From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:55242) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1d8nHU-0002Zl-Id for qemu-devel@nongnu.org; Thu, 11 May 2017 08:33:22 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1d8nHR-0004k7-GA for qemu-devel@nongnu.org; Thu, 11 May 2017 08:33:20 -0400 Received: from mx1.redhat.com ([209.132.183.28]:51038) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1d8nHR-0004jy-Av for qemu-devel@nongnu.org; Thu, 11 May 2017 08:33:17 -0400 From: Maxime Coquelin Date: Thu, 11 May 2017 14:32:40 +0200 Message-Id: <20170511123246.31308-1-maxime.coquelin@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Subject: [Qemu-devel] [PATCH 0/6] vhost-user: Specify and implement device IOTLB support List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: mst@redhat.com, peterx@redhat.com, marcandre.lureau@gmail.com, vkaplans@redhat.com, jasowang@redhat.com, wexu@redhat.com, yuanhan.liu@linux.intel.com, qemu-devel@nongnu.org, jfreiman@redhat.com Cc: Maxime Coquelin This series aims at specifying ans implementing the protocol update required to support device IOTLB with user backends. In this first non-RFC version, IOTLB messages forging and parsing have be= en extracted from the backend implementation; so that both user and kernel backends share as most code as possible, only the transport remains backe= nd specifics. Also various smaller fixes have been implementated, taking int= o account Peter's review comments. I tagged this third version as non-RFC, since there is now a DPDK vhost-user backend prototype supporting IOMMU[0= ], making possible testing of this series. The slave requests channel part is re-used from Marc-Andr=C3=A9's series = submitted last year[1], with main changes from original version being request/featu= re names renaming and addition of the REPLY_ACK feature support. Regarding IOTLB protocol, one noticeable change is the IOTLB miss request reply made optionnal (i.e. only if slave requests it by setting the VHOST_USER_NEED_REPLY flag in the message header). This change provides more flexibility in the backend implementation of the feature. The protocol is very close to kernel backends, except that a new communication channel is introduced to enable the slave to send requests to the master. [0]: https://gitlab.com/mcoquelin/dpdk-next-virtio/commits/vhost_iotlb_pr= oto [1]: https://lists.gnu.org/archive/html/qemu-devel/2016-04/msg00095.html Marc-Andr=C3=A9 Lureau (2): vhost-user: add vhost_user to hold the chr vhost-user: add slave-req-fd support Maxime Coquelin (4): vhost: propagate errors in vhost_device_iotlb_miss() vhost: rework IOTLB messaging vhost: Update rings information for IOTLB earlier spec/vhost-user spec: Add IOMMU support docs/specs/vhost-user.txt | 107 ++++++++++++++++++++++- hw/virtio/vhost-backend.c | 135 ++++++++++++++++------------- hw/virtio/vhost-user.c | 177 ++++++++++++++++++++++++++++++++= +++++- hw/virtio/vhost.c | 38 ++++---- include/hw/virtio/vhost-backend.h | 23 +++-- include/hw/virtio/vhost.h | 2 +- 6 files changed, 393 insertions(+), 89 deletions(-) --=20 2.9.3