Re: [PATCH v3 2/2] iio: adc: ti-ads1298: Add driver

[Mike Looijmans <mike.looijmans@topic.nl>]

On 06-02-2024 16:09, Andy Shevchenko wrote:
> On Tue, Feb 06, 2024 at 03:47:45PM +0100, Mike Looijmans wrote:
>> On 06-02-2024 15:25, Mike Looijmans wrote:
>>> On 06-02-2024 14:50, Andy Shevchenko wrote:
>>>> On Tue, Feb 06, 2024 at 02:33:47PM +0100, Mike Looijmans wrote:
>>>>> On 06-02-2024 13:57, Andy Shevchenko wrote:
>>>>>> On Tue, Feb 06, 2024 at 07:58:18AM +0100, Mike Looijmans wrote:
> ...
>
>>>>>>> +    wasbusy = --priv->rdata_xfer_busy;
>>>>>> Why preincrement? How would it be different from postincrement?
>>>>> Maybe better write this as:
>>>>>
>>>>> --priv->rdata_xfer_busy;
>>>>>
>>>>> wasbusy = priv->rdata_xfer_busy;
>>>>>
>>>>> I want the value after decrementing it.
>>>> Yes, looks more obvious.
>>> Having done that, and looking at it again, it's better to just eliminate
>>> the local "wasbusy" altogether. More concise.
>>
>> This removes the decrement operator, so the method now looks like this:
>>
>>
>> static void ads1298_rdata_release_busy_or_restart(struct ads1298_private
>> *priv)
>> {
>>      guard(spinlock_irqsave)(&priv->irq_busy_lock);
>>
>>      if (priv->rdata_xfer_busy > 1) {
>>          /*
>>           * DRDY interrupt occurred before SPI completion. Start a new
>>           * SPI transaction now to retrieve the data that wasn't latched
>>           * into the ADS1298 chip's transfer buffer yet.
>>           */
>>          spi_async(priv->spi, &priv->rdata_msg);
>>          /*
>>           * If more than one DRDY took place, there was an overrun. Since
>>           * the sample is already lost, reset the counter to 1 so that
>>           * we will wait for a DRDY interrupt after this SPI transaction.
>>           */
>>          priv->rdata_xfer_busy = 1;
>>      } else {
>>          /* No pending data, wait for DRDY */
>>          priv->rdata_xfer_busy = 0;
>>      }
>> }
> Yep and it looks like you reinvented atomics :-)
>
> 	atomic_t rdata_xfer_busy;
> 	...
>
> But it's up to you what to do with that.
> Maybe Jonathan can advice something different.
>
The spinlock also protects the call to spi_async().


-- 
Mike Looijmans
System Expert

TOPIC Embedded Products B.V.
Materiaalweg 4, 5681 RJ Best
The Netherlands

T: +31 (0) 499 33 69 69
E: mike.looijmans@topic.nl
W: www.topic.nl