ZX908 with EC800M-CN-LA module — flash chip mismatch on QuecPython firmware images

[AngelFreak]

Hi — running into a flashing block on a ZX908 board and hoping to get
guidance from anyone who's hit the same wall.

Hardware

• Board: ZX908 PCBA
• Module per ATI: EC800MCN_LA (not the LE variant referenced in some
  threads here)
• Stock firmware: LTE01R06A04_VOLTE_C_SDK_A
• AROM BootROM version: 2021.10.16
• QSPI flash chip ID (read by Quectel's flasher in BootROM mode):
  manufacturer 0x20, device 0x4117 — Micron N25Q064 (8 MB)

What works

• AT+QDOWNLOAD=1 cleanly enters AROM BootROM (USB 2ecc:3004)
• amitv87/asr_aboot speaks the SMUX
  protocol against the 2021.10.16 BootROM correctly
  (verified with getvar:product, getvar:version, getvar:version-bootrom)
• preboot.img and flasher.img from the LF firmware load successfully

What fails

The official QuecPython image QPY_OCPU_V0004_EC800M_CNLF_FW (the closest
public variant) gets through preboot + flasher, then fails at the
partition step:

[INFO: QspiNor] Manufacturer ID: 0x20, Device ID: 0x4117, [DTR mode disable].
[ERR : QspiNor] Flash device is different with specified!
[cmd] < FAILparse flash information table failed

So partition.bin in the LF firmware declares a different flash chip than
what's on this board. Device safely refuses to write — no damage.

Question

Has anyone successfully flashed an EC800M-CN-LA variant ZX908? If so,
which firmware image did you use, and where did you obtain it? Quectel's
public download portal lists firmware for LE / LF / GB / GD suffixes only —
no LA.

Issue #2 mentions a ZX908 with the LE module that flashed cleanly. Curious
whether the LA-variant boards are a known issue and what the escape route is.

Options I'm considering:

1. Ask Quectel support for an LA-targeted QuecPython image
2. Patch partition.bin to declare the actual flash chip
3. Stay on stock C-SDK and run everything via AT commands

Any pointers very welcome. Happy to share more diagnostic data if useful.