qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Jiang Wang <jiang.wang@bytedance.com>
To: mst@redhat.com, qemu-devel@nongnu.org
Cc: jasowang@redhat.com, arseny.krasnov@kaspersky.com,
	stefanha@redhat.com, sgarzare@redhat.com
Subject: [RFC v5] virtio/vsock: add two more queues for datagram types
Date: Sun, 12 Sep 2021 18:46:03 +0000	[thread overview]
Message-ID: <20210912184603.311817-1-jiang.wang@bytedance.com> (raw)

Datagram sockets are connectionless and unreliable.
The sender does not know the capacity of the receiver
and may send more packets than the receiver can handle.

Add two more dedicate virtqueues for datagram sockets,
so that it will not unfairly steal resources from
stream and future connection-oriented sockets.

Signed-off-by: Jiang Wang <jiang.wang@bytedance.com>
---
v1 -> v2: use qemu cmd option to control number of queues,
        removed configuration settings for dgram.
v2 -> v3: use ioctl to get features and decide number of
        virt queues, instead of qemu cmd option.
v3 -> v4: change DGRAM feature bit value to 2. Add an argument
        in vhost_vsock_common_realize to indicate dgram is supported or not.
v4 -> v5: don't open dev to get vhostfd. Removed leftover definition of
        enable_dgram

 hw/virtio/vhost-user-vsock.c                  |  2 +-
 hw/virtio/vhost-vsock-common.c                | 27 ++++++++++++++++---
 hw/virtio/vhost-vsock.c                       | 27 ++++++++++++++++++-
 include/hw/virtio/vhost-vsock-common.h        |  6 +++--
 include/hw/virtio/vhost-vsock.h               |  3 +++
 include/standard-headers/linux/virtio_vsock.h |  1 +
 6 files changed, 59 insertions(+), 7 deletions(-)

diff --git a/hw/virtio/vhost-user-vsock.c b/hw/virtio/vhost-user-vsock.c
index 6095ed7349..e9ec0e1c00 100644
--- a/hw/virtio/vhost-user-vsock.c
+++ b/hw/virtio/vhost-user-vsock.c
@@ -105,7 +105,7 @@ static void vuv_device_realize(DeviceState *dev, Error **errp)
         return;
     }
 
-    vhost_vsock_common_realize(vdev, "vhost-user-vsock");
+    vhost_vsock_common_realize(vdev, "vhost-user-vsock", false);
 
     vhost_dev_set_config_notifier(&vvc->vhost_dev, &vsock_ops);
 
diff --git a/hw/virtio/vhost-vsock-common.c b/hw/virtio/vhost-vsock-common.c
index 4ad6e234ad..f48b5a69df 100644
--- a/hw/virtio/vhost-vsock-common.c
+++ b/hw/virtio/vhost-vsock-common.c
@@ -17,6 +17,8 @@
 #include "hw/virtio/vhost-vsock.h"
 #include "qemu/iov.h"
 #include "monitor/monitor.h"
+#include <sys/ioctl.h>
+#include <linux/vhost.h>
 
 int vhost_vsock_common_start(VirtIODevice *vdev)
 {
@@ -196,9 +198,10 @@ int vhost_vsock_common_post_load(void *opaque, int version_id)
     return 0;
 }
 
-void vhost_vsock_common_realize(VirtIODevice *vdev, const char *name)
+void vhost_vsock_common_realize(VirtIODevice *vdev, const char *name, bool enable_dgram)
 {
     VHostVSockCommon *vvc = VHOST_VSOCK_COMMON(vdev);
+    int nvqs = MAX_VQS_WITHOUT_DGRAM;
 
     virtio_init(vdev, name, VIRTIO_ID_VSOCK,
                 sizeof(struct virtio_vsock_config));
@@ -209,12 +212,22 @@ void vhost_vsock_common_realize(VirtIODevice *vdev, const char *name)
     vvc->trans_vq = virtio_add_queue(vdev, VHOST_VSOCK_QUEUE_SIZE,
                                        vhost_vsock_common_handle_output);
 
+    if (!enable_dgram)
+        nvqs = MAX_VQS_WITHOUT_DGRAM;
+    else {
+        nvqs = MAX_VQS_WITH_DGRAM;
+        vvc->dgram_recv_vq = virtio_add_queue(vdev, VHOST_VSOCK_QUEUE_SIZE,
+                                              vhost_vsock_common_handle_output);
+        vvc->dgram_trans_vq = virtio_add_queue(vdev, VHOST_VSOCK_QUEUE_SIZE,
+                                               vhost_vsock_common_handle_output);
+    }
+
     /* The event queue belongs to QEMU */
     vvc->event_vq = virtio_add_queue(vdev, VHOST_VSOCK_QUEUE_SIZE,
                                        vhost_vsock_common_handle_output);
 
-    vvc->vhost_dev.nvqs = ARRAY_SIZE(vvc->vhost_vqs);
-    vvc->vhost_dev.vqs = vvc->vhost_vqs;
+    vvc->vhost_dev.nvqs = nvqs;
+    vvc->vhost_dev.vqs = g_new0(struct vhost_virtqueue, vvc->vhost_dev.nvqs);
 
     vvc->post_load_timer = NULL;
 }
@@ -227,6 +240,14 @@ void vhost_vsock_common_unrealize(VirtIODevice *vdev)
 
     virtio_delete_queue(vvc->recv_vq);
     virtio_delete_queue(vvc->trans_vq);
+    if (vvc->vhost_dev.nvqs == MAX_VQS_WITH_DGRAM) {
+        virtio_delete_queue(vvc->dgram_recv_vq);
+        virtio_delete_queue(vvc->dgram_trans_vq);
+    }
+
+    if (vvc->vhost_dev.vqs)
+        g_free(vvc->vhost_dev.vqs);
+
     virtio_delete_queue(vvc->event_vq);
     virtio_cleanup(vdev);
 }
diff --git a/hw/virtio/vhost-vsock.c b/hw/virtio/vhost-vsock.c
index 1b1a5c70ed..1fee25f144 100644
--- a/hw/virtio/vhost-vsock.c
+++ b/hw/virtio/vhost-vsock.c
@@ -20,9 +20,12 @@
 #include "hw/qdev-properties.h"
 #include "hw/virtio/vhost-vsock.h"
 #include "monitor/monitor.h"
+#include <sys/ioctl.h>
+#include <linux/vhost.h>
 
 const int feature_bits[] = {
     VIRTIO_VSOCK_F_SEQPACKET,
+    VIRTIO_VSOCK_F_DGRAM,
     VHOST_INVALID_FEATURE_BIT
 };
 
@@ -116,6 +119,8 @@ static uint64_t vhost_vsock_get_features(VirtIODevice *vdev,
     VHostVSockCommon *vvc = VHOST_VSOCK_COMMON(vdev);
 
     virtio_add_feature(&requested_features, VIRTIO_VSOCK_F_SEQPACKET);
+    if (vvc->vhost_dev.nvqs == MAX_VQS_WITH_DGRAM)
+        virtio_add_feature(&requested_features, VIRTIO_VSOCK_F_DGRAM);
     return vhost_get_features(&vvc->vhost_dev, feature_bits,
                                 requested_features);
 }
@@ -132,6 +137,24 @@ static const VMStateDescription vmstate_virtio_vhost_vsock = {
     .post_load = vhost_vsock_common_post_load,
 };
 
+static bool vhost_vsock_dgram_supported(int vhostfd)
+{
+    uint64_t features;
+    int ret;
+
+    ret = ioctl(vhostfd, VHOST_GET_FEATURES, &features);
+    if (ret) {
+        error_report("vhost-vsock: failed to read  device. %s", strerror(errno));
+        qemu_close(vhostfd);
+        return ret;
+    }
+
+    if (features & (1 << VIRTIO_VSOCK_F_DGRAM))
+        return true;
+
+    return false;
+}
+
 static void vhost_vsock_device_realize(DeviceState *dev, Error **errp)
 {
     VHostVSockCommon *vvc = VHOST_VSOCK_COMMON(dev);
@@ -139,6 +162,7 @@ static void vhost_vsock_device_realize(DeviceState *dev, Error **errp)
     VHostVSock *vsock = VHOST_VSOCK(dev);
     int vhostfd;
     int ret;
+    bool enable_dgram;
 
     /* Refuse to use reserved CID numbers */
     if (vsock->conf.guest_cid <= 2) {
@@ -175,7 +199,8 @@ static void vhost_vsock_device_realize(DeviceState *dev, Error **errp)
         qemu_set_nonblock(vhostfd);
     }
 
-    vhost_vsock_common_realize(vdev, "vhost-vsock");
+    enable_dgram = vhost_vsock_dgram_supported(vhostfd);
+    vhost_vsock_common_realize(vdev, "vhost-vsock", enable_dgram);
 
     ret = vhost_dev_init(&vvc->vhost_dev, (void *)(uintptr_t)vhostfd,
                          VHOST_BACKEND_TYPE_KERNEL, 0, errp);
diff --git a/include/hw/virtio/vhost-vsock-common.h b/include/hw/virtio/vhost-vsock-common.h
index e412b5ee98..6669d24714 100644
--- a/include/hw/virtio/vhost-vsock-common.h
+++ b/include/hw/virtio/vhost-vsock-common.h
@@ -27,12 +27,13 @@ enum {
 struct VHostVSockCommon {
     VirtIODevice parent;
 
-    struct vhost_virtqueue vhost_vqs[2];
     struct vhost_dev vhost_dev;
 
     VirtQueue *event_vq;
     VirtQueue *recv_vq;
     VirtQueue *trans_vq;
+    VirtQueue *dgram_recv_vq;
+    VirtQueue *dgram_trans_vq;
 
     QEMUTimer *post_load_timer;
 };
@@ -41,7 +42,8 @@ int vhost_vsock_common_start(VirtIODevice *vdev);
 void vhost_vsock_common_stop(VirtIODevice *vdev);
 int vhost_vsock_common_pre_save(void *opaque);
 int vhost_vsock_common_post_load(void *opaque, int version_id);
-void vhost_vsock_common_realize(VirtIODevice *vdev, const char *name);
+void vhost_vsock_common_realize(VirtIODevice *vdev, const char *name,
+			       bool enable_dgram);
 void vhost_vsock_common_unrealize(VirtIODevice *vdev);
 
 #endif /* _QEMU_VHOST_VSOCK_COMMON_H */
diff --git a/include/hw/virtio/vhost-vsock.h b/include/hw/virtio/vhost-vsock.h
index 84f4e727c7..7d16c0e218 100644
--- a/include/hw/virtio/vhost-vsock.h
+++ b/include/hw/virtio/vhost-vsock.h
@@ -33,4 +33,7 @@ struct VHostVSock {
     /*< public >*/
 };
 
+#define MAX_VQS_WITHOUT_DGRAM 2
+#define MAX_VQS_WITH_DGRAM 4
+
 #endif /* QEMU_VHOST_VSOCK_H */
diff --git a/include/standard-headers/linux/virtio_vsock.h b/include/standard-headers/linux/virtio_vsock.h
index 3a23488e42..7e35acf3d4 100644
--- a/include/standard-headers/linux/virtio_vsock.h
+++ b/include/standard-headers/linux/virtio_vsock.h
@@ -40,6 +40,7 @@
 
 /* The feature bitmap for virtio vsock */
 #define VIRTIO_VSOCK_F_SEQPACKET	1	/* SOCK_SEQPACKET supported */
+#define VIRTIO_VSOCK_F_DGRAM		2	/* SOCK_DGRAM supported */
 
 struct virtio_vsock_config {
 	uint64_t guest_cid;
-- 
2.20.1



             reply	other threads:[~2021-09-12 18:47 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-12 18:46 Jiang Wang [this message]
2021-09-13 13:28 ` [RFC v5] virtio/vsock: add two more queues for datagram types Stefano Garzarella

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=20210912184603.311817-1-jiang.wang@bytedance.com \
    --to=jiang.wang@bytedance.com \
    --cc=arseny.krasnov@kaspersky.com \
    --cc=jasowang@redhat.com \
    --cc=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=sgarzare@redhat.com \
    --cc=stefanha@redhat.com \
    /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).