| Commit message (Collapse) | Author | Age |
| ... | |
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
When WLAN driver unloading, reset IPA uC ready callback.
Change-Id: Icc973ae3e8a31bc01d49aeed799b1f0c1cea170a
CRs-Fixed: 956668
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
qcacld-3.0 to qcacld-2.0 propagation
RRM ie should not be present in association req if AP doesn't
support 11k. As a part of fix add the check to append RMM ie only
if AP support 11k.
Change-Id: I4e521dbdbdd5b083360405f26913278788c86555
CRs-Fixed: 961737
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Update scheduler parameters for NON_QOS_DATA catagory to
make it similar to BE.
CRs-Fixed: 962191
Change-Id: I3c3aee13152303f89c79f5946e015d2d96cc395a
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Wakelock feature has been removed from upstream kernel
and wakelock in Linux has changed to wakeup_source.
The wakelocks internally calls wakeup_source.
CRs-Fixed: 956999
Change-Id: I3863b5345fed7ed0be0536585c7dfb3deac19b8a
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
prima to qcacld-2.0 propagation
There are many info logs which are of
ERROR type and vice versa. As a part
of fix, clean up such logs.
Change-Id: I8114abb4deb10ed789d0f1000033a9b772a98519
CRs-Fixed: 920899
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Firmware can't support all station feature and 32 ibss clients
in one bin because of the mem limitation. So host supports to
download IBSS firmware bin if con_mode=6.
CRs-Fixed: 962246
Change-Id: I6c3dca17c82fdfd2bdadcee319cc157e6aa2ead9
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Release 4.0.10.255
Change-Id: I037e59f696eac34b2f18bcbabc62256df7f7ef9a
CRs-Fixed: 688141
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Roaming fresh assoc because the preauth frame has high possibility
sending on wrong channel if in MCC. Need to trigger scan to shcedule
to target AP channel if DUT in MCC state.
Change-Id: Ifee8bc00c10ea995093e3f4621b94e0d81a7c32e
CRs-Fixed: 962162
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
HL SDIO solution FW memory dump is too slow,
which doesn't meet the SSR latency requirement.
Shorten the vdev stop timeout value and shorten
the Target suspend timeout value to reduce the
total SSR time.
Change-Id: I7d994ab04bcc9ff29653464a05fddf1aa31f8d1f
CRs-Fixed: 960980
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Make the following enhancements to the EXTSCAN feature:
* Add buckets_scanned param while sending
on_full_scan_result/cached result event to user space.
* Configure REPORT_EVENTS_CONTEXT_HUB to firmware.
* Remove status while sending on_scan_event to user space.
* Add support for SCAN_THRESHOLD_NUM_SCANS /
SCAN_THRESHOLD_PERCENT while sending scan event.
Change-Id: Ib37ce7cef5b3ce9d9ee4ebdc5ea9bcadaad7b695
CRs-Fixed: 960456
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
WMI changes for gscan enhancements
CRs-Fixed: 962169
Change-Id: If319738b42d4e71aec8d21624546ca039232c413
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Make the following enhancements to the EPNO feature:
1) Implement the reset EPNO command handler
2) Add new parameters for candidate score calculation.
Change-Id: Iad1ff9c2b003c9e3ddbc3373366686040ccb55ca
CRs-Fixed: 956649
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
WMI changes for EPNO enhancement
CRs-Fixed: 962167
Change-Id: I8439c4f574738d89d54300e08fbd7e2d536863a9
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Add qtimer to TSF capture request and response WMI message/event.
CRs-Fixed: 962166
Change-Id: Ia10eca66cfbff821ef1f4c03a0fb964d504d53a2
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Vmalloc will introduce discontinuous memory area. This is not
convenient to extract FW ramdump from crashed RAM dump using a
HEX editor.
Vos_mem_malloc should be a better choice.
1. Memory allocated is continuouse;
2. We can use pre-alloc to avoid memory allocation failure.
Change-Id: I95cad84db75690c1c0aafbf787f95f68d5056086
CRs-Fixed: 962085
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Deactivate and clean up the logging thread during driver load failure and
ensure that logging thread is not enabled for FTM mode
CRs-Fixed: 961865
Change-Id: I53f85a30f52fc5715705624bf515ab5e4666db8e
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Currently in MDM platform when driver hit VOS_BUG, it just prints
out error msg, needs force crash when exception occurs.
Change-Id: I2e57889d38644e8bbf4b03ffa3891382dc699f54
CRs-Fixed: 962064
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
If extscan ini is not defined, then get valid channels for SAP
fails and default channellist is passed to SAP.This causes
SAP to be brought up only on 2.4G channels.This fix removes
extscan ini check in get valid channels API.
Change-Id: I9a6b6c3485e52ea4a18884a73572e68b79d1a79b
CRs-fixed: 961096
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Fix null pointer dereferencing while populating session
supported NSS
CRs-Fixed: 961984
Change-Id: Ic214f1432cc1913a1f65a217806f2dd2651ed85d
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
We need to stop SAP bss and destroy the SAP
session if the channel switch failed. Otherwise,
the vdev will leak and it will cause SoftAP
start failed in following test.
Change-Id: Ieb3a2d62c8698f245f787be3c215ee4365b2e8ba
CRs-Fixed: 944741
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
The current hardware name of QCA9367 and QCA9377 is the same name
"QCA9377_REV1_1". It will cause some confusion when use QCA9367.
Change the name to "QCA93x7_REV1_1" to avoid this confusion.
Change-Id: If36a7d718380bd0afd5cf2e49a8eca2fe12a6acf
CRs-Fixed: 961428
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
create_linux_regulatory_entry() does not return even if pHddCtx and
pVosContext are NULL.
Add return statements to avoid deferencing of NULL pointer when pHddCtx
and pVosContext NULL.
Change-Id: I25ad15d4c8e2544f7d3f9b22af4c8039a6da855c
CRs-Fixed: 957920
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
PTK exchange and install happens during preathentication if key_mgmt is
FT-PSK. STA TL state moves to athenticated only if ptk_installed is true.
Currently ptk_installed is false after roaming due to which STA TL state
does not move to athenticated state and also does not enter BMPS.
To fix this issue, set ptk_installed to true in case of 11R roaming
Change-Id: I59e4466994e59b055246e86613315b0299b767a7
CRs-Fixed: 961555
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
If a connect is received with previous connect already in progress,
sta connect state is moved from connecting to disconnecting and
disconnect is initiated to stop the previous connect.
If previous is present in sme/roam pending queue the disconnect
will abort the previous connect by removing the cmd from pending
queue. In this case the connect state is not moved back to
not connected from disconecting.
Thus the current connect command fails and all new connect command
will fail as the connect state will always be in disconnecting state.
To avoid this, move the connect state to not connected, once try
disconnect is issued internally.
Change-Id: Ia563f85df84cb9dbb3d3a82bc97127d39e7b1421
CRs-Fixed: 961559
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Release 4.0.10.254
Change-Id: Iaa438507dfe7c8a7fb8a1842255372a258c6d551
CRs-Fixed: 688141
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Fix static code analysis errors in hif and sdio modules.
Change-Id: I866368be1ad9006e41a24bf58310b7eb1203b49d
CRs-Fixed: 957920
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
In High dense environment roaming should be triggered early
at dense RSSI threshold which provide better throughput results.
Add support for dense roaming parameters to read from ini file
and send to FW with other roaming parameters.
CRs-Fixed: 953579
Change-Id: Ia63bfdd1248b4507d988b019f07164a7c75566af
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
add wmi_roam_dense_thres_param params defs
CRs-Fixed: 960961
Change-Id: I20b18babdde45b05f1675cf708ef9663141d8541
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Bus bandwidth voting level 0 was not handled originally.
Then lowest performance level was 1. But by design change,
lowest performance level changed as 0. If not handle level 0,
WLAN cannot clear pm_qos voting then, system consume more power.
To fix issue, WLAN bus bandwidth voting module should handle
performance level 0 also.
Change-Id: I8afea50b5a20c120feed5f0f2b4956e42a1799fb
CRs-fixed: 959871
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
prima to qcacld-2.0 propagation.
When WEP algo is AUTOSWITCH, driver will try OPEN AUTH first,
then SHARED AUTH.
If AP accepts both OPEN and SHARED AUTH, driver never tries to
connect to AP with SHARED AUTH, which makes user is not getting
the notification (about wrong password).
This change makes driver to try SHARED AUTH first, then OPEN AUTH
when WEP algo is AUTOSWITCH.
Change-Id: I21841f37d96316a261fd2977229390e751747c4c
CRs-fixed: 947898
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
prima to qcacld-2.0 propagation.
With invalid WEP password, some APs reply AUTH frame 4 with
invalid seqNumber. This AUTH frame will be dropped by driver,
then driver sends the generic status code instead of protocol
status code.
As a workaround, assign the seqNumber=4 for AUTH frame 4,
which ensures driver to send correct status code when WEP AUTH
fail.
Change-Id: I2fc96f0c253cf639d730d59d567be50bca51bd0b
CRs-Fixed: 947898
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Release 4.0.10.253
Change-Id: I4a226e72f469fc1ca82a13156d3eee779e68393f
CRs-Fixed: 688141
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
sdio core driver return NULL for sdio device func and device
ID if the Wi-Fi card unplugged or hardware programming is not
proper for Wi-Fi card.
Check against NULL pointer before use to avoid crash.
CRs-Fixed: 959670
Change-Id: Ibaa9687e3040191deef005dd97885dcde9893bcd
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
"log buffers are dropped" are consistently printed which
is affecting the tput KPI's. Masking the same with debug level
which is not enabled by default.
Change-Id: I6e9abc7d65f276d71810aedad9062152035699c0
CRs-Fixed: 956599
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
prima to qcacld-2.0 propagation
Currently driver has debug print for EAPOL
but it doesn't tell whether its M1, M2 etc.
As part of fix, add debug print for M1-M4 msg.
Change-Id: I1eb4f9dac6520b1ce00cdfc3e0fcef721ae498d6
CRs-Fixed: 919617
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
CONFIG_CNSS has been enabled for both PCIe & SDIO based
interface Wlan module and this feature add support to use
common wrapper API to call respective platform exported symbol
for both SDIO and PCIe based interface. Remove redundant wrapper
API and update conditional compilation config flag.
CRs-Fixed: 959151
Change-Id: Ie2f873512b4050577096abca04a339ae856209a7
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
prima to qcacld-2.0 propagation
This fix contains the necessary changes to detect thread stuck
issues in driver. The changes are to post message to MC thread
for every few seconds, which elicits a response from this thread.
A loss of response for consecutive three times is regarded as
thread stuck.
Change-Id: I3184db4dd7472f0c65dd607af7c197c52139cdb0
CRs-Fixed: 862223
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
prima to qcacld-2.0 propagation
Modify the timestamp format for the logs from
[Secondselapsedinaday.microseconds] to [hour:min:sec.msec].
This will help in correlating the logcat logs and
the Logging infra logs.
Change-Id: Ib0f72bc0e48c1f2e02816da24fca4fb10e7b26fe
CRs-Fixed: 771157
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Release 4.0.10.252
Change-Id: Ia0defa6b5edbc2538ebc0b16f0aa49542097b4e5
CRs-Fixed: 688141
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Do not allocate single large chunk of memory for rx hash
to avoid memory allocation failure. This change uses array
of pointers to reduce single large chunck of allocation.
Change-Id: I2aa7239be924b63c87f1044a74176a101d952108
CRs-Fixed: 954302
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Avoid logging of all module logs into kmsg during the driver load
phase. All module log level is set much before setting of the
flag that indicates whether multicast logging is enabled or not.
Because of this all module log levels are getting printed for a
brief duration in kmsg during driver load. Fix the same by
setting the multicast flag appropriately so that when multicast
logging is enabled, all the logs are multicast to user space
and only the error and fatal messages continue to get logged in
the kmsg.
CRs-Fixed: 958705
Change-Id: I260e38ef64f704dfd9ca1ca4dc68c56428526559
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
There was an issue in the earlier logic where, the counter num_rx_
excep will get incremented even for intra BSS packets which are not
meant to forward to local stack. So if get a mix of local and non-
local exception packets, the older logic can miss a call to 'netif_
rx_ni' for a sometime and can cause a spike in SKB's in local network
stack rx queue.
The fix is to introduce a new dedicated counter for the threshold
packet count.
Change-Id: I46447eba37f0152fcae8a0763bfea1f240826467
CRs-Fixed: 950379
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Add data length information in oem data request messages.
This is done to support TLV message format between LOWI
and firmware, which are passed through host driver.
Change-Id: Ibc132d31dd4345c4168dad23b4acf699f2a8c8e6
CRs-Fixed: 942260
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Add event handler for oem data response event from FW.
Add new boolean member to differentiate the response
generated internally and one coming from the FW.
Change-Id: Ia906c5dcbb7e560fd78ccd96fc47c7801b33aa6a
CRs-Fixed: 942261
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
prima to qcacld-2.0 propagation
If back to back connect and disconnect are issued by supplicant,
disconnect ends up queued at the head of roam/sme pending cmd
queue due to its high priority. Thus sme processes the disconnect
command before the connect command making sme and hdd out of sync.
Now if unload is triggered, with hdd in disconnected state,
disconnect is dropped in hdd. Thus no DEL BSS is sent before
HAL STOP leading to firmware crash.
This change removes the roam connect cmd from roam/sme pending list
when disconnect is issued by supplicant.
Change-Id: I34f192b3d549c6179ddc49bba376dbc007efa2c2
CRs-Fixed: 955178
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Release 4.0.10.251
Change-Id: I6cd056b62692b148c67eaef5a11fa03c79797829
CRs-Fixed: 688141
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Use SKB clone instead of using SKB copy for forward packets.
Addtional fix to release SKB to IPA when dropping Rx packets from IPA.
Change-Id: Ibfacf855b53148fd6b254e281f7163d03e3753ec
CRs-Fixed: 950379
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
prima to qcacld-2.0 propagation.
If MC thread stall while waiting on any event, In case of WPS,
probe requests may get accumulated in PE message queue.
Now every time MC thread process a probe request,
eWNI_SME_MGMT_FRM_IND msg is posted in SME message queue. But
unless PE queue is fully processed SME queue won't be processed.
This will gradually use all vos message wrapper.
To avoid this register a callback to send management frame
indication directly to HDD from PE which avoids posting message to
SME queue.
Change-Id: Ib83700825112cc52dade594909bfa8993909ac29
CRs-Fixed: 944961
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
If connect is received with previous connect in progress, driver
doesn't abort the previous connect and process the new connect.
In csr if new connect is for same profile, sme directly indicate
connection success to hdd and thus hdd increment the active
session count twice for both connect.
Now if one more connect is received it will try to disconnect the
previous connection and decrements the active session count.
So the count is 1 though no session is active.
Eventually if same scenario happens again the active session
count will become 2 with no active session. As max active session
limit is 2, no more connect is accepted.
To avoid this, if fresh connect is received and previous connect is
in progress abort the previous connect.
CRs-Fixed: 958963
Change-Id: I1544bc6e7894ffeda1956a19a553b6fe4e35f988
|
| | | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
If ACS is enabled in hostapd configuration,hostapd sends
a list of channels to HDD for softap bring-up. If hostapd
finds a invalid/unsupported channel in its channel list,
it will send a 0 for that channel. Since HDD always takes
the lowest channel as starting channel for scan, the scan
will always include 2.4Ghz channels even if channel mode
is set to a. Hence HDD should ignore these invalid entries
in channel list.
CRs-fixed: 959490
Change-Id: Iddaf6c4ba14e22b476f22f1f098faf726f4407ec
|