[PATCH v12 3/5] gpio: rpmsg: add generic rpmsg GPIO driver

[Shenwei Wang]

> -----Original Message-----
> From: Andrew Lunn <andrew at lunn.ch>
> Sent: Wednesday, March 18, 2026 3:11 PM
> To: Shenwei Wang <shenwei.wang at nxp.com>
> Cc: Mathieu Poirier <mathieu.poirier at linaro.org>; Arnaud POULIQUEN
> <arnaud.pouliquen at foss.st.com>; Linus Walleij <linusw at kernel.org>; Bartosz
> Golaszewski <brgl at kernel.org>; Jonathan Corbet <corbet at lwn.net>; Rob Herring
> <robh at kernel.org>; Krzysztof Kozlowski <krzk+dt at kernel.org>; Conor Dooley
> <conor+dt at kernel.org>; Bjorn Andersson <andersson at kernel.org>; Frank Li
> <frank.li at nxp.com>; Sascha Hauer <s.hauer at pengutronix.de>; Shuah Khan
> <skhan at linuxfoundation.org>; linux-gpio at vger.kernel.org; linux-
> doc at vger.kernel.org; linux-kernel at vger.kernel.org; Pengutronix Kernel Team
> <kernel at pengutronix.de>; Fabio Estevam <festevam at gmail.com>; Peng Fan
> <peng.fan at nxp.com>; devicetree at vger.kernel.org; linux-
> remoteproc at vger.kernel.org; imx at lists.linux.dev; linux-arm-
> kernel at lists.infradead.org; dl-linux-imx <linux-imx at nxp.com>; Bartosz
> Golaszewski <brgl at bgdev.pl>
> Subject: [EXT] Re: [PATCH v12 3/5] gpio: rpmsg: add generic rpmsg GPIO driver
> 
> > @Linus Walleij,
> > From a technical standpoint, this GPIO driver is no different from
> > gpio-mxc, gpio-omap, or gpio-rda.
> 
> Yes it is different. The example you list are all silicon GPIO blocks.
> This driver is not for silicon, is a protocol spoken over rpmsg.  The concept of two

A GPIO driver only needs to follow its defined interface, whether that interface is exposed 
via registers or via rpmsg. It doesn't need to know whether the backend implementation is 
hardware or firmware, and stable firmware can behave just like hardened logic from the 
driver's perspective.

> CPUs connected by rpmsg in a SoC is used by a number of vendors. Look in
> Documentation/devicetree/bindings/remoteproc you see:
> 
> amlogic, fsl, ingenic, mtk, qcom, reneses, st, ti, wkup and xlnx.
> 
> We want one generic protocol/implementation of GPIO over rpmsg which all

If you believe the current GPIO driver isn't generic, could you point to a real example 
that shows where it doesn't fit?

Thanks,
Shenwei 

> these vendors will use. That means we have one GPIO driver, not 10 drivers to
> maintain for the next 10-20 years. It also means those 10 vendors have 1/10 of a
> driver they need to maintain for the next 10-20 years. And likely less bugs to deal
> with, since the driver is more heavily tested by 10 vendors, etc.
> 
>      Andrew