On Wed, Nov 06, 2013 at 06:25:09PM -0800, Alex Wang wrote:
> Don't know if Guolin will review it or not,
I doubt it (I didn't really expect him to). I'd be happy to get a
review from you.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/m
On Tue, Nov 05, 2013 at 09:29:56PM -0800, Andy Zhou wrote:
> On Tue, Nov 5, 2013 at 4:27 PM, Ben Pfaff wrote:
>
> > On Wed, Nov 06, 2013 at 09:22:43AM +0900, Simon Horman wrote:
> > > On Mon, Nov 04, 2013 at 11:24:02AM -0800, Ben Pfaff wrote:
> > > > I'm actually hoping to update NEWS soon with s
Don't know if Guolin will review it or not,
LGTM,
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
Hi Jarno,
Thank you very much for the reply.
1. I want to recognize a new routing scheme over Ethertype, so keeping dl_type
as ethertype, I want to add a new field that recognizes the new routing
protocol. In this case, which files of ovs-ofctl do I have to modify?
2. Another way of seeing
A couple of controller vendors have mentioned to me that they would like to
have some part of the OpenFlow port number space reserved for the
controller to use. This commit reserves 32768 and up (roughly the upper
half of the OF1.0 port range) to the controller.
Bug #18753.
Signed-off-by: Ben Pfa
> .RS
> -.IP \fBoutput:\fIport\fR
> -Outputs the packet to \fIport\fR, which must be an OpenFlow port
> -number or keyword (e.g. \fBLOCAL\fR).
> +.IP \fIport\fR
> +.IQ \fBoutput:\fIport\fR
> +Outputs the packet to OpenFlow port number \fIport\fR. If \fIport\fR
> +is the packet's input port, the p
Yes, sorry, I mistakenly tested it with "enqueue:()".
Now looks good to me, thx
On Wed, Nov 6, 2013 at 1:41 PM, Ben Pfaff wrote:
> On Wed, Nov 06, 2013 at 01:35:06PM -0800, Alex Wang wrote:
> > On Sat, Nov 2, 2013 at 9:51 PM, Ben Pfaff wrote:
> >
> > > The formatting of the "enqueue" action u
On Wed, Nov 06, 2013 at 01:35:06PM -0800, Alex Wang wrote:
> On Sat, Nov 2, 2013 at 9:51 PM, Ben Pfaff wrote:
>
> > The formatting of the "enqueue" action uses a "q" to separate the port
> > number from the queue number, as in "enqueue:123q456". This is different
> > from every other action. Th
On Sat, Nov 2, 2013 at 9:51 PM, Ben Pfaff wrote:
> The formatting of the "enqueue" action uses a "q" to separate the port
> number from the queue number, as in "enqueue:123q456". This is different
> from every other action. This commit improves the situation by:
>
> * Switching the formatti
Thanks Alex.
Ethan, do you mind checking this? It is a followup to 76c4290 (cfm:
Add ovsdb column "cfm_flap_count".)
On 6 November 2013 10:39, Alex Wang wrote:
>
>> After digging in a little further, I think that the only trouble was the
>> above issue. With bools backed by ints in C, the other
On Sat, Nov 2, 2013 at 9:35 PM, Ben Pfaff wrote:
> On Sat, Sep 14, 2013 at 01:46:32PM -0700, Ben Pfaff wrote:
> > The update is incomplete, so document that also.
> >
> > Reported-by: Stephen Finucane
> > Signed-off-by: Ben Pfaff
>
> This could use a review.
Not sure what does this comment me
On Nov 6, 2013, at 11:01 AM, Anu Mercian wrote:
> Dear Developers,
>
> I am looking to add a new field to ovs-ofctl agent to set a particular flow.
> For example, we have the command to set a new flow according to set of rules.
>
> ovs-ofctl add-flow br0
> "in_port=LOCAL, table=0, idle_timeou
Dear Developers,
I am looking to add a new field to ovs-ofctl agent to set a particular flow.
For example, we have the command to set a new flow according to set of rules.
ovs-ofctl add-flow br0"in_port=LOCAL, table=0,
idle_timeout=60,ip,hard_timeout=69,vlan_tci=0x, dl_src=78:2b:cb:4b:db:c5,
> After digging in a little further, I think that the only trouble was the
> above issue. With bools backed by ints in C, the other issue I was
> concerned about does not exist. The "old_cfm_fault != cfm->fault" logic
> checks if there has been a change in fault condition, and the existing
> flap c
Previously, the flap count logic for CFM was dependent on the state of
the logging facility. This patch ensures flaps are always counted when
there is a transition between non-fault and fault (and vice versa).
Found by inspection.
Signed-off-by: Joe Stringer
---
v2: Don't modify cfm->fault logic
On second thought, I have a tidier proposal. I will send a follow up.
On 6 November 2013 10:03, Joe Stringer wrote:
> Right---there's two separate issues; One is that flaps are only
> counted when logging is on, and the other is that faults are only
> logged when the connection flaps (rather than
Right---there's two separate issues; One is that flaps are only
counted when logging is on, and the other is that faults are only
logged when the connection flaps (rather than on any difference in
error). Currently, old_cfm_fault is a bool while cfm->fault is not;
this makes it harder to follow.
I
Thanks very much for spotting this.
Just one thing,
Signed-off-by: Joe Stringer
> ---
> ---
> lib/cfm.c | 27 +--
> 1 file changed, 13 insertions(+), 14 deletions(-)
>
> diff --git a/lib/cfm.c b/lib/cfm.c
> index d256a5f..01c9a8c 100644
> --- a/lib/cfm.c
> +++ b/lib/cf
Previously, the flap count logic for CFM was dependent on the state of
the logging facility. This patch ensures that flaps are counted in all
cases.
Found by inspection.
Signed-off-by: Joe Stringer
---
---
lib/cfm.c | 27 +--
1 file changed, 13 insertions(+), 14 deleti
Alexander Wu:
This patch is to implement table features GET msg in openflow 1.3.
The spec says:
If the request body contains an array of one or more ofp_table_features
structs, the switch will attempt to change its flow tables to match the
requested flow table configuration. This operati
V2:
Fix type of NEXT_TABLE_MISS
V1:
Implement the at for OFPMP_TABLE_FEATURES.
(I've tested it via NOX-OF1.3 too.)
Signed-off-by: Alexander Wu
---
tests/ofp-print.at | 187
tests/ofproto.at | 25 +++
2 files changed, 212 inse
V2:
No change
V1:
Now the cli we implement is very crude. Maybe it could
display better.
Signed-off-by: Alexander Wu
---
utilities/ovs-ofctl.c | 17 +
1 files changed, 17 insertions(+), 0 deletions(-)
diff --git a/utilities/ovs-ofctl.c b/utilities/ovs-ofctl.c
index a0
V2:
Change calls of ofputil.
Fix CodingStyle accoring to Simon Horman's suggestions.
V1:
Add function to print OFPMP_TABLE_FEATURES.
But now the print is crude and dirty.
Fix it to bitmap or more desc later.
Signed-off-by: Alexander Wu
---
lib/ofp-print.c | 127
V2:
No change.
V1:
Make the enums allowed
Signed-off-by: Alexander Wu
---
lib/rconn.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/rconn.c b/lib/rconn.c
index f7f90f7..0f72a99 100644
--- a/lib/rconn.c
+++ b/lib/rconn.c
@@ -1381,8 +1381,6 @@ is_admitted_m
V2:
1. fix function align, align to 79 bytes, fix BE64 to UINT64, etc.
2. fix n_tables_miss init type.
3. Change calls from ofputil.
Simon Horman suggestions:
1. Fix CodingStyle, some coding error.
2. Fix the init functions to macros. - I'll fix it later.
V1:
Add some functions to i
V2:
No change.
V1:
Add table_feature structs in ofproto && oftable to initialize.
The struct in ofproto is used to get
The struct in oftable is used to set (set is not implement yet)
Signed-off-by: Alexander Wu
---
ofproto/ofproto-provider.h |3 +++
1 files changed, 3 insertions(+
Alexander Wu:
This patch is to implement table features GET msg in openflow 1.3.
The spec says:
If the request body contains an array of one or more ofp_table_features
structs, the switch will attempt to change its flow tables to match the
requested flow table con guration. This operatio
V2:
No change.
V1:
Fix the limits to right ones.
Signed-off-by: Alexander Wu
---
lib/ofp-msgs.h |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/ofp-msgs.h b/lib/ofp-msgs.h
index c526a15..26fd6a3 100644
--- a/lib/ofp-msgs.h
+++ b/lib/ofp-msgs.h
@@ -343,10 +3
V2:
Restructure implement of OFPMP_TABLE_FEATURES
Change decode_*_raw to normalized pull functions
1. add macros and funcs to en/decode table features
2. restructure OFPMP_TABLE_FEATURES en/decode function
now they act like others.
3. Change big array to defines.(oxm, table_featur
29 matches
Mail list logo
