From: Wei Wang <wei.w.wang@intel.com>
To: marcandre.lureau@gmail.com, mst@redhat.com, stefanha@redhat.com,
pbonzini@redhat.com, qemu-devel@nongnu.org,
virtio-dev@lists.oasis-open.org
Cc: Wei Wang <wei.w.wang@intel.com>
Subject: [Qemu-devel] [PATCH v1 18/37] vhost-user: send guest physical address of virtqueues to the slave
Date: Sat, 17 Dec 2016 18:43:28 +0800 [thread overview]
Message-ID: <1481971427-11094-19-git-send-email-wei.w.wang@intel.com> (raw)
In-Reply-To: <1481971427-11094-1-git-send-email-wei.w.wang@intel.com>
In the vhost-pci case, the slave needs the master side guest physical
address, rather than the qemu virtual address.
Signed-off-by: Wei Wang <wei.w.wang@intel.com>
---
hw/virtio/vhost.c | 61 ++++++++++++++++++++++++++++++++---------------
include/hw/virtio/vhost.h | 2 ++
2 files changed, 44 insertions(+), 19 deletions(-)
diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index f7f7023..df04375 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -26,6 +26,7 @@
#include "hw/virtio/virtio-bus.h"
#include "hw/virtio/virtio-access.h"
#include "migration/migration.h"
+#include "hw/virtio/vhost-user.h"
/* enabled until disconnected backend stabilizes */
#define _VHOST_DEBUG 1
@@ -858,6 +859,12 @@ static int vhost_virtqueue_set_vring_endian_legacy(struct vhost_dev *dev,
return -errno;
}
+bool vhost_pci_enabled(struct vhost_dev *dev)
+{
+ return ((dev->protocol_features &
+ (1ULL << VHOST_USER_PROTOCOL_F_VHOST_PCI)) != 0);
+}
+
static int vhost_virtqueue_start(struct vhost_dev *dev,
struct VirtIODevice *vdev,
struct vhost_virtqueue *vq,
@@ -901,26 +908,40 @@ static int vhost_virtqueue_start(struct vhost_dev *dev,
}
}
- vq->desc_size = s = l = virtio_queue_get_desc_size(vdev, idx);
vq->desc_phys = a = virtio_queue_get_desc_addr(vdev, idx);
- vq->desc = cpu_physical_memory_map(a, &l, 0);
- if (!vq->desc || l != s) {
- r = -ENOMEM;
- goto fail_alloc_desc;
+ if (vhost_pci_enabled(dev)) {
+ vq->desc = (void *)a;
+ } else {
+ vq->desc_size = s = l = virtio_queue_get_desc_size(vdev, idx);
+ vq->desc = cpu_physical_memory_map(a, &l, 0);
+ if (!vq->desc || l != s) {
+ r = -ENOMEM;
+ goto fail_alloc_desc;
+ }
}
+
vq->avail_size = s = l = virtio_queue_get_avail_size(vdev, idx);
vq->avail_phys = a = virtio_queue_get_avail_addr(vdev, idx);
- vq->avail = cpu_physical_memory_map(a, &l, 0);
- if (!vq->avail || l != s) {
- r = -ENOMEM;
- goto fail_alloc_avail;
+ if (vhost_pci_enabled(dev)) {
+ vq->avail = (void *)a;
+ } else {
+ vq->avail = cpu_physical_memory_map(a, &l, 0);
+ if (!vq->avail || l != s) {
+ r = -ENOMEM;
+ goto fail_alloc_avail;
+ }
}
+
vq->used_size = s = l = virtio_queue_get_used_size(vdev, idx);
vq->used_phys = a = virtio_queue_get_used_addr(vdev, idx);
- vq->used = cpu_physical_memory_map(a, &l, 1);
- if (!vq->used || l != s) {
- r = -ENOMEM;
- goto fail_alloc_used;
+ if (vhost_pci_enabled(dev)) {
+ vq->used = (void *)a;
+ } else {
+ vq->used = cpu_physical_memory_map(a, &l, 1);
+ if (!vq->used || l != s) {
+ r = -ENOMEM;
+ goto fail_alloc_used;
+ }
}
r = vhost_virtqueue_set_addr(dev, vq, vhost_vq_index, dev->log_enabled);
@@ -1003,12 +1024,14 @@ static void vhost_virtqueue_stop(struct vhost_dev *dev,
vhost_vq_index);
}
- cpu_physical_memory_unmap(vq->used, virtio_queue_get_used_size(vdev, idx),
- 1, virtio_queue_get_used_size(vdev, idx));
- cpu_physical_memory_unmap(vq->avail, virtio_queue_get_avail_size(vdev, idx),
- 0, virtio_queue_get_avail_size(vdev, idx));
- cpu_physical_memory_unmap(vq->desc, virtio_queue_get_desc_size(vdev, idx),
- 0, virtio_queue_get_desc_size(vdev, idx));
+ if (!vhost_pci_enabled(dev)) {
+ cpu_physical_memory_unmap(vq->used, virtio_queue_get_used_size(vdev, idx),
+ 1, virtio_queue_get_used_size(vdev, idx));
+ cpu_physical_memory_unmap(vq->avail, virtio_queue_get_avail_size(vdev, idx),
+ 0, virtio_queue_get_avail_size(vdev, idx));
+ cpu_physical_memory_unmap(vq->desc, virtio_queue_get_desc_size(vdev, idx),
+ 0, virtio_queue_get_desc_size(vdev, idx));
+ }
}
static void vhost_eventfd_add(MemoryListener *listener,
diff --git a/include/hw/virtio/vhost.h b/include/hw/virtio/vhost.h
index 9cf32e2..9955e07 100644
--- a/include/hw/virtio/vhost.h
+++ b/include/hw/virtio/vhost.h
@@ -92,4 +92,6 @@ bool vhost_has_free_slot(void);
int vhost_net_set_backend(struct vhost_dev *hdev,
struct vhost_vring_file *file);
+bool vhost_pci_enabled(struct vhost_dev *dev);
+
#endif
--
2.7.4
next prev parent reply other threads:[~2016-12-17 10:45 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-12-17 10:43 [Qemu-devel] [PATCH v1 00/37] Implementation of vhost-pci for inter-vm commucation Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 01/37] vhost-pci-net: the fundamental vhost-pci-net device emulation Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 02/37] vhost-pci-net: the fundamental implementation of vhost-pci-net-pci Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 03/37] vhost-user: share the vhost-user protocol related structures Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 04/37] vl: add the vhost-pci-slave command line option Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 05/37] vhost-pci-slave: start the implementation of vhost-pci-slave Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 06/37] vhost-pci-slave: set up the fundamental handlers for the server socket Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 07/37] vhost-pci-slave/msg: VHOST_USER_GET_FEATURES Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 08/37] vhost-pci-slave/msg: VHOST_USER_SET_FEATURES Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 09/37] vhost-pci-slave/msg: VHOST_USER_GET_PROTOCOL_FEATURES Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 10/37] vhost-pci-slave/msg: VHOST_USER_SET_PROTOCOL_FEATURES Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 11/37] vhost-user/msg: VHOST_USER_PROTOCOL_F_SET_DEVICE_ID Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 12/37] vhost-pci-slave/msg: VHOST_USER_SET_DEVICE_ID Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 13/37] vhost-pci-slave/msg: VHOST_USER_GET_QUEUE_NUM Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 14/37] vhost-pci-slave/msg: VHOST_USER_SET_OWNER Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 15/37] vhost-pci-slave/msg: VHOST_USER_SET_MEM_TABLE Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 16/37] vhost-pci-slave/msg: VHOST_USER_SET_VRING_NUM Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 17/37] vhost-pci-slave/msg: VHOST_USER_SET_VRING_BASE Wei Wang
2016-12-17 10:43 ` Wei Wang [this message]
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 19/37] vhost-pci-slave/msg: VHOST_USER_SET_VRING_ADDR Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 20/37] vhost-pci-slave/msg: VHOST_USER_SET_VRING_KICK Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 21/37] vhost-pci-slave/msg: VHOST_USER_SET_VRING_CALL Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 22/37] vhost-pci-slave/msg: VHOST_USER_SET_VRING_ENABLE Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 23/37] vhost-pci-slave/msg: VHOST_USER_SET_LOG_BASE Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 24/37] vhost-pci-slave/msg: VHOST_USER_SET_LOG_FD Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 25/37] vhost-pci-slave/msg: VHOST_USER_SEND_RARP Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 26/37] vhost-pci-slave/msg: VHOST_USER_GET_VRING_BASE Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 27/37] vhost-pci-net: pass the info collected by vp_slave to the device Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 28/37] vhost-pci-net: pass the mem and vring info to the driver Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 29/37] vhost-pci-slave/msg: VHOST_USER_SET_VHOST_PCI (start) Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 30/37] vhost-pci-slave/msg: VHOST_USER_SET_VHOST_PCI (stop) Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 31/37] vhost-user/msg: send VHOST_USER_SET_VHOST_PCI (start/stop) Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 32/37] vhost-user: add asynchronous read for the vhost-user master Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 33/37] vhost-pci-net: send the negotiated feature bits to the master Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 34/37] vhost-pci-slave: add "peer_reset" Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 35/37] vhost-pci-net: start the vhost-pci-net device Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 36/37] vhost-user/msg: handling VHOST_USER_SET_FEATURES Wei Wang
2016-12-17 10:43 ` [Qemu-devel] [PATCH v1 37/37] vl: enable vhost-pci-slave Wei Wang
2016-12-17 11:38 ` [Qemu-devel] [PATCH v1 00/37] Implementation of vhost-pci for inter-vm commucation no-reply
2016-12-17 11:55 ` no-reply
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1481971427-11094-19-git-send-email-wei.w.wang@intel.com \
--to=wei.w.wang@intel.com \
--cc=marcandre.lureau@gmail.com \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
--cc=virtio-dev@lists.oasis-open.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).