Usage:

	cache_dir Type Directory-Name Fs-specific-data [options]

	You can specify multiple cache_dir lines to spread the
	cache among different disk partitions.

	Type specifies the kind of storage system to use. Only "ufs"
	is built by default. To enable any of the other storage systems
	see the --enable-storeio configure option.

	'Directory' is a top-level directory where cache swap
	files will be stored. If you want to use an entire disk
	for caching, this can be the mount-point directory.
	The directory must exist and be writable by the Squid
	process. Squid will NOT create this directory for you.
	Only using COSS, a raw disk device or a stripe file can
	be specified, but the configuration of the "cache_swap_log"
	tag is mandatory.

	The ufs store type:

	"ufs" is the old well-known Squid storage format that has always
	been there.

	cache_dir ufs Directory-Name Mbytes L1 L2 [options]

	'Mbytes' is the amount of disk space (MB) to use under this
	directory.  The default is 100 MB.  Change this to suit your
	configuration.  Do NOT put the size of your disk drive here.
	Instead, if you want Squid to use the entire disk drive,
	subtract 20% and use that value.

	'Level-1' is the number of first-level subdirectories which
	will be created under the 'Directory'.  The default is 16.

	'Level-2' is the number of second-level subdirectories which
	will be created under each first-level directory.  The default
	is 256.

	The aufs store type:

	"aufs" uses the same storage format as "ufs", utilizing
	POSIX-threads to avoid blocking the main Squid process on
	disk-I/O. This was formerly known in Squid as async-io.

	cache_dir aufs Directory-Name Mbytes L1 L2 [options]

	see argument descriptions under ufs above

	The diskd store type:

	"diskd" uses the same storage format as "ufs", utilizing a
	separate process to avoid blocking the main Squid process on
	disk-I/O.

	cache_dir diskd Directory-Name Mbytes L1 L2 [options] [Q1=n] [Q2=n]

	see argument descriptions under ufs above

	Q1 specifies the number of unacknowledged I/O requests when Squid
	stops opening new files. If this many messages are in the queues,
	Squid won't open new files. Default is 64

	Q2 specifies the number of unacknowledged messages when Squid
	starts blocking.  If this many messages are in the queues,
	Squid blocks until it receives some replies. Default is 72

	When Q1 < Q2 (the default), the cache directory is optimized
	for lower response time at the expense of a decrease in hit
	ratio.  If Q1 > Q2, the cache directory is optimized for
	higher hit ratio at the expense of an increase in response
	time.

	The coss store type:

	block-size=n defines the "block size" for COSS cache_dir's.
	Squid uses file numbers as block numbers.  Since file numbers
	are limited to 24 bits, the block size determines the maximum
	size of the COSS partition.  The default is 512 bytes, which
	leads to a maximum cache_dir size of 512<<24, or 8 GB.  Note
	you should not change the COSS block size after Squid
	has written some objects to the cache_dir.

	overwrite-percent=n defines the percentage of disk that COSS
	must write to before a given object will be moved to the
	current stripe.  A value of "n" closer to 100 will cause COSS
	to waste less disk space by having multiple copies of an object
	on disk, but will increase the chances of overwriting a popular
	object as COSS overwrites stripes.  A value of "n" close to 0
	will cause COSS to keep all current objects in the current COSS
	stripe at the expense of the hit rate.  The default value of 50
	will allow any given object to be stored on disk a maximum of
	2 times.

	max-stripe-waste=n defines the maximum amount of space that COSS
	will waste in a given stripe (in bytes).  When COSS writes data
	to disk, it will potentially waste up to "max-size" worth of disk
	space for each 1MB of data written.  If "max-size" is set to a
	large value (ie >256k), this could potentially result in large
	amounts of wasted disk space. Setting this value to a lower value
	(ie 64k or 32k) will result in a COSS disk refusing to cache
	larger objects until the COSS stripe has been filled to within
	"max-stripe-waste" of the maximum size (1MB).

	membufs=n defines the number of "memory-only" stripes that COSS
	will use.  When an cache hit is performed on a COSS stripe before
	COSS has reached the overwrite-percent value for that object,
	COSS will use a series of memory buffers to hold the object in
	while the data is sent to the client.  This will define the maximum
	number of memory-only buffers that COSS will use.  The default value
	is 10, which will use a maximum of 10MB of memory for buffers.

	maxfullbufs=n defines the maximum number of stripes a COSS partition
	will have in memory waiting to be freed (either because the disk is
	under load and the stripe is unwritten, or because clients are still
	transferring data from objects using the memory).  In order to try
	and maintain a good hit rate under load, COSS will reserve the last
	2 full stripes for object hits. (ie a COSS cache_dir will reject
	new objects when the number of full stripes is 2 less than maxfullbufs)

	The null store type:

	no options are allowed or required

	Common options:

	no-store, no new objects should be stored to this cache_dir

	min-size=n, refers to the min object size this storedir will accept.
	It's used to restrict a storedir to only store large objects
	(e.g. aufs) while other storedirs are optimized for smaller objects
	(e.g. COSS). Defaults to 0.

	max-size=n, refers to the max object size this storedir supports.
	It is used to initially choose the storedir to dump the object.
	Note: To make optimal use of the max-size limits you should order
	the cache_dir lines with the smallest max-size value first and the
	ones with no max-size specification last.

	Note that for coss, max-size must be less than COSS_MEMBUF_SZ
	(hard coded at 1 MB).