1 THE LINUX/I386 BOOT PROTOCOL
2 ----------------------------
4 H. Peter Anvin <hpa@zytor.com>
7 On the i386 platform, the Linux kernel uses a rather complicated boot
8 convention. This has evolved partially due to historical aspects, as
9 well as the desire in the early days to have the kernel itself be a
10 bootable image, the complicated PC memory model and due to changed
11 expectations in the PC industry caused by the effective demise of
12 real-mode DOS as a mainstream operating system.
14 Currently, the following versions of the Linux/i386 boot protocol exist.
16 Old kernels: zImage/Image support only. Some very early kernels
17 may not even support a command line.
19 Protocol 2.00: (Kernel 1.3.73) Added bzImage and initrd support, as
20 well as a formalized way to communicate between the
21 boot loader and the kernel. setup.S made relocatable,
22 although the traditional setup area still assumed
25 Protocol 2.01: (Kernel 1.3.76) Added a heap overrun warning.
27 Protocol 2.02: (Kernel 2.4.0-test3-pre3) New command line protocol.
28 Lower the conventional memory ceiling. No overwrite
29 of the traditional setup area, thus making booting
30 safe for systems which use the EBDA from SMM or 32-bit
31 BIOS entry points. zImage deprecated but still
34 Protocol 2.03: (Kernel 2.4.18-pre1) Explicitly makes the highest possible
35 initrd address available to the bootloader.
37 Protocol 2.04: (Kernel 2.6.14) Extend the syssize field to four bytes.
39 Protocol 2.05: (Kernel 2.6.20) Make protected mode kernel relocatable.
40 Introduce relocatable_kernel and kernel_alignment fields.
42 Protocol 2.06: (Kernel 2.6.22) Added a field that contains the size of
48 The traditional memory map for the kernel loader, used for Image or
49 zImage kernels, typically looks like:
52 0A0000 +------------------------+
53 | Reserved for BIOS | Do not use. Reserved for BIOS EBDA.
54 09A000 +------------------------+
56 | Stack/heap | For use by the kernel real-mode code.
57 098000 +------------------------+
58 | Kernel setup | The kernel real-mode code.
59 090200 +------------------------+
60 | Kernel boot sector | The kernel legacy boot sector.
61 090000 +------------------------+
62 | Protected-mode kernel | The bulk of the kernel image.
63 010000 +------------------------+
64 | Boot loader | <- Boot sector entry point 0000:7C00
65 001000 +------------------------+
66 | Reserved for MBR/BIOS |
67 000800 +------------------------+
68 | Typically used by MBR |
69 000600 +------------------------+
71 000000 +------------------------+
74 When using bzImage, the protected-mode kernel was relocated to
75 0x100000 ("high memory"), and the kernel real-mode block (boot sector,
76 setup, and stack/heap) was made relocatable to any address between
77 0x10000 and end of low memory. Unfortunately, in protocols 2.00 and
78 2.01 the 0x90000+ memory range is still used internally by the kernel;
79 the 2.02 protocol resolves that problem.
81 It is desirable to keep the "memory ceiling" -- the highest point in
82 low memory touched by the boot loader -- as low as possible, since
83 some newer BIOSes have begun to allocate some rather large amounts of
84 memory, called the Extended BIOS Data Area, near the top of low
85 memory. The boot loader should use the "INT 12h" BIOS call to verify
86 how much low memory is available.
88 Unfortunately, if INT 12h reports that the amount of memory is too
89 low, there is usually nothing the boot loader can do but to report an
90 error to the user. The boot loader should therefore be designed to
91 take up as little space in low memory as it reasonably can. For
92 zImage or old bzImage kernels, which need data written into the
93 0x90000 segment, the boot loader should make sure not to use memory
94 above the 0x9A000 point; too many BIOSes will break above that point.
96 For a modern bzImage kernel with boot protocol version >= 2.02, a
97 memory layout like the following is suggested:
100 | Protected-mode kernel |
101 100000 +------------------------+
103 0A0000 +------------------------+
104 | Reserved for BIOS | Leave as much as possible unused
106 | Command line | (Can also be below the X+10000 mark)
107 X+10000 +------------------------+
108 | Stack/heap | For use by the kernel real-mode code.
109 X+08000 +------------------------+
110 | Kernel setup | The kernel real-mode code.
111 | Kernel boot sector | The kernel legacy boot sector.
112 X +------------------------+
113 | Boot loader | <- Boot sector entry point 0000:7C00
114 001000 +------------------------+
115 | Reserved for MBR/BIOS |
116 000800 +------------------------+
117 | Typically used by MBR |
118 000600 +------------------------+
120 000000 +------------------------+
122 ... where the address X is as low as the design of the boot loader
126 **** THE REAL-MODE KERNEL HEADER
128 In the following text, and anywhere in the kernel boot sequence, "a
129 sector" refers to 512 bytes. It is independent of the actual sector
130 size of the underlying medium.
132 The first step in loading a Linux kernel should be to load the
133 real-mode code (boot sector and setup code) and then examine the
134 following header at offset 0x01f1. The real-mode code can total up to
135 32K, although the boot loader may choose to load only the first two
136 sectors (1K) and then examine the bootup sector size.
138 The header looks like:
140 Offset Proto Name Meaning
143 01F1/1 ALL(1 setup_sects The size of the setup in sectors
144 01F2/2 ALL root_flags If set, the root is mounted readonly
145 01F4/4 2.04+(2 syssize The size of the 32-bit code in 16-byte paras
146 01F8/2 ALL ram_size DO NOT USE - for bootsect.S use only
147 01FA/2 ALL vid_mode Video mode control
148 01FC/2 ALL root_dev Default root device number
149 01FE/2 ALL boot_flag 0xAA55 magic number
150 0200/2 2.00+ jump Jump instruction
151 0202/4 2.00+ header Magic signature "HdrS"
152 0206/2 2.00+ version Boot protocol version supported
153 0208/4 2.00+ realmode_swtch Boot loader hook (see below)
154 020C/2 2.00+ start_sys The load-low segment (0x1000) (obsolete)
155 020E/2 2.00+ kernel_version Pointer to kernel version string
156 0210/1 2.00+ type_of_loader Boot loader identifier
157 0211/1 2.00+ loadflags Boot protocol option flags
158 0212/2 2.00+ setup_move_size Move to high memory size (used with hooks)
159 0214/4 2.00+ code32_start Boot loader hook (see below)
160 0218/4 2.00+ ramdisk_image initrd load address (set by boot loader)
161 021C/4 2.00+ ramdisk_size initrd size (set by boot loader)
162 0220/4 2.00+ bootsect_kludge DO NOT USE - for bootsect.S use only
163 0224/2 2.01+ heap_end_ptr Free memory after setup end
164 0226/2 N/A pad1 Unused
165 0228/4 2.02+ cmd_line_ptr 32-bit pointer to the kernel command line
166 022C/4 2.03+ initrd_addr_max Highest legal initrd address
167 0230/4 2.05+ kernel_alignment Physical addr alignment required for kernel
168 0234/1 2.05+ relocatable_kernel Whether kernel is relocatable or not
169 0235/3 N/A pad2 Unused
170 0238/4 2.06+ cmdline_size Maximum size of the kernel command line
171 023C/4 2.07+ hardware_subarch Hardware subarchitecture
172 0240/8 2.07+ hardware_subarch_data Subarchitecture-specific data
173 0248/4 2.08+ compressed_payload_offset
174 024C/4 2.08+ compressed_payload_length
176 (1) For backwards compatibility, if the setup_sects field contains 0, the
179 (2) For boot protocol prior to 2.04, the upper two bytes of the syssize
180 field are unusable, which means the size of a bzImage kernel
181 cannot be determined.
183 If the "HdrS" (0x53726448) magic number is not found at offset 0x202,
184 the boot protocol version is "old". Loading an old kernel, the
185 following parameters should be assumed:
189 Real-mode kernel must be located at 0x90000.
191 Otherwise, the "version" field contains the protocol version,
192 e.g. protocol version 2.01 will contain 0x0201 in this field. When
193 setting fields in the header, you must make sure only to set fields
194 supported by the protocol version in use.
197 **** DETAILS OF HEADER FIELDS
199 For each field, some are information from the kernel to the bootloader
200 ("read"), some are expected to be filled out by the bootloader
201 ("write"), and some are expected to be read and modified by the
202 bootloader ("modify").
204 All general purpose boot loaders should write the fields marked
205 (obligatory). Boot loaders who want to load the kernel at a
206 nonstandard address should fill in the fields marked (reloc); other
207 boot loaders can ignore those fields.
209 The byte order of all fields is littleendian (this is x86, after all.)
211 Field name: setup_sects
216 The size of the setup code in 512-byte sectors. If this field is
217 0, the real value is 4. The real-mode code consists of the boot
218 sector (always one 512-byte sector) plus the setup code.
220 Field name: root_flags
221 Type: modify (optional)
225 If this field is nonzero, the root defaults to readonly. The use of
226 this field is deprecated; use the "ro" or "rw" options on the
227 command line instead.
231 Offset/size: 0x1f4/4 (protocol 2.04+) 0x1f4/2 (protocol ALL)
234 The size of the protected-mode code in units of 16-byte paragraphs.
235 For protocol versions older than 2.04 this field is only two bytes
236 wide, and therefore cannot be trusted for the size of a kernel if
237 the LOAD_HIGH flag is set.
240 Type: kernel internal
244 This field is obsolete.
247 Type: modify (obligatory)
250 Please see the section on SPECIAL COMMAND LINE OPTIONS.
253 Type: modify (optional)
257 The default root device device number. The use of this field is
258 deprecated, use the "root=" option on the command line instead.
260 Field name: boot_flag
265 Contains 0xAA55. This is the closest thing old Linux kernels have
273 Contains an x86 jump instruction, 0xEB followed by a signed offset
274 relative to byte 0x202. This can be used to determine the size of
282 Contains the magic number "HdrS" (0x53726448).
289 Contains the boot protocol version, in (major << 8)+minor format,
290 e.g. 0x0204 for version 2.04, and 0x0a11 for a hypothetical version
293 Field name: readmode_swtch
294 Type: modify (optional)
298 Boot loader hook (see ADVANCED BOOT LOADER HOOKS below.)
300 Field name: start_sys
305 The load low segment (0x1000). Obsolete.
307 Field name: kernel_version
312 If set to a nonzero value, contains a pointer to a NUL-terminated
313 human-readable kernel version number string, less 0x200. This can
314 be used to display the kernel version to the user. This value
315 should be less than (0x200*setup_sects).
317 For example, if this value is set to 0x1c00, the kernel version
318 number string can be found at offset 0x1e00 in the kernel file.
319 This is a valid value if and only if the "setup_sects" field
320 contains the value 15 or higher, as:
322 0x1c00 < 15*0x200 (= 0x1e00) but
323 0x1c00 >= 14*0x200 (= 0x1c00)
325 0x1c00 >> 9 = 14, so the minimum value for setup_secs is 15.
327 Field name: type_of_loader
328 Type: write (obligatory)
332 If your boot loader has an assigned id (see table below), enter
333 0xTV here, where T is an identifier for the boot loader and V is
334 a version number. Otherwise, enter 0xFF here.
336 Assigned boot loader ids:
337 0 LILO (0x00 reserved for pre-2.00 bootloader)
339 2 bootsect-loader (0x20, all other values reserved)
349 Please contact <hpa@zytor.com> if you need a bootloader ID
352 Field name: loadflags
353 Type: modify (obligatory)
357 This field is a bitmask.
359 Bit 0 (read): LOADED_HIGH
360 - If 0, the protected-mode code is loaded at 0x10000.
361 - If 1, the protected-mode code is loaded at 0x100000.
363 Bit 6 (write): KEEP_SEGMENTS
365 - if 0, reload the segment registers in the 32bit entry point.
366 - if 1, do not reload the segment registers in the 32bit entry point.
367 Assume that %cs %ds %ss %es are all set to flat segments with
368 a base of 0 (or the equivalent for their environment).
370 Bit 7 (write): CAN_USE_HEAP
371 Set this bit to 1 to indicate that the value entered in the
372 heap_end_ptr is valid. If this field is clear, some setup code
373 functionality will be disabled.
375 Field name: setup_move_size
376 Type: modify (obligatory)
380 When using protocol 2.00 or 2.01, if the real mode kernel is not
381 loaded at 0x90000, it gets moved there later in the loading
382 sequence. Fill in this field if you want additional data (such as
383 the kernel command line) moved in addition to the real-mode kernel
386 The unit is bytes starting with the beginning of the boot sector.
388 This field is can be ignored when the protocol is 2.02 or higher, or
389 if the real-mode code is loaded at 0x90000.
391 Field name: code32_start
392 Type: modify (optional, reloc)
396 The address to jump to in protected mode. This defaults to the load
397 address of the kernel, and can be used by the boot loader to
398 determine the proper load address.
400 This field can be modified for two purposes:
402 1. as a boot loader hook (see ADVANCED BOOT LOADER HOOKS below.)
404 2. if a bootloader which does not install a hook loads a
405 relocatable kernel at a nonstandard address it will have to modify
406 this field to point to the load address.
408 Field name: ramdisk_image
409 Type: write (obligatory)
413 The 32-bit linear address of the initial ramdisk or ramfs. Leave at
414 zero if there is no initial ramdisk/ramfs.
416 Field name: ramdisk_size
417 Type: write (obligatory)
421 Size of the initial ramdisk or ramfs. Leave at zero if there is no
422 initial ramdisk/ramfs.
424 Field name: bootsect_kludge
425 Type: kernel internal
429 This field is obsolete.
431 Field name: heap_end_ptr
432 Type: write (obligatory)
436 Set this field to the offset (from the beginning of the real-mode
437 code) of the end of the setup stack/heap, minus 0x0200.
439 Field name: cmd_line_ptr
440 Type: write (obligatory)
444 Set this field to the linear address of the kernel command line.
445 The kernel command line can be located anywhere between the end of
446 the setup heap and 0xA0000; it does not have to be located in the
447 same 64K segment as the real-mode code itself.
449 Fill in this field even if your boot loader does not support a
450 command line, in which case you can point this to an empty string
451 (or better yet, to the string "auto".) If this field is left at
452 zero, the kernel will assume that your boot loader does not support
455 Field name: initrd_addr_max
460 The maximum address that may be occupied by the initial
461 ramdisk/ramfs contents. For boot protocols 2.02 or earlier, this
462 field is not present, and the maximum address is 0x37FFFFFF. (This
463 address is defined as the address of the highest safe byte, so if
464 your ramdisk is exactly 131072 bytes long and this field is
465 0x37FFFFFF, you can start your ramdisk at 0x37FE0000.)
467 Field name: kernel_alignment
472 Alignment unit required by the kernel (if relocatable_kernel is true.)
474 Field name: relocatable_kernel
479 If this field is nonzero, the protected-mode part of the kernel can
480 be loaded at any address that satisfies the kernel_alignment field.
481 After loading, the boot loader must set the code32_start field to
482 point to the loaded code, or to a boot loader hook.
484 Field name: cmdline_size
489 The maximum size of the command line without the terminating
490 zero. This means that the command line can contain at most
491 cmdline_size characters. With protocol version 2.05 and earlier, the
492 maximum size was 255.
494 Field name: hardware_subarch
499 In a paravirtualized environment the hardware low level architectural
500 pieces such as interrupt handling, page table handling, and
501 accessing process control registers needs to be done differently.
503 This field allows the bootloader to inform the kernel we are in one
504 one of those environments.
506 0x00000000 The default x86/PC environment
510 Field name: hardware_subarch_data
515 A pointer to data that is specific to hardware subarch
517 Field name: compressed_payload_offset
522 If non-zero then this field contains the offset from the end of the
523 real-mode code to the compressed payload. The compression format
524 should be determined using the standard magic number, currently only
527 Field name: compressed_payload_length
532 The length of the compressed payload.
534 **** THE KERNEL COMMAND LINE
536 The kernel command line has become an important way for the boot
537 loader to communicate with the kernel. Some of its options are also
538 relevant to the boot loader itself, see "special command line options"
541 The kernel command line is a null-terminated string. The maximum
542 length can be retrieved from the field cmdline_size. Before protocol
543 version 2.06, the maximum was 255 characters. A string that is too
544 long will be automatically truncated by the kernel.
546 If the boot protocol version is 2.02 or later, the address of the
547 kernel command line is given by the header field cmd_line_ptr (see
548 above.) This address can be anywhere between the end of the setup
551 If the protocol version is *not* 2.02 or higher, the kernel
552 command line is entered using the following protocol:
554 At offset 0x0020 (word), "cmd_line_magic", enter the magic
557 At offset 0x0022 (word), "cmd_line_offset", enter the offset
558 of the kernel command line (relative to the start of the
561 The kernel command line *must* be within the memory region
562 covered by setup_move_size, so you may need to adjust this
566 **** MEMORY LAYOUT OF THE REAL-MODE CODE
568 The real-mode code requires a stack/heap to be set up, as well as
569 memory allocated for the kernel command line. This needs to be done
570 in the real-mode accessible memory in bottom megabyte.
572 It should be noted that modern machines often have a sizable Extended
573 BIOS Data Area (EBDA). As a result, it is advisable to use as little
574 of the low megabyte as possible.
576 Unfortunately, under the following circumstances the 0x90000 memory
577 segment has to be used:
579 - When loading a zImage kernel ((loadflags & 0x01) == 0).
580 - When loading a 2.01 or earlier boot protocol kernel.
582 -> For the 2.00 and 2.01 boot protocols, the real-mode code
583 can be loaded at another address, but it is internally
584 relocated to 0x90000. For the "old" protocol, the
585 real-mode code must be loaded at 0x90000.
587 When loading at 0x90000, avoid using memory above 0x9a000.
589 For boot protocol 2.02 or higher, the command line does not have to be
590 located in the same 64K segment as the real-mode setup code; it is
591 thus permitted to give the stack/heap the full 64K segment and locate
592 the command line above it.
594 The kernel command line should not be located below the real-mode
595 code, nor should it be located in high memory.
598 **** SAMPLE BOOT CONFIGURATION
600 As a sample configuration, assume the following layout of the real
603 When loading below 0x90000, use the entire segment:
605 0x0000-0x7fff Real mode kernel
606 0x8000-0xdfff Stack and heap
607 0xe000-0xffff Kernel command line
609 When loading at 0x90000 OR the protocol version is 2.01 or earlier:
611 0x0000-0x7fff Real mode kernel
612 0x8000-0x97ff Stack and heap
613 0x9800-0x9fff Kernel command line
615 Such a boot loader should enter the following fields in the header:
617 unsigned long base_ptr; /* base address for real-mode segment */
619 if ( setup_sects == 0 ) {
623 if ( protocol >= 0x0200 ) {
624 type_of_loader = <type code>;
625 if ( loading_initrd ) {
626 ramdisk_image = <initrd_address>;
627 ramdisk_size = <initrd_size>;
630 if ( protocol >= 0x0202 && loadflags & 0x01 )
635 if ( protocol >= 0x0201 ) {
636 heap_end_ptr = heap_end - 0x200;
637 loadflags |= 0x80; /* CAN_USE_HEAP */
640 if ( protocol >= 0x0202 ) {
641 cmd_line_ptr = base_ptr + heap_end;
642 strcpy(cmd_line_ptr, cmdline);
644 cmd_line_magic = 0xA33F;
645 cmd_line_offset = heap_end;
646 setup_move_size = heap_end + strlen(cmdline)+1;
647 strcpy(base_ptr+cmd_line_offset, cmdline);
650 /* Very old kernel */
654 cmd_line_magic = 0xA33F;
655 cmd_line_offset = heap_end;
657 /* A very old kernel MUST have its real-mode code
660 if ( base_ptr != 0x90000 ) {
661 /* Copy the real-mode kernel */
662 memcpy(0x90000, base_ptr, (setup_sects+1)*512);
663 base_ptr = 0x90000; /* Relocated */
666 strcpy(0x90000+cmd_line_offset, cmdline);
668 /* It is recommended to clear memory up to the 32K mark */
669 memset(0x90000 + (setup_sects+1)*512, 0,
670 (64-(setup_sects+1))*512);
674 **** LOADING THE REST OF THE KERNEL
676 The 32-bit (non-real-mode) kernel starts at offset (setup_sects+1)*512
677 in the kernel file (again, if setup_sects == 0 the real value is 4.)
678 It should be loaded at address 0x10000 for Image/zImage kernels and
679 0x100000 for bzImage kernels.
681 The kernel is a bzImage kernel if the protocol >= 2.00 and the 0x01
682 bit (LOAD_HIGH) in the loadflags field is set:
684 is_bzImage = (protocol >= 0x0200) && (loadflags & 0x01);
685 load_address = is_bzImage ? 0x100000 : 0x10000;
687 Note that Image/zImage kernels can be up to 512K in size, and thus use
688 the entire 0x10000-0x90000 range of memory. This means it is pretty
689 much a requirement for these kernels to load the real-mode part at
690 0x90000. bzImage kernels allow much more flexibility.
693 **** SPECIAL COMMAND LINE OPTIONS
695 If the command line provided by the boot loader is entered by the
696 user, the user may expect the following command line options to work.
697 They should normally not be deleted from the kernel command line even
698 though not all of them are actually meaningful to the kernel. Boot
699 loader authors who need additional command line options for the boot
700 loader itself should get them registered in
701 Documentation/kernel-parameters.txt to make sure they will not
702 conflict with actual kernel options now or in the future.
705 <mode> here is either an integer (in C notation, either
706 decimal, octal, or hexadecimal) or one of the strings
707 "normal" (meaning 0xFFFF), "ext" (meaning 0xFFFE) or "ask"
708 (meaning 0xFFFD). This value should be entered into the
709 vid_mode field, as it is used by the kernel before the command
713 <size> is an integer in C notation optionally followed by
714 (case insensitive) K, M, G, T, P or E (meaning << 10, << 20,
715 << 30, << 40, << 50 or << 60). This specifies the end of
716 memory to the kernel. This affects the possible placement of
717 an initrd, since an initrd should be placed near end of
718 memory. Note that this is an option to *both* the kernel and
722 An initrd should be loaded. The meaning of <file> is
723 obviously bootloader-dependent, and some boot loaders
724 (e.g. LILO) do not have such a command.
726 In addition, some boot loaders add the following options to the
727 user-specified command line:
730 The boot image which was loaded. Again, the meaning of <file>
731 is obviously bootloader-dependent.
734 The kernel was booted without explicit user intervention.
736 If these options are added by the boot loader, it is highly
737 recommended that they are located *first*, before the user-specified
738 or configuration-specified command line. Otherwise, "init=/bin/sh"
739 gets confused by the "auto" option.
742 **** RUNNING THE KERNEL
744 The kernel is started by jumping to the kernel entry point, which is
745 located at *segment* offset 0x20 from the start of the real mode
746 kernel. This means that if you loaded your real-mode kernel code at
747 0x90000, the kernel entry point is 9020:0000.
749 At entry, ds = es = ss should point to the start of the real-mode
750 kernel code (0x9000 if the code is loaded at 0x90000), sp should be
751 set up properly, normally pointing to the top of the heap, and
752 interrupts should be disabled. Furthermore, to guard against bugs in
753 the kernel, it is recommended that the boot loader sets fs = gs = ds =
756 In our example from above, we would do:
758 /* Note: in the case of the "old" kernel protocol, base_ptr must
759 be == 0x90000 at this point; see the previous sample code */
763 cli(); /* Enter with interrupts disabled! */
765 /* Set up the real-mode kernel stack */
769 _DS = _ES = _FS = _GS = seg;
770 jmp_far(seg+0x20, 0); /* Run the kernel */
772 If your boot sector accesses a floppy drive, it is recommended to
773 switch off the floppy motor before running the kernel, since the
774 kernel boot leaves interrupts off and thus the motor will not be
775 switched off, especially if the loaded kernel has the floppy driver as
776 a demand-loaded module!
779 **** ADVANCED BOOT LOADER HOOKS
781 If the boot loader runs in a particularly hostile environment (such as
782 LOADLIN, which runs under DOS) it may be impossible to follow the
783 standard memory location requirements. Such a boot loader may use the
784 following hooks that, if set, are invoked by the kernel at the
785 appropriate time. The use of these hooks should probably be
786 considered an absolutely last resort!
788 IMPORTANT: All the hooks are required to preserve %esp, %ebp, %esi and
789 %edi across invocation.
792 A 16-bit real mode far subroutine invoked immediately before
793 entering protected mode. The default routine disables NMI, so
794 your routine should probably do so, too.
797 A 32-bit flat-mode routine *jumped* to immediately after the
798 transition to protected mode, but before the kernel is
799 uncompressed. No segments, except CS, are guaranteed to be
800 set up (current kernels do, but older ones do not); you should
801 set them up to BOOT_DS (0x18) yourself.
803 After completing your hook, you should jump to the address
804 that was in this field before your boot loader overwrote it
805 (relocated, if appropriate.)
808 **** 32-bit BOOT PROTOCOL
810 For machine with some new BIOS other than legacy BIOS, such as EFI,
811 LinuxBIOS, etc, and kexec, the 16-bit real mode setup code in kernel
812 based on legacy BIOS can not be used, so a 32-bit boot protocol needs
815 In 32-bit boot protocol, the first step in loading a Linux kernel
816 should be to setup the boot parameters (struct boot_params,
817 traditionally known as "zero page"). The memory for struct boot_params
818 should be allocated and initialized to all zero. Then the setup header
819 from offset 0x01f1 of kernel image on should be loaded into struct
820 boot_params and examined. The end of setup header can be calculated as
823 0x0202 + byte value at offset 0x0201
825 In addition to read/modify/write the setup header of the struct
826 boot_params as that of 16-bit boot protocol, the boot loader should
827 also fill the additional fields of the struct boot_params as that
828 described in zero-page.txt.
830 After setupping the struct boot_params, the boot loader can load the
831 32/64-bit kernel in the same way as that of 16-bit boot protocol.
833 In 32-bit boot protocol, the kernel is started by jumping to the
834 32-bit kernel entry point, which is the start address of loaded
837 At entry, the CPU must be in 32-bit protected mode with paging
838 disabled; a GDT must be loaded with the descriptors for selectors
839 __BOOT_CS(0x10) and __BOOT_DS(0x18); both descriptors must be 4G flat
840 segment; __BOOS_CS must have execute/read permission, and __BOOT_DS
841 must have read/write permission; CS must be __BOOT_CS and DS, ES, SS
842 must be __BOOT_DS; interrupt must be disabled; %esi must hold the base
843 address of the struct boot_params; %ebp, %edi and %ebx must be zero.