From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dpdk.org (dpdk.org [92.243.14.124]) by dpdk.space (Postfix) with ESMTP id 0642DA0096 for ; Mon, 6 May 2019 13:00:20 +0200 (CEST) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id DE3B12BAB; Mon, 6 May 2019 13:00:19 +0200 (CEST) Received: from alln-iport-6.cisco.com (alln-iport-6.cisco.com [173.37.142.93]) by dpdk.org (Postfix) with ESMTP id 02F491F28 for ; Mon, 6 May 2019 13:00:17 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=1072; q=dns/txt; s=iport; t=1557140418; x=1558350018; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=/6PLdEPOe+UpL5LoW3L5NY8aUaG7ddrmMOWohjRCGPE=; b=Rz18od4TRBt/sF8wuu/JZc6SDs+FqemXLsYXjL66u3lj8K1XOjkmwBtE K64CDBzNB2805fcYOMnQpHZxdFhvnuR0rNZt7Ag8Gw/poOUX0N5xDQg4S d22IHfvu8zm+/LBCxDJy564FfqOu7poP/3lHjfVjypfYHgW1whDX1SIWS 0=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: =?us-ascii?q?A0BdAAAiE9Bc/4ENJK1lGQEBAQEBAQE?= =?us-ascii?q?BAQEBAQcBAQEBAQGBVAEBAQEBAQsBghCBbSgKmSd+g2EBlW0OAQGEbQKCDiM?= =?us-ascii?q?3Bg4BAwEBBAEBAgECbSiFSgEBAQECAXkFCwIBCBEEAQEBCSUPIx0IAgQBCQQ?= =?us-ascii?q?FhR4PrEaKKIEyAYtNF4F/gRGCXTU+hESFYgSTDJQmCQIdgWySRRuVSIwblGk?= =?us-ascii?q?CERWBMDUigVZwFYMnkFFBMZAegSEBAQ?= X-IronPort-AV: E=Sophos;i="5.60,437,1549929600"; d="scan'208";a="272054552" Received: from alln-core-9.cisco.com ([173.36.13.129]) by alln-iport-6.cisco.com with ESMTP/TLS/DHE-RSA-SEED-SHA; 06 May 2019 11:00:16 +0000 Received: from XCH-ALN-016.cisco.com (xch-aln-016.cisco.com [173.36.7.26]) by alln-core-9.cisco.com (8.15.2/8.15.2) with ESMTPS id x46B0GvV012037 (version=TLSv1.2 cipher=AES256-SHA bits=256 verify=FAIL); Mon, 6 May 2019 11:00:16 GMT Received: from xch-rcd-017.cisco.com (173.37.102.27) by XCH-ALN-016.cisco.com (173.36.7.26) with Microsoft SMTP Server (TLS) id 15.0.1473.3; Mon, 6 May 2019 06:00:15 -0500 Received: from xch-rcd-017.cisco.com ([173.37.102.27]) by XCH-RCD-017.cisco.com ([173.37.102.27]) with mapi id 15.00.1473.003; Mon, 6 May 2019 06:00:15 -0500 From: "Jakub Grajciar -X (jgrajcia - PANTHEON TECHNOLOGIES at Cisco)" To: Honnappa Nagarahalli , Ferruh Yigit , "dev@dpdk.org" CC: nd , "Damjan Marion (damarion)" Thread-Topic: [dpdk-dev] [RFC v5] /net: memory interface (memif) Thread-Index: AQHU4KZ16NJCh3r9PESlg1+aACrGcKYdLYsAgBgjuyGAI70qEIAFIX7I Date: Mon, 6 May 2019 11:00:15 +0000 Message-ID: <1557140414503.9161@cisco.com> References: <20190220115254.18724-1-jgrajcia@cisco.com> <20190322115727.4358-1-jgrajcia@cisco.com>, <0762da59-4a97-474f-7d67-e3bd8daf50f2@intel.com> <1556800556778.14516@cisco.com>, In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-ms-exchange-transport-fromentityheader: Hosted x-originating-ip: [10.61.69.75] Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Outbound-SMTP-Client: 173.36.7.26, xch-aln-016.cisco.com X-Outbound-Node: alln-core-9.cisco.com Subject: Re: [dpdk-dev] [RFC v5] /net: memory interface (memif) X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" Message-ID: <20190506110015.Q6qcIs5Jx1G5omYdG8ONZVeYh07xU2JUa5PY0ihHhzI@z> =0A= ________________________________________=0A= From: Honnappa Nagarahalli =0A= Sent: Friday, May 3, 2019 6:27 AM=0A= To: Jakub Grajciar; Ferruh Yigit; dev@dpdk.org; Honnappa Nagarahalli=0A= Cc: nd; nd=0A= Subject: RE: [dpdk-dev] [RFC v5] /net: memory interface (memif)=0A= =0A= > On 3/22/2019 11:57 AM, Jakub Grajciar wrote:=0A= > > Memory interface (memif), provides high performance packet transfer=0A= > > over shared memory.=0A= > >=0A= > > Signed-off-by: Jakub Grajciar =0A= >=0A= =0A= <...>=0A= =0A= > With that in mind, I believe that 23Mpps is fine performance. No=0A= > performance target is=0A= > defined, the goal is to be as fast as possible.=0A= Use of C11 atomics have proven to provide better performance on weakly orde= red architectures (at least on Arm). IMO, C11 atomics should be used to imp= lement the fast path functions at least. This ensures optimal performance o= n all supported architectures in DPDK.=0A= =0A= Atomics are not required by memif driver.=