[Bug c/114931] [14/15 regression] ICE in get_alias_set when building tcl with -std=c23

["rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114931

--- Comment #10 from Richard Biener <rguenth at gcc dot gnu.org> ---
(In reply to Jakub Jelinek from comment #9)
> Do you really need to hash it on TYPE_CANON, rather than just on whether the
> type is
> TYPE_STRUCTURAL_EQUALITY_P or not?

It's enough for the testcases here, I didn't try to convince myself it's safe
but of course it's better than what we do right now.  So we can try that as
well.

> I'm worried that always computing the TYPE_CANON before calling
> type_hash_canon is going to be quite expensive, while just figuring out if
> it should have TYPE_STRUCTURAL_EQUALITY_P or not should be cheaper.  If it
> should have
> TYPE_STRUCTURAL_EQUALITY_P, we'd SET_TYPE_STRUCTURAL_EQUALITY, otherwise set
> TYPE_CANONICAL (t) = t; (I think that is what we generally get from
> make_node), look up
> and recompute TYPE_CANONICAL if TYPE_CANONICAL (t) == t.

I'd like to make it obvious at least that we only alter TYPE_CANONICAL if
we were the one building 't' in the first place.  The layouting of types
also seems to be redundant now as that's done by type_hash_canon.

Luckily there are not too many callers.

It's a bit late for 14.1, so we can refactor things on trunk and then backport
later.