Re: [PATCH v2 1/5] dmaengine: qcom: bam_dma: make bam clk optional

[Srinivas Kandagatla]

Thanks for the Review,

On 14/02/18 15:41, Bjorn Andersson wrote:
> >   
> >   	bdev->bamclk = devm_clk_get(bdev->dev, "bam_clk");
> > -	if (IS_ERR(bdev->bamclk))
> > -		return PTR_ERR(bdev->bamclk);
> > -
> > -	ret = clk_prepare_enable(bdev->bamclk);
> > -	if (ret) {
> > -		dev_err(bdev->dev, "failed to prepare/enable clock\n");
> > -		return ret;
> > +	if (IS_ERR(bdev->bamclk)) {
> In the case of !bdev->controlled_remotely I think this should still be
> an error.
Yep makes sense,

You mean something like this?

------------------------->cut<------------------------
        bdev->bamclk = devm_clk_get(bdev->dev, "bam_clk");
        if (IS_ERR(bdev->bamclk)) {
                if (!bdev->controlled_remotely);
                        return PTR_ERR(bdev->bamclk);

                bdev->bamclk = NULL;
        }

        ret = clk_prepare_enable(bdev->bamclk);
        if (ret) {
                dev_err(bdev->dev, "failed to prepare/enable clock\n");
                return ret;
        }

------------------------->cut<------------------------


> > +		bdev->bamclk = NULL;
> > +	} else {
> > +		ret = clk_prepare_enable(bdev->bamclk);
> > +		if (ret) {
> > +			dev_err(bdev->dev, "failed to prepare/enable clock\n");
> > +			return ret;
> > +		}
> The rest of the driver will keep operating the bamclk (which is okay),
> so for symmetry purposes I think you should just keep the
> clk_prepare_enable() block unmodified.
Yep, with above change, this block should be unchanged.


--srini
> Regards,
> Bjorn