Re: [RFC PATCH v1 4/4] rust: add PL031 RTC driver

[Greg Kroah-Hartman <gregkh@linuxfoundation.org>]

On Tue, Jan 06, 2026 at 04:04:00PM +0100, Danilo Krummrich wrote:
> On Tue Jan 6, 2026 at 3:44 PM CET, Greg Kroah-Hartman wrote:
> > On Tue, Jan 06, 2026 at 02:32:56PM +0100, Danilo Krummrich wrote:
> >> (Cc: Greg, Rafael)
> >> 
> >> On Tue Jan 6, 2026 at 3:51 AM CET, Ke Sun wrote:
> >> > Following the platform driver implementation, the AMBA driver stores its 
> >> > drvdata in amba_device->dev. However,
> >> > the RTC driver also stores its drvdata in the parent device (which is 
> >> > also amba_device->dev), causing a conflict.
> >
> > Wait, what?  That's broken in the C implementation, please, let's fix
> > that up now first.  drvdata is ONLY for the specific device that driver
> > is attached to, it can not be used by anyone else.
> 
> The C RTC class device implementation passes its parent device to all its
> callbacks, so the driver can obtain the device private data of the parent (bus)
> device.

Ugh :(

> The RTC core does only provide devres guarded registration functions and does
> some synchronization in devm_rtc_unregister_device() [1], such that no RTC class
> device callbacks can happen after the parent (bus) device is unbound (which is
> great).
> 
> So, technically it doesn't seem to be broken, but given the rationale in my
> previous reply, I dislike that there is no separation of ownership and lifetime
> due to not separating bus and class device private data.
> 
> [1] https://elixir.bootlin.com/linux/v6.19-rc4/source/drivers/rtc/class.c#L340

I agree, this should be fixed up somehow...