In some scenario, it is required to receive every ARP/NS packet even
it is not to solicit address itself. So add this vendor attribute
configuration to enable/disable ARP/NS offload on fly.
On some platform, if target suspend to WoW mode, APF filter will drop
such ARP/NS packet in the front, so prevent runtime suspend happen
if disable ARP/NS offload. System suspend still allow in consideration
of potential power impact.
It only support on STA/P2P-Client mode device, and this configure is
optional, while default behavior is ARP/NS offload enable. It will
restore to default behavior if interface being closed.
Change-Id: Icd49e230024bc1ce51519cd9fafee9bc9f79f382
CRs-Fixed: 3058494
Add ini support for tx_retry_multiplier and send
tx_retry_multiplier to fw as PDEV_STATS_TX_XRETRY_EXT
pdev param, so that fw can multiply the counter by the
percentage provided by user.
Change-Id: Ie2f06dc3aaa4a161a451b68abd085cc1537eb468
CRs-Fixed: 3066799
Currently in the case of roaming between latency-critical connection
and non-latency critical connection we are scheduling work to
update features and after feature update we are enabling network queues
which is introducing a delay in enabling queues due to work schedule.
To Fix the issue enable network queue in the current context and schedule
work to update features.
Change-Id: Idba5b81a58aaf2bd47d80da1d17ad0acf9939071
CRs-Fixed: 3070688
Userspace sends a connect request to host and connect
request contains either BSSID or BSSID hint.
Fix is to validate bssid_hint for non zero or broadcast
instead of checking for bssid.
Change-Id: I468f412c0b07c379b9719e175d5709c37b862a6d
CRs-Fixed: 3064383
Currently "is_forcescc_restart_required" flag is updated
for first go when any new go comes up. Force scc is
done as part of set key and it is expected that when CLI
joins second GO, first GO moves to new GO's channel but in
case of AUTO GO when no client is connected to first GO
and second GO comes up then "is_forcescc_restart_required"
is set for existing GO but when any client joins first GO
then as part of set key CSA is triggered for same target and
current frequency because "is_forcescc_restart_required" is true
for first AUTO GO. As target and current frequencies are same,
CSA fails and "is_forcescc_restart_required" sets to false. When
CLI joins second GO then CSA for first GO doesn't happen as
"is_forcescc_restart_required" is already changed to false as
part of first CSA attempt.
Fix is to trigger CSA only if current and target session ids are
different.
Change-Id: Ib875cdd93e08f4edc912589b867b733a1d57bdf3
CRs-Fixed: 3067847
Enable config flag to use multiple tx and tx completion
ring pairs for qca6490 when IPA is disabled.
Change-Id: Iafa6df504d7be55cbc5a233d6975963f0e118df5
CRs-Fixed: 3059736
Affine tx rings interrupts to gold cores at very high tput
level and to silver cores at tput levels lower than very
high.
Change-Id: I90006645acb82b71c63d2255722e2c67bb7a2f46
CRs-Fixed: 3059716
Currently we use sta_info TX/RX counters to update stats request from
upper layers. This requires these counters to be updated in per packet
path during TX/RX. These counters are already available from FW stats.
Hence use FW stats to update TX/RX packet and byte count.
Obtain last_tx_rx timestamp from DP layer.
CRs-Fixed: 3059244
Change-Id: Ic36c025992bf1dca6702e4b39efd797f5735f78a
Remove validate_context call from hdd_select_queue and replace it with
light-weight cds_is_driver_transitioning API.
Add SSR protection using hdd_dp_ssr_protect APIs to
hdd_wmm_select_queue.
CRs-Fixed: 3059244
Change-Id: Iff991ea0cc8afefc8711b233e864a47ba5fb4fec
Remove sta_info usage from _hard_xmit SAP API. The APIs are
taking/releasing a lock to access sta_info and adds lots of cpu cycles.
Replace it by accessing dp_peer from DP layer and checking for its
state.
Some of the existing stats being updated in sta_info during TX packet
path will be updated on demand using CDP APIs and stats obtained from FW.
CRs-Fixed: 3059244
Change-Id: Id33d7271244687e616ce08b80e1c540ec9b42084
Use the QDF abstraction API to get the cpu id instead
of using HLOS API directly.
Change-Id: I555ac130e153b8a5c458670a276d277c19a97104
CRs-Fixed: 3070758
In cm_roam_trigger_info_event, fw_timestamp is not copied to
wlan_log_record.
Add changes to copy fw_timestamp to wlan_log_record in
cm_roam_trigger_info_event.
Change-Id: I8ec7796b656d716e6ab3b7613be0c1332d5c9fa0
CRs-Fixed: 3068843
Add self bss roam check in roam invoke cmd handler,
If ini enable_self_bss_roam is 0, don't permit to roam
to current connected bss, just keep current connection.
Change-Id: I7fe724e26d15bcc008957829e4e3384c60b58773
CRs-Fixed: 3069587
This change sends the new wow event in the bitmap to FW only
when the roam stats suspend mode attribute is enabled from the
user space.
Also, the new wow reason is added to wake up the host while
sending the roam event stats.
Change-Id: Iaefb229929e8791870f33bf787487f7c20daf218
CRs-Fixed: 3050830
In cm_roam_scan_info_event, qdf_mem_copy is performed with
different sizes.
Fix is to assign frequency values instead of mem copy.
Change-Id: Ie7dc4eaf4d7c0b65100edd77d7c2306e53e2ea9b
CRs-Fixed: 3065698
Presently, whenever there is any kind of reference leak in the system
wlan driver doesn't recover and goes into a bad state.
Add support to forcefully clean the references in the builds where
the leak detection is disabled, so the system can gracefully recover
and wifi can be functional.
Change-Id: Ib9f59e71fe7f09ab11322aa6da9834c2c58fc32f
CRs-Fixed: 3012995
This change gathers the roam event stats from the FW.
New vendor event is used to fill in the vendor attributes
whenever the roam stats are received from Firmware.
Change-Id: I9a2ddef62d26b0b68897891788733df452ac1ceb
CRs-Fixed: 3036518
Add new vendor command to enable the roam events stats from the
driver/FW in both resume and suspend mode.
New roam param is introduced to indicate the FW to enable/disable
the roam stats using the new roam set command.
Change-Id: I6d04f0946e6735613a4cf3d4361bf31052314fa0
CRs-Fixed: 3036518
When 32 STAs connected to APUT, crash will happen because of lack
of resource. The max vdev number was decreased by 1 from the value
of 4 because of nan is not supported. Then FW will received the
value of 3 and finally the tid number is not enough. Change the
max value of vdev number to 5 to fix this issue.
Change-Id: Ia212a0b05922530526439206fd70a44f3ce03838
CRs-Fixed: 3064133
Update the concurrency policy for STA mode.
For an upcoming STA connection:
1. Disallow the 2nd STA connection if multiple STA connections are
not allowed.
2. Allow the secondary MLO link when primary link is connected.
3. Disallow ML STA when ML STA/SAP is present.
4. Disallow the 3rd STA.
5. Allow for the other cases.
Change-Id: I70c41177b7a3a9aa9cbac0aaced08c6eafca2b6d
CRs-Fixed: 3012577
Move the 6GHz support check so that the logic is invoked
regardless of the input band bitmap.
Change-Id: I45db2d2be8966ac073a8ede721d52bd689a98a88
CRs-fixed: 3069557
When the wait_for_ack flag is not set in deauth/disassoc
case, the Host does not send the management frame info
log event to userspace.
Make changes to send the deauth/disassoc events when
wait_for_ack flag is not set.
Change-Id: Id9f4f2e4af124644d919bf36b12376d292acfe3e
CRs-Fixed: 3065707
