Series enables async polling on io_uring command, and nvme passthrough (for io-commands) is wired up to leverage that. This gives nice IOPS hike (even 100% at times) particularly on lower queue-depths. Changes since v2: - rebase against for-next (which now has bio-cache that can operate on polled passthrough IO) - bit of rewording in commit description Changes since v1: - corrected variable name (Jens) - fix for a warning (test-robot) Kanchan Joshi (4): fs: add file_operations->uring_cmd_iopoll io_uring: add iopoll infrastructure for io_uring_cmd block: export blk_rq_is_poll nvme: wire up async polling for io passthrough commands block/blk-mq.c | 3 +- drivers/nvme/host/core.c | 1 + drivers/nvme/host/ioctl.c | 73 ++++++++++++++++++++++++++++++++--- drivers/nvme/host/multipath.c | 1 + drivers/nvme/host/nvme.h | 2 + include/linux/blk-mq.h | 1 + include/linux/fs.h | 1 + include/linux/io_uring.h | 8 +++- io_uring/io_uring.c | 6 +++ io_uring/opdef.c | 1 + io_uring/rw.c | 8 +++- io_uring/uring_cmd.c | 11 +++++- 12 files changed, 105 insertions(+), 11 deletions(-) base-commit: 15e543410e9ba86d36a0410bdaf0c02f59fb8936 -- 2.25.1
