Commit Graph
3 Commits
Author SHA1 Message Date
Jason 49c9aa28ad Fix Bug #11 (platform SPI transmit-only), FPGA B2 (chirp BRAM migration), FPGA B3 (DSP48 pipelining)
Bug #11: platform_noos_stm32.c used HAL_SPI_Transmit instead of
HAL_SPI_TransmitReceive — reads returned garbage. Changed to in-place
full-duplex. Dead code (never called), fixed per audit recommendation.
Test added: test_bug11_platform_spi_transmit_only.c. Mock infrastructure
updated with SPI spy types. All 11 firmware tests pass.

FPGA B2: Migrated long_chirp_lut[0:3599] from ~700 lines of hardcoded
assignments to BRAM with (* ram_style = "block" *) attribute and
$readmemh("long_chirp_lut.mem"). Added sync-only read block for proper
BRAM inference. 1-cycle read latency introduced. short_chirp_lut left
as distributed RAM (60 entries, too small for BRAM).

FPGA B3: Added BREG (window_val_reg) and MREG (mult_i_raw/mult_q_raw)
pipeline stages to doppler_processor.v. Eliminates DPIP-1 and DPOP-2
DRC warnings. S_LOAD_FFT retimed: fft_input_valid starts at sub=2,
+1 cycle total latency. BREG primed in S_PRE_READ at no extra cost.
Both FPGA files compile clean with Icarus Verilog.
2026-03-19 10:31:16 +02:00
Jason 397969348e Fix all 8 firmware bugs with regression tests
Bugs fixed in adf4382a_manager.c:
- Bug #1: Move initialized=true before sync setup, propagate sync failure
- Bug #3: Implement TriggerTimedSync with sw_sync pulse (was no-op)
- Bug #5: Replace GPIO-only placeholder with TIM3 PWM for DELADJ
- Bug #7: Correct GPIOG pin definitions to match CubeMX (pins 6-15)

Bugs fixed in main.cpp:
- Bug #2: Remove pre-reset ad9523_setup() call (keep only post-reset)
- Bug #4: Move init error check before phase shift calls
- Bug #6: Fix timer variable (last_check -> last_check1) in temp block
- Bug #8: Uncomment uart_print/uart_println debug helpers

Test harness updates:
- All 8 tests rewritten to assert correct post-fix behavior
- Added TIM PWM mock (SPY_TIM_PWM_START/STOP/SET_COMPARE)
- Added mock_adf4382_set_timed_sync_retval for failure injection
- Updated shims and Makefile for new test dependencies
- All 8 tests pass: make clean && make test -> 8/8 passed
2026-03-19 09:42:59 +02:00
Jason 28a66889ad Add MCU firmware test harness with 8 bug-confirming tests
Complete test infrastructure for the observe-before-fix methodology:

- stm32_hal_mock: HAL stub types + spy/recording ring buffer (512 entries)
- ad_driver_mock: ADF4382/AD9523 mock drivers with configurable returns
- 9 shim headers redirecting real #includes to mock types
- Makefile with individual (test_bug1..8) and aggregate (test) targets

All 8 tests pass, confirming:
  #1 Timed sync init ordering (SetupTimedSync before initialized=true)
  #2 AD9523 double setup (first call before reset release)
  #3 TriggerTimedSync no-op (prints messages, no HW action)
  #4 Phase shift before init error check
  #5 SetFinePhaseShift GPIO-only placeholder (no PWM)
  #6 Timer variable collision (last_check vs last_check1)
  #7 GPIO pin mapping conflict (manager.h vs CubeMX main.h)
  #8 uart_print/uart_println commented out
2026-03-19 09:28:19 +02:00