Re: [PATCH v2] dmaengine: mv_xor_v2: Fix clock resource by adding a register clock

[Gregory CLEMENT <gregory.clement@bootlin.com>]

Hi Gregory,
 
 On mer., mars 07 2018, Gregory CLEMENT <gregory.clement@bootlin.com> wrote:

> On the CP110 components which are present on the Armada 7K/8K SoC we need
> to explicitly enable the clock for the registers. However it is not
> needed for the AP8xx component, that's why this clock is optional.
>
> With this patch both clocks have now a name, but in order to be
> backward compatible, the name of the first clock is not used. It
> allows to still use this clock with a device tree using the old
> binding.
>
> Signed-off-by: Gregory CLEMENT <gregory.clement@bootlin.com>
> ---
> Changelog:
>
> v1 -> v2
>
>  - Fix typos in the commit log
>  - Add more explanation in te commit log
>  - Only skip the reg clock usage if the clock is not presetn in the
>    device tree (-ENOENT)
>  - Add dt list in CC
>
>
>  .../devicetree/bindings/dma/mv-xor-v2.txt          |  6 +++++-
>  drivers/dma/mv_xor_v2.c                            | 25 +++++++++++++++++-----
>  2 files changed, 25 insertions(+), 6 deletions(-)
>
> diff --git a/Documentation/devicetree/bindings/dma/mv-xor-v2.txt b/Documentation/devicetree/bindings/dma/mv-xor-v2.txt
> index 217a90eaabe7..9c38bbe7e6d7 100644
> --- a/Documentation/devicetree/bindings/dma/mv-xor-v2.txt
> +++ b/Documentation/devicetree/bindings/dma/mv-xor-v2.txt
> @@ -11,7 +11,11 @@ Required properties:
>    interrupts.
>  
>  Optional properties:
> -- clocks: Optional reference to the clock used by the XOR engine.
> +- clocks: Optional reference to the clocks used by the XOR engine.
> +- clock-names: mandatory if there is a second clock, in this case the
> +   name must be "core" for the first clock and "reg" for the second
> +   one
> +
>  
>  Example:
>  
> diff --git a/drivers/dma/mv_xor_v2.c b/drivers/dma/mv_xor_v2.c
> index f652a0e0f5a2..e3e854161dd0 100644
> --- a/drivers/dma/mv_xor_v2.c
> +++ b/drivers/dma/mv_xor_v2.c
> @@ -163,6 +163,7 @@ struct mv_xor_v2_device {
>  	void __iomem *dma_base;
>  	void __iomem *glob_base;
>  	struct clk *clk;
> +	struct clk *reg_clk;
>  	struct tasklet_struct irq_tasklet;
>  	struct list_head free_sw_desc;
>  	struct dma_device dmadev;
> @@ -749,13 +750,26 @@ static int mv_xor_v2_probe(struct platform_device *pdev)
>  	if (ret)
>  		return ret;
>  
> +	xor_dev->reg_clk = devm_clk_get(&pdev->dev, "reg");
> +	if (PTR_ERR(xor_dev->reg_clk) == -ENOENT) {

It should be "!= " there, a v3 is coming

Sorry for the noise,

Gregory

> +		if (!IS_ERR(xor_dev->reg_clk)) {
> +			ret = clk_prepare_enable(xor_dev->reg_clk);
> +			if (ret)
> +				return ret;
> +		} else {
> +			return PTR_ERR(xor_dev->reg_clk);
> +		}
> +	}
> +
>  	xor_dev->clk = devm_clk_get(&pdev->dev, NULL);
> -	if (IS_ERR(xor_dev->clk) && PTR_ERR(xor_dev->clk) == -EPROBE_DEFER)
> -		return -EPROBE_DEFER;
> +	if (IS_ERR(xor_dev->clk) && PTR_ERR(xor_dev->clk) == -EPROBE_DEFER) {
> +		ret = EPROBE_DEFER;
> +		goto disable_reg_clk;
> +	}
>  	if (!IS_ERR(xor_dev->clk)) {
>  		ret = clk_prepare_enable(xor_dev->clk);
>  		if (ret)
> -			return ret;
> +			goto disable_reg_clk;
>  	}
>  
>  	ret = platform_msi_domain_alloc_irqs(&pdev->dev, 1,
> @@ -866,8 +880,9 @@ static int mv_xor_v2_probe(struct platform_device *pdev)
>  free_msi_irqs:
>  	platform_msi_domain_free_irqs(&pdev->dev);
>  disable_clk:
> -	if (!IS_ERR(xor_dev->clk))
> -		clk_disable_unprepare(xor_dev->clk);
> +	clk_disable_unprepare(xor_dev->clk);
> +disable_reg_clk:
> +	clk_disable_unprepare(xor_dev->reg_clk);
>  	return ret;
>  }
>  
> -- 
> 2.16.1
>
> --
> To unsubscribe from this list: send the line "unsubscribe dmaengine" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 
Gregory Clement, Bootlin (formerly Free Electrons)
Embedded Linux and Kernel engineering
http://bootlin.com