A few instances of violating "Block comments should align the * on
each line" have been identified in target_if/dfs, so fix them.
Change-Id: Ic198912225b8bec587e30d00d7a2a5f76e03d366
CRs-Fixed: 2359528
The libc implementations of QDF were originally intended to be used for
off-target unit testing purposes. However, doing this involved
reimplementing large sections of QDF. Instead, unit testing is being
done in an emulator, which provides many of the benefits of off-target
testing without needing to reimplement any logic. As such, remove the
libc folder from QDF.
Change-Id: I829b4cae3f7bad7b24c97ed458e978eab2b0fbbd
CRs-Fixed: 2359499
MCL DP initialization path is running into issues due to change
I5732453f617bdc16995fda916b645c41845c3ecb. Fix the same.
Move flow control lock variables to re-use region. This is needed since
flow control deinit happens in dp_soc_detach, while in dp_soc_deinit, the
soc memory is being partially reset.
De-init HTT memory in dp_soc_deinit and free HTT handle
in dp_soc_detach. The existing code was not freeing HTT
packets for MCL case and freeing the HTT handle twice.
CRs-Fixed: 2359409
Change-Id: I52dcccc0444d4ce139a29ffcb884598bbacda88f
Change made I7fd67b02c3b7cb4f1bfe7c6f4641f6d881e25abd
to fix possible NULL dereference of pkt_stats_dump
caused pkt stats collection to fail as pkt_stats_dump
is NULL from cds_pkt_stats_to_logger_thread.
Move NULL check for pkt_stats_dump to just before
it is dereferenced
Change-Id: I0f4c9d58f0d4d17d6c26bfbbd79f6447dd52230b
CRs-Fixed: 2358139
While handling the WMI_SERVICE_READY_EXT_EVENTID WMI FW event, a NULL
pointer dereference can occur if param_buf->hal_reg_caps is not checked.
Check param_buf->hal_reg_caps before dereferencing it to avoid NULL
pointer dereference.
Change-Id: I00eba5e89fbdde78979d19f492df5ad4dca8b80c
CRs-Fixed: 2347673
Added missed hal_get_hw_hptp_generic
'Get HW head and tail pointer value for any ring' to ipq6018.
Change-Id: Idc4b595ab6c52558542cf6f87e36f590bf93123f
When CONFIG_MOBILE_ROUTER is enabled there are build failures
due to improper featurization of NAN, so fix the featurization.
Change-Id: I6bc11fb82394c2d32b328cb5d50ff974051755e1
CRs-Fixed: 2353170
Initialize the PN to the value which is given by upper layer
and plumb it to HW.
Change-Id: If6b14d2b9b92e4571a85cb20dc445608d5f56961
CRs-Fixed: 2350954
PCIe link up/down voting is critical hence log it using info level
log on console.
Change-Id: Iea2be05cf9ea5231b0e2de54b508d49aec341847
CRs-Fixed: 2357793
Check if ucast pkts count is more than success pkts count,
compute last packet error rate only if ucast is less than success
pkts.
Change-Id: I0aa4d41d09b1691e4d554beab2f2ed7a7f2ceb1e
CRs-Fixed: 2343479
Change max OL_TXRX_NUM_LOCAL_PEER_IDS to 35 to take care of
STA + SAP + P2P concurrency.
Each AP will occupy one ID, so it will occupy 2 IDs for AP-AP mode.
Clients will be assigned max 32 IDs and STA(associated)/
P2P DEV(self-PEER) will get one ID.
Change-Id: I4d3e74c550e34b1d3417825245990655ead87b28
CRs-Fixed: 2356032
Avoid memory fragmentation that happens during
attach-detach flow.
- Reuse transmit allocated static pool memory across soc up/down.
These memories are allocated during soc attach.
- Reuse DP source ring memory, DP soc context, DP pdev context
across soc up/down.
- Reorganise structure members of DP soc and DP pdev so that
we can zero out structure members across soc up/down
- Add cdp soc init/deinit and cdp pdev init/deinit that
will be active across soc up/down
Change-Id: I5732453f617bdc16995fda916b645c41845c3ecb
On MSM platforms, MSI vectors are always wakeable, so calling
enable_irq_wake() on an MSI vector interrupt line is a no-op.
Additionally, on some product lines, this call will fail. This causes
suspend to fail despite the MSI already being wakeable. Simply avoid
marking the wake MSI as wakeable during suspend as a work around.
Change-Id: I32b2b4fb231681ad29b4e98b2410d0eb6e68c671
CRs-Fixed: 2357784
Currently, qdf_nbuf_map_check_for_leaks() logs the leak information
under spinlock before inducing a panic. This can cause a confusing false
positive panic upon releasing the spinlock if the logging took longer
than the spinlock timeout threshold. Since we already know we are going
to panic, panic under spinlock to ensure the panic message is always
appropriate.
Change-Id: Ib98b21e8feac0314b7c1f00f08b9e81112c4f2ba
CRs-Fixed: 2357806
ASE entry aging for MEC entry type was same as
WDS type for 120 secs, this creates an issue when
device roams from MEC to WDS connection by not
aging out till 120secs. So now we reduce the MEC
aging time to 1sec
Change-Id: I6ab3b095a85c6d09f9fcd95b3b39f7152064d297
When x_psoc_create_handle return failed, need free all resources
before return, x_psoc_destroy_handler has no chance to be called
any more.
Change-Id: I63fd3743dd4ef4c9cc35c9ae8464c2b728447818
CRs-Fixed: 2353795
When x_psoc_create_handle return failed, need free all resources
before return, x_psoc_destroy_handler has no chance to be called
any more.
Change-Id: I4b9999dbc2597c4ea860f08a6f3334f7f5728146
CRs-Fixed: 2353795
When wlan_serialization_pdev_create_handle return failed, need
free all resources before return,
wlan_serialization_pdev_destroy_handler has no chance to be
called any more, same for psoc, pdev handlers.
Optimized destroy handlers to avoid double free and leak.
Change-Id: Iae2b7c2fa221502742cbe565b27bcb16b7fc56ed
CRs-Fixed: 2353795
Probe responses received from hidden AP after beacons
with non-zero ssid length will use a new entry in scan
db resulting in multiple scan entries for same AP.
Fix is to remove ssid check in util_is_scan_entry_match
api.
Change-Id: I1c094d483a40d40f7ca9d40b26d04a7f80099dd1
CRs-Fixed: 2353339
The driver has a default scan policy type of DBS scan
which is not overwritten in case of DBS disabled.
Fix is to have a default of NON-DBS scan, and overwrite the
scan policy if the low power, low span scan is triggered,
or if the DBS is disabled from ini.
Change-Id: I8c95b2ca5915e1a74ccdeae55dea5e65bcfe9664
CRs-Fixed: 2354103
objmgr should not call destroy callbacks for components
that have not successfully completed the creating callback.
Or the function wlan_objmgr_xxx_obj_destroy returns with
QDF_STATUS_E_FAILURE and doesn't call wlan_objmgr_xxx_obj_free
to free the psoc/pdev/vdev/peer.
Change-Id: I1ba5a6afd98254cbed4571d5fb1cd24543e1db54
CRs-Fixed: 2318184
A read buffer is allocated and used in
ath_procfs_diag_read and it is not freed
in one exit path which causes a memory leak
Free read buffer in all exit paths of
ath_procfs_diag_read
Change-Id: I1b2446a99f53c2f9130d395dd316eb4649eb24db
CRs-Fixed: 2352165
After gerrit#2436786, vdev in monitor mode is removed from vdev
list of pdev. So it is needed to add one more cdp API to get vdev
in monitor mode.
CRs-Fixed: 2343290
Change-Id: I9d74302a464280881ce0cf25dbc8c4f9d8b8d48d
When reap data from REO2SW ring and if the TP catches up to cached
HP, current logic will update HP again and give a second chance to
reap data. but the TP is not updated to ring register, if more data
is received on this ring then HP easily catches up to TP and the ring
is full.
Update TP as well if HP get chance be updated.
Change-Id: I92f381d2a31f3be987866a829e05b5578098649a
CRs-Fixed: 2334435
The ack_rssi_valid flag to set to 1 only when completion_status
value is 0 which shows that the TX was successful.
Change-Id: I9e520422995e374d7d35b9db0c060d1fce2f9067
In function init_deinit_chainmask_table_alloc, alloc size is declared
as uint32 and is assigned the value the multiplying non zero value
sizeof(struct wlan_psoc_host_chainmask_capabilities) and uint32
value ser_ext_par->chainmask_table[i].num_valid_chainmasks.
This can lead to overflow as multiplies value can be more than uint32.
Change-Id: I9a886d9ee5213ae8989a2c5d4502336cc275418a
CRs-Fixed: 2347650
Change rx defrag related log level from info to debug,
it's not necessary to print normal rx defrag operation related
log default, this can avoid panic caused by excessive logging
when receive lot of fragment data.
Change-Id: Id712d546a760377a6f59b321f73d7ae5ca4af564
CRs-Fixed: 2353869
Introduce a new wmi_send pdev param to enable/disable
"Subchannel Marking" in Firmware (only in Full Offload)
Change-Id: I3cd4f4f13ebca72c4505b6195cc8dc4856d41671
CRs-Fixed: 2334258
Introduce dfs_tgt API to send the WMI command to enable/disable
"Subchannel Marking" in Firmware (only in Full Offload)
Change-Id: If739409521c8704b612e52dfc8ed6b86ffe32bca
CRs-Fixed: 2334258
Changes configured to operate in Channel 100(HT80), if bin5 chirp
radar is injected at center frequency at 5530 MHz, instead of adding
channel 108(HT20), it adds channel 116(HT20) of adjacent 80 MHz band
to NOL list.
We have the following variable assignments in function
dfs_pulses_within_window:
dfs->dfs_min_sidx = pl->pl_elems[*index].p_sidx
dfs->dfs_max_sidx = pl->pl_elems[*index].p_sidx
Since the variable pl->pl_elems[*index].p_sidx is of signed type
and variables dfs->dfs_min_sidx/dfs->dfs_max_sidx are of unsigned type,
if the value of the variable pl->pl_elems[*index].p_sidx is negative,
the values of the variables dfs->dfs_min_sidx/dfs->dfs_max_sidx become
very large unsigned value.
When sidx is converted to frequency offset, the frequency offset
also becomes very large and it is greater than 40Mhz(half of HT80).
In this case, 50MHz is added to center frequency 5530 MHz, which
becomes 5580MHz(channel 116(HT20)). This results in the channel
116(HT20) of adjacent 80 MHz band to be added to NOL list.
Change the type of the variables dfs->dfs_min_sidx/dfs->dfs_max_sidx
from unsigned to signed so that assignments do not alter the actual
value of the frequency offset.
Change-Id: I13995888b618d9dd8a91447d5dffe27aa2e27a27
CRs-Fixed: 2342743
Current HTT_H2T messages from host driver does not have
consistency in message length set by host driver. Some
message types include HTC header length also within the
message length, while other types have message length
itself only, which causes difficulty in handling message
length in FW.
Change-Id: I885a21530a2d8f852387ae54cf7ee0751aad2516
CRs-Fixed: 2345075
Set TDLS channel offset correctly when TDLSSECONDARYCHANNELOFFSET
command is sent with offset 20.
Change-Id: Ifac48da3fea0c83415f3d30caa2919a559f5c3c5
CRs-Fixed: 2341883
Firmware could send same channel list for each pdev. Since SOURCE DRIVER
is default, it would be set for the second event. Avoid that. Remember the
source of the first event itself.
Change-Id: Ib9db58da1fa27673b04a4c62e10f2a05702348ad
CRs-Fixed: 2352151
In target_if_dbr_fill_ring(), min_buf_align (8 bytes) was added to
min_buf_size before allocating the memory for dbr_buf_pool. This leads to
extra 1MB allocation.
To avoid this extra allocation, call qdf_aligned_malloc() to allocate
aligned memory for direct rx buffers.
Change-Id: I72f7c616c7a699437b081de7149a4a67a3ad30c4
Acked-by: Shashikala Prabhu <pshashik@codeaurora.org>
CRs-Fixed : 2336697
Define qdf_aligned_mem_alloc_consistent() and qdf_aligned_malloc() which
allocates the memory and checks if allocated base address is aligned with
ring_base_align. If not, it frees the memory and re-allocates by adding
7 bytes to alloc_size and returns aligned address to the caller.
Change-Id: I412153c20e4e4566450b006356fe78d98f1fd3f0
Acked-by: Shashikala Prabhu <pshashik@codeaurora.org>
CRs-Fixed: 2336697
HKV1 does not have support to maintain per pdev AST table.
Hence workaround is required to ensure that AST entry is deleted before
adding an entry with same MAC. This syncronization needs to done across
pdev. This workaround in required on for HKV1, add changes to enable
workaround only for HKV1 and disabled for other platforms.
Change-Id: I9aa43ea51fdff2f02cdcc8ff7e906c7305446317
CRs-Fixed: 2344921
For HKv2, BA window size can be configured on per tid
basis as opposed to per peer basis on Hkv1. This per
peer per tid configuration on Hkv2 is for Tid 0 to 7.
Change-Id: Ic4afee92792af829a9ad8fdfc211625743b9d052
