re PR tree-optimization/68671 (gcc.dg/torture/pr66952.c FAILs with -fno-tree-dce)

	PR tree-optimization/68671
	* tree-ssa-reassoc.c (maybe_optimize_range_tests): For basic
	blocks starting with the successor of first bb we've modified
	and ending with last_bb call reset_flow_sensitive_info_in_bb.

	* gcc.dg/pr68671.c: New test.

From-SVN: r231278

gcc/ChangeLog

+2015-12-04  Jakub Jelinek  <jakub@redhat.com>
+
+	PR tree-optimization/68671
+	* tree-ssa-reassoc.c (maybe_optimize_range_tests): For basic
+	blocks starting with the successor of first bb we've modified
+	and ending with last_bb call reset_flow_sensitive_info_in_bb.
+
 2015-12-04  Jeff Law  <law@redhat.com>
 
 	* tree-ssa-reassoc.c (maybe_optimize_range_tests): Return boolean

gcc/testsuite/ChangeLog

+2015-12-04  Jakub Jelinek  <jakub@redhat.com>
+
+	PR tree-optimization/68671
+	* gcc.dg/pr68671.c: New test.
+
 2015-12-02  Jeff Law  <law@redhat.com>
 
 	* gcc.dg/tree-ssa/reassoc-43.c: New test.

gcc/testsuite/gcc.dg/pr68671.c

+/* PR tree-optimization/68671 */
+/* { dg-do run } */
+/* { dg-options " -O2 -fno-tree-dce" } */
+
+volatile int a = -1;
+volatile int b;
+
+static inline int
+fn1 (signed char p1, int p2)
+{
+  return (p1 < 0) || (p1 > (1 >> p2)) ? 0 : (p1 << 1);
+}
+
+int
+main ()
+{
+  signed char c = a;
+  b = fn1 (c, 1);
+  c = ((128 | c) < 0 ? 1 : 0);
+  if (c != 1)
+    __builtin_abort ();
+  return 0;
+}

gcc/tree-ssa-reassoc.c

@@ -3211,7 +3211,7 @@ maybe_optimize_range_tests (gimple *stmt)
     any_changes = optimize_range_tests (ERROR_MARK, &ops);
   if (any_changes)
     {
-      unsigned int idx;
+      unsigned int idx, max_idx = 0;
       /* update_ops relies on has_single_use predicates returning the
 	 same values as it did during get_ops earlier.  Additionally it
 	 never removes statements, only adds new ones and it should walk
@@ -3227,6 +3227,7 @@ maybe_optimize_range_tests (gimple *stmt)
 	    {
 	      tree new_op;
 
+	      max_idx = idx;
 	      stmt = last_stmt (bb);
 	      new_op = update_ops (bbinfo[idx].op,
 				   (enum tree_code)
@@ -3296,6 +3297,10 @@ maybe_optimize_range_tests (gimple *stmt)
 	      && ops[bbinfo[idx].first_idx]->op != NULL_TREE)
 	    {
 	      gcond *cond_stmt = as_a <gcond *> (last_stmt (bb));
+
+	      if (idx > max_idx)
+		max_idx = idx;
+
 	      /* If we collapse the conditional to a true/false
 		 condition, then bubble that knowledge up to our caller.  */
 	      if (integer_zerop (ops[bbinfo[idx].first_idx]->op))
@@ -3320,6 +3325,17 @@ maybe_optimize_range_tests (gimple *stmt)
 	  if (bb == first_bb)
 	    break;
 	}
+
+      /* The above changes could result in basic blocks after the first
+	 modified one, up to and including last_bb, to be executed even if
+	 they would not be in the original program.  If the value ranges of
+	 assignment lhs' in those bbs were dependent on the conditions
+	 guarding those basic blocks which now can change, the VRs might
+	 be incorrect.  As no_side_effect_bb should ensure those SSA_NAMEs
+	 are only used within the same bb, it should be not a big deal if
+	 we just reset all the VRs in those bbs.  See PR68671.  */
+      for (bb = last_bb, idx = 0; idx < max_idx; bb = single_pred (bb), idx++)
+	reset_flow_sensitive_info_in_bb (bb);
     }
   return cfg_cleanup_needed;
 }