Re: [PATCH] frv: fix fallout from "remove sched.h from mm.h" patch

[Alexey Dobriyan <adobriyan@sw.ru>]

On Fri, Jun 29, 2007 at 09:10:52AM -0400, Robert P. J. Day wrote:
> On Thu, 28 Jun 2007, Andrew Morton wrote:
> > On Thu, 28 Jun 2007 18:09:21 +0400
> > Alexey Dobriyan <adobriyan@sw.ru> wrote:
> >
> > > Signed-off-by: Alexey Dobriyan <adobriyan@sw.ru>
> > > ---
> > >
> > >  include/asm-frv/pgtable.h |    2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > --- a/include/asm-frv/pgtable.h
> > > +++ b/include/asm-frv/pgtable.h
> > > @@ -25,7 +25,7 @@
> > >  #include <linux/slab.h>
> > >  #include <linux/list.h>
> > >  #include <linux/spinlock.h>
> > > -struct mm_struct;
> > > +#include <linux/sched.h>
> > >  struct vm_area_struct;
> > >  #endif
> > >
> >
> > Please provide changelogs.
> >
> > I assume this patch fixes some build error or something.
>
>   i actually asked about this on either the janitors or newbies list
> the other day -- one of the early examples from the LDD3 device
> drivers book which built fine all this time suddenly stopped building,
> until i explicitly included <linux/sched.h> to be able to dereference
> a pointer to "task_struct":
>
> /home/rpjday/AMD/k/topics/0_hi/hi1.c:15: error: dereferencing pointer to incomplete type
> /home/rpjday/AMD/k/topics/0_hi/hi1.c:16: error: dereferencing pointer to incomplete type
>
>   clearly, until now, that header file was being picked up
> *somewhere*,

from include/linux/mm.h , yes

> but not now.

dammit!

> so i'm guessing there's going to be code
> all over the place that suddenly needs to add that inclusion of
> sched.h.

this is the last known to me bit of compile breakage that was caused by
commit e8edc6e03a5c8562dc70a6d969f732bdb355a7e7 aka
"[PATCH] Detach sched.h from mm.h"

And after you'll check diffstat and build current kernel for different archs
yourself, I hope you'll take that "all over the place" back.

P.S.: Al may know more because he is building different set of configs, but I
      was larted about frv only.