[PATCH v2] Led driver for technologic systems 5500 - Kernel

This is a discussion on [PATCH v2] Led driver for technologic systems 5500 - Kernel ; Here is a patch for the led on the Technologic Systems 5500 board. The first patch had a bug in an error path. Detection is still a bit flimsy; Is there anything that can be done about that? Signed-off-by: Sean ...

+ Reply to Thread
Results 1 to 4 of 4

Thread: [PATCH v2] Led driver for technologic systems 5500

  1. [PATCH v2] Led driver for technologic systems 5500

    Here is a patch for the led on the Technologic Systems 5500 board. The
    first patch had a bug in an error path. Detection is still a bit flimsy; Is
    there anything that can be done about that?

    Signed-off-by: Sean Young
    ---
    diff -urpN linux-2.6.26/drivers/leds/Kconfig /home/sean/tiger/linux-2.6.26/drivers/leds/Kconfig
    --- linux-2.6.26/drivers/leds/Kconfig 2008-07-13 22:51:29.000000000 +0100
    +++ /home/sean/tiger/linux-2.6.26/drivers/leds/Kconfig 2008-07-21 12:19:51.000000000 +0100
    @@ -147,6 +147,16 @@ config LEDS_CLEVO_MAIL
    To compile this driver as a module, choose M here: the
    module will be called leds-clevo-mail.

    +config LEDS_TS5500
    + tristate "LED Support for the TS-5500"
    + depends on LEDS_CLASS && X86
    + help
    + This driver makes the led on the Technologic Systems 5500
    + (and possibly other models) accessible for userspace.
    +
    + Note that this led is hardwired to be xor'ed with Compact Flash
    + activity.
    +
    comment "LED Triggers"

    config LEDS_TRIGGERS
    Binary files linux-2.6.26/drivers/leds/.leds-net48xx.c.swp and /home/sean/tiger/linux-2.6.26/drivers/leds/.leds-net48xx.c.swp differ
    diff -urpN linux-2.6.26/drivers/leds/leds-ts5500.c /home/sean/tiger/linux-2.6.26/drivers/leds/leds-ts5500.c
    --- linux-2.6.26/drivers/leds/leds-ts5500.c 1970-01-01 01:00:00.000000000 +0100
    +++ /home/sean/tiger/linux-2.6.26/drivers/leds/leds-ts5500.c 2008-07-21 21:15:41.000000000 +0100
    @@ -0,0 +1,115 @@
    +/*
    + * Copyright (C) 2008 Sean Young
    + */
    +
    +#include
    +#include
    +#include
    +#include
    +#include
    +#include
    +
    +#define DRVNAME "ts5500-leds"
    +
    +static struct platform_device *ts5500_pdev;
    +
    +static void ts5500_led_set(struct led_classdev *led_dev, enum led_brightness value)
    +{
    + outb(value ? 1 : 0, 0x77);
    +}
    +
    +static struct led_classdev ts5500_led = {
    + .name = "ts5500:green",
    + .brightness_set = ts5500_led_set,
    +};
    +
    +static int __devinit ts5500leds_probe(struct platform_device *pdev)
    +{
    + return led_classdev_register(&pdev->dev, &ts5500_led);
    +}
    +
    +static int __devexit ts5500leds_remove(struct platform_device *pdev)
    +{
    + led_classdev_unregister(&ts5500_led);
    + return 0;
    +}
    +
    +#ifdef CONFIG_PM
    +static int ts5500leds_suspend(struct platform_device *pdev, pm_message_t state)
    +{
    + led_classdev_suspend(&ts5500_led);
    + return 0;
    +}
    +
    +static int ts5500leds_resume(struct platform_device *pdev)
    +{
    + led_classdev_resume(&ts5500_led);
    + return 0;
    +}
    +#endif
    +
    +static struct platform_driver ts5500leds_driver = {
    + .driver = {
    + .name = DRVNAME,
    + .owner = THIS_MODULE,
    + },
    + .probe = ts5500leds_probe,
    + .remove = __devexit_p(ts5500leds_remove),
    +#ifdef CONFIG_PM
    + .suspend = ts5500leds_suspend,
    + .resume = ts5500leds_resume,
    +#endif
    +};
    +
    +
    +static int __init ts5500leds_init(void)
    +{
    + int rc;
    + uint8_t product_code;
    +
    + if (!request_region(0x74, 1, DRVNAME))
    + return -ENODEV;
    +
    + product_code = inb(0x74);
    +
    + release_region(0x74, 1);
    +
    + if (product_code != 0x60)
    + return -ENODEV;
    +
    + if (!request_region(0x77, 1, DRVNAME))
    + return -ENODEV;
    +
    + rc = platform_driver_register(&ts5500leds_driver);
    + if (rc)
    + goto free_port;
    +
    + ts5500_pdev = platform_device_register_simple(DRVNAME, -1, NULL, 0);
    + if (IS_ERR(ts5500_pdev)) {
    + rc = PTR_ERR(ts5500_pdev);
    + goto driver_unreg;
    + }
    +
    + return 0;
    +
    +driver_unreg:
    + platform_driver_unregister(&ts5500leds_driver);
    +free_port:
    + release_region(0x77, 1);
    +
    + return rc;
    +}
    +
    +static void __exit ts5500leds_exit(void)
    +{
    + platform_device_unregister(ts5500_pdev);
    + platform_driver_unregister(&ts5500leds_driver);
    + release_region(0x77, 1);
    +}
    +
    +module_init(ts5500leds_init);
    +module_exit(ts5500leds_exit);
    +
    +MODULE_AUTHOR("Sean Young ");
    +MODULE_DESCRIPTION("LED driver for the Technology Systems 5500");
    +MODULE_LICENSE("GPL");
    diff -urpN linux-2.6.26/drivers/leds/Makefile /home/sean/tiger/linux-2.6.26/drivers/leds/Makefile
    --- linux-2.6.26/drivers/leds/Makefile 2008-07-13 22:51:29.000000000 +0100
    +++ /home/sean/tiger/linux-2.6.26/drivers/leds/Makefile 2008-07-17 15:14:55.000000000 +0100
    @@ -21,6 +21,7 @@ obj-$(CONFIG_LEDS_CM_X270)
    obj-$(CONFIG_LEDS_CLEVO_MAIL) += leds-clevo-mail.o
    obj-$(CONFIG_LEDS_HP6XX) += leds-hp6xx.o
    obj-$(CONFIG_LEDS_FSG) += leds-fsg.o
    +obj-$(CONFIG_LEDS_TS5500) += leds-ts5500.o

    # LED Triggers
    obj-$(CONFIG_LEDS_TRIGGER_TIMER) += ledtrig-timer.o
    --
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/

  2. Re: [PATCH v2] Led driver for technologic systems 5500

    On Mon, 2008-07-21 at 21:14 +0000, Sean Young wrote:
    > Here is a patch for the led on the Technologic Systems 5500 board. The
    > first patch had a bug in an error path. Detection is still a bit flimsy; Is
    > there anything that can be done about that?


    I'm not really familiar with how x86 does the machine detection but I
    agree the detection looks a bit flimsy. Can anyone advise on how this
    hardware could be better detected?

    Cheers,

    Richard

    > Signed-off-by: Sean Young
    > ---
    > diff -urpN linux-2.6.26/drivers/leds/Kconfig /home/sean/tiger/linux-2.6.26/drivers/leds/Kconfig
    > --- linux-2.6.26/drivers/leds/Kconfig 2008-07-13 22:51:29.000000000 +0100
    > +++ /home/sean/tiger/linux-2.6.26/drivers/leds/Kconfig 2008-07-21 12:19:51.000000000 +0100
    > @@ -147,6 +147,16 @@ config LEDS_CLEVO_MAIL
    > To compile this driver as a module, choose M here: the
    > module will be called leds-clevo-mail.
    >
    > +config LEDS_TS5500
    > + tristate "LED Support for the TS-5500"
    > + depends on LEDS_CLASS && X86
    > + help
    > + This driver makes the led on the Technologic Systems 5500
    > + (and possibly other models) accessible for userspace.
    > +
    > + Note that this led is hardwired to be xor'ed with Compact Flash
    > + activity.
    > +
    > comment "LED Triggers"
    >
    > config LEDS_TRIGGERS
    > Binary files linux-2.6.26/drivers/leds/.leds-net48xx.c.swp and /home/sean/tiger/linux-2.6.26/drivers/leds/.leds-net48xx.c.swp differ
    > diff -urpN linux-2.6.26/drivers/leds/leds-ts5500.c /home/sean/tiger/linux-2.6.26/drivers/leds/leds-ts5500.c
    > --- linux-2.6.26/drivers/leds/leds-ts5500.c 1970-01-01 01:00:00.000000000 +0100
    > +++ /home/sean/tiger/linux-2.6.26/drivers/leds/leds-ts5500.c 2008-07-21 21:15:41.000000000 +0100
    > @@ -0,0 +1,115 @@
    > +/*
    > + * Copyright (C) 2008 Sean Young
    > + */
    > +
    > +#include
    > +#include
    > +#include
    > +#include
    > +#include
    > +#include
    > +
    > +#define DRVNAME "ts5500-leds"
    > +
    > +static struct platform_device *ts5500_pdev;
    > +
    > +static void ts5500_led_set(struct led_classdev *led_dev, enum led_brightness value)
    > +{
    > + outb(value ? 1 : 0, 0x77);
    > +}
    > +
    > +static struct led_classdev ts5500_led = {
    > + .name = "ts5500:green",
    > + .brightness_set = ts5500_led_set,
    > +};
    > +
    > +static int __devinit ts5500leds_probe(struct platform_device *pdev)
    > +{
    > + return led_classdev_register(&pdev->dev, &ts5500_led);
    > +}
    > +
    > +static int __devexit ts5500leds_remove(struct platform_device *pdev)
    > +{
    > + led_classdev_unregister(&ts5500_led);
    > + return 0;
    > +}
    > +
    > +#ifdef CONFIG_PM
    > +static int ts5500leds_suspend(struct platform_device *pdev, pm_message_t state)
    > +{
    > + led_classdev_suspend(&ts5500_led);
    > + return 0;
    > +}
    > +
    > +static int ts5500leds_resume(struct platform_device *pdev)
    > +{
    > + led_classdev_resume(&ts5500_led);
    > + return 0;
    > +}
    > +#endif
    > +
    > +static struct platform_driver ts5500leds_driver = {
    > + .driver = {
    > + .name = DRVNAME,
    > + .owner = THIS_MODULE,
    > + },
    > + .probe = ts5500leds_probe,
    > + .remove = __devexit_p(ts5500leds_remove),
    > +#ifdef CONFIG_PM
    > + .suspend = ts5500leds_suspend,
    > + .resume = ts5500leds_resume,
    > +#endif
    > +};
    > +
    > +
    > +static int __init ts5500leds_init(void)
    > +{
    > + int rc;
    > + uint8_t product_code;
    > +
    > + if (!request_region(0x74, 1, DRVNAME))
    > + return -ENODEV;
    > +
    > + product_code = inb(0x74);
    > +
    > + release_region(0x74, 1);
    > +
    > + if (product_code != 0x60)
    > + return -ENODEV;
    > +
    > + if (!request_region(0x77, 1, DRVNAME))
    > + return -ENODEV;
    > +
    > + rc = platform_driver_register(&ts5500leds_driver);
    > + if (rc)
    > + goto free_port;
    > +
    > + ts5500_pdev = platform_device_register_simple(DRVNAME, -1, NULL, 0);
    > + if (IS_ERR(ts5500_pdev)) {
    > + rc = PTR_ERR(ts5500_pdev);
    > + goto driver_unreg;
    > + }
    > +
    > + return 0;
    > +
    > +driver_unreg:
    > + platform_driver_unregister(&ts5500leds_driver);
    > +free_port:
    > + release_region(0x77, 1);
    > +
    > + return rc;
    > +}
    > +
    > +static void __exit ts5500leds_exit(void)
    > +{
    > + platform_device_unregister(ts5500_pdev);
    > + platform_driver_unregister(&ts5500leds_driver);
    > + release_region(0x77, 1);
    > +}
    > +
    > +module_init(ts5500leds_init);
    > +module_exit(ts5500leds_exit);
    > +
    > +MODULE_AUTHOR("Sean Young ");
    > +MODULE_DESCRIPTION("LED driver for the Technology Systems 5500");
    > +MODULE_LICENSE("GPL");
    > diff -urpN linux-2.6.26/drivers/leds/Makefile /home/sean/tiger/linux-2.6.26/drivers/leds/Makefile
    > --- linux-2.6.26/drivers/leds/Makefile 2008-07-13 22:51:29.000000000 +0100
    > +++ /home/sean/tiger/linux-2.6.26/drivers/leds/Makefile 2008-07-17 15:14:55.000000000 +0100
    > @@ -21,6 +21,7 @@ obj-$(CONFIG_LEDS_CM_X270)
    > obj-$(CONFIG_LEDS_CLEVO_MAIL) += leds-clevo-mail.o
    > obj-$(CONFIG_LEDS_HP6XX) += leds-hp6xx.o
    > obj-$(CONFIG_LEDS_FSG) += leds-fsg.o
    > +obj-$(CONFIG_LEDS_TS5500) += leds-ts5500.o
    >
    > # LED Triggers
    > obj-$(CONFIG_LEDS_TRIGGER_TIMER) += ledtrig-timer.o
    >


    --
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/

  3. Re: [PATCH v2] Led driver for technologic systems 5500

    On Tue, Jul 22, 2008 at 09:31:32AM +0100, Richard Purdie wrote:
    > On Mon, 2008-07-21 at 21:14 +0000, Sean Young wrote:
    > > Here is a patch for the led on the Technologic Systems 5500 board. The
    > > first patch had a bug in an error path. Detection is still a bit flimsy; Is
    > > there anything that can be done about that?

    >
    > I'm not really familiar with how x86 does the machine detection but I
    > agree the detection looks a bit flimsy. Can anyone advise on how this
    > hardware could be better detected?


    I think that's why DMI was invented, AFAIK. I have no idea whether how many
    X86 devices exist on these I/O ports (0x77/0x74). The only other thing I can
    think of is that this is an AMD Elan so the Kconfig could depend on ELAN
    and not X86.


    Sean
    --
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/

  4. Re: [PATCH v2] Led driver for technologic systems 5500

    > I think that's why DMI was invented, AFAIK. I have no idea whether how many
    > X86 devices exist on these I/O ports (0x77/0x74). The only other thing I can
    > think of is that this is an AMD Elan so the Kconfig could depend on ELAN
    > and not X86


    If it is ELAN hardware then checking for an ELAN system would certainly
    cut down a lot on the misdetects. Does the box have other unique
    signatures you might detect. Good candidates we've used in the past have
    been

    - PCI subvendor identifiers
    - BIOS strings (eg does the BIOS contain the sequence 'Technologic
    systems 5500' or similar anywhere)

    Alan
    --
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/

+ Reply to Thread