Advanced Micro Devices Inc (AMD)

[fastpathguru]

Ahhh, so a mere microcode edit used to bring a processor from x86 to x86-64 only requires an aborted incompatible 64-bit implementation to already exist on-die?

I wouldn't call that exactly supportive of your "IA32E is an instruction extension - a microcode edit." claim... From the context of the thread, it was not reasonable to assume your "microcode edit" depended upon already existing 64-bit infrastructure in Prescott.

Here's an earlier quote in the thread from you:

"I don't see IA32E as anything more than new instructions, like SSE-x."

Well duh, of course it's "just" new instructions. But do you think "SSE-x" could also be implemented with mere microcode changes?

fpg