From: Yuanhan Liu <yuanhan.liu@linux.intel.com>
To: Robert Wojciechowicz <robertx.wojciechowicz@intel.com>
Cc: stable@dpdk.org, "Mcnamara, John" <john.mcnamara@intel.com>,
"Gray, Mark D" <mark.d.gray@intel.com>,
"Devlin, Michelle" <michelle.devlin@intel.com>,
"Weglicki, MichalX" <michalx.weglicki@intel.com>,
Thomas Monjalon <thomas.monjalon@6wind.com>
Subject: Re: [dpdk-stable] vhost: many vhost user ports (backporting to v16.11)
Date: Tue, 14 Feb 2017 10:41:47 +0800 [thread overview]
Message-ID: <20170214024147.GQ20916@yliu-dev.sh.intel.com> (raw)
In-Reply-To: <20170209125628.GA10377@debian>
On Thu, Feb 09, 2017 at 07:56:29AM -0500, Robert Wojciechowicz wrote:
> Hi,
>
> in DPDK v17.02-rc2 there is available following patch:
>
> commit: 59317cef249c0b23098543df527b3e360ce9764f
>
> """
> vhost: allow many vhost-user ports
> Currently select() is used to monitor file descriptors for vhostuser
> ports. This limits the number of ports possible to create since the
> fd number is used as index in the fd_set and we have seen fds > 1023.
> This patch changes select() to poll(). This way we can keep an
> packed (pollfd) array for the fds, e.g. as many fds as the size of
> the array.
> Also see:
> http://dpdk.org/ml/archives/dev/2016-April/037024.html
> Reported-by: Patrik Andersson <patrik.r.andersson at ericsson.com>
> Signed-off-by: Jan Wickbom <jan.wickbom at ericsson.com>
> Signed-off-by: Yuanhan Liu <yuanhan.liu at linux.intel.com>
> """
>
> I tested this feature with OVS and DPDK v17.02-rc2.
> I applied also this patch to DPDK v16.11 without any issues
> and performed the same tests as for v17.02-rc2.
> It solves the problem with connecting many VMs to one vSwitch
> instance (in my tests 70 VMs) with vhost user ports.
>
> Is it possible to backport this feature to DPDK v16.11?
As a vhost maintainer, I think this patch could be backported to v16.11:
it resolves an issue you met.
Just let me know if anyone has objections.
> It is also worth pointing out that OVS 2.7 will use 16.11
One question though: will they use v16.11.1 or something like that?
--yliu
> which means we will have this bug for the next 6 months at least
> unless we can backport.
>
> What's your opinion?
>
> Br,
> Robert
next prev parent reply other threads:[~2017-02-14 2:39 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-02-09 12:56 Robert Wojciechowicz
2017-02-14 2:41 ` Yuanhan Liu [this message]
2017-02-14 8:50 ` Wojciechowicz, RobertX
2017-02-14 9:19 ` Gray, Mark D
2017-02-15 1:19 ` Yuanhan Liu
2017-02-15 14:57 ` Gray, Mark D
2017-02-22 15:28 ` Gray, Mark D
2017-02-23 3:51 ` Yuanhan Liu
2017-02-23 14:56 ` Gray, Mark D
2017-02-23 14:57 ` Stokes, Ian
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=20170214024147.GQ20916@yliu-dev.sh.intel.com \
--to=yuanhan.liu@linux.intel.com \
--cc=john.mcnamara@intel.com \
--cc=mark.d.gray@intel.com \
--cc=michalx.weglicki@intel.com \
--cc=michelle.devlin@intel.com \
--cc=robertx.wojciechowicz@intel.com \
--cc=stable@dpdk.org \
--cc=thomas.monjalon@6wind.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).