incubator-nuttx/fs/fat/Kconfig

#
# For a description of the syntax of this configuration file,
# see the file kconfig-language.txt in the NuttX tools repository.
#

config FS_FAT
	bool "FAT file system"
	default n
	depends on !DISABLE_MOUNTPOINT
	---help---
		Enable FAT filesystem support

if FS_FAT

config FAT_COMPUTE_FSINFO
	bool "FAT compute free space in FSINFO at mount time"
	default n
	---help---
		Enables the computation of free clusters at mount time as suggested by the
		white paper for FAT. The standard behavior of NuttX is to trust the stored
		value and only recompute it once required. This works if the file system
		is never mounted to another OS. SD-cards which are mounted to Windows to
		modify the content might report wrong space after reinserting it to NuttX.
		It is recommended to activate this setting if the "SD-Card" is swapped
		between systems.

config FAT_LCNAMES
	bool "FAT upper/lower names"
	default n
	---help---
		Enable use of the NT-style upper/lower case 8.3
		file name support.

config FAT_LFN
	bool "FAT long file names"
	default n
	---help---
		Enable FAT long file names.  NOTE:  Microsoft claims
		patents on FAT long file name technology.  Please read the
		disclaimer in the top-level NOTICE file and only enable this
		feature if you understand these issues.

config FAT_MAXFNAME
	int "FAT maximum file name size"
	depends on FAT_LFN
	default NAME_MAX
	range 12 255
	---help---
		If FAT_LFN is defined, then the default, maximum long file
		name is 255 bytes.  This can eat up a lot of memory (especially stack
		space).  If you are willing to live with some non-standard, short long
		file names, then define this value to be something more reasonable.  A
		good choice would be the same value as selected for NAME_MAX which will
		limit the visibility of longer file names anyway.

		This setting may not exceed NAME_MAX.  That will be verified at compile
		time.  The minimum values is 12 due to assumptions in internal logic.

config FAT_LFN_ALIAS_HASH
	bool "Use faster method for forming long filename 8.3 alias"
	depends on FAT_LFN
	default n
	---help---
		Long filenames on FAT filesystems have associated 8.3 character alias
		short filenames. The traditional form of these is FILENA~1.EXT with
		a running count of the number of similar names. However creating this
		unique count can take several seconds if there are many similarly named
		files in the directory. Enabling FAT_LFN_ALIAS_HASH uses an alternative
		format of FI0123~1.TXT where the four digits are a hash of the original
		filename. This method is similar to what is used by Windows 2000 and
		later.

config FAT_LFN_UTF8
	bool "Allow UTF8 long filenames"
	depends on FAT_LFN
	default n
	---help---
		UTF8 long filenames are accepted and converted to UCS2.

config FAT_LFN_ALIAS_TRAILCHARS
	int "Number of trailing characters to use for 8.3 alias"
	depends on FAT_LFN
	default 0
	---help---
		Traditional format for long filename 8.3 aliases takes first 6
		characters of long filename. If this option is set to N > 0,
		NuttX will instead take first 6-N and last N characters to form
		the short name. This is useful for filenames like "datafile12.txt"
		where the first characters would always remain the same.

config FS_FATTIME
	bool "FAT timestamps"
	default n
	---help---
		Support FAT date and time. NOTE:  There is not
		much sense in supporting FAT date and time unless you have a
		hardware RTC or other way to get the time and date.

config FAT_FORCE_INDIRECT
	bool "Force direct transfers"
	default n
	---help---
		Normally, the default behavior for the FAT file system is to perform
		data transfers indirectly though specially allocated sector buffers
		or, under certain circumstances, directly through user provided
		buffers .  These circumstances are:  (1) The transfer is being
		performed from the beginning of a sector (2) the user-provided
		buffer will hold the full sector of data.

		Some hardware, however, may require special DMA-capable memory or
		specially aligned memory in order to perform the transfers.  In this
		case, there may be no circumstance where the user buffer can be used.
		Selecting this option will disable all attempts to use the user-
		provided buffer:  All transfers will be force to be performed
		indirectly through the FAT file systems sector buffers.

		Note: This will have the negative impact of:  (1) An extra data
		copy to transfer the data between the user buffer and the FAT file
		systems internal sector buffers, and (2) A loss of performance
		because I/O will be limited to one sector at a time.

		This would typically be used with CONFIG_FAT_DMAMEMORY so that
		special memory allocators are also used and transfers are also
		performed using only that specially allocated memory.
		CONFIG_FAT_DMAMEMORY, on the other hand, is often used without
		CONFIG_FAT_FORCE_INDIRECT when the user memory buffers may come
		from mixed locations, some of which are DMA-able and some of
		which are not.  But CONFIG_FAT_FORCE_INDIRECT could be used
		without CONFIG_FAT_DMAMEMORY if there is, for example, only a
		memory alignment constraints.

			FORCE_   DMA    DIRECT EXAMPLE USAGE
			INDIRECT MEMORY RETRY
			  Y        Y      *    Use specially allocated memory;
			                       Never use caller provided buffer
			  Y        N      *    Not recommended
			  N        Y     **    Special memory required; user memory
			                       has mixed capability; sometimes
			                       caller memory is not usable
			  N        N      Y    No special memory but there are
			                       alignment requirements; return is
			                       caller buffer is not properly aligned
			  N        N      N    User memory can always be used for
			                       transfer.

			*  CONFIG_DIRECT_RETRY cannot be selected with CONFIG_FORCE_INDIRECT
			** CONFIG_DIRECT_RETRY is automatically selected with CONFIG_DMA_MEMORY

config FAT_DMAMEMORY
	bool "DMA memory allocator"
	default n
	select FAT_DIRECT_RETRY if !FAT_FORCE_INDIRECT
	---help---
		The FAT file system allocates two I/O buffers for data transfer, each
		are the size of one device sector.  One of the buffers is allocated
		once for each FAT volume that is mounted; the other buffers are
		allocated each time a FAT file is opened.

		Some hardware, however, may require special DMA-capable memory in
		order to perform the transfers.  If FAT_DMAMEMORY is defined
		then the architecture-specific hardware must provide the functions
		fat_dma_alloc() and fat_dma_free():  fat_dmalloc() will allocate
		DMA-capable memory of the specified size; fat_dmafree() is the
		corresponding function that will be called to free the DMA-capable
		memory.

			FORCE_   DMA    DIRECT EXAMPLE USAGE
			INDIRECT MEMORY RETRY
			  Y        Y      *    Use specially allocated memory;
			                       Never use caller provided buffer
			  Y        N      *    Not recommended
			  N        Y     **    Special memory required; user memory
			                       has mixed capability; sometimes
			                       caller memory is not usable
			  N        N      Y    No special memory but there are
			                       alignment requirements; return is
			                       caller buffer is not properly aligned
			  N        N      N    User memory can always be used for
			                       transfer.

			*  CONFIG_DIRECT_RETRY cannot be selected with CONFIG_FORCE_INDIRECT
			** CONFIG_DIRECT_RETRY is automatically selected with CONFIG_DMA_MEMORY

config FAT_DIRECT_RETRY
	bool "Direct transfer retry"
	default FAT_DMAMEMORY
	depends on !FAT_FORCE_INDIRECT
	---help---
		The FAT file system contains internal, well aligned sector buffers
		for indirect data transfer.  These transfers are indirect in the
		sense that that the actual transfer occurs into/out of the sector
		buffers and an additional copy is necessary to/from the user-
		provided I/O buffers.  But under certain conditions, the FAT file
		system will use the caller-provided I/O buffers directly to improve
		efficiency.  Those conditions are (1) CONFIG_FAT_FORCE_INDIRECT is
		not defined, (2) The access is to/from the beginning of a sector,
		and (3) the user provided buffer is large enough to hold an entire
		sector.

		The lower level SDIO driver may have, certain requirements on the
		memory buffer in order to perform the transfer.  Perhaps special
		DMA memory should be used (with CONFIG_FAT_DMAMEMORY) or perhaps
		some special memory alignment is required to interface with the
		hardware.

		If this option is selected, then the FAT file system will first
		try the user provided I/O buffer under above conditions.  If the
		transfer fails with -EFAULT. then the FAT file system will try one
		more time using the internal sector buffers.

			FORCE_   DMA    DIRECT EXAMPLE USAGE
			INDIRECT MEMORY RETRY
			  Y        Y      *    Use specially allocated memory;
			                       Never use caller provided buffer
			  Y        N      *    Not recommended
			  N        Y     **    Special memory required; user memory
			                       has mixed capability; sometimes
			                       caller memory is not usable
			  N        N      Y    No special memory but there are
			                       alignment requirements; return is
			                       caller buffer is not properly aligned
			  N        N      N    User memory can always be used for
			                       transfer.

			*  CONFIG_DIRECT_RETRY cannot be selected with CONFIG_FORCE_INDIRECT
			** CONFIG_DIRECT_RETRY is automatically selected with CONFIG_DMA_MEMORY

endif # FAT