Hello! I was wondering if anyone has had any luck installing plymouth for prettier boot process and graphical disk decryption prompt. I’m on a311d on pocket reform, for context, but I’m curious if this works on other SoCs too!
My first stab was installing plymouth and plymouth-themes, and adding “splash” to /etc/default/flash-kernel, (and running sudo flash-kernel). This slightly changes the disk decryption prompt, including some errors, but it still seems to be all text-mode. It looks like this:
Done.
Begin: Mounting root file system ... Begin: running: /scripts/local-top ... meson-pcie fc000000.pcie: error: wait linkup timeout
debugfs: Directory 'ff800280.cec' with parent 'regmap' already present!
panel-mnt-pocket-reform ffd07000.dsi.0: error -ENOENT: cannot get reset-gpios 0
panel-mnt-pocket-reform ffd07000.dsi.0: error -ENOENT: cannot get dcdc-en-gpio
panfrost ffe40000.gpu: error -ENODEV: _opp_set_regulators: no regulator (mali) found
panel-mnt-pocket-reform ffd07000.dsi.0: [mnt pocket reform display] read register 12a: 0x77
mdio_bus mdio_mux-0.0: MDIO device at address 0 is missing.
Please unlock disk reform_crypt:_
and noticeably, typing the disk password produces asterisks on the prompt, so something is definitely changing.
If I uninstall plymouth and plymouth-themes, remove the “splash” parameter, and reboot, I no longer see any errors on the disk decryption prompt and typing in my disk decryption passphrase doesn’t produce any asterisks.
I do note that those error messages still appear in dmesg so I suspect they’re happening regardless of plymouth, but somehow installing plymouth is changing the text-mode prompt but not making Plymouth itself work? Not sure, curious if others have had luck
For some reason, I remember hearing that the reason plymouth wasn’t be using was that it had issues on the Reform. I would image that this follows with the Pocket’s imx based SoC. Don’t quote me on it, but I’m pretty sure that is what was discussed in the past.
Thanks for trying it out, interesting that there’s an actual stacktrace on at least one cpu module
On a311d, the kernel seems to switch to full resolution during boot in a way that I assumed was before hitting userspace, which I thought was kms, but now I’m realizing I’m not sure? I wish it were easier to look up device and driver specific feature support, curious if anyone has any general advice on this.