Re: [PATCH] kbuild: drop check for CONFIG_ in headers_check

[Ingo Molnar]

* Arnd Bergmann <arnd@xxxxxxxx> wrote:

> On Saturday 31 January 2009, Ingo Molnar wrote:
> > Jaswinder, i think we now have a realistic chance to reach a zero baseline 
> > with these warnings, for v2.6.29. Here are the remaining warnings (summed 
> > up per file) on x86 (after Sam's patch):
> > 
> >      17 /dev/shm/tip/usr/include/linux/reiserfs_fs.h
> >      12 /dev/shm/tip/usr/include/linux/soundcard.h
> >       3 /dev/shm/tip/usr/include/linux/nubus.h
> >       2 /dev/shm/tip/usr/include/linux/in6.h
> >       1 /dev/shm/tip/usr/include/mtd/jffs2-user.h
> >       1 /dev/shm/tip/usr/include/linux/socket.h
> >       1 /dev/shm/tip/usr/include/linux/coda_psdev.h
> >       1 /dev/shm/tip/usr/include/linux/bsg.h
> > 
> 
> Ah, that doesn't look too bad. On a related subject, was there ever a 
> conclusion about whether we want to support __KERNEL_STRICT_NAMES in 
> every header? When that is set, headers cannot use most of the standard 
> types (dev_t, mode_t, off_t, pid_t, suseconds_t, timer_t, uid_t, gid_t, 
> loff_t, time_t, clock_t, caddr_t, pgoff_t, {u_,u,}int{8,16,32,64}_6}) 
> but have to use the __kernel_* versions of that.

hm, that looks quite ugly and unnatural. Could the usr/* exporting scripts 
convert these types automatically to the __kernel_* versions?

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/