On Mon, Dec 13, 2010 at 09:24:30PM +0100, L. Alberto Giménez wrote:
> On Mon, Dec 13, 2010 at 10:30:32AM -0800, Greg KH wrote:
> > On Mon, Dec 13, 2010 at 12:58:44AM +0100, L. Alberto Giménez wrote:
> > > Fix checkpatch complains.
> > >
> > > The change of the DBGPRINT_ERR macro causes the change on the callers.
> >
> > What change?
>
> -#define DBGPRINT_ERR(Fmt) \
> -{ \
> - printk("ERROR! "); \
> - printk Fmt; \
> -}
> +#define DBGPRINT_ERR(fmt, args...) printk(KERN_ERR fmt, ##args)
>
> Recommended by Joe. I agree that it is cleaner and more compliant.
Not really, it should be replaced with dev_err() or some such universal
replacement.
> That macro needed that the callers use two parens, as in:
>
> DBGPRINT_ERR(("MlmeInit failed, Status[=0x%08x]\n", Status));
>
> With the introduced change, all calls to the DBGPRINT_ERR macro must be
> translated to a regular single-parenthesized.
>
> DBGPRINT_ERR("NICInitializeAdapter failed, Status[=0x%08x]\n", Status);
>
> Yes, it's crap code, but I own an eeepc 901 and this was the perfect excuse to
> get my harnds dirty :)
That's fine.
How about splitting just the DBGPRINT_ERR() change into one patch, and
the other printk cleanups into another one? Remember, only do one thing
per patch please.
thanks,
greg k-h
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
