linux-kernel - Re: [PATCH] gpio: pca953x: mask interrupts in irq shutdown

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: <CAD++jLmOjdiDfLPqAA+6SmK2mfddh6O2YoPD10gkBbo=dCOf9Q@mail.gmail.com>
Date: Tue, 27 Jan 2026 10:29:24 +0100
From: Linus Walleij <linusw@...nel.org>
To: Martin Larsson <martin.larsson@...ia.se>
Cc: Bartosz Golaszewski <brgl@...nel.org>, 
	"linux-gpio@...r.kernel.org" <linux-gpio@...r.kernel.org>, 
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>, Michael Roth <mail@...th.net>, 
	Francesco Lavra <flavra@...libre.com>, Hugo Villeneuve <hvilleneuve@...onoff.com>, 
	Maria Garcia <mariagarcia7293@...il.com>, Emanuele Ghidoli <emanuele.ghidoli@...adex.com>
Subject: Re: [PATCH] gpio: pca953x: mask interrupts in irq shutdown

Add a few recent developers to CC, this is a very widely used GPIO chip.

Emanuele was dealing with IRQ storms and may want to look at this.

On Wed, Jan 21, 2026 at 2:12 PM Martin Larsson <martin.larsson@...ia.se> wrote:

> In the existing implementation irq_shutdown does not mask the interrupts
> in hardware. This can cause spurious interrupts from the IO expander.
> Add masking to irq_shutdown to prevent spurious interrupts.
>
> Signed-off-by: Martin Larsson <martin.larsson@...ia.se>
> ---
>  drivers/gpio/gpio-pca953x.c | 2 ++
>  1 file changed, 2 insertions(+)
>
> diff --git a/drivers/gpio/gpio-pca953x.c b/drivers/gpio/gpio-pca953x.c
> index 8727ae54bc57..f93a3dbb2daa 100644
> --- a/drivers/gpio/gpio-pca953x.c
> +++ b/drivers/gpio/gpio-pca953x.c
> @@ -914,6 +914,8 @@ static void pca953x_irq_shutdown(struct irq_data *d)
>         clear_bit(hwirq, chip->irq_trig_fall);
>         clear_bit(hwirq, chip->irq_trig_level_low);
>         clear_bit(hwirq, chip->irq_trig_level_high);
> +
> +       pca953x_irq_mask(d);

It makes sense to me.

Reviewed-by: Linus Walleij <linusw@...nel.org>

Yours,
Linus Walleij