NAME

fs, inode − format of a 4.2 (ufs) file system volume

SYNOPSIS

#include <sys/types.h>
#include <ufs/fs.h>
#include <ufs/inode.h>

DESCRIPTION

Standard 4.2 (ufs) file system storage volumes have a common format for certain vital information.  Every such volume is divided into a certain number of blocks.  The block size is a parameter of the file system.  Sectors 0 to 15 contain primary and secondary bootstrapping programs. 

The actual file system begins at sector 16 with the super-block. The layout of the super block is defined by the include file ufs/fs.h

Each disk drive contains some number of file systems.  A file system consists of a number of cylinder groups.  Each cylinder group contains inodes and data. 

A file system is described by its super-block, which in turn describes the cylinder groups.  The super-block is critical data and is replicated in each cylinder group to protect against catastrophic loss.  This is done at file system creation time and the critical super-block data does not change, so the copies need not be referenced further unless disaster strikes. 

Addresses stored in inodes are capable of addressing fragments of “blocks.” File system blocks of at most size MAXBSIZE can be optionally broken into 2, 4, or 8 pieces, each of which is addressable; these pieces may be DEV_BSIZE, or some multiple of a DEV_BSIZE unit. 

Large files consist of exclusively large data blocks.  To avoid undue wasted disk space, the last data block of a small file is allocated as only as many fragments of a large block as are necessary.  The file system format retains only a single pointer to such a fragment, which is a piece of a single large block that has been divided.  The size of such a fragment is determinable from information in the inode, using the ‘blksize(fs, ip, lbn)’ macro. 

The file system records space availability at the fragment level; to determine block availability, aligned fragments are examined. 

The root inode is the root of the file system.  Inode 0 cannot be used for normal purposes and historically bad blocks were linked to inode 1, thus the root inode is 2 (inode 1 is no longer used for this purpose, however numerous dump tapes make this assumption, so we are stuck with it).  The lost+found directory is given the next available inode when it is initially created by mkfs(8). 

fs_minfree gives the minimum acceptable percentage of file system blocks which may be free. If the freelist drops below this level only the super-user may continue to allocate blocks. This may be set to 0 if no reserve of free blocks is deemed necessary, however severe performance degradations will be observed if the file system is run at greater than 90% full; thus the default value of fs_minfree is 10%. 

Empirically the best trade-off between block fragmentation and overall disk utilization at a loading of 90% comes with a fragmentation of 4, thus the default fragment size is a fourth of the block size. 

Cylinder group related limits: Each cylinder keeps track of the availability of blocks at different rotational positions, so that sequential blocks can be laid out with minimum rotational latency. NRPOS is the number of rotational positions which are distinguished.  With NRPOS 8 the resolution of the summary information is 2ms for a typical 3600 rpm drive. 

fs_rotdelay gives the minimum number of milliseconds to initiate another disk transfer on the same cylinder.  It is used in determining the rotationally optimal layout for disk blocks within a file; the default value for fs_rotdelay is 2ms. 

Each file system has a statically allocated number of inodes.  An inode is allocated for each NBPI bytes of disk space.  The inode allocation strategy is extremely conservative. 

MAXIPG bounds the number of inodes per cylinder group, and is needed only to keep the structure simpler by having the only a single variable size element (the free bit map). 

Note: MAXIPG must be a multiple of INOPB(fs). 

MINBSIZE is the smallest allowable block size.  With a MINBSIZE of 4096 it is possible to create files of size 2^32 with only two levels of indirection.  MINBSIZE must be big enough to hold a cylinder group block, thus changes to (struct cg) must keep its size within INBSIZE.  MAXCPG is limited only to dimension an array in (struct cg); it can be made larger as long as that structure’s size remains within the bounds dictated by MINBSIZE.  Note: super blocks are never more than size SBSIZE. 

The path name on which the file system is mounted is maintained in fs_fsmnt.  MAXMNTLEN defines the amount of space allocated in the super block for this name.  The limit on the amount of summary information per file system is defined by MAXCSBUFS.  It is currently parameterized for a maximum of two million cylinders. 

Per cylinder group information is summarized in blocks allocated from the first cylinder group’s data blocks.  These blocks are read in from fs_csaddr (size fs_cssize) in addition to the super block. 

Note: sizeof (struct csum) must be a power of two in order for the fs_cs macro to work. 

Super block for a file system: MAXBPC bounds the size of the rotational layout tables and is limited by the fact that the super block is of size SBSIZE.  The size of these tables is inversely proportional to the block size of the file system. The size of the tables is increased when sector sizes are not powers of two, as this increases the number of cylinders included before the rotational pattern repeats ( fs_cpc).  The size of the rotational layout tables is derived from the number of bytes remaining in (struct fs). 

MAXBPG bounds the number of blocks of data per cylinder group, and is limited by the fact that cylinder groups are at most one block.  The size of the free block table is derived from the size of blocks and the number of remaining bytes in the cylinder group structure (struct cg). 

inode: The inode is the focus of all file activity in the file system.  There is a unique inode allocated for each active file, each current directory, each mounted-on file, text file, and the root. An inode is “named” by its device/i-number pair. For further information, see the include file ufs/inode.h. 

SEE ALSO

mkfs(8)

Sun Release 4.0  —  Last change: 22 March 1989