Re: [PATCH v2] rfkill-input doesn't work until 5 minutes after boot

[Andrew Morton <akpm@linux-foundation.org>]

On Sun, 5 Oct 2008 01:43:34 +0100 Matthew Garrett <mjg59@srcf.ucam.org> wrote:

> rfkill-input implements debounce as follows:
> 
>         if (time_after(jiffies, task->last + msecs_to_jiffies(200))) {
> 
> However, task->last is initialised to 0 while jiffies starts at -300*HZ. 
> Any input within 5 minutes of kernel start is therefore ignored. Fix by 
> initialising task->last correctly.
> 
> Signed-off-by: Matthew Garrett <mjg@redhat.com>
> 
> ---
> 
> Set the last event value at module load time, since otherwise we'll have 
> a window of failure if someone loads the module in a few hundred million 
> years. I look forward to being rewarded by the post-humans for caring so 
> much about them.

Jiffies wraparound is 49.7 days at HZ=1000.

> diff --git a/net/rfkill/rfkill-input.c b/net/rfkill/rfkill-input.c
> index e5b6955..86197bb 100644
> --- a/net/rfkill/rfkill-input.c
> +++ b/net/rfkill/rfkill-input.c
> @@ -255,6 +255,11 @@ static struct input_handler rfkill_handler = {
>  
>  static int __init rfkill_handler_init(void)
>  {
> +	rfkill_wlan.last = jiffies - HZ/5;
> +	rfkill_bt.last = jiffies - HZ/5;
> +	rfkill_uwb.last = jiffies - HZ/5;
> +	rfkill_wimax.last = jiffies - HZ/5;
> +	rfkill_wwan.last = jiffies - HZ/5;
>  	return input_register_handler(&rfkill_handler);
>  }

If someone adds a new rfkill_foo there's a risk that they'll forget to
make the corresponding change here.  A comment at the definition sites
would help.


Or, better, do something like

static struct rfkill_task rfkill_tasks[] = {
	DEFINE_RFKILL_TASK(RFKILL_TYPE_WLAN),
	...
};

#define rfkill_wlan	rfkill_tasks[0]
...

	for (i = 0; i < ARRAY_SIZE(rfkill_tasks); i++)
		...


but the new definition of DEFINE_RFKILL_TASK gets tricky.