OESF Portables Forum
Everything Else => Sharp Zaurus => Model Specific Forums => Distros, Development, and Model Specific Forums => Archived Forums => C1000/3x00 General discussions => Topic started by: xenophobe on February 16, 2005, 08:56:09 am
-
Is initrd image for SL-C3000 somewhere in Inet?
I try to understand booting process and MD management on SL-C3000 and port it to SL-6000. But I don't have SL-C3000 :-(.
Can somebody help me and post link to initrd image or tarred /etc directory?
--
Eugene
-
This url, put together by 'pelon' is probably going to serve your needs.
http://www.xlfag.com/zaurus (http://www.xlfag.com/zaurus)
It contains a link to a hdd1.tar.gz file which contains the files from the first partition of the hard disk.
Initrd is something else and contains the basic modules that are needed to boot a system if they aren't directly linked into the kernel, I think this is actually stored in the NAND image and since there is no CF/SD based update out for the SL-C3000 yet the NAND based OS file system is probably only accessable by mounting that NAND partition... I don't have one yet so I can't help you there.
-
This url, put together by 'pelon' is probably going to serve your needs.
http://www.xlfag.com/zaurus (http://www.xlfag.com/zaurus)
It contains a link to a hdd1.tar.gz file which contains the files from the first partition of the hard disk.
I have this file, but unfortunatly it does not contain most interesting part - /etc directory.
I have NAND backup image from http://www.xlfag.com/zaurus (http://www.xlfag.com/zaurus)
Is NAND a plain FLASH memory dump? And what is SL-C3000 memory layout?
Can I cut mtdblock3 part from this NAND and mount it as JFFS2?
--
Eugene
-
At a Linux desktop logon as root or su.
First check to see if you have a device called /dev/mtdblock0, if not then you want to...
mknod /dev/mtdblock0 b 31 0
then...
losetup -o 7576608 /dev/loop0 SYSTC300.DBK
modprobe mtdcore
modprobe jffs2
modprobe mtdram total_size=32768 erase_size=256
modprobe mtdchar
modprobe mtdblock
dd if=/dev/loop0 of=/dev/mtdblock0
losetup -d /dev/loop0
mount -t jffs2 /dev/mtdblock0 /mnt/test
Substitute your mount point in the last line.
Its a funny looking root with an initially sparse populated etc directory located in home/etc. Top level etc appears to be a script!
Have fun,
Andy
-
I had tried to follow your instruction exactly, except
losetup -o 7576608 SYSTC300.DBK
I use
losetup -o 7576608 /dev/loop0 SYSTC300.DBK
But I got "Segmentation fault" on mout. and dmesg returns:
kernel: ------------[ cut here ]------------
kernel: kernel BUG at fs/jffs2/nodemgmt.c:580!
kernel: invalid operand: 0000 [#1]
kernel: Modules linked in: jffs2 zlib_deflate loop mtdchar mtdblock mtd_blkdevs
kernel: CPU: 0
kernel: EIP: 0060:[] Not tainted VLI
kernel: EFLAGS: 00010282 (2.6.10-1.760_FC3)
kernel: EIP is at jffs2_mark_node_obsolete+0x5ee/0x644 [jffs2]
kernel: eax: dc15ad4c ebx: c0c8be00 ecx: dc15ad7c edx: dc15ad5c
kernel: esi: d52c60f0 edi: dc15ad6c ebp: 0000029e esp: ca7ecda4
kernel: ds: 007b es: 007b ss: 0068
kernel: Process mount (pid: 15867, threadinfo=ca7ec000 task=d25bb870)
kernel: Stack: 000002a0 0000000c c0021985 0000029e d29a768e 00000000 dc15ad6c d0kernel: ca7ecdec c0c8be00 e2ae6303 d09fc3c0 d09fc3c0 ca7ecdf0 c0c8be00 cakernel: e2ae61de 00000000 00000000 0000003e c0c8bf64 00000000 c0c8be00 d7kernel: Call Trace:
kernel: [] jffs2_build_remove_unlinked_inode+0x1b/0xa7 [jffs2]
kernel: [] jffs2_build_filesystem+0x172/0x27c [jffs2][CODE]
kernel: [] jffs2_do_mount_fs+0x31f/0x358 [jffs2]
kernel: [] jffs2_do_fill_super+0x116/0x1d0 [jffs2]
kernel: [] jffs2_get_sb_mtd+0x82/0xc4 [jffs2]
kernel: [] jffs2_get_sb+0x13f/0x16c [jffs2]
kernel: [] alloc_vfsmnt+0x78/0x9f
kernel: [] do_kern_mount+0x8a/0x13a
kernel: [] do_new_mount+0x61/0x90
kernel: [] do_mount+0x178/0x190
kernel: [] __alloc_pages+0xac/0x28e
kernel: [] sys_mount+0x6a/0xbd
kernel: [] syscall_call+0x7/0xb
kernel: Code: 39 8b 42 08 83 e0 03 48 75 30 8b 42 0c 01 47 0c 8b 42 04 89 47 04
If you can mount SYSTC300.DBK on your linux box by following your instructions, pls. tell me what is your kernel version and linux distro.
My is
[root@justtesting log]# uname -a
Linux justtesting 2.6.10-1.766_FC3 #1 Wed Feb 9 23:06:42 EST 2005 i686 athlon i386 GNU/Linuxand my distro is Fedora Core 3.
But I can change kernel to some of previous versions.
--
Eugene
-
I had tried to follow your instruction exactly, except
losetup -o 7576608 SYSTC300.DBK
I use
losetup -o 7576608 /dev/loop0 SYSTC300.DBK
But I got "Segmentation fault" on mout. and dmesg returns:
kernel: ------------[ cut here ]------------
kernel: kernel BUG at fs/jffs2/nodemgmt.c:580!
kernel: invalid operand: 0000 [#1]
If you can mount SYSTC300.DBK on your linux box by following your instructions, pls. tell me what is your kernel version and linux distro.
My is
[root@justtesting log]# uname -a
Linux justtesting 2.6.10-1.766_FC3 #1 Wed Feb 9 23:06:42 EST 2005 i686 athlon i386 GNU/Linuxand my distro is Fedora Core 3.
But I can change kernel to some of previous versions.
--
Eugene
[div align=\"right\"][a href=\"index.php?act=findpost&pid=67456\"][{POST_SNAPBACK}][/a][/div]
Eugene, well spotted on the typo (/dev/loop0 on the losetup command - updated the post for future reference)
I'm using SuSE 9.2 with all the latest patches and it works fine.
If you have a DVD burner then you might want to try the Live DVD or I'm sure that 'Linux User' or one of the magazines on the shelves probably has a Live DVD of the distro.
andrews@aslinhome:~> uname -a
Linux aslinhome 2.6.8-24.11-default #1 Fri Jan 14 13:01:26 UTC 2005 i686 i686 i386 GNU/Linux
andrews@aslinhome:~>
- Andy