[PATCH v2 0/3] staging: iio: replace sprintf() with sysfs_emit()

Gabriel Rondon posted 3 patches 2 weeks ago
drivers/staging/iio/adc/ad7816.c              | 16 ++++++-------
drivers/staging/iio/frequency/ad9834.c        | 20 +++++-----------
.../staging/iio/impedance-analyzer/ad5933.c   | 24 +++++++++----------
3 files changed, 26 insertions(+), 34 deletions(-)
[PATCH v2 0/3] staging: iio: replace sprintf() with sysfs_emit()
Posted by Gabriel Rondon 2 weeks ago
Replace the remaining sprintf() calls with sysfs_emit() in sysfs
show functions across three staging IIO drivers: ad7816, ad5933, and
ad9834.

sysfs_emit() is the preferred API for sysfs callbacks as it is aware
of the PAGE_SIZE buffer limit. Build tested with gcc 13.3.

Changes since v1:
- ad5933: remove unnecessary (int) cast in ad5933_show_frequency()
  and use %llu format specifier for unsigned long long (Andy Shevchenko)
- ad9834: simplify wavetype_available show functions by removing
  intermediate string variable and returning directly (Andy Shevchenko)

Gabriel Rondon (3):
  staging: iio: ad7816: use sysfs_emit() in show functions
  staging: iio: ad5933: use sysfs_emit() in show functions
  staging: iio: ad9834: use sysfs_emit() and simplify show functions

 drivers/staging/iio/adc/ad7816.c              | 16 ++++++-------
 drivers/staging/iio/frequency/ad9834.c        | 20 +++++-----------
 .../staging/iio/impedance-analyzer/ad5933.c   | 24 +++++++++----------
 3 files changed, 26 insertions(+), 34 deletions(-)

-- 
2.33.0
Re: [PATCH v2 0/3] staging: iio: replace sprintf() with sysfs_emit()
Posted by Jonathan Cameron 1 week, 6 days ago
On Fri, 20 Mar 2026 22:24:21 +0000
Gabriel Rondon <grondon@gmail.com> wrote:

> Replace the remaining sprintf() calls with sysfs_emit() in sysfs
> show functions across three staging IIO drivers: ad7816, ad5933, and
> ad9834.
> 
> sysfs_emit() is the preferred API for sysfs callbacks as it is aware
> of the PAGE_SIZE buffer limit. Build tested with gcc 13.3.
> 
Firstly a process thing. Don't send a v2 in reply to a v1. It should
be a new email thread.  I'm not sure where this style is coming from
as I've never encountered any kernel subsystem who ask for it.

The basic reason is threads get very confusing if they mix versions +
a more practical one is that your email appears a bunch of screens up
in the email client and many reviewers (who don't have time to look at
everything) start with most recent emails when looking for things to
review.  Thus you get ignored.

Jonathan

> Changes since v1:
> - ad5933: remove unnecessary (int) cast in ad5933_show_frequency()
>   and use %llu format specifier for unsigned long long (Andy Shevchenko)
> - ad9834: simplify wavetype_available show functions by removing
>   intermediate string variable and returning directly (Andy Shevchenko)
> 
> Gabriel Rondon (3):
>   staging: iio: ad7816: use sysfs_emit() in show functions
>   staging: iio: ad5933: use sysfs_emit() in show functions
>   staging: iio: ad9834: use sysfs_emit() and simplify show functions
> 
>  drivers/staging/iio/adc/ad7816.c              | 16 ++++++-------
>  drivers/staging/iio/frequency/ad9834.c        | 20 +++++-----------
>  .../staging/iio/impedance-analyzer/ad5933.c   | 24 +++++++++----------
>  3 files changed, 26 insertions(+), 34 deletions(-)
>
Re: [PATCH v2 0/3] staging: iio: replace sprintf() with sysfs_emit()
Posted by Dan Carpenter 1 week, 6 days ago
On Sat, Mar 21, 2026 at 01:03:07PM +0000, Jonathan Cameron wrote:
> On Fri, 20 Mar 2026 22:24:21 +0000
> Gabriel Rondon <grondon@gmail.com> wrote:
> 
> > Replace the remaining sprintf() calls with sysfs_emit() in sysfs
> > show functions across three staging IIO drivers: ad7816, ad5933, and
> > ad9834.
> > 
> > sysfs_emit() is the preferred API for sysfs callbacks as it is aware
> > of the PAGE_SIZE buffer limit. Build tested with gcc 13.3.
> > 
> Firstly a process thing. Don't send a v2 in reply to a v1. It should
> be a new email thread.  I'm not sure where this style is coming from
> as I've never encountered any kernel subsystem who ask for it.
> 

People used to do that in the early days of git, but now no one does.
It breaks patchwork as well.

regards,
dan carpenter