Previously all instances of the identifier "pAdapter" were renamed
within the driver. However the hdd_enable_default_pkt_filters()
prototype had a misspelling and hence it was not renamed, so rename it
now.
Change-Id: I6b2045496e5e6775dc12061254d06bcadf60be88
CRs-Fixed: 2362720
Previously change Idaaacac4fdaf1b964432c33c10b1c86f8a591c2d
("qcacld-3.0: Remove legacy P2P codes in HDD") removed function
wlan_hdd_roc_request_dequeue() but did not remove the prototype, so
remove it now.
Change-Id: Ic2eb217b3974b274e890c1c556eece9e688463b4
CRs-Fixed: 2362719
A prior version of the driver had LIBRA_LINUX_PC feature code, but it
has all been removed except for one piece, so remove that last bit
which conditionally defines HDD_TX_TIMEOUT.
Change-Id: Ie714b134b789a4649f8da7bbf5e98f36d27ed4ff
CRs-Fixed: 2362718
Change I15535827a03953231670d4138235c4876b16e045 ("qcacld-3.0: Add GRO
support for STA mode") removed the implementation of hdd_lro_init(),
but it failed to remove the inline stub implementation for use when
FEATURE_LRO was not enabled, so remove it now.
Change-Id: Ie835fdd08322c8968333e264d1228586261b5d59
CRs-Fixed: 2362716
If interface is not closed, Bus bandwidth compute timer is not
stopped during rmmod.
Stop bus bandwidth compute timer during driver remove
Change-Id: If9206f9dcfb6120394848715a6f31a810002d0c9
CRs-Fixed: 2355345
Scan request queueing has moved to the common project and
struct hdd_scan_req is no longer used, so remove it.
Change-Id: I88f2c6c313c7900406d6bddda87e612099d75e58
CRs-Fixed: 2361084
This reverts the change id- I3505c128276fa8e5e05ea5d9110eb75275a746e9.
When wifi is toggled from GUI, in order to support MAC randomization
security mechanism driver should not flush scan cache during interface
down. Framework has to set the scan flush flag so that driver will flush
the scan results while interface up and then do the scan.
Remove scan flush logic from driver on interface down.
Change-Id: I45ad1a257ab7ef4bd5474f52fd0dbfde5e3514c4
CRs-Fixed: 2360898
Per the Linux coding style "mixed-case names are frowned upon" so
remove the leading "e" from the eHDD_DRV_OP_* enumerations .
Change-Id: If756cbc89fb87f0dd8f4b0f7e8200d488a0215e7
CRs-Fixed: 2361015
The information previously contained within tdlsInfo_t has been
relocated to struct tdls_info in the TDLS component, and tdlsInfo_t is
no longer being used, so remove it.
Change-Id: I96a4b7c442f9f42e53d6eeb4a7ba3182737d9e2c
CRs-Fixed: 2361021
Add the following WLAN_REG ini configs and FE_BEAMFORM
cfg items to MLME cfg:
1. WNI_CFG_11D_ENABLED
2. WNI_CFG_11H_ENABLED
3. WNI_CFG_TX_BF_CAP
4. WNI_CFG_AS_CAP
5. WNI_CFG_DISABLE_LDPC_WITH_TXBF_AP
Introduce the basic infra APIs related to these configs
from mlme.
In csr_roam_open_session, lim_populate_vht_mcs_set,
lim_sta_send_add_bss, populate_dot11f_tdls_ht_vht_cap,
populate_dot11f_vht_caps and sme_update_nss, local structure
is updated instead of mlme global structure for vht caps and
thus the mlme vht caps contains only the default values.
Fix this by updating mlme vht caps in mlme_update_vht_cap.
Change-Id: I1a948d7a44ad2bcc7f711c46b278971d7e4026e7
CRs-Fixed: 2354562
Invalid channel_count may be sent to driver by user with
QCA_NL80211_VENDOR_SUBCMD_DCC_UPDATE_NDL vendor command, and may also
pass the sanity check with update_ndl_param->dcc_ndl_chan_list_len as
integer overflow, in send_dcc_update_ndl_cmd_tlv. This leads to heap
overflow while updating NDL per channel.
Fix is to validate channel count sent by user in
__wlan_hdd_cfg80211_dcc_update_ndl before accessing it.
Change-Id: Icf385dd5618fcdbdbf427062fe581fa7d3909882
CRs-Fixed: 2339182
If power-save is disabled then driver is returning failure to KERNEL
which causes framework and kernel to go out of sync and therefor
kernel doesn't accept any power-save enablement command from framework.
CRs-Fixed: 2357896
Change-Id: I36db68454f1ff6661563d021fc5b75b8161e8a78
NAN related OS IF, Target IF and component files have been
merged into CLD. Add Makefile changes, include correct
files, replace lmac tx/rx ops with the new ops and call
updated API's such as init/deinit.
Complete the remaining changes for moving NAN component
into CLD.
Change-Id: Idcf8816337aa49b399596ce3cdfc9cdab811d386
CRs-Fixed: 2338060
When CONFIG_MOBILE_ROUTER is enabled there are build failures
due to improper featurization of TDLS, so fix the featurization
build failures.
Change-Id: I0e6b21615e763f2c4c86e0092807062cc8397705
CRs-Fixed: 2355801
The copyright comment block in wlan_hdd_nud_tracking.h is not properly
aligned, so fix it.
Change-Id: I180980435c062fbdc43db58f84d9b48335f6bdc6
CRs-Fixed: 2359488
Add power items related CFG items in new file,
and store the respective inis in MLME component
Change-Id: Iac183da52495aca07c6270edeb9f383ab49c1ff5
CRs-Fixed: 2341900
There are multiple issues with how hdd_conn_set_connection_state()
is exported:
1) The prototype is defined in wlan_hdd_cfg80211.h even though the
implementation is in wlan_hdd_assoc.c.
2) The prototype uses the "extern" keyword which violates the Linux
Kernel coding style.
3) The implementation is documented instead of the interface.
To address these issues move the prototype (without "extern") as well
as the documentation to wlan_hdd_assoc.h.
Change-Id: Ie2b80e953cc500c10ae14d94c375cc1487241b6c
CRs-Fixed: 2357078
qcacld-2.0 to qcacld-3.0 Propagation.
the wma_send_time_stamp_sync_cmd will not be processed by fw,
so remove this wma_send_time_stamp_sync_cmd.
CRs-Fixed: 2280656
Change-Id: I3aee6e2da4783772ad9acf3fc20c3669829f9397
Since QDF TRACE INIs used infrequently, remove them and related codes.
Please file new FR to trace the requirement about host log level
setting.
Change-Id: I9c519cf2509e3dc22282f0c7ea6f52cbd2913ccd
CRs-Fixed: 2358850
qcacld-2.0 to qcacld-3.0 Propagation.
Connection fails after ssr due to the tx queues are paused by tx flow
control and ssr happens before tx resume, and the tx queues cannot be
resumed because of the unpaired status of pause map.
Resume tx queues and update pause map at the beginning of ssr if the tx
queues are paused by tx flow control.
CRs-Fixed: 2146129
Change-Id: Ib1ed8360c6effa4a47fc1702650375bf2380e65c
The 3rd party AP with Vendor IE OUI 001018, Vendor Data
02 FF 00 9C 00 00 and NSS 4x4 is not able to handle OMN/SMPS
frames sent by DUT in 2.4Ghz. This leads to the AP dropping
the data rates to 1Mbps and low throughput is seen.
This is because the connection is done with NSS 2x2 and when Coex
scenarios occur, the DUT switches dynamically to 1x1 by sending
OMN/SMPS frames. To overcome this issue, the workaround is to
blacklist the above AP and do connection in 1x1 only.
Add the vendor OUI of the 3rd party AP to gActionOUIConnect1x1
default INI string to connect in 1x1.
Change-Id: Idc0f3238e3521bb20c592b44de77216125e69504
CRs-Fixed: 2352946
Convert the current hard-coded list of unit-test callbacks in
hdd_we_unit_test() to a vtable. This streamlines future additions.
Change-Id: I216bbb6699ae50eaa96ac559999cb42ba080867c
CRs-Fixed: 2358606
Since refined second part of LFR INIs, apply these changes in HDD
layer. Delete related legacy codes in HDD layer.
Change-Id: Id39629e33e87ffe6ed58f375609a55e324618693
CRs-Fixed: 2346787
Based on new cfg component, refactor CFG_TX_POWER_CTRL_NAME and clean
up related legacy codes.
Change-Id: I6dd317d5065c4111b71a9fe94d9b9af1f7d3d664
CRs-Fixed: 2355918
When CONFIG_MOBILE_ROUTER is enabled there are build failures
due to improper featurization of NAN, so fix the featurization.
Change-Id: Ib5cd54b2d443ced73c9b99218ce28d3eb0ba8786
CRs-Fixed: 2355789
Currently if a vdev is destroyed, wow patterns are not
getting cleared for that vdev, so when a new vdev is
created, host tries to configure new wow patterns for
this new vdev but due to FW limitations on number of
wow patterns that can be maintained at FW, FW is not
able to act on this filter and will not maintain this
new wow filter.
To resolve this issue, clear the wow patterns for the
vdev before it gets destroyed.
Change-Id: Iae7ad8b683c95d8eaf11355bc0e36b91cece2202
CRs-Fixed: 2344337
In hdd_config.h file, wlan_tgt_def_config_hl.h is included without checking
for the flag CONFIG_HL_SUPPORT which could lead to compilation issues if
CONFIG_HL_SUPPORT is not defined for a PL.
Include wlan_tgt_def_config_hl.h under the correct CONFIG_HL_SUPPORT
flag.
Change-Id: Ia01b79e47b9ea504a93c96575599068218ef8671
CRs-Fixed: 2357292
Based on new cfg component, refactor INI CFG_OVERRIDE_HT40_20_24GHZ_NAME
and remove legacy codes.
Change-Id: I5b4ed96a4930a9358bd93ee809bf5cddb2787fb4
CRs-Fixed: 2354476
If NAN or NDP support is disabled through INI, repeated NAN/NDP
requests from upper layers flood console logs. Convert these
error prints to rate limited prints to avoid excessive logging.
Rate limit error logs in NAN and NDP cmd handlers.
Change-Id: Ie3e7fce9e829c64d3d21d856a72a957365a04aa8
CRs-Fixed: 2355727
Move the following INIs definition from current definition to new
HDD component model INI definitions in hdd_config.h.
gOperatingChannel
gNumVdevs
gEnableConcurrentSTA
gdbs_scan_selection
Change-Id: Id65ca1ff1954af2d220e7221bfc5181e04b3c63d
CRs-Fixed: 2354920
During Vdev start/restart LimProcessDefdMsgs is set to false. If
start/restart timeouts, it will initiate SSR but will not send
response to LIM and thus gLimProcessDefdMsgs will not set to true.
Thus no other commands will be processed in LIM. If any command is
queued in serialization during this time it will timeout.
This will eventually lead to timeout of the vdev destroy command as
it won't get processed until the command in active queue timeout and
it leads to peer and vdev refcount leaks.
To fix it purge all active and pending cmds during SSR, so that vdev
destroy cmd can be processed immediately and can release vdev/peer
ref count.
Change-Id: Id5c15ccb4c27dabfdd4a9e0d7ee9b765130e03bf
CRs-Fixed: 2354899
Cleanup multiple HT related CFG/INI Items which are already moved to the
MLME component.
Also move gShortSlotTimeEnabled INI from HDD to MLME component
Change-Id: Ie8c19f508401d4d72990f35ea5dcc18d45f0f16b
CRs-Fixed: 2354687
Refactor ini items gSapForce11NFor11AC and gGoForce11NFor11AC
into new mlme component.
Change-Id: I85e2f7f9ae14a23c322572290b095aae0f29b0ec
CRs-Fixed: 2354940
