Re: [vpp-dev] dpdk-input : serious load imbalance

Yeah, I did not expect change, but at least we know that RSS is confiugred on this card: rss avail: ipv4 ipv4-tcp ipv4-udp ipv6 ipv6-tcp ipv6-udp port vxlan geneve nvgre rss active:ipv4 ipv4-tcp ipv4-udp ipv6 ipv6-tcp ipv6-udp I'm affraid this

Re: [vpp-dev] Worker Thread Dead Lock on NAT44 IPFIX

Hi, NAT process all ipfix events in main thread. I think this should be reworked to per worker thread processing like flowprobe plugin Matus From: vpp-dev@lists.fd.io On Behalf Of david.leitch@gmail.com Sent: Wednesday, December 19, 2018 8:29 AM To: vpp-dev@lists.fd.io Subject: Re: [vpp-

[vpp-dev] regarding cross compiling

Hi all, I was confused with cross compiling, when I built vpp with the new version(18.10) using arm build tools. The compiling environment is listed below: - vpp 18.10 host : ubuntu 16.04 , x86-64 cpu target :

Re: [vpp-dev] regarding cross compiling

Somebody needs to spend a bit of time to teach CMake how to properly cross-compile -- Damjan > On 19 Dec 2018, at 11:19, saint_sun 孙 via Lists.Fd.Io > wrote: > > Hi all, > I was confused with cross compiling, when I built vpp with the new > version(18.10) using arm build tools. >

Re: [vpp-dev] regarding cross compiling

Hi Damjan, >>Somebody needs to spend a bit of time to teach CMake how to properly >>cross-compile Correct me if I am wrong but I think cross-compilation support was there before CMake transition. Also I am not finding capability to link VPP with externally compiled dpdk? Am I correct or mi

Re: [vpp-dev] regarding cross compiling

i > On 19 Dec 2018, at 14:00, Saxena, Nitin wrote: > > Hi Damjan, > > >>Somebody needs to spend a bit of time to teach CMake how to properly > >>cross-compile > > Correct me if I am wrong but I think cross-compilation support was there > before CMake transition. > I guess so, never use

Re: [vpp-dev] regarding cross compiling

On Wed, Dec 19, 2018 at 8:54 AM Damjan Marion via Lists.Fd.Io wrote: > > i > > On 19 Dec 2018, at 14:00, Saxena, Nitin wrote: > > Hi Damjan, > > >>Somebody needs to spend a bit of time to teach CMake how to properly > >>cross-compile > > Correct me if I am wrong but I think cross-compilation

Re: [vpp-dev] regarding cross compiling

According to this, https://cmake.org/cmake/help/latest/manual/cmake-toolchains.7.html#cross-compiling , you just need to set a few more variables for cross compilation. In the old days, that could be done in the ebuild *.mk files under build-data and build-root. However, I believe ebuild was optim

Re: [vpp-dev] Worker Thread Dead Lock on NAT44 IPFIX

hi Matus Thanks for your answer, can you explain more about this issue? You mean that I should rewrite "nat_ipfix_logging.c"  file to works per thread or  "vnet/ipfix-export/flow_report" -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#11697): https

[vpp-dev] NAT workers above 4 completely tanks performance

Hello everyone, This is on "19.01", I've yet to test with 18.10. I've setup dynamic NAT. I've tested this with a clean setup without specifying anything special in startup.conf. The results are irrelevant of other NAT settings except NAT workers. I've tried specifying various networks, single IP

Re: [vpp-dev] Worker Thread Dead Lock on NAT44 IPFIX

Hi, NAT code should be rewritten Matus From: vpp-dev@lists.fd.io On Behalf Of david.leitch@gmail.com Sent: Wednesday, December 19, 2018 4:56 PM To: vpp-dev@lists.fd.io Subject: Re: [vpp-dev] Worker Thread Dead Lock on NAT44 IPFIX hi Matus Thanks for your answer, can you explain more about

Re: [vpp-dev] regarding cross compiling

Please give the instructions at https://cmake.org/cmake/help/latest/manual/cmake-toolchains.7.html#cross-compiling a try and let us know what happens. Suffice it to say that the vpp committer community saw sufficient value in switching to cmake to do so. D. -Original Message- From:

Re: [vpp-dev] regarding cross compiling

On Wed, Dec 19, 2018 at 11:29 AM Dave Barach (dbarach) wrote: > > Please give the instructions at > https://cmake.org/cmake/help/latest/manual/cmake-toolchains.7.html#cross-compiling > a try and let us know what happens. Wow. So much harder to do than what autoconf provides. I think going to

Re: [vpp-dev] regarding cross compiling

Let me reiterate what Dave said: the committer community saw real value in moving to cmake. Those extra build minutes are subtracted from the lives of those who build vpp on a daily basis. Florin > On Dec 19, 2018, at 8:46 AM, Andrew Pinski wrote: > > On Wed, Dec 19, 2018 at 11:29 AM Dave Ba

Re: [vpp-dev] regarding cross compiling

> On 19 Dec 2018, at 16:40, Andrew Pinski wrote: > > On Wed, Dec 19, 2018 at 8:54 AM Damjan Marion via Lists.Fd.Io > wrote: >> >> i >> >> On 19 Dec 2018, at 14:00, Saxena, Nitin wrote: >> >> Hi Damjan, >> Somebody needs to spend a bit of time to teach CMake how to properly cross

Re: [vpp-dev] regarding cross compiling

> On 19 Dec 2018, at 17:46, Andrew Pinski wrote: > > On Wed, Dec 19, 2018 at 11:29 AM Dave Barach (dbarach) > wrote: >> >> Please give the instructions at >> https://cmake.org/cmake/help/latest/manual/cmake-toolchains.7.html#cross-compiling >> a try and let us know what happens. > > Wow.

Re: [vpp-dev] regarding cross compiling

On Wed, Dec 19, 2018 at 12:00 PM Damjan Marion wrote: > > > > > On 19 Dec 2018, at 17:46, Andrew Pinski wrote: > > > > On Wed, Dec 19, 2018 at 11:29 AM Dave Barach (dbarach) > > wrote: > >> > >> Please give the instructions at > >> https://cmake.org/cmake/help/latest/manual/cmake-toolchains.7.h

Re: [vpp-dev] Sanity check re: NAT for same-service mapping

​I've noted a potential modification as to how we do dynamic NAT that would alleviate issues seen by for example banks (they depend on this a lot and both current dynamic NAT implementations of VPP break this). Today, we use a 4-tuple session key for endpoint independent NAT, but, if the source

Re: [vpp-dev] regarding cross compiling

On Wed, Dec 19, 2018 at 12:05 PM Andrew Pinski via Lists.Fd.Io wrote: > > On Wed, Dec 19, 2018 at 12:00 PM Damjan Marion wrote: > > > > > > > > > On 19 Dec 2018, at 17:46, Andrew Pinski wrote: > > > > > > On Wed, Dec 19, 2018 at 11:29 AM Dave Barach (dbarach) > > > wrote: > > >> > > >> Please g

Re: [vpp-dev] regarding cross compiling

-- Damjan > On 19 Dec 2018, at 18:05, Andrew Pinski wrote: > > On Wed, Dec 19, 2018 at 12:00 PM Damjan Marion wrote: >> >> >> >>> On 19 Dec 2018, at 17:46, Andrew Pinski wrote: >>> >>> On Wed, Dec 19, 2018 at 11:29 AM Dave Barach (dbarach) >>> wrote: Please give the instructi

Re: [vpp-dev] regarding cross compiling

-- Damjan > On 19 Dec 2018, at 18:12, Andrew Pinski wrote: > > On Wed, Dec 19, 2018 at 12:05 PM Andrew Pinski via Lists.Fd.Io > wrote: >> >> On Wed, Dec 19, 2018 at 12:00 PM Damjan Marion wrote: >>> >>> >>> On 19 Dec 2018, at 17:46, Andrew Pinski wrote: On Wed, Dec 19, 2

Re: [vpp-dev] regarding cross compiling

> On 19 Dec 2018, at 18:38, Damjan Marion via Lists.Fd.Io > wrote: > > > -- > Damjan > >> On 19 Dec 2018, at 18:12, Andrew Pinski > > wrote: >> >> On Wed, Dec 19, 2018 at 12:05 PM Andrew Pinski via Lists.Fd.Io >> mailto:pinskia=gmail@lists.fd.io>> wrote: >>> >

Re: [vpp-dev] regarding cross compiling

Hi Andrew. As a non-cisco observer to group, I respectfully disagree. I'm sure the old code is still available in the repo for your use, should you wish to use it. You however took the position that the community here needs to take their resources and use them to maintain a solution that you fav

Re: [vpp-dev] regarding cross compiling

Thank you Paul, very well written, and please be assured that we are not looking into this as a cisco project and we are happy to se non-cisco people joining and contributing. Point here is that cmake story was discussed in the community, people had enough time to comment, and all comments were

Re: [vpp-dev] dpdk-input : serious load imbalance

The result is the same. The 2rd queue only received 24 packets as it is in 18.07. version : vpp v18.10-19~ga8e3001-dirty vpp# show hardware-interfaces NameIdx Link Hardware VirtualFunctionEthernet5/0/2 1 up VirtualFunctionEthernet5/0/2 Ethernet ad

Re: [vpp-dev] dpdk-input : serious load imbalance

The "show hardw" is as follow, the statistics may be different from yesterday. vpp# show hardware-interfaces Name Idx Link Hardware VirtualFunctionEthernet5/0/2 1 up VirtualFunctionEthernet5/0/2 Ethernet address 72:62:8a:40:43:12 Cavium ThunderX carrier up full duplex speed 1 mtu 9190 flags:

Re: [vpp-dev] dpdk-input : serious load imbalance

ok , thanks anyway. mik...@yeah.net From: Damjan Marion Date: 2018-12-19 16:04 To: mik...@yeah.net CC: vpp-dev Subject: Re: [vpp-dev] dpdk-input : serious load imbalance Yeah, I did not expect change, but at least we know that RSS is confiugred on this card: rss avail: ipv4 ipv4

Re: [vpp-dev] Status of PPPoE Plugin

Hi Raj, For the issue you mentioned, Alp Arslan in the To list will submit a patch to fix it. To Alp Arslan, Thank you for your great help! Thanks, Hongjun -Original Message- From: vpp-dev@lists.fd.io [mailto:vpp-dev@lists.fd.io] On Behalf Of Raj Sent: Tuesday, December 18, 2018 8:11

Re: [vpp-dev] regarding cross compiling

Thanks to all who have helped, I am trying, will resend the result later. saint_...@aliyun.com From: Dave Barach via Lists.Fd.Io Date: 2018-12-20 00:29 To: Andrew Pinski; dmar...@me.com CC: vpp-dev Subject: Re: [vpp-dev] regarding cross compiling Please give the instructions at https://cmake

Re: [vpp-dev] Sanity check re: NAT for same-service mapping

Hi, NAT address and port allocation is pluggable, you can write your own algorithm and use it instead of default (currently we support two additional port restricted algorithms map-e and port range) Matus From: John Biscevic Sent: Wednesday, December 19, 2018 6:12 PM To: Ole Troan ; Matus Fa

Re: [vpp-dev] Status of PPPoE Plugin

Thanks a lot to both of you! Raj On Thu, Dec 20, 2018 at 6:17 AM Ni, Hongjun wrote: > > Hi Raj, > > For the issue you mentioned, Alp Arslan in the To list will submit a patch to > fix it. > > To Alp Arslan, > Thank you for your great help! > > Thanks, > Hongjun > > -Original Message- >