pwm: pwm-mtk-disp: Switch to regmap for mmio access

Switch to using the generic regmap API instead of calls to readl/writel
for MMIO register access, allowing us to reduce code size.

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
---
 drivers/pwm/pwm-mtk-disp.c | 77 ++++++++++++++++----------------------
 1 file changed, 32 insertions(+), 45 deletions(-)

On Mon, Jan 03, 2022 at 04:35:06PM +0100, AngeloGioacchino Del Regno wrote:
> Switch to using the generic regmap API instead of calls to readl/writel
> for MMIO register access, allowing us to reduce code size.
> 
> Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
> ---
>  drivers/pwm/pwm-mtk-disp.c | 77 ++++++++++++++++----------------------
>  1 file changed, 32 insertions(+), 45 deletions(-)
> 
> diff --git a/drivers/pwm/pwm-mtk-disp.c b/drivers/pwm/pwm-mtk-disp.c
> index c605013e4114..4a6fab144cee 100644
> --- a/drivers/pwm/pwm-mtk-disp.c
> +++ b/drivers/pwm/pwm-mtk-disp.c
> @@ -14,6 +14,7 @@
>  #include <linux/of_device.h>
>  #include <linux/platform_device.h>
>  #include <linux/pwm.h>
> +#include <linux/regmap.h>
>  #include <linux/slab.h>
>  
>  #define DISP_PWM_EN		0x00
> @@ -47,7 +48,7 @@ struct mtk_disp_pwm {
>  	const struct mtk_pwm_data *data;
>  	struct clk *clk_main;
>  	struct clk *clk_mm;
> -	void __iomem *base;
> +	struct regmap *regmap;
>  	bool enabled;
>  };
>  
> @@ -56,22 +57,11 @@ static inline struct mtk_disp_pwm *to_mtk_disp_pwm(struct pwm_chip *chip)
>  	return container_of(chip, struct mtk_disp_pwm, chip);
>  }
>  
> -static void mtk_disp_pwm_update_bits(struct mtk_disp_pwm *mdp, u32 offset,
> -				     u32 mask, u32 data)
> -{
> -	void __iomem *address = mdp->base + offset;
> -	u32 value;
> -
> -	value = readl(address);
> -	value &= ~mask;
> -	value |= data;
> -	writel(value, address);
> -}
> -
>  static int mtk_disp_pwm_apply(struct pwm_chip *chip, struct pwm_device *pwm,
>  			      const struct pwm_state *state)
>  {
>  	struct mtk_disp_pwm *mdp = to_mtk_disp_pwm(chip);
> +	const struct mtk_pwm_data *pdata = mdp->data;
>  	u32 clk_div, period, high_width, value;
>  	u64 div, rate;
>  	int err;
> @@ -80,8 +70,7 @@ static int mtk_disp_pwm_apply(struct pwm_chip *chip, struct pwm_device *pwm,
>  		return -EINVAL;
>  
>  	if (!state->enabled) {
> -		mtk_disp_pwm_update_bits(mdp, DISP_PWM_EN, mdp->data->enable_mask,
> -					 0x0);
> +		regmap_clear_bits(mdp->regmap, DISP_PWM_EN, pdata->enable_mask);
>  
>  		if (mdp->enabled) {
>  			clk_disable_unprepare(mdp->clk_mm);
> @@ -138,37 +127,25 @@ static int mtk_disp_pwm_apply(struct pwm_chip *chip, struct pwm_device *pwm,
>  	high_width = mul_u64_u64_div_u64(state->duty_cycle, rate, div);
>  	value = period | (high_width << PWM_HIGH_WIDTH_SHIFT);
>  
> -	mtk_disp_pwm_update_bits(mdp, mdp->data->con0,
> -				 PWM_CLKDIV_MASK,
> -				 clk_div << PWM_CLKDIV_SHIFT);
> -	mtk_disp_pwm_update_bits(mdp, mdp->data->con1,
> -				 PWM_PERIOD_MASK | PWM_HIGH_WIDTH_MASK,
> -				 value);
> +	regmap_update_bits(mdp->regmap, pdata->con0, PWM_CLKDIV_MASK,
> +			   clk_div << PWM_CLKDIV_SHIFT);
> +	regmap_update_bits(mdp->regmap, pdata->con1,
> +			   PWM_PERIOD_MASK | PWM_HIGH_WIDTH_MASK, value);
>  
>  	if (mdp->data->has_commit) {
> -		mtk_disp_pwm_update_bits(mdp, mdp->data->commit,
> -					 mdp->data->commit_mask,
> -					 mdp->data->commit_mask);
> -		mtk_disp_pwm_update_bits(mdp, mdp->data->commit,
> -					 mdp->data->commit_mask,
> -					 0x0);
> +		regmap_set_bits(mdp->regmap, pdata->commit, pdata->commit_mask);
> +		regmap_clear_bits(mdp->regmap, pdata->commit, pdata->commit_mask);
>  	} else {
>  		/*
>  		 * For MT2701, disable double buffer before writing register
>  		 * and select manual mode and use PWM_PERIOD/PWM_HIGH_WIDTH.
>  		 */
> -		mtk_disp_pwm_update_bits(mdp, mdp->data->bls_debug,
> -					 mdp->data->bls_debug_mask,
> -					 mdp->data->bls_debug_mask);
> -		mtk_disp_pwm_update_bits(mdp, mdp->data->con0,
> -					 mdp->data->con0_sel,
> -					 mdp->data->con0_sel);
> +		regmap_set_bits(mdp->regmap, pdata->bls_debug, pdata->bls_debug_mask);
> +		regmap_set_bits(mdp->regmap, pdata->con0, pdata->con0_sel);
>  	}
>  
> -	mtk_disp_pwm_update_bits(mdp, DISP_PWM_EN, mdp->data->enable_mask,
> -				 mdp->data->enable_mask);
> +	regmap_set_bits(mdp->regmap, DISP_PWM_EN, pdata->enable_mask);
>  	mdp->enabled = true;
> -

Was the empty line here removed on purpose? I would have kept it.

>  	return 0;
>  }
>  
> @@ -195,8 +172,8 @@ static void mtk_disp_pwm_get_state(struct pwm_chip *chip,
>  	}
>  
>  	rate = clk_get_rate(mdp->clk_main);
> -	con0 = readl(mdp->base + mdp->data->con0);
> -	con1 = readl(mdp->base + mdp->data->con1);
> +	regmap_read(mdp->regmap, mdp->data->con0, &con0);
> +	regmap_read(mdp->regmap, mdp->data->con1, &con1);
>  	state->enabled = !!(con0 & BIT(0));
>  	clk_div = FIELD_GET(PWM_CLKDIV_MASK, con0);
>  	period = FIELD_GET(PWM_PERIOD_MASK, con1);
> @@ -219,9 +196,17 @@ static const struct pwm_ops mtk_disp_pwm_ops = {
>  	.owner = THIS_MODULE,
>  };
>  
> +static const struct regmap_config mtk_disp_pwm_regmap_config = {
> +	.reg_bits = 32,
> +	.reg_stride = 4,
> +	.val_bits = 32,
> +	.disable_locking = true,
> +};
> +
>  static int mtk_disp_pwm_probe(struct platform_device *pdev)
>  {
>  	struct mtk_disp_pwm *mdp;
> +	void __iomem *base;
>  	int ret;
>  
>  	mdp = devm_kzalloc(&pdev->dev, sizeof(*mdp), GFP_KERNEL);
> @@ -230,9 +215,13 @@ static int mtk_disp_pwm_probe(struct platform_device *pdev)
>  
>  	mdp->data = of_device_get_match_data(&pdev->dev);
>  
> -	mdp->base = devm_platform_ioremap_resource(pdev, 0);
> -	if (IS_ERR(mdp->base))
> -		return PTR_ERR(mdp->base);
> +	base = devm_platform_ioremap_resource(pdev, 0);
> +	if (IS_ERR(base))
> +		return PTR_ERR(base);
> +
> +	mdp->regmap = devm_regmap_init_mmio(&pdev->dev, base, &mtk_disp_pwm_regmap_config);
> +	if (IS_ERR(mdp->regmap))
> +		return PTR_ERR(mdp->regmap);
>  
>  	mdp->clk_main = devm_clk_get(&pdev->dev, "main");
>  	if (IS_ERR(mdp->clk_main))
> @@ -247,10 +236,8 @@ static int mtk_disp_pwm_probe(struct platform_device *pdev)
>  	mdp->chip.npwm = 1;
>  
>  	ret = pwmchip_add(&mdp->chip);
> -	if (ret < 0) {
> -		dev_err(&pdev->dev, "pwmchip_add() failed: %pe\n", ERR_PTR(ret));
> -		return ret;
> -	}
> +	if (ret < 0)
> +		return dev_err_probe(&pdev->dev, ret, "pwmchip_add() failed\n");

I wonder that you convert this to dev_err_probe, but don't use it in the
hunk above. If I'm not mistaken devm_regmap_init_mmio doesn't emit an
error message, so an error message would be appropriate there.

Best regards
Uwe