1 NAND FLASH commands and notes
6 # Dave Ellis, SIXNET, dge@sixnetio.com
8 # SPDX-License-Identifier: GPL-2.0+
13 Print a list of all of the bad blocks in the current device.
16 Print information about the current NAND device.
19 Make device `num' the current device and print information about it.
21 nand erase off|partition size
22 nand erase clean [off|partition size]
23 Erase `size' bytes starting at offset `off'. Alternatively partition
24 name can be specified, in this case size will be eventually limited
25 to not exceed partition size (this behaviour applies also to read
26 and write commands). Only complete erase blocks can be erased.
28 If `erase' is specified without an offset or size, the entire flash
29 is erased. If `erase' is specified with partition but without an
30 size, the entire partition is erased.
32 If `clean' is specified, a JFFS2-style clean marker is written to
33 each block after it is erased.
35 This command will not erase blocks that are marked bad. There is
36 a debug option in cmd_nand.c to allow bad blocks to be erased.
37 Please read the warning there before using it, as blocks marked
38 bad by the manufacturer must _NEVER_ be erased.
41 Print information about all of the NAND devices found.
43 nand read addr ofs|partition size
44 Read `size' bytes from `ofs' in NAND flash to `addr'. Blocks that
45 are marked bad are skipped. If a page cannot be read because an
46 uncorrectable data error is found, the command stops with an error.
48 nand read.oob addr ofs|partition size
49 Read `size' bytes from the out-of-band data area corresponding to
50 `ofs' in NAND flash to `addr'. This is limited to the 16 bytes of
51 data for one 512-byte page or 2 256-byte pages. There is no check
52 for bad blocks or ECC errors.
54 nand write addr ofs|partition size
55 Write `size' bytes from `addr' to `ofs' in NAND flash. Blocks that
56 are marked bad are skipped. If a page cannot be read because an
57 uncorrectable data error is found, the command stops with an error.
59 As JFFS2 skips blocks similarly, this allows writing a JFFS2 image,
60 as long as the image is short enough to fit even after skipping the
61 bad blocks. Compact images, such as those produced by mkfs.jffs2
62 should work well, but loading an image copied from another flash is
63 going to be trouble if there are any bad blocks.
65 nand write.trimffs addr ofs|partition size
66 Enabled by the CONFIG_CMD_NAND_TRIMFFS macro. This command will write to
67 the NAND flash in a manner identical to the 'nand write' command
68 described above -- with the additional check that all pages at the end
69 of eraseblocks which contain only 0xff data will not be written to the
70 NAND flash. This behaviour is required when flashing UBI images
71 containing UBIFS volumes as per the UBI FAQ[1].
73 [1] http://www.linux-mtd.infradead.org/doc/ubi.html#L_flasher_algo
75 nand write.oob addr ofs|partition size
76 Write `size' bytes from `addr' to the out-of-band data area
77 corresponding to `ofs' in NAND flash. This is limited to the 16 bytes
78 of data for one 512-byte page or 2 256-byte pages. There is no check
81 nand read.raw addr ofs|partition [count]
82 nand write.raw addr ofs|partition [count]
83 Read or write one or more pages at "ofs" in NAND flash, from or to
84 "addr" in memory. This is a raw access, so ECC is avoided and the
85 OOB area is transferred as well. If count is absent, it is assumed
86 to be one page. As with .yaffs2 accesses, the data is formatted as
87 a packed sequence of "data, oob, data, oob, ..." -- no alignment of
88 individual pages is maintained.
90 Configuration Options:
93 Enables NAND support and commmands.
95 CONFIG_CMD_NAND_TORTURE
96 Enables the torture command (see description of this command below).
98 CONFIG_MTD_NAND_ECC_JFFS2
99 Define this if you want the Error Correction Code information in
100 the out-of-band data to be formatted to match the JFFS2 file system.
101 CONFIG_MTD_NAND_ECC_YAFFS would be another useful choice for
102 someone to implement.
104 CONFIG_SYS_MAX_NAND_DEVICE
105 The maximum number of NAND devices you want to support.
107 CONFIG_SYS_NAND_MAX_ECCPOS
108 If specified, overrides the maximum number of ECC bytes
109 supported. Useful for reducing image size, especially with SPL.
110 This must be at least 48 if nand_base.c is used.
112 CONFIG_SYS_NAND_MAX_OOBFREE
113 If specified, overrides the maximum number of free OOB regions
114 supported. Useful for reducing image size, especially with SPL.
115 This must be at least 2 if nand_base.c is used.
117 CONFIG_SYS_NAND_MAX_CHIPS
118 The maximum number of NAND chips per device to be supported.
120 CONFIG_SYS_NAND_SELF_INIT
121 Traditionally, glue code in drivers/mtd/nand/nand.c has driven
122 the initialization process -- it provides the mtd and nand
123 structs, calls a board init function for a specific device,
124 calls nand_scan(), and registers with mtd.
126 This arrangement does not provide drivers with the flexibility to
127 run code between nand_scan_ident() and nand_scan_tail(), or other
128 deviations from the "normal" flow.
130 If a board defines CONFIG_SYS_NAND_SELF_INIT, drivers/mtd/nand/nand.c
131 will make one call to board_nand_init(), with no arguments. That
132 function is responsible for calling a driver init function for
133 each NAND device on the board, that performs all initialization
134 tasks except setting mtd->name, and registering with the rest of
135 U-Boot. Those last tasks are accomplished by calling nand_register()
136 on the new mtd device.
138 Example of new init to be added to the end of an existing driver
142 * devnum is the device number to be used in nand commands
143 * and in mtd->name. Must be less than
144 * CONFIG_SYS_NAND_MAX_DEVICE.
146 mtd = &nand_info[devnum];
148 /* chip is struct nand_chip, and is now provided by the driver. */
152 * Fill in appropriate values if this driver uses these fields,
153 * or uses the standard read_byte/write_buf/etc. functions from
154 * nand_base.c that use these fields.
156 chip.IO_ADDR_R = ...;
157 chip.IO_ADDR_W = ...;
159 if (nand_scan_ident(mtd, CONFIG_SYS_MAX_NAND_CHIPS, NULL))
163 * Insert here any code you wish to run after the chip has been
164 * identified, but before any other I/O is done.
167 if (nand_scan_tail(mtd))
170 if (nand_register(devnum))
173 In addition to providing more flexibility to the driver, it reduces
174 the difference between a U-Boot driver and its Linux counterpart.
175 nand_init() is now reduced to calling board_nand_init() once, and
176 printing a size summary. This should also make it easier to
177 transition to delayed NAND initialization.
179 Please convert your driver even if you don't need the extra
180 flexibility, so that one day we can eliminate the old mechanism.
183 CONFIG_SYS_NAND_ONFI_DETECTION
184 Enables detection of ONFI compliant devices during probe.
185 And fetching device parameters flashed on device, by parsing
189 Enables software based BCH ECC algorithm present in lib/bch.c
190 This is used by SoC platforms which do not have built-in ELM
191 hardware engine required for BCH ECC correction.
193 CONFIG_SYS_NAND_BUSWIDTH_16BIT
194 Indicates that NAND device has 16-bit wide data-bus. In absence of this
195 config, bus-width of NAND device is assumed to be either 8-bit and later
196 determined by reading ONFI params.
197 Above config is useful when NAND device's bus-width information cannot
198 be determined from on-chip ONFI params, like in following scenarios:
199 - SPL boot does not support reading of ONFI parameters. This is done to
200 keep SPL code foot-print small.
201 - In current U-Boot flow using nand_init(), driver initialization
202 happens in board_nand_init() which is called before any device probe
203 (nand_scan_ident + nand_scan_tail), thus device's ONFI parameters are
204 not available while configuring controller. So a static CONFIG_NAND_xx
205 is needed to know the device's bus-width in advance.
206 Some drivers using above config are:
207 drivers/mtd/nand/mxc_nand.c
208 drivers/mtd/nand/ndfc.c
209 drivers/mtd/nand/omap_gpmc.c
212 Platform specific options
213 =========================
214 CONFIG_NAND_OMAP_GPMC
215 Enables omap_gpmc.c driver for OMAPx and AMxxxx platforms.
216 GPMC controller is used for parallel NAND flash devices, and can
217 do ECC calculation (not ECC error detection) for HAM1, BCH4, BCH8
218 and BCH16 ECC algorithms.
221 Enables omap_elm.c driver for OMAPx and AMxxxx platforms.
222 ELM controller is used for ECC error detection (not ECC calculation)
223 of BCH4, BCH8 and BCH16 ECC algorithms.
224 Some legacy platforms like OMAP3xx do not have in-built ELM h/w engine,
225 thus such SoC platforms need to depend on software library for ECC error
226 detection. However ECC calculation on such plaforms would still be
227 done by GPMC controller.
229 CONFIG_NAND_OMAP_ECCSCHEME
230 On OMAP platforms, this CONFIG specifies NAND ECC scheme.
231 It can take following values:
232 OMAP_ECC_HAM1_CODE_SW
233 1-bit Hamming code using software lib.
234 (for legacy devices only)
235 OMAP_ECC_HAM1_CODE_HW
236 1-bit Hamming code using GPMC hardware.
237 (for legacy devices only)
238 OMAP_ECC_BCH4_CODE_HW_DETECTION_SW
239 4-bit BCH code (unsupported)
240 OMAP_ECC_BCH4_CODE_HW
241 4-bit BCH code (unsupported)
242 OMAP_ECC_BCH8_CODE_HW_DETECTION_SW
244 - ecc calculation using GPMC hardware engine,
245 - error detection using software library.
246 - requires CONFIG_BCH to enable software BCH library
247 (For legacy device which do not have ELM h/w engine)
248 OMAP_ECC_BCH8_CODE_HW
250 - ecc calculation using GPMC hardware engine,
251 - error detection using ELM hardware engine.
252 OMAP_ECC_BCH16_CODE_HW
254 - ecc calculation using GPMC hardware engine,
255 - error detection using ELM hardware engine.
257 How to select ECC scheme on OMAP and AMxx platforms ?
258 -----------------------------------------------------
259 Though higher ECC schemes have more capability to detect and correct
260 bit-flips, but still selection of ECC scheme is dependent on following
261 - hardware engines present in SoC.
262 Some legacy OMAP SoC do not have ELM h/w engine thus such
263 SoC cannot support BCHx_HW ECC schemes.
264 - size of OOB/Spare region
265 With higher ECC schemes, more OOB/Spare area is required to
266 store ECC. So choice of ECC scheme is limited by NAND oobsize.
268 In general following expression can help:
269 NAND_OOBSIZE >= 2 + (NAND_PAGESIZE / 512) * ECC_BYTES
271 NAND_OOBSIZE = number of bytes available in
272 OOB/spare area per NAND page.
273 NAND_PAGESIZE = bytes in main-area of NAND page.
274 ECC_BYTES = number of ECC bytes generated to
275 protect 512 bytes of data, which is:
276 3 for HAM1_xx ecc schemes
277 7 for BCH4_xx ecc schemes
278 14 for BCH8_xx ecc schemes
279 26 for BCH16_xx ecc schemes
281 example to check for BCH16 on 2K page NAND
284 2 + (2048 / 512) * 26 = 106 > NAND_OOBSIZE
285 Thus BCH16 cannot be supported on 2K page NAND.
287 However, for 4K pagesize NAND
291 2 + (4096 / 512) * 26 = 210 < NAND_OOBSIZE
292 Thus BCH16 can be supported on 4K page NAND.
298 The current NAND implementation is based on what is in recent
299 Linux kernels. The old legacy implementation has been removed.
301 If you have board code which used CONFIG_NAND_LEGACY, you'll need
302 to convert to the current NAND interface for it to continue to work.
304 The Disk On Chip driver is currently broken and has been for some time.
305 There is a driver in drivers/mtd/nand, taken from Linux, that works with
306 the current NAND system but has not yet been adapted to the u-boot
309 Additional improvements to the NAND subsystem by Guido Classen, 10-10-2006
311 JFFS2 related commands:
313 implement "nand erase clean" and old "nand erase"
314 using both the new code which is able to skip bad blocks
315 "nand erase clean" additionally writes JFFS2-cleanmarkers in the oob.
317 Miscellaneous and testing commands:
319 create an artificial bad block (for testing bad block handling)
321 "scrub [offset length]"
322 like "erase" but don't skip bad block. Instead erase them.
323 DANGEROUS!!! Factory set bad blocks will be lost. Use only
324 to remove artificial bad blocks created with the "markbad" command.
327 Torture block to determine if it is still reliable.
328 Enabled by the CONFIG_CMD_NAND_TORTURE configuration option.
329 This command returns 0 if the block is still reliable, else 1.
330 If the block is detected as unreliable, it is up to the user to decide to
331 mark this block as bad.
332 The analyzed block is put through 3 erase / write cycles (or less if the block
333 is detected as unreliable earlier).
334 This command can be used in scripts, e.g. together with the markbad command to
335 automate retries and handling of possibly newly detected bad blocks if the
336 nand write command fails.
337 It can also be used manually by users having seen some NAND errors in logs to
338 search the root cause of these errors.
339 The underlying nand_torture() function is also useful for code willing to
340 automate actions following a nand->write() error. This would e.g. be required
341 in order to program or update safely firmware to NAND, especially for the UBI
342 part of such firmware.
345 NAND locking command (for chips with active LOCKPRE pin)
348 set NAND chip to lock state (all pages locked)
351 set NAND chip to lock tight state (software can't change locking anymore)
354 displays current locking status of all pages
356 "nand unlock [offset] [size]"
357 unlock consecutive area (can be called multiple times for different areas)
359 "nand unlock.allexcept [offset] [size]"
360 unlock all except specified consecutive area
362 I have tested the code with board containing 128MiB NAND large page chips
363 and 32MiB small page chips.