Given the vast array of file systems out there, I'm certain that exceptions exist, but traditionally, the inode had an array of disk block numbers in it.
For example, in /usr/include/linux/ext3_fs.h, I see a definition of struct ext3_inode.
Inside struct ext3_inode, I see a member i_block[EXT3_N_BLOCKS];/* Pointers to blocks */
Different file systems have had different ways of keeping track of which disk blocks belong to an inode (the on-disk data structure that represents the file's data). Some have an array of block numbers, some have an array of runs or extents, a count plus the beginning block number of a run of contiguous blocks. The Berkeley FFS inode had an array of block numbers, and array of block numbers, each of those blocks contained data block numbers, and a block number that contained block numbers that contained data block numbers.
The whole thing gets a bit weirder for "log structured file systems", but those are the exception rather than the rule.
Take a look at linux-ftools. This suite of tools is specifically designed to analyze the buffers and cache. It includes the following tools:
fincore
$ fincore [options] files...
--pages=false Do not print pages
--summarize When comparing multiple files, print a summary report
--only-cached Only print stats for files that are actually in cache.
root@xxxxxx:/var/lib/mysql/blogindex# fincore --pages=false --summarize --only-cached *
stats for CLUSTER_LOG_2010_05_21.MYI: file size=93840384 , total pages=22910 , cached pages=1 , cached size=4096, cached perc=0.004365
stats for CLUSTER_LOG_2010_05_22.MYI: file size=417792 , total pages=102 , cached pages=1 , cached size=4096, cached perc=0.980392
stats for CLUSTER_LOG_2010_05_23.MYI: file size=826368 , total pages=201 , cached pages=1 , cached size=4096, cached perc=0.497512
stats for CLUSTER_LOG_2010_05_24.MYI: file size=192512 , total pages=47 , cached pages=1 , cached size=4096, cached perc=2.127660
...
NOTE: In the example output above, any files in the directory, /var/lib/mysql/blogindex, that are being cached, are displayed. In this case there are several files named CLUSTER_LOG_*.MYI.
fadvise
SYNTAX: filename mode [offset] [,length]
Where mode can be:
POSIX_FADV_NORMAL No further special treatment.
POSIX_FADV_RANDOM Expect random page references.
POSIX_FADV_SEQUENTIAL Expect sequential page references.
POSIX_FADV_WILLNEED Will need these pages.
POSIX_FADV_DONTNEED Dont need these pages.
POSIX_FADV_NOREUSE Data will be accessed once.
Allows an application to to tell the kernel how it expects to use a file handle,
so that the kernel can choose appropriate read-ahead and caching techniques for
access to the corresponding file. This is similar to the POSIX version of the
madvise system call, but for file access instead of memory access. The
sys_fadvise64() function is obsolete and corresponds to a broken glibc API,
sys_fadvise64_64() is the fixed version. The following are the values for the
advice parameter:
FADV_NORMAL
No special treatment.
FADV_RANDOM
Expect page references in random order.
FADV_SEQUENTIAL
Expect page references in sequential order.
FADV_WILLNEED
Expect access in the near future.
FADV_DONTNEED
Do not expect access in the near future. Subsequent access of pages in this
range will succeed, but will result either in reloading of the memory contents
from the underlying mapped file or zero-fill-in-demand pages for mappings
without an underlying file.
FADV_NOREUSE
Access data only once.
fallocate
SYNTAX: fallocate file length
fallocate() allows the caller to directly manipulate the allocated disk space
for the file referred to by fd for the byte range starting at offset and
continuing for len bytes.
The mode argument determines the operation to be performed on the given
range. Currently only one flag is supported for mode:
FALLOC_FL_KEEP_SIZE
This flag allocates and initializes to zero the disk space within the range
specified by offset and len. After a successful call, subsequent writes into
this range are guaranteed not to fail because of lack of disk
space. Preallocating zeroed blocks beyond the end of the file is useful for
optimizing append workloads. Preallocating blocks does not change the file size
(as reported by stat(2)) even if it is less than offset+len.
If FALLOC_FL_KEEP_SIZE flag is not specified in mode, the default behavior is
almost same as when this flag is specified. The only difference is that on
success, the file size will be changed if offset + len is greater than the file
size. This default behavior closely resembles the behavior of the
posix_fallocate(3) library function, and is intended as a method of optimally
implementing that function.
Because allocation is done in block size chunks, fallocate() may allocate a
larger range than that which was specified.
Emptying the buffers cache
If you ever want to empty it you can use this chain of commands.
$ free && sync && echo 3 > /proc/sys/vm/drop_caches && free
total used free shared buffers cached
Mem: 1018916 980832 38084 0 46924 355764
-/+ buffers/cache: 578144 440772
Swap: 2064376 128 2064248
total used free shared buffers cached
Mem: 1018916 685008 333908 0 224 108252
-/+ buffers/cache: 576532 442384
Swap: 2064376 128 2064248
You can signal the Linux Kernel to drop various aspects of cached items by changing the numeric argument to the above command.
NOTE: clean up memory of unnecessary things (Kernerl 2.6.16 or newer). Always make sure to run sync first to flush useful things out to disk!!!
To free pagecache:
$ echo 1 > /proc/sys/vm/drop_caches
To free dentries and inodes:
$ echo 2 > /proc/sys/vm/drop_caches
To free pagecache, dentries and inodes:
$ echo 3 > /proc/sys/vm/drop_caches
The above are meant to be run as root. If you're trying to do them using sudo then you'll need to change the syntax slightly to something like these:
$ sudo sh -c 'echo 1 >/proc/sys/vm/drop_caches'
$ sudo sh -c 'echo 2 >/proc/sys/vm/drop_caches'
$ sudo sh -c 'echo 3 >/proc/sys/vm/drop_caches'
Alternatives methods to the above:
# alternative #1
$ sudo tee /proc/sys/vm/drop_caches <<<1
# alternative #2
$ echo "echo 1 > /proc/sys/vm/drop_caches" | sudo sh
Why the change in syntax? The /bin/echo program is running as root, because of sudo, but the shell that's redirecting echo's output to the root-only file is still running as you. Your current shell does the redirection before sudo starts.
References
Best Answer
It's just another level of caching. Even if all the path components can be found in other caches, it still takes time to traverse them all. So the directory cache provides a way to cache whole paths and avoid looking things up path element by path element from the root each time.