diff options
Diffstat (limited to 'fs/btrfs/scrub.c')
| -rw-r--r-- | fs/btrfs/scrub.c | 3178 |
1 files changed, 1808 insertions, 1370 deletions
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index 61b37c56a7fb..196c4c6ed1ed 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -20,6 +20,7 @@ #include "rcu-string.h" #include "raid56.h" #include "block-group.h" +#include "zoned.h" /* * This is only the first step towards a full-features scrub. It reads all @@ -38,44 +39,39 @@ struct scrub_block; struct scrub_ctx; /* - * the following three values only influence the performance. + * The following three values only influence the performance. + * * The last one configures the number of parallel and outstanding I/O - * operations. The first two values configure an upper limit for the number + * operations. The first one configures an upper limit for the number * of (dynamically allocated) pages that are added to a bio. */ -#define SCRUB_PAGES_PER_RD_BIO 32 /* 128k per bio */ -#define SCRUB_PAGES_PER_WR_BIO 32 /* 128k per bio */ -#define SCRUB_BIOS_PER_SCTX 64 /* 8MB per device in flight */ +#define SCRUB_SECTORS_PER_BIO 32 /* 128KiB per bio for 4KiB pages */ +#define SCRUB_BIOS_PER_SCTX 64 /* 8MiB per device in flight for 4KiB pages */ /* - * the following value times PAGE_SIZE needs to be large enough to match the + * The following value times PAGE_SIZE needs to be large enough to match the * largest node/leaf/sector size that shall be supported. - * Values larger than BTRFS_STRIPE_LEN are not supported. */ -#define SCRUB_MAX_PAGES_PER_BLOCK 16 /* 64k per node/leaf/sector */ +#define SCRUB_MAX_SECTORS_PER_BLOCK (BTRFS_MAX_METADATA_BLOCKSIZE / SZ_4K) + +#define SCRUB_MAX_PAGES (DIV_ROUND_UP(BTRFS_MAX_METADATA_BLOCKSIZE, PAGE_SIZE)) struct scrub_recover { refcount_t refs; - struct btrfs_bio *bbio; + struct btrfs_io_context *bioc; u64 map_length; }; -struct scrub_page { +struct scrub_sector { struct scrub_block *sblock; - struct page *page; - struct btrfs_device *dev; struct list_head list; u64 flags; /* extent flags */ u64 generation; - u64 logical; - u64 physical; - u64 physical_for_dev_replace; + /* Offset in bytes to @sblock. */ + u32 offset; atomic_t refs; - struct { - unsigned int mirror_num:8; - unsigned int have_csum:1; - unsigned int io_error:1; - }; + unsigned int have_csum:1; + unsigned int io_error:1; u8 csum[BTRFS_CSUM_SIZE]; struct scrub_recover *recover; @@ -89,20 +85,30 @@ struct scrub_bio { blk_status_t status; u64 logical; u64 physical; -#if SCRUB_PAGES_PER_WR_BIO >= SCRUB_PAGES_PER_RD_BIO - struct scrub_page *pagev[SCRUB_PAGES_PER_WR_BIO]; -#else - struct scrub_page *pagev[SCRUB_PAGES_PER_RD_BIO]; -#endif - int page_count; + struct scrub_sector *sectors[SCRUB_SECTORS_PER_BIO]; + int sector_count; int next_free; - struct btrfs_work work; + struct work_struct work; }; struct scrub_block { - struct scrub_page *pagev[SCRUB_MAX_PAGES_PER_BLOCK]; - int page_count; - atomic_t outstanding_pages; + /* + * Each page will have its page::private used to record the logical + * bytenr. + */ + struct page *pages[SCRUB_MAX_PAGES]; + struct scrub_sector *sectors[SCRUB_MAX_SECTORS_PER_BLOCK]; + struct btrfs_device *dev; + /* Logical bytenr of the sblock */ + u64 logical; + u64 physical; + u64 physical_for_dev_replace; + /* Length of sblock in bytes */ + u32 len; + int sector_count; + int mirror_num; + + atomic_t outstanding_sectors; refcount_t refs; /* free mem on transition to zero */ struct scrub_ctx *sctx; struct scrub_parity *sparity; @@ -116,7 +122,7 @@ struct scrub_block { /* It is for the data with checksum */ unsigned int data_corrected:1; }; - struct btrfs_work work; + struct work_struct work; }; /* Used for the chunks with parity stripe such RAID5/6 */ @@ -131,25 +137,23 @@ struct scrub_parity { int nsectors; - u64 stripe_len; + u32 stripe_len; refcount_t refs; - struct list_head spages; + struct list_head sectors_list; /* Work of parity check and repair */ - struct btrfs_work work; + struct work_struct work; /* Mark the parity blocks which have data */ - unsigned long *dbitmap; + unsigned long dbitmap; /* * Mark the parity blocks which have data, but errors happen when * read data or check data */ - unsigned long *ebitmap; - - unsigned long bitmap[0]; + unsigned long ebitmap; }; struct scrub_ctx { @@ -161,17 +165,20 @@ struct scrub_ctx { atomic_t workers_pending; spinlock_t list_lock; wait_queue_head_t list_wait; - u16 csum_size; struct list_head csum_list; atomic_t cancel_req; int readonly; - int pages_per_rd_bio; + int sectors_per_bio; + + /* State of IO submission throttling affecting the associated device */ + ktime_t throttle_deadline; + u64 throttle_sent; int is_dev_replace; + u64 write_pointer; struct scrub_bio *wr_curr_bio; struct mutex wr_lock; - int pages_per_wr_bio; /* <= SCRUB_PAGES_PER_WR_BIO */ struct btrfs_device *wr_tgtdev; bool flush_all_writes; @@ -207,59 +214,217 @@ struct full_stripe_lock { struct mutex mutex; }; -static void scrub_pending_bio_inc(struct scrub_ctx *sctx); -static void scrub_pending_bio_dec(struct scrub_ctx *sctx); -static int scrub_handle_errored_block(struct scrub_block *sblock_to_check); +#ifndef CONFIG_64BIT +/* This structure is for archtectures whose (void *) is smaller than u64 */ +struct scrub_page_private { + u64 logical; +}; +#endif + +static int attach_scrub_page_private(struct page *page, u64 logical) +{ +#ifdef CONFIG_64BIT + attach_page_private(page, (void *)logical); + return 0; +#else + struct scrub_page_private *spp; + + spp = kmalloc(sizeof(*spp), GFP_KERNEL); + if (!spp) + return -ENOMEM; + spp->logical = logical; + attach_page_private(page, (void *)spp); + return 0; +#endif +} + +static void detach_scrub_page_private(struct page *page) +{ +#ifdef CONFIG_64BIT + detach_page_private(page); + return; +#else + struct scrub_page_private *spp; + + spp = detach_page_private(page); + kfree(spp); + return; +#endif +} + +static struct scrub_block *alloc_scrub_block(struct scrub_ctx *sctx, + struct btrfs_device *dev, + u64 logical, u64 physical, + u64 physical_for_dev_replace, + int mirror_num) +{ + struct scrub_block *sblock; + + sblock = kzalloc(sizeof(*sblock), GFP_KERNEL); + if (!sblock) + return NULL; + refcount_set(&sblock->refs, 1); + sblock->sctx = sctx; + sblock->logical = logical; + sblock->physical = physical; + sblock->physical_for_dev_replace = physical_for_dev_replace; + sblock->dev = dev; + sblock->mirror_num = mirror_num; + sblock->no_io_error_seen = 1; + /* + * Scrub_block::pages will be allocated at alloc_scrub_sector() when + * the corresponding page is not allocated. + */ + return sblock; +} + +/* + * Allocate a new scrub sector and attach it to @sblock. + * + * Will also allocate new pages for @sblock if needed. + */ +static struct scrub_sector *alloc_scrub_sector(struct scrub_block *sblock, + u64 logical, gfp_t gfp) +{ + const pgoff_t page_index = (logical - sblock->logical) >> PAGE_SHIFT; + struct scrub_sector *ssector; + + /* We must never have scrub_block exceed U32_MAX in size. */ + ASSERT(logical - sblock->logical < U32_MAX); + + ssector = kzalloc(sizeof(*ssector), gfp); + if (!ssector) + return NULL; + + /* Allocate a new page if the slot is not allocated */ + if (!sblock->pages[page_index]) { + int ret; + + sblock->pages[page_index] = alloc_page(gfp); + if (!sblock->pages[page_index]) { + kfree(ssector); + return NULL; + } + ret = attach_scrub_page_private(sblock->pages[page_index], + sblock->logical + (page_index << PAGE_SHIFT)); + if (ret < 0) { + kfree(ssector); + __free_page(sblock->pages[page_index]); + sblock->pages[page_index] = NULL; + return NULL; + } + } + + atomic_set(&ssector->refs, 1); + ssector->sblock = sblock; + /* The sector to be added should not be used */ + ASSERT(sblock->sectors[sblock->sector_count] == NULL); + ssector->offset = logical - sblock->logical; + + /* The sector count must be smaller than the limit */ + ASSERT(sblock->sector_count < SCRUB_MAX_SECTORS_PER_BLOCK); + + sblock->sectors[sblock->sector_count] = ssector; + sblock->sector_count++; + sblock->len += sblock->sctx->fs_info->sectorsize; + + return ssector; +} + +static struct page *scrub_sector_get_page(struct scrub_sector *ssector) +{ + struct scrub_block *sblock = ssector->sblock; + pgoff_t index; + /* + * When calling this function, ssector must be alreaday attached to the + * parent sblock. + */ + ASSERT(sblock); + + /* The range should be inside the sblock range */ + ASSERT(ssector->offset < sblock->len); + + index = ssector->offset >> PAGE_SHIFT; + ASSERT(index < SCRUB_MAX_PAGES); + ASSERT(sblock->pages[index]); + ASSERT(PagePrivate(sblock->pages[index])); + return sblock->pages[index]; +} + +static unsigned int scrub_sector_get_page_offset(struct scrub_sector *ssector) +{ + struct scrub_block *sblock = ssector->sblock; + + /* + * When calling this function, ssector must be already attached to the + * parent sblock. + */ + ASSERT(sblock); + + /* The range should be inside the sblock range */ + ASSERT(ssector->offset < sblock->len); + + return offset_in_page(ssector->offset); +} + +static char *scrub_sector_get_kaddr(struct scrub_sector *ssector) +{ + return page_address(scrub_sector_get_page(ssector)) + + scrub_sector_get_page_offset(ssector); +} + +static int bio_add_scrub_sector(struct bio *bio, struct scrub_sector *ssector, + unsigned int len) +{ + return bio_add_page(bio, scrub_sector_get_page(ssector), len, + scrub_sector_get_page_offset(ssector)); +} + static int scrub_setup_recheck_block(struct scrub_block *original_sblock, - struct scrub_block *sblocks_for_recheck); + struct scrub_block *sblocks_for_recheck[]); static void scrub_recheck_block(struct btrfs_fs_info *fs_info, struct scrub_block *sblock, int retry_failed_mirror); static void scrub_recheck_block_checksum(struct scrub_block *sblock); static int scrub_repair_block_from_good_copy(struct scrub_block *sblock_bad, struct scrub_block *sblock_good); -static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad, +static int scrub_repair_sector_from_good_copy(struct scrub_block *sblock_bad, struct scrub_block *sblock_good, - int page_num, int force_write); + int sector_num, int force_write); static void scrub_write_block_to_dev_replace(struct scrub_block *sblock); -static int scrub_write_page_to_dev_replace(struct scrub_block *sblock, - int page_num); +static int scrub_write_sector_to_dev_replace(struct scrub_block *sblock, + int sector_num); static int scrub_checksum_data(struct scrub_block *sblock); static int scrub_checksum_tree_block(struct scrub_block *sblock); static int scrub_checksum_super(struct scrub_block *sblock); -static void scrub_block_get(struct scrub_block *sblock); static void scrub_block_put(struct scrub_block *sblock); -static void scrub_page_get(struct scrub_page *spage); -static void scrub_page_put(struct scrub_page *spage); +static void scrub_sector_get(struct scrub_sector *sector); +static void scrub_sector_put(struct scrub_sector *sector); static void scrub_parity_get(struct scrub_parity *sparity); static void scrub_parity_put(struct scrub_parity *sparity); -static int scrub_add_page_to_rd_bio(struct scrub_ctx *sctx, - struct scrub_page *spage); -static int scrub_pages(struct scrub_ctx *sctx, u64 logical, u64 len, - u64 physical, struct btrfs_device *dev, u64 flags, - u64 gen, int mirror_num, u8 *csum, int force, - u64 physical_for_dev_replace); +static int scrub_sectors(struct scrub_ctx *sctx, u64 logical, u32 len, + u64 physical, struct btrfs_device *dev, u64 flags, + u64 gen, int mirror_num, u8 *csum, + u64 physical_for_dev_replace); static void scrub_bio_end_io(struct bio *bio); -static void scrub_bio_end_io_worker(struct btrfs_work *work); +static void scrub_bio_end_io_worker(struct work_struct *work); static void scrub_block_complete(struct scrub_block *sblock); -static void scrub_remap_extent(struct btrfs_fs_info *fs_info, - u64 extent_logical, u64 extent_len, - u64 *extent_physical, - struct btrfs_device **extent_dev, - int *extent_mirror_num); -static int scrub_add_page_to_wr_bio(struct scrub_ctx *sctx, - struct scrub_page *spage); +static void scrub_find_good_copy(struct btrfs_fs_info *fs_info, + u64 extent_logical, u32 extent_len, + u64 *extent_physical, + struct btrfs_device **extent_dev, + int *extent_mirror_num); +static int scrub_add_sector_to_wr_bio(struct scrub_ctx *sctx, + struct scrub_sector *sector); static void scrub_wr_submit(struct scrub_ctx *sctx); static void scrub_wr_bio_end_io(struct bio *bio); -static void scrub_wr_bio_end_io_worker(struct btrfs_work *work); -static void __scrub_blocked_if_needed(struct btrfs_fs_info *fs_info); -static void scrub_blocked_if_needed(struct btrfs_fs_info *fs_info); +static void scrub_wr_bio_end_io_worker(struct work_struct *work); static void scrub_put_ctx(struct scrub_ctx *sctx); -static inline int scrub_is_page_on_raid56(struct scrub_page *page) +static inline int scrub_is_page_on_raid56(struct scrub_sector *sector) { - return page->recover && - (page->recover->bbio->map_type & BTRFS_BLOCK_GROUP_RAID56_MASK); + return sector->recover && + (sector->recover->bioc->map_type & BTRFS_BLOCK_GROUP_RAID56_MASK); } static void scrub_pending_bio_inc(struct scrub_ctx *sctx) @@ -546,10 +711,8 @@ static noinline_for_stack void scrub_free_ctx(struct scrub_ctx *sctx) if (sctx->curr != -1) { struct scrub_bio *sbio = sctx->bios[sctx->curr]; - for (i = 0; i < sbio->page_count; i++) { - WARN_ON(!sbio->pagev[i]->page); - scrub_block_put(sbio->pagev[i]->sblock); - } + for (i = 0; i < sbio->sector_count; i++) + scrub_block_put(sbio->sectors[i]->sblock); bio_put(sbio->bio); } @@ -583,7 +746,7 @@ static noinline_for_stack struct scrub_ctx *scrub_setup_ctx( goto nomem; refcount_set(&sctx->refs, 1); sctx->is_dev_replace = is_dev_replace; - sctx->pages_per_rd_bio = SCRUB_PAGES_PER_RD_BIO; + sctx->sectors_per_bio = SCRUB_SECTORS_PER_BIO; sctx->curr = -1; sctx->fs_info = fs_info; INIT_LIST_HEAD(&sctx->csum_list); @@ -597,9 +760,8 @@ static noinline_for_stack struct scrub_ctx *scrub_setup_ctx( sbio->index = i; sbio->sctx = sctx; - sbio->page_count = 0; - btrfs_init_work(&sbio->work, scrub_bio_end_io_worker, NULL, - NULL); + sbio->sector_count = 0; + INIT_WORK(&sbio->work, scrub_bio_end_io_worker); if (i != SCRUB_BIOS_PER_SCTX - 1) sctx->bios[i]->next_free = i + 1; @@ -610,18 +772,17 @@ static noinline_for_stack struct scrub_ctx *scrub_setup_ctx( atomic_set(&sctx->bios_in_flight, 0); atomic_set(&sctx->workers_pending, 0); atomic_set(&sctx->cancel_req, 0); - sctx->csum_size = btrfs_super_csum_size(fs_info->super_copy); spin_lock_init(&sctx->list_lock); spin_lock_init(&sctx->stat_lock); init_waitqueue_head(&sctx->list_wait); + sctx->throttle_deadline = 0; WARN_ON(sctx->wr_curr_bio != NULL); mutex_init(&sctx->wr_lock); sctx->wr_curr_bio = NULL; if (is_dev_replace) { WARN_ON(!fs_info->dev_replace.tgtdev); - sctx->pages_per_wr_bio = SCRUB_PAGES_PER_WR_BIO; sctx->wr_tgtdev = fs_info->dev_replace.tgtdev; sctx->flush_all_writes = false; } @@ -636,7 +797,6 @@ nomem: static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, void *warn_ctx) { - u64 isize; u32 nlink; int ret; int i; @@ -647,13 +807,9 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, struct btrfs_fs_info *fs_info = swarn->dev->fs_info; struct inode_fs_paths *ipath = NULL; struct btrfs_root *local_root; - struct btrfs_key root_key; struct btrfs_key key; - root_key.objectid = root; - root_key.type = BTRFS_ROOT_ITEM_KEY; - root_key.offset = (u64)-1; - local_root = btrfs_read_fs_root_no_name(fs_info, &root_key); + local_root = btrfs_get_fs_root(fs_info, root, true); if (IS_ERR(local_root)) { ret = PTR_ERR(local_root); goto err; @@ -668,6 +824,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, ret = btrfs_search_slot(NULL, local_root, &key, swarn->path, 0, 0); if (ret) { + btrfs_put_root(local_root); btrfs_release_path(swarn->path); goto err; } @@ -675,7 +832,6 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, eb = swarn->path->nodes[0]; inode_item = btrfs_item_ptr(eb, swarn->path->slots[0], struct btrfs_inode_item); - isize = btrfs_inode_size(eb, inode_item); nlink = btrfs_inode_nlink(eb, inode_item); btrfs_release_path(swarn->path); @@ -688,6 +844,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, ipath = init_ipath(4096, local_root, swarn->path); memalloc_nofs_restore(nofs_flag); if (IS_ERR(ipath)) { + btrfs_put_root(local_root); ret = PTR_ERR(ipath); ipath = NULL; goto err; @@ -703,14 +860,15 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, */ for (i = 0; i < ipath->fspath->elem_cnt; ++i) btrfs_warn_in_rcu(fs_info, -"%s at logical %llu on dev %s, physical %llu, root %llu, inode %llu, offset %llu, length %llu, links %u (path: %s)", +"%s at logical %llu on dev %s, physical %llu, root %llu, inode %llu, offset %llu, length %u, links %u (path: %s)", swarn->errstr, swarn->logical, rcu_str_deref(swarn->dev->name), swarn->physical, root, inum, offset, - min(isize - offset, (u64)PAGE_SIZE), nlink, + fs_info->sectorsize, nlink, (char *)(unsigned long)ipath->fspath->val[i]); + btrfs_put_root(local_root); free_ipath(ipath); return 0; @@ -743,16 +901,23 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock) u8 ref_level = 0; int ret; - WARN_ON(sblock->page_count < 1); - dev = sblock->pagev[0]->dev; + WARN_ON(sblock->sector_count < 1); + dev = sblock->dev; fs_info = sblock->sctx->fs_info; + /* Super block error, no need to search extent tree. */ + if (sblock->sectors[0]->flags & BTRFS_EXTENT_FLAG_SUPER) { + btrfs_warn_in_rcu(fs_info, "%s on device %s, physical %llu", + errstr, rcu_str_deref(dev->name), + sblock->physical); + return; + } path = btrfs_alloc_path(); if (!path) return; - swarn.physical = sblock->pagev[0]->physical; - swarn.logical = sblock->pagev[0]->logical; + swarn.physical = sblock->physical; + swarn.logical = sblock->logical; swarn.errstr = errstr; swarn.dev = NULL; @@ -766,7 +931,7 @@ static void scrub_print_warning(const char *errstr, struct scrub_block *sblock) eb = path->nodes[0]; ei = btrfs_item_ptr(eb, path->slots[0], struct btrfs_extent_item); - item_size = btrfs_item_size_nr(eb, path->slots[0]); + item_size = btrfs_item_size(eb, path->slots[0]); if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) { do { @@ -806,15 +971,15 @@ static inline void scrub_put_recover(struct btrfs_fs_info *fs_info, { if (refcount_dec_and_test(&recover->refs)) { btrfs_bio_counter_dec(fs_info); - btrfs_put_bbio(recover->bbio); + btrfs_put_bioc(recover->bioc); kfree(recover); } } /* * scrub_handle_errored_block gets called when either verification of the - * pages failed or the bio failed to read, e.g. with EIO. In the latter - * case, this function handles all pages in the bio, even though only one + * sectors failed or the bio failed to read, e.g. with EIO. In the latter + * case, this function handles all sectors in the bio, even though only one * may be bad. * The goal of this function is to repair the errored block by using the * contents of one of the mirrors. @@ -822,43 +987,48 @@ static inline void scrub_put_recover(struct btrfs_fs_info *fs_info, static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) { struct scrub_ctx *sctx = sblock_to_check->sctx; - struct btrfs_device *dev; + struct btrfs_device *dev = sblock_to_check->dev; struct btrfs_fs_info *fs_info; u64 logical; unsigned int failed_mirror_index; unsigned int is_metadata; unsigned int have_csum; - struct scrub_block *sblocks_for_recheck; /* holds one for each mirror */ + /* One scrub_block for each mirror */ + struct scrub_block *sblocks_for_recheck[BTRFS_MAX_MIRRORS] = { 0 }; struct scrub_block *sblock_bad; int ret; int mirror_index; - int page_num; + int sector_num; int success; bool full_stripe_locked; unsigned int nofs_flag; - static DEFINE_RATELIMIT_STATE(_rs, DEFAULT_RATELIMIT_INTERVAL, + static DEFINE_RATELIMIT_STATE(rs, DEFAULT_RATELIMIT_INTERVAL, DEFAULT_RATELIMIT_BURST); - BUG_ON(sblock_to_check->page_count < 1); + BUG_ON(sblock_to_check->sector_count < 1); fs_info = sctx->fs_info; - if (sblock_to_check->pagev[0]->flags & BTRFS_EXTENT_FLAG_SUPER) { + if (sblock_to_check->sectors[0]->flags & BTRFS_EXTENT_FLAG_SUPER) { /* - * if we find an error in a super block, we just report it. + * If we find an error in a super block, we just report it. * They will get written with the next transaction commit * anyway */ + scrub_print_warning("super block error", sblock_to_check); spin_lock(&sctx->stat_lock); ++sctx->stat.super_errors; spin_unlock(&sctx->stat_lock); + btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_CORRUPTION_ERRS); return 0; } - logical = sblock_to_check->pagev[0]->logical; - BUG_ON(sblock_to_check->pagev[0]->mirror_num < 1); - failed_mirror_index = sblock_to_check->pagev[0]->mirror_num - 1; - is_metadata = !(sblock_to_check->pagev[0]->flags & + logical = sblock_to_check->logical; + ASSERT(sblock_to_check->mirror_num); + failed_mirror_index = sblock_to_check->mirror_num - 1; + is_metadata = !(sblock_to_check->sectors[0]->flags & BTRFS_EXTENT_FLAG_DATA); - have_csum = sblock_to_check->pagev[0]->have_csum; - dev = sblock_to_check->pagev[0]->dev; + have_csum = sblock_to_check->sectors[0]->have_csum; + + if (!sctx->is_dev_replace && btrfs_repair_one_zone(fs_info, logical)) + return 0; /* * We must use GFP_NOFS because the scrub task might be waiting for a @@ -866,7 +1036,7 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) * might be waiting the scrub task to pause (which needs to wait for all * the worker tasks to complete before pausing). * We do allocations in the workers through insert_full_stripe_lock() - * and scrub_add_page_to_wr_bio(), which happens down the call chain of + * and scrub_add_sector_to_wr_bio(), which happens down the call chain of * this function. */ nofs_flag = memalloc_nofs_save(); @@ -893,44 +1063,55 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) * read all mirrors one after the other. This includes to * re-read the extent or metadata block that failed (that was * the cause that this fixup code is called) another time, - * page by page this time in order to know which pages + * sector by sector this time in order to know which sectors * caused I/O errors and which ones are good (for all mirrors). * It is the goal to handle the situation when more than one * mirror contains I/O errors, but the errors do not * overlap, i.e. the data can be repaired by selecting the - * pages from those mirrors without I/O error on the - * particular pages. One example (with blocks >= 2 * PAGE_SIZE) - * would be that mirror #1 has an I/O error on the first page, - * the second page is good, and mirror #2 has an I/O error on - * the second page, but the first page is good. - * Then the first page of the first mirror can be repaired by - * taking the first page of the second mirror, and the - * second page of the second mirror can be repaired by - * copying the contents of the 2nd page of the 1st mirror. - * One more note: if the pages of one mirror contain I/O + * sectors from those mirrors without I/O error on the + * particular sectors. One example (with blocks >= 2 * sectorsize) + * would be that mirror #1 has an I/O error on the first sector, + * the second sector is good, and mirror #2 has an I/O error on + * the second sector, but the first sector is good. + * Then the first sector of the first mirror can be repaired by + * taking the first sector of the second mirror, and the + * second sector of the second mirror can be repaired by + * copying the contents of the 2nd sector of the 1st mirror. + * One more note: if the sectors of one mirror contain I/O * errors, the checksum cannot be verified. In order to get * the best data for repairing, the first attempt is to find * a mirror without I/O errors and with a validated checksum. - * Only if this is not possible, the pages are picked from + * Only if this is not possible, the sectors are picked from * mirrors with I/O errors without considering the checksum. * If the latter is the case, at the end, the checksum of the * repaired area is verified in order to correctly maintain * the statistics. */ - - sblocks_for_recheck = kcalloc(BTRFS_MAX_MIRRORS, - sizeof(*sblocks_for_recheck), GFP_KERNEL); - if (!sblocks_for_recheck) { - spin_lock(&sctx->stat_lock); - sctx->stat.malloc_errors++; - sctx->stat.read_errors++; - sctx->stat.uncorrectable_errors++; - spin_unlock(&sctx->stat_lock); - btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_READ_ERRS); - goto out; + for (mirror_index = 0; mirror_index < BTRFS_MAX_MIRRORS; mirror_index++) { + /* + * Note: the two members refs and outstanding_sectors are not + * used in the blocks that are used for the recheck procedure. + * + * But alloc_scrub_block() will initialize sblock::ref anyway, + * so we can use scrub_block_put() to clean them up. + * + * And here we don't setup the physical/dev for the sblock yet, + * they will be correctly initialized in scrub_setup_recheck_block(). + */ + sblocks_for_recheck[mirror_index] = alloc_scrub_block(sctx, NULL, + logical, 0, 0, mirror_index); + if (!sblocks_for_recheck[mirror_index]) { + spin_lock(&sctx->stat_lock); + sctx->stat.malloc_errors++; + sctx->stat.read_errors++; + sctx->stat.uncorrectable_errors++; + spin_unlock(&sctx->stat_lock); + btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_READ_ERRS); + goto out; + } } - /* setup the context, map the logical blocks and alloc the pages */ + /* Setup the context, map the logical blocks and alloc the sectors */ ret = scrub_setup_recheck_block(sblock_to_check, sblocks_for_recheck); if (ret) { spin_lock(&sctx->stat_lock); @@ -941,7 +1122,7 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) goto out; } BUG_ON(failed_mirror_index >= BTRFS_MAX_MIRRORS); - sblock_bad = sblocks_for_recheck + failed_mirror_index; + sblock_bad = sblocks_for_recheck[failed_mirror_index]; /* build and submit the bios for the failed mirror, check checksums */ scrub_recheck_block(fs_info, sblock_bad, 1); @@ -949,7 +1130,7 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) if (!sblock_bad->header_error && !sblock_bad->checksum_error && sblock_bad->no_io_error_seen) { /* - * the error disappeared after reading page by page, or + * The error disappeared after reading sector by sector, or * the area was part of a huge bio and other parts of the * bio caused I/O errors, or the block layer merged several * read requests into one and the error is caused by a @@ -970,14 +1151,14 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) spin_lock(&sctx->stat_lock); sctx->stat.read_errors++; spin_unlock(&sctx->stat_lock); - if (__ratelimit(&_rs)) + if (__ratelimit(&rs)) scrub_print_warning("i/o error", sblock_to_check); btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_READ_ERRS); } else if (sblock_bad->checksum_error) { spin_lock(&sctx->stat_lock); sctx->stat.csum_errors++; spin_unlock(&sctx->stat_lock); - if (__ratelimit(&_rs)) + if (__ratelimit(&rs)) scrub_print_warning("checksum error", sblock_to_check); btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_CORRUPTION_ERRS); @@ -985,7 +1166,7 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) spin_lock(&sctx->stat_lock); sctx->stat.verify_errors++; spin_unlock(&sctx->stat_lock); - if (__ratelimit(&_rs)) + if (__ratelimit(&rs)) scrub_print_warning("checksum/header error", sblock_to_check); if (sblock_bad->generation_error) @@ -1010,10 +1191,10 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) * that is known to contain an error is rewritten. Afterwards * the block is known to be corrected. * If a mirror is found which is completely correct, and no - * checksum is present, only those pages are rewritten that had + * checksum is present, only those sectors are rewritten that had * an I/O error in the block to be repaired, since it cannot be - * determined, which copy of the other pages is better (and it - * could happen otherwise that a correct page would be + * determined, which copy of the other sectors is better (and it + * could happen otherwise that a correct sector would be * overwritten by a bad one). */ for (mirror_index = 0; ;mirror_index++) { @@ -1023,26 +1204,25 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) continue; /* raid56's mirror can be more than BTRFS_MAX_MIRRORS */ - if (!scrub_is_page_on_raid56(sblock_bad->pagev[0])) { + if (!scrub_is_page_on_raid56(sblock_bad->sectors[0])) { if (mirror_index >= BTRFS_MAX_MIRRORS) break; - if (!sblocks_for_recheck[mirror_index].page_count) + if (!sblocks_for_recheck[mirror_index]->sector_count) break; - sblock_other = sblocks_for_recheck + mirror_index; + sblock_other = sblocks_for_recheck[mirror_index]; } else { - struct scrub_recover *r = sblock_bad->pagev[0]->recover; - int max_allowed = r->bbio->num_stripes - - r->bbio->num_tgtdevs; + struct scrub_recover *r = sblock_bad->sectors[0]->recover; + int max_allowed = r->bioc->num_stripes - r->bioc->num_tgtdevs; if (mirror_index >= max_allowed) break; - if (!sblocks_for_recheck[1].page_count) + if (!sblocks_for_recheck[1]->sector_count) break; ASSERT(failed_mirror_index == 0); - sblock_other = sblocks_for_recheck + 1; - sblock_other->pagev[0]->mirror_num = 1 + mirror_index; + sblock_other = sblocks_for_recheck[1]; + sblock_other->mirror_num = 1 + mirror_index; } /* build and submit the bios, check checksums */ @@ -1068,39 +1248,39 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) /* * In case of I/O errors in the area that is supposed to be - * repaired, continue by picking good copies of those pages. - * Select the good pages from mirrors to rewrite bad pages from + * repaired, continue by picking good copies of those sectors. + * Select the good sectors from mirrors to rewrite bad sectors from * the area to fix. Afterwards verify the checksum of the block * that is supposed to be repaired. This verification step is * only done for the purpose of statistic counting and for the * final scrub report, whether errors remain. * A perfect algorithm could make use of the checksum and try - * all possible combinations of pages from the different mirrors + * all possible combinations of sectors from the different mirrors * until the checksum verification succeeds. For example, when - * the 2nd page of mirror #1 faces I/O errors, and the 2nd page + * the 2nd sector of mirror #1 faces I/O errors, and the 2nd sector * of mirror #2 is readable but the final checksum test fails, - * then the 2nd page of mirror #3 could be tried, whether now + * then the 2nd sector of mirror #3 could be tried, whether now * the final checksum succeeds. But this would be a rare * exception and is therefore not implemented. At least it is * avoided that the good copy is overwritten. * A more useful improvement would be to pick the sectors * without I/O error based on sector sizes (512 bytes on legacy - * disks) instead of on PAGE_SIZE. Then maybe 512 byte of one + * disks) instead of on sectorsize. Then maybe 512 byte of one * mirror could be repaired by taking 512 byte of a different - * mirror, even if other 512 byte sectors in the same PAGE_SIZE + * mirror, even if other 512 byte sectors in the same sectorsize * area are unreadable. */ success = 1; - for (page_num = 0; page_num < sblock_bad->page_count; - page_num++) { - struct scrub_page *page_bad = sblock_bad->pagev[page_num]; + for (sector_num = 0; sector_num < sblock_bad->sector_count; + sector_num++) { + struct scrub_sector *sector_bad = sblock_bad->sectors[sector_num]; struct scrub_block *sblock_other = NULL; - /* skip no-io-error page in scrub */ - if (!page_bad->io_error && !sctx->is_dev_replace) + /* Skip no-io-error sectors in scrub */ + if (!sector_bad->io_error && !sctx->is_dev_replace) continue; - if (scrub_is_page_on_raid56(sblock_bad->pagev[0])) { + if (scrub_is_page_on_raid56(sblock_bad->sectors[0])) { /* * In case of dev replace, if raid56 rebuild process * didn't work out correct data, then copy the content @@ -1109,16 +1289,15 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) * sblock_for_recheck array to target device. */ sblock_other = NULL; - } else if (page_bad->io_error) { - /* try to find no-io-error page in mirrors */ + } else if (sector_bad->io_error) { + /* Try to find no-io-error sector in mirrors */ for (mirror_index = 0; mirror_index < BTRFS_MAX_MIRRORS && - sblocks_for_recheck[mirror_index].page_count > 0; + sblocks_for_recheck[mirror_index]->sector_count > 0; mirror_index++) { - if (!sblocks_for_recheck[mirror_index]. - pagev[page_num]->io_error) { - sblock_other = sblocks_for_recheck + - mirror_index; + if (!sblocks_for_recheck[mirror_index]-> + sectors[sector_num]->io_error) { + sblock_other = sblocks_for_recheck[mirror_index]; break; } } @@ -1128,27 +1307,26 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) if (sctx->is_dev_replace) { /* - * did not find a mirror to fetch the page - * from. scrub_write_page_to_dev_replace() - * handles this case (page->io_error), by - * filling the block with zeros before - * submitting the write request + * Did not find a mirror to fetch the sector from. + * scrub_write_sector_to_dev_replace() handles this + * case (sector->io_error), by filling the block with + * zeros before submitting the write request */ if (!sblock_other) sblock_other = sblock_bad; - if (scrub_write_page_to_dev_replace(sblock_other, - page_num) != 0) { + if (scrub_write_sector_to_dev_replace(sblock_other, + sector_num) != 0) { atomic64_inc( &fs_info->dev_replace.num_write_errors); success = 0; } } else if (sblock_other) { - ret = scrub_repair_page_from_good_copy(sblock_bad, - sblock_other, - page_num, 0); + ret = scrub_repair_sector_from_good_copy(sblock_bad, + sblock_other, + sector_num, 0); if (0 == ret) - page_bad->io_error = 0; + sector_bad->io_error = 0; else success = 0; } @@ -1193,27 +1371,28 @@ did_not_correct_error: } out: - if (sblocks_for_recheck) { - for (mirror_index = 0; mirror_index < BTRFS_MAX_MIRRORS; - mirror_index++) { - struct scrub_block *sblock = sblocks_for_recheck + - mirror_index; - struct scrub_recover *recover; - int page_index; - - for (page_index = 0; page_index < sblock->page_count; - page_index++) { - sblock->pagev[page_index]->sblock = NULL; - recover = sblock->pagev[page_index]->recover; - if (recover) { - scrub_put_recover(fs_info, recover); - sblock->pagev[page_index]->recover = - NULL; - } - scrub_page_put(sblock->pagev[page_index]); + for (mirror_index = 0; mirror_index < BTRFS_MAX_MIRRORS; mirror_index++) { + struct scrub_block *sblock = sblocks_for_recheck[mirror_index]; + struct scrub_recover *recover; + int sector_index; + + /* Not allocated, continue checking the next mirror */ + if (!sblock) + continue; + + for (sector_index = 0; sector_index < sblock->sector_count; + sector_index++) { + /* + * Here we just cleanup the recover, each sector will be + * properly cleaned up by later scrub_block_put() + */ + recover = sblock->sectors[sector_index]->recover; + if (recover) { + scrub_put_recover(fs_info, recover); + sblock->sectors[sector_index]->recover = NULL; } } - kfree(sblocks_for_recheck); + scrub_block_put(sblock); } ret = unlock_full_stripe(fs_info, logical, full_stripe_locked); @@ -1223,19 +1402,18 @@ out: return 0; } -static inline int scrub_nr_raid_mirrors(struct btrfs_bio *bbio) +static inline int scrub_nr_raid_mirrors(struct btrfs_io_context *bioc) { - if (bbio->map_type & BTRFS_BLOCK_GROUP_RAID5) + if (bioc->map_type & BTRFS_BLOCK_GROUP_RAID5) return 2; - else if (bbio->map_type & BTRFS_BLOCK_GROUP_RAID6) + else if (bioc->map_type & BTRFS_BLOCK_GROUP_RAID6) return 3; else - return (int)bbio->num_stripes; + return (int)bioc->num_stripes; } static inline void scrub_stripe_index_and_offset(u64 logical, u64 map_type, u64 *raid_map, - u64 mapped_length, int nstripes, int mirror, int *stripe_index, u64 *stripe_offset) @@ -1250,7 +1428,7 @@ static inline void scrub_stripe_index_and_offset(u64 logical, u64 map_type, continue; if (logical >= raid_map[i] && - logical < raid_map[i] + mapped_length) + logical < raid_map[i] + BTRFS_STRIPE_LEN) break; } @@ -1264,125 +1442,112 @@ static inline void scrub_stripe_index_and_offset(u64 logical, u64 map_type, } static int scrub_setup_recheck_block(struct scrub_block *original_sblock, - struct scrub_block *sblocks_for_recheck) + struct scrub_block *sblocks_for_recheck[]) { struct scrub_ctx *sctx = original_sblock->sctx; struct btrfs_fs_info *fs_info = sctx->fs_info; - u64 length = original_sblock->page_count * PAGE_SIZE; - u64 logical = original_sblock->pagev[0]->logical; - u64 generation = original_sblock->pagev[0]->generation; - u64 flags = original_sblock->pagev[0]->flags; - u64 have_csum = original_sblock->pagev[0]->have_csum; + u64 logical = original_sblock->logical; + u64 length = original_sblock->sector_count << fs_info->sectorsize_bits; + u64 generation = original_sblock->sectors[0]->generation; + u64 flags = original_sblock->sectors[0]->flags; + u64 have_csum = original_sblock->sectors[0]->have_csum; struct scrub_recover *recover; - struct btrfs_bio *bbio; + struct btrfs_io_context *bioc; u64 sublen; u64 mapped_length; u64 stripe_offset; int stripe_index; - int page_index = 0; + int sector_index = 0; int mirror_index; int nmirrors; int ret; - /* - * note: the two members refs and outstanding_pages - * are not used (and not set) in the blocks that are used for - * the recheck procedure - */ - while (length > 0) { - sublen = min_t(u64, length, PAGE_SIZE); + sublen = min_t(u64, length, fs_info->sectorsize); mapped_length = sublen; - bbio = NULL; + bioc = NULL; /* - * with a length of PAGE_SIZE, each returned stripe - * represents one mirror + * With a length of sectorsize, each returned stripe represents + * one mirror */ btrfs_bio_counter_inc_blocked(fs_info); ret = btrfs_map_sblock(fs_info, BTRFS_MAP_GET_READ_MIRRORS, - logical, &mapped_length, &bbio); - if (ret || !bbio || mapped_length < sublen) { - btrfs_put_bbio(bbio); + logical, &mapped_length, &bioc); + if (ret || !bioc || mapped_length < sublen) { + btrfs_put_bioc(bioc); btrfs_bio_counter_dec(fs_info); return -EIO; } recover = kzalloc(sizeof(struct scrub_recover), GFP_NOFS); if (!recover) { - btrfs_put_bbio(bbio); + btrfs_put_bioc(bioc); btrfs_bio_counter_dec(fs_info); return -ENOMEM; } refcount_set(&recover->refs, 1); - recover->bbio = bbio; + recover->bioc = bioc; recover->map_length = mapped_length; - BUG_ON(page_index >= SCRUB_MAX_PAGES_PER_BLOCK); + ASSERT(sector_index < SCRUB_MAX_SECTORS_PER_BLOCK); - nmirrors = min(scrub_nr_raid_mirrors(bbio), BTRFS_MAX_MIRRORS); + nmirrors = min(scrub_nr_raid_mirrors(bioc), BTRFS_MAX_MIRRORS); for (mirror_index = 0; mirror_index < nmirrors; mirror_index++) { struct scrub_block *sblock; - struct scrub_page *page; + struct scrub_sector *sector; - sblock = sblocks_for_recheck + mirror_index; + sblock = sblocks_for_recheck[mirror_index]; sblock->sctx = sctx; - page = kzalloc(sizeof(*page), GFP_NOFS); - if (!page) { -leave_nomem: + sector = alloc_scrub_sector(sblock, logical, GFP_NOFS); + if (!sector) { spin_lock(&sctx->stat_lock); sctx->stat.malloc_errors++; spin_unlock(&sctx->stat_lock); scrub_put_recover(fs_info, recover); return -ENOMEM; } - scrub_page_get(page); - sblock->pagev[page_index] = page; - page->sblock = sblock; - page->flags = flags; - page->generation = generation; - page->logical = logical; - page->have_csum = have_csum; + sector->flags = flags; + sector->generation = generation; + sector->have_csum = have_csum; if (have_csum) - memcpy(page->csum, - original_sblock->pagev[0]->csum, - sctx->csum_size); + memcpy(sector->csum, + original_sblock->sectors[0]->csum, + sctx->fs_info->csum_size); scrub_stripe_index_and_offset(logical, - bbio->map_type, - bbio->raid_map, - mapped_length, - bbio->num_stripes - - bbio->num_tgtdevs, + bioc->map_type, + bioc->raid_map, + bioc->num_stripes - + bioc->num_tgtdevs, mirror_index, &stripe_index, &stripe_offset); - page->physical = bbio->stripes[stripe_index].physical + - stripe_offset; - page->dev = bbio->stripes[stripe_index].dev; - - BUG_ON(page_index >= original_sblock->page_count); - page->physical_for_dev_replace = - original_sblock->pagev[page_index]-> - physical_for_dev_replace; - /* for missing devices, dev->bdev is NULL */ - page->mirror_num = mirror_index + 1; - sblock->page_count++; - page->page = alloc_page(GFP_NOFS); - if (!page->page) - goto leave_nomem; + /* + * We're at the first sector, also populate @sblock + * physical and dev. + */ + if (sector_index == 0) { + sblock->physical = + bioc->stripes[stripe_index].physical + + stripe_offset; + sblock->dev = bioc->stripes[stripe_index].dev; + sblock->physical_for_dev_replace = + original_sblock->physical_for_dev_replace; + } + BUG_ON(sector_index >= original_sblock->sector_count); scrub_get_recover(recover); - page->recover = recover; + sector->recover = recover; } scrub_put_recover(fs_info, recover); length -= sublen; logical += sublen; - page_index++; + sector_index++; } return 0; @@ -1395,22 +1560,15 @@ static void scrub_bio_wait_endio(struct bio *bio) static int scrub_submit_raid56_bio_wait(struct btrfs_fs_info *fs_info, struct bio *bio, - struct scrub_page *page) + struct scrub_sector *sector) { DECLARE_COMPLETION_ONSTACK(done); - int ret; - int mirror_num; - bio->bi_iter.bi_sector = page->logical >> 9; + bio->bi_iter.bi_sector = (sector->offset + sector->sblock->logical) >> + SECTOR_SHIFT; bio->bi_private = &done; bio->bi_end_io = scrub_bio_wait_endio; - - mirror_num = page->sblock->pagev[0]->mirror_num; - ret = raid56_parity_recover(fs_info, bio, page->recover->bbio, - page->recover->map_length, - mirror_num, 0); - if (ret) - return ret; + raid56_parity_recover(bio, sector->recover->bioc, sector->sblock->mirror_num); wait_for_completion_io(&done); return blk_status_to_errno(bio->bi_status); @@ -1419,26 +1577,24 @@ static int scrub_submit_raid56_bio_wait(struct btrfs_fs_info *fs_info, static void scrub_recheck_block_on_raid56(struct btrfs_fs_info *fs_info, struct scrub_block *sblock) { - struct scrub_page *first_page = sblock->pagev[0]; + struct scrub_sector *first_sector = sblock->sectors[0]; struct bio *bio; - int page_num; + int i; - /* All pages in sblock belong to the same stripe on the same device. */ - ASSERT(first_page->dev); - if (!first_page->dev->bdev) + /* All sectors in sblock belong to the same stripe on the same device. */ + ASSERT(sblock->dev); + if (!sblock->dev->bdev) goto out; - bio = btrfs_io_bio_alloc(BIO_MAX_PAGES); - bio_set_dev(bio, first_page->dev->bdev); + bio = bio_alloc(sblock->dev->bdev, BIO_MAX_VECS, REQ_OP_READ, GFP_NOFS); - for (page_num = 0; page_num < sblock->page_count; page_num++) { - struct scrub_page *page = sblock->pagev[page_num]; + for (i = 0; i < sblock->sector_count; i++) { + struct scrub_sector *sector = sblock->sectors[i]; - WARN_ON(!page->page); - bio_add_page(bio, page->page, PAGE_SIZE, 0); + bio_add_scrub_sector(bio, sector, fs_info->sectorsize); } - if (scrub_submit_raid56_bio_wait(fs_info, bio, first_page)) { + if (scrub_submit_raid56_bio_wait(fs_info, bio, first_sector)) { bio_put(bio); goto out; } @@ -1449,65 +1605,63 @@ static void scrub_recheck_block_on_raid56(struct btrfs_fs_info *fs_info, return; out: - for (page_num = 0; page_num < sblock->page_count; page_num++) - sblock->pagev[page_num]->io_error = 1; + for (i = 0; i < sblock->sector_count; i++) + sblock->sectors[i]->io_error = 1; sblock->no_io_error_seen = 0; } /* - * this function will check the on disk data for checksum errors, header - * errors and read I/O errors. If any I/O errors happen, the exact pages - * which are errored are marked as being bad. The goal is to enable scrub - * to take those pages that are not errored from all the mirrors so that - * the pages that are errored in the just handled mirror can be repaired. + * This function will check the on disk data for checksum errors, header errors + * and read I/O errors. If any I/O errors happen, the exact sectors which are + * errored are marked as being bad. The goal is to enable scrub to take those + * sectors that are not errored from all the mirrors so that the sectors that + * are errored in the just handled mirror can be repaired. */ static void scrub_recheck_block(struct btrfs_fs_info *fs_info, struct scrub_block *sblock, int retry_failed_mirror) { - int page_num; + int i; sblock->no_io_error_seen = 1; /* short cut for raid56 */ - if (!retry_failed_mirror && scrub_is_page_on_raid56(sblock->pagev[0])) + if (!retry_failed_mirror && scrub_is_page_on_raid56(sblock->sectors[0])) return scrub_recheck_block_on_raid56(fs_info, sblock); - for (page_num = 0; page_num < sblock->page_count; page_num++) { - struct bio *bio; - struct scrub_page *page = sblock->pagev[page_num]; + for (i = 0; i < sblock->sector_count; i++) { + struct scrub_sector *sector = sblock->sectors[i]; + struct bio bio; + struct bio_vec bvec; - if (page->dev->bdev == NULL) { - page->io_error = 1; + if (sblock->dev->bdev == NULL) { + sector->io_error = 1; sblock->no_io_error_seen = 0; continue; } - WARN_ON(!page->page); - bio = btrfs_io_bio_alloc(1); - bio_set_dev(bio, page->dev->bdev); + bio_init(&bio, sblock->dev->bdev, &bvec, 1, REQ_OP_READ); + bio_add_scrub_sector(&bio, sector, fs_info->sectorsize); + bio.bi_iter.bi_sector = (sblock->physical + sector->offset) >> + SECTOR_SHIFT; - bio_add_page(bio, page->page, PAGE_SIZE, 0); - bio->bi_iter.bi_sector = page->physical >> 9; - bio->bi_opf = REQ_OP_READ; - - if (btrfsic_submit_bio_wait(bio)) { - page->io_error = 1; + btrfsic_check_bio(&bio); + if (submit_bio_wait(&bio)) { + sector->io_error = 1; sblock->no_io_error_seen = 0; } - bio_put(bio); + bio_uninit(&bio); } if (sblock->no_io_error_seen) scrub_recheck_block_checksum(sblock); } -static inline int scrub_check_fsid(u8 fsid[], - struct scrub_page *spage) +static inline int scrub_check_fsid(u8 fsid[], struct scrub_sector *sector) { - struct btrfs_fs_devices *fs_devices = spage->dev->fs_devices; + struct btrfs_fs_devices *fs_devices = sector->sblock->dev->fs_devices; int ret; ret = memcmp(fsid, fs_devices->fsid, BTRFS_FSID_SIZE); @@ -1520,7 +1674,7 @@ static void scrub_recheck_block_checksum(struct scrub_block *sblock) sblock->checksum_error = 0; sblock->generation_error = 0; - if (sblock->pagev[0]->flags & BTRFS_EXTENT_FLAG_DATA) + if (sblock->sectors[0]->flags & BTRFS_EXTENT_FLAG_DATA) scrub_checksum_data(sblock); else scrub_checksum_tree_block(sblock); @@ -1529,15 +1683,14 @@ static void scrub_recheck_block_checksum(struct scrub_block *sblock) static int scrub_repair_block_from_good_copy(struct scrub_block *sblock_bad, struct scrub_block *sblock_good) { - int page_num; + int i; int ret = 0; - for (page_num = 0; page_num < sblock_bad->page_count; page_num++) { + for (i = 0; i < sblock_bad->sector_count; i++) { int ret_sub; - ret_sub = scrub_repair_page_from_good_copy(sblock_bad, - sblock_good, - page_num, 1); + ret_sub = scrub_repair_sector_from_good_copy(sblock_bad, + sblock_good, i, 1); if (ret_sub) ret = ret_sub; } @@ -1545,46 +1698,42 @@ static int scrub_repair_block_from_good_copy(struct scrub_block *sblock_bad, return ret; } -static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad, - struct scrub_block *sblock_good, - int page_num, int force_write) +static int scrub_repair_sector_from_good_copy(struct scrub_block *sblock_bad, + struct scrub_block *sblock_good, + int sector_num, int force_write) { - struct scrub_page *page_bad = sblock_bad->pagev[page_num]; - struct scrub_page *page_good = sblock_good->pagev[page_num]; + struct scrub_sector *sector_bad = sblock_bad->sectors[sector_num]; + struct scrub_sector *sector_good = sblock_good->sectors[sector_num]; struct btrfs_fs_info *fs_info = sblock_bad->sctx->fs_info; + const u32 sectorsize = fs_info->sectorsize; - BUG_ON(page_bad->page == NULL); - BUG_ON(page_good->page == NULL); if (force_write || sblock_bad->header_error || - sblock_bad->checksum_error || page_bad->io_error) { - struct bio *bio; + sblock_bad->checksum_error || sector_bad->io_error) { + struct bio bio; + struct bio_vec bvec; int ret; - if (!page_bad->dev->bdev) { + if (!sblock_bad->dev->bdev) { btrfs_warn_rl(fs_info, "scrub_repair_page_from_good_copy(bdev == NULL) is unexpected"); return -EIO; } - bio = btrfs_io_bio_alloc(1); - bio_set_dev(bio, page_bad->dev->bdev); - bio->bi_iter.bi_sector = page_bad->physical >> 9; - bio->bi_opf = REQ_OP_WRITE; + bio_init(&bio, sblock_bad->dev->bdev, &bvec, 1, REQ_OP_WRITE); + bio.bi_iter.bi_sector = (sblock_bad->physical + + sector_bad->offset) >> SECTOR_SHIFT; + ret = bio_add_scrub_sector(&bio, sector_good, sectorsize); - ret = bio_add_page(bio, page_good->page, PAGE_SIZE, 0); - if (PAGE_SIZE != ret) { - bio_put(bio); - return -EIO; - } + btrfsic_check_bio(&bio); + ret = submit_bio_wait(&bio); + bio_uninit(&bio); - if (btrfsic_submit_bio_wait(bio)) { - btrfs_dev_stat_inc_and_print(page_bad->dev, + if (ret) { + btrfs_dev_stat_inc_and_print(sblock_bad->dev, BTRFS_DEV_STAT_WRITE_ERRS); atomic64_inc(&fs_info->dev_replace.num_write_errors); - bio_put(bio); return -EIO; } - bio_put(bio); } return 0; @@ -1593,7 +1742,7 @@ static int scrub_repair_page_from_good_copy(struct scrub_block *sblock_bad, static void scrub_write_block_to_dev_replace(struct scrub_block *sblock) { struct btrfs_fs_info *fs_info = sblock->sctx->fs_info; - int page_num; + int i; /* * This block is used for the check of the parity on the source device, @@ -1602,36 +1751,60 @@ static void scrub_write_block_to_dev_replace(struct scrub_block *sblock) if (sblock->sparity) return; - for (page_num = 0; page_num < sblock->page_count; page_num++) { + for (i = 0; i < sblock->sector_count; i++) { int ret; - ret = scrub_write_page_to_dev_replace(sblock, page_num); + ret = scrub_write_sector_to_dev_replace(sblock, i); if (ret) atomic64_inc(&fs_info->dev_replace.num_write_errors); } } -static int scrub_write_page_to_dev_replace(struct scrub_block *sblock, - int page_num) +static int scrub_write_sector_to_dev_replace(struct scrub_block *sblock, int sector_num) +{ + const u32 sectorsize = sblock->sctx->fs_info->sectorsize; + struct scrub_sector *sector = sblock->sectors[sector_num]; + + if (sector->io_error) + memset(scrub_sector_get_kaddr(sector), 0, sectorsize); + + return scrub_add_sector_to_wr_bio(sblock->sctx, sector); +} + +static int fill_writer_pointer_gap(struct scrub_ctx *sctx, u64 physical) { - struct scrub_page *spage = sblock->pagev[page_num]; + int ret = 0; + u64 length; - BUG_ON(spage->page == NULL); - if (spage->io_error) { - void *mapped_buffer = kmap_atomic(spage->page); + if (!btrfs_is_zoned(sctx->fs_info)) + return 0; + + if (!btrfs_dev_is_sequential(sctx->wr_tgtdev, physical)) + return 0; - clear_page(mapped_buffer); - flush_dcache_page(spage->page); - kunmap_atomic(mapped_buffer); + if (sctx->write_pointer < physical) { + length = physical - sctx->write_pointer; + + ret = btrfs_zoned_issue_zeroout(sctx->wr_tgtdev, + sctx->write_pointer, length); + if (!ret) + sctx->write_pointer = physical; } - return scrub_add_page_to_wr_bio(sblock->sctx, spage); + return ret; } -static int scrub_add_page_to_wr_bio(struct scrub_ctx *sctx, - struct scrub_page *spage) +static void scrub_block_get(struct scrub_block *sblock) { + refcount_inc(&sblock->refs); +} + +static int scrub_add_sector_to_wr_bio(struct scrub_ctx *sctx, + struct scrub_sector *sector) +{ + struct scrub_block *sblock = sector->sblock; struct scrub_bio *sbio; int ret; + const u32 sectorsize = sctx->fs_info->sectorsize; mutex_lock(&sctx->wr_lock); again: @@ -1643,38 +1816,39 @@ again: return -ENOMEM; } sctx->wr_curr_bio->sctx = sctx; - sctx->wr_curr_bio->page_count = 0; + sctx->wr_curr_bio->sector_count = 0; } sbio = sctx->wr_curr_bio; - if (sbio->page_count == 0) { - struct bio *bio; + if (sbio->sector_count == 0) { + ret = fill_writer_pointer_gap(sctx, sector->offset + + sblock->physical_for_dev_replace); + if (ret) { + mutex_unlock(&sctx->wr_lock); + return ret; + } - sbio->physical = spage->physical_for_dev_replace; - sbio->logical = spage->logical; + sbio->physical = sblock->physical_for_dev_replace + sector->offset; + sbio->logical = sblock->logical + sector->offset; sbio->dev = sctx->wr_tgtdev; - bio = sbio->bio; - if (!bio) { - bio = btrfs_io_bio_alloc(sctx->pages_per_wr_bio); - sbio->bio = bio; + if (!sbio->bio) { + sbio->bio = bio_alloc(sbio->dev->bdev, sctx->sectors_per_bio, + REQ_OP_WRITE, GFP_NOFS); } - - bio->bi_private = sbio; - bio->bi_end_io = scrub_wr_bio_end_io; - bio_set_dev(bio, sbio->dev->bdev); - bio->bi_iter.bi_sector = sbio->physical >> 9; - bio->bi_opf = REQ_OP_WRITE; + sbio->bio->bi_private = sbio; + sbio->bio->bi_end_io = scrub_wr_bio_end_io; + sbio->bio->bi_iter.bi_sector = sbio->physical >> 9; sbio->status = 0; - } else if (sbio->physical + sbio->page_count * PAGE_SIZE != - spage->physical_for_dev_replace || - sbio->logical + sbio->page_count * PAGE_SIZE != - spage->logical) { + } else if (sbio->physical + sbio->sector_count * sectorsize != + sblock->physical_for_dev_replace + sector->offset || + sbio->logical + sbio->sector_count * sectorsize != + sblock->logical + sector->offset) { scrub_wr_submit(sctx); goto again; } - ret = bio_add_page(sbio->bio, spage->page, PAGE_SIZE, 0); - if (ret != PAGE_SIZE) { - if (sbio->page_count < 1) { + ret = bio_add_scrub_sector(sbio->bio, sector, sectorsize); + if (ret != sectorsize) { + if (sbio->sector_count < 1) { bio_put(sbio->bio); sbio->bio = NULL; mutex_unlock(&sctx->wr_lock); @@ -1684,10 +1858,17 @@ again: goto again; } - sbio->pagev[sbio->page_count] = spage; - scrub_page_get(spage); - sbio->page_count++; - if (sbio->page_count == sctx->pages_per_wr_bio) + sbio->sectors[sbio->sector_count] = sector; + scrub_sector_get(sector); + /* + * Since ssector no longer holds a page, but uses sblock::pages, we + * have to ensure the sblock had not been freed before our write bio + * finished. + */ + scrub_block_get(sector->sblock); + + sbio->sector_count++; + if (sbio->sector_count == sctx->sectors_per_bio) scrub_wr_submit(sctx); mutex_unlock(&sctx->wr_lock); @@ -1703,13 +1884,17 @@ static void scrub_wr_submit(struct scrub_ctx *sctx) sbio = sctx->wr_curr_bio; sctx->wr_curr_bio = NULL; - WARN_ON(!sbio->bio->bi_disk); scrub_pending_bio_inc(sctx); /* process all writes in a single worker thread. Then the block layer * orders the requests before sending them to the driver which * doubled the write performance on spinning disks when measured * with Linux 3.5 */ - btrfsic_submit_bio(sbio->bio); + btrfsic_check_bio(sbio->bio); + submit_bio(sbio->bio); + + if (btrfs_is_zoned(sctx->fs_info)) + sctx->write_pointer = sbio->physical + sbio->sector_count * + sctx->fs_info->sectorsize; } static void scrub_wr_bio_end_io(struct bio *bio) @@ -1720,31 +1905,37 @@ static void scrub_wr_bio_end_io(struct bio *bio) sbio->status = bio->bi_status; sbio->bio = bio; - btrfs_init_work(&sbio->work, scrub_wr_bio_end_io_worker, NULL, NULL); - btrfs_queue_work(fs_info->scrub_wr_completion_workers, &sbio->work); + INIT_WORK(&sbio->work, scrub_wr_bio_end_io_worker); + queue_work(fs_info->scrub_wr_completion_workers, &sbio->work); } -static void scrub_wr_bio_end_io_worker(struct btrfs_work *work) +static void scrub_wr_bio_end_io_worker(struct work_struct *work) { struct scrub_bio *sbio = container_of(work, struct scrub_bio, work); struct scrub_ctx *sctx = sbio->sctx; int i; - WARN_ON(sbio->page_count > SCRUB_PAGES_PER_WR_BIO); + ASSERT(sbio->sector_count <= SCRUB_SECTORS_PER_BIO); if (sbio->status) { struct btrfs_dev_replace *dev_replace = &sbio->sctx->fs_info->dev_replace; - for (i = 0; i < sbio->page_count; i++) { - struct scrub_page *spage = sbio->pagev[i]; + for (i = 0; i < sbio->sector_count; i++) { + struct scrub_sector *sector = sbio->sectors[i]; - spage->io_error = 1; + sector->io_error = 1; atomic64_inc(&dev_replace->num_write_errors); } } - for (i = 0; i < sbio->page_count; i++) - scrub_page_put(sbio->pagev[i]); + /* + * In scrub_add_sector_to_wr_bio() we grab extra ref for sblock, now in + * endio we should put the sblock. + */ + for (i = 0; i < sbio->sector_count; i++) { + scrub_block_put(sbio->sectors[i]->sblock); + scrub_sector_put(sbio->sectors[i]); + } bio_put(sbio->bio); kfree(sbio); @@ -1768,15 +1959,15 @@ static int scrub_checksum(struct scrub_block *sblock) sblock->generation_error = 0; sblock->checksum_error = 0; - WARN_ON(sblock->page_count < 1); - flags = sblock->pagev[0]->flags; + WARN_ON(sblock->sector_count < 1); + flags = sblock->sectors[0]->flags; ret = 0; if (flags & BTRFS_EXTENT_FLAG_DATA) ret = scrub_checksum_data(sblock); else if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) ret = scrub_checksum_tree_block(sblock); else if (flags & BTRFS_EXTENT_FLAG_SUPER) - (void)scrub_checksum_super(sblock); + ret = scrub_checksum_super(sblock); else WARN_ON(1); if (ret) @@ -1791,44 +1982,23 @@ static int scrub_checksum_data(struct scrub_block *sblock) struct btrfs_fs_info *fs_info = sctx->fs_info; SHASH_DESC_ON_STACK(shash, fs_info->csum_shash); u8 csum[BTRFS_CSUM_SIZE]; - u8 *on_disk_csum; - struct page *page; - void *buffer; - u64 len; - int index; + struct scrub_sector *sector; + char *kaddr; - BUG_ON(sblock->page_count < 1); - if (!sblock->pagev[0]->have_csum) + BUG_ON(sblock->sector_count < 1); + sector = sblock->sectors[0]; + if (!sector->have_csum) return 0; + kaddr = scrub_sector_get_kaddr(sector); + shash->tfm = fs_info->csum_shash; crypto_shash_init(shash); - on_disk_csum = sblock->pagev[0]->csum; - page = sblock->pagev[0]->page; - buffer = kmap_atomic(page); - - len = sctx->fs_info->sectorsize; - index = 0; - for (;;) { - u64 l = min_t(u64, len, PAGE_SIZE); - - crypto_shash_update(shash, buffer, l); - kunmap_atomic(buffer); - len -= l; - if (len == 0) - break; - index++; - BUG_ON(index >= sblock->page_count); - BUG_ON(!sblock->pagev[index]->page); - page = sblock->pagev[index]->page; - buffer = kmap_atomic(page); - } + crypto_shash_digest(shash, kaddr, fs_info->sectorsize, csum); - crypto_shash_final(shash, csum); - if (memcmp(csum, on_disk_csum, sctx->csum_size)) + if (memcmp(csum, sector->csum, fs_info->csum_size)) sblock->checksum_error = 1; - return sblock->checksum_error; } @@ -1840,65 +2010,59 @@ static int scrub_checksum_tree_block(struct scrub_block *sblock) SHASH_DESC_ON_STACK(shash, fs_info->csum_shash); u8 calculated_csum[BTRFS_CSUM_SIZE]; u8 on_disk_csum[BTRFS_CSUM_SIZE]; - struct page *page; - void *mapped_buffer; - u64 mapped_size; - void *p; - u64 len; - int index; + /* + * This is done in sectorsize steps even for metadata as there's a + * constraint for nodesize to be aligned to sectorsize. This will need + * to change so we don't misuse data and metadata units like that. + */ + const u32 sectorsize = sctx->fs_info->sectorsize; + const int num_sectors = fs_info->nodesize >> fs_info->sectorsize_bits; + int i; + struct scrub_sector *sector; + char *kaddr; - shash->tfm = fs_info->csum_shash; - crypto_shash_init(shash); + BUG_ON(sblock->sector_count < 1); + + /* Each member in sectors is just one sector */ + ASSERT(sblock->sector_count == num_sectors); - BUG_ON(sblock->page_count < 1); - page = sblock->pagev[0]->page; - mapped_buffer = kmap_atomic(page); - h = (struct btrfs_header *)mapped_buffer; - memcpy(on_disk_csum, h->csum, sctx->csum_size); + sector = sblock->sectors[0]; + kaddr = scrub_sector_get_kaddr(sector); + h = (struct btrfs_header *)kaddr; + memcpy(on_disk_csum, h->csum, sctx->fs_info->csum_size); /* * we don't use the getter functions here, as we * a) don't have an extent buffer and * b) the page is already kmapped */ - if (sblock->pagev[0]->logical != btrfs_stack_header_bytenr(h)) + if (sblock->logical != btrfs_stack_header_bytenr(h)) sblock->header_error = 1; - if (sblock->pagev[0]->generation != btrfs_stack_header_generation(h)) { + if (sector->generation != btrfs_stack_header_generation(h)) { sblock->header_error = 1; sblock->generation_error = 1; } - if (!scrub_check_fsid(h->fsid, sblock->pagev[0])) + if (!scrub_check_fsid(h->fsid, sector)) sblock->header_error = 1; if (memcmp(h->chunk_tree_uuid, fs_info->chunk_tree_uuid, BTRFS_UUID_SIZE)) sblock->header_error = 1; - len = sctx->fs_info->nodesize - BTRFS_CSUM_SIZE; - mapped_size = PAGE_SIZE - BTRFS_CSUM_SIZE; - p = ((u8 *)mapped_buffer) + BTRFS_CSUM_SIZE; - index = 0; - for (;;) { - u64 l = min_t(u64, len, mapped_size); + shash->tfm = fs_info->csum_shash; + crypto_shash_init(shash); + crypto_shash_update(shash, kaddr + BTRFS_CSUM_SIZE, + sectorsize - BTRFS_CSUM_SIZE); - crypto_shash_update(shash, p, l); - kunmap_atomic(mapped_buffer); - len -= l; - if (len == 0) - break; - index++; - BUG_ON(index >= sblock->page_count); - BUG_ON(!sblock->pagev[index]->page); - page = sblock->pagev[index]->page; - mapped_buffer = kmap_atomic(page); - mapped_size = PAGE_SIZE; - p = mapped_buffer; + for (i = 1; i < num_sectors; i++) { + kaddr = scrub_sector_get_kaddr(sblock->sectors[i]); + crypto_shash_update(shash, kaddr, sectorsize); } crypto_shash_final(shash, calculated_csum); - if (memcmp(calculated_csum, on_disk_csum, sctx->csum_size)) + if (memcmp(calculated_csum, on_disk_csum, sctx->fs_info->csum_size)) sblock->checksum_error = 1; return sblock->header_error || sblock->checksum_error; @@ -1911,84 +2075,36 @@ static int scrub_checksum_super(struct scrub_block *sblock) struct btrfs_fs_info *fs_info = sctx->fs_info; SHASH_DESC_ON_STACK(shash, fs_info->csum_shash); u8 calculated_csum[BTRFS_CSUM_SIZE]; - u8 on_disk_csum[BTRFS_CSUM_SIZE]; - struct page *page; - void *mapped_buffer; - u64 mapped_size; - void *p; + struct scrub_sector *sector; + char *kaddr; int fail_gen = 0; int fail_cor = 0; - u64 len; - int index; - - shash->tfm = fs_info->csum_shash; - crypto_shash_init(shash); - BUG_ON(sblock->page_count < 1); - page = sblock->pagev[0]->page; - mapped_buffer = kmap_atomic(page); - s = (struct btrfs_super_block *)mapped_buffer; - memcpy(on_disk_csum, s->csum, sctx->csum_size); + BUG_ON(sblock->sector_count < 1); + sector = sblock->sectors[0]; + kaddr = scrub_sector_get_kaddr(sector); + s = (struct btrfs_super_block *)kaddr; - if (sblock->pagev[0]->logical != btrfs_super_bytenr(s)) + if (sblock->logical != btrfs_super_bytenr(s)) ++fail_cor; - if (sblock->pagev[0]->generation != btrfs_super_generation(s)) + if (sector->generation != btrfs_super_generation(s)) ++fail_gen; - if (!scrub_check_fsid(s->fsid, sblock->pagev[0])) + if (!scrub_check_fsid(s->fsid, sector)) ++fail_cor; - len = BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE; - mapped_size = PAGE_SIZE - BTRFS_CSUM_SIZE; - p = ((u8 *)mapped_buffer) + BTRFS_CSUM_SIZE; - index = 0; - for (;;) { - u64 l = min_t(u64, len, mapped_size); - - crypto_shash_update(shash, p, l); - kunmap_atomic(mapped_buffer); - len -= l; - if (len == 0) - break; - index++; - BUG_ON(index >= sblock->page_count); - BUG_ON(!sblock->pagev[index]->page); - page = sblock->pagev[index]->page; - mapped_buffer = kmap_atomic(page); - mapped_size = PAGE_SIZE; - p = mapped_buffer; - } + shash->tfm = fs_info->csum_shash; + crypto_shash_init(shash); + crypto_shash_digest(shash, kaddr + BTRFS_CSUM_SIZE, + BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE, calculated_csum); - crypto_shash_final(shash, calculated_csum); - if (memcmp(calculated_csum, on_disk_csum, sctx->csum_size)) + if (memcmp(calculated_csum, s->csum, sctx->fs_info->csum_size)) ++fail_cor; - if (fail_cor + fail_gen) { - /* - * if we find an error in a super block, we just report it. - * They will get written with the next transaction commit - * anyway - */ - spin_lock(&sctx->stat_lock); - ++sctx->stat.super_errors; - spin_unlock(&sctx->stat_lock); - if (fail_cor) - btrfs_dev_stat_inc_and_print(sblock->pagev[0]->dev, - BTRFS_DEV_STAT_CORRUPTION_ERRS); - else - btrfs_dev_stat_inc_and_print(sblock->pagev[0]->dev, - BTRFS_DEV_STAT_GENERATION_ERRS); - } - return fail_cor + fail_gen; } -static void scrub_block_get(struct scrub_block *sblock) -{ - refcount_inc(&sblock->refs); -} - static void scrub_block_put(struct scrub_block *sblock) { if (refcount_dec_and_test(&sblock->refs)) { @@ -1997,24 +2113,86 @@ static void scrub_block_put(struct scrub_block *sblock) if (sblock->sparity) scrub_parity_put(sblock->sparity); - for (i = 0; i < sblock->page_count; i++) - scrub_page_put(sblock->pagev[i]); + for (i = 0; i < sblock->sector_count; i++) + scrub_sector_put(sblock->sectors[i]); + for (i = 0; i < DIV_ROUND_UP(sblock->len, PAGE_SIZE); i++) { + if (sblock->pages[i]) { + detach_scrub_page_private(sblock->pages[i]); + __free_page(sblock->pages[i]); + } + } kfree(sblock); } } -static void scrub_page_get(struct scrub_page *spage) +static void scrub_sector_get(struct scrub_sector *sector) +{ + atomic_inc(§or->refs); +} + +static void scrub_sector_put(struct scrub_sector *sector) { - atomic_inc(&spage->refs); + if (atomic_dec_and_test(§or->refs)) + kfree(sector); } -static void scrub_page_put(struct scrub_page *spage) +/* + * Throttling of IO submission, bandwidth-limit based, the timeslice is 1 + * second. Limit can be set via /sys/fs/UUID/devinfo/devid/scrub_speed_max. + */ +static void scrub_throttle(struct scrub_ctx *sctx) { - if (atomic_dec_and_test(&spage->refs)) { - if (spage->page) - __free_page(spage->page); - kfree(spage); + const int time_slice = 1000; + struct scrub_bio *sbio; + struct btrfs_device *device; + s64 delta; + ktime_t now; + u32 div; + u64 bwlimit; + + sbio = sctx->bios[sctx->curr]; + device = sbio->dev; + bwlimit = READ_ONCE(device->scrub_speed_max); + if (bwlimit == 0) + return; + + /* + * Slice is divided into intervals when the IO is submitted, adjust by + * bwlimit and maximum of 64 intervals. + */ + div = max_t(u32, 1, (u32)(bwlimit / (16 * 1024 * 1024))); + div = min_t(u32, 64, div); + + /* Start new epoch, set deadline */ + now = ktime_get(); + if (sctx->throttle_deadline == 0) { + sctx->throttle_deadline = ktime_add_ms(now, time_slice / div); + sctx->throttle_sent = 0; + } + + /* Still in the time to send? */ + if (ktime_before(now, sctx->throttle_deadline)) { + /* If current bio is within the limit, send it */ + sctx->throttle_sent += sbio->bio->bi_iter.bi_size; + if (sctx->throttle_sent <= div_u64(bwlimit, div)) + return; + + /* We're over the limit, sleep until the rest of the slice */ + delta = ktime_ms_delta(sctx->throttle_deadline, now); + } else { + /* New request after deadline, start new epoch */ + delta = 0; + } + + if (delta) { + long timeout; + + timeout = div_u64(delta * HZ, 1000); + schedule_timeout_interruptible(timeout); } + + /* Next call will start the deadline period */ + sctx->throttle_deadline = 0; } static void scrub_submit(struct scrub_ctx *sctx) @@ -2024,17 +2202,21 @@ static void scrub_submit(struct scrub_ctx *sctx) if (sctx->curr == -1) return; + scrub_throttle(sctx); + sbio = sctx->bios[sctx->curr]; sctx->curr = -1; scrub_pending_bio_inc(sctx); - btrfsic_submit_bio(sbio->bio); + btrfsic_check_bio(sbio->bio); + submit_bio(sbio->bio); } -static int scrub_add_page_to_rd_bio(struct scrub_ctx *sctx, - struct scrub_page *spage) +static int scrub_add_sector_to_rd_bio(struct scrub_ctx *sctx, + struct scrub_sector *sector) { - struct scrub_block *sblock = spage->sblock; + struct scrub_block *sblock = sector->sblock; struct scrub_bio *sbio; + const u32 sectorsize = sctx->fs_info->sectorsize; int ret; again: @@ -2047,7 +2229,7 @@ again: if (sctx->curr != -1) { sctx->first_free = sctx->bios[sctx->curr]->next_free; sctx->bios[sctx->curr]->next_free = -1; - sctx->bios[sctx->curr]->page_count = 0; + sctx->bios[sctx->curr]->sector_count = 0; spin_unlock(&sctx->list_lock); } else { spin_unlock(&sctx->list_lock); @@ -2055,37 +2237,31 @@ again: } } sbio = sctx->bios[sctx->curr]; - if (sbio->page_count == 0) { - struct bio *bio; - - sbio->physical = spage->physical; - sbio->logical = spage->logical; - sbio->dev = spage->dev; - bio = sbio->bio; - if (!bio) { - bio = btrfs_io_bio_alloc(sctx->pages_per_rd_bio); - sbio->bio = bio; + if (sbio->sector_count == 0) { + sbio->physical = sblock->physical + sector->offset; + sbio->logical = sblock->logical + sector->offset; + sbio->dev = sblock->dev; + if (!sbio->bio) { + sbio->bio = bio_alloc(sbio->dev->bdev, sctx->sectors_per_bio, + REQ_OP_READ, GFP_NOFS); } - - bio->bi_private = sbio; - bio->bi_end_io = scrub_bio_end_io; - bio_set_dev(bio, sbio->dev->bdev); - bio->bi_iter.bi_sector = sbio->physical >> 9; - bio->bi_opf = REQ_OP_READ; + sbio->bio->bi_private = sbio; + sbio->bio->bi_end_io = scrub_bio_end_io; + sbio->bio->bi_iter.bi_sector = sbio->physical >> 9; sbio->status = 0; - } else if (sbio->physical + sbio->page_count * PAGE_SIZE != - spage->physical || - sbio->logical + sbio->page_count * PAGE_SIZE != - spage->logical || - sbio->dev != spage->dev) { + } else if (sbio->physical + sbio->sector_count * sectorsize != + sblock->physical + sector->offset || + sbio->logical + sbio->sector_count * sectorsize != + sblock->logical + sector->offset || + sbio->dev != sblock->dev) { scrub_submit(sctx); goto again; } - sbio->pagev[sbio->page_count] = spage; - ret = bio_add_page(sbio->bio, spage->page, PAGE_SIZE, 0); - if (ret != PAGE_SIZE) { - if (sbio->page_count < 1) { + sbio->sectors[sbio->sector_count] = sector; + ret = bio_add_scrub_sector(sbio->bio, sector, sectorsize); + if (ret != sectorsize) { + if (sbio->sector_count < 1) { bio_put(sbio->bio); sbio->bio = NULL; return -EIO; @@ -2095,9 +2271,9 @@ again: } scrub_block_get(sblock); /* one for the page added to the bio */ - atomic_inc(&sblock->outstanding_pages); - sbio->page_count++; - if (sbio->page_count == sctx->pages_per_rd_bio) + atomic_inc(&sblock->outstanding_sectors); + sbio->sector_count++; + if (sbio->sector_count == sctx->sectors_per_bio) scrub_submit(sctx); return 0; @@ -2108,15 +2284,16 @@ static void scrub_missing_raid56_end_io(struct bio *bio) struct scrub_block *sblock = bio->bi_private; struct btrfs_fs_info *fs_info = sblock->sctx->fs_info; + btrfs_bio_counter_dec(fs_info); if (bio->bi_status) sblock->no_io_error_seen = 0; bio_put(bio); - btrfs_queue_work(fs_info->scrub_workers, &sblock->work); + queue_work(fs_info->scrub_workers, &sblock->work); } -static void scrub_missing_raid56_worker(struct btrfs_work *work) +static void scrub_missing_raid56_worker(struct work_struct *work) { struct scrub_block *sblock = container_of(work, struct scrub_block, work); struct scrub_ctx *sctx = sblock->sctx; @@ -2124,8 +2301,8 @@ static void scrub_missing_raid56_worker(struct btrfs_work *work) u64 logical; struct btrfs_device *dev; - logical = sblock->pagev[0]->logical; - dev = sblock->pagev[0]->dev; + logical = sblock->logical; + dev = sblock->dev; if (sblock->no_io_error_seen) scrub_recheck_block_checksum(sblock); @@ -2162,9 +2339,9 @@ static void scrub_missing_raid56_pages(struct scrub_block *sblock) { struct scrub_ctx *sctx = sblock->sctx; struct btrfs_fs_info *fs_info = sctx->fs_info; - u64 length = sblock->page_count * PAGE_SIZE; - u64 logical = sblock->pagev[0]->logical; - struct btrfs_bio *bbio = NULL; + u64 length = sblock->sector_count << fs_info->sectorsize_bits; + u64 logical = sblock->logical; + struct btrfs_io_context *bioc = NULL; struct bio *bio; struct btrfs_raid_bio *rbio; int ret; @@ -2172,61 +2349,66 @@ static void scrub_missing_raid56_pages(struct scrub_block *sblock) btrfs_bio_counter_inc_blocked(fs_info); ret = btrfs_map_sblock(fs_info, BTRFS_MAP_GET_READ_MIRRORS, logical, - &length, &bbio); - if (ret || !bbio || !bbio->raid_map) - goto bbio_out; + &length, &bioc); + if (ret || !bioc || !bioc->raid_map) + goto bioc_out; if (WARN_ON(!sctx->is_dev_replace || - !(bbio->map_type & BTRFS_BLOCK_GROUP_RAID56_MASK))) { + !(bioc->map_type & BTRFS_BLOCK_GROUP_RAID56_MASK))) { /* * We shouldn't be scrubbing a missing device. Even for dev * replace, we should only get here for RAID 5/6. We either * managed to mount something with no mirrors remaining or - * there's a bug in scrub_remap_extent()/btrfs_map_block(). + * there's a bug in scrub_find_good_copy()/btrfs_map_block(). */ - goto bbio_out; + goto bioc_out; } - bio = btrfs_io_bio_alloc(0); + bio = bio_alloc(NULL, BIO_MAX_VECS, REQ_OP_READ, GFP_NOFS); bio->bi_iter.bi_sector = logical >> 9; bio->bi_private = sblock; bio->bi_end_io = scrub_missing_raid56_end_io; - rbio = raid56_alloc_missing_rbio(fs_info, bio, bbio, length); + rbio = raid56_alloc_missing_rbio(bio, bioc); if (!rbio) goto rbio_out; - for (i = 0; i < sblock->page_count; i++) { - struct scrub_page *spage = sblock->pagev[i]; + for (i = 0; i < sblock->sector_count; i++) { + struct scrub_sector *sector = sblock->sectors[i]; - raid56_add_scrub_pages(rbio, spage->page, spage->logical); + raid56_add_scrub_pages(rbio, scrub_sector_get_page(sector), + scrub_sector_get_page_offset(sector), + sector->offset + sector->sblock->logical); } - btrfs_init_work(&sblock->work, scrub_missing_raid56_worker, NULL, NULL); + INIT_WORK(&sblock->work, scrub_missing_raid56_worker); scrub_block_get(sblock); scrub_pending_bio_inc(sctx); raid56_submit_missing_rbio(rbio); + btrfs_put_bioc(bioc); return; rbio_out: bio_put(bio); -bbio_out: +bioc_out: btrfs_bio_counter_dec(fs_info); - btrfs_put_bbio(bbio); + btrfs_put_bioc(bioc); spin_lock(&sctx->stat_lock); sctx->stat.malloc_errors++; spin_unlock(&sctx->stat_lock); } -static int scrub_pages(struct scrub_ctx *sctx, u64 logical, u64 len, +static int scrub_sectors(struct scrub_ctx *sctx, u64 logical, u32 len, u64 physical, struct btrfs_device *dev, u64 flags, - u64 gen, int mirror_num, u8 *csum, int force, + u64 gen, int mirror_num, u8 *csum, u64 physical_for_dev_replace) { struct scrub_block *sblock; + const u32 sectorsize = sctx->fs_info->sectorsize; int index; - sblock = kzalloc(sizeof(*sblock), GFP_KERNEL); + sblock = alloc_scrub_block(sctx, dev, logical, physical, + physical_for_dev_replace, mirror_num); if (!sblock) { spin_lock(&sctx->stat_lock); sctx->stat.malloc_errors++; @@ -2234,53 +2416,38 @@ static int scrub_pages(struct scrub_ctx *sctx, u64 logical, u64 len, return -ENOMEM; } - /* one ref inside this function, plus one for each page added to - * a bio later on */ - refcount_set(&sblock->refs, 1); - sblock->sctx = sctx; - sblock->no_io_error_seen = 1; - for (index = 0; len > 0; index++) { - struct scrub_page *spage; - u64 l = min_t(u64, len, PAGE_SIZE); + struct scrub_sector *sector; + /* + * Here we will allocate one page for one sector to scrub. + * This is fine if PAGE_SIZE == sectorsize, but will cost + * more memory for PAGE_SIZE > sectorsize case. + */ + u32 l = min(sectorsize, len); - spage = kzalloc(sizeof(*spage), GFP_KERNEL); - if (!spage) { -leave_nomem: + sector = alloc_scrub_sector(sblock, logical, GFP_KERNEL); + if (!sector) { spin_lock(&sctx->stat_lock); sctx->stat.malloc_errors++; spin_unlock(&sctx->stat_lock); scrub_block_put(sblock); return -ENOMEM; } - BUG_ON(index >= SCRUB_MAX_PAGES_PER_BLOCK); - scrub_page_get(spage); - sblock->pagev[index] = spage; - spage->sblock = sblock; - spage->dev = dev; - spage->flags = flags; - spage->generation = gen; - spage->logical = logical; - spage->physical = physical; - spage->physical_for_dev_replace = physical_for_dev_replace; - spage->mirror_num = mirror_num; + sector->flags = flags; + sector->generation = gen; if (csum) { - spage->have_csum = 1; - memcpy(spage->csum, csum, sctx->csum_size); + sector->have_csum = 1; + memcpy(sector->csum, csum, sctx->fs_info->csum_size); } else { - spage->have_csum = 0; + sector->have_csum = 0; } - sblock->page_count++; - spage->page = alloc_page(GFP_KERNEL); - if (!spage->page) - goto leave_nomem; len -= l; logical += l; physical += l; physical_for_dev_replace += l; } - WARN_ON(sblock->page_count == 0); + WARN_ON(sblock->sector_count == 0); if (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state)) { /* * This case should only be hit for RAID 5/6 device replace. See @@ -2288,18 +2455,18 @@ leave_nomem: */ scrub_missing_raid56_pages(sblock); } else { - for (index = 0; index < sblock->page_count; index++) { - struct scrub_page *spage = sblock->pagev[index]; + for (index = 0; index < sblock->sector_count; index++) { + struct scrub_sector *sector = sblock->sectors[index]; int ret; - ret = scrub_add_page_to_rd_bio(sctx, spage); + ret = scrub_add_sector_to_rd_bio(sctx, sector); if (ret) { scrub_block_put(sblock); return ret; } } - if (force) + if (flags & BTRFS_EXTENT_FLAG_SUPER) scrub_submit(sctx); } @@ -2316,31 +2483,31 @@ static void scrub_bio_end_io(struct bio *bio) sbio->status = bio->bi_status; sbio->bio = bio; - btrfs_queue_work(fs_info->scrub_workers, &sbio->work); + queue_work(fs_info->scrub_workers, &sbio->work); } -static void scrub_bio_end_io_worker(struct btrfs_work *work) +static void scrub_bio_end_io_worker(struct work_struct *work) { struct scrub_bio *sbio = container_of(work, struct scrub_bio, work); struct scrub_ctx *sctx = sbio->sctx; int i; - BUG_ON(sbio->page_count > SCRUB_PAGES_PER_RD_BIO); + ASSERT(sbio->sector_count <= SCRUB_SECTORS_PER_BIO); if (sbio->status) { - for (i = 0; i < sbio->page_count; i++) { - struct scrub_page *spage = sbio->pagev[i]; + for (i = 0; i < sbio->sector_count; i++) { + struct scrub_sector *sector = sbio->sectors[i]; - spage->io_error = 1; - spage->sblock->no_io_error_seen = 0; + sector->io_error = 1; + sector->sblock->no_io_error_seen = 0; } } - /* now complete the scrub_block items that have all pages completed */ - for (i = 0; i < sbio->page_count; i++) { - struct scrub_page *spage = sbio->pagev[i]; - struct scrub_block *sblock = spage->sblock; + /* Now complete the scrub_block items that have all pages completed */ + for (i = 0; i < sbio->sector_count; i++) { + struct scrub_sector *sector = sbio->sectors[i]; + struct scrub_block *sblock = sector->sblock; - if (atomic_dec_and_test(&sblock->outstanding_pages)) + if (atomic_dec_and_test(&sblock->outstanding_sectors)) scrub_block_complete(sblock); scrub_block_put(sblock); } @@ -2363,12 +2530,11 @@ static void scrub_bio_end_io_worker(struct btrfs_work *work) static inline void __scrub_mark_bitmap(struct scrub_parity *sparity, unsigned long *bitmap, - u64 start, u64 len) + u64 start, u32 len) { u64 offset; - u64 nsectors64; u32 nsectors; - int sectorsize = sparity->sctx->fs_info->sectorsize; + u32 sectorsize_bits = sparity->sctx->fs_info->sectorsize_bits; if (len >= sparity->stripe_len) { bitmap_set(bitmap, 0, sparity->nsectors); @@ -2377,11 +2543,8 @@ static inline void __scrub_mark_bitmap(struct scrub_parity *sparity, start -= sparity->logic_start; start = div64_u64_rem(start, sparity->stripe_len, &offset); - offset = div_u64(offset, sectorsize); - nsectors64 = div_u64(len, sectorsize); - - ASSERT(nsectors64 < UINT_MAX); - nsectors = (u32)nsectors64; + offset = offset >> sectorsize_bits; + nsectors = len >> sectorsize_bits; if (offset + nsectors <= sparity->nsectors) { bitmap_set(bitmap, offset, nsectors); @@ -2393,15 +2556,15 @@ static inline void __scrub_mark_bitmap(struct scrub_parity *sparity, } static inline void scrub_parity_mark_sectors_error(struct scrub_parity *sparity, - u64 start, u64 len) + u64 start, u32 len) { - __scrub_mark_bitmap(sparity, sparity->ebitmap, start, len); + __scrub_mark_bitmap(sparity, &sparity->ebitmap, start, len); } static inline void scrub_parity_mark_sectors_data(struct scrub_parity *sparity, - u64 start, u64 len) + u64 start, u32 len) { - __scrub_mark_bitmap(sparity, sparity->dbitmap, start, len); + __scrub_mark_bitmap(sparity, &sparity->dbitmap, start, len); } static void scrub_block_complete(struct scrub_block *sblock) @@ -2423,55 +2586,88 @@ static void scrub_block_complete(struct scrub_block *sblock) } if (sblock->sparity && corrupted && !sblock->data_corrected) { - u64 start = sblock->pagev[0]->logical; - u64 end = sblock->pagev[sblock->page_count - 1]->logical + - PAGE_SIZE; + u64 start = sblock->logical; + u64 end = sblock->logical + + sblock->sectors[sblock->sector_count - 1]->offset + + sblock->sctx->fs_info->sectorsize; + ASSERT(end - start <= U32_MAX); scrub_parity_mark_sectors_error(sblock->sparity, start, end - start); } } +static void drop_csum_range(struct scrub_ctx *sctx, struct btrfs_ordered_sum *sum) +{ + sctx->stat.csum_discards += sum->len >> sctx->fs_info->sectorsize_bits; + list_del(&sum->list); + kfree(sum); +} + +/* + * Find the desired csum for range [logical, logical + sectorsize), and store + * the csum into @csum. + * + * The search source is sctx->csum_list, which is a pre-populated list + * storing bytenr ordered csum ranges. We're responsible to cleanup any range + * that is before @logical. + * + * Return 0 if there is no csum for the range. + * Return 1 if there is csum for the range and copied to @csum. + */ static int scrub_find_csum(struct scrub_ctx *sctx, u64 logical, u8 *csum) { - struct btrfs_ordered_sum *sum = NULL; - unsigned long index; - unsigned long num_sectors; + bool found = false; while (!list_empty(&sctx->csum_list)) { + struct btrfs_ordered_sum *sum = NULL; + unsigned long index; + unsigned long num_sectors; + sum = list_first_entry(&sctx->csum_list, struct btrfs_ordered_sum, list); + /* The current csum range is beyond our range, no csum found */ if (sum->bytenr > logical) - return 0; - if (sum->bytenr + sum->len > logical) break; - ++sctx->stat.csum_discards; - list_del(&sum->list); - kfree(sum); - sum = NULL; - } - if (!sum) - return 0; + /* + * The current sum is before our bytenr, since scrub is always + * done in bytenr order, the csum will never be used anymore, + * clean it up so that later calls won't bother with the range, + * and continue search the next range. + */ + if (sum->bytenr + sum->len <= logical) { + drop_csum_range(sctx, sum); + continue; + } - index = div_u64(logical - sum->bytenr, sctx->fs_info->sectorsize); - ASSERT(index < UINT_MAX); + /* Now the csum range covers our bytenr, copy the csum */ + found = true; + index = (logical - sum->bytenr) >> sctx->fs_info->sectorsize_bits; + num_sectors = sum->len >> sctx->fs_info->sectorsize_bits; - num_sectors = sum->len / sctx->fs_info->sectorsize; - memcpy(csum, sum->sums + index * sctx->csum_size, sctx->csum_size); - if (index == num_sectors - 1) { - list_del(&sum->list); - kfree(sum); + memcpy(csum, sum->sums + index * sctx->fs_info->csum_size, + sctx->fs_info->csum_size); + + /* Cleanup the range if we're at the end of the csum range */ + if (index == num_sectors - 1) + drop_csum_range(sctx, sum); + break; } + if (!found) + return 0; return 1; } /* scrub extent tries to collect up to 64 kB for each bio */ static int scrub_extent(struct scrub_ctx *sctx, struct map_lookup *map, - u64 logical, u64 len, + u64 logical, u32 len, u64 physical, struct btrfs_device *dev, u64 flags, - u64 gen, int mirror_num, u64 physical_for_dev_replace) + u64 gen, int mirror_num) { + struct btrfs_device *src_dev = dev; + u64 src_physical = physical; + int src_mirror = mirror_num; int ret; u8 csum[BTRFS_CSUM_SIZE]; u32 blocksize; @@ -2499,8 +2695,20 @@ static int scrub_extent(struct scrub_ctx *sctx, struct map_lookup *map, WARN_ON(1); } + /* + * For dev-replace case, we can have @dev being a missing device. + * Regular scrub will avoid its execution on missing device at all, + * as that would trigger tons of read error. + * + * Reading from missing device will cause read error counts to + * increase unnecessarily. + * So here we change the read source to a good mirror. + */ + if (sctx->is_dev_replace && !dev->bdev) + scrub_find_good_copy(sctx->fs_info, logical, len, &src_physical, + &src_dev, &src_mirror); while (len) { - u64 l = min_t(u64, len, blocksize); + u32 l = min(len, blocksize); int have_csum = 0; if (flags & BTRFS_EXTENT_FLAG_DATA) { @@ -2509,29 +2717,32 @@ static int scrub_extent(struct scrub_ctx *sctx, struct map_lookup *map, if (have_csum == 0) ++sctx->stat.no_csum; } - ret = scrub_pages(sctx, logical, l, physical, dev, flags, gen, - mirror_num, have_csum ? csum : NULL, 0, - physical_for_dev_replace); + ret = scrub_sectors(sctx, logical, l, src_physical, src_dev, + flags, gen, src_mirror, + have_csum ? csum : NULL, physical); if (ret) return ret; len -= l; logical += l; physical += l; - physical_for_dev_replace += l; + src_physical += l; } return 0; } -static int scrub_pages_for_parity(struct scrub_parity *sparity, - u64 logical, u64 len, +static int scrub_sectors_for_parity(struct scrub_parity *sparity, + u64 logical, u32 len, u64 physical, struct btrfs_device *dev, u64 flags, u64 gen, int mirror_num, u8 *csum) { struct scrub_ctx *sctx = sparity->sctx; struct scrub_block *sblock; + const u32 sectorsize = sctx->fs_info->sectorsize; int index; - sblock = kzalloc(sizeof(*sblock), GFP_KERNEL); + ASSERT(IS_ALIGNED(len, sectorsize)); + + sblock = alloc_scrub_block(sctx, dev, logical, physical, physical, mirror_num); if (!sblock) { spin_lock(&sctx->stat_lock); sctx->stat.malloc_errors++; @@ -2539,75 +2750,58 @@ static int scrub_pages_for_parity(struct scrub_parity *sparity, return -ENOMEM; } - /* one ref inside this function, plus one for each page added to - * a bio later on */ - refcount_set(&sblock->refs, 1); - sblock->sctx = sctx; - sblock->no_io_error_seen = 1; sblock->sparity = sparity; scrub_parity_get(sparity); for (index = 0; len > 0; index++) { - struct scrub_page *spage; - u64 l = min_t(u64, len, PAGE_SIZE); + struct scrub_sector *sector; - spage = kzalloc(sizeof(*spage), GFP_KERNEL); - if (!spage) { -leave_nomem: + sector = alloc_scrub_sector(sblock, logical, GFP_KERNEL); + if (!sector) { spin_lock(&sctx->stat_lock); sctx->stat.malloc_errors++; spin_unlock(&sctx->stat_lock); scrub_block_put(sblock); return -ENOMEM; } - BUG_ON(index >= SCRUB_MAX_PAGES_PER_BLOCK); - /* For scrub block */ - scrub_page_get(spage); - sblock->pagev[index] = spage; + sblock->sectors[index] = sector; /* For scrub parity */ - scrub_page_get(spage); - list_add_tail(&spage->list, &sparity->spages); - spage->sblock = sblock; - spage->dev = dev; - spage->flags = flags; - spage->generation = gen; - spage->logical = logical; - spage->physical = physical; - spage->mirror_num = mirror_num; + scrub_sector_get(sector); + list_add_tail(§or->list, &sparity->sectors_list); + sector->flags = flags; + sector->generation = gen; if (csum) { - spage->have_csum = 1; - memcpy(spage->csum, csum, sctx->csum_size); + sector->have_csum = 1; + memcpy(sector->csum, csum, sctx->fs_info->csum_size); } else { - spage->have_csum = 0; + sector->have_csum = 0; } - sblock->page_count++; - spage->page = alloc_page(GFP_KERNEL); - if (!spage->page) - goto leave_nomem; - len -= l; - logical += l; - physical += l; + + /* Iterate over the stripe range in sectorsize steps */ + len -= sectorsize; + logical += sectorsize; + physical += sectorsize; } - WARN_ON(sblock->page_count == 0); - for (index = 0; index < sblock->page_count; index++) { - struct scrub_page *spage = sblock->pagev[index]; + WARN_ON(sblock->sector_count == 0); + for (index = 0; index < sblock->sector_count; index++) { + struct scrub_sector *sector = sblock->sectors[index]; int ret; - ret = scrub_add_page_to_rd_bio(sctx, spage); + ret = scrub_add_sector_to_rd_bio(sctx, sector); if (ret) { scrub_block_put(sblock); return ret; } } - /* last one frees, either here or in bio completion for last page */ + /* Last one frees, either here or in bio completion for last sector */ scrub_block_put(sblock); return 0; } static int scrub_extent_for_parity(struct scrub_parity *sparity, - u64 logical, u64 len, + u64 logical, u32 len, u64 physical, struct btrfs_device *dev, u64 flags, u64 gen, int mirror_num) { @@ -2631,7 +2825,7 @@ static int scrub_extent_for_parity(struct scrub_parity *sparity, } while (len) { - u64 l = min_t(u64, len, blocksize); + u32 l = min(len, blocksize); int have_csum = 0; if (flags & BTRFS_EXTENT_FLAG_DATA) { @@ -2640,7 +2834,7 @@ static int scrub_extent_for_parity(struct scrub_parity *sparity, if (have_csum == 0) goto skip; } - ret = scrub_pages_for_parity(sparity, logical, l, physical, dev, + ret = scrub_sectors_for_parity(sparity, logical, l, physical, dev, flags, gen, mirror_num, have_csum ? csum : NULL); if (ret) @@ -2700,10 +2894,10 @@ static int get_raid56_logic_offset(u64 physical, int num, static void scrub_free_parity(struct scrub_parity *sparity) { struct scrub_ctx *sctx = sparity->sctx; - struct scrub_page *curr, *next; + struct scrub_sector *curr, *next; int nbits; - nbits = bitmap_weight(sparity->ebitmap, sparity->nsectors); + nbits = bitmap_weight(&sparity->ebitmap, sparity->nsectors); if (nbits) { spin_lock(&sctx->stat_lock); sctx->stat.read_errors += nbits; @@ -2711,38 +2905,38 @@ static void scrub_free_parity(struct scrub_parity *sparity) spin_unlock(&sctx->stat_lock); } - list_for_each_entry_safe(curr, next, &sparity->spages, list) { + list_for_each_entry_safe(curr, next, &sparity->sectors_list, list) { list_del_init(&curr->list); - scrub_page_put(curr); + scrub_sector_put(curr); } kfree(sparity); } -static void scrub_parity_bio_endio_worker(struct btrfs_work *work) +static void scrub_parity_bio_endio_worker(struct work_struct *work) { struct scrub_parity *sparity = container_of(work, struct scrub_parity, work); struct scrub_ctx *sctx = sparity->sctx; + btrfs_bio_counter_dec(sctx->fs_info); scrub_free_parity(sparity); scrub_pending_bio_dec(sctx); } static void scrub_parity_bio_endio(struct bio *bio) { - struct scrub_parity *sparity = (struct scrub_parity *)bio->bi_private; + struct scrub_parity *sparity = bio->bi_private; struct btrfs_fs_info *fs_info = sparity->sctx->fs_info; if (bio->bi_status) - bitmap_or(sparity->ebitmap, sparity->ebitmap, sparity->dbitmap, - sparity->nsectors); + bitmap_or(&sparity->ebitmap, &sparity->ebitmap, + &sparity->dbitmap, sparity->nsectors); bio_put(bio); - btrfs_init_work(&sparity->work, scrub_parity_bio_endio_worker, NULL, - NULL); - btrfs_queue_work(fs_info->scrub_parity_workers, &sparity->work); + INIT_WORK(&sparity->work, scrub_parity_bio_endio_worker); + queue_work(fs_info->scrub_parity_workers, &sparity->work); } static void scrub_parity_check_and_repair(struct scrub_parity *sparity) @@ -2751,31 +2945,32 @@ static void scrub_parity_check_and_repair(struct scrub_parity *sparity) struct btrfs_fs_info *fs_info = sctx->fs_info; struct bio *bio; struct btrfs_raid_bio *rbio; - struct btrfs_bio *bbio = NULL; + struct btrfs_io_context *bioc = NULL; u64 length; int ret; - if (!bitmap_andnot(sparity->dbitmap, sparity->dbitmap, sparity->ebitmap, - sparity->nsectors)) + if (!bitmap_andnot(&sparity->dbitmap, &sparity->dbitmap, + &sparity->ebitmap, sparity->nsectors)) goto out; length = sparity->logic_end - sparity->logic_start; btrfs_bio_counter_inc_blocked(fs_info); ret = btrfs_map_sblock(fs_info, BTRFS_MAP_WRITE, sparity->logic_start, - &length, &bbio); - if (ret || !bbio || !bbio->raid_map) - goto bbio_out; + &length, &bioc); + if (ret || !bioc || !bioc->raid_map) + goto bioc_out; - bio = btrfs_io_bio_alloc(0); + bio = bio_alloc(NULL, BIO_MAX_VECS, REQ_OP_READ, GFP_NOFS); bio->bi_iter.bi_sector = sparity->logic_start >> 9; bio->bi_private = sparity; bio->bi_end_io = scrub_parity_bio_endio; - rbio = raid56_parity_alloc_scrub_rbio(fs_info, bio, bbio, - length, sparity->scrub_dev, - sparity->dbitmap, + rbio = raid56_parity_alloc_scrub_rbio(bio, bioc, + sparity->scrub_dev, + &sparity->dbitmap, sparity->nsectors); + btrfs_put_bioc(bioc); if (!rbio) goto rbio_out; @@ -2785,10 +2980,9 @@ static void scrub_parity_check_and_repair(struct scrub_parity *sparity) rbio_out: bio_put(bio); -bbio_out: +bioc_out: btrfs_bio_counter_dec(fs_info); - btrfs_put_bbio(bbio); - bitmap_or(sparity->ebitmap, sparity->ebitmap, sparity->dbitmap, + bitmap_or(&sparity->ebitmap, &sparity->ebitmap, &sparity->dbitmap, sparity->nsectors); spin_lock(&sctx->stat_lock); sctx->stat.malloc_errors++; @@ -2797,11 +2991,6 @@ out: scrub_free_parity(sparity); } -static inline int scrub_calc_parity_bitmap_len(int nsectors) -{ - return DIV_ROUND_UP(nsectors, BITS_PER_LONG) * sizeof(long); -} - static void scrub_parity_get(struct scrub_parity *sparity) { refcount_inc(&sparity->refs); @@ -2815,46 +3004,287 @@ static void scrub_parity_put(struct scrub_parity *sparity) scrub_parity_check_and_repair(sparity); } +/* + * Return 0 if the extent item range covers any byte of the range. + * Return <0 if the extent item is before @search_start. + * Return >0 if the extent item is after @start_start + @search_len. + */ +static int compare_extent_item_range(struct btrfs_path *path, + u64 search_start, u64 search_len) +{ + struct btrfs_fs_info *fs_info = path->nodes[0]->fs_info; + u64 len; + struct btrfs_key key; + + btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); + ASSERT(key.type == BTRFS_EXTENT_ITEM_KEY || + key.type == BTRFS_METADATA_ITEM_KEY); + if (key.type == BTRFS_METADATA_ITEM_KEY) + len = fs_info->nodesize; + else + len = key.offset; + + if (key.objectid + len <= search_start) + return -1; + if (key.objectid >= search_start + search_len) + return 1; + return 0; +} + +/* + * Locate one extent item which covers any byte in range + * [@search_start, @search_start + @search_length) + * + * If the path is not initialized, we will initialize the search by doing + * a btrfs_search_slot(). + * If the path is already initialized, we will use the path as the initial + * slot, to avoid duplicated btrfs_search_slot() calls. + * + * NOTE: If an extent item starts before @search_start, we will still + * return the extent item. This is for data extent crossing stripe boundary. + * + * Return 0 if we found such extent item, and @path will point to the extent item. + * Return >0 if no such extent item can be found, and @path will be released. + * Return <0 if hit fatal error, and @path will be released. + */ +static int find_first_extent_item(struct btrfs_root *extent_root, + struct btrfs_path *path, + u64 search_start, u64 search_len) +{ + struct btrfs_fs_info *fs_info = extent_root->fs_info; + struct btrfs_key key; + int ret; + + /* Continue using the existing path */ + if (path->nodes[0]) + goto search_forward; + + if (btrfs_fs_incompat(fs_info, SKINNY_METADATA)) + key.type = BTRFS_METADATA_ITEM_KEY; + else + key.type = BTRFS_EXTENT_ITEM_KEY; + key.objectid = search_start; + key.offset = (u64)-1; + + ret = btrfs_search_slot(NULL, extent_root, &key, path, 0, 0); + if (ret < 0) + return ret; + + ASSERT(ret > 0); + /* + * Here we intentionally pass 0 as @min_objectid, as there could be + * an extent item starting before @search_start. + */ + ret = btrfs_previous_extent_item(extent_root, path, 0); + if (ret < 0) + return ret; + /* + * No matter whether we have found an extent item, the next loop will + * properly do every check on the key. + */ +search_forward: + while (true) { + btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); + if (key.objectid >= search_start + search_len) + break; + if (key.type != BTRFS_METADATA_ITEM_KEY && + key.type != BTRFS_EXTENT_ITEM_KEY) + goto next; + + ret = compare_extent_item_range(path, search_start, search_len); + if (ret == 0) + return ret; + if (ret > 0) + break; +next: + path->slots[0]++; + if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) { + ret = btrfs_next_leaf(extent_root, path); + if (ret) { + /* Either no more item or fatal error */ + btrfs_release_path(path); + return ret; + } + } + } + btrfs_release_path(path); + return 1; +} + +static void get_extent_info(struct btrfs_path *path, u64 *extent_start_ret, + u64 *size_ret, u64 *flags_ret, u64 *generation_ret) +{ + struct btrfs_key key; + struct btrfs_extent_item *ei; + + btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); + ASSERT(key.type == BTRFS_METADATA_ITEM_KEY || + key.type == BTRFS_EXTENT_ITEM_KEY); + *extent_start_ret = key.objectid; + if (key.type == BTRFS_METADATA_ITEM_KEY) + *size_ret = path->nodes[0]->fs_info->nodesize; + else + *size_ret = key.offset; + ei = btrfs_item_ptr(path->nodes[0], path->slots[0], struct btrfs_extent_item); + *flags_ret = btrfs_extent_flags(path->nodes[0], ei); + *generation_ret = btrfs_extent_generation(path->nodes[0], ei); +} + +static bool does_range_cross_boundary(u64 extent_start, u64 extent_len, + u64 boundary_start, u64 boudary_len) +{ + return (extent_start < boundary_start && + extent_start + extent_len > boundary_start) || + (extent_start < boundary_start + boudary_len && + extent_start + extent_len > boundary_start + boudary_len); +} + +static int scrub_raid56_data_stripe_for_parity(struct scrub_ctx *sctx, + struct scrub_parity *sparity, + struct map_lookup *map, + struct btrfs_device *sdev, + struct btrfs_path *path, + u64 logical) +{ + struct btrfs_fs_info *fs_info = sctx->fs_info; + struct btrfs_root *extent_root = btrfs_extent_root(fs_info, logical); + struct btrfs_root *csum_root = btrfs_csum_root(fs_info, logical); + u64 cur_logical = logical; + int ret; + + ASSERT(map->type & BTRFS_BLOCK_GROUP_RAID56_MASK); + + /* Path must not be populated */ + ASSERT(!path->nodes[0]); + + while (cur_logical < logical + map->stripe_len) { + struct btrfs_io_context *bioc = NULL; + struct btrfs_device *extent_dev; + u64 extent_start; + u64 extent_size; + u64 mapped_length; + u64 extent_flags; + u64 extent_gen; + u64 extent_physical; + u64 extent_mirror_num; + + ret = find_first_extent_item(extent_root, path, cur_logical, + logical + map->stripe_len - cur_logical); + /* No more extent item in this data stripe */ + if (ret > 0) { + ret = 0; + break; + } + if (ret < 0) + break; + get_extent_info(path, &extent_start, &extent_size, &extent_flags, + &extent_gen); + + /* Metadata should not cross stripe boundaries */ + if ((extent_flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) && + does_range_cross_boundary(extent_start, extent_size, + logical, map->stripe_len)) { + btrfs_err(fs_info, + "scrub: tree block %llu spanning stripes, ignored. logical=%llu", + extent_start, logical); + spin_lock(&sctx->stat_lock); + sctx->stat.uncorrectable_errors++; + spin_unlock(&sctx->stat_lock); + cur_logical += extent_size; + continue; + } + + /* Skip hole range which doesn't have any extent */ + cur_logical = max(extent_start, cur_logical); + + /* Truncate the range inside this data stripe */ + extent_size = min(extent_start + extent_size, + logical + map->stripe_len) - cur_logical; + extent_start = cur_logical; + ASSERT(extent_size <= U32_MAX); + + scrub_parity_mark_sectors_data(sparity, extent_start, extent_size); + + mapped_length = extent_size; + ret = btrfs_map_block(fs_info, BTRFS_MAP_READ, extent_start, + &mapped_length, &bioc, 0); + if (!ret && (!bioc || mapped_length < extent_size)) + ret = -EIO; + if (ret) { + btrfs_put_bioc(bioc); + scrub_parity_mark_sectors_error(sparity, extent_start, + extent_size); + break; + } + extent_physical = bioc->stripes[0].physical; + extent_mirror_num = bioc->mirror_num; + extent_dev = bioc->stripes[0].dev; + btrfs_put_bioc(bioc); + + ret = btrfs_lookup_csums_range(csum_root, extent_start, + extent_start + extent_size - 1, + &sctx->csum_list, 1, false); + if (ret) { + scrub_parity_mark_sectors_error(sparity, extent_start, + extent_size); + break; + } + + ret = scrub_extent_for_parity(sparity, extent_start, + extent_size, extent_physical, + extent_dev, extent_flags, + extent_gen, extent_mirror_num); + scrub_free_csums(sctx); + + if (ret) { + scrub_parity_mark_sectors_error(sparity, extent_start, + extent_size); + break; + } + + cond_resched(); + cur_logical += extent_size; + } + btrfs_release_path(path); + return ret; +} + static noinline_for_stack int scrub_raid56_parity(struct scrub_ctx *sctx, struct map_lookup *map, struct btrfs_device *sdev, - struct btrfs_path *path, u64 logic_start, u64 logic_end) { struct btrfs_fs_info *fs_info = sctx->fs_info; - struct btrfs_root *root = fs_info->extent_root; - struct btrfs_root *csum_root = fs_info->csum_root; - struct btrfs_extent_item *extent; - struct btrfs_bio *bbio = NULL; - u64 flags; + struct btrfs_path *path; + u64 cur_logical; int ret; - int slot; - struct extent_buffer *l; - struct btrfs_key key; - u64 generation; - u64 extent_logical; - u64 extent_physical; - u64 extent_len; - u64 mapped_length; - struct btrfs_device *extent_dev; struct scrub_parity *sparity; int nsectors; - int bitmap_len; - int extent_mirror_num; - int stop_loop = 0; - nsectors = div_u64(map->stripe_len, fs_info->sectorsize); - bitmap_len = scrub_calc_parity_bitmap_len(nsectors); - sparity = kzalloc(sizeof(struct scrub_parity) + 2 * bitmap_len, - GFP_NOFS); + path = btrfs_alloc_path(); + if (!path) { + spin_lock(&sctx->stat_lock); + sctx->stat.malloc_errors++; + spin_unlock(&sctx->stat_lock); + return -ENOMEM; + } + path->search_commit_root = 1; + path->skip_locking = 1; + + ASSERT(map->stripe_len <= U32_MAX); + nsectors = map->stripe_len >> fs_info->sectorsize_bits; + ASSERT(nsectors <= BITS_PER_LONG); + sparity = kzalloc(sizeof(struct scrub_parity), GFP_NOFS); if (!sparity) { spin_lock(&sctx->stat_lock); sctx->stat.malloc_errors++; spin_unlock(&sctx->stat_lock); + btrfs_free_path(path); return -ENOMEM; } + ASSERT(map->stripe_len <= U32_MAX); sparity->stripe_len = map->stripe_len; sparity->nsectors = nsectors; sparity->sctx = sctx; @@ -2862,258 +3292,299 @@ static noinline_for_stack int scrub_raid56_parity(struct scrub_ctx *sctx, sparity->logic_start = logic_start; sparity->logic_end = logic_end; refcount_set(&sparity->refs, 1); - INIT_LIST_HEAD(&sparity->spages); - sparity->dbitmap = sparity->bitmap; - sparity->ebitmap = (void *)sparity->bitmap + bitmap_len; + INIT_LIST_HEAD(&sparity->sectors_list); ret = 0; - while (logic_start < logic_end) { - if (btrfs_fs_incompat(fs_info, SKINNY_METADATA)) - key.type = BTRFS_METADATA_ITEM_KEY; - else - key.type = BTRFS_EXTENT_ITEM_KEY; - key.objectid = logic_start; - key.offset = (u64)-1; - - ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + for (cur_logical = logic_start; cur_logical < logic_end; + cur_logical += map->stripe_len) { + ret = scrub_raid56_data_stripe_for_parity(sctx, sparity, map, + sdev, path, cur_logical); if (ret < 0) - goto out; + break; + } - if (ret > 0) { - ret = btrfs_previous_extent_item(root, path, 0); - if (ret < 0) - goto out; - if (ret > 0) { - btrfs_release_path(path); - ret = btrfs_search_slot(NULL, root, &key, - path, 0, 0); - if (ret < 0) - goto out; - } - } + scrub_parity_put(sparity); + scrub_submit(sctx); + mutex_lock(&sctx->wr_lock); + scrub_wr_submit(sctx); + mutex_unlock(&sctx->wr_lock); - stop_loop = 0; - while (1) { - u64 bytes; + btrfs_free_path(path); + return ret < 0 ? ret : 0; +} - l = path->nodes[0]; - slot = path->slots[0]; - if (slot >= btrfs_header_nritems(l)) { - ret = btrfs_next_leaf(root, path); - if (ret == 0) - continue; - if (ret < 0) - goto out; +static void sync_replace_for_zoned(struct scrub_ctx *sctx) +{ + if (!btrfs_is_zoned(sctx->fs_info)) + return; - stop_loop = 1; - break; - } - btrfs_item_key_to_cpu(l, &key, slot); + sctx->flush_all_writes = true; + scrub_submit(sctx); + mutex_lock(&sctx->wr_lock); + scrub_wr_submit(sctx); + mutex_unlock(&sctx->wr_lock); + + wait_event(sctx->list_wait, atomic_read(&sctx->bios_in_flight) == 0); +} - if (key.type != BTRFS_EXTENT_ITEM_KEY && - key.type != BTRFS_METADATA_ITEM_KEY) - goto next; +static int sync_write_pointer_for_zoned(struct scrub_ctx *sctx, u64 logical, + u64 physical, u64 physical_end) +{ + struct btrfs_fs_info *fs_info = sctx->fs_info; + int ret = 0; - if (key.type == BTRFS_METADATA_ITEM_KEY) - bytes = fs_info->nodesize; - else - bytes = key.offset; + if (!btrfs_is_zoned(fs_info)) + return 0; - if (key.objectid + bytes <= logic_start) - goto next; + wait_event(sctx->list_wait, atomic_read(&sctx->bios_in_flight) == 0); - if (key.objectid >= logic_end) { - stop_loop = 1; - break; - } + mutex_lock(&sctx->wr_lock); + if (sctx->write_pointer < physical_end) { + ret = btrfs_sync_zone_write_pointer(sctx->wr_tgtdev, logical, + physical, + sctx->write_pointer); + if (ret) + btrfs_err(fs_info, + "zoned: failed to recover write pointer"); + } + mutex_unlock(&sctx->wr_lock); + btrfs_dev_clear_zone_empty(sctx->wr_tgtdev, physical); - while (key.objectid >= logic_start + map->stripe_len) - logic_start += map->stripe_len; - - extent = btrfs_item_ptr(l, slot, - struct btrfs_extent_item); - flags = btrfs_extent_flags(l, extent); - generation = btrfs_extent_generation(l, extent); - - if ((flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) && - (key.objectid < logic_start || - key.objectid + bytes > - logic_start + map->stripe_len)) { - btrfs_err(fs_info, - "scrub: tree block %llu spanning stripes, ignored. logical=%llu", - key.objectid, logic_start); - spin_lock(&sctx->stat_lock); - sctx->stat.uncorrectable_errors++; - spin_unlock(&sctx->stat_lock); - goto next; - } -again: - extent_logical = key.objectid; - extent_len = bytes; + return ret; +} - if (extent_logical < logic_start) { - extent_len -= logic_start - extent_logical; - extent_logical = logic_start; - } +/* + * Scrub one range which can only has simple mirror based profile. + * (Including all range in SINGLE/DUP/RAID1/RAID1C*, and each stripe in + * RAID0/RAID10). + * + * Since we may need to handle a subset of block group, we need @logical_start + * and @logical_length parameter. + */ +static int scrub_simple_mirror(struct scrub_ctx *sctx, + struct btrfs_root *extent_root, + struct btrfs_root *csum_root, + struct btrfs_block_group *bg, + struct map_lookup *map, + u64 logical_start, u64 logical_length, + struct btrfs_device *device, + u64 physical, int mirror_num) +{ + struct btrfs_fs_info *fs_info = sctx->fs_info; + const u64 logical_end = logical_start + logical_length; + /* An artificial limit, inherit from old scrub behavior */ + const u32 max_length = SZ_64K; + struct btrfs_path path = { 0 }; + u64 cur_logical = logical_start; + int ret; - if (extent_logical + extent_len > - logic_start + map->stripe_len) - extent_len = logic_start + map->stripe_len - - extent_logical; - - scrub_parity_mark_sectors_data(sparity, extent_logical, - extent_len); - - mapped_length = extent_len; - bbio = NULL; - ret = btrfs_map_block(fs_info, BTRFS_MAP_READ, - extent_logical, &mapped_length, &bbio, - 0); - if (!ret) { - if (!bbio || mapped_length < extent_len) - ret = -EIO; - } - if (ret) { - btrfs_put_bbio(bbio); - goto out; - } - extent_physical = bbio->stripes[0].physical; - extent_mirror_num = bbio->mirror_num; - extent_dev = bbio->stripes[0].dev; - btrfs_put_bbio(bbio); - - ret = btrfs_lookup_csums_range(csum_root, - extent_logical, - extent_logical + extent_len - 1, - &sctx->csum_list, 1); - if (ret) - goto out; + /* The range must be inside the bg */ + ASSERT(logical_start >= bg->start && logical_end <= bg->start + bg->length); + + path.search_commit_root = 1; + path.skip_locking = 1; + /* Go through each extent items inside the logical range */ + while (cur_logical < logical_end) { + u64 extent_start; + u64 extent_len; + u64 extent_flags; + u64 extent_gen; + u64 scrub_len; - ret = scrub_extent_for_parity(sparity, extent_logical, - extent_len, - extent_physical, - extent_dev, flags, - generation, - extent_mirror_num); + /* Canceled? */ + if (atomic_read(&fs_info->scrub_cancel_req) || + atomic_read(&sctx->cancel_req)) { + ret = -ECANCELED; + break; + } + /* Paused? */ + if (atomic_read(&fs_info->scrub_pause_req)) { + /* Push queued extents */ + sctx->flush_all_writes = true; + scrub_submit(sctx); + mutex_lock(&sctx->wr_lock); + scrub_wr_submit(sctx); + mutex_unlock(&sctx->wr_lock); + wait_event(sctx->list_wait, + atomic_read(&sctx->bios_in_flight) == 0); + sctx->flush_all_writes = false; + scrub_blocked_if_needed(fs_info); + } + /* Block group removed? */ + spin_lock(&bg->lock); + if (test_bit(BLOCK_GROUP_FLAG_REMOVED, &bg->runtime_flags)) { + spin_unlock(&bg->lock); + ret = 0; + break; + } + spin_unlock(&bg->lock); - scrub_free_csums(sctx); + ret = find_first_extent_item(extent_root, &path, cur_logical, + logical_end - cur_logical); + if (ret > 0) { + /* No more extent, just update the accounting */ + sctx->stat.last_physical = physical + logical_length; + ret = 0; + break; + } + if (ret < 0) + break; + get_extent_info(&path, &extent_start, &extent_len, + &extent_flags, &extent_gen); + /* Skip hole range which doesn't have any extent */ + cur_logical = max(extent_start, cur_logical); + /* + * Scrub len has three limits: + * - Extent size limit + * - Scrub range limit + * This is especially imporatant for RAID0/RAID10 to reuse + * this function + * - Max scrub size limit + */ + scrub_len = min(min(extent_start + extent_len, + logical_end), cur_logical + max_length) - + cur_logical; + + if (extent_flags & BTRFS_EXTENT_FLAG_DATA) { + ret = btrfs_lookup_csums_range(csum_root, cur_logical, + cur_logical + scrub_len - 1, + &sctx->csum_list, 1, false); if (ret) - goto out; + break; + } + if ((extent_flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) && + does_range_cross_boundary(extent_start, extent_len, + logical_start, logical_length)) { + btrfs_err(fs_info, +"scrub: tree block %llu spanning boundaries, ignored. boundary=[%llu, %llu)", + extent_start, logical_start, logical_end); + spin_lock(&sctx->stat_lock); + sctx->stat.uncorrectable_errors++; + spin_unlock(&sctx->stat_lock); + cur_logical += scrub_len; + continue; + } + ret = scrub_extent(sctx, map, cur_logical, scrub_len, + cur_logical - logical_start + physical, + device, extent_flags, extent_gen, + mirror_num); + scrub_free_csums(sctx); + if (ret) + break; + if (sctx->is_dev_replace) + sync_replace_for_zoned(sctx); + cur_logical += scrub_len; + /* Don't hold CPU for too long time */ + cond_resched(); + } + btrfs_release_path(&path); + return ret; +} - if (extent_logical + extent_len < - key.objectid + bytes) { - logic_start += map->stripe_len; +/* Calculate the full stripe length for simple stripe based profiles */ +static u64 simple_stripe_full_stripe_len(const struct map_lookup *map) +{ + ASSERT(map->type & (BTRFS_BLOCK_GROUP_RAID0 | + BTRFS_BLOCK_GROUP_RAID10)); - if (logic_start >= logic_end) { - stop_loop = 1; - break; - } + return map->num_stripes / map->sub_stripes * map->stripe_len; +} - if (logic_start < key.objectid + bytes) { - cond_resched(); - goto again; - } - } -next: - path->slots[0]++; - } +/* Get the logical bytenr for the stripe */ +static u64 simple_stripe_get_logical(struct map_lookup *map, + struct btrfs_block_group *bg, + int stripe_index) +{ + ASSERT(map->type & (BTRFS_BLOCK_GROUP_RAID0 | + BTRFS_BLOCK_GROUP_RAID10)); + ASSERT(stripe_index < map->num_stripes); - btrfs_release_path(path); + /* + * (stripe_index / sub_stripes) gives how many data stripes we need to + * skip. + */ + return (stripe_index / map->sub_stripes) * map->stripe_len + bg->start; +} - if (stop_loop) - break; +/* Get the mirror number for the stripe */ +static int simple_stripe_mirror_num(struct map_lookup *map, int stripe_index) +{ + ASSERT(map->type & (BTRFS_BLOCK_GROUP_RAID0 | + BTRFS_BLOCK_GROUP_RAID10)); + ASSERT(stripe_index < map->num_stripes); - logic_start += map->stripe_len; - } -out: - if (ret < 0) - scrub_parity_mark_sectors_error(sparity, logic_start, - logic_end - logic_start); - scrub_parity_put(sparity); - scrub_submit(sctx); - mutex_lock(&sctx->wr_lock); - scrub_wr_submit(sctx); - mutex_unlock(&sctx->wr_lock); + /* For RAID0, it's fixed to 1, for RAID10 it's 0,1,0,1... */ + return stripe_index % map->sub_stripes + 1; +} - btrfs_release_path(path); - return ret < 0 ? ret : 0; +static int scrub_simple_stripe(struct scrub_ctx *sctx, + struct btrfs_root *extent_root, + struct btrfs_root *csum_root, + struct btrfs_block_group *bg, + struct map_lookup *map, + struct btrfs_device *device, + int stripe_index) +{ + const u64 logical_increment = simple_stripe_full_stripe_len(map); + const u64 orig_logical = simple_stripe_get_logical(map, bg, stripe_index); + const u64 orig_physical = map->stripes[stripe_index].physical; + const int mirror_num = simple_stripe_mirror_num(map, stripe_index); + u64 cur_logical = orig_logical; + u64 cur_physical = orig_physical; + int ret = 0; + + while (cur_logical < bg->start + bg->length) { + /* + * Inside each stripe, RAID0 is just SINGLE, and RAID10 is + * just RAID1, so we can reuse scrub_simple_mirror() to scrub + * this stripe. + */ + ret = scrub_simple_mirror(sctx, extent_root, csum_root, bg, map, + cur_logical, map->stripe_len, device, + cur_physical, mirror_num); + if (ret) + return ret; + /* Skip to next stripe which belongs to the target device */ + cur_logical += logical_increment; + /* For physical offset, we just go to next stripe */ + cur_physical += map->stripe_len; + } + return ret; } static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx, - struct map_lookup *map, + struct btrfs_block_group *bg, + struct extent_map *em, struct btrfs_device *scrub_dev, - int num, u64 base, u64 length) + int stripe_index) { - struct btrfs_path *path, *ppath; + struct btrfs_path *path; struct btrfs_fs_info *fs_info = sctx->fs_info; - struct btrfs_root *root = fs_info->extent_root; - struct btrfs_root *csum_root = fs_info->csum_root; - struct btrfs_extent_item *extent; + struct btrfs_root *root; + struct btrfs_root *csum_root; struct blk_plug plug; - u64 flags; + struct map_lookup *map = em->map_lookup; + const u64 profile = map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK; + const u64 chunk_logical = bg->start; int ret; - int slot; - u64 nstripes; - struct extent_buffer *l; - u64 physical; + u64 physical = map->stripes[stripe_index].physical; + const u64 dev_stripe_len = btrfs_calc_stripe_length(em); + const u64 physical_end = physical + dev_stripe_len; u64 logical; u64 logic_end; - u64 physical_end; - u64 generation; - int mirror_num; - struct reada_control *reada1; - struct reada_control *reada2; - struct btrfs_key key; - struct btrfs_key key_end; - u64 increment = map->stripe_len; + /* The logical increment after finishing one stripe */ + u64 increment; + /* Offset inside the chunk */ u64 offset; - u64 extent_logical; - u64 extent_physical; - u64 extent_len; u64 stripe_logical; u64 stripe_end; - struct btrfs_device *extent_dev; - int extent_mirror_num; int stop_loop = 0; - physical = map->stripes[num].physical; - offset = 0; - nstripes = div64_u64(length, map->stripe_len); - if (map->type & BTRFS_BLOCK_GROUP_RAID0) { - offset = map->stripe_len * num; - increment = map->stripe_len * map->num_stripes; - mirror_num = 1; - } else if (map->type & BTRFS_BLOCK_GROUP_RAID10) { - int factor = map->num_stripes / map->sub_stripes; - offset = map->stripe_len * (num / map->sub_stripes); - increment = map->stripe_len * factor; - mirror_num = num % map->sub_stripes + 1; - } else if (map->type & BTRFS_BLOCK_GROUP_RAID1_MASK) { - increment = map->stripe_len; - mirror_num = num % map->num_stripes + 1; - } else if (map->type & BTRFS_BLOCK_GROUP_DUP) { - increment = map->stripe_len; - mirror_num = num % map->num_stripes + 1; - } else if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) { - get_raid56_logic_offset(physical, num, map, &offset, NULL); - increment = map->stripe_len * nr_data_stripes(map); - mirror_num = 1; - } else { - increment = map->stripe_len; - mirror_num = 1; - } - path = btrfs_alloc_path(); if (!path) return -ENOMEM; - ppath = btrfs_alloc_path(); - if (!ppath) { - btrfs_free_path(path); - return -ENOMEM; - } - /* * work on commit root. The related disk blocks are static as * long as COW is applied. This means, it is save to rewrite @@ -3121,49 +3592,14 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx, */ path->search_commit_root = 1; path->skip_locking = 1; + path->reada = READA_FORWARD; - ppath->search_commit_root = 1; - ppath->skip_locking = 1; - /* - * trigger the readahead for extent tree csum tree and wait for - * completion. During readahead, the scrub is officially paused - * to not hold off transaction commits - */ - logical = base + offset; - physical_end = physical + nstripes * map->stripe_len; - if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) { - get_raid56_logic_offset(physical_end, num, - map, &logic_end, NULL); - logic_end += base; - } else { - logic_end = logical + increment * nstripes; - } wait_event(sctx->list_wait, atomic_read(&sctx->bios_in_flight) == 0); scrub_blocked_if_needed(fs_info); - /* FIXME it might be better to start readahead at commit root */ - key.objectid = logical; - key.type = BTRFS_EXTENT_ITEM_KEY; - key.offset = (u64)0; - key_end.objectid = logic_end; - key_end.type = BTRFS_METADATA_ITEM_KEY; - key_end.offset = (u64)-1; - reada1 = btrfs_reada_add(root, &key, &key_end); - - key.objectid = BTRFS_EXTENT_CSUM_OBJECTID; - key.type = BTRFS_EXTENT_CSUM_KEY; - key.offset = logical; - key_end.objectid = BTRFS_EXTENT_CSUM_OBJECTID; - key_end.type = BTRFS_EXTENT_CSUM_KEY; - key_end.offset = logic_end; - reada2 = btrfs_reada_add(csum_root, &key, &key_end); - - if (!IS_ERR(reada1)) - btrfs_reada_wait(reada1); - if (!IS_ERR(reada2)) - btrfs_reada_wait(reada2); - + root = btrfs_extent_root(fs_info, bg->start); + csum_root = btrfs_csum_root(fs_info, bg->start); /* * collect all data csums for the stripe to avoid seeking during @@ -3171,229 +3607,98 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx, */ blk_start_plug(&plug); + if (sctx->is_dev_replace && + btrfs_dev_is_sequential(sctx->wr_tgtdev, physical)) { + mutex_lock(&sctx->wr_lock); + sctx->write_pointer = physical; + mutex_unlock(&sctx->wr_lock); + sctx->flush_all_writes = true; + } + /* - * now find all extents for each stripe and scrub them + * There used to be a big double loop to handle all profiles using the + * same routine, which grows larger and more gross over time. + * + * So here we handle each profile differently, so simpler profiles + * have simpler scrubbing function. */ - ret = 0; - while (physical < physical_end) { - /* - * canceled? - */ - if (atomic_read(&fs_info->scrub_cancel_req) || - atomic_read(&sctx->cancel_req)) { - ret = -ECANCELED; - goto out; - } + if (!(profile & (BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID10 | + BTRFS_BLOCK_GROUP_RAID56_MASK))) { /* - * check to see if we have to pause + * Above check rules out all complex profile, the remaining + * profiles are SINGLE|DUP|RAID1|RAID1C*, which is simple + * mirrored duplication without stripe. + * + * Only @physical and @mirror_num needs to calculated using + * @stripe_index. */ - if (atomic_read(&fs_info->scrub_pause_req)) { - /* push queued extents */ - sctx->flush_all_writes = true; - scrub_submit(sctx); - mutex_lock(&sctx->wr_lock); - scrub_wr_submit(sctx); - mutex_unlock(&sctx->wr_lock); - wait_event(sctx->list_wait, - atomic_read(&sctx->bios_in_flight) == 0); - sctx->flush_all_writes = false; - scrub_blocked_if_needed(fs_info); - } - - if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) { - ret = get_raid56_logic_offset(physical, num, map, - &logical, - &stripe_logical); - logical += base; - if (ret) { - /* it is parity strip */ - stripe_logical += base; - stripe_end = stripe_logical + increment; - ret = scrub_raid56_parity(sctx, map, scrub_dev, - ppath, stripe_logical, - stripe_end); - if (ret) - goto out; - goto skip; - } - } - - if (btrfs_fs_incompat(fs_info, SKINNY_METADATA)) - key.type = BTRFS_METADATA_ITEM_KEY; - else - key.type = BTRFS_EXTENT_ITEM_KEY; - key.objectid = logical; - key.offset = (u64)-1; - - ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); - if (ret < 0) - goto out; - - if (ret > 0) { - ret = btrfs_previous_extent_item(root, path, 0); - if (ret < 0) - goto out; - if (ret > 0) { - /* there's no smaller item, so stick with the - * larger one */ - btrfs_release_path(path); - ret = btrfs_search_slot(NULL, root, &key, - path, 0, 0); - if (ret < 0) - goto out; - } - } - - stop_loop = 0; - while (1) { - u64 bytes; - - l = path->nodes[0]; - slot = path->slots[0]; - if (slot >= btrfs_header_nritems(l)) { - ret = btrfs_next_leaf(root, path); - if (ret == 0) - continue; - if (ret < 0) - goto out; - - stop_loop = 1; - break; - } - btrfs_item_key_to_cpu(l, &key, slot); - - if (key.type != BTRFS_EXTENT_ITEM_KEY && - key.type != BTRFS_METADATA_ITEM_KEY) - goto next; - - if (key.type == BTRFS_METADATA_ITEM_KEY) - bytes = fs_info->nodesize; - else - bytes = key.offset; - - if (key.objectid + bytes <= logical) - goto next; - - if (key.objectid >= logical + map->stripe_len) { - /* out of this device extent */ - if (key.objectid >= logic_end) - stop_loop = 1; - break; - } - - extent = btrfs_item_ptr(l, slot, - struct btrfs_extent_item); - flags = btrfs_extent_flags(l, extent); - generation = btrfs_extent_generation(l, extent); - - if ((flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) && - (key.objectid < logical || - key.objectid + bytes > - logical + map->stripe_len)) { - btrfs_err(fs_info, - "scrub: tree block %llu spanning stripes, ignored. logical=%llu", - key.objectid, logical); - spin_lock(&sctx->stat_lock); - sctx->stat.uncorrectable_errors++; - spin_unlock(&sctx->stat_lock); - goto next; - } - -again: - extent_logical = key.objectid; - extent_len = bytes; - - /* - * trim extent to this stripe - */ - if (extent_logical < logical) { - extent_len -= logical - extent_logical; - extent_logical = logical; - } - if (extent_logical + extent_len > - logical + map->stripe_len) { - extent_len = logical + map->stripe_len - - extent_logical; - } + ret = scrub_simple_mirror(sctx, root, csum_root, bg, map, + bg->start, bg->length, scrub_dev, + map->stripes[stripe_index].physical, + stripe_index + 1); + offset = 0; + goto out; + } + if (profile & (BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID10)) { + ret = scrub_simple_stripe(sctx, root, csum_root, bg, map, + scrub_dev, stripe_index); + offset = map->stripe_len * (stripe_index / map->sub_stripes); + goto out; + } - extent_physical = extent_logical - logical + physical; - extent_dev = scrub_dev; - extent_mirror_num = mirror_num; - if (sctx->is_dev_replace) - scrub_remap_extent(fs_info, extent_logical, - extent_len, &extent_physical, - &extent_dev, - &extent_mirror_num); - - ret = btrfs_lookup_csums_range(csum_root, - extent_logical, - extent_logical + - extent_len - 1, - &sctx->csum_list, 1); - if (ret) - goto out; + /* Only RAID56 goes through the old code */ + ASSERT(map->type & BTRFS_BLOCK_GROUP_RAID56_MASK); + ret = 0; - ret = scrub_extent(sctx, map, extent_logical, extent_len, - extent_physical, extent_dev, flags, - generation, extent_mirror_num, - extent_logical - logical + physical); + /* Calculate the logical end of the stripe */ + get_raid56_logic_offset(physical_end, stripe_index, + map, &logic_end, NULL); + logic_end += chunk_logical; - scrub_free_csums(sctx); + /* Initialize @offset in case we need to go to out: label */ + get_raid56_logic_offset(physical, stripe_index, map, &offset, NULL); + increment = map->stripe_len * nr_data_stripes(map); + /* + * Due to the rotation, for RAID56 it's better to iterate each stripe + * using their physical offset. + */ + while (physical < physical_end) { + ret = get_raid56_logic_offset(physical, stripe_index, map, + &logical, &stripe_logical); + logical += chunk_logical; + if (ret) { + /* it is parity strip */ + stripe_logical += chunk_logical; + stripe_end = stripe_logical + increment; + ret = scrub_raid56_parity(sctx, map, scrub_dev, + stripe_logical, + stripe_end); if (ret) goto out; + goto next; + } - if (extent_logical + extent_len < - key.objectid + bytes) { - if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) { - /* - * loop until we find next data stripe - * or we have finished all stripes. - */ -loop: - physical += map->stripe_len; - ret = get_raid56_logic_offset(physical, - num, map, &logical, - &stripe_logical); - logical += base; - - if (ret && physical < physical_end) { - stripe_logical += base; - stripe_end = stripe_logical + - increment; - ret = scrub_raid56_parity(sctx, - map, scrub_dev, ppath, - stripe_logical, - stripe_end); - if (ret) - goto out; - goto loop; - } - } else { - physical += map->stripe_len; - logical += increment; - } - if (logical < key.objectid + bytes) { - cond_resched(); - goto again; - } - - if (physical >= physical_end) { - stop_loop = 1; - break; - } - } + /* + * Now we're at a data stripe, scrub each extents in the range. + * + * At this stage, if we ignore the repair part, inside each data + * stripe it is no different than SINGLE profile. + * We can reuse scrub_simple_mirror() here, as the repair part + * is still based on @mirror_num. + */ + ret = scrub_simple_mirror(sctx, root, csum_root, bg, map, + logical, map->stripe_len, + scrub_dev, physical, 1); + if (ret < 0) + goto out; next: - path->slots[0]++; - } - btrfs_release_path(path); -skip: logical += increment; physical += map->stripe_len; spin_lock(&sctx->stat_lock); if (stop_loop) - sctx->stat.last_physical = map->stripes[num].physical + - length; + sctx->stat.last_physical = + map->stripes[stripe_index].physical + dev_stripe_len; else sctx->stat.last_physical = physical; spin_unlock(&sctx->stat_lock); @@ -3409,15 +3714,26 @@ out: blk_finish_plug(&plug); btrfs_free_path(path); - btrfs_free_path(ppath); + + if (sctx->is_dev_replace && ret >= 0) { + int ret2; + + ret2 = sync_write_pointer_for_zoned(sctx, + chunk_logical + offset, + map->stripes[stripe_index].physical, + physical_end); + if (ret2) + ret = ret2; + } + return ret < 0 ? ret : 0; } static noinline_for_stack int scrub_chunk(struct scrub_ctx *sctx, + struct btrfs_block_group *bg, struct btrfs_device *scrub_dev, - u64 chunk_offset, u64 length, u64 dev_offset, - struct btrfs_block_group *cache) + u64 dev_extent_len) { struct btrfs_fs_info *fs_info = sctx->fs_info; struct extent_map_tree *map_tree = &fs_info->mapping_tree; @@ -3427,7 +3743,7 @@ static noinline_for_stack int scrub_chunk(struct scrub_ctx *sctx, int ret = 0; read_lock(&map_tree->lock); - em = lookup_extent_mapping(map_tree, chunk_offset, 1); + em = lookup_extent_mapping(map_tree, bg->start, bg->length); read_unlock(&map_tree->lock); if (!em) { @@ -3435,26 +3751,23 @@ static noinline_for_stack int scrub_chunk(struct scrub_ctx *sctx, * Might have been an unused block group deleted by the cleaner * kthread or relocation. */ - spin_lock(&cache->lock); - if (!cache->removed) + spin_lock(&bg->lock); + if (!test_bit(BLOCK_GROUP_FLAG_REMOVED, &bg->runtime_flags)) ret = -EINVAL; - spin_unlock(&cache->lock); + spin_unlock(&bg->lock); return ret; } - - map = em->map_lookup; - if (em->start != chunk_offset) + if (em->start != bg->start) goto out; - - if (em->len < length) + if (em->len < dev_extent_len) goto out; + map = em->map_lookup; for (i = 0; i < map->num_stripes; ++i) { if (map->stripes[i].dev->bdev == scrub_dev->bdev && map->stripes[i].physical == dev_offset) { - ret = scrub_stripe(sctx, map, scrub_dev, i, - chunk_offset, length); + ret = scrub_stripe(sctx, bg, em, scrub_dev, i); if (ret) goto out; } @@ -3465,6 +3778,25 @@ out: return ret; } +static int finish_extent_writes_for_zoned(struct btrfs_root *root, + struct btrfs_block_group *cache) +{ + struct btrfs_fs_info *fs_info = cache->fs_info; + struct btrfs_trans_handle *trans; + + if (!btrfs_is_zoned(fs_info)) + return 0; + + btrfs_wait_block_group_reservations(cache); + btrfs_wait_nocow_writers(cache); + btrfs_wait_ordered_roots(fs_info, U64_MAX, cache->start, cache->length); + + trans = btrfs_join_transaction(root); + if (IS_ERR(trans)) + return PTR_ERR(trans); + return btrfs_commit_transaction(trans); +} + static noinline_for_stack int scrub_enumerate_chunks(struct scrub_ctx *sctx, struct btrfs_device *scrub_dev, u64 start, u64 end) @@ -3473,7 +3805,6 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, struct btrfs_path *path; struct btrfs_fs_info *fs_info = sctx->fs_info; struct btrfs_root *root = fs_info->dev_root; - u64 length; u64 chunk_offset; int ret = 0; int ro_set; @@ -3497,6 +3828,8 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, key.type = BTRFS_DEV_EXTENT_KEY; while (1) { + u64 dev_extent_len; + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); if (ret < 0) break; @@ -3533,9 +3866,9 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, break; dev_extent = btrfs_item_ptr(l, slot, struct btrfs_dev_extent); - length = btrfs_dev_extent_length(l, dev_extent); + dev_extent_len = btrfs_dev_extent_length(l, dev_extent); - if (found_key.offset + length <= start) + if (found_key.offset + dev_extent_len <= start) goto skip; chunk_offset = btrfs_dev_extent_chunk_offset(l, dev_extent); @@ -3551,6 +3884,55 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, if (!cache) goto skip; + ASSERT(cache->start <= chunk_offset); + /* + * We are using the commit root to search for device extents, so + * that means we could have found a device extent item from a + * block group that was deleted in the current transaction. The + * logical start offset of the deleted block group, stored at + * @chunk_offset, might be part of the logical address range of + * a new block group (which uses different physical extents). + * In this case btrfs_lookup_block_group() has returned the new + * block group, and its start address is less than @chunk_offset. + * + * We skip such new block groups, because it's pointless to + * process them, as we won't find their extents because we search + * for them using the commit root of the extent tree. For a device + * replace it's also fine to skip it, we won't miss copying them + * to the target device because we have the write duplication + * setup through the regular write path (by btrfs_map_block()), + * and we have committed a transaction when we started the device + * replace, right after setting up the device replace state. + */ + if (cache->start < chunk_offset) { + btrfs_put_block_group(cache); + goto skip; + } + + if (sctx->is_dev_replace && btrfs_is_zoned(fs_info)) { + if (!test_bit(BLOCK_GROUP_FLAG_TO_COPY, &cache->runtime_flags)) { + btrfs_put_block_group(cache); + goto skip; + } + } + + /* + * Make sure that while we are scrubbing the corresponding block + * group doesn't get its logical address and its device extents + * reused for another block group, which can possibly be of a + * different type and different profile. We do this to prevent + * false error detections and crashes due to bogus attempts to + * repair extents. + */ + spin_lock(&cache->lock); + if (test_bit(BLOCK_GROUP_FLAG_REMOVED, &cache->runtime_flags)) { + spin_unlock(&cache->lock); + btrfs_put_block_group(cache); + goto skip; + } + btrfs_freeze_block_group(cache); + spin_unlock(&cache->lock); + /* * we need call btrfs_inc_block_group_ro() with scrubs_paused, * to avoid deadlock caused by: @@ -3592,6 +3974,16 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, * group is not RO. */ ret = btrfs_inc_block_group_ro(cache, sctx->is_dev_replace); + if (!ret && sctx->is_dev_replace) { + ret = finish_extent_writes_for_zoned(root, cache); + if (ret) { + btrfs_dec_block_group_ro(cache); + scrub_pause_off(fs_info); + btrfs_put_block_group(cache); + break; + } + } + if (ret == 0) { ro_set = 1; } else if (ret == -ENOSPC && !sctx->is_dev_replace) { @@ -3603,9 +3995,17 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, * commit_transactions. */ ro_set = 0; + } else if (ret == -ETXTBSY) { + btrfs_warn(fs_info, + "skipping scrub of block group %llu due to active swapfile", + cache->start); + scrub_pause_off(fs_info); + ret = 0; + goto skip_unfreeze; } else { btrfs_warn(fs_info, "failed setting block group ro: %d", ret); + btrfs_unfreeze_block_group(cache); btrfs_put_block_group(cache); scrub_pause_off(fs_info); break; @@ -3624,13 +4024,13 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, scrub_pause_off(fs_info); down_write(&dev_replace->rwsem); - dev_replace->cursor_right = found_key.offset + length; + dev_replace->cursor_right = found_key.offset + dev_extent_len; dev_replace->cursor_left = found_key.offset; dev_replace->item_needs_writeback = 1; up_write(&dev_replace->rwsem); - ret = scrub_chunk(sctx, scrub_dev, chunk_offset, length, - found_key.offset, cache); + ret = scrub_chunk(sctx, cache, scrub_dev, found_key.offset, + dev_extent_len); /* * flush, submit all pending read and write bios, afterwards @@ -3664,6 +4064,11 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, scrub_pause_off(fs_info); + if (sctx->is_dev_replace && + !btrfs_finish_block_group_to_copy(dev_replace->srcdev, + cache, found_key.offset)) + ro_set = 0; + down_write(&dev_replace->rwsem); dev_replace->cursor_left = dev_replace->cursor_right; dev_replace->item_needs_writeback = 1; @@ -3680,8 +4085,8 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, * balance is triggered or it becomes used and unused again. */ spin_lock(&cache->lock); - if (!cache->removed && !cache->ro && cache->reserved == 0 && - cache->used == 0) { + if (!test_bit(BLOCK_GROUP_FLAG_REMOVED, &cache->runtime_flags) && + !cache->ro && cache->reserved == 0 && cache->used == 0) { spin_unlock(&cache->lock); if (btrfs_test_opt(fs_info, DISCARD_ASYNC)) btrfs_discard_queue_work(&fs_info->discard_ctl, @@ -3691,7 +4096,8 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, } else { spin_unlock(&cache->lock); } - +skip_unfreeze: + btrfs_unfreeze_block_group(cache); btrfs_put_block_group(cache); if (ret) break; @@ -3705,7 +4111,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, break; } skip: - key.offset = found_key.offset + length; + key.offset = found_key.offset + dev_extent_len; btrfs_release_path(path); } @@ -3723,8 +4129,8 @@ static noinline_for_stack int scrub_supers(struct scrub_ctx *sctx, int ret; struct btrfs_fs_info *fs_info = sctx->fs_info; - if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) - return -EIO; + if (BTRFS_FS_ERROR(fs_info)) + return -EROFS; /* Seed devices of a new filesystem has their own generation. */ if (scrub_dev->fs_devices != fs_info->fs_devices) @@ -3737,10 +4143,12 @@ static noinline_for_stack int scrub_supers(struct scrub_ctx *sctx, if (bytenr + BTRFS_SUPER_INFO_SIZE > scrub_dev->commit_total_bytes) break; + if (!btrfs_check_super_location(scrub_dev, bytenr)) + continue; - ret = scrub_pages(sctx, bytenr, BTRFS_SUPER_INFO_SIZE, bytenr, - scrub_dev, BTRFS_EXTENT_FLAG_SUPER, gen, i, - NULL, 1, bytenr); + ret = scrub_sectors(sctx, bytenr, BTRFS_SUPER_INFO_SIZE, bytenr, + scrub_dev, BTRFS_EXTENT_FLAG_SUPER, gen, i, + NULL, bytenr); if (ret) return ret; } @@ -3749,125 +4157,136 @@ static noinline_for_stack int scrub_supers(struct scrub_ctx *sctx, return 0; } +static void scrub_workers_put(struct btrfs_fs_info *fs_info) +{ + if (refcount_dec_and_mutex_lock(&fs_info->scrub_workers_refcnt, + &fs_info->scrub_lock)) { + struct workqueue_struct *scrub_workers = fs_info->scrub_workers; + struct workqueue_struct *scrub_wr_comp = + fs_info->scrub_wr_completion_workers; + struct workqueue_struct *scrub_parity = + fs_info->scrub_parity_workers; + + fs_info->scrub_workers = NULL; + fs_info->scrub_wr_completion_workers = NULL; + fs_info->scrub_parity_workers = NULL; + mutex_unlock(&fs_info->scrub_lock); + + if (scrub_workers) + destroy_workqueue(scrub_workers); + if (scrub_wr_comp) + destroy_workqueue(scrub_wr_comp); + if (scrub_parity) + destroy_workqueue(scrub_parity); + } +} + /* * get a reference count on fs_info->scrub_workers. start worker if necessary */ static noinline_for_stack int scrub_workers_get(struct btrfs_fs_info *fs_info, int is_dev_replace) { + struct workqueue_struct *scrub_workers = NULL; + struct workqueue_struct *scrub_wr_comp = NULL; + struct workqueue_struct *scrub_parity = NULL; unsigned int flags = WQ_FREEZABLE | WQ_UNBOUND; int max_active = fs_info->thread_pool_size; + int ret = -ENOMEM; - lockdep_assert_held(&fs_info->scrub_lock); + if (refcount_inc_not_zero(&fs_info->scrub_workers_refcnt)) + return 0; - if (refcount_read(&fs_info->scrub_workers_refcnt) == 0) { - ASSERT(fs_info->scrub_workers == NULL); - fs_info->scrub_workers = btrfs_alloc_workqueue(fs_info, "scrub", - flags, is_dev_replace ? 1 : max_active, 4); - if (!fs_info->scrub_workers) - goto fail_scrub_workers; - - ASSERT(fs_info->scrub_wr_completion_workers == NULL); - fs_info->scrub_wr_completion_workers = - btrfs_alloc_workqueue(fs_info, "scrubwrc", flags, - max_active, 2); - if (!fs_info->scrub_wr_completion_workers) - goto fail_scrub_wr_completion_workers; - - ASSERT(fs_info->scrub_parity_workers == NULL); - fs_info->scrub_parity_workers = - btrfs_alloc_workqueue(fs_info, "scrubparity", flags, - max_active, 2); - if (!fs_info->scrub_parity_workers) - goto fail_scrub_parity_workers; + scrub_workers = alloc_workqueue("btrfs-scrub", flags, + is_dev_replace ? 1 : max_active); + if (!scrub_workers) + goto fail_scrub_workers; + + scrub_wr_comp = alloc_workqueue("btrfs-scrubwrc", flags, max_active); + if (!scrub_wr_comp) + goto fail_scrub_wr_completion_workers; + scrub_parity = alloc_workqueue("btrfs-scrubparity", flags, max_active); + if (!scrub_parity) + goto fail_scrub_parity_workers; + + mutex_lock(&fs_info->scrub_lock); + if (refcount_read(&fs_info->scrub_workers_refcnt) == 0) { + ASSERT(fs_info->scrub_workers == NULL && + fs_info->scrub_wr_completion_workers == NULL && + fs_info->scrub_parity_workers == NULL); + fs_info->scrub_workers = scrub_workers; + fs_info->scrub_wr_completion_workers = scrub_wr_comp; + fs_info->scrub_parity_workers = scrub_parity; refcount_set(&fs_info->scrub_workers_refcnt, 1); - } else { - refcount_inc(&fs_info->scrub_workers_refcnt); + mutex_unlock(&fs_info->scrub_lock); + return 0; } - return 0; + /* Other thread raced in and created the workers for us */ + refcount_inc(&fs_info->scrub_workers_refcnt); + mutex_unlock(&fs_info->scrub_lock); + ret = 0; + destroy_workqueue(scrub_parity); fail_scrub_parity_workers: - btrfs_destroy_workqueue(fs_info->scrub_wr_completion_workers); + destroy_workqueue(scrub_wr_comp); fail_scrub_wr_completion_workers: - btrfs_destroy_workqueue(fs_info->scrub_workers); + destroy_workqueue(scrub_workers); fail_scrub_workers: - return -ENOMEM; + return ret; } int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start, u64 end, struct btrfs_scrub_progress *progress, int readonly, int is_dev_replace) { + struct btrfs_dev_lookup_args args = { .devid = devid }; struct scrub_ctx *sctx; int ret; struct btrfs_device *dev; unsigned int nofs_flag; - struct btrfs_workqueue *scrub_workers = NULL; - struct btrfs_workqueue *scrub_wr_comp = NULL; - struct btrfs_workqueue *scrub_parity = NULL; + bool need_commit = false; if (btrfs_fs_closing(fs_info)) return -EAGAIN; - if (fs_info->nodesize > BTRFS_STRIPE_LEN) { - /* - * in this case scrub is unable to calculate the checksum - * the way scrub is implemented. Do not handle this - * situation at all because it won't ever happen. - */ - btrfs_err(fs_info, - "scrub: size assumption nodesize <= BTRFS_STRIPE_LEN (%d <= %d) fails", - fs_info->nodesize, - BTRFS_STRIPE_LEN); - return -EINVAL; - } - - if (fs_info->sectorsize != PAGE_SIZE) { - /* not supported for data w/o checksums */ - btrfs_err_rl(fs_info, - "scrub: size assumption sectorsize != PAGE_SIZE (%d != %lu) fails", - fs_info->sectorsize, PAGE_SIZE); - return -EINVAL; - } + /* At mount time we have ensured nodesize is in the range of [4K, 64K]. */ + ASSERT(fs_info->nodesize <= BTRFS_STRIPE_LEN); - if (fs_info->nodesize > - PAGE_SIZE * SCRUB_MAX_PAGES_PER_BLOCK || - fs_info->sectorsize > PAGE_SIZE * SCRUB_MAX_PAGES_PER_BLOCK) { - /* - * would exhaust the array bounds of pagev member in - * struct scrub_block - */ - btrfs_err(fs_info, - "scrub: size assumption nodesize and sectorsize <= SCRUB_MAX_PAGES_PER_BLOCK (%d <= %d && %d <= %d) fails", - fs_info->nodesize, - SCRUB_MAX_PAGES_PER_BLOCK, - fs_info->sectorsize, - SCRUB_MAX_PAGES_PER_BLOCK); - return -EINVAL; - } + /* + * SCRUB_MAX_SECTORS_PER_BLOCK is calculated using the largest possible + * value (max nodesize / min sectorsize), thus nodesize should always + * be fine. + */ + ASSERT(fs_info->nodesize <= + SCRUB_MAX_SECTORS_PER_BLOCK << fs_info->sectorsize_bits); /* Allocate outside of device_list_mutex */ sctx = scrub_setup_ctx(fs_info, is_dev_replace); if (IS_ERR(sctx)) return PTR_ERR(sctx); + ret = scrub_workers_get(fs_info, is_dev_replace); + if (ret) + goto out_free_ctx; + mutex_lock(&fs_info->fs_devices->device_list_mutex); - dev = btrfs_find_device(fs_info->fs_devices, devid, NULL, NULL, true); + dev = btrfs_find_device(fs_info->fs_devices, &args); if (!dev || (test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state) && !is_dev_replace)) { mutex_unlock(&fs_info->fs_devices->device_list_mutex); ret = -ENODEV; - goto out_free_ctx; + goto out; } if (!is_dev_replace && !readonly && !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state)) { mutex_unlock(&fs_info->fs_devices->device_list_mutex); - btrfs_err_in_rcu(fs_info, "scrub: device %s is not writable", - rcu_str_deref(dev->name)); + btrfs_err_in_rcu(fs_info, + "scrub on devid %llu: filesystem on %s is not writable", + devid, rcu_str_deref(dev->name)); ret = -EROFS; - goto out_free_ctx; + goto out; } mutex_lock(&fs_info->scrub_lock); @@ -3876,7 +4295,7 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start, mutex_unlock(&fs_info->scrub_lock); mutex_unlock(&fs_info->fs_devices->device_list_mutex); ret = -EIO; - goto out_free_ctx; + goto out; } down_read(&fs_info->dev_replace.rwsem); @@ -3887,17 +4306,10 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start, mutex_unlock(&fs_info->scrub_lock); mutex_unlock(&fs_info->fs_devices->device_list_mutex); ret = -EINPROGRESS; - goto out_free_ctx; + goto out; } up_read(&fs_info->dev_replace.rwsem); - ret = scrub_workers_get(fs_info, is_dev_replace); - if (ret) { - mutex_unlock(&fs_info->scrub_lock); - mutex_unlock(&fs_info->fs_devices->device_list_mutex); - goto out_free_ctx; - } - sctx->readonly = readonly; dev->scrub_ctx = sctx; mutex_unlock(&fs_info->fs_devices->device_list_mutex); @@ -3913,7 +4325,7 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start, /* * In order to avoid deadlock with reclaim when there is a transaction * trying to pause scrub, make sure we use GFP_NOFS for all the - * allocations done at btrfs_scrub_pages() and scrub_pages_for_parity() + * allocations done at btrfs_scrub_sectors() and scrub_sectors_for_parity() * invoked by our callees. The pausing request is done when the * transaction commit starts, and it blocks the transaction until scrub * is paused (done at specific points at scrub_stripe() or right above @@ -3921,6 +4333,12 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start, */ nofs_flag = memalloc_nofs_save(); if (!is_dev_replace) { + u64 old_super_errors; + + spin_lock(&sctx->stat_lock); + old_super_errors = sctx->stat.super_errors; + spin_unlock(&sctx->stat_lock); + btrfs_info(fs_info, "scrub: started on devid %llu", devid); /* * by holding device list mutex, we can @@ -3929,6 +4347,16 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start, mutex_lock(&fs_info->fs_devices->device_list_mutex); ret = scrub_supers(sctx, dev); mutex_unlock(&fs_info->fs_devices->device_list_mutex); + + spin_lock(&sctx->stat_lock); + /* + * Super block errors found, but we can not commit transaction + * at current context, since btrfs_commit_transaction() needs + * to pause the current running scrub (hold by ourselves). + */ + if (sctx->stat.super_errors > old_super_errors && !sctx->readonly) + need_commit = true; + spin_unlock(&sctx->stat_lock); } if (!ret) @@ -3950,24 +4378,33 @@ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start, mutex_lock(&fs_info->scrub_lock); dev->scrub_ctx = NULL; - if (refcount_dec_and_test(&fs_info->scrub_workers_refcnt)) { - scrub_workers = fs_info->scrub_workers; - scrub_wr_comp = fs_info->scrub_wr_completion_workers; - scrub_parity = fs_info->scrub_parity_workers; - - fs_info->scrub_workers = NULL; - fs_info->scrub_wr_completion_workers = NULL; - fs_info->scrub_parity_workers = NULL; - } mutex_unlock(&fs_info->scrub_lock); - btrfs_destroy_workqueue(scrub_workers); - btrfs_destroy_workqueue(scrub_wr_comp); - btrfs_destroy_workqueue(scrub_parity); + scrub_workers_put(fs_info); scrub_put_ctx(sctx); + /* + * We found some super block errors before, now try to force a + * transaction commit, as scrub has finished. + */ + if (need_commit) { + struct btrfs_trans_handle *trans; + + trans = btrfs_start_transaction(fs_info->tree_root, 0); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + btrfs_err(fs_info, + "scrub: failed to start transaction to fix super block errors: %d", ret); + return ret; + } + ret = btrfs_commit_transaction(trans); + if (ret < 0) + btrfs_err(fs_info, + "scrub: failed to commit transaction to fix super block errors: %d", ret); + } return ret; - +out: + scrub_workers_put(fs_info); out_free_ctx: scrub_free_ctx(sctx); @@ -4042,11 +4479,12 @@ int btrfs_scrub_cancel_dev(struct btrfs_device *dev) int btrfs_scrub_progress(struct btrfs_fs_info *fs_info, u64 devid, struct btrfs_scrub_progress *progress) { + struct btrfs_dev_lookup_args args = { .devid = devid }; struct btrfs_device *dev; struct scrub_ctx *sctx = NULL; mutex_lock(&fs_info->fs_devices->device_list_mutex); - dev = btrfs_find_device(fs_info->fs_devices, devid, NULL, NULL, true); + dev = btrfs_find_device(fs_info->fs_devices, &args); if (dev) sctx = dev->scrub_ctx; if (sctx) @@ -4056,27 +4494,27 @@ int btrfs_scrub_progress(struct btrfs_fs_info *fs_info, u64 devid, return dev ? (sctx ? 0 : -ENOTCONN) : -ENODEV; } -static void scrub_remap_extent(struct btrfs_fs_info *fs_info, - u64 extent_logical, u64 extent_len, - u64 *extent_physical, - struct btrfs_device **extent_dev, - int *extent_mirror_num) +static void scrub_find_good_copy(struct btrfs_fs_info *fs_info, + u64 extent_logical, u32 extent_len, + u64 *extent_physical, + struct btrfs_device **extent_dev, + int *extent_mirror_num) { u64 mapped_length; - struct btrfs_bio *bbio = NULL; + struct btrfs_io_context *bioc = NULL; int ret; mapped_length = extent_len; ret = btrfs_map_block(fs_info, BTRFS_MAP_READ, extent_logical, - &mapped_length, &bbio, 0); - if (ret || !bbio || mapped_length < extent_len || - !bbio->stripes[0].dev->bdev) { - btrfs_put_bbio(bbio); + &mapped_length, &bioc, 0); + if (ret || !bioc || mapped_length < extent_len || + !bioc->stripes[0].dev->bdev) { + btrfs_put_bioc(bioc); return; } - *extent_physical = bbio->stripes[0].physical; - *extent_mirror_num = bbio->mirror_num; - *extent_dev = bbio->stripes[0].dev; - btrfs_put_bbio(bbio); + *extent_physical = bioc->stripes[0].physical; + *extent_mirror_num = bioc->mirror_num; + *extent_dev = bioc->stripes[0].dev; + btrfs_put_bioc(bioc); } |