Re: [gentoo-dev] QA

[Norberto Bensa]

Greg KH wrote:
> On Fri, Feb 20, 2004 at 01:19:26AM -0300, Norberto Bensa wrote:
> > Ernst Herzberg wrote:
> > > in file included from /usr/src/linux/include/linux/cdrom.h:14,
> > >                  from eject.c:66:
> > > /usr/src/linux/include/asm/byteorder.h:14: error: syntax error before
> > > "__u32"
> > > [...]
> > >
> > > (Reprocducable, try to emerge eject on stable)
> >
> > That's a know bug. Search {bf}.g.o. BTW, it's a bug in kernel headers;
> > complain to LKML.
>
> Yeah right, that will get you nowhere, they will laugh you out of
> there as it's not their problem.  Why are you including kernel headers
> in a userspace program.  That's a huge no-no.

I don't know why they do it, and honestly, I don't care. But there's something 
strange in /usr/src/linux/include/asm/byteorder.h. For example the following 
snippet:

	static __inline__ __attribute_const__ __u64 ___arch__swab64(__u64 val)
	{
	        union {
	                struct { __u32 a,b; } s;
	                __u64 u;
	        } v;
	        v.u = val;
	#ifdef CONFIG_X86_BSWAP
	        asm("bswapl %0 ; bswapl %1 ; xchgl %0,%1"
	            : "=r" (v.s.a), "=r" (v.s.b)
	            : "0" (v.s.a), "1" (v.s.b));
	#else
	   v.s.a = ___arch__swab32(v.s.a);
	        v.s.b = ___arch__swab32(v.s.b);
	        asm("xchgl %0,%1" : "=r" (v.s.a), "=r" (v.s.b) : "0" (v.s.a), 
	"1" (v.s.b));
	#endif
	        return v.u;
	}


Why "__inline__" instead of "inline" ? Or the other way around: why "asm" 
instead of "__asm__" ?

Ok, I know nothing about kernel's internals, so maybe you can enlight me :-)

Thanks,
Norberto