[PATCH] Disable pr46909 testcase on s390 and s390x

Jakub Jelinek jakub@redhat.com
Tue Feb 22 13:45:00 GMT 2011


On Tue, Feb 22, 2011 at 02:09:37PM +0100, Andreas Krebbel wrote:
> On 02/22/2011 01:57 PM, Jakub Jelinek wrote:
> > Please see a different patch I've just attached to PR47835 for this issue.
> 
> Your patch fixes the problem on s390x. Thanks!

Here it is, ok for trunk (tested on x86_64-linux and i686-linux with the
gimple-fold.c fix reverted and with vanilla gimple-fold.c and similarly with
a cross to powerpc64-linux (-m32/-m64))?

2011-02-22  Jakub Jelinek  <jakub@redhat.com>

	PR tree-optimization/47835
	* gcc.dg/pr46909.c: Check optimized dump instead of ifcombine.

--- gcc/testsuite/gcc.dg/pr46909.c.jj	2010-12-16 10:55:30.000000000 +0100
+++ gcc/testsuite/gcc.dg/pr46909.c	2011-02-22 13:37:21.000000000 +0100
@@ -1,6 +1,6 @@
 /* PR tree-optimization/46909 */
 /* { dg-do compile } */
-/* { dg-options "-O2 -fdump-tree-ifcombine" } */
+/* { dg-options "-O2 -fdump-tree-optimized" } */
 
 extern void abort ();
 
@@ -13,5 +13,8 @@ foo (unsigned int x)
   return -1;
 }
 
-/* { dg-final { scan-tree-dump "optimizing two comparisons to x_\[0-9\]+\\(D\\) != 4" "ifcombine" } } */
-/* { dg-final { cleanup-tree-dump "ifcombine" } } */
+/* { dg-final { scan-tree-dump-times "x_\[0-9\]+\\(D\\) != 4" 1 "optimized" } } */
+/* { dg-final { scan-tree-dump-times "x_\[0-9\]+\\(D\\) != 6" 0 "optimized" } } */
+/* { dg-final { scan-tree-dump-times "x_\[0-9\]+\\(D\\) == 2" 0 "optimized" } } */
+/* { dg-final { scan-tree-dump-times "x_\[0-9\]+\\(D\\) == 6" 0 "optimized" } } */
+/* { dg-final { cleanup-tree-dump "optimized" } } */


	Jakub



More information about the Gcc-patches mailing list