re PR tree-optimization/24142 (VRP miscompiles unzip inflate.c)

	PR 24142
	* tree-vrp.c (vrp_meet): Fix call to range_includes_zero_p in
	case of anti-ranges.

testsuite/

	PR 24142
	* gcc.c-torture/execute/pr24142.c: New test.

From-SVN: r104874

gcc/ChangeLog

+2005-10-02  Diego Novillo  <dnovillo@redhat.com>
+
+	PR 24142
+	* tree-vrp.c (vrp_meet): Fix call to range_includes_zero_p in
+	case of anti-ranges.
+
 2005-10-02  Andrew Pinski  <pinskia@physics.uc.edu>
 
 	PR c/18851

gcc/testsuite/ChangeLog

+2005-10-02  Diego Novillo  <dnovillo@redhat.com>
+
+	PR 24142
+	* gcc.c-torture/execute/pr24142.c: New test.
+
 2005-10-01  Diego Novillo  <dnovillo@redhat.com>
 
 	PR 24141

gcc/testsuite/gcc.c-torture/execute/pr24142.c

+void abort (void);
+
+int f (int a, int b)
+{
+  if (a == 1)
+    a = 0;
+  if (b == 0)
+    a = 1;
+  if (a != 0)
+    return 0;
+  return 1;
+}
+
+int main (void)
+{
+  if (f (1, 1) != 1)
+    abort ();
+  return 0;
+}

gcc/tree-vrp.c

@@ -3454,11 +3454,15 @@ vrp_meet (value_range_t *vr0, value_range_t *vr1)
 no_meet:
   /* The two range VR0 and VR1 do not meet.  Before giving up and
      setting the result to VARYING, see if we can at least derive a
-     useful anti-range.  */
+     useful anti-range.  FIXME, all this nonsense about distinguishing
+     anti-ranges from ranges is necessary because of the odd
+     semantics of range_includes_zero_p and friends.  */
   if (!symbolic_range_p (vr0)
-      && !range_includes_zero_p (vr0)
+      && ((vr0->type == VR_RANGE && !range_includes_zero_p (vr0))
+	  || (vr0->type == VR_ANTI_RANGE && range_includes_zero_p (vr0)))
       && !symbolic_range_p (vr1)
-      && !range_includes_zero_p (vr1))
+      && ((vr1->type == VR_RANGE && !range_includes_zero_p (vr1))
+	  || (vr1->type == VR_ANTI_RANGE && range_includes_zero_p (vr1))))
     {
       set_value_range_to_nonnull (vr0, TREE_TYPE (vr0->min));