The routine 'dfs_translate_radar_params()' translates the radar found
offsets given by a 160MHz detector which ranges from -80MHz to +80MHz
to that of the offsets given by an 80MHz detector that ranges from
-40MHz to +40Mhz. The same function also translates the segment ID from a
single-detector(160MHz) model to a two-detector model(80Mhz). This routine
is unnecessarily called for partial-offload chipsets which do not support
True 160MHz.
Do not call the function 'dfs_translate_radar_params()' from
'dfs_process_radar_ind()' which is common for both partial-offload and
full-offload. Call it from the full-offload specific radar action
function 'tgt_dfs_process_radar_ind()'.
Change-Id: I2c717219d0b0f9263734767bee6070f335032b04
CRs-Fixed: 2718507
Do not age out the scan entries of non-tx bss of connected AP
as the non-tx bss entries are updated based on connected AP
beacon.
Change-Id: Ib7276ff919f4c9a843c018caca1a90fd31168eac
CRs-Fixed: 2714982
New flag is added in PPDU user, this flag gets updated
on processing the user of a PPDU desc
Change-Id: Ib778b08833392bfd5651a83c2c1131f1d1ce3ea8
CRs-Fixed: 2738431
Ratelimit the defrag path error logs and add the stats
for fragmented packets received out-of-order which
inturn leads to sequence number mistamtch in defrag path.
Change-Id: I17d4c1cff214a8c8a05abf576701824b293d2883
CRs-Fixed: 2740805
Currently WLAN_CFG_MAC_PER_TARGET value is two which may cause
dp stats request sent to firmware with unsupported pdev mask.
Update WLAN_CFG_MAC_PER_TARGET value to one for QCA6750 target.
CRS-Fixed: 2738910
Change-Id: I14fdf483bf7d18bb2f8302251ac096885547b43b
Since struct spectral_samp_msg may be modified, so provide API to get
version information, then spectraltool can compatible with different
host based on version information.
Change-Id: If68dbd0d85b28f46056208d849fe2db4faea493a
CRs-Fixed: 2575818
To handle no-inter-band MCC scenarios for dual MAC non-DBS solutions,
add changes to detect capability at run-time based on service bits.
If the service is set then code for only intra-band MCC will be
exercised.
Change-Id: I402311fcecfd03bba67fedfb2b52f779f60b81ce
CRS-Fixed: 2730374
Currently, vendor scan request is enqueued in the scan queue first and
then posted into the scheduler queue. Now if there is an SSR in between,
then scan request will be freed twice. Once in SSR context and again
in the caller context because scan scheduling fails due to user disabled
the scan.
To resolve this, post the scan request in scheduler queue first. Once
scan request scheduling is success, then enqueue the scan request in
scan queue. This approach can lead to a race condition like after
scheduling scan request, host may get scan response from firmware even
before the scan request is queued. To avoid this race, scan request
scheduling and enqueue should be protected by scan_req_q_lock.
Change-Id: Ic26da0181fb9a902a8a4abb8f68d9fc668f1b3f2
CRs-Fixed: 2714495
Currently, per-user information in Tx PPDU completion is allocated for
37 users though FW may send information for fewer number of users. In
new scheme, FW provides number of users in USERS_INFO TLV and host can
allocate the per-user struct based on this number. This can save memory
while processing each completion message.
CRs-Fixed: 2730923
Change-Id: If04fd664a1ef6e77e8eb44c5d2bd1fa89fcaba72
Add 'tx_flag' param to wmi_mgmt_params structure for configuration
to use incorrect key for encryption of mgmt frame.
The 'tx_flag' param is primarily added to provide additional
test configuration options for mgmt frames.
Currently it only has support to encode the mgmt frame with
incorrect key for PMF protected connection.
Change-Id: Id894c735c4c612a75506a33938372c3fa2e7f24a
CRs-fixed: 2690147
A few functions in HIF component returns QDF status value with return
type as non QDF STATUS and vice versa. For such functions, update the
correct return type.
Change-Id: I806407dd0596cd03500242d97ca8220a7beb6c55
CRs-Fixed: 2734818
A few functions in HIF component returns QDF status value with return
type as non QDF STATUS and vice versa. For such functions, update the
correct return type.
Change-Id: Ifc1068d60e62f7405a15e2b4f0738d91243bd6de
CRs-Fixed: 2734818
A few functions in CE files returns QDF status value with return
type as non QDF STATUS. For such functions, update the return type as
QDF_STATUS.
Change-Id: I401c0a1e1c21ac0865cf4da018eb2d2fb6d86900
CRs-Fixed: 2734818
A few functions in HIF component returns QDF status value with return
type as non QDF STATUS. For such functions, update the return type as
QDF_STATUS.
Change-Id: I69644a2206266ffe2c2d105056f9fec452f5d972
CRs-Fixed: 2734818
Handle invalid HW ast index case inside function dp_peer_map_ast().
This change aids upcoming feature where status of HMWDS ast add is
reported to upper layer. This change eliminates extra peer ast list
traversal required for the new feature.
Change-Id: Ic6f45fedb53204c52719686aee9a90ae12695d09
CRs-Fixed: 2728611
If disable pktlog for hasting/genoa, we need to
mark the ce5/ce7 not used separately to save
some memory. But currently, if set target ce
entries to 0, it has wlan fw crash issue,
so reduce it from 32 to 2 first. Once wlan
fw fix the crash issue, and then change to 0
Change-Id: I7ab11f202fe353737a4104e71fc79ee749b80fc8
The Spectral module uses the first vdev of a pdev for scanning.
For MCL platforms all the vdevs are associated with a single pdev
and this may cause a mismatch between the vdev intended by user and
the vdev used by driver. To avoid this MCL platforms pass a reference
to vdev on which Spectral scan is triggered. WIN platforms
can continue to use the old approach of using the first vdev of a
pdev by passing NULL for vdev to vendor command handlers.
CRS-Fixed: 2725777
Change-Id: I1a7278bdcbcc2b706aa292b33c17546190dac082
It starts spectral scan without vdev object, and uses the first vdev
as current vdev. which cause vdev mismatch in some platforms. So pass
vdev object or id when finding current vdev.
Change-Id: Iab55afbe2c98906021a008918f186d3d2dce5871
CRS-Fixed: 2716460
CONFIG_CHAN_NUM_API macro will be disabled. Remove all the references to
any function defined/declared under the macro. The function
reg_get_channel_list_with_power was placed under CHAN_NUM_API by mistake.
This API does not have any CHAN_FREQ_API counterpart. Therefore, move
this function out of CHAN_NUM_API macro.
The function dfs_send_radar_ind_for_freq is redundant as its counterpart
dfs_send_radar_ind has the same arguments and same functionality,
removing the function dfs_send_radar_ind_for_freq. The function
dfs_send_radar_ind is moved out of CHAN_NUM_API macro.
The function wlan_reg_chan_to_freq, wlan_reg_legacy_chan_to_freq is
moved out of CONFIG_CHAN_NUM_API macro. Since it is applicable for legacy
bands.
CRs-Fixed: 2711600
Change-Id: Ib29be638c17ce51f928c865e362ac5b2b8954b42
Skip protocol tag updating from msdu end TLV protocol tag meta data
when there is RXDMA REO entrance push error on monitor destination ring.
The attention and msdu end TLV is not updated reliably when there
is RXDMA REO entrance push error such as FCS error on monitor
destination ring.
Change-Id: I4098c01251b4f4cb2e796d450c156ba8f2b9464d
Add support to get device name and invoke debugfs init
on post attach so that device name is updated at that point.
Change-Id: I0e0a06cae4bef4d6e43c0295f5db99220b185df2
Update reg domain rules to support three new
channels in 5.9 GHz for US.
New channels are 169, 173 and 177 with freq
5845, 5865 and 5885 MHz respectively.
These new channels will be controlled by an
ini fcc_5dot9_ghz_chan_in_master_mode and
a service bit WMI_SERVICE_5_DOT_9GHZ_SUPPORT.
If the service bit is not set we disable the
new channels, and if service bit is set but not
the ini, we make the channels passive. If both
are set, they will be enabled.
Change-Id: I575b02ad387cbda00b015f7075d80c907463bdef
CRs-Fixed: 2683702
Add new channel bonding pairs as well as frequency
pairs for 40, 80 and 160 MHz with the support of
three new 5.9 GHz channels 169, 173 and 177.
Add new API reg_modify_chan_list_for_5dot9_ghz_channels
to enable/disable these new channels based on
INI value and service bit.
Change-Id: I8598e705ba4047e96e7167661ba70d4236cde251
CRs-Fixed: 2696211
Add a count for the number of times we read a stale
cookie value from the REO2SW ring.
Change-Id: I4b20fa93f5261b4ccb9479b7b3469a294703a184
CRs-Fixed: 2736028
When a one-shot capture is triggered with CFR, the state is not
reset to DISABLED after capture has completed. This affects HW mode
switch behavior because if CFR capture is ongoing (ENABLED), mode
switch is not allowed. This change checks for period to return DISABLED
or ENABLED. Period is 0 in case of one-shot capture.
Change-Id: Iae8727959a1becd3b5f4da166e868c1592910b84
CRs-Fixed: 2735166
Move few scoring params to uint8 from uint32 to save memory.
Also move the req and resp related IE to req and resp structures
instead of global structure.
Change-Id: I67f8dd3f381642083fc5f288ebd9055d2919faad
CRs-Fixed: 2737146
In the TBTT info of RNR, one byte is added for power
spectral density. Add logic to get the PSD limit.
Change-Id: I7ce44a762c47d0d0526157fcb41a80ad3ef3c08e
CRs-Fixed: 2731433
On returning the msdu link descriptor to HW via WBM
release ring, the rx exception ring contents are not
zeroed. This could result in host reading stale ring
descriptor content in the scenario when HP is updated
even before the latest values are reflected in the ring
descriptor.
Fix is to set invalid bit in cookie for exception ring
descriptor and add cookie sanity check.
Change-Id: I01a294c92b260ebe8e584ef20e9550b1febec913
CRs-Fixed: 2730759
When an AP operates on a 6G channel. The channel frequency in the
radiotap header, in the beacon frame, shows an incorrect 6G frequency.
This is because, the function hal_rx_radiotap_num_to_freq is not updated
to handle the 6G channelization spec update ( IEEE 802.11 11AX Draft 6.1).
To fix this issue, update the function hal_rx_radiotap_num_to_freq to
handle the 6G channelization spec update (increase the 6G frequencies by
an offset of 10MHz).
Change-Id: I342dcb46b2627b3a2c3f12524077e63d811a5feb
CRs-Fixed: 2733367
For MCC, hash key related member in struct hal_rx_fst is not used,
it requires big chunk memory > 40K bytes that sometimes may fail to
allocate from system, wlan start up failed.
Remove hash key related member in hal_rx_fst by macro
WLAN_SUPPORT_RX_FISA for MCC.
Change-Id: I4214e18155c3ebc3dcc800c8c74f7eed16d580b4
CRs-Fixed: 2732990
In I3c82e2eb98cdaf0ff8754bd990529a1833276304, KW error is reported in
the function, reg_compute_chan_to_freq.
This is because min_chan_range and max_chan_range can be assigned values,
that can cause array out of bounds issue.
To fix the issue, add explicit range checks against min_chan_range and
max_chang_range before iterating through master_chan_list to
avoid out of bound access.
Change-Id: Id5cd032fb899475720080b29012a6de1b5d4a916
CRs-Fixed: 2727082
Make sure to drop the raw Rx frames as both driver and stack
are not expected to handle them.
Add counter for invalid fisa flow_idx packet received.
Change-Id: I5107c554b8ce6a9a7973f2aeca44bb0f360dc2df
CRs-Fixed: 2733981
Currently, host driver processes NAN events only if
wmi_service_nan_dbs_support is advertised by the firmware. But NAN
concurrency is handled in host in non-DBS platforms also when
wmi_service_nan_disable_support is advertised by the firmware.
So, process the events if any of these two are capabilities
are present.
Change-Id: Ib571ef69ea177ddd758f800448939122126339b8
CRs-Fixed: 2735465
Currently all the rx ring descriptor contents are left
intact even after these entries are processed. This can,
at times, lead to stale entries being processed, if the
head pointer of any ring is updated before the updated
contents of the ring descriptor gets reflected in the memory.
This can lead to scenarios where the host driver reads a
stale value of sw_cookie, and free/unmap a currently in-use
buffer, thereby leading to the hardware accessing unmapped
memory region.
The sw_cookie is the integral part of al the rx ring
processing. Hence we always mark the sw_cookie as invalid
after dequeuing an entry from the REO2SW ring. Every time
we check for the validity of the sw_cookie before we try to
process an entry from REO2SW ring. if the invalid bit in the
sw_cookie is set, we just skip this entry and move on to the
next entry in the ring.
Change-Id: I0e78fa662b8ba33e64687a4dee4d1a5875ddb4bf
CRs-Fixed: 2730718
