accelerometer is not working in redrix

[im5n]

My redrix is running on Win 11 23H2, and I believe that the accelerometer was working, but I can't remember which firmware version it was running on.
At the current versions, accelerometer is not working and the screen won't auto rotate.

[Zidrewndacht]

I'm having what seems to be the same issue. I posted on chrultrabook forums earlier believing it was a Windows issue, but it may actually be firmware, since neither Windows 11 LTSC 24H2 26100.7627 nor Fedora 43 LiveCD seem to "see" the tablet sensors (GOOG0006) at all in Redrix in this build while, based on a DriverIdentifier scan from someone in the past, it existed at some point ("Chromebook EC Sensor Hub" there).

I uploaded just cbmem in the forums. I'm attaching the full debug suite ran from Fedora LiveCD below instead. I can do other tests or gather more information as instructed if required. I have a SuzyQ if needed.

debug-logs-Redrix-2026-02-01_04h37m.tar.gz

[Zidrewndacht]

EC_ENABLE_TBMC_DEVICE is missing from
MrChromebox/coreboot/tree/MrChromebox-2512/src/mainboard/google/brya/variants/baseboard/brya/include/baseboard (which probably also explains someone else showing up recently in the forums with the same issue in a crota360) in the 2512 branch (somehow this file is from 3 years ago?) yet it was there in the 2509 branch

I don't have the right environment to compile coreboot from source (never did it before) to see if that's all there is to it right now, but, here's hope it is.

Edit: apparently this commit is missing from the 2512 branch.

[MrChromebox]

brya-based boards don't support TBMC at the EC level, so including the ACPI interface in earlier firmware was a mistake (and why Google doesn't do it for ChromeOS)

[Zidrewndacht]

@MrChromebox, what would be the correct mechanism for tablet sensors in these boards? VBTN (INT33D6) is also missing in my Redrix on 2512.

As there's some evidence from people that it was working in 2509, I'm considering flashing that version in my Redrix to get the debug logs (if it really ends up working in that version), so you can compare with the logs of 2512 I posted above. Would that help?

[MrChromebox]

@Zidrewndacht I will post a 25.12-based build with it added back tomorrow for you to test

[MrChromebox]

@Zidrewndacht try this and let me know if it fixes the issue

coreboot_edk2-redrix-mrchromebox_20260202.zip

[Zidrewndacht]

@MrChromebox yes, indeed this fixed it in both Fedora 43 LiveCD and Windows.

[im5n]

@Zidrewndacht try this and let me know if it fixes the issue

coreboot_edk2-redrix-mrchromebox_20260202.zip

this works in windows! Thanks a lot. BTW, any clue on auto disabling keyboard and trackpad in tablet mode?

[Zidrewndacht]

@im5n The keyboard already disables correctly in mine. Disabling trackpad in Windows, well, I read somewhere that it wasn't implemented in coolstar drivers (so it's "normal" that it stays working). I suppose it might be possible to do something ourselves to listen for the tablet mode event and disable/enable the device, but I don't know if someone did that already.

[im5n]

@Zidrewndacht thanks and i'll double check mine and try to gpt the trackpad thing

[literallyayden]

Hey there! I'm the person from this forum post. The issue seems to affect all brya mainboards in particular.

I recompiled a crota360 build of the latest mrchromebox firmware (2512.2 as of writing) with EC_ENABLE_TMBC_DEVICE in the baseboard file added. The sensors work perfectly fine and as expected like in previous versions following up to 2511/2512. (which is about when the sensors stopped working). I'm not entirely sure if it worked in 2510, though. Can't remember

Anyway, thanks a lot!

[xiaojianqaq]

Auto-rotation is now working. This UEFI firmware version will adds a new int33d6 device in Device Manager. Use this link to install the driver. Once the driver is installed, it will automatically detect the laptop flip status.