Gentile utente
La tua email è superato 2 GB, che viene creato dal webmaster, attualmente in
esecuzione a 2.30GB, non è possibile inviare o ricevere nuovi messaggi fino a
controllare il vostro account.
Compila il modulo sottostante per verificare il tuo account.
Compila il modulo per conferma
Simon,
Thanks for explaining. It is now clear what the test is about.
I spent some time today to study this. The root cause seems to be in
how packet out are
implemented, not with recirculation or bond. A reasonable solution
may be treating patch ports as
LOCAL ports for packet out. I have pro
On Wed, Apr 30, 2014 at 01:10:02AM -0700, Andy Zhou wrote:
> Simon,
>
> Thanks for explaining. It is now clear what the test is about.
>
> I spent some time today to study this. The root cause seems to be in
> how packet out are
> implemented, not with recirculation or bond. A reasonable solut
When controller send OFPT_PACKET_OUT message, with its input port set
to a patch port, and the packet handling requires recirculation,
the post recirculation flow is never set up since the xlate layer
rejects the up call triggered by such packet, on the bases those
up calls don't have valid datapat
On Wed, Apr 30, 2014 at 1:35 AM, Simon Horman wrote:
> On Wed, Apr 30, 2014 at 01:10:02AM -0700, Andy Zhou wrote:
>> Simon,
>>
>> Thanks for explaining. It is now clear what the test is about.
>>
>> I spent some time today to study this. The root cause seems to be in
>> how packet out are
>> impl
When controller sends OFPT_PACKET_OUT message with the in_port set
to a patch port or as CONTROLLER, and the message execution requires
recirculation, those packets will be dropped in the datapath.
This is because the post recirculation flow will not be set up by
Xlate layer that rejects up call wi
please ignore this patch. I've sent a v2 of it.
On Wed, Apr 30, 2014 at 2:09 AM, Andy Zhou wrote:
> When controller send OFPT_PACKET_OUT message, with its input port set
> to a patch port, and the packet handling requires recirculation,
> the post recirculation flow is never set up since the xlat
On Wed, Apr 30, 2014 at 02:22:24AM -0700, Andy Zhou wrote:
> On Wed, Apr 30, 2014 at 1:35 AM, Simon Horman wrote:
> > On Wed, Apr 30, 2014 at 01:10:02AM -0700, Andy Zhou wrote:
> >> Simon,
> >>
> >> Thanks for explaining. It is now clear what the test is about.
> >>
> >> I spent some time today to
On Wed, Apr 30, 2014 at 03:01:20AM -0700, Andy Zhou wrote:
> When controller sends OFPT_PACKET_OUT message with the in_port set
> to a patch port or as CONTROLLER, and the message execution requires
> recirculation, those packets will be dropped in the datapath.
> This is because the post recircula
Gentile utente
La tua email è superato 2 GB, che viene creato dal webmaster, attualmente in
esecuzione a 2.30GB, non è possibile inviare o ricevere nuovi messaggi fino a
controllare il vostro account.
Compila il modulo sottostante per verificare il tuo account.
Compila il modulo per conferma
On Wed, Apr 30, 2014 at 09:59:50AM +0900, YAMAMOTO Takashi wrote:
> > The Open vSwitch software switch now supports all the required features of
> > OpenFlow 1.0 through 1.3, with one known trivial exception[*]. Enable them
> > by default in ovs-vswitchd.
> >
> > For now, ovs-ofctl only enables O
> > +/* Bridge State Machine. */
> > +/* [17.28] Port Role Selection state machine. */
> > +
> > +void
> > +updt_role_disabled_tree(struct rstp * r)
> Is ther a specfc resn to contrct the first word in the nam of this functn?
We try to strictly follow the naming used in the IEEE 802.1D-2004 standa
On Wed, Apr 30, 2014 at 3:53 AM, Simon Horman wrote:
> On Wed, Apr 30, 2014 at 03:01:20AM -0700, Andy Zhou wrote:
>> When controller sends OFPT_PACKET_OUT message with the in_port set
>> to a patch port or as CONTROLLER, and the message execution requires
>> recirculation, those packets will be dr
An rconn has a human-readable name that typically designates both endpoints
of the connection. For a "reliable" rconn, that automatically reconnects,
the name remains constant regardless of whether the rconn is currently
connected. Until now, though, an "unreliable" rconn, that cannot
automatical
I've had a number of requests for more specific logging of flow_mod
information. It is useful for troubleshooting.
Signed-off-by: Ben Pfaff
---
ofproto/connmgr.c | 98
ofproto/connmgr.h |3 ++
ofproto/ofproto-provider.h |6
Signed-off-by: Ben Pfaff
---
ofproto/connmgr.h |2 --
1 file changed, 2 deletions(-)
diff --git a/ofproto/connmgr.h b/ofproto/connmgr.h
index 20c8160..c09c80f 100644
--- a/ofproto/connmgr.h
+++ b/ofproto/connmgr.h
@@ -158,8 +158,6 @@ enum ofperr ofconn_pktbuf_retrieve(struct ofconn *,
uint3
rconn objects do not cache IP address and port information any longer.
Signed-off-by: Ben Pfaff
---
lib/rconn.c |5 +
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/lib/rconn.c b/lib/rconn.c
index cb3cdd5..f19c14e 100644
--- a/lib/rconn.c
+++ b/lib/rconn.c
@@ -1085,10 +1085
Signed-off-by: Ben Pfaff
---
utilities/ovs-vsctl.8.in |4
1 file changed, 4 insertions(+)
diff --git a/utilities/ovs-vsctl.8.in b/utilities/ovs-vsctl.8.in
index 43b00bf..1701b48 100644
--- a/utilities/ovs-vsctl.8.in
+++ b/utilities/ovs-vsctl.8.in
@@ -975,6 +975,10 @@ Configure bridge \f
If a packet-in were to be queued, and the sequence number changed, after
grabbing the list of packet-ins, then the existing code could have gone to
sleep until something happened. By grabbing the sequence number before
the list of packet-ins, we avoid this race.
Found by inspection.
Signed-off-b
Thanks Ben for spotting and fixing this,
Acked-by: Alex Wang
On Wed, Apr 30, 2014 at 11:08 AM, Ben Pfaff wrote:
> If a packet-in were to be queued, and the sequence number changed, after
> grabbing the list of packet-ins, then the existing code could have gone to
> sleep until something happe
On Tue, Apr 29, 2014 at 7:08 PM, Jesse Gross wrote:
> On Tue, Apr 29, 2014 at 4:34 PM, Pravin B Shelar wrote:
>> Patch fixes following warning:
>> datapath/linux/flow_table.c:580:40: warning: symbol 'flow' shadows an
>> earlier one
>> datapath/linux/flow_table.c:558:24: originally declared here
This patch enables the client dpdk rings within the netdev-dpdk. It adds
a new
dpdk device called dpdkr (other naming suggestions?). This allows for the
use
of shared memory to communicate with other dpdk applications, on the host
or
within a virtual machine. Instructions for use are in INSTALL.
Acked-by: Ethan Jackson
On Tue, Apr 29, 2014 at 4:15 PM, Alex Wang wrote:
> This commit adds a new requirement that cfm session must receive
> at least one ccm every 100 * cfm_interval amount of time in demand
> mode. Otherwise, even if the data packets are received on the
> monitored interfac
Applied to master, branch-2.2, branch-2.1. Thanks!
On Wed, Apr 30, 2014 at 11:19:22AM -0700, Alex Wang wrote:
> Thanks Ben for spotting and fixing this,
>
> Acked-by: Alex Wang
>
>
> On Wed, Apr 30, 2014 at 11:08 AM, Ben Pfaff wrote:
>
> > If a packet-in were to be queued, and the sequence
Acked-by: Ethan Jackson
On Tue, Apr 29, 2014 at 4:15 PM, Alex Wang wrote:
> This commit adds a requirement that bfd session must receive at least
> one bfd control packet every 100 * bfd->cfg_min_rx amount of time in
> forwarding_if_rx mode. Otherwise, even if the data packets are received
> o
Pushed to master,
Jarno
On Apr 16, 2014, at 2:40 PM, Jarno Rajahalme wrote:
> Thanks for your review! I'll hold on to this until we have figured out when
> to enable this feature.
>
> Jarno
>
>> On Apr 16, 2014, at 2:28 PM, Ethan Jackson wrote:
>>
>> Acked-by: Ethan Jackson
>>
>> I su
On Wed, Apr 30, 2014 at 1:01 PM, Rogers, Gerald wrote:
> This patch enables the client dpdk rings within the netdev-dpdk. It adds
> a new
> dpdk device called dpdkr (other naming suggestions?). This allows for the
> use
> of shared memory to communicate with other dpdk applications, on the host
The poll_loop code has a feature that, when turned on manually or
automatically (due to high CPU use), logs the source file and line number
of the code that caused a thread to wake up from poll(). Until now, when
a function calls seq_wait(), the source file and line number logged was
the code insi
Signed-off-by: Justin Pettit
---
NEWS |4 ++--
debian/changelog |4 ++--
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/NEWS b/NEWS
index a31740e..e1fb093 100644
--- a/NEWS
+++ b/NEWS
@@ -1,6 +1,6 @@
-v2.1.2 - xx xxx
+v2.1.2 - 30 Apr 2014
This code created a cls_rule without destroying it. Found by
inspection.
Signed-off-by: Ethan Jackson
---
ofproto/ofproto.c | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c
index 3d788a6..208efc1 100644
--- a/ofproto/ofproto.c
+++
This configuration option was shifted in 2.0, then removed in 2.1.
Remove the misleading log message.
Signed-off-by: Joe Stringer
---
vswitchd/bridge.c |7 ---
1 file changed, 7 deletions(-)
diff --git a/vswitchd/bridge.c b/vswitchd/bridge.c
index 84e9ab8..12852b4 100644
--- a/vswitchd/
Acked-by: Ethan Jackson
On Wed, Apr 30, 2014 at 2:57 PM, Joe Stringer wrote:
> This configuration option was shifted in 2.0, then removed in 2.1.
> Remove the misleading log message.
>
> Signed-off-by: Joe Stringer
> ---
> vswitchd/bridge.c |7 ---
> 1 file changed, 7 deletions(-)
>
>
Acked-by: Jarno Rajahalme
I would put the match and priority parameters on the same line, though.
Jarno
On Apr 30, 2014, at 2:53 PM, Ethan Jackson wrote:
> This code created a cls_rule without destroying it. Found by
> inspection.
>
> Signed-off-by: Ethan Jackson
> ---
> ofproto/ofproto.
I'm not sure of a good way to make them fit . . .
At any rate, I'm getting rid of the rule_from_cls_rule() function in a
future patch so that will clean this up quite a bit.
Ethan
On Wed, Apr 30, 2014 at 3:02 PM, Jarno Rajahalme wrote:
> Acked-by: Jarno Rajahalme
>
> I would put the match and
On Apr 30, 2014, at 3:07 PM, Ethan Jackson wrote:
> I'm not sure of a good way to make them fit . . .
>
Oops, it was my email client with variable width font that created an “optical
illusion of space”…
Jarno
> At any rate, I'm getting rid of the rule_from_cls_rule() function in a
> futur
On Wed, Apr 30, 2014 at 02:51:43PM -0700, Justin Pettit wrote:
> Signed-off-by: Justin Pettit
Acked-by: Ben Pfaff
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
Pushed to master and branch-2.2.
On 1 May 2014 09:59, Ethan Jackson wrote:
> Acked-by: Ethan Jackson
>
>
> On Wed, Apr 30, 2014 at 2:57 PM, Joe Stringer
> wrote:
> > This configuration option was shifted in 2.0, then removed in 2.1.
> > Remove the misleading log message.
> >
> > Signed-off-by
On Tue, Apr 29, 2014 at 3:24 PM, Pritesh Kothari
wrote:
> diff --git a/datapath/linux/compat/include/linux/skbuff.h
> b/datapath/linux/compat/include/linux/skbuff.h
> index 714c955..de9b29d 100644
> --- a/datapath/linux/compat/include/linux/skbuff.h
> +++ b/datapath/linux/compat/include/linux/skb
On Tue, Apr 29, 2014 at 10:58 PM, Simon Horman wrote:
> On Tue, Apr 29, 2014 at 11:41:57AM -0700, Jesse Gross wrote:
>> On Mon, Apr 28, 2014 at 5:13 PM, Simon Horman wrote:
>> > On Mon, Apr 28, 2014 at 02:37:47PM -0700, Jesse Gross wrote:
>> >> On Mon, Apr 28, 2014 at 12:00 AM, Simon Horman wrot
On Fri, Apr 25, 2014 at 1:52 PM, wrote:
> Hi Jesse,
>
> 2014-04-25, Jesse Gross:
>>> Practically speaking, making your layer 3 port code work for
>>> MPLS-over-GRE is not entirely trivial:
>>> - on emission, compose_output_action__ pops the Ethernet header before
>>> push_mpls is called (which is
Thanks for the review, pushed both patches to master and branch-2.2
On Wed, Apr 30, 2014 at 1:38 PM, Ethan Jackson wrote:
> Acked-by: Ethan Jackson
>
>
> On Tue, Apr 29, 2014 at 4:15 PM, Alex Wang wrote:
> > This commit adds a requirement that bfd session must receive at least
> > one bfd con
Current datapath allows the same packet to be executed up to 5 times to
avoid blowing out the kernel stack. Recirculation is currently also
counted as one execution, but does not use same amount of stack compare
to other services, such as IPsec.
This patch introduces the concept of stack cost. Rec
If one tries to install a rule that's identical to another rule in
another OpenFlow table which is being deleted, it's possible that
is_flow_deletion_pending() might confuse them and block the
installation. This is such an edge case I doubt it has ever actually
happened.
Found by inspection.
Sig
> On Wed, Apr 30, 2014 at 10:24:46AM +0900, YAMAMOTO Takashi wrote:
>> The problem mentioned by Simon Horman in the following mail.
>> http://openvswitch.org/pipermail/dev/2014-April/039492.html
>>
>> Cc: Simon Horman
>> Signed-off-by: YAMAMOTO Takashi
>
> One of the most pointless differences
44 matches
Mail list logo
