Received-SPF: pass (google.com: domain of linux-kernel+bounces-173403-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Message-ID: <9f0a8fdb-dd34-4a53-948d-d4ed0410de6f@gmail.com>
Date: Wed, 8 May 2024 17:32:34 +0300
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH 7/7] drivers: iio: imu: Add support for adis1657x family
To: Jonathan Cameron <jic23@kernel.org>
Cc: linux-kernel@vger.kernel.org, linux-iio@vger.kernel.org,
 devicetree@vger.kernel.org, conor+dt@kernel.org,
 krzysztof.kozlowski+dt@linaro.org, robh@kernel.org, nuno.sa@analog.com
References: <20240426135339.185602-1-ramona.bolboaca13@gmail.com>
 <20240426135339.185602-8-ramona.bolboaca13@gmail.com>
 <20240428154523.17b27fa8@jic23-huawei>
Content-Language: en-US
From: Ramona Gradinariu <ramona.bolboaca13@gmail.com>
In-Reply-To: <20240428154523.17b27fa8@jic23-huawei>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit

Hello Jonathan,

Some explanations from my side.

>> @@ -437,6 +467,130 @@ static int adis16475_set_filter(struct adis16475 *st, const u32 filter)
>>  	return 0;
>>  }
>>  
>> +static ssize_t adis16475_get_fifo_enabled(struct device *dev,
>> +					  struct device_attribute *attr,
>> +					  char *buf)
>> +{
>> +	struct iio_dev *indio_dev = dev_to_iio_dev(dev);
>> +	struct adis16475 *st = iio_priv(indio_dev);
>> +	int ret;
>> +	u16 val;
>> +
>> +	ret = adis_read_reg_16(&st->adis, ADIS16475_REG_FIFO_CTRL, &val);
>> +	if (ret)
>> +		return ret;
>> +	val = FIELD_GET(ADIS16475_FIFO_EN_MASK, val);
>> +
>> +	return sysfs_emit(buf, "%d\n", val);
> As below, might as well put the FIELD_GET() in the sysfs_emit rather than
> writing the local parameter.

In all instances where I did, I did it to avoid casting. v2 inlines the values and
the cast is needed to avoid compilation errors.

> 	
>> +	if (adis->data->burst_max_len)
>> +		burst_max_length = adis->data->burst_max_len;
>> +	else
>> +		burst_max_length = burst_length;
>> +
>> +	tx = adis->buffer + burst_max_length;
>> +	tx[0] = ADIS_READ_REG(burst_req);
>> +
>> +	if (burst_req)
> If !burst_req does the rest of this do anything at all?
> If so flip the logic as
> 	if (!burst_req)
> 		return adis16475_push_single_sample(pf);
>
> 	the rest...
> 	return spi_sync(adis->spi, &adis->msg);

The update is needed even if burst_req is false. The adis message has to be updated
based on the burst request value, which is then used either in 
adis16475_push_single_sample or in spi_sync call.

> 		
>
>> +		return spi_sync(adis->spi, &adis->msg);
>> +
>> +	return adis16475_push_single_sample(pf);
>> +}
>> +
>> +/*
>> + * This handler is meant to be used for devices which support burst readings
>> + * from FIFO (namely devices from adis1657x family).
>> + * In order to pop the FIFO the 0x68 0x00 FIFO pop burst request has to be sent.
>> + * If the previous device command was not a FIFO pop burst request, the FIFO pop
>> + * burst request will simply pop the FIFO without returning valid data.
>> + * For the nth consecutive burst request, the
>> + * device will send the data popped with the (n-1)th consecutive burst request.
>> + * In order to read the data which was popped previously, without popping the FIFO,
>> + * the 0x00 0x00 burst request has to be sent.
>> + * If after a 0x68 0x00 FIFO pop burst request, there is any other device access
>> + * different from a 0x68 0x00 or a 0x00 0x00 burst request, the FIFO data popped
>> + * previously will be lost.
>> + */
>> +static irqreturn_t adis16475_trigger_handler_with_fifo(int irq, void *p)
>>  {
>>  	struct iio_poll_func *pf = p;
>>  	struct iio_dev *indio_dev = pf->indio_dev;
>> +	struct adis16475 *st = iio_priv(indio_dev);
>> +	struct adis *adis = &st->adis;
>> +	int ret;
>> +	u16 fifo_cnt, i;
>>  
>> -	adis16475_push_single_sample(pf);
>> +	adis_dev_lock(&st->adis);
>> +
>> +	ret = __adis_read_reg_16(adis, ADIS16475_REG_FIFO_CNT, &fifo_cnt);
>> +	if (ret || fifo_cnt < 2)
>> +		goto unlock;
> I would break these conditions and add a comment on why fifo_cnt < 2 is
> a reason to just return 0;

Updated this in v2, and actually fifo_cnt can also be 1 so the code simply
verifies if !fifo_cnt.

>
>> +
>> +	if (fifo_cnt > st->fifo_watermark)
>> +		fifo_cnt = st->fifo_watermark;
> fifo_cnt = min(fifo_cnt, st->fifo_watermark);
>
> This confuses me though as normally overreading after a fifo watermark is
> both safe and the right thing to do (as reduces chance of overflow etc).
> If we need to clamp to the watermark for some reason, add a comment.

Removed this in v2.

>>  
>>  			ret = __adis_write_reg_16(&st->adis,
>>  						  ADIS16475_REG_UP_SCALE,
>> @@ -1467,7 +1888,23 @@ static int adis16475_config_irq_pin(struct adis16475 *st)
>>  	 */
>>  	usleep_range(250, 260);
>>  
>> -	return 0;
>> +	/*
>> +	 * If the device has FIFO support, configure the watermark polarity
>> +	 * pin as well.
> The pin is for polarity or the polarity is for the watermark signalling on that
> pin?  I'm not seeing a datasheet yet for these parts so I couldn't check.

The device has a watermark pin, which can be used as a trigger source.
In this case we set the polarity for the watermark pin.

>
>> +	 */
>> +	if (st->info->flags & ADIS16475_HAS_FIFO) {
>> +		val = ADIS16475_WM_POL(polarity);
>> +		ret = adis_update_bits(&st->adis, ADIS16475_REG_FIFO_CTRL,
>> +				       ADIS16475_WM_POL_MASK, val);
>> +		if (ret)
>> +			return ret;
>> +
>> +		/* Enable watermark interrupt pin. */
>> +		val = ADIS16475_WM_EN(1);
>> +		ret = adis_update_bits(&st->adis, ADIS16475_REG_FIFO_CTRL, ADIS16475_WM_EN_MASK, val);

Best Regards,
Ramona