2020-01-07 11:43:00

by Peter Ujfalusi

Subject: [PATCH v2] iio: adc: stm32-adc: Use dma_request_chan() instead dma_request_slave_channel()

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]>
---
Hi,

Changes since v1:
- Fall back to IRQ mode only in case of ENODEV

Regards,
Peter

drivers/iio/adc/stm32-adc.c | 16 ++++++++++++++--
1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/drivers/iio/adc/stm32-adc.c b/drivers/iio/adc/stm32-adc.c
index 3b291d72701c..df5f5d61f9f9 100644
--- a/drivers/iio/adc/stm32-adc.c
+++ b/drivers/iio/adc/stm32-adc.c
@@ -1746,9 +1746,21 @@ static int stm32_adc_dma_request(struct iio_dev *indio_dev)
struct dma_slave_config config;
int ret;

- adc->dma_chan = dma_request_slave_channel(&indio_dev->dev, "rx");
- if (!adc->dma_chan)
+ adc->dma_chan = dma_request_chan(&indio_dev->dev, "rx");
+ if (IS_ERR(adc->dma_chan)) {
+ ret = PTR_ERR(adc->dma_chan);
+ if (ret != -ENODEV) {
+ if (ret != -EPROBE_DEFER)
+ dev_err(&indio_dev->dev,
+ "DMA channel request failed with %d\n",
+ ret);
+ return ret;
+ }
+
+ /* Ignore errors to fall back to IRQ mode */
+ adc->dma_chan = NULL;
return 0;
+ }

adc->rx_buf = dma_alloc_coherent(adc->dma_chan->device->dev,
STM32_DMA_BUFFER_SIZE,
--
Peter

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

2020-01-07 17:17:23

by Fabrice Gasnier

Subject: Re: [PATCH v2] iio: adc: stm32-adc: Use dma_request_chan() instead dma_request_slave_channel()

On 1/7/20 12:41 PM, Peter Ujfalusi 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]>
> ---
> Hi,
>
> Changes since v1:
> - Fall back to IRQ mode only in case of ENODEV
>
> Regards,
> Peter

Hi Peter,

Thanks for the patch,

In case you send another version... I've just a minor suggestion
regarding the comment (see after). Apart from that, you can add my:

Acked-by: Fabrice Gasnier <[email protected]>

Best Regards,
Fabrice

>
> drivers/iio/adc/stm32-adc.c | 16 ++++++++++++++--
> 1 file changed, 14 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/iio/adc/stm32-adc.c b/drivers/iio/adc/stm32-adc.c
> index 3b291d72701c..df5f5d61f9f9 100644
> --- a/drivers/iio/adc/stm32-adc.c
> +++ b/drivers/iio/adc/stm32-adc.c
> @@ -1746,9 +1746,21 @@ static int stm32_adc_dma_request(struct iio_dev *indio_dev)
> struct dma_slave_config config;
> int ret;
>
> - adc->dma_chan = dma_request_slave_channel(&indio_dev->dev, "rx");
> - if (!adc->dma_chan)
> + adc->dma_chan = dma_request_chan(&indio_dev->dev, "rx");
> + if (IS_ERR(adc->dma_chan)) {
> + ret = PTR_ERR(adc->dma_chan);
> + if (ret != -ENODEV) {
> + if (ret != -EPROBE_DEFER)
> + dev_err(&indio_dev->dev,
> + "DMA channel request failed with %d\n",
> + ret);
> + return ret;
> + }
> +
> + /* Ignore errors to fall back to IRQ mode */
^
error
alternate suggestion:
/* DMA is optional: fall back to IRQ mode */

> + adc->dma_chan = NULL;
> return 0;
> + }
>
> adc->rx_buf = dma_alloc_coherent(adc->dma_chan->device->dev,
> STM32_DMA_BUFFER_SIZE,
>

2020-01-08 08:14:53

by Peter Ujfalusi

Subject: Re: [PATCH v2] iio: adc: stm32-adc: Use dma_request_chan() instead dma_request_slave_channel()

Hi Fabrice,

On 07/01/2020 19.15, Fabrice Gasnier wrote:
> On 1/7/20 12:41 PM, Peter Ujfalusi 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]>
>> ---
>> Hi,
>>
>> Changes since v1:
>> - Fall back to IRQ mode only in case of ENODEV
>>
>> Regards,
>> Peter
>
> Hi Peter,
>
> Thanks for the patch,
>
> In case you send another version... I've just a minor suggestion
> regarding the comment (see after). Apart from that, you can add my:

Thanks, I'll take your suggested update and send v3.

> Acked-by: Fabrice Gasnier <[email protected]>
>
> Best Regards,
> Fabrice
>
>>
>> drivers/iio/adc/stm32-adc.c | 16 ++++++++++++++--
>> 1 file changed, 14 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/iio/adc/stm32-adc.c b/drivers/iio/adc/stm32-adc.c
>> index 3b291d72701c..df5f5d61f9f9 100644
>> --- a/drivers/iio/adc/stm32-adc.c
>> +++ b/drivers/iio/adc/stm32-adc.c
>> @@ -1746,9 +1746,21 @@ static int stm32_adc_dma_request(struct iio_dev *indio_dev)
>> struct dma_slave_config config;
>> int ret;
>>
>> - adc->dma_chan = dma_request_slave_channel(&indio_dev->dev, "rx");
>> - if (!adc->dma_chan)
>> + adc->dma_chan = dma_request_chan(&indio_dev->dev, "rx");
>> + if (IS_ERR(adc->dma_chan)) {
>> + ret = PTR_ERR(adc->dma_chan);
>> + if (ret != -ENODEV) {
>> + if (ret != -EPROBE_DEFER)
>> + dev_err(&indio_dev->dev,
>> + "DMA channel request failed with %d\n",
>> + ret);
>> + return ret;
>> + }
>> +
>> + /* Ignore errors to fall back to IRQ mode */
> ^
> error
> alternate suggestion:
> /* DMA is optional: fall back to IRQ mode */

Makes more sense in the context.

>
>> + adc->dma_chan = NULL;
>> return 0;
>> + }
>>
>> adc->rx_buf = dma_alloc_coherent(adc->dma_chan->device->dev,
>> STM32_DMA_BUFFER_SIZE,
>>

- Péter

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