KVM Archive on lore.kernel.org
 help / color / Atom feed
From: Jason Wang <jasowang@redhat.com>
To: mst@redhat.com, jasowang@redhat.com
Cc: kvm@vger.kernel.org, virtualization@lists.linux-foundation.org,
	netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
	jfreimann@redhat.com, tiwei.bie@intel.com,
	maxime.coquelin@redhat.com
Subject: [PATCH V3 00/15] Packed virtqueue support for vhost
Date: Wed, 17 Jul 2019 06:52:40 -0400
Message-ID: <20190717105255.63488-1-jasowang@redhat.com> (raw)

Hi all:

This series implements packed virtqueues which were described
at [1]. In this version we try to address the performance regression
saw by V2. The root cause is packed virtqueue need more times of
userspace memory accesssing which turns out to be very
expensive. Thanks to the help of 7f466032dc9e ("vhost: access vq
metadata through kernel virtual address"), such overhead cold be
eliminated. So in this version, we can see about 2% improvement for
packed virtqueue on PPS.

More optimizations (e.g IN_ORDER) is on the road.

Please review.

[1] https://docs.oasis-open.org/virtio/virtio/v1.1/csprd01/virtio-v1.1-csprd01.html#x1-610007

This version were tested with:
- zercopy/datacopy
- mergeable buffer on/off
- TCP stream & virtio-user

Changes from V2:
- rebase on top of vhost metadata accelreation series
- introduce shadow used ring API
- new SET_VRING_BASE/GET_VRING_BASE that takes care about warp counter
  and index for both avail and used
- various twaeaks

Changes from V1:
- drop uapi patch and use Tiwei's
- split the enablement of packed virtqueue into a separate patch

Changes from RFC V5:
- save unnecessary barriers during vhost_add_used_packed_n()
- more compact math for event idx
- fix failure of SET_VRING_BASE when avail_wrap_counter is true
- fix not copy avail_wrap_counter during GET_VRING_BASE
- introduce SET_VRING_USED_BASE/GET_VRING_USED_BASE for syncing
- last_used_idx
- rename used_wrap_counter to last_used_wrap_counter
- rebase to net-next

Changes from RFC V4:
- fix signalled_used index recording
- track avail index correctly
- various minor fixes

Changes from RFC V3:
- Fix math on event idx checking
- Sync last avail wrap counter through GET/SET_VRING_BASE
- remove desc_event prefix in the driver/device structure

Changes from RFC V2:
- do not use & in checking desc_event_flags
- off should be most significant bit
- remove the workaround of mergeable buffer for dpdk prototype
- id should be in the last descriptor in the chain
- keep _F_WRITE for write descriptor when adding used
- device flags updating should use ADDR_USED type
- return error on unexpected unavail descriptor in a chain
- return false in vhost_ve_avail_empty is descriptor is available
- track last seen avail_wrap_counter
- correctly examine available descriptor in get_indirect_packed()
- vhost_idx_diff should return u16 instead of bool

Changes from RFC V1:
- Refactor vhost used elem code to avoid open coding on used elem
- Event suppression support (compile test only).
- Indirect descriptor support (compile test only).
- Zerocopy support.
- vIOMMU support.
- SCSI/VSOCK support (compile test only).
- Fix several bugs

Jason Wang (15):
  vhost: simplify meta data pointer accessing
  vhost: remove the unnecessary parameter of vhost_vq_avail_empty()
  vhost: remove unnecessary parameter of
    vhost_enable_notify()/vhost_disable_notify
  vhost-net: don't use vhost_add_used_n() for zerocopy
  vhost: introduce helpers to manipulate shadow used ring
  vhost_net: switch TX to use shadow used ring API
  vhost_net: calculate last used length once for mergeable buffer
  vhost_net: switch to use shadow used ring API for RX
  vhost: do not export vhost_add_used_n() and
    vhost_add_used_and_signal_n()
  vhost: hide used ring layout from device
  vhost: do not use vring_used_elem
  vhost: vhost_put_user() can accept metadata type
  vhost: packed ring support
  vhost: event suppression for packed ring
  vhost: enable packed virtqueues

 drivers/vhost/net.c   |  200 +++---
 drivers/vhost/scsi.c  |   72 +-
 drivers/vhost/test.c  |    6 +-
 drivers/vhost/vhost.c | 1508 +++++++++++++++++++++++++++++++++++------
 drivers/vhost/vhost.h |   78 ++-
 drivers/vhost/vsock.c |   57 +-
 6 files changed, 1513 insertions(+), 408 deletions(-)

-- 
2.18.1


             reply index

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-07-17 10:52 Jason Wang [this message]
2019-07-17 10:52 ` [PATCH V3 01/15] vhost: simplify meta data pointer accessing Jason Wang
2019-07-17 10:52 ` [PATCH V3 02/15] vhost: remove the unnecessary parameter of vhost_vq_avail_empty() Jason Wang
2019-07-17 10:52 ` [PATCH V3 03/15] vhost: remove unnecessary parameter of vhost_enable_notify()/vhost_disable_notify Jason Wang
2019-07-17 10:52 ` [PATCH V3 04/15] vhost-net: don't use vhost_add_used_n() for zerocopy Jason Wang
2019-07-17 10:52 ` [PATCH V3 05/15] vhost: introduce helpers to manipulate shadow used ring Jason Wang
2019-07-17 10:52 ` [PATCH V3 06/15] vhost_net: switch TX to use shadow used ring API Jason Wang
2019-07-17 10:52 ` [PATCH V3 07/15] vhost_net: calculate last used length once for mergeable buffer Jason Wang
2019-07-17 10:52 ` [PATCH V3 08/15] vhost_net: switch to use shadow used ring API for RX Jason Wang
2019-07-17 10:52 ` [PATCH V3 09/15] vhost: do not export vhost_add_used_n() and vhost_add_used_and_signal_n() Jason Wang
2019-07-17 10:52 ` [PATCH V3 10/15] vhost: hide used ring layout from device Jason Wang
2019-07-17 10:52 ` [PATCH V3 11/15] vhost: do not use vring_used_elem Jason Wang
2019-07-17 10:52 ` [PATCH V3 12/15] vhost: vhost_put_user() can accept metadata type Jason Wang
2019-07-17 10:52 ` [PATCH V3 13/15] vhost: packed ring support Jason Wang
2019-07-17 10:52 ` [PATCH V3 14/15] vhost: event suppression for packed ring Jason Wang
2019-07-17 10:52 ` [PATCH V3 15/15] vhost: enable packed virtqueues Jason Wang
2019-07-17 11:02 ` [PATCH V3 00/15] Packed virtqueue support for vhost Michael S. Tsirkin
2019-07-17 12:27   ` Jason Wang
2019-07-17 14:28     ` Michael S. Tsirkin

Reply instructions:

You may reply publically 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 \
    --in-reply-to=20190717105255.63488-1-jasowang@redhat.com \
    --to=jasowang@redhat.com \
    --cc=jfreimann@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maxime.coquelin@redhat.com \
    --cc=mst@redhat.com \
    --cc=netdev@vger.kernel.org \
    --cc=tiwei.bie@intel.com \
    --cc=virtualization@lists.linux-foundation.org \
    /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

KVM Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/kvm/0 kvm/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 kvm kvm/ https://lore.kernel.org/kvm \
		kvm@vger.kernel.org kvm@archiver.kernel.org
	public-inbox-index kvm

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.kvm


AGPL code for this site: git clone https://public-inbox.org/ public-inbox