RPi 4 deferred probe timeout of V3D PM domain

Marek Szyprowski m.szyprowski at samsung.com
Fri Oct 31 06:54:15 PDT 2025


On 31.10.2025 13:39, Mark Brown wrote:
> On Thu, Oct 30, 2025 at 09:12:57PM +0100, Marek Szyprowski wrote:
>> On 30.10.2025 20:13, Stefan Wahren wrote:
>>> Any ideas how to solve this?
>> As I mentioned in the other thread, this is imho a configuration issue.
>> If core modules are distributed on NFS rootfs, then one should increase
>> deferred probe timeout by adding deferred_probe_timeout=60 to cmdline.
> Like I say I just don't think we should apply timeouts that take things
> out of commision entirely.
>
>> On the other hand drivers built into the kernel should not depend on the
>> resources provided by the drivers built as modules, so maybe it would
>> make sense to change CONFIG_CLK_RASPBERRYPI from 'm' to 'y' in
>> arch/arm64/configs/defconfig.
> Or the GPU driver should be moved to a module to match the clock.

V3D already is being built as a module in ARM64's defconfig. The problem 
is in CONFIG_BCM2835_POWER driver (defaults to 'y'), which depends on 
resources provided by CONFIG_CLK_RASPBERRYPI (defaults to 'm'). IMHO 
both of them should be built-in and this will solve this issue.

Best regards
-- 
Marek Szyprowski, PhD
Samsung R&D Institute Poland




More information about the linux-arm-kernel mailing list