linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: u.kleine-koenig@pengutronix.de (Uwe Kleine-König)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v2 1/2] ARM: mxc: Introduce imx_add_gpio_leds
Date: Tue, 5 Apr 2011 09:30:30 +0200	[thread overview]
Message-ID: <20110405073030.GI13963@pengutronix.de> (raw)
In-Reply-To: <0D753D10438DA54287A00B027084269764D1C89CC6@AUSP01VMBX24.collaborationhost.net>

Hello,

On Mon, Apr 04, 2011 at 04:52:43PM -0500, H Hartley Sweeten wrote:
> On Monday, April 04, 2011 11:18 AM, Russell King wrote:
> >>> If you go to the effort of making it _that_ generic, then it shouldn't
> >>> even be in plat-mxc, but somewhere *everyone* can benefit from it.  See
> >>> Linus' complaints in the OMAP pull request thread about the amount of
> >>> largely similar arch/arm code.
> >> 
> >> If it's made generic, mach-ep93xx/core.c could use it to register it's two
> >> platform leds.  Please keep me in the loop if this is done.
> >
> > In which case we don't want "if it's made generic", we *require* it to be
> > generic.

Yeah, definitely. I already tried once to make the device registration
that we use on mxc globally available.

> > As I say, we *must* get away from the idea that something in mach-xyz has
> > nothing to do with our own development in mach-abc, and start taking an
> > interest in consolidating some of this stuff.  Otherwise we may get to
> > the point where further arch/arm development is barred from the mainline
> > kernel.
> 
> How about something like this:
> 
> int __init gpio_led_platform_register(int id, struct gpio_led *leds, int nr)
> {
> 	struct gpio_led_platform_data *pdata;
> 	struct platform_device *pdev;
> 	int ret;
> 
> 	pdata = kzalloc(sizeof(*pdata), GFP_KERNEL);
> 	if (!pdata)
> 		return -ENOMEM;
> 
> 	pdev = platform_device_alloc("leds-gpio", id);
> 	if (!pdev) {
> 		ret = -ENOMEM;
> 		goto out_free;
> 	}
> 
> 	pdata->leds = leds;
> 	pdata->num_leds = nr;
> 	pdev->dev.platform_data = pdata;
> 	ret = platform_device_add(pdev);
> 	if (ret)
> 		goto out_pdev;
> 	return 0;
> 
> out_pdev:
> 	platform_device_put(pdev);
> out_free:
> 	kfree(pdata);
> 	return ret;
> }
I like my approach better, without using mxc specific functions it would
look as follows:

struct platform_device *__init gpio_led_register_device(
		const gpio_led_platform_data *pdata)
{
	struct platform_device *ret = ERR_PTR(-ENOMEM);
	struct gpio_led_platform_data *_pdata = *pdata;
	_pdata->leds = kmemdup(pdata->leds,
			 pdata->num_leds * sizeof(*pdata->leds), GFP_KERNEL);

	if (!_pdata->leds)
		return ERR_PTR(-ENOMEM);

	ret = platform_device_register_resndata(NULL, "leds-gpio", -1,
		NULL, 0, _pdata, sizeof(_pdata));

	if (IS_ERR(ret))
		kfree(_pdata->leds);

	return ret;
}

> The the mach-xyz init code just has to:
> 
> static struct gpio_led gpio_leds[] = {
> 	{
> 		.name	= "foo:fooclr",
> 		.gpio	= FOO_GPIO_NUM,
> 	}, {
> 		.name	= "bar:barclr",
> 		.gpio	= BAR_LED_NUM,
> 	},
> 	/* etc. */
> };
> 
> 
> 	gpio_led_platform_register(-1, gpio_leds, ARRAY_SIZE(gpio_leds));
Compared to your approach with mine
 - gpio_leds could be const __initconst;
 - uses only a single parameter which AFAIK saves a few bytes to call
   the function;
 - you don't have control about the .id member (which IMHO is OK but
   might be changed easily);
 - you have to call a function with a different name;

(Note all of these are no hard requirements for me, just my thoughts.)

> If some one could suggest a good place for the gpio_led_platform_register function
> I can create a proper patch.  Maybe just add it to the end of the leds-gpio.c driver?
mxc uses a central place to register all devices
(arch/arm/plat-mxc/devices).  I think for a global approach this won't
scale (e.g. because a single header file is used for all devices), so a
place near the driver is the right thing to do. leds-gpio.c doesn't work
though because then it might end in a module.

I'll send a patch as a reply to this mail.

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-K?nig            |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |

  reply	other threads:[~2011-04-05  7:30 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-04-04 17:06 [PATCH v2 1/2] ARM: mxc: Introduce imx_add_gpio_leds Fabio Estevam
2011-04-04 17:06 ` [PATCH v2 2/2] ARM: mx5/mx53_evk.c: Add LED support Fabio Estevam
2011-04-04 17:19 ` [PATCH v2 1/2] ARM: mxc: Introduce imx_add_gpio_leds Sascha Hauer
2011-04-04 17:42   ` Uwe Kleine-König
2011-04-04 17:52     ` Russell King - ARM Linux
2011-04-04 18:06       ` H Hartley Sweeten
2011-04-04 18:17         ` Russell King - ARM Linux
2011-04-04 21:52           ` H Hartley Sweeten
2011-04-05  7:30             ` Uwe Kleine-König [this message]
2011-04-05  7:40               ` Russell King - ARM Linux
2011-04-05  7:43                 ` Uwe Kleine-König
2011-04-05  7:47                   ` Russell King - ARM Linux
2011-04-05  7:51                     ` Uwe Kleine-König
2011-04-05  7:59                       ` Russell King - ARM Linux
2011-04-05  8:32                         ` Uwe Kleine-König
2011-04-05  8:43                           ` Russell King - ARM Linux
2011-04-05  8:46                             ` Uwe Kleine-König
2011-04-05  8:37               ` [PATCH] leds: provide helper to register "leds-gpio" devices Uwe Kleine-König
2011-04-05 16:13                 ` Fabio Estevam
2011-04-05 16:29                   ` Fabio Estevam
2011-04-05 18:12                     ` H Hartley Sweeten
2011-04-05 16:33                 ` Russell King - ARM Linux
2011-04-05 20:24                   ` [PATCH v2] " Uwe Kleine-König
2011-04-06 11:45                     ` Fabio Estevam
2011-04-06 11:52                     ` Richard Purdie
2011-04-06 12:33                       ` Uwe Kleine-König
2011-04-06 13:38                         ` Richard Purdie
2011-04-11 20:35                           ` [PATCH v3] " Uwe Kleine-König
2011-04-12 21:48                             ` Russell King - ARM Linux
2011-04-13  6:23                               ` Uwe Kleine-König
2011-05-06 21:03                                 ` Richard Purdie
2011-05-09  8:00                                   ` Uwe Kleine-König
2011-04-26 15:08                             ` Uwe Kleine-König
2011-05-06  8:25                               ` Uwe Kleine-König
2011-05-09 22:02                             ` Andrew Morton
2011-05-09 22:17                               ` Russell King - ARM Linux
2011-05-10  6:45                                 ` Uwe Kleine-König
2011-05-10  7:31                               ` Uwe Kleine-König
2011-05-10  8:50                                 ` [PATCH v4] " Uwe Kleine-König
2011-05-10  8:50                                   ` [PATCH] [wip] ARM: imx: register "leds-gpio" device using new helper function Uwe Kleine-König
2011-05-10 22:26                                     ` H Hartley Sweeten
2011-05-11  6:22                                       ` Uwe Kleine-König
2011-05-10 23:02                                     ` H Hartley Sweeten
2011-04-19 23:19                   ` [PATCH] leds: provide helper to register "leds-gpio" devices Andrew Morton
2011-04-19 23:24                     ` Russell King - ARM Linux
2011-04-19 23:50                       ` Andrew Morton
2011-04-05 16:41               ` [PATCH v2 1/2] ARM: mxc: Introduce imx_add_gpio_leds H Hartley Sweeten

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=20110405073030.GI13963@pengutronix.de \
    --to=u.kleine-koenig@pengutronix.de \
    --cc=linux-arm-kernel@lists.infradead.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).