Release 1.8.1 after some man page updates and other fixes.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Create.c

@@ -205,9 +205,9 @@ int Create(struct supertype *st, char *mddev, int mdfd,
 		}
 		else {
 			ldsize = dsize;
-			dsize <<= 9;
+			ldsize <<= 9;
 		}
-		freesize = st->ss->avail_size(ldsize);
+		freesize = st->ss->avail_size(ldsize >> 9);
 		if (freesize == 0) {
 			fprintf(stderr, Name ": %s is too small: %luK\n",
 				dname, (unsigned long)(ldsize>>10));

Makefile

@@ -62,7 +62,7 @@ OBJS =  mdadm.o config.o mdstat.o  ReadMe.o util.o Manage.o Assemble.o Build.o \
 SRCS =  mdadm.c config.c mdstat.c  ReadMe.c util.c Manage.c Assemble.c Build.c \
 	Create.c Detail.c Examine.c Grow.c Monitor.c dlink.c Kill.c Query.c mdopen.c super0.c super1.c
 
-ASSEMBLE_SRCS := mdassemble.c Assemble.c config.c dlink.c util.c
+ASSEMBLE_SRCS := mdassemble.c Assemble.c config.c dlink.c util.c super0.c super1.c
 ifdef MDASSEMBLE_AUTO
 ASSEMBLE_SRCS += mdopen.c mdstat.c
 ASSEMBLE_FLAGS = -DMDASSEMBLE_AUTO

md.4

@@ -16,7 +16,7 @@ Array of Independent Devices.
 .B md
 supports RAID levels 1 (mirroring) 4 (striped array with parity
 device), 5 (striped array with distributed parity information) and 6
-(striped array with distributed dual redundancy information.)  If a
+(striped array with distributed dual redundancy information.)  If
 some number of underlying devices fails while using one of these
 levels, the array will continue to function; this number is one for
 RAID levels 4 and 5, two for RAID level 6, and all but one (N-1) for
@@ -24,24 +24,42 @@ RAID level 1.
 .PP
 .B md
 also supports a number of pseudo RAID (non-redundant) configurations
-including RAID0 (striped array), LINEAR (catenated array) and
+including RAID0 (striped array), LINEAR (catenated array),
-MULTIPATH (a set of different interfaces to the same device).
+MULTIPATH (a set of different interfaces to the same device),
+and FAULTY (a layer over a single device into which errors can be injected).
 
 .SS MD SUPER BLOCK
-With the exception of Legacy Arrays described below, each device that
+Each device in an array may have a
-is incorporated into an MD array has a
+.I superblock
-.I super block
+which records information about the structure and state of the array.
-written towards the end of the device.  This superblock records
+This allows the array to be reliably re-assembled after a shutdown.
-information about the structure and state of the array so that the
-array can be reliably re-assembled after a shutdown.
 
-The superblock is 4K long and is written into a 64K aligned block that
+From Linux kernel version 2.6.10,
+.B md
+provides support for two different formats of this superblock, and
+other formats can be added.  Prior to this release, only one format is
+supported.
+
+The common format - known as version 0.90 - has
+a superblock that is 4K long and is written into a 64K aligned block that
 starts at least 64K and less than 128K from the end of the device
 (i.e. to get the address of the superblock round the size of the
 device down to a multiple of 64K and then subtract 64K).
 The available size of each device is the amount of space before the
 super block, so between 64K and 128K is lost when a device in
 incorporated into an MD array.
+This superblock stores multi-byte fields in a processor-dependant
+manner, so arrays cannot easily be moved between computers with
+different processors.
+
+The new format - known as version 1 - has a superblock that is
+normally 1K long, but can be longer.  It is normally stored between 8K
+and 12K from the end of the device, on a 4K boundary, though
+variations can be stored at the start of the device (version 1.1) or 4K from
+the start of the device (version 1.2).
+This superblock format stores multibyte data in a
+processor-independant format and has supports upto hundreds of
+component devices (version 0.90 only supports 28).
 
 The superblock contains, among other things:
 .TP
@@ -53,17 +71,34 @@ UUID
 a 128 bit Universally Unique Identifier that identifies the array that
 this device is part of.
 
-.SS LEGACY ARRAYS
+.SS ARRAYS WITHOUT SUPERBLOCKS
+While it is usually best to create arrays with superblocks so that
+they can be assembled reliably, there are some circumstances where an
+array without superblocks in preferred.  This include:
+.TP
+LEGACY ARRAYS
 Early versions of the
 .B md
-driver only supported Linear and Raid0 configurations and so
+driver only supported Linear and Raid0 configurations and did not use
-did not use an MD superblock (as there is no state that needs to be
+a superblock (which is less critical with these configurations).
-recorded).  While it is strongly recommended that all newly created
+While such arrays should be rebuilt with superblocks if possible,
-arrays utilise a superblock to help ensure that they are assembled
-properly, the
 .B md
-driver still supports legacy linear and raid0 md arrays that
+continues to support them.
-do not have a superblock.
+.TP
+FAULTY
+Being a largely transparent layer over a different device, the FAULTY
+personality doesn't gain anything from having a superblock.
+.TP
+MULTIPATH
+It is often possible to detect devices which are different paths to
+the same storage directly rather than having a distinctive superblock
+written to the device and searched for on all paths.  In this case,
+a MULTIPATH array with no superblock makes sense.
+.TP
+RAID1
+In some configurations it might be desired to create a raid1
+configuration that does use a superblock, and to maintain the state of
+the array elsewhere.  While not encouraged, this is supported.
 
 .SS LINEAR
 

mdadm.8

@@ -1,5 +1,5 @@
 .\" -*- nroff -*-
-.TH MDADM 8 "" v1.11.0
+.TH MDADM 8 "" v1.11.1
 .SH NAME
 mdadm \- manage MD devices
 .I aka
@@ -92,7 +92,7 @@ information so as to assemble a faulty array.
 
 .TP
 .B Build
-Build a legacy array without per-device superblocks.
+Build an array without per-device superblocks.
 
 .TP
 .B Create
@@ -234,6 +234,24 @@ in which case
 says to get a list of array devices from
 .BR /proc/mdstat .
 
+.TP
+.B -e ", " --metadata=
+Declare the style of superblock (raid metadata) to be used.  The
+default is 0.90 for --create, and to guess for other operations.
+
+Options are:
+.RS
+.IP "0, 0.90, default"
+Use the original 0.90 format superblock.  This format limits arrays to
+28 componenet devices and limits component devices of levels 1 and
+greater to 2 terabytes.
+.IP "1, 1.0, 1.1, 1.2"
+Use the new version-1 format superblock.  This has few restrictions.
+The different subversion store the superblock at different locations
+on the device, either at the end (for 1.0), at the start (for 1.1) or
+4K from the start (for 1.2).
+.RE
+
 .SH For create or build:
 
 .TP

mdadm.spec

@@ -1,6 +1,6 @@
 Summary:     mdadm is used for controlling Linux md devices (aka RAID arrays)
 Name:        mdadm
-Version:     1.11.0
+Version:     1.11.1
 Release:     1
 Source:      http://www.cse.unsw.edu.au/~neilb/source/mdadm/mdadm-%{version}.tgz
 URL:         http://www.cse.unsw.edu.au/~neilb/source/mdadm/

mdassemble.c

@@ -95,7 +95,7 @@ int main() {
 			if (ioctl(mdfd, GET_ARRAY_INFO, &array)>=0)
 				/* already assembled, skip */
 				continue;
-			rv |= Assemble(array_list->devname, mdfd,
+			rv |= Assemble(array_list->st, array_list->devname, mdfd,
 					   array_list, configfile,
 					   NULL,
 					   readonly, runstop, NULL, verbose, force);

super0.c

@@ -415,6 +415,7 @@ static int store_super0(int fd, void *sbv)
 	if (write(fd, super, sizeof(*super)) != sizeof(*super))
 		return 4;
 
+	fsync(fd);
 	return 0;
 }
 

super1.c

@@ -83,7 +83,9 @@ struct mdp_superblock_1 {
 	__u16	dev_roles[0];	/* role in array, or 0xffff for a spare, or 0xfffe for faulty */
 };
 
+#ifndef offsetof
 #define offsetof(t,f) ((int)&(((t*)0)->f))
+#endif
 static unsigned int calc_sb_1_csum(struct mdp_superblock_1 * sb)
 {
 	unsigned int disk_csum, csum;
@@ -139,11 +141,11 @@ static void examine_super1(void *sbv)
 	printf("     Raid Level : %s\n", c?c:"-unknown-");
 	printf("   Raid Devices : %d\n", __le32_to_cpu(sb->raid_disks));
 	printf("\n");
-	printf("    Device Size : %llu%s\n", sb->data_size, human_size(sb->data_size<<9));
+	printf("    Device Size : %llu%s\n", (unsigned long long)sb->data_size, human_size(sb->data_size<<9));
 	if (sb->data_offset)
-		printf("    Data Offset : %llu sectors\n", __le64_to_cpu(sb->data_offset));
+		printf("    Data Offset : %llu sectors\n", (unsigned long long)__le64_to_cpu(sb->data_offset));
 	if (sb->super_offset)
-		printf("   Super Offset : %llu sectors\n", __le64_to_cpu(sb->super_offset));
+		printf("   Super Offset : %llu sectors\n", (unsigned long long)__le64_to_cpu(sb->super_offset));
 	printf("    Device UUID : ");
 	for (i=0; i<16; i++) {
 		printf("%02x", sb->set_uuid[i]);
@@ -159,7 +161,7 @@ static void examine_super1(void *sbv)
 	else
 		printf("       Checksum : %x - expected %x\n", __le32_to_cpu(sb->sb_csum),
 		       __le32_to_cpu(calc_sb_1_csum(sb)));
-	printf("         Events : %llu\n", __le64_to_cpu(sb->events));
+	printf("         Events : %llu\n", (unsigned long long)__le64_to_cpu(sb->events));
 	printf("\n");
 	if (__le32_to_cpu(sb->level) == 5) {
 		c = map_num(r5layout, __le32_to_cpu(sb->layout));
@@ -235,7 +237,7 @@ static void detail_super1(void *sbv)
 		printf("%02x", sb->set_uuid[i]);
 		if ((i&3)==0 && i != 0) printf(":");
 	}
-	printf("\n         Events : %llu\n\n", __le64_to_cpu(sb->events));
+	printf("\n         Events : %llu\n\n", (unsigned long long)__le64_to_cpu(sb->events));
 }
 
 static void brief_detail_super1(void *sbv)
@@ -452,6 +454,7 @@ static int store_super1(int fd, void *sbv)
 	if (write(fd, sb, sbsize) != sbsize)
 		return 4;
 
+	fsync(fd);
 	return 0;
 }
 
@@ -575,15 +578,33 @@ static int load_super1(struct supertype *st, int fd, void **sbp, char *devname)
 
 
 	if (st->ss == NULL) {
-		/* guess... */
+		int bestvers = -1;
+		__u64 bestctime = 0;
+		/* guess... choose latest ctime */
 		st->ss = &super1;
 		for (st->minor_version = 0; st->minor_version <= 2 ; st->minor_version++) {
 			switch(load_super1(st, fd, sbp, devname)) {
-			case 0: return 0; /* good */
+			case 0: super = *sbp;
+				if (bestvers == -1 ||
+				    bestctime < __le64_to_cpu(super->ctime)) {
+					bestvers = st->minor_version;
+					bestctime = __le64_to_cpu(super->ctime);
+				}
+				free(super);
+				*sbp = NULL;
+				break;
 			case 1: st->ss = NULL; return 1; /*bad device */
 			case 2: break; /* bad, try next */
 			}
 		}
+		if (bestvers != -1) {
+			int rv;
+			st->minor_version = bestvers;
+			st->ss = &super1;
+			rv = load_super1(st, fd, sbp, devname);
+			if (rv) st->ss = NULL;
+			return rv;
+		}
 		st->ss = NULL;
 		return 2;
 	}

util.c

@@ -386,6 +386,15 @@ unsigned long calc_csum(void *super, int bytes)
 	for(i=0; i<bytes/4; i++)
 		newcsum+= superc[i];
 	csum = (newcsum& 0xffffffff) + (newcsum>>32);
+#ifdef __alpha__
+/* The in-kernel checksum calculation is always 16bit on 
+ * the alpha, though it is 32 bit on i386...
+ * I wonder what it is elsewhere... (it uses and API in
+ * a way that it shouldn't).
+ */
+	csum = (csum & 0xffff) + (csum >> 16);
+	csum = (csum & 0xffff) + (csum >> 16);
+#endif
 	return csum;
 }
 
@@ -544,6 +553,8 @@ struct supertype *guess_super(int fd)
 	 */
 	struct superswitch  *ss;
 	struct supertype *st;
+	unsigned long besttime = 0;
+	int bestsuper = -1;
 	
 	void *sbp = NULL;
 	int i;
@@ -554,10 +565,26 @@ struct supertype *guess_super(int fd)
 		int rv;
 		ss = superlist[i];
 		rv = ss->load_super(st, fd, &sbp, NULL);
+		if (rv == 0) {
+			struct mdinfo info;
+			ss->getinfo_super(&info, sbp);
+			if (bestsuper == -1 ||
+			    besttime < info.array.ctime) {
+				bestsuper = i;
+				besttime = info.array.ctime;
+				st->ss = NULL;
+			}
+			free(sbp);
+		}
+	}
+	if (bestsuper != -1) {
+		int rv;
+		rv = superlist[bestsuper]->load_super(st, fd, &sbp, NULL);
 		if (rv == 0) {
 			free(sbp);
 			return st;
 		}
 	}
+	free(st);
 	return NULL;
 }