2017-11-20 19:49:47 +00:00
|
|
|
/*-
|
|
|
|
* SPDX-License-Identifier: BSD-3-Clause
|
|
|
|
*
|
1994-05-26 06:35:07 +00:00
|
|
|
* Copyright (c) 1980, 1986, 1993
|
|
|
|
* The Regents of the University of California. All rights reserved.
|
|
|
|
*
|
|
|
|
* Redistribution and use in source and binary forms, with or without
|
|
|
|
* modification, are permitted provided that the following conditions
|
|
|
|
* are met:
|
|
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer.
|
|
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
|
|
* documentation and/or other materials provided with the distribution.
|
2017-02-28 23:42:47 +00:00
|
|
|
* 3. Neither the name of the University nor the names of its contributors
|
1994-05-26 06:35:07 +00:00
|
|
|
* may be used to endorse or promote products derived from this software
|
|
|
|
* without specific prior written permission.
|
|
|
|
*
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
|
|
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
|
|
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
|
|
* SUCH DAMAGE.
|
|
|
|
*/
|
|
|
|
|
1998-12-03 02:41:11 +00:00
|
|
|
#if 0
|
2003-05-03 18:41:59 +00:00
|
|
|
#ifndef lint
|
1997-03-11 12:20:21 +00:00
|
|
|
static const char sccsid[] = "@(#)pass3.c 8.2 (Berkeley) 4/27/95";
|
1994-05-26 06:35:07 +00:00
|
|
|
#endif /* not lint */
|
2003-05-03 18:41:59 +00:00
|
|
|
#endif
|
|
|
|
#include <sys/cdefs.h>
|
|
|
|
__FBSDID("$FreeBSD$");
|
1994-05-26 06:35:07 +00:00
|
|
|
|
|
|
|
#include <sys/param.h>
|
1997-03-11 12:20:21 +00:00
|
|
|
|
1994-05-26 06:35:07 +00:00
|
|
|
#include <ufs/ufs/dinode.h>
|
1998-12-03 02:27:35 +00:00
|
|
|
#include <ufs/ufs/dir.h>
|
|
|
|
#include <ufs/ffs/fs.h>
|
|
|
|
|
|
|
|
#include <string.h>
|
1997-03-11 12:20:21 +00:00
|
|
|
|
1994-05-26 06:35:07 +00:00
|
|
|
#include "fsck.h"
|
|
|
|
|
1995-04-02 14:52:29 +00:00
|
|
|
void
|
2002-03-20 22:57:10 +00:00
|
|
|
pass3(void)
|
1994-05-26 06:35:07 +00:00
|
|
|
{
|
1998-12-03 02:27:35 +00:00
|
|
|
struct inoinfo *inp;
|
|
|
|
int loopcnt, inpindex, state;
|
1994-05-26 06:35:07 +00:00
|
|
|
ino_t orphan;
|
2021-01-07 01:37:08 +00:00
|
|
|
struct inode ip;
|
1998-12-03 02:27:35 +00:00
|
|
|
struct inodesc idesc;
|
2017-02-09 17:47:01 +00:00
|
|
|
char namebuf[UFS_MAXNAMLEN+1];
|
1994-05-26 06:35:07 +00:00
|
|
|
|
1998-12-03 02:27:35 +00:00
|
|
|
for (inpindex = inplast - 1; inpindex >= 0; inpindex--) {
|
2000-12-15 14:23:55 +00:00
|
|
|
if (got_siginfo) {
|
|
|
|
printf("%s: phase 3: dir %d of %d (%d%%)\n", cdevname,
|
2001-11-17 23:48:21 +00:00
|
|
|
(int)(inplast - inpindex - 1), (int)inplast,
|
|
|
|
(int)((inplast - inpindex - 1) * 100 / inplast));
|
2000-12-15 14:23:55 +00:00
|
|
|
got_siginfo = 0;
|
|
|
|
}
|
2004-02-28 07:50:42 +00:00
|
|
|
if (got_sigalarm) {
|
|
|
|
setproctitle("%s p3 %d%%", cdevname,
|
|
|
|
(int)((inplast - inpindex - 1) * 100 / inplast));
|
|
|
|
got_sigalarm = 0;
|
|
|
|
}
|
1998-12-03 02:27:35 +00:00
|
|
|
inp = inpsort[inpindex];
|
|
|
|
state = inoinfo(inp->i_number)->ino_state;
|
2017-02-15 19:50:26 +00:00
|
|
|
if (inp->i_number == UFS_ROOTINO ||
|
2004-10-08 20:44:47 +00:00
|
|
|
(inp->i_parent != 0 && !S_IS_DUNFOUND(state)))
|
1998-12-03 02:27:35 +00:00
|
|
|
continue;
|
Improvement in UFS/FFS directory placement when doing mkdir(2).
The algorithm for laying out new directories was devised in the 1980s
and markedly improved the performance of the filesystem. In those days
large disks had at most 100 cylinder groups and often as few as 10-20.
Modern multi-terrabyte disks have thousands of cylinder groups. The
original algorithm does not handle these large sizes well. This change
attempts to expand the scope of the original algorithm to work well
with these much larger disks while still retaining the properties
of the original algorithm for small disks.
The filesystem implementation is divided into policy routines and
implementation routines. The policy routines can be changed in any
way desired without risk of corrupting the filesystem. The policy
requests are handled by the implementation layer. If the policy
asks for an available resource, it is granted. But if it asks for
an already in-use resource, then the implementation will provide
an available one nearby the request. Thus it is impossible for a
policy to double allocate. This change is limited to the policy
implementation.
This change updates the ffs_dirpref() routine which is responsible
for selecting the cylinder group into which a new directory should
be placed. If we are near the root of the filesystem we aim to
spread them out as much as possible. As we descend deeper from the
root we cluster them closer together around their parent as we
expect them to be more closely interactive. Higher-level directories
like usr/src/sys and usr/src/bin should be separated while the
directories in these areas are more likely to be accessed together
so should be closer. And directories within commands or kernel
subsystems should be closer still.
We pick a range of cylinder groups around the cylinder group of the
directory in which we are being created. The size of the range for
our search is based on our depth from the root of our filesystem.
We then probe that range based on how many directories are already
present. The first new directory is at 1/2 (middle) of the range;
the second is in the first 1/4 of the range, then at 3/4, 1/8, 3/8,
5/8, 7/8, 1/16, 3/16, 5/16, etc.
It is desirable to store the depth of a directory in its on-disk
inode so that it is available when we need it. We add a new field
di_dirdepth to track the depth of each directory. Because there are
few spare fields left in the inode, we choose to share an existing
field in the inode rather than having one of our own. Specifically
we create a union with the di_freelink field. The di_freelink field
is used to track inodes that have been unlinked but remain referenced.
It is not needed until a rmdir(2) operation has been done on a
directory. At that point, the directory has no contents and even
if it is kept active as a current directory is no longer able to
have any new directories or files created in it. Thus the use of
di_dirdepth and di_freelink will never coincide.
Reported by: Timo Voelker
Reviewed by: kib
Tested by: Peter Holm
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D39246
2023-03-30 04:09:39 +00:00
|
|
|
if (state == DCLEAR || state == DZLINK)
|
1994-05-26 06:35:07 +00:00
|
|
|
continue;
|
1998-12-03 02:27:35 +00:00
|
|
|
/*
|
|
|
|
* If we are running with soft updates and we come
|
|
|
|
* across unreferenced directories, we just leave
|
|
|
|
* them in DSTATE which will cause them to be pitched
|
|
|
|
* in pass 4.
|
|
|
|
*/
|
2001-03-21 09:48:03 +00:00
|
|
|
if ((preen || bkgrdflag) &&
|
2004-10-08 20:44:47 +00:00
|
|
|
resolved && usedsoftdep && S_IS_DUNFOUND(state)) {
|
2017-02-15 19:50:26 +00:00
|
|
|
if (inp->i_dotdot >= UFS_ROOTINO)
|
1998-12-03 02:27:35 +00:00
|
|
|
inoinfo(inp->i_dotdot)->ino_linkcnt++;
|
1994-05-26 06:35:07 +00:00
|
|
|
continue;
|
1998-12-03 02:27:35 +00:00
|
|
|
}
|
1994-05-26 06:35:07 +00:00
|
|
|
for (loopcnt = 0; ; loopcnt++) {
|
|
|
|
orphan = inp->i_number;
|
|
|
|
if (inp->i_parent == 0 ||
|
2004-10-08 20:44:47 +00:00
|
|
|
!INO_IS_DUNFOUND(inp->i_parent) ||
|
1998-12-03 02:27:35 +00:00
|
|
|
loopcnt > countdirs)
|
1994-05-26 06:35:07 +00:00
|
|
|
break;
|
|
|
|
inp = getinoinfo(inp->i_parent);
|
|
|
|
}
|
1998-12-03 02:27:35 +00:00
|
|
|
if (loopcnt <= countdirs) {
|
|
|
|
if (linkup(orphan, inp->i_dotdot, NULL)) {
|
|
|
|
inp->i_parent = inp->i_dotdot = lfdir;
|
|
|
|
inoinfo(lfdir)->ino_linkcnt--;
|
|
|
|
}
|
|
|
|
inoinfo(orphan)->ino_state = DFOUND;
|
Improvement in UFS/FFS directory placement when doing mkdir(2).
The algorithm for laying out new directories was devised in the 1980s
and markedly improved the performance of the filesystem. In those days
large disks had at most 100 cylinder groups and often as few as 10-20.
Modern multi-terrabyte disks have thousands of cylinder groups. The
original algorithm does not handle these large sizes well. This change
attempts to expand the scope of the original algorithm to work well
with these much larger disks while still retaining the properties
of the original algorithm for small disks.
The filesystem implementation is divided into policy routines and
implementation routines. The policy routines can be changed in any
way desired without risk of corrupting the filesystem. The policy
requests are handled by the implementation layer. If the policy
asks for an available resource, it is granted. But if it asks for
an already in-use resource, then the implementation will provide
an available one nearby the request. Thus it is impossible for a
policy to double allocate. This change is limited to the policy
implementation.
This change updates the ffs_dirpref() routine which is responsible
for selecting the cylinder group into which a new directory should
be placed. If we are near the root of the filesystem we aim to
spread them out as much as possible. As we descend deeper from the
root we cluster them closer together around their parent as we
expect them to be more closely interactive. Higher-level directories
like usr/src/sys and usr/src/bin should be separated while the
directories in these areas are more likely to be accessed together
so should be closer. And directories within commands or kernel
subsystems should be closer still.
We pick a range of cylinder groups around the cylinder group of the
directory in which we are being created. The size of the range for
our search is based on our depth from the root of our filesystem.
We then probe that range based on how many directories are already
present. The first new directory is at 1/2 (middle) of the range;
the second is in the first 1/4 of the range, then at 3/4, 1/8, 3/8,
5/8, 7/8, 1/16, 3/16, 5/16, etc.
It is desirable to store the depth of a directory in its on-disk
inode so that it is available when we need it. We add a new field
di_dirdepth to track the depth of each directory. Because there are
few spare fields left in the inode, we choose to share an existing
field in the inode rather than having one of our own. Specifically
we create a union with the di_freelink field. The di_freelink field
is used to track inodes that have been unlinked but remain referenced.
It is not needed until a rmdir(2) operation has been done on a
directory. At that point, the directory has no contents and even
if it is kept active as a current directory is no longer able to
have any new directories or files created in it. Thus the use of
di_dirdepth and di_freelink will never coincide.
Reported by: Timo Voelker
Reviewed by: kib
Tested by: Peter Holm
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D39246
2023-03-30 04:09:39 +00:00
|
|
|
check_dirdepth(inp);
|
1998-12-03 02:27:35 +00:00
|
|
|
propagate();
|
|
|
|
continue;
|
|
|
|
}
|
2001-11-17 23:48:21 +00:00
|
|
|
pfatal("ORPHANED DIRECTORY LOOP DETECTED I=%lu",
|
|
|
|
(u_long)orphan);
|
1998-12-03 02:27:35 +00:00
|
|
|
if (reply("RECONNECT") == 0)
|
|
|
|
continue;
|
|
|
|
memset(&idesc, 0, sizeof(struct inodesc));
|
|
|
|
idesc.id_type = DATA;
|
|
|
|
idesc.id_number = inp->i_parent;
|
|
|
|
idesc.id_parent = orphan;
|
|
|
|
idesc.id_func = findname;
|
|
|
|
idesc.id_name = namebuf;
|
2021-01-07 01:37:08 +00:00
|
|
|
ginode(inp->i_parent, &ip);
|
|
|
|
if ((ckinode(ip.i_dp, &idesc) & FOUND) == 0)
|
1998-12-03 02:27:35 +00:00
|
|
|
pfatal("COULD NOT FIND NAME IN PARENT DIRECTORY");
|
|
|
|
if (linkup(orphan, inp->i_parent, namebuf)) {
|
|
|
|
idesc.id_func = clearentry;
|
2021-01-07 01:37:08 +00:00
|
|
|
if (ckinode(ip.i_dp, &idesc) & FOUND)
|
1998-12-03 02:27:35 +00:00
|
|
|
inoinfo(orphan)->ino_linkcnt++;
|
|
|
|
inp->i_parent = inp->i_dotdot = lfdir;
|
|
|
|
inoinfo(lfdir)->ino_linkcnt--;
|
|
|
|
}
|
2021-01-07 01:37:08 +00:00
|
|
|
irelse(&ip);
|
1998-12-03 02:27:35 +00:00
|
|
|
inoinfo(orphan)->ino_state = DFOUND;
|
Improvement in UFS/FFS directory placement when doing mkdir(2).
The algorithm for laying out new directories was devised in the 1980s
and markedly improved the performance of the filesystem. In those days
large disks had at most 100 cylinder groups and often as few as 10-20.
Modern multi-terrabyte disks have thousands of cylinder groups. The
original algorithm does not handle these large sizes well. This change
attempts to expand the scope of the original algorithm to work well
with these much larger disks while still retaining the properties
of the original algorithm for small disks.
The filesystem implementation is divided into policy routines and
implementation routines. The policy routines can be changed in any
way desired without risk of corrupting the filesystem. The policy
requests are handled by the implementation layer. If the policy
asks for an available resource, it is granted. But if it asks for
an already in-use resource, then the implementation will provide
an available one nearby the request. Thus it is impossible for a
policy to double allocate. This change is limited to the policy
implementation.
This change updates the ffs_dirpref() routine which is responsible
for selecting the cylinder group into which a new directory should
be placed. If we are near the root of the filesystem we aim to
spread them out as much as possible. As we descend deeper from the
root we cluster them closer together around their parent as we
expect them to be more closely interactive. Higher-level directories
like usr/src/sys and usr/src/bin should be separated while the
directories in these areas are more likely to be accessed together
so should be closer. And directories within commands or kernel
subsystems should be closer still.
We pick a range of cylinder groups around the cylinder group of the
directory in which we are being created. The size of the range for
our search is based on our depth from the root of our filesystem.
We then probe that range based on how many directories are already
present. The first new directory is at 1/2 (middle) of the range;
the second is in the first 1/4 of the range, then at 3/4, 1/8, 3/8,
5/8, 7/8, 1/16, 3/16, 5/16, etc.
It is desirable to store the depth of a directory in its on-disk
inode so that it is available when we need it. We add a new field
di_dirdepth to track the depth of each directory. Because there are
few spare fields left in the inode, we choose to share an existing
field in the inode rather than having one of our own. Specifically
we create a union with the di_freelink field. The di_freelink field
is used to track inodes that have been unlinked but remain referenced.
It is not needed until a rmdir(2) operation has been done on a
directory. At that point, the directory has no contents and even
if it is kept active as a current directory is no longer able to
have any new directories or files created in it. Thus the use of
di_dirdepth and di_freelink will never coincide.
Reported by: Timo Voelker
Reviewed by: kib
Tested by: Peter Holm
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D39246
2023-03-30 04:09:39 +00:00
|
|
|
check_dirdepth(inp);
|
1994-05-26 06:35:07 +00:00
|
|
|
propagate();
|
|
|
|
}
|
|
|
|
}
|