From: "Randy.Dunlap" <rdunlap@xenotime.net>
To: Yu Luming <luming.yu@intel.com>
Cc: linux-acpi@vger.kernel.org
Subject: Re: [RFC PATCH]: video output sysfs class and acpi video driver support
Date: Wed, 30 Aug 2006 08:54:30 -0700 [thread overview]
Message-ID: <20060830085430.60e05b4e.rdunlap@xenotime.net> (raw)
In-Reply-To: <200608301542.16035.luming.yu@intel.com>
On Wed, 30 Aug 2006 15:42:15 +0800 Yu Luming wrote:
> Hi,
>
> This is a proposal for a unified sysfs interface to control
> video output device switch. The patch includes two parts:
> 1. output sysfs class. 2. acpi video driver support.
> Any comments are appreciated.
Could you be a bit more verbose in the description of the
patch? (i.e., what it does) Both here and in the Kconfig
help text....
> Thanks,
> Luming
>
> diff -BruN 2.6/drivers/acpi/Kconfig edited/drivers/acpi/Kconfig
> --- 2.6/drivers/acpi/Kconfig 2006-08-30 15:10:05.000000000 +0800
> +++ edited/drivers/acpi/Kconfig 2006-08-30 15:19:09.000000000 +0800
> @@ -106,7 +106,7 @@
>
> config ACPI_VIDEO
> tristate "Video"
> - depends on X86 && BACKLIGHT_DEVICE
> + depends on X86 && ( BACKLIGHT_CLASS_DEVICE || VIDEO_OUTPUT_CONTROL)
s/( /(/
> help
> This driver implement the ACPI Extensions For Display Adapters
s/implement/implements/
(That entire help section needs help...)
> for integrated graphics devices on motherboard, as specified in
> diff -BruN 2.6/drivers/acpi/video.c edited/drivers/acpi/video.c
> --- 2.6/drivers/acpi/video.c 2006-08-30 15:10:05.000000000 +0800
> +++ edited/drivers/acpi/video.c 2006-08-30 15:19:09.000000000 +0800
> @@ -31,6 +31,7 @@
> #include <linux/proc_fs.h>
> #include <linux/seq_file.h>
> #include <linux/backlight.h>
> +#include <linux/output.h>
>
> #include <asm/uaccess.h>
>
> @@ -163,6 +164,7 @@
> struct acpi_video_bus *video;
> struct acpi_device *dev;
> struct acpi_video_device_brightness *brightness;
> + struct output_device *output_dev;
> };
>
> /* bus */
> @@ -266,6 +268,10 @@
> /*backlight device sysfs support*/
> static int acpi_video_get_brightness(struct backlight_device *bd);
> static int acpi_video_set_brightness(struct backlight_device *bd);
> +static int acpi_video_output_get(struct output_device *);
> +static int acpi_video_output_set(struct output_device *);
> +static int acpi_video_device_get_state(struct acpi_video_device *, unsigned long *);
> +static int acpi_video_device_set_state(struct acpi_video_device *, int);
Don't put just parameter types, put a meaningful/useful identifier
name with them also. (See the functions just above the new lines here.)
>
> static struct backlight_device *acpi_video_backlight;
> static struct acpi_video_device *backlight_acpi_device;
> @@ -275,6 +281,10 @@
> .get_brightness = acpi_video_get_brightness,
> .update_status = acpi_video_set_brightness,
> };
> +static struct output_properties acpi_output_properties = {
> + .set_state = acpi_video_output_set,
> + .get_status = acpi_video_output_get,
> +};
> static int acpi_video_get_brightness(struct backlight_device *bd)
> {
> unsigned long cur_level;
> @@ -289,6 +299,20 @@
> return 0;
> }
>
> +static int acpi_video_output_get(struct output_device *od)
> +{
> + unsigned long state;
> + struct acpi_video_device *vd = (struct acpi_video_device *)class_get_devdata(&od->class_dev);
> + acpi_video_device_get_state(vd, &state);
> + return (int)state;
> +}
> +
> +static int acpi_video_output_set(struct output_device *od)
> +{
> + unsigned long state = od->request_state;
> + struct acpi_video_device *vd = (struct acpi_video_device *)class_get_devdata(&od->class_dev);
> + return acpi_video_device_set_state(vd, state);
> +}
>
> /* --------------------------------------------------------------------------
> Video Management
> @@ -614,6 +638,13 @@
> backlight_acpi_device = device;
> }
>
> + if (device->cap._DCS && device->cap._DSS){
) {
> + char name[16];
> + strcat(name, acpi_device_bid(device->dev->parent));
> + strcat(name, "_");
> + strcpy(name, acpi_device_bid(device->dev));
I would feel safer with some length-safe str* functions being
used there.
> + device->output_dev = video_output_register(name, device, &acpi_output_properties);
> + }
> return;
> }
>
> @@ -1617,6 +1648,8 @@
> if (device == backlight_acpi_device)
> backlight_device_unregister(acpi_video_backlight);
>
> + video_output_unregister(device->output_dev);
> +
> return 0;
> }
>
> diff -BruN 2.6/drivers/video/Kconfig edited/drivers/video/Kconfig
> --- 2.6/drivers/video/Kconfig 2006-08-28 11:41:48.000000000 +0800
> +++ edited/drivers/video/Kconfig 2006-08-30 15:19:09.000000000 +0800
> @@ -1616,5 +1616,13 @@
> source "drivers/video/backlight/Kconfig"
> endif
>
> +
> +config VIDEO_OUTPUT_CONTROL
> + tristate "video output device switch control"
> + depends on SYSFS
> + ---help---
> + sysfs support for video output device switching.
This is only useful if someone already knows what it does.
It doesn't help someone who is uninformed about it.
> +
> +
> endmenu
>
> diff -BruN 2.6/drivers/video/Makefile edited/drivers/video/Makefile
> --- 2.6/drivers/video/Makefile 2006-08-28 11:41:48.000000000 +0800
> +++ edited/drivers/video/Makefile 2006-08-30 15:19:09.000000000 +0800
> @@ -12,7 +12,7 @@
>
> obj-$(CONFIG_VT) += console/
> obj-$(CONFIG_LOGO) += logo/
> -obj-$(CONFIG_SYSFS) += backlight/
> +obj-$(CONFIG_SYSFS) += backlight/
Don't add whitespace at end of line (as above).
>
> obj-$(CONFIG_FB_CFB_FILLRECT) += cfbfillrect.o
> obj-$(CONFIG_FB_CFB_COPYAREA) += cfbcopyarea.o
> @@ -107,3 +107,4 @@
>
> # the test framebuffer is last
> obj-$(CONFIG_FB_VIRTUAL) += vfb.o
> +obj-$(CONFIG_VIDEO_OUTPUT_CONTROL) += output.o
> diff -BruN 2.6/drivers/video/output.c edited/drivers/video/output.c
> --- 2.6/drivers/video/output.c 1970-01-01 08:00:00.000000000 +0800
> +++ edited/drivers/video/output.c 2006-08-30 15:20:04.000000000 +0800
> @@ -0,0 +1,109 @@
> +/*
> + * Video output switch support
> + */
> +
> +#include <linux/module.h>
> +#include <linux/output.h>
> +#include <linux/err.h>
> +#include <linux/ctype.h>
> +
> +
> +MODULE_DESCRIPTION("Backlight Lowlevel Control Abstraction");
> +MODULE_LICENSE("GPL");
> +MODULE_AUTHOR("Luming Yu <luming.yu@intel.com>");
> +
> +static ssize_t video_output_show_state (struct class_device *dev, char *buf)
func_name(...) // no space between func_name and '('
> +{
> + ssize_t ret_size = 0;
> + struct output_device *od = to_output_device(dev);
> +
> + if(od->props)
if (
> + ret_size = sprintf(buf, "%.8x\n", od->props->get_status(od));
> +
> + return ret_size;
> +}
> +
> +static ssize_t video_output_store_state (struct class_device *dev, const char *buf, size_t count)
no space after function name
> +{
> + char *endp;
> + struct output_device *od = to_output_device(dev);
> + int request_state = simple_strtoul(buf, &endp, 0);
> + size_t size = endp -buf;
> +
> + printk("in video_output_store_state %.8x\n", request_state);
> + if (*endp && isspace(*endp))
> + size++;
> + if (size != count)
> + return -EINVAL;
There seems to be some off whitespace/tabs there.
> +
> + if(od->props){
if (
> + od->request_state = request_state;
> + od->props->set_state(od);
> + }
> + return count;
> +}
> +
> +static void video_output_class_release(struct class_device *dev)
> +{
> + struct output_device *od = to_output_device(dev);
> + kfree(od);
> +}
> +
> +static struct class_device_attribute video_output_attributes[] = {
> + __ATTR(state, 0644, video_output_show_state, video_output_store_state),
> + __ATTR_NULL,
> +};
> +
> +static struct class video_output_class = {
> + .name = "video_output",
> + .release = video_output_class_release,
> + .class_dev_attrs = video_output_attributes,
> +};
> +
> +
> +struct output_device *video_output_register(const char *name, void *devdata, struct output_properties *op)
> +{
> + struct output_device *new_dev;
> + int ret_code = 0;
> +
> + new_dev = (struct output_device *) kzalloc( sizeof(struct output_device), GFP_KERNEL);
kzalloc(sizeof(...
> + if (!new_dev) {
> + ret_code = -ENOMEM;
> + goto error_return;
> + }
> + new_dev->props = op;
> + new_dev->class_dev.class = &video_output_class;
> + strlcpy(new_dev->class_dev.class_id, name, KOBJ_NAME_LEN);
> + class_set_devdata(&new_dev->class_dev, devdata);
> + ret_code = class_device_register(&new_dev->class_dev);
> + if (ret_code){
) {
> + kfree (new_dev);
kfree(new_dev);
> + goto error_return;
> + }
> + return new_dev;
> +
> +error_return:
> + return ERR_PTR(ret_code);
> +}
> +EXPORT_SYMBOL(video_output_register);
> +
> +void video_output_unregister(struct output_device *dev)
> +{
> + if (!dev)
> + return;
> + class_device_unregister(&dev->class_dev);
> +}
> +EXPORT_SYMBOL(video_output_unregister);
> +
> +static void __exit video_output_class_exit(void)
> +{
> + class_unregister(&video_output_class);
> +}
> +
> +static int __init video_output_class_init(void)
> +{
> + return class_register(&video_output_class);
> +}
> +
> +postcore_initcall(video_output_class_init);
> +module_exit(video_output_class_exit);
> diff -BruN 2.6/include/linux/output.h edited/include/linux/output.h
> --- 2.6/include/linux/output.h 1970-01-01 08:00:00.000000000 +0800
> +++ edited/include/linux/output.h 2006-08-30 15:19:38.000000000 +0800
> @@ -0,0 +1,23 @@
> +#ifndef _LINUX_OUTPUT_H
> +#define _LINUX_OUTPUT_H
> +
> +#include <linux/device.h>
> +
> +struct output_device;
> +
> +struct output_properties {
> + int (*set_state)(struct output_device *);
> + int (*get_status)(struct output_device *);
> +};
> +
> +struct output_device {
> + int request_state;
> + struct output_properties *props;
> + struct class_device class_dev;
> +};
> +
> +#define to_output_device(obj) container_of(obj, struct output_device, class_dev)
> +
> +struct output_device *video_output_register(const char *, void *, struct output_properties *);
> +void video_output_unregister(struct output_device *);
> +#endif
---
~Randy
prev parent reply other threads:[~2006-08-30 15:51 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-08-30 7:42 [RFC PATCH]: video output sysfs class and acpi video driver support Yu Luming
2006-08-30 15:54 ` Randy.Dunlap [this message]
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=20060830085430.60e05b4e.rdunlap@xenotime.net \
--to=rdunlap@xenotime.net \
--cc=linux-acpi@vger.kernel.org \
--cc=luming.yu@intel.com \
/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