[Bug tree-optimization/59164] [4.7/4.8/4.9 Regression] ice: tree check:=?UTF-8?Q?=20expected=20tree=20that=20contains=20=E2=80=98decl=20minimal=E2=80=99=20structure?=,=?UTF-8?Q?=20have=20=E2=80=98integer=5Fcst=E2=80=99=20in=20get=5Fvar=5Finfo?=, at tree-into-ssa.c:380

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59164

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |congh at gcc dot gnu.org

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
Interestingly the commented

@@ -1672,7 +1670,7 @@ vect_update_ivs_after_vectorizer (loop_v
   gimple_stmt_iterator gsi, gsi1;
   basic_block update_bb = update_e->dest;

-  /* gcc_assert (vect_can_advance_ivs_p (loop_vinfo)); */
+  gcc_assert (vect_can_advance_ivs_p (loop_vinfo));

   /* Make sure there exists a single-predecessor exit bb:  */
   gcc_assert (single_pred_p (exit_bb));

triggers here after I fix the original ICE (we can deal with that).
This points to vect_analyze_loop_operations which has a mismatched
condition to check peeling ...