Re: [PATCH] tree-optimization/106593 - fix ICE with backward threading

[Aldy Hernandez <aldyh@redhat.com>]

On Fri, Aug 12, 2022 at 1:36 PM Richard Biener <rguenther@suse.de> wrote:
>
> On Fri, 12 Aug 2022, Aldy Hernandez wrote:
>
> > On Fri, Aug 12, 2022 at 12:59 PM Richard Biener <rguenther@suse.de> wrote:
> > >
> > > With the last re-org I failed to make sure to not add SSA names
> > > nor supported by ranger into m_imports which then triggers an
> > > ICE in range_on_path_entry because range_of_expr returns false.  I've
> > > noticed that range_on_path_entry does mightly complicated things
> > > that don't make sense to me and the commentary might just be
> > > out of date.  For the sake of it I replaced it with range_on_entry
> > > and statistics show we thread _more_ jumps with that, so better
> > > not do magic there.
> >
> > Hang on, hang on.  range_on_path_entry was written that way for a
> > reason.  Andrew and I had numerous discussions about this.  For that
> > matter, my first implementation did exactly what you're proposing, but
> > he had reservations about using range_on_entry, which IIRC he thought
> > should be removed from the (public) API because it had a tendency to
> > blow up lookups.
> >
> > Let's wait for Andrew to chime in on this.  If indeed the commentary
> > is out of date, I would much rather use range_on_entry like you
> > propose, but he and I have fought many times about this... over
> > various versions of the path solver :).
> >
> > For now I would return VARYING in range_on_path_entry if range_of_expr
> > returns false.  We shouldn't be ICEing when we can gracefully handle
> > things.  This gcc_unreachable was there to catch implementation issues
> > during development.
> >
> > I would keep your gimple_range_ssa_p check regardless.  No sense doing
> > extra work if we're absolutely sure we won't handle it.
>
> OK, I'll push just the gimple_range_ssa_p then since that resolves
> the PR on its own.  I was first misled about the gcc_unreachable
> and my brain hurt understanding this function ... (also as to
> why using range_of_expr on a _random_ stmt would be OK).

Calling range_of_expr on a random stmt, is not OK, and is bound to
lead to subtle issues.  As I mentioned earlier, and both in the
comments for class path_range_query and
path_range_query::internal_range_of_expr, all we really support is
querying range_of_stmt and range_of_expr as it would appear at the end
of the path.

Internally to the path solver, if it uses range_of_expr and the SSA is
defined out side the path, we'll ignore the statement altogether and
return the range on entry to the path.  So yeah... feeding random
statements is not good.  It's meant to be used to query ranges of SSA
names at the end of the path.

Hmmm, perhaps I should rewrite
path_range_query::internal_range_of_expr() to explicitly ignore the
STMT, or even put some asserts if it's being used nonsensibly.

Aldy