Dear Stefan Roese,
In message 200905201201.50810.sr@denx.de you wrote:
=> fsinfo ### filesystem type is JFFS2 Scanning JFFS2 FS: read_nand_cached: error reading nand off 0xabfe00 size 8192 bytes read_nand_cached: error reading nand off 0xadfe00 size 8192 bytes read_nand_cached: error reading nand off 0xac0000 size 8192 bytes read_nand_cached: error reading nand off 0xaffe00 size 8192 bytes
<snip>
The addresses where the read errors occur don't seem to be related to the bad blocks reported by Linux.
No. What does a "normal" "nand read" return when reading this complete partition into RAM? On PPC4xx it returned -117 here. I fixed it with the already posted patches for the 4xx NDFC ECC byte ordering. Perhaps the TQM8548 has similar incompatibility issues with the Linux kernel here too?
I get this:
=> nand read 200000 0 1000000
NAND read: device 0 offset 0x0, size 0x1000000 NAND read from offset 0 failed -74 16777216 bytes read: ERROR
Hm, this is indeed a very helpful message. What does it tell you? I can't parse that. May we please have more readable error messages from the NAND code? Please?!
Is this -74 decimal as EBADMSG or -0x74 = NAND_CMD_STATUS_ERROR1 ?
Best regards,
Wolfgang Denk