* Initialize stats in dp_txrx_peer and dp_monitor_peer as per
the stats re-organization.
* Update peer calibrated stats on every timer expiry of 1 second.
* Define APIs to reset monitor_peer stats and to fetch rdkstats_ctx
from monitor peer.
* Place check to make sure that monitor peer related APIs are
are called only for legacy peer or MLO Link peer.
* Update peer based pktlog flag, tx and rx capture feature flags
in monitor_peer.
Change-Id: Idadbe19e85d4a9fdf77fb6dabe0c8a4952b024b8
CRs-Fixed: 3092123
Segregate the stats among dp_peer, dp_txrx_peer and dp_mon_peer
structures.
Segregation is done as per below:
* Move stats updated in per pkt path to dp_txrx_peer
* Consider stats updated purely in per pkt path as per_pkt_stats
* Consider stats for which there is an alternate path of
updation as extd_stats
* Move delay stats and jitter stats to dp_txrx_peer
* Maintain calibrated stats in dp_peer
* Move rdkstats_ctx from dp_peer to dp_mon_peer
* Move invalid peer from dp_pdev to dp_mon_pdev
As per the above segregation, if there is not support for
alternate path of stats updation, then stats in dp_txrx_peer is
combination of per_pkt_stats and extd_stats and dp_mon_peer
does not hold any stats.
If support is there, then dp_txrx_peer holds only per_pkt_stats and
dp_mon_peer holds extd_stats.
Change-Id: I2eb81d1b93a178985c05ff4f6557a173cc8a262d
CRs-Fixed: 3092123
Move the stats parameters from the dp_peer which are used
in per packet path to txrx_peer
Change-Id: Ieb68b6950740791b37bbd2bfdc4815c3d7bc4834
CRs-Fixed: 3095637
Move the parameters from the dp_peer which are used
in per packet path to txrx_peer and params related
to monitor path to monitor peer.
Initialize txrx_peer and monitor peers.
Change-Id: I0449c67c1cd47deb76bc89b5ddc64174f6694eb0
CRs-Fixed: 3095637
Check physical address of nbuf with ring addr. If they
don't match, move to the next descriptor. Activated with
build flag.
Change-Id: Ibd6b38679d677a59f14450db3224104396c88d2e
In NAWDS, if ast for self bridge is present in AST table,
then the mcast arp frames are getting dropped if peerid
for nawds peer is matching ast peerid of self bridge.
To fix this, add intra-bss check for NAWDS in tx path.
Change-Id: I6f1b44634536c558520aaf0bd0c0155c4b2a4e28
In order to support flow overide feature,
AST table has to be split between RxPCU and DDR.
With this split, RX monitor cannot make use of
ast_index to fetch peer as it is not from DDR.
So make use of sw_peer_id to fetch peer.
This sw_peer_id is derived from RX_MPDU_STAT_START_TLV
Remove the spinlock on unsused variable i.e ast->lock.
These changes were overwritten by monitor segregation
effort, hence bringing back again.
Change-Id: I339f4296562e3b4dfd6f3a71102385e395f00268
CRs-Fixed: 3108200
1. Replenish path issue:
a. when number of required buffers passed to replenish API is '1',
host does not replenish '1' buffer.
b. If no. of available entries in ring are '0', return sw desc
back to pool.
2. 32-bit compilation issue
3. Block ppdu desc deliver API when pktlog is enabled for BE
Change-Id: I79ebf1f84830b4c2f5e5b727e3e08c5ee1870898
For the non standard htt tx completion, extract the tx completion
status based on the CDP arch type.
For Lithium based architecture, use HTT_TX_WBM_COMPLETION_V2
For Beryllium based architecture, use HTT_TX_WBM_COMPLETION_V3
Change-Id: I050e8a144d72c2a9fcb6174c690c38c4e2c88c25
CRs-Fixed: 3119153
Change Ext stats for msdu_missed to reflect reo error.
Update EXT stats transmit enqueue failure value to reflect
enqueue errors on all transmit rings.
Change-Id: I532b2b0adaa3216809978e3d558a3b11e5bc8eb6
CRs-Fixed: 3115517
Add support to update tsf timestamp on driver entry and
exit in data packet. This helps debug latency issue in
XR usecases
Change-Id: I9c966c1b8cb09dc5eab6104fdad36c19a1d68045
CRs-Fixed: 3090108
Reduce the log level of tx flow control threshold pool size
from error to info. This is done to reduce the delay due to
console logging in connection/roaming path.
Change-Id: I5332bd179edc92006976a6421121cfce38b89b70
CRs-Fixed: 3119430
Race condition is observed when runtime suspend is aborted
due to pending tx completions, in this context all tx desc
are checked for corruption or leakage. In parallel tx completion
is received and descriptor timestamp is set to null. As a
result, while checking for leakage in runtime suspend context,
time difference for the tx completion is current jiffies which
is huge, resulting in triggering recovery.
To fix this, tx_desc timestamp is checked before calculating
time latency. Also, printing tx descriptor id for which
completion is not received.
Change-Id: I38b487a47170af374f43c44a96e2a0753f5d57ef
CRs-Fixed: 3117549
EAPOL packets on MLO are destined to MLD mac address of the VDEV. So
we have to accept these packets as well, in case of MLO
Change-Id: If424d145341073806ba313f415e88a738360214d
CRs-Fixed: 3121164
simple Alloc is being used in RX path which avoids
certain debug logic. during free of nbuf we should
avoid this debug logic else it will report it as
double free, this will be triggered only when debug
is enabled
Change-Id: Iadb40071fb733cc4de3291784df5075d5a099a8e
In WIN BE chipsets, replace the REO tid
queue programming in FW via WMI with writing to a
Host managed table shared by HW and SW. REO HW will
pick the tid queue address from the table indexed by
peer id and tid number.
Change-Id: I8107ca5116425538329b11ae3519f02b32573bac
Enable the 4th Tx. completion ring to save CPU load
Initialization and interrupt handling for 4th completion ring
is done here.
Change-Id: I2db27218a3c3e14d719d012f03454a6a7aa647fe
Due to usage of lithium specific structure, the filter settings
were not getting reflected in the htt msg that is being sent to
FW. Usage of BE specific monitor pdev structure is helping out
in configuring the monitor filter settings for various modes.
Change-Id: Ie4a59ec16717d752c6c49624142966722bfacc08
CRs-Fixed: 3118596
Add function pointers to free buffer address stored in status buffer
and support tx monitor in legacy and Waikiki flow.
Change-Id: I28612d388009292ff751fe514183fb801909f485
Introduce a new flag "BE_PKTLOG_SUPPORT" for beryllium specific
pktlog support.
Change-Id: If91156541b322905ad5d97cf2d9d7ae687290ecb
CRs-Fixed: 3115832
Send only eapol packets to stack and drop other packets
when found in rx err path when either REO or RXDMA push
reason is set to ROUTE type.
Change-Id: If1f43426adf21f7d00f17d369cd7fde7f7f85866
CRs-Fixed: 3114312
Add qdf wrapper for dma_clean_range and avoid use of DMA
map/unmap APIs in Tx. simple fast path.
Gain of ~150 Mbps is observed while running 4 radio traffic.
Change-Id: I0e96bfee43fe810da8c861cb0f4535fb0ba7f6f7
1. avoid accessing CFG parameter in per pkt path.
2. Access peer, vdev and pdev only when current peer_id
is different from previous peer_id.
3. Avoid HKv1 and v2 checks in WDS learning path.
Change-Id: Id7fb7a29c642480f8ca58738dbb0c92130e44ecd
DMA unmap for scatter gather packets is done using
only frag0 dma address and the entire packet length
which will lead to incorrect behaviors. Add handling
to do dma unmap for all the fragments of SG packets.
Change-Id: I614553d9f96c573f0095c9be38706fd4ac3223ab
CRs-Fixed: 3112424
Tx descriptor is not being freed and put back in tx desc
pool for a long time. As a result, system is not going
into suspended state.
To find this descriptor, during dp_bus_suspend and
dp_runtime_suspend, parse through the descriptor pool to
check for any descriptors not freed for a long and trigger
self recovery when found.
Change-Id: Id97c5c8537c9bec922f4e254b5bf094505ee61ff
CRs-Fixed: 3109868
use dma_inv_range call instead of dma map and unmap
APIs during replenish and reap. This complete code is
under a compile time macro.
before change: 4160Mbps @ 96.5% (core 3)
after change: 4160Mbps @ 90.5% (core 3)
Change-Id: I61bca349a369ace06cd86f353880108cee013cb1
