linux-kernel - Re: [PATCH v3 2/2] leds: add sgm3140 driver

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]

Date:   Mon, 27 Apr 2020 23:10:26 +0200
From:   Jacek Anaszewski <jacek.anaszewski@...il.com>
To:     Luca Weiss <luca@...tu.xyz>, Pavel Machek <pavel@....cz>
Cc:     linux-leds@...r.kernel.org, Dan Murphy <dmurphy@...com>,
        Heiko Stuebner <heiko@...ech.de>,
        Icenowy Zheng <icenowy@...c.io>,
        Laurent Pinchart <laurent.pinchart@...asonboard.com>,
        Mark Rutland <mark.rutland@....com>,
        Maxime Ripard <mripard@...nel.org>,
        Rob Herring <robh+dt@...nel.org>,
        Shawn Guo <shawnguo@...nel.org>, devicetree@...r.kernel.org,
        linux-kernel@...r.kernel.org, ~postmarketos/upstreaming@...ts.sr.ht
Subject: Re: [PATCH v3 2/2] leds: add sgm3140 driver

Hi Luca,

On 4/27/20 9:49 PM, Luca Weiss wrote:
> On Montag, 27. April 2020 11:51:02 CEST Pavel Machek wrote:
>> Hi!
>>
>>> Add a driver for the SGMICRO SGM3140 Buck/Boost Charge Pump LED driver.
>>>
>>> This device is controlled by two GPIO pins, one for enabling and the
>>> second one for switching between torch and flash mode.
>>>
>>> Signed-off-by: Luca Weiss <luca@...tu.xyz>
>>
>> Thanks, applied, but... I may remove it again.
>>
>>> +++ b/drivers/leds/leds-sgm3140.c
>>> @@ -0,0 +1,320 @@
>>> +// SPDX-License-Identifier: GPL-2.0
>>
>> Would you consider GPL-2+?
> 
> I don't really have a preference either way but GPL-2.0-or-later is fine for
> me.
> 
>>
>>> +#if IS_ENABLED(CONFIG_V4L2_FLASH_LED_CLASS)
>>> +static void sgm3140_init_v4l2_flash_config(struct sgm3140 *priv,
>>
>> Ok.
>>
>>> +static int sgm3140_probe(struct platform_device *pdev)
>>> +{
>>
>> ...
>>
>>> +	led_cdev->brightness_set_blocking = sgm3140_brightness_set;
>>> +	led_cdev->max_brightness = LED_ON;
>>
>> Don't do this, unless you really have 255 levels of brightness.
> 
> LED_ON is 1, so the brightness available is 0 - 1.
> 
>>
>>> +	/* Create V4L2 Flash subdev */
>>> +	priv->v4l2_flash = v4l2_flash_init(&pdev->dev,
>>> +					   child_node,
>>> +					   fled_cdev, NULL,
>>> +					   &v4l2_sd_cfg);
>>> +	if (IS_ERR(priv->v4l2_flash)) {
>>
>> Does this need some #ifdef guards?
> 
> v4l2_flash_init has a NULL-returning version when CONFIG_V4L2_FLASH_LED_CLASS
> is not defined (see https://elixir.bootlin.com/linux/latest/source/include/
> media/v4l2-flash-led-class.h#L166 )

Correct.

>>
>>> +		ret = PTR_ERR(priv->v4l2_flash);
>>> +		goto err;
>>> +	}
>>> +
>>> +	return ret;
>>
>> Should this return 0?
> 
> ret should be 0 here, so it shouldn't matter much.
> 
>>
>>> +err:
>>> +	fwnode_handle_put(child_node);
>>> +	return ret;
>>> +}
>>
>> Does non-error path needs handle_put, too?
> 
> I don't think so, I'm passing child_node to v4l2_flash_init which then saves
> the pointer to v4l2_subdev->fwnode.
> 
> The devm_led_classdev_flash_register_ext function also seems to store the
> pointer (led_cdev->dev->fwnode = init_data->fwnode; in
> led_classdev_register_ext)

The node needs to be held only during parsing, so you
have to call fwnode_handle_put() for both paths.

-- 
Best regards,
Jacek Anaszewski