Propagation from qcacld-2.0 to qcacld-3.0
In random testing cases, the SAP close may be called even before the BSS
is started. In such cases the SAP callback may not have been registered
and such scenarios may lead to unnecessary assertion. Just returning an
errror should do.
Remove the assert but retain the error notification.
Change-Id: Ief9ea45d2d7f3d910766e73a9e0dca5e34c85905
CRs-Fixed: 2121720
Propagation from qcacld-2.0 to qcacld-3.0
During BSS frame update, frame_len is calculated as size of ieee80211_mgmt
and ielen. Since ieee80211_mgmt is a generic frame structure and different
frame structures are defined under union this may exceed the actual frame
len.
Fix by calculatiing offset of variable(ies) and ies length.
Change-Id: Ied8e4e604e41de1ac5ccc047ef5cc3cdb05a9445
CRs-Fixed: 2121711
Update ini param g_auto_detect_power_failure_mode to incorporate below
values,
0 - Don't register wow wakeup event and FW crashes on power failure
1 - Register wow wakeup event and FW sends failure event to host on
power failure
2 - Don't register wow wakeup event and FW silently rejuvenate on
power failure
3 - Don't register wow wakeup event and the auto power failure detect
feature is disabled in FW.
Change-Id: I8a704954ecbacadbc035c1523fa41a18b6300f66
CRs-Fixed: 2087144
Current driver has roam id uninitialized in anticipation that roam id
will be filled by SME APIs to correct value but in error conditions
that value may not be filled at all. In those kind of cases
initializing to invalid value will help to avoid any security breach.
CRs-Fixed: 2119198
Change-Id: I96e55cb91ef76df63dd6ba267130e1092fdcf899
Make sure num_vdev_mac_entries which is coming from firmware within
MAX_VDEV_SUPPORTED to avoid any buffer overflow or OOB read.
Change-Id: I92793a6bcfd46b288c3f496a6f6cc9b372f60c48
CRs-Fixed: 2119432
Current driver forming a unit test command within HDD layer and uses
message passing method to deliver it to WMA layer which
requires HDD layer to have a knowledge of how to form a unit test command.
User SME API to send params to WMA layer and let WMA form a unit test
command and send it directly to FW through WMI layer instead of HDD
forming and passing it down.
CRs-Fixed: 2118725
Change-Id: Id1838939813e6cd2d52cee8720a1f4e0ca34329b
Checkpatch has detected multiple instances of "line over 80
characters" so fix them.
Change-Id: I240fcea5af2cfb4f3c1b660784fc7d37bc948c3c
CRs-Fixed: 2120625
Checkpatch has detected multiple instances of "line over 80
characters" so fix them.
Change-Id: If7b0e259a4399cc829fa4d0186c4699ae7980706
CRs-Fixed: 2120617
Current doesn't perform boundary check on num_vdev_mac_entries param
which coming from firmware. Without boundary check, driver may be
exposed to buffer overflow.
Check against the boundary limit before using it.
CRs-Fixed: 2119430
Change-Id: I502926a7f783acc7b73a3fbbbd70386a099b48b3
Currently if_id used in the for loop is incremented based on vdev_map != 0
and vdev_map is a uint_32, received from FW, and is right shifted by one bit
for each iteration. This could result in if_id going upto max of 31 and cause
OOB read.
Add sanity check to make sure if_id is less than max_bssid.
Change-Id: I7e0c4e9a26cb67f41e35c60c2756d7ad02cf43ea
CRs-Fixed: 2119443
Add sanity check for vdev id in wma_roam_event_callback() to prevent
out of bound access of memory in wma_roam_better_ap_handler().
Change-Id: If3cf06a8eca767201fdd8b056bee6d773938a2a6
CRs-Fixed: 2119400
In case of FILS connection, form 802.3 packet out of
HLP IE info parsed from assoc response(if present) and
send the same to IP layer.
Change-Id: I0e077ee48030da84bfe9f987722f96ac2e05ae36
CRs-Fixed: 2034452
This change adds support to append/store the HLP IE(if present)
in req_ie as part of connect request from Supplicant and adds the
same in Association request for FILS connection.
Change-Id: I8cd3e28b462a8ac2bd9eee2a383f9d0886adfa83
CRs-Fixed: 2034452
Add new kek_len param used for FILS in the pmo_gtk_req structure under
PMO GTK public structures
Change-Id: I4fc25026d9fc004f267b9461d1671f8b1ac57f53
CRs-Fixed: 2081252
Add support in wma_flush_complete_evt_handler to capture
data stall event from Firmware and post the message to
sys queue.
CRs-Fixed: 2086176
Change-Id: I4e819b1ae711b3867fa46ff638d4bfd2054519ed
Add support to post data stall event to sys message queue
to handle it in data detection module and finally post
diag event to QXDM. Add support to send NUD failure
diag event.
CRs-Fixed: 2086176
Change-Id: I72ba36d4c2f6ef2eb495ad1586f74af0f3c69254
Add data stall detection module in hdd which handles data
stall reported by host or fw and take necessary recovery
steps based on the data stall type.
Change-Id: Idf6c43f55d1bc115a0c06b4c6ef766ff3ed09bee
CRs-Fixed: 2090643
Host should keep the wake lock from the time it sends
WMI_PDEV_SET_MAC_CONFIG_CMDID to FW till it receives the
WMI_PDEV_SET_MAC_CONFIG_RESP_EVENTID. This will avoid any fatal
crash condition.
Change-Id: Id16a1957b38acee6cf45c123ea9dbab25aae9b39
CRs-Fixed: 2070779
If Probe resp is received just after the Probe after
HB timer expires then it results in disconnection.
Fix is to increase the probe after hb timer default
value to 70msecs.
Change-Id: I0d4ac3e567def348c6b03b41c54579f332099de8
CRs-Fixed: 2111579
wlan_hdd_cfg80211_get_fw_mem_dump() currently returns ENOTSUPP
if the feature is not supported by firmware. Return EOPNOTSUPP
instead of ENOTSUPP, which looks more appropriate.
Change-Id: I1ca2ebd3fa572ba3caae29ef9f1e396693378fb2
CRs-Fixed: 2088667
Add Vendor Event to get the driver hang reason indicating to the
user space that the driver has detected an internal failure.
This event carries the information indicating the reason that triggered
this detection.
Change-Id: I3934f2a18c796ed3b53175dcbe7efd7f4d1409b9
CRs-fixed: 2098498
The Linux Coding Style frowns upon mixed-case names and
so-called Hungarian notation, so rename pAdapter to align
with the Coding Style.
Change-Id: Id6ab1b413a3eec66c88dce52ce14dd3693d15a34
CRs-Fixed: 2119860
The Linux Coding Style frowns upon mixed-case names and
so-called Hungarian notation, so rename pAdapter to align
with the Coding Style.
Change-Id: I29e0573d41628692838037c79007ef983a0cbf9f
CRs-Fixed: 2119855
The Linux Coding Style frowns upon mixed-case names and
so-called Hungarian notation, so rename pAdapter to align
with the Coding Style.
Change-Id: I3d3c52f80a402231dd2a742cf019bf5813791fe7
CRs-Fixed: 2119854
The Linux Coding Style frowns upon mixed-case names and
so-called Hungarian notation, so rename pAdapter to align
with the Coding Style.
Change-Id: I846cf0b6effcfacce22ce122bf73064a230a05c1
CRs-Fixed: 2119853
The Linux Coding Style frowns upon mixed-case names and
so-called Hungarian notation, so rename pAdapter to align
with the Coding Style.
Change-Id: Iee939c8df097967143b77cbffe46fba6e69cd52a
CRs-Fixed: 2119849
Check the length of the IE's before appending them
and storing them in the session in
lim_process_update_add_ies.
Change-Id: I70d26638a58998c82a8810d7c2181d1f24c56e19
CRs-Fixed: 2119729
Currently the size of array ch_list in sme_set_plm_request is
defined as WNI_CFG_VALID_CHANNEL_LIST and this is incorrect.
This is just an index to the corresponding CFG item. Fix the
size to WNI_CFG_VALID_CHANNEL_LIST_LEN which is the maximum
size that can be passed from the source buffer.
Change-Id: I90086f2c73ee09cfc9d63a327b464f4017f5b37f
CRs-Fixed: 2120725
Currently, mem_dump feature support is advertized just based
on the compile macro, if the firmware does not support this
feature, still code ends up sending true to the user space.
Fix this by properly advertising the mem_dump feature support
based on the the FW support.
Change-Id: I4d601f764c6598c51c48d43e24e15fb6c35875bd
CRs-Fixed: 2077931
After deriving the vdev_id from the vdev map in
wma_beacon_swba_handler check for the validity
of the vdev_id
Change-Id: Ifc4577d8a00f447e2bcfa4e01fce5ac2dbe96a4d
CRs-Fixed: 2120751
The Linux Coding Style frowns upon mixed-case names and
so-called Hungarian notation, so rename pAdapter to align
with the Coding Style.
Change-Id: I262e70dfaf24e11c61c848b8d53e39f9be838baa
CRs-Fixed: 2119844
