Re: ghc 4.02 compilation failure (linux i386 glibc)

[Giuliano P Procida]

Hi.

On Tue, May 18, 1999 at 10:10:20AM -0700, Sigbjorn Finne (Intl Vendor) wrote:
> 
> Giuliano P Procida <[EMAIL PROTECTED]> writes: 
> > Easier said than done... it was a while before I realised 
> > that removing the fflush and typing make just wasn't going to
> > hack it. This is because hsc is linked against the installed
> > libHSrts (argh! scream!) rather than the one just re-built.
> > The build remakes the fresh rts lib and re-links against the
> > old one -- almost certainly not what is intended.
> > 
> 
> That's the way it should be, really - how do you expect to successfully
> combine 4.03's RTS with ghc/compiler/ sources that has been compiled with
> 3.02 (say)? The RTS and the compiled code needs to agree 100% on the
> layout of heap objects, for example. If they don't, well, then you get
> crashes such as the one you're now experiencing with 'hsc' ;-)

So the build goes like:

1. Build new RTS (not yet used) using C compiler.
2. Build new hsc (dependent on old RTS) using old hsc.
3. New hsc produces newer hsc which depends on new RTS.
4. Newer hsc and new RTS are now usable together.

It would be less confusing of the RTS was built after the "new hsc".

Right? Unless the new and old RTS happen to remain unchanged between
releases.

And there is no way for compiler X to produce output for RTS Y where X
and Y are different? I imagine cross-compilation is just not there at
all then.

I believe I am compiling 4.02 with 4.02, so all the above shouldn't
apply. This suggests something is properly broken (probably libc).

Giuliano.