On Fri, 19 Jun 2026 10:16:31 +0100 Nuno Sá <[email protected]> wrote:
> On Thu, Jun 18, 2026 at 05:14:19PM +0100, Rodrigo Alencar wrote: > > On 18/06/26 16:06, Nuno Sá wrote: > > > On Thu, Jun 18, 2026 at 02:27:22PM +0100, Rodrigo Alencar via B4 Relay > > > wrote: > > > > From: Rodrigo Alencar <[email protected]> > > > > > > > > Move logic to create a channel prefix for naming attribute files into a > > > > separate __iio_chan_prefix_emit() function for reuse. > > > > ... > > > > > > +static int __iio_chan_prefix_emit(const struct iio_chan_spec *chan, > > > > + enum iio_shared_by shared_by, > > > > + char *buf, size_t len) > > > > +{ > > > > + const char *dir = iio_direction[chan->output]; > > > > + const char *type = iio_chan_type_name_spec[chan->type]; > > > > + int n = 0; > > > > + > > > > + switch (shared_by) { > > > > + case IIO_SHARED_BY_ALL: > > > > + buf[0] = '\0'; /* empty channel prefix */ > > > > + break; > > > > + case IIO_SHARED_BY_DIR: > > > > + n = scnprintf(buf, len, "%s", dir); > > > > + break; > > > > + case IIO_SHARED_BY_TYPE: > > > > + n = scnprintf(buf, len, "%s_%s", dir, type); > > > > + if (chan->differential) > > > > + n += scnprintf(buf + n, len - n, "-%s", type); > > > > + break; > > > > + case IIO_SEPARATE: > > > > + if (chan->indexed) { > > > > + n = scnprintf(buf, len, "%s_%s%d", dir, type, > > > > + chan->channel); > > > > + if (chan->differential) > > > > + n += scnprintf(buf + n, len - n, > > > > "-%s%d", type, > > > > + chan->channel2); > > > > + } else { > > > > + if (chan->differential) { > > > > + WARN(1, "Differential channels must be > > > > indexed\n"); > > > > + return -EINVAL; > > > > + } > > > > + n = scnprintf(buf, len, "%s_%s", dir, type); > > > > + } > > > > + > > > > + if (chan->modified) { > > > > + if (chan->differential) { > > > > + WARN(1, "Differential channels can not > > > > have modifier\n"); > > > > + return -EINVAL; > > > > > > WARN() looks too much to me. dev_error() as we're treating it as such. I > > > guess you don't want to pass struct device but not really an issue IMHO. > > > > __iio_device_attr_init() also used WARN(), probably because it didnt have > > access to a dev pointer. It would not be a problem to add an extra param. > > Hmm, fair enough. Maybe a chance to change it. Not sure how others feel > about it. I was young and knew less back then (either as reviewer or author). WARN() on a lot of systems actually means panic, so for stuff like this it isn't appropriate. I'd definitely support a series flipping any WARN() to dev_err() or similar. Thanks, Jonathan
