Re: m68k build failure

[Geert Uytterhoeven <geert@linux-m68k.org>]

On Wed, 28 Nov 2007, Pierre Ossman wrote:
> On Wed, 28 Nov 2007 09:28:56 +0000
> Al Viro <viro@ftp.linux.org.uk> wrote:
> > 
> > Eh...  m68k has 16bit alignment for unsigned long.  
> > 
> > diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
> > --- a/include/linux/mod_devicetable.h
> > +++ b/include/linux/mod_devicetable.h
> > @@ -343,7 +343,8 @@ struct sdio_device_id {
> >  	__u8	class;			/* Standard interface or SDIO_ANY_ID */
> >  	__u16	vendor;			/* Vendor or SDIO_ANY_ID */
> >  	__u16	device;			/* Device ID or SDIO_ANY_ID */
> > -	kernel_ulong_t driver_data;	/* Data private to the driver */
> > +	kernel_ulong_t driver_data	/* Data private to the driver */
> > +		__attribute__((aligned(sizeof(kernel_ulong_t))));
> >  };
> >  
> >  /* SSB core, see drivers/ssb/ */
> 
> Unfortunately, that just papers over the symptom and doesn't solve the underlying issue. If you cross-compile on/for an arch with byte alignment, then the issue is back. Or one that uses 4-byte alignment even for u16.
> 
> Is there no directive we can stick in there that forces a reasonable alignment (e.g. alignment == sizeof(type)) independently of arch?

We could use something like is used for compat_*.
E.g. compare compat_s64 in <asm/compat.h> for x86 and powerpc.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

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