From: "Rob van der Heij" <[email protected]>
Sent: Tuesday, June 03, 2014 4:14 PM
On 3 June 2014 01:56, Robin Vowels <[email protected]> wrote:
XR Rn,Rn is faster than SR.
But does it matter?
Such an instruction should be executed only once, and once only.
It shouldn't be in the loop.
So why is it faster? I suppose "faster" means it would allow more options
for it to overlap with previous or following instructions? Not obvious to
me. Wouldn't the SR know the condition code upfront as well?
I briefly looked at shifting the bytes with SLDA to replace the IC for
every byte. But the penalty seemed the need to clear the register before
shifting the byte in. An XR or SR or whatever in the loop. I don't see how
one could prevent having that in the loop in such a case.
See my earlier example. SR (or XR or whatever) is required only once -- before
the loop.
---
This email is free from viruses and malware because avast! Antivirus protection
is active.
http://www.avast.com
