Customer reported their HLOS LTE-WLAN transition algo sometimes go wrong
for wrong low layer stats, Txretry delta is bigger than Txtotal delta.
Txretry unit is msdu while Txtotal unit is mpdu.
Add vdev/peer stats for mpdu number of successfully transmitted after
more than one retransmission attempt, to align mpdu tx_packets, tx_failed.
CRs-Fixed: 3102117
Change-Id: Id4b9d8fd81af4bf0c141dad481bfd2b3f6c1db2a
Allow idle shutdown in case it is triggered due to driver mode change
even if any of interface is up.
Change-Id: I29db749d40729a7e7633836bd23083f7652ac845
CRs-Fixed: 3106971
Currently for 6GHz, nan is allowed on LPI channels as well,
nan should be allowed only on VLP channels.
With this change allow nan only on VLP channels.
CRs-Fixed: 3072585
Change-Id: Ic8cd8a5ca5c096530762cc0f9d199cef30155838
When STA+SAP concurrency, sta roaming pcl will remove all channels of SAP
band except scc channel to avoid mcc.
When SAP stopped, STA roaming pcl isn't updated for wrong vdev id is used,
so can't roam to all channels of old SAP band except scc channel to avoid
mcc.
Change-Id: I35adb55ac155e4c84f858f16796dbec12589ab2f
CRs-Fixed: 3102601
When host resume, check arp offload status wrongly, so arp offload
disable cmd isn't sent to F/W. when next suspend comes, arp offload
enable cmd isn't sent too.
Change-Id: I06ea617df49287ee87e7b5bf8b20971b581dedf1
CRs-Fixed: 3106172
Currently firmware is crashing due to levelconf structure received NULL
from host.
Hence initialize num_thermal_conf param to fill levelconf structure part of
WMI_THERM_THROT_SET_CONF_CMDID.
Change-Id: Ifcfb49253222fff6656327d9be3977a707c18af4
CRs-Fixed: 3064400
This reverts commit I7704a3b9b3e8ad187677705e1b11fbd82be73cfb.
This change cause CXPC failure, as each ll_status will cause
runtime pm resume.
Change-Id: I3547181fb8cfee1e5e40992dfaf5547eaac29a2a
CRs-Fixed: 3110499
Currently when ll_stats or get station request comes and while fw
sends the response of the request, it is possible that driver can go into
runtime suspend state while fw still wants to send few more stats
events. Because of this ll_stats request or station stats request gets
time out in host.
To address above issue, add logic to prevent runtime suspend when
ll_stats or station stats request is received in host driver.
Change-Id: I7704a3b9b3e8ad187677705e1b11fbd82be73cfb
CRs-FixeD: 3096165
Initialize sap bw as MAX BW in
policy_mgr_valid_sap_conc_channel_check() as
wlansap_get_csa_chanwidth_from_phymode() take care
of selecting BW for sap and it has a check that will
take minimum of selected BW and initialized BW.
If this BW is not initialized with Max BW then in cases
where original BW is 0 (20 Mhz) (cases such as channel
switch happens from 2.4Ghz to 5Ghz), the above mentioned
check will result in selected BW being 0 (20 Mhz) for
non-DBS HW.
Change-Id: I29febf09036ffa0163df58ce51b399abe2a43fe2
CRs-Fixed: 3101285
Return success when firmware doesn't support 11k offload, or else
subsequent roaming related wmi commands have no chance to issue.
Change-Id: I69d55cca63f2b7ab4787295a2825159395cf2a11
CRs-Fixed: 2976163
LL stats request is not supported for SAP mode. But driver is not checking
for device mode for LL stats request. Instead it is depending on LL stats
request set flag which won't be set for SAP mode. In case of mode switch
from STA to SAP, it is possible that LL stats request set flag will be set
for STA mode and LL stats are requested after switching to SAP mode. This
will bypass the mode SAP.
To avoid this, reject LL stats request for SAP mode.
Change-Id: I8c680288991126f0b349584f502b22e18af72287
CRs-Fixed: 3081453
In the process of sar req-resp event the timer callback will wait
for the userspace response due to which time scheduler thread is paused
and can't process any other message. Because of this there is delay and
some command may fail.
To address above issue, move the process of sar req-resp
event to work context as we are waiting in the timer handler.
Change-Id: Idadd99ff28dcb6f5ec2d925cb381f223124300e6
CRs-fixed: 3091293
Ignore idle_shutdown if any interface is up. This is done
to handle special scenario in which platform driver triggers
idle shutdown when it receives Deep sleep/Hibernate entry
notification from modem subsystem in wearable devices.
Change-Id: I0f2a3bd4cbf0d7c0d3577dfbd15c7b3001d827d1
CRs-Fixed: 3091085
Deep Sleep or Hibernate are system level mode power
modes in which wlan firmware is shutdown gracefully
and linux kernel is suspended. Currently, suspend
is not allowed if firmware is down, with this change
suspend is allowed if firmware is down due to Deep
Sleep or Hibernate low power state.
A new pld api is introduced to know the low power
state from platform driver, this api is enabled based
on CONFIG_ENABLE_LOW_POWER_MODE flag. Currently, this
feature is enabled only for wearable targets.
Change-Id: Iddaa4da8ee4880a4223420f59cbe9c365c30344a
CRs-Fixed: 3090481
Host send HTT_H2T_MSG_TYPE_WDI_IPA_CFG cmd to fw, need update
high 32bit addr for non-smmu platform which alloc physical
address more than 4G, like addr 0x1 5000 0000.It will cut
off bit32~bit63, fw get memory is 0x5000 0000, it's maybe
use for peripheral memory.when fw write addr 0x5000 0000,
it will meet some error.
32bit platform compile error,for right shift must less
than system width.
Change-Id: If1e84cfdcd15494be3f7327986ff0e91ec7b2b3f
CRs-Fixed: 3097240
Host send HTT_H2T_MSG_TYPE_WDI_IPA_CFG cmd to fw, need update
high 32bit addr for non-smmu platform which alloc physical
address more than 4G, like addr 0x1 5000 0000.It will cut
off bit32~bit63, fw get memory is 0x5000 0000, it's maybe
use for peripheral memory.when fw write addr 0x5000 0000,
it will meet some error.
Change-Id: Ic5c5c00c85bdf08e61c9ff0f539989d2e96bb153
CRs-Fixed: 3077016
Add RSO state changing log to roam history log
so that help debug roam state change issues.
Change-Id: Ifb11013d83de6ee35241dd796df9a7ccbb1464e7
CRs-Fixed: 3094770
Currently if sta_sap_scc_on_dfs_chnl is disabled,
standalone sap can come up on DFS channel in non-dbs
hw but standalone sap channel switch to dfs is rejected
by hdd in case of non-dbs hw and with disabled
sta_sap_scc_on_dfs_chnl.
Fix is not to reject standalone sap csa request for
DFS channel on non-dbs hw if sta_sap_scc_on_dfs_chnl is
disabled, as standalone sap can come up on DFS channel
when sta_sap_scc_on_dfs_chnl is set to 0.
Change-Id: Ibd1b0af25309aa5cc7d9629d2cb78311c20fffcb
CRs-Fixed: 3095298
In qdf_nbuf_update_radiotap rssi is filled based on flag
DP_MON_RSSI_IN_DBM. When this flag is enabled
qdf_nbuf_update_radiotap is expecting rssi value in dbm.
So fill rssi_comb based on flag DP_MON_RSSI_IN_DBM.
Change-Id: I9a6542cc27a5e19072e89ae78ff05f4392c64157
CRs-Fixed: 3083133
Add support to send qos null frames received from firmware on
STA interface to mon interface based on config value. The config
value get update based on vendor command set by user.
The packet filter check in ucfg_pkt_capture_process_mgmt_tx_data
moved to target_if_mgmt_offload_data_event_handler. So that we
will not allocate any buffer if filter is not set.
Change-Id: I426b340c5a65711ada971062af95ae039d18d0bd
CRs-Fixed: 3076241
Currently the global vdev gp_pkt_capture_vdev used in packet capture
does not have ref count.
Add ref count for global vdev used in packet capture component.
Change-Id: I1cc619b31c81a77af0842ce219cfcc96060626a0
CRs-Fixed: 3049225
In packet capture mode currently the cck and ofdm flags are
filled by checking phy mode received from FW. But now FW is
sending below rate codes instead of phy mode. So update the
check in host to check rate code and based on rate code fill
cck and ofdm. The proper filling of cck and ofdm will help
Wireshark to derive correct phy type in the packet capture.
WIFI_HW_RATECODE_PREAM_OFDM,
WIFI_HW_RATECODE_PREAM_CCK,
WIFI_HW_RATECODE_PREAM_HT,
WIFI_HW_RATECODE_PREAM_VHT,
WIFI_HW_RATECODE_PREAM_HE,
Change-Id: Ie9b38dd403a0bf39397ff22e80dd24dfa152fdcd
CRs-Fixed: 3079456
The global aggregation size is only set to firmware once
when vdev is created. And the size may be modified
dynamically for a specified AP during association, according
to the OUI based aggregation size configured by ini
'cfg_tx_iot_aggr'.
If global AMSDU size is 0, considering the case as below:
1. connect to AP-1 which is included in 'cfg_tx_iot_aggr',
the AMSDU size will be set to the specifed value.
2. connect to AP-2 which is NOT in 'cfg_tx_iot_aggr',
it doesn't reset the AMSDU size to 0, and firmware is
still using the value set in step #1, it's wrong.
To fix this issue, set the global size for each vdev start,
as init values for each connection.
Change-Id: I790d580fc5762e6816e840ba5484b3cd758334df
CRs-Fixed: 2918046
Check the configured IOT OUIs in Beacon/Probe Response
IEs, but not in Associate Response IEs, because some APs
won't add its specific OUI in Associate Response frame.
Change-Id: I3aedeef0a5dbecd4bed41eb47a9374e654509746
CRs-Fixed: 2955964
Add ini 'cfg_tx_iot_aggr', it gives an option to
configure Tx aggregation size in no. of MPDUs/MSDUs
for specified OUI. This can be useful for IOT issues.
Format of the configuration:
cfg_tx_iot_aggr=<OUI-1>,<MSDU-1>,<MPDU-1>,<OUI-2>,<MSDU-2>,<MPDU-2>...
MSDU/MPDU means the max tx aggregation size in no. of MSDUs/MPDUs,
0 means not specified.
For example:
cfg_tx_iot_aggr=112233,2,0,445566,3,32,778899,0,64
If vendor OUI-1("\x11\x22\x33") is found in assoc resp,
set tx amsdu size to 2;
If vendor OUI-2("\x44\x55\x66") is found in assoc resp,
set tx amsdu size to 3, set tx ampdu size to 32;
If vendor OUI-3("\x77\x88\x99") is found in assoc resp,
set tx ampdu size to 64.
Change-Id: Idcf370a4bf93ca299ce1126eaba4394be1ab5b9d
CRs-Fixed: 2849203
A race between cfg80211 disconnect and wiphy system suspend can lead
to either DPM WD or serialization VDEV disconnect active command
timeout since scheduler thread gets suspended as part of wiphy
suspend and both cfg80211_disconnect and wiphy suspend/resume acquire
RTNL lock. To address this race condition avoid disconnect when wiphy
suspend is already completed since scheduler thread gets suspended
as part of wiphy suspend and it can't process vdev disconnect.
Change-Id: Ia7e42cffb3f6b08b33c878b68122dbdc00bad251
CRs-Fixed: 3042442
Currently there is no provision to get the calibration failure
information from the driver. Because of this it is getting difficult
to debug the calibration failure issues.
To simplify the calibration failure issue debugging, add support to
parse calibration failures events in the driver.
Change-Id: I6d831804cca259862fea3e8bb4af33d556138d43
CRs-Fixed: 3078926
It is possible tx_time_per_power_level is not freed
in last event, and it is reused in
__wma_unified_radio_tx_power_level_stats_event_handler,
the buffer size may be different for each event by
manually test.
Fix is to check if memory is freed before malloc, if
not null, free it before malloc.
Change-Id: I51064734cf97b9ff0ecbbaf27f38d5a223c91d3b
CRs-Fixed: 3057436
In hdd_config_tx_rx_nss and hdd_config_vdev_chains, it
missed to check if tx_attr or rx_attr is null, which will
cause invalid memory access.
Change-Id: Ic3427d714e240507cf4253588f706d06d355ba93
CRs-Fixed: 3086252
Add INI - mgmt_frame_hw_tx_retry_count to configure MGMT
frame HW tx retry count for certain frame types.
The INI String format:
mgmt_frame_hw_tx_retry_count="<frame type>,<retry count>,..."
The supported frame types are defined by enum mlme_cfg_frame_type.
Retry count max value is 127.
For example:
mgmt_frame_hw_tx_retry_count="0,64,2,32"
The above input string means:
For p2p go negotiation request fame, hw retry count 64
For p2p provision discovery request, hw retry count 32
Change-Id: I32f6c7d83ede9b28484c7a0b29824bde32e06422
CRs-Fixed: 3082532
Currently, SAP state machine is moved to INIT state when SAP
start is aborted due to timeout or SSR. SAP is not cleaned up
in such cases as eWNI_SME_STOP_BSS_REQ (which further issues
vdev down and cleanup the peers properly) is not sent to
below layers(firmware).
Replace qdf_wait_for_event_completion to
wait_for_completion_timeout in start_bss which will wait
till the normal operation execute irrespective of firmware
state/SSR.
Change-Id: Ifff2ca9658769cb1145f1c5cb278cd7551cb8c00
CRs-Fixed: 3058550
