1 # SPDX-License-Identifier: GPL-2.0-only
3 tristate "Enable UBI - Unsorted block images"
6 UBI is a software layer above MTD layer which admits use of LVM-like
7 logical volumes on top of MTD devices, hides some complexities of
8 flash chips like wear and bad blocks and provides some other useful
9 capabilities. Please, consult the MTD web site for more details
10 (www.linux-mtd.infradead.org).
14 config MTD_UBI_WL_THRESHOLD
15 int "UBI wear-leveling threshold"
19 This parameter defines the maximum difference between the highest
20 erase counter value and the lowest erase counter value of eraseblocks
21 of UBI devices. When this threshold is exceeded, UBI starts performing
22 wear leveling by means of moving data from eraseblock with low erase
23 counter to eraseblocks with high erase counter.
25 The default value should be OK for SLC NAND flashes, NOR flashes and
26 other flashes which have eraseblock life-cycle 100000 or more.
27 However, in case of MLC NAND flashes which typically have eraseblock
28 life-cycle less than 10000, the threshold should be lessened (e.g.,
29 to 128 or 256, although it does not have to be power of 2).
31 config MTD_UBI_BEB_LIMIT
32 int "Maximum expected bad eraseblock count per 1024 eraseblocks"
36 This option specifies the maximum bad physical eraseblocks UBI
37 expects on the MTD device (per 1024 eraseblocks). If the underlying
38 flash does not admit of bad eraseblocks (e.g. NOR flash), this value
41 NAND datasheets often specify the minimum and maximum NVM (Number of
42 Valid Blocks) for the flashes' endurance lifetime. The maximum
43 expected bad eraseblocks per 1024 eraseblocks then can be calculated
44 as "1024 * (1 - MinNVB / MaxNVB)", which gives 20 for most NANDs
45 (MaxNVB is basically the total count of eraseblocks on the chip).
47 To put it differently, if this value is 20, UBI will try to reserve
48 about 1.9% of physical eraseblocks for bad blocks handling. And that
49 will be 1.9% of eraseblocks on the entire NAND chip, not just the MTD
50 partition UBI attaches. This means that if you have, say, a NAND
51 flash chip admits maximum 40 bad eraseblocks, and it is split on two
52 MTD partitions of the same size, UBI will reserve 40 eraseblocks when
53 attaching a partition.
55 This option can be overridden by the "mtd=" UBI module parameter or
56 by the "attach" ioctl.
58 Leave the default value if unsure.
60 config MTD_UBI_FASTMAP
61 bool "UBI Fastmap (Experimental feature)"
64 Important: this feature is experimental so far and the on-flash
65 format for fastmap may change in the next kernel versions
67 Fastmap is a mechanism which allows attaching an UBI device
68 in nearly constant time. Instead of scanning the whole MTD device it
69 only has to locate a checkpoint (called fastmap) on the device.
70 The on-flash fastmap contains all information needed to attach
71 the device. Using fastmap makes only sense on large devices where
72 attaching by scanning takes long. UBI will not automatically install
73 a fastmap on old images, but you can set the UBI module parameter
74 fm_autoconvert to 1 if you want so. Please note that fastmap-enabled
75 images are still usable with UBI implementations without
76 fastmap support. On typical flash devices the whole fastmap fits
77 into one PEB. UBI will reserve PEBs to hold two fastmaps.
82 tristate "MTD devices emulation driver (gluebi)"
84 This option enables gluebi - an additional driver which emulates MTD
85 devices on top of UBI volumes: for each UBI volumes an MTD device is
86 created, and all I/O to this MTD device is redirected to the UBI
87 volume. This is handy to make MTD-oriented software (like JFFS2)
88 work on top of UBI. Do not enable this unless you use legacy
92 bool "Read-only block devices on top of UBI volumes"
96 This option enables read-only UBI block devices support. UBI block
97 devices will be layered on top of UBI volumes, which means that the
98 UBI driver will transparently handle things like bad eraseblocks and
99 bit-flips. You can put any block-oriented file system on top of UBI
100 volumes in read-only mode (e.g., ext4), but it is probably most
101 practical for read-only file systems, like squashfs.
103 When selected, this feature will be built in the UBI driver.
105 If in doubt, say "N".