linux-mips
[Top] [All Lists]

Re: ddb5477 fixes for 2.6

To: "Maciej W. Rozycki" <macro@ds2.pg.gda.pl>
Subject: Re: ddb5477 fixes for 2.6
From: Daniel Jacobowitz <dan@debian.org>
Date: Thu, 18 Sep 2003 13:06:42 -0400
Cc: linux-mips@linux-mips.org, ralf@linux-mips.org
In-reply-to: <Pine.GSO.3.96.1030918185742.20533A-100000@delta.ds2.pg.gda.pl>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <20030918163344.GA22013@nevyn.them.org> <Pine.GSO.3.96.1030918185742.20533A-100000@delta.ds2.pg.gda.pl>
Sender: linux-mips-bounce@linux-mips.org
User-agent: Mutt/1.5.1i
On Thu, Sep 18, 2003 at 07:04:43PM +0200, Maciej W. Rozycki wrote:
> On Thu, 18 Sep 2003, Daniel Jacobowitz wrote:
> 
> > --- arch/mips/pci/pci.c     22 Jun 2003 23:09:48 -0000      1.3
> > +++ arch/mips/pci/pci.c     18 Sep 2003 16:25:08 -0000
> > @@ -66,6 +66,8 @@
> >  extern void pcibios_fixup(void);
> >  extern void pcibios_fixup_irqs(void);
> >  
> > +#if 0
> > +
> >  void __init pcibios_fixup_irqs(void)
> >  {
> >     struct pci_dev *dev = NULL;
> > @@ -126,6 +128,8 @@ void __init pcibios_fixup_resources(stru
> >     }
> >  
> >  }
> > +
> > +#endif
> >  
> >  struct pci_fixup pcibios_fixups[] = {
> >     {PCI_FIXUP_HEADER, PCI_ANY_ID, PCI_ANY_ID,
> 
>  Is it OK for other PCI systems?

Yes, I think so.  Those two functions seem to have migrated in from
pci-hplj.c; they can't possibly compile, since they use constants only
defined in that file.

> > --- include/asm-mips/addrspace.h    9 Aug 2003 21:16:38 -0000       1.11
> > +++ include/asm-mips/addrspace.h    18 Sep 2003 16:25:13 -0000
> > @@ -75,14 +75,14 @@
> >   * The compatibility segments use the full 64-bit sign extended value.  
> > Note
> >   * the R8000 doesn't have them so don't reference these in generic MIPS 
> > code.
> >   */
> > -#define XKUSEG                     0x0000000000000000
> > -#define XKSSEG                     0x4000000000000000
> > -#define XKPHYS                     0x8000000000000000
> > -#define XKSEG                      0xc000000000000000
> > -#define CKSEG0                     0xffffffff80000000
> > -#define CKSEG1                     0xffffffffa0000000
> > -#define CKSSEG                     0xffffffffc0000000
> > -#define CKSEG3                     0xffffffffe0000000
> > +#define XKUSEG                     0x0000000000000000ULL
> > +#define XKSSEG                     0x4000000000000000ULL
> > +#define XKPHYS                     0x8000000000000000ULL
> > +#define XKSEG                      0xc000000000000000ULL
> > +#define CKSEG0                     0xffffffff80000000ULL
> > +#define CKSEG1                     0xffffffffa0000000ULL
> > +#define CKSSEG                     0xffffffffc0000000ULL
> > +#define CKSEG3                     0xffffffffe0000000ULL
> >  
> >  /*
> >   * Cache modes for XKPHYS address conversion macros
> 
>  Why do you want these suffixes?  They don't work for assembly sources.

Because otherwise uses of XKPHYS in a 32-bit kernel generate noisy
warnings.  I don't remember where it was offhand.  Wrap it in
__ASSEMBLY__ if you like.

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer

<Prev in Thread] Current Thread [Next in Thread>