[vpp-dev] API trace memory

[Chandra Mohan, Vijay Mohan via lists.fd.io]

Hello,

I am on 21.01 version and looking to understand the memory allocation around 
API tracing. I’ve got both Rx and Tx API trace enabled. The heap memory is 
continuously going up which, I guess, is expected. “nitems” is configured to 
1024 which is the default.  I do see that the trace wraps around but the 
heap-memory still keeps growing and eventually process crashes as the memory 
runs out. Was expecting the memory to be reused after the trace limit of 1024 
items has reached.

root@cnfp:/# cnfp-dbg vpp show api trace-status
RX Trace enabled
RX trace: used 1024 of 1024 items, is enabled, has wrapped
TX Trace enabled
TX trace: used 1024 of 1024 items, is enabled, has wrapped


Seeing this behavior as part of calling vl_msg_api_trace() nested in the below 
macro:

  REPLY_MACRO3(VL_API_CLI_INBAND_REPLY, vec_len (out_vec),
  ({
vl_api_vec_to_api_string(out_vec, >reply);
  }));

Thanks,
Vijay

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#21243): https://lists.fd.io/g/vpp-dev/message/21243
Mute This Topic: https://lists.fd.io/mt/90425185/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] How to check how many packets per vector is VPP processing ? #vpp

[Pim van Pelt]

Hoi David,

*show runtime* shows vectors/sec per worker thread and packets/vector per
node and clock cycles per packet.
See https://fd.io/docs/vpp/v2101/troubleshooting/cpuusage.html for more
details

groet,
Pim

On Tue, Apr 12, 2022 at 6:29 PM  wrote:

> Hi everyone,
>
> Is there any command to check how many packets per vector is VPP
> processing ? Thanks in advance.
>
> David.
>
>
>
>
>
>
>
>
>
>
>
>
>
> 
>
>

-- 
Pim van Pelt 
PBVP1-RIPE - http://www.ipng.nl/

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#21242): https://lists.fd.io/g/vpp-dev/message/21242
Mute This Topic: https://lists.fd.io/mt/90421955/21656
Mute #vpp:https://lists.fd.io/g/vpp-dev/mutehashtag/vpp
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] How to check how many packets per vector is VPP processing ? #vpp

[david . fernandes]

Hi everyone,

Is there any command to check how many packets per vector is VPP processing ? 
Thanks in advance.

David.

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#21241): https://lists.fd.io/g/vpp-dev/message/21241
Mute This Topic: https://lists.fd.io/mt/90421955/21656
Mute #vpp:https://lists.fd.io/g/vpp-dev/mutehashtag/vpp
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] How does VPP proceeds to fill Vectors with packets ? #vpp

[david . fernandes]

Hi everyone,

I would like to understand how VPP proceeds to fill Vectors with packets. Does 
it follow any criteria to decide which packets it will group in a same vector 
in order to process them together ? Is there any kind of classification ?
I can't find this information in the documentation. Thanks in advance for your 
help.

David

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#21240): https://lists.fd.io/g/vpp-dev/message/21240
Mute This Topic: https://lists.fd.io/mt/90421366/21656
Mute #vpp:https://lists.fd.io/g/vpp-dev/mutehashtag/vpp
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] CSIT topics for VPP meeting today

[Maciek Konstantynowicz (mkonstan) via lists.fd.io]

Hi,

We would like to spend 5-10min doing the CSIT-2202 vpp rls notes review, 
specifically recorder failures and regressions.
Vratko and me will lead that.

If there is no time on the call agenda today, no sweat, we can do it on the 
next call.

Cheers,
Maciek
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#21239): https://lists.fd.io/g/vpp-dev/message/21239
Mute This Topic: https://lists.fd.io/mt/90418453/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] Queries regarding mfib and multicast support in VPP

[arunmittal88]

Hi,

Just want to know if the mfib-lookup node is attached to the default packet 
processing node graph or do we need to explicitly enable this.

Also wanted to understand if the following things are achievable with mfib in 
VPP (To understand the mfib in VPP I am following 
"https://wiki.fd.io/view/VPP/MFIB; page).

Lets assume I have 3 interfaces Eth0/1 Eth0/2 and Eth0/3 and I want to 
configure the multicast rule for Group G (236.1.2.3) as follow:
1.) The group G stream is acceptable on interface Eth0/1 and Eth0/2 and it is 
blocked on interface Eth0/3.
2.) If group G stream is received on Eth0/1, it is forwarded to Eth0/2 and if 
Eth0/2 receives it then, it should be forwarded to Eth0/1 as well.

Thanks,
Arun

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#21238): https://lists.fd.io/g/vpp-dev/message/21238
Mute This Topic: https://lists.fd.io/mt/90417743/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] VPP SRv6 steering packets into a SR policy based on classifier of SRH tag field

[Igor Mjasojedov]

Hello,

Is it possible to steer packets based on ‘tag’ field of IPv6 SR extension 
header? I am using VPP v18.10.
I found that there are some options to do so, but there is not much explanation 
of the arguments of the commands ‘classify table’, ‘classify session’.
https://docs.fd.io/vpp/18.04/srv6_steering_doc.html


I would like to somehow define hex mask in order to match tag field and gather 
the value from it. Based on the value, I would specify which SRv6 policy should 
be used for that tagged traffic.

Can you please give me some advice?

Thank you.
Best regards,
Igor Mjasojedov
mjasojedov.igo...@gmail.com
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#21237): https://lists.fd.io/g/vpp-dev/message/21237
Mute This Topic: https://lists.fd.io/mt/90415835/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] Issues with failsafe dpdk pmd in Azure

[Peter Morrow]

Thanks Long,

We are good now that the guidance is to stick with the netvsc pmd. Replying 
here so others might see this message on the vpp mailing list.

Peter.

From: Long Li 
Sent: 11 April 2022 22:39
To: Peter Morrow ; Stephen Hemminger 

Cc: vpp-dev@lists.fd.io 
Subject: RE: Issues with failsafe dpdk pmd in Azure


Hi Peter,



The failure you see on Failsafe/MLX5:

> 2022/04/09 09:11:16:424 notice dpdk   net_failsafe: Link update 
> failed for sub_device 0 with error 1
> 2022/04/09 09:12:32:380 notice dpdk   common_mlx5: Unable to find 
> virtually contiguous chunk for address (0x10). 
> rte_memseg_contig_walk() failed.



This is a known issue with MLX5 registering a buffer not allocated through DPDK 
API. This can happen if other driver or other code pass a buffer to MLX5 but 
MLX5 can’t register it with the hardware.

The error detail is at:

https://github.com/DPDK/dpdk/blob/7cf3d07c3adcb015c303e4cdf2ef9712a65ce46d/drivers/common/mlx5/mlx5_common_mr.c#L626



If this is troubling on your setup, can you set “mr_ext_memseg_en=0” for mlx5? 
This should get rid of this error.



Long



From: Peter Morrow 
Sent: Monday, April 11, 2022 1:49 AM
To: Stephen Hemminger 
Cc: vpp-dev@lists.fd.io; Long Li 
Subject: Re: Issues with failsafe dpdk pmd in Azure



You don't often get email from pe...@graphiant.com. 
Learn why this is important

Thanks Stephen,



Very happy to stick with netvsc in that case.



Peter.



From: Stephen Hemminger 
mailto:step...@networkplumber.org>>
Sent: 10 April 2022 18:59
To: Peter Morrow mailto:pe...@graphiant.com>>
Cc: vpp-dev@lists.fd.io 
mailto:vpp-dev@lists.fd.io>>; Long Li 
mailto:lon...@microsoft.com>>
Subject: Re: Issues with failsafe dpdk pmd in Azure



On Sat, 9 Apr 2022 09:20:43 +
Peter Morrow mailto:pe...@graphiant.com>> wrote:

> Hello,
>
> I currently have vpp working nicely in Azure via the netsvc pmd, however 
> after reading 
> https://docs.microsoft.com/en-us/azure/virtual-network/setup-dpdk
>  and 
> https://fd.io/docs/vpp/v2101/usecases/vppinazure.html
>  it sounds like I should be using the failsafe pmd instead. So, I gave this a 
> try but ran into some issues, some of which I've seen discussed on this email 
> group but none of the solutions have worked for me thus far. I was able to 
> make the failsafe pmd work via dpdk-testpmd with dpdk standalone from debian 
> bullseye (dpdk 20.11).

You have it backwards.  Failsafe is the older driver which was developed to be 
generic.
Faisafe is slower because it has to go through the kernel for the slow path.
Would like to deprecate use of failsafe but there are some use cases such as 
rte_flow which
netvsc PMD does not support. Supporting rte_flow in a software driver would 
require a significant
amount of work (same problem as virtio).


>
> I'm running vpp 22.06 and an external dpdk at version 21.11, though also see 
> the same thing when downgrading to 20.11. We have 3 interfaces, the first is 
> a non accelerated networking interface which is not to be controlled by vpp 
> (eth0) the 2nd are data interfaces which are vpp owned. My dpdk section of 
> the vpp startup config looks like this:
>
> dpdk {
>
> dev 0ed6:00:02.0
> vdev net_vdev_netvsc0,iface=eth1
>
> dev 6fa1:00:02.0
> vdev net_vdev_netvsc1,iface=eth2
>
> base-virtaddr 0x2
> }
>
> When vpp starts up the 2 interfaces are shown:
>
> me@azure:~$ sudo vppctl sh hard
>   NameIdx   Link  Hardware
> FailsafeEthernet2