[Git][ghc/ghc][wip/romes/isNullaryRepDataCon] Check DataCon wrapper's LFInfo rather than its arity

[Matthew Craven (@clyring)]

Matthew Craven pushed to branch wip/romes/isNullaryRepDataCon at Glasgow Haskell Compiler / GHC


Commits:
869faa23 by Matthew Craven at 2024-02-23T13:06:06-05:00
Check DataCon wrapper's LFInfo rather than its arity

...avoid questions about whether arity 0 could ever mean "unknown"

- - - - -


1 changed file:

- compiler/GHC/StgToCmm/DataCon.hs


Changes:

=====================================
compiler/GHC/StgToCmm/DataCon.hs
=====================================
@@ -404,7 +404,9 @@ precomputedStaticConInfo_maybe cfg binder con []
   | isNullaryRepDataCon con
   = use_name (dataConName con)
   | Just wrapper <- dataConWrapId_maybe con
-  , idArity wrapper == 0 -- This might be pre-unarise arity, but that's OK
+  -- check the wrapper's LFInfo to avoid wondering whether an arity of
+  -- zero could possibly mean "unknown arity" for a DataCon wrapper
+  , Just (LFCon {}) <- idLFInfo_maybe wrapper
   = use_name (varName wrapper)
   where
     use_name name



View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/869faa23120af52ee08e49bf60e39b7e3f71cd54

-- 
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/869faa23120af52ee08e49bf60e39b7e3f71cd54
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/20240223/e690e4d2/attachment.html>