From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:48231) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZebVI-0005kD-1u for qemu-devel@nongnu.org; Wed, 23 Sep 2015 00:18:01 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ZebVE-0000vM-T0 for qemu-devel@nongnu.org; Wed, 23 Sep 2015 00:18:00 -0400 Received: from mga11.intel.com ([192.55.52.93]:57780) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZebVE-0000vD-Nl for qemu-devel@nongnu.org; Wed, 23 Sep 2015 00:17:56 -0400 From: Yuanhan Liu Date: Wed, 23 Sep 2015 12:19:54 +0800 Message-Id: <1442982001-10669-1-git-send-email-yuanhan.liu@linux.intel.com> Subject: [Qemu-devel] [PATCH v11 0/7] vhost-user multiple queue support List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: Yuanhan Liu , jasowang@redhat.com, Changchun.ouyang@hotmail.com, mst@redhat.com Hi, Here is the updated patch set for enabling vhost-user multiple queue. And I did proper and formal testing this time. This patch set introduces 2 more vhost user messages: VHOST_USER_GET_QUEUE_NUM, for querying how many queues the backend supports, and VHOST_USER_SET_VRING_ENABLE, for enabling/disabling a specific virt queue. Both of the two new messages are treated as vhost protocol extension, and that's why Michaels's patch "vhost-user: add protocol feature negotiation" is also included here. Patch 1-5 are all prepare works for actually enabling multiple queue. Patch 6 is the major patch for enabling multiple queue, which also tries to address two major concerns from Michael: no feedback from backend if it can't support # of requested queues, and all messages are sent N time. It also fixes a hidden bug. Patch 7 introduces the VHOST_USER_SET_VRING_ENABLE message, to enable or disable a specific vring. v11: - typo fixes pointed out by Eric - define a dummy vhost_net_get_max_queues() when !CONFIG_VHOST_NET. - Per suggested by Jason Wang, invoke qmp_set_link() directly to remove duplicate code. v10: - typo fixes pointed out by Eric - [PATCH 6]: don't treat VHOST_USER_SET/GET_[PROTOCOL]_FEATURES as one time request, as the two feature bits need to be stored at per-device. v9: - Per suggested by Jason Wang, patch 5 introduces a new vhost backend method: vhost_backend_get_vq_index(). - Use qemu_find_net_clients_except() at net/vhost-user.c for gathering all related ncs so that we could register chr dev event handler once. Which is also suggested by Jason Wang. Thanks. --yliu --- Changchun Ouyang (2): vhost-user: add multiple queue support vhost-user: add a new message to disable/enable a specific virt queue. Michael S. Tsirkin (1): vhost-user: add protocol feature negotiation Yuanhan Liu (4): vhost-user: use VHOST_USER_XXX macro for switch statement vhost: rename VHOST_RESET_OWNER to VHOST_RESET_DEVICE vhost-user: add VHOST_USER_GET_QUEUE_NUM message vhost: introduce vhost_backend_get_vq_index method docs/specs/vhost-user.txt | 77 ++++++++++++++++++++- hw/net/vhost_net.c | 44 ++++++++++-- hw/net/virtio-net.c | 8 +++ hw/virtio/vhost-backend.c | 10 ++- hw/virtio/vhost-user.c | 139 +++++++++++++++++++++++++++++++------ hw/virtio/vhost.c | 20 +++--- include/hw/virtio/vhost-backend.h | 4 ++ include/hw/virtio/vhost.h | 2 + include/net/vhost_net.h | 3 + linux-headers/linux/vhost.h | 2 +- net/vhost-user.c | 141 +++++++++++++++++++++++++------------- qapi-schema.json | 6 +- qemu-options.hx | 5 +- tests/vhost-user-test.c | 2 +- 14 files changed, 371 insertions(+), 92 deletions(-) -- 1.9.0