Hawkboard

From eLinux.org
Revision as of 09:48, 19 December 2010 by Gaston (Talk | contribs)

Jump to: navigation, search

Hawk sticker1.JPG

Access Beginners Guide to Hawkboard if you are new to Development on Hawkboard http://elinux.org/Hawkboard/BeginnersGuide


Hardware Details

Hawkboard hw.jpg


Software Details

Tools & Softwares Downloads

Note:

1.Systems running "full" Linux, i.e., Linux on CPUs with an MMU. Use this to build both the Linux kernel and applications.

2.New Releases : http://www.codesourcery.com/sgpp/lite/arm/portal/release1039



Note:

1.This is for RTOS systems or "bare metal" systems where no operating system is present. These toolchains should not be used to build Linux kernels or Linux applications.

2.New Releases : http://www.codesourcery.com/sgpp/lite/arm/portal/release1033


Tools From Other Vendors

JTAG emulators

Misc Tools

Sources

Sample Rootfs

1.Ubuntu Jaunty http://hawktool.googlecode.com/files/RootFS_v1.tar.bz2 (username/password =hawk/password)

2.Fedora RootFS -http://ftp.linux.org.uk/pub/linux/arm/fedora/rootfs/rootfs-f12.tar.bz2 (username/password =root/fedoraarm)

3.Impactlinux http://impactlinux.com/fwl/downloads/binaries/root-filesystem-armv5l.tar.bz2


Getting Started

1.Powering the Device: Hawkboard can be powered through USB OTG port(mini USB) or Separate 5V Source. Since the USB port of a PC/laptop or hub are often limited to 500 mA, it is advisable to use a separate power supply of 5VDC that supplies at least 1 A of current.

Note: It is recommended to use a separate 5VDC Power supply with at least a 1 Amps current rating instead of USB Power for Normal Operation of the Board

2.Connecting VGA Monitor: Once the power supply has been attached, Hawkboard boots to Uboot Bootloader and will show Hawkboard Logo on Screen.

3.Connecting UART: To transfer images to Hawkboard and to set other parameters, the UART/serial cable needs to be attached to Hawkboard and PC. A NULL Modem(Crossed Cable i.e 2 and 3 Crossed) is needed for that. Run a terminal session (such as Minicom on Linux or TeraTerm on Windows) on the Host PC and configure it to connect to that serial port with the following characteristics:

Bits per Second: 115200
Data Bits: 8
Parity: None
Stop Bits: 1
Flow Control: None
Transmit delay: 0 msec/char, 10 msec/line

Hawk Serial Set.JPG

After powering the device you should be able to see the following:

Hawk uboot msg.JPG Hawk vga logo.jpg


  • On Hyperterminal (Any other Serial Console) (Left) On VGA Monitor you will see HawkBoard Logo (Right)


Command prompts in this guide

In this guide, commands are preceded by prompts that indicate the environment where the command is to be typed. For example:

  • host$
Indicates command to be typed into the shell window of the host Linux workstation.
  • Hawkboard.org>
Indicates commands to be typed into the U-Boot shell in a console window connected to the Hawkboard's serial port.
  • target$
Indicates commands to be typed into the Linux shell in the terminal window connected to the Hawkboard's serial port or TTY..

Building & Booting Uboot images from source

HawkBoard comes with pre-installed U-boot and normally you don't want to Compile and Load U-boot unless you have Bricked the Bootloader or want to change something.So You can Skip this Section and can go Directly to Compiling Linux Kernel or Booting Kernel Section.

Compiling u-boot (bootloader)

  1. Issue compile commands with make:
  host$ make ARCH=arm CROSS_COMPILE=arm-none-linux-gnueabi- distclean
  host$ make ARCH=arm CROSS_COMPILE=arm-none-linux-gnueabi- omapl_hawkboard_config 
  host$ make ARCH=arm CROSS_COMPILE=arm-none-linux-gnueabi-

If you get an error while compiling, see the FAQor the mailing list thread

Signing u-boot for UART boot

  1. Copy u-boot (elf) file to the host machine where AIS Generator is installed. Follow the below steps
  2. AIS gen uart1.JPG
  3. AIS gen uart2.JPG
  4. AIS gen uart3.JPG
  5. AIS gen uart4.JPG

Signing u-boot for NAND boot

  1. Copy u-boot (elf) file to the host machine where AIS Generator is installed. Follow the below steps
  2. AIS gen nand1.JPG
  3. Configure PLL0, PLL1, DDR tabs as shown above. The File name to generate remains as in the step above.

Booting

Booting u-boot over UART

  • Power OFF the board
  • Close any previously running terminal programs
  • Connect the UART cable(Null/Cross Cable) to the Host machine.

Following process used on Windows family of Operating System:

  • User can use mono utility with dvflasher.ext program to load u-boot on Linux system.
  • Details To be updated

Following process used on Windows family of Operating System:

  • Configure the Boot Switches as 1-OFF 2-ON 3-OFF 4-ON
  • Start the UART Host Utility, should have been installed with AIS Generator / UART Host Tool
  • Select the AIS FILE Generated for UART

UART HOST BOOT.JPG

  • Click on the Start button
  • Power UP the Device
  • If you see errors click on stop and press start again and give a board reset.
  • Wait till you all the below messages and COM is closed:
  (File IO): Read 414964 bytes from file D:\hawk_release\u-boot_uart_ais.bin.
  (Serial Port): Opening COM1 at 115200 baud...
  (AIS Parse): Read magic word 0x41504954.
  (AIS Parse): Waiting for BOOTME...
  (AIS Parse): Performing Start-Word Sync...
  (AIS Parse): Performing Ping Opcode Sync...
  (AIS Parse): Processing command 0: 0x5853590D.
  (AIS Parse): Performing Opcode Sync...
  (AIS Parse): Executing function...
  (AIS Parse): Processing command 1: 0x5853590D.
  (AIS Parse): Loaded 1512-byte section to address 0xC10E4BEC.
  ...
  ...
  ...
  (AIS Parse): Processing command 15: 0x58535906.
  (AIS Parse): Performing Opcode Sync...
  (AIS Parse): Performing jump and close...
  (AIS Parse): AIS complete. Jump to address 0xC1080000.
  (AIS Parse): Waiting for DONE...
  (AIS Parse): Boot completed successfully.
  (Serial Port): Closing COM1.
  • Now Start any standard UART Terminal and hit enter key, should see the u-boot prompt

Procedure to flash u-boot on NAND

After booting the u-boot over UART as mentioned above,

  • On the u-boot prompt in the terminal window
  • Configure the Ethernet server and Client IP addresses, For e.g.
  hawkboard.org > setenv serverip 172.24.156.199
  hawkboard.org > setenv ipaddr 172.24.190.58
  • Download the u-boot generated for NAND
  hawkboard.org > tftpboot 0xc0700000 u-boot_nand_ais.bin
  
  TFTP from server 172.24.156.199; our IP address is 172.24.190.58
  Filename 'u-boot_nand_ais.bin'.
  Load address: 0xc0700000
  Loading: #################################################################
           #################
  done
  Bytes transferred = 414988 (6550c hex)  
  • Erase NAND Flash
  hawkboard.org > nand erase

  NAND erase: device 0 whole chip
  OK
  • Flash the NAND with u-boot
  hawkboard.org > nand write.e 0xc0700000 0x20000 0x70000
  NAND write: device 0 offset 0x20000, size 0x70000
  458752 bytes written: OK
  hawkboard.org >
  • Switch off the board
  • Change the DIP Switches for NAND boot 1-ON 2-OFF 3-OFF 4-OFF

Booting u-boot over NAND

Flash u-boot into NAND as mentioned above

  • Switch off the board
  • Change the DIP Switches for NAND boot 1-ON 2-OFF 3-OFF 4-OFF
  • Switch on the board

Building & Booting Kernel images from source

Compiling Linux Kernel

  1. Issue compile commands with make:
  make ARCH=arm CROSS_COMPILE=arm-none-linux-gnueabi- distclean
  make ARCH=arm CROSS_COMPILE=arm-none-linux-gnueabi- omapl138_hawkboard_defconfig
  make ARCH=arm CROSS_COMPILE=arm-none-linux-gnueabi- uImage

If you get the "mkimage" command not found - U-Boot images will not be built error,build uboot from the source first and then copy the mkimage binary from ~u-boot-omapl1/tools to a directory in your PATH

Booting

Booting Linux Kernel and Mounting RAMDISK

You should find a kernel and a ramdisk image if you haven't found already. You can download samples from [1] (uImage_v1 & ramdisk_v1.gz)). Then copy these files to your "/tftpboot" directory assuming you will use tftp for transferring files to your board. In order to do so you need to run a tftp server on your system. Under linux a good choice is atfptd, but you can also use others. Typically a linux tftp server exports the directory /tftpboot. Make sure before you continue that your ftp server is running.
At the u-boot prompt:

Set ethernet connection:

$setenv serverip <PC ipaddress>
$setenv ipaddr <board ipaddress>

<PC ipaddress> should be the address of the PC running the tftp server.
<board ipaaddress> should be an address that is not in use on the same subnet as the tftp server.
Transfer files to the board:

$tftp c0700000 uImage_v1
$tftp c1180000 ramdisk_v1.gz

If you broke your bootargs previously then:

$setenv bootargs "mem=128M console=ttyS2,115200n8 root=/dev/ram0 rw initrd=0xc1180000,4M"

And finally boot the images:

$bootm c0700000

The TFTP transferred image can be written to the NAND Flash for future use to perform kernel upgrades and enable booting without network connection.

To do this erase the section of the NAND containing the kernel and write the kernel image from RAM to the NAND.

$nand erase 200000 200000
$nand write.e 0xc0700000 0x200000 0x200000

The written image can be read back using

$nand read.e 0xc0700000 0x200000 0x200000

Booting Linux Kernel and Mounting NFS (Network File System)

Load the Kernel through NAND Flash or TFTP Method ,Once kernel has been loaded to RAM using tftp c0700000 uImage_v1 or similar Next. Follow following steps to Mount RootFS through NFS.(Ubuntu)

Setting NFS Server

1. Install NFS Server

host$ sudo apt-get install portmap nfs-kernel-server

2. Share the Folder containing extracted RootFS (e.g /nfsroot contains RootFS)

Edit /etc/exports and add the shares:

/nfsroot <NETWROK_IP><SUBNET_MASK>(rw,sync,no_subtree_check,no_root_squash)

e.g, /nfsroot 192.168.0.0/255.255.255.0(rw,sync,no_subtree_check,no_root_squash)

Assuming your network is is 192.168.0.0

NB:- no_root_squash might be necessary if you are getting sudo erros in Ubuntu Rootfs

3.After setting up /etc/exports, export the shares:

 host$ sudo exportfs -ra

restart the nfs-server if required

 host$ sudo /etc/init.d/nfs-kernel-server restart
Booting Board using NFS

Under Uboot prompt once the Kernel has been loaded through

 hawkboard.org> tftp c0700000 uImage_v1

1.Set bootargs

Set boot arguments:

a)If Board uses Static IP


hawkboard.org> setenv bootargs 'console=ttyS2,115200n8 noinitrd rw ip=<HawkboardIP>:<server-ip>:<gateway-ip>:<netmask>::eth0: root=/dev/nfs nfsroot=<NFS Server ipaddress>:/nfsroot'

e.g

 hawkboard.org> setenv bootargs 'console=ttyS2,115200n8 noinitrd rw ip=192.168.0.125:192.168.0.1:192.168.0.1:255.255.255.0::eth0: root=/dev/nfs nfsroot=<NFS Server ipaddress>:/nfsroot'

where 192.168.0.125 is board IP & 192.168.0.1 is gateway IP& Router IP.


b)DHCP IP Board

 hawkboard.org> setenv bootargs "mem=128M console=ttyS2,115200n8 root=/dev/nfs nfsroot=<Server ipaddress>:/nfsroot ip=dhcp"


2.And finally boot the image:

 hawkboard.org> bootm c0700000

Booting Linux Kernel from USB

This steps work for USB Pendrive connected to Hub or Directly to USB Standard Port and rootfs is Ubuntu jaunty made through rootstock.Should work for other distribution too.

1. Load uimage through tftp ( as usual)

setenv serverip <PC IP>
setenv ipaddr <BOARD IP>
tftp c0700000 uImage_v1

2.Sent environment variable

setenv bootargs console=ttyS2,115200n8 noinitrd root=/dev/sda1 rootwait rw init=/sbin/init

3.bootm

bootm c0700000

Here /dev/sda1 is USB Drive with only one ext2 partition and rootfs lying init.

Booting Linux Kernel from SATA

Currently the provided u-boot does not allow booting from either SATA. However, it is possible to boot from nand and have the root filesystem loaded from SATA.
If you want to do so, you have to figure out the right partition for the root filesystem and add that to your bootargs.

E.g. in u-boot say something like:

setenv bootargs mem=128M console=ttyS2,115200n8 root=/dev/sda1 rootwait

Where you replace /dev/sda1 with the name of the device that contains your root filesystem. /dev/sda1 is a good name for sata partition 1 if no usb or sd devices are present. Note that the rootwait argument is needed. It tells the kernel to wait until the disk has settled.

Booting Linux Kernel from SD OR MMC

Currently the provided u-boot does not allow direct booting from either SD or MMC However, it is possible to boot from nand and have the root filesystem loaded from SD
If you want to do so, you have to figure out the right partition for the root filesystem and add that to your bootargs.


E.g. in u-boot say something like:

setenv bootargs console=ttyS2,115200n8 console=tty1 noinitrd root=/dev/mmcblk0p1 rootwait rw

or

setenv bootargs console=ttyS2,115200n8 console=tty1 noinitrd root=/dev/mmcblk0p1 rootdelay = 2 rootfstype = ext2 rw

Here p1 in mmcblk0p1 says that Rootfs is present in first partition of SD Card.

MISC

1.If you find screen broken, do this before bootm in u-boot, ideally this should go into kernel code

mw.l 0x01c14110 0x44442222 1;mw.l 0x01c14114 0x44400000 1;mw.l 0x01c14118 0x04604404 1;

2.For booting android over MMC the bootargs should be

setenv bootargs mem=128M console=ttyS2,115200n8 noinitrd root=/dev/mmcblk0p1 rootwait ip=off init=/init androidboot.console=ttyS2

3.You probably want to avoid that you have to retype these commands every time you want to boot. This can be achieved easily by issuing the following commands on the u-boot prompt:

setenv serverip <PC ipaddress>
setenv ipaddr <board ipaddress>
setenv bootargs_nfs mem=128M console=ttyS2,115200n8 root=/dev/nfs nfsroot=<PC ipaddress>:/nfsroot ip=dhcp
setenv bootcmd 'setenv bootargs $bootargs_nfs;tftp c0700000 uImage.v1; bootm c0700000'
saveenv

Make sure to use single quotes in the last setenv command.


Custom RootFS

1.Fedora http://fedoraproject.org/wiki/Architectures/ARM/RfsBuild

2.http://arago-project.org

3.Ubuntu Use Rootstock

4.Angstrom uses Openembedded http://www.angstrom-distribution.org/narcissus/


Making Use of the C6740 DSP Core

The C6740 DSP core inside the OMAP-L138 processor is extremely capable, providing fixed-point and full double-precision floating-point operations. Probably the simplest way to get started with the C6740 DSP core on the Hawkboard is the C6Run project. It provides a GCC-like front end for building applications, which, when executed from the Linux command prompt, run transparently on the DSP. See the C6Run main page on the TI Embedded Processor wiki for more details and to get started.


FAQs

For HawkBoard frequently asked questions (FAQ) see community FAQ.


Common Issues

At this stage it appears that the latest (late 2010) batch of boards have serious problems many problems have been discussed on the mailing list regarding failure to boot linux and problems with tftp There has been a hardware fix posted on www.innovatesolutions.net. Summary of the fix is either to send the faulty board to distributor who will send them on to innovate alternatively, the fix can also be done by customers without voiding guarantees. The fix is made up by removing two ferrite beads and shortening the pads afterwards. Hawkboard as deliveredHawkboard as delivered

Projects Based on HawkBoard

Live Links


Guide Links


Subpages

Updated List is here http://elinux.org/Category:Hawkboard <splist


/>