[vpp-dev] vppctl issue

[Kevin Yan]

Hello,
 Does anyone met below errors when using vppctl and what is the possible 
reason?

clib_socket_init: connect (fd 3, '/run/vpp/cli.sock'): Connection refused

  BTW,  I changed a little vpp code but I suppose it doesn't not related to 
this issue and VPP itself is started without any problem

My startup configuration:

cpu {
main-core 0
corelist-workers 1-3
}

unix {
nodaemon
log /tmp/vpp.log
full-coredump
cli-listen /run/vpp/cli.sock
}

dpdk {
socket-mem 2048
uio-driver uio_pci_generic
dev :00:06.0 {
num-rx-queues 2
num-tx-queues 2
}
dev :00:09.0
vdev cryptodev_aesni_mb_pmd,socket_id=0
vdev cryptodev_aesni_mb_pmd,socket_id=0
}

api-trace {
on
}

tap-inject {
enable
}


BRs,
Kevin



This e-mail message may contain confidential or proprietary information of 
Mavenir Systems, Inc. or its affiliates and is intended solely for the use of 
the intended recipient(s). If you are not the intended recipient of this 
message, you are hereby notified that any review, use or distribution of this 
information is absolutely prohibited and we request that you delete all copies 
in your control and contact us by e-mailing to secur...@mavenir.com. This 
message contains the views of its author and may not necessarily reflect the 
views of Mavenir Systems, Inc. or its affiliates, who employ systems to monitor 
email messages, but make no representation that such messages are authorized, 
secure, uncompromised, or free from computer viruses, malware, or other 
defects. Thank You
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11134): https://lists.fd.io/g/vpp-dev/message/11134
Mute This Topic: https://lists.fd.io/mt/28021014/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] vppctl issue

[alp . arslan]

Check if VPP is running properly, "service vpp status" if its installed
using the binaries. It normally occurs when the VPP service is not running.


 

From: vpp-dev@lists.fd.io [mailto:vpp-dev@lists.fd.io] On Behalf Of Kevin
Yan
Sent: Wednesday, November 7, 2018 2:10 PM
To: vpp-dev@lists.fd.io
Subject: [vpp-dev] vppctl issue

 

Hello, 

 Does anyone met below errors when using vppctl and what is the possible
reason?

 

clib_socket_init: connect (fd 3, '/run/vpp/cli.sock'): Connection refused

 

  BTW,  I changed a little vpp code but I suppose it doesn't not related to
this issue and VPP itself is started without any problem

 

My startup configuration:

 

cpu {

main-core 0

corelist-workers 1-3

}

 

unix {

nodaemon

log /tmp/vpp.log

full-coredump

cli-listen /run/vpp/cli.sock

}

 

dpdk {

socket-mem 2048

uio-driver uio_pci_generic

dev :00:06.0 {

num-rx-queues 2

num-tx-queues 2

}

dev :00:09.0

vdev cryptodev_aesni_mb_pmd,socket_id=0

vdev cryptodev_aesni_mb_pmd,socket_id=0

}

 

api-trace {

on

}

 

tap-inject {

enable

}

 

 

BRs,

Kevin

 

 

  _  

This e-mail message may contain confidential or proprietary information of
Mavenir Systems, Inc. or its affiliates and is intended solely for the use
of the intended recipient(s). If you are not the intended recipient of this
message, you are hereby notified that any review, use or distribution of
this information is absolutely prohibited and we request that you delete all
copies in your control and contact us by e-mailing to secur...@mavenir.com
 . This message contains the views of its
author and may not necessarily reflect the views of Mavenir Systems, Inc. or
its affiliates, who employ systems to monitor email messages, but make no
representation that such messages are authorized, secure, uncompromised, or
free from computer viruses, malware, or other defects. Thank You

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11135): https://lists.fd.io/g/vpp-dev/message/11135
Mute This Topic: https://lists.fd.io/mt/28021014/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] inet raw socket

[saint_sun 孙 via Lists . Fd . Io]

I am confused, your suggestion is that I can use 
The API interface?




可用于iOS的myMail发送


2018年11月7日 星期三 +0800 15:49 发件人 Ole Troan  :
>> <81.gif>thanks very much! And I  have another question: does VPP support 
>> INET RAW socket, I find the UDP and TCP socket except the raw socket.
>
>No, but you do get full access to the frame in the buffer. Which is generally 
>a much more flexible interface than INET RAW.
>
>Cheers,
>Ole-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11136): https://lists.fd.io/g/vpp-dev/message/11136
Mute This Topic: https://lists.fd.io/mt/27857220/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] inet raw socket

[Ole Troan]

> I am confused, your suggestion is that I can use 
> The API interface?

No. What I said was that you were writing a graph node, you have direct access 
to the frames.

Instead of playing hide and seak. Could you tell us what you are trying to do?
What problem are you trying to solve?

Ole


> 
> 
> 
> 
> 可用于iOS的myMail发送
> 
> 
> 2018年11月7日 星期三 +0800 15:49 发件人 Ole Troan :
> > <81.gif>thanks very much! And I have another question: does VPP support 
> > INET RAW socket, I find the UDP and TCP socket except the raw socket.
> 
> No, but you do get full access to the frame in the buffer. Which is generally 
> a much more flexible interface than INET RAW.
> 
> Cheers,
> Ole

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11137): https://lists.fd.io/g/vpp-dev/message/11137
Mute This Topic: https://lists.fd.io/mt/27857220/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] vppctl issue

[Dave Barach via Lists.Fd.Io]

+1.

If the vpp service is running - and you're using the distributed version of 
/etc/vpp/startup.conf - you'll see this stanza:

api-segment {
  gid vpp
}

Make sure that the user running vppctl is either in the vpp group, or is root.

HTH... Dave

From: vpp-dev@lists.fd.io  On Behalf Of 
alp.ars...@xflowresearch.com
Sent: Wednesday, November 7, 2018 4:48 AM
To: 'Kevin Yan' ; vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] vppctl issue

Check if VPP is running properly, "service vpp status" if its installed using 
the binaries. It normally occurs when the VPP service is not running.

From: vpp-dev@lists.fd.io 
[mailto:vpp-dev@lists.fd.io] On Behalf Of Kevin Yan
Sent: Wednesday, November 7, 2018 2:10 PM
To: vpp-dev@lists.fd.io
Subject: [vpp-dev] vppctl issue

Hello,
 Does anyone met below errors when using vppctl and what is the possible 
reason?

clib_socket_init: connect (fd 3, '/run/vpp/cli.sock'): Connection refused

  BTW,  I changed a little vpp code but I suppose it doesn't not related to 
this issue and VPP itself is started without any problem

My startup configuration:

cpu {
main-core 0
corelist-workers 1-3
}

unix {
nodaemon
log /tmp/vpp.log
full-coredump
cli-listen /run/vpp/cli.sock
}

dpdk {
socket-mem 2048
uio-driver uio_pci_generic
dev :00:06.0 {
num-rx-queues 2
num-tx-queues 2
}
dev :00:09.0
vdev cryptodev_aesni_mb_pmd,socket_id=0
vdev cryptodev_aesni_mb_pmd,socket_id=0
}

api-trace {
on
}

tap-inject {
enable
}


BRs,
Kevin



This e-mail message may contain confidential or proprietary information of 
Mavenir Systems, Inc. or its affiliates and is intended solely for the use of 
the intended recipient(s). If you are not the intended recipient of this 
message, you are hereby notified that any review, use or distribution of this 
information is absolutely prohibited and we request that you delete all copies 
in your control and contact us by e-mailing to 
secur...@mavenir.com. This message contains the 
views of its author and may not necessarily reflect the views of Mavenir 
Systems, Inc. or its affiliates, who employ systems to monitor email messages, 
but make no representation that such messages are authorized, secure, 
uncompromised, or free from computer viruses, malware, or other defects. Thank 
You
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11138): https://lists.fd.io/g/vpp-dev/message/11138
Mute This Topic: https://lists.fd.io/mt/28021014/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] Can program have more than 1 VAPI connection?

[wangchuan...@163.com]

Hi all,
Can i create more than 1 connection using vapi_connect in my program ?
ctx was defined at .so (more than 1) 



wangchuan...@163.com
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11139): https://lists.fd.io/g/vpp-dev/message/11139
Mute This Topic: https://lists.fd.io/mt/2803/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] Can program have more than 1 VAPI connection?

[Klement Sekera via Lists.Fd.Io]

Hi,

sadly, no, because the underlying low-level library uses global variables.

Regards,
Klement

Quoting wangchuan...@163.com (2018-11-07 13:52:39)
>Hi all,
>    Can i create more than 1 connection using vapi_connect in my program ?
>ctx was defined at .so (more than 1) 
> 
>--
> 
>wangchuan...@163.com
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11140): https://lists.fd.io/g/vpp-dev/message/11140
Mute This Topic: https://lists.fd.io/mt/2803/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] inet raw socket

[saint_sun 孙 via Lists . Fd . Io]

Some protocols or functions need inet raw socket to transmit and receive 
packets such as ping/vrrp/pim, and I want to debug these functions.



可用于iOS的myMail发送


2018年11月7日 星期三 +0800 19:40 发件人 Ole Troan  :
>> I am confused, your suggestion is that I can use 
>> The API interface?
>
>No. What I said was that you were writing a graph node, you have direct access 
>to the frames.
>
>Instead of playing hide and seak. Could you tell us what you are trying to do?
>What problem are you trying to solve?
>
>Ole
>
>
>> 
>> 
>> 
>> 
>> 可用于iOS的myMail发送
>> 
>> 
>> 2018年11月7日 星期三 +0800 15:49 发件人 Ole Troan < otr...@employees.org >:
>> > <81.gif>thanks very much! And I have another question: does VPP support 
>> > INET RAW socket, I find the UDP and TCP socket except the raw socket.
>> 
>> No, but you do get full access to the frame in the buffer. Which is 
>> generally a much more flexible interface than INET RAW.
>> 
>> Cheers,
>> Ole
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11141): https://lists.fd.io/g/vpp-dev/message/11141
Mute This Topic: https://lists.fd.io/mt/27857220/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] VPP unit test failures (l2fib, ip4 reassemble) on environments other than x86 Ubuntu 1804

[Juraj Linkeš]

Hi vpp-devs,

There are some failures in unit tests that have cropped up in master on ARM:

*https://jira.fd.io/browse/VPP-1475

*https://jira.fd.io/browse/VPP-1476

I tried Ubuntu1604 and 1804 on ARM and the issues are reproducible on both. 
I've noticed that we switched testing from ubuntu1604 to 1804 on x86 in CI, so 
I've also tried some other OS's on x86 and sure enough, I saw both issues on 
ubuntu1604 and Debian9.

Could someone look into these?

Thanks,
Juraj
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11142): https://lists.fd.io/g/vpp-dev/message/11142
Mute This Topic: https://lists.fd.io/mt/28023151/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] Faulty java version check for JVPP GEN

[Juraj Linkeš]

There's actually a JIRA ticket opened for this, so it seems I'm not the only 
one who's looking for a solution: https://jira.fd.io/browse/VPP-1477

Juraj

From: Juraj Linkeš [mailto:juraj.lin...@pantheon.tech]
Sent: Monday, November 5, 2018 5:36 PM
To: vpp-dev 
Cc: sirshak@arm.com
Subject: [vpp-dev] Faulty java version check for JVPP GEN

Hi vpp-devs,

I'm trying to get VPP build (actually make verify for ci purposes) to run on 
Ubuntu1804 and I've hit an issue with java:
checking /usr/lib/jvm/java-11-openjdk-arm64 for Java 8 compiler... found 
version 11
...
/vpp/build-data/../extras/japi/java/jvpp-core/io/fd/vpp/jvpp/core/examples/L2AclExample.java:37:
 error: package javax.xml.bind does not exist
import javax.xml.bind.DatatypeConverter;
 ^
/vpp/build-data/../extras/japi/java/jvpp-core/io/fd/vpp/jvpp/core/examples/L2AclExample.java:123:
 error: cannot find symbol
System.out.println("Mask hex: " + 
DatatypeConverter.printHexBinary(reply.mask));
  ^
  symbol:   variable DatatypeConverter
  location: class L2AclExample

The package in question was deprecated in Java 9 and removed in Java 11, so 
that's the reason for the failure.

Java 8 present on the system and Java 11 gets installed as part of dependency 
installation of make verify (default-jdk-headless), so a workaround along the 
lines of removing Java 11 or reconfiguring Java is not really on the table.

If we want to install default-jdk-headless as part of make install-dep, we 
should be prepared for Java 11 being installed. What's the right way to fix 
this? Use a package other than javax.xml.bind or modify how the build searches 
for Java compilers?

Thanks,
Juraj
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11143): https://lists.fd.io/g/vpp-dev/message/11143
Mute This Topic: https://lists.fd.io/mt/27860462/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [**EXTERNAL**] [vpp-dev] TAP v1/v2 API questions

[Coulson, Ken]

Our normal pattern for forwarding plane entities is create/modify/delete, the 
TAP v1 API fit nicely with that.
The desire is to support changing the interface after it's created without 
deleting and re-creating to avoid as much potential traffic disruption as 
possible.
An example is MTU (which neither v1 or v2 support).  The desired maximum MTU 
may not be known at creation time or may change depending on routing so a 
modify capability seems appropriate.
So be specific on your questions, want to control MTU and admin/link 
enable/disable, would like to know link status without polling.

DPDK TAP PMD (and vHost PMD) provide the same APIs as physical devices using 
the rte_eth_dev_*() functions allowing the next layer of code to very similar 
for physical and virtual devices.
DPDK TAP PMD supports MTU, jumbo frames, extended statistics, link status 
change notifications, plus likely other features.  These seem to me to be an 
advantage and not a drawback.
I'm curious about the drawbacks mentioned and advantages of the VPP 
implementation, could you provide a more detail?  More information will 
certainly help!

Ken Coulson


From: Damjan Marion 
Date: Tuesday, November 6, 2018 at 9:36 AM
To: "Coulson, Ken" 
Cc: "vpp-dev@lists.fd.io" 
Subject: Re: [**EXTERNAL**] [vpp-dev] TAP v1/v2 API questions




On 6 Nov 2018, at 18:19, Coulson, Ken 
mailto:kcoul...@ciena.com>> wrote:

I didn’t hear anything back on questions below so re-posting again for another 
attempt at answers.

Ken Coulson



From: mailto:vpp-dev@lists.fd.io>> on behalf of "Coulson, 
Ken" mailto:kcoul...@ciena.com>>
Date: Wednesday, October 31, 2018 at 11:10 AM
To: "vpp-dev@lists.fd.io" 
mailto:vpp-dev@lists.fd.io>>
Subject: [**EXTERNAL**] [vpp-dev] TAP v1/v2 API questions

Per emails on this list tap v1 is being deprecated.
tap v1 create/modify/delete and tap v2 is different in that it only has 
create/delete.

Some questions:
Is the intent to create a tap v2 modify API and/or to add capabilities to set 
interface API?

What exactly would you like to modify? What capabilities?

tap APIs presently have no MTU setting.  Is there any plan to add that 
capability?

Isn't there generic API for that?


Is there any plan to use DPDK tap PMD?

I don't see any benefits and I see several drawbacks of doing that.
Why do you ask?


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11144): https://lists.fd.io/g/vpp-dev/message/11144
Mute This Topic: https://lists.fd.io/mt/27873113/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] #vpp sctp echo test crash

[Yao, Chengqiang]

Hi Macro,

I use following CLI command to setup SCTP server, but VPP crashed. While this 
command works for TCP/UDP.

Can you tell me  the correct cli commands to run echo test for SCTP?


vpp# set int ip address FortyGigabitEthernet18/0/1 192.168.60.73/24
vpp# set int state FortyGigabitEthernet18/0/1 up
vpp# set ip arp FortyGigabitEthernet18/0/1 192.168.60.74 3c:fd:fe:c2:12:68
vpp# test echo server fifo-size 4096 uri sctp://192.168.60.73/1234 no-echo


After debugging, I found it’s caused by invalid sw_if_index in 
sctp_connection_bind function of sctp.c.



Best Regards,
Chengqiang Yao



From: vpp-dev@lists.fd.io [mailto:vpp-dev@lists.fd.io] On Behalf Of Yao, 
Chengqiang
Sent: Thursday, October 25, 2018 9:15 PM
To: Marco Varlese ; vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] #vpp

Hi Macro,

Thanks for your information, I will have a try.


Best Regards,
Chengqiang Yao



From: Marco Varlese [mailto:mvarl...@suse.de]
Sent: Thursday, October 25, 2018 9:09 PM
To: Yao, Chengqiang 
mailto:chengqiang@intel.com>>; 
vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] #vpp

Hi,

On Thu, 2018-10-25 at 04:59 +, Yao, Chengqiang wrote:
Hi Macro,

Is there any sample code to show how to use SCTP (such as association, packet 
transmission/reception, etc.)? And is there any performance report for SCTP?
You can have a look at the test for SCTP:
1) test/test_sctp.py
2) src/vnet/session-apps/echo_client.c & echo_server.c

The echo_client and echo_server are used by TCP / UDP and SCTP to exercise the 
networking stack implemented.

With regards to the state-machine (I assumed that's what you're asking by 
"association", "packet transmssion/reception", etc.), the whole of it is gone 
through when running the tests.
If you are interested in seeing/learning how that's implemented I am afraid you 
need to look into the code tho.

With regards to performance: no, there're no reports. However, something can be 
retrieved by running the test_sctp.py since at the end it prints out how much 
data has been transferred and how long that took.



Best Regards,
Chengqiang Yao
Cheers,
Marco





From: vpp-dev@lists.fd.io 
[mailto:vpp-dev@lists.fd.io] On Behalf Of Marco Varlese
Sent: Wednesday, October 10, 2018 5:47 PM
To: srivastava.rac...@gmail.com; 
vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] #vpp

Hi Srivastava,

On Wed, 2018-10-10 at 01:23 -0700, 
srivastava.rac...@gmail.com wrote:
Hi

Does the VPP support SCTP. If yes what is the configuration for SCTP. Does it 
support the same features as the linux kernel SCTP ?
VPP has an implementation for SCTP. You can look into /src/vnet/sctp folder to 
see what we currently support in terms of RFC.
Any feedback (and contribution) is very much welcome!


Thanks
Rachit Srivastava
Thanks,
Marco


-=-=-=-=-=-=-=-=-=-=-=-

Links: You receive all messages sent to this group.



View/Reply Online (#10787): https://lists.fd.io/g/vpp-dev/message/10787

Mute This Topic: https://lists.fd.io/mt/27155313/675056

Mute #vpp: https://lists.fd.io/mk?hashtag=vpp&subid=1480704

Group Owner: vpp-dev+ow...@lists.fd.io

Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  
[mvarl...@suse.de]

-=-=-=-=-=-=-=-=-=-=-=-


-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11145): https://lists.fd.io/g/vpp-dev/message/11145
Mute This Topic: https://lists.fd.io/mt/28023358/21656
Mute #vpp: https://lists.fd.io/mk?hashtag=vpp&subid=1480452
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [**EXTERNAL**] [vpp-dev] TAP v1/v2 API questions

[Damjan Marion via Lists.Fd.Io]

-- 
Damjan

> On 7 Nov 2018, at 16:01, Coulson, Ken  wrote:
> 
> Our normal pattern for forwarding plane entities is create/modify/delete, the 
> TAP v1 API fit nicely with that.
> The desire is to support changing the interface after it's created without 
> deleting and re-creating to avoid as much potential traffic disruption as 
> possible.
> An example is MTU (which neither v1 or v2 support).  The desired maximum MTU 
> may not be known at creation time or may change depending on routing so a 
> modify capability seems appropriate.

VPP have universal API (hw_interface_set_mtu ) for setting MTU on all 
interfaces.

> So be specific on your questions, want to control MTU and admin/link 
> enable/disable, would like to know link status without polling.

VPP have universal API (sw_interface_set_flags ) for setting link and admin 
states.
You can subscribe to interface events with want_interface_events API to be 
notified on change.

>  
> DPDK TAP PMD (and vHost PMD) provide the same APIs as physical devices using 
> the rte_eth_dev_*() functions allowing the next layer of code to very similar 
> for physical and virtual devices.

In some cases we run vPP without DPDK, DPDK is just one device driver option 
for VPP.

> DPDK TAP PMD supports MTU,

see above

> jumbo frames

we support jumbo frames with tapv2.

> , extended statistics,

what exact extended statistic you are looking for?

> link status change notifications

see above

> plus likely other features.

like?

>   These seem to me to be an advantage and not a drawback. 
> I'm curious about the drawbacks mentioned and advantages of the VPP 
> implementation, could you provide a more detail?  More information will 
> certainly help!

main drawback  that it requires external dependency on DPDK which in some use 
cases we don't want.
Not sure if TAP PMD supports interrupt mode and dynamic interface 
creation/deletion but VPP native driver does.

Also all DPDK drivers have extra cost of converting metadata and dealing with 2 
more cachelines per packet in LLC (not big deal in case of tap)
so unless we really do bad job native driver will always be faster and simpler.

TAP interface is not rocket science, it may be that we are missing few features 
mainly due to my ignorance, but i don't expect anything what cannot be 
added/extended in few dozen lines of code.
I hope people will pick it up and contribute...

-- 
Damjan-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11146): https://lists.fd.io/g/vpp-dev/message/11146
Mute This Topic: https://lists.fd.io/mt/27873113/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] Published: FD.io CSIT-18.10 Open-Source Benchmarking Report

[Maciek Konstantynowicz (mkonstan) via Lists.Fd.Io]

Hi All,

FD.io CSIT-18.10.W45 report has been published on FD.io docs site:

 html: https://docs.fd.io/csit/rls1810/report/
 pdf: https://docs.fd.io/csit/rls1810/report/_static/archive/csit_rls1810.pdf

Many thanks to All in CSIT and FD.io community who contributed, worked
very hard to make CSIT-18.10 happen!

CSIT-18.10 is not the First release able using latest Intel Xeon Skylake
processors.  But it is the first release to present them in all of their
I/O optimized glory. See point Three below for throughput comparison
between Skylake and Haswell testbeds. This is finally, for the Second
time making VPP code shine in a modern software data world :)

Note that FD.io CSIT reports are now versioned using a notation for
ordinal weeks of the year specified in ISO 8601 (http://bit.ly/2yZYI0h).
All FD.io CSIT releases will be suffixed with '.Www', a week number in
the format ww prefixed by the letter 'W'.
This takes care even of leap second by accommodating leap week. :)

FD.io VPP points of note in the report:

 1. VPP release notes
a. Changes in CSIT-18.10: http://bit.ly/2OxSUA6
b. Known issues: http://bit.ly/2REvB9D

 2. VPP performance graphs
a. Throughput: http://bit.ly/2OwQPEn
b. Speedup Multi-Core: http://bit.ly/2zztqgi
c. Latency: http://bit.ly/2F9j4K6

 3. VPP performance comparisons
a. VPP-18.10 vs. VPP-18.04: http://bit.ly/2RGBvXM
b. 3-Node Skylake vs. 3-Node Haswell testbeds: http://bit.ly/2DqjrOJ

 4. VPP performance test details
a. Detailed results: http://bit.ly/2yVWMGf
b. Configuration: http://bit.ly/2qy7w8U
c. Run-time telemetry: http://bit.ly/2PQlkKw

DPDK Testpmd and L3fwd performance sections follow similar structure.

 1. DPDK release notes: 
   a. Changes in CSIT-18.10: http://bit.ly/2Oy7E1U
   b. Known issues: http://bit.ly/2RGGE22

 2. DPDK performance graphs for Testpmd and L3fwd demonstrator apps
   a. Throughput: http://bit.ly/2PeSlAz
   b. Latency: http://bit.ly/2z5OSdx

Functional tests, including VPP_Device (functional device tests),
VPP_VIRL, HoneyComb, NSH_SFC and DMM are all included in the report.

Welcome all comments, best by email to csit-...@lists.fd.io.

Cheers,
-Maciek
On behalf of FD.io CSIT Open-Source Benchmarking Project.

P.S. CSIT-18.10 report includes Ever More benchmarking data, more graphs
(744 on the last count) that are now more orderly and more beautifully
presented than Ever. Report HTML and PDF layouts got revamped more than
a bit, striving to improve presentation clarity and browsability. And
equally portability and consumability got improved. We are aiming the
FD.io CSIT Open-Source Benchmarking reports to be consumable by
everyone. And by everyone we mean Everyone. Those who code it, those who
test it, those who use it, those who want to use it but are afraid to
touch it. Everyone who likes it.

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11147): https://lists.fd.io/g/vpp-dev/message/11147
Mute This Topic: https://lists.fd.io/mt/28024401/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [**EXTERNAL**] [vpp-dev] TAP v1/v2 API questions

[Coulson, Ken]

Latest comments inline in yellow below.


From: Damjan Marion 
Date: Wednesday, November 7, 2018 at 7:34 AM
To: "Coulson, Ken" 
Cc: "vpp-dev@lists.fd.io" 
Subject: Re: [**EXTERNAL**] [vpp-dev] TAP v1/v2 API questions


--
Damjan


On 7 Nov 2018, at 16:01, Coulson, Ken 
mailto:kcoul...@ciena.com>> wrote:

Our normal pattern for forwarding plane entities is create/modify/delete, the 
TAP v1 API fit nicely with that.
The desire is to support changing the interface after it's created without 
deleting and re-creating to avoid as much potential traffic disruption as 
possible.
An example is MTU (which neither v1 or v2 support).  The desired maximum MTU 
may not be known at creation time or may change depending on routing so a 
modify capability seems appropriate.

VPP have universal API (hw_interface_set_mtu ) for setting MTU on all 
interfaces.
  So the intent is that from cli the following would set mtu?  ‘set 
interface mtu 2000 tap0’ does not generate any errors but does not appear to do 
anything.  ‘ifconfig tap0’ still shows default mtu of 1500.


So be specific on your questions, want to control MTU and admin/link 
enable/disable, would like to know link status without polling.

VPP have universal API (sw_interface_set_flags ) for setting link and admin 
states.
You can subscribe to interface events with want_interface_events API to be 
notified on change.
  I’ll give this a try.



DPDK TAP PMD (and vHost PMD) provide the same APIs as physical devices using 
the rte_eth_dev_*() functions allowing the next layer of code to very similar 
for physical and virtual devices.

In some cases we run vPP without DPDK, DPDK is just one device driver option 
for VPP.
  Good reason!  I hadn’t considered that.


DPDK TAP PMD supports MTU,

see above


jumbo frames

we support jumbo frames with tapv2.
  Do you have some cli examples?  I have tried with stock vpp and 
as the mtu set does not appear to work I’m unclear how jumbos would work.


, extended statistics,

what exact extended statistic you are looking for?
   DPDK maintains some statistics for TAPs (and vHost) but since 
they are software counters and cost cpu cycles to maintain the list is pretty 
thin.
   There are rx packets,bytes,errors,no_mem_buff and tx 
packets,bytes,errors.
   An example of an error I don’t seem to see is if the rx packet 
is greater than mtu.


link status change notifications

see above


plus likely other features.

like?


  These seem to me to be an advantage and not a drawback.
I'm curious about the drawbacks mentioned and advantages of the VPP 
implementation, could you provide a more detail?  More information will 
certainly help!

main drawback  that it requires external dependency on DPDK which in some use 
cases we don't want.
Not sure if TAP PMD supports interrupt mode and dynamic interface 
creation/deletion but VPP native driver does.
I believe TAP PMD does not support an ‘interrupt’ mode.  It’s actually not 
really a true interrupt but a signal indicating the driver has packets.

Also all DPDK drivers have extra cost of converting metadata and dealing with 2 
more cachelines per packet in LLC (not big deal in case of tap)
so unless we really do bad job native driver will always be faster and simpler.

TAP interface is not rocket science, it may be that we are missing few features 
mainly due to my ignorance, but i don't expect anything what cannot be 
added/extended in few dozen lines of code.
I hope people will pick it up and contribute...

--
Damjan
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11148): https://lists.fd.io/g/vpp-dev/message/11148
Mute This Topic: https://lists.fd.io/mt/27873113/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [**EXTERNAL**] [vpp-dev] TAP v1/v2 API questions

[Damjan Marion via Lists.Fd.Io]

-- 
Damjan

> On 7 Nov 2018, at 19:36, Coulson, Ken  wrote:
> 
> Latest comments inline in yellow below.
>  
>  
> From: Damjan Marion mailto:dmar...@me.com>>
> Date: Wednesday, November 7, 2018 at 7:34 AM
> To: "Coulson, Ken" mailto:kcoul...@ciena.com>>
> Cc: "vpp-dev@lists.fd.io "  >
> Subject: Re: [**EXTERNAL**] [vpp-dev] TAP v1/v2 API questions
>  
>  
> -- 
> Damjan
> 
> 
>> On 7 Nov 2018, at 16:01, Coulson, Ken > > wrote:
>>  
>> Our normal pattern for forwarding plane entities is create/modify/delete, 
>> the TAP v1 API fit nicely with that.
>> The desire is to support changing the interface after it's created without 
>> deleting and re-creating to avoid as much potential traffic disruption as 
>> possible.
>> An example is MTU (which neither v1 or v2 support).  The desired maximum MTU 
>> may not be known at creation time or may change depending on routing so a 
>> modify capability seems appropriate.
>  
> VPP have universal API (hw_interface_set_mtu ) for setting MTU on all 
> interfaces.
>   So the intent is that from cli the following would set mtu?  
> ‘set interface mtu 2000 tap0’ does not generate any errors but does not 
> appear to do anything.  ‘ifconfig tap0’ still shows default mtu of 1500.

TAP interface have 2 ends, one is on VPP another one is in linux. that API/CLI 
sets MTU in VPP->linux direction. In opposite direction you need to use linux 
tools to set MTU for linux->VPP direction.
We may consider sending netlink message to kernel asking for different MTU on 
remote side, but i'm not sure if that is right thing to do or not, as MTU in 
theory can bi different on each end


> 
> 
>> So be specific on your questions, want to control MTU and admin/link 
>> enable/disable, would like to know link status without polling.
>  
> VPP have universal API (sw_interface_set_flags ) for setting link and admin 
> states.
> You can subscribe to interface events with want_interface_events API to be 
> notified on change.
>   I’ll give this a try.
> 
> 
>>  
>> DPDK TAP PMD (and vHost PMD) provide the same APIs as physical devices using 
>> the rte_eth_dev_*() functions allowing the next layer of code to very 
>> similar for physical and virtual devices.
>  
> In some cases we run vPP without DPDK, DPDK is just one device driver option 
> for VPP.
>   Good reason!  I hadn’t considered that.
> 
> 
>> DPDK TAP PMD supports MTU,
>  
> see above
> 
> 
>> jumbo frames
>  
> we support jumbo frames with tapv2.
>   Do you have some cli examples?  I have tried with stock vpp and 
> as the mtu set does not appear to work I’m unclear how jumbos would work.

See above... You can manually change MTU on linux side with "ifconfig" or "ip 
link".

-- 
Damjan



-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11149): https://lists.fd.io/g/vpp-dev/message/11149
Mute This Topic: https://lists.fd.io/mt/27873113/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] inet raw socket

[Ole Troan]

> Some protocols or functions need inet raw socket to transmit and receive 
> packets such as ping/vrrp/pim, and I want to debug these functions.

You are still leaving me guessing at what you are trying to do.

You can implement these inside of VPP, where you obviously don’t need or use a 
socket API.

Sounds like you are talking about having these protocols implemented on the 
control plane side.
There are many ways of integrating a control plane with VPP.
With e.g. a port mirroring approach, you would use Linux applications 
unchanged. That’s not necessarily where I would go. We have some ideas around 
improvements for the CP/DP interface, and I’m happy to engage in a discussion 
with that, but you would have to share what problem you are trying to solve in 
more detail.

Best regards,
Ole

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#11150): https://lists.fd.io/g/vpp-dev/message/11150
Mute This Topic: https://lists.fd.io/mt/27857220/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] [tsc] Project Proposal for Sweetcomb

[Ni, Hongjun]

Welcome Cisco to join Sweetcomb project!

https://wiki.fd.io/view/Project_Proposals/Sweetcomb

Thanks a lot,
Hongjun

From: t...@lists.fd.io [mailto:t...@lists.fd.io] On Behalf Of Ni, Hongjun
Sent: Thursday, November 1, 2018 8:40 AM
To: Ni, Hongjun ; Edward Warnicke 
Cc: t...@lists.fd.io; vpp-dev@lists.fd.io; Wang, Drenfong 
; ??? ; 
chen...@huachentel.com; lizhuo...@cmhi.chinamobile.com; ??? 
; zhijl@chinatelecom.cn; changlin...@nxp.com; Wang 
Tianyi ; davidfgao(?? ; 
lixin...@huachentel.com; jingqing@alibaba-inc.com; gu.ji...@zte.com.cn
Subject: Re: [vpp-dev] [tsc] Project Proposal for Sweetcomb


Welcome ZTE to join!

Thanks,
Hongjun

From: vpp-dev@lists.fd.io 
[mailto:vpp-dev@lists.fd.io] On Behalf Of Ni, Hongjun
Sent: Thursday, October 25, 2018 9:22 AM
To: Edward Warnicke mailto:hagb...@gmail.com>>
Cc: t...@lists.fd.io; 
vpp-dev@lists.fd.io; Wang, Drenfong 
mailto:drenfong.w...@intel.com>>; ??? 
mailto:wangchuan...@huachentel.com>>; 
chen...@huachentel.com; 
lizhuo...@cmhi.chinamobile.com; ??? 
mailto:lihf...@chinaunicom.cn>>; 
zhijl@chinatelecom.cn; 
changlin...@nxp.com; Wang Tianyi 
mailto:tianyi.w...@tieto.com>>; davidfgao(?? 
mailto:davidf...@tencent.com>>; 
lixin...@huachentel.com; 
jingqing@alibaba-inc.com
Subject: Re: [vpp-dev] [tsc] Project Proposal for Sweetcomb

Hi all,

Some guys are asking for the original code in private.
Here is our answer:


We are working on reworking the original code, and doing internal legal review.

When it is done,  we will submit the code to FD.io community for IPR review in 
one or two weeks.

Thanks a lot,
Hongjun

From: vpp-dev@lists.fd.io 
[mailto:vpp-dev@lists.fd.io] On Behalf Of Edward Warnicke
Sent: Tuesday, October 23, 2018 9:47 PM
To: Ni, Hongjun mailto:hongjun...@intel.com>>
Cc: t...@lists.fd.io; 
vpp-dev@lists.fd.io; Wang, Drenfong 
mailto:drenfong.w...@intel.com>>; ??? 
mailto:wangchuan...@huachentel.com>>; 
chen...@huachentel.com; 
lizhuo...@cmhi.chinamobile.com; ??? 
mailto:lihf...@chinaunicom.cn>>; 
zhijl@chinatelecom.cn; 
changlin...@nxp.com; Wang Tianyi 
mailto:tianyi.w...@tieto.com>>; davidfgao(?? 
mailto:davidf...@tencent.com>>; 
lixin...@huachentel.com; 
jingqing@alibaba-inc.com
Subject: Re: [vpp-dev] [tsc] Project Proposal for Sweetcomb

I look forward to it :)

Ed

On Tue, Oct 23, 2018 at 8:42 AM Ni, Hongjun 
mailto:hongjun...@intel.com>> wrote:
Hi Ed,

OK. I or some project proposer will join the Nov 8 8am PT TSC meeting and 
present the proposal.

Thanks a lot,
Hongjun

From: t...@lists.fd.io 
[mailto:t...@lists.fd.io] On Behalf Of Edward Warnicke
Sent: Tuesday, October 23, 2018 9:15 PM
To: Ni, Hongjun mailto:hongjun...@intel.com>>
Cc: t...@lists.fd.io; 
vpp-dev@lists.fd.io; Wang, Drenfong 
mailto:drenfong.w...@intel.com>>; ??? 
mailto:wangchuan...@huachentel.com>>; 
chen...@huachentel.com; 
lizhuo...@cmhi.chinamobile.com; ??? 
mailto:lihf...@chinaunicom.cn>>; 
zhijl@chinatelecom.cn; 
changlin...@nxp.com; Wang Tianyi 
mailto:tianyi.w...@tieto.com>>; davidfgao(?? 
mailto:davidf...@tencent.com>>; 
lixin...@huachentel.com; 
jingqing@alibaba-inc.com
Subject: Re: [tsc] Project Proposal for Sweetcomb

Hongjun,

Thank you for the proposal :)  Per the FD.io technical charter, all proposals 
must be out for public review for two weeks prior to approval by the TSC.  I 
believe this makes Nov 8 the earliest TSC meeting where we could approve at the 
TSC.  Would you like to schedule the project creation review for the Nov 8 8am 
PT TSC meeting?  Will you (or other proposers of the project) be able to make 
that time to present the proposal?

Ed

On Tue, Oct 23, 2018 at 7:01 AM Ni, Hongjun 
mailto:hongjun...@intel.com>> wrote:
Hello FD.io TSCs

Please accept this project proposal for Sweetcomb for consideration.
https://wiki.fd.io/view/Project_Proposals/Sweetcomb

This project has nine founding companies:
Intel, HuachenTel, China Mobile, China Unicom, China Telecom, NXP, Tieto, 
Tencent, Alibaba.

If possible, I would like to present this on TSC meeting.

Thanks,
Hongjun

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#8