DPDK usage discussions
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber.org>
To: 宋捷 <songj@zctt.com>
Cc: "'users'" <users@dpdk.org>
Subject: Re: [dpdk-users] How to make virtual NIC (virtio and vmxnet3) support IP header checksum?
Date: Tue, 26 Feb 2019 07:48:15 -0800	[thread overview]
Message-ID: <20190226074815.6203533d@shemminger-XPS-13-9360> (raw)
In-Reply-To: <5c74f8e3.1c69fb81.b9311.cde6SMTPIN_ADDED_BROKEN@mx.google.com>

On Tue, 26 Feb 2019 16:29:00 +0800
宋捷 <songj@zctt.com> wrote:

> Hi All
> 
>     I test the DPDK virtual nics both in Esxi and KVM, I used DPDK  vmxnet3,
> virtio and e1000 nic. I need create IP packets and TX them.
> 
>  
> 
> I set the rte_mbuf as below before TX.
> 
> mb->l2_len = len(out_eth)
> 
> mb->l3_len = len(out_ip)
> 
> mb->ol_flags |= PKT_TX_IPV4 | PKT_TX_IP_CSUM
> 
>  
> 
> It worked well with DPDK virtual e1000 nic, the IP header checksum will be
> set correct value. but vmxnet3 and virtio just keep 0 for IP header
> checksum. 
> 
> I disable above code and call rte_ipv4_cksum for each IP header checksum. it
> worked with vmxnet3 and virtio when TX slowly, but checksum invalid once TX
> quicklyL
> 
>  
> 
> Why the DPDK vmxnet3 and virtio drivers don't support IP checksum?
> 
> If there’s an easy way for adding the IP header checksum with vmxnet3 and
> virtio ?
> 
>  
> 
>  
> 
> Thanks
> 
>  
> 
> Jie
> 

You need to look at the TX_OFFLOAD capability flags in dev_info.
Many drivers do not support IP checksum offload.

       reply	other threads:[~2019-02-26 15:48 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <5c74f8e3.1c69fb81.b9311.cde6SMTPIN_ADDED_BROKEN@mx.google.com>
2019-02-26 15:48 ` Stephen Hemminger [this message]
2019-02-27  2:40   ` [dpdk-users] 答复: " 宋捷
2019-02-26  8:29 [dpdk-users] " 宋捷

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=20190226074815.6203533d@shemminger-XPS-13-9360 \
    --to=stephen@networkplumber.org \
    --cc=songj@zctt.com \
    --cc=users@dpdk.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
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).