#include <linux/slab.h>
#include <linux/highmem.h>
#include <linux/pagemap.h>
-#include <linux/smp_lock.h>
+#include <linux/quotaops.h>
#include <asm/byteorder.h>
#include "ocfs2.h"
#include "alloc.h"
+#include "dir.h"
+#include "blockcheck.h"
#include "dlmglue.h"
#include "extent_map.h"
#include "file.h"
#include "symlink.h"
#include "sysfile.h"
#include "uptodate.h"
-#include "vote.h"
+#include "xattr.h"
+#include "refcounttree.h"
#include "buffer_head_io.h"
u64 fi_blkno;
unsigned long fi_ino;
unsigned int fi_flags;
+ unsigned int fi_sysfile_type;
};
+static struct lock_class_key ocfs2_sysfile_lock_key[NUM_SYSTEM_INODES];
+
static int ocfs2_read_locked_inode(struct inode *inode,
struct ocfs2_find_inode_args *args);
static int ocfs2_init_locked_inode(struct inode *inode, void *opaque);
inode->i_flags |= S_DIRSYNC;
}
-struct inode *ocfs2_iget(struct ocfs2_super *osb, u64 blkno, int flags)
+/* Propagate flags from i_flags to OCFS2_I(inode)->ip_attr */
+void ocfs2_get_inode_flags(struct ocfs2_inode_info *oi)
+{
+ unsigned int flags = oi->vfs_inode.i_flags;
+
+ oi->ip_attr &= ~(OCFS2_SYNC_FL|OCFS2_APPEND_FL|
+ OCFS2_IMMUTABLE_FL|OCFS2_NOATIME_FL|OCFS2_DIRSYNC_FL);
+ if (flags & S_SYNC)
+ oi->ip_attr |= OCFS2_SYNC_FL;
+ if (flags & S_APPEND)
+ oi->ip_attr |= OCFS2_APPEND_FL;
+ if (flags & S_IMMUTABLE)
+ oi->ip_attr |= OCFS2_IMMUTABLE_FL;
+ if (flags & S_NOATIME)
+ oi->ip_attr |= OCFS2_NOATIME_FL;
+ if (flags & S_DIRSYNC)
+ oi->ip_attr |= OCFS2_DIRSYNC_FL;
+}
+
+struct inode *ocfs2_ilookup(struct super_block *sb, u64 blkno)
+{
+ struct ocfs2_find_inode_args args;
+
+ args.fi_blkno = blkno;
+ args.fi_flags = 0;
+ args.fi_ino = ino_from_blkno(sb, blkno);
+ args.fi_sysfile_type = 0;
+
+ return ilookup5(sb, blkno, ocfs2_find_actor, &args);
+}
+struct inode *ocfs2_iget(struct ocfs2_super *osb, u64 blkno, unsigned flags,
+ int sysfile_type)
{
struct inode *inode = NULL;
struct super_block *sb = osb->sb;
args.fi_blkno = blkno;
args.fi_flags = flags;
args.fi_ino = ino_from_blkno(sb, blkno);
+ args.fi_sysfile_type = sysfile_type;
inode = iget5_locked(sb, args.fi_ino, ocfs2_find_actor,
ocfs2_init_locked_inode, &args);
static int ocfs2_init_locked_inode(struct inode *inode, void *opaque)
{
struct ocfs2_find_inode_args *args = opaque;
+ static struct lock_class_key ocfs2_quota_ip_alloc_sem_key,
+ ocfs2_file_ip_alloc_sem_key;
mlog_entry("inode = %p, opaque = %p\n", inode, opaque);
inode->i_ino = args->fi_ino;
OCFS2_I(inode)->ip_blkno = args->fi_blkno;
+ if (args->fi_sysfile_type != 0)
+ lockdep_set_class(&inode->i_mutex,
+ &ocfs2_sysfile_lock_key[args->fi_sysfile_type]);
+ if (args->fi_sysfile_type == USER_QUOTA_SYSTEM_INODE ||
+ args->fi_sysfile_type == GROUP_QUOTA_SYSTEM_INODE ||
+ args->fi_sysfile_type == LOCAL_USER_QUOTA_SYSTEM_INODE ||
+ args->fi_sysfile_type == LOCAL_GROUP_QUOTA_SYSTEM_INODE)
+ lockdep_set_class(&OCFS2_I(inode)->ip_alloc_sem,
+ &ocfs2_quota_ip_alloc_sem_key);
+ else
+ lockdep_set_class(&OCFS2_I(inode)->ip_alloc_sem,
+ &ocfs2_file_ip_alloc_sem_key);
mlog_exit(0);
return 0;
}
-int ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe,
- int create_ino)
+void ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe,
+ int create_ino)
{
struct super_block *sb;
struct ocfs2_super *osb;
- int status = -EINVAL;
+ int use_plocks = 1;
mlog_entry("(0x%p, size:%llu)\n", inode,
- (unsigned long long)fe->i_size);
+ (unsigned long long)le64_to_cpu(fe->i_size));
sb = inode->i_sb;
osb = OCFS2_SB(sb);
- /* this means that read_inode cannot create a superblock inode
- * today. change if needed. */
- if (!OCFS2_IS_VALID_DINODE(fe) ||
- !(fe->i_flags & cpu_to_le32(OCFS2_VALID_FL))) {
- mlog(0, "Invalid dinode: i_ino=%lu, i_blkno=%llu, "
- "signature = %.*s, flags = 0x%x\n",
- inode->i_ino,
- (unsigned long long)le64_to_cpu(fe->i_blkno), 7,
- fe->i_signature, le32_to_cpu(fe->i_flags));
- goto bail;
- }
+ if ((osb->s_mount_opt & OCFS2_MOUNT_LOCALFLOCKS) ||
+ ocfs2_mount_local(osb) || !ocfs2_stack_supports_plocks())
+ use_plocks = 0;
+
+ /*
+ * These have all been checked by ocfs2_read_inode_block() or set
+ * by ocfs2_mknod_locked(), so a failure is a code bug.
+ */
+ BUG_ON(!OCFS2_IS_VALID_DINODE(fe)); /* This means that read_inode
+ cannot create a superblock
+ inode today. change if
+ that is needed. */
+ BUG_ON(!(fe->i_flags & cpu_to_le32(OCFS2_VALID_FL)));
+ BUG_ON(le32_to_cpu(fe->i_fs_generation) != osb->fs_generation);
- if (le32_to_cpu(fe->i_fs_generation) != osb->fs_generation) {
- mlog(ML_ERROR, "file entry generation does not match "
- "superblock! osb->fs_generation=%x, "
- "fe->i_fs_generation=%x\n",
- osb->fs_generation, le32_to_cpu(fe->i_fs_generation));
- goto bail;
- }
+
+ OCFS2_I(inode)->ip_clusters = le32_to_cpu(fe->i_clusters);
+ OCFS2_I(inode)->ip_attr = le32_to_cpu(fe->i_attr);
+ OCFS2_I(inode)->ip_dyn_features = le16_to_cpu(fe->i_dyn_features);
inode->i_version = 1;
inode->i_generation = le32_to_cpu(fe->i_generation);
if (S_ISLNK(inode->i_mode) && !fe->i_clusters)
inode->i_blocks = 0;
else
- inode->i_blocks =
- ocfs2_align_bytes_to_sectors(le64_to_cpu(fe->i_size));
+ inode->i_blocks = ocfs2_inode_sector_count(inode);
inode->i_mapping->a_ops = &ocfs2_aops;
inode->i_atime.tv_sec = le64_to_cpu(fe->i_atime);
inode->i_atime.tv_nsec = le32_to_cpu(fe->i_atime_nsec);
mlog(ML_ERROR,
"ip_blkno %llu != i_blkno %llu!\n",
(unsigned long long)OCFS2_I(inode)->ip_blkno,
- (unsigned long long)fe->i_blkno);
+ (unsigned long long)le64_to_cpu(fe->i_blkno));
- OCFS2_I(inode)->ip_clusters = le32_to_cpu(fe->i_clusters);
- OCFS2_I(inode)->ip_attr = le32_to_cpu(fe->i_attr);
+ inode->i_nlink = ocfs2_read_links_count(fe);
- inode->i_nlink = le16_to_cpu(fe->i_links_count);
-
- if (fe->i_flags & cpu_to_le32(OCFS2_SYSTEM_FL))
+ if (fe->i_flags & cpu_to_le32(OCFS2_SYSTEM_FL)) {
OCFS2_I(inode)->ip_flags |= OCFS2_INODE_SYSTEM_FILE;
+ inode->i_flags |= S_NOQUOTA;
+ }
if (fe->i_flags & cpu_to_le32(OCFS2_LOCAL_ALLOC_FL)) {
OCFS2_I(inode)->ip_flags |= OCFS2_INODE_BITMAP;
mlog(0, "local alloc inode: i_ino=%lu\n", inode->i_ino);
} else if (fe->i_flags & cpu_to_le32(OCFS2_BITMAP_FL)) {
OCFS2_I(inode)->ip_flags |= OCFS2_INODE_BITMAP;
+ } else if (fe->i_flags & cpu_to_le32(OCFS2_QUOTA_FL)) {
+ inode->i_flags |= S_NOQUOTA;
} else if (fe->i_flags & cpu_to_le32(OCFS2_SUPER_BLOCK_FL)) {
mlog(0, "superblock inode: i_ino=%lu\n", inode->i_ino);
/* we can't actually hit this as read_inode can't
switch (inode->i_mode & S_IFMT) {
case S_IFREG:
- inode->i_fop = &ocfs2_fops;
+ if (use_plocks)
+ inode->i_fop = &ocfs2_fops;
+ else
+ inode->i_fop = &ocfs2_fops_no_plocks;
inode->i_op = &ocfs2_file_iops;
i_size_write(inode, le64_to_cpu(fe->i_size));
break;
case S_IFDIR:
inode->i_op = &ocfs2_dir_iops;
- inode->i_fop = &ocfs2_dops;
+ if (use_plocks)
+ inode->i_fop = &ocfs2_dops;
+ else
+ inode->i_fop = &ocfs2_dops_no_plocks;
i_size_write(inode, le64_to_cpu(fe->i_size));
break;
case S_IFLNK:
* the generation argument to
* ocfs2_inode_lock_res_init() will have to change.
*/
- BUG_ON(fe->i_flags & cpu_to_le32(OCFS2_SYSTEM_FL));
+ BUG_ON(le32_to_cpu(fe->i_flags) & OCFS2_SYSTEM_FL);
- ocfs2_inode_lock_res_init(&OCFS2_I(inode)->ip_meta_lockres,
+ ocfs2_inode_lock_res_init(&OCFS2_I(inode)->ip_inode_lockres,
OCFS2_LOCK_TYPE_META, 0, inode);
ocfs2_inode_lock_res_init(&OCFS2_I(inode)->ip_open_lockres,
OCFS2_LOCK_TYPE_RW, inode->i_generation,
inode);
- ocfs2_inode_lock_res_init(&OCFS2_I(inode)->ip_data_lockres,
- OCFS2_LOCK_TYPE_DATA, inode->i_generation,
- inode);
-
ocfs2_set_inode_flags(inode);
- status = 0;
-bail:
- mlog_exit(status);
- return status;
+ OCFS2_I(inode)->ip_last_used_slot = 0;
+ OCFS2_I(inode)->ip_last_used_group = 0;
+ mlog_exit_void();
}
static int ocfs2_read_locked_inode(struct inode *inode,
if (args->fi_flags & OCFS2_FI_FLAG_SYSFILE)
generation = osb->fs_generation;
- ocfs2_inode_lock_res_init(&OCFS2_I(inode)->ip_meta_lockres,
+ ocfs2_inode_lock_res_init(&OCFS2_I(inode)->ip_inode_lockres,
OCFS2_LOCK_TYPE_META,
generation, inode);
mlog_errno(status);
return status;
}
- status = ocfs2_meta_lock(inode, NULL, 0);
+ status = ocfs2_inode_lock(inode, NULL, 0);
if (status) {
make_bad_inode(inode);
mlog_errno(status);
if (args->fi_flags & OCFS2_FI_FLAG_ORPHAN_RECOVERY) {
status = ocfs2_try_open_lock(inode, 0);
if (status) {
- make_bad_inode(inode);
+ make_bad_inode(inode);
return status;
}
}
- status = ocfs2_read_block(osb, args->fi_blkno, &bh, 0,
- can_lock ? inode : NULL);
+ if (can_lock) {
+ status = ocfs2_read_inode_block_full(inode, &bh,
+ OCFS2_BH_IGNORE_CACHE);
+ } else {
+ status = ocfs2_read_blocks_sync(osb, args->fi_blkno, 1, &bh);
+ if (!status)
+ status = ocfs2_validate_inode_block(osb->sb, bh);
+ }
if (status < 0) {
mlog_errno(status);
goto bail;
status = -EINVAL;
fe = (struct ocfs2_dinode *) bh->b_data;
- if (!OCFS2_IS_VALID_DINODE(fe)) {
- mlog(ML_ERROR, "Invalid dinode #%llu: signature = %.*s\n",
- (unsigned long long)fe->i_blkno, 7, fe->i_signature);
- goto bail;
- }
/*
* This is a code bug. Right now the caller needs to
if (S_ISCHR(le16_to_cpu(fe->i_mode)) ||
S_ISBLK(le16_to_cpu(fe->i_mode)))
- inode->i_rdev = huge_decode_dev(le64_to_cpu(fe->id1.dev1.i_rdev));
+ inode->i_rdev = huge_decode_dev(le64_to_cpu(fe->id1.dev1.i_rdev));
- if (ocfs2_populate_inode(inode, fe, 0) < 0)
- goto bail;
+ ocfs2_populate_inode(inode, fe, 0);
BUG_ON(args->fi_blkno != le64_to_cpu(fe->i_blkno));
bail:
if (can_lock)
- ocfs2_meta_unlock(inode, 0);
+ ocfs2_inode_unlock(inode, 0);
if (status < 0)
make_bad_inode(inode);
fe = (struct ocfs2_dinode *) fe_bh->b_data;
+ /*
+ * This check will also skip truncate of inodes with inline
+ * data and fast symlinks.
+ */
if (fe->i_clusters) {
+ if (ocfs2_should_order_data(inode))
+ ocfs2_begin_ordered_truncate(inode, 0);
+
handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
goto out;
}
- status = ocfs2_journal_access(handle, inode, fe_bh,
- OCFS2_JOURNAL_ACCESS_WRITE);
+ status = ocfs2_journal_access_di(handle, INODE_CACHE(inode),
+ fe_bh,
+ OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto out;
}
mutex_lock(&inode_alloc_inode->i_mutex);
- status = ocfs2_meta_lock(inode_alloc_inode, &inode_alloc_bh, 1);
+ status = ocfs2_inode_lock(inode_alloc_inode, &inode_alloc_bh, 1);
if (status < 0) {
mutex_unlock(&inode_alloc_inode->i_mutex);
goto bail;
}
- handle = ocfs2_start_trans(osb, OCFS2_DELETE_INODE_CREDITS);
+ handle = ocfs2_start_trans(osb, OCFS2_DELETE_INODE_CREDITS +
+ ocfs2_quota_trans_credits(inode->i_sb));
if (IS_ERR(handle)) {
status = PTR_ERR(handle);
mlog_errno(status);
}
/* set the inodes dtime */
- status = ocfs2_journal_access(handle, inode, di_bh,
- OCFS2_JOURNAL_ACCESS_WRITE);
+ status = ocfs2_journal_access_di(handle, INODE_CACHE(inode), di_bh,
+ OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto bail_commit;
}
di->i_dtime = cpu_to_le64(CURRENT_TIME.tv_sec);
- le32_and_cpu(&di->i_flags, ~(OCFS2_VALID_FL | OCFS2_ORPHANED_FL));
+ di->i_flags &= cpu_to_le32(~(OCFS2_VALID_FL | OCFS2_ORPHANED_FL));
status = ocfs2_journal_dirty(handle, di_bh);
if (status < 0) {
goto bail_commit;
}
- ocfs2_remove_from_cache(inode, di_bh);
+ ocfs2_remove_from_cache(INODE_CACHE(inode), di_bh);
+ dquot_free_inode(inode);
status = ocfs2_free_dinode(handle, inode_alloc_inode,
inode_alloc_bh, di);
bail_commit:
ocfs2_commit_trans(osb, handle);
bail_unlock:
- ocfs2_meta_unlock(inode_alloc_inode, 1);
+ ocfs2_inode_unlock(inode_alloc_inode, 1);
mutex_unlock(&inode_alloc_inode->i_mutex);
brelse(inode_alloc_bh);
bail:
return status;
}
-/*
+/*
* Serialize with orphan dir recovery. If the process doing
* recovery on this orphan dir does an iget() with the dir
* i_mutex held, we'll deadlock here. Instead we detect this
* delete_inode operation. We do this now to avoid races with
* recovery completion on other nodes. */
mutex_lock(&orphan_dir_inode->i_mutex);
- status = ocfs2_meta_lock(orphan_dir_inode, &orphan_dir_bh, 1);
+ status = ocfs2_inode_lock(orphan_dir_inode, &orphan_dir_bh, 1);
if (status < 0) {
mutex_unlock(&orphan_dir_inode->i_mutex);
}
/* we do this while holding the orphan dir lock because we
- * don't want recovery being run from another node to vote for
- * an inode delete on us -- this will result in two nodes
+ * don't want recovery being run from another node to try an
+ * inode delete underneath us -- this will result in two nodes
* truncating the same file! */
status = ocfs2_truncate_for_delete(osb, inode, di_bh);
if (status < 0) {
goto bail_unlock_dir;
}
+ /* Remove any dir index tree */
+ if (S_ISDIR(inode->i_mode)) {
+ status = ocfs2_dx_dir_truncate(inode, di_bh);
+ if (status) {
+ mlog_errno(status);
+ goto bail_unlock_dir;
+ }
+ }
+
+ /*Free extended attribute resources associated with this inode.*/
+ status = ocfs2_xattr_remove(inode, di_bh);
+ if (status < 0) {
+ mlog_errno(status);
+ goto bail_unlock_dir;
+ }
+
+ status = ocfs2_remove_refcount_tree(inode, di_bh);
+ if (status < 0) {
+ mlog_errno(status);
+ goto bail_unlock_dir;
+ }
+
status = ocfs2_remove_inode(inode, di_bh, orphan_dir_inode,
orphan_dir_bh);
if (status < 0)
mlog_errno(status);
bail_unlock_dir:
- ocfs2_meta_unlock(orphan_dir_inode, 1);
+ ocfs2_inode_unlock(orphan_dir_inode, 1);
mutex_unlock(&orphan_dir_inode->i_mutex);
brelse(orphan_dir_bh);
bail:
}
/* There is a series of simple checks that should be done before a
- * vote is even considered. Encapsulate those in this function. */
+ * trylock is even considered. Encapsulate those in this function. */
static int ocfs2_inode_is_valid_to_delete(struct inode *inode)
{
int ret = 0;
goto bail;
}
- /* If we're coming from process_vote we can't go into our own
+ /* If we're coming from downconvert_thread we can't go into our own
* voting [hello, deadlock city!], so unforuntately we just
* have to skip deleting this guy. That's OK though because
* the node who's doing the actual deleting should handle it
* anyway. */
- if (current == osb->vote_task) {
+ if (current == osb->dc_task) {
mlog(0, "Skipping delete of %lu because we're currently "
- "in process_vote\n", inode->i_ino);
+ "in downconvert\n", inode->i_ino);
goto bail;
}
goto bail_unlock;
}
- /* If we have voted "yes" on the wipe of this inode for
- * another node, it will be marked here so we can safely skip
- * it. Recovery will cleanup any inodes we might inadvertantly
- * skip here. */
+ /* If we have allowd wipe of this inode for another node, it
+ * will be marked here so we can safely skip it. Recovery will
+ * cleanup any inodes we might inadvertantly skip here. */
if (oi->ip_flags & OCFS2_INODE_SKIP_DELETE) {
mlog(0, "Skipping delete of %lu because another node "
"has done this for us.\n", inode->i_ino);
"Inode %llu (on-disk %llu) not orphaned! "
"Disk flags 0x%x, inode flags 0x%x\n",
(unsigned long long)oi->ip_blkno,
- (unsigned long long)di->i_blkno, di->i_flags,
- oi->ip_flags);
+ (unsigned long long)le64_to_cpu(di->i_blkno),
+ le32_to_cpu(di->i_flags), oi->ip_flags);
goto bail;
}
status = ocfs2_try_open_lock(inode, 1);
if (status == -EAGAIN) {
status = 0;
- mlog(0, "Skipping delete of %llu because it is in use on"
+ mlog(0, "Skipping delete of %llu because it is in use on "
"other nodes\n", (unsigned long long)oi->ip_blkno);
goto bail;
}
mlog_entry("(inode->i_ino = %lu)\n", inode->i_ino);
- if (is_bad_inode(inode)) {
+ /* When we fail in read_inode() we mark inode as bad. The second test
+ * catches the case when inode allocation fails before allocating
+ * a block for inode. */
+ if (is_bad_inode(inode) || !OCFS2_I(inode)->ip_blkno) {
mlog(0, "Skipping delete of bad inode\n");
goto bail;
}
goto bail;
}
+ /*
+ * Synchronize us against ocfs2_get_dentry. We take this in
+ * shared mode so that all nodes can still concurrently
+ * process deletes.
+ */
+ status = ocfs2_nfs_sync_lock(OCFS2_SB(inode->i_sb), 0);
+ if (status < 0) {
+ mlog(ML_ERROR, "getting nfs sync lock(PR) failed %d\n", status);
+ ocfs2_cleanup_delete_inode(inode, 0);
+ goto bail_unblock;
+ }
/* Lock down the inode. This gives us an up to date view of
* it's metadata (for verification), and allows us to
- * serialize delete_inode votes.
+ * serialize delete_inode on multiple nodes.
*
* Even though we might be doing a truncate, we don't take the
* allocation lock here as it won't be needed - nobody will
* have the file open.
*/
- status = ocfs2_meta_lock(inode, &di_bh, 1);
+ status = ocfs2_inode_lock(inode, &di_bh, 1);
if (status < 0) {
if (status != -ENOENT)
mlog_errno(status);
ocfs2_cleanup_delete_inode(inode, 0);
- goto bail_unblock;
+ goto bail_unlock_nfs_sync;
}
/* Query the cluster. This will be the final decision made
* before we go ahead and wipe the inode. */
status = ocfs2_query_inode_wipe(inode, di_bh, &wipe);
if (!wipe || status < 0) {
- /* Error and inode busy vote both mean we won't be
+ /* Error and remote inode busy both mean we won't be
* removing the inode, so they take almost the same
* path. */
if (status < 0)
mlog_errno(status);
- /* Someone in the cluster has voted to not wipe this
- * inode, or it was never completely orphaned. Write
- * out the pages and exit now. */
+ /* Someone in the cluster has disallowed a wipe of
+ * this inode, or it was never completely
+ * orphaned. Write out the pages and exit now. */
ocfs2_cleanup_delete_inode(inode, 1);
goto bail_unlock_inode;
}
OCFS2_I(inode)->ip_flags |= OCFS2_INODE_DELETED;
bail_unlock_inode:
- ocfs2_meta_unlock(inode, 1);
+ ocfs2_inode_unlock(inode, 1);
brelse(di_bh);
+
+bail_unlock_nfs_sync:
+ ocfs2_nfs_sync_unlock(OCFS2_SB(inode->i_sb), 0);
+
bail_unblock:
status = sigprocmask(SIG_SETMASK, &oldset, NULL);
if (status < 0)
mlog_bug_on_msg(OCFS2_SB(inode->i_sb) == NULL,
"Inode=%lu\n", inode->i_ino);
- /* For remove delete_inode vote, we hold open lock before,
- * now it is time to unlock PR and EX open locks. */
+ dquot_drop(inode);
+
+ /* To preven remote deletes we hold open lock before, now it
+ * is time to unlock PR and EX open locks. */
ocfs2_open_unlock(inode);
/* Do these before all the other work so that we don't bounce
- * the vote thread while waiting to destroy the locks. */
+ * the downconvert thread while waiting to destroy the locks. */
ocfs2_mark_lockres_freeing(&oi->ip_rw_lockres);
- ocfs2_mark_lockres_freeing(&oi->ip_meta_lockres);
- ocfs2_mark_lockres_freeing(&oi->ip_data_lockres);
+ ocfs2_mark_lockres_freeing(&oi->ip_inode_lockres);
ocfs2_mark_lockres_freeing(&oi->ip_open_lockres);
/* We very well may get a clear_inode before all an inodes
"Clear inode of %llu, inode has io markers\n",
(unsigned long long)oi->ip_blkno);
+ ocfs2_extent_map_trunc(inode, 0);
+
status = ocfs2_drop_inode_locks(inode);
if (status < 0)
mlog_errno(status);
ocfs2_lock_res_free(&oi->ip_rw_lockres);
- ocfs2_lock_res_free(&oi->ip_meta_lockres);
- ocfs2_lock_res_free(&oi->ip_data_lockres);
+ ocfs2_lock_res_free(&oi->ip_inode_lockres);
ocfs2_lock_res_free(&oi->ip_open_lockres);
- ocfs2_metadata_cache_purge(inode);
+ ocfs2_metadata_cache_exit(INODE_CACHE(inode));
- mlog_bug_on_msg(oi->ip_metadata_cache.ci_num_cached,
+ mlog_bug_on_msg(INODE_CACHE(inode)->ci_num_cached,
"Clear inode of %llu, inode has %u cache items\n",
- (unsigned long long)oi->ip_blkno, oi->ip_metadata_cache.ci_num_cached);
+ (unsigned long long)oi->ip_blkno,
+ INODE_CACHE(inode)->ci_num_cached);
- mlog_bug_on_msg(!(oi->ip_flags & OCFS2_INODE_CACHE_INLINE),
+ mlog_bug_on_msg(!(INODE_CACHE(inode)->ci_flags & OCFS2_CACHE_FL_INLINE),
"Clear inode of %llu, inode has a bad flag\n",
(unsigned long long)oi->ip_blkno);
(unsigned long long)oi->ip_blkno, oi->ip_open_count);
/* Clear all other flags. */
- oi->ip_flags = OCFS2_INODE_CACHE_INLINE;
- oi->ip_created_trans = 0;
- oi->ip_last_trans = 0;
+ oi->ip_flags = 0;
oi->ip_dir_start_lookup = 0;
oi->ip_blkno = 0ULL;
+ /*
+ * ip_jinode is used to track txns against this inode. We ensure that
+ * the journal is flushed before journal shutdown. Thus it is safe to
+ * have inodes get cleaned up after journal shutdown.
+ */
+ jbd2_journal_release_jbd_inode(OCFS2_SB(inode->i_sb)->journal->j_journal,
+ &oi->ip_jinode);
+
bail:
mlog_exit_void();
}
}
/*
- * TODO: this should probably be merged into ocfs2_get_block
- *
- * However, you now need to pay attention to the cont_prepare_write()
- * stuff in ocfs2_get_block (that is, ocfs2_get_block pretty much
- * expects never to extend).
- */
-struct buffer_head *ocfs2_bread(struct inode *inode,
- int block, int *err, int reada)
-{
- struct buffer_head *bh = NULL;
- int tmperr;
- u64 p_blkno;
- int readflags = OCFS2_BH_CACHED;
-
- if (reada)
- readflags |= OCFS2_BH_READAHEAD;
-
- if (((u64)block << inode->i_sb->s_blocksize_bits) >=
- i_size_read(inode)) {
- BUG_ON(!reada);
- return NULL;
- }
-
- tmperr = ocfs2_extent_map_get_blocks(inode, block, &p_blkno, NULL,
- NULL);
- if (tmperr < 0) {
- mlog_errno(tmperr);
- goto fail;
- }
-
- tmperr = ocfs2_read_block(OCFS2_SB(inode->i_sb), p_blkno, &bh,
- readflags, inode);
- if (tmperr < 0)
- goto fail;
-
- tmperr = 0;
-
- *err = 0;
- return bh;
-
-fail:
- if (bh) {
- brelse(bh);
- bh = NULL;
- }
- *err = -EIO;
- return NULL;
-}
-
-/*
* This is called from our getattr.
*/
int ocfs2_inode_revalidate(struct dentry *dentry)
}
spin_unlock(&OCFS2_I(inode)->ip_lock);
- /* Let ocfs2_meta_lock do the work of updating our struct
+ /* Let ocfs2_inode_lock do the work of updating our struct
* inode for us. */
- status = ocfs2_meta_lock(inode, NULL, 0);
+ status = ocfs2_inode_lock(inode, NULL, 0);
if (status < 0) {
if (status != -ENOENT)
mlog_errno(status);
goto bail;
}
- ocfs2_meta_unlock(inode, 0);
+ ocfs2_inode_unlock(inode, 0);
bail:
mlog_exit(status);
mlog_entry("(inode %llu)\n",
(unsigned long long)OCFS2_I(inode)->ip_blkno);
- status = ocfs2_journal_access(handle, inode, bh,
- OCFS2_JOURNAL_ACCESS_WRITE);
+ status = ocfs2_journal_access_di(handle, INODE_CACHE(inode), bh,
+ OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto leave;
spin_lock(&OCFS2_I(inode)->ip_lock);
fe->i_clusters = cpu_to_le32(OCFS2_I(inode)->ip_clusters);
+ ocfs2_get_inode_flags(OCFS2_I(inode));
fe->i_attr = cpu_to_le32(OCFS2_I(inode)->ip_attr);
+ fe->i_dyn_features = cpu_to_le16(OCFS2_I(inode)->ip_dyn_features);
spin_unlock(&OCFS2_I(inode)->ip_lock);
fe->i_size = cpu_to_le64(i_size_read(inode));
- fe->i_links_count = cpu_to_le16(inode->i_nlink);
+ ocfs2_set_links_count(fe, inode->i_nlink);
fe->i_uid = cpu_to_le32(inode->i_uid);
fe->i_gid = cpu_to_le32(inode->i_gid);
fe->i_mode = cpu_to_le16(inode->i_mode);
OCFS2_I(inode)->ip_clusters = le32_to_cpu(fe->i_clusters);
OCFS2_I(inode)->ip_attr = le32_to_cpu(fe->i_attr);
+ OCFS2_I(inode)->ip_dyn_features = le16_to_cpu(fe->i_dyn_features);
ocfs2_set_inode_flags(inode);
i_size_write(inode, le64_to_cpu(fe->i_size));
- inode->i_nlink = le16_to_cpu(fe->i_links_count);
+ inode->i_nlink = ocfs2_read_links_count(fe);
inode->i_uid = le32_to_cpu(fe->i_uid);
inode->i_gid = le32_to_cpu(fe->i_gid);
inode->i_mode = le16_to_cpu(fe->i_mode);
if (S_ISLNK(inode->i_mode) && le32_to_cpu(fe->i_clusters) == 0)
inode->i_blocks = 0;
else
- inode->i_blocks = ocfs2_align_bytes_to_sectors(i_size_read(inode));
+ inode->i_blocks = ocfs2_inode_sector_count(inode);
inode->i_atime.tv_sec = le64_to_cpu(fe->i_atime);
inode->i_atime.tv_nsec = le32_to_cpu(fe->i_atime_nsec);
inode->i_mtime.tv_sec = le64_to_cpu(fe->i_mtime);
spin_unlock(&OCFS2_I(inode)->ip_lock);
}
+
+int ocfs2_validate_inode_block(struct super_block *sb,
+ struct buffer_head *bh)
+{
+ int rc;
+ struct ocfs2_dinode *di = (struct ocfs2_dinode *)bh->b_data;
+
+ mlog(0, "Validating dinode %llu\n",
+ (unsigned long long)bh->b_blocknr);
+
+ BUG_ON(!buffer_uptodate(bh));
+
+ /*
+ * If the ecc fails, we return the error but otherwise
+ * leave the filesystem running. We know any error is
+ * local to this block.
+ */
+ rc = ocfs2_validate_meta_ecc(sb, bh->b_data, &di->i_check);
+ if (rc) {
+ mlog(ML_ERROR, "Checksum failed for dinode %llu\n",
+ (unsigned long long)bh->b_blocknr);
+ goto bail;
+ }
+
+ /*
+ * Errors after here are fatal.
+ */
+
+ rc = -EINVAL;
+
+ if (!OCFS2_IS_VALID_DINODE(di)) {
+ ocfs2_error(sb, "Invalid dinode #%llu: signature = %.*s\n",
+ (unsigned long long)bh->b_blocknr, 7,
+ di->i_signature);
+ goto bail;
+ }
+
+ if (le64_to_cpu(di->i_blkno) != bh->b_blocknr) {
+ ocfs2_error(sb, "Invalid dinode #%llu: i_blkno is %llu\n",
+ (unsigned long long)bh->b_blocknr,
+ (unsigned long long)le64_to_cpu(di->i_blkno));
+ goto bail;
+ }
+
+ if (!(di->i_flags & cpu_to_le32(OCFS2_VALID_FL))) {
+ ocfs2_error(sb,
+ "Invalid dinode #%llu: OCFS2_VALID_FL not set\n",
+ (unsigned long long)bh->b_blocknr);
+ goto bail;
+ }
+
+ if (le32_to_cpu(di->i_fs_generation) !=
+ OCFS2_SB(sb)->fs_generation) {
+ ocfs2_error(sb,
+ "Invalid dinode #%llu: fs_generation is %u\n",
+ (unsigned long long)bh->b_blocknr,
+ le32_to_cpu(di->i_fs_generation));
+ goto bail;
+ }
+
+ rc = 0;
+
+bail:
+ return rc;
+}
+
+int ocfs2_read_inode_block_full(struct inode *inode, struct buffer_head **bh,
+ int flags)
+{
+ int rc;
+ struct buffer_head *tmp = *bh;
+
+ rc = ocfs2_read_blocks(INODE_CACHE(inode), OCFS2_I(inode)->ip_blkno,
+ 1, &tmp, flags, ocfs2_validate_inode_block);
+
+ /* If ocfs2_read_blocks() got us a new bh, pass it up. */
+ if (!rc && !*bh)
+ *bh = tmp;
+
+ return rc;
+}
+
+int ocfs2_read_inode_block(struct inode *inode, struct buffer_head **bh)
+{
+ return ocfs2_read_inode_block_full(inode, bh, 0);
+}
+
+
+static u64 ocfs2_inode_cache_owner(struct ocfs2_caching_info *ci)
+{
+ struct ocfs2_inode_info *oi = cache_info_to_inode(ci);
+
+ return oi->ip_blkno;
+}
+
+static struct super_block *ocfs2_inode_cache_get_super(struct ocfs2_caching_info *ci)
+{
+ struct ocfs2_inode_info *oi = cache_info_to_inode(ci);
+
+ return oi->vfs_inode.i_sb;
+}
+
+static void ocfs2_inode_cache_lock(struct ocfs2_caching_info *ci)
+{
+ struct ocfs2_inode_info *oi = cache_info_to_inode(ci);
+
+ spin_lock(&oi->ip_lock);
+}
+
+static void ocfs2_inode_cache_unlock(struct ocfs2_caching_info *ci)
+{
+ struct ocfs2_inode_info *oi = cache_info_to_inode(ci);
+
+ spin_unlock(&oi->ip_lock);
+}
+
+static void ocfs2_inode_cache_io_lock(struct ocfs2_caching_info *ci)
+{
+ struct ocfs2_inode_info *oi = cache_info_to_inode(ci);
+
+ mutex_lock(&oi->ip_io_mutex);
+}
+
+static void ocfs2_inode_cache_io_unlock(struct ocfs2_caching_info *ci)
+{
+ struct ocfs2_inode_info *oi = cache_info_to_inode(ci);
+
+ mutex_unlock(&oi->ip_io_mutex);
+}
+
+const struct ocfs2_caching_operations ocfs2_inode_caching_ops = {
+ .co_owner = ocfs2_inode_cache_owner,
+ .co_get_super = ocfs2_inode_cache_get_super,
+ .co_cache_lock = ocfs2_inode_cache_lock,
+ .co_cache_unlock = ocfs2_inode_cache_unlock,
+ .co_io_lock = ocfs2_inode_cache_io_lock,
+ .co_io_unlock = ocfs2_inode_cache_io_unlock,
+};
+
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff