projects / ghc-hetmet.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 8f0a33d) | patch
Fix long-standing bug in CPR analysis
authorsimonpj@microsoft.com <unknown>
Tue, 31 Jan 2006 15:32:47 +0000 (15:32 +0000)
committersimonpj@microsoft.com <unknown>
Tue, 31 Jan 2006 15:32:47 +0000 (15:32 +0000)
commit95c13506a9988cfe613618ef5c76fe95f4048d22
tree74dde0cee7c802af4ba1db90b8a6884626df2111tree | snapshot
parent8f0a33dd7c6c054df760d4c728677d3ed87b02edcommit | diff
Fix long-standing bug in CPR analysis

MERGE TO STABLE

For a long time (2002!) the CPR analysis done by
dmdAnalTopRhs has been bogus.  In particular, it's possible
for a newtype constructor to look CPR-ish when it simply isn't.

This fixes it.  Test is stranal/newtype
ghc/compiler/stranal/DmdAnal.lhs diff | blob | history
Glasgow Haskell Compiler with heterogeneous metaprogramming (generalized arrows and modal types)