OpenVZ / Virtuozzo kernel development (devel@openvz.org)
 help / color / mirror / Atom feed
From: Konstantin Khorenko <khorenko@virtuozzo.com>
To: Eva Kurchatova <eva.kurchatova@virtuozzo.com>
Cc: devel@openvz.org
Subject: Re: [Devel] [PATCH hci-8.1 v3 1/1] selftests: net: skip the veth GRO checks where they cannot run
Date: Wed, 30 Sep 2026 15:51:44 +0200	[thread overview]
Message-ID: <179077630438.3598112.6223753507691542141.b4-review@b4> (raw)
In-Reply-To: <20260928193506.549055-2-eva.kurchatova@virtuozzo.com>

> The three rx-udp-gro-forwarding checks in veth.sh count 10 packets
> where they expect 1 on a host that has a vxlan or another UDP tunnel,
> in any namespace, and pass on a host that has none.
> 
> Any UDP encapsulation socket on the machine moves UDP GRO to the socket
> lookup path, see commit f2e9de210d50 ("udp: gro behind static key"),
> and rx-udp-gro-forwarding is then not consulted at all, see commit
> 36707061d6ba ("udp: allow forwarding of plain (non-fraglisted) UDP GRO
> packets"). The switch is global, so the test namespaces do not help,
> and nothing the test can set brings the device flag back.
> 
> Skip those three checks when UDP encapsulation is in use: a vxlan,
> geneve or bareudp device, a fou port or an l2tp tunnel in any
> namespace, or a socket on the vxlan or geneve port.
> 
> https://virtuozzo.atlassian.net/browse/VSTOR-139651
> Feature: fix selftests
> Signed-off-by: Eva Kurchatova <eva.kurchatova@virtuozzo.com>
>
> diff --git a/tools/testing/selftests/net/veth.sh b/tools/testing/selftests/net/veth.sh
> index 6bb7dfaa30b6a..3fcd63f93be72 100755
> --- a/tools/testing/selftests/net/veth.sh
> +++ b/tools/testing/selftests/net/veth.sh
> @@ -97,6 +97,42 @@ chk_channels() {
>  	fi
>  }
>  
> +udp_encap_in_use() {
> +	local ns cmd t
> +
> +	for ns in "" `ip netns list 2>/dev/null | awk '{print $1}'`; do
> +		if [ -n "$ns" ]; then
> +			cmd="ip netns exec $ns"
> +		else
> +			cmd=""
> +		fi
> +
> +		for t in vxlan geneve bareudp; do
> +			$cmd ip -o link show type $t 2>/dev/null | \
> +				grep -q . && return 0
> +		done
> +		$cmd ip fou show 2>/dev/null | grep -q . && return 0
> +		$cmd ip l2tp show tunnel 2>/dev/null | grep -q . && return 0
> +		$cmd ss -uapn 2>/dev/null | \
> +			grep -qE ':(4789|6081)([^0-9]|$)' && return 0
> +	done
> +	return 1
> +}
> +
> +# Only meaningful where nothing else has turned udp_encap_needed_key on
> +chk_gro_fwd() {
> +	local msg="$1"
> +	local expected=$2
> +
> +	if [ $UDP_ENCAP -ne 0 ]; then

Bug: chk_gro_fwd skips the checks on a clean host and runs them on a host with a tunnel.

udp_encap_in_use returns 0 when encapsulation is found, so UDP_ENCAP=$? yields 0 = "in use". The top level gets this
right:

[ $UDP_ENCAP -eq 0 ] && \
      echo "UDP encapsulation in use, forwarding tests will be skipped"

but the helper has it backwards:

if [ $UDP_ENCAP -ne 0 ]; then
      ...
      echo " skip - UDP encapsulation in use"

-ne 0 is true when there is no encapsulation.

-- 
Konstantin Khorenko <khorenko@virtuozzo.com>
_______________________________________________
Devel mailing list
Devel@openvz.org
https://lists.openvz.org/mailman/listinfo/devel

      reply	other threads:[~2026-09-30 13:53 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-28 19:33 [Devel] [PATCH hci-8.1 v3 0/1] " Eva Kurchatova
2026-09-28 19:33 ` [Devel] [PATCH hci-8.1 v3 1/1] " Eva Kurchatova
2026-09-30 13:51   ` Konstantin Khorenko [this message]

Reply instructions:

You may reply publicly 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=179077630438.3598112.6223753507691542141.b4-review@b4 \
    --to=khorenko@virtuozzo.com \
    --cc=devel@openvz.org \
    --cc=eva.kurchatova@virtuozzo.com \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox