On Wed, 2005-04-27 at 09:35 -0700, Richard Henderson wrote: > On Wed, Apr 27, 2005 at 12:36:41AM -0600, Jeffrey A Law wrote: > > Anyway, I'm going to look into why we're seeing so many *& expressions > > during TER. > > We have an open PR for this. We don't propagate the & when > it's not a constant. Like in &x->y. yea, I figured that out about 10 minutes ago :-) I'm going to dig out some old patches from Andrew Pinskii which might address this problem. I'll note that a tree combiner would fix this problem :-)
jeff > > > r~
