Hi, all
I know NORMAL action means do "mac learning and do switching" ?
But in my cause, I want to learn mac --> do something --> do switching.
Could anybody tell me how to define "Learn MAC Address" without doing switching?
Thanks.
___
dev mailin
Where is it set? I think normally it is set in the route module, but I could
not find where.
Could you tell me?
At 2015-09-24 03:48:53, "Pravin Shelar" wrote:
>On Tue, Sep 22, 2015 at 4:45 PM, openvswitcher wrote:
>> Thank you for your reply.
>>
i, Sep 18, 2015 at 04:32:05PM +0800, openvswitcher wrote:
>> Why the vxlan offloading NICs can improve the bandwith performance?
>> Could you explain me by the source code?
>
>You have TSO and checksumming offloading on your card at least
>and very likely enabled by default.
Hi, all
I see the function skb_udp_tunnel_segment in file linux-3.14\net\ipv4\udp.c
struct sk_buff *skb_udp_tunnel_segment(struct sk_buff *skb,
netdev_features_t features)
(...skip...)
/* segment inner packet. */
enc_features = skb->dev->hw_enc_features & netif_skb_features(skb);
Hi, all
I want to know the principle of the vxlan checksum offload NIC.
I know the vxlan packet is composed of outer udp header, vxlan header and innel
l2 frame.
And I read one description of a vxlan checksum offload NIC:
VXLAN overlay performance challenge:
Hypervisor IP stack and standard
Why the vxlan offloading NICs can improve the bandwith performance?
Could you explain me by the source code?
Thanks.
At 2015-09-09 21:09:26, "YaoJun" wrote:
Try vxlan offloading firstly if you have the proper NICs.
On Wed, Sep 9, 2015 at 6:35 PM, openvswitcher wrote:
f (!swkey->eth.tci)
goto unencap;
} else
encap = NULL;
(...skip)
unencap:
if (encap)
nla_nest_end(skb, encap);
return 0;
nla_put_failure:
return -EMSGSIZE;
}
At 2015-09-17 23:43:49, "openvswitcher" wrote:
Hi, all
I want to ask a question about the fu
Hi, all
I want to ask a question about the function parse_8021q_onward.
(...skip...)
if (!is_mask) {
if (!(present_attrs & (UINT64_C(1) << OVS_KEY_ATTR_VLAN))) {
return ODP_FIT_TOO_LITTLE;
} else if (flow->vlan_tci == htons(0)) {
/* Corner case for a
Hi, all
Could you tell me when the attribute OVS_KEY_ATTR_ENCAP will be used?
If the packet is 802.1q, then OVS_KEY_ATTR_VLAN will be used to store the vlan
tci,
so what the OVS_KEY_ATTR_ENCAP is used for?
Thanks.
___
dev mailing list
dev@openvsw
Hi, all
I want to ask one question about parse_8021q_onward.
parse_8021q_onward:
(...skip...)
/* Get the VLAN TCI value. */
if (!is_mask && !(present_attrs & (UINT64_C(1) << OVS_KEY_ATTR_VLAN))) {
return ODP_FIT_TOO_LITTLE;
} else {
tci = (present_attrs & (UINT6
Thank you very much. It's helpful.
At 2015-09-16 23:42:08, "Ben Pfaff" wrote:
>On Wed, Sep 16, 2015 at 11:36:09PM +0800, openvswitcher wrote:
>> So what will happen in userspace if the patch port exists between two
>> bridges?
>> For example:
>>
So what will happen in userspace if the patch port exists between two bridges?
For example:
bridge1 patch-port1- patch-port2 bridge2
bridge1:
flowtable1: match: all action: normal;
bridge2:
flowtable1: match in-port=patch-port2 ; vlan=1; action: set_tunnel_id(vxlan
id)= 101, output: v
Hi, all
Could anybody explain the difference in function odp_flow_key_to_flow__ for if
the is_mask flag is true or flase?
Thanks.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
flow (check
>ovs-ofctl dump-flows br-tun), so packet is not flooded there.
>
>Best regards,
>Han
>
>On Tue, Sep 15, 2015 at 12:33 AM, openvswitcher wrote:
>> I want to ask one question about patch port and flooding.
>>
>>
>> The topology used is below:
>
I want to ask one question about patch port and flooding.
The topology used is below:
namespace 1(tap516ed44b-29) --br-int--br-tun
I ping an unreachable address 192.168.10.20 from 192.168.10.2(namespace port
tap516ed44b-29),so the arp flow is installed in kernel.
But the action is "push_vla
2015-09-11 11:35:07, "openvswitcher" wrote:
>Could you tell me why I am wrong?
>
>
>
>
>
>
>
>
>At 2015-09-11 11:08:29, "Ben Pfaff" wrote:
>>On Fri, Sep 11, 2015 at 09:52:19AM +0800, openvswitcher wrote:
>>> >When the datapath
Could any body tell me how the count 20 of 'up to 20 tuples' is calculated?
"
As an example, we examined the OpenFlow table installed by a production
deployment of VMware’s NVP controller [19]. This table contained 29 tuples. Of
those 29 tuples, 26 contained flows of a single priority, which m
Could you tell me why I am wrong?
At 2015-09-11 11:08:29, "Ben Pfaff" wrote:
>On Fri, Sep 11, 2015 at 09:52:19AM +0800, openvswitcher wrote:
>> >When the datapath flow is deleted, the next packet will cause a
>> >flow-miss, which will cause a corrected flow
"Ben Pfaff" wrote:
>On Thu, Sep 10, 2015 at 11:41:43AM +0800, openvswitcher wrote:
>> >If the OpenFlow table changes, the kernel flow cache has to be
>> >reexamined and sometimes updated.
>>
>> You mean the revalidate thread ? I read the source and I thoug
d if I add the
rule for '9.8.8.8' unless the flow for
'8.8.8.8' timeout which will cause new flow-miss.
Could you tell me more about it?
At 2015-09-09 23:46:12, "Ben Pfaff" wrote:
>On Wed, Sep 09, 2015 at 02:34:43PM +0800, o
.8', so I think the the 'action output:200' will never be applied
to 'nw_dst=9.8.8.8' unless the flow-cache in kernel get timeout.
The treatment is not right for 'nw_dst=9.8.8.8', is it?
At 2015-09-10 01:26:36, "Jarno Rajahalme" wrote:
On Sep 8,
I use the openvswitch vxlan tunnel as the basic overlay service.
But I find the bandwith between two virtual machine only reaches 4G/s using
iperf to test.
So could anybody tell me how to improve it or where is the bottleneck?
Looking forward for your reply. Thank you!
_
When I am reading the source of openvswitch I find the logic about fragment.
rule_dpif_lookup_in_table function:
(...skip...)
if (frag && ofproto->up.frag_handling == OFPC_FRAG_NORMAL) {
/* We must pretend that transport ports are unavailable. */
struct flow ofpc_norma
ad
>>to do with the kernel interface. We should really document the logic. I'll
>>try to >>remember to add some documentation around that.
That makes looking forward to it!
Thanks.
At 2015-09-09 15:18:54, "Justin Pettit" wrote:
>
>> On Sep 8, 2015, at
--Justin
>
>
>> On Sep 8, 2015, at 7:33 PM, openvswitcher wrote:
>>
>> Thank you for your suggestions but I could not use the command you suggested.
>>
>>
>>
>>
>> root@comm:~# ovs-appctl upcall/enable-megaflows
>> "upcall/enable-
ernel. Is this right?
Looking for your reply. Thanks.
At 2015-09-09 14:12:37, "Ben Pfaff" wrote:
>On Wed, Sep 09, 2015 at 01:43:19PM +0800, openvswitcher wrote:
>> Thank you for your reply.
>>
>>
>> >The megaflow match 0.0.0.0/2.0.0.0 checks for:
>&
Thank you for your reply.
>The megaflow match 0.0.0.0/2.0.0.0 checks for:
> 0xxx...
But is there a typo? why 0xxx == 0/2 ?
At 2015-09-09 13:33:23, "Ben Pfaff" wrote:
>On Wed, Sep 09, 2015 at 10:47:14AM +0800, openvswitcher wrote:
Hi, all
I wonder why the installed megaflow is the one bellow:
> The installed megaflow i see is this:
> Megaflow: recirc_id=0,ip,in_port=2,dl_dst=fa:16:3e:e2:16:63,nw_dst=
> 0.0.0.0/2.0.0.0,nw_frag=no
Can 'nw_dst=0.0.0.0/2.0.0.0' matches any flow of which the prefix is
'network address:
appctl upcall/enable-megaflows` (also disable-megaflows) can be used to
turn it
on and off,
Thanks,
Alex Wang,
On 8 September 2015 at 00:28, openvswitcher wrote:
Hi,all
Could anybody tell me what the megaflow feature actually used for?
For example, if I configure one match rule bellow,
&quo
Hi,all
Could anybody tell me what the megaflow feature actually used for?
For example, if I configure one match rule bellow,
"in_port=1, dl_vlan=0x20, action set_vlan=0x10, output:2"
then, if one packet 'dl_vlan=0x20, src_mac=A, dst_mac=B' arrived.
what will be the difference of flow inst
ecause gdb cannot map the function name
>to the code address space until it's able to read the debug info for vswitchd.
>
>openvswitcher wrote:
>> Hi, all
>>
>>
>>I want to ask about how to make the vswitchd break on the breakpoint by
>> gdb.
>>I
Hi, all
I want to ask about how to make the vswitchd break on the breakpoint by gdb.
I attached the vswitchd process using 'gdb -p' as bellow:
gdb --pid=31212
(gdb) b dpif_linux.c:dpif_linux_port_add
No symbol table is loaded. Use the "file" command.
Make breakpoint pending on future share
32 matches
Mail list logo
