Re: [PATCH 2/2] Input: tsc2007 - Add a z1_low_threshhold platform data parameter

[Feng Tang <feng.tang@intel.com>]

Hi Dmitry,

On Mon, 5 Dec 2011 13:35:41 +0800
Feng Tang <feng.tang@intel.com> wrote:

> > > 			/*
> > > 			 * Sample found inconsistent by debouncing
> > > or pressure is
> > > 			 * beyond the maximum. Don't report it to
> > > user space,
> > > 			 * repeat at least once more the
> > > measurement. */
> > > 			dev_dbg(&ts->client->dev, "ignored pressure
> > > %d\n", rt); }
> > 
> > What I meant that we need to ajust the logic to _exit_ the loop if we
> > receive several samples with rt > max_rt instead of adding a new
> > parameter.
> 
> Yes, I did try a similar way to set a retry limit which also works
> basically, code is like this
> 
> @@ -169,6 +169,7 @@ static irqreturn_t tsc2007_soft_irq(int irq, void *handle)
>         struct tsc2007 *ts = handle;
>         struct input_dev *input = ts->input;
>         struct ts_event tc;
> +       u32 max_retry = 2;
>         u32 rt;
>  
>         while (!ts->stopped && tsc2007_is_pen_down(ts)) {
> @@ -206,6 +207,8 @@ static irqreturn_t tsc2007_soft_irq(int irq, void *handle)
>                          * repeat at least once more the measurement.
>                          */
>                         dev_dbg(&ts->client->dev, "ignored pressure %d\n",
> rt);
> +                       if (!--max_retry)
> +                               break;
>                 }
> 
> 
> But I still have some concerns:
> 1. How many retries should we try? the tsc2007_read_values() will take about
> 70 ms on our platform, plus the "poll_period", one retry will take about
> 100ms. 2. I checked the noise data, its z1 value is always in a range from 9
> to 13, while the real data's z1 is always bigger than 300. So I think there
> is a very clear gap to tell the noise data from valid data by z1 value.
> 
> How do you think about it?

Any further comments?

Thanks,
Feng