currently rx_mon_ring and REO are part of the same
interrupt group, due to which in case of mission
mode when IRQ is received from REO, even the lmac
rings are processed in the same context.
This results in dp_mon_reap_timer_handler and
dp_process_lmac_rings calling dp_rx_mon_status_process_tlv()
in different context and access global list at the same time,
which may lead to skb double free issue.
To fix the issue if current mode is not monitor mode
do not set rx_mon_ring_mask MSI interrupt mask.
Change-Id: Icc64455824f919e17d3daa741be1cbdafd6fe88f
CRs-Fixed: 3274531
Multiple instances of the misspelling "recevied" are present, so
replace them with the correct spelling "received."
Change-Id: Ibd7f4a136541be8aa98f5bee4db7f75af0271f53
CRs-Fixed: 3273526
When EasyMesh feature is enabled, ARP replies from WDS node are
getting dropped by the AP, as there is no proper way to find the
DA_PEER with given WDS mac address.
Fix is to get the peer mac address for wds node based on ast entry.
CRs-Fixed: 3265158
Change-Id: I5ea890ce37dfde89d067b471417ba7aeb8c5d6f5
Update tx_fast callback with dp_tx_drop in
Umac pre-reset handler to pause the tx.
Change-Id: I39a74179a3deb8fba34acb7149046471e3f4797b
CRs-Fixed: 3267249
Add framework to use different RX hash values and ring masks
for ML and non-ML peers
Change-Id: I098cb50b8873eb137ce096011d01a5c21aaf854f
CRs-Fixed: 3269916
Maping and unmaping of the MLO DP peer to the partner soc
will be initialized based on architecture.
dp_mlo_partner_chips_map
dp_mlo_partner_chips_unmap
Change-Id: I918f6d552acd9d64086abf4c38f913348e13c381
CRs-Fixed: 3267904
Check peer_monitor is valid before processing.
This is required for new boot flow where vap creation starts
before monitor ko insmod.
This check makes sure that monitor_peer is valid which indirectly
means that monitor ko is insmoded.
Change-Id: Ifd132cd6717e3537dd1aa7b7b37e95012d7fe8fc
CRs-Fixed: 3272747
This is a new FAST TX API which avoids various checks.
This API will be called when SFE tags a pkt as fast_forwarded
and vap's fast_tx flag is set.
avoid additional re-checks in the wifi TX function
CRs-Fixed: 3218650
Change-Id: Iba17ede59652a1ff2af553f57de21dc58946298e
Error Handling in Event Based Mechanism to reduce
delay by removing sleep while getting response
from firmware.
Change-Id: I9ee8e03b0d0ebdaae150417fb0fb0531e5bf4a86
CRs-Fixed: 3268808
Use WLAN_CONFIG_TX_DELAY to enable HW Tx delay for both
SAWF and non-SAWF stats.
Change-Id: I6db7d386c0e3654f53398dcba1db7c5568544b0c
CRs-Fixed: 3262425
Based on traffic end indication marked packet coming
from network stack, send indication packet to fw via
exception path to terminate ongoing SP.
Change-Id: Ia2be60d485be4b05665cf6b1684af9258417ffb1
CRs-Fixed: 3207392
Access peer, vdev only if peer_id in current pkt is
different from previous pkt.
CRs-Fixed: 3257844
Change-Id: If10356a01030511c7473112f67f3434d328d4586
Enable subtype filtering for all types of frames. Check the subtype
for all types of frames and drop the frames which are not requested.
Also enable PPDU level filtering by dropping the MPDUs after the first
MPDUs.
Change-Id: I94496ec278724a7023a79b50c2cd7a77b2707e74
CRs-Fixed: 3247309
Fix the mpdu logging option in htt command so that msdu/mpdu
logging is correctly written into the register.
Change-Id: I98b8a1a78cc0cfae37e91e9e3e489ac54139ff75
CRs-Fixed: 3265015
In some conditions REO error ring may have RX flood, but
there is no force break scheme for that ring process. This
leads to starvation of RX refill buffer ring and then FW
crashes due to no refill buffer. Change is aimed to force
break of ring processing and refill buffers in small batch
count.
Change-Id: I3c543ebb16bd1709e782be759c9063a356a96562
CRs-Fixed: 3262685
In function dp_rx_defrag_ccmp_decap(), it will remove the 8 bytes
ccmp header for each fragment, but later at the end of dp_rx_defrag(),
8 bytes ccmp header will be removed again, then actual 8 bytes packet
data will be lost, de-frag failed.
Do not remove ccmp header in dp_rx_defrag_ccmp_decap() as it is not
necessary.
Change-Id: Ibf7beca38e9979a3203d4de702716c8e24e435fe
CRs-Fixed: 3263872
Loop is continuing as chain index is getting assigned to zero inside loop.
Adding check for chain index to avoid infinite looping.
Change-Id: Ifff67c00b15d12195e5a3bf554de1529511c400f
CRs-Fixed: 3261913
Pool id validation should not be against MAX_PDEV_CNT because pool id
1 is valid in case of monitor status ring process. Hence, added a new
API dp_get_rx_mon_status_desc_from_cookie() which validate the pool id
against NUM_RXDMA_RINGS_PER_PDEV to get rx_desc from rx_desc_status pool.
Change-Id: I20baf8c667c1983eff08a204afe3736708d844d1
CRs-Fixed: 3262544
Driver is able to get rates info per MSDU, which can
exclude special frames like ARP, DHCP, etc. This change
is to refill the rates info parsing by rateflags from FW,
including legacy rate, MCS, NSS, BW, preamble, GI, etc.
Change-Id: I129d5f0dc48758d1a92ac0c3622c1761cdae8286
CRs-Fixed: 3255467
When nbuf allocation failure happens, there is no retry scheme.
So RX buffers in refill ring may shrink and not got enlarged.
This change is aimed to replenish more RX buffers when watermark
is less than critical low threshold.
Change-Id: I201b9e252ba08ba1bff47e0b5ec819a45f1b1ddf
CRs-Fixed: 3245915
Add HW, SW and nbuf prefetch support in Berryllium, this will
ensure we have prefetched the HW desc, SW desc and nbuf by
the time we are in the 3rd iteration of the dp_rx_be_process
first loop.
CRs-Fixed: 3218647
Change-Id: I27d371c5d1c9a37d61e4fc00d5eb03609fad589c
Features that rely on monitor status rings do not work when
num_rxdma_status_rings_per_pdev is assign to 1. This change
will fix this issue by removing this assignment statement.
Change-Id: Ic7b8b84be6fa9c34c15a86cb6ec5174cc512ddd9
CRs-Fixed: 3258392
In QCN9224 fetch the peer meta data from the msdu end tlv
instead of MPDU start
Change-Id: Icd9420cd83e06abe5e54e9e05cc8cbf8d8312ae1
CRs-Fixed: 3245626
Added fix,
a. To populate response timestamp from phy timestamp.
b. Based on window type swap address1 and address2 to match the sniffer.
c. For VHT frames, increment 1 to the received nss value.
Change-Id: I68dffadb609d82e2ad2cf433fcee2a76b0be98e7
CRs-Fixed: 3235718
skb queue is not always mapped to the proper CPU
hence it is good to map the ring_id to CPU through get_cpu
API.
Change-Id: I3bf76452cc1fcb2033ef1a8640c4dd31ecba61b3
CRs-Fixed: 3255144
Throughput based RTPM logic of ring access may fail and
lead to NOC error in some corner cases. Change is aimed
to add additional link states check to prevent invalid SRNG
access.
Change-Id: I7e7edbadfd21b4857efa4faff0ada6d94d682f2c
CRs-Fixed: 3256702
When TX completion entires are pending over 60 seconds,
Perform force free for these entries that belong to invalid
vdev. This could allow system to continue to go to suspend.
Change-Id: I34a39b8c7a9385fea0bed087e0644564f426bf90
CRs-Fixed: 3252807
Added fix for two leak,
a) During wifi unload, buffer used in init need to be free during deinit.
b) Before buffer free, unmap is required for the buffer mapped to DMA.
Change-Id: I4e5a63b0c2aabb69367365e74961a63194144057
CRs-Fixed: 3157424
use msdu_count and index to occupy buffer headroom within mpdu.
new mechanism to debug buffer while using pf tags enabled
Change-Id: I1b7d8a1d7ec93f51869119294d2f0b14e3909fe9
CRs-Fixed: 3217284
Call UMAC reset initialization API from soc_attach_target().
FW exposes a service bit for this feature, use that to conditionally
enable this feature. Also, add default log levels for this feature.
CRs-Fixed: 3253464
Change-Id: Ia7c9cf07a7ab7b000ebe452ab074a82173b70129
DP uses multi page allocation for tx/rx descriptor.
ID and offset mask of decriptor is based on 4096 which
couples with Kernel's MMU PAGE_SIZE. This cause trouble
when deploy driver on none-4K page size kernel.
Set qdf_dp_blockmem_size to 4096 so that DP won't
depend on kernel page size.
Change-Id: I17f5c10b394e8709e6b4b153f3dd094cf792787f
CRs-Fixed: 3235246
