Feature development

Matthew Pickering matthewtpickering at gmail.com
Tue Apr 5 18:23:17 UTC 2016


Seems like this issue - https://ghc.haskell.org/trac/ghc/ticket/11155

I also build on OS X and haven't had any problems recently fwiw.

On Tue, Apr 5, 2016 at 6:58 PM, Sazanovich Nikita
<nikitasazanovich at gmail.com> wrote:
> Thanks for your reply, Matthew.
>
> I’ve tried to merge my local repository with ghc/master branch and after
> this I couldn't build it (even checked out to a successful commit). I am
> getting  this kind of error.
>
> Undefined symbols for architecture x86_64:
> "_stg_sel_17_upd_info", referenced from:
> _s1yAM_info in UI.dyn_o
> _s1yC7_info in UI.dyn_o
> _s1yKD_info in UI.dyn_o
> _s1yLa_info in UI.dyn_o
>
> It seems like something specific to OS X. Can you help me?
>
> On 05 Apr 2016, at 2:25 AM, Matthew Pickering <matthewtpickering at gmail.com>
> wrote:
>
> Hello Nikita,
>
> I don't know the answers to your questions but I do know that it's
> easier to give help if there is some code to look at. I would suggest
> you put a diff up on phab with what you have done so far so that it is
> easy for others to give specific comments about where you are stuck.
>
> Matt
>
> On Mon, Apr 4, 2016 at 7:22 PM, Sazanovich Nikita
> <nikitasazanovich at gmail.com> wrote:
>
>  Hello!
>  I am working on ticket https://ghc.haskell.org/trac/ghc/ticket/5850 now.
> I’ve come upon converting the type, which GHC.compileExpr returns, to my
> prompt function type [PModule] -> Int -> IO String. But this function
> returns ghc-prim-0.5.0.0:GHC.Prim.Any. I don’t see a clear way to convert
> it. So my questions are:
> - What this type is about? I know Any type in Haskell. Is it a specific
> version of it?
> - Or maybe it is not the best way to check a function correctness? I’ve
> checked out InteractiveEval module and it seems that there are only four
> quite similar functions to compile Expressions.
>  Last person, who was trying to solve this problem, used Unsafe.Coerce and
> it is also doing this job pretty well now, but the name speaks for itself.
>  Thank you.
>
> _______________________________________________
> ghc-devs mailing list
> ghc-devs at haskell.org
> http://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-devs
>
>


More information about the ghc-devs mailing list