acked-by: Kalle Valo <kvalo@kernel.org>
Link: https://patch.msgid.link/20241115062854.1919672-3-quic_rdevanat@quicinc.com
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Add support to request downlink pager stats from firmware through HTT
stats type 36. These stats give paging information like number of pages,
their timestamp, number of locked and free pages, synchronous and
asynchronous locked pages.
Note: MCC firmware version -
WLAN.HMT.1.0-03427-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.15378.4 responds to
the event requesting stats, but it does not give any data.
Sample output:
-------------
echo 36 > /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats_type
cat /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats
HTT_DLPAGER_STATS_TLV:
ASYNC locked pages = 2
SYNC locked pages = 0
Total locked pages = 2
Total free pages = 127
LOCKED PAGES HISTORY
last_locked_page_idx = 0
Index - 0 ; Page Number - 8495 ; Num of pages - 1 ; Timestamp - 4031009360us
Index - 1 ; Page Number - 7219 ; Num of pages - 2 ; Timestamp - 885379515us
Index - 2 ; Page Number - 0 ; Num of pages - 0 ; Timestamp - 0us
Index - 3 ; Page Number - 0 ; Num of pages - 0 ; Timestamp - 0us
.....
UNLOCKED PAGES HISTORY
last_unlocked_page_idx = 0
Index - 0 ; Page Number - 7144 ; Num of pages - 2 ; Timestamp - 4032070008us
Index - 1 ; Page Number - 7214 ; Num of pages - 2 ; Timestamp - 885379512us
Index - 2 ; Page Number - 0 ; Num of pages - 0 ; Timestamp - 0us
Index - 3 ; Page Number - 0 ; Num of pages - 0 ; Timestamp - 0us
.....
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Signed-off-by: Dinesh Karthikeyan <quic_dinek@quicinc.com>
Signed-off-by: Roopni Devanathan <quic_rdevanat@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://patch.msgid.link/20241115062854.1919672-2-quic_rdevanat@quicinc.com
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Some ath9k chips can, seemingly at random, end up in a state which can
be described as "deaf". No or nearly no interrupts are generated anymore
for incoming packets. Existing links either break down after a while and
new links will not be established.
The circumstances leading to this "deafness" is still unclear, but some
particular chips (especially 2-stream 11n SoCs, but also others) can go
'deaf' when running AP or mesh (or both) after some time. It's probably
a hardware issue, and doing a channel scan to trigger a chip
reset (which one normally can't do on an AP interface) recovers the
hardware.
The only way the driver can detect this state, is by detecting if there
has been no RX activity for a while. In this case we can proactively
reset the chip (which only takes a small number of milliseconds, so
shouldn't interrupt things too much if it has been idle for several
seconds), which functions as a workaround.
OpenWrt, and various derivatives, have been carrying versions of this
workaround for years, that were never upstreamed. One version[0],
written by Felix Fietkau, used a simple counter and only reset if there
was precisely zero RX activity for a long period of time. This had the
problem that in some cases a small number of interrupts would appear
even if the device was otherwise not responsive. For this reason,
another version[1], written by Simon Wunderlich and Sven Eckelmann, used
a time-based approach to calculate the average number of RX interrupts
over a longer (four-second) interval, and reset the chip when seeing
less than one interrupt per second over this period. However, that
version relied on debugfs counters to keep track of the number of
interrupts, which means it didn't work at all if debugfs was not
enabled.
This patch unifies the two versions: it uses the same approach as Felix'
patch to count the number of RX handler invocations, but uses the same
time-based windowing approach as Simon and Sven's patch to still handle
the case where occasional interrupts appear but the device is otherwise
deaf.
Since this is based on ideas by all three people, but not actually
directly derived from any of the patches, I'm including Suggested-by
tags from Simon, Sven and Felix below, which should hopefully serve as
proper credit.
[0] https://patchwork.kernel.org/project/linux-wireless/patch/20170125163654.66431-3-nbd@nbd.name/
[1] https://patchwork.kernel.org/project/linux-wireless/patch/20161117083614.19188-2-sven.eckelmann@open-mesh.com/
Suggested-by: Simon Wunderlich <sw@simonwunderlich.de>
Suggested-by: Sven Eckelmann <se@simonwunderlich.de>
Suggested-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Tested-by: Sven Eckelmann <se@simonwunderlich.de>
Reviewed-by: Sven Eckelmann <se@simonwunderlich.de>
Tested-by: Issam Hamdi <ih@simonwunderlich.de>
Acked-by: Simon Wunderlich <sw@simonwunderlich.de>
Link: https://patch.msgid.link/20241106-ath9k-deaf-detection-v1-1-736a150d2425@redhat.com
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
The following error messages were encountered while parsing fragmented RX
packets for WCN6750/WCN6855:
ath11k 17a10040.wifi: invalid return buffer manager 4
This issue arose due to a hardcoded check for HAL_RX_BUF_RBM_SW3_BM
introduced in 'commit 71c748b5e0 ("ath11k: Fix unexpected return buffer
manager error for QCA6390")'
For WCN6750 and WCN6855, the return buffer manager ID should be
HAL_RX_BUF_RBM_SW1_BM. The incorrect conditional check caused fragmented
packets to be dropped, resulting in the above error log.
Fix this by adding a check for HAL_RX_BUF_RBM_SW1_BM.
Tested-on: WCN6750 hw1.0 AHB WLAN.MSL.2.0.c2-00258-QCAMSLSWPL-1
Tested-on: WCN6855 hw2.1 WLAN.HSP.1.1-04479-QCAHSPSWPL_V1_V2_SILICONZ_IOE-1
Fixes: 71c748b5e0 ("ath11k: Fix unexpected return buffer manager error for QCA6390")
Signed-off-by: Balaji Pothunoori <quic_bpothuno@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Acked-by: Kalle Valo <kvalo@kernel.org>
Link: https://patch.msgid.link/20241030114625.2416942-1-quic_bpothuno@quicinc.com
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Selfgen stats are placed in a buffer using print_array_to_buf_index() function.
Array length parameter passed to the function is too big, resulting in possible
out-of bound memory error.
Decreasing buffer size by one fixes faulty upper bound of passed array.
Discovered in coverity scan, CID 1600742 and CID 1600758
Signed-off-by: Karol Przybylski <karprzy7@gmail.com>
Acked-by: Kalle Valo <kvalo@kernel.org>
Link: https://patch.msgid.link/20241105101132.374372-1-karprzy7@gmail.com
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Add helper functions for multi link peer addition and deletion. And add address
validation to ensure we are not creating link peers (belonging to different
clients) with same MLD address. To aid in this validation for faster lookup,
add a new list of ML peers to struct ath12k_hw::ml_peers and use the same for
parsing for the above address validation use cases.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3
Signed-off-by: Sriram R <quic_srirrama@quicinc.com>
Signed-off-by: Harshitha Prem <quic_hprem@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://patch.msgid.link/20241101151705.165987-6-kvalo@kernel.org
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
In the following patch we need to use ath12k_warn() but don't easily have
access to struct ath12k_base (ab) but do have access to struct ath12k_hw (ah).
So add a new warning helper ath12_hw_warn() which takes the latter but the log
output is still identical but uses the struct device pointer stored to struct
ath12k_hw.
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://patch.msgid.link/20241101151705.165987-5-kvalo@kernel.org
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Add changes to add the link vdevs dynamically whenever a channel is assigned
from mac80211 for a link vdev. During vdev create, update ML address of the
vdev to firmware using the new WMI parameter (WMI_TAG_MLO_VDEV_CREATE_PARAMS).
During vdev start, notify the firmware that this link vdev is newly added and
also indicate all its known partners so that the firmware can take necessary
actions to internally update the partners on the new link being added.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3
Signed-off-by: Sriram R <quic_srirrama@quicinc.com>
Co-developed-by: Rameshkumar Sundaram <quic_ramess@quicinc.com>
Signed-off-by: Rameshkumar Sundaram <quic_ramess@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://patch.msgid.link/20241101151705.165987-3-kvalo@kernel.org
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
As described in [1], compiling the ath12k driver using clang with
KASAN enabled warns about some functions with excessive stack usage,
with the worst case being:
drivers/net/wireless/ath/ath12k/qmi.c:3546:13: warning: stack frame size (2456) exceeds limit (1024) in 'ath12k_qmi_driver_event_work' [-Wframe-larger-than]
Nathan [2] highlighted work done by Arnd [3] to address similar
issues in other portions of the kernel.
ath12k_qmi_driver_event_work() itself is a pretty lightweight
function, but it dispatches to several other functions which do the
real work:
ath12k_qmi_driver_event_work()
ath12k_qmi_event_server_arrive()
ath12k_qmi_host_cap_send()
ath12k_qmi_event_mem_request()
ath12k_qmi_respond_fw_mem_request()
ath12k_qmi_event_load_bdf()
ath12k_qmi_request_target_cap()
ath12k_qmi_load_bdf_qmi()
ath12k_qmi_wlanfw_m3_info_send()
Mark all of those underlying functions as 'noinline_for_stack' to
prevent them from being inlined in ath12k_qmi_driver_event_work(),
thereby eliminating the excessive stack usage.
Link: https://msgid.link/bc214795-1c51-4cb7-922f-67d6ef98bff2@quicinc.com # [1]
Link: https://msgid.link/20241025223321.GA3647469@thelio-3990X # [2]
Link: https://lore.kernel.org/all/?q=f:arnd@kernel.org+Wframe-larger-than # [3]
Acked-by: Kalle Valo <kvalo@kernel.org>
Link: https://patch.msgid.link/20241028-ath12k_qmi_driver_event_work-v1-1-0d532eb593fa@quicinc.com
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
During rmmod, the ath11k host driver sends a QMI MODE OFF command
to firmware.
As part of this command, firmware initiates WLAN de-initialization
and accesses certain UMAC registers during this process.
Currently, on WCN6750 WLAN hardware, the system is in a sleep state when
firmware receives the QMI MODE OFF command.
This results in a firmware/hardware reset while accessing the UMAC hardware
registers during sleep state.
To avoid this, add logic to send WCN6750 hardware specific
WMI_PDEV_SUSPEND_AND_DISABLE_INTR command to firmware prior to sending
the QMI MODE OFF command.
This will cause firmware to cease all activities and put the device in
a powered-on state that prevents access to registers which have been
powered off.
Signed-off-by: Balaji Pothunoori <quic_bpothuno@quicinc.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://patch.msgid.link/20241029083340.3010798-1-quic_bpothuno@quicinc.com
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
ath.git patches for v6.13
This development cycle featured phase 1 of patches to ath12k to
support the new 802.11be MLO feature, along with other ath12k feature
patches. In older drivers, support for some additional devices were
added. And there was the usual set of bug fixes and cleanups across
most drivers.
Per-driver highlights:
ath12k
* Switch to using wiphy_lock() and remove ar->conf_mutex
* Convert struct ath12k_sta::update_wk to use struct wiphy_work
* Add phase 1 of 802.11be MLO support
* Add firmware coredump collection support
* Add debugfs support for a multitude of statistics
* Fix host representation of multiple hal_rx structs
* Fix use-after-free in ath12k_dp_cc_cleanup()
* Skip Rx TID cleanup for self peer
* Fix warning and crash when unloading in a VM
* Convert CE interrupt handling from tasklet to BH workqueue
* Fix A-MSDU indication in monitor mode
ath11k
* Fix double free issue during SRNG deinit
* Enable firmware diagnostic events for WCN6750
* Fix CE offset address calculation for WCN6750 during SSR
* Fix stack frame size warning in ath11k_vif_wow_set_wakeups()
* Document the inputs for ath11k on WCN6855
ath10k
* Fix multiple stack frame size warnings
* Fix invalid VHT parameters in supported_vht_mcs_rate_nss* structs
* Avoid NULL pointer error during SDIO remove
ath5k
* Add support for Arcadyan ARV45XX AR2417 & Gigaset SX76[23] AR241[34]A
Kalle Valo says:
====================
wireless-next patches for v6.13
The first -next "new features" pull request for v6.13. This is a big
one as we have not been able to send one earlier. We have also some
patches affecting other subsystems: in staging we deleted the rtl8192e
driver and in debugfs added a new interface to save struct
file_operations memory; both were acked by GregKH.
Because of the lib80211/libipw move there were quite a lot of
conflicts and to solve those we decided to merge net-next into
wireless-next.
Major changes:
cfg80211/mac80211
* stop exporting wext symbols
* new mac80211 op to indicate that a new interface is to be added
* support radio separation of multi-band devices
Wireless Extensions
* move wext spy implementation to libiw
* remove iw_public_data from struct net_device
brcmfmac
* optional LPO clock support
ipw2x00
* move remaining lib80211 code into libiw
wilc1000
* WILC3000 support
rtw89
* RTL8852BE and RTL8852BE-VT BT-coexistence improvements
* tag 'wireless-next-2024-10-25' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next: (126 commits)
mac80211: Remove NOP call to ieee80211_hw_config
wifi: iwlwifi: work around -Wenum-compare-conditional warning
wifi: mac80211: re-order assigning channel in activate links
wifi: mac80211: convert debugfs files to short fops
debugfs: add small file operations for most files
wifi: mac80211: remove misleading j_0 construction parts
wifi: mac80211_hwsim: use hrtimer_active()
wifi: mac80211: refactor BW limitation check for CSA parsing
wifi: mac80211: filter on monitor interfaces based on configured channel
wifi: mac80211: refactor ieee80211_rx_monitor
wifi: mac80211: add support for the monitor SKIP_TX flag
wifi: cfg80211: add monitor SKIP_TX flag
wifi: mac80211: add flag to opt out of virtual monitor support
wifi: cfg80211: pass net_device to .set_monitor_channel
wifi: mac80211: remove status->ampdu_delimiter_crc
wifi: cfg80211: report per wiphy radio antenna mask
wifi: mac80211: use vif radio mask to limit creating chanctx
wifi: mac80211: use vif radio mask to limit ibss scan frequencies
wifi: cfg80211: add option for vif allowed radios
wifi: iwlwifi: allow IWL_FW_CHECK() with just a string
...
====================
Link: https://patch.msgid.link/20241025170705.5F6B2C4CEC3@smtp.kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
I suggested to put DEBUG_NET_WARN_ON_ONCE() in __sock_create() to
catch possible use-after-free.
But the warning itself was not useful because our interest is in
the callee than the caller.
Let's define DEBUG_NET_WARN_ONCE() and print the name of pf->create()
and the socket identifier.
While at it, we enclose DEBUG_NET_WARN_ON_ONCE() in parentheses too
to avoid a checkpatch error.
Note that %pf or %pF were obsoleted and will be removed later as per
comment in lib/vsprintf.c.
Link: https://lore.kernel.org/netdev/202410231427.633734b3-lkp@intel.com/
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Link: https://patch.msgid.link/20241024201458.49412-1-kuniyu@amazon.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Russell King says:
====================
net: phylink: simplify SFP PHY attachment
These two patches simplify how we attach SFP PHYs.
The first patch notices that at the two sites where we call
sfp_select_interface(), if that fails, we always print the same error.
Move this into its own function.
The second patch adds an additional level of validation, checking that
the returned interface is one that is supported by the MAC/PCS.
The last patch simplifies how SFP PHYs are attached, reducing the
number of times that we do validation in this path.
====================
Link: https://patch.msgid.link/Zxj8_clRmDA_G7uH@shell.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
There are a few issues with how SFP PHYs are attached:
a) The phylink_sfp_connect_phy() and phylink_sfp_config_phy() code
validates the configuration three times:
1. To discover the support/advertising masks that the PHY/PCS/MAC
can support in order to select an interface.
2. To validate the selected interface.
3. When the PHY is brought up after being attached, another validation
is done.
This is needlessly complex.
b) The configuration is set prior to the PHY being attached, which
means we don't have the PHY available in phylink_major_config()
for phylink_pcs_neg_mode() to make decisions upon.
We have already added an extra step to validate the selected interface,
so we can now move the attachment and bringup of the PHY earlier,
inside phylink_sfp_config_phy(). This results in the validation at
step 2 above becoming entirely unnecessary, so remove that too.
Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Tested-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Link: https://patch.msgid.link/E1t3bcb-000c8H-3e@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
phylink_parse_fixedlink() wants to preserve the pause, asym_pause and
autoneg bits in pl->supported. Rather than reading the bits into
separate bools, zeroing pl->supported, and then setting them if they
were previously set, use a mask and linkmode_and() to achieve the same
result.
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/E1t3Fh5-000aQi-Nk@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Tariq Toukan says:
====================
mlx5e update features on config changes
This small patchset by Dragos adds a call to netdev_update_features()
in configuration changes that could impact the features status.
====================
Link: https://patch.msgid.link/20241024164134.299646-1-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When the ring size changes successfully, trigger
netdev_update_features() to enable features in wanted state if
applicable.
An example of such scenario:
$ ip link set dev eth1 up
$ ethtool --set-ring eth1 rx 8192
$ ip link set dev eth1 mtu 9000
$ ethtool --features eth1 rx-gro-hw on --> fails
$ ethtool --set-ring eth1 rx 1024
With this patch, HW GRO will be turned on automatically because
it is set in the device's wanted_features.
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20241024164134.299646-3-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
