From: "Liu, Yong" <yong.liu@intel.com>
To: "maxime.coquelin@redhat.com" <maxime.coquelin@redhat.com>,
"Bie, Tiwei" <tiwei.bie@intel.com>
Cc: "Wang, Zhihong" <zhihong.wang@intel.com>, "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH 0/7] support VIRTIO_F_IN_ORDER feature
Date: Fri, 8 Jun 2018 02:11:46 +0000 [thread overview]
Message-ID: <86228AFD5BCD8E4EBFD2B90117B5E81E630BF9B0@SHSMSX103.ccr.corp.intel.com> (raw)
In-Reply-To: <20180608090724.20855-1-yong.liu@intel.com>
Sorry, add missing loopback performance comparison.
+---------+----------+----------------+----------------+
|Loopback |1 Queue |2 Queues |4 Queues |
+---------+----------+----------------+----------------+
|Inorder |6.2Mpps |9.5 ~ 11.9Mpps |10.6 ~ 11.3Mpps |
+---------+----------+----------------+----------------+
|Normal |7.5Mpps |7.7 ~ 9.0Mpps |7.6 ~ 7.8Mpps |
+---------+----------+----------------+----------------+
Thanks,
Marvin
> -----Original Message-----
> From: Liu, Yong
> Sent: Friday, June 08, 2018 5:07 PM
> To: maxime.coquelin@redhat.com; Bie, Tiwei <tiwei.bie@intel.com>
> Cc: Wang, Zhihong <zhihong.wang@intel.com>; dev@dpdk.org; Liu, Yong
> <yong.liu@intel.com>
> Subject: [PATCH 0/7] support VIRTIO_F_IN_ORDER feature
>
> In latest virtio-spec, new feature bit VIRTIO_F_IN_ORDER was introduced.
> When this feature has been negotiated, virtio driver will use
> descriptors in ring order: starting from offset 0 in the table, and
> wrapping around at the end of the table. Vhost devices will always use
> descriptors in the same order in which they have been made available.
> This can simplify vhost and device code as desc/used/avail ring are
> using same index.
>
> Based on updated virtio-spec, this series realized IN_ORDER prototype
> in virtio driver. Due to new [RT]x path added into selection, also add
> two new parameters mrg_rx and in_order into virtio-user vdev parameters
> list. This will allow user to configure feature bits thus can impact
> [RT]x path selection.
>
> IN_ORDER can improve virtio/vhost performance. This patch implement
> virtio driver part only, we can see significant gain there. As a result
> it impacts vhost performance a little since there’s no dedicated Rx/Tx
> implementation on vhost side. This can be added in another patch later.
>
> Performance of virtio user with IN_ORDER feature:
>
> Platform: Purely
> CPU: Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
> DPDK baseline: 18.05
> Setup: testpmd with vhost vdev + testpmd with virtio vdev
>
> +--------------+----------+----------+---------+
> |Vhost->Virtio |1 Queue |2 Queues |4 Queues |
> +--------------+----------+----------+---------+
> |Inorder |12.9Mpps |25.4Mpps |31.6Mpps |
> |Normal |12.1Mpps |18.5Mpps |18.9Mpps |
> +--------------+----------+----------+---------+
>
> +--------------+----------+----------------+---------+
> |Virtio->Vhost |1 Queue |2 Queues |4 Queues |
> +--------------+----------+----------------+---------+
> |Inorder |16.4Mpps |11.9 ~ 19.5Mpps |11.8Mpps |
> |Normal |13.3Mpps |9.8 ~ 14Mpps |10.5Mpps |
> +--------------+----------+----------------+---------+
>
> +---------+----------+----------------+----------------+
> |Loopback |1 Queue |2 Queues |4 Queues |
> +---------+----------+----------------+----------------+
> |Inorder |6.2Mpps |9.5 ~ 11.9Mpps |10.6 ~ 11.3Mpps |
> +---------+----------+----------------+----------------+
>
> Marvin Liu (7):
> vhost: announce VIRTIO_F_IN_ORDER support
> net/virtio: add VIRTIO_F_IN_ORDER definition
> net/virtio-user: add mgr_rxbuf and in_order vdev parameters
> net/virtio: free IN_ORDER descriptors
> net/virtio: support IN_ORDER Rx and Tx
> net/virtio: add IN_ORDER Rx/Tx into selection
> net/virtio: annouce VIRTIO_F_IN_ORDER support
>
> drivers/net/virtio/virtio_ethdev.c | 31 +-
> drivers/net/virtio/virtio_ethdev.h | 7 +
> drivers/net/virtio/virtio_pci.h | 8 +
> drivers/net/virtio/virtio_rxtx.c | 618 ++++++++++++++++--
> .../net/virtio/virtio_user/virtio_user_dev.c | 10 +-
> .../net/virtio/virtio_user/virtio_user_dev.h | 3 +-
> drivers/net/virtio/virtio_user_ethdev.c | 33 +-
> drivers/net/virtio/virtqueue.c | 8 +
> drivers/net/virtio/virtqueue.h | 2 +
> lib/librte_vhost/socket.c | 4 +
> lib/librte_vhost/vhost.h | 10 +-
> 11 files changed, 668 insertions(+), 66 deletions(-)
>
> --
> 2.17.0
next prev parent reply other threads:[~2018-06-08 2:11 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-06-08 9:07 Marvin Liu
2018-06-08 2:11 ` Liu, Yong [this message]
2018-06-08 9:07 ` [dpdk-dev] [PATCH 1/7] vhost: announce VIRTIO_F_IN_ORDER support Marvin Liu
2018-06-13 6:18 ` Tiwei Bie
2018-06-08 9:07 ` [dpdk-dev] [PATCH 2/7] net/virtio: add VIRTIO_F_IN_ORDER definition Marvin Liu
2018-06-08 9:07 ` [dpdk-dev] [PATCH 3/7] net/virtio-user: add mgr_rxbuf and in_order vdev parameters Marvin Liu
2018-06-13 6:37 ` Tiwei Bie
2018-06-08 9:07 ` [dpdk-dev] [PATCH 4/7] net/virtio: free IN_ORDER descriptors Marvin Liu
2018-06-08 9:07 ` [dpdk-dev] [PATCH 5/7] net/virtio: support IN_ORDER Rx and Tx Marvin Liu
2018-06-20 7:41 ` Tiwei Bie
2018-06-22 8:05 ` Liu, Yong
2018-06-08 9:07 ` [dpdk-dev] [PATCH 6/7] net/virtio: add IN_ORDER Rx/Tx into selection Marvin Liu
2018-06-20 7:44 ` Tiwei Bie
2018-06-20 16:13 ` Liu, Yong
2018-06-08 9:07 ` [dpdk-dev] [PATCH 7/7] net/virtio: annouce VIRTIO_F_IN_ORDER support Marvin Liu
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 \
--in-reply-to=86228AFD5BCD8E4EBFD2B90117B5E81E630BF9B0@SHSMSX103.ccr.corp.intel.com \
--to=yong.liu@intel.com \
--cc=dev@dpdk.org \
--cc=maxime.coquelin@redhat.com \
--cc=tiwei.bie@intel.com \
--cc=zhihong.wang@intel.com \
/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;
as well as URLs for NNTP newsgroup(s).