DP peer changes required for multi-chip MLO.
This change includes
1) Adding MLO peer to global peer hash at ML context
2) Add ML peer to all partner chips id to objtable
Change-Id: I230a6c1b14484c587b190a9a318fe9ffb1caea11
Changes needed for MLO soc attach to pass chip_id,
dp_ml_context from upper layer.
This change also takes care of assigning appropriate
RBM id for IDLE link descriptors based on chip_id.
Change-Id: I8f5f08c524d91942e6e458f048700b7bdd900107
Changes to have HW cookie conversion context per
desc pool.
This context will be used to program CMEM of the
other SOC in case multi-chip MLO.
Change-Id: I5ec68813e8fcb6d124698a52f5553acf9a7b1795
Prefetch RX HW desc, SW desc and SKB in pipeline
fasion in the first loop of RX processing.
This has improved TPUT by 200Mbps and provided a
10% gain in CPU (single core)
PINE with other optimizations: 3960Mbps @ 100% core-3
PINE + pipeline prefetch: 4130Mbps @ 90% core-3
Change-Id: I47f351601b264eb3a2b50e4154229d55da738724
Added an API to do a batch invalidation of REO descs
saw an improvement of 40 to 45 Mbps.
Note: this change is applicable only for cached
descriptors
PINE with Default driver: 3189 @ 100% core-3
PINE with skb prefetch: 3469 @ 100% core-3
PINE with skb pre + batch inv: 3506 @ 100% core-3
Change-Id: Ic2cf294972acfe5765448a18bed7e903562836c3
This force use BA64 ini config is no longer needed, because another
gRxAggregationSize can do the same settings and more flexible.
Change is used to remove this config.
Change-Id: Ie780489849f8b701481a628a9bca2b4112460bd8
CRs-Fixed: 3076982
Dynamic GRO feature is enabled by default and aimed for specific
customers. Add an ini control to allow other customers to config
this feature enable/disable.
Change-Id: I7f505599327ac131b3cdac9b4d9e038861b1aeb6
CRs-Fixed: 3074689
For IPQ products, there is 1 refill ring which is of hardware type
and host replenishes the buffers onto this ring so that hardware can
use these buffers for Rx.
In IPA offload mode, the buffer replenishment model is different from
the one mentioned above. There are 3 refill rings, out of which,
2 are software refill rings (1 for host and 1 for IPA), and last ring
is hardware ring given to FW.
Ring given to IPA is to refill the buffers after processing the
regular Rx packets and ring given to host is to refill the buffers
after processing of exception packets. Since there are 2 entities to
refill the buffers, the hardware ring given to FW multiplexes these 2
software rings and provides the buffers to hardware.
Make changes to follow above replenishment model for SDX+Pine
integration.
Change-Id: I0d9e4ec811a3023a258e0a6b9ee22ccdffcebafa
CRs-Fixed: 3049633
Tid in RX frame header may be larger than MAX TID allowed
value, this will lead a out of boundary array access and
lead to kernel crash at last. Change is aimed to do a TID
check and discard such frame when necessary.
Change-Id: Ie9e7a1816d197d05cf845e81251ef7772721b849
CRs-Fixed: 3071743
Some platforms don't have mon_register_intr_ops enabled so add
a similar macro around dp_mon_register_intr_ops to resolve
compilation issues.
Change-Id: Id9c7bb45d965005d4dd0dde3a08f254464244147
CRs-fixed: 3075651
Check of Invalid QOS tag which can be set for non-mscs clients
connected on same VAP as MSCS clients. TID override needs to be
avoided for those clients
Change-Id: I651a354e740fe6aee74f94b59ac2e6f154a80c6d
remove the assert when HTT_TX_FW2WBM_TX_STATUS_MEC_NOTIFY
is received and mec_fw_offload is enabled.
Change-Id: I1b9c876822e3e9c05b5035af82afa484106e880a
For DA MC/BC RX frame, host should not use da_idx to access soc->ast_table
as this da_idx is not valid, OB accessing might happen.
Change-Id: I5c78d869e32536effe2635e561cb6881cdc97c38
CRs-Fixed: 3072841
As part of code change we are removing csum_enabled flag check.
In that case, csum_enabled becomes redundant since we are not using it
in the code base, So remove redundant code as part of the cleanup.
Change-Id: Iac411b20f06436053b1969a1af9e3b3ee418c34c
CRs-Fixed: 3070858
Currently, to set checksum enable flags in Tx descriptor
we are checking the csum_enabled flag along with stack checksum
offload request. In case of roaming from latency-critical connection
to non-latency critical connection we request stack to calculate
checksum and set csum_enabled flag to 0. For some packets which
are already queued where the stack has not calculated checksum and
csum_enabled flag is set to 0, we will not set the checksum enable
flags in Tx descriptors for those packets.
To fix the issue remove the csum_enabled flag check and directly
check if the stack has calculated checksum for those packets or not.
Change-Id: I8d7754afc3d0a33315b85b0113cd3062e5783e28
CRs-Fixed: 3070858
Define a new cdp interface to fetch peer delay stats.
Define a new cdp interface to fetch peer jitter stats.
These interface APIs will be used by upper layers to fetch delay and
jitter stats per peer for telemetry stats feature.
Change-Id: I96ee6a861fa2626e7e1fba3df7df9ec64ff7e946
CRs-Fixed: 3071493
In WBM error processing read peer_id from peer_meta_data
instead of sw_peer_id.
This changes is needed because we need to process Rx packet
on ML peer. But in MLO case sw_peer_id field contains
link_peer_id where as peer_meta_data has ml_peer_id.
Change-Id: I3f469adfdf7efa88cb081e94fa9fe0c54c1fb078
When monitor KO not loaded Return QDF_STATUS_E_FAILURE
from dp_monitor_tx_add_to_comp_queue() so that caller
will free the buffer
Change-Id: Idebcf81121767ccd93d95308433241fcf0a93c93
Currently, we are Flushing fragments for peer as part of rx defrag
clean up. There is the possibility that frag flush is called while
we are accessing frags in rx defrag store fragment path in parallel.
Which will cause access after the free issue.
Extending lock in rx defrag store fragment path to avoid any
parallel flush while accessing the frags.
Change-Id: I1e897b195f61d80ea6738e9a93f7bcaaa04adc97
CRs-Fixed: 3065414
There is a race condition as tid_lock is unlocked early in
dp_rx_defrag_store_fragment and descriptors are replenished
from dp_peer_flush_frags and same descriptor is again
being replenished in fragmented path.
To resolve this, extending lock period till all the operations
on tid.head_frag_desc are done.
Change-Id: I6d2abb2119e3bebf739de9e41334d58ba87ee391
CRs-Fixed: 3068165
Avoid tracing data packets at high throughput levels. This helps save
CPU cycles.
CRs-Fixed: 3059712
Change-Id: Ideca04e142a2f4080cd5758fe27f0b526ea4f9ad
Add timestamps for last TX/RX in peer stats. This stat is needed to be
passed on to upper layers on demand.
CRs-Fixed: 3059712
Change-Id: If0d8b7dc3c139e4d1b670bf98f3f574f02ea9715
Add support to invoke HIF runtime PM APIs in TX packet path based on
throughput level. Essentially at high throughputs, where runtime PM is
not expected to provide power gains, invoking the APIs is deterimental to
performance.
Hence invoke runtime_pm APIs based on throughput level.
CRs-Fixed: 3059712
Change-Id: I49d111bffa8a37c68abbdd54911f9ecc22430562
Repurpose the IPA tx and tx completions rings for
normal use when IPA is disabled either via config
flag or ini.
Change-Id: Ia4b6a89c73d888a217bdef40e3c05435c3bb1bb2
CRs-Fixed: 3059730
For T-put test case, if near_full IRQ is fired, ring_near_full
flag will be set and host will always reap reo2sw ring as much
as possible, but if quota 0, this will lead to dead loop
in dp_rx_process_be.
stop reaping reo2sw ring if quota is 0 even ring_near_full is set.
Change-Id: Ifea885162be3ccc6ae0515fc3c2538c459372887
CRs-Fixed: 3067231
Add APIs in order to support TR-069 commands to return all the
supported rates for a specific mode based on gi setting, no. of
spatial streams and channel bandwidth.
Change-Id: If6999e59b8c83d0b5a15054bfdaa97465f8992e4
CRs-Fixed: 3066766
In ifconfig athX stats, tx packets and bytes were getting double
incremented in NSS offload disabled case. It was due to tx.comp_pkt
stats getting updated both in per packet Tx completion path and in
Tx per ppdu indication from HTT path.
Fix by updating stats only in the Tx. completion path.
Change-Id: Id75d914874480a4e24ccb13223818b67869eb378
Add fisa deletion support by invalidating the fisa
DDR entry when adding a new entry.
Change-Id: I02189e22e09ca0ef5e1fdb5952c7e72cd87d3673
CRs-Fixed: 2954060
In case of successful transmission If MPDU tx
is retried more than the threshold then
increment MPDU retry count by transmit_cnt / threshold
Change-Id: Idbc7e2d0579ff68e272132fe4ffe9f951c4aacf5
CRs-Fixed: 3058464
Avoid using ast_entry in intra_bss handling, instead use
da_peer_id directly from the msdu_metadata
Change-Id: Ic586f297e8e393504d7d399cff7a67c3035aa11f
sw_exception bit will be marked in reo dest ring descriptor for
FW re-injected frame, get this bit and save it in skb->cb,
this bit value can be used for FISA further check.
there is no reo_dest_indication field in reo dest ring on beryllium,
so share same cb member for sw_exception and reo_dest_indication.
Change-Id: I2321121be7dda68ed19faca177d868c7e8ba1dbf
CRs-Fixed: 3056156
If ppdu rssi unit is dbm, need convert to db for API:
qdf_nbuf_update_radiotap, only need convert once for first mdpu in the
ppdu. if convert for each mpdu, wrong rssi will be saved.
Change-Id: I85e92121ddb289e3513022ff092083825197efb4
CRs-Fixed: 3057041
Add API to get dp peer authorize state.
Also modify dp_tx_get_rbm_id_li to update rbm for IPA offload
scenario
Change-Id: I0f8cca4623a1c3b840f336aa6d67740951cb6700
Adding a nbuf and nbuf->data prefetches in 2nd loop
of Lithium datapath RX improved UDP throughput by
about 250Mbps.
PINE default driver: 3189Mbps @ 100% core-3
PINE with prefetch: 3469Mbps @ 100% core-3
Note: PINE reo ring is mapped to core-3.
Change-Id: I7f166b52c3697facdce3954994755c9c1412c1f3
