Implemented driver interface for NL command for setting
and getting TX and RX chains. The chains can be set to
either chain 0(1) or chain 1(2) or both(3). This is only valid if
antenna mode is set to 2X2.
Below command can be used to set the chainmask:
iw phy <phyname> set antenna <bitmap> | all | <tx bitmap> <rx bitmap>
Example:
iw phy0 set antenna 1 -- to set both Tx and Rx chain 0.
iw phy0 set antenna all -- to set all/both chains for rx/tx
iw phy0 set antenna 1 2 -- to set chain 0 for tx and chain 1 for rx
Below command can be used to get the chainmask after set:
iw phy <phyname> info
check for "Configured Antennas: TX 0x1 RX 0x1"
Below conditions should be met for setting chainmask:
1. Antenna sharing should be disabled.
2. gEnable2x2 in ini file should be 0
3. enableBTChainSeparation in ini file should be 1
4. DBS should be disabled (gDualMacFeatureDisable = 1)
or HW should be DBS 2x2 capable
5. set TX/RX chainmask 3 in ini (gSetTxChainmask1x1/gSetRxChainmask1x1)
6. HW should support TX/RX chainmask as 3
Change-Id: I04f6667398eac405a6be61bc039e048090578abf
CRs-Fixed: 2693370
chan_freq and mhz variable represent the same thing so use chan_freq
instead of mhz.
Change-Id: I4162504562c30ece71f5ebc465b1af3a048b01b2
CRs-Fixed: 2696795
As part of this change I2e7caec268776a79d7166020c64af93fee2f8aec,
the feature flag was not properly defined and there were some compilation
issues. Hence, this change is fixing these issues. Also, add unit test
ini check in fw unit test command.
Change-Id: Ia8dfedec19d4d97c8c100321a0a012ee1a1b6a00
CRs-Fixed: 2695208
Bandwidth level was set correctly to PLD_BUS_WIDTH_VERY_HIGH
but current vote level was not updated for monitor mode.
Due to this during PCI link down the reg writes for SW2WBM
release ring were enqueued in delayed register write work
queue instead of being written directly.
Delay in scheduling the workqueue results in SW2WBM release
ring getting full which results in invoking panic.
Hence update the current vote level along with the bus
bandwidth in HDD open function for monitor mode.
Change-Id: Ib9861a038afa6ca7758ea35afcae2bcfe9e69b60
CRs-Fixed: 2690247
This change is to set txpower to specific interface
from iw.
Ex:
"iw dev <interface name> set txpower fixed <value>"
Change-Id: I5a3195a36cd2d7252eb3ef0312545d338b4000ec
CRs-Fixed: 2677223
Rx mpdu count received as part of peer link stats
event from firmware does not have the proper values.
Fix is to fetch the rx mpdu count from dp layer and
update the rx mpdu count before sending the stats to
HDD.
Change-Id: I945d32c7701f5f5c9bfbbaa6ab4576b94389c84c
CRs-Fixed: 2691648
Enable band preference for 6ghz and give default
max weightage to 6ghz to push it for STA operation.
Change-Id: I2fb1d68bb0d8535b4e9813fdd1551723a1fe4f56
CRs-Fixed: 2688585
Android build system is using toybox “realpath” which doesn’t have
“—relative-to” option as in GNU tools. Use more generic python API.
Change-Id: Ib129bd721056e2fcb563817ac61cc6811568f68d
CRs-Fixed: 2676451
Make include paths relative to kernel build directory. This will fix
the following error due to long absolute path in include directories,
make[2]: execvp: /bin/sh: Argument list too long
make[2]: *** [path/to/object/file] Error 127
Change-Id: Ic930384f38f2166e6bede8d16ac37b359dd48a0d
CRs-Fixed: 2661829
Check WLAN_CFG_USE_DEFAULT to use default profile in case
TARGET_MULTI_WLAN is defined. Default configuration will be used
to build WLAN driver for multiple chipsets.
Change-Id: Ib5b64fd8213f401b043aa7792cc8ac734cd722b8
CRs-Fixed: 2661827
Currently host driver sends the data rate of connected
peer in the units of kbps whereas user space expects the
data rate to be in the units of mbps.
To address above issue, add a change in the host driver
to send the data rate in the units of mbps.
Change-Id: I022659f1cd16ff62dc5b9379405c35c651381e4f
CRs-Fixed: 2688150
The current bss color value is 0 in color
detection config which is not valid. Set
it accordingly in lim to share the correct
bss color config to FW.
Change-Id: I4387c7eb661846d70b62cde41b64b9c174e2f48c
CRs-Fixed: 2689632
When kernel panic, irq will be disabled, when dump Rome
firmware, dma buffer is malloced and freed every time
hif_diag_read_mem is called to dump 20K mem, dma_free_attrs is
called 100 times for 2M mem dump, WARN_ON(irqs_disabled())
will printk stack about 100 times in short time, watchdog bite.
Fix: Change dma mem malloc size from 20KB to 256KB, then only
malloc/free less than 10 times totally when dump Rome firmware.
Change-Id: I803b2cd30a1c6cdc1ce8f8f1073cdb6c17b650af
CRs-Fixed: 2693690
Kernel 5.4 has applied batched GRO_NORMAL packets processing
for all napi_gro_receive() users. this requires NAPI users to
call napi_complete_done() or napi_complete() at the end of every
polling to flush batched GRO_NORMAL packets.
However, by current wlan driver design, napi_gro_recevie()
is not happened in same NAPI polling context and done in another
thread context, it likely packets which have not been flushed from
napi->rx_list will get stall until next RX cycle.
Fix this by adding a manual flushing of the list right after
napi_gro_flush() call to mimic napi_complete() logics.
Change-Id: Ib3e851c0822a85c4712a1b817cc19dfecf6d0e7a
CRs-Fixed: 2673959
Previously pm_dbs would be queried via iwpriv
--> iwpriv wlan0 pm_dbs [dbs] [system_pref]
it is now changed to support via sysfs:
--> echo [dbs] [system_pref] > /sys/kernel/wifi/pm_dbs
Change-Id: I3f6b153a58be6932cafbc1df15ef4a62971ba954
CRs-Fixed: 2692150
Previously pm_cinfo,pm_pcl would be queried via iwpriv
--> iwpriv wlan0 pm_cinfo
it is now changed to support via sysfs:
--> cat /sys/class/net/wifi/pm_cinfo
--> echo 1 > /sys/kernel/wifi/pm_pcl
Change-Id: I437eb6d4e9424f3aed9d1cf218333371c9a9f93b
CRs-Fixed: 2684258
Add the sysfs entry for the pktlog.
Previously get_ampdu,get_amsdu would be queried via iwpriv
--> iwpriv wlan0 1 1 pktlog
it is now changed to support via sysfs:
--> echo 1 1 > /sys/class/net/wifi/pktlog
Change-Id: Id5461c86bb1369a04bc68514e874df6df1230b85
CRs-Fixed: 2684106
Add the sysfs entry for the ipaucstat.
Previously ipaucstat would be queried via iwpriv
--> iwpriv wlan0 ipaucstat 1
it is now changed to support via sysfs:
--> echo "1" > /sys/class/net/wlanxx/ipaucstat
Change-Id: I5a3195a36cd2d7252eb3ef0312545d338b40d3ec
CRs-Fixed: 2676440
Add the sysfs entry for dumpStats and clearStats.
Previously these commands would be queried via iwpriv
--> iwpriv wlan0 dumpStats [stats_id]
--> iwpriv wlan0 clearStats [stats_id]
it is now changed to support via sysfs:
--> echo [stats_id] > /sys/class/net/wlanxx/dump_stats
--> cat /sys/class/net/wlanxx/dump_stats
--> echo [stats_id] > /sys/class/net/wlanxx/clear_stats
Change-Id: If8e4d997bc621df98732461802fdd9cea89bd1ce
CRs-Fixed: 2685995
If the STA-KICKOUT event comes too frequently for
a certain AP, then the AP might have some genuine
issue and should not be tried again and again in
intial connection and roaming case scenarios.
So it is better to avoid this AP and if the AP
still faces a kickout then after 3 attempts which
is ini configurable the driver would blacklist
it to avoid ping pong.
Change-Id: I8221cca9adc777ca275fdd76c2f8ba17cd12c14f
CRs-Fixed: 2683615
Add connection check in set_power_mgmt() before enabling dhcp
protection since set_power_mgmt() can be called in disconnected
state as well for enabling power save.
Change-Id: I4b1796ba45ebf0561f4b58b066bfa63c0ca697ea
CRs-Fixed: 2690325
Add the sysfs entry for dumpStats and clearStats.
Previously these commands would be queried via iwpriv
enable / disable --> iwpriv wlan0 dump_dp_trace [1/3]
show --> iwpriv wlan0 dump_dp_trace 0
clear --> iwpriv wlan0 dump_dp_trace 2
it is now changed to support via sysfs:
enable / disable --> echo [1/0] > /sys/kernel/wifi/dump_dp_trace
show --> echo 2 [count] > /sys/kernel/wifi/dump_dp_trace
cat /sys/kernel/wifi/dump_dp_trace
clear --> echo 1 > /sys/kernel/wifi/clear_dp_trace
Change-Id: I2e7caec268776a79d7166020c64af93fee2f8aec
CRs-Fixed: 2691705
Add a config option to set maximum number of clients allowed.
And set max clients allowed to 64 on SDXPRAIRIE platform.
Change-Id: I4f475abbf5fba8d01f4121e55a87324fdd2d5d70
CRs-Fixed: 2692639
As part of WEXT replacement, replace range_ext with a sysfs file.
This feature can be controlled using configuration flag
CONFIG_WLAN_SET_RANGE_EXT
file path: /sys/class/net/wlanxx/range_ext
where wlanxx is adapter name
example command: echo 1 > /sys/class/net/wlan1/range_ext
Change-Id: I389da429e1a0cd3a2ed1efede6cb958628eb7e6e
CRs-Fixed: 2689879
Issue: Host is sending VDEV_INSTALL_KEY command without validating vdev
state. If there is channel change request, in function lim_change_channel
on WLAN_VDEV_SM_EV_CSA_COMPLETE, WLAN_VDEV_SM_EV_RESTART_REQ i.e vdev
restart request is triggred to fw and if there is any ongoing sta
connection parallely, then it may send vdev install without validating vdev
state. If firmware receives VDEV_INSTALL_KEY, it stops critical channel
request. Firmware sends VDEV_START_RESPONSE/VDEV_RESTART_RESPONSE once it
gets channel grant. If host sends VDEV_INSTALL_KEY to fw before it gets
channel grant then fw stops channel request and does not send
VDEV_RESTART_RESPONSE.
Fix: Valdiate vdev state before sending VDEV_INSTALL_KEY command.
In function wlan_hdd_add_key_sap validate the state of vdev and send
command only if vdev is in between restarting state.
Change-Id: I427ab9bfab499818290dc36b6b4b3a4974b7d3a2
CRs-Fixed: 2605369
hdd_bus_bw_work_handler API is stuck waiting on mutex
sme_global_lock acquired by scheduler thread.
WIFI logger start request API doesn't need to acquire
sme global lock because the request is posted to scheudler
thread and the request is processed in schduler thread context.
Change-Id: I34522abdc54d1a7d6b5acb546179194b4ee869d8
CRs-Fixed: 2690483
Remove wowlan_deauth_enable and wowlan_disassoc_enable INI params as
these are unused
Change-Id: I90d10995365771359f1bcd029b9193437ad67a18
CRs-Fixed: 2678156
