From: Jiayu Hu <jiayu.hu@intel.com>
To: dev@dpdk.org
Cc: konstantin.ananyev@intel.com, jiayu.hu@intel.com
Subject: [dpdk-dev] [RFC] Add UDP GSO Support
Date: Mon, 14 May 2018 11:35:37 +0800 [thread overview]
Message-ID: <20180514033537.GA64815@dpdk15.sh.intel.com> (raw)
With the support of UDP Fragmentation Offload (UFO) and TCP Segmentation
Offload (TSO) in virtio, VMs can exchange large UDP and TCP packets
exceeding MTU between each other, which can greatly reduce per-packet
processing overheads.
When the destination of the large TCP and UDP packets is crossing
machines, the host application needs to call two different libraries,
GSO and IP fragmentation, to split the large packets respectively.
However,the GSO and IP fragmentation library have quite different APIs,
which greatly complicates the host application implementation.
To simplify application development, we propose to support UDP segmentation
offload in the GSO library. With supporting UDP GSO, host applicationss can
use the unified APIs to split large UDP and TCP packets.
Since we already have the IP fragmentation library, we can implement UDP
GSO in two ways:
1. Calling the IP fragmentation library inside the GSO library. This
method can resue the existed code, but it also introduces some
changes. Specifically, the GSO library ignores checksum, but the IP
fragmentation library sets PKT_TX_IP_CKSUM for output packets by
default. To avoid confusing applications, we need to change the GSO
or IP fragmentation library design, but both of them would change the
behaviors of the applications which already use the two libraries.
2. Implementing a new UDP fragmentation function in the GSO library. This
method can keep current designs for the two libraries, but we cannot
reuse the existed code.
Look forward to your opinions about this proposal.
Thanks,
Jiayu
reply other threads:[~2018-05-14 3:28 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=20180514033537.GA64815@dpdk15.sh.intel.com \
--to=jiayu.hu@intel.com \
--cc=dev@dpdk.org \
--cc=konstantin.ananyev@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).