Re: [PATCH v2 dmaengine 1/1] dmaengine: imx-sdma: revert: add clock ratio 1:1 check

[Angus Ainslie <angus@xxxxxxxx>]

Hi Lucas,

On 2019-03-21 20:36, Andy Duan wrote:
> This reverts commit 25aaa75df1e6 (dmaengine: imx-sdma: add clock ratio
> 1:1 check) due to break SDMA function on i.MX6SX platform.
>
> Log:
> imx-sdma 20ec000.sdma: Timeout waiting for CH0 ready
>
> Fixes: 25aaa75df1e6 (dmaengine: imx-sdma: add clock ratio 1:1 check)
> Signed-off-by: Fugang Duan <fugang.duan@xxxxxxx>
> ---
>  drivers/dma/imx-sdma.c | 18 ++++--------------
>  1 file changed, 4 insertions(+), 14 deletions(-)
>
> diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c
> index 5f3c137..7fae4bf 100644
> --- a/drivers/dma/imx-sdma.c
> +++ b/drivers/dma/imx-sdma.c
> @@ -441,8 +441,6 @@ struct sdma_engine {
>  	unsigned int			irq;
>  	dma_addr_t			bd0_phys;
>  	struct sdma_buffer_descriptor	*bd0;
> -	/* clock ratio for AHB:SDMA core. 1:1 is 1, 2:1 is 0*/
> -	bool				clk_ratio;
>  };
>
>  static int sdma_config_write(struct dma_chan *chan,
> @@ -665,11 +663,8 @@ static int sdma_run_channel0(struct sdma_engine 
> *sdma)
>  		dev_err(sdma->dev, "Timeout waiting for CH0 ready\n");
>
>  	/* Set bits of CONFIG register with dynamic context switching */
> -	reg = readl(sdma->regs + SDMA_H_CONFIG);
> -	if ((reg & SDMA_H_CONFIG_CSM) == 0) {
> -		reg |= SDMA_H_CONFIG_CSM;
> -		writel_relaxed(reg, sdma->regs + SDMA_H_CONFIG);
> -	}
> +	if (readl(sdma->regs + SDMA_H_CONFIG) == 0)
> +		writel_relaxed(SDMA_H_CONFIG_CSM, sdma->regs + SDMA_H_CONFIG);
>
>  	return ret;
>  }
> @@ -1852,9 +1847,6 @@ static int sdma_init(struct sdma_engine *sdma)
>  	if (ret)
>  		goto disable_clk_ipg;
>
> -	if (clk_get_rate(sdma->clk_ahb) == clk_get_rate(sdma->clk_ipg))
> -		sdma->clk_ratio = 1;
> -
>  	/* Be sure SDMA has not started yet */
>  	writel_relaxed(0, sdma->regs + SDMA_H_C0PTR);
>
> @@ -1895,10 +1887,8 @@ static int sdma_init(struct sdma_engine *sdma)
>  	writel_relaxed(0x4050, sdma->regs + SDMA_CHN0ADDR);
>
>  	/* Set bits of CONFIG register but with static context switching */
> -	if (sdma->clk_ratio)
> -		writel_relaxed(SDMA_H_CONFIG_ACR, sdma->regs + SDMA_H_CONFIG);
> -	else
> -		writel_relaxed(0, sdma->regs + SDMA_H_CONFIG);
> +	/* FIXME: Check whether to set ACR bit depending on clock ratios */
> +	writel_relaxed(0, sdma->regs + SDMA_H_CONFIG);

It looks like the automated check breaks some imx chips. Are you ok if I 
resubmit with the original devicetree switch ?

I guess another way would be to have a different compatible string and 
have some kind of "quirks".

Thanks
Angus

>  	writel_relaxed(ccb_phys, sdma->regs + SDMA_H_C0PTR);