UBI IO errors with GPMI NAND

I'm using a i.Mx6 Solo based custom board with the below NAND chip.
<<NAND device: Manufacturer ID: 0x98, Chip ID: 0xac (Toshiba NAND 512MiB 1,8V 8-bit), page size: 4096, OOB size: 128>>

I'm getting warnings like the one below while attaching the mtd to ubi, with a backtrace.

<<UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 969:4096, read only 4096 bytes, retry>>

According to [1], I ran mtd-tests. But the mtd_oobtest failed to proceed as the GPMI-NAND doesn't define a valid nand_ecclayout structure. (drivers/mtd/nand/gpmi-nand/gpmi-nand.c) [l2-mtd.git]

I guess that the mtd_oobtest can't be performed without a valid nand_ecclayout as the ooblen, oobavail are taken from this struct when trying to write to the OOB area in the test.

Did someone try to execute this test with gpmi-nand driver?

The weird behavior is, I see the errors while mounting an empty flash as well. I need to do a ubiattach followed by ubiformat. Please see the detailed log.

[1] http://www.linux-mtd.infradead.org/faq/ubi.html#L_ecc_error

<Detailed log>

root@BOX:/# ubiattach -m 6 /dev/ubi_ctrl
[   77.053308] UBI: attaching mtd6 to ubi2
[   77.058095] UBI: physical eraseblock size:   262144 bytes (256 KiB)
[   77.066010] UBI: logical eraseblock size:    253952 bytes
[   77.072237] UBI: smallest flash I/O unit:    4096
[   77.077782] UBI: VID header offset:          4096 (aligned 4096)
[   77.085361] UBI: data offset:                8192
[   78.653120] UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 714:4096, read only 4096 bytes, retry
[   78.668408] UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 714:4096, read only 4096 bytes, retry
[   78.683707] UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 714:4096, read only 4096 bytes, retry
[   78.699016] UBI error: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 714:4096, read 4096 bytes
[   78.711264] Backtrace:
[   78.714606] [<80011464>] (dump_backtrace+0x0/0x10c) from [<80414774>] (dump_stack+0x18/0x1c)
[   78.724676]  r6:00001000 r5:ffffffb6 r4:8a311800 r3:60000013
[   78.732046] [<8041475c>] (dump_stack+0x0/0x1c) from [<802bb030>] (ubi_io_read+0x1fc/0x2d8)
[   78.741991] [<802bae34>] (ubi_io_read+0x0/0x2d8) from [<802bb56c>] (ubi_io_read_vid_hdr+0x78/0x220)
[   78.752734] [<802bb4f4>] (ubi_io_read_vid_hdr+0x0/0x220) from [<802bfbdc>] (ubi_attach+0x31c/0x1448)
[   78.763593] [<802bf8c0>] (ubi_attach+0x0/0x1448) from [<802b5b6c>] (ubi_attach_mtd_dev+0x6a8/0xcf4)
[   78.774390] [<802b54c4>] (ubi_attach_mtd_dev+0x0/0xcf4) from [<802b6460>] (ctrl_cdev_ioctl+0xd8/0x184)
[   78.785485] [<802b6388>] (ctrl_cdev_ioctl+0x0/0x184) from [<800dbaa4>] (do_vfs_ioctl+0x570/0x5e4)
[   78.796100]  r7:00000003 r6:8a610580 r5:8a21fc58 r4:7e9ccc2c
[   78.803564] [<800db534>] (do_vfs_ioctl+0x0/0x5e4) from [<800dbb58>] (sys_ioctl+0x40/0x64)
[   78.813471]  r9:8a616000 r8:8000d944 r7:00000003 r6:40186f40 r5:7e9ccc2c
r4:8a610580
[   78.823947] [<800dbb18>] (sys_ioctl+0x0/0x64) from [<8000d7c0>] (ret_fast_syscall+0x0/0x30)
[   78.833987]  r7:00000036 r6:00008b20 r5:00000000 r4:00000000
[   79.467228] UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 969:4096, read only 4096 bytes, retry
[   79.482716] UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 969:4096, read only 4096 bytes, retry
[   79.498185] UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 969:4096, read only 4096 bytes, retry
[   79.513668] UBI error: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 969:4096, read 4096 bytes
[   79.526009] Backtrace:
[   79.529352] [<80011464>] (dump_backtrace+0x0/0x10c) from [<80414774>] (dump_stack+0x18/0x1c)
[   79.539499]  r6:00001000 r5:ffffffb6 r4:8a311800 r3:60000013
[   79.546945] [<8041475c>] (dump_stack+0x0/0x1c) from [<802bb030>] (ubi_io_read+0x1fc/0x2d8)
[   79.556920] [<802bae34>] (ubi_io_read+0x0/0x2d8) from [<802bb56c>] (ubi_io_read_vid_hdr+0x78/0x220)
[   79.567730] [<802bb4f4>] (ubi_io_read_vid_hdr+0x0/0x220) from [<802bfbdc>] (ubi_attach+0x31c/0x1448)
[   79.578628] [<802bf8c0>] (ubi_attach+0x0/0x1448) from [<802b5b6c>] (ubi_attach_mtd_dev+0x6a8/0xcf4)
[   79.589458] [<802b54c4>] (ubi_attach_mtd_dev+0x0/0xcf4) from [<802b6460>] (ctrl_cdev_ioctl+0xd8/0x184)
[   79.600529] [<802b6388>] (ctrl_cdev_ioctl+0x0/0x184) from [<800dbaa4>] (do_vfs_ioctl+0x570/0x5e4)
[   79.611135]  r7:00000003 r6:8a610580 r5:8a21fc58 r4:7e9ccc2c
[   79.618593] [<800db534>] (do_vfs_ioctl+0x0/0x5e4) from [<800dbb58>] (sys_ioctl+0x40/0x64)
[   79.628488]  r9:8a616000 r8:8000d944 r7:00000003 r6:40186f40 r5:7e9ccc2c
r4:8a610580
[   79.638971] [<800dbb18>] (sys_ioctl+0x0/0x64) from [<8000d7c0>] (ret_fast_syscall+0x0/0x30)
[   79.649009]  r7:00000036 r6:00008b20 r5:00000000 r4:00000000
[   80.160882] UBI error: scan_peb: bad image sequence number -1059128657 in PEB 1168, expected -2128599336
[   80.172534] UBI error: ubi_attach_mtd_dev: failed to attach mtd6, error -22
ubiattach: error!: cannot attach mtd6
           error 22 (Invalid argument)
root@BOX:/# ubiformat /dev/mtd6
ubiformat: mtd6 (nand), size 307232768 bytes (293.0 MiB), 1172 eraseblocks of 262144 bytes (256.0 KiB), min. I/O size 4096 bytes
libscan: scanning eraseblock 1171 -- 100 % complete
ubiformat: 1170 eraseblocks have valid erase counter, mean value is 1
ubiformat: warning!: 2 of 1172 eraseblocks contain non-ubifs data
ubiformat: continue? (yes/no)  yes
ubiformat: formatting eraseblock 1171 -- 100 % complete
root@BOX:/# ubiattach -m 6 /dev/ubi_ctrl
[  112.615497] UBI: attaching mtd6 to ubi2
[  112.620299] UBI: physical eraseblock size:   262144 bytes (256 KiB)
[  112.628266] UBI: logical eraseblock size:    253952 bytes
[  112.634542] UBI: smallest flash I/O unit:    4096
[  112.640071] UBI: VID header offset:          4096 (aligned 4096)
[  112.647673] UBI: data offset:                8192
[  115.211159] UBI: max. sequence number:       0
[  115.231875] UBI: attached mtd6 to ubi2
[  115.236460] UBI: MTD device name:            "user_data2"
[  115.242453] UBI: MTD device size:            293 MiB
[  115.248174] UBI: number of good PEBs:        1172
[  115.253620] UBI: number of bad PEBs:         0
[  115.258763] UBI: number of corrupted PEBs:   0
[  115.263925] UBI: max. allowed volumes:       128
[  115.269230] UBI: wear-leveling threshold:    4096
[  115.274623] UBI: number of internal volumes: 1
[  115.279726] UBI: number of user volumes:     0
[  115.284826] UBI: available PEBs:             1157
[  115.290158] UBI: total number of reserved PEBs: 15
[  115.295582] UBI: number of PEBs reserved for bad PEB handling: 11
[  115.302857] UBI: max/mean erase counter: 3/2
[  115.307752] UBI: image sequence number:  552854966
[  115.313163] UBI: background thread "ubi_bgt2d" started, PID 134
UBI device number 2, total 1172 LEBs (297631744 bytes, 283.8 MiB), available 1157 LEBs (293822464 bytes, 280.2 MiB), LEB size 253952 bytes (248.0 KiB)
</Detailed log>

The gpmi-nand can not run in the imx6 solo board. Please check the PINMUX for the gpmi. It is not ready.

thanks.

Huang Shijie

Hello Huang,

I adapted the PINMUX for Solo variant. Is there any other thing other than PINUMX which restricts the usage on i.Mx6 solo?

Regards,
Vikram

Hi Vikram,

Can you please post your patches for adding mx6solo support?

Regards,

Fabio Estevam

It's no where mainline ready. But still I can post it for your reference.

Regards,
Vikram