Skip to content

Support for 06cb:00a2 Synaptics, Inc. Metallica MOH Touch Fingerprint Reader #170

Open

Description

@blackinitial

hope anyone success on 06cb:00a2 Synaptics, Inc. Metallica MOH Touch Fingerprint Reader or Thinkpad L390 Yoga or any reference ?

Activity

  1. qkyubit commented on Sep 20, 2026

    @qkyubit

    I have a 06cb:00a2 (Synaptics Metallica MOH) on a ThinkPad L390 and have got python-validity most of the way through init on it. Posting the details in case they're useful to anyone else with this chip.

    Environment

    • ThinkPad L390, Arch Linux
    • Sensor: 06cb:00a2 Synaptics, Inc. Metallica MOH Touch Fingerprint Reader
    • python-validity master, Python 3.14.7, pyusb 1.3.1
    • No Windows install on this machine (so I can't capture reference traffic myself)

    Most important finding: the ROM info is identical to a 009a apart from the serial.

    Comparing my cmd 01 response against the one posted in #272 for a 06cb:009a (T480s):

    009a:  0000f0b05e54a4000000060701300001 000091278b684d37 0023000000000100000003d10007
    00a2:  0000f0b05e54a4000000060701300001 0000ee464a7064eb 0023000000000100000003d10007
    

    Byte-for-byte identical except the 6-byte serial field in the middle. Same firmware version (060701), same sensor type. So MIS and MOH appear to be the same silicon and the same ROM, the distinction must live above this layer.

    Changes I made to get this far

    1. usb.py: added DEV_a2 = (0x06cb, 0x00a2) to SupportedDevices
    2. firmware_tables.py: copied the DEV_9a entries into FIRMWARE_URIS and FIRMWARE_NAMES verbatim (same installer, same sha512, same 6_07f_lenovo_mis_qm.xpfwext)
    3. blobs.py: added an 0x00a2 branch importing blobs_9a
    4. usb.py open_dev(): added self.dev.reset() + 1s sleep after self.dev = dev, before set_configuration()
    5. init.py open_common(): added a usb.cmd(b'\x01') before init_flash()

    On (4) and (5): without the wake command, get_flash_info (cmd 3e) fails with status 0304. With it, 3e returns a full 76-byte header + partition table and status 0000. Probing by hand, cmd 01 answers correctly on the first attempt and 3e answers correctly once 01 has been sent first, so this looks like a variant of the transient-busy behaviour described in #271/#272, but with a different status code.

    The firmware fetcher works unmodified once the table entries are added, it downloads nz3gf07w.exe and extracts the standard MIS blob, hash matches.

    Where it stops

    init_flash() now completes successfully (flash formatted, partition table written). It then fails in usb.send_init():

    File "validitysensor/usb.py", line 92, in send_init
        assert_status(self.cmd(init_hardcoded))
    Exception: Failed: 04be
    

    So every generic command works, and the one device-specific payload, init_hardcoded from blobs_9a, is rejected with 04be. Which is roughly what you'd expect if the MOH part needs its own configuration blob.

    Questions

    1. Does anyone recognise status 04be, or 0304 from get_flash_info?
    2. Does anyone with a 00a2 and a Windows install have the ability to capture the init traffic? What I'm missing is the init_hardcoded payload the Windows driver sends to this device. That's the one piece blocking further progress, and I can't produce it myself.
    3. Is init_hardcoded known to be a fixed per-device constant, or is it derived from something at runtime? If it's a constant, a single capture would likely be enough to add a blobs_a2.py.

    I'd be happy to run any experiments on this hardware and report back with my findings!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions