Problem accessing mtdblock0 from Debian Squeeze

[maisondouf]

I'd remark some erreors messages in 'dmesg' when booting Debian Squeeze on my plug.

Code:

[  172.572294] uncorrectable error :
[  172.575545] end_request: I/O error, dev mtdblock0, sector 0
[  172.581325] Buffer I/O error on device mtdblock0, logical block 0
[  172.587883] uncorrectable error :
[  172.591134] end_request: I/O error, dev mtdblock0, sector 8
[  172.596915] Buffer I/O error on device mtdblock0, logical block 1
[  172.603855] uncorrectable error :
[  172.607101] end_request: I/O error, dev mtdblock0, sector 16
[  172.612969] Buffer I/O error on device mtdblock0, logical block 2
[  172.619626] uncorrectable error :
[  172.622896] end_request: I/O error, dev mtdblock0, sector 24
[  172.628769] Buffer I/O error on device mtdblock0, logical block 3
[  172.635427] uncorrectable error :
[  172.638676] end_request: I/O error, dev mtdblock0, sector 0
[  172.644456] Buffer I/O error on device mtdblock0, logical block 0

So I try from OS level to have access into /dev/mtdblock0 and I receive an error.

Code:

root@sheeva-nand:~# hd -n 256 /dev/mtdblock0
hd: /dev/mtdblock0: Input/output error
root@sheeva-nand:~#

This action generate other error messages in 'dmesg'.

I can access /dev/mtblock1 and 2 without any problem

Code:

root@sheeva-nand:~# hd -n 256 /dev/mtdblock1
00000000  55 42 49 23 01 00 00 00  00 00 00 00 00 00 00 07  |UBI#............|
00000010  00 00 02 00 00 00 08 00  00 00 00 00 00 00 00 00  |................|
00000020  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
00000030  00 00 00 00 00 00 00 00  00 00 00 00 92 84 9c e2  |................|
00000040  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
*
00000100

My partitions are difined as this :

mtdparts=orion_nand:0x400000@0x100000(uImage),0x1ef00000@0x500000(rootfs),0xc00000@0x1f400000(debian)

'uImage' and 'rootfs' are standard partition for sheeva and le small last 'debian' (12Mo) contain uImage and uInitrd for Debian on usb disk (/dev/sda2).

so mtd reported correctly as this

Code:

root@sheeva-nand:~# cat /proc/mtd
dev:    size   erasesize  name
mtd0: 00400000 00020000 "uImage"
mtd1: 1ef00000 00020000 "rootfs"
mtd2: 00c00000 00020000 "debian"

Is there a special attribute for mtd0 ?
because devices seem to have the sames

Code:

root@sheeva-nand:~# ls -l /dev/mtdb*
brw-rw---- 1 root disk 31, 0 Sep  3 11:48 /dev/mtdblock0
brw-rw---- 1 root disk 31, 1 Sep  3 11:48 /dev/mtdblock1
brw-rw---- 1 root disk 31, 2 Sep  3 11:48 /dev/mtdblock2

I don't get any error with character devices, I can make "cat /dev/mtd0" or "cat /dev/mtd0ro" and the content seem to be really uImage ....

[aleandro]

same type of errors. Not worth to post them. Cannot say if it seems hardware or software related. I got this first time when I changed the FS on the SD card. I have and updated debian on it. Without updating debian the system was more stable and I didn't see that kind of errors.