kernel: use gettime(9)/getuptime(9) in lieu of time_second(9)/time_uptime(9)

time_second(9) and time_uptime(9) are widely used in the kernel to
quickly get the system UTC or system uptime as a time_t.  However,
time_t is 64-bit everywhere, so it is not generally safe to use them
on 32-bit platforms: you have a split-read problem if your hardware
cannot perform atomic 64-bit reads.

This patch replaces time_second(9) with gettime(9), a safer successor
interface, throughout the kernel.  Similarly, time_uptime(9) is replaced
with getuptime(9).

There is a performance cost on 32-bit platforms in exchange for
eliminating the split-read problem: instead of two register reads you
now have a lockless read loop to pull the values from the timehands.
This is really not *too* bad in the grand scheme of things, but
compared to what we were doing before it is several times slower.

There is no performance cost on 64-bit (__LP64__) platforms.

With input from visa@, dlg@, and tedu@.

Several bugs squashed by visa@.

ok kettenis@

1 files changed, 5 insertions, 5 deletions

diff --git a/sys/net/rtsock.c b/sys/net/rtsock.c
index e00e94997e7..9d4ccadd4e2 100644
--- a/sys/net/rtsock.c
+++ b/sys/net/rtsock.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: rtsock.c,v 1.298 2020/03/24 13:50:18 tobhe Exp $	*/
+/*	$OpenBSD: rtsock.c,v 1.299 2020/06/24 22:03:42 cheloha Exp $	*/
 /*	$NetBSD: rtsock.c,v 1.18 1996/03/29 00:32:10 cgd Exp $	*/
 
 /*
@@ -1325,8 +1325,8 @@ rtm_setmetrics(u_long which, const struct rt_metrics *in,
 	if (which & RTV_EXPIRE) {
 		expire = in->rmx_expire;
 		if (expire != 0) {
-			expire -= time_second;
-			expire += time_uptime;
+			expire -= gettime();
+			expire += getuptime();
 		}
 
 		out->rmx_expire = expire;
@@ -1340,8 +1340,8 @@ rtm_getmetrics(const struct rt_kmetrics *in, struct rt_metrics *out)
 
 	expire = in->rmx_expire;
 	if (expire != 0) {
-		expire -= time_uptime;
-		expire += time_second;
+		expire -= getuptime();
+		expire += gettime();
 	}
 
 	bzero(out, sizeof(*out));