This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
committed: xfail for PR 30768
- From: Hans-Peter Nilsson <hans-peter dot nilsson at axis dot com>
- To: gcc-patches at gcc dot gnu dot org, libstdc++ at gcc dot gnu dot org
- Date: Wed, 14 Feb 2007 21:16:15 +0100
- Subject: committed: xfail for PR 30768
See the PR. It seems open for interpretation whether
*regressions* are supposed to be xfailed when a PR has been
opened, but I don't see the logic in that (like, it's not a
regression for you if you never ran the testsuite). It's not
codified (at least I can't find it), but there are to me clear
statements regarding PR-then-xfail in the archives, like in
<URL:http://gcc.gnu.org/ml/gcc/2004-10/msg00617.html> and
<URL:http://gcc.gnu.org/ml/gcc-patches/2003-11/msg00271.html>.
Can we codify procedure on the regular pages? Either way, it
helps my maintenance and it's a target-specific tweak so it's
according to rules. Still, I guess most targets see this; at
least x86_64 and ppc too.
PR middle-end/30768
* testsuite/ext/pb_ds/regression/list_update_data_map_rand.cc:
Xfail ICE for cris-*-*.
Index: testsuite/ext/pb_ds/regression/list_update_data_map_rand.cc
===================================================================
--- testsuite/ext/pb_ds/regression/list_update_data_map_rand.cc (revision 121954)
+++ testsuite/ext/pb_ds/regression/list_update_data_map_rand.cc (working copy)
@@ -1,4 +1,9 @@
// { dg-options "-DPB_DS_REGRESSION" }
+// { dg-bogus "In member function" "PR 30768, ICE" { xfail cris-*-* } 0 }
+// { dg-bogus "internal compiler error" "ICE" { xfail cris-*-* } 0 }
+// { dg-bogus "submit a full bug report" "ICE" { xfail cris-*-* } 0 }
+// { dg-bogus "preprocessed source" "ICE" { xfail cris-*-* } 0 }
+// { dg-bogus "bugs.html" "ICE" { xfail cris-*-* } 0 }
// -*- C++ -*-
// Copyright (C) 2005, 2006 Free Software Foundation, Inc.
brgds, H-P