Re: [PATCH v1 3/5] pinctrl: st: Add software edge trigger interrupt support.

[srinivas kandagatla <srinivas.kandagatla@st.com>]

Thankyou for reviewing the patch.

On 15/01/14 14:27, Linus Walleij wrote:
> On Tue, Jan 14, 2014 at 3:52 PM,  <srinivas.kandagatla@st.com> wrote:
> 
>> From: Srinivas Kandagatla <srinivas.kandagatla@st.com>
>>
>> ST pin controller does not have hardware support for detecting edge
>> triggered interrupts, It only has level triggering support.
>> This patch attempts to fake up edge triggers from hw level trigger
>> support in software. With this facility now the gpios can be easily used
>> for keypads, otherwise it would be difficult for drivers like keypads to
>> work with level trigger interrupts.
>>
>> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@st.com>
> 
> Clever! Mostly I like the patch as it is but:
> 
>>         for_each_set_bit(n, &port_active, BITS_PER_LONG) {
>> +               /* check if we are detecting fake edges ... */
>> +               pin_edge_cfg = ST_IRQ_EDGE_CONF(bank_edge_mask, n);
>> +
>> +               if (pin_edge_cfg) {
>> +                       /* edge detection. */
>> +                       val = st_gpio_get(&bank->gpio_chip, n);
>> +                       if (val)
>> +                               writel(BIT(n), bank->base + REG_PIO_SET_PCOMP);
>> +                       else
>> +                               writel(BIT(n), bank->base + REG_PIO_CLR_PCOMP);
>> +
>> +                       if (pin_edge_cfg != ST_IRQ_EDGE_BOTH &&
>> +                               !((pin_edge_cfg & ST_IRQ_EDGE_FALLING) ^ val))
>> +                                       continue;
>> +               }
>> +
> 
> Please insert comments here to explain what you are actually doing
> because I sure as hell do not understand this code without comments
> describing the trick used.

I agree, I will document this logic in next version.

Thanks,
srini
> 
> Yours,
> Linus Walleij
> 
>