Fix nfs snapdir automount

The current implementation for allowing nfs to access snapdir is very buggy.
It uses a special fh for snapdirs, such that the next time nfsd does
fh_to_dentry, it actually returns the root inode inside the snapshot. So nfsd
never knows it cross a mountpoint.

The problem is that nfsd will not hold a reference on the vfsmount of the
snapshot. This cause auto unmounter to unmount the snapshot even though nfs is
still holding dentries in it.

To fix this, we return the inode for the snapdirs themselves. However, we also
trigger automount upon fh_to_dentry, and return ESTALE so nfsd will revalidate
and see the mountpoint and do crossmnt.

Because nfsd will now be aware that these are different filesystems users
must add crossmnt to their export options to access snapshot directories.

Reviewed-by: Brian Behlendorf <[email protected]>
Signed-off-by: Chunwei Chen <[email protected]>
Closes #3794
Closes #4716
Closes #5810 
Closes #5833 

1 files changed, 24 insertions, 17 deletions

diff --git a/module/zfs/zfs_vfsops.c b/module/zfs/zfs_vfsops.c
index 85c275822..3135319cb 100644
--- a/module/zfs/zfs_vfsops.c
+++ b/module/zfs/zfs_vfsops.c
@@ -1597,8 +1597,19 @@ zfs_vget(struct super_block *sb, struct inode **ipp, fid_t *fidp)
 
 	*ipp = NULL;
 
-	ZFS_ENTER(zsb);
+	if (fidp->fid_len == SHORT_FID_LEN || fidp->fid_len == LONG_FID_LEN) {
+		zfid_short_t	*zfid = (zfid_short_t *)fidp;
+
+		for (i = 0; i < sizeof (zfid->zf_object); i++)
+			object |= ((uint64_t)zfid->zf_object[i]) << (8 * i);
 
+		for (i = 0; i < sizeof (zfid->zf_gen); i++)
+			fid_gen |= ((uint64_t)zfid->zf_gen[i]) << (8 * i);
+	} else {
+		return (SET_ERROR(EINVAL));
+	}
+
+	/* LONG_FID_LEN means snapdirs */
 	if (fidp->fid_len == LONG_FID_LEN) {
 		zfid_long_t	*zlfid = (zfid_long_t *)fidp;
 		uint64_t	objsetid = 0;
@@ -1610,28 +1621,24 @@ zfs_vget(struct super_block *sb, struct inode **ipp, fid_t *fidp)
 		for (i = 0; i < sizeof (zlfid->zf_setgen); i++)
 			setgen |= ((uint64_t)zlfid->zf_setgen[i]) << (8 * i);
 
-		ZFS_EXIT(zsb);
+		if (objsetid != ZFSCTL_INO_SNAPDIRS - object) {
+			dprintf("snapdir fid: objsetid (%llu) != "
+			    "ZFSCTL_INO_SNAPDIRS (%llu) - object (%llu)\n",
+			    objsetid, ZFSCTL_INO_SNAPDIRS, object);
 
-		err = zfsctl_lookup_objset(sb, objsetid, &zsb);
-		if (err)
 			return (SET_ERROR(EINVAL));
+		}
 
-		ZFS_ENTER(zsb);
-	}
-
-	if (fidp->fid_len == SHORT_FID_LEN || fidp->fid_len == LONG_FID_LEN) {
-		zfid_short_t	*zfid = (zfid_short_t *)fidp;
-
-		for (i = 0; i < sizeof (zfid->zf_object); i++)
-			object |= ((uint64_t)zfid->zf_object[i]) << (8 * i);
+		if (fid_gen > 1 || setgen != 0) {
+			dprintf("snapdir fid: fid_gen (%llu) and setgen "
+			    "(%llu)\n", fid_gen, setgen);
+			return (SET_ERROR(EINVAL));
+		}
 
-		for (i = 0; i < sizeof (zfid->zf_gen); i++)
-			fid_gen |= ((uint64_t)zfid->zf_gen[i]) << (8 * i);
-	} else {
-		ZFS_EXIT(zsb);
-		return (SET_ERROR(EINVAL));
+		return (zfsctl_snapdir_vget(sb, objsetid, fid_gen, ipp));
 	}
 
+	ZFS_ENTER(zsb);
 	/* A zero fid_gen means we are in the .zfs control directories */
 	if (fid_gen == 0 &&
 	    (object == ZFSCTL_INO_ROOT || object == ZFSCTL_INO_SNAPDIR)) {