date:20170830

Re: [vpp-dev] multiple ips for an interface

2017-08-30 Thread Pragash Vijayaragavan

Hi Ole,

Is there any cli command for source address selection in the vpp?

Thanks,

Pragash Vijayaragavan
Grad Student at Rochester Institute of Technology
email : pxv3...@rit.edu
ph : 585 764 4662


On Wed, Aug 30, 2017 at 4:14 AM, Ole Troan  wrote:

> Pragash,
>
> > I have a quick question, i am able to assign multiple ips for an
> interface in vpp. Is this correct behavior.
>
> Yes. Shouldn't it be?
> Note that VPP itself doesn't implement source address selection so the
> local application would have to deal with that.
>
> Best regards,
> Ole
>
>
> >
> > pragash@revvit:~/VPP$ sudo vppctl sh interfaces address
> > FortyGigabitEthernet2/0/0 (up):
> >   10.1.1.1/24
> >   10.2.1.1/24
> >   2001::1/48
> >   fe80::3efd:feff:fea6:8ef0/128
> >   2001::2/48
> >   2003::2/48
> >   2004::2/48
> > FortyGigabitEthernet4/0/0 (up):
> >   2002::1/48
> >   fe80::3efd:feff:fea6:8e78/128
> >   9001::2/48
> > local0 (dn):
> >
> >
> >
> > Thanks,
> >
> > Pragash Vijayaragavan
> > Grad Student at Rochester Institute of Technology
> > email : pxv3...@rit.edu
> > ph : 585 764 4662
> >
> > ___
> > vpp-dev mailing list
> > vpp-dev@lists.fd.io
> > https://lists.fd.io/mailman/listinfo/vpp-dev
>
>
___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

[vpp-dev] ip6 routing issue

2017-08-30 Thread Pragash Vijayaragavan

Hi,

We are sending ip6 traffic as below, but traffic is not forwarded to the
destination port.

-  [  Traffic gen  (dst ip - port 2) ]---[ (port 1)
vpp (port 2) ]

We tried to ping from port 2 from port 1, 98 percent packet loss.

Can someone let us know whether we are missing any configuration.
I can give more outputs if needed.


vpp# sh interfaces addr
FortyGigabitEthernet2/0/0 (up):
  2002::1/48
  fe80::3efd:feff:fea6:8ef0/128
FortyGigabitEthernet4/0/0 (up):
  9002::1/48
  fe80::3efd:feff:fea6:8e78/128
local0 (up):

vpp# ping 9002::1 source FortyGigabitEthernet2/0/0
5 bytes from 9002::1: icmp_seq=0 ttl=64 time=20630217.5218 ms
5 bytes from 9002::1: icmp_seq=0 ttl=64 time=20631217.5077 ms
5 bytes from 9002::1: icmp_seq=0 ttl=64 time=20632217.5011 ms
5 bytes from 9002::1: icmp_seq=0 ttl=64 time=20633217.5006 ms
5 bytes from 9002::1: icmp_seq=0 ttl=64 time=20634220.4604 ms
5 bytes from 9002::1: icmp_seq=0 ttl=64 time=20634220.4636 ms

Statistics: 300 sent, 6 received, 98% packet loss



Thanks,

Pragash Vijayaragavan
Grad Student at Rochester Institute of Technology
email : pxv3...@rit.edu
ph : 585 764 4662
___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

[vpp-dev] VPP Communications Library demo instructions

2017-08-30 Thread Keith Burns

Folks,

Updated README.md in VPPSB for VCL-preload.

Also put a version on the VPP wiki:
https://wiki.fd.io/view/VPP/VPPCommunicationsLibrary

Enjoy!
___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

2017-08-30 Thread Thomas F Herbert




On 08/30/2017 01:57 PM, Gabriel Ganne wrote:


https://gerrit.fd.io/r/#/c/8260/

It adds bc to the dependency list, but also replaces a $(shell $(echo 
…)) by $(shell `echo …`)


They should be the same, but it seems to change the way the Makefile 
variables are interpreted.


The problem I saw was on Fedora and Centos both of which have bc. The 
problem it needs a double $
to escape the "$()" from make because it "thinks" it is a reference 
to a variable.


commit 14afc64629e9b35a2e5c5941232236a78c2ecd75
Author: Thomas F Herbert 
Date:   Wed Aug 30 10:13:51 2017 -0400

Fix shell error.

Change-Id: I06af51eef20c2191199613f951f569ef1727b9c4
Signed-off-by: Thomas F Herbert 

diff --git a/Makefile b/Makefile
index 1548f36..ef92e2f 100644
--- a/Makefile
+++ b/Makefile
@@ -78,7 +78,7 @@ ifeq ($(OS_ID)-$(OS_VERSION_ID),fedora-25)
RPM_DEPENDS += python-devel
RPM_DEPENDS += python2-virtualenv
RPM_DEPENDS_GROUPS = 'C Development Tools and Libraries'
-else ifeq ($(shell if [ $(echo "$(OS_VERSION_ID) > 25" | bc) -eq 1 ] ; 
then echo "y" ; fi),"y")
+else ifeq ($(shell if [ $$(echo "$(OS_VERSION_ID) > 25" | bc) -eq 1 ] ; 
then echo "y" ; fi),"y")

RPM_DEPENDS += python2-devel
RPM_DEPENDS += python2-virtualenv
RPM_DEPENDS_GROUPS = 'C Development Tools and Libraries'


I tested on rhel 7, Ubuntu 16.04, and debian 8.

Best regards,

--

Gabriel Ganne

*From:*Dave Wallace [mailto:dwallac...@gmail.com]
*Sent:* mercredi 30 août 2017 16:27
*To:* Burt Silverman ; Gabriel Ganne 

*Cc:* Thomas F Herbert ; vpp-dev 


*Subject:* Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

I agree.

@Gabriel, please push a patch which adds bc to DEB_DEPENDS in 
vpp/Makefile.


Thanks,
-daw-

On 08/30/2017 09:14 AM, Burt Silverman wrote:

To me, it doesn't seem to be a crime to add bc to the
dependencies. I guess another approach would be to remove the dot
in 16dot04 and then just use shell arithmetic. The release numbers
are always in the same format, 2 digits DOT 2 digits, so I would
think that should work.

Burt

On Wed, Aug 30, 2017 at 3:47 AM, Gabriel Ganne
> wrote:

Hi,

This probably is be because you don't have bc.

It is not in the dependency list. I'm so used to having it
around that I did not think to check.

Sorry.

If so, the best thing probably is to revert and not to
increase the dependency list just to silence a warning.

Regards,

--

Gabriel Ganne



*From:*Dave Wallace >
*Sent:* Wednesday, August 30, 2017 6:25:32 AM
*To:* Thomas F Herbert; Gabriel Ganne
*Cc:* vpp-dev
*Subject:* Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/




Tom,

What OS are you running?

Thanks,
-daw-

On 08/29/2017 07:58 PM, Thomas F Herbert wrote:

This patch creates may have fixed one problem with Ubuntu
16.04 but created another:

$ make
/bin/sh: line 0: [: -eq: unary operator expected

...

-- 
*Thomas F Herbert*

NFV and Fast Data Planes
Office of Technology
*Red Hat*



___

vpp-dev mailing list

vpp-dev@lists.fd.io 

https://lists.fd.io/mailman/listinfo/vpp-dev




___
vpp-dev mailing list
vpp-dev@lists.fd.io 
https://lists.fd.io/mailman/listinfo/vpp-dev





--
*Thomas F Herbert*
NFV and Fast

Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

2017-08-30 Thread Dave Wallace


Thanks Gabriel, I'll merge it when the Verify job completes.

-daw-

On 08/30/2017 01:57 PM, Gabriel Ganne wrote:


https://gerrit.fd.io/r/#/c/8260/

It adds bc to the dependency list, but also replaces a $(shell $(echo 
…)) by $(shell `echo …`)


They should be the same, but it seems to change the way the Makefile 
variables are interpreted.


I tested on rhel 7, Ubuntu 16.04, and debian 8.

Best regards,

--

Gabriel Ganne

*From:*Dave Wallace [mailto:dwallac...@gmail.com]
*Sent:* mercredi 30 août 2017 16:27
*To:* Burt Silverman ; Gabriel Ganne 

*Cc:* Thomas F Herbert ; vpp-dev 


*Subject:* Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

I agree.

@Gabriel, please push a patch which adds bc to DEB_DEPENDS in 
vpp/Makefile.


Thanks,
-daw-

On 08/30/2017 09:14 AM, Burt Silverman wrote:

To me, it doesn't seem to be a crime to add bc to the
dependencies. I guess another approach would be to remove the dot
in 16dot04 and then just use shell arithmetic. The release numbers
are always in the same format, 2 digits DOT 2 digits, so I would
think that should work.

Burt

On Wed, Aug 30, 2017 at 3:47 AM, Gabriel Ganne
> wrote:

Hi,

This probably is be because you don't have bc.

It is not in the dependency list. I'm so used to having it
around that I did not think to check.

Sorry.

If so, the best thing probably is to revert and not to
increase the dependency list just to silence a warning.

Regards,

--

Gabriel Ganne



*From:*Dave Wallace >
*Sent:* Wednesday, August 30, 2017 6:25:32 AM
*To:* Thomas F Herbert; Gabriel Ganne
*Cc:* vpp-dev
*Subject:* Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/




Tom,

What OS are you running?

Thanks,
-daw-

On 08/29/2017 07:58 PM, Thomas F Herbert wrote:

This patch creates may have fixed one problem with Ubuntu
16.04 but created another:

$ make
/bin/sh: line 0: [: -eq: unary operator expected

...

-- 
*Thomas F Herbert*

NFV and Fast Data Planes
Office of Technology
*Red Hat*



___

vpp-dev mailing list

vpp-dev@lists.fd.io 

https://lists.fd.io/mailman/listinfo/vpp-dev




___
vpp-dev mailing list
vpp-dev@lists.fd.io 
https://lists.fd.io/mailman/listinfo/vpp-dev





___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

2017-08-30 Thread Gabriel Ganne

https://gerrit.fd.io/r/#/c/8260/

It adds bc to the dependency list, but also replaces a $(shell $(echo …)) by 
$(shell `echo …`)
They should be the same, but it seems to change the way the Makefile variables 
are interpreted.

I tested on rhel 7, Ubuntu 16.04, and debian 8.

Best regards,

--
Gabriel Ganne

From: Dave Wallace [mailto:dwallac...@gmail.com]
Sent: mercredi 30 août 2017 16:27
To: Burt Silverman ; Gabriel Ganne 
Cc: Thomas F Herbert ; vpp-dev 
Subject: Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

I agree.

@Gabriel, please push a patch which adds bc to DEB_DEPENDS in vpp/Makefile.

Thanks,
-daw-
On 08/30/2017 09:14 AM, Burt Silverman wrote:
To me, it doesn't seem to be a crime to add bc to the dependencies. I guess 
another approach would be to remove the dot in 16dot04 and then just use shell 
arithmetic. The release numbers are always in the same format, 2 digits DOT 2 
digits, so I would think that should work.

Burt

On Wed, Aug 30, 2017 at 3:47 AM, Gabriel Ganne 
> wrote:

Hi,



This probably is be because you don't have bc.

It is not in the dependency list. I'm so used to having it around that I did 
not think to check.

Sorry.



If so, the best thing probably is to revert and not to increase the dependency 
list just to silence a warning.



Regards,



--

Gabriel Ganne


From: Dave Wallace >
Sent: Wednesday, August 30, 2017 6:25:32 AM
To: Thomas F Herbert; Gabriel Ganne
Cc: vpp-dev
Subject: Re: [vpp-dev] 
https://gerrit.fd.io/r/#/c/8236/

Tom,

What OS are you running?

Thanks,
-daw-
On 08/29/2017 07:58 PM, Thomas F Herbert wrote:

This patch creates may have fixed one problem with Ubuntu 16.04 but created 
another:

$ make
/bin/sh: line 0: [: -eq: unary operator expected

...


--
Thomas F Herbert
NFV and Fast Data Planes
Office of Technology
Red Hat



___

vpp-dev mailing list

vpp-dev@lists.fd.io

https://lists.fd.io/mailman/listinfo/vpp-dev


___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev


___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Query for IPSec support on VPP

2017-08-30 Thread Mukesh Yadav (mukyadav)

HI Sergio,

I haven’t done any extra config besides mentioned in two scenario.

1. 

dpdk {
socket-mem 1024,1024
num-mbufs 131072
dev :00:08.0
}
/usr/bin/vpp[3844]: dpdk_ipsec_process:239: not enough Cryptodevs, default to 
OpenSSL IPsec

2. 
dpdk {
socket-mem 1024,1024
num-mbufs 131072
dev :00:08.0
   vdev cryptodev_aesni_mb_pmd,socket_id=1
vdev cryptodev_aesni_mb_pmd,socket_id=1
}

CRYPTODEV: [crypto_aesni_mb] cryptodev_aesni_mb_create() line 704: failed to 
create cryptodev vdev
CRYPTODEV: [crypto_aesni_mb] cryptodev_aesni_mb_create() line 746: driver 
cryptodev_aesni_mb_pmd: cryptodev_aesni_create failed
EAL: failed to initialize cryptodev_aesni_mb_pmd device


I am not aware of significance of various paramters mentioned above. Just 
trying from reading DPDK documentation.
Will be helpful if you can point me some sample config for DPDK IPSec or any 
link which shall have these pointers.

Thanks
Mukesh

On 30/08/17, 9:14 PM, "Sergio Gonzalez Monroy" 
 wrote:

I was just reproducing your steps.

Check that you have created a 'vdev crypto_aeasni_mb0' in 
/etc/vpp/startup.conf inside the 'dpdk' section.

Then to check if you have support, you can run:
vppctl show crypto device mapping verbose

Thanks,
Sergio

On 30/08/2017 13:07, Mukesh Yadav (mukyadav) wrote:
>
> HI Sergio,
>
> I am using VPP version 17.10
>
> vpp# show ver
>
> vpp v17.10-rc0~103-g42e6b09 built by vagrant on localhost at Sun Aug 
> 27 22:06:20 PDT 2017
>
> I am cloned vpp on my mac machine using below command
>
> git clone https://gerrit.fd.io/r/vpp
>
> Then I have ran vagrant up/vagrant ssh
>
> Referred: 
> 
https://wiki.fd.io/view/VPP/Pulling,_Building,_Running,_Hacking_and_Pushing_VPP_Code
>
> On vagrant ssh, I have started vpp service.
>
> Note: I have ran scenario mentioned on below links properly.
>
> https://wiki.fd.io/view/VPP/IPSec_and_IKEv2
>
> Only issue is coming when trying to use DPDK.
>
> Once/twice I have ran below as well after vagrant ssh. But haven’t 
> made any code changes in VPP Source.
>
> cd /vpp
>
> make wipe-release
>
> make install-dep
>
> make build-release (or make build)
>
> make pkg-deb
>
> sudo dpkg -i /vpp/build-root/*.deb
>
> sudo service vpp start
>
> Thanks
>
> Mukesh
>
> *From: *Sergio Gonzalez Monroy 
> *Date: *Wednesday, 30 August 2017 at 4:18 PM
> *To: *Mukesh Yadav , "vpp-dev@lists.fd.io" 
> 
> *Subject: *Re: [vpp-dev] Query for IPSec support on VPP
>
> Which VPP are you using?
>
> If you are using/installing VPP package (rpm/deb), then it is likely 
> that you do not have DPDK SW crypto support.
>
> As a summary, if you want DPDK SW crypto support:
> - VPP package: 17.07.1 (AFAIK not released yet) or newer.
> - From source:
> - 17.04 or 17.07: need to build VPP with DPDK SW crypto support
>  make build-release vpp_uses_dpdk_cryptodev_sw=yes
> - 17.07.1 or newer:  DPDK SW crypto support built by default
>  make build-release
>
> Thanks,
> Sergio
>
> On 30/08/2017 10:57, Mukesh Yadav (mukyadav) wrote:
>
> Thanks Sergio.
>
> I tried below combinations of dpdk config.
>
> dpdk {
>
>   socket-mem 1024,1024
>
>   num-mbufs 131072
>
>   dev :00:08.0
>
> #   enable-cryptodev
>
> #   vdev cryptodev_aesni_mb_pmd,socket_id=1
>
> #   vdev cryptodev_aesni_mb_pmd,socket_id=1
>
> }
>
> I get below error:
>
> /usr/bin/vpp[3844]: dpdk_ipsec_process:239: not enough Cryptodevs,
> default to OpenSSL IPsec
>
> dpdk {
>
>   socket-mem 1024,1024
>
>   num-mbufs 131072
>
>   dev :00:08.0
>
> #   enable-cryptodev
>
>   vdev cryptodev_aesni_mb_pmd,socket_id=1
>
>   vdev cryptodev_aesni_mb_pmd,socket_id=1
>
> }
>
> CRYPTODEV: [crypto_aesni_mb] cryptodev_aesni_mb_create() line 704:
> failed to create cryptodev vdev
>
> CRYPTODEV: [crypto_aesni_mb] cryptodev_aesni_mb_create() line 746:
> driver cryptodev_aesni_mb_pmd: cryptodev_aesni_create failed
>
> EAL: failed to initialize cryptodev_aesni_mb_pmd device
>
> Seems still config is missing something.
>
> Can you point me out any working wiki or document for making IPSec
> work along DPDK in VPP.
>
> Thanks
>
> Mukesh
>
> *From: *Sergio

Re: [vpp-dev] Dynamically change number of cores used by VPP?

2017-08-30 Thread Damjan Marion (damarion)

On 30 Aug 2017, at 13:59, Tobias Sundqvist 
> wrote:

Hi I guess the silence tells that there are no way of dynamically scale the 
number of cores used by VPP when VPP already has started.
If anyone has any idea if it is dynamically possible to change the way the 
cores is used just send me an email. Perhaps it could be possible to assign a 
certain core to not poll the interface or something similar, then the core 
would still be in use of VPP but the CPU load would go down.

BR /Tobias

On 24 August 2017 at 10:36, Tobias Sundqvist 
> wrote:
Hi we are building an application that uses vpp and when I have some questions 
that concerns multi core usage.

In the run that we are doing now we are using 4 cores and we use some DPDK 
polling of an interface which makes the cpu load go 100% on all cores. We are 
interested in energy consumption in the application we are creating and would 
like to be able to scale up and down the cpu usage during runtime.

Is it possible to dynamically change the number of cores used by VPP or specify 
somehow which cores that are used for polling of an certain interface during 
runtime?

During low and high traffic we would like to scale up and down the amount of 
cores that is used.

If the polling is causing the cpu to go to 100% would it help using Turboboost 
and Speedstep, would it actually lower the frequency when the traffic is low?

BR /Tobias

I did some work to enable interrupt mode with DPDK devices, but it is a bit 
hacky (it is digging some data form dpdk internal structures)
so it is not published. Together with adaptive interrupt/polling mode which VPP 
supports it might be solution to your problem.

Please note that you can define today which worker is poling which interface, 
so you can effectively remove all interfaces form  specific core ant that core 
should go to sleep.
This works on runtime without the need to switch interface off so it can also 
be a way to address your problem.

See “set interface rx-placement / show interface rx-placement” commands for 
details…

Thanks,

Damjan

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Query for IPSec support on VPP

2017-08-30 Thread Sergio Gonzalez Monroy


I was just reproducing your steps.

Check that you have created a 'vdev crypto_aeasni_mb0' in 
/etc/vpp/startup.conf inside the 'dpdk' section.


Then to check if you have support, you can run:
vppctl show crypto device mapping verbose

Thanks,
Sergio

On 30/08/2017 13:07, Mukesh Yadav (mukyadav) wrote:


HI Sergio,

I am using VPP version 17.10

vpp# show ver

vpp v17.10-rc0~103-g42e6b09 built by vagrant on localhost at Sun Aug 
27 22:06:20 PDT 2017


I am cloned vpp on my mac machine using below command

git clone https://gerrit.fd.io/r/vpp

Then I have ran vagrant up/vagrant ssh

Referred: 
https://wiki.fd.io/view/VPP/Pulling,_Building,_Running,_Hacking_and_Pushing_VPP_Code


On vagrant ssh, I have started vpp service.

Note: I have ran scenario mentioned on below links properly.

https://wiki.fd.io/view/VPP/IPSec_and_IKEv2

Only issue is coming when trying to use DPDK.

Once/twice I have ran below as well after vagrant ssh. But haven’t 
made any code changes in VPP Source.


cd /vpp

make wipe-release

make install-dep

make build-release (or make build)

make pkg-deb

sudo dpkg -i /vpp/build-root/*.deb

sudo service vpp start

Thanks

Mukesh

*From: *Sergio Gonzalez Monroy 
*Date: *Wednesday, 30 August 2017 at 4:18 PM
*To: *Mukesh Yadav , "vpp-dev@lists.fd.io" 


*Subject: *Re: [vpp-dev] Query for IPSec support on VPP

Which VPP are you using?

If you are using/installing VPP package (rpm/deb), then it is likely 
that you do not have DPDK SW crypto support.


As a summary, if you want DPDK SW crypto support:
- VPP package: 17.07.1 (AFAIK not released yet) or newer.
- From source:
- 17.04 or 17.07: need to build VPP with DPDK SW crypto support
 make build-release vpp_uses_dpdk_cryptodev_sw=yes
- 17.07.1 or newer:  DPDK SW crypto support built by default
 make build-release

Thanks,
Sergio

On 30/08/2017 10:57, Mukesh Yadav (mukyadav) wrote:

Thanks Sergio.

I tried below combinations of dpdk config.

dpdk {

  socket-mem 1024,1024

  num-mbufs 131072

  dev :00:08.0

#   enable-cryptodev

#   vdev cryptodev_aesni_mb_pmd,socket_id=1

#   vdev cryptodev_aesni_mb_pmd,socket_id=1

}

I get below error:

/usr/bin/vpp[3844]: dpdk_ipsec_process:239: not enough Cryptodevs,
default to OpenSSL IPsec

dpdk {

  socket-mem 1024,1024

  num-mbufs 131072

  dev :00:08.0

#   enable-cryptodev

  vdev cryptodev_aesni_mb_pmd,socket_id=1

  vdev cryptodev_aesni_mb_pmd,socket_id=1

}

CRYPTODEV: [crypto_aesni_mb] cryptodev_aesni_mb_create() line 704:
failed to create cryptodev vdev

CRYPTODEV: [crypto_aesni_mb] cryptodev_aesni_mb_create() line 746:
driver cryptodev_aesni_mb_pmd: cryptodev_aesni_create failed

EAL: failed to initialize cryptodev_aesni_mb_pmd device

Seems still config is missing something.

Can you point me out any working wiki or document for making IPSec
work along DPDK in VPP.

Thanks

Mukesh

*From: *Sergio Gonzalez Monroy 

*Date: *Wednesday, 30 August 2017 at 2:06 PM
*To: *Mukesh Yadav 
, "vpp-dev@lists.fd.io"
 

*Subject: *Re: [vpp-dev] Query for IPSec support on VPP

Just for reference and clarity, there is no wiki entry for the
DPDK crypto, only documentation page.
The wiki page is for general VPP/IPSec/IKEv2.

Thanks,
Sergio

On 30/08/2017 09:19, Sergio Gonzalez Monroy wrote:

On 29/08/2017 18:25, Mukesh Yadav (mukyadav) wrote:

Thanks Sergio for quick help. Pointers cleared few of
basic doubts.


You are flushing out few left overs from previous releases, so
thank you.



I will use below wiki to proceed.

https://docs.fd.io/vpp/17.10/dpdk_crypto_ipsec_doc.html

Since vpp version is 17.10

I added below in /vpp/src/vpp/conf/startup.conf

dpdk {

  socket-mem 1024,1024

  num-mbufs 131072

  dev :00:08.0

  enable-cryptodev

  vdev cryptodev_aesni_mb_pmd,socket_id=1

  vdev cryptodev_aesni_mb_pmd,socket_id=1

}

And did a restart of vpp service using “sudo service vpp
restart”. It didn’t restarted service.

Till now I was using vanilla config which came with Source
where dpdk config section was commented.

Next I tried below.

/usr/bin/vpp -c /etc/vpp/startup.conf

I can see logs on command terminal where above command is ran:

load_one_plugin:63: Loaded

Re: [vpp-dev] About the order of VLIB_INIT_FUNCTION called between different plugins

2017-08-30 Thread Damjan Marion (damarion)

Yes, also please note that you can at any time use vlib_get_plugin_symbol(..) 
function to get pointer to symbol in another plugin. If you get NULL then 
another plugin is not loaded.

So something like this should work, assuming that you want to go that way...

static clib_error_t *
bar_init (vlib_main_t * vm)
 {
clib_error_t *error = 0;
if (vlib_get_plugin_symbol (“foo_plugin.so”, “foo_init”) == 0)
{
  clib_warning ( “foo plugin not loaded. bar disabled”);
  bar_main.disabled = 1;
  return 0;
}

if ((error = vlib_call_init_function (vm, foo_init)))
 return error;

  // continue with bar init…
}



> On 30 Aug 2017, at 12:37, Dave Barach (dbarach)  wrote:
> 
> Explicit dependencies between plugins is probably not a good idea. There is 
> little to guarantee that both A and B will be loaded.
>  
> Please describe the use-case in more detail.  
>  
> Thanks… Dave
>  
> From: vpp-dev-boun...@lists.fd.io [mailto:vpp-dev-boun...@lists.fd.io] On 
> Behalf Of wang.hu...@zte.com.cn
> Sent: Wednesday, August 30, 2017 4:01 AM
> To: vpp-dev@lists.fd.io
> Cc: zhao.qingl...@zte.com.cn; wu.bi...@zte.com.cn; gu.ji...@zte.com.cn; 
> dong.ju...@zte.com.cn
> Subject: [vpp-dev] About the order of  VLIB_INIT_FUNCTION called between 
> different plugins
>  
> Hi all:
> 
> How to control the order of  VLIB_INIT_FUNCTION (user xxx_init function) 
> called between Different plugins?
> 
> It depends on plugin name?or the sequence of loading plugin ?
> 
>  or is there any other way to adjust the order?
> 
>  
> 
> Thanks~
> 
>  
> 
>  
> 
>  
> 
> 王辉 wanghui
> 
>  
> 
> IT开发工程师 IT Development Engineer
> 虚拟化南京四部/无线研究院/无线产品经营部 NIV Nanjing Dept. IV/Wireless Product R＆D 
> Institute/Wireless Product Operation Division
> 
>  
> 
> ___
> vpp-dev mailing list
> vpp-dev@lists.fd.io
> https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Issue forwarding TCP packets

2017-08-30 Thread Dave Barach (dbarach)

This is a system problem; vpp can’t solve it all by itself.

When forwarding packets at L2, vpp doesn’t look past the ethernet header. It’s 
simply delivering packets generated by the Linux kernel on one interface to 
another Linux kernel interface.

The kernel cheats by not generating L4 checksums. When we deliver such packets 
to a Linux kernel interface which isn’t part of a Linux bridge, it throws them 
on the floor. Linux bridging winks and nods, knowing that the L4 checksum won’t 
be set.

Two options: compute the checksums in software, or ignore them across the 
board. Neither option is perfect. The checksum computation is slow. I’m not a 
big believer in checksums, to be honest, but ignoring the L4 checksum across 
the board is probably not acceptable.

Thanks… Dave

From: vpp-dev-boun...@lists.fd.io [mailto:vpp-dev-boun...@lists.fd.io] On 
Behalf Of Prabhjot Singh Sethi
Sent: Wednesday, August 30, 2017 9:45 AM
To: Florin Coras ; Prabhjot Singh Sethi 

Cc: vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] Issue forwarding TCP packets

Hi Florin,
For now we can proceed with a requirement to disable offload on VMs. but still 
keeping performance in view i think VPP should have option similar to linux 
bridge to be able to do forwarding without correct checksums, while offloading 
is enable the real checksum calculations should be done only when the packet is 
egressing out of the server to other compute nodes or gateway.

i think we can leave the topic open for now, and it can be considered later for 
performance improvements for TCP

Regards,
Prabhjot

- Original Message -
From:
"Florin Coras" >

To:
"Prabhjot Singh Sethi" 
>.
Cc:
>
Sent:
Tue, 29 Aug 2017 11:13:14 -0700
Subject:
Re: [vpp-dev] Issue forwarding TCP packets

Hi Prabhjot,

In your setup, VPP just switches packets from one interface to the other, 
irrespective of their L4 checksum. That is, it does not read nor update them. 
Therefore, since the source Linux interface does not provide correct checksums, 
the destination rejects all packets.

As you’ve noted lower, although the packets are ultimately delivered to the 
app, even with the Linux bridge, the checksum is not correctly computed when 
offloading is on. I suspect the reason for this is an underlying optimization 
in the Linux kernel whereby local interfaces can exchange L4 packets without 
the use of checksums. In fact, the approach makes sense because packets never 
leave the kernel between the sndmsg() and recvmsg() calls so there’s no need to 
protect data integrity. However, this condition does not hold anymore when vpp 
does the switching, so the checksums must be computed before the packets leave 
the kernel (hence the need for checksum offload disabling).

Hope this helps,
Florin

> On Aug 28, 2017, at 11:46 PM, 
> prabh...@techtrueup.com wrote:
>
> Thanks Florin,
> it works with offload disabled, earlier when i tried changing offload 
> settings i missed doing it on one machine.
>
> However my question is still about why VPP is unable to handle it, is it a 
> missing configuration or missing functionality in VPP?
>
> i can ssh from one vm to another without turning off offload while using 
> linux bridge, and i don't see any change in the packet as well after being 
> forwarded. Though tcpdump complains about checksum, but everything works fine.
>
> packet ingressing into linux bridge from vm-1
> 06:41:58.626869 02:53:71:ef:2f:2a > 02:91:fb:46:9e:43, ethertype IPv4 
> (0x0800), length 74: (tos 0x0, ttl 64, id 6341, offset 0, flags [DF], proto 
> TCP (6), length 60)
> 1.1.1.3.51912 > 1.1.1.4.22: Flags [S], cksum 0x0437 (incorrect -> 0xdeae), 
> seq 2306393024, win 29200, options [mss 1460,sackOK,TS val 1825541 ecr 
> 0,nop,wscale 7], length 0
> 0x: 0291 fb46 9e43 0253 71ef 2f2a 0800 4500
> 0x0010: 003c 18c5 4000 4006 1def 0101 0103 0101
> 0x0020: 0104 cac8 0016 8978 c3c0   a002
> 0x0030: 7210 0437  0204 05b4 0402 080a 001b
> 0x0040: db05   0103 0307
>
> packet egressing from linux bridge to vm-2
> 06:41:58.627130 02:53:71:ef:2f:2a > 02:91:fb:46:9e:43, ethertype IPv4 
> (0x0800), length 74: (tos 0x0, ttl 64, id 6341, offset 0, flags [DF], proto 
> TCP (6), length 60)
> 1.1.1.3.51912 > 1.1.1.4.22: Flags [S], cksum 0x0437 (incorrect -> 0xdeae), 
> seq 2306393024, win 29200, options [mss 1460,sackOK,TS val 1825541 ecr 
> 0,nop,wscale 7], length 0
> 0x: 0291 fb46 9e43 0253 71ef 2f2a 0800 4500
> 0x0010: 003c 18c5 4000 4006 1def 0101 0103 0101
> 0x0020: 0104 cac8 0016 8978 c3c0   a002
> 0x0030: 7210 0437  0204 05b4 0402 080a 001b
> 0x0040: db05   0103 0307
>
>
> Regards,
> Prabhjot
>
> Quoting Florin Coras >:
>
>> Hi Prabhjot,
>>
>> From your

Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

2017-08-30 Thread Dave Wallace


I agree.

@Gabriel, please push a patch which adds bc to DEB_DEPENDS in vpp/Makefile.

Thanks,
-daw-

On 08/30/2017 09:14 AM, Burt Silverman wrote:
To me, it doesn't seem to be a crime to add bc to the dependencies. I 
guess another approach would be to remove the dot in 16dot04 and then 
just use shell arithmetic. The release numbers are always in the same 
format, 2 digits DOT 2 digits, so I would think that should work.


Burt

On Wed, Aug 30, 2017 at 3:47 AM, Gabriel Ganne > wrote:


Hi,


This probably is be because you don't have bc.

It is not in the dependency list. I'm so used to having it around
that I did not think to check.

Sorry.


If so, the best thing probably is to revert and not to increase
the dependency list just to silence a warning.


Regards,


--

Gabriel Ganne


*From:* Dave Wallace >
*Sent:* Wednesday, August 30, 2017 6:25:32 AM
*To:* Thomas F Herbert; Gabriel Ganne
*Cc:* vpp-dev
*Subject:* Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

Tom,

What OS are you running?

Thanks,
-daw-

On 08/29/2017 07:58 PM, Thomas F Herbert wrote:


This patch creates may have fixed one problem with Ubuntu 16.04
but created another:

$ make
/bin/sh: line 0: [: -eq: unary operator expected

...


-- 
*Thomas F Herbert*

NFV and Fast Data Planes
Office of Technology
*Red Hat*


___
vpp-dev mailing list
vpp-dev@lists.fd.io 
https://lists.fd.io/mailman/listinfo/vpp-dev





___
vpp-dev mailing list
vpp-dev@lists.fd.io 
https://lists.fd.io/mailman/listinfo/vpp-dev





___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Issue forwarding TCP packets

2017-08-30 Thread Prabhjot Singh Sethi

Hi Florin,

For now we can proceed with a requirement to disable offload on VMs.
but still keeping performance in view i think VPP should have option
similar to linux bridge to be able to do forwarding without correct
checksums, while offloading is enable the real checksum calculations
should be done only when the packet is egressing out of the server to
other compute nodes or gateway.

i think we can leave the topic open for now, and it can be considered
later for performance improvements for TCP

Regards,
Prabhjot

- Original Message -
From:
 "Florin Coras" 

To:
"Prabhjot Singh Sethi" .
Cc:

Sent:
Tue, 29 Aug 2017 11:13:14 -0700
Subject:
Re: [vpp-dev] Issue forwarding TCP packets

 Hi Prabhjot, 

 In your setup, VPP just switches packets from one interface to the
other, irrespective of their L4 checksum. That is, it does not read
nor update them. Therefore, since the source Linux interface does not
provide correct checksums, the destination rejects all packets. 

 As you’ve noted lower, although the packets are ultimately
delivered to the app, even with the Linux bridge, the checksum is not
correctly computed when offloading is on. I suspect the reason for
this is an underlying optimization in the Linux kernel whereby local
interfaces can exchange L4 packets without the use of checksums. In
fact, the approach makes sense because packets never leave the kernel
between the sndmsg() and recvmsg() calls so there’s no need to
protect data integrity. However, this condition does not hold anymore
when vpp does the switching, so the checksums must be computed before
the packets leave the kernel (hence the need for checksum offload
disabling). 

 Hope this helps, 
 Florin 

 > On Aug 28, 2017, at 11:46 PM, prabh...@techtrueup.com wrote:
 > 
 > Thanks Florin,
 > it works with offload disabled, earlier when i tried changing
offload settings i missed doing it on one machine.
 > 
 > However my question is still about why VPP is unable to handle it,
is it a missing configuration or missing functionality in VPP?
 > 
 > i can ssh from one vm to another without turning off offload while
using linux bridge, and i don't see any change in the packet as well
after being forwarded. Though tcpdump complains about checksum, but
everything works fine.
 > 
 > packet ingressing into linux bridge from vm-1
 > 06:41:58.626869 02:53:71:ef:2f:2a > 02:91:fb:46:9e:43, ethertype
IPv4 (0x0800), length 74: (tos 0x0, ttl 64, id 6341, offset 0, flags
[DF], proto TCP (6), length 60)
 > 1.1.1.3.51912 > 1.1.1.4.22: Flags [S], cksum 0x0437 (incorrect ->
0xdeae), seq 2306393024, win 29200, options [mss 1460,sackOK,TS val
1825541 ecr 0,nop,wscale 7], length 0
 > 0x: 0291 fb46 9e43 0253 71ef 2f2a 0800 4500
 > 0x0010: 003c 18c5 4000 4006 1def 0101 0103 0101
 > 0x0020: 0104 cac8 0016 8978 c3c0   a002
 > 0x0030: 7210 0437  0204 05b4 0402 080a 001b
 > 0x0040: db05   0103 0307
 > 
 > packet egressing from linux bridge to vm-2
 > 06:41:58.627130 02:53:71:ef:2f:2a > 02:91:fb:46:9e:43, ethertype
IPv4 (0x0800), length 74: (tos 0x0, ttl 64, id 6341, offset 0, flags
[DF], proto TCP (6), length 60)
 > 1.1.1.3.51912 > 1.1.1.4.22: Flags [S], cksum 0x0437 (incorrect ->
0xdeae), seq 2306393024, win 29200, options [mss 1460,sackOK,TS val
1825541 ecr 0,nop,wscale 7], length 0
 > 0x: 0291 fb46 9e43 0253 71ef 2f2a 0800 4500
 > 0x0010: 003c 18c5 4000 4006 1def 0101 0103 0101
 > 0x0020: 0104 cac8 0016 8978 c3c0   a002
 > 0x0030: 7210 0437  0204 05b4 0402 080a 001b
 > 0x0040: db05   0103 0307
 > 
 > 
 > Regards,
 > Prabhjot
 > 
 > Quoting Florin Coras :
 > 
 >> Hi Prabhjot,
 >> 
 >> From your description, I suspect it may be a linux tcp checksum
offload issue. Could you try disabling it for all interfaces with:
 >> 
 >> ethtool --offload  rx off tx off
 >> 
 >> Hope this helps,
 >> Florin
 >> 
 >>> On Aug 28, 2017, at 10:56 AM, Prabhjot Singh Sethi
 wrote:
 >>> 
 >>> We have been trying to use VPP as a L2/L3 forwarding data path
between VMs.
 >>> 
 >>> where we have connected tap interfaces from VM to VPP using
"create host-interface name "
 >>> after adding both the interfaces to same bridge-domain (bd_id =
2), we are able to ping from
 >>> one VM to another (both in same subnet). However when we try to
do ssh we observe that all the
 >>> tcp packets are transmitted/forwarded by VPP to the other end but
they are dropped as
 >>> "bad segments received".
 >>> 
 >>> Can anyone help with what could be wrong here, are we missing any
other required config?
 >>> 
 >>> Note:- same VMs works well when connected to linux bridge
 >>> 
 >>> root@vm-2:~# netstat -s | grep -A11 -i tcp:
 >>> Tcp:
 >>> 33 active connections openings
 >>> 4 passive connection openings
 >>> 0 failed connection attempts
 >>> 1 connection resets received
 >>> 1 connections established
 >>> 885 segments received
 >>> 650 segments

Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

2017-08-30 Thread Burt Silverman

To me, it doesn't seem to be a crime to add bc to the dependencies. I guess
another approach would be to remove the dot in 16dot04 and then just use
shell arithmetic. The release numbers are always in the same format, 2
digits DOT 2 digits, so I would think that should work.

Burt

On Wed, Aug 30, 2017 at 3:47 AM, Gabriel Ganne 
wrote:

> Hi,
>
>
> This probably is be because you don't have bc.
>
> It is not in the dependency list. I'm so used to having it around that I
> did not think to check.
>
> Sorry.
>
>
> If so, the best thing probably is to revert and not to increase the
> dependency list just to silence a warning.
>
>
> Regards,
>
>
> --
>
> Gabriel Ganne
> --
> *From:* Dave Wallace 
> *Sent:* Wednesday, August 30, 2017 6:25:32 AM
> *To:* Thomas F Herbert; Gabriel Ganne
> *Cc:* vpp-dev
> *Subject:* Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/
>
> Tom,
>
> What OS are you running?
>
> Thanks,
> -daw-
>
> On 08/29/2017 07:58 PM, Thomas F Herbert wrote:
>
> This patch creates may have fixed one problem with Ubuntu 16.04 but
> created another:
>
> $ make
> /bin/sh: line 0: [: -eq: unary operator expected
>
> ...
>
>
> --
> *Thomas F Herbert*
> NFV and Fast Data Planes
> Office of Technology
> *Red Hat*
>
>
> ___
> vpp-dev mailing 
> listvpp-...@lists.fd.iohttps://lists.fd.io/mailman/listinfo/vpp-dev 
> 
>
>
>
> ___
> vpp-dev mailing list
> vpp-dev@lists.fd.io
> https://lists.fd.io/mailman/listinfo/vpp-dev
>
___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Dynamically change number of cores used by VPP?

2017-08-30 Thread Sergio Gonzalez Monroy

AFAIK there is no support to adjust the number of cores used or the 
cores frequency in runtime.
Not sure if there is full support to dynamically switch between polling 
and interrupt mode, but I believe most of the infra is already there 
(read below thread).

https://lists.fd.io/pipermail/vpp-dev/2017-January/003868.html

Thanks,
Sergio

On 30/08/2017 12:59, Tobias Sundqvist wrote:
Hi I guess the silence tells that there are no way of dynamically 
scale the number of cores used by VPP when VPP already has started.
If anyone has any idea if it is dynamically possible to change the way 
the cores is used just send me an email. Perhaps it could be possible 
to assign a certain core to not poll the interface or something 
similar, then the core would still be in use of VPP but the CPU load 
would go down.


BR /Tobias

On 24 August 2017 at 10:36, Tobias Sundqvist 
> wrote:


Hi we are building an application that uses vpp and when I have
some questions that concerns multi core usage.

In the run that we are doing now we are using 4 cores and we use
some DPDK polling of an interface which makes the cpu load go 100%
on all cores. We are interested in energy consumption in the
application we are creating and would like to be able to scale up
and down the cpu usage during runtime.

Is it possible to dynamically change the number of cores used by
VPP or specify somehow which cores that are used for polling of an
certain interface during runtime?

During low and high traffic we would like to scale up and down the
amount of cores that is used.

If the polling is causing the cpu to go to 100% would it help
using Turboboost and Speedstep, would it actually lower the
frequency when the traffic is low?

BR /Tobias




___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev



___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Query for IPSec support on VPP

2017-08-30 Thread Mukesh Yadav (mukyadav)

HI Sergio,

I am using VPP version 17.10

vpp# show ver
vpp v17.10-rc0~103-g42e6b09 built by vagrant on localhost at Sun Aug 27 
22:06:20 PDT 2017

I am cloned vpp on my mac machine using below command
git clone https://gerrit.fd.io/r/vpp

Then I have ran vagrant up/vagrant ssh
Referred: 
https://wiki.fd.io/view/VPP/Pulling,_Building,_Running,_Hacking_and_Pushing_VPP_Code

On vagrant ssh, I have started vpp service.

Note: I have ran scenario mentioned on below links properly.
https://wiki.fd.io/view/VPP/IPSec_and_IKEv2

Only issue is coming when trying to use DPDK.

Once/twice I have ran below as well after vagrant ssh. But haven’t made any 
code changes in VPP Source.
cd /vpp
make wipe-release
make install-dep
make build-release (or make build)
make pkg-deb
sudo dpkg -i /vpp/build-root/*.deb
sudo service vpp start



Thanks
Mukesh

From: Sergio Gonzalez Monroy 
Date: Wednesday, 30 August 2017 at 4:18 PM
To: Mukesh Yadav , "vpp-dev@lists.fd.io" 

Subject: Re: [vpp-dev] Query for IPSec support on VPP

Which VPP are you using?

If you are using/installing VPP package (rpm/deb), then it is likely that you 
do not have DPDK SW crypto support.

As a summary, if you want DPDK SW crypto support:
- VPP package: 17.07.1 (AFAIK not released yet) or newer.
- From source:
- 17.04 or 17.07: need to build VPP with DPDK SW crypto support
 make build-release vpp_uses_dpdk_cryptodev_sw=yes
- 17.07.1 or newer:  DPDK SW crypto support built by default
 make build-release

Thanks,
Sergio

On 30/08/2017 10:57, Mukesh Yadav (mukyadav) wrote:
Thanks Sergio.

I tried below combinations of dpdk config.

dpdk {
socket-mem 1024,1024
num-mbufs 131072
dev :00:08.0
#enable-cryptodev
#vdev cryptodev_aesni_mb_pmd,socket_id=1
#vdev cryptodev_aesni_mb_pmd,socket_id=1
}

I get below error:
/usr/bin/vpp[3844]: dpdk_ipsec_process:239: not enough Cryptodevs, default to 
OpenSSL IPsec


dpdk {
socket-mem 1024,1024
num-mbufs 131072
dev :00:08.0
#enable-cryptodev
vdev cryptodev_aesni_mb_pmd,socket_id=1
vdev cryptodev_aesni_mb_pmd,socket_id=1
}

CRYPTODEV: [crypto_aesni_mb] cryptodev_aesni_mb_create() line 704: failed to 
create cryptodev vdev
CRYPTODEV: [crypto_aesni_mb] cryptodev_aesni_mb_create() line 746: driver 
cryptodev_aesni_mb_pmd: cryptodev_aesni_create failed
EAL: failed to initialize cryptodev_aesni_mb_pmd device

Seems still config is missing something.
Can you point me out any working wiki or document for making IPSec work along 
DPDK in VPP.

Thanks
Mukesh

From: Sergio Gonzalez Monroy 

Date: Wednesday, 30 August 2017 at 2:06 PM
To: Mukesh Yadav , 
"vpp-dev@lists.fd.io" 

Subject: Re: [vpp-dev] Query for IPSec support on VPP

Just for reference and clarity, there is no wiki entry for the DPDK crypto, 
only documentation page.
The wiki page is for general VPP/IPSec/IKEv2.

Thanks,
Sergio

On 30/08/2017 09:19, Sergio Gonzalez Monroy wrote:
On 29/08/2017 18:25, Mukesh Yadav (mukyadav) wrote:
Thanks Sergio for quick help. Pointers cleared few of basic doubts.

You are flushing out few left overs from previous releases, so thank you.




I will use below wiki to proceed.
https://docs.fd.io/vpp/17.10/dpdk_crypto_ipsec_doc.html
Since vpp version is 17.10


I added below in /vpp/src/vpp/conf/startup.conf

dpdk {
socket-mem 1024,1024
num-mbufs 131072
dev :00:08.0
enable-cryptodev
vdev cryptodev_aesni_mb_pmd,socket_id=1
vdev cryptodev_aesni_mb_pmd,socket_id=1
}





And did a restart of vpp service using “sudo service vpp restart”. It didn’t 
restarted service.
Till now I was using vanilla config which came with Source where dpdk config 
section was commented.


Next I tried below.
/usr/bin/vpp -c /etc/vpp/startup.conf



I can see logs on command terminal where above command is ran:
load_one_plugin:63: Loaded plugin: 
/usr/lib/vpp_api_test_plugins/ioam_pot_test_plugin.so
dpdk_config: unknown input `enable-cryptodev vdev cryptode...'

That is my bad. The option is not supported anymore, the doc is incorrect.






In wiki it is mentioned to have config like below:
dpdk {
2 socket-mem 1024,1024
3 num-mbufs 131072
4 dev :81:00.0
5 dev :81:00.1
6 enable-cryptodev
7 dev :85:01.0
8 dev :85:01.1
9 vdev cryptodev_aesni_mb_pmd,socket_id=1
   10 vdev cryptodev_aesni_mb_pmd,socket_id=1
   11 }



I see the wiki needs update too, at least for the latest VPP release.






It is mandatory to have 4 dev config, or It can work with single dev config as 
I mentioned in my sample config.

I reckon it should work with a single

Re: [vpp-dev] Dynamically change number of cores used by VPP?

2017-08-30 Thread Tobias Sundqvist

Hi I guess the silence tells that there are no way of dynamically scale the
number of cores used by VPP when VPP already has started.
If anyone has any idea if it is dynamically possible to change the way the
cores is used just send me an email. Perhaps it could be possible to assign
a certain core to not poll the interface or something similar, then the
core would still be in use of VPP but the CPU load would go down.

BR /Tobias

On 24 August 2017 at 10:36, Tobias Sundqvist 
wrote:

> Hi we are building an application that uses vpp and when I have some
> questions that concerns multi core usage.
>
> In the run that we are doing now we are using 4 cores and we use some DPDK
> polling of an interface which makes the cpu load go 100% on all cores. We
> are interested in energy consumption in the application we are creating and
> would like to be able to scale up and down the cpu usage during runtime.
>
> Is it possible to dynamically change the number of cores used by VPP or
> specify somehow which cores that are used for polling of an certain
> interface during runtime?
>
> During low and high traffic we would like to scale up and down the amount
> of cores that is used.
>
> If the polling is causing the cpu to go to 100% would it help using
> Turboboost and Speedstep, would it actually lower the frequency when the
> traffic is low?
>
> BR /Tobias
>
___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Query for IPSec support on VPP

2017-08-30 Thread Sergio Gonzalez Monroy


Which VPP are you using?

If you are using/installing VPP package (rpm/deb), then it is likely 
that you do not have DPDK SW crypto support.


As a summary, if you want DPDK SW crypto support:
- VPP package: 17.07.1 (AFAIK not released yet) or newer.
- From source:
- 17.04 or 17.07: need to build VPP with DPDK SW crypto support
 make build-release vpp_uses_dpdk_cryptodev_sw=yes
- 17.07.1 or newer:  DPDK SW crypto support built by default
 make build-release

Thanks,
Sergio

On 30/08/2017 10:57, Mukesh Yadav (mukyadav) wrote:


Thanks Sergio.

I tried below combinations of dpdk config.

dpdk {

  socket-mem 1024,1024

  num-mbufs 131072

  dev :00:08.0

#   enable-cryptodev

#   vdev cryptodev_aesni_mb_pmd,socket_id=1

#   vdev cryptodev_aesni_mb_pmd,socket_id=1

}

I get below error:

/usr/bin/vpp[3844]: dpdk_ipsec_process:239: not enough Cryptodevs, 
default to OpenSSL IPsec


dpdk {

  socket-mem 1024,1024

  num-mbufs 131072

  dev :00:08.0

#   enable-cryptodev

  vdev cryptodev_aesni_mb_pmd,socket_id=1

  vdev cryptodev_aesni_mb_pmd,socket_id=1

}

CRYPTODEV: [crypto_aesni_mb] cryptodev_aesni_mb_create() line 704: 
failed to create cryptodev vdev


CRYPTODEV: [crypto_aesni_mb] cryptodev_aesni_mb_create() line 746: 
driver cryptodev_aesni_mb_pmd: cryptodev_aesni_create failed


EAL: failed to initialize cryptodev_aesni_mb_pmd device

Seems still config is missing something.

Can you point me out any working wiki or document for making IPSec 
work along DPDK in VPP.


Thanks

Mukesh

*From: *Sergio Gonzalez Monroy 
*Date: *Wednesday, 30 August 2017 at 2:06 PM
*To: *Mukesh Yadav , "vpp-dev@lists.fd.io" 


*Subject: *Re: [vpp-dev] Query for IPSec support on VPP

Just for reference and clarity, there is no wiki entry for the DPDK 
crypto, only documentation page.

The wiki page is for general VPP/IPSec/IKEv2.

Thanks,
Sergio

On 30/08/2017 09:19, Sergio Gonzalez Monroy wrote:

On 29/08/2017 18:25, Mukesh Yadav (mukyadav) wrote:

Thanks Sergio for quick help. Pointers cleared few of basic
doubts.


You are flushing out few left overs from previous releases, so
thank you.


I will use below wiki to proceed.

https://docs.fd.io/vpp/17.10/dpdk_crypto_ipsec_doc.html

Since vpp version is 17.10

I added below in /vpp/src/vpp/conf/startup.conf

dpdk {

  socket-mem 1024,1024

  num-mbufs 131072

  dev :00:08.0

  enable-cryptodev

  vdev cryptodev_aesni_mb_pmd,socket_id=1

  vdev cryptodev_aesni_mb_pmd,socket_id=1

}

And did a restart of vpp service using “sudo service vpp
restart”. It didn’t restarted service.

Till now I was using vanilla config which came with Source
where dpdk config section was commented.

Next I tried below.

/usr/bin/vpp -c /etc/vpp/startup.conf

I can see logs on command terminal where above command is ran:

load_one_plugin:63: Loaded plugin:
/usr/lib/vpp_api_test_plugins/ioam_pot_test_plugin.so

dpdk_config: unknown input `enable-cryptodev vdev cryptode...'


That is my bad. The option is not supported anymore, the doc is
incorrect.


In wiki it is mentioned to have config like below:

dpdk {

  2  socket-mem 1024,1024

  3  num-mbufs 131072

  4  dev :81:00.0

  5  dev :81:00.1

  6  enable-cryptodev

  7  dev :85:01.0

  8  dev :85:01.1

  9  vdev cryptodev_aesni_mb_pmd,socket_id=1

10  vdev cryptodev_aesni_mb_pmd,socket_id=1

11 }


I see the wiki needs update too, at least for the latest VPP release.


It is mandatory to have 4 dev config, or It can work with
single dev config as I mentioned in my sample config.


I reckon it should work with a single port, should entirely depend
on your config (fib, ipsec-tunnels, etc).
The example in the wiki was just trying to cover the different
options with a single configuration.

Thanks,
Sergio


Thanks

Mukesh




___

vpp-dev mailing list

vpp-dev@lists.fd.io 

https://lists.fd.io/mailman/listinfo/vpp-dev



___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] About the order of VLIB_INIT_FUNCTION called between different plugins

2017-08-30 Thread Dave Barach (dbarach)

Explicit dependencies between plugins is probably not a good idea. There is 
little to guarantee that both A and B will be loaded.

Please describe the use-case in more detail.

Thanks… Dave

From: vpp-dev-boun...@lists.fd.io [mailto:vpp-dev-boun...@lists.fd.io] On 
Behalf Of wang.hu...@zte.com.cn
Sent: Wednesday, August 30, 2017 4:01 AM
To: vpp-dev@lists.fd.io
Cc: zhao.qingl...@zte.com.cn; wu.bi...@zte.com.cn; gu.ji...@zte.com.cn; 
dong.ju...@zte.com.cn
Subject: [vpp-dev] About the order of  VLIB_INIT_FUNCTION called between 
different plugins


Hi all:

How to control the order of  VLIB_INIT_FUNCTION (user xxx_init function) called 
between Different plugins?

It depends on plugin name?or the sequence of loading plugin ?

 or is there any other way to adjust the order?



Thanks~







王辉 wanghui



IT开发工程师 IT Development Engineer
虚拟化南京四部/无线研究院/无线产品经营部 NIV Nanjing Dept. IV/Wireless Product R＆D 
Institute/Wireless Product Operation Division


___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] Query for IPSec support on VPP

2017-08-30 Thread Mukesh Yadav (mukyadav)

Thanks Sergio.

I tried below combinations of dpdk config.

dpdk {
socket-mem 1024,1024
num-mbufs 131072
dev :00:08.0
#enable-cryptodev
#vdev cryptodev_aesni_mb_pmd,socket_id=1
#vdev cryptodev_aesni_mb_pmd,socket_id=1
}

I get below error:
/usr/bin/vpp[3844]: dpdk_ipsec_process:239: not enough Cryptodevs, default to 
OpenSSL IPsec


dpdk {
socket-mem 1024,1024
num-mbufs 131072
dev :00:08.0
#enable-cryptodev
vdev cryptodev_aesni_mb_pmd,socket_id=1
vdev cryptodev_aesni_mb_pmd,socket_id=1
}

CRYPTODEV: [crypto_aesni_mb] cryptodev_aesni_mb_create() line 704: failed to 
create cryptodev vdev
CRYPTODEV: [crypto_aesni_mb] cryptodev_aesni_mb_create() line 746: driver 
cryptodev_aesni_mb_pmd: cryptodev_aesni_create failed
EAL: failed to initialize cryptodev_aesni_mb_pmd device

Seems still config is missing something.
Can you point me out any working wiki or document for making IPSec work along 
DPDK in VPP.

Thanks
Mukesh

From: Sergio Gonzalez Monroy 
Date: Wednesday, 30 August 2017 at 2:06 PM
To: Mukesh Yadav , "vpp-dev@lists.fd.io" 

Subject: Re: [vpp-dev] Query for IPSec support on VPP

Just for reference and clarity, there is no wiki entry for the DPDK crypto, 
only documentation page.
The wiki page is for general VPP/IPSec/IKEv2.

Thanks,
Sergio

On 30/08/2017 09:19, Sergio Gonzalez Monroy wrote:
On 29/08/2017 18:25, Mukesh Yadav (mukyadav) wrote:
Thanks Sergio for quick help. Pointers cleared few of basic doubts.

You are flushing out few left overs from previous releases, so thank you.



I will use below wiki to proceed.
https://docs.fd.io/vpp/17.10/dpdk_crypto_ipsec_doc.html
Since vpp version is 17.10


I added below in /vpp/src/vpp/conf/startup.conf

dpdk {
socket-mem 1024,1024
num-mbufs 131072
dev :00:08.0
enable-cryptodev
vdev cryptodev_aesni_mb_pmd,socket_id=1
vdev cryptodev_aesni_mb_pmd,socket_id=1
}





And did a restart of vpp service using “sudo service vpp restart”. It didn’t 
restarted service.
Till now I was using vanilla config which came with Source where dpdk config 
section was commented.


Next I tried below.
/usr/bin/vpp -c /etc/vpp/startup.conf



I can see logs on command terminal where above command is ran:
load_one_plugin:63: Loaded plugin: 
/usr/lib/vpp_api_test_plugins/ioam_pot_test_plugin.so
dpdk_config: unknown input `enable-cryptodev vdev cryptode...'

That is my bad. The option is not supported anymore, the doc is incorrect.





In wiki it is mentioned to have config like below:
dpdk {
2 socket-mem 1024,1024
3 num-mbufs 131072
4 dev :81:00.0
5 dev :81:00.1
6 enable-cryptodev
7 dev :85:01.0
8 dev :85:01.1
9 vdev cryptodev_aesni_mb_pmd,socket_id=1
   10 vdev cryptodev_aesni_mb_pmd,socket_id=1
   11 }



I see the wiki needs update too, at least for the latest VPP release.





It is mandatory to have 4 dev config, or It can work with single dev config as 
I mentioned in my sample config.

I reckon it should work with a single port, should entirely depend on your 
config (fib, ipsec-tunnels, etc).
The example in the wiki was just trying to cover the different options with a 
single configuration.

Thanks,
Sergio







Thanks

Mukesh








___

vpp-dev mailing list

vpp-dev@lists.fd.io

https://lists.fd.io/mailman/listinfo/vpp-dev


___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] multiple ips for an interface

2017-08-30 Thread Ole Troan

Pragash,

> I have a quick question, i am able to assign multiple ips for an interface in 
> vpp. Is this correct behavior.

Yes. Shouldn't it be?
Note that VPP itself doesn't implement source address selection so the local 
application would have to deal with that.

Best regards,
Ole


> 
> pragash@revvit:~/VPP$ sudo vppctl sh interfaces address
> FortyGigabitEthernet2/0/0 (up):
>   10.1.1.1/24
>   10.2.1.1/24
>   2001::1/48
>   fe80::3efd:feff:fea6:8ef0/128
>   2001::2/48
>   2003::2/48
>   2004::2/48
> FortyGigabitEthernet4/0/0 (up):
>   2002::1/48
>   fe80::3efd:feff:fea6:8e78/128
>   9001::2/48
> local0 (dn):
> 
> 
> 
> Thanks,
> 
> Pragash Vijayaragavan
> Grad Student at Rochester Institute of Technology
> email : pxv3...@rit.edu
> ph : 585 764 4662
> 
> ___
> vpp-dev mailing list
> vpp-dev@lists.fd.io
> https://lists.fd.io/mailman/listinfo/vpp-dev



signature.asc
Description: Message signed with OpenPGP
___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

[vpp-dev] About the order of VLIB_INIT_FUNCTION called between different plugins

2017-08-30 Thread wang.hui56

Hi all:

How to control the order of  VLIB_INIT_FUNCTION (user xxx_init function) called 
between Different plugins?

It depends on plugin name?or the sequence of loading plugin ?

 or is there any other way to adjust the order?




Thanks~













王辉 wanghui






IT开发工程师 IT Development
Engineer
虚拟化南京四部/无线研究院/无线产品经营部 NIV Nanjing Dept. IV/Wireless Product R＆D 
Institute/Wireless Product Operation Division___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

2017-08-30 Thread Gabriel Ganne

Hi,


This probably is be because you don't have bc.

It is not in the dependency list. I'm so used to having it around that I did 
not think to check.

Sorry.


If so, the best thing probably is to revert and not to increase the dependency 
list just to silence a warning.


Regards,


--

Gabriel Ganne


From: Dave Wallace 
Sent: Wednesday, August 30, 2017 6:25:32 AM
To: Thomas F Herbert; Gabriel Ganne
Cc: vpp-dev
Subject: Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

Tom,

What OS are you running?

Thanks,
-daw-

On 08/29/2017 07:58 PM, Thomas F Herbert wrote:

This patch creates may have fixed one problem with Ubuntu 16.04 but created 
another:

$ make
/bin/sh: line 0: [: -eq: unary operator expected

...


--
Thomas F Herbert
NFV and Fast Data Planes
Office of Technology
Red Hat



___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

___
vpp-dev mailing list
vpp-dev@lists.fd.io
https://lists.fd.io/mailman/listinfo/vpp-dev

Re: [vpp-dev] multiple ips for an interface

[vpp-dev] ip6 routing issue

[vpp-dev] VPP Communications Library demo instructions

Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

Re: [vpp-dev] Query for IPSec support on VPP

Re: [vpp-dev] Dynamically change number of cores used by VPP?

Re: [vpp-dev] Query for IPSec support on VPP

Re: [vpp-dev] About the order of VLIB_INIT_FUNCTION called between different plugins

Re: [vpp-dev] Issue forwarding TCP packets

Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

Re: [vpp-dev] Issue forwarding TCP packets

Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

Re: [vpp-dev] Dynamically change number of cores used by VPP?

Re: [vpp-dev] Query for IPSec support on VPP

Re: [vpp-dev] Dynamically change number of cores used by VPP?

Re: [vpp-dev] Query for IPSec support on VPP

Re: [vpp-dev] About the order of VLIB_INIT_FUNCTION called between different plugins

Re: [vpp-dev] Query for IPSec support on VPP

Re: [vpp-dev] multiple ips for an interface

[vpp-dev] About the order of VLIB_INIT_FUNCTION called between different plugins

Re: [vpp-dev] https://gerrit.fd.io/r/#/c/8236/

23 matches

Site Navigation

Mail list logo

Footer information