Re: [PATCH] iio: si1133: fix format string warnings

[Joe Perches]

On Fri, 2021-05-28 at 23:59 +0300, Andy Shevchenko wrote:
> On Thursday, May 27, 2021, Joe Perches <joe@xxxxxxxxxxx> wrote:
> > On Sun, 2021-05-16 at 10:36 +0100, Jonathan Cameron wrote:
> > > On Fri, 14 May 2021 10:45:02 -0700
> > > Nathan Chancellor <nathan@xxxxxxxxxx> wrote:
> > > > On 5/14/2021 6:59 AM, Arnd Bergmann wrote:
> > > > > From: Arnd Bergmann <arnd@xxxxxxxx>
> > > > > 
> > > > > clang complains about multiple instances of printing an integer
> > > > > using the %hhx format string:
> > > > > 
> > > > > drivers/iio/light/si1133.c:982:4: error: format specifies type
> > 'unsigned char' but the argument has type 'unsigned int' [-Werror,-Wformat]
> > > > >                   part_id, rev_id, mfr_id);
> > > > >                   ^~~~~~~
> > > > > 
> > > > > Print them as a normal integer instead, leaving the "#02"
> > > > > length modifier.
> > []
> > > > > diff --git a/drivers/iio/light/si1133.c b/drivers/iio/light/si1133.c
> > []
> > > > > @@ -978,11 +978,11 @@ static int si1133_validate_ids(struct iio_dev
> > *iio_dev)
> > > > >                   return err;
> > > > > 
> > > > > 
> > > > >           dev_info(&iio_dev->dev,
> > > > > -          "Device ID part %#02hhx rev %#02hhx mfr %#02hhx\n",
> > > > > +          "Device ID part %#02x rev %#02x mfr %#02x\n",
> > > > >                    part_id, rev_id, mfr_id);
> > > > >           if (part_id != SI1133_PART_ID) {
> > > > >                   dev_err(&iio_dev->dev,
> > > > > -                 "Part ID mismatch got %#02hhx, expected %#02x\n",
> > > > > +                 "Part ID mismatch got %#02x, expected %#02x\n",
> > 
> > which is almost certainly wrong.
> > the length specification includes the # which is already 2 bytes.
> > 
> > Likely these should be 0x%02x
> 
> What’s the difference (except printing 0)?

(assuming the argument is unsigned char/u8)

%#02x will always emit more than the specified length (3 or 4 chars)
values < 16 are 0x<hexdigit>, values >= 16 are 0x<hexdigit><hexdigit>

0x%02x will always emit 4 chars

It's very likely the writer didn't know the difference and assumed
that the # did not count in the specified width.