Re: [RFC] New kernel-message logging API (take 2)

["Vegard Nossum" <vegard.nossum@gmail.com>]

On 9/28/07, Miguel Ojeda <maxextreme@gmail.com> wrote:
> On 9/28/07, Vegard Nossum <vegard.nossum@gmail.com> wrote:
> > reason we can't use KBUILD_MODNAME is that this is defined on the
> > command line. The declaration inside the header would thus be horribly
> > wrong. We can, however, use KBUILD_MODNAME as a default value for
> > KPRINT_DRIVER, like:
> > static const char *KPRINT_DRIVER = KBUILD_MODNAME;
> > which would pre-process to something like:
> > static const char *KPRINT_DRIVER = "bcm43xx";
> >
> > This value can still be overridden using #define KPRINT_DRIVER "new
> > name". In this case, it is possible that the original KPRINT_DRIVER
> > symbol can cause an "unused variable"-warning. I guess this is fixable
> > with the gcc "unused" variable attribute.
>
> Yep, then, in a year or two, we will be able to delete such attribute.

Actually, no, since it will throw a warning only if a source file
#defines KPRINT_SUBSYSTEM (i.e. overrides the constant variable
(oxymoron!) with the same name). What you're hoping is that some time
in the future, EVERY source file will come equipped with these
definitions, and yes, at that point, the entire declaration can be
removed, BUT I think that's... well. Yes.

> Will there be a team to change main subsystems/drivers to the new API?

No. First of all, this is a specification draft; there is no code yet.
Also, very possibly, this is such a violent change that nobody really
wants to use it anyway. But we can hope. ;-)

Vegard