[S390] Inline assembly cleanup.

Major cleanup of all s390 inline assemblies. They now have a common
coding style. Quite a few have been shortened, mainly by using register
asm variables. Use of the EX_TABLE macro helps  as well. The atomic ops,
bit ops and locking inlines new use the Q-constraint if a newer gcc
is used.  That results in slightly better code.

Thanks to Christian Borntraeger for proof reading the changes.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
diff --git a/include/asm-s390/timex.h b/include/asm-s390/timex.h
index 5d0332a..4df4a41 100644
--- a/include/asm-s390/timex.h
+++ b/include/asm-s390/timex.h
@@ -15,20 +15,21 @@
 
 typedef unsigned long long cycles_t;
 
-static inline cycles_t get_cycles(void)
-{
-	cycles_t cycles;
-
-	__asm__ __volatile__ ("stck 0(%1)" : "=m" (cycles) : "a" (&cycles) : "cc");
-	return cycles >> 2;
-}
-
 static inline unsigned long long get_clock (void)
 {
 	unsigned long long clk;
 
-	__asm__ __volatile__ ("stck 0(%1)" : "=m" (clk) : "a" (&clk) : "cc");
+#if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ > 2)
+	asm volatile("stck %0" : "=Q" (clk) : : "cc");
+#else /* __GNUC__ */
+	asm volatile("stck 0(%1)" : "=m" (clk) : "a" (&clk) : "cc");
+#endif /* __GNUC__ */
 	return clk;
 }
 
+static inline cycles_t get_cycles(void)
+{
+	return (cycles_t) get_clock() >> 2;
+}
+
 #endif