From: Maxime Coquelin <maxime.coquelin@redhat.com>
To: "Wang, Zhihong" <zhihong.wang@intel.com>,
Yuanhan Liu <yuanhan.liu@linux.intel.com>
Cc: "stephen@networkplumber.org" <stephen@networkplumber.org>,
"Pierre Pfister (ppfister)" <ppfister@cisco.com>,
"Xie, Huawei" <huawei.xie@intel.com>,
"dev@dpdk.org" <dev@dpdk.org>,
"vkaplans@redhat.com" <vkaplans@redhat.com>,
"mst@redhat.com" <mst@redhat.com>
Subject: Re: [dpdk-dev] [PATCH v4] vhost: Add indirect descriptors support to the TX path
Date: Fri, 4 Nov 2016 13:54:00 +0100 [thread overview]
Message-ID: <75d2c094-11e8-99aa-ccb5-60ded04047d7@redhat.com> (raw)
In-Reply-To: <8F6C2BD409508844A0EFC19955BE09414E7DC66F@SHSMSX103.ccr.corp.intel.com>
On 11/04/2016 01:30 PM, Wang, Zhihong wrote:
>
>
>> -----Original Message-----
>> From: Maxime Coquelin [mailto:maxime.coquelin@redhat.com]
>> Sent: Friday, November 4, 2016 7:23 PM
>> To: Wang, Zhihong <zhihong.wang@intel.com>; Yuanhan Liu
>> <yuanhan.liu@linux.intel.com>
>> Cc: stephen@networkplumber.org; Pierre Pfister (ppfister)
>> <ppfister@cisco.com>; Xie, Huawei <huawei.xie@intel.com>; dev@dpdk.org;
>> vkaplans@redhat.com; mst@redhat.com
>> Subject: Re: [dpdk-dev] [PATCH v4] vhost: Add indirect descriptors support to the
>> TX path
>>
>>
>>
>>>>>> Hi Maxime,
>>>>>>
>>>>>> I did a little more macswap test and found out more stuff here:
>>>>> Thanks for doing more tests.
>>>>>
>>>>>>
>>>>>> 1. I did loopback test on another HSW machine with the same H/W,
>>>>>> and indirect_desc on and off seems have close perf
>>>>>>
>>>>>> 2. So I checked the gcc version:
>>>>>>
>>>>>> * Previous: gcc version 6.2.1 20160916 (Fedora 24)
>>>>>>
>>>>>> * New: gcc version 5.4.0 20160609 (Ubuntu 16.04.1 LTS)
>>>>>
>>>>> On my side, I tested with RHEL7.3:
>>>>> - gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-11)
>>>>>
>>>>> It certainly contains some backports from newer GCC versions.
>>>>>
>>>>>>
>>>>>> On previous one indirect_desc has 20% drop
>>>>>>
>>>>>> 3. Then I compiled binary on Ubuntu and scp to Fedora, and as
>>>>>> expected I got the same perf as on Ubuntu, and the perf gap
>>>>>> disappeared, so gcc is definitely one factor here
>>>>>>
>>>>>> 4. Then I use the Ubuntu binary on Fedora for PVP test, then the
>>>>>> perf gap comes back again and the same with the Fedora binary
>>>>>> results, indirect_desc causes about 20% drop
>>>>>
>>>>> Let me know if I understand correctly:
>>>
>>> Yes, and it's hard to breakdown further at this time.
>>>
>>> Also we may need to check whether it's caused by certain NIC
>>> model. Unfortunately I don't have the right setup right now.
>>>
>>>>> Loopback test with macswap:
>>>>> - gcc version 6.2.1 : 20% perf drop
>>>>> - gcc version 5.4.0 : No drop
>>>>>
>>>>> PVP test with macswap:
>>>>> - gcc version 6.2.1 : 20% perf drop
>>>>> - gcc version 5.4.0 : 20% perf drop
>>>>
>>>> I forgot to ask, did you recompile only host, or both host and guest
>>>> testmpd's in your test?
>>
>>> Both.
>>
>> I recompiled testpmd on a Fedora 24 machine using GCC6:
>> gcc (GCC) 6.1.1 20160621 (Red Hat 6.1.1-3)
>> Testing loopback with macswap on my Haswell RHEL7.3 machine gives me the
>> following results:
>> - indirect on: 7.75Mpps
>> - indirect off: 7.35Mpps
>>
>> Surprisingly, I get better results with indirect on my setup (I
>> reproduced the tests multiple times).
>>
>> Do you have a document explaining the tuning/config you apply to both
>> the host and the guest (isolation, HT, hugepage size, ...) in your
>> setup?
>
>
> The setup where it goes wrong:
> 1. Xeon E5-2699, HT on, turbo off, 1GB hugepage for both host and guest
On the Haswell machine (on which I don't have BIOS access), HT is on,
but I unplug siblings at runtime.
I also have 1G pages on both sides, and I isolate the cores used by both
testpmd and vCPUS.
> 2. Fortville 40G
> 3. Fedora 4.7.5-200.fc24.x86_64
> 4. gcc version 6.2.1
> 5. 16.11 RC2 for both host and guest
> 6. PVP, testpmd macswap for both host and guest
>
> BTW, I do see indirect_desc gives slightly better performance for loopback
> in tests on other platforms, but don't know how PVP performs yet.
Interesting, other platforms are also Haswell/Broadwell?
For PVP benchmarks, are your figures with 0% pkt loss?
Thanks,
Maxime
>
>
>>
>> Regards,
>> Maxime
next prev parent reply other threads:[~2016-11-04 12:54 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-09-23 8:28 [dpdk-dev] [PATCH v3] " Maxime Coquelin
2016-09-23 15:49 ` Michael S. Tsirkin
2016-09-23 18:02 ` Maxime Coquelin
2016-09-23 18:06 ` Michael S. Tsirkin
2016-09-23 18:16 ` Maxime Coquelin
2016-09-23 18:22 ` Michael S. Tsirkin
2016-09-23 20:24 ` Stephen Hemminger
2016-09-26 3:03 ` Yuanhan Liu
2016-09-26 12:25 ` Michael S. Tsirkin
2016-09-26 13:04 ` Yuanhan Liu
2016-09-27 4:15 ` Yuanhan Liu
2016-09-27 7:25 ` Maxime Coquelin
2016-09-27 8:42 ` [dpdk-dev] [PATCH v4] " Maxime Coquelin
2016-09-27 12:18 ` Yuanhan Liu
2016-10-14 7:24 ` Wang, Zhihong
2016-10-14 7:34 ` Wang, Zhihong
2016-10-14 15:50 ` Maxime Coquelin
2016-10-17 11:23 ` Maxime Coquelin
2016-10-17 13:21 ` Yuanhan Liu
2016-10-17 14:14 ` Maxime Coquelin
2016-10-27 9:00 ` Wang, Zhihong
2016-10-27 9:10 ` Maxime Coquelin
2016-10-27 9:55 ` Maxime Coquelin
2016-10-27 10:19 ` Wang, Zhihong
2016-10-28 7:32 ` Pierre Pfister (ppfister)
2016-10-28 7:58 ` Maxime Coquelin
2016-11-01 8:15 ` Yuanhan Liu
2016-11-01 9:39 ` Thomas Monjalon
2016-11-02 2:44 ` Yuanhan Liu
2016-10-27 10:33 ` Yuanhan Liu
2016-10-27 10:35 ` Maxime Coquelin
2016-10-27 10:46 ` Yuanhan Liu
2016-10-28 0:49 ` Wang, Zhihong
2016-10-28 7:42 ` Maxime Coquelin
2016-10-31 10:01 ` Wang, Zhihong
2016-11-02 10:51 ` Maxime Coquelin
2016-11-03 8:11 ` Maxime Coquelin
2016-11-04 6:18 ` Xu, Qian Q
2016-11-04 7:41 ` Maxime Coquelin
2016-11-04 7:20 ` Wang, Zhihong
2016-11-04 7:57 ` Maxime Coquelin
2016-11-04 7:59 ` Maxime Coquelin
2016-11-04 10:43 ` Wang, Zhihong
2016-11-04 11:22 ` Maxime Coquelin
2016-11-04 11:36 ` Yuanhan Liu
2016-11-04 11:39 ` Maxime Coquelin
2016-11-04 12:30 ` Wang, Zhihong
2016-11-04 12:54 ` Maxime Coquelin [this message]
2016-11-04 13:09 ` Wang, Zhihong
2016-11-08 10:51 ` Wang, Zhihong
2016-10-27 10:53 ` Maxime Coquelin
2016-10-28 6:05 ` Xu, Qian Q
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=75d2c094-11e8-99aa-ccb5-60ded04047d7@redhat.com \
--to=maxime.coquelin@redhat.com \
--cc=dev@dpdk.org \
--cc=huawei.xie@intel.com \
--cc=mst@redhat.com \
--cc=ppfister@cisco.com \
--cc=stephen@networkplumber.org \
--cc=vkaplans@redhat.com \
--cc=yuanhan.liu@linux.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).