DPDK CI discussions
 help / color / Atom feed
From: David Marchand <david.marchand@redhat.com>
To: Jeremy Plsek <jplsek@iol.unh.edu>
Cc: ci@dpdk.org
Subject: Re: [dpdk-ci] CI Test Cases
Date: Tue, 19 Mar 2019 09:38:08 +0100
Message-ID: <CAJFAV8y8-LSh5vniZXR812ckKNa2ELEJVRKRzT53PVu2zO902w@mail.gmail.com> (raw)
In-Reply-To: <CA+xUZB67UM0b2kN=sSPAYBQMOuDisVWx8D7MSKjbP9FmUNE-tA@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 2218 bytes --]

Hello Jeremy,

On Mon, Mar 18, 2019 at 4:56 PM Jeremy Plsek <jplsek@iol.unh.edu> wrote:

> Based on last weeks meeting, I've created a document to help start
> documenting what test cases
> should be worked on for cross platform support between NIC vendors and
> which test cases can be used now in the Community Lab.
>
> This list of test cases that should be worked on can be used when
> discussing the budget with the board.
>
> This also gives the Community Lab a list of tests that can be
> implemented in the mean time.
>
>
> https://docs.google.com/document/d/1WbRoIq0f5SmHM6D5sLbeh7hxRJdE2ppwB5JECHllLjg/edit?usp=sharing
> (This document is publicly editable. If I notice the document changes
> outside the scope of the goal, I'll try to revert it and give
> individual access instead.)
>
> If creating this list over email is wanted instead, then we can do it
> in this thread instead of the document. (And I'll update the document
> as emails come in.)
>

Well, for now, I prefer emails because I did not follow the CI discussions
so far and I might be off topic.

Here I go, the list is large, I tried to write from the basic to the most
advanced/less used.

- Non regression tests on basic startup (we can reuse the dpdk ut and/or
enhance them), this part would be hw agnostic:
  - core checks: playing with cpus online/offline + dpdk coremask/list +
checking cpu affinity of all dpdk threads
  - memory checks: a big topic because we have multiple modes at the moment
  - startup devices blacklist/whitelist
  - hotplug (unplug) of devices

- Non regression tests on eth drivers (testpmd seems the best candidate),
this must be ran against a lot of nics/drivers:
  - reception of unicast packets: check that we receive packets for primary
and secondary addresses (and only for them, meaning with checks that
undesired packets are dropped following updates)
  - promisc support
  - idem with multicast packets
  - link status checks
  - rte_eth_link_up/down checks
  - mtu checks
  - rss checks
  - ptypes classification
  - stats checks
  - rte_flow checks
  - port restart checks + associated configuration persistency
  - flow control
  - rx/tx offloads
  - rx/tx descriptor status


-- 
David Marchand

[-- Attachment #2: Type: text/html, Size: 3181 bytes --]

<div dir="ltr"><div dir="ltr">Hello Jeremy,<br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Mon, Mar 18, 2019 at 4:56 PM Jeremy Plsek &lt;<a href="mailto:jplsek@iol.unh.edu">jplsek@iol.unh.edu</a>&gt; wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Based on last weeks meeting, I&#39;ve created a document to help start<br>
documenting what test cases<br>
should be worked on for cross platform support between NIC vendors and<br>
which test cases can be used now in the Community Lab.<br>
<br>
This list of test cases that should be worked on can be used when<br>
discussing the budget with the board.<br>
<br>
This also gives the Community Lab a list of tests that can be<br>
implemented in the mean time.<br>
<br>
<a href="https://docs.google.com/document/d/1WbRoIq0f5SmHM6D5sLbeh7hxRJdE2ppwB5JECHllLjg/edit?usp=sharing" rel="noreferrer" target="_blank">https://docs.google.com/document/d/1WbRoIq0f5SmHM6D5sLbeh7hxRJdE2ppwB5JECHllLjg/edit?usp=sharing</a><br>
(This document is publicly editable. If I notice the document changes<br>
outside the scope of the goal, I&#39;ll try to revert it and give<br>
individual access instead.)<br>
<br>
If creating this list over email is wanted instead, then we can do it<br>
in this thread instead of the document. (And I&#39;ll update the document<br>
as emails come in.)<br></blockquote><div><br></div><div>Well, for now, I prefer emails because I did not follow the CI discussions so far and I might be off topic.<br></div><div><br></div><div></div><div>Here I go, the list is large, I tried to write from the basic to the most advanced/less used.</div><div></div><div><br></div><div></div><div>- Non regression tests on basic startup (we can reuse the dpdk ut and/or enhance them), this part would be hw agnostic:<br></div><div>  - core checks: playing with cpus online/offline + dpdk coremask/list + checking cpu affinity of all dpdk threads<br></div><div>  - memory checks: a big topic because we have multiple modes at the moment</div><div>  - startup devices blacklist/whitelist</div><div>  - hotplug (unplug) of devices</div><div><br></div><div>- Non regression tests on eth drivers (testpmd seems the best candidate), this must be ran against a lot of nics/drivers:</div><div></div>  - reception of unicast packets: check that we receive packets for primary and secondary addresses (and only for them, meaning with checks that undesired packets are dropped following updates)<br><div>  - promisc support</div><div>  - idem with multicast packets</div><div><div>  - link status checks</div><div>  - rte_eth_link_up/down checks<br></div>  - mtu checks</div><div>  - rss checks<br></div><div></div>  - ptypes classification<div><div><div>  - stats checks<br></div><div>  - rte_flow checks<br></div>  - port restart checks + associated configuration persistency<br></div>  - flow control<br></div><div>  - rx/tx offloads</div><div></div><div>  - rx/tx descriptor status<br></div><div><br></div><div></div><div><br></div><div>-- <br></div><div>David Marchand<br></div></div></div>

  reply index

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-18 15:55 Jeremy Plsek
2019-03-19  8:38 ` David Marchand [this message]
2019-04-09 16:44 ` Thomas Monjalon

Reply instructions:

You may reply publically to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAJFAV8y8-LSh5vniZXR812ckKNa2ELEJVRKRzT53PVu2zO902w@mail.gmail.com \
    --to=david.marchand@redhat.com \
    --cc=ci@dpdk.org \
    --cc=jplsek@iol.unh.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

DPDK CI discussions

Archives are clonable:
	git clone --mirror http://inbox.dpdk.org/ci/0 ci/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 ci ci/ http://inbox.dpdk.org/ci \
		ci@dpdk.org
	public-inbox-index ci


Newsgroup available over NNTP:
	nntp://inbox.dpdk.org/inbox.dpdk.ci


AGPL code for this site: git clone https://public-inbox.org/ public-inbox