kernel.git/fs/ext4/namei.c, branch linux-6.9.y

ext4: avoid excessive credit estimate in ext4_tmpfile()

2024-05-30T07:44:57Z

[ Upstream commit 35a1f12f0ca857fee1d7a04ef52cbd5f1f84de13 ] A user with minimum journal size (1024 blocks these days) complained about the following error triggered by generic/697 test in ext4_tmpfile(): run fstests generic/697 at 2024-02-28 05:34:46 JBD2: vfstest wants too many credits credits:260 rsv_credits:0 max:256 EXT4-fs error (device loop0) in __ext4_new_inode:1083: error 28 Indeed the credit estimate in ext4_tmpfile() is huge. EXT4_MAXQUOTAS_INIT_BLOCKS() is 219, then 10 credits from ext4_tmpfile() itself and then ext4_xattr_credits_for_new_inode() adds more credits needed for security attributes and ACLs. Now the EXT4_MAXQUOTAS_INIT_BLOCKS() is in fact unnecessary because we've already initialized quotas with dquot_init() shortly before and so EXT4_MAXQUOTAS_TRANS_BLOCKS() is enough (which boils down to 3 credits). Fixes: af51a2ac36d1 ("ext4: ->tmpfile() support") Signed-off-by: Jan Kara Tested-by: Luis Henriques Tested-by: Disha Goel Link: https://lore.kernel.org/r/20240307115320.28949-1-jack@suse.cz Signed-off-by: Theodore Ts'o Signed-off-by: Sasha Levin

ext4: Configure dentry operations at dentry-creation time

2024-02-27T21:55:34Z

This was already the case for case-insensitive before commit bb9cd9106b22 ("fscrypt: Have filesystems handle their d_ops"), but it was changed to set at lookup-time to facilitate the integration with fscrypt. But it's a problem because dentries that don't get created through ->lookup() won't have any visibility of the operations. Since fscrypt now also supports configuring dentry operations at creation-time, do it for any encrypted and/or casefold volume, simplifying the implementation across these features. Acked-by: Theodore Ts'o Reviewed-by: Eric Biggers Link: https://lore.kernel.org/r/20240221171412.10710-8-krisman@suse.de Signed-off-by: Gabriel Krisman Bertazi

Merge tag 'pull-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs

2024-01-12T04:23:50Z

Pull misc filesystem updates from Al Viro: "Misc cleanups (the part that hadn't been picked by individual fs trees)" * tag 'pull-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: apparmorfs: don't duplicate kfree_link() orangefs: saner arguments passing in readdir guts ocfs2_find_match(): there's no such thing as NULL or negative ->d_parent reiserfs_add_entry(): get rid of pointless namelen checks __ocfs2_add_entry(), ocfs2_prepare_dir_for_insert(): namelen checks ext4_add_entry(): ->d_name.len is never 0 befs: d_obtain_alias(ERR_PTR(...)) will do the right thing affs: d_obtain_alias(ERR_PTR(...)) will do the right thing /proc/sys: use d_splice_alias() calling conventions to simplify failure exits hostfs: use d_splice_alias() calling conventions to simplify failure exits udf_fiiter_add_entry(): check for zero ->d_name.len is bogus... udf: d_obtain_alias(ERR_PTR(...)) will do the right thing... udf: d_splice_alias() will do the right thing on ERR_PTR() inode nfsd: kill stale comment about simple_fill_super() requirements bfs_add_entry(): get rid of pointless ->d_name.len checks nilfs2: d_obtain_alias(ERR_PTR(...)) will do the right thing... zonefs: d_splice_alias() will do the right thing on ERR_PTR() inode

ext4_add_entry(): ->d_name.len is never 0

2023-12-21T17:53:11Z

That bogosity goes back to the initial merge of ext3. Once upon a time ext2 used to have a similar check; that got taken out during the switch to page cache (June 2001). ext3 got merged into mainline 5 months later, still using buffer cache for directories; removal of the pointless check in ext2 should've been done as a separate patch, but it hadn't been, so that thing got missed... Signed-off-by: Al Viro

ext4: don't access the source subdirectory content on same-directory rename

2023-11-25T07:53:20Z

We can't really afford locking the source on same-directory rename; currently vfs_rename() tries to do that, but it will have to be changed. The logics in ext4 is lazy and goes looking for ".." in source even in same-directory case. It's not hard to get rid of that, leaving that behaviour only for cross-directory case; that VFS can get locks safely (and will keep doing that after the coming changes). Reviewed-by: Jan Kara Signed-off-by: Al Viro

Merge tag 'ext4_for_linus-6.7-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

2023-11-02T17:45:14Z

Pull ext4 updates from Ted Ts'o: "Cleanup ext4's multi-block allocator, including adding some unit tests, as well as cleaning how we update the backup superblock after online resizes or updating the label or uuid. Optimize handling of released data blocks in ext4's commit machinery to avoid a potential lock contention on s_md_lock spinlock. Fix a number of ext4 bugs: - fix race between writepages and remount - fix racy may inline data check in dio write - add missed brelse in an error path in update_backups - fix umask handling when ACL support is disabled - fix lost EIO error when a journal commit races with a fsync of the blockdev - fix potential improper i_size when there is a crash right after an O_SYNC direct write. - check extent node for validity before potentially using what might be an invalid pointer - fix potential stale data exposure when writing to an unwritten extent and the file system is nearly out of space - fix potential accounting error around block reservations when writing partial delayed allocation writes to a bigalloc cluster - avoid memory allocation failure when tracking partial delayed allocation writes to a bigalloc cluster - fix various debugging print messages" * tag 'ext4_for_linus-6.7-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (41 commits) ext4: properly sync file size update after O_SYNC direct IO ext4: fix racy may inline data check in dio write ext4: run mballoc test with different layouts setting ext4: add first unit test for ext4_mb_new_blocks_simple in mballoc ext4: add some kunit stub for mballoc kunit test ext4: call ext4_mb_mark_context in ext4_group_add_blocks() ext4: Separate block bitmap and buddy bitmap freeing in ext4_group_add_blocks() ext4: call ext4_mb_mark_context in ext4_mb_clear_bb ext4: Separate block bitmap and buddy bitmap freeing in ext4_mb_clear_bb() ext4: call ext4_mb_mark_context in ext4_mb_mark_diskspace_used ext4: extend ext4_mb_mark_context to support allocation under journal ext4: call ext4_mb_mark_context in ext4_free_blocks_simple ext4: factor out codes to update block bitmap and group descriptor on disk from ext4_mb_mark_bb ext4: make state in ext4_mb_mark_bb to be bool jbd2: fix potential data lost in recovering journal raced with synchronizing fs bdev ext4: apply umask if ACL support is disabled ext4: mark buffer new if it is unwritten to avoid stale data exposure ext4: move 'ix' sanity check to corrent position jbd2: fix printk format type for 'io_block' in do_one_pass() jbd2: print io_block if check data block checksum failed when do recovery ...

ext4: convert to new timestamp accessors

2023-10-18T11:26:21Z

Convert to using the new inode timestamp accessor functions. Signed-off-by: Jeff Layton Link: https://lore.kernel.org/r/20231004185347.80880-33-jlayton@kernel.org Signed-off-by: Christian Brauner

ext4: fix incorrect offset

2023-10-06T02:32:14Z

The last argument of ext4_check_dir_entry is dentry offset int the file. Luckily this error only results in the wrong offset being printed in the eventual error message. Signed-off-by: Wang Jianjian Reviewed-by: Jan Kara Link: https://lore.kernel.org/r/tencent_F992989953734FD5DE3F88ECB2191A856206@qq.com Signed-off-by: Theodore Ts'o

Merge tag 'ext4_for_linus-6.6-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

2023-09-17T17:33:53Z

Pull ext4 fixes from Ted Ts'o: "Regression and bug fixes for ext4" * tag 'ext4_for_linus-6.6-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: ext4: fix rec_len verify error ext4: do not let fstrim block system suspend ext4: move setting of trimmed bit into ext4_try_to_trim_range() jbd2: Fix memory leak in journal_init_common() jbd2: Remove page size assumptions buffer: Make bh_offset() work for compound pages

ext4: fix rec_len verify error

2023-09-14T16:07:07Z

With the configuration PAGE_SIZE 64k and filesystem blocksize 64k, a problem occurred when more than 13 million files were directly created under a directory: EXT4-fs error (device xx): ext4_dx_csum_set:492: inode #xxxx: comm xxxxx: dir seems corrupt? Run e2fsck -D. EXT4-fs error (device xx): ext4_dx_csum_verify:463: inode #xxxx: comm xxxxx: dir seems corrupt? Run e2fsck -D. EXT4-fs error (device xx): dx_probe:856: inode #xxxx: block 8188: comm xxxxx: Directory index failed checksum When enough files are created, the fake_dirent->reclen will be 0xffff. it doesn't equal to the blocksize 65536, i.e. 0x10000. But it is not the same condition when blocksize equals to 4k. when enough files are created, the fake_dirent->reclen will be 0x1000. it equals to the blocksize 4k, i.e. 0x1000. The problem seems to be related to the limitation of the 16-bit field when the blocksize is set to 64k. To address this, helpers like ext4_rec_len_{from,to}_disk has already been introduced to complete the conversion between the encoded and the plain form of rec_len. So fix this one by using the helper, and all the other in this file too. Cc: stable@kernel.org Fixes: dbe89444042a ("ext4: Calculate and verify checksums for htree nodes") Suggested-by: Andreas Dilger Suggested-by: Darrick J. Wong Signed-off-by: Shida Zhang Reviewed-by: Andreas Dilger Reviewed-by: Darrick J. Wong Link: https://lore.kernel.org/r/20230803060938.1929759-1-zhangshida@kylinos.cn Signed-off-by: Theodore Ts'o