diff options
Diffstat (limited to 'fs/iomap/buffered-io.c')
| -rw-r--r-- | fs/iomap/buffered-io.c | 1919 |
1 files changed, 923 insertions, 996 deletions
diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index ae8673ce08b1..e5c1ca440d93 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -1,37 +1,26 @@ // SPDX-License-Identifier: GPL-2.0 /* * Copyright (C) 2010 Red Hat, Inc. - * Copyright (C) 2016-2019 Christoph Hellwig. + * Copyright (C) 2016-2023 Christoph Hellwig. */ -#include <linux/module.h> -#include <linux/compiler.h> -#include <linux/fs.h> #include <linux/iomap.h> -#include <linux/pagemap.h> -#include <linux/uio.h> #include <linux/buffer_head.h> -#include <linux/dax.h> #include <linux/writeback.h> -#include <linux/list_sort.h> #include <linux/swap.h> -#include <linux/bio.h> -#include <linux/sched/signal.h> #include <linux/migrate.h> +#include "internal.h" #include "trace.h" #include "../internal.h" -#define IOEND_BATCH_SIZE 4096 - -typedef int (*iomap_punch_t)(struct inode *inode, loff_t offset, loff_t length); /* * Structure allocated for each folio to track per-block uptodate, dirty state * and I/O completions. */ struct iomap_folio_state { - atomic_t read_bytes_pending; - atomic_t write_bytes_pending; spinlock_t state_lock; + unsigned int read_bytes_pending; + atomic_t write_bytes_pending; /* * Each block has two bits in this bitmap: @@ -41,8 +30,6 @@ struct iomap_folio_state { unsigned long state[]; }; -static struct bio_set iomap_ioend_bioset; - static inline bool ifs_is_fully_uptodate(struct folio *folio, struct iomap_folio_state *ifs) { @@ -51,46 +38,127 @@ static inline bool ifs_is_fully_uptodate(struct folio *folio, return bitmap_full(ifs->state, i_blocks_per_folio(inode, folio)); } -static inline bool ifs_block_is_uptodate(struct iomap_folio_state *ifs, - unsigned int block) +/* + * Find the next uptodate block in the folio. end_blk is inclusive. + * If no uptodate block is found, this will return end_blk + 1. + */ +static unsigned ifs_next_uptodate_block(struct folio *folio, + unsigned start_blk, unsigned end_blk) { - return test_bit(block, ifs->state); + struct iomap_folio_state *ifs = folio->private; + + return find_next_bit(ifs->state, end_blk + 1, start_blk); } -static void ifs_set_range_uptodate(struct folio *folio, +/* + * Find the next non-uptodate block in the folio. end_blk is inclusive. + * If no non-uptodate block is found, this will return end_blk + 1. + */ +static unsigned ifs_next_nonuptodate_block(struct folio *folio, + unsigned start_blk, unsigned end_blk) +{ + struct iomap_folio_state *ifs = folio->private; + + return find_next_zero_bit(ifs->state, end_blk + 1, start_blk); +} + +static bool ifs_set_range_uptodate(struct folio *folio, struct iomap_folio_state *ifs, size_t off, size_t len) { struct inode *inode = folio->mapping->host; unsigned int first_blk = off >> inode->i_blkbits; unsigned int last_blk = (off + len - 1) >> inode->i_blkbits; unsigned int nr_blks = last_blk - first_blk + 1; - unsigned long flags; - spin_lock_irqsave(&ifs->state_lock, flags); bitmap_set(ifs->state, first_blk, nr_blks); - if (ifs_is_fully_uptodate(folio, ifs)) - folio_mark_uptodate(folio); - spin_unlock_irqrestore(&ifs->state_lock, flags); + return ifs_is_fully_uptodate(folio, ifs); } static void iomap_set_range_uptodate(struct folio *folio, size_t off, size_t len) { struct iomap_folio_state *ifs = folio->private; + unsigned long flags; + bool uptodate = true; - if (ifs) - ifs_set_range_uptodate(folio, ifs, off, len); - else + if (folio_test_uptodate(folio)) + return; + + if (ifs) { + spin_lock_irqsave(&ifs->state_lock, flags); + uptodate = ifs_set_range_uptodate(folio, ifs, off, len); + spin_unlock_irqrestore(&ifs->state_lock, flags); + } + + if (uptodate) folio_mark_uptodate(folio); } -static inline bool ifs_block_is_dirty(struct folio *folio, - struct iomap_folio_state *ifs, int block) +/* + * Find the next dirty block in the folio. end_blk is inclusive. + * If no dirty block is found, this will return end_blk + 1. + */ +static unsigned ifs_next_dirty_block(struct folio *folio, + unsigned start_blk, unsigned end_blk) { + struct iomap_folio_state *ifs = folio->private; struct inode *inode = folio->mapping->host; - unsigned int blks_per_folio = i_blocks_per_folio(inode, folio); + unsigned int blks = i_blocks_per_folio(inode, folio); + + return find_next_bit(ifs->state, blks + end_blk + 1, + blks + start_blk) - blks; +} + +/* + * Find the next clean block in the folio. end_blk is inclusive. + * If no clean block is found, this will return end_blk + 1. + */ +static unsigned ifs_next_clean_block(struct folio *folio, + unsigned start_blk, unsigned end_blk) +{ + struct iomap_folio_state *ifs = folio->private; + struct inode *inode = folio->mapping->host; + unsigned int blks = i_blocks_per_folio(inode, folio); + + return find_next_zero_bit(ifs->state, blks + end_blk + 1, + blks + start_blk) - blks; +} + +static unsigned ifs_find_dirty_range(struct folio *folio, + struct iomap_folio_state *ifs, u64 *range_start, u64 range_end) +{ + struct inode *inode = folio->mapping->host; + unsigned start_blk = + offset_in_folio(folio, *range_start) >> inode->i_blkbits; + unsigned end_blk = min_not_zero( + offset_in_folio(folio, range_end) >> inode->i_blkbits, + i_blocks_per_folio(inode, folio)) - 1; + unsigned nblks; + + start_blk = ifs_next_dirty_block(folio, start_blk, end_blk); + if (start_blk > end_blk) + return 0; + if (start_blk == end_blk) + nblks = 1; + else + nblks = ifs_next_clean_block(folio, start_blk + 1, end_blk) - + start_blk; - return test_bit(block + blks_per_folio, ifs->state); + *range_start = folio_pos(folio) + (start_blk << inode->i_blkbits); + return nblks << inode->i_blkbits; +} + +static unsigned iomap_find_dirty_range(struct folio *folio, u64 *range_start, + u64 range_end) +{ + struct iomap_folio_state *ifs = folio->private; + + if (*range_start >= range_end) + return 0; + + if (ifs) + return ifs_find_dirty_range(folio, ifs, range_start, range_end); + return range_end - *range_start; } static void ifs_clear_range_dirty(struct folio *folio, @@ -181,7 +249,7 @@ static void ifs_free(struct folio *folio) if (!ifs) return; - WARN_ON_ONCE(atomic_read(&ifs->read_bytes_pending)); + WARN_ON_ONCE(ifs->read_bytes_pending != 0); WARN_ON_ONCE(atomic_read(&ifs->write_bytes_pending)); WARN_ON_ONCE(ifs_is_fully_uptodate(folio, ifs) != folio_test_uptodate(folio)); @@ -189,6 +257,22 @@ static void ifs_free(struct folio *folio) } /* + * Calculate how many bytes to truncate based off the number of blocks to + * truncate and the end position to start truncating from. + */ +static size_t iomap_bytes_to_truncate(loff_t end_pos, unsigned block_bits, + unsigned blocks_truncated) +{ + unsigned block_size = 1 << block_bits; + unsigned block_offset = end_pos & (block_size - 1); + + if (!block_offset) + return blocks_truncated << block_bits; + + return ((blocks_truncated - 1) << block_bits) + block_offset; +} + +/* * Calculate the range inside the folio that we actually need to read. */ static void iomap_adjust_read_range(struct inode *inode, struct folio *folio, @@ -201,6 +285,7 @@ static void iomap_adjust_read_range(struct inode *inode, struct folio *folio, unsigned block_size = (1 << block_bits); size_t poff = offset_in_folio(folio, *pos); size_t plen = min_t(loff_t, folio_size(folio) - poff, length); + size_t orig_plen = plen; unsigned first = poff >> block_bits; unsigned last = (poff + plen - 1) >> block_bits; @@ -210,24 +295,29 @@ static void iomap_adjust_read_range(struct inode *inode, struct folio *folio, * to avoid reading in already uptodate ranges. */ if (ifs) { - unsigned int i; - - /* move forward for each leading block marked uptodate */ - for (i = first; i <= last; i++) { - if (!ifs_block_is_uptodate(ifs, i)) - break; - *pos += block_size; - poff += block_size; - plen -= block_size; - first++; + unsigned int next, blocks_skipped; + + next = ifs_next_nonuptodate_block(folio, first, last); + blocks_skipped = next - first; + + if (blocks_skipped) { + unsigned long block_offset = *pos & (block_size - 1); + unsigned bytes_skipped = + (blocks_skipped << block_bits) - block_offset; + + *pos += bytes_skipped; + poff += bytes_skipped; + plen -= bytes_skipped; } + first = next; /* truncate len if we find any trailing uptodate block(s) */ - for ( ; i <= last; i++) { - if (ifs_block_is_uptodate(ifs, i)) { - plen -= (last - i + 1) * block_size; - last = i - 1; - break; + if (++next <= last) { + next = ifs_next_uptodate_block(folio, next, last); + if (next <= last) { + plen -= iomap_bytes_to_truncate(*pos + plen, + block_bits, last - next + 1); + last = next - 1; } } } @@ -237,50 +327,28 @@ static void iomap_adjust_read_range(struct inode *inode, struct folio *folio, * handle both halves separately so that we properly zero data in the * page cache for blocks that are entirely outside of i_size. */ - if (orig_pos <= isize && orig_pos + length > isize) { + if (orig_pos <= isize && orig_pos + orig_plen > isize) { unsigned end = offset_in_folio(folio, isize - 1) >> block_bits; if (first <= end && last > end) - plen -= (last - end) * block_size; + plen -= iomap_bytes_to_truncate(*pos + plen, block_bits, + last - end); } *offp = poff; *lenp = plen; } -static void iomap_finish_folio_read(struct folio *folio, size_t offset, - size_t len, int error) -{ - struct iomap_folio_state *ifs = folio->private; - - if (unlikely(error)) { - folio_clear_uptodate(folio); - folio_set_error(folio); - } else { - iomap_set_range_uptodate(folio, offset, len); - } - - if (!ifs || atomic_sub_and_test(len, &ifs->read_bytes_pending)) - folio_unlock(folio); -} - -static void iomap_read_end_io(struct bio *bio) +static inline bool iomap_block_needs_zeroing(const struct iomap_iter *iter, + loff_t pos) { - int error = blk_status_to_errno(bio->bi_status); - struct folio_iter fi; + const struct iomap *srcmap = iomap_iter_srcmap(iter); - bio_for_each_folio_all(fi, bio) - iomap_finish_folio_read(fi.folio, fi.offset, fi.length, error); - bio_put(bio); + return srcmap->type != IOMAP_MAPPED || + (srcmap->flags & IOMAP_F_NEW) || + pos >= i_size_read(iter->inode); } -struct iomap_readpage_ctx { - struct folio *cur_folio; - bool cur_folio_in_bio; - struct bio *bio; - struct readahead_control *rac; -}; - /** * iomap_read_inline_data - copy inline data into the page cache * @iter: iteration structure @@ -295,179 +363,251 @@ static int iomap_read_inline_data(const struct iomap_iter *iter, { const struct iomap *iomap = iomap_iter_srcmap(iter); size_t size = i_size_read(iter->inode) - iomap->offset; - size_t poff = offset_in_page(iomap->offset); size_t offset = offset_in_folio(folio, iomap->offset); - void *addr; + + if (WARN_ON_ONCE(!iomap->inline_data)) + return -EIO; if (folio_test_uptodate(folio)) return 0; - if (WARN_ON_ONCE(size > PAGE_SIZE - poff)) - return -EIO; - if (WARN_ON_ONCE(size > PAGE_SIZE - - offset_in_page(iomap->inline_data))) - return -EIO; if (WARN_ON_ONCE(size > iomap->length)) return -EIO; if (offset > 0) ifs_alloc(iter->inode, folio, iter->flags); - addr = kmap_local_folio(folio, offset); - memcpy(addr, iomap->inline_data, size); - memset(addr + size, 0, PAGE_SIZE - poff - size); - kunmap_local(addr); - iomap_set_range_uptodate(folio, offset, PAGE_SIZE - poff); + folio_fill_tail(folio, offset, iomap->inline_data, size); + iomap_set_range_uptodate(folio, offset, folio_size(folio) - offset); return 0; } -static inline bool iomap_block_needs_zeroing(const struct iomap_iter *iter, - loff_t pos) +void iomap_finish_folio_read(struct folio *folio, size_t off, size_t len, + int error) { - const struct iomap *srcmap = iomap_iter_srcmap(iter); + struct iomap_folio_state *ifs = folio->private; + bool uptodate = !error; + bool finished = true; - return srcmap->type != IOMAP_MAPPED || - (srcmap->flags & IOMAP_F_NEW) || - pos >= i_size_read(iter->inode); + if (ifs) { + unsigned long flags; + + spin_lock_irqsave(&ifs->state_lock, flags); + if (!error) + uptodate = ifs_set_range_uptodate(folio, ifs, off, len); + ifs->read_bytes_pending -= len; + finished = !ifs->read_bytes_pending; + spin_unlock_irqrestore(&ifs->state_lock, flags); + } + + if (finished) + folio_end_read(folio, uptodate); } +EXPORT_SYMBOL_GPL(iomap_finish_folio_read); -static loff_t iomap_readpage_iter(const struct iomap_iter *iter, - struct iomap_readpage_ctx *ctx, loff_t offset) +static void iomap_read_init(struct folio *folio) +{ + struct iomap_folio_state *ifs = folio->private; + + if (ifs) { + size_t len = folio_size(folio); + + /* + * ifs->read_bytes_pending is used to track how many bytes are + * read in asynchronously by the IO helper. We need to track + * this so that we can know when the IO helper has finished + * reading in all the necessary ranges of the folio and can end + * the read. + * + * Increase ->read_bytes_pending by the folio size to start, and + * add a +1 bias. We'll subtract the bias and any uptodate / + * zeroed ranges that did not require IO in iomap_read_end() + * after we're done processing the folio. + * + * We do this because otherwise, we would have to increment + * ifs->read_bytes_pending every time a range in the folio needs + * to be read in, which can get expensive since the spinlock + * needs to be held whenever modifying ifs->read_bytes_pending. + * + * We add the bias to ensure the read has not been ended on the + * folio when iomap_read_end() is called, even if the IO helper + * has already finished reading in the entire folio. + */ + spin_lock_irq(&ifs->state_lock); + WARN_ON_ONCE(ifs->read_bytes_pending != 0); + ifs->read_bytes_pending = len + 1; + spin_unlock_irq(&ifs->state_lock); + } +} + +/* + * This ends IO if no bytes were submitted to an IO helper. + * + * Otherwise, this calibrates ifs->read_bytes_pending to represent only the + * submitted bytes (see comment in iomap_read_init()). If all bytes submitted + * have already been completed by the IO helper, then this will end the read. + * Else the IO helper will end the read after all submitted ranges have been + * read. + */ +static void iomap_read_end(struct folio *folio, size_t bytes_submitted) +{ + struct iomap_folio_state *ifs = folio->private; + + if (ifs) { + bool end_read, uptodate; + + spin_lock_irq(&ifs->state_lock); + if (!ifs->read_bytes_pending) { + WARN_ON_ONCE(bytes_submitted); + spin_unlock_irq(&ifs->state_lock); + folio_unlock(folio); + return; + } + + /* + * Subtract any bytes that were initially accounted to + * read_bytes_pending but skipped for IO. The +1 accounts for + * the bias we added in iomap_read_init(). + */ + ifs->read_bytes_pending -= + (folio_size(folio) + 1 - bytes_submitted); + + /* + * If !ifs->read_bytes_pending, this means all pending reads by + * the IO helper have already completed, which means we need to + * end the folio read here. If ifs->read_bytes_pending != 0, + * the IO helper will end the folio read. + */ + end_read = !ifs->read_bytes_pending; + if (end_read) + uptodate = ifs_is_fully_uptodate(folio, ifs); + spin_unlock_irq(&ifs->state_lock); + if (end_read) + folio_end_read(folio, uptodate); + } else if (!bytes_submitted) { + /* + * If there were no bytes submitted, this means we are + * responsible for unlocking the folio here, since no IO helper + * has taken ownership of it. If there were bytes submitted, + * then the IO helper will end the read via + * iomap_finish_folio_read(). + */ + folio_unlock(folio); + } +} + +static int iomap_read_folio_iter(struct iomap_iter *iter, + struct iomap_read_folio_ctx *ctx, size_t *bytes_submitted) { const struct iomap *iomap = &iter->iomap; - loff_t pos = iter->pos + offset; - loff_t length = iomap_length(iter) - offset; + loff_t pos = iter->pos; + loff_t length = iomap_length(iter); struct folio *folio = ctx->cur_folio; - struct iomap_folio_state *ifs; - loff_t orig_pos = pos; size_t poff, plen; - sector_t sector; + loff_t pos_diff; + int ret; - if (iomap->type == IOMAP_INLINE) - return iomap_read_inline_data(iter, folio); + if (iomap->type == IOMAP_INLINE) { + ret = iomap_read_inline_data(iter, folio); + if (ret) + return ret; + return iomap_iter_advance(iter, length); + } - /* zero post-eof blocks as the page may be mapped */ - ifs = ifs_alloc(iter->inode, folio, iter->flags); - iomap_adjust_read_range(iter->inode, folio, &pos, length, &poff, &plen); - if (plen == 0) - goto done; + ifs_alloc(iter->inode, folio, iter->flags); - if (iomap_block_needs_zeroing(iter, pos)) { - folio_zero_range(folio, poff, plen); - iomap_set_range_uptodate(folio, poff, plen); - goto done; - } + length = min_t(loff_t, length, + folio_size(folio) - offset_in_folio(folio, pos)); + while (length) { + iomap_adjust_read_range(iter->inode, folio, &pos, length, &poff, + &plen); - ctx->cur_folio_in_bio = true; - if (ifs) - atomic_add(plen, &ifs->read_bytes_pending); - - sector = iomap_sector(iomap, pos); - if (!ctx->bio || - bio_end_sector(ctx->bio) != sector || - !bio_add_folio(ctx->bio, folio, plen, poff)) { - gfp_t gfp = mapping_gfp_constraint(folio->mapping, GFP_KERNEL); - gfp_t orig_gfp = gfp; - unsigned int nr_vecs = DIV_ROUND_UP(length, PAGE_SIZE); - - if (ctx->bio) - submit_bio(ctx->bio); - - if (ctx->rac) /* same as readahead_gfp_mask */ - gfp |= __GFP_NORETRY | __GFP_NOWARN; - ctx->bio = bio_alloc(iomap->bdev, bio_max_segs(nr_vecs), - REQ_OP_READ, gfp); - /* - * If the bio_alloc fails, try it again for a single page to - * avoid having to deal with partial page reads. This emulates - * what do_mpage_read_folio does. - */ - if (!ctx->bio) { - ctx->bio = bio_alloc(iomap->bdev, 1, REQ_OP_READ, - orig_gfp); + pos_diff = pos - iter->pos; + if (WARN_ON_ONCE(pos_diff + plen > length)) + return -EIO; + + ret = iomap_iter_advance(iter, pos_diff); + if (ret) + return ret; + + if (plen == 0) + return 0; + + /* zero post-eof blocks as the page may be mapped */ + if (iomap_block_needs_zeroing(iter, pos)) { + folio_zero_range(folio, poff, plen); + iomap_set_range_uptodate(folio, poff, plen); + } else { + if (!*bytes_submitted) + iomap_read_init(folio); + ret = ctx->ops->read_folio_range(iter, ctx, plen); + if (ret) + return ret; + *bytes_submitted += plen; } - if (ctx->rac) - ctx->bio->bi_opf |= REQ_RAHEAD; - ctx->bio->bi_iter.bi_sector = sector; - ctx->bio->bi_end_io = iomap_read_end_io; - bio_add_folio_nofail(ctx->bio, folio, plen, poff); - } -done: - /* - * Move the caller beyond our range so that it keeps making progress. - * For that, we have to include any leading non-uptodate ranges, but - * we can skip trailing ones as they will be handled in the next - * iteration. - */ - return pos - orig_pos + plen; + ret = iomap_iter_advance(iter, plen); + if (ret) + return ret; + length -= pos_diff + plen; + pos = iter->pos; + } + return 0; } -int iomap_read_folio(struct folio *folio, const struct iomap_ops *ops) +void iomap_read_folio(const struct iomap_ops *ops, + struct iomap_read_folio_ctx *ctx) { + struct folio *folio = ctx->cur_folio; struct iomap_iter iter = { .inode = folio->mapping->host, .pos = folio_pos(folio), .len = folio_size(folio), }; - struct iomap_readpage_ctx ctx = { - .cur_folio = folio, - }; + size_t bytes_submitted = 0; int ret; trace_iomap_readpage(iter.inode, 1); while ((ret = iomap_iter(&iter, ops)) > 0) - iter.processed = iomap_readpage_iter(&iter, &ctx, 0); + iter.status = iomap_read_folio_iter(&iter, ctx, + &bytes_submitted); - if (ret < 0) - folio_set_error(folio); - - if (ctx.bio) { - submit_bio(ctx.bio); - WARN_ON_ONCE(!ctx.cur_folio_in_bio); - } else { - WARN_ON_ONCE(ctx.cur_folio_in_bio); - folio_unlock(folio); - } + if (ctx->ops->submit_read) + ctx->ops->submit_read(ctx); - /* - * Just like mpage_readahead and block_read_full_folio, we always - * return 0 and just set the folio error flag on errors. This - * should be cleaned up throughout the stack eventually. - */ - return 0; + iomap_read_end(folio, bytes_submitted); } EXPORT_SYMBOL_GPL(iomap_read_folio); -static loff_t iomap_readahead_iter(const struct iomap_iter *iter, - struct iomap_readpage_ctx *ctx) +static int iomap_readahead_iter(struct iomap_iter *iter, + struct iomap_read_folio_ctx *ctx, size_t *cur_bytes_submitted) { - loff_t length = iomap_length(iter); - loff_t done, ret; + int ret; - for (done = 0; done < length; done += ret) { + while (iomap_length(iter)) { if (ctx->cur_folio && - offset_in_folio(ctx->cur_folio, iter->pos + done) == 0) { - if (!ctx->cur_folio_in_bio) - folio_unlock(ctx->cur_folio); + offset_in_folio(ctx->cur_folio, iter->pos) == 0) { + iomap_read_end(ctx->cur_folio, *cur_bytes_submitted); ctx->cur_folio = NULL; } if (!ctx->cur_folio) { ctx->cur_folio = readahead_folio(ctx->rac); - ctx->cur_folio_in_bio = false; + if (WARN_ON_ONCE(!ctx->cur_folio)) + return -EINVAL; + *cur_bytes_submitted = 0; } - ret = iomap_readpage_iter(iter, ctx, done); - if (ret <= 0) + ret = iomap_read_folio_iter(iter, ctx, cur_bytes_submitted); + if (ret) return ret; } - return done; + return 0; } /** * iomap_readahead - Attempt to read pages from a file. - * @rac: Describes the pages to be read. * @ops: The operations vector for the filesystem. + * @ctx: The ctx used for issuing readahead. * * This function is for filesystems to call to implement their readahead * address_space operation. @@ -479,28 +619,28 @@ static loff_t iomap_readahead_iter(const struct iomap_iter *iter, * function is called with memalloc_nofs set, so allocations will not cause * the filesystem to be reentered. */ -void iomap_readahead(struct readahead_control *rac, const struct iomap_ops *ops) +void iomap_readahead(const struct iomap_ops *ops, + struct iomap_read_folio_ctx *ctx) { + struct readahead_control *rac = ctx->rac; struct iomap_iter iter = { .inode = rac->mapping->host, .pos = readahead_pos(rac), .len = readahead_length(rac), }; - struct iomap_readpage_ctx ctx = { - .rac = rac, - }; + size_t cur_bytes_submitted; trace_iomap_readahead(rac->mapping->host, readahead_count(rac)); while (iomap_iter(&iter, ops) > 0) - iter.processed = iomap_readahead_iter(&iter, &ctx); + iter.status = iomap_readahead_iter(&iter, ctx, + &cur_bytes_submitted); - if (ctx.bio) - submit_bio(ctx.bio); - if (ctx.cur_folio) { - if (!ctx.cur_folio_in_bio) - folio_unlock(ctx.cur_folio); - } + if (ctx->ops->submit_read) + ctx->ops->submit_read(ctx); + + if (ctx->cur_folio) + iomap_read_end(ctx->cur_folio, cur_bytes_submitted); } EXPORT_SYMBOL_GPL(iomap_readahead); @@ -515,7 +655,7 @@ bool iomap_is_partially_uptodate(struct folio *folio, size_t from, size_t count) { struct iomap_folio_state *ifs = folio->private; struct inode *inode = folio->mapping->host; - unsigned first, last, i; + unsigned first, last; if (!ifs) return false; @@ -527,10 +667,7 @@ bool iomap_is_partially_uptodate(struct folio *folio, size_t from, size_t count) first = from >> inode->i_blkbits; last = (from + count - 1) >> inode->i_blkbits; - for (i = first; i <= last; i++) - if (!ifs_block_is_uptodate(ifs, i)) - return false; - return true; + return ifs_next_nonuptodate_block(folio, first, last) > last; } EXPORT_SYMBOL_GPL(iomap_is_partially_uptodate); @@ -549,6 +686,8 @@ struct folio *iomap_get_folio(struct iomap_iter *iter, loff_t pos, size_t len) if (iter->flags & IOMAP_NOWAIT) fgp |= FGP_NOWAIT; + if (iter->flags & IOMAP_DONTCACHE) + fgp |= FGP_DONTCACHE; fgp |= fgf_set_order(len); return __filemap_get_folio(iter->inode->i_mapping, pos >> PAGE_SHIFT, @@ -615,23 +754,12 @@ iomap_write_failed(struct inode *inode, loff_t pos, unsigned len) pos + len - 1); } -static int iomap_read_folio_sync(loff_t block_start, struct folio *folio, - size_t poff, size_t plen, const struct iomap *iomap) -{ - struct bio_vec bvec; - struct bio bio; - - bio_init(&bio, iomap->bdev, &bvec, 1, REQ_OP_READ); - bio.bi_iter.bi_sector = iomap_sector(iomap, block_start); - bio_add_folio_nofail(&bio, folio, plen, poff); - return submit_bio_wait(&bio); -} - -static int __iomap_write_begin(const struct iomap_iter *iter, loff_t pos, - size_t len, struct folio *folio) +static int __iomap_write_begin(const struct iomap_iter *iter, + const struct iomap_write_ops *write_ops, size_t len, + struct folio *folio) { - const struct iomap *srcmap = iomap_iter_srcmap(iter); struct iomap_folio_state *ifs; + loff_t pos = iter->pos; loff_t block_size = i_blocksize(iter->inode); loff_t block_start = round_down(pos, block_size); loff_t block_end = round_up(pos + len, block_size); @@ -640,12 +768,14 @@ static int __iomap_write_begin(const struct iomap_iter *iter, loff_t pos, size_t poff, plen; /* - * If the write completely overlaps the current folio, then + * If the write or zeroing completely overlaps the current folio, then * entire folio will be dirtied so there is no need for * per-block state tracking structures to be attached to this folio. + * For the unshare case, we must read in the ondisk contents because we + * are not changing pagecache contents. */ - if (pos <= folio_pos(folio) && - pos + len >= folio_pos(folio) + folio_size(folio)) + if (!(iter->flags & IOMAP_UNSHARE) && pos <= folio_pos(folio) && + pos + len >= folio_next_pos(folio)) return 0; ifs = ifs_alloc(iter->inode, folio, iter->flags); @@ -654,7 +784,6 @@ static int __iomap_write_begin(const struct iomap_iter *iter, loff_t pos, if (folio_test_uptodate(folio)) return 0; - folio_clear_error(folio); do { iomap_adjust_read_range(iter->inode, folio, &block_start, @@ -662,9 +791,12 @@ static int __iomap_write_begin(const struct iomap_iter *iter, loff_t pos, if (plen == 0) break; - if (!(iter->flags & IOMAP_UNSHARE) && - (from <= poff || from >= poff + plen) && - (to <= poff || to >= poff + plen)) + /* + * If the read range will be entirely overwritten by the write, + * we can skip having to zero/read it in. + */ + if (!(iter->flags & IOMAP_UNSHARE) && from <= poff && + to >= poff + plen) continue; if (iomap_block_needs_zeroing(iter, block_start)) { @@ -677,8 +809,12 @@ static int __iomap_write_begin(const struct iomap_iter *iter, loff_t pos, if (iter->flags & IOMAP_NOWAIT) return -EAGAIN; - status = iomap_read_folio_sync(block_start, folio, - poff, plen, srcmap); + if (write_ops && write_ops->read_folio_range) + status = write_ops->read_folio_range(iter, + folio, block_start, plen); + else + status = iomap_bio_read_folio_range_sync(iter, + folio, block_start, plen); if (status) return status; } @@ -688,30 +824,71 @@ static int __iomap_write_begin(const struct iomap_iter *iter, loff_t pos, return 0; } -static struct folio *__iomap_get_folio(struct iomap_iter *iter, loff_t pos, - size_t len) +static struct folio *__iomap_get_folio(struct iomap_iter *iter, + const struct iomap_write_ops *write_ops, size_t len) { - const struct iomap_folio_ops *folio_ops = iter->iomap.folio_ops; + loff_t pos = iter->pos; - if (folio_ops && folio_ops->get_folio) - return folio_ops->get_folio(iter, pos, len); - else - return iomap_get_folio(iter, pos, len); + if (!mapping_large_folio_support(iter->inode->i_mapping)) + len = min_t(size_t, len, PAGE_SIZE - offset_in_page(pos)); + + if (iter->fbatch) { + struct folio *folio = folio_batch_next(iter->fbatch); + + if (!folio) + return NULL; + + /* + * The folio mapping generally shouldn't have changed based on + * fs locks, but be consistent with filemap lookup and retry + * the iter if it does. + */ + folio_lock(folio); + if (unlikely(folio->mapping != iter->inode->i_mapping)) { + iter->iomap.flags |= IOMAP_F_STALE; + folio_unlock(folio); + return NULL; + } + + folio_get(folio); + return folio; + } + + if (write_ops && write_ops->get_folio) + return write_ops->get_folio(iter, pos, len); + return iomap_get_folio(iter, pos, len); } -static void __iomap_put_folio(struct iomap_iter *iter, loff_t pos, size_t ret, +static void __iomap_put_folio(struct iomap_iter *iter, + const struct iomap_write_ops *write_ops, size_t ret, struct folio *folio) { - const struct iomap_folio_ops *folio_ops = iter->iomap.folio_ops; + loff_t pos = iter->pos; - if (folio_ops && folio_ops->put_folio) { - folio_ops->put_folio(iter->inode, pos, ret, folio); + if (write_ops && write_ops->put_folio) { + write_ops->put_folio(iter->inode, pos, ret, folio); } else { folio_unlock(folio); folio_put(folio); } } +/* trim pos and bytes to within a given folio */ +static loff_t iomap_trim_folio_range(struct iomap_iter *iter, + struct folio *folio, size_t *offset, u64 *bytes) +{ + loff_t pos = iter->pos; + size_t fsize = folio_size(folio); + + WARN_ON_ONCE(pos < folio_pos(folio)); + WARN_ON_ONCE(pos >= folio_pos(folio) + fsize); + + *offset = offset_in_folio(folio, pos); + *bytes = min(*bytes, fsize - *offset); + + return pos; +} + static int iomap_write_begin_inline(const struct iomap_iter *iter, struct folio *folio) { @@ -721,29 +898,42 @@ static int iomap_write_begin_inline(const struct iomap_iter *iter, return iomap_read_inline_data(iter, folio); } -static int iomap_write_begin(struct iomap_iter *iter, loff_t pos, - size_t len, struct folio **foliop) +/* + * Grab and prepare a folio for write based on iter state. Returns the folio, + * offset, and length. Callers can optionally pass a max length *plen, + * otherwise init to zero. + */ +static int iomap_write_begin(struct iomap_iter *iter, + const struct iomap_write_ops *write_ops, struct folio **foliop, + size_t *poffset, u64 *plen) { - const struct iomap_folio_ops *folio_ops = iter->iomap.folio_ops; const struct iomap *srcmap = iomap_iter_srcmap(iter); + loff_t pos; + u64 len = min_t(u64, SIZE_MAX, iomap_length(iter)); struct folio *folio; int status = 0; - BUG_ON(pos + len > iter->iomap.offset + iter->iomap.length); - if (srcmap != &iter->iomap) - BUG_ON(pos + len > srcmap->offset + srcmap->length); + len = min_not_zero(len, *plen); + *foliop = NULL; + *plen = 0; if (fatal_signal_pending(current)) return -EINTR; - if (!mapping_large_folio_support(iter->inode->i_mapping)) - len = min_t(size_t, len, PAGE_SIZE - offset_in_page(pos)); - - folio = __iomap_get_folio(iter, pos, len); + folio = __iomap_get_folio(iter, write_ops, len); if (IS_ERR(folio)) return PTR_ERR(folio); /* + * No folio means we're done with a batch. We still have range to + * process so return and let the caller iterate and refill the batch. + */ + if (!folio) { + WARN_ON_ONCE(!iter->fbatch); + return 0; + } + + /* * Now we have a locked folio, before we do anything with it we need to * check that the iomap we have cached is not stale. The inode extent * mapping can change due to concurrent IO in flight (e.g. @@ -753,8 +943,8 @@ static int iomap_write_begin(struct iomap_iter *iter, loff_t pos, * could do the wrong thing here (zero a page range incorrectly or fail * to zero) and corrupt data. */ - if (folio_ops && folio_ops->iomap_valid) { - bool iomap_valid = folio_ops->iomap_valid(iter->inode, + if (write_ops && write_ops->iomap_valid) { + bool iomap_valid = write_ops->iomap_valid(iter->inode, &iter->iomap); if (!iomap_valid) { iter->iomap.flags |= IOMAP_F_STALE; @@ -763,30 +953,44 @@ static int iomap_write_begin(struct iomap_iter *iter, loff_t pos, } } - if (pos + len > folio_pos(folio) + folio_size(folio)) - len = folio_pos(folio) + folio_size(folio) - pos; + /* + * The folios in a batch may not be contiguous. If we've skipped + * forward, advance the iter to the pos of the current folio. If the + * folio starts beyond the end of the mapping, it may have been trimmed + * since the lookup for whatever reason. Return a NULL folio to + * terminate the op. + */ + if (folio_pos(folio) > iter->pos) { + len = min_t(u64, folio_pos(folio) - iter->pos, + iomap_length(iter)); + status = iomap_iter_advance(iter, len); + len = iomap_length(iter); + if (status || !len) + goto out_unlock; + } + + pos = iomap_trim_folio_range(iter, folio, poffset, &len); if (srcmap->type == IOMAP_INLINE) status = iomap_write_begin_inline(iter, folio); else if (srcmap->flags & IOMAP_F_BUFFER_HEAD) status = __block_write_begin_int(folio, pos, len, NULL, srcmap); else - status = __iomap_write_begin(iter, pos, len, folio); + status = __iomap_write_begin(iter, write_ops, len, folio); if (unlikely(status)) goto out_unlock; *foliop = folio; + *plen = len; return 0; out_unlock: - __iomap_put_folio(iter, pos, 0, folio); - iomap_write_failed(iter->inode, pos, len); - + __iomap_put_folio(iter, write_ops, 0, folio); return status; } -static size_t __iomap_write_end(struct inode *inode, loff_t pos, size_t len, +static bool __iomap_write_end(struct inode *inode, loff_t pos, size_t len, size_t copied, struct folio *folio) { flush_dcache_folio(folio); @@ -803,14 +1007,14 @@ static size_t __iomap_write_end(struct inode *inode, loff_t pos, size_t len, * redo the whole thing. */ if (unlikely(copied < len && !folio_test_uptodate(folio))) - return 0; + return false; iomap_set_range_uptodate(folio, offset_in_folio(folio, pos), len); iomap_set_range_dirty(folio, offset_in_folio(folio, pos), copied); filemap_dirty_folio(inode->i_mapping, folio); - return copied; + return true; } -static size_t iomap_write_end_inline(const struct iomap_iter *iter, +static bool iomap_write_end_inline(const struct iomap_iter *iter, struct folio *folio, loff_t pos, size_t copied) { const struct iomap *iomap = &iter->iomap; @@ -819,75 +1023,71 @@ static size_t iomap_write_end_inline(const struct iomap_iter *iter, WARN_ON_ONCE(!folio_test_uptodate(folio)); BUG_ON(!iomap_inline_data_valid(iomap)); + if (WARN_ON_ONCE(!iomap->inline_data)) + return false; + flush_dcache_folio(folio); addr = kmap_local_folio(folio, pos); memcpy(iomap_inline_data(iomap, pos), addr, copied); kunmap_local(addr); mark_inode_dirty(iter->inode); - return copied; + return true; } -/* Returns the number of bytes copied. May be 0. Cannot be an errno. */ -static size_t iomap_write_end(struct iomap_iter *iter, loff_t pos, size_t len, - size_t copied, struct folio *folio) +/* + * Returns true if all copied bytes have been written to the pagecache, + * otherwise return false. + */ +static bool iomap_write_end(struct iomap_iter *iter, size_t len, size_t copied, + struct folio *folio) { const struct iomap *srcmap = iomap_iter_srcmap(iter); - loff_t old_size = iter->inode->i_size; - size_t ret; - - if (srcmap->type == IOMAP_INLINE) { - ret = iomap_write_end_inline(iter, folio, pos, copied); - } else if (srcmap->flags & IOMAP_F_BUFFER_HEAD) { - ret = block_write_end(NULL, iter->inode->i_mapping, pos, len, - copied, &folio->page, NULL); - } else { - ret = __iomap_write_end(iter->inode, pos, len, copied, folio); - } + loff_t pos = iter->pos; - /* - * Update the in-memory inode size after copying the data into the page - * cache. It's up to the file system to write the updated size to disk, - * preferably after I/O completion so that no stale data is exposed. - */ - if (pos + ret > old_size) { - i_size_write(iter->inode, pos + ret); - iter->iomap.flags |= IOMAP_F_SIZE_CHANGED; + if (srcmap->type == IOMAP_INLINE) + return iomap_write_end_inline(iter, folio, pos, copied); + + if (srcmap->flags & IOMAP_F_BUFFER_HEAD) { + size_t bh_written; + + bh_written = block_write_end(pos, len, copied, folio); + WARN_ON_ONCE(bh_written != copied && bh_written != 0); + return bh_written == copied; } - __iomap_put_folio(iter, pos, ret, folio); - if (old_size < pos) - pagecache_isize_extended(iter->inode, old_size, pos); - if (ret < len) - iomap_write_failed(iter->inode, pos + ret, len - ret); - return ret; + return __iomap_write_end(iter->inode, pos, len, copied, folio); } -static loff_t iomap_write_iter(struct iomap_iter *iter, struct iov_iter *i) +static int iomap_write_iter(struct iomap_iter *iter, struct iov_iter *i, + const struct iomap_write_ops *write_ops) { - loff_t length = iomap_length(iter); - size_t chunk = PAGE_SIZE << MAX_PAGECACHE_ORDER; - loff_t pos = iter->pos; - ssize_t written = 0; - long status = 0; + ssize_t total_written = 0; + int status = 0; struct address_space *mapping = iter->inode->i_mapping; + size_t chunk = mapping_max_folio_size(mapping); unsigned int bdp_flags = (iter->flags & IOMAP_NOWAIT) ? BDP_ASYNC : 0; do { struct folio *folio; + loff_t old_size; size_t offset; /* Offset into folio */ - size_t bytes; /* Bytes to write to folio */ + u64 bytes; /* Bytes to write to folio */ size_t copied; /* Bytes copied from user */ + u64 written; /* Bytes have been written */ + loff_t pos; - offset = pos & (chunk - 1); - bytes = min(chunk - offset, iov_iter_count(i)); + bytes = iov_iter_count(i); +retry: + offset = iter->pos & (chunk - 1); + bytes = min(chunk - offset, bytes); status = balance_dirty_pages_ratelimited_flags(mapping, bdp_flags); if (unlikely(status)) break; - if (bytes > length) - bytes = length; + if (bytes > iomap_length(iter)) + bytes = iomap_length(iter); /* * Bring in the user page that we'll copy from _first_. @@ -904,68 +1104,88 @@ static loff_t iomap_write_iter(struct iomap_iter *iter, struct iov_iter *i) break; } - status = iomap_write_begin(iter, pos, bytes, &folio); - if (unlikely(status)) + status = iomap_write_begin(iter, write_ops, &folio, &offset, + &bytes); + if (unlikely(status)) { + iomap_write_failed(iter->inode, iter->pos, bytes); break; + } if (iter->iomap.flags & IOMAP_F_STALE) break; - offset = offset_in_folio(folio, pos); - if (bytes > folio_size(folio) - offset) - bytes = folio_size(folio) - offset; + pos = iter->pos; if (mapping_writably_mapped(mapping)) flush_dcache_folio(folio); copied = copy_folio_from_iter_atomic(folio, offset, bytes, i); - status = iomap_write_end(iter, pos, bytes, copied, folio); + written = iomap_write_end(iter, bytes, copied, folio) ? + copied : 0; - if (unlikely(copied != status)) - iov_iter_revert(i, copied - status); + /* + * Update the in-memory inode size after copying the data into + * the page cache. It's up to the file system to write the + * updated size to disk, preferably after I/O completion so that + * no stale data is exposed. Only once that's done can we + * unlock and release the folio. + */ + old_size = iter->inode->i_size; + if (pos + written > old_size) { + i_size_write(iter->inode, pos + written); + iter->iomap.flags |= IOMAP_F_SIZE_CHANGED; + } + __iomap_put_folio(iter, write_ops, written, folio); + + if (old_size < pos) + pagecache_isize_extended(iter->inode, old_size, pos); cond_resched(); - if (unlikely(status == 0)) { + if (unlikely(written == 0)) { /* * A short copy made iomap_write_end() reject the * thing entirely. Might be memory poisoning * halfway through, might be a race with munmap, * might be severe memory pressure. */ - if (copied) - bytes = copied; + iomap_write_failed(iter->inode, pos, bytes); + iov_iter_revert(i, copied); + if (chunk > PAGE_SIZE) chunk /= 2; + if (copied) { + bytes = copied; + goto retry; + } } else { - pos += status; - written += status; - length -= status; + total_written += written; + iomap_iter_advance(iter, written); } - } while (iov_iter_count(i) && length); + } while (iov_iter_count(i) && iomap_length(iter)); - if (status == -EAGAIN) { - iov_iter_revert(i, written); - return -EAGAIN; - } - return written ? written : status; + return total_written ? 0 : status; } ssize_t iomap_file_buffered_write(struct kiocb *iocb, struct iov_iter *i, - const struct iomap_ops *ops) + const struct iomap_ops *ops, + const struct iomap_write_ops *write_ops, void *private) { struct iomap_iter iter = { .inode = iocb->ki_filp->f_mapping->host, .pos = iocb->ki_pos, .len = iov_iter_count(i), .flags = IOMAP_WRITE, + .private = private, }; ssize_t ret; if (iocb->ki_flags & IOCB_NOWAIT) iter.flags |= IOMAP_NOWAIT; + if (iocb->ki_flags & IOCB_DONTCACHE) + iter.flags |= IOMAP_DONTCACHE; while ((ret = iomap_iter(&iter, ops)) > 0) - iter.processed = iomap_write_iter(&iter, i); + iter.status = iomap_write_iter(&iter, i, write_ops); if (unlikely(iter.pos == iocb->ki_pos)) return ret; @@ -975,15 +1195,14 @@ iomap_file_buffered_write(struct kiocb *iocb, struct iov_iter *i, } EXPORT_SYMBOL_GPL(iomap_file_buffered_write); -static int iomap_write_delalloc_ifs_punch(struct inode *inode, +static void iomap_write_delalloc_ifs_punch(struct inode *inode, struct folio *folio, loff_t start_byte, loff_t end_byte, - iomap_punch_t punch) + struct iomap *iomap, iomap_punch_t punch) { - unsigned int first_blk, last_blk, i; + unsigned int first_blk, last_blk; loff_t last_byte; u8 blkbits = inode->i_blkbits; struct iomap_folio_state *ifs; - int ret = 0; /* * When we have per-block dirty tracking, there can be @@ -993,61 +1212,46 @@ static int iomap_write_delalloc_ifs_punch(struct inode *inode, */ ifs = folio->private; if (!ifs) - return ret; + return; - last_byte = min_t(loff_t, end_byte - 1, - folio_pos(folio) + folio_size(folio) - 1); + last_byte = min_t(loff_t, end_byte - 1, folio_next_pos(folio) - 1); first_blk = offset_in_folio(folio, start_byte) >> blkbits; last_blk = offset_in_folio(folio, last_byte) >> blkbits; - for (i = first_blk; i <= last_blk; i++) { - if (!ifs_block_is_dirty(folio, ifs, i)) { - ret = punch(inode, folio_pos(folio) + (i << blkbits), - 1 << blkbits); - if (ret) - return ret; - } + while ((first_blk = ifs_next_clean_block(folio, first_blk, last_blk)) + <= last_blk) { + punch(inode, folio_pos(folio) + (first_blk << blkbits), + 1 << blkbits, iomap); + first_blk++; } - - return ret; } - -static int iomap_write_delalloc_punch(struct inode *inode, struct folio *folio, +static void iomap_write_delalloc_punch(struct inode *inode, struct folio *folio, loff_t *punch_start_byte, loff_t start_byte, loff_t end_byte, - iomap_punch_t punch) + struct iomap *iomap, iomap_punch_t punch) { - int ret = 0; - if (!folio_test_dirty(folio)) - return ret; + return; /* if dirty, punch up to offset */ if (start_byte > *punch_start_byte) { - ret = punch(inode, *punch_start_byte, - start_byte - *punch_start_byte); - if (ret) - return ret; + punch(inode, *punch_start_byte, start_byte - *punch_start_byte, + iomap); } /* Punch non-dirty blocks within folio */ - ret = iomap_write_delalloc_ifs_punch(inode, folio, start_byte, - end_byte, punch); - if (ret) - return ret; + iomap_write_delalloc_ifs_punch(inode, folio, start_byte, end_byte, + iomap, punch); /* * Make sure the next punch start is correctly bound to * the end of this data range, not the end of the folio. */ - *punch_start_byte = min_t(loff_t, end_byte, - folio_pos(folio) + folio_size(folio)); - - return ret; + *punch_start_byte = min_t(loff_t, end_byte, folio_next_pos(folio)); } /* * Scan the data range passed to us for dirty page cache folios. If we find a - * dirty folio, punch out the preceeding range and update the offset from which + * dirty folio, punch out the preceding range and update the offset from which * the next punch will start from. * * We can punch out storage reservations under clean pages because they either @@ -1062,13 +1266,12 @@ static int iomap_write_delalloc_punch(struct inode *inode, struct folio *folio, * This function uses [start_byte, end_byte) intervals (i.e. open ended) to * simplify range iterations. */ -static int iomap_write_delalloc_scan(struct inode *inode, +static void iomap_write_delalloc_scan(struct inode *inode, loff_t *punch_start_byte, loff_t start_byte, loff_t end_byte, - iomap_punch_t punch) + struct iomap *iomap, iomap_punch_t punch) { while (start_byte < end_byte) { struct folio *folio; - int ret; /* grab locked page */ folio = filemap_lock_folio(inode->i_mapping, @@ -1079,27 +1282,47 @@ static int iomap_write_delalloc_scan(struct inode *inode, continue; } - ret = iomap_write_delalloc_punch(inode, folio, punch_start_byte, - start_byte, end_byte, punch); - if (ret) { - folio_unlock(folio); - folio_put(folio); - return ret; - } + iomap_write_delalloc_punch(inode, folio, punch_start_byte, + start_byte, end_byte, iomap, punch); /* move offset to start of next folio in range */ - start_byte = folio_next_index(folio) << PAGE_SHIFT; + start_byte = folio_next_pos(folio); folio_unlock(folio); folio_put(folio); } - return 0; } /* + * When a short write occurs, the filesystem might need to use ->iomap_end + * to remove space reservations created in ->iomap_begin. + * + * For filesystems that use delayed allocation, there can be dirty pages over + * the delalloc extent outside the range of a short write but still within the + * delalloc extent allocated for this iomap if the write raced with page + * faults. + * * Punch out all the delalloc blocks in the range given except for those that * have dirty data still pending in the page cache - those are going to be * written and so must still retain the delalloc backing for writeback. * + * The punch() callback *must* only punch delalloc extents in the range passed + * to it. It must skip over all other types of extents in the range and leave + * them completely unchanged. It must do this punch atomically with respect to + * other extent modifications. + * + * The punch() callback may be called with a folio locked to prevent writeback + * extent allocation racing at the edge of the range we are currently punching. + * The locked folio may or may not cover the range being punched, so it is not + * safe for the punch() callback to lock folios itself. + * + * Lock order is: + * + * inode->i_rwsem (shared or exclusive) + * inode->i_mapping->invalidate_lock (exclusive) + * folio_lock() + * ->punch + * internal filesystem allocation lock + * * As we are scanning the page cache for data, we don't need to reimplement the * wheel - mapping_seek_hole_data() does exactly what we need to identify the * start and end of data ranges correctly even for sub-folio block sizes. This @@ -1128,20 +1351,21 @@ static int iomap_write_delalloc_scan(struct inode *inode, * require sprinkling this code with magic "+ 1" and "- 1" arithmetic and expose * the code to subtle off-by-one bugs.... */ -static int iomap_write_delalloc_release(struct inode *inode, - loff_t start_byte, loff_t end_byte, iomap_punch_t punch) +void iomap_write_delalloc_release(struct inode *inode, loff_t start_byte, + loff_t end_byte, unsigned flags, struct iomap *iomap, + iomap_punch_t punch) { loff_t punch_start_byte = start_byte; loff_t scan_end_byte = min(i_size_read(inode), end_byte); - int error = 0; /* - * Lock the mapping to avoid races with page faults re-instantiating - * folios and dirtying them via ->page_mkwrite whilst we walk the - * cache and perform delalloc extent removal. Failing to do this can - * leave dirty pages with no space reservation in the cache. + * The caller must hold invalidate_lock to avoid races with page faults + * re-instantiating folios and dirtying them via ->page_mkwrite whilst + * we walk the cache and perform delalloc extent removal. Failing to do + * this can leave dirty pages with no space reservation in the cache. */ - filemap_invalidate_lock(inode->i_mapping); + lockdep_assert_held_write(&inode->i_mapping->invalidate_lock); + while (start_byte < scan_end_byte) { loff_t data_end; @@ -1150,13 +1374,15 @@ static int iomap_write_delalloc_release(struct inode *inode, /* * If there is no more data to scan, all that is left is to * punch out the remaining range. + * + * Note that mapping_seek_hole_data is only supposed to return + * either an offset or -ENXIO, so WARN on any other error as + * that would be an API change without updating the callers. */ if (start_byte == -ENXIO || start_byte == scan_end_byte) break; - if (start_byte < 0) { - error = start_byte; - goto out_unlock; - } + if (WARN_ON_ONCE(start_byte < 0)) + return; WARN_ON_ONCE(start_byte < punch_start_byte); WARN_ON_ONCE(start_byte > scan_end_byte); @@ -1166,221 +1392,233 @@ static int iomap_write_delalloc_release(struct inode *inode, */ data_end = mapping_seek_hole_data(inode->i_mapping, start_byte, scan_end_byte, SEEK_HOLE); - if (data_end < 0) { - error = data_end; - goto out_unlock; - } - WARN_ON_ONCE(data_end <= start_byte); + if (WARN_ON_ONCE(data_end < 0)) + return; + + /* + * If we race with post-direct I/O invalidation of the page cache, + * there might be no data left at start_byte. + */ + if (data_end == start_byte) + continue; + + WARN_ON_ONCE(data_end < start_byte); WARN_ON_ONCE(data_end > scan_end_byte); - error = iomap_write_delalloc_scan(inode, &punch_start_byte, - start_byte, data_end, punch); - if (error) - goto out_unlock; + iomap_write_delalloc_scan(inode, &punch_start_byte, start_byte, + data_end, iomap, punch); /* The next data search starts at the end of this one. */ start_byte = data_end; } if (punch_start_byte < end_byte) - error = punch(inode, punch_start_byte, - end_byte - punch_start_byte); -out_unlock: - filemap_invalidate_unlock(inode->i_mapping); - return error; -} - -/* - * When a short write occurs, the filesystem may need to remove reserved space - * that was allocated in ->iomap_begin from it's ->iomap_end method. For - * filesystems that use delayed allocation, we need to punch out delalloc - * extents from the range that are not dirty in the page cache. As the write can - * race with page faults, there can be dirty pages over the delalloc extent - * outside the range of a short write but still within the delalloc extent - * allocated for this iomap. - * - * This function uses [start_byte, end_byte) intervals (i.e. open ended) to - * simplify range iterations. - * - * The punch() callback *must* only punch delalloc extents in the range passed - * to it. It must skip over all other types of extents in the range and leave - * them completely unchanged. It must do this punch atomically with respect to - * other extent modifications. - * - * The punch() callback may be called with a folio locked to prevent writeback - * extent allocation racing at the edge of the range we are currently punching. - * The locked folio may or may not cover the range being punched, so it is not - * safe for the punch() callback to lock folios itself. - * - * Lock order is: - * - * inode->i_rwsem (shared or exclusive) - * inode->i_mapping->invalidate_lock (exclusive) - * folio_lock() - * ->punch - * internal filesystem allocation lock - */ -int iomap_file_buffered_write_punch_delalloc(struct inode *inode, - struct iomap *iomap, loff_t pos, loff_t length, - ssize_t written, iomap_punch_t punch) -{ - loff_t start_byte; - loff_t end_byte; - unsigned int blocksize = i_blocksize(inode); - - if (iomap->type != IOMAP_DELALLOC) - return 0; - - /* If we didn't reserve the blocks, we're not allowed to punch them. */ - if (!(iomap->flags & IOMAP_F_NEW)) - return 0; - - /* - * start_byte refers to the first unused block after a short write. If - * nothing was written, round offset down to point at the first block in - * the range. - */ - if (unlikely(!written)) - start_byte = round_down(pos, blocksize); - else - start_byte = round_up(pos + written, blocksize); - end_byte = round_up(pos + length, blocksize); - - /* Nothing to do if we've written the entire delalloc extent */ - if (start_byte >= end_byte) - return 0; - - return iomap_write_delalloc_release(inode, start_byte, end_byte, - punch); + punch(inode, punch_start_byte, end_byte - punch_start_byte, + iomap); } -EXPORT_SYMBOL_GPL(iomap_file_buffered_write_punch_delalloc); +EXPORT_SYMBOL_GPL(iomap_write_delalloc_release); -static loff_t iomap_unshare_iter(struct iomap_iter *iter) +static int iomap_unshare_iter(struct iomap_iter *iter, + const struct iomap_write_ops *write_ops) { struct iomap *iomap = &iter->iomap; - const struct iomap *srcmap = iomap_iter_srcmap(iter); - loff_t pos = iter->pos; - loff_t length = iomap_length(iter); - long status = 0; - loff_t written = 0; + u64 bytes = iomap_length(iter); + int status; - /* don't bother with blocks that are not shared to start with */ - if (!(iomap->flags & IOMAP_F_SHARED)) - return length; - /* don't bother with holes or unwritten extents */ - if (srcmap->type == IOMAP_HOLE || srcmap->type == IOMAP_UNWRITTEN) - return length; + if (!iomap_want_unshare_iter(iter)) + return iomap_iter_advance(iter, bytes); do { - unsigned long offset = offset_in_page(pos); - unsigned long bytes = min_t(loff_t, PAGE_SIZE - offset, length); struct folio *folio; + size_t offset; + bool ret; - status = iomap_write_begin(iter, pos, bytes, &folio); + bytes = min_t(u64, SIZE_MAX, bytes); + status = iomap_write_begin(iter, write_ops, &folio, &offset, + &bytes); if (unlikely(status)) return status; - if (iter->iomap.flags & IOMAP_F_STALE) + if (iomap->flags & IOMAP_F_STALE) break; - status = iomap_write_end(iter, pos, bytes, bytes, folio); - if (WARN_ON_ONCE(status == 0)) + ret = iomap_write_end(iter, bytes, bytes, folio); + __iomap_put_folio(iter, write_ops, bytes, folio); + if (WARN_ON_ONCE(!ret)) return -EIO; cond_resched(); - pos += status; - written += status; - length -= status; - balance_dirty_pages_ratelimited(iter->inode->i_mapping); - } while (length); - return written; + status = iomap_iter_advance(iter, bytes); + if (status) + break; + } while ((bytes = iomap_length(iter)) > 0); + + return status; } int iomap_file_unshare(struct inode *inode, loff_t pos, loff_t len, - const struct iomap_ops *ops) + const struct iomap_ops *ops, + const struct iomap_write_ops *write_ops) { struct iomap_iter iter = { .inode = inode, .pos = pos, - .len = len, .flags = IOMAP_WRITE | IOMAP_UNSHARE, }; + loff_t size = i_size_read(inode); int ret; + if (pos < 0 || pos >= size) + return 0; + + iter.len = min(len, size - pos); while ((ret = iomap_iter(&iter, ops)) > 0) - iter.processed = iomap_unshare_iter(&iter); + iter.status = iomap_unshare_iter(&iter, write_ops); return ret; } EXPORT_SYMBOL_GPL(iomap_file_unshare); -static loff_t iomap_zero_iter(struct iomap_iter *iter, bool *did_zero) +/* + * Flush the remaining range of the iter and mark the current mapping stale. + * This is used when zero range sees an unwritten mapping that may have had + * dirty pagecache over it. + */ +static inline int iomap_zero_iter_flush_and_stale(struct iomap_iter *i) { - const struct iomap *srcmap = iomap_iter_srcmap(iter); - loff_t pos = iter->pos; - loff_t length = iomap_length(iter); - loff_t written = 0; + struct address_space *mapping = i->inode->i_mapping; + loff_t end = i->pos + i->len - 1; + + i->iomap.flags |= IOMAP_F_STALE; + return filemap_write_and_wait_range(mapping, i->pos, end); +} - /* already zeroed? we're done. */ - if (srcmap->type == IOMAP_HOLE || srcmap->type == IOMAP_UNWRITTEN) - return length; +static int iomap_zero_iter(struct iomap_iter *iter, bool *did_zero, + const struct iomap_write_ops *write_ops) +{ + u64 bytes = iomap_length(iter); + int status; do { struct folio *folio; - int status; size_t offset; - size_t bytes = min_t(u64, SIZE_MAX, length); + bool ret; - status = iomap_write_begin(iter, pos, bytes, &folio); + bytes = min_t(u64, SIZE_MAX, bytes); + status = iomap_write_begin(iter, write_ops, &folio, &offset, + &bytes); if (status) return status; if (iter->iomap.flags & IOMAP_F_STALE) break; - offset = offset_in_folio(folio, pos); - if (bytes > folio_size(folio) - offset) - bytes = folio_size(folio) - offset; + /* a NULL folio means we're done with a folio batch */ + if (!folio) { + status = iomap_iter_advance_full(iter); + break; + } + + /* warn about zeroing folios beyond eof that won't write back */ + WARN_ON_ONCE(folio_pos(folio) > iter->inode->i_size); + + trace_iomap_zero_iter(iter->inode, folio_pos(folio) + offset, + bytes); folio_zero_range(folio, offset, bytes); folio_mark_accessed(folio); - bytes = iomap_write_end(iter, pos, bytes, bytes, folio); - if (WARN_ON_ONCE(bytes == 0)) + ret = iomap_write_end(iter, bytes, bytes, folio); + __iomap_put_folio(iter, write_ops, bytes, folio); + if (WARN_ON_ONCE(!ret)) return -EIO; - pos += bytes; - length -= bytes; - written += bytes; - } while (length > 0); + status = iomap_iter_advance(iter, bytes); + if (status) + break; + } while ((bytes = iomap_length(iter)) > 0); if (did_zero) *did_zero = true; - return written; + return status; } +loff_t +iomap_fill_dirty_folios( + struct iomap_iter *iter, + loff_t offset, + loff_t length) +{ + struct address_space *mapping = iter->inode->i_mapping; + pgoff_t start = offset >> PAGE_SHIFT; + pgoff_t end = (offset + length - 1) >> PAGE_SHIFT; + + iter->fbatch = kmalloc(sizeof(struct folio_batch), GFP_KERNEL); + if (!iter->fbatch) + return offset + length; + folio_batch_init(iter->fbatch); + + filemap_get_folios_dirty(mapping, &start, end, iter->fbatch); + return (start << PAGE_SHIFT); +} +EXPORT_SYMBOL_GPL(iomap_fill_dirty_folios); + int iomap_zero_range(struct inode *inode, loff_t pos, loff_t len, bool *did_zero, - const struct iomap_ops *ops) + const struct iomap_ops *ops, + const struct iomap_write_ops *write_ops, void *private) { struct iomap_iter iter = { .inode = inode, .pos = pos, .len = len, .flags = IOMAP_ZERO, + .private = private, }; + struct address_space *mapping = inode->i_mapping; int ret; + bool range_dirty; - while ((ret = iomap_iter(&iter, ops)) > 0) - iter.processed = iomap_zero_iter(&iter, did_zero); + /* + * To avoid an unconditional flush, check pagecache state and only flush + * if dirty and the fs returns a mapping that might convert on + * writeback. + */ + range_dirty = filemap_range_needs_writeback(mapping, iter.pos, + iter.pos + iter.len - 1); + while ((ret = iomap_iter(&iter, ops)) > 0) { + const struct iomap *srcmap = iomap_iter_srcmap(&iter); + + if (WARN_ON_ONCE(iter.fbatch && + srcmap->type != IOMAP_UNWRITTEN)) + return -EIO; + + if (!iter.fbatch && + (srcmap->type == IOMAP_HOLE || + srcmap->type == IOMAP_UNWRITTEN)) { + s64 status; + + if (range_dirty) { + range_dirty = false; + status = iomap_zero_iter_flush_and_stale(&iter); + } else { + status = iomap_iter_advance_full(&iter); + } + iter.status = status; + continue; + } + + iter.status = iomap_zero_iter(&iter, did_zero, write_ops); + } return ret; } EXPORT_SYMBOL_GPL(iomap_zero_range); int iomap_truncate_page(struct inode *inode, loff_t pos, bool *did_zero, - const struct iomap_ops *ops) + const struct iomap_ops *ops, + const struct iomap_write_ops *write_ops, void *private) { unsigned int blocksize = i_blocksize(inode); unsigned int off = pos & (blocksize - 1); @@ -1388,11 +1626,12 @@ iomap_truncate_page(struct inode *inode, loff_t pos, bool *did_zero, /* Block boundary? Nothing to do */ if (!off) return 0; - return iomap_zero_range(inode, pos, blocksize - off, did_zero, ops); + return iomap_zero_range(inode, pos, blocksize - off, did_zero, ops, + write_ops, private); } EXPORT_SYMBOL_GPL(iomap_truncate_page); -static loff_t iomap_folio_mkwrite_iter(struct iomap_iter *iter, +static int iomap_folio_mkwrite_iter(struct iomap_iter *iter, struct folio *folio) { loff_t length = iomap_length(iter); @@ -1403,20 +1642,22 @@ static loff_t iomap_folio_mkwrite_iter(struct iomap_iter *iter, &iter->iomap); if (ret) return ret; - block_commit_write(&folio->page, 0, length); + block_commit_write(folio, 0, length); } else { WARN_ON_ONCE(!folio_test_uptodate(folio)); folio_mark_dirty(folio); } - return length; + return iomap_iter_advance(iter, length); } -vm_fault_t iomap_page_mkwrite(struct vm_fault *vmf, const struct iomap_ops *ops) +vm_fault_t iomap_page_mkwrite(struct vm_fault *vmf, const struct iomap_ops *ops, + void *private) { struct iomap_iter iter = { .inode = file_inode(vmf->vma->vm_file), .flags = IOMAP_WRITE | IOMAP_FAULT, + .private = private, }; struct folio *folio = page_folio(vmf->page); ssize_t ret; @@ -1428,7 +1669,7 @@ vm_fault_t iomap_page_mkwrite(struct vm_fault *vmf, const struct iomap_ops *ops) iter.pos = folio_pos(folio); iter.len = ret; while ((ret = iomap_iter(&iter, ops)) > 0) - iter.processed = iomap_folio_mkwrite_iter(&iter, folio); + iter.status = iomap_folio_mkwrite_iter(&iter, folio); if (ret < 0) goto out_unlock; @@ -1440,397 +1681,173 @@ out_unlock: } EXPORT_SYMBOL_GPL(iomap_page_mkwrite); -static void iomap_finish_folio_write(struct inode *inode, struct folio *folio, - size_t len, int error) +static void iomap_writeback_init(struct inode *inode, struct folio *folio) { struct iomap_folio_state *ifs = folio->private; - if (error) { - folio_set_error(folio); - mapping_set_error(inode->i_mapping, error); - } - WARN_ON_ONCE(i_blocks_per_folio(inode, folio) > 1 && !ifs); - WARN_ON_ONCE(ifs && atomic_read(&ifs->write_bytes_pending) <= 0); - - if (!ifs || atomic_sub_and_test(len, &ifs->write_bytes_pending)) - folio_end_writeback(folio); -} - -/* - * We're now finished for good with this ioend structure. Update the page - * state, release holds on bios, and finally free up memory. Do not use the - * ioend after this. - */ -static u32 -iomap_finish_ioend(struct iomap_ioend *ioend, int error) -{ - struct inode *inode = ioend->io_inode; - struct bio *bio = &ioend->io_inline_bio; - struct bio *last = ioend->io_bio, *next; - u64 start = bio->bi_iter.bi_sector; - loff_t offset = ioend->io_offset; - bool quiet = bio_flagged(bio, BIO_QUIET); - u32 folio_count = 0; - - for (bio = &ioend->io_inline_bio; bio; bio = next) { - struct folio_iter fi; - + if (ifs) { + WARN_ON_ONCE(atomic_read(&ifs->write_bytes_pending) != 0); /* - * For the last bio, bi_private points to the ioend, so we - * need to explicitly end the iteration here. + * Set this to the folio size. After processing the folio for + * writeback in iomap_writeback_folio(), we'll subtract any + * ranges not written back. + * + * We do this because otherwise, we would have to atomically + * increment ifs->write_bytes_pending every time a range in the + * folio needs to be written back. */ - if (bio == last) - next = NULL; - else - next = bio->bi_private; - - /* walk all folios in bio, ending page IO on them */ - bio_for_each_folio_all(fi, bio) { - iomap_finish_folio_write(inode, fi.folio, fi.length, - error); - folio_count++; - } - bio_put(bio); - } - /* The ioend has been freed by bio_put() */ - - if (unlikely(error && !quiet)) { - printk_ratelimited(KERN_ERR -"%s: writeback error on inode %lu, offset %lld, sector %llu", - inode->i_sb->s_id, inode->i_ino, offset, start); - } - return folio_count; -} - -/* - * Ioend completion routine for merged bios. This can only be called from task - * contexts as merged ioends can be of unbound length. Hence we have to break up - * the writeback completions into manageable chunks to avoid long scheduler - * holdoffs. We aim to keep scheduler holdoffs down below 10ms so that we get - * good batch processing throughput without creating adverse scheduler latency - * conditions. - */ -void -iomap_finish_ioends(struct iomap_ioend *ioend, int error) -{ - struct list_head tmp; - u32 completions; - - might_sleep(); - - list_replace_init(&ioend->io_list, &tmp); - completions = iomap_finish_ioend(ioend, error); - - while (!list_empty(&tmp)) { - if (completions > IOEND_BATCH_SIZE * 8) { - cond_resched(); - completions = 0; - } - ioend = list_first_entry(&tmp, struct iomap_ioend, io_list); - list_del_init(&ioend->io_list); - completions += iomap_finish_ioend(ioend, error); - } -} -EXPORT_SYMBOL_GPL(iomap_finish_ioends); - -/* - * We can merge two adjacent ioends if they have the same set of work to do. - */ -static bool -iomap_ioend_can_merge(struct iomap_ioend *ioend, struct iomap_ioend *next) -{ - if (ioend->io_bio->bi_status != next->io_bio->bi_status) - return false; - if ((ioend->io_flags & IOMAP_F_SHARED) ^ - (next->io_flags & IOMAP_F_SHARED)) - return false; - if ((ioend->io_type == IOMAP_UNWRITTEN) ^ - (next->io_type == IOMAP_UNWRITTEN)) - return false; - if (ioend->io_offset + ioend->io_size != next->io_offset) - return false; - /* - * Do not merge physically discontiguous ioends. The filesystem - * completion functions will have to iterate the physical - * discontiguities even if we merge the ioends at a logical level, so - * we don't gain anything by merging physical discontiguities here. - * - * We cannot use bio->bi_iter.bi_sector here as it is modified during - * submission so does not point to the start sector of the bio at - * completion. - */ - if (ioend->io_sector + (ioend->io_size >> 9) != next->io_sector) - return false; - return true; -} - -void -iomap_ioend_try_merge(struct iomap_ioend *ioend, struct list_head *more_ioends) -{ - struct iomap_ioend *next; - - INIT_LIST_HEAD(&ioend->io_list); - - while ((next = list_first_entry_or_null(more_ioends, struct iomap_ioend, - io_list))) { - if (!iomap_ioend_can_merge(ioend, next)) - break; - list_move_tail(&next->io_list, &ioend->io_list); - ioend->io_size += next->io_size; + atomic_set(&ifs->write_bytes_pending, folio_size(folio)); } } -EXPORT_SYMBOL_GPL(iomap_ioend_try_merge); -static int -iomap_ioend_compare(void *priv, const struct list_head *a, - const struct list_head *b) +void iomap_finish_folio_write(struct inode *inode, struct folio *folio, + size_t len) { - struct iomap_ioend *ia = container_of(a, struct iomap_ioend, io_list); - struct iomap_ioend *ib = container_of(b, struct iomap_ioend, io_list); + struct iomap_folio_state *ifs = folio->private; - if (ia->io_offset < ib->io_offset) - return -1; - if (ia->io_offset > ib->io_offset) - return 1; - return 0; -} + WARN_ON_ONCE(i_blocks_per_folio(inode, folio) > 1 && !ifs); + WARN_ON_ONCE(ifs && atomic_read(&ifs->write_bytes_pending) <= 0); -void -iomap_sort_ioends(struct list_head *ioend_list) -{ - list_sort(NULL, ioend_list, iomap_ioend_compare); + if (!ifs || atomic_sub_and_test(len, &ifs->write_bytes_pending)) + folio_end_writeback(folio); } -EXPORT_SYMBOL_GPL(iomap_sort_ioends); +EXPORT_SYMBOL_GPL(iomap_finish_folio_write); -static void iomap_writepage_end_bio(struct bio *bio) +static int iomap_writeback_range(struct iomap_writepage_ctx *wpc, + struct folio *folio, u64 pos, u32 rlen, u64 end_pos, + size_t *bytes_submitted) { - struct iomap_ioend *ioend = bio->bi_private; - - iomap_finish_ioend(ioend, blk_status_to_errno(bio->bi_status)); -} + do { + ssize_t ret; -/* - * Submit the final bio for an ioend. - * - * If @error is non-zero, it means that we have a situation where some part of - * the submission process has failed after we've marked pages for writeback - * and unlocked them. In this situation, we need to fail the bio instead of - * submitting it. This typically only happens on a filesystem shutdown. - */ -static int -iomap_submit_ioend(struct iomap_writepage_ctx *wpc, struct iomap_ioend *ioend, - int error) -{ - ioend->io_bio->bi_private = ioend; - ioend->io_bio->bi_end_io = iomap_writepage_end_bio; + ret = wpc->ops->writeback_range(wpc, folio, pos, rlen, end_pos); + if (WARN_ON_ONCE(ret == 0 || ret > rlen)) + return -EIO; + if (ret < 0) + return ret; + rlen -= ret; + pos += ret; - if (wpc->ops->prepare_ioend) - error = wpc->ops->prepare_ioend(ioend, error); - if (error) { /* - * If we're failing the IO now, just mark the ioend with an - * error and finish it. This will run IO completion immediately - * as there is only one reference to the ioend at this point in - * time. + * Holes are not written back by ->writeback_range, so track + * if we did handle anything that is not a hole here. */ - ioend->io_bio->bi_status = errno_to_blk_status(error); - bio_endio(ioend->io_bio); - return error; - } + if (wpc->iomap.type != IOMAP_HOLE) + *bytes_submitted += ret; + } while (rlen); - submit_bio(ioend->io_bio); return 0; } -static struct iomap_ioend * -iomap_alloc_ioend(struct inode *inode, struct iomap_writepage_ctx *wpc, - loff_t offset, sector_t sector, struct writeback_control *wbc) -{ - struct iomap_ioend *ioend; - struct bio *bio; - - bio = bio_alloc_bioset(wpc->iomap.bdev, BIO_MAX_VECS, - REQ_OP_WRITE | wbc_to_write_flags(wbc), - GFP_NOFS, &iomap_ioend_bioset); - bio->bi_iter.bi_sector = sector; - wbc_init_bio(wbc, bio); - - ioend = container_of(bio, struct iomap_ioend, io_inline_bio); - INIT_LIST_HEAD(&ioend->io_list); - ioend->io_type = wpc->iomap.type; - ioend->io_flags = wpc->iomap.flags; - ioend->io_inode = inode; - ioend->io_size = 0; - ioend->io_folios = 0; - ioend->io_offset = offset; - ioend->io_bio = bio; - ioend->io_sector = sector; - return ioend; -} - /* - * Allocate a new bio, and chain the old bio to the new one. + * Check interaction of the folio with the file end. * - * Note that we have to perform the chaining in this unintuitive order - * so that the bi_private linkage is set up in the right direction for the - * traversal in iomap_finish_ioend(). + * If the folio is entirely beyond i_size, return false. If it straddles + * i_size, adjust end_pos and zero all data beyond i_size. */ -static struct bio * -iomap_chain_bio(struct bio *prev) +static bool iomap_writeback_handle_eof(struct folio *folio, struct inode *inode, + u64 *end_pos) { - struct bio *new; + u64 isize = i_size_read(inode); - new = bio_alloc(prev->bi_bdev, BIO_MAX_VECS, prev->bi_opf, GFP_NOFS); - bio_clone_blkg_association(new, prev); - new->bi_iter.bi_sector = bio_end_sector(prev); - - bio_chain(prev, new); - bio_get(prev); /* for iomap_finish_ioend */ - submit_bio(prev); - return new; -} - -static bool -iomap_can_add_to_ioend(struct iomap_writepage_ctx *wpc, loff_t offset, - sector_t sector) -{ - if ((wpc->iomap.flags & IOMAP_F_SHARED) != - (wpc->ioend->io_flags & IOMAP_F_SHARED)) - return false; - if (wpc->iomap.type != wpc->ioend->io_type) - return false; - if (offset != wpc->ioend->io_offset + wpc->ioend->io_size) - return false; - if (sector != bio_end_sector(wpc->ioend->io_bio)) - return false; - /* - * Limit ioend bio chain lengths to minimise IO completion latency. This - * also prevents long tight loops ending page writeback on all the - * folios in the ioend. - */ - if (wpc->ioend->io_folios >= IOEND_BATCH_SIZE) - return false; - return true; -} + if (*end_pos > isize) { + size_t poff = offset_in_folio(folio, isize); + pgoff_t end_index = isize >> PAGE_SHIFT; -/* - * Test to see if we have an existing ioend structure that we could append to - * first; otherwise finish off the current ioend and start another. - */ -static void -iomap_add_to_ioend(struct inode *inode, loff_t pos, struct folio *folio, - struct iomap_folio_state *ifs, struct iomap_writepage_ctx *wpc, - struct writeback_control *wbc, struct list_head *iolist) -{ - sector_t sector = iomap_sector(&wpc->iomap, pos); - unsigned len = i_blocksize(inode); - size_t poff = offset_in_folio(folio, pos); - - if (!wpc->ioend || !iomap_can_add_to_ioend(wpc, pos, sector)) { - if (wpc->ioend) - list_add(&wpc->ioend->io_list, iolist); - wpc->ioend = iomap_alloc_ioend(inode, wpc, pos, sector, wbc); - } + /* + * If the folio is entirely ouside of i_size, skip it. + * + * This can happen due to a truncate operation that is in + * progress and in that case truncate will finish it off once + * we've dropped the folio lock. + * + * Note that the pgoff_t used for end_index is an unsigned long. + * If the given offset is greater than 16TB on a 32-bit system, + * then if we checked if the folio is fully outside i_size with + * "if (folio->index >= end_index + 1)", "end_index + 1" would + * overflow and evaluate to 0. Hence this folio would be + * redirtied and written out repeatedly, which would result in + * an infinite loop; the user program performing this operation + * would hang. Instead, we can detect this situation by + * checking if the folio is totally beyond i_size or if its + * offset is just equal to the EOF. + */ + if (folio->index > end_index || + (folio->index == end_index && poff == 0)) + return false; - if (!bio_add_folio(wpc->ioend->io_bio, folio, len, poff)) { - wpc->ioend->io_bio = iomap_chain_bio(wpc->ioend->io_bio); - bio_add_folio_nofail(wpc->ioend->io_bio, folio, len, poff); + /* + * The folio straddles i_size. + * + * It must be zeroed out on each and every writepage invocation + * because it may be mmapped: + * + * A file is mapped in multiples of the page size. For a + * file that is not a multiple of the page size, the + * remaining memory is zeroed when mapped, and writes to that + * region are not written out to the file. + * + * Also adjust the end_pos to the end of file and skip writeback + * for all blocks entirely beyond i_size. + */ + folio_zero_segment(folio, poff, folio_size(folio)); + *end_pos = isize; } - if (ifs) - atomic_add(len, &ifs->write_bytes_pending); - wpc->ioend->io_size += len; - wbc_account_cgroup_owner(wbc, &folio->page, len); + return true; } -/* - * We implement an immediate ioend submission policy here to avoid needing to - * chain multiple ioends and hence nest mempool allocations which can violate - * the forward progress guarantees we need to provide. The current ioend we're - * adding blocks to is cached in the writepage context, and if the new block - * doesn't append to the cached ioend, it will create a new ioend and cache that - * instead. - * - * If a new ioend is created and cached, the old ioend is returned and queued - * locally for submission once the entire page is processed or an error has been - * detected. While ioends are submitted immediately after they are completed, - * batching optimisations are provided by higher level block plugging. - * - * At the end of a writeback pass, there will be a cached ioend remaining on the - * writepage context that the caller will need to submit. - */ -static int -iomap_writepage_map(struct iomap_writepage_ctx *wpc, - struct writeback_control *wbc, struct inode *inode, - struct folio *folio, u64 end_pos) +int iomap_writeback_folio(struct iomap_writepage_ctx *wpc, struct folio *folio) { struct iomap_folio_state *ifs = folio->private; - struct iomap_ioend *ioend, *next; - unsigned len = i_blocksize(inode); - unsigned nblocks = i_blocks_per_folio(inode, folio); + struct inode *inode = wpc->inode; u64 pos = folio_pos(folio); - int error = 0, count = 0, i; - LIST_HEAD(submit_list); + u64 end_pos = pos + folio_size(folio); + u64 end_aligned = 0; + size_t bytes_submitted = 0; + int error = 0; + u32 rlen; + + WARN_ON_ONCE(!folio_test_locked(folio)); + WARN_ON_ONCE(folio_test_dirty(folio)); + WARN_ON_ONCE(folio_test_writeback(folio)); + + trace_iomap_writeback_folio(inode, pos, folio_size(folio)); + if (!iomap_writeback_handle_eof(folio, inode, &end_pos)) + return 0; WARN_ON_ONCE(end_pos <= pos); - if (!ifs && nblocks > 1) { - ifs = ifs_alloc(inode, folio, 0); - iomap_set_range_dirty(folio, 0, end_pos - pos); - } + if (i_blocks_per_folio(inode, folio) > 1) { + if (!ifs) { + ifs = ifs_alloc(inode, folio, 0); + iomap_set_range_dirty(folio, 0, end_pos - pos); + } - WARN_ON_ONCE(ifs && atomic_read(&ifs->write_bytes_pending) != 0); + iomap_writeback_init(inode, folio); + } /* - * Walk through the folio to find areas to write back. If we - * run off the end of the current map or find the current map - * invalid, grab a new one. + * Set the writeback bit ASAP, as the I/O completion for the single + * block per folio case happen hit as soon as we're submitting the bio. */ - for (i = 0; i < nblocks && pos < end_pos; i++, pos += len) { - if (ifs && !ifs_block_is_dirty(folio, ifs, i)) - continue; + folio_start_writeback(folio); - error = wpc->ops->map_blocks(wpc, inode, pos); + /* + * Walk through the folio to find dirty areas to write back. + */ + end_aligned = round_up(end_pos, i_blocksize(inode)); + while ((rlen = iomap_find_dirty_range(folio, &pos, end_aligned))) { + error = iomap_writeback_range(wpc, folio, pos, rlen, end_pos, + &bytes_submitted); if (error) break; - trace_iomap_writepage_map(inode, &wpc->iomap); - if (WARN_ON_ONCE(wpc->iomap.type == IOMAP_INLINE)) - continue; - if (wpc->iomap.type == IOMAP_HOLE) - continue; - iomap_add_to_ioend(inode, pos, folio, ifs, wpc, wbc, - &submit_list); - count++; + pos += rlen; } - if (count) - wpc->ioend->io_folios++; - WARN_ON_ONCE(!wpc->ioend && !list_empty(&submit_list)); - WARN_ON_ONCE(!folio_test_locked(folio)); - WARN_ON_ONCE(folio_test_writeback(folio)); - WARN_ON_ONCE(folio_test_dirty(folio)); - - /* - * We cannot cancel the ioend directly here on error. We may have - * already set other pages under writeback and hence we have to run I/O - * completion to mark the error state of the pages under writeback - * appropriately. - */ - if (unlikely(error)) { - /* - * Let the filesystem know what portion of the current page - * failed to map. If the page hasn't been added to ioend, it - * won't be affected by I/O completion and we must unlock it - * now. - */ - if (wpc->ops->discard_folio) - wpc->ops->discard_folio(folio, pos); - if (!count) { - folio_unlock(folio); - goto done; - } - } + if (bytes_submitted) + wpc->nr_folios++; /* * We can have dirty bits set past end of file in page_mkwrite path @@ -1838,155 +1855,65 @@ iomap_writepage_map(struct iomap_writepage_ctx *wpc, * all the dirty bits in the folio here. */ iomap_clear_range_dirty(folio, 0, folio_size(folio)); - folio_start_writeback(folio); - folio_unlock(folio); /* - * Preserve the original error if there was one; catch - * submission errors here and propagate into subsequent ioend - * submissions. + * Usually the writeback bit is cleared by the I/O completion handler. + * But we may end up either not actually writing any blocks, or (when + * there are multiple blocks in a folio) all I/O might have finished + * already at this point. In that case we need to clear the writeback + * bit ourselves right after unlocking the page. */ - list_for_each_entry_safe(ioend, next, &submit_list, io_list) { - int error2; + if (ifs) { + /* + * Subtract any bytes that were initially accounted to + * write_bytes_pending but skipped for writeback. + */ + size_t bytes_not_submitted = folio_size(folio) - + bytes_submitted; - list_del_init(&ioend->io_list); - error2 = iomap_submit_ioend(wpc, ioend, error); - if (error2 && !error) - error = error2; + if (bytes_not_submitted) + iomap_finish_folio_write(inode, folio, + bytes_not_submitted); + } else if (!bytes_submitted) { + folio_end_writeback(folio); } - /* - * We can end up here with no error and nothing to write only if we race - * with a partial page truncate on a sub-page block sized filesystem. - */ - if (!count) - folio_end_writeback(folio); -done: mapping_set_error(inode->i_mapping, error); return error; } +EXPORT_SYMBOL_GPL(iomap_writeback_folio); -/* - * Write out a dirty page. - * - * For delalloc space on the page, we need to allocate space and flush it. - * For unwritten space on the page, we need to start the conversion to - * regular allocated space. - */ -static int iomap_do_writepage(struct folio *folio, - struct writeback_control *wbc, void *data) +int +iomap_writepages(struct iomap_writepage_ctx *wpc) { - struct iomap_writepage_ctx *wpc = data; - struct inode *inode = folio->mapping->host; - u64 end_pos, isize; - - trace_iomap_writepage(inode, folio_pos(folio), folio_size(folio)); + struct address_space *mapping = wpc->inode->i_mapping; + struct folio *folio = NULL; + int error; /* - * Refuse to write the folio out if we're called from reclaim context. - * - * This avoids stack overflows when called from deeply used stacks in - * random callers for direct reclaim or memcg reclaim. We explicitly - * allow reclaim from kswapd as the stack usage there is relatively low. - * - * This should never happen except in the case of a VM regression so - * warn about it. + * Writeback from reclaim context should never happen except in the case + * of a VM regression so warn about it and refuse to write the data. */ - if (WARN_ON_ONCE((current->flags & (PF_MEMALLOC|PF_KSWAPD)) == + if (WARN_ON_ONCE((current->flags & (PF_MEMALLOC | PF_KSWAPD)) == PF_MEMALLOC)) - goto redirty; + return -EIO; + + while ((folio = writeback_iter(mapping, wpc->wbc, folio, &error))) { + error = iomap_writeback_folio(wpc, folio); + folio_unlock(folio); + } /* - * Is this folio beyond the end of the file? + * If @error is non-zero, it means that we have a situation where some + * part of the submission process has failed after we've marked pages + * for writeback. * - * The folio index is less than the end_index, adjust the end_pos - * to the highest offset that this folio should represent. - * ----------------------------------------------------- - * | file mapping | <EOF> | - * ----------------------------------------------------- - * | Page ... | Page N-2 | Page N-1 | Page N | | - * ^--------------------------------^----------|-------- - * | desired writeback range | see else | - * ---------------------------------^------------------| + * We cannot cancel the writeback directly in that case, so always call + * ->writeback_submit to run the I/O completion handler to clear the + * writeback bit and let the file system proess the errors. */ - isize = i_size_read(inode); - end_pos = folio_pos(folio) + folio_size(folio); - if (end_pos > isize) { - /* - * Check whether the page to write out is beyond or straddles - * i_size or not. - * ------------------------------------------------------- - * | file mapping | <EOF> | - * ------------------------------------------------------- - * | Page ... | Page N-2 | Page N-1 | Page N | Beyond | - * ^--------------------------------^-----------|--------- - * | | Straddles | - * ---------------------------------^-----------|--------| - */ - size_t poff = offset_in_folio(folio, isize); - pgoff_t end_index = isize >> PAGE_SHIFT; - - /* - * Skip the page if it's fully outside i_size, e.g. - * due to a truncate operation that's in progress. We've - * cleaned this page and truncate will finish things off for - * us. - * - * Note that the end_index is unsigned long. If the given - * offset is greater than 16TB on a 32-bit system then if we - * checked if the page is fully outside i_size with - * "if (page->index >= end_index + 1)", "end_index + 1" would - * overflow and evaluate to 0. Hence this page would be - * redirtied and written out repeatedly, which would result in - * an infinite loop; the user program performing this operation - * would hang. Instead, we can detect this situation by - * checking if the page is totally beyond i_size or if its - * offset is just equal to the EOF. - */ - if (folio->index > end_index || - (folio->index == end_index && poff == 0)) - goto unlock; - - /* - * The page straddles i_size. It must be zeroed out on each - * and every writepage invocation because it may be mmapped. - * "A file is mapped in multiples of the page size. For a file - * that is not a multiple of the page size, the remaining - * memory is zeroed when mapped, and writes to that region are - * not written out to the file." - */ - folio_zero_segment(folio, poff, folio_size(folio)); - end_pos = isize; - } - - return iomap_writepage_map(wpc, wbc, inode, folio, end_pos); - -redirty: - folio_redirty_for_writepage(wbc, folio); -unlock: - folio_unlock(folio); - return 0; -} - -int -iomap_writepages(struct address_space *mapping, struct writeback_control *wbc, - struct iomap_writepage_ctx *wpc, - const struct iomap_writeback_ops *ops) -{ - int ret; - - wpc->ops = ops; - ret = write_cache_pages(mapping, wbc, iomap_do_writepage, wpc); - if (!wpc->ioend) - return ret; - return iomap_submit_ioend(wpc, wpc->ioend, ret); + if (wpc->wb_ctx) + return wpc->ops->writeback_submit(wpc, error); + return error; } EXPORT_SYMBOL_GPL(iomap_writepages); - -static int __init iomap_init(void) -{ - return bioset_init(&iomap_ioend_bioset, 4 * (PAGE_SIZE / SECTOR_SIZE), - offsetof(struct iomap_ioend, io_inline_bio), - BIOSET_NEED_BVECS); -} -fs_initcall(iomap_init); |