THE LINUX/I386 BOOT PROTOCOL

qfox

文件:Linux-i386-Boot-Code-HOWTO.pdf大小:178KB下载:
下载
                     THE LINUX/I386 BOOT PROTOCOL
                     ----------------------------
                    H. Peter Anvin
                        Last update 2005-09-02
On the i386 platform, the Linux kernel uses a rather complicated boot
convention.  This has evolved partially due to historical aspects, as
well as the desire in the early days to have the kernel itself be a
bootable image, the complicated PC memory model and due to changed
expectations in the PC industry caused by the effective demise of
real-mode DOS as a mainstream operating system.
Currently, four versions of the Linux/i386 boot protocol exist.
Old kernels:        zImage/Image support only.  Some very early kernels
                may not even support a command line.
Protocol 2.00:        (Kernel 1.3.73) Added bzImage and initrd support, as
                well as a formalized way to communicate between the
                boot loader and the kernel.  setup.S made relocatable,
                although the traditional setup area still assumed
                writable.
Protocol 2.01:        (Kernel 1.3.76) Added a heap overrun warning.
Protocol 2.02:        (Kernel 2.4.0-test3-pre3) New command line protocol.
                Lower the conventional memory ceiling.        No overwrite
                of the traditional setup area, thus making booting
                safe for systems which use the EBDA from SMM or 32-bit
                BIOS entry points.  zImage deprecated but still
                supported.
Protocol 2.03:        (Kernel 2.4.18-pre1) Explicitly makes the highest possible
                initrd address available to the bootloader.
Protocol 2.04:        (Kernel 2.6.14) Extend the syssize field to four bytes.
**** MEMORY LAYOUT
The traditional memory map for the kernel loader, used for Image or
zImage kernels, typically looks like:
        |                         |
0A0000        +------------------------+
        |  Reserved for BIOS         |        Do not use.  Reserved for BIOS EBDA.
09A000        +------------------------+
        |  Stack/heap/cmdline         |        For use by the kernel real-mode code.
098000        +------------------------+       
        |  Kernel setup                 |        The kernel real-mode code.
090200        +------------------------+
        |  Kernel boot sector         |        The kernel legacy boot sector.
090000        +------------------------+
        |  Protected-mode kernel |        The bulk of the kernel image.
010000        +------------------------+
        |  Boot loader                 |         if you need a bootloader ID
        value assigned.
  loadflags, heap_end_ptr:
        If the protocol version is 2.01 or higher, enter the
        offset limit of the setup heap into heap_end_ptr and set the
        0x80 bit (CAN_USE_HEAP) of loadflags.  heap_end_ptr appears to
        be relative to the start of setup (offset 0x0200).
  setup_move_size:
        When using protocol 2.00 or 2.01, if the real mode
        kernel is not loaded at 0x90000, it gets moved there later in
        the loading sequence.  Fill in this field if you want
        additional data (such as the kernel command line) moved in
        addition to the real-mode kernel itself.
  ramdisk_image, ramdisk_size:
        If your boot loader has loaded an initial ramdisk (initrd),
        set ramdisk_image to the 32-bit pointer to the ramdisk data
        and the ramdisk_size to the size of the ramdisk data.
        The initrd should typically be located as high in memory as
        possible, as it may otherwise get overwritten by the early
        kernel initialization sequence.         However, it must never be
        located above the address specified in the initrd_addr_max
        field.        The initrd should be at least 4K page aligned.
  cmd_line_ptr:
        If the protocol version is 2.02 or higher, this is a 32-bit
        pointer to the kernel command line.  The kernel command line
        can be located anywhere between the end of setup and 0xA0000.
        Fill in this field even if your boot loader does not support a
        command line, in which case you can point this to an empty
        string (or better yet, to the string "auto".)  If this field
        is left at zero, the kernel will assume that your boot loader
        does not support the 2.02+ protocol.
  ramdisk_max:
        The maximum address that may be occupied by the initrd
        contents.  For boot protocols 2.02 or earlier, this field is
        not present, and the maximum address is 0x37FFFFFF.  (This
        address is defined as the address of the highest safe byte, so
        if your ramdisk is exactly 131072 bytes long and this field is
        0x37FFFFFF, you can start your ramdisk at 0x37FE0000.)
**** THE KERNEL COMMAND LINE
The kernel command line has become an important way for the boot
loader to communicate with the kernel.  Some of its options are also
relevant to the boot loader itself, see "special command line options"
below.
The kernel command line is a null-terminated string currently up to
255 characters long, plus the final null.  A string that is too long
will be automatically truncated by the kernel, a boot loader may allow
a longer command line to be passed to permit future kernels to extend
this limit.
If the boot protocol version is 2.02 or later, the address of the
kernel command line is given by the header field cmd_line_ptr (see
above.)  This address can be anywhere between the end of the setup
heap and 0xA0000.
If the protocol version is *not* 2.02 or higher, the kernel
command line is entered using the following protocol:
        At offset 0x0020 (word), "cmd_line_magic", enter the magic
        number 0xA33F.
        At offset 0x0022 (word), "cmd_line_offset", enter the offset
        of the kernel command line (relative to the start of the
        real-mode kernel).
       
        The kernel command line *must* be within the memory region
        covered by setup_move_size, so you may need to adjust this
        field.
**** SAMPLE BOOT CONFIGURATION
As a sample configuration, assume the following layout of the real
mode segment (this is a typical, and recommended layout):
        0x0000-0x7FFF        Real mode kernel
        0x8000-0x8FFF        Stack and heap
        0x9000-0x90FF        Kernel command line
Such a boot loader should enter the following fields in the header:
        unsigned long base_ptr;        /* base address for real-mode segment */
        if ( setup_sects == 0 ) {
                setup_sects = 4;
        }
        if ( protocol >= 0x0200 ) {
                type_of_loader = ;
                if ( loading_initrd ) {
                        ramdisk_image = ;
                        ramdisk_size = ;
                }
                if ( protocol >= 0x0201 ) {
                        heap_end_ptr = 0x9000 - 0x200;
                        loadflags |= 0x80; /* CAN_USE_HEAP */
                }
                if ( protocol >= 0x0202 ) {
                        cmd_line_ptr = base_ptr + 0x9000;
                } else {
                        cmd_line_magic        = 0xA33F;
                        cmd_line_offset = 0x9000;
                        setup_move_size = 0x9100;
                }
        } else {
                /* Very old kernel */
                cmd_line_magic        = 0xA33F;
                cmd_line_offset = 0x9000;
                /* A very old kernel MUST have its real-mode code
                   loaded at 0x90000 */
                if ( base_ptr != 0x90000 ) {
                        /* Copy the real-mode kernel */
                        memcpy(0x90000, base_ptr, (setup_sects+1)*512);
                        /* Copy the command line */
                        memcpy(0x99000, base_ptr+0x9000, 256);
                        base_ptr = 0x90000;                 /* Relocated */
                }
                /* It is recommended to clear memory up to the 32K mark */
                memset(0x90000 + (setup_sects+1)*512, 0,
                       (64-(setup_sects+1))*512);
        }
**** LOADING THE REST OF THE KERNEL
The 32-bit (non-real-mode) kernel starts at offset (setup_sects+1)*512
in the kernel file (again, if setup_sects == 0 the real value is 4.)
It should be loaded at address 0x10000 for Image/zImage kernels and
0x100000 for bzImage kernels.
The kernel is a bzImage kernel if the protocol >= 2.00 and the 0x01
bit (LOAD_HIGH) in the loadflags field is set:
        is_bzImage = (protocol >= 0x0200) && (loadflags & 0x01);
        load_address = is_bzImage ? 0x100000 : 0x10000;
Note that Image/zImage kernels can be up to 512K in size, and thus use
the entire 0x10000-0x90000 range of memory.  This means it is pretty
much a requirement for these kernels to load the real-mode part at
0x90000.  bzImage kernels allow much more flexibility.
**** SPECIAL COMMAND LINE OPTIONS
If the command line provided by the boot loader is entered by the
user, the user may expect the following command line options to work.
They should normally not be deleted from the kernel command line even
though not all of them are actually meaningful to the kernel.  Boot
loader authors who need additional command line options for the boot
loader itself should get them registered in
Documentation/kernel-parameters.txt to make sure they will not
conflict with actual kernel options now or in the future.
  vga=
         here is either an integer (in C notation, either
        decimal, octal, or hexadecimal) or one of the strings
        "normal" (meaning 0xFFFF), "ext" (meaning 0xFFFE) or "ask"
        (meaning 0xFFFD).  This value should be entered into the
        vid_mode field, as it is used by the kernel before the command
        line is parsed.
  mem=
         is an integer in C notation optionally followed by K, M
        or G (meaning
        An initrd should be loaded.  The meaning of  is
        obviously bootloader-dependent, and some boot loaders
        (e.g. LILO) do not have such a command.
In addition, some boot loaders add the following options to the
user-specified command line:
  BOOT_IMAGE=
        The boot image which was loaded.  Again, the meaning of
        is obviously bootloader-dependent.
  auto
        The kernel was booted without explicit user intervention.
If these options are added by the boot loader, it is highly
recommended that they are located *first*, before the user-specified
or configuration-specified command line.  Otherwise, "init=/bin/sh"
gets confused by the "auto" option.
**** RUNNING THE KERNEL
The kernel is started by jumping to the kernel entry point, which is
located at *segment* offset 0x20 from the start of the real mode
kernel.  This means that if you loaded your real-mode kernel code at
0x90000, the kernel entry point is 9020:0000.
At entry, ds = es = ss should point to the start of the real-mode
kernel code (0x9000 if the code is loaded at 0x90000), sp should be
set up properly, normally pointing to the top of the heap, and
interrupts should be disabled.  Furthermore, to guard against bugs in
the kernel, it is recommended that the boot loader sets fs = gs = ds =
es = ss.
In our example from above, we would do:
        /* Note: in the case of the "old" kernel protocol, base_ptr must
           be == 0x90000 at this point; see the previous sample code */
        seg = base_ptr >> 4;
        cli();        /* Enter with interrupts disabled! */
        /* Set up the real-mode kernel stack */
        _SS = seg;
        _SP = 0x9000;        /* Load SP immediately after loading SS! */
        _DS = _ES = _FS = _GS = seg;
        jmp_far(seg+0x20, 0);        /* Run the kernel */
If your boot sector accesses a floppy drive, it is recommended to
switch off the floppy motor before running the kernel, since the
kernel boot leaves interrupts off and thus the motor will not be
switched off, especially if the loaded kernel has the floppy driver as
a demand-loaded module!
**** ADVANCED BOOT TIME HOOKS
If the boot loader runs in a particularly hostile environment (such as
LOADLIN, which runs under DOS) it may be impossible to follow the
standard memory location requirements.  Such a boot loader may use the
following hooks that, if set, are invoked by the kernel at the
appropriate time.  The use of these hooks should probably be
considered an absolutely last resort!
IMPORTANT: All the hooks are required to preserve %esp, %ebp, %esi and
%edi across invocation.
  realmode_swtch:
        A 16-bit real mode far subroutine invoked immediately before
        entering protected mode.  The default routine disables NMI, so
        your routine should probably do so, too.
  code32_start:
        A 32-bit flat-mode routine *jumped* to immediately after the
        transition to protected mode, but before the kernel is
        uncompressed.  No segments, except CS, are set up; you should
        set them up to KERNEL_DS (0x18) yourself.
        After completing your hook, you should jump to the address
        that was in this field before your boot loader overwrote it.


本文来自ChinaUnix博客，如果查看原文请点：http://blog.chinaunix.net/u/19881/showart_219334.html