Thanks, this works!

I guess this is the kind of problems I run into by using `mkIface_`
directly instead of via `mkIfaceTc`. Unfortunately, if I try that, I end up
with a panic in `GHC.Iface.Recomp.mkHashFun` in my real program (see
separate email earlier).

On Tue, Jul 27, 2021, 16:14 Sylvain Henry <[email protected]> wrote:

> > What am I doing wrong? Is filling the `mi_fixities` field of the
> `ModIface` not enough to let importers see the correct fixities?
>
> It seems like the renamer is looking for the fixities via `mi_fix_fn
> (mi_final_exts iface)`, not `mi_fixities`.
>
> You should try to replace:
>
>   , mi_final_exts = mi_final_exts empty
>
> with:
>
>   , mi_final_exts = (mi_final_exts empty){ mi_fix_fn = mkIfaceFixCache
> (mi_fixities partial)
>
>
> _______________________________________________
> ghc-devs mailing list
> [email protected]
> http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs
>
_______________________________________________
ghc-devs mailing list
[email protected]
http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs

Reply via email to