^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) .. SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) Inline Data
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) -----------
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) The inline data feature was designed to handle the case that a file's
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) data is so tiny that it readily fits inside the inode, which
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) (theoretically) reduces disk block consumption and reduces seeks. If the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) file is smaller than 60 bytes, then the data are stored inline in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) ``inode.i_block``. If the rest of the file would fit inside the extended
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) attribute space, then it might be found as an extended attribute
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) “system.data” within the inode body (“ibody EA”). This of course
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) constrains the amount of extended attributes one can attach to an inode.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) If the data size increases beyond i\_block + ibody EA, a regular block
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) is allocated and the contents moved to that block.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) Pending a change to compact the extended attribute key used to store
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) inline data, one ought to be able to store 160 bytes of data in a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) 256-byte inode (as of June 2015, when i\_extra\_isize is 28). Prior to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) that, the limit was 156 bytes due to inefficient use of inode space.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) The inline data feature requires the presence of an extended attribute
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) for “system.data”, even if the attribute value is zero length.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) Inline Directories
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) ~~~~~~~~~~~~~~~~~~
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) The first four bytes of i\_block are the inode number of the parent
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) directory. Following that is a 56-byte space for an array of directory
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) entries; see ``struct ext4_dir_entry``. If there is a “system.data”
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) attribute in the inode body, the EA value is an array of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) ``struct ext4_dir_entry`` as well. Note that for inline directories, the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) i\_block and EA space are treated as separate dirent blocks; directory
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) entries cannot span the two.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) Inline directory entries are not checksummed, as the inode checksum
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) should protect all inline data contents.