#include "xattr.h"
#include "acl.h"
#include "quota.h"
+ #include "refcounttree.h"
#include "buffer_head_io.h"
goto out;
}
- ret = ocfs2_journal_access_di(handle, inode, bh,
+ ret = ocfs2_journal_access_di(handle, INODE_CACHE(inode), bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (ret) {
mlog_errno(ret);
return ret;
}
+ static int ocfs2_cow_file_pos(struct inode *inode,
+ struct buffer_head *fe_bh,
+ u64 offset)
+ {
+ int status;
+ u32 phys, cpos = offset >> OCFS2_SB(inode->i_sb)->s_clustersize_bits;
+ unsigned int num_clusters = 0;
+ unsigned int ext_flags = 0;
+
+ /*
+ * If the new offset is aligned to the range of the cluster, there is
+ * no space for ocfs2_zero_range_for_truncate to fill, so no need to
+ * CoW either.
+ */
+ if ((offset & (OCFS2_SB(inode->i_sb)->s_clustersize - 1)) == 0)
+ return 0;
+
+ status = ocfs2_get_clusters(inode, cpos, &phys,
+ &num_clusters, &ext_flags);
+ if (status) {
+ mlog_errno(status);
+ goto out;
+ }
+
+ if (!(ext_flags & OCFS2_EXT_REFCOUNTED))
+ goto out;
+
+ return ocfs2_refcount_cow(inode, fe_bh, cpos, 1, cpos+1);
+
+ out:
+ return status;
+ }
+
static int ocfs2_orphan_for_truncate(struct ocfs2_super *osb,
struct inode *inode,
struct buffer_head *fe_bh,
mlog_entry_void();
+ /*
+ * We need to CoW the cluster contains the offset if it is reflinked
+ * since we will call ocfs2_zero_range_for_truncate later which will
+ * write "0" from offset to the end of the cluster.
+ */
+ status = ocfs2_cow_file_pos(inode, fe_bh, new_i_size);
+ if (status) {
+ mlog_errno(status);
+ return status;
+ }
+
/* TODO: This needs to actually orphan the inode in this
* transaction. */
goto out;
}
- status = ocfs2_journal_access_di(handle, inode, fe_bh,
+ status = ocfs2_journal_access_di(handle, INODE_CACHE(inode), fe_bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
up_write(&OCFS2_I(inode)->ip_alloc_sem);
bail:
+ if (!status && OCFS2_I(inode)->ip_clusters == 0)
+ status = ocfs2_try_remove_refcount_tree(inode, di_bh);
mlog_exit(status);
return status;
int ret;
struct ocfs2_extent_tree et;
- ocfs2_init_dinode_extent_tree(&et, inode, fe_bh);
- ret = ocfs2_add_clusters_in_btree(osb, inode, logical_offset,
- clusters_to_add, mark_unwritten,
- &et, handle,
- data_ac, meta_ac, reason_ret);
+ ocfs2_init_dinode_extent_tree(&et, INODE_CACHE(inode), fe_bh);
+ ret = ocfs2_add_clusters_in_btree(handle, &et, logical_offset,
+ clusters_to_add, mark_unwritten,
+ data_ac, meta_ac, reason_ret);
return ret;
}
(unsigned long long)OCFS2_I(inode)->ip_blkno,
(long long)i_size_read(inode), le32_to_cpu(fe->i_clusters),
clusters_to_add);
- ocfs2_init_dinode_extent_tree(&et, inode, bh);
+ ocfs2_init_dinode_extent_tree(&et, INODE_CACHE(inode), bh);
status = ocfs2_lock_allocators(inode, &et, clusters_to_add, 0,
&data_ac, &meta_ac);
if (status) {
/* reserve a write to the file entry early on - that we if we
* run out of credits in the allocation path, we can still
* update i_size. */
- status = ocfs2_journal_access_di(handle, inode, bh,
+ status = ocfs2_journal_access_di(handle, INODE_CACHE(inode), bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (status < 0) {
mlog_errno(status);
goto out;
}
- ret = ocfs2_journal_access_di(handle, inode, bh,
+ ret = ocfs2_journal_access_di(handle, INODE_CACHE(inode), bh,
OCFS2_JOURNAL_ACCESS_WRITE);
if (ret < 0) {
mlog_errno(ret);
struct address_space *mapping = inode->i_mapping;
struct ocfs2_extent_tree et;
- ocfs2_init_dinode_extent_tree(&et, inode, di_bh);
+ ocfs2_init_dinode_extent_tree(&et, INODE_CACHE(inode), di_bh);
ocfs2_init_dealloc_ctxt(&dealloc);
if (byte_len == 0)
OCFS2_IOC_RESVSP64, &sr, change_size);
}
+ int ocfs2_check_range_for_refcount(struct inode *inode, loff_t pos,
+ size_t count)
+ {
+ int ret = 0;
+ unsigned int extent_flags;
+ u32 cpos, clusters, extent_len, phys_cpos;
+ struct super_block *sb = inode->i_sb;
+
+ if (!ocfs2_refcount_tree(OCFS2_SB(inode->i_sb)) ||
+ !(OCFS2_I(inode)->ip_dyn_features & OCFS2_HAS_REFCOUNT_FL))
+ return 0;
+
+ cpos = pos >> OCFS2_SB(sb)->s_clustersize_bits;
+ clusters = ocfs2_clusters_for_bytes(sb, pos + count) - cpos;
+
+ while (clusters) {
+ ret = ocfs2_get_clusters(inode, cpos, &phys_cpos, &extent_len,
+ &extent_flags);
+ if (ret < 0) {
+ mlog_errno(ret);
+ goto out;
+ }
+
+ if (phys_cpos && (extent_flags & OCFS2_EXT_REFCOUNTED)) {
+ ret = 1;
+ break;
+ }
+
+ if (extent_len > clusters)
+ extent_len = clusters;
+
+ clusters -= extent_len;
+ cpos += extent_len;
+ }
+ out:
+ return ret;
+ }
+
+ static int ocfs2_prepare_inode_for_refcount(struct inode *inode,
+ loff_t pos, size_t count,
+ int *meta_level)
+ {
+ int ret;
+ struct buffer_head *di_bh = NULL;
+ u32 cpos = pos >> OCFS2_SB(inode->i_sb)->s_clustersize_bits;
+ u32 clusters =
+ ocfs2_clusters_for_bytes(inode->i_sb, pos + count) - cpos;
+
+ ret = ocfs2_inode_lock(inode, &di_bh, 1);
+ if (ret) {
+ mlog_errno(ret);
+ goto out;
+ }
+
+ *meta_level = 1;
+
+ ret = ocfs2_refcount_cow(inode, di_bh, cpos, clusters, UINT_MAX);
+ if (ret)
+ mlog_errno(ret);
+ out:
+ brelse(di_bh);
+ return ret;
+ }
+
static int ocfs2_prepare_inode_for_write(struct dentry *dentry,
loff_t *ppos,
size_t count,
end = saved_pos + count;
+ ret = ocfs2_check_range_for_refcount(inode, saved_pos, count);
+ if (ret == 1) {
+ ocfs2_inode_unlock(inode, meta_level);
+ meta_level = -1;
+
+ ret = ocfs2_prepare_inode_for_refcount(inode,
+ saved_pos,
+ count,
+ &meta_level);
+ }
+
+ if (ret < 0) {
+ mlog_errno(ret);
+ goto out_unlock;
+ }
+
/*
* Skip the O_DIRECT checks if we don't need
* them.
*ppos = saved_pos;
out_unlock:
- ocfs2_inode_unlock(inode, meta_level);
+ if (meta_level >= 0)
+ ocfs2_inode_unlock(inode, meta_level);
out:
return ret;
goto out_dio;
}
} else {
- written = generic_file_aio_write_nolock(iocb, iov, nr_segs,
- *ppos);
+ written = __generic_file_aio_write(iocb, iov, nr_segs, ppos);
}
out_dio:
BUG_ON(ret == -EIOCBQUEUED && !(file->f_flags & O_DIRECT));
if ((file->f_flags & O_SYNC && !direct_io) || IS_SYNC(inode)) {
- /*
- * The generic write paths have handled getting data
- * to disk, but since we don't make use of the dirty
- * inode list, a manual journal commit is necessary
- * here.
- */
- if (old_size != i_size_read(inode) ||
- old_clusters != OCFS2_I(inode)->ip_clusters) {
+ ret = filemap_fdatawrite_range(file->f_mapping, pos,
+ pos + count - 1);
+ if (ret < 0)
+ written = ret;
+
+ if (!ret && (old_size != i_size_read(inode) ||
+ old_clusters != OCFS2_I(inode)->ip_clusters)) {
ret = jbd2_journal_force_commit(osb->journal->j_journal);
if (ret < 0)
written = ret;
}
+
+ if (!ret)
+ ret = filemap_fdatawait_range(file->f_mapping, pos,
+ pos + count - 1);
}
/*
if (ret > 0) {
unsigned long nr_pages;
+ int err;
- *ppos += ret;
nr_pages = (ret + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
- /*
- * If file or inode is SYNC and we actually wrote some data,
- * sync it.
- */
- if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(inode))) {
- int err;
-
- mutex_lock(&inode->i_mutex);
- err = ocfs2_rw_lock(inode, 1);
- if (err < 0) {
- mlog_errno(err);
- } else {
- err = generic_osync_inode(inode, mapping,
- OSYNC_METADATA|OSYNC_DATA);
- ocfs2_rw_unlock(inode, 1);
- }
- mutex_unlock(&inode->i_mutex);
+ err = generic_write_sync(out, *ppos, ret);
+ if (err)
+ ret = err;
+ else
+ *ppos += ret;
- if (err)
- ret = err;
- }
balance_dirty_pages_ratelimited_nr(mapping, nr_pages);
}
#include "ver.h"
#include "xattr.h"
#include "quota.h"
+ #include "refcounttree.h"
#include "buffer_head_io.h"
return vfs_quota_disable(sb, type, DQUOT_LIMITS_ENABLED);
}
-static struct quotactl_ops ocfs2_quotactl_ops = {
+static const struct quotactl_ops ocfs2_quotactl_ops = {
.quota_on = ocfs2_quota_on,
.quota_off = ocfs2_quota_off,
.quota_sync = vfs_quota_sync,
spin_lock_init(&oi->ip_lock);
ocfs2_extent_map_init(&oi->vfs_inode);
INIT_LIST_HEAD(&oi->ip_io_markers);
- oi->ip_created_trans = 0;
- oi->ip_last_trans = 0;
oi->ip_dir_start_lookup = 0;
init_rwsem(&oi->ip_alloc_sem);
ocfs2_lock_res_init_once(&oi->ip_inode_lockres);
ocfs2_lock_res_init_once(&oi->ip_open_lockres);
- ocfs2_metadata_cache_init(&oi->vfs_inode);
+ ocfs2_metadata_cache_init(INODE_CACHE(&oi->vfs_inode),
+ &ocfs2_inode_caching_ops);
inode_init_once(&oi->vfs_inode);
}
ocfs2_sync_blockdev(sb);
+ ocfs2_purge_refcount_trees(osb);
+
/* No cluster connection means we've failed during mount, so skip
* all the steps which depended on that to complete. */
if (osb->cconn) {
goto bail;
}
+ osb->osb_rf_lock_tree = RB_ROOT;
+
osb->s_feature_compat =
le32_to_cpu(OCFS2_RAW_SB(di)->s_feature_compat);
osb->s_feature_ro_compat =
/* Force a panic(). This stinks, but it's better than letting
* things continue without having a proper hard readonly
* here. */
- OCFS2_SB(sb)->s_mount_opt |= OCFS2_MOUNT_ERRORS_PANIC;
+ if (!ocfs2_mount_local(OCFS2_SB(sb)))
+ OCFS2_SB(sb)->s_mount_opt |= OCFS2_MOUNT_ERRORS_PANIC;
ocfs2_handle_error(sb);
}