Re: [PATCH net v1 1/1] net: dsa: realtek: use gpiod_set_value_cansleep for reset GPIO

On Fri, Aug 14, 2026 at 10:00:14AM -0700, Jakub Kicinski wrote:

On Fri, 14 Aug 2026 16:09:04 +0200 Andrew Lunn wrote:

quoted

On Fri, Aug 14, 2026 at 01:01:02PM +0200, Oleksij Rempel wrote:

quoted

From: Ahmad Fatoum <a.fatoum@pengutronix.de>

rtl83xx_reset_assert() and rtl83xx_reset_deassert() are only called from
the probe path, which may sleep and is not timing-critical.  When the
reset GPIO is provided by a sleeping controller such as an I2C I/O
expander, gpiod_set_value() warns:

Switch both helpers to gpiod_set_value_cansleep() so such a reset GPIO can
be used without triggering the warning.

The reset GPIO has been driven with the non-sleeping gpiod_set_value()
since the driver was added in v4.19.  The call has since been refactored
across several files - from realtek-smi.c / realtek-mdio.c into the common
rtl83xx.c module and then into the rtl83xx_reset_assert() and
rtl83xx_reset_deassert() helpers (both in v6.9).  This patch therefore
applies as-is only to kernels that carry those helpers (v6.9+); older
stable kernels need the same gpiod_set_value_cansleep() conversion at the
corresponding open-coded call sites.  

This is one of those fuzzy cases. It never worked, so this is not a
regression. Nobody has needed it up until now. So it probably should
be considered a new feature, and merged via net-next.

Sounds like it's a false positive, but it's a WARN, so if it hits on
real platforms -- I think it's a real fix?

We can back port it for related product by our self. Otherwise, the
LLM back port bot from Sasha Levin will do it any way, as soon as it is
taken. no matter as net or net-next. Flip a coin :)

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |