2019-12-23 15:46:39

by Jonathan Cameron

Subject: Re: [PATCH] iio: adc: stm32-dfsdm: Use dma_request_chan() instead dma_request_slave_channel()

On Tue, 17 Dec 2019 09:52:21 +0200
Peter Ujfalusi <[email protected]> wrote:

> dma_request_slave_channel() is a wrapper on top of dma_request_chan()
> eating up the error code.
>
> By using dma_request_chan() directly the driver can support deferred
> probing against DMA.
>
> Signed-off-by: Peter Ujfalusi <[email protected]>
I don't think this one goes far enough.

One of the paths does:

/* Optionally request DMA */
if (stm32_dfsdm_dma_request(indio_dev)) {
dev_dbg(&indio_dev->dev, "No DMA support\n");
return 0;
}

That needs specific handling for deferred probing.

> ---
> drivers/iio/adc/stm32-dfsdm-adc.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/iio/adc/stm32-dfsdm-adc.c b/drivers/iio/adc/stm32-dfsdm-adc.c
> index e493242c266e..3aac1a21f9d0 100644
> --- a/drivers/iio/adc/stm32-dfsdm-adc.c
> +++ b/drivers/iio/adc/stm32-dfsdm-adc.c
> @@ -1383,9 +1383,9 @@ static int stm32_dfsdm_dma_request(struct iio_dev *indio_dev)
> {
> struct stm32_dfsdm_adc *adc = iio_priv(indio_dev);
>
> - adc->dma_chan = dma_request_slave_channel(&indio_dev->dev, "rx");
> - if (!adc->dma_chan)
> - return -EINVAL;
> + adc->dma_chan = dma_request_chan(&indio_dev->dev, "rx");
> + if (IS_ERR(adc->dma_chan))
> + return PTR_ERR(adc->dma_chan);
>
> adc->rx_buf = dma_alloc_coherent(adc->dma_chan->device->dev,
> DFSDM_DMA_BUFFER_SIZE,

2019-12-23 22:00:48

by Peter Ujfalusi

Subject: Re: [PATCH] iio: adc: stm32-dfsdm: Use dma_request_chan() instead dma_request_slave_channel()

Hi Jonathan,

On 12/23/19 5:45 PM, Jonathan Cameron wrote:
> On Tue, 17 Dec 2019 09:52:21 +0200
> Peter Ujfalusi <[email protected]> wrote:
>
>> dma_request_slave_channel() is a wrapper on top of dma_request_chan()
>> eating up the error code.
>>
>> By using dma_request_chan() directly the driver can support deferred
>> probing against DMA.
>>
>> Signed-off-by: Peter Ujfalusi <[email protected]>
> I don't think this one goes far enough.
>
> One of the paths does:
>
> /* Optionally request DMA */
> if (stm32_dfsdm_dma_request(indio_dev)) {
> dev_dbg(&indio_dev->dev, "No DMA support\n");
> return 0;
> }
>
> That needs specific handling for deferred probing.

Right, I have checked the stm32_dfsdm_audio_init() path, via that the
error is propagated correctly.

I'll send an update next year ;)

>
>> ---
>> drivers/iio/adc/stm32-dfsdm-adc.c | 6 +++---
>> 1 file changed, 3 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/iio/adc/stm32-dfsdm-adc.c b/drivers/iio/adc/stm32-dfsdm-adc.c
>> index e493242c266e..3aac1a21f9d0 100644
>> --- a/drivers/iio/adc/stm32-dfsdm-adc.c
>> +++ b/drivers/iio/adc/stm32-dfsdm-adc.c
>> @@ -1383,9 +1383,9 @@ static int stm32_dfsdm_dma_request(struct iio_dev *indio_dev)
>> {
>> struct stm32_dfsdm_adc *adc = iio_priv(indio_dev);
>>
>> - adc->dma_chan = dma_request_slave_channel(&indio_dev->dev, "rx");
>> - if (!adc->dma_chan)
>> - return -EINVAL;
>> + adc->dma_chan = dma_request_chan(&indio_dev->dev, "rx");
>> + if (IS_ERR(adc->dma_chan))
>> + return PTR_ERR(adc->dma_chan);
>>
>> adc->rx_buf = dma_alloc_coherent(adc->dma_chan->device->dev,
>> DFSDM_DMA_BUFFER_SIZE,
>

- Peter

Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki.
Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki