From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 036F541C61 for ; Fri, 10 Feb 2023 14:18:41 +0100 (CET) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 4EE6740EE6; Fri, 10 Feb 2023 14:18:41 +0100 (CET) Received: from mail-qv1-f47.google.com (mail-qv1-f47.google.com [209.85.219.47]) by mails.dpdk.org (Postfix) with ESMTP id A22E640687 for ; Fri, 10 Feb 2023 14:18:40 +0100 (CET) Received: by mail-qv1-f47.google.com with SMTP id q10so3461449qvt.10 for ; Fri, 10 Feb 2023 05:18:40 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:from:to:cc:subject:date:message-id:reply-to; bh=iS8/Lfb3a8sM52LSfhqbb/XhVzL1IuCeGwWbFjISXV4=; b=IlrV/uxW3k8fyyn3saI5t2h5y4eI8h/qD7uZIDMnqOtNxW7/tTciuM0oXo4vX56Bi+ 6yo6wMH3qUfxXm9cPno1ObcEZB427w8dPh66LfvVm3g+1OdfwZ5M1XfrE7wrs+uyfFJC SKr+1xcUI2OR+OaR3kUie2vDeqUu8Fr4pLqhSl14EZTbjKN6slUZZsZUbMYESWDJ0Rsz nl4QpBooMpmheuuI6txkVn/k2hoZtXAR303dLHE6ALcbYYOVdzE8naBfp4kU0P6CQo9b a9TSBuAil1sYDjlTrFl3Tb5Po6UhLOOR7ykhixpFO8iTixaKUPNZgACIyLc41scsUhPk 8B3A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; 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=iS8/Lfb3a8sM52LSfhqbb/XhVzL1IuCeGwWbFjISXV4=; b=T8N3A5Km+4hJu+EPxluhT+XbHvRcexAufw05VcuDwm8Eu9EWfyyluZrrnz7J0xGHeP j82SjAmEo0S16HNFTGeWxUkMJ5K+57vr+UqVhdimAcREjA/OP4/QpYapfaOexlnouQKy /gR96oA9VT4PiHPpU0rePjy7S21c+eIr8vC9eB8uxwDyM11IuqoOIJaqeMjKQJ6eo7HY kblKfM3VZo0l6gApilmCTpm/GYwoq1kNZ5e/JQIq3dZiEu76FvmMW3lhBdrajcB8X5Ym lIZ88CrWdB1kHOC7wViD/JRpMIYillPoKZkq/obDUu5GfrDxW3dJDkUQi/wzPbiioBbG DFeg== X-Gm-Message-State: AO0yUKXx//wypQfmGxeYBzonJw1UIzm+RXBwv8vjFn+9adfAECzZu1ML jsoPA+7ei25wlnQEKzDUbxDWmjB7l06itQ/EPhZMWpBb+Xk0A4IO X-Google-Smtp-Source: AK7set9jpbYRnRwt8rpLIwtsKDZdJk2ab58THbiGyL4Ph5BoS/xqxVmXEbUp1mNlwjXkYAjpkWYUuaE2feXIz1daolI= X-Received: by 2002:a0c:8ec8:0:b0:537:7bd7:29c3 with SMTP id y8-20020a0c8ec8000000b005377bd729c3mr1037203qvb.4.1676035120023; Fri, 10 Feb 2023 05:18:40 -0800 (PST) MIME-Version: 1.0 References: <484931676028579@mail.yandex.ru> In-Reply-To: <484931676028579@mail.yandex.ru> From: Pavel Vazharov Date: Fri, 10 Feb 2023 15:18:26 +0200 Message-ID: Subject: Re: RSS on 82599ES/X520-2 To: "Ruslan R. Laishev" Cc: "users@dpdk.org" Content-Type: multipart/alternative; boundary="000000000000f5cca905f4585593" X-BeenThere: users@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK usage discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: users-bounces@dpdk.org --000000000000f5cca905f4585593 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hi, I'm almost sure 82599ES doesn't support RSS with the L4 protocols (TCP, UDP, etc). As far as I remember I hit this issue about a year ago. What is the value of `dev_info.flow_type_rss_offloads` obtained like: rte_eth_dev_info dev_info; rte_eth_dev_info_get(nic_port_id, &dev_info); The `flow_type_rss_offloads` contains the supported RSS flags by the device= . On Fri, Feb 10, 2023 at 3:10 PM Ruslan R. Laishev wrote: > Hello! > > I have tried to use RSS, I do the same piece of code on NXP ls2160 and on > Intel with the 82599ES, on NXP I see that incoming traffic is distributed > across queues. > On Intel I see that all incoming packet still in the queue #0. > So, what I'm need to check additionally (at initialization/setup time) = ? > > Thanks-in-advance. > > > > 10-02-2023 15:17:24.718 128431 [CGWPROC\tootoo2_cgw_start:1373] > %TTR2CGW-I: 3---TTR3 [Device:0000:02:00.1, _PEA03:, WAN] --- modified RS= S > hash function based on hardware support,requested:0x3bffc configured:0x38= d34 > > Translated bits for readability: > 10-02-2023 15:17:24.718 128431 [CGWPROC\tootoo2_cgw_start:1383] > RSS Offloads: > RTE_ETH_RSS_IPV4|RTE_ETH_RSS_NONFRAG_IPV4_TCP|RTE_ETH_RSS_NONFRAG_IPV4_UD= P|RTE_ETH_RSS_IPV6|RTE_ETH_RSS_NONFRAG_IPV6_TCP|RTE_ETH_RSS_NONFRAG_IPV6_UD= P|RTE_ETH_RSS_IPV6_EX|RTE_ETH_RSS_IPV6_TCP_EX|RTE_ETH_RSS_IPV6_UDP_EX > > Test IP traffic is generated by IPERF3 (TCP). Counters from *struct* rte_= eth_stats > follows: > > at NXP lx216010-02-2023; 15:19:18; dpni.2; _PEA01:; 1---PUZZLE; LAN; > 272194; 8086631; 18122106; 12242301932; 0; 0; 0; 0 > q0[33929; 1446431; 0; 0; 0]; > q1[46998; 1328423; 0; 0; 0]; > q2[28776; 1623997; 0; 0; 0]; > q3[41012; 1195463; 0; 0; 0]; > q4[23007; 427604; 0; 0; 0]; > q5[35160; 1150040; 0; 0; 0]; > q6[16089; 760039; 0; 0; 0]; > q7[47240; 154326; 0; 0; 0]; > > on Intel with 82599ES/520-2 > 10-02-2023; 15:19:44; 0000:02:00.0; _PEA02:; 4---PUZZLE; LAN; 8122102; > 442634; 12296494642; 29295238; 0; 0; 0; 0 > q0[8122102; 442635; 12296494642; 29295286; 0]; > q1[0; 0; 0; 0; 0]; > q2[0; 0; 0; 0; 0]; > q3[0; 0; 0; 0; 0]; > q4[0; 0; 0; 0; 0]; > q5[0; 0; 0; 0; 0]; > q6[0; 0; 0; 0; 0]; > q7[0; 0; 0; 0; 0]; > > > > > > --- > =D0=A1 =D1=83=D0=B2=D0=B0=D0=B6=D0=B5=D0=BD=D0=B8=D0=B5=D0=BC, > Ruslan R. Laishev > OpenVMS bigot, natural born system/network progger, C contractor. > +79013163222 > +79910009922 > > --000000000000f5cca905f4585593 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Hi,

I'm almost sure 8= 2599ES doesn't support RSS with the L4 protocols (TCP, UDP, etc). As fa= r as I remember I hit this issue about a year ago.

What = is the value of `dev_info.flow_type_rss_offloads` obtained like:

rt= e_eth_dev_info dev_info;
rte_eth_dev_info_get(nic_port_id, &dev_info= );


The `flow_type_rss_offloads` contains the supported RSS= flags by the device.

On Fri, Feb 10, 2023 at 3:10 PM Ruslan R. Laishev = <zator@yandex.ru> wrote:
Hello!
=C2=A0
<= div>I have tried to u= se RSS, I do the same piece of code on NXP ls2160 and on Intel with the 825= 99ES, on NXP I see that incoming traffic is distributed across queues.
On Intel= I see that all incoming packet still in the queue #0.
So, what I'm need t= o check additionally (at initialization/setup time) =C2=A0 ?
=C2=A0
T= hanks-in-advance.
=C2=A0
=C2=A0
=C2=A0=
10-02-2023 15:17:24.718 128431 [CGWPROC\tootoo2_cgw_start:1373] %TTR2CGW= -I: =C2=A03---TTR3 [Device:0000:02:00.1, _PEA03:, WAN] --- modified RSS has= h function based on hardware support,requested:0x3bffc configured:0x38d34
=C2=A0
Translated bits for readability:
10-02-2023 15:17:2= 4.718 128431 [CGWPROC\tootoo2_cgw_start:1383] =C2=A0 =C2=A0 =C2=A0 =C2=A0 R= SS Offloads: RTE_ETH_RSS_IPV4|RTE_ETH_RSS_NONFRAG_IPV4_TCP|RTE_ETH_RSS_NONF= RAG_IPV4_UDP|RTE_ETH_RSS_IPV6|RTE_ETH_RSS_NONFRAG_IPV6_TCP|RTE_ETH_RSS_NONF= RAG_IPV6_UDP|RTE_ETH_RSS_IPV6_EX|RTE_ETH_RSS_IPV6_TCP_EX|RTE_ETH_RSS_IPV6_U= DP_EX
=C2=A0
Test IP traffic is generated by IPERF3 (TCP). Co= unters from struct= rte_eth_stats follows:
=C2=A0
at NXP lx216010-02-2023;= 15:19:18; dpni.2; _PEA01:; 1---PUZZLE; LAN; 272194; 8086631; 18122106; 122= 42301932; 0; 0; 0; 0
q0[33929; 1446431; 0; 0; 0];
q1[46998; 1328423= ; 0; 0; 0];
q2[28776; 1623997; 0; 0; 0];
q3[41012; 1195463; 0; 0;= 0];
q4[23007; 427604; 0; 0; 0];
q5[35160; 1150040; 0; 0; 0];
q= 6[16089; 760039; 0; 0; 0];
q7[47240; 154326; 0; 0; 0];
=C2=A0
on Intel with 82599ES/520-2
10-02-2023; 15:19:44; 0000:02:00.0; _P= EA02:; 4---PUZZLE; LAN; 8122102; 442634; 12296494642; 29295238; 0; 0; 0; 0<= /span>
q0[8122102; 442635; 12296494642; 29295286; 0];
q1[0; 0; 0; 0; 0];
q2[0; 0; 0; 0; 0];
q3[0; 0; 0; 0; 0];
q4[0; 0; 0; 0; 0];
q= 5[0; 0; 0; 0; 0];
q6[0; 0; 0; 0; 0];
q7[0; 0; 0; 0; 0];
=C2=A0
=C2=A0
=C2=A0
=C2=A0
=
=C2=A0
---=C2=A0
=D0=A1 =D1=83=D0=B2=D0=B0=D0=B6=D0=B5=D0=BD=D0=B8=D0=B5=D0=BC,
Ru= slan R. Laishev
OpenVMS bigot, natural born system/network progger,=C2=A0C cont= ractor.
+79013163222
+79910009922
=C2=A0
--000000000000f5cca905f4585593--