[dpdk-dev] [PATCH v3 00/21] implement packed virtqueues
Jens Freimann
jfreimann at redhat.com
Thu Apr 5 12:10:10 CEST 2018
This is a basic implementation of packed virtqueues as specified in the
Virtio 1.1 draft. A compiled version of the current draft is available
at https://github.com/oasis-tcs/virtio-docs.git (or as .pdf at
https://github.com/oasis-tcs/virtio-docs/blob/master/virtio-v1.1-packed-wd10.pdf
It does not implement yet indirect descriptors and checksum offloading.
A packed virtqueue is different from a split virtqueue in that it
consists of only a single descriptor ring that replaces available and
used ring, index and descriptor buffer.
Each descriptor is readable and writable and has a flags field. These flags
will mark if a descriptor is available or used. To detect new available descriptors
even after the ring has wrapped, device and driver each have a
single-bit wrap counter that is flipped from 0 to 1 and vice versa every time
the last descriptor in the ring is used/made available.
The idea behind this is to 1. improve performance by avoiding cache misses
and 2. be easier for devices to implement.
Regarding performance: with these patches I get 21.13 Mpps on my system
as compared to 18.8 Mpps with the virtio 1.0 code. Packet size was 64
bytes, 0.05% acceptable loss. Test setup is described as in
http://dpdk.org/doc/guides/howto/pvp_reference_benchmark.html
Packet generator:
MoonGen
Intel(R) Xeon(R) CPU E5-2665 0 @ 2.40GHz
Intel X710 NIC
RHEL 7.4
Device under test:
Intel(R) Xeon(R) CPU E5-2667 v4 @ 3.20GHz
Intel X710 NIC
RHEL 7.4
VM on DuT: RHEL7.4
I plan to do more performance test with bigger frame sizes.
This patch series is based on a prototype implemented by Yuanhan Liu and
Tiwei Bie.
changes from v2->v3:
* implement event suppression
* add code do dump packed virtqueues
* don't use assert in vhost code
* rename virtio-user parameter to packed-vq
* support rxvf flush
changes from v1->v2:
* don't use VIRTQ_DESC_F_NEXT in used descriptors (Jason)
* no rte_panice() in guest triggerable code (Maxime)
* use unlikely when checking for vq (Maxime)
* rename everything from _1_1 to _packed (Yuanhan)
* add two more patches to implement mergeable receive buffers
regards,
Jens
Jens Freimann (17):
net/virtio: by default disable packed virtqueues
net/virtio: vring init for packed queues
net/virtio: add virtio 1.1 defines
net/virtio: add packed virtqueue helpers
net/virtio: dump packed virtqueue data
net/virtio: implement transmit path for packed queues
vhost: disable packed virtqueues by default
vhost: turn of indirect descriptors for packed virtqueues
vhost: add virtio 1.1 defines
vhost: add helpers for packed virtqueues
vhost: dequeue for packed queues
vhost: packed queue enqueue path
net/virtio: disable ctrl virtqueue for packed rings
net/virtio: add support for mergeable buffers with packed virtqueues
vhost: support mergeable rx buffers with packed queues
net/virtio: add support for event suppression
vhost: add event suppression for packed queues
Yuanhan Liu (4):
net/virtio-user: add option to use packed queues
net/virtio: implement receive path for packed queues
vhost: vring address setup for packed queues
vhost: enable packed virtqueues
config/common_base | 2 +
drivers/net/virtio/virtio_ethdev.c | 51 ++-
drivers/net/virtio/virtio_ethdev.h | 7 +-
drivers/net/virtio/virtio_pci.h | 8 +
drivers/net/virtio/virtio_ring.h | 91 +++-
drivers/net/virtio/virtio_rxtx.c | 359 +++++++++++++++-
drivers/net/virtio/virtio_user/virtio_user_dev.c | 12 +-
drivers/net/virtio/virtio_user/virtio_user_dev.h | 3 +-
drivers/net/virtio/virtio_user_ethdev.c | 15 +-
drivers/net/virtio/virtqueue.c | 10 +
drivers/net/virtio/virtqueue.h | 91 +++-
lib/librte_vhost/socket.c | 5 +
lib/librte_vhost/vhost.c | 16 +-
lib/librte_vhost/vhost.h | 69 ++-
lib/librte_vhost/vhost_user.c | 40 +-
lib/librte_vhost/virtio-1.1.h | 62 +++
lib/librte_vhost/virtio_net.c | 507 +++++++++++++++++++++--
17 files changed, 1271 insertions(+), 77 deletions(-)
create mode 100644 lib/librte_vhost/virtio-1.1.h
--
2.14.3
More information about the dev
mailing list