[PATCH v3 0/7] Fix RK3588 GPU domain

[Heiko Stübner]

Am Dienstag, 22. Oktober 2024, 17:41:45 CEST schrieb Sebastian Reichel:
> Hi,
> 
> I got a report, that the Linux kernel crashes on Rock 5B when the panthor
> driver is loaded late after booting. The crash starts with the following
> shortened error print:
> 
> rockchip-pm-domain fd8d8000.power-management:power-controller: failed to set domain 'gpu', val=0
> rockchip-pm-domain fd8d8000.power-management:power-controller: failed to get ack on domain 'gpu', val=0xa9fff
> SError Interrupt on CPU4, code 0x00000000be000411 -- SError
> 
> This series first does some cleanups in the Rockchip power domain
> driver and changes the driver, so that it no longer tries to continue
> when it fails to enable a domain. This gets rid of the SError interrupt
> and long backtraces. But the kernel still hangs when it fails to enable
> a power domain. I have not done further analysis to check if that can
> be avoided.
> 
> Last but not least this provides a fix for the GPU power domain failing
> to get enabled - after some testing from my side it seems to require the
> GPU voltage supply to be enabled.
> 
> This series is now based on the pull request from Mark Brown:
> https://lore.kernel.org/linux-pm/ZvsVfQ1fuSVZpF6A@finisterre.sirena.org.uk/
> 
> I added one more patch, which adds devm_of_regulator_get without the
> _optional suffix, since that is more sensible for the Rockchip usecase.
> Longer explanation can be seen in patch 6, which adds the handling to
> the Rockchip driver. My merge suggestion would be that Mark adds the
> regulator patch on top of the immutable branch and creates a new pull
> request.
> 
> The last patch, which updates the RK3588 board files only covers the
> boards from 6.12-rc1. Any board missing the update will behave as before,
> so it is perfectly fine not to update all DT files at once.

My rk3588 jaguar somehow developed some delay when dhcp'ing for its nfs
root and with that actually started running into that gpu-regulator-issue.

With this series applied, that issue goes away:

Tested-by: Heiko Stuebner <heiko at sntech.de>