The AD7192 datasheet (Rev. A, page 34, RESET section) says:
"When a reset is initiated, the user must allow a period of
500 us before accessing any of the on-chip registers."

Use fsleep(500) instead of usleep_range(500, 1000). The 500 us
minimum stays the same; fsleep() picks the upper slack itself
(about +25% on a default config).

Add a code comment with the datasheet reference so the "why"
of the wait is visible at the call site.

Signed-off-by: Stepan Ionichev <sozdayvek@gmail.com>
---
v4:
- Move the datasheet citation into a code comment (per Andy)
- Add a blank line before the fsleep call (per Andy)
- Drop the "No functional change" claim — fsleep's upper bound is
  about +25%, narrower than the original usleep_range(500, 1000)

v3:
- Cite AD7192 datasheet for the 500 us minimum
- Add changelog (was missing in v2)

v2:
- Drop the redundant "Wait for at least 500us" comment

 drivers/iio/adc/ad7192.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/iio/adc/ad7192.c b/drivers/iio/adc/ad7192.c
index 8b1664f6b..ba27614c8 100644
--- a/drivers/iio/adc/ad7192.c
+++ b/drivers/iio/adc/ad7192.c
@@ -576,7 +576,12 @@ static int ad7192_setup(struct iio_dev *indio_dev, struct device *dev)
 	ret = ad_sd_reset(&st->sd);
 	if (ret < 0)
 		return ret;
-	usleep_range(500, 1000); /* Wait for at least 500us */
+
+	/*
+	 * Per AD7192 datasheet (Rev. A, page 34, RESET section), allow
+	 * 500 us after a reset before accessing on-chip registers.
+	 */
+	fsleep(500);
 
 	/* write/read test for device presence */
 	ret = ad_sd_read_reg(&st->sd, AD7192_REG_ID, 1, &id);
-- 
2.43.0

On Mon, May 11, 2026 at 07:55:44AM +0500, Stepan Ionichev wrote:
> The AD7192 datasheet (Rev. A, page 34, RESET section) says:
> "When a reset is initiated, the user must allow a period of
> 500 us before accessing any of the on-chip registers."

> Use fsleep(500) instead of usleep_range(500, 1000). The 500 us
> minimum stays the same; fsleep() picks the upper slack itself
> (about +25% on a default config).

The second sentence here...

> Add a code comment with the datasheet reference so the "why"
> of the wait is visible at the call site.

...and this paragraph are unneeded details.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>

-- 
With Best Regards,
Andy Shevchenko

On Mon, 11 May 2026 16:28:39 +0300
Andy Shevchenko <andriy.shevchenko@intel.com> wrote:

> On Mon, May 11, 2026 at 07:55:44AM +0500, Stepan Ionichev wrote:
> > The AD7192 datasheet (Rev. A, page 34, RESET section) says:
> > "When a reset is initiated, the user must allow a period of
> > 500 us before accessing any of the on-chip registers."  
> 
> > Use fsleep(500) instead of usleep_range(500, 1000). The 500 us
> > minimum stays the same; fsleep() picks the upper slack itself
> > (about +25% on a default config).  
> 
> The second sentence here...
> 
> > Add a code comment with the datasheet reference so the "why"
> > of the wait is visible at the call site.  
> 
> ...and this paragraph are unneeded details.
> 
> Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com>
> 

Tweaked and applied.