5 files changed, 136 insertions, 106 deletions

diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c
index 87eccbbca255..f522425aaa24 100644
--- a/fs/ext3/dir.c
+++ b/fs/ext3/dir.c
@@ -28,8 +28,7 @@ static unsigned char ext3_filetype_table[] = {
         DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK
 };
 
-static int ext3_dx_readdir(struct file * filp,
-                           void * dirent, filldir_t filldir);
+static int ext3_dx_readdir(struct file *, struct dir_context *);
 
 static unsigned char get_dtype(struct super_block *sb, int filetype)
 {
@@ -91,36 +90,30 @@ int ext3_check_dir_entry (const char * function, struct inode * dir,
         return error_msg == NULL ? 1 : 0;
 }
 
-static int ext3_readdir(struct file * filp,
-                         void * dirent, filldir_t filldir)
+static int ext3_readdir(struct file *file, struct dir_context *ctx)
 {
-        int error = 0;
         unsigned long offset;
-        int i, stored;
+        int i;
         struct ext3_dir_entry_2 *de;
         int err;
-        struct inode *inode = file_inode(filp);
+        struct inode *inode = file_inode(file);
         struct super_block *sb = inode->i_sb;
-        int ret = 0;
         int dir_has_error = 0;
 
         if (is_dx_dir(inode)) {
-                err = ext3_dx_readdir(filp, dirent, filldir);
-                if (err != ERR_BAD_DX_DIR) {
-                        ret = err;
-                        goto out;
-                }
+                err = ext3_dx_readdir(file, ctx);
+                if (err != ERR_BAD_DX_DIR)
+                        return err;
                 /*
                  * We don't set the inode dirty flag since it's not
                  * critical that it get flushed back to the disk.
                  */
-                EXT3_I(file_inode(filp))->i_flags &= ~EXT3_INDEX_FL;
+                EXT3_I(inode)->i_flags &= ~EXT3_INDEX_FL;
         }
-        stored = 0;
-        offset = filp->f_pos & (sb->s_blocksize - 1);
+        offset = ctx->pos & (sb->s_blocksize - 1);
 
-        while (!error && !stored && filp->f_pos < inode->i_size) {
-                unsigned long blk = filp->f_pos >> EXT3_BLOCK_SIZE_BITS(sb);
+        while (ctx->pos < inode->i_size) {
+                unsigned long blk = ctx->pos >> EXT3_BLOCK_SIZE_BITS(sb);
                 struct buffer_head map_bh;
                 struct buffer_head *bh = NULL;
 
@@ -129,12 +122,12 @@ static int ext3_readdir(struct file * filp,
                 if (err > 0) {
                         pgoff_t index = map_bh.b_blocknr >>
                                         (PAGE_CACHE_SHIFT - inode->i_blkbits);
-                        if (!ra_has_index(&filp->f_ra, index))
+                        if (!ra_has_index(&file->f_ra, index))
                                 page_cache_sync_readahead(
                                         sb->s_bdev->bd_inode->i_mapping,
-                                        &filp->f_ra, filp,
+                                        &file->f_ra, file,
                                         index, 1);
-                        filp->f_ra.prev_pos = (loff_t)index << PAGE_CACHE_SHIFT;
+                        file->f_ra.prev_pos = (loff_t)index << PAGE_CACHE_SHIFT;
                         bh = ext3_bread(NULL, inode, blk, 0, &err);
                 }
 
@@ -146,22 +139,21 @@ static int ext3_readdir(struct file * filp,
                         if (!dir_has_error) {
                                 ext3_error(sb, __func__, "directory #%lu "
                                         "contains a hole at offset %lld",
-                                        inode->i_ino, filp->f_pos);
+                                        inode->i_ino, ctx->pos);
                                 dir_has_error = 1;
                         }
                         /* corrupt size?  Maybe no more blocks to read */
-                        if (filp->f_pos > inode->i_blocks << 9)
+                        if (ctx->pos > inode->i_blocks << 9)
                                 break;
-                        filp->f_pos += sb->s_blocksize - offset;
+                        ctx->pos += sb->s_blocksize - offset;
                         continue;
                 }
 
-revalidate:
                 /* If the dir block has changed since the last call to
                  * readdir(2), then we might be pointing to an invalid
                  * dirent right now.  Scan from the start of the block
                  * to make sure. */
-                if (filp->f_version != inode->i_version) {
+                if (offset && file->f_version != inode->i_version) {
                         for (i = 0; i < sb->s_blocksize && i < offset; ) {
                                 de = (struct ext3_dir_entry_2 *)
                                         (bh->b_data + i);
@@ -177,53 +169,40 @@ revalidate:
                                 i += ext3_rec_len_from_disk(de->rec_len);
                         }
                         offset = i;
-                        filp->f_pos = (filp->f_pos & ~(sb->s_blocksize - 1))
+                        ctx->pos = (ctx->pos & ~(sb->s_blocksize - 1))
                                 | offset;
-                        filp->f_version = inode->i_version;
+                        file->f_version = inode->i_version;
                 }
 
-                while (!error && filp->f_pos < inode->i_size
+                while (ctx->pos < inode->i_size
                        && offset < sb->s_blocksize) {
                         de = (struct ext3_dir_entry_2 *) (bh->b_data + offset);
                         if (!ext3_check_dir_entry ("ext3_readdir", inode, de,
                                                    bh, offset)) {
-                                /* On error, skip the f_pos to the
+                                /* On error, skip the to the
                                    next block. */
-                                filp->f_pos = (filp->f_pos |
+                                ctx->pos = (ctx->pos |
                                                 (sb->s_blocksize - 1)) + 1;
-                                brelse (bh);
-                                ret = stored;
-                                goto out;
+                                break;
                         }
                         offset += ext3_rec_len_from_disk(de->rec_len);
                         if (le32_to_cpu(de->inode)) {
-                                /* We might block in the next section
-                                 * if the data destination is
-                                 * currently swapped out.  So, use a
-                                 * version stamp to detect whether or
-                                 * not the directory has been modified
-                                 * during the copy operation.
-                                 */
-                                u64 version = filp->f_version;
-
-                                error = filldir(dirent, de->name,
-                                                de->name_len,
-                                                filp->f_pos,
-                                                le32_to_cpu(de->inode),
-                                                get_dtype(sb, de->file_type));
-                                if (error)
-                                        break;
-                                if (version != filp->f_version)
-                                        goto revalidate;
-                                stored ++;
+                                if (!dir_emit(ctx, de->name, de->name_len,
+                                              le32_to_cpu(de->inode),
+                                              get_dtype(sb, de->file_type))) {
+                                        brelse(bh);
+                                        return 0;
+                                }
                         }
-                        filp->f_pos += ext3_rec_len_from_disk(de->rec_len);
+                        ctx->pos += ext3_rec_len_from_disk(de->rec_len);
                 }
                 offset = 0;
                 brelse (bh);
+                if (ctx->pos < inode->i_size)
+                        if (!dir_relax(inode))
+                                return 0;
         }
-out:
-        return ret;
+        return 0;
 }
 
 static inline int is_32bit_api(void)
@@ -452,62 +431,54 @@ int ext3_htree_store_dirent(struct file *dir_file, __u32 hash,
  * for all entres on the fname linked list.  (Normally there is only
  * one entry on the linked list, unless there are 62 bit hash collisions.)
  */
-static int call_filldir(struct file * filp, void * dirent,
-                        filldir_t filldir, struct fname *fname)
+static bool call_filldir(struct file *file, struct dir_context *ctx,
+                        struct fname *fname)
 {
-        struct dir_private_info *info = filp->private_data;
-        loff_t  curr_pos;
-        struct inode *inode = file_inode(filp);
-        struct super_block * sb;
-        int error;
-
-        sb = inode->i_sb;
+        struct dir_private_info *info = file->private_data;
+        struct inode *inode = file_inode(file);
+        struct super_block *sb = inode->i_sb;
 
         if (!fname) {
                 printk("call_filldir: called with null fname?!?\n");
-                return 0;
+                return true;
         }
-        curr_pos = hash2pos(filp, fname->hash, fname->minor_hash);
+        ctx->pos = hash2pos(file, fname->hash, fname->minor_hash);
         while (fname) {
-                error = filldir(dirent, fname->name,
-                                fname->name_len, curr_pos,
+                if (!dir_emit(ctx, fname->name, fname->name_len,
                                 fname->inode,
-                                get_dtype(sb, fname->file_type));
-                if (error) {
-                        filp->f_pos = curr_pos;
+                                get_dtype(sb, fname->file_type))) {
                         info->extra_fname = fname;
-                        return error;
+                        return false;
                 }
                 fname = fname->next;
         }
-        return 0;
+        return true;
 }
 
-static int ext3_dx_readdir(struct file * filp,
-                         void * dirent, filldir_t filldir)
+static int ext3_dx_readdir(struct file *file, struct dir_context *ctx)
 {
-        struct dir_private_info *info = filp->private_data;
-        struct inode *inode = file_inode(filp);
+        struct dir_private_info *info = file->private_data;
+        struct inode *inode = file_inode(file);
         struct fname *fname;
         int     ret;
 
         if (!info) {
-                info = ext3_htree_create_dir_info(filp, filp->f_pos);
+                info = ext3_htree_create_dir_info(file, ctx->pos);
                 if (!info)
                         return -ENOMEM;
-                filp->private_data = info;
+                file->private_data = info;
         }
 
-        if (filp->f_pos == ext3_get_htree_eof(filp))
+        if (ctx->pos == ext3_get_htree_eof(file))
                 return 0;       /* EOF */
 
         /* Some one has messed with f_pos; reset the world */
-        if (info->last_pos != filp->f_pos) {
+        if (info->last_pos != ctx->pos) {
                 free_rb_tree_fname(&info->root);
                 info->curr_node = NULL;
                 info->extra_fname = NULL;
-                info->curr_hash = pos2maj_hash(filp, filp->f_pos);
-                info->curr_minor_hash = pos2min_hash(filp, filp->f_pos);
+                info->curr_hash = pos2maj_hash(file, ctx->pos);
+                info->curr_minor_hash = pos2min_hash(file, ctx->pos);
         }
 
         /*
@@ -515,7 +486,7 @@ static int ext3_dx_readdir(struct file * filp,
          * chain, return them first.
          */
         if (info->extra_fname) {
-                if (call_filldir(filp, dirent, filldir, info->extra_fname))
+                if (!call_filldir(file, ctx, info->extra_fname))
                         goto finished;
                 info->extra_fname = NULL;
                 goto next_node;
@@ -529,17 +500,17 @@ static int ext3_dx_readdir(struct file * filp,
                  * cached entries.
                  */
                 if ((!info->curr_node) ||
-                    (filp->f_version != inode->i_version)) {
+                    (file->f_version != inode->i_version)) {
                         info->curr_node = NULL;
                         free_rb_tree_fname(&info->root);
-                        filp->f_version = inode->i_version;
-                        ret = ext3_htree_fill_tree(filp, info->curr_hash,
+                        file->f_version = inode->i_version;
+                        ret = ext3_htree_fill_tree(file, info->curr_hash,
                                                    info->curr_minor_hash,
                                                    &info->next_hash);
                         if (ret < 0)
                                 return ret;
                         if (ret == 0) {
-                                filp->f_pos = ext3_get_htree_eof(filp);
+                                ctx->pos = ext3_get_htree_eof(file);
                                 break;
                         }
                         info->curr_node = rb_first(&info->root);
@@ -548,7 +519,7 @@ static int ext3_dx_readdir(struct file * filp,
                 fname = rb_entry(info->curr_node, struct fname, rb_hash);
                 info->curr_hash = fname->hash;
                 info->curr_minor_hash = fname->minor_hash;
-                if (call_filldir(filp, dirent, filldir, fname))
+                if (!call_filldir(file, ctx, fname))
                         break;
         next_node:
                 info->curr_node = rb_next(info->curr_node);
@@ -559,7 +530,7 @@ static int ext3_dx_readdir(struct file * filp,
                         info->curr_minor_hash = fname->minor_hash;
                 } else {
                         if (info->next_hash == ~0) {
-                                filp->f_pos = ext3_get_htree_eof(filp);
+                                ctx->pos = ext3_get_htree_eof(file);
                                 break;
                         }
                         info->curr_hash = info->next_hash;
@@ -567,7 +538,7 @@ static int ext3_dx_readdir(struct file * filp,
                 }
         }
 finished:
-        info->last_pos = filp->f_pos;
+        info->last_pos = ctx->pos;
         return 0;
 }
 
@@ -582,7 +553,7 @@ static int ext3_release_dir (struct inode * inode, struct file * filp)
 const struct file_operations ext3_dir_operations = {
         .llseek         = ext3_dir_llseek,
         .read           = generic_read_dir,
-        .readdir        = ext3_readdir,
+        .iterate        = ext3_readdir,
         .unlocked_ioctl = ext3_ioctl,
 #ifdef CONFIG_COMPAT
         .compat_ioctl   = ext3_compat_ioctl,
diff --git a/fs/ext3/fsync.c b/fs/ext3/fsync.c
index b31dbd4c46ad..1cb9c7e10c6f 100644
--- a/fs/ext3/fsync.c
+++ b/fs/ext3/fsync.c
@@ -48,9 +48,13 @@ int ext3_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 
         trace_ext3_sync_file_enter(file, datasync);
 
-        if (inode->i_sb->s_flags & MS_RDONLY)
+        if (inode->i_sb->s_flags & MS_RDONLY) {
+                /* Make sure that we read updated state */
+                smp_rmb();
+                if (EXT3_SB(inode->i_sb)->s_mount_state & EXT3_ERROR_FS)
+                        return -EROFS;
                 return 0;
-
+        }
         ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
         if (ret)
                 goto out;
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 23c712825640..2bd85486b879 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -1825,19 +1825,20 @@ ext3_readpages(struct file *file, struct address_space *mapping,
         return mpage_readpages(mapping, pages, nr_pages, ext3_get_block);
 }
 
-static void ext3_invalidatepage(struct page *page, unsigned long offset)
+static void ext3_invalidatepage(struct page *page, unsigned int offset,
+                                unsigned int length)
 {
         journal_t *journal = EXT3_JOURNAL(page->mapping->host);
 
-        trace_ext3_invalidatepage(page, offset);
+        trace_ext3_invalidatepage(page, offset, length);
 
         /*
          * If it's a full truncate we just forget about the pending dirtying
          */
-        if (offset == 0)
+        if (offset == 0 && length == PAGE_CACHE_SIZE)
                 ClearPageChecked(page);
 
-        journal_invalidatepage(journal, page, offset);
+        journal_invalidatepage(journal, page, offset, length);
 }
 
 static int ext3_releasepage(struct page *page, gfp_t wait)
@@ -1984,6 +1985,7 @@ static const struct address_space_operations ext3_ordered_aops = {
         .direct_IO              = ext3_direct_IO,
         .migratepage            = buffer_migrate_page,
         .is_partially_uptodate  = block_is_partially_uptodate,
+        .is_dirty_writeback     = buffer_check_dirty_writeback,
         .error_remove_page      = generic_error_remove_page,
 };
 
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 692de13e3596..1194b1f0f839 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -576,11 +576,8 @@ static int htree_dirblock_to_tree(struct file *dir_file,
                 if (!ext3_check_dir_entry("htree_dirblock_to_tree", dir, de, bh,
                                         (block<<EXT3_BLOCK_SIZE_BITS(dir->i_sb))
                                                 +((char *)de - bh->b_data))) {
-                        /* On error, skip the f_pos to the next block. */
-                        dir_file->f_pos = (dir_file->f_pos |
-                                        (dir->i_sb->s_blocksize - 1)) + 1;
-                        brelse (bh);
-                        return count;
+                        /* silently ignore the rest of the block */
+                        break;
                 }
                 ext3fs_dirhash(de->name, de->name_len, hinfo);
                 if ((hinfo->hash < start_hash) ||
@@ -1762,6 +1759,45 @@ retry:
         return err;
 }
 
+static int ext3_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode)
+{
+        handle_t *handle;
+        struct inode *inode;
+        int err, retries = 0;
+
+        dquot_initialize(dir);
+
+retry:
+        handle = ext3_journal_start(dir, EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb) +
+                          4 + EXT3_XATTR_TRANS_BLOCKS);
+
+        if (IS_ERR(handle))
+                return PTR_ERR(handle);
+
+        inode = ext3_new_inode (handle, dir, NULL, mode);
+        err = PTR_ERR(inode);
+        if (!IS_ERR(inode)) {
+                inode->i_op = &ext3_file_inode_operations;
+                inode->i_fop = &ext3_file_operations;
+                ext3_set_aops(inode);
+                d_tmpfile(dentry, inode);
+                err = ext3_orphan_add(handle, inode);
+                if (err)
+                        goto err_drop_inode;
+                mark_inode_dirty(inode);
+                unlock_new_inode(inode);
+        }
+        ext3_journal_stop(handle);
+        if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries))
+                goto retry;
+        return err;
+err_drop_inode:
+        ext3_journal_stop(handle);
+        unlock_new_inode(inode);
+        iput(inode);
+        return err;
+}
+
 static int ext3_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode)
 {
         handle_t *handle;
@@ -2303,7 +2339,7 @@ static int ext3_link (struct dentry * old_dentry,
 
 retry:
         handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
-                                        EXT3_INDEX_EXTRA_TRANS_BLOCKS);
+                                        EXT3_INDEX_EXTRA_TRANS_BLOCKS + 1);
         if (IS_ERR(handle))
                 return PTR_ERR(handle);
 
@@ -2317,6 +2353,11 @@ retry:
         err = ext3_add_entry(handle, dentry, inode);
         if (!err) {
                 ext3_mark_inode_dirty(handle, inode);
+                /* this can happen only for tmpfile being
+                 * linked the first time
+                 */
+                if (inode->i_nlink == 1)
+                        ext3_orphan_del(handle, inode);
                 d_instantiate(dentry, inode);
         } else {
                 drop_nlink(inode);
@@ -2519,6 +2560,7 @@ const struct inode_operations ext3_dir_inode_operations = {
         .mkdir          = ext3_mkdir,
         .rmdir          = ext3_rmdir,
         .mknod          = ext3_mknod,
+        .tmpfile        = ext3_tmpfile,
         .rename         = ext3_rename,
         .setattr        = ext3_setattr,
 #ifdef CONFIG_EXT3_FS_XATTR
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 6356665a74bb..c47f14750722 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -174,6 +174,11 @@ static void ext3_handle_error(struct super_block *sb)
         if (test_opt (sb, ERRORS_RO)) {
                 ext3_msg(sb, KERN_CRIT,
                         "error: remounting filesystem read-only");
+                /*
+                 * Make sure updated value of ->s_mount_state will be visible
+                 * before ->s_flags update.
+                 */
+                smp_wmb();
                 sb->s_flags |= MS_RDONLY;
         }
         ext3_commit_super(sb, es, 1);
@@ -291,8 +296,14 @@ void ext3_abort(struct super_block *sb, const char *function,
         ext3_msg(sb, KERN_CRIT,
                 "error: remounting filesystem read-only");
         EXT3_SB(sb)->s_mount_state |= EXT3_ERROR_FS;
-        sb->s_flags |= MS_RDONLY;
         set_opt(EXT3_SB(sb)->s_mount_opt, ABORT);
+        /*
+         * Make sure updated value of ->s_mount_state will be visible
+         * before ->s_flags update.
+         */
+        smp_wmb();
+        sb->s_flags |= MS_RDONLY;
+
         if (EXT3_SB(sb)->s_journal)
                 journal_abort(EXT3_SB(sb)->s_journal, -EIO);
 }