bootxx_cd9660 (cdboot), like bootxx_fat16 (fatboot), is not a traditional
libsa-based program. As such, build bootxx_cd9660 where its source lives.
This has been done by moving bootxx/bootxx_cd9660/Makefile to cdboot/Makefile
and adjusting the relative paths appropriately, so as to minimize binary change.

Build and install bootxx_ext2fs, primary boot for ext2fs.

Ext2fs doesn't have enough free space (it has only 1KB)
to store this primary loader, but we can put it into
an independent small 'boot' partition as NetBSD/hp300 does
if it's really necessary.

It could be a fun project to add smaller pre-primary loader
that loads ~8KB primary loader for ext2fs, but probably it's unlikely
because there is a functional wheel named GRUB for ext2fs.

Introduce .WAIT into the cleandir rule for blasting the libsa/libkern/etc.
trees, because it can race with the libsa/libkern/etc. makefiles' own
cleandir rules. I think I've found all of the uses of the offending idiom...
Closes PR 43360.

branches: 1.11.56; 1.11.70; 1.11.76; 1.11.78;
use "=" instead of ":=" where .OBJDIR is used because it might change
later in <bsd.obj.mk>, thanks to Alan Barrett for the hint

branches: 1.10.20; 1.10.22;
merge ktrace-lwp.

Ensure we don't build the libraries twice: move the .WAIT after
the first subdirectory that uses the libraries. That subdirectory
is bootxx_ffsv1, not bootxx_cd9660.

Build and install /usr/mdec/bootxx_cd9660, a primary boot loader
for ISO9660 filesystems.

branches: 1.7.12;
Need to use := when evaluating ${.OBJDIR}, then the correct value
is passed into the subdir builds and the 'lib' objects are correctly shared.
Fixes PR/26554

Add .WAIT after first subdirectory to ensure we don't build the libraries
twice.

Add bsd.obj.mk so these directories get their own obj directory (for the
library objects shared by their subdirectories).
Needed when the obj directory isn't a parent of the subdirectories obj directory

Change the way the shared lib directory is handled.
Should now work if ${.OBJDIR} = ${.CURDIR}/obj.

Overhaul MBR handling (part 1):

<sys/bootblock.h>:
* Added definitions for the Master Boot Record (MBR) used by
a variety of systems (primarily i386), including the format
of the BIOS Parameter Block (BPB).
This information was cribbed from a variety of sources
including <sys/disklabel_mbr.h> which this is a superset of.

As part of this, some data structure elements and #defines
were renamed to be more "namespace friendly" and consistent
with other bootblocks and MBR documentation.
Update all uses of the old names to the new names.

<sys/disklabel_mbr.h>:
* Deprecated in favor of <sys/bootblock.h> (the latter is more
"host tool" friendly).

amd64 & i386:
* Renamed /usr/mdec/bootxx_dosfs to /usr/mdec/bootxx_msdos, to
be consistent with the naming convention of the msdosfs tools.

* Removed /usr/mdec/bootxx_ufs, as it's equivalent to bootxx_ffsv1
and it's confusing to have two functionally equivalent bootblocks,
especially given that "ufs" has multiple meanings (it could be
a synonym for "ffs", or the group of ffs/lfs/ext2fs file systems).

* Rework pbr.S (the first sector of bootxx_*):
+ Ensure that BPB (bytes 11..89) and the partition table
(bytes 446..509) do not contain code.
+ Add support for booting from FAT partitions if BOOT_FROM_FAT
is defined. (Only set for bootxx_msdos).
+ Remove "dummy" partition 3; if people want to installboot(8)
these to the start of the disk they can use fdisk(8) to
create a real MBR partition table...
+ Compile with TERSE_ERROR so it fits because of the above.
Whilst this is less user friendly, I feel it's important
to have a valid partition table and BPB in the MBR/PBR.

* Renamed /usr/mdec/biosboot to /usr/mdec/boot, to be consistent
with other platforms.

* Enable SUPPORT_DOSFS in /usr/mdec/boot (stage2), so that
we can boot off FAT partitions.

* Crank version of /usr/mdec/boot to 3.1, and fix some of the other
entries in the version file.

installboot(8) (i386):
* Read the existing MBR of the filesystem and retain the BIOS
Parameter Block (BPB) in bytes 11..89 and the MBR partition
table in bytes 446..509. (Previously installboot(8) would
trash those two sections of the MBR.)

mbrlabel(8):
* Use sys/lib/libkern/xlat_mbr_fstype.c instead of homegrown code
to map the MBR partition type to the NetBSD disklabel type.


Test built "make release" for i386, and new bootblocks verified to work
(even off FAT!).

branches: 1.2.2;
Remove "lib" directory in "cleandir" target.

bootxx: bootstrap code - installable by MI installboot.
Loads /boot using filesystem accesses (rather than a list of block numbers)

bsd.own.mk: rename to CC_WNO_ADDRESS_OF_PACKED_MEMBER

Provide a single variable
CC_WNO_ADDRESS_OF_PACKED_MEMBER
with options for both clang and gcc, to replace
CLANG_NO_ADDR_OF_PACKED_MEMBER
CC_NO_ADDR_OF_PACKED_MEMBER
GCC_NO_ADDR_OF_PACKED_MEMBER

Using the convention CC_compilerflag, where compilerflag
is based on the full compiler flag name.

add support for new GCC 9 warnings that may be too much to fix
right now. new address-of-packed-member and format-overflow
warnings have new GCC_NO_ADDR_OF_PACKED_MEMBER amd
GCC_NO_FORMAT_OVERFLOW variables to remove these warnings.

apply to a bunch of the tree. mostly, these are real bugs that
should be fixed, but in many cases, only by removing the 'packed'
attribute from some structure that doesn't really need it. (i
looked at many different ones, and while perhaps 60-80% were
already properly aligned, it wasn't clear to me that the uses
were always coming from sane data vs network alignment, so it
doesn't seem safe to remove packed without careful research for
each affect struct.) clang already warned (and was not erroring)
for many of these cases, but gcc picked up dozens more.

Specify NOLIBCSANITIZER in x86 bootloader-like code under sys/arch/

Set NOLIBCSANITIZER for i386 and amd64 specific bootloader-like code.

branches: 1.50.2;
- Disable MKSANITIZER
- Redo using NOPIE
- Add NORELRO

branches: 1.49.16;
${PROG} depends on ${LDSCRIPT}.

branches: 1.48.6;
Reduce amount of -no-integrated-as on x86 as .code16 is now supported by
LLVM.

Add empty LIBCRTI= as LIBCRT0 to build sa programs without installed DESTDIR.

XXX: probabry we should have bsd.saprog.mk or something.

provide an ldscript to strip the unwind sections
XXX: untested

Add a missing KLINK_MACHINE= i386

Use <bsd.klinks.mk>

branches: 1.43.2; 1.43.4;
Don't depend on HAVE_GCC being always present.

branches: 1.42.2;
remove all the code that supported HAVE_GCC=3. mostly from chuq.

Refactor compiler-specific optimizer flags. Mark bootxx as supported
with clang.

branches: 1.40.2;
Disable integrated assembler for files that use .code16 or .code32 for
now. Disable Clang completely for bootxx and netboot for now until the
size issue has been sorted out.

branches: 1.39.2;
Introduce .WAIT into the cleandir rule for blasting the libsa/libkern/etc.
trees, because it can race with the libsa/libkern/etc. makefiles' own
cleandir rules. I think I've found all of the uses of the offending idiom...
Closes PR 43360.

branches: 1.38.2; 1.38.4;
Move relocation address to lower memory (0x1000) further away from the
BIOS 40:0 segment than previously but giving more room for heap that 0x8800.

Change relay address for mbr and bootxx code to be 0x8800.
I'm not sure why I used 0x600, but I have a feeling that might
sometimes corrupt bios data.
0x8800 is far enough above 0x7e00 for a sector read to the latter address.

Remove obsolete LIBSA_USE_MEMCPY and LIBSA_USE_MEMSET.
They were removed from <lib/libsa/stand.h> on December 2007.

Remove extra trailing slash in ${S} path.

Prefer MACHINE_ARCH to MACHINE in some tests

branches: 1.33.2; 1.33.8;
disable PIE for boot code.

Wrap compiler-specific flags with HAVE_GCC and HAVE_PCC as necessary. Add a few flags for PCC.

branches: 1.31.4; 1.31.6; 1.31.10;
Add -momit-leaf-frame-pointer, saves a few bytes.

branches: 1.30.2; 1.30.6;
Add EPIA_HACK so that boot loaders work on EDEN 5000 processors.
This worked in NetBSD 3, but broke by NetBSD 4.

My offer of a development machine to fix this one and for all is still open!

branches: 1.29.2;
Merge the ppcoea-renovation branch to HEAD.

This branch was a major cleanup and rototill of many of the various OEA
cpu based PPC ports that focused on sharing as much code as possible
between the various ports to eliminate near-identical copies of files in
every tree. Additionally there is a new PIC system that unifies the
interface to interrupt code for all different OEA ppc arches. The work
for this branch was done by a variety of people, too long to list here.

TODO:
bebox still needs work to complete the transition to -renovation.
ofppc still needs a bunch of work, which I will be looking at.
ev64260 still needs to be renovated
amigappc was not attempted.

NOTES:
pmppc was removed as an arch, and moved to a evbppc target.

Remove -DBOOT_ELF64 to make build on amd64.

branches: 1.27.6; 1.27.14; 1.27.22; 1.27.24; 1.27.26;
Don't define UNALIGNED_ACCESS, it's not used anymore.

Add -DUNALIGNED_ACCESS here, libsa/dosfs.c seems to need it.
Probably it ought to come from some generic .h file, but I've no idea
which - nor if any existing define has the same effect.
In any case this saves oodles of bytes in bootxx_msdos.

branches: 1.25.2; 1.25.4;
nuke $DBG, since it contains -O2; suggested by uwe.

branches: 1.24.4; 1.24.6;
Use ${TOOL_STAT} to get the size, instead of ls | tr | cut.

Use -Wl,-Ttext,foo instead of -Ttext foo when using $(CC)

branches: 1.22.4;
Convert to using CC instead of LD, as LDFLAGS is for CC not LD.

support MAKEVERBOSE

- -mcpu=i386 is gone in GCC4
- we need -Wno-attributes (to avoid __packed__ warnings)
- use -Wno-pointer-sign
- remove some redundancy from pxeboot/Makefile

branches: 1.19.4; 1.19.6; 1.19.8; 1.19.12;
merge ktrace-lwp.

use a .BEGIN target to make the various symlinks,
the previous way had problems with parallel make.

branches: 1.17.2;
add dependall and realdepend to the targets that we need to call in the
beginning. XXX: The real fix (handle dependencies in .BEGIN) is forthcoming.

More .BEGIN lossage.

branches: 1.15.10;
More complete fix for overriding CPUFLAGS and also apply fix to bootxx,
from Valeriy Ushakov.

Ignore errors on some rm -rf's for platforms that aren't quite
POSIX compliant
and return errors on r/o source (FreeBSD) for -rf. Fixes PR#25022

branches: 1.13.2;
If a bootxx program already happened to be a multiple of 512 bytes in
length before padding, the shell command to conditionally pad it to
a multiple of 512 bytes returned a nonzero exit status, causing the
build to fail.

Instead of defining TERSE_ERROR by default (due to space constraints),
define NO_LBA_CHECK. bootxx_msdos still needs TERSE_ERROR because its
BPB is larger than the default.

Pick up X86_BOOT_MAGIC_* from bootblock.h
Move boot_params.S to lib (seems useful to use it in pxeboot).

Change the way the shared lib directory is handled.
Should now work if ${.OBJDIR} = ${.CURDIR}/obj.

Overhaul MBR handling (part 1):

<sys/bootblock.h>:
* Added definitions for the Master Boot Record (MBR) used by
a variety of systems (primarily i386), including the format
of the BIOS Parameter Block (BPB).
This information was cribbed from a variety of sources
including <sys/disklabel_mbr.h> which this is a superset of.

As part of this, some data structure elements and #defines
were renamed to be more "namespace friendly" and consistent
with other bootblocks and MBR documentation.
Update all uses of the old names to the new names.

<sys/disklabel_mbr.h>:
* Deprecated in favor of <sys/bootblock.h> (the latter is more
"host tool" friendly).

amd64 & i386:
* Renamed /usr/mdec/bootxx_dosfs to /usr/mdec/bootxx_msdos, to
be consistent with the naming convention of the msdosfs tools.

* Removed /usr/mdec/bootxx_ufs, as it's equivalent to bootxx_ffsv1
and it's confusing to have two functionally equivalent bootblocks,
especially given that "ufs" has multiple meanings (it could be
a synonym for "ffs", or the group of ffs/lfs/ext2fs file systems).

* Rework pbr.S (the first sector of bootxx_*):
+ Ensure that BPB (bytes 11..89) and the partition table
(bytes 446..509) do not contain code.
+ Add support for booting from FAT partitions if BOOT_FROM_FAT
is defined. (Only set for bootxx_msdos).
+ Remove "dummy" partition 3; if people want to installboot(8)
these to the start of the disk they can use fdisk(8) to
create a real MBR partition table...
+ Compile with TERSE_ERROR so it fits because of the above.
Whilst this is less user friendly, I feel it's important
to have a valid partition table and BPB in the MBR/PBR.

* Renamed /usr/mdec/biosboot to /usr/mdec/boot, to be consistent
with other platforms.

* Enable SUPPORT_DOSFS in /usr/mdec/boot (stage2), so that
we can boot off FAT partitions.

* Crank version of /usr/mdec/boot to 3.1, and fix some of the other
entries in the version file.

installboot(8) (i386):
* Read the existing MBR of the filesystem and retain the BIOS
Parameter Block (BPB) in bytes 11..89 and the MBR partition
table in bytes 446..509. (Previously installboot(8) would
trash those two sections of the MBR.)

mbrlabel(8):
* Use sys/lib/libkern/xlat_mbr_fstype.c instead of homegrown code
to map the MBR partition type to the NetBSD disklabel type.


Test built "make release" for i386, and new bootblocks verified to work
(even off FAT!).

Add -DBOOT_ELF64 for amd64.

Add -m elf_i386 to ${LD}, not ${LDFLAGS}, since bsd.lib.mk doesn't
include ${LDFLAGS} when using ld -x on an object file.

XXX what's the point of these ld -x rules anyway?

Report actual $MACHINE (so it will be amd64, not i386)
and filesystem type (could be useful)
in primary bootstrap banner message.

branches: 1.5.2;
Make this work without a populated destdir:
machine and x86 symlinks.
Include from right places.
Don't depend on LIBCRT0,etc.

Also for the bootxx programs, keep the ELF object during the build
process.

Install into /usr/mdec

x86_64 -> amd64

Uncomment line that deletes tempory file

bootxx: bootstrap code - installable by MI installboot.
Loads /boot using filesystem accesses (rather than a list of block numbers)

Primary bootstrap is now able to read a GPT inside RAIDframe.

Previously, primary bootstrap was able to boot on RAID-1 RAIDframe set
with the limitation that the FFS filesystem had to start at bloc 0 in the
RAID. That allowed inner RAID partitionning with a disklabel, but not with
a GPT.

When booting on a RAID-1 RAIDframe, primary bootstrap now first try a
filesystem at bloc 0 of the RAID as before. On failure, it tries to
read a GPT and load secondary bootstrap from, by priority;
1) the first partition with the bootme attribute set
2) the first partition of type FFS, LFS, CCD or CGD
3) the first partition present in the GPT

branches: 1.21.10;
boot1.c: remove a comment that's no longer relevant/correct

In r. 1.13, a check of the return value in fd was removed, and a comment
about this ("...so keep going") added. Then in r. 1.19, the fd return
value check was reinstated (as the true underlying errno could be masked
by the fstat() call), so there is no "keep going" happening anymore.

branches: 1.20.64; 1.20.76;
Support booting from GPT-partioned disks on PC-BIOS-compatible systems.

Much of the work in this commit was done by Mike Volokhov during GSoC 2009.

branches: 1.19.4;
If the open fails, don't call fstat, because this changes the errno to EINVAL
from eg. ENOENT.

Reinstate the 'banner'.
It is a useful diagnostic that the boot sequence is proceeding.

branches: 1.17.14;
Tidy up console output slightly.

Remove clause 3 and 4 from TNF licenses

branches: 1.15.16; 1.15.18; 1.15.20;
Merge the ppcoea-renovation branch to HEAD.

This branch was a major cleanup and rototill of many of the various OEA
cpu based PPC ports that focused on sharing as much code as possible
between the various ports to eliminate near-identical copies of files in
every tree. Additionally there is a new PIC system that unifies the
interface to interrupt code for all different OEA ppc arches. The work
for this branch was done by a variety of people, too long to list here.

TODO:
bebox still needs work to complete the transition to -renovation.
ofppc still needs a bunch of work, which I will be looking at.
ev64260 still needs to be renovated
amigappc was not attempted.

NOTES:
pmppc was removed as an arch, and moved to a evbppc target.

- Make i386 bootblocks boot amd64 kernels, and change to say "NetBSD/x86..."
- Don't bother booting a.out kernels any more.

branches: 1.13.8; 1.13.16; 1.13.26; 1.13.28; 1.13.30;
remove a bit of code and a label and explain why in a comment; no code size
change.

Fix a logic bug in the bootblock code.

OK'ed by christos@

shrink further.

shave 32 bytes.

branches: 1.9.20; 1.9.22;
merge ktrace-lwp.

branches: 1.8.2;
Define BIOSDISK_DEFAULT_SECSIZE in biosdisk_ll.h and replace BIOSDISK_SECSIZE
with it.

Fix build breakage.

Change interface between bootxx.S and boot1() - always linked together.
This allows boot1() to change the sector number (of the boot partition)
that bootxx.S passes through to boot2().
This means that boot2() will find the correct partition when boot1()
reads /boot from the 'a' partition instead of the mbr boot partition.
This all happens when you update a system that used a small 'wd0h' partition
to boot a raid1 set to the new bootcode. Deleting /boot from the 'wd0h'
partition will make the new bootcode find /boot and the root filesystem
inside the raid set.

Look for /boot in the 'a' partition (from the label in the mbr partition)
if it can't be found in a filesystem (or raid set) at the start of the
mbr partition.

Fix speling mistake in commennt.

Try looking RF_PROTECTED_SECTORS further down the partition if /boot
cannot be found.
Lets the system boot directly from a rad set.

Report actual $MACHINE (so it will be amd64, not i386)
and filesystem type (could be useful)
in primary bootstrap banner message.

branches: 1.1.2;
bootxx: bootstrap code - installable by MI installboot.
Loads /boot using filesystem accesses (rather than a list of block numbers)

Pick up X86_BOOT_MAGIC_* from bootblock.h
Move boot_params.S to lib (seems useful to use it in pxeboot).

branches: 1.1.2;
bootxx: bootstrap code - installable by MI installboot.
Loads /boot using filesystem accesses (rather than a list of block numbers)

pass errno through switch to protected mode.

Fix typo (s/supress/suppress/).

branches: 1.10.60;
Support booting from GPT-partioned disks on PC-BIOS-compatible systems.

Much of the work in this commit was done by Mike Volokhov during GSoC 2009.

branches: 1.9.14; 1.9.22;
Remove clause 3 and 4 from TNF licenses

branches: 1.8.74; 1.8.76; 1.8.78;
merge ktrace-lwp.

Increase space for boot parameters - needed for specifying serial port
IO address and long keyboard translation map.

Rename bootparams to boot_params for consistency.

Change interface between bootxx.S and boot1() - always linked together.
This allows boot1() to change the sector number (of the boot partition)
that bootxx.S passes through to boot2().
This means that boot2() will find the correct partition when boot1()
reads /boot from the 'a' partition instead of the mbr boot partition.
This all happens when you update a system that used a small 'wd0h' partition
to boot a raid1 set to the new bootcode. Deleting /boot from the 'wd0h'
partition will make the new bootcode find /boot and the root filesystem
inside the raid set.

Report errno value from boot1() - even though the standalone FS code
seems to translate everything into ENOENT.

Pick up X86_BOOT_MAGIC_* from bootblock.h
Move boot_params.S to lib (seems useful to use it in pxeboot).

Overhaul MBR handling (part 1):

<sys/bootblock.h>:
* Added definitions for the Master Boot Record (MBR) used by
a variety of systems (primarily i386), including the format
of the BIOS Parameter Block (BPB).
This information was cribbed from a variety of sources
including <sys/disklabel_mbr.h> which this is a superset of.

As part of this, some data structure elements and #defines
were renamed to be more "namespace friendly" and consistent
with other bootblocks and MBR documentation.
Update all uses of the old names to the new names.

<sys/disklabel_mbr.h>:
* Deprecated in favor of <sys/bootblock.h> (the latter is more
"host tool" friendly).

amd64 & i386:
* Renamed /usr/mdec/bootxx_dosfs to /usr/mdec/bootxx_msdos, to
be consistent with the naming convention of the msdosfs tools.

* Removed /usr/mdec/bootxx_ufs, as it's equivalent to bootxx_ffsv1
and it's confusing to have two functionally equivalent bootblocks,
especially given that "ufs" has multiple meanings (it could be
a synonym for "ffs", or the group of ffs/lfs/ext2fs file systems).

* Rework pbr.S (the first sector of bootxx_*):
+ Ensure that BPB (bytes 11..89) and the partition table
(bytes 446..509) do not contain code.
+ Add support for booting from FAT partitions if BOOT_FROM_FAT
is defined. (Only set for bootxx_msdos).
+ Remove "dummy" partition 3; if people want to installboot(8)
these to the start of the disk they can use fdisk(8) to
create a real MBR partition table...
+ Compile with TERSE_ERROR so it fits because of the above.
Whilst this is less user friendly, I feel it's important
to have a valid partition table and BPB in the MBR/PBR.

* Renamed /usr/mdec/biosboot to /usr/mdec/boot, to be consistent
with other platforms.

* Enable SUPPORT_DOSFS in /usr/mdec/boot (stage2), so that
we can boot off FAT partitions.

* Crank version of /usr/mdec/boot to 3.1, and fix some of the other
entries in the version file.

installboot(8) (i386):
* Read the existing MBR of the filesystem and retain the BIOS
Parameter Block (BPB) in bytes 11..89 and the MBR partition
table in bytes 446..509. (Previously installboot(8) would
trash those two sections of the MBR.)

mbrlabel(8):
* Use sys/lib/libkern/xlat_mbr_fstype.c instead of homegrown code
to map the MBR partition type to the NetBSD disklabel type.


Test built "make release" for i386, and new bootblocks verified to work
(even off FAT!).

branches: 1.1.2;
bootxx: bootstrap code - installable by MI installboot.
Loads /boot using filesystem accesses (rather than a list of block numbers)

merge ktrace-lwp.

Look for /boot in the 'a' partition (from the label in the mbr partition)
if it can't be found in a filesystem (or raid set) at the start of the
mbr partition.

branches: 1.1.2;
bootxx: bootstrap code - installable by MI installboot.
Loads /boot using filesystem accesses (rather than a list of block numbers)

Replace several magic numbers with macro to describe GPT's hybrid MBR boot.

fix various typos in comments.

revert, this has to do with the bootloader protocol version and should
stay the same until there is a reason for it to change.

change 60 to 70 which is the current release. Noticed by Rares Aioanei.

branches: 1.20.30;
Remove home grown GPT MBR handoff support in favor of T13 EDD-4 annex A,
which does the same thing in a more interoperable way.

Support booting from GPT-partioned disks on PC-BIOS-compatible systems.

Much of the work in this commit was done by Mike Volokhov during GSoC 2009.

Correct comments referencing PRIMARY_LOAD_ADDRESS to match current reality.

branches: 1.17.4;
Update 'oemname' to NetBSD60.

branches: 1.16.14;
Remove clause 3 and 4 from TNF licenses

branches: 1.15.34; 1.15.36; 1.15.38;
Change the default 'oemname' from NetBSD20 to NetBSD40

branches: 1.14.4; 1.14.8; 1.14.10;
s/apparant/apparent/, from Zafer.

branches: 1.13.20; 1.13.22;
merge ktrace-lwp.

branches: 1.12.8;
Fix comment, code is requesting CHS geometry, not max LBA sector

branches: 1.11.8;
Move mbr_bootsel from offset 404 to offset 400 in struct mbr_sector to
leave 4 bytes for the Windows NT Drive Serial Number (DSN) at 440-443
(as mbr_sector.mbr_dsn).

Ensure that all the MBR & PBR code reserves space for mbr_sector.mbr_dsn.

Leave the bootsel magic number at 444-445 as mbr_sector.mbr_bootsel_magic
(instead of mbr_sector.mbr_bootsel.mbrbs_magic), but use 0xb5e1 (MBR_BS_MAGIC)
instead of 0xaa55 (MBR_MAGIC) to indicate that this change has occurred.

Rework MBR_BS_NEWMBR to mean "mbr_bootsel has moved to 400".

Modify fdisk(8) to automatically relocate the mbr_bootsel from 404 to 400
if mbr_bootsel_magic is the old value (0xaa55), and unset MBR_BS_NEWMBR
to flag that new mbr_bootsel code must be used if updating the MBR.


These changes fixes a problem where Windows 2000 or Windows XP would corrupt
the last 3 bytes + NUL of MBR partition 3's bootsel name if the bootsel name
was 5 characters long, replacing bytes 6-9 with the DSN.
Also, by explicitly reserving the space for the DSN we prevent problems in the
future if non bootsel MBR or PBR code had other information at bytes 440-443.

Fix problems booting disks with extended partition from grub (and other
non-netbsd mbr code) - thanks to Joe Thiemann for finding this...

The reserved space for the BIOS Parameter Block (BPB) only needs to be for
FAT16 (11+51) except when booting from FAT{12,16,32}, which needs FAT32 (11+79).
We still reserve the BPB for non-bootxx_msdos PBR bootblocks because
they may be installed as a floppy boot record (and those need a BPB).

Remove some redundant wording in an error messsage, saving 6 bytes.

Overhaul MBR handling (part 1):

<sys/bootblock.h>:
* Added definitions for the Master Boot Record (MBR) used by
a variety of systems (primarily i386), including the format
of the BIOS Parameter Block (BPB).
This information was cribbed from a variety of sources
including <sys/disklabel_mbr.h> which this is a superset of.

As part of this, some data structure elements and #defines
were renamed to be more "namespace friendly" and consistent
with other bootblocks and MBR documentation.
Update all uses of the old names to the new names.

<sys/disklabel_mbr.h>:
* Deprecated in favor of <sys/bootblock.h> (the latter is more
"host tool" friendly).

amd64 & i386:
* Renamed /usr/mdec/bootxx_dosfs to /usr/mdec/bootxx_msdos, to
be consistent with the naming convention of the msdosfs tools.

* Removed /usr/mdec/bootxx_ufs, as it's equivalent to bootxx_ffsv1
and it's confusing to have two functionally equivalent bootblocks,
especially given that "ufs" has multiple meanings (it could be
a synonym for "ffs", or the group of ffs/lfs/ext2fs file systems).

* Rework pbr.S (the first sector of bootxx_*):
+ Ensure that BPB (bytes 11..89) and the partition table
(bytes 446..509) do not contain code.
+ Add support for booting from FAT partitions if BOOT_FROM_FAT
is defined. (Only set for bootxx_msdos).
+ Remove "dummy" partition 3; if people want to installboot(8)
these to the start of the disk they can use fdisk(8) to
create a real MBR partition table...
+ Compile with TERSE_ERROR so it fits because of the above.
Whilst this is less user friendly, I feel it's important
to have a valid partition table and BPB in the MBR/PBR.

* Renamed /usr/mdec/biosboot to /usr/mdec/boot, to be consistent
with other platforms.

* Enable SUPPORT_DOSFS in /usr/mdec/boot (stage2), so that
we can boot off FAT partitions.

* Crank version of /usr/mdec/boot to 3.1, and fix some of the other
entries in the version file.

installboot(8) (i386):
* Read the existing MBR of the filesystem and retain the BIOS
Parameter Block (BPB) in bytes 11..89 and the MBR partition
table in bytes 446..509. (Previously installboot(8) would
trash those two sections of the MBR.)

mbrlabel(8):
* Use sys/lib/libkern/xlat_mbr_fstype.c instead of homegrown code
to map the MBR partition type to the NetBSD disklabel type.


Test built "make release" for i386, and new bootblocks verified to work
(even off FAT!).

Placate people who insist on using non-standard disk layouts by adding
a valid MBR entry for partition 3 (covering the first 15 sectors) into the
end of the partition boot code.

Some systems (esp. those with a Promise IDE controller card) seem to
destroy %dl (drive number) across the 'disk reset' command.
Preserve %dl across that call and all registers across the disk reads.
Reorder the code to remove some long conditional jumps to save space.

Dont need jmpl, ljmp or jmp is fine.

Some bios manage to load the mbr code to 7c0:0 (instead of 0:7c00) especially
when loading from CDROM. Change so we don't care what %cs is.
Adjust code layout so that anyone who feels so inclinded can use this for
the mbr code. Note that mbr slot 0 overlays a data area that must be zero
when this is read from the pbr.

branches: 1.3.2;
Incorrect sector number passed to boot1 in floppy (no mbr) case

Fix code that scans extended partition list

bootxx: bootstrap code - installable by MI installboot.
Loads /boot using filesystem accesses (rather than a list of block numbers)

branches: 1.1.2; 1.1.4; 1.1.6; 1.1.12;
Build and install bootxx_ext2fs, primary boot for ext2fs.

Ext2fs doesn't have enough free space (it has only 1KB)
to store this primary loader, but we can put it into
an independent small 'boot' partition as NetBSD/hp300 does
if it's really necessary.

It could be a fun project to add smaller pre-primary loader
that loads ~8KB primary loader for ext2fs, but probably it's unlikely
because there is a functional wheel named GRUB for ext2fs.

bootxx: bootstrap code - installable by MI installboot.
Loads /boot using filesystem accesses (rather than a list of block numbers)

bootxx: bootstrap code - installable by MI installboot.
Loads /boot using filesystem accesses (rather than a list of block numbers)

bootxx: bootstrap code - installable by MI installboot.
Loads /boot using filesystem accesses (rather than a list of block numbers)

bootxx: bootstrap code - installable by MI installboot.
Loads /boot using filesystem accesses (rather than a list of block numbers)

Override these two booters with -Oz for clang since it produces smaller code
here.

branches: 1.4.24;
Clamp FAT partition size to 32 bit byte offsets to make the code small
enough to fit.

branches: 1.3.166; 1.3.172;
merge ktrace-lwp.

branches: 1.2.4;
Instead of defining TERSE_ERROR by default (due to space constraints),
define NO_LBA_CHECK. bootxx_msdos still needs TERSE_ERROR because its
BPB is larger than the default.

Overhaul MBR handling (part 1):

<sys/bootblock.h>:
* Added definitions for the Master Boot Record (MBR) used by
a variety of systems (primarily i386), including the format
of the BIOS Parameter Block (BPB).
This information was cribbed from a variety of sources
including <sys/disklabel_mbr.h> which this is a superset of.

As part of this, some data structure elements and #defines
were renamed to be more "namespace friendly" and consistent
with other bootblocks and MBR documentation.
Update all uses of the old names to the new names.

<sys/disklabel_mbr.h>:
* Deprecated in favor of <sys/bootblock.h> (the latter is more
"host tool" friendly).

amd64 & i386:
* Renamed /usr/mdec/bootxx_dosfs to /usr/mdec/bootxx_msdos, to
be consistent with the naming convention of the msdosfs tools.

* Removed /usr/mdec/bootxx_ufs, as it's equivalent to bootxx_ffsv1
and it's confusing to have two functionally equivalent bootblocks,
especially given that "ufs" has multiple meanings (it could be
a synonym for "ffs", or the group of ffs/lfs/ext2fs file systems).

* Rework pbr.S (the first sector of bootxx_*):
+ Ensure that BPB (bytes 11..89) and the partition table
(bytes 446..509) do not contain code.
+ Add support for booting from FAT partitions if BOOT_FROM_FAT
is defined. (Only set for bootxx_msdos).
+ Remove "dummy" partition 3; if people want to installboot(8)
these to the start of the disk they can use fdisk(8) to
create a real MBR partition table...
+ Compile with TERSE_ERROR so it fits because of the above.
Whilst this is less user friendly, I feel it's important
to have a valid partition table and BPB in the MBR/PBR.

* Renamed /usr/mdec/biosboot to /usr/mdec/boot, to be consistent
with other platforms.

* Enable SUPPORT_DOSFS in /usr/mdec/boot (stage2), so that
we can boot off FAT partitions.

* Crank version of /usr/mdec/boot to 3.1, and fix some of the other
entries in the version file.

installboot(8) (i386):
* Read the existing MBR of the filesystem and retain the BIOS
Parameter Block (BPB) in bytes 11..89 and the MBR partition
table in bytes 446..509. (Previously installboot(8) would
trash those two sections of the MBR.)

mbrlabel(8):
* Use sys/lib/libkern/xlat_mbr_fstype.c instead of homegrown code
to map the MBR partition type to the NetBSD disklabel type.


Test built "make release" for i386, and new bootblocks verified to work
(even off FAT!).

Override these two booters with -Oz for clang since it produces smaller code
here.

branches: 1.3.94;
Reinstate the 'banner'.
It is a useful diagnostic that the boot sequence is proceeding.

branches: 1.2.14;
Tidy up console output slightly.

branches: 1.1.104; 1.1.106; 1.1.108;
bootxx: bootstrap code - installable by MI installboot.
Loads /boot using filesystem accesses (rather than a list of block numbers)