1E2fsprogs 1.44.0 (March 7, 2018)
2================================
3
4Updates/Fixes since v1.43.9:
5
6UI and Features
7---------------
8
9E2fsprogs now supports the large_dir (INCOMPAT_LARGEDIR) feature.  This
10feature allows larger directories to be created, both with directory
11sizes over 2GB and and a maximum htree depth of 3 instead of the
12current limit of 2.  These features are needed in order to exceed
13the currently limit of approximately 10M entries in a single
14directory for 4KB blocksize (~100k for 1KB).
15
16E2fsprogs now supports the ea_inode (INCOMPAT_EA_INODE) feature, which
17stores large extended attributes in an external inode instead of a
18single data block.  This is backwards compatible with a Lustre file
19system feature, but it has been extended to support deduplicating
20extended attribute values which are used in multiple inodes.
21
22The metadata_csum feature (RO_COMPAT_METADATA_CSUM) is now enabled by
23default by mke2fs.
24
25The resize2fs program will no longer complain about resizing bigalloc
26file systems as being dangerous when doing an on-line resize (since this
27is purely a kernel feature).
28
29Debugfs's ls command will now print the high bits of the inode's mode
30bits.
31
32Fixes
33-----
34
35Fixed some potential buffer overrun bugs in the blkid library and in the
36fsck program.
37
38There were some corner cases which tune2fs wouldn't correctly handle
39when operating on a file system which was not cleanly mounted and thus
40required running the journal because it could be modified.  Tune2fs will
41now run the journal before trying to change the superblock, and if the
42file system needs to be checked after the journal has been replayed, it
43will require it before proceeding.  Also if the file system is busy (for
44example, because it is in use by Lustre), but is not present in the
45/proc/mounts or /etc/mtab, tune2fs will not try to replay the journal,
46since this could do real damage if it the file system is being actively
47modified by the kernel.
48
49The libss library (used by debugfs) now supports the version 7 of the
50libreadline shared library.
51
52The e2freefrag program will use the GETFSMAP ioctl if it is available
53and the file system is mounted, so that the free space statistics will
54be more accurate.
55
56The consistency checks for symlinks (especially for encrypted and inline
57data file sytems) in e2fsck are now much more careful/stringent.  Also
58fixed a bug where creating a symlink using ext2fs_symlink() which was
59exactly 60 bytes long when inline data was enabled would result in a
60missing system.data xattr.
61
62
63Performance, Internal Implementation, Development Support etc.
64--------------------------------------------------------------
65
66Mke2fs now uses io_channel_flush() instead of sync() for the sync_kludge
67feature.
68
69E2fsprogs now uses i_size as the primary way of determining whether a
70symlink is a fast symlink.
71
72Fix various compiler and UBSAN warnings.
73
74Long running test are now skipped with "make check" and only run when
75"make fullcheck".  The test runner will warn when potentially long
76running tests are being run via "make check".
77
78E2fsck is now much faster for bigalloc file systems when scanning
79extents for bigalloc file systems.  (Addresses Google Bug #36886699)
80
81Update Czech, Spanish, French, Malay, and Ukrainian translations.
82
83