mbox series

[PATCHv3,0/5] block integrity: directly map user space addresses

Message ID 20231120224058.2750705-1-kbusch@meta.com (mailing list archive)
Headers show
Series block integrity: directly map user space addresses | expand

Message

Keith Busch Nov. 20, 2023, 10:40 p.m. UTC 
From: Keith Busch <kbusch@kernel.org>

Handling passthrough metadata ("integrity") today introduces overhead
and complications that we can avoid if we just map user space addresses
directly. This patch series implements that, falling back to a kernel
bounce buffer if necessary.

v2->v3:

  Introduces a multi-page bvec iterator.

  Fix leaking pinned user pages (Kanchan)

  Fix final unpaired 'put' on user pages (Kanchan)

  Doesn't increase the size of 'struct bio_integrity_profile'; if the
  'copy_vec' pointer is needed, it gets appended to the existing bvec.

  Fix compiler warnings

  Fix compiler error for !CONFIG_BLK_INTEGRITY

Keith Busch (5):
  bvec: introduce multi-page bvec iterating
  block: bio-integrity: directly map user buffers
  nvme: use bio_integrity_map_user
  iouring: remove IORING_URING_CMD_POLLED
  io_uring: remove uring_cmd cookie

 block/bio-integrity.c     | 212 ++++++++++++++++++++++++++++++++++++++
 drivers/nvme/host/ioctl.c | 197 ++++++-----------------------------
 include/linux/bio.h       |  12 +++
 include/linux/bvec.h      |   6 ++
 include/linux/io_uring.h  |   9 +-
 io_uring/uring_cmd.c      |   1 -
 6 files changed, 261 insertions(+), 176 deletions(-)