Support full-precision file timestamps. Until now, only the seconds

have been maintained, and that is still the default.  A new sysctl
variable "vfs.timestamp_precision" can be used to enable higher
levels of precision:

      0 = seconds only; nanoseconds zeroed (default).
      1 = seconds and nanoseconds, accurate within 1/HZ.
      2 = seconds and nanoseconds, truncated to microseconds.
    >=3 = seconds and nanoseconds, maximum precision.

Level 1 uses getnanotime(), which is fast but can be wrong by up
to 1/HZ.  Level 2 uses microtime().  It might be desirable for
consistency with utimes() and friends, which take timeval structures
rather than timespecs.  Level 3 uses nanotime() for the higest
precision.

I benchmarked levels 0, 1, and 3 by copying a 550 MB tree with
"cpio -pdu".  There was almost negligible difference in the system
times -- much less than 1%, and less than the variation among
multiple runs at the same level.  Bruce Evans dreamed up a torture
test involving 1-byte reads with intervening fstat() calls, but
the cpio test seems more realistic to me.

This feature is currently implemented only for the UFS (FFS and
MFS) filesystems.  But I think it should be easy to support it in
the others as well.

An earlier version of this was reviewed by Bruce.  He's not to
blame for any breakage I've introduced since then.

Reviewed by:	bde (an earlier version of the code)

4 files changed, 107 insertions, 13 deletions

diff --git a/sys/kern/vfs_export.c b/sys/kern/vfs_export.c
index 21bb57b..043a5bf 100644
--- a/sys/kern/vfs_export.c
+++ b/sys/kern/vfs_export.c
@@ -36,7 +36,7 @@
  * SUCH DAMAGE.
  *
  *	@(#)vfs_subr.c	8.31 (Berkeley) 5/26/95
- * $Id: vfs_subr.c,v 1.216 1999/08/13 10:10:01 phk Exp $
+ * $Id: vfs_subr.c,v 1.217 1999/08/13 10:29:21 phk Exp $
  */
 
 /*
@@ -361,6 +361,48 @@ vfs_getnewfsid(mp)
 }
 
 /*
+ * Knob to control the precision of file timestamps:
+ *
+ *   0 = seconds only; nanoseconds zeroed.
+ *   1 = seconds and nanoseconds, accurate within 1/HZ.
+ *   2 = seconds and nanoseconds, truncated to microseconds.
+ * >=3 = seconds and nanoseconds, maximum precision.
+ */
+enum { TSP_SEC, TSP_HZ, TSP_USEC, TSP_NSEC };
+
+static int timestamp_precision = TSP_SEC;
+SYSCTL_INT(_vfs, OID_AUTO, timestamp_precision, CTLFLAG_RW,
+    &timestamp_precision, 0, "");
+
+/*
+ * Get a current timestamp.
+ */
+void
+vfs_timestamp(tsp)
+	struct timespec *tsp;
+{
+	struct timeval tv;
+
+	switch (timestamp_precision) {
+	case TSP_SEC:
+		tsp->tv_sec = time_second;
+		tsp->tv_nsec = 0;
+		break;
+	case TSP_HZ:
+		getnanotime(tsp);
+		break;
+	case TSP_USEC:
+		microtime(&tv);
+		TIMEVAL_TO_TIMESPEC(&tv, tsp);
+		break;
+	case TSP_NSEC:
+	default:
+		nanotime(tsp);
+		break;
+	}
+}
+
+/*
  * Set vnode attributes to VNOVAL
  */
 void
diff --git a/sys/kern/vfs_subr.c b/sys/kern/vfs_subr.c
index 21bb57b..043a5bf 100644
--- a/sys/kern/vfs_subr.c
+++ b/sys/kern/vfs_subr.c
@@ -36,7 +36,7 @@
  * SUCH DAMAGE.
  *
  *	@(#)vfs_subr.c	8.31 (Berkeley) 5/26/95
- * $Id: vfs_subr.c,v 1.216 1999/08/13 10:10:01 phk Exp $
+ * $Id: vfs_subr.c,v 1.217 1999/08/13 10:29:21 phk Exp $
  */
 
 /*
@@ -361,6 +361,48 @@ vfs_getnewfsid(mp)
 }
 
 /*
+ * Knob to control the precision of file timestamps:
+ *
+ *   0 = seconds only; nanoseconds zeroed.
+ *   1 = seconds and nanoseconds, accurate within 1/HZ.
+ *   2 = seconds and nanoseconds, truncated to microseconds.
+ * >=3 = seconds and nanoseconds, maximum precision.
+ */
+enum { TSP_SEC, TSP_HZ, TSP_USEC, TSP_NSEC };
+
+static int timestamp_precision = TSP_SEC;
+SYSCTL_INT(_vfs, OID_AUTO, timestamp_precision, CTLFLAG_RW,
+    &timestamp_precision, 0, "");
+
+/*
+ * Get a current timestamp.
+ */
+void
+vfs_timestamp(tsp)
+	struct timespec *tsp;
+{
+	struct timeval tv;
+
+	switch (timestamp_precision) {
+	case TSP_SEC:
+		tsp->tv_sec = time_second;
+		tsp->tv_nsec = 0;
+		break;
+	case TSP_HZ:
+		getnanotime(tsp);
+		break;
+	case TSP_USEC:
+		microtime(&tv);
+		TIMEVAL_TO_TIMESPEC(&tv, tsp);
+		break;
+	case TSP_NSEC:
+	default:
+		nanotime(tsp);
+		break;
+	}
+}
+
+/*
  * Set vnode attributes to VNOVAL
  */
 void
diff --git a/sys/sys/vnode.h b/sys/sys/vnode.h
index fc9dbe9..aa02ac4 100644
--- a/sys/sys/vnode.h
+++ b/sys/sys/vnode.h
@@ -31,7 +31,7 @@
  * SUCH DAMAGE.
  *
  *	@(#)vnode.h	8.7 (Berkeley) 2/4/94
- * $Id: vnode.h,v 1.93 1999/08/08 18:43:00 phk Exp $
+ * $Id: vnode.h,v 1.94 1999/08/13 10:10:11 phk Exp $
  */
 
 #ifndef _SYS_VNODE_H_
@@ -544,6 +544,7 @@ dev_t	vn_todev __P((struct vnode *vp));
 int	vfs_cache_lookup __P((struct vop_lookup_args *ap));
 int	vfs_object_create __P((struct vnode *vp, struct proc *p,
                 struct ucred *cred));
+void	vfs_timestamp __P((struct timespec *));
 int 	vn_writechk __P((struct vnode *vp));
 int	vop_stdbwrite __P((struct vop_bwrite_args *ap));
 int	vop_stdislocked __P((struct vop_islocked_args *));
diff --git a/sys/ufs/ufs/ufs_vnops.c b/sys/ufs/ufs/ufs_vnops.c
index e65f22b..f68e6da 100644
--- a/sys/ufs/ufs/ufs_vnops.c
+++ b/sys/ufs/ufs/ufs_vnops.c
@@ -36,7 +36,7 @@
  * SUCH DAMAGE.
  *
  *	@(#)ufs_vnops.c	8.27 (Berkeley) 5/27/95
- * $Id: ufs_vnops.c,v 1.118 1999/08/13 10:10:12 phk Exp $
+ * $Id: ufs_vnops.c,v 1.119 1999/08/13 10:56:02 phk Exp $
  */
 
 #include "opt_quota.h"
@@ -138,26 +138,31 @@ ufs_itimes(vp)
 	struct vnode *vp;
 {
 	struct inode *ip;
-	time_t tv_sec;
+	struct timespec ts;
 
 	ip = VTOI(vp);
 	if ((ip->i_flag & (IN_ACCESS | IN_CHANGE | IN_UPDATE)) == 0)
 		return;
 	if ((vp->v_mount->mnt_flag & MNT_RDONLY) == 0) {
-		tv_sec = time_second;
+		vfs_timestamp(&ts);
 		if ((vp->v_type == VBLK || vp->v_type == VCHR) &&
 		    !DOINGSOFTDEP(vp))
 			ip->i_flag |= IN_LAZYMOD;
 		else
 			ip->i_flag |= IN_MODIFIED;
-		if (ip->i_flag & IN_ACCESS)
-			ip->i_atime = tv_sec;
+		if (ip->i_flag & IN_ACCESS) {
+			ip->i_atime = ts.tv_sec;
+			ip->i_atimensec = ts.tv_nsec;
+		}
 		if (ip->i_flag & IN_UPDATE) {
-			ip->i_mtime = tv_sec;
+			ip->i_mtime = ts.tv_sec;
+			ip->i_mtimensec = ts.tv_nsec;
 			ip->i_modrev++;
 		}
-		if (ip->i_flag & IN_CHANGE)
-			ip->i_ctime = tv_sec;
+		if (ip->i_flag & IN_CHANGE) {
+			ip->i_ctime = ts.tv_sec;
+			ip->i_ctimensec = ts.tv_nsec;
+		}
 	}
 	ip->i_flag &= ~(IN_ACCESS | IN_CHANGE | IN_UPDATE);
 }
@@ -500,10 +505,14 @@ ufs_setattr(ap)
 		if (vap->va_mtime.tv_sec != VNOVAL)
 			ip->i_flag |= IN_CHANGE | IN_UPDATE;
 		ufs_itimes(vp);
-		if (vap->va_atime.tv_sec != VNOVAL)
+		if (vap->va_atime.tv_sec != VNOVAL) {
 			ip->i_atime = vap->va_atime.tv_sec;
-		if (vap->va_mtime.tv_sec != VNOVAL)
+			ip->i_atimensec = vap->va_atime.tv_nsec;
+		}
+		if (vap->va_mtime.tv_sec != VNOVAL) {
 			ip->i_mtime = vap->va_mtime.tv_sec;
+			ip->i_mtimensec = vap->va_mtime.tv_nsec;
+		}
 		error = UFS_UPDATE(vp, 0);
 		if (error)
 			return (error);