[Ocfs2-devel] [PATCH 4/7 v4] ocfs2: add and remove inode in orphan dir in ocfs2_direct_IO

[Andrew Morton <akpm@linux-foundation.org>]

On Sat, 11 Oct 2014 20:29:08 +0800 WeiWei Wang <wangww631@huawei.com> wrote:

> Add the inode to orphan dir first, and then delete it once append
> O_DIRECT finished.
> This is to make sure block allocation and inode size are consistent.
> 
> ...
>
> +static ssize_t ocfs2_direct_IO_write(struct kiocb *iocb,
> +		struct iov_iter *iter,
> +		loff_t offset)
> +{
> +	ssize_t ret = 0;
> +	int orphan = 0;
> +	int is_overwrite = 0;
> +	struct file *file = iocb->ki_filp;
> +	struct inode *inode = file->f_path.dentry->d_inode->i_mapping->host;
> +	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
> +	struct buffer_head *di_bh = NULL;
> +	size_t count = iter->count;
> +	journal_t *journal = osb->journal->j_journal;
> +	u32 p_cpos = 0;
> +	u32 v_cpos = ocfs2_clusters_for_bytes(osb->sb, offset);
> +	u32 zero_len = offset % (1 << osb->s_clustersize_bits);

On i386 this generates a call to _moddi3, which doesn't exist.  You'll
need to use do_div() or something similar.

> +	int cluster_align = offset % (1 << osb->s_clustersize_bits) ? 0 : 1;

Similar here.  Why not just do cluster_align = !!zero_len?

> +	int append_write = offset >= i_size_read(inode) ? 1 : 0;
> +	unsigned int num_clusters = 0;
> +	unsigned int ext_flags = 0;
>
> ...
>