Hi Simon,
Sorry I haven't responded to your patches yet. I'm not ignoring them
- I've just been very busy lately. Your high level descriptions all
sounds good so far though. I'm hoping to have a chance to look at
them this weekend.
On Wed, May 23, 2012 at 5:26 PM, Simon Horman wrote:
> Hi Jes
I was wrong on handling v4/v6 tos. I will re-check man pages for -Q option,
maybe it's using old TOS format? I will assume it's going be DSCP for IPv4 and
TClass for IPv6 and copy lower 3 bits into MPLS EXP bits.
Thanks,
Ravi
-Original Message-
From: dev-boun...@openvswitch.org [mailto:
On Thu, May 24, 2012 at 09:29:11AM -0700, Ben Pfaff wrote:
> On Thu, May 24, 2012 at 06:08:56PM +0900, Simon Horman wrote:
> > Cc: Kyle Mestery
>
> But I don't see him CCed?
Strange. I asked git send-mail to CC him explicitly.
> > +ovs_be32 ipv4_src;
> > +ovs_be32 ipv4_dst;
> >
On Thu, May 24, 2012 at 09:47:38AM -0700, Ben Pfaff wrote:
> The concept seems OK to me here. I have only a few minor comments.
>
> On Thu, May 24, 2012 at 06:08:57PM +0900, Simon Horman wrote:
> > +#define TNL_F_CSUM (1 << 0) /* Checksum packets. */
> > +#define TNL_F_TOS_INHERIT (1 <<
Thanks Ben for your thorough review. Comments inline
-Original Message-
From: dev-boun...@openvswitch.org [mailto:dev-boun...@openvswitch.org] On
Behalf Of Ben Pfaff
Sent: Thursday, May 24, 2012 4:02 PM
To: Kerur, Ravi
Cc: dev@openvswitch.org
Subject: Re: [ovs-dev] MPLS and VLAN QinQ pat
On Thu, May 24, 2012 at 11:58:53PM +0200, ravi.ke...@telekom.com wrote:
> Ben Pfaff writes:
> > get_l3_ttl_and_tos()
> >
> >
> > Why does get_l3_ttl_and_tos() consider an unknown IP version as
> > success, with a default? This seems odd.
>
> To handle non-IP traffic i.e push
Replies Inline
-Original Message-
From: Ben Pfaff [mailto:b...@nicira.com]
Sent: Thursday, May 24, 2012 1:45 PM
To: Kerur, Ravi
Cc: dev@openvswitch.org
Subject: Re: [ovs-dev] MPLS and VLAN QinQ patch
On Tue, May 22, 2012 at 07:36:32PM +0200, ravi.ke...@telekom.com wrote:
> Attached MPLS
On Tue, May 22, 2012 at 02:23:13PM -0700, Ethan Jackson wrote:
> There's some trailing whitespace added to the ofctl man page.
Thanks, fixed.
> Otherwise looks good. May benefit from a unit test.
Good idea, I added this to the existing unit test:
diff --git a/tests/ofproto.at b/tests/ofproto.a
Thanks, merged to master.
Ethan
On Thu, May 24, 2012 at 1:47 PM, Ben Pfaff wrote:
> On Tue, May 22, 2012 at 04:44:58PM -0700, Ethan Jackson wrote:
>> Signed-off-by: Ethan Jackson
>> ---
>>
>> Mildly ironic.
>
> Indeed.
>
> Thanks, this looks good, please commit.
On Tue, May 22, 2012 at 04:44:58PM -0700, Ethan Jackson wrote:
> Signed-off-by: Ethan Jackson
> ---
>
> Mildly ironic.
Indeed.
Thanks, this looks good, please commit.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Tue, May 22, 2012 at 07:36:32PM +0200, ravi.ke...@telekom.com wrote:
> Attached MPLS and VLAN QinQ patch after rebasing to following commit
>
> commit 046f1f89e6d7716581de207dd0c54421926bc25b
> Author: Ethan Jackson mailto:et...@nicira.com>>
> Date: Mon May 21 13:20:18 2012 -0700
>
> Patch(s
Signed-off-by: Ethan Jackson
---
Mildly ironic.
---
lib/memory.c |1 +
1 file changed, 1 insertion(+)
diff --git a/lib/memory.c b/lib/memory.c
index 779860e..1babfe9 100644
--- a/lib/memory.c
+++ b/lib/memory.c
@@ -119,6 +119,7 @@ compose_report(const struct simap *usage, struct ds *s)
I don't have any specific use-case in mind for mpls ttl match. I think I
misunderstood nw_ttl implementation.
Thanks,
Ravi
-Original Message-
From: Ben Pfaff [mailto:b...@nicira.com]
Sent: Thursday, May 24, 2012 10:41 AM
To: Kerur, Ravi
Cc: dev@openvswitch.org
Subject: Re: [ovs-dev] MPL
On Thu, May 24, 2012 at 07:25:52PM +0200, ravi.ke...@telekom.com wrote:
> > As for the TTL, I don't think that any OpenFlow version allows for
> > matching on an IP TTL, so there is no support-by-analogy for
> > matching on the MPLS TTL. Is there another reason to support
> > matching on the MPLS
Inline
-Original Message-
From: Ben Pfaff [mailto:b...@nicira.com]
Sent: Thursday, May 24, 2012 10:06 AM
To: Kerur, Ravi
Cc: dev@openvswitch.org
Subject: Re: [ovs-dev] MPLS and VLAN QinQ patch
I've fixed your quoting.
On Thu, May 24, 2012 at 06:38:58PM +0200, ravi.ke...@telekom.com
wro
These changes seem logically separate to me. If a single change
addresses both issues (as you suggest), then that's fine with me.
On Thu, May 24, 2012 at 06:55:51PM +0200, ravi.ke...@telekom.com wrote:
> In the list you mention VLAN QinQ and handling ox88a8 as separate activity,
> is there a rea
I've fixed your quoting.
On Thu, May 24, 2012 at 06:38:58PM +0200, ravi.ke...@telekom.com
wrote:
> Ben Pfaff writes:
> > OF1.3 adds the ability to match on the MPLS "BOS" bit. Should we add that
> > here too?
>
> My plan was to first get to 1.1 and later on to 1.2/1.3. I
> think it is good to ha
In the list you mention VLAN QinQ and handling ox88a8 as separate activity, is
there a reason for that i.e. some special handling needed for 0x88a8 as outer
vlan tag in OVS?
In the VLAN QinQ patch I sent earlier, it handles 0x88a8 and 0x8100 as outer
tag and can match on either one. There might
The concept seems OK to me here. I have only a few minor comments.
On Thu, May 24, 2012 at 06:08:57PM +0900, Simon Horman wrote:
> +#define TNL_F_CSUM (1 << 0) /* Checksum packets. */
> +#define TNL_F_TOS_INHERIT(1 << 1) /* Inherit ToS from inner packet. */
> +#define TNL_F_TTL_INHER
Couple of things... Since it is not technical discussion I have unicasted.
Could you please change my email contact for MPLS/QinQ to rke...@gmail.com.
Gmail is much more reliable than our company email.
Thanks,
Ravi
-Original Message-
From: dev-boun...@openvswitch.org [mailto:dev-boun.
On 5/23/12 11:50 AM, Phan Xuan Thien wrote:
- Can we deploy an OpenFlow network with multiple Controllers over
multiple network Domains, each Controller manage a subset of OF
Switches? (Can we deploy an OpenFlow network in which some OF Switches
of this network connect to a Controller ( call
Thanks Luiz, Please update on your test results.
Thanks,
Ravi
-Original Message-
From: Luiz Ozaki [mailto:luiz.oz...@locaweb.com.br]
Sent: Wednesday, May 23, 2012 2:38 PM
To: Kerur, Ravi
Cc: b...@nicira.com; dev@openvswitch.org
Subject: Re: [ovs-dev] MPLS and VLAN QinQ patch
On 5/22/12
Please see inline . I have removed the ones which I fixed it. Some
questions/clarifications inline.
-Original Message-
From: Ben Pfaff [mailto:b...@nicira.com]
Sent: Wednesday, May 23, 2012 11:10 AM
To: Kerur, Ravi
Cc: dev@openvswitch.org
Subject: Re: [ovs-dev] MPLS and VLAN QinQ patch
Vealo en directo por
internet
On Thu, May 24, 2012 at 06:08:56PM +0900, Simon Horman wrote:
> Cc: Kyle Mestery
But I don't see him CCed?
> +ovs_be32 ipv4_src;
> +ovs_be32 ipv4_dst;
> +unsigned long long tun_flags;
> +int ipv4_tos;
> +int ipv4_ttl;
> +int n = -1;
> +
> +
On Thu, May 24, 2012 at 05:15:04PM +0900, Simon Horman wrote:
> The inclusion include/openvswitch/tunnel.h does not seem to be needed any
> more.
>
> Signed-off-by: Simon Horman
Pushed to master, thank you!
___
dev mailing list
dev@openvswitch.org
http
CC: je...@nicira.com; dev@openvswitch.org; roberto.sa...@polito.it;
paolo.smirag...@polito.it; disc...@openvswitch.org
From: sergius_kvan...@yahoo.co.uk
Subject: Re: [ovs-discuss] [ovs-dev] Problem with DHCP and TUNNEL
Date: Thu, 24 May 2012 15:30:26 +0300
To: dano1...@hotmail.it
On May 24,
On May 24, 2012, at 15:04, Daniele Milani wrote:
>
>
> > From: je...@nicira.com
> > Date: Wed, 23 May 2012 09:20:21 -0700
> > Subject: Re: [ovs-dev] Problem with DHCP and TUNNEL
> > To: dano1...@hotmail.it
> > CC: disc...@openvswitch.org; dev@openvswitch.org; roberto.sa...@polito.it;
> > pao
> From: je...@nicira.com
> Date: Wed, 23 May 2012 09:20:21 -0700
> Subject: Re: [ovs-dev] Problem with DHCP and TUNNEL
> To: dano1...@hotmail.it
> CC: disc...@openvswitch.org; dev@openvswitch.org; roberto.sa...@polito.it;
> paolo.smirag...@polito.it
>
> On Wed, May 23, 2012 at 8:59 AM, Daniele
This duplicates parse_tunnel_config, the duplication will later be minimised.
iface_parse_tunnel() is currently only used to verify the configuration
by passing NULL as its third argument. It will later be used in storing
the configuration by passing a non-NULL argument. The purpose of verificatio
Use the tun_key, which is the basis of flow-based tunnelling, on transmit.
Cc: Kyle Mestery
Signed-off-by: Simon Horman
---
datapath/tunnel.c | 45 -
1 file changed, 32 insertions(+), 13 deletions(-)
diff --git a/datapath/tunnel.c b/datapath/tunnel.c
The use of these flags in the tnl_mutable_config structure
are no longer correct as a tunnel device may be used to
transmit packets for many different tunnels.
This change restores the checksum and out key behavior of
tunneling.
Cc: Kyle Mestery
Signed-of-by: Simon Horman
---
datapath/tunnel.c
As tunndevs no longer have a daddr the cache can no longer built in this way.
Furthermore, its not clear to me what the value of keeping the cache is in
the context of moving towards allowing use of in-tree tunnelling.
Cc: Kyle Mestery
Signed-off-by: Simon Horman
---
datapath/tunnel.c | 384 +++
These flags should always be valid and allows the flags
element of tnl_mutable_config to be removed.
The flags in mutable were actually not being set due to a previous patch in
this series, so all flag-related features, except outgoing ken and csum
which were restored in a previous patch, were dis
The lookup is now only based on the net and tunnel type.
It should be possible to either get rid of the lookup alltogether
or push it into the GRE and CAPWAP implementations, but this
change is simpler for now
Cc: Kyle Mestery
Signed-off-by: Simon Horman
---
datapath/tunnel.c | 110 +++---
OVS_KEY_ATTR_TUN_ID may now be removed as it is
no longer used in any meaningful way.
Cc: Kyle Mestery
Signed-off-by: Simon Horman
---
datapath/datapath.c | 1 -
datapath/flow.c | 1 -
include/linux/openvswitch.h | 1 -
lib/dpif-netdev.c | 1 -
lib/odp-util.c
On the user-space side of things, the existing tunnel classes become tunnel
realdev classes and new classes are added to provide tunnel tundevs.
On the datpath side of things, the existing tunnel vports are used as
tundev vports. A new vport is added for tunnel realdevs.
It should be possible to
The tun_key should always be present and correct.
Mutable no longer stores correct address information
and the saddr and daddr fields will be removed.
Cc: Kyle Mestery
Signed-off-by: Simon Horman
---
datapath/tunnel.c | 42 +-
1 file changed, 17 insertion
tun_key should always be present and correct in ovs_tnl_send()
It ought to be possible to handle the ttl entirely
in user-space. This is not implemented yet. However, the
TNL_F_TOS_INHERIT is currently never set.
Cc: Kyle Mestery
Signed-off-by: Simon Horman
---
datapath/tunnel.c | 10 ++---
In essence this attached the tun_key, if any,
to the output processing of a packet. This allows
it the packet to be transmitted using flow-based
tunneling as necessary.
Cc: Kyle Mestery
Signed-off-by: Simon Horman
---
v4
* Set tun_flags field of flow.tun_key
* Remove debugging message
v3
* In
This allows the tun_key tp be bassed throughout user-space,
attached to a flow. This is the essence of flow-based tunneling.
This does not add tun_key or wildcards, other than the existing match for
the tun_id. It is envisaged that most if not all fields of the tun_key
could be wildcarded.
Cc: Ky
In essence this is a duplication of ovs_tnl_find_port(),
copying code from the datapath to vswitchd. It is planned
that the datapath version will be removed.
It is used to map from the tundev interface that a
packet is recieved by in the datapath to the tunnel realdev
interface used in user-sapce.
This is used to map a tunnel or VLAN realdevs to
tundev and vlandevs respectively. This is used
on transmit to map fromt the interface used
in user-space to the interface used in the datapath.
In the case where an interface is not a tunnel
and does not have VLAN splinters configured
a identity map
Multicast may be handled in user-space (but isn't yet).
Cc: Kyle Mestery
Signed-off-by: Simon Horman
---
datapath/tunnel.c | 22 --
datapath/tunnel.h | 3 ---
2 files changed, 25 deletions(-)
diff --git a/datapath/tunnel.c b/datapath/tunnel.c
index f07ec69..cdcb0a7 100644
For tunnel realdevs this sets the remote IP and type,
and optionally source IP, ttl and tos. The remote IP
must non-zero.
For tunnel tundevs only the type is configured.
The remote IP must be zero.
Cc: Kyle Mestery
Signed-off-by: Simon Horman
---
vswitchd/bridge.c | 69
Cc: Kyle Mestery
Signed-off-by: Simon Horman
---
include/linux/openvswitch.h | 11 +++
lib/odp-util.c | 12 ++--
2 files changed, 17 insertions(+), 6 deletions(-)
diff --git a/include/linux/openvswitch.h b/include/linux/openvswitch.h
index 87a3e22..f2d56ec 100644
--
Add tunnel tundevs for tunnel realdevs as needed.
In general the notion is that realdevs may be configured by users
and from an end-user point of view are compatible with the existing
port-based tunneling code. And that tundevs exist in the datapath
arnd are actually used to send and recieve packe
Allow configuration of tunneling in ofproto_port instances.
For tunnel realdevs this includes the remote IP of the and type tunnel,
and optionally the local IP, tos and ttl.
For tunnel tundevs it only includes the type.
realdevs and tundevs can be differentiated by examining the remote IP,
which
this is a first pass at providing a tun_key which can be used
as the basis for flow-based tunnelling. The tun_key includes and
replaces the tun_id in both struct ovs_skb_cb and struct sw_tun_key.
In ovs_skb_cb tun_key is a pointer as it is envisaged that it will grow
when support for IPv6 to an ex
Hi,
This series comprises a fresh batch of proposed changes to introduce
flow-based tunnelling.
At the heart of these changes is the following structure, which
is attached as a pointer to skb->cb.
struct ovs_key_ipv4_tunnel {
__be64 tun_id;
__u32 tun_flags;
__be32 ipv4_s
Cc: Kyle Mestery
Signed-off-by: Simon Horman
---
v4
Correct parsing of tunnel key in parse_odp_key_attr()
so that it matches the out put of format_odp_key_attr()
TODO: fix test suite
v3
* Initial post
---
lib/odp-util.c | 29 +
1 file changed, 29 insertions(+)
di
The inclusion include/openvswitch/tunnel.h does not seem to be needed any
more.
Signed-off-by: Simon Horman
---
lib/odp-util.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/lib/odp-util.c b/lib/odp-util.c
index 13cdf8e..8693d3c 100644
--- a/lib/odp-util.c
+++ b/lib/odp-util.c
@@ -30,7 +30,6
52 matches
Mail list logo
