[Bug c++/70594] [6 Regression] -fcompare-debug failure

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70594

--- Comment #15 from Patrick Palka <ppalka at gcc dot gnu.org> ---
I tried building latest gcc with BOOT_CFLAGS="-O2 -fcompare-debug" and the
build fails with an -fcompare-debug error while compiling dwarf2out.c.  The
failure doesn't go away if I make the constexpr_call_table non-deletable
either.  However, if I "disable" the fundef_copies_table with the following
patch as well then the failure goes away.  Like in Tobias' case, passing
-save-temps to the compiler makes the failure mysteriously go away too.

diff --git a/gcc/cp/constexpr.c b/gcc/cp/constexpr.c
index d84377b..234b09c 100644
--- a/gcc/cp/constexpr.c
+++ b/gcc/cp/constexpr.c
@@ -915,7 +915,7 @@ struct constexpr_ctx {
 /* A table of all constexpr calls that have been evaluated by the
    compiler in this translation unit.  */

-static GTY ((deletable)) hash_table<constexpr_call_hasher>
*constexpr_call_table;
+static GTY (()) hash_table<constexpr_call_hasher> *constexpr_call_table;

 static tree cxx_eval_constant_expression (const constexpr_ctx *, tree,
                                          bool, bool *, bool *, tree * = NULL);
@@ -1012,7 +1012,7 @@ get_fundef_copy (tree fun)

   fundef_copy *copy;
   fundef_copy **slot = &fundef_copies_table.map->get_or_insert (fun, NULL);
-  if (*slot == NULL)
+  if (1)
     {
       copy = ggc_alloc<fundef_copy> ();
       copy->body = copy_fn (fun, copy->parms, copy->res);

So it looks like marking constexpr_call_table as non-deletable is not
sufficient.