From: Marc Zyngier <maz@kernel.org>
To: Linus Walleij <linus.walleij@linaro.org>
Cc: linux-gpio@vger.kernel.org
Subject: Re: [PATCH] pinctrl: nomadik: Make gpio irqchip immutable
Date: Sat, 17 Sep 2022 17:27:01 +0100 [thread overview]
Message-ID: <87o7vekluy.wl-maz@kernel.org> (raw)
In-Reply-To: <20220917131907.86899-1-linus.walleij@linaro.org>
On Sat, 17 Sep 2022 14:19:07 +0100,
Linus Walleij <linus.walleij@linaro.org> wrote:
>
> This makes the Nomadik GPIO irqchip immutable.
>
> Tested on the Samsung Galaxy SIII mini GT-I8190.
>
> Cc: Marc Zyngier <maz@kernel.org>
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
> ---
> drivers/pinctrl/nomadik/pinctrl-nomadik.c | 51 +++++++++++------------
> 1 file changed, 24 insertions(+), 27 deletions(-)
>
> diff --git a/drivers/pinctrl/nomadik/pinctrl-nomadik.c b/drivers/pinctrl/nomadik/pinctrl-nomadik.c
> index f5014d09d81a..1ee3b45dd6bc 100644
> --- a/drivers/pinctrl/nomadik/pinctrl-nomadik.c
> +++ b/drivers/pinctrl/nomadik/pinctrl-nomadik.c
> @@ -244,7 +244,6 @@ enum nmk_gpio_slpm {
>
> struct nmk_gpio_chip {
> struct gpio_chip chip;
> - struct irq_chip irqchip;
> void __iomem *addr;
> struct clk *clk;
> unsigned int bank;
> @@ -675,15 +674,11 @@ static void __nmk_gpio_set_wake(struct nmk_gpio_chip *nmk_chip,
> __nmk_gpio_irq_modify(nmk_chip, offset, WAKE, on);
> }
>
> -static int nmk_gpio_irq_maskunmask(struct irq_data *d, bool enable)
> +static void nmk_gpio_irq_maskunmask(struct nmk_gpio_chip *nmk_chip,
> + struct irq_data *d, bool enable)
> {
> - struct nmk_gpio_chip *nmk_chip;
> unsigned long flags;
>
> - nmk_chip = irq_data_get_irq_chip_data(d);
> - if (!nmk_chip)
> - return -EINVAL;
> -
> clk_enable(nmk_chip->clk);
> spin_lock_irqsave(&nmk_gpio_slpm_lock, flags);
> spin_lock(&nmk_chip->lock);
> @@ -696,18 +691,22 @@ static int nmk_gpio_irq_maskunmask(struct irq_data *d, bool enable)
> spin_unlock(&nmk_chip->lock);
> spin_unlock_irqrestore(&nmk_gpio_slpm_lock, flags);
> clk_disable(nmk_chip->clk);
> -
> - return 0;
> }
>
> static void nmk_gpio_irq_mask(struct irq_data *d)
> {
> - nmk_gpio_irq_maskunmask(d, false);
> + struct nmk_gpio_chip *nmk_chip = irq_data_get_irq_chip_data(d);
As far as I can tell, the per-chip data for a GPIO chip points to the
gpio_chip, and not the driver-specific view that contains the generic
structure. Here you're lucky that they are the same address as the
gpio_chip is the first one. This is the same bug as the one fixed by
d1e972ace423 ("gpio: tegra186: Fix chip_data type confusion").
> +
> + nmk_gpio_irq_maskunmask(nmk_chip, d, false);
> + gpiochip_disable_irq(&nmk_chip->chip, irqd_to_hwirq(d));
> }
>
> static void nmk_gpio_irq_unmask(struct irq_data *d)
> {
> - nmk_gpio_irq_maskunmask(d, true);
> + struct nmk_gpio_chip *nmk_chip = irq_data_get_irq_chip_data(d);
> +
> + gpiochip_enable_irq(&nmk_chip->chip, irqd_to_hwirq(d));
> + nmk_gpio_irq_maskunmask(nmk_chip, d, true);
> }
>
> static int nmk_gpio_irq_set_wake(struct irq_data *d, unsigned int on)
> @@ -1078,13 +1077,25 @@ static struct nmk_gpio_chip *nmk_gpio_populate_chip(struct device_node *np,
> return nmk_chip;
> }
>
> +static const struct irq_chip nmk_irq_chip = {
> + .name = "nomadik-gpio",
Maybe not, see below.
> + .irq_ack = nmk_gpio_irq_ack,
> + .irq_mask = nmk_gpio_irq_mask,
> + .irq_unmask = nmk_gpio_irq_unmask,
> + .irq_set_type = nmk_gpio_irq_set_type,
> + .irq_set_wake = nmk_gpio_irq_set_wake,
> + .irq_startup = nmk_gpio_irq_startup,
> + .irq_shutdown = nmk_gpio_irq_shutdown,
> + .flags = IRQCHIP_MASK_ON_SUSPEND | IRQCHIP_IMMUTABLE,
> + GPIOCHIP_IRQ_RESOURCE_HELPERS,
> +};
> +
> static int nmk_gpio_probe(struct platform_device *dev)
> {
> struct device_node *np = dev->dev.of_node;
> struct nmk_gpio_chip *nmk_chip;
> struct gpio_chip *chip;
> struct gpio_irq_chip *girq;
> - struct irq_chip *irqchip;
> bool supports_sleepmode;
> int irq;
> int ret;
> @@ -1125,22 +1136,8 @@ static int nmk_gpio_probe(struct platform_device *dev)
> chip->can_sleep = false;
> chip->owner = THIS_MODULE;
>
> - irqchip = &nmk_chip->irqchip;
> - irqchip->irq_ack = nmk_gpio_irq_ack;
> - irqchip->irq_mask = nmk_gpio_irq_mask;
> - irqchip->irq_unmask = nmk_gpio_irq_unmask;
> - irqchip->irq_set_type = nmk_gpio_irq_set_type;
> - irqchip->irq_set_wake = nmk_gpio_irq_set_wake;
> - irqchip->irq_startup = nmk_gpio_irq_startup;
> - irqchip->irq_shutdown = nmk_gpio_irq_shutdown;
> - irqchip->flags = IRQCHIP_MASK_ON_SUSPEND;
> - irqchip->name = kasprintf(GFP_KERNEL, "nmk%u-%u-%u",
> - dev->id,
> - chip->base,
> - chip->base + chip->ngpio - 1);
> -
Dropping this is unfortunately a userspace visible change. I'm 99%
sure nobody will ever care, but we've been burned by the remaining 1%,
and I'd rather you keep it using the irq_print_chip() callback with
something like:
static void nmk_gpio_irq_print_chip(struct irq_data *data, struct seq_file *p)
{
struct gpio_chip *chip = irq_data_get_irq_chip_data(d);
struct nmk_gpio_chip *nmk_chip = container_of(chip,
struct nmk_gpio_chip,
chip);
seq_printf(p, "nmk%u-%u-%u", nmk_chip->bank,
chip->base, chip->base + chip->ngpio - 1);
}
Thanks,
M.
--
Without deviation from the norm, progress is not possible.
prev parent reply other threads:[~2022-09-17 16:27 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-17 13:19 [PATCH] pinctrl: nomadik: Make gpio irqchip immutable Linus Walleij
2022-09-17 16:27 ` Marc Zyngier [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87o7vekluy.wl-maz@kernel.org \
--to=maz@kernel.org \
--cc=linus.walleij@linaro.org \
--cc=linux-gpio@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.