docs/multiboot.texi

\input texinfo @c -*-texinfo-*-
@c -*-texinfo-*-
@c %**start of header
@setfilename multiboot.info
@settitle Multiboot Specification
@c %**end of header

@c Unify all our little indices for now.
@syncodeindex fn cp
@syncodeindex vr cp
@syncodeindex ky cp
@syncodeindex pg cp
@syncodeindex tp cp

@footnotestyle separate
@paragraphindent 3
@finalout


@dircategory Kernel
@direntry
* Multiboot Specification: (multiboot).		Multiboot Specification.
@end direntry

@ifinfo
Copyright @copyright{} 1995, 96 Bryan Ford <baford@@cs.utah.edu>
Copyright @copyright{} 1995, 96 Erich Stefan Boleyn <erich@@uruk.org>
Copyright @copyright{} 1999, 2000, 2001, 2002 Free Software Foundation, Inc.

Permission is granted to make and distribute verbatim copies of
this manual provided the copyright notice and this permission notice
are preserved on all copies.

@ignore
Permission is granted to process this file through TeX and print the
results, provided the printed document carries a copying permission
notice identical to this one except for the removal of this paragraph
(this paragraph not being relevant to the printed manual).

@end ignore

Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided also that
the entire resulting derived work is distributed under the terms of a
permission notice identical to this one.

Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions.
@end ifinfo

@titlepage
@sp 10
@title The Multiboot Specification
@author Yoshinori K. Okuji, Bryan Ford, Erich Stefan Boleyn, Kunihiro Ishiguro
@page

@vskip 0pt plus 1filll
Copyright @copyright{} 1995, 96 Bryan Ford <baford@@cs.utah.edu>
Copyright @copyright{} 1995, 96 Erich Stefan Boleyn <erich@@uruk.org>
Copyright @copyright{} 1999, 2000, 2001, 2002 Free Software Foundation, Inc.

Permission is granted to make and distribute verbatim copies of
this manual provided the copyright notice and this permission notice
are preserved on all copies.

Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided also that
the entire resulting derived work is distributed under the terms of a
permission notice identical to this one.

Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions.
@end titlepage

@finalout
@headings double

@ifnottex
@node Top
@top Multiboot Specification

This file documents Multiboot Specification, the proposal for the boot
sequence standard. This edition documents version 0.6.93.
@end ifnottex

@menu
* Overview::                    
* Terminology::                 
* Specification::               
* Examples::                    
* History::                     
* Index::                       
@end menu


@node Overview
@chapter Introduction to Multiboot Specification

This chapter describes some rough information on the Multiboot
Specification. Note that this is not a part of the specification itself.

@menu
* Motivation::                  
* Architecture::                
* Operating systems::           
* Boot sources::                
* Boot-time configuration::     
* Convenience to operating systems::  
* Boot modules::                
@end menu


@node Motivation
@section The background of Multiboot Specification

Every operating system ever created tends to have its own boot loader.
Installing a new operating system on a machine generally involves
installing a whole new set of boot mechanisms, each with completely
different install-time and boot-time user interfaces. Getting multiple
operating systems to coexist reliably on one machine through typical
@dfn{chaining} mechanisms can be a nightmare. There is little or no
choice of boot loaders for a particular operating system --- if the one
that comes with the operating system doesn't do exactly what you want,
or doesn't work on your machine, you're screwed.

While we may not be able to fix this problem in existing commercial
operating systems, it shouldn't be too difficult for a few people in the
free operating system communities to put their heads together and solve
this problem for the popular free operating systems. That's what this
specification aims for. Basically, it specifies an interface between a
boot loader and a operating system, such that any complying boot loader
should be able to load any complying operating system. This
specification does @emph{not} specify how boot loaders should work ---
only how they must interface with the operating system being loaded.


@node Architecture
@section The target architecture

This specification is primarily targeted at @sc{pc}, since they are the
most common and have the largest variety of operating systems and boot
loaders. However, to the extent that certain other architectures may
need a boot specification and do not have one already, a variation of
this specification, stripped of the x86-specific details, could be
adopted for them as well.


@node Operating systems
@section The target operating systems

This specification is targeted toward free 32-bit operating systems
that can be fairly easily modified to support the specification without
going through lots of bureaucratic rigmarole. The particular free
operating systems that this specification is being primarily designed
for are Linux, FreeBSD, NetBSD, Mach, and VSTa. It is hoped that other
emerging free operating systems will adopt it from the start, and thus
immediately be able to take advantage of existing boot loaders. It would
be nice if commercial operating system vendors eventually adopted this
specification as well, but that's probably a pipe dream.


@node Boot sources
@section Boot sources

It should be possible to write compliant boot loaders that load the OS
image from a variety of sources, including floppy disk, hard disk, and
across a network.

Disk-based boot loaders may use a variety of techniques to find the
relevant OS image and boot module data on disk, such as by
interpretation of specific file systems (e.g. the BSD/Mach boot loader),
using precalculated @dfn{block lists} (e.g. LILO), loading from a
special @dfn{boot partition} (e.g. OS/2), or even loading from within
another operating system (e.g. the VSTa boot code, which loads from
DOS). Similarly, network-based boot loaders could use a variety of
network hardware and protocols.

It is hoped that boot loaders will be created that support multiple
loading mechanisms, increasing their portability, robustness, and
user-friendliness.


@node Boot-time configuration
@section Configure an operating system at boot-time

It is often necessary for one reason or another for the user to be able
to provide some configuration information to an operating system
dynamically at boot time. While this specification should not dictate
how this configuration information is obtained by the boot loader, it
should provide a standard means for the boot loader to pass such
information to the operating system.


@node Convenience to operating systems
@section How to make OS development easier

OS images should be easy to generate. Ideally, an OS image should simply
be an ordinary 32-bit executable file in whatever file format the
operating system normally uses. It should be possible to @code{nm} or
disassemble OS images just like normal executables. Specialized tools
should not be required to create OS images in a @emph{special} file
format. If this means shifting some work from the operating system to
a boot loader, that is probably appropriate, because all the memory
consumed by the boot loader will typically be made available again after
the boot process is created, whereas every bit of code in the OS image
typically has to remain in memory forever. The operating system should
not have to worry about getting into 32-bit mode initially, because mode
switching code generally needs to be in the boot loader anyway in order
to load operating system data above the 1MB boundary, and forcing the
operating system to do this makes creation of OS images much more
difficult.

Unfortunately, there is a horrendous variety of executable file formats
even among free Unix-like @sc{pc}-based operating systems --- generally
a different format for each operating system. Most of the relevant free
operating systems use some variant of a.out format, but some are moving
to @sc{elf}. It is highly desirable for boot loaders not to have to be
able to interpret all the different types of executable file formats in
existence in order to load the OS image --- otherwise the boot loader
effectively becomes operating system specific again.

This specification adopts a compromise solution to this
problem. Multiboot-compliant OS images always contain a magic
@dfn{Multiboot header} (@pxref{OS image format}), which allows the boot
loader to load the image without having to understand numerous a.out
variants or other executable formats. This magic header does not need to
be at the very beginning of the executable file, so kernel images can
still conform to the local a.out format variant in addition to being
Multiboot-compliant.


@node Boot modules
@section Boot modules

Many modern operating system kernels, such as those of VSTa and Mach, do
not by themselves contain enough mechanism to get the system fully
operational: they require the presence of additional software modules at
boot time in order to access devices, mount file systems, etc. While
these additional modules could be embedded in the main OS image along
with the kernel itself, and the resulting image be split apart manually
by the operating system when it receives control, it is often more
flexible, more space-efficient, and more convenient to the operating
system and user if the boot loader can load these additional modules
independently in the first place.

Thus, this specification should provide a standard method for a boot
loader to indicate to the operating system what auxiliary boot modules
were loaded, and where they can be found. Boot loaders don't have to
support multiple boot modules, but they are strongly encouraged to,
because some operating systems will be unable to boot without them.


@node Terminology
@chapter The definitions of terms used through the specification

@table @dfn
@item must
We use the term @dfn{must}, when any boot loader or OS image needs to
follow a rule --- otherwise, the boot loader or OS image is @emph{not}
Multiboot-compliant.

@item should
We use the term @dfn{should}, when any boot loader or OS image is
recommended to follow a rule, but it doesn't need to follow the rule.

@item may
We use the term @dfn{may}, when any boot loader or OS image is allowed
to follow a rule.

@item boot loader
Whatever program or set of programs loads the image of the final
operating system to be run on the machine. The boot loader may itself
consist of several stages, but that is an implementation detail not
relevant to this specification. Only the @emph{final} stage of the boot
loader --- the stage that eventually transfers control to an operating
system --- must follow the rules specified in this document in order
to be @dfn{Multiboot-compliant}; earlier boot loader stages may be
designed in whatever way is most convenient.

@item OS image
The initial binary image that a boot loader loads into memory and
transfers control to start an operating system. The OS image is
typically an executable containing the operating system kernel.

@item boot module
Other auxiliary files that a boot loader loads into memory along with
an OS image, but does not interpret in any way other than passing their
locations to the operating system when it is invoked.

@item Multiboot-compliant
A boot loader or an OS image which follows the rules defined as
@dfn{must} is Multiboot-compliant. When this specification specifies a
rule as @dfn{should} or @dfn{may}, a Multiboot-complaint boot loader/OS
image doesn't need to follow the rule.

@item u8
The type of unsigned 8-bit data.

@item u16
The type of unsigned 16-bit data. Because the target architecture is
little-endian, u16 is coded in little-endian.

@item u32
The type of unsigned 32-bit data. Because the target architecture is
little-endian, u32 is coded in little-endian.

@item u64
The type of unsigned 64-bit data. Because the target architecture is
little-endian, u64 is coded in little-endian.
@end table


@node Specification
@chapter The exact definitions of Multiboot Specification

There are three main aspects of a boot loader/OS image interface:

@enumerate
@item
The format of an OS image as seen by a boot loader.

@item
The state of a machine when a boot loader starts an operating
system.

@item
The format of information passed by a boot loader to an operating
system.
@end enumerate

@menu
* OS image format::             
* Machine state::               
* Boot information format::     
@end menu


@node OS image format
@section OS image format

An OS image may be an ordinary 32-bit executable file in the standard
format for that particular operating system, except that it may be
linked at a non-default load address to avoid loading on top of the
@sc{pc}'s I/O region or other reserved areas, and of course it should
not use shared libraries or other fancy features.

An OS image must contain an additional header called @dfn{Multiboot
header}, besides the headers of the format used by the OS image. The
Multiboot header must be contained completely within the first 8192
bytes of the OS image, and must be longword (32-bit) aligned. In
general, it should come @emph{as early as possible}, and may be
embedded in the beginning of the text segment after the @emph{real}
executable header.

@menu
* Header layout::               The layout of Multiboot header
* Header magic fields::         The magic fields of Multiboot header
* Header address fields::       
* Header graphics fields::      
@end menu


@node Header layout
@subsection The layout of Multiboot header

The layout of the Multiboot header must be as follows:

@multitable @columnfractions .1 .1 .2 .5
@item Offset @tab Type  @tab Field Name    @tab Note
@item 0      @tab u32 @tab magic         @tab required
@item 4      @tab u32 @tab flags         @tab required
@item 8      @tab u32 @tab checksum      @tab required
@item 12     @tab u32 @tab header_addr   @tab if flags[16] is set
@item 16     @tab u32 @tab load_addr     @tab if flags[16] is set
@item 20     @tab u32 @tab load_end_addr @tab if flags[16] is set
@item 24     @tab u32 @tab bss_end_addr  @tab if flags[16] is set
@item 28     @tab u32 @tab entry_addr    @tab if flags[16] is set
@item 32     @tab u32 @tab mode_type     @tab if flags[2] is set
@item 36     @tab u32 @tab width         @tab if flags[2] is set
@item 40     @tab u32 @tab height        @tab if flags[2] is set
@item 44     @tab u32 @tab depth         @tab if flags[2] is set
@end multitable

The fields @samp{magic}, @samp{flags} and @samp{checksum} are defined in
@ref{Header magic fields}, the fields @samp{header_addr},
@samp{load_addr}, @samp{load_end_addr}, @samp{bss_end_addr} and
@samp{entry_addr} are defined in @ref{Header address fields}, and the
fields @samp{mode_type}, @samp{width}, @samp{height} and @samp{depth} are
defind in @ref{Header graphics fields}.


@node Header magic fields
@subsection The magic fields of Multiboot header

@table @samp
@item magic
The field @samp{magic} is the magic number identifying the header,
which must be the hexadecimal value @code{0x1BADB002}.

@item flags
The field @samp{flags} specifies features that the OS image requests or
requires of an boot loader. Bits 0-15 indicate requirements; if the
boot loader sees any of these bits set but doesn't understand the flag
or can't fulfill the requirements it indicates for some reason, it must
notify the user and fail to load the OS image. Bits 16-31 indicate
optional features; if any bits in this range are set but the boot loader
doesn't understand them, it may simply ignore them and proceed as
usual. Naturally, all as-yet-undefined bits in the @samp{flags} word
must be set to zero in OS images. This way, the @samp{flags} fields
serves for version control as well as simple feature selection.

If bit 0 in the @samp{flags} word is set, then all boot modules loaded
along with the operating system must be aligned on page (4KB)
boundaries. Some operating systems expect to be able to map the pages
containing boot modules directly into a paged address space during
startup, and thus need the boot modules to be page-aligned.

If bit 1 in the @samp{flags} word is set, then information on available
memory via at least the @samp{mem_*} fields of the Multiboot information
structure (@pxref{Boot information format}) must be included. If the
boot loader is capable of passing a memory map (the @samp{mmap_*} fields)
and one exists, then it may be included as well.

If bit 2 in the @samp{flags} word is set, information about the video
mode table (@pxref{Boot information format}) must be available to the
kernel.

If bit 16 in the @samp{flags} word is set, then the fields at offsets
8-24 in the Multiboot header are valid, and the boot loader should use
them instead of the fields in the actual executable header to calculate
where to load the OS image. This information does not need to be
provided if the kernel image is in @sc{elf} format, but it @emph{must}
be provided if the images is in a.out format or in some other
format. Compliant boot loaders must be able to load images that either
are in @sc{elf} format or contain the load address information embedded
in the Multiboot header; they may also directly support other executable
formats, such as particular a.out variants, but are not required to.

@item checksum
The field @samp{checksum} is a 32-bit unsigned value which, when added
to the other magic fields (i.e. @samp{magic} and @samp{flags}), must
have a 32-bit unsigned sum of zero.
@end table


@node Header address fields
@subsection The address fields of Multiboot header

All of the address fields enabled by flag bit 16 are physical addresses.
The meaning of each is as follows:

@table @code
@item header_addr
Contains the address corresponding to the beginning of the Multiboot
header --- the physical memory location at which the magic value is
supposed to be loaded. This field serves to @dfn{synchronize} the
mapping between OS image offsets and physical memory addresses.

@item load_addr
Contains the physical address of the beginning of the text segment. The
offset in the OS image file at which to start loading is defined by the
offset at which the header was found, minus (header_addr -
load_addr). load_addr must be less than or equal to header_addr.

@item load_end_addr
Contains the physical address of the end of the data
segment. (load_end_addr - load_addr) specifies how much data to load.
This implies that the text and data segments must be consecutive in the
OS image; this is true for existing a.out executable formats.
If this field is zero, the boot loader assumes that the text and data
segments occupy the whole OS image file.

@item bss_end_addr
Contains the physical address of the end of the bss segment. The boot
loader initializes this area to zero, and reserves the memory it
occupies to avoid placing boot modules and other data relevant to the
operating system in that area. If this field is zero, the boot loader
assumes that no bss segment is present.

@item entry_addr
The physical address to which the boot loader should jump in order to
start running the operating system.
@end table


@node Header graphics fields
@subsection The graphics fields of Multiboot header

Also, information about a preferred default mode can be specified in the
@samp{mode_type}, @samp{width}, @samp{height} and @samp{depth}
fields. This is only a @emph{recommended} mode by the kernel. If the
mode exists, the boot loader should set it if the user hasn't specified
a mode. If not, it should fall back to a similar mode, if available.

Valid numbers for @samp{mode_type} is 0 for linear graphics mode and 1
for EGA-standard text mode. Everything else is reserved for future
expansion. Please note that even if you set this field to indicate that
you want a graphics mode, you might get a text mode.

@samp{width} and @samp{height} is specified in pixels, if graphics mode,
or characters in EGA text mode. @samp{depth} is given in bits per pixel
for graphics, or zero for EGA text modes.

Any, some or all of those three fields may be set to zero, indicating to
the boot loader that no preference is given. This makes it possible for
a kernel to just say it wants a 32-bit resolution, for example.


@node Machine state
@section Machine state

FIXME: Use a table like the chapter ``Header layout''.

When the boot loader invokes the 32-bit operating system, the machine
must have the following state:

@itemize @bullet
@item
@code{CS} must be a 32-bit read/execute code segment with an offset of 0
and a limit of 0xFFFFFFFF.

@item
@code{DS}, @code{ES}, @code{FS}, @code{GS}, and @code{SS} must be a
32-bit read/write data segment with an offset of 0 and a limit of
0xFFFFFFFF.

@item
The address 20 line must be usable for standard linear 32-bit addressing
of memory (in standard @sc{pc} hardware, it is wired to 0 at bootup,
forcing addresses in the 1-2 MB range to be mapped to the 0-1 MB range,
3-4 is mapped to 2-3, etc.).

@item
Paging must be turned off.

@item
The processor interrupt flag must be turned off.

@item
@code{EAX} must contain the magic value 0x2BADB002; the presence of this
value indicates to the operating system that it was loaded by a
Multiboot-compliant boot loader (e.g. as opposed to another type of boot
loader that the operating system can also be loaded from).

@item
@code{EBX} must contain the 32-bit physical address of the Multiboot
information structure provided by the boot loader (@pxref{Boot
information format}).
@end itemize

All other processor registers and flag bits are undefined. This
includes, in particular:

@itemize @bullet
@item
@code{ESP}: the 32-bit operating system must create its own stack as
soon as it needs one.

@item
@code{GDTR}: Even though the segment registers are set up as described
above, the @code{GDTR} may be invalid, so the operating system must not
load any segment registers (even just reloading the same values!) until
it sets up its own @code{GDT}.

@item
@code{IDTR}: The operating system must leave interrupts disabled until
it sets up its own @code{IDT}.
@end itemize

However, other machine state should be left by the boot loader in
@dfn{normal working order}, i.e. as initialized by the @sc{bios} (or
DOS, if that's what the boot loader runs from). In other words, the
operating system should be able to make @sc{bios} calls and such after
being loaded, as long as it does not overwrite the @sc{bios} data
structures before doing so. Also, the boot loader must leave the
@sc{pic} programmed with the normal @sc{bios}/DOS values, even if it
changed them during the switch to 32-bit mode.


@node Boot information format
@section Boot information format

FIXME: Split this chapter like the chapter ``OS image format''.

Upon entry to the operating system, the @code{EBX} register contains the
physical address of a @dfn{Multiboot information} data structure,
through which the boot loader communicates vital information to the
operating system. The operating system can use or ignore any parts of
the structure as it chooses; all information passed by the boot loader
is advisory only.

The Multiboot information structure and its related substructures may be
placed anywhere in memory by the boot loader (with the exception of the
memory reserved for the kernel and boot modules, of course). It is the
operating system's responsibility to avoid overwriting this memory until
it is done using it.

The format of the Multiboot information structure (as defined so far)
follows:

@example
@group
        +-------------------+
0       | flags             |    (required)
        +-------------------+
4       | mem_lower         |    (present if flags[0] is set)
8       | mem_upper         |    (present if flags[0] is set)
        +-------------------+
12      | boot_device       |    (present if flags[1] is set)
        +-------------------+
16      | cmdline           |    (present if flags[2] is set)
        +-------------------+
20      | mods_count        |    (present if flags[3] is set)
24      | mods_addr         |    (present if flags[3] is set)
        +-------------------+
28 - 40 | syms              |    (present if flags[4] or
        |                   |                flags[5] is set)
        +-------------------+
44      | mmap_length       |    (present if flags[6] is set)
48      | mmap_addr         |    (present if flags[6] is set)
        +-------------------+
52      | drives_length     |    (present if flags[7] is set)
56      | drives_addr       |    (present if flags[7] is set)
        +-------------------+
60      | config_table      |    (present if flags[8] is set)
        +-------------------+
64      | boot_loader_name  |    (present if flags[9] is set)
        +-------------------+
68      | apm_table         |    (present if flags[10] is set)
        +-------------------+
72      | vbe_control_info  |    (present if flags[11] is set)
76      | vbe_mode_info     |
80      | vbe_mode          |
82      | vbe_interface_seg |
84      | vbe_interface_off |
86      | vbe_interface_len |
        +-------------------+
@end group
@end example

The first longword indicates the presence and validity of other fields
in the Multiboot information structure. All as-yet-undefined bits must
be set to zero by the boot loader. Any set bits that the operating
system does not understand should be ignored. Thus, the @samp{flags}
field also functions as a version indicator, allowing the Multiboot
information structure to be expanded in the future without breaking
anything.

If bit 0 in the @samp{flags} word is set, then the @samp{mem_*} fields
are valid. @samp{mem_lower} and @samp{mem_upper} indicate the amount of
lower and upper memory, respectively, in kilobytes. Lower memory starts
at address 0, and upper memory starts at address 1 megabyte. The maximum
possible value for lower memory is 640 kilobytes. The value returned for
upper memory is maximally the address of the first upper memory hole
minus 1 megabyte. It is not guaranteed to be this value.

If bit 1 in the @samp{flags} word is set, then the @samp{boot_device}
field is valid, and indicates which @sc{bios} disk device the boot
loader loaded the OS image from. If the OS image was not loaded from a
@sc{bios} disk, then this field must not be present (bit 3 must be
clear). The operating system may use this field as a hint for
determining its own @dfn{root} device, but is not required to. The
@samp{boot_device} field is laid out in four one-byte subfields as
follows:

@example
@group
+-------+-------+-------+-------+
| drive | part1 | part2 | part3 |
+-------+-------+-------+-------+
@end group
@end example

The first byte contains the @sc{bios} drive number as understood by the
@sc{bios} INT 0x13 low-level disk interface: e.g. 0x00 for the first
floppy disk or 0x80 for the first hard disk.

The three remaining bytes specify the boot partition. @samp{part1}
specifies the @dfn{top-level} partition number, @samp{part2} specifies a
@dfn{sub-partition} in the top-level partition, etc. Partition numbers
always start from zero. Unused partition bytes must be set to 0xFF. For
example, if the disk is partitioned using a simple one-level DOS
partitioning scheme, then @samp{part1} contains the DOS partition
number, and @samp{part2} and @samp{part3} are both 0xFF. As another
example, if a disk is partitioned first into DOS partitions, and then
one of those DOS partitions is subdivided into several BSD partitions
using BSD's @dfn{disklabel} strategy, then @samp{part1} contains the DOS
partition number, @samp{part2} contains the BSD sub-partition within
that DOS partition, and @samp{part3} is 0xFF.

DOS extended partitions are indicated as partition numbers starting from
4 and increasing, rather than as nested sub-partitions, even though the
underlying disk layout of extended partitions is hierarchical in
nature. For example, if the boot loader boots from the second extended
partition on a disk partitioned in conventional DOS style, then
@samp{part1} will be 5, and @samp{part2} and @samp{part3} will both be
0xFF.

If bit 2 of the @samp{flags} longword is set, the @samp{cmdline} field
is valid, and contains the physical address of the command line to
be passed to the kernel. The command line is a normal C-style
zero-terminated string.

If bit 3 of the @samp{flags} is set, then the @samp{mods} fields
indicate to the kernel what boot modules were loaded along with the
kernel image, and where they can be found. @samp{mods_count} contains
the number of modules loaded; @samp{mods_addr} contains the physical
address of the first module structure. @samp{mods_count} may be zero,
indicating no boot modules were loaded, even if bit 1 of @samp{flags} is
set. Each module structure is formatted as follows:

@example
@group
        +-------------------+
0       | mod_start         |
4       | mod_end           |
        +-------------------+
8       | string            |
        +-------------------+
12      | reserved (0)      |
        +-------------------+
@end group
@end example

The first two fields contain the start and end addresses of the boot
module itself. The @samp{string} field provides an arbitrary string to
be associated with that particular boot module; it is a zero-terminated
ASCII string, just like the kernel command line. The @samp{string} field
may be 0 if there is no string associated with the module. Typically the
string might be a command line (e.g. if the operating system treats boot
modules as executable programs), or a pathname (e.g. if the operating
system treats boot modules as files in a file system), but its exact use
is specific to the operating system. The @samp{reserved} field must be
set to 0 by the boot loader and ignored by the operating system.

@strong{Caution:} Bits 4 & 5 are mutually exclusive.

If bit 4 in the @samp{flags} word is set, then the following fields in
the Multiboot information structure starting at byte 28 are valid:

@example
@group
        +-------------------+
28      | tabsize           |
32      | strsize           |
36      | addr              |
40      | reserved (0)      |
        +-------------------+
@end group
@end example

These indicate where the symbol table from an a.out kernel image can be
found. @samp{addr} is the physical address of the size (4-byte unsigned
long) of an array of a.out format @dfn{nlist} structures, followed
immediately by the array itself, then the size (4-byte unsigned long) of
a set of zero-terminated @sc{ascii} strings (plus sizeof(unsigned long) in
this case), and finally the set of strings itself. @samp{tabsize} is
equal to its size parameter (found at the beginning of the symbol
section), and @samp{strsize} is equal to its size parameter (found at
the beginning of the string section) of the following string table to
which the symbol table refers. Note that @samp{tabsize} may be 0,
indicating no symbols, even if bit 4 in the @samp{flags} word is set.

If bit 5 in the @samp{flags} word is set, then the following fields in
the Multiboot information structure starting at byte 28 are valid:

@example
@group
        +-------------------+
28      | num               |
32      | size              |
36      | addr              |
40      | shndx             |
        +-------------------+
@end group
@end example

These indicate where the section header table from an ELF kernel is, the
size of each entry, number of entries, and the string table used as the
index of names. They correspond to the @samp{shdr_*} entries
(@samp{shdr_num}, etc.) in the Executable and Linkable Format (@sc{elf})
specification in the program header. All sections are loaded, and the
physical address fields of the @sc{elf} section header then refer to where
the sections are in memory (refer to the i386 @sc{elf} documentation for
details as to how to read the section header(s)). Note that
@samp{shdr_num} may be 0, indicating no symbols, even if bit 5 in the
@samp{flags} word is set.

If bit 6 in the @samp{flags} word is set, then the @samp{mmap_*} fields
are valid, and indicate the address and length of a buffer containing a
memory map of the machine provided by the @sc{bios}. @samp{mmap_addr} is
the address, and @samp{mmap_length} is the total size of the buffer. The
buffer consists of one or more of the following size/structure pairs
(@samp{size} is really used for skipping to the next pair):

@example
@group
        +-------------------+
-4      | size              |
        +-------------------+
0       | base_addr_low     |
4       | base_addr_high    |
8       | length_low        |
12      | length_high       |
16      | type              |
        +-------------------+
@end group
@end example

where @samp{size} is the size of the associated structure in bytes, which
can be greater than the minimum of 20 bytes. @samp{base_addr_low} is the
lower 32 bits of the starting address, and @samp{base_addr_high} is the
upper 32 bits, for a total of a 64-bit starting address. @samp{length_low}
is the lower 32 bits of the size of the memory region in bytes, and
@samp{length_high} is the upper 32 bits, for a total of a 64-bit
length. @samp{type} is the variety of address range represented, where a
value of 1 indicates available @sc{ram}, and all other values currently
indicated a reserved area.

The map provided is guaranteed to list all standard @sc{ram} that should
be available for normal use.

If bit 7 in the @samp{flags} is set, then the @samp{drives_*} fields
are valid, and indicate the address of the physical address of the first
drive structure and the size of drive structures. @samp{drives_addr}
is the address, and @samp{drives_length} is the total size of drive
structures. Note that @samp{drives_length} may be zero. Each drive
structure is formatted as follows:

@example
@group
        +-------------------+
0       | size              |
        +-------------------+
4       | drive_number      |
        +-------------------+
5       | drive_mode        |
        +-------------------+
6       | drive_cylinders   |
8       | drive_heads       |
9       | drive_sectors     |
        +-------------------+
10 - xx | drive_ports       |
        +-------------------+
@end group
@end example

The @samp{size} field specifies the size of this structure. The size
varies, depending on the number of ports. Note that the size may not be
equal to (10 + 2 * the number of ports), because of an alignment.

The @samp{drive_number} field contains the BIOS drive number. The
@samp{drive_mode} field represents the access mode used by the boot
loader. Currently, the following modes are defined:

@table @samp
@item 0
CHS mode (traditional cylinder/head/sector addressing mode).

@item 1
LBA mode (Logical Block Addressing mode).
@end table

The three fields, @samp{drive_cylinders}, @samp{drive_heads} and
@samp{drive_sectors}, indicate the geometry of the drive detected by the
@sc{bios}. @samp{drive_cylinders} contains the number of the
cylinders. @samp{drive_heads} contains the number of the
heads. @samp{drive_sectors} contains the number of the sectors per
track.

The @samp{drive_ports} field contains the array of the I/O ports used
for the drive in the @sc{bios} code. The array consists of zero or more
unsigned two-bytes integers, and is terminated with zero. Note that the
array may contain any number of I/O ports that are not related to the
drive actually (such as @sc{dma} controller's ports).

If bit 8 in the @samp{flags} is set, then the @samp{config_table} field
is valid, and indicates the address of the @sc{rom} configuration table
returned by the @dfn{GET CONFIGURATION} @sc{bios} call. If the @sc{bios}
call fails, then the size of the table must be @emph{zero}.

If bit 9 in the @samp{flags} is set, the @samp{boot_loader_name} field
is valid, and contains the physical address of the name of a boot
loader booting the kernel. The name is a normal C-style zero-terminated
string.

If bit 10 in the @samp{flags} is set, the @samp{apm_table} field is
valid, and contains the physical address of an @sc{apm} table defined as
below:

@example
@group
        +----------------------+
0       | version              |
2       | cseg                 |
4       | offset               |
8       | cseg_16              |
10      | dseg                 |
12      | flags                |
14      | cseg_len             |
16      | cseg_16_len          |
18      | dseg_len             |
        +----------------------+
@end group
@end example

The fields @samp{version}, @samp{cseg}, @samp{offset}, @samp{cseg_16},
@samp{dseg}, @samp{flags}, @samp{cseg_len}, @samp{cseg_16_len},
@samp{dseg_len} indicate the version number, the protected mode 32-bit
code segment, the offset of the entry point, the protected mode 16-bit
code segment, the protected mode 16-bit data segment, the flags, the
length of the protected mode 32-bit code segment, the length of the
protected mode 16-bit code segment, and the length of the protected mode
16-bit data segment, respectively. Only the field @samp{offset} is 4
bytes, and the others are 2 bytes. See
@uref{http://www.microsoft.com/hwdev/busbios/amp_12.htm, Advanced Power
Management (APM) BIOS Interface Specification}, for more information.

If bit 11 in the @samp{flags} is set, the graphics table is available.
This must only be done if the kernel has indicated in the
@samp{Multiboot Header} that it accepts a graphics mode.

The fields @samp{vbe_control_info} and @samp{vbe_mode_info} contain
the physical addresses of @sc{vbe} control information returned by the
@sc{vbe} Function 00h and @sc{vbe} mode information returned by the
@sc{vbe} Function 01h, respectively.

The field @samp{vbe_mode} indicates current video mode in the format
specified in @sc{vbe} 3.0.

The rest fields @samp{vbe_interface_seg}, @samp{vbe_interface_off}, and
@samp{vbe_interface_len} contain the table of a protected mode interface
defined in @sc{vbe} 2.0+. If this information is not available, those
fields contain zero. Note that @sc{vbe} 3.0 defines another protected
mode interface which is incompatible with the old one. If you want to
use the new protected mode interface, you will have to find the table
yourself.

The fields for the graphics table are designed for @sc{vbe}, but
Multiboot boot loaders may simulate @sc{vbe} on non-@sc{vbe} modes, as
if they were @sc{vbe} modes.


@node Examples
@chapter Examples

@strong{Caution:} The following items are not part of the specification
document, but are included for prospective operating system and boot
loader writers.

@menu
* Notes on PC::                 
* BIOS device mapping techniques::  
* Example OS code::             
* Example boot loader code::    
@end menu


@node Notes on PC
@section Notes on PC

In reference to bit 0 of the @samp{flags} parameter in the Multiboot
information structure, if the bootloader in question uses older
@sc{bios} interfaces, or the newest ones are not available (see
description about bit 6), then a maximum of either 15 or 63 megabytes of
memory may be reported. It is @emph{highly} recommended that boot
loaders perform a thorough memory probe.

In reference to bit 1 of the @samp{flags} parameter in the Multiboot
information structure, it is recognized that determination of which
@sc{bios} drive maps to which device driver in an operating system is
non-trivial, at best. Many kludges have been made to various operating
systems instead of solving this problem, most of them breaking under
many conditions. To encourage the use of general-purpose solutions to
this problem, there are 2 @sc{bios} device mapping techniques
(@pxref{BIOS device mapping techniques}). 

In reference to bit 6 of the @samp{flags} parameter in the Multiboot
information structure, it is important to note that the data structure
used there (starting with @samp{BaseAddrLow}) is the data returned by
the INT 15h, AX=E820h --- Query System Address Map call. See @xref{Query
System Address Map, , Query System Address Map, grub.info, The GRUB
Manual}, for more information. The interface here is meant to allow a
boot loader to work unmodified with any reasonable extensions of the
@sc{bios} interface, passing along any extra data to be interpreted by
the operating system as desired.


@node BIOS device mapping techniques
@section BIOS device mapping techniques

Both of these techniques should be usable from any PC operating system,
and neither require any special support in the drivers themselves. This
section will be flushed out into detailed explanations, particularly for
the I/O restriction technique.

The general rule is that the data comparison technique is the quick and
dirty solution. It works most of the time, but doesn't cover all the
bases, and is relatively simple.

The I/O restriction technique is much more complex, but it has potential
to solve the problem under all conditions, plus allow access of the
remaining @sc{bios} devices when not all of them have operating system
drivers.

@menu
* Data comparison technique::   
* I/O restriction technique::   
@end menu


@node Data comparison technique
@subsection Data comparison technique

Before activating @emph{any} of the device drivers, gather enough data
from similar sectors on each of the disks such that each one can be
uniquely identified.

After activating the device drivers, compare data from the drives using
the operating system drivers. This should hopefully be sufficient to
provide such a mapping.

Problems:

@enumerate
@item
The data on some @sc{bios} devices might be identical (so the part
reading the drives from the @sc{bios} should have some mechanism to give
up).

@item
There might be extra drives not accessible from the @sc{bios} which are
identical to some drive used by the @sc{bios} (so it should be capable
of giving up there as well).
@end enumerate


@node I/O restriction technique
@subsection I/O restriction technique

This first step may be unnecessary, but first create copy-on-write
mappings for the device drivers writing into @sc{pc} @sc{ram}. Keep the
original copies for the @dfn{clean @sc{bios} virtual machine} to be
created later.

For each device driver brought online, determine which @sc{bios} devices
become inaccessible by:

@enumerate
@item
Create a @dfn{clean @sc{bios} virtual machine}.

@item
Set the I/O permission map for the I/O area claimed by the device driver
to no permissions (neither read nor write).

@item
Access each device.

@item
Record which devices succeed, and those which try to access the
@dfn{restricted} I/O areas (hopefully, this will be an @dfn{xor}
situation).
@end enumerate

For each device driver, given how many of the @sc{bios} devices were
subsumed by it (there should be no gaps in this list), it should be easy
to determine which devices on the controller these are.

In general, you have at most 2 disks from each controller given
@sc{bios} numbers, but they pretty much always count from the lowest
logically numbered devices on the controller.


@node Example OS code
@section Example OS code

In this distribution, the example Multiboot kernel @file{kernel} is
included. The kernel just prints out the Multiboot information structure
on the screen, so you can make use of the kernel to test a
Multiboot-compliant boot loader and for reference to how to implement a
Multiboot kernel. The source files can be found under the directory
@file{docs} in the GRUB distribution.

The kernel @file{kernel} consists of only three files: @file{boot.S},
@file{kernel.c} and @file{multiboot.h}. The assembly source
@file{boot.S} is written in GAS (@pxref{Top, , GNU assembler, as.info,
The GNU assembler}), and contains the Multiboot information structure to
comply with the specification. When a Multiboot-compliant boot loader
loads and execute it, it initialize the stack pointer and @code{EFLAGS},
and then call the function @code{cmain} defined in @file{kernel.c}. If
@code{cmain} returns to the callee, then it shows a message to inform
the user of the halt state and stops forever until you push the reset
key. The file @file{kernel.c} contains the function @code{cmain},
which checks if the magic number passed by the boot loader is valid and
so on, and some functions to print messages on the screen. The file
@file{multiboot.h} defines some macros, such as the magic number for the
Multiboot header, the Multiboot header structure and the Multiboot
information structure.

@menu
* multiboot.h::                 
* boot.S::                      
* kernel.c::                    
* Other Multiboot kernels::     
@end menu


@node multiboot.h
@subsection multiboot.h

This is the source code in the file @file{multiboot.h}:

@example
@include multiboot.h.texi
@end example


@node boot.S
@subsection boot.S

In the file @file{boot.S}:

@example
@include boot.S.texi
@end example


@node kernel.c
@subsection kernel.c

And, in the file @file{kernel.c}:

@example
@include kernel.c.texi
@end example


@node Other Multiboot kernels
@subsection Other Multiboot kernels

Other useful information should be available in Multiboot kernels, such
as GNU Mach and Fiasco @url{http://os.inf.tu-dresden.de/fiasco/}. And,
it is worth mentioning the OSKit
@url{http://www.cs.utah.edu/projects/flux/oskit/}, which provides a
library supporting the specification.


@node Example boot loader code
@section Example boot loader code

The GNU GRUB (@pxref{Top, , GRUB, grub.info, The GRUB manual}) project
is a full Multiboot-compliant boot loader, supporting all required and
optional features present in this specification. A public release has
not been made, but the test release is available from:

@url{ftp://alpha.gnu.org/gnu/hurd/src/}

See the webpage @url{http://www.gnu.org/software/grub/grub.html}, for
more information.


@node History
@chapter The change log of this specification

@table @asis
@item 0.7
@itemize @bullet
@item
@dfn{Multiboot Standard} is renamed to @dfn{Multiboot Specification}.

@item
Graphics fields are added to Multiboot header.

@item
BIOS drive information, BIOS configuration table, the name of a boot
loader, APM information, and graphics information are added to Multiboot
information.

@item
Rewritten in Texinfo format.

@item
Rewritten, using more strict words.

@item
The maintainer changes to the GNU GRUB maintainer team
@email{bug-grub@@gnu.org}, from Bryan Ford and Erich Stefan Boleyn.
@end itemize

@item 0.6
@itemize @bullet
@item
A few wording changes.

@item
Header checksum.

@item
Clasification of machine state passed to an operating system.
@end itemize

@item 0.5
@itemize @bullet
@item
Name change.
@end itemize

@item 0.4
@itemize @bullet
@item
Major changes plus HTMLification.
@end itemize
@end table


@node Index
@unnumbered Index

@printindex cp

@contents
@bye