Re: [PATCH]: Cache immediate dominators

[Jan Hubicka <hubicka at kam dot mff dot cuni dot cz>]

> 	I tried this patch in the trunk and I do not see any noticeable
> difference in time, cycle count, or instruction count.  Why can't I
> duplicate your results?  Is this specific to the tree-ssa-branch?
I believe this is specific to SSA that is only so heavy user of the
function.
Concerning the implementation I was thinkig about slightly different
approach.  Currently the et_forest is using nodes for basic blocks and
occurences for each edge to the father in the tree.

THere is exactly one occurence (or two one for the empty tree?) for each
non-root node representing the edge to the father.  It should make sense
to put occurences into the nodes structure itself saving some memory
allocation overhead and making father lookup constant as well.

I didn't had time to do this yet.. Just an idea.

Honza
> 
> David