Re: [PATCH v3 2/3] pwm: imx27: Use clk_bulk_*() API to simplify clock handling

[Frank Li <Frank.li@nxp.com>]

On Tue, Oct 22, 2024 at 08:53:40AM +0200, Uwe Kleine-König wrote:
> Hello,
>
> On Tue, Sep 10, 2024 at 03:07:19PM -0400, Frank Li wrote:
> > @@ -229,7 +209,7 @@ static int pwm_imx27_apply(struct pwm_chip *chip, struct pwm_device *pwm,
> >  	int ret;
> >  	u32 cr;
> >
> > -	clkrate = clk_get_rate(imx->clk_per);
> > +	clkrate = clk_get_rate(imx->clks[PWM_IMX27_PER].clk);
> >  	c = clkrate * state->period;
>
> Unrelated to this patch: clk_get_rate() should only be called on enabled
> clocks. Given that further down in that function (see next hunk)
> pwm_imx27_clk_prepare_enable() (or clk_bulk_prepare_enable()
> respectively) is called, that clk might be off?!
>
> >  	do_div(c, NSEC_PER_SEC);
> > @@ -259,7 +239,7 @@ static int pwm_imx27_apply(struct pwm_chip *chip, struct pwm_device *pwm,
> >  	if (pwm->state.enabled) {
> >  		pwm_imx27_wait_fifo_slot(chip, pwm);
> >  	} else {
> > -		ret = pwm_imx27_clk_prepare_enable(imx);
> > +		ret = clk_bulk_prepare_enable(imx->clks_cnt, imx->clks);
> >  		if (ret)
> >  			return ret;
> >
>
> I applied just this patch to
> https://git.kernel.org/pub/scm/linux/kernel/git/ukleinek/linux.git pwm/for-next
> . The others are still under discussion, right?

Yes, thanks. I think 32k is not necessary and need more research.

>
> I see you signed your patch (which is fine!), but I couldn't find your
> key, neither on hkps://keys.openpgp.org/ nor on

Thanks. Fixed.

> hkps://keyserver.ubuntu.com nor in the kernel keyring. At least the
> first two should be easy to fix.
>
> Best regards
> Uwe