[vpp-dev] VPP CSIT robot framework #automation

[Devis Reagan]

Dear Team ,

I could see with the current VPP CSIT framework we have traffic generator 
support using trex . Do we have IXIA support API's ?
If any private git branch for IXIA support , could you pls point to me ?

https://docs.fd.io/csit/master/report/csit_framework_documentation/csit_design.html
https://wiki.fd.io/view/Project_Proposals/CSIT

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

View/Reply Online (#13314): https://lists.fd.io/g/vpp-dev/message/13314
Mute This Topic: https://lists.fd.io/mt/32104144/21656
Mute #automation: https://lists.fd.io/mk?hashtag=automation&subid=1480452
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 tap interface issue on Arm servers

[Lijian Zhang]

Ping vpp tap interface via host tap interface also failed.
I checked vpp counters with ‘show runtime’, but it seems vpp doesn’t receive 
any packets.
‘u16 n_left = vring->used->idx – last;’ in function 
virtio_device_input_inline() is always 0.
Could you please give suggestions on root-causing the issue?

DBGvpp# show runtime
Thread 0 vpp_main (lcore 2)
Time 1.3, average vectors/node 0.00, last 128 main loops 0.00 per node 0.00
  vector rates in 0.e0, out 0.e0, drop 0.e0, punt 0.e0
 Name State Calls  Vectors
Suspends Clocks   Vectors/Call
dpdk-processany wait 0   0  
 1  5.10e10.00
ikev2-manager-process   any wait 0   0  
 1  5.20e10.00
ip6-icmp-neighbor-discovery-ev  any wait 0   0  
 1  4.70e10.00
unix-cli-local:0 active  0   0  
 9  2.36e30.00
unix-epoll-input polling382236   0  
 0  3.18e10.00
---
Thread 1 vpp_wk_0 (lcore 6)
Time 1.3, average vectors/node 0.00, last 128 main loops 0.00 per node 0.00
  vector rates in 0.e0, out 0.e0, drop 0.e0, punt 0.e0
 Name State Calls  Vectors
Suspends Clocks   Vectors/Call
unix-epoll-input polling   701   0  
 0  1.27e10.00
virtio-input polling718881   0  
 0  2.96e10.00

Thanks.

From: Lijian Zhang (Arm Technology China)
Sent: 2019年6月18日 13:35
To: vpp-dev@lists.fd.io
Cc: nd 
Subject: VPP tap interface issue on Arm servers

Hi,
We tried to create a VPP tap interface, assigned ip address to both VPP and 
Linux side, and then ping VPP tap interface via host interface, and vice versa.
But ping failed on both side.

With below ping via VPP tap interface for example, it seems VPP has sent out 
packets, but there’s no any counter increasing on the interface of host side.

Could you please suggest on this issue?

DBGvpp# create tap
DBGvpp# set int ip address tap0 16.0.10.2/24
DBGvpp# set int state tap0 up

$ sudo ip a add 16.0.10.1/24 dev tap0

DBGvpp# clear interfaces
DBGvpp# clear errors
DBGvpp# clear runtime
DBGvpp# ping 16.0.10.1

Statistics: 5 sent, 0 received, 100% packet loss

DBGvpp# show runtime
Thread 0 vpp_main (lcore 2)
Time 88.2, average vectors/node 1.00, last 128 main loops 0.00 per node 0.00
  vector rates in 0.e0, out 5.6688e-2, drop 5.6688e-2, punt 0.e0
 Name State Calls  Vectors
Suspends Clocks   Vectors/Call
api-rx-from-ringany wait 0   0  
 4  8.05e10.00
dhcp-client-process any wait 0   0  
 1  8.00e10.00
dpdk-processany wait 0   0  
30  3.45e10.00
drop active  5   5  
 0  6.30e11.00
error-drop   active  5   5  
 0  7.24e11.00
fib-walkany wait 0   0  
44  4.35e10.00
ikev2-manager-process   any wait 0   0  
88  2.78e10.00
ip-neighbor-scan-processany wait 0   0  
 2  6.50e10.00
ip-route-resolver-process   any wait 0   0  
 1  8.20e10.00
ip4-gleanactive  5   5  
 0  9.50e21.00
ip4-lookup   active  5   5  
 0  1.39e21.00
ip4-reassembly-expire-walk  any wait 0   0  
 9  4.09e10.00
ip6-icmp-neighbor-discovery-ev  any wait 0   0  
88  2.46e10.00
ip6-reassembly-expire-walk  any wait 0   0  
 9  4.06e10.00
statseg-collector-process   time wait0   0  
 9  6.33e10.00
tap0-output  active  5   5  
 0  8.16e11.00
tap0-tx  active  5

[vpp-dev] VPP tap interface issue on Arm servers

[Lijian Zhang]

Hi,
We tried to create a VPP tap interface, assigned ip address to both VPP and 
Linux side, and then ping VPP tap interface via host interface, and vice versa.
But ping failed on both side.

With below ping via VPP tap interface for example, it seems VPP has sent out 
packets, but there’s no any counter increasing on the interface of host side.

Could you please suggest on this issue?

DBGvpp# create tap
DBGvpp# set int ip address tap0 16.0.10.2/24
DBGvpp# set int state tap0 up

$ sudo ip a add 16.0.10.1/24 dev tap0

DBGvpp# clear interfaces
DBGvpp# clear errors
DBGvpp# clear runtime
DBGvpp# ping 16.0.10.1

Statistics: 5 sent, 0 received, 100% packet loss

DBGvpp# show runtime
Thread 0 vpp_main (lcore 2)
Time 88.2, average vectors/node 1.00, last 128 main loops 0.00 per node 0.00
  vector rates in 0.e0, out 5.6688e-2, drop 5.6688e-2, punt 0.e0
 Name State Calls  Vectors
Suspends Clocks   Vectors/Call
api-rx-from-ringany wait 0   0  
 4  8.05e10.00
dhcp-client-process any wait 0   0  
 1  8.00e10.00
dpdk-processany wait 0   0  
30  3.45e10.00
drop active  5   5  
 0  6.30e11.00
error-drop   active  5   5  
 0  7.24e11.00
fib-walkany wait 0   0  
44  4.35e10.00
ikev2-manager-process   any wait 0   0  
88  2.78e10.00
ip-neighbor-scan-processany wait 0   0  
 2  6.50e10.00
ip-route-resolver-process   any wait 0   0  
 1  8.20e10.00
ip4-gleanactive  5   5  
 0  9.50e21.00
ip4-lookup   active  5   5  
 0  1.39e21.00
ip4-reassembly-expire-walk  any wait 0   0  
 9  4.09e10.00
ip6-icmp-neighbor-discovery-ev  any wait 0   0  
88  2.46e10.00
ip6-reassembly-expire-walk  any wait 0   0  
 9  4.06e10.00
statseg-collector-process   time wait0   0  
 9  6.33e10.00
tap0-output  active  5   5  
 0  8.16e11.00
tap0-tx  active  5   5  
 0  7.20e11.00
unix-cli-stdin   active  0   0  
56  6.18e20.00
unix-epoll-input polling  25126436   0  
 0  3.14e10.00
---
Thread 1 vpp_wk_0 (lcore 6)
Time 88.2, average vectors/node 0.00, last 128 main loops 0.00 per node 0.00
  vector rates in 0.e0, out 0.e0, drop 0.e0, punt 0.e0
 Name State Calls  Vectors
Suspends Clocks   Vectors/Call
unix-epoll-input polling 45817   0  
 0  1.47e10.00
virtio-input polling  46962827   0  
 0  2.96e10.00
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#13312): https://lists.fd.io/g/vpp-dev/message/13312
Mute This Topic: https://lists.fd.io/mt/32103947/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] Commit Subjects

[Jon Loeliger via Lists.Fd.Io]

On Mon, Jun 17, 2019 at 3:43 PM Damjan Marion  wrote:

>
> On 17 Jun 2019, at 22:41, Jon Loeliger  wrote:
>
> On Mon, Jun 17, 2019 at 3:33 PM Damjan Marion  wrote:
>
>>
>> I tried to be as much descriptive as possible in those messages, but
>> looks like I failed :(
>>
>
> Damjan,
>
> No problem.  I am working on some improvements.
>
> For example, fixing these problems:
>
> $ echo "Consolidate IPv6 link-local-address functions." |  sed -e
> 's/\(.*\):.*/\1/'
> Consolidate IPv6 link-local-address functions.
>
> $ echo "foo: bar: baz: Consolidate IPv6 link-local-address functions." |
>  sed -e 's/\(.*\):.*/\1/'
> foo: bar: baz
>
>
> Your commit message needs to be:
>
> foo bar baz: consolidate...
>

Damjan,

I think there will be unexpected complications with this approach.
Consider this:

$ echo "docs: fix http:/foo.example.com:8080/fred use." |  sed -e
's/\(.*\):.*/\1/'
docs: fix http:/foo.example.com

That won't match "docs:", nor any other valid I: word in any case where
there is an
additional colon later in the subject line.  It will also allow false
positives like this
subject:  "whanged on ip stuff: it was messed up".

Historically, these things have been keywords individually terminated by a
colon, as
I gave in my example.  Once the collection of keywords is gathered, strip
the trailing
colons for validity comparisons.

Just a thought.

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

View/Reply Online (#13311): https://lists.fd.io/g/vpp-dev/message/13311
Mute This Topic: https://lists.fd.io/mt/32096793/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] Commit Subjects

[Damjan Marion via Lists.Fd.Io]

> On 17 Jun 2019, at 22:41, Jon Loeliger  wrote:
> 
> On Mon, Jun 17, 2019 at 3:33 PM Damjan Marion  > wrote:
> 
> I tried to be as much descriptive as possible in those messages, but looks 
> like I failed :(
> 
> Damjan,
> 
> No problem.  I am working on some improvements.
> 
> For example, fixing these problems:
> 
> $ echo "Consolidate IPv6 link-local-address functions." |  sed -e 
> 's/\(.*\):.*/\1/'
> Consolidate IPv6 link-local-address functions.
>  
> $ echo "foo: bar: baz: Consolidate IPv6 link-local-address functions." |  sed 
> -e 's/\(.*\):.*/\1/'
> foo: bar: baz

Your commit message needs to be:

foo bar baz: consolidate...

with lowercase 'c'


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

View/Reply Online (#13310): https://lists.fd.io/g/vpp-dev/message/13310
Mute This Topic: https://lists.fd.io/mt/32096793/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] Commit Subjects

[Jon Loeliger via Lists.Fd.Io]

On Mon, Jun 17, 2019 at 3:33 PM Damjan Marion  wrote:

>
> I tried to be as much descriptive as possible in those messages, but looks
> like I failed :(
>

Damjan,

No problem.  I am working on some improvements.

For example, fixing these problems:

$ echo "Consolidate IPv6 link-local-address functions." |  sed -e
's/\(.*\):.*/\1/'
Consolidate IPv6 link-local-address functions.

$ echo "foo: bar: baz: Consolidate IPv6 link-local-address functions." |
 sed -e 's/\(.*\):.*/\1/'
foo: bar: baz

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

View/Reply Online (#13309): https://lists.fd.io/g/vpp-dev/message/13309
Mute This Topic: https://lists.fd.io/mt/32096793/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] Commit Subjects

[Damjan Marion via Lists.Fd.Io]

I tried to be as much descriptive as possible in those messages, but looks like 
I failed :(

> On 17 Jun 2019, at 22:20, Florin Coras  wrote:
> 
> Hi Jon, 
> 
> Yup, it has to match a maintainer “I” entry and yes, that’s where the script 
> is. :-)
> 
> Florin
> 
>> On Jun 17, 2019, at 1:16 PM, Jon Loeliger via Lists.Fd.Io 
>> mailto:jdl=netgate@lists.fd.io>> wrote:
>> 
>> And now this: 
>> 
>> === ERROR ===
>> Unknown feature 'Consolidate' in commit 'Subject:' line.
>> Feature must exist in MAINTAINERS file. If this commit intruduces 
>> new feature, then this commit must add new entry into the 
>> MAINTAINERS file.
>> === ERROR ===
>> Does it just check the first word on the subject line regardless?
>> What else is being checked?  Poorly?
>> Does it _have_ to match a MAINTAINER I: word?
>> 
>> Where is this script?  Rhetorical.  It is here:
>> extras/scripts/check_commit_msg.sh
>> 
>> Now I at least know what I am up against.
>> 
>> jdl
>> 
>> -=-=-=-=-=-=-=-=-=-=-=-
>> Links: You receive all messages sent to this group.
>> 
>> View/Reply Online (#13306): https://lists.fd.io/g/vpp-dev/message/13306 
>> 
>> Mute This Topic: https://lists.fd.io/mt/32096793/675152 
>> 
>> Group Owner: vpp-dev+ow...@lists.fd.io 
>> Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub 
>>   [fcoras.li...@gmail.com 
>> ]
>> -=-=-=-=-=-=-=-=-=-=-=-
> 
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> 
> View/Reply Online (#13307): https://lists.fd.io/g/vpp-dev/message/13307
> Mute This Topic: https://lists.fd.io/mt/32096793/675642
> Group Owner: vpp-dev+ow...@lists.fd.io
> Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [dmar...@me.com]
> -=-=-=-=-=-=-=-=-=-=-=-

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

View/Reply Online (#13308): https://lists.fd.io/g/vpp-dev/message/13308
Mute This Topic: https://lists.fd.io/mt/32096793/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] Commit Subjects

[Florin Coras]

Hi Jon, 

Yup, it has to match a maintainer “I” entry and yes, that’s where the script 
is. :-)

Florin

> On Jun 17, 2019, at 1:16 PM, Jon Loeliger via Lists.Fd.Io 
>  wrote:
> 
> And now this: 
> 
> === ERROR ===
> Unknown feature 'Consolidate' in commit 'Subject:' line.
> Feature must exist in MAINTAINERS file. If this commit intruduces 
> new feature, then this commit must add new entry into the 
> MAINTAINERS file.
> === ERROR ===
> Does it just check the first word on the subject line regardless?
> What else is being checked?  Poorly?
> Does it _have_ to match a MAINTAINER I: word?
> 
> Where is this script?  Rhetorical.  It is here:
> extras/scripts/check_commit_msg.sh
> 
> Now I at least know what I am up against.
> 
> jdl
> 
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> 
> View/Reply Online (#13306): https://lists.fd.io/g/vpp-dev/message/13306
> Mute This Topic: https://lists.fd.io/mt/32096793/675152
> Group Owner: vpp-dev+ow...@lists.fd.io
> Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [fcoras.li...@gmail.com]
> -=-=-=-=-=-=-=-=-=-=-=-

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

View/Reply Online (#13307): https://lists.fd.io/g/vpp-dev/message/13307
Mute This Topic: https://lists.fd.io/mt/32096793/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] Commit Subjects

[Jon Loeliger via Lists.Fd.Io]

And now this:

=== ERROR ===

Unknown feature 'Consolidate' in commit 'Subject:' line.
Feature must exist in MAINTAINERS file. If this commit intruduces
new feature, then this commit must add new entry into the
MAINTAINERS file.
=== ERROR ===

Does it just check the first word on the subject line regardless?
What else is being checked?  Poorly?
Does it _have_ to match a MAINTAINER I: word?

Where is this script?  Rhetorical.  It is here:
extras/scripts/check_commit_msg.sh

Now I at least know what I am up against.

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

View/Reply Online (#13306): https://lists.fd.io/g/vpp-dev/message/13306
Mute This Topic: https://lists.fd.io/mt/32096793/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] Emergency Maintenance - FD.io Jenkins Slowness

[Vanessa Valderrama]

Jenkins is up. We're still determining the root cause and will monitor
closely.

Thank you,

Vanessa


On 06/17/2019 01:58 PM, Vanessa Valderrama wrote:
> We are placing Jenkins in shutdown mode to troubleshoot the slowness we
> are experiencing.
>
> We apologize for the inconvenience.
>
> Thank you,
>
> Vanessa
>
>
> On 06/17/2019 01:32 PM, Vanessa Valderrama wrote:
>> We're currently experience slowness and intermittent HTTPError: 504
>> Server Error: Gateway Time-outs on build jobs.
>>
>> We are working on resolving this issue as quickly as possible.
>>
>> Thank you,
>> Vanessa

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

View/Reply Online (#13305): https://lists.fd.io/g/vpp-dev/message/13305
Mute This Topic: https://lists.fd.io/mt/32098808/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] Emergency Maintenance - FD.io Jenkins Slowness

[Vanessa Valderrama]

We are placing Jenkins in shutdown mode to troubleshoot the slowness we
are experiencing.

We apologize for the inconvenience.

Thank you,

Vanessa


On 06/17/2019 01:32 PM, Vanessa Valderrama wrote:
> We're currently experience slowness and intermittent HTTPError: 504
> Server Error: Gateway Time-outs on build jobs.
>
> We are working on resolving this issue as quickly as possible.
>
> Thank you,
> Vanessa

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

View/Reply Online (#13304): https://lists.fd.io/g/vpp-dev/message/13304
Mute This Topic: https://lists.fd.io/mt/32098808/21656
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 Jenkins Slowness

[Vanessa Valderrama]

We're currently experience slowness and intermittent HTTPError: 504
Server Error: Gateway Time-outs on build jobs.

We are working on resolving this issue as quickly as possible.

Thank you,
Vanessa
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#13303): https://lists.fd.io/g/vpp-dev/message/13303
Mute This Topic: https://lists.fd.io/mt/32098451/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] Triggering a Verify Job?

[Jon Loeliger via Lists.Fd.Io]

Folks,

I am failing to trigger a "verify job" using the Review keyword "recheck"
on one of my patches.

Can someone remind me how to clear up my dumbness?

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

View/Reply Online (#13302): https://lists.fd.io/g/vpp-dev/message/13302
Mute This Topic: https://lists.fd.io/mt/32098401/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

[vpp-dev] Commit Subjects

[Jon Loeliger via Lists.Fd.Io]

Folks,

I received this during a build of a patch I submitted:

=== ERROR ===
Unknown feature 'IPv6' in commit 'Subject:' line.
Feature must exist in MAINTAINERS file. If this commit intruduces
new feature, then this commit must add new entry into the
MAINTAINERS file.
=== ERROR ===

Can we modify the scripts/check-something to verify this before
it is submitted?

Can we fix the spelling error in the message?

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

View/Reply Online (#13301): https://lists.fd.io/g/vpp-dev/message/13301
Mute This Topic: https://lists.fd.io/mt/32096793/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] timer wheel usage queries

[Dave Barach via Lists.Fd.Io]

Add a vector of tw_timer_wheel_xxx_t *’s to e.g. my_main_t. Vec_validate (...) 
so you have one wheel per thread. Initialize all of the wheels.

Use e.g. vm->thread_index or os_get_cpu_index(..) to pick the wheel to process 
and/or use to start/stop timers.


From: vpp-dev@lists.fd.io  On Behalf Of hari_akkin via 
Lists.Fd.Io
Sent: Monday, June 17, 2019 11:37 AM
To: vpp-dev@lists.fd.io
Cc: vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] timer wheel usage queries

Hi Dave,
Thanks for the quick response.
Can you please point me how to tie a timer wheel to a thread?

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

View/Reply Online (#13300): https://lists.fd.io/g/vpp-dev/message/13300
Mute This Topic: https://lists.fd.io/mt/32039603/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] timer wheel usage queries

[hari_akkin via Lists.Fd.Io]

Hi Dave,
Thanks for the quick response.
Can you please point me how to tie a timer wheel to a thread?

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

View/Reply Online (#13299): https://lists.fd.io/g/vpp-dev/message/13299
Mute This Topic: https://lists.fd.io/mt/32039603/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] PerfMon Plugin Issue

[Dave Barach via Lists.Fd.Io]

Sounds like this should work. Please try to debug it and report results.

From: vpp-dev@lists.fd.io  On Behalf Of Satya Murthy
Sent: Monday, June 17, 2019 3:10 AM
To: vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] PerfMon Plugin Issue

Hi Dave,

Thanks for the reply.

I am still seeing the same issue.
Here is what I am doing.

1) Created 2 VPP instances

2) Created Memif interfaces on each of them and assigned IP addresses to them

3) Able to ping the IPs from one another successfully.

4) On VPP-1, started ping of interface on VPP-2 instance as below
DBGvpp# ping 1.1.1.2 repeat 10 interval 0.0001

5) While the ping is running on VPP-1 CLI, ran following commands on VPP-2 CLI

DBGvpp# set pmc cpu-cycles
Start collection for 1 events, wait 1.00 seconds
DEADMAN: collection still running...
Data collection complete...

6) Ran show pmc to check the data

DBGvpp# show pmc
Data collection in progress...

This is always showing it as "Data collection in progress".

I am using official version of fdio.1904 in this activity.

Am I missing anything obvious ?
--
Thanks & Regards,
Murthy
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#13298): https://lists.fd.io/g/vpp-dev/message/13298
Mute This Topic: https://lists.fd.io/mt/32064990/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] timer wheel usage queries

[Dave Barach via Lists.Fd.Io]

The only requirement is to call tw_timer_expire_timers_xxx (...) often enough 
so that timers expire accurately enough for correct operation. You can spin up 
an input node (as you describe), you can arrange to schedule a frame for an 
interior graph node, and so on.

I would suggest placing sessions, etc. onto specific threads; use one wheel per 
thread. Running one timer per thread is not suggested.

D.

From: vpp-dev@lists.fd.io  On Behalf Of hari_akkin via 
Lists.Fd.Io
Sent: Monday, June 17, 2019 6:54 AM
To: vpp-dev@lists.fd.io
Cc: vpp-dev@lists.fd.io
Subject: Re: [vpp-dev] timer wheel usage queries

Thanks Dave for the information on calling tw_timer_expire_timers_xxx(tw, now). 
However i have to make the node type as VLIB_NODE_TYPE_INPUT and set node state 
to VLIB_NODE_STATE_POLLING for polling to work. is this the right approach?

I have one more query on how to implement per thread timer wheels. Lets I say I 
have N worker threads and want to have one timer per thread, what are the best 
ways to achieve this using TW framework. In my prototype the expiry function 
always lands on main thread.

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

View/Reply Online (#13297): https://lists.fd.io/g/vpp-dev/message/13297
Mute This Topic: https://lists.fd.io/mt/32039603/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] timer wheel usage queries

[hari_akkin via Lists.Fd.Io]

Thanks Dave for the information on calling tw_timer_expire_timers_xxx(tw, now). 
However i have to make the node type as VLIB_NODE_TYPE_INPUT and set node state 
to VLIB_NODE_STATE_POLLING for polling to work. is this the right approach?

I have one more query on how to implement per thread timer wheels. Lets I say I 
have N worker threads and want to have one timer per thread, what are the best 
ways to achieve this using TW framework. In my prototype the expiry function 
always lands on main thread.

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

View/Reply Online (#13296): https://lists.fd.io/g/vpp-dev/message/13296
Mute This Topic: https://lists.fd.io/mt/32039603/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] identifying packets sent to a loopback address

[Neale Ranns via Lists.Fd.Io]

Hi Matt,

I have no explanation for why adj_index[VLIB_RX] is set to the same value as 
the TX, AFAICT it’s been that way since time immemorial 😉 I think we should 
change that logic to RX=TX; TX=lbi. I don’t see anything in the code that 
relies on RX being set to the TX.
In that case the RX will be the receive DPO which has the sw_if_index 
associated with the for-us FIB entry.

Is your plugin running as a for-us feature? That’s good if you want to see all 
the for-us packets. If you want per-prefix for-us you might consider a 
interposing your own DPO in that prefixes output chain. See the interposed 
gbp_policy_dpo_t for an example.

Regards,
neale



De : "Dave Barach (dbarach)" 
Date : vendredi 14 juin 2019 à 20:30
À : "mgsm...@netgate.com" , "Neale Ranns (nranns)" 

Objet : Re: [vpp-dev] identifying packets sent to a loopback address

This is a Neale Ranns question. Looping him in... HTH... Dave


From: vpp-dev@lists.fd.io  on behalf of Matthew Smith via 
Lists.Fd.Io 
Sent: Friday, June 14, 2019 1:52 PM
To: vpp-dev
Cc: vpp-dev@lists.fd.io
Subject: [vpp-dev] identifying packets sent to a loopback address

Hi,

I have a plugin node in which I want to process some received packets. When a 
packet is destined to an IP address that is configured on a loopback interface, 
I want to be able to lookup some data based on the sw_if_index of the loopback 
interface. Packets that are sent to the loopback address are received on 
hardware interfaces, so I cannot look at sw_if_index[VLIB_RX] in the vnet 
buffer opaque data to find out that the packet destination address belongs to a 
loopback. I was hoping that ip.adj_index[VLIB_TX] would contain data that would 
help me do what I'm trying to do, but that does not seem to be the case. In 
ip4-lookup, a FIB entry is found which matches the loopback address. The index 
of the adjacency for the path is stored in ip.adj_index[VLIB_TX]. But the 
stored value is subsequently overwritten in ip4-local during the call to 
ip4_local_check_src(). That function does a lookup on the source IP address and 
overwrites both ip.adj_index[VLIB_RX] and ip.adj_index[VLIB_TX] with the index 
retrieved by the lookup.

I can work around this in my plugin code if I need to but it would be nice to 
be able to use the adjacency data that was retrieved by ip4-lookup. Would it 
break anything if ip4_local_check_src() only updates ip.adj_index[VLIB_RX] and 
leaves the contents of ip.adj_index[VLIB_TX] intact?

Thanks,
-Matt

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

View/Reply Online (#13295): https://lists.fd.io/g/vpp-dev/message/13295
Mute This Topic: https://lists.fd.io/mt/32066650/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Re: [vpp-dev] PerfMon Plugin Issue

[Satya Murthy]

Hi Dave,

Thanks for the reply.

I am still seeing the same issue.
Here is what I am doing.

1) Created 2 VPP instances

2) Created Memif interfaces on each of them and assigned IP addresses to them

3) Able to ping the IPs from one another successfully.

4) On VPP-1, started ping of interface on VPP-2 instance as below
DBGvpp# ping 1.1.1.2 repeat 10 interval 0.0001

5) While the ping is running on VPP-1 CLI, ran following commands on VPP-2 CLI

DBGvpp# set pmc cpu-cycles
Start collection for 1 events, wait 1.00 seconds
DEADMAN: collection still running...
Data collection complete...

6) Ran show pmc to check the data

DBGvpp# show pmc
Data collection in progress...

This is always showing it as "Data collection in progress".

I am using official version of fdio.1904 in this activity.

Am I missing anything obvious ?
--
Thanks & Regards,
Murthy
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.

View/Reply Online (#13294): https://lists.fd.io/g/vpp-dev/message/13294
Mute This Topic: https://lists.fd.io/mt/32064990/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe: https://lists.fd.io/g/vpp-dev/unsub  [arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-