arguments, including a clarification that `secondary' is relative to the root
of the filesystem, not a path name to that file if `filesystem' is mounted.
Resolves PR 20977 from Jim Bernard.
all the options have been processed that they are supported my the
specific system.
Add i386 MD options (console, password, speed and timeout)
(Approved by christos and lukem)
o Correct the order of arguments to ffs_read_disk_block; the second one
is blocknumber to read, not the size to be read. This would affect the
UFS2 code, and is thus not much excercised at the moment.
o The offsets in SBLOCKSEARCH are in bytes, but ffs_read_disk_block wants
its location in terms of DEV_BSIZE blocks. Fix this bug in ffs_match().
The latter together with the recent magic number fix to <ufs/ffs/fs.h>
fixes a cross build problem seen when doing a i386->sparc build.
64 bit block pointers, extended attribute storage, and a few
other things.
This commit does not yet include the code to manipulate the extended
storage (for e.g. ACLs), this will be done later.
Originally written by Kirk McKusick and Network Associates Laboratories for
FreeBSD.
bbinfo.c module; news* boot blocks occupy all of the first 16 sectors,
with a jump instruction to skip the label at bytes 64..511.
Replace news_clearboot() and news_setboot() callbacks with common
news_copydisklable() callback, and set bbinfo_params->offset to
NEWS_BOOT_BLOCK_OFFSET (which is now 0), and ->headeroffset to 0.
(Thanks to Izumi for picking this up; the perils of working on code at 2am)
- bbinfo_params:
- replace "int littleendian" with "bbinfo_endian endian"
- add comments
- shared_bbinfo_clearboot():
- add callback method to shared_bbinfo_clearboot()
- don't clear from 0..headeroffset; use a callback to do that
- add news68k and newsmips support.
From Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>, with a rework by me to
take advantage of the new shared_bbinfo_clearboot() callback.
(XXX: untested yet)
column gets a trailing "," if there's more entries to come.
- if getmachines() or getfstypes() is called with 3 NULL params, just
print the list of supported items.
- in usage(), call getmachines() and getfstypes() with 3 NULL params.
(i got sick of typing "installboot -m asdf" to get a list of
supported machines during testing :)
- unify sparc_bbinfo (1064 bytes, with 256 block entries)
and sun68k_bbinfo (296 byte, with 64 block entries)
into shared_bbinfo (512 bytes, with 118 block entries),
which will be also shared by future bbinfo-using platforms
(including macppc)
- add datestamp to *_BBINFO_MAGIC strings, to prevent installboot vs
bootxx version skew.
- add macppc support
*/bootxx.c:
- migrate to new shared_bbinfo structure
installboot:
- add macppc support (still needs applepartmap support and testing)
- improve and add some more warnings & errors to installboot
- implement shared_bbinfo_clearboot() and shared_bbinfo_setboot(), which
perform the majority of the work for bbinfo-using back-ends
(rather than replicating that across multiple back-ends).
- unify sparc_bbinfo (1064 bytes, with 256 block entries)
and sun68k_bbinfo (296 byte, with 64 block entries)
into shared_bbinfo (512 bytes, with 118 block entries),
which will be also shared by future bbinfo-using platforms
(including macppc)
- add datestamp to *_BBINFO_MAGIC strings, to prevent installboot vs
bootxx version skew.
- add macppc support
*/bootxx.c:
- migrate to new shared_bbinfo structure
installboot:
- add macppc support (still needs applepartmap support and testing)
- improve and add some more warnings & errors to installboot
- implement shared_bbinfo_clearboot() and shared_bbinfo_setboot(), which
perform the majority of the work for bbinfo-using back-ends
(rather than replicating that across multiple back-ends).
and use instead of replicating the effort in various locations
- if filesystem is not a regular file, use sync(2) instead of fsync(2)
after the bootstrap has been written
- move <sys/stat.h> and <stdint.h> into "installboot.h"
and use instead of replicating the effort in various locations
- if filesystem is not a regular file, use sync(2) instead of fsync(2)
after the bootstrap has been written
- add `-B s2bno', to provide the starting block for the secondary bootstrap.
intended for use on platforms where the blocks of the stage2 bootstrap
are hardcoded into the stage1 bootstrap (currently: sparc, sun2)
- don't support `-b s1bno' for sparc or sun2, since the primary is always
at a fixed location on the disk.
- if `filesystem' is a regular file, use fsync(2) instead of sync(2)
code changes:
- add hardcode_stage2(), which uses -B s2bno and the size of the
provided secondary bootstrap (as an actual file, not as part of the
`filesystem' argument) to provide a sequential list of blocks from s2bno,
each block being the appropriate file system size (from
params->fstype->blocksize)
- add blocksize and needswap run-time parameters to ib_fs
- in *_match(), set params->fstype->blocksize to the underlying block size
(8KB for raw), and params->fstype->needswap as appropriate
- rename IB_STARTBLOCK to IB_STAGE1START, and add IB_STAGE2START
- use hardcode_stage2() to implement raw_findstage2() and IB_STAGE2BLOCK
support for ffs_findstage2()
- improve some comments, add some prototypes, ...
lukem