From 3c64a1aba7cfcb04f79e76f859b3d66660275d59 Mon Sep 17 00:00:00 2001 From: Stefan Behrens Date: Mon, 13 May 2013 13:53:35 +0000 Subject: Btrfs: cleanup: don't check the same thing twice btrfs_read_fs_root_no_name() already checks if btrfs_root_refs() is zero and returns ENOENT in this case. There is no need to do it again in six places. Signed-off-by: Stefan Behrens Signed-off-by: Josef Bacik --- fs/btrfs/file.c | 4 ---- 1 file changed, 4 deletions(-) (limited to 'fs/btrfs/file.c') diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index b3e359bc8e6..185af15ad9e 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -308,10 +308,6 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info, ret = PTR_ERR(inode_root); goto cleanup; } - if (btrfs_root_refs(&inode_root->root_item) == 0) { - ret = -ENOENT; - goto cleanup; - } key.objectid = defrag->ino; btrfs_set_key_type(&key, BTRFS_INODE_ITEM_KEY); -- cgit v1.2.3-70-g09d2 From a71754fc68f740b7ed46bb83123c63fbbc130611 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Mon, 17 Jun 2013 17:14:39 -0400 Subject: Btrfs: move btrfs_truncate_page to btrfs_cont_expand instead of btrfs_truncate This has plagued us forever and I'm so over working around it. When we truncate down to a non-page aligned offset we will call btrfs_truncate_page to zero out the end of the page and write it back to disk, this will keep us from exposing stale data if we truncate back up from that point. The problem with this is it requires data space to do this, and people don't really expect to get ENOSPC from truncate() for these sort of things. This also tends to bite the orphan cleanup stuff too which keeps people from mounting. To get around this we can just move this into btrfs_cont_expand() to make sure if we are truncating up from a non-page size aligned i_size we will zero out the rest of this page so that we don't expose stale data. This will give ENOSPC if you try to truncate() up or if you try to write past the end of isize, which is much more reasonable. This fixes xfstests generic/083 failing to mount because of the orphan cleanup failing. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/extent_io.c | 11 ++++++++++- fs/btrfs/file.c | 21 +++++++++++++++------ fs/btrfs/inode.c | 15 ++++++++++----- 3 files changed, 35 insertions(+), 12 deletions(-) (limited to 'fs/btrfs/file.c') diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 03ca3ab9589..a83d7019ede 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2477,11 +2477,12 @@ static void end_bio_extent_readpage(struct bio *bio, int err) struct extent_state *cached = NULL; struct extent_state *state; struct btrfs_io_bio *io_bio = btrfs_io_bio(bio); + struct inode *inode = page->mapping->host; pr_debug("end_bio_extent_readpage: bi_sector=%llu, err=%d, " "mirror=%lu\n", (u64)bio->bi_sector, err, io_bio->mirror_num); - tree = &BTRFS_I(page->mapping->host)->io_tree; + tree = &BTRFS_I(inode)->io_tree; /* We always issue full-page reads, but if some block * in a page fails to read, blk_update_request() will @@ -2555,6 +2556,14 @@ static void end_bio_extent_readpage(struct bio *bio, int err) unlock_extent_cached(tree, start, end, &cached, GFP_ATOMIC); if (uptodate) { + loff_t i_size = i_size_read(inode); + pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT; + unsigned offset; + + /* Zero out the end if this page straddles i_size */ + offset = i_size & (PAGE_CACHE_SIZE-1); + if (page->index == end_index && offset) + zero_user_segment(page, offset, PAGE_CACHE_SIZE); SetPageUptodate(page); } else { ClearPageUptodate(page); diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 185af15ad9e..5ffde560368 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -2173,12 +2173,6 @@ static long btrfs_fallocate(struct file *file, int mode, goto out_reserve_fail; } - /* - * wait for ordered IO before we have any locks. We'll loop again - * below with the locks held. - */ - btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start); - mutex_lock(&inode->i_mutex); ret = inode_newsize_ok(inode, alloc_end); if (ret) @@ -2189,8 +2183,23 @@ static long btrfs_fallocate(struct file *file, int mode, alloc_start); if (ret) goto out; + } else { + /* + * If we are fallocating from the end of the file onward we + * need to zero out the end of the page if i_size lands in the + * middle of a page. + */ + ret = btrfs_truncate_page(inode, inode->i_size, 0, 0); + if (ret) + goto out; } + /* + * wait for ordered IO before we have any locks. We'll loop again + * below with the locks held. + */ + btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start); + locked_end = alloc_end - 1; while (1) { struct btrfs_ordered_extent *ordered; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 509112da611..b7fa96f72ec 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4253,6 +4253,15 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) u64 hole_size; int err = 0; + /* + * If our size started in the middle of a page we need to zero out the + * rest of the page before we expand the i_size, otherwise we could + * expose stale data. + */ + err = btrfs_truncate_page(inode, oldsize, 0, 0); + if (err) + return err; + if (size <= hole_start) return 0; @@ -7565,16 +7574,12 @@ static int btrfs_truncate(struct inode *inode) { struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_block_rsv *rsv; - int ret; + int ret = 0; int err = 0; struct btrfs_trans_handle *trans; u64 mask = root->sectorsize - 1; u64 min_size = btrfs_calc_trunc_metadata_size(root, 1); - ret = btrfs_truncate_page(inode, inode->i_size, 0, 0); - if (ret) - return ret; - btrfs_wait_ordered_range(inode, inode->i_size & (~mask), (u64)-1); btrfs_ordered_update_i_size(inode, inode->i_size, NULL); -- cgit v1.2.3-70-g09d2 From 7ee9e4405f264e9eda808aa5ca4522746a1af9c1 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Fri, 21 Jun 2013 16:37:03 -0400 Subject: Btrfs: check if we can nocow if we don't have data space We always just try and reserve data space when we write, but if we are out of space but have prealloc'ed extents we should still successfully write. This patch will try and see if we can write to prealloc'ed space and if we can go ahead and allow the write to continue. With this patch we now pass xfstests generic/274. Thanks, Signed-off-by: Josef Bacik --- fs/btrfs/ctree.h | 4 ++ fs/btrfs/extent-tree.c | 1 + fs/btrfs/extent_io.c | 3 ++ fs/btrfs/extent_io.h | 1 + fs/btrfs/file.c | 125 ++++++++++++++++++++++++++++++++++++++++++++----- fs/btrfs/inode.c | 40 ++++++++++------ 6 files changed, 148 insertions(+), 26 deletions(-) (limited to 'fs/btrfs/file.c') diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index b528a5509cb..e795bf135e8 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -3552,6 +3552,10 @@ void btrfs_wait_and_free_delalloc_work(struct btrfs_delalloc_work *work); struct extent_map *btrfs_get_extent_fiemap(struct inode *inode, struct page *page, size_t pg_offset, u64 start, u64 len, int create); +noinline int can_nocow_extent(struct btrfs_trans_handle *trans, + struct inode *inode, u64 offset, u64 *len, + u64 *orig_start, u64 *orig_block_len, + u64 *ram_bytes); /* RHEL and EL kernels have a patch that renames PG_checked to FsMisc */ #if defined(ClearPageFsMisc) && !defined(ClearPageChecked) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 5154b91f638..11ba82e43e8 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3666,6 +3666,7 @@ void btrfs_free_reserved_data_space(struct inode *inode, u64 bytes) data_sinfo = root->fs_info->data_sinfo; spin_lock(&data_sinfo->lock); + WARN_ON(data_sinfo->bytes_may_use < bytes); data_sinfo->bytes_may_use -= bytes; trace_btrfs_space_reservation(root->fs_info, "space_info", data_sinfo->flags, bytes, 0); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index a83d7019ede..f8586a957a0 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -543,6 +543,9 @@ int clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 end, btrfs_debug_check_extent_io_range(tree->mapping->host, start, end); + if (bits & EXTENT_DELALLOC) + bits |= EXTENT_NORESERVE; + if (delete) bits |= ~EXTENT_CTLBITS; bits |= EXTENT_FIRST_DELALLOC; diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index 41fb81e7ec5..3b8c4e26e1d 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -19,6 +19,7 @@ #define EXTENT_FIRST_DELALLOC (1 << 12) #define EXTENT_NEED_WAIT (1 << 13) #define EXTENT_DAMAGED (1 << 14) +#define EXTENT_NORESERVE (1 << 15) #define EXTENT_IOBITS (EXTENT_LOCKED | EXTENT_WRITEBACK) #define EXTENT_CTLBITS (EXTENT_DO_ACCOUNTING | EXTENT_FIRST_DELALLOC) diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 5ffde560368..2d70849cec9 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1312,6 +1312,56 @@ fail: } +static noinline int check_can_nocow(struct inode *inode, loff_t pos, + size_t *write_bytes) +{ + struct btrfs_trans_handle *trans; + struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_ordered_extent *ordered; + u64 lockstart, lockend; + u64 num_bytes; + int ret; + + lockstart = round_down(pos, root->sectorsize); + lockend = lockstart + round_up(*write_bytes, root->sectorsize) - 1; + + while (1) { + lock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend); + ordered = btrfs_lookup_ordered_range(inode, lockstart, + lockend - lockstart + 1); + if (!ordered) { + break; + } + unlock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend); + btrfs_start_ordered_extent(inode, ordered, 1); + btrfs_put_ordered_extent(ordered); + } + + trans = btrfs_join_transaction(root); + if (IS_ERR(trans)) { + unlock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend); + return PTR_ERR(trans); + } + + num_bytes = lockend - lockstart + 1; + ret = can_nocow_extent(trans, inode, lockstart, &num_bytes, NULL, NULL, + NULL); + btrfs_end_transaction(trans, root); + if (ret <= 0) { + ret = 0; + } else { + clear_extent_bit(&BTRFS_I(inode)->io_tree, lockstart, lockend, + EXTENT_DIRTY | EXTENT_DELALLOC | + EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 0, 0, + NULL, GFP_NOFS); + *write_bytes = min_t(size_t, *write_bytes, num_bytes); + } + + unlock_extent(&BTRFS_I(inode)->io_tree, lockstart, lockend); + + return ret; +} + static noinline ssize_t __btrfs_buffered_write(struct file *file, struct iov_iter *i, loff_t pos) @@ -1319,10 +1369,12 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, struct inode *inode = file_inode(file); struct btrfs_root *root = BTRFS_I(inode)->root; struct page **pages = NULL; + u64 release_bytes = 0; unsigned long first_index; size_t num_written = 0; int nrptrs; int ret = 0; + bool only_release_metadata = false; bool force_page_uptodate = false; nrptrs = min((iov_iter_count(i) + PAGE_CACHE_SIZE - 1) / @@ -1343,6 +1395,7 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, offset); size_t num_pages = (write_bytes + offset + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + size_t reserve_bytes; size_t dirty_pages; size_t copied; @@ -1357,11 +1410,41 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, break; } - ret = btrfs_delalloc_reserve_space(inode, - num_pages << PAGE_CACHE_SHIFT); + reserve_bytes = num_pages << PAGE_CACHE_SHIFT; + ret = btrfs_check_data_free_space(inode, reserve_bytes); + if (ret == -ENOSPC && + (BTRFS_I(inode)->flags & (BTRFS_INODE_NODATACOW | + BTRFS_INODE_PREALLOC))) { + ret = check_can_nocow(inode, pos, &write_bytes); + if (ret > 0) { + only_release_metadata = true; + /* + * our prealloc extent may be smaller than + * write_bytes, so scale down. + */ + num_pages = (write_bytes + offset + + PAGE_CACHE_SIZE - 1) >> + PAGE_CACHE_SHIFT; + reserve_bytes = num_pages << PAGE_CACHE_SHIFT; + ret = 0; + } else { + ret = -ENOSPC; + } + } + if (ret) break; + ret = btrfs_delalloc_reserve_metadata(inode, reserve_bytes); + if (ret) { + if (!only_release_metadata) + btrfs_free_reserved_data_space(inode, + reserve_bytes); + break; + } + + release_bytes = reserve_bytes; + /* * This is going to setup the pages array with the number of * pages we want, so we don't really need to worry about the @@ -1370,11 +1453,8 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, ret = prepare_pages(root, file, pages, num_pages, pos, first_index, write_bytes, force_page_uptodate); - if (ret) { - btrfs_delalloc_release_space(inode, - num_pages << PAGE_CACHE_SHIFT); + if (ret) break; - } copied = btrfs_copy_from_user(pos, num_pages, write_bytes, pages, i); @@ -1404,30 +1484,46 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, * managed to copy. */ if (num_pages > dirty_pages) { + release_bytes = (num_pages - dirty_pages) << + PAGE_CACHE_SHIFT; if (copied > 0) { spin_lock(&BTRFS_I(inode)->lock); BTRFS_I(inode)->outstanding_extents++; spin_unlock(&BTRFS_I(inode)->lock); } - btrfs_delalloc_release_space(inode, - (num_pages - dirty_pages) << - PAGE_CACHE_SHIFT); + if (only_release_metadata) + btrfs_delalloc_release_metadata(inode, + release_bytes); + else + btrfs_delalloc_release_space(inode, + release_bytes); } + release_bytes = dirty_pages << PAGE_CACHE_SHIFT; if (copied > 0) { ret = btrfs_dirty_pages(root, inode, pages, dirty_pages, pos, copied, NULL); if (ret) { - btrfs_delalloc_release_space(inode, - dirty_pages << PAGE_CACHE_SHIFT); btrfs_drop_pages(pages, num_pages); break; } } + release_bytes = 0; btrfs_drop_pages(pages, num_pages); + if (only_release_metadata && copied > 0) { + u64 lockstart = round_down(pos, root->sectorsize); + u64 lockend = lockstart + + (dirty_pages << PAGE_CACHE_SHIFT) - 1; + + set_extent_bit(&BTRFS_I(inode)->io_tree, lockstart, + lockend, EXTENT_NORESERVE, NULL, + NULL, GFP_NOFS); + only_release_metadata = false; + } + cond_resched(); balance_dirty_pages_ratelimited(inode->i_mapping); @@ -1440,6 +1536,13 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, kfree(pages); + if (release_bytes) { + if (only_release_metadata) + btrfs_delalloc_release_metadata(inode, release_bytes); + else + btrfs_delalloc_release_space(inode, release_bytes); + } + return num_written ? num_written : ret; } diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 8edcdf6910f..4d7c0225839 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -1641,7 +1641,7 @@ static void btrfs_clear_bit_hook(struct inode *inode, btrfs_delalloc_release_metadata(inode, len); if (root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID - && do_list) + && do_list && !(state->state & EXTENT_NORESERVE)) btrfs_free_reserved_data_space(inode, len); __percpu_counter_add(&root->fs_info->delalloc_bytes, -len, @@ -6396,10 +6396,10 @@ out: * returns 1 when the nocow is safe, < 1 on error, 0 if the * block must be cow'd */ -static noinline int can_nocow_odirect(struct btrfs_trans_handle *trans, - struct inode *inode, u64 offset, u64 *len, - u64 *orig_start, u64 *orig_block_len, - u64 *ram_bytes) +noinline int can_nocow_extent(struct btrfs_trans_handle *trans, + struct inode *inode, u64 offset, u64 *len, + u64 *orig_start, u64 *orig_block_len, + u64 *ram_bytes) { struct btrfs_path *path; int ret; @@ -6413,7 +6413,7 @@ static noinline int can_nocow_odirect(struct btrfs_trans_handle *trans, u64 num_bytes; int slot; int found_type; - + bool nocow = (BTRFS_I(inode)->flags & BTRFS_INODE_NODATACOW); path = btrfs_alloc_path(); if (!path) return -ENOMEM; @@ -6453,18 +6453,28 @@ static noinline int can_nocow_odirect(struct btrfs_trans_handle *trans, /* not a regular extent, must cow */ goto out; } + + if (!nocow && found_type == BTRFS_FILE_EXTENT_REG) + goto out; + disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi); + if (disk_bytenr == 0) + goto out; + + if (btrfs_file_extent_compression(leaf, fi) || + btrfs_file_extent_encryption(leaf, fi) || + btrfs_file_extent_other_encoding(leaf, fi)) + goto out; + backref_offset = btrfs_file_extent_offset(leaf, fi); - *orig_start = key.offset - backref_offset; - *orig_block_len = btrfs_file_extent_disk_num_bytes(leaf, fi); - *ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi); + if (orig_start) { + *orig_start = key.offset - backref_offset; + *orig_block_len = btrfs_file_extent_disk_num_bytes(leaf, fi); + *ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi); + } extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi); - if (extent_end < offset + *len) { - /* extent doesn't include our full range, must cow */ - goto out; - } if (btrfs_extent_readonly(root, disk_bytenr)) goto out; @@ -6708,8 +6718,8 @@ static int btrfs_get_blocks_direct(struct inode *inode, sector_t iblock, if (IS_ERR(trans)) goto must_cow; - if (can_nocow_odirect(trans, inode, start, &len, &orig_start, - &orig_block_len, &ram_bytes) == 1) { + if (can_nocow_extent(trans, inode, start, &len, &orig_start, + &orig_block_len, &ram_bytes) == 1) { if (type == BTRFS_ORDERED_PREALLOC) { free_extent_map(em); em = create_pinned_em(inode, start, len, -- cgit v1.2.3-70-g09d2 From 46a1c2c7ae53de2a5676754b54a73c591a3951d2 Mon Sep 17 00:00:00 2001 From: Jie Liu Date: Tue, 25 Jun 2013 12:02:13 +0800 Subject: vfs: export lseek_execute() to modules For those file systems(btrfs/ext4/ocfs2/tmpfs) that support SEEK_DATA/SEEK_HOLE functions, we end up handling the similar matter in lseek_execute() to update the current file offset to the desired offset if it is valid, ceph also does the simliar things at ceph_llseek(). To reduce the duplications, this patch make lseek_execute() public accessible so that we can call it directly from the underlying file systems. Thanks Dave Chinner for this suggestion. [AV: call it vfs_setpos(), don't bring the removed 'inode' argument back] v2->v1: - Add kernel-doc comments for lseek_execute() - Call lseek_execute() in ceph->llseek() Signed-off-by: Jie Liu Cc: Dave Chinner Cc: Al Viro Cc: Andi Kleen Cc: Andrew Morton Cc: Christoph Hellwig Cc: Chris Mason Cc: Josef Bacik Cc: Ben Myers Cc: Ted Tso Cc: Hugh Dickins Cc: Mark Fasheh Cc: Joel Becker Cc: Sage Weil Signed-off-by: Al Viro --- fs/btrfs/file.c | 15 +-------------- fs/ceph/file.c | 11 +---------- fs/ext4/file.c | 24 ++---------------------- fs/ocfs2/file.c | 12 +----------- fs/read_write.c | 19 ++++++++++++++++--- fs/xfs/xfs_file.c | 6 ++---- include/linux/fs.h | 1 + mm/shmem.c | 5 +---- 8 files changed, 25 insertions(+), 68 deletions(-) (limited to 'fs/btrfs/file.c') diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 4205ba752d4..89da56a58b6 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -2425,20 +2425,7 @@ static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int whence) } } - if (offset < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) { - offset = -EINVAL; - goto out; - } - if (offset > inode->i_sb->s_maxbytes) { - offset = -EINVAL; - goto out; - } - - /* Special lock needed here? */ - if (offset != file->f_pos) { - file->f_pos = offset; - file->f_version = 0; - } + offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); out: mutex_unlock(&inode->i_mutex); return offset; diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 656e1690743..16c989d3e23 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -866,16 +866,7 @@ static loff_t ceph_llseek(struct file *file, loff_t offset, int whence) break; } - if (offset < 0 || offset > inode->i_sb->s_maxbytes) { - offset = -EINVAL; - goto out; - } - - /* Special lock needed here? */ - if (offset != file->f_pos) { - file->f_pos = offset; - file->f_version = 0; - } + offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); out: mutex_unlock(&inode->i_mutex); diff --git a/fs/ext4/file.c b/fs/ext4/file.c index b1b4d51b5d8..469361dbe61 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -494,17 +494,7 @@ static loff_t ext4_seek_data(struct file *file, loff_t offset, loff_t maxsize) if (dataoff > isize) return -ENXIO; - if (dataoff < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) - return -EINVAL; - if (dataoff > maxsize) - return -EINVAL; - - if (dataoff != file->f_pos) { - file->f_pos = dataoff; - file->f_version = 0; - } - - return dataoff; + return vfs_setpos(file, dataoff, maxsize); } /* @@ -580,17 +570,7 @@ static loff_t ext4_seek_hole(struct file *file, loff_t offset, loff_t maxsize) if (holeoff > isize) holeoff = isize; - if (holeoff < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) - return -EINVAL; - if (holeoff > maxsize) - return -EINVAL; - - if (holeoff != file->f_pos) { - file->f_pos = holeoff; - file->f_version = 0; - } - - return holeoff; + return vfs_setpos(file, holeoff, maxsize); } /* diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 8a38714f1d9..41000f223ca 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2646,17 +2646,7 @@ static loff_t ocfs2_file_llseek(struct file *file, loff_t offset, int whence) goto out; } - if (offset < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) - ret = -EINVAL; - if (!ret && offset > inode->i_sb->s_maxbytes) - ret = -EINVAL; - if (ret) - goto out; - - if (offset != file->f_pos) { - file->f_pos = offset; - file->f_version = 0; - } + offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); out: mutex_unlock(&inode->i_mutex); diff --git a/fs/read_write.c b/fs/read_write.c index 37d16e82b57..122a3846d9e 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -41,7 +41,19 @@ static inline int unsigned_offsets(struct file *file) return file->f_mode & FMODE_UNSIGNED_OFFSET; } -static loff_t lseek_execute(struct file *file, loff_t offset, loff_t maxsize) +/** + * vfs_setpos - update the file offset for lseek + * @file: file structure in question + * @offset: file offset to seek to + * @maxsize: maximum file size + * + * This is a low-level filesystem helper for updating the file offset to + * the value specified by @offset if the given offset is valid and it is + * not equal to the current file offset. + * + * Return the specified offset on success and -EINVAL on invalid offset. + */ +loff_t vfs_setpos(struct file *file, loff_t offset, loff_t maxsize) { if (offset < 0 && !unsigned_offsets(file)) return -EINVAL; @@ -54,6 +66,7 @@ static loff_t lseek_execute(struct file *file, loff_t offset, loff_t maxsize) } return offset; } +EXPORT_SYMBOL(vfs_setpos); /** * generic_file_llseek_size - generic llseek implementation for regular files @@ -94,7 +107,7 @@ generic_file_llseek_size(struct file *file, loff_t offset, int whence, * like SEEK_SET. */ spin_lock(&file->f_lock); - offset = lseek_execute(file, file->f_pos + offset, maxsize); + offset = vfs_setpos(file, file->f_pos + offset, maxsize); spin_unlock(&file->f_lock); return offset; case SEEK_DATA: @@ -116,7 +129,7 @@ generic_file_llseek_size(struct file *file, loff_t offset, int whence, break; } - return lseek_execute(file, offset, maxsize); + return vfs_setpos(file, offset, maxsize); } EXPORT_SYMBOL(generic_file_llseek_size); diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 0ad2b95fca1..de3dc98f4e8 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -1268,8 +1268,7 @@ xfs_seek_data( } out: - if (offset != file->f_pos) - file->f_pos = offset; + offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); out_unlock: xfs_iunlock_map_shared(ip, lock); @@ -1377,8 +1376,7 @@ out: * situation in particular. */ offset = min_t(loff_t, offset, isize); - if (offset != file->f_pos) - file->f_pos = offset; + offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); out_unlock: xfs_iunlock_map_shared(ip, lock); diff --git a/include/linux/fs.h b/include/linux/fs.h index a137a73fc1f..bccb1924ec9 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2426,6 +2426,7 @@ extern void file_ra_state_init(struct file_ra_state *ra, struct address_space *mapping); extern loff_t noop_llseek(struct file *file, loff_t offset, int whence); extern loff_t no_llseek(struct file *file, loff_t offset, int whence); +extern loff_t vfs_setpos(struct file *file, loff_t offset, loff_t maxsize); extern loff_t generic_file_llseek(struct file *file, loff_t offset, int whence); extern loff_t generic_file_llseek_size(struct file *file, loff_t offset, int whence, loff_t maxsize, loff_t eof); diff --git a/mm/shmem.c b/mm/shmem.c index f887358dabc..118dfa4952f 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1798,10 +1798,7 @@ static loff_t shmem_file_llseek(struct file *file, loff_t offset, int whence) } } - if (offset >= 0 && offset != file->f_pos) { - file->f_pos = offset; - file->f_version = 0; - } + offset = vfs_setpos(file, offset, MAX_LFS_FILESIZE); mutex_unlock(&inode->i_mutex); return offset; } -- cgit v1.2.3-70-g09d2 From ee20a98314e52a6675e94d1a07ca205ffdf09a72 Mon Sep 17 00:00:00 2001 From: Josef Bacik Date: Thu, 11 Jul 2013 10:34:59 -0400 Subject: Btrfs: allow splitting of hole em's when dropping extent cache I noticed while running multi-threaded fsync tests that sometimes fsck would complain about an improper gap. This happens because we fail to add a hole extent to the file, which was happening when we'd split a hole EM because btrfs_drop_extent_cache was just discarding the whole em instead of splitting it. So this patch fixes this by allowing us to split a hole em properly, which means that added holes actually get logged properly and we no longer see this fsck error. Thankfully we're tolerant of these sort of problems so a user would not see any adverse effects of this bug, other than fsck complaining. Thanks, Signed-off-by: Josef Bacik Signed-off-by: Chris Mason --- fs/btrfs/file.c | 62 +++++++++++++++++++++++++++++++++++++-------------------- 1 file changed, 40 insertions(+), 22 deletions(-) (limited to 'fs/btrfs/file.c') diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index a005fe2c072..8e686a427ce 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -596,20 +596,29 @@ void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, if (no_splits) goto next; - if (em->block_start < EXTENT_MAP_LAST_BYTE && - em->start < start) { + if (em->start < start) { split->start = em->start; split->len = start - em->start; - split->orig_start = em->orig_start; - split->block_start = em->block_start; - if (compressed) - split->block_len = em->block_len; - else - split->block_len = split->len; - split->ram_bytes = em->ram_bytes; - split->orig_block_len = max(split->block_len, - em->orig_block_len); + if (em->block_start < EXTENT_MAP_LAST_BYTE) { + split->orig_start = em->orig_start; + split->block_start = em->block_start; + + if (compressed) + split->block_len = em->block_len; + else + split->block_len = split->len; + split->orig_block_len = max(split->block_len, + em->orig_block_len); + split->ram_bytes = em->ram_bytes; + } else { + split->orig_start = split->start; + split->block_len = 0; + split->block_start = em->block_start; + split->orig_block_len = 0; + split->ram_bytes = split->len; + } + split->generation = gen; split->bdev = em->bdev; split->flags = flags; @@ -620,8 +629,7 @@ void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, split = split2; split2 = NULL; } - if (em->block_start < EXTENT_MAP_LAST_BYTE && - testend && em->start + em->len > start + len) { + if (testend && em->start + em->len > start + len) { u64 diff = start + len - em->start; split->start = start + len; @@ -630,18 +638,28 @@ void btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end, split->flags = flags; split->compress_type = em->compress_type; split->generation = gen; - split->orig_block_len = max(em->block_len, + + if (em->block_start < EXTENT_MAP_LAST_BYTE) { + split->orig_block_len = max(em->block_len, em->orig_block_len); - split->ram_bytes = em->ram_bytes; - if (compressed) { - split->block_len = em->block_len; - split->block_start = em->block_start; - split->orig_start = em->orig_start; + split->ram_bytes = em->ram_bytes; + if (compressed) { + split->block_len = em->block_len; + split->block_start = em->block_start; + split->orig_start = em->orig_start; + } else { + split->block_len = split->len; + split->block_start = em->block_start + + diff; + split->orig_start = em->orig_start; + } } else { - split->block_len = split->len; - split->block_start = em->block_start + diff; - split->orig_start = em->orig_start; + split->ram_bytes = split->len; + split->orig_start = split->start; + split->block_len = 0; + split->block_start = em->block_start; + split->orig_block_len = 0; } ret = add_extent_mapping(em_tree, split, modified); -- cgit v1.2.3-70-g09d2