Talat,

I've just gone through a few cycles trying to get the lh79520 BDK to build on my Fedora Core 3 machine, and was successful.

What I found is that if you have remenents of other Embedix builds on your system, then those might conflict with the rpm packages that the lh79520 BDK is trying to install.

Do you have the *full* log of the install starting with the original BDK install? That may shed some light on what is going wrong for you.

--
Peter Barada

Hello,
The following is the log of my installation. Please let me know if I am doing anything wrong.
Thanks!

T.O
********************************************************************
OS: Fedora Core 3
Started on March 30, 2005 by Talat Ozyagcilar.

--Starting...

-- found out installed rpm packages
# rpm -qa | grep Embedix
** Output:
Embedix-tools-i386-2.0-4
Embedix-tcspoof-0.1-1
Embedix-tools-arm-doc-1.0-0
Embedix-lwiz-Python-1.0.1-3
Embedix-tools-common-2.0-4
Embedix-rpm-2.5.5-3
Embedix-tools-arm-1.0-0
Embedix-tools-native-2.0-0
** Output end

-- Removed all by
# rpm -e ...

-- Removed /opt/Embedix
# rm -fvr /opt/Embedix/

# cd /media/cdrom/
# ./install.sh
** Output:
Embedix BDK installation for the LH79520
Your distribution must contain a version of RPM.
This install must be run as root
You are root and we will continue
/
/
/
Installing ARM toolchains...
Installing i386 toolchains...
Installing Python...
Installing Native toolchains...
Installing Common...
Installing Spoofing...
Installing RPM...
Extract bdk...
You now have the LH79520 BDK installed. Type the following command:
# chroot /opt/Embedix
# cd /bdk/lh79520
# make -f Makefile.bdk firsttime
# make -f Makefile.bdk deploy
** Output end

-- Running the patch
# cd /home/tozyagcilar/lpd-linux-LLH79520-1.0.1b
# ./install-lpd-linux79530-lineo-overlay.sh modern

-- making firsttime
# make -f Makefile.bdk firsttime

-- Going home...
-- Back in the morning, the compilation failed with error
+ STATUS=0
+ [ 0 -ne 0 ]
+ cd glibc-2.2.1
+ chown -Rh root.root .
Bad exit status from /var/tmp/rpm-tmp.8217 (%prep)
make: *** [firsttime] Error 1
*****************************************************************

Talat,

Those steps look pretty good. I'm trying to repoduce those steps with a slight difference, namely installing the modified Logic scripts/src *outside* of the chroot.

So in my case I'm trying:

#install the base BDK
mount -o loop sharp_lh79520_embedix_bdk.iso /mnt/fs
cd /mnt/fs
./install.sh

#overlay the Logic changes
cd /tmp
tar zxf lpd-linux-LLH79520-1.0.1b.tgz
cd lpd-linux-LLH79520-1.0.1b
./install-lpd-linux79530-lineo-overlay.sh modern

# build the BDK
chroot /opt/Embedix
make -f Makefile.bdk firsttime

I'll report what I find when it completes...

--
Peter Barada

Actually Peter, I skipped chroot'ing in my log somehow, but the way I do it is exactly the same as you do. For me the error comes within like 5 mins.

One more thing, whenI chroot to /opt/Embedix/ I tried chown command on some file and chown did not change anything. I did

"chown tozyagcilar.root Makefile.bdk"

it returned without any error messages, but when I checked the ownership on Makefile.bdk, it was still root.root. This is weird...

>Actually Peter, I skipped chroot'ing in my log somehow, but the way I do it is exactly the >same as you do. For me the error comes within like 5 mins.

Mine completed. The log where you got your error looks like:

-rw-r--r-- ev/users 1956 1999-11-22 20:52:54 glibc-2.2.1/linuxthreads_db/\thread_dbP.h
-rw-r--r-- ev/users 1813 1999-11-08 22:52:10 glibc-2.2.1/linuxthreads_db/\td_ta_set_event.c
-rw-r--r-- ev/users 1821 1999-11-08 22:52:10 glibc-2.2.1/linuxthreads_db/\td_ta_clear_event.c
-rw-r--r-- ev/users 3982 1999-11-08 22:52:10 glibc-2.2.1/linuxthreads_db/\td_ta_event_getmsg.c
+ STATUS=0
+ [ 0 -ne 0 ]
+ cd glibc-2.2.1
+ chown -Rh root.root . # <----- after this you had your error.
+ chmod -R a+rX,g-w,o-w .
+ echo Patch #1:
Patch #1:
+ patch -p1 -s
+ echo Patch #2:
Patch #2:
+ patch -p1 -s
+ perl -pi -e s,/var/mail,/var/spool/mail,; sysdeps/unix/sysv/linux/paths.h
+ find . -type d -a -name CVS -print0
+ xargs -0 rm -rf
+ exit 0
+ [ 1 = 1 ]
+ set -x -e
+ export PATH=/opt/Embedix/tools/native-linux/bin/:/opt/Embedix/tools/bin:/opt/\Embedix/bin:/opt/Embedix/usr/bin:/opt/Embedix/usr/local/bin
+ PATH=/opt/Embedix/tools/native-linux/bin/:/opt/Embedix/tools/bin:/opt/Embedix\/bin:/opt/Embedix/usr/bin:/opt/Embedix/usr/local/bin
+ cd /bdk/lh79520/build/rpmdir/BUILD/glibc-2.2.1


1) I'm wondering if you ran out of disk space in the midst of building the tools. Can you check the log in /var/log/messages for any such messages?

2) Is your Fedora Core 3 installation up-to-date? (have you done 'yum update' to pick up any fixed Fedora RPMs) lately?

3) Can you mail me the *complete* log(everything starting from the ./install.sh) so I can see where they differ?

--
Peter Barada

Hello!

I did a "yum update", erased all the previous installation stuff, and finally re-did everything and the compilation finished sucessfully!

I wonder what yum update did to fix it really, I am curious, If you have any idea please let me know.

Now though, I have another problem. I saw people talking about this in several threads, but none of them really came to a solid resolution. When I try to boot from CFlash, I get the following error:

-----------------------------------------------
losh> mount fatfs /f
losh> cd /f
losh> load elf VMLINUX
loading from /f/VMLINUX:
R...........
elf file type : 0x0002
machine type: 0028 version: 1
prog start addr : 0x200c8000
num prog headers: 1
num sect headers: 10
offset : 0x00008000 disk length: 0x00157c73 mem len: 0x0017f39c
phyaddr: 0x200c8000 vaddr : 0x200c8000 dl addr: 0x200c8000
losh> exec root=/dev/hda1
kernel cmdline: 'root=/dev/hda1' at 200c0100
jumping to: 200c8000...
Linux version 2.4.17-rmk2-lineo5 (root@coop-desk03) (gcc version 2.95.3 20010315Processor: ARM ARM720T revision 3
Architecture: LogicPD LH79520 Card Engine
On node 0 totalpages: 8192
zone(0): 8192 pages.
zone(1): 0 pages.
zone(2): 0 pages.
Kernel command line: console=ttyAM1,115200 root=/dev/hda1
setup_timer(): timerLoad=32256
Console: colour dummy device 80x30
Calibrating delay loop... 25.65 BogoMIPS
Memory: 32MB = 32MB total
Memory: 14288KB available (1206K code, 250K data, 76K init)
Dentry-cache hash table entries: 4096 (order: 3, 32768 bytes)
Inode-cache hash table entries: 2048 (order: 2, 16384 bytes)
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
Buffer-cache hash table entries: 1024 (order: 0, 4096 bytes)
Page-cache hash table entries: 8192 (order: 3, 32768 bytes)
POSIX conformance testing by UNIFIX
Linux NET4.0 for Linux 2.4
Based upon Swansea University Computer Society NET3.039
Initializing RT netlink socket
Starting kswapd
ttyAM0 at MEM 0xfffc0000 (irq = 23) is a AMBA PrimeCell PL011
ttyAM1 at MEM 0xfffc1000 (irq = 24) is a AMBA PrimeCell PL011
ttyAM2 at MEM 0xfffc2000 (irq = 25) is a AMBA PrimeCell PL011
Console: switching to colour frame buffer device 30x40
LH79520 Watchdog Timer: timer margin 60 sec
block: 64 slots per queue, batch=16
RAMDISK driver initialized: 16 RAM disks of 8192K size 1024 blocksize
hd: no drives specified - use hd=cyl,head,sectors on kernel command line
hd?: check_status: status=0xd0 { Busy DriveReady SeekComplete }
hda: 61MB, CHS=978/4/32
Partition check:
hda:hda: check_status: status=0xd0 { Busy DriveReady SeekComplete }
hda: check_status: status=0xd0 { Busy DriveReady SeekComplete }
hda1 hda2
eth0: SMC91C11xFD(rev:1) at 0xf4800000 IRQ:4 DMA:0 (16-bit) MEM:8192b NOWAIT:1
eth0: ADDR: 00:08:ee:00:12:1f
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP
IP: routing cache hash table of 512 buckets, 4Kbytes
TCP: Hash tables configured (established 2048 bind 2048)
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
NetWinder Floating Point Emulator V0.95 (c) 1998-1999 Rebel.com
RAMDISK: Couldn't find valid RAM disk image starting at 0.
Freeing initrd memory: 16384K
EXT2-fs warning: mounting unchecked fs, running e2fsck is recommended
VFS: Mounted root (ext2 filesystem).
Freeing init memory: 76K
attempt to access beyond end of device
03:01: rw=0, want=1647259183, limit=43989
attempt to access beyond end of device
03:01: rw=0, want=1647259183, limit=43989
attempt to access beyond end of device
03:01: rw=0, want=6845282, limit=43989
attempt to access beyond end of device
03:01: rw=0, want=6845282, limit=43989
Kernel panic: No init found. Try passing init= option to kernel.
----------------------------------------------------

I tried the same thing for hda2 and including the x appended to it, it did not work. I have 2 Compact flashes, sandisk and kingston. It is the same behaviour on both. Any suggestions?

Thanks!

You command line has:

root=/dev/hda1

1) Is the root filesystem in the first partition, or is that where the kernel is?
2) What's the layout of your CF(i.e. how is it partitioned)?

Quote:
1) Is the root filesystem in the first partition, or is that where the kernel is?

I am assuming it is in the first partition, considering that I copied the rootfs under sda1 on the host.

Quote:
2) What's the layout of your CF(i.e. how is it partitioned)?

Following is the partition table...
---------------------------------------------------
[root@coop-desk03 tozyagcilar]# /sbin/fdisk /dev/sda

Command (m for help): p

Disk /dev/sda: 256 MB, 256901120 bytes
16 heads, 32 sectors/track, 980 cylinders
Units = cylinders of 512 * 512 = 262144 bytes

Device Boot Start End Blocks Id System
/dev/sda1 1 245 62704 83 Linux
/dev/sda2 * 246 980 188160 6 FAT16

Command (m for help):
-----------------------------------------------------

- I formatted sda1 with mke2fs, and sda2 with mkfs.msdos.
- cp -a did not give any errors at all, and I looked at the ownerships in the CF, they seemed fine.

Regards,

Talat.

>Any suggestions?

Hmm, not at the moment. perhaps you can try googling for some suggestions...

Sorry I can't be of more help at the moment...

Hello,

From the information I have found from other forums on the web, I suspect that the error message I am getting during the startup may be from an unsupported filesystem by the kernel. I am getting a couple error messages when doing "make -f Makefile.bdk deploy". These are as follows:
*********************
tar: ./sbin/e2fsck: Cannot link to `./sbin/fsck.ext3': No such file or directory
tar: ./sbin/fsck.ext2: Cannot link to `./sbin/fsck.ext3': No such file or directory
tar: Error exit delayed from previous errors
**********************

Could these be a reason why I am getting the error messages I specified in my previous postings (kernel startup errors)? How can I fix this if so?

Thanks.

Following is all the output:

bash# make -f Makefile.bdk deploy
sh scripts/extract.sh
Processing package kernel...
Processing package glibc...
Processing package elvis_tiny...
Processing package iproute2...
Processing package ext2fs...
tar: ./sbin/e2fsck: Cannot link to `./sbin/fsck.ext3': No such file or directory
tar: ./sbin/fsck.ext2: Cannot link to `./sbin/fsck.ext3': No such file or directory
tar: Error exit delayed from previous errors
Processing package libz...
Processing package libstdcpp...
Processing package ncurses...
Processing package netkit_ftp...
Processing package flex...
Processing package ash...
Processing package tinylogin...
Processing package libjpeg...
Processing package qte...
Processing package qpe...
Processing package konqembedded...
Processing package tcp_wrappers...
Processing package portmap...
execute chmod 755 ./etc/rc.d/init.d/S23portmap ; chown root.root ./etc/rc.d/init.d/S23portmap ;
Processing package Device_Files...
Processing package grep...
Processing package skellinux...
Processing package util_linux...
Processing package busybox...
execute chmod 644 ./var/log/wtmp ; chown root.utmp ./var/log/wtmp ;
Processing package default_passwd...
Processing package terminfo...
Processing package strace...
Processing package netkit_base...
Processing package netkit_telnet...
Processing package modutils...
Processing package blob...
Processing package dhcpcd...
Processing package madplay...
Processing package pwm_examples...
/bdk/lh79520/emb-bin/post-fs-build /bdk/lh79520
post-fs-build: device nodes successfully created in target root filesystem
/opt/Embedix/tools/bin/arm-linux-nm: /bdk/lh79520/src/linux/vmlinux: no symbols
post-fs-build: depmod successfully run on target root filesystem
post-fs-build: symlinks successfully created on target root filesystem
echo "root filesystem done..."
root filesystem done...
sh scripts/deploy.sh
scripts/deploy.sh: noinitrd: command not found
Welcome to the Sharp LH79520 deployment script
a few questions will be asked to help deploy the rootfs
May we suggest inquiring about our Embedix SDK and BSP for the LH79520
The SDK will make rebuilding and configuring the OS and associated tools
a whole lot easier...

First let's create the param file. The param file contains the kernel
commandline and tftp settings.

What is your Host Computer's IP (Enter for default: 172.27.150.60):
What is your Target Computer's IP (Enter for default: 172.27.150.100):
What is your Gateway (Enter for default: 172.27.255.254):
What is your Netmask (Enter for default: 255.255.0.0):
Tftpboot file (Enter for default: /tftpboot/zImage):

Here is the default commandline: console=ttyAM1,115200 noinitrd root=/dev/nfs nfsroot=172.27.150.60:/tftpboot/172.27.150.100 ip=172.27.150.100:172.27.150.60:172.27.255.254:255.255.0.0:lh79520::off

Enter a new commandline (Press EMTER to use default):
Save defaults...
Kernel Commandline is: console=ttyAM1,115200 noinitrd root=/dev/nfs nfsroot=172.27.150.60:/tftpboot/172.27.150.100 ip=172.27.150.100:172.27.150.60:172.27.255.254:255.255.0.0:lh79520::off

You have created a paramblock file for the BLOB boot loader.
We suggest you now setup the tftpd server and add links to the
the param and zImage files to the tftpboot directory

Would you like help on setting up a tftpd server(y/n)? n
Great now reboot the lh79520 with the following switch settings and you
are ready to to boot the kernel and rootfs

S1: off|on|on|off|off|off|offoff

NOTE: SW3 is for downloading a param file. SW2 is for NFS root, turn off both SW2 and SW3 to get a commandline for BLOB on the serial port. Serial port settings are 115200 8N1. Use minicom or kermit to connect to the lh79520 on DTE port 1

Talat,

the switch setting mentioned in your log file

S1: off|on|on|off|off|off|offoff

refers to an old SHARP Board (KEV79520)
which is - with regard to the hardware -
totally different from Logic´s boards.

Let me know if you need more help.

Gunter Wagschal
SHARP Microlecetronics Europe