With current implementation in_use bit for last descriptor is not
getting set as 0. Removing redundant code and setting in_use bit
for last descriptor as 0.
Change-Id: Ia1945fcb814136128d18a8aec473ed1b4786dd9e
While draining the pending reo cmd's during wifi down,
the tid array from the peer structure is used in debug
statement.However,the peer is freed much before
the drain operation.This leads to use-after-free access.
Change-Id: Idf92ccd0fe4eba3eed8a97ac83485de8fccb0f24
Local peer_id is being cleaned up across DP, HDD and PS/WMA.
So, any references to local peer_id/sta_id will be replaced
by peer mac address and all interactions between the layers
will be based on peer mac address.
Cleanup references to peer_local_id from the
following APIs:
1) dp_rx_process
2) dp_rx_process_rxdma_err
Change-Id: Ia1c8bd091759132367718e6c6d5f67cf5a98e507
CRs-Fixed: 2512706
Pass pdev to dp_peer_stats_notify API instead of de-referencing
this from peer structure.
Change-Id: I5dc679283f5822f44b4140aefb3aee44a826c720
CRs-Fixed: 2500647
Adds new INI variable for threshold control.
Host sends message to target with new threshold info.
If buffer reaches threshold then MAC drops incoming monitor frames.
Change-Id: Id659d7df68a5ec0b22fb571dc1ffa4990f8fcc4f
In the sdx platform, WLAN monitor mode is disabed,
in other words, CONFIG_FEATURE_MONITOR_MODE_SUPPORT
will be set to n, but dp_mon_buf_delayed_replenish
hasn't be defineded if it set to n, which can cause the
runtime unknown symbol issue. So add the definition
for it.
Change-Id: Iec50f7cc5ed6f94e2f5a40fbc6594c2830d1359f
Replace dp_srng_mem_alloc() by qdf_aligned_mem_alloc_consistent() and
qdf_aligned_malloc() in dp_srng_setup().
Change-Id: Iae236a3cc19ec7a9f94b04382d37fc95e6ed622f
CRs-Fixed: 2462455
STATUS_DONE is missing sometimes in monitor status ring,
but logging for each entry in the ring increases the time
for which the hal_ring access lock is held. If there are
many entries in the monitor status ring, this can lead to
the hal ring access lock being held for too long and cause
inconsistencies.
Reduce the logging level of the tlv tag STATUS_DONE errors
and have a counter for the cases when this happens.
CRs-Fixed: 2502422
Change-Id: Ica77ce62bbe2c6b7691d638ac44ab0452dd2fd97
mac_id is used to get rx_desc_pool for monitor ring and mac_for_pdev is
used for srng for monitor buf. Use appropriate id to get correct
combination of hw ring and sw desc for monitor rings.
Change-Id: I7e4467e9568056e55cbabfef8f97ce94ad9c1b1c
CRs-Fixed: 2512292
tx_failed is not updated from ppdu completions owing to HW limitation.
Hence update tx_failed from data path
Change-Id: I808223005047af9b474ce423caac27fd9a83677e
In cases where NSS offload is enabled, we currently do not support
flow-based search (flow tagging). Check for NSS offload enabled flag
prior to configuring the flow search table in target/hardware.
CRs-Fixed: 2502311
Change-Id: I291c610fe568ac36bacb3c21fd14feb544332b20
MDNS packets if forwarded for a NAN vdev can lead to potential flooding
of the air interface. Hence donot forward them.
Change-Id: Idfdedfb0b5b553745440587448230013f3b56a7d
CRs-Fixed: 2503360
1. Assign correct first msdu payload
2. Reset mpdu fcs ok bitmap upon reception of next ppdu
3. Free rx_ppdu_buf_q in error cases
Change-Id: I4f2e687d51d1e10693adc9cfcdee49190ba6815c
CRs-Fixed: 2502889
Cleanup dp_get_vdev_by_sta_id to be peer mac address based
from local peer id based.
Also, modify the API cdp_peer_get_vdev_by_sta_id with additional
argument for peer mac address.
Rename cdp_peer_get_vdev_by_sta_id to cdp_peer_get_vdev_by_peer_addr.
Change-Id: I889538b5ea7ebac8973dcaaa5ebdad1ac495e791
CRs-Fixed: 2504569
Process smart antenna fields from tlv
HTT_PPDU_STATS_USR_COMPLTN_COMMON_TLV in tx
completion path and export it to tx_completion.
Change-Id: If14e711b0aee8e583fd329f3c3915904649ffe47
CRs-Fixed: 2491429
Since tx_monitor mode and rx_monitor mode consume signficant MIPS
when enabled on a per-peer basis, new requirement is to enable such
functionality on a per-peer basis. New CDP APIs are added to set these
per-peer parameters.
CRs-Fixed: 2494471
Change-Id: Ib44845ec98481cc4df36c30f47335522e5274568
In dp_rx_process there are two cases in which vdev will be
null. One is when peer is invalid case which happens when
packets comes on reo with out any valid peer and vdev set.
Second case is fetching more data in case of napi no yield
condition and there are no buffers to fetch. In current
dp_rx_process gro flush is called at the end of nbuff process
and with out checking for vdev sanity it is referenced. So
add vdev null sanity check to prevent NULL pointer dereference.
Change-Id: Ie2d480108118d9b83373a450aecabee57675c41d
CRs-Fixed: 2507067
In case of TCP packets being processed by dp_rx_process, send out GRO
flush indication to the thread.
CRs-Fixed: 2500152
Change-Id: I4f464456d423e4680955992c0acf0ed5f4e618b8
Cleanup dp_clear_peer to be peer mac address based
from local peer id based.
Also, modify the API cdp_clear_peer with addition argument
for peer mac address.
Change-Id: I1bd3fab1b9510b5f1d4476e11b617fcb6417f2c4
CRs-Fixed: 2503774
rxhost ring backpressure:
identifying rings causing rx backpressure after being notified
by FW message. Adding logs to be able to see at what state
the ap was after a backpressure event was triggered.
Adding radio stats (261) as well as napi stats for better
state description.
Change-Id: I395450be6faaf959f91729516a7b229c5b3396ce
Add device ID change and target type checks for pine.
Also remove memory war added for Hk emulation.
Change-Id: Idf531a48a03202d4fb241a92a1d671ee2b94cfbd
CRs-fixed: 2453899
In case of peer reuse case rx_tid locks are destroyed as
part of peer cleanup and are not initialized again. So while
deleting this peer as part of unref delete peer cleanup is done
which again tries to destroy the rx_tid locks which were already
destroyed. This leads to assert complaining destroying lock with
out initialization. So dont destroy the rx tid locks in case of
peer reference reuse.
Change-Id: I860de558950009cae6f62385dd665badfe9964e0
CRs-Fixed: 2504973
Read extra bits of fcs okay bit map from
RX_PPDU_END_USER_STATS_EXT TLV and use the same
to send first fcs okay packet in case of M COPY
CRs-Fixed: 2499150
Change-Id: I1a25971ea789dd7fddddb312af8a18a4cccdc178
Cleanup dp_register_peer to be peer mac address based
from local peer id based.
Change-Id: Ifc0166b8ace5aabd7c287941476df47c806e8663
CRs-Fixed: 2503680
Map mac_for_pdev appropriately for both MCL & WIN,
since MCL have multiple monitor buffer rings and
WIN just has one mon buffer ring, below API helps
identify accurate monitor buffer ring for both cases
CRs-Fixed: 2502051
Change-Id: I3929f3f7717d795850a41f4148397d87b0415118
This feature enables user to change HW mode dynamically
from DBS to DBS_SBS mode and vice-versa. Currently, HW
mode configuration is only possible through INI setting
requiring a subsequent reboot.
Relevant DP changes are:
1. Add API cdp_txrx_handle_pdev_status_change to pass
pdev 'up' or 'down' status to DP module
2. Add pdev-status check in dp_rx_process_invalid_peer
3. Add pdev-status check in dp_tx_comp_handler to free
buffer and release descriptor
Change-Id: I74b144abb1b0dc41a26a18ad28f872e6457e9653
CRs-fixed: 2490212
Tags are programmed using wlanconfig commands. Rx IPv4/v6
TCP/UDP packets matching a 5-tuple are tagged using HawkeyeV2 hardware.
Tags are populated in the skb->cb in the REO/exception/monitor data
path and sent to upper stack
CRs-Fixed: 2502311
Change-Id: I7c999e75fab43b6ecb6f9d9fd4b0351f0b9cfda8
When the scheduler thread initiates the WMA_SET_BSSKEY_REQ
we send CMD_UPDATE_RX_REO_QUEUE to REO srng. This is done by
posting a descriptor to the reo command ring and then we
update the HP so that the HW can consume the descriptor.
Avoid accessing HP shadow address when we are in runtime
suspend state. Perform a hif_pm_runtime_get to resume the
link and access the shadow register and once done initiate a
hif_pm_runtime_put to allow device to go into runtime
suspend.
Change-Id: I24c3e046a5769f03a0f1969360cccdbe55b81d45
CRs-Fixed: 2495720
Do not send the WMI command WMI_PEER_REORDER_QUEUE_SETUP_CMDID
to target if roaming is in progress, as there is a discrepancy
in peer information between the host and firmware when roaming
is in progress
Change-Id: I8e91635b56257bb29972898a29c454123f7cdb8e
CRs-Fixed: 2490238
HW access to MSDU link Ext descriptor with NULL address,
whie from host MSDU Ext pool and related ring entry history,
not find NULL address recorded. In case the MSDU Ext descriptor
has been overwritten before issue happened, add DMA address
validity checking in MSDU Ext descriptor filled from TSO
segment fragments and also buffer address in ring entry.
Change-Id: I11dc1d0211530a5f01b515473b2458df15a5f4cc
CRs-Fixed: 2498359
This changes takes care of the below issues.
1. Skip the host allocation of software Tx descriptors
for single band NSS offloaded radio configuration.
2. Reset the reo frag destination to alternate select
for single band offloaded radio configuration.
Change-Id: I3930f266ff9d81d876dda437b29b1ca83d542e12
1. Remove vlan tag in tx and enqueue to hardware.
2. Add vlan tag in rx after peer-vlan_id lookup.
Change-Id: I932202540ac03cabdd20ffd4849fe759ea8a7abb
Limit the total number of tx descriptors allocated using tx_device_limit
ini configurable parameter.
Change-Id: I00b4aa089b235e45bc6c445c57ce32b977d872c3
Add tx device limit configurable from INI.
This is used to limit the overall tx descriptors used in the device
Change-Id: I51e9611ca14bdc146de914317b38b98329130c9f
