Segmentation fault during build of 5.04.3

[Ed Osinski]

On Tue, Apr 22, 2003 at 03:32:10PM +0100, Simon Marlow wrote:
>  
> > * I'm using gcc 3.2.2 (which I've built lots of other software with
> >   successfully).
> 
> That gives me an idea: try recompiling genprimopcode with -fasm and see
> if that solves the crashes:
> 
>   cd ghc/utils/genprimopcode
>   make clean
>   make EXTRA_HC_OPTS=-fasm

Yes, this indeed takes care of the problem with genprimopcode.  I
tried "make all EXTRA_HC_OPTS=-fasm" at the top level, and reissued
compilations manually for modules that required -fvia-C.
Unfortunately, I eventually got:

------------------------------------------------------------------------
==fptools== make way=p all;
PWD = /systems/sware/src/build/ghc-5.04.3-build/libraries/base
------------------------------------------------------------------------
rm -f GHC/Base.p_o; if [ ! -d GHC/Base_split ]; then mkdir GHC/Base_split; else /usr/bin/find GHC/Base_split -name '*.p_o' | xargs rm -f __rm_food; fi;   
../../ghc/compiler/ghc-inplace -ldl -fglasgow-exts -cpp -Iinclude -funbox-strict-fields -package-name base -O -Rghc-timing  -split-objs -hisuf p_hi -hcsuf p_hc -osuf p_o -prof  -fasm -optl-R/usr/local/lib -c GHC/Base.lhs -o GHC/Base.p_o
ghc-5.04.3: panic! (the `impossible' happened, GHC version 5.04.3):
        AbsCStixGen.gencode
    CC_DECLARE(GHCziBase_CAFs_cc,"CAF","GHC.Base",CC_IS_CAF,);

Please report it as a compiler bug to glasgow-haskell-bugs at haskell.org,
or http://sourceforge.net/projects/ghc/.


<<ghc: 96890036 bytes, 26 GCs, 876009/1533616 avg/max bytes residency (3 samples), 9M in use, 0.02 INIT (0.99 elapsed), 3.89 MUT (4.84 elapsed), 1.05 GC (3.47 elapsed) :ghc>>
make[3]: *** [GHC/Base.p_o] Error 1
make[2]: *** [all] Error 1
make[1]: *** [all] Error 1
make[1]: Leaving directory `/systems/sware/src/build/ghc-5.04.3-build/libraries'
make: *** [all] Error 1

Has anyone compiled ghc 5.04.3 successfully on a Sun running Solaris
8, with gcc?  If so, which version of gcc?

- Ed