On Tue, 2019-01-29 at 09:35 +-0100, Christoph Hellwig wrote: +AD4 I disagree with investing further effort into BIDI support. It is +AD4 dead for all practical purposes in standards and implementation, +AD4 and the fact that we found all these bugs in it just further confirms +AD4 that. The only answer to that is to bite the bullet and remove it. Hi Christoph, It would help if you could explain why you are so strongly opposed against keeping BIDI support in the kernel. If the reason is that removing BIDI support would allow to improve NVMe processing performance in the kernel by a small margin, that is not good enough as a reason. Users who care about that level of performance will access the NVMe queues directly from user space anyway to avoid the system call overhead. Bart.
