OpenCloudOS-Kernel/fs/reiserfs/super.c

2655 lines
70 KiB
C
Raw Permalink Normal View History

/*
* Copyright 2000 by Hans Reiser, licensing governed by reiserfs/README
*
* Trivial changes by Alan Cox to add the LFS fixes
*
* Trivial Changes:
* Rights granted to Hans Reiser to redistribute under other terms providing
* he accepts all liability including but not limited to patent, fitness
* for purpose, and direct or indirect claims arising from failure to perform.
*
* NO WARRANTY
*/
#include <linux/module.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
#include <linux/vmalloc.h>
#include <linux/time.h>
#include <linux/uaccess.h>
#include "reiserfs.h"
#include "acl.h"
#include "xattr.h"
#include <linux/init.h>
#include <linux/blkdev.h>
#include <linux/backing-dev.h>
#include <linux/buffer_head.h>
#include <linux/exportfs.h>
#include <linux/quotaops.h>
#include <linux/vfs.h>
#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/crc32.h>
#include <linux/seq_file.h>
struct file_system_type reiserfs_fs_type;
static const char reiserfs_3_5_magic_string[] = REISERFS_SUPER_MAGIC_STRING;
static const char reiserfs_3_6_magic_string[] = REISER2FS_SUPER_MAGIC_STRING;
static const char reiserfs_jr_magic_string[] = REISER2FS_JR_SUPER_MAGIC_STRING;
int is_reiserfs_3_5(struct reiserfs_super_block *rs)
{
return !strncmp(rs->s_v1.s_magic, reiserfs_3_5_magic_string,
strlen(reiserfs_3_5_magic_string));
}
int is_reiserfs_3_6(struct reiserfs_super_block *rs)
{
return !strncmp(rs->s_v1.s_magic, reiserfs_3_6_magic_string,
strlen(reiserfs_3_6_magic_string));
}
int is_reiserfs_jr(struct reiserfs_super_block *rs)
{
return !strncmp(rs->s_v1.s_magic, reiserfs_jr_magic_string,
strlen(reiserfs_jr_magic_string));
}
static int is_any_reiserfs_magic_string(struct reiserfs_super_block *rs)
{
return (is_reiserfs_3_5(rs) || is_reiserfs_3_6(rs) ||
is_reiserfs_jr(rs));
}
static int reiserfs_remount(struct super_block *s, int *flags, char *data);
static int reiserfs_statfs(struct dentry *dentry, struct kstatfs *buf);
static int reiserfs_sync_fs(struct super_block *s, int wait)
{
struct reiserfs_transaction_handle th;
/*
* Writeback quota in non-journalled quota case - journalled quota has
* no dirty dquots
*/
dquot_writeback_dquots(s, -1);
reiserfs_write_lock(s);
if (!journal_begin(&th, s, 1))
if (!journal_end_sync(&th))
reiserfs_flush_old_commits(s);
reiserfs_write_unlock(s);
return 0;
}
static void flush_old_commits(struct work_struct *work)
{
struct reiserfs_sb_info *sbi;
struct super_block *s;
sbi = container_of(work, struct reiserfs_sb_info, old_work.work);
s = sbi->s_journal->j_work_sb;
/*
* We need s_umount for protecting quota writeback. We have to use
* trylock as reiserfs_cancel_old_flush() may be waiting for this work
* to complete with s_umount held.
*/
if (!down_read_trylock(&s->s_umount)) {
/* Requeue work if we are not cancelling it */
spin_lock(&sbi->old_work_lock);
if (sbi->work_queued == 1)
queue_delayed_work(system_long_wq, &sbi->old_work, HZ);
spin_unlock(&sbi->old_work_lock);
return;
}
spin_lock(&sbi->old_work_lock);
/* Avoid clobbering the cancel state... */
if (sbi->work_queued == 1)
sbi->work_queued = 0;
spin_unlock(&sbi->old_work_lock);
reiserfs_sync_fs(s, 1);
up_read(&s->s_umount);
}
void reiserfs_schedule_old_flush(struct super_block *s)
{
struct reiserfs_sb_info *sbi = REISERFS_SB(s);
unsigned long delay;
/*
* Avoid scheduling flush when sb is being shut down. It can race
* with journal shutdown and free still queued delayed work.
*/
Rename superblock flags (MS_xyz -> SB_xyz) This is a pure automated search-and-replace of the internal kernel superblock flags. The s_flags are now called SB_*, with the names and the values for the moment mirroring the MS_* flags that they're equivalent to. Note how the MS_xyz flags are the ones passed to the mount system call, while the SB_xyz flags are what we then use in sb->s_flags. The script to do this was: # places to look in; re security/*: it generally should *not* be # touched (that stuff parses mount(2) arguments directly), but # there are two places where we really deal with superblock flags. FILES="drivers/mtd drivers/staging/lustre fs ipc mm \ include/linux/fs.h include/uapi/linux/bfs_fs.h \ security/apparmor/apparmorfs.c security/apparmor/include/lib.h" # the list of MS_... constants SYMS="RDONLY NOSUID NODEV NOEXEC SYNCHRONOUS REMOUNT MANDLOCK \ DIRSYNC NOATIME NODIRATIME BIND MOVE REC VERBOSE SILENT \ POSIXACL UNBINDABLE PRIVATE SLAVE SHARED RELATIME KERNMOUNT \ I_VERSION STRICTATIME LAZYTIME SUBMOUNT NOREMOTELOCK NOSEC BORN \ ACTIVE NOUSER" SED_PROG= for i in $SYMS; do SED_PROG="$SED_PROG -e s/MS_$i/SB_$i/g"; done # we want files that contain at least one of MS_..., # with fs/namespace.c and fs/pnode.c excluded. L=$(for i in $SYMS; do git grep -w -l MS_$i $FILES; done| sort|uniq|grep -v '^fs/namespace.c'|grep -v '^fs/pnode.c') for f in $L; do sed -i $f $SED_PROG; done Requested-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-28 05:05:09 +08:00
if (sb_rdonly(s) || !(s->s_flags & SB_ACTIVE))
return;
spin_lock(&sbi->old_work_lock);
if (!sbi->work_queued) {
delay = msecs_to_jiffies(dirty_writeback_interval * 10);
queue_delayed_work(system_long_wq, &sbi->old_work, delay);
sbi->work_queued = 1;
}
spin_unlock(&sbi->old_work_lock);
}
void reiserfs_cancel_old_flush(struct super_block *s)
{
struct reiserfs_sb_info *sbi = REISERFS_SB(s);
spin_lock(&sbi->old_work_lock);
/* Make sure no new flushes will be queued */
sbi->work_queued = 2;
spin_unlock(&sbi->old_work_lock);
cancel_delayed_work_sync(&REISERFS_SB(s)->old_work);
}
filesystem freeze: add error handling of write_super_lockfs/unlockfs Currently, ext3 in mainline Linux doesn't have the freeze feature which suspends write requests. So, we cannot take a backup which keeps the filesystem's consistency with the storage device's features (snapshot and replication) while it is mounted. In many case, a commercial filesystem (e.g. VxFS) has the freeze feature and it would be used to get the consistent backup. If Linux's standard filesystem ext3 has the freeze feature, we can do it without a commercial filesystem. So I have implemented the ioctls of the freeze feature. I think we can take the consistent backup with the following steps. 1. Freeze the filesystem with the freeze ioctl. 2. Separate the replication volume or create the snapshot with the storage device's feature. 3. Unfreeze the filesystem with the unfreeze ioctl. 4. Take the backup from the separated replication volume or the snapshot. This patch: VFS: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that they can return an error. Rename write_super_lockfs and unlockfs of the super block operation freeze_fs and unfreeze_fs to avoid a confusion. ext3, ext4, xfs, gfs2, jfs: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that write_super_lockfs returns an error if needed, and unlockfs always returns 0. reiserfs: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that they always return 0 (success) to keep a current behavior. Signed-off-by: Takashi Sato <t-sato@yk.jp.nec.com> Signed-off-by: Masayuki Hamaguchi <m-hamaguchi@ys.jp.nec.com> Cc: <xfs-masters@oss.sgi.com> Cc: <linux-ext4@vger.kernel.org> Cc: Christoph Hellwig <hch@lst.de> Cc: Dave Kleikamp <shaggy@austin.ibm.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Alasdair G Kergon <agk@redhat.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-10 08:40:58 +08:00
static int reiserfs_freeze(struct super_block *s)
{
struct reiserfs_transaction_handle th;
reiserfs_cancel_old_flush(s);
reiserfs_write_lock(s);
if (!sb_rdonly(s)) {
int err = journal_begin(&th, s, 1);
if (err) {
reiserfs_block_writes(&th);
} else {
reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s),
1);
journal_mark_dirty(&th, SB_BUFFER_WITH_SB(s));
reiserfs_block_writes(&th);
journal_end_sync(&th);
}
}
reiserfs_write_unlock(s);
filesystem freeze: add error handling of write_super_lockfs/unlockfs Currently, ext3 in mainline Linux doesn't have the freeze feature which suspends write requests. So, we cannot take a backup which keeps the filesystem's consistency with the storage device's features (snapshot and replication) while it is mounted. In many case, a commercial filesystem (e.g. VxFS) has the freeze feature and it would be used to get the consistent backup. If Linux's standard filesystem ext3 has the freeze feature, we can do it without a commercial filesystem. So I have implemented the ioctls of the freeze feature. I think we can take the consistent backup with the following steps. 1. Freeze the filesystem with the freeze ioctl. 2. Separate the replication volume or create the snapshot with the storage device's feature. 3. Unfreeze the filesystem with the unfreeze ioctl. 4. Take the backup from the separated replication volume or the snapshot. This patch: VFS: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that they can return an error. Rename write_super_lockfs and unlockfs of the super block operation freeze_fs and unfreeze_fs to avoid a confusion. ext3, ext4, xfs, gfs2, jfs: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that write_super_lockfs returns an error if needed, and unlockfs always returns 0. reiserfs: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that they always return 0 (success) to keep a current behavior. Signed-off-by: Takashi Sato <t-sato@yk.jp.nec.com> Signed-off-by: Masayuki Hamaguchi <m-hamaguchi@ys.jp.nec.com> Cc: <xfs-masters@oss.sgi.com> Cc: <linux-ext4@vger.kernel.org> Cc: Christoph Hellwig <hch@lst.de> Cc: Dave Kleikamp <shaggy@austin.ibm.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Alasdair G Kergon <agk@redhat.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-10 08:40:58 +08:00
return 0;
}
filesystem freeze: add error handling of write_super_lockfs/unlockfs Currently, ext3 in mainline Linux doesn't have the freeze feature which suspends write requests. So, we cannot take a backup which keeps the filesystem's consistency with the storage device's features (snapshot and replication) while it is mounted. In many case, a commercial filesystem (e.g. VxFS) has the freeze feature and it would be used to get the consistent backup. If Linux's standard filesystem ext3 has the freeze feature, we can do it without a commercial filesystem. So I have implemented the ioctls of the freeze feature. I think we can take the consistent backup with the following steps. 1. Freeze the filesystem with the freeze ioctl. 2. Separate the replication volume or create the snapshot with the storage device's feature. 3. Unfreeze the filesystem with the unfreeze ioctl. 4. Take the backup from the separated replication volume or the snapshot. This patch: VFS: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that they can return an error. Rename write_super_lockfs and unlockfs of the super block operation freeze_fs and unfreeze_fs to avoid a confusion. ext3, ext4, xfs, gfs2, jfs: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that write_super_lockfs returns an error if needed, and unlockfs always returns 0. reiserfs: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that they always return 0 (success) to keep a current behavior. Signed-off-by: Takashi Sato <t-sato@yk.jp.nec.com> Signed-off-by: Masayuki Hamaguchi <m-hamaguchi@ys.jp.nec.com> Cc: <xfs-masters@oss.sgi.com> Cc: <linux-ext4@vger.kernel.org> Cc: Christoph Hellwig <hch@lst.de> Cc: Dave Kleikamp <shaggy@austin.ibm.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Alasdair G Kergon <agk@redhat.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-10 08:40:58 +08:00
static int reiserfs_unfreeze(struct super_block *s)
{
struct reiserfs_sb_info *sbi = REISERFS_SB(s);
reiserfs_allow_writes(s);
spin_lock(&sbi->old_work_lock);
/* Allow old_work to run again */
sbi->work_queued = 0;
spin_unlock(&sbi->old_work_lock);
filesystem freeze: add error handling of write_super_lockfs/unlockfs Currently, ext3 in mainline Linux doesn't have the freeze feature which suspends write requests. So, we cannot take a backup which keeps the filesystem's consistency with the storage device's features (snapshot and replication) while it is mounted. In many case, a commercial filesystem (e.g. VxFS) has the freeze feature and it would be used to get the consistent backup. If Linux's standard filesystem ext3 has the freeze feature, we can do it without a commercial filesystem. So I have implemented the ioctls of the freeze feature. I think we can take the consistent backup with the following steps. 1. Freeze the filesystem with the freeze ioctl. 2. Separate the replication volume or create the snapshot with the storage device's feature. 3. Unfreeze the filesystem with the unfreeze ioctl. 4. Take the backup from the separated replication volume or the snapshot. This patch: VFS: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that they can return an error. Rename write_super_lockfs and unlockfs of the super block operation freeze_fs and unfreeze_fs to avoid a confusion. ext3, ext4, xfs, gfs2, jfs: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that write_super_lockfs returns an error if needed, and unlockfs always returns 0. reiserfs: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that they always return 0 (success) to keep a current behavior. Signed-off-by: Takashi Sato <t-sato@yk.jp.nec.com> Signed-off-by: Masayuki Hamaguchi <m-hamaguchi@ys.jp.nec.com> Cc: <xfs-masters@oss.sgi.com> Cc: <linux-ext4@vger.kernel.org> Cc: Christoph Hellwig <hch@lst.de> Cc: Dave Kleikamp <shaggy@austin.ibm.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Alasdair G Kergon <agk@redhat.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-10 08:40:58 +08:00
return 0;
}
extern const struct in_core_key MAX_IN_CORE_KEY;
/*
* this is used to delete "save link" when there are no items of a
* file it points to. It can either happen if unlink is completed but
* "save unlink" removal, or if file has both unlink and truncate
* pending and as unlink completes first (because key of "save link"
* protecting unlink is bigger that a key lf "save link" which
* protects truncate), so there left no items to make truncate
* completion on
*/
static int remove_save_link_only(struct super_block *s,
struct reiserfs_key *key, int oid_free)
{
struct reiserfs_transaction_handle th;
int err;
/* we are going to do one balancing */
err = journal_begin(&th, s, JOURNAL_PER_BALANCE_CNT);
if (err)
return err;
reiserfs_delete_solid_item(&th, NULL, key);
if (oid_free)
/* removals are protected by direct items */
reiserfs_release_objectid(&th, le32_to_cpu(key->k_objectid));
return journal_end(&th);
}
#ifdef CONFIG_QUOTA
static int reiserfs_quota_on_mount(struct super_block *, int);
#endif
reiserfs: Unlock superblock before calling reiserfs_quota_on_mount() If we hold the superblock lock while calling reiserfs_quota_on_mount(), we can deadlock our own worker - mount blocks kworker/3:2, sleeps forever more. crash> ps|grep UN 715 2 3 ffff880220734d30 UN 0.0 0 0 [kworker/3:2] 9369 9341 2 ffff88021ffb7560 UN 1.3 493404 123184 Xorg 9665 9664 3 ffff880225b92ab0 UN 0.0 47368 812 udisks-daemon 10635 10403 3 ffff880222f22c70 UN 0.0 14904 936 mount crash> bt ffff880220734d30 PID: 715 TASK: ffff880220734d30 CPU: 3 COMMAND: "kworker/3:2" #0 [ffff8802244c3c20] schedule at ffffffff8144584b #1 [ffff8802244c3cc8] __rt_mutex_slowlock at ffffffff814472b3 #2 [ffff8802244c3d28] rt_mutex_slowlock at ffffffff814473f5 #3 [ffff8802244c3dc8] reiserfs_write_lock at ffffffffa05f28fd [reiserfs] #4 [ffff8802244c3de8] flush_async_commits at ffffffffa05ec91d [reiserfs] #5 [ffff8802244c3e08] process_one_work at ffffffff81073726 #6 [ffff8802244c3e68] worker_thread at ffffffff81073eba #7 [ffff8802244c3ec8] kthread at ffffffff810782e0 #8 [ffff8802244c3f48] kernel_thread_helper at ffffffff81450064 crash> rd ffff8802244c3cc8 10 ffff8802244c3cc8: ffffffff814472b3 ffff880222f23250 .rD.....P2.".... ffff8802244c3cd8: 0000000000000000 0000000000000286 ................ ffff8802244c3ce8: ffff8802244c3d30 ffff880220734d80 0=L$.....Ms .... ffff8802244c3cf8: ffff880222e8f628 0000000000000000 (.."............ ffff8802244c3d08: 0000000000000000 0000000000000002 ................ crash> struct rt_mutex ffff880222e8f628 struct rt_mutex { wait_lock = { raw_lock = { slock = 65537 } }, wait_list = { node_list = { next = 0xffff8802244c3d48, prev = 0xffff8802244c3d48 } }, owner = 0xffff880222f22c71, save_state = 0 } crash> bt 0xffff880222f22c70 PID: 10635 TASK: ffff880222f22c70 CPU: 3 COMMAND: "mount" #0 [ffff8802216a9868] schedule at ffffffff8144584b #1 [ffff8802216a9910] schedule_timeout at ffffffff81446865 #2 [ffff8802216a99a0] wait_for_common at ffffffff81445f74 #3 [ffff8802216a9a30] flush_work at ffffffff810712d3 #4 [ffff8802216a9ab0] schedule_on_each_cpu at ffffffff81074463 #5 [ffff8802216a9ae0] invalidate_bdev at ffffffff81178aba #6 [ffff8802216a9af0] vfs_load_quota_inode at ffffffff811a3632 #7 [ffff8802216a9b50] dquot_quota_on_mount at ffffffff811a375c #8 [ffff8802216a9b80] finish_unfinished at ffffffffa05dd8b0 [reiserfs] #9 [ffff8802216a9cc0] reiserfs_fill_super at ffffffffa05de825 [reiserfs] RIP: 00007f7b9303997a RSP: 00007ffff443c7a8 RFLAGS: 00010202 RAX: 00000000000000a5 RBX: ffffffff8144ef12 RCX: 00007f7b932e9ee0 RDX: 00007f7b93d9a400 RSI: 00007f7b93d9a3e0 RDI: 00007f7b93d9a3c0 RBP: 00007f7b93d9a2c0 R8: 00007f7b93d9a550 R9: 0000000000000001 R10: ffffffffc0ed040e R11: 0000000000000202 R12: 000000000000040e R13: 0000000000000000 R14: 00000000c0ed040e R15: 00007ffff443ca20 ORIG_RAX: 00000000000000a5 CS: 0033 SS: 002b Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Acked-by: Mike Galbraith <mgalbraith@suse.de> Cc: <stable@vger.kernel.org> Signed-off-by: Jan Kara <jack@suse.cz>
2012-08-13 21:21:23 +08:00
/*
* Look for uncompleted unlinks and truncates and complete them
*
* Called with superblock write locked. If quotas are enabled, we have to
* release/retake lest we call dquot_quota_on_mount(), proceed to
* schedule_on_each_cpu() in invalidate_bdev() and deadlock waiting for the per
* cpu worklets to complete flush_async_commits() that in turn wait for the
* superblock write lock.
*/
static int finish_unfinished(struct super_block *s)
{
INITIALIZE_PATH(path);
struct cpu_key max_cpu_key, obj_key;
struct reiserfs_key save_link_key, last_inode_key;
int retval = 0;
struct item_head *ih;
struct buffer_head *bh;
int item_pos;
char *item;
int done;
struct inode *inode;
int truncate;
#ifdef CONFIG_QUOTA
int i;
int ms_active_set;
int quota_enabled[REISERFS_MAXQUOTAS];
#endif
/* compose key to look for "save" links */
max_cpu_key.version = KEY_FORMAT_3_5;
max_cpu_key.on_disk_key.k_dir_id = ~0U;
max_cpu_key.on_disk_key.k_objectid = ~0U;
set_cpu_key_k_offset(&max_cpu_key, ~0U);
max_cpu_key.key_length = 3;
memset(&last_inode_key, 0, sizeof(last_inode_key));
#ifdef CONFIG_QUOTA
/* Needed for iput() to work correctly and not trash data */
Rename superblock flags (MS_xyz -> SB_xyz) This is a pure automated search-and-replace of the internal kernel superblock flags. The s_flags are now called SB_*, with the names and the values for the moment mirroring the MS_* flags that they're equivalent to. Note how the MS_xyz flags are the ones passed to the mount system call, while the SB_xyz flags are what we then use in sb->s_flags. The script to do this was: # places to look in; re security/*: it generally should *not* be # touched (that stuff parses mount(2) arguments directly), but # there are two places where we really deal with superblock flags. FILES="drivers/mtd drivers/staging/lustre fs ipc mm \ include/linux/fs.h include/uapi/linux/bfs_fs.h \ security/apparmor/apparmorfs.c security/apparmor/include/lib.h" # the list of MS_... constants SYMS="RDONLY NOSUID NODEV NOEXEC SYNCHRONOUS REMOUNT MANDLOCK \ DIRSYNC NOATIME NODIRATIME BIND MOVE REC VERBOSE SILENT \ POSIXACL UNBINDABLE PRIVATE SLAVE SHARED RELATIME KERNMOUNT \ I_VERSION STRICTATIME LAZYTIME SUBMOUNT NOREMOTELOCK NOSEC BORN \ ACTIVE NOUSER" SED_PROG= for i in $SYMS; do SED_PROG="$SED_PROG -e s/MS_$i/SB_$i/g"; done # we want files that contain at least one of MS_..., # with fs/namespace.c and fs/pnode.c excluded. L=$(for i in $SYMS; do git grep -w -l MS_$i $FILES; done| sort|uniq|grep -v '^fs/namespace.c'|grep -v '^fs/pnode.c') for f in $L; do sed -i $f $SED_PROG; done Requested-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-28 05:05:09 +08:00
if (s->s_flags & SB_ACTIVE) {
ms_active_set = 0;
} else {
ms_active_set = 1;
Rename superblock flags (MS_xyz -> SB_xyz) This is a pure automated search-and-replace of the internal kernel superblock flags. The s_flags are now called SB_*, with the names and the values for the moment mirroring the MS_* flags that they're equivalent to. Note how the MS_xyz flags are the ones passed to the mount system call, while the SB_xyz flags are what we then use in sb->s_flags. The script to do this was: # places to look in; re security/*: it generally should *not* be # touched (that stuff parses mount(2) arguments directly), but # there are two places where we really deal with superblock flags. FILES="drivers/mtd drivers/staging/lustre fs ipc mm \ include/linux/fs.h include/uapi/linux/bfs_fs.h \ security/apparmor/apparmorfs.c security/apparmor/include/lib.h" # the list of MS_... constants SYMS="RDONLY NOSUID NODEV NOEXEC SYNCHRONOUS REMOUNT MANDLOCK \ DIRSYNC NOATIME NODIRATIME BIND MOVE REC VERBOSE SILENT \ POSIXACL UNBINDABLE PRIVATE SLAVE SHARED RELATIME KERNMOUNT \ I_VERSION STRICTATIME LAZYTIME SUBMOUNT NOREMOTELOCK NOSEC BORN \ ACTIVE NOUSER" SED_PROG= for i in $SYMS; do SED_PROG="$SED_PROG -e s/MS_$i/SB_$i/g"; done # we want files that contain at least one of MS_..., # with fs/namespace.c and fs/pnode.c excluded. L=$(for i in $SYMS; do git grep -w -l MS_$i $FILES; done| sort|uniq|grep -v '^fs/namespace.c'|grep -v '^fs/pnode.c') for f in $L; do sed -i $f $SED_PROG; done Requested-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-28 05:05:09 +08:00
s->s_flags |= SB_ACTIVE;
}
/* Turn on quotas so that they are updated correctly */
for (i = 0; i < REISERFS_MAXQUOTAS; i++) {
quota_enabled[i] = 1;
if (REISERFS_SB(s)->s_qf_names[i]) {
int ret;
if (sb_has_quota_active(s, i)) {
quota_enabled[i] = 0;
continue;
}
reiserfs: Unlock superblock before calling reiserfs_quota_on_mount() If we hold the superblock lock while calling reiserfs_quota_on_mount(), we can deadlock our own worker - mount blocks kworker/3:2, sleeps forever more. crash> ps|grep UN 715 2 3 ffff880220734d30 UN 0.0 0 0 [kworker/3:2] 9369 9341 2 ffff88021ffb7560 UN 1.3 493404 123184 Xorg 9665 9664 3 ffff880225b92ab0 UN 0.0 47368 812 udisks-daemon 10635 10403 3 ffff880222f22c70 UN 0.0 14904 936 mount crash> bt ffff880220734d30 PID: 715 TASK: ffff880220734d30 CPU: 3 COMMAND: "kworker/3:2" #0 [ffff8802244c3c20] schedule at ffffffff8144584b #1 [ffff8802244c3cc8] __rt_mutex_slowlock at ffffffff814472b3 #2 [ffff8802244c3d28] rt_mutex_slowlock at ffffffff814473f5 #3 [ffff8802244c3dc8] reiserfs_write_lock at ffffffffa05f28fd [reiserfs] #4 [ffff8802244c3de8] flush_async_commits at ffffffffa05ec91d [reiserfs] #5 [ffff8802244c3e08] process_one_work at ffffffff81073726 #6 [ffff8802244c3e68] worker_thread at ffffffff81073eba #7 [ffff8802244c3ec8] kthread at ffffffff810782e0 #8 [ffff8802244c3f48] kernel_thread_helper at ffffffff81450064 crash> rd ffff8802244c3cc8 10 ffff8802244c3cc8: ffffffff814472b3 ffff880222f23250 .rD.....P2.".... ffff8802244c3cd8: 0000000000000000 0000000000000286 ................ ffff8802244c3ce8: ffff8802244c3d30 ffff880220734d80 0=L$.....Ms .... ffff8802244c3cf8: ffff880222e8f628 0000000000000000 (.."............ ffff8802244c3d08: 0000000000000000 0000000000000002 ................ crash> struct rt_mutex ffff880222e8f628 struct rt_mutex { wait_lock = { raw_lock = { slock = 65537 } }, wait_list = { node_list = { next = 0xffff8802244c3d48, prev = 0xffff8802244c3d48 } }, owner = 0xffff880222f22c71, save_state = 0 } crash> bt 0xffff880222f22c70 PID: 10635 TASK: ffff880222f22c70 CPU: 3 COMMAND: "mount" #0 [ffff8802216a9868] schedule at ffffffff8144584b #1 [ffff8802216a9910] schedule_timeout at ffffffff81446865 #2 [ffff8802216a99a0] wait_for_common at ffffffff81445f74 #3 [ffff8802216a9a30] flush_work at ffffffff810712d3 #4 [ffff8802216a9ab0] schedule_on_each_cpu at ffffffff81074463 #5 [ffff8802216a9ae0] invalidate_bdev at ffffffff81178aba #6 [ffff8802216a9af0] vfs_load_quota_inode at ffffffff811a3632 #7 [ffff8802216a9b50] dquot_quota_on_mount at ffffffff811a375c #8 [ffff8802216a9b80] finish_unfinished at ffffffffa05dd8b0 [reiserfs] #9 [ffff8802216a9cc0] reiserfs_fill_super at ffffffffa05de825 [reiserfs] RIP: 00007f7b9303997a RSP: 00007ffff443c7a8 RFLAGS: 00010202 RAX: 00000000000000a5 RBX: ffffffff8144ef12 RCX: 00007f7b932e9ee0 RDX: 00007f7b93d9a400 RSI: 00007f7b93d9a3e0 RDI: 00007f7b93d9a3c0 RBP: 00007f7b93d9a2c0 R8: 00007f7b93d9a550 R9: 0000000000000001 R10: ffffffffc0ed040e R11: 0000000000000202 R12: 000000000000040e R13: 0000000000000000 R14: 00000000c0ed040e R15: 00007ffff443ca20 ORIG_RAX: 00000000000000a5 CS: 0033 SS: 002b Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Acked-by: Mike Galbraith <mgalbraith@suse.de> Cc: <stable@vger.kernel.org> Signed-off-by: Jan Kara <jack@suse.cz>
2012-08-13 21:21:23 +08:00
reiserfs_write_unlock(s);
ret = reiserfs_quota_on_mount(s, i);
reiserfs: Unlock superblock before calling reiserfs_quota_on_mount() If we hold the superblock lock while calling reiserfs_quota_on_mount(), we can deadlock our own worker - mount blocks kworker/3:2, sleeps forever more. crash> ps|grep UN 715 2 3 ffff880220734d30 UN 0.0 0 0 [kworker/3:2] 9369 9341 2 ffff88021ffb7560 UN 1.3 493404 123184 Xorg 9665 9664 3 ffff880225b92ab0 UN 0.0 47368 812 udisks-daemon 10635 10403 3 ffff880222f22c70 UN 0.0 14904 936 mount crash> bt ffff880220734d30 PID: 715 TASK: ffff880220734d30 CPU: 3 COMMAND: "kworker/3:2" #0 [ffff8802244c3c20] schedule at ffffffff8144584b #1 [ffff8802244c3cc8] __rt_mutex_slowlock at ffffffff814472b3 #2 [ffff8802244c3d28] rt_mutex_slowlock at ffffffff814473f5 #3 [ffff8802244c3dc8] reiserfs_write_lock at ffffffffa05f28fd [reiserfs] #4 [ffff8802244c3de8] flush_async_commits at ffffffffa05ec91d [reiserfs] #5 [ffff8802244c3e08] process_one_work at ffffffff81073726 #6 [ffff8802244c3e68] worker_thread at ffffffff81073eba #7 [ffff8802244c3ec8] kthread at ffffffff810782e0 #8 [ffff8802244c3f48] kernel_thread_helper at ffffffff81450064 crash> rd ffff8802244c3cc8 10 ffff8802244c3cc8: ffffffff814472b3 ffff880222f23250 .rD.....P2.".... ffff8802244c3cd8: 0000000000000000 0000000000000286 ................ ffff8802244c3ce8: ffff8802244c3d30 ffff880220734d80 0=L$.....Ms .... ffff8802244c3cf8: ffff880222e8f628 0000000000000000 (.."............ ffff8802244c3d08: 0000000000000000 0000000000000002 ................ crash> struct rt_mutex ffff880222e8f628 struct rt_mutex { wait_lock = { raw_lock = { slock = 65537 } }, wait_list = { node_list = { next = 0xffff8802244c3d48, prev = 0xffff8802244c3d48 } }, owner = 0xffff880222f22c71, save_state = 0 } crash> bt 0xffff880222f22c70 PID: 10635 TASK: ffff880222f22c70 CPU: 3 COMMAND: "mount" #0 [ffff8802216a9868] schedule at ffffffff8144584b #1 [ffff8802216a9910] schedule_timeout at ffffffff81446865 #2 [ffff8802216a99a0] wait_for_common at ffffffff81445f74 #3 [ffff8802216a9a30] flush_work at ffffffff810712d3 #4 [ffff8802216a9ab0] schedule_on_each_cpu at ffffffff81074463 #5 [ffff8802216a9ae0] invalidate_bdev at ffffffff81178aba #6 [ffff8802216a9af0] vfs_load_quota_inode at ffffffff811a3632 #7 [ffff8802216a9b50] dquot_quota_on_mount at ffffffff811a375c #8 [ffff8802216a9b80] finish_unfinished at ffffffffa05dd8b0 [reiserfs] #9 [ffff8802216a9cc0] reiserfs_fill_super at ffffffffa05de825 [reiserfs] RIP: 00007f7b9303997a RSP: 00007ffff443c7a8 RFLAGS: 00010202 RAX: 00000000000000a5 RBX: ffffffff8144ef12 RCX: 00007f7b932e9ee0 RDX: 00007f7b93d9a400 RSI: 00007f7b93d9a3e0 RDI: 00007f7b93d9a3c0 RBP: 00007f7b93d9a2c0 R8: 00007f7b93d9a550 R9: 0000000000000001 R10: ffffffffc0ed040e R11: 0000000000000202 R12: 000000000000040e R13: 0000000000000000 R14: 00000000c0ed040e R15: 00007ffff443ca20 ORIG_RAX: 00000000000000a5 CS: 0033 SS: 002b Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Acked-by: Mike Galbraith <mgalbraith@suse.de> Cc: <stable@vger.kernel.org> Signed-off-by: Jan Kara <jack@suse.cz>
2012-08-13 21:21:23 +08:00
reiserfs_write_lock(s);
if (ret < 0)
reiserfs_warning(s, "reiserfs-2500",
"cannot turn on journaled "
"quota: error %d", ret);
}
}
#endif
done = 0;
REISERFS_SB(s)->s_is_unlinked_ok = 1;
while (!retval) {
int depth;
retval = search_item(s, &max_cpu_key, &path);
if (retval != ITEM_NOT_FOUND) {
reiserfs_error(s, "vs-2140",
"search_by_key returned %d", retval);
break;
}
bh = get_last_bh(&path);
item_pos = get_item_pos(&path);
if (item_pos != B_NR_ITEMS(bh)) {
reiserfs_warning(s, "vs-2060",
"wrong position found");
break;
}
item_pos--;
ih = item_head(bh, item_pos);
if (le32_to_cpu(ih->ih_key.k_dir_id) != MAX_KEY_OBJECTID)
/* there are no "save" links anymore */
break;
save_link_key = ih->ih_key;
if (is_indirect_le_ih(ih))
truncate = 1;
else
truncate = 0;
/* reiserfs_iget needs k_dirid and k_objectid only */
item = ih_item_body(bh, ih);
obj_key.on_disk_key.k_dir_id = le32_to_cpu(*(__le32 *) item);
obj_key.on_disk_key.k_objectid =
le32_to_cpu(ih->ih_key.k_objectid);
obj_key.on_disk_key.k_offset = 0;
obj_key.on_disk_key.k_type = 0;
pathrelse(&path);
inode = reiserfs_iget(s, &obj_key);
if (IS_ERR_OR_NULL(inode)) {
/*
* the unlink almost completed, it just did not
* manage to remove "save" link and release objectid
*/
reiserfs_warning(s, "vs-2180", "iget failed for %K",
&obj_key);
retval = remove_save_link_only(s, &save_link_key, 1);
continue;
}
if (!truncate && inode->i_nlink) {
/* file is not unlinked */
reiserfs_warning(s, "vs-2185",
"file %K is not unlinked",
&obj_key);
retval = remove_save_link_only(s, &save_link_key, 0);
continue;
}
depth = reiserfs_write_unlock_nested(inode->i_sb);
dquot_initialize(inode);
reiserfs_write_lock_nested(inode->i_sb, depth);
if (truncate && S_ISDIR(inode->i_mode)) {
/*
* We got a truncate request for a dir which
* is impossible. The only imaginable way is to
* execute unfinished truncate request then boot
* into old kernel, remove the file and create dir
* with the same key.
*/
reiserfs_warning(s, "green-2101",
"impossible truncate on a "
"directory %k. Please report",
INODE_PKEY(inode));
retval = remove_save_link_only(s, &save_link_key, 0);
truncate = 0;
iput(inode);
continue;
}
if (truncate) {
REISERFS_I(inode)->i_flags |=
i_link_saved_truncate_mask;
/*
* not completed truncate found. New size was
* committed together with "save" link
*/
reiserfs_info(s, "Truncating %k to %lld ..",
INODE_PKEY(inode), inode->i_size);
/* don't update modification time */
reiserfs_truncate_file(inode, 0);
retval = remove_save_link(inode, truncate);
} else {
REISERFS_I(inode)->i_flags |= i_link_saved_unlink_mask;
/* not completed unlink (rmdir) found */
reiserfs_info(s, "Removing %k..", INODE_PKEY(inode));
if (memcmp(&last_inode_key, INODE_PKEY(inode),
sizeof(last_inode_key))){
last_inode_key = *INODE_PKEY(inode);
/* removal gets completed in iput */
retval = 0;
} else {
reiserfs_warning(s, "super-2189", "Dead loop "
"in finish_unfinished "
"detected, just remove "
"save link\n");
retval = remove_save_link_only(s,
&save_link_key, 0);
}
}
iput(inode);
printk("done\n");
done++;
}
REISERFS_SB(s)->s_is_unlinked_ok = 0;
#ifdef CONFIG_QUOTA
/* Turn quotas off */
reiserfs_write_unlock(s);
for (i = 0; i < REISERFS_MAXQUOTAS; i++) {
if (sb_dqopt(s)->files[i] && quota_enabled[i])
dquot_quota_off(s, i);
}
reiserfs_write_lock(s);
if (ms_active_set)
/* Restore the flag back */
Rename superblock flags (MS_xyz -> SB_xyz) This is a pure automated search-and-replace of the internal kernel superblock flags. The s_flags are now called SB_*, with the names and the values for the moment mirroring the MS_* flags that they're equivalent to. Note how the MS_xyz flags are the ones passed to the mount system call, while the SB_xyz flags are what we then use in sb->s_flags. The script to do this was: # places to look in; re security/*: it generally should *not* be # touched (that stuff parses mount(2) arguments directly), but # there are two places where we really deal with superblock flags. FILES="drivers/mtd drivers/staging/lustre fs ipc mm \ include/linux/fs.h include/uapi/linux/bfs_fs.h \ security/apparmor/apparmorfs.c security/apparmor/include/lib.h" # the list of MS_... constants SYMS="RDONLY NOSUID NODEV NOEXEC SYNCHRONOUS REMOUNT MANDLOCK \ DIRSYNC NOATIME NODIRATIME BIND MOVE REC VERBOSE SILENT \ POSIXACL UNBINDABLE PRIVATE SLAVE SHARED RELATIME KERNMOUNT \ I_VERSION STRICTATIME LAZYTIME SUBMOUNT NOREMOTELOCK NOSEC BORN \ ACTIVE NOUSER" SED_PROG= for i in $SYMS; do SED_PROG="$SED_PROG -e s/MS_$i/SB_$i/g"; done # we want files that contain at least one of MS_..., # with fs/namespace.c and fs/pnode.c excluded. L=$(for i in $SYMS; do git grep -w -l MS_$i $FILES; done| sort|uniq|grep -v '^fs/namespace.c'|grep -v '^fs/pnode.c') for f in $L; do sed -i $f $SED_PROG; done Requested-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-28 05:05:09 +08:00
s->s_flags &= ~SB_ACTIVE;
#endif
pathrelse(&path);
if (done)
reiserfs_info(s, "There were %d uncompleted unlinks/truncates. "
"Completed\n", done);
return retval;
}
/*
* to protect file being unlinked from getting lost we "safe" link files
* being unlinked. This link will be deleted in the same transaction with last
* item of file. mounting the filesystem we scan all these links and remove
* files which almost got lost
*/
void add_save_link(struct reiserfs_transaction_handle *th,
struct inode *inode, int truncate)
{
INITIALIZE_PATH(path);
int retval;
struct cpu_key key;
struct item_head ih;
__le32 link;
BUG_ON(!th->t_trans_id);
/* file can only get one "save link" of each kind */
RFALSE(truncate &&
(REISERFS_I(inode)->i_flags & i_link_saved_truncate_mask),
"saved link already exists for truncated inode %lx",
(long)inode->i_ino);
RFALSE(!truncate &&
(REISERFS_I(inode)->i_flags & i_link_saved_unlink_mask),
"saved link already exists for unlinked inode %lx",
(long)inode->i_ino);
/* setup key of "save" link */
key.version = KEY_FORMAT_3_5;
key.on_disk_key.k_dir_id = MAX_KEY_OBJECTID;
key.on_disk_key.k_objectid = inode->i_ino;
if (!truncate) {
/* unlink, rmdir, rename */
set_cpu_key_k_offset(&key, 1 + inode->i_sb->s_blocksize);
set_cpu_key_k_type(&key, TYPE_DIRECT);
/* item head of "safe" link */
make_le_item_head(&ih, &key, key.version,
1 + inode->i_sb->s_blocksize, TYPE_DIRECT,
4 /*length */ , 0xffff /*free space */ );
} else {
/* truncate */
if (S_ISDIR(inode->i_mode))
reiserfs_warning(inode->i_sb, "green-2102",
"Adding a truncate savelink for "
"a directory %k! Please report",
INODE_PKEY(inode));
set_cpu_key_k_offset(&key, 1);
set_cpu_key_k_type(&key, TYPE_INDIRECT);
/* item head of "safe" link */
make_le_item_head(&ih, &key, key.version, 1, TYPE_INDIRECT,
4 /*length */ , 0 /*free space */ );
}
key.key_length = 3;
/* look for its place in the tree */
retval = search_item(inode->i_sb, &key, &path);
if (retval != ITEM_NOT_FOUND) {
if (retval != -ENOSPC)
reiserfs_error(inode->i_sb, "vs-2100",
"search_by_key (%K) returned %d", &key,
retval);
pathrelse(&path);
return;
}
/* body of "save" link */
link = INODE_PKEY(inode)->k_dir_id;
/* put "save" link into tree, don't charge quota to anyone */
retval =
reiserfs_insert_item(th, &path, &key, &ih, NULL, (char *)&link);
if (retval) {
if (retval != -ENOSPC)
reiserfs_error(inode->i_sb, "vs-2120",
"insert_item returned %d", retval);
} else {
if (truncate)
REISERFS_I(inode)->i_flags |=
i_link_saved_truncate_mask;
else
REISERFS_I(inode)->i_flags |= i_link_saved_unlink_mask;
}
}
/* this opens transaction unlike add_save_link */
int remove_save_link(struct inode *inode, int truncate)
{
struct reiserfs_transaction_handle th;
struct reiserfs_key key;
int err;
/* we are going to do one balancing only */
err = journal_begin(&th, inode->i_sb, JOURNAL_PER_BALANCE_CNT);
if (err)
return err;
/* setup key of "save" link */
key.k_dir_id = cpu_to_le32(MAX_KEY_OBJECTID);
key.k_objectid = INODE_PKEY(inode)->k_objectid;
if (!truncate) {
/* unlink, rmdir, rename */
set_le_key_k_offset(KEY_FORMAT_3_5, &key,
1 + inode->i_sb->s_blocksize);
set_le_key_k_type(KEY_FORMAT_3_5, &key, TYPE_DIRECT);
} else {
/* truncate */
set_le_key_k_offset(KEY_FORMAT_3_5, &key, 1);
set_le_key_k_type(KEY_FORMAT_3_5, &key, TYPE_INDIRECT);
}
if ((truncate &&
(REISERFS_I(inode)->i_flags & i_link_saved_truncate_mask)) ||
(!truncate &&
(REISERFS_I(inode)->i_flags & i_link_saved_unlink_mask)))
/* don't take quota bytes from anywhere */
reiserfs_delete_solid_item(&th, NULL, &key);
if (!truncate) {
reiserfs_release_objectid(&th, inode->i_ino);
REISERFS_I(inode)->i_flags &= ~i_link_saved_unlink_mask;
} else
REISERFS_I(inode)->i_flags &= ~i_link_saved_truncate_mask;
return journal_end(&th);
}
static void reiserfs_kill_sb(struct super_block *s)
{
if (REISERFS_SB(s)) {
reiserfs_proc_info_done(s);
/*
* Force any pending inode evictions to occur now. Any
* inodes to be removed that have extended attributes
* associated with them need to clean them up before
* we can release the extended attribute root dentries.
* shrink_dcache_for_umount will BUG if we don't release
* those before it's called so ->put_super is too late.
*/
shrink_dcache_sb(s);
dput(REISERFS_SB(s)->xattr_root);
REISERFS_SB(s)->xattr_root = NULL;
dput(REISERFS_SB(s)->priv_root);
REISERFS_SB(s)->priv_root = NULL;
}
kill_block_super(s);
}
#ifdef CONFIG_QUOTA
static int reiserfs_quota_off(struct super_block *sb, int type);
static void reiserfs_quota_off_umount(struct super_block *s)
{
int type;
for (type = 0; type < REISERFS_MAXQUOTAS; type++)
reiserfs_quota_off(s, type);
}
#else
static inline void reiserfs_quota_off_umount(struct super_block *s)
{
}
#endif
static void reiserfs_put_super(struct super_block *s)
{
struct reiserfs_transaction_handle th;
th.t_trans_id = 0;
reiserfs_quota_off_umount(s);
reiserfs: kill-the-BKL This patch is an attempt to remove the Bkl based locking scheme from reiserfs and is intended. It is a bit inspired from an old attempt by Peter Zijlstra: http://lkml.indiana.edu/hypermail/linux/kernel/0704.2/2174.html The bkl is heavily used in this filesystem to prevent from concurrent write accesses on the filesystem. Reiserfs makes a deep use of the specific properties of the Bkl: - It can be acqquired recursively by a same task - It is released on the schedule() calls and reacquired when schedule() returns The two properties above are a roadmap for the reiserfs write locking so it's very hard to simply replace it with a common mutex. - We need a recursive-able locking unless we want to restructure several blocks of the code. - We need to identify the sites where the bkl was implictly relaxed (schedule, wait, sync, etc...) so that we can in turn release and reacquire our new lock explicitly. Such implicit releases of the lock are often required to let other resources producer/consumer do their job or we can suffer unexpected starvations or deadlocks. So the new lock that replaces the bkl here is a per superblock mutex with a specific property: it can be acquired recursively by a same task, like the bkl. For such purpose, we integrate a lock owner and a lock depth field on the superblock information structure. The first axis on this patch is to turn reiserfs_write_(un)lock() function into a wrapper to manage this mutex. Also some explicit calls to lock_kernel() have been converted to reiserfs_write_lock() helpers. The second axis is to find the important blocking sites (schedule...(), wait_on_buffer(), sync_dirty_buffer(), etc...) and then apply an explicit release of the write lock on these locations before blocking. Then we can safely wait for those who can give us resources or those who need some. Typically this is a fight between the current writer, the reiserfs workqueue (aka the async commiter) and the pdflush threads. The third axis is a consequence of the second. The write lock is usually on top of a lock dependency chain which can include the journal lock, the flush lock or the commit lock. So it's dangerous to release and trying to reacquire the write lock while we still hold other locks. This is fine with the bkl: T1 T2 lock_kernel() mutex_lock(A) unlock_kernel() // do something lock_kernel() mutex_lock(A) -> already locked by T1 schedule() (and then unlock_kernel()) lock_kernel() mutex_unlock(A) .... This is not fine with a mutex: T1 T2 mutex_lock(write) mutex_lock(A) mutex_unlock(write) // do something mutex_lock(write) mutex_lock(A) -> already locked by T1 schedule() mutex_lock(write) -> already locked by T2 deadlock The solution in this patch is to provide a helper which releases the write lock and sleep a bit if we can't lock a mutex that depend on it. It's another simulation of the bkl behaviour. The last axis is to locate the fs callbacks that are called with the bkl held, according to Documentation/filesystem/Locking. Those are: - reiserfs_remount - reiserfs_fill_super - reiserfs_put_super Reiserfs didn't need to explicitly lock because of the context of these callbacks. But now we must take care of that with the new locking. After this patch, reiserfs suffers from a slight performance regression (for now). On UP, a high volume write with dd reports an average of 27 MB/s instead of 30 MB/s without the patch applied. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Reviewed-by: Ingo Molnar <mingo@elte.hu> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Bron Gondwana <brong@fastmail.fm> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> LKML-Reference: <1239070789-13354-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-07 10:19:49 +08:00
reiserfs_write_lock(s);
/*
* change file system state to current state if it was mounted
* with read-write permissions
*/
if (!sb_rdonly(s)) {
if (!journal_begin(&th, s, 10)) {
reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s),
1);
set_sb_umount_state(SB_DISK_SUPER_BLOCK(s),
REISERFS_SB(s)->s_mount_state);
journal_mark_dirty(&th, SB_BUFFER_WITH_SB(s));
}
}
/*
* note, journal_release checks for readonly mount, and can
* decide not to do a journal_end
*/
journal_release(&th, s);
reiserfs_free_bitmap_cache(s);
brelse(SB_BUFFER_WITH_SB(s));
print_statistics(s);
if (REISERFS_SB(s)->reserved_blocks != 0) {
reiserfs_warning(s, "green-2005", "reserved blocks left %d",
REISERFS_SB(s)->reserved_blocks);
}
reiserfs: kill-the-BKL This patch is an attempt to remove the Bkl based locking scheme from reiserfs and is intended. It is a bit inspired from an old attempt by Peter Zijlstra: http://lkml.indiana.edu/hypermail/linux/kernel/0704.2/2174.html The bkl is heavily used in this filesystem to prevent from concurrent write accesses on the filesystem. Reiserfs makes a deep use of the specific properties of the Bkl: - It can be acqquired recursively by a same task - It is released on the schedule() calls and reacquired when schedule() returns The two properties above are a roadmap for the reiserfs write locking so it's very hard to simply replace it with a common mutex. - We need a recursive-able locking unless we want to restructure several blocks of the code. - We need to identify the sites where the bkl was implictly relaxed (schedule, wait, sync, etc...) so that we can in turn release and reacquire our new lock explicitly. Such implicit releases of the lock are often required to let other resources producer/consumer do their job or we can suffer unexpected starvations or deadlocks. So the new lock that replaces the bkl here is a per superblock mutex with a specific property: it can be acquired recursively by a same task, like the bkl. For such purpose, we integrate a lock owner and a lock depth field on the superblock information structure. The first axis on this patch is to turn reiserfs_write_(un)lock() function into a wrapper to manage this mutex. Also some explicit calls to lock_kernel() have been converted to reiserfs_write_lock() helpers. The second axis is to find the important blocking sites (schedule...(), wait_on_buffer(), sync_dirty_buffer(), etc...) and then apply an explicit release of the write lock on these locations before blocking. Then we can safely wait for those who can give us resources or those who need some. Typically this is a fight between the current writer, the reiserfs workqueue (aka the async commiter) and the pdflush threads. The third axis is a consequence of the second. The write lock is usually on top of a lock dependency chain which can include the journal lock, the flush lock or the commit lock. So it's dangerous to release and trying to reacquire the write lock while we still hold other locks. This is fine with the bkl: T1 T2 lock_kernel() mutex_lock(A) unlock_kernel() // do something lock_kernel() mutex_lock(A) -> already locked by T1 schedule() (and then unlock_kernel()) lock_kernel() mutex_unlock(A) .... This is not fine with a mutex: T1 T2 mutex_lock(write) mutex_lock(A) mutex_unlock(write) // do something mutex_lock(write) mutex_lock(A) -> already locked by T1 schedule() mutex_lock(write) -> already locked by T2 deadlock The solution in this patch is to provide a helper which releases the write lock and sleep a bit if we can't lock a mutex that depend on it. It's another simulation of the bkl behaviour. The last axis is to locate the fs callbacks that are called with the bkl held, according to Documentation/filesystem/Locking. Those are: - reiserfs_remount - reiserfs_fill_super - reiserfs_put_super Reiserfs didn't need to explicitly lock because of the context of these callbacks. But now we must take care of that with the new locking. After this patch, reiserfs suffers from a slight performance regression (for now). On UP, a high volume write with dd reports an average of 27 MB/s instead of 30 MB/s without the patch applied. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Reviewed-by: Ingo Molnar <mingo@elte.hu> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Bron Gondwana <brong@fastmail.fm> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> LKML-Reference: <1239070789-13354-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-07 10:19:49 +08:00
reiserfs_write_unlock(s);
mutex_destroy(&REISERFS_SB(s)->lock);
destroy_workqueue(REISERFS_SB(s)->commit_wq);
kfree(REISERFS_SB(s)->s_jdev);
kfree(s->s_fs_info);
s->s_fs_info = NULL;
}
static struct kmem_cache *reiserfs_inode_cachep;
static struct inode *reiserfs_alloc_inode(struct super_block *sb)
{
struct reiserfs_inode_info *ei;
ei = kmem_cache_alloc(reiserfs_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
atomic_set(&ei->openers, 0);
mutex_init(&ei->tailpack);
#ifdef CONFIG_QUOTA
memset(&ei->i_dquot, 0, sizeof(ei->i_dquot));
#endif
return &ei->vfs_inode;
}
static void reiserfs_free_inode(struct inode *inode)
{
kmem_cache_free(reiserfs_inode_cachep, REISERFS_I(inode));
}
static void init_once(void *foo)
{
struct reiserfs_inode_info *ei = (struct reiserfs_inode_info *)foo;
INIT_LIST_HEAD(&ei->i_prealloc_list);
inode_init_once(&ei->vfs_inode);
}
static int __init init_inodecache(void)
{
reiserfs_inode_cachep = kmem_cache_create("reiser_inode_cache",
sizeof(struct
reiserfs_inode_info),
0, (SLAB_RECLAIM_ACCOUNT|
2016-01-15 07:18:21 +08:00
SLAB_MEM_SPREAD|
SLAB_ACCOUNT),
init_once);
if (reiserfs_inode_cachep == NULL)
return -ENOMEM;
return 0;
}
static void destroy_inodecache(void)
{
/*
* Make sure all delayed rcu free inodes are flushed before we
* destroy cache.
*/
rcu_barrier();
kmem_cache_destroy(reiserfs_inode_cachep);
}
/* we don't mark inodes dirty, we just log them */
static void reiserfs_dirty_inode(struct inode *inode, int flags)
{
struct reiserfs_transaction_handle th;
int err = 0;
if (sb_rdonly(inode->i_sb)) {
reiserfs_warning(inode->i_sb, "clm-6006",
"writing inode %lu on readonly FS",
inode->i_ino);
return;
}
reiserfs_write_lock(inode->i_sb);
/*
* this is really only used for atime updates, so they don't have
* to be included in O_SYNC or fsync
*/
err = journal_begin(&th, inode->i_sb, 1);
if (err)
goto out;
reiserfs_update_sd(&th, inode);
journal_end(&th);
out:
reiserfs_write_unlock(inode->i_sb);
}
static int reiserfs_show_options(struct seq_file *seq, struct dentry *root)
{
struct super_block *s = root->d_sb;
struct reiserfs_journal *journal = SB_JOURNAL(s);
long opts = REISERFS_SB(s)->s_mount_opt;
if (opts & (1 << REISERFS_LARGETAIL))
seq_puts(seq, ",tails=on");
else if (!(opts & (1 << REISERFS_SMALLTAIL)))
seq_puts(seq, ",notail");
/* tails=small is default so we don't show it */
if (!(opts & (1 << REISERFS_BARRIER_FLUSH)))
seq_puts(seq, ",barrier=none");
/* barrier=flush is default so we don't show it */
if (opts & (1 << REISERFS_ERROR_CONTINUE))
seq_puts(seq, ",errors=continue");
else if (opts & (1 << REISERFS_ERROR_PANIC))
seq_puts(seq, ",errors=panic");
/* errors=ro is default so we don't show it */
if (opts & (1 << REISERFS_DATA_LOG))
seq_puts(seq, ",data=journal");
else if (opts & (1 << REISERFS_DATA_WRITEBACK))
seq_puts(seq, ",data=writeback");
/* data=ordered is default so we don't show it */
if (opts & (1 << REISERFS_ATTRS))
seq_puts(seq, ",attrs");
if (opts & (1 << REISERFS_XATTRS_USER))
seq_puts(seq, ",user_xattr");
if (opts & (1 << REISERFS_EXPOSE_PRIVROOT))
seq_puts(seq, ",expose_privroot");
if (opts & (1 << REISERFS_POSIXACL))
seq_puts(seq, ",acl");
if (REISERFS_SB(s)->s_jdev)
fs: create and use seq_show_option for escaping Many file systems that implement the show_options hook fail to correctly escape their output which could lead to unescaped characters (e.g. new lines) leaking into /proc/mounts and /proc/[pid]/mountinfo files. This could lead to confusion, spoofed entries (resulting in things like systemd issuing false d-bus "mount" notifications), and who knows what else. This looks like it would only be the root user stepping on themselves, but it's possible weird things could happen in containers or in other situations with delegated mount privileges. Here's an example using overlay with setuid fusermount trusting the contents of /proc/mounts (via the /etc/mtab symlink). Imagine the use of "sudo" is something more sneaky: $ BASE="ovl" $ MNT="$BASE/mnt" $ LOW="$BASE/lower" $ UP="$BASE/upper" $ WORK="$BASE/work/ 0 0 none /proc fuse.pwn user_id=1000" $ mkdir -p "$LOW" "$UP" "$WORK" $ sudo mount -t overlay -o "lowerdir=$LOW,upperdir=$UP,workdir=$WORK" none /mnt $ cat /proc/mounts none /root/ovl/mnt overlay rw,relatime,lowerdir=ovl/lower,upperdir=ovl/upper,workdir=ovl/work/ 0 0 none /proc fuse.pwn user_id=1000 0 0 $ fusermount -u /proc $ cat /proc/mounts cat: /proc/mounts: No such file or directory This fixes the problem by adding new seq_show_option and seq_show_option_n helpers, and updating the vulnerable show_option handlers to use them as needed. Some, like SELinux, need to be open coded due to unusual existing escape mechanisms. [akpm@linux-foundation.org: add lost chunk, per Kees] [keescook@chromium.org: seq_show_option should be using const parameters] Signed-off-by: Kees Cook <keescook@chromium.org> Acked-by: Serge Hallyn <serge.hallyn@canonical.com> Acked-by: Jan Kara <jack@suse.com> Acked-by: Paul Moore <paul@paul-moore.com> Cc: J. R. Okajima <hooanon05g@gmail.com> Signed-off-by: Kees Cook <keescook@chromium.org> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-09-05 06:44:57 +08:00
seq_show_option(seq, "jdev", REISERFS_SB(s)->s_jdev);
if (journal->j_max_commit_age != journal->j_default_max_commit_age)
seq_printf(seq, ",commit=%d", journal->j_max_commit_age);
#ifdef CONFIG_QUOTA
if (REISERFS_SB(s)->s_qf_names[USRQUOTA])
fs: create and use seq_show_option for escaping Many file systems that implement the show_options hook fail to correctly escape their output which could lead to unescaped characters (e.g. new lines) leaking into /proc/mounts and /proc/[pid]/mountinfo files. This could lead to confusion, spoofed entries (resulting in things like systemd issuing false d-bus "mount" notifications), and who knows what else. This looks like it would only be the root user stepping on themselves, but it's possible weird things could happen in containers or in other situations with delegated mount privileges. Here's an example using overlay with setuid fusermount trusting the contents of /proc/mounts (via the /etc/mtab symlink). Imagine the use of "sudo" is something more sneaky: $ BASE="ovl" $ MNT="$BASE/mnt" $ LOW="$BASE/lower" $ UP="$BASE/upper" $ WORK="$BASE/work/ 0 0 none /proc fuse.pwn user_id=1000" $ mkdir -p "$LOW" "$UP" "$WORK" $ sudo mount -t overlay -o "lowerdir=$LOW,upperdir=$UP,workdir=$WORK" none /mnt $ cat /proc/mounts none /root/ovl/mnt overlay rw,relatime,lowerdir=ovl/lower,upperdir=ovl/upper,workdir=ovl/work/ 0 0 none /proc fuse.pwn user_id=1000 0 0 $ fusermount -u /proc $ cat /proc/mounts cat: /proc/mounts: No such file or directory This fixes the problem by adding new seq_show_option and seq_show_option_n helpers, and updating the vulnerable show_option handlers to use them as needed. Some, like SELinux, need to be open coded due to unusual existing escape mechanisms. [akpm@linux-foundation.org: add lost chunk, per Kees] [keescook@chromium.org: seq_show_option should be using const parameters] Signed-off-by: Kees Cook <keescook@chromium.org> Acked-by: Serge Hallyn <serge.hallyn@canonical.com> Acked-by: Jan Kara <jack@suse.com> Acked-by: Paul Moore <paul@paul-moore.com> Cc: J. R. Okajima <hooanon05g@gmail.com> Signed-off-by: Kees Cook <keescook@chromium.org> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-09-05 06:44:57 +08:00
seq_show_option(seq, "usrjquota",
REISERFS_SB(s)->s_qf_names[USRQUOTA]);
else if (opts & (1 << REISERFS_USRQUOTA))
seq_puts(seq, ",usrquota");
if (REISERFS_SB(s)->s_qf_names[GRPQUOTA])
fs: create and use seq_show_option for escaping Many file systems that implement the show_options hook fail to correctly escape their output which could lead to unescaped characters (e.g. new lines) leaking into /proc/mounts and /proc/[pid]/mountinfo files. This could lead to confusion, spoofed entries (resulting in things like systemd issuing false d-bus "mount" notifications), and who knows what else. This looks like it would only be the root user stepping on themselves, but it's possible weird things could happen in containers or in other situations with delegated mount privileges. Here's an example using overlay with setuid fusermount trusting the contents of /proc/mounts (via the /etc/mtab symlink). Imagine the use of "sudo" is something more sneaky: $ BASE="ovl" $ MNT="$BASE/mnt" $ LOW="$BASE/lower" $ UP="$BASE/upper" $ WORK="$BASE/work/ 0 0 none /proc fuse.pwn user_id=1000" $ mkdir -p "$LOW" "$UP" "$WORK" $ sudo mount -t overlay -o "lowerdir=$LOW,upperdir=$UP,workdir=$WORK" none /mnt $ cat /proc/mounts none /root/ovl/mnt overlay rw,relatime,lowerdir=ovl/lower,upperdir=ovl/upper,workdir=ovl/work/ 0 0 none /proc fuse.pwn user_id=1000 0 0 $ fusermount -u /proc $ cat /proc/mounts cat: /proc/mounts: No such file or directory This fixes the problem by adding new seq_show_option and seq_show_option_n helpers, and updating the vulnerable show_option handlers to use them as needed. Some, like SELinux, need to be open coded due to unusual existing escape mechanisms. [akpm@linux-foundation.org: add lost chunk, per Kees] [keescook@chromium.org: seq_show_option should be using const parameters] Signed-off-by: Kees Cook <keescook@chromium.org> Acked-by: Serge Hallyn <serge.hallyn@canonical.com> Acked-by: Jan Kara <jack@suse.com> Acked-by: Paul Moore <paul@paul-moore.com> Cc: J. R. Okajima <hooanon05g@gmail.com> Signed-off-by: Kees Cook <keescook@chromium.org> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-09-05 06:44:57 +08:00
seq_show_option(seq, "grpjquota",
REISERFS_SB(s)->s_qf_names[GRPQUOTA]);
else if (opts & (1 << REISERFS_GRPQUOTA))
seq_puts(seq, ",grpquota");
if (REISERFS_SB(s)->s_jquota_fmt) {
if (REISERFS_SB(s)->s_jquota_fmt == QFMT_VFS_OLD)
seq_puts(seq, ",jqfmt=vfsold");
else if (REISERFS_SB(s)->s_jquota_fmt == QFMT_VFS_V0)
seq_puts(seq, ",jqfmt=vfsv0");
}
#endif
/* Block allocator options */
if (opts & (1 << REISERFS_NO_BORDER))
seq_puts(seq, ",block-allocator=noborder");
if (opts & (1 << REISERFS_NO_UNHASHED_RELOCATION))
seq_puts(seq, ",block-allocator=no_unhashed_relocation");
if (opts & (1 << REISERFS_HASHED_RELOCATION))
seq_puts(seq, ",block-allocator=hashed_relocation");
if (opts & (1 << REISERFS_TEST4))
seq_puts(seq, ",block-allocator=test4");
show_alloc_options(seq, s);
return 0;
}
#ifdef CONFIG_QUOTA
static ssize_t reiserfs_quota_write(struct super_block *, int, const char *,
size_t, loff_t);
static ssize_t reiserfs_quota_read(struct super_block *, int, char *, size_t,
loff_t);
static struct dquot **reiserfs_get_dquots(struct inode *inode)
{
return REISERFS_I(inode)->i_dquot;
}
#endif
static const struct super_operations reiserfs_sops = {
.alloc_inode = reiserfs_alloc_inode,
.free_inode = reiserfs_free_inode,
.write_inode = reiserfs_write_inode,
.dirty_inode = reiserfs_dirty_inode,
.evict_inode = reiserfs_evict_inode,
.put_super = reiserfs_put_super,
.sync_fs = reiserfs_sync_fs,
filesystem freeze: add error handling of write_super_lockfs/unlockfs Currently, ext3 in mainline Linux doesn't have the freeze feature which suspends write requests. So, we cannot take a backup which keeps the filesystem's consistency with the storage device's features (snapshot and replication) while it is mounted. In many case, a commercial filesystem (e.g. VxFS) has the freeze feature and it would be used to get the consistent backup. If Linux's standard filesystem ext3 has the freeze feature, we can do it without a commercial filesystem. So I have implemented the ioctls of the freeze feature. I think we can take the consistent backup with the following steps. 1. Freeze the filesystem with the freeze ioctl. 2. Separate the replication volume or create the snapshot with the storage device's feature. 3. Unfreeze the filesystem with the unfreeze ioctl. 4. Take the backup from the separated replication volume or the snapshot. This patch: VFS: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that they can return an error. Rename write_super_lockfs and unlockfs of the super block operation freeze_fs and unfreeze_fs to avoid a confusion. ext3, ext4, xfs, gfs2, jfs: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that write_super_lockfs returns an error if needed, and unlockfs always returns 0. reiserfs: Changed the type of write_super_lockfs and unlockfs from "void" to "int" so that they always return 0 (success) to keep a current behavior. Signed-off-by: Takashi Sato <t-sato@yk.jp.nec.com> Signed-off-by: Masayuki Hamaguchi <m-hamaguchi@ys.jp.nec.com> Cc: <xfs-masters@oss.sgi.com> Cc: <linux-ext4@vger.kernel.org> Cc: Christoph Hellwig <hch@lst.de> Cc: Dave Kleikamp <shaggy@austin.ibm.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Alasdair G Kergon <agk@redhat.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-10 08:40:58 +08:00
.freeze_fs = reiserfs_freeze,
.unfreeze_fs = reiserfs_unfreeze,
.statfs = reiserfs_statfs,
.remount_fs = reiserfs_remount,
.show_options = reiserfs_show_options,
#ifdef CONFIG_QUOTA
.quota_read = reiserfs_quota_read,
.quota_write = reiserfs_quota_write,
.get_dquots = reiserfs_get_dquots,
#endif
};
#ifdef CONFIG_QUOTA
#define QTYPE2NAME(t) ((t)==USRQUOTA?"user":"group")
static int reiserfs_write_dquot(struct dquot *);
static int reiserfs_acquire_dquot(struct dquot *);
static int reiserfs_release_dquot(struct dquot *);
static int reiserfs_mark_dquot_dirty(struct dquot *);
static int reiserfs_write_info(struct super_block *, int);
static int reiserfs_quota_on(struct super_block *, int, int, const struct path *);
static const struct dquot_operations reiserfs_quota_operations = {
.write_dquot = reiserfs_write_dquot,
.acquire_dquot = reiserfs_acquire_dquot,
.release_dquot = reiserfs_release_dquot,
.mark_dirty = reiserfs_mark_dquot_dirty,
.write_info = reiserfs_write_info,
.alloc_dquot = dquot_alloc,
.destroy_dquot = dquot_destroy,
.get_next_id = dquot_get_next_id,
};
static const struct quotactl_ops reiserfs_qctl_operations = {
.quota_on = reiserfs_quota_on,
.quota_off = reiserfs_quota_off,
.quota_sync = dquot_quota_sync,
.get_state = dquot_get_state,
.set_info = dquot_set_dqinfo,
.get_dqblk = dquot_get_dqblk,
.set_dqblk = dquot_set_dqblk,
};
#endif
static const struct export_operations reiserfs_export_ops = {
.encode_fh = reiserfs_encode_fh,
.fh_to_dentry = reiserfs_fh_to_dentry,
.fh_to_parent = reiserfs_fh_to_parent,
.get_parent = reiserfs_get_parent,
};
/*
* this struct is used in reiserfs_getopt () for containing the value for
* those mount options that have values rather than being toggles.
*/
typedef struct {
char *value;
/*
* bitmask which is to set on mount_options bitmask
* when this value is found, 0 is no bits are to be changed.
*/
int setmask;
/*
* bitmask which is to clear on mount_options bitmask
* when this value is found, 0 is no bits are to be changed.
* This is applied BEFORE setmask
*/
int clrmask;
} arg_desc_t;
/* Set this bit in arg_required to allow empty arguments */
#define REISERFS_OPT_ALLOWEMPTY 31
/*
* this struct is used in reiserfs_getopt() for describing the
* set of reiserfs mount options
*/
typedef struct {
char *option_name;
/* 0 if argument is not required, not 0 otherwise */
int arg_required;
/* list of values accepted by an option */
const arg_desc_t *values;
/*
* bitmask which is to set on mount_options bitmask
* when this value is found, 0 is no bits are to be changed.
*/
int setmask;
/*
* bitmask which is to clear on mount_options bitmask
* when this value is found, 0 is no bits are to be changed.
* This is applied BEFORE setmask
*/
int clrmask;
} opt_desc_t;
/* possible values for -o data= */
static const arg_desc_t logging_mode[] = {
{"ordered", 1 << REISERFS_DATA_ORDERED,
(1 << REISERFS_DATA_LOG | 1 << REISERFS_DATA_WRITEBACK)},
{"journal", 1 << REISERFS_DATA_LOG,
(1 << REISERFS_DATA_ORDERED | 1 << REISERFS_DATA_WRITEBACK)},
{"writeback", 1 << REISERFS_DATA_WRITEBACK,
(1 << REISERFS_DATA_ORDERED | 1 << REISERFS_DATA_LOG)},
{.value = NULL}
};
/* possible values for -o barrier= */
static const arg_desc_t barrier_mode[] = {
{"none", 1 << REISERFS_BARRIER_NONE, 1 << REISERFS_BARRIER_FLUSH},
{"flush", 1 << REISERFS_BARRIER_FLUSH, 1 << REISERFS_BARRIER_NONE},
{.value = NULL}
};
/*
* possible values for "-o block-allocator=" and bits which are to be set in
* s_mount_opt of reiserfs specific part of in-core super block
*/
static const arg_desc_t balloc[] = {
{"noborder", 1 << REISERFS_NO_BORDER, 0},
{"border", 0, 1 << REISERFS_NO_BORDER},
{"no_unhashed_relocation", 1 << REISERFS_NO_UNHASHED_RELOCATION, 0},
{"hashed_relocation", 1 << REISERFS_HASHED_RELOCATION, 0},
{"test4", 1 << REISERFS_TEST4, 0},
{"notest4", 0, 1 << REISERFS_TEST4},
{NULL, 0, 0}
};
static const arg_desc_t tails[] = {
{"on", 1 << REISERFS_LARGETAIL, 1 << REISERFS_SMALLTAIL},
{"off", 0, (1 << REISERFS_LARGETAIL) | (1 << REISERFS_SMALLTAIL)},
{"small", 1 << REISERFS_SMALLTAIL, 1 << REISERFS_LARGETAIL},
{NULL, 0, 0}
};
static const arg_desc_t error_actions[] = {
{"panic", 1 << REISERFS_ERROR_PANIC,
(1 << REISERFS_ERROR_RO | 1 << REISERFS_ERROR_CONTINUE)},
{"ro-remount", 1 << REISERFS_ERROR_RO,
(1 << REISERFS_ERROR_PANIC | 1 << REISERFS_ERROR_CONTINUE)},
#ifdef REISERFS_JOURNAL_ERROR_ALLOWS_NO_LOG
{"continue", 1 << REISERFS_ERROR_CONTINUE,
(1 << REISERFS_ERROR_PANIC | 1 << REISERFS_ERROR_RO)},
#endif
{NULL, 0, 0},
};
/*
* proceed only one option from a list *cur - string containing of mount
* options
* opts - array of options which are accepted
* opt_arg - if option is found and requires an argument and if it is specifed
* in the input - pointer to the argument is stored here
* bit_flags - if option requires to set a certain bit - it is set here
* return -1 if unknown option is found, opt->arg_required otherwise
*/
static int reiserfs_getopt(struct super_block *s, char **cur, opt_desc_t * opts,
char **opt_arg, unsigned long *bit_flags)
{
char *p;
/*
* foo=bar,
* ^ ^ ^
* | | +-- option_end
* | +-- arg_start
* +-- option_start
*/
const opt_desc_t *opt;
const arg_desc_t *arg;
p = *cur;
/* assume argument cannot contain commas */
*cur = strchr(p, ',');
if (*cur) {
*(*cur) = '\0';
(*cur)++;
}
if (!strncmp(p, "alloc=", 6)) {
/*
* Ugly special case, probably we should redo options
* parser so that it can understand several arguments for
* some options, also so that it can fill several bitfields
* with option values.
*/
if (reiserfs_parse_alloc_options(s, p + 6)) {
return -1;
} else {
return 0;
}
}
/* for every option in the list */
for (opt = opts; opt->option_name; opt++) {
if (!strncmp(p, opt->option_name, strlen(opt->option_name))) {
if (bit_flags) {
if (opt->clrmask ==
(1 << REISERFS_UNSUPPORTED_OPT))
reiserfs_warning(s, "super-6500",
"%s not supported.\n",
p);
else
*bit_flags &= ~opt->clrmask;
if (opt->setmask ==
(1 << REISERFS_UNSUPPORTED_OPT))
reiserfs_warning(s, "super-6501",
"%s not supported.\n",
p);
else
*bit_flags |= opt->setmask;
}
break;
}
}
if (!opt->option_name) {
reiserfs_warning(s, "super-6502",
"unknown mount option \"%s\"", p);
return -1;
}
p += strlen(opt->option_name);
switch (*p) {
case '=':
if (!opt->arg_required) {
reiserfs_warning(s, "super-6503",
"the option \"%s\" does not "
"require an argument\n",
opt->option_name);
return -1;
}
break;
case 0:
if (opt->arg_required) {
reiserfs_warning(s, "super-6504",
"the option \"%s\" requires an "
"argument\n", opt->option_name);
return -1;
}
break;
default:
reiserfs_warning(s, "super-6505",
"head of option \"%s\" is only correct\n",
opt->option_name);
return -1;
}
/*
* move to the argument, or to next option if argument is not
* required
*/
p++;
if (opt->arg_required
&& !(opt->arg_required & (1 << REISERFS_OPT_ALLOWEMPTY))
&& !strlen(p)) {
/* this catches "option=," if not allowed */
reiserfs_warning(s, "super-6506",
"empty argument for \"%s\"\n",
opt->option_name);
return -1;
}
if (!opt->values) {
/* *=NULLopt_arg contains pointer to argument */
*opt_arg = p;
return opt->arg_required & ~(1 << REISERFS_OPT_ALLOWEMPTY);
}
/* values possible for this option are listed in opt->values */
for (arg = opt->values; arg->value; arg++) {
if (!strcmp(p, arg->value)) {
if (bit_flags) {
*bit_flags &= ~arg->clrmask;
*bit_flags |= arg->setmask;
}
return opt->arg_required;
}
}
reiserfs_warning(s, "super-6506",
"bad value \"%s\" for option \"%s\"\n", p,
opt->option_name);
return -1;
}
/* returns 0 if something is wrong in option string, 1 - otherwise */
static int reiserfs_parse_options(struct super_block *s,
/* string given via mount's -o */
char *options,
/*
* after the parsing phase, contains the
* collection of bitflags defining what
* mount options were selected.
*/
unsigned long *mount_options,
/* strtol-ed from NNN of resize=NNN */
unsigned long *blocks,
char **jdev_name,
unsigned int *commit_max_age,
char **qf_names,
unsigned int *qfmt)
{
int c;
char *arg = NULL;
char *pos;
opt_desc_t opts[] = {
/*
* Compatibility stuff, so that -o notail for old
* setups still work
*/
{"tails",.arg_required = 't',.values = tails},
{"notail",.clrmask =
(1 << REISERFS_LARGETAIL) | (1 << REISERFS_SMALLTAIL)},
{"conv",.setmask = 1 << REISERFS_CONVERT},
{"attrs",.setmask = 1 << REISERFS_ATTRS},
{"noattrs",.clrmask = 1 << REISERFS_ATTRS},
{"expose_privroot", .setmask = 1 << REISERFS_EXPOSE_PRIVROOT},
#ifdef CONFIG_REISERFS_FS_XATTR
{"user_xattr",.setmask = 1 << REISERFS_XATTRS_USER},
{"nouser_xattr",.clrmask = 1 << REISERFS_XATTRS_USER},
#else
{"user_xattr",.setmask = 1 << REISERFS_UNSUPPORTED_OPT},
{"nouser_xattr",.clrmask = 1 << REISERFS_UNSUPPORTED_OPT},
#endif
#ifdef CONFIG_REISERFS_FS_POSIX_ACL
{"acl",.setmask = 1 << REISERFS_POSIXACL},
{"noacl",.clrmask = 1 << REISERFS_POSIXACL},
#else
{"acl",.setmask = 1 << REISERFS_UNSUPPORTED_OPT},
{"noacl",.clrmask = 1 << REISERFS_UNSUPPORTED_OPT},
#endif
{.option_name = "nolog"},
{"replayonly",.setmask = 1 << REPLAYONLY},
{"block-allocator",.arg_required = 'a',.values = balloc},
{"data",.arg_required = 'd',.values = logging_mode},
{"barrier",.arg_required = 'b',.values = barrier_mode},
{"resize",.arg_required = 'r',.values = NULL},
{"jdev",.arg_required = 'j',.values = NULL},
{"nolargeio",.arg_required = 'w',.values = NULL},
{"commit",.arg_required = 'c',.values = NULL},
{"usrquota",.setmask = 1 << REISERFS_USRQUOTA},
{"grpquota",.setmask = 1 << REISERFS_GRPQUOTA},
{"noquota",.clrmask = 1 << REISERFS_USRQUOTA | 1 << REISERFS_GRPQUOTA},
{"errors",.arg_required = 'e',.values = error_actions},
{"usrjquota",.arg_required =
'u' | (1 << REISERFS_OPT_ALLOWEMPTY),.values = NULL},
{"grpjquota",.arg_required =
'g' | (1 << REISERFS_OPT_ALLOWEMPTY),.values = NULL},
{"jqfmt",.arg_required = 'f',.values = NULL},
{.option_name = NULL}
};
*blocks = 0;
if (!options || !*options)
/*
* use default configuration: create tails, journaling on, no
* conversion to newest format
*/
return 1;
for (pos = options; pos;) {
c = reiserfs_getopt(s, &pos, opts, &arg, mount_options);
if (c == -1)
/* wrong option is given */
return 0;
if (c == 'r') {
char *p;
p = NULL;
/* "resize=NNN" or "resize=auto" */
if (!strcmp(arg, "auto")) {
/* From JFS code, to auto-get the size. */
*blocks =
i_size_read(s->s_bdev->bd_inode) >> s->
s_blocksize_bits;
} else {
*blocks = simple_strtoul(arg, &p, 0);
if (*p != '\0') {
/* NNN does not look like a number */
reiserfs_warning(s, "super-6507",
"bad value %s for "
"-oresize\n", arg);
return 0;
}
}
}
if (c == 'c') {
char *p = NULL;
unsigned long val = simple_strtoul(arg, &p, 0);
/* commit=NNN (time in seconds) */
if (*p != '\0' || val >= (unsigned int)-1) {
reiserfs_warning(s, "super-6508",
"bad value %s for -ocommit\n",
arg);
return 0;
}
*commit_max_age = (unsigned int)val;
}
if (c == 'w') {
reiserfs_warning(s, "super-6509", "nolargeio option "
"is no longer supported");
return 0;
}
if (c == 'j') {
if (arg && *arg && jdev_name) {
/* Hm, already assigned? */
if (*jdev_name) {
reiserfs_warning(s, "super-6510",
"journal device was "
"already specified to "
"be %s", *jdev_name);
return 0;
}
*jdev_name = arg;
}
}
#ifdef CONFIG_QUOTA
if (c == 'u' || c == 'g') {
int qtype = c == 'u' ? USRQUOTA : GRPQUOTA;
if (sb_any_quota_loaded(s) &&
(!*arg != !REISERFS_SB(s)->s_qf_names[qtype])) {
reiserfs_warning(s, "super-6511",
"cannot change journaled "
"quota options when quota "
"turned on.");
return 0;
}
if (qf_names[qtype] !=
REISERFS_SB(s)->s_qf_names[qtype])
kfree(qf_names[qtype]);
qf_names[qtype] = NULL;
if (*arg) { /* Some filename specified? */
if (REISERFS_SB(s)->s_qf_names[qtype]
&& strcmp(REISERFS_SB(s)->s_qf_names[qtype],
arg)) {
reiserfs_warning(s, "super-6512",
"%s quota file "
"already specified.",
QTYPE2NAME(qtype));
return 0;
}
if (strchr(arg, '/')) {
reiserfs_warning(s, "super-6513",
"quotafile must be "
"on filesystem root.");
return 0;
}
qf_names[qtype] = kstrdup(arg, GFP_KERNEL);
if (!qf_names[qtype]) {
reiserfs_warning(s, "reiserfs-2502",
"not enough memory "
"for storing "
"quotafile name.");
return 0;
}
if (qtype == USRQUOTA)
*mount_options |= 1 << REISERFS_USRQUOTA;
else
*mount_options |= 1 << REISERFS_GRPQUOTA;
} else {
if (qtype == USRQUOTA)
*mount_options &= ~(1 << REISERFS_USRQUOTA);
else
*mount_options &= ~(1 << REISERFS_GRPQUOTA);
}
}
if (c == 'f') {
if (!strcmp(arg, "vfsold"))
*qfmt = QFMT_VFS_OLD;
else if (!strcmp(arg, "vfsv0"))
*qfmt = QFMT_VFS_V0;
else {
reiserfs_warning(s, "super-6514",
"unknown quota format "
"specified.");
return 0;
}
if (sb_any_quota_loaded(s) &&
*qfmt != REISERFS_SB(s)->s_jquota_fmt) {
reiserfs_warning(s, "super-6515",
"cannot change journaled "
"quota options when quota "
"turned on.");
return 0;
}
}
#else
if (c == 'u' || c == 'g' || c == 'f') {
reiserfs_warning(s, "reiserfs-2503", "journaled "
"quota options not supported.");
return 0;
}
#endif
}
#ifdef CONFIG_QUOTA
if (!REISERFS_SB(s)->s_jquota_fmt && !*qfmt
&& (qf_names[USRQUOTA] || qf_names[GRPQUOTA])) {
reiserfs_warning(s, "super-6515",
"journaled quota format not specified.");
return 0;
}
if ((!(*mount_options & (1 << REISERFS_USRQUOTA)) &&
sb_has_quota_loaded(s, USRQUOTA)) ||
(!(*mount_options & (1 << REISERFS_GRPQUOTA)) &&
sb_has_quota_loaded(s, GRPQUOTA))) {
reiserfs_warning(s, "super-6516", "quota options must "
"be present when quota is turned on.");
return 0;
}
#endif
return 1;
}
static void switch_data_mode(struct super_block *s, unsigned long mode)
{
REISERFS_SB(s)->s_mount_opt &= ~((1 << REISERFS_DATA_LOG) |
(1 << REISERFS_DATA_ORDERED) |
(1 << REISERFS_DATA_WRITEBACK));
REISERFS_SB(s)->s_mount_opt |= (1 << mode);
}
static void handle_data_mode(struct super_block *s, unsigned long mount_options)
{
if (mount_options & (1 << REISERFS_DATA_LOG)) {
if (!reiserfs_data_log(s)) {
switch_data_mode(s, REISERFS_DATA_LOG);
reiserfs_info(s, "switching to journaled data mode\n");
}
} else if (mount_options & (1 << REISERFS_DATA_ORDERED)) {
if (!reiserfs_data_ordered(s)) {
switch_data_mode(s, REISERFS_DATA_ORDERED);
reiserfs_info(s, "switching to ordered data mode\n");
}
} else if (mount_options & (1 << REISERFS_DATA_WRITEBACK)) {
if (!reiserfs_data_writeback(s)) {
switch_data_mode(s, REISERFS_DATA_WRITEBACK);
reiserfs_info(s, "switching to writeback data mode\n");
}
}
}
static void handle_barrier_mode(struct super_block *s, unsigned long bits)
{
int flush = (1 << REISERFS_BARRIER_FLUSH);
int none = (1 << REISERFS_BARRIER_NONE);
int all_barrier = flush | none;
if (bits & all_barrier) {
REISERFS_SB(s)->s_mount_opt &= ~all_barrier;
if (bits & flush) {
REISERFS_SB(s)->s_mount_opt |= flush;
printk("reiserfs: enabling write barrier flush mode\n");
} else if (bits & none) {
REISERFS_SB(s)->s_mount_opt |= none;
printk("reiserfs: write barriers turned off\n");
}
}
}
static void handle_attrs(struct super_block *s)
{
struct reiserfs_super_block *rs = SB_DISK_SUPER_BLOCK(s);
if (reiserfs_attrs(s)) {
if (old_format_only(s)) {
reiserfs_warning(s, "super-6517", "cannot support "
"attributes on 3.5.x disk format");
REISERFS_SB(s)->s_mount_opt &= ~(1 << REISERFS_ATTRS);
return;
}
if (!(le32_to_cpu(rs->s_flags) & reiserfs_attrs_cleared)) {
reiserfs_warning(s, "super-6518", "cannot support "
"attributes until flag is set in "
"super-block");
REISERFS_SB(s)->s_mount_opt &= ~(1 << REISERFS_ATTRS);
}
}
}
#ifdef CONFIG_QUOTA
static void handle_quota_files(struct super_block *s, char **qf_names,
unsigned int *qfmt)
{
int i;
for (i = 0; i < REISERFS_MAXQUOTAS; i++) {
if (qf_names[i] != REISERFS_SB(s)->s_qf_names[i])
kfree(REISERFS_SB(s)->s_qf_names[i]);
REISERFS_SB(s)->s_qf_names[i] = qf_names[i];
}
if (*qfmt)
REISERFS_SB(s)->s_jquota_fmt = *qfmt;
}
#endif
static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg)
{
struct reiserfs_super_block *rs;
struct reiserfs_transaction_handle th;
unsigned long blocks;
unsigned long mount_options = REISERFS_SB(s)->s_mount_opt;
unsigned long safe_mask = 0;
unsigned int commit_max_age = (unsigned int)-1;
struct reiserfs_journal *journal = SB_JOURNAL(s);
int err;
char *qf_names[REISERFS_MAXQUOTAS];
unsigned int qfmt = 0;
#ifdef CONFIG_QUOTA
int i;
reiserfs: kill-the-BKL This patch is an attempt to remove the Bkl based locking scheme from reiserfs and is intended. It is a bit inspired from an old attempt by Peter Zijlstra: http://lkml.indiana.edu/hypermail/linux/kernel/0704.2/2174.html The bkl is heavily used in this filesystem to prevent from concurrent write accesses on the filesystem. Reiserfs makes a deep use of the specific properties of the Bkl: - It can be acqquired recursively by a same task - It is released on the schedule() calls and reacquired when schedule() returns The two properties above are a roadmap for the reiserfs write locking so it's very hard to simply replace it with a common mutex. - We need a recursive-able locking unless we want to restructure several blocks of the code. - We need to identify the sites where the bkl was implictly relaxed (schedule, wait, sync, etc...) so that we can in turn release and reacquire our new lock explicitly. Such implicit releases of the lock are often required to let other resources producer/consumer do their job or we can suffer unexpected starvations or deadlocks. So the new lock that replaces the bkl here is a per superblock mutex with a specific property: it can be acquired recursively by a same task, like the bkl. For such purpose, we integrate a lock owner and a lock depth field on the superblock information structure. The first axis on this patch is to turn reiserfs_write_(un)lock() function into a wrapper to manage this mutex. Also some explicit calls to lock_kernel() have been converted to reiserfs_write_lock() helpers. The second axis is to find the important blocking sites (schedule...(), wait_on_buffer(), sync_dirty_buffer(), etc...) and then apply an explicit release of the write lock on these locations before blocking. Then we can safely wait for those who can give us resources or those who need some. Typically this is a fight between the current writer, the reiserfs workqueue (aka the async commiter) and the pdflush threads. The third axis is a consequence of the second. The write lock is usually on top of a lock dependency chain which can include the journal lock, the flush lock or the commit lock. So it's dangerous to release and trying to reacquire the write lock while we still hold other locks. This is fine with the bkl: T1 T2 lock_kernel() mutex_lock(A) unlock_kernel() // do something lock_kernel() mutex_lock(A) -> already locked by T1 schedule() (and then unlock_kernel()) lock_kernel() mutex_unlock(A) .... This is not fine with a mutex: T1 T2 mutex_lock(write) mutex_lock(A) mutex_unlock(write) // do something mutex_lock(write) mutex_lock(A) -> already locked by T1 schedule() mutex_lock(write) -> already locked by T2 deadlock The solution in this patch is to provide a helper which releases the write lock and sleep a bit if we can't lock a mutex that depend on it. It's another simulation of the bkl behaviour. The last axis is to locate the fs callbacks that are called with the bkl held, according to Documentation/filesystem/Locking. Those are: - reiserfs_remount - reiserfs_fill_super - reiserfs_put_super Reiserfs didn't need to explicitly lock because of the context of these callbacks. But now we must take care of that with the new locking. After this patch, reiserfs suffers from a slight performance regression (for now). On UP, a high volume write with dd reports an average of 27 MB/s instead of 30 MB/s without the patch applied. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Reviewed-by: Ingo Molnar <mingo@elte.hu> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Bron Gondwana <brong@fastmail.fm> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> LKML-Reference: <1239070789-13354-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-07 10:19:49 +08:00
#endif
fs: push sync_filesystem() down to the file system's remount_fs() Previously, the no-op "mount -o mount /dev/xxx" operation when the file system is already mounted read-write causes an implied, unconditional syncfs(). This seems pretty stupid, and it's certainly documented or guaraunteed to do this, nor is it particularly useful, except in the case where the file system was mounted rw and is getting remounted read-only. However, it's possible that there might be some file systems that are actually depending on this behavior. In most file systems, it's probably fine to only call sync_filesystem() when transitioning from read-write to read-only, and there are some file systems where this is not needed at all (for example, for a pseudo-filesystem or something like romfs). Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: linux-fsdevel@vger.kernel.org Cc: Christoph Hellwig <hch@infradead.org> Cc: Artem Bityutskiy <dedekind1@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Evgeniy Dushistov <dushistov@mail.ru> Cc: Jan Kara <jack@suse.cz> Cc: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Cc: Anders Larsen <al@alarsen.net> Cc: Phillip Lougher <phillip@squashfs.org.uk> Cc: Kees Cook <keescook@chromium.org> Cc: Mikulas Patocka <mikulas@artax.karlin.mff.cuni.cz> Cc: Petr Vandrovec <petr@vandrovec.name> Cc: xfs@oss.sgi.com Cc: linux-btrfs@vger.kernel.org Cc: linux-cifs@vger.kernel.org Cc: samba-technical@lists.samba.org Cc: codalist@coda.cs.cmu.edu Cc: linux-ext4@vger.kernel.org Cc: linux-f2fs-devel@lists.sourceforge.net Cc: fuse-devel@lists.sourceforge.net Cc: cluster-devel@redhat.com Cc: linux-mtd@lists.infradead.org Cc: jfs-discussion@lists.sourceforge.net Cc: linux-nfs@vger.kernel.org Cc: linux-nilfs@vger.kernel.org Cc: linux-ntfs-dev@lists.sourceforge.net Cc: ocfs2-devel@oss.oracle.com Cc: reiserfs-devel@vger.kernel.org
2014-03-13 22:14:33 +08:00
sync_filesystem(s);
reiserfs: kill-the-BKL This patch is an attempt to remove the Bkl based locking scheme from reiserfs and is intended. It is a bit inspired from an old attempt by Peter Zijlstra: http://lkml.indiana.edu/hypermail/linux/kernel/0704.2/2174.html The bkl is heavily used in this filesystem to prevent from concurrent write accesses on the filesystem. Reiserfs makes a deep use of the specific properties of the Bkl: - It can be acqquired recursively by a same task - It is released on the schedule() calls and reacquired when schedule() returns The two properties above are a roadmap for the reiserfs write locking so it's very hard to simply replace it with a common mutex. - We need a recursive-able locking unless we want to restructure several blocks of the code. - We need to identify the sites where the bkl was implictly relaxed (schedule, wait, sync, etc...) so that we can in turn release and reacquire our new lock explicitly. Such implicit releases of the lock are often required to let other resources producer/consumer do their job or we can suffer unexpected starvations or deadlocks. So the new lock that replaces the bkl here is a per superblock mutex with a specific property: it can be acquired recursively by a same task, like the bkl. For such purpose, we integrate a lock owner and a lock depth field on the superblock information structure. The first axis on this patch is to turn reiserfs_write_(un)lock() function into a wrapper to manage this mutex. Also some explicit calls to lock_kernel() have been converted to reiserfs_write_lock() helpers. The second axis is to find the important blocking sites (schedule...(), wait_on_buffer(), sync_dirty_buffer(), etc...) and then apply an explicit release of the write lock on these locations before blocking. Then we can safely wait for those who can give us resources or those who need some. Typically this is a fight between the current writer, the reiserfs workqueue (aka the async commiter) and the pdflush threads. The third axis is a consequence of the second. The write lock is usually on top of a lock dependency chain which can include the journal lock, the flush lock or the commit lock. So it's dangerous to release and trying to reacquire the write lock while we still hold other locks. This is fine with the bkl: T1 T2 lock_kernel() mutex_lock(A) unlock_kernel() // do something lock_kernel() mutex_lock(A) -> already locked by T1 schedule() (and then unlock_kernel()) lock_kernel() mutex_unlock(A) .... This is not fine with a mutex: T1 T2 mutex_lock(write) mutex_lock(A) mutex_unlock(write) // do something mutex_lock(write) mutex_lock(A) -> already locked by T1 schedule() mutex_lock(write) -> already locked by T2 deadlock The solution in this patch is to provide a helper which releases the write lock and sleep a bit if we can't lock a mutex that depend on it. It's another simulation of the bkl behaviour. The last axis is to locate the fs callbacks that are called with the bkl held, according to Documentation/filesystem/Locking. Those are: - reiserfs_remount - reiserfs_fill_super - reiserfs_put_super Reiserfs didn't need to explicitly lock because of the context of these callbacks. But now we must take care of that with the new locking. After this patch, reiserfs suffers from a slight performance regression (for now). On UP, a high volume write with dd reports an average of 27 MB/s instead of 30 MB/s without the patch applied. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Reviewed-by: Ingo Molnar <mingo@elte.hu> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Bron Gondwana <brong@fastmail.fm> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> LKML-Reference: <1239070789-13354-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-07 10:19:49 +08:00
reiserfs_write_lock(s);
reiserfs: kill-the-BKL This patch is an attempt to remove the Bkl based locking scheme from reiserfs and is intended. It is a bit inspired from an old attempt by Peter Zijlstra: http://lkml.indiana.edu/hypermail/linux/kernel/0704.2/2174.html The bkl is heavily used in this filesystem to prevent from concurrent write accesses on the filesystem. Reiserfs makes a deep use of the specific properties of the Bkl: - It can be acqquired recursively by a same task - It is released on the schedule() calls and reacquired when schedule() returns The two properties above are a roadmap for the reiserfs write locking so it's very hard to simply replace it with a common mutex. - We need a recursive-able locking unless we want to restructure several blocks of the code. - We need to identify the sites where the bkl was implictly relaxed (schedule, wait, sync, etc...) so that we can in turn release and reacquire our new lock explicitly. Such implicit releases of the lock are often required to let other resources producer/consumer do their job or we can suffer unexpected starvations or deadlocks. So the new lock that replaces the bkl here is a per superblock mutex with a specific property: it can be acquired recursively by a same task, like the bkl. For such purpose, we integrate a lock owner and a lock depth field on the superblock information structure. The first axis on this patch is to turn reiserfs_write_(un)lock() function into a wrapper to manage this mutex. Also some explicit calls to lock_kernel() have been converted to reiserfs_write_lock() helpers. The second axis is to find the important blocking sites (schedule...(), wait_on_buffer(), sync_dirty_buffer(), etc...) and then apply an explicit release of the write lock on these locations before blocking. Then we can safely wait for those who can give us resources or those who need some. Typically this is a fight between the current writer, the reiserfs workqueue (aka the async commiter) and the pdflush threads. The third axis is a consequence of the second. The write lock is usually on top of a lock dependency chain which can include the journal lock, the flush lock or the commit lock. So it's dangerous to release and trying to reacquire the write lock while we still hold other locks. This is fine with the bkl: T1 T2 lock_kernel() mutex_lock(A) unlock_kernel() // do something lock_kernel() mutex_lock(A) -> already locked by T1 schedule() (and then unlock_kernel()) lock_kernel() mutex_unlock(A) .... This is not fine with a mutex: T1 T2 mutex_lock(write) mutex_lock(A) mutex_unlock(write) // do something mutex_lock(write) mutex_lock(A) -> already locked by T1 schedule() mutex_lock(write) -> already locked by T2 deadlock The solution in this patch is to provide a helper which releases the write lock and sleep a bit if we can't lock a mutex that depend on it. It's another simulation of the bkl behaviour. The last axis is to locate the fs callbacks that are called with the bkl held, according to Documentation/filesystem/Locking. Those are: - reiserfs_remount - reiserfs_fill_super - reiserfs_put_super Reiserfs didn't need to explicitly lock because of the context of these callbacks. But now we must take care of that with the new locking. After this patch, reiserfs suffers from a slight performance regression (for now). On UP, a high volume write with dd reports an average of 27 MB/s instead of 30 MB/s without the patch applied. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Reviewed-by: Ingo Molnar <mingo@elte.hu> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Bron Gondwana <brong@fastmail.fm> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> LKML-Reference: <1239070789-13354-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-07 10:19:49 +08:00
#ifdef CONFIG_QUOTA
memcpy(qf_names, REISERFS_SB(s)->s_qf_names, sizeof(qf_names));
#endif
rs = SB_DISK_SUPER_BLOCK(s);
if (!reiserfs_parse_options
(s, arg, &mount_options, &blocks, NULL, &commit_max_age,
qf_names, &qfmt)) {
#ifdef CONFIG_QUOTA
for (i = 0; i < REISERFS_MAXQUOTAS; i++)
if (qf_names[i] != REISERFS_SB(s)->s_qf_names[i])
kfree(qf_names[i]);
#endif
err = -EINVAL;
goto out_err_unlock;
}
#ifdef CONFIG_QUOTA
handle_quota_files(s, qf_names, &qfmt);
#endif
handle_attrs(s);
/* Add options that are safe here */
safe_mask |= 1 << REISERFS_SMALLTAIL;
safe_mask |= 1 << REISERFS_LARGETAIL;
safe_mask |= 1 << REISERFS_NO_BORDER;
safe_mask |= 1 << REISERFS_NO_UNHASHED_RELOCATION;
safe_mask |= 1 << REISERFS_HASHED_RELOCATION;
safe_mask |= 1 << REISERFS_TEST4;
safe_mask |= 1 << REISERFS_ATTRS;
safe_mask |= 1 << REISERFS_XATTRS_USER;
safe_mask |= 1 << REISERFS_POSIXACL;
safe_mask |= 1 << REISERFS_BARRIER_FLUSH;
safe_mask |= 1 << REISERFS_BARRIER_NONE;
safe_mask |= 1 << REISERFS_ERROR_RO;
safe_mask |= 1 << REISERFS_ERROR_CONTINUE;
safe_mask |= 1 << REISERFS_ERROR_PANIC;
safe_mask |= 1 << REISERFS_USRQUOTA;
safe_mask |= 1 << REISERFS_GRPQUOTA;
/*
* Update the bitmask, taking care to keep
* the bits we're not allowed to change here
*/
REISERFS_SB(s)->s_mount_opt =
(REISERFS_SB(s)->
s_mount_opt & ~safe_mask) | (mount_options & safe_mask);
if (commit_max_age != 0 && commit_max_age != (unsigned int)-1) {
journal->j_max_commit_age = commit_max_age;
journal->j_max_trans_age = commit_max_age;
} else if (commit_max_age == 0) {
/* 0 means restore defaults. */
journal->j_max_commit_age = journal->j_default_max_commit_age;
journal->j_max_trans_age = JOURNAL_MAX_TRANS_AGE;
}
if (blocks) {
err = reiserfs_resize(s, blocks);
if (err != 0)
goto out_err_unlock;
}
Rename superblock flags (MS_xyz -> SB_xyz) This is a pure automated search-and-replace of the internal kernel superblock flags. The s_flags are now called SB_*, with the names and the values for the moment mirroring the MS_* flags that they're equivalent to. Note how the MS_xyz flags are the ones passed to the mount system call, while the SB_xyz flags are what we then use in sb->s_flags. The script to do this was: # places to look in; re security/*: it generally should *not* be # touched (that stuff parses mount(2) arguments directly), but # there are two places where we really deal with superblock flags. FILES="drivers/mtd drivers/staging/lustre fs ipc mm \ include/linux/fs.h include/uapi/linux/bfs_fs.h \ security/apparmor/apparmorfs.c security/apparmor/include/lib.h" # the list of MS_... constants SYMS="RDONLY NOSUID NODEV NOEXEC SYNCHRONOUS REMOUNT MANDLOCK \ DIRSYNC NOATIME NODIRATIME BIND MOVE REC VERBOSE SILENT \ POSIXACL UNBINDABLE PRIVATE SLAVE SHARED RELATIME KERNMOUNT \ I_VERSION STRICTATIME LAZYTIME SUBMOUNT NOREMOTELOCK NOSEC BORN \ ACTIVE NOUSER" SED_PROG= for i in $SYMS; do SED_PROG="$SED_PROG -e s/MS_$i/SB_$i/g"; done # we want files that contain at least one of MS_..., # with fs/namespace.c and fs/pnode.c excluded. L=$(for i in $SYMS; do git grep -w -l MS_$i $FILES; done| sort|uniq|grep -v '^fs/namespace.c'|grep -v '^fs/pnode.c') for f in $L; do sed -i $f $SED_PROG; done Requested-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-28 05:05:09 +08:00
if (*mount_flags & SB_RDONLY) {
reiserfs_write_unlock(s);
reiserfs_xattr_init(s, *mount_flags);
/* remount read-only */
if (sb_rdonly(s))
/* it is read-only already */
goto out_ok_unlocked;
err = dquot_suspend(s, -1);
if (err < 0)
goto out_err;
/* try to remount file system with read-only permissions */
if (sb_umount_state(rs) == REISERFS_VALID_FS
|| REISERFS_SB(s)->s_mount_state != REISERFS_VALID_FS) {
goto out_ok_unlocked;
}
reiserfs_write_lock(s);
err = journal_begin(&th, s, 10);
if (err)
goto out_err_unlock;
/* Mounting a rw partition read-only. */
reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s), 1);
set_sb_umount_state(rs, REISERFS_SB(s)->s_mount_state);
journal_mark_dirty(&th, SB_BUFFER_WITH_SB(s));
} else {
/* remount read-write */
if (!sb_rdonly(s)) {
reiserfs_write_unlock(s);
reiserfs_xattr_init(s, *mount_flags);
goto out_ok_unlocked; /* We are read-write already */
}
if (reiserfs_is_journal_aborted(journal)) {
err = journal->j_errno;
goto out_err_unlock;
}
handle_data_mode(s, mount_options);
handle_barrier_mode(s, mount_options);
REISERFS_SB(s)->s_mount_state = sb_umount_state(rs);
/* now it is safe to call journal_begin */
Rename superblock flags (MS_xyz -> SB_xyz) This is a pure automated search-and-replace of the internal kernel superblock flags. The s_flags are now called SB_*, with the names and the values for the moment mirroring the MS_* flags that they're equivalent to. Note how the MS_xyz flags are the ones passed to the mount system call, while the SB_xyz flags are what we then use in sb->s_flags. The script to do this was: # places to look in; re security/*: it generally should *not* be # touched (that stuff parses mount(2) arguments directly), but # there are two places where we really deal with superblock flags. FILES="drivers/mtd drivers/staging/lustre fs ipc mm \ include/linux/fs.h include/uapi/linux/bfs_fs.h \ security/apparmor/apparmorfs.c security/apparmor/include/lib.h" # the list of MS_... constants SYMS="RDONLY NOSUID NODEV NOEXEC SYNCHRONOUS REMOUNT MANDLOCK \ DIRSYNC NOATIME NODIRATIME BIND MOVE REC VERBOSE SILENT \ POSIXACL UNBINDABLE PRIVATE SLAVE SHARED RELATIME KERNMOUNT \ I_VERSION STRICTATIME LAZYTIME SUBMOUNT NOREMOTELOCK NOSEC BORN \ ACTIVE NOUSER" SED_PROG= for i in $SYMS; do SED_PROG="$SED_PROG -e s/MS_$i/SB_$i/g"; done # we want files that contain at least one of MS_..., # with fs/namespace.c and fs/pnode.c excluded. L=$(for i in $SYMS; do git grep -w -l MS_$i $FILES; done| sort|uniq|grep -v '^fs/namespace.c'|grep -v '^fs/pnode.c') for f in $L; do sed -i $f $SED_PROG; done Requested-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-28 05:05:09 +08:00
s->s_flags &= ~SB_RDONLY;
err = journal_begin(&th, s, 10);
if (err)
goto out_err_unlock;
/* Mount a partition which is read-only, read-write */
reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s), 1);
REISERFS_SB(s)->s_mount_state = sb_umount_state(rs);
Rename superblock flags (MS_xyz -> SB_xyz) This is a pure automated search-and-replace of the internal kernel superblock flags. The s_flags are now called SB_*, with the names and the values for the moment mirroring the MS_* flags that they're equivalent to. Note how the MS_xyz flags are the ones passed to the mount system call, while the SB_xyz flags are what we then use in sb->s_flags. The script to do this was: # places to look in; re security/*: it generally should *not* be # touched (that stuff parses mount(2) arguments directly), but # there are two places where we really deal with superblock flags. FILES="drivers/mtd drivers/staging/lustre fs ipc mm \ include/linux/fs.h include/uapi/linux/bfs_fs.h \ security/apparmor/apparmorfs.c security/apparmor/include/lib.h" # the list of MS_... constants SYMS="RDONLY NOSUID NODEV NOEXEC SYNCHRONOUS REMOUNT MANDLOCK \ DIRSYNC NOATIME NODIRATIME BIND MOVE REC VERBOSE SILENT \ POSIXACL UNBINDABLE PRIVATE SLAVE SHARED RELATIME KERNMOUNT \ I_VERSION STRICTATIME LAZYTIME SUBMOUNT NOREMOTELOCK NOSEC BORN \ ACTIVE NOUSER" SED_PROG= for i in $SYMS; do SED_PROG="$SED_PROG -e s/MS_$i/SB_$i/g"; done # we want files that contain at least one of MS_..., # with fs/namespace.c and fs/pnode.c excluded. L=$(for i in $SYMS; do git grep -w -l MS_$i $FILES; done| sort|uniq|grep -v '^fs/namespace.c'|grep -v '^fs/pnode.c') for f in $L; do sed -i $f $SED_PROG; done Requested-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-28 05:05:09 +08:00
s->s_flags &= ~SB_RDONLY;
set_sb_umount_state(rs, REISERFS_ERROR_FS);
if (!old_format_only(s))
set_sb_mnt_count(rs, sb_mnt_count(rs) + 1);
/* mark_buffer_dirty (SB_BUFFER_WITH_SB (s), 1); */
journal_mark_dirty(&th, SB_BUFFER_WITH_SB(s));
REISERFS_SB(s)->s_mount_state = REISERFS_VALID_FS;
}
/* this will force a full flush of all journal lists */
SB_JOURNAL(s)->j_must_wait = 1;
err = journal_end(&th);
if (err)
goto out_err_unlock;
reiserfs_write_unlock(s);
Rename superblock flags (MS_xyz -> SB_xyz) This is a pure automated search-and-replace of the internal kernel superblock flags. The s_flags are now called SB_*, with the names and the values for the moment mirroring the MS_* flags that they're equivalent to. Note how the MS_xyz flags are the ones passed to the mount system call, while the SB_xyz flags are what we then use in sb->s_flags. The script to do this was: # places to look in; re security/*: it generally should *not* be # touched (that stuff parses mount(2) arguments directly), but # there are two places where we really deal with superblock flags. FILES="drivers/mtd drivers/staging/lustre fs ipc mm \ include/linux/fs.h include/uapi/linux/bfs_fs.h \ security/apparmor/apparmorfs.c security/apparmor/include/lib.h" # the list of MS_... constants SYMS="RDONLY NOSUID NODEV NOEXEC SYNCHRONOUS REMOUNT MANDLOCK \ DIRSYNC NOATIME NODIRATIME BIND MOVE REC VERBOSE SILENT \ POSIXACL UNBINDABLE PRIVATE SLAVE SHARED RELATIME KERNMOUNT \ I_VERSION STRICTATIME LAZYTIME SUBMOUNT NOREMOTELOCK NOSEC BORN \ ACTIVE NOUSER" SED_PROG= for i in $SYMS; do SED_PROG="$SED_PROG -e s/MS_$i/SB_$i/g"; done # we want files that contain at least one of MS_..., # with fs/namespace.c and fs/pnode.c excluded. L=$(for i in $SYMS; do git grep -w -l MS_$i $FILES; done| sort|uniq|grep -v '^fs/namespace.c'|grep -v '^fs/pnode.c') for f in $L; do sed -i $f $SED_PROG; done Requested-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-28 05:05:09 +08:00
if (!(*mount_flags & SB_RDONLY)) {
dquot_resume(s, -1);
reiserfs_write_lock(s);
finish_unfinished(s);
reiserfs_write_unlock(s);
reiserfs_xattr_init(s, *mount_flags);
}
out_ok_unlocked:
return 0;
out_err_unlock:
reiserfs_write_unlock(s);
out_err:
return err;
}
static int read_super_block(struct super_block *s, int offset)
{
struct buffer_head *bh;
struct reiserfs_super_block *rs;
int fs_blocksize;
bh = sb_bread(s, offset / s->s_blocksize);
if (!bh) {
reiserfs_warning(s, "sh-2006",
"bread failed (dev %s, block %lu, size %lu)",
s->s_id, offset / s->s_blocksize,
s->s_blocksize);
return 1;
}
rs = (struct reiserfs_super_block *)bh->b_data;
if (!is_any_reiserfs_magic_string(rs)) {
brelse(bh);
return 1;
}
/*
* ok, reiserfs signature (old or new) found in at the given offset
*/
fs_blocksize = sb_blocksize(rs);
brelse(bh);
sb_set_blocksize(s, fs_blocksize);
bh = sb_bread(s, offset / s->s_blocksize);
if (!bh) {
reiserfs_warning(s, "sh-2007",
"bread failed (dev %s, block %lu, size %lu)",
s->s_id, offset / s->s_blocksize,
s->s_blocksize);
return 1;
}
rs = (struct reiserfs_super_block *)bh->b_data;
if (sb_blocksize(rs) != s->s_blocksize) {
reiserfs_warning(s, "sh-2011", "can't find a reiserfs "
"filesystem on (dev %s, block %llu, size %lu)",
s->s_id,
(unsigned long long)bh->b_blocknr,
s->s_blocksize);
brelse(bh);
return 1;
}
if (rs->s_v1.s_root_block == cpu_to_le32(-1)) {
brelse(bh);
reiserfs_warning(s, "super-6519", "Unfinished reiserfsck "
"--rebuild-tree run detected. Please run\n"
"reiserfsck --rebuild-tree and wait for a "
"completion. If that fails\n"
"get newer reiserfsprogs package");
return 1;
}
SB_BUFFER_WITH_SB(s) = bh;
SB_DISK_SUPER_BLOCK(s) = rs;
/*
* magic is of non-standard journal filesystem, look at s_version to
* find which format is in use
*/
if (is_reiserfs_jr(rs)) {
if (sb_version(rs) == REISERFS_VERSION_2)
reiserfs_info(s, "found reiserfs format \"3.6\""
" with non-standard journal\n");
else if (sb_version(rs) == REISERFS_VERSION_1)
reiserfs_info(s, "found reiserfs format \"3.5\""
" with non-standard journal\n");
else {
reiserfs_warning(s, "sh-2012", "found unknown "
"format \"%u\" of reiserfs with "
"non-standard magic", sb_version(rs));
return 1;
}
} else
/*
* s_version of standard format may contain incorrect
* information, so we just look at the magic string
*/
reiserfs_info(s,
"found reiserfs format \"%s\" with standard journal\n",
is_reiserfs_3_5(rs) ? "3.5" : "3.6");
s->s_op = &reiserfs_sops;
s->s_export_op = &reiserfs_export_ops;
#ifdef CONFIG_QUOTA
s->s_qcop = &reiserfs_qctl_operations;
s->dq_op = &reiserfs_quota_operations;
s->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP;
#endif
/*
* new format is limited by the 32 bit wide i_blocks field, want to
* be one full block below that.
*/
s->s_maxbytes = (512LL << 32) - s->s_blocksize;
return 0;
}
/* after journal replay, reread all bitmap and super blocks */
static int reread_meta_blocks(struct super_block *s)
{
ll_rw_block(REQ_OP_READ, 0, 1, &SB_BUFFER_WITH_SB(s));
wait_on_buffer(SB_BUFFER_WITH_SB(s));
if (!buffer_uptodate(SB_BUFFER_WITH_SB(s))) {
reiserfs_warning(s, "reiserfs-2504", "error reading the super");
return 1;
}
return 0;
}
/* hash detection stuff */
/*
* if root directory is empty - we set default - Yura's - hash and
* warn about it
* FIXME: we look for only one name in a directory. If tea and yura
* both have the same value - we ask user to send report to the
* mailing list
*/
static __u32 find_hash_out(struct super_block *s)
{
int retval;
struct inode *inode;
struct cpu_key key;
INITIALIZE_PATH(path);
struct reiserfs_dir_entry de;
struct reiserfs_de_head *deh;
__u32 hash = DEFAULT_HASH;
__u32 deh_hashval, teahash, r5hash, yurahash;
inode = d_inode(s->s_root);
make_cpu_key(&key, inode, ~0, TYPE_DIRENTRY, 3);
retval = search_by_entry_key(s, &key, &path, &de);
if (retval == IO_ERROR) {
pathrelse(&path);
return UNSET_HASH;
}
if (retval == NAME_NOT_FOUND)
de.de_entry_num--;
set_de_name_and_namelen(&de);
deh = de.de_deh + de.de_entry_num;
if (deh_offset(deh) == DOT_DOT_OFFSET) {
/* allow override in this case */
if (reiserfs_rupasov_hash(s))
hash = YURA_HASH;
reiserfs_info(s, "FS seems to be empty, autodetect is using the default hash\n");
goto out;
}
deh_hashval = GET_HASH_VALUE(deh_offset(deh));
r5hash = GET_HASH_VALUE(r5_hash(de.de_name, de.de_namelen));
teahash = GET_HASH_VALUE(keyed_hash(de.de_name, de.de_namelen));
yurahash = GET_HASH_VALUE(yura_hash(de.de_name, de.de_namelen));
if ((teahash == r5hash && deh_hashval == r5hash) ||
(teahash == yurahash && deh_hashval == yurahash) ||
(r5hash == yurahash && deh_hashval == yurahash)) {
reiserfs_warning(s, "reiserfs-2506",
"Unable to automatically detect hash "
"function. Please mount with -o "
"hash={tea,rupasov,r5}");
hash = UNSET_HASH;
goto out;
}
if (deh_hashval == yurahash)
hash = YURA_HASH;
else if (deh_hashval == teahash)
hash = TEA_HASH;
else if (deh_hashval == r5hash)
hash = R5_HASH;
else {
reiserfs_warning(s, "reiserfs-2506",
"Unrecognised hash function");
hash = UNSET_HASH;
}
out:
pathrelse(&path);
return hash;
}
/* finds out which hash names are sorted with */
static int what_hash(struct super_block *s)
{
__u32 code;
code = sb_hash_function_code(SB_DISK_SUPER_BLOCK(s));
/*
* reiserfs_hash_detect() == true if any of the hash mount options
* were used. We must check them to make sure the user isn't
* using a bad hash value
*/
if (code == UNSET_HASH || reiserfs_hash_detect(s))
code = find_hash_out(s);
if (code != UNSET_HASH && reiserfs_hash_detect(s)) {
/*
* detection has found the hash, and we must check against the
* mount options
*/
if (reiserfs_rupasov_hash(s) && code != YURA_HASH) {
reiserfs_warning(s, "reiserfs-2507",
"Error, %s hash detected, "
"unable to force rupasov hash",
reiserfs_hashname(code));
code = UNSET_HASH;
} else if (reiserfs_tea_hash(s) && code != TEA_HASH) {
reiserfs_warning(s, "reiserfs-2508",
"Error, %s hash detected, "
"unable to force tea hash",
reiserfs_hashname(code));
code = UNSET_HASH;
} else if (reiserfs_r5_hash(s) && code != R5_HASH) {
reiserfs_warning(s, "reiserfs-2509",
"Error, %s hash detected, "
"unable to force r5 hash",
reiserfs_hashname(code));
code = UNSET_HASH;
}
} else {
/*
* find_hash_out was not called or
* could not determine the hash
*/
if (reiserfs_rupasov_hash(s)) {
code = YURA_HASH;
} else if (reiserfs_tea_hash(s)) {
code = TEA_HASH;
} else if (reiserfs_r5_hash(s)) {
code = R5_HASH;
}
}
/*
* if we are mounted RW, and we have a new valid hash code, update
* the super
*/
if (code != UNSET_HASH &&
!sb_rdonly(s) &&
code != sb_hash_function_code(SB_DISK_SUPER_BLOCK(s))) {
set_sb_hash_function_code(SB_DISK_SUPER_BLOCK(s), code);
}
return code;
}
/* return pointer to appropriate function */
static hashf_t hash_function(struct super_block *s)
{
switch (what_hash(s)) {
case TEA_HASH:
reiserfs_info(s, "Using tea hash to sort names\n");
return keyed_hash;
case YURA_HASH:
reiserfs_info(s, "Using rupasov hash to sort names\n");
return yura_hash;
case R5_HASH:
reiserfs_info(s, "Using r5 hash to sort names\n");
return r5_hash;
}
return NULL;
}
/* this is used to set up correct value for old partitions */
static int function2code(hashf_t func)
{
if (func == keyed_hash)
return TEA_HASH;
if (func == yura_hash)
return YURA_HASH;
if (func == r5_hash)
return R5_HASH;
BUG(); /* should never happen */
return 0;
}
#define SWARN(silent, s, id, ...) \
if (!(silent)) \
reiserfs_warning(s, id, __VA_ARGS__)
static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
{
struct inode *root_inode;
struct reiserfs_transaction_handle th;
int old_format = 0;
unsigned long blocks;
unsigned int commit_max_age = 0;
int jinit_done = 0;
struct reiserfs_iget_args args;
struct reiserfs_super_block *rs;
char *jdev_name;
struct reiserfs_sb_info *sbi;
int errval = -EINVAL;
char *qf_names[REISERFS_MAXQUOTAS] = {};
unsigned int qfmt = 0;
sbi = kzalloc(sizeof(struct reiserfs_sb_info), GFP_KERNEL);
if (!sbi)
return -ENOMEM;
s->s_fs_info = sbi;
/* Set default values for options: non-aggressive tails, RO on errors */
sbi->s_mount_opt |= (1 << REISERFS_SMALLTAIL);
sbi->s_mount_opt |= (1 << REISERFS_ERROR_RO);
sbi->s_mount_opt |= (1 << REISERFS_BARRIER_FLUSH);
/* no preallocation minimum, be smart in reiserfs_file_write instead */
sbi->s_alloc_options.preallocmin = 0;
/* Preallocate by 16 blocks (17-1) at once */
sbi->s_alloc_options.preallocsize = 17;
/* setup default block allocator options */
reiserfs_init_alloc_options(s);
spin_lock_init(&sbi->old_work_lock);
INIT_DELAYED_WORK(&sbi->old_work, flush_old_commits);
mutex_init(&sbi->lock);
sbi->lock_depth = -1;
reiserfs: kill-the-BKL This patch is an attempt to remove the Bkl based locking scheme from reiserfs and is intended. It is a bit inspired from an old attempt by Peter Zijlstra: http://lkml.indiana.edu/hypermail/linux/kernel/0704.2/2174.html The bkl is heavily used in this filesystem to prevent from concurrent write accesses on the filesystem. Reiserfs makes a deep use of the specific properties of the Bkl: - It can be acqquired recursively by a same task - It is released on the schedule() calls and reacquired when schedule() returns The two properties above are a roadmap for the reiserfs write locking so it's very hard to simply replace it with a common mutex. - We need a recursive-able locking unless we want to restructure several blocks of the code. - We need to identify the sites where the bkl was implictly relaxed (schedule, wait, sync, etc...) so that we can in turn release and reacquire our new lock explicitly. Such implicit releases of the lock are often required to let other resources producer/consumer do their job or we can suffer unexpected starvations or deadlocks. So the new lock that replaces the bkl here is a per superblock mutex with a specific property: it can be acquired recursively by a same task, like the bkl. For such purpose, we integrate a lock owner and a lock depth field on the superblock information structure. The first axis on this patch is to turn reiserfs_write_(un)lock() function into a wrapper to manage this mutex. Also some explicit calls to lock_kernel() have been converted to reiserfs_write_lock() helpers. The second axis is to find the important blocking sites (schedule...(), wait_on_buffer(), sync_dirty_buffer(), etc...) and then apply an explicit release of the write lock on these locations before blocking. Then we can safely wait for those who can give us resources or those who need some. Typically this is a fight between the current writer, the reiserfs workqueue (aka the async commiter) and the pdflush threads. The third axis is a consequence of the second. The write lock is usually on top of a lock dependency chain which can include the journal lock, the flush lock or the commit lock. So it's dangerous to release and trying to reacquire the write lock while we still hold other locks. This is fine with the bkl: T1 T2 lock_kernel() mutex_lock(A) unlock_kernel() // do something lock_kernel() mutex_lock(A) -> already locked by T1 schedule() (and then unlock_kernel()) lock_kernel() mutex_unlock(A) .... This is not fine with a mutex: T1 T2 mutex_lock(write) mutex_lock(A) mutex_unlock(write) // do something mutex_lock(write) mutex_lock(A) -> already locked by T1 schedule() mutex_lock(write) -> already locked by T2 deadlock The solution in this patch is to provide a helper which releases the write lock and sleep a bit if we can't lock a mutex that depend on it. It's another simulation of the bkl behaviour. The last axis is to locate the fs callbacks that are called with the bkl held, according to Documentation/filesystem/Locking. Those are: - reiserfs_remount - reiserfs_fill_super - reiserfs_put_super Reiserfs didn't need to explicitly lock because of the context of these callbacks. But now we must take care of that with the new locking. After this patch, reiserfs suffers from a slight performance regression (for now). On UP, a high volume write with dd reports an average of 27 MB/s instead of 30 MB/s without the patch applied. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Reviewed-by: Ingo Molnar <mingo@elte.hu> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Bron Gondwana <brong@fastmail.fm> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> LKML-Reference: <1239070789-13354-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-07 10:19:49 +08:00
sbi->commit_wq = alloc_workqueue("reiserfs/%s", WQ_MEM_RECLAIM, 0,
s->s_id);
if (!sbi->commit_wq) {
SWARN(silent, s, "", "Cannot allocate commit workqueue");
errval = -ENOMEM;
goto error_unlocked;
}
jdev_name = NULL;
if (reiserfs_parse_options
(s, (char *)data, &sbi->s_mount_opt, &blocks, &jdev_name,
&commit_max_age, qf_names, &qfmt) == 0) {
goto error_unlocked;
}
if (jdev_name && jdev_name[0]) {
sbi->s_jdev = kstrdup(jdev_name, GFP_KERNEL);
if (!sbi->s_jdev) {
SWARN(silent, s, "", "Cannot allocate memory for "
"journal device name");
goto error_unlocked;
}
}
#ifdef CONFIG_QUOTA
handle_quota_files(s, qf_names, &qfmt);
#endif
if (blocks) {
SWARN(silent, s, "jmacd-7", "resize option for remount only");
goto error_unlocked;
}
/*
* try old format (undistributed bitmap, super block in 8-th 1k
* block of a device)
*/
if (!read_super_block(s, REISERFS_OLD_DISK_OFFSET_IN_BYTES))
old_format = 1;
/*
* try new format (64-th 1k block), which can contain reiserfs
* super block
*/
else if (read_super_block(s, REISERFS_DISK_OFFSET_IN_BYTES)) {
SWARN(silent, s, "sh-2021", "can not find reiserfs on %s",
s->s_id);
goto error_unlocked;
}
s->s_time_min = 0;
s->s_time_max = U32_MAX;
rs = SB_DISK_SUPER_BLOCK(s);
/*
* Let's do basic sanity check to verify that underlying device is not
* smaller than the filesystem. If the check fails then abort and
* scream, because bad stuff will happen otherwise.
*/
if (s->s_bdev && s->s_bdev->bd_inode
&& i_size_read(s->s_bdev->bd_inode) <
sb_block_count(rs) * sb_blocksize(rs)) {
SWARN(silent, s, "", "Filesystem cannot be "
"mounted because it is bigger than the device");
SWARN(silent, s, "", "You may need to run fsck "
"or increase size of your LVM partition");
SWARN(silent, s, "", "Or may be you forgot to "
"reboot after fdisk when it told you to");
goto error_unlocked;
}
sbi->s_mount_state = SB_REISERFS_STATE(s);
sbi->s_mount_state = REISERFS_VALID_FS;
if ((errval = reiserfs_init_bitmap_cache(s))) {
SWARN(silent, s, "jmacd-8", "unable to read bitmap");
goto error_unlocked;
}
errval = -EINVAL;
#ifdef CONFIG_REISERFS_CHECK
SWARN(silent, s, "", "CONFIG_REISERFS_CHECK is set ON");
SWARN(silent, s, "", "- it is slow mode for debugging.");
#endif
/* make data=ordered the default */
if (!reiserfs_data_log(s) && !reiserfs_data_ordered(s) &&
!reiserfs_data_writeback(s)) {
sbi->s_mount_opt |= (1 << REISERFS_DATA_ORDERED);
}
if (reiserfs_data_log(s)) {
reiserfs_info(s, "using journaled data mode\n");
} else if (reiserfs_data_ordered(s)) {
reiserfs_info(s, "using ordered data mode\n");
} else {
reiserfs_info(s, "using writeback data mode\n");
}
if (reiserfs_barrier_flush(s)) {
printk("reiserfs: using flush barriers\n");
}
if (journal_init(s, jdev_name, old_format, commit_max_age)) {
SWARN(silent, s, "sh-2022",
"unable to initialize journal space");
goto error_unlocked;
} else {
/*
* once this is set, journal_release must be called
* if we error out of the mount
*/
jinit_done = 1;
}
if (reread_meta_blocks(s)) {
SWARN(silent, s, "jmacd-9",
"unable to reread meta blocks after journal init");
reiserfs: don't lock root inode searching Nothing requires that we lock the filesystem until the root inode is provided. Also iget5_locked() triggers a warning because we are holding the filesystem lock while allocating the inode, which result in a lockdep suspicion that we have a lock inversion against the reclaim path: [ 1986.896979] ================================= [ 1986.896990] [ INFO: inconsistent lock state ] [ 1986.896997] 3.1.1-main #8 [ 1986.897001] --------------------------------- [ 1986.897007] inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage. [ 1986.897016] kswapd0/16 [HC0[0]:SC0[0]:HE1:SE1] takes: [ 1986.897023] (&REISERFS_SB(s)->lock){+.+.?.}, at: [<c01f8bd4>] reiserfs_write_lock+0x20/0x2a [ 1986.897044] {RECLAIM_FS-ON-W} state was registered at: [ 1986.897050] [<c014a5b9>] mark_held_locks+0xae/0xd0 [ 1986.897060] [<c014aab3>] lockdep_trace_alloc+0x7d/0x91 [ 1986.897068] [<c0190ee0>] kmem_cache_alloc+0x1a/0x93 [ 1986.897078] [<c01e7728>] reiserfs_alloc_inode+0x13/0x3d [ 1986.897088] [<c01a5b06>] alloc_inode+0x14/0x5f [ 1986.897097] [<c01a5cb9>] iget5_locked+0x62/0x13a [ 1986.897106] [<c01e99e0>] reiserfs_fill_super+0x410/0x8b9 [ 1986.897114] [<c01953da>] mount_bdev+0x10b/0x159 [ 1986.897123] [<c01e764d>] get_super_block+0x10/0x12 [ 1986.897131] [<c0195b38>] mount_fs+0x59/0x12d [ 1986.897138] [<c01a80d1>] vfs_kern_mount+0x45/0x7a [ 1986.897147] [<c01a83e3>] do_kern_mount+0x2f/0xb0 [ 1986.897155] [<c01a987a>] do_mount+0x5c2/0x612 [ 1986.897163] [<c01a9a72>] sys_mount+0x61/0x8f [ 1986.897170] [<c044060c>] sysenter_do_call+0x12/0x32 [ 1986.897181] irq event stamp: 7509691 [ 1986.897186] hardirqs last enabled at (7509691): [<c0190f34>] kmem_cache_alloc+0x6e/0x93 [ 1986.897197] hardirqs last disabled at (7509690): [<c0190eea>] kmem_cache_alloc+0x24/0x93 [ 1986.897209] softirqs last enabled at (7508896): [<c01294bd>] __do_softirq+0xee/0xfd [ 1986.897222] softirqs last disabled at (7508859): [<c01030ed>] do_softirq+0x50/0x9d [ 1986.897234] [ 1986.897235] other info that might help us debug this: [ 1986.897242] Possible unsafe locking scenario: [ 1986.897244] [ 1986.897250] CPU0 [ 1986.897254] ---- [ 1986.897257] lock(&REISERFS_SB(s)->lock); [ 1986.897265] <Interrupt> [ 1986.897269] lock(&REISERFS_SB(s)->lock); [ 1986.897276] [ 1986.897277] *** DEADLOCK *** [ 1986.897278] [ 1986.897286] no locks held by kswapd0/16. [ 1986.897291] [ 1986.897292] stack backtrace: [ 1986.897299] Pid: 16, comm: kswapd0 Not tainted 3.1.1-main #8 [ 1986.897306] Call Trace: [ 1986.897314] [<c0439e76>] ? printk+0xf/0x11 [ 1986.897324] [<c01482d1>] print_usage_bug+0x20e/0x21a [ 1986.897332] [<c01479b8>] ? print_irq_inversion_bug+0x172/0x172 [ 1986.897341] [<c014855c>] mark_lock+0x27f/0x483 [ 1986.897349] [<c0148d88>] __lock_acquire+0x628/0x1472 [ 1986.897358] [<c0149fae>] lock_acquire+0x47/0x5e [ 1986.897366] [<c01f8bd4>] ? reiserfs_write_lock+0x20/0x2a [ 1986.897384] [<c01f8bd4>] ? reiserfs_write_lock+0x20/0x2a [ 1986.897397] [<c043b5ef>] mutex_lock_nested+0x35/0x26f [ 1986.897409] [<c01f8bd4>] ? reiserfs_write_lock+0x20/0x2a [ 1986.897421] [<c01f8bd4>] reiserfs_write_lock+0x20/0x2a [ 1986.897433] [<c01e2edd>] map_block_for_writepage+0xc9/0x590 [ 1986.897448] [<c01b1706>] ? create_empty_buffers+0x33/0x8f [ 1986.897461] [<c0121124>] ? get_parent_ip+0xb/0x31 [ 1986.897472] [<c043ef7f>] ? sub_preempt_count+0x81/0x8e [ 1986.897485] [<c043cae0>] ? _raw_spin_unlock+0x27/0x3d [ 1986.897496] [<c0121124>] ? get_parent_ip+0xb/0x31 [ 1986.897508] [<c01e355d>] reiserfs_writepage+0x1b9/0x3e7 [ 1986.897521] [<c0173b40>] ? clear_page_dirty_for_io+0xcb/0xde [ 1986.897533] [<c014a6e3>] ? trace_hardirqs_on_caller+0x108/0x138 [ 1986.897546] [<c014a71e>] ? trace_hardirqs_on+0xb/0xd [ 1986.897559] [<c0177b38>] shrink_page_list+0x34f/0x5e2 [ 1986.897572] [<c01780a7>] shrink_inactive_list+0x172/0x22c [ 1986.897585] [<c0178464>] shrink_zone+0x303/0x3b1 [ 1986.897597] [<c043cae0>] ? _raw_spin_unlock+0x27/0x3d [ 1986.897611] [<c01788c9>] kswapd+0x3b7/0x5f2 The deadlock shouldn't happen since we are doing that allocation in the mount path, the filesystem is not available for any reclaim. Still the warning is annoying. To solve this, acquire the lock later only where we need it, right before calling reiserfs_read_locked_inode() that wants to lock to walk the tree. Reported-by: Knut Petersen <Knut_Petersen@t-online.de> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Christoph Hellwig <hch@lst.de> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-11 07:11:11 +08:00
goto error_unlocked;
}
if (replay_only(s))
reiserfs: don't lock root inode searching Nothing requires that we lock the filesystem until the root inode is provided. Also iget5_locked() triggers a warning because we are holding the filesystem lock while allocating the inode, which result in a lockdep suspicion that we have a lock inversion against the reclaim path: [ 1986.896979] ================================= [ 1986.896990] [ INFO: inconsistent lock state ] [ 1986.896997] 3.1.1-main #8 [ 1986.897001] --------------------------------- [ 1986.897007] inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage. [ 1986.897016] kswapd0/16 [HC0[0]:SC0[0]:HE1:SE1] takes: [ 1986.897023] (&REISERFS_SB(s)->lock){+.+.?.}, at: [<c01f8bd4>] reiserfs_write_lock+0x20/0x2a [ 1986.897044] {RECLAIM_FS-ON-W} state was registered at: [ 1986.897050] [<c014a5b9>] mark_held_locks+0xae/0xd0 [ 1986.897060] [<c014aab3>] lockdep_trace_alloc+0x7d/0x91 [ 1986.897068] [<c0190ee0>] kmem_cache_alloc+0x1a/0x93 [ 1986.897078] [<c01e7728>] reiserfs_alloc_inode+0x13/0x3d [ 1986.897088] [<c01a5b06>] alloc_inode+0x14/0x5f [ 1986.897097] [<c01a5cb9>] iget5_locked+0x62/0x13a [ 1986.897106] [<c01e99e0>] reiserfs_fill_super+0x410/0x8b9 [ 1986.897114] [<c01953da>] mount_bdev+0x10b/0x159 [ 1986.897123] [<c01e764d>] get_super_block+0x10/0x12 [ 1986.897131] [<c0195b38>] mount_fs+0x59/0x12d [ 1986.897138] [<c01a80d1>] vfs_kern_mount+0x45/0x7a [ 1986.897147] [<c01a83e3>] do_kern_mount+0x2f/0xb0 [ 1986.897155] [<c01a987a>] do_mount+0x5c2/0x612 [ 1986.897163] [<c01a9a72>] sys_mount+0x61/0x8f [ 1986.897170] [<c044060c>] sysenter_do_call+0x12/0x32 [ 1986.897181] irq event stamp: 7509691 [ 1986.897186] hardirqs last enabled at (7509691): [<c0190f34>] kmem_cache_alloc+0x6e/0x93 [ 1986.897197] hardirqs last disabled at (7509690): [<c0190eea>] kmem_cache_alloc+0x24/0x93 [ 1986.897209] softirqs last enabled at (7508896): [<c01294bd>] __do_softirq+0xee/0xfd [ 1986.897222] softirqs last disabled at (7508859): [<c01030ed>] do_softirq+0x50/0x9d [ 1986.897234] [ 1986.897235] other info that might help us debug this: [ 1986.897242] Possible unsafe locking scenario: [ 1986.897244] [ 1986.897250] CPU0 [ 1986.897254] ---- [ 1986.897257] lock(&REISERFS_SB(s)->lock); [ 1986.897265] <Interrupt> [ 1986.897269] lock(&REISERFS_SB(s)->lock); [ 1986.897276] [ 1986.897277] *** DEADLOCK *** [ 1986.897278] [ 1986.897286] no locks held by kswapd0/16. [ 1986.897291] [ 1986.897292] stack backtrace: [ 1986.897299] Pid: 16, comm: kswapd0 Not tainted 3.1.1-main #8 [ 1986.897306] Call Trace: [ 1986.897314] [<c0439e76>] ? printk+0xf/0x11 [ 1986.897324] [<c01482d1>] print_usage_bug+0x20e/0x21a [ 1986.897332] [<c01479b8>] ? print_irq_inversion_bug+0x172/0x172 [ 1986.897341] [<c014855c>] mark_lock+0x27f/0x483 [ 1986.897349] [<c0148d88>] __lock_acquire+0x628/0x1472 [ 1986.897358] [<c0149fae>] lock_acquire+0x47/0x5e [ 1986.897366] [<c01f8bd4>] ? reiserfs_write_lock+0x20/0x2a [ 1986.897384] [<c01f8bd4>] ? reiserfs_write_lock+0x20/0x2a [ 1986.897397] [<c043b5ef>] mutex_lock_nested+0x35/0x26f [ 1986.897409] [<c01f8bd4>] ? reiserfs_write_lock+0x20/0x2a [ 1986.897421] [<c01f8bd4>] reiserfs_write_lock+0x20/0x2a [ 1986.897433] [<c01e2edd>] map_block_for_writepage+0xc9/0x590 [ 1986.897448] [<c01b1706>] ? create_empty_buffers+0x33/0x8f [ 1986.897461] [<c0121124>] ? get_parent_ip+0xb/0x31 [ 1986.897472] [<c043ef7f>] ? sub_preempt_count+0x81/0x8e [ 1986.897485] [<c043cae0>] ? _raw_spin_unlock+0x27/0x3d [ 1986.897496] [<c0121124>] ? get_parent_ip+0xb/0x31 [ 1986.897508] [<c01e355d>] reiserfs_writepage+0x1b9/0x3e7 [ 1986.897521] [<c0173b40>] ? clear_page_dirty_for_io+0xcb/0xde [ 1986.897533] [<c014a6e3>] ? trace_hardirqs_on_caller+0x108/0x138 [ 1986.897546] [<c014a71e>] ? trace_hardirqs_on+0xb/0xd [ 1986.897559] [<c0177b38>] shrink_page_list+0x34f/0x5e2 [ 1986.897572] [<c01780a7>] shrink_inactive_list+0x172/0x22c [ 1986.897585] [<c0178464>] shrink_zone+0x303/0x3b1 [ 1986.897597] [<c043cae0>] ? _raw_spin_unlock+0x27/0x3d [ 1986.897611] [<c01788c9>] kswapd+0x3b7/0x5f2 The deadlock shouldn't happen since we are doing that allocation in the mount path, the filesystem is not available for any reclaim. Still the warning is annoying. To solve this, acquire the lock later only where we need it, right before calling reiserfs_read_locked_inode() that wants to lock to walk the tree. Reported-by: Knut Petersen <Knut_Petersen@t-online.de> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Christoph Hellwig <hch@lst.de> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-11 07:11:11 +08:00
goto error_unlocked;
s->s_xattr = reiserfs_xattr_handlers;
if (bdev_read_only(s->s_bdev) && !sb_rdonly(s)) {
SWARN(silent, s, "clm-7000",
"Detected readonly device, marking FS readonly");
Rename superblock flags (MS_xyz -> SB_xyz) This is a pure automated search-and-replace of the internal kernel superblock flags. The s_flags are now called SB_*, with the names and the values for the moment mirroring the MS_* flags that they're equivalent to. Note how the MS_xyz flags are the ones passed to the mount system call, while the SB_xyz flags are what we then use in sb->s_flags. The script to do this was: # places to look in; re security/*: it generally should *not* be # touched (that stuff parses mount(2) arguments directly), but # there are two places where we really deal with superblock flags. FILES="drivers/mtd drivers/staging/lustre fs ipc mm \ include/linux/fs.h include/uapi/linux/bfs_fs.h \ security/apparmor/apparmorfs.c security/apparmor/include/lib.h" # the list of MS_... constants SYMS="RDONLY NOSUID NODEV NOEXEC SYNCHRONOUS REMOUNT MANDLOCK \ DIRSYNC NOATIME NODIRATIME BIND MOVE REC VERBOSE SILENT \ POSIXACL UNBINDABLE PRIVATE SLAVE SHARED RELATIME KERNMOUNT \ I_VERSION STRICTATIME LAZYTIME SUBMOUNT NOREMOTELOCK NOSEC BORN \ ACTIVE NOUSER" SED_PROG= for i in $SYMS; do SED_PROG="$SED_PROG -e s/MS_$i/SB_$i/g"; done # we want files that contain at least one of MS_..., # with fs/namespace.c and fs/pnode.c excluded. L=$(for i in $SYMS; do git grep -w -l MS_$i $FILES; done| sort|uniq|grep -v '^fs/namespace.c'|grep -v '^fs/pnode.c') for f in $L; do sed -i $f $SED_PROG; done Requested-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-28 05:05:09 +08:00
s->s_flags |= SB_RDONLY;
}
args.objectid = REISERFS_ROOT_OBJECTID;
args.dirid = REISERFS_ROOT_PARENT_OBJECTID;
root_inode =
iget5_locked(s, REISERFS_ROOT_OBJECTID, reiserfs_find_actor,
reiserfs_init_locked_inode, (void *)&args);
if (!root_inode) {
SWARN(silent, s, "jmacd-10", "get root inode failed");
reiserfs: don't lock root inode searching Nothing requires that we lock the filesystem until the root inode is provided. Also iget5_locked() triggers a warning because we are holding the filesystem lock while allocating the inode, which result in a lockdep suspicion that we have a lock inversion against the reclaim path: [ 1986.896979] ================================= [ 1986.896990] [ INFO: inconsistent lock state ] [ 1986.896997] 3.1.1-main #8 [ 1986.897001] --------------------------------- [ 1986.897007] inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage. [ 1986.897016] kswapd0/16 [HC0[0]:SC0[0]:HE1:SE1] takes: [ 1986.897023] (&REISERFS_SB(s)->lock){+.+.?.}, at: [<c01f8bd4>] reiserfs_write_lock+0x20/0x2a [ 1986.897044] {RECLAIM_FS-ON-W} state was registered at: [ 1986.897050] [<c014a5b9>] mark_held_locks+0xae/0xd0 [ 1986.897060] [<c014aab3>] lockdep_trace_alloc+0x7d/0x91 [ 1986.897068] [<c0190ee0>] kmem_cache_alloc+0x1a/0x93 [ 1986.897078] [<c01e7728>] reiserfs_alloc_inode+0x13/0x3d [ 1986.897088] [<c01a5b06>] alloc_inode+0x14/0x5f [ 1986.897097] [<c01a5cb9>] iget5_locked+0x62/0x13a [ 1986.897106] [<c01e99e0>] reiserfs_fill_super+0x410/0x8b9 [ 1986.897114] [<c01953da>] mount_bdev+0x10b/0x159 [ 1986.897123] [<c01e764d>] get_super_block+0x10/0x12 [ 1986.897131] [<c0195b38>] mount_fs+0x59/0x12d [ 1986.897138] [<c01a80d1>] vfs_kern_mount+0x45/0x7a [ 1986.897147] [<c01a83e3>] do_kern_mount+0x2f/0xb0 [ 1986.897155] [<c01a987a>] do_mount+0x5c2/0x612 [ 1986.897163] [<c01a9a72>] sys_mount+0x61/0x8f [ 1986.897170] [<c044060c>] sysenter_do_call+0x12/0x32 [ 1986.897181] irq event stamp: 7509691 [ 1986.897186] hardirqs last enabled at (7509691): [<c0190f34>] kmem_cache_alloc+0x6e/0x93 [ 1986.897197] hardirqs last disabled at (7509690): [<c0190eea>] kmem_cache_alloc+0x24/0x93 [ 1986.897209] softirqs last enabled at (7508896): [<c01294bd>] __do_softirq+0xee/0xfd [ 1986.897222] softirqs last disabled at (7508859): [<c01030ed>] do_softirq+0x50/0x9d [ 1986.897234] [ 1986.897235] other info that might help us debug this: [ 1986.897242] Possible unsafe locking scenario: [ 1986.897244] [ 1986.897250] CPU0 [ 1986.897254] ---- [ 1986.897257] lock(&REISERFS_SB(s)->lock); [ 1986.897265] <Interrupt> [ 1986.897269] lock(&REISERFS_SB(s)->lock); [ 1986.897276] [ 1986.897277] *** DEADLOCK *** [ 1986.897278] [ 1986.897286] no locks held by kswapd0/16. [ 1986.897291] [ 1986.897292] stack backtrace: [ 1986.897299] Pid: 16, comm: kswapd0 Not tainted 3.1.1-main #8 [ 1986.897306] Call Trace: [ 1986.897314] [<c0439e76>] ? printk+0xf/0x11 [ 1986.897324] [<c01482d1>] print_usage_bug+0x20e/0x21a [ 1986.897332] [<c01479b8>] ? print_irq_inversion_bug+0x172/0x172 [ 1986.897341] [<c014855c>] mark_lock+0x27f/0x483 [ 1986.897349] [<c0148d88>] __lock_acquire+0x628/0x1472 [ 1986.897358] [<c0149fae>] lock_acquire+0x47/0x5e [ 1986.897366] [<c01f8bd4>] ? reiserfs_write_lock+0x20/0x2a [ 1986.897384] [<c01f8bd4>] ? reiserfs_write_lock+0x20/0x2a [ 1986.897397] [<c043b5ef>] mutex_lock_nested+0x35/0x26f [ 1986.897409] [<c01f8bd4>] ? reiserfs_write_lock+0x20/0x2a [ 1986.897421] [<c01f8bd4>] reiserfs_write_lock+0x20/0x2a [ 1986.897433] [<c01e2edd>] map_block_for_writepage+0xc9/0x590 [ 1986.897448] [<c01b1706>] ? create_empty_buffers+0x33/0x8f [ 1986.897461] [<c0121124>] ? get_parent_ip+0xb/0x31 [ 1986.897472] [<c043ef7f>] ? sub_preempt_count+0x81/0x8e [ 1986.897485] [<c043cae0>] ? _raw_spin_unlock+0x27/0x3d [ 1986.897496] [<c0121124>] ? get_parent_ip+0xb/0x31 [ 1986.897508] [<c01e355d>] reiserfs_writepage+0x1b9/0x3e7 [ 1986.897521] [<c0173b40>] ? clear_page_dirty_for_io+0xcb/0xde [ 1986.897533] [<c014a6e3>] ? trace_hardirqs_on_caller+0x108/0x138 [ 1986.897546] [<c014a71e>] ? trace_hardirqs_on+0xb/0xd [ 1986.897559] [<c0177b38>] shrink_page_list+0x34f/0x5e2 [ 1986.897572] [<c01780a7>] shrink_inactive_list+0x172/0x22c [ 1986.897585] [<c0178464>] shrink_zone+0x303/0x3b1 [ 1986.897597] [<c043cae0>] ? _raw_spin_unlock+0x27/0x3d [ 1986.897611] [<c01788c9>] kswapd+0x3b7/0x5f2 The deadlock shouldn't happen since we are doing that allocation in the mount path, the filesystem is not available for any reclaim. Still the warning is annoying. To solve this, acquire the lock later only where we need it, right before calling reiserfs_read_locked_inode() that wants to lock to walk the tree. Reported-by: Knut Petersen <Knut_Petersen@t-online.de> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Christoph Hellwig <hch@lst.de> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-11 07:11:11 +08:00
goto error_unlocked;
}
reiserfs: don't lock root inode searching Nothing requires that we lock the filesystem until the root inode is provided. Also iget5_locked() triggers a warning because we are holding the filesystem lock while allocating the inode, which result in a lockdep suspicion that we have a lock inversion against the reclaim path: [ 1986.896979] ================================= [ 1986.896990] [ INFO: inconsistent lock state ] [ 1986.896997] 3.1.1-main #8 [ 1986.897001] --------------------------------- [ 1986.897007] inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage. [ 1986.897016] kswapd0/16 [HC0[0]:SC0[0]:HE1:SE1] takes: [ 1986.897023] (&REISERFS_SB(s)->lock){+.+.?.}, at: [<c01f8bd4>] reiserfs_write_lock+0x20/0x2a [ 1986.897044] {RECLAIM_FS-ON-W} state was registered at: [ 1986.897050] [<c014a5b9>] mark_held_locks+0xae/0xd0 [ 1986.897060] [<c014aab3>] lockdep_trace_alloc+0x7d/0x91 [ 1986.897068] [<c0190ee0>] kmem_cache_alloc+0x1a/0x93 [ 1986.897078] [<c01e7728>] reiserfs_alloc_inode+0x13/0x3d [ 1986.897088] [<c01a5b06>] alloc_inode+0x14/0x5f [ 1986.897097] [<c01a5cb9>] iget5_locked+0x62/0x13a [ 1986.897106] [<c01e99e0>] reiserfs_fill_super+0x410/0x8b9 [ 1986.897114] [<c01953da>] mount_bdev+0x10b/0x159 [ 1986.897123] [<c01e764d>] get_super_block+0x10/0x12 [ 1986.897131] [<c0195b38>] mount_fs+0x59/0x12d [ 1986.897138] [<c01a80d1>] vfs_kern_mount+0x45/0x7a [ 1986.897147] [<c01a83e3>] do_kern_mount+0x2f/0xb0 [ 1986.897155] [<c01a987a>] do_mount+0x5c2/0x612 [ 1986.897163] [<c01a9a72>] sys_mount+0x61/0x8f [ 1986.897170] [<c044060c>] sysenter_do_call+0x12/0x32 [ 1986.897181] irq event stamp: 7509691 [ 1986.897186] hardirqs last enabled at (7509691): [<c0190f34>] kmem_cache_alloc+0x6e/0x93 [ 1986.897197] hardirqs last disabled at (7509690): [<c0190eea>] kmem_cache_alloc+0x24/0x93 [ 1986.897209] softirqs last enabled at (7508896): [<c01294bd>] __do_softirq+0xee/0xfd [ 1986.897222] softirqs last disabled at (7508859): [<c01030ed>] do_softirq+0x50/0x9d [ 1986.897234] [ 1986.897235] other info that might help us debug this: [ 1986.897242] Possible unsafe locking scenario: [ 1986.897244] [ 1986.897250] CPU0 [ 1986.897254] ---- [ 1986.897257] lock(&REISERFS_SB(s)->lock); [ 1986.897265] <Interrupt> [ 1986.897269] lock(&REISERFS_SB(s)->lock); [ 1986.897276] [ 1986.897277] *** DEADLOCK *** [ 1986.897278] [ 1986.897286] no locks held by kswapd0/16. [ 1986.897291] [ 1986.897292] stack backtrace: [ 1986.897299] Pid: 16, comm: kswapd0 Not tainted 3.1.1-main #8 [ 1986.897306] Call Trace: [ 1986.897314] [<c0439e76>] ? printk+0xf/0x11 [ 1986.897324] [<c01482d1>] print_usage_bug+0x20e/0x21a [ 1986.897332] [<c01479b8>] ? print_irq_inversion_bug+0x172/0x172 [ 1986.897341] [<c014855c>] mark_lock+0x27f/0x483 [ 1986.897349] [<c0148d88>] __lock_acquire+0x628/0x1472 [ 1986.897358] [<c0149fae>] lock_acquire+0x47/0x5e [ 1986.897366] [<c01f8bd4>] ? reiserfs_write_lock+0x20/0x2a [ 1986.897384] [<c01f8bd4>] ? reiserfs_write_lock+0x20/0x2a [ 1986.897397] [<c043b5ef>] mutex_lock_nested+0x35/0x26f [ 1986.897409] [<c01f8bd4>] ? reiserfs_write_lock+0x20/0x2a [ 1986.897421] [<c01f8bd4>] reiserfs_write_lock+0x20/0x2a [ 1986.897433] [<c01e2edd>] map_block_for_writepage+0xc9/0x590 [ 1986.897448] [<c01b1706>] ? create_empty_buffers+0x33/0x8f [ 1986.897461] [<c0121124>] ? get_parent_ip+0xb/0x31 [ 1986.897472] [<c043ef7f>] ? sub_preempt_count+0x81/0x8e [ 1986.897485] [<c043cae0>] ? _raw_spin_unlock+0x27/0x3d [ 1986.897496] [<c0121124>] ? get_parent_ip+0xb/0x31 [ 1986.897508] [<c01e355d>] reiserfs_writepage+0x1b9/0x3e7 [ 1986.897521] [<c0173b40>] ? clear_page_dirty_for_io+0xcb/0xde [ 1986.897533] [<c014a6e3>] ? trace_hardirqs_on_caller+0x108/0x138 [ 1986.897546] [<c014a71e>] ? trace_hardirqs_on+0xb/0xd [ 1986.897559] [<c0177b38>] shrink_page_list+0x34f/0x5e2 [ 1986.897572] [<c01780a7>] shrink_inactive_list+0x172/0x22c [ 1986.897585] [<c0178464>] shrink_zone+0x303/0x3b1 [ 1986.897597] [<c043cae0>] ? _raw_spin_unlock+0x27/0x3d [ 1986.897611] [<c01788c9>] kswapd+0x3b7/0x5f2 The deadlock shouldn't happen since we are doing that allocation in the mount path, the filesystem is not available for any reclaim. Still the warning is annoying. To solve this, acquire the lock later only where we need it, right before calling reiserfs_read_locked_inode() that wants to lock to walk the tree. Reported-by: Knut Petersen <Knut_Petersen@t-online.de> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Christoph Hellwig <hch@lst.de> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-01-11 07:11:11 +08:00
/*
* This path assumed to be called with the BKL in the old times.
* Now we have inherited the big reiserfs lock from it and many
* reiserfs helpers called in the mount path and elsewhere require
* this lock to be held even if it's not always necessary. Let's be
* conservative and hold it early. The window can be reduced after
* careful review of the code.
*/
reiserfs_write_lock(s);
if (root_inode->i_state & I_NEW) {
reiserfs_read_locked_inode(root_inode, &args);
unlock_new_inode(root_inode);
}
if (!S_ISDIR(root_inode->i_mode) || !inode_get_bytes(root_inode) ||
!root_inode->i_size) {
SWARN(silent, s, "", "corrupt root inode, run fsck");
iput(root_inode);
errval = -EUCLEAN;
goto error;
}
s->s_root = d_make_root(root_inode);
if (!s->s_root)
goto error;
/* define and initialize hash function */
sbi->s_hash_function = hash_function(s);
if (sbi->s_hash_function == NULL) {
dput(s->s_root);
s->s_root = NULL;
goto error;
}
if (is_reiserfs_3_5(rs)
|| (is_reiserfs_jr(rs) && SB_VERSION(s) == REISERFS_VERSION_1))
set_bit(REISERFS_3_5, &sbi->s_properties);
else if (old_format)
set_bit(REISERFS_OLD_FORMAT, &sbi->s_properties);
else
set_bit(REISERFS_3_6, &sbi->s_properties);
if (!sb_rdonly(s)) {
errval = journal_begin(&th, s, 1);
if (errval) {
dput(s->s_root);
s->s_root = NULL;
goto error;
}
reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s), 1);
set_sb_umount_state(rs, REISERFS_ERROR_FS);
set_sb_fs_state(rs, 0);
/*
* Clear out s_bmap_nr if it would wrap. We can handle this
* case, but older revisions can't. This will cause the
* file system to fail mount on those older implementations,
* avoiding corruption. -jeffm
*/
if (bmap_would_wrap(reiserfs_bmap_count(s)) &&
sb_bmap_nr(rs) != 0) {
reiserfs_warning(s, "super-2030", "This file system "
"claims to use %u bitmap blocks in "
"its super block, but requires %u. "
"Clearing to zero.", sb_bmap_nr(rs),
reiserfs_bmap_count(s));
set_sb_bmap_nr(rs, 0);
}
if (old_format_only(s)) {
/*
* filesystem of format 3.5 either with standard
* or non-standard journal
*/
if (convert_reiserfs(s)) {
/* and -o conv is given */
if (!silent)
reiserfs_info(s,
"converting 3.5 filesystem to the 3.6 format");
if (is_reiserfs_3_5(rs))
/*
* put magic string of 3.6 format.
* 2.2 will not be able to
* mount this filesystem anymore
*/
memcpy(rs->s_v1.s_magic,
reiserfs_3_6_magic_string,
sizeof
(reiserfs_3_6_magic_string));
set_sb_version(rs, REISERFS_VERSION_2);
reiserfs_convert_objectid_map_v1(s);
set_bit(REISERFS_3_6, &sbi->s_properties);
clear_bit(REISERFS_3_5, &sbi->s_properties);
} else if (!silent) {
reiserfs_info(s, "using 3.5.x disk format\n");
}
} else
set_sb_mnt_count(rs, sb_mnt_count(rs) + 1);
journal_mark_dirty(&th, SB_BUFFER_WITH_SB(s));
errval = journal_end(&th);
if (errval) {
dput(s->s_root);
s->s_root = NULL;
goto error;
}
reiserfs_write_unlock(s);
if ((errval = reiserfs_lookup_privroot(s)) ||
(errval = reiserfs_xattr_init(s, s->s_flags))) {
dput(s->s_root);
s->s_root = NULL;
goto error_unlocked;
}
reiserfs_write_lock(s);
/*
* look for files which were to be removed in previous session
*/
finish_unfinished(s);
} else {
if (old_format_only(s) && !silent) {
reiserfs_info(s, "using 3.5.x disk format\n");
}
reiserfs_write_unlock(s);
if ((errval = reiserfs_lookup_privroot(s)) ||
(errval = reiserfs_xattr_init(s, s->s_flags))) {
dput(s->s_root);
s->s_root = NULL;
goto error_unlocked;
}
reiserfs_write_lock(s);
}
/*
* mark hash in super block: it could be unset. overwrite should be ok
*/
set_sb_hash_function_code(rs, function2code(sbi->s_hash_function));
handle_attrs(s);
reiserfs_proc_info_init(s);
init_waitqueue_head(&(sbi->s_wait));
spin_lock_init(&sbi->bitmap_lock);
reiserfs: kill-the-BKL This patch is an attempt to remove the Bkl based locking scheme from reiserfs and is intended. It is a bit inspired from an old attempt by Peter Zijlstra: http://lkml.indiana.edu/hypermail/linux/kernel/0704.2/2174.html The bkl is heavily used in this filesystem to prevent from concurrent write accesses on the filesystem. Reiserfs makes a deep use of the specific properties of the Bkl: - It can be acqquired recursively by a same task - It is released on the schedule() calls and reacquired when schedule() returns The two properties above are a roadmap for the reiserfs write locking so it's very hard to simply replace it with a common mutex. - We need a recursive-able locking unless we want to restructure several blocks of the code. - We need to identify the sites where the bkl was implictly relaxed (schedule, wait, sync, etc...) so that we can in turn release and reacquire our new lock explicitly. Such implicit releases of the lock are often required to let other resources producer/consumer do their job or we can suffer unexpected starvations or deadlocks. So the new lock that replaces the bkl here is a per superblock mutex with a specific property: it can be acquired recursively by a same task, like the bkl. For such purpose, we integrate a lock owner and a lock depth field on the superblock information structure. The first axis on this patch is to turn reiserfs_write_(un)lock() function into a wrapper to manage this mutex. Also some explicit calls to lock_kernel() have been converted to reiserfs_write_lock() helpers. The second axis is to find the important blocking sites (schedule...(), wait_on_buffer(), sync_dirty_buffer(), etc...) and then apply an explicit release of the write lock on these locations before blocking. Then we can safely wait for those who can give us resources or those who need some. Typically this is a fight between the current writer, the reiserfs workqueue (aka the async commiter) and the pdflush threads. The third axis is a consequence of the second. The write lock is usually on top of a lock dependency chain which can include the journal lock, the flush lock or the commit lock. So it's dangerous to release and trying to reacquire the write lock while we still hold other locks. This is fine with the bkl: T1 T2 lock_kernel() mutex_lock(A) unlock_kernel() // do something lock_kernel() mutex_lock(A) -> already locked by T1 schedule() (and then unlock_kernel()) lock_kernel() mutex_unlock(A) .... This is not fine with a mutex: T1 T2 mutex_lock(write) mutex_lock(A) mutex_unlock(write) // do something mutex_lock(write) mutex_lock(A) -> already locked by T1 schedule() mutex_lock(write) -> already locked by T2 deadlock The solution in this patch is to provide a helper which releases the write lock and sleep a bit if we can't lock a mutex that depend on it. It's another simulation of the bkl behaviour. The last axis is to locate the fs callbacks that are called with the bkl held, according to Documentation/filesystem/Locking. Those are: - reiserfs_remount - reiserfs_fill_super - reiserfs_put_super Reiserfs didn't need to explicitly lock because of the context of these callbacks. But now we must take care of that with the new locking. After this patch, reiserfs suffers from a slight performance regression (for now). On UP, a high volume write with dd reports an average of 27 MB/s instead of 30 MB/s without the patch applied. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Reviewed-by: Ingo Molnar <mingo@elte.hu> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Bron Gondwana <brong@fastmail.fm> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> LKML-Reference: <1239070789-13354-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-04-07 10:19:49 +08:00
reiserfs_write_unlock(s);
return (0);
error:
reiserfs_write_unlock(s);
error_unlocked:
/* kill the commit thread, free journal ram */
if (jinit_done) {
reiserfs_write_lock(s);
journal_release_error(NULL, s);
reiserfs_write_unlock(s);
}
if (sbi->commit_wq)
destroy_workqueue(sbi->commit_wq);
reiserfs_cancel_old_flush(s);
reiserfs_free_bitmap_cache(s);
if (SB_BUFFER_WITH_SB(s))
brelse(SB_BUFFER_WITH_SB(s));
#ifdef CONFIG_QUOTA
{
int j;
for (j = 0; j < REISERFS_MAXQUOTAS; j++)
kfree(qf_names[j]);
}
#endif
kfree(sbi->s_jdev);
kfree(sbi);
s->s_fs_info = NULL;
return errval;
}
static int reiserfs_statfs(struct dentry *dentry, struct kstatfs *buf)
{
struct reiserfs_super_block *rs = SB_DISK_SUPER_BLOCK(dentry->d_sb);
buf->f_namelen = (REISERFS_MAX_NAME(s->s_blocksize));
buf->f_bfree = sb_free_blocks(rs);
buf->f_bavail = buf->f_bfree;
buf->f_blocks = sb_block_count(rs) - sb_bmap_nr(rs) - 1;
buf->f_bsize = dentry->d_sb->s_blocksize;
/* changed to accommodate gcc folks. */
buf->f_type = REISERFS_SUPER_MAGIC;
buf->f_fsid.val[0] = (u32)crc32_le(0, rs->s_uuid, sizeof(rs->s_uuid)/2);
buf->f_fsid.val[1] = (u32)crc32_le(0, rs->s_uuid + sizeof(rs->s_uuid)/2,
sizeof(rs->s_uuid)/2);
return 0;
}
#ifdef CONFIG_QUOTA
static int reiserfs_write_dquot(struct dquot *dquot)
{
struct reiserfs_transaction_handle th;
int ret, err;
int depth;
reiserfs_write_lock(dquot->dq_sb);
ret =
journal_begin(&th, dquot->dq_sb,
REISERFS_QUOTA_TRANS_BLOCKS(dquot->dq_sb));
if (ret)
goto out;
depth = reiserfs_write_unlock_nested(dquot->dq_sb);
ret = dquot_commit(dquot);
reiserfs_write_lock_nested(dquot->dq_sb, depth);
err = journal_end(&th);
if (!ret && err)
ret = err;
out:
reiserfs_write_unlock(dquot->dq_sb);
return ret;
}
static int reiserfs_acquire_dquot(struct dquot *dquot)
{
struct reiserfs_transaction_handle th;
int ret, err;
int depth;
reiserfs_write_lock(dquot->dq_sb);
ret =
journal_begin(&th, dquot->dq_sb,
REISERFS_QUOTA_INIT_BLOCKS(dquot->dq_sb));
if (ret)
goto out;
depth = reiserfs_write_unlock_nested(dquot->dq_sb);
ret = dquot_acquire(dquot);
reiserfs_write_lock_nested(dquot->dq_sb, depth);
err = journal_end(&th);
if (!ret && err)
ret = err;
out:
reiserfs_write_unlock(dquot->dq_sb);
return ret;
}
static int reiserfs_release_dquot(struct dquot *dquot)
{
struct reiserfs_transaction_handle th;
int ret, err;
reiserfs_write_lock(dquot->dq_sb);
ret =
journal_begin(&th, dquot->dq_sb,
REISERFS_QUOTA_DEL_BLOCKS(dquot->dq_sb));
reiserfs_write_unlock(dquot->dq_sb);
if (ret) {
/* Release dquot anyway to avoid endless cycle in dqput() */
dquot_release(dquot);
goto out;
}
ret = dquot_release(dquot);
reiserfs_write_lock(dquot->dq_sb);
err = journal_end(&th);
if (!ret && err)
ret = err;
reiserfs_write_unlock(dquot->dq_sb);
out:
return ret;
}
static int reiserfs_mark_dquot_dirty(struct dquot *dquot)
{
/* Are we journaling quotas? */
if (REISERFS_SB(dquot->dq_sb)->s_qf_names[USRQUOTA] ||
REISERFS_SB(dquot->dq_sb)->s_qf_names[GRPQUOTA]) {
dquot_mark_dquot_dirty(dquot);
return reiserfs_write_dquot(dquot);
} else
return dquot_mark_dquot_dirty(dquot);
}
static int reiserfs_write_info(struct super_block *sb, int type)
{
struct reiserfs_transaction_handle th;
int ret, err;
int depth;
/* Data block + inode block */
reiserfs_write_lock(sb);
ret = journal_begin(&th, sb, 2);
if (ret)
goto out;
depth = reiserfs_write_unlock_nested(sb);
ret = dquot_commit_info(sb, type);
reiserfs_write_lock_nested(sb, depth);
err = journal_end(&th);
if (!ret && err)
ret = err;
out:
reiserfs_write_unlock(sb);
return ret;
}
/*
* Turn on quotas during mount time - we need to find the quota file and such...
*/
static int reiserfs_quota_on_mount(struct super_block *sb, int type)
{
return dquot_quota_on_mount(sb, REISERFS_SB(sb)->s_qf_names[type],
REISERFS_SB(sb)->s_jquota_fmt, type);
}
/*
* Standard function to be called on quota_on
*/
static int reiserfs_quota_on(struct super_block *sb, int type, int format_id,
const struct path *path)
{
int err;
struct inode *inode;
struct reiserfs_transaction_handle th;
int opt = type == USRQUOTA ? REISERFS_USRQUOTA : REISERFS_GRPQUOTA;
reiserfs_write_lock(sb);
if (!(REISERFS_SB(sb)->s_mount_opt & (1 << opt))) {
err = -EINVAL;
goto out;
}
/* Quotafile not on the same filesystem? */
if (path->dentry->d_sb != sb) {
err = -EXDEV;
goto out;
}
inode = d_inode(path->dentry);
/*
* We must not pack tails for quota files on reiserfs for quota
* IO to work
*/
if (!(REISERFS_I(inode)->i_flags & i_nopack_mask)) {
err = reiserfs_unpack(inode, NULL);
if (err) {
reiserfs_warning(sb, "super-6520",
"Unpacking tail of quota file failed"
" (%d). Cannot turn on quotas.", err);
err = -EINVAL;
goto out;
}
mark_inode_dirty(inode);
}
/* Journaling quota? */
if (REISERFS_SB(sb)->s_qf_names[type]) {
/* Quotafile not of fs root? */
if (path->dentry->d_parent != sb->s_root)
reiserfs_warning(sb, "super-6521",
"Quota file not on filesystem root. "
"Journalled quota will not work.");
}
/*
* When we journal data on quota file, we have to flush journal to see
* all updates to the file when we bypass pagecache...
*/
if (reiserfs_file_data_log(inode)) {
/* Just start temporary transaction and finish it */
err = journal_begin(&th, sb, 1);
if (err)
goto out;
err = journal_end_sync(&th);
if (err)
goto out;
}
reiserfs_write_unlock(sb);
err = dquot_quota_on(sb, type, format_id, path);
if (!err) {
inode_lock(inode);
REISERFS_I(inode)->i_attrs |= REISERFS_IMMUTABLE_FL |
REISERFS_NOATIME_FL;
inode_set_flags(inode, S_IMMUTABLE | S_NOATIME,
S_IMMUTABLE | S_NOATIME);
inode_unlock(inode);
mark_inode_dirty(inode);
}
return err;
out:
reiserfs_write_unlock(sb);
return err;
}
static int reiserfs_quota_off(struct super_block *sb, int type)
{
int err;
struct inode *inode = sb_dqopt(sb)->files[type];
if (!inode || !igrab(inode))
goto out;
err = dquot_quota_off(sb, type);
if (err)
goto out_put;
inode_lock(inode);
REISERFS_I(inode)->i_attrs &= ~(REISERFS_IMMUTABLE_FL |
REISERFS_NOATIME_FL);
inode_set_flags(inode, 0, S_IMMUTABLE | S_NOATIME);
inode_unlock(inode);
mark_inode_dirty(inode);
out_put:
iput(inode);
return err;
out:
return dquot_quota_off(sb, type);
}
/*
* Read data from quotafile - avoid pagecache and such because we cannot afford
* acquiring the locks... As quota files are never truncated and quota code
* itself serializes the operations (and no one else should touch the files)
* we don't have to be afraid of races
*/
static ssize_t reiserfs_quota_read(struct super_block *sb, int type, char *data,
size_t len, loff_t off)
{
struct inode *inode = sb_dqopt(sb)->files[type];
unsigned long blk = off >> sb->s_blocksize_bits;
int err = 0, offset = off & (sb->s_blocksize - 1), tocopy;
size_t toread;
struct buffer_head tmp_bh, *bh;
loff_t i_size = i_size_read(inode);
if (off > i_size)
return 0;
if (off + len > i_size)
len = i_size - off;
toread = len;
while (toread > 0) {
tocopy =
sb->s_blocksize - offset <
toread ? sb->s_blocksize - offset : toread;
tmp_bh.b_state = 0;
/*
* Quota files are without tails so we can safely
* use this function
*/
reiserfs_write_lock(sb);
err = reiserfs_get_block(inode, blk, &tmp_bh, 0);
reiserfs_write_unlock(sb);
if (err)
return err;
if (!buffer_mapped(&tmp_bh)) /* A hole? */
memset(data, 0, tocopy);
else {
bh = sb_bread(sb, tmp_bh.b_blocknr);
if (!bh)
return -EIO;
memcpy(data, bh->b_data + offset, tocopy);
brelse(bh);
}
offset = 0;
toread -= tocopy;
data += tocopy;
blk++;
}
return len;
}
/*
* Write to quotafile (we know the transaction is already started and has
* enough credits)
*/
static ssize_t reiserfs_quota_write(struct super_block *sb, int type,
const char *data, size_t len, loff_t off)
{
struct inode *inode = sb_dqopt(sb)->files[type];
unsigned long blk = off >> sb->s_blocksize_bits;
int err = 0, offset = off & (sb->s_blocksize - 1), tocopy;
int journal_quota = REISERFS_SB(sb)->s_qf_names[type] != NULL;
size_t towrite = len;
struct buffer_head tmp_bh, *bh;
if (!current->journal_info) {
printk(KERN_WARNING "reiserfs: Quota write (off=%llu, len=%llu) cancelled because transaction is not started.\n",
(unsigned long long)off, (unsigned long long)len);
return -EIO;
}
while (towrite > 0) {
tocopy = sb->s_blocksize - offset < towrite ?
sb->s_blocksize - offset : towrite;
tmp_bh.b_state = 0;
reiserfs_write_lock(sb);
err = reiserfs_get_block(inode, blk, &tmp_bh, GET_BLOCK_CREATE);
reiserfs_write_unlock(sb);
if (err)
goto out;
if (offset || tocopy != sb->s_blocksize)
bh = sb_bread(sb, tmp_bh.b_blocknr);
else
bh = sb_getblk(sb, tmp_bh.b_blocknr);
if (!bh) {
err = -EIO;
goto out;
}
lock_buffer(bh);
memcpy(bh->b_data + offset, data, tocopy);
flush_dcache_page(bh->b_page);
set_buffer_uptodate(bh);
unlock_buffer(bh);
reiserfs_write_lock(sb);
reiserfs_prepare_for_journal(sb, bh, 1);
journal_mark_dirty(current->journal_info, bh);
if (!journal_quota)
reiserfs_add_ordered_list(inode, bh);
reiserfs_write_unlock(sb);
brelse(bh);
offset = 0;
towrite -= tocopy;
data += tocopy;
blk++;
}
out:
if (len == towrite)
return err;
if (inode->i_size < off + len - towrite)
i_size_write(inode, off + len - towrite);
inode->i_mtime = inode->i_ctime = current_time(inode);
mark_inode_dirty(inode);
return len - towrite;
}
#endif
static struct dentry *get_super_block(struct file_system_type *fs_type,
[PATCH] VFS: Permit filesystem to override root dentry on mount Extend the get_sb() filesystem operation to take an extra argument that permits the VFS to pass in the target vfsmount that defines the mountpoint. The filesystem is then required to manually set the superblock and root dentry pointers. For most filesystems, this should be done with simple_set_mnt() which will set the superblock pointer and then set the root dentry to the superblock's s_root (as per the old default behaviour). The get_sb() op now returns an integer as there's now no need to return the superblock pointer. This patch permits a superblock to be implicitly shared amongst several mount points, such as can be done with NFS to avoid potential inode aliasing. In such a case, simple_set_mnt() would not be called, and instead the mnt_root and mnt_sb would be set directly. The patch also makes the following changes: (*) the get_sb_*() convenience functions in the core kernel now take a vfsmount pointer argument and return an integer, so most filesystems have to change very little. (*) If one of the convenience function is not used, then get_sb() should normally call simple_set_mnt() to instantiate the vfsmount. This will always return 0, and so can be tail-called from get_sb(). (*) generic_shutdown_super() now calls shrink_dcache_sb() to clean up the dcache upon superblock destruction rather than shrink_dcache_anon(). This is required because the superblock may now have multiple trees that aren't actually bound to s_root, but that still need to be cleaned up. The currently called functions assume that the whole tree is rooted at s_root, and that anonymous dentries are not the roots of trees which results in dentries being left unculled. However, with the way NFS superblock sharing are currently set to be implemented, these assumptions are violated: the root of the filesystem is simply a dummy dentry and inode (the real inode for '/' may well be inaccessible), and all the vfsmounts are rooted on anonymous[*] dentries with child trees. [*] Anonymous until discovered from another tree. (*) The documentation has been adjusted, including the additional bit of changing ext2_* into foo_* in the documentation. [akpm@osdl.org: convert ipath_fs, do other stuff] Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: Al Viro <viro@zeniv.linux.org.uk> Cc: Nathan Scott <nathans@sgi.com> Cc: Roland Dreier <rolandd@cisco.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-23 17:02:57 +08:00
int flags, const char *dev_name,
void *data)
{
return mount_bdev(fs_type, flags, dev_name, data, reiserfs_fill_super);
}
static int __init init_reiserfs_fs(void)
{
int ret;
ret = init_inodecache();
if (ret)
return ret;
reiserfs_proc_info_global_init();
ret = register_filesystem(&reiserfs_fs_type);
if (ret)
goto out;
return 0;
out:
reiserfs_proc_info_global_done();
destroy_inodecache();
return ret;
}
static void __exit exit_reiserfs_fs(void)
{
reiserfs_proc_info_global_done();
unregister_filesystem(&reiserfs_fs_type);
destroy_inodecache();
}
struct file_system_type reiserfs_fs_type = {
.owner = THIS_MODULE,
.name = "reiserfs",
.mount = get_super_block,
.kill_sb = reiserfs_kill_sb,
.fs_flags = FS_REQUIRES_DEV,
};
fs: Limit sys_mount to only request filesystem modules. Modify the request_module to prefix the file system type with "fs-" and add aliases to all of the filesystems that can be built as modules to match. A common practice is to build all of the kernel code and leave code that is not commonly needed as modules, with the result that many users are exposed to any bug anywhere in the kernel. Looking for filesystems with a fs- prefix limits the pool of possible modules that can be loaded by mount to just filesystems trivially making things safer with no real cost. Using aliases means user space can control the policy of which filesystem modules are auto-loaded by editing /etc/modprobe.d/*.conf with blacklist and alias directives. Allowing simple, safe, well understood work-arounds to known problematic software. This also addresses a rare but unfortunate problem where the filesystem name is not the same as it's module name and module auto-loading would not work. While writing this patch I saw a handful of such cases. The most significant being autofs that lives in the module autofs4. This is relevant to user namespaces because we can reach the request module in get_fs_type() without having any special permissions, and people get uncomfortable when a user specified string (in this case the filesystem type) goes all of the way to request_module. After having looked at this issue I don't think there is any particular reason to perform any filtering or permission checks beyond making it clear in the module request that we want a filesystem module. The common pattern in the kernel is to call request_module() without regards to the users permissions. In general all a filesystem module does once loaded is call register_filesystem() and go to sleep. Which means there is not much attack surface exposed by loading a filesytem module unless the filesystem is mounted. In a user namespace filesystems are not mounted unless .fs_flags = FS_USERNS_MOUNT, which most filesystems do not set today. Acked-by: Serge Hallyn <serge.hallyn@canonical.com> Acked-by: Kees Cook <keescook@chromium.org> Reported-by: Kees Cook <keescook@google.com> Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
2013-03-03 11:39:14 +08:00
MODULE_ALIAS_FS("reiserfs");
MODULE_DESCRIPTION("ReiserFS journaled filesystem");
MODULE_AUTHOR("Hans Reiser <reiser@namesys.com>");
MODULE_LICENSE("GPL");
module_init(init_reiserfs_fs);
module_exit(exit_reiserfs_fs);