On Mon, Oct 16, 2023 at 02:32:25PM +0200, Andrey Albershteyn wrote: > >From the discussion in v2 [1] I understood that btrfs would like to > have this bio_set/submit_io interface for iomap transition; For btrfs it would not be a transitional thing, but forever as it has it's own equivalent of a device mapper at this level. But now that the fsverity interfaces work so that we don't have other file system dependencies I would not want to design around it. If/when btrfs migrates to iomap for the buffer read path we can easily hook this into the existing code. > and any > other filesystem deferrals would also be possible. Is it no more the > case with btrfs? Would fs-verity verification in iomap_read_end_io > combine both solutions (fs-verity verification in iomap + > submit_io/bio_set interface). btrfs does also need to do I/O completion from a workqueue, but it needs it's own. fsverity OTOH is a pretty generic feature.
