FW: HEAD build failure: incomplete patterns

Simon Peyton-Jones simonpj at microsoft.com
Mon Aug 15 14:00:48 CEST 2011


Ian

Can you reproduce?  I don't have a 7.2 conveniently available (Simon M -- should it become default on cam-04?), and there is some fix to binary to do?

I tried with a standalone test case but that worked fine.

Simon

-----Original Message-----
From: cvs-ghc-bounces at haskell.org [mailto:cvs-ghc-bounces at haskell.org] On Behalf Of Daniel Fischer
Sent: 14 August 2011 16:58
To: cvs-ghc at haskell.org
Subject: HEAD build failure: incomplete patterns

After (locally) fixing the imports in binary, building HEAD with 7.2.1 
fails with

> compiler/cmm/CmmNode.hs:187:3:
>     Warning: Pattern match(es) are non-exhaustive
>     
>              In an equation for `entryLabel':
>                  Patterns not matched:
>                      CmmComment _
>                      CmmAssign _ _
>                      CmmStore _ _
>                      CmmUnsafeForeignCall _ _ _
>                      ...
> 
> compiler/cmm/CmmNode.hs:189:3:
>     Warning: Pattern match(es) are non-exhaustive
>     
>              In an equation for `successors':
>                  Patterns not matched:
>                      CmmEntry _
>                      CmmComment _
>                      CmmAssign _ _
>                      CmmStore _ _
>                      ...
> 
> <no location info>:
> Failing due to -Werror.


At the top of CmmNode.hs, there is

{-# OPTIONS_GHC -fno-warn-incomplete-patterns #-}
#if __GLASGOW_HASKELL__ >= 701
-- GHC 7.0.1 improved incomplete pattern warnings with GADTs
{-# OPTIONS_GHC -fwarn-incomplete-patterns #-}
#endif

but 7.2.1 doesn't recognise that the incriminated patterns have the wrong 
type.

_______________________________________________
Cvs-ghc mailing list
Cvs-ghc at haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc




More information about the Cvs-ghc mailing list