Hamish Marson wrote:
> Timothy Miller wrote:
>>> On 7/10/06, Hamish <[EMAIL PROTECTED]> wrote:
>>>
>>>>> NOOP.vh
>>>>>
>>>>> Is that good, or would that confuse people, since it's the same
>>>>>
>>>> syntax
>>>>
>>>> Actually calling it a NOOP confuses me... NOOP to me means
>>>> increment the program counter. Do nothing else. Affect no other
>>>> flags... Why separate by a .? Why not something like
>>>>
>>>> ASSERT   v,h
>>>>
>>>> It's clear, does what it says on the tin. etc.
>>> I see your point, but at the same time, all instructions assert
>>> flags. It's just that the NOOP does nothing BUT assert flags.  See?
>>> :)
>>>
> 
> Yes, but AFAIK on any other architecture, it doesn't. I know we're
> arging semantics here, and it's not even hardware really, but NOOP's
> have always been described as NO OPERATION. Including
> setting/resetting flags.
> 
> To redefine what a NOOP is, even for a private architecture like
> this.. It just doesn't feel right...
> 

Maybe it would make some sense treating these instructions as VLIW -
instruction composed of several sections. Then in one of them, you could
have NOOP as really nothing, and in another V/H sync and so on. Is there
some "standard" how to write an VLIW instruction in text to one line?

Daniel
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)

Reply via email to