There was a reason for 2 calls for dp_netdev_pmd_reload_done() inside
pmd_thread_main(). The reason is that we must wait until PMD thread
completely done with reloading. This patch introduces race condition
for pmd->exit_latch. While removing last port on numa node
dp_netdev_reload_pmd__(pmd) will
On 19.04.2016 00:19, Daniele Di Proietto wrote:
>
>
> On 18/04/2016 07:50, "Ilya Maximets" wrote:
>
>> On 08.04.2016 06:13, Daniele Di Proietto wrote:
>>> Signed-off-by: Daniele Di Proietto
>>> ---
>>> lib/dpif-netdev.c | 243
>>> +++---
>>> 1 f
From: Benli Ye
This patch enables port level IPFIX. Before this patch, OVS supported
per bridge IPFIX and per flow IPFX, and exporting packet tunnel headers
is only supported by bridge IPFIX. This patch adds port level IPFIX
for easy configuration and port level IPFIX also supports exporting
pack
From: Guru Shetty
static routes are useful when connecting multiple
routers with each other.
Signed-off-by: Gurucharan Shetty
Signed-off-by: steve.ruan
---
ovn/northd/ovn-northd.8.xml | 5 +-
ovn/northd/ovn-northd.c | 101 +
ovn/ovn-nb.ovsschema | 15 ++
I want to add a field in the flow, and i checked the FAQ but i have a
problem is that field is not related to the packet, and when i use
"ovs-ofctl add-flow br0
idle_timeout=180,priority=33000,new_field=1547,actions=output:2" it give me
that ovs extract the field but when i use "ovs-ofctl dump-flow
On 19.04.2016 10:18, Ilya Maximets wrote:
> There was a reason for 2 calls for dp_netdev_pmd_reload_done() inside
> pmd_thread_main(). The reason is that we must wait until PMD thread
> completely done with reloading. This patch introduces race condition
> for pmd->exit_latch. While removing last p
From: Johnson Li
In user space, only standard VxLAN was support. This patch will
add the VxLAN-GBP support for the user space data path.
How to use:
1> Create VxLAN port with GBP extension
$ovs-vsctl add-port br-int vxlan0 -- set interface vxlan0 \
type=vxlan options:dst_port=6633 \
For a given release this is listed in the INSTALL.DPDK.md
but it gets asked quite a bit on the mailing list, so create
a table in the FAQ.
Signed-off-by: Kevin Traynor
---
FAQ.md | 13 +
1 files changed, 13 insertions(+), 0 deletions(-)
diff --git a/FAQ.md b/FAQ.md
index 0fee992..
Fix build warning: 'flags_mask' may be used uninitialized.
Signed-off-by: Antonio Fischetti
---
lib/odp-util.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/odp-util.c b/lib/odp-util.c
index b4689cc..10fb6c2 100644
--- a/lib/odp-util.c
+++ b/lib/odp-util.c
@@ -2366,7 +2
This message was undeliverable due to the following reason(s):
Your message was not delivered because the destination computer was
unreachable within the allowed queue period. The amount of time
a message is queued before it is returned depends on local configura-
tion parameters.
Most likely the
Hi lifuqiong,
Could you provide the output of the following commands in your setup?:
ovs-vsctl show
ovs-ofctl dump-flows
ovs-ofctl dump-ports
Mauricio Vasquez,
On Sat, Apr 16, 2016 at 10:23 AM, lifuqiong
wrote:
> Hi Mauricio:
>
> I changed my qemu version from 2.2.1 to 2.5.0 and the v
On Mon, Apr 18, 2016 at 2:55 PM, Aaron Rosen wrote:
> I like this idea as well.
>
> The one question I have with is is how we should determine which ip address
> to select for the 'distributed' port?
Aaron, Thanks for your review, and feedback.
We can use the dhcp-port (and its IP address) for t
On Tue, Apr 19, 2016 at 3:20 AM, Johnson.Li wrote:
> diff --git a/lib/netdev-vport.c b/lib/netdev-vport.c
> index e398562..e6b35ed 100644
> --- a/lib/netdev-vport.c
> +++ b/lib/netdev-vport.c
> @@ -1297,10 +1298,16 @@ netdev_vxlan_pop_header(struct dp_packet *packet)
> return EINVAL;
>
Mickey, Can you give a first shot at the review. You had some issues with
the my code structure and looks like Steve decided to use something
similar.
On 19 April 2016 at 02:36, steve.ruan wrote:
> From: Guru Shetty
>
Steve,
You should be the author (when you respin after Mickey's review). Yo
>
> We can use the dhcp-port (and its IP address) for the distributed-port.
>
> The current native-dhcp proposal
> (http://openvswitch.org/pipermail/dev/2016-April/069787.html)
> assumes that a dhcp-server ip-address "server_id" is defined on the subnet.
>
> action=(dhcp_offer(offerip = 10.0.0.2, r
On Mon, 18 Apr 2016 11:30:43 -0700, Jesse Gross wrote:
> (Even on new kernels I don't know how this would interact
> with OVS tunnel ports that are not of type 'flow'. Would they also use
> the compat code?)
This is a very good point. Although, on the other hand, how much common
are tunnel ports w
On Tue, Apr 19, 2016 at 07:00:00PM +0200, Jiri Benc wrote:
> On Mon, 18 Apr 2016 11:30:43 -0700, Jesse Gross wrote:
> > (Even on new kernels I don't know how this would interact
> > with OVS tunnel ports that are not of type 'flow'. Would they also use
> > the compat code?)
>
> This is a very good
Signed-off-by: Hui Kang
---
ovn/utilities/ovn-sbctl.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/ovn/utilities/ovn-sbctl.c b/ovn/utilities/ovn-sbctl.c
index a257729..a888333 100644
--- a/ovn/utilities/ovn-sbctl.c
+++ b/ovn/utilities/ovn-sbctl.c
@@ -329,7 +329,7 @@ Options
Hi Ramu,
On Tue, Apr 19, 2016 at 8:12 AM, Ramu Ramamurthy
wrote:
> On Mon, Apr 18, 2016 at 2:55 PM, Aaron Rosen
> wrote:
> > I like this idea as well.
> >
> > The one question I have with is is how we should determine which ip
> address
> > to select for the 'distributed' port?
>
> Aaron, Than
On 15 April 2016 at 17:02, Daniele Di Proietto wrote:
> Linux and FreeBSD have slightly different names for these constants.
> Windows doesn't define them. It is simpler to redefine them from
> scratch for OVS. The new names are different than those used in Linux
> and FreeBSD.
>
> These definit
On 15 April 2016 at 17:02, Daniele Di Proietto wrote:
> This will be used by a future commit.
>
> Signed-off-by: Daniele Di Proietto
> ---
> lib/flow.c | 140
> ++---
> lib/flow.h | 3 ++
> 2 files changed, 81 insertions(+), 62 deletions
On 15 April 2016 at 17:02, Daniele Di Proietto wrote:
> The function simply returns the ethernet type of the packet (after
> eventually discarding the VLAN tag). It will be used by a following
> commit.
>
> Signed-off-by: Daniele Di Proietto
> ---
> lib/flow.c | 14 --
> lib/flow.h
On 15 April 2016 at 17:02, Daniele Di Proietto wrote:
> This commit adds the conntrack module.
>
> It is a connection tracker that resides entirely in userspace. Its
> primary user will be the dpif-netdev datapath.
>
> The module main goal is to provide conntrack_execute(), which offers a
> conve
On 15 April 2016 at 17:02, Daniele Di Proietto wrote:
> Signed-off-by: Daniele Di Proietto
Looks like this can be combined with patch #11 (to plumb it up through appctl).
Acked-by: Joe Stringer
___
dev mailing list
dev@openvswitch.org
http://openvswi
On 15 April 2016 at 17:02, Daniele Di Proietto wrote:
> Signed-off-by: Daniele Di Proietto
> +static void
> +conn_key_to_tuple(const struct conn_key *key, struct ct_dpif_tuple *tuple)
> +{
> +if (key->dl_type == htons(ETH_TYPE_IP)) {
> +tuple->l3_type = AF_INET;
> +} else if (k
On 15 April 2016 at 17:02, Daniele Di Proietto wrote:
> This commit implements the OVS_ACTION_ATTR_CT action in dpif-netdev.
>
> To allow ofproto-dpif to detect the conntrack feature, flow_put will not
> discard anymore flows with ct_* fields set. We still shouldn't allow
> flows with NAT bits set
On 15 April 2016 at 17:02, Daniele Di Proietto wrote:
> Signed-off-by: Daniele Di Proietto
Could be combined with the earlier patch that provides the actual
implementation?
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinf
On 15 April 2016 at 17:02, Daniele Di Proietto wrote:
> The system userspace testsuite uses the userspace datapath with
> netdev-linux devices, connected to veth pairs with the AF_PACKET socket:
>
> (veth pair) (AF_PACKET)
> TCP stack -> p0 ---> ovs-p0 -> netdev-linux
On 15 April 2016 at 17:02, Daniele Di Proietto wrote:
> Useful to test the datapath ability to forward tcp packets without the
> complexity of connection tracking.
>
> Signed-off-by: Daniele Di Proietto
I think we could do with a few more of these incremental steps on the
way up to the more comp
Currently we treat set_multiq() calls specially in netdev and dpif-netdev:
every pmd thread must be stopped and set_multiq() is allowed to destroy and
recreate the device.
I think we can improve this by:
* Generalizing the mechanism to allow changing other parameters at runtime
(such as MTU).
*
Signed-off-by: Daniele Di Proietto
---
lib/dpif-netdev.c | 14 +-
1 file changed, 5 insertions(+), 9 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 24717cc..060f5e0 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -430,8 +430,6 @@ struct dp_netdev_pmd_
This commit introduces some functions to add/remove rxqs from pmd
threads without reloading them. They will be used by next commits.
Signed-off-by: Daniele Di Proietto
---
lib/dpif-netdev.c | 77 ---
1 file changed, 56 insertions(+), 21 deleti
Found by inspection.
Signed-off-by: Daniele Di Proietto
---
lib/dpif-netdev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 1e8a37c..24717cc 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -986,6 +986,7 @@ dp_netdev_free(struct dp_netde
Instead of performing every operation inside do_port_add() it seems
clearer to introduce port_create(), since we already have
port_destroy().
No functional change.
Signed-off-by: Daniele Di Proietto
---
lib/dpif-netdev.c | 69 ++-
1 file chang
The pmds and the main threads are synchronized using a condition
variable. The main thread writes a new configuration, then it waits on
the condition variable. A pmd thread reads the new configuration, then
it calls signal() on the condition variable. To make sure that the pmds
and the main threa
Instead of duplicating code to add ports in
dp_netdev_set_pmds_on_numa(), we can always use
dp_netdev_add_port_to_pmds__().
Signed-off-by: Daniele Di Proietto
---
lib/dpif-netdev.c | 58 +--
1 file changed, 22 insertions(+), 36 deletions(-)
di
A future commit will stop using RCU for 'dp->ports' and use a mutex for
reading/writing them. To avoid taking a mutex in dp_execute_cb(), which
is called in the fast path, this commit introduces a pmd thread local
cache of ports.
The downside is that every port add/remove now needs to synchronize
Makes popping each member of the hmap a bit easier.
Signed-off-by: Daniele Di Proietto
---
lib/cfm.c| 5 ++---
lib/hmap.h | 4
lib/id-pool.c| 5 ++---
lib/learning-switch.c| 5 ++---
lib/netdev-linux.c | 5 ++---
l
The interface will be more similar to the cmap.
Signed-off-by: Daniele Di Proietto
---
lib/hmap.c | 26 --
lib/hmap.h | 7 ++-
lib/sset.c | 12 +---
lib/sset.h | 7 ++-
ofproto/ofproto-dpif.c | 8 +++-
netdev objects are hard to use with RCU, because it's not possible to
split removal and reclamation. Postponing the removal means that the
port is not removed and cannot be readded immediately. Waiting for
reclamation means introducing a quiescent state, and that may introduce
subtle bugs, due to
ovs_mutex_cond_wait() is used in many functions in dpif-netdev to
synchronize with pmd threads, but we can't guarantee that the callers do
not hold RCU references, so it's better to avoid quiescing.
In system_stats_thread_func() the code relied on ovs_mutex_cond_wait()
to introduce a quiescent sta
Signed-off-by: Daniele Di Proietto
---
lib/dpif-netdev.c | 144 ++--
lib/dpif-provider.h | 3 +-
2 files changed, 84 insertions(+), 63 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 8cc37e2..c905d1d 100644
--- a/lib/dpif-
A netdev provider, especially a PMD provider (like netdev DPDK) might
not be able to change some of its parameters (such as MTU, or number of
queues) without stopping everything and restarting.
This commit introduces a mechanism that allows a netdev provider to
request a restart (netdev_request_re
An upcoming commit will make dpif_poll_threads_set() record the
requested configuration and dpif_run() apply it, so it makes sense to
change the order.
Signed-off-by: Daniele Di Proietto
Tested-by: Ilya Maximets
Acked-by: Ilya Maximets
Acked-by: Mark Kavanagh
---
ofproto/ofproto-dpif.c | 4 ++
Errors returned by netdev_set_multiq() and netdev_rxq_open() weren't
handled properly in reconfigure_pmd_threads(). In case of error now we
remove the port from the datapath.
Also, part of the code is moved in a new function, port_reconfigure().
Signed-off-by: Daniele Di Proietto
---
lib/dpif-
This introduces in dpif-netdev and netdev-dpdk the first use for the
newly introduce reconfigure netdev call.
When a request to change the number of queues comes, netdev-dpdk will
remember this and notify the upper layer via
netdev_request_reconfigure().
The datapath, instead of periodically call
On 19/04/2016 02:48, "Ilya Maximets" wrote:
>On 19.04.2016 10:18, Ilya Maximets wrote:
>> There was a reason for 2 calls for dp_netdev_pmd_reload_done() inside
>> pmd_thread_main(). The reason is that we must wait until PMD thread
>> completely done with reloading. This patch introduces race co
On Mon, Apr 18, 2016 at 2:57 AM, Thadeu Lima de Souza Cascardo
wrote:
> On Fri, Apr 15, 2016 at 08:36:51PM -0700, Jesse Gross wrote:
>> On Fri, Apr 15, 2016 at 2:30 PM, Thadeu Lima de Souza Cascardo
>> wrote:
>> > Hi, this is an RFC patch (could probably be 2 patches - more below), that
>> > crea
George went oï the top of food.What's up my lovely peckeri found yr images on
instagram! you are rogue!are u down to f$ck? i hate commitment !! just want to have
fun... if ur good in bed, we should chat..The screenname - Kassie1994my profile is over
there: http://jaxzbivh。DatingSurfer。ru/?acc=K
On 15 April 2016 at 17:02, Daniele Di Proietto wrote:
> The userspace connection tracker doesn't support ALGs, frag reassembly
> or NAT yet, so skip those tests.
>
> Also, connection tracking state input from a local port is not possible
> in userspace.
>
> Finally, the userspace datapath pads all
Hi,
As part of this patch the following scenario takes place:
An ovsdb-server acts as a client of another ovsdb-server: Server A is
monitoring
Server B tables, so Server B constantly sends update notifications to Server A.
When an update notification arrives to Server A, I'd like to process th
On 19 April 2016 at 16:53, Joe Stringer wrote:
> On 15 April 2016 at 17:02, Daniele Di Proietto wrote:
>> The userspace connection tracker doesn't support ALGs, frag reassembly
>> or NAT yet, so skip those tests.
>>
>> Also, connection tracking state input from a local port is not possible
>> in
Hi Mauricio Vasquez:
root@host52:~# ovs-vsctl show
f2664a74-7523-4240-812a-4f022051346a
Bridge "ovsbr0"
Port "vhost-user-1"
Interface "vhost-user-1"
type: dpdkvhostuser
Port "dpdk0"
Interface "dpdk0"
typ
> -Original Message-
> From: Jesse Gross [mailto:je...@kernel.org]
> Sent: Tuesday, April 19, 2016 11:22 PM
> To: Li, Johnson
> Cc: ovs dev
> Subject: Re: [ovs-dev] [PATCH] Add VxLAN-GBP support for user space data
> path
>
> On Tue, Apr 19, 2016 at 3:20 AM, Johnson.Li wrote:
> > diff -
54 matches
Mail list logo