JFFS2 noise okay or a problem?

Andy Green andy at openmoko.com
Sat Jan 26 10:48:48 CET 2008

Hash: SHA1

Hi folks -

I don't know if it is because I am running a config the hardware ECC
that made trouble for Werner and Xora, but I am noticing some sticky
warnings coming from the rootfs JFFS2 and the return of an error that
disappeared for a long while.

[    4.830000] Empty flash at 0x05d64a08 ends at 0x05d65000
[    4.835000] jffs2_scan_inode_node(): CRC failed on node at
0x05d6cfe8: Read 0xffffffff, calculated 0x28d8485d
[    4.855000] Empty flash at 0x05d6d02c ends at 0x05d6d800
[    4.870000] Empty flash at 0x05d74a60 ends at 0x05d75000
[    4.875000] Empty flash at 0x05d77260 ends at 0x05d77800
[    4.880000] Empty flash at 0x05d7d260 ends at 0x05d7d800
[    5.415000] Empty flash at 0x09741260 ends at 0x09741800
[    5.550000] jffs2_scan_inode_node(): CRC failed on node at
0x0a1807cc: Read 0xffffffff, calculated 0x45c94d91
[    5.580000] Empty flash at 0x0a180810 ends at 0x0a181000
[    5.600000] Empty flash at 0x0a186100 ends at 0x0a186800
[    5.620000] Empty flash at 0x0a188814 ends at 0x0a189000
[    5.635000] Empty flash at 0x0a18fb18 ends at 0x0a190000
[    5.650000] Empty flash at 0x0a193820 ends at 0x0a194000
[    5.650000] jffs2_scan_eraseblock(): Node at 0x0a19b7fc {0x1985,
0xe002, 0xe0021985) has invalid CRC 0x00000044 (calculated 0xd7cd6a7b)
[    5.655000] jffs2_scan_inode_node(): CRC failed on node at
0x0a19efe8: Read 0xffffffff, calculated 0xc55098de
[    5.660000] Empty flash at 0x0a19f270 ends at 0x0a19f800
[    5.790000] jffs2_scan_inode_node(): CRC failed on node at
0x0acb8fec: Read 0xffffffff, calculated 0x048226cc
[    5.800000] Empty flash at 0x0acb9030 ends at 0x0acb9800
[    5.810000] Empty flash at 0x0acbd94c ends at 0x0acbe000
[    6.540000] fork failed for JFFS2 garbage collect thread: 513  <= GAH
[    6.540000] VFS: Mounted root (jffs2 filesystem).

Initially these Empty flash notices disappeared entirely after using the
canned filesystem from Werner, now there seems to be more of them.  But
it is true I am randomly resetting the board all day.

The failure of the garbage collector to fork has reappeared as well.

The filesystem is not blowing chunks so maybe it's all fine... Googling
about the "Empty flash" thing says it is "harmless".

Does this represent a problem or is it okay for us?

- -Andy
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org


More information about the openmoko-kernel mailing list