[committed, PATCH] Remove Disp16|Disp32 from 64-bit direct branches

Michael Matz matz@suse.de
Tue May 12 15:32:00 GMT 2015


Hi,

On Tue, 12 May 2015, H.J. Lu wrote:

> >> I thought it did jump to "(nextip + offset) & 0xffff" on AMD.  Can 
> >> you verify if it is true?
> >
> > Sorry, yes, this is true, the 16bit operand size prefix truncates RIP 
> > to 16 bit.  Nevertheless it's a valid instruction and works as 
> > documented.
> 
> It doesn't work on Intel processors.  Can you show me a real usage for 
> this?

Why should I?  Obviously in linux there are no uses, but perhaps in BIOS 
or boot loader code, or micro-OSes and the like.  Can you instead show me 
why you want to remove support for it?


Ciao,
Michael.



More information about the Binutils mailing list