What is the status of this project?

[TheMojoMan]

First: Thank you so much for your ongoing fantastic work for the Xiaomi Pad 5 and now also for the 6S Pro!

Can you please give a short overview of what is already working on the 6S Pro and what is still missing?
[I would offer my help in testing but I do not own a 6S Pro (yet). I am also not 100% sure if I will buy one but if Linux would work on it, it might clear up my mind.]

Thanks again!!!

[map220v]

It currently has working:
UFS,
Type-C,
DisplayPort(2 lane),
Charger,
FuelGauge,
Display(144hz only),
Backlight,
GPU(mesa 24.2 and higher for OpenGL),
WiFi,
Bluetooth

And not working:
Touchscreen(it sends raw frames instead of regular x/y),
Microphones(will fix all of them after touchscreen),
Fingerprint,
Hall sensors(will try after touchscreen fix),
Sensors(untested),
Flash Light & RGB Led(will add after touchscreen),
Speakers(will try fixing after touchscreen, currently i2s clock doesn't seem to work),
Cameras(will try when cams support for sm8550 upstreamed)

[TheMojoMan]

That is so impressive!
Already useable as a full computer with keyboard and mouse. And media consumption should be possible as well if we could use bluetooth headphones.

It would be a real shame if Xiaomi really disables the ability to unlock the boot loader as I have recently read about.

[alghiffaryfa19]

@map220v what is the current status?

[map220v]

Touchscreen now partially works. Currently only one finger is supported and position detection is a bit inaccurate.
All Microphones are working, but in mono mode, so only one microphone can be used at time.
Hall sensor for magnetic cover works.
Flashlight and RGB Led are working.
All Speakers are working.
Analog headset/headphones maybe also works, if type-c audio jack detection is working.
Cameras work in sheng-6.13-camera-wip branch and in future sheng-6.14 branch.
Suspend works now.

UCM2 configs for Speakers and Microphones can be found here https://github.com/map220v/alsa-ucm-conf and firmwares are here https://github.com/map220v/sheng-firmware

[alghiffaryfa19]

@map220v xiaomi official keyboard work?

[map220v]

@map220v xiaomi official keyboard work?

It doesn't. There is no source code for that driver in sheng branch of Xiaomi repos.

[alghiffaryfa19]

Maybe this ?
https://github.com/MiCode/kernel_devicetree/blob/sheng-u-oss/qcom/sheng-sm8550.dtsi
@map220v

[alghiffaryfa19]

https://github.com/MiCode/kernel_devicetree/blob/c4e85e1569f8acbbfe8e6dd694623aa56e424c60/qcom/sheng-sm8550.dtsi#L760
https://t.me/pipa_mainline/413
pipa-mainline/linux@6301a92
pipa-mainline/linux@8384ccf
https://github.com/MiCode/Xiaomi_Kernel_OpenSource/blob/sheng-u-oss/drivers/power/supply/pogoChg/extSOC/nanosic_platform.c
https://github.com/MiCode/Xiaomi_Kernel_OpenSource/blob/29db5d592c5b577aa9d9611da988a5e071d100dd/arch/arm64/configs/vendor/sheng_GKI.config#L421

do the pipa and sheng have the same use of nanosic 803?

[map220v]

Tested nanosic from pipa-mainline with some changes, even with correct pin config and all needed regulators enabled it fails to read nanosic version with "regmap write cmd failed" error. Also on sheng nanosic driver seem to upload nanosic firmware after each boot, so it's probably required for it to work.

[alghiffaryfa19]

what about the ones in the sheng device tree that I sent?

[alghiffaryfa19]

I want to try adapting it for Sheng, but I haven't succeeded in unlocking the bootloader, HyperOS is quite difficult

[map220v]

what about the ones in the sheng device tree that I sent?

https://github.com/MiCode/Xiaomi_Kernel_OpenSource/blob/sheng-u-oss/drivers/power/supply/pogoChg/extSOC/nanosic_platform.c is for charging through pogo pin. The one that is for pogo pin keyboard is in vendor_opensource_touch-drivers repo I guess, but Xiaomi didn't shared it.
Edit: sheng's nanosic driver build path is "vendor/xiaomi/proprietary/keyboard/keyboard_driver/sheng/nanosic_driver/", so they didn't upload this driver on purpose.

[99degree]

good job man!

if you find touch driver not very functional, you might try below driver:
https://lore.kernel.org/all/20241015-nt36xxx-v1-0-3919d0bffee6@gmail.com/

or a more resent one at
https://github.com/99degree/linux/tree/nt36xxx_stable

[alghiffaryfa19]

good job man!

if you find touch driver not very functional, you might try below driver: https://lore.kernel.org/all/20241015-nt36xxx-v1-0-3919d0bffee6@gmail.com/

or a more resent one at https://github.com/99degree/linux/tree/nt36xxx_stable

support for nt36532e ?

[99degree]

good job man!
if you find touch driver not very functional, you might try below driver: https://lore.kernel.org/all/20241015-nt36xxx-v1-0-3919d0bffee6@gmail.com/
or a more resent one at https://github.com/99degree/linux/tree/nt36xxx_stable

support for nt36532e ?

not sure if it need a big code add. i am feeling positive with nt36532e.

my rewrite of nt36xxx driver is based on prev version of nvt driver and this newer nvt driver still compatible with nt36675 and add support for nt36532e so they might shared many common logic.