diff options
Diffstat (limited to 'drivers/md/raid5.c')
| -rw-r--r-- | drivers/md/raid5.c | 1308 |
1 files changed, 1118 insertions, 190 deletions
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 31843604049..f920e50ea12 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -2,8 +2,11 @@ * raid5.c : Multiple Devices driver for Linux * Copyright (C) 1996, 1997 Ingo Molnar, Miguel de Icaza, Gadi Oxman * Copyright (C) 1999, 2000 Ingo Molnar + * Copyright (C) 2002, 2003 H. Peter Anvin * - * RAID-5 management functions. + * RAID-4/5/6 management functions. + * Thanks to Penguin Computing for making the RAID-6 development possible + * by donating a test server! * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -19,11 +22,11 @@ #include <linux/config.h> #include <linux/module.h> #include <linux/slab.h> -#include <linux/raid/raid5.h> #include <linux/highmem.h> #include <linux/bitops.h> #include <linux/kthread.h> #include <asm/atomic.h> +#include "raid6.h" #include <linux/raid/bitmap.h> @@ -68,6 +71,16 @@ #define __inline__ #endif +#if !RAID6_USE_EMPTY_ZERO_PAGE +/* In .bss so it's zeroed */ +const char raid6_empty_zero_page[PAGE_SIZE] __attribute__((aligned(256))); +#endif + +static inline int raid6_next_disk(int disk, int raid_disks) +{ + disk++; + return (disk < raid_disks) ? disk : 0; +} static void print_raid5_conf (raid5_conf_t *conf); static void __release_stripe(raid5_conf_t *conf, struct stripe_head *sh) @@ -104,7 +117,7 @@ static void release_stripe(struct stripe_head *sh) { raid5_conf_t *conf = sh->raid_conf; unsigned long flags; - + spin_lock_irqsave(&conf->device_lock, flags); __release_stripe(conf, sh); spin_unlock_irqrestore(&conf->device_lock, flags); @@ -117,7 +130,7 @@ static inline void remove_hash(struct stripe_head *sh) hlist_del_init(&sh->hash); } -static void insert_hash(raid5_conf_t *conf, struct stripe_head *sh) +static inline void insert_hash(raid5_conf_t *conf, struct stripe_head *sh) { struct hlist_head *hp = stripe_hash(conf, sh->sector); @@ -190,7 +203,7 @@ static void init_stripe(struct stripe_head *sh, sector_t sector, int pd_idx, int (unsigned long long)sh->sector); remove_hash(sh); - + sh->sector = sector; sh->pd_idx = pd_idx; sh->state = 0; @@ -269,8 +282,9 @@ static struct stripe_head *get_active_stripe(raid5_conf_t *conf, sector_t sector } else { if (!test_bit(STRIPE_HANDLE, &sh->state)) atomic_inc(&conf->active_stripes); - if (!list_empty(&sh->lru)) - list_del_init(&sh->lru); + if (list_empty(&sh->lru)) + BUG(); + list_del_init(&sh->lru); } } } while (sh == NULL); @@ -321,10 +335,9 @@ static int grow_stripes(raid5_conf_t *conf, int num) return 1; conf->slab_cache = sc; conf->pool_size = devs; - while (num--) { + while (num--) if (!grow_one_stripe(conf)) return 1; - } return 0; } @@ -631,8 +644,7 @@ static void raid5_build_block (struct stripe_head *sh, int i) dev->req.bi_private = sh; dev->flags = 0; - if (i != sh->pd_idx) - dev->sector = compute_blocknr(sh, i); + dev->sector = compute_blocknr(sh, i); } static void error(mddev_t *mddev, mdk_rdev_t *rdev) @@ -659,7 +671,7 @@ static void error(mddev_t *mddev, mdk_rdev_t *rdev) " Operation continuing on %d devices\n", bdevname(rdev->bdev,b), conf->working_disks); } -} +} /* * Input: a 'big' sector number, @@ -697,9 +709,12 @@ static sector_t raid5_compute_sector(sector_t r_sector, unsigned int raid_disks, /* * Select the parity disk based on the user selected algorithm. */ - if (conf->level == 4) + switch(conf->level) { + case 4: *pd_idx = data_disks; - else switch (conf->algorithm) { + break; + case 5: + switch (conf->algorithm) { case ALGORITHM_LEFT_ASYMMETRIC: *pd_idx = data_disks - stripe % raid_disks; if (*dd_idx >= *pd_idx) @@ -721,6 +736,39 @@ static sector_t raid5_compute_sector(sector_t r_sector, unsigned int raid_disks, default: printk(KERN_ERR "raid5: unsupported algorithm %d\n", conf->algorithm); + } + break; + case 6: + + /**** FIX THIS ****/ + switch (conf->algorithm) { + case ALGORITHM_LEFT_ASYMMETRIC: + *pd_idx = raid_disks - 1 - (stripe % raid_disks); + if (*pd_idx == raid_disks-1) + (*dd_idx)++; /* Q D D D P */ + else if (*dd_idx >= *pd_idx) + (*dd_idx) += 2; /* D D P Q D */ + break; + case ALGORITHM_RIGHT_ASYMMETRIC: + *pd_idx = stripe % raid_disks; + if (*pd_idx == raid_disks-1) + (*dd_idx)++; /* Q D D D P */ + else if (*dd_idx >= *pd_idx) + (*dd_idx) += 2; /* D D P Q D */ + break; + case ALGORITHM_LEFT_SYMMETRIC: + *pd_idx = raid_disks - 1 - (stripe % raid_disks); + *dd_idx = (*pd_idx + 2 + *dd_idx) % raid_disks; + break; + case ALGORITHM_RIGHT_SYMMETRIC: + *pd_idx = stripe % raid_disks; + *dd_idx = (*pd_idx + 2 + *dd_idx) % raid_disks; + break; + default: + printk (KERN_CRIT "raid6: unsupported algorithm %d\n", + conf->algorithm); + } + break; } /* @@ -742,12 +790,17 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i) int chunk_number, dummy1, dummy2, dd_idx = i; sector_t r_sector; + chunk_offset = sector_div(new_sector, sectors_per_chunk); stripe = new_sector; BUG_ON(new_sector != stripe); - - switch (conf->algorithm) { + if (i == sh->pd_idx) + return 0; + switch(conf->level) { + case 4: break; + case 5: + switch (conf->algorithm) { case ALGORITHM_LEFT_ASYMMETRIC: case ALGORITHM_RIGHT_ASYMMETRIC: if (i > sh->pd_idx) @@ -761,7 +814,37 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i) break; default: printk(KERN_ERR "raid5: unsupported algorithm %d\n", + conf->algorithm); + } + break; + case 6: + data_disks = raid_disks - 2; + if (i == raid6_next_disk(sh->pd_idx, raid_disks)) + return 0; /* It is the Q disk */ + switch (conf->algorithm) { + case ALGORITHM_LEFT_ASYMMETRIC: + case ALGORITHM_RIGHT_ASYMMETRIC: + if (sh->pd_idx == raid_disks-1) + i--; /* Q D D D P */ + else if (i > sh->pd_idx) + i -= 2; /* D D P Q D */ + break; + case ALGORITHM_LEFT_SYMMETRIC: + case ALGORITHM_RIGHT_SYMMETRIC: + if (sh->pd_idx == raid_disks-1) + i--; /* Q D D D P */ + else { + /* D D P Q D */ + if (i < sh->pd_idx) + i += raid_disks; + i -= (sh->pd_idx + 2); + } + break; + default: + printk (KERN_CRIT "raid6: unsupported algorithm %d\n", conf->algorithm); + } + break; } chunk_number = stripe * data_disks + i; @@ -778,10 +861,11 @@ static sector_t compute_blocknr(struct stripe_head *sh, int i) /* - * Copy data between a page in the stripe cache, and a bio. - * There are no alignment or size guarantees between the page or the - * bio except that there is some overlap. - * All iovecs in the bio must be considered. + * Copy data between a page in the stripe cache, and one or more bion + * The page could align with the middle of the bio, or there could be + * several bion, each with several bio_vecs, which cover part of the page + * Multiple bion are linked together on bi_next. There may be extras + * at the end of this list. We ignore them. */ static void copy_data(int frombio, struct bio *bio, struct page *page, @@ -810,7 +894,7 @@ static void copy_data(int frombio, struct bio *bio, if (len > 0 && page_offset + len > STRIPE_SIZE) clen = STRIPE_SIZE - page_offset; else clen = len; - + if (clen > 0) { char *ba = __bio_kmap_atomic(bio, i, KM_USER0); if (frombio) @@ -862,14 +946,14 @@ static void compute_block(struct stripe_head *sh, int dd_idx) set_bit(R5_UPTODATE, &sh->dev[dd_idx].flags); } -static void compute_parity(struct stripe_head *sh, int method) +static void compute_parity5(struct stripe_head *sh, int method) { raid5_conf_t *conf = sh->raid_conf; int i, pd_idx = sh->pd_idx, disks = sh->disks, count; void *ptr[MAX_XOR_BLOCKS]; struct bio *chosen; - PRINTK("compute_parity, stripe %llu, method %d\n", + PRINTK("compute_parity5, stripe %llu, method %d\n", (unsigned long long)sh->sector, method); count = 1; @@ -956,9 +1040,195 @@ static void compute_parity(struct stripe_head *sh, int method) clear_bit(R5_UPTODATE, &sh->dev[pd_idx].flags); } +static void compute_parity6(struct stripe_head *sh, int method) +{ + raid6_conf_t *conf = sh->raid_conf; + int i, pd_idx = sh->pd_idx, qd_idx, d0_idx, disks = conf->raid_disks, count; + struct bio *chosen; + /**** FIX THIS: This could be very bad if disks is close to 256 ****/ + void *ptrs[disks]; + + qd_idx = raid6_next_disk(pd_idx, disks); + d0_idx = raid6_next_disk(qd_idx, disks); + + PRINTK("compute_parity, stripe %llu, method %d\n", + (unsigned long long)sh->sector, method); + + switch(method) { + case READ_MODIFY_WRITE: + BUG(); /* READ_MODIFY_WRITE N/A for RAID-6 */ + case RECONSTRUCT_WRITE: + for (i= disks; i-- ;) + if ( i != pd_idx && i != qd_idx && sh->dev[i].towrite ) { + chosen = sh->dev[i].towrite; + sh->dev[i].towrite = NULL; + + if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags)) + wake_up(&conf->wait_for_overlap); + + if (sh->dev[i].written) BUG(); + sh->dev[i].written = chosen; + } + break; + case CHECK_PARITY: + BUG(); /* Not implemented yet */ + } + + for (i = disks; i--;) + if (sh->dev[i].written) { + sector_t sector = sh->dev[i].sector; + struct bio *wbi = sh->dev[i].written; + while (wbi && wbi->bi_sector < sector + STRIPE_SECTORS) { + copy_data(1, wbi, sh->dev[i].page, sector); + wbi = r5_next_bio(wbi, sector); + } + + set_bit(R5_LOCKED, &sh->dev[i].flags); + set_bit(R5_UPTODATE, &sh->dev[i].flags); + } + +// switch(method) { +// case RECONSTRUCT_WRITE: +// case CHECK_PARITY: +// case UPDATE_PARITY: + /* Note that unlike RAID-5, the ordering of the disks matters greatly. */ + /* FIX: Is this ordering of drives even remotely optimal? */ + count = 0; + i = d0_idx; + do { + ptrs[count++] = page_address(sh->dev[i].page); + if (count <= disks-2 && !test_bit(R5_UPTODATE, &sh->dev[i].flags)) + printk("block %d/%d not uptodate on parity calc\n", i,count); + i = raid6_next_disk(i, disks); + } while ( i != d0_idx ); +// break; +// } + + raid6_call.gen_syndrome(disks, STRIPE_SIZE, ptrs); + + switch(method) { + case RECONSTRUCT_WRITE: + set_bit(R5_UPTODATE, &sh->dev[pd_idx].flags); + set_bit(R5_UPTODATE, &sh->dev[qd_idx].flags); + set_bit(R5_LOCKED, &sh->dev[pd_idx].flags); + set_bit(R5_LOCKED, &sh->dev[qd_idx].flags); + break; + case UPDATE_PARITY: + set_bit(R5_UPTODATE, &sh->dev[pd_idx].flags); + set_bit(R5_UPTODATE, &sh->dev[qd_idx].flags); + break; + } +} + + +/* Compute one missing block */ +static void compute_block_1(struct stripe_head *sh, int dd_idx, int nozero) +{ + raid6_conf_t *conf = sh->raid_conf; + int i, count, disks = conf->raid_disks; + void *ptr[MAX_XOR_BLOCKS], *p; + int pd_idx = sh->pd_idx; + int qd_idx = raid6_next_disk(pd_idx, disks); + + PRINTK("compute_block_1, stripe %llu, idx %d\n", + (unsigned long long)sh->sector, dd_idx); + + if ( dd_idx == qd_idx ) { + /* We're actually computing the Q drive */ + compute_parity6(sh, UPDATE_PARITY); + } else { + ptr[0] = page_address(sh->dev[dd_idx].page); + if (!nozero) memset(ptr[0], 0, STRIPE_SIZE); + count = 1; + for (i = disks ; i--; ) { + if (i == dd_idx || i == qd_idx) + continue; + p = page_address(sh->dev[i].page); + if (test_bit(R5_UPTODATE, &sh->dev[i].flags)) + ptr[count++] = p; + else + printk("compute_block() %d, stripe %llu, %d" + " not present\n", dd_idx, + (unsigned long long)sh->sector, i); + + check_xor(); + } + if (count != 1) + xor_block(count, STRIPE_SIZE, ptr); + if (!nozero) set_bit(R5_UPTODATE, &sh->dev[dd_idx].flags); + else clear_bit(R5_UPTODATE, &sh->dev[dd_idx].flags); + } +} + +/* Compute two missing blocks */ +static void compute_block_2(struct stripe_head *sh, int dd_idx1, int dd_idx2) +{ + raid6_conf_t *conf = sh->raid_conf; + int i, count, disks = conf->raid_disks; + int pd_idx = sh->pd_idx; + int qd_idx = raid6_next_disk(pd_idx, disks); + int d0_idx = raid6_next_disk(qd_idx, disks); + int faila, failb; + + /* faila and failb are disk numbers relative to d0_idx */ + /* pd_idx become disks-2 and qd_idx become disks-1 */ + faila = (dd_idx1 < d0_idx) ? dd_idx1+(disks-d0_idx) : dd_idx1-d0_idx; + failb = (dd_idx2 < d0_idx) ? dd_idx2+(disks-d0_idx) : dd_idx2-d0_idx; + + BUG_ON(faila == failb); + if ( failb < faila ) { int tmp = faila; faila = failb; failb = tmp; } + + PRINTK("compute_block_2, stripe %llu, idx %d,%d (%d,%d)\n", + (unsigned long long)sh->sector, dd_idx1, dd_idx2, faila, failb); + + if ( failb == disks-1 ) { + /* Q disk is one of the missing disks */ + if ( faila == disks-2 ) { + /* Missing P+Q, just recompute */ + compute_parity6(sh, UPDATE_PARITY); + return; + } else { + /* We're missing D+Q; recompute D from P */ + compute_block_1(sh, (dd_idx1 == qd_idx) ? dd_idx2 : dd_idx1, 0); + compute_parity6(sh, UPDATE_PARITY); /* Is this necessary? */ + return; + } + } + + /* We're missing D+P or D+D; build pointer table */ + { + /**** FIX THIS: This could be very bad if disks is close to 256 ****/ + void *ptrs[disks]; + + count = 0; + i = d0_idx; + do { + ptrs[count++] = page_address(sh->dev[i].page); + i = raid6_next_disk(i, disks); + if (i != dd_idx1 && i != dd_idx2 && + !test_bit(R5_UPTODATE, &sh->dev[i].flags)) + printk("compute_2 with missing block %d/%d\n", count, i); + } while ( i != d0_idx ); + + if ( failb == disks-2 ) { + /* We're missing D+P. */ + raid6_datap_recov(disks, STRIPE_SIZE, faila, ptrs); + } else { + /* We're missing D+D. */ + raid6_2data_recov(disks, STRIPE_SIZE, faila, failb, ptrs); + } + + /* Both the above update both missing blocks */ + set_bit(R5_UPTODATE, &sh->dev[dd_idx1].flags); + set_bit(R5_UPTODATE, &sh->dev[dd_idx2].flags); + } +} + + + /* * Each stripe/dev can have one or more bion attached. - * toread/towrite point to the first in a chain. + * toread/towrite point to the first in a chain. * The bi_next chain must be in order. */ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, int forwrite) @@ -1031,6 +1301,13 @@ static int add_stripe_bio(struct stripe_head *sh, struct bio *bi, int dd_idx, in static void end_reshape(raid5_conf_t *conf); +static int page_is_zero(struct page *p) +{ + char *a = page_address(p); + return ((*(u32*)a) == 0 && + memcmp(a, a+4, STRIPE_SIZE-4)==0); +} + static int stripe_to_pdidx(sector_t stripe, raid5_conf_t *conf, int disks) { int sectors_per_chunk = conf->chunk_size >> 9; @@ -1062,7 +1339,7 @@ static int stripe_to_pdidx(sector_t stripe, raid5_conf_t *conf, int disks) * */ -static void handle_stripe(struct stripe_head *sh) +static void handle_stripe5(struct stripe_head *sh) { raid5_conf_t *conf = sh->raid_conf; int disks = sh->disks; @@ -1394,7 +1671,7 @@ static void handle_stripe(struct stripe_head *sh) if (locked == 0 && (rcw == 0 ||rmw == 0) && !test_bit(STRIPE_BIT_DELAY, &sh->state)) { PRINTK("Computing parity...\n"); - compute_parity(sh, rcw==0 ? RECONSTRUCT_WRITE : READ_MODIFY_WRITE); + compute_parity5(sh, rcw==0 ? RECONSTRUCT_WRITE : READ_MODIFY_WRITE); /* now every locked buffer is ready to be written */ for (i=disks; i--;) if (test_bit(R5_LOCKED, &sh->dev[i].flags)) { @@ -1421,13 +1698,10 @@ static void handle_stripe(struct stripe_head *sh) !test_bit(STRIPE_INSYNC, &sh->state)) { set_bit(STRIPE_HANDLE, &sh->state); if (failed == 0) { - char *pagea; BUG_ON(uptodate != disks); - compute_parity(sh, CHECK_PARITY); + compute_parity5(sh, CHECK_PARITY); uptodate--; - pagea = page_address(sh->dev[sh->pd_idx].page); - if ((*(u32*)pagea) == 0 && - !memcmp(pagea, pagea+4, STRIPE_SIZE-4)) { + if (page_is_zero(sh->dev[sh->pd_idx].page)) { /* parity is correct (on disc, not in buffer any more) */ set_bit(STRIPE_INSYNC, &sh->state); } else { @@ -1487,7 +1761,7 @@ static void handle_stripe(struct stripe_head *sh) /* Need to write out all blocks after computing parity */ sh->disks = conf->raid_disks; sh->pd_idx = stripe_to_pdidx(sh->sector, conf, conf->raid_disks); - compute_parity(sh, RECONSTRUCT_WRITE); + compute_parity5(sh, RECONSTRUCT_WRITE); for (i= conf->raid_disks; i--;) { set_bit(R5_LOCKED, &sh->dev[i].flags); locked++; @@ -1615,6 +1889,569 @@ static void handle_stripe(struct stripe_head *sh) } } +static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page) +{ + raid6_conf_t *conf = sh->raid_conf; + int disks = conf->raid_disks; + struct bio *return_bi= NULL; + struct bio *bi; + int i; + int syncing; + int locked=0, uptodate=0, to_read=0, to_write=0, failed=0, written=0; + int non_overwrite = 0; + int failed_num[2] = {0, 0}; + struct r5dev *dev, *pdev, *qdev; + int pd_idx = sh->pd_idx; + int qd_idx = raid6_next_disk(pd_idx, disks); + int p_failed, q_failed; + + PRINTK("handling stripe %llu, state=%#lx cnt=%d, pd_idx=%d, qd_idx=%d\n", + (unsigned long long)sh->sector, sh->state, atomic_read(&sh->count), + pd_idx, qd_idx); + + spin_lock(&sh->lock); + clear_bit(STRIPE_HANDLE, &sh->state); + clear_bit(STRIPE_DELAYED, &sh->state); + + syncing = test_bit(STRIPE_SYNCING, &sh->state); + /* Now to look around and see what can be done */ + + rcu_read_lock(); + for (i=disks; i--; ) { + mdk_rdev_t *rdev; + dev = &sh->dev[i]; + clear_bit(R5_Insync, &dev->flags); + + PRINTK("check %d: state 0x%lx read %p write %p written %p\n", + i, dev->flags, dev->toread, dev->towrite, dev->written); + /* maybe we can reply to a read */ + if (test_bit(R5_UPTODATE, &dev->flags) && dev->toread) { + struct bio *rbi, *rbi2; + PRINTK("Return read for disc %d\n", i); + spin_lock_irq(&conf->device_lock); + rbi = dev->toread; + dev->toread = NULL; + if (test_and_clear_bit(R5_Overlap, &dev->flags)) + wake_up(&conf->wait_for_overlap); + spin_unlock_irq(&conf->device_lock); + while (rbi && rbi->bi_sector < dev->sector + STRIPE_SECTORS) { + copy_data(0, rbi, dev->page, dev->sector); + rbi2 = r5_next_bio(rbi, dev->sector); + spin_lock_irq(&conf->device_lock); + if (--rbi->bi_phys_segments == 0) { + rbi->bi_next = return_bi; + return_bi = rbi; + } + spin_unlock_irq(&conf->device_lock); + rbi = rbi2; + } + } + + /* now count some things */ + if (test_bit(R5_LOCKED, &dev->flags)) locked++; + if (test_bit(R5_UPTODATE, &dev->flags)) uptodate++; + + + if (dev->toread) to_read++; + if (dev->towrite) { + to_write++; + if (!test_bit(R5_OVERWRITE, &dev->flags)) + non_overwrite++; + } + if (dev->written) written++; + rdev = rcu_dereference(conf->disks[i].rdev); + if (!rdev || !test_bit(In_sync, &rdev->flags)) { + /* The ReadError flag will just be confusing now */ + clear_bit(R5_ReadError, &dev->flags); + clear_bit(R5_ReWrite, &dev->flags); + } + if (!rdev || !test_bit(In_sync, &rdev->flags) + || test_bit(R5_ReadError, &dev->flags)) { + if ( failed < 2 ) + failed_num[failed] = i; + failed++; + } else + set_bit(R5_Insync, &dev->flags); + } + rcu_read_unlock(); + PRINTK("locked=%d uptodate=%d to_read=%d" + " to_write=%d failed=%d failed_num=%d,%d\n", + locked, uptodate, to_read, to_write, failed, + failed_num[0], failed_num[1]); + /* check if the array has lost >2 devices and, if so, some requests might + * need to be failed + */ + if (failed > 2 && to_read+to_write+written) { + for (i=disks; i--; ) { + int bitmap_end = 0; + + if (test_bit(R5_ReadError, &sh->dev[i].flags)) { + mdk_rdev_t *rdev; + rcu_read_lock(); + rdev = rcu_dereference(conf->disks[i].rdev); + if (rdev && test_bit(In_sync, &rdev->flags)) + /* multiple read failures in one stripe */ + md_error(conf->mddev, rdev); + rcu_read_unlock(); + } + + spin_lock_irq(&conf->device_lock); + /* fail all writes first */ + bi = sh->dev[i].towrite; + sh->dev[i].towrite = NULL; + if (bi) { to_write--; bitmap_end = 1; } + + if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags)) + wake_up(&conf->wait_for_overlap); + + while (bi && bi->bi_sector < sh->dev[i].sector + STRIPE_SECTORS){ + struct bio *nextbi = r5_next_bio(bi, sh->dev[i].sector); + clear_bit(BIO_UPTODATE, &bi->bi_flags); + if (--bi->bi_phys_segments == 0) { + md_write_end(conf->mddev); + bi->bi_next = return_bi; + return_bi = bi; + } + bi = nextbi; + } + /* and fail all 'written' */ + bi = sh->dev[i].written; + sh->dev[i].written = NULL; + if (bi) bitmap_end = 1; + while (bi && bi->bi_sector < sh->dev[i].sector + STRIPE_SECTORS) { + struct bio *bi2 = r5_next_bio(bi, sh->dev[i].sector); + clear_bit(BIO_UPTODATE, &bi->bi_flags); + if (--bi->bi_phys_segments == 0) { + md_write_end(conf->mddev); + bi->bi_next = return_bi; + return_bi = bi; + } + bi = bi2; + } + + /* fail any reads if this device is non-operational */ + if (!test_bit(R5_Insync, &sh->dev[i].flags) || + test_bit(R5_ReadError, &sh->dev[i].flags)) { + bi = sh->dev[i].toread; + sh->dev[i].toread = NULL; + if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags)) + wake_up(&conf->wait_for_overlap); + if (bi) to_read--; + while (bi && bi->bi_sector < sh->dev[i].sector + STRIPE_SECTORS){ + struct bio *nextbi = r5_next_bio(bi, sh->dev[i].sector); + clear_bit(BIO_UPTODATE, &bi->bi_flags); + if (--bi->bi_phys_segments == 0) { + bi->bi_next = return_bi; + return_bi = bi; + } + bi = nextbi; + } + } + spin_unlock_irq(&conf->device_lock); + if (bitmap_end) + bitmap_endwrite(conf->mddev->bitmap, sh->sector, + STRIPE_SECTORS, 0, 0); + } + } + if (failed > 2 && syncing) { + md_done_sync(conf->mddev, STRIPE_SECTORS,0); + clear_bit(STRIPE_SYNCING, &sh->state); + syncing = 0; + } + + /* + * might be able to return some write requests if the parity blocks + * are safe, or on a failed drive + */ + pdev = &sh->dev[pd_idx]; + p_failed = (failed >= 1 && failed_num[0] == pd_idx) + || (failed >= 2 && failed_num[1] == pd_idx); + qdev = &sh->dev[qd_idx]; + q_failed = (failed >= 1 && failed_num[0] == qd_idx) + || (failed >= 2 && failed_num[1] == qd_idx); + + if ( written && + ( p_failed || ((test_bit(R5_Insync, &pdev->flags) + && !test_bit(R5_LOCKED, &pdev->flags) + && test_bit(R5_UPTODATE, &pdev->flags))) ) && + ( q_failed || ((test_bit(R5_Insync, &qdev->flags) + && !test_bit(R5_LOCKED, &qdev->flags) + && test_bit(R5_UPTODATE, &qdev->flags))) ) ) { + /* any written block on an uptodate or failed drive can be + * returned. Note that if we 'wrote' to a failed drive, + * it will be UPTODATE, but never LOCKED, so we don't need + * to test 'failed' directly. + */ + for (i=disks; i--; ) + if (sh->dev[i].written) { + dev = &sh->dev[i]; + if (!test_bit(R5_LOCKED, &dev->flags) && + test_bit(R5_UPTODATE, &dev->flags) ) { + /* We can return any write requests */ + int bitmap_end = 0; + struct bio *wbi, *wbi2; + PRINTK("Return write for stripe %llu disc %d\n", + (unsigned long long)sh->sector, i); + spin_lock_irq(&conf->device_lock); + wbi = dev->written; + dev->written = NULL; + while (wbi && wbi->bi_sector < dev->sector + STRIPE_SECTORS) { + wbi2 = r5_next_bio(wbi, dev->sector); + if (--wbi->bi_phys_segments == 0) { + md_write_end(conf->mddev); + wbi->bi_next = return_bi; + return_bi = wbi; + } + wbi = wbi2; + } + if (dev->towrite == NULL) + bitmap_end = 1; + spin_unlock_irq(&conf->device_lock); + if (bitmap_end) + bitmap_endwrite(conf->mddev->bitmap, sh->sector, + STRIPE_SECTORS, + !test_bit(STRIPE_DEGRADED, &sh->state), 0); + } + } + } + + /* Now we might consider reading some blocks, either to check/generate + * parity, or to satisfy requests + * or to load a block that is being partially written. + */ + if (to_read || non_overwrite || (to_write && failed) || (syncing && (uptodate < disks))) { + for (i=disks; i--;) { + dev = &sh->dev[i]; + if (!test_bit(R5_LOCKED, &dev->flags) && !test_bit(R5_UPTODATE, &dev->flags) && + (dev->toread || + (dev->towrite && !test_bit(R5_OVERWRITE, &dev->flags)) || + syncing || + (failed >= 1 && (sh->dev[failed_num[0]].toread || to_write)) || + (failed >= 2 && (sh->dev[failed_num[1]].toread || to_write)) + ) + ) { + /* we would like to get this block, possibly + * by computing it, but we might not be able to + */ + if (uptodate == disks-1) { + PRINTK("Computing stripe %llu block %d\n", + (unsigned long long)sh->sector, i); + compute_block_1(sh, i, 0); + uptodate++; + } else if ( uptodate == disks-2 && failed >= 2 ) { + /* Computing 2-failure is *very* expensive; only do it if failed >= 2 */ + int other; + for (other=disks; other--;) { + if ( other == i ) + continue; + if ( !test_bit(R5_UPTODATE, &sh->dev[other].flags) ) + break; + } + BUG_ON(other < 0); + PRINTK("Computing stripe %llu blocks %d,%d\n", + (unsigned long long)sh->sector, i, other); + compute_block_2(sh, i, other); + uptodate += 2; + } else if (test_bit(R5_Insync, &dev->flags)) { + set_bit(R5_LOCKED, &dev->flags); + set_bit(R5_Wantread, &dev->flags); +#if 0 + /* if I am just reading this block and we don't have + a failed drive, or any pending writes then sidestep the cache */ + if (sh->bh_read[i] && !sh->bh_read[i]->b_reqnext && + ! syncing && !failed && !to_write) { + sh->bh_cache[i]->b_page = sh->bh_read[i]->b_page; + sh->bh_cache[i]->b_data = sh->bh_read[i]->b_data; + } +#endif + locked++; + PRINTK("Reading block %d (sync=%d)\n", + i, syncing); + } + } + } + set_bit(STRIPE_HANDLE, &sh->state); + } + + /* now to consider writing and what else, if anything should be read */ + if (to_write) { + int rcw=0, must_compute=0; + for (i=disks ; i--;) { + dev = &sh->dev[i]; + /* Would I have to read this buffer for reconstruct_write */ + if (!test_bit(R5_OVERWRITE, &dev->flags) + && i != pd_idx && i != qd_idx + && (!test_bit(R5_LOCKED, &dev->flags) +#if 0 + || sh->bh_page[i] != bh->b_page +#endif + ) && + !test_bit(R5_UPTODATE, &dev->flags)) { + if (test_bit(R5_Insync, &dev->flags)) rcw++; + else { + PRINTK("raid6: must_compute: disk %d flags=%#lx\n", i, dev->flags); + must_compute++; + } + } + } + PRINTK("for sector %llu, rcw=%d, must_compute=%d\n", + (unsigned long long)sh->sector, rcw, must_compute); + set_bit(STRIPE_HANDLE, &sh->state); + + if (rcw > 0) + /* want reconstruct write, but need to get some data */ + for (i=disks; i--;) { + dev = &sh->dev[i]; + if (!test_bit(R5_OVERWRITE, &dev->flags) + && !(failed == 0 && (i == pd_idx || i == qd_idx)) + && !test_bit(R5_LOCKED, &dev->flags) && !test_bit(R5_UPTODATE, &dev->flags) && + test_bit(R5_Insync, &dev->flags)) { + if (test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) + { + PRINTK("Read_old stripe %llu block %d for Reconstruct\n", + (unsigned long long)sh->sector, i); + set_bit(R5_LOCKED, &dev->flags); + set_bit(R5_Wantread, &dev->flags); + locked++; + } else { + PRINTK("Request delayed stripe %llu block %d for Reconstruct\n", + (unsigned long long)sh->sector, i); + set_bit(STRIPE_DELAYED, &sh->state); + set_bit(STRIPE_HANDLE, &sh->state); + } + } + } + /* now if nothing is locked, and if we have enough data, we can start a write request */ + if (locked == 0 && rcw == 0 && + !test_bit(STRIPE_BIT_DELAY, &sh->state)) { + if ( must_compute > 0 ) { + /* We have failed blocks and need to compute them */ + switch ( failed ) { + case 0: BUG(); + case 1: compute_block_1(sh, failed_num[0], 0); break; + case 2: compute_block_2(sh, failed_num[0], failed_num[1]); break; + default: BUG(); /* This request should have been failed? */ + } + } + + PRINTK("Computing parity for stripe %llu\n", (unsigned long long)sh->sector); + compute_parity6(sh, RECONSTRUCT_WRITE); + /* now every locked buffer is ready to be written */ + for (i=disks; i--;) + if (test_bit(R5_LOCKED, &sh->dev[i].flags)) { + PRINTK("Writing stripe %llu block %d\n", + (unsigned long long)sh->sector, i); + locked++; + set_bit(R5_Wantwrite, &sh->dev[i].flags); + } + /* after a RECONSTRUCT_WRITE, the stripe MUST be in-sync */ + set_bit(STRIPE_INSYNC, &sh->state); + + if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) { + atomic_dec(&conf->preread_active_stripes); + if (atomic_read(&conf->preread_active_stripes) < IO_THRESHOLD) + md_wakeup_thread(conf->mddev->thread); + } + } + } + + /* maybe we need to check and possibly fix the parity for this stripe + * Any reads will already have been scheduled, so we just see if enough data + * is available + */ + if (syncing && locked == 0 && !test_bit(STRIPE_INSYNC, &sh->state)) { + int update_p = 0, update_q = 0; + struct r5dev *dev; + + set_bit(STRIPE_HANDLE, &sh->state); + + BUG_ON(failed>2); + BUG_ON(uptodate < disks); + /* Want to check and possibly repair P and Q. + * However there could be one 'failed' device, in which + * case we can only check one of them, possibly using the + * other to generate missing data + */ + + /* If !tmp_page, we cannot do the calculations, + * but as we have set STRIPE_HANDLE, we will soon be called + * by stripe_handle with a tmp_page - just wait until then. + */ + if (tmp_page) { + if (failed == q_failed) { + /* The only possible failed device holds 'Q', so it makes + * sense to check P (If anything else were failed, we would + * have used P to recreate it). + */ + compute_block_1(sh, pd_idx, 1); + if (!page_is_zero(sh->dev[pd_idx].page)) { + compute_block_1(sh,pd_idx,0); + update_p = 1; + } + } + if (!q_failed && failed < 2) { + /* q is not failed, and we didn't use it to generate + * anything, so it makes sense to check it + */ + memcpy(page_address(tmp_page), + page_address(sh->dev[qd_idx].page), + STRIPE_SIZE); + compute_parity6(sh, UPDATE_PARITY); + if (memcmp(page_address(tmp_page), + page_address(sh->dev[qd_idx].page), + STRIPE_SIZE)!= 0) { + clear_bit(STRIPE_INSYNC, &sh->state); + update_q = 1; + } + } + if (update_p || update_q) { + conf->mddev->resync_mismatches += STRIPE_SECTORS; + if (test_bit(MD_RECOVERY_CHECK, &conf->mddev->recovery)) + /* don't try to repair!! */ + update_p = update_q = 0; + } + + /* now write out any block on a failed drive, + * or P or Q if they need it + */ + + if (failed == 2) { + dev = &sh->dev[failed_num[1]]; + locked++; + set_bit(R5_LOCKED, &dev->flags); + set_bit(R5_Wantwrite, &dev->flags); + } + if (failed >= 1) { + dev = &sh->dev[failed_num[0]]; + locked++; + set_bit(R5_LOCKED, &dev->flags); + set_bit(R5_Wantwrite, &dev->flags); + } + + if (update_p) { + dev = &sh->dev[pd_idx]; + locked ++; + set_bit(R5_LOCKED, &dev->flags); + set_bit(R5_Wantwrite, &dev->flags); + } + if (update_q) { + dev = &sh->dev[qd_idx]; + locked++; + set_bit(R5_LOCKED, &dev->flags); + set_bit(R5_Wantwrite, &dev->flags); + } + clear_bit(STRIPE_DEGRADED, &sh->state); + + set_bit(STRIPE_INSYNC, &sh->state); + } + } + + if (syncing && locked == 0 && test_bit(STRIPE_INSYNC, &sh->state)) { + md_done_sync(conf->mddev, STRIPE_SECTORS,1); + clear_bit(STRIPE_SYNCING, &sh->state); + } + + /* If the failed drives are just a ReadError, then we might need + * to progress the repair/check process + */ + if (failed <= 2 && ! conf->mddev->ro) + for (i=0; i<failed;i++) { + dev = &sh->dev[failed_num[i]]; + if (test_bit(R5_ReadError, &dev->flags) + && !test_bit(R5_LOCKED, &dev->flags) + && test_bit(R5_UPTODATE, &dev->flags) + ) { + if (!test_bit(R5_ReWrite, &dev->flags)) { + set_bit(R5_Wantwrite, &dev->flags); + set_bit(R5_ReWrite, &dev->flags); + set_bit(R5_LOCKED, &dev->flags); + } else { + /* let's read it back */ + set_bit(R5_Wantread, &dev->flags); + set_bit(R5_LOCKED, &dev->flags); + } + } + } + spin_unlock(&sh->lock); + + while ((bi=return_bi)) { + int bytes = bi->bi_size; + + return_bi = bi->bi_next; + bi->bi_next = NULL; + bi->bi_size = 0; + bi->bi_end_io(bi, bytes, 0); + } + for (i=disks; i-- ;) { + int rw; + struct bio *bi; + mdk_rdev_t *rdev; + if (test_and_clear_bit(R5_Wantwrite, &sh->dev[i].flags)) + rw = 1; + else if (test_and_clear_bit(R5_Wantread, &sh->dev[i].flags)) + rw = 0; + else + continue; + + bi = &sh->dev[i].req; + + bi->bi_rw = rw; + if (rw) + bi->bi_end_io = raid5_end_write_request; + else + bi->bi_end_io = raid5_end_read_request; + + rcu_read_lock(); + rdev = rcu_dereference(conf->disks[i].rdev); + if (rdev && test_bit(Faulty, &rdev->flags)) + rdev = NULL; + if (rdev) + atomic_inc(&rdev->nr_pending); + rcu_read_unlock(); + + if (rdev) { + if (syncing) + md_sync_acct(rdev->bdev, STRIPE_SECTORS); + + bi->bi_bdev = rdev->bdev; + PRINTK("for %llu schedule op %ld on disc %d\n", + (unsigned long long)sh->sector, bi->bi_rw, i); + atomic_inc(&sh->count); + bi->bi_sector = sh->sector + rdev->data_offset; + bi->bi_flags = 1 << BIO_UPTODATE; + bi->bi_vcnt = 1; + bi->bi_max_vecs = 1; + bi->bi_idx = 0; + bi->bi_io_vec = &sh->dev[i].vec; + bi->bi_io_vec[0].bv_len = STRIPE_SIZE; + bi->bi_io_vec[0].bv_offset = 0; + bi->bi_size = STRIPE_SIZE; + bi->bi_next = NULL; + if (rw == WRITE && + test_bit(R5_ReWrite, &sh->dev[i].flags)) + atomic_add(STRIPE_SECTORS, &rdev->corrected_errors); + generic_make_request(bi); + } else { + if (rw == 1) + set_bit(STRIPE_DEGRADED, &sh->state); + PRINTK("skip op %ld on disc %d for sector %llu\n", + bi->bi_rw, i, (unsigned long long)sh->sector); + clear_bit(R5_LOCKED, &sh->dev[i].flags); + set_bit(STRIPE_HANDLE, &sh->state); + } + } +} + +static void handle_stripe(struct stripe_head *sh, struct page *tmp_page) +{ + if (sh->raid_conf->level == 6) + handle_stripe6(sh, tmp_page); + else + handle_stripe5(sh); +} + + + static void raid5_activate_delayed(raid5_conf_t *conf) { if (atomic_read(&conf->preread_active_stripes) < IO_THRESHOLD) { @@ -1753,7 +2590,7 @@ static int make_request(request_queue_t *q, struct bio * bi) for (;logical_sector < last_sector; logical_sector += STRIPE_SECTORS) { DEFINE_WAIT(w); - int disks; + int disks, data_disks; retry: prepare_to_wait(&conf->wait_for_overlap, &w, TASK_UNINTERRUPTIBLE); @@ -1781,7 +2618,9 @@ static int make_request(request_queue_t *q, struct bio * bi) } spin_unlock_irq(&conf->device_lock); } - new_sector = raid5_compute_sector(logical_sector, disks, disks - 1, + data_disks = disks - conf->max_degraded; + + new_sector = raid5_compute_sector(logical_sector, disks, data_disks, &dd_idx, &pd_idx, conf); PRINTK("raid5: make_request, sector %llu logical %llu\n", (unsigned long long)new_sector, @@ -1833,7 +2672,7 @@ static int make_request(request_queue_t *q, struct bio * bi) } finish_wait(&conf->wait_for_overlap, &w); raid5_plug_device(conf); - handle_stripe(sh); + handle_stripe(sh, NULL); release_stripe(sh); } else { /* cannot get stripe for read-ahead, just give-up */ @@ -1849,7 +2688,7 @@ static int make_request(request_queue_t *q, struct bio * bi) if (remaining == 0) { int bytes = bi->bi_size; - if ( bio_data_dir(bi) == WRITE ) + if ( rw == WRITE ) md_write_end(mddev); bi->bi_size = 0; bi->bi_end_io(bi, bytes, 0); @@ -1857,17 +2696,142 @@ static int make_request(request_queue_t *q, struct bio * bi) return 0; } -/* FIXME go_faster isn't used */ -static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster) +static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped) { + /* reshaping is quite different to recovery/resync so it is + * handled quite separately ... here. + * + * On each call to sync_request, we gather one chunk worth of + * destination stripes and flag them as expanding. + * Then we find all the source stripes and request reads. + * As the reads complete, handle_stripe will copy the data + * into the destination stripe and release that stripe. + */ raid5_conf_t *conf = (raid5_conf_t *) mddev->private; struct stripe_head *sh; int pd_idx; sector_t first_sector, last_sector; + int raid_disks; + int data_disks; + int i; + int dd_idx; + sector_t writepos, safepos, gap; + + if (sector_nr == 0 && + conf->expand_progress != 0) { + /* restarting in the middle, skip the initial sectors */ + sector_nr = conf->expand_progress; + sector_div(sector_nr, conf->raid_disks-1); + *skipped = 1; + return sector_nr; + } + + /* we update the metadata when there is more than 3Meg + * in the block range (that is rather arbitrary, should + * probably be time based) or when the data about to be + * copied would over-write the source of the data at + * the front of the range. + * i.e. one new_stripe forward from expand_progress new_maps + * to after where expand_lo old_maps to + */ + writepos = conf->expand_progress + + conf->chunk_size/512*(conf->raid_disks-1); + sector_div(writepos, conf->raid_disks-1); + safepos = conf->expand_lo; + sector_div(safepos, conf->previous_raid_disks-1); + gap = conf->expand_progress - conf->expand_lo; + + if (writepos >= safepos || + gap > (conf->raid_disks-1)*3000*2 /*3Meg*/) { + /* Cannot proceed until we've updated the superblock... */ + wait_event(conf->wait_for_overlap, + atomic_read(&conf->reshape_stripes)==0); + mddev->reshape_position = conf->expand_progress; + mddev->sb_dirty = 1; + md_wakeup_thread(mddev->thread); + wait_event(mddev->sb_wait, mddev->sb_dirty == 0 || + kthread_should_stop()); + spin_lock_irq(&conf->device_lock); + conf->expand_lo = mddev->reshape_position; + spin_unlock_irq(&conf->device_lock); + wake_up(&conf->wait_for_overlap); + } + + for (i=0; i < conf->chunk_size/512; i+= STRIPE_SECTORS) { + int j; + int skipped = 0; + pd_idx = stripe_to_pdidx(sector_nr+i, conf, conf->raid_disks); + sh = get_active_stripe(conf, sector_nr+i, + conf->raid_disks, pd_idx, 0); + set_bit(STRIPE_EXPANDING, &sh->state); + atomic_inc(&conf->reshape_stripes); + /* If any of this stripe is beyond the end of the old + * array, then we need to zero those blocks + */ + for (j=sh->disks; j--;) { + sector_t s; + if (j == sh->pd_idx) + continue; + s = compute_blocknr(sh, j); + if (s < (mddev->array_size<<1)) { + skipped = 1; + continue; + } + memset(page_address(sh->dev[j].page), 0, STRIPE_SIZE); + set_bit(R5_Expanded, &sh->dev[j].flags); + set_bit(R5_UPTODATE, &sh->dev[j].flags); + } + if (!skipped) { + set_bit(STRIPE_EXPAND_READY, &sh->state); + set_bit(STRIPE_HANDLE, &sh->state); + } + release_stripe(sh); + } + spin_lock_irq(&conf->device_lock); + conf->expand_progress = (sector_nr + i)*(conf->raid_disks-1); + spin_unlock_irq(&conf->device_lock); + /* Ok, those stripe are ready. We can start scheduling + * reads on the source stripes. + * The source stripes are determined by mapping the first and last + * block on the destination stripes. + */ + raid_disks = conf->previous_raid_disks; + data_disks = raid_disks - 1; + first_sector = + raid5_compute_sector(sector_nr*(conf->raid_disks-1), + raid_disks, data_disks, + &dd_idx, &pd_idx, conf); + last_sector = + raid5_compute_sector((sector_nr+conf->chunk_size/512) + *(conf->raid_disks-1) -1, + raid_disks, data_disks, + &dd_idx, &pd_idx, conf); + if (last_sector >= (mddev->size<<1)) + last_sector = (mddev->size<<1)-1; + while (first_sector <= last_sector) { + pd_idx = stripe_to_pdidx(first_sector, conf, conf->previous_raid_disks); + sh = get_active_stripe(conf, first_sector, + conf->previous_raid_disks, pd_idx, 0); + set_bit(STRIPE_EXPAND_SOURCE, &sh->state); + set_bit(STRIPE_HANDLE, &sh->state); + release_stripe(sh); + first_sector += STRIPE_SECTORS; + } + return conf->chunk_size>>9; +} + +/* FIXME go_faster isn't used */ +static inline sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, int go_faster) +{ + raid5_conf_t *conf = (raid5_conf_t *) mddev->private; + struct stripe_head *sh; + int pd_idx; int raid_disks = conf->raid_disks; - int data_disks = raid_disks-1; + int data_disks = raid_disks - conf->max_degraded; sector_t max_sector = mddev->size << 1; int sync_blocks; + int still_degraded = 0; + int i; if (sector_nr >= max_sector) { /* just being told to finish up .. nothing much to do */ @@ -1880,134 +2844,22 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i if (mddev->curr_resync < max_sector) /* aborted */ bitmap_end_sync(mddev->bitmap, mddev->curr_resync, &sync_blocks, 1); - else /* compelted sync */ + else /* completed sync */ conf->fullsync = 0; bitmap_close_sync(mddev->bitmap); return 0; } - if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery)) { - /* reshaping is quite different to recovery/resync so it is - * handled quite separately ... here. - * - * On each call to sync_request, we gather one chunk worth of - * destination stripes and flag them as expanding. - * Then we find all the source stripes and request reads. - * As the reads complete, handle_stripe will copy the data - * into the destination stripe and release that stripe. - */ - int i; - int dd_idx; - sector_t writepos, safepos, gap; - - if (sector_nr == 0 && - conf->expand_progress != 0) { - /* restarting in the middle, skip the initial sectors */ - sector_nr = conf->expand_progress; - sector_div(sector_nr, conf->raid_disks-1); - *skipped = 1; - return sector_nr; - } - - /* we update the metadata when there is more than 3Meg - * in the block range (that is rather arbitrary, should - * probably be time based) or when the data about to be - * copied would over-write the source of the data at - * the front of the range. - * i.e. one new_stripe forward from expand_progress new_maps - * to after where expand_lo old_maps to - */ - writepos = conf->expand_progress + - conf->chunk_size/512*(conf->raid_disks-1); - sector_div(writepos, conf->raid_disks-1); - safepos = conf->expand_lo; - sector_div(safepos, conf->previous_raid_disks-1); - gap = conf->expand_progress - conf->expand_lo; - - if (writepos >= safepos || - gap > (conf->raid_disks-1)*3000*2 /*3Meg*/) { - /* Cannot proceed until we've updated the superblock... */ - wait_event(conf->wait_for_overlap, - atomic_read(&conf->reshape_stripes)==0); - mddev->reshape_position = conf->expand_progress; - mddev->sb_dirty = 1; - md_wakeup_thread(mddev->thread); - wait_event(mddev->sb_wait, mddev->sb_dirty == 0 || - kthread_should_stop()); - spin_lock_irq(&conf->device_lock); - conf->expand_lo = mddev->reshape_position; - spin_unlock_irq(&conf->device_lock); - wake_up(&conf->wait_for_overlap); - } - - for (i=0; i < conf->chunk_size/512; i+= STRIPE_SECTORS) { - int j; - int skipped = 0; - pd_idx = stripe_to_pdidx(sector_nr+i, conf, conf->raid_disks); - sh = get_active_stripe(conf, sector_nr+i, - conf->raid_disks, pd_idx, 0); - set_bit(STRIPE_EXPANDING, &sh->state); - atomic_inc(&conf->reshape_stripes); - /* If any of this stripe is beyond the end of the old - * array, then we need to zero those blocks - */ - for (j=sh->disks; j--;) { - sector_t s; - if (j == sh->pd_idx) - continue; - s = compute_blocknr(sh, j); - if (s < (mddev->array_size<<1)) { - skipped = 1; - continue; - } - memset(page_address(sh->dev[j].page), 0, STRIPE_SIZE); - set_bit(R5_Expanded, &sh->dev[j].flags); - set_bit(R5_UPTODATE, &sh->dev[j].flags); - } - if (!skipped) { - set_bit(STRIPE_EXPAND_READY, &sh->state); - set_bit(STRIPE_HANDLE, &sh->state); - } - release_stripe(sh); - } - spin_lock_irq(&conf->device_lock); - conf->expand_progress = (sector_nr + i)*(conf->raid_disks-1); - spin_unlock_irq(&conf->device_lock); - /* Ok, those stripe are ready. We can start scheduling - * reads on the source stripes. - * The source stripes are determined by mapping the first and last - * block on the destination stripes. - */ - raid_disks = conf->previous_raid_disks; - data_disks = raid_disks - 1; - first_sector = - raid5_compute_sector(sector_nr*(conf->raid_disks-1), - raid_disks, data_disks, - &dd_idx, &pd_idx, conf); - last_sector = - raid5_compute_sector((sector_nr+conf->chunk_size/512) - *(conf->raid_disks-1) -1, - raid_disks, data_disks, - &dd_idx, &pd_idx, conf); - if (last_sector >= (mddev->size<<1)) - last_sector = (mddev->size<<1)-1; - while (first_sector <= last_sector) { - pd_idx = stripe_to_pdidx(first_sector, conf, conf->previous_raid_disks); - sh = get_active_stripe(conf, first_sector, - conf->previous_raid_disks, pd_idx, 0); - set_bit(STRIPE_EXPAND_SOURCE, &sh->state); - set_bit(STRIPE_HANDLE, &sh->state); - release_stripe(sh); - first_sector += STRIPE_SECTORS; - } - return conf->chunk_size>>9; - } - /* if there is 1 or more failed drives and we are trying + if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery)) + return reshape_request(mddev, sector_nr, skipped); + + /* if there is too many failed drives and we are trying * to resync, then assert that we are finished, because there is * nothing we can do. */ - if (mddev->degraded >= 1 && test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) { + if (mddev->degraded >= conf->max_degraded && + test_bit(MD_RECOVERY_SYNC, &mddev->recovery)) { sector_t rv = (mddev->size << 1) - sector_nr; *skipped = 1; return rv; @@ -2026,17 +2878,26 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i if (sh == NULL) { sh = get_active_stripe(conf, sector_nr, raid_disks, pd_idx, 0); /* make sure we don't swamp the stripe cache if someone else - * is trying to get access + * is trying to get access */ schedule_timeout_uninterruptible(1); } - bitmap_start_sync(mddev->bitmap, sector_nr, &sync_blocks, 0); - spin_lock(&sh->lock); + /* Need to check if array will still be degraded after recovery/resync + * We don't need to check the 'failed' flag as when that gets set, + * recovery aborts. + */ + for (i=0; i<mddev->raid_disks; i++) + if (conf->disks[i].rdev == NULL) + still_degraded = 1; + + bitmap_start_sync(mddev->bitmap, sector_nr, &sync_blocks, still_degraded); + + spin_lock(&sh->lock); set_bit(STRIPE_SYNCING, &sh->state); clear_bit(STRIPE_INSYNC, &sh->state); spin_unlock(&sh->lock); - handle_stripe(sh); + handle_stripe(sh, NULL); release_stripe(sh); return STRIPE_SECTORS; @@ -2091,7 +2952,7 @@ static void raid5d (mddev_t *mddev) spin_unlock_irq(&conf->device_lock); handled++; - handle_stripe(sh); + handle_stripe(sh, conf->spare_page); release_stripe(sh); spin_lock_irq(&conf->device_lock); @@ -2181,8 +3042,8 @@ static int run(mddev_t *mddev) struct disk_info *disk; struct list_head *tmp; - if (mddev->level != 5 && mddev->level != 4) { - printk(KERN_ERR "raid5: %s: raid level not set to 4/5 (%d)\n", + if (mddev->level != 5 && mddev->level != 4 && mddev->level != 6) { + printk(KERN_ERR "raid5: %s: raid level not set to 4/5/6 (%d)\n", mdname(mddev), mddev->level); return -EIO; } @@ -2251,6 +3112,11 @@ static int run(mddev_t *mddev) if ((conf->stripe_hashtbl = kzalloc(PAGE_SIZE, GFP_KERNEL)) == NULL) goto abort; + if (mddev->level == 6) { + conf->spare_page = alloc_page(GFP_KERNEL); + if (!conf->spare_page) + goto abort; + } spin_lock_init(&conf->device_lock); init_waitqueue_head(&conf->wait_for_stripe); init_waitqueue_head(&conf->wait_for_overlap); @@ -2282,12 +3148,16 @@ static int run(mddev_t *mddev) } /* - * 0 for a fully functional array, 1 for a degraded array. + * 0 for a fully functional array, 1 or 2 for a degraded array. */ mddev->degraded = conf->failed_disks = conf->raid_disks - conf->working_disks; conf->mddev = mddev; conf->chunk_size = mddev->chunk_size; conf->level = mddev->level; + if (conf->level == 6) + conf->max_degraded = 2; + else + conf->max_degraded = 1; conf->algorithm = mddev->layout; conf->max_nr_stripes = NR_STRIPES; conf->expand_progress = mddev->reshape_position; @@ -2296,6 +3166,11 @@ static int run(mddev_t *mddev) mddev->size &= ~(mddev->chunk_size/1024 -1); mddev->resync_max_sectors = mddev->size << 1; + if (conf->level == 6 && conf->raid_disks < 4) { + printk(KERN_ERR "raid6: not enough configured devices for %s (%d, minimum 4)\n", + mdname(mddev), conf->raid_disks); + goto abort; + } if (!conf->chunk_size || conf->chunk_size % 4) { printk(KERN_ERR "raid5: invalid chunk size %d for %s\n", conf->chunk_size, mdname(mddev)); @@ -2307,14 +3182,14 @@ static int run(mddev_t *mddev) conf->algorithm, mdname(mddev)); goto abort; } - if (mddev->degraded > 1) { + if (mddev->degraded > conf->max_degraded) { printk(KERN_ERR "raid5: not enough operational devices for %s" " (%d/%d failed)\n", mdname(mddev), conf->failed_disks, conf->raid_disks); goto abort; } - if (mddev->degraded == 1 && + if (mddev->degraded > 0 && mddev->recovery_cp != MaxSector) { if (mddev->ok_start_degraded) printk(KERN_WARNING @@ -2379,11 +3254,12 @@ static int run(mddev_t *mddev) } /* read-ahead size must cover two whole stripes, which is - * 2 * (n-1) * chunksize where 'n' is the number of raid devices + * 2 * (datadisks) * chunksize where 'n' is the number of raid devices */ { - int stripe = (mddev->raid_disks-1) * mddev->chunk_size - / PAGE_SIZE; + int data_disks = conf->previous_raid_disks - conf->max_degraded; + int stripe = data_disks * + (mddev->chunk_size / PAGE_SIZE); if (mddev->queue->backing_dev_info.ra_pages < 2 * stripe) mddev->queue->backing_dev_info.ra_pages = 2 * stripe; } @@ -2393,12 +3269,14 @@ static int run(mddev_t *mddev) mddev->queue->unplug_fn = raid5_unplug_device; mddev->queue->issue_flush_fn = raid5_issue_flush; - mddev->array_size = mddev->size * (conf->previous_raid_disks - 1); + mddev->array_size = mddev->size * (conf->previous_raid_disks - + conf->max_degraded); return 0; abort: if (conf) { print_raid5_conf(conf); + safe_put_page(conf->spare_page); kfree(conf->disks); kfree(conf->stripe_hashtbl); kfree(conf); @@ -2427,23 +3305,23 @@ static int stop(mddev_t *mddev) } #if RAID5_DEBUG -static void print_sh (struct stripe_head *sh) +static void print_sh (struct seq_file *seq, struct stripe_head *sh) { int i; - printk("sh %llu, pd_idx %d, state %ld.\n", - (unsigned long long)sh->sector, sh->pd_idx, sh->state); - printk("sh %llu, count %d.\n", - (unsigned long long)sh->sector, atomic_read(&sh->count)); - printk("sh %llu, ", (unsigned long long)sh->sector); + seq_printf(seq, "sh %llu, pd_idx %d, state %ld.\n", + (unsigned long long)sh->sector, sh->pd_idx, sh->state); + seq_printf(seq, "sh %llu, count %d.\n", + (unsigned long long)sh->sector, atomic_read(&sh->count)); + seq_printf(seq, "sh %llu, ", (unsigned long long)sh->sector); for (i = 0; i < sh->disks; i++) { - printk("(cache%d: %p %ld) ", - i, sh->dev[i].page, sh->dev[i].flags); + seq_printf(seq, "(cache%d: %p %ld) ", + i, sh->dev[i].page, sh->dev[i].flags); } - printk("\n"); + seq_printf(seq, "\n"); } -static void printall (raid5_conf_t *conf) +static void printall (struct seq_file *seq, raid5_conf_t *conf) { struct stripe_head *sh; struct hlist_node *hn; @@ -2454,7 +3332,7 @@ static void printall (raid5_conf_t *conf) hlist_for_each_entry(sh, hn, &conf->stripe_hashtbl[i], hash) { if (sh->raid_conf != conf) continue; - print_sh(sh); + print_sh(seq, sh); } } spin_unlock_irq(&conf->device_lock); @@ -2474,9 +3352,8 @@ static void status (struct seq_file *seq, mddev_t *mddev) test_bit(In_sync, &conf->disks[i].rdev->flags) ? "U" : "_"); seq_printf (seq, "]"); #if RAID5_DEBUG -#define D(x) \ - seq_printf (seq, "<"#x":%d>", atomic_read(&conf->x)) - printall(conf); + seq_printf (seq, "\n"); + printall(seq, conf); #endif } @@ -2560,14 +3437,20 @@ static int raid5_add_disk(mddev_t *mddev, mdk_rdev_t *rdev) int disk; struct disk_info *p; - if (mddev->degraded > 1) + if (mddev->degraded > conf->max_degraded) /* no point adding a device */ return 0; /* - * find the disk ... + * find the disk ... but prefer rdev->saved_raid_disk + * if possible. */ - for (disk=0; disk < conf->raid_disks; disk++) + if (rdev->saved_raid_disk >= 0 && + conf->disks[rdev->saved_raid_disk].rdev == NULL) + disk = rdev->saved_raid_disk; + else + disk = 0; + for ( ; disk < conf->raid_disks; disk++) if ((p=conf->disks + disk)->rdev == NULL) { clear_bit(In_sync, &rdev->flags); rdev->raid_disk = disk; @@ -2590,8 +3473,10 @@ static int raid5_resize(mddev_t *mddev, sector_t sectors) * any io in the removed space completes, but it hardly seems * worth it. */ + raid5_conf_t *conf = mddev_to_conf(mddev); + sectors &= ~((sector_t)mddev->chunk_size/512 - 1); - mddev->array_size = (sectors * (mddev->raid_disks-1))>>1; + mddev->array_size = (sectors * (mddev->raid_disks-conf->max_degraded))>>1; set_capacity(mddev->gendisk, mddev->array_size << 1); mddev->changed = 1; if (sectors/2 > mddev->size && mddev->recovery_cp == MaxSector) { @@ -2680,6 +3565,7 @@ static int raid5_start_reshape(mddev_t *mddev) set_bit(In_sync, &rdev->flags); conf->working_disks++; added_devices++; + rdev->recovery_offset = 0; sprintf(nm, "rd%d", rdev->raid_disk); sysfs_create_link(&mddev->kobj, &rdev->kobj, nm); } else @@ -2731,6 +3617,17 @@ static void end_reshape(raid5_conf_t *conf) conf->expand_progress = MaxSector; spin_unlock_irq(&conf->device_lock); conf->mddev->reshape_position = MaxSector; + + /* read-ahead size must cover two whole stripes, which is + * 2 * (datadisks) * chunksize where 'n' is the number of raid devices + */ + { + int data_disks = conf->previous_raid_disks - conf->max_degraded; + int stripe = data_disks * + (conf->mddev->chunk_size / PAGE_SIZE); + if (conf->mddev->queue->backing_dev_info.ra_pages < 2 * stripe) + conf->mddev->queue->backing_dev_info.ra_pages = 2 * stripe; + } } } @@ -2762,6 +3659,23 @@ static void raid5_quiesce(mddev_t *mddev, int state) } } +static struct mdk_personality raid6_personality = +{ + .name = "raid6", + .level = 6, + .owner = THIS_MODULE, + .make_request = make_request, + .run = run, + .stop = stop, + .status = status, + .error_handler = error, + .hot_add_disk = raid5_add_disk, + .hot_remove_disk= raid5_remove_disk, + .spare_active = raid5_spare_active, + .sync_request = sync_request, + .resize = raid5_resize, + .quiesce = raid5_quiesce, +}; static struct mdk_personality raid5_personality = { .name = "raid5", @@ -2804,6 +3718,12 @@ static struct mdk_personality raid4_personality = static int __init raid5_init(void) { + int e; + + e = raid6_select_algo(); + if ( e ) + return e; + register_md_personality(&raid6_personality); register_md_personality(&raid5_personality); register_md_personality(&raid4_personality); return 0; @@ -2811,6 +3731,7 @@ static int __init raid5_init(void) static void raid5_exit(void) { + unregister_md_personality(&raid6_personality); unregister_md_personality(&raid5_personality); unregister_md_personality(&raid4_personality); } @@ -2823,3 +3744,10 @@ MODULE_ALIAS("md-raid5"); MODULE_ALIAS("md-raid4"); MODULE_ALIAS("md-level-5"); MODULE_ALIAS("md-level-4"); +MODULE_ALIAS("md-personality-8"); /* RAID6 */ +MODULE_ALIAS("md-raid6"); +MODULE_ALIAS("md-level-6"); + +/* This used to be two separate modules, they were: */ +MODULE_ALIAS("raid5"); +MODULE_ALIAS("raid6"); |