From: Alexei A. Frounze on 21 Aug 2008 04:20 On Aug 20, 10:59 pm, "H. Peter Anvin" <h...(a)zytor.com> wrote: > Alexei A. Frounze wrote: > > >> 48 90 = rex.w nop = xchg rax,rax > > > Should be a NOP effectively. > > >> 49 90 = rex.wb nop = xchg rax,r8 > > > Same here. > > Not a NOP, certainly... Oops, sorry, missed B. So, it's an XCHG, > > So, did you do the last two under a debugger? If you did, on what CPU > > brand? Intel, AMD or both? > > Not under a debugger, but yes, I executed them (inside a small C > program). AMD Athlon X2 4200+ (socket 939). Then what? The AMD doc confirmed and opcode 0x90 can indeed represent either NOP or XCHG? Alex
From: Wolfgang Kern on 21 Aug 2008 09:39 Alexei A. Frounze wrote: ..... >> 48 90 = rex.w nop = xchg rax,rax > Should be a NOP effectively. Yes, but this NOP may cost a few cycles in opposition to single bytes. __ wolfgang
From: H. Peter Anvin on 21 Aug 2008 12:11 Alexei A. Frounze wrote: > > Then what? The AMD doc confirmed and opcode 0x90 can indeed represent > either NOP or XCHG? > Yes. -hpa
From: Alexei A. Frounze on 21 Aug 2008 14:29
On Aug 21, 9:11 am, "H. Peter Anvin" <h...(a)zytor.com> wrote: > Alexei A. Frounze wrote: > > > Then what? The AMD doc confirmed and opcode 0x90 can indeed represent > > either NOP or XCHG? > > Yes. I had an impression you were opposing this and it now seems like you aren't/weren't. I must've misspoken something else... Oh, well. :) Alex |