Re: [PATCH] btrfs: remove code duplication in ordered extent finishing

[Filipe Manana]

On Thu, Sep 26, 2024 at 1:27 PM Johannes Thumshirn <jth@xxxxxxxxxx> wrote:
>
> From: Johannes Thumshirn <johannes.thumshirn@xxxxxxx>
>
> Remove the duplicated transaction joining, block reserve setting and raid
> extent inserting in btrfs_finish_ordered_extent().
>
> While at it, also abort the transaction in case inserting a RAID
> stripe-tree entry fails.
>
> Suggested-by: Naohiro Aota <naohiro.aota@xxxxxxx>
> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@xxxxxxx>
> ---
>  fs/btrfs/inode.c | 45 ++++++++++++++++-----------------------------
>  1 file changed, 16 insertions(+), 29 deletions(-)
>
> diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
> index 353fb58c83da..35a03a671fc6 100644
> --- a/fs/btrfs/inode.c
> +++ b/fs/btrfs/inode.c
> @@ -3068,34 +3068,6 @@ int btrfs_finish_one_ordered(struct btrfs_ordered_extent *ordered_extent)
>                         goto out;
>         }
>
> -       if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags)) {
> -               BUG_ON(!list_empty(&ordered_extent->list)); /* Logic error */
> -
> -               btrfs_inode_safe_disk_i_size_write(inode, 0);
> -               if (freespace_inode)
> -                       trans = btrfs_join_transaction_spacecache(root);
> -               else
> -                       trans = btrfs_join_transaction(root);
> -               if (IS_ERR(trans)) {
> -                       ret = PTR_ERR(trans);
> -                       trans = NULL;
> -                       goto out;
> -               }
> -               trans->block_rsv = &inode->block_rsv;
> -               ret = btrfs_update_inode_fallback(trans, inode);
> -               if (ret) /* -ENOMEM or corruption */
> -                       btrfs_abort_transaction(trans, ret);
> -
> -               ret = btrfs_insert_raid_extent(trans, ordered_extent);
> -               if (ret)
> -                       btrfs_abort_transaction(trans, ret);
> -
> -               goto out;
> -       }
> -
> -       clear_bits |= EXTENT_LOCKED;
> -       lock_extent(io_tree, start, end, &cached_state);
> -
>         if (freespace_inode)
>                 trans = btrfs_join_transaction_spacecache(root);
>         else
> @@ -3109,8 +3081,23 @@ int btrfs_finish_one_ordered(struct btrfs_ordered_extent *ordered_extent)
>         trans->block_rsv = &inode->block_rsv;
>
>         ret = btrfs_insert_raid_extent(trans, ordered_extent);
> -       if (ret)
> +       if (ret) {
> +               btrfs_abort_transaction(trans, ret);
> +               goto out;
> +       }
> +
> +       if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags)) {
> +               BUG_ON(!list_empty(&ordered_extent->list)); /* Logic error */
> +
> +               btrfs_inode_safe_disk_i_size_write(inode, 0);
> +               ret = btrfs_update_inode_fallback(trans, inode);
> +               if (ret) /* -ENOMEM or corruption */

While at it we can change the comment to comply with the preferred
style, to leave it on the line above or inside the if:

if (ret) {
   /* -ENOMEM or corruption */
  btrfs_abort_transaction(trans, ret);
}

Same for the other comment after the BUG_ON() (and we could change the
BUG_ON() to ASSERT() plus abort transaction maybe, but that probably
as a separate patch).

Otherwise it looks fine, thanks.

Reviewed-by: Filipe Manana <fdmanana@xxxxxxxx>


> +                       btrfs_abort_transaction(trans, ret);
>                 goto out;
> +       }
> +
> +       clear_bits |= EXTENT_LOCKED;
> +       lock_extent(io_tree, start, end, &cached_state);
>
>         if (test_bit(BTRFS_ORDERED_COMPRESSED, &ordered_extent->flags))
>                 compress_type = ordered_extent->compress_type;
> --
> 2.43.0
>
>