[RFC 0/5] big-cqe based uring-passthru

[Kanchan Joshi]

So this is a trimmed-down series primarily to test the
waters with 32-byte CQE (patch 4).
This big-cqe is created by combining two adjacent 16b CQEs.
And that is done when ring is setup with IORING_SETUP_CQE32 flag.

nvme-passthru sends one result to bigcqe->res and another result (8 bytes)
gets updated to bigcqe->res2.
As always, fio is modified to test the interface:
https://github.com/joshkan/fio/tree/big-cqe

Limited testing ATM, as plumbing itself is in early stage with patch 4.
Patches are against for-next (linux-block),
on top of 9e9d83faa ("io_uring: Remove unneeded test in io_run_task_work_sig")

Jens Axboe (3):
  io_uring: add support for 128-byte SQEs
  fs: add file_operations->async_cmd()
  io_uring: add infra and support for IORING_OP_URING_CMD

Kanchan Joshi (2):
  io_uring: add support for big-cqe
  nvme: wire-up support for async-passthru on char-device.

 drivers/nvme/host/core.c      |   1 +
 drivers/nvme/host/ioctl.c     | 187 ++++++++++++++++++++++++++++------
 drivers/nvme/host/multipath.c |   1 +
 drivers/nvme/host/nvme.h      |   3 +
 fs/io_uring.c                 | 168 ++++++++++++++++++++++++++----
 include/linux/fs.h            |   2 +
 include/linux/io_uring.h      |  33 ++++++
 include/uapi/linux/io_uring.h |  26 ++++-
 8 files changed, 369 insertions(+), 52 deletions(-)

-- 
2.25.1