RE: Help: Segfault | Where is the *.hc for 4.02 ?

[Jose Romildo Malaquias]

I am also having segmentation faults with ghc on a RedHat Linux 6.0
(glibc-2.1.1) machine on every Haskell program I try. I have tried
both the binary distsribution (version 4.02) prepared by the GHC
team and the binary RPMS provided by Sven Panne at
ftp://ftp.informatik.uni-muenchen.de/pub/local/pms/ghc-4.03-38.i386.rpm

My next attempt would be recompile myself from sources so the binaries
could use the newest libraries from my system. Where is the .hc files
for 4.03 or 4.02? The most recent .hc files I could locate were for
4.00. So I am going to try compiling 4.00 and them 4.03.

Prof. Jose Romildo Malaquias <[EMAIL PROTECTED]>
-----------------------------------------------------------------------
> > I've tried `ghc-4.02-i386-unknown-linux.tar.gz' in-place, but even
> > "hello world" will dump core.
> > 
> > Some info about my box:
> > 
> > glibc-2.1.1
> 
> I think the conclusion is that recent glibcs don't work with ghc.
> Unfortunately we don't have a box here to test on, so if anyone can help
> narrow down the problem and/or mail those responsible we'd be grateful.
> 
> > BTW, I've tried to give ghc `-optc-g', but all ghc-4.02 versions which
> > I have, include those being compiled by myself, cause `ld' to dump
> > core, even with glibc-2.0.7 and gcc 2.7.2.3 :-(
> 
> We don't support/reccommend compiling with -optc-g :-)  The reason is that
> ghc  post-processes the assembly output from gcc, and the script simply
> doesn't understand the debugging info inserted by -g.  Nevertheless, ld
> shouldn't dump core.
> 
> As per the well-known joke, "Don't do that then!"
> 
> Simon