[PATCH 12/14] x86: remove address space overrides using set_fs()

Christoph Hellwig hch at lst.de
Fri Sep 4 16:38:13 AEST 2020


On Fri, Sep 04, 2020 at 03:55:10AM +0100, Al Viro wrote:
> On Thu, Sep 03, 2020 at 04:22:40PM +0200, Christoph Hellwig wrote:
> 
> > diff --git a/arch/x86/lib/getuser.S b/arch/x86/lib/getuser.S
> > index c8a85b512796e1..94f7be4971ed04 100644
> > --- a/arch/x86/lib/getuser.S
> > +++ b/arch/x86/lib/getuser.S
> > @@ -35,10 +35,19 @@
> >  #include <asm/smap.h>
> >  #include <asm/export.h>
> >  
> > +#ifdef CONFIG_X86_5LEVEL
> > +#define LOAD_TASK_SIZE_MINUS_N(n) \
> > +	ALTERNATIVE "mov $((1 << 47) - 4096 - (n)),%rdx", \
> > +		    "mov $((1 << 56) - 4096 - (n)),%rdx", X86_FEATURE_LA57
> > +#else
> > +#define LOAD_TASK_SIZE_MINUS_N(n) \
> > +	mov $(TASK_SIZE_MAX - (n)),%_ASM_DX
> > +#endif
> 
> Wait a sec... how is that supposed to build with X86_5LEVEL?  Do you mean
> 
> #define LOAD_TASK_SIZE_MINUS_N(n) \
> 	ALTERNATIVE __stringify(mov $((1 << 47) - 4096 - (n)),%rdx), \
> 		    __stringify(mov $((1 << 56) - 4096 - (n)),%rdx), X86_FEATURE_LA57
> 
> there?

Don't ask me about the how, but it builds and works with X86_5LEVEL,
and the style is copied from elsewhere..


More information about the Linuxppc-dev mailing list