Work-around a compiler optimization bug, that broke libthr. Massive

inlining resulted in constant propagation to the extend that cmpval
was known to the compiler to be URWLOCK_WRITE_OWNER (= 0x80000000U).
Unfortunately, instead of zero-extending the unsigned constant, it
was sign-extended. As such, the cmpxchg instruction was comparing
0x0000000080000000LU to 0xffffffff80000000LU and obviously didn't
perform the exchange.
But, since the value returned by cmpxhg equalled cmpval (when zero-
extended), the _thr_rtld_lock_release() function thought the exchange
did happen and as such returned as if having released the lock. This
was not the case. Subsequent locking requests found rw_state non-zero
and the thread in question entered the kernel and block indefinitely.

The work-around is to zero-extend by casting to uint64_t.

sys/ia64/include/atomic.h

@@ -42,7 +42,7 @@
 		"mov ar.ccv=%2;;\n\t"					\
 		"cmpxchg" #sz "." #sem " %0=%4,%3,ar.ccv\n\t"		\
 		: "=r" (ret), "=m" (*p)					\
-		: "r" (cmpval), "r" (newval), "m" (*p)			\
+		: "r" ((uint64_t)cmpval), "r" (newval), "m" (*p)	\
 		: "memory")
 
 /*