Re: [MiNT] Some mintlib patches

[Alan Hourihane <alanh@fairlite.demon.co.uk>]

On Wed, 2008-02-20 at 18:21 -0500, Keith Scroggins wrote:
> >>> Hi,
> >>>
> >>> here are is a patch for mintlib configuration I use, I think it's useful
> >>> for others, too:
> >>>
> >>> - set premilitary stack to 64KB (instead of 1K) which makes gcc linking
> >>> happy again
> >>
> >> I'd really like to know what the real problem is here, this just seems
> >> like another bandage for a real problem.
> >>
> >> For me I've actually reverted your original patch to crt0.S and I don't
> >> see a problem. So it'd be worth finding out if you reverted the file
> >> back to the original and find out why gcc is crashing (if it really
> >> still does).
> >
> > Oh, and if the problem is 100% reproducible, can you outline the exact
> > steps you take to replication so others can take a look ?
> 
> Hello Alan,
> 
> I believe the main problem is that during the bootstrap process of GCC 
> natively, the stack of the newly created compiler is not set properly in a 
> fashion for it to run the whole way thru (stage 2), so the stack size is 
> not enough on cc1.  If you modify the stack size after the crash, GCC will 
> in turn relink the cc1 executable, and then crash again.  So, I believe 
> this was Miro's solution to that.

Mmmm. But the new mintlib's have their default stack size set to 64K
anyway now.

> My solution was to modify the stack size while the process was still 
> running.  This is only a problem in the bootstrap process as far as I know 
> and it could possibly be a problem with compiled programs that depend on 
> building little pieces of software to generate source files.

I'm confused why Miro is still changing crt0.S ?

Alan.