Repository navigation
rx filters: Jaguar1 port, identity refusal, caps flag (#597) - #599
Conversation
Jaguar1 applies rx.drop_ctrl and rx.own_addr_only. own_addr_only goes through the shared monitor_rcr(); bring-up already programs MACID from EFUSE, so it needs no identity arm. drop_ctrl clears RCR ACF as well as RXFLTMAP1: on an 8812AU the map alone filtered nothing. BlockAcks still arrive, since that MAC passes them with data frames. Jaguar2/3: rx.own_addr_only with a beamformee arm and an ACK responder on a different address is refused at bring-up, since MACID holds one address. Jaguar3: InitWrite clears the "RX filters closed" state after its bring-up too, as Init does. AdapterCaps::rx_mac_filters_ok (adapter.caps "rx_mac_filters") says whether a backend applies the two knobs: Jaguar1, Jaguar2, Jaguar3, the RTL8733B. DeviceConfig.h records that a few control frames admitted during bring-up can still reach the host under rx.drop_ctrl. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
PR Summary by QodoApply Jaguar1 RX filters and reject conflicting MAC identities
AI Description
Merge Risk
|
Code Review by Qodo
1.
|
…own_addr_only The Jaguar1 beamformee answers on the EFUSE address bring-up leaves in MACID, and SetAckResponder moves MACID. With rx.own_addr_only that is the only unicast accept path, so the combination is refused at bring-up, as on Jaguar2/3. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
|
Addressed Qodo's finding (Jaguar1 beamformee + ACK responder under Checked on the 8812AU (
So on the 8812 die the three cannot share a session at all; |
What this is
The rest of #597, on top of #598 (which landed the Jaguar3
Initflag clear, the MACID-zero warning and the comments).rx.drop_ctrlandrx.own_addr_only.own_addr_onlyuses the sharedmonitor_rcr(); Jaguar1 bring-up already programs MACID from EFUSE, so it needs no identity arm.drop_ctrlclears RCRACFas well as RXFLTMAP1, because on this MAC the map alone filtered nothing.own_addr_onlywith a beamformee arm and an ACK responder on a different address. MACID holds one address, so that session could not receive both. Same address, or promiscuous RX, is unaffected. rx: follow-ups to the monitor RX filters (#595 review notes 1-3) #598 documented this; this makes it an error (std::runtime_error, like a failed responder arm).InitWriteclears the "RX filters closed" state after its bring-up, asInitnow does.AdapterCaps::rx_mac_filters_ok(rx_mac_filtersinadapter.caps): whether the backend applies the two knobs. Jaguar1, Jaguar2, Jaguar3, RTL8733B.DeviceConfig.h: a few control frames admitted during bring-up can still reach the host underdrop_ctrl.Measured
All ch36, one unit per die. The flood is a T3U sending unicast QoS-Data to address X; a hardware ACK responder for X puts control frames on air. Counts are the DUT's own per-frame events over a 10 s cell (8 s for the last two rows).
Jaguar1, 8812AU (on a second host, RF-coupled to the bench):
drop_ctrlown_addr_only, EFUSE identityown_addr_only, identity Xown_addr_only, identity YManagement frames were 364–368 in every cell. With RXFLTMAP1 = 0 alone,
drop_ctrlread 4184 control frames against 4234 off.8812AU as transmitter,
drop_ctrlon its own RX (peer: an 8812BU hardware responder; TX reports OK):drop_ctrlBlockAck under A-MPDU with
drop_ctrlon the transmitter (the item #595 left unmeasured), TX reports OK, open /drop_ctrl/ no responder: 8812CU 96 / 96 / 0%, 8812BU 95 / 97 / 0%, 8733BU 95 / 97 / 0%. Under aggregation the per-frame retry count reads 0 in every cell, including no responder, so only the OK share discriminates.The refusal,
rxdemowithDEVOURER_RX_OWN_ADDR_ONLY=1 DEVOURER_BF_ARM_BFEE=…, on an 8812CU and an 8812BU: responder on another address exits with the error; responder on the beamformee address runs; the same pair withoutown_addr_onlyruns.The leaked control frames from the #595 review (1–3 per cell on Jaguar3 and the RTL8733B): all sat in the first ~40 events of a session, while an ambient NDPA source ran at ~60/s for the whole unfiltered cell. They are frames admitted during bring-up before the monitor filter is written. Later runs had 0 or 1.
Regression on this build: the #595 filter cells on the 8812CU and 8733BU:
drop_ctrl1 and 0 control frames against 3396 and 2678;own_addr_only0 flood frames without the identity, 1545 and 1470 with it.ctest96/96.Not verified
drop_ctrl. The cross-rig loop did not close with both knobs off either (the transmitter retried every frame to its limit with the responder armed), so the run says nothing about the knob.InitWritethenIniton one Jaguar3 object, thenStop, logged a clean de-init and did not exit in two of three runs.🤖 Generated with Claude Code