Pre-fix S_IDLE had two independent if-branches: one for frame_start_pulse
(resets pointers) and one for data_valid (transitions to S_ACCUMULATE).
A data_valid arriving before frame_start_pulse would advance the FSM with
whatever pointers happened to be live, and the BRAM write block would write
the sample into mem_write_addr = (write_chirp_index*RANGE_BINS) + 0.
In current operation the race is benign — end-of-S_ACCUMULATE always zeros
write_chirp_index/write_range_bin (line 287-288) and the MF pipeline latency
(~165 µs) is millions of cycles longer than the frame_start CDC latency
(~50 ns), so frame_start always arrives first. But the FSM relies on an
undocumented system-level invariant; a future code path that leaves
pointers stale on entry to S_IDLE would silently corrupt the first sample.
Fix: add a `frame_armed` register set when frame_start_pulse arrives in
S_IDLE, cleared on transition to S_ACCUMULATE. Both the FSM transition and
the BRAM write block gate on `(frame_start_pulse || frame_armed)`. The OR
admits the same-cycle case where both arrive together (write to addr 0
still resolves correctly because both blocks use the same gate).
Verification: tb_doppler_frame_start_gate 21/21 PASS, quick regression
32/32 PASS (was 31/31; +1 new test, 0 regressions). tb_doppler_realdata
(full FFT pipeline) still passes — gate transparent to normal operation.
Bug: 16-bit detect_count was reset only on power-on; increments at three
sites (ST_IDLE/ST_BUFFER simple-threshold paths and ST_CFAR_CMP) accumulate
across frames. At 178 fps with even 2-3 average detections per frame the
counter wraps in 100-180 seconds, breaking any rate-based host telemetry
or health check that reads it.
Fix: add `detect_count <= 16'd0` in ST_DONE so the counter represents
"detections this frame" instead of cumulative-since-boot. Updated $display
wording from "total detections" to "frame detections".
T13 flipped from "count keeps growing" to "identical-scene frames produce
identical counts" (the actual contract a per-frame counter must satisfy).
TB snapshots detect_count during ST_DONE because cfar_busy only goes low
on ST_IDLE entry — after the reset has fired.
Verification: tb_cfar_ca 24/24 PASS, quick regression 31/31 PASS.
Note: detect_count output port is now "live" (accumulates during frame,
0 between frames). Audit confirmed no current host telemetry consumes
this port. If future host code needs a stable last-frame total, add a
detect_count_last_frame snapshot register then.
AUDIT-C12: usb_data_interface_ft2232h had a misleading single-buffer comment
that overstated the timing slack and referenced a frame_ack_toggle CDC that
was never implemented. Re-verified actual numbers: at 178 fps the slack is
1.14 ms (20%), not "much shorter than gap". No data corruption today (write
order matches read order, addresses don't collide), but frame_complete
firing while WR_FSM is still draining the previous frame causes silent
frame drops via the missed frame_ready_toggle edge.
Fix is instrumentation, not architectural rework: add wr_done_toggle
(ft_clk -> clk CDC) on WR_DONE -> WR_IDLE, track frame_pending in clk
domain, count drops in 7-bit saturating frame_drop_count, surface in
unused upper 7 bits of status_words[5]. Host now has visibility into the
failure mode if margin ever shrinks (faster frame rate or USB bandwidth
shortfall). Replaced misleading comment with corrected timing breakdown.
AUDIT-S22: cfar_ca emits one detection per 3 cycles (THR/MUL/CMP); the
detection RMW takes 3 cycles. Match by construction today, fragile against
any CFAR speedup. Added a header comment in cfar_ca.v documenting the
dependency, and a SIMULATION-only assertion in usb_data_interface_ft2232h.v
that fires [ASSERT FAIL] AUDIT-S22 if cfar_valid arrives while RMW busy.
Catches silent-drop regressions in the test suite.
Verification: new tb_ft2232h_frame_drop.v with 5 scenarios (no drops /
stalled drops / multi-drop / recovery / saturation at 127) - 10/10 PASS.
Quick regression 31/31 PASS (was 30/30; +1 new test, 0 regressions).
Closes AUDIT-C15. The 200T XDC `adc_or_p/n` PACKAGE_PIN was a TODO
placeholder that blocked all 200T synth/impl with unplaced-IO errors.
Pins U20/V20 are L11P/L11N_T1_SRCC_14 — same T1 clock tile as adc_dco_p
on L12_MRCC (W19/W20), so OR captures with the same IBUFDS->BUFIO->IDDR
source-synchronous topology as adc_d_p[*]. Free-pair confirmed by Vivado
get_package_pins query against xc7a200tfbg484-2.
Adds DIFF_TERM TRUE on adc_or_n (was only on the p-side; explicit on
both is safer). Adds input_delay constraints mirroring adc_d_p
(max 1.0 ns / min 0.2 ns on both edges).
Header pin counts updated: Bank 14 21/50 used, total 184/285.
This is the FPGA-team RECOMMENDATION for the production PCB (NEW
design); the PCB designer must route AD9484 OR+ -> U20 and OR- -> V20.
Validation:
- read_xdc + link_design -part xc7a200tfbg484-2 -> READ OK on both
xc7a200t_fbg484.xdc and adc_clk_mmcm.xdc; no PACKAGE_PIN errors.
- ./run_regression.sh --quick: 29/29 PASS (RTL untouched).
The GUI's radar_protocol.py parsed 11-byte legacy packets only. The
production board (50T, USB_MODE=1) emits ~35 KB bulk frames from
usb_data_interface_ft2232h.v, so the legacy parser saw a random walk
of false 11-byte boundaries through bulk data — no usable display on
production hardware.
Bulk parser added (radar_protocol.py):
- parse_bulk_frame validates header, reserved bits, n_range=512,
n_doppler=32, footer-at-flag-derived-offset; unpacks range_profile
/ doppler_mag / cfar_dense per the format-flags byte.
- find_bulk_frame_boundaries is the bulk counterpart of
find_packet_boundaries; status packets (0xBB) handled in the same
stream since FT2232H emits them too.
- RadarAcquisition dispatches on isinstance(conn, FT2232HConnection):
bulk path skips the per-sample state machine and fills RadarFrame
in one shot. FT601 / 200T keeps legacy 11-byte (USB 3.0 has 50x
bandwidth headroom; per-sample format is correct and already works).
- RadarFrame.mag_only flag carries the wire's mag_only bit so
downstream consumers can skip I/Q panels cleanly.
- FT2232HConnection._mock_read now emits synthetic bulk frames
(was misleading legacy 11-byte).
RTL alignment (AUDIT-C9 RTL stub option):
- usb_data_interface_ft2232h.v header no longer promises the
unimplemented mag_only=0 (full-I/Q) and sparse_det=1 paths;
explicit INERT FLAGS note distinguishes the two reasons:
* Full-I/Q is constrained by hardware — needs ~28-BRAM18 I/Q
buffer (50T currently 78% BRAM utilised after FFT IP) AND
USB 2.0 bandwidth (12.21 MB/s vs 8 MB/s conservative budget).
* Sparse-list is feasible — smaller than dense for typical
scenes (<341 detections), ~1 BRAM18 cost. Just unimplemented
RTL work (small list BRAM + new WR_DETECT_SPARSE state).
- New SIMULATION-only assertion fires if stream_mag_only ever
becomes 0 or stream_sparse_det ever becomes 1 — backstop for
any future regression that bypasses the host-register clamp.
- radar_system_top.v opcode 0x04 force-clamps mag_only=1 and
sparse_det=0 in host_stream_control when USB_MODE=1, so a
Custom-Command host write can't push the FPGA into a wire-format
vs FSM divergence.
Bandwidth math (verified for 27c9c22+):
Frame rate = 1 / (16x167 us + 175.4 us + 16x175 us) = ~178 fps
Mag-only frame = 8+1024+32768+2048+1 = 35849 B = 6.38 MB/s
FT2232H 245-Sync-FIFO sustained budget (FTDI AN_232B-04
conservative): 8 MB/s. Headroom 20%.
Tests: test_GUI_V65_Tk.py TestBulkFrameParser — 18 new cases covering
round-trip per stream-flag combo, header/footer/n_range/n_doppler/
reserved-bit/truncation rejection, multi-frame boundaries, bulk+status
mixed streams, byte-drop resync, dispatch-by-connection-type,
ingest-to-RadarFrame end-to-end. GUI 117/117 PASS, v7 83/83 PASS,
FPGA quick regression 29/29 PASS, ruff clean.
Refs: AUDIT-C9 (GUI parses legacy 11-byte vs FT2232H bulk).
Follow-ups (separate patches):
- Sparse-detection write FSM (~1 BRAM18 + ~100 RTL lines).
Bandwidth- and memory-feasible; just unimplemented work.
- Full-I/Q write FSM. Constrained: needs ~28-BRAM18 I/Q buffer
AND USB 2.0 bandwidth headroom (50T post-FFT-IP at 78% BRAM).
The DDC hard-coded an offset-binary->2C subtract on the AD9484 path. The
chip's output format is selected by the SCLK/DFS strap (jumper SJ1 on
RADAR_Main_Board.sch), and CSB is hard-tied HIGH so SPI cannot be used
to confirm or change it from firmware. If the board is assembled with
SJ1 on pins 2-3 (two's-complement), the existing RTL silently mis-
converts every sample.
Add a 2-bit adc_format input to ddc_400m_enhanced (2-FF synchronized
clk_100m -> clk_400m, ASYNC_REG attribute), drive it from a new top-
level register host_adc_format written by host opcode 0x33, and wire
it through radar_receiver_final. Default 2'b00 matches the SJ1 default
strap (offset-binary) and preserves pre-patch behavior. Opcode 0x32 is
intentionally left unused; reserved for the future S-25 fix
(host-driven adc_pwdn).
Tests: tb/tb_ddc_400m.v Test Group 5 — 7 new assertions covering
offset-binary at {0x80, 0x00, 0xFF}, two's-complement at
{0x00, 0x80, 0x7F}, and reserved 2'b10 fallback. 14/14 PASS.
Refs: AUDIT-C3 (DDC offset-binary hardcoded).
Schematic ref: RADAR_Main_Board.sch:46719 (CSB on +1V8_CLOCK_F),
:46845 (SCLK/DFS via SJ1).
tb_radar_receiver_final had three pre-existing issues that all surfaced as
fails in regression (32 passed, 2 failed before; 34 passed, 0 after):
1. host_range_mode was undriven (floating 2'bzz); rmc log confirmed
"Auto-scan starting, range_mode=z". Add explicit 2'b01 (long-range
dual-chirp) for the test scenario.
2. DDC_MAX_ENERGY threshold (2^56) was sized for an unspecified earlier
stimulus; the test feeds a deliberately-loud 120 MHz sawtooth that
produces ~1.27e17 energy over 2M samples. Raised to 2^60 (~10x
observed) so B1b catches true overflow without false-firing.
3. The 9 doppler-frame-dependent checks (S4-S9, G1, B2a, B3, B4) need
~108 ms simulated time to fill a 32-chirp Doppler frame because the
in-house fft_engine takes ~340 K cycles per multi-segment chirp
(RX-NEW-3, commit 5c8cc8c). Iverilog can't elaborate the Xilinx FFT IP
that would make this tractable. Guard those checks behind
`ifdef FFT_USE_XILINX_IP` so iverilog cleanly SKIPs them with an
explanatory line; XSim with the IP runs them normally.
Also tightens run_regression.sh's pass/fail regex from
^\[(PASS|FAIL)([^]]*)\] to ^\[(PASS|FAIL)( [0-9]+)?\] so informational
tags like [FAIL-INFO] (used to document the known RX-NEW-1 fft_engine
bin-shift in tb_matched_filter_processing_chain.v) no longer false-fire
as real failures. The Matched Filter Chain test goes from FAIL (40 pass,
2 false-fails) to PASS (40 checks).
Regression: 34 passed, 0 failed.
The DAC short/long chirp LUTs are 10..30 MHz upchirps (Hilbert-confirmed).
With TX_LO=10.500 GHz, RX_LO=10.380 GHz (adf4382a_manager.h) and the
120 MHz DDC NCO (ddc_400m.v), high-side mixing places the post-DDC echo
at 10..30 MHz baseband. The matched-filter reference (gen_chirp_mem.py)
was generating 0..20 MHz, implicitly assuming the chirp's low edge mixed
to DC. This caused a 10 MHz spectral offset and ~5 dB matched-filter loss.
Adds F_BASEBAND_LOW=10e6 in both gen_chirp_mem.py and radar_scene.py,
with phase formula 2*pi*F_BASEBAND_LOW*t + pi*rate*t^2 in all chirp
generators. Regenerates the 6 .mem files. Adds analyze_short_chirp_mismatch.py
for the Hilbert-based diagnosis. Fixes the misleading "30MHz to 10MHz"
comment in plfm_chirp_controller.v and adds an end-to-end frequency plan
in the LUT header.
Sideband orientation (high-side at both mixers) is the conventional choice
and consistent with antenna match (10.25..10.75 GHz, 8x16 patch designed
at 10.5 GHz). Loopback capture would settle definitively; if either mixer
is low-side the F_BASEBAND_LOW sign flips and/or chirp direction reverses.
latency_buffer.v has had zero non-tb instantiations since RX-B (2026-04-23)
replaced its hookup in radar_receiver_final with a 1-FF alignment register.
The module was being kept "for potential future use" — exactly the kind of
dead weight the codebase does not need. Deleted, along with all build /
test infrastructure that dragged it along:
- 9_Firmware/9_2_FPGA/latency_buffer.v
- 9_Firmware/9_2_FPGA/tb/tb_latency_buffer.v
- run_regression.sh: removed from RTL_FILES and RECEIVER_RTL
- scripts/200t/build_200t.tcl: removed from synthesis source list
- tb/tb_system_e2e.v: removed from header compile-string example
- tb/cosim/validate_mem_files.py: deleted test_latency_buffer() (~75 lines),
its call site, and the corresponding entry in the module docstring
Historical RX-B comments referencing latency_buffer in radar_receiver_final.v,
tb_rxb_fullchain_latency.v, and tb_rxb_latency_measure.v are kept — they
explain WHY the module was removed, which is still useful design archaeology.
Two doc-only housekeeping touches bundled in:
- plfm_chirp_controller.v: replaced two empty "CRITICAL FIX: Generate
valid signal" labels at LONG_CHIRP and SHORT_CHIRP with one shared
chirp_valid policy comment block above LONG_CHIRP that explains the
actual rationale (downstream FIFO underrun on trailing samples).
- v7/models.py: replaced the "range_resolution and velocity_resolution
should be calibrated" docstring (sounded like an open TODO but was a
documented placeholder) with a clear pointer to the GUI-C3 fix in
workers.py:RadarDataWorker so future readers know the live path
derives correct values from WaveformConfig.
FPGA quick regression unchanged: 28/29 (1 fail is the unrelated iverilog/
Xilinx-IP RX-NEW-3 gap). GUI suite 180/180. Ruff clean.
cmd_data / cmd_opcode / cmd_addr / cmd_value feed downstream CDC sync
chains; the safety property is that they only change on the cycle
cmd_valid rises (RD_PROCESS), and stay held on every other cycle so the
receiver's 2-FF synchronizer sees a clean payload regardless of where
its sample window lands. The FSM satisfies this implicitly today, but
nothing flagged a regression that introduced a stray write somewhere
in the same always block.
Added an `ifdef SIMULATION block at the bottom of both
usb_data_interface.v (FT601 / ft601_clk_in / ft601_reset_n) and
usb_data_interface_ft2232h.v (FT2232H / ft_clk / ft_reset_n). It
snapshots the payload + cmd_valid each cycle and fires
[ASSERT FAIL] TX-N9: cmd_<field> changed while cmd_valid=0 (old -> new)
on any payload change while cmd_valid is low. Local regs suffixed _n9
to avoid future name collisions. Synthesis-inert.
Quick FPGA regression unchanged: USB Data Interface 91/91 PASS, overall
28/29 (same baseline; the 1 fail is the pre-existing iverilog/Xilinx-IP
RX-NEW-3 gap).
matched_filter_processing_chain declared `input wire [5:0] chirp_counter`
but never read it inside the module. matched_filter_multi_segment passed
its own chirp_counter through to that dead port.
Removed the port from the chain and the corresponding hookup at the
multi_segment instantiation site. Five testbenches also referenced the
port (tb_mf_cosim, tb_matched_filter_processing_chain, tb_rxb_latency
_measure plus the four MF cosim variants that share tb_mf_cosim) — the
reg/connection/init lines were dropped, and the now-stale "Test Group 8:
Chirp Counter Passthrough" was repurposed as a port-removal smoke test
that confirms the chain still produces FFT_SIZE outputs without that
input.
multi_segment.chirp_counter input remains on the port list (it could
plausibly be wired to per-chirp logic in the future); it is now formally
unused but iverilog/Vivado do not flag unused module inputs.
Quick regression: 28/29 PASS (same as baseline; the 1 fail is the known
iverilog/Xilinx-IP RX-NEW-3 gap unchanged by this commit).
`chirps_mismatch_error` was set in radar_system_top when the host
requested chirps_per_elev != Doppler FFT size, but never wired into the
USB status response — a latent silent failure.
Wired the flag through both USB interfaces (FT601 + FT2232H) into bit
[10] of status word 4 (was reserved). GUI parser exposes it as
StatusResponse.chirps_mismatch.
- usb_data_interface*.v: new status_chirps_mismatch input, packed at [10]
- radar_system_top.v: connect chirps_mismatch_error to both USB instances
- radar_protocol.py + test_GUI_V65_Tk.py: parse new bit, +1 round-trip test
- tb_usb_data_interface.v: drive the new port, update word-4 expectation
Tests: GUI 92/92 (was 91), MCU 75/75, USB TB 91/91, ruff clean repo-wide.
The 2 remaining FPGA regression failures (Receiver Integration, MF Chain)
are the pre-existing iverilog-can't-link-Xilinx-IP issue tracked
separately as the open RX-NEW-3 follow-up.
mti_canceller previously armed has_previous and refreshed
prev_chirp_was_long only when range_bin_d1 == NUM_RANGE_BINS - 1.
range_bin_decimator can early-terminate a chirp before reaching the
last bin (overflow guard at range_bin_decimator.v:306, watchdog at
:314), so on every such chirp MTI never armed and stayed muted forever
on every subsequent chirp until reset.
Detect chirp boundary internally using bin-0 arrival after at least
one non-zero bin in the prior chirp. effective_has_previous lifts
has_previous=1 the cycle chirp_boundary fires so the new chirp's
bin-0 is subtracted (read-before-write on prev[0] correctly returns
the previous chirp's bin-0). prev_chirp_was_long now updates on every
range_valid_d1 (no-op within a chirp; OLD value still visible at the
chirp_boundary cycle for the waveform_changed compare). Pass-through
clears saw_nonzero_bin_in_chirp so the first MTI-enabled chirp after
a pass-through run is correctly muted.
No port changes. tb_mti_canceller T13 added: feed a 32/64-bin partial
chirp followed by a full chirp, verify the second chirp is NOT muted
(would fail without the fix). MTI Canceller goes from 40 -> 43 checks,
all passing. Local regression: 32/34 PASS (same as baseline; the two
failing tests are pre-existing RX-NEW-3 FFT throughput).
Replaces the in-house iterative fft_engine.v in the matched-filter chain
with the Pipelined Streaming Xilinx FFT IP, closing RX-NEW-3 (FFT chain
~11x too slow vs PRI budget).
Components:
* ip/xfft_2048_ip/xfft_2048_ip.xci — committed IP definition
(16-bit fixed point, BFP scaling, convergent rounding, natural order,
pipelined-streaming, BRAM data/reorder/phase factors). Vivado
regenerates .dcp / sim-netlist from this on each build.
* scripts/50t/gen_xfft_2048_ip.tcl — IP-Catalog generation script
* scripts/50t/run_xfft_xsim.sh — XSim batch runner for tb_xfft_2048_xsim
* xfft_2048.v — AXI-Stream wrapper. FFT_USE_XILINX_IP define routes to
real LogiCORE for synth/XSim; falls back to fft_engine batched
one-shot for iverilog (unit coverage only).
* fft_engine_axi_bridge.v — exposes legacy fft_engine port surface on
top of the xfft_2048 AXI wrapper, so the chain swap is a 1-line
module-name change.
* matched_filter_processing_chain.v — fft_engine -> fft_engine_axi_bridge
* scripts/50t/build_50t.tcl — read_ip + generate_target + synth_ip;
adds FFT_USE_XILINX_IP to verilog defines.
* tb/tb_xfft_2048_xsim.v — XSim verification (DC, impulse, tone bin 128).
All 5 assertions PASS on remote with the real IP; tuser=0x0a (BLK_EXP=10)
confirms BFP scaling working.
Local iverilog regression: 32/34 PASS — identical to baseline. Same two
RX-NEW-3 failures (Receiver Integration, Matched Filter Chain) — these
only resolve in remote XSim with the real IP, since iverilog uses the
fft_engine fallback inside xfft_2048 (~150K cycles/pass, not the
~2200-cycle Pipelined Streaming throughput). MF cosim 4/4 PASS confirms
bridge bit-exact in fallback mode.
Pending: remote XSim of tb_radar_receiver_final to demonstrate Doppler
frames produced within PRI budget; remote synth to confirm DSP/timing
post-IP.
Strip the explicit DSP48E1 instance from comb stage 0 and the
(* use_dsp = "yes" *) attribute from comb stages 1-4. The combs are
gated by data_valid_comb_pipe (fires once every 4 clk_400m cycles
post-decimation), so a multicycle path of 4 -setup / 3 -hold scoped
to the comb registers in xc7a50t_ftg256.xdc gives STA 10 ns of slack
for fabric carry-chain to close 28-bit subtracts comfortably.
Pipeline depth and bit-widths unchanged: the new fabric model mirrors
the prior CREG+AREG+BREG+PREG structure exactly, so data_valid_comb_0_out
alignment and downstream stages 1-4 see bit-identical samples. CIC
behavioral simulation model now lives outside the SIMULATION ifdef
branch (used unconditionally) since there is no longer a synthesis-only
DSP48E1 to replace.
50T post-impl results (Vivado 2025.2):
DSPs: 80 → 70 / 120 (66.7% → 58.3%, freed 10)
LUTs: 22114 / 32600 (67.8%)
BRAM: 55.5 / 75 (74.0%, unchanged)
adc_dco_p WNS: +0.022 ns → +0.906 ns (margin improved)
All clocks meet timing, 0 failing endpoints.
Local regression: 32/34 PASS — same as baseline; the two failures
(Receiver Integration, Matched Filter Chain) are pre-existing
RX-NEW-3 (FFT throughput) and unaffected by this change. Bit-exact
through DDC chain (NCO→CIC→FIR) and MF cosim verified.
Cumulative DSP savings today: 112 → 70 (freed 42), enough headroom
for Xilinx LogiCORE FFT Pipelined Streaming swap (~33 DSPs for the
3-instance matched-filter chain) with 17 DSPs to spare.
Re-group the 32-tap symmetric lowpass into 16 (D+A)*B operations using
the DSP48E1 pre-adder, exploiting coeff[k] == coeff[31-k]. Production
silicon (XC7A50T) drops from 112/120 DSPs (93.3%) to 80/120 (66.7%),
freeing the budget needed for the matched-filter FFT swap (RX-NEW-3).
Bit-exact contract preserved at non-saturating signal levels: DC=5000
→ 8847 and 45 MHz tone → ±16 LSB match the unfolded design and the
Python golden model. Throughput unchanged (1 sample/cycle, 100 MSPS);
latency +2 cycles for the pre-adder stage.
Saturation thresholds rebuilt via bit concatenation to dodge the
Verilog 32-bit-literal trap (1 <<< 34 silently wraps to 0, which
made the earlier symmetric draft assert positive saturation on all
non-negative accumulator values).
Local regression: 32/34 PASS — same as baseline; the two failures
(Receiver Integration, Matched Filter Chain) are pre-existing
RX-NEW-3 (FFT throughput) and unaffected by this change.
The header had two claims that "valid samples arrive every ~4 cycles" at
the FIR boundary. That is false in the production wiring: the CIC `_4x`
decimator turns clk_400m into a 100 M-pulse-per-second stream, then
cdc_adc_to_processing crosses that into clk_100m where dst_valid asserts
every cycle in steady state. The 4:1 ratio applies between the two clock
domains, not as further sub-sampling inside clk_100m.
This matters because the 32-tap coefficients were designed for the
25 MSPS rate the wrong comment described, but the FIR is actually being
driven at 100 MSPS. The cutoff sits 4x higher than intended; existing
tests pass because the 36-bit accumulator silently wraps on large
sustained inputs (see RX-NEW-3 in the project ledger).
Comment-only commit. No RTL behaviour change. Any future DSP-saving
rework — symmetric pre-adder, 4:1 fold, Xilinx FIR Compiler — needs a
designer call on whether to redesign coefficients for 100 MSPS, add a
real decimation stage to hit 25 MSPS, or keep the current accidental
behaviour.
The Group 3 (tone autocorrelation), Group 10 (golden DC autocorr), and
Group 11 (golden tone autocorr) tests asserted cap_max_abs > mean_abs * 2,
which is mathematically impossible for those stimuli regardless of FFT
precision:
- DC autocorrelation produces a constant-magnitude time-domain output
(peak/mean ≡ 1.0 by definition).
- Single-tone autocorrelation produces a constant-magnitude rotating
phasor; |I|+|Q| envelope varies in [|X|^2, sqrt(2)*|X|^2], so
peak/mean is bounded by ~1.41x.
Empirical RTL output ratios from this regression: DC=1.07x, Tone5=1.18x,
Chirp=3.14x, Impulse=2015x — confirming theory and confirming the FFT
engine is correct for narrow-spectrum inputs.
Replace each ">2x" check with mean>0 && peak<=mean*2 (flatness bound).
Still catches flat-zero output (mean=0) but admits the correct constant-
magnitude result.
Matched Filter Chain regression: 5 failures -> 2 failures.
- run_regression.sh: add frequency_matched_filter.v to PROD_RTL and RECEIVER_RTL
compile groups (was implicitly required after inline behavioural FFT in
matched_filter_processing_chain.v was removed); empty EXTRA_RTL with set -u
guards; bump Matched Filter Chain timeout to 600s.
- run_regression.sh: add two PHASE 3 tests — tb_rxb_latency_measure (chain
pipeline depth) and tb_rxb_fullchain_latency (multi-segment + chain).
- radar_receiver_final.v: replace dangling delayed_ref_i/q references (left
over from latency_buffer removal) with ref_chirp_real/imag.
- tb/tb_radar_receiver_final.v: chain-state debug uses production
collect_count/out_count signals instead of the deleted SIMULATION-only
fwd_in_count.
- tb/tb_rxb_latency_measure.v: add explicit [PASS]/[FAIL] markers around the
2007..2107 cycle expected-latency window.
FPGA — RX chain
matched_filter_multi_segment.v: drop the gratuitous /4 scaling on
DDC sign-extended input (was ddc_i[17:2] + ddc_i[1]); use
ddc_i[15:0] directly. fft_engine has INTERNAL_W=32 with
saturating 16-bit output, so full 16-bit input is safe. Restores
~12 dB of MF input dynamic range.
radar_receiver_final.v: remove latency_buffer (count-N-pulses-then-
prime FIFO that left frame 1 with all-zero ref). Replaced with
a single-FF alignment register on ref_i/ref_q that matches the
1-FF stage multi_segment ST_PROCESSING uses on adc_data.
Verified by tb/tb_rxb_fullchain_latency.v — autocorrelation peak
at bin 0 with peak/mean ~88x.
doppler_processor.v / mti_canceller.v / cfar_ca.v /
range_bin_decimator.v / radar_receiver_final.v / radar_system_top.v
/ usb_data_interface_ft2232h.v: switch port and parameter widths
from RP_NUM_RANGE_BINS / RP_RANGE_BIN_BITS (always 512 / 9-bit)
to RP_MAX_OUTPUT_BINS / RP_RANGE_BIN_WIDTH_MAX (auto-scales:
50T 512 / 9-bit, 200T 4096 / 12-bit). Unblocks 200T 20 km mode
at the RX module boundary; USB wire-protocol extension still
pending.
radar_receiver_final.v: doppler_frame_done_prev reset value 0 -> 1
to prevent false done pulse on cycle 1 when level signal is
HIGH at reset.
matched_filter_processing_chain.v: delete the broken `ifdef
SIMULATION inline behavioural FFT (482 lines removed). It
produced wrong-bin peaks and 100-1000x weak magnitudes. Chain
now uses production fft_engine.v + frequency_matched_filter.v
in both iverilog and Vivado. Iverilog tests are ~38x slower per
chain pass but produce correct results. Misleading "OK with
Xilinx IP" comments at three test sites updated since the FFT
is in-house, not an IP placeholder.
FPGA — testbenches
tb/tb_rxb_latency_measure.v (new): measures chain internal pipeline
depth (~2057 cycles, chirp-agnostic).
tb/tb_rxb_fullchain_latency.v (new): full-chain autocorrelation
verification — drives ddc with the same chirp samples the loader
serves as ref, finds peak position and peak/mean.
tb/tb_matched_filter_processing_chain.v: wait timeouts bumped
50000 -> 500000 cycles to accommodate production FFT pipeline.
MCU
main.cpp checkSystemHealthStatus: latch system_emergency_state on
the error_count > 10 path so the SAFE-MODE blink loop in main()
actually engages (was bypassed because predicate was false).
main.cpp: move FPGA reset BEFORE the if(PowerAmplifier) block so
adar_tr_x is driven LOW (RX commanded externally) before PA Vdd
reaches 22 V. Old reset block at the original location removed.
main.cpp MX_GPIO_Init: add GPIO_PIN_12 (FPGA reset) to the
explicit WritePin(LOW) list so the safe initial state is no
longer implicit.
main.cpp checkSystemHealth: rate-limit ADAR1000
verifyDeviceCommunication (HAL_Delay 1ms x 4 devices = 4 ms
blocking SPI burst per main-loop iteration) from every-loop to
every 2 s. readTemperature stays per-loop so over-temp
detection latency is unchanged.
USBHandler.cpp processSettingsData: dispatch threshold bumped
74 -> 82 (matches parser minimum); buffer drained after parse
attempt (slide remaining bytes left) so a false END find no
longer sticks the buffer until 256-byte overflow.
GUI
radar_protocol.py: NUM_RANGE_BINS 64 -> 512 (matches FPGA
RP_NUM_RANGE_BINS); NUM_CELLS 2048 -> 16384.
radar_protocol.py _ingest_sample: honor FPGA frame_start bit for
resync after a USB drop; capture range_profile[rbin] once per
range bin at dbin == 0 (FPGA emits the same range_i/range_q for
all 32 Doppler cells of a given range bin; previous accumulator
inflated the profile 32x).
v7/models.py RadarSettings: range_resolution 24 -> 6 m (matches
c/(2*100MHz)*4); max_distance and coverage_radius 1536 -> 3072 m;
map_size 2000 -> 4000.
v7/models.py WaveformConfig: n_range_bins 64 -> 512, fft_size
1024 -> 2048, decimation_factor 16 -> 4.
GUI_V65_Tk.py: _RANGE_PER_BIN math and stale "~24 m / ~1536 m"
comments updated.
test_v7.py: assertion values updated to match new defaults.
Tests
test_ddc_cosim_fuzz.py: remove unused os/tempfile imports, wrap
three long lines for ruff E501 compliance.
Two bugs fixed recently had no tests that would have failed before the
fix. Add direct regressions so either cannot silently return:
1. tb_chirp_controller Group 3b (multi-frame, C-3): run a second full
frame back-to-back after DONE and assert chirp_counter returns to 0,
frame 2 reaches GUARD_TIME after exactly CHIRP_MAX/2 long chirps,
and frame 2 reaches DONE. Before the fix, chirp_counter held at
CHIRP_MAX after frame 1, the LONG_LISTEN -> GUARD guard (=CHIRP_MAX/2-1)
never matched, and frame 2 ran extra chirps until the 6-bit counter
wrapped — these checks fail loudly if that regresses.
2. tb_usb_data_interface frame-sync width + value pins: assert
$bits(uut.sample_counter) >= 15 and uut.NUM_CELLS == 15'd16384.
Protects against reintroducing the 12-bit / 2048-cell constants
that fired 8 false frame-start markers per real 512 x 32 frame.
Regression: 32/32 PASS; USB TB 89 -> 91 checks.
C-3: plfm_chirp_controller_enhanced never reset chirp_counter when the
frame completed. Counter sat at CHIRP_MAX after frame 1, so the
LONG_LISTEN -> GUARD transition guard (== CHIRP_MAX/2-1) never matched
correctly on subsequent frames and frame 2+ ran extra chirps until the
6-bit counter wrapped. Reset chirp_counter in the DONE state.
S-2: Replace hardcoded CHIRP_MAX = 32 with RP_CHIRPS_PER_FRAME from
radar_params.vh so the TX FSM tracks the single source of truth.
S-1: Correct misleading labels in tb_system_e2e G14.1-G14.3. Per
radar_params.vh the range_mode encoding is 2'b00 = 3 km, 2'b01 =
long-range, 2'b10/2'b11 = reserved. The TB strings previously called
2'b01 "short" and 2'b10 "long", which is inverted and inconsistent
with the RTL comments in radar_mode_controller.v.
Regression: 32/32 PASS.
usb_data_interface.v NUM_CELLS was still 12'd2048 (64 range x 32 doppler)
from the pre-2048-FFT architecture. With 512 range bins x 32 Doppler, the
12-bit counter wrapped every 2048 packets and the host received 8 false
frame-start markers per real frame via the sample_counter==0 bit packed
into the detection byte. Widen counter to 15 bits and set NUM_CELLS to
16384. Sister file usb_data_interface_ft2232h.v was already correct.
Remove three stale testbenches hardcoded to the old 1024-pt / 64-bin
architecture (tb_mf_chain_synth, tb_fullchain_mti_cfar_realdata,
tb_range_fft_realdata). Equivalent current-architecture coverage already
exists in tb_matched_filter_processing_chain, tb_fullchain_realdata,
tb_fft_engine, tb_multiseg_cosim, and tb_mf_cosim.
rx_final_doppler_out.csv is written by tb_radar_receiver_final.v on
every run via $fopen — it is a test-run artifact, not an oracle. It
was mistakenly tracked in an earlier commit, causing unnecessary
churn on every sim. Remove from the index and ignore going forward.
Also ignore stray a.out from iverilog one-shot compiles.
Golden references (.hex, .mem, doppler_golden_py_*.csv) remain
tracked — they are load-bearing oracles used by MF / Doppler /
receiver cosim testbenches.
RTL (P0 pre-bringup findings R-1/R-2/R-3/R-5/R-6):
- mti_canceller: add use_long_chirp input and waveform-boundary mute
so the long->short transition in mode 01 no longer subtracts across
heterogeneous waveforms (R-1). Prev buffer is overwritten in-flight
at the boundary so the next same-waveform chirp subtracts cleanly.
- ad9484_interface_400m: 2FF sync of mmcm_locked into the 400 MHz
domain before gating reset_n_gated (R-6).
- cic_decimator_4x_enhanced: correct max_fanout narrative (R-3).
- ad9484_interface_400m: strip stale pblock comment, note 3.0 ns
max_delay instead (R-2).
- mti_canceller / doppler_processor: 200T-20km WARNING banners
flagging the broken 4096-bin path (R-5). 9-bit BRAM address aliases
silently until rewritten.
- adc_clk_mmcm.xdc: relax set_max_delay from 2.700 -> 3.000 ns,
closes WNS with headroom on 50T build.
- radar_receiver_final: wire use_long_chirp into mti_inst.
Architecture-bump finalization (2048-pt range FFT, 512 range bins,
32 Doppler bins -> 16384 output cells per frame):
- tb/cosim/radar_scene.py: FFT_SIZE 1024 -> 2048, RANGE_BINS 64 -> 512.
- tb/gen_mf_golden_ref.py: N 1024 -> 2048.
- Regenerate all affected hex goldens (MF cases 1-4, Doppler inputs
+ py goldens, receiver integration golden_doppler.mem 2048 -> 16384).
- tb_radar_receiver_final: widen range_bin_out 6 -> 9 bits, bump
GOLDEN_ENTRIES 2048 -> 16384, expand bitmaps/arrays to 512 bins,
update all check messages and thresholds.
- tb_mti_canceller, tb_fullchain_mti_cfar_realdata: tie/pass
use_long_chirp so compile still works after RTL port add.
Test-suite hardening (coverage audit findings):
- tb_mti_canceller T12: 10 new assertions exercising R-1 waveform-
boundary mute across a long/long/short/short/long sequence. Catches
a regression that re-enables subtraction across the boundary.
- tb_fir_lowpass: replace tautological check(1'b1, ...) on coefficient
symmetry with a real hierarchical check coeff[k]===coeff[31-k];
replace always-pass overflow check with a well-driven (not X/Z)
assertion on filter_overflow.
- tb_matched_filter_processing_chain: replace three always-pass peak-
bin placeholders with peak-to-mean-|out| > 2x ratio checks (catches
flat/zero output that the old tautologies silently accepted).
- tb_cdc_modules M2: replace always-pass narrow-pulse check with a
well-defined-output assertion on the synchronizer.
- tb_nco_400m: replace always-pass freq-switch check with a swing +
no-X assertion across 200 post-switch samples.
- tb_system_e2e G12.1: replace check(1, ...) with test_num > 20 so
it catches a stalled TB that skipped prior groups.
- tb_multiseg_cosim TEST 4: replace always-pass placeholder with a
bitmap that asserts segment_request visited all 4 values.
- tb_mf_chain_synth and tb_fullchain_mti_cfar_realdata: add DEPRECATED
headers plus \$fatal guards (ifndef ALLOW_STALE_*) so they cannot
be silently re-enabled in CI with stale 1024-bin goldens against
current 2048-pt RTL.
Regression: 32 passed, 0 failed. MTI TB grew 30 -> 39 checks;
receiver integration grew 17 -> 18 checks with 16384/16384 golden
match at tolerance +/- 2 LSB.
Hand-merged files modified on both fix/pre-bringup-audit-p0 and
feat/fft-2048-upgrade. Wave 1 (commit 60e49c7) took 20 files from fft
verbatim; this wave resolves the overlap.
- run_regression.sh: 3-way merge. Adopts fft's ${RECEIVER_RTL[@]} array
refactor and drops the self-blessing golden pair from p0. Skip count
bumped to 5.
- usb_data_interface.v (FT601/200T): p0 FSM + clock-loss watchdog kept
wholesale; widened stream_control 3 -> 6 bits to carry fft's extended
mode bits through the CDC sync chain and the 0xFF status word.
- mti_canceller.v: fft's BRAM-inferred 512-range-bin implementation as
the base, with p0's F-6.3 saturation counter grafted onto the d1
pipeline stage. Overflow detection uses the top-two-bits disagreement
on diff_{i,q}_full (DATA_WIDTH+1 signed).
- radar_receiver_final.v: fft's 2048-pt / 512-bin structure + p0
diagnostic plumbing (ADC overrange sticky+CDC, DDC diagnostics,
tx_frame_start edge detector replacing chirp_counter frame sync,
mti_saturation_count, range_decim_watchdog).
- radar_system_top.v: clean 3-way merge, orthogonal regions
(+38 / -27).
- usb_data_interface_ft2232h.v (FT2232H/50T): fft's per-frame bulk BRAM
rewrite kept wholesale. Ported two p0 items that are orthogonal to
the write FSM:
* ft_clk-loss watchdog (heartbeat + 2FF ASYNC_REG sync + 16-bit
timeout) ORed into a 2FF sync'd ft_effective_reset_n for the FSM.
* rd_cmd_complete flag so RD_DEASSERT can distinguish a legitimate
3-byte completion from an ft_rxf_n abort that also zeros
rd_byte_cnt.
Deliberately NOT taken from 2401f5f: cic_decimator_4x_enhanced.v and
ddc_400m.v reset-strategy changes. Those conflict with p0's shipped
registered-sync-reset + max_fanout=25 distribution, which is already
timing-clean on the production build.
adc_or_p (overrange pin, added in commit 70067c6 for audit finding F-0.1)
uses the same IBUFDS→BUFIO source-synchronous capture topology as the 8
data pins adc_d_p[*]. STA reports identical -1.913 ns hold on this path
for the same reason (clock insertion ~4.0 ns via BUFIO vs data IBUFDS
~0.9 ns). External PCB layout guarantees hold, not FPGA clock tree.
Extends the existing adc_d_p[*] false_path waiver to cover adc_or_p.
Post-route now clean: WNS +0.034 ns, WHS positive.
Previous output_delay of 11.667 ns was a synthetic back-calculation
(period − 5 ns), not a datasheet number. It over-constrained FPGA
launch by ~8 ns vs the actual FT2232H 245-Sync FIFO setup requirement.
Per FTDI TN_167:
- t_su (data to CLKOUT rising): 3.5 ns (was 11.667 — too tight)
- t_h (data hold after CLKOUT): 1.0 ns (was 0.0 — no hold check)
- t_co (CLKOUT to data valid): 10.0 ns (was 9.667 — close)
- t_coh (CLKOUT to data hold): 0.5 ns (was 0.0 — no hold check)
NB: values must be verified against the exact TN_167 revision in use
before shipping. If the engineer's revision differs, numbers change
but the direction (big relaxation of output_delay_max) is correct.
The previous attempt put BUFIO inside u_core/gen_ft_bufr, but the pad
(ft_clkout) and its inferred IBUF live at the top wrapper level. Vivado
shape-packs IBUF↔BUFIO into the same IOB tile, and it couldn't do that
across the wrapper→u_core hierarchy boundary — producing CRITICAL
WARNING [12-1411] "Illegal to place BUFIO on TIEOFF site" and WNS=-5.737
(worse than the CLOCK_DEDICATED_ROUTE=FALSE baseline).
Fix: instantiate IBUF+BUFIO+BUFR explicitly in radar_system_top_50t.v
and pass the BUFR output into u_core.ft601_clk_in. radar_system_top.v
now does a pass-through wire assign for USB_MODE=1 (no BUFG) so the
clock net doesn't get double-buffered.
C4 is an SRCC pin (IS_CLK_CAPABLE=1, IS_MASTER=0 in the Vivado device
model), not an MRCC as earlier comments claimed. SRCC cannot drive BUFG
through dedicated routing, so the previous CLOCK_DEDICATED_ROUTE=FALSE
override forced fabric routing and burned ~5 ns on the ft_clkout path
(WNS -5.362 ns in the d36a4c9 build).
Swap to BUFIO + BUFR for USB_MODE=1 (50T/FT2232H): SRCC → BUFIO → BUFR
is the standard 7-series path for regional clock distribution. All
ft_clkout-domain logic (FT2232H FSM, toggle CDCs, USB FIFO flops) is
contained in bank 35 / one clock region, so regional distribution is
sufficient. USB_MODE=0 (200T/FT601) keeps the BUFG because D17 is a
proper MRCC pin.
Removed CLOCK_DEDICATED_ROUTE=FALSE from both the XDC and the build
script — no longer needed with dedicated BUFIO/BUFR routing.
A: cic_decimator_4x_enhanced.v reset_h max_fanout 50→25. More replicas
mean each drives fewer DSP48 RSTB loads, letting Vivado place each
closer to its consumers. Targets the rep__24 → comb_reg[4]/RSTB path
that failed clk_mmcm_out0 intra by -10 ps (1.4 ns of pure routing).
B: adc_clk_mmcm.xdc BUFIO↔BUFG max_delay 2.500→2.700 ns. The 2.5 ns
target was tighter than achievable for the IDDR (ILOGIC) → FDRE (fabric
SLICE) re-registration. The effective window is the BUFIO↔BUFG phase
relationship (not the clock period), so 2.7 ns remains safe. Fixes the
adc_dco_p→clk_mmcm_out0 inter path -113 ps failure on lane 7.
Post-route WNS = -5.355 ns on path group ft_clkout, net
u_core/gen_ft2232h.usb_inst/ft_data_TRI[0]_repN_1
FT2232H 245-sync FIFO input setup (t_su = 11.667 ns on a 16.667 ns
CLKOUT) leaves the FPGA only ~5 ns from clock edge to pad. Without
IOB=TRUE, the output / tristate FFs live in fabric and FF→OBUFT
routing eats 2–3 ns, forcing Vivado to replicate the tristate
driver (ft_data_TRI[*]_repN) and still miss timing.
The FSM in usb_data_interface_ft2232h.v already registers
ft_data_out / ft_data_oe / ft_{rd,wr,oe}_n at the output boundary
in the ft_clk domain, so packing them into the IOB is safe with
no RTL change.
Build-blocking fixes surfaced by gpu-server synth:
1. radar_system_top_50t.v wrapper was missing adc_or_p/n ports and the
u_core instantiation left them unconnected. Every XDC line in the 50T
anchor block (PACKAGE_PIN M6/N6, IOSTANDARD, DIFF_TERM, set_input_delay)
therefore matched no ports and emitted CRITICAL WARNINGs, leaving the
overrange pin effectively tied off. Added the two inputs and wired them
through to the core.
2. adc_clk_mmcm.xdc used foreach / unset — Vivado's XDC parser only
accepts a restricted Tcl subset and rejected them as
[Designutils 20-1307]. Moved the clk_mmcm_out0 ↔ USB-clock false paths
into each board XDC (ft_clkout for 50T, ft601_clk_in for 200T) where
the clock name is already known.
A new SCENARIO_FUZZ branch in tb_ddc_cosim.v accepts +hex / +csv / +tag
plusargs so an external runner can pick stimulus and output paths per
iteration. The three path registers are widened to 4 kbit each so long
temp-directory paths (e.g. /private/var/folders/...) do not overflow
the MSB and emerge truncated — a real failure mode caught while writing
this runner.
test_ddc_cosim_fuzz.py is a pytest-driven fuzz harness:
- Generates a random plausible radar scene per seed (1-4 targets with
random range/velocity/RCS/phase, random noise level 0.5-6.0 LSB
stddev) via radar_scene.generate_adc_samples, fully deterministic.
- Compiles tb_ddc_cosim.v once per session (module-scope fixture),
then runs vvp per seed.
- Asserts sample-count bounds consistent with 4x CIC decimation,
signed-18 range on every baseband I/Q word, and non-zero output
(catches silent pipeline stalls).
- Ships with two tiers: test_ddc_fuzz_fast (8 seeds, default CI) and
test_ddc_fuzz_full (100 seeds, opt-in via -m slow) matching the
audit ask.
Registers the "slow" marker in pyproject.toml for the 100-seed opt-in.
G9B adds a 4-iteration reset sweep on top of the existing e2e harness:
- Reset is injected at four offsets (3/7/12/18 us) into a steady-state
auto-scan burst, with mixed short/long hold durations (20-120 clk_100m)
to exercise asynchronous assert paths through the FSM + CDCs.
- Each iteration asserts: system_status drops to 0 during reset,
new_chirp_frame resumes post-release, and obs_range_valid_count
advances — proving the full DDC->MF chain recovers, not just the
transmitter FSM.
The stub and three existing testbenches are updated to drive the new
adc_or_p/n ports tied to 1'b0/1'b1, matching the F-0.1 RTL change.
The AD9484 OR (overrange) LVDS pair is routed on the 50T main board to
xc7a50t-ftg256 bank-14 pins M6/N6 but was previously left unconnected at
the top level. Plumb it through the full stack so saturation at the raw
ADC boundary shows up in the existing overflow aggregation:
- ad9484_interface_400m: add adc_or_p/n inputs, IBUFDS + IDDR capture of
both phases in the BUFIO domain, re-register into the clk_400m BUFG
domain, OR rise|fall into adc_overrange_400m output.
- radar_receiver_final: stickify adc_overrange_400m in clk_400m, CDC to
clk_100m via a 2FF ASYNC_REG chain (same reasoning as F-1.2's
cdc_cic_fir_overrun — single-bit, latched low→high, GPIO-class
diagnostic), OR into the existing ddc_overflow_any aggregation.
- radar_system_top: expose adc_or_p/n top-level ports and pass through.
- xc7a50t_ftg256.xdc: anchor M6/N6 as LVDS_25 DIFF_TERM, with the same
DCO-relative input-delay constraints as adc_d_p[*].
- xc7a200t_fbg484.xdc: IOSTANDARD/DIFF_TERM set; PACKAGE_PIN left as a
documented TODO — the 200T dev-board schematic has not been checked
and the 200T build will need the anchor filled in before place/route.
Addresses the remaining actionable items from
docs/DEVELOP_AUDIT_2026-04-19.md after commit 3f47d1e.
XDC (dead waivers — F-0.4, F-0.5, F-0.6, F-0.7):
- ft_clkout_IBUF CLOCK_DEDICATED_ROUTE now uses hierarchical filter;
flat net name did not exist post-synth.
- reset_sync_reg[*] false-path rewritten to walk hierarchy and filter
on CLR/PRE pins.
- adc_clk_mmcm.xdc ft601_clk_in references replaced with foreach-loop
over real USB clock names, gated on -quiet existence.
- MMCM LOCKED waiver uses REF_PIN_NAME filter instead of the
previously-missing u_core/ literal path.
CDC (F-1.1, F-1.2, F-1.3):
- Documented the quasi-static-bus stability invariant above the
FT601 cmd_valid toggle block.
- cdc_adc_to_processing gains an `overrun` output; the two CIC->FIR
instances feed a sticky cdc_cic_fir_overrun flag surfaced on
gpio_dig5 so silent sample drops become visible to the MCU.
- Removed the dead mixers_enable synchronizer in ddc_400m.v; the _sync
output was unused and every caller ties the port to 1'b1.
Diagnostics (F-6.4):
- range_bin_decimator watchdog_timeout plumbed through receiver
and top-level, OR'd into gpio_dig5.
ADAR (F-4.7):
- delayUs() replaced with DWT cycle counter; self-initialising
TRCENA/CYCCNTENA, overflow-safe unsigned subtraction.
Regression: tb_cdc_modules.v 57/57 passes under iverilog after
the cdc_modules.v change. Remote Vivado verification in progress.
Addresses findings from docs/DEVELOP_AUDIT_2026-04-19.md:
P0 source-level:
- F-4.3 ADAR1000_Manager::adarSetTxPhase now writes REG_LOAD_WORKING
with LD_WRK_REGS_LDTX_OVERRIDE (0x02) instead of 0x01. Previous value
toggled the LDRX latch on a TX-phase write, so host TX phase updates
never reached the working registers.
- F-6.1 DDC mixer_saturation / filter_overflow / diagnostics were deleted
at the receiver boundary. Now plumbed to new outputs on
radar_receiver_final (ddc_overflow_any, ddc_saturation_count) and
aggregated into gpio_dig5 in radar_system_top. Added mark_debug
attributes for ILA visibility. Test/debug inputs tied low explicitly.
- F-0.8 adc_clk_mmcm.xdc set_clock_uncertainty: removed invalid -add
flag (Vivado silently rejected it, applying zero guardband). Now uses
absolute 0.150 ns which covers 53 ps jitter + ~100 ps PVT margin.
P1:
- F-4.2 adarSetBit / adarResetBit reject broadcast=ON — the RMW sampled
a single device but wrote to all four, clobbering the other three's
state.
- F-4.4 initializeSingleDevice returns false and leaves initialized=false
when scratchpad verification fails; previously marked the device
initialized anyway so downstream PA enable could drive a dead bus.
- F-6.2 FIR I/Q filter_overflow ports, previously unconnected, now OR'd
into the module-level filter_overflow output.
- F-6.3 mti_canceller exposes 8-bit saturation counter. Saturation was
previously invisible and produces spurious Doppler harmonics.
Verification:
- 27/27 iverilog testbenches pass
- 228/228 pytest pass (cross-layer contract + cosim)
- MCU unit tests 51/51 + 24/24 pass
- Remote Vivado 2025.2 build: bitstream writes; 400 MHz mixer pipeline
now shows WNS -0.109 ns which MATCHES the audit's F-0.9 prediction
that the design only closed because F-0.8's guardband was silently
dropped. ft_clkout F-0.9 remains a show-stopper (requires MRCC pin
move), tracked separately.
Not addressed in this PR (larger scope, follow-up tickets):
F-0.4, F-0.5, F-0.6, F-0.7, F-0.9, F-1.1, F-1.2, F-2.2, F-3.2, F-4.1,
F-4.7, F-6.4, F-6.5.
Three conflicts — all resolved in favor of develop, which has a more
refined version of the same work this branch introduced:
- radar_system_top.v: develop's cleaner USB_MODE=1 comment (same value).
- run_regression.sh: develop's ${SYSTEM_RTL[@]} refactor + added
USB_MODE=1 test variants.
- tb/radar_system_tb.v: develop's ifdef USB_MODE_1 to dump the correct
USB instance based on mode.
The 400 MHz reset fan-out fix (nco_400m_enhanced, cic_decimator_4x_enhanced,
ddc_400m) and ADAR1000 channel-indexing fix remain intact on this branch.
Replace direct !reset_n async sense with a registered active-high reset_h
(max_fanout=50) in nco_400m_enhanced, cic_decimator_4x_enhanced, and
ddc_400m. The prior single-LUT1 / 700+ load net was the root cause of
WNS=-0.626 ns in the 400 MHz clock domain on the xc7a50t build. Vivado
replicates the constrained register into ≈14 regional copies, each driving
≤50 loads, closing timing at 2.5 ns.
Change radar_system_top default USB_MODE from 0 (FT601) to 1 (FT2232H).
FT601 remains available for the 200T premium board via explicit parameter
override; the 50T production wrapper already hard-codes USB_MODE=1.
Regression: add usb_data_interface_ft2232h.v to PROD_RTL lint list and
both system-top TB compile commands; fix legacy radar_system_tb hierarchical
probe from gen_ft601.usb_inst to gen_ft2232h.usb_inst.
Golden reference files (rtl_bb_dc.csv, rx_final_doppler_out.csv,
golden_doppler.mem) regenerated to reflect the +1-cycle registered-reset
boundary behaviour; Receiver golden-compare passes 18/18 checks.
All 25 regression tests pass (0 failures, 0 skipped).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Resolve cross-layer AGC control mismatch where opcode 0x28 only
controlled the FPGA inner-loop AGC but the STM32 outer-loop AGC
(ADAR1000_AGC) ran independently with its own enable state.
FPGA: Drive gpio_dig6 from host_agc_enable instead of tied low,
making the FPGA register the single source of truth for AGC state.
MCU: Change ADAR1000_AGC constructor default from enabled(true) to
enabled(false) so boot state matches FPGA reset default (AGC off).
Read DIG_6 GPIO every frame with 2-frame confirmation debounce to
sync outerAgc.enabled — prevents single-sample glitch from causing
spurious AGC state transitions.
Tests: Update MCU unit tests for new default, add 6 cross-layer
contract tests verifying the FPGA-MCU-GUI AGC invariant chain.
Jason