From mboxrd@z Thu Jan 1 00:00:00 1970
Return-Path: <dev-bounces@dpdk.org>
Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124])
by inbox.dpdk.org (Postfix) with ESMTP id A3EC544037;
Wed, 15 May 2024 17:52:12 +0200 (CEST)
Received: from mails.dpdk.org (localhost [127.0.0.1])
by mails.dpdk.org (Postfix) with ESMTP id 3DA0F4025C;
Wed, 15 May 2024 17:52:12 +0200 (CEST)
Received: from mail-ed1-f49.google.com (mail-ed1-f49.google.com
[209.85.208.49]) by mails.dpdk.org (Postfix) with ESMTP id 91AE54021D
for <dev@dpdk.org>; Wed, 15 May 2024 17:49:27 +0200 (CEST)
Received: by mail-ed1-f49.google.com with SMTP id
4fb4d7f45d1cf-574ea5184abso2170774a12.3
for <dev@dpdk.org>; Wed, 15 May 2024 08:49:27 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=gmail.com; s=20230601; t=1715788167; x=1716392967; darn=dpdk.org;
h=cc:to:subject:message-id:date:from:in-reply-to:references
:mime-version:from:to:cc:subject:date:message-id:reply-to;
bh=c2hm0IxWjt84AcgOM+feasE2fB+cmHke0+pANzkWZ80=;
b=Oa0hGVxbpHUGWYwzd0JM2aNiLC/oSQhjkWTdI+fV14x7kEAbWQ8PRw91IhjHSgMlUS
W6MT+C+dKrjNM6F/yOksN9qf8ySH3kd9GEbabjskq0uGmepc282+KQDRYeA7vcsNcxjg
OamL692BXlBlzUp+ewRY7Ks0IYJFMDKfE34l22xSnahmUPe4AZRzFMuGd3EwNNoeW/Ec
DnHYj7izivCvqUYiwNFrbXMA94ZUR/ZWQ9qRiIqtu/j33uNgZAzrF8nX2VpUYFqAbNE+
jJb3mj8hEnG/trKuIhNFp77Vjkb9fo53X+bJ6HI8niNbmQdXd/J3XVWrb7qOGc2+DQKx
6mlQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
d=1e100.net; s=20230601; t=1715788167; x=1716392967;
h=cc:to:subject:message-id:date:from:in-reply-to:references
:mime-version:x-gm-message-state:from:to:cc:subject:date:message-id
:reply-to;
bh=c2hm0IxWjt84AcgOM+feasE2fB+cmHke0+pANzkWZ80=;
b=s/GvrfyjQLw3XnYJWIbKxXTCCVqot1tI7JjCQJ9CNIMJJyP4ClWoGUs/j3XvFTC6gO
SOeWca9Nav7H0pytO/Ci+fUop5Umml61J51vu3n5vU3hwWOsEvH40XMcnR5yIIJIcv+Z
Cu0MPJgD6+OrUhCo9V3/eM+Rcmkqx7iUcL/TCC4RyM6TB7bIol1/rfCFAr4IfpXisNU6
IvUjWEZ6hOkZNSzhIk40CFaIQ6KrraSrIbU+nsOmqyX6XRvfrvY7X6uxBFQuGkaU+4x/
Y8dZoXn7++1lMGVWudCn8ASh39yLRuuwGIw68PXktSt2LHDFoN89+PMSBxBOLjqhFKcW
qYWg==
X-Gm-Message-State: AOJu0YzmUmZ0recgA7Taxk4gvfgzGEmUh9QGRSqd3uNEDceA6/dE8uMw
JSa/fyhD8JD9W/D9qq32fixKrgSbrNmPdVGXEnTP1JF+V+2BkYgymqyuneI+cWMcyql2jG1sduT
HizKYOo12ogFR7PugIeoE7EwM4QoP
X-Google-Smtp-Source: AGHT+IH5EDHKF/KwOeVYn3/9cekoMSkSB8L4ap/5rWCR//7qOT2vlClirtHJgU86HHofl0uCkGrqTpX7K0YoHzSCx+s=
X-Received: by 2002:a17:906:2b0d:b0:a59:be21:3575 with SMTP id
a640c23a62f3a-a5a2d672ee4mr1198234066b.51.1715788166731; Wed, 15 May 2024
08:49:26 -0700 (PDT)
MIME-Version: 1.0
References: <CABH5nGYTEJweZrockVxuU2YLgjjVi3QTL_0T__jyotxRt5mpSA@mail.gmail.com>
<20240515080434.33abfd44@hermes.local>
In-Reply-To: <20240515080434.33abfd44@hermes.local>
From: Nagesh Mallappa <mallappanagesh@gmail.com>
Date: Wed, 15 May 2024 21:19:15 +0530
Message-ID: <CABH5nGaP==N+4hKj0B-d97QoRqzBQ_Ffg_O602DAViMxprZGbQ@mail.gmail.com>
Subject: Re: RSS Hash value returned by the gvnic pmd dpdk driver not matching
Toeplitz
To: Stephen Hemminger <stephen@networkplumber.org>
Cc: dev@dpdk.org
Content-Type: multipart/alternative; boundary="000000000000301d6b06188010ef"
X-Mailman-Approved-At: Wed, 15 May 2024 17:52:10 +0200
X-BeenThere: dev@dpdk.org
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: DPDK patches and discussions <dev.dpdk.org>
List-Unsubscribe: <https://mails.dpdk.org/options/dev>,
<mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://mails.dpdk.org/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <https://mails.dpdk.org/listinfo/dev>,
<mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
--000000000000301d6b06188010ef
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Hi Stephen,
Thanks for your response,
I have verified that too, but that is not the case.
Thanks,
Nagesh
On Wed, May 15, 2024 at 8:34=E2=80=AFPM Stephen Hemminger <
stephen@networkplumber.org> wrote:
> On Wed, 15 May 2024 13:50:38 +0530
> Nagesh Mallappa <mallappanagesh@gmail.com> wrote:
>
> > Hi Team,
> >
> > We are using a C3 GCP instance having GVNIC with DPDK enabled.
> > We are using latest version of DPDK ( dpdk-24.03 ) to configure RSS key=
,
> we
> > understand C3 GCP instance with gvnic supports Toeplitz rss algorithm (
> > Reference -
> >
> https://github.com/google/compute-virtual-ethernet-dpdk/blob/main/gve_eth=
dev.c
> > ),
> > but we are hitting below issue.
> >
> > *Problem Statement - The RSS Hash value returned by the gvnic driver is
> > different and is not matching the one computed by Toeplitz.*
> >
> > Using below script to compute rss hash value from Toeplitz.
> > (https://gist.github.com/joongh/16867705b03b49e393cbf91da3cb42a7)
> >
> > Consider below data to compute RSS Hash value -
> >
> > Source IP - 79.124.62.122
> > Destination IP - 192.168.7.28
> > Source Port - 49442
> > Destination Port - 55327
> > RSS Key - {0x94e3e97e, 0x4abac401, 0x1e0edbf, 0xe059e1f4, 0xb6e27dff,
> > 0xdf55c916, 0xe9e6e4db, 0x1f9fb82c, 0x1e0edbf, 0xe059e1f4}
> >
> > *Hash Value returned by Toeplitz - 0x8bbafd0f*
> >
> > Gvnic dpdk pmd driver which uses Toeplitz RSS hash algorithm too is
> > expected to compute above RSS Hash value but we see the Hash value stor=
ed
> > in dpdk mbuf to be 0xc6a15941 which is not right.
> > From the DPDK side we have ensured RSS offloading configuration is done
> > appropriately, but still we are seeing above mismatch.
> >
> > We request you to please help us understand why this mismatch is
> happening
> > and is this expected behavior of gvnic on C3 instance.
>
> Could you see if it is a driver byte order issue by byteswapping the key?
>
--000000000000301d6b06188010ef
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
<div dir=3D"ltr">Hi Stephen,<div><br></div><div>Thanks for your response,</=
div><div>I have verified that too, but that is not the case.</div><div><br>=
</div><div>Thanks,</div><div>Nagesh</div></div><br><div class=3D"gmail_quot=
e"><div dir=3D"ltr" class=3D"gmail_attr">On Wed, May 15, 2024 at 8:34=E2=80=
=AFPM Stephen Hemminger <<a href=3D"mailto:stephen@networkplumber.org">s=
tephen@networkplumber.org</a>> wrote:<br></div><blockquote class=3D"gmai=
l_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,20=
4,204);padding-left:1ex">On Wed, 15 May 2024 13:50:38 +0530<br>
Nagesh Mallappa <<a href=3D"mailto:mallappanagesh@gmail.com" target=3D"_=
blank">mallappanagesh@gmail.com</a>> wrote:<br>
<br>
> Hi Team,<br>
> <br>
> We are using a C3 GCP instance having GVNIC with DPDK enabled.<br>
> We are using latest version of DPDK ( dpdk-24.03 ) to configure RSS ke=
y, we<br>
> understand C3 GCP instance with gvnic supports Toeplitz rss algorithm =
(<br>
> Reference -<br>
> <a href=3D"https://github.com/google/compute-virtual-ethernet-dpdk/blo=
b/main/gve_ethdev.c" rel=3D"noreferrer" target=3D"_blank">https://github.co=
m/google/compute-virtual-ethernet-dpdk/blob/main/gve_ethdev.c</a><br>
> ),<br>
> but we are hitting below issue.<br>
> <br>
> *Problem Statement - The RSS Hash value returned by the gvnic driver i=
s<br>
> different and is not matching the one computed by Toeplitz.*<br>
> <br>
> Using below script to compute rss hash value from Toeplitz.<br>
> (<a href=3D"https://gist.github.com/joongh/16867705b03b49e393cbf91da3c=
b42a7" rel=3D"noreferrer" target=3D"_blank">https://gist.github.com/joongh/=
16867705b03b49e393cbf91da3cb42a7</a>)<br>
> <br>
> Consider below data to compute RSS Hash value -<br>
> <br>
> Source IP - 79.124.62.122<br>
> Destination IP - 192.168.7.28<br>
> Source Port - 49442<br>
> Destination Port - 55327<br>
> RSS Key - {0x94e3e97e, 0x4abac401, 0x1e0edbf, 0xe059e1f4, 0xb6e27dff,<=
br>
> 0xdf55c916, 0xe9e6e4db, 0x1f9fb82c, 0x1e0edbf, 0xe059e1f4}<br>
> <br>
> *Hash Value returned by Toeplitz - 0x8bbafd0f*<br>
> <br>
> Gvnic dpdk pmd driver which uses Toeplitz RSS hash algorithm too is<br=
>
> expected to compute above RSS Hash value but we see the Hash value sto=
red<br>
> in dpdk mbuf to be 0xc6a15941 which is not right.<br>
> From the DPDK side we have ensured RSS offloading configuration is don=
e<br>
> appropriately, but still we are seeing above mismatch.<br>
> <br>
> We request you to please help us understand why this mismatch is happe=
ning<br>
> and is this expected behavior of gvnic on C3 instance.<br>
<br>
Could you see if it is a driver byte order issue by byteswapping the key?<b=
r>
</blockquote></div>
--000000000000301d6b06188010ef--