Commit Graph

121 Commits

Author SHA1 Message Date
Ian Dowse
9710700cb1 Remove the ancient STANDALONE code.
Approved by:	phk
2002-03-19 16:47:20 +00:00
Ian Dowse
af53d6d86e Remove yet more vestiges of mount_mfs. 2002-03-18 15:31:44 +00:00
Bruce Evans
0c08079e26 Fixed some style bugs (mainly ones not fixed or made worse by rev.1.44).
Don't use ISO string concatentation to obfuscate long single-line
messages...
2002-03-18 03:04:58 +00:00
Bruce Evans
63dab85cea Fixed some style bugs (mainly ones not fixed or made worse by rev.1.41).
Old code obfuscates long (but single-line) messages by printing them in
pieces using %s.  Rev.1.41 obfuscated some new long messages using ISO
string concatenation.  This commit only fixes the new obfuscations.
2002-03-18 02:43:14 +00:00
Bruce Evans
3ac7f11229 Removed vestiges of mount_mfs. Sorted the Makefile a bit. 2002-03-18 02:23:43 +00:00
Bruce Evans
3f305db0a2 Fixed 2 layers of breakage of WARNS. Setting WARNS unconditionally to
0 was bad and setting it unconditionally to 2 was worse.
2002-03-18 02:13:38 +00:00
Poul-Henning Kamp
345b78a301 Remove __P() and register.
Set WARNS=2

This is the beginning of a pre-UFS2 cleanup of newfs.

Sponsored by:	DARPA, NAI Labs
2002-03-17 09:01:41 +00:00
Sheldon Hearn
3626f83327 Update the default newfs block and fragment sizes from 8192/1024 to
16384/2048.

Following recent discussions on the -arch mailing list, involving dillon
and mckusick, this change parallels the one made over a decade ago when
the default was bumped up from 4096/512.

This should provide significant performance improvements for most
folks, less significant performance losses for a few folks and
wasted space lost to large fragments for many folks.

For discussion, please see the following thread in the -arch archive:

Subject: Using a larger block size on large filesystems

The discussion ceases to be relevant when the issue of partitioning
schemes is raised.
2001-12-11 16:21:40 +00:00
Sheldon Hearn
be42c56328 Fix typo: 'fragement' -> 'fragment' 2001-12-07 13:18:28 +00:00
David E. O'Brien
2d68bf45bf Default to WARNS=2.
Binary builds that cannot handle this must explicitly set WARNS=0.

Reviewed by:	mike
2001-12-04 02:19:58 +00:00
Sheldon Hearn
629c25d4a9 Fix the example of suggested default settings. It stated that settings
were only of benefit to large filesystems, which recent research
suggests is not the case, and which the original author of the text
no longer endorses.
2001-11-27 19:39:07 +00:00
Sheldon Hearn
2441e154ef Correct the example introduced in rev 1.29, which suggested a block:frag
size ratio other than 8:1.  Currently, we only recommend an 8:1
ratio, because the impact of others ratios has not been adequately
investigated.

Also, do not recommend the use of the -c option in the example, since
newfs now automatically calculates the best cyl:cylgrp ratio.

This change was discussed with the author of rev 1.29.
2001-11-27 17:01:17 +00:00
Peter Wemm
1165817956 Remove support for FreeBSD/tahoe
Submitted by:	phk
2001-11-03 08:35:11 +00:00
Poul-Henning Kamp
bfd1f63d44 style(9) cleanup.
Submitted by:	j mckitrick <jcm@freebsd-uk.eu.org>
Reviewed by:	phk, /sbin/md5
2001-11-02 09:23:34 +00:00
Ollivier Robert
4f63c70a2b Fix diskless clients by removing the code for calculating the minimum
value for cpg. The change was bogus.

Submitted by:	bde
MFC after:	2 days
2001-10-18 09:48:28 +00:00
Doug Barton
eb9d1d2716 Document the optimal block:fragment ratio, per discussion
on -arch and cvs-all.

Reviewed by:	dillon
2001-10-15 22:47:55 +00:00
Ollivier Robert
08870345f5 Following the discussion in -arch and the submission of a patch by bde, here
it is. I added the manpage change.

Submitted by:	bde
MFC after:	1 week
2001-10-04 12:15:50 +00:00
Brian Somers
9cfe90fe1f Handle snprintf() returning < 0 (not just -1)
MFC after:	2 weeks
2001-08-20 14:53:05 +00:00
Brian Somers
327e849ae1 Handle snprintf() returning -1.
MFC after:	2 weeks
2001-08-20 12:56:45 +00:00
Kris Kennaway
5979df34a6 Silence non-constant format string warnings by marking functions
as __printflike()/__printf0like(), adding const, or adding missing "%s"
format strings, as appropriate.

MFC after:	2 weeks
2001-08-19 08:19:37 +00:00
Dima Dorfman
7ebcc426ef Remove whitespace at EOL. 2001-07-15 07:53:42 +00:00
Ruslan Ermilov
9fe48c6e8d mdoc(7) police: removed HISTORY info from the .Os call. 2001-07-10 11:04:34 +00:00
Dima Dorfman
70d51341bf mdoc(7) police: remove extraneous .Pp before and/or after .Sh. 2001-07-09 09:54:33 +00:00
Bruce Evans
e38c0bd643 Don't link ${BINDIR}/newfs to nowhere.
Don't clutter this Makefile (not to mention the error output) with
$(BDECFLAGS}.
2001-05-30 09:31:24 +00:00
Dima Dorfman
f628a4b10e Remove -DMFS from CFLAGS. 2001-05-29 23:57:23 +00:00
Dima Dorfman
7ccb741f6a Remove all references to MFS. 2001-05-29 23:55:43 +00:00
Poul-Henning Kamp
80f86e526b A more complete removal of MFS related code.
XXX: This program badly needs a style(9) + BDECFLAGS treatment.
2001-05-29 19:40:39 +00:00
Poul-Henning Kamp
a383ba34c2 Initial cleanout of MFS from newfs. More complete wash needed. 2001-05-29 18:52:39 +00:00
Kris Kennaway
1fef4cc97d sprintf() -> snprintf()
Partially submitted by:	"Andrew R. Reiter" <arr@watson.org>
Obtained from:	OpenBSD
2001-04-24 10:26:00 +00:00
Kris Kennaway
c3b1df1293 Add a missing argument to an error message format string. 2001-04-17 07:21:48 +00:00
Nik Clayton
044479f5ad Add information about the new options to newfs and tunefs which set the
expected average file size and number of files per directory.  Could do
with some fleshing out.
2001-04-10 10:36:44 +00:00
Kirk McKusick
a61ab64ac4 Directory layout preference improvements from Grigoriy Orlov <gluk@ptci.ru>.
His description of the problem and solution follow. My own tests show
speedups on typical filesystem intensive workloads of 5% to 12% which
is very impressive considering the small amount of code change involved.

------

  One day I noticed that some file operations run much faster on
small file systems then on big ones. I've looked at the ffs
algorithms, thought about them, and redesigned the dirpref algorithm.

  First I want to describe the results of my tests. These results are old
and I have improved the algorithm after these tests were done. Nevertheless
they show how big the perfomance speedup may be. I have done two file/directory
intensive tests on a two OpenBSD systems with old and new dirpref algorithm.
The first test is "tar -xzf ports.tar.gz", the second is "rm -rf ports".
The ports.tar.gz file is the ports collection from the OpenBSD 2.8 release.
It contains 6596 directories and 13868 files. The test systems are:

1. Celeron-450, 128Mb, two IDE drives, the system at wd0, file system for
   test is at wd1. Size of test file system is 8 Gb, number of cg=991,
   size of cg is 8m, block size = 8k, fragment size = 1k OpenBSD-current
   from Dec 2000 with BUFCACHEPERCENT=35

2. PIII-600, 128Mb, two IBM DTLA-307045 IDE drives at i815e, the system
   at wd0, file system for test is at wd1. Size of test file system is 40 Gb,
   number of cg=5324, size of cg is 8m, block size = 8k, fragment size = 1k
   OpenBSD-current from Dec 2000 with BUFCACHEPERCENT=50

You can get more info about the test systems and methods at:
http://www.ptci.ru/gluk/dirpref/old/dirpref.html

                              Test Results

             tar -xzf ports.tar.gz               rm -rf ports
  mode  old dirpref new dirpref speedup old dirprefnew dirpref speedup
                             First system
 normal     667         472      1.41       477        331       1.44
 async      285         144      1.98       130         14       9.29
 sync       768         616      1.25       477        334       1.43
 softdep    413         252      1.64       241         38       6.34
                             Second system
 normal     329         81       4.06       263.5       93.5     2.81
 async      302         25.7    11.75       112          2.26   49.56
 sync       281         57.0     4.93       263         90.5     2.9
 softdep    341         40.6     8.4        284          4.76   59.66

"old dirpref" and "new dirpref" columns give a test time in seconds.
speedup - speed increasement in times, ie. old dirpref / new dirpref.

------

Algorithm description

The old dirpref algorithm is described in comments:

/*
 * Find a cylinder to place a directory.
 *
 * The policy implemented by this algorithm is to select from
 * among those cylinder groups with above the average number of
 * free inodes, the one with the smallest number of directories.
 */

A new directory is allocated in a different cylinder groups than its
parent directory resulting in a directory tree that is spreaded across
all the cylinder groups. This spreading out results in a non-optimal
access to the directories and files. When we have a small filesystem
it is not a problem but when the filesystem is big then perfomance
degradation becomes very apparent.

What I mean by a big file system ?

  1. A big filesystem is a filesystem which occupy 20-30 or more percent
     of total drive space, i.e. first and last cylinder are physically
     located relatively far from each other.
  2. It has a relatively large number of cylinder groups, for example
     more cylinder groups than 50% of the buffers in the buffer cache.

The first results in long access times, while the second results in
many buffers being used by metadata operations. Such operations use
cylinder group blocks and on-disk inode blocks. The cylinder group
block (fs->fs_cblkno) contains struct cg, inode and block bit maps.
It is 2k in size for the default filesystem parameters. If new and
parent directories are located in different cylinder groups then the
system performs more input/output operations and uses more buffers.
On filesystems with many cylinder groups, lots of cache buffers are
used for metadata operations.

My solution for this problem is very simple. I allocate many directories
in one cylinder group. I also do some things, so that the new allocation
method does not cause excessive fragmentation and all directory inodes
will not be located at a location far from its file's inodes and data.
The algorithm is:
/*
 * Find a cylinder group to place a directory.
 *
 * The policy implemented by this algorithm is to allocate a
 * directory inode in the same cylinder group as its parent
 * directory, but also to reserve space for its files inodes
 * and data. Restrict the number of directories which may be
 * allocated one after another in the same cylinder group
 * without intervening allocation of files.
 *
 * If we allocate a first level directory then force allocation
 * in another cylinder group.
 */

  My early versions of dirpref give me a good results for a wide range of
file operations and different filesystem capacities except one case:
those applications that create their entire directory structure first
and only later fill this structure with files.

  My solution for such and similar cases is to limit a number of
directories which may be created one after another in the same cylinder
group without intervening file creations. For this purpose, I allocate
an array of counters at mount time. This array is linked to the superblock
fs->fs_contigdirs[cg]. Each time a directory is created the counter
increases and each time a file is created the counter decreases. A 60Gb
filesystem with 8mb/cg requires 10kb of memory for the counters array.

  The maxcontigdirs is a maximum number of directories which may be created
without an intervening file creation. I found in my tests that the best
performance occurs when I restrict the number of directories in one cylinder
group such that all its files may be located in the same cylinder group.
There may be some deterioration in performance if all the file inodes
are in the same cylinder group as its containing directory, but their
data partially resides in a different cylinder group. The maxcontigdirs
value is calculated to try to prevent this condition. Since there is
no way to know how many files and directories will be allocated later
I added two optimization parameters in superblock/tunefs. They are:

        int32_t  fs_avgfilesize;   /* expected average file size */
        int32_t  fs_avgfpdir;      /* expected # of files per directory */

These parameters have reasonable defaults but may be tweeked for special
uses of a filesystem. They are only necessary in rare cases like better
tuning a filesystem being used to store a squid cache.

I have been using this algorithm for about 3 months. I have done
a lot of testing on filesystems with different capacities, average
filesize, average number of files per directory, and so on. I think
this algorithm has no negative impact on filesystem perfomance. It
works better than the default one in all cases. The new dirpref
will greatly improve untarring/removing/coping of big directories,
decrease load on cvs servers and much more. The new dirpref doesn't
speedup a compilation process, but also doesn't slow it down.

Obtained from:	Grigoriy Orlov <gluk@ptci.ru>
2001-04-10 08:38:59 +00:00
Bruce Evans
5f98b5af89 Fixed style bugs in previous commit. 2001-04-03 09:35:36 +00:00
David E. O'Brien
8af4736afe Document the newfs.c rev 1.33 changing the default c/g from 16 to 22. 2001-04-02 22:48:54 +00:00
David E. O'Brien
991bf32193 Fix patch merge braino. 2001-04-02 22:46:02 +00:00
David E. O'Brien
b2cd1ce8ee Allow enabling soft updates (with -U) on a new filesystem.
[I first added this functionality, and thought to check prior art.  Seeing
OpenBSD had already done this, I changed my addition to reduce the diffs
between the two and went with their option letter.]
Obtained from:	OpenBSD
2001-04-02 01:25:55 +00:00
David E. O'Brien
094ab93715 The common wisdom is to use the largest number of cylinders per group.
So bump the default from `16' to `22', which is the largest value allowed
with the current default block size.  This change increases the the
group size from 32MB/g to 44MB/g on a 4GB SCSI disk.
2001-03-27 01:34:58 +00:00
Ruslan Ermilov
0a5779d45b - Backout botched attempt to introduce MANSECT feature.
- MAN[1-9] -> MAN.
2001-03-26 14:33:27 +00:00
Ruslan Ermilov
fe655281c5 Set the default manual section for sbin/ to 8. 2001-03-20 18:13:31 +00:00
Poul-Henning Kamp
7cf109c1b4 Make mount_mfs annoy users for 15 seconds and point them at mdconfig(8). 2001-01-30 10:21:20 +00:00
Ruslan Ermilov
9884911506 mdoc(7) police: fixed broken references. 2001-01-16 11:52:00 +00:00
Ian Dowse
f55ff3f3ef The ffs superblock includes a 128-byte region for use by temporary
in-core pointers to summary information. An array in this region
(fs_csp) could overflow on filesystems with a very large number of
cylinder groups (~16000 on i386 with 8k blocks). When this happens,
other fields in the superblock get corrupted, and fsck refuses to
check the filesystem.

Solve this problem by replacing the fs_csp array in 'struct fs'
with a single pointer, and add padding to keep the length of the
128-byte region fixed. Update the kernel and userland utilities
to use just this single pointer.

With this change, the kernel no longer makes use of the superblock
fields 'fs_csshift' and 'fs_csmask'. Add a comment to newfs/mkfs.c
to indicate that these fields must be calculated for compatibility
with older kernels.

Reviewed by:	mckusick
2001-01-15 18:30:40 +00:00
Eivind Eklund
10185bdc57 Replace reference to replacing mkfs(8) with a paragraph actually
describing what newfs *does*.
2001-01-15 03:13:26 +00:00
Ruslan Ermilov
d90d7015f9 Prepare for mdoc(7)NG. 2000-12-27 14:40:52 +00:00
Warner Losh
64dee60f8c o Add an example for a large file system.
o Remove bug about boot blocks hating non-8k file systems.  This hasn't been
  the case for a long time.

Not Objected to by: hackers, doc
2000-12-19 21:55:07 +00:00
Ruslan Ermilov
7c7fb079b9 mdoc(7) police: use the new features of the Nm macro. 2000-11-20 16:52:27 +00:00
Ruslan Ermilov
726b61ab5f Avoid use of direct troff requests in mdoc(7) manual pages. 2000-11-10 17:46:15 +00:00
John W. De Boskey
929f494bc7 Cast block number to off_t to avoid possible overflow bugs.
Pointed out by: Bruce Evans <bde@zeta.org.au>
2000-10-24 03:28:59 +00:00
John W. De Boskey
45c29d5cda The write combining code in revision 1.30 needs a few additional
touch ups.  The cache needs to be flushed against block
reads, and a final flush at process termination to force the
backup superblocks to disk.

I believe this will allow 'make release' to complete.

Submitted by:	Tor.Egge@fast.no
2000-10-24 00:08:30 +00:00
Peter Wemm
3927beeda1 Implement simple write combining for newfs - this is particularly useful
for large scsi disks with WCE = 0.  This yields around a 7 times speedup
on elapsed newfs time on test disks here.  64k clusters seems to be the
sweet spot for scsi disks using our present drivers.
2000-10-17 00:41:36 +00:00