frame control for second and subsequent user are zero which overide the
master frame control. Added fix to update master frame control from user
position zero.
Change-Id: Ic0331dd5d486fbffa44d29c74b4f1f15414d0a65
Currently, when an extension descriptor is used during TX datapath, we
use HTT metadata length + msdu extension descriptor length, even we
do not fill HTT meta data. This is inefficient and hardware will end up
doing DMA of extra HTT metadata, even when not needed.
Include length of metadata in TCLdata command only if HTT metadata is
filled in.
Change-Id: Idd1b053c1c6bd8b39880efe5743980667454df0d
CRs-Fixed: 2738733
This change will acquire MSCS parameters from the
MSCS request sent by the STA and provide it to the
data path which will later use it for building flow
database.
Change-Id: I972154792c682527f8df04de693077188726e8da
CRs-Fixed: 2669386
As index based search is protected with refcnt now
reverting the changes which are added as a WAR to protect
peer memory
Change-Id: Iac837d45d0d4b2509bbf7d60401d9497c0d998b0
New approach requires module IDs to be passed to
acquire and release peer references
Hence pass module ID to acquire and release the
peer reference
CRs-Fixed: 2755560
Change-Id: Id351179da68f6ca30f13710a47575f3ff550ada3
Add new APIs to iterate through the vdev peer list
by taking the peer reference.
A call back has to be passed to these APIs which will
be called for every peer in the lisst
dp_vdev_iterate_peer() -> iterates through peer_list of vdev
dp_pdev_iterate_peer() -> iterates through peer_list of all
vdevs in pdev
dp_soc_iterate_peer() -> iterates through peer_lists of all vdevs
in soc
Additonal APIs are added to for itearations where the callback
will be called outside the vdev->peer_list_lock, as these APIs
have additional memory allocations suggested to use only in case
the iteration need to happen outside lock
dp_vdev_iterate_peer_lock_safe()
dp_pdev_iterate_peer_lock_safe()
dp_soc_iterate_peer_lock_safe
Change-Id: I24632fe611355cc3e93b7f16d90913d4b8686ca9
Remove the peer backpointer in ast entry and store
peer_id instead
Assign peer_id in AST entry in AST MAP event,
also add the ast entry to peers ast list
In AST map & AST unmap APIs use ast find by vdev_id
Change-Id: I74d9828dc309149d98f6f577b5c8304cb087fd76
Add support to get the peer reference with module id
To help debug the peer reference related issues
Change-Id: Ie20c7e710b9784b52f2e0f3d7488509282528a00
Use unified version of dp_peer_find_by_id API
which will take peer reference
Also use unified peer ref release API dp_peer_unref_delete
Change-Id: Ibb516a933020a42a5584dbbbba59f8d9b72dcaa4
Update peer DB as per new design
peer_hash_table: add at peer_create and delete at peer delete
peer_id_to_obj_map: add at peer map and delete at peer unmap
peer_list: vdev level peer list add at peer create delete at
peer delete
Call dp_peer_cleanup() at unmap event
dp_peer_cleanup -> dp_peer_tx_cleanup -> Tx capture cleanup
|
---------> dp_peer_rx_cleanup -> Cleanup Rx REO TID queues &
defrag queue free
Introduce ref_cnt variable per VAP which will be
incremented and decremented at peer alloc and free. This counter
will be used to delay the vdev free if there are any active
peers for vdev exist in system
Change-Id: Iec13bceed0db0db71655fdbf6f96df8795b5077b
Remove the global soc->peer_ref_mutex lock to protect
peer access
Instead use granular locks to protect the individual
peer DBs
1) soc→peer_hash_lock : To protect peer hash table
2) soc→peer_map_lock : To protect id to objmap array
3) vdev→peer_list_lock : To protect vdev level peer list
Change-Id: I9f044c4d71fe1155817a2b22a5a050545ec37445
Peer wlanstats context was accessed directly from
PPDU cookie, with this approach the cookie pointer
may become stale if peer goes for deletion
Hence access the wlanstats context using CDP API
with stats context protection
CRs-Fixed: 2740261
Change-Id: I54c785325f8852ad422f05527b98ebca6e4d6cf0
hdr_ptr is in skb_buffer data, it's assigned with 6B array,
use uint8_t point convert to avoid SA overflow warnning.
tid has asseration protect, but need to break execute to avoid
of SA warnning
Fix use-after-free of ast_entry
Change-Id: I0835f93291cf3da2b4fd57d8c9a90f20a60c11ee
CRs-Fixed: 2751678
UPDATE_RX_REO_QUEUE commands are queued in the REO_CMD
ring as part of peer cleanup for each tid. The cb
registered defers pocessing the entries in the
reo_desc_freelist based on either number of pending
entries or time threshold. In the scenario when
these entries from the freelist are processed as part
of next disconnection and if the REO_CMD enqueue fails,
the while loop over reo_desc_freelist does not exit
and will end up in a perpetual loop.
Fix is to break from the desc_freelist loop on REO
cmd enqueue failure.
Change-Id: I18ee31ec6a6150cd6952903cc154634b7de4ce60
CRs-Fixed: 2751223
Add allocation check failure before accessing index in
dp_soc_rx_history. In case of allocation failure for a particular ring,
continue with allocation and history recording of other rings.
Change-Id: Ifae46ddf38b3af271900651cdbba662325220ef9
CRs-Fixed: 2747891
In dp_srng_init, max_buffer_length and prefetch_timer are used
while uninitialized.
In dp_bucket_index, overrunning array cdp_sw_enq_delay leads to
out-of-bounds access.
In dp_rx_defrag_fraglist_insert, cur is first NULL checked but
cur is again set to qdf_nbuf_next and is accessed without
NULL check. Thus do a NULL check again before dereferencing
cur to avoid potential NULL pointer dereference.
In htt_t2h_stats_handler, soc could be NULL while cmn_init_done
is dereferenced. Thus fix it by NULL check soc first and then
dereference cmn_init_done.
Change-Id: Ie6a33347d34862f30ba04a10096d3892af7571d3
CRs-Fixed: 2751573
We have come across scenarios where rxdma is pushing
a certain entry more than once to the reo exception
ring. In this scenario, when we try to unmap a buffer,
it can lead to multiple unmap of the same buffer.
Handle this case, by skipping the process of this
duplicate entry, if alrady unmapped, and proceed to the
next entry.
Change-Id: Iae66f27e432f795ba4730911029fa1d63a75cb06
CRs-Fixed: 2739176
Add support to retrieve per PPDU phy tx duration without response time.
Add support to retrieve per User phy tx duration for MU-MIMO and OFDMA.
These metrics are useful for ATF stats feature. Stats processing done
only if atf stats flag is set.
CRs-Fixed: 2709102
Change-Id: I90db41aa45843808e07343a80750c154abbde9cd
Memory optimization of monitor status ring by allocating buffers during
replenish using alloc_skb (linux API).
It creates buffer of required size rather than 4k size(dev_alloc_skb)
Change-Id: I3ae5e403de28c4570f8ac3b50d3ca878a9e4b2f9
CRs-Fixed: 2733931
Rx desc pool have a flag to identify whether frag or nbuf
operation needs to be performed for alloc, map, prep and
free buffer for monitor dest buffer.
This flag will be set only for mon destination desc pool,
if RX_MON_MEM_FRAG feature is enabled.
In all other case, It will be set to Zero and default nbuf
operation will be taken.
This flag get initialized at the time of pdev rx_desc_pool
initialization and gets reset while pdev deinit.
Mon destination buffer will have support for frag if
RX_MON_MEM_FRAG flag is set.
Change-Id: I67c6c823ee4f114035b884c024a1a9054a40665b
CRs-Fixed: 2741757
Currently when query REO stats, each time REO will response
cumulative REO stats per TID, while host side count them as
delta stats between each query.
Reset host stats to 0 before each query, and if query timeout
happened, use the last saved stats for this time query.
Change-Id: I29e986b3c8859f5c3d1b557d9cc980a52219d664
CRs-Fixed: 2743352
Currently in_use and unmapped flag from rx_desc can not know is
the rx nbuf has done IPA smmu map/unmap accurately.
add rx nbuf IPA smmu map/unmap flag in qdf_nbuf_cb, and use
this flag to check if IPA smmu map/unmap need to be done or not.
Change-Id: I19bbc6cc69ba752ef7c58d41fbe26299ca96d29e
CRs-Fixed: 2731313
Deliver fragmented frames directly to stack without reinjecting back
to REO. Handle PN check for fragmented frames before delivery.
Drop the frames on PN check failure.
Change-Id: I7865def0d39fa83378073e07d318c34dccc6c6e5
CRs-Fixed: 2739870
Move RU and MCS strings from dp_stats.c to cdp header file so that these
strings can be used in user applications including the CDP header. This
will remove redundant definitions of these strings in user application
Change-Id: Ie74283f54131378447324b927e19e9716ef7519a
CRs-Fixed: 2749062
When matching ppdu_id is detected during HAL processing, then print the
error message instead of going for assert.
Change-Id: I8f58359a66fbeaf2ebc98477f65bab446c0a784b
CRs-Fixed: 2747953
Remove self and BSS peer back pointers and added APIs to
get self and BSS peer from VDEV
These new APIs will iterate through peer list and return
the appropriate peer by taking a reference
Caller has to take care of releasing the reference
Change-Id: I550ff83d665f3ad0a185bea1a1e6270c5474ff07
wdi event has to be called before dp_txrx_ppdu_stats_handler.
this will handle all pktlog htt messages including cases for mgmt frame
Change-Id: Iefba15b61278fe77cbb6d6df9190e66b46ed573c
Add WMI support for WMI_TWT_SESSION_STATS_EVENTID. This event
contains stats for a given TWT session.
Change-Id: I01d5f7b30da803ee713a14c1d1124b8af7161bca
CRs-Fixed: 2609951
Same back to back link descriptor address/cookie is observed in
WBM idle link desc ring.
add duplicate link desc check when host
refill link descriptor to WBM through SW2WBM release ring,
also REO reinject ring.
Change-Id: Iaf9defd87670776fa9488d7f650efa3c08fefa60
CRs-Fixed: 2739879
Notification is sent in following two scenarios.
- After call to dp_peer_add_ast() if we are not expecting any
peer map event.
- When HTT peer map event is received.
Change-Id: Id926d1cec1ee64520b0368fcc5000988c24e0513
CRs-Fixed: 2745542
The rx rings are relatively of smaller size. Any
duplicate entry sent by hardware cannot be back-tracked
by just looking at the ring contents alone.
Hence add a history to track the entries for the
REO2SW, REO exception and SW2REO ring.
Change-Id: I9b0b311950d60a9421378ce0fcc0535be450f713
CRs-Fixed: 2739181
In monitor mode, radiotap header contains VHT/HE information in management
and control frames after association of STA to AP.For this,reset of
ppdu_info is required before processing of ppdu in order to avoid reuse
of ppdu_info.
Change-Id: Ia36ec2664f79d645ff76db016f986dfb92ec8bbd
CRs-Fixed: 2734804
Th nbuf sanity can fail in case when HW posts the
same buffer twice. This case can be handled gracefully
by just skipping the processing of the corresponding rx
descriptor.
Change-Id: I471bb9f364a51937e85249996e427f15872bda97
CRs-Fixed: 2738558
DP RX changes to support RX buffer pool, this is a pre-allocated pool
of buffers which will be utilized during low memory conditions.
Change-Id: I8d89a865f989d4e88c10390861e9d4be72ae0299
CRs-Fixed: 2731517
Add support for RX buffer pool, this is a pre-allocated pool
of buffers which will be utilized during low memory conditions.
RX buffer replenish will fail when the system is low in memory;
this is okay if the condition is momentary, but if the
system persists in such state for longer duration, then HW will
run out of free buffers and this results in a crash.
RX buffer pool will be helpful in such scenarios. Buffers from
the pool will be given to the HW when the system is running low
in memory.
Change-Id: I7ca1159d858227a0a2021235d24d763e56ac0b27
CRs-Fixed: 2731517
copying mu data info is done inside schedule command status tlv
and handling based on completion status is added.
Change-Id: I2792d754a4fa3e2992d68f569c9ce40be8ff02e2
HTT interface to enable FISA has been modified to accomadate only
FISA enable and aggregation limit fields, rest of the fields will
be initialized by the firmware; this change is in accordance with
it.
Change-Id: Ia5950541b8d852029650c02e81c097bf049e54c5
CRs-Fixed: 2731543
Currently when running downlink traffic with
fragmentation enabled, the SW2REO is getting
full. Increase the SW2REO ring size to 128 entries.
Change-Id: If43bc72a8cc173d44953ca367573800243b1cc5d
CRs-Fixed: 2738309
Ratelimit the defrag path error logs and add the stats
for fragmented packets received out-of-order which
inturn leads to sequence number mistamtch in defrag path.
Change-Id: I17d4c1cff214a8c8a05abf576701824b293d2883
CRs-Fixed: 2740805
Currently, per-user information in Tx PPDU completion is allocated for
37 users though FW may send information for fewer number of users. In
new scheme, FW provides number of users in USERS_INFO TLV and host can
allocate the per-user struct based on this number. This can save memory
while processing each completion message.
CRs-Fixed: 2730923
Change-Id: If04fd664a1ef6e77e8eb44c5d2bd1fa89fcaba72
