<!-- retain these comments for translator revision tracking -->
<!-- $Id: arm.xml 69818 2015-04-17 15:47:40Z 93sam $ -->


  <sect2 arch="armhf;armel" id="boot-image-formats">
    <title>Boot image formats</title>
    <para>
      On ARM-based systems in most cases one of two formats for boot images
      is used: a) standard Linux zImage-format kernels
      (<quote>vmlinuz</quote>) in conjunction with standard Linux initial
      ramdisks (<quote>initrd.gz</quote>) or b) uImage-format kernels
      (<quote>uImage</quote>) in conjunction with corresponding initial
      ramdisks (<quote>uInitrd</quote>).
    </para>
    <para>
      uImage/uInitrd are image formats designed for the U-Boot firmware that
      is used on many ARM-based systems (mostly 32-bit ones). Older U-Boot versions can only
      boot files in uImage/uInitrd format, so these are often used on
      older armel systems. Newer U-Boot versions can - besides booting
      uImages/uInitrds - also boot standard Linux kernels and ramdisk images,
      but the command syntax to do that is slightly different from that
      for booting uImages.
    </para>
    <para>
      For systems using a multiplatform kernel, besides kernel and initial
      ramdisk a so-called device-tree file (or device-tree blob,
      <quote>dtb</quote>) is needed. It is specific to each supported system
      and contains a description of the particular hardware. The dtb
      should be supplied on the device by the firmware, but in practice a
      newer one often needs to be loaded.
    </para>
  </sect2>

  <sect2 arch="armhf" id="armhf-console-setup">
    <title>Console configuration</title>
    <para>
      The netboot tarball (<xref
      linkend="boot-armhf-netboot.tar.gz"/>), and the installer
      SD-card images (<xref linkend="boot-installer-sd-image"/>) use
      the (platform-specific) default console that is defined by
      U-Boot in the <quote>console</quote> variable.  In most cases
      that is a serial console, so on those platforms you by default
      need a serial console cable to use the installer.
    </para>
    <para>
      On platforms which also support a video console, you can modify the
      U-Boot <quote>console</quote> variable accordingly if you would like
      the installer to start on the video console.
    </para>
  </sect2>

  <sect2 arch="arm64" id="arm64-console-setup">
    <title>Console configuration</title>
    <para>
      The graphical installer is not enabled on the arm64 &d-i; images
      for jessie so the serial console is used. The console device
      should be detected automatically from the firmware, but if it is
      not then after you boot linux from the GRUB menu you will see a
      'Booting Linux' message then nothing more.
    </para>
    <para>
      If you hit this issue you will need to set a specific console
      config on the kernel command line. Hit <userinput>e</userinput>
      for 'Edit Kernel command-line' at the GRUB menu, and change
      <informalexample><screen>--- quiet</screen></informalexample> to
      <informalexample><screen>console=&lt;device&gt;,&lt;speed&gt;</screen></informalexample>
      e.g. <informalexample><screen>console=ttyAMA0,115200n8</screen></informalexample>
      When finished hit <keycombo><keycap>Control</keycap>
      <keycap>x</keycap></keycombo> to continue booting with new
      setting.
    </para>
  </sect2>

  <sect2 arch="arm64" id="juno-installation">
    <title>Juno Installation</title>
    <para>
      Juno has UEFI so the install is straightforward.  The most
      practical method is installing from USB stick. You need up to
      date firmware for USB-booting to work. Builds from <ulink
      url="&url-juno-firmware;">&url-juno-firmware;</ulink> after March
      2015 tested OK. Consult Juno documentation on firmware updating.
    </para>
    <para>
      Prepare a standard arm64 CD image on a USB stick. Insert it in
      one of the USB ports on the back. Plug a serial cable into the
      upper 9-pin serial port on the back. If you need networking
      (netboot image) plug the ethernet cable into the socket on the
      front of the machine.
    </para>
    <para>
      Run a serial console at 115200, 8bit no parity, and boot the
      Juno. It should boot from the USB stick to a GRUB menu.
      The console config is not correctly detected on Juno so just hitting
      return will show no kernel output. Set the console to  
<informalexample><screen>console=ttyAMA0,115200n8</screen></informalexample>
      as described in (<xref linkend="arm64-console-setup"/>). <keycombo><keycap>Control</keycap>
<keycap>x</keycap></keycombo> to boot should show you the &d-i; screens,
      and allow you to proceed with a standard installation.
    </para>
  </sect2>

  <sect2 arch="arm64" id="apm-installation">
    <title>Applied Micro Mustang Installation</title>
    <para>
      UEFI is available for this machine but it is normally shipped
      with U-Boot so you will need to either install UEFI firmware
      first then use standard boot/install methods, or use U-Boot boot
      methods. Also USB is not supported in the jessie kernel so
      installing from a USB stick does not work. You must use a serial
      console to control the installation because the graphical
      installer is not enabled on the arm64 architecture.
    </para>
    <para>
      The recommended install method is to copy the &d-i; kernel and
      initrd onto the hard drive, using the openembedded system
      supplied with the machine, then boot from that to run the
      installer. Alternatively use TFTP to get the kernel/dtb/initrd
      copied over and booted (<xref linkend="boot-tftp-uboot"/>). After
      installation, manual changes to boot from the installed image
      are needed.
    </para>
    <para>
      Run a serial console at 115200, 8bit no parity, and boot the
      machine. Reboot the machine and when you see "Hit any key to
      stop autoboot:" hit a key to get a Mustang# prompt. Then use
      U-Boot commands to load and boot the kernel, dtb and initrd.
     </para>
  </sect2>


  <sect2 arch="arm" id="boot-tftp"><title>Booting by TFTP</title>

&boot-installer-intro-net.xml;

    <sect3 arch="arm" id="boot-tftp-uboot">
      <title>TFTP-booting in U-Boot</title>
      <para>
        Network booting on systems using the U-Boot firmware consists of
        three steps: a) configuring the network, b) loading the images
        (kernel/initial ramdisk/dtb) into memory and c) actually executing
        the previosly loaded code.
      </para>
      <para>
        First you have to configure the network, either automatically via
        DHCP by running
<informalexample><screen>
setenv autoload no
dhcp
</screen></informalexample>
        or manually by setting several environment variables
<informalexample><screen>
setenv ipaddr &lt;ip address of the client&gt;
setenv netmask &lt;netmask&gt;
setenv serverip &lt;ip address of the tftp server&gt;
setenv dnsip &lt;ip address of the nameserver&gt;
setenv gatewayip &lt;ip address of the default gateway&gt;
</screen></informalexample>
        If you prefer, you can make these settings permanent by running
<informalexample><screen>
saveenv
</screen></informalexample>
      </para>
      <para>
        Afterwards you need to load the images (kernel/initial
        ramdisk/dtb) into memory. This is done with the tftpboot command,
        which has to be provided with the address at which the image
        shall be stored in memory. Unfortunately the memory map can vary
        from system to system, so there is no general rule which addresses
        can be used for this.                           
      </para>
      <para>
        On some systems, U-Boot predefines a set of environment variables
        with suitable load addresses: kernel_addr_r, ramdisk_addr_r and
        fdt_addr_r. You can check whether they are defined by running
<informalexample><screen>
printenv kernel_addr_r ramdisk_addr_r fdt_addr_r
</screen></informalexample>
        If they are not defined, you have to check your system's
        documentation for appropriate values and set them manually.  For
        systems based on Allwinner SunXi SOCs (e.g.  the Allwinner A10,
        architecture name <quote>sun4i</quote> or the Allwinner A20,
        architecture name <quote>sun7i</quote>), you can e.g.  use the
        follwing values:
<informalexample><screen>
setenv kernel_addr_r 0x46000000
setenv fdt_addr_r 0x47000000
setenv ramdisk_addr_r 0x48000000
</screen></informalexample>
      </para>
      <para>
        When the load addresses are defined, you can load the images
        into memory from the previously defined tftp server with
<informalexample><screen>
tftpboot ${kernel_addr_r} &lt;filename of the kernel image&gt;
tftpboot ${fdt_addr_r} &lt;filename of the dtb&gt;
tftpboot ${ramdisk_addr_r} &lt;filename of the initial ramdisk image&gt;
</screen></informalexample>
      </para>
      <para>
        The third part is setting the kernel commandline and actually
        executing the loaded code.  U-Boot passes the content of the
        <quote>bootargs</quote> environment variable as commandline to the
        kernel, so any parameters for the kernel and the installer - such as
        the console device (see <xref linkend="boot-console"/>) or
        preseeding options (see <xref linkend="installer-args"/> and <xref
        linkend="appendix-preseed"/>) - can be set with a command like
<informalexample><screen>
setenv bootargs console=ttyS0,115200 rootwait panic=10
</screen></informalexample>
        The exact command to execute the previously loaded code depends on
        the image format used.  With uImage/uInitrd, the command is
<informalexample><screen>
bootm ${kernel_addr_r} ${ramdisk_addr_r} ${fdt_addr_r}
</screen></informalexample>
       and with native Linux images it is
<informalexample><screen>
bootz ${kernel_addr_r} ${ramdisk_addr_r}:${filesize} ${fdt_addr_r}
</screen></informalexample>
      </para>
      <para>
        Note: When booting standard linux images, it is important to load
        the initial ramdisk image after the kernel and the dtb as U-Boot
        sets the filesize variable to the size of the last file loaded and
        the bootz command requires the size of the ramdisk image to work
        correctly.  In case of booting a platform-specific kernel, i.e.  a
        kernel without device-tree, simply omit the ${fdt_addr_r} parameter. 
        </para>
    </sect3>

    <sect3 arch="armhf" id="boot-armhf-netboot.tar.gz">
      <title>Pre-built netboot tarball</title>
      <para>
        &debian; provides a pre-built tarball (&armmp-netboot-tarball;)
        that can simply be unpacked on your tftp server and contains
        all files necessary for netbooting.  It also includes a boot
        script that automates all steps to load the installer.  Modern
        U-Boot versions contain a tftp autoboot feature that becomes
        active if there is no bootable local storage device (MMC/SD,
        USB, IDE/SATA/SCSI) and then loads this boot script from the
        tftp server.  Prerequisite for using this feature is that you
        have a dhcp server in your network which provides the client
        with the address of the tftp server.
      </para>
      <para>
        If you would like to trigger the tftp autoboot feature from the
        U-Boot commandline, you can use the follwing command:
        <informalexample><screen>run bootcmd_dhcp</screen></informalexample>
      </para>
      <para>
        To manually load the bootscript provided by the tarball, you can
        alternatively issue the following commands at the U-Boot prompt:

<informalexample><screen>
setenv autoload no
dhcp
tftpboot ${scriptaddr} /debian-installer/armhf/tftpboot.scr
source ${scriptaddr}
</screen></informalexample>

      </para>
    </sect3>
  </sect2> 


  <sect2 arch="arm64" condition="bootable-usb" id="usb-boot">
  <title>Booting from USB Memory Stick with UEFI</title>

&boot-installer-intro-usb.xml;

  </sect2>

  <sect2 arch="armel;armhf" id="boot-hd-media">
    <title>Booting from a USB stick in U-Boot</title>
    <para>
    
      Many modern U-Boot versions have USB support and allow booting from
      USB mass storage devices such as USB sticks.  Unfortunately the exact
      steps required to do that can vary quite a bit from device to device.
      
    </para>
    <para>
    
      U-Boot v2014.10 has introduced a common commandline handling and
      autoboot framework.  This allows building generic boot images that
      work on any system implementing this framework.  The &d-i; supports
      installation from a USB stick on such systems, but unfortunately not
      all platforms have adopted this new framework yet.

    </para>
    <para>
    
      To build a bootable USB stick for installing &debian;, unpack the
      hd-media tarball (see <xref linkend="where-files"/>) onto a
      USB stick formatted with a filesystem supported by the U-Boot version
      on your device.  For modern U-Boot versions, any of FAT16 / FAT32 /
      ext2 / ext3 / ext4 usually works. Then copy the ISO image file of the
      first &debian; installation CD or DVD onto the stick.

    </para>
    <para>
      The autoboot framework in modern U-Boot versions works similar to the
      boot ordering options in a PC BIOS, i.e.  it checks a list of possible
      boot devices for a valid boot image and starts the first one it finds. 
      If there is no operating system installed, plugging in the USB stick
      and powering up the system should result in starting the installer. 
      You can also initiate the USB-boot process any time from the U-Boot
      prompt by entering the <quote>run bootcmd_usb0</quote> command.
    </para>
    
    <para>
      One problem that can come up when booting from a USB stick while using
      a serial console can be a console baudrate mismatch.  If a console
      variable is defined in U-Boot, the &d-i; boot script automatically
      passes it to the kernel to set the primary console device and, if
      applicable, the console baudrate.  Unfortunately the handling of the
      console variable varies from platform to platform - on some platforms,
      the console variable includes the baudrate (as in
      <quote>console=ttyS0,115200</quote>), while on other platforms the
      console variable contains only the device (as in
      <quote>console=ttyS0</quote>).  The latter case leads to a garbled
      console output when the default baudrate differs between U-Boot and
      the kernel.  Modern U-Boot versions often use 115200 baud while the
      kernel still defaults to the traditional 9600 baud. If this happens,
      you should manually set the console variable to contain the correct
      baudrate for your system and then start the installer with the
      <quote>run bootcmd_usb0</quote> command.
    </para>
  </sect2>     

  <sect2 arch="armhf" id="boot-installer-sd-image">
    <title>Using pre-built SD-card images with the installer</title>
    <para>
      For a number of systems, Debian provides SD card images that contain
      both U-Boot and the &d-i;.  These images are provided in two variants
      - one for downloading the software packages over the network
      (available at &armmp-netboot-sd-img;) and one for offline
      installations using a Debian CD/DVD (available at
      &armmp-hd-media-sd-img;).  To save space and network bandwidth, the
      images consist of two parts - a system-dependent part named
      <quote>firmware.&lt;system-type&gt;.img.gz</quote>, and a
      system-independent part named <quote>partition.img.gz</quote>.
    </para>
    <para>
      To create a complete image from the two parts on Linux systems, you
      can use zcat as follows:
      
      <informalexample><screen>zcat firmware.&lt;system-type&gt;.img.gz partition.img.gz > complete_image.img</screen></informalexample>
            
      On Windows systems, you have to first decompress the two parts
      separately, which can be done e.g.  by using 7-Zip, and then
      concatenate the decompressed parts together by running the command
      
      <informalexample><screen>copy /b firmware.&lt;system-type&gt;.img + partition.img complete_image.img</screen></informalexample>
          
      in a Windows CMD.exe window.
    </para>
    <para>
      Write the resulting image onto an SD card, e.g. by running the
      following command on a Linux system:
      
      <informalexample><screen>cat complete_image.img > /dev/SD_CARD_DEVICE</screen></informalexample>

      After plugging the SD card into the target system and powering the
      system up, the installer is loaded from the SD card.  If you use the
      hd-media variant for offline installations, you must provide the
      installer with access to the first &debian; CD/DVD on a separate
      medium, which can e.g. be a CD/DVD ISO image on a USB stick.
    </para>
    <para>
      When you come to the partitioning step in the installer (see <xref
      linkend="di-partition"/>), you can delete or replace any previous
      partitions on the card.  Once the installer is started, it runs
      completely in the system's main memory and does not need to access the
      SD card anymore, so you can use the full card for installing &debian;. 
      The easiest way to create a proper partition layout on the SD card is
      to let the installer automatically create one for you (see
      <xref linkend="partman-auto"/>).
    </para>
  </sect2>
    
<!-- # None of the arm systems supported in Jessie is able to boot from
     # CD/DVD -> commenting out the "Booting from CD-ROM section" for arm
     
  <sect2 arch="arm"><title>Booting from CD-ROM</title>

&boot-installer-intro-cd.xml;

  </sect2>
-->

<!--

  <sect2 arch="arm" id="boot-firmware"><title>Booting from Firmware</title>

&boot-installer-intro-firmware.xml;

   <sect3 arch="arm" id="boot-firmware-ss4000e">
   <title>Booting the SS4000-E</title>
<para>

Due to limitations in the SS4000-E firmware, it unfortunately is not
possible to boot the installer without the use of a serial port at
this time. To boot the installer, you will need a serial nullmodem
cable; a computer with a serial port<footnote id="arm-s4ke-port">

<para>
A USB serial converter will also work.
</para>

</footnote>; and a ribbon cable with a male DB9 connector at one end,
and a 10-pin .1" IDC header at the other<footnote id="arm-s4k-rib">

<para>
This cable is often found in older desktop machines with builtin 9-pin
serial ports.
</para>

</footnote>.

</para><para>

To boot the SS4000-E, use your serial nullmodem cable and the ribbon
cable to connect to the serial port of the SS4000-E, and reboot the
machine. You need to use a serial terminal application to communicate
with the machine; a good option on a &debian; GNU/Linux is to use the
<command>cu</command> program, in the package of the same name. Assuming
the serial port on your computer is to be found on
<filename>/dev/ttyS0</filename>, use the following command line:

</para>

<informalexample><screen>
cu -lttyS0 -s115200
</screen></informalexample>

<para>

If using Windows, you may want to consider using the program
<classname>hyperterminal</classname>. Use a baud rate of 115200,
8 bits word length, no stop bits, and one parity bit.

</para><para>

When the machine boots, you will see the following line of output:

</para>

<informalexample><screen>
No network interfaces found

EM-7210 ver.T04 2005-12-12 (For ver.AA)
== Executing boot script in 1.000 seconds - enter ^C to abort
</screen></informalexample>

<para>

At this point, hit Control-C to interrupt the boot
loader<footnote id="arm-s4ke-sec">

<para>
Note that you have only one second to do so; if you miss this window,
just powercycle the machine and try again.
</para>

</footnote>. This will give you the RedBoot prompt. Enter the
following commands:

<informalexample><screen>
load -v -r -b 0x01800000 -m ymodem ramdisk.gz
load -v -r -b 0x01008000 -m ymodem zImage
exec -c "console=ttyS0,115200 rw root=/dev/ram mem=256M@0xa0000000" -r 0x01800000
</screen></informalexample>

</para><para>

After every <command>load</command> command, the system will expect a
file to be transmitted using the YMODEM protocol. When using cu, make
sure you have the package <classname>lrzsz</classname> installed, then
hit enter, followed by the <quote>~&lt;</quote> escape sequence to start
an external program, and run <command>sb initrd.gz</command> or
<command>sb vmlinuz</command>.

</para><para>

Alternatively, it is possible to load the kernel and ramdisk using
HTTP rather than YMODEM. This is faster, but requires a working HTTP
server on the network. To do so, first switch the bootloader to RAM mode:

<informalexample><screen>
fis load rammode
g
</screen></informalexample>

</para><para>

This will seemingly restart the machine; but in reality, it loads
redboot to RAM and restarts it from there. Not doing this step will cause
the system to hang in the necessary ip_address step that comes next.

</para><para>

You will need to hit Ctrl-C again to interrupt the boot. Then:

<informalexample><screen>
ip_address -l <replaceable>192.168.2.249</replaceable> -h <replaceable>192.168.2.4</replaceable>
load -v -r -b 0x01800000 -m http /initrd.gz
load -v -r -b 0x01008000 -m http /zImage
exec -c "console=ttyS0,115200 rw root=/dev/ram mem=256M@0xa0000000" -r 0x01800000
</screen></informalexample>

Where <replaceable>192.168.2.249</replaceable> is the IP address of the
installed system and <replaceable>192.168.2.4</replaceable> the IP address
of the HTTP server containing the kernel and ramdisk files.

</para><para>

The installer will now start as usual.

</para>
   </sect3>
  </sect2>

-->