linux-apus-devel

Re: Bug#271878: asm/setup.h non-existent #include on ppc

[Geert U. <ge...@li...>]

On Sat, 25 Sep 2004, GOTO Masanori wrote:
> At Sat, 25 Sep 2004 11:58:56 +1000,
> Benjamin Herrenschmidt wrote:
> > On Sat, 2004-09-25 at 11:44, GOTO Masanori wrote:
> > > At Wed, 15 Sep 2004 14:51:00 -0500,
> > > Stephen R. Marenka <sma...@de...> wrote:
> > > > asm/setup.h contains #include <asm-m68k/setup.h>, 
> > > > which is non-existent on ppc.
> > > 
> > > Then what's the problem?  I guess you didn't get any problems; if so I
> > > close this bug.
> > > 
> > > BTW, PPC guys, is it intentional to include asm-m68k/setup.h in
> > > /usr/include/asm-ppc/setup.h ?  At a glance, we have no need to
> > > include this header in setup.h.
> > 
> > This is historically used by the APUS platform, which I think is an m68k
> > amiga with a ppc CPU stucked in it, but then, ask Geert, he'll know
> > better.
> 
> Thanks for your reply.  Geert, is it APUS still alive?  If so, I guess
> asm-m68k/setup.h is necessary for asm-ppc/setup.h.

Yes, AFAIK APUS is still alive.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- ge...@li...

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: Bug#271878: asm/setup.h non-existent #include on ppc

[Tom R. <tr...@ke...>]

On Sun, Sep 26, 2004 at 02:02:59PM -0700, Tom Rini wrote:

> Hmmm.. How many of these <asm-ppc/foo.h> #includes <asm-m68k/foo.h> (or
> #else's them in) does glibc (or really, any non-kernel user) need?
> Would someone have the time to try casing all of these inside of #ifdef
> __KERNEL__ ?   I've finished doing similar for SH recently (asm/cpu/ and
> asm/mach/ symlinks in this case) and found none of it was really needed.

How old were the kernel headers in question here?  In 2.6 and 2.4
current, <asm-ppc/setup.h> is already protected with a __KERNEL__ test.

-- 
Tom Rini
http://gate.crashing.org/~trini/

Re: Bug#271878: asm/setup.h non-existent #include on ppc

[GOTO M. <go...@de...>]

At Fri, 1 Oct 2004 14:10:09 -0700,
Tom Rini wrote:
> On Sun, Sep 26, 2004 at 02:02:59PM -0700, Tom Rini wrote:
> 
> > Hmmm.. How many of these <asm-ppc/foo.h> #includes <asm-m68k/foo.h> (or
> > #else's them in) does glibc (or really, any non-kernel user) need?
> > Would someone have the time to try casing all of these inside of #ifdef
> > __KERNEL__ ?   I've finished doing similar for SH recently (asm/cpu/ and
> > asm/mach/ symlinks in this case) and found none of it was really needed.
> 
> How old were the kernel headers in question here?  In 2.6 and 2.4
> current, <asm-ppc/setup.h> is already protected with a __KERNEL__ test.

We were discussing about 2.6.0-test7.  Thanks for your checking - even
if that version is not enclosed __KERNEL__, we've just known it'll be
fixed in the next linux-kernel-headers update.

Regards,
-- gotom

Re: Bug#271878: asm/setup.h non-existent #include on ppc

[Sven L. <sve...@wa...>]

On Sun, Sep 26, 2004 at 08:42:38PM +0200, Geert Uytterhoeven wrote:
> On Sat, 25 Sep 2004, GOTO Masanori wrote:
> > At Sat, 25 Sep 2004 11:58:56 +1000,
> > Benjamin Herrenschmidt wrote:
> > > On Sat, 2004-09-25 at 11:44, GOTO Masanori wrote:
> > > > At Wed, 15 Sep 2004 14:51:00 -0500,
> > > > Stephen R. Marenka <sma...@de...> wrote:
> > > > > asm/setup.h contains #include <asm-m68k/setup.h>, 
> > > > > which is non-existent on ppc.
> > > > 
> > > > Then what's the problem?  I guess you didn't get any problems; if so I
> > > > close this bug.
> > > > 
> > > > BTW, PPC guys, is it intentional to include asm-m68k/setup.h in
> > > > /usr/include/asm-ppc/setup.h ?  At a glance, we have no need to
> > > > include this header in setup.h.
> > > 
> > > This is historically used by the APUS platform, which I think is an m68k
> > > amiga with a ppc CPU stucked in it, but then, ask Geert, he'll know
> > > better.
> > 
> > Thanks for your reply.  Geert, is it APUS still alive?  If so, I guess
> > asm-m68k/setup.h is necessary for asm-ppc/setup.h.
> 
> Yes, AFAIK APUS is still alive.

Indeed, and thanks to Roman, we should even ressurect debian support for it
for sarge :)

Friendly,

Sven Luther

Re: Bug#271878: asm/setup.h non-existent #include on ppc

[Tom R. <tr...@ke...>]

On Sun, Sep 26, 2004 at 08:42:38PM +0200, Geert Uytterhoeven wrote:
> On Sat, 25 Sep 2004, GOTO Masanori wrote:
> > At Sat, 25 Sep 2004 11:58:56 +1000,
> > Benjamin Herrenschmidt wrote:
> > > On Sat, 2004-09-25 at 11:44, GOTO Masanori wrote:
> > > > At Wed, 15 Sep 2004 14:51:00 -0500,
> > > > Stephen R. Marenka <sma...@de...> wrote:
> > > > > asm/setup.h contains #include <asm-m68k/setup.h>, 
> > > > > which is non-existent on ppc.
> > > > 
> > > > Then what's the problem?  I guess you didn't get any problems; if so I
> > > > close this bug.
> > > > 
> > > > BTW, PPC guys, is it intentional to include asm-m68k/setup.h in
> > > > /usr/include/asm-ppc/setup.h ?  At a glance, we have no need to
> > > > include this header in setup.h.
> > > 
> > > This is historically used by the APUS platform, which I think is an m68k
> > > amiga with a ppc CPU stucked in it, but then, ask Geert, he'll know
> > > better.
> > 
> > Thanks for your reply.  Geert, is it APUS still alive?  If so, I guess
> > asm-m68k/setup.h is necessary for asm-ppc/setup.h.
> 
> Yes, AFAIK APUS is still alive.

Hmmm.. How many of these <asm-ppc/foo.h> #includes <asm-m68k/foo.h> (or
#else's them in) does glibc (or really, any non-kernel user) need?
Would someone have the time to try casing all of these inside of #ifdef
__KERNEL__ ?   I've finished doing similar for SH recently (asm/cpu/ and
asm/mach/ symlinks in this case) and found none of it was really needed.

-- 
Tom Rini
http://gate.crashing.org/~trini/

Re: Bug#271878: asm/setup.h non-existent #include on ppc

[GOTO M. <go...@de...>]

At Sun, 26 Sep 2004 14:02:59 -0700,
Tom Rini wrote:
> On Sun, Sep 26, 2004 at 08:42:38PM +0200, Geert Uytterhoeven wrote:
> > On Sat, 25 Sep 2004, GOTO Masanori wrote:
> > > At Sat, 25 Sep 2004 11:58:56 +1000,
> > > Benjamin Herrenschmidt wrote:
> > > > On Sat, 2004-09-25 at 11:44, GOTO Masanori wrote:
> > > > > At Wed, 15 Sep 2004 14:51:00 -0500,
> > > > > Stephen R. Marenka <sma...@de...> wrote:
> > > > > > asm/setup.h contains #include <asm-m68k/setup.h>, 
> > > > > > which is non-existent on ppc.
> > > > > 
> > > > > Then what's the problem?  I guess you didn't get any problems; if so I
> > > > > close this bug.
> > > > > 
> > > > > BTW, PPC guys, is it intentional to include asm-m68k/setup.h in
> > > > > /usr/include/asm-ppc/setup.h ?  At a glance, we have no need to
> > > > > include this header in setup.h.
> > > > 
> > > > This is historically used by the APUS platform, which I think is an m68k
> > > > amiga with a ppc CPU stucked in it, but then, ask Geert, he'll know
> > > > better.
> > > 
> > > Thanks for your reply.  Geert, is it APUS still alive?  If so, I guess
> > > asm-m68k/setup.h is necessary for asm-ppc/setup.h.
> > 
> > Yes, AFAIK APUS is still alive.

Cool, thanks Geert.

> Hmmm.. How many of these <asm-ppc/foo.h> #includes <asm-m68k/foo.h> (or
> #else's them in) does glibc (or really, any non-kernel user) need?
> Would someone have the time to try casing all of these inside of #ifdef
> __KERNEL__ ?   I've finished doing similar for SH recently (asm/cpu/ and
> asm/mach/ symlinks in this case) and found none of it was really needed.

Exactly, AFAIK, there are no headers to include them.

Regards,
-- gotom