[project @ 2001-07-24 09:55:16 by simonpj]
authorsimonpj <unknown>
Tue, 24 Jul 2001 09:55:16 +0000 (09:55 +0000)
committersimonpj <unknown>
Tue, 24 Jul 2001 09:55:16 +0000 (09:55 +0000)
commitba4b4b51476e2bbeb0de04f3c8f2a70e36a020f4
treef82b48a47937f70bffc47cf801222fc28f9ea8b9
parent198ee273db2c8ff68874a7af139f0a2a56a812e8
[project @ 2001-07-24 09:55:16 by simonpj]
Two fixes to the demand analyis.

1.  Don't look inside Coerces.  These wrap recursive newtypes,
and we might fail to find a fixpoint if we look inside them.

2.  Polymorphic 'seq' is represented by a (Seq _ _ []) demand,
so lub/both must be prepared to take a Seq with an empty list
of demands without complaint.
ghc/compiler/stranal/DmdAnal.lhs