[GHC] #13474: GHC HEAD regression: Prelude.!!: index too large

GHC ghc-devs at haskell.org
Thu Mar 23 22:34:49 UTC 2017


#13474: GHC HEAD regression: Prelude.!!: index too large
-------------------------------------+-------------------------------------
        Reporter:  RyanGlScott       |                Owner:  (none)
            Type:  bug               |               Status:  new
        Priority:  highest           |            Milestone:  8.2.1
       Component:  Compiler          |              Version:  8.1
      Resolution:                    |             Keywords:
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  GHC rejects       |  Unknown/Multiple
  valid program                      |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |
-------------------------------------+-------------------------------------
Description changed by RyanGlScott:

Old description:

> `git-annex-6.20170321` from Hackage currently fails to build with GHC
> 8.2.1 and HEAD because it triggers this panic. Here's a simplified
> example that exhibits the issue:
>
> {{{#!hs
> {-# LANGUAGE BangPatterns #-}
> module Command.Vicfg where
>
> import qualified Data.Map as M
>
> class Default a where
>   def :: a
>
> instance Default Int where
>   def = 0
>
> data Cfg = Cfg
>     { cfgTrustMap :: M.Map Int Int
>     , cfgGroupMap :: M.Map Int Int
>     }
>
> defCfg :: Cfg -> Cfg
> defCfg curcfg = Cfg
>     { cfgTrustMap = mapdef $ cfgTrustMap curcfg
>     , cfgGroupMap = mapdef $ cfgGroupMap curcfg
>     }
>   where
>     mapdef :: Default v => M.Map k v -> M.Map k v
>     mapdef = M.map (const def)
> }}}
>
> {{{
> $ /opt/ghc/8.2.1/bin/ghc Bug.hs -O1
> [1 of 1] Compiling Command.Vicfg    ( Bug.hs, Bug.o )
> ghc: panic! (the 'impossible' happened)
>   (GHC version 8.2.0.20170321 for x86_64-unknown-linux):
>         Prelude.!!: index too large
> }}}
>
> Note:
>
> * You need optimization (`-O1` or `-O2`) enabled.
> * You need to import `Map`/`map` for this panic to trigger, it seems, as
> redefining `Map`/`map` locally makes it work again.

New description:

 `git-annex-6.20170321` from Hackage currently fails to build with GHC
 8.2.1 and HEAD because it triggers this panic. Here's a simplified example
 that exhibits the issue:

 {{{#!hs
 module Command.Vicfg where

 import qualified Data.Map as M

 class Default a where
   def :: a

 instance Default Int where
   def = 0

 data Cfg = Cfg
     { cfgTrustMap :: M.Map Int Int
     , cfgGroupMap :: M.Map Int Int
     }

 defCfg :: Cfg -> Cfg
 defCfg curcfg = Cfg
     { cfgTrustMap = mapdef $ cfgTrustMap curcfg
     , cfgGroupMap = mapdef $ cfgGroupMap curcfg
     }
   where
     mapdef :: Default v => M.Map k v -> M.Map k v
     mapdef = M.map (const def)
 }}}

 {{{
 $ /opt/ghc/8.2.1/bin/ghc Bug.hs -O1
 [1 of 1] Compiling Command.Vicfg    ( Bug.hs, Bug.o )
 ghc: panic! (the 'impossible' happened)
   (GHC version 8.2.0.20170321 for x86_64-unknown-linux):
         Prelude.!!: index too large
 }}}

 Note:

 * You need optimization (`-O1` or `-O2`) enabled.
 * You need to import `Map`/`map` for this panic to trigger, it seems, as
 redefining `Map`/`map` locally makes it work again.

--

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13474#comment:1>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list