As noted in <http://gcc.gnu.org/ml/gcc-patches/2012-08/msg00940.html>
for CRIS PR54261:

> This should have been trivially caught by the test-suite, but
> wasn't, for three reasons.
>  One, the sync_* test-suite effective-targets use hard-coded
> target tuples, and I haven't added cris and crisv32 there, doh
> (I see also m68k missing).

Fixed thusly:

gcc/testsuite/

2013-03-23  Mikael Pettersson  <mikpe@it.uu.se>

	* lib/target-supports.exp: (check_effective_target_sync_char_short):
	Adjust m68k-*-linux* addition for aarch64*-*-* changes.

2012-12-23  Mikael Pettersson  <mikpe@it.uu.se>

	* lib/target-supports.exp: (check_effective_target_sync_int_long):
	Move m68k-*-linux* check to avoid context conflict with arm check
	adjusted post gcc-4.7.2.
	(check_effective_target_sync_char_short): Likewise.

2012-09-01  Mikael Pettersson  <mikpe@it.uu.se>

	* lib/target-supports.exp: (check_effective_target_sync_int_long):
	Add m68k-*-linux*.
	(check_effective_target_sync_char_short): Likewise.

--- gcc-4.8.0/gcc/testsuite/lib/target-supports.exp.~1~	2013-02-12 00:11:12.000000000 +0100
+++ gcc-4.8.0/gcc/testsuite/lib/target-supports.exp	2013-03-23 17:26:48.590310991 +0100
@@ -4366,6 +4366,7 @@ proc check_effective_target_sync_int_lon
 # This is intentionally powerpc but not rs6000, rs6000 doesn't have the
 # load-reserved/store-conditional instructions.
         if { [istarget ia64-*-*]
+	     || [istarget m68k-*-linux*]
 	     || [istarget i?86-*-*]
 	     || [istarget x86_64-*-*]
 	     || [istarget aarch64*-*-*]
@@ -4399,6 +4400,7 @@ proc check_effective_target_sync_char_sh
 # load-reserved/store-conditional instructions.
         if { [istarget aarch64*-*-*]
 	     || [istarget ia64-*-*]
+	     || [istarget m68k-*-linux*]
 	     || [istarget i?86-*-*]
 	     || [istarget x86_64-*-*]
 	     || [istarget alpha*-*-*]