[Bug tree-optimization/106936] [13 Regression] ICE in get_value_range, at value-query.cc:170 since r13-1815-g8b8103dcd2624936

[aldyh at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106936

--- Comment #2 from Aldy Hernandez <aldyh at gcc dot gnu.org> ---
This assert was put here to make sure that the legacy get_value_range() wasn't
being called on stuff that legacy couldn't handle (floats, etc), because the
result would ultimately be copied into a value_range_equiv.

In this case, simplify_casted_cond() is calling it on an offset_type which
isn't either an integer nor a pointer, so the assert is failing.  However,
range_of_expr happily punted on it because it couldn't handle it, so we're just
returning VARYING.  As value_range_equiv can store VARYING types of anything
(including types it can't handle), this is fine.

I think the easiest thing to do is remove the assert.  If someone from the non
legacy world tries to get a non integer/pointer range here, it's going to blow
up anyhow because the temporary in get_value_range is int_range_max.

Anywho.  This should do the trick.  I'm in transit to Cauldron.  Could someone
test and push this?  I'd hate to leave what looks like an ICE on valid code
open.

diff --git a/gcc/value-query.cc b/gcc/value-query.cc
index 06ad5fe9708..0bdd670982b 100644
--- a/gcc/value-query.cc
+++ b/gcc/value-query.cc
@@ -167,7 +167,6 @@ range_query::free_value_range_equiv (value_range_equiv *v)
 const class value_range_equiv *
 range_query::get_value_range (const_tree expr, gimple *stmt)
 {
-  gcc_checking_assert (value_range_equiv::supports_p (TREE_TYPE (expr)));
   int_range_max r;
   if (range_of_expr (r, const_cast<tree> (expr), stmt))
     return new (equiv_alloc->allocate ()) value_range_equiv (r);