so is the problem in the code generator or the slowpopcount?


On Thu, Sep 5, 2013 at 2:00 PM, Muhaimin Ahsan <leroux@fezrev.com> wrote:
Here's the dumped asm: http://ghc.haskell.org/trac/ghc/attachment/ticket/7684/cgrun071.S

Muhaimin Ahsan

On Sep 5, 2013, at 12:06 PM, "GHC" <ghc-devs@haskell.org> wrote:

> #7684: cgrun071 segfaults
> ----------------------------------+----------------------------------
>        Reporter:  tibbe          |            Owner:  leroux
>            Type:  bug            |           Status:  new
>        Priority:  normal         |        Milestone:  7.8.1
>       Component:  Compiler       |          Version:  7.7
>      Resolution:                 |         Keywords:
> Operating System:  MacOS X        |     Architecture:  x86_64 (amd64)
> Type of failure:  Runtime crash  |       Difficulty:  Unknown
>       Test Case:  cgrun071       |       Blocked By:
>        Blocking:                 |  Related Tickets:
> ----------------------------------+----------------------------------
>
> Comment (by tibbe):
>
> I'm confused. The test isn't compiled with -msse42 and thus the popcnt
> instruction should never be omitted (check the output assembly to make
> sure) so it seems more likely that the failure has to do with narrowing.
>
> --
> Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7684#comment:20>
> GHC <http://www.haskell.org/ghc/>
> The Glasgow Haskell Compiler


_______________________________________________
ghc-devs mailing list
ghc-devs@haskell.org
http://www.haskell.org/mailman/listinfo/ghc-devs