On 05/15/2017 04:18 PM, Jason Wang wrote:
On 2017年05月15日 16:03, Zhang Chen wrote:
On 05/15/2017 12:05 PM, Jason Wang wrote:
On 2017年05月12日 09:41, Zhang Chen wrote:
We can use this property flush and send packet with vnet_hdr_len.
Signed-off-by: Zhang Chen <zhangchen.f...@cn.fujitsu.com>
Then I think it's not necessary to store vnet_hdr_len in
SocketReadState?
Do you means we keep the patch 05/12 in original?
Thanks
Zhang Chen
I mean we could fetch vnet_hdr_len from the buf directly. Or is there
any advantage to store it in SocketReadState?
No, The rs->buf did't have the vnet_hdr_len field.
In the net_fill_rstate(), when case = 2, we override the rs->buf by real
net packet data,
So, the vnet_hdr_len field and packet_len field didn't be include in the
last rs->buf.
Thanks
Zhang Chen
Thanks
--
Thanks
Zhang Chen
