base/SOURCES/nfs-utils-1.3.0-blkmapd-pnfs.patch

diff --git a/Makefile.am b/Makefile.am
index ae7cd16..c9e9f87 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -2,7 +2,7 @@
 
 AUTOMAKE_OPTIONS = foreign
 
-SUBDIRS = tools support utils linux-nfs tests
+SUBDIRS = tools support utils linux-nfs tests systemd
 
 MAINTAINERCLEANFILES = Makefile.in
 
diff --git a/configure.ac b/configure.ac
index 7b93de6..4ee4db5 100644
--- a/configure.ac
+++ b/configure.ac
@@ -54,6 +54,16 @@ AC_ARG_WITH(start-statd,
 	)
 	AC_SUBST(startstatd)
 	AC_DEFINE_UNQUOTED(START_STATD, "$startstatd", [Define this to a script which can start statd on mount])
+unitdir=/usr/lib/systemd/system
+AC_ARG_WITH(systemd,
+	[AC_HELP_STRING([--with-systemd@<:@=unit-dir-path@:>@],
+			[install systemd unit files @<:@Default: no, and path defaults to /usr/lib/systemd/system if not given@:>@])],
+	test "$withval" = "no" && use_systemd=0 || unitdir=$withval use_systemd=1
+	use_systemd=0
+	)
+	AM_CONDITIONAL(INSTALL_SYSTEMD, [test "$use_systemd" = 1])
+	AC_SUBST(unitdir)
+
 AC_ARG_ENABLE(nfsv4,
 	[AC_HELP_STRING([--enable-nfsv4],
                         [enable support for NFSv4 @<:@default=yes@:>@])],
@@ -506,6 +516,7 @@ AC_CONFIG_FILES([
 	utils/showmount/Makefile
 	utils/statd/Makefile
 	utils/osd_login/Makefile
+	systemd/Makefile
 	tests/Makefile
 	tests/nsm_client/Makefile])
 AC_OUTPUT
diff --git a/support/include/nfs/export.h b/support/include/nfs/export.h
index 2f59e6a..1194255 100644
--- a/support/include/nfs/export.h
+++ b/support/include/nfs/export.h
@@ -26,6 +26,7 @@
 #define	NFSEXP_CROSSMOUNT	0x4000
 #define NFSEXP_NOACL		0x8000 /* reserved for possible ACL related use */
 #define NFSEXP_V4ROOT		0x10000
+#define NFSEXP_PNFS            0x20000
 /*
  * All flags supported by the kernel before addition of the
  * export_features interface:
diff --git a/support/nfs/exports.c b/support/nfs/exports.c
index 5451ed7..9399a12 100644
--- a/support/nfs/exports.c
+++ b/support/nfs/exports.c
@@ -275,6 +275,7 @@ putexportent(struct exportent *ep)
 		"no_" : "");
 	if (ep->e_flags & NFSEXP_NOREADDIRPLUS)
 		fprintf(fp, "nordirplus,");
+	fprintf(fp, "%spnfs,", (ep->e_flags & NFSEXP_PNFS)? "" : "no_");
 	if (ep->e_flags & NFSEXP_FSID) {
 		fprintf(fp, "fsid=%d,", ep->e_fsid);
 	}
@@ -581,6 +582,10 @@ parseopts(char *cp, struct exportent *ep, int warn, int *had_subtree_opt_ptr)
 			clearflags(NFSEXP_NOACL, active, ep);
 		else if (strcmp(opt, "no_acl") == 0)
 			setflags(NFSEXP_NOACL, active, ep);
+		else if (!strcmp(opt, "pnfs"))
+			setflags(NFSEXP_PNFS, active, ep);
+		else if (!strcmp(opt, "no_pnfs"))
+			clearflags(NFSEXP_PNFS, active, ep);
 		else if (strncmp(opt, "anonuid=", 8) == 0) {
 			char *oe;
 			ep->e_anonuid = strtol(opt+8, &oe, 10);
diff --git a/systemd/Makefile.am b/systemd/Makefile.am
new file mode 100644
index 0000000..fbcabb1
--- /dev/null
+++ b/systemd/Makefile.am
@@ -0,0 +1,31 @@
+## Process this file with automake to produce Makefile.in
+
+MAINTAINERCLEANFILES = Makefile.in
+
+unit_files =  \
+    nfs-client.target \
+    \
+    auth-rpcgss-module.service \
+    nfs-blkmap.service \
+    nfs-config.service \
+    nfs-idmapd.service \
+    nfs-mountd.service \
+    nfs-server.service \
+    nfs-utils.service \
+    rpc-gssd.service \
+    rpc-statd-notify.service \
+    rpc-statd.service \
+    rpc-svcgssd.service \
+    \
+    proc-fs-nfsd.mount \
+    var-lib-nfs-rpc_pipefs.mount
+
+EXTRA_DIST = $(unit_files)
+
+unit_dir = /usr/lib/systemd/system
+
+if INSTALL_SYSTEMD
+install-data-hook: $(unit_files)
+	mkdir -p $(DESTDIR)/$(unitdir)
+	cp $(unit_files) $(DESTDIR)/$(unitdir)
+endif
diff --git a/systemd/README b/systemd/README
index a2a5f06..bbd7790 100644
--- a/systemd/README
+++ b/systemd/README
@@ -24,7 +24,7 @@ by a suitable 'preset' setting:
     is started by /usr/sbin/start-statd which mount.nfs will run
     if statd is needed.
 
- nfs-blkmap.target
+ nfs-blkmap.service
     If enabled, then blkmapd will be run when nfs-client.target is
     started.
 
diff --git a/systemd/nfs-blkmap.service b/systemd/nfs-blkmap.service
index f470e3d..ddbf4e9 100644
--- a/systemd/nfs-blkmap.service
+++ b/systemd/nfs-blkmap.service
@@ -5,12 +5,13 @@ Conflicts=umount.target
 After=var-lib-nfs-rpc_pipefs.mount
 Requires=var-lib-nfs-rpc_pipefs.mount
 
-Requisite=nfs-blkmap.target
-After=nfs-blkmap.target
-
 PartOf=nfs-utils.service
 
 [Service]
 Type=forking
 PIDFile=/var/run/blkmapd.pid
+EnvironmentFile=-/run/sysconfig/nfs-utils
 ExecStart=/usr/sbin/blkmapd $BLKMAPDARGS
+
+[Install]
+WantedBy=nfs-client.target
diff --git a/systemd/nfs-blkmap.target b/systemd/nfs-blkmap.target
deleted file mode 100644
index fbcc111..0000000
--- a/systemd/nfs-blkmap.target
+++ /dev/null
@@ -1,8 +0,0 @@
-[Unit]
-Description= PNFS blkmaping enablement.
-# If this target is enabled, then blkmapd will be started
-# as required.  If it is not enabled it won't.
-
-[Install]
-WantedBy=remote-fs.target
-WantedBy=multi-user.target
\ No newline at end of file
diff --git a/systemd/nfs-client.target b/systemd/nfs-client.target
index 9b792a3..8a8300a 100644
--- a/systemd/nfs-client.target
+++ b/systemd/nfs-client.target
@@ -5,8 +5,7 @@ Wants=remote-fs-pre.target
 
 # Note: we don't "Wants=rpc-statd.service" as "mount.nfs" will arrange to
 # start that on demand if needed.
-Wants=nfs-blkmap.service rpc-statd-notify.service
-After=nfs-blkmap.service
+Wants=rpc-statd-notify.service
 
 # GSS services dependencies and ordering
 Wants=auth-rpcgss-module.service
diff --git a/utils/blkmapd/device-discovery.c b/utils/blkmapd/device-discovery.c
index df4627e..b52afe2 100644
--- a/utils/blkmapd/device-discovery.c
+++ b/utils/blkmapd/device-discovery.c
@@ -77,16 +77,6 @@ struct bl_disk_path *bl_get_path(const char *filepath,
 	return tmp;
 }
 
-/* Check whether valid_path is a substring(partition) of path */
-int bl_is_partition(struct bl_disk_path *valid_path, struct bl_disk_path *path)
-{
-	if (!strncmp(valid_path->full_path, path->full_path,
-		     strlen(valid_path->full_path)))
-		return 1;
-
-	return 0;
-}
-
 /*
  * For multipath devices, devices state could be PASSIVE/ACTIVE/PSEUDO,
  * where PSEUDO > ACTIVE > PASSIVE. Device with highest state is used to
@@ -95,19 +85,13 @@ int bl_is_partition(struct bl_disk_path *valid_path, struct bl_disk_path *path)
  * If device-mapper multipath support is a must, pseudo devices should
  * exist for each multipath device. If not, active device path will be
  * chosen for device creation.
- * Treat partition as invalid path.
  */
-int bl_update_path(struct bl_disk_path *path, enum bl_path_state_e state,
-		   struct bl_disk *disk)
+int bl_update_path(enum bl_path_state_e state, struct bl_disk *disk)
 {
 	struct bl_disk_path *valid_path = disk->valid_path;
 
-	if (valid_path) {
-		if (valid_path->state >= state) {
-			if (bl_is_partition(valid_path, path))
-				return 0;
-		}
-	}
+	if (valid_path && valid_path->state >= state)
+		return 0;
 	return 1;
 }
 
@@ -164,15 +148,16 @@ void bl_add_disk(char *filepath)
 
 	dev = sb.st_rdev;
 	serial = bldev_read_serial(fd, filepath);
-	if (dm_is_dm_major(major(dev)))
+	if (!serial) {
+		BL_LOG_ERR("%s: no serial found for %s\n",
+				 __func__, filepath);
+		ap_state = BL_PATH_STATE_PASSIVE;
+	} else if (dm_is_dm_major(major(dev)))
 		ap_state = BL_PATH_STATE_PSEUDO;
 	else
 		ap_state = bldev_read_ap_state(fd);
 	close(fd);
 
-	if (ap_state != BL_PATH_STATE_ACTIVE)
-		return;
-
 	for (disk = visible_disk_list; disk != NULL; disk = disk->next) {
 		/* Already scanned or a partition?
 		 * XXX: if released each time, maybe not need to compare
@@ -216,7 +201,7 @@ void bl_add_disk(char *filepath)
 		path->next = disk->paths;
 		disk->paths = path;
 		/* check whether we need to update disk info */
-		if (bl_update_path(path, path->state, disk)) {
+		if (bl_update_path(path->state, disk)) {
 			disk->dev = dev;
 			disk->size = size;
 			disk->valid_path = path;
diff --git a/utils/blkmapd/device-inq.c b/utils/blkmapd/device-inq.c
index eabc70c..c5bf71f 100644
--- a/utils/blkmapd/device-inq.c
+++ b/utils/blkmapd/device-inq.c
@@ -179,6 +179,7 @@ struct bl_serial *bldev_read_serial(int fd, const char *filename)
 	char *buffer;
 	struct bl_dev_id *dev_root, *dev_id;
 	unsigned int pos, len, current_id = 0;
+	size_t devid_len = sizeof(struct bl_dev_id) - sizeof(unsigned char);
 
 	status = bldev_inquire_pages(fd, 0x83, &buffer);
 	if (status)
@@ -189,7 +190,11 @@ struct bl_serial *bldev_read_serial(int fd, const char *filename)
 	pos = 0;
 	current_id = 0;
 	len = dev_root->len;
-	while (pos < (len - sizeof(struct bl_dev_id) + sizeof(unsigned char))) {
+
+	if (len < devid_len)
+		goto out;
+
+	while (pos < (len - devid_len)) {
 		dev_id = (struct bl_dev_id *)&(dev_root->data[pos]);
 		if ((dev_id->ids & 0xf) < current_id)
 			continue;
@@ -221,8 +226,7 @@ struct bl_serial *bldev_read_serial(int fd, const char *filename)
 		}
 		if (current_id == 3)
 			break;
-		pos += (dev_id->len + sizeof(struct bl_dev_id) -
-			sizeof(unsigned char));
+		pos += (dev_id->len + devid_len);
 	}
  out:
 	if (!serial_out)
diff --git a/utils/blkmapd/device-process.c b/utils/blkmapd/device-process.c
index 5fe3dff..f53a616 100644
--- a/utils/blkmapd/device-process.c
+++ b/utils/blkmapd/device-process.c
@@ -181,6 +181,8 @@ static int map_sig_to_device(struct bl_sig *sig, struct bl_volume *vol)
 		/* FIXME: should we use better algorithm for disk scan? */
 		mapped = verify_sig(disk, sig);
 		if (mapped) {
+			BL_LOG_INFO("%s: using device %s\n",
+					__func__, disk->valid_path->full_path);
 			vol->param.bv_dev = disk->dev;
 			vol->bv_size = disk->size;
 			break;
diff --git a/utils/blkmapd/dm-device.c b/utils/blkmapd/dm-device.c
index 0f4f148..24ffcbf 100644
--- a/utils/blkmapd/dm-device.c
+++ b/utils/blkmapd/dm-device.c
@@ -400,6 +400,8 @@ uint64_t dm_device_create(struct bl_volume *vols, int num_vols)
 			}
 			dev = node->bv_vols[0]->param.bv_dev;
 			tmp = table->params;
+			BL_LOG_INFO("%s: major %lu minor %lu", __func__,
+					MAJOR(dev), MINOR(dev));
 			if (!dm_format_dev(tmp, DM_PARAMS_LEN,
 					   MAJOR(dev), MINOR(dev))) {
 				free(table);
@@ -459,6 +461,8 @@ uint64_t dm_device_create(struct bl_volume *vols, int num_vols)
 				strcpy(table->target_type, "linear");
 				tmp = table->params;
 				dev = node->bv_vols[i]->param.bv_dev;
+				BL_LOG_INFO("%s: major %lu minor %lu", __func__,
+					MAJOR(dev), MINOR(dev));
 				if (!dm_format_dev(tmp, DM_PARAMS_LEN,
 						   MAJOR(dev), MINOR(dev))) {
 					free(table);
diff --git a/utils/exportfs/exportfs.c b/utils/exportfs/exportfs.c
index 8391615..53e86ec 100644
--- a/utils/exportfs/exportfs.c
+++ b/utils/exportfs/exportfs.c
@@ -815,6 +815,8 @@ dump(int verbose, int export_format)
 				c = dumpopt(c, "insecure_locks");
 			if (ep->e_flags & NFSEXP_NOACL)
 				c = dumpopt(c, "no_acl");
+			if (ep->e_flags & NFSEXP_PNFS)
+				c = dumpopt(c, "pnfs");
 			if (ep->e_flags & NFSEXP_FSID)
 				c = dumpopt(c, "fsid=%d", ep->e_fsid);
 			if (ep->e_uuid)
diff --git a/utils/exportfs/exports.man b/utils/exportfs/exports.man
index 3d974d9..59358e6 100644
--- a/utils/exportfs/exports.man
+++ b/utils/exportfs/exports.man
@@ -378,6 +378,15 @@ If the client asks for alternative locations for the export point, it
 will be given this list of alternatives. (Note that actual replication
 of the filesystem must be handled elsewhere.)
 
+.TP
+.IR pnfs
+This option allows enables the use of pNFS extension if protocol level
+is NFSv4.1 or higher, and the filesystem supports pNFS exports.  With
+pNFS clients can bypass the server and perform I/O directly to storage
+devices. The default can be explicitly requested with the
+.I no_pnfs
+option.
+
 .SS User ID Mapping
 .PP
 .B nfsd