DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Cao, Waterman" <waterman.cao@intel.com>
To: 'Yong Wang' <yongwang@vmware.com>,
	Thomas Monjalon <thomas.monjalon@6wind.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH 0/5] vmxnet3 pmd fixes/improvement
Date: Wed, 5 Nov 2014 01:32:16 +0000	[thread overview]
Message-ID: <AA3F441F262C58498CD6D0C1801DE7EB0ABC1E40@SHSMSX103.ccr.corp.intel.com> (raw)
In-Reply-To: <edc61efd1bbf4798ac333a979405ed8b@EX13-MBX-026.vmware.com>

Hi Yong,

	We tested your patch with VMWare ESX 5.5.
	It works fine with R1.8 RC1. 
	You can find more details from Xiaonan's reports.

Regards

Waterman 
>-----Original Message-----
>From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Yong Wang
>Sent: Tuesday, October 14, 2014 5:00 AM
>To: Thomas Monjalon
>Cc: dev@dpdk.org
>Subject: Re: [dpdk-dev] [PATCH 0/5] vmxnet3 pmd fixes/improvement
>
>Only the last one is performance related and it merely tries to give hints to the compiler to hopefully make branch prediction more efficient.  It also moves a constant assignment out of the pkt polling loop.
>
>We did performance evaluation on a Nehalem box with 4cores@2.8GHz x 2 socket:
>On the DPDK-side, it's running some l3 forwarding apps in a VM on ESXi with one core assigned for polling.  The client side is pktgen/dpdk, pumping 64B tcp packets at line rate.  Before the patch, we are seeing ~900K PPS with 65% cpu of a core used for DPDK.  After the patch, we are seeing the same pkt rate with only 45% of a core used.  CPU usage is collected factoring our the idle loop cost.  The packet rate is a result of the mode we used for vmxnet3 (pure emulation mode running default number of hypervisor contexts).  I can add these info in the review request.
>
>Yong
>________________________________________
>From: Thomas Monjalon <thomas.monjalon@6wind.com>
>Sent: Monday, October 13, 2014 1:29 PM
>To: Yong Wang
>Cc: dev@dpdk.org
>Subject: Re: [dpdk-dev] [PATCH 0/5] vmxnet3 pmd fixes/improvement
>
>Hi,
>
>2014-10-12 23:23, Yong Wang:
>> This patch series include various fixes and improvement to the
>> vmxnet3 pmd driver.
>>
>> Yong Wang (5):
>>   vmxnet3: Fix VLAN Rx stripping
>>   vmxnet3: Add VLAN Tx offload
>>   vmxnet3: Fix dev stop/restart bug
>>   vmxnet3: Add rx pkt check offloads
>>   vmxnet3: Some perf improvement on the rx path
>
>Please, could describe what is the performance gain for these patches?
>Benchmark numbers would be appreciated.
>
>Thanks
>--
>Thomas

-----Original Message-----
From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Yong Wang
Sent: Tuesday, October 14, 2014 5:00 AM
To: Thomas Monjalon
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] [PATCH 0/5] vmxnet3 pmd fixes/improvement

Only the last one is performance related and it merely tries to give hints to the compiler to hopefully make branch prediction more efficient.  It also moves a constant assignment out of the pkt polling loop.

We did performance evaluation on a Nehalem box with 4cores@2.8GHz x 2 socket:
On the DPDK-side, it's running some l3 forwarding apps in a VM on ESXi with one core assigned for polling.  The client side is pktgen/dpdk, pumping 64B tcp packets at line rate.  Before the patch, we are seeing ~900K PPS with 65% cpu of a core used for DPDK.  After the patch, we are seeing the same pkt rate with only 45% of a core used.  CPU usage is collected factoring our the idle loop cost.  The packet rate is a result of the mode we used for vmxnet3 (pure emulation mode running default number of hypervisor contexts).  I can add these info in the review request.

Yong
________________________________________
From: Thomas Monjalon <thomas.monjalon@6wind.com>
Sent: Monday, October 13, 2014 1:29 PM
To: Yong Wang
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] [PATCH 0/5] vmxnet3 pmd fixes/improvement

Hi,

2014-10-12 23:23, Yong Wang:
> This patch series include various fixes and improvement to the
> vmxnet3 pmd driver.
>
> Yong Wang (5):
>   vmxnet3: Fix VLAN Rx stripping
>   vmxnet3: Add VLAN Tx offload
>   vmxnet3: Fix dev stop/restart bug
>   vmxnet3: Add rx pkt check offloads
>   vmxnet3: Some perf improvement on the rx path

Please, could describe what is the performance gain for these patches?
Benchmark numbers would be appreciated.

Thanks
--
Thomas

  parent reply	other threads:[~2014-11-05  1:24 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-10-13  6:23 Yong Wang
2014-10-13  6:23 ` [dpdk-dev] [PATCH 1/5] vmxnet3: Fix VLAN Rx stripping Yong Wang
2014-10-13  9:31   ` Stephen Hemminger
2014-10-13 18:42     ` Yong Wang
2014-10-22 13:39       ` Stephen Hemminger
2014-10-28 21:57         ` Yong Wang
2014-10-29  9:04           ` Bruce Richardson
2014-10-29  9:41             ` Thomas Monjalon
2014-10-29 17:57               ` Yong Wang
2014-10-29 18:51                 ` Thomas Monjalon
2014-10-13  6:23 ` [dpdk-dev] [PATCH 2/5] vmxnet3: Add VLAN Tx offload Yong Wang
2014-10-13  6:23 ` [dpdk-dev] [PATCH 3/5] vmxnet3: Fix dev stop/restart bug Yong Wang
2014-10-13  6:23 ` [dpdk-dev] [PATCH 4/5] vmxnet3: Add rx pkt check offloads Yong Wang
2014-10-13  6:23 ` [dpdk-dev] [PATCH 5/5] vmxnet3: Some perf improvement on the rx path Yong Wang
2014-11-05  0:13   ` Thomas Monjalon
2014-10-13 20:29 ` [dpdk-dev] [PATCH 0/5] vmxnet3 pmd fixes/improvement Thomas Monjalon
2014-10-13 21:00   ` Yong Wang
2014-10-21 22:10     ` Yong Wang
2014-10-22  7:07       ` Cao, Waterman
2014-10-28 14:40         ` Thomas Monjalon
2014-10-28 19:59           ` Yong Wang
2014-10-29  0:33             ` Cao, Waterman
2014-11-05  1:32     ` Cao, Waterman [this message]
2014-11-04  5:57 ` Zhang, XiaonanX
2014-11-04 22:50   ` Thomas Monjalon
2014-11-05  5:26     ` Cao, Waterman

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=AA3F441F262C58498CD6D0C1801DE7EB0ABC1E40@SHSMSX103.ccr.corp.intel.com \
    --to=waterman.cao@intel.com \
    --cc=dev@dpdk.org \
    --cc=thomas.monjalon@6wind.com \
    --cc=yongwang@vmware.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).