[Git][ghc/ghc][wip/T25657] More updates

[Simon Peyton Jones (@simonpj)]

Simon Peyton Jones pushed to branch wip/T25657 at Glasgow Haskell Compiler / GHC


Commits:
14118210 by Simon Peyton Jones at 2025-03-01T11:01:12+00:00
More updates

- - - - -


2 changed files:

- compiler/GHC/Core/InstEnv.hs
- compiler/GHC/Utils/Panic.hs


Changes:

=====================================
compiler/GHC/Core/InstEnv.hs
=====================================
@@ -1163,7 +1163,10 @@ instEnvMatchesAndUnifiers (InstEnv rm) vis_mods cls tys
                 -- Unification will break badly if the variables overlap
                 -- They shouldn't because we allocate separate uniques for them
                 -- See Note [Template tyvars are fresh]
-        case tcUnifyTysFG dontCareBindFam instanceBindFun tpl_tys tys of
+        case tcUnifyTysFG alwaysBindFam instanceBindFun tpl_tys tys of
+          -- alwaysBindFam: the family-application can't be in the instance head,
+          -- but it certainly can be in the Wanted constraint we are matching!
+          --
           -- We consider MaybeApart to be a case where the instance might
           -- apply in the future. This covers an instance like C Int and
           -- a target like [W] C (F a), where F is a type family.


=====================================
compiler/GHC/Utils/Panic.hs
=====================================
@@ -191,7 +191,7 @@ pprPanic :: HasCallStack => String -> SDoc -> a
 pprPanic s doc = panicDoc s (doc $$ callStackDoc)
 
 -- | Throw an exception saying "bug in GHC"
-panicDoc :: String -> SDoc -> a
+panicDoc :: HasCallStack => String -> SDoc -> a
 panicDoc x doc = throwGhcException (PprPanic x doc)
 
 -- | Throw an exception saying "this isn't finished yet"



View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/14118210a0d812b1ad8511feb7919d9dc270a0ef

-- 
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/commit/14118210a0d812b1ad8511feb7919d9dc270a0ef
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/20250301/37c07631/attachment-0001.html>