/*

 * CDDL HEADER START

 *

 * The contents of this file are subject to the terms of the

 * Common Development and Distribution License (the "License").

 * You may not use this file except in compliance with the License.

 *

 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE

 * or http://www.opensolaris.org/os/licensing.

 * See the License for the specific language governing permissions

 * and limitations under the License.

 *

 * When distributing Covered Code, include this CDDL HEADER in each

 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.

 * If applicable, add the following below this CDDL HEADER, with the

 * fields enclosed by brackets "[]" replaced with your own identifying

 * information: Portions Copyright [yyyy] [name of copyright owner]

 *

 * CDDL HEADER END

 */

/*

 * Copyright 2008 Sun Microsystems, Inc.  All rights reserved.

 * Use is subject to license terms.

 */

#pragma ident	"%Z%%M%	%I%	%E% SMI"

/*

 * Pool import support functions.

 *

 * To import a pool, we rely on reading the configuration information from the

 * ZFS label of each device.  If we successfully read the label, then we

 * organize the configuration information in the following hierarchy:

 *

 * 	pool guid -> toplevel vdev guid -> label txg

 *

 * Duplicate entries matching this same tuple will be discarded.  Once we have

 * examined every device, we pick the best label txg config for each toplevel

 * vdev.  We then arrange these toplevel vdevs into a complete pool config, and

 * update any paths that have changed.  Finally, we attempt to import the pool

 * using our derived config, and record the results.

 */

#include <devid.h>

#include <dirent.h>

#include <errno.h>

#include <libintl.h>

#include <stdlib.h>

#include <string.h>

#include <sys/stat.h>

#include <unistd.h>

#include <fcntl.h>

#include <sys/vdev_impl.h>

#include "libzfs.h"

#include "libzfs_impl.h"

/*

 * Intermediate structures used to gather configuration information.

 */

typedef struct config_entry {

	uint64_t		ce_txg;

	nvlist_t		*ce_config;

	struct config_entry	*ce_next;

} config_entry_t;

typedef struct vdev_entry {

	uint64_t		ve_guid;

	config_entry_t		*ve_configs;

	struct vdev_entry	*ve_next;

} vdev_entry_t;

typedef struct pool_entry {

	uint64_t		pe_guid;

	vdev_entry_t		*pe_vdevs;

	struct pool_entry	*pe_next;

} pool_entry_t;

typedef struct name_entry {

	char			*ne_name;

	uint64_t		ne_guid;

	struct name_entry	*ne_next;

} name_entry_t;

typedef struct pool_list {

	pool_entry_t		*pools;

	name_entry_t		*names;

} pool_list_t;

static char *

get_devid(const char *path)

{

	int fd;

	ddi_devid_t devid;

	char *minor, *ret;

	if ((fd = open(path, O_RDONLY)) < 0)

		return (NULL);

	minor = NULL;

	ret = NULL;

	if (devid_get(fd, &devid) == 0) {

		if (devid_get_minor_name(fd, &minor) == 0)

			ret = devid_str_encode(devid, minor);

		if (minor != NULL)

			devid_str_free(minor);

		devid_free(devid);

	}

	(void) close(fd);

	return (ret);

}

/*

 * Go through and fix up any path and/or devid information for the given vdev

 * configuration.

 */

static int

fix_paths(nvlist_t *nv, name_entry_t *names)

{

	nvlist_t **child;

	uint_t c, children;

	uint64_t guid;

	name_entry_t *ne, *best;

	char *path, *devid;

	int matched;

	if (nvlist_lookup_nvlist_array(nv, ZPOOL_CONFIG_CHILDREN,

	    &child, &children) == 0) {

		for (c = 0; c < children; c++)

			if (fix_paths(child[c], names) != 0)

				return (-1);

		return (0);

	}

	/*

	 * This is a leaf (file or disk) vdev.  In either case, go through

	 * the name list and see if we find a matching guid.  If so, replace

	 * the path and see if we can calculate a new devid.

	 *

	 * There may be multiple names associated with a particular guid, in

	 * which case we have overlapping slices or multiple paths to the same

	 * disk.  If this is the case, then we want to pick the path that is

	 * the most similar to the original, where "most similar" is the number

	 * of matching characters starting from the end of the path.  This will

	 * preserve slice numbers even if the disks have been reorganized, and

	 * will also catch preferred disk names if multiple paths exist.

	 */

	verify(nvlist_lookup_uint64(nv, ZPOOL_CONFIG_GUID, &guid) == 0);

	if (nvlist_lookup_string(nv, ZPOOL_CONFIG_PATH, &path) != 0)

		path = NULL;

	matched = 0;

	best = NULL;

	for (ne = names; ne != NULL; ne = ne->ne_next) {

		if (ne->ne_guid == guid) {

			const char *src, *dst;

			int count;

			if (path == NULL) {

				best = ne;

				break;

			}

			src = ne->ne_name + strlen(ne->ne_name) - 1;

			dst = path + strlen(path) - 1;

			for (count = 0; src >= ne->ne_name && dst >= path;

			    src--, dst--, count++)

				if (*src != *dst)

					break;

			/*

			 * At this point, 'count' is the number of characters

			 * matched from the end.

			 */

			if (count > matched || best == NULL) {

				best = ne;

				matched = count;

			}

		}

	}

	if (best == NULL)

		return (0);

	if (nvlist_add_string(nv, ZPOOL_CONFIG_PATH, best->ne_name) != 0)

		return (-1);

	if ((devid = get_devid(best->ne_name)) == NULL) {

		(void) nvlist_remove_all(nv, ZPOOL_CONFIG_DEVID);

	} else {

		if (nvlist_add_string(nv, ZPOOL_CONFIG_DEVID, devid) != 0)

			return (-1);

		devid_str_free(devid);

	}

	return (0);

}

/*

 * Add the given configuration to the list of known devices.

 */

static int

add_config(libzfs_handle_t *hdl, pool_list_t *pl, const char *path,

    nvlist_t *config)

{

	uint64_t pool_guid, vdev_guid, top_guid, txg, state;

	pool_entry_t *pe;

	vdev_entry_t *ve;

	config_entry_t *ce;

	name_entry_t *ne;

	/*

	 * If this is a hot spare not currently in use or level 2 cache

	 * device, add it to the list of names to translate, but don't do

	 * anything else.

	 */

	if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_POOL_STATE,

	    &state) == 0 &&

	    (state == POOL_STATE_SPARE || state == POOL_STATE_L2CACHE) &&

	    nvlist_lookup_uint64(config, ZPOOL_CONFIG_GUID, &vdev_guid) == 0) {

		if ((ne = zfs_alloc(hdl, sizeof (name_entry_t))) == NULL)

			return (-1);

		if ((ne->ne_name = zfs_strdup(hdl, path)) == NULL) {

			free(ne);

			return (-1);

		}

		ne->ne_guid = vdev_guid;

		ne->ne_next = pl->names;

		pl->names = ne;

		return (0);

	}

	/*

	 * If we have a valid config but cannot read any of these fields, then

	 * it means we have a half-initialized label.  In vdev_label_init()

	 * we write a label with txg == 0 so that we can identify the device

	 * in case the user refers to the same disk later on.  If we fail to

	 * create the pool, we'll be left with a label in this state

	 * which should not be considered part of a valid pool.

	 */

	if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_POOL_GUID,

	    &pool_guid) != 0 ||

	    nvlist_lookup_uint64(config, ZPOOL_CONFIG_GUID,

	    &vdev_guid) != 0 ||

	    nvlist_lookup_uint64(config, ZPOOL_CONFIG_TOP_GUID,

	    &top_guid) != 0 ||

	    nvlist_lookup_uint64(config, ZPOOL_CONFIG_POOL_TXG,

	    &txg) != 0 || txg == 0) {

		nvlist_free(config);

		return (0);

	}

	/*

	 * First, see if we know about this pool.  If not, then add it to the

	 * list of known pools.

	 */

	for (pe = pl->pools; pe != NULL; pe = pe->pe_next) {

		if (pe->pe_guid == pool_guid)

			break;

	}

	if (pe == NULL) {

		if ((pe = zfs_alloc(hdl, sizeof (pool_entry_t))) == NULL) {

			nvlist_free(config);

			return (-1);

		}

		pe->pe_guid = pool_guid;

		pe->pe_next = pl->pools;

		pl->pools = pe;

	}

	/*

	 * Second, see if we know about this toplevel vdev.  Add it if its

	 * missing.

	 */

	for (ve = pe->pe_vdevs; ve != NULL; ve = ve->ve_next) {

		if (ve->ve_guid == top_guid)

			break;

	}

	if (ve == NULL) {

		if ((ve = zfs_alloc(hdl, sizeof (vdev_entry_t))) == NULL) {

			nvlist_free(config);

			return (-1);

		}

		ve->ve_guid = top_guid;

		ve->ve_next = pe->pe_vdevs;

		pe->pe_vdevs = ve;

	}

	/*

	 * Third, see if we have a config with a matching transaction group.  If

	 * so, then we do nothing.  Otherwise, add it to the list of known

	 * configs.

	 */

	for (ce = ve->ve_configs; ce != NULL; ce = ce->ce_next) {

		if (ce->ce_txg == txg)

			break;

	}

	if (ce == NULL) {

		if ((ce = zfs_alloc(hdl, sizeof (config_entry_t))) == NULL) {

			nvlist_free(config);

			return (-1);

		}

		ce->ce_txg = txg;

		ce->ce_config = config;

		ce->ce_next = ve->ve_configs;

		ve->ve_configs = ce;

	} else {

		nvlist_free(config);

	}

	/*

	 * At this point we've successfully added our config to the list of

	 * known configs.  The last thing to do is add the vdev guid -> path

	 * mappings so that we can fix up the configuration as necessary before

	 * doing the import.

	 */

	if ((ne = zfs_alloc(hdl, sizeof (name_entry_t))) == NULL)

		return (-1);

	if ((ne->ne_name = zfs_strdup(hdl, path)) == NULL) {

		free(ne);

		return (-1);

	}

	ne->ne_guid = vdev_guid;

	ne->ne_next = pl->names;

	pl->names = ne;

	return (0);

}

/*

 * Returns true if the named pool matches the given GUID.

 */

static int

pool_active(libzfs_handle_t *hdl, const char *name, uint64_t guid,

    boolean_t *isactive)

{

	zpool_handle_t *zhp;

	uint64_t theguid;

	if (zpool_open_silent(hdl, name, &zhp) != 0)

		return (-1);

	if (zhp == NULL) {

		*isactive = B_FALSE;

		return (0);

	}

	verify(nvlist_lookup_uint64(zhp->zpool_config, ZPOOL_CONFIG_POOL_GUID,

	    &theguid) == 0);

	zpool_close(zhp);

	*isactive = (theguid == guid);

	return (0);

}

static nvlist_t *

refresh_config(libzfs_handle_t *hdl, nvlist_t *config)

{

	nvlist_t *nvl;

	zfs_cmd_t zc = { 0 };

	int err;

	if (zcmd_write_conf_nvlist(hdl, &zc, config) != 0)

		return (NULL);

	if (zcmd_alloc_dst_nvlist(hdl, &zc,

	    zc.zc_nvlist_conf_size * 2) != 0) {

		zcmd_free_nvlists(&zc);

		return (NULL);

	}

	while ((err = ioctl(hdl->libzfs_fd, ZFS_IOC_POOL_TRYIMPORT,

	    &zc)) != 0 && errno == ENOMEM) {

		if (zcmd_expand_dst_nvlist(hdl, &zc) != 0) {

			zcmd_free_nvlists(&zc);

			return (NULL);

		}

	}

	if (err) {

		(void) zpool_standard_error(hdl, errno,

		    dgettext(TEXT_DOMAIN, "cannot discover pools"));

		zcmd_free_nvlists(&zc);

		return (NULL);

	}

	if (zcmd_read_dst_nvlist(hdl, &zc, &nvl) != 0) {

		zcmd_free_nvlists(&zc);

		return (NULL);

	}

	zcmd_free_nvlists(&zc);

	return (nvl);

}

/*

 * Convert our list of pools into the definitive set of configurations.  We

 * start by picking the best config for each toplevel vdev.  Once that's done,

 * we assemble the toplevel vdevs into a full config for the pool.  We make a

 * pass to fix up any incorrect paths, and then add it to the main list to

 * return to the user.

 */

static nvlist_t *

get_configs(libzfs_handle_t *hdl, pool_list_t *pl, boolean_t active_ok)

{

	pool_entry_t *pe;

	vdev_entry_t *ve;

	config_entry_t *ce;

	nvlist_t *ret = NULL, *config = NULL, *tmp, *nvtop, *nvroot;

	nvlist_t **spares, **l2cache;

	uint_t i, nspares, nl2cache;

	boolean_t config_seen;

	uint64_t best_txg;

	char *name, *hostname;

	uint64_t version, guid;

	uint_t children = 0;

	nvlist_t **child = NULL;

	uint_t c;

	boolean_t isactive;

	uint64_t hostid;

	nvlist_t *nvl;

	boolean_t found_one = B_FALSE;

	if (nvlist_alloc(&ret, 0, 0) != 0)

		goto nomem;

	for (pe = pl->pools; pe != NULL; pe = pe->pe_next) {

		uint64_t id;

		if (nvlist_alloc(&config, NV_UNIQUE_NAME, 0) != 0)

			goto nomem;

		config_seen = B_FALSE;

		/*

		 * Iterate over all toplevel vdevs.  Grab the pool configuration

		 * from the first one we find, and then go through the rest and

		 * add them as necessary to the 'vdevs' member of the config.

		 */

		for (ve = pe->pe_vdevs; ve != NULL; ve = ve->ve_next) {

			/*

			 * Determine the best configuration for this vdev by

			 * selecting the config with the latest transaction

			 * group.

			 */

			best_txg = 0;

			for (ce = ve->ve_configs; ce != NULL;

			    ce = ce->ce_next) {

				if (ce->ce_txg > best_txg) {

					tmp = ce->ce_config;

					best_txg = ce->ce_txg;

				}

			}

			if (!config_seen) {

				/*

				 * Copy the relevant pieces of data to the pool

				 * configuration:

				 *

				 *	version

				 * 	pool guid

				 * 	name

				 * 	pool state

				 *	hostid (if available)

				 *	hostname (if available)

				 */

				uint64_t state;

				verify(nvlist_lookup_uint64(tmp,

				    ZPOOL_CONFIG_VERSION, &version) == 0);

				if (nvlist_add_uint64(config,

				    ZPOOL_CONFIG_VERSION, version) != 0)

					goto nomem;

				verify(nvlist_lookup_uint64(tmp,

				    ZPOOL_CONFIG_POOL_GUID, &guid) == 0);

				if (nvlist_add_uint64(config,

				    ZPOOL_CONFIG_POOL_GUID, guid) != 0)

					goto nomem;

				verify(nvlist_lookup_string(tmp,

				    ZPOOL_CONFIG_POOL_NAME, &name) == 0);

				if (nvlist_add_string(config,

				    ZPOOL_CONFIG_POOL_NAME, name) != 0)

					goto nomem;

				verify(nvlist_lookup_uint64(tmp,

				    ZPOOL_CONFIG_POOL_STATE, &state) == 0);

				if (nvlist_add_uint64(config,

				    ZPOOL_CONFIG_POOL_STATE, state) != 0)

					goto nomem;

				hostid = 0;

				if (nvlist_lookup_uint64(tmp,

				    ZPOOL_CONFIG_HOSTID, &hostid) == 0) {

					if (nvlist_add_uint64(config,

					    ZPOOL_CONFIG_HOSTID, hostid) != 0)

						goto nomem;

					verify(nvlist_lookup_string(tmp,