Aren
/
NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
NetBSD/sys/ufs/lfs/lfs_balloc.c

425 lines
13 KiB
C
Raw Normal View History

Deal with fragment size changes better. For each fragment that can exist on an on-disk inode, we keep a record of its size in struct inode, which is updated when we write the block to disk. The cleaner routines thus have ready access to what size is the correct size for this block, on disk. Fixed a related bug: if a file with fragments is being cleaned (fragments being cleaned) at the same time it is being extended beyond NDADDR blocks, we could write a bogus FINFO record that has a frag in the middle; when it was cleaned this would give back bogus file data. Don't write the indirect blocks in this case, since there is no need. lfs_fragextend and lfs_truncate no longer require the seglock, but instead take a shared lock, which the seglock locks exclusively.
2002-07-06 05:30:11 +04:00
/* $NetBSD: lfs_balloc.c,v 1.33 2002/07/06 01:30:12 perseant Exp $ */
New RCS ID's, take two. they're more aesthecially pleasant, and use 'NetBSD'
1994-06-29 10:39:25 +04:00
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
/*-
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 08:49:54 +04:00
* Copyright (c) 1999, 2000 The NetBSD Foundation, Inc.
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
* All rights reserved.
*
* This code is derived from software contributed to The NetBSD Foundation
* by Konrad E. Schroder <perseant@hhhh.org>.
*
* 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.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the NetBSD
* Foundation, Inc. and its contributors.
* 4. Neither the name of The NetBSD Foundation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. 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 FOUNDATION 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.
*/
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
/*
* Copyright (c) 1989, 1991, 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.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by the University of
* California, Berkeley and its contributors.
* 4. Neither the name of the University nor the names of its contributors
* 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.
*
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
* @(#)lfs_balloc.c 8.4 (Berkeley) 5/8/95
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
*/
Use the newly-defined opt_quota.h.
1998-06-08 08:27:50 +04:00
add RCSID
2001-11-08 05:39:06 +03:00
#include <sys/cdefs.h>
Deal with fragment size changes better. For each fragment that can exist on an on-disk inode, we keep a record of its size in struct inode, which is updated when we write the block to disk. The cleaner routines thus have ready access to what size is the correct size for this block, on disk. Fixed a related bug: if a file with fragments is being cleaned (fragments being cleaned) at the same time it is being extended beyond NDADDR blocks, we could write a bogus FINFO record that has a frag in the middle; when it was cleaned this would give back bogus file data. Don't write the indirect blocks in this case, since there is no need. lfs_fragextend and lfs_truncate no longer require the seglock, but instead take a shared lock, which the seglock locks exclusively.
2002-07-06 05:30:11 +04:00
__KERNEL_RCSID(0, "$NetBSD: lfs_balloc.c,v 1.33 2002/07/06 01:30:12 perseant Exp $");
add RCSID
2001-11-08 05:39:06 +03:00
use _KERNEL_OPT
2001-05-30 15:57:16 +04:00
#if defined(_KERNEL_OPT)
Use the newly-defined opt_quota.h.
1998-06-08 08:27:50 +04:00
#include "opt_quota.h"
Protect various config(8)-generated files from inclusion while building LKMs. Fixes PR 5557.
1998-06-09 11:46:31 +04:00
#endif
Use the newly-defined opt_quota.h.
1998-06-08 08:27:50 +04:00
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
#include <sys/param.h>
lfs prototypes
1996-02-10 01:28:45 +03:00
#include <sys/systm.h>
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
#include <sys/buf.h>
#include <sys/proc.h>
#include <sys/vnode.h>
#include <sys/mount.h>
#include <sys/resourcevar.h>
#include <sys/trace.h>
#include <miscfs/specfs/specdev.h>
#include <ufs/ufs/quota.h>
#include <ufs/ufs/inode.h>
#include <ufs/ufs/ufsmount.h>
lfs prototypes
1996-02-10 01:28:45 +03:00
#include <ufs/ufs/ufs_extern.h>
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
#include <ufs/lfs/lfs.h>
#include <ufs/lfs/lfs_extern.h>
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.
2001-07-14 00:30:18 +04:00
int lfs_fragextend(struct vnode *, int, int, ufs_daddr_t, struct buf **, struct ucred *);
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
Change the way LFS does block accounting, from trying to infer from the buffer cache flags, to marking the inode and/or indirect blocks with a special disk address UNWRITTEN==-2 when a block is accounted for. (This address is never written to disk, but only used in-core. This is essentially the same method of block accounting as on the UBC branch, where the buffer headers don't exist.) Make sure that truncation is handled properly, especially in the case of holey files. Fixes PR#9994.
2000-05-06 00:59:20 +04:00
/*
* Allocate a block, and to inode and filesystem block accounting for it
* and for any indirect blocks the may need to be created in order for
* this block to be created.
*
* Blocks which have never been accounted for (i.e., which "do not exist")
* have disk address 0, which is translated by ufs_bmap to the special value
* UNASSIGNED == -1, as in the historical UFS.
*
* Blocks which have been accounted for but which have not yet been written
* to disk are given the new special disk address UNWRITTEN == -2, so that
* they can be differentiated from completely new blocks.
*/
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 08:49:54 +04:00
/* VOP_BWRITE NIADDR+2 times */
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
int
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.
2001-07-14 00:30:18 +04:00
lfs_balloc(void *v)
Add Kirk McKusick's soft updates code to the trunk. Not enabled by default, as the copyright on the main file (ffs_softdep.c) is such that is has been put into gnusrc. options SOFTDEP will pull this in. This code also contains the trickle syncer. Bump version number to 1.4O
1999-11-15 21:49:07 +03:00
{
struct vop_balloc_args /* {
struct vnode *a_vp;
off_t a_startoffset;
int a_size;
struct ucred *a_cred;
int a_flags;
struct buf *a_bpp;
} */ *ap = v;
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
struct vnode *vp;
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
int offset;
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
u_long iosize;
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
daddr_t daddr, idaddr;
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
struct buf *ibp, *bp;
struct inode *ip;
struct lfs *fs;
Change the way LFS does block accounting, from trying to infer from the buffer cache flags, to marking the inode and/or indirect blocks with a special disk address UNWRITTEN==-2 when a block is accounted for. (This address is never written to disk, but only used in-core. This is essentially the same method of block accounting as on the UBC branch, where the buffer headers don't exist.) Make sure that truncation is handled properly, especially in the case of holey files. Fixes PR#9994.
2000-05-06 00:59:20 +04:00
struct indir indirs[NIADDR+2], *idp;
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
ufs_daddr_t lbn, lastblock;
int bb, bcount;
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
int error, frags, i, nsize, osize, num;
Add Kirk McKusick's soft updates code to the trunk. Not enabled by default, as the copyright on the main file (ffs_softdep.c) is such that is has been put into gnusrc. options SOFTDEP will pull this in. This code also contains the trickle syncer. Bump version number to 1.4O
1999-11-15 21:49:07 +03:00
vp = ap->a_vp;
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
ip = VTOI(vp);
fs = ip->i_lfs;
Add Kirk McKusick's soft updates code to the trunk. Not enabled by default, as the copyright on the main file (ffs_softdep.c) is such that is has been put into gnusrc. options SOFTDEP will pull this in. This code also contains the trickle syncer. Bump version number to 1.4O
1999-11-15 21:49:07 +03:00
offset = blkoff(fs, ap->a_startoffset);
iosize = ap->a_size;
lbn = lblkno(fs, ap->a_startoffset);
(void)lfs_check(vp, lbn, 0);
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
/*
* Three cases: it's a block beyond the end of file, it's a block in
* the file that may or may not have been assigned a disk address or
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
* we're writing an entire block.
*
* Note, if the daddr is UNWRITTEN, the block already exists in
* the cache (it was read or written earlier). If so, make sure
* we don't count it as a new block or zero out its contents. If
* it did not, make sure we allocate any necessary indirect
* blocks.
*
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
* If we are writing a block beyond the end of the file, we need to
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
* check if the old last block was a fragment. If it was, we need
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
* to rewrite it.
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
*/
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
Add Kirk McKusick's soft updates code to the trunk. Not enabled by default, as the copyright on the main file (ffs_softdep.c) is such that is has been put into gnusrc. options SOFTDEP will pull this in. This code also contains the trickle syncer. Bump version number to 1.4O
1999-11-15 21:49:07 +03:00
*ap->a_bpp = NULL;
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
/* Check for block beyond end of file and fragment extension needed. */
lastblock = lblkno(fs, ip->i_ffs_size);
if (lastblock < NDADDR && lastblock < lbn) {
osize = blksize(fs, ip, lastblock);
if (osize < fs->lfs_bsize && osize > 0) {
if ((error = lfs_fragextend(vp, osize, fs->lfs_bsize,
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.
2001-07-14 00:30:18 +04:00
lastblock, &bp,
ap->a_cred)))
add spaces for KNF. confirmed to produce identical objects.
2001-11-24 00:44:25 +03:00
return (error);
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
ip->i_ffs_size = (lastblock + 1) * fs->lfs_bsize;
Don't cast the quad_t file size to u_long, this can cause overflows.
1998-03-03 12:02:51 +03:00
uvm_vnp_setsize(vp, ip->i_ffs_size);
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
ip->i_flag |= IN_CHANGE | IN_UPDATE;
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 08:49:54 +04:00
(void) VOP_BWRITE(bp);
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
}
}
/*
* If the block we are writing is a direct block, it's the last
* block in the file, and offset + iosize is less than a full
* block, we can write one or more fragments. There are two cases:
* the block is brand new and we should allocate it the correct
* size or it already exists and contains some fragments and
* may need to extend it.
*/
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
if (lbn < NDADDR && lblkno(fs, ip->i_ffs_size) <= lbn) {
osize = blksize(fs, ip, lbn);
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
nsize = fragroundup(fs, offset + iosize);
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
if (lblktosize(fs, lbn) >= ip->i_ffs_size) {
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
/* Brand new block or fragment */
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
frags = numfrags(fs, nsize);
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.
2001-07-14 00:30:18 +04:00
bb = fragstofsb(fs, frags);
Add Kirk McKusick's soft updates code to the trunk. Not enabled by default, as the copyright on the main file (ffs_softdep.c) is such that is has been put into gnusrc. options SOFTDEP will pull this in. This code also contains the trickle syncer. Bump version number to 1.4O
1999-11-15 21:49:07 +03:00
*ap->a_bpp = bp = getblk(vp, lbn, nsize, 0, 0);
Allow the number of free segments reserved for the cleaner to be parametrized in the filesystem, defaulting to MIN_FREE_SEGS = 2 but set to something more reasonable at newfs_lfs time. Note the number of blocks that have been scheduled for writing but which are not yet on disk in an inode extension, i_lfs_effnblks. Move i_ffs_effnlink out of the ffs extension and onto the main inode, since it's used all over the shared code and the lfs extension would clobber it. At inode write time, indirect blocks and inode-held blocks of inodes that have i_lfs_effnblks != i_ffs_blocks are cleansed of UNWRITTEN disk addresses, so that these never make it to disk.
2000-07-03 05:45:46 +04:00
ip->i_lfs_effnblks += bb;
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
ip->i_lfs->lfs_bfree -= bb;
ip->i_ffs_db[lbn] = bp->b_blkno = UNWRITTEN;
} else {
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
if (nsize <= osize) {
/* No need to extend */
if ((error = bread(vp, lbn, osize, NOCRED, &bp)))
return error;
} else {
/* Extend existing block */
if ((error =
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.
2001-07-14 00:30:18 +04:00
lfs_fragextend(vp, osize, nsize, lbn, &bp,
ap->a_cred)))
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
return error;
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
}
Add Kirk McKusick's soft updates code to the trunk. Not enabled by default, as the copyright on the main file (ffs_softdep.c) is such that is has been put into gnusrc. options SOFTDEP will pull this in. This code also contains the trickle syncer. Bump version number to 1.4O
1999-11-15 21:49:07 +03:00
*ap->a_bpp = bp;
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
}
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
return 0;
}
error = ufs_bmaparray(vp, lbn, &daddr, &indirs[0], &num, NULL );
if (error)
return (error);
/*
* Do byte accounting all at once, so we can gracefully fail *before*
* we start assigning blocks.
*/
bb = VFSTOUFS(vp->v_mount)->um_seqinc;
bcount = 0;
if (daddr == UNASSIGNED) {
bcount = bb;
}
for (i = 1; i < num; ++i) {
if (!indirs[i].in_exists) {
bcount += bb;
}
}
if (ISSPACE(fs, bcount, ap->a_cred)) {
ip->i_lfs->lfs_bfree -= bcount;
Allow the number of free segments reserved for the cleaner to be parametrized in the filesystem, defaulting to MIN_FREE_SEGS = 2 but set to something more reasonable at newfs_lfs time. Note the number of blocks that have been scheduled for writing but which are not yet on disk in an inode extension, i_lfs_effnblks. Move i_ffs_effnlink out of the ffs extension and onto the main inode, since it's used all over the shared code and the lfs extension would clobber it. At inode write time, indirect blocks and inode-held blocks of inodes that have i_lfs_effnblks != i_ffs_blocks are cleansed of UNWRITTEN disk addresses, so that these never make it to disk.
2000-07-03 05:45:46 +04:00
ip->i_lfs_effnblks += bcount;
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
} else {
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
return ENOSPC;
}
if (daddr == UNASSIGNED) {
if (num > 0 && ip->i_ffs_ib[indirs[0].in_off] == 0) {
ip->i_ffs_ib[indirs[0].in_off] = UNWRITTEN;
}
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
/*
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
* Create new indirect blocks if necessary
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
*/
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
if (num > 1)
idaddr = ip->i_ffs_ib[indirs[0].in_off];
for (i = 1; i < num; ++i) {
ibp = getblk(vp, indirs[i].in_lbn, fs->lfs_bsize, 0,0);
if (!indirs[i].in_exists) {
clrbuf(ibp);
ibp->b_blkno = UNWRITTEN;
} else if (!(ibp->b_flags & (B_DELWRI | B_DONE))) {
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.
2001-07-14 00:30:18 +04:00
ibp->b_blkno = fsbtodb(fs, idaddr);
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
ibp->b_flags |= B_READ;
VOP_STRATEGY(ibp);
biowait(ibp);
}
/*
* This block exists, but the next one may not.
* If that is the case mark it UNWRITTEN to keep
* the accounting straight.
*/
add spaces for KNF. confirmed to produce identical objects.
2001-11-24 00:44:25 +03:00
if (((daddr_t *)ibp->b_data)[indirs[i].in_off] == 0)
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
((daddr_t *)ibp->b_data)[indirs[i].in_off] =
UNWRITTEN;
idaddr = ((daddr_t *)ibp->b_data)[indirs[i].in_off];
if ((error = VOP_BWRITE(ibp))) {
return error;
}
}
}
/*
* Get the existing block from the cache.
*/
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.
2001-07-14 00:30:18 +04:00
frags = fsbtofrags(fs, bb);
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
*ap->a_bpp = bp = getblk(vp, lbn, blksize(fs, ip, lbn), 0, 0);
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
/*
* The block we are writing may be a brand new block
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
* in which case we need to do accounting.
Fix problems outlined in PR#9926: - lfs_truncate extends the file if called with length > i_ffs_size; - lfs_truncate errors out if called with length < 0; - lfs_balloc block accounting corrected for the case of blocks read into the cache before they exist on disk; - mp->mnt_stat.f_iosize is initialized in lfs_mountfs.
2000-04-24 01:10:26 +04:00
*
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
* We can tell a truly new block because ufs_bmaparray will say
* it is UNASSIGNED. Once we allocate it we will assign it the
* disk address UNWRITTEN.
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
*/
Change the way LFS does block accounting, from trying to infer from the buffer cache flags, to marking the inode and/or indirect blocks with a special disk address UNWRITTEN==-2 when a block is accounted for. (This address is never written to disk, but only used in-core. This is essentially the same method of block accounting as on the UBC branch, where the buffer headers don't exist.) Make sure that truncation is handled properly, especially in the case of holey files. Fixes PR#9994.
2000-05-06 00:59:20 +04:00
if (daddr == UNASSIGNED) {
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
if (iosize != fs->lfs_bsize)
clrbuf(bp);
/* Note the new address */
bp->b_blkno = UNWRITTEN;
switch (num) {
case 0:
ip->i_ffs_db[lbn] = UNWRITTEN;
break;
case 1:
ip->i_ffs_ib[indirs[0].in_off] = UNWRITTEN;
break;
default:
idp = &indirs[num - 1];
if (bread(vp, idp->in_lbn, fs->lfs_bsize, NOCRED,
&ibp))
panic("lfs_balloc: bread bno %d", idp->in_lbn);
((ufs_daddr_t *)ibp->b_data)[idp->in_off] = UNWRITTEN;
VOP_BWRITE(ibp);
Fix problems outlined in PR#9926: - lfs_truncate extends the file if called with length > i_ffs_size; - lfs_truncate errors out if called with length < 0; - lfs_balloc block accounting corrected for the case of blocks read into the cache before they exist on disk; - mp->mnt_stat.f_iosize is initialized in lfs_mountfs.
2000-04-24 01:10:26 +04:00
}
} else if (!(bp->b_flags & (B_DONE|B_DELWRI))) {
/*
* Not a brand new block, also not in the cache;
* read it in from disk.
*/
if (iosize == fs->lfs_bsize)
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
/* Optimization: I/O is unnecessary. */
bp->b_blkno = daddr;
Fix problems outlined in PR#9926: - lfs_truncate extends the file if called with length > i_ffs_size; - lfs_truncate errors out if called with length < 0; - lfs_balloc block accounting corrected for the case of blocks read into the cache before they exist on disk; - mp->mnt_stat.f_iosize is initialized in lfs_mountfs.
2000-04-24 01:10:26 +04:00
else {
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
/*
* We need to read the block to preserve the
* existing bytes.
*/
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
bp->b_blkno = daddr;
bp->b_flags |= B_READ;
VOP_STRATEGY(bp);
add spaces for KNF. confirmed to produce identical objects.
2001-11-24 00:44:25 +03:00
return (biowait(bp));
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
}
}
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
return (0);
}
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 08:49:54 +04:00
/* VOP_BWRITE 1 time */
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
int
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.
2001-07-14 00:30:18 +04:00
lfs_fragextend(struct vnode *vp, int osize, int nsize, ufs_daddr_t lbn, struct buf **bpp, struct ucred *cred)
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
{
struct inode *ip;
struct lfs *fs;
long bb;
int error;
New sources should leave the LFS in a more-or-less working state. Changes include: - DIROP segregation is enabled, and greater care is taken to make sure that a checkpoint completes. Fsck is not needed to remount the filesystem. - Several checks to make sure that the LFS subsystem does not overuse various resources (memory, in particular). - The cleaner routines, lfs_markv in particular, are completely rewritten. A buffer overflow is removed. Greater care is taken to ensure that inodes come from where lfs_cleanerd say they come from (so we know nothing has changed since lfs_bmapv was called). - Fragment allocation is fixed, so that writes beyond end-of-file do the right thing.
1999-03-10 03:20:00 +03:00
extern long locked_queue_bytes;
Correct accounting of lfs_avail, locked_queue_count, and locked_queue_bytes. (PR #11468). In the case of fragment allocation, check to see if enough space is available before extending a fragment already scheduled for writing. The locked_queue_* variables indicate the number of buffer headers and bytes, respectively, that are unavailable to getnewbuf() because they are locked up waiting for LFS to flush them; make sure that that is actually what we're counting, i.e., never count malloced buffers, and always use b_bufsize instead of b_bcount. If DEBUG is defined, the periodic calls to lfs_countlocked will now complain if either counter is incorrect. (In the future lfs_countlocked will not need to be called at all if DEBUG is not defined.)
2000-11-17 22:14:41 +03:00
size_t obufsize;
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
ip = VTOI(vp);
fs = ip->i_lfs;
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.
2001-07-14 00:30:18 +04:00
bb = (long)fragstofsb(fs, numfrags(fs, nsize - osize));
Protect inode free list with seglock, instead of separate lock, so that the head of the inode free list (on the superblock) always matches the rest of the free list (in the ifile). Protect lfs_fragextend with seglock, to prevent the segment byte count fudging from making its way to disk. Don't try to inactivate dirop vnodes that are still in the middle of their dirop (may address PR#10285).
2000-06-07 00:19:14 +04:00
error = 0;
/*
* Get the seglock so we don't enlarge blocks or change the segment
* accounting information while a segment is being written.
*/
Correct accounting of lfs_avail, locked_queue_count, and locked_queue_bytes. (PR #11468). In the case of fragment allocation, check to see if enough space is available before extending a fragment already scheduled for writing. The locked_queue_* variables indicate the number of buffer headers and bytes, respectively, that are unavailable to getnewbuf() because they are locked up waiting for LFS to flush them; make sure that that is actually what we're counting, i.e., never count malloced buffers, and always use b_bufsize instead of b_bcount. If DEBUG is defined, the periodic calls to lfs_countlocked will now complain if either counter is incorrect. (In the future lfs_countlocked will not need to be called at all if DEBUG is not defined.)
2000-11-17 22:14:41 +03:00
top:
Deal with fragment size changes better. For each fragment that can exist on an on-disk inode, we keep a record of its size in struct inode, which is updated when we write the block to disk. The cleaner routines thus have ready access to what size is the correct size for this block, on disk. Fixed a related bug: if a file with fragments is being cleaned (fragments being cleaned) at the same time it is being extended beyond NDADDR blocks, we could write a bogus FINFO record that has a frag in the middle; when it was cleaned this would give back bogus file data. Don't write the indirect blocks in this case, since there is no need. lfs_fragextend and lfs_truncate no longer require the seglock, but instead take a shared lock, which the seglock locks exclusively.
2002-07-06 05:30:11 +04:00
#ifdef LFS_MALLOC_SEGLOCK
Protect inode free list with seglock, instead of separate lock, so that the head of the inode free list (on the superblock) always matches the rest of the free list (in the ifile). Protect lfs_fragextend with seglock, to prevent the segment byte count fudging from making its way to disk. Don't try to inactivate dirop vnodes that are still in the middle of their dirop (may address PR#10285).
2000-06-07 00:19:14 +04:00
lfs_seglock(fs, SEGM_PROT);
Deal with fragment size changes better. For each fragment that can exist on an on-disk inode, we keep a record of its size in struct inode, which is updated when we write the block to disk. The cleaner routines thus have ready access to what size is the correct size for this block, on disk. Fixed a related bug: if a file with fragments is being cleaned (fragments being cleaned) at the same time it is being extended beyond NDADDR blocks, we could write a bogus FINFO record that has a frag in the middle; when it was cleaned this would give back bogus file data. Don't write the indirect blocks in this case, since there is no need. lfs_fragextend and lfs_truncate no longer require the seglock, but instead take a shared lock, which the seglock locks exclusively.
2002-07-06 05:30:11 +04:00
#else
lockmgr(&fs->lfs_fraglock, LK_SHARED, 0);
#endif
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.
2001-07-14 00:30:18 +04:00
if (!ISSPACE(fs, bb, cred)) {
Protect inode free list with seglock, instead of separate lock, so that the head of the inode free list (on the superblock) always matches the rest of the free list (in the ifile). Protect lfs_fragextend with seglock, to prevent the segment byte count fudging from making its way to disk. Don't try to inactivate dirop vnodes that are still in the middle of their dirop (may address PR#10285).
2000-06-07 00:19:14 +04:00
error = ENOSPC;
goto out;
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
}
if ((error = bread(vp, lbn, osize, NOCRED, bpp))) {
brelse(*bpp);
Protect inode free list with seglock, instead of separate lock, so that the head of the inode free list (on the superblock) always matches the rest of the free list (in the ifile). Protect lfs_fragextend with seglock, to prevent the segment byte count fudging from making its way to disk. Don't try to inactivate dirop vnodes that are still in the middle of their dirop (may address PR#10285).
2000-06-07 00:19:14 +04:00
goto out;
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
}
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
#ifdef QUOTA
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.
2001-07-14 00:30:18 +04:00
if ((error = chkdq(ip, bb, cred, 0))) {
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
brelse(*bpp);
goto out;
}
#endif
Correct accounting of lfs_avail, locked_queue_count, and locked_queue_bytes. (PR #11468). In the case of fragment allocation, check to see if enough space is available before extending a fragment already scheduled for writing. The locked_queue_* variables indicate the number of buffer headers and bytes, respectively, that are unavailable to getnewbuf() because they are locked up waiting for LFS to flush them; make sure that that is actually what we're counting, i.e., never count malloced buffers, and always use b_bufsize instead of b_bcount. If DEBUG is defined, the periodic calls to lfs_countlocked will now complain if either counter is incorrect. (In the future lfs_countlocked will not need to be called at all if DEBUG is not defined.)
2000-11-17 22:14:41 +03:00
/*
* Adjust accounting for lfs_avail. If there's not enough room,
* we will have to wait for the cleaner, which we can't do while
* holding a block busy or while holding the seglock. In that case,
* release both and start over after waiting.
*/
Deal with fragment size changes better. For each fragment that can exist on an on-disk inode, we keep a record of its size in struct inode, which is updated when we write the block to disk. The cleaner routines thus have ready access to what size is the correct size for this block, on disk. Fixed a related bug: if a file with fragments is being cleaned (fragments being cleaned) at the same time it is being extended beyond NDADDR blocks, we could write a bogus FINFO record that has a frag in the middle; when it was cleaned this would give back bogus file data. Don't write the indirect blocks in this case, since there is no need. lfs_fragextend and lfs_truncate no longer require the seglock, but instead take a shared lock, which the seglock locks exclusively.
2002-07-06 05:30:11 +04:00
Correct accounting of lfs_avail, locked_queue_count, and locked_queue_bytes. (PR #11468). In the case of fragment allocation, check to see if enough space is available before extending a fragment already scheduled for writing. The locked_queue_* variables indicate the number of buffer headers and bytes, respectively, that are unavailable to getnewbuf() because they are locked up waiting for LFS to flush them; make sure that that is actually what we're counting, i.e., never count malloced buffers, and always use b_bufsize instead of b_bcount. If DEBUG is defined, the periodic calls to lfs_countlocked will now complain if either counter is incorrect. (In the future lfs_countlocked will not need to be called at all if DEBUG is not defined.)
2000-11-17 22:14:41 +03:00
if ((*bpp)->b_flags & B_DELWRI) {
if (!lfs_fits(fs, bb)) {
brelse(*bpp);
#ifdef QUOTA
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.
2001-07-14 00:30:18 +04:00
chkdq(ip, -bb, cred, 0);
Correct accounting of lfs_avail, locked_queue_count, and locked_queue_bytes. (PR #11468). In the case of fragment allocation, check to see if enough space is available before extending a fragment already scheduled for writing. The locked_queue_* variables indicate the number of buffer headers and bytes, respectively, that are unavailable to getnewbuf() because they are locked up waiting for LFS to flush them; make sure that that is actually what we're counting, i.e., never count malloced buffers, and always use b_bufsize instead of b_bcount. If DEBUG is defined, the periodic calls to lfs_countlocked will now complain if either counter is incorrect. (In the future lfs_countlocked will not need to be called at all if DEBUG is not defined.)
2000-11-17 22:14:41 +03:00
#endif
Deal with fragment size changes better. For each fragment that can exist on an on-disk inode, we keep a record of its size in struct inode, which is updated when we write the block to disk. The cleaner routines thus have ready access to what size is the correct size for this block, on disk. Fixed a related bug: if a file with fragments is being cleaned (fragments being cleaned) at the same time it is being extended beyond NDADDR blocks, we could write a bogus FINFO record that has a frag in the middle; when it was cleaned this would give back bogus file data. Don't write the indirect blocks in this case, since there is no need. lfs_fragextend and lfs_truncate no longer require the seglock, but instead take a shared lock, which the seglock locks exclusively.
2002-07-06 05:30:11 +04:00
#ifdef LFS_FRAGSIZE_SEGLOCK
Correct accounting of lfs_avail, locked_queue_count, and locked_queue_bytes. (PR #11468). In the case of fragment allocation, check to see if enough space is available before extending a fragment already scheduled for writing. The locked_queue_* variables indicate the number of buffer headers and bytes, respectively, that are unavailable to getnewbuf() because they are locked up waiting for LFS to flush them; make sure that that is actually what we're counting, i.e., never count malloced buffers, and always use b_bufsize instead of b_bcount. If DEBUG is defined, the periodic calls to lfs_countlocked will now complain if either counter is incorrect. (In the future lfs_countlocked will not need to be called at all if DEBUG is not defined.)
2000-11-17 22:14:41 +03:00
lfs_segunlock(fs);
Deal with fragment size changes better. For each fragment that can exist on an on-disk inode, we keep a record of its size in struct inode, which is updated when we write the block to disk. The cleaner routines thus have ready access to what size is the correct size for this block, on disk. Fixed a related bug: if a file with fragments is being cleaned (fragments being cleaned) at the same time it is being extended beyond NDADDR blocks, we could write a bogus FINFO record that has a frag in the middle; when it was cleaned this would give back bogus file data. Don't write the indirect blocks in this case, since there is no need. lfs_fragextend and lfs_truncate no longer require the seglock, but instead take a shared lock, which the seglock locks exclusively.
2002-07-06 05:30:11 +04:00
#else
lockmgr(&fs->lfs_fraglock, LK_RELEASE, 0);
#endif
Correct accounting of lfs_avail, locked_queue_count, and locked_queue_bytes. (PR #11468). In the case of fragment allocation, check to see if enough space is available before extending a fragment already scheduled for writing. The locked_queue_* variables indicate the number of buffer headers and bytes, respectively, that are unavailable to getnewbuf() because they are locked up waiting for LFS to flush them; make sure that that is actually what we're counting, i.e., never count malloced buffers, and always use b_bufsize instead of b_bcount. If DEBUG is defined, the periodic calls to lfs_countlocked will now complain if either counter is incorrect. (In the future lfs_countlocked will not need to be called at all if DEBUG is not defined.)
2000-11-17 22:14:41 +03:00
lfs_availwait(fs, bb);
goto top;
}
fs->lfs_avail -= bb;
}
Fixes associated with filling an LFS: Change the space computation to appear to change the size of the *disk* rather than the *bytes used* when more segment summaries and inode blocks are written. Try to estimate the amount of space that these will take up when more files are written, so the disk size doesn't change too much. Regularize error returns from lfs_valloc, lfs_balloc, lfs_truncate: they now fail entirely, rather than succeeding half-way and leaving the fs in an inconsistent state. Rewrite lfs_truncate, mostly stealing from ffs_truncate. The old lfs_truncate had difficulty truncating a large file to a non-zero size (indirect blocks were not handled appropriately). Unmark VDIROP on fvp after ufs_remove, ufs_rmdir, so these can be reclaimed immediately: this vnode would not be written to disk again anyway if the removal succeeded, and if it failed, no directory operation occurred. ufs_makeinode and ufs_mkdir now remove IN_ADIROP on error.
2000-06-28 00:57:11 +04:00
fs->lfs_bfree -= bb;
i_lfs_effnblks fixes. Put debugging printfs under #ifdef DEBUG_LFS.
2000-07-03 12:20:58 +04:00
ip->i_lfs_effnblks += bb;
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
ip->i_flag |= IN_CHANGE | IN_UPDATE;
Correct accounting of lfs_avail, locked_queue_count, and locked_queue_bytes. (PR #11468). In the case of fragment allocation, check to see if enough space is available before extending a fragment already scheduled for writing. The locked_queue_* variables indicate the number of buffer headers and bytes, respectively, that are unavailable to getnewbuf() because they are locked up waiting for LFS to flush them; make sure that that is actually what we're counting, i.e., never count malloced buffers, and always use b_bufsize instead of b_bcount. If DEBUG is defined, the periodic calls to lfs_countlocked will now complain if either counter is incorrect. (In the future lfs_countlocked will not need to be called at all if DEBUG is not defined.)
2000-11-17 22:14:41 +03:00
Phase one of my three-phase plan to make LFS play nice with UBC, and bug-fixes I found while making sure there weren't any new ones. * Make the write clusters keep track of the buffers whose blocks they contain. This should make it possible to (1) write clusters using a page mapping instead of malloc, if desired, and (2) schedule blocks for rewriting (somewhere else) if a write error occurs. Code is present to use pagemove() to construct the clusters but that is untested and will go away anyway in favor of page mapping. * DEBUG now keeps a log of Ifile writes, so that any lingering instances of the "dirty bufs" problem can be properly debugged. * Keep track of whether the Ifile has been dirtied by various routines that can be called by lfs_segwrite, and loop on that until it is clean, for a checkpoint. Checkpoints need to be squeaky clean. * Warn the user (once) if the Ifile grows larger than is reasonable for their buffer cache. Both lfs_mountfs and lfs_unmount check since the Ifile can grow. * If an inode is not found in a disk block, try rereading the block, under the assumption that the block was copied to a cluster and then freed. * Protect WRITEINPROG() with splbio() to fix a hang in lfs_update.
2002-05-15 00:03:53 +04:00
LFS_DEBUG_COUNTLOCKED("frag1");
Correct accounting of lfs_avail, locked_queue_count, and locked_queue_bytes. (PR #11468). In the case of fragment allocation, check to see if enough space is available before extending a fragment already scheduled for writing. The locked_queue_* variables indicate the number of buffer headers and bytes, respectively, that are unavailable to getnewbuf() because they are locked up waiting for LFS to flush them; make sure that that is actually what we're counting, i.e., never count malloced buffers, and always use b_bufsize instead of b_bcount. If DEBUG is defined, the periodic calls to lfs_countlocked will now complain if either counter is incorrect. (In the future lfs_countlocked will not need to be called at all if DEBUG is not defined.)
2000-11-17 22:14:41 +03:00
obufsize = (*bpp)->b_bufsize;
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
allocbuf(*bpp, nsize);
Correct accounting of lfs_avail, locked_queue_count, and locked_queue_bytes. (PR #11468). In the case of fragment allocation, check to see if enough space is available before extending a fragment already scheduled for writing. The locked_queue_* variables indicate the number of buffer headers and bytes, respectively, that are unavailable to getnewbuf() because they are locked up waiting for LFS to flush them; make sure that that is actually what we're counting, i.e., never count malloced buffers, and always use b_bufsize instead of b_bcount. If DEBUG is defined, the periodic calls to lfs_countlocked will now complain if either counter is incorrect. (In the future lfs_countlocked will not need to be called at all if DEBUG is not defined.)
2000-11-17 22:14:41 +03:00
/* Adjust locked-list accounting */
if (((*bpp)->b_flags & (B_LOCKED | B_CALL)) == B_LOCKED)
More locked_queue_* and lfs_avail accounting fixes from Jesse Off <joff@gci-net.com>. Remove a specious btodb() in lfs_fragextend, and count blocks shrunk or removed by VOP_TRUNCATE in lfs_avail.
2000-11-21 03:00:31 +03:00
locked_queue_bytes += (*bpp)->b_bufsize - obufsize;
Correct accounting of lfs_avail, locked_queue_count, and locked_queue_bytes. (PR #11468). In the case of fragment allocation, check to see if enough space is available before extending a fragment already scheduled for writing. The locked_queue_* variables indicate the number of buffer headers and bytes, respectively, that are unavailable to getnewbuf() because they are locked up waiting for LFS to flush them; make sure that that is actually what we're counting, i.e., never count malloced buffers, and always use b_bufsize instead of b_bcount. If DEBUG is defined, the periodic calls to lfs_countlocked will now complain if either counter is incorrect. (In the future lfs_countlocked will not need to be called at all if DEBUG is not defined.)
2000-11-17 22:14:41 +03:00
Phase one of my three-phase plan to make LFS play nice with UBC, and bug-fixes I found while making sure there weren't any new ones. * Make the write clusters keep track of the buffers whose blocks they contain. This should make it possible to (1) write clusters using a page mapping instead of malloc, if desired, and (2) schedule blocks for rewriting (somewhere else) if a write error occurs. Code is present to use pagemove() to construct the clusters but that is untested and will go away anyway in favor of page mapping. * DEBUG now keeps a log of Ifile writes, so that any lingering instances of the "dirty bufs" problem can be properly debugged. * Keep track of whether the Ifile has been dirtied by various routines that can be called by lfs_segwrite, and loop on that until it is clean, for a checkpoint. Checkpoints need to be squeaky clean. * Warn the user (once) if the Ifile grows larger than is reasonable for their buffer cache. Both lfs_mountfs and lfs_unmount check since the Ifile can grow. * If an inode is not found in a disk block, try rereading the block, under the assumption that the block was copied to a cluster and then freed. * Protect WRITEINPROG() with splbio() to fix a hang in lfs_update.
2002-05-15 00:03:53 +04:00
LFS_DEBUG_COUNTLOCKED("frag2");
Merge with Lite2 + local changes
1998-03-01 05:20:01 +03:00
bzero((char *)((*bpp)->b_data) + osize, (u_int)(nsize - osize));
Protect inode free list with seglock, instead of separate lock, so that the head of the inode free list (on the superblock) always matches the rest of the free list (in the ifile). Protect lfs_fragextend with seglock, to prevent the segment byte count fudging from making its way to disk. Don't try to inactivate dirop vnodes that are still in the middle of their dirop (may address PR#10285).
2000-06-07 00:19:14 +04:00
out:
Deal with fragment size changes better. For each fragment that can exist on an on-disk inode, we keep a record of its size in struct inode, which is updated when we write the block to disk. The cleaner routines thus have ready access to what size is the correct size for this block, on disk. Fixed a related bug: if a file with fragments is being cleaned (fragments being cleaned) at the same time it is being extended beyond NDADDR blocks, we could write a bogus FINFO record that has a frag in the middle; when it was cleaned this would give back bogus file data. Don't write the indirect blocks in this case, since there is no need. lfs_fragextend and lfs_truncate no longer require the seglock, but instead take a shared lock, which the seglock locks exclusively.
2002-07-06 05:30:11 +04:00
#ifdef LFS_FRAGSIZE_SEGLOCK
Protect inode free list with seglock, instead of separate lock, so that the head of the inode free list (on the superblock) always matches the rest of the free list (in the ifile). Protect lfs_fragextend with seglock, to prevent the segment byte count fudging from making its way to disk. Don't try to inactivate dirop vnodes that are still in the middle of their dirop (may address PR#10285).
2000-06-07 00:19:14 +04:00
lfs_segunlock(fs);
Deal with fragment size changes better. For each fragment that can exist on an on-disk inode, we keep a record of its size in struct inode, which is updated when we write the block to disk. The cleaner routines thus have ready access to what size is the correct size for this block, on disk. Fixed a related bug: if a file with fragments is being cleaned (fragments being cleaned) at the same time it is being extended beyond NDADDR blocks, we could write a bogus FINFO record that has a frag in the middle; when it was cleaned this would give back bogus file data. Don't write the indirect blocks in this case, since there is no need. lfs_fragextend and lfs_truncate no longer require the seglock, but instead take a shared lock, which the seglock locks exclusively.
2002-07-06 05:30:11 +04:00
#else
lockmgr(&fs->lfs_fraglock, LK_RELEASE, 0);
#endif
Protect inode free list with seglock, instead of separate lock, so that the head of the inode free list (on the superblock) always matches the rest of the free list (in the ifile). Protect lfs_fragextend with seglock, to prevent the segment byte count fudging from making its way to disk. Don't try to inactivate dirop vnodes that are still in the middle of their dirop (may address PR#10285).
2000-06-07 00:19:14 +04:00
return (error);
Update to 4.4-Lite fs code, with local changes.
1994-06-08 15:41:58 +04:00
}