[vpp-dev] VPP Memory usage

2018-08-18 Thread Rubina Bianchi
Hi dear VPP


I configured vpp stable/1807 and added permit+reflect acl on input and output 
of my network interfaces. I configured vpp with 9 cpu (1 main and 8 worker 
cpu). My init.conf is:


vppctl>

set acl-plugin session table max-entries 4000
set acl-plugin session table hash-table-buckets 100
set acl-plugin session table hash-table-memory 17179869184
set acl-plugin session timeout udp idle 20
set acl-plugin session timeout tcp idle 120
set acl-plugin session timeout tcp transient 30


vpp_api_test>

acl_add_replace permit
acl_add_replace permit+reflect

acl_interface_add_del TenGigabitEthernet3/0/0 add output acl 1
acl_interface_add_del TenGigabitEthernet3/0/1 add output acl 1
acl_interface_add_del TenGigabitEthernet3/0/0 add input acl 1
acl_interface_add_del TenGigabitEthernet3/0/1 add input acl 1

exec set interface l2 bridge TenGigabitEthernet3/0/0 1
exec set interface l2 bridge TenGigabitEthernet3/0/1 1
exec set int state TenGigabitEthernet3/0/0 up
exec set int state TenGigabitEthernet3/0/1 up

My startup.conf is pasted in this link: https://paste.ubuntu.com/p/MhQDyqF6Xd/


I used Trex as traffic generator as following:

./t-rex-64 --cfg cfg/trex_config.yaml  -f cap2/sfr.yaml -m 50 -c 3 -d 3600 -p


During execution of my test, Total-rx continuously decreased and after a while, 
it reached to 0. I checked vpp status and it got SIGKILL signal from OS.

I monitored vpp memory and it was increasing until it crashed.

Does acl_plugin session management have any memory leak problem?


Regards,

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

View/Reply Online (#10213): https://lists.fd.io/g/vpp-dev/message/10213
Mute This Topic: https://lists.fd.io/mt/24729023/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-


[vpp-dev] FD.io CSIT-18.07 v1.0 report published

2018-08-18 Thread Maciek Konstantynowicz (mkonstan) via Lists.Fd.Io
Hi All,

FD.io CSIT-18.07 report has been published on FD.io docs site:

  html: https://docs.fd.io/csit/rls1807/report/
  pdf: https://docs.fd.io/csit/rls1807/report/_static/archive/csit_rls1807.pdf

Many thanks to All in CSIT and FD.io community who contributed and made
CSIT-18.07 happen!

CSIT-18.07 is the first release able to use the brand new FD.io labs and
testing servers equipped the latest Intel Xeon Skylake processors. See
point 3. below for throughput comparison between Skylake and Haswell
testbeds.
This is finally making VPP code shine in a modern compute world :)

The labs were long time in the making and they have finally
materialized. Thanks again to All who made it happen!

CSIT-18.07 report includes much more test results data and report html
layout got revamped quite a bit, striving to improve presentation
clarity and browsability. Test environment descriptions also got a
substantial makeover.

VPP points of note in the report:

  1. VPP release notes
a. Changes in CSIT-18.07 [1]
b. Known issues [2]

  2. VPP performance graphs
a. Throughput [3]
b. Speedup Multi-Core [4]
c. Latency [5]

  3. VPP performance comparisons
b. VPP-18.07 vs. VPP-18.04 [6]
c. 3-Node Skylake vs. 3-Node Haswell testbed types [7]

  4. VPP performance test details
a. Results [8]
b. Configuration [9]
c. Run-time telemetry [10]

DPDK Testpmd and L3fwd performance sections follow similar structure.

Functional tests, including VPP, 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.

[1] 
https://docs.fd.io/csit/rls1807/report/vpp_performance_tests/csit_release_notes.html#changes-in-csit-release
[2] 
https://docs.fd.io/csit/rls1807/report/vpp_performance_tests/csit_release_notes.html#known-issues

[3] 
https://docs.fd.io/csit/rls1807/report/vpp_performance_tests/packet_throughput_graphs/index.html
[4] 
https://docs.fd.io/csit/rls1807/report/vpp_performance_tests/throughput_speedup_multi_core/index.html
[5] 
https://docs.fd.io/csit/rls1807/report/vpp_performance_tests/packet_latency_graphs/index.html

[6] 
https://docs.fd.io/csit/rls1807/report/vpp_performance_tests/comparisons/current_vs_previous_release.html
[7] 
https://docs.fd.io/csit/rls1807/report/vpp_performance_tests/comparisons/3n-skx_vs_3n-hsw_testbeds.html

[8] 
https://docs.fd.io/csit/rls1807/report/detailed_test_results/vpp_performance_results/index.html
[9] 
https://docs.fd.io/csit/rls1807/report/test_configuration/vpp_performance_configuration/index.html
[10] 
https://docs.fd.io/csit/rls1807/report/test_operational_data/vpp_performance_operational_data/index.html-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#10212): https://lists.fd.io/g/vpp-dev/message/10212
Mute This Topic: https://lists.fd.io/mt/24677295/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-


Re: [vpp-dev] CMake

2018-08-18 Thread Damjan Marion via Lists.Fd.Io


> On 18 Aug 2018, at 17:36, Andrew Pinski  wrote:
> 
> On Sat, Aug 18, 2018 at 3:38 AM Damjan Marion via Lists.Fd.Io
> mailto:dmarion=me@lists.fd.io>> wrote:
>> [snip]
>> Feedback, issue reports and contributions to get CMake be a full replacement 
>> for autotools are welcome.
> 
> I think this is backwards.  CMake is worse than the autotools when it
> comes to build systems maintaince and debugging.

Unless you have some data points, it is just another opinion.
My (another) opinion is that autotools is one of ugliest piece of software
I had to deal with, and it is extremely hard to maintain and debug.
Beside that it is damn slow…

To support my opinion with some data points:
Clean VPP build (with cache cleared)  with make takes 14 seconds, same thing 
with autotools takes 54 seconds.

Please note that we definitely don’t need another autotools vs cmake discussion 
here.
If you are able to show same speed and simplicity with autotools in this 
project, I will be more than happy to take into your patch and re-consider use 
of cmake.

> Can someone explain to me the benifit to cmake over autotools?
> Because I see none.


Speed, much simpler configs, specially for people who just need to add new 
plugin into tree, embedded test and packaging.

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

View/Reply Online (#10211): https://lists.fd.io/g/vpp-dev/message/10211
Mute This Topic: https://lists.fd.io/mt/24665874/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-


Re: [vpp-dev] CMake

2018-08-18 Thread Andrew Pinski
On Sat, Aug 18, 2018 at 3:38 AM Damjan Marion via Lists.Fd.Io
 wrote:
>
>
> Dear all,
>
> Yesterday we merged a patch which introduces CMake build system in VPP.
>
> https://git.fd.io/vpp/commit/?id=612dd6a9
>
> It is initial patch, and additional work is needed, but in general it manages 
> to compile VPP binaries significantly faster
> than what we have today with autotools.
>
> In addition it brings following benefits which we can leverage in the future:
>  - use of CPack to generate deb/rpms vs. hardcoded scripts we have today
>  - much simpler and smaller build definitions specially for plugins
>  - support for different build types (Debug, Release, Release with symbols, 
> Coverage, ...)
>  - support for external components (i.e. we can likely build dpdk and his 
> dependencies)
>
> Autotools is still default way to run verify jobs and build images, but 
> people can give CMake a try in few ways:
>
> Step 0: "apt-get install cmake ninja-build"
>
> - "make {build,rebuild,build-release,rebuild-release}" are already using 
> cmake as default, old behaviour can be restored by saying "make 
> vpp_uses_cmake=no ..."
>
> - make -C build-root vpp_uses_cmake=yes PLATFORM=vpp TAG={vpp|vpp_debug}
>
> - uncommenting "vpp_uses_cmake=yes" in build-data/platforms/vpp.mk
>
> - without ebuild:
> mkdir _build
> cd _build
> cmake [-G Ninja] /path/to/vpp/src -DCMAKE_BUILD_TYPE={Release|Debug}
> (optionally "ccmake ." to change parameters)
> {ninja|make}
>
> ./bin/vpp unix interactive
>
> Building packages (incomplete / work in progress):
> $ ninja package
> [0/1] Run CPack packaging tool...
> CPack: Create package using DEB
> CPack: Install projects
> CPack: - Install project: vpp
> CPack: -   Install component: dev
> CPack: -   Install component: plugins
> CPack: -   Install component: vpp
> CPack: Create package
> CPack: - package: /home/damarion/tmp/_build/vpp-dev.deb generated.
> CPack: - package: /home/damarion/tmp/_build/vpp-plugins.deb generated.
> CPack: - package: /home/damarion/tmp/_build/vpp.deb generated.
>
> Feedback, issue reports and contributions to get CMake be a full replacement 
> for autotools are welcome.

I think this is backwards.  CMake is worse than the autotools when it
comes to build systems maintaince and debugging.
Can someone explain to me the benifit to cmake over autotools?
Because I see none.

Thanks,
Andrew Pinski

>
> Thanks,
>
> --
> Damjan
>
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
>
> View/Reply Online (#10209): https://lists.fd.io/g/vpp-dev/message/10209
> Mute This Topic: https://lists.fd.io/mt/24665874/912176
> Group Owner: vpp-dev+ow...@lists.fd.io
> Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [pins...@gmail.com]
> -=-=-=-=-=-=-=-=-=-=-=-
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#10210): https://lists.fd.io/g/vpp-dev/message/10210
Mute This Topic: https://lists.fd.io/mt/24665874/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-


[vpp-dev] CMake

2018-08-18 Thread Damjan Marion via Lists.Fd.Io

Dear all,

Yesterday we merged a patch which introduces CMake build system in VPP.

https://git.fd.io/vpp/commit/?id=612dd6a9

It is initial patch, and additional work is needed, but in general it manages 
to compile VPP binaries significantly faster 
than what we have today with autotools.

In addition it brings following benefits which we can leverage in the future:
 - use of CPack to generate deb/rpms vs. hardcoded scripts we have today
 - much simpler and smaller build definitions specially for plugins
 - support for different build types (Debug, Release, Release with symbols, 
Coverage, ...)
 - support for external components (i.e. we can likely build dpdk and his 
dependencies)

Autotools is still default way to run verify jobs and build images, but people 
can give CMake a try in few ways:

Step 0: "apt-get install cmake ninja-build"

- "make {build,rebuild,build-release,rebuild-release}" are already using cmake 
as default, old behaviour can be restored by saying "make vpp_uses_cmake=no ..."

- make -C build-root vpp_uses_cmake=yes PLATFORM=vpp TAG={vpp|vpp_debug} 

- uncommenting "vpp_uses_cmake=yes" in build-data/platforms/vpp.mk

- without ebuild:
mkdir _build
cd _build
cmake [-G Ninja] /path/to/vpp/src -DCMAKE_BUILD_TYPE={Release|Debug}
(optionally "ccmake ." to change parameters)
{ninja|make}

./bin/vpp unix interactive

Building packages (incomplete / work in progress):
$ ninja package
[0/1] Run CPack packaging tool...
CPack: Create package using DEB
CPack: Install projects
CPack: - Install project: vpp
CPack: -   Install component: dev
CPack: -   Install component: plugins
CPack: -   Install component: vpp
CPack: Create package
CPack: - package: /home/damarion/tmp/_build/vpp-dev.deb generated.
CPack: - package: /home/damarion/tmp/_build/vpp-plugins.deb generated.
CPack: - package: /home/damarion/tmp/_build/vpp.deb generated.

Feedback, issue reports and contributions to get CMake be a full replacement 
for autotools are welcome.

Thanks,

-- 
Damjan

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

View/Reply Online (#10209): https://lists.fd.io/g/vpp-dev/message/10209
Mute This Topic: https://lists.fd.io/mt/24665874/21656
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 crashes on multi core

2018-08-18 Thread Mehran Memarnejad
Here is the system characteristics:

==
NODE: DUT1

CPU:
  Model name:Intel(R) Xeon(R) CPU E5-2650 v2 @ 2.60GHz
  CPU(s):16
  Thread(s) per core:2
  Core(s) per socket:8
   Socket(s):1
   NUMA node0 CPU(s):0-15
 CPU max MHz:3400.
 CPU min MHz:1200.
 SMT:Enabled

VPP Threads: (Name: Cpu Number)
  vpp_wk_2  : 3
  vpp_main  : 1
  vpp_wk_1  : 2
  vpp_wk_0  : 0
  vpp_stats : 0

Grub Command Line:
  Current: BOOT_IMAGE=/boot/vmlinuz-4.4.0-127-generic
root=UUID=c1dc6de4-d9fe-46ac-b679-58576c59a655 ro net.ifnames=0
biosdevname=0
  Configured: GRUB_CMDLINE_LINUX_DEFAULT=""

Huge Pages:
  Total System Memory   : 16396212 kB
  Total Free Memory : 13334516 kB
  Actual Huge Page Total: 1024
  Configured Huge Page Total: 1024
  Huge Pages Free   : 971
  Huge Page Size: 2048 kB

Devices:

Devices with link up (can not be used with VPP):
:0a:00.0mgmt1 82583V Gigabit Network Connection

No devices bound to kernel drivers

Devices bound to DPDK drivers:
:33:00.3  82580 Gigabit Network Connection
:0b:00.0  82583V Gigabit Network Connection
:32:00.1  82580 Gigabit Network Connection
:32:00.3  82580 Gigabit Network Connection
:33:00.0  82580 Gigabit Network Connection
:33:00.1  82580 Gigabit Network Connection
:33:00.2  82580 Gigabit Network Connection
:32:00.0  82580 Gigabit Network Connection
:06:00.0  82580 Gigabit Network Connection
:06:00.1  82580 Gigabit Network Connection
:06:00.2  82580 Gigabit Network Connection
:06:00.3  82580 Gigabit Network Connection
:05:00.1  82580 Gigabit Network Connection
:05:00.0  82580 Gigabit Network Connection
:05:00.3  82580 Gigabit Network Connection
:05:00.2  82580 Gigabit Network Connection
:01:00.1  82599ES 10-Gigabit SFI/SFP+
Network Connection
:01:00.0  82599ES 10-Gigabit SFI/SFP+
Network Connection
:32:00.2  82580 Gigabit Network Connection

Devices in use by VPP:
Name   Socket RXQs RXDescs TXQs TXDescs
GigabitEthernet32/0/0   01102441024
GigabitEthernet32/0/1   01102441024
GigabitEthernet32/0/2   01102441024
GigabitEthernet32/0/3   01102441024
GigabitEthernet33/0/0   01102441024
GigabitEthernet33/0/1   01102441024
GigabitEthernet33/0/2   01102441024
GigabitEthernet33/0/3   01102441024
GigabitEthernet5/0/001102441024
GigabitEthernet5/0/101102441024
GigabitEthernet5/0/201102441024
GigabitEthernet5/0/301102441024
GigabitEthernet6/0/001102441024
GigabitEthernet6/0/101102441024
GigabitEthernet6/0/201102441024
GigabitEthernet6/0/301102441024
GigabitEtherneta/0/001102411024
GigabitEthernetb/0/001102411024
TenGigabitEthernet1/0/0 01102441024
TenGigabitEthernet1/0/1 01102441024

VPP Service Status:
  active (running)

==

VPP : the latest one (18.11)
OS : Ubuntu 16.04.4 LTS


On Fri, Aug 17, 2018 at 5:02 PM, Dave Barach (dbarach) 
wrote:

> Please refer to https://wiki.fd.io/view/VPP/BugReports for bug reporting,
> also https://wiki.fd.io/view/VPP/Configuration_Tool
>
>
>
> If you expect folks to *volunteer* to help you: write up your system
> configuration, vpp version and configuration, and so forth to a reasonable
> standard.
>
>
>
> *From:* vpp-dev@lists.fd.io  *On Behalf Of *Mehran
> Memarnejad
> *Sent:* Friday, August 17, 2018 5:51 AM
> *To:* vpp-dev@lists.fd.io
> *Subject:* [vpp-dev] VPP crashes on multi core
>
>
>
> Hi,
>
> In default configuration of vpp on single core, everything seems ok but
> when I want to get the most out of my hardware resources and modify vpp
> startup config, it crashes at some stages.
>
> Here is some carshes:
>
> 1- I set interface's IP and then when I plugged cables in ports, It
> suddenly carshes and