Re: joypad interrupt question

Light bulb on. Thanks again.

E

On Tue, Mar 29, 2005 at 10:07:03PM +0100, Phil Blundell wrote:
> On Tue, 2005-03-29 at 11:57 -0800, Erik Hovland wrote:
> > - The GPIO joypad interrupts are set up to interrupt on rising and
> > falling edge and for 'maximum' filtering.
> > 1. What is 'maximum' filtering or what is filtering in this context?
>
> I think the "filtering" in this context is glitch suppression, or
> debouncing. You should be able to find some details in the SAMCOP
> and/or s3c2410 datasheets.
>
> > In the case of say the center press of the joypad the code attempts to
> > register a 'action' key.
> > 2. If someone pressed the center of the joypad I would expect there to
> > to be a interrupt generated for the rising edge and for the isr to
> > read a 1 from the proper bit for the center press. Then another
> > interrupt generated for the falling edge and for the isr to read a
> > 1 again from the proper bit. The isr would use make_keypress for
> > 'action' and 1 for the rising edge and a make_keypress for 'action'
> > and 0 for the falling edge. Is this sequence of events correct?
>
> Yes, except that it's the other way around. The bit is 1 when the key
> is not pressed, and goes to 0 when the key is pressed.
>
> > - The function gpio_to_keys flips the bits of the value read from the
> > GPA data register, why? That register shouldn't have anything
> > different then for application buttons which don't do this to the
> > value read.
>
> For the same reason as above: these are active low signals. Actually,
> the code does do this for the application buttons too, just not in quite
> the same way. This line:
>
> down = (h5400_asic_read_register (H5400_ASIC_GPIO_GPA_DAT) &
> b->bitmask) ? 0 : 1;
>
> will yield 0 if the bit in the register is set, and 1 if the bit is
> clear.
>
> > - The map of buttons is written such that if the joypad NW press is done
> > there is the possibility for both a 'up' and a 'left' key press. Is
> > this done for a reason?
>
> Older iPAQs were able to detect "diagonal" button presses and generate
> the appropriate pair of key events. There are a few applications for
> which this is handy. You might be right that trying to do the same on
> the h5400 is causing more trouble than it's worth.
>
> p.
>
>

-- 
Erik Hovland
mail: erik AT hovland DOT org
web: http://hovland.org/
PGP/GPG public key available on request