linux-kernel - Re: [PATCH v3 2/2] pinctrl: cy8c95x0: Add reset 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] [day] [month] [year] [list]

Message-ID: <ZLmUrA1ANfAVJYHi@surfacebook>
Date:   Thu, 20 Jul 2023 23:10:20 +0300
From:   andy.shevchenko@...il.com
To:     Naresh Solanki <naresh.solanki@...ements.com>
Cc:     Patrick Rudolph <patrick.rudolph@...ements.com>,
        Linus Walleij <linus.walleij@...aro.org>,
        linux-gpio@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3 2/2] pinctrl: cy8c95x0: Add reset support

Fri, Jul 14, 2023 at 10:19:01AM +0200, Naresh Solanki kirjoitti:
> From: Patrick Rudolph <patrick.rudolph@...ements.com>
> 
> This patch adds support for an optional "reset" GPIO pin in the cy8c95x0

Find "This patch" in the Submitting Patches documentation and modify this
accordingly.

> pinctrl driver. On probe, the reset pin is pulled low to bring chip out
> of reset. The reset pin has an internal pull-down and can be left
> floating if not required.
> 
> The datasheet doesn't mention any timing related to the reset pin.
> 
> Based on empirical tests, it was found that the chip requires a
> delay of 250 milliseconds before accepting I2C transfers after driving
> the reset pin low. Therefore, a delay of 250ms is added before
> proceeding with I2C transfers.

...

> +	/* bring the chip out of reset if reset pin is provided */
> +	chip->gpio_reset = devm_gpiod_get_optional(&client->dev, "reset", GPIOD_OUT_HIGH);

&client->dev

> +	if (IS_ERR(chip->gpio_reset)) {
> +		ret = dev_err_probe(chip->dev, PTR_ERR(chip->gpio_reset),

chip->dev

Why inconsistent?

> +				    "Failed to get GPIO 'reset'\n");
> +		goto err_exit;

I haven't looked at the context, but this seems like a smoke, we usually don't
have such in devm-enabled drivers.

> +	} else if (chip->gpio_reset) {
> +		usleep_range(1000, 2000);
> +		gpiod_set_value_cansleep(chip->gpio_reset, 0);

> +		usleep_range(250000, 300000);

This is incorrect use of usleep_range(). Please, read documentation
Documentation/timers/timers-howto.rst on the topic. The fix can be switching to
fsleep() in both branches.


> +		gpiod_set_consumer_name(chip->gpio_reset, "CY8C95X0 RESET");

This can be done uncoditionally before toggling the reset.

> +	}

...

This change wasn't Cc'ed to me, however I was contributing a lot into this driver.
I would recommend to use my "smart" script [1] to send patches. It has some
heuristics which works well in 98% of the cases for me.

[1]: https://github.com/andy-shev/home-bin-tools/blob/master/ge2maintainer.sh


-- 
With Best Regards,
Andy Shevchenko