On Mon, Jan 13, 2020 at 1:08 PM Jason Wang [off-list ref] wrote:

On 2020/1/13 下午4:17, Yuri Benditovich wrote:

quoted

VIRTIO_NET_F_RSC_EXT feature bit indicates that the device
is able to provide extended RSC information. When the feature
is negotiatede and 'gso_type' field in received packet is not
GSO_NONE, the device reports number of coalesced packets in
'csum_start' field and number of duplicated acks in 'csum_offset'
field and sets VIRTIO_NET_HDR_F_RSC_INFO in 'flags' field.

Signed-off-by: Yuri Benditovich <redacted>

Hi Yuri:

Is the feature used by Linux? If yes, it's better to include the real user.

It is not used by Linux. Mainly needed for certification under Windows.

quoted

---
  include/uapi/linux/virtio_net.h | 10 +++++++++-
  1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/include/uapi/linux/virtio_net.h b/include/uapi/linux/virtio_net.h
index a3715a3224c1..2bdd26f8a4ed 100644
--- a/include/uapi/linux/virtio_net.h
+++ b/include/uapi/linux/virtio_net.h

@@ -56,7 +56,7 @@
  #define VIRTIO_NET_F_MQ     22      /* Device supports Receive Flow
                                       * Steering */
  #define VIRTIO_NET_F_CTRL_MAC_ADDR 23       /* Set MAC address */
-
+#define VIRTIO_NET_F_RSC_EXT   61    /* Provides extended RSC info */

Is this better to keep the newline around?

No problem, let's wait until the rest is accepted.

quoted

  #define VIRTIO_NET_F_STANDBY          62    /* Act as standby for another device
                                       * with the same MAC.
                                       */

@@ -104,6 +104,7 @@ struct virtio_net_config {
  struct virtio_net_hdr_v1 {
  #define VIRTIO_NET_HDR_F_NEEDS_CSUM 1       /* Use csum_start, csum_offset */
  #define VIRTIO_NET_HDR_F_DATA_VALID 2       /* Csum is valid */
+#define VIRTIO_NET_HDR_F_RSC_INFO    4       /* rsc_ext data in csum_ fields */
      __u8 flags;
  #define VIRTIO_NET_HDR_GSO_NONE             0       /* Not a GSO frame */
  #define VIRTIO_NET_HDR_GSO_TCPV4    1       /* GSO frame, IPv4 TCP (TSO) */

@@ -118,6 +119,13 @@ struct virtio_net_hdr_v1 {
      __virtio16 num_buffers; /* Number of merged rx buffers */
  };

+/*
+ * if VIRTIO_NET_F_RSC_EXT feature has been negotiated and
+ * VIRTIO_NET_HDR_F_RSC_INFO is set in RX packet
+ */
+#define virtio_net_rsc_ext_num_packets       csum_start
+#define virtio_net_rsc_ext_num_dupacks       csum_offset

This looks sub-optimal, it looks to me union is better?

This was discussed in v1, MST decided the define is better.

Thanks

quoted

+
  #ifndef VIRTIO_NET_NO_LEGACY
  /* This header comes first in the scatter-gather list.
   * For legacy virtio, if VIRTIO_F_ANY_LAYOUT is not negotiated, it must