[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-arm-kernel
Subject:    Re: Strange problem while mount JFFS2 fs on MTD, 2.6.9 kernel
From:       James Steward <jamessteward () optusnet ! com ! au>
Date:       2006-02-23 20:59:57
Message-ID: 1140728397.4171.9.camel () JS2-LINUX
[Download RAW message or body]

On Thu, 2006-02-23 at 13:49 +0800, tao yuan wrote:
> it's equal to the jffs2 file I generated with "
> mkfs.jffs2 -r ./rootfs -o 16m.jffs2 -e 0x40000 --pad=0x1000000".    dump
> command of blob also can show expected thing on the place.

Is it really a 256KiB block size device?  Have you tried specifying the
erase block size as --eraseblock%6KiB ? (If it really is a 256 KiB
device).

> During kernel boots up it can detect Flash by CFI and know it support
> Intel's extend feature and know it 2x16 32 bits width.  The strange thing
> is, I can't mount JFFS2 partition, it always report:
> jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0xxxxxxxxx:
> 0x0fd7 instead
> jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0xxxxxxxxx:
> 0x6614 instead
> .....
> Cowardly refusing to erase blocks on filesystem with no valid JFFS2 nodes
> and then kernel panic b/c without root filesystem.

So the kernel is reading something (0x0fd7 and 0x6614) but just not what
it expects, and certainly not zeros.

>
> Since failed to use JFFS2 as root filesystem I tried to use NFS as root
> filesystem. After boot the system up I tried to "hexdump /dev/mtdblock0" to
> try to show something where blob was located, but displaied context is
> always 0, which is not the case since blob is there and functional.
>
Do the boot messages show that the kernel has found all partitions in
the mtd flash?  I use command line params to specify the partitions, as
in..
mtdparts=phys_mapped_flash:256k(uboot)ro,128k(env)ro,-(root)
(Yours may be different).

Then during the boot sequence I see..
physmap flash device: 1000000 at 0
phys_mapped_flash: Found 1 x16 devices at 0x0 in 16-bit bank
 Intel/Sharp Extended Query Table at 0x0031
Using buffer write method
cfi_cmdset_0001: Erase suspend on write enabled
3 cmdlinepart partitions found on MTD device phys_mapped_flash
Creating 3 MTD partitions on "phys_mapped_flash":
0x00000000-0x00040000 : "uboot"
0x00040000-0x00060000 : "env"
0x00060000-0x01000000 : "root"

Do you get something similar?

Regards,
James.


-------------------------------------------------------------------
List admin: http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm-kernel
FAQ:        http://www.arm.linux.org.uk/mailinglists/faq.php
Etiquette:  http://www.arm.linux.org.uk/mailinglists/etiquette.php
[prev in list] [next in list] [prev in thread] [next in thread] 

Configure | About | News | Add a list | Sponsored by KoreLogic