#14880: GHC panic: updateRole -------------------------------------+------------------------------------- Reporter: RyanGlScott | Owner: goldfire Type: bug | Status: new Priority: normal | Milestone: 8.8.1 Component: Compiler (Type | Version: 8.2.2 checker) | Resolution: | Keywords: TypeInType Operating System: Unknown/Multiple | Architecture: Type of failure: Compile-time | Unknown/Multiple crash or panic | Test Case: Blocked By: | Blocking: Related Tickets: #15076 | Differential Rev(s): Phab:D4769 Wiki Page: | -------------------------------------+------------------------------------- Comment (by tdammers): Replying to [comment:84 simonpj]:
2. I think you mean that `NDFV` keeps a set of forall-bound variables, and checks in that set before adding to the accumulator. But foralls are rare (I think), so the set will typically be empty.
Well, considering the `del...` function for `FV` / `NDFV`: {{{#!hs delNDFV :: Var -> NDFV -> NDFV delNDFV var fv fv_cand !in_scope acc = fv fv_cand (extendVarSet in_scope var) acc }}} ...it turns out that *all* "deletions" from the "set" are implemented as adding to the `in_scope` set. Despite the name, the data structure doesn't really care whether they are really forall-bound variables. They might be though - I don't know if there are other reasons to delete from a free-var set.
3. That leaves the accumulator-style as the main candidate.
As Richard says, though, the way to find out is to try it one thing at a time.
Yes, absolutely. Will do exactly that. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/14880#comment:85 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler