X-Git-Url: http://git.ipfire.org/?a=blobdiff_plain;f=Assemble.c;h=9f75c68c9d49b9b2a9a7daf8201e1a6de4a71657;hb=757e55435997e355ee9b03e5d913b5496a3c39a8;hp=05ace561fb507bb6a557b3d15e6d35f18517400b;hpb=56bbc588f7f0f3bdd3ec23f02109b427c1d3b8f1;p=thirdparty%2Fmdadm.git

diff --git a/Assemble.c b/Assemble.c
index 05ace561..9f75c68c 100644
--- a/Assemble.c
+++ b/Assemble.c
@@ -1,7 +1,7 @@
 /*
  * mdadm - manage Linux "md" devices aka RAID arrays.
  *
- * Copyright (C) 2001-2013 Neil Brown <neilb@suse.de>
+ * Copyright (C) 2001-2016 Neil Brown <neilb@suse.com>
  *
  *
  *    This program is free software; you can redistribute it and/or modify
@@ -25,21 +25,27 @@
 #include	"mdadm.h"
 #include	<ctype.h>
 
-static int name_matches(char *found, char *required, char *homehost)
+static int name_matches(char *found, char *required, char *homehost, int require_homehost)
 {
 	/* See if the name found matches the required name, possibly
 	 * prefixed with 'homehost'
 	 */
-	char fnd[33];
+	char *sep;
+	unsigned int l;
 
-	strncpy(fnd, found, 32);
-	fnd[32] = 0;
 	if (strcmp(found, required)==0)
 		return 1;
-	if (homehost) {
-		int l = strlen(homehost);
-		if (l < 32 && fnd[l] == ':' &&
-		    strcmp(fnd+l+1, required)==0)
+	sep = strchr(found, ':');
+	if (!sep)
+		return 0;
+	l = sep - found;
+	if (strncmp(found, "any:", 4) == 0 ||
+	    (homehost && strcmp(homehost, "any") == 0) ||
+	    !require_homehost ||
+	    (homehost && strlen(homehost) == l &&
+	     strncmp(found, homehost, l) == 0)) {
+		/* matching homehost */
+		if (strcmp(sep+1, required) == 0)
 			return 1;
 	}
 	return 0;
@@ -73,7 +79,7 @@ static int is_member_busy(char *metadata_version)
 static int ident_matches(struct mddev_ident *ident,
 			 struct mdinfo *content,
 			 struct supertype *tst,
-			 char *homehost,
+			 char *homehost, int require_homehost,
 			 char *update, char *devname)
 {
 
@@ -85,7 +91,7 @@ static int ident_matches(struct mddev_ident *ident,
 		return 0;
 	}
 	if (ident->name[0] && (!update || strcmp(update, "name")!= 0) &&
-	    name_matches(content->name, ident->name, homehost)==0) {
+	    name_matches(content->name, ident->name, homehost, require_homehost)==0) {
 		if (devname)
 			pr_err("%s has wrong name.\n", devname);
 		return 0;
@@ -143,6 +149,7 @@ static int select_devices(struct mddev_dev *devlist,
 	struct mdinfo *content = NULL;
 	int report_mismatch = ((inargv && c->verbose >= 0) || c->verbose > 0);
 	struct domainlist *domains = NULL;
+	dev_t rdev;
 
 	tmpdev = devlist; num_devs = 0;
 	while (tmpdev) {
@@ -163,7 +170,6 @@ static int select_devices(struct mddev_dev *devlist,
 	     tmpdev = tmpdev ? tmpdev->next : NULL) {
 		char *devname = tmpdev->devname;
 		int dfd;
-		struct stat stb;
 		struct supertype *tst;
 		struct dev_policy *pol = NULL;
 		int found_container = 0;
@@ -198,14 +204,7 @@ static int select_devices(struct mddev_dev *devlist,
 				pr_err("cannot open device %s: %s\n",
 				       devname, strerror(errno));
 			tmpdev->used = 2;
-		} else if (fstat(dfd, &stb)< 0) {
-			/* Impossible! */
-			pr_err("fstat failed for %s: %s\n",
-			       devname, strerror(errno));
-			tmpdev->used = 2;
-		} else if ((stb.st_mode & S_IFMT) != S_IFBLK) {
-			pr_err("%s is not a block device.\n",
-			       devname);
+		} else if (!fstat_is_blkdev(dfd, devname, &rdev)) {
 			tmpdev->used = 2;
 		} else if (must_be_container(dfd)) {
 			if (st) {
@@ -216,25 +215,23 @@ static int select_devices(struct mddev_dev *devlist,
 					pr_err("%s is a container, but we are looking for components\n",
 					       devname);
 				tmpdev->used = 2;
-#if !defined(MDASSEMBLE) || defined(MDASSEMBLE) && defined(MDASSEMBLE_AUTO)
 			} if (!tst && (tst = super_by_fd(dfd, NULL)) == NULL) {
 				if (report_mismatch)
 					pr_err("not a recognisable container: %s\n",
 					       devname);
 				tmpdev->used = 2;
-#endif
-			} else if (!tst->ss->load_container
-				   || tst->ss->load_container(tst, dfd, NULL)) {
+			} else if (!tst->ss->load_container ||
+				   tst->ss->load_container(tst, dfd, NULL)) {
 				if (report_mismatch)
 					pr_err("no correct container type: %s\n",
 					       devname);
 				tmpdev->used = 2;
 			} else if (auto_assem &&
-				   !conf_test_metadata(tst->ss->name, (pol = devid_policy(stb.st_rdev)),
+				   !conf_test_metadata(tst->ss->name,
+						       (pol = devid_policy(rdev)),
 						       tst->ss->match_home(tst, c->homehost) == 1)) {
 				if (report_mismatch)
-					pr_err("%s has metadata type %s for which "
-					       "auto-assembly is disabled\n",
+					pr_err("%s has metadata type %s for which auto-assembly is disabled\n",
 					       devname, tst->ss->name);
 				tmpdev->used = 2;
 			} else
@@ -245,7 +242,9 @@ static int select_devices(struct mddev_dev *devlist,
 					pr_err("no recogniseable superblock on %s\n",
 					       devname);
 				tmpdev->used = 2;
-			} else if (tst->ss->load_super(tst,dfd, NULL)) {
+			} else if ((tst->ignore_hw_compat = 0),
+				   tst->ss->load_super(tst, dfd,
+						       report_mismatch ? devname : NULL)) {
 				if (report_mismatch)
 					pr_err("no RAID superblock on %s\n",
 					       devname);
@@ -256,11 +255,11 @@ static int select_devices(struct mddev_dev *devlist,
 					       tst->ss->name, devname);
 				tmpdev->used = 2;
 			} else if (auto_assem && st == NULL &&
-				   !conf_test_metadata(tst->ss->name, (pol = devid_policy(stb.st_rdev)),
+				   !conf_test_metadata(tst->ss->name,
+						       (pol = devid_policy(rdev)),
 						       tst->ss->match_home(tst, c->homehost) == 1)) {
 				if (report_mismatch)
-					pr_err("%s has metadata type %s for which "
-					       "auto-assembly is disabled\n",
+					pr_err("%s has metadata type %s for which auto-assembly is disabled\n",
 					       devname, tst->ss->name);
 				tmpdev->used = 2;
 			}
@@ -282,6 +281,8 @@ static int select_devices(struct mddev_dev *devlist,
 				st->ss->free_super(st);
 			dev_policy_free(pol);
 			domain_free(domains);
+			if (tst)
+				tst->ss->free_super(tst);
 			return -1;
 		}
 
@@ -327,7 +328,8 @@ static int select_devices(struct mddev_dev *devlist,
 			     content = content->next) {
 
 				if (!ident_matches(ident, content, tst,
-						   c->homehost, c->update,
+						   c->homehost, c->require_homehost,
+						   c->update,
 						   report_mismatch ? devname : NULL))
 					/* message already printed */;
 				else if (is_member_busy(content->text_version)) {
@@ -350,8 +352,7 @@ static int select_devices(struct mddev_dev *devlist,
 
 			st = tst; tst = NULL;
 			if (!auto_assem && inargv && tmpdev->next != NULL) {
-				pr_err("%s is a container, but is not "
-				       "only device given: confused and aborting\n",
+				pr_err("%s is a container, but is not only device given: confused and aborting\n",
 				       devname);
 				st->ss->free_super(st);
 				dev_policy_free(pol);
@@ -366,36 +367,43 @@ static int select_devices(struct mddev_dev *devlist,
 			tmpdev = NULL;
 			goto loop;
 		} else {
-			int rv = 0;
-			struct mddev_ident *match;
-
 			content = *contentp;
 			tst->ss->getinfo_super(tst, content, NULL);
 
 			if (!ident_matches(ident, content, tst,
-					   c->homehost, c->update,
+					   c->homehost, c->require_homehost,
+					   c->update,
 					   report_mismatch ? devname : NULL))
 				goto loop;
 
-			match = conf_match(tst, content, devname,
-					   report_mismatch ? c->verbose : -1,
-					   &rv);
-			if (!match && rv == 2)
-				goto loop;
-			if (match && match->devname &&
-			    strcasecmp(match->devname, "<ignore>") == 0) {
-				if (report_mismatch)
-					pr_err("%s is a member of an explicitly ignored array\n",
-					       devname);
-				goto loop;
-			}
-			if (match && !ident_matches(match, content, tst,
-						    c->homehost, c->update,
-						    report_mismatch ? devname : NULL))
-				/* Array exists  in mdadm.conf but some
-				 * details don't match, so reject it
+			if (auto_assem) {
+				/* Never auto-assemble things that conflict
+				 * with mdadm.conf in some way
 				 */
-				goto loop;
+				struct mddev_ident *match;
+				int rv = 0;
+
+				match = conf_match(tst, content, devname,
+						   report_mismatch ? c->verbose : -1,
+						   &rv);
+				if (!match && rv == 2)
+					goto loop;
+				if (match && match->devname &&
+				    strcasecmp(match->devname, "<ignore>") == 0) {
+					if (report_mismatch)
+						pr_err("%s is a member of an explicitly ignored array\n",
+						       devname);
+					goto loop;
+				}
+				if (match && !ident_matches(match, content, tst,
+							    c->homehost, c->require_homehost,
+							    c->update,
+							    report_mismatch ? devname : NULL))
+					/* Array exists  in mdadm.conf but some
+					 * details don't match, so reject it
+					 */
+					goto loop;
+			}
 
 			/* should be safe to try an exclusive open now, we
 			 * have rejected anything that some other mdadm might
@@ -473,7 +481,7 @@ static int select_devices(struct mddev_dev *devlist,
 		/* Collect domain information from members only */
 		if (tmpdev && tmpdev->used == 1) {
 			if (!pol)
-				pol = devid_policy(stb.st_rdev);
+				pol = devid_policy(rdev);
 			domain_merge(&domains, pol, tst?tst->ss->name:NULL);
 		}
 		dev_policy_free(pol);
@@ -506,15 +514,12 @@ static int select_devices(struct mddev_dev *devlist,
 
 	/* Now reject spares that don't match domains of identified members */
 	for (tmpdev = devlist; tmpdev; tmpdev = tmpdev->next) {
-		struct stat stb;
 		if (tmpdev->used != 3)
 			continue;
-		if (stat(tmpdev->devname, &stb)< 0) {
-			pr_err("fstat failed for %s: %s\n",
-			       tmpdev->devname, strerror(errno));
+		if (!stat_is_blkdev(tmpdev->devname, &rdev)) {
 			tmpdev->used = 2;
 		} else {
-			struct dev_policy *pol = devid_policy(stb.st_rdev);
+			struct dev_policy *pol = devid_policy(rdev);
 			int dt = domain_test(domains, pol, NULL);
 			if (inargv && dt != 0)
 				/* take this spare as domains match
@@ -561,9 +566,7 @@ static int load_devices(struct devs *devices, char *devmap,
 	struct mddev_dev *tmpdev;
 	int devcnt = 0;
 	int nextspare = 0;
-#ifndef MDASSEMBLE
 	int bitmap_done = 0;
-#endif
 	int most_recent = -1;
 	int bestcnt = 0;
 	int *best = *bestp;
@@ -574,30 +577,30 @@ static int load_devices(struct devs *devices, char *devmap,
 		struct stat stb;
 		struct supertype *tst;
 		int i;
+		int dfd;
+		int disk_state;
 
 		if (tmpdev->used != 1)
 			continue;
 		/* looks like a good enough match to update the super block if needed */
-#ifndef MDASSEMBLE
 		if (c->update) {
-			int dfd;
 			/* prepare useful information in info structures */
 			struct stat stb2;
 			int err;
 			fstat(mdfd, &stb2);
 
-			if (strcmp(c->update, "uuid")==0 &&
-			    !ident->uuid_set) {
-				int rfd;
-				if ((rfd = open("/dev/urandom", O_RDONLY)) < 0 ||
-				    read(rfd, ident->uuid, 16) != 16) {
-					*(__u32*)(ident->uuid) = random();
-					*(__u32*)(ident->uuid+1) = random();
-					*(__u32*)(ident->uuid+2) = random();
-					*(__u32*)(ident->uuid+3) = random();
-				}
-				if (rfd >= 0) close(rfd);
+			if (strcmp(c->update, "uuid") == 0 && !ident->uuid_set)
+				random_uuid((__u8 *)ident->uuid);
+
+			if (strcmp(c->update, "ppl") == 0 &&
+			    ident->bitmap_fd >= 0) {
+				pr_err("PPL is not compatible with bitmap\n");
+				close(mdfd);
+				free(devices);
+				free(devmap);
+				return -1;
 			}
+
 			dfd = dev_open(devname,
 				       tmpdev->disposition == 'I'
 				       ? O_RDWR : (O_RDWR|O_EXCL));
@@ -624,6 +627,19 @@ static int load_devices(struct devs *devices, char *devmap,
 
 			if (strcmp(c->update, "byteorder") == 0)
 				err = 0;
+			else if (strcmp(c->update, "home-cluster") == 0) {
+				tst->cluster_name = c->homecluster;
+				err = tst->ss->write_bitmap(tst, dfd, NameUpdate);
+			} else if (strcmp(c->update, "nodes") == 0) {
+				tst->nodes = c->nodes;
+				err = tst->ss->write_bitmap(tst, dfd, NodeNumUpdate);
+			} else if (strcmp(c->update, "revert-reshape") == 0 &&
+				   c->invalid_backup)
+				err = tst->ss->update_super(tst, content,
+							    "revert-reshape-nobackup",
+							    devname, c->verbose,
+							    ident->uuid_set,
+							    c->homehost);
 			else
 				err = tst->ss->update_super(tst, content, c->update,
 							    devname, c->verbose,
@@ -631,8 +647,7 @@ static int load_devices(struct devs *devices, char *devmap,
 							    c->homehost);
 			if (err < 0) {
 				if (err == -1)
-					pr_err("--update=%s not understood"
-					       " for %s metadata\n",
+					pr_err("--update=%s not understood for %s metadata\n",
 					       c->update, tst->ss->name);
 				tst->ss->free_super(tst);
 				free(tst);
@@ -651,7 +666,6 @@ static int load_devices(struct devs *devices, char *devmap,
 			if (tst->ss->store_super(tst, dfd))
 				pr_err("Could not re-write superblock on %s.\n",
 				       devname);
-			close(dfd);
 
 			if (strcmp(c->update, "uuid")==0 &&
 			    ident->bitmap_fd >= 0 && !bitmap_done) {
@@ -662,12 +676,10 @@ static int load_devices(struct devs *devices, char *devmap,
 				else
 					bitmap_done = 1;
 			}
-		} else
-#endif
-		{
-			int dfd = dev_open(devname,
-					   tmpdev->disposition == 'I'
-					   ? O_RDWR : (O_RDWR|O_EXCL));
+		} else {
+			dfd = dev_open(devname,
+				       tmpdev->disposition == 'I'
+				       ? O_RDWR : (O_RDWR|O_EXCL));
 			tst = dup_super(st);
 
 			if (dfd < 0 || tst->ss->load_super(tst, dfd, NULL) != 0) {
@@ -684,10 +696,10 @@ static int load_devices(struct devs *devices, char *devmap,
 				return -1;
 			}
 			tst->ss->getinfo_super(tst, content, devmap + devcnt * content->array.raid_disks);
-			close(dfd);
 		}
 
-		stat(devname, &stb);
+		fstat(dfd, &stb);
+		close(dfd);
 
 		if (c->verbose > 0)
 			pr_err("%s is identified as a member of %s, slot %d%s.\n",
@@ -700,7 +712,9 @@ static int load_devices(struct devs *devices, char *devmap,
 		devices[devcnt].i.disk.major = major(stb.st_rdev);
 		devices[devcnt].i.disk.minor = minor(stb.st_rdev);
 
-		if (devices[devcnt].i.disk.state == 6) {
+		disk_state = devices[devcnt].i.disk.state & ~((1<<MD_DISK_FAILFAST) |
+							      (1<<MD_DISK_WRITEMOSTLY));
+		if (disk_state == ((1<<MD_DISK_ACTIVE) | (1<<MD_DISK_SYNC))) {
 			if (most_recent < 0 ||
 			    devices[devcnt].i.events
 			    > devices[most_recent].i.events) {
@@ -718,7 +732,7 @@ static int load_devices(struct devs *devices, char *devmap,
 			i = devcnt;
 		else
 			i = devices[devcnt].i.disk.raid_disk;
-		if (i+1 == 0) {
+		if (i+1 == 0 || i == MD_DISK_ROLE_JOURNAL) {
 			if (nextspare < content->array.raid_disks*2)
 				nextspare = content->array.raid_disks*2;
 			i = nextspare++;
@@ -747,22 +761,19 @@ static int load_devices(struct devs *devices, char *devmap,
 				bestcnt = newbestcnt;
 			}
 			if (best[i] >=0 &&
-			    devices[best[i]].i.events
-			    == devices[devcnt].i.events
-			    && (devices[best[i]].i.disk.minor
-				!= devices[devcnt].i.disk.minor)
-			    && st->ss == &super0
-			    && content->array.level != LEVEL_MULTIPATH) {
+			    devices[best[i]].i.events ==
+			    devices[devcnt].i.events &&
+			    (devices[best[i]].i.disk.minor !=
+			     devices[devcnt].i.disk.minor) &&
+			    st->ss == &super0 &&
+			    content->array.level != LEVEL_MULTIPATH) {
 				/* two different devices with identical superblock.
 				 * Could be a mis-detection caused by overlapping
 				 * partitions.  fail-safe.
 				 */
-				pr_err("WARNING %s and %s appear"
-				       " to have very similar superblocks.\n"
-				       "      If they are really different, "
-				       "please --zero the superblock on one\n"
-				       "      If they are the same or overlap,"
-				       " please remove one from %s.\n",
+				pr_err("WARNING %s and %s appear to have very similar superblocks.\n"
+				       "      If they are really different, please --zero the superblock on one\n"
+				       "      If they are the same or overlap, please remove one from %s.\n",
 				       devices[best[i]].devname, devname,
 				       inargv ? "the list" :
 				       "the\n      DEVICE list in mdadm.conf"
@@ -773,10 +784,11 @@ static int load_devices(struct devs *devices, char *devmap,
 				*stp = st;
 				return -1;
 			}
-			if (best[i] == -1
-			    || (devices[best[i]].i.events
-				< devices[devcnt].i.events))
+			if (best[i] == -1 || (devices[best[i]].i.events
+					      < devices[devcnt].i.events))
 				best[i] = devcnt;
+			else if (st->ss == &super_imsm)
+				best[i+1] = devcnt;
 		}
 		devcnt++;
 	}
@@ -798,14 +810,11 @@ static int force_array(struct mdinfo *content,
 	int okcnt = 0;
 	while (!enough(content->array.level, content->array.raid_disks,
 		       content->array.layout, 1,
-		       avail)
-	       ||
+		       avail) ||
 	       (content->reshape_active && content->delta_disks > 0 &&
 		!enough(content->array.level, (content->array.raid_disks
 					       - content->delta_disks),
-			content->new_layout, 1,
-			avail)
-		       )) {
+			content->new_layout, 1, avail))) {
 		/* Choose the newest best drive which is
 		 * not up-to-date, update the superblock
 		 * and add it.
@@ -820,12 +829,49 @@ static int force_array(struct mdinfo *content,
 		     i < content->array.raid_disks * 2 && i < bestcnt;
 		     i += 2) {
 			int j = best[i];
-			if (j>=0 &&
-			    !devices[j].uptodate &&
-			    devices[j].i.recovery_start == MaxSector &&
-			    (chosen_drive < 0 ||
+			if (j < 0)
+				continue;
+			if (devices[j].uptodate)
+				continue;
+			if (devices[j].i.recovery_start != MaxSector) {
+				int delta;
+				if (!devices[j].i.reshape_active ||
+				    devices[j].i.delta_disks <= 0)
+					continue;
+				/* When increasing number of devices, an
+				 * added device also appears to be
+				 * recovering.  It is safe to include it
+				 * as long as it won't be a source of
+				 * data.
+				 * For now, just allow for last data
+				 * devices in RAID4 or last devices in RAID4/5/6.
+				 */
+				delta = devices[j].i.delta_disks;
+				if (devices[j].i.array.level >= 4 &&
+				    devices[j].i.array.level <= 6 &&
+				    i/2 >= content->array.raid_disks - delta)
+					/* OK */;
+				else if (devices[j].i.array.level == 4 &&
+					 i/2 >= content->array.raid_disks - delta - 1)
+					/* OK */;
+				else
+					continue;
+			} else if (devices[j].i.reshape_active !=
+			    content->reshape_active ||
+			    (devices[j].i.reshape_active &&
+			    devices[j].i.reshape_progress !=
+			    content->reshape_progress))
+				/* Here, it may be a source of data. If two
+				 * devices claim different progresses, it
+				 * means that reshape boundaries differ for
+				 * their own devices. Kernel will only treat
+				 * the first one as reshape progress and
+				 * go on. It may cause disaster, so avoid it.
+				 */
+				continue;
+			if (chosen_drive < 0 ||
 			     devices[j].i.events
-			     > devices[chosen_drive].i.events))
+			    > devices[chosen_drive].i.events)
 				chosen_drive = j;
 		}
 		if (chosen_drive < 0)
@@ -874,7 +920,6 @@ static int force_array(struct mdinfo *content,
 		avail[chosen_drive] = 1;
 		okcnt++;
 		tst->ss->free_super(tst);
-
 		/* If there are any other drives of the same vintage,
 		 * add them in as well.  We can't lose and we might gain
 		 */
@@ -885,7 +930,13 @@ static int force_array(struct mdinfo *content,
 			if (j >= 0 &&
 			    !devices[j].uptodate &&
 			    devices[j].i.recovery_start == MaxSector &&
-			    devices[j].i.events == current_events) {
+			    devices[j].i.events == current_events &&
+			    ((!devices[j].i.reshape_active &&
+			    !content->reshape_active) ||
+			    (devices[j].i.reshape_active ==
+			    content->reshape_active &&
+			    devices[j].i.reshape_progress ==
+			    content->reshape_progress))) {
 				chosen_drive = j;
 				goto add_another;
 			}
@@ -905,6 +956,7 @@ static int start_array(int mdfd,
 		       unsigned int okcnt,
 		       unsigned int sparecnt,
 		       unsigned int rebuilding_cnt,
+		       unsigned int journalcnt,
 		       struct context *c,
 		       int clean, char *avail,
 		       int start_partial_ok,
@@ -916,6 +968,18 @@ static int start_array(int mdfd,
 	int i;
 	unsigned int req_cnt;
 
+	if (content->journal_device_required && (content->journal_clean == 0)) {
+		if (!c->force) {
+			pr_err("Not safe to assemble with missing or stale journal device, consider --force.\n");
+			return 1;
+		}
+		pr_err("Journal is missing or stale, starting array read only.\n");
+		c->readonly = 1;
+	}
+
+	if (content->consistency_policy == CONSISTENCY_POLICY_PPL)
+		clean = 1;
+
 	rv = set_array_info(mdfd, st, content);
 	if (rv && !err_ok) {
 		pr_err("failed to set array info for %s: %s\n",
@@ -944,7 +1008,7 @@ static int start_array(int mdfd,
 	}
 
 	/* First, add the raid disks, but add the chosen one last */
-	for (i=0; i<= bestcnt; i++) {
+	for (i = 0; i <= bestcnt; i++) {
 		int j;
 		if (i < bestcnt) {
 			j = best[i];
@@ -954,8 +1018,9 @@ static int start_array(int mdfd,
 			j = chosen_drive;
 
 		if (j >= 0 && !devices[j].included) {
-			int dfd = dev_open(devices[j].devname,
-					   O_RDWR|O_EXCL);
+			int dfd;
+
+			dfd = dev_open(devices[j].devname, O_RDWR|O_EXCL);
 			if (dfd >= 0) {
 				remove_partitions(dfd);
 				close(dfd);
@@ -963,44 +1028,64 @@ static int start_array(int mdfd,
 			rv = add_disk(mdfd, st, content, &devices[j].i);
 
 			if (rv) {
-				pr_err("failed to add "
-				       "%s to %s: %s\n",
-				       devices[j].devname,
-				       mddev,
+				pr_err("failed to add %s to %s: %s\n",
+				       devices[j].devname, mddev,
 				       strerror(errno));
-				if (i < content->array.raid_disks * 2
-				    || i == bestcnt)
+				if (i < content->array.raid_disks * 2 ||
+				    i == bestcnt)
 					okcnt--;
 				else
 					sparecnt--;
-			} else if (c->verbose > 0)
+			} else if (c->verbose > 0) {
 				pr_err("added %s to %s as %d%s%s\n",
 				       devices[j].devname, mddev,
 				       devices[j].i.disk.raid_disk,
 				       devices[j].uptodate?"":
 				       " (possibly out of date)",
-				       (devices[j].i.disk.state & (1<<MD_DISK_REPLACEMENT))?" replacement":"");
+				       (devices[j].i.disk.state &
+					(1<<MD_DISK_REPLACEMENT)) ?
+				       " replacement":"");
+			}
 		} else if (j >= 0) {
 			if (c->verbose > 0)
 				pr_err("%s is already in %s as %d\n",
 				       devices[j].devname, mddev,
 				       devices[j].i.disk.raid_disk);
-		} else if (c->verbose > 0 && i < content->array.raid_disks*2
-			   && (i&1) == 0)
+		} else if (c->verbose > 0 &&
+			   i < content->array.raid_disks * 2 && (i & 1) == 0)
 			pr_err("no uptodate device for slot %d of %s\n",
-			       i, mddev);
+			       i/2, mddev);
 	}
 
 	if (content->array.level == LEVEL_CONTAINER) {
 		if (c->verbose >= 0) {
-			pr_err("Container %s has been "
-			       "assembled with %d drive%s",
-			       mddev, okcnt+sparecnt, okcnt+sparecnt==1?"":"s");
+			pr_err("Container %s has been assembled with %d drive%s",
+			       mddev, okcnt + sparecnt + journalcnt,
+			       okcnt + sparecnt + journalcnt == 1 ? "" : "s");
 			if (okcnt < (unsigned)content->array.raid_disks)
 				fprintf(stderr, " (out of %d)",
 					content->array.raid_disks);
 			fprintf(stderr, "\n");
 		}
+
+		if (st->ss->validate_container) {
+			struct mdinfo *devices_list;
+			struct mdinfo *info_devices;
+			unsigned int count;
+
+			devices_list = NULL;
+			info_devices = xmalloc(sizeof(struct mdinfo) *
+					       (okcnt + sparecnt));
+			for (count = 0; count < okcnt + sparecnt; count++) {
+				info_devices[count] = devices[count].i;
+				info_devices[count].next = devices_list;
+				devices_list = &info_devices[count];
+			}
+			if (st->ss->validate_container(devices_list))
+				pr_err("Mismatch detected!\n");
+			free(info_devices);
+		}
+
 		st->ss->free_super(st);
 		sysfs_uevent(content, "change");
 		if (err_ok && okcnt < (unsigned)content->array.raid_disks)
@@ -1017,17 +1102,16 @@ static int start_array(int mdfd,
 
 	if (c->runstop == 1 ||
 	    (c->runstop <= 0 &&
-	     ( enough(content->array.level, content->array.raid_disks,
-		      content->array.layout, clean, avail) &&
-	       (okcnt + rebuilding_cnt >= req_cnt || start_partial_ok)
-		     ))) {
+	     (enough(content->array.level, content->array.raid_disks,
+		     content->array.layout, clean, avail) &&
+	       (okcnt + rebuilding_cnt >= req_cnt || start_partial_ok)))) {
 		/* This array is good-to-go.
 		 * If a reshape is in progress then we might need to
 		 * continue monitoring it.  In that case we start
 		 * it read-only and let the grow code make it writable.
 		 */
 		int rv;
-#ifndef MDASSEMBLE
+
 		if (content->reshape_active &&
 		    !(content->reshape_active & RESHAPE_NO_BACKUP) &&
 		    content->delta_disks <= 0) {
@@ -1044,15 +1128,14 @@ static int start_array(int mdfd,
 					   "array_state", "readonly");
 			if (rv == 0)
 				rv = Grow_continue(mdfd, st, content,
-						   c->backup_file,
+						   c->backup_file, 0,
 						   c->freeze_reshape);
 		} else if (c->readonly &&
-			   sysfs_attribute_available(
-				   content, NULL, "array_state")) {
+			   sysfs_attribute_available(content, NULL,
+						     "array_state")) {
 			rv = sysfs_set_str(content, NULL,
 					   "array_state", "readonly");
 		} else
-#endif
 			rv = ioctl(mdfd, RUN_ARRAY, NULL);
 		reopen_mddev(mdfd); /* drop O_EXCL */
 		if (rv == 0) {
@@ -1060,11 +1143,19 @@ static int start_array(int mdfd,
 				pr_err("%s has been started with %d drive%s",
 				       mddev, okcnt, okcnt==1?"":"s");
 				if (okcnt < (unsigned)content->array.raid_disks)
-					fprintf(stderr, " (out of %d)", content->array.raid_disks);
+					fprintf(stderr, " (out of %d)",
+						content->array.raid_disks);
 				if (rebuilding_cnt)
-					fprintf(stderr, "%s %d rebuilding", sparecnt?",":" and", rebuilding_cnt);
+					fprintf(stderr, "%s %d rebuilding",
+						sparecnt?",":" and",
+						rebuilding_cnt);
 				if (sparecnt)
-					fprintf(stderr, " and %d spare%s", sparecnt, sparecnt==1?"":"s");
+					fprintf(stderr, " and %d spare%s",
+						sparecnt,
+						sparecnt == 1 ? "" : "s");
+				if (content->journal_clean)
+					fprintf(stderr, " and %d journal",
+						journalcnt);
 				fprintf(stderr, ".\n");
 			}
 			if (content->reshape_active &&
@@ -1073,12 +1164,19 @@ static int start_array(int mdfd,
 				/* might need to increase the size
 				 * of the stripe cache - default is 256
 				 */
-				if (256 < 4 * (content->array.chunk_size/4096)) {
-					struct mdinfo *sra = sysfs_read(mdfd, NULL, 0);
+				int chunk_size = content->array.chunk_size;
+
+				if (content->reshape_active &&
+				    content->new_chunk > chunk_size)
+					chunk_size = content->new_chunk;
+				if (256 < 4 * ((chunk_size+4065)/4096)) {
+					struct mdinfo *sra;
+
+					sra = sysfs_read(mdfd, NULL, 0);
 					if (sra)
 						sysfs_set_num(sra, NULL,
 							      "stripe_cache_size",
-							      (4 * content->array.chunk_size / 4096) + 1);
+							      (4 * chunk_size / 4096) + 1);
 					sysfs_free(sra);
 				}
 			}
@@ -1107,7 +1205,9 @@ static int start_array(int mdfd,
 			if (content->array.level == 6 &&
 			    okcnt + 1 == (unsigned)content->array.raid_disks &&
 			    was_forced) {
-				struct mdinfo *sra = sysfs_read(mdfd, NULL, 0);
+				struct mdinfo *sra;
+
+				sra = sysfs_read(mdfd, NULL, 0);
 				if (sra)
 					sysfs_set_str(sra, NULL,
 						      "sync_action", "repair");
@@ -1115,50 +1215,47 @@ static int start_array(int mdfd,
 			}
 			return 0;
 		}
-		pr_err("failed to RUN_ARRAY %s: %s\n",
-		       mddev, strerror(errno));
+		pr_err("failed to RUN_ARRAY %s: %s\n", mddev, strerror(errno));
 
 		if (!enough(content->array.level, content->array.raid_disks,
 			    content->array.layout, 1, avail))
-			pr_err("Not enough devices to "
-			       "start the array.\n");
+			pr_err("Not enough devices to start the array.\n");
 		else if (!enough(content->array.level,
 				 content->array.raid_disks,
-				 content->array.layout, clean,
-				 avail))
-			pr_err("Not enough devices to "
-			       "start the array while not clean "
-			       "- consider --force.\n");
+				 content->array.layout, clean, avail))
+			pr_err("Not enough devices to start the array while not clean - consider --force.\n");
 
 		return 1;
 	}
 	if (c->runstop == -1) {
 		pr_err("%s assembled from %d drive%s",
-		       mddev, okcnt, okcnt==1?"":"s");
+		       mddev, okcnt, okcnt == 1 ? "" : "s");
 		if (okcnt != (unsigned)content->array.raid_disks)
-			fprintf(stderr, " (out of %d)", content->array.raid_disks);
+			fprintf(stderr, " (out of %d)",
+				content->array.raid_disks);
 		fprintf(stderr, ", but not started.\n");
 		return 2;
 	}
 	if (c->verbose >= -1) {
-		pr_err("%s assembled from %d drive%s", mddev, okcnt, okcnt==1?"":"s");
+		pr_err("%s assembled from %d drive%s",
+		       mddev, okcnt, okcnt == 1 ? "" : "s");
 		if (rebuilding_cnt)
-			fprintf(stderr, "%s %d rebuilding", sparecnt?",":" and", rebuilding_cnt);
+			fprintf(stderr, "%s %d rebuilding",
+				sparecnt ? "," : " and", rebuilding_cnt);
 		if (sparecnt)
-			fprintf(stderr, " and %d spare%s", sparecnt, sparecnt==1?"":"s");
+			fprintf(stderr, " and %d spare%s", sparecnt,
+				sparecnt == 1 ? "" : "s");
 		if (!enough(content->array.level, content->array.raid_disks,
 			    content->array.layout, 1, avail))
 			fprintf(stderr, " - not enough to start the array.\n");
 		else if (!enough(content->array.level,
 				 content->array.raid_disks,
-				 content->array.layout, clean,
-				 avail))
-			fprintf(stderr, " - not enough to start the "
-				"array while not clean - consider "
-				"--force.\n");
+				 content->array.layout, clean, avail))
+			fprintf(stderr, " - not enough to start the array while not clean - consider --force.\n");
 		else {
 			if (req_cnt == (unsigned)content->array.raid_disks)
-				fprintf(stderr, " - need all %d to start it", req_cnt);
+				fprintf(stderr, " - need all %d to start it",
+					req_cnt);
 			else
 				fprintf(stderr, " - need %d to start", req_cnt);
 			fprintf(stderr, " (use --run to insist).\n");
@@ -1226,18 +1323,19 @@ int Assemble(struct supertype *st, char *mddev,
 	 *    START_ARRAY
 	 *
 	 */
-	int rv;
-	int mdfd;
+	int rv = -1;
+	int mdfd = -1;
 	int clean;
 	int auto_assem = (mddev == NULL && !ident->uuid_set &&
-			  ident->super_minor == UnSet && ident->name[0] == 0
-			  && (ident->container == NULL || ident->member == NULL));
-	struct devs *devices;
+			  ident->super_minor == UnSet && ident->name[0] == 0 &&
+			  (ident->container == NULL || ident->member == NULL));
+	struct devs *devices = NULL;
 	char *devmap;
 	int *best = NULL; /* indexed by raid_disk */
 	int bestcnt = 0;
 	int devcnt;
-	unsigned int okcnt, sparecnt, rebuilding_cnt, replcnt;
+	unsigned int okcnt, sparecnt, rebuilding_cnt, replcnt, journalcnt;
+	int journal_clean = 0;
 	int i;
 	int was_forced = 0;
 	int most_recent = 0;
@@ -1293,7 +1391,7 @@ try_again:
 		       mddev ? mddev : "further assembly");
 
 	content = &info;
-	if (st)
+	if (st && c->force)
 		st->ignore_hw_compat = 1;
 	num_devs = select_devices(devlist, ident, &st, &content, c,
 				  inargv, auto_assem);
@@ -1317,7 +1415,10 @@ try_again:
 	 */
 	if (map_lock(&map))
 		pr_err("failed to get exclusive lock on mapfile - continue anyway...\n");
-	mp = map_by_uuid(&map, content->uuid);
+	if (c->update && strcmp(c->update,"uuid") == 0)
+		mp = NULL;
+	else
+		mp = map_by_uuid(&map, content->uuid);
 	if (mp) {
 		struct mdinfo *dv;
 		/* array already exists. */
@@ -1326,7 +1427,7 @@ try_again:
 			pr_err("Found some drive for an array that is already active: %s\n",
 			       mp->path);
 			pr_err("giving up.\n");
-			return 1;
+			goto out;
 		}
 		for (dv = pre_exist->devs; dv; dv = dv->next) {
 			/* We want to add this device to our list,
@@ -1393,23 +1494,15 @@ try_again:
 			name = strchr(name, ':')+1;
 
 		mdfd = create_mddev(mddev, name, ident->autof, trustworthy,
-				    chosen_name);
+				    chosen_name, 0);
 	}
 	if (mdfd < 0) {
 		st->ss->free_super(st);
 		if (auto_assem)
 			goto try_again;
-		return 1;
+		goto out;
 	}
 	mddev = chosen_name;
-	if (get_linux_version() < 2004000 ||
-	    md_get_version(mdfd) < 9000) {
-		pr_err("Assemble requires Linux 2.4 or later, and\n"
-		       "     md driver version 0.90.0 or later.\n"
-		       "    Upgrade your kernel or try --build\n");
-		close(mdfd);
-		return 1;
-	}
 	if (pre_exist == NULL) {
 		if (mddev_busy(fd2devnm(mdfd))) {
 			pr_err("%s already active, cannot restart it!\n",
@@ -1426,13 +1519,12 @@ try_again:
 			st->ss->free_super(st);
 			if (auto_assem)
 				goto try_again;
-			return 1;
+			goto out;
 		}
 		/* just incase it was started but has no content */
 		ioctl(mdfd, STOP_ARRAY, NULL);
 	}
 
-#ifndef MDASSEMBLE
 	if (content != &info) {
 		/* This is a member of a container.  Try starting the array. */
 		int err;
@@ -1441,25 +1533,30 @@ try_again:
 		close(mdfd);
 		return err;
 	}
-#endif
+
 	/* Ok, no bad inconsistancy, we can try updating etc */
 	devices = xcalloc(num_devs, sizeof(*devices));
 	devmap = xcalloc(num_devs, content->array.raid_disks);
 	devcnt = load_devices(devices, devmap, ident, &st, devlist,
 			      c, content, mdfd, mddev,
 			      &most_recent, &bestcnt, &best, inargv);
-	if (devcnt < 0)
-		return 1;
+	if (devcnt < 0) {
+		mdfd = -3;
+		/*
+		 * devices is already freed in load_devices, so set devices
+		 * to NULL to avoid double free devices.
+		 */
+		devices = NULL;
+		goto out;
+	}
 
 	if (devcnt == 0) {
 		pr_err("no devices found for %s\n",
 		       mddev);
 		if (st)
 			st->ss->free_super(st);
-		close(mdfd);
-		free(devices);
 		free(devmap);
-		return 1;
+		goto out;
 	}
 
 	if (c->update && strcmp(c->update, "byteorder")==0)
@@ -1475,6 +1572,7 @@ try_again:
 	okcnt = 0;
 	replcnt = 0;
 	sparecnt=0;
+	journalcnt=0;
 	rebuilding_cnt=0;
 	for (i=0; i< bestcnt; i++) {
 		int j = best[i];
@@ -1485,8 +1583,13 @@ try_again:
 		/* note: we ignore error flags in multipath arrays
 		 * as they don't make sense
 		 */
-		if (content->array.level != LEVEL_MULTIPATH)
-			if (!(devices[j].i.disk.state & (1<<MD_DISK_ACTIVE))) {
+		if (content->array.level != LEVEL_MULTIPATH) {
+			if (devices[j].i.disk.state & (1<<MD_DISK_JOURNAL)) {
+				if (content->journal_device_required)
+					journalcnt++;
+				else	/* unexpected journal, mark as faulty */
+					devices[j].i.disk.state |= (1<<MD_DISK_FAULTY);
+			} else if (!(devices[j].i.disk.state & (1<<MD_DISK_ACTIVE))) {
 				if (!(devices[j].i.disk.state
 				      & (1<<MD_DISK_FAULTY))) {
 					devices[j].uptodate = 1;
@@ -1494,6 +1597,7 @@ try_again:
 				}
 				continue;
 			}
+		}
 		/* If this device thinks that 'most_recent' has failed, then
 		 * we must reject this device.
 		 */
@@ -1517,12 +1621,12 @@ try_again:
 		    devices[most_recent].i.events
 			) {
 			devices[j].uptodate = 1;
+			if (devices[j].i.disk.state & (1<<MD_DISK_JOURNAL))
+				journal_clean = 1;
 			if (i < content->array.raid_disks * 2) {
 				if (devices[j].i.recovery_start == MaxSector ||
 				    (content->reshape_active &&
-				     ((i >= content->array.raid_disks - content->delta_disks) ||
-				      (i >= content->array.raid_disks - content->delta_disks - 1
-				       && content->array.level == 4)))) {
+				     i >= content->array.raid_disks - content->delta_disks)) {
 					if (!avail[i/2]) {
 						okcnt++;
 						avail[i/2]=1;
@@ -1530,7 +1634,7 @@ try_again:
 						replcnt++;
 				} else
 					rebuilding_cnt++;
-			} else
+			} else if (devices[j].i.disk.raid_disk != MD_DISK_ROLE_JOURNAL)
 				sparecnt++;
 		}
 	}
@@ -1566,43 +1670,46 @@ try_again:
 				 : (O_RDONLY|O_EXCL)))< 0) {
 			pr_err("Cannot open %s: %s\n",
 			       devices[j].devname, strerror(errno));
-			close(mdfd);
-			free(devices);
-			return 1;
+			goto out;
 		}
 		if (st->ss->load_super(st,fd, NULL)) {
 			close(fd);
 			pr_err("RAID superblock has disappeared from %s\n",
 			       devices[j].devname);
-			close(mdfd);
-			free(devices);
-			return 1;
+			goto out;
 		}
 		close(fd);
 	}
 	if (st->sb == NULL) {
 		pr_err("No suitable drives found for %s\n", mddev);
-		close(mdfd);
-		free(devices);
-		return 1;
+		goto out;
 	}
 	st->ss->getinfo_super(st, content, NULL);
-#ifndef MDASSEMBLE
-	sysfs_init(content, mdfd, NULL);
-#endif
+	if (sysfs_init(content, mdfd, NULL)) {
+		pr_err("Unable to initialize sysfs\n");
+		goto out;
+	}
+
+	/* after reload context, store journal_clean in context */
+	content->journal_clean = journal_clean;
 	for (i=0; i<bestcnt; i++) {
 		int j = best[i];
 		unsigned int desired_state;
 
-		if (i >= content->array.raid_disks * 2)
+		if (j < 0)
+			continue;
+		if (devices[j].i.disk.raid_disk == MD_DISK_ROLE_JOURNAL)
+			desired_state = (1<<MD_DISK_JOURNAL);
+		else if (i >= content->array.raid_disks * 2)
 			desired_state = 0;
 		else if (i & 1)
 			desired_state = (1<<MD_DISK_ACTIVE) | (1<<MD_DISK_REPLACEMENT);
 		else
 			desired_state = (1<<MD_DISK_ACTIVE) | (1<<MD_DISK_SYNC);
 
-		if (j<0)
-			continue;
+		desired_state |= devices[j].i.disk.state & ((1<<MD_DISK_FAILFAST) |
+							    (1<<MD_DISK_WRITEMOSTLY));
+
 		if (!devices[j].uptodate)
 			continue;
 
@@ -1649,17 +1756,13 @@ try_again:
 		if (fd < 0) {
 			pr_err("Could not open %s for write - cannot Assemble array.\n",
 			       devices[chosen_drive].devname);
-			close(mdfd);
-			free(devices);
-			return 1;
+			goto out;
 		}
 		if (st->ss->store_super(st, fd)) {
 			close(fd);
 			pr_err("Could not re-write superblock on %s\n",
 			       devices[chosen_drive].devname);
-			close(mdfd);
-			free(devices);
-			return 1;
+			goto out;
 		}
 		if (c->verbose >= 0)
 			pr_err("Marking array %s as 'clean'\n",
@@ -1671,15 +1774,15 @@ try_again:
 	 * that was moved aside due to the reshape overwriting live data
 	 * The code of doing this lives in Grow.c
 	 */
-#ifndef MDASSEMBLE
 	if (content->reshape_active &&
 	    !(content->reshape_active & RESHAPE_NO_BACKUP)) {
 		int err = 0;
 		int *fdlist = xmalloc(sizeof(int)* bestcnt);
 		if (c->verbose > 0)
-			pr_err(":%s has an active reshape - checking "
-			       "if critical section needs to be restored\n",
+			pr_err("%s has an active reshape - checking if critical section needs to be restored\n",
 			       chosen_name);
+		if (!c->backup_file)
+			c->backup_file = locate_backup(content->sys_name);
 		enable_fds(bestcnt/2);
 		for (i = 0; i < bestcnt/2; i++) {
 			int j = best[i*2];
@@ -1704,8 +1807,7 @@ try_again:
 						   c->backup_file, c->verbose > 0);
 			if (err && c->invalid_backup) {
 				if (c->verbose > 0)
-					pr_err("continuing"
-					       " without restoring backup\n");
+					pr_err("continuing without restoring backup\n");
 				err = 0;
 			}
 		}
@@ -1718,12 +1820,9 @@ try_again:
 			pr_err("Failed to restore critical section for reshape, sorry.\n");
 			if (c->backup_file == NULL)
 				cont_err("Possibly you needed to specify the --backup-file\n");
-			close(mdfd);
-			free(devices);
-			return err;
+			goto out;
 		}
 	}
-#endif
 
 	/* Almost ready to actually *do* something */
 	/* First, fill in the map, so that udev can find our name
@@ -1741,7 +1840,7 @@ try_again:
 	rv = start_array(mdfd, mddev, content,
 			 st, ident, best, bestcnt,
 			 chosen_drive, devices, okcnt, sparecnt,
-			 rebuilding_cnt,
+			 rebuilding_cnt, journalcnt,
 			 c,
 			 clean, avail, start_partial_ok,
 			 pre_exist != NULL,
@@ -1749,6 +1848,7 @@ try_again:
 	if (rv == 1 && !pre_exist)
 		ioctl(mdfd, STOP_ARRAY, NULL);
 	free(devices);
+out:
 	map_unlock(&map);
 	if (rv == 0) {
 		wait_for(chosen_name, mdfd);
@@ -1779,39 +1879,43 @@ try_again:
 				usecs <<= 1;
 			}
 		}
-	} else
+	} else if (mdfd >= 0)
 		close(mdfd);
 
 	/* '2' means 'OK, but not started yet' */
+	if (rv == -1) {
+		free(devices);
+		return 1;
+	}
 	return rv == 2 ? 0 : rv;
 }
 
-#ifndef MDASSEMBLE
 int assemble_container_content(struct supertype *st, int mdfd,
 			       struct mdinfo *content, struct context *c,
 			       char *chosen_name, int *result)
 {
-	struct mdinfo *dev, *sra;
+	struct mdinfo *dev, *sra, *dev2;
 	int working = 0, preexist = 0;
 	int expansion = 0;
-	struct map_ent *map = NULL;
 	int old_raid_disks;
 	int start_reshape;
-	char *avail = NULL;
+	char *avail;
 	int err;
 
-	sysfs_init(content, mdfd, NULL);
+	if (sysfs_init(content, mdfd, NULL)) {
+		pr_err("Unable to initialize sysfs\n");
+		return 1;
+	}
 
-	sra = sysfs_read(mdfd, NULL, GET_VERSION);
+	sra = sysfs_read(mdfd, NULL, GET_VERSION|GET_DEVS);
 	if (sra == NULL || strcmp(sra->text_version, content->text_version) != 0) {
 		if (content->array.major_version == -1 &&
 		    content->array.minor_version == -2 &&
 		    c->readonly &&
 		    content->text_version[0] == '/')
 			content->text_version[0] = '-';
-		if (sysfs_set_array(content, md_get_version(mdfd)) != 0) {
-			if (sra)
-				sysfs_free(sra);
+		if (sysfs_set_array(content, 9003) != 0) {
+			sysfs_free(sra);
 			return 1;
 		}
 	}
@@ -1829,8 +1933,22 @@ int assemble_container_content(struct supertype *st, int mdfd,
 	if (st->ss->external && content->recovery_blocked && start_reshape)
 		block_subarray(content);
 
-	if (sra)
-		sysfs_free(sra);
+	for (dev2 = sra->devs; dev2; dev2 = dev2->next) {
+		for (dev = content->devs; dev; dev = dev->next)
+			if (dev2->disk.major == dev->disk.major &&
+			    dev2->disk.minor == dev->disk.minor)
+				break;
+		if (dev)
+			continue;
+		/* Don't want this one any more */
+		if (sysfs_set_str(sra, dev2, "slot", "none") < 0 &&
+		    errno == EBUSY) {
+			pr_err("Cannot remove old device %s: not updating %s\n", dev2->sys_name, sra->sys_name);
+			sysfs_free(sra);
+			return 1;
+		}
+		sysfs_set_str(sra, dev2, "state", "remove");
+	}
 	old_raid_disks = content->array.raid_disks - content->delta_disks;
 	avail = xcalloc(content->array.raid_disks, 1);
 	for (dev = content->devs; dev; dev = dev->next) {
@@ -1845,14 +1963,62 @@ int assemble_container_content(struct supertype *st, int mdfd,
 		} else if (errno == EEXIST)
 			preexist++;
 	}
+	sysfs_free(sra);
 	if (working + expansion == 0 && c->runstop <= 0) {
 		free(avail);
 		return 1;/* Nothing new, don't try to start */
 	}
-	map_update(&map, fd2devnm(mdfd),
-		   content->text_version,
+	map_update(NULL, fd2devnm(mdfd), content->text_version,
 		   content->uuid, chosen_name);
 
+	if (content->consistency_policy == CONSISTENCY_POLICY_PPL &&
+	    st->ss->validate_ppl) {
+		content->array.state |= 1;
+		err = 0;
+
+		for (dev = content->devs; dev; dev = dev->next) {
+			int dfd;
+			char *devpath;
+			int ret;
+
+			ret = st->ss->validate_ppl(st, content, dev);
+			if (ret == 0)
+				continue;
+
+			if (ret < 0) {
+				err = 1;
+				break;
+			}
+
+			if (!c->force) {
+				pr_err("%s contains invalid PPL - consider --force or --update-subarray with --update=no-ppl\n",
+					chosen_name);
+				content->array.state &= ~1;
+				avail[dev->disk.raid_disk] = 0;
+				break;
+			}
+
+			/* have --force - overwrite the invalid ppl */
+			devpath = map_dev(dev->disk.major, dev->disk.minor, 0);
+			dfd = dev_open(devpath, O_RDWR);
+			if (dfd < 0) {
+				pr_err("Failed to open %s\n", devpath);
+				err = 1;
+				break;
+			}
+
+			err = st->ss->write_init_ppl(st, content, dfd);
+			close(dfd);
+
+			if (err)
+				break;
+		}
+
+		if (err) {
+			free(avail);
+			return err;
+		}
+	}
 
 	if (enough(content->array.level, content->array.raid_disks,
 		   content->array.layout, content->array.state & 1, avail) == 0) {
@@ -1892,7 +2058,7 @@ int assemble_container_content(struct supertype *st, int mdfd,
 		int spare = content->array.raid_disks + expansion;
 		if (restore_backup(st, content,
 				   working,
-				   spare, c->backup_file, c->verbose) == 1)
+				   spare, &c->backup_file, c->verbose) == 1)
 			return 1;
 
 		err = sysfs_set_str(content, NULL,
@@ -1910,7 +2076,7 @@ int assemble_container_content(struct supertype *st, int mdfd,
 		}
 
 		err = Grow_continue(mdfd, st, content, c->backup_file,
-				    c->freeze_reshape);
+				    0, c->freeze_reshape);
 	} else switch(content->array.level) {
 		case LEVEL_LINEAR:
 		case LEVEL_MULTIPATH:
@@ -1966,4 +2132,3 @@ int assemble_container_content(struct supertype *st, int mdfd,
 	return err;
 	/* FIXME should have an O_EXCL and wait for read-auto */
 }
-#endif