bug#34655: 26.1.92; Segfault in module with --module-assertions

[Philipp Stephani]

Am Do., 21. März 2019 um 18:00 Uhr schrieb Eli Zaretskii <eliz@gnu.org>:
>
> > From: Philipp Stephani <p.stephani2@gmail.com>
> > Date: Thu, 21 Mar 2019 17:11:41 +0100
> > Cc: "Basil L. Contovounesios" <contovob@tcd.ie>, 34655@debbugs.gnu.org
> >
> > I haven't checked everything in detail, but my impression is that this
> > is rather another instance of bug#31238. Fixing this only when module
> > assertions are enabled will probably not fix anything, but rather mask
> > issues. Reverting commit 3eb93c07f7a60ac9ce8a16f10c3afd5a3a31243a is
> > still the right approach here. Can you please hold off a bit? I've
> > almost completed the revert, but haven't pushed it yet. Once that's in
> > we can check whether it also fixes this issue.
>
> I will CC Stefan, who committed 3eb93c07f7a60ac9ce8a16f10c3afd5a3a31243a.
>
> I'm not sure we should revert that; we could instead add GC protection
> for those parts that need it.

Yes, that's what reverting that commit does :-)
We need to mark the objects in all cases, not just when module
assertions are enabled.
Note that both the designer of the module API (Daniel) and I as one of
its main implementers disagree with commit
3eb93c07f7a60ac9ce8a16f10c3afd5a3a31243a. I'm happy to discuss
alternatives, but for now we should revert it and discuss the
alternatives *before* implementing them. I've already confirmed that
reverting commit 3eb93c07f7a60ac9ce8a16f10c3afd5a3a31243a fixes
bug#31238, and I can try it with this bug as well.