On Mon, Jan 11, 2021 at 10:17:53AM +0000, Johannes Thumshirn wrote:

On 22/12/2020 14:40, Christoph Hellwig wrote:

quoted

I just did a very quick look, but didn't see anything (based on the
subjects) that deals with ITER_BVEC direct I/O.  Is the support for that
hidden somewhere?

I couldn't reproduce the problem you reported and asked for a reproducer. Probably
the mail got lost somewhere.

Do you have a reproducer for me that triggers the problem? fio with --ioengine=io_uring
didn't do the job on my end.

I've only found this by auditing the code.  But the reproducer should
be io_uring with fixed buffers, alternatively using the nvme target
driver with a btrfs file as the backend.