diff options
| -rw-r--r-- | compiler/stranal/DmdAnal.lhs | 21 |
1 files changed, 21 insertions, 0 deletions
diff --git a/compiler/stranal/DmdAnal.lhs b/compiler/stranal/DmdAnal.lhs index adda04156f..5140eb89a6 100644 --- a/compiler/stranal/DmdAnal.lhs +++ b/compiler/stranal/DmdAnal.lhs @@ -866,6 +866,27 @@ iterate twice to be sure of not getting over-optimistic CPR info, in the case where t turns out to be not-demanded. This is handled by dmdAnalTopBind. +Note [On the nature of call demands] +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +In a signature environment P, variables f are associated with a usage +signature <k; t; \phi>, where k is the arity of f. Why does t not +have enough information to make k redundant? + +Consider an example: + +h = \f -> \g -> case y of True -> \a -> f (a 1) + False -> \b -> b (g 2) + +For h the analysis produces the following demand type (handling of +case-expressions is defined in S3.8): + +1*C1(U) -> 1*C1(U) -> 1*C1(U) -> * + +At the call site (h f1 g1) having two arguments provided (k = 2) is +already enough to unleash the inferred demand type. Had we taken k = +3, as suggested by the demand type, we would not be able to treat this +case with a non-trivial demand. Note [NOINLINE and strictness] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |