Marge Bot pushed to branch master at Glasgow Haskell Compiler / GHC
Commits:
ec263a59 by Sebastian Graf at 2023-03-10T23:25:25-05:00
Simplify: Move `wantEtaExpansion` before expensive `do_eta_expand` check
There is no need to run arity analysis and what not if we are not in a
Simplifier phase that eta-expands or if we don't want to eta-expand the
expression in the first place.
Purely a refactoring with the goal of improving compiler perf.
- - - - -
1 changed file:
- compiler/GHC/Core/Opt/Simplify/Utils.hs
Changes:
=====================================
compiler/GHC/Core/Opt/Simplify/Utils.hs
=====================================
@@ -1859,11 +1859,12 @@ tryEtaExpandRhs :: SimplEnv -> BindContext -> OutId -> OutExpr
-> SimplM (ArityType, OutExpr)
-- See Note [Eta-expanding at let bindings]
tryEtaExpandRhs env bind_cxt bndr rhs
- | do_eta_expand -- If the current manifest arity isn't enough
- -- (never true for join points)
- , seEtaExpand env -- and eta-expansion is on
- , wantEtaExpansion rhs
- = -- Do eta-expansion.
+ | seEtaExpand env -- If Eta-expansion is on
+ , wantEtaExpansion rhs -- and we'd like to eta-expand e
+ , do_eta_expand -- and e's manifest arity is lower than
+ -- what it could be
+ -- (never true for join points)
+ = -- Do eta-expansion.
assertPpr( not (isJoinBC bind_cxt) ) (ppr bndr) $
-- assert: this never happens for join points; see GHC.Core.Opt.Arity
-- Note [Do not eta-expand join points]
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/ec263a59b886ea616dabce349df7a377d5356dd5
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/ec263a59b886ea616dabce349df7a377d5356dd5
You're receiving this email because of your account on gitlab.haskell.org.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/ghc-commits/attachments/20230310/ee16f52a/attachment-0001.html>