^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) * Copyright (C) 2010-2011 Neil Brown
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * Copyright (C) 2010-2018 Red Hat, Inc. All rights reserved.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * This file is released under the GPL.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) #include <linux/slab.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) #include <linux/module.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include "md.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include "raid1.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include "raid5.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include "raid10.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include "md-bitmap.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <linux/device-mapper.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #define DM_MSG_PREFIX "raid"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #define MAX_RAID_DEVICES 253 /* md-raid kernel limit */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) * Minimum sectors of free reshape space per raid device
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) #define MIN_FREE_RESHAPE_SPACE to_sector(4*4096)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) * Minimum journal space 4 MiB in sectors.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) #define MIN_RAID456_JOURNAL_SPACE (4*2048)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) static bool devices_handle_discard_safely = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) * The following flags are used by dm-raid.c to set up the array state.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) * They must be cleared before md_run is called.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) #define FirstUse 10 /* rdev flag */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) struct raid_dev {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) * Two DM devices, one to hold metadata and one to hold the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) * actual data/parity. The reason for this is to not confuse
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) * ti->len and give more flexibility in altering size and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) * characteristics.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) * While it is possible for this device to be associated
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) * with a different physical device than the data_dev, it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) * is intended for it to be the same.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) * |--------- Physical Device ---------|
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) * |- meta_dev -|------ data_dev ------|
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) struct dm_dev *meta_dev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) struct dm_dev *data_dev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) struct md_rdev rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) * Bits for establishing rs->ctr_flags
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) * 1 = no flag value
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) * 2 = flag with value
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) #define __CTR_FLAG_SYNC 0 /* 1 */ /* Not with raid0! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) #define __CTR_FLAG_NOSYNC 1 /* 1 */ /* Not with raid0! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) #define __CTR_FLAG_REBUILD 2 /* 2 */ /* Not with raid0! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) #define __CTR_FLAG_DAEMON_SLEEP 3 /* 2 */ /* Not with raid0! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) #define __CTR_FLAG_MIN_RECOVERY_RATE 4 /* 2 */ /* Not with raid0! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) #define __CTR_FLAG_MAX_RECOVERY_RATE 5 /* 2 */ /* Not with raid0! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) #define __CTR_FLAG_MAX_WRITE_BEHIND 6 /* 2 */ /* Only with raid1! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) #define __CTR_FLAG_WRITE_MOSTLY 7 /* 2 */ /* Only with raid1! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) #define __CTR_FLAG_STRIPE_CACHE 8 /* 2 */ /* Only with raid4/5/6! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) #define __CTR_FLAG_REGION_SIZE 9 /* 2 */ /* Not with raid0! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) #define __CTR_FLAG_RAID10_COPIES 10 /* 2 */ /* Only with raid10 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) #define __CTR_FLAG_RAID10_FORMAT 11 /* 2 */ /* Only with raid10 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) /* New for v1.9.0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) #define __CTR_FLAG_DELTA_DISKS 12 /* 2 */ /* Only with reshapable raid1/4/5/6/10! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) #define __CTR_FLAG_DATA_OFFSET 13 /* 2 */ /* Only with reshapable raid4/5/6/10! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) #define __CTR_FLAG_RAID10_USE_NEAR_SETS 14 /* 2 */ /* Only with raid10! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) /* New for v1.10.0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) #define __CTR_FLAG_JOURNAL_DEV 15 /* 2 */ /* Only with raid4/5/6 (journal device)! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) /* New for v1.11.1 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) #define __CTR_FLAG_JOURNAL_MODE 16 /* 2 */ /* Only with raid4/5/6 (journal mode)! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) * Flags for rs->ctr_flags field.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) #define CTR_FLAG_SYNC (1 << __CTR_FLAG_SYNC)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) #define CTR_FLAG_NOSYNC (1 << __CTR_FLAG_NOSYNC)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) #define CTR_FLAG_REBUILD (1 << __CTR_FLAG_REBUILD)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) #define CTR_FLAG_DAEMON_SLEEP (1 << __CTR_FLAG_DAEMON_SLEEP)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) #define CTR_FLAG_MIN_RECOVERY_RATE (1 << __CTR_FLAG_MIN_RECOVERY_RATE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) #define CTR_FLAG_MAX_RECOVERY_RATE (1 << __CTR_FLAG_MAX_RECOVERY_RATE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) #define CTR_FLAG_MAX_WRITE_BEHIND (1 << __CTR_FLAG_MAX_WRITE_BEHIND)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) #define CTR_FLAG_WRITE_MOSTLY (1 << __CTR_FLAG_WRITE_MOSTLY)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) #define CTR_FLAG_STRIPE_CACHE (1 << __CTR_FLAG_STRIPE_CACHE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) #define CTR_FLAG_REGION_SIZE (1 << __CTR_FLAG_REGION_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) #define CTR_FLAG_RAID10_COPIES (1 << __CTR_FLAG_RAID10_COPIES)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) #define CTR_FLAG_RAID10_FORMAT (1 << __CTR_FLAG_RAID10_FORMAT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) #define CTR_FLAG_DELTA_DISKS (1 << __CTR_FLAG_DELTA_DISKS)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) #define CTR_FLAG_DATA_OFFSET (1 << __CTR_FLAG_DATA_OFFSET)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) #define CTR_FLAG_RAID10_USE_NEAR_SETS (1 << __CTR_FLAG_RAID10_USE_NEAR_SETS)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) #define CTR_FLAG_JOURNAL_DEV (1 << __CTR_FLAG_JOURNAL_DEV)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) #define CTR_FLAG_JOURNAL_MODE (1 << __CTR_FLAG_JOURNAL_MODE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) * Definitions of various constructor flags to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) * be used in checks of valid / invalid flags
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) * per raid level.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) /* Define all any sync flags */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) #define CTR_FLAGS_ANY_SYNC (CTR_FLAG_SYNC | CTR_FLAG_NOSYNC)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) /* Define flags for options without argument (e.g. 'nosync') */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) #define CTR_FLAG_OPTIONS_NO_ARGS (CTR_FLAGS_ANY_SYNC | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) CTR_FLAG_RAID10_USE_NEAR_SETS)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) /* Define flags for options with one argument (e.g. 'delta_disks +2') */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) #define CTR_FLAG_OPTIONS_ONE_ARG (CTR_FLAG_REBUILD | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) CTR_FLAG_WRITE_MOSTLY | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) CTR_FLAG_DAEMON_SLEEP | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) CTR_FLAG_MIN_RECOVERY_RATE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) CTR_FLAG_MAX_RECOVERY_RATE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) CTR_FLAG_MAX_WRITE_BEHIND | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) CTR_FLAG_STRIPE_CACHE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) CTR_FLAG_REGION_SIZE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) CTR_FLAG_RAID10_COPIES | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) CTR_FLAG_RAID10_FORMAT | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) CTR_FLAG_DELTA_DISKS | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) CTR_FLAG_DATA_OFFSET | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) CTR_FLAG_JOURNAL_DEV | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) CTR_FLAG_JOURNAL_MODE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) /* Valid options definitions per raid level... */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) /* "raid0" does only accept data offset */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) #define RAID0_VALID_FLAGS (CTR_FLAG_DATA_OFFSET)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) /* "raid1" does not accept stripe cache, data offset, delta_disks or any raid10 options */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) #define RAID1_VALID_FLAGS (CTR_FLAGS_ANY_SYNC | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) CTR_FLAG_REBUILD | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) CTR_FLAG_WRITE_MOSTLY | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) CTR_FLAG_DAEMON_SLEEP | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) CTR_FLAG_MIN_RECOVERY_RATE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) CTR_FLAG_MAX_RECOVERY_RATE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) CTR_FLAG_MAX_WRITE_BEHIND | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) CTR_FLAG_REGION_SIZE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) CTR_FLAG_DELTA_DISKS | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) CTR_FLAG_DATA_OFFSET)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) /* "raid10" does not accept any raid1 or stripe cache options */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) #define RAID10_VALID_FLAGS (CTR_FLAGS_ANY_SYNC | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) CTR_FLAG_REBUILD | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) CTR_FLAG_DAEMON_SLEEP | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) CTR_FLAG_MIN_RECOVERY_RATE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) CTR_FLAG_MAX_RECOVERY_RATE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) CTR_FLAG_REGION_SIZE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) CTR_FLAG_RAID10_COPIES | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) CTR_FLAG_RAID10_FORMAT | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) CTR_FLAG_DELTA_DISKS | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) CTR_FLAG_DATA_OFFSET | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) CTR_FLAG_RAID10_USE_NEAR_SETS)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) * "raid4/5/6" do not accept any raid1 or raid10 specific options
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) * "raid6" does not accept "nosync", because it is not guaranteed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) * that both parity and q-syndrome are being written properly with
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) * any writes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) #define RAID45_VALID_FLAGS (CTR_FLAGS_ANY_SYNC | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) CTR_FLAG_REBUILD | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) CTR_FLAG_DAEMON_SLEEP | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) CTR_FLAG_MIN_RECOVERY_RATE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) CTR_FLAG_MAX_RECOVERY_RATE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) CTR_FLAG_STRIPE_CACHE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) CTR_FLAG_REGION_SIZE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) CTR_FLAG_DELTA_DISKS | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) CTR_FLAG_DATA_OFFSET | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) CTR_FLAG_JOURNAL_DEV | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) CTR_FLAG_JOURNAL_MODE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) #define RAID6_VALID_FLAGS (CTR_FLAG_SYNC | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) CTR_FLAG_REBUILD | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) CTR_FLAG_DAEMON_SLEEP | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) CTR_FLAG_MIN_RECOVERY_RATE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) CTR_FLAG_MAX_RECOVERY_RATE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) CTR_FLAG_STRIPE_CACHE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) CTR_FLAG_REGION_SIZE | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) CTR_FLAG_DELTA_DISKS | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) CTR_FLAG_DATA_OFFSET | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) CTR_FLAG_JOURNAL_DEV | \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) CTR_FLAG_JOURNAL_MODE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) /* ...valid options definitions per raid level */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) * Flags for rs->runtime_flags field
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) * (RT_FLAG prefix meaning "runtime flag")
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) * These are all internal and used to define runtime state,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) * e.g. to prevent another resume from preresume processing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) * the raid set all over again.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) #define RT_FLAG_RS_PRERESUMED 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) #define RT_FLAG_RS_RESUMED 1
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) #define RT_FLAG_RS_BITMAP_LOADED 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) #define RT_FLAG_UPDATE_SBS 3
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) #define RT_FLAG_RESHAPE_RS 4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) #define RT_FLAG_RS_SUSPENDED 5
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) #define RT_FLAG_RS_IN_SYNC 6
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) #define RT_FLAG_RS_RESYNCING 7
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) #define RT_FLAG_RS_GROW 8
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) /* Array elements of 64 bit needed for rebuild/failed disk bits */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) #define DISKS_ARRAY_ELEMS ((MAX_RAID_DEVICES + (sizeof(uint64_t) * 8 - 1)) / sizeof(uint64_t) / 8)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) * raid set level, layout and chunk sectors backup/restore
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) struct rs_layout {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) int new_level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) int new_layout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) int new_chunk_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) struct raid_set {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) struct dm_target *ti;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) uint32_t stripe_cache_entries;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) unsigned long ctr_flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) unsigned long runtime_flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) uint64_t rebuild_disks[DISKS_ARRAY_ELEMS];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) int raid_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) int delta_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) int data_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) int raid10_copies;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) int requested_bitmap_chunk_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) struct mddev md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) struct raid_type *raid_type;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) sector_t array_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) sector_t dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) /* Optional raid4/5/6 journal device */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) struct journal_dev {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) struct dm_dev *dev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) struct md_rdev rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) int mode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) } journal_dev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) struct raid_dev dev[];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) static void rs_config_backup(struct raid_set *rs, struct rs_layout *l)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) l->new_level = mddev->new_level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) l->new_layout = mddev->new_layout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) l->new_chunk_sectors = mddev->new_chunk_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) static void rs_config_restore(struct raid_set *rs, struct rs_layout *l)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) mddev->new_level = l->new_level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) mddev->new_layout = l->new_layout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) mddev->new_chunk_sectors = l->new_chunk_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) /* raid10 algorithms (i.e. formats) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) #define ALGORITHM_RAID10_DEFAULT 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) #define ALGORITHM_RAID10_NEAR 1
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) #define ALGORITHM_RAID10_OFFSET 2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) #define ALGORITHM_RAID10_FAR 3
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) /* Supported raid types and properties. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) static struct raid_type {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) const char *name; /* RAID algorithm. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) const char *descr; /* Descriptor text for logging. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) const unsigned int parity_devs; /* # of parity devices. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) const unsigned int minimal_devs;/* minimal # of devices in set. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) const unsigned int level; /* RAID level. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) const unsigned int algorithm; /* RAID algorithm. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) } raid_types[] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) {"raid0", "raid0 (striping)", 0, 2, 0, 0 /* NONE */},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) {"raid1", "raid1 (mirroring)", 0, 2, 1, 0 /* NONE */},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) {"raid10_far", "raid10 far (striped mirrors)", 0, 2, 10, ALGORITHM_RAID10_FAR},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) {"raid10_offset", "raid10 offset (striped mirrors)", 0, 2, 10, ALGORITHM_RAID10_OFFSET},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) {"raid10_near", "raid10 near (striped mirrors)", 0, 2, 10, ALGORITHM_RAID10_NEAR},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) {"raid10", "raid10 (striped mirrors)", 0, 2, 10, ALGORITHM_RAID10_DEFAULT},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) {"raid4", "raid4 (dedicated first parity disk)", 1, 2, 5, ALGORITHM_PARITY_0}, /* raid4 layout = raid5_0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) {"raid5_n", "raid5 (dedicated last parity disk)", 1, 2, 5, ALGORITHM_PARITY_N},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) {"raid5_ls", "raid5 (left symmetric)", 1, 2, 5, ALGORITHM_LEFT_SYMMETRIC},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) {"raid5_rs", "raid5 (right symmetric)", 1, 2, 5, ALGORITHM_RIGHT_SYMMETRIC},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) {"raid5_la", "raid5 (left asymmetric)", 1, 2, 5, ALGORITHM_LEFT_ASYMMETRIC},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) {"raid5_ra", "raid5 (right asymmetric)", 1, 2, 5, ALGORITHM_RIGHT_ASYMMETRIC},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) {"raid6_zr", "raid6 (zero restart)", 2, 4, 6, ALGORITHM_ROTATING_ZERO_RESTART},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) {"raid6_nr", "raid6 (N restart)", 2, 4, 6, ALGORITHM_ROTATING_N_RESTART},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) {"raid6_nc", "raid6 (N continue)", 2, 4, 6, ALGORITHM_ROTATING_N_CONTINUE},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) {"raid6_n_6", "raid6 (dedicated parity/Q n/6)", 2, 4, 6, ALGORITHM_PARITY_N_6},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) {"raid6_ls_6", "raid6 (left symmetric dedicated Q 6)", 2, 4, 6, ALGORITHM_LEFT_SYMMETRIC_6},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) {"raid6_rs_6", "raid6 (right symmetric dedicated Q 6)", 2, 4, 6, ALGORITHM_RIGHT_SYMMETRIC_6},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) {"raid6_la_6", "raid6 (left asymmetric dedicated Q 6)", 2, 4, 6, ALGORITHM_LEFT_ASYMMETRIC_6},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) {"raid6_ra_6", "raid6 (right asymmetric dedicated Q 6)", 2, 4, 6, ALGORITHM_RIGHT_ASYMMETRIC_6}
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) /* True, if @v is in inclusive range [@min, @max] */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) static bool __within_range(long v, long min, long max)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) return v >= min && v <= max;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) /* All table line arguments are defined here */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) static struct arg_name_flag {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) const unsigned long flag;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) const char *name;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) } __arg_name_flags[] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) { CTR_FLAG_SYNC, "sync"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) { CTR_FLAG_NOSYNC, "nosync"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) { CTR_FLAG_REBUILD, "rebuild"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) { CTR_FLAG_DAEMON_SLEEP, "daemon_sleep"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) { CTR_FLAG_MIN_RECOVERY_RATE, "min_recovery_rate"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) { CTR_FLAG_MAX_RECOVERY_RATE, "max_recovery_rate"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) { CTR_FLAG_MAX_WRITE_BEHIND, "max_write_behind"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) { CTR_FLAG_WRITE_MOSTLY, "write_mostly"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) { CTR_FLAG_STRIPE_CACHE, "stripe_cache"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) { CTR_FLAG_REGION_SIZE, "region_size"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) { CTR_FLAG_RAID10_COPIES, "raid10_copies"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) { CTR_FLAG_RAID10_FORMAT, "raid10_format"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) { CTR_FLAG_DATA_OFFSET, "data_offset"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) { CTR_FLAG_DELTA_DISKS, "delta_disks"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) { CTR_FLAG_RAID10_USE_NEAR_SETS, "raid10_use_near_sets"},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) { CTR_FLAG_JOURNAL_DEV, "journal_dev" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) { CTR_FLAG_JOURNAL_MODE, "journal_mode" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) /* Return argument name string for given @flag */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) static const char *dm_raid_arg_name_by_flag(const uint32_t flag)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) if (hweight32(flag) == 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) struct arg_name_flag *anf = __arg_name_flags + ARRAY_SIZE(__arg_name_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) while (anf-- > __arg_name_flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) if (flag & anf->flag)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) return anf->name;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) DMERR("%s called with more than one flag!", __func__);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) /* Define correlation of raid456 journal cache modes and dm-raid target line parameters */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) static struct {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) const int mode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) const char *param;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) } _raid456_journal_mode[] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) { R5C_JOURNAL_MODE_WRITE_THROUGH , "writethrough" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) { R5C_JOURNAL_MODE_WRITE_BACK , "writeback" }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) /* Return MD raid4/5/6 journal mode for dm @journal_mode one */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) static int dm_raid_journal_mode_to_md(const char *mode)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) int m = ARRAY_SIZE(_raid456_journal_mode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) while (m--)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) if (!strcasecmp(mode, _raid456_journal_mode[m].param))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) return _raid456_journal_mode[m].mode;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) /* Return dm-raid raid4/5/6 journal mode string for @mode */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) static const char *md_journal_mode_to_dm_raid(const int mode)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) int m = ARRAY_SIZE(_raid456_journal_mode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) while (m--)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) if (mode == _raid456_journal_mode[m].mode)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) return _raid456_journal_mode[m].param;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) return "unknown";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) * Bool helpers to test for various raid levels of a raid set.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) * It's level as reported by the superblock rather than
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) * the requested raid_type passed to the constructor.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) /* Return true, if raid set in @rs is raid0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) static bool rs_is_raid0(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) return !rs->md.level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) /* Return true, if raid set in @rs is raid1 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) static bool rs_is_raid1(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) return rs->md.level == 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) /* Return true, if raid set in @rs is raid10 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) static bool rs_is_raid10(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) return rs->md.level == 10;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) /* Return true, if raid set in @rs is level 6 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) static bool rs_is_raid6(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) return rs->md.level == 6;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) /* Return true, if raid set in @rs is level 4, 5 or 6 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) static bool rs_is_raid456(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) return __within_range(rs->md.level, 4, 6);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) /* Return true, if raid set in @rs is reshapable */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) static bool __is_raid10_far(int layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) static bool rs_is_reshapable(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) return rs_is_raid456(rs) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) (rs_is_raid10(rs) && !__is_raid10_far(rs->md.new_layout));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) /* Return true, if raid set in @rs is recovering */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) static bool rs_is_recovering(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) return rs->md.recovery_cp < rs->md.dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) /* Return true, if raid set in @rs is reshaping */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) static bool rs_is_reshaping(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) return rs->md.reshape_position != MaxSector;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) * bool helpers to test for various raid levels of a raid type @rt
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) /* Return true, if raid type in @rt is raid0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) static bool rt_is_raid0(struct raid_type *rt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) return !rt->level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) /* Return true, if raid type in @rt is raid1 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) static bool rt_is_raid1(struct raid_type *rt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) return rt->level == 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) /* Return true, if raid type in @rt is raid10 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) static bool rt_is_raid10(struct raid_type *rt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) return rt->level == 10;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) /* Return true, if raid type in @rt is raid4/5 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471) static bool rt_is_raid45(struct raid_type *rt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) return __within_range(rt->level, 4, 5);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) /* Return true, if raid type in @rt is raid6 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) static bool rt_is_raid6(struct raid_type *rt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) return rt->level == 6;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) /* Return true, if raid type in @rt is raid4/5/6 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) static bool rt_is_raid456(struct raid_type *rt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) return __within_range(rt->level, 4, 6);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) /* END: raid level bools */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) /* Return valid ctr flags for the raid level of @rs */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) static unsigned long __valid_flags(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) if (rt_is_raid0(rs->raid_type))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) return RAID0_VALID_FLAGS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) else if (rt_is_raid1(rs->raid_type))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) return RAID1_VALID_FLAGS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) else if (rt_is_raid10(rs->raid_type))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) return RAID10_VALID_FLAGS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) else if (rt_is_raid45(rs->raid_type))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) return RAID45_VALID_FLAGS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) else if (rt_is_raid6(rs->raid_type))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) return RAID6_VALID_FLAGS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) * Check for valid flags set on @rs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) * Has to be called after parsing of the ctr flags!
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) static int rs_check_for_valid_flags(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) if (rs->ctr_flags & ~__valid_flags(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) rs->ti->error = "Invalid flags combination";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) /* MD raid10 bit definitions and helpers */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) #define RAID10_OFFSET (1 << 16) /* stripes with data copies area adjacent on devices */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) #define RAID10_BROCKEN_USE_FAR_SETS (1 << 17) /* Broken in raid10.c: use sets instead of whole stripe rotation */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) #define RAID10_USE_FAR_SETS (1 << 18) /* Use sets instead of whole stripe rotation */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525) #define RAID10_FAR_COPIES_SHIFT 8 /* raid10 # far copies shift (2nd byte of layout) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) /* Return md raid10 near copies for @layout */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) static unsigned int __raid10_near_copies(int layout)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) return layout & 0xFF;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) /* Return md raid10 far copies for @layout */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) static unsigned int __raid10_far_copies(int layout)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) return __raid10_near_copies(layout >> RAID10_FAR_COPIES_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) /* Return true if md raid10 offset for @layout */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) static bool __is_raid10_offset(int layout)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) return !!(layout & RAID10_OFFSET);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) /* Return true if md raid10 near for @layout */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) static bool __is_raid10_near(int layout)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) return !__is_raid10_offset(layout) && __raid10_near_copies(layout) > 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) /* Return true if md raid10 far for @layout */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) static bool __is_raid10_far(int layout)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) return !__is_raid10_offset(layout) && __raid10_far_copies(layout) > 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557) /* Return md raid10 layout string for @layout */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) static const char *raid10_md_layout_to_format(int layout)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) * Bit 16 stands for "offset"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) * (i.e. adjacent stripes hold copies)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) * Refer to MD's raid10.c for details
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) if (__is_raid10_offset(layout))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) return "offset";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) if (__raid10_near_copies(layout) > 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) return "near";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) if (__raid10_far_copies(layout) > 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) return "far";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) return "unknown";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578) /* Return md raid10 algorithm for @name */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579) static int raid10_name_to_format(const char *name)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581) if (!strcasecmp(name, "near"))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) return ALGORITHM_RAID10_NEAR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) else if (!strcasecmp(name, "offset"))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) return ALGORITHM_RAID10_OFFSET;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) else if (!strcasecmp(name, "far"))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) return ALGORITHM_RAID10_FAR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) /* Return md raid10 copies for @layout */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) static unsigned int raid10_md_layout_to_copies(int layout)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) return max(__raid10_near_copies(layout), __raid10_far_copies(layout));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) /* Return md raid10 format id for @format string */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598) static int raid10_format_to_md_layout(struct raid_set *rs,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599) unsigned int algorithm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) unsigned int copies)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) unsigned int n = 1, f = 1, r = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) * MD resilienece flaw:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) * enabling use_far_sets for far/offset formats causes copies
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) * to be colocated on the same devs together with their origins!
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610) * -> disable it for now in the definition above
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) if (algorithm == ALGORITHM_RAID10_DEFAULT ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613) algorithm == ALGORITHM_RAID10_NEAR)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) n = copies;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616) else if (algorithm == ALGORITHM_RAID10_OFFSET) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617) f = copies;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) r = RAID10_OFFSET;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619) if (!test_bit(__CTR_FLAG_RAID10_USE_NEAR_SETS, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) r |= RAID10_USE_FAR_SETS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622) } else if (algorithm == ALGORITHM_RAID10_FAR) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) f = copies;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) if (!test_bit(__CTR_FLAG_RAID10_USE_NEAR_SETS, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625) r |= RAID10_USE_FAR_SETS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) return r | (f << RAID10_FAR_COPIES_SHIFT) | n;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632) /* END: MD raid10 bit definitions and helpers */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) /* Check for any of the raid10 algorithms */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) static bool __got_raid10(struct raid_type *rtp, const int layout)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) if (rtp->level == 10) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638) switch (rtp->algorithm) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639) case ALGORITHM_RAID10_DEFAULT:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640) case ALGORITHM_RAID10_NEAR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) return __is_raid10_near(layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) case ALGORITHM_RAID10_OFFSET:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) return __is_raid10_offset(layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644) case ALGORITHM_RAID10_FAR:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) return __is_raid10_far(layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) /* Return raid_type for @name */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) static struct raid_type *get_raid_type(const char *name)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) struct raid_type *rtp = raid_types + ARRAY_SIZE(raid_types);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) while (rtp-- > raid_types)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) if (!strcasecmp(rtp->name, name))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) return rtp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666) /* Return raid_type for @name based derived from @level and @layout */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667) static struct raid_type *get_raid_type_by_ll(const int level, const int layout)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669) struct raid_type *rtp = raid_types + ARRAY_SIZE(raid_types);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) while (rtp-- > raid_types) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672) /* RAID10 special checks based on @layout flags/properties */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) if (rtp->level == level &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674) (__got_raid10(rtp, layout) || rtp->algorithm == layout))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) return rtp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) /* Adjust rdev sectors */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682) static void rs_set_rdev_sectors(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685) struct md_rdev *rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688) * raid10 sets rdev->sector to the device size, which
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689) * is unintended in case of out-of-place reshaping
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691) rdev_for_each(rdev, mddev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) if (!test_bit(Journal, &rdev->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) rdev->sectors = mddev->dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697) * Change bdev capacity of @rs in case of a disk add/remove reshape
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699) static void rs_set_capacity(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701) struct gendisk *gendisk = dm_disk(dm_table_get_md(rs->ti->table));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703) set_capacity(gendisk, rs->md.array_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) revalidate_disk_size(gendisk, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) * Set the mddev properties in @rs to the current
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) * ones retrieved from the freshest superblock
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711) static void rs_set_cur(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715) mddev->new_level = mddev->level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) mddev->new_layout = mddev->layout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) mddev->new_chunk_sectors = mddev->chunk_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721) * Set the mddev properties in @rs to the new
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) * ones requested by the ctr
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724) static void rs_set_new(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728) mddev->level = mddev->new_level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729) mddev->layout = mddev->new_layout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730) mddev->chunk_sectors = mddev->new_chunk_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731) mddev->raid_disks = rs->raid_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732) mddev->delta_disks = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735) static struct raid_set *raid_set_alloc(struct dm_target *ti, struct raid_type *raid_type,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736) unsigned int raid_devs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) unsigned int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) struct raid_set *rs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741) if (raid_devs <= raid_type->parity_devs) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) ti->error = "Insufficient number of devices";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743) return ERR_PTR(-EINVAL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) rs = kzalloc(struct_size(rs, dev, raid_devs), GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747) if (!rs) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) ti->error = "Cannot allocate raid context";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) return ERR_PTR(-ENOMEM);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752) mddev_init(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754) rs->raid_disks = raid_devs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) rs->delta_disks = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757) rs->ti = ti;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758) rs->raid_type = raid_type;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759) rs->stripe_cache_entries = 256;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) rs->md.raid_disks = raid_devs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) rs->md.level = raid_type->level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762) rs->md.new_level = rs->md.level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763) rs->md.layout = raid_type->algorithm;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764) rs->md.new_layout = rs->md.layout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) rs->md.delta_disks = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766) rs->md.recovery_cp = MaxSector;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768) for (i = 0; i < raid_devs; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) md_rdev_init(&rs->dev[i].rdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) * Remaining items to be initialized by further RAID params:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773) * rs->md.persistent
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774) * rs->md.external
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775) * rs->md.chunk_sectors
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776) * rs->md.new_chunk_sectors
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) * rs->md.dev_sectors
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) return rs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783) /* Free all @rs allocations */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784) static void raid_set_free(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) if (rs->journal_dev.dev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789) md_rdev_clear(&rs->journal_dev.rdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) dm_put_device(rs->ti, rs->journal_dev.dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) for (i = 0; i < rs->raid_disks; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) if (rs->dev[i].meta_dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795) dm_put_device(rs->ti, rs->dev[i].meta_dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796) md_rdev_clear(&rs->dev[i].rdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) if (rs->dev[i].data_dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798) dm_put_device(rs->ti, rs->dev[i].data_dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) kfree(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805) * For every device we have two words
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806) * <meta_dev>: meta device name or '-' if missing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) * <data_dev>: data device name or '-' if missing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809) * The following are permitted:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810) * - -
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) * - <data_dev>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812) * <meta_dev> <data_dev>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) * The following is not allowed:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815) * <meta_dev> -
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817) * This code parses those words. If there is a failure,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) * the caller must use raid_set_free() to unwind the operations.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820) static int parse_dev_params(struct raid_set *rs, struct dm_arg_set *as)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823) int rebuild = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) int metadata_available = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825) int r = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826) const char *arg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) /* Put off the number of raid devices argument to get to dev pairs */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) arg = dm_shift_arg(as);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830) if (!arg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833) for (i = 0; i < rs->raid_disks; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834) rs->dev[i].rdev.raid_disk = i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) rs->dev[i].meta_dev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837) rs->dev[i].data_dev = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840) * There are no offsets initially.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841) * Out of place reshape will set them accordingly.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) rs->dev[i].rdev.data_offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844) rs->dev[i].rdev.new_data_offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845) rs->dev[i].rdev.mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847) arg = dm_shift_arg(as);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848) if (!arg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851) if (strcmp(arg, "-")) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852) r = dm_get_device(rs->ti, arg, dm_table_get_mode(rs->ti->table),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) &rs->dev[i].meta_dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) if (r) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855) rs->ti->error = "RAID metadata device lookup failure";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859) rs->dev[i].rdev.sb_page = alloc_page(GFP_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860) if (!rs->dev[i].rdev.sb_page) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861) rs->ti->error = "Failed to allocate superblock page";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866) arg = dm_shift_arg(as);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) if (!arg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870) if (!strcmp(arg, "-")) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) if (!test_bit(In_sync, &rs->dev[i].rdev.flags) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872) (!rs->dev[i].rdev.recovery_offset)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873) rs->ti->error = "Drive designated for rebuild not specified";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877) if (rs->dev[i].meta_dev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878) rs->ti->error = "No data device supplied with metadata device";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885) r = dm_get_device(rs->ti, arg, dm_table_get_mode(rs->ti->table),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886) &rs->dev[i].data_dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887) if (r) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888) rs->ti->error = "RAID device lookup failure";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892) if (rs->dev[i].meta_dev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893) metadata_available = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) rs->dev[i].rdev.meta_bdev = rs->dev[i].meta_dev->bdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) rs->dev[i].rdev.bdev = rs->dev[i].data_dev->bdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897) list_add_tail(&rs->dev[i].rdev.same_set, &rs->md.disks);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898) if (!test_bit(In_sync, &rs->dev[i].rdev.flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899) rebuild++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902) if (rs->journal_dev.dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903) list_add_tail(&rs->journal_dev.rdev.same_set, &rs->md.disks);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905) if (metadata_available) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906) rs->md.external = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907) rs->md.persistent = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908) rs->md.major_version = 2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909) } else if (rebuild && !rs->md.recovery_cp) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) * Without metadata, we will not be able to tell if the array
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912) * is in-sync or not - we must assume it is not. Therefore,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) * it is impossible to rebuild a drive.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915) * Even if there is metadata, the on-disk information may
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916) * indicate that the array is not in-sync and it will then
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) * fail at that time.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 919) * User could specify 'nosync' option if desperate.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 920) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 921) rs->ti->error = "Unable to rebuild drive while array is not in-sync";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 922) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 923) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 924)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 925) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 926) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 927)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 928) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 929) * validate_region_size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 930) * @rs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 931) * @region_size: region size in sectors. If 0, pick a size (4MiB default).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 932) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 933) * Set rs->md.bitmap_info.chunksize (which really refers to 'region size').
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 934) * Ensure that (ti->len/region_size < 2^21) - required by MD bitmap.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 935) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 936) * Returns: 0 on success, -EINVAL on failure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 937) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 938) static int validate_region_size(struct raid_set *rs, unsigned long region_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 939) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 940) unsigned long min_region_size = rs->ti->len / (1 << 21);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 941)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 942) if (rs_is_raid0(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 943) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 944)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 945) if (!region_size) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 946) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 947) * Choose a reasonable default. All figures in sectors.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 948) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 949) if (min_region_size > (1 << 13)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 950) /* If not a power of 2, make it the next power of 2 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 951) region_size = roundup_pow_of_two(min_region_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 952) DMINFO("Choosing default region size of %lu sectors",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 953) region_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 954) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 955) DMINFO("Choosing default region size of 4MiB");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 956) region_size = 1 << 13; /* sectors */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 957) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 958) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 959) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 960) * Validate user-supplied value.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 961) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 962) if (region_size > rs->ti->len) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 963) rs->ti->error = "Supplied region size is too large";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 964) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 965) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 966)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 967) if (region_size < min_region_size) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 968) DMERR("Supplied region_size (%lu sectors) below minimum (%lu)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 969) region_size, min_region_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 970) rs->ti->error = "Supplied region size is too small";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 971) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 972) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 973)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 974) if (!is_power_of_2(region_size)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 975) rs->ti->error = "Region size is not a power of 2";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 976) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 977) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 978)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 979) if (region_size < rs->md.chunk_sectors) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 980) rs->ti->error = "Region size is smaller than the chunk size";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 981) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 982) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 983) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 984)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 985) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 986) * Convert sectors to bytes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 987) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 988) rs->md.bitmap_info.chunksize = to_bytes(region_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 989)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 990) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 991) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 992)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 993) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 994) * validate_raid_redundancy
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 995) * @rs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 996) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 997) * Determine if there are enough devices in the array that haven't
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 998) * failed (or are being rebuilt) to form a usable array.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 999) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1000) * Returns: 0 on success, -EINVAL on failure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1001) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1002) static int validate_raid_redundancy(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1003) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1004) unsigned int i, rebuild_cnt = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1005) unsigned int rebuilds_per_group = 0, copies;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1006) unsigned int group_size, last_group_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1007)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1008) for (i = 0; i < rs->md.raid_disks; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1009) if (!test_bit(In_sync, &rs->dev[i].rdev.flags) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1010) !rs->dev[i].rdev.sb_page)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1011) rebuild_cnt++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1012)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1013) switch (rs->md.level) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1014) case 0:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1015) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1016) case 1:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1017) if (rebuild_cnt >= rs->md.raid_disks)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1018) goto too_many;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1019) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1020) case 4:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1021) case 5:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1022) case 6:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1023) if (rebuild_cnt > rs->raid_type->parity_devs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1024) goto too_many;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1025) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1026) case 10:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1027) copies = raid10_md_layout_to_copies(rs->md.new_layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1028) if (copies < 2) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1029) DMERR("Bogus raid10 data copies < 2!");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1030) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1031) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1032)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1033) if (rebuild_cnt < copies)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1034) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1035)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1036) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1037) * It is possible to have a higher rebuild count for RAID10,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1038) * as long as the failed devices occur in different mirror
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1039) * groups (i.e. different stripes).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1040) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1041) * When checking "near" format, make sure no adjacent devices
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1042) * have failed beyond what can be handled. In addition to the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1043) * simple case where the number of devices is a multiple of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1044) * number of copies, we must also handle cases where the number
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1045) * of devices is not a multiple of the number of copies.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1046) * E.g. dev1 dev2 dev3 dev4 dev5
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1047) * A A B B C
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1048) * C D D E E
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1049) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1050) if (__is_raid10_near(rs->md.new_layout)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1051) for (i = 0; i < rs->md.raid_disks; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1052) if (!(i % copies))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1053) rebuilds_per_group = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1054) if ((!rs->dev[i].rdev.sb_page ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1055) !test_bit(In_sync, &rs->dev[i].rdev.flags)) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1056) (++rebuilds_per_group >= copies))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1057) goto too_many;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1058) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1059) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1060) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1061)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1062) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1063) * When checking "far" and "offset" formats, we need to ensure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1064) * that the device that holds its copy is not also dead or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1065) * being rebuilt. (Note that "far" and "offset" formats only
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1066) * support two copies right now. These formats also only ever
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1067) * use the 'use_far_sets' variant.)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1068) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1069) * This check is somewhat complicated by the need to account
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1070) * for arrays that are not a multiple of (far) copies. This
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1071) * results in the need to treat the last (potentially larger)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1072) * set differently.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1073) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1074) group_size = (rs->md.raid_disks / copies);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1075) last_group_start = (rs->md.raid_disks / group_size) - 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1076) last_group_start *= group_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1077) for (i = 0; i < rs->md.raid_disks; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1078) if (!(i % copies) && !(i > last_group_start))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1079) rebuilds_per_group = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1080) if ((!rs->dev[i].rdev.sb_page ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1081) !test_bit(In_sync, &rs->dev[i].rdev.flags)) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1082) (++rebuilds_per_group >= copies))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1083) goto too_many;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1084) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1085) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1086) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1087) if (rebuild_cnt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1088) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1089) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1090)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1091) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1092)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1093) too_many:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1094) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1095) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1096)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1097) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1098) * Possible arguments are...
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1099) * <chunk_size> [optional_args]
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1100) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1101) * Argument definitions
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1102) * <chunk_size> The number of sectors per disk that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1103) * will form the "stripe"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1104) * [[no]sync] Force or prevent recovery of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1105) * entire array
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1106) * [rebuild <idx>] Rebuild the drive indicated by the index
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1107) * [daemon_sleep <ms>] Time between bitmap daemon work to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1108) * clear bits
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1109) * [min_recovery_rate <kB/sec/disk>] Throttle RAID initialization
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1110) * [max_recovery_rate <kB/sec/disk>] Throttle RAID initialization
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1111) * [write_mostly <idx>] Indicate a write mostly drive via index
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1112) * [max_write_behind <sectors>] See '-write-behind=' (man mdadm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1113) * [stripe_cache <sectors>] Stripe cache size for higher RAIDs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1114) * [region_size <sectors>] Defines granularity of bitmap
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1115) * [journal_dev <dev>] raid4/5/6 journaling deviice
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1116) * (i.e. write hole closing log)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1117) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1118) * RAID10-only options:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1119) * [raid10_copies <# copies>] Number of copies. (Default: 2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1120) * [raid10_format <near|far|offset>] Layout algorithm. (Default: near)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1121) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1122) static int parse_raid_params(struct raid_set *rs, struct dm_arg_set *as,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1123) unsigned int num_raid_params)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1124) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1125) int value, raid10_format = ALGORITHM_RAID10_DEFAULT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1126) unsigned int raid10_copies = 2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1127) unsigned int i, write_mostly = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1128) unsigned int region_size = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1129) sector_t max_io_len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1130) const char *arg, *key;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1131) struct raid_dev *rd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1132) struct raid_type *rt = rs->raid_type;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1133)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1134) arg = dm_shift_arg(as);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1135) num_raid_params--; /* Account for chunk_size argument */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1136)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1137) if (kstrtoint(arg, 10, &value) < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1138) rs->ti->error = "Bad numerical argument given for chunk_size";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1139) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1140) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1141)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1142) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1143) * First, parse the in-order required arguments
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1144) * "chunk_size" is the only argument of this type.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1145) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1146) if (rt_is_raid1(rt)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1147) if (value)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1148) DMERR("Ignoring chunk size parameter for RAID 1");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1149) value = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1150) } else if (!is_power_of_2(value)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1151) rs->ti->error = "Chunk size must be a power of 2";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1152) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1153) } else if (value < 8) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1154) rs->ti->error = "Chunk size value is too small";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1155) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1156) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1157)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1158) rs->md.new_chunk_sectors = rs->md.chunk_sectors = value;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1159)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1160) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1161) * We set each individual device as In_sync with a completed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1162) * 'recovery_offset'. If there has been a device failure or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1163) * replacement then one of the following cases applies:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1164) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1165) * 1) User specifies 'rebuild'.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1166) * - Device is reset when param is read.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1167) * 2) A new device is supplied.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1168) * - No matching superblock found, resets device.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1169) * 3) Device failure was transient and returns on reload.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1170) * - Failure noticed, resets device for bitmap replay.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1171) * 4) Device hadn't completed recovery after previous failure.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1172) * - Superblock is read and overrides recovery_offset.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1173) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1174) * What is found in the superblocks of the devices is always
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1175) * authoritative, unless 'rebuild' or '[no]sync' was specified.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1176) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1177) for (i = 0; i < rs->raid_disks; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1178) set_bit(In_sync, &rs->dev[i].rdev.flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1179) rs->dev[i].rdev.recovery_offset = MaxSector;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1180) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1181)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1182) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1183) * Second, parse the unordered optional arguments
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1184) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1185) for (i = 0; i < num_raid_params; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1186) key = dm_shift_arg(as);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1187) if (!key) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1188) rs->ti->error = "Not enough raid parameters given";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1189) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1190) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1191)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1192) if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_NOSYNC))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1193) if (test_and_set_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1194) rs->ti->error = "Only one 'nosync' argument allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1195) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1196) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1197) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1198) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1199) if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_SYNC))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1200) if (test_and_set_bit(__CTR_FLAG_SYNC, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1201) rs->ti->error = "Only one 'sync' argument allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1202) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1203) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1204) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1205) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1206) if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_RAID10_USE_NEAR_SETS))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1207) if (test_and_set_bit(__CTR_FLAG_RAID10_USE_NEAR_SETS, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1208) rs->ti->error = "Only one 'raid10_use_new_sets' argument allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1209) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1210) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1211) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1212) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1213)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1214) arg = dm_shift_arg(as);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1215) i++; /* Account for the argument pairs */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1216) if (!arg) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1217) rs->ti->error = "Wrong number of raid parameters given";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1218) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1219) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1220)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1221) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1222) * Parameters that take a string value are checked here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1223) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1224) /* "raid10_format {near|offset|far} */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1225) if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_RAID10_FORMAT))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1226) if (test_and_set_bit(__CTR_FLAG_RAID10_FORMAT, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1227) rs->ti->error = "Only one 'raid10_format' argument pair allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1228) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1229) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1230) if (!rt_is_raid10(rt)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1231) rs->ti->error = "'raid10_format' is an invalid parameter for this RAID type";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1232) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1233) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1234) raid10_format = raid10_name_to_format(arg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1235) if (raid10_format < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1236) rs->ti->error = "Invalid 'raid10_format' value given";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1237) return raid10_format;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1238) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1239) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1240) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1241)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1242) /* "journal_dev <dev>" */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1243) if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_JOURNAL_DEV))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1244) int r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1245) struct md_rdev *jdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1246)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1247) if (test_and_set_bit(__CTR_FLAG_JOURNAL_DEV, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1248) rs->ti->error = "Only one raid4/5/6 set journaling device allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1249) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1250) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1251) if (!rt_is_raid456(rt)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1252) rs->ti->error = "'journal_dev' is an invalid parameter for this RAID type";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1253) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1254) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1255) r = dm_get_device(rs->ti, arg, dm_table_get_mode(rs->ti->table),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1256) &rs->journal_dev.dev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1257) if (r) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1258) rs->ti->error = "raid4/5/6 journal device lookup failure";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1259) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1260) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1261) jdev = &rs->journal_dev.rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1262) md_rdev_init(jdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1263) jdev->mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1264) jdev->bdev = rs->journal_dev.dev->bdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1265) jdev->sectors = to_sector(i_size_read(jdev->bdev->bd_inode));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1266) if (jdev->sectors < MIN_RAID456_JOURNAL_SPACE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1267) rs->ti->error = "No space for raid4/5/6 journal";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1268) return -ENOSPC;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1269) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1270) rs->journal_dev.mode = R5C_JOURNAL_MODE_WRITE_THROUGH;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1271) set_bit(Journal, &jdev->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1272) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1273) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1274)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1275) /* "journal_mode <mode>" ("journal_dev" mandatory!) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1276) if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_JOURNAL_MODE))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1277) int r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1278)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1279) if (!test_bit(__CTR_FLAG_JOURNAL_DEV, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1280) rs->ti->error = "raid4/5/6 'journal_mode' is invalid without 'journal_dev'";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1281) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1282) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1283) if (test_and_set_bit(__CTR_FLAG_JOURNAL_MODE, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1284) rs->ti->error = "Only one raid4/5/6 'journal_mode' argument allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1285) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1286) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1287) r = dm_raid_journal_mode_to_md(arg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1288) if (r < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1289) rs->ti->error = "Invalid 'journal_mode' argument";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1290) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1291) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1292) rs->journal_dev.mode = r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1293) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1294) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1295)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1296) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1297) * Parameters with number values from here on.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1298) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1299) if (kstrtoint(arg, 10, &value) < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1300) rs->ti->error = "Bad numerical argument given in raid params";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1301) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1302) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1303)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1304) if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_REBUILD))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1305) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1306) * "rebuild" is being passed in by userspace to provide
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1307) * indexes of replaced devices and to set up additional
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1308) * devices on raid level takeover.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1309) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1310) if (!__within_range(value, 0, rs->raid_disks - 1)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1311) rs->ti->error = "Invalid rebuild index given";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1312) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1313) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1314)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1315) if (test_and_set_bit(value, (void *) rs->rebuild_disks)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1316) rs->ti->error = "rebuild for this index already given";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1317) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1318) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1319)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1320) rd = rs->dev + value;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1321) clear_bit(In_sync, &rd->rdev.flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1322) clear_bit(Faulty, &rd->rdev.flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1323) rd->rdev.recovery_offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1324) set_bit(__CTR_FLAG_REBUILD, &rs->ctr_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1325) } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_WRITE_MOSTLY))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1326) if (!rt_is_raid1(rt)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1327) rs->ti->error = "write_mostly option is only valid for RAID1";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1328) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1329) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1330)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1331) if (!__within_range(value, 0, rs->md.raid_disks - 1)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1332) rs->ti->error = "Invalid write_mostly index given";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1333) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1334) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1335)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1336) write_mostly++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1337) set_bit(WriteMostly, &rs->dev[value].rdev.flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1338) set_bit(__CTR_FLAG_WRITE_MOSTLY, &rs->ctr_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1339) } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_MAX_WRITE_BEHIND))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1340) if (!rt_is_raid1(rt)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1341) rs->ti->error = "max_write_behind option is only valid for RAID1";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1342) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1343) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1344)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1345) if (test_and_set_bit(__CTR_FLAG_MAX_WRITE_BEHIND, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1346) rs->ti->error = "Only one max_write_behind argument pair allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1347) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1348) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1349)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1350) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1351) * In device-mapper, we specify things in sectors, but
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1352) * MD records this value in kB
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1353) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1354) if (value < 0 || value / 2 > COUNTER_MAX) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1355) rs->ti->error = "Max write-behind limit out of range";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1356) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1357) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1358)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1359) rs->md.bitmap_info.max_write_behind = value / 2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1360) } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_DAEMON_SLEEP))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1361) if (test_and_set_bit(__CTR_FLAG_DAEMON_SLEEP, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1362) rs->ti->error = "Only one daemon_sleep argument pair allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1363) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1364) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1365) if (value < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1366) rs->ti->error = "daemon sleep period out of range";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1367) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1368) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1369) rs->md.bitmap_info.daemon_sleep = value;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1370) } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_DATA_OFFSET))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1371) /* Userspace passes new data_offset after having extended the the data image LV */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1372) if (test_and_set_bit(__CTR_FLAG_DATA_OFFSET, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1373) rs->ti->error = "Only one data_offset argument pair allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1374) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1375) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1376) /* Ensure sensible data offset */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1377) if (value < 0 ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1378) (value && (value < MIN_FREE_RESHAPE_SPACE || value % to_sector(PAGE_SIZE)))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1379) rs->ti->error = "Bogus data_offset value";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1380) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1381) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1382) rs->data_offset = value;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1383) } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_DELTA_DISKS))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1384) /* Define the +/-# of disks to add to/remove from the given raid set */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1385) if (test_and_set_bit(__CTR_FLAG_DELTA_DISKS, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1386) rs->ti->error = "Only one delta_disks argument pair allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1387) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1388) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1389) /* Ensure MAX_RAID_DEVICES and raid type minimal_devs! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1390) if (!__within_range(abs(value), 1, MAX_RAID_DEVICES - rt->minimal_devs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1391) rs->ti->error = "Too many delta_disk requested";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1392) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1393) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1394)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1395) rs->delta_disks = value;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1396) } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_STRIPE_CACHE))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1397) if (test_and_set_bit(__CTR_FLAG_STRIPE_CACHE, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1398) rs->ti->error = "Only one stripe_cache argument pair allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1399) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1400) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1402) if (!rt_is_raid456(rt)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1403) rs->ti->error = "Inappropriate argument: stripe_cache";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1404) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1405) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1406)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1407) if (value < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1408) rs->ti->error = "Bogus stripe cache entries value";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1409) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1410) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1411) rs->stripe_cache_entries = value;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1412) } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_MIN_RECOVERY_RATE))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1413) if (test_and_set_bit(__CTR_FLAG_MIN_RECOVERY_RATE, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1414) rs->ti->error = "Only one min_recovery_rate argument pair allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1415) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1416) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1417)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1418) if (value < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1419) rs->ti->error = "min_recovery_rate out of range";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1420) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1421) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1422) rs->md.sync_speed_min = value;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1423) } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_MAX_RECOVERY_RATE))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1424) if (test_and_set_bit(__CTR_FLAG_MAX_RECOVERY_RATE, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1425) rs->ti->error = "Only one max_recovery_rate argument pair allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1426) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1427) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1428)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1429) if (value < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1430) rs->ti->error = "max_recovery_rate out of range";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1431) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1432) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1433) rs->md.sync_speed_max = value;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1434) } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_REGION_SIZE))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1435) if (test_and_set_bit(__CTR_FLAG_REGION_SIZE, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1436) rs->ti->error = "Only one region_size argument pair allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1437) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1438) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1439)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1440) region_size = value;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1441) rs->requested_bitmap_chunk_sectors = value;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1442) } else if (!strcasecmp(key, dm_raid_arg_name_by_flag(CTR_FLAG_RAID10_COPIES))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1443) if (test_and_set_bit(__CTR_FLAG_RAID10_COPIES, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1444) rs->ti->error = "Only one raid10_copies argument pair allowed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1445) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1446) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1447)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1448) if (!__within_range(value, 2, rs->md.raid_disks)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1449) rs->ti->error = "Bad value for 'raid10_copies'";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1450) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1451) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1452)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1453) raid10_copies = value;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1454) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1455) DMERR("Unable to parse RAID parameter: %s", key);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1456) rs->ti->error = "Unable to parse RAID parameter";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1457) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1458) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1459) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1460)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1461) if (test_bit(__CTR_FLAG_SYNC, &rs->ctr_flags) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1462) test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1463) rs->ti->error = "sync and nosync are mutually exclusive";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1464) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1465) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1466)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1467) if (test_bit(__CTR_FLAG_REBUILD, &rs->ctr_flags) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1468) (test_bit(__CTR_FLAG_SYNC, &rs->ctr_flags) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1469) test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1470) rs->ti->error = "sync/nosync and rebuild are mutually exclusive";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1471) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1472) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1473)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1474) if (write_mostly >= rs->md.raid_disks) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1475) rs->ti->error = "Can't set all raid1 devices to write_mostly";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1476) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1477) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1478)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1479) if (rs->md.sync_speed_max &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1480) rs->md.sync_speed_min > rs->md.sync_speed_max) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1481) rs->ti->error = "Bogus recovery rates";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1482) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1483) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1484)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1485) if (validate_region_size(rs, region_size))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1486) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1487)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1488) if (rs->md.chunk_sectors)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1489) max_io_len = rs->md.chunk_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1490) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1491) max_io_len = region_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1492)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1493) if (dm_set_target_max_io_len(rs->ti, max_io_len))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1494) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1495)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1496) if (rt_is_raid10(rt)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1497) if (raid10_copies > rs->md.raid_disks) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1498) rs->ti->error = "Not enough devices to satisfy specification";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1499) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1500) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1501)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1502) rs->md.new_layout = raid10_format_to_md_layout(rs, raid10_format, raid10_copies);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1503) if (rs->md.new_layout < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1504) rs->ti->error = "Error getting raid10 format";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1505) return rs->md.new_layout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1506) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1507)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1508) rt = get_raid_type_by_ll(10, rs->md.new_layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1509) if (!rt) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1510) rs->ti->error = "Failed to recognize new raid10 layout";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1511) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1512) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1513)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1514) if ((rt->algorithm == ALGORITHM_RAID10_DEFAULT ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1515) rt->algorithm == ALGORITHM_RAID10_NEAR) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1516) test_bit(__CTR_FLAG_RAID10_USE_NEAR_SETS, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1517) rs->ti->error = "RAID10 format 'near' and 'raid10_use_near_sets' are incompatible";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1518) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1519) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1520) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1521)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1522) rs->raid10_copies = raid10_copies;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1523)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1524) /* Assume there are no metadata devices until the drives are parsed */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1525) rs->md.persistent = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1526) rs->md.external = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1527)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1528) /* Check, if any invalid ctr arguments have been passed in for the raid level */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1529) return rs_check_for_valid_flags(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1530) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1531)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1532) /* Set raid4/5/6 cache size */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1533) static int rs_set_raid456_stripe_cache(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1534) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1535) int r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1536) struct r5conf *conf;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1537) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1538) uint32_t min_stripes = max(mddev->chunk_sectors, mddev->new_chunk_sectors) / 2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1539) uint32_t nr_stripes = rs->stripe_cache_entries;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1540)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1541) if (!rt_is_raid456(rs->raid_type)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1542) rs->ti->error = "Inappropriate raid level; cannot change stripe_cache size";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1543) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1544) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1545)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1546) if (nr_stripes < min_stripes) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1547) DMINFO("Adjusting requested %u stripe cache entries to %u to suit stripe size",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1548) nr_stripes, min_stripes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1549) nr_stripes = min_stripes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1550) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1551)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1552) conf = mddev->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1553) if (!conf) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1554) rs->ti->error = "Cannot change stripe_cache size on inactive RAID set";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1555) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1556) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1557)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1558) /* Try setting number of stripes in raid456 stripe cache */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1559) if (conf->min_nr_stripes != nr_stripes) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1560) r = raid5_set_cache_size(mddev, nr_stripes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1561) if (r) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1562) rs->ti->error = "Failed to set raid4/5/6 stripe cache size";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1563) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1564) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1565)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1566) DMINFO("%u stripe cache entries", nr_stripes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1567) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1568)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1569) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1570) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1571)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1572) /* Return # of data stripes as kept in mddev as of @rs (i.e. as of superblock) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1573) static unsigned int mddev_data_stripes(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1574) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1575) return rs->md.raid_disks - rs->raid_type->parity_devs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1576) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1577)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1578) /* Return # of data stripes of @rs (i.e. as of ctr) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1579) static unsigned int rs_data_stripes(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1580) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1581) return rs->raid_disks - rs->raid_type->parity_devs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1582) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1583)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1584) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1585) * Retrieve rdev->sectors from any valid raid device of @rs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1586) * to allow userpace to pass in arbitray "- -" device tupples.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1587) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1588) static sector_t __rdev_sectors(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1589) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1590) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1591)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1592) for (i = 0; i < rs->md.raid_disks; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1593) struct md_rdev *rdev = &rs->dev[i].rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1594)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1595) if (!test_bit(Journal, &rdev->flags) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1596) rdev->bdev && rdev->sectors)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1597) return rdev->sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1598) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1599)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1600) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1601) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1602)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1603) /* Check that calculated dev_sectors fits all component devices. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1604) static int _check_data_dev_sectors(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1605) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1606) sector_t ds = ~0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1607) struct md_rdev *rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1608)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1609) rdev_for_each(rdev, &rs->md)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1610) if (!test_bit(Journal, &rdev->flags) && rdev->bdev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1611) ds = min(ds, to_sector(i_size_read(rdev->bdev->bd_inode)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1612) if (ds < rs->md.dev_sectors) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1613) rs->ti->error = "Component device(s) too small";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1614) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1615) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1616) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1617)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1618) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1619) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1620)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1621) /* Calculate the sectors per device and per array used for @rs */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1622) static int rs_set_dev_and_array_sectors(struct raid_set *rs, sector_t sectors, bool use_mddev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1623) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1624) int delta_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1625) unsigned int data_stripes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1626) sector_t array_sectors = sectors, dev_sectors = sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1627) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1628)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1629) if (use_mddev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1630) delta_disks = mddev->delta_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1631) data_stripes = mddev_data_stripes(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1632) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1633) delta_disks = rs->delta_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1634) data_stripes = rs_data_stripes(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1635) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1636)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1637) /* Special raid1 case w/o delta_disks support (yet) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1638) if (rt_is_raid1(rs->raid_type))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1639) ;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1640) else if (rt_is_raid10(rs->raid_type)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1641) if (rs->raid10_copies < 2 ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1642) delta_disks < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1643) rs->ti->error = "Bogus raid10 data copies or delta disks";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1644) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1645) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1646)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1647) dev_sectors *= rs->raid10_copies;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1648) if (sector_div(dev_sectors, data_stripes))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1649) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1650)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1651) array_sectors = (data_stripes + delta_disks) * dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1652) if (sector_div(array_sectors, rs->raid10_copies))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1653) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1654)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1655) } else if (sector_div(dev_sectors, data_stripes))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1656) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1657)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1658) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1659) /* Striped layouts */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1660) array_sectors = (data_stripes + delta_disks) * dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1661)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1662) mddev->array_sectors = array_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1663) mddev->dev_sectors = dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1664) rs_set_rdev_sectors(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1665)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1666) return _check_data_dev_sectors(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1667) bad:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1668) rs->ti->error = "Target length not divisible by number of data devices";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1669) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1670) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1671)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1672) /* Setup recovery on @rs */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1673) static void rs_setup_recovery(struct raid_set *rs, sector_t dev_sectors)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1674) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1675) /* raid0 does not recover */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1676) if (rs_is_raid0(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1677) rs->md.recovery_cp = MaxSector;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1678) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1679) * A raid6 set has to be recovered either
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1680) * completely or for the grown part to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1681) * ensure proper parity and Q-Syndrome
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1682) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1683) else if (rs_is_raid6(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1684) rs->md.recovery_cp = dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1685) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1686) * Other raid set types may skip recovery
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1687) * depending on the 'nosync' flag.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1688) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1689) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1690) rs->md.recovery_cp = test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1691) ? MaxSector : dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1692) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1693)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1694) static void do_table_event(struct work_struct *ws)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1695) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1696) struct raid_set *rs = container_of(ws, struct raid_set, md.event_work);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1697)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1698) smp_rmb(); /* Make sure we access most actual mddev properties */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1699) if (!rs_is_reshaping(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1700) if (rs_is_raid10(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1701) rs_set_rdev_sectors(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1702) rs_set_capacity(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1703) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1704) dm_table_event(rs->ti->table);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1705) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1706)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1707) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1708) * Make sure a valid takover (level switch) is being requested on @rs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1709) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1710) * Conversions of raid sets from one MD personality to another
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1711) * have to conform to restrictions which are enforced here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1712) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1713) static int rs_check_takeover(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1714) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1715) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1716) unsigned int near_copies;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1717)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1718) if (rs->md.degraded) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1719) rs->ti->error = "Can't takeover degraded raid set";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1720) return -EPERM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1721) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1722)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1723) if (rs_is_reshaping(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1724) rs->ti->error = "Can't takeover reshaping raid set";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1725) return -EPERM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1726) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1727)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1728) switch (mddev->level) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1729) case 0:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1730) /* raid0 -> raid1/5 with one disk */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1731) if ((mddev->new_level == 1 || mddev->new_level == 5) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1732) mddev->raid_disks == 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1733) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1734)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1735) /* raid0 -> raid10 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1736) if (mddev->new_level == 10 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1737) !(rs->raid_disks % mddev->raid_disks))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1738) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1739)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1740) /* raid0 with multiple disks -> raid4/5/6 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1741) if (__within_range(mddev->new_level, 4, 6) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1742) mddev->new_layout == ALGORITHM_PARITY_N &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1743) mddev->raid_disks > 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1744) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1745)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1746) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1747)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1748) case 10:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1749) /* Can't takeover raid10_offset! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1750) if (__is_raid10_offset(mddev->layout))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1751) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1752)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1753) near_copies = __raid10_near_copies(mddev->layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1754)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1755) /* raid10* -> raid0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1756) if (mddev->new_level == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1757) /* Can takeover raid10_near with raid disks divisable by data copies! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1758) if (near_copies > 1 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1759) !(mddev->raid_disks % near_copies)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1760) mddev->raid_disks /= near_copies;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1761) mddev->delta_disks = mddev->raid_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1762) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1763) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1764)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1765) /* Can takeover raid10_far */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1766) if (near_copies == 1 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1767) __raid10_far_copies(mddev->layout) > 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1768) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1769)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1770) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1771) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1772)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1773) /* raid10_{near,far} -> raid1 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1774) if (mddev->new_level == 1 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1775) max(near_copies, __raid10_far_copies(mddev->layout)) == mddev->raid_disks)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1776) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1777)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1778) /* raid10_{near,far} with 2 disks -> raid4/5 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1779) if (__within_range(mddev->new_level, 4, 5) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1780) mddev->raid_disks == 2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1781) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1782) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1783)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1784) case 1:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1785) /* raid1 with 2 disks -> raid4/5 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1786) if (__within_range(mddev->new_level, 4, 5) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1787) mddev->raid_disks == 2) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1788) mddev->degraded = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1789) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1790) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1791)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1792) /* raid1 -> raid0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1793) if (mddev->new_level == 0 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1794) mddev->raid_disks == 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1795) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1796)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1797) /* raid1 -> raid10 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1798) if (mddev->new_level == 10)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1799) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1800) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1801)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1802) case 4:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1803) /* raid4 -> raid0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1804) if (mddev->new_level == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1805) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1806)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1807) /* raid4 -> raid1/5 with 2 disks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1808) if ((mddev->new_level == 1 || mddev->new_level == 5) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1809) mddev->raid_disks == 2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1810) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1811)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1812) /* raid4 -> raid5/6 with parity N */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1813) if (__within_range(mddev->new_level, 5, 6) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1814) mddev->layout == ALGORITHM_PARITY_N)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1815) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1816) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1817)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1818) case 5:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1819) /* raid5 with parity N -> raid0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1820) if (mddev->new_level == 0 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1821) mddev->layout == ALGORITHM_PARITY_N)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1822) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1823)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1824) /* raid5 with parity N -> raid4 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1825) if (mddev->new_level == 4 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1826) mddev->layout == ALGORITHM_PARITY_N)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1827) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1828)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1829) /* raid5 with 2 disks -> raid1/4/10 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1830) if ((mddev->new_level == 1 || mddev->new_level == 4 || mddev->new_level == 10) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1831) mddev->raid_disks == 2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1832) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1833)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1834) /* raid5_* -> raid6_*_6 with Q-Syndrome N (e.g. raid5_ra -> raid6_ra_6 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1835) if (mddev->new_level == 6 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1836) ((mddev->layout == ALGORITHM_PARITY_N && mddev->new_layout == ALGORITHM_PARITY_N) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1837) __within_range(mddev->new_layout, ALGORITHM_LEFT_ASYMMETRIC_6, ALGORITHM_RIGHT_SYMMETRIC_6)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1838) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1839) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1840)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1841) case 6:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1842) /* raid6 with parity N -> raid0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1843) if (mddev->new_level == 0 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1844) mddev->layout == ALGORITHM_PARITY_N)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1845) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1846)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1847) /* raid6 with parity N -> raid4 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1848) if (mddev->new_level == 4 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1849) mddev->layout == ALGORITHM_PARITY_N)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1850) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1851)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1852) /* raid6_*_n with Q-Syndrome N -> raid5_* */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1853) if (mddev->new_level == 5 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1854) ((mddev->layout == ALGORITHM_PARITY_N && mddev->new_layout == ALGORITHM_PARITY_N) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1855) __within_range(mddev->new_layout, ALGORITHM_LEFT_ASYMMETRIC, ALGORITHM_RIGHT_SYMMETRIC)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1856) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1857)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1858) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1859) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1860) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1861)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1862) rs->ti->error = "takeover not possible";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1863) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1864) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1865)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1866) /* True if @rs requested to be taken over */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1867) static bool rs_takeover_requested(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1868) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1869) return rs->md.new_level != rs->md.level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1870) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1871)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1872) /* True if layout is set to reshape. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1873) static bool rs_is_layout_change(struct raid_set *rs, bool use_mddev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1874) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1875) return (use_mddev ? rs->md.delta_disks : rs->delta_disks) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1876) rs->md.new_layout != rs->md.layout ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1877) rs->md.new_chunk_sectors != rs->md.chunk_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1878) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1879)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1880) /* True if @rs is requested to reshape by ctr */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1881) static bool rs_reshape_requested(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1882) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1883) bool change;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1884) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1885)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1886) if (rs_takeover_requested(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1887) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1888)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1889) if (rs_is_raid0(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1890) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1891)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1892) change = rs_is_layout_change(rs, false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1893)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1894) /* Historical case to support raid1 reshape without delta disks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1895) if (rs_is_raid1(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1896) if (rs->delta_disks)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1897) return !!rs->delta_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1898)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1899) return !change &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1900) mddev->raid_disks != rs->raid_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1901) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1902)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1903) if (rs_is_raid10(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1904) return change &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1905) !__is_raid10_far(mddev->new_layout) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1906) rs->delta_disks >= 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1907)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1908) return change;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1909) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1910)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1911) /* Features */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1912) #define FEATURE_FLAG_SUPPORTS_V190 0x1 /* Supports extended superblock */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1913)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1914) /* State flags for sb->flags */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1915) #define SB_FLAG_RESHAPE_ACTIVE 0x1
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1916) #define SB_FLAG_RESHAPE_BACKWARDS 0x2
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1917)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1918) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1919) * This structure is never routinely used by userspace, unlike md superblocks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1920) * Devices with this superblock should only ever be accessed via device-mapper.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1921) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1922) #define DM_RAID_MAGIC 0x64526D44
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1923) struct dm_raid_superblock {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1924) __le32 magic; /* "DmRd" */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1925) __le32 compat_features; /* Used to indicate compatible features (like 1.9.0 ondisk metadata extension) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1926)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1927) __le32 num_devices; /* Number of devices in this raid set. (Max 64) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1928) __le32 array_position; /* The position of this drive in the raid set */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1929)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1930) __le64 events; /* Incremented by md when superblock updated */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1931) __le64 failed_devices; /* Pre 1.9.0 part of bit field of devices to */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1932) /* indicate failures (see extension below) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1933)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1934) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1935) * This offset tracks the progress of the repair or replacement of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1936) * an individual drive.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1937) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1938) __le64 disk_recovery_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1939)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1940) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1941) * This offset tracks the progress of the initial raid set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1942) * synchronisation/parity calculation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1943) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1944) __le64 array_resync_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1945)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1946) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1947) * raid characteristics
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1948) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1949) __le32 level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1950) __le32 layout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1951) __le32 stripe_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1952)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1953) /********************************************************************
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1954) * BELOW FOLLOW V1.9.0 EXTENSIONS TO THE PRISTINE SUPERBLOCK FORMAT!!!
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1955) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1956) * FEATURE_FLAG_SUPPORTS_V190 in the compat_features member indicates that those exist
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1957) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1958)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1959) __le32 flags; /* Flags defining array states for reshaping */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1960)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1961) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1962) * This offset tracks the progress of a raid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1963) * set reshape in order to be able to restart it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1964) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1965) __le64 reshape_position;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1966)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1967) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1968) * These define the properties of the array in case of an interrupted reshape
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1969) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1970) __le32 new_level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1971) __le32 new_layout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1972) __le32 new_stripe_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1973) __le32 delta_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1974)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1975) __le64 array_sectors; /* Array size in sectors */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1976)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1977) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1978) * Sector offsets to data on devices (reshaping).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1979) * Needed to support out of place reshaping, thus
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1980) * not writing over any stripes whilst converting
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1981) * them from old to new layout
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1982) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1983) __le64 data_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1984) __le64 new_data_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1985)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1986) __le64 sectors; /* Used device size in sectors */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1987)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1988) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1989) * Additonal Bit field of devices indicating failures to support
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1990) * up to 256 devices with the 1.9.0 on-disk metadata format
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1991) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1992) __le64 extended_failed_devices[DISKS_ARRAY_ELEMS - 1];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1993)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1994) __le32 incompat_features; /* Used to indicate any incompatible features */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1995)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1996) /* Always set rest up to logical block size to 0 when writing (see get_metadata_device() below). */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1997) } __packed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1998)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1999) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2000) * Check for reshape constraints on raid set @rs:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2001) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2002) * - reshape function non-existent
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2003) * - degraded set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2004) * - ongoing recovery
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2005) * - ongoing reshape
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2006) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2007) * Returns 0 if none or -EPERM if given constraint
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2008) * and error message reference in @errmsg
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2009) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2010) static int rs_check_reshape(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2011) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2012) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2013)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2014) if (!mddev->pers || !mddev->pers->check_reshape)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2015) rs->ti->error = "Reshape not supported";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2016) else if (mddev->degraded)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2017) rs->ti->error = "Can't reshape degraded raid set";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2018) else if (rs_is_recovering(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2019) rs->ti->error = "Convert request on recovering raid set prohibited";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2020) else if (rs_is_reshaping(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2021) rs->ti->error = "raid set already reshaping!";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2022) else if (!(rs_is_raid1(rs) || rs_is_raid10(rs) || rs_is_raid456(rs)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2023) rs->ti->error = "Reshaping only supported for raid1/4/5/6/10";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2024) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2025) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2026)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2027) return -EPERM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2028) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2029)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2030) static int read_disk_sb(struct md_rdev *rdev, int size, bool force_reload)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2031) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2032) BUG_ON(!rdev->sb_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2033)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2034) if (rdev->sb_loaded && !force_reload)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2035) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2036)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2037) rdev->sb_loaded = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2038)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2039) if (!sync_page_io(rdev, 0, size, rdev->sb_page, REQ_OP_READ, 0, true)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2040) DMERR("Failed to read superblock of device at position %d",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2041) rdev->raid_disk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2042) md_error(rdev->mddev, rdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2043) set_bit(Faulty, &rdev->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2044) return -EIO;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2045) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2046)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2047) rdev->sb_loaded = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2048)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2049) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2050) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2051)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2052) static void sb_retrieve_failed_devices(struct dm_raid_superblock *sb, uint64_t *failed_devices)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2053) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2054) failed_devices[0] = le64_to_cpu(sb->failed_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2055) memset(failed_devices + 1, 0, sizeof(sb->extended_failed_devices));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2056)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2057) if (le32_to_cpu(sb->compat_features) & FEATURE_FLAG_SUPPORTS_V190) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2058) int i = ARRAY_SIZE(sb->extended_failed_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2059)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2060) while (i--)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2061) failed_devices[i+1] = le64_to_cpu(sb->extended_failed_devices[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2062) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2063) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2064)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2065) static void sb_update_failed_devices(struct dm_raid_superblock *sb, uint64_t *failed_devices)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2066) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2067) int i = ARRAY_SIZE(sb->extended_failed_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2068)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2069) sb->failed_devices = cpu_to_le64(failed_devices[0]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2070) while (i--)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2071) sb->extended_failed_devices[i] = cpu_to_le64(failed_devices[i+1]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2072) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2073)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2074) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2075) * Synchronize the superblock members with the raid set properties
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2076) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2077) * All superblock data is little endian.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2078) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2079) static void super_sync(struct mddev *mddev, struct md_rdev *rdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2080) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2081) bool update_failed_devices = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2082) unsigned int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2083) uint64_t failed_devices[DISKS_ARRAY_ELEMS];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2084) struct dm_raid_superblock *sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2085) struct raid_set *rs = container_of(mddev, struct raid_set, md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2086)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2087) /* No metadata device, no superblock */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2088) if (!rdev->meta_bdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2089) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2090)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2091) BUG_ON(!rdev->sb_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2092)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2093) sb = page_address(rdev->sb_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2094)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2095) sb_retrieve_failed_devices(sb, failed_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2096)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2097) for (i = 0; i < rs->raid_disks; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2098) if (!rs->dev[i].data_dev || test_bit(Faulty, &rs->dev[i].rdev.flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2099) update_failed_devices = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2100) set_bit(i, (void *) failed_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2101) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2102)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2103) if (update_failed_devices)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2104) sb_update_failed_devices(sb, failed_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2105)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2106) sb->magic = cpu_to_le32(DM_RAID_MAGIC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2107) sb->compat_features = cpu_to_le32(FEATURE_FLAG_SUPPORTS_V190);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2108)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2109) sb->num_devices = cpu_to_le32(mddev->raid_disks);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2110) sb->array_position = cpu_to_le32(rdev->raid_disk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2111)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2112) sb->events = cpu_to_le64(mddev->events);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2113)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2114) sb->disk_recovery_offset = cpu_to_le64(rdev->recovery_offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2115) sb->array_resync_offset = cpu_to_le64(mddev->recovery_cp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2117) sb->level = cpu_to_le32(mddev->level);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2118) sb->layout = cpu_to_le32(mddev->layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2119) sb->stripe_sectors = cpu_to_le32(mddev->chunk_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2120)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2121) /********************************************************************
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2122) * BELOW FOLLOW V1.9.0 EXTENSIONS TO THE PRISTINE SUPERBLOCK FORMAT!!!
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2123) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2124) * FEATURE_FLAG_SUPPORTS_V190 in the compat_features member indicates that those exist
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2125) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2126) sb->new_level = cpu_to_le32(mddev->new_level);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2127) sb->new_layout = cpu_to_le32(mddev->new_layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2128) sb->new_stripe_sectors = cpu_to_le32(mddev->new_chunk_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2129)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2130) sb->delta_disks = cpu_to_le32(mddev->delta_disks);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2131)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2132) smp_rmb(); /* Make sure we access most recent reshape position */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2133) sb->reshape_position = cpu_to_le64(mddev->reshape_position);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2134) if (le64_to_cpu(sb->reshape_position) != MaxSector) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2135) /* Flag ongoing reshape */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2136) sb->flags |= cpu_to_le32(SB_FLAG_RESHAPE_ACTIVE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2137)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2138) if (mddev->delta_disks < 0 || mddev->reshape_backwards)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2139) sb->flags |= cpu_to_le32(SB_FLAG_RESHAPE_BACKWARDS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2140) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2141) /* Clear reshape flags */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2142) sb->flags &= ~(cpu_to_le32(SB_FLAG_RESHAPE_ACTIVE|SB_FLAG_RESHAPE_BACKWARDS));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2143) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2144)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2145) sb->array_sectors = cpu_to_le64(mddev->array_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2146) sb->data_offset = cpu_to_le64(rdev->data_offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2147) sb->new_data_offset = cpu_to_le64(rdev->new_data_offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2148) sb->sectors = cpu_to_le64(rdev->sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2149) sb->incompat_features = cpu_to_le32(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2150)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2151) /* Zero out the rest of the payload after the size of the superblock */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2152) memset(sb + 1, 0, rdev->sb_size - sizeof(*sb));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2153) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2154)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2155) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2156) * super_load
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2157) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2158) * This function creates a superblock if one is not found on the device
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2159) * and will decide which superblock to use if there's a choice.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2160) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2161) * Return: 1 if use rdev, 0 if use refdev, -Exxx otherwise
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2162) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2163) static int super_load(struct md_rdev *rdev, struct md_rdev *refdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2164) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2165) int r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2166) struct dm_raid_superblock *sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2167) struct dm_raid_superblock *refsb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2168) uint64_t events_sb, events_refsb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2169)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2170) r = read_disk_sb(rdev, rdev->sb_size, false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2171) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2172) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2173)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2174) sb = page_address(rdev->sb_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2175)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2176) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2177) * Two cases that we want to write new superblocks and rebuild:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2178) * 1) New device (no matching magic number)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2179) * 2) Device specified for rebuild (!In_sync w/ offset == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2180) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2181) if ((sb->magic != cpu_to_le32(DM_RAID_MAGIC)) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2182) (!test_bit(In_sync, &rdev->flags) && !rdev->recovery_offset)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2183) super_sync(rdev->mddev, rdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2184)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2185) set_bit(FirstUse, &rdev->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2186) sb->compat_features = cpu_to_le32(FEATURE_FLAG_SUPPORTS_V190);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2187)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2188) /* Force writing of superblocks to disk */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2189) set_bit(MD_SB_CHANGE_DEVS, &rdev->mddev->sb_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2190)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2191) /* Any superblock is better than none, choose that if given */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2192) return refdev ? 0 : 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2193) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2194)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2195) if (!refdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2196) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2197)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2198) events_sb = le64_to_cpu(sb->events);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2199)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2200) refsb = page_address(refdev->sb_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2201) events_refsb = le64_to_cpu(refsb->events);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2202)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2203) return (events_sb > events_refsb) ? 1 : 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2204) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2205)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2206) static int super_init_validation(struct raid_set *rs, struct md_rdev *rdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2207) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2208) int role;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2209) unsigned int d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2210) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2211) uint64_t events_sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2212) uint64_t failed_devices[DISKS_ARRAY_ELEMS];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2213) struct dm_raid_superblock *sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2214) uint32_t new_devs = 0, rebuild_and_new = 0, rebuilds = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2215) struct md_rdev *r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2216) struct dm_raid_superblock *sb2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2217)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2218) sb = page_address(rdev->sb_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2219) events_sb = le64_to_cpu(sb->events);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2220)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2221) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2222) * Initialise to 1 if this is a new superblock.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2223) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2224) mddev->events = events_sb ? : 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2225)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2226) mddev->reshape_position = MaxSector;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2227)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2228) mddev->raid_disks = le32_to_cpu(sb->num_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2229) mddev->level = le32_to_cpu(sb->level);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2230) mddev->layout = le32_to_cpu(sb->layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2231) mddev->chunk_sectors = le32_to_cpu(sb->stripe_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2232)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2233) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2234) * Reshaping is supported, e.g. reshape_position is valid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2235) * in superblock and superblock content is authoritative.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2236) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2237) if (le32_to_cpu(sb->compat_features) & FEATURE_FLAG_SUPPORTS_V190) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2238) /* Superblock is authoritative wrt given raid set layout! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2239) mddev->new_level = le32_to_cpu(sb->new_level);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2240) mddev->new_layout = le32_to_cpu(sb->new_layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2241) mddev->new_chunk_sectors = le32_to_cpu(sb->new_stripe_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2242) mddev->delta_disks = le32_to_cpu(sb->delta_disks);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2243) mddev->array_sectors = le64_to_cpu(sb->array_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2244)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2245) /* raid was reshaping and got interrupted */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2246) if (le32_to_cpu(sb->flags) & SB_FLAG_RESHAPE_ACTIVE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2247) if (test_bit(__CTR_FLAG_DELTA_DISKS, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2248) DMERR("Reshape requested but raid set is still reshaping");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2249) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2250) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2251)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2252) if (mddev->delta_disks < 0 ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2253) (!mddev->delta_disks && (le32_to_cpu(sb->flags) & SB_FLAG_RESHAPE_BACKWARDS)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2254) mddev->reshape_backwards = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2255) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2256) mddev->reshape_backwards = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2257)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2258) mddev->reshape_position = le64_to_cpu(sb->reshape_position);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2259) rs->raid_type = get_raid_type_by_ll(mddev->level, mddev->layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2260) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2261)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2262) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2263) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2264) * No takeover/reshaping, because we don't have the extended v1.9.0 metadata
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2265) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2266) struct raid_type *rt_cur = get_raid_type_by_ll(mddev->level, mddev->layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2267) struct raid_type *rt_new = get_raid_type_by_ll(mddev->new_level, mddev->new_layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2268)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2269) if (rs_takeover_requested(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2270) if (rt_cur && rt_new)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2271) DMERR("Takeover raid sets from %s to %s not yet supported by metadata. (raid level change)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2272) rt_cur->name, rt_new->name);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2273) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2274) DMERR("Takeover raid sets not yet supported by metadata. (raid level change)");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2275) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2276) } else if (rs_reshape_requested(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2277) DMERR("Reshaping raid sets not yet supported by metadata. (raid layout change keeping level)");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2278) if (mddev->layout != mddev->new_layout) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2279) if (rt_cur && rt_new)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2280) DMERR(" current layout %s vs new layout %s",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2281) rt_cur->name, rt_new->name);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2282) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2283) DMERR(" current layout 0x%X vs new layout 0x%X",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2284) le32_to_cpu(sb->layout), mddev->new_layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2285) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2286) if (mddev->chunk_sectors != mddev->new_chunk_sectors)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2287) DMERR(" current stripe sectors %u vs new stripe sectors %u",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2288) mddev->chunk_sectors, mddev->new_chunk_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2289) if (rs->delta_disks)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2290) DMERR(" current %u disks vs new %u disks",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2291) mddev->raid_disks, mddev->raid_disks + rs->delta_disks);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2292) if (rs_is_raid10(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2293) DMERR(" Old layout: %s w/ %u copies",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2294) raid10_md_layout_to_format(mddev->layout),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2295) raid10_md_layout_to_copies(mddev->layout));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2296) DMERR(" New layout: %s w/ %u copies",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2297) raid10_md_layout_to_format(mddev->new_layout),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2298) raid10_md_layout_to_copies(mddev->new_layout));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2299) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2300) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2301) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2302)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2303) DMINFO("Discovered old metadata format; upgrading to extended metadata format");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2304) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2305)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2306) if (!test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2307) mddev->recovery_cp = le64_to_cpu(sb->array_resync_offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2308)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2309) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2310) * During load, we set FirstUse if a new superblock was written.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2311) * There are two reasons we might not have a superblock:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2312) * 1) The raid set is brand new - in which case, all of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2313) * devices must have their In_sync bit set. Also,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2314) * recovery_cp must be 0, unless forced.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2315) * 2) This is a new device being added to an old raid set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2316) * and the new device needs to be rebuilt - in which
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2317) * case the In_sync bit will /not/ be set and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2318) * recovery_cp must be MaxSector.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2319) * 3) This is/are a new device(s) being added to an old
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2320) * raid set during takeover to a higher raid level
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2321) * to provide capacity for redundancy or during reshape
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2322) * to add capacity to grow the raid set.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2323) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2324) d = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2325) rdev_for_each(r, mddev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2326) if (test_bit(Journal, &rdev->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2327) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2328)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2329) if (test_bit(FirstUse, &r->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2330) new_devs++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2331)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2332) if (!test_bit(In_sync, &r->flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2333) DMINFO("Device %d specified for rebuild; clearing superblock",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2334) r->raid_disk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2335) rebuilds++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2336)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2337) if (test_bit(FirstUse, &r->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2338) rebuild_and_new++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2339) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2340)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2341) d++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2342) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2343)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2344) if (new_devs == rs->raid_disks || !rebuilds) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2345) /* Replace a broken device */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2346) if (new_devs == rs->raid_disks) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2347) DMINFO("Superblocks created for new raid set");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2348) set_bit(MD_ARRAY_FIRST_USE, &mddev->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2349) } else if (new_devs != rebuilds &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2350) new_devs != rs->delta_disks) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2351) DMERR("New device injected into existing raid set without "
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2352) "'delta_disks' or 'rebuild' parameter specified");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2353) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2354) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2355) } else if (new_devs && new_devs != rebuilds) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2356) DMERR("%u 'rebuild' devices cannot be injected into"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2357) " a raid set with %u other first-time devices",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2358) rebuilds, new_devs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2359) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2360) } else if (rebuilds) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2361) if (rebuild_and_new && rebuilds != rebuild_and_new) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2362) DMERR("new device%s provided without 'rebuild'",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2363) new_devs > 1 ? "s" : "");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2364) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2365) } else if (!test_bit(__CTR_FLAG_REBUILD, &rs->ctr_flags) && rs_is_recovering(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2366) DMERR("'rebuild' specified while raid set is not in-sync (recovery_cp=%llu)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2367) (unsigned long long) mddev->recovery_cp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2368) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2369) } else if (rs_is_reshaping(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2370) DMERR("'rebuild' specified while raid set is being reshaped (reshape_position=%llu)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2371) (unsigned long long) mddev->reshape_position);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2372) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2373) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2374) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2375)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2376) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2377) * Now we set the Faulty bit for those devices that are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2378) * recorded in the superblock as failed.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2379) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2380) sb_retrieve_failed_devices(sb, failed_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2381) rdev_for_each(r, mddev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2382) if (test_bit(Journal, &rdev->flags) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2383) !r->sb_page)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2384) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2385) sb2 = page_address(r->sb_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2386) sb2->failed_devices = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2387) memset(sb2->extended_failed_devices, 0, sizeof(sb2->extended_failed_devices));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2388)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2389) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2390) * Check for any device re-ordering.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2391) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2392) if (!test_bit(FirstUse, &r->flags) && (r->raid_disk >= 0)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2393) role = le32_to_cpu(sb2->array_position);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2394) if (role < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2395) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2396)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2397) if (role != r->raid_disk) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2398) if (rs_is_raid10(rs) && __is_raid10_near(mddev->layout)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2399) if (mddev->raid_disks % __raid10_near_copies(mddev->layout) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2400) rs->raid_disks % rs->raid10_copies) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2401) rs->ti->error =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2402) "Cannot change raid10 near set to odd # of devices!";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2403) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2404) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2405)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2406) sb2->array_position = cpu_to_le32(r->raid_disk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2407)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2408) } else if (!(rs_is_raid10(rs) && rt_is_raid0(rs->raid_type)) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2409) !(rs_is_raid0(rs) && rt_is_raid10(rs->raid_type)) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2410) !rt_is_raid1(rs->raid_type)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2411) rs->ti->error = "Cannot change device positions in raid set";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2412) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2413) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2414)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2415) DMINFO("raid device #%d now at position #%d", role, r->raid_disk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2416) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2417)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2418) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2419) * Partial recovery is performed on
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2420) * returning failed devices.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2421) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2422) if (test_bit(role, (void *) failed_devices))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2423) set_bit(Faulty, &r->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2424) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2425) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2426)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2427) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2428) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2429)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2430) static int super_validate(struct raid_set *rs, struct md_rdev *rdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2431) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2432) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2433) struct dm_raid_superblock *sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2434)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2435) if (rs_is_raid0(rs) || !rdev->sb_page || rdev->raid_disk < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2436) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2437)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2438) sb = page_address(rdev->sb_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2439)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2440) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2441) * If mddev->events is not set, we know we have not yet initialized
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2442) * the array.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2443) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2444) if (!mddev->events && super_init_validation(rs, rdev))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2445) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2446)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2447) if (le32_to_cpu(sb->compat_features) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2448) le32_to_cpu(sb->compat_features) != FEATURE_FLAG_SUPPORTS_V190) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2449) rs->ti->error = "Unable to assemble array: Unknown flag(s) in compatible feature flags";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2450) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2451) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2452)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2453) if (sb->incompat_features) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2454) rs->ti->error = "Unable to assemble array: No incompatible feature flags supported yet";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2455) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2456) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2457)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2458) /* Enable bitmap creation on @rs unless no metadevs or raid0 or journaled raid4/5/6 set. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2459) mddev->bitmap_info.offset = (rt_is_raid0(rs->raid_type) || rs->journal_dev.dev) ? 0 : to_sector(4096);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2460) mddev->bitmap_info.default_offset = mddev->bitmap_info.offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2461)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2462) if (!test_and_clear_bit(FirstUse, &rdev->flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2463) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2464) * Retrieve rdev size stored in superblock to be prepared for shrink.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2465) * Check extended superblock members are present otherwise the size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2466) * will not be set!
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2467) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2468) if (le32_to_cpu(sb->compat_features) & FEATURE_FLAG_SUPPORTS_V190)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2469) rdev->sectors = le64_to_cpu(sb->sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2470)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2471) rdev->recovery_offset = le64_to_cpu(sb->disk_recovery_offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2472) if (rdev->recovery_offset == MaxSector)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2473) set_bit(In_sync, &rdev->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2474) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2475) * If no reshape in progress -> we're recovering single
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2476) * disk(s) and have to set the device(s) to out-of-sync
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2477) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2478) else if (!rs_is_reshaping(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2479) clear_bit(In_sync, &rdev->flags); /* Mandatory for recovery */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2480) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2481)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2482) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2483) * If a device comes back, set it as not In_sync and no longer faulty.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2484) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2485) if (test_and_clear_bit(Faulty, &rdev->flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2486) rdev->recovery_offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2487) clear_bit(In_sync, &rdev->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2488) rdev->saved_raid_disk = rdev->raid_disk;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2489) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2490)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2491) /* Reshape support -> restore repective data offsets */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2492) rdev->data_offset = le64_to_cpu(sb->data_offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2493) rdev->new_data_offset = le64_to_cpu(sb->new_data_offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2494)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2495) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2496) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2497)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2498) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2499) * Analyse superblocks and select the freshest.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2500) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2501) static int analyse_superblocks(struct dm_target *ti, struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2502) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2503) int r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2504) struct md_rdev *rdev, *freshest;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2505) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2506)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2507) freshest = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2508) rdev_for_each(rdev, mddev) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2509) if (test_bit(Journal, &rdev->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2510) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2511)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2512) if (!rdev->meta_bdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2513) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2514)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2515) /* Set superblock offset/size for metadata device. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2516) rdev->sb_start = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2517) rdev->sb_size = bdev_logical_block_size(rdev->meta_bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2518) if (rdev->sb_size < sizeof(struct dm_raid_superblock) || rdev->sb_size > PAGE_SIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2519) DMERR("superblock size of a logical block is no longer valid");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2520) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2521) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2522)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2523) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2524) * Skipping super_load due to CTR_FLAG_SYNC will cause
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2525) * the array to undergo initialization again as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2526) * though it were new. This is the intended effect
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2527) * of the "sync" directive.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2528) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2529) * With reshaping capability added, we must ensure that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2530) * that the "sync" directive is disallowed during the reshape.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2531) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2532) if (test_bit(__CTR_FLAG_SYNC, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2533) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2534)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2535) r = super_load(rdev, freshest);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2536)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2537) switch (r) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2538) case 1:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2539) freshest = rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2540) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2541) case 0:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2542) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2543) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2544) /* This is a failure to read the superblock from the metadata device. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2545) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2546) * We have to keep any raid0 data/metadata device pairs or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2547) * the MD raid0 personality will fail to start the array.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2548) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2549) if (rs_is_raid0(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2550) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2551)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2552) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2553) * We keep the dm_devs to be able to emit the device tuple
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2554) * properly on the table line in raid_status() (rather than
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2555) * mistakenly acting as if '- -' got passed into the constructor).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2556) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2557) * The rdev has to stay on the same_set list to allow for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2558) * the attempt to restore faulty devices on second resume.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2559) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2560) rdev->raid_disk = rdev->saved_raid_disk = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2561) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2562) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2563) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2564)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2565) if (!freshest)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2566) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2567)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2568) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2569) * Validation of the freshest device provides the source of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2570) * validation for the remaining devices.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2571) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2572) rs->ti->error = "Unable to assemble array: Invalid superblocks";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2573) if (super_validate(rs, freshest))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2574) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2575)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2576) if (validate_raid_redundancy(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2577) rs->ti->error = "Insufficient redundancy to activate array";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2578) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2579) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2580)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2581) rdev_for_each(rdev, mddev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2582) if (!test_bit(Journal, &rdev->flags) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2583) rdev != freshest &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2584) super_validate(rs, rdev))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2585) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2586) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2587) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2588)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2589) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2590) * Adjust data_offset and new_data_offset on all disk members of @rs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2591) * for out of place reshaping if requested by contructor
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2592) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2593) * We need free space at the beginning of each raid disk for forward
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2594) * and at the end for backward reshapes which userspace has to provide
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2595) * via remapping/reordering of space.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2596) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2597) static int rs_adjust_data_offsets(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2598) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2599) sector_t data_offset = 0, new_data_offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2600) struct md_rdev *rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2601)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2602) /* Constructor did not request data offset change */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2603) if (!test_bit(__CTR_FLAG_DATA_OFFSET, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2604) if (!rs_is_reshapable(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2605) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2606)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2607) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2608) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2609)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2610) /* HM FIXME: get In_Sync raid_dev? */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2611) rdev = &rs->dev[0].rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2612)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2613) if (rs->delta_disks < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2614) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2615) * Removing disks (reshaping backwards):
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2616) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2617) * - before reshape: data is at offset 0 and free space
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2618) * is at end of each component LV
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2619) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2620) * - after reshape: data is at offset rs->data_offset != 0 on each component LV
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2621) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2622) data_offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2623) new_data_offset = rs->data_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2624)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2625) } else if (rs->delta_disks > 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2626) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2627) * Adding disks (reshaping forwards):
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2628) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2629) * - before reshape: data is at offset rs->data_offset != 0 and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2630) * free space is at begin of each component LV
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2631) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2632) * - after reshape: data is at offset 0 on each component LV
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2633) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2634) data_offset = rs->data_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2635) new_data_offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2636)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2637) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2638) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2639) * User space passes in 0 for data offset after having removed reshape space
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2640) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2641) * - or - (data offset != 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2642) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2643) * Changing RAID layout or chunk size -> toggle offsets
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2644) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2645) * - before reshape: data is at offset rs->data_offset 0 and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2646) * free space is at end of each component LV
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2647) * -or-
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2648) * data is at offset rs->data_offset != 0 and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2649) * free space is at begin of each component LV
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2650) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2651) * - after reshape: data is at offset 0 if it was at offset != 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2652) * or at offset != 0 if it was at offset 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2653) * on each component LV
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2654) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2655) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2656) data_offset = rs->data_offset ? rdev->data_offset : 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2657) new_data_offset = data_offset ? 0 : rs->data_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2658) set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2659) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2660)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2661) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2662) * Make sure we got a minimum amount of free sectors per device
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2663) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2664) if (rs->data_offset &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2665) to_sector(i_size_read(rdev->bdev->bd_inode)) - rs->md.dev_sectors < MIN_FREE_RESHAPE_SPACE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2666) rs->ti->error = data_offset ? "No space for forward reshape" :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2667) "No space for backward reshape";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2668) return -ENOSPC;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2669) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2670) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2671) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2672) * Raise recovery_cp in case data_offset != 0 to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2673) * avoid false recovery positives in the constructor.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2674) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2675) if (rs->md.recovery_cp < rs->md.dev_sectors)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2676) rs->md.recovery_cp += rs->dev[0].rdev.data_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2677)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2678) /* Adjust data offsets on all rdevs but on any raid4/5/6 journal device */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2679) rdev_for_each(rdev, &rs->md) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2680) if (!test_bit(Journal, &rdev->flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2681) rdev->data_offset = data_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2682) rdev->new_data_offset = new_data_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2683) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2684) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2685)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2686) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2687) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2688)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2689) /* Userpace reordered disks -> adjust raid_disk indexes in @rs */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2690) static void __reorder_raid_disk_indexes(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2691) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2692) int i = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2693) struct md_rdev *rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2694)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2695) rdev_for_each(rdev, &rs->md) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2696) if (!test_bit(Journal, &rdev->flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2697) rdev->raid_disk = i++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2698) rdev->saved_raid_disk = rdev->new_raid_disk = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2699) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2700) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2701) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2702)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2703) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2704) * Setup @rs for takeover by a different raid level
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2705) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2706) static int rs_setup_takeover(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2707) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2708) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2709) struct md_rdev *rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2710) unsigned int d = mddev->raid_disks = rs->raid_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2711) sector_t new_data_offset = rs->dev[0].rdev.data_offset ? 0 : rs->data_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2712)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2713) if (rt_is_raid10(rs->raid_type)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2714) if (rs_is_raid0(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2715) /* Userpace reordered disks -> adjust raid_disk indexes */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2716) __reorder_raid_disk_indexes(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2717)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2718) /* raid0 -> raid10_far layout */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2719) mddev->layout = raid10_format_to_md_layout(rs, ALGORITHM_RAID10_FAR,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2720) rs->raid10_copies);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2721) } else if (rs_is_raid1(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2722) /* raid1 -> raid10_near layout */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2723) mddev->layout = raid10_format_to_md_layout(rs, ALGORITHM_RAID10_NEAR,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2724) rs->raid_disks);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2725) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2726) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2727)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2728) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2729)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2730) clear_bit(MD_ARRAY_FIRST_USE, &mddev->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2731) mddev->recovery_cp = MaxSector;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2732)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2733) while (d--) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2734) rdev = &rs->dev[d].rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2735)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2736) if (test_bit(d, (void *) rs->rebuild_disks)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2737) clear_bit(In_sync, &rdev->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2738) clear_bit(Faulty, &rdev->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2739) mddev->recovery_cp = rdev->recovery_offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2740) /* Bitmap has to be created when we do an "up" takeover */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2741) set_bit(MD_ARRAY_FIRST_USE, &mddev->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2742) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2743)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2744) rdev->new_data_offset = new_data_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2745) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2746)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2747) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2748) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2749)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2750) /* Prepare @rs for reshape */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2751) static int rs_prepare_reshape(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2752) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2753) bool reshape;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2754) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2755)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2756) if (rs_is_raid10(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2757) if (rs->raid_disks != mddev->raid_disks &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2758) __is_raid10_near(mddev->layout) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2759) rs->raid10_copies &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2760) rs->raid10_copies != __raid10_near_copies(mddev->layout)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2761) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2762) * raid disk have to be multiple of data copies to allow this conversion,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2763) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2764) * This is actually not a reshape it is a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2765) * rebuild of any additional mirrors per group
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2766) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2767) if (rs->raid_disks % rs->raid10_copies) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2768) rs->ti->error = "Can't reshape raid10 mirror groups";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2769) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2770) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2771)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2772) /* Userpace reordered disks to add/remove mirrors -> adjust raid_disk indexes */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2773) __reorder_raid_disk_indexes(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2774) mddev->layout = raid10_format_to_md_layout(rs, ALGORITHM_RAID10_NEAR,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2775) rs->raid10_copies);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2776) mddev->new_layout = mddev->layout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2777) reshape = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2778) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2779) reshape = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2780)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2781) } else if (rs_is_raid456(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2782) reshape = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2783)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2784) else if (rs_is_raid1(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2785) if (rs->delta_disks) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2786) /* Process raid1 via delta_disks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2787) mddev->degraded = rs->delta_disks < 0 ? -rs->delta_disks : rs->delta_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2788) reshape = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2789) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2790) /* Process raid1 without delta_disks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2791) mddev->raid_disks = rs->raid_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2792) reshape = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2793) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2794) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2795) rs->ti->error = "Called with bogus raid type";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2796) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2797) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2798)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2799) if (reshape) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2800) set_bit(RT_FLAG_RESHAPE_RS, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2801) set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2802) } else if (mddev->raid_disks < rs->raid_disks)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2803) /* Create new superblocks and bitmaps, if any new disks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2804) set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2805)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2806) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2807) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2808)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2809) /* Get reshape sectors from data_offsets or raid set */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2810) static sector_t _get_reshape_sectors(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2811) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2812) struct md_rdev *rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2813) sector_t reshape_sectors = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2814)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2815) rdev_for_each(rdev, &rs->md)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2816) if (!test_bit(Journal, &rdev->flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2817) reshape_sectors = (rdev->data_offset > rdev->new_data_offset) ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2818) rdev->data_offset - rdev->new_data_offset :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2819) rdev->new_data_offset - rdev->data_offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2820) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2821) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2822)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2823) return max(reshape_sectors, (sector_t) rs->data_offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2824) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2825)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2826) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2827) * Reshape:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2828) * - change raid layout
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2829) * - change chunk size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2830) * - add disks
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2831) * - remove disks
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2832) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2833) static int rs_setup_reshape(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2834) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2835) int r = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2836) unsigned int cur_raid_devs, d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2837) sector_t reshape_sectors = _get_reshape_sectors(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2838) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2839) struct md_rdev *rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2840)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2841) mddev->delta_disks = rs->delta_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2842) cur_raid_devs = mddev->raid_disks;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2843)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2844) /* Ignore impossible layout change whilst adding/removing disks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2845) if (mddev->delta_disks &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2846) mddev->layout != mddev->new_layout) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2847) DMINFO("Ignoring invalid layout change with delta_disks=%d", rs->delta_disks);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2848) mddev->new_layout = mddev->layout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2849) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2850)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2851) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2852) * Adjust array size:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2853) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2854) * - in case of adding disk(s), array size has
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2855) * to grow after the disk adding reshape,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2856) * which'll hapen in the event handler;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2857) * reshape will happen forward, so space has to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2858) * be available at the beginning of each disk
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2859) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2860) * - in case of removing disk(s), array size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2861) * has to shrink before starting the reshape,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2862) * which'll happen here;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2863) * reshape will happen backward, so space has to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2864) * be available at the end of each disk
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2865) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2866) * - data_offset and new_data_offset are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2867) * adjusted for aforementioned out of place
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2868) * reshaping based on userspace passing in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2869) * the "data_offset <sectors>" key/value
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2870) * pair via the constructor
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2871) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2872)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2873) /* Add disk(s) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2874) if (rs->delta_disks > 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2875) /* Prepare disks for check in raid4/5/6/10 {check|start}_reshape */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2876) for (d = cur_raid_devs; d < rs->raid_disks; d++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2877) rdev = &rs->dev[d].rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2878) clear_bit(In_sync, &rdev->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2879)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2880) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2881) * save_raid_disk needs to be -1, or recovery_offset will be set to 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2882) * by md, which'll store that erroneously in the superblock on reshape
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2883) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2884) rdev->saved_raid_disk = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2885) rdev->raid_disk = d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2886)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2887) rdev->sectors = mddev->dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2888) rdev->recovery_offset = rs_is_raid1(rs) ? 0 : MaxSector;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2889) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2890)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2891) mddev->reshape_backwards = 0; /* adding disk(s) -> forward reshape */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2892)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2893) /* Remove disk(s) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2894) } else if (rs->delta_disks < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2895) r = rs_set_dev_and_array_sectors(rs, rs->ti->len, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2896) mddev->reshape_backwards = 1; /* removing disk(s) -> backward reshape */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2897)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2898) /* Change layout and/or chunk size */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2899) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2900) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2901) * Reshape layout (e.g. raid5_ls -> raid5_n) and/or chunk size:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2902) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2903) * keeping number of disks and do layout change ->
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2904) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2905) * toggle reshape_backward depending on data_offset:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2906) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2907) * - free space upfront -> reshape forward
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2908) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2909) * - free space at the end -> reshape backward
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2910) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2911) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2912) * This utilizes free reshape space avoiding the need
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2913) * for userspace to move (parts of) LV segments in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2914) * case of layout/chunksize change (for disk
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2915) * adding/removing reshape space has to be at
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2916) * the proper address (see above with delta_disks):
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2917) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2918) * add disk(s) -> begin
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2919) * remove disk(s)-> end
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2920) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2921) mddev->reshape_backwards = rs->dev[0].rdev.data_offset ? 0 : 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2922) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2923)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2924) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2925) * Adjust device size for forward reshape
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2926) * because md_finish_reshape() reduces it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2927) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2928) if (!mddev->reshape_backwards)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2929) rdev_for_each(rdev, &rs->md)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2930) if (!test_bit(Journal, &rdev->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2931) rdev->sectors += reshape_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2932)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2933) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2934) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2935)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2936) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2937) * If the md resync thread has updated superblock with max reshape position
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2938) * at the end of a reshape but not (yet) reset the layout configuration
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2939) * changes -> reset the latter.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2940) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2941) static void rs_reset_inconclusive_reshape(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2942) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2943) if (!rs_is_reshaping(rs) && rs_is_layout_change(rs, true)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2944) rs_set_cur(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2945) rs->md.delta_disks = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2946) rs->md.reshape_backwards = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2947) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2948) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2949)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2950) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2951) * Enable/disable discard support on RAID set depending on
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2952) * RAID level and discard properties of underlying RAID members.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2953) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2954) static void configure_discard_support(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2955) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2956) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2957) bool raid456;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2958) struct dm_target *ti = rs->ti;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2959)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2960) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2961) * XXX: RAID level 4,5,6 require zeroing for safety.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2962) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2963) raid456 = rs_is_raid456(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2964)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2965) for (i = 0; i < rs->raid_disks; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2966) struct request_queue *q;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2967)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2968) if (!rs->dev[i].rdev.bdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2969) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2970)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2971) q = bdev_get_queue(rs->dev[i].rdev.bdev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2972) if (!q || !blk_queue_discard(q))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2973) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2974)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2975) if (raid456) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2976) if (!devices_handle_discard_safely) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2977) DMERR("raid456 discard support disabled due to discard_zeroes_data uncertainty.");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2978) DMERR("Set dm-raid.devices_handle_discard_safely=Y to override.");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2979) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2980) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2981) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2982) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2983)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2984) ti->num_discard_bios = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2985) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2986)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2987) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2988) * Construct a RAID0/1/10/4/5/6 mapping:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2989) * Args:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2990) * <raid_type> <#raid_params> <raid_params>{0,} \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2991) * <#raid_devs> [<meta_dev1> <dev1>]{1,}
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2992) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2993) * <raid_params> varies by <raid_type>. See 'parse_raid_params' for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2994) * details on possible <raid_params>.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2995) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2996) * Userspace is free to initialize the metadata devices, hence the superblocks to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2997) * enforce recreation based on the passed in table parameters.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2998) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2999) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3000) static int raid_ctr(struct dm_target *ti, unsigned int argc, char **argv)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3001) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3002) int r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3003) bool resize = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3004) struct raid_type *rt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3005) unsigned int num_raid_params, num_raid_devs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3006) sector_t sb_array_sectors, rdev_sectors, reshape_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3007) struct raid_set *rs = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3008) const char *arg;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3009) struct rs_layout rs_layout;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3010) struct dm_arg_set as = { argc, argv }, as_nrd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3011) struct dm_arg _args[] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3012) { 0, as.argc, "Cannot understand number of raid parameters" },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3013) { 1, 254, "Cannot understand number of raid devices parameters" }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3014) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3015)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3016) arg = dm_shift_arg(&as);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3017) if (!arg) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3018) ti->error = "No arguments";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3019) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3020) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3021)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3022) rt = get_raid_type(arg);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3023) if (!rt) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3024) ti->error = "Unrecognised raid_type";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3025) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3026) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3027)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3028) /* Must have <#raid_params> */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3029) if (dm_read_arg_group(_args, &as, &num_raid_params, &ti->error))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3030) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3031)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3032) /* number of raid device tupples <meta_dev data_dev> */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3033) as_nrd = as;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3034) dm_consume_args(&as_nrd, num_raid_params);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3035) _args[1].max = (as_nrd.argc - 1) / 2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3036) if (dm_read_arg(_args + 1, &as_nrd, &num_raid_devs, &ti->error))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3037) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3038)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3039) if (!__within_range(num_raid_devs, 1, MAX_RAID_DEVICES)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3040) ti->error = "Invalid number of supplied raid devices";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3041) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3042) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3043)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3044) rs = raid_set_alloc(ti, rt, num_raid_devs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3045) if (IS_ERR(rs))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3046) return PTR_ERR(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3047)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3048) r = parse_raid_params(rs, &as, num_raid_params);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3049) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3050) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3051)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3052) r = parse_dev_params(rs, &as);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3053) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3054) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3055)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3056) rs->md.sync_super = super_sync;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3057)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3058) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3059) * Calculate ctr requested array and device sizes to allow
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3060) * for superblock analysis needing device sizes defined.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3061) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3062) * Any existing superblock will overwrite the array and device sizes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3063) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3064) r = rs_set_dev_and_array_sectors(rs, rs->ti->len, false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3065) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3066) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3067)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3068) /* Memorize just calculated, potentially larger sizes to grow the raid set in preresume */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3069) rs->array_sectors = rs->md.array_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3070) rs->dev_sectors = rs->md.dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3071)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3072) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3073) * Backup any new raid set level, layout, ...
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3074) * requested to be able to compare to superblock
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3075) * members for conversion decisions.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3076) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3077) rs_config_backup(rs, &rs_layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3078)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3079) r = analyse_superblocks(ti, rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3080) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3081) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3082)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3083) /* All in-core metadata now as of current superblocks after calling analyse_superblocks() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3084) sb_array_sectors = rs->md.array_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3085) rdev_sectors = __rdev_sectors(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3086) if (!rdev_sectors) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3087) ti->error = "Invalid rdev size";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3088) r = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3089) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3090) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3091)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3092)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3093) reshape_sectors = _get_reshape_sectors(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3094) if (rs->dev_sectors != rdev_sectors) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3095) resize = (rs->dev_sectors != rdev_sectors - reshape_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3096) if (rs->dev_sectors > rdev_sectors - reshape_sectors)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3097) set_bit(RT_FLAG_RS_GROW, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3098) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3099)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3100) INIT_WORK(&rs->md.event_work, do_table_event);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3101) ti->private = rs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3102) ti->num_flush_bios = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3103)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3104) /* Restore any requested new layout for conversion decision */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3105) rs_config_restore(rs, &rs_layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3106)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3107) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3108) * Now that we have any superblock metadata available,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3109) * check for new, recovering, reshaping, to be taken over,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3110) * to be reshaped or an existing, unchanged raid set to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3111) * run in sequence.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3112) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3113) if (test_bit(MD_ARRAY_FIRST_USE, &rs->md.flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3114) /* A new raid6 set has to be recovered to ensure proper parity and Q-Syndrome */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3115) if (rs_is_raid6(rs) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3116) test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3117) ti->error = "'nosync' not allowed for new raid6 set";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3118) r = -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3119) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3120) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3121) rs_setup_recovery(rs, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3122) set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3123) rs_set_new(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3124) } else if (rs_is_recovering(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3125) /* A recovering raid set may be resized */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3126) goto size_check;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3127) } else if (rs_is_reshaping(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3128) /* Have to reject size change request during reshape */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3129) if (resize) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3130) ti->error = "Can't resize a reshaping raid set";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3131) r = -EPERM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3132) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3133) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3134) /* skip setup rs */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3135) } else if (rs_takeover_requested(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3136) if (rs_is_reshaping(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3137) ti->error = "Can't takeover a reshaping raid set";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3138) r = -EPERM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3139) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3140) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3141)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3142) /* We can't takeover a journaled raid4/5/6 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3143) if (test_bit(__CTR_FLAG_JOURNAL_DEV, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3144) ti->error = "Can't takeover a journaled raid4/5/6 set";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3145) r = -EPERM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3146) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3147) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3148)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3149) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3150) * If a takeover is needed, userspace sets any additional
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3151) * devices to rebuild and we can check for a valid request here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3152) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3153) * If acceptible, set the level to the new requested
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3154) * one, prohibit requesting recovery, allow the raid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3155) * set to run and store superblocks during resume.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3156) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3157) r = rs_check_takeover(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3158) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3159) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3160)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3161) r = rs_setup_takeover(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3162) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3163) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3164)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3165) set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3166) /* Takeover ain't recovery, so disable recovery */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3167) rs_setup_recovery(rs, MaxSector);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3168) rs_set_new(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3169) } else if (rs_reshape_requested(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3170) /* Only request grow on raid set size extensions, not on reshapes. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3171) clear_bit(RT_FLAG_RS_GROW, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3172)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3173) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3174) * No need to check for 'ongoing' takeover here, because takeover
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3175) * is an instant operation as oposed to an ongoing reshape.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3176) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3177)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3178) /* We can't reshape a journaled raid4/5/6 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3179) if (test_bit(__CTR_FLAG_JOURNAL_DEV, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3180) ti->error = "Can't reshape a journaled raid4/5/6 set";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3181) r = -EPERM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3182) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3183) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3184)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3185) /* Out-of-place space has to be available to allow for a reshape unless raid1! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3186) if (reshape_sectors || rs_is_raid1(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3187) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3188) * We can only prepare for a reshape here, because the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3189) * raid set needs to run to provide the repective reshape
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3190) * check functions via its MD personality instance.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3191) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3192) * So do the reshape check after md_run() succeeded.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3193) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3194) r = rs_prepare_reshape(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3195) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3196) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3197)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3198) /* Reshaping ain't recovery, so disable recovery */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3199) rs_setup_recovery(rs, MaxSector);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3200) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3201) rs_set_cur(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3202) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3203) size_check:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3204) /* May not set recovery when a device rebuild is requested */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3205) if (test_bit(__CTR_FLAG_REBUILD, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3206) clear_bit(RT_FLAG_RS_GROW, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3207) set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3208) rs_setup_recovery(rs, MaxSector);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3209) } else if (test_bit(RT_FLAG_RS_GROW, &rs->runtime_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3210) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3211) * Set raid set to current size, i.e. size as of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3212) * superblocks to grow to larger size in preresume.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3213) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3214) r = rs_set_dev_and_array_sectors(rs, sb_array_sectors, false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3215) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3216) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3217)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3218) rs_setup_recovery(rs, rs->md.recovery_cp < rs->md.dev_sectors ? rs->md.recovery_cp : rs->md.dev_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3219) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3220) /* This is no size change or it is shrinking, update size and record in superblocks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3221) r = rs_set_dev_and_array_sectors(rs, rs->ti->len, false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3222) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3223) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3224)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3225) if (sb_array_sectors > rs->array_sectors)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3226) set_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3227) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3228) rs_set_cur(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3229) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3230)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3231) /* If constructor requested it, change data and new_data offsets */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3232) r = rs_adjust_data_offsets(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3233) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3234) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3235)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3236) /* Catch any inconclusive reshape superblock content. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3237) rs_reset_inconclusive_reshape(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3238)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3239) /* Start raid set read-only and assumed clean to change in raid_resume() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3240) rs->md.ro = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3241) rs->md.in_sync = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3242)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3243) /* Keep array frozen until resume. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3244) set_bit(MD_RECOVERY_FROZEN, &rs->md.recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3245)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3246) /* Has to be held on running the array */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3247) mddev_lock_nointr(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3248) r = md_run(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3249) rs->md.in_sync = 0; /* Assume already marked dirty */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3250) if (r) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3251) ti->error = "Failed to run raid array";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3252) mddev_unlock(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3253) goto bad;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3254) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3255)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3256) r = md_start(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3257) if (r) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3258) ti->error = "Failed to start raid array";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3259) mddev_unlock(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3260) goto bad_md_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3261) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3262)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3263) /* If raid4/5/6 journal mode explicitly requested (only possible with journal dev) -> set it */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3264) if (test_bit(__CTR_FLAG_JOURNAL_MODE, &rs->ctr_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3265) r = r5c_journal_mode_set(&rs->md, rs->journal_dev.mode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3266) if (r) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3267) ti->error = "Failed to set raid4/5/6 journal mode";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3268) mddev_unlock(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3269) goto bad_journal_mode_set;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3270) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3271) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3272)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3273) mddev_suspend(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3274) set_bit(RT_FLAG_RS_SUSPENDED, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3275)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3276) /* Try to adjust the raid4/5/6 stripe cache size to the stripe size */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3277) if (rs_is_raid456(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3278) r = rs_set_raid456_stripe_cache(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3279) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3280) goto bad_stripe_cache;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3281) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3282)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3283) /* Now do an early reshape check */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3284) if (test_bit(RT_FLAG_RESHAPE_RS, &rs->runtime_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3285) r = rs_check_reshape(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3286) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3287) goto bad_check_reshape;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3288)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3289) /* Restore new, ctr requested layout to perform check */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3290) rs_config_restore(rs, &rs_layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3291)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3292) if (rs->md.pers->start_reshape) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3293) r = rs->md.pers->check_reshape(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3294) if (r) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3295) ti->error = "Reshape check failed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3296) goto bad_check_reshape;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3297) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3298) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3299) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3300)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3301) /* Disable/enable discard support on raid set. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3302) configure_discard_support(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3303)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3304) mddev_unlock(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3305) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3306)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3307) bad_md_start:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3308) bad_journal_mode_set:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3309) bad_stripe_cache:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3310) bad_check_reshape:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3311) md_stop(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3312) bad:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3313) raid_set_free(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3314)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3315) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3316) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3317)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3318) static void raid_dtr(struct dm_target *ti)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3319) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3320) struct raid_set *rs = ti->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3321)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3322) md_stop(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3323) raid_set_free(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3324) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3325)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3326) static int raid_map(struct dm_target *ti, struct bio *bio)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3327) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3328) struct raid_set *rs = ti->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3329) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3330)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3331) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3332) * If we're reshaping to add disk(s)), ti->len and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3333) * mddev->array_sectors will differ during the process
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3334) * (ti->len > mddev->array_sectors), so we have to requeue
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3335) * bios with addresses > mddev->array_sectors here or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3336) * there will occur accesses past EOD of the component
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3337) * data images thus erroring the raid set.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3338) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3339) if (unlikely(bio_end_sector(bio) > mddev->array_sectors))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3340) return DM_MAPIO_REQUEUE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3341)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3342) md_handle_request(mddev, bio);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3343)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3344) return DM_MAPIO_SUBMITTED;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3345) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3346)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3347) /* Return sync state string for @state */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3348) enum sync_state { st_frozen, st_reshape, st_resync, st_check, st_repair, st_recover, st_idle };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3349) static const char *sync_str(enum sync_state state)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3350) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3351) /* Has to be in above sync_state order! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3352) static const char *sync_strs[] = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3353) "frozen",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3354) "reshape",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3355) "resync",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3356) "check",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3357) "repair",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3358) "recover",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3359) "idle"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3360) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3361)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3362) return __within_range(state, 0, ARRAY_SIZE(sync_strs) - 1) ? sync_strs[state] : "undef";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3363) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3364)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3365) /* Return enum sync_state for @mddev derived from @recovery flags */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3366) static enum sync_state decipher_sync_action(struct mddev *mddev, unsigned long recovery)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3367) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3368) if (test_bit(MD_RECOVERY_FROZEN, &recovery))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3369) return st_frozen;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3370)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3371) /* The MD sync thread can be done with io or be interrupted but still be running */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3372) if (!test_bit(MD_RECOVERY_DONE, &recovery) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3373) (test_bit(MD_RECOVERY_RUNNING, &recovery) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3374) (!mddev->ro && test_bit(MD_RECOVERY_NEEDED, &recovery)))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3375) if (test_bit(MD_RECOVERY_RESHAPE, &recovery))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3376) return st_reshape;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3377)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3378) if (test_bit(MD_RECOVERY_SYNC, &recovery)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3379) if (!test_bit(MD_RECOVERY_REQUESTED, &recovery))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3380) return st_resync;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3381) if (test_bit(MD_RECOVERY_CHECK, &recovery))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3382) return st_check;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3383) return st_repair;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3384) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3385)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3386) if (test_bit(MD_RECOVERY_RECOVER, &recovery))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3387) return st_recover;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3388)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3389) if (mddev->reshape_position != MaxSector)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3390) return st_reshape;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3391) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3392)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3393) return st_idle;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3394) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3395)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3396) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3397) * Return status string for @rdev
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3398) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3399) * Status characters:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3400) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3401) * 'D' = Dead/Failed raid set component or raid4/5/6 journal device
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3402) * 'a' = Alive but not in-sync raid set component _or_ alive raid4/5/6 'write_back' journal device
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3403) * 'A' = Alive and in-sync raid set component _or_ alive raid4/5/6 'write_through' journal device
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3404) * '-' = Non-existing device (i.e. uspace passed '- -' into the ctr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3405) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3406) static const char *__raid_dev_status(struct raid_set *rs, struct md_rdev *rdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3407) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3408) if (!rdev->bdev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3409) return "-";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3410) else if (test_bit(Faulty, &rdev->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3411) return "D";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3412) else if (test_bit(Journal, &rdev->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3413) return (rs->journal_dev.mode == R5C_JOURNAL_MODE_WRITE_THROUGH) ? "A" : "a";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3414) else if (test_bit(RT_FLAG_RS_RESYNCING, &rs->runtime_flags) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3415) (!test_bit(RT_FLAG_RS_IN_SYNC, &rs->runtime_flags) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3416) !test_bit(In_sync, &rdev->flags)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3417) return "a";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3418) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3419) return "A";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3420) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3421)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3422) /* Helper to return resync/reshape progress for @rs and runtime flags for raid set in sync / resynching */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3423) static sector_t rs_get_progress(struct raid_set *rs, unsigned long recovery,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3424) enum sync_state state, sector_t resync_max_sectors)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3425) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3426) sector_t r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3427) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3428)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3429) clear_bit(RT_FLAG_RS_IN_SYNC, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3430) clear_bit(RT_FLAG_RS_RESYNCING, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3431)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3432) if (rs_is_raid0(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3433) r = resync_max_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3434) set_bit(RT_FLAG_RS_IN_SYNC, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3435)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3436) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3437) if (state == st_idle && !test_bit(MD_RECOVERY_INTR, &recovery))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3438) r = mddev->recovery_cp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3439) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3440) r = mddev->curr_resync_completed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3441)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3442) if (state == st_idle && r >= resync_max_sectors) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3443) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3444) * Sync complete.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3445) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3446) /* In case we have finished recovering, the array is in sync. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3447) if (test_bit(MD_RECOVERY_RECOVER, &recovery))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3448) set_bit(RT_FLAG_RS_IN_SYNC, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3449)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3450) } else if (state == st_recover)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3451) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3452) * In case we are recovering, the array is not in sync
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3453) * and health chars should show the recovering legs.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3454) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3455) * Already retrieved recovery offset from curr_resync_completed above.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3456) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3457) ;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3458)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3459) else if (state == st_resync || state == st_reshape)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3460) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3461) * If "resync/reshape" is occurring, the raid set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3462) * is or may be out of sync hence the health
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3463) * characters shall be 'a'.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3464) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3465) set_bit(RT_FLAG_RS_RESYNCING, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3466)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3467) else if (state == st_check || state == st_repair)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3468) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3469) * If "check" or "repair" is occurring, the raid set has
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3470) * undergone an initial sync and the health characters
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3471) * should not be 'a' anymore.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3472) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3473) set_bit(RT_FLAG_RS_IN_SYNC, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3474)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3475) else if (test_bit(MD_RECOVERY_NEEDED, &recovery))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3476) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3477) * We are idle and recovery is needed, prevent 'A' chars race
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3478) * caused by components still set to in-sync by constructor.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3479) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3480) set_bit(RT_FLAG_RS_RESYNCING, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3481)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3482) else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3483) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3484) * We are idle and the raid set may be doing an initial
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3485) * sync, or it may be rebuilding individual components.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3486) * If all the devices are In_sync, then it is the raid set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3487) * that is being initialized.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3488) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3489) struct md_rdev *rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3490)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3491) set_bit(RT_FLAG_RS_IN_SYNC, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3492) rdev_for_each(rdev, mddev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3493) if (!test_bit(Journal, &rdev->flags) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3494) !test_bit(In_sync, &rdev->flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3495) clear_bit(RT_FLAG_RS_IN_SYNC, &rs->runtime_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3496) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3497) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3498) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3499) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3500)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3501) return min(r, resync_max_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3502) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3503)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3504) /* Helper to return @dev name or "-" if !@dev */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3505) static const char *__get_dev_name(struct dm_dev *dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3506) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3507) return dev ? dev->name : "-";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3508) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3509)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3510) static void raid_status(struct dm_target *ti, status_type_t type,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3511) unsigned int status_flags, char *result, unsigned int maxlen)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3512) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3513) struct raid_set *rs = ti->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3514) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3515) struct r5conf *conf = mddev->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3516) int i, max_nr_stripes = conf ? conf->max_nr_stripes : 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3517) unsigned long recovery;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3518) unsigned int raid_param_cnt = 1; /* at least 1 for chunksize */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3519) unsigned int sz = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3520) unsigned int rebuild_writemostly_count = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3521) sector_t progress, resync_max_sectors, resync_mismatches;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3522) enum sync_state state;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3523) struct raid_type *rt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3524)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3525) switch (type) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3526) case STATUSTYPE_INFO:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3527) /* *Should* always succeed */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3528) rt = get_raid_type_by_ll(mddev->new_level, mddev->new_layout);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3529) if (!rt)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3530) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3531)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3532) DMEMIT("%s %d ", rt->name, mddev->raid_disks);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3533)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3534) /* Access most recent mddev properties for status output */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3535) smp_rmb();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3536) /* Get sensible max sectors even if raid set not yet started */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3537) resync_max_sectors = test_bit(RT_FLAG_RS_PRERESUMED, &rs->runtime_flags) ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3538) mddev->resync_max_sectors : mddev->dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3539) recovery = rs->md.recovery;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3540) state = decipher_sync_action(mddev, recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3541) progress = rs_get_progress(rs, recovery, state, resync_max_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3542) resync_mismatches = (mddev->last_sync_action && !strcasecmp(mddev->last_sync_action, "check")) ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3543) atomic64_read(&mddev->resync_mismatches) : 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3544)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3545) /* HM FIXME: do we want another state char for raid0? It shows 'D'/'A'/'-' now */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3546) for (i = 0; i < rs->raid_disks; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3547) DMEMIT(__raid_dev_status(rs, &rs->dev[i].rdev));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3548)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3549) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3550) * In-sync/Reshape ratio:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3551) * The in-sync ratio shows the progress of:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3552) * - Initializing the raid set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3553) * - Rebuilding a subset of devices of the raid set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3554) * The user can distinguish between the two by referring
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3555) * to the status characters.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3556) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3557) * The reshape ratio shows the progress of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3558) * changing the raid layout or the number of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3559) * disks of a raid set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3560) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3561) DMEMIT(" %llu/%llu", (unsigned long long) progress,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3562) (unsigned long long) resync_max_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3563)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3564) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3565) * v1.5.0+:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3566) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3567) * Sync action:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3568) * See Documentation/admin-guide/device-mapper/dm-raid.rst for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3569) * information on each of these states.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3570) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3571) DMEMIT(" %s", sync_str(state));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3572)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3573) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3574) * v1.5.0+:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3575) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3576) * resync_mismatches/mismatch_cnt
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3577) * This field shows the number of discrepancies found when
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3578) * performing a "check" of the raid set.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3579) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3580) DMEMIT(" %llu", (unsigned long long) resync_mismatches);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3581)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3582) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3583) * v1.9.0+:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3584) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3585) * data_offset (needed for out of space reshaping)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3586) * This field shows the data offset into the data
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3587) * image LV where the first stripes data starts.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3588) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3589) * We keep data_offset equal on all raid disks of the set,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3590) * so retrieving it from the first raid disk is sufficient.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3591) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3592) DMEMIT(" %llu", (unsigned long long) rs->dev[0].rdev.data_offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3593)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3594) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3595) * v1.10.0+:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3596) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3597) DMEMIT(" %s", test_bit(__CTR_FLAG_JOURNAL_DEV, &rs->ctr_flags) ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3598) __raid_dev_status(rs, &rs->journal_dev.rdev) : "-");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3599) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3600)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3601) case STATUSTYPE_TABLE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3602) /* Report the table line string you would use to construct this raid set */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3603)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3604) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3605) * Count any rebuild or writemostly argument pairs and subtract the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3606) * hweight count being added below of any rebuild and writemostly ctr flags.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3607) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3608) for (i = 0; i < rs->raid_disks; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3609) rebuild_writemostly_count += (test_bit(i, (void *) rs->rebuild_disks) ? 2 : 0) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3610) (test_bit(WriteMostly, &rs->dev[i].rdev.flags) ? 2 : 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3611) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3612) rebuild_writemostly_count -= (test_bit(__CTR_FLAG_REBUILD, &rs->ctr_flags) ? 2 : 0) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3613) (test_bit(__CTR_FLAG_WRITE_MOSTLY, &rs->ctr_flags) ? 2 : 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3614) /* Calculate raid parameter count based on ^ rebuild/writemostly argument counts and ctr flags set. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3615) raid_param_cnt += rebuild_writemostly_count +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3616) hweight32(rs->ctr_flags & CTR_FLAG_OPTIONS_NO_ARGS) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3617) hweight32(rs->ctr_flags & CTR_FLAG_OPTIONS_ONE_ARG) * 2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3618) /* Emit table line */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3619) /* This has to be in the documented order for userspace! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3620) DMEMIT("%s %u %u", rs->raid_type->name, raid_param_cnt, mddev->new_chunk_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3621) if (test_bit(__CTR_FLAG_SYNC, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3622) DMEMIT(" %s", dm_raid_arg_name_by_flag(CTR_FLAG_SYNC));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3623) if (test_bit(__CTR_FLAG_NOSYNC, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3624) DMEMIT(" %s", dm_raid_arg_name_by_flag(CTR_FLAG_NOSYNC));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3625) if (test_bit(__CTR_FLAG_REBUILD, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3626) for (i = 0; i < rs->raid_disks; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3627) if (test_bit(i, (void *) rs->rebuild_disks))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3628) DMEMIT(" %s %u", dm_raid_arg_name_by_flag(CTR_FLAG_REBUILD), i);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3629) if (test_bit(__CTR_FLAG_DAEMON_SLEEP, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3630) DMEMIT(" %s %lu", dm_raid_arg_name_by_flag(CTR_FLAG_DAEMON_SLEEP),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3631) mddev->bitmap_info.daemon_sleep);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3632) if (test_bit(__CTR_FLAG_MIN_RECOVERY_RATE, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3633) DMEMIT(" %s %d", dm_raid_arg_name_by_flag(CTR_FLAG_MIN_RECOVERY_RATE),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3634) mddev->sync_speed_min);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3635) if (test_bit(__CTR_FLAG_MAX_RECOVERY_RATE, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3636) DMEMIT(" %s %d", dm_raid_arg_name_by_flag(CTR_FLAG_MAX_RECOVERY_RATE),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3637) mddev->sync_speed_max);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3638) if (test_bit(__CTR_FLAG_WRITE_MOSTLY, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3639) for (i = 0; i < rs->raid_disks; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3640) if (test_bit(WriteMostly, &rs->dev[i].rdev.flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3641) DMEMIT(" %s %d", dm_raid_arg_name_by_flag(CTR_FLAG_WRITE_MOSTLY),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3642) rs->dev[i].rdev.raid_disk);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3643) if (test_bit(__CTR_FLAG_MAX_WRITE_BEHIND, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3644) DMEMIT(" %s %lu", dm_raid_arg_name_by_flag(CTR_FLAG_MAX_WRITE_BEHIND),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3645) mddev->bitmap_info.max_write_behind);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3646) if (test_bit(__CTR_FLAG_STRIPE_CACHE, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3647) DMEMIT(" %s %d", dm_raid_arg_name_by_flag(CTR_FLAG_STRIPE_CACHE),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3648) max_nr_stripes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3649) if (test_bit(__CTR_FLAG_REGION_SIZE, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3650) DMEMIT(" %s %llu", dm_raid_arg_name_by_flag(CTR_FLAG_REGION_SIZE),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3651) (unsigned long long) to_sector(mddev->bitmap_info.chunksize));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3652) if (test_bit(__CTR_FLAG_RAID10_COPIES, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3653) DMEMIT(" %s %d", dm_raid_arg_name_by_flag(CTR_FLAG_RAID10_COPIES),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3654) raid10_md_layout_to_copies(mddev->layout));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3655) if (test_bit(__CTR_FLAG_RAID10_FORMAT, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3656) DMEMIT(" %s %s", dm_raid_arg_name_by_flag(CTR_FLAG_RAID10_FORMAT),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3657) raid10_md_layout_to_format(mddev->layout));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3658) if (test_bit(__CTR_FLAG_DELTA_DISKS, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3659) DMEMIT(" %s %d", dm_raid_arg_name_by_flag(CTR_FLAG_DELTA_DISKS),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3660) max(rs->delta_disks, mddev->delta_disks));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3661) if (test_bit(__CTR_FLAG_DATA_OFFSET, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3662) DMEMIT(" %s %llu", dm_raid_arg_name_by_flag(CTR_FLAG_DATA_OFFSET),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3663) (unsigned long long) rs->data_offset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3664) if (test_bit(__CTR_FLAG_JOURNAL_DEV, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3665) DMEMIT(" %s %s", dm_raid_arg_name_by_flag(CTR_FLAG_JOURNAL_DEV),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3666) __get_dev_name(rs->journal_dev.dev));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3667) if (test_bit(__CTR_FLAG_JOURNAL_MODE, &rs->ctr_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3668) DMEMIT(" %s %s", dm_raid_arg_name_by_flag(CTR_FLAG_JOURNAL_MODE),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3669) md_journal_mode_to_dm_raid(rs->journal_dev.mode));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3670) DMEMIT(" %d", rs->raid_disks);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3671) for (i = 0; i < rs->raid_disks; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3672) DMEMIT(" %s %s", __get_dev_name(rs->dev[i].meta_dev),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3673) __get_dev_name(rs->dev[i].data_dev));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3674) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3675) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3676)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3677) static int raid_message(struct dm_target *ti, unsigned int argc, char **argv,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3678) char *result, unsigned maxlen)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3679) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3680) struct raid_set *rs = ti->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3681) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3682)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3683) if (!mddev->pers || !mddev->pers->sync_request)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3684) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3685)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3686) if (!strcasecmp(argv[0], "frozen"))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3687) set_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3688) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3689) clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3690)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3691) if (!strcasecmp(argv[0], "idle") || !strcasecmp(argv[0], "frozen")) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3692) if (mddev->sync_thread) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3693) set_bit(MD_RECOVERY_INTR, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3694) md_reap_sync_thread(mddev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3695) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3696) } else if (decipher_sync_action(mddev, mddev->recovery) != st_idle)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3697) return -EBUSY;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3698) else if (!strcasecmp(argv[0], "resync"))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3699) ; /* MD_RECOVERY_NEEDED set below */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3700) else if (!strcasecmp(argv[0], "recover"))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3701) set_bit(MD_RECOVERY_RECOVER, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3702) else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3703) if (!strcasecmp(argv[0], "check")) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3704) set_bit(MD_RECOVERY_CHECK, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3705) set_bit(MD_RECOVERY_REQUESTED, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3706) set_bit(MD_RECOVERY_SYNC, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3707) } else if (!strcasecmp(argv[0], "repair")) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3708) set_bit(MD_RECOVERY_REQUESTED, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3709) set_bit(MD_RECOVERY_SYNC, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3710) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3711) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3712) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3713) if (mddev->ro == 2) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3714) /* A write to sync_action is enough to justify
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3715) * canceling read-auto mode
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3716) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3717) mddev->ro = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3718) if (!mddev->suspended && mddev->sync_thread)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3719) md_wakeup_thread(mddev->sync_thread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3720) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3721) set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3722) if (!mddev->suspended && mddev->thread)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3723) md_wakeup_thread(mddev->thread);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3724)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3725) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3726) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3727)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3728) static int raid_iterate_devices(struct dm_target *ti,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3729) iterate_devices_callout_fn fn, void *data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3730) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3731) struct raid_set *rs = ti->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3732) unsigned int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3733) int r = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3734)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3735) for (i = 0; !r && i < rs->md.raid_disks; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3736) if (rs->dev[i].data_dev)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3737) r = fn(ti,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3738) rs->dev[i].data_dev,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3739) 0, /* No offset on data devs */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3740) rs->md.dev_sectors,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3741) data);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3742)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3743) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3744) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3745)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3746) static void raid_io_hints(struct dm_target *ti, struct queue_limits *limits)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3747) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3748) struct raid_set *rs = ti->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3749) unsigned int chunk_size_bytes = to_bytes(rs->md.chunk_sectors);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3750)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3751) blk_limits_io_min(limits, chunk_size_bytes);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3752) blk_limits_io_opt(limits, chunk_size_bytes * mddev_data_stripes(rs));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3753)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3754) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3755) * RAID0 and RAID10 personalities require bio splitting,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3756) * RAID1/4/5/6 don't and process large discard bios properly.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3757) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3758) if (rs_is_raid0(rs) || rs_is_raid10(rs)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3759) limits->discard_granularity = chunk_size_bytes;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3760) limits->max_discard_sectors = rs->md.chunk_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3761) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3762) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3763)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3764) static void raid_postsuspend(struct dm_target *ti)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3765) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3766) struct raid_set *rs = ti->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3767)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3768) if (!test_and_set_bit(RT_FLAG_RS_SUSPENDED, &rs->runtime_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3769) /* Writes have to be stopped before suspending to avoid deadlocks. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3770) if (!test_bit(MD_RECOVERY_FROZEN, &rs->md.recovery))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3771) md_stop_writes(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3772)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3773) mddev_lock_nointr(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3774) mddev_suspend(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3775) mddev_unlock(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3776) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3777) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3778)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3779) static void attempt_restore_of_faulty_devices(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3780) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3781) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3782) uint64_t cleared_failed_devices[DISKS_ARRAY_ELEMS];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3783) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3784) bool cleared = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3785) struct dm_raid_superblock *sb;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3786) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3787) struct md_rdev *r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3788)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3789) /* RAID personalities have to provide hot add/remove methods or we need to bail out. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3790) if (!mddev->pers || !mddev->pers->hot_add_disk || !mddev->pers->hot_remove_disk)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3791) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3792)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3793) memset(cleared_failed_devices, 0, sizeof(cleared_failed_devices));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3794)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3795) for (i = 0; i < mddev->raid_disks; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3796) r = &rs->dev[i].rdev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3797) /* HM FIXME: enhance journal device recovery processing */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3798) if (test_bit(Journal, &r->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3799) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3800)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3801) if (test_bit(Faulty, &r->flags) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3802) r->meta_bdev && !read_disk_sb(r, r->sb_size, true)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3803) DMINFO("Faulty %s device #%d has readable super block."
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3804) " Attempting to revive it.",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3805) rs->raid_type->name, i);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3806)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3807) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3808) * Faulty bit may be set, but sometimes the array can
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3809) * be suspended before the personalities can respond
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3810) * by removing the device from the array (i.e. calling
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3811) * 'hot_remove_disk'). If they haven't yet removed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3812) * the failed device, its 'raid_disk' number will be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3813) * '>= 0' - meaning we must call this function
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3814) * ourselves.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3815) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3816) flags = r->flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3817) clear_bit(In_sync, &r->flags); /* Mandatory for hot remove. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3818) if (r->raid_disk >= 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3819) if (mddev->pers->hot_remove_disk(mddev, r)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3820) /* Failed to revive this device, try next */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3821) r->flags = flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3822) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3823) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3824) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3825) r->raid_disk = r->saved_raid_disk = i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3826)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3827) clear_bit(Faulty, &r->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3828) clear_bit(WriteErrorSeen, &r->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3829)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3830) if (mddev->pers->hot_add_disk(mddev, r)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3831) /* Failed to revive this device, try next */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3832) r->raid_disk = r->saved_raid_disk = -1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3833) r->flags = flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3834) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3835) clear_bit(In_sync, &r->flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3836) r->recovery_offset = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3837) set_bit(i, (void *) cleared_failed_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3838) cleared = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3839) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3840) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3841) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3842)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3843) /* If any failed devices could be cleared, update all sbs failed_devices bits */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3844) if (cleared) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3845) uint64_t failed_devices[DISKS_ARRAY_ELEMS];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3846)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3847) rdev_for_each(r, &rs->md) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3848) if (test_bit(Journal, &r->flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3849) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3850)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3851) sb = page_address(r->sb_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3852) sb_retrieve_failed_devices(sb, failed_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3853)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3854) for (i = 0; i < DISKS_ARRAY_ELEMS; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3855) failed_devices[i] &= ~cleared_failed_devices[i];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3856)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3857) sb_update_failed_devices(sb, failed_devices);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3858) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3859) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3860) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3861)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3862) static int __load_dirty_region_bitmap(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3863) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3864) int r = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3865)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3866) /* Try loading the bitmap unless "raid0", which does not have one */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3867) if (!rs_is_raid0(rs) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3868) !test_and_set_bit(RT_FLAG_RS_BITMAP_LOADED, &rs->runtime_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3869) r = md_bitmap_load(&rs->md);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3870) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3871) DMERR("Failed to load bitmap");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3872) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3873)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3874) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3875) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3876)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3877) /* Enforce updating all superblocks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3878) static void rs_update_sbs(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3879) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3880) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3881) int ro = mddev->ro;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3882)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3883) set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3884) mddev->ro = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3885) md_update_sb(mddev, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3886) mddev->ro = ro;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3887) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3888)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3889) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3890) * Reshape changes raid algorithm of @rs to new one within personality
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3891) * (e.g. raid6_zr -> raid6_nc), changes stripe size, adds/removes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3892) * disks from a raid set thus growing/shrinking it or resizes the set
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3893) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3894) * Call mddev_lock_nointr() before!
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3895) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3896) static int rs_start_reshape(struct raid_set *rs)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3897) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3898) int r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3899) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3900) struct md_personality *pers = mddev->pers;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3901)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3902) /* Don't allow the sync thread to work until the table gets reloaded. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3903) set_bit(MD_RECOVERY_WAIT, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3904)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3905) r = rs_setup_reshape(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3906) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3907) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3908)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3909) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3910) * Check any reshape constraints enforced by the personalility
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3911) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3912) * May as well already kick the reshape off so that * pers->start_reshape() becomes optional.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3913) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3914) r = pers->check_reshape(mddev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3915) if (r) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3916) rs->ti->error = "pers->check_reshape() failed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3917) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3918) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3919)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3920) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3921) * Personality may not provide start reshape method in which
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3922) * case check_reshape above has already covered everything
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3923) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3924) if (pers->start_reshape) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3925) r = pers->start_reshape(mddev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3926) if (r) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3927) rs->ti->error = "pers->start_reshape() failed";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3928) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3929) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3930) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3931)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3932) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3933) * Now reshape got set up, update superblocks to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3934) * reflect the fact so that a table reload will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3935) * access proper superblock content in the ctr.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3936) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3937) rs_update_sbs(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3938)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3939) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3940) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3941)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3942) static int raid_preresume(struct dm_target *ti)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3943) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3944) int r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3945) struct raid_set *rs = ti->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3946) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3947)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3948) /* This is a resume after a suspend of the set -> it's already started. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3949) if (test_and_set_bit(RT_FLAG_RS_PRERESUMED, &rs->runtime_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3950) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3951)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3952) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3953) * The superblocks need to be updated on disk if the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3954) * array is new or new devices got added (thus zeroed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3955) * out by userspace) or __load_dirty_region_bitmap
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3956) * will overwrite them in core with old data or fail.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3957) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3958) if (test_bit(RT_FLAG_UPDATE_SBS, &rs->runtime_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3959) rs_update_sbs(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3960)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3961) /* Load the bitmap from disk unless raid0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3962) r = __load_dirty_region_bitmap(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3963) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3964) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3965)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3966) /* We are extending the raid set size, adjust mddev/md_rdev sizes and set capacity. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3967) if (test_bit(RT_FLAG_RS_GROW, &rs->runtime_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3968) mddev->array_sectors = rs->array_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3969) mddev->dev_sectors = rs->dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3970) rs_set_rdev_sectors(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3971) rs_set_capacity(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3972) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3973)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3974) /* Resize bitmap to adjust to changed region size (aka MD bitmap chunksize) or grown device size */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3975) if (test_bit(RT_FLAG_RS_BITMAP_LOADED, &rs->runtime_flags) && mddev->bitmap &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3976) (test_bit(RT_FLAG_RS_GROW, &rs->runtime_flags) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3977) (rs->requested_bitmap_chunk_sectors &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3978) mddev->bitmap_info.chunksize != to_bytes(rs->requested_bitmap_chunk_sectors)))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3979) int chunksize = to_bytes(rs->requested_bitmap_chunk_sectors) ?: mddev->bitmap_info.chunksize;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3980)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3981) r = md_bitmap_resize(mddev->bitmap, mddev->dev_sectors, chunksize, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3982) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3983) DMERR("Failed to resize bitmap");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3984) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3985)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3986) /* Check for any resize/reshape on @rs and adjust/initiate */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3987) /* Be prepared for mddev_resume() in raid_resume() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3988) set_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3989) if (mddev->recovery_cp && mddev->recovery_cp < MaxSector) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3990) set_bit(MD_RECOVERY_REQUESTED, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3991) mddev->resync_min = mddev->recovery_cp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3992) if (test_bit(RT_FLAG_RS_GROW, &rs->runtime_flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3993) mddev->resync_max_sectors = mddev->dev_sectors;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3994) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3995)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3996) /* Check for any reshape request unless new raid set */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3997) if (test_bit(RT_FLAG_RESHAPE_RS, &rs->runtime_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3998) /* Initiate a reshape. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3999) rs_set_rdev_sectors(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4000) mddev_lock_nointr(mddev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4001) r = rs_start_reshape(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4002) mddev_unlock(mddev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4003) if (r)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4004) DMWARN("Failed to check/start reshape, continuing without change");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4005) r = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4006) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4007)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4008) return r;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4009) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4010)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4011) static void raid_resume(struct dm_target *ti)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4012) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4013) struct raid_set *rs = ti->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4014) struct mddev *mddev = &rs->md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4015)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4016) if (test_and_set_bit(RT_FLAG_RS_RESUMED, &rs->runtime_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4017) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4018) * A secondary resume while the device is active.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4019) * Take this opportunity to check whether any failed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4020) * devices are reachable again.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4021) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4022) attempt_restore_of_faulty_devices(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4023) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4024)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4025) if (test_and_clear_bit(RT_FLAG_RS_SUSPENDED, &rs->runtime_flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4026) /* Only reduce raid set size before running a disk removing reshape. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4027) if (mddev->delta_disks < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4028) rs_set_capacity(rs);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4029)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4030) mddev_lock_nointr(mddev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4031) clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4032) mddev->ro = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4033) mddev->in_sync = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4034) mddev_resume(mddev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4035) mddev_unlock(mddev);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4036) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4037) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4038)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4039) static struct target_type raid_target = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4040) .name = "raid",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4041) .version = {1, 15, 1},
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4042) .module = THIS_MODULE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4043) .ctr = raid_ctr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4044) .dtr = raid_dtr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4045) .map = raid_map,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4046) .status = raid_status,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4047) .message = raid_message,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4048) .iterate_devices = raid_iterate_devices,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4049) .io_hints = raid_io_hints,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4050) .postsuspend = raid_postsuspend,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4051) .preresume = raid_preresume,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4052) .resume = raid_resume,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4053) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4054)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4055) static int __init dm_raid_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4056) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4057) DMINFO("Loading target version %u.%u.%u",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4058) raid_target.version[0],
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4059) raid_target.version[1],
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4060) raid_target.version[2]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4061) return dm_register_target(&raid_target);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4062) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4063)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4064) static void __exit dm_raid_exit(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4065) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4066) dm_unregister_target(&raid_target);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4067) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4068)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4069) module_init(dm_raid_init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4070) module_exit(dm_raid_exit);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4071)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4072) module_param(devices_handle_discard_safely, bool, 0644);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4073) MODULE_PARM_DESC(devices_handle_discard_safely,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4074) "Set to Y if all devices in each array reliably return zeroes on reads from discarded regions");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4075)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4076) MODULE_DESCRIPTION(DM_NAME " raid0/1/10/4/5/6 target");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4077) MODULE_ALIAS("dm-raid0");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4078) MODULE_ALIAS("dm-raid1");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4079) MODULE_ALIAS("dm-raid10");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4080) MODULE_ALIAS("dm-raid4");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4081) MODULE_ALIAS("dm-raid5");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4082) MODULE_ALIAS("dm-raid6");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4083) MODULE_AUTHOR("Neil Brown <dm-devel@redhat.com>");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4084) MODULE_AUTHOR("Heinz Mauelshagen <dm-devel@redhat.com>");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4085) MODULE_LICENSE("GPL");
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags