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, 2 insertions, 2 deletions

diff --git a/sys/ufs/ffs/ffs_vfsops.c b/sys/ufs/ffs/ffs_vfsops.c
index b8a7b0a590b..86ff869783d 100644
--- a/sys/ufs/ffs/ffs_vfsops.c
+++ b/sys/ufs/ffs/ffs_vfsops.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: ffs_vfsops.c,v 1.184 2020/05/28 15:48:29 otto Exp $	*/
+/*	$OpenBSD: ffs_vfsops.c,v 1.185 2020/06/24 22:03:45 cheloha Exp $	*/
 /*	$NetBSD: ffs_vfsops.c,v 1.19 1996/02/09 22:22:26 christos Exp $	*/
 
 /*
@@ -1525,7 +1525,7 @@ ffs_sbupdate(struct ufsmount *mp, int waitfor)
 	    fs->fs_sblockloc >> (fs->fs_fshift - fs->fs_fsbtodb),
 	    (int)fs->fs_sbsize, 0, INFSLP);
 	fs->fs_fmod = 0;
-	fs->fs_time = time_second;
+	fs->fs_time = gettime();
 	memcpy(bp->b_data, fs, fs->fs_sbsize);
 	/* Restore compatibility to old file systems.		   XXX */
 	dfs = (struct fs *)bp->b_data;				/* XXX */