http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59490
--- Comment #3 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
The following patch silences the failures:
--- ../_clean/gcc/testsuite/g++.dg/cilk-plus/CK/catch_exc.cc 2013-12-11
19:35:38.000000000 +0100
+++ gcc/testsuite/g++.dg/cilk-plus/CK/catch_exc.cc 2013-12-14
14:03:33.000000000 +0100
@@ -1,6 +1,6 @@
/* { dg-options "-fcilkplus" } */
/* { dg-do run { target i?86-*-* x86_64-*-* arm*-*-* } } */
-/* { dg-options "-fcilkplus -lcilkrts" { target { i?86-*-* x86_64-*-* arm*-*-*
} } } */
+/* { dg-options "-mtune=generic -fcilkplus -lcilkrts" { target { i?86-*-*
x86_64-*-* arm*-*-* } } } */
#include <assert.h>
#include <unistd.h>
However, while I understand that -mtune=* can change the content of dump files
as in pr59494 without any side effect on the generated code, I think the above
patch is only papering over a real bug.
