Hello Michael,

On Mon, Nov 30, 2020 at 09:17:52AM +0100, Michael Walle wrote:
> Am 2020-11-24 22:24, schrieb Uwe Kleine-König:
> > Currently .get_state() and .apply() use dev_get_drvdata() on the struct
> > device related to the pwm chip. This only works after .probe() called
> > platform_set_drvdata() which in this driver happens only after
> > pwmchip_add() and so comes possibly too late.
> > 
> > Instead of setting the driver data earlier use the traditional
> > container_of approach as this way the driver data is conceptually and
> > computational nearer.
> > 
> > Fixes: 9db33d221efc ("pwm: Add support for sl28cpld PWM controller")
> > Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> > ---
> > Hello,
> > 
> > in v4 of the driver Michael still used container_of and then changed to
> > use dev_get_drvdata() as Lee suggested. I didn't notice this suggestion,
> > otherwise I would have expressed my variance about this already earlier.
> > 
> > I noticed this problem because Michael contacted me via irc and showed
> > me the resulting oops, so I think applying this before 5.10 would be
> > good. It's not entirely clear to me why .get_state() is called that
> > early in his case, but the theory is clear: The callbacks can be called
> > as soon as pwmchip_add() is called.
> > 
> > Best regards
> > Uwe
> 
> Ping. Would be nice if this makes it into v5.10.

Yes, fully agreed. @Thierry: Do you care to send it to Linus? Or are you
ok if I do that?

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | https://www.pengutronix.de/ |