Re: [PATCHv7 1/2] pwm: Add Allwinner SoC support

[Thierry Reding <thierry.reding@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 1974 bytes --]

On Wed, Sep 17, 2014 at 09:53:57PM +0200, Alexandre Belloni wrote:
[...]
> diff --git a/drivers/pwm/pwm-sunxi.c b/drivers/pwm/pwm-sunxi.c
> new file mode 100644
> index 000000000000..643f84ea013e
> --- /dev/null
> +++ b/drivers/pwm/pwm-sunxi.c
> @@ -0,0 +1,371 @@
> +/*
> + * Driver for Allwinner Pulse Width Modulation Controller
> + *
> + * Copyright (C) 2014 Alexandre Belloni <alexandre.belloni@free-electrons.com>

This is somewhat weird. So you are the copyright holder, not your
employer? The email address is somewhat misleading.

> +#define BIT_CH(bit, chan)	(bit << (chan * PWMCH_OFFSET))

bit and chan could use additional parentheses here for extra safety.

> +static const u32 prescaler_table[] = {
> +	120,
> +	180,
> +	240,
> +	360,
> +	480,
> +	0,
> +	0,
> +	0,
> +	12000,
> +	24000,
> +	36000,
> +	48000,
> +	72000,
> +	0,
> +	0,
> +	0, /* Actually 1 but tested separately */
> +};

Did I already mention that this was really weird?

> +static inline u32 sunxi_pwm_readl(struct sunxi_pwm_chip *chip,
> +				  unsigned long offset)
> +{
> +	return readl(chip->base + offset);
> +}
> +
> +

There's a gratuitous blank line here.

> +static int sunxi_pwm_probe(struct platform_device *pdev)
> +{
[...]
> +
> +	ret = pwmchip_add(&pwm->chip);
> +	if (ret < 0) {
> +		dev_err(&pdev->dev, "failed to add PWM chip %d\n", ret);

"chip: %d" to make it clear that it's not some kind of chip ID.

> +		goto error;
> +	}
> +
> +	ret = clk_prepare_enable(pwm->clk);
> +	if (ret) {
> +		dev_err(&pdev->dev, "failed to enable PWM clock\n");
> +		goto error;

Don't you want to remove the stale PWM chip here? Why not do this at the
very end, after everything's been set up properly?

> +MODULE_ALIAS("platform:sunxi-pwm");
> +MODULE_AUTHOR("Alexandre Belloni <alexandre.belloni@free-electrons.com>");
> +MODULE_DESCRIPTION("Allwinner PWM driver");

Perhaps "Allwinner sunxi PWM driver"? Presumably Allwinner could at some
point release a different family of SoCs.

Thierry

[-- Attachment #2: Type: application/pgp-signature, Size: 819 bytes --]