> On Mon, Oct 01, 2001 at 10:52:32AM +0100, Simon Marlow wrote:
> > 
> > > > This is a known bug, but thanks for reporting it anyway.  
> > > The workaround
> > > > is as suggested, try compiling with -O.
> > > 
> > > I *was* compiling with -O. IIRC the problem occured with 
> any level of
> > > optimisation I tried if and only if profiling was enabled.
> > > 
> > > Any other suggestions?
> > 
> > You could try breaking up the structure into smaller pieces.  If you
> > have a local GHC build, you can also bump the BLOCK_SIZE to 8k in
> > ghc/rts/Constants.h.
> 
> Did you mean ghc/includes/Constants.h?

Yes.

> Changing
> #define BLOCK_SIZE   0x1000
> to
> #define BLOCK_SIZE   0x2000
> in there and rebuilding produced a build which produces code that
> segfaults.

Ah, I forgot to mention you need to adjust BLOCK_SHIFT accordingly (on
the next line).  Changing it from 12 to 13 should do the trick.

Cheers,
        Simon

_______________________________________________
Glasgow-haskell-bugs mailing list
[EMAIL PROTECTED]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs

Reply via email to