Re: [PATCH V2] Btrfs: flush all dirty inodes if writeback can not start

[Josef Bacik <jbacik@fusionio.com>]

On Mon, Dec 24, 2012 at 02:57:19AM -0700, Miao Xie wrote:
> We may try to flush some dirty pages when there is no enough space to reserve.
> But it is possible that this operation fails, in order to get enough space to
> reserve successfully, we will sync all the delalloc file. This operation is
> safe, we needn't worry about the case that the filesystem goes from r/w to r/o.
> because the filesystem should guarantee all the dirty pages have been written
> into the disk after it becomes readonly, so the sync operation will do nothing
> if the filesystem is already readonly. Though it may waste lots of time,
> as a corner case, we needn't care.
> 
> Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
> ---
> Changelog v1 -> v2:
> - make the function static
> ---
>  fs/btrfs/extent-tree.c | 40 +++++++++++++++++++++++++++++++---------
>  1 file changed, 31 insertions(+), 9 deletions(-)
> 
> diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
> index b6ed965..2d9fe27 100644
> --- a/fs/btrfs/extent-tree.c
> +++ b/fs/btrfs/extent-tree.c
> @@ -3695,12 +3695,15 @@ static int can_overcommit(struct btrfs_root *root,
>  	return 0;
>  }
>  
> -static int writeback_inodes_sb_nr_if_idle_safe(struct super_block *sb,
> -					       unsigned long nr_pages,
> -					       enum wb_reason reason)
> +static inline int writeback_inodes_sb_nr_if_idle_safe(struct super_block *sb,
> +						      unsigned long nr_pages,
> +						      enum wb_reason reason)
>  {
> -	if (!writeback_in_progress(sb->s_bdi) &&
> -	    down_read_trylock(&sb->s_umount)) {
> +	/* the flusher is dealing with the dirty inodes now. */
> +	if (writeback_in_progress(sb->s_bdi))
> +		return 1;
> +
> +	if (down_read_trylock(&sb->s_umount)) {
>  		writeback_inodes_sb_nr(sb, nr_pages, reason);
>  		up_read(&sb->s_umount);
>  		return 1;
> @@ -3709,6 +3712,28 @@ static int writeback_inodes_sb_nr_if_idle_safe(struct super_block *sb,
>  	return 0;
>  }
>  
> +static void btrfs_writeback_inodes_sb_nr(struct btrfs_root *root,
> +					 unsigned long nr_pages)
> +{
> +	struct super_block *sb = root->fs_info->sb;
> +	int started;
> +
> +	/* If we can not start writeback, just sync all the delalloc file. */
> +	started = writeback_inodes_sb_nr_if_idle_safe(sb, nr_pages,
> +						      WB_REASON_FS_FREE_SPACE);
> +	if (!started) {
> +		/*
> +		 * We needn't worry the filesystem going from r/w to r/o though
> +		 * we don't acquire ->s_umount mutex, because the filesystem
> +		 * should guarantee the delalloc inodes list be empty after
> +		 * the filesystem is readonly(all dirty pages are written to
> +		 * the disk).
> +		 */
> +		btrfs_start_delalloc_inodes(root, 0);
> +		btrfs_wait_ordered_extents(root, 0);

We can't just call wait_ordered_extents, we may have an open trans handle which
could make us deadlock if a transaction commit starts.  Thanks,

Josef