From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail5.wrs.com (mail5.windriver.com [192.103.53.11]) by dpdk.org (Postfix) with ESMTP id 03ECC2A5E for ; Mon, 27 Feb 2017 18:09:43 +0100 (CET) Received: from ALA-HCB.corp.ad.wrs.com (ala-hcb.corp.ad.wrs.com [147.11.189.41]) by mail5.wrs.com (8.15.2/8.15.2) with ESMTPS id v1RH9hLo026277 (version=TLSv1 cipher=AES128-SHA bits=128 verify=OK); Mon, 27 Feb 2017 09:09:43 -0800 Received: from ALA-MBC.corp.ad.wrs.com ([fe80::fcbe:9b7:1141:89a1]) by ALA-HCB.corp.ad.wrs.com ([147.11.189.41]) with mapi id 14.03.0294.000; Mon, 27 Feb 2017 09:09:42 -0800 From: "Legacy, Allain" To: Stephen Hemminger CC: "YIGIT, FERRUH" , "dev@dpdk.org" Thread-Topic: [dpdk-dev] [PATCH v2 07/15] net/avp: driver registration Thread-Index: AQHSkRoLGNwhLi7UXkOSz4o+ss+I06F9Fb7A Date: Mon, 27 Feb 2017 17:09:42 +0000 Message-ID: <70A7408C6E1BFB41B192A929744D85238A757FFC@ALA-MBC.corp.ad.wrs.com> References: <1487985795-136044-1-git-send-email-allain.legacy@windriver.com> <1488136143-116389-1-git-send-email-allain.legacy@windriver.com> <1488136143-116389-8-git-send-email-allain.legacy@windriver.com> <20170227085339.3f14a3fb@xeon-e3> In-Reply-To: <20170227085339.3f14a3fb@xeon-e3> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [128.224.140.166] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [PATCH v2 07/15] net/avp: driver registration 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: , X-List-Received-Date: Mon, 27 Feb 2017 17:09:44 -0000 > -----Original Message----- > From: Stephen Hemminger [mailto:stephen@networkplumber.org] >=20 > What exactly is the spinlock protecting? The control operations in DPDK = are > defined to be not thread safe. I.e it is responsibility of caller to sync= hronize. > Therefore is lock really needed? The implementation assumes that interrupts (for VM live-migration) may be s= erviced on a thread other than the thread used to do normal device operatio= ns (i.e., configure, start, stop, etc...). We use the spinlock to protect= the device flags and the request/response queue against concurrent access = from the management thread and the interrupt thread.=20