#3435: ghc-stage2 panic while compiling ghc-paths-0.1.0.5
-------------------------+--------------------------------------------------
    Reporter:  jsch      |        Owner:  simonmar
        Type:  bug       |       Status:  new     
    Priority:  high      |    Milestone:  6.12.1  
   Component:  Compiler  |      Version:  6.11    
    Severity:  normal    |   Resolution:          
    Keywords:            |   Difficulty:  Unknown 
    Testcase:            |           Os:  Linux   
Architecture:  x86       |  
-------------------------+--------------------------------------------------
Changes (by simonmar):

  * priority:  normal => high
  * difficulty:  => Unknown
  * owner:  => simonmar
  * milestone:  => 6.12.1

Comment:

 There are two things going on here.  The panic is because GHC has tried to
 read an interface file generated by an older version of GHC (do you have
 any idea how this happened?).

 The second thing is that GHC is supposed to report a helpful error message
 rather than panic when it reads an old interface file.  I've been
 suspicious that this isn't working properly for some time, given the
 number of panics we've been seeing.  I think I now understand what's
 happened.  The external binary format for `Int` was changed recently on
 32-bit platforms to be the same as on 64-bit platforms; unfortunately
 because GHC reads an `Int` before the version check on the `.hi` file, GHC
 sees a corrupt binary representation for the version and panics while
 reading it.

 So I'll fix the binary format to be stable again; unfortunately this will
 also change the binary format once more, so we'll have another round of
 panics before things settle down (note this is only an issue if you're
 working with HEAD, not a released version of GHC).

-- 
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/3435#comment:2>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs

Reply via email to