Le dimanche 19 octobre 2014 à 15:59 +0200, Hans de Goede a écrit : > Hi, > > On 10/08/2014 07:11 PM, Paul Kocialkowski wrote: > > Signed-off-by: Paul Kocialkowski <[email protected]> > > --- > > drivers/input/keyboard/sun4i-keyboard.c | 4 ++-- > > 1 file changed, 2 insertions(+), 2 deletions(-) > > > > diff --git a/drivers/input/keyboard/sun4i-keyboard.c > > b/drivers/input/keyboard/sun4i-keyboard.c > > index 74869e1..fa3bd45 100644 > > --- a/drivers/input/keyboard/sun4i-keyboard.c > > +++ b/drivers/input/keyboard/sun4i-keyboard.c > > @@ -194,7 +194,7 @@ static void sun4i_keyboard_resume(struct early_suspend > > *h) > > printk("enter laterresume: sun4i_keyboard_resume. \n"); > > #endif > > > > - > > writel(FIRST_CONCERT_DLY|LEVELB_VOL|KEY_MODE_SELECT|LRADC_HOLD_EN|ADC_CHAN_SELECT|LRADC_SAMPLE_62HZ|LRADC_EN,KEY_BASSADDRESS > > + LRADC_CTRL); > > + > > writel(FIRST_CONCERT_DLY|LEVELB_VOL|KEY_MODE_SELECT|LRADC_HOLD_EN|ADC_CHAN_SELECT|LRADC_SAMPLE_250HZ|LRADC_EN,KEY_BASSADDRESS > > + LRADC_CTRL); > > return ; > > } > > #else > > Hmm, all the debounce times are in samplerate times to, so this seriously > shortens the > debounce time, which is not a good idea IMHO, this will make the debounce so > short, that > we are likely to see false keypresses from bounces.
I agree, a high frequency involves a shorter debounce time, but it doesn't necessarily mean that this is a problem. I haven't seen any false key report on any of the 4 devices I've tested this on. Seeing how slow 62Hz was, I'm confident 250Hz still provides enough time to have stable reads. Please let me know if you have results that show otherwise. I do need that patch to be able to use navigation by keys on CloclworkMod recovery. Any value lower than 250Hz requires a break of half a second between 2 presses to make them distinguishable. This is unbearable. -- Paul Kocialkowski, Replicant developer Replicant is a fully free Android distribution Website: http://www.replicant.us/ Redmine: http://redmine.replicant.us/
signature.asc
Description: This is a digitally signed message part
