It seems that the behaviour is not so intuitive.
cf. https://bugs.launchpad.net/neutron/+bug/1346861
Signed-off-by: YAMAMOTO Takashi
---
FAQ | 21 +
1 file changed, 21 insertions(+)
diff --git a/FAQ b/FAQ
index df5ac0e..21a04f3 100644
--- a/FAQ
+++ b/FAQ
@@ -731,6 +731,27 @@
On Thu, Sep 18, 2014 at 2:30 AM, Ben Pfaff wrote:
> As it turns out, when output goes through a tunnel, the pkt_mark comes
> from the tunnel configuration. Currently, OVS forces it to 1 for IPSEC
> tunnel traffic and 0 for other tunnel traffic. This could be changed,
> and it probably should, b
On Wed, Sep 17, 2014 at 9:02 PM, Joe Stringer wrote:
>
>
> On 18 September 2014 13:02, Joe Stringer wrote:
>>>
>>>
>>> > @@ -229,17 +244,19 @@ static bool match_validate(const struct
>>> > sw_flow_match *match,
>>> > }
>>> > }
>>> >
>>> > - if ((key_attrs & key_expec
On 18 September 2014 13:02, Joe Stringer wrote:
>
>> > @@ -229,17 +244,19 @@ static bool match_validate(const struct
>> sw_flow_match *match,
>> > }
>> > }
>> >
>> > - if ((key_attrs & key_expected) != key_expected) {
>> > + attrs = attrs_to_bitmask(key_attrs,
Alin,
Thanks for incorporating all the comments.
Does it get any better if you don't do the renaming in this patch?
I looked at the patch and there's only deleted code in dpif-linux.c and new
code in dpif-netlink.c. It is hard to do any comparison about what was changed.
I am OK with this thoug
Hi,
I am currently working on a project that needs to use Open vSwitch, and
modify the data payload of the OpenFlow messages that are sent back and
forward between the controller and Open vSwitch (data payload here means
the rest of the bits in an OpenFlow message except the 8 bytes OpenFlow
messa
The original message was received at Thu, 18 Sep 2014 10:59:31 +0900
from 216.168.47.192
- The following addresses had permanent fatal errors -
dev@openvswitch.org
- Transcript of session follows -
... while talking to openvswitch.org.:
554 Service unavailable; [91.75.108.198] blo
OVS keeps pointer to packet key in skb->cb, but the packet key is
store on stack. This could make code bit tricky. So it is better to
get rid of the pointer.
Signed-off-by: Pravin B Shelar
---
datapath/actions.c| 294 +++---
datapath/datapath.c |
This add support for the select group action to ovs-vswtichd.
This new feature is currently disabled in xlate_select_group()
because ctx->xbridge->dp_select_group is always false.
This patch is a prototype and has several limitations:
* It assumes that no actions follow a select group action
b
Allow validation and copying of select group actions.
This completes the prototype select group action implementation
in the datapath. Subsequent patches will add support to ovs-vswtichd.
Signed-off-by: Simon Horman
---
datapath/flow_netlink.c | 102 ++
This is a quick hack to enable the datapath group select action.
It is in lieu of some combination of:
* probing
* run-time configuration by the end-use.
* run-time heuristic to use the action as appropriate
Signed-off-by: Simon Horman
---
ofproto/ofproto-dpif-xlate.c | 2 +-
1 file changed, 1 i
This is in preparation for using last_action() from
more than one C file as part of supporting an odp select group action.
Signed-off-by: Simon Horman
---
datapath/actions.c | 6 --
datapath/datapath.h | 5 +
2 files changed, 5 insertions(+), 6 deletions(-)
diff --git a/datapath/action
Allow execution of select group action in the datapath.
A subsequent patch will add validation and copying of
the select group action in the datapath.
The selection algorithm used is based on the RSS hash.
This was chosen because it resembles the algorithm currently
used by the implementation of s
Allow formatting of select group action. This is used
when pretty-printing datapath flows. Subsequent patches
will add support for the select group action to the datapath
and ovs-vswtichd.
Signed-off-by: Simon Horman
---
lib/odp-util.c | 67 +++
Add a multiple field to struct nl_policy which if set suppresses
warning of duplicate attributes in nl_parse_nested().
As is the case without this patch only the last occurrence of an
attribute is stored in attrs by nl_parse_nested(). As such
if the multiple field of struct nl_policy is set then i
This is the core of a proposed ODP select group action.
It models OpenFlow select groups without any extensions.
Further work:
We believe there is scope to add OVS_SELECT_GROUP_ATTR_* attributes
to supply parameters to the selection method: for example which
selection algorithm to use. This relat
[repost with correct dev@openvswitch.org address]
Hi,
the purpose of this patch-set is to provide a prototype of a select group
action in the Open vSwitch datapath. And the motivation for that is to
allow offloading of selection either in the datapath or by any hooks
provided by the datapath for
On Thu, Sep 11, 2014 at 05:46:03PM -0700, Jesse Gross wrote:
> On Sun, Sep 7, 2014 at 7:18 PM, Simon Horman
> wrote:
> > On Fri, Sep 05, 2014 at 12:07:17PM -0700, Jesse Gross wrote:
> >> On Thu, Sep 4, 2014 at 12:28 AM, Simon Horman
> >> wrote:
> >> > On Tue, Sep 02, 2014 at 07:20:30PM -0700, Pr
Unfortunately there some special characters in dpif-linux.c
that do not allow me to send the patch normally.
Trying via attachment.
Short description:
The patch contains the necessary modifications to compile and also to run
under MSVC.
Added the files to the build system and also changed dpif_
Thanks for the review, comments inline
On 17 September 2014 17:37, Pravin Shelar wrote:
> On Wed, Sep 10, 2014 at 3:16 AM, Joe Stringer
> wrote:
> > Reduce duplicate code by using nla_policy and nla_parse_strict().
> >
> Thanks for simplifying this code. I have couple of comments.
>
> > Signed-
On 18 September 2014 11:03, Ben Pfaff wrote:
> On Mon, Sep 15, 2014 at 02:25:13PM +1200, Joe Stringer wrote:
> > Add the 128-bit murmurhash by Austin Appleby, for 32-bit systems from:
> > http://code.google.com/p/smhasher/source/browse/trunk/MurmurHash3.cpp
> >
> > Signed-off-by: Joe Stringer
>
On 18 September 2014 10:53, Ben Pfaff wrote:
> On Mon, Sep 15, 2014 at 02:25:12PM +1200, Joe Stringer wrote:
> > This allows us to ignore most fields of a flow_dump, requiring only the
> > flow key for looking up the ukey. Fetching flows can also be avoided in
> > the corner case where a flow is
Thanks for the review,
On 18 September 2014 10:33, Ben Pfaff wrote:
> On Mon, Sep 15, 2014 at 02:25:11PM +1200, Joe Stringer wrote:
> > Currently, when a revalidator thread first dumps a flow, it creates a
> > 'udpif_key' object and caches a copy of a kernel flow key. This allows
> > us to perfo
I'm OK with it, I'll take your word for it that it's a better approach.
Acked-by: Ben Pfaff
On Wed, Sep 17, 2014 at 03:04:11PM -0700, Ethan Jackson wrote:
> I think something like that would be useful. However, once the number
> of flows gets really high, I like the idea of having something tha
On 18 September 2014 11:13, Ben Pfaff wrote:
> On Mon, Sep 15, 2014 at 02:25:14PM +1200, Joe Stringer wrote:
> > One of the limiting factors on the number of flows that can be supported
> > in the datapath is the overhead of assembling flow dump messages in the
> > datapath. This patch adds a new
On Tue, Sep 16, 2014 at 10:44 AM, Thomas Graf wrote:
> Port datapath to work with kernrels up to 3.17 and use 3.16.2 as
> the new kernel for CI testing.
>
> Tested with 3.14, 3.16.2, and net-next (3.17).
>
Thanks for working on this. I have couple of comments.
> Signed-off-by: Thomas Graf
> Co-a
On Mon, Sep 15, 2014 at 02:25:14PM +1200, Joe Stringer wrote:
> One of the limiting factors on the number of flows that can be supported
> in the datapath is the overhead of assembling flow dump messages in the
> datapath. This patch adds a new alternative to dumping the key, mask and
> actions fro
Hi Justin,
I got a chance to try out your code with the example flows and measure raw
packet throughput with and without connection tracking. The code works and
the performance numbers look pretty good.
I ran two sets of tests both with 4 iptable rules in chain input and 8 in
chain forward, all d
This was required for old compatibility code. Now vswitchd
has dropped it. This support was always deprecated, so
finally removing it.
Signed-off-by: Pravin B Shelar
---
datapath/datapath.c | 8
datapath/vport.c| 46 ++
datapath/vport.h
On Mon, Sep 15, 2014 at 02:25:13PM +1200, Joe Stringer wrote:
> Add the 128-bit murmurhash by Austin Appleby, for 32-bit systems from:
> http://code.google.com/p/smhasher/source/browse/trunk/MurmurHash3.cpp
>
> Signed-off-by: Joe Stringer
Why does the loop in hash_words128 count up from a negati
On Mon, Sep 15, 2014 at 02:25:12PM +1200, Joe Stringer wrote:
> This allows us to ignore most fields of a flow_dump, requiring only the
> flow key for looking up the ukey. Fetching flows can also be avoided in
> the corner case where a flow is missed from a dump but revalidation is
> required.
>
>
On Sep 17, 2014, at 11:12 PM, Eitan Eliahu
wrote:
> This code handles an event notification subscription for a user mode thread
> which joins an MC group. The event wait handler queues an IRP which is
> completed upon change in a port state.
>
> Signed-off-by: Eitan Eliahu
LG. Thanks for doin
Ok will update it to dpif_netlink_vport*.
-Mesaj original-
De la: Ben Pfaff [mailto:b...@nicira.com]
Trimis: Wednesday, September 17, 2014 3:37 PM
Către: Alin Serdean
Cc: Nithin Raju; Eitan Eliahu; dev@openvswitch.org
Subiect: Re: [ovs-dev] [PATCH] Changes needed for dpif-linux to compil
On Wed, Sep 17, 2014 at 10:33:04PM +, Alin Serdean wrote:
> 1. I did not change the dpif_linux_vport* functions because that means
> also touching netdev-linux.c. If that is ok with Ben, I can also add
> them to a v2 of the patch.
I only see one reference to dpif_linux_vport* from netdev-linux
Hi Nithin,
Thank you for looking over the patch.
1. I did not change the dpif_linux_vport* functions because that means also
touching netdev-linux.c. If that is ok with Ben, I can also add them to a v2 of
the patch.
2. I will look over the spacing in v2.
Thank you again,
Nithin.
-Mesaj or
On Mon, Sep 15, 2014 at 02:25:11PM +1200, Joe Stringer wrote:
> Currently, when a revalidator thread first dumps a flow, it creates a
> 'udpif_key' object and caches a copy of a kernel flow key. This allows
> us to perform lookups in the classifier to attribute stats and validate
> the correctness
Just wanted to summarize the meeting for future reference:
Attendees: Samuel Ghinet, Alin Serdean, Ankur Sharma, Eitan Eliahu, Nithin Raju
> 1. statuses: DP commands, VPORT commands, Flow commands, Event commands
- Review for DP commands SET_DP and GET_DP along with transaction semantics has
be
This code handles an event notification subscription for a user mode thread
which joins an MC group. The event wait handler queues an IRP which is
completed upon change in a port state.
Signed-off-by: Eitan Eliahu
---
datapath-windows/ovsext/Datapath.c | 103 +
Thanks!
On 9/17/14, 3:19 PM, "Jarno Rajahalme" wrote:
>Looks good, thanks again!
>
>Acked-by: Jarno Rajahalme
>
>& pushed to master,
>
> Jarno
>
>On Sep 5, 2014, at 5:10 PM, Daniele Di Proietto
>wrote:
>
>> This optimization is done to avoid calling count_1bits(), which, if the
>>popcnt
>> is
Looks good, thanks again!
Acked-by: Jarno Rajahalme
& pushed to master,
Jarno
On Sep 5, 2014, at 5:10 PM, Daniele Di Proietto wrote:
> This optimization is done to avoid calling count_1bits(), which, if the popcnt
> istruction, is not available might is slow. popcnt may not be available
>
Looks good, thanks!
Acked-by: Jarno Rajahalme
& pushed to master,
Jarno
On Sep 5, 2014, at 5:10 PM, Daniele Di Proietto wrote:
> netdev_flow_key is a miniflow with the following constraints:
>
> 1) It is used only inside dpif-netdev.c.
> 2) It always has inline values.
> 3) It contains on
On Sep 17, 2014, at 1:11 PM, Alin Serdean
wrote:
> The patch contains the necessary modifications to compile and also to run
> under MSVC.
>
> Added the files to the build system and also changed dpif_linux to be under
> a more generic name dpif_windows.
>
> Added a TODO under the windows par
On Wed, Sep 17, 2014 at 1:34 AM, Jiri Pirko wrote:
> Thu, Sep 04, 2014 at 10:46:28PM CEST, pshe...@nicira.com wrote:
>>On the other hand if vswitchd uses common interface (switchdev) there
>>is no need to extend ovs kernel interface. For example specifying
>>extra metadata, like (sw only, hw olny,
On Wed, Sep 17, 2014 at 2:56 PM, Pravin Shelar wrote:
> On Wed, Sep 17, 2014 at 9:13 AM, Nicolas Dichtel
> wrote:
>> From: Samuel Gauthier
>>
>> Since commit fb5d1e9e127a ("openvswitch: Build flow cmd netlink reply only
>> if needed."),
>> the new flows are not notified to the listeners of OVS_
I think something like that would be useful. However, once the number
of flows gets really high, I like the idea of having something that's
a bit more in your face. When debugging these things, people tend to
tail the OVS log file instead of grepping for warnings and error
messages. I think some
Looks good to me, thanks!
Acked-by: Jarno Rajahalme
Pushed to master,
Jarno
On Sep 17, 2014, at 2:14 PM, Daniele Di Proietto wrote:
> This optimization should give a small performance benefit to the userspace
> datapath.
>
> Signed-off-by: Daniele Di Proietto
> ---
> ofproto/ofproto-dpif
On Wed, Sep 17, 2014 at 9:13 AM, Nicolas Dichtel
wrote:
> From: Samuel Gauthier
>
> Since commit fb5d1e9e127a ("openvswitch: Build flow cmd netlink reply only if
> needed."),
> the new flows are not notified to the listeners of OVS_FLOW_MCGROUP.
>
> This commit fixes the problem by checking that
I will rebase and also the filename change in V2.
-Mesaj original-
De la: Ben Pfaff [mailto:b...@nicira.com]
Trimis: Wednesday, September 17, 2014 2:06 PM
Către: Alin Serdean
Cc: Eitan Eliahu; 'Nithin Raju'; dev@openvswitch.org
Subiect: Re: [ovs-dev] [PATCH] Changes needed for dpif-linux
This optimization should give a small performance benefit to the userspace
datapath.
Signed-off-by: Daniele Di Proietto
---
ofproto/ofproto-dpif-xlate.c | 25 ++--
tests/classifier.at | 20 +++
tests/dpif-netdev.at | 16 ++---
tests/ofproto-dpif.at| 136 ++
We can do the file re-naming after this checkin. I'm reviewing the change. Let
me get back in a few minutes.
thanks,
Nithin
On Sep 17, 2014, at 1:38 PM, Ben Pfaff
wrote:
> I'm happy to do that. I don't care whether it happens before or after
> or simultaneously with the change under conside
On Wed, Sep 17, 2014 at 08:11:08PM +, Alin Serdean wrote:
> The patch contains the necessary modifications to compile and also to run
> under MSVC.
>
> Added the files to the build system and also changed dpif_linux to be under
> a more generic name dpif_windows.
>
> Added a TODO under the wi
Acked-by: Eitan Eliahu
-Original Message-
From: Alin Serdean [mailto:aserd...@cloudbasesolutions.com]
Sent: Wednesday, September 17, 2014 1:50 PM
To: Eitan Eliahu; Ben Pfaff
Cc: Nithin Raju; dev@openvswitch.org
Subject: RE: [ovs-dev] [PATCH] Changes needed for dpif-linux to compile under
As it turns out, when output goes through a tunnel, the pkt_mark comes
from the tunnel configuration. Currently, OVS forces it to 1 for IPSEC
tunnel traffic and 0 for other tunnel traffic. This could be changed,
and it probably should, because it is surprising, but this is the first
that I've not
Did a full clone->autoreconf->make->make check under Linux did not see any
issues.
Alin.
-Mesaj original-
De la: Eitan Eliahu [mailto:elia...@vmware.com]
Trimis: Wednesday, September 17, 2014 1:47 PM
Către: Ben Pfaff
Cc: Alin Serdean; Nithin Raju; dev@openvswitch.org
Subiect: RE: [ovs-d
Once Alin confirms we didn't break anything with Linux we are ready to go.
Thanks,
Eitan
-Original Message-
From: Ben Pfaff [mailto:b...@nicira.com]
Sent: Wednesday, September 17, 2014 1:38 PM
To: Eitan Eliahu
Cc: Alin Serdean; Nithin Raju; dev@openvswitch.org
Subject: Re: [ovs-dev] [PATC
On Wed, Sep 17, 2014 at 01:25:19PM -0700, Ethan Jackson wrote:
> Frequently we've run into controller bugs which result in hundreds of
> thousands, or even millions of rules being installed in an OpenFlow
> table. This isn't something trouble-shooters naturally think of to
> check for, so it's nic
I can change the filename after the patch gets merged no problem :).
Thank you,
Alin.
-Mesaj original-
De la: Ben Pfaff [mailto:b...@nicira.com]
Trimis: Wednesday, September 17, 2014 11:38 PM
Către: Eitan Eliahu
Cc: Alin Serdean; Nithin Raju; dev@openvswitch.org
Subiect: Re: [ovs-dev] [P
I'm happy to do that. I don't care whether it happens before or after
or simultaneously with the change under consideration.
Did you review the change?
On Wed, Sep 17, 2014 at 08:30:57PM +, Eitan Eliahu wrote:
> Hi Alin, should we change the file name to dpif_netlink?
> Thanks,
> Eitan
>
>
FYI All I've dropped this patch.
Ethan
On Mon, Sep 8, 2014 at 2:42 PM, Ben Pfaff wrote:
> On Mon, Sep 08, 2014 at 12:51:11PM -0700, Ethan Jackson wrote:
>> > Is there any point to testing OVS in a configuration that we won't
>> > distribute?
>>
>> This is a subject in which there could legitimat
Hi Alin, should we change the file name to dpif_netlink?
Thanks,
Eitan
-Original Message-
From: Alin Serdean [mailto:aserd...@cloudbasesolutions.com]
Sent: Wednesday, September 17, 2014 1:11 PM
To: Eitan Eliahu; Nithin Raju; 'Ben Pfaff'; dev@openvswitch.org
Subject: [ovs-dev] [PATCH] Chan
Frequently we've run into controller bugs which result in hundreds of
thousands, or even millions of rules being installed in an OpenFlow
table. This isn't something trouble-shooters naturally think of to
check for, so it's nice to have a low rate warning message to hint at
the potential problem.
The patch contains the necessary modifications to compile and also to run
under MSVC.
Added the files to the build system and also changed dpif_linux to be under
a more generic name dpif_windows.
Added a TODO under the windows part in case we want to implement another
counterpart for epoll functi
OK, this may be a bug. I'll try to investigate soon.
On Wed, Sep 17, 2014 at 11:53:16PM +0530, Nirmalanand Jebakumar wrote:
> Hi Ben,
>
> The ovs-dpctl command output does not show any skb_mark being set for the
> datapath flow:
>
> # ovs-dpctl dump-flows
> skb_priority(0),in_port(4),eth(src=52
Hi Ben,
The ovs-dpctl command output does not show any skb_mark being set for the
datapath flow:
# ovs-dpctl dump-flows
skb_priority(0),in_port(4),eth(src=52:54:00:74:f9:41,dst=52:54:00:aa:93:9e),eth_type(0x0800),ipv4(src=
2.2.2.3/255.255.255.255,dst=2.2.2.2/255.255.255.255,proto=1/0xff,tos=0/0x3
On Wed, Sep 17, 2014 at 05:12:20PM +0530, Nirmalanand Jebakumar wrote:
> In this link below, I read that pkt_mark is supported as a MATCH field.
> Does OVS support pkt_mark as a SET action?
Yes.
> The command I ran is
> /usr/bin/ovs-ofctl add-flow -O Openflow13 br0 "
> hard_timeout=3600,idle_tim
I'll try to review the remaining patches this afternoon.
On Wed, Sep 17, 2014 at 05:32:40PM +1200, Joe Stringer wrote:
> Bump. Any chance of some feedback?
>
> On 15 September 2014 14:25, Joe Stringer wrote:
>
> > Patches #1-4 have been reviewed and Ack'd by Ben, which I have held out
> > from
On Fri, Sep 12, 2014 at 11:24 AM, Jarno Rajahalme wrote:
> This new flag is useful for suppressing error logging while probing
> for datapath features using flow commands. For backwards
> compatibility reasons the commands are executed normally, but error
> logging is suppressed.
>
> Signed-off-b
Hi Sam,
Looks good. Some comments:
Shouldn't we check the return code from NlMsgPutTailU32() ?
Do we check the size of the out message before we populate it with the port NL
attributes?
Thanks,
Eitan
-Original Message-
From: Samuel Ghinet [mailto:sghi...@cloudbasesolutions.com]
Sent: Wed
From: Samuel Gauthier
Since commit fb5d1e9e127a ("openvswitch: Build flow cmd netlink reply only if
needed."),
the new flows are not notified to the listeners of OVS_FLOW_MCGROUP.
This commit fixes the problem by checking that there are listeners in
the actual OVS_FLOW_MCGROUP group, instead of
How about using RtlEqualString() ?
Str1.MaximumLength = 16;
Str1.Buffer = &AnsiString;
if (!RtlEqualString( &str1, &str2, TRUE ){
}
-Original Message-
From: Nithin Raju
Sent: Wednesday, September 17, 2014 8:34 AM
To: Samuel Ghinet
Cc: dev@openvswitch.o
Use this area to offer a short teaser of your email's content. Text here will
show in the preview area of some email clients.
Is this email not displaying correctly?
View it in your browser
(http://us9.campaign-archive1.com/?u=35917e482
On Sep 17, 2014, at 6:16 AM, Samuel Ghinet
wrote:
> I don't think there's a need to implement such a function.
> You can use memcmp, which behaves quite like strncmp.
hi Samuel,
Thanks for the pointer. I was trying to use Rtl* functions, but if memcmp()
works, that is great.
I have a couple o
Hi Nithin, look good. Few comments:
if (request_nlmsg->nlmsg_seq != reply_nlmsg->nlmsg_seq)
Unlink the Linux implementation the above condition should never happen. I
would put just an assert there.
Before accessing request_nlmsg we need to check its size against the reply_len
parameter retu
Oh, except a little typo in the commit message:
"Eg. the Windows kernel does not send embed an error"
Sam
From: Samuel Ghinet
Sent: Wednesday, September 17, 2014 4:34 PM
To: Nithin Raju; dev@openvswitch.org; elia...@vmware.com; ankursha...@vmware.com
Subjec
Hello Nithin,
This patch looks good!
Acked-by: Samuel Ghinet
From: Nithin Raju [nit...@vmware.com]
Sent: Wednesday, September 17, 2014 5:06 AM
To: dev@openvswitch.org; Samuel Ghinet; elia...@vmware.com;
ankursha...@vmware.com
Cc: Nithin Raju
Subject: [P
Nithin,
I don't think there's a need to implement such a function.
You can use memcmp, which behaves quite like strncmp.
Regards,
Sam
From: Nithin Raju [nit...@vmware.com]
Sent: Wednesday, September 17, 2014 5:06 AM
To: dev@openvswitch.org; Samuel Ghinet;
Functionality for vport dump.
Later, when we will add more netlink dump commands, some common code will need
to be split to functions.
Notes:
a) the current implementation of vport assumes the datapath feature
"multiple upcall pids" is not used. A single upcall pid is used instead.
c) the vxlan de
The windows ovs kernel uses an OVS_VPORT_TYPE enum that is incompatible with
the userspace counterpart (enum ovs_vport_type from openvswitch.h). We must use
the same enum type - enum ovs_vport_type - for the netlink communication to work
properly.
This patch makes the fix: "typedef enum ovs_vport_
Work-in-Progress
Attach conntrack template to packet if a non standard zone setting has
been provided. Conntrack template is stored in action.
---
datapath/actions.c| 12 ++-
datapath/datapath.c | 18 +++--
datapath/flow.h
From: Justin Pettit
An RFC to get back early feedback on exposing Linux's kernel connection
tracker to OVS. The code has a few rough spots that will be addressed
in the next version:
- Need better interface than setting individual flags for the state.
- Need support for IP frags.
-
Justin,
I've extended your conntrack path with zone support. It's unfinished
and still has open issues such as how to calculate the internal
sw_flow_actions length.
Storing nf_conn in the actions is somewhat of a mess due to the
possible misalignment of the pointer (only 4byte alignment is guaran
Hello everyone,
For some implementation of ours, we need to mark the socket
buffer(sk_buff->skb_mark) using flows in an openvswitch (v 2.0) before
sending it through an egress vxlan port on the OVS, for setting ToS and CoS
values using iptables command later.
VXLAN offload is not enabled on the h
Thu, Sep 04, 2014 at 10:46:28PM CEST, pshe...@nicira.com wrote:
>On Thu, Sep 4, 2014 at 5:33 AM, Jiri Pirko wrote:
>> Wed, Sep 03, 2014 at 08:41:39PM CEST, pshe...@nicira.com wrote:
>>>On Wed, Sep 3, 2014 at 2:24 AM, Jiri Pirko wrote:
After this, flow related structures can be used in other
83 matches
Mail list logo
