From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <stephen@networkplumber.org>
Received: from mail-pg1-f179.google.com (mail-pg1-f179.google.com
 [209.85.215.179]) by dpdk.org (Postfix) with ESMTP id 781CF1BBE4
 for <dev@dpdk.org>; Sat, 12 Jan 2019 00:37:17 +0100 (CET)
Received: by mail-pg1-f179.google.com with SMTP id c25so6948124pgb.4
 for <dev@dpdk.org>; Fri, 11 Jan 2019 15:37:17 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
 d=networkplumber-org.20150623.gappssmtp.com; s=20150623;
 h=date:from:to:cc:subject:message-id:in-reply-to:references
 :mime-version:content-transfer-encoding;
 bh=DczlQFgt92Aw3pmdlg/ulerTYdTZLAIwKI5jvNmoSB8=;
 b=tMjrudrnk2WhFQZCAzHD0Kj6PmDDmgfI3g+UOxxq2RAJBiI50A9+NXaw7unMdogLuj
 wEAMiROxv7kjll7dgeCrx8LrnPNEhihHchmIU975Ectrdgpe68g9nWMJ15x/PCiwMzID
 rheJCwwV1Gzj9vfw56Z5D4sJn7FlCeJuL4yrrhVHPEDMhWKq0jFtOkVmGof+91tvuHww
 7C+5ltOQs3N9MbGqkG8RcaYvuaxrYbHDAUR2inLEbA7SMBYcLAuoMvvZZ3Rf7Do/UP4v
 hIbLHHGsKbhyeOE05sSWc46ECG+OoUMxD5YxbONZLCHutPnkEIrTVi3Xaza791Elmat+
 d4DQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
 d=1e100.net; s=20161025;
 h=x-gm-message-state:date:from:to:cc:subject:message-id:in-reply-to
 :references:mime-version:content-transfer-encoding;
 bh=DczlQFgt92Aw3pmdlg/ulerTYdTZLAIwKI5jvNmoSB8=;
 b=F5km4JGPjg/b2hCLsKPSUv7qHi/6C384Gu7O15XGVeJz0zjgoUopjv6Sats83r/PoK
 bKg4HZjqXGgoRRTxJ2hQtuL8AU7YG+cdmOp+TZfSuzG1SRz87rzvZJ+KqdAeLq3VWndV
 vSczZdUkOQBcSatz30Nk0CcDasquU3NLHM/RsxaIEbwI8/9O6JcfMqIhDaJt8MeEvebT
 io9trBbTjy6XKmd9jS6qSJJfC+xAdlJUWi9C37C5XRNky+Vklxk4FM5nuS0RfHSejeVW
 mM91LeI8L7nEkEUjtNhRphG0PiKUi3qYlVLY3A6+b2a/MurmAnOArRrA1ndPzRXItZdv
 unbQ==
X-Gm-Message-State: AJcUukdzlDXSu/YEpnlXfm/lVHr+QIoGLv8BpF6EXwyEwcH8rBQ/Wa5c
 dbUFQw3C1tPo6ug+8Jk48iWTbg==
X-Google-Smtp-Source: ALg8bN4APLEjqZr5PMxP7ZNqR/W0bGEwbML1YDgjfGAp9MYfpCC+5EnYrMU3EHUvznS/XxR+B/eOjw==
X-Received: by 2002:a62:61c3:: with SMTP id v186mr16641066pfb.55.1547249836055; 
 Fri, 11 Jan 2019 15:37:16 -0800 (PST)
Received: from hermes.lan (204-195-22-127.wavecable.com. [204.195.22.127])
 by smtp.gmail.com with ESMTPSA id d80sm220786553pfm.146.2019.01.11.15.37.15
 (version=TLS1_2 cipher=ECDHE-RSA-CHACHA20-POLY1305 bits=256/256);
 Fri, 11 Jan 2019 15:37:15 -0800 (PST)
Date: Fri, 11 Jan 2019 15:37:09 -0800
From: Stephen Hemminger <stephen@networkplumber.org>
To: "Soni, Shivam" <shivsoni@amazon.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>, "users@dpdk.org" <users@dpdk.org>
Message-ID: <20190111153709.3117c539@hermes.lan>
In-Reply-To: <454CCFA5-5843-441F-9C6D-33E807419574@amazon.com>
References: <454CCFA5-5843-441F-9C6D-33E807419574@amazon.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Subject: Re: [dpdk-dev] TX unable to enqueue packets to NIC due to no free
 TX descriptor
X-BeenThere: dev@dpdk.org
X-Mailman-Version: 2.1.15
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>
X-List-Received-Date: Fri, 11 Jan 2019 23:37:17 -0000

On Fri, 11 Jan 2019 22:10:39 +0000
"Soni, Shivam" <shivsoni@amazon.com> wrote:

> Hi All,
>=20
> We are trying to debug and fix an issue. After the deployment, in few of =
the hosts we see an issue where TX is unable to enqueue packets to NIC. On =
rebouncing or restarting our packet processor daemon, issue gets resolved.
>=20
> We are using IntelDPDK version 17.11.4 and i40e drivers.
>=20
> On looking into driver=E2=80=99s code, we found that whenever the issue i=
s happening the value for nb_tx_free is =E2=80=980=E2=80=99. And then it tr=
ies to free the buffer by calling function =E2=80=98i40e_tx_free_bufs=E2=80=
=99.
>=20
> This method returns early as the buffer its trying to free says it hasn=
=E2=80=99t finished transmitting yet. The method returns at this if conditi=
on:
>=20
> /* check DD bits on threshold descriptor */
> if ((txq->tx_ring[txq->tx_next_dd].cmd_type_offset_bsz &
>                 rte_cpu_to_le_64(I40E_TXD_QW1_DTYPE_MASK)) !=3D
>                 rte_cpu_to_le_64(I40E_TX_DESC_DTYPE_DESC_DONE)) {
> return 0;
> }
>=20
> Hence nb_tx_free remains 0.
>=20
> Our tx descriptor count is 1024.
>=20
> How can we fix this issue.  Can someone help us out here please

Use bigger mbuf pool.  For safety the mbuf pool has to be big enough
for Nports * (NRxd + NTxd) + NCore * (mbuf_pool_cache_size + burst_size)

Each NIC might get full receive ring and full transmit ring
and each active core might be processing a burst of packets and have
free buffers sitting in the mbuf pool cache. This doesn't account for addit=
ional
mbuf's created if doing things like reassembly, encryption, re-encapsulatio=
n, or compression

Anything smaller and your application is relying on statistical averages
to never see resource exhaustion; overcommitment