ghc 5.04.2 and seg fault

> Could be a problem with the recompilation checking.  We know of one
> such bug: if you compile first with -O, then later without -O,
> inconsistencies can arise. =20

No, I'm afraid its nothing so obvious. I can
rm *.o *.exe
build, run

and it runs fine.  However, if I alter one of the files and rebuild
exactly the same way, I get the segfault, so I make clean and build
from scratch each time.

> If your problem is not of this form, then it would help greatly if
> you could provide us with a repeatable example (indeed, it will
> probably be impossible to track down without one).

I'll work on it.  I've been hesitating about posting this bug report
for a few weeks because I couldn't come up with a small, repeatable
example.

> The next time it happens

At this point, it always happens.

> just tar up the tree and send it to us, and if you can provide a
> description of which changes were made recently to the tree that
> would help too.

I'd love to be able to do that.  Unfortunately I'm working on
proprietary code, and I cannot.  However as a big advocate of free
software, I find it refreshing that most people make that assumption
:)

So anyway, I'll work on coming up with a smaller version of the bug
that I can send you.  In the meantime, if anyone has any other ideas
of what I could do to isolate the bug, I'd be glad to try them.

peace,

isaac