From patchwork Tue Dec 14 21:16:41 2010 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jonthan Brassow X-Patchwork-Id: 411491 Received: from mx3-phx2.redhat.com (mx3-phx2.redhat.com [209.132.183.24]) by demeter1.kernel.org (8.14.4/8.14.3) with ESMTP id oBELJp6U029869 for ; Tue, 14 Dec 2010 21:20:12 GMT Received: from lists01.pubmisc.prod.ext.phx2.redhat.com (lists01.pubmisc.prod.ext.phx2.redhat.com [10.5.19.33]) by mx3-phx2.redhat.com (8.13.8/8.13.8) with ESMTP id oBELHmY2009307; Tue, 14 Dec 2010 16:17:48 -0500 Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22]) by lists01.pubmisc.prod.ext.phx2.redhat.com (8.13.8/8.13.8) with ESMTP id oBELGlsl001040 for ; Tue, 14 Dec 2010 16:16:47 -0500 Received: from hydrogen.msp.redhat.com (hydrogen.msp.redhat.com [10.15.80.1]) by int-mx09.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id oBELGfYw029047 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for ; Tue, 14 Dec 2010 16:16:42 -0500 Received: from hydrogen.msp.redhat.com ([127.0.0.1]) by hydrogen.msp.redhat.com (8.14.1/8.14.1) with ESMTP id oBELGfMf015525 for ; Tue, 14 Dec 2010 15:16:41 -0600 Received: (from jbrassow@localhost) by hydrogen.msp.redhat.com (8.14.1/8.14.1/Submit) id oBELGfRI015524 for dm-devel@redhat.com; Tue, 14 Dec 2010 15:16:41 -0600 Date: Tue, 14 Dec 2010 15:16:41 -0600 From: Jonathan Brassow Message-Id: <201012142116.oBELGfRI015524@hydrogen.msp.redhat.com> To: dm-devel@redhat.com X-Scanned-By: MIMEDefang 2.68 on 10.5.11.22 X-loop: dm-devel@redhat.com Subject: [dm-devel] [PATCH 4 of 8] dm raid seed module X-BeenThere: dm-devel@redhat.com X-Mailman-Version: 2.1.12 Precedence: junk Reply-To: device-mapper development List-Id: device-mapper development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Sender: dm-devel-bounces@redhat.com Errors-To: dm-devel-bounces@redhat.com X-Greylist: IP, sender and recipient auto-whitelisted, not delayed by milter-greylist-4.2.3 (demeter1.kernel.org [140.211.167.41]); Tue, 14 Dec 2010 21:20:12 +0000 (UTC) Index: linux-2.6/drivers/md/Kconfig =================================================================== --- linux-2.6.orig/drivers/md/Kconfig +++ linux-2.6/drivers/md/Kconfig @@ -240,6 +240,28 @@ config DM_MIRROR Allow volume managers to mirror logical volumes, also needed for live data migration tools such as 'pvmove'. +config DM_RAID + tristate "RAID 4/5/6 target (EXPERIMENTAL)" + depends on BLK_DEV_DM && MD_RAID456 && EXPERIMENTAL + ---help--- + A dm target that supports RAID4, RAID5 and RAID6 mappings + + A RAID-5 set of N drives with a capacity of C MB per drive provides + the capacity of C * (N - 1) MB, and protects against a failure + of a single drive. For a given sector (row) number, (N - 1) drives + contain data sectors, and one drive contains the parity protection. + For a RAID-4 set, the parity blocks are present on a single drive, + while a RAID-5 set distributes the parity across the drives in one + of the available parity distribution methods. + + A RAID-6 set of N drives with a capacity of C MB per drive + provides the capacity of C * (N - 2) MB, and protects + against a failure of any two drives. For a given sector + (row) number, (N - 2) drives contain data sectors, and two + drives contains two independent redundancy syndromes. Like + RAID-5, RAID-6 distributes the syndromes across the drives + in one of the available parity distribution methods. + config DM_LOG_USERSPACE tristate "Mirror userspace logging (EXPERIMENTAL)" depends on DM_MIRROR && EXPERIMENTAL && NET Index: linux-2.6/drivers/md/Makefile =================================================================== --- linux-2.6.orig/drivers/md/Makefile +++ linux-2.6/drivers/md/Makefile @@ -36,6 +36,7 @@ obj-$(CONFIG_DM_SNAPSHOT) += dm-snapshot obj-$(CONFIG_DM_MIRROR) += dm-mirror.o dm-log.o dm-region-hash.o obj-$(CONFIG_DM_LOG_USERSPACE) += dm-log-userspace.o obj-$(CONFIG_DM_ZERO) += dm-zero.o +obj-$(CONFIG_DM_RAID) += dm-raid.o ifeq ($(CONFIG_DM_UEVENT),y) dm-mod-objs += dm-uevent.o Index: linux-2.6/drivers/md/dm-raid.c =================================================================== --- /dev/null +++ linux-2.6/drivers/md/dm-raid.c @@ -0,0 +1,578 @@ + +/* + * dm-raid - implemented as wrapper for md/raid456 + * + */ +#include +#include "md.h" +#include "raid5.h" +#include "dm.h" +#include "bitmap.h" + +#define DM_MSG_PREFIX "raid" + +/* + * If the MD doesn't support MD_SYNC_STATE_FORCED yet, then + * make it so the flag doesn't set anything. + */ +#ifndef MD_SYNC_STATE_FORCED +#define MD_SYNC_STATE_FORCED 0 +#endif + +struct raid_dev { + /* + * Two DM devices, one to hold metadata and one to hold the + * actual data/parity. The reason for this is to not confuse + * ti->len and give more flexibility in altering size and + * characteristics. + * + * While it is possible for this device to be associated + * with a different physical device than the data_dev, it + * is intended for it to be the same. + * |--------- Physical Device ---------| + * |- meta_dev -|------ data_dev ------| + */ + struct dm_dev *meta_dev; + struct dm_dev *data_dev; + struct mdk_rdev_s rdev; +}; + +struct raid_set { + struct dm_target *ti; + +#define DMPF_DAEMON_SLEEP 0x1 +#define DMPF_MAX_WRITE_BEHIND 0x2 +#define DMPF_SYNC 0x4 +#define DMPF_NOSYNC 0x8 +#define DMPF_STRIPE_CACHE 0x10 +#define DMPF_MIN_RECOVERY_RATE 0x20 +#define DMPF_MAX_RECOVERY_RATE 0x40 + uint64_t print_flags; + + struct mddev_s md; + struct raid_type *raid_type; + + struct raid_dev dev[0]; +}; + +/* Supported raid types and properties. */ +static struct raid_type { + const char *name; /* RAID algorithm. */ + const char *descr; /* Descriptor text for logging. */ + const unsigned parity_devs; /* # of parity devices. */ + const unsigned minimal_devs; /* minimal # of devices in set. */ + const unsigned level; /* RAID level. */ + const unsigned algorithm; /* RAID algorithm. */ +} raid_types[] = { + {"raid4", "RAID4 (dedicated parity disk)", 1, 2, 5, ALGORITHM_PARITY_0}, + {"raid5_la", "RAID5 (left asymmetric)", 1, 2, 5, ALGORITHM_LEFT_ASYMMETRIC}, + {"raid5_ra", "RAID5 (right asymmetric)", 1, 2, 5, ALGORITHM_RIGHT_ASYMMETRIC}, + {"raid5_ls", "RAID5 (left symmetric)", 1, 2, 5, ALGORITHM_LEFT_SYMMETRIC}, + {"raid5_rs", "RAID5 (right symmetric)", 1, 2, 5, ALGORITHM_RIGHT_SYMMETRIC}, + {"raid6_zr", "RAID6 (zero restart)", 2, 4, 6, ALGORITHM_ROTATING_ZERO_RESTART}, + {"raid6_nr", "RAID6 (N restart)", 2, 4, 6, ALGORITHM_ROTATING_N_RESTART}, + {"raid6_nc", "RAID6 (N continue)", 2, 4, 6, ALGORITHM_ROTATING_N_CONTINUE} +}; + +static struct raid_type *get_raid_type(char *name) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(raid_types); i++) + if (strcmp(raid_types[i].name, name) == 0) + return &raid_types[i]; + return NULL; +} + +static struct raid_set * +context_alloc(struct dm_target *ti, struct raid_type *raid_type, long raid_devs) +{ + int i; + struct raid_set *rs; + sector_t sectors_per_dev; + + sectors_per_dev = ti->len; + if (sector_div(sectors_per_dev, (raid_devs - raid_type->parity_devs))) { + ti->error = "Target length not divisible by number of data devices"; + return ERR_PTR(-EINVAL); + } + + rs = kzalloc(sizeof(*rs) + raid_devs * sizeof(rs->dev[0]), GFP_KERNEL); + if (!rs) { + ti->error = "Cannot allocate raid context"; + return ERR_PTR(-ENOMEM); + } + + mddev_init(&rs->md); + + rs->ti = ti; + rs->raid_type = raid_type; + rs->md.raid_disks = raid_devs; + rs->md.level = raid_type->level; + rs->md.new_level = rs->md.level; + rs->md.dev_sectors = sectors_per_dev; + rs->md.layout = raid_type->algorithm; + rs->md.new_layout = rs->md.layout; + rs->md.delta_disks = 0; + rs->md.recovery_cp = 0; + + for (i = 0; i < raid_devs; i++) + md_rdev_init(&rs->dev[i].rdev); + + /* + * Remaining items to be initialized by further RAID params: + * rs->md.persistent + * rs->md.external + * rs->md.chunk_sectors + * rs->md.new_chunk_sectors + */ + + return rs; +} + +static void context_free(struct raid_set *rs) +{ + int i; + for (i = 0; i < rs->md.raid_disks; i++) { + if (rs->dev[i].data_dev) + dm_put_device(rs->ti, rs->dev[i].data_dev); + } + kfree(rs); +} + +/* For every device we have two words + * : meta device name or '-' if missing + * : data device name or '-' if missing + * + * This code parses those words. + */ +static int dev_parms(struct raid_set *rs, char **argv) +{ + int i; + int rebuild = 0; + int metadata_available = 0; + + for (i = 0; i < rs->md.raid_disks; i++, argv += 2) { + int err = 0; + + rs->dev[i].rdev.raid_disk = i; + + rs->dev[i].meta_dev = NULL; + rs->dev[i].data_dev = NULL; + + /* + * There are no offsets, since there is a separate device + * for data and metadata. + */ + rs->dev[i].rdev.data_offset = 0; + rs->dev[i].rdev.mddev = &rs->md; + + if (strcmp(argv[0], "-") != 0) { + rs->ti->error = "Metadata devices not supported"; + return -EINVAL; + } + + if (strcmp(argv[1], "-") == 0) { + rs->ti->error = "Drive designated for rebuild not specified"; + if (!test_bit(In_sync, &rs->dev[i].rdev.flags) && + (rs->dev[i].rdev.recovery_offset == 0)) + return -EINVAL; + + continue; + } + + err = dm_get_device(rs->ti, argv[1], + dm_table_get_mode(rs->ti->table), + &rs->dev[i].data_dev); + rs->ti->error = "RAID device lookup failure"; + if (err) + return err; + + rs->dev[i].rdev.bdev = rs->dev[i].data_dev->bdev; + list_add(&rs->dev[i].rdev.same_set, &rs->md.disks); + if (!test_bit(In_sync, &rs->dev[i].rdev.flags)) + rebuild++; + } + + if (metadata_available) { + rs->md.external = 0; + rs->md.persistent = 1; + rs->md.major_version = 2; + } else if (rebuild && !rs->md.recovery_cp) { + /* + * Without metadata, we will not be able to tell if the array + * is in-sync or not - we must assume it is not. Therefore, + * it is impossible to rebuild a drive. + * + * Even if there is metadata, the on-disk information may + * indicate that the array is not in-sync and it will then + * fail at that time. + * + * User could specify 'nosync' option if desperate. + */ + DMERR("Unable to rebuild drive while array is not in-sync"); + rs->ti->error = "RAID device lookup failure"; + return -EINVAL; + } + + rs->ti->error = NULL; + return 0; +} + +/* + * Possible arguments are... + * RAID456: + * [optional_args] + * + * Optional args include: + * [[no]sync] Force or prevent recovery of the entire array + * [rebuild ] Rebuild the drive indicated by the index + * [daemon_sleep ] Time between bitmap daemon work to clear bits + * [min_recovery_rate ] + * [max_recovery_rate ] + * [max_write_behind ] + * [stripe_cache ] + */ +static int parse_raid_params(struct raid_set *rs, char **argv, + unsigned long num_raid_params) +{ + int i, rebuild_cnt = 0; + unsigned long value; + char *key; + char *reason = NULL; + struct mddev_s *mddev = &rs->md; + + /* + * First, parse the in-order required arguments + */ + rs->ti->error = "Bad chunk size"; + if ((strict_strtoul(argv[0], 10, &value) < 0) || + !is_power_of_2(value) || (value < 8)) + return -EINVAL; + rs->md.new_chunk_sectors = rs->md.chunk_sectors = value; + argv++; + num_raid_params--; + + /* + * Second, parse the unordered optional arguments + */ + for (i = 0; i < rs->md.raid_disks; i++) + set_bit(In_sync, &rs->dev[i].rdev.flags); + + for (i = 0; i < num_raid_params; i++) { + if (!strcmp(argv[i], "nosync")) { + rs->md.recovery_cp = MaxSector; + rs->print_flags |= DMPF_NOSYNC; + rs->md.flags |= MD_SYNC_STATE_FORCED; + continue; + } + if (!strcmp(argv[i], "sync")) { + rs->md.recovery_cp = 0; + rs->print_flags |= DMPF_SYNC; + rs->md.flags |= MD_SYNC_STATE_FORCED; + continue; + } + + /* The rest of the optional arguments come in key/value pairs */ + if ((i + 1) >= num_raid_params) { + reason = "Wrong number of raid parameters given"; + goto out; + } + key = argv[i]; + i++; + if (strict_strtoul(argv[i], 10, &value) < 0) { + reason = "Bad numerical argument given in raid params"; + goto out; + } + + if (!strcmp(key, "rebuild")) { + if (++rebuild_cnt > rs->raid_type->parity_devs) + reason = "Too many rebuild drives given"; + else if ((value < 0) || (value > rs->md.raid_disks)) + reason = "Invalid rebuild index given"; + else { + clear_bit(In_sync, &rs->dev[value].rdev.flags); + rs->dev[value].rdev.recovery_offset = 0; + } + } else if (!strcmp(key, "max_write_behind")) { + rs->print_flags |= DMPF_MAX_WRITE_BEHIND; + + /* + * In device-mapper, we specify things in sectors, but + * MD records this value in kB + */ + value /= 2; + if (value > COUNTER_MAX) + reason = "Max write-behind limit out of range"; + else + mddev->bitmap_info.max_write_behind = value; + } else if (!strcmp(key, "daemon_sleep")) { + rs->print_flags |= DMPF_DAEMON_SLEEP; + if ((value < 1) || (value > MAX_SCHEDULE_TIMEOUT)) + reason = "daemon sleep period out of range"; + else + mddev->bitmap_info.daemon_sleep = value; + } else if (!strcmp(key, "stripe_cache")) { + rs->print_flags |= DMPF_STRIPE_CACHE; + + /* + * In device-mapper, we specify things in sectors, but + * MD records this value in kB + */ + value /= 2; + if (rs->raid_type->level < 5) + reason = "Inappropriate argument, stripe_cache"; + else if (raid5_set_cache_size(&rs->md, (int)value)) + reason = "Bad stripe_cache size"; + } else if (!strcmp(key, "min_recovery_rate")) { + rs->print_flags |= DMPF_MIN_RECOVERY_RATE; + rs->md.sync_speed_min = (int)value; + } else if (!strcmp(key, "max_recovery_rate")) { + rs->print_flags |= DMPF_MAX_RECOVERY_RATE; + rs->md.sync_speed_max = (int)value; + rs->md.flags |= MD_SYNC_STATE_FORCED; + } else { + DMERR("Unable to parse RAID parameter, %s", key); + reason = "Unable to parse RAID parameters"; + } + if (reason) + goto out; + } + + /* Assume there are no metadata devices until the drives are parsed */ + rs->md.persistent = 0; + rs->md.external = 1; + +out: + rs->ti->error = reason; + return (reason) ? -EINVAL : 0; +} + +static void do_table_event(struct work_struct *ws) +{ + struct raid_set *rs = container_of(ws, struct raid_set, + md.event_work); + dm_table_event(rs->ti->table); +} + +/* + * Construct a RAID4/5/6 mapping: + * Args: + * <#raid_params> \ + * <#raid_devs> { .. } + * + * ** metadata devices are not supported yet, use '-' instead ** + * + * varies by . See 'parse_raid_params' for + * details on possible . + */ +static int raid_ctr(struct dm_target *ti, unsigned argc, char **argv) +{ + char *err = NULL; + int errnum = -EINVAL; + struct raid_type *rt; + unsigned long num_raid_params, num_raid_devs; + struct raid_set *rs = NULL; + + /* Must have at least <#raid_params> */ + err = "Too few arguments"; + if (argc < 2) + goto err; + + /* raid type */ + err = "Cannot find raid_type"; + rt = get_raid_type(argv[0]); + if (!rt) + goto err; + argc--; argv++; + + /* number of RAID parameters */ + err = "Cannot understand number of RAID parameters"; + if (strict_strtoul(argv[0], 10, &num_raid_params) < 0) + goto err; + argc--; argv++; + + /* Skip over RAID params for now and find out # of devices */ + err = "Arguments do not agree with counts given"; + if (num_raid_params + 1 > argc) + goto err; + + err = "Bad number of raid devices"; + if (strict_strtol(argv[num_raid_params], 10, &num_raid_devs) < 0) + goto err; + + rs = context_alloc(ti, rt, num_raid_devs); + if (IS_ERR(rs)) + return PTR_ERR(rs); + + errnum = parse_raid_params(rs, argv, num_raid_params); + if (errnum) { + err = ti->error; + goto err; + } + errnum = -EINVAL; + + argc -= num_raid_params + 1; /* +1: we already have num_raid_devs */ + argv += num_raid_params + 1; + + err = "Supplied RAID devices does not match the count given"; + if (argc != (num_raid_devs * 2)) + goto err; + + errnum = dev_parms(rs, argv); + if (errnum) { + err = ti->error; + goto err; + } + + INIT_WORK(&rs->md.event_work, do_table_event); + ti->split_io = rs->md.chunk_sectors; + ti->private = rs; + + mutex_lock(&rs->md.reconfig_mutex); + err = "Fail to run raid array"; + errnum = md_run(&rs->md); + rs->md.in_sync = 0; /* Assume already marked dirty */ + mutex_unlock(&rs->md.reconfig_mutex); + + if (!errnum) + return 0; + +err: + if (rs) + context_free(rs); + ti->error = err; + return errnum; +} + +static void raid_dtr(struct dm_target *ti) +{ + struct raid_set *rs = ti->private; + + md_stop(&rs->md); + context_free(rs); +} + +static int raid_map(struct dm_target *ti, struct bio *bio, + union map_info *map_context) +{ + struct raid_set *rs = ti->private; + mddev_t *mddev = &rs->md; + + mddev->pers->make_request(mddev, bio); + return DM_MAPIO_SUBMITTED; +} + +static int raid_status(struct dm_target *ti, status_type_t type, + char *result, unsigned maxlen) +{ + struct raid_set *rs = ti->private; + struct raid5_private_data *conf = rs->md.private; + int sz = 0; + int raid_param_cnt = 1; /* at least 1 for chunksize */ + int i; + sector_t sync; + + switch (type) { + case STATUSTYPE_INFO: + DMEMIT("%s %d ", rs->raid_type->name, rs->md.raid_disks); + for (i = 0; i < rs->md.raid_disks; i++) { + if (test_bit(Faulty, &rs->dev[i].rdev.flags)) + DMEMIT("D"); + else if (test_bit(In_sync, &rs->dev[i].rdev.flags)) + DMEMIT("A"); + else + DMEMIT("a"); + } + if (test_bit(MD_RECOVERY_RUNNING, &rs->md.recovery)) + sync = rs->md.curr_resync_completed; + else + sync = rs->md.recovery_cp; + if (sync > rs->md.resync_max_sectors) + sync = rs->md.resync_max_sectors; + DMEMIT(" %llu/%llu", + (unsigned long long) sync, + (unsigned long long) rs->md.resync_max_sectors); + + break; + case STATUSTYPE_TABLE: + /* The string you would use to construct this array */ + for (i = 0; i < rs->md.raid_disks; i++) + if (rs->dev[i].data_dev && + !test_bit(In_sync, &rs->dev[i].rdev.flags)) + raid_param_cnt++; /* for rebuilds */ + + raid_param_cnt += (hweight64(rs->print_flags) * 2); + if (rs->print_flags & (DMPF_SYNC | DMPF_NOSYNC)) + raid_param_cnt--; + + DMEMIT("%s %d %u", rs->raid_type->name, + raid_param_cnt, rs->md.chunk_sectors); + + for (i = 0; i < rs->md.raid_disks; i++) + if (rs->dev[i].data_dev && + !test_bit(In_sync, &rs->dev[i].rdev.flags)) + DMEMIT(" rebuild=%u", i); + + if (rs->print_flags & DMPF_DAEMON_SLEEP) + DMEMIT(" daemon_sleep %lu", + rs->md.bitmap_info.daemon_sleep); + if (rs->print_flags & DMPF_MAX_WRITE_BEHIND) + DMEMIT(" max_write_behind %lu", + rs->md.bitmap_info.max_write_behind); + if (rs->print_flags & DMPF_STRIPE_CACHE) + DMEMIT(" stripe_cache "); + if (rs->print_flags & DMPF_MIN_RECOVERY_RATE) + DMEMIT(" min_recovery_rate %d", rs->md.sync_speed_min); + if (rs->print_flags & DMPF_MAX_RECOVERY_RATE) + DMEMIT(" max_recovery_rate %d", rs->md.sync_speed_max); + if ((rs->print_flags & DMPF_SYNC) && + (rs->md.recovery_cp == MaxSector)) + DMEMIT(" sync"); + if (rs->print_flags & DMPF_NOSYNC) + DMEMIT(" nosync"); + + DMEMIT(" %d", rs->md.raid_disks); + for (i = 0; i < rs->md.raid_disks; i++) { + DMEMIT(" -"); /* metadata device */ + + if (rs->dev[i].data_dev) + DMEMIT(" %s", rs->dev[i].data_dev->name); + else + DMEMIT(" -"); + } + break; + } + return 0; +} + +static struct target_type raid_target = { + .name = "raid", + .version = {1, 0, 0}, + .module = THIS_MODULE, + .ctr = raid_ctr, + .dtr = raid_dtr, + .map = raid_map, + .status = raid_status, +}; + +static int __init dm_raid_init(void) +{ + int r = dm_register_target(&raid_target); + + return r; +} + +static void __exit dm_raid_exit(void) +{ + dm_unregister_target(&raid_target); +} + +module_init(dm_raid_init); +module_exit(dm_raid_exit); + +MODULE_DESCRIPTION(DM_NAME " raid4/5/6 target"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("dm-raid4"); +MODULE_ALIAS("dm-raid5"); +MODULE_ALIAS("dm-raid6");