diff options
Diffstat (limited to 'fs/f2fs/data.c')
| -rw-r--r-- | fs/f2fs/data.c | 1251 |
1 files changed, 799 insertions, 452 deletions
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index de4da6d9cd93..8d4f1e75dee3 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -10,7 +10,7 @@ #include <linux/sched/mm.h> #include <linux/mpage.h> #include <linux/writeback.h> -#include <linux/pagevec.h> +#include <linux/folio_batch.h> #include <linux/blkdev.h> #include <linux/bio.h> #include <linux/blk-crypto.h> @@ -31,9 +31,15 @@ static struct kmem_cache *bio_post_read_ctx_cache; static struct kmem_cache *bio_entry_slab; +static struct kmem_cache *ffs_entry_slab; static mempool_t *bio_post_read_ctx_pool; static struct bio_set f2fs_bioset; +struct f2fs_folio_state { + spinlock_t state_lock; + unsigned int read_pages_pending; +}; + #define F2FS_BIO_POOL_SIZE NR_CURSEG_TYPE int __init f2fs_init_bioset(void) @@ -47,14 +53,14 @@ void f2fs_destroy_bioset(void) bioset_exit(&f2fs_bioset); } -bool f2fs_is_cp_guaranteed(struct page *page) +bool f2fs_is_cp_guaranteed(const struct folio *folio) { - struct address_space *mapping = page->mapping; + struct address_space *mapping = folio->mapping; struct inode *inode; struct f2fs_sb_info *sbi; - if (!mapping) - return false; + if (fscrypt_is_bounce_folio(folio)) + return folio_test_f2fs_gcing(fscrypt_pagecache_folio(folio)); inode = mapping->host; sbi = F2FS_I_SB(inode); @@ -65,7 +71,7 @@ bool f2fs_is_cp_guaranteed(struct page *page) return true; if ((S_ISREG(inode->i_mode) && IS_NOQUOTA(inode)) || - page_private_gcing(page)) + folio_test_f2fs_gcing(folio)) return true; return false; } @@ -109,6 +115,7 @@ enum bio_post_read_step { struct bio_post_read_ctx { struct bio *bio; struct f2fs_sb_info *sbi; + struct fsverity_info *vi; struct work_struct work; unsigned int enabled_steps; /* @@ -138,20 +145,42 @@ static void f2fs_finish_read_bio(struct bio *bio, bool in_task) { struct folio_iter fi; struct bio_post_read_ctx *ctx = bio->bi_private; + unsigned long flags; bio_for_each_folio_all(fi, bio) { struct folio *folio = fi.folio; + unsigned nr_pages = fi.length >> PAGE_SHIFT; + bool finished = true; - if (f2fs_is_compressed_page(&folio->page)) { + if (!folio_test_large(folio) && + f2fs_is_compressed_page(folio)) { if (ctx && !ctx->decompression_attempted) - f2fs_end_read_compressed_page(&folio->page, true, 0, + f2fs_end_read_compressed_page(folio, true, 0, in_task); - f2fs_put_page_dic(&folio->page, in_task); + f2fs_put_folio_dic(folio, in_task); continue; } - dec_page_count(F2FS_F_SB(folio), __read_io_type(folio)); - folio_end_read(folio, bio->bi_status == 0); + if (folio_test_large(folio)) { + struct f2fs_folio_state *ffs = folio->private; + + spin_lock_irqsave(&ffs->state_lock, flags); + ffs->read_pages_pending -= nr_pages; + finished = !ffs->read_pages_pending; + spin_unlock_irqrestore(&ffs->state_lock, flags); + } + + while (nr_pages--) + dec_page_count(F2FS_F_SB(folio), __read_io_type(folio)); + + if (bio->bi_status == BLK_STS_OK && + F2FS_F_SB(folio)->node_inode && is_node_folio(folio) && + f2fs_sanity_check_node_footer(F2FS_F_SB(folio), + folio, folio->index, NODE_TYPE_REGULAR, true)) + bio->bi_status = BLK_STS_IOERR; + + if (finished) + folio_end_read(folio, bio->bi_status == BLK_STS_OK); } if (ctx) @@ -165,6 +194,7 @@ static void f2fs_verify_bio(struct work_struct *work) container_of(work, struct bio_post_read_ctx, work); struct bio *bio = ctx->bio; bool may_have_compressed_pages = (ctx->enabled_steps & STEP_DECOMPRESS); + struct fsverity_info *vi = ctx->vi; /* * fsverity_verify_bio() may call readahead() again, and while verity @@ -181,20 +211,19 @@ static void f2fs_verify_bio(struct work_struct *work) * as those were handled separately by f2fs_end_read_compressed_page(). */ if (may_have_compressed_pages) { - struct bio_vec *bv; - struct bvec_iter_all iter_all; + struct folio_iter fi; - bio_for_each_segment_all(bv, bio, iter_all) { - struct page *page = bv->bv_page; + bio_for_each_folio_all(fi, bio) { + struct folio *folio = fi.folio; - if (!f2fs_is_compressed_page(page) && - !fsverity_verify_page(page)) { + if (!f2fs_is_compressed_page(folio) && + !fsverity_verify_folio(vi, folio)) { bio->bi_status = BLK_STS_IOERR; break; } } } else { - fsverity_verify_bio(bio); + fsverity_verify_bio(vi, bio); } f2fs_finish_read_bio(bio, true); @@ -233,16 +262,15 @@ static void f2fs_verify_and_finish_bio(struct bio *bio, bool in_task) static void f2fs_handle_step_decompress(struct bio_post_read_ctx *ctx, bool in_task) { - struct bio_vec *bv; - struct bvec_iter_all iter_all; + struct folio_iter fi; bool all_compressed = true; block_t blkaddr = ctx->fs_blkaddr; - bio_for_each_segment_all(bv, ctx->bio, iter_all) { - struct page *page = bv->bv_page; + bio_for_each_folio_all(fi, ctx->bio) { + struct folio *folio = fi.folio; - if (f2fs_is_compressed_page(page)) - f2fs_end_read_compressed_page(page, false, blkaddr, + if (f2fs_is_compressed_page(folio)) + f2fs_end_read_compressed_page(folio, false, blkaddr, in_task); else all_compressed = false; @@ -280,9 +308,9 @@ static void f2fs_post_read_work(struct work_struct *work) static void f2fs_read_end_io(struct bio *bio) { - struct f2fs_sb_info *sbi = F2FS_P_SB(bio_first_page_all(bio)); + struct f2fs_sb_info *sbi = F2FS_F_SB(bio_first_folio_all(bio)); struct bio_post_read_ctx *ctx; - bool intask = in_task(); + bool intask = in_task() && !irqs_disabled(); iostat_update_and_unbind_ctx(bio); ctx = bio->bi_private; @@ -290,7 +318,7 @@ static void f2fs_read_end_io(struct bio *bio) if (time_to_inject(sbi, FAULT_READ_IO)) bio->bi_status = BLK_STS_IOERR; - if (bio->bi_status) { + if (bio->bi_status != BLK_STS_OK) { f2fs_finish_read_bio(bio, intask); return; } @@ -319,8 +347,7 @@ static void f2fs_read_end_io(struct bio *bio) static void f2fs_write_end_io(struct bio *bio) { struct f2fs_sb_info *sbi; - struct bio_vec *bvec; - struct bvec_iter_all iter_all; + struct folio_iter fi; iostat_update_and_unbind_ctx(bio); sbi = bio->bi_private; @@ -328,38 +355,54 @@ static void f2fs_write_end_io(struct bio *bio) if (time_to_inject(sbi, FAULT_WRITE_IO)) bio->bi_status = BLK_STS_IOERR; - bio_for_each_segment_all(bvec, bio, iter_all) { - struct page *page = bvec->bv_page; - enum count_type type = WB_DATA_TYPE(page, false); + bio_for_each_folio_all(fi, bio) { + struct folio *folio = fi.folio; + enum count_type type; - fscrypt_finalize_bounce_page(&page); + if (fscrypt_is_bounce_folio(folio)) { + struct folio *io_folio = folio; + + folio = fscrypt_pagecache_folio(io_folio); + fscrypt_free_bounce_page(&io_folio->page); + } #ifdef CONFIG_F2FS_FS_COMPRESSION - if (f2fs_is_compressed_page(page)) { - f2fs_compress_write_end_io(bio, page); + if (f2fs_is_compressed_page(folio)) { + f2fs_compress_write_end_io(bio, folio); continue; } #endif - if (unlikely(bio->bi_status)) { - mapping_set_error(page->mapping, -EIO); + type = WB_DATA_TYPE(folio, false); + + if (unlikely(bio->bi_status != BLK_STS_OK)) { + mapping_set_error(folio->mapping, -EIO); if (type == F2FS_WB_CP_DATA) f2fs_stop_checkpoint(sbi, true, STOP_CP_REASON_WRITE_FAIL); } - f2fs_bug_on(sbi, page->mapping == NODE_MAPPING(sbi) && - page_folio(page)->index != nid_of_node(page)); + if (is_node_folio(folio)) { + f2fs_sanity_check_node_footer(sbi, folio, + folio->index, NODE_TYPE_REGULAR, true); + f2fs_bug_on(sbi, folio->index != nid_of_node(folio)); + } + if (f2fs_in_warm_node_list(folio)) + f2fs_del_fsync_node_entry(sbi, folio); dec_page_count(sbi, type); - if (f2fs_in_warm_node_list(sbi, page)) - f2fs_del_fsync_node_entry(sbi, page); - clear_page_private_gcing(page); - end_page_writeback(page); - } - if (!get_pages(sbi, F2FS_WB_CP_DATA) && + + /* + * we should access sbi before folio_end_writeback() to + * avoid racing w/ kill_f2fs_super() + */ + if (type == F2FS_WB_CP_DATA && !get_pages(sbi, type) && wq_has_sleeper(&sbi->cp_wait)) - wake_up(&sbi->cp_wait); + wake_up(&sbi->cp_wait); + + folio_clear_f2fs_gcing(folio); + folio_end_writeback(folio); + } bio_put(bio); } @@ -438,6 +481,11 @@ static blk_opf_t f2fs_io_flags(struct f2fs_io_info *fio) op_flags |= REQ_META; if (BIT(fio->temp) & fua_flag) op_flags |= REQ_FUA; + + if (fio->type == DATA && + F2FS_I(fio->folio->mapping->host)->ioprio_hint == F2FS_IOPRIO_WRITE) + op_flags |= REQ_PRIO; + return op_flags; } @@ -480,7 +528,9 @@ static void f2fs_set_bio_crypt_ctx(struct bio *bio, const struct inode *inode, * read/write raw data without encryption. */ if (!fio || !fio->encrypted_page) - fscrypt_set_bio_crypt_ctx(bio, inode, first_idx, gfp_mask); + fscrypt_set_bio_crypt_ctx(bio, inode, + (loff_t)first_idx << inode->i_blkbits, + gfp_mask); } static bool f2fs_crypt_mergeable_bio(struct bio *bio, const struct inode *inode, @@ -494,17 +544,21 @@ static bool f2fs_crypt_mergeable_bio(struct bio *bio, const struct inode *inode, if (fio && fio->encrypted_page) return !bio_has_crypt_ctx(bio); - return fscrypt_mergeable_bio(bio, inode, next_idx); + return fscrypt_mergeable_bio(bio, inode, + (loff_t)next_idx << inode->i_blkbits); } void f2fs_submit_read_bio(struct f2fs_sb_info *sbi, struct bio *bio, enum page_type type) { + if (!bio) + return; + WARN_ON_ONCE(!is_read_io(bio_op(bio))); trace_f2fs_submit_read_bio(sbi->sb, type, bio); iostat_update_submit_ctx(bio, type); - submit_bio(bio); + blk_crypto_submit_bio(bio); } static void f2fs_submit_write_bio(struct f2fs_sb_info *sbi, struct bio *bio, @@ -513,7 +567,7 @@ static void f2fs_submit_write_bio(struct f2fs_sb_info *sbi, struct bio *bio, WARN_ON_ONCE(is_read_io(bio_op(bio))); trace_f2fs_submit_write_bio(sbi->sb, type, bio); iostat_update_submit_ctx(bio, type); - submit_bio(bio); + blk_crypto_submit_bio(bio); } static void __submit_merged_bio(struct f2fs_bio_info *io) @@ -534,34 +588,33 @@ static void __submit_merged_bio(struct f2fs_bio_info *io) } static bool __has_merged_page(struct bio *bio, struct inode *inode, - struct page *page, nid_t ino) + struct folio *folio, nid_t ino) { - struct bio_vec *bvec; - struct bvec_iter_all iter_all; + struct folio_iter fi; if (!bio) return false; - if (!inode && !page && !ino) + if (!inode && !folio && !ino) return true; - bio_for_each_segment_all(bvec, bio, iter_all) { - struct page *target = bvec->bv_page; + bio_for_each_folio_all(fi, bio) { + struct folio *target = fi.folio; - if (fscrypt_is_bounce_page(target)) { - target = fscrypt_pagecache_page(target); + if (fscrypt_is_bounce_folio(target)) { + target = fscrypt_pagecache_folio(target); if (IS_ERR(target)) continue; } if (f2fs_is_compressed_page(target)) { - target = f2fs_compress_control_page(target); + target = f2fs_compress_control_folio(target); if (IS_ERR(target)) continue; } if (inode && inode == target->mapping->host) return true; - if (page && page == target) + if (folio && folio == target) return true; if (ino && ino == ino_of_node(target)) return true; @@ -587,7 +640,8 @@ int f2fs_init_write_merge_io(struct f2fs_sb_info *sbi) for (j = HOT; j < n; j++) { struct f2fs_bio_info *io = &sbi->write_io[i][j]; - init_f2fs_rwsem(&io->io_rwsem); + init_f2fs_rwsem_trace(&io->io_rwsem, sbi, + LOCK_NAME_IO_RWSEM); io->sbi = sbi; io->bio = NULL; io->last_block_in_bio = 0; @@ -611,8 +665,9 @@ static void __f2fs_submit_merged_write(struct f2fs_sb_info *sbi, { enum page_type btype = PAGE_TYPE_OF_BIO(type); struct f2fs_bio_info *io = sbi->write_io[btype] + temp; + struct f2fs_lock_context lc; - f2fs_down_write(&io->io_rwsem); + f2fs_down_write_trace(&io->io_rwsem, &lc); if (!io->bio) goto unlock_out; @@ -626,27 +681,37 @@ static void __f2fs_submit_merged_write(struct f2fs_sb_info *sbi, } __submit_merged_bio(io); unlock_out: - f2fs_up_write(&io->io_rwsem); + f2fs_up_write_trace(&io->io_rwsem, &lc); } static void __submit_merged_write_cond(struct f2fs_sb_info *sbi, - struct inode *inode, struct page *page, - nid_t ino, enum page_type type, bool force) + struct inode *inode, struct folio *folio, + nid_t ino, enum page_type type, bool writeback) { enum temp_type temp; bool ret = true; + bool force = !inode && !folio && !ino; for (temp = HOT; temp < NR_TEMP_TYPE; temp++) { if (!force) { enum page_type btype = PAGE_TYPE_OF_BIO(type); struct f2fs_bio_info *io = sbi->write_io[btype] + temp; + struct f2fs_lock_context lc; - f2fs_down_read(&io->io_rwsem); - ret = __has_merged_page(io->bio, inode, page, ino); - f2fs_up_read(&io->io_rwsem); + f2fs_down_read_trace(&io->io_rwsem, &lc); + ret = __has_merged_page(io->bio, inode, folio, ino); + f2fs_up_read_trace(&io->io_rwsem, &lc); } - if (ret) + if (ret) { __f2fs_submit_merged_write(sbi, type, temp); + /* + * For waitting writebck case, if the bio owned by the + * folio is already submitted, we do not need to submit + * other types of bios. + */ + if (writeback) + break; + } /* TODO: use HOT temp only for meta pages now. */ if (type >= META) @@ -656,14 +721,20 @@ static void __submit_merged_write_cond(struct f2fs_sb_info *sbi, void f2fs_submit_merged_write(struct f2fs_sb_info *sbi, enum page_type type) { - __submit_merged_write_cond(sbi, NULL, NULL, 0, type, true); + __submit_merged_write_cond(sbi, NULL, NULL, 0, type, false); } void f2fs_submit_merged_write_cond(struct f2fs_sb_info *sbi, - struct inode *inode, struct page *page, + struct inode *inode, struct folio *folio, nid_t ino, enum page_type type) { - __submit_merged_write_cond(sbi, inode, page, ino, type, false); + __submit_merged_write_cond(sbi, inode, folio, ino, type, false); +} + +void f2fs_submit_merged_write_folio(struct f2fs_sb_info *sbi, + struct folio *folio, enum page_type type) +{ + __submit_merged_write_cond(sbi, NULL, folio, 0, type, true); } void f2fs_flush_merged_writes(struct f2fs_sb_info *sbi) @@ -680,7 +751,7 @@ void f2fs_flush_merged_writes(struct f2fs_sb_info *sbi) int f2fs_submit_page_bio(struct f2fs_io_info *fio) { struct bio *bio; - struct folio *fio_folio = page_folio(fio->page); + struct folio *fio_folio = fio->folio; struct folio *data_folio = fio->encrypted_page ? page_folio(fio->encrypted_page) : fio_folio; @@ -702,7 +773,7 @@ int f2fs_submit_page_bio(struct f2fs_io_info *fio) wbc_account_cgroup_owner(fio->io_wbc, fio_folio, PAGE_SIZE); inc_page_count(fio->sbi, is_read_io(fio->op) ? - __read_io_type(data_folio) : WB_DATA_TYPE(fio->page, false)); + __read_io_type(data_folio) : WB_DATA_TYPE(fio->folio, false)); if (is_read_io(bio_op(bio))) f2fs_submit_read_bio(fio->sbi, bio, fio->type); @@ -725,9 +796,11 @@ static bool page_is_mergeable(struct f2fs_sb_info *sbi, struct bio *bio, static bool io_type_is_mergeable(struct f2fs_bio_info *io, struct f2fs_io_info *fio) { + blk_opf_t mask = ~(REQ_PREFLUSH | REQ_FUA); + if (io->fio.op != fio->op) return false; - return io->fio.op_flags == fio->op_flags; + return (io->fio.op_flags & mask) == (fio->op_flags & mask); } static bool io_is_mergeable(struct f2fs_sb_info *sbi, struct bio *bio, @@ -742,7 +815,7 @@ static bool io_is_mergeable(struct f2fs_sb_info *sbi, struct bio *bio, } static void add_bio_entry(struct f2fs_sb_info *sbi, struct bio *bio, - struct page *page, enum temp_type temp) + struct folio *folio, enum temp_type temp) { struct f2fs_bio_info *io = sbi->write_io[DATA] + temp; struct bio_entry *be; @@ -751,8 +824,7 @@ static void add_bio_entry(struct f2fs_sb_info *sbi, struct bio *bio, be->bio = bio; bio_get(bio); - if (bio_add_page(bio, page, PAGE_SIZE, 0) != PAGE_SIZE) - f2fs_bug_on(sbi, 1); + bio_add_folio_nofail(bio, folio, folio_size(folio), 0); f2fs_down_write(&io->bio_list_lock); list_add_tail(&be->list, &io->bio_list); @@ -766,8 +838,9 @@ static void del_bio_entry(struct bio_entry *be) } static int add_ipu_page(struct f2fs_io_info *fio, struct bio **bio, - struct page *page) + struct folio *folio) { + struct folio *fio_folio = fio->folio; struct f2fs_sb_info *sbi = fio->sbi; enum temp_type temp; bool found = false; @@ -789,10 +862,9 @@ static int add_ipu_page(struct f2fs_io_info *fio, struct bio **bio, *fio->last_block, fio->new_blkaddr)); if (f2fs_crypt_mergeable_bio(*bio, - fio->page->mapping->host, - page_folio(fio->page)->index, fio) && - bio_add_page(*bio, page, PAGE_SIZE, 0) == - PAGE_SIZE) { + fio_folio->mapping->host, + fio_folio->index, fio) && + bio_add_folio(*bio, folio, folio_size(folio), 0)) { ret = 0; break; } @@ -814,13 +886,13 @@ static int add_ipu_page(struct f2fs_io_info *fio, struct bio **bio, } void f2fs_submit_merged_ipu_write(struct f2fs_sb_info *sbi, - struct bio **bio, struct page *page) + struct bio **bio, struct folio *folio) { enum temp_type temp; bool found = false; struct bio *target = bio ? *bio : NULL; - f2fs_bug_on(sbi, !target && !page); + f2fs_bug_on(sbi, !target && !folio); for (temp = HOT; temp < NR_TEMP_TYPE && !found; temp++) { struct f2fs_bio_info *io = sbi->write_io[DATA] + temp; @@ -836,7 +908,7 @@ void f2fs_submit_merged_ipu_write(struct f2fs_sb_info *sbi, found = (target == be->bio); else found = __has_merged_page(be->bio, NULL, - page, 0); + folio, 0); if (found) break; } @@ -853,7 +925,7 @@ void f2fs_submit_merged_ipu_write(struct f2fs_sb_info *sbi, found = (target == be->bio); else found = __has_merged_page(be->bio, NULL, - page, 0); + folio, 0); if (found) { target = be->bio; del_bio_entry(be); @@ -874,14 +946,15 @@ void f2fs_submit_merged_ipu_write(struct f2fs_sb_info *sbi, int f2fs_merge_page_bio(struct f2fs_io_info *fio) { struct bio *bio = *fio->bio; - struct page *page = fio->encrypted_page ? - fio->encrypted_page : fio->page; + struct folio *data_folio = fio->encrypted_page ? + page_folio(fio->encrypted_page) : fio->folio; + struct folio *folio = fio->folio; if (!f2fs_is_valid_blkaddr(fio->sbi, fio->new_blkaddr, __is_meta_io(fio) ? META_GENERIC : DATA_GENERIC)) return -EFSCORRUPTED; - trace_f2fs_submit_folio_bio(page_folio(page), fio); + trace_f2fs_submit_folio_bio(data_folio, fio); if (bio && !page_is_mergeable(fio->sbi, bio, *fio->last_block, fio->new_blkaddr)) @@ -889,20 +962,19 @@ int f2fs_merge_page_bio(struct f2fs_io_info *fio) alloc_new: if (!bio) { bio = __bio_alloc(fio, BIO_MAX_VECS); - f2fs_set_bio_crypt_ctx(bio, fio->page->mapping->host, - page_folio(fio->page)->index, fio, GFP_NOIO); + f2fs_set_bio_crypt_ctx(bio, folio->mapping->host, + folio->index, fio, GFP_NOIO); - add_bio_entry(fio->sbi, bio, page, fio->temp); + add_bio_entry(fio->sbi, bio, data_folio, fio->temp); } else { - if (add_ipu_page(fio, &bio, page)) + if (add_ipu_page(fio, &bio, data_folio)) goto alloc_new; } if (fio->io_wbc) - wbc_account_cgroup_owner(fio->io_wbc, page_folio(fio->page), - PAGE_SIZE); + wbc_account_cgroup_owner(fio->io_wbc, folio, folio_size(folio)); - inc_page_count(fio->sbi, WB_DATA_TYPE(page, false)); + inc_page_count(fio->sbi, WB_DATA_TYPE(folio, false)); *fio->last_block = fio->new_blkaddr; *fio->bio = bio; @@ -937,12 +1009,13 @@ void f2fs_submit_page_write(struct f2fs_io_info *fio) struct f2fs_sb_info *sbi = fio->sbi; enum page_type btype = PAGE_TYPE_OF_BIO(fio->type); struct f2fs_bio_info *io = sbi->write_io[btype] + fio->temp; - struct page *bio_page; + struct folio *bio_folio; + struct f2fs_lock_context lc; enum count_type type; f2fs_bug_on(sbi, is_read_io(fio->op)); - f2fs_down_write(&io->io_rwsem); + f2fs_down_write_trace(&io->io_rwsem, &lc); next: #ifdef CONFIG_BLK_DEV_ZONED if (f2fs_sb_has_blkzoned(sbi) && btype < META && io->zone_pending_bio) { @@ -968,44 +1041,44 @@ next: verify_fio_blkaddr(fio); if (fio->encrypted_page) - bio_page = fio->encrypted_page; + bio_folio = page_folio(fio->encrypted_page); else if (fio->compressed_page) - bio_page = fio->compressed_page; + bio_folio = page_folio(fio->compressed_page); else - bio_page = fio->page; + bio_folio = fio->folio; /* set submitted = true as a return value */ fio->submitted = 1; - type = WB_DATA_TYPE(bio_page, fio->compressed_page); + type = WB_DATA_TYPE(bio_folio, fio->compressed_page); inc_page_count(sbi, type); if (io->bio && (!io_is_mergeable(sbi, io->bio, io, fio, io->last_block_in_bio, fio->new_blkaddr) || - !f2fs_crypt_mergeable_bio(io->bio, fio->page->mapping->host, - page_folio(bio_page)->index, fio))) + !f2fs_crypt_mergeable_bio(io->bio, fio_inode(fio), + bio_folio->index, fio))) __submit_merged_bio(io); alloc_new: if (io->bio == NULL) { io->bio = __bio_alloc(fio, BIO_MAX_VECS); - f2fs_set_bio_crypt_ctx(io->bio, fio->page->mapping->host, - page_folio(bio_page)->index, fio, GFP_NOIO); + f2fs_set_bio_crypt_ctx(io->bio, fio_inode(fio), + bio_folio->index, fio, GFP_NOIO); io->fio = *fio; } - if (bio_add_page(io->bio, bio_page, PAGE_SIZE, 0) < PAGE_SIZE) { + if (!bio_add_folio(io->bio, bio_folio, folio_size(bio_folio), 0)) { __submit_merged_bio(io); goto alloc_new; } if (fio->io_wbc) - wbc_account_cgroup_owner(fio->io_wbc, page_folio(fio->page), - PAGE_SIZE); + wbc_account_cgroup_owner(fio->io_wbc, fio->folio, + folio_size(fio->folio)); io->last_block_in_bio = fio->new_blkaddr; - trace_f2fs_submit_folio_write(page_folio(fio->page), fio); + trace_f2fs_submit_folio_write(fio->folio, fio); #ifdef CONFIG_BLK_DEV_ZONED if (f2fs_sb_has_blkzoned(sbi) && btype < META && is_end_zone_blkaddr(sbi, fio->new_blkaddr)) { @@ -1024,10 +1097,11 @@ out: if (is_sbi_flag_set(sbi, SBI_IS_SHUTDOWN) || !f2fs_is_checkpoint_ready(sbi)) __submit_merged_bio(io); - f2fs_up_write(&io->io_rwsem); + f2fs_up_write_trace(&io->io_rwsem, &lc); } -static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr, +static struct bio *f2fs_grab_read_bio(struct inode *inode, + struct fsverity_info *vi, block_t blkaddr, unsigned nr_pages, blk_opf_t op_flag, pgoff_t first_idx, bool for_write) { @@ -1041,8 +1115,6 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr, bio = bio_alloc_bioset(bdev, bio_max_segs(nr_pages), REQ_OP_READ | op_flag, for_write ? GFP_NOIO : GFP_KERNEL, &f2fs_bioset); - if (!bio) - return ERR_PTR(-ENOMEM); bio->bi_iter.bi_sector = sector; f2fs_set_bio_crypt_ctx(bio, inode, first_idx, NULL, GFP_NOFS); bio->bi_end_io = f2fs_read_end_io; @@ -1050,7 +1122,7 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr, if (fscrypt_inode_uses_fs_layer_crypto(inode)) post_read_steps |= STEP_DECRYPT; - if (f2fs_need_verity(inode, first_idx)) + if (vi) post_read_steps |= STEP_VERITY; /* @@ -1065,6 +1137,7 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr, ctx = mempool_alloc(bio_post_read_ctx_pool, GFP_NOFS); ctx->bio = bio; ctx->sbi = sbi; + ctx->vi = vi; ctx->enabled_steps = post_read_steps; ctx->fs_blkaddr = blkaddr; ctx->decompression_attempted = false; @@ -1076,37 +1149,30 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr, } /* This can handle encryption stuffs */ -static int f2fs_submit_page_read(struct inode *inode, struct folio *folio, - block_t blkaddr, blk_opf_t op_flags, - bool for_write) +static void f2fs_submit_page_read(struct inode *inode, struct fsverity_info *vi, + struct folio *folio, block_t blkaddr, + blk_opf_t op_flags, bool for_write) { struct f2fs_sb_info *sbi = F2FS_I_SB(inode); struct bio *bio; - bio = f2fs_grab_read_bio(inode, blkaddr, 1, op_flags, - folio->index, for_write); - if (IS_ERR(bio)) - return PTR_ERR(bio); + bio = f2fs_grab_read_bio(inode, vi, blkaddr, 1, op_flags, folio->index, + for_write); /* wait for GCed page writeback via META_MAPPING */ f2fs_wait_on_block_writeback(inode, blkaddr); - if (!bio_add_folio(bio, folio, PAGE_SIZE, 0)) { - iostat_update_and_unbind_ctx(bio); - if (bio->bi_private) - mempool_free(bio->bi_private, bio_post_read_ctx_pool); - bio_put(bio); - return -EFAULT; - } + if (!bio_add_folio(bio, folio, PAGE_SIZE, 0)) + f2fs_bug_on(sbi, 1); + inc_page_count(sbi, F2FS_RD_DATA); f2fs_update_iostat(sbi, NULL, FS_DATA_READ_IO, F2FS_BLKSIZE); f2fs_submit_read_bio(sbi, bio, DATA); - return 0; } static void __set_data_blkaddr(struct dnode_of_data *dn, block_t blkaddr) { - __le32 *addr = get_dnode_addr(dn->inode, dn->node_page); + __le32 *addr = get_dnode_addr(dn->inode, dn->node_folio); dn->data_blkaddr = blkaddr; addr[dn->ofs_in_node] = cpu_to_le32(dn->data_blkaddr); @@ -1115,14 +1181,14 @@ static void __set_data_blkaddr(struct dnode_of_data *dn, block_t blkaddr) /* * Lock ordering for the change of data block address: * ->data_page - * ->node_page + * ->node_folio * update block addresses in the node page */ void f2fs_set_data_blkaddr(struct dnode_of_data *dn, block_t blkaddr) { - f2fs_wait_on_page_writeback(dn->node_page, NODE, true, true); + f2fs_folio_wait_writeback(dn->node_folio, NODE, true, true); __set_data_blkaddr(dn, blkaddr); - if (set_page_dirty(dn->node_page)) + if (folio_mark_dirty(dn->node_folio)) dn->node_changed = true; } @@ -1150,7 +1216,7 @@ int f2fs_reserve_new_blocks(struct dnode_of_data *dn, blkcnt_t count) trace_f2fs_reserve_new_blocks(dn->inode, dn->nid, dn->ofs_in_node, count); - f2fs_wait_on_page_writeback(dn->node_page, NODE, true, true); + f2fs_folio_wait_writeback(dn->node_folio, NODE, true, true); for (; count > 0; dn->ofs_in_node++) { block_t blkaddr = f2fs_data_blkaddr(dn); @@ -1161,7 +1227,7 @@ int f2fs_reserve_new_blocks(struct dnode_of_data *dn, blkcnt_t count) } } - if (set_page_dirty(dn->node_page)) + if (folio_mark_dirty(dn->node_folio)) dn->node_changed = true; return 0; } @@ -1179,7 +1245,7 @@ int f2fs_reserve_new_block(struct dnode_of_data *dn) int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index) { - bool need_put = dn->inode_page ? false : true; + bool need_put = dn->inode_folio ? false : true; int err; err = f2fs_get_dnode_of_data(dn, index, ALLOC_NODE); @@ -1193,18 +1259,35 @@ int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index) return err; } -struct page *f2fs_get_read_data_page(struct inode *inode, pgoff_t index, - blk_opf_t op_flags, bool for_write, - pgoff_t *next_pgofs) +static inline struct fsverity_info *f2fs_need_verity(const struct inode *inode, + pgoff_t idx) +{ + if (idx < DIV_ROUND_UP(inode->i_size, PAGE_SIZE)) + return fsverity_get_info(inode); + return NULL; +} + +struct folio *f2fs_get_read_data_folio(struct inode *inode, pgoff_t index, + blk_opf_t op_flags, bool for_write, pgoff_t *next_pgofs) { struct address_space *mapping = inode->i_mapping; struct dnode_of_data dn; - struct page *page; + struct folio *folio; int err; +retry: + folio = f2fs_grab_cache_folio(mapping, index, for_write); + if (IS_ERR(folio)) + return folio; - page = f2fs_grab_cache_page(mapping, index, for_write); - if (!page) - return ERR_PTR(-ENOMEM); + if (folio_test_large(folio)) { + pgoff_t folio_index = mapping_align_index(mapping, index); + + f2fs_folio_put(folio, true); + invalidate_inode_pages2_range(mapping, folio_index, + folio_index + folio_nr_pages(folio) - 1); + f2fs_schedule_timeout(DEFAULT_SCHEDULE_TIMEOUT); + goto retry; + } if (f2fs_lookup_read_extent_cache_block(inode, index, &dn.data_blkaddr)) { @@ -1239,61 +1322,62 @@ struct page *f2fs_get_read_data_page(struct inode *inode, pgoff_t index, goto put_err; } got_it: - if (PageUptodate(page)) { - unlock_page(page); - return page; + if (folio_test_uptodate(folio)) { + folio_unlock(folio); + return folio; } /* * A new dentry page is allocated but not able to be written, since its * new inode page couldn't be allocated due to -ENOSPC. * In such the case, its blkaddr can be remained as NEW_ADDR. - * see, f2fs_add_link -> f2fs_get_new_data_page -> + * see, f2fs_add_link -> f2fs_get_new_data_folio -> * f2fs_init_inode_metadata. */ if (dn.data_blkaddr == NEW_ADDR) { - zero_user_segment(page, 0, PAGE_SIZE); - if (!PageUptodate(page)) - SetPageUptodate(page); - unlock_page(page); - return page; + folio_zero_segment(folio, 0, folio_size(folio)); + if (!folio_test_uptodate(folio)) + folio_mark_uptodate(folio); + folio_unlock(folio); + return folio; } - err = f2fs_submit_page_read(inode, page_folio(page), dn.data_blkaddr, - op_flags, for_write); - if (err) - goto put_err; - return page; + f2fs_submit_page_read(inode, f2fs_need_verity(inode, folio->index), + folio, dn.data_blkaddr, op_flags, for_write); + return folio; put_err: - f2fs_put_page(page, 1); + f2fs_folio_put(folio, true); return ERR_PTR(err); } -struct page *f2fs_find_data_page(struct inode *inode, pgoff_t index, +struct folio *f2fs_find_data_folio(struct inode *inode, pgoff_t index, pgoff_t *next_pgofs) { struct address_space *mapping = inode->i_mapping; - struct page *page; + struct folio *folio; - page = find_get_page_flags(mapping, index, FGP_ACCESSED); - if (page && PageUptodate(page)) - return page; - f2fs_put_page(page, 0); + folio = f2fs_filemap_get_folio(mapping, index, FGP_ACCESSED, 0); + if (IS_ERR(folio)) + goto read; + if (folio_test_uptodate(folio)) + return folio; + f2fs_folio_put(folio, false); - page = f2fs_get_read_data_page(inode, index, 0, false, next_pgofs); - if (IS_ERR(page)) - return page; +read: + folio = f2fs_get_read_data_folio(inode, index, 0, false, next_pgofs); + if (IS_ERR(folio)) + return folio; - if (PageUptodate(page)) - return page; + if (folio_test_uptodate(folio)) + return folio; - wait_on_page_locked(page); - if (unlikely(!PageUptodate(page))) { - f2fs_put_page(page, 0); + folio_wait_locked(folio); + if (unlikely(!folio_test_uptodate(folio))) { + f2fs_folio_put(folio, false); return ERR_PTR(-EIO); } - return page; + return folio; } /* @@ -1301,23 +1385,23 @@ struct page *f2fs_find_data_page(struct inode *inode, pgoff_t index, * Because, the callers, functions in dir.c and GC, should be able to know * whether this page exists or not. */ -struct page *f2fs_get_lock_data_page(struct inode *inode, pgoff_t index, +struct folio *f2fs_get_lock_data_folio(struct inode *inode, pgoff_t index, bool for_write) { struct address_space *mapping = inode->i_mapping; - struct page *page; + struct folio *folio; - page = f2fs_get_read_data_page(inode, index, 0, for_write, NULL); - if (IS_ERR(page)) - return page; + folio = f2fs_get_read_data_folio(inode, index, 0, for_write, NULL); + if (IS_ERR(folio)) + return folio; /* wait for read completion */ - lock_page(page); - if (unlikely(page->mapping != mapping || !PageUptodate(page))) { - f2fs_put_page(page, 1); + folio_lock(folio); + if (unlikely(folio->mapping != mapping || !folio_test_uptodate(folio))) { + f2fs_folio_put(folio, true); return ERR_PTR(-EIO); } - return page; + return folio; } /* @@ -1326,57 +1410,57 @@ struct page *f2fs_get_lock_data_page(struct inode *inode, pgoff_t index, * * Also, caller should grab and release a rwsem by calling f2fs_lock_op() and * f2fs_unlock_op(). - * Note that, ipage is set only by make_empty_dir, and if any error occur, - * ipage should be released by this function. + * Note that, ifolio is set only by make_empty_dir, and if any error occur, + * ifolio should be released by this function. */ -struct page *f2fs_get_new_data_page(struct inode *inode, - struct page *ipage, pgoff_t index, bool new_i_size) +struct folio *f2fs_get_new_data_folio(struct inode *inode, + struct folio *ifolio, pgoff_t index, bool new_i_size) { struct address_space *mapping = inode->i_mapping; - struct page *page; + struct folio *folio; struct dnode_of_data dn; int err; - page = f2fs_grab_cache_page(mapping, index, true); - if (!page) { + folio = f2fs_grab_cache_folio(mapping, index, true); + if (IS_ERR(folio)) { /* - * before exiting, we should make sure ipage will be released + * before exiting, we should make sure ifolio will be released * if any error occur. */ - f2fs_put_page(ipage, 1); + f2fs_folio_put(ifolio, true); return ERR_PTR(-ENOMEM); } - set_new_dnode(&dn, inode, ipage, NULL, 0); + set_new_dnode(&dn, inode, ifolio, NULL, 0); err = f2fs_reserve_block(&dn, index); if (err) { - f2fs_put_page(page, 1); + f2fs_folio_put(folio, true); return ERR_PTR(err); } - if (!ipage) + if (!ifolio) f2fs_put_dnode(&dn); - if (PageUptodate(page)) + if (folio_test_uptodate(folio)) goto got_it; if (dn.data_blkaddr == NEW_ADDR) { - zero_user_segment(page, 0, PAGE_SIZE); - if (!PageUptodate(page)) - SetPageUptodate(page); + folio_zero_segment(folio, 0, folio_size(folio)); + if (!folio_test_uptodate(folio)) + folio_mark_uptodate(folio); } else { - f2fs_put_page(page, 1); + f2fs_folio_put(folio, true); - /* if ipage exists, blkaddr should be NEW_ADDR */ - f2fs_bug_on(F2FS_I_SB(inode), ipage); - page = f2fs_get_lock_data_page(inode, index, true); - if (IS_ERR(page)) - return page; + /* if ifolio exists, blkaddr should be NEW_ADDR */ + f2fs_bug_on(F2FS_I_SB(inode), ifolio); + folio = f2fs_get_lock_data_folio(inode, index, true); + if (IS_ERR(folio)) + return folio; } got_it: if (new_i_size && i_size_read(inode) < ((loff_t)(index + 1) << PAGE_SHIFT)) f2fs_i_size_write(inode, ((loff_t)(index + 1) << PAGE_SHIFT)); - return page; + return folio; } static int __allocate_data_block(struct dnode_of_data *dn, int seg_type) @@ -1416,32 +1500,37 @@ static int __allocate_data_block(struct dnode_of_data *dn, int seg_type) return 0; } -static void f2fs_map_lock(struct f2fs_sb_info *sbi, int flag) +static void f2fs_map_lock(struct f2fs_sb_info *sbi, + struct f2fs_lock_context *lc, + int flag) { if (flag == F2FS_GET_BLOCK_PRE_AIO) - f2fs_down_read(&sbi->node_change); + f2fs_down_read_trace(&sbi->node_change, lc); else - f2fs_lock_op(sbi); + f2fs_lock_op(sbi, lc); } -static void f2fs_map_unlock(struct f2fs_sb_info *sbi, int flag) +static void f2fs_map_unlock(struct f2fs_sb_info *sbi, + struct f2fs_lock_context *lc, + int flag) { if (flag == F2FS_GET_BLOCK_PRE_AIO) - f2fs_up_read(&sbi->node_change); + f2fs_up_read_trace(&sbi->node_change, lc); else - f2fs_unlock_op(sbi); + f2fs_unlock_op(sbi, lc); } int f2fs_get_block_locked(struct dnode_of_data *dn, pgoff_t index) { struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode); + struct f2fs_lock_context lc; int err = 0; - f2fs_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO); + f2fs_map_lock(sbi, &lc, F2FS_GET_BLOCK_PRE_AIO); if (!f2fs_lookup_read_extent_cache_block(dn->inode, index, &dn->data_blkaddr)) err = f2fs_reserve_block(dn, index); - f2fs_map_unlock(sbi, F2FS_GET_BLOCK_PRE_AIO); + f2fs_map_unlock(sbi, &lc, F2FS_GET_BLOCK_PRE_AIO); return err; } @@ -1490,13 +1579,14 @@ static bool f2fs_map_blocks_cached(struct inode *inode, f2fs_wait_on_block_writeback_range(inode, map->m_pblk, map->m_len); - if (f2fs_allow_multi_device_dio(sbi, flag)) { + map->m_multidev_dio = f2fs_allow_multi_device_dio(sbi, flag); + if (map->m_multidev_dio) { int bidx = f2fs_target_device_index(sbi, map->m_pblk); struct f2fs_dev_info *dev = &sbi->devs[bidx]; map->m_bdev = dev->bdev; - map->m_pblk -= dev->start_blk; map->m_len = min(map->m_len, dev->end_blk + 1 - map->m_pblk); + map->m_pblk -= dev->start_blk; } else { map->m_bdev = inode->i_sb->s_bdev; } @@ -1532,6 +1622,7 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag) unsigned int maxblocks = map->m_len; struct dnode_of_data dn; struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + struct f2fs_lock_context lc; int mode = map->m_may_create ? ALLOC_NODE : LOOKUP_NODE; pgoff_t pgofs, end_offset, end; int err = 0, ofs = 1; @@ -1541,12 +1632,34 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag) unsigned int start_pgofs; int bidx = 0; bool is_hole; + bool lfs_dio_write; if (!maxblocks) return 0; - if (!map->m_may_create && f2fs_map_blocks_cached(inode, map, flag)) - goto out; + lfs_dio_write = (flag == F2FS_GET_BLOCK_DIO && f2fs_lfs_mode(sbi) && + map->m_may_create); + + if (!map->m_may_create && f2fs_map_blocks_cached(inode, map, flag)) { + struct extent_info ei; + + /* + * 1. If map->m_multidev_dio is true, map->m_pblk cannot be + * waitted by f2fs_wait_on_block_writeback_range() and are not + * mergeable. + * 2. If pgofs hits the read extent cache, it means the mapping + * is already cached in the extent cache, but it is not + * mergeable, and there is no need to query the mapping again + * via f2fs_get_dnode_of_data(). + */ + pgofs = (pgoff_t)map->m_lblk + map->m_len; + if (map->m_len == maxblocks || + map->m_multidev_dio || + f2fs_lookup_read_extent_cache(inode, pgofs, &ei)) + goto out; + ofs = map->m_len; + goto map_more; + } map->m_bdev = inode->i_sb->s_bdev; map->m_multidev_dio = @@ -1557,11 +1670,18 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map, int flag) /* it only supports block size == page size */ pgofs = (pgoff_t)map->m_lblk; - end = pgofs + maxblocks; +map_more: + end = (pgoff_t)map->m_lblk + maxblocks; + + if (flag == F2FS_GET_BLOCK_PRECACHE) + mode = LOOKUP_NODE_RA; next_dnode: - if (map->m_may_create) - f2fs_map_lock(sbi, flag); + if (map->m_may_create) { + if (f2fs_lfs_mode(sbi)) + f2fs_balance_fs(sbi, true); + f2fs_map_lock(sbi, &lc, flag); + } /* When reading holes, we need its node page */ set_new_dnode(&dn, inode, NULL, NULL, 0); @@ -1577,7 +1697,7 @@ next_dnode: start_pgofs = pgofs; prealloc = 0; last_ofs_in_node = ofs_in_node = dn.ofs_in_node; - end_offset = ADDRS_PER_PAGE(dn.node_page, inode); + end_offset = ADDRS_PER_PAGE(dn.node_folio, inode); next_block: blkaddr = f2fs_data_blkaddr(&dn); @@ -1591,7 +1711,7 @@ next_block: /* use out-place-update for direct IO under LFS mode */ if (map->m_may_create && (is_hole || (flag == F2FS_GET_BLOCK_DIO && f2fs_lfs_mode(sbi) && - !f2fs_is_pinned_file(inode)))) { + !f2fs_is_pinned_file(inode) && map->m_last_pblk != blkaddr))) { if (unlikely(f2fs_cp_error(sbi))) { err = -EIO; goto sync_out; @@ -1675,10 +1795,15 @@ next_block: if (map->m_multidev_dio) map->m_bdev = FDEV(bidx).bdev; + + if (lfs_dio_write) + map->m_last_pblk = NULL_ADDR; } else if (map_is_mergeable(sbi, map, blkaddr, flag, bidx, ofs)) { ofs++; map->m_len++; } else { + if (lfs_dio_write && !f2fs_is_pinned_file(inode)) + map->m_last_pblk = blkaddr; goto sync_out; } @@ -1703,14 +1828,6 @@ skip: dn.ofs_in_node = end_offset; } - if (flag == F2FS_GET_BLOCK_DIO && f2fs_lfs_mode(sbi) && - map->m_may_create) { - /* the next block to be allocated may not be contiguous. */ - if (GET_SEGOFF_FROM_SEG0(sbi, blkaddr) % BLKS_PER_SEC(sbi) == - CAP_BLKS_PER_SEC(sbi) - 1) - goto sync_out; - } - if (pgofs >= end) goto sync_out; else if (dn.ofs_in_node < end_offset) @@ -1729,7 +1846,7 @@ skip: f2fs_put_dnode(&dn); if (map->m_may_create) { - f2fs_map_unlock(sbi, flag); + f2fs_map_unlock(sbi, &lc, flag); f2fs_balance_fs(sbi, dn.node_changed); } goto next_dnode; @@ -1765,17 +1882,18 @@ sync_out: if (map->m_flags & F2FS_MAP_MAPPED) { unsigned int ofs = start_pgofs - map->m_lblk; - f2fs_update_read_extent_cache_range(&dn, - start_pgofs, map->m_pblk + ofs, - map->m_len - ofs); + if (map->m_len > ofs) + f2fs_update_read_extent_cache_range(&dn, + start_pgofs, map->m_pblk + ofs, + map->m_len - ofs); } if (map->m_next_extent) - *map->m_next_extent = pgofs + 1; + *map->m_next_extent = is_hole ? pgofs + 1 : pgofs; } f2fs_put_dnode(&dn); unlock_out: if (map->m_may_create) { - f2fs_map_unlock(sbi, flag); + f2fs_map_unlock(sbi, &lc, flag); f2fs_balance_fs(sbi, dn.node_changed); } out: @@ -1783,7 +1901,8 @@ out: return err; } -bool f2fs_overwrite_io(struct inode *inode, loff_t pos, size_t len) +static bool __f2fs_overwrite_io(struct inode *inode, loff_t pos, size_t len, + bool check_first) { struct f2fs_map_blocks map; block_t last_lblk; @@ -1805,15 +1924,21 @@ bool f2fs_overwrite_io(struct inode *inode, loff_t pos, size_t len) if (err || map.m_len == 0) return false; map.m_lblk += map.m_len; + if (check_first) + break; } return true; } +bool f2fs_overwrite_io(struct inode *inode, loff_t pos, size_t len) +{ + return __f2fs_overwrite_io(inode, pos, len, false); +} + static int f2fs_xattr_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo) { struct f2fs_sb_info *sbi = F2FS_I_SB(inode); - struct page *page; struct node_info ni; __u64 phys = 0, len; __u32 flags; @@ -1822,15 +1947,15 @@ static int f2fs_xattr_fiemap(struct inode *inode, if (f2fs_has_inline_xattr(inode)) { int offset; + struct folio *folio = f2fs_grab_cache_folio(NODE_MAPPING(sbi), + inode->i_ino, false); - page = f2fs_grab_cache_page(NODE_MAPPING(sbi), - inode->i_ino, false); - if (!page) - return -ENOMEM; + if (IS_ERR(folio)) + return PTR_ERR(folio); err = f2fs_get_node_info(sbi, inode->i_ino, &ni, false); if (err) { - f2fs_put_page(page, 1); + f2fs_folio_put(folio, true); return err; } @@ -1842,7 +1967,7 @@ static int f2fs_xattr_fiemap(struct inode *inode, phys += offset; len = inline_xattr_size(inode); - f2fs_put_page(page, 1); + f2fs_folio_put(folio, true); flags = FIEMAP_EXTENT_DATA_INLINE | FIEMAP_EXTENT_NOT_ALIGNED; @@ -1856,20 +1981,22 @@ static int f2fs_xattr_fiemap(struct inode *inode, } if (xnid) { - page = f2fs_grab_cache_page(NODE_MAPPING(sbi), xnid, false); - if (!page) - return -ENOMEM; + struct folio *folio = f2fs_grab_cache_folio(NODE_MAPPING(sbi), + xnid, false); + + if (IS_ERR(folio)) + return PTR_ERR(folio); err = f2fs_get_node_info(sbi, xnid, &ni, false); if (err) { - f2fs_put_page(page, 1); + f2fs_folio_put(folio, true); return err; } phys = F2FS_BLK_TO_BYTES(ni.blk_addr); len = inode->i_sb->s_blocksize; - f2fs_put_page(page, 1); + f2fs_folio_put(folio, true); flags = FIEMAP_EXTENT_LAST; } @@ -2052,12 +2179,12 @@ static inline blk_opf_t f2fs_ra_op_flags(struct readahead_control *rac) return rac ? REQ_RAHEAD : 0; } -static int f2fs_read_single_page(struct inode *inode, struct folio *folio, - unsigned nr_pages, - struct f2fs_map_blocks *map, - struct bio **bio_ret, - sector_t *last_block_in_bio, - struct readahead_control *rac) +static int f2fs_read_single_page(struct inode *inode, struct fsverity_info *vi, + struct folio *folio, unsigned int nr_pages, + struct f2fs_map_blocks *map, + struct bio **bio_ret, + sector_t *last_block_in_bio, + struct readahead_control *rac) { struct bio *bio = *bio_ret; const unsigned int blocksize = F2FS_BLKSIZE; @@ -2065,7 +2192,7 @@ static int f2fs_read_single_page(struct inode *inode, struct folio *folio, sector_t last_block; sector_t last_block_in_file; sector_t block_nr; - pgoff_t index = folio_index(folio); + pgoff_t index = folio->index; int ret = 0; block_in_file = (sector_t)index; @@ -2081,10 +2208,13 @@ static int f2fs_read_single_page(struct inode *inode, struct folio *folio, /* * Map blocks using the previous result first. */ - if ((map->m_flags & F2FS_MAP_MAPPED) && - block_in_file > map->m_lblk && + if (map->m_flags & F2FS_MAP_MAPPED) { + if (block_in_file > map->m_lblk && block_in_file < (map->m_lblk + map->m_len)) + goto got_it; + } else if (block_in_file < *map->m_next_pgofs) { goto got_it; + } /* * Then do more f2fs_map_blocks() calls until we are @@ -2109,8 +2239,7 @@ got_it: } else { zero_out: folio_zero_segment(folio, 0, folio_size(folio)); - if (f2fs_need_verity(inode, index) && - !fsverity_verify_folio(folio)) { + if (vi && !fsverity_verify_folio(vi, folio)) { ret = -EIO; goto out; } @@ -2131,16 +2260,9 @@ submit_and_realloc: f2fs_submit_read_bio(F2FS_I_SB(inode), bio, DATA); bio = NULL; } - if (bio == NULL) { - bio = f2fs_grab_read_bio(inode, block_nr, nr_pages, - f2fs_ra_op_flags(rac), index, - false); - if (IS_ERR(bio)) { - ret = PTR_ERR(bio); - bio = NULL; - goto out; - } - } + if (bio == NULL) + bio = f2fs_grab_read_bio(inode, vi, block_nr, nr_pages, + f2fs_ra_op_flags(rac), index, false); /* * If the page is under writeback, we need to wait for @@ -2178,6 +2300,12 @@ int f2fs_read_multi_pages(struct compress_ctx *cc, struct bio **bio_ret, int i; int ret = 0; + if (unlikely(f2fs_cp_error(sbi))) { + ret = -EIO; + from_dnode = false; + goto out_put_dnode; + } + f2fs_bug_on(sbi, f2fs_cluster_is_empty(cc)); last_block_in_file = F2FS_BYTES_TO_BLK(f2fs_readpage_limit(inode) + @@ -2221,17 +2349,13 @@ int f2fs_read_multi_pages(struct compress_ctx *cc, struct bio **bio_ret, if (ret) goto out; - if (unlikely(f2fs_cp_error(sbi))) { - ret = -EIO; - goto out_put_dnode; - } f2fs_bug_on(sbi, dn.data_blkaddr != COMPRESS_ADDR); skip_reading_dnode: for (i = 1; i < cc->cluster_size; i++) { block_t blkaddr; - blkaddr = from_dnode ? data_blkaddr(dn.inode, dn.node_page, + blkaddr = from_dnode ? data_blkaddr(dn.inode, dn.node_folio, dn.ofs_in_node + i) : ei.blk + i - 1; @@ -2265,14 +2389,13 @@ skip_reading_dnode: block_t blkaddr; struct bio_post_read_ctx *ctx; - blkaddr = from_dnode ? data_blkaddr(dn.inode, dn.node_page, + blkaddr = from_dnode ? data_blkaddr(dn.inode, dn.node_folio, dn.ofs_in_node + i + 1) : ei.blk + i; f2fs_wait_on_block_writeback(inode, blkaddr); - if (f2fs_load_compressed_page(sbi, folio_page(folio, 0), - blkaddr)) { + if (f2fs_load_compressed_folio(sbi, folio, blkaddr)) { if (atomic_dec_and_test(&dic->remaining_pages)) { f2fs_decompress_cluster(dic, true); break; @@ -2288,18 +2411,11 @@ submit_and_realloc: bio = NULL; } - if (!bio) { - bio = f2fs_grab_read_bio(inode, blkaddr, nr_pages, - f2fs_ra_op_flags(rac), - folio->index, for_write); - if (IS_ERR(bio)) { - ret = PTR_ERR(bio); - f2fs_decompress_end_io(dic, ret, true); - f2fs_put_dnode(&dn); - *bio_ret = NULL; - return ret; - } - } + if (!bio) + bio = f2fs_grab_read_bio(inode, cc->vi, blkaddr, + nr_pages - i, + f2fs_ra_op_flags(rac), + folio->index, for_write); if (!bio_add_folio(bio, folio, blocksize, 0)) goto submit_and_realloc; @@ -2334,11 +2450,192 @@ out: } #endif +static struct f2fs_folio_state *ffs_find_or_alloc(struct folio *folio) +{ + struct f2fs_folio_state *ffs = folio->private; + + if (ffs) + return ffs; + + ffs = f2fs_kmem_cache_alloc(ffs_entry_slab, + GFP_NOIO | __GFP_ZERO, true, NULL); + + spin_lock_init(&ffs->state_lock); + folio_attach_private(folio, ffs); + return ffs; +} + +static void ffs_detach_free(struct folio *folio) +{ + struct f2fs_folio_state *ffs; + + if (!folio_test_large(folio)) { + folio_detach_private(folio); + return; + } + + ffs = folio_detach_private(folio); + if (!ffs) + return; + + WARN_ON_ONCE(ffs->read_pages_pending != 0); + kmem_cache_free(ffs_entry_slab, ffs); +} + +static int f2fs_read_data_large_folio(struct inode *inode, + struct fsverity_info *vi, + struct readahead_control *rac, struct folio *folio) +{ + struct bio *bio = NULL; + sector_t last_block_in_bio = 0; + struct f2fs_map_blocks map = {0, }; + pgoff_t index, offset, next_pgofs = 0; + unsigned max_nr_pages = rac ? readahead_count(rac) : + folio_nr_pages(folio); + unsigned nrpages; + struct f2fs_folio_state *ffs; + int ret = 0; + bool folio_in_bio; + + if (!IS_IMMUTABLE(inode) || f2fs_compressed_file(inode)) { + if (folio) + folio_unlock(folio); + return -EOPNOTSUPP; + } + + map.m_seg_type = NO_CHECK_TYPE; + + if (rac) + folio = readahead_folio(rac); +next_folio: + if (!folio) + goto out; + + f2fs_update_read_folio_count(F2FS_I_SB(inode), folio); + + folio_in_bio = false; + index = folio->index; + offset = 0; + ffs = NULL; + nrpages = folio_nr_pages(folio); + + for (; nrpages; nrpages--, max_nr_pages--, index++, offset++) { + sector_t block_nr; + /* + * Map blocks using the previous result first. + */ + if (map.m_flags & F2FS_MAP_MAPPED) { + if (index > map.m_lblk && + index < (map.m_lblk + map.m_len)) + goto got_it; + } else if (index < next_pgofs) { + /* hole case */ + goto got_it; + } + + /* + * Then do more f2fs_map_blocks() calls until we are + * done with this page. + */ + memset(&map, 0, sizeof(map)); + map.m_next_pgofs = &next_pgofs; + map.m_seg_type = NO_CHECK_TYPE; + map.m_lblk = index; + map.m_len = max_nr_pages; + + ret = f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_DEFAULT); + if (ret) + goto err_out; +got_it: + if ((map.m_flags & F2FS_MAP_MAPPED)) { + block_nr = map.m_pblk + index - map.m_lblk; + if (!f2fs_is_valid_blkaddr(F2FS_I_SB(inode), block_nr, + DATA_GENERIC_ENHANCE_READ)) { + ret = -EFSCORRUPTED; + goto err_out; + } + } else { + size_t page_offset = offset << PAGE_SHIFT; + folio_zero_range(folio, page_offset, PAGE_SIZE); + if (vi && !fsverity_verify_blocks(vi, folio, PAGE_SIZE, page_offset)) { + ret = -EIO; + goto err_out; + } + continue; + } + + /* We must increment read_pages_pending before possible BIOs submitting + * to prevent from premature folio_end_read() call on folio + */ + if (folio_test_large(folio)) { + ffs = ffs_find_or_alloc(folio); + + /* set the bitmap to wait */ + spin_lock_irq(&ffs->state_lock); + ffs->read_pages_pending++; + spin_unlock_irq(&ffs->state_lock); + } + + /* + * This page will go to BIO. Do we need to send this + * BIO off first? + */ + if (bio && (!page_is_mergeable(F2FS_I_SB(inode), bio, + last_block_in_bio, block_nr) || + !f2fs_crypt_mergeable_bio(bio, inode, index, NULL))) { +submit_and_realloc: + f2fs_submit_read_bio(F2FS_I_SB(inode), bio, DATA); + bio = NULL; + } + if (bio == NULL) + bio = f2fs_grab_read_bio(inode, vi, + block_nr, max_nr_pages, + f2fs_ra_op_flags(rac), + index, false); + + /* + * If the page is under writeback, we need to wait for + * its completion to see the correct decrypted data. + */ + f2fs_wait_on_block_writeback(inode, block_nr); + + if (!bio_add_folio(bio, folio, F2FS_BLKSIZE, + offset << PAGE_SHIFT)) + goto submit_and_realloc; + + folio_in_bio = true; + inc_page_count(F2FS_I_SB(inode), F2FS_RD_DATA); + f2fs_update_iostat(F2FS_I_SB(inode), NULL, FS_DATA_READ_IO, + F2FS_BLKSIZE); + last_block_in_bio = block_nr; + } + trace_f2fs_read_folio(folio, DATA); +err_out: + if (!folio_in_bio) { + folio_end_read(folio, !ret); + if (ret) + return ret; + } + if (rac) { + folio = readahead_folio(rac); + goto next_folio; + } +out: + f2fs_submit_read_bio(F2FS_I_SB(inode), bio, DATA); + if (ret) { + /* Wait bios and clear uptodate. */ + folio_lock(folio); + folio_clear_uptodate(folio); + folio_unlock(folio); + } + return ret; +} + /* * This function was originally taken from fs/mpage.c, and customized for f2fs. * Major change was from block_size == page_size in f2fs by default. */ -static int f2fs_mpage_readpages(struct inode *inode, +static int f2fs_mpage_readpages(struct inode *inode, struct fsverity_info *vi, struct readahead_control *rac, struct folio *folio) { struct bio *bio = NULL; @@ -2358,15 +2655,28 @@ static int f2fs_mpage_readpages(struct inode *inode, pgoff_t nc_cluster_idx = NULL_CLUSTER; pgoff_t index; #endif + pgoff_t next_pgofs = 0; unsigned nr_pages = rac ? readahead_count(rac) : 1; + struct address_space *mapping = rac ? rac->mapping : folio->mapping; unsigned max_nr_pages = nr_pages; int ret = 0; + if (mapping_large_folio_support(mapping)) + return f2fs_read_data_large_folio(inode, vi, rac, folio); + +#ifdef CONFIG_F2FS_FS_COMPRESSION + if (f2fs_compressed_file(inode)) { + index = rac ? readahead_index(rac) : folio->index; + max_nr_pages = round_up(index + nr_pages, cc.cluster_size) - + round_down(index, cc.cluster_size); + } +#endif + map.m_pblk = 0; map.m_lblk = 0; map.m_len = 0; map.m_flags = 0; - map.m_next_pgofs = NULL; + map.m_next_pgofs = &next_pgofs; map.m_next_extent = NULL; map.m_seg_type = NO_CHECK_TYPE; map.m_may_create = false; @@ -2377,14 +2687,17 @@ static int f2fs_mpage_readpages(struct inode *inode, prefetchw(&folio->flags); } + f2fs_update_read_folio_count(F2FS_I_SB(inode), folio); + #ifdef CONFIG_F2FS_FS_COMPRESSION - index = folio_index(folio); + index = folio->index; if (!f2fs_compressed_file(inode)) goto read_single_page; /* there are remained compressed pages, submit them */ if (!f2fs_cluster_can_merge_page(&cc, index)) { + cc.vi = vi; ret = f2fs_read_multi_pages(&cc, &bio, max_nr_pages, &last_block_in_bio, @@ -2418,8 +2731,9 @@ static int f2fs_mpage_readpages(struct inode *inode, read_single_page: #endif - ret = f2fs_read_single_page(inode, folio, max_nr_pages, &map, - &bio, &last_block_in_bio, rac); + ret = f2fs_read_single_page(inode, vi, folio, max_nr_pages, + &map, &bio, &last_block_in_bio, + rac); if (ret) { #ifdef CONFIG_F2FS_FS_COMPRESSION set_error_page: @@ -2435,6 +2749,7 @@ next_page: if (f2fs_compressed_file(inode)) { /* last page */ if (nr_pages == 1 && !f2fs_cluster_is_empty(&cc)) { + cc.vi = vi; ret = f2fs_read_multi_pages(&cc, &bio, max_nr_pages, &last_block_in_bio, @@ -2444,15 +2759,15 @@ next_page: } #endif } - if (bio) - f2fs_submit_read_bio(F2FS_I_SB(inode), bio, DATA); + f2fs_submit_read_bio(F2FS_I_SB(inode), bio, DATA); return ret; } static int f2fs_read_data_folio(struct file *file, struct folio *folio) { struct inode *inode = folio->mapping->host; - int ret = -EAGAIN; + struct fsverity_info *vi = NULL; + int ret; trace_f2fs_readpage(folio, DATA); @@ -2462,16 +2777,22 @@ static int f2fs_read_data_folio(struct file *file, struct folio *folio) } /* If the file has inline data, try to read it directly */ - if (f2fs_has_inline_data(inode)) + if (f2fs_has_inline_data(inode)) { ret = f2fs_read_inline_data(inode, folio); - if (ret == -EAGAIN) - ret = f2fs_mpage_readpages(inode, NULL, folio); - return ret; + if (ret != -EAGAIN) + return ret; + } + + vi = f2fs_need_verity(inode, folio->index); + if (vi) + fsverity_readahead(vi, folio->index, folio_nr_pages(folio)); + return f2fs_mpage_readpages(inode, vi, NULL, folio); } static void f2fs_readahead(struct readahead_control *rac) { struct inode *inode = rac->mapping->host; + struct fsverity_info *vi = NULL; trace_f2fs_readpages(inode, readahead_index(rac), readahead_count(rac)); @@ -2482,14 +2803,18 @@ static void f2fs_readahead(struct readahead_control *rac) if (f2fs_has_inline_data(inode)) return; - f2fs_mpage_readpages(inode, rac, NULL); + vi = f2fs_need_verity(inode, readahead_index(rac)); + if (vi) + fsverity_readahead(vi, readahead_index(rac), + readahead_count(rac)); + f2fs_mpage_readpages(inode, vi, rac, NULL); } int f2fs_encrypt_one_page(struct f2fs_io_info *fio) { - struct inode *inode = fio->page->mapping->host; - struct page *mpage, *page; - gfp_t gfp_flags = GFP_NOFS; + struct inode *inode = fio_inode(fio); + struct folio *mfolio; + struct page *page; if (!f2fs_encrypted_file(inode)) return 0; @@ -2499,26 +2824,17 @@ int f2fs_encrypt_one_page(struct f2fs_io_info *fio) if (fscrypt_inode_uses_inline_crypto(inode)) return 0; -retry_encrypt: - fio->encrypted_page = fscrypt_encrypt_pagecache_blocks(page, - PAGE_SIZE, 0, gfp_flags); - if (IS_ERR(fio->encrypted_page)) { - /* flush pending IOs and wait for a while in the ENOMEM case */ - if (PTR_ERR(fio->encrypted_page) == -ENOMEM) { - f2fs_flush_merged_writes(fio->sbi); - memalloc_retry_wait(GFP_NOFS); - gfp_flags |= __GFP_NOFAIL; - goto retry_encrypt; - } + fio->encrypted_page = fscrypt_encrypt_pagecache_blocks(page_folio(page), + PAGE_SIZE, 0, GFP_NOFS); + if (IS_ERR(fio->encrypted_page)) return PTR_ERR(fio->encrypted_page); - } - mpage = find_lock_page(META_MAPPING(fio->sbi), fio->old_blkaddr); - if (mpage) { - if (PageUptodate(mpage)) - memcpy(page_address(mpage), + mfolio = filemap_lock_folio(META_MAPPING(fio->sbi), fio->old_blkaddr); + if (!IS_ERR(mfolio)) { + if (folio_test_uptodate(mfolio)) + memcpy(folio_address(mfolio), page_address(fio->encrypted_page), PAGE_SIZE); - f2fs_put_page(mpage, 1); + f2fs_folio_put(mfolio, true); } return 0; } @@ -2617,7 +2933,7 @@ bool f2fs_should_update_outplace(struct inode *inode, struct f2fs_io_info *fio) static inline bool need_inplace_update(struct f2fs_io_info *fio) { - struct inode *inode = fio->page->mapping->host; + struct inode *inode = fio_inode(fio); if (f2fs_should_update_outplace(inode, fio)) return false; @@ -2627,17 +2943,18 @@ static inline bool need_inplace_update(struct f2fs_io_info *fio) int f2fs_do_write_data_page(struct f2fs_io_info *fio) { - struct folio *folio = page_folio(fio->page); + struct folio *folio = fio->folio; struct inode *inode = folio->mapping->host; struct dnode_of_data dn; struct node_info ni; + struct f2fs_lock_context lc; bool ipu_force = false; bool atomic_commit; int err = 0; /* Use COW inode to make dnode_of_data for atomic write */ atomic_commit = f2fs_is_atomic_file(inode) && - page_private_atomic(folio_page(folio, 0)); + folio_test_f2fs_atomic(folio); if (atomic_commit) set_new_dnode(&dn, F2FS_I(inode)->cow_inode, NULL, NULL, 0); else @@ -2655,8 +2972,12 @@ int f2fs_do_write_data_page(struct f2fs_io_info *fio) goto got_it; } + if (is_sbi_flag_set(fio->sbi, SBI_ENABLE_CHECKPOINT) && + time_to_inject(fio->sbi, FAULT_SKIP_WRITE)) + return -EINVAL; + /* Deadlock due to between page->lock and f2fs_lock_op */ - if (fio->need_lock == LOCK_REQ && !f2fs_trylock_op(fio->sbi)) + if (fio->need_lock == LOCK_REQ && !f2fs_trylock_op(fio->sbi, &lc)) return -EAGAIN; err = f2fs_get_dnode_of_data(&dn, folio->index, LOOKUP_NODE); @@ -2668,7 +2989,7 @@ int f2fs_do_write_data_page(struct f2fs_io_info *fio) /* This page is already truncated */ if (fio->old_blkaddr == NULL_ADDR) { folio_clear_uptodate(folio); - clear_page_private_gcing(folio_page(folio, 0)); + folio_clear_f2fs_gcing(folio); goto out_writepage; } got_it: @@ -2697,7 +3018,7 @@ got_it: folio_start_writeback(folio); f2fs_put_dnode(&dn); if (fio->need_lock == LOCK_REQ) - f2fs_unlock_op(fio->sbi); + f2fs_unlock_op(fio->sbi, &lc); err = f2fs_inplace_write_data(fio); if (err) { if (fscrypt_inode_uses_fs_layer_crypto(inode)) @@ -2711,7 +3032,7 @@ got_it: } if (fio->need_lock == LOCK_RETRY) { - if (!f2fs_trylock_op(fio->sbi)) { + if (!f2fs_trylock_op(fio->sbi, &lc)) { err = -EAGAIN; goto out_writepage; } @@ -2738,12 +3059,12 @@ got_it: trace_f2fs_do_write_data_page(folio, OPU); set_inode_flag(inode, FI_APPEND_WRITE); if (atomic_commit) - clear_page_private_atomic(folio_page(folio, 0)); + folio_clear_f2fs_atomic(folio); out_writepage: f2fs_put_dnode(&dn); out: if (fio->need_lock == LOCK_REQ) - f2fs_unlock_op(fio->sbi); + f2fs_unlock_op(fio->sbi, &lc); return err; } @@ -2756,7 +3077,6 @@ int f2fs_write_single_data_page(struct folio *folio, int *submitted, bool allow_balance) { struct inode *inode = folio->mapping->host; - struct page *page = folio_page(folio, 0); struct f2fs_sb_info *sbi = F2FS_I_SB(inode); loff_t i_size = i_size_read(inode); const pgoff_t end_index = ((unsigned long long)i_size) @@ -2773,7 +3093,7 @@ int f2fs_write_single_data_page(struct folio *folio, int *submitted, .op = REQ_OP_WRITE, .op_flags = wbc_to_write_flags(wbc), .old_blkaddr = NULL_ADDR, - .page = page, + .folio = folio, .encrypted_page = NULL, .submitted = 0, .compr_blocks = compr_blocks, @@ -2824,30 +3144,26 @@ int f2fs_write_single_data_page(struct folio *folio, int *submitted, write: /* Dentry/quota blocks are controlled by checkpoint */ if (S_ISDIR(inode->i_mode) || quota_inode) { + struct f2fs_lock_context lc; + /* * We need to wait for node_write to avoid block allocation during * checkpoint. This can only happen to quota writes which can cause * the below discard race condition. */ if (quota_inode) - f2fs_down_read(&sbi->node_write); + f2fs_down_read_trace(&sbi->node_write, &lc); fio.need_lock = LOCK_DONE; err = f2fs_do_write_data_page(&fio); if (quota_inode) - f2fs_up_read(&sbi->node_write); + f2fs_up_read_trace(&sbi->node_write, &lc); goto done; } - if (!wbc->for_reclaim) - need_balance_fs = true; - else if (has_not_enough_free_secs(sbi, 0, 0)) - goto redirty_out; - else - set_inode_flag(inode, FI_HOT_DATA); - + need_balance_fs = true; err = -EAGAIN; if (f2fs_has_inline_data(inode)) { err = f2fs_write_inline_data(inode, folio); @@ -2881,14 +3197,7 @@ out: inode_dec_dirty_pages(inode); if (err) { folio_clear_uptodate(folio); - clear_page_private_gcing(page); - } - - if (wbc->for_reclaim) { - f2fs_submit_merged_write_cond(sbi, NULL, page, 0, DATA); - clear_inode_flag(inode, FI_HOT_DATA); - f2fs_remove_dirty_inode(inode); - submitted = NULL; + folio_clear_f2fs_gcing(folio); } folio_unlock(folio); if (!S_ISDIR(inode->i_mode) && !IS_NOQUOTA(inode) && @@ -2915,35 +3224,12 @@ redirty_out: * file_write_and_wait_range() will see EIO error, which is critical * to return value of fsync() followed by atomic_write failure to user. */ - if (!err || wbc->for_reclaim) - return AOP_WRITEPAGE_ACTIVATE; folio_unlock(folio); + if (!err) + return 1; return err; } -static int f2fs_write_data_page(struct page *page, - struct writeback_control *wbc) -{ - struct folio *folio = page_folio(page); -#ifdef CONFIG_F2FS_FS_COMPRESSION - struct inode *inode = folio->mapping->host; - - if (unlikely(f2fs_cp_error(F2FS_I_SB(inode)))) - goto out; - - if (f2fs_compressed_file(inode)) { - if (f2fs_is_compressed_cluster(inode, folio->index)) { - folio_redirty_for_writepage(wbc, folio); - return AOP_WRITEPAGE_ACTIVATE; - } - } -out: -#endif - - return f2fs_write_single_data_page(folio, NULL, NULL, NULL, - wbc, FS_DATA_IO, 0, true); -} - /* * This function was copied from write_cache_pages from mm/page-writeback.c. * The major change is making write step of cold data page separately from @@ -3016,10 +3302,7 @@ static int f2fs_write_cache_pages(struct address_space *mapping, if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) range_whole = 1; } - if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages) - tag = PAGECACHE_TAG_TOWRITE; - else - tag = PAGECACHE_TAG_DIRTY; + tag = wbc_to_tag(wbc); retry: retry = 0; if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages) @@ -3137,7 +3420,7 @@ continue_unlock: if (folio_test_writeback(folio)) { if (wbc->sync_mode == WB_SYNC_NONE) goto continue_unlock; - f2fs_wait_on_page_writeback(&folio->page, DATA, true, true); + f2fs_folio_wait_writeback(folio, DATA, true, true); } if (!folio_clear_dirty_for_io(folio)) @@ -3154,8 +3437,6 @@ continue_unlock: ret = f2fs_write_single_data_page(folio, &submitted, &bio, &last_block, wbc, io_type, 0, true); - if (ret == AOP_WRITEPAGE_ACTIVATE) - folio_unlock(folio); #ifdef CONFIG_F2FS_FS_COMPRESSION result: #endif @@ -3167,14 +3448,14 @@ result: * keep nr_to_write, since vfs uses this to * get # of written pages. */ - if (ret == AOP_WRITEPAGE_ACTIVATE) { + if (ret == 1) { ret = 0; goto next; } else if (ret == -EAGAIN) { ret = 0; if (wbc->sync_mode == WB_SYNC_ALL) { - f2fs_io_schedule_timeout( - DEFAULT_IO_TIMEOUT); + f2fs_schedule_timeout( + DEFAULT_SCHEDULE_TIMEOUT); goto retry_write; } goto next; @@ -3247,6 +3528,8 @@ static inline bool __should_serialize_io(struct inode *inode, if (IS_NOQUOTA(inode)) return false; + if (f2fs_is_pinned_file(inode)) + return false; if (f2fs_need_compress_data(inode)) return true; if (wbc->sync_mode != WB_SYNC_ALL) @@ -3256,6 +3539,29 @@ static inline bool __should_serialize_io(struct inode *inode, return false; } +static inline void account_writeback(struct inode *inode, bool inc) +{ + if (!f2fs_sb_has_compression(F2FS_I_SB(inode))) + return; + + f2fs_down_read(&F2FS_I(inode)->i_sem); + if (inc) + atomic_inc(&F2FS_I(inode)->writeback); + else + atomic_dec(&F2FS_I(inode)->writeback); + f2fs_up_read(&F2FS_I(inode)->i_sem); +} + +static inline void update_skipped_write(struct f2fs_sb_info *sbi, + struct writeback_control *wbc) +{ + long skipped = wbc->pages_skipped; + + if (is_sbi_flag_set(sbi, SBI_ENABLE_CHECKPOINT) && skipped && + wbc->sync_mode == WB_SYNC_ALL) + atomic_add(skipped, &sbi->nr_pages[F2FS_SKIPPED_WRITE]); +} + static int __f2fs_write_data_pages(struct address_space *mapping, struct writeback_control *wbc, enum iostat_type io_type) @@ -3266,10 +3572,6 @@ static int __f2fs_write_data_pages(struct address_space *mapping, int ret; bool locked = false; - /* deal with chardevs and other special file */ - if (!mapping->a_ops->writepage) - return 0; - /* skip writing if there is no dirty page in this inode */ if (!get_dirty_pages(inode) && wbc->sync_mode == WB_SYNC_NONE) return 0; @@ -3305,10 +3607,14 @@ static int __f2fs_write_data_pages(struct address_space *mapping, locked = true; } + account_writeback(inode, true); + blk_start_plug(&plug); ret = f2fs_write_cache_pages(mapping, wbc, io_type); blk_finish_plug(&plug); + account_writeback(inode, false); + if (locked) mutex_unlock(&sbi->writepages); @@ -3320,10 +3626,19 @@ static int __f2fs_write_data_pages(struct address_space *mapping, */ f2fs_remove_dirty_inode(inode); + + /* + * f2fs_write_cache_pages() has retry logic for EAGAIN case which is + * common when racing w/ checkpoint, so only update skipped write + * when ret is non-zero. + */ + if (ret) + update_skipped_write(sbi, wbc); return ret; skip_write: wbc->pages_skipped += get_dirty_pages(inode); + update_skipped_write(sbi, wbc); trace_f2fs_writepages(mapping->host, wbc, DATA); return 0; } @@ -3365,7 +3680,8 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi, struct inode *inode = folio->mapping->host; pgoff_t index = folio->index; struct dnode_of_data dn; - struct page *ipage; + struct f2fs_lock_context lc; + struct folio *ifolio; bool locked = false; int flag = F2FS_GET_BLOCK_PRE_AIO; int err = 0; @@ -3381,32 +3697,32 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi, if (f2fs_has_inline_data(inode)) { if (pos + len > MAX_INLINE_DATA(inode)) flag = F2FS_GET_BLOCK_DEFAULT; - f2fs_map_lock(sbi, flag); + f2fs_map_lock(sbi, &lc, flag); locked = true; } else if ((pos & PAGE_MASK) >= i_size_read(inode)) { - f2fs_map_lock(sbi, flag); + f2fs_map_lock(sbi, &lc, flag); locked = true; } restart: /* check inline_data */ - ipage = f2fs_get_node_page(sbi, inode->i_ino); - if (IS_ERR(ipage)) { - err = PTR_ERR(ipage); + ifolio = f2fs_get_inode_folio(sbi, inode->i_ino); + if (IS_ERR(ifolio)) { + err = PTR_ERR(ifolio); goto unlock_out; } - set_new_dnode(&dn, inode, ipage, ipage, 0); + set_new_dnode(&dn, inode, ifolio, ifolio, 0); if (f2fs_has_inline_data(inode)) { if (pos + len <= MAX_INLINE_DATA(inode)) { - f2fs_do_read_inline_data(folio, ipage); + f2fs_do_read_inline_data(folio, ifolio); set_inode_flag(inode, FI_DATA_EXIST); if (inode->i_nlink) - set_page_private_inline(ipage); + folio_set_f2fs_inline(ifolio); goto out; } - err = f2fs_convert_inline_page(&dn, folio_page(folio, 0)); + err = f2fs_convert_inline_folio(&dn, folio); if (err || dn.data_blkaddr != NULL_ADDR) goto out; } @@ -3428,7 +3744,7 @@ restart: if (!err && dn.data_blkaddr != NULL_ADDR) goto out; f2fs_put_dnode(&dn); - f2fs_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO); + f2fs_map_lock(sbi, &lc, F2FS_GET_BLOCK_PRE_AIO); WARN_ON(flag != F2FS_GET_BLOCK_PRE_AIO); locked = true; goto restart; @@ -3442,7 +3758,7 @@ out: f2fs_put_dnode(&dn); unlock_out: if (locked) - f2fs_map_unlock(sbi, flag); + f2fs_map_unlock(sbi, &lc, flag); return err; } @@ -3450,14 +3766,14 @@ static int __find_data_block(struct inode *inode, pgoff_t index, block_t *blk_addr) { struct dnode_of_data dn; - struct page *ipage; + struct folio *ifolio; int err = 0; - ipage = f2fs_get_node_page(F2FS_I_SB(inode), inode->i_ino); - if (IS_ERR(ipage)) - return PTR_ERR(ipage); + ifolio = f2fs_get_inode_folio(F2FS_I_SB(inode), inode->i_ino); + if (IS_ERR(ifolio)) + return PTR_ERR(ifolio); - set_new_dnode(&dn, inode, ipage, ipage, 0); + set_new_dnode(&dn, inode, ifolio, ifolio, 0); if (!f2fs_lookup_read_extent_cache_block(inode, index, &dn.data_blkaddr)) { @@ -3478,17 +3794,18 @@ static int __reserve_data_block(struct inode *inode, pgoff_t index, { struct f2fs_sb_info *sbi = F2FS_I_SB(inode); struct dnode_of_data dn; - struct page *ipage; + struct f2fs_lock_context lc; + struct folio *ifolio; int err = 0; - f2fs_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO); + f2fs_map_lock(sbi, &lc, F2FS_GET_BLOCK_PRE_AIO); - ipage = f2fs_get_node_page(sbi, inode->i_ino); - if (IS_ERR(ipage)) { - err = PTR_ERR(ipage); + ifolio = f2fs_get_inode_folio(sbi, inode->i_ino); + if (IS_ERR(ifolio)) { + err = PTR_ERR(ifolio); goto unlock_out; } - set_new_dnode(&dn, inode, ipage, ipage, 0); + set_new_dnode(&dn, inode, ifolio, ifolio, 0); if (!f2fs_lookup_read_extent_cache_block(dn.inode, index, &dn.data_blkaddr)) @@ -3499,7 +3816,7 @@ static int __reserve_data_block(struct inode *inode, pgoff_t index, f2fs_put_dnode(&dn); unlock_out: - f2fs_map_unlock(sbi, F2FS_GET_BLOCK_PRE_AIO); + f2fs_map_unlock(sbi, &lc, F2FS_GET_BLOCK_PRE_AIO); return err; } @@ -3546,8 +3863,10 @@ reserve_block: return 0; } -static int f2fs_write_begin(struct file *file, struct address_space *mapping, - loff_t pos, unsigned len, struct folio **foliop, void **fsdata) +static int f2fs_write_begin(const struct kiocb *iocb, + struct address_space *mapping, + loff_t pos, unsigned len, struct folio **foliop, + void **fsdata) { struct inode *inode = mapping->host; struct f2fs_sb_info *sbi = F2FS_I_SB(inode); @@ -3603,8 +3922,9 @@ repeat: * Do not use FGP_STABLE to avoid deadlock. * Will wait that below with our IO control. */ - folio = __filemap_get_folio(mapping, index, - FGP_LOCK | FGP_WRITE | FGP_CREAT, GFP_NOFS); + folio = f2fs_filemap_get_folio(mapping, index, + FGP_LOCK | FGP_WRITE | FGP_CREAT | FGP_NOFS, + mapping_gfp_mask(mapping)); if (IS_ERR(folio)) { err = PTR_ERR(folio); goto fail; @@ -3636,7 +3956,7 @@ repeat: } } - f2fs_wait_on_page_writeback(&folio->page, DATA, false, true); + f2fs_folio_wait_writeback(folio, DATA, false, true); if (len == folio_size(folio) || folio_test_uptodate(folio)) return 0; @@ -3656,11 +3976,10 @@ repeat: err = -EFSCORRUPTED; goto put_folio; } - err = f2fs_submit_page_read(use_cow ? - F2FS_I(inode)->cow_inode : inode, - folio, blkaddr, 0, true); - if (err) - goto put_folio; + f2fs_submit_page_read(use_cow ? F2FS_I(inode)->cow_inode : + inode, + NULL, /* can't write to fsverity files */ + folio, blkaddr, 0, true); folio_lock(folio); if (unlikely(folio->mapping != mapping)) { @@ -3676,14 +3995,13 @@ repeat: return 0; put_folio: - folio_unlock(folio); - folio_put(folio); + f2fs_folio_put(folio, true); fail: f2fs_write_failed(inode, pos + len); return err; } -static int f2fs_write_end(struct file *file, +static int f2fs_write_end(const struct kiocb *iocb, struct address_space *mapping, loff_t pos, unsigned len, unsigned copied, struct folio *folio, void *fsdata) @@ -3723,7 +4041,7 @@ static int f2fs_write_end(struct file *file, folio_mark_dirty(folio); if (f2fs_is_atomic_file(inode)) - set_page_private_atomic(folio_page(folio, 0)); + folio_set_f2fs_atomic(folio); if (pos + copied > i_size_read(inode) && !f2fs_verity_in_progress(inode)) { @@ -3733,8 +4051,7 @@ static int f2fs_write_end(struct file *file, pos + copied); } unlock_out: - folio_unlock(folio); - folio_put(folio); + f2fs_folio_put(folio, true); f2fs_update_time(F2FS_I_SB(inode), REQ_TIME); return copied; } @@ -3758,7 +4075,12 @@ void f2fs_invalidate_folio(struct folio *folio, size_t offset, size_t length) f2fs_remove_dirty_inode(inode); } } - clear_page_private_all(&folio->page); + + if (offset || length != folio_size(folio)) + return; + + folio_cancel_dirty(folio); + ffs_detach_free(folio); } bool f2fs_release_folio(struct folio *folio, gfp_t wait) @@ -3767,7 +4089,7 @@ bool f2fs_release_folio(struct folio *folio, gfp_t wait) if (folio_test_dirty(folio)) return false; - clear_page_private_all(&folio->page); + ffs_detach_free(folio); return true; } @@ -3891,18 +4213,18 @@ static int f2fs_migrate_blocks(struct inode *inode, block_t start_blk, set_inode_flag(inode, FI_SKIP_WRITES); for (blkofs = 0; blkofs <= blkofs_end; blkofs++) { - struct page *page; + struct folio *folio; unsigned int blkidx = secidx * blk_per_sec + blkofs; - page = f2fs_get_lock_data_page(inode, blkidx, true); - if (IS_ERR(page)) { + folio = f2fs_get_lock_data_folio(inode, blkidx, true); + if (IS_ERR(folio)) { f2fs_up_write(&sbi->pin_sem); - ret = PTR_ERR(page); + ret = PTR_ERR(folio); goto done; } - set_page_dirty(page); - f2fs_put_page(page, 1); + folio_mark_dirty(folio); + f2fs_folio_put(folio, true); } clear_inode_flag(inode, FI_SKIP_WRITES); @@ -3952,6 +4274,7 @@ static int check_swap_activate(struct swap_info_struct *sis, while (cur_lblock < last_lblock && cur_lblock < sis->max) { struct f2fs_map_blocks map; + bool last_extent = false; retry: cond_resched(); @@ -3977,11 +4300,10 @@ retry: pblock = map.m_pblk; nr_pblocks = map.m_len; - if ((pblock - SM_I(sbi)->main_blkaddr) % blks_per_sec || - nr_pblocks % blks_per_sec || - !f2fs_valid_pinned_area(sbi, pblock)) { - bool last_extent = false; - + if (!last_extent && + ((pblock - SM_I(sbi)->main_blkaddr) % blks_per_sec || + nr_pblocks % blks_per_sec || + f2fs_is_sequential_zone_area(sbi, pblock))) { not_aligned++; nr_pblocks = roundup(nr_pblocks, blks_per_sec); @@ -4002,8 +4324,8 @@ retry: goto out; } - if (!last_extent) - goto retry; + /* lookup block mapping info after block migration */ + goto retry; } if (cur_lblock + nr_pblocks >= sis->max) @@ -4101,7 +4423,6 @@ static void f2fs_swap_deactivate(struct file *file) const struct address_space_operations f2fs_dblock_aops = { .read_folio = f2fs_read_data_folio, .readahead = f2fs_readahead, - .writepage = f2fs_write_data_page, .writepages = f2fs_write_data_pages, .write_begin = f2fs_write_begin, .write_end = f2fs_write_end, @@ -4174,19 +4495,32 @@ int __init f2fs_init_bio_entry_cache(void) { bio_entry_slab = f2fs_kmem_cache_create("f2fs_bio_entry_slab", sizeof(struct bio_entry)); - return bio_entry_slab ? 0 : -ENOMEM; + + if (!bio_entry_slab) + return -ENOMEM; + + ffs_entry_slab = f2fs_kmem_cache_create("f2fs_ffs_slab", + sizeof(struct f2fs_folio_state)); + + if (!ffs_entry_slab) { + kmem_cache_destroy(bio_entry_slab); + return -ENOMEM; + } + + return 0; } void f2fs_destroy_bio_entry_cache(void) { kmem_cache_destroy(bio_entry_slab); + kmem_cache_destroy(ffs_entry_slab); } static int f2fs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, unsigned int flags, struct iomap *iomap, struct iomap *srcmap) { - struct f2fs_map_blocks map = {}; + struct f2fs_map_blocks map = { NULL, }; pgoff_t next_pgofs = 0; int err; @@ -4195,7 +4529,17 @@ static int f2fs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, map.m_next_pgofs = &next_pgofs; map.m_seg_type = f2fs_rw_hint_to_seg_type(F2FS_I_SB(inode), inode->i_write_hint); - if (flags & IOMAP_WRITE) + if (flags & IOMAP_WRITE && iomap->private) { + map.m_last_pblk = (unsigned long)iomap->private; + iomap->private = NULL; + } + + /* + * If the blocks being overwritten are already allocated, + * f2fs_map_lock and f2fs_balance_fs are not necessary. + */ + if ((flags & IOMAP_WRITE) && + !__f2fs_overwrite_io(inode, offset, length, true)) map.m_may_create = true; err = f2fs_map_blocks(inode, &map, F2FS_GET_BLOCK_DIO); @@ -4227,6 +4571,9 @@ static int f2fs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, iomap->flags |= IOMAP_F_MERGED; iomap->bdev = map.m_bdev; iomap->addr = F2FS_BLK_TO_BYTES(map.m_pblk); + + if (flags & IOMAP_WRITE && map.m_last_pblk) + iomap->private = (void *)map.m_last_pblk; } else { if (flags & IOMAP_WRITE) return -ENOTBLK; @@ -4246,7 +4593,7 @@ static int f2fs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, if (map.m_flags & F2FS_MAP_NEW) iomap->flags |= IOMAP_F_NEW; - if ((inode->i_state & I_DIRTY_DATASYNC) || + if ((inode_state_read_once(inode) & I_DIRTY_DATASYNC) || offset + length > i_size_read(inode)) iomap->flags |= IOMAP_F_DIRTY; |