[Git][ghc/ghc][wip/T22272] Apply 2 suggestion(s) to 1 file(s)
Matthew Pickering (@mpickering)
gitlab at gitlab.haskell.org
Mon Dec 12 12:58:40 UTC 2022
Matthew Pickering pushed to branch wip/T22272 at Glasgow Haskell Compiler / GHC
Commits:
0071f57e by sheaf at 2022-12-12T12:58:38+00:00
Apply 2 suggestion(s) to 1 file(s)
- - - - -
1 changed file:
- compiler/GHC/IfaceToCore.hs
Changes:
=====================================
compiler/GHC/IfaceToCore.hs
=====================================
@@ -1821,7 +1821,7 @@ variables in `e`. Now we can see the problem: to set the `uf_is_value`
field of `foo`'s unfolding, we look at its unfolding (in this case
just `foo` itself!). Loop. This is the root cause of ticket #22272.
-The simple solution we chose is to serialise the various auxilary
+The simple solution we chose is to serialise the various auxiliary
fields of `CoreUnfolding` so that we don't need to recreate them when
rehydrating. Specifically, the following fields are moved to the
'UnfoldingCache', which is persisted in the interface file:
@@ -1835,7 +1835,7 @@ These four bits make the interface files only one byte larger per
unfolding; on the other hand, this does save calls to 'exprIsValue',
'exprIsExpandable' etc for every imported Id.
-We could choose to do this only for loop breakers. But thats a bit
+We could choose to do this only for loop breakers. But that's a bit
more complicated and it seems good all round.
-}
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/0071f57e9d0427eb540f36f577fc2926a61fe7d2
--
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/0071f57e9d0427eb540f36f577fc2926a61fe7d2
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/20221212/b7999de4/attachment.html>
More information about the ghc-commits
mailing list