Re: [PATCH 12/19] iio & mfd & input: ti_tscadc: Match mfd sub devices to regmap interface

What make you think I am the original author?

I just fixed whatever I had to fix to get it working.

Namely that the multi function device was using regmap in one place and
not in another, which had the nice effect of the regmap cache getting out of sync.

Regards

-- Pantelis

On Jun 4, 2013, at 1:52 PM, Sebastian Andrzej Siewior wrote:

* Pantelis Antoniou | 2013-06-04 13:25:28 [+0300]:

quoted

quoted

quoted

quoted

diff --git a/drivers/mfd/ti_am335x_tscadc.c b/drivers/mfd/ti_am335x_tscadc.c
index bd127bd..a27401a 100644
--- a/drivers/mfd/ti_am335x_tscadc.c
+++ b/drivers/mfd/ti_am335x_tscadc.c

@@ -31,6 +31,7 @@ static unsigned int tscadc_readl(struct ti_tscadc_dev *tsadc, unsigned int reg)

{
	unsigned int val;

????  What is this doing here? It's not doing the move to regmap but rather setting a default value.

No lo sé :) I saw that and was wondering myself a little and planned the
romval of this default for later. Now I think I do this earlier.

The tscadc_read function doesn't have a failure mode; doesn't return an error in case of a regmap_read
fail. Rather that change all the callers, we make sure we don't ever return an uninitialized value 
in case of an error (when val should be unchanged).

One stupid question: Why did you start using regmap in the first place?

quoted

quoted

quoted

quoted

+	val = (unsigned int)-1;
	regmap_read(tsadc->regmap_tscadc, reg, &val);
	return val;
}

Sebastian