The Glasgow Haskell Users Mailing List

Dominic Steinitz | 26 Dec 2012 19:14

Is there a workaround for this bug?

AFAICT this bug fix http://hackage.haskell.org/trac/ghc/ticket/7239 did not make it into 7.6.1. Also I
am happily working on the Haskell Platform with 7.4.1 and I'd rather avoid upgrading if possible.

Is there a workaround? I've attached my code below along with the error message (which is the same as in the
above bug report). I'm rather hoping I won't have to build HEAD.

Thanks, Dominic.

bash-3.2$ ghc -fext-core --make Test.hs
[1 of 1] Compiling Main             ( Test.hs, Test.o )
ghc: panic! (the 'impossible' happened)
 (GHC version 7.4.1 for x86_64-apple-darwin):
	MkExternalCore died: make_lit

{-# LANGUAGE FlexibleContexts #-}

{-# OPTIONS_GHC -Wall -fno-warn-name-shadowing -fno-warn-type-defaults #-}

import Data.Array.Repa as Repa
import Data.Array.Repa.Eval
import Control.Monad

r, sigma, k, t, xMax, deltaX, deltaT :: Double
m, n :: Int
r = 0.05
sigma = 0.2
k = 50.0
t = 3.0
m = 80
xMax = 150

(Continue reading)

headers

Simon Peyton-Jones | 1 Jan 2013 20:48

RE: Is there a workaround for this bug?

I think the patch did get into 7.6.2 (which is about to be released) though.

I don't think there's a workaround, except by not using External Core, or not using Integer literals (use
Ints?).  Sorry.

Simon

|  -----Original Message-----
|  From: glasgow-haskell-users-bounces <at> haskell.org [mailto:glasgow-haskell-users-
|  bounces <at> haskell.org] On Behalf Of Dominic Steinitz
|  Sent: 26 December 2012 18:14
|  To: glasgow-haskell-users <at> haskell.org
|  Subject: Is there a workaround for this bug?
|  
|  AFAICT this bug fix http://hackage.haskell.org/trac/ghc/ticket/7239 did not make
|  it into 7.6.1. Also I am happily working on the Haskell Platform with 7.4.1 and I'd
|  rather avoid upgrading if possible.
|  
|  Is there a workaround? I've attached my code below along with the error message
|  (which is the same as in the above bug report). I'm rather hoping I won't have to
|  build HEAD.
|  
|  Thanks, Dominic.
|  
|  bash-3.2$ ghc -fext-core --make Test.hs
|  [1 of 1] Compiling Main             ( Test.hs, Test.o )
|  ghc: panic! (the 'impossible' happened)
|   (GHC version 7.4.1 for x86_64-apple-darwin):
|  	MkExternalCore died: make_lit
|

(Continue reading)

headers

Dominic Steinitz | 1 Jan 2013 21:30

Re: Is there a workaround for this bug?

Thanks - I'll probably wait for the next release.

On 1 Jan 2013, at 19:48, Simon Peyton-Jones <simonpj <at> microsoft.com> wrote:

> I think the patch did get into 7.6.2 (which is about to be released) though.
> 
> I don't think there's a workaround, except by not using External Core, or not using Integer literals (use
Ints?).  Sorry.
> 
> Simon
> 
> |  -----Original Message-----
> |  From: glasgow-haskell-users-bounces <at> haskell.org [mailto:glasgow-haskell-users-
> |  bounces <at> haskell.org] On Behalf Of Dominic Steinitz
> |  Sent: 26 December 2012 18:14
> |  To: glasgow-haskell-users <at> haskell.org
> |  Subject: Is there a workaround for this bug?
> |  
> |  AFAICT this bug fix http://hackage.haskell.org/trac/ghc/ticket/7239 did not make
> |  it into 7.6.1. Also I am happily working on the Haskell Platform with 7.4.1 and I'd
> |  rather avoid upgrading if possible.
> |  
> |  Is there a workaround? I've attached my code below along with the error message
> |  (which is the same as in the above bug report). I'm rather hoping I won't have to
> |  build HEAD.
> |  
> |  Thanks, Dominic.
> |  
> |  bash-3.2$ ghc -fext-core --make Test.hs
> |  [1 of 1] Compiling Main             ( Test.hs, Test.o )

(Continue reading)

headers

Roman Cheplyaka | 1 Jan 2013 21:51

Re: Is there a workaround for this bug?

Hi Dom,

I can confirm that your example compiles (with minimal adjustments)
under GHC 7.6.2 RC1. You can get it here:

http://www.haskell.org/ghc/dist/7.6.2-rc1/

Roman

* Dominic Steinitz <dominic <at> steinitz.org> [2013-01-01 20:30:10+0000]
> Thanks - I'll probably wait for the next release.
> 
> On 1 Jan 2013, at 19:48, Simon Peyton-Jones <simonpj <at> microsoft.com> wrote:
> 
> > I think the patch did get into 7.6.2 (which is about to be released) though.
> > 
> > I don't think there's a workaround, except by not using External Core, or not using Integer literals (use
Ints?).  Sorry.
> > 
> > Simon
> > 
> > |  -----Original Message-----
> > |  From: glasgow-haskell-users-bounces <at> haskell.org [mailto:glasgow-haskell-users-
> > |  bounces <at> haskell.org] On Behalf Of Dominic Steinitz
> > |  Sent: 26 December 2012 18:14
> > |  To: glasgow-haskell-users <at> haskell.org
> > |  Subject: Is there a workaround for this bug?
> > |  
> > |  AFAICT this bug fix http://hackage.haskell.org/trac/ghc/ticket/7239 did not make
> > |  it into 7.6.1. Also I am happily working on the Haskell Platform with 7.4.1 and I'd

(Continue reading)