GHC 7.0.4 recursion while trying to derive type

[Simon Peyton-Jones]

If you think there's a bug here, could you open a ticket a repro case, please?

Many thanks

Simon

| -----Original Message-----
| From: glasgow-haskell-users-bounces at haskell.org [mailto:glasgow-haskell-
| users-bounces at haskell.org] On Behalf Of Bogdan Opanchuk
| Sent: 09 January 2012 13:00
| To: Christian Maeder
| Cc: glasgow-haskell-users at haskell.org
| Subject: Re: GHC 7.0.4 recursion while trying to derive type
| 
| Hello Christian,
| 
| On Mon, Jan 9, 2012 at 11:31 PM, Christian Maeder
| <Christian.Maeder at dfki.de> wrote:
| > We have also such (cruel) code using these extension.
| 
| I understand that this code is not very good; I am just learning
| Haskell, and it is an experiment to see what its type system can
| handle.
| 
| >> I cannot answer this, but our code also compiles with ghc-7.2.2 using a
| >> lower context-stack (of 26).
| > Apologies, I've just re-checked and noticed that our code changed and needs
| > consistently -fcontext-stack=26 for ghc-7.2.2 and ghc-7.0.4 (and ghc-
| 7.0.2).
| 
| I just tried compiling my code with ghc7.0.4 and stack size 100, and
| it produces the same error. In fact, I would be surprised if it
| helped, because I would expect either infinite recursion (when type
| derivation fails) or successful compilation with some small stack size
| (definitely less than 26). The latter is the case for ghc7.2.2, which
| successfully compiles my code with stack size 3.
| 
| Best regards,
| Bogdan
| 
| _______________________________________________
| Glasgow-haskell-users mailing list
| Glasgow-haskell-users at haskell.org
| http://www.haskell.org/mailman/listinfo/glasgow-haskell-users