Hi people, I just got a FTBFS bug report for my haskell-edison package [1] with the title: "haskell-edison: FTBFS: multiple definition of `EdisonAPIzm1zi2zi1_DataziEdisonziPrelude_zdp1Measured_closure'"
I see that there a couple of Haskell packages which also have FTBFS bug
reports with the same problem (which bug reports I have CC'ed). One of
these packages is ghc6 and that bugreport [2] tells me the problem is
caused by the 'new' gcc-4.2 we have in unstable and testing.
The upstream bug report [3] says something about gcc 4.2 needing the
-fno-toplevel-reorder flag and that the bug is fixed for ghc in version
6.8:
> status changed from new to closed.
> resolution set to fixed.
> Fixed:
>
> Tue Aug 21 16:15:53 BST 2007 Simon Marlow <[EMAIL PROTECTED]>
> * FIX #1427, #1569: gcc 4.2.x needs -fno-toplevel-reorder
I tried to fix the FTBFS of haskell-edison by explicitly adding the
-fno-toplevel-reorder flag to the call to cc (ghc ... -opt-c
-fno-toplevel-reorder) but this had no effect.
After I also stopped using the --enable-split-objs ghc flag, I could
compile haskell-edison again.
I have a couple of questions though which you might be able to awnser:
* I assume this workaround is only needed until ghc 6.8 to enters
the archive?
* Also in the upstream bug report I see nothing about
--enable-split-objs. Is this tested with cvs version of ghc?
Greetings Arjan
[1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=445754
[2] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=428060
[3] http://hackage.haskell.org/trac/ghc/ticket/1427
signature.asc
Description: Dit berichtdeel is digitaal ondertekend
_______________________________________________ debian-haskell mailing list [email protected] http://urchin.earth.li/mailman/listinfo/debian-haskell
