[E3-hacking] [PATCH v2 1/6] SoC Camera: add driver for OMAP1 camera interface

[Janusz Krzysztofik]

Tuesday 21 September 2010 15:48:33 Ralph Corderoy napisał(a):
> Hi Janusz,
>
> > > > +				"0x%0x\n", __func__, it_status);
> > > > +				"0x%0x\n", __func__, it_status);
> > > > +		dev_warn(dev, "%s: format %#x not found\n", __func__,
> > > > +		dev_warn(icd->dev.parent, "Format %#x not found\n",
> > >
> > > Two bytes could be saved with consistent use of %#x.  :-)
> >
> > ...
> >
> > In the first two cases, I intended to keep the "0x" prepended even if
> > 0. Am I missing something?
>
> No, I didn't realise that was the intent.  I agree, if you want zero to
> come out as "0x0" then you have to supply your own "0x".
>
> > In next iteration, I'll probably use "0x%.8x" when printing 32-bit
> > register values. What do you think?
>
> I'm probably more used to seeing "0x%08x", which is what I used to use
> myself, but either gives the same result.  I stopped using it because I
> found that the shape of the word is as relevant with hex as it is with
> English.
>
>     0 0x80 0x800080
>     0x00000000 0x00000080 0x00800080
>
> With the first line I can instantly see it's
>
>     Zero.
>     128 with the top-three bytes all zero.
>     Three bytes with the top byte zero.
>
> With the second line I have to scan the 0 digits carefully, looking for
> non-zeroes.

Fair enough, I'll convert them to "%#x".

Thanks,
Janusz