From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by dpdk.org (Postfix) with ESMTP id 092D05A6A for ; Thu, 12 Feb 2015 06:26:46 +0100 (CET) Received: from orsmga001.jf.intel.com ([10.7.209.18]) by orsmga102.jf.intel.com with ESMTP; 11 Feb 2015 21:22:56 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.09,563,1418112000"; d="scan'208";a="651050851" Received: from pgsmsx105.gar.corp.intel.com ([10.221.44.96]) by orsmga001.jf.intel.com with ESMTP; 11 Feb 2015 21:26:43 -0800 Received: from shsmsx152.ccr.corp.intel.com (10.239.6.52) by PGSMSX105.gar.corp.intel.com (10.221.44.96) with Microsoft SMTP Server (TLS) id 14.3.195.1; Thu, 12 Feb 2015 13:26:41 +0800 Received: from shsmsx101.ccr.corp.intel.com ([169.254.1.192]) by SHSMSX152.ccr.corp.intel.com ([169.254.6.46]) with mapi id 14.03.0195.001; Thu, 12 Feb 2015 13:26:39 +0800 From: "Xie, Huawei" To: "dev@dpdk.org" Thread-Topic: [dpdk-dev] [PATCH v2 00/11] qemu vhost-user support Thread-Index: AdBGhHohL6mUXlC3TMCqi4WrmIqIkw== Date: Thu, 12 Feb 2015 05:26:39 +0000 Message-ID: References: <1423717649-11818-1-git-send-email-huawei.xie@intel.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.239.127.40] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [PATCH v2 00/11] qemu vhost-user support X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 12 Feb 2015 05:26:47 -0000 Sorry, forget to add --in-reply-to and add v2 changes.=0A= =0A= v2 changes:=0A= * vhost-cuse and vhost-user have their own set_mem_table message handler.= =0A= * rework and refine mutex operation in fd management to avoid race conditio= n=0A= * increase listen backlog=0A= * use memset to fix initialization compiler errors reported by haifeng lin= =0A= * code style fixes=0A= On 2/12/2015 1:08 PM, Huawei Xie wrote:=0A= > vhost-user supports passing vring information to a seperate vhost enabled= =0A= > user space process, normally a user space vSwitch, through unix domain so= cket.=0A= >=0A= > In previous DPDK version, we implement a user space character device driv= er=0A= > vhost-cuse in user space DPDK process. vring information is passed to the= =0A= > cuse driver through ioctl call, including eventfds for interrupt injectio= n and=0A= > host notification. A kernel module is developed to copy these fds from=0A= > qemu process into our process. We also need some trick to map guest memor= y.=0A= > (TODO: kickfd/callfd is reversed which causes confusion)=0A= >=0A= > known issue in vhost-user implementation in QEMU, reported by haifeng.lin= @huawei.com=0A= > * QEMU doesn't send correct memory region information with multiple numa = node configuration=0A= > http://lists.gnu.org/archive/html/qemu-devel/2014-12/msg01454.htm= l=0A= >=0A= > Thanks Tetsuya for reporting the issue that "FD_ISSET would crash when re= ceive -1=0A= > as fd on Ubuntu 14.04".=0A= >=0A= > Huawei Xie (11):=0A= > enable VIRTIO_NET_F_CTRL_RX=0A= > create vhost_cuse directory and move vhost-net-cdev.c into vhost_cuse=0A= > rename vhost-net-cdev.h to vhost-net.h=0A= > move fd copying(from qemu process into vhost process) to eventfd_copy.c= =0A= > copy host_memory_map from virtio-net.c to a new file virtio-net-cdev.c= =0A= > make host_memory_map a more generic function.=0A= > implement cuse_set_memory_table in virtio-net-cdev.c=0A= > add select based event driven processing=0A= > vhost user support=0A= > support dev->ifname=0A= > support calling rte_vhost_driver_register after rte_vhost_driver_session= _start=0A= >=0A= > lib/librte_vhost/Makefile | 8 +-=0A= > lib/librte_vhost/rte_virtio_net.h | 5 +-=0A= > lib/librte_vhost/vhost-net-cdev.c | 389 --------------------= =0A= > lib/librte_vhost/vhost-net-cdev.h | 113 ------=0A= > lib/librte_vhost/vhost-net.h | 118 +++++++=0A= > lib/librte_vhost/vhost_cuse/eventfd_copy.c | 88 +++++=0A= > lib/librte_vhost/vhost_cuse/eventfd_copy.h | 39 ++=0A= > lib/librte_vhost/vhost_cuse/vhost-net-cdev.c | 417 ++++++++++++++++++++= ++=0A= > lib/librte_vhost/vhost_cuse/virtio-net-cdev.c | 423 ++++++++++++++++++++= ++=0A= > lib/librte_vhost/vhost_cuse/virtio-net-cdev.h | 48 +++=0A= > lib/librte_vhost/vhost_rxtx.c | 2 +-=0A= > lib/librte_vhost/vhost_user/fd_man.c | 258 ++++++++++++++=0A= > lib/librte_vhost/vhost_user/fd_man.h | 67 ++++=0A= > lib/librte_vhost/vhost_user/vhost-net-user.c | 472 ++++++++++++++++++++= +++++=0A= > lib/librte_vhost/vhost_user/vhost-net-user.h | 106 ++++++=0A= > lib/librte_vhost/vhost_user/virtio-net-user.c | 314 ++++++++++++++++=0A= > lib/librte_vhost/vhost_user/virtio-net-user.h | 49 +++=0A= > lib/librte_vhost/virtio-net.c | 491 ++------------------= ------=0A= > lib/librte_vhost/virtio-net.h | 43 +++=0A= > 19 files changed, 2491 insertions(+), 959 deletions(-)=0A= > delete mode 100644 lib/librte_vhost/vhost-net-cdev.c=0A= > delete mode 100644 lib/librte_vhost/vhost-net-cdev.h=0A= > create mode 100644 lib/librte_vhost/vhost-net.h=0A= > create mode 100644 lib/librte_vhost/vhost_cuse/eventfd_copy.c=0A= > create mode 100644 lib/librte_vhost/vhost_cuse/eventfd_copy.h=0A= > create mode 100644 lib/librte_vhost/vhost_cuse/vhost-net-cdev.c=0A= > create mode 100644 lib/librte_vhost/vhost_cuse/virtio-net-cdev.c=0A= > create mode 100644 lib/librte_vhost/vhost_cuse/virtio-net-cdev.h=0A= > create mode 100644 lib/librte_vhost/vhost_user/fd_man.c=0A= > create mode 100644 lib/librte_vhost/vhost_user/fd_man.h=0A= > create mode 100644 lib/librte_vhost/vhost_user/vhost-net-user.c=0A= > create mode 100644 lib/librte_vhost/vhost_user/vhost-net-user.h=0A= > create mode 100644 lib/librte_vhost/vhost_user/virtio-net-user.c=0A= > create mode 100644 lib/librte_vhost/vhost_user/virtio-net-user.h=0A= > create mode 100644 lib/librte_vhost/virtio-net.h=0A= >=0A= =0A=