On Mon, Jun 8, 2015 at 7:42 PM, Pravin Shelar wrote:
> On Mon, Jun 8, 2015 at 6:13 PM, Wei li 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 tx will be accessed without
On Mon, Jun 8, 2015 at 7:42 PM, Pravin Shelar wrote:
> On Mon, Jun 8, 2015 at 6:13 PM, Wei li 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 tx will be accessed without
On Tuesday 16 June 2015 11:39 AM, Wei Li wrote:
On 2015/6/15 22:33, Gowrishankar wrote:
From: Gowrishankar M
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).
On 2015/6/15 22:33, Gowrishankar wrote:
From: Gowrishankar M
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 correction is on right command to use to
Also updated an additional correction as in :
http://openvswitch.org/pipermail/dev/2015-June/056379.html
Could you include that as well.
Thanks,
Gowrishankar
On Tuesday 16 June 2015 11:19 AM, Wei li wrote:
Signed-off-by: Wei li
---
INSTALL.DPDK.md | 4 ++--
1 file changed, 2 insertions(+),
Signed-off-by: Wei li
---
INSTALL.DPDK.md | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/INSTALL.DPDK.md b/INSTALL.DPDK.md
index cdef6cf..1287fea 100644
--- a/INSTALL.DPDK.md
+++ b/INSTALL.DPDK.md
@@ -355,7 +355,7 @@ have arbitrary names.
- For vhost-user, the name of
On Wed, Jun 10, 2015 at 07:07:28AM +0300, Liran Schour wrote:
> I want to add learning capabilities for lport with unknown macs set..
>
> When an unknown mac source address is going out from a logical port that
> its macs sets to "unknown", an additional controller action will be take
> and the
OK, great, I applied this to master.
On Fri, Jun 12, 2015 at 1:54 PM, Alin Serdean
wrote:
> Acked-by: Alin Gabriel Serdean
>
>
>
> -Mesaj original-
> De la: dev [mailto:dev-boun...@openvswitch.org] În numele Sorin Vinturis
> Trimis: Thursday, May 28, 2015 11:31 PM
> Către: dev@openvswitc
Any ovsdb programs that accepts a single schema file name will be
extended to accepts multiple file names. This patch implements the
file name parsing logic for future patches.
Signed-off-by: Andy Zhou
---
ovsdb/ovsdb.c | 39 +++
ovsdb/ovsdb.h | 2 ++
2 files
ovsdb-tool now accepts multiple schema files for all applicable options.
If these schema can be joined together without compatibility error,
the joined schema will be used as the runtime DB schema.
When storing joined schemata into a database file, A JSON array will
be store as its first record, w
Add unit tests for ovsdb-tools that deals with multiple schemas.
Signed-off-by: Andy Zhou
---
tests/ovsdb-tool.at | 152
1 file changed, 152 insertions(+)
diff --git a/tests/ovsdb-tool.at b/tests/ovsdb-tool.at
index 0d3219b..e61ae79 100644
--
OVSDB can be more flexible by supporting multiple schemas. Those multiple
schemas can be taken
from different releases, or on different feature sets. If the schema tables
don't overlap,
or for tables do overlap, but columns are not in conflict, those schemas can be
joined together
as single sche
Add functions to support dealing with multiple schemas as a set. These
functions will be useful in the following patch.
Signed-off-by: Andy Zhou
---
ovsdb/ovsdb.c | 153 ++
ovsdb/ovsdb.h | 34 -
2 files changed, 186 insertions(
Add ovsdb_schema_join that joins two schemas into a new schema that
is the union of the originally tow schemas.
Signed-off-by: Andy Zhou
---
ovsdb/column.c | 13 +
ovsdb/column.h | 2 ++
ovsdb/ovsdb.c | 28 +++-
ovsdb/ovsdb.h | 3 +++
ovsdb/table.c | 39 +
nQ óeXÙ¯%{qëÚ:¯/Û°ÝW,áí½'<ÒD1Hbw
448Ç{âòôì]ï¬xtÚÔ 8uËË
ìÝ8YØÛpÛ'Ï(ÂߢÃÇÅvÇry]Ä&×Výr.ȽµÊ¤]ìï´HCÚgôü{eõSC;HZêúå×è.FÙËóÐ`wGò*¿ðËH'çQd«G,dVgh¯Ô`ó³ø#a
7-¬*l¯M6'ÝAçDº$´
ÏÓßîYëÍݤOØÓÇÁÉÏù¡5×»«*UnòùzÒ¶}´×jÝBO³õSh¹:£¡ÏNønML«èBü÷FÑ)ò¯åñ)Ú&¬½ç>ÝR§ã{ÕÊõìaÙzhøkA³¦"Z/Lµr¾Ýmm
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
regardless of if they also output to non-tunnel vports which is allowed for
MPLS packets
On Wed, Jun 10, 2015 at 03:13:54PM -0400, Russell Bryant wrote:
> Provider Networks
> =
>
> OpenStack Neutron currently has a feature referred to as "provider
> networks". This is used as a way to define existing physical networks
> that you would like to integrate into your envir
Thanks, applied.
On Mon, Jun 15, 2015 at 03:33:34PM -0700, Alex Wang wrote:
> Acked-by: Alex Wang
>
> On Mon, Jun 15, 2015 at 3:28 PM, Ben Pfaff wrote:
>
> > The keys in the Chassis table's "gateway_ports" column report the same
> > information as the Gateway table's "attached_port" column, so
Acked-by: Alex Wang
On Mon, Jun 15, 2015 at 3:28 PM, Ben Pfaff wrote:
> The keys in the Chassis table's "gateway_ports" column report the same
> information as the Gateway table's "attached_port" column, so this commit
> removes the latter.
>
> Signed-off-by: Ben Pfaff
> ---
> ovn/ovn-sb.ovss
On Thu, Jun 11, 2015 at 07:36:56PM -0700, Alex Wang wrote:
> Signed-off-by: Alex Wang
Acked-by: Ben Pfaff
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
The keys in the Chassis table's "gateway_ports" column report the same
information as the Gateway table's "attached_port" column, so this commit
removes the latter.
Signed-off-by: Ben Pfaff
---
ovn/ovn-sb.ovsschema | 3 +--
ovn/ovn-sb.xml | 9 ++---
2 files changed, 3 insertions(+), 9
Acked-by: Ethan Jackson
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 I don't know if
it would actually affect things.
Merged.
Ethan
On Mon, Jun 15, 2015 at 11:0
Acked-by: Alex Wang
On Sun, Jun 14, 2015 at 12:19 PM, Ben Pfaff wrote:
> This makes life easier for testing at the point you start to separate your
> environment into multiple machines.
>
> Also work on the manpage a little.
>
> Signed-off-by: Ben Pfaff
> ---
> ovn/ovn-nbctl.8.xml | 38
On Thu, Jun 11, 2015 at 04:07:00PM -0700, Sabyasachi Sengupta wrote:
>
> Abort openvswitch startup script if ovsdb startup fails for
> some reason. This helps in getting the system startup to NOT hang
> indefinitely, as was seen in a recent report when ovsdb failed with
> "I/O error: /etc/openvswi
The current Geneve code in OVS only supports a feature set equal
to VXLAN - the ability to set and match on the VNI. This patch series
provides the more interesting part, which is the ability to use
options.
This series should be generally complete and stable but I thought that
it would be a good
This adds support for receiving variable length fields encoded in
NXM/OXM and mapping them into OVS internal structures. In order
for this to make sense, we need to define some semantics:
There are three lengths that matter in this process: the maximum
size of the field (represented as the existin
Thanks! Applied to master.
On Fri, Jun 12, 2015 at 04:49:07PM -0700, Jarno Rajahalme wrote:
> Found this from junk mail...
>
> Acked-by: Jarno Rajahalme
>
> Jarno
>
>
> > On Jun 9, 2015, at 11:32, Ben Pfaff wrote:
> >
> > Signed-off-by: Ben Pfaff
> > ---
> > v1->v2: Found and converted
Can this patch be considered? We keep hitting it in our data center where a
number of fresh installed EL7 hosts fail to boot up due to default selinux
permissions, but ovs should not prevent the system to bootup to multiuser..
On Thu, 11 Jun 2015, Sabyasachi Sengupta wrote:
Abort openvswit
The addition of Geneve metadata requires a large amount of additional
space to handle the maximum set of options. In most cases, this is
not a big deal since it is only temporary storage on the stack or
can be automatically stripped out for miniflows. However, userspace
tunnels need to deal with th
The current support for Geneve in OVS is exactly equivalent to VXLAN:
it is possible to set and match on the VNI but not on any options
contained in the header. This patch enables the use of options.
The goal for Geneve support is not to add support for any particular option
but to allow end users
In order to work with Geneve options, we need to maintain a mapping
table between an option (defined by ) and
an NXM field that can be operated on for the purposes of matches,
actions, etc. This mapping must be explicitly specified by the
user.
Conceptually, this table could be communicated using
When reading in hex strings that form NXM fields, we don't need to
enforce size constraints if the fields are variable length.
Instead, we can set the header size based on the string length.
Signed-off-by: Jesse Gross
---
lib/nx-match.c | 20 ++--
1 file changed, 18 insertions(+)
It is technically correct to send the entire maximum length of
a field when it is variable length. However, it is awkward to
do so and not what one would naively expect. Since receivers will
internally zero-extend fields, we can do the opposite and trim
off leading zeros. This results in encodings
Currently when an NXM field is encoded, the caller must specify
the length of the data being provided. However, this data is
always placed into a field of standard length. In order to
support variable length options, the length field must also
alter the size in the header. The previous implementati
Until now, all fields that OVS can match against have been fixed
size (variable length headers can be skipped during parsing but
the match is fixed). However, Geneve options can vary in size
so we must not require the size of these fields to be known
at compile time.
This allows data types to be a
Currently we treat the entire NXM/OXM header, including length,
as an ID to define a field. However, this does not allow for
multiple lengths of a particular field.
If a field has been marked as variable, we should ignore the length
when looking up the field and only use the class and field. We
co
Acked-by: Eitan Eliahu
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Gurucharan Shetty
Sent: Monday, June 15, 2015 12:52 PM
To: dev@openvswitch.org
Cc: Gurucharan Shetty
Subject: [ovs-dev] [PATCH ovn] ofctrl: Don't use designated initializer
On Fri, Jun 12, 2015 at 09:08:33PM -0400, Russell Bryant wrote:
> The git repository for the neutron plugin has been renamed to reflect
> that it is now officially part of the OpenStack Neutron project. The
> repo now lives in the "openstack" namespace.
>
> Also remove the link to the todo file a
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 ovn-controller:
>
> # systemctl start ovn-controller
>
> If you wan
On Fri, Jun 12, 2015 at 12:51:23PM -0400, Russell Bryant wrote:
> This patch introduces ovn-ctl, which is similar to ovs-ctl. I opted
> for a new script as everything in OVN so far is nicely isolated, so a
> new script seemed to make the most sense.
>
> If you'd like to run ovn-controller on a ho
On Mon, Jun 15, 2015 at 12:51:57PM -0700, Gurucharan Shetty wrote:
> MSVC 2013 does not like designated initializers when
> structs are initialized inside structs.
> Apparently it has been fixed in MSCV 2015.
>
> Signed-off-by: Gurucharan Shetty
It's necessary to also memset the structures to 0
MSVC 2013 does not like designated initializers when
structs are initialized inside structs.
Apparently it has been fixed in MSCV 2015.
Signed-off-by: Gurucharan Shetty
---
ovn/controller/ofctrl.c | 41 +++--
1 file changed, 19 insertions(+), 22 deletions(-)
It doesn't make sense to mix "conjunction" actions with most other kinds
of actions. That's because flows with "conjunction" actions aren't ever
actually executed, so any actions mixed up with them would never do
anything useful. "note" actions are a little different because they never
do anythin
On Sun, Jun 14, 2015 at 12:19 PM, Ben Pfaff 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.
>
> Signed-off-by: Ben Pfaff
> ---
> build-aux/xml2nroff | 30
Acked-by: Alex Wang
On Sun, Jun 14, 2015 at 12:19 PM, Ben Pfaff wrote:
> When --enable-dummy=system or --enable-dummy=override is in use, dpifs
> other than "dummy" are actually dummy dpifs, so use a more reliable test.
>
> Signed-off-by: Ben Pfaff
> ---
> lib/dpif-netdev.c | 9 -
> l
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 mine, looks good,
On Mon, Jun 15, 2015 at 11:36 AM, Ben Pfa
I think it would be a little better, but I know of third-party trees
that use --enable-dummy (I don't know whether they use
--enable-dummy=override specifically) for testing. I don't want to
break their use cases just for a little bit of better naming.
On Mon, Jun 15, 2015 at 11:04:04AM -0700, Al
Thanks for the ack!
The FAQ entries I had in mind are about IP addresses and why you
shouldn't put them on physical ports, not about where you should put
your dummy connections (I don't think dummies are mentioned at all in
the FAQ, since they're a dev testing only feature).
Here are the ones I w
On Mon, Jun 15, 2015 at 11:28:12AM -0700, Gurucharan Shetty wrote:
> On Mon, Jun 15, 2015 at 11:09 AM, Daniele Di Proietto
> 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
On Mon, Jun 15, 2015 at 07:09:29PM +0100, Daniele Di Proietto 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
> datapath.
>
> Si
On Mon, Jun 15, 2015 at 11:09 AM, Daniele Di Proietto
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
> datapath.
>
> Signed-off
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
---
vtep/ovs-vtep | 12 ++--
1 file ch
It appears that miniflow_extract() in emc_processing() spends a lot of
cycles waiting for the packet's data to be read.
Prefetching the next packet's data while parsing removes this delay.
For a single flow pipeline the throughput improves by ~10%. With a
more realistic pipeline the change has a
Do you think it is better to change the optarg to "override-all" and
"override-system"?
Acked-by: Alex Wang
On Sun, Jun 14, 2015 at 12:19 PM, Ben Pfaff wrote:
> Until now there have been two variants for --enable-dummy:
>
> * --enable-dummy: This adds support for "dummy" dpif and netdev.
>
On 15/06/2015 12:16, "Traynor, Kevin" wrote:
>
>> -Original Message-
>> From: Traynor, Kevin
>> Sent: Thursday, June 11, 2015 4:55 PM
>> To: Daniele Di Proietto; Flavio Leitner
>> Cc: dev@openvswitch.org; Gray, Mark D
>> Subject: RE: [ovs-dev] Is this an issue for DPDK vhost rss?
>>
>>
Acked-by: Alex Wang
On Sun, Jun 14, 2015 at 12:18 PM, Ben Pfaff wrote:
> I found this invaluable for debugging native tunnels over dummy devices.
>
> Signed-off-by: Ben Pfaff
> ---
> ofproto/ofproto-dpif-xlate.c | 17 -
> 1 file changed, 16 insertions(+), 1 deletion(-)
>
> dif
Acked-by: Alex Wang
On Sun, Jun 14, 2015 at 12:19 PM, Ben Pfaff wrote:
> Otherwise, if netdev_unregister_provider() is called before any other
> netdev function, netdev_class_mutex is not initialized and the attempt to
> lock it aborts.
>
> This doesn't fix an existing bug but with the followin
Thx for your explanation, makes sense!
Could you point me to the FAQ explaining this? Did not find it by
searching key words:
remote, conn
Acked-by: Alex Wang
On Mon, Jun 15, 2015 at 8:49 AM, Ben Pfaff wrote:
> I think that it would be unusual to have a single interface with both an
> IP add
Recent tests added for openflow atomic bundle support
looks for Linux specific output in logs. Adjust them
so that tests pass in Windows too.
Signed-off-by: Gurucharan Shetty
---
tests/ofproto-macros.at |9 -
tests/ofproto.at|4 ++--
tests/ovs-ofctl.at |4 ++--
3
On Windows, we don't do a good job of translating errors to Linux
type errors. So in the tests, don't look for a specific type of
error.
Signed-off-by: Gurucharan Shetty
---
tests/ofproto.at |5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/tests/ofproto.at b/tests/ofp
I think that it would be unusual to have a single interface with both an
IP address assigned and a connection to a remote switch. This is
because an IP address is normally on an internal port (most commonly the
port named the same as the bridge), whereas connections to remote
switches (which are a
Hi Sorin,
Can you please forward stack trace?
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Sorin Vinturis
Sent: Monday, June 15, 2015 7:49 AM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH v2] datapath-windows: BSOD when disabling the
ext
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 behaviour is correct and expected,
but the BSOD is caused in this cas
From: Gowrishankar M
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 correction is on right command to use to add vhost-user port.
Signed-off-by: Gowr
Hi Ben,
Thanks for the confirmation. TCS team will contribute these features to
OpenVSwitch.
Thanks & Regards
Niti Rohilla
Tata Consultancy Services
On Fri, Jun 12, 2015 at 8:34 PM, Ben Pfaff wrote:
> On Fri, Jun 12, 2015 at 05:22:19PM +0530, niti Rohilla wrote:
> > I want to implement and con
> -Original Message-
> From: Traynor, Kevin
> Sent: Thursday, June 11, 2015 4:55 PM
> To: Daniele Di Proietto; Flavio Leitner
> Cc: dev@openvswitch.org; Gray, Mark D
> Subject: RE: [ovs-dev] Is this an issue for DPDK vhost rss?
>
>
> > -Original Message-
> > From: Daniele Di Proi
66 matches
Mail list logo
