[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20120520100015.GC20652@opensource.wolfsonmicro.com>
Date: Sun, 20 May 2012 11:00:15 +0100
From: Mark Brown <broonie@...nsource.wolfsonmicro.com>
To: Yadwinder Singh Brar <yadi.brar01@...il.com>
Cc: linux-kernel@...r.kernel.org, linux-samsung-soc@...r.kernel.org,
linux-arm-kernel@...ts.infradead.org, Liam Girdwood <lrg@...com>,
Yadwinder Singh Brar <yadi.brar@...sung.com>
Subject: Re: [PATCH v2 2/2] regulator: Add support for MAX77686.
On Thu, May 17, 2012 at 07:09:27PM +0530, Yadwinder Singh Brar wrote:
Looks mostly good. A couple of fairly small things:
> +static int max77686_get_voltage_sel(struct regulator_dev *rdev)
> +{
This looks like it should be regulator_get_voltage_sel_regmap().
> +static int max77686_set_voltage_sel(struct regulator_dev *rdev,
> + unsigned sel)
> +{
This looks like it should be regulator_set_voltage_sel_regmap().
> + max77686->ramp_delay = pdata->ramp_delay - 1;
> + max77686_update_reg(i2c, MAX77686_REG_BUCK2CTRL1,
> + RAMP_VALUE, RAMP_MASK);
> + max77686_update_reg(i2c, MAX77686_REG_BUCK3CTRL1,
> + RAMP_VALUE, RAMP_MASK);
> + max77686_update_reg(i2c, MAX77686_REG_BUCK4CTRL1,
> + RAMP_VALUE, RAMP_MASK);
This code is unclear because RAMP_VALUE looks like a constant that has
nothing to do with ramp_delay when in fact it's actually this:
> +#define RAMP_VALUE (max77686->ramp_delay << 6)
which isn't constant - this is why I queried this last time. Just
remove the define and write this out directly. The way the code is
written at the minute it looks like ramp_delay is just stored and not
referred to unless you go searching around the rest of the driver.
Download attachment "signature.asc" of type "application/pgp-signature" (837 bytes)
Powered by blists - more mailing lists