🚀 go-pugleaf

In article <324@dumbo.UUCP>, hansen@mips.UUCP (Craig Hansen) writes:
[ Stuff ]
> 
> The feature that most caught my eye was that of branch prediction,
> for which I quote from the "Technical Introduction":
> 
> 	The selection is based on the branch condition and
> 	direction; forward branches are predicted as taken,
> 	backward branches as non-taken.
> 
[ Alot more stuff ]
I too had wondered about this branch prediction mess, and I agree with this
poster about this being the WRONG way to do things.  I was prepared to make
the argument that trying to find the branch direction would require fetching
the offset (assuming relative branches), which would obligate the processor
to a possibly unneeded fetch, along with running this offset through some logic
(one or two more micro instructions) to figure out the direction, which
sounded like a real loser.  However, I suspect that someone at AMD would tell
me how wonderfully pipelined their machine was and so on ...

[ Now for the revised posting ]

Craig stated that, like the rest of us, he was raised to believe that
backwards branches make nice branches after while loops and such, and
forward branches make nice branches before if-thens.  Has anyone bothered
to think about adding branch instructions that tell the CPU explicitly
whether or not they expect to be taken?  For example,

	MBLEQ - most of the time branch less or equal
	SBLEQ - seldom branch less or equal

Compilers would be able to pick the instructions they wanted.  (Speaking of
patents, if this idea is patentable and has not yet been so, anyone but
Nat-Semi can have it ... :-) :-) so long as I get credit ... :-) :-)

- John		(jfh@killer.UUCP)

Disclaimer:
	No disclaimer.  Whatcha gonna do, sue me?