[GHC] #5980: Name of compiler mismatch in safeHaskell/ghci/p5 p7 p8

[GHC]

#5980: Name of compiler mismatch in safeHaskell/ghci/p5 p7 p8
---------------------------------+------------------------------------------
    Reporter:  goldfire          |       Owner:                  
        Type:  bug               |      Status:  new             
    Priority:  highest           |   Milestone:  7.6.1           
   Component:  Test Suite        |     Version:  7.5             
    Keywords:                    |          Os:  Unknown/Multiple
Architecture:  Unknown/Multiple  |     Failure:  None/Unknown    
  Difficulty:  Unknown           |    Testcase:                  
   Blockedby:                    |    Blocking:                  
     Related:                    |  
---------------------------------+------------------------------------------
Changes (by igloo):

  * priority:  normal => highest
  * difficulty:  => Unknown
  * os:  MacOS X => Unknown/Multiple
  * architecture:  x86_64 (amd64) => Unknown/Multiple
  * milestone:  => 7.6.1


Comment:

 The tests should use "ghc", not "ghc-stage2". Currently this would fix the
 tests for `BINDIST=YES`, but break them without it.

 The driver is meant to normalise `ghc-stage2` to `ghc`, which would fix
 this. However, it only does this on stderr, and this message is going to
 stdout.

 As per [http://www.haskell.org/pipermail/cvs-ghc/2012-March/071033.html
 this thread], I think the right fix is for `ghc/InteractiveUI.hs
 showException` to send this message to stderr. Someone will need to look
 at what, if anything, `showException` should continue sending to stdout.

 I'm marking this as highest, because we should really get validate clean
 again, regardless of whether or not `--fast` is being used.

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5980#comment:1>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler