Add thread-local storage support for arm:

- Switch to Variant I TCB layout
- Use function from rtld for TCB allocation/deallocation

lib/libthr/arch/arm/arm/pthread_md.c

@@ -37,14 +37,17 @@ _tcb_ctor(struct pthread *thread, int initial)
 {
 	struct tcb *tcb;
 
-	tcb = malloc(sizeof(struct tcb));
+	tcb = _rtld_allocate_tls((initial) ? _tcb_get() :  NULL,
+	    sizeof(struct tcb), 16);
 	if (tcb)
 		tcb->tcb_thread = thread;
+
 	return (tcb);
 }
 
 void
 _tcb_dtor(struct tcb *tcb)
 {
-	free(tcb);
+
+	_rtld_free_tls(tcb, sizeof(struct tcb), 16);
 }

lib/libthr/arch/arm/include/pthread_md.h

@@ -43,10 +43,8 @@
  * Variant II tcb, first two members are required by rtld.
  */
 struct tcb {
-	struct tcb		*tcb_self;	/* required by rtld */
 	void			*tcb_dtv;	/* required by rtld */
 	struct pthread		*tcb_thread;	/* our hook */
-	void			*tcb_spare[1];
 };
 
 /*