Forums » Software Development »
mtd->read (...) returned ECC error
Added by Mads Lind Christiansen about 13 years ago
Hi
We have one mitydsp L138 board which keeps coming up with these mtd->read (....) returned ECC error.
We have several other boards that boots without any errors.
This has been seen on other boards as well, but might have gone when system was last updated (reflashed).
Attached is a boot log of 2 consecutive boots.
According to some stories on the internet it could be related to flashing the partitions from u-boot (but I am not sure if this is the case).
Like http://lists.denx.de/pipermail/u-boot/2007-April/020711.html or http://groups.google.com/group/hawkboard/msg/d989303623486f55
Please notice that we have built the kernel and filesystem our selves.
Any thoughts someone?
Thanks,
Mads
| ecc_errors.txt (25.8 KB) ecc_errors.txt |
Replies (3)
RE: mtd->read (...) returned ECC error - Added by Tim Iskander about 13 years ago
Mads
Did you use the filesystem creation notes from the wiki Linux_Root_File_System ?
We have found that you need to use the pre-fill memory and write in page sizes
to keep the jffs2 filesystem correct.
Hope this helps..
cheers
/Tim
RE: mtd->read (...) returned ECC error - Added by Christian Rückl over 11 years ago
Hi Tim,
since today, the same problems as already mentioned are happening at the moment on my board.
Before I started with my development, I flashed the mityomap-base.jffs2 (MDK_2012-08-10) fs on the NAND. This works pretty well until today.
During the startup sequence I got errors like this:
mtd->read(0x1fb94 bytes from 0x37a046c) returned ECC error
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x037a1b2c: 0x2bb8 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x037a1b30: 0x2bb8 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x037a1b34: 0xd000 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x037a1b38: 0x073a instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x037a1b3c: 0x1000 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x037a1b40: 0x0006 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x037a1b44: 0xacde instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x037a1b48: 0x100a instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x037a1b4c: 0x5e78 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x037a1b50: 0x8c4d instead
Further such events for this erase block will not be printed
mtd->read(0x1e020 bytes from 0x37a1fe0) returned ECC error
Empty flash at 0x037a1fdc ends at 0x037a2000
mtd->read(0x681c bytes from 0x37b97e4) returned ECC error
Empty flash at 0x037b97e0 ends at 0x037b9800
and the mtd->read(...) returned ECC error's also when linux was booting up.
How can things like this happen, suddenly?????
Is the only chance to go back to a working filesystem to make an image of this and build a new jffs2 one?
What is your recommendation?
I hope you can help me.
Thanks
Christian
RE: mtd->read (...) returned ECC error - Added by Jonathan Cormier over 11 years ago
Christian,
It is helpful for us if you create a new post so that we see it otherwise your might be missed. If you are still having this problem could you post your full boot log as well as the steps you took to create this problem.
-Jonathan
