On 15/06/2015 19:28, Gurucharan Shetty shet...@nicira.com wrote:
On Mon, Jun 15, 2015 at 11:09 AM, Daniele Di Proietto
diproiet...@vmware.com wrote:
With this commit, the VTEP emulator detects the datapath_type of the
bridge used as a physical switch, and creates subsequent bridges
with the
With this commit, the VTEP emulator detects the datapath_type of the
bridge used as a physical switch, and creates subsequent bridges
with the same type. This allows ovs-vtep to work with the userspace
datapath.
Signed-off-by: Daniele Di Proietto diproiet...@vmware.com
---
v1 - v2:
Applied
On 15/06/2015 20:40, Alex Wang al...@nicira.com wrote:
I actually have a similar patch in my ovn for vtep series. but instead, I
added
a new command-line option for that... But goal is the same, = use
different
datapath type (dummy in my case).
So, I'm fine with this change, and will drop
On Tue, Jun 16, 2015 at 08:16:33AM -0700, Ben Pfaff wrote:
On Mon, Jun 08, 2015 at 01:05:41PM -0300, Thadeu Lima de Souza Cascardo wrote:
Support IGMPv3 messages with multiple records. Make sure all IGMPv3
messages go through slow path, since they may carry multiple multicast
addresses,
On 16/06/2015 09:55, Gray, Mark D mark.d.g...@intel.com wrote:
Acked-by: Ethan Jackson et...@nicira.com
One question I had for a future patch. Have you considered prefetching
more packets at once? I.E. 4 at a time or something?
That's how these things are typically written, though
With this commit, the VTEP emulator detects the datapath_type of the
bridge used as a physical switch, and creates subsequent bridges
with the same type. This allows ovs-vtep to work with the userspace
datapath.
Signed-off-by: Daniele Di Proietto diproiet...@vmware.com
---
v2 - v3:
* Use
On Tue, Jun 16, 2015 at 12:38:06PM -0300, Thadeu Lima de Souza Cascardo wrote:
On Tue, Jun 16, 2015 at 08:16:33AM -0700, Ben Pfaff wrote:
On Mon, Jun 08, 2015 at 01:05:41PM -0300, Thadeu Lima de Souza Cascardo
wrote:
Support IGMPv3 messages with multiple records. Make sure all IGMPv3
On Tue, Jun 16, 2015 at 8:25 AM, Daniele Di Proietto
diproiet...@vmware.com wrote:
With this commit, the VTEP emulator detects the datapath_type of the
bridge used as a physical switch, and creates subsequent bridges
with the same type. This allows ovs-vtep to work with the userspace
On Tue, Jun 16, 2015 at 06:44:35AM -0700, Gurucharan Shetty wrote:
MSVC 2013 does not like designated initializers when
structs are initialized inside structs.
Apparently it has been fixed in MSVC 2015.
Signed-off-by: Gurucharan Shetty gshe...@nicira.com
Acked-by: Ben Pfaff b...@nicira.com
On 16/06/2015 15:39, Gurucharan Shetty shet...@nicira.com wrote:
+global ps_type
+ps_type = ovs_vsctl(get Bridge %s datapath_type % ps_name)
+if ps_type == \\:
+ps_type =
Else where in the code, this is handled by calling strip().
e.g: port_type = ovs_vsctl(get
On Tue, Jun 16, 2015 at 6:20 AM, Daniele Di Proietto
diproiet...@vmware.com wrote:
With this commit, the VTEP emulator detects the datapath_type of the
bridge used as a physical switch, and creates subsequent bridges
with the same type. This allows ovs-vtep to work with the userspace
On Mon, Jun 15, 2015 at 01:22:59PM -0700, Alex Wang wrote:
On Sun, Jun 14, 2015 at 12:19 PM, Ben Pfaff b...@nicira.com wrote:
This allows XML-generated manpages in the source tree to include correct
directory names for the local configuration, instead of just the plain
nroff ones.
On Mon, Jun 08, 2015 at 01:05:41PM -0300, Thadeu Lima de Souza Cascardo wrote:
Support IGMPv3 messages with multiple records. Make sure all IGMPv3
messages go through slow path, since they may carry multiple multicast
addresses, unlike IGMPv2.
Tests done:
* multiple addresses in IGMPv3
On Mon, Jun 15, 2015 at 02:49:04PM +, Sorin Vinturis wrote:
When the filter detach routine is called while there are packets
still in processing, the OvsUninitSwitchContext function call will
decrement the switch context reference count without releasing the
switch context structure. This
+set X $1; shift
+if test $# != 1; then
+echo 2 $FUNCNAME: sandbox name must be a single word
+return 1
+fi
+
This is a cool check~
Thanks.
I think that this is a positive review, so I'm going to push this to
'ovn' in a minute.
Sounds
MSVC 2013 does not like designated initializers when
structs are initialized inside structs.
Apparently it has been fixed in MSVC 2015.
Signed-off-by: Gurucharan Shetty gshe...@nicira.com
---
ovn/controller/ofctrl.c | 44 ++--
1 file changed, 22
On Tue, Jun 16, 2015 at 01:07:36AM -0700, Alex Wang wrote:
Really like this patch~!
Thanks!
+ dd
+Adds a new port to varbridge/var in the default sandbox (as
set
+with codeas/code) that plugs it into the varnetwork/var
+interconnection network.
Acked-by: Eitan Eliahu elia...@vmware.com
Thanks Sorin.
Eitan
-Original Message-
From: Sorin Vinturis [mailto:svintu...@cloudbasesolutions.com]
Sent: Tuesday, June 16, 2015 2:36 AM
To: Eitan Eliahu; dev@openvswitch.org
Subject: RE: [ovs-dev] [PATCH v2] datapath-windows: BSOD when
The packed annotation doesn't do anything here because all of the members
in the structure are naturally aligned.
Signed-off-by: Ben Pfaff b...@nicira.com
---
lib/lacp.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/lib/lacp.c b/lib/lacp.c
index 65149fd..29b8b43 100644
On Tue, Jun 16, 2015 at 11:23:07AM +0530, gowrishankar wrote:
Also updated an additional correction as in :
http://openvswitch.org/pipermail/dev/2015-June/056379.html
Could you include that as well.
Yeap. One patch with all the s/ovs-vsctl/ovs-vsctl/
and another for the scheme
On Mon, Jun 15, 2015 at 05:55:13PM +, Daniele Di Proietto wrote:
On 15/06/2015 12:16, Traynor, Kevin kevin.tray...@intel.com wrote:
There is a dpdk patchset that contains a potential fix for this and lots
of
other changes, but I haven't tested yet.
On Mon, Jun 15, 2015 at 02:31:45PM -0700, Ben Pfaff wrote:
On Fri, Jun 12, 2015 at 12:51:24PM -0400, Russell Bryant wrote:
This patch creates a new subpackage for OVN, openvswitch-ovn. It also
installs systemd unit files for ovn-controller and ovn-northd.
If you want to run
Sorry for this very delayed reply, thx so much for the review~!
Addressed all comments to this series,
I think I have three more things to do here:
1. adding more unit tests,
2. making this command run in a separate thread,
3. received a request to add another command to detect unused openflow
On Tue, Jun 16, 2015 at 08:47:34AM -0700, Ben Pfaff wrote:
The packed annotation doesn't do anything here because all of the members
in the structure are naturally aligned.
Signed-off-by: Ben Pfaff b...@nicira.com
---
Acked-by: Flavio Leitner f...@sysclose.org
On 16/06/2015 07:40, Pravin Shelar pshe...@nicira.com wrote:
On Mon, Jun 8, 2015 at 7:42 PM, Pravin Shelar pshe...@nicira.com wrote:
On Mon, Jun 8, 2015 at 6:13 PM, Wei li l...@dtdream.com wrote:
When tx queue is shared among CPUS,the pkts always be flush in
'netdev_dpdk_eth_send'
So it is
Acked-by: Ethan Jackson et...@nicira.com
On Tue, Jun 16, 2015 at 10:53 AM, Flavio Leitner f...@sysclose.org wrote:
On Tue, Jun 16, 2015 at 08:47:34AM -0700, Ben Pfaff wrote:
The packed annotation doesn't do anything here because all of the members
in the structure are naturally aligned.
Good to see these done!~
Acked-by: Alex Wang al...@nicira.com
On Tue, Jun 16, 2015 at 10:33 AM, Ben Pfaff b...@nicira.com wrote:
Signed-off-by: Ben Pfaff b...@nicira.com
---
ovn/TODO | 85
1 file changed, 85 deletions(-)
On 16/06/2015 18:27, Flavio Leitner f...@sysclose.org wrote:
On Mon, Jun 15, 2015 at 05:55:13PM +, Daniele Di Proietto wrote:
On 15/06/2015 12:16, Traynor, Kevin kevin.tray...@intel.com wrote:
There is a dpdk patchset that contains a potential fix for this and
lots
of
other changes, but
Signed-off-by: Ben Pfaff b...@nicira.com
---
ovn/TODO | 85
1 file changed, 85 deletions(-)
diff --git a/ovn/TODO b/ovn/TODO
index fe296b4..07d66da 100644
--- a/ovn/TODO
+++ b/ovn/TODO
@@ -1,64 +1,5 @@
* ovn-controller
-** Flow
IGMP packets need to take the slow path. Otherwise, packets that match
the same flow will not be processed by OVS. That might prevent OVS from
updating the expire time for entries already in the mdb, but also to
lose packets with different addresses in the payload.
Signed-off-by: Thadeu Lima de
From: Dennis Flynn drfl...@avaya.com
This commit fixes a bug where the i-sid/vlan mapping structures associated with
an lldp-enabled port were not being freed during general port cleanup.
Signed-off-by: Dennis Flynn drfl...@avaya.com
diff --git a/lib/lldp/lldpd-structs.c
On Tue, Jun 16, 2015 at 06:25:47PM -0300, Flavio Leitner wrote:
On Tue, Jun 16, 2015 at 06:01:09PM -0300, Thadeu Lima de Souza Cascardo wrote:
IGMP packets need to take the slow path. Otherwise, packets that match
the same flow will not be processed by OVS. That might prevent OVS from
If you tried it and it didn't improve, then I think what we have now is great.
Ethan
On Tue, Jun 16, 2015 at 9:08 AM, Daniele Di Proietto
diproiet...@vmware.com wrote:
On 16/06/2015 09:55, Gray, Mark D mark.d.g...@intel.com wrote:
Acked-by: Ethan Jackson et...@nicira.com
One question I
DPDK mbufs contain a valid RSS hash only if PKT_RX_RSS_HASH is
set in 'ol_flags'. Otherwise the hash is garbage and doesn't
relate to the packet.
This fixes an issue with vhost, which, being a virtual NIC, doesn't
compute the hash.
Unfortunately the ixgbe vPMD doesn't set the PKT_RX_RSS_HASH,
On Tue, Jun 16, 2015 at 06:01:09PM -0300, Thadeu Lima de Souza Cascardo wrote:
IGMP packets need to take the slow path. Otherwise, packets that match
the same flow will not be processed by OVS. That might prevent OVS from
updating the expire time for entries already in the mdb, but also to
Support IGMPv3 messages with multiple records. Make sure all IGMPv3
messages go through slow path, since they may carry multiple multicast
addresses, unlike IGMPv2.
Tests done:
* multiple addresses in IGMPv3 report are inserted in mdb;
* address is removed from IGMPv3 if record is INCLUDE_MODE;
IGMP packets need to take the slow path. Otherwise, packets that match
the same flow will not be processed by OVS. That might prevent OVS from
updating the expire time for entries already in the mdb, but also to
lose packets with different addresses in the payload.
Signed-off-by: Thadeu Lima de
On Tue, Jun 16, 2015 at 05:16:19PM -0300, Thadeu Lima de Souza Cascardo wrote:
IGMP packets need to take the slow path. Otherwise, packets that match
the same flow will not be processed by OVS. That might prevent OVS from
updating the expire time for entries already in the mdb, but also to
On Tue, Jun 16, 2015 at 05:16:20PM -0300, Thadeu Lima de Souza Cascardo wrote:
Support IGMPv3 messages with multiple records. Make sure all IGMPv3
messages go through slow path, since they may carry multiple multicast
addresses, unlike IGMPv2.
Tests done:
* multiple addresses in IGMPv3
On Tue, Jun 16, 2015 at 11:39 AM, Daniele Di Proietto
diproiet...@vmware.com wrote:
DPDK mbufs contain a valid RSS hash only if PKT_RX_RSS_HASH is
set in 'ol_flags'. Otherwise the hash is garbage and doesn't
relate to the packet.
This fixes an issue with vhost, which, being a virtual NIC,
On Mon, Jun 15, 2015 at 5:39 PM, Simon Horman
simon.hor...@netronome.com wrote:
Currently output of MPLS packets on tunnel vports is not allowed by the
datapath and, moreover, flows that match on MPLS packets and output to
tunnel vports are rejected by the datapath. The flows are rejected
On Tue, Jun 16, 2015 at 06:48:13PM +, Daniele Di Proietto wrote:
On 16/06/2015 18:27, Flavio Leitner f...@sysclose.org wrote:
Could you point me to the vector and non-vector rx routines?
I feel like I am missing something.
ixgbe_recv_pkts_vec() in drivers/net/ixgbe/ixgbe_rxtx.c is the
On 06/12/15 20:31, Ansis Atteka wrote:
On Fri, Jun 12, 2015 at 5:50 AM, Franck Baudin franck.bau...@qosmos.com wrote:
Hi Ansis,
On 06/09/15 22:59, Ansis Atteka wrote:
Hi Franck
On 8 June 2015 at 09:34, Franck BAUDIN franck.bau...@qosmos.com
mailto:franck.bau...@qosmos.com wrote:
Hello,
On Tue, Jun 16, 2015 at 07:39:00PM +0100, Daniele Di Proietto wrote:
DPDK mbufs contain a valid RSS hash only if PKT_RX_RSS_HASH is
set in 'ol_flags'. Otherwise the hash is garbage and doesn't
relate to the packet.
This fixes an issue with vhost, which, being a virtual NIC, doesn't
compute
On Tue, Jun 16, 2015 at 07:09:25PM -0300, Thadeu Lima de Souza Cascardo wrote:
On Tue, Jun 16, 2015 at 06:25:47PM -0300, Flavio Leitner wrote:
On Tue, Jun 16, 2015 at 06:01:09PM -0300, Thadeu Lima de Souza Cascardo
wrote:
IGMP packets need to take the slow path. Otherwise, packets that
On Tuesday 16 June 2015 11:07 PM, Flavio Leitner wrote:
On Tue, Jun 16, 2015 at 11:23:07AM +0530, gowrishankar wrote:
Also updated an additional correction as in :
http://openvswitch.org/pipermail/dev/2015-June/056379.html
Could you include that as well.
Yeap. One patch with all the
As per the testcase included in this patch it has been observed
that ovs-vswtichd may segfault when deleting a bridge.
Analysing the output of valgrind and gdb it appears that
this is caused by the connmgr of a ofproto being accessed
after the latter has been freed.
It appears that this is
On 2015/6/17 1:44, Daniele Di Proietto wrote:
On 16/06/2015 07:40, Pravin Shelar pshe...@nicira.com wrote:
On Mon, Jun 8, 2015 at 7:42 PM, Pravin Shelar pshe...@nicira.com wrote:
On Mon, Jun 8, 2015 at 6:13 PM, Wei li l...@dtdream.com wrote:
When tx queue is shared among CPUS,the pkts
On Tuesday 16 June 2015 11:39 AM, Wei Li wrote:
On 2015/6/15 22:33, Gowrishankar wrote:
From: Gowrishankar M gowrishanka...@linux.vnet.ibm.com
For the first time while creating OVS db, ovsdb-tool has to be given
schema file existing in source (and not the one installed in distro
or not even
On 2015/6/15 22:33, Gowrishankar wrote:
From: Gowrishankar M gowrishanka...@linux.vnet.ibm.com
For the first time while creating OVS db, ovsdb-tool has to be given
schema file existing in source (and not the one installed in distro
or not even existing if not installed earlier).
Another
On Mon, Jun 8, 2015 at 7:42 PM, Pravin Shelar pshe...@nicira.com wrote:
On Mon, Jun 8, 2015 at 6:13 PM, Wei li l...@dtdream.com wrote:
When tx queue is shared among CPUS,the pkts always be flush in
'netdev_dpdk_eth_send'
So it is unnecessarily for flushing in netdev_dpdk_rxq_recv
Otherwise
On Mon, Jun 8, 2015 at 7:42 PM, Pravin Shelar pshe...@nicira.com wrote:
On Mon, Jun 8, 2015 at 6:13 PM, Wei li l...@dtdream.com wrote:
When tx queue is shared among CPUS,the pkts always be flush in
'netdev_dpdk_eth_send'
So it is unnecessarily for flushing in netdev_dpdk_rxq_recv
Otherwise
Hi Eitan,
Please see below the stacktrace of the BSOD. The FilterDetach routine was
called while the requests were being processed and the gOvsSwitchContext global
pointer was set to NULL. In this case the gOvsSwitchContext was not released,
but only the gOvsSwitchContextRefCount reference
Really like this patch~!
+h2Interconnection Network Commands/h2
+
+p
+ When multiple sandboxed Open vSwitch instances exist, one will
inevitably
+ want to connect them together. These commands allow for that.
+ Conceptually, an interconnection network is a switch
54 matches
Mail list logo