linux-kernel - Re: [PATCH v5 3/3] pwm: mtk_disp: implement atomic API .get

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20210625175907.mkm5hef3rqyn2xu2@pengutronix.de>
Date:   Fri, 25 Jun 2021 19:59:07 +0200
From:   Uwe Kleine-König <u.kleine-koenig@...gutronix.de>
To:     Jitao Shi <jitao.shi@...iatek.com>
Cc:     Thierry Reding <thierry.reding@...il.com>,
        Lee Jones <lee.jones@...aro.org>,
        Matthias Brugger <matthias.bgg@...il.com>,
        linux-pwm@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
        linux-mediatek@...ts.infradead.org, linux-kernel@...r.kernel.org,
        srv_heupstream@...iatek.com, yingjoe.chen@...iatek.com,
        eddie.huang@...iatek.com, cawa.cheng@...iatek.com,
        bibby.hsieh@...iatek.com, ck.hu@...iatek.com, stonea168@....com,
        huijuan.xie@...iatek.com
Subject: Re: [PATCH v5 3/3] pwm: mtk_disp: implement atomic API .get_state()

On Wed, Jun 16, 2021 at 04:52:24PM +0800, Jitao Shi wrote:
> Switch the driver to support the .get_state() method.
> 
> Signed-off-by: Jitao Shi <jitao.shi@...iatek.com>
> ---
>  drivers/pwm/pwm-mtk-disp.c | 39 ++++++++++++++++++++++++++++++++++++++
>  1 file changed, 39 insertions(+)
> 
> diff --git a/drivers/pwm/pwm-mtk-disp.c b/drivers/pwm/pwm-mtk-disp.c
> index a4766e931b68..d60a6b379683 100644
> --- a/drivers/pwm/pwm-mtk-disp.c
> +++ b/drivers/pwm/pwm-mtk-disp.c
> @@ -159,8 +159,47 @@ static int mtk_disp_pwm_apply(struct pwm_chip *chip, struct pwm_device *pwm,
>  	return 0;
>  }
>  
> +static void mtk_disp_pwm_get_state(struct pwm_chip *chip,
> +				   struct pwm_device *pwm,
> +				   struct pwm_state *state)
> +{
> +	struct mtk_disp_pwm *mdp = to_mtk_disp_pwm(chip);
> +	u32 clk_div, con0, con1;
> +	u64 rate, period, high_width;
> +	int err;
> +
> +	if (!mdp->enabled) {
> +		err = clk_prepare_enable(mdp->clk_main);
> +		if (err < 0) {
> +			dev_err(chip->dev, "Can't enable mdp->clk_main: %pe\n", ERR_PTR(err));
> +			return;
> +		}
> +		err = clk_prepare_enable(mdp->clk_mm);
> +		if (err < 0) {
> +			dev_err(chip->dev, "Can't enable mdp->clk_mm: %pe\n", ERR_PTR(err));
> +			clk_disable_unprepare(mdp->clk_main);
> +			return;
> +		}
> +	}

As already written in reply to one of the previous patches: The
clk_prepare_enable() can be done unconditionally.

> +	rate = clk_get_rate(mdp->clk_main);
> +	con0 = readl(mdp->base + mdp->data->con0);
> +	con1 = readl(mdp->base + mdp->data->con1);
> +	state->enabled = !!(con0 & BIT(0));

Maybe introduce a name for that.

> +	clk_div = FIELD_GET(PWM_CLKDIV_MASK, con0);
> +	period = con1 & PWM_PERIOD_MASK;

FIELD_GET(PWM_PERIOD_MASK, con1) for consistency.

> +	state->period = DIV64_U64_ROUND_UP(period * (clk_div + 1) * NSEC_PER_SEC, rate);

period has 12 bits, clk_div 11 and NSEC_PER_SEC has 30, so this cannot
overflow. Maybe mention this in a comment.

> +	high_width = FIELD_GET(PWM_HIGH_WIDTH_MASK, con1);
> +	state->duty_cycle = DIV64_U64_ROUND_UP(high_width * (clk_div + 1) * NSEC_PER_SEC,
> +					       rate);

state->polarity = PWM_POLARITY_NORMAL;

> +	if (!mdp->enabled) {
> +		clk_disable_unprepare(mdp->clk_mm);
> +		clk_disable_unprepare(mdp->clk_main);
> +	}
> +}
> +
>  static const struct pwm_ops mtk_disp_pwm_ops = {
>  	.apply = mtk_disp_pwm_apply,
> +	.get_state = mtk_disp_pwm_get_state,
>  	.owner = THIS_MODULE,
>  };

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | https://www.pengutronix.de/ |

Download attachment "signature.asc" of type "application/pgp-signature" (489 bytes)