On Wed, 28 Feb 2007, Linus Torvalds wrote:
>
> In other words, there is *zero* excuse for that braindamage.
To be clear:
- in header files, we put "common definitions":
* #defines
* data structure declarations
* external function and data declarations
* inline functions ("nicer but otherwise equivalent to a #define")
- but we do *not* put
* actual real code
* actual real data
because those go into C files.
Yes, yes, all rules have exceptions, and sometimes we have ugly header
files. For an example of a pre-existing ugly header file that breaks these
rules, just look at <asm-i386/bugs.h> for example. Yeah, it only gets
included from one place, but it *still* shouldn't have code in it. It grew
over time, and none of the individual events were ever really big enough
for anybody to say "ok, we should clean this up and create a bugs.c file
in arch/i386/kernel".
I'm sure there are other examples of the exceptions too. But I do not want
to add *new* ugly stuff, and I certainly refuse to do it after we're
already long past a merge window.
Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
