You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The builder was wiring the AD9371 reset/sysref_req lines to GPIOs
130/136 — which is the ZCU102+ADRV9009 pinout, not ZC706+AD9371.
On bq the Mykonos driver logged ``ad9371 spi1.1: RESET Failed``
during probe, then the jesd204 FSM bounced off ``link_pre_setup``
with error -19 (ENODEV) in a retry loop, and the TPL DMA never
streamed. Fix: move all three GPIO numbers to match the Kuiper
``zc706-adrv9371`` reference DT on bq's SD card:
- ``trx_reset_gpio`` 130 → 106 (AD9371 reset)
- ``trx_sysref_req_gpio`` 136 → 112 (AD9371 SYSREF_REQ)
- new ``ad9528_reset_gpio`` 113 (AD9528 reset — wasn't
emitted before).
Additions to support ``ad9528_reset_gpio``:
- ``ADRV937xBuilder`` now passes a ``_GpioLine(prop="reset-gpios",
…)`` into ``AD9528_1(gpio_lines=[…])`` so the ad9528 node
renders ``reset-gpios = <&gpio0 113 0>;`` alongside its channel
subnodes.
- ``_ADRV9009_BOARD_ALLOWED_KEYS`` + ``_ADRV9009_BOARD_INT_KEYS``
in ``adidt/xsa/profiles.py`` gain ``ad9528_reset_gpio`` so the
profile-JSON loader doesn't reject it.
- ``adidt/xsa/profiles/adrv937x_zc706.json`` defaults updated.
- System-API side (``adidt/eval/adrv937x_fmc.py``) updated in
lockstep so the dts-parity test stays green.
- Test cfg (``test/hw/test_adrv9371_zc706_hw.py``) updated too —
it was overriding the profile defaults with the old 130/136
values.
Fixture ``test/devices/fixtures/adrv9371_zc706_xsa_reference.dts``
regenerated from the new builder output. All 449 unit tests
still pass.
0 commit comments