freebsd-dev/sys/ufs/ffs/ffs_inode.c

770 lines
23 KiB
C
Raw Normal View History

/*-
* SPDX-License-Identifier: BSD-3-Clause
*
1994-05-24 10:09:53 +00:00
* Copyright (c) 1982, 1986, 1989, 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. Neither the name of the University nor the names of its contributors
1994-05-24 10:09:53 +00:00
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* @(#)ffs_inode.c 8.13 (Berkeley) 4/21/95
1994-05-24 10:09:53 +00:00
*/
2003-06-11 06:34:30 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
1996-01-05 18:31:58 +00:00
#include "opt_quota.h"
1994-05-24 10:09:53 +00:00
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/bio.h>
1994-05-24 10:09:53 +00:00
#include <sys/buf.h>
#include <sys/malloc.h>
Huge cleanup of random(4) code. * GENERAL - Update copyright. - Make kernel options for RANDOM_YARROW and RANDOM_DUMMY. Set neither to ON, which means we want Fortuna - If there is no 'device random' in the kernel, there will be NO random(4) device in the kernel, and the KERN_ARND sysctl will return nothing. With RANDOM_DUMMY there will be a random(4) that always blocks. - Repair kern.arandom (KERN_ARND sysctl). The old version went through arc4random(9) and was a bit weird. - Adjust arc4random stirring a bit - the existing code looks a little suspect. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Redo read_random(9) so as to duplicate random(4)'s read internals. This makes it a first-class citizen rather than a hack. - Move stuff out of locked regions when it does not need to be there. - Trim RANDOM_DEBUG printfs. Some are excess to requirement, some behind boot verbose. - Use SYSINIT to sequence the startup. - Fix init/deinit sysctl stuff. - Make relevant sysctls also tunables. - Add different harvesting "styles" to allow for different requirements (direct, queue, fast). - Add harvesting of FFS atime events. This needs to be checked for weighing down the FS code. - Add harvesting of slab allocator events. This needs to be checked for weighing down the allocator code. - Fix the random(9) manpage. - Loadable modules are not present for now. These will be re-engineered when the dust settles. - Use macros for locks. - Fix comments. * src/share/man/... - Update the man pages. * src/etc/... - The startup/shutdown work is done in D2924. * src/UPDATING - Add UPDATING announcement. * src/sys/dev/random/build.sh - Add copyright. - Add libz for unit tests. * src/sys/dev/random/dummy.c - Remove; no longer needed. Functionality incorporated into randomdev.*. * live_entropy_sources.c live_entropy_sources.h - Remove; content moved. - move content to randomdev.[ch] and optimise. * src/sys/dev/random/random_adaptors.c src/sys/dev/random/random_adaptors.h - Remove; plugability is no longer used. Compile-time algorithm selection is the way to go. * src/sys/dev/random/random_harvestq.c src/sys/dev/random/random_harvestq.h - Add early (re)boot-time randomness caching. * src/sys/dev/random/randomdev_soft.c src/sys/dev/random/randomdev_soft.h - Remove; no longer needed. * src/sys/dev/random/uint128.h - Provide a fake uint128_t; if a real one ever arrived, we can use that instead. All that is needed here is N=0, N++, N==0, and some localised trickery is used to manufacture a 128-bit 0ULLL. * src/sys/dev/random/unit_test.c src/sys/dev/random/unit_test.h - Improve unit tests; previously the testing human needed clairvoyance; now the test will do a basic check of compressibility. Clairvoyant talent is still a good idea. - This is still a long way off a proper unit test. * src/sys/dev/random/fortuna.c src/sys/dev/random/fortuna.h - Improve messy union to just uint128_t. - Remove unneeded 'static struct fortuna_start_cache'. - Tighten up up arithmetic. - Provide a method to allow eternal junk to be introduced; harden it against blatant by compress/hashing. - Assert that locks are held correctly. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Turn into self-sufficient module (no longer requires randomdev_soft.[ch]) * src/sys/dev/random/yarrow.c src/sys/dev/random/yarrow.h - Improve messy union to just uint128_t. - Remove unneeded 'staic struct start_cache'. - Tighten up up arithmetic. - Provide a method to allow eternal junk to be introduced; harden it against blatant by compress/hashing. - Assert that locks are held correctly. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Turn into self-sufficient module (no longer requires randomdev_soft.[ch]) - Fix some magic numbers elsewhere used as FAST and SLOW. Differential Revision: https://reviews.freebsd.org/D2025 Reviewed by: vsevolod,delphij,rwatson,trasz,jmg Approved by: so (delphij)
2015-06-30 17:00:45 +00:00
#include <sys/mount.h>
#include <sys/proc.h>
#include <sys/racct.h>
Huge cleanup of random(4) code. * GENERAL - Update copyright. - Make kernel options for RANDOM_YARROW and RANDOM_DUMMY. Set neither to ON, which means we want Fortuna - If there is no 'device random' in the kernel, there will be NO random(4) device in the kernel, and the KERN_ARND sysctl will return nothing. With RANDOM_DUMMY there will be a random(4) that always blocks. - Repair kern.arandom (KERN_ARND sysctl). The old version went through arc4random(9) and was a bit weird. - Adjust arc4random stirring a bit - the existing code looks a little suspect. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Redo read_random(9) so as to duplicate random(4)'s read internals. This makes it a first-class citizen rather than a hack. - Move stuff out of locked regions when it does not need to be there. - Trim RANDOM_DEBUG printfs. Some are excess to requirement, some behind boot verbose. - Use SYSINIT to sequence the startup. - Fix init/deinit sysctl stuff. - Make relevant sysctls also tunables. - Add different harvesting "styles" to allow for different requirements (direct, queue, fast). - Add harvesting of FFS atime events. This needs to be checked for weighing down the FS code. - Add harvesting of slab allocator events. This needs to be checked for weighing down the allocator code. - Fix the random(9) manpage. - Loadable modules are not present for now. These will be re-engineered when the dust settles. - Use macros for locks. - Fix comments. * src/share/man/... - Update the man pages. * src/etc/... - The startup/shutdown work is done in D2924. * src/UPDATING - Add UPDATING announcement. * src/sys/dev/random/build.sh - Add copyright. - Add libz for unit tests. * src/sys/dev/random/dummy.c - Remove; no longer needed. Functionality incorporated into randomdev.*. * live_entropy_sources.c live_entropy_sources.h - Remove; content moved. - move content to randomdev.[ch] and optimise. * src/sys/dev/random/random_adaptors.c src/sys/dev/random/random_adaptors.h - Remove; plugability is no longer used. Compile-time algorithm selection is the way to go. * src/sys/dev/random/random_harvestq.c src/sys/dev/random/random_harvestq.h - Add early (re)boot-time randomness caching. * src/sys/dev/random/randomdev_soft.c src/sys/dev/random/randomdev_soft.h - Remove; no longer needed. * src/sys/dev/random/uint128.h - Provide a fake uint128_t; if a real one ever arrived, we can use that instead. All that is needed here is N=0, N++, N==0, and some localised trickery is used to manufacture a 128-bit 0ULLL. * src/sys/dev/random/unit_test.c src/sys/dev/random/unit_test.h - Improve unit tests; previously the testing human needed clairvoyance; now the test will do a basic check of compressibility. Clairvoyant talent is still a good idea. - This is still a long way off a proper unit test. * src/sys/dev/random/fortuna.c src/sys/dev/random/fortuna.h - Improve messy union to just uint128_t. - Remove unneeded 'static struct fortuna_start_cache'. - Tighten up up arithmetic. - Provide a method to allow eternal junk to be introduced; harden it against blatant by compress/hashing. - Assert that locks are held correctly. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Turn into self-sufficient module (no longer requires randomdev_soft.[ch]) * src/sys/dev/random/yarrow.c src/sys/dev/random/yarrow.h - Improve messy union to just uint128_t. - Remove unneeded 'staic struct start_cache'. - Tighten up up arithmetic. - Provide a method to allow eternal junk to be introduced; harden it against blatant by compress/hashing. - Assert that locks are held correctly. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Turn into self-sufficient module (no longer requires randomdev_soft.[ch]) - Fix some magic numbers elsewhere used as FAST and SLOW. Differential Revision: https://reviews.freebsd.org/D2025 Reviewed by: vsevolod,delphij,rwatson,trasz,jmg Approved by: so (delphij)
2015-06-30 17:00:45 +00:00
#include <sys/random.h>
1994-05-24 10:09:53 +00:00
#include <sys/resourcevar.h>
#include <sys/rwlock.h>
#include <sys/stat.h>
Huge cleanup of random(4) code. * GENERAL - Update copyright. - Make kernel options for RANDOM_YARROW and RANDOM_DUMMY. Set neither to ON, which means we want Fortuna - If there is no 'device random' in the kernel, there will be NO random(4) device in the kernel, and the KERN_ARND sysctl will return nothing. With RANDOM_DUMMY there will be a random(4) that always blocks. - Repair kern.arandom (KERN_ARND sysctl). The old version went through arc4random(9) and was a bit weird. - Adjust arc4random stirring a bit - the existing code looks a little suspect. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Redo read_random(9) so as to duplicate random(4)'s read internals. This makes it a first-class citizen rather than a hack. - Move stuff out of locked regions when it does not need to be there. - Trim RANDOM_DEBUG printfs. Some are excess to requirement, some behind boot verbose. - Use SYSINIT to sequence the startup. - Fix init/deinit sysctl stuff. - Make relevant sysctls also tunables. - Add different harvesting "styles" to allow for different requirements (direct, queue, fast). - Add harvesting of FFS atime events. This needs to be checked for weighing down the FS code. - Add harvesting of slab allocator events. This needs to be checked for weighing down the allocator code. - Fix the random(9) manpage. - Loadable modules are not present for now. These will be re-engineered when the dust settles. - Use macros for locks. - Fix comments. * src/share/man/... - Update the man pages. * src/etc/... - The startup/shutdown work is done in D2924. * src/UPDATING - Add UPDATING announcement. * src/sys/dev/random/build.sh - Add copyright. - Add libz for unit tests. * src/sys/dev/random/dummy.c - Remove; no longer needed. Functionality incorporated into randomdev.*. * live_entropy_sources.c live_entropy_sources.h - Remove; content moved. - move content to randomdev.[ch] and optimise. * src/sys/dev/random/random_adaptors.c src/sys/dev/random/random_adaptors.h - Remove; plugability is no longer used. Compile-time algorithm selection is the way to go. * src/sys/dev/random/random_harvestq.c src/sys/dev/random/random_harvestq.h - Add early (re)boot-time randomness caching. * src/sys/dev/random/randomdev_soft.c src/sys/dev/random/randomdev_soft.h - Remove; no longer needed. * src/sys/dev/random/uint128.h - Provide a fake uint128_t; if a real one ever arrived, we can use that instead. All that is needed here is N=0, N++, N==0, and some localised trickery is used to manufacture a 128-bit 0ULLL. * src/sys/dev/random/unit_test.c src/sys/dev/random/unit_test.h - Improve unit tests; previously the testing human needed clairvoyance; now the test will do a basic check of compressibility. Clairvoyant talent is still a good idea. - This is still a long way off a proper unit test. * src/sys/dev/random/fortuna.c src/sys/dev/random/fortuna.h - Improve messy union to just uint128_t. - Remove unneeded 'static struct fortuna_start_cache'. - Tighten up up arithmetic. - Provide a method to allow eternal junk to be introduced; harden it against blatant by compress/hashing. - Assert that locks are held correctly. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Turn into self-sufficient module (no longer requires randomdev_soft.[ch]) * src/sys/dev/random/yarrow.c src/sys/dev/random/yarrow.h - Improve messy union to just uint128_t. - Remove unneeded 'staic struct start_cache'. - Tighten up up arithmetic. - Provide a method to allow eternal junk to be introduced; harden it against blatant by compress/hashing. - Assert that locks are held correctly. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Turn into self-sufficient module (no longer requires randomdev_soft.[ch]) - Fix some magic numbers elsewhere used as FAST and SLOW. Differential Revision: https://reviews.freebsd.org/D2025 Reviewed by: vsevolod,delphij,rwatson,trasz,jmg Approved by: so (delphij)
2015-06-30 17:00:45 +00:00
#include <sys/vmmeter.h>
#include <sys/vnode.h>
1994-05-24 10:09:53 +00:00
#include <vm/vm.h>
#include <vm/vm_extern.h>
#include <vm/vm_object.h>
1994-05-24 10:09:53 +00:00
#include <ufs/ufs/extattr.h>
1994-05-24 10:09:53 +00:00
#include <ufs/ufs/quota.h>
#include <ufs/ufs/ufsmount.h>
1994-05-24 10:09:53 +00:00
#include <ufs/ufs/inode.h>
#include <ufs/ufs/ufs_extern.h>
1994-05-24 10:09:53 +00:00
#include <ufs/ffs/fs.h>
#include <ufs/ffs/ffs_extern.h>
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
static int ffs_indirtrunc(struct inode *, ufs2_daddr_t, ufs2_daddr_t,
ufs2_daddr_t, int, ufs2_daddr_t *);
1994-05-24 10:09:53 +00:00
/*
* Update the access, modified, and inode change times as specified by the
* IN_ACCESS, IN_UPDATE, and IN_CHANGE flags respectively. Write the inode
* to disk if the IN_MODIFIED flag is set (it may be set initially, or by
* the timestamp update). The IN_LAZYMOD flag is set to force a write
* later if not now. The IN_LAZYACCESS is set instead of IN_MODIFIED if the fs
* is currently being suspended (or is suspended) and vnode has been accessed.
* If we write now, then clear IN_MODIFIED, IN_LAZYACCESS and IN_LAZYMOD to
* reflect the presumably successful write, and if waitfor is set, then wait
* for the write to complete.
1994-05-24 10:09:53 +00:00
*/
int
ffs_update(vp, waitfor)
struct vnode *vp;
int waitfor;
1994-05-24 10:09:53 +00:00
{
struct fs *fs;
1994-05-24 10:09:53 +00:00
struct buf *bp;
struct inode *ip;
This change avoids a kernel deadlock on "snaplk" when using snapshots on UFS filesystems running with journaled soft updates. This is the first of several bugs that need to be fixed before removing the restriction added in -r230250 to prevent the use of snapshots on filesystems running with journaled soft updates. The deadlock occurs when holding the snapshot lock (snaplk) and then trying to flush an inode via ffs_update(). We become blocked by another process trying to flush a different inode contained in the same inode block that we need. It holds the inode block for which we are waiting locked. When it tries to write the inode block, it gets blocked waiting for the our snaplk when it calls ffs_copyonwrite() to see if the inode block needs to be copied in our snapshot. The most obvious place that this deadlock arises is in the ffs_copyonwrite() routine when it updates critical metadata in a snapshot and tries to write it out before proceeding. The fix here is to write the data and indirect block pointer for the snapshot, but to skip the call to ffs_update() to write the snapshot inode. To ensure that we will never have to update a pointer in the inode itself, the ffs_snapshot() routine that creates the snapshot has to ensure that all the direct blocks are allocated as part of the creation of the snapshot. A less obvious place that this deadlock occurs is when we hold the snaplk because we are deleting a snapshot. In the course of doing the deletion, we need to allocate various soft update dependency structures and allocate some journal space. If we hit a resource limit while doing this we decrease the resources in use by flushing out an existing dirty file to get it to give up the soft dependency resources that it holds. The flush can cause an ffs_update() to be done on the inode for the file that we have selected to flush resulting in the same deadlock as described above when the inode that we have chosen to flush resides in the same inode block as the snapshot inode that we hold. The fix is to defer cleaning up any time that the inode on which we are operating is a snapshot. Help and review by: Jeff Roberson Tested by: Peter Holm MFC (to 9 only) after: 2 weeks
2012-03-01 18:45:25 +00:00
int flags, error;
1994-05-24 10:09:53 +00:00
ASSERT_VOP_ELOCKED(vp, "ffs_update");
ufs_itimes(vp);
ip = VTOI(vp);
if ((ip->i_flag & IN_MODIFIED) == 0 && waitfor == 0)
1994-05-24 10:09:53 +00:00
return (0);
ip->i_flag &= ~(IN_LAZYACCESS | IN_LAZYMOD | IN_MODIFIED);
fs = ITOFS(ip);
if (fs->fs_ronly && ITOUMP(ip)->um_fsckpid == 0)
return (0);
This change avoids a kernel deadlock on "snaplk" when using snapshots on UFS filesystems running with journaled soft updates. This is the first of several bugs that need to be fixed before removing the restriction added in -r230250 to prevent the use of snapshots on filesystems running with journaled soft updates. The deadlock occurs when holding the snapshot lock (snaplk) and then trying to flush an inode via ffs_update(). We become blocked by another process trying to flush a different inode contained in the same inode block that we need. It holds the inode block for which we are waiting locked. When it tries to write the inode block, it gets blocked waiting for the our snaplk when it calls ffs_copyonwrite() to see if the inode block needs to be copied in our snapshot. The most obvious place that this deadlock arises is in the ffs_copyonwrite() routine when it updates critical metadata in a snapshot and tries to write it out before proceeding. The fix here is to write the data and indirect block pointer for the snapshot, but to skip the call to ffs_update() to write the snapshot inode. To ensure that we will never have to update a pointer in the inode itself, the ffs_snapshot() routine that creates the snapshot has to ensure that all the direct blocks are allocated as part of the creation of the snapshot. A less obvious place that this deadlock occurs is when we hold the snaplk because we are deleting a snapshot. In the course of doing the deletion, we need to allocate various soft update dependency structures and allocate some journal space. If we hit a resource limit while doing this we decrease the resources in use by flushing out an existing dirty file to get it to give up the soft dependency resources that it holds. The flush can cause an ffs_update() to be done on the inode for the file that we have selected to flush resulting in the same deadlock as described above when the inode that we have chosen to flush resides in the same inode block as the snapshot inode that we hold. The fix is to defer cleaning up any time that the inode on which we are operating is a snapshot. Help and review by: Jeff Roberson Tested by: Peter Holm MFC (to 9 only) after: 2 weeks
2012-03-01 18:45:25 +00:00
/*
* If we are updating a snapshot and another process is currently
* writing the buffer containing the inode for this snapshot then
* a deadlock can occur when it tries to check the snapshot to see
* if that block needs to be copied. Thus when updating a snapshot
* we check to see if the buffer is already locked, and if it is
* we drop the snapshot lock until the buffer has been written
* and is available to us. We have to grab a reference to the
* snapshot vnode to prevent it from being removed while we are
* waiting for the buffer.
*/
flags = 0;
if (IS_SNAPSHOT(ip))
flags = GB_LOCK_NOWAIT;
loop:
Continuing efforts to provide hardening of FFS, this change adds a check hash to cylinder groups. If a check hash fails when a cylinder group is read, no further allocations are attempted in that cylinder group until it has been fixed by fsck. This avoids a class of filesystem panics related to corrupted cylinder group maps. The hash is done using crc32c. Check hases are added only to UFS2 and not to UFS1 as UFS1 is primarily used in embedded systems with small memories and low-powered processors which need as light-weight a filesystem as possible. Specifics of the changes: sys/sys/buf.h: Add BX_FSPRIV to reserve a set of eight b_xflags that may be used by individual filesystems for their own purpose. Their specific definitions are found in the header files for each filesystem that uses them. Also add fields to struct buf as noted below. sys/kern/vfs_bio.c: It is only necessary to compute a check hash for a cylinder group when it is actually read from disk. When calling bread, you do not know whether the buffer was found in the cache or read. So a new flag (GB_CKHASH) and a pointer to a function to perform the hash has been added to breadn_flags to say that the function should be called to calculate a hash if the data has been read. The check hash is placed in b_ckhash and the B_CKHASH flag is set to indicate that a read was done and a check hash calculated. Though a rather elaborate mechanism, it should also work for check hashing other metadata in the future. A kernel internal API change was to change breada into a static fucntion and add flags and a function pointer to a check-hash function. sys/ufs/ffs/fs.h: Add flags for types of check hashes; stored in a new word in the superblock. Define corresponding BX_ flags for the different types of check hashes. Add a check hash word in the cylinder group. sys/ufs/ffs/ffs_alloc.c: In ffs_getcg do the dance with breadn_flags to get a check hash and if one is provided, check it. sys/ufs/ffs/ffs_vfsops.c: Copy across the BX_FFSTYPES flags in background writes. Update the check hash when writing out buffers that need them. sys/ufs/ffs/ffs_snapshot.c: Recompute check hash when updating snapshot cylinder groups. sys/libkern/crc32.c: lib/libufs/Makefile: lib/libufs/libufs.h: lib/libufs/cgroup.c: Include libkern/crc32.c in libufs and use it to compute check hashes when updating cylinder groups. Four utilities are affected: sbin/newfs/mkfs.c: Add the check hashes when building the cylinder groups. sbin/fsck_ffs/fsck.h: sbin/fsck_ffs/fsutil.c: Verify and update check hashes when checking and writing cylinder groups. sbin/fsck_ffs/pass5.c: Offer to add check hashes to existing filesystems. Precompute check hashes when rebuilding cylinder group (although this will be done when it is written in fsutil.c it is necessary to do it early before comparing with the old cylinder group) sbin/dumpfs/dumpfs.c Print out the new check hash flag(s) sbin/fsdb/Makefile: Needs to add libufs now used by pass5.c imported from fsck_ffs. Reviewed by: kib Tested by: Peter Holm (pho)
2017-09-22 12:45:15 +00:00
error = bread_gb(ITODEVVP(ip),
This change avoids a kernel deadlock on "snaplk" when using snapshots on UFS filesystems running with journaled soft updates. This is the first of several bugs that need to be fixed before removing the restriction added in -r230250 to prevent the use of snapshots on filesystems running with journaled soft updates. The deadlock occurs when holding the snapshot lock (snaplk) and then trying to flush an inode via ffs_update(). We become blocked by another process trying to flush a different inode contained in the same inode block that we need. It holds the inode block for which we are waiting locked. When it tries to write the inode block, it gets blocked waiting for the our snaplk when it calls ffs_copyonwrite() to see if the inode block needs to be copied in our snapshot. The most obvious place that this deadlock arises is in the ffs_copyonwrite() routine when it updates critical metadata in a snapshot and tries to write it out before proceeding. The fix here is to write the data and indirect block pointer for the snapshot, but to skip the call to ffs_update() to write the snapshot inode. To ensure that we will never have to update a pointer in the inode itself, the ffs_snapshot() routine that creates the snapshot has to ensure that all the direct blocks are allocated as part of the creation of the snapshot. A less obvious place that this deadlock occurs is when we hold the snaplk because we are deleting a snapshot. In the course of doing the deletion, we need to allocate various soft update dependency structures and allocate some journal space. If we hit a resource limit while doing this we decrease the resources in use by flushing out an existing dirty file to get it to give up the soft dependency resources that it holds. The flush can cause an ffs_update() to be done on the inode for the file that we have selected to flush resulting in the same deadlock as described above when the inode that we have chosen to flush resides in the same inode block as the snapshot inode that we hold. The fix is to defer cleaning up any time that the inode on which we are operating is a snapshot. Help and review by: Jeff Roberson Tested by: Peter Holm MFC (to 9 only) after: 2 weeks
2012-03-01 18:45:25 +00:00
fsbtodb(fs, ino_to_fsba(fs, ip->i_number)),
Continuing efforts to provide hardening of FFS, this change adds a check hash to cylinder groups. If a check hash fails when a cylinder group is read, no further allocations are attempted in that cylinder group until it has been fixed by fsck. This avoids a class of filesystem panics related to corrupted cylinder group maps. The hash is done using crc32c. Check hases are added only to UFS2 and not to UFS1 as UFS1 is primarily used in embedded systems with small memories and low-powered processors which need as light-weight a filesystem as possible. Specifics of the changes: sys/sys/buf.h: Add BX_FSPRIV to reserve a set of eight b_xflags that may be used by individual filesystems for their own purpose. Their specific definitions are found in the header files for each filesystem that uses them. Also add fields to struct buf as noted below. sys/kern/vfs_bio.c: It is only necessary to compute a check hash for a cylinder group when it is actually read from disk. When calling bread, you do not know whether the buffer was found in the cache or read. So a new flag (GB_CKHASH) and a pointer to a function to perform the hash has been added to breadn_flags to say that the function should be called to calculate a hash if the data has been read. The check hash is placed in b_ckhash and the B_CKHASH flag is set to indicate that a read was done and a check hash calculated. Though a rather elaborate mechanism, it should also work for check hashing other metadata in the future. A kernel internal API change was to change breada into a static fucntion and add flags and a function pointer to a check-hash function. sys/ufs/ffs/fs.h: Add flags for types of check hashes; stored in a new word in the superblock. Define corresponding BX_ flags for the different types of check hashes. Add a check hash word in the cylinder group. sys/ufs/ffs/ffs_alloc.c: In ffs_getcg do the dance with breadn_flags to get a check hash and if one is provided, check it. sys/ufs/ffs/ffs_vfsops.c: Copy across the BX_FFSTYPES flags in background writes. Update the check hash when writing out buffers that need them. sys/ufs/ffs/ffs_snapshot.c: Recompute check hash when updating snapshot cylinder groups. sys/libkern/crc32.c: lib/libufs/Makefile: lib/libufs/libufs.h: lib/libufs/cgroup.c: Include libkern/crc32.c in libufs and use it to compute check hashes when updating cylinder groups. Four utilities are affected: sbin/newfs/mkfs.c: Add the check hashes when building the cylinder groups. sbin/fsck_ffs/fsck.h: sbin/fsck_ffs/fsutil.c: Verify and update check hashes when checking and writing cylinder groups. sbin/fsck_ffs/pass5.c: Offer to add check hashes to existing filesystems. Precompute check hashes when rebuilding cylinder group (although this will be done when it is written in fsutil.c it is necessary to do it early before comparing with the old cylinder group) sbin/dumpfs/dumpfs.c Print out the new check hash flag(s) sbin/fsdb/Makefile: Needs to add libufs now used by pass5.c imported from fsck_ffs. Reviewed by: kib Tested by: Peter Holm (pho)
2017-09-22 12:45:15 +00:00
(int) fs->fs_bsize, NOCRED, flags, &bp);
This change avoids a kernel deadlock on "snaplk" when using snapshots on UFS filesystems running with journaled soft updates. This is the first of several bugs that need to be fixed before removing the restriction added in -r230250 to prevent the use of snapshots on filesystems running with journaled soft updates. The deadlock occurs when holding the snapshot lock (snaplk) and then trying to flush an inode via ffs_update(). We become blocked by another process trying to flush a different inode contained in the same inode block that we need. It holds the inode block for which we are waiting locked. When it tries to write the inode block, it gets blocked waiting for the our snaplk when it calls ffs_copyonwrite() to see if the inode block needs to be copied in our snapshot. The most obvious place that this deadlock arises is in the ffs_copyonwrite() routine when it updates critical metadata in a snapshot and tries to write it out before proceeding. The fix here is to write the data and indirect block pointer for the snapshot, but to skip the call to ffs_update() to write the snapshot inode. To ensure that we will never have to update a pointer in the inode itself, the ffs_snapshot() routine that creates the snapshot has to ensure that all the direct blocks are allocated as part of the creation of the snapshot. A less obvious place that this deadlock occurs is when we hold the snaplk because we are deleting a snapshot. In the course of doing the deletion, we need to allocate various soft update dependency structures and allocate some journal space. If we hit a resource limit while doing this we decrease the resources in use by flushing out an existing dirty file to get it to give up the soft dependency resources that it holds. The flush can cause an ffs_update() to be done on the inode for the file that we have selected to flush resulting in the same deadlock as described above when the inode that we have chosen to flush resides in the same inode block as the snapshot inode that we hold. The fix is to defer cleaning up any time that the inode on which we are operating is a snapshot. Help and review by: Jeff Roberson Tested by: Peter Holm MFC (to 9 only) after: 2 weeks
2012-03-01 18:45:25 +00:00
if (error != 0) {
if (error != EBUSY)
This change avoids a kernel deadlock on "snaplk" when using snapshots on UFS filesystems running with journaled soft updates. This is the first of several bugs that need to be fixed before removing the restriction added in -r230250 to prevent the use of snapshots on filesystems running with journaled soft updates. The deadlock occurs when holding the snapshot lock (snaplk) and then trying to flush an inode via ffs_update(). We become blocked by another process trying to flush a different inode contained in the same inode block that we need. It holds the inode block for which we are waiting locked. When it tries to write the inode block, it gets blocked waiting for the our snaplk when it calls ffs_copyonwrite() to see if the inode block needs to be copied in our snapshot. The most obvious place that this deadlock arises is in the ffs_copyonwrite() routine when it updates critical metadata in a snapshot and tries to write it out before proceeding. The fix here is to write the data and indirect block pointer for the snapshot, but to skip the call to ffs_update() to write the snapshot inode. To ensure that we will never have to update a pointer in the inode itself, the ffs_snapshot() routine that creates the snapshot has to ensure that all the direct blocks are allocated as part of the creation of the snapshot. A less obvious place that this deadlock occurs is when we hold the snaplk because we are deleting a snapshot. In the course of doing the deletion, we need to allocate various soft update dependency structures and allocate some journal space. If we hit a resource limit while doing this we decrease the resources in use by flushing out an existing dirty file to get it to give up the soft dependency resources that it holds. The flush can cause an ffs_update() to be done on the inode for the file that we have selected to flush resulting in the same deadlock as described above when the inode that we have chosen to flush resides in the same inode block as the snapshot inode that we hold. The fix is to defer cleaning up any time that the inode on which we are operating is a snapshot. Help and review by: Jeff Roberson Tested by: Peter Holm MFC (to 9 only) after: 2 weeks
2012-03-01 18:45:25 +00:00
return (error);
KASSERT((IS_SNAPSHOT(ip)), ("EBUSY from non-snapshot"));
/*
* Wait for our inode block to become available.
*
* Hold a reference to the vnode to protect against
* ffs_snapgone(). Since we hold a reference, it can only
* get reclaimed (VI_DOOMED flag) in a forcible downgrade
* or unmount. For an unmount, the entire filesystem will be
* gone, so we cannot attempt to touch anything associated
* with it while the vnode is unlocked; all we can do is
* pause briefly and try again. If when we relock the vnode
* we discover that it has been reclaimed, updating it is no
* longer necessary and we can just return an error.
*/
vref(vp);
This change avoids a kernel deadlock on "snaplk" when using snapshots on UFS filesystems running with journaled soft updates. This is the first of several bugs that need to be fixed before removing the restriction added in -r230250 to prevent the use of snapshots on filesystems running with journaled soft updates. The deadlock occurs when holding the snapshot lock (snaplk) and then trying to flush an inode via ffs_update(). We become blocked by another process trying to flush a different inode contained in the same inode block that we need. It holds the inode block for which we are waiting locked. When it tries to write the inode block, it gets blocked waiting for the our snaplk when it calls ffs_copyonwrite() to see if the inode block needs to be copied in our snapshot. The most obvious place that this deadlock arises is in the ffs_copyonwrite() routine when it updates critical metadata in a snapshot and tries to write it out before proceeding. The fix here is to write the data and indirect block pointer for the snapshot, but to skip the call to ffs_update() to write the snapshot inode. To ensure that we will never have to update a pointer in the inode itself, the ffs_snapshot() routine that creates the snapshot has to ensure that all the direct blocks are allocated as part of the creation of the snapshot. A less obvious place that this deadlock occurs is when we hold the snaplk because we are deleting a snapshot. In the course of doing the deletion, we need to allocate various soft update dependency structures and allocate some journal space. If we hit a resource limit while doing this we decrease the resources in use by flushing out an existing dirty file to get it to give up the soft dependency resources that it holds. The flush can cause an ffs_update() to be done on the inode for the file that we have selected to flush resulting in the same deadlock as described above when the inode that we have chosen to flush resides in the same inode block as the snapshot inode that we hold. The fix is to defer cleaning up any time that the inode on which we are operating is a snapshot. Help and review by: Jeff Roberson Tested by: Peter Holm MFC (to 9 only) after: 2 weeks
2012-03-01 18:45:25 +00:00
VOP_UNLOCK(vp, 0);
pause("ffsupd", 1);
This change avoids a kernel deadlock on "snaplk" when using snapshots on UFS filesystems running with journaled soft updates. This is the first of several bugs that need to be fixed before removing the restriction added in -r230250 to prevent the use of snapshots on filesystems running with journaled soft updates. The deadlock occurs when holding the snapshot lock (snaplk) and then trying to flush an inode via ffs_update(). We become blocked by another process trying to flush a different inode contained in the same inode block that we need. It holds the inode block for which we are waiting locked. When it tries to write the inode block, it gets blocked waiting for the our snaplk when it calls ffs_copyonwrite() to see if the inode block needs to be copied in our snapshot. The most obvious place that this deadlock arises is in the ffs_copyonwrite() routine when it updates critical metadata in a snapshot and tries to write it out before proceeding. The fix here is to write the data and indirect block pointer for the snapshot, but to skip the call to ffs_update() to write the snapshot inode. To ensure that we will never have to update a pointer in the inode itself, the ffs_snapshot() routine that creates the snapshot has to ensure that all the direct blocks are allocated as part of the creation of the snapshot. A less obvious place that this deadlock occurs is when we hold the snaplk because we are deleting a snapshot. In the course of doing the deletion, we need to allocate various soft update dependency structures and allocate some journal space. If we hit a resource limit while doing this we decrease the resources in use by flushing out an existing dirty file to get it to give up the soft dependency resources that it holds. The flush can cause an ffs_update() to be done on the inode for the file that we have selected to flush resulting in the same deadlock as described above when the inode that we have chosen to flush resides in the same inode block as the snapshot inode that we hold. The fix is to defer cleaning up any time that the inode on which we are operating is a snapshot. Help and review by: Jeff Roberson Tested by: Peter Holm MFC (to 9 only) after: 2 weeks
2012-03-01 18:45:25 +00:00
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
vrele(vp);
if ((vp->v_iflag & VI_DOOMED) != 0)
return (ENOENT);
goto loop;
1994-05-24 10:09:53 +00:00
}
if (DOINGSOFTDEP(vp))
softdep_update_inodeblock(ip, bp, waitfor);
else if (ip->i_effnlink != ip->i_nlink)
panic("ffs_update: bad link cnt");
if (I_IS_UFS1(ip)) {
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
*((struct ufs1_dinode *)bp->b_data +
ino_to_fsbo(fs, ip->i_number)) = *ip->i_din1;
Huge cleanup of random(4) code. * GENERAL - Update copyright. - Make kernel options for RANDOM_YARROW and RANDOM_DUMMY. Set neither to ON, which means we want Fortuna - If there is no 'device random' in the kernel, there will be NO random(4) device in the kernel, and the KERN_ARND sysctl will return nothing. With RANDOM_DUMMY there will be a random(4) that always blocks. - Repair kern.arandom (KERN_ARND sysctl). The old version went through arc4random(9) and was a bit weird. - Adjust arc4random stirring a bit - the existing code looks a little suspect. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Redo read_random(9) so as to duplicate random(4)'s read internals. This makes it a first-class citizen rather than a hack. - Move stuff out of locked regions when it does not need to be there. - Trim RANDOM_DEBUG printfs. Some are excess to requirement, some behind boot verbose. - Use SYSINIT to sequence the startup. - Fix init/deinit sysctl stuff. - Make relevant sysctls also tunables. - Add different harvesting "styles" to allow for different requirements (direct, queue, fast). - Add harvesting of FFS atime events. This needs to be checked for weighing down the FS code. - Add harvesting of slab allocator events. This needs to be checked for weighing down the allocator code. - Fix the random(9) manpage. - Loadable modules are not present for now. These will be re-engineered when the dust settles. - Use macros for locks. - Fix comments. * src/share/man/... - Update the man pages. * src/etc/... - The startup/shutdown work is done in D2924. * src/UPDATING - Add UPDATING announcement. * src/sys/dev/random/build.sh - Add copyright. - Add libz for unit tests. * src/sys/dev/random/dummy.c - Remove; no longer needed. Functionality incorporated into randomdev.*. * live_entropy_sources.c live_entropy_sources.h - Remove; content moved. - move content to randomdev.[ch] and optimise. * src/sys/dev/random/random_adaptors.c src/sys/dev/random/random_adaptors.h - Remove; plugability is no longer used. Compile-time algorithm selection is the way to go. * src/sys/dev/random/random_harvestq.c src/sys/dev/random/random_harvestq.h - Add early (re)boot-time randomness caching. * src/sys/dev/random/randomdev_soft.c src/sys/dev/random/randomdev_soft.h - Remove; no longer needed. * src/sys/dev/random/uint128.h - Provide a fake uint128_t; if a real one ever arrived, we can use that instead. All that is needed here is N=0, N++, N==0, and some localised trickery is used to manufacture a 128-bit 0ULLL. * src/sys/dev/random/unit_test.c src/sys/dev/random/unit_test.h - Improve unit tests; previously the testing human needed clairvoyance; now the test will do a basic check of compressibility. Clairvoyant talent is still a good idea. - This is still a long way off a proper unit test. * src/sys/dev/random/fortuna.c src/sys/dev/random/fortuna.h - Improve messy union to just uint128_t. - Remove unneeded 'static struct fortuna_start_cache'. - Tighten up up arithmetic. - Provide a method to allow eternal junk to be introduced; harden it against blatant by compress/hashing. - Assert that locks are held correctly. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Turn into self-sufficient module (no longer requires randomdev_soft.[ch]) * src/sys/dev/random/yarrow.c src/sys/dev/random/yarrow.h - Improve messy union to just uint128_t. - Remove unneeded 'staic struct start_cache'. - Tighten up up arithmetic. - Provide a method to allow eternal junk to be introduced; harden it against blatant by compress/hashing. - Assert that locks are held correctly. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Turn into self-sufficient module (no longer requires randomdev_soft.[ch]) - Fix some magic numbers elsewhere used as FAST and SLOW. Differential Revision: https://reviews.freebsd.org/D2025 Reviewed by: vsevolod,delphij,rwatson,trasz,jmg Approved by: so (delphij)
2015-06-30 17:00:45 +00:00
/* XXX: FIX? The entropy here is desirable, but the harvesting may be expensive */
random_harvest_queue(&(ip->i_din1), sizeof(ip->i_din1), 1, RANDOM_FS_ATIME);
} else {
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
*((struct ufs2_dinode *)bp->b_data +
ino_to_fsbo(fs, ip->i_number)) = *ip->i_din2;
Huge cleanup of random(4) code. * GENERAL - Update copyright. - Make kernel options for RANDOM_YARROW and RANDOM_DUMMY. Set neither to ON, which means we want Fortuna - If there is no 'device random' in the kernel, there will be NO random(4) device in the kernel, and the KERN_ARND sysctl will return nothing. With RANDOM_DUMMY there will be a random(4) that always blocks. - Repair kern.arandom (KERN_ARND sysctl). The old version went through arc4random(9) and was a bit weird. - Adjust arc4random stirring a bit - the existing code looks a little suspect. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Redo read_random(9) so as to duplicate random(4)'s read internals. This makes it a first-class citizen rather than a hack. - Move stuff out of locked regions when it does not need to be there. - Trim RANDOM_DEBUG printfs. Some are excess to requirement, some behind boot verbose. - Use SYSINIT to sequence the startup. - Fix init/deinit sysctl stuff. - Make relevant sysctls also tunables. - Add different harvesting "styles" to allow for different requirements (direct, queue, fast). - Add harvesting of FFS atime events. This needs to be checked for weighing down the FS code. - Add harvesting of slab allocator events. This needs to be checked for weighing down the allocator code. - Fix the random(9) manpage. - Loadable modules are not present for now. These will be re-engineered when the dust settles. - Use macros for locks. - Fix comments. * src/share/man/... - Update the man pages. * src/etc/... - The startup/shutdown work is done in D2924. * src/UPDATING - Add UPDATING announcement. * src/sys/dev/random/build.sh - Add copyright. - Add libz for unit tests. * src/sys/dev/random/dummy.c - Remove; no longer needed. Functionality incorporated into randomdev.*. * live_entropy_sources.c live_entropy_sources.h - Remove; content moved. - move content to randomdev.[ch] and optimise. * src/sys/dev/random/random_adaptors.c src/sys/dev/random/random_adaptors.h - Remove; plugability is no longer used. Compile-time algorithm selection is the way to go. * src/sys/dev/random/random_harvestq.c src/sys/dev/random/random_harvestq.h - Add early (re)boot-time randomness caching. * src/sys/dev/random/randomdev_soft.c src/sys/dev/random/randomdev_soft.h - Remove; no longer needed. * src/sys/dev/random/uint128.h - Provide a fake uint128_t; if a real one ever arrived, we can use that instead. All that is needed here is N=0, N++, N==0, and some localised trickery is used to manufacture a 128-bit 0ULLL. * src/sys/dev/random/unit_test.c src/sys/dev/random/unit_test.h - Improve unit tests; previously the testing human needed clairvoyance; now the test will do a basic check of compressibility. Clairvoyant talent is still a good idea. - This is still a long way off a proper unit test. * src/sys/dev/random/fortuna.c src/sys/dev/random/fortuna.h - Improve messy union to just uint128_t. - Remove unneeded 'static struct fortuna_start_cache'. - Tighten up up arithmetic. - Provide a method to allow eternal junk to be introduced; harden it against blatant by compress/hashing. - Assert that locks are held correctly. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Turn into self-sufficient module (no longer requires randomdev_soft.[ch]) * src/sys/dev/random/yarrow.c src/sys/dev/random/yarrow.h - Improve messy union to just uint128_t. - Remove unneeded 'staic struct start_cache'. - Tighten up up arithmetic. - Provide a method to allow eternal junk to be introduced; harden it against blatant by compress/hashing. - Assert that locks are held correctly. - Fix the nasty pre- and post-read overloading by providing explictit functions to do these tasks. - Turn into self-sufficient module (no longer requires randomdev_soft.[ch]) - Fix some magic numbers elsewhere used as FAST and SLOW. Differential Revision: https://reviews.freebsd.org/D2025 Reviewed by: vsevolod,delphij,rwatson,trasz,jmg Approved by: so (delphij)
2015-06-30 17:00:45 +00:00
/* XXX: FIX? The entropy here is desirable, but the harvesting may be expensive */
random_harvest_queue(&(ip->i_din2), sizeof(ip->i_din2), 1, RANDOM_FS_ATIME);
}
Sprinkle DOINGASYNC() checks so as to do delayed writes for async mounts in almost all cases instead of in most cases. Don't override DOINGASYNC() by any condition except IO_SYNC. Fix previous sprinking of DOINGASYNC() checks. Don't override IO_SYNC by DOINGASYNC(). In ffs_write() and ffs_extwrite(), there were intentional overrides that just broke O_SYNC of data. In ffs_truncate(), there are 5 calls to ffs_update(), 4 with apparently-unintentional overrides and 1 without; this had no effect due to the main async mount hack descibed below. Fix 1 place in ffs_truncate() where the caller's IO_ASYNC was overridden for the soft updates case too (to do a delayed write instead of a sync write). This is supposed to be the only change that affects anything except async mounts. In ffs_update(), remove the 19 year old efficiency hack of ignoring the waitfor flag for async mounts, so that fsync() almost works for async mounts. All callers are supposed to be fixed to not ask for a sync update unless they are for fsync() or [I]O_SYNC operations. fsync() now almost works for async mounts. It used to sync the data but not the most important metdata (the inode). It still doesn't sync associated directories. This gave 10-20% fewer writes for my makeworld benchmark with async mounted tmp and obj directories from an already small number. Style fixes: - in ffs_balloc.c, remove rotted quadruplicated comments about the simplest part of the DOING*() decisions and rearrange the nearly- quadruplicated code to be more nearly so. - in ufs_vnops.c, use a consistent style with less negative logic and no manual "optimization" of || to | in DOING*() expressions. Reviewed by: kib (previous version)
2016-09-08 17:40:40 +00:00
if (waitfor)
This change avoids a kernel deadlock on "snaplk" when using snapshots on UFS filesystems running with journaled soft updates. This is the first of several bugs that need to be fixed before removing the restriction added in -r230250 to prevent the use of snapshots on filesystems running with journaled soft updates. The deadlock occurs when holding the snapshot lock (snaplk) and then trying to flush an inode via ffs_update(). We become blocked by another process trying to flush a different inode contained in the same inode block that we need. It holds the inode block for which we are waiting locked. When it tries to write the inode block, it gets blocked waiting for the our snaplk when it calls ffs_copyonwrite() to see if the inode block needs to be copied in our snapshot. The most obvious place that this deadlock arises is in the ffs_copyonwrite() routine when it updates critical metadata in a snapshot and tries to write it out before proceeding. The fix here is to write the data and indirect block pointer for the snapshot, but to skip the call to ffs_update() to write the snapshot inode. To ensure that we will never have to update a pointer in the inode itself, the ffs_snapshot() routine that creates the snapshot has to ensure that all the direct blocks are allocated as part of the creation of the snapshot. A less obvious place that this deadlock occurs is when we hold the snaplk because we are deleting a snapshot. In the course of doing the deletion, we need to allocate various soft update dependency structures and allocate some journal space. If we hit a resource limit while doing this we decrease the resources in use by flushing out an existing dirty file to get it to give up the soft dependency resources that it holds. The flush can cause an ffs_update() to be done on the inode for the file that we have selected to flush resulting in the same deadlock as described above when the inode that we have chosen to flush resides in the same inode block as the snapshot inode that we hold. The fix is to defer cleaning up any time that the inode on which we are operating is a snapshot. Help and review by: Jeff Roberson Tested by: Peter Holm MFC (to 9 only) after: 2 weeks
2012-03-01 18:45:25 +00:00
error = bwrite(bp);
else if (vm_page_count_severe() || buf_dirty_count_severe()) {
bawrite(bp);
error = 0;
} else {
if (bp->b_bufsize == fs->fs_bsize)
bp->b_flags |= B_CLUSTEROK;
1994-05-24 10:09:53 +00:00
bdwrite(bp);
This change avoids a kernel deadlock on "snaplk" when using snapshots on UFS filesystems running with journaled soft updates. This is the first of several bugs that need to be fixed before removing the restriction added in -r230250 to prevent the use of snapshots on filesystems running with journaled soft updates. The deadlock occurs when holding the snapshot lock (snaplk) and then trying to flush an inode via ffs_update(). We become blocked by another process trying to flush a different inode contained in the same inode block that we need. It holds the inode block for which we are waiting locked. When it tries to write the inode block, it gets blocked waiting for the our snaplk when it calls ffs_copyonwrite() to see if the inode block needs to be copied in our snapshot. The most obvious place that this deadlock arises is in the ffs_copyonwrite() routine when it updates critical metadata in a snapshot and tries to write it out before proceeding. The fix here is to write the data and indirect block pointer for the snapshot, but to skip the call to ffs_update() to write the snapshot inode. To ensure that we will never have to update a pointer in the inode itself, the ffs_snapshot() routine that creates the snapshot has to ensure that all the direct blocks are allocated as part of the creation of the snapshot. A less obvious place that this deadlock occurs is when we hold the snaplk because we are deleting a snapshot. In the course of doing the deletion, we need to allocate various soft update dependency structures and allocate some journal space. If we hit a resource limit while doing this we decrease the resources in use by flushing out an existing dirty file to get it to give up the soft dependency resources that it holds. The flush can cause an ffs_update() to be done on the inode for the file that we have selected to flush resulting in the same deadlock as described above when the inode that we have chosen to flush resides in the same inode block as the snapshot inode that we hold. The fix is to defer cleaning up any time that the inode on which we are operating is a snapshot. Help and review by: Jeff Roberson Tested by: Peter Holm MFC (to 9 only) after: 2 weeks
2012-03-01 18:45:25 +00:00
error = 0;
1994-05-24 10:09:53 +00:00
}
This change avoids a kernel deadlock on "snaplk" when using snapshots on UFS filesystems running with journaled soft updates. This is the first of several bugs that need to be fixed before removing the restriction added in -r230250 to prevent the use of snapshots on filesystems running with journaled soft updates. The deadlock occurs when holding the snapshot lock (snaplk) and then trying to flush an inode via ffs_update(). We become blocked by another process trying to flush a different inode contained in the same inode block that we need. It holds the inode block for which we are waiting locked. When it tries to write the inode block, it gets blocked waiting for the our snaplk when it calls ffs_copyonwrite() to see if the inode block needs to be copied in our snapshot. The most obvious place that this deadlock arises is in the ffs_copyonwrite() routine when it updates critical metadata in a snapshot and tries to write it out before proceeding. The fix here is to write the data and indirect block pointer for the snapshot, but to skip the call to ffs_update() to write the snapshot inode. To ensure that we will never have to update a pointer in the inode itself, the ffs_snapshot() routine that creates the snapshot has to ensure that all the direct blocks are allocated as part of the creation of the snapshot. A less obvious place that this deadlock occurs is when we hold the snaplk because we are deleting a snapshot. In the course of doing the deletion, we need to allocate various soft update dependency structures and allocate some journal space. If we hit a resource limit while doing this we decrease the resources in use by flushing out an existing dirty file to get it to give up the soft dependency resources that it holds. The flush can cause an ffs_update() to be done on the inode for the file that we have selected to flush resulting in the same deadlock as described above when the inode that we have chosen to flush resides in the same inode block as the snapshot inode that we hold. The fix is to defer cleaning up any time that the inode on which we are operating is a snapshot. Help and review by: Jeff Roberson Tested by: Peter Holm MFC (to 9 only) after: 2 weeks
2012-03-01 18:45:25 +00:00
return (error);
1994-05-24 10:09:53 +00:00
}
#define SINGLE 0 /* index of single indirect block */
#define DOUBLE 1 /* index of double indirect block */
#define TRIPLE 2 /* index of triple indirect block */
/*
* Truncate the inode ip to at most length size, freeing the
1994-05-24 10:09:53 +00:00
* disk blocks.
*/
int
ffs_truncate(vp, length, flags, cred)
struct vnode *vp;
off_t length;
int flags;
struct ucred *cred;
1994-05-24 10:09:53 +00:00
{
struct inode *ip;
ufs2_daddr_t bn, lbn, lastblock, lastiblock[UFS_NIADDR];
ufs2_daddr_t indir_lbn[UFS_NIADDR], oldblks[UFS_NDADDR + UFS_NIADDR];
ufs2_daddr_t newblks[UFS_NDADDR + UFS_NIADDR];
ufs2_daddr_t count, blocksreleased = 0, datablocks, blkno;
struct bufobj *bo;
struct fs *fs;
1994-05-24 10:09:53 +00:00
struct buf *bp;
struct ufsmount *ump;
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
int softdeptrunc, journaltrunc;
int needextclean, extblocks;
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
int offset, size, level, nblocks;
Sprinkle DOINGASYNC() checks so as to do delayed writes for async mounts in almost all cases instead of in most cases. Don't override DOINGASYNC() by any condition except IO_SYNC. Fix previous sprinking of DOINGASYNC() checks. Don't override IO_SYNC by DOINGASYNC(). In ffs_write() and ffs_extwrite(), there were intentional overrides that just broke O_SYNC of data. In ffs_truncate(), there are 5 calls to ffs_update(), 4 with apparently-unintentional overrides and 1 without; this had no effect due to the main async mount hack descibed below. Fix 1 place in ffs_truncate() where the caller's IO_ASYNC was overridden for the soft updates case too (to do a delayed write instead of a sync write). This is supposed to be the only change that affects anything except async mounts. In ffs_update(), remove the 19 year old efficiency hack of ignoring the waitfor flag for async mounts, so that fsync() almost works for async mounts. All callers are supposed to be fixed to not ask for a sync update unless they are for fsync() or [I]O_SYNC operations. fsync() now almost works for async mounts. It used to sync the data but not the most important metdata (the inode). It still doesn't sync associated directories. This gave 10-20% fewer writes for my makeworld benchmark with async mounted tmp and obj directories from an already small number. Style fixes: - in ffs_balloc.c, remove rotted quadruplicated comments about the simplest part of the DOING*() decisions and rearrange the nearly- quadruplicated code to be more nearly so. - in ufs_vnops.c, use a consistent style with less negative logic and no manual "optimization" of || to | in DOING*() expressions. Reviewed by: kib (previous version)
2016-09-08 17:40:40 +00:00
int i, error, allerror, indiroff, waitforupdate;
1994-05-24 10:09:53 +00:00
off_t osize;
ip = VTOI(vp);
ump = VFSTOUFS(vp->v_mount);
fs = ump->um_fs;
bo = &vp->v_bufobj;
ASSERT_VOP_LOCKED(vp, "ffs_truncate");
if (length < 0)
return (EINVAL);
if (length > fs->fs_maxfilesize)
return (EFBIG);
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
#ifdef QUOTA
error = getinoquota(ip);
if (error)
return (error);
#endif
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
/*
* Historically clients did not have to specify which data
* they were truncating. So, if not specified, we assume
* traditional behavior, e.g., just the normal data.
*/
if ((flags & (IO_EXT | IO_NORMAL)) == 0)
flags |= IO_NORMAL;
if (!DOINGSOFTDEP(vp) && !DOINGASYNC(vp))
flags |= IO_SYNC;
Sprinkle DOINGASYNC() checks so as to do delayed writes for async mounts in almost all cases instead of in most cases. Don't override DOINGASYNC() by any condition except IO_SYNC. Fix previous sprinking of DOINGASYNC() checks. Don't override IO_SYNC by DOINGASYNC(). In ffs_write() and ffs_extwrite(), there were intentional overrides that just broke O_SYNC of data. In ffs_truncate(), there are 5 calls to ffs_update(), 4 with apparently-unintentional overrides and 1 without; this had no effect due to the main async mount hack descibed below. Fix 1 place in ffs_truncate() where the caller's IO_ASYNC was overridden for the soft updates case too (to do a delayed write instead of a sync write). This is supposed to be the only change that affects anything except async mounts. In ffs_update(), remove the 19 year old efficiency hack of ignoring the waitfor flag for async mounts, so that fsync() almost works for async mounts. All callers are supposed to be fixed to not ask for a sync update unless they are for fsync() or [I]O_SYNC operations. fsync() now almost works for async mounts. It used to sync the data but not the most important metdata (the inode). It still doesn't sync associated directories. This gave 10-20% fewer writes for my makeworld benchmark with async mounted tmp and obj directories from an already small number. Style fixes: - in ffs_balloc.c, remove rotted quadruplicated comments about the simplest part of the DOING*() decisions and rearrange the nearly- quadruplicated code to be more nearly so. - in ufs_vnops.c, use a consistent style with less negative logic and no manual "optimization" of || to | in DOING*() expressions. Reviewed by: kib (previous version)
2016-09-08 17:40:40 +00:00
waitforupdate = (flags & IO_SYNC) != 0 || !DOINGASYNC(vp);
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
/*
* If we are truncating the extended-attributes, and cannot
* do it with soft updates, then do it slowly here. If we are
* truncating both the extended attributes and the file contents
* (e.g., the file is being unlinked), then pick it off with
* soft updates below.
*/
allerror = 0;
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
needextclean = 0;
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
softdeptrunc = 0;
journaltrunc = DOINGSUJ(vp);
if (journaltrunc == 0 && DOINGSOFTDEP(vp) && length == 0)
softdeptrunc = !softdep_slowdown(vp);
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
extblocks = 0;
datablocks = DIP(ip, i_blocks);
if (fs->fs_magic == FS_UFS2_MAGIC && ip->i_din2->di_extsize > 0) {
extblocks = btodb(fragroundup(fs, ip->i_din2->di_extsize));
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
datablocks -= extblocks;
}
if ((flags & IO_EXT) && extblocks > 0) {
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
if (length != 0)
panic("ffs_truncate: partial trunc of extdata");
if (softdeptrunc || journaltrunc) {
if ((flags & IO_NORMAL) == 0)
goto extclean;
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
needextclean = 1;
} else {
if ((error = ffs_syncvnode(vp, MNT_WAIT, 0)) != 0)
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
return (error);
#ifdef QUOTA
(void) chkdq(ip, -extblocks, NOCRED, 0);
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
#endif
vinvalbuf(vp, V_ALT, 0, 0);
vn_pages_remove(vp,
OFF_TO_IDX(lblktosize(fs, -extblocks)), 0);
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
osize = ip->i_din2->di_extsize;
ip->i_din2->di_blocks -= extblocks;
ip->i_din2->di_extsize = 0;
for (i = 0; i < UFS_NXADDR; i++) {
oldblks[i] = ip->i_din2->di_extb[i];
ip->i_din2->di_extb[i] = 0;
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
}
ip->i_flag |= IN_CHANGE;
Sprinkle DOINGASYNC() checks so as to do delayed writes for async mounts in almost all cases instead of in most cases. Don't override DOINGASYNC() by any condition except IO_SYNC. Fix previous sprinking of DOINGASYNC() checks. Don't override IO_SYNC by DOINGASYNC(). In ffs_write() and ffs_extwrite(), there were intentional overrides that just broke O_SYNC of data. In ffs_truncate(), there are 5 calls to ffs_update(), 4 with apparently-unintentional overrides and 1 without; this had no effect due to the main async mount hack descibed below. Fix 1 place in ffs_truncate() where the caller's IO_ASYNC was overridden for the soft updates case too (to do a delayed write instead of a sync write). This is supposed to be the only change that affects anything except async mounts. In ffs_update(), remove the 19 year old efficiency hack of ignoring the waitfor flag for async mounts, so that fsync() almost works for async mounts. All callers are supposed to be fixed to not ask for a sync update unless they are for fsync() or [I]O_SYNC operations. fsync() now almost works for async mounts. It used to sync the data but not the most important metdata (the inode). It still doesn't sync associated directories. This gave 10-20% fewer writes for my makeworld benchmark with async mounted tmp and obj directories from an already small number. Style fixes: - in ffs_balloc.c, remove rotted quadruplicated comments about the simplest part of the DOING*() decisions and rearrange the nearly- quadruplicated code to be more nearly so. - in ufs_vnops.c, use a consistent style with less negative logic and no manual "optimization" of || to | in DOING*() expressions. Reviewed by: kib (previous version)
2016-09-08 17:40:40 +00:00
if ((error = ffs_update(vp, waitforupdate)))
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
return (error);
for (i = 0; i < UFS_NXADDR; i++) {
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
if (oldblks[i] == 0)
continue;
ffs_blkfree(ump, fs, ITODEVVP(ip), oldblks[i],
sblksize(fs, osize, i), ip->i_number,
vp->v_type, NULL);
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
}
}
}
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
if ((flags & IO_NORMAL) == 0)
return (0);
if (vp->v_type == VLNK &&
(ip->i_size < vp->v_mount->mnt_maxsymlinklen ||
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
datablocks == 0)) {
#ifdef INVARIANTS
1994-05-24 10:09:53 +00:00
if (length != 0)
panic("ffs_truncate: partial truncate of symlink");
#endif
bzero(SHORTLINK(ip), (u_int)ip->i_size);
ip->i_size = 0;
DIP_SET(ip, i_size, 0);
ip->i_flag |= IN_CHANGE | IN_UPDATE;
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
if (needextclean)
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
goto extclean;
Sprinkle DOINGASYNC() checks so as to do delayed writes for async mounts in almost all cases instead of in most cases. Don't override DOINGASYNC() by any condition except IO_SYNC. Fix previous sprinking of DOINGASYNC() checks. Don't override IO_SYNC by DOINGASYNC(). In ffs_write() and ffs_extwrite(), there were intentional overrides that just broke O_SYNC of data. In ffs_truncate(), there are 5 calls to ffs_update(), 4 with apparently-unintentional overrides and 1 without; this had no effect due to the main async mount hack descibed below. Fix 1 place in ffs_truncate() where the caller's IO_ASYNC was overridden for the soft updates case too (to do a delayed write instead of a sync write). This is supposed to be the only change that affects anything except async mounts. In ffs_update(), remove the 19 year old efficiency hack of ignoring the waitfor flag for async mounts, so that fsync() almost works for async mounts. All callers are supposed to be fixed to not ask for a sync update unless they are for fsync() or [I]O_SYNC operations. fsync() now almost works for async mounts. It used to sync the data but not the most important metdata (the inode). It still doesn't sync associated directories. This gave 10-20% fewer writes for my makeworld benchmark with async mounted tmp and obj directories from an already small number. Style fixes: - in ffs_balloc.c, remove rotted quadruplicated comments about the simplest part of the DOING*() decisions and rearrange the nearly- quadruplicated code to be more nearly so. - in ufs_vnops.c, use a consistent style with less negative logic and no manual "optimization" of || to | in DOING*() expressions. Reviewed by: kib (previous version)
2016-09-08 17:40:40 +00:00
return (ffs_update(vp, waitforupdate));
1994-05-24 10:09:53 +00:00
}
if (ip->i_size == length) {
ip->i_flag |= IN_CHANGE | IN_UPDATE;
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
if (needextclean)
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
goto extclean;
return (ffs_update(vp, 0));
1994-05-24 10:09:53 +00:00
}
if (fs->fs_ronly)
panic("ffs_truncate: read-only filesystem");
This change avoids a kernel deadlock on "snaplk" when using snapshots on UFS filesystems running with journaled soft updates. This is the first of several bugs that need to be fixed before removing the restriction added in -r230250 to prevent the use of snapshots on filesystems running with journaled soft updates. The deadlock occurs when holding the snapshot lock (snaplk) and then trying to flush an inode via ffs_update(). We become blocked by another process trying to flush a different inode contained in the same inode block that we need. It holds the inode block for which we are waiting locked. When it tries to write the inode block, it gets blocked waiting for the our snaplk when it calls ffs_copyonwrite() to see if the inode block needs to be copied in our snapshot. The most obvious place that this deadlock arises is in the ffs_copyonwrite() routine when it updates critical metadata in a snapshot and tries to write it out before proceeding. The fix here is to write the data and indirect block pointer for the snapshot, but to skip the call to ffs_update() to write the snapshot inode. To ensure that we will never have to update a pointer in the inode itself, the ffs_snapshot() routine that creates the snapshot has to ensure that all the direct blocks are allocated as part of the creation of the snapshot. A less obvious place that this deadlock occurs is when we hold the snaplk because we are deleting a snapshot. In the course of doing the deletion, we need to allocate various soft update dependency structures and allocate some journal space. If we hit a resource limit while doing this we decrease the resources in use by flushing out an existing dirty file to get it to give up the soft dependency resources that it holds. The flush can cause an ffs_update() to be done on the inode for the file that we have selected to flush resulting in the same deadlock as described above when the inode that we have chosen to flush resides in the same inode block as the snapshot inode that we hold. The fix is to defer cleaning up any time that the inode on which we are operating is a snapshot. Help and review by: Jeff Roberson Tested by: Peter Holm MFC (to 9 only) after: 2 weeks
2012-03-01 18:45:25 +00:00
if (IS_SNAPSHOT(ip))
ffs_snapremove(vp);
vp->v_lasta = vp->v_clen = vp->v_cstart = vp->v_lastw = 0;
osize = ip->i_size;
/*
* Lengthen the size of the file. We must ensure that the
* last byte of the file is allocated. Since the smallest
* value of osize is 0, length will be at least 1.
*/
if (osize < length) {
vnode_pager_setsize(vp, length);
flags |= BA_CLRBUF;
error = UFS_BALLOC(vp, length - 1, 1, cred, flags, &bp);
if (error) {
vnode_pager_setsize(vp, osize);
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
return (error);
}
ip->i_size = length;
DIP_SET(ip, i_size, length);
if (bp->b_bufsize == fs->fs_bsize)
bp->b_flags |= B_CLUSTEROK;
if (flags & IO_SYNC)
bwrite(bp);
else if (DOINGASYNC(vp))
bdwrite(bp);
else
bawrite(bp);
ip->i_flag |= IN_CHANGE | IN_UPDATE;
Sprinkle DOINGASYNC() checks so as to do delayed writes for async mounts in almost all cases instead of in most cases. Don't override DOINGASYNC() by any condition except IO_SYNC. Fix previous sprinking of DOINGASYNC() checks. Don't override IO_SYNC by DOINGASYNC(). In ffs_write() and ffs_extwrite(), there were intentional overrides that just broke O_SYNC of data. In ffs_truncate(), there are 5 calls to ffs_update(), 4 with apparently-unintentional overrides and 1 without; this had no effect due to the main async mount hack descibed below. Fix 1 place in ffs_truncate() where the caller's IO_ASYNC was overridden for the soft updates case too (to do a delayed write instead of a sync write). This is supposed to be the only change that affects anything except async mounts. In ffs_update(), remove the 19 year old efficiency hack of ignoring the waitfor flag for async mounts, so that fsync() almost works for async mounts. All callers are supposed to be fixed to not ask for a sync update unless they are for fsync() or [I]O_SYNC operations. fsync() now almost works for async mounts. It used to sync the data but not the most important metdata (the inode). It still doesn't sync associated directories. This gave 10-20% fewer writes for my makeworld benchmark with async mounted tmp and obj directories from an already small number. Style fixes: - in ffs_balloc.c, remove rotted quadruplicated comments about the simplest part of the DOING*() decisions and rearrange the nearly- quadruplicated code to be more nearly so. - in ufs_vnops.c, use a consistent style with less negative logic and no manual "optimization" of || to | in DOING*() expressions. Reviewed by: kib (previous version)
2016-09-08 17:40:40 +00:00
return (ffs_update(vp, waitforupdate));
}
/*
* Lookup block number for a given offset. Zero length files
* have no blocks, so return a blkno of -1.
*/
lbn = lblkno(fs, length - 1);
if (length == 0) {
blkno = -1;
} else if (lbn < UFS_NDADDR) {
blkno = DIP(ip, i_db[lbn]);
} else {
error = UFS_BALLOC(vp, lblktosize(fs, (off_t)lbn), fs->fs_bsize,
cred, BA_METAONLY, &bp);
if (error)
return (error);
indiroff = (lbn - UFS_NDADDR) % NINDIR(fs);
if (I_IS_UFS1(ip))
blkno = ((ufs1_daddr_t *)(bp->b_data))[indiroff];
else
blkno = ((ufs2_daddr_t *)(bp->b_data))[indiroff];
/*
* If the block number is non-zero, then the indirect block
* must have been previously allocated and need not be written.
* If the block number is zero, then we may have allocated
* the indirect block and hence need to write it out.
*/
if (blkno != 0)
brelse(bp);
Sprinkle DOINGASYNC() checks so as to do delayed writes for async mounts in almost all cases instead of in most cases. Don't override DOINGASYNC() by any condition except IO_SYNC. Fix previous sprinking of DOINGASYNC() checks. Don't override IO_SYNC by DOINGASYNC(). In ffs_write() and ffs_extwrite(), there were intentional overrides that just broke O_SYNC of data. In ffs_truncate(), there are 5 calls to ffs_update(), 4 with apparently-unintentional overrides and 1 without; this had no effect due to the main async mount hack descibed below. Fix 1 place in ffs_truncate() where the caller's IO_ASYNC was overridden for the soft updates case too (to do a delayed write instead of a sync write). This is supposed to be the only change that affects anything except async mounts. In ffs_update(), remove the 19 year old efficiency hack of ignoring the waitfor flag for async mounts, so that fsync() almost works for async mounts. All callers are supposed to be fixed to not ask for a sync update unless they are for fsync() or [I]O_SYNC operations. fsync() now almost works for async mounts. It used to sync the data but not the most important metdata (the inode). It still doesn't sync associated directories. This gave 10-20% fewer writes for my makeworld benchmark with async mounted tmp and obj directories from an already small number. Style fixes: - in ffs_balloc.c, remove rotted quadruplicated comments about the simplest part of the DOING*() decisions and rearrange the nearly- quadruplicated code to be more nearly so. - in ufs_vnops.c, use a consistent style with less negative logic and no manual "optimization" of || to | in DOING*() expressions. Reviewed by: kib (previous version)
2016-09-08 17:40:40 +00:00
else if (flags & IO_SYNC)
bwrite(bp);
Sprinkle DOINGASYNC() checks so as to do delayed writes for async mounts in almost all cases instead of in most cases. Don't override DOINGASYNC() by any condition except IO_SYNC. Fix previous sprinking of DOINGASYNC() checks. Don't override IO_SYNC by DOINGASYNC(). In ffs_write() and ffs_extwrite(), there were intentional overrides that just broke O_SYNC of data. In ffs_truncate(), there are 5 calls to ffs_update(), 4 with apparently-unintentional overrides and 1 without; this had no effect due to the main async mount hack descibed below. Fix 1 place in ffs_truncate() where the caller's IO_ASYNC was overridden for the soft updates case too (to do a delayed write instead of a sync write). This is supposed to be the only change that affects anything except async mounts. In ffs_update(), remove the 19 year old efficiency hack of ignoring the waitfor flag for async mounts, so that fsync() almost works for async mounts. All callers are supposed to be fixed to not ask for a sync update unless they are for fsync() or [I]O_SYNC operations. fsync() now almost works for async mounts. It used to sync the data but not the most important metdata (the inode). It still doesn't sync associated directories. This gave 10-20% fewer writes for my makeworld benchmark with async mounted tmp and obj directories from an already small number. Style fixes: - in ffs_balloc.c, remove rotted quadruplicated comments about the simplest part of the DOING*() decisions and rearrange the nearly- quadruplicated code to be more nearly so. - in ufs_vnops.c, use a consistent style with less negative logic and no manual "optimization" of || to | in DOING*() expressions. Reviewed by: kib (previous version)
2016-09-08 17:40:40 +00:00
else
bdwrite(bp);
}
/*
* If the block number at the new end of the file is zero,
* then we must allocate it to ensure that the last block of
* the file is allocated. Soft updates does not handle this
* case, so here we have to clean up the soft updates data
* structures describing the allocation past the truncation
* point. Finding and deallocating those structures is a lot of
* work. Since partial truncation with a hole at the end occurs
* rarely, we solve the problem by syncing the file so that it
* will have no soft updates data structures left.
*/
if (blkno == 0 && (error = ffs_syncvnode(vp, MNT_WAIT, 0)) != 0)
return (error);
if (blkno != 0 && DOINGSOFTDEP(vp)) {
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
if (softdeptrunc == 0 && journaltrunc == 0) {
/*
* If soft updates cannot handle this truncation,
* clean up soft dependency data structures and
* fall through to the synchronous truncation.
*/
if ((error = ffs_syncvnode(vp, MNT_WAIT, 0)) != 0)
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
return (error);
} else {
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
flags = IO_NORMAL | (needextclean ? IO_EXT: 0);
if (journaltrunc)
softdep_journal_freeblocks(ip, cred, length,
flags);
else
softdep_setup_freeblocks(ip, length, flags);
ASSERT_VOP_LOCKED(vp, "ffs_truncate1");
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
if (journaltrunc == 0) {
ip->i_flag |= IN_CHANGE | IN_UPDATE;
error = ffs_update(vp, 0);
}
return (error);
}
1994-05-24 10:09:53 +00:00
}
/*
* Shorten the size of the file. If the last block of the
* shortened file is unallocated, we must allocate it.
* Additionally, if the file is not being truncated to a
* block boundary, the contents of the partial block
* following the end of the file must be zero'ed in
* case it ever becomes accessible again because of
* subsequent file growth. Directories however are not
* zero'ed as they should grow back initialized to empty.
1994-05-24 10:09:53 +00:00
*/
offset = blkoff(fs, length);
if (blkno != 0 && offset == 0) {
ip->i_size = length;
DIP_SET(ip, i_size, length);
1994-05-24 10:09:53 +00:00
} else {
lbn = lblkno(fs, length);
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
flags |= BA_CLRBUF;
error = UFS_BALLOC(vp, length - 1, 1, cred, flags, &bp);
if (error)
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
return (error);
/*
* When we are doing soft updates and the UFS_BALLOC
* above fills in a direct block hole with a full sized
* block that will be truncated down to a fragment below,
* we must flush out the block dependency with an FSYNC
* so that we do not get a soft updates inconsistency
* when we create the fragment below.
*/
if (DOINGSOFTDEP(vp) && lbn < UFS_NDADDR &&
fragroundup(fs, blkoff(fs, length)) < fs->fs_bsize &&
(error = ffs_syncvnode(vp, MNT_WAIT, 0)) != 0)
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
return (error);
ip->i_size = length;
DIP_SET(ip, i_size, length);
size = blksize(fs, ip, lbn);
if (vp->v_type != VDIR && offset != 0)
bzero((char *)bp->b_data + offset,
(u_int)(size - offset));
/* Kirk's code has reallocbuf(bp, size, 1) here */
allocbuf(bp, size);
if (bp->b_bufsize == fs->fs_bsize)
bp->b_flags |= B_CLUSTEROK;
Add support to UFS2 to provide storage for extended attributes. As this code is not actually used by any of the existing interfaces, it seems unlikely to break anything (famous last words). The internal kernel interface to manipulate these attributes is invoked using two new IO_ flags: IO_NORMAL and IO_EXT. These flags may be specified in the ioflags word of VOP_READ, VOP_WRITE, and VOP_TRUNCATE. Specifying IO_NORMAL means that you want to do I/O to the normal data part of the file and IO_EXT means that you want to do I/O to the extended attributes part of the file. IO_NORMAL and IO_EXT are mutually exclusive for VOP_READ and VOP_WRITE, but may be specified individually or together in the case of VOP_TRUNCATE. For example, when removing a file, VOP_TRUNCATE is called with both IO_NORMAL and IO_EXT set. For backward compatibility, if neither IO_NORMAL nor IO_EXT is set, then IO_NORMAL is assumed. Note that the BA_ and IO_ flags have been `merged' so that they may both be used in the same flags word. This merger is possible by assigning the IO_ flags to the low sixteen bits and the BA_ flags the high sixteen bits. This works because the high sixteen bits of the IO_ word is reserved for read-ahead and help with write clustering so will never be used for flags. This merge lets us get away from code of the form: if (ioflags & IO_SYNC) flags |= BA_SYNC; For the future, I have considered adding a new field to the vattr structure, va_extsize. This addition could then be exported through the stat structure to allow applications to find out the size of the extended attribute storage and also would provide a more standard interface for truncating them (via VOP_SETATTR rather than VOP_TRUNCATE). I am also contemplating adding a pathconf parameter (for concreteness, lets call it _PC_MAX_EXTSIZE) which would let an application determine the maximum size of the extended atribute storage. Sponsored by: DARPA & NAI Labs.
2002-07-19 07:29:39 +00:00
if (flags & IO_SYNC)
1994-05-24 10:09:53 +00:00
bwrite(bp);
else if (DOINGASYNC(vp))
bdwrite(bp);
1994-05-24 10:09:53 +00:00
else
bawrite(bp);
}
/*
* Calculate index into inode's block list of
* last direct and indirect blocks (if any)
* which we want to keep. Lastblock is -1 when
* the file is truncated to 0.
*/
lastblock = lblkno(fs, length + fs->fs_bsize - 1) - 1;
lastiblock[SINGLE] = lastblock - UFS_NDADDR;
1994-05-24 10:09:53 +00:00
lastiblock[DOUBLE] = lastiblock[SINGLE] - NINDIR(fs);
lastiblock[TRIPLE] = lastiblock[DOUBLE] - NINDIR(fs) * NINDIR(fs);
nblocks = btodb(fs->fs_bsize);
/*
* Update file and block pointers on disk before we start freeing
* blocks. If we crash before free'ing blocks below, the blocks
* will be returned to the free list. lastiblock values are also
* normalized to -1 for calls to ffs_indirtrunc below.
*/
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
for (level = TRIPLE; level >= SINGLE; level--) {
oldblks[UFS_NDADDR + level] = DIP(ip, i_ib[level]);
1994-05-24 10:09:53 +00:00
if (lastiblock[level] < 0) {
DIP_SET(ip, i_ib[level], 0);
1994-05-24 10:09:53 +00:00
lastiblock[level] = -1;
}
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
}
for (i = 0; i < UFS_NDADDR; i++) {
oldblks[i] = DIP(ip, i_db[i]);
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
if (i > lastblock)
DIP_SET(ip, i_db[i], 0);
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
}
ip->i_flag |= IN_CHANGE | IN_UPDATE;
Sprinkle DOINGASYNC() checks so as to do delayed writes for async mounts in almost all cases instead of in most cases. Don't override DOINGASYNC() by any condition except IO_SYNC. Fix previous sprinking of DOINGASYNC() checks. Don't override IO_SYNC by DOINGASYNC(). In ffs_write() and ffs_extwrite(), there were intentional overrides that just broke O_SYNC of data. In ffs_truncate(), there are 5 calls to ffs_update(), 4 with apparently-unintentional overrides and 1 without; this had no effect due to the main async mount hack descibed below. Fix 1 place in ffs_truncate() where the caller's IO_ASYNC was overridden for the soft updates case too (to do a delayed write instead of a sync write). This is supposed to be the only change that affects anything except async mounts. In ffs_update(), remove the 19 year old efficiency hack of ignoring the waitfor flag for async mounts, so that fsync() almost works for async mounts. All callers are supposed to be fixed to not ask for a sync update unless they are for fsync() or [I]O_SYNC operations. fsync() now almost works for async mounts. It used to sync the data but not the most important metdata (the inode). It still doesn't sync associated directories. This gave 10-20% fewer writes for my makeworld benchmark with async mounted tmp and obj directories from an already small number. Style fixes: - in ffs_balloc.c, remove rotted quadruplicated comments about the simplest part of the DOING*() decisions and rearrange the nearly- quadruplicated code to be more nearly so. - in ufs_vnops.c, use a consistent style with less negative logic and no manual "optimization" of || to | in DOING*() expressions. Reviewed by: kib (previous version)
2016-09-08 17:40:40 +00:00
allerror = ffs_update(vp, waitforupdate);
1994-05-24 10:09:53 +00:00
/*
* Having written the new inode to disk, save its new configuration
* and put back the old block pointers long enough to process them.
* Note that we save the new block configuration so we can check it
* when we are done.
*/
for (i = 0; i < UFS_NDADDR; i++) {
newblks[i] = DIP(ip, i_db[i]);
DIP_SET(ip, i_db[i], oldblks[i]);
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
}
for (i = 0; i < UFS_NIADDR; i++) {
newblks[UFS_NDADDR + i] = DIP(ip, i_ib[i]);
DIP_SET(ip, i_ib[i], oldblks[UFS_NDADDR + i]);
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
}
ip->i_size = osize;
DIP_SET(ip, i_size, osize);
error = vtruncbuf(vp, cred, length, fs->fs_bsize);
if (error && (allerror == 0))
allerror = error;
1994-05-24 10:09:53 +00:00
/*
* Indirect blocks first.
*/
indir_lbn[SINGLE] = -UFS_NDADDR;
1994-05-24 10:09:53 +00:00
indir_lbn[DOUBLE] = indir_lbn[SINGLE] - NINDIR(fs) - 1;
indir_lbn[TRIPLE] = indir_lbn[DOUBLE] - NINDIR(fs) * NINDIR(fs) - 1;
for (level = TRIPLE; level >= SINGLE; level--) {
bn = DIP(ip, i_ib[level]);
1994-05-24 10:09:53 +00:00
if (bn != 0) {
error = ffs_indirtrunc(ip, indir_lbn[level],
1994-05-24 10:09:53 +00:00
fsbtodb(fs, bn), lastiblock[level], level, &count);
if (error)
allerror = error;
blocksreleased += count;
if (lastiblock[level] < 0) {
DIP_SET(ip, i_ib[level], 0);
ffs_blkfree(ump, fs, ump->um_devvp, bn,
fs->fs_bsize, ip->i_number,
vp->v_type, NULL);
1994-05-24 10:09:53 +00:00
blocksreleased += nblocks;
}
}
if (lastiblock[level] >= 0)
goto done;
}
/*
* All whole direct blocks or frags.
*/
for (i = UFS_NDADDR - 1; i > lastblock; i--) {
long bsize;
1994-05-24 10:09:53 +00:00
bn = DIP(ip, i_db[i]);
1994-05-24 10:09:53 +00:00
if (bn == 0)
continue;
DIP_SET(ip, i_db[i], 0);
bsize = blksize(fs, ip, i);
ffs_blkfree(ump, fs, ump->um_devvp, bn, bsize, ip->i_number,
vp->v_type, NULL);
1994-05-24 10:09:53 +00:00
blocksreleased += btodb(bsize);
}
if (lastblock < 0)
goto done;
/*
* Finally, look for a change in size of the
* last direct block; release any frags.
*/
bn = DIP(ip, i_db[lastblock]);
1994-05-24 10:09:53 +00:00
if (bn != 0) {
long oldspace, newspace;
/*
* Calculate amount of space we're giving
* back as old block size minus new block size.
*/
oldspace = blksize(fs, ip, lastblock);
ip->i_size = length;
DIP_SET(ip, i_size, length);
newspace = blksize(fs, ip, lastblock);
1994-05-24 10:09:53 +00:00
if (newspace == 0)
panic("ffs_truncate: newspace");
1994-05-24 10:09:53 +00:00
if (oldspace - newspace > 0) {
/*
* Block number of space to be free'd is
* the old block # plus the number of frags
* required for the storage we're keeping.
*/
bn += numfrags(fs, newspace);
ffs_blkfree(ump, fs, ump->um_devvp, bn,
oldspace - newspace, ip->i_number, vp->v_type, NULL);
1994-05-24 10:09:53 +00:00
blocksreleased += btodb(oldspace - newspace);
}
}
done:
#ifdef INVARIANTS
1994-05-24 10:09:53 +00:00
for (level = SINGLE; level <= TRIPLE; level++)
if (newblks[UFS_NDADDR + level] != DIP(ip, i_ib[level]))
panic("ffs_truncate1");
for (i = 0; i < UFS_NDADDR; i++)
if (newblks[i] != DIP(ip, i_db[i]))
panic("ffs_truncate2");
BO_LOCK(bo);
1994-05-24 10:09:53 +00:00
if (length == 0 &&
(fs->fs_magic != FS_UFS2_MAGIC || ip->i_din2->di_extsize == 0) &&
(bo->bo_dirty.bv_cnt > 0 || bo->bo_clean.bv_cnt > 0))
panic("ffs_truncate3");
BO_UNLOCK(bo);
#endif /* INVARIANTS */
1994-05-24 10:09:53 +00:00
/*
* Put back the real size.
*/
ip->i_size = length;
DIP_SET(ip, i_size, length);
if (DIP(ip, i_blocks) >= blocksreleased)
DIP_SET(ip, i_blocks, DIP(ip, i_blocks) - blocksreleased);
else /* sanity */
DIP_SET(ip, i_blocks, 0);
ip->i_flag |= IN_CHANGE;
1994-05-24 10:09:53 +00:00
#ifdef QUOTA
(void) chkdq(ip, -blocksreleased, NOCRED, 0);
1994-05-24 10:09:53 +00:00
#endif
Implement fully asynchronous partial truncation with softupdates journaling to resolve errors which can cause corruption on recovery with the old synchronous mechanism. - Append partial truncation freework structures to indirdeps while truncation is proceeding. These prevent new block pointers from becoming valid until truncation completes and serialize truncations. - On completion of a partial truncate journal work waits for zeroed pointers to hit indirects. - softdep_journal_freeblocks() handles last frag allocation and last block zeroing. - vtruncbuf/ffs_page_remove moved into softdep_*_freeblocks() so it is only implemented in one place. - Block allocation failure handling moved up one level so it does not proceed with buf locks held. This permits us to do more extensive reclaims when filesystem space is exhausted. - softdep_sync_metadata() is broken into two parts, the first executes once at the start of ffs_syncvnode() and flushes truncations and inode dependencies. The second is called on each locked buf. This eliminates excessive looping and rollbacks. - Improve the mechanism in process_worklist_item() that handles acquiring vnode locks for handle_workitem_remove() so that it works more generally and does not loop excessively over the same worklist items on each call. - Don't corrupt directories by zeroing the tail in fsck. This is only done for regular files. - Push a fsync complete record for files that need it so the checker knows a truncation in the journal is no longer valid. Discussed with: mckusick, kib (ffs_pages_remove and ffs_truncate parts) Tested by: pho
2011-06-10 22:48:35 +00:00
return (allerror);
extclean:
if (journaltrunc)
softdep_journal_freeblocks(ip, cred, length, IO_EXT);
else
softdep_setup_freeblocks(ip, length, IO_EXT);
Sprinkle DOINGASYNC() checks so as to do delayed writes for async mounts in almost all cases instead of in most cases. Don't override DOINGASYNC() by any condition except IO_SYNC. Fix previous sprinking of DOINGASYNC() checks. Don't override IO_SYNC by DOINGASYNC(). In ffs_write() and ffs_extwrite(), there were intentional overrides that just broke O_SYNC of data. In ffs_truncate(), there are 5 calls to ffs_update(), 4 with apparently-unintentional overrides and 1 without; this had no effect due to the main async mount hack descibed below. Fix 1 place in ffs_truncate() where the caller's IO_ASYNC was overridden for the soft updates case too (to do a delayed write instead of a sync write). This is supposed to be the only change that affects anything except async mounts. In ffs_update(), remove the 19 year old efficiency hack of ignoring the waitfor flag for async mounts, so that fsync() almost works for async mounts. All callers are supposed to be fixed to not ask for a sync update unless they are for fsync() or [I]O_SYNC operations. fsync() now almost works for async mounts. It used to sync the data but not the most important metdata (the inode). It still doesn't sync associated directories. This gave 10-20% fewer writes for my makeworld benchmark with async mounted tmp and obj directories from an already small number. Style fixes: - in ffs_balloc.c, remove rotted quadruplicated comments about the simplest part of the DOING*() decisions and rearrange the nearly- quadruplicated code to be more nearly so. - in ufs_vnops.c, use a consistent style with less negative logic and no manual "optimization" of || to | in DOING*() expressions. Reviewed by: kib (previous version)
2016-09-08 17:40:40 +00:00
return (ffs_update(vp, waitforupdate));
1994-05-24 10:09:53 +00:00
}
/*
* Release blocks associated with the inode ip and stored in the indirect
* block bn. Blocks are free'd in LIFO order up to (but not including)
* lastbn. If level is greater than SINGLE, the block is an indirect block
* and recursive calls to indirtrunc must be used to cleanse other indirect
* blocks.
*/
static int
ffs_indirtrunc(ip, lbn, dbn, lastbn, level, countp)
struct inode *ip;
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
ufs2_daddr_t lbn, lastbn;
ufs2_daddr_t dbn;
1994-05-24 10:09:53 +00:00
int level;
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
ufs2_daddr_t *countp;
1994-05-24 10:09:53 +00:00
{
struct buf *bp;
struct fs *fs;
1994-05-24 10:09:53 +00:00
struct vnode *vp;
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
caddr_t copy = NULL;
int i, nblocks, error = 0, allerror = 0;
ufs2_daddr_t nb, nlbn, last;
ufs2_daddr_t blkcount, factor, blocksreleased = 0;
ufs1_daddr_t *bap1 = NULL;
ufs2_daddr_t *bap2 = NULL;
#define BAP(ip, i) (I_IS_UFS1(ip) ? bap1[i] : bap2[i])
fs = ITOFS(ip);
1994-05-24 10:09:53 +00:00
/*
* Calculate index in current block of last
* block to be kept. -1 indicates the entire
* block so we need not calculate the index.
*/
factor = lbn_offset(fs, level);
1994-05-24 10:09:53 +00:00
last = lastbn;
if (lastbn > 0)
last /= factor;
nblocks = btodb(fs->fs_bsize);
/*
* Get buffer of block pointers, zero those entries corresponding
* to blocks to be free'd, and update on disk copy first. Since
* double(triple) indirect before single(double) indirect, calls
* to bmap on these blocks will fail. However, we already have
* the on disk address, so we have to set the b_blkno field
* explicitly instead of letting bread do everything for us.
*/
vp = ITOV(ip);
bp = getblk(vp, lbn, (int)fs->fs_bsize, 0, 0, 0);
1995-03-04 03:24:45 +00:00
if ((bp->b_flags & B_CACHE) == 0) {
#ifdef RACCT
if (racct_enable) {
PROC_LOCK(curproc);
racct_add_buf(curproc, bp, 0);
PROC_UNLOCK(curproc);
}
#endif /* RACCT */
curthread->td_ru.ru_inblock++; /* pay for read */
bp->b_iocmd = BIO_READ;
bp->b_flags &= ~B_INVAL;
bp->b_ioflags &= ~BIO_ERROR;
1994-05-24 10:09:53 +00:00
if (bp->b_bcount > bp->b_bufsize)
panic("ffs_indirtrunc: bad buffer size");
bp->b_blkno = dbn;
These changes embody the support of the fully coherent merged VM buffer cache, much higher filesystem I/O performance, and much better paging performance. It represents the culmination of over 6 months of R&D. The majority of the merged VM/cache work is by John Dyson. The following highlights the most significant changes. Additionally, there are (mostly minor) changes to the various filesystem modules (nfs, msdosfs, etc) to support the new VM/buffer scheme. vfs_bio.c: Significant rewrite of most of vfs_bio to support the merged VM buffer cache scheme. The scheme is almost fully compatible with the old filesystem interface. Significant improvement in the number of opportunities for write clustering. vfs_cluster.c, vfs_subr.c Upgrade and performance enhancements in vfs layer code to support merged VM/buffer cache. Fixup of vfs_cluster to eliminate the bogus pagemove stuff. vm_object.c: Yet more improvements in the collapse code. Elimination of some windows that can cause list corruption. vm_pageout.c: Fixed it, it really works better now. Somehow in 2.0, some "enhancements" broke the code. This code has been reworked from the ground-up. vm_fault.c, vm_page.c, pmap.c, vm_object.c Support for small-block filesystems with merged VM/buffer cache scheme. pmap.c vm_map.c Dynamic kernel VM size, now we dont have to pre-allocate excessive numbers of kernel PTs. vm_glue.c Much simpler and more effective swapping code. No more gratuitous swapping. proc.h Fixed the problem that the p_lock flag was not being cleared on a fork. swap_pager.c, vnode_pager.c Removal of old vfs_bio cruft to support the past pseudo-coherency. Now the code doesn't need it anymore. machdep.c Changes to better support the parameter values for the merged VM/buffer cache scheme. machdep.c, kern_exec.c, vm_glue.c Implemented a seperate submap for temporary exec string space and another one to contain process upages. This eliminates all map fragmentation problems that previously existed. ffs_inode.c, ufs_inode.c, ufs_readwrite.c Changes for merged VM/buffer cache. Add "bypass" support for sneaking in on busy buffers. Submitted by: John Dyson and David Greenman
1995-01-09 16:06:02 +00:00
vfs_busy_pages(bp, 0);
bp->b_iooffset = dbtob(bp->b_blkno);
bstrategy(bp);
error = bufwait(bp);
1994-05-24 10:09:53 +00:00
}
if (error) {
brelse(bp);
*countp = 0;
return (error);
}
if (I_IS_UFS1(ip))
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
bap1 = (ufs1_daddr_t *)bp->b_data;
else
bap2 = (ufs2_daddr_t *)bp->b_data;
if (lastbn != -1) {
copy = malloc(fs->fs_bsize, M_TEMP, M_WAITOK);
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
bcopy((caddr_t)bp->b_data, copy, (u_int)fs->fs_bsize);
for (i = last + 1; i < NINDIR(fs); i++)
if (I_IS_UFS1(ip))
bap1[i] = 0;
else
bap2[i] = 0;
if (DOINGASYNC(vp)) {
bdwrite(bp);
} else {
error = bwrite(bp);
if (error)
allerror = error;
}
if (I_IS_UFS1(ip))
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
bap1 = (ufs1_daddr_t *)copy;
else
bap2 = (ufs2_daddr_t *)copy;
}
1994-05-24 10:09:53 +00:00
/*
* Recursively free totally unused blocks.
*/
for (i = NINDIR(fs) - 1, nlbn = lbn + 1 - i * factor; i > last;
i--, nlbn += factor) {
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
nb = BAP(ip, i);
1994-05-24 10:09:53 +00:00
if (nb == 0)
continue;
if (level > SINGLE) {
if ((error = ffs_indirtrunc(ip, nlbn, fsbtodb(fs, nb),
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
(ufs2_daddr_t)-1, level - 1, &blkcount)) != 0)
1994-05-24 10:09:53 +00:00
allerror = error;
blocksreleased += blkcount;
}
ffs_blkfree(ITOUMP(ip), fs, ITODEVVP(ip), nb, fs->fs_bsize,
ip->i_number, vp->v_type, NULL);
1994-05-24 10:09:53 +00:00
blocksreleased += nblocks;
}
/*
* Recursively free last partial block.
*/
if (level > SINGLE && lastbn >= 0) {
last = lastbn % factor;
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
nb = BAP(ip, i);
1994-05-24 10:09:53 +00:00
if (nb != 0) {
error = ffs_indirtrunc(ip, nlbn, fsbtodb(fs, nb),
last, level - 1, &blkcount);
if (error)
1994-05-24 10:09:53 +00:00
allerror = error;
blocksreleased += blkcount;
}
}
if (copy != NULL) {
free(copy, M_TEMP);
} else {
bp->b_flags |= B_INVAL | B_NOCACHE;
brelse(bp);
}
This commit adds basic support for the UFS2 filesystem. The UFS2 filesystem expands the inode to 256 bytes to make space for 64-bit block pointers. It also adds a file-creation time field, an ability to use jumbo blocks per inode to allow extent like pointer density, and space for extended attributes (up to twice the filesystem block size worth of attributes, e.g., on a 16K filesystem, there is space for 32K of attributes). UFS2 fully supports and runs existing UFS1 filesystems. New filesystems built using newfs can be built in either UFS1 or UFS2 format using the -O option. In this commit UFS1 is the default format, so if you want to build UFS2 format filesystems, you must specify -O 2. This default will be changed to UFS2 when UFS2 proves itself to be stable. In this commit the boot code for reading UFS2 filesystems is not compiled (see /sys/boot/common/ufsread.c) as there is insufficient space in the boot block. Once the size of the boot block is increased, this code can be defined. Things to note: the definition of SBSIZE has changed to SBLOCKSIZE. The header file <ufs/ufs/dinode.h> must be included before <ufs/ffs/fs.h> so as to get the definitions of ufs2_daddr_t and ufs_lbn_t. Still TODO: Verify that the first level bootstraps work for all the architectures. Convert the utility ffsinfo to understand UFS2 and test growfs. Add support for the extended attribute storage. Update soft updates to ensure integrity of extended attribute storage. Switch the current extended attribute interfaces to use the extended attribute storage. Add the extent like functionality (framework is there, but is currently never used). Sponsored by: DARPA & NAI Labs. Reviewed by: Poul-Henning Kamp <phk@freebsd.org>
2002-06-21 06:18:05 +00:00
1994-05-24 10:09:53 +00:00
*countp = blocksreleased;
return (allerror);
}
int
ffs_rdonly(struct inode *ip)
{
return (ITOFS(ip)->fs_ronly != 0);
}