Re: [EXT] [vpp-dev] 128 byte cache line support

> On 22 Mar 2019, at 09:35, Nitin Saxena wrote: > > Hi Damjan, > > >> Can we conclude that there is no technical arguments? > > On technical aspects: > Changing VLIB structure to 128 byte from current 256 byte (for our target) > In case of forwarding we only use single Dcache line till

Re: [EXT] [vpp-dev] 128 byte cache line support

, March 21, 2019 8:47 PM >> > To: Nitin Saxena >> > Cc: vpp-dev@lists.fd.io; Narayana Prasad Raju Athreya >> > Subject: Re: [EXT] Re: [vpp-dev] 128 byte cache line support >> > >> > >> > >> >> On 21 Mar 2019, at 16:04, Nitin Saxena

Re: [vpp-dev] Status of PPPoE Plugin

roblem was MAC > address resolution for PPPoE sessions. -John > > From: vpp-dev@lists.fd.io On Behalf Of Damjan Marion > via Lists.Fd.Io > Sent: Thursday, March 21, 2019 11:08 AM > To: Ni, Hongjun > Cc: vpp-dev@lists.fd.io > Subject: Re: [vpp-dev] Status of PPPoE Plugin

Re: [EXT] [vpp-dev] 128 byte cache line support

gt; > > > > > >> On 21 Mar 2019, at 16:04, Nitin Saxena >> <mailto:nsax...@marvell.com>> wrote: > >> > >> Hi Damjan, > >> > >> From: vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io> > >> mailto:vpp-dev@lists.fd.i

Re: [EXT] Re: [vpp-dev] 128 byte cache line support

port > > > >> On 21 Mar 2019, at 16:04, Nitin Saxena wrote: >> >> Hi Damjan, >> >> From: vpp-dev@lists.fd.io on behalf of Damjan Marion >> via Lists.Fd.Io >> Sent: Thursday, March 21, 2019 6:03 PM >> To: Nitin Saxena >&g

Re: [EXT] Re: [vpp-dev] 128 byte cache line support

> On 21 Mar 2019, at 16:04, Nitin Saxena wrote: > > Hi Damjan, > > From: vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io> <mailto:vpp-dev@lists.fd.io>> on behalf of Damjan Marion via Lists.Fd.Io > mailto:dmarion=me@lists.fd.io>> > Sent: Th

Re: [vpp-dev] Status of PPPoE Plugin

> On 21 Mar 2019, at 06:20, Ni, Hongjun wrote: > > Hi Raj, > > I think you may check the Ethernet type, if it is PPPoE packet, then MAC > check will be skipped. > > Thanks, > Hongjun > > -Original Message- > From: vpp-dev@lists.fd.io [mailto:vpp-dev@lists.fd.io] On Behalf Of Raj >

Re: [EXT] Re: [vpp-dev] 128 byte cache line support

> On 21 Mar 2019, at 06:51, Nitin Saxena wrote: > > Hi, > > First all sorry for responding late to this mail chain. Please see my answers > inline in blue > > Thanks, > Nitin > > > From: Damjan Marion > Sent: Monday, March 18, 2019 4:48 PM > To: Honnappa Nagarahalli > Cc: vpp-dev; Nitin

Re: [vpp-dev] 128 byte cache line support

> On 15 Mar 2019, at 04:52, Honnappa Nagarahalli > wrote: > > > > Related to change 18278[1], I was wondering if there is really a benefit of > dealing with 128-byte cachelines like we do today. > Compiling VPP with cacheline size set to 128 will basically just add 64 bytes > of unused

Re: [vpp-dev] possible buffer deposit if allocating buffer via CPU on one NUMA and freeing buffer via CPU on another NUMA

-- Damjan On 15 Mar 2019, at 11:42, Lollita Liu mailto:lollita@ericsson.com>> wrote: Hi, Damjan. After reading the buffer management mechanism of master’s latest code, I think there is possibility that some buffers can’t be used anymore. Let me an example to explain the

[vpp-dev] 128 byte cache line support

Related to change 18278[1], I was wondering if there is really a benefit of dealing with 128-byte cachelines like we do today. Compiling VPP with cacheline size set to 128 will basically just add 64 bytes of unused space at the end of each cacheline so vlib_buffer_t for example will grow from

Re: [vpp-dev] I want to disable acl plugin but VPP is not coming up

Will be nice from ABF plugin to check if ACL plugin is loaded and avoid doing something wrong if ACL plugin is not there. -- Damjan > On 14 Mar 2019, at 11:08, Andrew Yourtchenko wrote: > > hi Chetan, > > ABF plugin uses ACL plugin, so you need to disable also abf_plugin.so. > > --a > >

Re: [vpp-dev] VPP plugins development in Go

> On 13 Mar 2019, at 19:42, OMER AHARONIAN wrote: > > Hi, > > My name is Omer and I'm planning to start using VPP as part of a project I'm > working on. I was trying to find an answer to the following question but > without much success. I was hoping you might have the answer. > VPP

Re: [vpp-dev] Build VPP packages on aarch64

On 13 Mar 2019, at 16:10, Lijian Zhang (Arm Technology China) mailto:lijian.zh...@arm.com>> wrote: Yes, Damjan, will submit the patch to gerrit for public review, after finishing internal review inside VPP/Aarch64. Just for my curiosity, why your “internal” reviewers cannot just use gerrit

Re: [vpp-dev] Build VPP packages on aarch64

Dear Lijian, We use gerrit for code review, so please submit your patch there… Thanks, -- Damjan On 13 Mar 2019, at 16:01, Lijian Zhang (Arm Technology China) mailto:lijian.zh...@arm.com>> wrote: Hi Maintainers, The current aarch64 version of VPP package distro in cloud repository

Re: [vpp-dev] Standard VPP functionality for AD4,AD6,AS4,AS6

I bet that more than 90% of subscribers to this list doesn’t know what AD4,AD6,AS4,AS6 means (including myself). Out of code snippets you provided, looks like it is SR related. So might be worth adding existing code author(s) to CC, as maybe they are not reading this list very frequently. “git

Re: [vpp-dev] vlib_buffer_free_inline with CLIB_HAVE_VEC128 will always free buffers one by one, not via the 4 packets batch fast path

On 13 Mar 2019, at 10:19, Lollita Liu mailto:lollita@ericsson.com>> wrote: Hi, Damjan. We found in vlib_buffer_free_inline with CLIB_HAVE_VEC128, buffers will always be freed one by one, not via the 4 packets batch fast path. The phenomenon is the same by calling via

Re: [vpp-dev] NAT handoff node graph question

> On 11 Mar 2019, at 19:53, Ranadip Das wrote: > > Hello, > > Once handoff node decides to handoff a pkt to another core, which node on the > other core picks the pkt up? > I am using 1807, and it seems like nst44-handoff-classify will send the pkt > to either nat44-in2out-worker-handoff

Re: [vpp-dev] VPP Crash in dpdk node

> On 8 Mar 2019, at 17:41, satish.gu...@gmail.com wrote: > > Hi VPP-Dev, > > We are seeing VPP crash with the following stack trace. We believe that this > crash might be occurring during a memif message transaction. > Is this crash a known issue by any chance. Please let us know. Appreciate

Re: [vpp-dev] vmxnet3 devices

> On 7 Mar 2019, at 18:57, Jim Thompson wrote: > > > >> On Mar 7, 2019, at 11:52 AM, Damjan Marion via Lists.Fd.Io >> mailto:dmarion=me@lists.fd.io>> wrote: >> >> >> >>> On 7 Mar 2019, at 18:33, Matthew Smith >

Re: [vpp-dev] vmxnet3 devices

> On 7 Mar 2019, at 18:33, Matthew Smith wrote: > > > Hi all, > > At startup, the DPDK plugin attempts to take over most PCI network devices > which are supported by a PMD and are not in use by the kernel. It seems that > vmxnet3 is an exception to this - if vmxnet3 devices are present,

Re: [vpp-dev] Build failure question

> On 5 Mar 2019, at 15:14, Thomas F Herbert wrote: > > > > > On 03/05/2019 02:37 AM, Damjan Marion wrote: >> >> >>> On 4 Mar 2019, at 22:52, Thomas F Herbert >> > wrote: >>> >>> Damjan et. al. >>> >>> I get build failures when builging from a tarball outside

Re: [vpp-dev] One question about vpp configuration file

> On 5 Mar 2019, at 08:01, Zhang, Yuwei1 wrote: > > Hello All, > I noticed there’s a config option in startup.conf file named > scheduler-priority, from the comments, it seems related to scheduling policy > about main and worker threads, could anybody give me some more detail >

Re: [vpp-dev] Question about the ipsec policy matching

> On 5 Mar 2019, at 04:25, Zhang, Yuwei1 wrote: > > Hello All, > I’m doing some investigation on ipsec in vpp recently, I noticed > that the ipsec policy matching part using a very simple linear search > algorithm which works fine in the case only have a few policy entries, but >

Re: [vpp-dev] Build failure question

> On 4 Mar 2019, at 22:52, Thomas F Herbert wrote: > > Damjan et. al. > > I get build failures when builging from a tarball outside of git tree on > 19.01 > > I documented this in https://jira.fd.io/browse/VPP-1577 > > This is not reproduceable when

Re: [vpp-dev] VPP crashed on servers with unusual numa config

Can you please try: https://gerrit.fd.io/r/#/c/17950/ I don't have such "special" configuration :) Thanks, Damjan > On 28 Feb 2019, at 06:31, Lijian Zhang wrote: > > Hi Maintainers, > I have two supermicro x86 servers with unusual numa configs. The first

Re: [vpp-dev] VPP crashed on servers with unusual numa config

Current code assumes that all numa nodes have memory present, which is reasonable assumption for any production use. It is on my todo list to improve handling of such cases, but it is not 5-minute task, so I cannot promise anything soon. Simply changing from "online" to "has_memory" will not

Re: [vpp-dev] New vpp project committer nomination: Paul Vinciguerra

+1 > On 27 Feb 2019, at 14:20, Dave Barach via Lists.Fd.Io > wrote: > > +1... Dave > > From: vpp-dev@lists.fd.io > On Behalf Of Dave Barach via Lists.Fd.Io > Sent: Wednesday, February 27, 2019 7:38 AM > To: vpp-dev@lists.fd.io

Re: [vpp-dev] One question about "worker-handoff" graph node.

gt; Zhiyong > From: vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io> > [mailto:vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io>] On Behalf Of Damjan > Marion via Lists.Fd.Io > Sent: Thursday, February 21, 2019 6:54 PM > To: Yang, Zhiyong mailto:zhiyong.y...@intel.com>>

Re: [vpp-dev] One question about "worker-handoff" graph node.

iyong <> > <>From: vpp-dev@lists.fd.io [mailto:vpp-dev@lists.fd.io] On Behalf Of Damjan > Marion via Lists.Fd.Io > Sent: Thursday, February 21, 2019 6:54 PM > To: Yang, Zhiyong > Cc: vpp-dev@lists.fd.io > Subject: Re: [vpp-dev] One question about "worker-handof

Re: [vpp-dev] VPP doesn't seem to recognise 25GbE link speed NICs logging warning: (18.01)

> On 22 Feb 2019, at 16:51, chetan bhasin wrote: > > Hi Damajan, > > I need your guidance. > > I am working on vpp 18.01 . > > VPP doesn't seem to recognise 25GbE as per the link speed NICs logging > warning. > > Query1 Does it have any impact on the packet processing ? likely not, it

Re: [vpp-dev] :: Finding right ECCN

> On 20 Feb 2019, at 09:37, Omer Majeed wrote: > > Hi, > > Can anyone help with figuring out the right Export Control Classification > Number ("ECCN") for VPP. We are going through certification and following are > the options > > - EAR99 > - Not Subject to the EAR in accordance with

Re: [vpp-dev] One question about "worker-handoff" graph node.

dev@lists.fd.io [mailto:vpp-dev@lists.fd.io] On Behalf Of Damjan > Marion via Lists.Fd.Io > Sent: Wednesday, February 20, 2019 6:23 PM > To: Yang, Zhiyong > Cc: vpp-dev@lists.fd.io > Subject: Re: [vpp-dev] One question about "worker-handoff" graph node. > > &

Re: [vpp-dev] vpp on raspberry pi

> On 21 Feb 2019, at 10:06, Umberto Fattore wrote: > > > Hello everybody, > > I’m trying to install vpp on my Raspberry Pi. To do this, I downloaded vpp > from the github repository (https://github.com/FDio/vpp > ) and then followed the instruction for >

Re: [vpp-dev] One question about "worker-handoff" graph node.

> On 20 Feb 2019, at 09:49, Zhiyong Yang wrote: > > Hi VPP guys, > There exists a “handoff” graph node in VPP that is named > “worker-handoff”. From my understanding, it looks to be acting as a > distributing packets received role in the multi thread working environment ,

[vpp-dev] gerrit - unable to login

Folks, Myself and at least one more person is not able to login to gerrit.fd.io . Error message is: Authentication unavailable at this time. Please take a a look, -- Damjan -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online

Re: [vpp-dev] [19.01] Multiple tests fail

jvpp is in process of moving to separate fd.io project, so we are not building jars anymore. Those tests should be removed soon... > On 18 Feb 2019, at 07:27, jean-christophe manciot > wrote: > > On Ubuntu disco & Debian buster: > Testing with: > {code:java} > make V=0 PLATFORM=vpp TAG=vpp

Re: [vpp-dev] CPU load on containerized VPP instances #vpp

You need to switch memif to polling or adaptive mode, cli command is “set interface rx-mode [polling|adaptive|interrupt] — Damjan > On Feb 17, 2019, at 9:37 AM, charlesb...@gmail.com wrote: > > Hello all. > > I am attempting to run VPP in LXD managed containers with memif interfaces to >

Re: [vpp-dev] Error Starting VPP (ERROR: This system does not support "SSSE3")

Your vm is not passing host cpuid flags. You likely need to pass “-cpu native” to qemu.. — Damjan > On Feb 13, 2019, at 8:50 PM, Ukponmwan, Itohan > wrote: > > Hi All, > > I am trying to start VPP on a VM running Ubuntu-16.04 but It does not start > and I see the following in the logs;

Re: [vpp-dev] Issue coming with memif

Yeah, unless you are not setting length correctly only on some packets... > On 14 Feb 2019, at 08:05, chetan bhasin wrote: > > Thanks Damjan for looking into this. > > I also review the code for the same but If that is the case then app will not > work properly even first time. > > Thanks,

Re: [vpp-dev] Issue coming with memif

> On 14 Feb 2019, at 07:54, chetan bhasin wrote: > > Hi, > > We are using vpp 18.0.1 . We have created a memif client app , which is > connecting to vpp via memif and sending UDP packet . > > One of our plugin is receiving that packet, processing it and generating an > UDP response. > >

Re: [vpp-dev] VLAN tag-rewrite question

> On 13 Feb 2019, at 09:17, Jon Loeliger wrote: > > VPPers, > > Is it intended that VPP supports placing VLAN tag rewrite rules on any > interface? > > The "Create ACCESS Ports" example on this page: > https://wiki.fd.io/view/VPP/Interconnecting_vRouters_with_VPP >

Re: [vpp-dev] DPDK and SR-IOV #vpp

> On 6 Feb 2019, at 05:37, amir...@rad.com wrote: > > Thanks, > Is there any option to attach \ detach interface from the VPP (without > restarting the service) ? With DPDK interfaces, no. Native AVF driver supports that but that is intel x722/xl710/x710/xxv710 only (and 100G intel E810

Re: [vpp-dev] DPDK and SR-IOV #vpp

> On 5 Feb 2019, at 05:08, amir...@rad.com wrote: > > Hi Damjan, > > My main goal is to use the PF inside the VPP using DPDK in addition to SR-IOV > connection to the VMs. > Currently (before enabling SR-IOV VFs ) all interfaces is connected to VPP > with DPDK using VFIO-PCI driver. > I've

Re: [EXT] [vpp-dev] RFC: buffer manager rework

> On 4 Feb 2019, at 19:38, Jerin Jacob Kollanukkaran wrote: > > On Mon, 2019-02-04 at 19:32 +0100, Damjan Marion wrote: >> >> >>> On 4 Feb 2019, at 14:19, Jerin Jacob Kollanukkaran >> > wrote: >>> >>> On Sun, 2019-02-03 at 21:13 +0100, Damjan Marion wrote:

Re: [EXT] [vpp-dev] RFC: buffer manager rework

> On 4 Feb 2019, at 14:19, Jerin Jacob Kollanukkaran wrote: > > On Sun, 2019-02-03 at 21:13 +0100, Damjan Marion wrote: >> External Email >> >> >>> On 3 Feb 2019, at 20:13, Saxena, Nitin >> > wrote: >>> >>> Hi Damjan, >>> >>> See function

Re: [vpp-dev] ipsec dpdk backend

; > A quick debug session shows it might be a problem of dpdk1902 API changes. > > Will come back with a patch later. Probably need a week. Holiday... > > > 原始邮件 > 主题： Re: [vpp-dev] VPP register node change upper limit > 来自： "Damjan Marion via Lists.F

Re: [vpp-dev] VPP register node change upper limit

eate interfaces that do not need TX > nodes, but I am not sure how to use that. > > Also I can not figure out where the nodes created along with the pppoe > sessions are being used as they do not show up in the “show runtime” or the > trace of packets. > > > > F

Re: [vpp-dev] RFC: buffer manager rework

our hardware do. "It is hardware so it is fast" is not real argument, we need real datapoints before investing time into this area > > Thanks, > Nitin > >> On 03-Feb-2019, at 11:39 PM, Damjan Marion via Lists.Fd.Io >> mailto:dmarion=me@li

Re: [vpp-dev] RFC: buffer manager rework

nt and that you are confident that you can do significantly better, I will be happy to work with you on implementing support for hardware buffer manager. > > Thanks, > Nitin > >> On 03-Feb-2019, at 10:34 PM, Damjan Marion via Lists.Fd.Io >> mailto:dmarion=me@lists.

Re: [vpp-dev] RFC: buffer manager rework

o have problem in scaling, for example if you have hundreds of virtual interfaces... > > Thanks, > Nitin > > From: vpp-dev@lists.fd.io <mailto:vpp-dev@lists.fd.io> <mailto:vpp-dev@lists.fd.io>> on behalf of Damjan Marion via Lists.Fd.Io > mailto:dmarion=me@li

Re: [vpp-dev] vpp-verify-master-centos failures

Problem is likely caused by old multiarch scheme + gcc7 (6 or 8 work fine) + march=(cpu type which doesn't do avx2). https://gerrit.fd.io/r/#/c/17252 Should fix the issue > On 1 Feb 2019, at 18:52, Florin Coras wrote: > > Just a heads up to everyone.

Re: [vpp-dev] gcc versions

> On 2 Feb 2019, at 00:20, Paul Vinciguerra wrote: > > What versions of gcc are supported for building vpp? Pretty much anything newer than 4 -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#12135):

Re: [vpp-dev] Understanding vlib_frame_t

And to answer biffer question, buffer memory is allocated by vpp. — Damjan > On Feb 1, 2019, at 3:51 PM, Dave Barach via Lists.Fd.Io > wrote: > > Next frames are the physical manifestations of graph arcs. Pending frames are > graph scheduler run-queue entries. > > -Original

Re: [vpp-dev] VPP register node change upper limit

e is created so same limitation applies > > Also I can not figure out where the nodes created along with the pppoe > sessions are being used as they do not show up in the “show runtime” or the > trace of packets. > > > > From: Damjan Marion via Lists.Fd.Io <mailto

Re: [vpp-dev] VPP register node change upper limit

> On 1 Feb 2019, at 11:32, Abeeha Aqeel wrote: > > Dear All, > > I am trying to create 64k PPPoE sessions with VPP but VPP crashes after > creating 216 sessions each time. From the system logs it seems that it > crashes while trying to register a node and that node’s index is greater than

Re: [vpp-dev] Connecting Cellular modem (wwan) interface to VPP

can you paste output of: brctl show ip link ip addr > On 31 Jan 2019, at 20:36, amir...@rad.com wrote: > > Hi Mohsin Kazmi, > Thanks for the quick answer. > I'm not sure I got your point, > I've two PCs each one connect to different physical port in a network device > in the middle. > This

Re: [vpp-dev] opensuse expert needed

> On 31 Jan 2019, at 14:12, Marco Varlese wrote: > > Damjan, > > On 1/31/19 12:41 PM, Damjan Marion wrote: >> >> >>> On 31 Jan 2019, at 08:57, Marco Varlese >> <mailto:mvarl...@suse.de>> wrote: >>> >>> Dam

Re: [vpp-dev] Understanding vlib_frame_t

> On 31 Jan 2019, at 15:26, Raj wrote: > > Hello all, > > Continuing my reading of VPP code, I have a query regarding packet handling. > > When a node gets a vector of packets to process, if I understand > correctly, this would be a vlib_frame_t instance. From this > vlib_frame_t instance

Re: [vpp-dev] opensuse expert needed

> On 31 Jan 2019, at 08:57, Marco Varlese wrote: > > Damjan, > > On 1/30/19 3:29 PM, Damjan Marion via Lists.Fd.Io wrote: >> >> Folks, >> >> Anybody can help with understanding why opensuse jobs are failing with: >> >> *13:54:57* &g

Re: [vpp-dev] opensuse expert needed

> On 30 Jan 2019, at 16:26, Paul Vinciguerra wrote: > > I would agree that the error seems to indicate that the build-host has > insufficient /tmp space. > > There is no doubt, that the build system, like the test system, could use > some love. > What is your feeling on moving the CI logic

[vpp-dev] opensuse expert needed

Folks, Anybody can help with understanding why opensuse jobs are failing with: 13:54:57 /w/workspace/vpp-verify-master-osleap15/build-root/rpmbuild/BUILD/vpp-19.04/src/plugins/nsim/nsim.c:620:1: fatal error: error writing to /tmp/ccJaw6Cw.s: Cannot allocate memory 13:54:57 }; 13:54:57 ^

[vpp-dev] PEP8 expert needed

Can somebody with python skills take care for this checkstyle errors, not sure why they started popping out now... Thanks! 19:13:35 /w/workspace/vpp-checkstyle-verify-master/test/test_syslog.py:132:17: E117 over-indented 19:13:35 self.logger.error(ppp("invalid packet:", capture[0]))

Re: [vpp-dev] How do I get the "dpdk-shared" in VPP ?

PM, Damjan Marion via Lists.Fd.Io wrote: >> >> With this change, I'm able to compile VPP out of tarball produced by >> "make dist". >> >> https://gerrit.fd.io/r/#/c/17125/ <https://gerrit.fd.io/r/#/c/17125/> >> >> >>> On 28

Re: [vpp-dev] Getting core in vec_resize (vpp 18.01)

Please search this mailing list archive, Dave provided some hints some time ago 90M is not terribly high, but it can also be victim of something else holding memory. > On 29 Jan 2019, at 12:54, chetan bhasin wrote: > > Hi Damjan, > > Thanks for the reply. > > what should be a typical

Re: [vpp-dev] Getting core in vec_resize (vpp 18.01)

typically this happens when you run out of memory / main heap size or you have corrupted vector pointer.. It will be easier to read your traceback if it is captured with debug image, but according to frame 11, your vector is already 90MB big. Is this expected to be? > On 29 Jan 2019, at

Re: [vpp-dev] How do I get the "dpdk-shared" in VPP ?

Thanks, > Marco > > On 1/28/19 5:57 PM, Damjan Marion via Lists.Fd.Io wrote: >> >> With this change, I'm able to compile VPP out of tarball produced by >> "make dist". >> >> https://gerrit.fd.io/r/#/c/17125/ >> >> >>> On 28 Jan 2

Re: [vpp-dev] How do I get the "dpdk-shared" in VPP ?

With this change, I'm able to compile VPP out of tarball produced by "make dist". https://gerrit.fd.io/r/#/c/17125/ <https://gerrit.fd.io/r/#/c/17125/> > On 28 Jan 2019, at 13:35, Damjan Marion via Lists.Fd.Io > wrote: > > > >> On 28 Jan 2019, at

Re: [vpp-dev] How do I get the "dpdk-shared" in VPP ?

ot;make dist" (required indeed for > downstream consumption). that should be easy fixable > > On 1/26/19 2:22 PM, Damjan Marion via Lists.Fd.Io wrote: >> >> Here it is: https://gerrit.fd.io/r/17094 >> >> >> $ mkdir build-vpp stage >> >>

[vpp-dev] RFC: buffer manager rework

I am very close to the finish line with buffer management rework patch, and would like to ask people to take a look before it is merged. https://gerrit.fd.io/r/16638 It significantly improves performance of buffer alloc free and introduces numa awareness. On my skylake platinum 8180 system,

Re: [vpp-dev] Question about vlib_next_frame_change_ownership

Just to avoid potential confusion, my recent ethernet-input change mandates that frames are not aggregated when they are coming from input nodes. Reason for that is much faster processing of the frame when we know in advance that all packets in frame share same sw_if_index. In other words,

Re: [vpp-dev] How do I get the "dpdk-shared" in VPP ?

> On 25 Jan 2019, at 10:49, Marco Varlese wrote: > > Hi Damjan, > > On 1/24/19 10:46 PM, Damjan Marion via Lists.Fd.Io wrote: >> >> In theory like any other cmake project: >> >> $ mkdir build >> $ cd build >> $ cmake /path/to/vpp/src

Re: [vpp-dev] CPU usage on vpp_main thread is high when using tap cli between vpp and another userspace application #vpp

Don't use tapcli code, it is outdated and I just submitted patch to deprecate it (as agreed few months ago on the community call). https://gerrit.fd.io/r/17073 tapv2 (create interface tap) should be able to run on worker, but don't expect magic, it's jus tap

Re: [vpp-dev] How do I get the "dpdk-shared" in VPP ?

In theory like any other cmake project: $ mkdir build $ cd build $ cmake /path/to/vpp/src $ make $ make install In practice probably few lines should be modified in src/plugins/dodkCMakeLists.txt to enable linking with shared libs, as today we do static... — Damjan > On Jan 24, 2019, at

Re: [vpp-dev] l2 input/output feature arcs

> On 24 Jan 2019, at 11:37, Benoit Ganne (bganne) wrote: > >>> 1) add a new feature arc flavor 'all' alongside nonip, ip4 and ip6. >>> Nodes on this arc will gets all packets regardless of the headers. It keeps >>> backward compat but should add a small performance hit when L2 feature >>> arcs

Re: [vpp-dev] l2 input/output feature arcs

> On 24 Jan 2019, at 10:39, Benoit Ganne (bganne) via Lists.Fd.Io > wrote: > > Hi all, > > While refactoring the GBP plugin to use feature arcs instead of hardcoded L2 > feature bits, I had to basically duplicate my feature arc nodes 3x (and > disable/enable them 3x etc.) because the L2

Re: [vpp-dev] vpp-verify-master-clang now bionic based

On 22 Jan 2019, at 23:47, Ed Kern (ejk) mailto:e...@cisco.com>> wrote: As discussed in the vpp call this morning I have moved this from from running on xenial to bionic. Ran cleanly on the sandbox…but please let me know if anyone sees issues in production. Thanks! -- Damjan

[vpp-dev] debian/ubuntu packaging rework

Dear all, I just submitted debian/ubuntu packaging rework patch[1], and would like to hear feedback from people: Instead of hardcoded shell scripts it uses CMake components to put file into right package, so for example: install(FILES foo.txt DESTINATION share/vpp COMPONENT vpp-dev) means

Re: [vpp-dev] DPDK and SR-IOV #vpp

> On 16 Jan 2019, at 15:07, amir...@rad.com wrote: > > Hello, > > I would like to enable SR-IOV on interfaces that assigned to VPP. > I was able to create VFs using igb_uio driver (using "echo 8 > > /sys/bus/pci/devices/\:07\:00.1/max_vfs " command) > > The problem is that some of the

Re: [vpp-dev] Using prefetch and buffer_enqueue

> On 10 Jan 2019, at 07:15, Raj wrote: > > On Wed, Jan 9, 2019 at 4:18 PM Damjan Marion wrote: >>> On 9 Jan 2019, at 09:53, Raj wrote: >>> 1. static_always_inline void >>> vlib_buffer_enqueue_to_next (vlib_main_t * vm, vlib_node_runtime_t * node, >>> u32 *

Re: [vpp-dev] Question about #16465 patch

> On 9 Jan 2019, at 16:38, Alexander Chernavin wrote: > > Hello, > I have a question about this path: https://gerrit.fd.io/r/#/c/16465/ > > > There was a comment about used terminology ("dev instance" vs "ID") that I > addressed but no further actions

Re: [vpp-dev] Using prefetch and buffer_enqueue

> On 9 Jan 2019, at 09:53, Raj wrote: > > Hello all, > > While going through the VPP code, I have not fully understood what > the following functions mean and when should they be used? > > 1. static_always_inline void > vlib_buffer_enqueue_to_next (vlib_main_t * vm, vlib_node_runtime_t

Re: [vpp-dev] Performance Improvement

> On 9 Jan 2019, at 10:24, Gunaseelan Venkatachary - ERS, HCL Tech > wrote: > > Hello Team > Wish to reach out to bigger audience for some support in terms of performance > fine tuning, the context being , We are trying to insert a sample node, in > addition to whats already available in

Re: [vpp-dev] wwan0 intarface - vpp

> On 7 Jan 2019, at 16:03, amitmulay...@gmail.com wrote: > > Hello all , im new to vpp and just started to reseach about it > > i was wandering regarding an issue , i want that the vpp will recognaize in > his port list > a wwan0 intarface (wifi) that i have in the linux device. > it has its

Re: [vpp-dev] ethernet-input on master branch

> On 7 Jan 2019, at 12:38, Kingwel Xie wrote: > > thanks, Damjan, that is very clear. I checked the device-input and > ethernet-input code, and yes I totally understand your point. > > Two more questions, I can see you spend some time on avf plugin, do you think > it will eventually become

Re: [vpp-dev] ethernet-input on master branch

> On 5 Jan 2019, at 04:55, Kingwel Xie wrote: > > Hi Damjan, > > I noticed you removed the quick path from dpdk-input to ip-input/mpls-input, > after you merged the patch of ethernet-input optimization. Therefore, all > packets now have to go through ethernet-input. It would take a few

Re: [vpp-dev] Regarding vpp core health monitoring

> On Jan 4, 2019, at 10:45 AM, Kathiresan Jayabalan wrote: > > Hi Team, > > How can an external process monitor the health of vpp cores? Basically we > want to periodically check the health of vpp cores and eventually use it for > high availability. Please define “health” Thanks, —

Re: [vpp-dev] VPP/DPDK performance with Madvise (Transparent Huge pages)

Sorry, cannot help much wit details you provided... — Damjan > On Jan 4, 2019, at 9:33 AM, chetan bhasin wrote: > > Hi Damjan, > > I want to ensure that VPP uses THP and other Linux applications don't use THP > feature. > > Thanks, > Chetan Bhasin > >> On Thu, Jan 3, 2019 at 4:46 PM

Re: [vpp-dev] Question about duplicate nodes in the graph

> On 28 Dec 2018, at 16:21, Jon Loeliger wrote: > > Folks, > > Is there a valid use for duplicate nodes in the node graph? > Specifically, if I type this command three times: > > vpp# set interface ip vxlan-bypass TenGigabitEthernet6/0/0 > vpp# set interface ip vxlan-bypass

Re: [vpp-dev] VPP/DPDK performance with Madvise (Transparent Huge pages)

> On 2 Jan 2019, at 16:28, chetan bhasin wrote: > > Hi, > > We are using VPP 18.01. We have seen that in case Transparent Hugepage is > enable with Madvise . VPP does not take benefit of Annonymous Huge Pages. > Does anybody have any idea about the same ? Can you provide more details, What

Re: [vpp-dev] found some issue in pci vfio

Dear Ping, I already agreed that code needs to be fixed, but the point is that looking for container fd is not right way to detect if vfio-pci can be used or not... -- Damjan > On 2 Jan 2019, at 02:49, Yu, Ping wrote: > > Hi, Damjan, > > The issue existing in the code is container fd is

Re: [vpp-dev] VPP coding techniques

> On 1 Jan 2019, at 10:38, Raj wrote: > > Hello list and a very happy new year to all. > > I am starting to learn VPP, initially by configuring for common > scenarios and then trying to understand the code and may be eventually > into hacking VPP code. > > I know that VPP uses quite a bit of

Re: [vpp-dev] found some issue in pci vfio

Fact that you can open vfio container doesn't actually mean that you can use vfio-pci module on specific PCI device. vfio-pci module can be used in 2 cases: - when /sys/bus/pci/devices//iommu_group exists - when /sys/module/vfio/parameters/enable_unsafe_noiommu_mode is set to Y (introduced

Re: [vpp-dev] NAT handoff mechanism

> On 29 Dec 2018, at 07:26, david.leitch@gmail.com wrote: > > Hi ... > > I know that we need handoff mechanism when running multithread because > traffic for specific inside network user must be processed always on same > thread in both directions and We can not remove handoff node from

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

> On 22 Dec 2018, at 09:00, david.leitch@gmail.com wrote: > > Hi > What is difference between "plugins/flowprobe" and "vnet/flow"? > -=-=-=-=-=-=-=-=-=-=-=- vnet/flow is my attempt to create flow offload infra which supports different device driver types (i.e. dpdk rte_flow) and allows

Re: [vpp-dev] regarding cross compiling

> On 21 Dec 2018, at 11:21, saint_...@aliyun.com > wrote: > > Hi, > I modified several positions, then the make can be done successfully with > some minor problems, the details list below: Please submit to gerrit for review... -=-=-=-=-=-=-=-=-=-=-=- Links:

Re: [vpp-dev] dpdk: switch to in-memory mode, deprecate use of socket-mem

en support for socket-mem goes away, what will need to be done to ensure > that enough memory is available for crypto devices? DPDK is supposed to dynamically alloc more memory if needed, Unless I missed something... > > Thanks, > -Matt > > > > On Thu, Dec 20, 2018 at

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

> On 20 Dec 2018, at 05:19, mik...@yeah.net wrote: > >Thanks for you advice. That helps a lot. The result of DPDK testpmd is > almost the same. It seems something wrong with DPDK. > Or the card. Have you tried to play with RSS options? -- Damjan -=-=-=-=-=-=-=-=-=-=-=- Links: You

Re: [vpp-dev] build errors on CenOS 7.5

> On 20 Dec 2018, at 11:04, Zhiyong Yang wrote: > > Hi VPP guys, > > Could you notice this building errors on CentOs as below? > Could anybody help fix it? > > Regards > Zhiyong > Does https://gerrit.fd.io/r/#/c/16562/ helps?

[vpp-dev] dpdk: switch to in-memory mode, deprecate use of socket-mem

Regarding: https://gerrit.fd.io/r/#/c/16543/ This patch switches dpdk to new in-memory mode, and reduces dpdk memory footprint as pages are allocated dynamically on-demand. I tested on both Ubuntu and Centos 7.5 and everything looks good but will appreciate feedback from people using

<    1   2   3   4   5   6   7   >