[GHC] #11520: GHC falls into a hole if given incorrect kind signature

GHC ghc-devs at haskell.org
Mon Feb 1 14:10:36 UTC 2016


#11520: GHC falls into a hole if given incorrect kind signature
-------------------------------------+-------------------------------------
        Reporter:  bgamari           |                Owner:
            Type:  bug               |               Status:  new
        Priority:  highest           |            Milestone:  8.0.1
       Component:  Compiler (Type    |              Version:  8.0.1-rc1
  checker)                           |
      Resolution:                    |             Keywords:  TypeInType
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Incorrect         |  Unknown/Multiple
  warning at compile-time            |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by goldfire):

 The `but does not fail` bit is just to accurately describe the function.
 As usual, failing is fine, but it means you get fewer error messages at a
 go. In other words, we can just change the behavior.

 But I much favor just fixing `fvProv` to do something other than panic on
 a coercion hole, like return an empty list. There are a bunch of these
 panics on coercion holes where I thought coercion holes should never be
 seen. But I've learned that erroneous code can make GHC take paths I
 hadn't considered. So I'll remove all of these panics and replace them
 with something that won't crash GHC.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/11520#comment:6>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list