[Git][ghc/ghc][wip/clyring/seqHash-DmdTransformer] Why is seq# absent in its own defining module???

[Matthew Craven (@clyring)]

Matthew Craven pushed to branch wip/clyring/seqHash-DmdTransformer at Glasgow Haskell Compiler / GHC


Commits:
be5dfb89 by Matthew Craven at 2024-04-22T14:46:45-04:00
Why is seq# absent in its own defining module???

- - - - -


1 changed file:

- compiler/GHC/Core/Opt/DmdAnal.hs


Changes:

=====================================
compiler/GHC/Core/Opt/DmdAnal.hs
=====================================
@@ -1025,7 +1025,10 @@ dmdTransform env var sd
   = -- pprTrace "dmdTransform:DictSel" (ppr var $$ ppr (idDmdSig var) $$ ppr sd) $
     dmdTransformDictSelSig (idDmdSig var) sd
   | var `hasKey` seqHashKey
-  = dmdTransformSeqHash sd
+  = if isGlobalId var
+      then dmdTransformSeqHash sd
+      else -- stupid hack to make sure seq# isn't found absent in its own module
+           addVarDmd (dmdTransformSeqHash sd) var topDmd
   -- Imported functions
   | isGlobalId var
   , let res = dmdTransformSig (idDmdSig var) sd



View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/be5dfb893fa82043bc56443cbc457ea886719e42

-- 
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/be5dfb893fa82043bc56443cbc457ea886719e42
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/20240422/9666e595/attachment.html>