> Even weirder, I am now trying to compile array.sml with all the > -drop-passes, and it appears to be in an infinite loop within > checkScopes after commonSubexpSSA. Yeah, I get that. It's somewhere in main_0, so it's a pain to figure out. I mean to ask, could the new dominator version that doesn't care about the graph being connected and single rooted result in a dominator tree that has a cycle?