The log subtype field is not filled for the connecting fail
subtype logs. This causes WLAN_CONNECTING event to be sent
in case of connection failure also.
Fill the log subtype as WLAN_CONNECTING_FAIL for connection
failures.
Change-Id: I36d5c756274d77f562f1acfc1d75a79fcda210d9
CRs-Fixed: 3060642
Set PCL command was sent from CSR->LIM->WMA in
legacy implementation. With current flow set
pcl follows policy_mgr->WMA path.
So remove unused LIM set roam PCL code.
Change-Id: I300f901fbd62ebab14527f05d98528c108823cd0
CRs-Fixed: 3062176
In OSIF every north bound entry is protected via dsc. DSC op start/stop
takes a spinlock for every call into the driver, the framework
also searches through a osif vdev list for every into driver call.
In per-packet path this is operation of taking/releasing the spinlocks
and traversal of the linked lists is costly. To avoid this and
protect the DP operations against SSR/load unload define a new
atomic variable. Before SSR/driver unload wait for the DP op
also to complete.
Change-Id: I4e9061b9e6b454b20789f9d7d3b7e3955a9bfaa5
CRs-Fixed: 3060552
Host sends the EXT CAP IE based on the capability
coming from the supplicant in join request. But, after
roaming the cached join request will be freed. Therefore,
after roaming few capabilities sent by the userspace are
not saved and therefore lost in reassoc request.
Save the EXT CAP IE from the join request in the VDEV
and use the cached capability for sending Ext cap IE.
Change-Id: I71947388b4c6e9e56a5832557416d9b462de70e2
CRs-Fixed: 3038496
During components psoc open, if the Nth component psoc open fails,
then previous (N-1)th psoc's must be closed in reverse order as
part of error handling.
Change-Id: Id12a6c667b74b0ac17e45c2aa2451add18fbdae1
CRs-Fixed: 3061676
If ini gDisableDFSChSwitch=1 is set to setup dfs test mode,
should disable dfs nol function regardless sap operation
channel dfs or not, because sap's bonded channels may have
dfs channels if bandwidth is 160M.
Fix it by invoke wlan_mlme_check_chan_param_has_dfs().
Change-Id: If42c85a3dd1b83e495d7e6a556a6769c3e9e28b6
CRs-Fixed: 3038252
There is frequent output when get unexpected vendor specific action
frame. So, add this change and reduce the log level.
Change-Id: I9d832546a4436bdf7dd25cb00a7de5cce1295b31
CRs-Fixed: 3060750
Currently, target_if forwards the blacklist event to upper layers
even if there is no AP info present. This event can be dropped in
target_if to avoid accessing unallocated memory.
Change-Id: Icfaabb0a8e3304882979ffde7f5cd27fd4c8a067
CRs-Fixed: 3057424
In WPA3 SAE TX case, auth_algo, auth_type and auth_seq_num
are not filled to the connectivity logging.
In SAE RX case, the auth frames are not getting logged.
Fix connectivity logging to handle SAE RX and TX case
by adding the params mentiioned above.
Change-Id: I388dd154f9d80aaf51f45bb571139072c77bf782
CRs-Fixed: 3057861
Add support to send beacon miss disconnection logging event.
Fix the diag attribute end marker in the event table.
Change-Id: I11fc8859af383b69bcba42654fb3a2bc597fef83
CRs-Fixed: 3056266
Issue is, if the req->ssid_len is greater than 32 then
record->conn_info.ssid_len is limited to 32, but while copying the ssid to
the destination array, req->ssid_len is used, which may cause out of
bounds access.
Fix is to use destination array ssid length record->conn_info.ssid_len
Change-Id: I2a661b18fc1d27eec8b2f85877681fc1b782ad39
CRs-Fixed: 3059161
For resetting NOA, "P2P_SET_NOA 0 0 0" command will be
used. Since duration and interval are same, currently the
set will be rejected.
Handle the count 0 case and bypass the existing validation
for periodic NOA.
Change-Id: I5294d7bcdd70d9280e50167c961d2e9c525fca4c
CRs-Fixed: 3054252
Currently FASTREASSOC issues roam command to trigger fw process
LFR3 roaming. When roam event comes back, it deletes the old pe
session and creates a new one hence the bss desc info is lost.
Funtion lim_is_medium_time_valid could not get desc info and
cause addTS fail.
To resolve this issue, store the wmm info in the pe session if
it presents.
Change-Id: I6c7c79c96fed7384a686a1fa0f49069440096e90
CRs-Fixed: 3058801
Currently WMI_DBGLOG_TIME_STAMP_SYNC_CMDID command sent
every 1 sec causes RTPM collapse and inturn has power
impact. So to avoid this enable time of the day synchronization
only after connection/roaming.
Send WMI_DBGLOG_TIME_STAMP_SYNC_CMDID as part of RSO
start command.
Change-Id: Iee3cefe1318a6e5507a43cefb9666cc2a439fe06
CRs-Fixed: 3060762
This reverts commit I784e4b1735f4f217713f65c30092bac2ad6cb698.
Some AP reports HT20 in beacon frame but HT40 in assoc response
frame. FW roaming will decide BW HT40 based on assoc response.
In previous change host will decide BW based on assoc response
as FW does. But the issue is after roaming, the host will check
beacon for AP BW changing and then do vdev restart. But the vdev
restart may be rejected by FW if the FW roaming is still in
progress. To avoid the complication, revert the change.
The original issue will be fixed by validation the channel
segment by Ia41c51feca855cab54e908438c51a8d12075f864 to avoid
the kernel check channel failure and disconnect.
Change-Id: Ib170acb3588cf07bf666d446f5dde73de2d4705b
CRs-Fixed: 3061233
pdev is declared in nan_enable_req after a variable-size array.
Place the pdev declaration before the array to avoid corruption
while copying contents to the buffer.
Change-Id: I92766c40909bcecaed5f1eb6e2f015447f3c8221
CRs-Fixed: 3060146
Currently there are multiple driver calls in per packet path
w.r.t recovery, bad state, load/unload which is leading tput impact.
Move all the cds driver checks into one function to avoid multiple
calls.
Change-Id: I70f4fcc6bfbd1df32ae17ce6a0485fa3f55ca561
CRs-Fixed: 3057573
As part of target if convergence, roam stats event handling
is moved to target if from wma. BTM request logs are not
sent in the new path.
Add support for BTM request logs in connection manager.
Change-Id: Icc66c5dc608260f2a0f75c5d1a85c799195f9f3e
CRs-Fixed: 3054987
The value of btcoex is 0 when bluetooth device is
connected.
Add bt_profile_con param in enum WLAN_COEX_EVENT, add
a check for bluetooth device connection and update the
value of btcoex on bluetooth connection.
Change-Id: I72d4ef4fd0eb15d37132c11b7aa0a0c7f1452aa2
CRs-Fixed: 3050655
Currently the DA for EAPOl packets is being compared
with the adapter mac address. For MLO connection, the
EAPOL frames are exchanges using MLD mac address and
not the link address. hence this leads to drop of all
the EAPOL packets in receive path.
Add MLD address check for received EAPOL frames.
Change-Id: If158da1338169958c5a74cfb7ee125d7365a8202
CRs-Fixed: 3039333
EAP length attribute is of type NLA_U8, but the table entry
has this attribute type as NLA_U16. This causes parsing
failure of this attribute and the length print is not
included in the logs.
Fix the NLA type for the QCA_WLAN_VENDOR_ATTR_DIAG_EAP_LEN
attribute in the connectivity logging table.
Change-Id: I3570931799811787169c9403b1f26488379524bc
CRs-Fixed: 3054835
In current implementation, validation code of
SCC, MCC, DBS or SBS is repetitive for all
combinations.
Remove repetitive validation code and make
common apis for scc, mcc and dbs validation.
Change-Id: Ic88f01dd9bbcc3d5dcc3453b2c6fef030cfe0305
CRs-Fixed: 3052959
Framework disables p2p before enabling NAN as P2P+NAN concurrency
is not supported. NAN operations happens through wifihal and p2p
operations happen through wpa_supplicant which are two different
userspace components. Wifihal call to enable NAN may come to host
driver before disabling p2p through wpa_supplicant. There are high
chances for p2p scan to be in running state while NAN enable is
received. Firmware NAN state machine goes to inconsistent state
and disables NAN as p2p scan is running in such cases.
So, stop the ongoing p2p scan before enabling NAN as P2P+NAN
concurrency is not supported currently.
Also, forward the pdev to NAN component inorder to iterate through
all P2P vdevs.
Change-Id: Ibe30a5ebe90514aee4f6721cdc5476570524cad8
CRs-Fixed: 3054576
Update HDD API definitions under MLO feature as per 5.10 kernel version.
Also use kernel feature falgs along with MLO feature flag. This will make
sure that if MLO is not enabled in the kernel, then driver won't register
ML APIs from driver.
Change-Id: Ie77e82487d99c8053ba0ca4aabea2e435cc8cb2c
CRs-Fixed: 3056715
When STA connects with 2.4GHz band AP, shortGI ie for 80MHz
& 160MHz becomes 1 instead of 0 as 2.4GHz supports max 40MHz
channel width.
As part of fix, if STA connects in 2.4GHz band AP then
a. Disable shortgi80 and shortgi160and80plus80 for vht caps,
b. Disable ul_2x996_tone_ru_supp, num_sounding_gt_80,
bfee_sts_gt_80, tb_ppdu_tx_stbc_gt_80mhz, rx_stbc_gt_80mhz,
he_ppdu_20_in_160_80p80Mhz, he_ppdu_80_in_160_80p80Mhz for he
caps.
Change-Id: I27fdd6e949e738b33ced940f08664ed7b61c5af8
CRs-Fixed: 3054133
Currently based on the INI option we are scheduling
work to dynamically disable/enable checksum offload and TSO.
But in the case of roaming from latency critical connection
to latency critical connection or non-latency critical connection
to non-latency critical connection, we do not need to schedule work.
So adding a condition to check :
For non-latency critical connection: If checksum offload and TSO are
not disabled then schedule work.
For latency critical: If checksum offload and TSO are
not enabled then schedule work.
Change-Id: I75a51707774c3428971dfe5cc0b0d3cdc2a17ac8
CRs-Fixed: 3054183
FW may experience HWDTIM mode enter and exit, if HOST can't queue
frames fast enough, this will bring 10ms latency for TX under WoW.
This change adds an indication bit to notify FW whether a TX happens
after WoW wakeup, then FW can discard HWDITM and reduce TX latency.
Change-Id: I74f7311fc0639186ee0bca711160bb8c0d5d9fb3
CRs-Fixed: 3055501
