Ok thanks! It's not stopping me from anything so it is not super urgent for me.
d-


> -----Original Message-----
> From: Manuel M T Chakravarty [mailto:[email protected]]
> Sent: 16 March 2012 06:17
> To: Dimitrios Vytiniotis
> Cc: [email protected]
> Subject: Re: head panic when building dph?
> 
> Ok, it only happens when you compile the stage 2 compiler with -DDEBUG.
> 
> I think I know what the problem is.  I'll fix it, but -in the meantime- it
> shouldn't affect anyone unless you use -DDEBUG for the stage 2 compiler
> *and* use that compiler to compile the DPH libraries.  (In particular, it 
> doesn't
> affect validation.)
> 
> Manuel
> 
> Dimitrios Vytiniotis:
> > I did nothing special, just  'make'  (not validate if that matters)
> > with the default devel2 flavor in build.mk
> >
> >> -----Original Message-----
> >> From: Manuel M T Chakravarty [mailto:[email protected]]
> >> Sent: Thursday, March 15, 2012 12:52 PM
> >> To: Dimitrios Vytiniotis
> >> Cc: Ben Lippmeier; [email protected]
> >> Subject: Re: head panic when building dph?
> >>
> >> Dimitrios,
> >>
> >> How did you compile?  I just ran validate on the HEAD (on OS X) and
> >> it works just fine including DPH.  I merely get the following failures:
> >>
> >>> Unexpected failures:
> >>>   ffi/should_run    capi_value [exit code non-0] (normal)
> >>>   perf/compiler     T4801 [stat too good] (normal)
> >>>   safeHaskell/ghci  p5 [bad stdout] (ghci)
> >>>   safeHaskell/ghci  p7 [bad stdout] (ghci)
> >>>   safeHaskell/ghci  p8 [bad stdout] (ghci)
> >>>
> >>
> >> Manuel
> >>
> >> Dimitrios Vytiniotis:
> >>> Hmm ok I will build w/o dph for now. FYI someone seems to be calling
> >>> vectScalarFun on an expression which returns a superclass (I've
> >>> added
> >> some tracing in my local copy). Weird.
> >>>
> >>> vectScalarFun:
> >>> ghc-prim:GHC.Classes.$p1Ord{v r9v} [gid[ClassOp]]
> >>>   @ ghc-prim:GHC.Types.Int{(w) tc 3J}
> >>>   ghc-prim:GHC.Classes.$fOrdInt{v r3O} [gid[DFunId]]
> >>>   :: ghc-prim:GHC.Classes.Eq{tc 23}
> >>>        ghc-prim:GHC.Types.Int{(w) tc 3J}
> >>> ghc-stage2: panic! (the 'impossible' happened)  (GHC version 7.5 for
> >>> x86_64-unknown-linux):
> >>>       ASSERT failed! file compiler/vectorise/Vectorise/Exp.hs, line
> >>> 335
> >>>
> >>> thanks!
> >>> d-
> >>>
> >>>
> >>>
> >>>
> >>>> -----Original Message-----
> >>>> From: Ben Lippmeier [mailto:[email protected]]
> >>>> Sent: 14 March 2012 03:39
> >>>> To: Dimitrios Vytiniotis
> >>>> Cc: [email protected]
> >>>> Subject: Re: head panic when building dph?
> >>>>
> >>>>
> >>>> On 14/03/2012, at 2:24 AM, Dimitrios Vytiniotis wrote:
> >>>>
> >>>>>
> >>>>> Hi all, just checked out a fresh copy of HEAD and got the following:
> >>>>>
> >>>>> "inplace/bin/ghc-stage2"   -H64m -O -fasm    -package-name dph-
> lifted-
> >>>> vseg-0.6.0.1 -hide-all-packages -i -ilibraries/dph/dph-lifted-
> >>>> vseg/./Data/Array/Parallel/Prelude/Int.hs -o
> >>>> libraries/dph/dph-lifted-
> >>>> vseg/dist-install/build/Data/Array/Parallel/Prelude/Int.o
> >>>>> ghc-stage2: panic! (the 'impossible' happened)  (GHC version 7.5
> >>>>> for
> >>>>> x86_64-unknown-linux):
> >>>>>      ASSERT failed! file compiler/vectorise/Vectorise/Exp.hs, line
> >>>>> 334
> >>>>>
> >>>>> Please report this as a GHC bug:
> >>>>> http://www.haskell.org/ghc/reportabug
> >>>>>
> >>>>> make[1]: ***
> >>>>> [libraries/dph/dph-lifted-vseg/dist-install/build/Data/Array/Paral
> >>>>> le
> >>>>> l/
> >>>>> Prelude/Int.o] Error 1
> >>>>> make: *** [all] Error 2
> >>>>> dimitris@cam-04-unx:/64playpen/dimitris/GHC-HEAD/ghc$
> >>>>>
> >>>>> I am just building with flavor=devel2 and no other special options
> >>>>> on a Linux
> >>>> 64-bit machine.
> >>>>> Any ideas?
> >>>>
> >>>> I'm getting the same thing but I don't know what's wrong yet. There
> >>>> haven't been any changes to the DPH library for a couple of weeks
> >>>> -- I pushed a patch on the weekend but this was to dph-examples
> >>>> which is ignored by the build system.
> >>>>
> >>>> Ben.
> >>>>
> >>>
> >>>
> >>>
> >>> _______________________________________________
> >>> Cvs-ghc mailing list
> >>> [email protected]
> >>> http://www.haskell.org/mailman/listinfo/cvs-ghc
> >>
> >
> >
> >
> > _______________________________________________
> > Cvs-ghc mailing list
> > [email protected]
> > http://www.haskell.org/mailman/listinfo/cvs-ghc
> 



_______________________________________________
Cvs-ghc mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/cvs-ghc

Reply via email to