Merge git://git.kernel.org/pub/scm/linux/kernel/git/agk/linux-2.6-dm

* git://git.kernel.org/pub/scm/linux/kernel/git/agk/linux-2.6-dm:
dm snapshot: fix on disk chunk size validation
dm exception store: split set_chunk_size
dm snapshot: fix header corruption race on invalidation
dm snapshot: refactor zero_disk_area to use chunk_io
dm log: userspace add luid to distinguish between concurrent log instances
dm raid1: do not allow log_failure variable to unset after being set
dm log: remove incorrect field from userspace table output
dm log: fix userspace status output
dm stripe: expose correct io hints
dm table: add more context to terse warning messages
dm table: fix queue_limit checking device iterator
dm snapshot: implement iterate devices
dm multipath: fix oops when request based io fails when no paths

drivers/md/dm-exception-store.c

···
 	 */
 	chunk_size_ulong = round_up(chunk_size_ulong, PAGE_SIZE >> 9);
 
+	return dm_exception_store_set_chunk_size(store, chunk_size_ulong,
+						 error);
+}
+
+int dm_exception_store_set_chunk_size(struct dm_exception_store *store,
+				      unsigned long chunk_size_ulong,
+				      char **error)
+{
 	/* Check chunk_size is a power of 2 */
 	if (!is_power_of_2(chunk_size_ulong)) {
 		*error = "Chunk size is not a power of 2";
···
 	/* Validate the chunk size against the device block size */
 	if (chunk_size_ulong % (bdev_logical_block_size(store->cow->bdev) >> 9)) {
 		*error = "Chunk size is not a multiple of device blocksize";
+		return -EINVAL;
+	}
+
+	if (chunk_size_ulong > INT_MAX >> SECTOR_SHIFT) {
+		*error = "Chunk size is too high";
 		return -EINVAL;
 	}
 

drivers/md/dm-exception-store.h

···
 int dm_exception_store_type_register(struct dm_exception_store_type *type);
 int dm_exception_store_type_unregister(struct dm_exception_store_type *type);
 
+int dm_exception_store_set_chunk_size(struct dm_exception_store *store,
+				      unsigned long chunk_size_ulong,
+				      char **error);
+
 int dm_exception_store_create(struct dm_target *ti, int argc, char **argv,
 			      unsigned *args_used,
 			      struct dm_exception_store **store);

drivers/md/dm-log-userspace-base.c

···
 	struct dm_target *ti;
 	uint32_t region_size;
 	region_t region_count;
+	uint64_t luid;
 	char uuid[DM_UUID_LEN];
 
 	char *usr_argv_str;
···
 	 * restored.
 	 */
 retry:
-	r = dm_consult_userspace(uuid, request_type, data,
+	r = dm_consult_userspace(uuid, lc->luid, request_type, data,
 				 data_size, rdata, rdata_size);
 
 	if (r != -ESRCH)
···
 		set_current_state(TASK_INTERRUPTIBLE);
 		schedule_timeout(2*HZ);
 		DMWARN("Attempting to contact userspace log server...");
-		r = dm_consult_userspace(uuid, DM_ULOG_CTR, lc->usr_argv_str,
+		r = dm_consult_userspace(uuid, lc->luid, DM_ULOG_CTR,
+					 lc->usr_argv_str,
 					 strlen(lc->usr_argv_str) + 1,
 					 NULL, NULL);
 		if (!r)
 			break;
 	}
 	DMINFO("Reconnected to userspace log server... DM_ULOG_CTR complete");
-	r = dm_consult_userspace(uuid, DM_ULOG_RESUME, NULL,
+	r = dm_consult_userspace(uuid, lc->luid, DM_ULOG_RESUME, NULL,
 				 0, NULL, NULL);
 	if (!r)
 		goto retry;
···
 		return -ENOMEM;
 	}
 
-	for (i = 0, str_size = 0; i < argc; i++)
-		str_size += sprintf(str + str_size, "%s ", argv[i]);
-	str_size += sprintf(str + str_size, "%llu",
-			    (unsigned long long)ti->len);
+	str_size = sprintf(str, "%llu", (unsigned long long)ti->len);
+	for (i = 0; i < argc; i++)
+		str_size += sprintf(str + str_size, " %s", argv[i]);
 
 	*ctr_str = str;
 	return str_size;
···
 		return -ENOMEM;
 	}
 
+	/* The ptr value is sufficient for local unique id */
+	lc->luid = (uint64_t)lc;
+
 	lc->ti = ti;
 
 	if (strlen(argv[0]) > (DM_UUID_LEN - 1)) {
···
 	}
 
 	/* Send table string */
-	r = dm_consult_userspace(lc->uuid, DM_ULOG_CTR,
+	r = dm_consult_userspace(lc->uuid, lc->luid, DM_ULOG_CTR,
 				 ctr_str, str_size, NULL, NULL);
 
 	if (r == -ESRCH) {
···
 
 	/* Since the region size does not change, get it now */
 	rdata_size = sizeof(rdata);
-	r = dm_consult_userspace(lc->uuid, DM_ULOG_GET_REGION_SIZE,
+	r = dm_consult_userspace(lc->uuid, lc->luid, DM_ULOG_GET_REGION_SIZE,
 				 NULL, 0, (char *)&rdata, &rdata_size);
 
 	if (r) {
···
 	int r;
 	struct log_c *lc = log->context;
 
-	r = dm_consult_userspace(lc->uuid, DM_ULOG_DTR,
+	r = dm_consult_userspace(lc->uuid, lc->luid, DM_ULOG_DTR,
 				 NULL, 0,
 				 NULL, NULL);
 
···
 	int r;
 	struct log_c *lc = log->context;
 
-	r = dm_consult_userspace(lc->uuid, DM_ULOG_PRESUSPEND,
+	r = dm_consult_userspace(lc->uuid, lc->luid, DM_ULOG_PRESUSPEND,
 				 NULL, 0,
 				 NULL, NULL);
 
···
 	int r;
 	struct log_c *lc = log->context;
 
-	r = dm_consult_userspace(lc->uuid, DM_ULOG_POSTSUSPEND,
+	r = dm_consult_userspace(lc->uuid, lc->luid, DM_ULOG_POSTSUSPEND,
 				 NULL, 0,
 				 NULL, NULL);
 
···
 	struct log_c *lc = log->context;
 
 	lc->in_sync_hint = 0;
-	r = dm_consult_userspace(lc->uuid, DM_ULOG_RESUME,
+	r = dm_consult_userspace(lc->uuid, lc->luid, DM_ULOG_RESUME,
 				 NULL, 0,
 				 NULL, NULL);
 
···
 			    char *result, unsigned maxlen)
 {
 	int r = 0;
+	char *table_args;
 	size_t sz = (size_t)maxlen;
 	struct log_c *lc = log->context;
 
···
 		break;
 	case STATUSTYPE_TABLE:
 		sz = 0;
-		DMEMIT("%s %u %s %s", log->type->name, lc->usr_argc + 1,
-		       lc->uuid, lc->usr_argv_str);
+		table_args = strstr(lc->usr_argv_str, " ");
+		BUG_ON(!table_args); /* There will always be a ' ' */
+		table_args++;
+
+		DMEMIT("%s %u %s %s ", log->type->name, lc->usr_argc,
+		       lc->uuid, table_args);
 		break;
 	}
 	return (r) ? 0 : (int)sz;

drivers/md/dm-log-userspace-transfer.c

···
 
 /**
  * dm_consult_userspace
- * @uuid: log's uuid (must be DM_UUID_LEN in size)
+ * @uuid: log's universal unique identifier (must be DM_UUID_LEN in size)
+ * @luid: log's local unique identifier
  * @request_type:  found in include/linux/dm-log-userspace.h
  * @data: data to tx to the server
  * @data_size: size of data in bytes
···
  *
  * Returns: 0 on success, -EXXX on failure
  **/
-int dm_consult_userspace(const char *uuid, int request_type,
+int dm_consult_userspace(const char *uuid, uint64_t luid, int request_type,
 			 char *data, size_t data_size,
 			 char *rdata, size_t *rdata_size)
 {
···
 
 	memset(tfr, 0, DM_ULOG_PREALLOCED_SIZE - overhead_size);
 	memcpy(tfr->uuid, uuid, DM_UUID_LEN);
+	tfr->luid = luid;
 	tfr->seq = dm_ulog_seq++;
 
 	/*

drivers/md/dm-log-userspace-transfer.h

···
 
 int dm_ulog_tfr_init(void);
 void dm_ulog_tfr_exit(void);
-int dm_consult_userspace(const char *uuid, int request_type,
+int dm_consult_userspace(const char *uuid, uint64_t luid, int request_type,
 			 char *data, size_t data_size,
 			 char *rdata, size_t *rdata_size);
 

drivers/md/dm-raid1.c

···
 	 */
 	dm_rh_inc_pending(ms->rh, &sync);
 	dm_rh_inc_pending(ms->rh, &nosync);
-	ms->log_failure = dm_rh_flush(ms->rh) ? 1 : 0;
+
+	/*
+	 * If the flush fails on a previous call and succeeds here,
+	 * we must not reset the log_failure variable.  We need
+	 * userspace interaction to do that.
+	 */
+	ms->log_failure = dm_rh_flush(ms->rh) ? 1 : ms->log_failure;
 
 	/*
 	 * Dispatch io.

drivers/md/dm-snap-persistent.c

···
 	void *zero_area;
 
 	/*
+	 * An area used for header. The header can be written
+	 * concurrently with metadata (when invalidating the snapshot),
+	 * so it needs a separate buffer.
+	 */
+	void *header_area;
+
+	/*
 	 * Used to keep track of which metadata area the data in
 	 * 'chunk' refers to.
 	 */
···
 	 */
 	ps->area = vmalloc(len);
 	if (!ps->area)
-		return r;
+		goto err_area;
 
 	ps->zero_area = vmalloc(len);
-	if (!ps->zero_area) {
-		vfree(ps->area);
-		return r;
-	}
+	if (!ps->zero_area)
+		goto err_zero_area;
 	memset(ps->zero_area, 0, len);
 
+	ps->header_area = vmalloc(len);
+	if (!ps->header_area)
+		goto err_header_area;
+
 	return 0;
+
+err_header_area:
+	vfree(ps->zero_area);
+
+err_zero_area:
+	vfree(ps->area);
+
+err_area:
+	return r;
 }
 
 static void free_area(struct pstore *ps)
···
 	if (ps->zero_area)
 		vfree(ps->zero_area);
 	ps->zero_area = NULL;
+
+	if (ps->header_area)
+		vfree(ps->header_area);
+	ps->header_area = NULL;
 }
 
 struct mdata_req {
···
 /*
  * Read or write a chunk aligned and sized block of data from a device.
  */
-static int chunk_io(struct pstore *ps, chunk_t chunk, int rw, int metadata)
+static int chunk_io(struct pstore *ps, void *area, chunk_t chunk, int rw,
+		    int metadata)
 {
 	struct dm_io_region where = {
 		.bdev = ps->store->cow->bdev,
···
 	struct dm_io_request io_req = {
 		.bi_rw = rw,
 		.mem.type = DM_IO_VMA,
-		.mem.ptr.vma = ps->area,
+		.mem.ptr.vma = area,
 		.client = ps->io_client,
 		.notify.fn = NULL,
 	};
···
 
 	chunk = area_location(ps, ps->current_area);
 
-	r = chunk_io(ps, chunk, rw, 0);
+	r = chunk_io(ps, ps->area, chunk, rw, 0);
 	if (r)
 		return r;
 
···
 
 static int zero_disk_area(struct pstore *ps, chunk_t area)
 {
-	struct dm_io_region where = {
-		.bdev = ps->store->cow->bdev,
-		.sector = ps->store->chunk_size * area_location(ps, area),
-		.count = ps->store->chunk_size,
-	};
-	struct dm_io_request io_req = {
-		.bi_rw = WRITE,
-		.mem.type = DM_IO_VMA,
-		.mem.ptr.vma = ps->zero_area,
-		.client = ps->io_client,
-		.notify.fn = NULL,
-	};
-
-	return dm_io(&io_req, 1, &where, NULL);
+	return chunk_io(ps, ps->zero_area, area_location(ps, area), WRITE, 0);
 }
 
 static int read_header(struct pstore *ps, int *new_snapshot)
···
 	struct disk_header *dh;
 	chunk_t chunk_size;
 	int chunk_size_supplied = 1;
+	char *chunk_err;
 
 	/*
 	 * Use default chunk size (or hardsect_size, if larger) if none supplied
···
 	if (r)
 		return r;
 
-	r = chunk_io(ps, 0, READ, 1);
+	r = chunk_io(ps, ps->header_area, 0, READ, 1);
 	if (r)
 		goto bad;
 
-	dh = (struct disk_header *) ps->area;
+	dh = ps->header_area;
 
 	if (le32_to_cpu(dh->magic) == 0) {
 		*new_snapshot = 1;
···
 	ps->version = le32_to_cpu(dh->version);
 	chunk_size = le32_to_cpu(dh->chunk_size);
 
-	if (!chunk_size_supplied || ps->store->chunk_size == chunk_size)
+	if (ps->store->chunk_size == chunk_size)
 		return 0;
 
-	DMWARN("chunk size %llu in device metadata overrides "
-	       "table chunk size of %llu.",
-	       (unsigned long long)chunk_size,
-	       (unsigned long long)ps->store->chunk_size);
+	if (chunk_size_supplied)
+		DMWARN("chunk size %llu in device metadata overrides "
+		       "table chunk size of %llu.",
+		       (unsigned long long)chunk_size,
+		       (unsigned long long)ps->store->chunk_size);
 
 	/* We had a bogus chunk_size. Fix stuff up. */
 	free_area(ps);
 
-	ps->store->chunk_size = chunk_size;
-	ps->store->chunk_mask = chunk_size - 1;
-	ps->store->chunk_shift = ffs(chunk_size) - 1;
+	r = dm_exception_store_set_chunk_size(ps->store, chunk_size,
+					      &chunk_err);
+	if (r) {
+		DMERR("invalid on-disk chunk size %llu: %s.",
+		      (unsigned long long)chunk_size, chunk_err);
+		return r;
+	}
 
 	r = dm_io_client_resize(sectors_to_pages(ps->store->chunk_size),
 				ps->io_client);
···
 {
 	struct disk_header *dh;
 
-	memset(ps->area, 0, ps->store->chunk_size << SECTOR_SHIFT);
+	memset(ps->header_area, 0, ps->store->chunk_size << SECTOR_SHIFT);
 
-	dh = (struct disk_header *) ps->area;
+	dh = ps->header_area;
 	dh->magic = cpu_to_le32(SNAP_MAGIC);
 	dh->valid = cpu_to_le32(ps->valid);
 	dh->version = cpu_to_le32(ps->version);
 	dh->chunk_size = cpu_to_le32(ps->store->chunk_size);
 
-	return chunk_io(ps, 0, WRITE, 1);
+	return chunk_io(ps, ps->header_area, 0, WRITE, 1);
 }
 
 /*
···
 	ps->valid = 1;
 	ps->version = SNAPSHOT_DISK_VERSION;
 	ps->area = NULL;
+	ps->zero_area = NULL;
+	ps->header_area = NULL;
 	ps->next_free = 2;	/* skipping the header and first area */
 	ps->current_committed = 0;
 

drivers/md/dm-snap.c

···
 	return 0;
 }
 
+static int snapshot_iterate_devices(struct dm_target *ti,
+				    iterate_devices_callout_fn fn, void *data)
+{
+	struct dm_snapshot *snap = ti->private;
+
+	return fn(ti, snap->origin, 0, ti->len, data);
+}
+
+
 /*-----------------------------------------------------------------
  * Origin methods
  *---------------------------------------------------------------*/
···
 	return 0;
 }
 
+static int origin_iterate_devices(struct dm_target *ti,
+				  iterate_devices_callout_fn fn, void *data)
+{
+	struct dm_dev *dev = ti->private;
+
+	return fn(ti, dev, 0, ti->len, data);
+}
+
 static struct target_type origin_target = {
 	.name    = "snapshot-origin",
-	.version = {1, 6, 0},
+	.version = {1, 7, 0},
 	.module  = THIS_MODULE,
 	.ctr     = origin_ctr,
 	.dtr     = origin_dtr,
 	.map     = origin_map,
 	.resume  = origin_resume,
 	.status  = origin_status,
+	.iterate_devices = origin_iterate_devices,
 };
 
 static struct target_type snapshot_target = {
 	.name    = "snapshot",
-	.version = {1, 6, 0},
+	.version = {1, 7, 0},
 	.module  = THIS_MODULE,
 	.ctr     = snapshot_ctr,
 	.dtr     = snapshot_dtr,
···
 	.end_io  = snapshot_end_io,
 	.resume  = snapshot_resume,
 	.status  = snapshot_status,
+	.iterate_devices = snapshot_iterate_devices,
 };
 
 static int __init dm_snapshot_init(void)

drivers/md/dm-stripe.c

···
 	return ret;
 }
 
+static void stripe_io_hints(struct dm_target *ti,
+			    struct queue_limits *limits)
+{
+	struct stripe_c *sc = ti->private;
+	unsigned chunk_size = (sc->chunk_mask + 1) << 9;
+
+	blk_limits_io_min(limits, chunk_size);
+	limits->io_opt = chunk_size * sc->stripes;
+}
+
 static struct target_type stripe_target = {
 	.name   = "striped",
-	.version = {1, 2, 0},
+	.version = {1, 3, 0},
 	.module = THIS_MODULE,
 	.ctr    = stripe_ctr,
 	.dtr    = stripe_dtr,
···
 	.end_io = stripe_end_io,
 	.status = stripe_status,
 	.iterate_devices = stripe_iterate_devices,
+	.io_hints = stripe_io_hints,
 };
 
 int __init dm_stripe_init(void)

drivers/md/dm-table.c

···
 }
 
 /*
- * If possible, this checks an area of a destination device is valid.
+ * If possible, this checks an area of a destination device is invalid.
  */
-static int device_area_is_valid(struct dm_target *ti, struct dm_dev *dev,
-				sector_t start, sector_t len, void *data)
+static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
+				  sector_t start, sector_t len, void *data)
 {
 	struct queue_limits *limits = data;
 	struct block_device *bdev = dev->bdev;
···
 	char b[BDEVNAME_SIZE];
 
 	if (!dev_size)
-		return 1;
+		return 0;
 
 	if ((start >= dev_size) || (start + len > dev_size)) {
-		DMWARN("%s: %s too small for target",
-		       dm_device_name(ti->table->md), bdevname(bdev, b));
-		return 0;
+		DMWARN("%s: %s too small for target: "
+		       "start=%llu, len=%llu, dev_size=%llu",
+		       dm_device_name(ti->table->md), bdevname(bdev, b),
+		       (unsigned long long)start,
+		       (unsigned long long)len,
+		       (unsigned long long)dev_size);
+		return 1;
 	}
 
 	if (logical_block_size_sectors <= 1)
-		return 1;
+		return 0;
 
 	if (start & (logical_block_size_sectors - 1)) {
 		DMWARN("%s: start=%llu not aligned to h/w "
-		       "logical block size %hu of %s",
+		       "logical block size %u of %s",
 		       dm_device_name(ti->table->md),
 		       (unsigned long long)start,
 		       limits->logical_block_size, bdevname(bdev, b));
-		return 0;
+		return 1;
 	}
 
 	if (len & (logical_block_size_sectors - 1)) {
 		DMWARN("%s: len=%llu not aligned to h/w "
-		       "logical block size %hu of %s",
+		       "logical block size %u of %s",
 		       dm_device_name(ti->table->md),
 		       (unsigned long long)len,
 		       limits->logical_block_size, bdevname(bdev, b));
-		return 0;
+		return 1;
 	}
 
-	return 1;
+	return 0;
 }
 
 /*
···
 	}
 
 	if (blk_stack_limits(limits, &q->limits, start << 9) < 0)
-		DMWARN("%s: target device %s is misaligned",
-		       dm_device_name(ti->table->md), bdevname(bdev, b));
+		DMWARN("%s: target device %s is misaligned: "
+		       "physical_block_size=%u, logical_block_size=%u, "
+		       "alignment_offset=%u, start=%llu",
+		       dm_device_name(ti->table->md), bdevname(bdev, b),
+		       q->limits.physical_block_size,
+		       q->limits.logical_block_size,
+		       q->limits.alignment_offset,
+		       (unsigned long long) start << 9);
+
 
 	/*
 	 * Check if merge fn is supported.
···
 
 	if (remaining) {
 		DMWARN("%s: table line %u (start sect %llu len %llu) "
-		       "not aligned to h/w logical block size %hu",
+		       "not aligned to h/w logical block size %u",
 		       dm_device_name(table->md), i,
 		       (unsigned long long) ti->begin,
 		       (unsigned long long) ti->len,
···
 		ti->type->iterate_devices(ti, dm_set_device_limits,
 					  &ti_limits);
 
+		/* Set I/O hints portion of queue limits */
+		if (ti->type->io_hints)
+			ti->type->io_hints(ti, &ti_limits);
+
 		/*
 		 * Check each device area is consistent with the target's
 		 * overall queue limits.
 		 */
-		if (!ti->type->iterate_devices(ti, device_area_is_valid,
-					       &ti_limits))
+		if (ti->type->iterate_devices(ti, device_area_is_invalid,
+					      &ti_limits))
 			return -EINVAL;
 
 combine_limits:

drivers/md/dm.c

···
 	dm_put(md);
 }
 
+static void free_rq_clone(struct request *clone)
+{
+	struct dm_rq_target_io *tio = clone->end_io_data;
+
+	blk_rq_unprep_clone(clone);
+	free_rq_tio(tio);
+}
+
 static void dm_unprep_request(struct request *rq)
 {
 	struct request *clone = rq->special;
-	struct dm_rq_target_io *tio = clone->end_io_data;
 
 	rq->special = NULL;
 	rq->cmd_flags &= ~REQ_DONTPREP;
 
-	blk_rq_unprep_clone(clone);
-	free_rq_tio(tio);
+	free_rq_clone(clone);
 }
 
 /*
···
 			rq->sense_len = clone->sense_len;
 	}
 
-	BUG_ON(clone->bio);
-	free_rq_tio(tio);
+	free_rq_clone(clone);
 
 	blk_end_request_all(rq, error);
 

include/linux/device-mapper.h

···
 				      iterate_devices_callout_fn fn,
 				      void *data);
 
+typedef void (*dm_io_hints_fn) (struct dm_target *ti,
+				struct queue_limits *limits);
+
 /*
  * Returns:
  *    0: The target can handle the next I/O immediately.
···
 	dm_merge_fn merge;
 	dm_busy_fn busy;
 	dm_iterate_devices_fn iterate_devices;
+	dm_io_hints_fn io_hints;
 
 	/* For internal device-mapper use. */
 	struct list_head list;

include/linux/dm-log-userspace.h

···
 	(DM_ULOG_REQUEST_MASK & (request_type))
 
 struct dm_ulog_request {
-	char uuid[DM_UUID_LEN]; /* Ties a request to a specific mirror log */
+	/*
+	 * The local unique identifier (luid) and the universally unique
+	 * identifier (uuid) are used to tie a request to a specific
+	 * mirror log.  A single machine log could probably make due with
+	 * just the 'luid', but a cluster-aware log must use the 'uuid' and
+	 * the 'luid'.  The uuid is what is required for node to node
+	 * communication concerning a particular log, but the 'luid' helps
+	 * differentiate between logs that are being swapped and have the
+	 * same 'uuid'.  (Think "live" and "inactive" device-mapper tables.)
+	 */
+	uint64_t luid;
+	char uuid[DM_UUID_LEN];
 	char padding[7];        /* Padding because DM_UUID_LEN = 129 */
 
 	int32_t error;          /* Used to report back processing errors */