This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: A latent bug - use of BRANCH_EDGE on trees.
On Sunday 24 October 2004 22:01, Zdenek Dvorak wrote:
> > There is nothing specifically ugly about IR independence in principle,
> > except that:
> > 1) people expect BRANCH_EDGE to be cheap and you'd add a function
> > call to it so it is needlessly expensive.
>
> you really believe that this would make any measurable difference
> in compile time???? Uhhh...
Not really, but it still is an unnecessary abstraction penalty.
> > 2) BRANCH_EDGE really doesn't have much of a meaning on trees, so
> > making it work on trees just does not make sense.
>
> You may rename it to something more senseful (like EDGE_TRUE, or
> whatever); in a sense what I find ugly is having both
> BRANCH_EDGE/FALLTHRU_EDGE and extract_true_false_edges_from_block.
IMHO having EDGE_{TRUE,FALSE}_VALUE on trees and {BRANCH,FALLTRHU}_EDGE
on RTL is an unfortunate thing that we'll probably need to clean up.
Oh well, as long as whatever Kazu comes up with allows us to finally
take more advantage of the edge-vector work...
Gr.
Steven