getdirentries.2: fix for NFS mounts

It was reported that getdirentries(2) was
returning dirents with d_off set to 0 for an NFS
mount.

This is believed to be correct behaviour at
this time (it may change for some NFS mounts
in the future), but is inconsistent with what the
getdirentries(2) man page says.

This patch fixes the man page.

This is a content change.

PR:	253428
Reviewed by:	asomers
MFC after:	2 weeks
Differential Revision:	https://reviews.freebsd.org/D28664

1 files changed, 8 insertions, 6 deletions

diff --git a/lib/libc/sys/getdirentries.2 b/lib/libc/sys/getdirentries.2
index f2d1c05240d5..658be9459af5 100644
--- a/lib/libc/sys/getdirentries.2
+++ b/lib/libc/sys/getdirentries.2
@@ -28,7 +28,7 @@
 .\"	@(#)getdirentries.2	8.2 (Berkeley) 5/3/95
 .\" $FreeBSD$
 .\"
-.Dd March 30, 2020
+.Dd February 14, 2021
 .Dt GETDIRENTRIES 2
 .Os
 .Sh NAME
@@ -89,7 +89,7 @@ have the same
 .Fa d_fileno .
 The
 .Fa d_off
-field returns a cookie which can be used with
+field returns a cookie which, if non-zero, can be used with
 .Xr lseek 2
 to position the directory descriptor to the next entry.
 The
@@ -148,14 +148,16 @@ only
 .Pc ,
 a value returned in the
 .Fa d_off
-field,
+field if it is non-zero,
 or zero.
 .Sh IMPLEMENTATION NOTES
 The
 .Fa d_off
-field is being used as a cookie to readdir for nfs servers.
-These cookies can be cached and allow to read directory entries at a specific
-offset on demand.
+field is currently set to 0 by the NFS client, since the
+directory offset cookies returned by an NFS server cannot
+be used by
+.Xr lseek 2
+at this time.
 .Sh RETURN VALUES
 If successful, the number of bytes actually transferred is returned.
 Otherwise, -1 is returned and the global variable