[GHC] #4160: Panic! at the HEAD

[GHC]

#4160: Panic! at the HEAD
-------------------------------------------------+--------------------------
  Reporter:  LouisWasserman                      |          Owner:  simonpj
      Type:  bug                                 |         Status:  closed 
  Priority:  high                                |      Milestone:  7.0.1  
 Component:  Compiler                            |        Version:  6.13   
Resolution:  fixed                               |       Keywords:         
  Testcase:  indexed-types/should_compile/T4160  |      Blockedby:  4232   
Difficulty:                                      |             Os:  Linux  
  Blocking:                                      |   Architecture:  x86    
   Failure:  Compile-time crash                  |  
-------------------------------------------------+--------------------------
Changes (by simonpj):

  * status:  new => closed
  * testcase:  => indexed-types/should_compile/T4160
  * resolution:  => fixed


Comment:

 This turned out to be a very subtle bug in the coercion optimiser, which
 only manifested in a rather complicated situation.  Thank you for
 providing a nice small test case; it still took me a good chunk of
 Thursday to find.  And, sadly, it's in code that will probably die when
 Brent Yorgey's new coercion patch comes on stream.

 Oh well, it's good to have the HEAD in working order meanwhile.
 {{{
 Thu Sep 16 18:04:52 BST 2010  simonpj at microsoft.com
   * Fix a very subtle shadowing bug in optCoercion

   See Note [Subtle shadowing in coercions]

   This is what was going wrong in Trac 4160.

     M ./compiler/types/OptCoercion.lhs -20 +67
 }}}

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4160#comment:6>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler