Re: [PATCH] bluetooth hidp is broken on s390

[Martin Schwidefsky <schwidefsky@de.ibm.com>]

On Sat, 2005-10-29 at 19:47 +0100, Al Viro wrote:
> On Sat, Oct 29, 2005 at 01:05:49PM +0200, Marcel Holtmann wrote:
> > Hi Al,
> > 
> > > 	Bluetooth HIDP selects INPUT and it really needs it to be
> > > there - module depends on input core.  And input core is never
> > > built on s390...  Marked as broken on s390, for now; if somebody
> > > has better ideas, feel free to fix it and remove dependency...
> > > 
> > > Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
> > 
> > basically I think someone should fix the input layer on S390, but I am
> > fine with your fix.

And since there isn't a Bluetooth device on s390 (yet), we can live with
the broken input core for now. Would be nicer if the input layer would
work but there are more important things..

> There's a problem with that patch, though ;-/  S390 is never defined by
> arch/s390/Kconfig; ARCH_S390 is.  From the fast grep it looks like there
> is one more place with the same problem: drivers/char/Kconfig has
> config HW_CONSOLE
>         bool
>         depends on VT && !S390 && !UML
>         default y
> and the second term is always true here.  Why do we have that dependency,
> anyway, when s390 doesn't include drivers/char/Kconfig at all?

.. e.g. this one. I tried to cleanup the Kconfig for s390 once. After
the patch has grown over 2000 lines of code I gave up. If you have the
feeling that this should be fixed I'll give it another try. At least the
issue with ARCH_S390/ARCH_S390X vs. S390/S390_64 should be fixed.

-- 
blue skies,
   Martin

Martin Schwidefsky
Linux for zSeries Development & Services
IBM Deutschland Entwicklung GmbH