From: Yuanhan Liu <yuanhan.liu@linux.intel.com>
To: Panu Matilainen <pmatilai@redhat.com>
Cc: "Yang, Zhiyong" <zhiyong.yang@intel.com>, "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH] vhost: add pmd xstats
Date: Wed, 24 Aug 2016 13:46:02 +0800 [thread overview]
Message-ID: <20160824054602.GS30752@yliu-dev.sh.intel.com> (raw)
In-Reply-To: <45bf611c-fbd5-d485-566a-5752cd58ed7c@redhat.com>
On Tue, Aug 23, 2016 at 12:45:54PM +0300, Panu Matilainen wrote:
> >>>Since collecting data of vhost_update_packet_xstats will have some
> >>>effect on RX/TX performance, so, Setting compiling switch
> >>>CONFIG_RTE_LIBRTE_PMD_VHOST_UPDATE_XSTATS=n by default in the
> >>file
> >>>config/common_base, if needing xstats data, you can enable it(y).
> >>
> >>NAK, such things need to be switchable at run-time.
> >>
> >> - Panu -
> >
> >Considering the following reasons using the compiler switch, not
> >command-line at run-time.
> >
> >1.Similar xstats update functions are always collecting stats data in the
> >background when rx/tx are running, such as the physical NIC or virtio,
> >which have no switch. Compiler switch for vhost pmd xstats is added
> >as a option when performance is viewed as critical factor.
> >
> >2. No data structure and API in any layer support the xstats update switch
> >at run-time. Common data structure (struct rte_eth_dev_data) has no
> >device-specific data member, if implementing enable/disable of vhost_update
> >_packet_xstats at run-time, must define a flag(device-specific) in it,
> >because the definition of struct vhost_queue in the driver code
> >(eth_vhost_rx/eth_vhost_tx processing)is not visible from device perspective.
> >
> >3. I tested RX/TX with v1 patch (y) as reference based on Intel(R)
> >Xeon(R) CPU E5-2699 v3 @ 2.30GHz, for 64byts packets in burst mode, 32 packets
> >in one RX/TX processing. Overhead of vhost_update_packet_xstats is less than
> >3% for the rx/tx processing. It looks that vhost_update_packet_xstats has a
> >limited effect on performance drop.
>
> Well, either the performance overhead is acceptable and it should always be
> on (like with physical NICs I think). Or it is not. In which case it needs
> to be turnable on and off, at run-time. Rebuilding is not an option in the
> world of distros.
I think the less than 3% overhead is acceptable here, that I agree with
Panu we should always keep it on. If someone compains it later that even
3% is too big for them, let's consider to make it be switchable at
run-time. Either we could introduce a generic eth API for that, Or
just introduce a vhost one if that doesn't make too much sense to other
eth drivers.
--yliu
next prev parent reply other threads:[~2016-08-24 5:36 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-08-19 12:16 Zhiyong Yang
2016-08-22 7:52 ` Panu Matilainen
2016-08-23 8:04 ` Yang, Zhiyong
2016-08-23 9:45 ` Panu Matilainen
2016-08-24 5:46 ` Yuanhan Liu [this message]
2016-08-24 8:44 ` Thomas Monjalon
2016-08-24 12:37 ` Panu Matilainen
2016-08-25 9:21 ` Yang, Zhiyong
2016-08-30 2:45 ` Yao, Lei A
2016-08-30 3:03 ` Xu, Qian Q
2016-08-30 3:21 ` Yao, Lei A
2016-08-31 7:18 ` Yang, Zhiyong
2016-09-01 6:37 ` Yang, Zhiyong
2016-09-09 8:15 ` [dpdk-dev] [PATCH v2] net/vhost: " Zhiyong Yang
2016-09-09 8:40 ` Van Haaren, Harry
2016-09-09 8:54 ` Yang, Zhiyong
2016-09-14 6:20 ` Yuanhan Liu
2016-09-14 7:43 ` Yang, Zhiyong
2016-09-18 13:16 ` Yuanhan Liu
2016-09-19 2:48 ` Yang, Zhiyong
2016-09-14 8:30 ` Yang, Zhiyong
2016-09-20 9:36 ` [dpdk-dev] [PATCH v3 0/2] net/vhost: add pmd xstats support Zhiyong Yang
2016-09-20 9:36 ` [dpdk-dev] [PATCH v3 1/2] net/vhost: add a new stats struct Zhiyong Yang
2016-09-20 10:44 ` Yuanhan Liu
2016-09-21 5:12 ` Yang, Zhiyong
2016-09-21 10:05 ` [dpdk-dev] [PATCH v4 0/2] net/vhost: add pmd xstats support Zhiyong Yang
2016-09-21 10:05 ` Zhiyong Yang
2016-09-21 10:05 ` [dpdk-dev] [PATCH v4 2/2] net/vhost: add pmd xstats Zhiyong Yang
2016-09-21 10:57 ` Yuanhan Liu
2016-09-22 1:42 ` Yang, Zhiyong
2016-09-22 2:09 ` Yuanhan Liu
2016-09-21 10:13 ` [dpdk-dev] [PATCH v4 1/2] net/vhost: add a new defined stats struct Zhiyong Yang
2016-09-22 8:19 ` [dpdk-dev] [PATCH v5 0/2] net/vhost: add pmd xstats support Zhiyong Yang
2016-09-22 8:19 ` [dpdk-dev] [PATCH v5 1/2] net/vhost: add a new defined stats struct Zhiyong Yang
2016-09-28 8:33 ` [dpdk-dev] [PATCH v6 0/2] net/vhost: add pmd xstats support Zhiyong Yang
2016-09-28 8:33 ` [dpdk-dev] [PATCH v6 1/2] net/vhost: add a new defined stats struct Zhiyong Yang
2016-09-28 13:26 ` [dpdk-dev] [PATCH v7 0/2] net/vhost: add pmd xstats support Zhiyong Yang
2016-09-28 13:26 ` [dpdk-dev] [PATCH v7 1/2] net/vhost: add a new defined stats struct Zhiyong Yang
2016-09-29 1:55 ` Yuanhan Liu
2016-09-29 12:35 ` [dpdk-dev] [PATCH v8 0/2] net/vhost: add pmd xstats support Zhiyong Yang
2016-09-29 12:35 ` [dpdk-dev] [PATCH v8 1/2] net/vhost: add a new defined stats struct Zhiyong Yang
2016-09-29 12:35 ` [dpdk-dev] [PATCH v8 2/2] net/vhost: add pmd xstats Zhiyong Yang
2016-09-29 13:04 ` Yuanhan Liu
2016-09-29 13:50 ` Yang, Zhiyong
2016-09-28 13:26 ` [dpdk-dev] [PATCH v7 " Zhiyong Yang
2016-09-29 8:48 ` Loftus, Ciara
2016-09-29 12:02 ` Yuanhan Liu
2016-09-29 12:22 ` Yang, Zhiyong
2016-09-28 8:33 ` [dpdk-dev] [PATCH v6 " Zhiyong Yang
2016-09-22 8:19 ` [dpdk-dev] [PATCH v5 " Zhiyong Yang
2016-09-23 3:56 ` [dpdk-dev] [PATCH v5 0/2] net/vhost: add pmd xstats support Yuanhan Liu
2016-09-28 2:35 ` Yuanhan Liu
2016-09-20 9:36 ` [dpdk-dev] [PATCH v3 2/2] net/vhost: add pmd xstats Zhiyong Yang
2016-09-20 10:56 ` Yuanhan Liu
2016-09-21 7:22 ` Yang, Zhiyong
2016-09-20 11:50 ` Yuanhan Liu
2016-09-21 6:15 ` Yang, Zhiyong
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=20160824054602.GS30752@yliu-dev.sh.intel.com \
--to=yuanhan.liu@linux.intel.com \
--cc=dev@dpdk.org \
--cc=pmatilai@redhat.com \
--cc=zhiyong.yang@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).