Commit 64bfc011 authored by simonpj@microsoft.com's avatar simonpj@microsoft.com
Browse files

Wibble to the OccurAnal fix for RULEs and loop-breakers

parent eea143f8
......@@ -221,7 +221,7 @@ However things are made quite a bit more complicated by RULES. Remember
Remmber that we simplify the RULES before any RHS (see Note
[Rules are visible in their own rec group] above).
So we must *not* postInlineUnconditinoally 'g', even though
So we must *not* postInlineUnconditionally 'g', even though
its RHS turns out to be trivial. (I'm assuming that 'g' is
not choosen as a loop breaker.)
......@@ -235,7 +235,7 @@ However things are made quite a bit more complicated by RULES. Remember
other yes yes
The **sole** reason for this kind of loop breaker is so that
postInlineUnconditioanlly does not fire. Ugh.
postInlineUnconditionally does not fire. Ugh.
* Note [Rule dependency info]
~~~~~~~~~~~~~~~~~~~~~~~~~~~
......@@ -307,10 +307,14 @@ occAnalBind env (Rec pairs) body_usage
= body_usage +++ addRuleUsage rhs_usage bndr
(final_usage, tagged_bndrs) = tagBinders total_usage bndrs
final_bndrs | no_rules = tagged_bndrs
final_bndrs | isEmptyVarSet all_rule_fvs = tagged_bndrs
| otherwise = map tag_rule_var tagged_bndrs
tag_rule_var bndr | bndr `elemVarSet` all_rule_fvs = makeLoopBreaker True bndr
| otherwise = bndr
all_rule_fvs = bndr_set `intersectVarSet` foldr (unionVarSet . idRuleVars) emptyVarSet bndrs
-- Mark the binder with OccInfo saying "no preInlineUnconditionally" if
-- it is used in any rule (lhs or rhs) of the recursive group
---- stuff for dependency analysis of binds -------------------------------
sccs :: [SCC (Node Details)]
......@@ -353,7 +357,6 @@ occAnalBind env (Rec pairs) body_usage
rule_fv_env = rule_loop init_rule_fvs
no_rules = null init_rule_fvs
all_rule_fvs = foldr (unionVarSet . snd) emptyVarSet init_rule_fvs
init_rule_fvs = [(b, rule_fvs)
| b <- bndrs
, let rule_fvs = idRuleRhsVars b `intersectVarSet` bndr_set
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment