linux-kernel - Re: [PATCH v2 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] [thread-next>] [day] [month] [year] [list]

Message-ID: <4437676.LvFx2qVVIh@g550jk>
Date:   Sun, 05 Apr 2020 20:45:14 +0200
From:   Luca Weiss <luca@...tu.xyz>
To:     Andy Shevchenko <andy.shevchenko@...il.com>
Cc:     Linux LED Subsystem <linux-leds@...r.kernel.org>,
        Dan Murphy <dmurphy@...com>, Heiko Stuebner <heiko@...ech.de>,
        Icenowy Zheng <icenowy@...c.io>,
        Jacek Anaszewski <jacek.anaszewski@...il.com>,
        Laurent Pinchart <laurent.pinchart@...asonboard.com>,
        Mark Rutland <mark.rutland@....com>,
        Maxime Ripard <mripard@...nel.org>,
        Pavel Machek <pavel@....cz>, Rob Herring <robh+dt@...nel.org>,
        Shawn Guo <shawnguo@...nel.org>,
        devicetree <devicetree@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        ~postmarketos/upstreaming@...ts.sr.ht
Subject: Re: [PATCH v2 2/2] leds: add sgm3140 driver

Hi Andy,

On Samstag, 4. April 2020 11:58:31 CEST Andy Shevchenko wrote:
> On Mon, Mar 30, 2020 at 10:49 PM Luca Weiss <luca@...tu.xyz> wrote:
> > 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.
> 
> ...
> 
> > +config LEDS_SGM3140
> > +       tristate "LED support for the SGM3140"
> > +       depends on LEDS_CLASS_FLASH
> > +       depends on V4L2_FLASH_LED_CLASS || !V4L2_FLASH_LED_CLASS
> > 
> > +       depends on OF
> 
> depends on OF || COMPILE_TEST ?
> But hold on...
> 
> ...
> 
> > +#include <linux/of.h>
> 
> Perhaps switch this to property.h and replace OF with more generic
> device property / fwnode API?
> 

I didn't find clear documentation on this, the functions in drivers/base/
property.c can be used instead of the of_* (device tree) functions?

As far as I can tell, the device_property_* functions are supposed to be used 
for simple "give me a property for this 'struct device*'" while the fwnode_* 
functions are used as generic equivalent of the of_* functions?

So in this case I can replace 

struct device_node *child_node;
child_node = of_get_next_available_child(pdev->dev.of_node, NULL);

with

struct fwnode_handle *child_node;
child_node = fwnode_get_next_available_child_node(pdev->dev.fwnode, NULL);

and then instead of

ret = of_property_read_u32(child_node, "flash-max-timeout-us",
		   &priv->max_timeout);

use

ret = fwnode_property_read_u32(child_node, "flash-max-timeout-us",
		            &priv->max_timeout);

and finally instead of

init_data.fwnode = of_fwnode_handle(child_node);

I can probably directly do

init_data.fwnode = child_node;

Does that sound correct?

> ...
> 
> > +struct sgm3140 {
> > +       bool enabled;
> > +       struct gpio_desc *flash_gpio;
> > +       struct gpio_desc *enable_gpio;
> > +       struct regulator *vin_regulator;
> > +
> > +       /* current timeout in us */
> > +       u32 timeout;
> > +       /* maximum timeout in us */
> > +       u32 max_timeout;
> > +
> > 
> > +       struct led_classdev_flash fled_cdev;
> 
> I guess it might be slightly better to make it first member of the
> struct (I didn't check but the rationale is to put more often used
> members at the beginning to utilize cachelines).
> 
> > +       struct v4l2_flash *v4l2_flash;
> > +
> > +       struct timer_list powerdown_timer;
> > +};
> 
> ...
> 
> > +static struct sgm3140 *flcdev_to_sgm3140(struct led_classdev_flash
> > *flcdev) +{
> > +       return container_of(flcdev, struct sgm3140, fled_cdev);
> > +}
> 
> ...and this becomes a no-op AFAICS (doesn't mean you need to remove it).
> 
> ...
> 
> > +       struct device_node *child_node;
> > 
> > +       child_node = of_get_next_available_child(pdev->dev.of_node, NULL);
> > 
> > +       ret = of_property_read_u32(child_node, "flash-max-timeout-us",
> > +                                  &priv->max_timeout);
> > 
> > +       init_data.fwnode = of_fwnode_handle(child_node);
> > 
> > +       of_node_put(child_node);
> 
> Device property / fwnode API?
> 
> --
> With Best Regards,
> Andy Shevchenko

Regards
Luca