Re: [PATCH v1 22/43] dma: cirrus: add DT support for Cirrus EP93xx

[andy.shevchenko@gmail.com]

Thu, Jun 01, 2023 at 08:45:27AM +0300, Nikita Shubin kirjoitti:
> - find register range from the device tree
> - get clocks, interrupts from device tree

...

> --- a/arch/arm/mach-ep93xx/dma.c
> +++ b/arch/arm/mach-ep93xx/dma.c
> @@ -19,6 +19,7 @@
>  #include <linux/init.h>
>  #include <linux/interrupt.h>
>  #include <linux/kernel.h>
> +#include <linux/of.h>
>  #include <linux/platform_device.h>
>  
>  #include <linux/platform_data/dma-ep93xx.h>

Stray change.

...

> --- a/drivers/dma/ep93xx_dma.c
> +++ b/drivers/dma/ep93xx_dma.c
> @@ -20,6 +20,7 @@
>  #include <linux/dmaengine.h>
>  #include <linux/module.h>
>  #include <linux/mod_devicetable.h>
> +#include <linux/of_device.h>
>  #include <linux/platform_device.h>
>  #include <linux/slab.h>

...

> -	struct ep93xx_dma_chan	channels[];
> +	struct ep93xx_dma_chan	*channels;

Why? This is helpful to allocate main structure and channels in one go.

...

> @@ -875,9 +881,11 @@ static int ep93xx_dma_alloc_chan_resources(struct dma_chan *chan)
>  	if (!edmac->edma->m2m) {
>  		if (!data)
>  			return -EINVAL;
> +
>  		if (data->port < EP93XX_DMA_I2S1 ||
>  		    data->port > EP93XX_DMA_IRDA)
>  			return -EINVAL;
> +
>  		if (data->direction != ep93xx_dma_chan_direction(chan))
>  			return -EINVAL;
>  	} else {

Seems unrelated change.

>  	ep93xx_dma_advance_work(to_ep93xx_dma_chan(chan));
>  }

...

> +static const struct of_device_id ep93xx_dma_of_ids[] = {
> +	{ .compatible = "cirrus,ep9301-dma-m2p", .data = &edma_m2p },
> +	{ .compatible = "cirrus,ep9301-dma-m2m", .data = &edma_m2m },
> +	{ /* sentinel */ }
> +};
> +MODULE_DEVICE_TABLE(of, ep93xx_dma_of_ids);

Move this closer to the real user (see below).
Also this ID table shouldn't be under ifdeffery.

...

> +	struct device_node *np = pdev->dev.of_node;
> +	const struct of_device_id *match = of_match_node(ep93xx_dma_of_ids, pdev->dev.of_node);
> +	const struct ep93xx_edma_data *data = match->data;

No NULL check? Why do you have this duplication, btw?

> +	struct dma_device *dma_dev = &edma->dma_dev;
> +	int num_channels;
> +	int i;
> +
> +	match = of_match_device((ep93xx_dma_of_ids), &pdev->dev);
> +	if (!match || !match->data) {
> +		dev_err(&pdev->dev, "No device match found\n");
> +		return -ENODEV;
> +	}

Use of_device_get_match_data().

...

> +	edma->channels = devm_kzalloc(&pdev->dev,
> +				      num_channels * sizeof(struct ep93xx_dma_chan),
> +				      GFP_KERNEL);
> +	if (!edma->channels)
> +		return -ENOMEM;

Simply no. See below what to do.

...

> +	for (i = 0; i < num_channels; i++) {
> +		struct ep93xx_dma_chan *edmac = &edma->channels[i];
> +
> +		edmac->chan.device = dma_dev;
> +		edmac->regs = devm_platform_ioremap_resource(pdev, i);

No error check?

> +		edmac->irq = platform_get_irq(pdev, i);

No error check?

> +		edmac->edma = edma;
> +
> +		edmac->clk = of_clk_get(np, i);

Can this actually use clk_get() or its devm_*() variant?

> +

Redundant blank line.

> +		if (IS_ERR(edmac->clk)) {
> +			dev_warn(&pdev->dev, "failed to get clock\n");
> +			continue;
> +		}
> +
> +		spin_lock_init(&edmac->lock);
> +		INIT_LIST_HEAD(&edmac->active);
> +		INIT_LIST_HEAD(&edmac->queue);
> +		INIT_LIST_HEAD(&edmac->free_list);
> +		tasklet_setup(&edmac->tasklet, ep93xx_dma_tasklet);
> +
> +		list_add_tail(&edmac->chan.device_node,
> +			      &dma_dev->channels);
> +	}

...

> -	edma_size = pdata->num_channels * sizeof(struct ep93xx_dma_chan);
> -	edma = kzalloc(sizeof(*edma) + edma_size, GFP_KERNEL);
> -	if (!edma)
> +	edma->channels = devm_kzalloc(&pdev->dev,
> +				      pdata->num_channels * sizeof(struct ep93xx_dma_chan),
> +				      GFP_KERNEL);
> +	if (!edma->channels)
>  		return -ENOMEM;

No. Just include overflow.h and use struct_size().

...

> +	edma = devm_kzalloc(&pdev->dev, sizeof(*edma), GFP_KERNEL);
> +

No error check?!

...

> +	if (platform_get_device_id(pdev))
> +		ret = ep93xx_init_from_pdata(pdev, edma);
> +	else
> +		ret = ep93xx_dma_of_probe(pdev, edma);

> +

Redundant blank line.

> +	if (ret)
> +		return ret;

-- 
With Best Regards,
Andy Shevchenko