Commit Graph

54 Commits

Author SHA1 Message Date
wiz ee1b406595 Spell address with two d's. Inspired by similar changes in OpenBSD,
originating from Jonathon Gray and forwarded by jmc@openbsd.
2003-11-10 08:51:51 +00:00
yamt d59161d069 - don't try to write alt super blocks beyond the end of the partition.
- flush stdout when printing each super block addresses.
2003-10-09 14:28:34 +00:00
itojun e2c411ce3c use arc4random 2003-09-06 12:42:00 +00:00
dsl b4c1c6117c Stop superblock being written to sector 0 for small filesystems.
Honour the '-s size' command option.
Allow an LFS filesystem to be created in a file.
2003-08-12 08:41:36 +00:00
agc 276d62f603 Move UCB-licensed code from 4-clause to 3-clause licence.
Patches provided by Joel Baker in PR 22308, verified by myself.
2003-08-07 10:04:22 +00:00
itojun d90a2369ab use bounded string ops 2003-07-13 07:39:39 +00:00
fvdl 42614ed3f3 Add support for UFS2. UFS2 is an enhanced FFS, adding support for
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.
2003-04-02 10:39:19 +00:00
simonb 668973e677 Use the PRId64 format to print a daddr_t. 2003-02-23 05:21:18 +00:00
perseant 0cfa2c5369 Discourage the creation of new LFSv1 filesystems. 2003-02-23 04:53:17 +00:00
perseant 8685c52d63 Make the "-O" (start filesystem offset) flag to newfs_lfs work correctly,
and update fsck_lfs and dumplfs to deal with it.  Note that while the argument
to -O is given in disk sectors, it must be a multiple of the fragment size,
and although it can be lower than the label or superblock, it can't intersect
either.
2003-02-23 04:32:05 +00:00
fvdl a3ff3a3038 Bump daddr_t to 64 bits. Replace it with int32_t in all places where
it was used on-disk, so that on-disk formats remain the same.
Remove ufs_daddr_t and ufs_lbn_t for the time being.
2003-01-24 21:55:02 +00:00
scw 6054b38a2b Use getlabeloffset() instead of LABELOFFSET. 2002-12-12 11:40:17 +00:00
wiz 2fb4b1db52 New sentence, new line. By Robert Elz with minimal fixes. 2002-10-01 13:40:23 +00:00
perseant d484b67904 Don't use adaptive measuring techniques to discover a good segment size,
if the device is not a character device (they will give bogus results
if they interact with the buffer cache).
2002-08-29 19:51:58 +00:00
lukem 7360d7b6ae Use ${NETBSDSRCDIR}/some/path instead of ${.CURDIR}/../../some/path 2002-08-19 10:16:51 +00:00
wiz c632c8bea4 __STDC__ is always defined on NetBSD. 2002-05-25 23:45:12 +00:00
wiz 65c4d36952 Use mdoc macros instead of man ones. 2002-01-21 18:23:09 +00:00
wiz 39304f2bbb Punctuation fixes, sort SEE ALSO, use an mdoc macros instead of two man macros. 2001-11-16 12:05:00 +00:00
wiz 4ce43ae0cd Whitespace nits 2001-11-16 10:25:42 +00:00
lukem 3d6fc56715 fix -Wshadow warning by moving "int version" from global to main() scope 2001-11-01 07:44:05 +00:00
mason 44c196b374 The default version created by newfs_lfs without arguments is, in fact, 2. 2001-09-08 03:49:44 +00:00
thorpej 8f43f0ca06 Fix printf format on LP64. 2001-07-13 21:09:55 +00:00
perseant 4e3fced95b Merge the short-lived perseant-lfsv2 branch into the trunk.
Kernels and tools understand both v1 and v2 filesystems; newfs_lfs
generates v2 by default.  Changes for the v2 layout include:

- Segments of non-PO2 size and arbitrary block offset, so these can be
  matched to convenient physical characteristics of the partition (e.g.,
  stripe or track size and offset).

- Address by fragment instead of by disk sector, paving the way for
  non-512-byte-sector devices.  In theory fragments can be as large
  as you like, though in reality they must be smaller than MAXBSIZE in size.

- Use serial number and filesystem identifier to ensure that roll-forward
  doesn't get old data and think it's new.  Roll-forward is enabled for
  v2 filesystems, though not for v1 filesystems by default.

- The inode free list is now a tailq, paving the way for undelete (undelete
  is not yet implemented, but can be without further non-backwards-compatible
  changes to disk structures).

- Inode atime information is kept in the Ifile, instead of on the inode;
  that is, the inode is never written *just* because atime was changed.
  Because of this the inodes remain near the file data on the disk, rather
  than wandering all over as the disk is read repeatedly.  This speeds up
  repeated reads by a small but noticeable amount.

Other changes of note include:

- The ifile written by newfs_lfs can now be of arbitrary length, it is no
  longer restricted to a single indirect block.

- Fixed an old bug where ctime was changed every time a vnode was created.
  I need to look more closely to make sure that the times are only updated
  during write(2) and friends, not after-the-fact during a segment write,
  and certainly not by the cleaner.
2001-07-13 20:30:18 +00:00
wiz 73f545bb5b Drop arguments of .Os. 2001-06-05 11:22:41 +00:00
perseant 8a3e9fdf9d Add "-A" option to autoconfigure the segment size based on measured
bandwidth and seek time of the disk, using the "4 * bandwidth * seek
time" formula from Neefe-Matthews' 1997 paper.  An RZ25 disk with this
option gets 200K segments.  Reference the paper in the manual page.
2000-12-05 19:51:14 +00:00
perseant 262e70b7e7 Initialize inodes' generation number to 1, consistent with their version
number.
2000-11-23 23:09:57 +00:00
hubertf 1320b80b0e xref lfs_cleanerd(8) in SEE ALSO 2000-11-08 19:43:16 +00:00
hubertf 4fdbe215a1 In SEE ALSO:
* it's dump_lfs(8), not dumplfs(8)
 * xref mount_lfs(8)
2000-11-08 19:37:46 +00:00
he 5b7047ce18 The type of sizeof() can be u_long, so cast to that and print with %ld. 2000-10-11 21:08:54 +00:00
he e3ab5a24f9 Format fixup. size_t can be long, so cast to u_long and print with %ld. 2000-10-11 21:02:05 +00:00
is 9979da6cbb Format string cleanups by Bill Sommerfeld. 2000-10-10 20:24:49 +00:00
perseant 7d6de39e20 Handle "-m 0" properly (PR #11112). 2000-10-02 16:06:48 +00:00
perseant 15c84d56e5 Do not report write errors if the user specified the -N flag.
Also, change the segment size report to include the total size of the disk,
similar to newfs, e.g.

  newfs_lfs -N -F -B 65536 /dev/rsd0b
  272.7MB in 4363 segments of size 65536
  super-block backups (for fsck -b #) at:
  16, 55824, 111632, 167440, 223248, 279056, 334864, 390672, 446480, 502288
2000-09-11 01:32:21 +00:00
perseant 9c7f8050f4 Various bug-fixes to LFS, to wit:
Kernel:

* Add runtime quantity lfs_ravail, the number of disk-blocks reserved
  for writing.  Writes to the filesystem first reserve a maximum amount
  of blocks before their write is allowed to proceed; after the blocks
  are allocated the reserved total is reduced by a corresponding amount.

  If the lfs_reserve function cannot immediately reserve the requested
  number of blocks, the inode is unlocked, and the thread sleeps until
  the cleaner has made enough space available for the blocks to be
  reserved.  In this way large files can be written to the filesystem
  (or, smaller files can be written to a nearly-full but thoroughly
  clean filesystem) and the cleaner can still function properly.

* Remove explicit switching on dlfs_minfreeseg from the kernel code; it
  is now merely a fs-creation parameter used to compute dlfs_avail and
  dlfs_bfree (and used by fsck_lfs(8) to check their accuracy).  Its
  former role is better assumed by a properly computed dlfs_avail.

* Bounds-check inode numbers submitted through lfs_bmapv and lfs_markv.
  This prevents a panic, but, if the cleaner is feeding the filesystem
  the wrong data, you are still in a world of hurt.

* Cleanup: remove explicit references of DEV_BSIZE in favor of
  btodb()/dbtob().

lfs_cleanerd:

* Make -n mean "send N segments' blocks through a single call to
  lfs_markv".  Previously it had meant "clean N segments though N calls
  to lfs_markv, before looking again to see if more need to be cleaned".
  The new behavior gives better packing of direct data on disk with as
  little metadata as possible, largely alleviating the problem that the
  cleaner can consume more disk through inefficient use of metadata than
  it frees by moving dirty data away from clean "holes" to produce
  entirely clean segments.

* Make -b mean "read as many segments as necessary to write N segments
  of dirty data back to disk", rather than its former meaning of "read
  as many segments as necessary to free N segments worth of space".  The
  new meaning, combined with the new -n behavior described above,
  further aids in cleaning storage efficiency as entire segments can be
  written at once, using as few blocks as possible for segment summaries
  and inode blocks.

* Make the cleaner take note of segments which could not be cleaned due
  to error, and not attempt to clean them until they are entirely free
  of dirty blocks.  This prevents the case in which a cleanerd running
  with -n 1 and without -b (formerly the default) would spin trying
  repeatedly to clean a corrupt segment, while the remaining space
  filled and deadlocked the filesystem.

* Update the lfs_cleanerd manual page to describe all the options,
  including the changes mentioned here (in particular, the -b and -n
  flags were previously undocumented).

fsck_lfs:

* Check, and optionally fix, lfs_avail (to an exact figure) and
  lfs_bfree (within a margin of error) in pass 5.

newfs_lfs:

* Reduce the default dlfs_minfreeseg to 1/20 of the total segments.

* Add a warning if the sgs disklabel field is 16 (the default for FFS'
  cpg, but not usually desirable for LFS' sgs: 5--8 is a better range).

* Change the calculation of lfs_avail and lfs_bfree, corresponding to
  the kernel changes mentioned above.

mount_lfs:

* Add -N and -b options to pass corresponding -n and -b options to
  lfs_cleanerd.

* Default to calling lfs_cleanerd with "-b -n 4".


[All of these changes were largely tested in the 1.5 branch, with the
idea that they (along with previous un-pulled-up work) could be applied
to the branch while it was still in ALPHA2; however my test system has
experienced corruption on another filesystem (/dev/console has gone
missing :^), and, while I believe this unrelated to the LFS changes, I
cannot with good conscience request that the changes be pulled up.]
2000-09-09 04:49:54 +00:00
perseant 48d7c317e8 Tweak how lfs_avail is initialized, corresponding to changes in the
kernel.

Don't make more superblock segments than we have a record of in
the superblock; and print these out as we go, like newfs.

Add am "-M" flag to specify the number of reserved segments, with a note
in the man page not to use it.
2000-07-04 22:35:04 +00:00
thorpej a00a01e514 Default the root directory and lost+found directory to mode 0755,
as in FFS.
2000-07-03 21:51:05 +00:00
perseant 9a38f49c57 User-level changes corrseponding to my latest kernel changes.
newfs_lfs gives lfs_minfreeseg a value of 1/8 of the total segments on
the disk, based on rough empirical data, but this should be refined in
the future.
2000-07-03 01:49:11 +00:00
perseant 61d1fe8df3 User-level changes for filling the disk.
Set MINFREE to 80, since that's a more reasonable value according to the
literature than FFS' 90.  Remove a bunch of other unused FFS cruft from
config.h.

Initialize lfs_bfree correctly vis-a-vis MIN_FREE_SEGS, so the
filesystem doesn't over-represent the amount of free space it has.
Initialize lfs_dmeta so the kernel can estimate starting from a
reasonable value.
2000-06-27 21:06:24 +00:00
perseant c534307563 Count the Ifile indirect block too (part of PR#9357) 2000-05-23 18:17:20 +00:00
perseant 0b785d0c0d More corrections to newfs_lfs' handling of segment 0. 2000-05-18 19:45:46 +00:00
perseant 9f31357cd9 Make newfs_lfs use fragments when creating the root and lost+found
directories, rather than full blocks; corrects some minor misaccounting
for segment zero.
2000-05-17 20:24:12 +00:00
perseant 42bce177e0 Add -F flag to override newfs_lfs' preference for 4.4LFS-labelled partitions. 2000-02-12 23:58:09 +00:00
enami 3fb18954ab - The -B flag should be followed by logical segment size.
- Remove unnecessary comma at the end of SEE ALSO list.
2000-02-05 11:57:30 +00:00
perseant 88d28e9b98 Fix frag -> fsize typo 2000-01-19 21:27:08 +00:00
perseant 43a7fde6d9 newfs_lfs now recognizes a zero p_sgs field to mean "use the default segment
size" (for consistency with bsize/fsize, and since segment size == block
size is never a valid combination).

Updated the man page to include explicit reasonable values for fsize, bsize,
and sgs, at suggestion from Hubert Feyrer.
2000-01-18 21:57:59 +00:00
perseant f1dfdc6927 disklabel now understands the p_sgs partition field (shift to compute
segment size from block size).

newfs_lfs now reads the disklabel to find segment, block, and fragment
sizes.  Because reading this info from the wrong fs type could result in
very poor fs layout (e.g. ffs has "16" where the segshift would go,
resulting in 512-*megabyte* segments for 8K blocks), newfs_lfs refuses
to create a filesystem on a partition not labeled "4.4LFS".

Man pages for newfs_lfs updated to reflect this change.
2000-01-18 00:02:28 +00:00
hubertf 5471344299 Document the proper fstype for LFS. (If the "default" 4.2BSD is used,
newfs_lfs runs fine, but I get hangs when writing to the disk)
2000-01-16 00:44:59 +00:00
perseant f52d155148 Make newfs_lfs try increasingly smaller segment sizes down to 2*block size,
if the partition is so small that not enough segments can be created to
place the second superblock.  Addresses PR#7623.
1999-11-05 18:59:12 +00:00
perseant eda6e4cca4 Address PR bin/7623: if the filesystem is so small (or segment size so large)
that not enough segments are available for the second superblock, or to have
MIN_FREE_SEGS free for work room for the cleaner, newfs_lfs will now exit
with an error.
1999-07-22 20:14:29 +00:00
perseant 7fa6616468 Address PR 7412: fix newfs_lfs.8 to list all real options for newfs_lfs;
also fix newfs_lfs to get rid of all sorts of useless options that applied
only to newfs_ffs.  Corrected reference to the FFS paper to the reference
to the BSD-LFS paper.
1999-07-15 19:09:40 +00:00