[PATCH 8/19] powerpc: interfaces to the hypervisor of Celleb

Ishizaki Kou kou.ishizaki at toshiba.co.jp
Wed Dec 20 19:28:44 EST 2006


Geert-san,

> On Thu, 14 Dec 2006, Ishizaki Kou wrote:
> > --- /dev/null      Wed Dec 13 21:32:01 2006
> > +++ linux-powerpc-git/arch/powerpc/platforms/celleb/hvCall_macro.S	Wed Dec 13 18:37:14 2006
> 
> > +#define	__define_beat_callup_handler_NOOUTS(name, id)		\
> > +_GLOBAL(beat_##name)										\
> > +													HCALL_MEDIUM;	     \
> > +													mflr		     r0;      \
> > +													std		     r0, 16(r1);     \
> > +													HCALL_INST_PRECALL;		       \
> > +													LOAD_CONST64(r11,id);		           \
> > +													HVSC;					         \
> > +													HCALL_INST_POSTCALL;				     \
> > +													ld						     r0, 16(r1);     \
> > +													mtlr						     r0;	         \
> > +													blr
> 
> `id' is always the name prefixed with `HV_'? In that case you can use
> `HV_##name', so you don't have to pass the number.

It's in that case, since `id' is assumed to be a hypervisor-call number,
and defined in celleb/beat_syscall.h.

> > --- /dev/null	Wed Dec 13 21:32:02 2006
> > +++ linux-powerpc-git/arch/powerpc/platforms/celleb/hvCall.S	Wed Dec 13 18:37:14 2006
> 
> > +__define_beat_callup_handler_OUT2(4, allocate_memory, (0))
>                                                          ^^^
> I.e., these can go away.

I'll modify them so.
These sources are generated from our 'hvcall table', so we have to
modify the generator.

Best Regards,
Kou Ishizaki



More information about the Linuxppc-dev mailing list