If driver SSR has happened after system suspend then RTPM lock
acquired during system suspend is not getting released which
leads to mismatch in RTPM reference count.
Use RTPM prevent suspend sync API to ensure system is in
resume state before acquiring RTPM prevent suspend lock
and in the driver SSR sequence, call to RTPM lock deinit
cleans up any acquired lock.
Change-Id: Icf1d564aa74d9c081a71fe173895e4d29b95c90b
CRs-Fixed: 3254539
Parse and store new ACS vendor attribute
QCA_WLAN_VENDOR_ATTR_ACS_LAST_SCAN_AGEOUT_TIME in SAP ACS
config.
Change-Id: If50224780192ed7b587505e973a8dabeab3459b2
CRs-Fixed: 3239372
Currently sap is not starting if all channels configured as coex
unsafe and if restriction mask is not set.
This change is to select unsafe channel if restriction mask is
not set and start the sap.
Change-Id: I4ccf887655066da6d6b50715f631dde6edfda598
CRs-Fixed: 3260305
Currently, host driver validates pe_session in lim to handle
set key response and send the set context response to sme if there
is no valid session. This is fine for STA, P2P and SAP modes as
there would be a valid session. But host driver does not need to
send set context response for NAN iface in NAN pairing protocol
as it doesn't have a valid pe_session and peer does not exist for
NAN iface. So, exempt the pe_session validation for NAN interface.
Change-Id: Iff3ca8b6a281fc0ab94ed8be683ae507e20a214e
CRs-Fixed: 3257689
With the new requirement, host can receive oem data event
immediately after it sends WMI INIT command to fw.
To process this event, initialize async oem data event before
it sends the WMI INIT command to fw.
Change-Id: I58cc4fac6c6c0f65305b80d04ab22ba867f2b679
CRs-Fixed: 3262660
In wlan_cm_mlo_update_disconnecting_vdev_id(), add an extra
check of wlan_cm_is_vdev_disconnecting with
wlan_cm_get_active_req_type, to check the vdev state as well,
along with the active_cm_id flag.
Change-Id: I80d59d419555864d6c9712e0d93efce0a6c203e4
CRs-Fixed: 3262543
To avoid unmeaningful roaming, when low RSSI trigger,
only roam when rx linkspeed is also bad.
Steps:
1. F/W indicates feature supported by:
wmi_service_linkspeed_roam_trigger_support
2. App sets vdev rx link speed threshold by vendor cmd.
3. Bus_bw_work gets rx link speed from data path periodically.
4. If found rx link speed change from good to poor, or poor to good, send
to F/W.
5. F/W low rssi roaming is triggered only when both RSSI and link speed are
poor.
Change-Id: I2c2a22c1f24a730783ee09455f4d70b099d9bff1
CRs-Fixed: 3255036
Enhance NUD failure handling by also adding
check for packet received from AP in the case
where AP is gateway
Change-Id: I45f4c11f34eddf536756edc4f81009227ae64c35
CRs-Fixed: 3239391
Currently host driver does not cleanup previous ap beacon ie
info if it does not support 11AX, which is leading to memleak.
To address above issue, free previous ap beacon ie info
irrespective of 11AX support.
Change-Id: Ia96e4790a7b6fc8805272367ef7aee985ef931b1
CRs-Fixed: 3258173
The vdev id might be invalid when sending WMI_BPF_SET_VDEV_ENABLE_CMDID
to firmware, which will cause assert in firmware.
To fix this issue, validate vdev id before sending the command.
Change-Id: Ie1b60c5eeb6704de963f23d7fef0649c208081a4
CRs-Fixed: 3260462
Sometimes fw sends roam event with RSO status 7, it does not need
host driver trigger in function target_if_cm_roam_send_stop. When
this scenario happens, driver tries to stop the timer which does
not started and dumps wrong info.
Change-Id: I82f723bb8e8c5b6108caff48a9189eeaf2686a2c
CRs-Fixed: 3261664
Use cdp_wdi_event_unsub() to unsubscribe
PKT_CAPTURE_RX_NO_PEER_SUBSCRIBER in
pkt_capture_wdi_event_unsubscribe().
Change-Id: I43575df1402611d468b8523bf948d2dc86f31dd7
CRs-Fixed: 3260272
Add code to support modularizing CNSS2 CNSS_GENL and CNSS_UTILS
for auto qca6390.
Change-Id: I3d9c5c7a41863d75c91a17b110e7cce29792bdba
CRs-Fixed: 3249464
Currently there is compilation issue in case of IPA
offload enable case due to code changes done as
part of DP componentization.
Fix compilation issue by adding calls to DP component APIs
Change-Id: Id545f67bff75682ceccba5725f956a89c6b9b9f8
CRs-Fixed: 3259293
Authentication frames(with type PASN, value 7) get
exchanged between NAN peers in NAN pairing protocol.
These exchanges happen on NAN interface.
But currently, NAN interface doesn't support auth mgmt
frame processing.
Add entries in wiphy->mgmt_stypes for NAN iface also
to support this.
Change-Id: I2e4f53af737c80e8b7eae5883e95f32ff98d4f00
CRs-Fixed: 3257646
Currently, if AP advertises it supports BW 40 & 80 MHz in
5 GHz and 6 GHz in HE Phy Capability IE, driver does not set
the varibles such as bw_20_rx_max_nss_for_mcs_0_to_7 etc
in function lim_ieee80211_unpack_ehtcap. When STA is only
set to support channel width 20 MHz, then it invokes function
lim_populate_eht_20only_mcs_set and causes the intersection
result EHT RxMCSmap 0 TxMCSmap 0 when set to fw.
Change-Id: I9e52fad7f5e429b05402e4413f7599c01d2e50d0
CRs-Fixed: 3261970
Incorrect ini values are read for the roam_scan_inactivity_time
and roam_scan_period_after_inactivity. So wrong values are
populated to firmware. This is seen with wlan-connection-roaming.ini
ini file is present on the device.
Read the value for roam_scan_inactivity_time from
CFG_ROAM_SCAN_INACTIVE_TIMER and read roam_scan_period_after_inactivity
from CFG_ROAM_SCAN_SECOND_TIMER.
Change-Id: I98a9d99f09200a021a123226cfb714b1c49c51d5
CRs-Fixed: 3259600
In DBS-disabled scenario start DFS AP will be blocked because
function policy_mgr_update_dfs_master_dynamic_enabled return
false when sta_sap_scc_on_dfs_chnl equals 1.
Firmware will limit scan channel in 2 GHz when DFS AP plus STA.
Hence it is reasonable to allow this policy in Non-DBS mode
Change-Id: I31cada8caf74694a228cad215893cf442f1a5564
CRs-Fixed: 3253870
Currently, there is no print to check local constrtaint which
may be reason of tx_power updation based on power constarint is
absolute or not. Add print to check local constarint, so that
tx power updation can be monitored.
Change-Id: I463828aab69e189c6cd6c4e61e01af1ba140ab29
CRs-Fixed: 3260542
In kernel 5.4, the arch name for sa515m is CONFIG_ARCH_SA515M, enable
mobile router for this arch in the default defconfig file. The
qca6490 uses the default defconfig, to support qca6490 in sa515m, this
change is required.
Change-Id: I1a26762f8453ea8028c626938599c1f7ddfd3ed4
CRs-Fixed: 3255024
Userspace may send RSNXE also in connect request irrespective
of the connecting AP capabilities. This allows the driver to
choose best candidate based on score. But the chosen candidate
may not support the RSNXE features and may not advertise RSNXE
in beacon/probe response. Station is not supposed to include
the RSNX IE in assoc request in such cases as legacy APs
may misbehave due to the new IE. It's observed that few
legacy APs which don't support the RSNXE reject the
connection at EAPOL stage.
So, strip the IE when below conditions are met to avoid
sending the RSNXE to legacy APs,
1. If AP doesn't support/advertise the RSNXE
2. If the connection is legacy than WPA3 mode
3. If no other bits are set than SAE_H2E, SAE_PK, SECURE_LTF,
SECURE_RTT, PROT_RANGE_NEGOTIOATION in RSNXE capabilities
field.
This is a continuation to the partial
fix Ia7dee29792718a6514da44b13b36b46586ef25ab (which is reverted
as it's not a complete fix).
Change-Id: Ib857e9358a5414841002966ca14bdbb30d6587c6
CRs-Fixed: 3255353
At present, 6 GHz allow check is not applicable to P2P GO interface
type in API hdd_get_ap_6 GHz_capable, thus start GO on 6 GHz failed.
Fix by call policy_mgr_is_6ghz_conc_mode_supported API to check
interface type support 6 GHz or not.
Change-Id: I96bda834e65d0e1fe7301ef695234df9088f50a6
CRs-Fixed: 3253893
As is_disconn_stats_completed is not reset in start of get stats
process if get stats fails the failure handling is blocked
as is_disconn_stats_completed can be 1 from previous disconnect
and incrementing it again make it 2. Thus the peer disconnect
is not processed leading to active command timeout.
So reset is_disconn_stats_completed before error handlings.
Change-Id: I4c7a23a698f063a002ee0a193180dcacd8dea8de
CRs-Fixed: 3258387
Currently, there is a race condition where packet gets
enqueued when WOW handshake is being done in parallel during
bus suspend resulting in assert in FW as Tx is aborted when
WOW command is received in FW.
To fix this, suspend DP tx path before suspend sequence starts
which will drop tx packets.
Change-Id: I38a6a6bdf9858f627205f75f75f351f7debbb7f7
CRs-Fixed: 3227752
If AP EHT cap doesn't support 320M, DUT STA session EHT cap shouldn't
support it, EHT 320M MCS/NSS set shouldn't be included in assoc req.
Change-Id: I874e1167ea6b6903b4341a90acc8528965522d9c
CRs-Fixed: 3257502
currently when the frame is received driver checks if its a ml
association and if the frame is received on non-assoc link it
gets the assoc link adapter and sends the frame to the userspace.
In the current issue when the sta is not connected and a frame is
received on the secondary link for which netdevice is not registered
it is sending on the same unregistered netdevice resulting
in system instability.
If the frame is received on the non registered adapter, get the
netdev adapter and send the frame to userspace.
Change-Id: I8b2b6315df211622cdcd3853b5ca0ece7edb84fc
CRs-Fixed: 3258365
Currently while attaching vdev to DP interface
we are searching DP interface based on the Adapter
MAC address, but if STA MLD and Link address is
same, we were getting the Link Interface instead of
MLD interface, which was causing the EAPOL packet to drop.
To Fix the issue, instead of searching for DP interface
based on MAC address, use netdev to search DP interface.
Change-Id: I50c7974c053717b496fab0eeb89774af382b4313
CRs-Fixed: 3258684
