Re: [PATCH] block: fix disordered IO in the case recursive split

[Christoph Hellwig]

On Mon, Aug 25, 2025 at 02:15:49PM +0800, Yu Kuai wrote:
> > be resubmitted to current->bio_list, hence this patch won't break this
> > case, right?
> 
> And xfs_rw_bdev() is not under submit_bio() context, current->bio_list
> is still NULL, means xfs_rw_bdev() is submitting bio one by one in the
> right lba order, the bio recursive handling is not related in this case.

Yes, usually not - unless we somehow eventually get the loop device
out a separate workque context.