From: Maxime Coquelin <maxime.coquelin@redhat.com>
To: patrick.fu@intel.com, dev@dpdk.org, chenbo.xia@intel.com,
zhihong.wang@intel.com, xiaolong.ye@intel.com
Cc: cheng1.jiang@intel.com, cunming.liang@intel.com
Subject: Re: [dpdk-dev] [PATCH v1 0/2] introduce asynchronous data path for vhost
Date: Fri, 26 Jun 2020 16:42:37 +0200 [thread overview]
Message-ID: <0a2fdf05-685d-0f09-4af9-4b7dc56e22c0@redhat.com> (raw)
In-Reply-To: <1591869725-13331-1-git-send-email-patrick.fu@intel.com>
Hi Patrick,
On 6/11/20 12:02 PM, patrick.fu@intel.com wrote:
> From: Patrick Fu <patrick.fu@intel.com>
>
> Performing large memory copies usually takes up a major part of CPU
> cycles and becomes the hot spot in vhost-user enqueue operation. To
> offload expensive memory operations from the CPU, this patch set
> proposes to leverage DMA engines, e.g., I/OAT, a DMA engine in the
> Intel's processor, to accelerate large copies.
>
> Large copies are offloaded from the CPU to the DMA in an asynchronous
> manner. The CPU just submits copy jobs to the DMA but without waiting
> for its copy completion. Thus, there is no CPU intervention during
> data transfer; we can save precious CPU cycles and improve the overall
> throughput for vhost-user based applications, like OVS. During packet
> transmission, it offloads large copies to the DMA and performs small
> copies by the CPU, due to startup overheads associated with the DMA.
>
> This patch set construct a general framework that applications can
> leverage to attach DMA channels with vhost-user transmit queues. Four
> new RTE APIs are introduced to vhost library for applications to
> register and use the asynchronous data path. In addition, two new DMA
> operation callbacks are defined, by which vhost-user asynchronous data
> path can interact with DMA hardware. Currently only enqueue operation
> for split queue is implemented, but the frame is flexible to extend
> support for dequeue & packed queue.
Thanks for this big rework of the Vhost DMA series.
It looks overall good to me and is consistent with what design you
suggested few months back.
I don't see a big risk to integrate it in v20.08 once the few comments
are taken into account.
I'll try to make another pass before you send a v2.
Maxime
> Patrick Fu (2):
> vhost: introduce async data path registration API
> vhost: introduce async enqueue for split ring
>
> lib/librte_vhost/Makefile | 3 +-
> lib/librte_vhost/rte_vhost.h | 1 +
> lib/librte_vhost/rte_vhost_async.h | 172 ++++++++++++
> lib/librte_vhost/socket.c | 20 ++
> lib/librte_vhost/vhost.c | 74 ++++-
> lib/librte_vhost/vhost.h | 30 ++-
> lib/librte_vhost/vhost_user.c | 28 +-
> lib/librte_vhost/virtio_net.c | 538 ++++++++++++++++++++++++++++++++++++-
> 8 files changed, 857 insertions(+), 9 deletions(-)
> create mode 100644 lib/librte_vhost/rte_vhost_async.h
>
next prev parent reply other threads:[~2020-06-26 14:42 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-11 10:02 patrick.fu
2020-06-11 10:02 ` [dpdk-dev] [PATCH v1 1/2] vhost: introduce async data path registration API patrick.fu
2020-06-18 5:50 ` Liu, Yong
2020-06-18 9:08 ` Fu, Patrick
2020-06-19 0:40 ` Liu, Yong
2020-06-25 13:42 ` Maxime Coquelin
2020-06-26 14:28 ` Maxime Coquelin
2020-06-29 1:15 ` Fu, Patrick
2020-06-26 14:44 ` Maxime Coquelin
2020-06-11 10:02 ` [dpdk-dev] [PATCH v1 2/2] vhost: introduce async enqueue for split ring patrick.fu
2020-06-18 6:56 ` Liu, Yong
2020-06-18 11:36 ` Fu, Patrick
2020-06-26 14:39 ` Maxime Coquelin
2020-06-26 14:46 ` Maxime Coquelin
2020-06-29 1:25 ` Fu, Patrick
2020-06-26 14:42 ` Maxime Coquelin [this message]
2020-07-03 10:27 ` [dpdk-dev] [PATCH v3 0/2] introduce asynchronous data path for vhost patrick.fu
2020-07-03 10:27 ` [dpdk-dev] [PATCH v3 1/2] vhost: introduce async enqueue registration API patrick.fu
2020-07-03 10:27 ` [dpdk-dev] [PATCH v3 2/2] vhost: introduce async enqueue for split ring patrick.fu
2020-07-03 12:21 ` [dpdk-dev] [PATCH v4 0/2] introduce asynchronous data path for vhost patrick.fu
2020-07-03 12:21 ` [dpdk-dev] [PATCH v4 1/2] vhost: introduce async enqueue registration API patrick.fu
2020-07-06 3:05 ` Liu, Yong
2020-07-06 9:08 ` Fu, Patrick
2020-07-03 12:21 ` [dpdk-dev] [PATCH v4 2/2] vhost: introduce async enqueue for split ring patrick.fu
2020-07-06 11:53 ` [dpdk-dev] [PATCH v5 0/2] introduce asynchronous data path for vhost patrick.fu
2020-07-06 11:53 ` [dpdk-dev] [PATCH v5 1/2] vhost: introduce async enqueue registration API patrick.fu
2020-07-06 11:53 ` [dpdk-dev] [PATCH v5 2/2] vhost: introduce async enqueue for split ring patrick.fu
2020-07-07 5:07 ` [dpdk-dev] [PATCH v6 0/2] introduce asynchronous data path for vhost patrick.fu
2020-07-07 5:07 ` [dpdk-dev] [PATCH v6 1/2] vhost: introduce async enqueue registration API patrick.fu
2020-07-07 8:22 ` Xia, Chenbo
2020-07-07 5:07 ` [dpdk-dev] [PATCH v6 2/2] vhost: introduce async enqueue for split ring patrick.fu
2020-07-07 8:22 ` Xia, Chenbo
2020-07-07 16:45 ` [dpdk-dev] [PATCH v6 0/2] introduce asynchronous data path for vhost Ferruh Yigit
2020-07-20 13:26 ` Maxime Coquelin
2020-07-21 2:28 ` Fu, Patrick
2020-07-21 8:28 ` Maxime Coquelin
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=0a2fdf05-685d-0f09-4af9-4b7dc56e22c0@redhat.com \
--to=maxime.coquelin@redhat.com \
--cc=chenbo.xia@intel.com \
--cc=cheng1.jiang@intel.com \
--cc=cunming.liang@intel.com \
--cc=dev@dpdk.org \
--cc=patrick.fu@intel.com \
--cc=xiaolong.ye@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).