hif_ext_group and soc->intr_ctx[i].lro_ctx are allocated
but not free while fail to register parts of ext_group in
hif_register_ext_group, which cause memory leak. Fix is
to detach according resource before return.
Change-Id: I03ecc37437d09bf9275f128b1f2b320124a7df64
CRs-Fixed: 2988891
-Add flag to inidicate special vap configuration
-Add frame type counters while processing tlv
WIFIRX_MPDU_START_E
-Add function to update special vap rx stats
-Add dp pdev param to enable/disable special vap
stats reset
-Add function to reset special vap stats
-Add fucntion to get special vap stats
-Add CDP function to retreive special vap stats
Change-Id: Ia5de6743e472dc86c9e66b9e789c909a57025e35
CRs-Fixed: 3005425
The following configurations are changed
- Change numer of WBM2SWRELEASE rings from 7 to 8
- Use configurable RBM value when enqueuing packets for TX. This is needed
since WBM release ring numbers do not have an easy mapping to RBM values
for HMT1.0.
Change-Id: Idcf9e48e00b7039331fc1837bb1e900b12f19eb3
CRs-Fixed: 2984362
In dp_service_srngs, the current logic assumes that WBM2SWRELEASE ring
number as obtained from interrupt_ctx->tx_mask matches the index of
soc->tx_comp[] array. However this may not be true, esp for HMT.
Add logic to fix the same.
Use a separate macro to enable use of single TX ring.
Change-Id: I1bee27b800ad4e4ab1a1fe5e2b01b5b43acfe1f7
CRs-Fixed: 2984362
PATCH[7/7]:
This patch consists following changes:
-Conditionally compile all monitor destination ring related code
Macro used QCA_MONITOR_PKT_SUPPORT
-Add QCA_ENHANCED_STATS_SUPPORT macro to conditionally compile
enhanced stats support
-Use QCA_MCOPY_SUPPORT and QCA_TX_CAPTURE_SUPPORT macros
to conditionally compile MCOPY and Tx capture features
respectively
-Use QCN_IE macro to conditionally compile BPR
feature
-Use QCA_ADVANCE_MON_FILTER_SUPPORT macro to conditionally
compile advance monitor filter feature
-Fix vdev attach issue for special and smart monitor vap
-Fix status ring initialization issue.
Change-Id: I0deaa28a9a54bf34b0f41e6dd510fdd8d4992db2
CRs-Fixed: 2983780
PATCH[6/7]:
This patch consists following changes:
-Move monitor related pdev variables to struct dp_mon_pdev
-Move monitor relted soc variables to struct dp_mon_soc
-Move cookie to monitor link desc va to monitor file
-Add APIs to access monitor related variables
-Add dummy APIs to build without monitor support.
Change-Id: I032a480b1383d061f984cee3ba6b10ac7a0ff350
CRs-Fixed: 2983781
PATCH[5/7]:
This patch consists following changes:
-Monitor vdev context allocation while vdev attach and free
while vdev delete
-Monitor peer context allocation while peer create and free while
peer delete
-Move monitor vdev timer function to monitor file
-Move monitor reap timer handler function to monitor file
-Move monitor timer related variables to monitor file
-Add timer init/deinit and start/stop in monitor file.
Change-Id: I9c7910671d3678c53ca9ec44a57bc10e892008d9
CRs-Fixed: 2983994
PATCH[3/7]:
This patch consists following changes:
-Move struct cdp_mon_ops to dp monitor file
-Add new function for dp monitor soc cfg initialization
-Move full monitor mode function to dp monitor file
-Move vdev set monitor mode function to dp monitor file
-Move config debug sniffer function to dp monitor file
-Move set advance monitor filter function to dp monitor file
-Set monitor delayed replenish entries for target types
Change-Id: If8dac4eedc0c8eb4e956df34c7c3adae8712c840
CRs-Fixed: 2983939
IPA and rx_err ring are processed in same CPU resulting
in low Tput.
To fix this, not allow processing of rx_err ring in
CPU 0
Change-Id: Id53a03c9290607beb1a595c84bfb0fd8d9f5d105
CRs-Fixed: 2949569
When vdev deleted and interface down, there are pending pkts in nbuf
queue of dp rx thread in cornel cases, when flush cmd comes, even vdev
is not found by vdev id, rx stack callback isn't set to NULL, pkts are
still delivered to kernel stack, but rx queue is invalid now, assert
will happen when GRO flush happens.
Fix by setting rx stack callback as NULL when vdev isn't found by vdev
id.
Change-Id: Iad3b71dbc86563a4f8b5369474b2a901cdf8b325
CRs-Fixed: 2995694
Mark vdev as delete in process after flush out the packets in rx thread
queue currently.
If rx packet received after flush and before mark vdev as delete in
process, the packet is enqueued into rx thread queue and not flushed after
vdev deleted physically and net interface down. After that, if flush
pkts after interface down, assert will happen.
Fix:
Mark vdev as delete in process before flush out the packets in the rx
thread queues.
If the vdev is marked as delete in progress, then the
packets received on that particular vdev should not be
enqueued to dp rx thread any more.
Change-Id: If651a9e97b8501efc26400079194570ba6407f13
CRs-Fixed: 2996534
Add support to calculate uplink delay on a per vdev basis.
The delay calculated excludes bus delays, where packet
enqueueing time and complete time are fetched from HW release
ring.
Change-Id: I4bbcbf28b197f7a87cb822a841f1f957ebcf063b
CRs-Fixed: 2958833
In MISSION mode, MSI address and data are still
configured for monitor status rings even though MCC
uses only polling mode for this ring.
Fix is to skip MSI configuration for monitor sts
ring in MISSION mode for MCC.
Change-Id: If295a5afc6731bbec1ad7182590e0015f3a911ba
CRs-Fixed: 2992167
Define a new cdp interface structure and API for soc stats for use
by upper layers to fetch dp soc stats.
Change-Id: I314e952546551003da2ca7f69b2e1aa23c126e66
This change will receive SCS procedure based parameters from
the control path and the peer will start building it's database
for traffic classification using SCS procedures.
Change-Id: I4fd7db595e0645285a95d0561eb46343238fa337
CRs-Fixed: 2970408
Currently the EAPOL/ARP/DHCP frames arriving as 2K-jump
or out-of-order frmaes are being delivered to the network
stack, without checking for the packet-number sequence.
WCN7850 has hardware support to provide the packet number
of the previous successful re-ordered packet from hardware.
Use this feature to check if the packet-number are in proper
sequence for these EAPOL/ARP/DHCP packets arriving as 2k-jump
or out-of-order packets before submitting it to the network
stack.
Change-Id: I1078452afce4bc00b2509436295e5bd80000feb4
CRs-Fixed: 2965086
Add the handlers to process the near-full irqs for
the rx, tx_completion, wbm_error and reo_status rings.
Change-Id: Ia5a2abb6d66a96e8dcb5c651d24769382db0d666
CRs-Fixed: 2965081
WCN7850 has support for near full indication for
the consumer srngs. This interrupt is used to take
preventive actions to avoid ring full watchdog irq
trigger.
Register for the near full irq and add the necessary
ext groups for these near-full irqs.
Change-Id: Ic16381fceabc54e6c52b34dd13abea74cad4d38c
CRs-Fixed: 2965081
Smart monitor mode requires buffers on monitor status ring only.
It does not require any buffers on monitor buffer ring.
As an optimization, initialization of descriptor pool and allocation
of buffers for monitor buffer ring is not required. Also filter
settings for monitor buffer ring are not required in smart monitor
mode.
Change-Id: If328b08184865e5472425b5520dd0a364a478c40
CRs-Fixed: 2965717
In case of aging out wds entries, avoid sending all
the wds delete wmi commands in a huge single burst.
Instead limit the maximum commands to half of the
wmi queue size. Inactivity will continue to be
marked every 120 secs but the inactive entries
will be flushed every 5 secs if there are pending
entries.
Change-Id: I6735704a0750ef466f1df798f0b354f4382098d6
CRs-Fixed: 2952682
Currently IPA tx buffers are fixed at 1007 buffers.
Adding INI support to change number of TX buffers at load time.
Change-Id: Ide0357554f34759198276b9699e0e6f4a3e20051
CRs-Fixed: 2957916
Implement core DP rx processing to functions in to corresponding
architecture specific be/li rx files. Keep common utility functions
in DP common files.
Change-Id: I40083e10772fd2b6ce2f1fa9e197f2ad92d0522a
CRs-Fixed: 2891021
Move DP TX target specific functionality to dp/wifi3.0/be
and dp/wifi3.0/li folders. DP Functionality common to both lithium and
beryllium targets stays in dp/wifi3.0.
Change-Id: I3497284153e2ea30a9cb1faf05bd41422329b804
CRs-Fixed: 2891038
Add changes in datapath rx monitor mode APIs
to handle modification in HAL APIs for WCN7850.
Change-Id: Iec7fbf9389033dfa86e71b1081e6a64607125ecc
CRs-Fixed: 2888556
Memory allocated for tx_hw_desc_history uses incorrect
size parameter resulting in much lower memory to get
assigned. This will result in OOB access and corruptions
in memory, regions post the trailing boundary when updating
tx hw desc events via dp_tx_hw_desc_update_evt.
Fix is to use the appropriate memory size for tx_hw_desc_history
and add NULL check in dp_tx_hw_desc_update_evt.
Change-Id: I97af7898cf8bf1b24978d559f84a2a3d00227ed8
CRs-Fixed: 2952859
As per current design, in failure path of hdd_wlan_start_modules,
mem_free of hif_ext_group is done (in cds_dp_close) before
free_irq (in hdd_hif_close), during next hdd_wlan_start_modules,
request_irq adds new handler entry to the list in irq_desc, this
leads to a crash on accessing older stale entry from irq handler,
so adding a bus_ops hif_grp_irq_deconfigure to free ext grp IRQs.
Change-Id: I4d0a2bee1fabee388cea8a85226fae641165a8d5
CRs-Fixed: 2949400
There is a timing race condition between RTPM suspend flow and
DP TX flow. When TX is queued during RTPM suspend flow. TX
ring update may be delayed. Add a force TX HP flush when RTPM
is rejected due to TX pending frame. This can help to improve
TX pending frames delay when race condition happens.
Change-Id: I6f60f2902dfda630f81528dcf978da6644d18ba7
CRs-Fixed: 2942744
Provide multiple combinations to configure the msi interrupts
of DP and CE based on the number of MSIs available in the platform.
Number of MSIs used for CE and DP can be changed by modifying the
MSI assignment table in platform driver. Best possible mask for that
MSI is automatically chosen based on predetermined settings.
Change-Id: I02b44fb033631d69d97f2d8d2d3f698541d37aad
In some RX backpressure cases, we see the HW accessing REO
queue descriptors of a deleted peer(after the queue descriptors
are unmapped/freed), this is leading to SMMU faults. There are
cases where the HW is accessing the stale REO queue descriptors
after ~12seconds after the queue descriptors were freed.
In order to avoid the problem, HW team has suggested to defer
unmapping/free of REO Queue descriptors. Add the logic for the
same.
Change-Id: I5b1fb966dc75b963ccc9d22c40272c8d1d8d6026
CRs-Fixed: 2939223
Register dp_peer_flush_frags API in dp peer ops
for flushing fragments for a particular peer.
Change-Id: Ia179d3160bdc306ec965c465134042c66a0c40a6
CRs-Fixed: 2874366
Add the history support to log Tx descriptors programmed
in Tx and completion HW rings.
Change-Id: I60954c93e2595e7dad1251c459eed8afc761e917
CRs-Fixed: 2924614
