Re: [PATCH v2] btrfs: report filemap_fdata<write|wait>_range error

[Qu Wenruo <quwenruo.btrfs@gmx.com>]

在 2024/4/18 15:44, Anand Jain 写道:
> In the function btrfs_write_marked_extents() and in __btrfs_wait_marked_extents()
> return the actual error if when filemap_fdata<write|wait>_range() fails.
>
> Suggested-by: Josef Bacik <josef@toxicpanda.com>
> Signed-off-by: Anand Jain <anand.jain@oracle.com>

Looks fine for the patch, although I have a small question.

If we failed to write some metadata extents, we break out, meaning there
would be dirty metadata still hanging there.

Would it be a problem?

Thanks,
Qu
> ---
> v2: add __btrfs_wait_marked_extents() as well.
>
>   fs/btrfs/transaction.c | 4 ++++
>   1 file changed, 4 insertions(+)
>
> diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
> index 3e3bcc5f64c6..8c3b3cda1390 100644
> --- a/fs/btrfs/transaction.c
> +++ b/fs/btrfs/transaction.c
> @@ -1156,6 +1156,8 @@ int btrfs_write_marked_extents(struct btrfs_fs_info *fs_info,
>   		else if (wait_writeback)
>   			werr = filemap_fdatawait_range(mapping, start, end);
>   		free_extent_state(cached_state);
> +		if (werr)
> +			break;
>   		cached_state = NULL;
>   		cond_resched();
>   		start = end + 1;
> @@ -1198,6 +1200,8 @@ static int __btrfs_wait_marked_extents(struct btrfs_fs_info *fs_info,
>   		if (err)
>   			werr = err;
>   		free_extent_state(cached_state);
> +		if (werr)
> +			break;
>   		cached_state = NULL;
>   		cond_resched();
>   		start = end + 1;