[GHC] #10083: ghc: panic! (the 'impossible' happened)

GHC ghc-devs at haskell.org
Thu May 12 08:05:08 UTC 2016


#10083: ghc: panic! (the 'impossible' happened)
-------------------------------------+-------------------------------------
        Reporter:  hedayaty          |                Owner:
            Type:  bug               |               Status:  new
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  7.8.4
      Resolution:                    |             Keywords:  hs-boot
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Compile-time      |  Unknown/Multiple
  crash                              |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by ezyang):

 Simon and I agreed on our call that we should pursue inserting `noinline
 f` for any references to `f` from an `hs-boot` file. The `noinline` gets
 removed during `CorePrep` and as a result we won't infinitely try to
 unfold when we compile the module.

 How do we know when to insert these `noinline`s? It should happen in the
 typechecker or renamer, and the idea will be to mark `Id`s which come from
 a boot file as "boot identities"; then we know to insert `noinline`.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10083#comment:27>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list