bwlp/qemu.git/docs/specs/vhost-user.txt, branch spice_video_codecs

docs: create interop/ subdirectory

2017-06-15T09:18:39+00:00

This is for the future interoperability & management guide. It includes the QAPI docs, including the automatically generated ones, other socket protocols (vhost-user, VNC), and the qcow2 file format. Signed-off-by: Paolo Bonzini

spec/vhost-user spec: Add IOMMU support

2017-06-02T15:57:17+00:00

This patch specifies and implements the master/slave communication to support device IOTLB in slave. The vhost_iotlb_msg structure introduced for kernel backends is re-used, making the design close between the two backends. An exception is the use of the secondary channel to enable the slave to send IOTLB miss requests to the master. Signed-off-by: Maxime Coquelin Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin

vhost-user: add slave-req-fd support

2017-06-02T15:57:17+00:00

Learn to give a socket to the slave to let him make requests to the master. Signed-off-by: Marc-André Lureau Signed-off-by: Maxime Coquelin Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin

vhost-user: Add MTU protocol feature and op

2017-01-10T05:02:53+00:00

This patch implements VHOST_USER_PROTOCOL_F_NET_MTU protocol feature and VHOST_USER_NET_SET_MTU request so that the backend gets notified of the user defined host MTU. If backend supports VHOST_USER_PROTOCOL_F_REPLY_ACK, QEMU assumes MTU is valid if success is returned. Vhost-net driver sends this request through a new vhost_net_set_mtu vhost_ops entry. Cc: Michael S. Tsirkin Cc: Aaron Conole Signed-off-by: Maxime Coquelin Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin

spec/vhost-user: fix the VHOST_USER prefix

2016-11-30T02:20:57+00:00

Signed-off-by: Wei Wang Reviewed-by: Marc-André Lureau Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin

vhost-user: Introduce a new protocol feature REPLY_ACK.

2016-08-10T14:47:29+00:00

This introduces the VHOST_USER_PROTOCOL_F_REPLY_ACK. If negotiated, client applications should send a u64 payload in response to any message that contains the "need_reply" bit set on the message flags. Setting the payload to "zero" indicates the command finished successfully. Likewise, setting it to "non-zero" indicates an error. Currently implemented only for SET_MEM_TABLE. Reviewed-by: Marc-André Lureau Signed-off-by: Prerna Saxena Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin

specs/vhost-user: spelling fix

2016-04-13T16:52:34+00:00

Signed-off-by: Marc-André Lureau Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin

specs/vhost-user: improve VHOST_SET_VRING_NUM documentation

2016-04-13T16:52:34+00:00

"number of vrings" doesn't help me understand the purpose of this message. My understanding is that it is rather the size of the queue (in modern terms). Signed-off-by: Marc-André Lureau Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin

vhost-user: clarify start and enable

2015-11-25T11:42:38+00:00

It seems that we currently have some duplication between started and enabled states. The actual reason is that enable is not documented correctly: what it does is connecting ring to the backend. This is important for MQ, because a Linux guest expects TX packets to be completed even if it disables some queues temporarily. Cc: Yuanhan Liu Cc: Victor Kaplansky Signed-off-by: Michael S. Tsirkin

specs/vhost-user: fix spec to match reality

2015-11-17T13:41:13+00:00

We wanted to start/stop rings on VRING_ENABLE, but that is not what QEMU does. Rather than tweaking code some more, with risk to stability, let's just document it as it is. We'll be able to fix this in the future with a new protocol feature bit. Reported-by: Victor Kaplansky Signed-off-by: Michael S. Tsirkin