On Mon, Apr 4, 2022 at 12:51 PM Christoph Hellwig <hch@xxxxxx> wrote: > > I really can't get excited about the pdu thingy. Here is a patch > (on top of the series and the patch sent in reply to patch 4) that > does away with it and just adds a oob_user field to struct io_uring_cmd > to simplify the handling a fair bit: Fine. We discussed the tradeoff before, and the concern was generic-space reduction for other use-cases. But this is kernel-only change, so we can alter this if/when a new usecase requires more space. It seems fine to move ahead with what you proposed.
