Re: branch predictions

[Jason Eckhardt <jle at cygnus dot com>]

On Thu, 4 May 2000, Jeffrey A Law wrote:

> 
> 
> I can think of three heuristics which would work to resolve this problem.
> 
> 	1. Given two successors, if one unconditionally exits and the
> 	other does not, then predict the path which does not unconditionally
> 	exit.
> 
> 	2. Given two successors, if one passes control to a loop that it
> 	dominates and the other does not, then predict the path to the
> 	dominated loop.
> 
> 	3. If we do not have any kind of prediction and one of the blocks
> 	physically follows the current block, then predict the block which
> 	physically follows the current block.  ie, don't rearrange code
> 	at random ;-)  
> 
> I did a very quick and dirty implementation of #1.  That allowed us to
> properly predict the key branch and we got the desired code.
> 

   this is something that just fell through the cracks. The Ball paper
   discusses heuristics similar to #1,#2. I had on my todo list all of them,
   but never got to it due to reordering itself. Might as well install
   it if you've done the work. Thanks!
   jaosn.