Re: [PATCH v3 2/3] i2c: stm32f7: unmap DMA mapped buffer

[Clement LE GOFFIC <clement.legoffic@foss.st.com>]

Hi Andi,

On 7/2/25 19:08, Andi Shyti wrote:
> Hi Clement,
> 
> On Mon, Jun 30, 2025 at 02:55:14PM +0200, Clément Le Goffic wrote:
>> Fix an issue where the mapped DMA buffer was not unmapped.
> 
> "Fix an issue..." is too generic. Can you be more specific? Where
> was it mapped? Where was it left unmapped?
> 
> Please, do consider that the user needs to understand what
> happens in the patch without needing to look into the patch.

Ok sure I'll refine the commit message.

> 
>> Fixes: 7ecc8cfde553 ("i2c: i2c-stm32f7: Add DMA support")
>> Acked-by: Alain Volmat <alain.volmat@foss.st.com>
>> Signed-off-by: Clément Le Goffic <clement.legoffic@foss.st.com>
>> ---
>>   drivers/i2c/busses/i2c-stm32f7.c | 6 ++++++
>>   1 file changed, 6 insertions(+)
>>
>> diff --git a/drivers/i2c/busses/i2c-stm32f7.c b/drivers/i2c/busses/i2c-stm32f7.c
>> index e4aaeb2262d0..042386b4cabe 100644
>> --- a/drivers/i2c/busses/i2c-stm32f7.c
>> +++ b/drivers/i2c/busses/i2c-stm32f7.c
>> @@ -1554,6 +1554,8 @@ static irqreturn_t stm32f7_i2c_handle_isr_errs(struct stm32f7_i2c_dev *i2c_dev,
>>   	if (i2c_dev->use_dma) {
>>   		stm32f7_i2c_disable_dma_req(i2c_dev);
>>   		dmaengine_terminate_async(dma->chan_using);
>> +		dma_unmap_single(i2c_dev->dev, dma->dma_buf, dma->dma_len,
>> +				 dma->dma_data_dir);
>>   	}
>>   
>>   	i2c_dev->master_mode = false;
>> @@ -1622,6 +1624,8 @@ static irqreturn_t stm32f7_i2c_isr_event_thread(int irq, void *data)
>>   		if (i2c_dev->use_dma) {
>>   			stm32f7_i2c_disable_dma_req(i2c_dev);
>>   			dmaengine_terminate_async(dma->chan_using);
>> +			dma_unmap_single(i2c_dev->dev, dma->dma_buf, dma->dma_len,
>> +					 dma->dma_data_dir);
>>   		}
>>   		f7_msg->result = -ENXIO;
>>   	}
>> @@ -1642,6 +1646,8 @@ static irqreturn_t stm32f7_i2c_isr_event_thread(int irq, void *data)
>>   				dev_dbg(i2c_dev->dev, "<%s>: Timed out\n", __func__);
>>   				stm32f7_i2c_disable_dma_req(i2c_dev);
>>   				dmaengine_terminate_async(dma->chan_using);
>> +				dma_unmap_single(i2c_dev->dev, dma->dma_buf, dma->dma_len,
>> +						 dma->dma_data_dir);
> 
> Can't we use the dma_callback here, or similar? I see some
> similar patterns and I think the code can be improved.

Yes, it seems the code can be factorized.
I'll submit a new version with the factorization.

Best regards,
Clément