public inbox for [email protected]
 help / color / mirror / Atom feed
From: David Wei <[email protected]>
To: [email protected], [email protected]
Cc: Jens Axboe <[email protected]>,
	Pavel Begunkov <[email protected]>,
	Jakub Kicinski <[email protected]>, Paolo Abeni <[email protected]>,
	"David S. Miller" <[email protected]>,
	Eric Dumazet <[email protected]>,
	Jesper Dangaard Brouer <[email protected]>,
	David Ahern <[email protected]>,
	Mina Almasry <[email protected]>
Subject: [RFC PATCH v4 00/16] Zero copy Rx using io_uring
Date: Tue, 12 Mar 2024 14:44:14 -0700	[thread overview]
Message-ID: <[email protected]> (raw)

This patchset is a proposal that adds zero copy network Rx to io_uring.
With it, userspace can register a region of host memory for receiving
data directly from a NIC using DMA, without needing a kernel to user
copy.

This is still a WIP and has a list of known issues. We're looking for
feedback on the overall approach.

Full kernel tree including some out of tree BNXT changes:

https://github.com/spikeh/linux/tree/zcrx

On the userspace side, support is added to both liburing and Netbench:

https://github.com/spikeh/liburing/tree/zcrx2
https://github.com/spikeh/netbench/tree/zcrx

Hardware support is added to the Broadcom BNXT driver. This patchset +
userspace code was tested on an Intel Xeon Platinum 8321HC CPU and
Broadcom BCM57504 NIC.

Early benchmarks using this prototype, with iperf3 as a load generator,
showed a ~50% reduction in overall system memory bandwidth as measured
using perf counters. Note that DDIO must be disabled on Intel systems.
Build Netbench using the modified liburing above.

This patchset is based on the work by Jonathan Lemon
<[email protected]>:
https://lore.kernel.org/io-uring/[email protected]/

Changes in RFC v4:
------------------

* Rebased on top of Mina Almasry's TCP devmem patchset and latest
  net-next, now sharing common infra e.g.:
    * netmem_t and net_iovs
    * Page pool memory provider
* The registered buffer (rbuf) completion queue where completions from
  io_recvzc requests are posted is removed. Now these post into the main
  completion queue, using big (32-byte) CQEs. The first 16 bytes is an
  ordinary CQE, while the latter 16 bytes contain the io_uring_rbuf_cqe
  as before. This vastly simplifies the uAPI and removes a level of
  indirection in userspace when looking for payloads.
  * The rbuf refill queue is still needed for userspace to return
    buffers to kernel.
* Simplified code and uAPI on the io_uring side, particularly
  io_recvzc() and io_zc_rx_recv(). Many unnecessary lines were removed
  e.g. extra msg flags, readlen, etc.

Changes in RFC v3:
------------------

* Rebased on top of Jakub Kicinski's memory provider API RFC. The ZC
  pool added is now a backend for memory provider.
* We're also reusing ppiov infrastructure. The refcounting rules stay
  the same but it's shifted into ppiov->refcount. That lets us to
  flexibly manage buffer lifetimes without adding any extra code to the
  common networking paths. It'd also make it easier to support dmabufs
  and device memory in the future.
  * io_uring also knows about pages, and so ppiovs might unnecessarily
    break tools inspecting data, that can easily be solved later.

Many patches are not for upstream as they depend on work in progress,
namely from Mina:

* struct netmem_t
* Driver ndo commands for Rx queue configs
* struct page_pool_iov and shared pp infra

Changes in RFC v2:
------------------

* Added copy fallback support if userspace memory allocated for ZC Rx
  runs out, or if header splitting or flow steering fails.
* Added veth support for ZC Rx, for testing and demonstration. We will
  need to figure out what driver would be best for such testing
  functionality in the future. Perhaps netdevsim?
* Added socket registration API to io_uring to associate specific
  sockets with ifqs/Rx queues for ZC.
* Added multi-socket support, such that multiple connections can be
  steered into the same hardware Rx queue.
* Added Netbench server/client support.

Known deficiencies that we will address in a future patchset:

* Proper test driver + selftests, maybe netdevsim.
* Revisiting userspace API.
* Multi-region support.
* Steering setup.
* Further optimisation work.
* ...and more.

If you would like to try out this patchset, build and run the kernel
tree then build Netbench using liburing, all from forks above.

Run setup.sh first:

https://gist.github.com/isilence/e6a28ce41a545a261566672104afa461

Then run the following commands:

sudo ip netns exec nsserv ./netbench --server_only 1 --v6 false \
    --rx "io_uring --provide_buffers 0 --use_zc 1 \
    --zc_pool_pages 16384 --zc_ifname ptp-serv" --use_port 9999

sudo ip netns exec nscl ./netbench --client_only 1 --v6 false \
    --tx "epoll --threads 1 --per_thread 1 --size 2800" \
    --host 10.10.10.20 --use_port 9999

David Wei (7):
  io_uring: introduce interface queue
  io_uring: add mmap support for shared ifq ringbuffers
  netdev: add XDP_SETUP_ZC_RX command
  io_uring: setup ZC for an Rx queue when registering an ifq
  io_uring: add zero copy buf representation and pool
  io_uring: add io_recvzc request
  io_uring/zcrx: add copy fallback

Pavel Begunkov (9):
  net: generalise pp provider params passing
  io_uring: delayed cqe commit
  net: page_pool: add ->scrub mem provider callback
  io_uring: separate header for exported net bits
  io_uring/zcrx: implement socket registration
  io_uring: implement pp memory provider for zc rx
  io_uring/zcrx: implement PP_FLAG_DMA_* handling
  net: execute custom callback from napi
  veth: add support for io_uring zc rx

 drivers/net/veth.c             | 214 +++++++-
 include/linux/io_uring.h       |   6 -
 include/linux/io_uring/net.h   |  30 ++
 include/linux/io_uring_types.h |   5 +
 include/linux/net.h            |   2 +
 include/linux/netdevice.h      |   6 +
 include/net/busy_poll.h        |   7 +
 include/net/netdev_rx_queue.h  |   3 +
 include/net/page_pool/types.h  |   2 +
 include/uapi/linux/io_uring.h  |  50 ++
 io_uring/Makefile              |   3 +-
 io_uring/io_uring.c            |  15 +-
 io_uring/io_uring.h            |  10 +
 io_uring/net.c                 | 108 +++-
 io_uring/opdef.c               |  16 +
 io_uring/register.c            |  13 +
 io_uring/uring_cmd.c           |   1 +
 io_uring/zc_rx.c               | 916 +++++++++++++++++++++++++++++++++
 io_uring/zc_rx.h               |  83 +++
 net/core/dev.c                 |  48 +-
 net/core/page_pool.c           |   8 +-
 net/socket.c                   |   3 +-
 22 files changed, 1530 insertions(+), 19 deletions(-)
 create mode 100644 include/linux/io_uring/net.h
 create mode 100644 io_uring/zc_rx.c
 create mode 100644 io_uring/zc_rx.h

-- 
2.43.0


             reply	other threads:[~2024-03-12 21:44 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-12 21:44 David Wei [this message]
2024-03-12 21:44 ` [RFC PATCH v4 01/16] net: generalise pp provider params passing David Wei
2024-03-12 21:44 ` [RFC PATCH v4 02/16] io_uring: delayed cqe commit David Wei
2024-03-12 21:44 ` [RFC PATCH v4 03/16] net: page_pool: add ->scrub mem provider callback David Wei
2024-03-12 21:44 ` [RFC PATCH v4 04/16] io_uring: separate header for exported net bits David Wei
2024-03-12 21:44 ` [RFC PATCH v4 05/16] io_uring: introduce interface queue David Wei
2024-03-12 21:44 ` [RFC PATCH v4 06/16] io_uring: add mmap support for shared ifq ringbuffers David Wei
2024-03-12 21:44 ` [RFC PATCH v4 07/16] netdev: add XDP_SETUP_ZC_RX command David Wei
2024-03-12 21:44 ` [RFC PATCH v4 08/16] io_uring: setup ZC for an Rx queue when registering an ifq David Wei
2024-03-12 21:44 ` [RFC PATCH v4 09/16] io_uring/zcrx: implement socket registration David Wei
2024-03-12 21:44 ` [RFC PATCH v4 10/16] io_uring: add zero copy buf representation and pool David Wei
2024-03-12 21:44 ` [RFC PATCH v4 11/16] io_uring: implement pp memory provider for zc rx David Wei
2024-03-12 21:44 ` [RFC PATCH v4 12/16] io_uring/zcrx: implement PP_FLAG_DMA_* handling David Wei
2024-03-12 21:44 ` [RFC PATCH v4 13/16] io_uring: add io_recvzc request David Wei
2024-03-13 20:25   ` Jens Axboe
2024-03-13 20:26     ` Pavel Begunkov
2024-03-13 21:03       ` Jens Axboe
2024-03-14 16:14       ` Jens Axboe
2024-03-15 17:34         ` Pavel Begunkov
2024-03-15 18:38           ` Jens Axboe
2024-03-15 23:52             ` Pavel Begunkov
2024-03-16 16:59               ` Jens Axboe
2024-03-17 21:22                 ` Pavel Begunkov
2024-03-17 21:30                   ` Jens Axboe
2024-03-12 21:44 ` [RFC PATCH v4 14/16] net: execute custom callback from napi David Wei
2024-03-12 21:44 ` [RFC PATCH v4 15/16] io_uring/zcrx: add copy fallback David Wei
2024-03-12 21:44 ` [RFC PATCH v4 16/16] veth: add support for io_uring zc rx David Wei

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox