[solution exits] no PCMCIA in 4.0-rc3

[One Thousand Gnomes]

On Tue, 10 Mar 2015 21:31:58 +0100
> > Probably a PCMCIA not a block layer change - well I guess could be
> > either. I fixed a pile of PCMCIA bugs and also reworked the allocator for
> > pure PCI boxes so it didn't do various bad things.
> > 
> > Can you tweak drivers/pcmcia/Kconfig for config YENTA and make
> > 
> >         select PCCARD_PCI if PCMCIA !=n && !ISA   
> > 
> > instead do
> > 
> > 	select PCCARD_NONSTATIC if PCMCIA !=n && !ISA
> > 
> > rebuild and re-test
> 
> If I do that, it breaks compile. If I do that, and delete "config
> PCCARD_PCI", it fixes PCMCIA. Thanks!

Greg - can you revert 

commit 02b03846bb2befc558bfd0665749d6bb26f4c2f1
Author: Alan Cox <alan at linux.intel.com>
Date:   Wed Dec 10 15:07:36 2014 +0000

    pcmcia: add a new resource manager for non ISA systems
    
    On a pure PCI platform we don't actually need all the complexity of
    the rsrc_nonstatic manager, in fact we can just work directly with
    the pci allocators and avoid all the complexity (and code bloat).
    
    Signed-off-by: Alan Cox <alan at linux.intel.com>
    Signed-off-by: Greg Kroah-Hartman <gregkh at linuxfoundation.org>


and the one on top of it

commit d885d4f3728f386034bb2f7a61b7f2054c49b2d4
Author: Arnd Bergmann <arnd at arndb.de>
Date:   Tue Jan 13 15:11:55 2015 +0100

    pcmcia: add missing include for new pci resource handler
    
    The recently added rsrc_pci.c file calls pci_bus_alloc_resource without
    including the header file that declares it, and that sometimes causes
    a build warning on ARM:
    
    drivers/pcmcia/rsrc_pci.c: In function 'find_io_region':
    drivers/pcmcia/rsrc_pci.c:40:2: error: implicit declaration of function 'pci_bus_alloc_resource' [-Werror=implicit-function-declaration]
    
    This adds the missing include statement.
    
    Signed-off-by: Arnd Bergmann <arnd at arndb.de>
    Fixes: 02b03846bb2be ("pcmcia: add a new resource manager for non ISA systems")
    Acked-by: Alan Cox <alan at linux.intel.com>
    Reported-by: Jim Davis <jim.epost at gmail.com>
    Signed-off-by: Greg Kroah-Hartman <gregkh at linuxfoundation.org>


as it seems there is a regression in there for some configuration of I/O
based devices. I'll take a look at it over the next couple of kernel
releases and see what is up then resubmit it with fixes.