linux-kernel - Re: [RFC v7 4/6] gpio: mpfs: add polarfire soc gpio support

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20241016-dandelion-hypnosis-9d989bb2fdd1@spud>
Date: Wed, 16 Oct 2024 11:29:03 +0100
From: Conor Dooley <conor@...nel.org>
To: Linus Walleij <linus.walleij@...aro.org>
Cc: Conor Dooley <conor.dooley@...rochip.com>, linux-kernel@...r.kernel.org,
	Marc Zyngier <maz@...nel.org>,
	Daire McNamara <daire.mcnamara@...rochip.com>,
	Bartosz Golaszewski <brgl@...ev.pl>, Rob Herring <robh@...nel.org>,
	Krzysztof Kozlowski <krzk+dt@...nel.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Paul Walmsley <paul.walmsley@...ive.com>,
	Palmer Dabbelt <palmer@...belt.com>,
	linux-riscv@...ts.infradead.org, linux-gpio@...r.kernel.org,
	devicetree@...r.kernel.org, Lewis Hanly <lewis.hanly@...rochip.com>
Subject: Re: [RFC v7 4/6] gpio: mpfs: add polarfire soc gpio support

On Wed, Oct 16, 2024 at 10:56:32AM +0100, Conor Dooley wrote:
> On Mon, Aug 05, 2024 at 10:04:53AM +0200, Linus Walleij wrote:
> > On Tue, Jul 23, 2024 at 1:28 PM Conor Dooley <conor.dooley@...rochip.com> wrote:
> > 
> > 
> > > From: Lewis Hanly <lewis.hanly@...rochip.com>
> > >
> > > Add a driver to support the Polarfire SoC gpio controller
> > >
> > > Signed-off-by: Lewis Hanly <lewis.hanly@...rochip.com>
> > > Signed-off-by: Conor Dooley <conor.dooley@...rochip.com>
> > 
> > Just a comment on second thought:
> > 
> > > +config GPIO_POLARFIRE_SOC
> > > +       bool "Microchip FPGA GPIO support"
> > > +       depends on OF_GPIO
> > > +       select GPIOLIB_IRQCHIP
> > 
> > select GPIO_GENERIC?
> > 
> > > +static int mpfs_gpio_direction_input(struct gpio_chip *gc, unsigned int gpio_index)
> > > +{
> > > +       struct mpfs_gpio_chip *mpfs_gpio = gpiochip_get_data(gc);
> > > +       u32 gpio_cfg;
> > > +       unsigned long flags;
> > > +
> > > +       raw_spin_lock_irqsave(&mpfs_gpio->lock, flags);
> > > +
> > > +       gpio_cfg = readl(mpfs_gpio->base + MPFS_GPIO_CTRL(gpio_index));
> > > +       gpio_cfg |= MPFS_GPIO_EN_IN;
> > > +       gpio_cfg &= ~(MPFS_GPIO_EN_OUT | MPFS_GPIO_EN_OUT_BUF);
> > 
> > OK this part is unique...
> > 
> > > +static int mpfs_gpio_direction_output(struct gpio_chip *gc, unsigned int gpio_index, int value)
> > > +{
> > > +       struct mpfs_gpio_chip *mpfs_gpio = gpiochip_get_data(gc);
> > > +       u32 gpio_cfg;
> > > +       unsigned long flags;
> > > +
> > > +       raw_spin_lock_irqsave(&mpfs_gpio->lock, flags);
> > > +
> > > +       gpio_cfg = readl(mpfs_gpio->base + MPFS_GPIO_CTRL(gpio_index));
> > > +       gpio_cfg |= MPFS_GPIO_EN_OUT | MPFS_GPIO_EN_OUT_BUF;
> > 
> > Also here
> > 
> > > +static int mpfs_gpio_get_direction(struct gpio_chip *gc,
> > > +                                  unsigned int gpio_index)
> > > +static int mpfs_gpio_get(struct gpio_chip *gc,
> > > +                        unsigned int gpio_index)
> > > +static void mpfs_gpio_set(struct gpio_chip *gc, unsigned int gpio_index, int value)
> > 
> > But these are just MMIO functions.
> > 
> > Is it possible to use augmented generic MMIO, i.e just override these
> > two functions that
> > need special handling?
> 
> So, I've been looking into this again (finally), with an eye to stripping
> the interrupt handling bits out, and trying to upstream this in pieces.
> I dunno if I'm making a mistake here, but I don't know if there's much
> value in implementing this suggestion - as far as I can tell only the
> get()/set() functions can be replaced by what's provided by gpio-mmio.c.
> There are no controller wide registers that control direction and so
> bgpio_get_dir() can't be used - direction is read from the same
> mpfs_gpio->base + MPFS_GPIO_CTRL(gpio_index) registers that it is set
> using. Adding bgpio stuff, to just go ahead and overwrite it, to save on
> trivial get()/set() implementations seems to me like adding complication
> rather than removing it. What am I missing here?

What does bring a nice simplification though, IMO, is regmap. I am
pretty sure that using it was one of the suggestions made last time
Lewis submitted this - so I think I'm going to do that instead.

Download attachment "signature.asc" of type "application/pgp-signature" (229 bytes)