OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

virtio-dev message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Re: [virtio-dev] Re: [Qemu-devel] [PATCH v3 1/2] virtio-net: enable configurable tx queue size

On 2017年06月26日 12:55, Wei Wang wrote:
On 06/26/2017 11:18 AM, Jason Wang wrote:

On 2017年06月23日 10:32, Wei Wang wrote:
This patch enables the virtio-net tx queue size to be configurable
between 256 (the default queue size) and 1024 by the user when the
vhost-user backend is used.

Currently, the maximum tx queue size for other backends is 512 due
to the following limitations:
- QEMU backend: the QEMU backend implementation in some cases may
send 1024+1 iovs to writev.
- Vhost_net backend: there are possibilities that the guest sends
a vring_desc of memory which corsses a MemoryRegion thereby
generating more than 1024 iovs after translattion from guest-physical
address in the backend.

Signed-off-by: Wei Wang <wei.w.wang@intel.com>
hw/net/virtio-net.c | 45 +++++++++++++++++++++++++++++++++---------
  include/hw/virtio/virtio-net.h |  1 +
  2 files changed, 37 insertions(+), 9 deletions(-)

diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index 91eddaf..d13ca60 100644
--- a/hw/net/virtio-net.c
+++ b/hw/net/virtio-net.c
@@ -34,8 +34,11 @@
    /* previously fixed value */
+    /*
+ * Currently, backends other than vhost-user don't support 1024 queue
+     * size.
+     */
+    if (n->net_conf.tx_queue_size == VIRTQUEUE_MAX_SIZE &&
+        nc->peer->info->type != NET_CLIENT_DRIVER_VHOST_USER) {
+        n->net_conf.tx_queue_size = VIRTIO_NET_TX_QUEUE_DEFAULT_SIZE;

Do we really want assume all vhost-user backend support 1024 queue size?

 Do you know any vhost-user backend implementation that doesn't support
1024 tx queue size?

I don't but the issue is vhost-user uditis an open protocol, there could be even closed source implementation on this. So we can't audit all kinds of backends.

+    }
+    for (i = 0; i < n->max_queues; i++) {
+        virtio_net_add_queue(n, i);
+    }

Any reason to move virtio_net_add_queue() here?

Please check the whole init steps. It was moved here (after qemu_new_nic())
to make sure nc->peer is not NULL.

I don't get here, can't you get peer just from nic_conf.peers?



To unsubscribe, e-mail: virtio-dev-unsubscribe@lists.oasis-open.org
For additional commands, e-mail: virtio-dev-help@lists.oasis-open.org

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]