| Commit message (Collapse) | Author | Age |
|---|
| ... | |
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Provide fix for the unexpected kernel trace while unloading
the driver. The kernel trace was caused due to access of vos
lock after the closure of VOSS as a layer. This happens when
the kernel invokes module_exit() followed by ndo_close()
callbacks. ndo_close() does not take unloading operation in
to consideration and attmpts to clean up the driver by stopping
the interface and also stopping the scan which results in
accessing of VOSS data structures
Change-Id: I63efa15f243cabf290bae31a6fb43fbfb74b0920
CRs-Fixed: 675991
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Previously Change-Id I2be31e9e5bd3ce5c8821156b28578fec39c719bc,
"qcacld: hdd: Fix memory corruption while updating config dat", fixed
potential stack corruptions when modifying WNI_CFG_HT_CAP_INFO.
Recent comparison of the prima and qcacld implementions of this fix
revealed a slight difference in the two fixes. Since the prima fix
looks "more correct", align the qcacld fix to mirror the prima fix.
Change-Id: Id2225f15717b4b939c00b5a0bad1aee4180bdbe8
CRs-fixed: 679359
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
This commit updates host_interest_s struct to save SW ROM version.
Change-Id: Id48a0f2b16e35690765ce6240666acdc235a01d4
CRs-Fixed: 682930
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
This commit adds WMI changes for supporting heterogeneous IBSS network.
Change-Id: I7eb700cb6697e67a10dd9b2042a7d90744cd1343
CRs-Fixed: 682921
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
In case of SAP running in DFS channel, ic_curchan would be allocated
while it was first brought up. When it detected the radar and
restarted the SAP after randomly picking channel, the ic_curchan was
still holding the memory allocated. Add sanity checking and release the
memory if it was allocated.
Change-Id: Ie9f92c636ad0e15444403ae2a5ed670eca23c554
CRs-fixed: 680517
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
In the CAC_WAIT state, when stop bss, we should transit to
disconnecting state to clean up the resource that it allocated when
creating the bss, instead of moving to disconnected state, which
will leave the timer and some callback resource behind, that causes
the unexpected post message coming and hence assert happens.
Change-Id: I04bb2ee476864a7fcab977a1b651551ce4968d6c
CRs-fixed: 681993
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
When attempt of connection fails, driver sends the generic
reason code instead of sending protocol reason code.
This restricts application not to reconfigure station to
SHARED-WEP when OPEN-WEP authentication fails.
This change has been done in a way that driver will send exact
protocol reason code in case of OPEN-WEP or SHARED-WEP besides
station interface is not in connected state.
Change-Id: I33fb59a3654fbfbf113376a737ab4d60bcc12749
CRs-fixed: 679890
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Compat handling is required for some ioctl handlers when used in
32U/64K environment.
Adding a common function hdd_priv_get_data() for compat handling
so that when the driver is part of 64-bit kernel, handlers can
use it to work with 32-bit userspace applications
Change-Id: Ifb1e2022301e6b34ae9c1a7aca89a891642d834e
CRs-Fixed: 681631
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Since ula_complete variable is used in wait call without initializing
the same leads to device crash, avoid the same by initializing the
ula_completion completion variable in P2P GO/SAP context.
Change-Id: Ib1f86cb57e53eac01401601817518ffaca30b856
CRs-Fixed: 680923
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Currently cfgState->remain_on_chan_ctx is assigned
to pRemainChanCtx without NULL check of cfgState.
Added a NULL check prior to assignment.
Change-Id: I5b530c15f1e611b287e702b0c69aaff16fd08915
CRs-Fixed: 680921
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
-Add SSR protection for resume and suspend
indication processing.
-Increase the SSR WAIT time in WD thread.
Change-Id: Icfbbb983fa995564027a2dcd17709369e70776ac
CRs-Fixed: 680446
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Reduce log level in hdd_roamRegisterSTA() which is coming
because of the test script which is issuing connect/disconnect
back to back without any delay so by the time STA is registered
with TL disconnect is coming form CFG80211 which is changing STA
connection state to disconnecting state and because of it ASSERT
log is seen.There is no functional failure during the test because
STA is getting disconnected and connected properly.
Change-Id: I20f4fcbdbbac1bf8677abe8ca76398a706ce96bc
CRs-Fixed: 681999
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
While driver loading failure case we are freeing up the adf context
but adf context in the scn handle is not getting deiinitilaized.
Beacuse of this in hif_recv_buffer_cleanup_on_pipe() we are acccessing
the inavlid adf context and results in crash.
So deinitialize the adf context and validate it in
hif_recv_buffer_cleanup_on_pipe() as well.
Change-Id: Ifad54e00fd896bf05d77061460217ba9e7af7a7a
CRs-Fixed: 666102
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
p2pcli is 2.4ghz band, sta is 5ghz band, p2p side throughput
is much less than sta throughput
problem is when channel switching from sta to p2p sta interface
could not drain pending frames enough. so lbw interface could not
have chance to push frame to firmware.
lower LBW interface low watermark value then lbw interface will
have more change to push packets.
Change-Id: Iec8b0571db02f8695cd9529d0b827f5e9c71d803
CRs-fixed: 672262
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
For saving more power different configurations are configured
during suspend.
Dynamic DTIM:gEnableDynamicDTIM
By default this value is 0.
If set to any non zero value during suspend if the beacon dtim period is
less that dtim threshold i.e default 2 dtim policy will be changed to
Normal Dtim and the LI will be set to the value in gEnableDynamicDTIM.
Modulated DTIM:gEnableModulatedDTIM
By Default this value is 0.
If set to non zero value then DTIM will be confugured as follows:-
If maxModulatedDTIM ((MAX_LI_VAL = 10) / AP_DTIM)
equal or larger than MDTIM (configured in WCNSS_qcom_cfg.ini)
Set LI to MDTIM * AP_DTIM
If Dtim = 2 and Mdtim = 2 then LI is 4
Else
Set LI to maxModulatedDTIM * AP_DTIM
In system resume the dtim policy will be restored back to stick dtim.
Change-Id: I1b24cf49fdc7a9ff636fcab3e66343bb7d820d4c
CRs-Fixed: 674826
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Dynamic DTIM and Modulated DTIM parameters need to passed from
umac inorder to configure these params during system suspend case.
Inaddition pass last beacon dtim period also which is required
to compute the listen interval.
Change-Id: Ia451835051a4aea8fc81195458356a1332626d37
CRs-Fixed: 674826
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
When driver issues active scan command, the firmware will do
active scan on non-DFS channels and passive scan on DFS channels.
In the current implementation the dwell time for passive scan on
DFS channels is same as dwell time of active scan.
This dwell time is not enough to receive the beacon from the AP
on DFS channels. So these APs on DFS channels are getting
discovered after fews scans.
The dwell time for DFS channels should be sufficient to receive
the beacon from the AP.
Change-Id: Id79b579dff2036a61912a6b635ad3fda6bf02483
CRs-Fixed: 676024
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Driver is sending the WMI_PEER_AUTHORISE command twice in open
and security modes. It is sending once after association and
again after key installation. This is affecting the key exchange
as firmware is cancelling channel request after association
once it receives the first WMI_PEER_AUTHORISE. So 4-way handshake
is failing sometimes.
So removing the redundant code to send the WMI_PEER_AUTHORISE
command twice.
CRs-Fixed: 681209
Change-Id: If2837a69dfbe7cd1f5bdc8ab1e6fe9197b193231
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
The change introduces some regression when operating at security mode
in ROME SDIO/USB. Revert this change.
Change-Id: If2f99394ecc05374b5c3e95a9ea28f9339ec8254
CRs-fixed: 683388
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
LIM processes beacon parameter only for the first active
AP session. So the beacon template is not updated for the
guest AP when the primary AP is active.
To fix this:
1) Retrieve all AP sessions using peFindSessionBySessionId()
and send Beacon Params to each of them.
2) Beacon/ProbeRsp templates buffers are now moved under PE
Session, so that each SAP interface have a copy and they
cannot overwrite other's template.
Change-Id: Id744f929cd284db3f6420e7f20317f648eee5f37
CRs-fixed: 647476
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Add CONFIG_CNSS flag to protect SSR related functions for PCIe which
should not be compiled if there is no CNSS platform driver.
Change-Id: Ifb0b381fd87510bd3d52f1251e753e7a96d67c00
CRs-fixed: 644535
|
| | |\ \ \ \ \ |
|
| | |/| | | | |
| | |/ / / /
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Merge remote-tracking branch 'origin/caf/caf-wlan/master'
* origin/caf/caf-wlan/master:
Cafstaging Release 1.0.0.129
CLD: Fix overwrites of mbssid sap config merge
qcacld: Add SAP Channel Change iwpriv with CSA
qcacld: HDD: Fix identical mac address issue (STA + STA)
wlan:Protocol Stack Add Passpoint 2.0R2 support in the QCACLD WLAN driver
wlan:HDD: Add Passpoint 2.0R2 support in the QCACLD WLAN driver
qcacld: Fix to retrive proper vdev id for PMF operation in TL layer
qcacld: Set correct AC when ACM is set in high latency mode
qcacld: Do not suspend wlan in the middle of roaming.
qcacld: CL 973382 973435 - update fw common interface files
qca_cld: Copy firmware dump after kernel panic
qcacld: Delete MCBC filters configured in resume.
qcacld: Selectively enable active scans on DFS channels
qcacld: Serialize the wma tbtt offset update event
Change-Id: I59952824c31be296c65aa69fd7d1dfd472988bfa
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Cafstaging Release 1.0.0.129
Change-Id: I8a18e28540f6f8b42b8f898825bcd254a6c98256
CRs-Fixed: 682625
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Fix some code overwrriten during merge conflict resolving during
propagation of mbssid SAP config update.
Change-Id: Ieb57c362def787065324d2e211c3974b6fe198e4
CRs-fixed: 682618
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Adding support for iwpriv command to allow SAP
channel change with CSA IE and Channel Switch
wrapper IE in the beacons and probe response.
Channel change is allowed only from one 5Ghz
channel to another 5Ghz channel. Channel switch
wrapper element is present only for channel width
of 40 or wider channel switch operation.
Change-Id: Ied87b98d7016a1a1a12a6a99596f66642d5f55ba
CRs-Fixed: 638484
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Recent fix of randomly generating valid mac address for second
STA interface was resulting in identifal mac address of both
STA interfaces.
Change-Id: Ia3d413bc602aed2f1984e314f945f53ccfdbcd16
CRs-Fixed: 680384
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Port the changes introduced to add Passpoint 2.0 support to the Pronto
WLAN driver into the QCACLD driver. this commit incorporates the
changes required in protocol stack for registration for Passpoint 2.0
Action frames (ANQP and WNM), and support for OSEN and QoS mapping.
Change-Id: Ib2dfdc398a666965ef69981b33ae21a8ac40eca9
CRs-Fixed: 672071
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Port the changes introduced to add Passpoint 2.0 support to the Pronto
WLAN driver into the QCACLD driver. This Commit incorporates the
changes required in HDD module for registration for Passpoint 2.0
Action frames (ANQP and WNM), and support for OSEN and QoS mapping.
Change-Id: I92fe7ad6d64fc8fea1504782668d4239af36804d
CRs-Fixed: 672071
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
with current design, we are passing default vdev_id as 0 which is
in-correct. with this fix, we will retrive the vdev_id based on
received frame's bssid and use it in PMF operation.
CRs-Fixed: 661624
Change-Id: I3b7fd9ff322447a149e491d913c272e0baad875c
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Due to in the Tx path of high latency mode, tid/ac will be judged
by ol_tx_classify.
Since adf nbuf has carried the info of tid, clasify the tid again
only when tid info is ADF_NBUF_TX_EXT_TID_INVALID.
Change-Id: I06dcc7851b997d2bccc47d28cd48af814bbf8c2e
CRs-fixed: 668705
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
The wlan suspend code path is different in pronto and CLD.
In CLD driver, during cfg80211 suspend wlan there is no check
to make sure if driver is in the middle of roaming, hence
if the wlan suspend happens while roaming is also in progress, then
sending data via hdd_hard_start_xmit() after PCIe is suspended
caused L2 master port error.
Fix this in wlan_hdd_cfg80211_suspend_wlan by adding a check
to see if driver is in the middle of roaming.
reject the suspend if roaming is in progress so that kernel
shall try to suspend again later
Change-Id: If4f769c0e328df360863c65bd16e0d36434dcec5
CRs-Fixed: 681513
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Adding OEM types and dfs header changes for fwcommon files.
Change-Id: I57513d20937bb330707e4f87de0fc17423087fc6
CRs-Fixed: 680941
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Directly dump CE registers and copy firmware memory after kernel panic.
Change-Id: I8154e37b2571f20e4f7fba646952cccceb983ca3
CRs-fixed: 644535
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Delete the MCBC filters in resume which are configured in suspend.
CRs-Fixed: 681150
Change-Id: I17588134dfa5c6050c9ce9e6a5d36343130d2f5b
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Roaming scan on DFS channel is supported, but it performs a passive
scan only.
Added another option to SETDFSSCANMODE command. Allow active
scanning on DFS channels when mode is set to 2.
Change-Id: I4c67939e93f9821aaf96b226babe01aa0073f952
CRs-Fixed: 673873
|
| | | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Currently the wma tbtt offset update event is not serialized to
VosMC thread.So the crash is seen during the race conditions.
Changes are done to serialize this event to MC thread to avoid
the crash.
Change-Id: I3be0c1698fba6b882065361bb304f4e723381269
CRs-Fixed: 680907
|
| | |\ \ \ \ \ |
|
| | | |\| | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Merge commit 'fdcbd508df2843b9345b3e55a93b7302dbb7b27e' into HEAD
* commit 'fdcbd508df2843b9345b3e55a93b7302dbb7b27e':
Cafstaging Release 1.0.0.128
qcacld: Compilation error on x86-64 PCIE
qcacld: Fix compilation error
Skip channel 14 only once in channel selection initialization
Droidsec: sscanf parameter count check
wlan: Limit the Deauth Frames sent by AP to STA
qcacld:Fix compilation issues on enabling CONFIG_ATH_PCIE_ACCESS_DEBUG
qcacld: CL 986653 - update fw common interface files
Timer leak for tx_flow_control_timer in wlan host driver
qca-cld:sap: Improve HT80 primary channel selection
qcacld: HDD: Provide fix kernel panic while adding virtual intf
Change-Id: If723b0b437f2d9e19f9a47459221fc8fd6ea0d8b
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Cafstaging Release 1.0.0.128
Change-Id: If723b0b437f2d9e19f9a47459221fc8fd6ea0d8b
CRs-Fixed: 681755
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Fix a build error observed on x86-64 PCI platform.
Change-Id: Icb675abc1c112e5e50adc5cf6dc7362db4286efb
CRs-Fixed: 681267
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Fix compilation error
Change-Id: I91b539f7cd73c536d704402c0efd69ce3b30f84c
CRs-Fixed: 681547
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Channel 14 was implemented as skipping twice which should be just once.
Made the change so that it is done only once, so as to no addtional
channel is skipped accidentally.
Change-Id: I14ab93c5a11f7045f84678c383fc506406ff7bf9
CRs-Fixed: 678952
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Must check return value from sscanf to confirm that all variables
were correctly assigned before subsequent referencing.
Change-Id: Ifbe3a18aa60fbe2c83d40e7f0ecb4800548c439a
CRs-fixed: 678384
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Currently when Scan is ongoing and Deauth is sent by AP to STA,
LIM post the DEAUTH IND message to CSR.Because of scan, FW keeps
on sending the NULL frame and AP sends the DEAUTH frame in
response to this NULL frame.This results in flood of DEAUTH frame
and PE message queue always have this DEAUTH frames because of which
CSR is not able to get chance to process the DEAUTH indication.
This eventually results into crash.
So now the deauth frames RX are limited to 20 so that CSR can
process the DEAUTH IND.
Change-Id: I7e67be11daaaf7b10a2f6e4f0b75420f3721e641
CRs-Fixed: 675950
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
On enabling CONFIG_ATH_PCIE_ACCESS_DEBUG flag, Driver keeps the
record of register read/write access information like register address
and value.
Apart from fixing the compilation errors this fix provides the
dump_pcie_log iwpriv command to dump the PCI register access records.
Change-Id: I97fa1b41fcfc81db1c9af55fc3ac6947a6426189
CRs-Fixed: 673865
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
This commit updates WMI command for setting the DTIM policy.
Change-Id: I1998bd94df15c1765e6522136aea59d7a19b7e97
CRs-Fixed: 681421
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Initialization of tx_flow_control_timer with change interface is added.
Change-Id: I74c72017bb7c5ca443741a214c457a38e6903ca9
CRs-FIXED: 672878
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
This change improves the selection of the primary channel in HT80
mode for ACS. In HT80 mode 4 channels are bonded, compare the
sum of upper two channels with the lower two channels, and then
compare the two channels within the selected channel pair to
choose the best primary channel.
Change-Id: I1520a4a5501d7b166c351d7423630874741af852
CRs-fixed: 675584
|
| | | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Provide fix for the kernel panic observed while adding virtual
interface. The panic is a result of caller performing invalid
memory region access due to the handler returning NULL instead
of ERR_PTR.
Change-Id: If19ca4b1c058cda673bfdbc3301f49c16605e1c9
CRs-Fixed: 679629
|
